==================================================Ascend
============================= test session starts ==============================
platform linux -- Python 3.9.19, pytest-6.2.5, py-1.11.0, pluggy-1.5.0
rootdir: /home/jenkins/mindspore/testcases/testcases/tests/st/ops/allcases_onecard, configfile: ../../../../../../../sault/virtual_test/virtualenv_003/sault/config/pytest.ini
plugins: mock-3.14.0, hydra-core-1.3.2, forked-1.6.0, anyio-4.9.0, xdist-1.32.0
collected 1 item

test_ops_group_cases.py 
platform: ascend910b, max workers: 2147483647, memory threshold: 51200M
level: level0, group_name: unary_op
Start group testing...

ops group_cases_0 with 182 cases start to running, all cases are below:
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.astype')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.byte')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.clone')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.contiguous')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.is_contiguous')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.numpy')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.to')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.clone')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.empty')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.empty_like')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.nn.functional.gelu')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.nn.GELU')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.nn.functional.glu')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.nn.GLU')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.nn.functional.hardsigmoid')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.nn.functional.hardswish')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.nn.functional.leaky_relu')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.nn.functional.logsigmoid')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.nn.Identity')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.nn.LogSigmoid')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.nn.PReLU')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.nn.functional.relu')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.nn.ReLU')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.nn.functional.relu6')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.nn.ReLU6')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.tanh')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.tanh')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.nn.Tanh')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.floor')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.ceil')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.erf')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.erfc')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.erfinv')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.exp2')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.exp')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.log')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.neg')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.sigmoid')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.sqrt')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.abs')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.floor')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.ceil')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.exp')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.log')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.neg')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.sigmoid')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.sqrt')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.abs')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.square')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.ones_like')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.zeros_like')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.tan')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.sin')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.sinc')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.sin')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.bfloat16')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.bool')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.double')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.float')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.half')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.int')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.long')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.logical_not')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.logical_not')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.nn.functional.selu')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.tile')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.tile')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.acos')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.atanh')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.atan')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.sinh')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.acosh')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.asinh')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.asin')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.cosh')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.cos')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.cos')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.imag')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.real')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.reciprocal')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.reciprocal')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.trunc')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.expm1')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.rsqrt')
case: (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'mint.sign')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.astype')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.byte')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.clone')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.contiguous')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.is_contiguous')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.numpy')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.to')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.clone')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.empty')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.empty_like')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.nn.functional.gelu')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.nn.GELU')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.nn.functional.glu')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.nn.GLU')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.nn.functional.hardsigmoid')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.nn.functional.hardswish')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.nn.functional.leaky_relu')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.nn.functional.logsigmoid')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.nn.Identity')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.nn.LogSigmoid')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.nn.PReLU')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.nn.functional.relu')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.nn.ReLU')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.nn.functional.relu6')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.nn.ReLU6')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.tanh')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.tanh')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.nn.Tanh')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.floor')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.ceil')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.erf')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.erfc')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.erfinv')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.exp2')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.exp')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.log')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.neg')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.sigmoid')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.sqrt')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.abs')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.floor')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.ceil')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.exp')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.log')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.neg')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.sigmoid')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.sqrt')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.abs')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.square')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.ones_like')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.zeros_like')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.tan')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.sin')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.sinc')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.sin')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.bfloat16')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.bool')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.double')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.float')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.half')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.int')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.long')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.logical_not')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.logical_not')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.nn.functional.selu')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.tile')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.tile')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.acos')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.atanh')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.atan')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.sinh')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.acosh')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.asinh')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.asin')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.cosh')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.cos')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.cos')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.imag')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.real')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'Tensor.reciprocal')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.reciprocal')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.trunc')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.expm1')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.rsqrt')
case: (<function test_unary_op_reference_backward at 0xffff7d3a2f70>, 'pynative', 'mint.sign')
case: (<function test_unary_op_reference_forward_kbk at 0xffff7d369040>, 'kbk', 'mint.tanh')
case: (<function test_unary_op_reference_forward_kbk at 0xffff7d369040>, 'kbk', 'mint.nn.functional.relu')
case: (<function test_unary_op_reference_forward_kbk at 0xffff7d369040>, 'kbk', 'mint.tile')
case: (<function test_unary_op_reference_forward_kbk at 0xffff7d369040>, 'kbk', 'mint.acosh')
case: (<function test_unary_op_reference_forward_kbk at 0xffff7d369040>, 'kbk', 'mint.clone')
case: (<function test_unary_op_reference_forward_kbk at 0xffff7d369040>, 'kbk', 'Tensor.contiguous')
case: (<function test_unary_op_reference_backward_kbk at 0xffff7d3690d0>, 'kbk', 'mint.tanh')
case: (<function test_unary_op_reference_backward_kbk at 0xffff7d3690d0>, 'kbk', 'mint.nn.functional.relu')
case: (<function test_unary_op_reference_backward_kbk at 0xffff7d3690d0>, 'kbk', 'mint.tile')
case: (<function test_unary_op_reference_backward_kbk at 0xffff7d3690d0>, 'kbk', 'mint.acosh')
case: (<function test_unary_op_reference_backward_kbk at 0xffff7d3690d0>, 'kbk', 'mint.clone')
case: (<function test_unary_op_reference_backward_kbk at 0xffff7d3690d0>, 'kbk', 'Tensor.contiguous')
ops group_cases_0 total running memory: 2912M, memory threshold: 51200M

op_name: Tensor.is_contiguous, mode:pynative, test_op_reference...

op_name: Tensor.numpy, mode:pynative, test_op_reference...

op_name: Tensor.astype, mode:pynative, test_op_reference...

op_name: Tensor.byte, mode:pynative, test_op_reference...

op_name: Tensor.contiguous, mode:pynative, test_op_reference...

op_name: Tensor.clone, mode:pynative, test_op_reference...

op_name: Tensor.to, mode:pynative, test_op_reference...

op_name: mint.clone, mode:pynative, test_op_reference...
.[LOG_WARNING] can not open file, file: /home/jenkins/ascend/log/debug/plog/plog-159072_20260129173646295.log, possible reason: Permission denied.............[LOG_WARNING] can not open file, file: /home/jenkins/ascend/log/debug/plog/plog-159072_20260129173646295.log, possible reason: Permission denied.....[LOG_WARNING] can not open file, file: /home/jenkins/ascend/log/debug/plog/plog-159072_20260129173646295.log, possible reason: Permission denied...................................................................
TotalTime = 0.0351329, [30]
[bootstrap]: 0.00049239
[type_inference]: 0.0180954
[event_method]: 2.214e-05
[auto_monad]: 0.00016564
[graph_reusing]: 6.85998e-06
[pre_auto_parallel]: 1.036e-05
[py_interpret_to_execute]: 9.184e-05
[rewriter_before_opt_a]: 0.00014739
[expand_dump_flag]: 3.36001e-06
[jit_opt_a]: 0.0131949, [2]
    [Cycle 1]: 0.0027015, [27]
        [switch_simplify]: 0.00022024
        [loop_unroll]: 3.73e-05
        [a_1]: 0.00094774
        [with_stream_mark]: 2.583e-05
        [recompute_prepare]: 8.67998e-06
        [updatestate_depend_eliminate]: 4.28001e-06
        [updatestate_assign_eliminate]: 3.26001e-06
        [updatestate_loads_eliminate]: 2.68998e-06
        [parameter_eliminate]: 2.39001e-06
        [specialize_transform]: 6.74001e-06
        [updatestate_useless_node_eliminater]: 5.30999e-06
        [accelerated_algorithm]: 6.61999e-06
        [meta_shard_fg_expand]: 1.644e-05
        [get_grad_eliminate_]: 1.703e-05
        [merge_forward]: 3.75e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.124e-05
        [j_node_and_user_rematch]: 1.185e-05
        [meta_fg_expand]: 2.94999e-06
        [replace_old_param]: 1.265e-05
        [inline_without_move]: 6.98e-06
        [renormalize]: 0.00101349
        [add_forward_monad_depend]: 2.51e-05
        [auto_monad_grad]: 2.79999e-06
        [auto_monad_eliminator]: 3.156e-05
        [cse]: 3.407e-05
        [replace_applicator]: 1.308e-05
    [Cycle 2]: 0.0003273, [27]
        [switch_simplify]: 5.92999e-06
        [loop_unroll]: 5.05999e-06
        [a_1]: 9.41e-05
        [with_stream_mark]: 1.091e-05
        [recompute_prepare]: 5.26998e-06
        [updatestate_depend_eliminate]: 2.61e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.18001e-06
        [specialize_transform]: 5.21002e-06
        [updatestate_useless_node_eliminater]: 5.28002e-06
        [accelerated_algorithm]: 5.53002e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 4.48999e-06
        [merge_forward]: 2.64001e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.332e-05
        [j_node_and_user_rematch]: 8.46002e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 8.89003e-06
        [inline_without_move]: 4.84e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 8.60018e-07
        [auto_monad_eliminator]: 4.68999e-06
        [cse]: 9.69e-06
        [replace_applicator]: 5.00999e-06
[py_interpret_to_execute_after_opt_a]: 1.427e-05
[rewriter_after_opt_a]: 9.311e-05
[convert_after_rewriter]: 7.95998e-06
[order_py_execute_after_rewriter]: 5.40001e-06
[mutable_eliminate]: 0.00085994
[jit_opt_b]: 5.427e-05, [1]
    [Cycle 1]: 4.639e-05, [2]
        [frontend_op_eliminate]: 1.766e-05
        [inline_after_opt_a]: 1.557e-05
[cconv]: 2.745e-05
[loop_unroll]: 0.00045978
[jit_opt_after_cconv]: 0.00014562, [1]
    [Cycle 1]: 0.00013879, [11]
        [c_1]: 2.06e-05
        [parameter_eliminate]: 3.14001e-06
        [updatestate_depend_eliminate]: 5.86e-06
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [cse]: 2.078e-05
        [call_graph_tuple_transform]: 1.916e-05
        [tuple_list_get_item_eliminator]: 5.29e-06
        [none_parameter_eliminate]: 1.97001e-06
        [renormalize]: 4.40021e-07
        [switch_simplify]: 5.61e-06
[remove_dup_value]: 1.56e-05
[partial_unused_args_eliminate]: 2.17001e-06
[environ_conv]: 3.66e-05
[add_recomputation]: 6.197e-05
[cse_after_recomputation]: 2.361e-05, [1]
    [Cycle 1]: 1.777e-05, [1]
        [cse]: 1.128e-05
[auto_monad_reorder]: 2.354e-05
[get_jit_bprop_graph]: 2.27001e-06
[rewriter_after_jit_bprop_graph]: 0.00017225
[opt_after_jit_grad]: 0.00051267
[symbol_engine_optimizer]: 7.493e-05, [1]
    [Cycle 1]: 6.855e-05, [6]
        [build]: 4.27e-06
        [elim_shapecalc]: 7.75e-06
        [elim_not_effective]: 1.294e-05
        [opt_reshape]: 5.75001e-06
        [fold_const_symbol]: 8.77e-06
        [renormalize]: 6.19999e-07
[validate]: 8.91e-05
Sums
    bootstrap                            :     0.000492s :  2.03%
    type_inference                       :     0.018095s : 74.49%
    event_method                         :     0.000022s :  0.09%
    auto_monad                           :     0.000166s :  0.68%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000010s :  0.04%
    py_interpret_to_execute              :     0.000092s :  0.38%
    rewriter_before_opt_a                :     0.000147s :  0.61%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000226s :  0.93%
    jit_opt_a.loop_unroll                :     0.000042s :  0.17%
    jit_opt_a.a_1                        :     0.001042s :  4.29%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000018s :  0.07%
    jit_opt_a.get_grad_eliminate_        :     0.000022s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.09%
    jit_opt_a.inline_without_move        :     0.000012s :  0.05%
    jit_opt_a.renormalize                :     0.001014s :  4.17%
    jit_opt_a.add_forward_monad_depend   :     0.000026s :  0.11%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000036s :  0.15%
    jit_opt_a.cse                        :     0.000044s :  0.18%
    jit_opt_a.replace_applicator         :     0.000018s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.06%
    rewriter_after_opt_a                 :     0.000093s :  0.38%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000860s :  3.54%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.06%
    cconv                                :     0.000027s :  0.11%
    loop_unroll                          :     0.000460s :  1.89%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000037s :  0.15%
    add_recomputation                    :     0.000062s :  0.26%
    cse_after_recomputation.cse          :     0.000011s :  0.05%
    auto_monad_reorder                   :     0.000024s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000172s :  0.71%
    opt_after_jit_grad                   :     0.000513s :  2.11%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000089s :  0.37%


Time group info:
------[substitution.]   0.000484    26
 0.43% :     0.000002s :      2: substitution.elim_not_effective
 0.34% :     0.000002s :      2: substitution.fold_const_symbol
 1.18% :     0.000006s :      3: substitution.graph_param_transform
84.41% :     0.000408s :      6: substitution.inline
 0.87% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.08% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.38% :     0.000007s :      2: substitution.replace_old_param
 7.33% :     0.000035s :      1: substitution.switch_simplify
 2.99% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.017981     2
90.74% :     0.016317s :      1: type_inference.infer
 9.26% :     0.001664s :      1: type_inference.specialize
------[replace.]   0.000196     9
49.93% :     0.000098s :      6: replace.inline
26.68% :     0.000052s :      1: replace.switch_simplify
23.40% :     0.000046s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000452     9
89.58% :     0.000405s :      6: match.inline
 7.62% :     0.000034s :      1: match.switch_simplify
 2.80% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000149   801
 1.38% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.10% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.32% :     0.000002s :     12: predicate.addn_check_dump
 1.45% :     0.000002s :     12: predicate.addn_zero_filter
 3.32% :     0.000005s :     12: predicate.arithmetic_simplify
 1.18% :     0.000002s :     12: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000002s :     12: predicate.compare_switch_simplify
 1.36% :     0.000002s :     12: predicate.depend_value_elim
 1.07% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.16% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.72% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.02% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.07% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     12: predicate.environ_get_eliminate
 1.06% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.67% :     0.000008s :     26: predicate.inline
 0.76% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :      6: predicate.less_batch_normalization
 1.69% :     0.000003s :     14: predicate.list_to_tuple_eliminator_
 2.05% :     0.000003s :     17: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.34% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.74% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     12: predicate.merge_addn
 1.13% :     0.000002s :     12: predicate.minmaximum_grad
 2.10% :     0.000003s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     17: predicate.partial_eliminate
 1.20% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.82% :     0.000003s :     12: predicate.reduce_eliminate
 1.44% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.47% :     0.000002s :     20: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     12: predicate.reshape_eliminate
 1.32% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.58% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.11% :     0.000002s :     12: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.75% :     0.000001s :      6: predicate.special_op_eliminate
 0.82% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.31% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.88% :     0.000004s :     20: predicate.switch_defer_inline
 2.19% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.43% :     0.000013s :     65: predicate.switch_simplify
 1.21% :     0.000002s :     12: predicate.tile_eliminate
 1.22% :     0.000002s :     12: predicate.transpose_eliminate
 1.41% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.89% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.95% :     0.000003s :     14: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.39% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     12: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.92% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001391    15
49.26% :     0.000685s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.20% :     0.000184s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.54% :     0.000522s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.037562    72
 0.18% :     0.000066s :      1: add_recomputation
 0.45% :     0.000170s :      1: auto_monad
 0.07% :     0.000026s :      1: auto_monad_reorder
 1.37% :     0.000516s :      1: bootstrap
 0.08% :     0.000030s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000026s :      1: cse_after_recomputation
 0.11% :     0.000040s :      1: environ_conv
 0.07% :     0.000027s :      1: event_method
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
35.14% :     0.013199s :      1: jit_opt_a
 0.40% :     0.000149s :      1: jit_opt_after_cconv
 0.15% :     0.000057s :      1: jit_opt_b
 1.25% :     0.000468s :      1: loop_unroll
 2.31% :     0.000869s :      1: mutable_eliminate
 3.83% :     0.001438s :     26: opt.transform.jit_opt_a
 0.13% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000026s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.39% :     0.000521s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000012s :      1: pre_auto_parallel
 0.25% :     0.000095s :      1: py_interpret_to_execute
 0.04% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 1.64% :     0.000614s :      1: renormalize.infer
 1.04% :     0.000389s :      1: renormalize.specialize
 0.47% :     0.000176s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000097s :      1: rewriter_after_opt_a
 0.40% :     0.000151s :      1: rewriter_before_opt_a
 0.21% :     0.000077s :      1: symbol_engine_optimizer
48.22% :     0.018112s :      1: type_inference


TotalTime = 0.667847, [33]
[bootstrap]: 0.00059582
[type_inference]: 0.0158508
[event_method]: 2.378e-05
[auto_monad]: 0.00019473
[graph_reusing]: 6.71e-06
[pre_auto_parallel]: 1.266e-05
[py_interpret_to_execute]: 7.068e-05
[rewriter_before_opt_a]: 0.00032435
[expand_dump_flag]: 5.07e-06
[jit_opt_a]: 0.0145123, [2]
    [Cycle 1]: 0.00247438, [27]
        [switch_simplify]: 0.00019428
        [loop_unroll]: 3.399e-05
        [a_1]: 0.00068814
        [with_stream_mark]: 2.506e-05
        [recompute_prepare]: 8.21002e-06
        [updatestate_depend_eliminate]: 4.13001e-06
        [updatestate_assign_eliminate]: 2.82002e-06
        [updatestate_loads_eliminate]: 2.89001e-06
        [parameter_eliminate]: 2.34999e-06
        [specialize_transform]: 6.12001e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 6.76999e-06
        [meta_shard_fg_expand]: 1.245e-05
        [get_grad_eliminate_]: 5.99e-06
        [merge_forward]: 4.22e-06
        [cell_reuse_recompute_pass]: 1.04998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.058e-05
        [j_node_and_user_rematch]: 1.017e-05
        [meta_fg_expand]: 2.66e-06
        [replace_old_param]: 1.058e-05
        [inline_without_move]: 5.59e-06
        [renormalize]: 0.00109422
        [add_forward_monad_depend]: 3.091e-05
        [auto_monad_grad]: 2.43998e-06
        [auto_monad_eliminator]: 2.871e-05
        [cse]: 3.192e-05
        [replace_applicator]: 1.486e-05
    [Cycle 2]: 0.00036569, [27]
        [switch_simplify]: 6.83e-06
        [loop_unroll]: 5.16002e-06
        [a_1]: 9.987e-05
        [with_stream_mark]: 1.401e-05
        [recompute_prepare]: 5.91e-06
        [updatestate_depend_eliminate]: 3.48999e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 1.98997e-06
        [parameter_eliminate]: 1.47999e-06
        [specialize_transform]: 6.43e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.25001e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 4.63999e-06
        [merge_forward]: 2.98998e-06
        [cell_reuse_recompute_pass]: 1.47999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.482e-05
        [j_node_and_user_rematch]: 8.08999e-06
        [meta_fg_expand]: 1.82001e-06
        [replace_old_param]: 8.13001e-06
        [inline_without_move]: 4.97e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.67999e-06
        [auto_monad_grad]: 9.89996e-07
        [auto_monad_eliminator]: 6.04001e-06
        [cse]: 1.273e-05
        [replace_applicator]: 5.30001e-06
[py_interpret_to_execute_after_opt_a]: 1.48e-05
[rewriter_after_opt_a]: 0.00019717
[convert_after_rewriter]: 6.155e-05
[order_py_execute_after_rewriter]: 6.69001e-06
[mutable_eliminate]: 0.00082741
[jit_opt_b]: 5.747e-05, [1]
    [Cycle 1]: 4.87e-05, [2]
        [frontend_op_eliminate]: 1.948e-05
        [inline_after_opt_a]: 1.668e-05
[cconv]: 3.173e-05
[loop_unroll]: 0.00048066
[jit_opt_after_cconv]: 0.00015489, [1]
    [Cycle 1]: 0.00014831, [11]
        [c_1]: 2.282e-05
        [parameter_eliminate]: 3.9e-06
        [updatestate_depend_eliminate]: 6.98e-06
        [updatestate_assign_eliminate]: 2.59001e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 2.356e-05
        [call_graph_tuple_transform]: 2.192e-05
        [tuple_list_get_item_eliminator]: 5.36002e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 8.2e-07
        [switch_simplify]: 6.74001e-06
[remove_dup_value]: 1.579e-05
[partial_unused_args_eliminate]: 2.74001e-06
[environ_conv]: 3.662e-05
[add_recomputation]: 6.549e-05
[cse_after_recomputation]: 2.494e-05, [1]
    [Cycle 1]: 1.76e-05, [1]
        [cse]: 1.045e-05
[auto_monad_reorder]: 2.225e-05
[get_jit_bprop_graph]: 2.14e-06
[rewriter_after_jit_bprop_graph]: 0.00013731
[opt_after_jit_grad]: 0.00051106
[symbol_engine_optimizer]: 7.638e-05, [1]
    [Cycle 1]: 6.947e-05, [6]
        [build]: 4.47e-06
        [elim_shapecalc]: 7.45e-06
        [elim_not_effective]: 1.248e-05
        [opt_reshape]: 5.74999e-06
        [fold_const_symbol]: 9.57001e-06
        [renormalize]: 8.39995e-07
[validate]: 9.565e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.63309
[execute]: 1.201e-05
Sums
    bootstrap                            :     0.000596s :  0.09%
    type_inference                       :     0.015851s :  2.42%
    event_method                         :     0.000024s :  0.00%
    auto_monad                           :     0.000195s :  0.03%
    graph_reusing                        :     0.000007s :  0.00%
    pre_auto_parallel                    :     0.000013s :  0.00%
    py_interpret_to_execute              :     0.000071s :  0.01%
    rewriter_before_opt_a                :     0.000324s :  0.05%
    expand_dump_flag                     :     0.000005s :  0.00%
    jit_opt_a.switch_simplify            :     0.000201s :  0.03%
    jit_opt_a.loop_unroll                :     0.000039s :  0.01%
    jit_opt_a.a_1                        :     0.000788s :  0.12%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.01%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.00%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.00%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.00%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.00%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.00%
    jit_opt_a.specialize_transform       :     0.000013s :  0.00%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.00%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.00%
    jit_opt_a.meta_shard_fg_expand       :     0.000014s :  0.00%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.00%
    jit_opt_a.merge_forward              :     0.000007s :  0.00%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.00%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.01%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.00%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.00%
    jit_opt_a.replace_old_param          :     0.000019s :  0.00%
    jit_opt_a.inline_without_move        :     0.000011s :  0.00%
    jit_opt_a.renormalize                :     0.001094s :  0.17%
    jit_opt_a.add_forward_monad_depend   :     0.000033s :  0.00%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000035s :  0.01%
    jit_opt_a.cse                        :     0.000045s :  0.01%
    jit_opt_a.replace_applicator         :     0.000020s :  0.00%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.00%
    rewriter_after_opt_a                 :     0.000197s :  0.03%
    convert_after_rewriter               :     0.000062s :  0.01%
    order_py_execute_after_rewriter      :     0.000007s :  0.00%
    mutable_eliminate                    :     0.000827s :  0.13%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.00%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.00%
    cconv                                :     0.000032s :  0.00%
    loop_unroll                          :     0.000481s :  0.07%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.00%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.00%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000024s :  0.00%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.00%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.00%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.00%
    remove_dup_value                     :     0.000016s :  0.00%
    partial_unused_args_eliminate        :     0.000003s :  0.00%
    environ_conv                         :     0.000037s :  0.01%
    add_recomputation                    :     0.000065s :  0.01%
    cse_after_recomputation.cse          :     0.000010s :  0.00%
    auto_monad_reorder                   :     0.000022s :  0.00%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000137s :  0.02%
    opt_after_jit_grad                   :     0.000511s :  0.08%
    symbol_engine_optimizer.build        :     0.000004s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.00%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.00%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.00%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.00%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000096s :  0.01%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.633090s : 96.60%
    execute                              :     0.000012s :  0.00%


Time group info:
------[substitution.]   0.000305    23
 0.71% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.09% :     0.000006s :      3: substitution.graph_param_transform
87.09% :     0.000265s :      4: substitution.inline
 1.25% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.93% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.98% :     0.000006s :      2: substitution.replace_old_param
 4.48% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.015756     2
90.46% :     0.014253s :      1: type_inference.infer
 9.54% :     0.001503s :      1: type_inference.specialize
------[replace.]   0.000085     6
65.94% :     0.000056s :      4: replace.inline
34.06% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000275     6
95.44% :     0.000263s :      4: match.inline
 4.56% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   691
 1.69% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 2.40% :     0.000003s :     10: predicate.arithmetic_simplify
 1.11% :     0.000002s :     10: predicate.cast_eliminate
 0.66% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.92% :     0.000001s :     10: predicate.compare_switch_simplify
 1.58% :     0.000002s :     10: predicate.depend_value_elim
 0.98% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.07% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.03% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      3: predicate.elim_not_effective
 0.70% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.05% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000002s :     10: predicate.environ_get_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.57% :     0.000001s :      3: predicate.graph_param_transform
 5.25% :     0.000007s :     22: predicate.inline
 0.73% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.48% :     0.000002s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.25% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.42% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 2.66% :     0.000004s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.30% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.57% :     0.000001s :      3: predicate.reset_defer_inline
 1.83% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000002s :     10: predicate.same_eliminate
 0.67% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.58% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.02% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.75% :     0.000004s :     16: predicate.switch_defer_inline
 1.82% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.23% :     0.000011s :     53: predicate.switch_simplify
 1.04% :     0.000001s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.22% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.05% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.68% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.42% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.41% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001080    12
44.73% :     0.000483s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.41% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.86% :     0.000538s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.670051    76
 0.01% :     0.000069s :      1: add_recomputation
 0.03% :     0.000201s :      1: auto_monad
 0.00% :     0.000025s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.09% :     0.000620s :      1: bootstrap
 0.01% :     0.000034s :      1: cconv
 0.01% :     0.000065s :      1: convert_after_rewriter
 0.00% :     0.000027s :      1: cse_after_recomputation
 0.01% :     0.000041s :      1: environ_conv
 0.00% :     0.000030s :      1: event_method
 0.00% :     0.000020s :      1: execute
 0.00% :     0.000007s :      1: expand_dump_flag
 0.00% :     0.000005s :      1: get_jit_bprop_graph
 0.00% :     0.000010s :      1: graph_reusing
 2.17% :     0.014516s :      1: jit_opt_a
 0.02% :     0.000158s :      1: jit_opt_after_cconv
 0.01% :     0.000060s :      1: jit_opt_b
 0.07% :     0.000489s :      1: loop_unroll
 0.13% :     0.000838s :      1: mutable_eliminate
 0.16% :     0.001049s :     26: opt.transform.jit_opt_a
 0.01% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.00% :     0.000029s :      4: opt.transform.jit_opt_b
 0.00% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.00% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.00% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.00% :     0.000032s :      4: opt.transform.symbol_engine_opt
 0.08% :     0.000519s :      1: opt_after_jit_grad
 0.00% :     0.000009s :      1: order_py_execute_after_rewriter
 0.00% :     0.000005s :      1: partial_unused_args_eliminate
 0.00% :     0.000015s :      1: pre_auto_parallel
 0.01% :     0.000074s :      1: py_interpret_to_execute
 0.00% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.00% :     0.000018s :      1: remove_dup_value
 0.09% :     0.000605s :      1: renormalize.infer
 0.07% :     0.000479s :      1: renormalize.specialize
 0.02% :     0.000141s :      1: rewriter_after_jit_bprop_graph
 0.03% :     0.000203s :      1: rewriter_after_opt_a
 0.05% :     0.000331s :      1: rewriter_before_opt_a
 0.01% :     0.000079s :      1: symbol_engine_optimizer
94.49% :     0.633125s :      1: task_emit
 2.37% :     0.015874s :      1: type_inference
 0.02% :     0.000122s :      1: validate


TotalTime = 0.0218297, [33]
[bootstrap]: 0.00041755
[type_inference]: 0.005432
[event_method]: 1.923e-05
[auto_monad]: 6.574e-05
[graph_reusing]: 5.85002e-06
[pre_auto_parallel]: 2.73998e-06
[py_interpret_to_execute]: 3.805e-05
[rewriter_before_opt_a]: 7.336e-05
[expand_dump_flag]: 3.43e-06
[jit_opt_a]: 0.00658159, [2]
    [Cycle 1]: 0.0019217, [27]
        [switch_simplify]: 6.279e-05
        [loop_unroll]: 3.004e-05
        [a_1]: 0.00059806
        [with_stream_mark]: 2.326e-05
        [recompute_prepare]: 8.2e-06
        [updatestate_depend_eliminate]: 3.7e-06
        [updatestate_assign_eliminate]: 3.16999e-06
        [updatestate_loads_eliminate]: 3.01999e-06
        [parameter_eliminate]: 1.90001e-06
        [specialize_transform]: 5.83002e-06
        [updatestate_useless_node_eliminater]: 5.51998e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 1.99999e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 3.99002e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.518e-05
        [j_node_and_user_rematch]: 9.66e-06
        [meta_fg_expand]: 2.45002e-06
        [replace_old_param]: 8.57998e-06
        [inline_without_move]: 5.20999e-06
        [renormalize]: 0.00086432
        [add_forward_monad_depend]: 9.52001e-06
        [auto_monad_grad]: 2.17001e-06
        [auto_monad_eliminator]: 1.429e-05
        [cse]: 3.188e-05
        [replace_applicator]: 1.204e-05
    [Cycle 2]: 0.00032267, [27]
        [switch_simplify]: 5.66e-06
        [loop_unroll]: 5.19e-06
        [a_1]: 9.227e-05
        [with_stream_mark]: 9.90002e-06
        [recompute_prepare]: 5.51e-06
        [updatestate_depend_eliminate]: 2.84001e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 9.00007e-07
        [specialize_transform]: 4.79002e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 4.82e-06
        [meta_shard_fg_expand]: 1.02998e-06
        [get_grad_eliminate_]: 4.63999e-06
        [merge_forward]: 2.42001e-06
        [cell_reuse_recompute_pass]: 1.73002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.329e-05
        [j_node_and_user_rematch]: 7.7e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 8.00999e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.04e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 4.99e-06
        [cse]: 1.056e-05
        [replace_applicator]: 5.64998e-06
[py_interpret_to_execute_after_opt_a]: 1.13e-05
[rewriter_after_opt_a]: 4.384e-05
[convert_after_rewriter]: 7.27002e-06
[order_py_execute_after_rewriter]: 5.24e-06
[mutable_eliminate]: 0.00062595
[jit_opt_b]: 5.184e-05, [1]
    [Cycle 1]: 4.411e-05, [2]
        [frontend_op_eliminate]: 1.667e-05
        [inline_after_opt_a]: 1.548e-05
[cconv]: 2.705e-05
[loop_unroll]: 0.00044051
[jit_opt_after_cconv]: 0.00014428, [1]
    [Cycle 1]: 0.00013785, [11]
        [c_1]: 2.233e-05
        [parameter_eliminate]: 2.59999e-06
        [updatestate_depend_eliminate]: 5.57999e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 2.153e-05
        [call_graph_tuple_transform]: 2.028e-05
        [tuple_list_get_item_eliminator]: 5.25999e-06
        [none_parameter_eliminate]: 1.30001e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 5.10999e-06
[remove_dup_value]: 1.478e-05
[partial_unused_args_eliminate]: 2.22999e-06
[environ_conv]: 5.09e-06
[add_recomputation]: 4.998e-05
[cse_after_recomputation]: 2.1e-05, [1]
    [Cycle 1]: 1.549e-05, [1]
        [cse]: 9.52001e-06
[auto_monad_reorder]: 1.549e-05
[get_jit_bprop_graph]: 1.96998e-06
[rewriter_after_jit_bprop_graph]: 5.22e-06
[opt_after_jit_grad]: 0.00047987
[symbol_engine_optimizer]: 7.658e-05, [1]
    [Cycle 1]: 7.072e-05, [6]
        [build]: 3.87998e-06
        [elim_shapecalc]: 7.71999e-06
        [elim_not_effective]: 1.375e-05
        [opt_reshape]: 6.08002e-06
        [fold_const_symbol]: 1.023e-05
        [renormalize]: 4.39992e-07
[validate]: 3.663e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00690082
[execute]: 9.89999e-06
Sums
    bootstrap                            :     0.000418s :  2.48%
    type_inference                       :     0.005432s : 32.21%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000066s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.23%
    rewriter_before_opt_a                :     0.000073s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.41%
    jit_opt_a.loop_unroll                :     0.000035s :  0.21%
    jit_opt_a.a_1                        :     0.000690s :  4.09%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000864s :  5.13%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.11%
    jit_opt_a.cse                        :     0.000042s :  0.25%
    jit_opt_a.replace_applicator         :     0.000018s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000044s :  0.26%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000626s :  3.71%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000027s :  0.16%
    loop_unroll                          :     0.000441s :  2.61%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.30%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000480s :  2.85%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006901s : 40.92%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000252    23
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000002s :      2: substitution.fold_const_symbol
 2.28% :     0.000006s :      3: substitution.graph_param_transform
86.26% :     0.000218s :      4: substitution.inline
 1.38% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.18% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.74% :     0.000004s :      2: substitution.replace_old_param
 4.74% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005358     2
83.83% :     0.004492s :      1: type_inference.infer
16.17% :     0.000866s :      1: type_inference.specialize
------[replace.]   0.000081     6
65.96% :     0.000053s :      4: replace.inline
34.04% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000226     6
95.20% :     0.000215s :      4: match.inline
 4.80% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.30% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.53% :     0.000002s :     10: predicate.addn_zero_filter
 1.87% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.47% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.75% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.34% :     0.000000s :      3: predicate.graph_param_transform
 5.34% :     0.000006s :     22: predicate.inline
 1.08% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 1.64% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.96% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.27% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.79% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.33% :     0.000002s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.82% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.66% :     0.000009s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.77% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.41% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.05% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000671    12
52.12% :     0.000350s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.23% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.65% :     0.000266s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023709    76
 0.22% :     0.000053s :      1: add_recomputation
 0.29% :     0.000069s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.82% :     0.000432s :      1: bootstrap
 0.12% :     0.000030s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
27.78% :     0.006585s :      1: jit_opt_a
 0.62% :     0.000148s :      1: jit_opt_after_cconv
 0.23% :     0.000055s :      1: jit_opt_b
 1.89% :     0.000448s :      1: loop_unroll
 2.68% :     0.000636s :      1: mutable_eliminate
 3.84% :     0.000911s :     26: opt.transform.jit_opt_a
 0.21% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.06% :     0.000489s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000041s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.24% :     0.000530s :      1: renormalize.infer
 1.37% :     0.000325s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000047s :      1: rewriter_after_opt_a
 0.32% :     0.000076s :      1: rewriter_before_opt_a
 0.33% :     0.000079s :      1: symbol_engine_optimizer
29.18% :     0.006918s :      1: task_emit
22.98% :     0.005449s :      1: type_inference
 0.24% :     0.000058s :      1: validate


TotalTime = 0.0188829, [30]
[bootstrap]: 0.00040286
[type_inference]: 0.00926456
[event_method]: 2.262e-05
[auto_monad]: 8.642e-05
[graph_reusing]: 7.1e-06
[pre_auto_parallel]: 3.09001e-06
[py_interpret_to_execute]: 6.176e-05
[rewriter_before_opt_a]: 8.891e-05
[expand_dump_flag]: 3.38e-06
[jit_opt_a]: 0.00662418, [2]
    [Cycle 1]: 0.00200893, [27]
        [switch_simplify]: 0.00012948
        [loop_unroll]: 3.583e-05
        [a_1]: 0.00072375
        [with_stream_mark]: 1.313e-05
        [recompute_prepare]: 7.10002e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.16997e-06
        [specialize_transform]: 5.78997e-06
        [updatestate_useless_node_eliminater]: 5.34e-06
        [accelerated_algorithm]: 5.72999e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 5.14003e-06
        [merge_forward]: 2.64001e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.24e-05
        [j_node_and_user_rematch]: 9.15001e-06
        [meta_fg_expand]: 1.91998e-06
        [replace_old_param]: 7.93001e-06
        [inline_without_move]: 5.26002e-06
        [renormalize]: 0.0007975
        [add_forward_monad_depend]: 1.01e-05
        [auto_monad_grad]: 1.61998e-06
        [auto_monad_eliminator]: 1.025e-05
        [cse]: 1.791e-05
        [replace_applicator]: 1.233e-05
    [Cycle 2]: 0.00031946, [27]
        [switch_simplify]: 5.69e-06
        [loop_unroll]: 5.22999e-06
        [a_1]: 9.402e-05
        [with_stream_mark]: 9.07001e-06
        [recompute_prepare]: 4.99e-06
        [updatestate_depend_eliminate]: 2.59999e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.80001e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 4.91002e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.37e-06
        [merge_forward]: 2.52001e-06
        [cell_reuse_recompute_pass]: 1.86998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.261e-05
        [j_node_and_user_rematch]: 7.85e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 7.8e-06
        [inline_without_move]: 4.90999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 4.82e-06
        [cse]: 9.96e-06
        [replace_applicator]: 5.02e-06
[py_interpret_to_execute_after_opt_a]: 1.049e-05
[rewriter_after_opt_a]: 2.763e-05
[convert_after_rewriter]: 5.36002e-06
[order_py_execute_after_rewriter]: 4.16001e-06
[mutable_eliminate]: 0.00058231
[jit_opt_b]: 5.421e-05, [1]
    [Cycle 1]: 4.71e-05, [2]
        [frontend_op_eliminate]: 1.837e-05
        [inline_after_opt_a]: 1.613e-05
[cconv]: 1.933e-05
[loop_unroll]: 0.00048504
[jit_opt_after_cconv]: 0.0001483, [1]
    [Cycle 1]: 0.00014162, [11]
        [c_1]: 2.195e-05
        [parameter_eliminate]: 3.11001e-06
        [updatestate_depend_eliminate]: 6.46e-06
        [updatestate_assign_eliminate]: 2.72001e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 1.982e-05
        [call_graph_tuple_transform]: 2.023e-05
        [tuple_list_get_item_eliminator]: 5.57999e-06
        [none_parameter_eliminate]: 9.20001e-07
        [renormalize]: 5.79981e-07
        [switch_simplify]: 5.40001e-06
[remove_dup_value]: 9.74e-06
[partial_unused_args_eliminate]: 1.59e-06
[environ_conv]: 4.23001e-06
[add_recomputation]: 4.004e-05
[cse_after_recomputation]: 2.271e-05, [1]
    [Cycle 1]: 1.738e-05, [1]
        [cse]: 1.107e-05
[auto_monad_reorder]: 1.065e-05
[get_jit_bprop_graph]: 1.55999e-06
[rewriter_after_jit_bprop_graph]: 6.66e-06
[opt_after_jit_grad]: 0.00053306
[symbol_engine_optimizer]: 8.326e-05, [1]
    [Cycle 1]: 7.681e-05, [6]
        [build]: 3.86999e-06
        [elim_shapecalc]: 7.63999e-06
        [elim_not_effective]: 1.595e-05
        [opt_reshape]: 6.58e-06
        [fold_const_symbol]: 1.012e-05
        [renormalize]: 6.19999e-07
[validate]: 3.227e-05
Sums
    bootstrap                            :     0.000403s :  2.89%
    type_inference                       :     0.009265s : 66.49%
    event_method                         :     0.000023s :  0.16%
    auto_monad                           :     0.000086s :  0.62%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000062s :  0.44%
    rewriter_before_opt_a                :     0.000089s :  0.64%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000135s :  0.97%
    jit_opt_a.loop_unroll                :     0.000041s :  0.29%
    jit_opt_a.a_1                        :     0.000818s :  5.87%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000798s :  5.72%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000028s :  0.20%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.08%
    rewriter_after_opt_a                 :     0.000028s :  0.20%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000582s :  4.18%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.12%
    cconv                                :     0.000019s :  0.14%
    loop_unroll                          :     0.000485s :  3.48%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000010s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000040s :  0.29%
    cse_after_recomputation.cse          :     0.000011s :  0.08%
    auto_monad_reorder                   :     0.000011s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.05%
    opt_after_jit_grad                   :     0.000533s :  3.83%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.23%


Time group info:
------[substitution.]   0.000296    26
 0.60% :     0.000002s :      2: substitution.elim_not_effective
 0.33% :     0.000001s :      2: substitution.fold_const_symbol
 1.63% :     0.000005s :      3: substitution.graph_param_transform
87.89% :     0.000260s :      6: substitution.inline
 1.06% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.41% :     0.000004s :      4: substitution.remove_not_recompute_node
 0.91% :     0.000003s :      2: substitution.replace_old_param
 3.16% :     0.000009s :      1: substitution.switch_simplify
 3.00% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009190     2
81.41% :     0.007482s :      1: type_inference.infer
18.59% :     0.001708s :      1: type_inference.specialize
------[replace.]   0.000150     9
46.80% :     0.000070s :      6: replace.inline
28.61% :     0.000043s :      1: replace.switch_simplify
24.59% :     0.000037s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000272     9
94.03% :     0.000256s :      6: match.inline
 3.15% :     0.000009s :      1: match.switch_simplify
 2.82% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   801
 1.41% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     12: predicate.addn_check_dump
 1.90% :     0.000003s :     12: predicate.addn_zero_filter
 2.01% :     0.000003s :     12: predicate.arithmetic_simplify
 1.36% :     0.000002s :     12: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     12: predicate.compare_switch_simplify
 1.48% :     0.000002s :     12: predicate.depend_value_elim
 1.17% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.10% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     12: predicate.environ_get_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 4.73% :     0.000007s :     26: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000002s :      6: predicate.less_batch_normalization
 1.93% :     0.000003s :     14: predicate.list_to_tuple_eliminator_
 1.82% :     0.000003s :     17: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.45% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.05% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.33% :     0.000002s :     12: predicate.merge_addn
 1.12% :     0.000002s :     12: predicate.minmaximum_grad
 1.69% :     0.000002s :      3: predicate.mutable_eliminate
 0.67% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     17: predicate.partial_eliminate
 1.19% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.77% :     0.000002s :     12: predicate.reduce_eliminate
 1.43% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.40% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     20: predicate.replace_applicator
 0.43% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     12: predicate.reshape_eliminate
 1.46% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     12: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.78% :     0.000001s :      6: predicate.special_op_eliminate
 0.81% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.59% :     0.000004s :     20: predicate.switch_defer_inline
 2.23% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.06% :     0.000011s :     65: predicate.switch_simplify
 1.30% :     0.000002s :     12: predicate.tile_eliminate
 1.30% :     0.000002s :     12: predicate.transpose_eliminate
 1.64% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.35% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.64% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 2.11% :     0.000003s :     12: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001376    15
45.93% :     0.000632s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 9.13% :     0.000126s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.93% :     0.000618s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020807    72
 0.21% :     0.000044s :      1: add_recomputation
 0.43% :     0.000090s :      1: auto_monad
 0.06% :     0.000013s :      1: auto_monad_reorder
 2.02% :     0.000421s :      1: bootstrap
 0.11% :     0.000022s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.13% :     0.000027s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000010s :      1: graph_reusing
31.86% :     0.006628s :      1: jit_opt_a
 0.73% :     0.000151s :      1: jit_opt_after_cconv
 0.27% :     0.000057s :      1: jit_opt_b
 2.37% :     0.000494s :      1: loop_unroll
 2.85% :     0.000593s :      1: mutable_eliminate
 5.31% :     0.001105s :     26: opt.transform.jit_opt_a
 0.24% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000028s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.61% :     0.000544s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000064s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000013s :      1: remove_dup_value
 2.20% :     0.000458s :      1: renormalize.infer
 1.60% :     0.000333s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000031s :      1: rewriter_after_opt_a
 0.44% :     0.000092s :      1: rewriter_before_opt_a
 0.41% :     0.000086s :      1: symbol_engine_optimizer
44.60% :     0.009280s :      1: type_inference


TotalTime = 0.0204836, [33]
[bootstrap]: 0.00028588
[type_inference]: 0.00461562
[event_method]: 1.666e-05
[auto_monad]: 5.02e-05
[graph_reusing]: 3.83999e-06
[pre_auto_parallel]: 2.22999e-06
[py_interpret_to_execute]: 3.404e-05
[rewriter_before_opt_a]: 8.823e-05
[expand_dump_flag]: 2.44999e-06
[jit_opt_a]: 0.00550993, [2]
    [Cycle 1]: 0.00177927, [27]
        [switch_simplify]: 5.806e-05
        [loop_unroll]: 3.322e-05
        [a_1]: 0.00057507
        [with_stream_mark]: 2.193e-05
        [recompute_prepare]: 7.32002e-06
        [updatestate_depend_eliminate]: 3.54002e-06
        [updatestate_assign_eliminate]: 3.12002e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 1.77999e-06
        [specialize_transform]: 6.02999e-06
        [updatestate_useless_node_eliminater]: 5.21002e-06
        [accelerated_algorithm]: 6.28e-06
        [meta_shard_fg_expand]: 2.06998e-06
        [get_grad_eliminate_]: 5.71e-06
        [merge_forward]: 3.97e-06
        [cell_reuse_recompute_pass]: 1.03001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.336e-05
        [j_node_and_user_rematch]: 8.65001e-06
        [meta_fg_expand]: 2.35002e-06
        [replace_old_param]: 8.90999e-06
        [inline_without_move]: 5.67999e-06
        [renormalize]: 0.00076005
        [add_forward_monad_depend]: 5.51998e-06
        [auto_monad_grad]: 2.63003e-06
        [auto_monad_eliminator]: 1.431e-05
        [cse]: 3.112e-05
        [replace_applicator]: 1.142e-05
    [Cycle 2]: 0.00033627, [27]
        [switch_simplify]: 5.97999e-06
        [loop_unroll]: 4.82e-06
        [a_1]: 9.42e-05
        [with_stream_mark]: 9.00001e-06
        [recompute_prepare]: 4.94e-06
        [updatestate_depend_eliminate]: 2.59999e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 5.34e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 5.02e-06
        [meta_shard_fg_expand]: 1.09003e-06
        [get_grad_eliminate_]: 4.60001e-06
        [merge_forward]: 2.56e-06
        [cell_reuse_recompute_pass]: 1.76998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.295e-05
        [j_node_and_user_rematch]: 8.37e-06
        [meta_fg_expand]: 1.64998e-06
        [replace_old_param]: 7.3e-06
        [inline_without_move]: 4.99e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.52999e-06
        [auto_monad_grad]: 9.80013e-07
        [auto_monad_eliminator]: 5.29e-06
        [cse]: 1.142e-05
        [replace_applicator]: 5.25999e-06
[py_interpret_to_execute_after_opt_a]: 1.199e-05
[rewriter_after_opt_a]: 3.66e-05
[convert_after_rewriter]: 6.74999e-06
[order_py_execute_after_rewriter]: 5.31998e-06
[mutable_eliminate]: 0.00059759
[jit_opt_b]: 5.253e-05, [1]
    [Cycle 1]: 4.456e-05, [2]
        [frontend_op_eliminate]: 1.772e-05
        [inline_after_opt_a]: 1.523e-05
[cconv]: 2.983e-05
[loop_unroll]: 0.00048182
[jit_opt_after_cconv]: 0.00015755, [1]
    [Cycle 1]: 0.00015088, [11]
        [c_1]: 2.172e-05
        [parameter_eliminate]: 3.93001e-06
        [updatestate_depend_eliminate]: 7.67998e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.64001e-06
        [cse]: 2.62e-05
        [call_graph_tuple_transform]: 1.991e-05
        [tuple_list_get_item_eliminator]: 6.04999e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 1.00001e-06
        [switch_simplify]: 5.76998e-06
[remove_dup_value]: 1.597e-05
[partial_unused_args_eliminate]: 3.07002e-06
[environ_conv]: 5.64e-06
[add_recomputation]: 5.435e-05
[cse_after_recomputation]: 2.244e-05, [1]
    [Cycle 1]: 1.64e-05, [1]
        [cse]: 1.005e-05
[auto_monad_reorder]: 1.637e-05
[get_jit_bprop_graph]: 2.61e-06
[rewriter_after_jit_bprop_graph]: 1.031e-05
[opt_after_jit_grad]: 0.00057316
[symbol_engine_optimizer]: 8.627e-05, [1]
    [Cycle 1]: 7.769e-05, [6]
        [build]: 5.76e-06
        [elim_shapecalc]: 8.07e-06
        [elim_not_effective]: 1.599e-05
        [opt_reshape]: 8.38001e-06
        [fold_const_symbol]: 9.54e-06
        [renormalize]: 1.28002e-06
[validate]: 4.412e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.00742388
[execute]: 1.005e-05
Sums
    bootstrap                            :     0.000286s :  1.74%
    type_inference                       :     0.004616s : 28.09%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000050s :  0.31%
    graph_reusing                        :     0.000004s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000034s :  0.21%
    rewriter_before_opt_a                :     0.000088s :  0.54%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000064s :  0.39%
    jit_opt_a.loop_unroll                :     0.000038s :  0.23%
    jit_opt_a.a_1                        :     0.000669s :  4.07%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000760s :  4.63%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000043s :  0.26%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000598s :  3.64%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000030s :  0.18%
    loop_unroll                          :     0.000482s :  2.93%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000026s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000016s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000054s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.06%
    opt_after_jit_grad                   :     0.000573s :  3.49%
    symbol_engine_optimizer.build        :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007424s : 45.17%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000208    23
 1.26% :     0.000003s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.56% :     0.000005s :      3: substitution.graph_param_transform
84.09% :     0.000175s :      4: substitution.inline
 1.39% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.13% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.90% :     0.000004s :      2: substitution.replace_old_param
 6.08% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004559     2
82.09% :     0.003742s :      1: type_inference.infer
17.91% :     0.000816s :      1: type_inference.specialize
------[replace.]   0.000071     6
65.32% :     0.000046s :      4: replace.inline
34.68% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000184     6
93.74% :     0.000172s :      4: match.inline
 6.26% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   691
 1.24% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.67% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.40% :     0.000002s :     10: predicate.addn_check_dump
 2.28% :     0.000003s :     10: predicate.addn_zero_filter
 2.19% :     0.000003s :     10: predicate.arithmetic_simplify
 1.52% :     0.000002s :     10: predicate.cast_eliminate
 0.56% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.22% :     0.000002s :     10: predicate.depend_value_elim
 1.46% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.50% :     0.000001s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 4.69% :     0.000006s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.78% :     0.000002s :      6: predicate.less_batch_normalization
 1.73% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.88% :     0.000002s :     15: predicate.load_eliminater
 1.72% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.85% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 2.39% :     0.000003s :      3: predicate.mutable_eliminate
 0.69% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.59% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.44% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.32% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.13% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.37% :     0.000010s :     53: predicate.switch_simplify
 1.02% :     0.000001s :     10: predicate.tile_eliminate
 1.43% :     0.000002s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.62% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.52% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.36% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.89% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000582    12
47.42% :     0.000276s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.54% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.04% :     0.000262s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022245    76
 0.26% :     0.000057s :      1: add_recomputation
 0.24% :     0.000054s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.37% :     0.000304s :      1: bootstrap
 0.15% :     0.000033s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
24.79% :     0.005514s :      1: jit_opt_a
 0.72% :     0.000161s :      1: jit_opt_after_cconv
 0.25% :     0.000056s :      1: jit_opt_b
 2.21% :     0.000491s :      1: loop_unroll
 2.73% :     0.000608s :      1: mutable_eliminate
 3.98% :     0.000885s :     26: opt.transform.jit_opt_a
 0.22% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000038s :      4: opt.transform.symbol_engine_opt
 2.63% :     0.000585s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 1.91% :     0.000426s :      1: renormalize.infer
 1.47% :     0.000326s :      1: renormalize.specialize
 0.06% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000040s :      1: rewriter_after_opt_a
 0.41% :     0.000092s :      1: rewriter_before_opt_a
 0.40% :     0.000089s :      1: symbol_engine_optimizer
33.46% :     0.007444s :      1: task_emit
20.81% :     0.004629s :      1: type_inference
 0.30% :     0.000067s :      1: validate


TotalTime = 0.0138659, [30]
[bootstrap]: 0.00029314
[type_inference]: 0.00642784
[event_method]: 2.063e-05
[auto_monad]: 6.234e-05
[graph_reusing]: 4.68001e-06
[pre_auto_parallel]: 2.26998e-06
[py_interpret_to_execute]: 4.633e-05
[rewriter_before_opt_a]: 7.699e-05
[expand_dump_flag]: 2.06e-06
[jit_opt_a]: 0.00494308, [2]
    [Cycle 1]: 0.00163984, [27]
        [switch_simplify]: 9.452e-05
        [loop_unroll]: 3.467e-05
        [a_1]: 0.00064295
        [with_stream_mark]: 1.149e-05
        [recompute_prepare]: 7.21001e-06
        [updatestate_depend_eliminate]: 2.90998e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 6.04001e-06
        [updatestate_useless_node_eliminater]: 4.94003e-06
        [accelerated_algorithm]: 5.15001e-06
        [meta_shard_fg_expand]: 1.40001e-06
        [get_grad_eliminate_]: 5.09998e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.255e-05
        [j_node_and_user_rematch]: 8.42e-06
        [meta_fg_expand]: 1.80001e-06
        [replace_old_param]: 8.13001e-06
        [inline_without_move]: 5.07e-06
        [renormalize]: 0.00057364
        [add_forward_monad_depend]: 4.58999e-06
        [auto_monad_grad]: 1.51002e-06
        [auto_monad_eliminator]: 9.27999e-06
        [cse]: 1.556e-05
        [replace_applicator]: 1.095e-05
    [Cycle 2]: 0.00031127, [27]
        [switch_simplify]: 5.74999e-06
        [loop_unroll]: 5.03002e-06
        [a_1]: 8.949e-05
        [with_stream_mark]: 8.33999e-06
        [recompute_prepare]: 5.20001e-06
        [updatestate_depend_eliminate]: 2.64001e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.86998e-06
        [parameter_eliminate]: 9.49978e-07
        [specialize_transform]: 5.05999e-06
        [updatestate_useless_node_eliminater]: 4.97999e-06
        [accelerated_algorithm]: 4.79e-06
        [meta_shard_fg_expand]: 1.13001e-06
        [get_grad_eliminate_]: 4.57e-06
        [merge_forward]: 2.54999e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.198e-05
        [j_node_and_user_rematch]: 7.60998e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 7.06001e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.29998e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 4.76002e-06
        [cse]: 9.12001e-06
        [replace_applicator]: 4.84998e-06
[py_interpret_to_execute_after_opt_a]: 8.85999e-06
[rewriter_after_opt_a]: 2.547e-05
[convert_after_rewriter]: 5.13002e-06
[order_py_execute_after_rewriter]: 4.47e-06
[mutable_eliminate]: 0.00050147
[jit_opt_b]: 5.161e-05, [1]
    [Cycle 1]: 4.442e-05, [2]
        [frontend_op_eliminate]: 1.765e-05
        [inline_after_opt_a]: 1.494e-05
[cconv]: 1.711e-05
[loop_unroll]: 0.00042536
[jit_opt_after_cconv]: 0.00013269, [1]
    [Cycle 1]: 0.00012648, [11]
        [c_1]: 2.053e-05
        [parameter_eliminate]: 2.42001e-06
        [updatestate_depend_eliminate]: 5.21998e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 1.627e-05
        [call_graph_tuple_transform]: 1.647e-05
        [tuple_list_get_item_eliminator]: 5.44998e-06
        [none_parameter_eliminate]: 7.49977e-07
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.30999e-06
[remove_dup_value]: 8.67e-06
[partial_unused_args_eliminate]: 1.70001e-06
[environ_conv]: 3.8e-06
[add_recomputation]: 3.464e-05
[cse_after_recomputation]: 2.107e-05, [1]
    [Cycle 1]: 1.554e-05, [1]
        [cse]: 9.59999e-06
[auto_monad_reorder]: 1.045e-05
[get_jit_bprop_graph]: 1.47999e-06
[rewriter_after_jit_bprop_graph]: 4.52998e-06
[opt_after_jit_grad]: 0.00046563
[symbol_engine_optimizer]: 7.239e-05, [1]
    [Cycle 1]: 6.636e-05, [6]
        [build]: 3.47002e-06
        [elim_shapecalc]: 7.39002e-06
        [elim_not_effective]: 1.227e-05
        [opt_reshape]: 5.54e-06
        [fold_const_symbol]: 9.31e-06
        [renormalize]: 5.60016e-07
[validate]: 3.026e-05
Sums
    bootstrap                            :     0.000293s :  2.84%
    type_inference                       :     0.006428s : 62.29%
    event_method                         :     0.000021s :  0.20%
    auto_monad                           :     0.000062s :  0.60%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000046s :  0.45%
    rewriter_before_opt_a                :     0.000077s :  0.75%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000100s :  0.97%
    jit_opt_a.loop_unroll                :     0.000040s :  0.38%
    jit_opt_a.a_1                        :     0.000732s :  7.10%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000574s :  5.56%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.14%
    jit_opt_a.cse                        :     0.000025s :  0.24%
    jit_opt_a.replace_applicator         :     0.000016s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.09%
    rewriter_after_opt_a                 :     0.000025s :  0.25%
    convert_after_rewriter               :     0.000005s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000501s :  4.86%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.14%
    cconv                                :     0.000017s :  0.17%
    loop_unroll                          :     0.000425s :  4.12%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000009s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000035s :  0.34%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000010s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000466s :  4.51%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000030s :  0.29%


Time group info:
------[substitution.]   0.000229    26
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000001s :      2: substitution.fold_const_symbol
 1.63% :     0.000004s :      3: substitution.graph_param_transform
85.43% :     0.000196s :      6: substitution.inline
 1.12% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.11% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.38% :     0.000003s :      2: substitution.replace_old_param
 3.43% :     0.000008s :      1: substitution.switch_simplify
 3.66% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006378     2
84.48% :     0.005388s :      1: type_inference.infer
15.52% :     0.000990s :      1: type_inference.specialize
------[replace.]   0.000118     9
55.82% :     0.000066s :      6: replace.inline
19.37% :     0.000023s :      1: replace.switch_simplify
24.81% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000207     9
93.05% :     0.000193s :      6: match.inline
 3.42% :     0.000007s :      1: match.switch_simplify
 3.53% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   801
 1.55% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.18% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.26% :     0.000002s :     12: predicate.addn_check_dump
 1.41% :     0.000002s :     12: predicate.addn_zero_filter
 1.86% :     0.000002s :     12: predicate.arithmetic_simplify
 1.37% :     0.000002s :     12: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.21% :     0.000002s :     12: predicate.compare_switch_simplify
 1.43% :     0.000002s :     12: predicate.depend_value_elim
 1.25% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.31% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     12: predicate.environ_get_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.15% :     0.000007s :     26: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     17: predicate.load_eliminater
 1.27% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.35% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000002s :     12: predicate.merge_addn
 1.16% :     0.000001s :     12: predicate.minmaximum_grad
 1.49% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     17: predicate.partial_eliminate
 1.22% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     12: predicate.reduce_eliminate
 1.85% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     20: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     12: predicate.reshape_eliminate
 1.25% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.73% :     0.000001s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.70% :     0.000003s :     20: predicate.switch_defer_inline
 2.47% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.22% :     0.000011s :     65: predicate.switch_simplify
 1.29% :     0.000002s :     12: predicate.tile_eliminate
 1.30% :     0.000002s :     12: predicate.transpose_eliminate
 1.49% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.41% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.34% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.83% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000002s :     12: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000770    15
50.83% :     0.000391s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.11% :     0.000078s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.06% :     0.000301s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015494    72
 0.24% :     0.000038s :      1: add_recomputation
 0.43% :     0.000066s :      1: auto_monad
 0.09% :     0.000013s :      1: auto_monad_reorder
 1.98% :     0.000307s :      1: bootstrap
 0.13% :     0.000020s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
31.92% :     0.004946s :      1: jit_opt_a
 0.87% :     0.000136s :      1: jit_opt_after_cconv
 0.35% :     0.000054s :      1: jit_opt_b
 2.79% :     0.000433s :      1: loop_unroll
 3.30% :     0.000511s :      1: mutable_eliminate
 6.34% :     0.000982s :     26: opt.transform.jit_opt_a
 0.29% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.14% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.05% :     0.000473s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.32% :     0.000049s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000011s :      1: remove_dup_value
 1.99% :     0.000308s :      1: renormalize.infer
 1.67% :     0.000259s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000028s :      1: rewriter_after_opt_a
 0.52% :     0.000080s :      1: rewriter_before_opt_a
 0.49% :     0.000075s :      1: symbol_engine_optimizer
41.57% :     0.006440s :      1: type_inference


TotalTime = 0.0210955, [33]
[bootstrap]: 0.00031154
[type_inference]: 0.0047038
[event_method]: 1.768e-05
[auto_monad]: 6.665e-05
[graph_reusing]: 6.07999e-06
[pre_auto_parallel]: 2.69999e-06
[py_interpret_to_execute]: 3.445e-05
[rewriter_before_opt_a]: 7.747e-05
[expand_dump_flag]: 3.09001e-06
[jit_opt_a]: 0.00520297, [2]
    [Cycle 1]: 0.00150326, [27]
        [switch_simplify]: 5.407e-05
        [loop_unroll]: 3.068e-05
        [a_1]: 0.00051639
        [with_stream_mark]: 1.546e-05
        [recompute_prepare]: 7.33e-06
        [updatestate_depend_eliminate]: 3.33e-06
        [updatestate_assign_eliminate]: 3.37002e-06
        [updatestate_loads_eliminate]: 2.69001e-06
        [parameter_eliminate]: 1.97001e-06
        [specialize_transform]: 5.95002e-06
        [updatestate_useless_node_eliminater]: 5.02999e-06
        [accelerated_algorithm]: 5.64998e-06
        [meta_shard_fg_expand]: 1.86998e-06
        [get_grad_eliminate_]: 5.40999e-06
        [merge_forward]: 3.59002e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.431e-05
        [j_node_and_user_rematch]: 9.28002e-06
        [meta_fg_expand]: 2.21e-06
        [replace_old_param]: 8.35999e-06
        [inline_without_move]: 5.15001e-06
        [renormalize]: 0.00053529
        [add_forward_monad_depend]: 5.71e-06
        [auto_monad_grad]: 2.14999e-06
        [auto_monad_eliminator]: 1.371e-05
        [cse]: 2.949e-05
        [replace_applicator]: 4.13e-05
    [Cycle 2]: 0.00034728, [27]
        [switch_simplify]: 6.73998e-06
        [loop_unroll]: 5.40001e-06
        [a_1]: 9.456e-05
        [with_stream_mark]: 9.99999e-06
        [recompute_prepare]: 5.62001e-06
        [updatestate_depend_eliminate]: 2.74001e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 1.173e-05
        [updatestate_useless_node_eliminater]: 5.44e-06
        [accelerated_algorithm]: 5.22999e-06
        [meta_shard_fg_expand]: 1.31002e-06
        [get_grad_eliminate_]: 4.81002e-06
        [merge_forward]: 2.82002e-06
        [cell_reuse_recompute_pass]: 1.69e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.417e-05
        [j_node_and_user_rematch]: 8.74e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 8.45999e-06
        [inline_without_move]: 4.80001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.57999e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 5.69999e-06
        [cse]: 1.161e-05
        [replace_applicator]: 5.32001e-06
[py_interpret_to_execute_after_opt_a]: 9.20999e-06
[rewriter_after_opt_a]: 3.282e-05
[convert_after_rewriter]: 7.14001e-06
[order_py_execute_after_rewriter]: 5.71e-06
[mutable_eliminate]: 0.00049375
[jit_opt_b]: 4.904e-05, [1]
    [Cycle 1]: 4.23e-05, [2]
        [frontend_op_eliminate]: 1.631e-05
        [inline_after_opt_a]: 1.431e-05
[cconv]: 2.519e-05
[loop_unroll]: 0.00042184
[jit_opt_after_cconv]: 0.0001389, [1]
    [Cycle 1]: 0.00013244, [11]
        [c_1]: 2.118e-05
        [parameter_eliminate]: 2.53e-06
        [updatestate_depend_eliminate]: 5.30999e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [cse]: 1.671e-05
        [call_graph_tuple_transform]: 1.808e-05
        [tuple_list_get_item_eliminator]: 5.40001e-06
        [none_parameter_eliminate]: 1.56998e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 5.39998e-06
[remove_dup_value]: 1.451e-05
[partial_unused_args_eliminate]: 2.86e-06
[environ_conv]: 4.97e-06
[add_recomputation]: 4.905e-05
[cse_after_recomputation]: 2.047e-05, [1]
    [Cycle 1]: 1.479e-05, [1]
        [cse]: 8.89e-06
[auto_monad_reorder]: 1.523e-05
[get_jit_bprop_graph]: 1.65001e-06
[rewriter_after_jit_bprop_graph]: 5.40001e-06
[opt_after_jit_grad]: 0.00045917
[symbol_engine_optimizer]: 7.629e-05, [1]
    [Cycle 1]: 6.92e-05, [6]
        [build]: 4.74e-06
        [elim_shapecalc]: 7.78001e-06
        [elim_not_effective]: 1.333e-05
        [opt_reshape]: 5.66e-06
        [fold_const_symbol]: 9.04998e-06
        [renormalize]: 4.7998e-07
[validate]: 3.598e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.0085781
[execute]: 9.09e-06
Sums
    bootstrap                            :     0.000312s :  1.82%
    type_inference                       :     0.004704s : 27.48%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000067s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.20%
    rewriter_before_opt_a                :     0.000077s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.36%
    jit_opt_a.loop_unroll                :     0.000036s :  0.21%
    jit_opt_a.a_1                        :     0.000611s :  3.57%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000018s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000535s :  3.13%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.11%
    jit_opt_a.cse                        :     0.000041s :  0.24%
    jit_opt_a.replace_applicator         :     0.000047s :  0.27%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.05%
    rewriter_after_opt_a                 :     0.000033s :  0.19%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000494s :  2.88%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.08%
    cconv                                :     0.000025s :  0.15%
    loop_unroll                          :     0.000422s :  2.46%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000459s :  2.68%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008578s : 50.12%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000181    23
 1.12% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.68% :     0.000005s :      3: substitution.graph_param_transform
82.62% :     0.000150s :      4: substitution.inline
 1.99% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.73% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000003s :      2: substitution.replace_old_param
 6.27% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004638     2
82.10% :     0.003808s :      1: type_inference.infer
17.90% :     0.000830s :      1: type_inference.specialize
------[replace.]   0.000069     6
66.96% :     0.000046s :      4: replace.inline
33.04% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000157     6
93.48% :     0.000147s :      4: match.inline
 6.52% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.38% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 1.79% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.42% :     0.000002s :     10: predicate.depend_value_elim
 1.20% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.53% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.22% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.88% :     0.000002s :     15: predicate.load_eliminater
 1.61% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.57% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.32% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     10: predicate.same_eliminate
 0.67% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000001s :      6: predicate.special_op_eliminate
 1.12% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 2.24% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.98% :     0.000010s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.34% :     0.000002s :     10: predicate.transpose_eliminate
 1.57% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.64% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000569    12
52.95% :     0.000301s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.91% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.13% :     0.000240s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022588    76
 0.23% :     0.000052s :      1: add_recomputation
 0.31% :     0.000070s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.47% :     0.000333s :      1: bootstrap
 0.12% :     0.000028s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
23.05% :     0.005206s :      1: jit_opt_a
 0.63% :     0.000142s :      1: jit_opt_after_cconv
 0.23% :     0.000052s :      1: jit_opt_b
 1.90% :     0.000430s :      1: loop_unroll
 2.22% :     0.000502s :      1: mutable_eliminate
 3.81% :     0.000860s :     26: opt.transform.jit_opt_a
 0.21% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.07% :     0.000467s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000038s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.20% :     0.000271s :      1: renormalize.infer
 1.13% :     0.000256s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000036s :      1: rewriter_after_opt_a
 0.36% :     0.000081s :      1: rewriter_before_opt_a
 0.35% :     0.000079s :      1: symbol_engine_optimizer
38.06% :     0.008596s :      1: task_emit
20.88% :     0.004717s :      1: type_inference
 0.25% :     0.000056s :      1: validate


TotalTime = 0.0212919, [33]
[bootstrap]: 0.00034254
[type_inference]: 0.00537184
[event_method]: 1.914e-05
[auto_monad]: 7.143e-05
[graph_reusing]: 6.23998e-06
[pre_auto_parallel]: 2.92002e-06
[py_interpret_to_execute]: 4.356e-05
[rewriter_before_opt_a]: 8.879e-05
[expand_dump_flag]: 3.81999e-06
[jit_opt_a]: 0.00639998, [2]
    [Cycle 1]: 0.00182797, [27]
        [switch_simplify]: 5.984e-05
        [loop_unroll]: 3.134e-05
        [a_1]: 0.00058894
        [with_stream_mark]: 2.182e-05
        [recompute_prepare]: 7.23e-06
        [updatestate_depend_eliminate]: 3.86999e-06
        [updatestate_assign_eliminate]: 3.16999e-06
        [updatestate_loads_eliminate]: 2.98e-06
        [parameter_eliminate]: 2.23998e-06
        [specialize_transform]: 6.12999e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 5.38002e-06
        [meta_shard_fg_expand]: 2.04e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 3.63999e-06
        [cell_reuse_recompute_pass]: 1.00999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 7.009e-05
        [j_node_and_user_rematch]: 9.71998e-06
        [meta_fg_expand]: 2.93e-06
        [replace_old_param]: 9.81998e-06
        [inline_without_move]: 5.27001e-06
        [renormalize]: 0.00073963
        [add_forward_monad_depend]: 5.90002e-06
        [auto_monad_grad]: 2.43998e-06
        [auto_monad_eliminator]: 1.421e-05
        [cse]: 2.975e-05
        [replace_applicator]: 1.266e-05
    [Cycle 2]: 0.00032941, [27]
        [switch_simplify]: 5.59e-06
        [loop_unroll]: 4.82e-06
        [a_1]: 9.148e-05
        [with_stream_mark]: 1.019e-05
        [recompute_prepare]: 5.03002e-06
        [updatestate_depend_eliminate]: 2.66e-06
        [updatestate_assign_eliminate]: 2.14999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.20999e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 4.94e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 5.06002e-06
        [merge_forward]: 2.96999e-06
        [cell_reuse_recompute_pass]: 1.33002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.472e-05
        [j_node_and_user_rematch]: 7.71999e-06
        [meta_fg_expand]: 1.76e-06
        [replace_old_param]: 7.3e-06
        [inline_without_move]: 4.97e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 5.02999e-06
        [cse]: 1.012e-05
        [replace_applicator]: 5.04998e-06
[py_interpret_to_execute_after_opt_a]: 1.224e-05
[rewriter_after_opt_a]: 3.522e-05
[convert_after_rewriter]: 6.86999e-06
[order_py_execute_after_rewriter]: 1.819e-05
[mutable_eliminate]: 0.00059152
[jit_opt_b]: 5.025e-05, [1]
    [Cycle 1]: 4.298e-05, [2]
        [frontend_op_eliminate]: 1.699e-05
        [inline_after_opt_a]: 1.48e-05
[cconv]: 2.8e-05
[loop_unroll]: 0.00046689
[jit_opt_after_cconv]: 0.00014473, [1]
    [Cycle 1]: 0.00013774, [11]
        [c_1]: 2.145e-05
        [parameter_eliminate]: 2.63e-06
        [updatestate_depend_eliminate]: 6.10002e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.834e-05
        [call_graph_tuple_transform]: 1.841e-05
        [tuple_list_get_item_eliminator]: 6.19001e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 5.90022e-07
        [switch_simplify]: 6.09999e-06
[remove_dup_value]: 1.519e-05
[partial_unused_args_eliminate]: 2.76e-06
[environ_conv]: 5.58002e-06
[add_recomputation]: 5.131e-05
[cse_after_recomputation]: 2.108e-05, [1]
    [Cycle 1]: 1.54e-05, [1]
        [cse]: 9.51e-06
[auto_monad_reorder]: 1.484e-05
[get_jit_bprop_graph]: 2.03997e-06
[rewriter_after_jit_bprop_graph]: 5.04e-06
[opt_after_jit_grad]: 0.00047367
[symbol_engine_optimizer]: 8.181e-05, [1]
    [Cycle 1]: 7.472e-05, [6]
        [build]: 4.93001e-06
        [elim_shapecalc]: 8.42e-06
        [elim_not_effective]: 1.49e-05
        [opt_reshape]: 6.36e-06
        [fold_const_symbol]: 9.12001e-06
        [renormalize]: 4.00003e-07
[validate]: 3.789e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00665163
[execute]: 7e-06
Sums
    bootstrap                            :     0.000343s :  2.09%
    type_inference                       :     0.005372s : 32.73%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000071s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000044s :  0.27%
    rewriter_before_opt_a                :     0.000089s :  0.54%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.40%
    jit_opt_a.loop_unroll                :     0.000036s :  0.22%
    jit_opt_a.a_1                        :     0.000680s :  4.15%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000085s :  0.52%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000740s :  4.51%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000040s :  0.24%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000018s :  0.11%
    mutable_eliminate                    :     0.000592s :  3.60%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000028s :  0.17%
    loop_unroll                          :     0.000467s :  2.85%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000051s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000474s :  2.89%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006652s : 40.53%
    execute                              :     0.000007s :  0.04%


Time group info:
------[substitution.]   0.000300    23
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000002s :      2: substitution.fold_const_symbol
 1.81% :     0.000005s :      3: substitution.graph_param_transform
69.75% :     0.000209s :      4: substitution.inline
 1.09% :     0.000003s :      4: substitution.j_node_and_user_rematch
20.37% :     0.000061s :      4: substitution.remove_not_recompute_node
 1.60% :     0.000005s :      2: substitution.replace_old_param
 4.10% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005288     2
81.54% :     0.004311s :      1: type_inference.infer
18.46% :     0.000976s :      1: type_inference.specialize
------[replace.]   0.000077     6
68.45% :     0.000052s :      4: replace.inline
31.55% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000217     6
94.88% :     0.000206s :      4: match.inline
 5.12% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.48% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.07% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 1.79% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.48% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.94% :     0.000007s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.44% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.80% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.04% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.02% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.79% :     0.000002s :      3: predicate.mutable_eliminate
 0.57% :     0.000001s :      3: predicate.opt_reshape
 2.38% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.56% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.43% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.65% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.89% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.99% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000674    12
50.19% :     0.000338s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.92% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.89% :     0.000282s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023093    76
 0.24% :     0.000054s :      1: add_recomputation
 0.33% :     0.000076s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.58% :     0.000366s :      1: bootstrap
 0.13% :     0.000031s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
27.73% :     0.006404s :      1: jit_opt_a
 0.64% :     0.000148s :      1: jit_opt_after_cconv
 0.23% :     0.000053s :      1: jit_opt_b
 2.06% :     0.000476s :      1: loop_unroll
 2.60% :     0.000601s :      1: mutable_eliminate
 4.13% :     0.000955s :     26: opt.transform.jit_opt_a
 0.21% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.09% :     0.000482s :      1: opt_after_jit_grad
 0.09% :     0.000021s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000048s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.80% :     0.000415s :      1: renormalize.infer
 1.37% :     0.000316s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000038s :      1: rewriter_after_opt_a
 0.40% :     0.000093s :      1: rewriter_before_opt_a
 0.36% :     0.000084s :      1: symbol_engine_optimizer
28.86% :     0.006664s :      1: task_emit
23.33% :     0.005388s :      1: type_inference
 0.25% :     0.000058s :      1: validate


TotalTime = 0.0143444, [30]
[bootstrap]: 0.00031214
[type_inference]: 0.00648287
[event_method]: 1.991e-05
[auto_monad]: 8.268e-05
[graph_reusing]: 6.36998e-06
[pre_auto_parallel]: 2.63003e-06
[py_interpret_to_execute]: 4.736e-05
[rewriter_before_opt_a]: 8.517e-05
[expand_dump_flag]: 3.38e-06
[jit_opt_a]: 0.00533847, [2]
    [Cycle 1]: 0.00178245, [27]
        [switch_simplify]: 0.00010965
        [loop_unroll]: 3.356e-05
        [a_1]: 0.00067147
        [with_stream_mark]: 1.619e-05
        [recompute_prepare]: 6.93e-06
        [updatestate_depend_eliminate]: 3.52002e-06
        [updatestate_assign_eliminate]: 2.89999e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [parameter_eliminate]: 2.24001e-06
        [specialize_transform]: 6.08002e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 5.41002e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 5.44998e-06
        [merge_forward]: 3.5e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.393e-05
        [j_node_and_user_rematch]: 8.64e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 8.17998e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 0.00064778
        [add_forward_monad_depend]: 4.75001e-06
        [auto_monad_grad]: 2.10002e-06
        [auto_monad_eliminator]: 1.328e-05
        [cse]: 2.469e-05
        [replace_applicator]: 1.114e-05
    [Cycle 2]: 0.00033863, [27]
        [switch_simplify]: 5.29998e-06
        [loop_unroll]: 5.00001e-06
        [a_1]: 8.813e-05
        [with_stream_mark]: 8.65001e-06
        [recompute_prepare]: 4.80001e-06
        [updatestate_depend_eliminate]: 2.63e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 5.20999e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 4.71002e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 7.62998e-06
        [merge_forward]: 2.76e-06
        [cell_reuse_recompute_pass]: 1.44998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.295e-05
        [j_node_and_user_rematch]: 8.71002e-06
        [meta_fg_expand]: 1.47999e-06
        [replace_old_param]: 7.46999e-06
        [inline_without_move]: 5.08002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 5.09e-06
        [cse]: 1.116e-05
        [replace_applicator]: 5.09e-06
[py_interpret_to_execute_after_opt_a]: 8.01001e-06
[rewriter_after_opt_a]: 3.159e-05
[convert_after_rewriter]: 6.73e-06
[order_py_execute_after_rewriter]: 5.04e-06
[mutable_eliminate]: 0.00047265
[jit_opt_b]: 4.769e-05, [1]
    [Cycle 1]: 4.105e-05, [2]
        [frontend_op_eliminate]: 1.548e-05
        [inline_after_opt_a]: 1.439e-05
[cconv]: 2.361e-05
[loop_unroll]: 0.00041405
[jit_opt_after_cconv]: 0.00014338, [1]
    [Cycle 1]: 0.00013713, [11]
        [c_1]: 2.034e-05
        [parameter_eliminate]: 2.29999e-06
        [updatestate_depend_eliminate]: 5.07999e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 2.405e-05
        [call_graph_tuple_transform]: 1.692e-05
        [tuple_list_get_item_eliminator]: 5.27001e-06
        [none_parameter_eliminate]: 1.86e-06
        [renormalize]: 5.39992e-07
        [switch_simplify]: 5.30999e-06
[remove_dup_value]: 1.282e-05
[partial_unused_args_eliminate]: 2.11998e-06
[environ_conv]: 5.00999e-06
[add_recomputation]: 4.587e-05
[cse_after_recomputation]: 1.92e-05, [1]
    [Cycle 1]: 1.393e-05, [1]
        [cse]: 8.22e-06
[auto_monad_reorder]: 1.499e-05
[get_jit_bprop_graph]: 1.39e-06
[rewriter_after_jit_bprop_graph]: 4.16001e-06
[opt_after_jit_grad]: 0.00044692
[symbol_engine_optimizer]: 7.201e-05, [1]
    [Cycle 1]: 6.519e-05, [6]
        [build]: 2.81e-06
        [elim_shapecalc]: 7.68999e-06
        [elim_not_effective]: 1.182e-05
        [opt_reshape]: 5.84e-06
        [fold_const_symbol]: 8.45001e-06
        [renormalize]: 4.50003e-07
[validate]: 2.983e-05
Sums
    bootstrap                            :     0.000312s :  2.96%
    type_inference                       :     0.006483s : 61.42%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000083s :  0.78%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000047s :  0.45%
    rewriter_before_opt_a                :     0.000085s :  0.81%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000115s :  1.09%
    jit_opt_a.loop_unroll                :     0.000039s :  0.37%
    jit_opt_a.a_1                        :     0.000760s :  7.20%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000013s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000648s :  6.14%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.17%
    jit_opt_a.cse                        :     0.000036s :  0.34%
    jit_opt_a.replace_applicator         :     0.000016s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.08%
    rewriter_after_opt_a                 :     0.000032s :  0.30%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000473s :  4.48%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.14%
    cconv                                :     0.000024s :  0.22%
    loop_unroll                          :     0.000414s :  3.92%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000046s :  0.43%
    cse_after_recomputation.cse          :     0.000008s :  0.08%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.04%
    opt_after_jit_grad                   :     0.000447s :  4.23%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000030s :  0.28%


Time group info:
------[substitution.]   0.000265    26
 0.71% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 1.80% :     0.000005s :      3: substitution.graph_param_transform
85.52% :     0.000226s :      6: substitution.inline
 1.10% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.84% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.24% :     0.000003s :      2: substitution.replace_old_param
 2.86% :     0.000008s :      1: substitution.switch_simplify
 4.46% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006423     2
84.16% :     0.005405s :      1: type_inference.infer
15.84% :     0.001018s :      1: type_inference.specialize
------[replace.]   0.000130     9
50.12% :     0.000065s :      6: replace.inline
25.85% :     0.000034s :      1: replace.switch_simplify
24.03% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000240     9
92.72% :     0.000223s :      6: match.inline
 2.83% :     0.000007s :      1: match.switch_simplify
 4.45% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   801
 1.49% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.20% :     0.000002s :     12: predicate.addn_check_dump
 1.54% :     0.000002s :     12: predicate.addn_zero_filter
 1.81% :     0.000002s :     12: predicate.arithmetic_simplify
 1.39% :     0.000002s :     12: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.22% :     0.000002s :     12: predicate.compare_switch_simplify
 1.43% :     0.000002s :     12: predicate.depend_value_elim
 1.16% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.71% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.76% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     12: predicate.environ_get_eliminate
 1.12% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.40% :     0.000007s :     26: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     17: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.27% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.08% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000002s :     12: predicate.merge_addn
 1.17% :     0.000002s :     12: predicate.minmaximum_grad
 1.40% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     17: predicate.partial_eliminate
 1.24% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.86% :     0.000002s :     12: predicate.reduce_eliminate
 1.65% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     20: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     12: predicate.reshape_eliminate
 1.51% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.81% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.73% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.66% :     0.000003s :     20: predicate.switch_defer_inline
 2.32% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.04% :     0.000011s :     65: predicate.switch_simplify
 1.29% :     0.000002s :     12: predicate.tile_eliminate
 1.24% :     0.000002s :     12: predicate.transpose_eliminate
 1.44% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.43% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000002s :     12: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000787    15
49.66% :     0.000391s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.35% :     0.000089s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.98% :     0.000307s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016078    72
 0.30% :     0.000049s :      1: add_recomputation
 0.54% :     0.000087s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 2.01% :     0.000323s :      1: bootstrap
 0.16% :     0.000026s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000021s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.23% :     0.005342s :      1: jit_opt_a
 0.91% :     0.000146s :      1: jit_opt_after_cconv
 0.31% :     0.000050s :      1: jit_opt_b
 2.62% :     0.000421s :      1: loop_unroll
 2.99% :     0.000481s :      1: mutable_eliminate
 6.39% :     0.001027s :     26: opt.transform.jit_opt_a
 0.28% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.83% :     0.000455s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000050s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000015s :      1: remove_dup_value
 2.12% :     0.000341s :      1: renormalize.infer
 1.86% :     0.000300s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000035s :      1: rewriter_after_opt_a
 0.55% :     0.000088s :      1: rewriter_before_opt_a
 0.46% :     0.000075s :      1: symbol_engine_optimizer
40.40% :     0.006496s :      1: type_inference


TotalTime = 0.0178112, [33]
[bootstrap]: 0.00026106
[type_inference]: 0.00401349
[event_method]: 1.569e-05
[auto_monad]: 5.726e-05
[graph_reusing]: 4.62e-06
[pre_auto_parallel]: 2.66e-06
[py_interpret_to_execute]: 4.904e-05
[rewriter_before_opt_a]: 6.786e-05
[expand_dump_flag]: 2.76999e-06
[jit_opt_a]: 0.00518201, [2]
    [Cycle 1]: 0.00164419, [27]
        [switch_simplify]: 4.825e-05
        [loop_unroll]: 3.084e-05
        [a_1]: 0.00059004
        [with_stream_mark]: 1.353e-05
        [recompute_prepare]: 7.80998e-06
        [updatestate_depend_eliminate]: 3.17002e-06
        [updatestate_assign_eliminate]: 2.45997e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [parameter_eliminate]: 1.67999e-06
        [specialize_transform]: 6.46999e-06
        [updatestate_useless_node_eliminater]: 5.42001e-06
        [accelerated_algorithm]: 5.50001e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 5.25001e-06
        [merge_forward]: 2.73003e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.302e-05
        [j_node_and_user_rematch]: 5.169e-05
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 8.27998e-06
        [inline_without_move]: 5.32999e-06
        [renormalize]: 0.00059973
        [add_forward_monad_depend]: 4.92e-06
        [auto_monad_grad]: 2.12999e-06
        [auto_monad_eliminator]: 1.178e-05
        [cse]: 1.798e-05
        [replace_applicator]: 1.138e-05
    [Cycle 2]: 0.0003376, [27]
        [switch_simplify]: 5.59e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 9.573e-05
        [with_stream_mark]: 9.10001e-06
        [recompute_prepare]: 5.35001e-06
        [updatestate_depend_eliminate]: 2.44001e-06
        [updatestate_assign_eliminate]: 2.23998e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [parameter_eliminate]: 1.17999e-06
        [specialize_transform]: 5.94999e-06
        [updatestate_useless_node_eliminater]: 5.29998e-06
        [accelerated_algorithm]: 5.61e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 5.59998e-06
        [merge_forward]: 2.65997e-06
        [cell_reuse_recompute_pass]: 1.89999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.276e-05
        [j_node_and_user_rematch]: 7.90998e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 7.18998e-06
        [inline_without_move]: 4.74e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 5.23002e-06
        [cse]: 1.065e-05
        [replace_applicator]: 5.22e-06
[py_interpret_to_execute_after_opt_a]: 9.47001e-06
[rewriter_after_opt_a]: 3.173e-05
[convert_after_rewriter]: 6.36e-06
[order_py_execute_after_rewriter]: 5.49e-06
[mutable_eliminate]: 0.00052261
[jit_opt_b]: 5.135e-05, [1]
    [Cycle 1]: 4.358e-05, [2]
        [frontend_op_eliminate]: 1.708e-05
        [inline_after_opt_a]: 1.529e-05
[cconv]: 2.395e-05
[loop_unroll]: 0.00042386
[jit_opt_after_cconv]: 0.00013887, [1]
    [Cycle 1]: 0.00013264, [11]
        [c_1]: 2.124e-05
        [parameter_eliminate]: 2.80002e-06
        [updatestate_depend_eliminate]: 5.77001e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 1.682e-05
        [call_graph_tuple_transform]: 1.784e-05
        [tuple_list_get_item_eliminator]: 5.25001e-06
        [none_parameter_eliminate]: 1.35001e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.47999e-06
[remove_dup_value]: 1.344e-05
[partial_unused_args_eliminate]: 2.04e-06
[environ_conv]: 4.87e-06
[add_recomputation]: 4.212e-05
[cse_after_recomputation]: 2.022e-05, [1]
    [Cycle 1]: 1.496e-05, [1]
        [cse]: 9.31e-06
[auto_monad_reorder]: 1.479e-05
[get_jit_bprop_graph]: 1.57001e-06
[rewriter_after_jit_bprop_graph]: 3.86001e-06
[opt_after_jit_grad]: 0.00045846
[symbol_engine_optimizer]: 7.415e-05, [1]
    [Cycle 1]: 6.744e-05, [6]
        [build]: 3.65e-06
        [elim_shapecalc]: 7.88999e-06
        [elim_not_effective]: 1.298e-05
        [opt_reshape]: 5.57001e-06
        [fold_const_symbol]: 9.07001e-06
        [renormalize]: 4.7998e-07
[validate]: 3.384e-05
[backend_pass]: 1.03001e-06
[task_emit]: 0.00608863
[execute]: 7.98001e-06
Sums
    bootstrap                            :     0.000261s :  1.86%
    type_inference                       :     0.004013s : 28.66%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000057s :  0.41%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000049s :  0.35%
    rewriter_before_opt_a                :     0.000068s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.38%
    jit_opt_a.loop_unroll                :     0.000036s :  0.26%
    jit_opt_a.a_1                        :     0.000686s :  4.90%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000060s :  0.43%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000600s :  4.28%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.12%
    jit_opt_a.cse                        :     0.000029s :  0.20%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000523s :  3.73%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000424s :  3.03%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000042s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000458s :  3.27%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006089s : 43.48%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000185    23
 1.19% :     0.000002s :      2: substitution.elim_not_effective
 0.92% :     0.000002s :      2: substitution.fold_const_symbol
 2.75% :     0.000005s :      3: substitution.graph_param_transform
83.76% :     0.000155s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.39% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.76% :     0.000003s :      2: substitution.replace_old_param
 5.65% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003965     2
81.52% :     0.003232s :      1: type_inference.infer
18.48% :     0.000733s :      1: type_inference.specialize
------[replace.]   0.000126     6
82.81% :     0.000104s :      4: replace.inline
17.19% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     6
94.29% :     0.000153s :      4: match.inline
 5.71% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.30% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000001s :     10: predicate.addn_check_dump
 1.46% :     0.000002s :     10: predicate.addn_zero_filter
 1.85% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.26% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.74% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.28% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.94% :     0.000002s :     15: predicate.load_eliminater
 1.17% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.02% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.17% :     0.000001s :     10: predicate.minmaximum_grad
 1.88% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     15: predicate.partial_eliminate
 1.36% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     10: predicate.reduce_eliminate
 1.65% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.41% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     10: predicate.same_eliminate
 0.76% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.59% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.46% :     0.000003s :     16: predicate.switch_defer_inline
 2.26% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.54% :     0.000009s :     53: predicate.switch_simplify
 1.33% :     0.000002s :     10: predicate.tile_eliminate
 1.33% :     0.000002s :     10: predicate.transpose_eliminate
 1.57% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.39% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000499    12
46.47% :     0.000232s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.70% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.83% :     0.000234s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019457    76
 0.23% :     0.000045s :      1: add_recomputation
 0.31% :     0.000060s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.39% :     0.000270s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
26.65% :     0.005186s :      1: jit_opt_a
 0.73% :     0.000142s :      1: jit_opt_after_cconv
 0.28% :     0.000054s :      1: jit_opt_b
 2.22% :     0.000431s :      1: loop_unroll
 2.73% :     0.000531s :      1: mutable_eliminate
 4.80% :     0.000934s :     26: opt.transform.jit_opt_a
 0.24% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.40% :     0.000466s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000052s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.60% :     0.000311s :      1: renormalize.infer
 1.45% :     0.000282s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000035s :      1: rewriter_after_opt_a
 0.36% :     0.000071s :      1: rewriter_before_opt_a
 0.39% :     0.000077s :      1: symbol_engine_optimizer
31.36% :     0.006101s :      1: task_emit
20.67% :     0.004022s :      1: type_inference
 0.27% :     0.000052s :      1: validate


TotalTime = 0.0134817, [30]
[bootstrap]: 0.00028943
[type_inference]: 0.00615087
[event_method]: 1.871e-05
[auto_monad]: 5.637e-05
[graph_reusing]: 4.16001e-06
[pre_auto_parallel]: 2.05002e-06
[py_interpret_to_execute]: 4.556e-05
[rewriter_before_opt_a]: 7.311e-05
[expand_dump_flag]: 1.92999e-06
[jit_opt_a]: 0.00488492, [2]
    [Cycle 1]: 0.0016096, [27]
        [switch_simplify]: 0.00010673
        [loop_unroll]: 3.293e-05
        [a_1]: 0.00062094
        [with_stream_mark]: 9.02e-06
        [recompute_prepare]: 6.91001e-06
        [updatestate_depend_eliminate]: 2.64001e-06
        [updatestate_assign_eliminate]: 2.01998e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 9.09989e-07
        [specialize_transform]: 5.64e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 5.45001e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 5.56e-06
        [merge_forward]: 2.31e-06
        [cell_reuse_recompute_pass]: 8.2e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.034e-05
        [j_node_and_user_rematch]: 8.40999e-06
        [meta_fg_expand]: 1.69998e-06
        [replace_old_param]: 7.83999e-06
        [inline_without_move]: 5.35001e-06
        [renormalize]: 0.00057722
        [add_forward_monad_depend]: 3.86999e-06
        [auto_monad_grad]: 1.22e-06
        [auto_monad_eliminator]: 8.67998e-06
        [cse]: 1.435e-05
        [replace_applicator]: 9.78998e-06
    [Cycle 2]: 0.00029832, [27]
        [switch_simplify]: 5.35001e-06
        [loop_unroll]: 4.83001e-06
        [a_1]: 8.887e-05
        [with_stream_mark]: 6.60002e-06
        [recompute_prepare]: 5.24e-06
        [updatestate_depend_eliminate]: 2.11998e-06
        [updatestate_assign_eliminate]: 1.87001e-06
        [updatestate_loads_eliminate]: 1.66998e-06
        [parameter_eliminate]: 7.89994e-07
        [specialize_transform]: 4.98001e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 4.4e-06
        [meta_shard_fg_expand]: 8.29983e-07
        [get_grad_eliminate_]: 4.51002e-06
        [merge_forward]: 2.01998e-06
        [cell_reuse_recompute_pass]: 1.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.06e-05
        [j_node_and_user_rematch]: 7.32002e-06
        [meta_fg_expand]: 1.37e-06
        [replace_old_param]: 6.51e-06
        [inline_without_move]: 4.53001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 8.70001e-07
        [auto_monad_grad]: 6.30011e-07
        [auto_monad_eliminator]: 4.02e-06
        [cse]: 8.32998e-06
        [replace_applicator]: 4.83001e-06
[py_interpret_to_execute_after_opt_a]: 6.43e-06
[rewriter_after_opt_a]: 2.095e-05
[convert_after_rewriter]: 5.24e-06
[order_py_execute_after_rewriter]: 3.66999e-06
[mutable_eliminate]: 0.00047324
[jit_opt_b]: 5.668e-05, [1]
    [Cycle 1]: 5.009e-05, [2]
        [frontend_op_eliminate]: 2.497e-05
        [inline_after_opt_a]: 1.364e-05
[cconv]: 1.46e-05
[loop_unroll]: 0.00041239
[jit_opt_after_cconv]: 0.00013306, [1]
    [Cycle 1]: 0.00012638, [11]
        [c_1]: 2.02e-05
        [parameter_eliminate]: 2.28002e-06
        [updatestate_depend_eliminate]: 5.86e-06
        [updatestate_assign_eliminate]: 2.31998e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [cse]: 1.651e-05
        [call_graph_tuple_transform]: 1.687e-05
        [tuple_list_get_item_eliminator]: 5.20001e-06
        [none_parameter_eliminate]: 1.09e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.09e-06
[remove_dup_value]: 1.22e-05
[partial_unused_args_eliminate]: 2.23002e-06
[environ_conv]: 4.70999e-06
[add_recomputation]: 4.2e-05
[cse_after_recomputation]: 1.883e-05, [1]
    [Cycle 1]: 1.383e-05, [1]
        [cse]: 8.35001e-06
[auto_monad_reorder]: 1.344e-05
[get_jit_bprop_graph]: 1.87001e-06
[rewriter_after_jit_bprop_graph]: 5.05999e-06
[opt_after_jit_grad]: 0.00047042
[symbol_engine_optimizer]: 7.195e-05, [1]
    [Cycle 1]: 6.607e-05, [6]
        [build]: 3.42002e-06
        [elim_shapecalc]: 7.3e-06
        [elim_not_effective]: 1.272e-05
        [opt_reshape]: 5.59e-06
        [fold_const_symbol]: 8.69e-06
        [renormalize]: 4.19997e-07
[validate]: 3.232e-05
Sums
    bootstrap                            :     0.000289s :  2.90%
    type_inference                       :     0.006151s : 61.67%
    event_method                         :     0.000019s :  0.19%
    auto_monad                           :     0.000056s :  0.57%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000046s :  0.46%
    rewriter_before_opt_a                :     0.000073s :  0.73%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000112s :  1.12%
    jit_opt_a.loop_unroll                :     0.000038s :  0.38%
    jit_opt_a.a_1                        :     0.000710s :  7.12%
    jit_opt_a.with_stream_mark           :     0.000016s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000021s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000577s :  5.79%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.13%
    jit_opt_a.cse                        :     0.000023s :  0.23%
    jit_opt_a.replace_applicator         :     0.000015s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000006s :  0.06%
    rewriter_after_opt_a                 :     0.000021s :  0.21%
    convert_after_rewriter               :     0.000005s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000473s :  4.75%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.25%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.14%
    cconv                                :     0.000015s :  0.15%
    loop_unroll                          :     0.000412s :  4.13%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000012s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000042s :  0.42%
    cse_after_recomputation.cse          :     0.000008s :  0.08%
    auto_monad_reorder                   :     0.000013s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000470s :  4.72%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.32%


Time group info:
------[substitution.]   0.000221    26
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.20% :     0.000005s :      3: substitution.graph_param_transform
85.72% :     0.000190s :      6: substitution.inline
 1.19% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.65% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.05% :     0.000002s :      2: substitution.replace_old_param
 3.51% :     0.000008s :      1: substitution.switch_simplify
 3.44% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006103     2
84.28% :     0.005144s :      1: type_inference.infer
15.72% :     0.000959s :      1: type_inference.specialize
------[replace.]   0.000124     9
48.45% :     0.000060s :      6: replace.inline
26.68% :     0.000033s :      1: replace.switch_simplify
24.87% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     9
93.27% :     0.000186s :      6: match.inline
 3.53% :     0.000007s :      1: match.switch_simplify
 3.20% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   801
 1.44% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.18% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.20% :     0.000002s :     12: predicate.addn_check_dump
 1.55% :     0.000002s :     12: predicate.addn_zero_filter
 1.90% :     0.000002s :     12: predicate.arithmetic_simplify
 1.42% :     0.000002s :     12: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.22% :     0.000002s :     12: predicate.compare_switch_simplify
 1.51% :     0.000002s :     12: predicate.depend_value_elim
 1.22% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.47% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.32% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.22% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     12: predicate.environ_get_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 4.99% :     0.000006s :     26: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     17: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.15% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.98% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.25% :     0.000002s :     12: predicate.merge_addn
 1.17% :     0.000001s :     12: predicate.minmaximum_grad
 1.68% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     17: predicate.partial_eliminate
 1.26% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     12: predicate.reduce_eliminate
 1.57% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     20: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     12: predicate.reshape_eliminate
 1.33% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.61% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     12: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.57% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.60% :     0.000003s :     20: predicate.switch_defer_inline
 2.34% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.43% :     0.000011s :     65: predicate.switch_simplify
 1.29% :     0.000002s :     12: predicate.tile_eliminate
 1.32% :     0.000002s :     12: predicate.transpose_eliminate
 1.62% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.37% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.02% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     12: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000722    15
51.00% :     0.000368s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.23% :     0.000081s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.78% :     0.000273s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015089    72
 0.30% :     0.000045s :      1: add_recomputation
 0.40% :     0.000060s :      1: auto_monad
 0.11% :     0.000016s :      1: auto_monad_reorder
 2.01% :     0.000303s :      1: bootstrap
 0.11% :     0.000017s :      1: cconv
 0.05% :     0.000007s :      1: convert_after_rewriter
 0.14% :     0.000021s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.14% :     0.000022s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
32.39% :     0.004888s :      1: jit_opt_a
 0.90% :     0.000136s :      1: jit_opt_after_cconv
 0.39% :     0.000059s :      1: jit_opt_b
 2.78% :     0.000420s :      1: loop_unroll
 3.19% :     0.000481s :      1: mutable_eliminate
 6.38% :     0.000963s :     26: opt.transform.jit_opt_a
 0.29% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000033s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.17% :     0.000478s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.32% :     0.000048s :      1: py_interpret_to_execute
 0.06% :     0.000009s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000014s :      1: remove_dup_value
 2.13% :     0.000321s :      1: renormalize.infer
 1.67% :     0.000252s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000024s :      1: rewriter_after_opt_a
 0.50% :     0.000076s :      1: rewriter_before_opt_a
 0.49% :     0.000074s :      1: symbol_engine_optimizer
40.82% :     0.006160s :      1: type_inference


TotalTime = 0.0170826, [33]
[bootstrap]: 0.00024283
[type_inference]: 0.00389968
[event_method]: 1.576e-05
[auto_monad]: 5.488e-05
[graph_reusing]: 4.57e-06
[pre_auto_parallel]: 2.40002e-06
[py_interpret_to_execute]: 3.149e-05
[rewriter_before_opt_a]: 6.685e-05
[expand_dump_flag]: 2.73e-06
[jit_opt_a]: 0.00505022, [2]
    [Cycle 1]: 0.00151487, [27]
        [switch_simplify]: 4.901e-05
        [loop_unroll]: 3.002e-05
        [a_1]: 0.00050758
        [with_stream_mark]: 1.475e-05
        [recompute_prepare]: 6.81999e-06
        [updatestate_depend_eliminate]: 3.09001e-06
        [updatestate_assign_eliminate]: 2.69001e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [parameter_eliminate]: 1.40001e-06
        [specialize_transform]: 5.81998e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 5.37999e-06
        [meta_shard_fg_expand]: 2.33002e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 2.96001e-06
        [cell_reuse_recompute_pass]: 1.07998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.282e-05
        [j_node_and_user_rematch]: 8.70999e-06
        [meta_fg_expand]: 2.20002e-06
        [replace_old_param]: 8.38001e-06
        [inline_without_move]: 5.32001e-06
        [renormalize]: 0.00062037
        [add_forward_monad_depend]: 4.90999e-06
        [auto_monad_grad]: 2.29001e-06
        [auto_monad_eliminator]: 1.187e-05
        [cse]: 1.75e-05
        [replace_applicator]: 1.188e-05
    [Cycle 2]: 0.00032564, [27]
        [switch_simplify]: 5.78997e-06
        [loop_unroll]: 5.05001e-06
        [a_1]: 9.339e-05
        [with_stream_mark]: 8.08999e-06
        [recompute_prepare]: 4.85001e-06
        [updatestate_depend_eliminate]: 2.39999e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 9.60019e-07
        [specialize_transform]: 5.07e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 1.172e-05
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.53999e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.206e-05
        [j_node_and_user_rematch]: 8.28999e-06
        [meta_fg_expand]: 1.45999e-06
        [replace_old_param]: 7.43e-06
        [inline_without_move]: 4.79e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.05001e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 4.53001e-06
        [cse]: 9.07999e-06
        [replace_applicator]: 4.95999e-06
[py_interpret_to_execute_after_opt_a]: 1.05e-05
[rewriter_after_opt_a]: 3.191e-05
[convert_after_rewriter]: 6.73998e-06
[order_py_execute_after_rewriter]: 5.31002e-06
[mutable_eliminate]: 0.00052065
[jit_opt_b]: 4.895e-05, [1]
    [Cycle 1]: 4.23e-05, [2]
        [frontend_op_eliminate]: 1.6e-05
        [inline_after_opt_a]: 1.472e-05
[cconv]: 2.285e-05
[loop_unroll]: 0.00041231
[jit_opt_after_cconv]: 0.00013293, [1]
    [Cycle 1]: 0.00012666, [11]
        [c_1]: 2.022e-05
        [parameter_eliminate]: 2.06e-06
        [updatestate_depend_eliminate]: 5.76e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [cse]: 1.601e-05
        [call_graph_tuple_transform]: 1.765e-05
        [tuple_list_get_item_eliminator]: 5.22999e-06
        [none_parameter_eliminate]: 6.39993e-07
        [renormalize]: 2.69996e-07
        [switch_simplify]: 5.04e-06
[remove_dup_value]: 1.254e-05
[partial_unused_args_eliminate]: 1.50999e-06
[environ_conv]: 4.43999e-06
[add_recomputation]: 4.154e-05
[cse_after_recomputation]: 1.862e-05, [1]
    [Cycle 1]: 1.346e-05, [1]
        [cse]: 8.05e-06
[auto_monad_reorder]: 1.365e-05
[get_jit_bprop_graph]: 1.82001e-06
[rewriter_after_jit_bprop_graph]: 5.14e-06
[opt_after_jit_grad]: 0.00047047
[symbol_engine_optimizer]: 7.266e-05, [1]
    [Cycle 1]: 6.625e-05, [6]
        [build]: 3.48999e-06
        [elim_shapecalc]: 7.53999e-06
        [elim_not_effective]: 1.235e-05
        [opt_reshape]: 5.84e-06
        [fold_const_symbol]: 9.20001e-06
        [renormalize]: 4.69998e-07
[validate]: 3.175e-05
[backend_pass]: 1.05001e-06
[task_emit]: 0.00564945
[execute]: 8.90999e-06
Sums
    bootstrap                            :     0.000243s :  1.83%
    type_inference                       :     0.003900s : 29.34%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000055s :  0.41%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.24%
    rewriter_before_opt_a                :     0.000067s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.41%
    jit_opt_a.loop_unroll                :     0.000035s :  0.26%
    jit_opt_a.a_1                        :     0.000601s :  4.52%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000620s :  4.67%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.12%
    jit_opt_a.cse                        :     0.000027s :  0.20%
    jit_opt_a.replace_applicator         :     0.000017s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.08%
    rewriter_after_opt_a                 :     0.000032s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000521s :  3.92%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000412s :  3.10%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000042s :  0.31%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000470s :  3.54%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005649s : 42.51%
    execute                              :     0.000009s :  0.07%


Time group info:
------[substitution.]   0.000180    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.77% :     0.000001s :      2: substitution.fold_const_symbol
 2.67% :     0.000005s :      3: substitution.graph_param_transform
83.20% :     0.000150s :      4: substitution.inline
 1.66% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.36% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.94% :     0.000003s :      2: substitution.replace_old_param
 6.43% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003846     2
81.32% :     0.003128s :      1: type_inference.infer
18.68% :     0.000718s :      1: type_inference.specialize
------[replace.]   0.000065     6
66.62% :     0.000044s :      4: replace.inline
33.38% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000158     6
93.35% :     0.000147s :      4: match.inline
 6.65% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.21% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 1.99% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000000s :      3: predicate.elim_not_effective
 1.02% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.33% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.04% :     0.000006s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.27% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.90% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.27% :     0.000001s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.11% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.54% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.67% :     0.000003s :     16: predicate.switch_defer_inline
 2.42% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.37% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 2.00% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.54% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000507    12
45.95% :     0.000233s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.39% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.67% :     0.000237s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018613    76
 0.24% :     0.000044s :      1: add_recomputation
 0.31% :     0.000058s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.37% :     0.000255s :      1: bootstrap
 0.14% :     0.000025s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.07% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
27.15% :     0.005053s :      1: jit_opt_a
 0.73% :     0.000135s :      1: jit_opt_after_cconv
 0.28% :     0.000051s :      1: jit_opt_b
 2.26% :     0.000420s :      1: loop_unroll
 2.84% :     0.000529s :      1: mutable_eliminate
 4.34% :     0.000808s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.57% :     0.000478s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000034s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.77% :     0.000329s :      1: renormalize.infer
 1.53% :     0.000284s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000035s :      1: rewriter_after_opt_a
 0.37% :     0.000070s :      1: rewriter_before_opt_a
 0.40% :     0.000075s :      1: symbol_engine_optimizer
30.44% :     0.005666s :      1: task_emit
21.01% :     0.003911s :      1: type_inference
 0.27% :     0.000051s :      1: validate


TotalTime = 0.0134385, [30]
[bootstrap]: 0.00025595
[type_inference]: 0.00593124
[event_method]: 1.916e-05
[auto_monad]: 6.261e-05
[graph_reusing]: 4.75999e-06
[pre_auto_parallel]: 1.91e-06
[py_interpret_to_execute]: 4.96e-05
[rewriter_before_opt_a]: 7.727e-05
[expand_dump_flag]: 2.53e-06
[jit_opt_a]: 0.00502719, [2]
    [Cycle 1]: 0.001632, [27]
        [switch_simplify]: 9.267e-05
        [loop_unroll]: 3.428e-05
        [a_1]: 0.00062683
        [with_stream_mark]: 1.371e-05
        [recompute_prepare]: 7.63001e-06
        [updatestate_depend_eliminate]: 2.99001e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.17999e-06
        [specialize_transform]: 5.82999e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.34e-06
        [meta_shard_fg_expand]: 1.45999e-06
        [get_grad_eliminate_]: 5.53002e-06
        [merge_forward]: 2.39999e-06
        [cell_reuse_recompute_pass]: 8.90024e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.116e-05
        [j_node_and_user_rematch]: 8.30999e-06
        [meta_fg_expand]: 1.85001e-06
        [replace_old_param]: 7.5e-06
        [inline_without_move]: 5.13002e-06
        [renormalize]: 0.00058227
        [add_forward_monad_depend]: 4.37998e-06
        [auto_monad_grad]: 1.37e-06
        [auto_monad_eliminator]: 1.016e-05
        [cse]: 1.511e-05
        [replace_applicator]: 1.112e-05
    [Cycle 2]: 0.00030205, [27]
        [switch_simplify]: 5.21002e-06
        [loop_unroll]: 4.68001e-06
        [a_1]: 8.783e-05
        [with_stream_mark]: 8.01001e-06
        [recompute_prepare]: 4.62e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 1.94999e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 9.10019e-07
        [specialize_transform]: 4.79002e-06
        [updatestate_useless_node_eliminater]: 4.61002e-06
        [accelerated_algorithm]: 4.55999e-06
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.39002e-06
        [merge_forward]: 2.16e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.173e-05
        [j_node_and_user_rematch]: 7.51001e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 6.71e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 3.99887e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 4.53999e-06
        [cse]: 9.24998e-06
        [replace_applicator]: 4.94e-06
[py_interpret_to_execute_after_opt_a]: 7.71001e-06
[rewriter_after_opt_a]: 2.468e-05
[convert_after_rewriter]: 5.52001e-06
[order_py_execute_after_rewriter]: 3.98999e-06
[mutable_eliminate]: 0.0004867
[jit_opt_b]: 5.64e-05, [1]
    [Cycle 1]: 5.004e-05, [2]
        [frontend_op_eliminate]: 2.355e-05
        [inline_after_opt_a]: 1.466e-05
[cconv]: 1.524e-05
[loop_unroll]: 0.00044173
[jit_opt_after_cconv]: 0.0001338, [1]
    [Cycle 1]: 0.00012773, [11]
        [c_1]: 2.108e-05
        [parameter_eliminate]: 2.71e-06
        [updatestate_depend_eliminate]: 4.94e-06
        [updatestate_assign_eliminate]: 2.31998e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 1.672e-05
        [call_graph_tuple_transform]: 1.611e-05
        [tuple_list_get_item_eliminator]: 5.41998e-06
        [none_parameter_eliminate]: 7.39994e-07
        [renormalize]: 4.2998e-07
        [switch_simplify]: 5.40001e-06
[remove_dup_value]: 7.30998e-06
[partial_unused_args_eliminate]: 1.76e-06
[environ_conv]: 4.16001e-06
[add_recomputation]: 3.142e-05
[cse_after_recomputation]: 2.011e-05, [1]
    [Cycle 1]: 1.458e-05, [1]
        [cse]: 8.84e-06
[auto_monad_reorder]: 9.14998e-06
[get_jit_bprop_graph]: 1.44998e-06
[rewriter_after_jit_bprop_graph]: 5.05999e-06
[opt_after_jit_grad]: 0.00045665
[symbol_engine_optimizer]: 7.029e-05, [1]
    [Cycle 1]: 6.425e-05, [6]
        [build]: 3.25e-06
        [elim_shapecalc]: 7.16001e-06
        [elim_not_effective]: 1.224e-05
        [opt_reshape]: 5.33002e-06
        [fold_const_symbol]: 7.95e-06
        [renormalize]: 4.39992e-07
[validate]: 2.636e-05
Sums
    bootstrap                            :     0.000256s :  2.62%
    type_inference                       :     0.005931s : 60.79%
    event_method                         :     0.000019s :  0.20%
    auto_monad                           :     0.000063s :  0.64%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000050s :  0.51%
    rewriter_before_opt_a                :     0.000077s :  0.79%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000098s :  1.00%
    jit_opt_a.loop_unroll                :     0.000039s :  0.40%
    jit_opt_a.a_1                        :     0.000715s :  7.32%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000582s :  5.97%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.15%
    jit_opt_a.cse                        :     0.000024s :  0.25%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.08%
    rewriter_after_opt_a                 :     0.000025s :  0.25%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000487s :  4.99%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.24%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.15%
    cconv                                :     0.000015s :  0.16%
    loop_unroll                          :     0.000442s :  4.53%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000007s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000031s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000009s :  0.09%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000457s :  4.68%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000026s :  0.27%


Time group info:
------[substitution.]   0.000225    26
 0.61% :     0.000001s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 1.49% :     0.000003s :      3: substitution.graph_param_transform
86.18% :     0.000194s :      6: substitution.inline
 1.13% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.80% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.12% :     0.000003s :      2: substitution.replace_old_param
 3.42% :     0.000008s :      1: substitution.switch_simplify
 3.83% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005881     2
82.88% :     0.004874s :      1: type_inference.infer
17.12% :     0.001007s :      1: type_inference.specialize
------[replace.]   0.000117     9
53.89% :     0.000063s :      6: replace.inline
19.46% :     0.000023s :      1: replace.switch_simplify
26.65% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000205     9
93.00% :     0.000191s :      6: match.inline
 3.34% :     0.000007s :      1: match.switch_simplify
 3.66% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   801
 1.41% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.22% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.18% :     0.000001s :     12: predicate.addn_check_dump
 1.53% :     0.000002s :     12: predicate.addn_zero_filter
 1.75% :     0.000002s :     12: predicate.arithmetic_simplify
 1.45% :     0.000002s :     12: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     12: predicate.compare_switch_simplify
 1.42% :     0.000002s :     12: predicate.depend_value_elim
 1.24% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.24% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     12: predicate.environ_get_eliminate
 1.18% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.11% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.24% :     0.000007s :     26: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     17: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.38% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000002s :     12: predicate.merge_addn
 1.15% :     0.000001s :     12: predicate.minmaximum_grad
 1.60% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     17: predicate.partial_eliminate
 1.33% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     12: predicate.reduce_eliminate
 1.51% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     20: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.21% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     12: predicate.reshape_eliminate
 1.24% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     12: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.77% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.66% :     0.000003s :     20: predicate.switch_defer_inline
 2.31% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.24% :     0.000010s :     65: predicate.switch_simplify
 1.41% :     0.000002s :     12: predicate.tile_eliminate
 1.32% :     0.000002s :     12: predicate.transpose_eliminate
 1.55% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.28% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000002s :     12: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000724    15
48.82% :     0.000354s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.57% :     0.000077s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.61% :     0.000294s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015051    72
 0.23% :     0.000034s :      1: add_recomputation
 0.44% :     0.000066s :      1: auto_monad
 0.08% :     0.000012s :      1: auto_monad_reorder
 2.04% :     0.000307s :      1: bootstrap
 0.12% :     0.000018s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
33.42% :     0.005030s :      1: jit_opt_a
 0.91% :     0.000136s :      1: jit_opt_after_cconv
 0.39% :     0.000059s :      1: jit_opt_b
 2.98% :     0.000449s :      1: loop_unroll
 3.29% :     0.000495s :      1: mutable_eliminate
 6.36% :     0.000958s :     26: opt.transform.jit_opt_a
 0.30% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000032s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000029s :      4: opt.transform.symbol_engine_opt
 3.08% :     0.000464s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.35% :     0.000053s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000010s :      1: remove_dup_value
 2.13% :     0.000320s :      1: renormalize.infer
 1.71% :     0.000257s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000028s :      1: rewriter_after_opt_a
 0.53% :     0.000080s :      1: rewriter_before_opt_a
 0.48% :     0.000073s :      1: symbol_engine_optimizer
39.50% :     0.005946s :      1: type_inference


TotalTime = 0.0171268, [33]
[bootstrap]: 0.00025547
[type_inference]: 0.00402757
[event_method]: 1.717e-05
[auto_monad]: 4.156e-05
[graph_reusing]: 3.91999e-06
[pre_auto_parallel]: 1.82001e-06
[py_interpret_to_execute]: 3.162e-05
[rewriter_before_opt_a]: 6.527e-05
[expand_dump_flag]: 2.02999e-06
[jit_opt_a]: 0.00481705, [2]
    [Cycle 1]: 0.00137813, [27]
        [switch_simplify]: 4.635e-05
        [loop_unroll]: 2.983e-05
        [a_1]: 0.00049819
        [with_stream_mark]: 1.229e-05
        [recompute_prepare]: 7.26001e-06
        [updatestate_depend_eliminate]: 2.89999e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 5.76e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.07999e-06
        [meta_shard_fg_expand]: 1.28002e-06
        [get_grad_eliminate_]: 5.18002e-06
        [merge_forward]: 2.33998e-06
        [cell_reuse_recompute_pass]: 9.50007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.089e-05
        [j_node_and_user_rematch]: 8.68001e-06
        [meta_fg_expand]: 1.99999e-06
        [replace_old_param]: 7.26999e-06
        [inline_without_move]: 5.35999e-06
        [renormalize]: 0.00051225
        [add_forward_monad_depend]: 4.32998e-06
        [auto_monad_grad]: 1.62001e-06
        [auto_monad_eliminator]: 9.79e-06
        [cse]: 1.773e-05
        [replace_applicator]: 1.105e-05
    [Cycle 2]: 0.00032768, [27]
        [switch_simplify]: 5.46e-06
        [loop_unroll]: 4.74998e-06
        [a_1]: 9.154e-05
        [with_stream_mark]: 9.10001e-06
        [recompute_prepare]: 5.10999e-06
        [updatestate_depend_eliminate]: 2.44001e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.86003e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 4.81997e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 4.94e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 1.175e-05
        [merge_forward]: 2.34999e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.19e-05
        [j_node_and_user_rematch]: 8.05999e-06
        [meta_fg_expand]: 1.47999e-06
        [replace_old_param]: 7.35e-06
        [inline_without_move]: 4.75001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.30999e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 5.33002e-06
        [cse]: 1.095e-05
        [replace_applicator]: 5.61003e-06
[py_interpret_to_execute_after_opt_a]: 8.64003e-06
[rewriter_after_opt_a]: 2.598e-05
[convert_after_rewriter]: 5.81e-06
[order_py_execute_after_rewriter]: 4.02e-06
[mutable_eliminate]: 0.00050348
[jit_opt_b]: 4.778e-05, [1]
    [Cycle 1]: 4.155e-05, [2]
        [frontend_op_eliminate]: 1.577e-05
        [inline_after_opt_a]: 1.477e-05
[cconv]: 1.547e-05
[loop_unroll]: 0.00042005
[jit_opt_after_cconv]: 0.00013427, [1]
    [Cycle 1]: 0.000128, [11]
        [c_1]: 1.993e-05
        [parameter_eliminate]: 2.56e-06
        [updatestate_depend_eliminate]: 5.52999e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 1.659e-05
        [call_graph_tuple_transform]: 1.684e-05
        [tuple_list_get_item_eliminator]: 5.57001e-06
        [none_parameter_eliminate]: 9.70002e-07
        [renormalize]: 4.09986e-07
        [switch_simplify]: 5.31002e-06
[remove_dup_value]: 7.33e-06
[partial_unused_args_eliminate]: 1.67999e-06
[environ_conv]: 3.83001e-06
[add_recomputation]: 3.231e-05
[cse_after_recomputation]: 1.973e-05, [1]
    [Cycle 1]: 1.507e-05, [1]
        [cse]: 9.36e-06
[auto_monad_reorder]: 9.51003e-06
[get_jit_bprop_graph]: 1.34e-06
[rewriter_after_jit_bprop_graph]: 4.48999e-06
[opt_after_jit_grad]: 0.00045661
[symbol_engine_optimizer]: 7.169e-05, [1]
    [Cycle 1]: 6.516e-05, [6]
        [build]: 3.03e-06
        [elim_shapecalc]: 8.09002e-06
        [elim_not_effective]: 1.235e-05
        [opt_reshape]: 5.66e-06
        [fold_const_symbol]: 8.13999e-06
        [renormalize]: 4.09986e-07
[validate]: 2.658e-05
[backend_pass]: 1.02e-06
[task_emit]: 0.00586886
[execute]: 9.02e-06
Sums
    bootstrap                            :     0.000255s :  1.90%
    type_inference                       :     0.004028s : 29.97%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000042s :  0.31%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000032s :  0.24%
    rewriter_before_opt_a                :     0.000065s :  0.49%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000052s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.26%
    jit_opt_a.a_1                        :     0.000590s :  4.39%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.13%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000512s :  3.81%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000029s :  0.21%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000026s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000503s :  3.75%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000015s :  0.12%
    loop_unroll                          :     0.000420s :  3.13%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000007s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000032s :  0.24%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000010s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000457s :  3.40%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005869s : 43.67%
    execute                              :     0.000009s :  0.07%


Time group info:
------[substitution.]   0.000149    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.46% :     0.000004s :      3: substitution.graph_param_transform
84.41% :     0.000126s :      4: substitution.inline
 1.79% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.67% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.77% :     0.000003s :      2: substitution.replace_old_param
 5.18% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003978     2
81.13% :     0.003228s :      1: type_inference.infer
18.87% :     0.000750s :      1: type_inference.specialize
------[replace.]   0.000061     6
65.13% :     0.000040s :      4: replace.inline
34.87% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000130     6
94.93% :     0.000124s :      4: match.inline
 5.07% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.61% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.44% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.19% :     0.000001s :     10: predicate.addn_check_dump
 1.50% :     0.000002s :     10: predicate.addn_zero_filter
 1.89% :     0.000002s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.12% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.64% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.19% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.54% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.58% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000002s :     15: predicate.partial_eliminate
 1.26% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.71% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.62% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.67% :     0.000009s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.57% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.57% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000506    12
47.88% :     0.000242s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.07% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.05% :     0.000238s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018532    76
 0.19% :     0.000035s :      1: add_recomputation
 0.24% :     0.000045s :      1: auto_monad
 0.07% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.45% :     0.000268s :      1: bootstrap
 0.10% :     0.000018s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
26.01% :     0.004820s :      1: jit_opt_a
 0.74% :     0.000137s :      1: jit_opt_after_cconv
 0.27% :     0.000050s :      1: jit_opt_b
 2.31% :     0.000428s :      1: loop_unroll
 2.76% :     0.000512s :      1: mutable_eliminate
 4.28% :     0.000793s :     26: opt.transform.jit_opt_a
 0.24% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.51% :     0.000465s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000010s :      1: remove_dup_value
 1.35% :     0.000250s :      1: renormalize.infer
 1.38% :     0.000255s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000029s :      1: rewriter_after_opt_a
 0.37% :     0.000068s :      1: rewriter_before_opt_a
 0.40% :     0.000074s :      1: symbol_engine_optimizer
31.76% :     0.005885s :      1: task_emit
21.79% :     0.004038s :      1: type_inference
 0.23% :     0.000042s :      1: validate


TotalTime = 0.0186793, [33]
[bootstrap]: 0.00030513
[type_inference]: 0.00461656
[event_method]: 1.672e-05
[auto_monad]: 6.278e-05
[graph_reusing]: 5.16998e-06
[pre_auto_parallel]: 2.55002e-06
[py_interpret_to_execute]: 3.463e-05
[rewriter_before_opt_a]: 7.275e-05
[expand_dump_flag]: 2.89001e-06
[jit_opt_a]: 0.0052977, [2]
    [Cycle 1]: 0.00153842, [27]
        [switch_simplify]: 5.295e-05
        [loop_unroll]: 2.96e-05
        [a_1]: 0.00051893
        [with_stream_mark]: 1.622e-05
        [recompute_prepare]: 7.20998e-06
        [updatestate_depend_eliminate]: 3.38e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 2.73998e-06
        [parameter_eliminate]: 2.06e-06
        [specialize_transform]: 6.01e-06
        [updatestate_useless_node_eliminater]: 5.11002e-06
        [accelerated_algorithm]: 5.67001e-06
        [meta_shard_fg_expand]: 1.98002e-06
        [get_grad_eliminate_]: 5.56e-06
        [merge_forward]: 3.46999e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.417e-05
        [j_node_and_user_rematch]: 8.73001e-06
        [meta_fg_expand]: 2.32001e-06
        [replace_old_param]: 9.25999e-06
        [inline_without_move]: 5.14e-06
        [renormalize]: 0.0005938
        [add_forward_monad_depend]: 5.46998e-06
        [auto_monad_grad]: 2.27001e-06
        [auto_monad_eliminator]: 1.391e-05
        [cse]: 3.119e-05
        [replace_applicator]: 1.33e-05
    [Cycle 2]: 0.00033971, [27]
        [switch_simplify]: 5.89999e-06
        [loop_unroll]: 5.08002e-06
        [a_1]: 9.019e-05
        [with_stream_mark]: 8.82e-06
        [recompute_prepare]: 5.15999e-06
        [updatestate_depend_eliminate]: 2.78998e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 5.07999e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 4.99998e-06
        [meta_shard_fg_expand]: 1.20001e-06
        [get_grad_eliminate_]: 1.427e-05
        [merge_forward]: 2.74999e-06
        [cell_reuse_recompute_pass]: 1.29998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.341e-05
        [j_node_and_user_rematch]: 7.71001e-06
        [meta_fg_expand]: 1.80001e-06
        [replace_old_param]: 6.94001e-06
        [inline_without_move]: 7.06999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 1.07e-06
        [auto_monad_eliminator]: 5.67999e-06
        [cse]: 1.212e-05
        [replace_applicator]: 5.78002e-06
[py_interpret_to_execute_after_opt_a]: 1.107e-05
[rewriter_after_opt_a]: 5.712e-05
[convert_after_rewriter]: 7.8e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.00053139
[jit_opt_b]: 4.781e-05, [1]
    [Cycle 1]: 4.16e-05, [2]
        [frontend_op_eliminate]: 1.541e-05
        [inline_after_opt_a]: 1.434e-05
[cconv]: 2.518e-05
[loop_unroll]: 0.00042479
[jit_opt_after_cconv]: 0.00014208, [1]
    [Cycle 1]: 0.00013562, [11]
        [c_1]: 2.089e-05
        [parameter_eliminate]: 2.71999e-06
        [updatestate_depend_eliminate]: 5.49e-06
        [updatestate_assign_eliminate]: 2.41998e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [cse]: 2.005e-05
        [call_graph_tuple_transform]: 1.881e-05
        [tuple_list_get_item_eliminator]: 5.15001e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 5.55001e-06
[remove_dup_value]: 1.502e-05
[partial_unused_args_eliminate]: 2.29001e-06
[environ_conv]: 5.60001e-06
[add_recomputation]: 4.66e-05
[cse_after_recomputation]: 2.105e-05, [1]
    [Cycle 1]: 1.559e-05, [1]
        [cse]: 9.46e-06
[auto_monad_reorder]: 1.506e-05
[get_jit_bprop_graph]: 2.11e-06
[rewriter_after_jit_bprop_graph]: 6.21e-06
[opt_after_jit_grad]: 0.00045759
[symbol_engine_optimizer]: 7.364e-05, [1]
    [Cycle 1]: 6.637e-05, [6]
        [build]: 3.88001e-06
        [elim_shapecalc]: 7.65e-06
        [elim_not_effective]: 1.29e-05
        [opt_reshape]: 5.50001e-06
        [fold_const_symbol]: 8.74e-06
        [renormalize]: 3.89991e-07
[validate]: 3.653e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00612231
[execute]: 8.23001e-06
Sums
    bootstrap                            :     0.000305s :  2.08%
    type_inference                       :     0.004617s : 31.51%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000063s :  0.43%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.24%
    rewriter_before_opt_a                :     0.000073s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000609s :  4.16%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000020s :  0.14%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000012s :  0.08%
    jit_opt_a.renormalize                :     0.000594s :  4.05%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000043s :  0.30%
    jit_opt_a.replace_applicator         :     0.000019s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.08%
    rewriter_after_opt_a                 :     0.000057s :  0.39%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000531s :  3.63%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000425s :  2.90%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000015s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000047s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000458s :  3.12%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006122s : 41.79%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000189    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.74% :     0.000005s :      3: substitution.graph_param_transform
82.56% :     0.000156s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.37% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.47% :     0.000005s :      2: substitution.replace_old_param
 6.67% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004550     2
82.62% :     0.003759s :      1: type_inference.infer
17.38% :     0.000791s :      1: type_inference.specialize
------[replace.]   0.000068     6
67.55% :     0.000046s :      4: replace.inline
32.45% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000165     6
93.03% :     0.000154s :      4: match.inline
 6.97% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.51% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.36% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 1.86% :     0.000002s :     10: predicate.arithmetic_simplify
 1.16% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.14% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.06% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.24% :     0.000006s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.02% :     0.000002s :     15: predicate.load_eliminater
 1.64% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.09% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.22% :     0.000001s :     10: predicate.minmaximum_grad
 1.70% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.68% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.70% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.11% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.10% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000002s :     10: predicate.value_based_eliminate
 0.43% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000547    12
51.52% :     0.000282s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.96% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.53% :     0.000238s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020203    76
 0.24% :     0.000049s :      1: add_recomputation
 0.33% :     0.000066s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.60% :     0.000324s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.24% :     0.005301s :      1: jit_opt_a
 0.72% :     0.000145s :      1: jit_opt_after_cconv
 0.25% :     0.000050s :      1: jit_opt_b
 2.14% :     0.000433s :      1: loop_unroll
 2.67% :     0.000539s :      1: mutable_eliminate
 4.10% :     0.000828s :     26: opt.transform.jit_opt_a
 0.23% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.30% :     0.000465s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000037s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000017s :      1: remove_dup_value
 1.47% :     0.000297s :      1: renormalize.infer
 1.43% :     0.000289s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.30% :     0.000061s :      1: rewriter_after_opt_a
 0.37% :     0.000076s :      1: rewriter_before_opt_a
 0.38% :     0.000077s :      1: symbol_engine_optimizer
30.36% :     0.006134s :      1: task_emit
22.93% :     0.004632s :      1: type_inference
 0.27% :     0.000055s :      1: validate


TotalTime = 0.0181201, [33]
[bootstrap]: 0.00028089
[type_inference]: 0.00454491
[event_method]: 1.843e-05
[auto_monad]: 6.373e-05
[graph_reusing]: 5.40001e-06
[pre_auto_parallel]: 2.50002e-06
[py_interpret_to_execute]: 3.462e-05
[rewriter_before_opt_a]: 7.169e-05
[expand_dump_flag]: 3.11999e-06
[jit_opt_a]: 0.00517453, [2]
    [Cycle 1]: 0.00146974, [27]
        [switch_simplify]: 5.214e-05
        [loop_unroll]: 3.084e-05
        [a_1]: 0.00051421
        [with_stream_mark]: 1.679e-05
        [recompute_prepare]: 6.93998e-06
        [updatestate_depend_eliminate]: 3.54002e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [parameter_eliminate]: 1.77001e-06
        [specialize_transform]: 5.78002e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.93997e-06
        [get_grad_eliminate_]: 5.22999e-06
        [merge_forward]: 3.43e-06
        [cell_reuse_recompute_pass]: 1.06997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.441e-05
        [j_node_and_user_rematch]: 9.00999e-06
        [meta_fg_expand]: 2.29001e-06
        [replace_old_param]: 8.59998e-06
        [inline_without_move]: 5.25001e-06
        [renormalize]: 0.00054423
        [add_forward_monad_depend]: 5.04e-06
        [auto_monad_grad]: 2.06998e-06
        [auto_monad_eliminator]: 1.276e-05
        [cse]: 2.815e-05
        [replace_applicator]: 1.084e-05
    [Cycle 2]: 0.00035372, [27]
        [switch_simplify]: 5.62001e-06
        [loop_unroll]: 4.85001e-06
        [a_1]: 9.546e-05
        [with_stream_mark]: 8.18001e-06
        [recompute_prepare]: 5.04e-06
        [updatestate_depend_eliminate]: 2.54001e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 4.96002e-06
        [updatestate_useless_node_eliminater]: 1.118e-05
        [accelerated_algorithm]: 4.74e-06
        [meta_shard_fg_expand]: 9.89996e-07
        [get_grad_eliminate_]: 4.23999e-06
        [merge_forward]: 2.38002e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.211e-05
        [j_node_and_user_rematch]: 8.95001e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 8.79e-06
        [inline_without_move]: 4.84e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 5.94e-06
        [cse]: 1.204e-05
        [replace_applicator]: 5.77999e-06
[py_interpret_to_execute_after_opt_a]: 1.012e-05
[rewriter_after_opt_a]: 3.3e-05
[convert_after_rewriter]: 7.18e-06
[order_py_execute_after_rewriter]: 5.19e-06
[mutable_eliminate]: 0.00048375
[jit_opt_b]: 4.794e-05, [1]
    [Cycle 1]: 4.12e-05, [2]
        [frontend_op_eliminate]: 1.555e-05
        [inline_after_opt_a]: 1.457e-05
[cconv]: 2.411e-05
[loop_unroll]: 0.00045542
[jit_opt_after_cconv]: 0.00013657, [1]
    [Cycle 1]: 0.00013048, [11]
        [c_1]: 2.047e-05
        [parameter_eliminate]: 2.52001e-06
        [updatestate_depend_eliminate]: 4.82e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [cse]: 1.717e-05
        [call_graph_tuple_transform]: 1.89e-05
        [tuple_list_get_item_eliminator]: 5.25999e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 7.09988e-07
        [switch_simplify]: 5.34998e-06
[remove_dup_value]: 1.433e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 4.37998e-06
[add_recomputation]: 4.518e-05
[cse_after_recomputation]: 1.909e-05, [1]
    [Cycle 1]: 1.399e-05, [1]
        [cse]: 8.74998e-06
[auto_monad_reorder]: 1.52e-05
[get_jit_bprop_graph]: 1.64e-06
[rewriter_after_jit_bprop_graph]: 5.00999e-06
[opt_after_jit_grad]: 0.00044904
[symbol_engine_optimizer]: 7.134e-05, [1]
    [Cycle 1]: 6.49e-05, [6]
        [build]: 3.26001e-06
        [elim_shapecalc]: 7.66001e-06
        [elim_not_effective]: 1.33e-05
        [opt_reshape]: 5.34e-06
        [fold_const_symbol]: 8.32e-06
        [renormalize]: 6.59988e-07
[validate]: 3.186e-05
[backend_pass]: 1.06002e-06
[task_emit]: 0.00586168
[execute]: 7.88001e-06
Sums
    bootstrap                            :     0.000281s :  1.98%
    type_inference                       :     0.004545s : 32.09%
    event_method                         :     0.000018s :  0.13%
    auto_monad                           :     0.000064s :  0.45%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.24%
    rewriter_before_opt_a                :     0.000072s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.41%
    jit_opt_a.loop_unroll                :     0.000036s :  0.25%
    jit_opt_a.a_1                        :     0.000610s :  4.30%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000544s :  3.84%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000040s :  0.28%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000484s :  3.42%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000455s :  3.22%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000045s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000449s :  3.17%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005862s : 41.39%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000183    23
 1.25% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.92% :     0.000005s :      3: substitution.graph_param_transform
83.04% :     0.000152s :      4: substitution.inline
 1.60% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.27% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.00% :     0.000004s :      2: substitution.replace_old_param
 6.27% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004478     2
81.92% :     0.003668s :      1: type_inference.infer
18.08% :     0.000809s :      1: type_inference.specialize
------[replace.]   0.000066     6
67.86% :     0.000045s :      4: replace.inline
32.14% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000160     6
93.52% :     0.000150s :      4: match.inline
 6.48% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.49% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 1.82% :     0.000002s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000001s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.52% :     0.000002s :     10: predicate.environ_get_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.83% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.00% :     0.000002s :     15: predicate.load_eliminater
 1.65% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.32% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.36% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.36% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.88% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.28% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.55% :     0.000009s :     53: predicate.switch_simplify
 1.10% :     0.000001s :     10: predicate.tile_eliminate
 1.25% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.00% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.89% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000578    12
53.75% :     0.000311s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.86% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.39% :     0.000239s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019591    76
 0.24% :     0.000048s :      1: add_recomputation
 0.34% :     0.000067s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.52% :     0.000298s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.43% :     0.005178s :      1: jit_opt_a
 0.71% :     0.000139s :      1: jit_opt_after_cconv
 0.26% :     0.000051s :      1: jit_opt_b
 2.36% :     0.000463s :      1: loop_unroll
 2.51% :     0.000492s :      1: mutable_eliminate
 4.20% :     0.000823s :     26: opt.transform.jit_opt_a
 0.24% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.33% :     0.000457s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000017s :      1: remove_dup_value
 1.36% :     0.000266s :      1: renormalize.infer
 1.38% :     0.000271s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000036s :      1: rewriter_after_opt_a
 0.38% :     0.000074s :      1: rewriter_before_opt_a
 0.38% :     0.000074s :      1: symbol_engine_optimizer
29.97% :     0.005871s :      1: task_emit
23.28% :     0.004561s :      1: type_inference
 0.25% :     0.000049s :      1: validate


TotalTime = 0.0137785, [30]
[bootstrap]: 0.00026793
[type_inference]: 0.00594235
[event_method]: 1.978e-05
[auto_monad]: 7.611e-05
[graph_reusing]: 6.32001e-06
[pre_auto_parallel]: 2.69999e-06
[py_interpret_to_execute]: 4.285e-05
[rewriter_before_opt_a]: 8.123e-05
[expand_dump_flag]: 3.06001e-06
[jit_opt_a]: 0.00531403, [2]
    [Cycle 1]: 0.0017079, [27]
        [switch_simplify]: 0.00010557
        [loop_unroll]: 3.37e-05
        [a_1]: 0.00066824
        [with_stream_mark]: 1.563e-05
        [recompute_prepare]: 6.88e-06
        [updatestate_depend_eliminate]: 3.65e-06
        [updatestate_assign_eliminate]: 2.73003e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [parameter_eliminate]: 1.72999e-06
        [specialize_transform]: 5.82001e-06
        [updatestate_useless_node_eliminater]: 5.18002e-06
        [accelerated_algorithm]: 5.82999e-06
        [meta_shard_fg_expand]: 2.04e-06
        [get_grad_eliminate_]: 5.35999e-06
        [merge_forward]: 3.51999e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.423e-05
        [j_node_and_user_rematch]: 9.36002e-06
        [meta_fg_expand]: 2.54999e-06
        [replace_old_param]: 8.79e-06
        [inline_without_move]: 5.43002e-06
        [renormalize]: 0.00058425
        [add_forward_monad_depend]: 4.96002e-06
        [auto_monad_grad]: 2.16e-06
        [auto_monad_eliminator]: 1.312e-05
        [cse]: 1.999e-05
        [replace_applicator]: 1.137e-05
    [Cycle 2]: 0.00030198, [27]
        [switch_simplify]: 5.40001e-06
        [loop_unroll]: 4.96002e-06
        [a_1]: 8.59e-05
        [with_stream_mark]: 8.80999e-06
        [recompute_prepare]: 5.10001e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 2.06998e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 9.90025e-07
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 4.62e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.21e-06
        [cell_reuse_recompute_pass]: 1.84e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.157e-05
        [j_node_and_user_rematch]: 7.55998e-06
        [meta_fg_expand]: 1.42999e-06
        [replace_old_param]: 7.4e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.03001e-06
        [auto_monad_grad]: 7.80012e-07
        [auto_monad_eliminator]: 4.70001e-06
        [cse]: 9.12001e-06
        [replace_applicator]: 4.77998e-06
[py_interpret_to_execute_after_opt_a]: 8.85999e-06
[rewriter_after_opt_a]: 3.17e-05
[convert_after_rewriter]: 7.01001e-06
[order_py_execute_after_rewriter]: 4.62998e-06
[mutable_eliminate]: 0.00049092
[jit_opt_b]: 5.6e-05, [1]
    [Cycle 1]: 4.949e-05, [2]
        [frontend_op_eliminate]: 2.192e-05
        [inline_after_opt_a]: 1.587e-05
[cconv]: 2.293e-05
[loop_unroll]: 0.00042949
[jit_opt_after_cconv]: 0.00013451, [1]
    [Cycle 1]: 0.00012843, [11]
        [c_1]: 2.01e-05
        [parameter_eliminate]: 2.36998e-06
        [updatestate_depend_eliminate]: 5.22e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.01003e-06
        [cse]: 1.635e-05
        [call_graph_tuple_transform]: 1.845e-05
        [tuple_list_get_item_eliminator]: 5.22e-06
        [none_parameter_eliminate]: 1.37999e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 5.18002e-06
[remove_dup_value]: 1.258e-05
[partial_unused_args_eliminate]: 2.44001e-06
[environ_conv]: 5.59e-06
[add_recomputation]: 4.559e-05
[cse_after_recomputation]: 2.076e-05, [1]
    [Cycle 1]: 1.532e-05, [1]
        [cse]: 9.48997e-06
[auto_monad_reorder]: 1.483e-05
[get_jit_bprop_graph]: 1.67999e-06
[rewriter_after_jit_bprop_graph]: 4.98001e-06
[opt_after_jit_grad]: 0.00046183
[symbol_engine_optimizer]: 7.441e-05, [1]
    [Cycle 1]: 6.826e-05, [6]
        [build]: 3.06001e-06
        [elim_shapecalc]: 7.88999e-06
        [elim_not_effective]: 1.335e-05
        [opt_reshape]: 6.08998e-06
        [fold_const_symbol]: 8.92999e-06
        [renormalize]: 5.19998e-07
[validate]: 3.361e-05
Sums
    bootstrap                            :     0.000268s :  2.70%
    type_inference                       :     0.005942s : 59.86%
    event_method                         :     0.000020s :  0.20%
    auto_monad                           :     0.000076s :  0.77%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000043s :  0.43%
    rewriter_before_opt_a                :     0.000081s :  0.82%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000111s :  1.12%
    jit_opt_a.loop_unroll                :     0.000039s :  0.39%
    jit_opt_a.a_1                        :     0.000754s :  7.60%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000584s :  5.89%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.18%
    jit_opt_a.cse                        :     0.000029s :  0.29%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.09%
    rewriter_after_opt_a                 :     0.000032s :  0.32%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000491s :  4.95%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.16%
    cconv                                :     0.000023s :  0.23%
    loop_unroll                          :     0.000429s :  4.33%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000046s :  0.46%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000462s :  4.65%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000034s :  0.34%


Time group info:
------[substitution.]   0.000263    26
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.12% :     0.000006s :      3: substitution.graph_param_transform
84.39% :     0.000222s :      6: substitution.inline
 1.10% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.69% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.40% :     0.000004s :      2: substitution.replace_old_param
 3.26% :     0.000009s :      1: substitution.switch_simplify
 4.66% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005888     2
83.79% :     0.004933s :      1: type_inference.infer
16.21% :     0.000955s :      1: type_inference.specialize
------[replace.]   0.000119     9
51.20% :     0.000061s :      6: replace.inline
23.66% :     0.000028s :      1: replace.switch_simplify
25.14% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000238     9
92.05% :     0.000219s :      6: match.inline
 3.28% :     0.000008s :      1: match.switch_simplify
 4.67% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   801
 1.39% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.06% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     12: predicate.addn_check_dump
 1.42% :     0.000002s :     12: predicate.addn_zero_filter
 2.45% :     0.000003s :     12: predicate.arithmetic_simplify
 1.42% :     0.000002s :     12: predicate.cast_eliminate
 0.35% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.20% :     0.000002s :     12: predicate.compare_switch_simplify
 1.38% :     0.000002s :     12: predicate.depend_value_elim
 1.27% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     12: predicate.environ_get_eliminate
 1.16% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.31% :     0.000007s :     26: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     17: predicate.load_eliminater
 1.23% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.23% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     12: predicate.merge_addn
 1.17% :     0.000002s :     12: predicate.minmaximum_grad
 1.36% :     0.000002s :      3: predicate.mutable_eliminate
 0.58% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     17: predicate.partial_eliminate
 1.19% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.98% :     0.000003s :     12: predicate.reduce_eliminate
 1.45% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     20: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.60% :     0.000002s :     12: predicate.reshape_eliminate
 1.29% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 1.06% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     12: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.69% :     0.000003s :     20: predicate.switch_defer_inline
 2.37% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.52% :     0.000011s :     65: predicate.switch_simplify
 1.21% :     0.000002s :     12: predicate.tile_eliminate
 1.24% :     0.000002s :     12: predicate.transpose_eliminate
 1.43% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.59% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     12: predicate.value_based_eliminate
 0.27% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000730    15
49.50% :     0.000361s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.49% :     0.000091s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.00% :     0.000277s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015439    72
 0.31% :     0.000048s :      1: add_recomputation
 0.52% :     0.000080s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.80% :     0.000279s :      1: bootstrap
 0.17% :     0.000026s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
34.44% :     0.005317s :      1: jit_opt_a
 0.89% :     0.000137s :      1: jit_opt_after_cconv
 0.38% :     0.000059s :      1: jit_opt_b
 2.83% :     0.000437s :      1: loop_unroll
 3.23% :     0.000499s :      1: mutable_eliminate
 6.57% :     0.001015s :     26: opt.transform.jit_opt_a
 0.30% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000031s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.04% :     0.000469s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000046s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000015s :      1: remove_dup_value
 2.07% :     0.000320s :      1: renormalize.infer
 1.67% :     0.000257s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000035s :      1: rewriter_after_opt_a
 0.54% :     0.000084s :      1: rewriter_before_opt_a
 0.50% :     0.000077s :      1: symbol_engine_optimizer
38.57% :     0.005955s :      1: type_inference


TotalTime = 0.0177171, [33]
[bootstrap]: 0.00025866
[type_inference]: 0.00428552
[event_method]: 1.74e-05
[auto_monad]: 6.209e-05
[graph_reusing]: 5.49e-06
[pre_auto_parallel]: 2.49001e-06
[py_interpret_to_execute]: 3.375e-05
[rewriter_before_opt_a]: 7.176e-05
[expand_dump_flag]: 3.11001e-06
[jit_opt_a]: 0.00500687, [2]
    [Cycle 1]: 0.00145352, [27]
        [switch_simplify]: 5.235e-05
        [loop_unroll]: 2.945e-05
        [a_1]: 0.00050924
        [with_stream_mark]: 1.646e-05
        [recompute_prepare]: 6.78998e-06
        [updatestate_depend_eliminate]: 3.58e-06
        [updatestate_assign_eliminate]: 3.6e-06
        [updatestate_loads_eliminate]: 2.63003e-06
        [parameter_eliminate]: 1.55999e-06
        [specialize_transform]: 5.77999e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 2.02001e-06
        [get_grad_eliminate_]: 5.21002e-06
        [merge_forward]: 3.48e-06
        [cell_reuse_recompute_pass]: 1.10999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.423e-05
        [j_node_and_user_rematch]: 8.92e-06
        [meta_fg_expand]: 2.11998e-06
        [replace_old_param]: 8.90001e-06
        [inline_without_move]: 5.00001e-06
        [renormalize]: 0.00053667
        [add_forward_monad_depend]: 5.41998e-06
        [auto_monad_grad]: 2.29001e-06
        [auto_monad_eliminator]: 1.395e-05
        [cse]: 2.444e-05
        [replace_applicator]: 1.297e-05
    [Cycle 2]: 0.00031342, [27]
        [switch_simplify]: 5.90002e-06
        [loop_unroll]: 4.98001e-06
        [a_1]: 9.104e-05
        [with_stream_mark]: 1.006e-05
        [recompute_prepare]: 5.02e-06
        [updatestate_depend_eliminate]: 2.61999e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 4.74e-06
        [updatestate_useless_node_eliminater]: 4.70999e-06
        [accelerated_algorithm]: 4.60999e-06
        [meta_shard_fg_expand]: 1.32999e-06
        [get_grad_eliminate_]: 4.43001e-06
        [merge_forward]: 2.74001e-06
        [cell_reuse_recompute_pass]: 2.22001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.239e-05
        [j_node_and_user_rematch]: 7.44002e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 6.93e-06
        [inline_without_move]: 4.60999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 4.70001e-06
        [cse]: 1.017e-05
        [replace_applicator]: 4.93001e-06
[py_interpret_to_execute_after_opt_a]: 8.33999e-06
[rewriter_after_opt_a]: 3.195e-05
[convert_after_rewriter]: 6.73998e-06
[order_py_execute_after_rewriter]: 5.16998e-06
[mutable_eliminate]: 0.00055775
[jit_opt_b]: 5.09e-05, [1]
    [Cycle 1]: 4.438e-05, [2]
        [frontend_op_eliminate]: 1.735e-05
        [inline_after_opt_a]: 1.563e-05
[cconv]: 2.503e-05
[loop_unroll]: 0.00046727
[jit_opt_after_cconv]: 0.00015408, [1]
    [Cycle 1]: 0.00014774, [11]
        [c_1]: 2.89e-05
        [parameter_eliminate]: 2.44999e-06
        [updatestate_depend_eliminate]: 5.57001e-06
        [updatestate_assign_eliminate]: 2.69999e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [cse]: 1.85e-05
        [call_graph_tuple_transform]: 2.203e-05
        [tuple_list_get_item_eliminator]: 5.80002e-06
        [none_parameter_eliminate]: 1.51002e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 5.54e-06
[remove_dup_value]: 1.46e-05
[partial_unused_args_eliminate]: 2.24001e-06
[environ_conv]: 5.82001e-06
[add_recomputation]: 5.214e-05
[cse_after_recomputation]: 2.236e-05, [1]
    [Cycle 1]: 1.651e-05, [1]
        [cse]: 9.99001e-06
[auto_monad_reorder]: 1.527e-05
[get_jit_bprop_graph]: 1.50001e-06
[rewriter_after_jit_bprop_graph]: 5.59998e-06
[opt_after_jit_grad]: 0.00046477
[symbol_engine_optimizer]: 7.764e-05, [1]
    [Cycle 1]: 7.176e-05, [6]
        [build]: 3.3e-06
        [elim_shapecalc]: 7.2e-06
        [elim_not_effective]: 1.301e-05
        [opt_reshape]: 5.51e-06
        [fold_const_symbol]: 1.412e-05
        [renormalize]: 5.00004e-07
[validate]: 3.594e-05
[backend_pass]: 8.49977e-07
[task_emit]: 0.00578098
[execute]: 7.82e-06
Sums
    bootstrap                            :     0.000259s :  1.86%
    type_inference                       :     0.004286s : 30.82%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000062s :  0.45%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.24%
    rewriter_before_opt_a                :     0.000072s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.42%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000600s :  4.32%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000537s :  3.86%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000035s :  0.25%
    jit_opt_a.replace_applicator         :     0.000018s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000032s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000558s :  4.01%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000025s :  0.18%
    loop_unroll                          :     0.000467s :  3.36%
    jit_opt_after_cconv.c_1              :     0.000029s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000015s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000052s :  0.37%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000465s :  3.34%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005781s : 41.57%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000184    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.82% :     0.000005s :      3: substitution.graph_param_transform
82.45% :     0.000152s :      4: substitution.inline
 1.72% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.41% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.27% :     0.000004s :      2: substitution.replace_old_param
 6.59% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004230     2
81.42% :     0.003444s :      1: type_inference.infer
18.58% :     0.000786s :      1: type_inference.specialize
------[replace.]   0.000062     6
66.91% :     0.000042s :      4: replace.inline
33.09% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000160     6
93.21% :     0.000149s :      4: match.inline
 6.79% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.32% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 2.01% :     0.000003s :     10: predicate.arithmetic_simplify
 1.39% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.22% :     0.000002s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.04% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.45% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.13% :     0.000006s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.32% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.90% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.75% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 2.08% :     0.000003s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 7.51% :     0.000009s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.55% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.24% :     0.000002s :     10: predicate.reshape_eliminate
 1.42% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.24% :     0.000003s :     16: predicate.switch_defer_inline
 2.00% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.22% :     0.000009s :     53: predicate.switch_simplify
 1.08% :     0.000001s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.39% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.93% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000526    12
47.96% :     0.000252s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.54% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.50% :     0.000245s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019185    76
 0.29% :     0.000055s :      1: add_recomputation
 0.34% :     0.000065s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.41% :     0.000270s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.11% :     0.005010s :      1: jit_opt_a
 0.82% :     0.000157s :      1: jit_opt_after_cconv
 0.28% :     0.000054s :      1: jit_opt_b
 2.48% :     0.000476s :      1: loop_unroll
 2.95% :     0.000566s :      1: mutable_eliminate
 4.19% :     0.000803s :     26: opt.transform.jit_opt_a
 0.31% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.46% :     0.000473s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000017s :      1: remove_dup_value
 1.48% :     0.000284s :      1: renormalize.infer
 1.28% :     0.000245s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000035s :      1: rewriter_after_opt_a
 0.39% :     0.000075s :      1: rewriter_before_opt_a
 0.42% :     0.000080s :      1: symbol_engine_optimizer
30.19% :     0.005791s :      1: task_emit
22.40% :     0.004297s :      1: type_inference
 0.28% :     0.000053s :      1: validate


TotalTime = 0.0154866, [30]
[bootstrap]: 0.00027788
[type_inference]: 0.00676648
[event_method]: 2.086e-05
[auto_monad]: 8.098e-05
[graph_reusing]: 7.06999e-06
[pre_auto_parallel]: 3.18e-06
[py_interpret_to_execute]: 5.475e-05
[rewriter_before_opt_a]: 8.928e-05
[expand_dump_flag]: 3.58e-06
[jit_opt_a]: 0.00587651, [2]
    [Cycle 1]: 0.0019465, [27]
        [switch_simplify]: 0.00012461
        [loop_unroll]: 3.512e-05
        [a_1]: 0.00070946
        [with_stream_mark]: 2.024e-05
        [recompute_prepare]: 7.82002e-06
        [updatestate_depend_eliminate]: 3.55e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [parameter_eliminate]: 2.06e-06
        [specialize_transform]: 6.12001e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 2.64001e-06
        [get_grad_eliminate_]: 5.52999e-06
        [merge_forward]: 4.03999e-06
        [cell_reuse_recompute_pass]: 1.89e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.524e-05
        [j_node_and_user_rematch]: 8.84e-06
        [meta_fg_expand]: 2.09999e-06
        [replace_old_param]: 8.45001e-06
        [inline_without_move]: 5.30999e-06
        [renormalize]: 0.00068259
        [add_forward_monad_depend]: 6.21e-06
        [auto_monad_grad]: 2.07999e-06
        [auto_monad_eliminator]: 1.43e-05
        [cse]: 2.796e-05
        [replace_applicator]: 1.369e-05
    [Cycle 2]: 0.00041367, [27]
        [switch_simplify]: 6.29999e-06
        [loop_unroll]: 4.77e-06
        [a_1]: 0.00012979
        [with_stream_mark]: 1.142e-05
        [recompute_prepare]: 6.06e-06
        [updatestate_depend_eliminate]: 2.83e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.72999e-06
        [specialize_transform]: 2.334e-05
        [updatestate_useless_node_eliminater]: 4.81002e-06
        [accelerated_algorithm]: 4.50001e-06
        [meta_shard_fg_expand]: 1.72999e-06
        [get_grad_eliminate_]: 4.69002e-06
        [merge_forward]: 3.06001e-06
        [cell_reuse_recompute_pass]: 1.93002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.414e-05
        [j_node_and_user_rematch]: 7.78001e-06
        [meta_fg_expand]: 1.40001e-06
        [replace_old_param]: 8.38999e-06
        [inline_without_move]: 4.76002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.32999e-06
        [auto_monad_grad]: 1.55999e-06
        [auto_monad_eliminator]: 6.80002e-06
        [cse]: 1.308e-05
        [replace_applicator]: 5.04e-06
[py_interpret_to_execute_after_opt_a]: 1.127e-05
[rewriter_after_opt_a]: 3.524e-05
[convert_after_rewriter]: 7.45998e-06
[order_py_execute_after_rewriter]: 5.12e-06
[mutable_eliminate]: 0.00059922
[jit_opt_b]: 5.881e-05, [1]
    [Cycle 1]: 5.13e-05, [2]
        [frontend_op_eliminate]: 2.317e-05
        [inline_after_opt_a]: 1.513e-05
[cconv]: 2.72e-05
[loop_unroll]: 0.00049129
[jit_opt_after_cconv]: 0.00014651, [1]
    [Cycle 1]: 0.00013932, [11]
        [c_1]: 2.082e-05
        [parameter_eliminate]: 2.69999e-06
        [updatestate_depend_eliminate]: 6.28998e-06
        [updatestate_assign_eliminate]: 2.48002e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 2.014e-05
        [call_graph_tuple_transform]: 1.827e-05
        [tuple_list_get_item_eliminator]: 5.71e-06
        [none_parameter_eliminate]: 1.76e-06
        [renormalize]: 8.00006e-07
        [switch_simplify]: 6.09999e-06
[remove_dup_value]: 1.652e-05
[partial_unused_args_eliminate]: 2.51998e-06
[environ_conv]: 5.22999e-06
[add_recomputation]: 5.001e-05
[cse_after_recomputation]: 2.469e-05, [1]
    [Cycle 1]: 1.904e-05, [1]
        [cse]: 1.153e-05
[auto_monad_reorder]: 1.472e-05
[get_jit_bprop_graph]: 2.18998e-06
[rewriter_after_jit_bprop_graph]: 6.86999e-06
[opt_after_jit_grad]: 0.00051305
[symbol_engine_optimizer]: 7.685e-05, [1]
    [Cycle 1]: 6.989e-05, [6]
        [build]: 4.72e-06
        [elim_shapecalc]: 7.97e-06
        [elim_not_effective]: 1.435e-05
        [opt_reshape]: 5.51998e-06
        [fold_const_symbol]: 7.97e-06
        [renormalize]: 4.89992e-07
[validate]: 3.803e-05
Sums
    bootstrap                            :     0.000278s :  2.46%
    type_inference                       :     0.006766s : 59.83%
    event_method                         :     0.000021s :  0.18%
    auto_monad                           :     0.000081s :  0.72%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000055s :  0.48%
    rewriter_before_opt_a                :     0.000089s :  0.79%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000131s :  1.16%
    jit_opt_a.loop_unroll                :     0.000040s :  0.35%
    jit_opt_a.a_1                        :     0.000839s :  7.42%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000029s :  0.26%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000683s :  6.04%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.19%
    jit_opt_a.cse                        :     0.000041s :  0.36%
    jit_opt_a.replace_applicator         :     0.000019s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.10%
    rewriter_after_opt_a                 :     0.000035s :  0.31%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000599s :  5.30%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.13%
    cconv                                :     0.000027s :  0.24%
    loop_unroll                          :     0.000491s :  4.34%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000017s :  0.15%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000050s :  0.44%
    cse_after_recomputation.cse          :     0.000012s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000513s :  4.54%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.34%


Time group info:
------[substitution.]   0.000279    26
 0.80% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 1.71% :     0.000005s :      3: substitution.graph_param_transform
84.60% :     0.000236s :      6: substitution.inline
 1.15% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.54% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000005s :      2: substitution.replace_old_param
 2.92% :     0.000008s :      1: substitution.switch_simplify
 4.94% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006702     2
84.26% :     0.005647s :      1: type_inference.infer
15.74% :     0.001055s :      1: type_inference.specialize
------[replace.]   0.000150     9
49.12% :     0.000074s :      6: replace.inline
25.37% :     0.000038s :      1: replace.switch_simplify
25.52% :     0.000038s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000252     9
92.22% :     0.000232s :      6: match.inline
 2.90% :     0.000007s :      1: match.switch_simplify
 4.89% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000140   801
 1.45% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.49% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     12: predicate.addn_check_dump
 1.39% :     0.000002s :     12: predicate.addn_zero_filter
 1.96% :     0.000003s :     12: predicate.arithmetic_simplify
 1.41% :     0.000002s :     12: predicate.cast_eliminate
 0.59% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     12: predicate.compare_switch_simplify
 1.41% :     0.000002s :     12: predicate.depend_value_elim
 1.09% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.11% :     0.000002s :     12: predicate.dict_set_item_eliminator
 1.07% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.92% :     0.000003s :     12: predicate.environ_get_depend_swap
 1.12% :     0.000002s :     12: predicate.environ_get_eliminate
 1.07% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.17% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.19% :     0.000000s :      3: predicate.graph_param_transform
 5.24% :     0.000007s :     26: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.02% :     0.000001s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 2.16% :     0.000003s :     17: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.81% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000002s :     12: predicate.merge_addn
 1.10% :     0.000002s :     12: predicate.minmaximum_grad
 1.71% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     17: predicate.partial_eliminate
 1.11% :     0.000002s :     12: predicate.print_const_string_wrapper
 2.66% :     0.000004s :     12: predicate.reduce_eliminate
 1.34% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.52% :     0.000002s :     20: predicate.replace_applicator
 0.41% :     0.000001s :      6: predicate.replace_old_param
 0.38% :     0.000001s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     12: predicate.reshape_eliminate
 1.17% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 2.23% :     0.000003s :     12: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 0.82% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.90% :     0.000003s :     12: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.61% :     0.000004s :     20: predicate.switch_defer_inline
 2.25% :     0.000003s :     20: predicate.switch_layer_defer_inline
 7.98% :     0.000011s :     65: predicate.switch_simplify
 1.14% :     0.000002s :     12: predicate.tile_eliminate
 1.11% :     0.000002s :     12: predicate.transpose_eliminate
 1.30% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.22% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.32% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.54% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.49% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000802    15
49.91% :     0.000400s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.12% :     0.000089s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.97% :     0.000312s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017382    72
 0.31% :     0.000054s :      1: add_recomputation
 0.49% :     0.000085s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 1.66% :     0.000289s :      1: bootstrap
 0.17% :     0.000030s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.14% :     0.000024s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.83% :     0.005881s :      1: jit_opt_a
 0.86% :     0.000149s :      1: jit_opt_after_cconv
 0.35% :     0.000062s :      1: jit_opt_b
 2.89% :     0.000502s :      1: loop_unroll
 3.51% :     0.000610s :      1: mutable_eliminate
 6.59% :     0.001146s :     26: opt.transform.jit_opt_a
 0.27% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000032s :      4: opt.transform.jit_opt_b
 0.08% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.01% :     0.000523s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.33% :     0.000058s :      1: py_interpret_to_execute
 0.08% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000019s :      1: remove_dup_value
 2.22% :     0.000385s :      1: renormalize.infer
 1.65% :     0.000287s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000039s :      1: rewriter_after_opt_a
 0.53% :     0.000092s :      1: rewriter_before_opt_a
 0.46% :     0.000080s :      1: symbol_engine_optimizer
38.99% :     0.006778s :      1: type_inference


TotalTime = 0.0190717, [33]
[bootstrap]: 0.00029577
[type_inference]: 0.00489058
[event_method]: 1.659e-05
[auto_monad]: 6.252e-05
[graph_reusing]: 5.32999e-06
[pre_auto_parallel]: 2.32001e-06
[py_interpret_to_execute]: 4.078e-05
[rewriter_before_opt_a]: 7.226e-05
[expand_dump_flag]: 2.85002e-06
[jit_opt_a]: 0.00573134, [2]
    [Cycle 1]: 0.00170954, [27]
        [switch_simplify]: 5.391e-05
        [loop_unroll]: 2.96e-05
        [a_1]: 0.00054121
        [with_stream_mark]: 1.655e-05
        [recompute_prepare]: 7.22997e-06
        [updatestate_depend_eliminate]: 2.88e-06
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.66998e-06
        [specialize_transform]: 5.67999e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 5.66e-06
        [meta_shard_fg_expand]: 3.28e-06
        [get_grad_eliminate_]: 5.30999e-06
        [merge_forward]: 3.51999e-06
        [cell_reuse_recompute_pass]: 1.67001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.316e-05
        [j_node_and_user_rematch]: 9.46e-06
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 7.77e-06
        [inline_without_move]: 5.87001e-06
        [renormalize]: 0.00068869
        [add_forward_monad_depend]: 6.29001e-06
        [auto_monad_grad]: 2.12001e-06
        [auto_monad_eliminator]: 1.363e-05
        [cse]: 2.249e-05
        [replace_applicator]: 1.494e-05
    [Cycle 2]: 0.00035013, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 5.12e-06
        [a_1]: 9.768e-05
        [with_stream_mark]: 1.051e-05
        [recompute_prepare]: 4.94e-06
        [updatestate_depend_eliminate]: 2.63e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 1.96998e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 5.00999e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 5.52001e-06
        [meta_shard_fg_expand]: 1.62001e-06
        [get_grad_eliminate_]: 4.66002e-06
        [merge_forward]: 2.86e-06
        [cell_reuse_recompute_pass]: 3.68e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.271e-05
        [j_node_and_user_rematch]: 8.62e-06
        [meta_fg_expand]: 1.82001e-06
        [replace_old_param]: 8.26002e-06
        [inline_without_move]: 4.80999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.80001e-06
        [auto_monad_grad]: 1.27e-06
        [auto_monad_eliminator]: 7.15e-06
        [cse]: 1.233e-05
        [replace_applicator]: 5.07999e-06
[py_interpret_to_execute_after_opt_a]: 1.27e-05
[rewriter_after_opt_a]: 3.531e-05
[convert_after_rewriter]: 6.63998e-06
[order_py_execute_after_rewriter]: 5.19e-06
[mutable_eliminate]: 0.0007052
[jit_opt_b]: 5.256e-05, [1]
    [Cycle 1]: 4.495e-05, [2]
        [frontend_op_eliminate]: 1.65e-05
        [inline_after_opt_a]: 1.638e-05
[cconv]: 2.536e-05
[loop_unroll]: 0.00052777
[jit_opt_after_cconv]: 0.00015158, [1]
    [Cycle 1]: 0.00014441, [11]
        [c_1]: 2.217e-05
        [parameter_eliminate]: 3.13e-06
        [updatestate_depend_eliminate]: 6.70002e-06
        [updatestate_assign_eliminate]: 2.91999e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [cse]: 2.214e-05
        [call_graph_tuple_transform]: 2.144e-05
        [tuple_list_get_item_eliminator]: 5.70001e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 1.05001e-06
        [switch_simplify]: 5.39e-06
[remove_dup_value]: 1.253e-05
[partial_unused_args_eliminate]: 1.89e-06
[environ_conv]: 4.68001e-06
[add_recomputation]: 4.885e-05
[cse_after_recomputation]: 2.116e-05, [1]
    [Cycle 1]: 1.556e-05, [1]
        [cse]: 9.77001e-06
[auto_monad_reorder]: 1.363e-05
[get_jit_bprop_graph]: 1.63002e-06
[rewriter_after_jit_bprop_graph]: 6.10002e-06
[opt_after_jit_grad]: 0.00056635
[symbol_engine_optimizer]: 7.912e-05, [1]
    [Cycle 1]: 7.058e-05, [6]
        [build]: 4.68999e-06
        [elim_shapecalc]: 7.88001e-06
        [elim_not_effective]: 1.485e-05
        [opt_reshape]: 6.20002e-06
        [fold_const_symbol]: 8.42e-06
        [renormalize]: 4.89992e-07
[validate]: 3.923e-05
[backend_pass]: 7.50006e-07
[task_emit]: 0.00542792
[execute]: 5.86e-06
Sums
    bootstrap                            :     0.000296s :  2.01%
    type_inference                       :     0.004891s : 33.21%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000063s :  0.42%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000041s :  0.28%
    rewriter_before_opt_a                :     0.000072s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000060s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000639s :  4.34%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000689s :  4.68%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.14%
    jit_opt_a.cse                        :     0.000035s :  0.24%
    jit_opt_a.replace_applicator         :     0.000020s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.09%
    rewriter_after_opt_a                 :     0.000035s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000705s :  4.79%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000528s :  3.58%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000566s :  3.85%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000039s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005428s : 36.86%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000215    23
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.32% :     0.000005s :      3: substitution.graph_param_transform
81.05% :     0.000174s :      4: substitution.inline
 1.90% :     0.000004s :      4: substitution.j_node_and_user_rematch
 5.81% :     0.000012s :      4: substitution.remove_not_recompute_node
 2.04% :     0.000004s :      2: substitution.replace_old_param
 5.26% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004821     2
81.73% :     0.003940s :      1: type_inference.infer
18.27% :     0.000881s :      1: type_inference.specialize
------[replace.]   0.000069     6
65.48% :     0.000045s :      4: replace.inline
34.52% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000182     6
94.37% :     0.000172s :      4: match.inline
 5.63% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.65% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.67% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.60% :     0.000002s :     10: predicate.addn_zero_filter
 2.30% :     0.000003s :     10: predicate.arithmetic_simplify
 1.32% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.21% :     0.000002s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.20% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.75% :     0.000002s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.30% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.02% :     0.000003s :     15: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.76% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.08% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.91% :     0.000002s :      3: predicate.mutable_eliminate
 0.57% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     10: predicate.reshape_eliminate
 1.39% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.29% :     0.000002s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.70% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.71% :     0.000008s :     53: predicate.switch_simplify
 1.31% :     0.000002s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.28% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.96% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.75% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000626    12
46.93% :     0.000294s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.80% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.27% :     0.000296s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020741    76
 0.25% :     0.000052s :      1: add_recomputation
 0.32% :     0.000066s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.50% :     0.000311s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.04% :     0.000009s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.65% :     0.005735s :      1: jit_opt_a
 0.75% :     0.000155s :      1: jit_opt_after_cconv
 0.27% :     0.000056s :      1: jit_opt_b
 2.60% :     0.000539s :      1: loop_unroll
 3.45% :     0.000715s :      1: mutable_eliminate
 4.15% :     0.000862s :     26: opt.transform.jit_opt_a
 0.25% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.78% :     0.000577s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.21% :     0.000043s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 1.63% :     0.000339s :      1: renormalize.infer
 1.65% :     0.000342s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000038s :      1: rewriter_after_opt_a
 0.36% :     0.000075s :      1: rewriter_before_opt_a
 0.40% :     0.000082s :      1: symbol_engine_optimizer
26.22% :     0.005438s :      1: task_emit
23.64% :     0.004904s :      1: type_inference
 0.29% :     0.000060s :      1: validate


TotalTime = 0.0146107, [30]
[bootstrap]: 0.00029855
[type_inference]: 0.0066769
[event_method]: 1.909e-05
[auto_monad]: 6.328e-05
[graph_reusing]: 4.4e-06
[pre_auto_parallel]: 1.94999e-06
[py_interpret_to_execute]: 4.644e-05
[rewriter_before_opt_a]: 8.23e-05
[expand_dump_flag]: 1.86e-06
[jit_opt_a]: 0.00523543, [2]
    [Cycle 1]: 0.00171595, [27]
        [switch_simplify]: 0.00010133
        [loop_unroll]: 3.243e-05
        [a_1]: 0.00065271
        [with_stream_mark]: 1.49e-05
        [recompute_prepare]: 7.15003e-06
        [updatestate_depend_eliminate]: 2.94001e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 1.56998e-06
        [specialize_transform]: 5.96e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 5.15001e-06
        [meta_shard_fg_expand]: 1.48002e-06
        [get_grad_eliminate_]: 5.07999e-06
        [merge_forward]: 2.66e-06
        [cell_reuse_recompute_pass]: 1.09998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.197e-05
        [j_node_and_user_rematch]: 9.24998e-06
        [meta_fg_expand]: 1.77999e-06
        [replace_old_param]: 8.25e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.00062252
        [add_forward_monad_depend]: 4.60999e-06
        [auto_monad_grad]: 1.11997e-06
        [auto_monad_eliminator]: 9.81998e-06
        [cse]: 1.628e-05
        [replace_applicator]: 1.175e-05
    [Cycle 2]: 0.00031069, [27]
        [switch_simplify]: 5.46e-06
        [loop_unroll]: 4.62e-06
        [a_1]: 9.098e-05
        [with_stream_mark]: 8.73001e-06
        [recompute_prepare]: 4.93001e-06
        [updatestate_depend_eliminate]: 2.31e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 5.66998e-06
        [updatestate_useless_node_eliminater]: 4.59002e-06
        [accelerated_algorithm]: 4.66002e-06
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.45999e-06
        [merge_forward]: 2.20002e-06
        [cell_reuse_recompute_pass]: 2.16e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.293e-05
        [j_node_and_user_rematch]: 7.70998e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 7.67998e-06
        [inline_without_move]: 4.92999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 9.79984e-07
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 4.69998e-06
        [cse]: 9.40001e-06
        [replace_applicator]: 4.97e-06
[py_interpret_to_execute_after_opt_a]: 9.32999e-06
[rewriter_after_opt_a]: 2.634e-05
[convert_after_rewriter]: 5.50001e-06
[order_py_execute_after_rewriter]: 3.96001e-06
[mutable_eliminate]: 0.00055882
[jit_opt_b]: 5.785e-05, [1]
    [Cycle 1]: 5.071e-05, [2]
        [frontend_op_eliminate]: 2.384e-05
        [inline_after_opt_a]: 1.479e-05
[cconv]: 2.601e-05
[loop_unroll]: 0.00046874
[jit_opt_after_cconv]: 0.00014321, [1]
    [Cycle 1]: 0.0001368, [11]
        [c_1]: 2.118e-05
        [parameter_eliminate]: 3.25002e-06
        [updatestate_depend_eliminate]: 6.00002e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 1.991e-05
        [call_graph_tuple_transform]: 1.899e-05
        [tuple_list_get_item_eliminator]: 5.59e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 5.25001e-06
[remove_dup_value]: 1.468e-05
[partial_unused_args_eliminate]: 2.64999e-06
[environ_conv]: 5.82999e-06
[add_recomputation]: 5.02e-05
[cse_after_recomputation]: 2.03e-05, [1]
    [Cycle 1]: 1.44e-05, [1]
        [cse]: 8.87999e-06
[auto_monad_reorder]: 1.584e-05
[get_jit_bprop_graph]: 1.92999e-06
[rewriter_after_jit_bprop_graph]: 6.07999e-06
[opt_after_jit_grad]: 0.00047534
[symbol_engine_optimizer]: 7.321e-05, [1]
    [Cycle 1]: 6.642e-05, [6]
        [build]: 3.93999e-06
        [elim_shapecalc]: 7.35003e-06
        [elim_not_effective]: 1.264e-05
        [opt_reshape]: 5.34998e-06
        [fold_const_symbol]: 8.67e-06
        [renormalize]: 4.10015e-07
[validate]: 4.116e-05
Sums
    bootstrap                            :     0.000299s :  2.76%
    type_inference                       :     0.006677s : 61.66%
    event_method                         :     0.000019s :  0.18%
    auto_monad                           :     0.000063s :  0.58%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000046s :  0.43%
    rewriter_before_opt_a                :     0.000082s :  0.76%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000107s :  0.99%
    jit_opt_a.loop_unroll                :     0.000037s :  0.34%
    jit_opt_a.a_1                        :     0.000744s :  6.87%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000623s :  5.75%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.13%
    jit_opt_a.cse                        :     0.000026s :  0.24%
    jit_opt_a.replace_applicator         :     0.000017s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.09%
    rewriter_after_opt_a                 :     0.000026s :  0.24%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000559s :  5.16%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.14%
    cconv                                :     0.000026s :  0.24%
    loop_unroll                          :     0.000469s :  4.33%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000015s :  0.14%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000050s :  0.46%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000475s :  4.39%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000041s :  0.38%


Time group info:
------[substitution.]   0.000245    26
 0.80% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000001s :      2: substitution.fold_const_symbol
 2.12% :     0.000005s :      3: substitution.graph_param_transform
85.56% :     0.000210s :      6: substitution.inline
 1.06% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.03% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.23% :     0.000003s :      2: substitution.replace_old_param
 3.19% :     0.000008s :      1: substitution.switch_simplify
 3.48% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006620     2
84.88% :     0.005619s :      1: type_inference.infer
15.12% :     0.001001s :      1: type_inference.specialize
------[replace.]   0.000128     9
51.10% :     0.000065s :      6: replace.inline
22.51% :     0.000029s :      1: replace.switch_simplify
26.39% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000221     9
93.48% :     0.000207s :      6: match.inline
 3.16% :     0.000007s :      1: match.switch_simplify
 3.36% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   801
 1.36% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     12: predicate.addn_check_dump
 1.45% :     0.000002s :     12: predicate.addn_zero_filter
 1.84% :     0.000002s :     12: predicate.arithmetic_simplify
 1.47% :     0.000002s :     12: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     12: predicate.compare_switch_simplify
 1.37% :     0.000002s :     12: predicate.depend_value_elim
 1.20% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.55% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     12: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.43% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.49% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     12: predicate.environ_get_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.78% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.04% :     0.000007s :     26: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     17: predicate.load_eliminater
 1.72% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.93% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000002s :     12: predicate.merge_addn
 1.11% :     0.000001s :     12: predicate.minmaximum_grad
 2.27% :     0.000003s :      3: predicate.mutable_eliminate
 0.39% :     0.000001s :      3: predicate.opt_reshape
 1.99% :     0.000003s :     17: predicate.partial_eliminate
 1.17% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     12: predicate.reduce_eliminate
 1.43% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     20: predicate.replace_applicator
 0.42% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     12: predicate.reshape_eliminate
 1.23% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.66% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     12: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.57% :     0.000003s :     20: predicate.switch_defer_inline
 2.30% :     0.000003s :     20: predicate.switch_layer_defer_inline
 7.69% :     0.000010s :     65: predicate.switch_simplify
 1.50% :     0.000002s :     12: predicate.tile_eliminate
 1.26% :     0.000002s :     12: predicate.transpose_eliminate
 1.35% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.92% :     0.000003s :     12: predicate.tuple_list_get_item_depend_reorder
 3.31% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 2.33% :     0.000003s :     12: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000786    15
50.83% :     0.000400s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.45% :     0.000090s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.72% :     0.000296s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016291    72
 0.33% :     0.000053s :      1: add_recomputation
 0.41% :     0.000067s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 1.94% :     0.000316s :      1: bootstrap
 0.18% :     0.000029s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.02% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
32.17% :     0.005241s :      1: jit_opt_a
 0.90% :     0.000146s :      1: jit_opt_after_cconv
 0.37% :     0.000061s :      1: jit_opt_b
 2.93% :     0.000477s :      1: loop_unroll
 3.49% :     0.000568s :      1: mutable_eliminate
 6.14% :     0.001001s :     26: opt.transform.jit_opt_a
 0.29% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000032s :      4: opt.transform.jit_opt_b
 0.08% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.97% :     0.000484s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.30% :     0.000050s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.08% :     0.000339s :      1: renormalize.infer
 1.69% :     0.000276s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000029s :      1: rewriter_after_opt_a
 0.52% :     0.000085s :      1: rewriter_before_opt_a
 0.46% :     0.000076s :      1: symbol_engine_optimizer
41.06% :     0.006689s :      1: type_inference


TotalTime = 0.0194736, [33]
[bootstrap]: 0.00028322
[type_inference]: 0.00451805
[event_method]: 1.731e-05
[auto_monad]: 6.624e-05
[graph_reusing]: 5.99e-06
[pre_auto_parallel]: 3.11001e-06
[py_interpret_to_execute]: 3.669e-05
[rewriter_before_opt_a]: 7.527e-05
[expand_dump_flag]: 3.49001e-06
[jit_opt_a]: 0.00569475, [2]
    [Cycle 1]: 0.00168635, [27]
        [switch_simplify]: 5.574e-05
        [loop_unroll]: 2.862e-05
        [a_1]: 0.00055219
        [with_stream_mark]: 1.862e-05
        [recompute_prepare]: 7.17002e-06
        [updatestate_depend_eliminate]: 3.61001e-06
        [updatestate_assign_eliminate]: 3.58e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 2.16e-06
        [specialize_transform]: 5.51998e-06
        [updatestate_useless_node_eliminater]: 5.06002e-06
        [accelerated_algorithm]: 5.15999e-06
        [meta_shard_fg_expand]: 1.81e-06
        [get_grad_eliminate_]: 5.17e-06
        [merge_forward]: 4.02e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.452e-05
        [j_node_and_user_rematch]: 8.80001e-06
        [meta_fg_expand]: 2.29001e-06
        [replace_old_param]: 9.24e-06
        [inline_without_move]: 5.47001e-06
        [renormalize]: 0.00071187
        [add_forward_monad_depend]: 5.89999e-06
        [auto_monad_grad]: 2.99999e-06
        [auto_monad_eliminator]: 1.435e-05
        [cse]: 2.171e-05
        [replace_applicator]: 1.307e-05
    [Cycle 2]: 0.00031497, [27]
        [switch_simplify]: 5.65001e-06
        [loop_unroll]: 5.07e-06
        [a_1]: 9.161e-05
        [with_stream_mark]: 9.41003e-06
        [recompute_prepare]: 5.00999e-06
        [updatestate_depend_eliminate]: 2.89999e-06
        [updatestate_assign_eliminate]: 1.92999e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.29998e-06
        [specialize_transform]: 5.03002e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 4.42e-06
        [meta_shard_fg_expand]: 1.06002e-06
        [get_grad_eliminate_]: 4.35999e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 1.86e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.226e-05
        [j_node_and_user_rematch]: 7.66001e-06
        [meta_fg_expand]: 1.47999e-06
        [replace_old_param]: 7.57998e-06
        [inline_without_move]: 4.70001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 5.09998e-06
        [cse]: 1.002e-05
        [replace_applicator]: 5.30001e-06
[py_interpret_to_execute_after_opt_a]: 1.309e-05
[rewriter_after_opt_a]: 3.305e-05
[convert_after_rewriter]: 6.62002e-06
[order_py_execute_after_rewriter]: 5.09998e-06
[mutable_eliminate]: 0.00074423
[jit_opt_b]: 4.985e-05, [1]
    [Cycle 1]: 4.245e-05, [2]
        [frontend_op_eliminate]: 1.531e-05
        [inline_after_opt_a]: 1.526e-05
[cconv]: 2.488e-05
[loop_unroll]: 0.00049494
[jit_opt_after_cconv]: 0.00014064, [1]
    [Cycle 1]: 0.00013411, [11]
        [c_1]: 2.141e-05
        [parameter_eliminate]: 3.10002e-06
        [updatestate_depend_eliminate]: 6.13002e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 2.004e-05
        [call_graph_tuple_transform]: 1.77e-05
        [tuple_list_get_item_eliminator]: 5.22e-06
        [none_parameter_eliminate]: 8.59989e-07
        [renormalize]: 4.39992e-07
        [switch_simplify]: 5.54e-06
[remove_dup_value]: 8.12003e-06
[partial_unused_args_eliminate]: 1.42e-06
[environ_conv]: 4e-06
[add_recomputation]: 3.412e-05
[cse_after_recomputation]: 2.184e-05, [1]
    [Cycle 1]: 1.543e-05, [1]
        [cse]: 9.45001e-06
[auto_monad_reorder]: 1.494e-05
[get_jit_bprop_graph]: 2.10002e-06
[rewriter_after_jit_bprop_graph]: 6.53e-06
[opt_after_jit_grad]: 0.00050213
[symbol_engine_optimizer]: 7.353e-05, [1]
    [Cycle 1]: 6.703e-05, [6]
        [build]: 3.83001e-06
        [elim_shapecalc]: 8.35999e-06
        [elim_not_effective]: 1.306e-05
        [opt_reshape]: 5.32001e-06
        [fold_const_symbol]: 8.40001e-06
        [renormalize]: 9.39996e-07
[validate]: 3.053e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00633539
[execute]: 9.66e-06
Sums
    bootstrap                            :     0.000283s :  1.87%
    type_inference                       :     0.004518s : 29.80%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000066s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.24%
    rewriter_before_opt_a                :     0.000075s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000644s :  4.25%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000712s :  4.70%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000032s :  0.21%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.09%
    rewriter_after_opt_a                 :     0.000033s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000744s :  4.91%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000025s :  0.16%
    loop_unroll                          :     0.000495s :  3.26%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000008s :  0.05%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000034s :  0.23%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000502s :  3.31%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000031s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006335s : 41.78%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000225    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 1.71% :     0.000004s :      3: substitution.graph_param_transform
86.06% :     0.000193s :      4: substitution.inline
 1.35% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.86% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.22% :     0.000005s :      2: substitution.replace_old_param
 5.43% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004444     2
81.46% :     0.003620s :      1: type_inference.infer
18.54% :     0.000824s :      1: type_inference.specialize
------[replace.]   0.000066     6
65.82% :     0.000044s :      4: replace.inline
34.18% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000202     6
94.45% :     0.000191s :      4: match.inline
 5.55% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 2.03% :     0.000002s :     10: predicate.arithmetic_simplify
 1.54% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000001s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.91% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.46% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.98% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.86% :     0.000002s :     15: predicate.load_eliminater
 1.76% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.10% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.81% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000000s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.85% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.29% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.93% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.20% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.83% :     0.000009s :     53: predicate.switch_simplify
 1.27% :     0.000001s :     10: predicate.tile_eliminate
 1.25% :     0.000001s :     10: predicate.transpose_eliminate
 1.60% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.84% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.48% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000556    12
46.04% :     0.000256s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.38% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.58% :     0.000253s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021137    76
 0.17% :     0.000037s :      1: add_recomputation
 0.33% :     0.000070s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.42% :     0.000300s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.96% :     0.005699s :      1: jit_opt_a
 0.68% :     0.000143s :      1: jit_opt_after_cconv
 0.25% :     0.000052s :      1: jit_opt_b
 2.39% :     0.000504s :      1: loop_unroll
 3.57% :     0.000755s :      1: mutable_eliminate
 4.03% :     0.000851s :     26: opt.transform.jit_opt_a
 0.22% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.42% :     0.000511s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000040s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000010s :      1: remove_dup_value
 1.87% :     0.000395s :      1: renormalize.infer
 1.46% :     0.000308s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000036s :      1: rewriter_after_opt_a
 0.37% :     0.000078s :      1: rewriter_before_opt_a
 0.36% :     0.000076s :      1: symbol_engine_optimizer
30.07% :     0.006356s :      1: task_emit
21.46% :     0.004535s :      1: type_inference
 0.23% :     0.000048s :      1: validate


TotalTime = 0.014735, [30]
[bootstrap]: 0.0002602
[type_inference]: 0.00629531
[event_method]: 2.037e-05
[auto_monad]: 7.679e-05
[graph_reusing]: 5.81e-06
[pre_auto_parallel]: 2.62001e-06
[py_interpret_to_execute]: 4.966e-05
[rewriter_before_opt_a]: 8.167e-05
[expand_dump_flag]: 2.83e-06
[jit_opt_a]: 0.00565756, [2]
    [Cycle 1]: 0.00184623, [27]
        [switch_simplify]: 0.00010901
        [loop_unroll]: 3.407e-05
        [a_1]: 0.00070428
        [with_stream_mark]: 1.489e-05
        [recompute_prepare]: 8.35001e-06
        [updatestate_depend_eliminate]: 2.86e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 1.42e-06
        [specialize_transform]: 6.27001e-06
        [updatestate_useless_node_eliminater]: 5.45001e-06
        [accelerated_algorithm]: 5.47999e-06
        [meta_shard_fg_expand]: 1.19e-06
        [get_grad_eliminate_]: 5.04e-06
        [merge_forward]: 2.68998e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.275e-05
        [j_node_and_user_rematch]: 8.82e-06
        [meta_fg_expand]: 1.68002e-06
        [replace_old_param]: 8.87e-06
        [inline_without_move]: 5.06002e-06
        [renormalize]: 0.00066606
        [add_forward_monad_depend]: 5.25001e-06
        [auto_monad_grad]: 1.99e-06
        [auto_monad_eliminator]: 1.258e-05
        [cse]: 2.067e-05
        [replace_applicator]: 1.238e-05
    [Cycle 2]: 0.00032347, [27]
        [switch_simplify]: 5.18002e-06
        [loop_unroll]: 4.99e-06
        [a_1]: 8.972e-05
        [with_stream_mark]: 1.148e-05
        [recompute_prepare]: 5.46e-06
        [updatestate_depend_eliminate]: 2.69001e-06
        [updatestate_assign_eliminate]: 2.10002e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [parameter_eliminate]: 9.99979e-07
        [specialize_transform]: 5.22999e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 4.69998e-06
        [meta_shard_fg_expand]: 1.56002e-06
        [get_grad_eliminate_]: 4.58999e-06
        [merge_forward]: 3.3e-06
        [cell_reuse_recompute_pass]: 2.06e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.229e-05
        [j_node_and_user_rematch]: 8.07e-06
        [meta_fg_expand]: 1.44998e-06
        [replace_old_param]: 7.14001e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.45001e-06
        [auto_monad_grad]: 1.04e-06
        [auto_monad_eliminator]: 5.54998e-06
        [cse]: 1.133e-05
        [replace_applicator]: 6.49999e-06
[py_interpret_to_execute_after_opt_a]: 1.164e-05
[rewriter_after_opt_a]: 3.097e-05
[convert_after_rewriter]: 6.31e-06
[order_py_execute_after_rewriter]: 4.65001e-06
[mutable_eliminate]: 0.0005661
[jit_opt_b]: 5.833e-05, [1]
    [Cycle 1]: 5.15e-05, [2]
        [frontend_op_eliminate]: 2.23e-05
        [inline_after_opt_a]: 1.625e-05
[cconv]: 2.292e-05
[loop_unroll]: 0.00047763
[jit_opt_after_cconv]: 0.00014171, [1]
    [Cycle 1]: 0.00013566, [11]
        [c_1]: 2.055e-05
        [parameter_eliminate]: 3.21001e-06
        [updatestate_depend_eliminate]: 5.51e-06
        [updatestate_assign_eliminate]: 2.76e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.93e-05
        [call_graph_tuple_transform]: 1.789e-05
        [tuple_list_get_item_eliminator]: 5.69e-06
        [none_parameter_eliminate]: 1.62001e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 5.32999e-06
[remove_dup_value]: 1.037e-05
[partial_unused_args_eliminate]: 1.84e-06
[environ_conv]: 4.63999e-06
[add_recomputation]: 3.883e-05
[cse_after_recomputation]: 2.287e-05, [1]
    [Cycle 1]: 1.658e-05, [1]
        [cse]: 9.25001e-06
[auto_monad_reorder]: 1.196e-05
[get_jit_bprop_graph]: 1.96e-06
[rewriter_after_jit_bprop_graph]: 5.70001e-06
[opt_after_jit_grad]: 0.00058446
[symbol_engine_optimizer]: 7.259e-05, [1]
    [Cycle 1]: 6.593e-05, [6]
        [build]: 3.35e-06
        [elim_shapecalc]: 7.34002e-06
        [elim_not_effective]: 1.315e-05
        [opt_reshape]: 5.51e-06
        [fold_const_symbol]: 8.18999e-06
        [renormalize]: 4.69998e-07
[validate]: 3.208e-05
Sums
    bootstrap                            :     0.000260s :  2.44%
    type_inference                       :     0.006295s : 59.11%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000077s :  0.72%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000050s :  0.47%
    rewriter_before_opt_a                :     0.000082s :  0.77%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000114s :  1.07%
    jit_opt_a.loop_unroll                :     0.000039s :  0.37%
    jit_opt_a.a_1                        :     0.000794s :  7.46%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000666s :  6.25%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.17%
    jit_opt_a.cse                        :     0.000032s :  0.30%
    jit_opt_a.replace_applicator         :     0.000019s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.11%
    rewriter_after_opt_a                 :     0.000031s :  0.29%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000566s :  5.32%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.15%
    cconv                                :     0.000023s :  0.22%
    loop_unroll                          :     0.000478s :  4.48%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000010s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000039s :  0.36%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000012s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000584s :  5.49%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.30%


Time group info:
------[substitution.]   0.000279    26
 0.66% :     0.000002s :      2: substitution.elim_not_effective
 0.38% :     0.000001s :      2: substitution.fold_const_symbol
 1.38% :     0.000004s :      3: substitution.graph_param_transform
80.72% :     0.000225s :      6: substitution.inline
 1.08% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.51% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.53% :     0.000004s :      2: substitution.replace_old_param
 3.07% :     0.000009s :      1: substitution.switch_simplify
 9.67% :     0.000027s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006232     2
82.91% :     0.005167s :      1: type_inference.infer
17.09% :     0.001065s :      1: type_inference.specialize
------[replace.]   0.000143     9
52.39% :     0.000075s :      6: replace.inline
22.63% :     0.000032s :      1: replace.switch_simplify
24.99% :     0.000036s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000255     9
87.00% :     0.000222s :      6: match.inline
 3.01% :     0.000008s :      1: match.switch_simplify
 9.98% :     0.000025s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   801
 1.33% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.21% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     12: predicate.addn_check_dump
 1.43% :     0.000002s :     12: predicate.addn_zero_filter
 1.99% :     0.000003s :     12: predicate.arithmetic_simplify
 1.29% :     0.000002s :     12: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     12: predicate.compare_switch_simplify
 1.41% :     0.000002s :     12: predicate.depend_value_elim
 1.11% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.65% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.99% :     0.000003s :     12: predicate.environ_get_eliminate
 1.37% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 5.35% :     0.000007s :     26: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.92% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 2.77% :     0.000004s :     17: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.29% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     12: predicate.merge_addn
 1.10% :     0.000001s :     12: predicate.minmaximum_grad
 1.74% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.04% :     0.000003s :     17: predicate.partial_eliminate
 1.16% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     12: predicate.reduce_eliminate
 1.42% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.82% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     20: predicate.replace_applicator
 0.44% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     12: predicate.reshape_eliminate
 1.21% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     12: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.60% :     0.000003s :     20: predicate.switch_defer_inline
 2.40% :     0.000003s :     20: predicate.switch_layer_defer_inline
 7.95% :     0.000010s :     65: predicate.switch_simplify
 1.28% :     0.000002s :     12: predicate.tile_eliminate
 1.27% :     0.000002s :     12: predicate.transpose_eliminate
 1.33% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.34% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.59% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.43% :     0.000002s :     12: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000756    15
46.60% :     0.000352s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.58% :     0.000087s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.82% :     0.000316s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016536    72
 0.25% :     0.000042s :      1: add_recomputation
 0.49% :     0.000081s :      1: auto_monad
 0.09% :     0.000015s :      1: auto_monad_reorder
 1.66% :     0.000274s :      1: bootstrap
 0.15% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
34.24% :     0.005662s :      1: jit_opt_a
 0.87% :     0.000144s :      1: jit_opt_after_cconv
 0.37% :     0.000061s :      1: jit_opt_b
 2.95% :     0.000487s :      1: loop_unroll
 3.49% :     0.000577s :      1: mutable_eliminate
 6.43% :     0.001063s :     26: opt.transform.jit_opt_a
 0.28% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000032s :      4: opt.transform.jit_opt_b
 0.08% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.59% :     0.000594s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.32% :     0.000052s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000013s :      1: remove_dup_value
 2.26% :     0.000373s :      1: renormalize.infer
 1.71% :     0.000284s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000034s :      1: rewriter_after_opt_a
 0.52% :     0.000085s :      1: rewriter_before_opt_a
 0.45% :     0.000075s :      1: symbol_engine_optimizer
38.15% :     0.006309s :      1: type_inference


TotalTime = 0.0203155, [33]
[bootstrap]: 0.00032858
[type_inference]: 0.00482302
[event_method]: 1.71e-05
[auto_monad]: 5.999e-05
[graph_reusing]: 5.22e-06
[pre_auto_parallel]: 2.06998e-06
[py_interpret_to_execute]: 3.862e-05
[rewriter_before_opt_a]: 6.962e-05
[expand_dump_flag]: 2.79999e-06
[jit_opt_a]: 0.00589998, [2]
    [Cycle 1]: 0.00178026, [27]
        [switch_simplify]: 5.389e-05
        [loop_unroll]: 2.853e-05
        [a_1]: 0.00055385
        [with_stream_mark]: 1.906e-05
        [recompute_prepare]: 8.17e-06
        [updatestate_depend_eliminate]: 3.33e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 6.03002e-06
        [updatestate_useless_node_eliminater]: 5.13002e-06
        [accelerated_algorithm]: 5.40999e-06
        [meta_shard_fg_expand]: 2.06998e-06
        [get_grad_eliminate_]: 5.52999e-06
        [merge_forward]: 3.81001e-06
        [cell_reuse_recompute_pass]: 1.21002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.327e-05
        [j_node_and_user_rematch]: 9.12001e-06
        [meta_fg_expand]: 2.30002e-06
        [replace_old_param]: 9.05001e-06
        [inline_without_move]: 5.00001e-06
        [renormalize]: 0.00078586
        [add_forward_monad_depend]: 6.71e-06
        [auto_monad_grad]: 2.51e-06
        [auto_monad_eliminator]: 1.46e-05
        [cse]: 2.779e-05
        [replace_applicator]: 1.459e-05
    [Cycle 2]: 0.00035169, [27]
        [switch_simplify]: 6.33002e-06
        [loop_unroll]: 5.06002e-06
        [a_1]: 9.726e-05
        [with_stream_mark]: 1.229e-05
        [recompute_prepare]: 4.87e-06
        [updatestate_depend_eliminate]: 2.65002e-06
        [updatestate_assign_eliminate]: 2.47001e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 5.30999e-06
        [updatestate_useless_node_eliminater]: 5.54998e-06
        [accelerated_algorithm]: 5.93002e-06
        [meta_shard_fg_expand]: 1.92001e-06
        [get_grad_eliminate_]: 5.06002e-06
        [merge_forward]: 2.88e-06
        [cell_reuse_recompute_pass]: 2.39001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.397e-05
        [j_node_and_user_rematch]: 8.72e-06
        [meta_fg_expand]: 1.82999e-06
        [replace_old_param]: 7.98999e-06
        [inline_without_move]: 4.62998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.76e-06
        [auto_monad_grad]: 1.25001e-06
        [auto_monad_eliminator]: 5.40999e-06
        [cse]: 1.297e-05
        [replace_applicator]: 5.04e-06
[py_interpret_to_execute_after_opt_a]: 1.336e-05
[rewriter_after_opt_a]: 3.564e-05
[convert_after_rewriter]: 7.39002e-06
[order_py_execute_after_rewriter]: 5.37001e-06
[mutable_eliminate]: 0.00075738
[jit_opt_b]: 5.36e-05, [1]
    [Cycle 1]: 4.507e-05, [2]
        [frontend_op_eliminate]: 1.822e-05
        [inline_after_opt_a]: 1.493e-05
[cconv]: 3.068e-05
[loop_unroll]: 0.00053921
[jit_opt_after_cconv]: 0.0001516, [1]
    [Cycle 1]: 0.00014389, [11]
        [c_1]: 2.139e-05
        [parameter_eliminate]: 3.5e-06
        [updatestate_depend_eliminate]: 6.41e-06
        [updatestate_assign_eliminate]: 2.49999e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 2.139e-05
        [call_graph_tuple_transform]: 2.224e-05
        [tuple_list_get_item_eliminator]: 5.96e-06
        [none_parameter_eliminate]: 1.71e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 5.56e-06
[remove_dup_value]: 1.351e-05
[partial_unused_args_eliminate]: 2.49001e-06
[environ_conv]: 4.97999e-06
[add_recomputation]: 4.993e-05
[cse_after_recomputation]: 2.062e-05, [1]
    [Cycle 1]: 1.439e-05, [1]
        [cse]: 8.74e-06
[auto_monad_reorder]: 1.478e-05
[get_jit_bprop_graph]: 2.23998e-06
[rewriter_after_jit_bprop_graph]: 6.63998e-06
[opt_after_jit_grad]: 0.00061005
[symbol_engine_optimizer]: 7.768e-05, [1]
    [Cycle 1]: 6.968e-05, [6]
        [build]: 4.35e-06
        [elim_shapecalc]: 7.85e-06
        [elim_not_effective]: 1.483e-05
        [opt_reshape]: 5.79999e-06
        [fold_const_symbol]: 9.06002e-06
        [renormalize]: 5.00004e-07
[validate]: 4.018e-05
[backend_pass]: 7.50006e-07
[task_emit]: 0.00640653
[execute]: 1.056e-05
Sums
    bootstrap                            :     0.000329s :  2.07%
    type_inference                       :     0.004823s : 30.34%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000060s :  0.38%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.24%
    rewriter_before_opt_a                :     0.000070s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000060s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.21%
    jit_opt_a.a_1                        :     0.000651s :  4.10%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000786s :  4.94%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000041s :  0.26%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000036s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000757s :  4.76%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000031s :  0.19%
    loop_unroll                          :     0.000539s :  3.39%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000610s :  3.84%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.25%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006407s : 40.30%
    execute                              :     0.000011s :  0.07%


Time group info:
------[substitution.]   0.000224    23
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000002s :      2: substitution.fold_const_symbol
 2.86% :     0.000006s :      3: substitution.graph_param_transform
84.71% :     0.000189s :      4: substitution.inline
 1.49% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.02% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000005s :      2: substitution.replace_old_param
 5.24% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004760     2
82.64% :     0.003934s :      1: type_inference.infer
17.36% :     0.000826s :      1: type_inference.specialize
------[replace.]   0.000069     6
67.08% :     0.000046s :      4: replace.inline
32.92% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000198     6
94.63% :     0.000187s :      4: match.inline
 5.37% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.19% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.50% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.21% :     0.000003s :     10: predicate.arithmetic_simplify
 1.43% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.25% :     0.000002s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.31% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.57% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.36% :     0.000002s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.23% :     0.000006s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.71% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.80% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 1.97% :     0.000002s :     15: predicate.partial_eliminate
 1.27% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.09% :     0.000003s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.49% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.45% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.21% :     0.000002s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.71% :     0.000008s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.74% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.90% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.97% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000623    12
47.40% :     0.000295s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.38% :     0.000046s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.22% :     0.000282s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022080    76
 0.24% :     0.000053s :      1: add_recomputation
 0.29% :     0.000063s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.55% :     0.000343s :      1: bootstrap
 0.15% :     0.000033s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.74% :     0.005904s :      1: jit_opt_a
 0.70% :     0.000155s :      1: jit_opt_after_cconv
 0.26% :     0.000056s :      1: jit_opt_b
 2.48% :     0.000548s :      1: loop_unroll
 3.48% :     0.000769s :      1: mutable_eliminate
 3.91% :     0.000864s :     26: opt.transform.jit_opt_a
 0.24% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.81% :     0.000620s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000041s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.87% :     0.000413s :      1: renormalize.infer
 1.65% :     0.000364s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000039s :      1: rewriter_after_opt_a
 0.33% :     0.000072s :      1: rewriter_before_opt_a
 0.36% :     0.000080s :      1: symbol_engine_optimizer
29.10% :     0.006425s :      1: task_emit
21.90% :     0.004835s :      1: type_inference
 0.28% :     0.000063s :      1: validate


TotalTime = 0.0183916, [33]
[bootstrap]: 0.00031117
[type_inference]: 0.00477814
[event_method]: 1.783e-05
[auto_monad]: 5.619e-05
[graph_reusing]: 5.72001e-06
[pre_auto_parallel]: 2.19999e-06
[py_interpret_to_execute]: 3.252e-05
[rewriter_before_opt_a]: 7.157e-05
[expand_dump_flag]: 3.08998e-06
[jit_opt_a]: 0.00535092, [2]
    [Cycle 1]: 0.0015597, [27]
        [switch_simplify]: 4.667e-05
        [loop_unroll]: 3.051e-05
        [a_1]: 0.00050635
        [with_stream_mark]: 1.646e-05
        [recompute_prepare]: 7.94002e-06
        [updatestate_depend_eliminate]: 3.20998e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [parameter_eliminate]: 1.54e-06
        [specialize_transform]: 6.64001e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 5.44e-06
        [merge_forward]: 2.84999e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.215e-05
        [j_node_and_user_rematch]: 9.85002e-06
        [meta_fg_expand]: 2.10002e-06
        [replace_old_param]: 8.45999e-06
        [inline_without_move]: 5.67999e-06
        [renormalize]: 0.00063054
        [add_forward_monad_depend]: 6.24999e-06
        [auto_monad_grad]: 2.51e-06
        [auto_monad_eliminator]: 1.255e-05
        [cse]: 3.338e-05
        [replace_applicator]: 1.425e-05
    [Cycle 2]: 0.00033654, [27]
        [switch_simplify]: 6.28e-06
        [loop_unroll]: 5.05001e-06
        [a_1]: 9.184e-05
        [with_stream_mark]: 1.135e-05
        [recompute_prepare]: 5.24e-06
        [updatestate_depend_eliminate]: 2.78e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 9.99979e-07
        [specialize_transform]: 5.10999e-06
        [updatestate_useless_node_eliminater]: 5.12999e-06
        [accelerated_algorithm]: 4.97e-06
        [meta_shard_fg_expand]: 1.56002e-06
        [get_grad_eliminate_]: 4.37998e-06
        [merge_forward]: 2.86e-06
        [cell_reuse_recompute_pass]: 1.79e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.297e-05
        [j_node_and_user_rematch]: 8.08999e-06
        [meta_fg_expand]: 1.89e-06
        [replace_old_param]: 7.93001e-06
        [inline_without_move]: 4.89e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.53002e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 5.27001e-06
        [cse]: 1.194e-05
        [replace_applicator]: 5.34998e-06
[py_interpret_to_execute_after_opt_a]: 1.057e-05
[rewriter_after_opt_a]: 3.401e-05
[convert_after_rewriter]: 8.28001e-06
[order_py_execute_after_rewriter]: 5.04998e-06
[mutable_eliminate]: 0.00065828
[jit_opt_b]: 4.85e-05, [1]
    [Cycle 1]: 4.169e-05, [2]
        [frontend_op_eliminate]: 1.616e-05
        [inline_after_opt_a]: 1.445e-05
[cconv]: 2.513e-05
[loop_unroll]: 0.00044289
[jit_opt_after_cconv]: 0.00014213, [1]
    [Cycle 1]: 0.00013559, [11]
        [c_1]: 2.066e-05
        [parameter_eliminate]: 2.91e-06
        [updatestate_depend_eliminate]: 5.74e-06
        [updatestate_assign_eliminate]: 2.31998e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [cse]: 2.077e-05
        [call_graph_tuple_transform]: 1.78e-05
        [tuple_list_get_item_eliminator]: 5.66e-06
        [none_parameter_eliminate]: 1.42999e-06
        [renormalize]: 4.40021e-07
        [switch_simplify]: 5.32999e-06
[remove_dup_value]: 1.12e-05
[partial_unused_args_eliminate]: 1.92999e-06
[environ_conv]: 5.25999e-06
[add_recomputation]: 4.467e-05
[cse_after_recomputation]: 2.159e-05, [1]
    [Cycle 1]: 1.56e-05, [1]
        [cse]: 9.64e-06
[auto_monad_reorder]: 1.288e-05
[get_jit_bprop_graph]: 1.87999e-06
[rewriter_after_jit_bprop_graph]: 5.74999e-06
[opt_after_jit_grad]: 0.00047207
[symbol_engine_optimizer]: 7.194e-05, [1]
    [Cycle 1]: 6.518e-05, [6]
        [build]: 2.84001e-06
        [elim_shapecalc]: 7.56999e-06
        [elim_not_effective]: 1.298e-05
        [opt_reshape]: 5.17e-06
        [fold_const_symbol]: 8.87999e-06
        [renormalize]: 3.50003e-07
[validate]: 3.093e-05
[backend_pass]: 7.50006e-07
[task_emit]: 0.00549927
[execute]: 7.38e-06
Sums
    bootstrap                            :     0.000311s :  2.17%
    type_inference                       :     0.004778s : 33.37%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000056s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.23%
    rewriter_before_opt_a                :     0.000072s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.37%
    jit_opt_a.loop_unroll                :     0.000036s :  0.25%
    jit_opt_a.a_1                        :     0.000598s :  4.18%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000631s :  4.40%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000045s :  0.32%
    jit_opt_a.replace_applicator         :     0.000020s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.24%
    convert_after_rewriter               :     0.000008s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000658s :  4.60%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.18%
    loop_unroll                          :     0.000443s :  3.09%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000011s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000045s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000013s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000472s :  3.30%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005499s : 38.41%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000178    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000001s :      2: substitution.fold_const_symbol
 2.66% :     0.000005s :      3: substitution.graph_param_transform
82.73% :     0.000147s :      4: substitution.inline
 1.72% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.61% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.58% :     0.000005s :      2: substitution.replace_old_param
 5.94% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004716     2
83.30% :     0.003928s :      1: type_inference.infer
16.70% :     0.000788s :      1: type_inference.specialize
------[replace.]   0.000064     6
64.91% :     0.000042s :      4: replace.inline
35.09% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000154     6
93.87% :     0.000145s :      4: match.inline
 6.13% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.21% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.67% :     0.000002s :     10: predicate.addn_zero_filter
 1.84% :     0.000002s :     10: predicate.arithmetic_simplify
 1.69% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.17% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.55% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.71% :     0.000007s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.90% :     0.000002s :     15: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.96% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.75% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.25% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.84% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000000s :      3: predicate.opt_reshape
 2.15% :     0.000002s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.46% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.66% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.51% :     0.000002s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.40% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.97% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000554    12
51.65% :     0.000286s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.95% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.41% :     0.000240s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019929    76
 0.24% :     0.000048s :      1: add_recomputation
 0.30% :     0.000060s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.66% :     0.000330s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.87% :     0.005354s :      1: jit_opt_a
 0.73% :     0.000145s :      1: jit_opt_after_cconv
 0.26% :     0.000051s :      1: jit_opt_b
 2.26% :     0.000451s :      1: loop_unroll
 3.36% :     0.000669s :      1: mutable_eliminate
 4.04% :     0.000805s :     26: opt.transform.jit_opt_a
 0.23% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.41% :     0.000480s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000035s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000013s :      1: remove_dup_value
 1.64% :     0.000326s :      1: renormalize.infer
 1.49% :     0.000297s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000037s :      1: rewriter_after_opt_a
 0.37% :     0.000075s :      1: rewriter_before_opt_a
 0.37% :     0.000074s :      1: symbol_engine_optimizer
27.65% :     0.005511s :      1: task_emit
24.06% :     0.004794s :      1: type_inference
 0.24% :     0.000048s :      1: validate


TotalTime = 0.0153682, [30]
[bootstrap]: 0.00029016
[type_inference]: 0.00673648
[event_method]: 1.952e-05
[auto_monad]: 8.059e-05
[graph_reusing]: 6.54001e-06
[pre_auto_parallel]: 2.31e-06
[py_interpret_to_execute]: 4.75e-05
[rewriter_before_opt_a]: 8.16e-05
[expand_dump_flag]: 2.88e-06
[jit_opt_a]: 0.00578316, [2]
    [Cycle 1]: 0.00201538, [27]
        [switch_simplify]: 0.00011795
        [loop_unroll]: 3.295e-05
        [a_1]: 0.00069523
        [with_stream_mark]: 1.946e-05
        [recompute_prepare]: 7.97e-06
        [updatestate_depend_eliminate]: 3.36999e-06
        [updatestate_assign_eliminate]: 2.71999e-06
        [updatestate_loads_eliminate]: 2.34999e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 6.34001e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.86e-06
        [meta_shard_fg_expand]: 2.12001e-06
        [get_grad_eliminate_]: 5.14e-06
        [merge_forward]: 3.63999e-06
        [cell_reuse_recompute_pass]: 1.79e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.435e-05
        [j_node_and_user_rematch]: 9.64e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 9.13002e-06
        [inline_without_move]: 5.27001e-06
        [renormalize]: 0.00077459
        [add_forward_monad_depend]: 7.13e-06
        [auto_monad_grad]: 3.53e-06
        [auto_monad_eliminator]: 1.657e-05
        [cse]: 2.41e-05
        [replace_applicator]: 2.416e-05
    [Cycle 2]: 0.00033832, [27]
        [switch_simplify]: 5.44e-06
        [loop_unroll]: 5.31002e-06
        [a_1]: 9.096e-05
        [with_stream_mark]: 1.323e-05
        [recompute_prepare]: 5.44998e-06
        [updatestate_depend_eliminate]: 2.67001e-06
        [updatestate_assign_eliminate]: 2.01998e-06
        [updatestate_loads_eliminate]: 2.28002e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 5.16002e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 5.05001e-06
        [meta_shard_fg_expand]: 1.40001e-06
        [get_grad_eliminate_]: 4.61002e-06
        [merge_forward]: 3.04001e-06
        [cell_reuse_recompute_pass]: 1.74e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.492e-05
        [j_node_and_user_rematch]: 7.75e-06
        [meta_fg_expand]: 1.65001e-06
        [replace_old_param]: 8.52e-06
        [inline_without_move]: 6.68e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.38002e-06
        [auto_monad_grad]: 1.05001e-06
        [auto_monad_eliminator]: 5.69e-06
        [cse]: 1.104e-05
        [replace_applicator]: 5.62001e-06
[py_interpret_to_execute_after_opt_a]: 1.269e-05
[rewriter_after_opt_a]: 3.472e-05
[convert_after_rewriter]: 6.56e-06
[order_py_execute_after_rewriter]: 5.24e-06
[mutable_eliminate]: 0.00072917
[jit_opt_b]: 5.119e-05, [1]
    [Cycle 1]: 4.352e-05, [2]
        [frontend_op_eliminate]: 1.702e-05
        [inline_after_opt_a]: 1.492e-05
[cconv]: 3.109e-05
[loop_unroll]: 0.00045152
[jit_opt_after_cconv]: 0.00014445, [1]
    [Cycle 1]: 0.00013765, [11]
        [c_1]: 2.041e-05
        [parameter_eliminate]: 3.55998e-06
        [updatestate_depend_eliminate]: 6.46e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [cse]: 2.049e-05
        [call_graph_tuple_transform]: 1.992e-05
        [tuple_list_get_item_eliminator]: 4.99998e-06
        [none_parameter_eliminate]: 1.79998e-06
        [renormalize]: 6.89994e-07
        [switch_simplify]: 5.15001e-06
[remove_dup_value]: 1.419e-05
[partial_unused_args_eliminate]: 2.24999e-06
[environ_conv]: 5.15999e-06
[add_recomputation]: 5.056e-05
[cse_after_recomputation]: 2.021e-05, [1]
    [Cycle 1]: 1.452e-05, [1]
        [cse]: 8.70001e-06
[auto_monad_reorder]: 1.555e-05
[get_jit_bprop_graph]: 2.04999e-06
[rewriter_after_jit_bprop_graph]: 6.63e-06
[opt_after_jit_grad]: 0.00046389
[symbol_engine_optimizer]: 7.217e-05, [1]
    [Cycle 1]: 6.494e-05, [6]
        [build]: 3.59002e-06
        [elim_shapecalc]: 7.18e-06
        [elim_not_effective]: 1.255e-05
        [opt_reshape]: 5.43002e-06
        [fold_const_symbol]: 8.31002e-06
        [renormalize]: 5.50004e-07
[validate]: 3.674e-05
Sums
    bootstrap                            :     0.000290s :  2.56%
    type_inference                       :     0.006736s : 59.48%
    event_method                         :     0.000020s :  0.17%
    auto_monad                           :     0.000081s :  0.71%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000048s :  0.42%
    rewriter_before_opt_a                :     0.000082s :  0.72%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000123s :  1.09%
    jit_opt_a.loop_unroll                :     0.000038s :  0.34%
    jit_opt_a.a_1                        :     0.000786s :  6.94%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.16%
    jit_opt_a.inline_without_move        :     0.000012s :  0.11%
    jit_opt_a.renormalize                :     0.000775s :  6.84%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.20%
    jit_opt_a.cse                        :     0.000035s :  0.31%
    jit_opt_a.replace_applicator         :     0.000030s :  0.26%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.11%
    rewriter_after_opt_a                 :     0.000035s :  0.31%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000729s :  6.44%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.13%
    cconv                                :     0.000031s :  0.27%
    loop_unroll                          :     0.000452s :  3.99%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000051s :  0.45%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000464s :  4.10%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.32%


Time group info:
------[substitution.]   0.000295    26
 0.59% :     0.000002s :      2: substitution.elim_not_effective
 0.44% :     0.000001s :      2: substitution.fold_const_symbol
 1.84% :     0.000005s :      3: substitution.graph_param_transform
85.55% :     0.000252s :      6: substitution.inline
 1.17% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.70% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.73% :     0.000005s :      2: substitution.replace_old_param
 2.74% :     0.000008s :      1: substitution.switch_simplify
 4.25% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006676     2
84.13% :     0.005617s :      1: type_inference.infer
15.87% :     0.001059s :      1: type_inference.specialize
------[replace.]   0.000130     9
48.15% :     0.000063s :      6: replace.inline
27.72% :     0.000036s :      1: replace.switch_simplify
24.13% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000268     9
92.98% :     0.000249s :      6: match.inline
 2.75% :     0.000007s :      1: match.switch_simplify
 4.26% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   801
 1.54% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.04% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     12: predicate.addn_check_dump
 1.44% :     0.000002s :     12: predicate.addn_zero_filter
 2.24% :     0.000003s :     12: predicate.arithmetic_simplify
 1.37% :     0.000002s :     12: predicate.cast_eliminate
 0.63% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.21% :     0.000002s :     12: predicate.compare_switch_simplify
 1.34% :     0.000002s :     12: predicate.depend_value_elim
 1.13% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.73% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.42% :     0.000002s :     12: predicate.environ_get_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.33% :     0.000000s :      3: predicate.graph_param_transform
 5.93% :     0.000008s :     26: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     17: predicate.load_eliminater
 1.17% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.20% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     12: predicate.merge_addn
 1.10% :     0.000001s :     12: predicate.minmaximum_grad
 1.88% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     17: predicate.partial_eliminate
 1.17% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     12: predicate.reduce_eliminate
 1.72% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     20: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.63% :     0.000002s :     12: predicate.reshape_eliminate
 1.19% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 1.11% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     12: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.80% :     0.000001s :      6: predicate.special_op_eliminate
 1.12% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.76% :     0.000004s :     20: predicate.switch_defer_inline
 2.30% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.09% :     0.000011s :     65: predicate.switch_simplify
 1.22% :     0.000002s :     12: predicate.tile_eliminate
 1.20% :     0.000002s :     12: predicate.transpose_eliminate
 1.40% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.75% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.66% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.93% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000797    15
46.34% :     0.000370s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.60% :     0.000108s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.06% :     0.000319s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017281    72
 0.31% :     0.000053s :      1: add_recomputation
 0.49% :     0.000084s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 1.75% :     0.000302s :      1: bootstrap
 0.20% :     0.000034s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.13% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.48% :     0.005786s :      1: jit_opt_a
 0.85% :     0.000147s :      1: jit_opt_after_cconv
 0.31% :     0.000054s :      1: jit_opt_b
 2.66% :     0.000460s :      1: loop_unroll
 4.28% :     0.000739s :      1: mutable_eliminate
 6.25% :     0.001081s :     26: opt.transform.jit_opt_a
 0.27% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.73% :     0.000472s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.29% :     0.000050s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.44% :     0.000421s :      1: renormalize.infer
 1.99% :     0.000345s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000038s :      1: rewriter_after_opt_a
 0.49% :     0.000084s :      1: rewriter_before_opt_a
 0.43% :     0.000075s :      1: symbol_engine_optimizer
39.06% :     0.006749s :      1: type_inference


TotalTime = 0.0175855, [30]
[bootstrap]: 0.00038117
[type_inference]: 0.00837845
[event_method]: 3.677e-05
[auto_monad]: 9.618e-05
[graph_reusing]: 6.97002e-06
[pre_auto_parallel]: 2.76999e-06
[py_interpret_to_execute]: 7.831e-05
[rewriter_before_opt_a]: 0.00010229
[expand_dump_flag]: 4.22e-06
[jit_opt_a]: 0.00603207, [2]
    [Cycle 1]: 0.00191752, [27]
        [switch_simplify]: 0.00011234
        [loop_unroll]: 3.415e-05
        [a_1]: 0.00076292
        [with_stream_mark]: 1.499e-05
        [recompute_prepare]: 7.46001e-06
        [updatestate_depend_eliminate]: 2.85998e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.09998e-06
        [specialize_transform]: 5.94e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 5.60001e-06
        [meta_shard_fg_expand]: 1.35001e-06
        [get_grad_eliminate_]: 5.52001e-06
        [merge_forward]: 2.50997e-06
        [cell_reuse_recompute_pass]: 9.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.207e-05
        [j_node_and_user_rematch]: 9.01002e-06
        [meta_fg_expand]: 2.19999e-06
        [replace_old_param]: 8.02e-06
        [inline_without_move]: 5.72999e-06
        [renormalize]: 0.00066106
        [add_forward_monad_depend]: 4.43001e-06
        [auto_monad_grad]: 3.16e-05
        [auto_monad_eliminator]: 1.176e-05
        [cse]: 1.764e-05
        [replace_applicator]: 1.268e-05
    [Cycle 2]: 0.00031532, [27]
        [switch_simplify]: 5.96998e-06
        [loop_unroll]: 4.95999e-06
        [a_1]: 9.11e-05
        [with_stream_mark]: 9.24998e-06
        [recompute_prepare]: 5.37999e-06
        [updatestate_depend_eliminate]: 2.43998e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 1.26002e-06
        [specialize_transform]: 4.87e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 4.64002e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 4.58001e-06
        [merge_forward]: 2.59999e-06
        [cell_reuse_recompute_pass]: 1.81003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.229e-05
        [j_node_and_user_rematch]: 8.15999e-06
        [meta_fg_expand]: 1.99e-06
        [replace_old_param]: 6.98e-06
        [inline_without_move]: 4.70999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 7.80012e-07
        [auto_monad_eliminator]: 4.57998e-06
        [cse]: 1.024e-05
        [replace_applicator]: 4.70999e-06
[py_interpret_to_execute_after_opt_a]: 8.72e-06
[rewriter_after_opt_a]: 2.672e-05
[convert_after_rewriter]: 5.29998e-06
[order_py_execute_after_rewriter]: 4.1e-06
[mutable_eliminate]: 0.00066782
[jit_opt_b]: 8.605e-05, [1]
    [Cycle 1]: 7.869e-05, [2]
        [frontend_op_eliminate]: 1.577e-05
        [inline_after_opt_a]: 4.9e-05
[cconv]: 1.886e-05
[loop_unroll]: 0.00048009
[jit_opt_after_cconv]: 0.0001893, [1]
    [Cycle 1]: 0.00018208, [11]
        [c_1]: 4.693e-05
        [parameter_eliminate]: 3.13e-06
        [updatestate_depend_eliminate]: 6.26e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [cse]: 2.036e-05
        [call_graph_tuple_transform]: 1.781e-05
        [tuple_list_get_item_eliminator]: 5.20999e-06
        [none_parameter_eliminate]: 1.05999e-06
        [renormalize]: 5.09986e-07
        [switch_simplify]: 5.57999e-06
[remove_dup_value]: 8.37e-06
[partial_unused_args_eliminate]: 1.77999e-06
[environ_conv]: 4.22e-06
[add_recomputation]: 3.828e-05
[cse_after_recomputation]: 2.152e-05, [1]
    [Cycle 1]: 1.578e-05, [1]
        [cse]: 1.026e-05
[auto_monad_reorder]: 9.70002e-06
[get_jit_bprop_graph]: 1.59998e-06
[rewriter_after_jit_bprop_graph]: 5.82001e-06
[opt_after_jit_grad]: 0.00059018
[symbol_engine_optimizer]: 7.661e-05, [1]
    [Cycle 1]: 6.935e-05, [6]
        [build]: 3.78001e-06
        [elim_shapecalc]: 8.25999e-06
        [elim_not_effective]: 1.347e-05
        [opt_reshape]: 5.96998e-06
        [fold_const_symbol]: 8.15e-06
        [renormalize]: 4.30009e-07
[validate]: 3.302e-05
Sums
    bootstrap                            :     0.000381s :  2.90%
    type_inference                       :     0.008378s : 63.64%
    event_method                         :     0.000037s :  0.28%
    auto_monad                           :     0.000096s :  0.73%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000078s :  0.59%
    rewriter_before_opt_a                :     0.000102s :  0.78%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000118s :  0.90%
    jit_opt_a.loop_unroll                :     0.000039s :  0.30%
    jit_opt_a.a_1                        :     0.000854s :  6.49%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000661s :  5.02%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000032s :  0.25%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.12%
    jit_opt_a.cse                        :     0.000028s :  0.21%
    jit_opt_a.replace_applicator         :     0.000017s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000027s :  0.20%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000668s :  5.07%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000049s :  0.37%
    cconv                                :     0.000019s :  0.14%
    loop_unroll                          :     0.000480s :  3.65%
    jit_opt_after_cconv.c_1              :     0.000047s :  0.36%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000008s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000038s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.08%
    auto_monad_reorder                   :     0.000010s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000590s :  4.48%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.25%


Time group info:
------[substitution.]   0.000271    26
 0.55% :     0.000002s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 1.45% :     0.000004s :      3: substitution.graph_param_transform
87.25% :     0.000236s :      6: substitution.inline
 1.07% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.66% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.14% :     0.000003s :      2: substitution.replace_old_param
 3.19% :     0.000009s :      1: substitution.switch_simplify
 3.25% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008269     2
84.98% :     0.007027s :      1: type_inference.infer
15.02% :     0.001242s :      1: type_inference.specialize
------[replace.]   0.000164     9
57.26% :     0.000094s :      6: replace.inline
21.69% :     0.000036s :      1: replace.switch_simplify
21.05% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000248     9
93.78% :     0.000233s :      6: match.inline
 3.17% :     0.000008s :      1: match.switch_simplify
 3.05% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   801
 1.30% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.11% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     12: predicate.addn_check_dump
 1.37% :     0.000002s :     12: predicate.addn_zero_filter
 2.02% :     0.000003s :     12: predicate.arithmetic_simplify
 1.35% :     0.000002s :     12: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     12: predicate.compare_switch_simplify
 1.66% :     0.000002s :     12: predicate.depend_value_elim
 1.14% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.34% :     0.000002s :     12: predicate.dict_set_item_eliminator
 1.21% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.13% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     12: predicate.environ_get_eliminate
 1.11% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.04% :     0.000007s :     26: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.07% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     17: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.17% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.52% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     12: predicate.merge_addn
 1.13% :     0.000002s :     12: predicate.minmaximum_grad
 1.97% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     17: predicate.partial_eliminate
 1.22% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.77% :     0.000002s :     12: predicate.reduce_eliminate
 1.52% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     20: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.32% :     0.000002s :     12: predicate.reshape_eliminate
 1.30% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.59% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.61% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.70% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.66% :     0.000004s :     20: predicate.switch_defer_inline
 2.46% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.08% :     0.000011s :     65: predicate.switch_simplify
 1.21% :     0.000002s :     12: predicate.tile_eliminate
 1.55% :     0.000002s :     12: predicate.transpose_eliminate
 1.51% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.60% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.43% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     12: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.43% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000922    15
50.27% :     0.000464s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.10% :     0.000102s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.63% :     0.000356s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019489    72
 0.21% :     0.000041s :      1: add_recomputation
 0.51% :     0.000100s :      1: auto_monad
 0.06% :     0.000013s :      1: auto_monad_reorder
 2.04% :     0.000397s :      1: bootstrap
 0.11% :     0.000021s :      1: cconv
 0.04% :     0.000007s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.22% :     0.000043s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000010s :      1: graph_reusing
30.98% :     0.006038s :      1: jit_opt_a
 0.99% :     0.000192s :      1: jit_opt_after_cconv
 0.46% :     0.000089s :      1: jit_opt_b
 2.51% :     0.000489s :      1: loop_unroll
 3.48% :     0.000678s :      1: mutable_eliminate
 5.77% :     0.001124s :     26: opt.transform.jit_opt_a
 0.37% :     0.000072s :      4: opt.transform.jit_opt_after_cconv
 0.30% :     0.000058s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.08% :     0.000601s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.42% :     0.000081s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000011s :      1: remove_dup_value
 1.81% :     0.000353s :      1: renormalize.infer
 1.54% :     0.000300s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000030s :      1: rewriter_after_opt_a
 0.54% :     0.000106s :      1: rewriter_before_opt_a
 0.41% :     0.000079s :      1: symbol_engine_optimizer
43.10% :     0.008399s :      1: type_inference


TotalTime = 0.019505, [33]
[bootstrap]: 0.00034951
[type_inference]: 0.00475958
[event_method]: 1.625e-05
[auto_monad]: 4.186e-05
[graph_reusing]: 4.90999e-06
[pre_auto_parallel]: 1.91e-06
[py_interpret_to_execute]: 3.287e-05
[rewriter_before_opt_a]: 6.581e-05
[expand_dump_flag]: 2.14e-06
[jit_opt_a]: 0.00543765, [2]
    [Cycle 1]: 0.00156176, [27]
        [switch_simplify]: 5.545e-05
        [loop_unroll]: 2.817e-05
        [a_1]: 0.00054029
        [with_stream_mark]: 1.874e-05
        [recompute_prepare]: 6.79999e-06
        [updatestate_depend_eliminate]: 3.65998e-06
        [updatestate_assign_eliminate]: 3.05002e-06
        [updatestate_loads_eliminate]: 2.67001e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 5.61003e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.20001e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 5.14e-06
        [merge_forward]: 4.12e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.419e-05
        [j_node_and_user_rematch]: 8.57e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 8.94e-06
        [inline_without_move]: 5.17e-06
        [renormalize]: 0.00060666
        [add_forward_monad_depend]: 5.94e-06
        [auto_monad_grad]: 2.29001e-06
        [auto_monad_eliminator]: 1.438e-05
        [cse]: 2.611e-05
        [replace_applicator]: 1.249e-05
    [Cycle 2]: 0.00031847, [27]
        [switch_simplify]: 5.11002e-06
        [loop_unroll]: 5.09e-06
        [a_1]: 9.672e-05
        [with_stream_mark]: 1.029e-05
        [recompute_prepare]: 5.27001e-06
        [updatestate_depend_eliminate]: 2.34001e-06
        [updatestate_assign_eliminate]: 1.97999e-06
        [updatestate_loads_eliminate]: 1.71e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.85999e-06
        [updatestate_useless_node_eliminater]: 4.55001e-06
        [accelerated_algorithm]: 4.70999e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 4.36002e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 1.50999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.206e-05
        [j_node_and_user_rematch]: 7.31001e-06
        [meta_fg_expand]: 1.64e-06
        [replace_old_param]: 6.58e-06
        [inline_without_move]: 4.79e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.40999e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 5.42999e-06
        [cse]: 9.84999e-06
        [replace_applicator]: 5.15999e-06
[py_interpret_to_execute_after_opt_a]: 1.063e-05
[rewriter_after_opt_a]: 3.183e-05
[convert_after_rewriter]: 6.81001e-06
[order_py_execute_after_rewriter]: 5.07999e-06
[mutable_eliminate]: 0.00063379
[jit_opt_b]: 5.653e-05, [1]
    [Cycle 1]: 4.885e-05, [2]
        [frontend_op_eliminate]: 1.6e-05
        [inline_after_opt_a]: 1.557e-05
[cconv]: 2.593e-05
[loop_unroll]: 0.00043717
[jit_opt_after_cconv]: 0.00014199, [1]
    [Cycle 1]: 0.00013498, [11]
        [c_1]: 1.978e-05
        [parameter_eliminate]: 3.08e-06
        [updatestate_depend_eliminate]: 6.51999e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [cse]: 2.003e-05
        [call_graph_tuple_transform]: 1.896e-05
        [tuple_list_get_item_eliminator]: 5.27001e-06
        [none_parameter_eliminate]: 1.54998e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.24e-06
[remove_dup_value]: 1.482e-05
[partial_unused_args_eliminate]: 2.49001e-06
[environ_conv]: 4.97e-06
[add_recomputation]: 4.832e-05
[cse_after_recomputation]: 3.515e-05, [1]
    [Cycle 1]: 1.628e-05, [1]
        [cse]: 9.82001e-06
[auto_monad_reorder]: 1.509e-05
[get_jit_bprop_graph]: 2.01003e-06
[rewriter_after_jit_bprop_graph]: 5.61e-06
[opt_after_jit_grad]: 0.00046308
[symbol_engine_optimizer]: 7.096e-05, [1]
    [Cycle 1]: 6.493e-05, [6]
        [build]: 3.87998e-06
        [elim_shapecalc]: 7.69002e-06
        [elim_not_effective]: 1.293e-05
        [opt_reshape]: 5.25999e-06
        [fold_const_symbol]: 8.02e-06
        [renormalize]: 5.19998e-07
[validate]: 3.756e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00649724
[execute]: 9.00001e-06
Sums
    bootstrap                            :     0.000350s :  2.29%
    type_inference                       :     0.004760s : 31.12%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000042s :  0.27%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000033s :  0.21%
    rewriter_before_opt_a                :     0.000066s :  0.43%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000061s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.22%
    jit_opt_a.a_1                        :     0.000637s :  4.16%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000607s :  3.97%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000036s :  0.24%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000634s :  4.14%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000026s :  0.17%
    loop_unroll                          :     0.000437s :  2.86%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000048s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000463s :  3.03%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006497s : 42.48%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000198    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.84% :     0.000006s :      3: substitution.graph_param_transform
83.50% :     0.000165s :      4: substitution.inline
 1.39% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.47% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000004s :      2: substitution.replace_old_param
 6.13% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004702     2
83.14% :     0.003910s :      1: type_inference.infer
16.86% :     0.000793s :      1: type_inference.specialize
------[replace.]   0.000059     6
66.20% :     0.000039s :      4: replace.inline
33.80% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000174     6
93.69% :     0.000163s :      4: match.inline
 6.31% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000142   691
 1.03% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.12% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.83% :     0.000001s :     10: predicate.addn_check_dump
 1.05% :     0.000001s :     10: predicate.addn_zero_filter
 1.70% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.89% :     0.000001s :     10: predicate.compare_switch_simplify
 1.04% :     0.000001s :     10: predicate.depend_value_elim
 0.89% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.09% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.89% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.92% :     0.000001s :     10: predicate.environ_get_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 4.61% :     0.000007s :     22: predicate.inline
 0.74% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.91% :     0.000001s :      6: predicate.less_batch_normalization
 1.36% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.38% :     0.000002s :     15: predicate.load_eliminater
 1.02% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.20% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.37% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 0.87% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.34% :     0.000000s :      3: predicate.opt_reshape
 1.89% :     0.000003s :     15: predicate.partial_eliminate
 0.90% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.35% :     0.000002s :     10: predicate.reduce_eliminate
 1.26% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.40% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.29% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.08% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.00% :     0.000001s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 0.77% :     0.000001s :      6: predicate.specialize_transform
 0.99% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 0.96% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 1.85% :     0.000003s :     16: predicate.switch_defer_inline
 1.76% :     0.000003s :     16: predicate.switch_layer_defer_inline
26.02% :     0.000037s :     53: predicate.switch_simplify
 0.85% :     0.000001s :     10: predicate.tile_eliminate
 0.95% :     0.000001s :     10: predicate.transpose_eliminate
 0.98% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.89% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.06% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.05% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.31% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.14% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.18% :     0.000002s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.48% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000566    12
48.79% :     0.000276s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.29% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.91% :     0.000249s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021018    76
 0.24% :     0.000051s :      1: add_recomputation
 0.22% :     0.000045s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.73% :     0.000363s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000037s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.09% :     0.000020s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.89% :     0.005441s :      1: jit_opt_a
 0.69% :     0.000145s :      1: jit_opt_after_cconv
 0.28% :     0.000059s :      1: jit_opt_b
 2.12% :     0.000445s :      1: loop_unroll
 3.06% :     0.000644s :      1: mutable_eliminate
 4.00% :     0.000840s :     26: opt.transform.jit_opt_a
 0.22% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.24% :     0.000471s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.56% :     0.000327s :      1: renormalize.infer
 1.29% :     0.000272s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000035s :      1: rewriter_after_opt_a
 0.33% :     0.000069s :      1: rewriter_before_opt_a
 0.35% :     0.000073s :      1: symbol_engine_optimizer
31.00% :     0.006516s :      1: task_emit
22.70% :     0.004771s :      1: type_inference
 0.28% :     0.000058s :      1: validate


TotalTime = 0.626518, [33]
[bootstrap]: 0.00044917
[type_inference]: 0.0113233
[event_method]: 1.706e-05
[auto_monad]: 0.00013089
[graph_reusing]: 6.12999e-06
[pre_auto_parallel]: 1.041e-05
[py_interpret_to_execute]: 5.56e-05
[rewriter_before_opt_a]: 0.00011389
[expand_dump_flag]: 3.71001e-06
[jit_opt_a]: 0.0143549, [2]
    [Cycle 1]: 0.00215354, [27]
        [switch_simplify]: 9.043e-05
        [loop_unroll]: 3.237e-05
        [a_1]: 0.00053985
        [with_stream_mark]: 1.435e-05
        [recompute_prepare]: 7.71999e-06
        [updatestate_depend_eliminate]: 4.02e-06
        [updatestate_assign_eliminate]: 2.54001e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.35999e-06
        [specialize_transform]: 6.81999e-06
        [updatestate_useless_node_eliminater]: 5.32999e-06
        [accelerated_algorithm]: 1.002e-05
        [meta_shard_fg_expand]: 1.03e-05
        [get_grad_eliminate_]: 6.06e-06
        [merge_forward]: 2.93e-06
        [cell_reuse_recompute_pass]: 1.12999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.41e-05
        [j_node_and_user_rematch]: 1.005e-05
        [meta_fg_expand]: 2.40002e-06
        [replace_old_param]: 9.82999e-06
        [inline_without_move]: 5.77999e-06
        [renormalize]: 0.00109993
        [add_forward_monad_depend]: 3.433e-05
        [auto_monad_grad]: 2.26e-06
        [auto_monad_eliminator]: 2.179e-05
        [cse]: 2.705e-05
        [replace_applicator]: 1.132e-05
    [Cycle 2]: 0.00039912, [27]
        [switch_simplify]: 6.52001e-06
        [loop_unroll]: 5.34e-06
        [a_1]: 9.325e-05
        [with_stream_mark]: 1.029e-05
        [recompute_prepare]: 5.27001e-06
        [updatestate_depend_eliminate]: 6.242e-05
        [updatestate_assign_eliminate]: 2.30002e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.35001e-06
        [specialize_transform]: 5.74e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.21997e-06
        [get_grad_eliminate_]: 4.85999e-06
        [merge_forward]: 2.93998e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.273e-05
        [j_node_and_user_rematch]: 8.94e-06
        [meta_fg_expand]: 1.73002e-06
        [replace_old_param]: 7.88001e-06
        [inline_without_move]: 4.85999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.16997e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 4.92e-06
        [cse]: 1.167e-05
        [replace_applicator]: 1.144e-05
[py_interpret_to_execute_after_opt_a]: 8.74e-06
[rewriter_after_opt_a]: 4.613e-05
[convert_after_rewriter]: 7.3e-06
[order_py_execute_after_rewriter]: 5.17999e-06
[mutable_eliminate]: 0.00057359
[jit_opt_b]: 5.017e-05, [1]
    [Cycle 1]: 4.3e-05, [2]
        [frontend_op_eliminate]: 1.632e-05
        [inline_after_opt_a]: 1.514e-05
[cconv]: 2.284e-05
[loop_unroll]: 0.00043139
[jit_opt_after_cconv]: 0.00013965, [1]
    [Cycle 1]: 0.00013346, [11]
        [c_1]: 2.094e-05
        [parameter_eliminate]: 2.57001e-06
        [updatestate_depend_eliminate]: 5.85002e-06
        [updatestate_assign_eliminate]: 2.64999e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [cse]: 1.777e-05
        [call_graph_tuple_transform]: 1.805e-05
        [tuple_list_get_item_eliminator]: 5.40001e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 4.20026e-07
        [switch_simplify]: 5.77999e-06
[remove_dup_value]: 1.345e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 3.175e-05
[add_recomputation]: 5.582e-05
[cse_after_recomputation]: 2.386e-05, [1]
    [Cycle 1]: 1.805e-05, [1]
        [cse]: 1.097e-05
[auto_monad_reorder]: 2.212e-05
[get_jit_bprop_graph]: 1.50001e-06
[rewriter_after_jit_bprop_graph]: 0.00015304
[opt_after_jit_grad]: 0.00047792
[symbol_engine_optimizer]: 7.673e-05, [1]
    [Cycle 1]: 7.034e-05, [6]
        [build]: 3.11999e-06
        [elim_shapecalc]: 7.91001e-06
        [elim_not_effective]: 1.377e-05
        [opt_reshape]: 5.84e-06
        [fold_const_symbol]: 9.31e-06
        [renormalize]: 4.50003e-07
[validate]: 8.734e-05
[backend_pass]: 1.17e-06
[task_emit]: 0.597533
[execute]: 9.86998e-06
Sums
    bootstrap                            :     0.000449s :  0.07%
    type_inference                       :     0.011323s :  1.84%
    event_method                         :     0.000017s :  0.00%
    auto_monad                           :     0.000131s :  0.02%
    graph_reusing                        :     0.000006s :  0.00%
    pre_auto_parallel                    :     0.000010s :  0.00%
    py_interpret_to_execute              :     0.000056s :  0.01%
    rewriter_before_opt_a                :     0.000114s :  0.02%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000097s :  0.02%
    jit_opt_a.loop_unroll                :     0.000038s :  0.01%
    jit_opt_a.a_1                        :     0.000633s :  0.10%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.00%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.00%
    jit_opt_a.updatestate_depend_eliminate :     0.000066s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.00%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.00%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.00%
    jit_opt_a.specialize_transform       :     0.000013s :  0.00%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.00%
    jit_opt_a.accelerated_algorithm      :     0.000015s :  0.00%
    jit_opt_a.meta_shard_fg_expand       :     0.000012s :  0.00%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.00%
    jit_opt_a.merge_forward              :     0.000006s :  0.00%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.00%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.00%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.00%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.00%
    jit_opt_a.replace_old_param          :     0.000018s :  0.00%
    jit_opt_a.inline_without_move        :     0.000011s :  0.00%
    jit_opt_a.renormalize                :     0.001100s :  0.18%
    jit_opt_a.add_forward_monad_depend   :     0.000035s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.00%
    jit_opt_a.cse                        :     0.000039s :  0.01%
    jit_opt_a.replace_applicator         :     0.000023s :  0.00%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.00%
    rewriter_after_opt_a                 :     0.000046s :  0.01%
    convert_after_rewriter               :     0.000007s :  0.00%
    order_py_execute_after_rewriter      :     0.000005s :  0.00%
    mutable_eliminate                    :     0.000574s :  0.09%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.00%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.00%
    cconv                                :     0.000023s :  0.00%
    loop_unroll                          :     0.000431s :  0.07%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.00%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.00%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000018s :  0.00%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.00%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.00%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.00%
    remove_dup_value                     :     0.000013s :  0.00%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000032s :  0.01%
    add_recomputation                    :     0.000056s :  0.01%
    cse_after_recomputation.cse          :     0.000011s :  0.00%
    auto_monad_reorder                   :     0.000022s :  0.00%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000153s :  0.02%
    opt_after_jit_grad                   :     0.000478s :  0.08%
    symbol_engine_optimizer.build        :     0.000003s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.00%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.00%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.00%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.00%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000087s :  0.01%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.597533s : 97.32%
    execute                              :     0.000010s :  0.00%


Time group info:
------[substitution.]   0.000193    23
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.66% :     0.000005s :      3: substitution.graph_param_transform
83.17% :     0.000161s :      4: substitution.inline
 2.05% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.32% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.22% :     0.000004s :      2: substitution.replace_old_param
 5.81% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011254     2
90.30% :     0.010162s :      1: type_inference.infer
 9.70% :     0.001092s :      1: type_inference.specialize
------[replace.]   0.000067     6
65.61% :     0.000044s :      4: replace.inline
34.39% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000168     6
93.94% :     0.000158s :      4: match.inline
 6.06% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 2.20% :     0.000003s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.72% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.51% :     0.000001s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 4.87% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.62% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.85% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.20% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.83% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.38% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000002s :     10: predicate.reshape_eliminate
 1.39% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.14% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.54% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.93% :     0.000010s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.80% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000748    12
46.15% :     0.000345s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 3.49% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.36% :     0.000377s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.628593    76
 0.01% :     0.000059s :      1: add_recomputation
 0.02% :     0.000135s :      1: auto_monad
 0.00% :     0.000025s :      1: auto_monad_reorder
 0.00% :     0.000005s :      1: backend_pass
 0.08% :     0.000474s :      1: bootstrap
 0.00% :     0.000025s :      1: cconv
 0.00% :     0.000010s :      1: convert_after_rewriter
 0.00% :     0.000026s :      1: cse_after_recomputation
 0.01% :     0.000035s :      1: environ_conv
 0.00% :     0.000021s :      1: event_method
 0.00% :     0.000015s :      1: execute
 0.00% :     0.000006s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.00% :     0.000009s :      1: graph_reusing
 2.28% :     0.014358s :      1: jit_opt_a
 0.02% :     0.000143s :      1: jit_opt_after_cconv
 0.01% :     0.000053s :      1: jit_opt_b
 0.07% :     0.000439s :      1: loop_unroll
 0.09% :     0.000582s :      1: mutable_eliminate
 0.14% :     0.000886s :     26: opt.transform.jit_opt_a
 0.01% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.00% :     0.000025s :      4: opt.transform.jit_opt_b
 0.00% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.00% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.00% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.01% :     0.000033s :      4: opt.transform.symbol_engine_opt
 0.08% :     0.000487s :      1: opt_after_jit_grad
 0.00% :     0.000007s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.00% :     0.000013s :      1: pre_auto_parallel
 0.01% :     0.000058s :      1: py_interpret_to_execute
 0.00% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.00% :     0.000016s :      1: remove_dup_value
 0.12% :     0.000779s :      1: renormalize.infer
 0.05% :     0.000314s :      1: renormalize.specialize
 0.02% :     0.000157s :      1: rewriter_after_jit_bprop_graph
 0.01% :     0.000050s :      1: rewriter_after_opt_a
 0.02% :     0.000117s :      1: rewriter_before_opt_a
 0.01% :     0.000080s :      1: symbol_engine_optimizer
95.06% :     0.597557s :      1: task_emit
 1.80% :     0.011338s :      1: type_inference
 0.02% :     0.000144s :      1: validate


TotalTime = 0.0170416, [30]
[bootstrap]: 0.00040245
[type_inference]: 0.00751584
[event_method]: 6.779e-05
[auto_monad]: 8.813e-05
[graph_reusing]: 6.53e-06
[pre_auto_parallel]: 3.06001e-06
[py_interpret_to_execute]: 5.56e-05
[rewriter_before_opt_a]: 8.782e-05
[expand_dump_flag]: 3.4e-06
[jit_opt_a]: 0.00637691, [2]
    [Cycle 1]: 0.0021046, [27]
        [switch_simplify]: 0.00013334
        [loop_unroll]: 3.396e-05
        [a_1]: 0.00077335
        [with_stream_mark]: 2.209e-05
        [recompute_prepare]: 7.97e-06
        [updatestate_depend_eliminate]: 3.76001e-06
        [updatestate_assign_eliminate]: 3.21001e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 5.95002e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 2.40002e-06
        [get_grad_eliminate_]: 5.26998e-06
        [merge_forward]: 4.02002e-06
        [cell_reuse_recompute_pass]: 1.12999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.511e-05
        [j_node_and_user_rematch]: 1.028e-05
        [meta_fg_expand]: 2.47001e-06
        [replace_old_param]: 9.27001e-06
        [inline_without_move]: 5.71e-06
        [renormalize]: 0.00078621
        [add_forward_monad_depend]: 6.86001e-06
        [auto_monad_grad]: 3.19001e-06
        [auto_monad_eliminator]: 1.509e-05
        [cse]: 3.154e-05
        [replace_applicator]: 1.488e-05
    [Cycle 2]: 0.00032, [27]
        [switch_simplify]: 5.74e-06
        [loop_unroll]: 4.86002e-06
        [a_1]: 9.1e-05
        [with_stream_mark]: 1.066e-05
        [recompute_prepare]: 5.10999e-06
        [updatestate_depend_eliminate]: 2.54999e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 4.62e-06
        [accelerated_algorithm]: 4.77e-06
        [meta_shard_fg_expand]: 1.60999e-06
        [get_grad_eliminate_]: 4.27998e-06
        [merge_forward]: 2.64001e-06
        [cell_reuse_recompute_pass]: 1.84998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.468e-05
        [j_node_and_user_rematch]: 7.91001e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 7.39002e-06
        [inline_without_move]: 4.80999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.33002e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 5.04998e-06
        [cse]: 1.012e-05
        [replace_applicator]: 5.13002e-06
[py_interpret_to_execute_after_opt_a]: 1.263e-05
[rewriter_after_opt_a]: 3.622e-05
[convert_after_rewriter]: 7.28e-06
[order_py_execute_after_rewriter]: 4.63999e-06
[mutable_eliminate]: 0.00069394
[jit_opt_b]: 5.058e-05, [1]
    [Cycle 1]: 4.338e-05, [2]
        [frontend_op_eliminate]: 1.605e-05
        [inline_after_opt_a]: 1.546e-05
[cconv]: 2.884e-05
[loop_unroll]: 0.00045067
[jit_opt_after_cconv]: 0.00022774, [1]
    [Cycle 1]: 0.00022058, [11]
        [c_1]: 2.245e-05
        [parameter_eliminate]: 3.16001e-06
        [updatestate_depend_eliminate]: 7.33999e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.39001e-06
        [cse]: 2.466e-05
        [call_graph_tuple_transform]: 2.008e-05
        [tuple_list_get_item_eliminator]: 5.57001e-06
        [none_parameter_eliminate]: 1.94e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 5.49e-06
[remove_dup_value]: 1.604e-05
[partial_unused_args_eliminate]: 2.36998e-06
[environ_conv]: 5.56002e-06
[add_recomputation]: 5.308e-05
[cse_after_recomputation]: 2.166e-05, [1]
    [Cycle 1]: 1.553e-05, [1]
        [cse]: 9.51e-06
[auto_monad_reorder]: 1.69e-05
[get_jit_bprop_graph]: 2.01e-06
[rewriter_after_jit_bprop_graph]: 7.1e-06
[opt_after_jit_grad]: 0.0004879
[symbol_engine_optimizer]: 7.701e-05, [1]
    [Cycle 1]: 6.989e-05, [6]
        [build]: 3.85e-06
        [elim_shapecalc]: 7.66999e-06
        [elim_not_effective]: 1.421e-05
        [opt_reshape]: 5.36998e-06
        [fold_const_symbol]: 9.10999e-06
        [renormalize]: 4.09986e-07
[validate]: 3.661e-05
Sums
    bootstrap                            :     0.000402s :  3.25%
    type_inference                       :     0.007516s : 60.66%
    event_method                         :     0.000068s :  0.55%
    auto_monad                           :     0.000088s :  0.71%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000056s :  0.45%
    rewriter_before_opt_a                :     0.000088s :  0.71%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000139s :  1.12%
    jit_opt_a.loop_unroll                :     0.000039s :  0.31%
    jit_opt_a.a_1                        :     0.000864s :  6.98%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.13%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.000786s :  6.35%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.16%
    jit_opt_a.cse                        :     0.000042s :  0.34%
    jit_opt_a.replace_applicator         :     0.000020s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.10%
    rewriter_after_opt_a                 :     0.000036s :  0.29%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000694s :  5.60%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000029s :  0.23%
    loop_unroll                          :     0.000451s :  3.64%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000025s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000016s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000053s :  0.43%
    cse_after_recomputation.cse          :     0.000010s :  0.08%
    auto_monad_reorder                   :     0.000017s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000488s :  3.94%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.30%


Time group info:
------[substitution.]   0.000353    26
 0.63% :     0.000002s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 1.59% :     0.000006s :      3: substitution.graph_param_transform
86.81% :     0.000306s :      6: substitution.inline
 1.14% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.48% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.21% :     0.000004s :      2: substitution.replace_old_param
 2.86% :     0.000010s :      1: substitution.switch_simplify
 3.86% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007438     2
84.65% :     0.006296s :      1: type_inference.infer
15.35% :     0.001142s :      1: type_inference.specialize
------[replace.]   0.000148     9
48.26% :     0.000072s :      6: replace.inline
27.97% :     0.000041s :      1: replace.switch_simplify
23.77% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000325     9
93.31% :     0.000303s :      6: match.inline
 2.87% :     0.000009s :      1: match.switch_simplify
 3.82% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   801
 1.45% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.21% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     12: predicate.addn_check_dump
 1.32% :     0.000002s :     12: predicate.addn_zero_filter
 1.91% :     0.000003s :     12: predicate.arithmetic_simplify
 1.48% :     0.000002s :     12: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     12: predicate.compare_switch_simplify
 1.49% :     0.000002s :     12: predicate.depend_value_elim
 1.11% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.36% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     12: predicate.environ_get_eliminate
 1.11% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.81% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 6.14% :     0.000008s :     26: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.99% :     0.000001s :      6: predicate.less_batch_normalization
 1.80% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     17: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.21% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.86% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     12: predicate.merge_addn
 1.12% :     0.000002s :     12: predicate.minmaximum_grad
 1.68% :     0.000002s :      3: predicate.mutable_eliminate
 0.39% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     17: predicate.partial_eliminate
 1.17% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     12: predicate.reduce_eliminate
 1.46% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     20: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.20% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     12: predicate.reshape_eliminate
 1.35% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000002s :     12: predicate.same_eliminate
 0.64% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 0.85% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.62% :     0.000004s :     20: predicate.switch_defer_inline
 2.50% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.44% :     0.000011s :     65: predicate.switch_simplify
 1.31% :     0.000002s :     12: predicate.tile_eliminate
 1.68% :     0.000002s :     12: predicate.transpose_eliminate
 1.56% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.60% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.85% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000002s :     12: predicate.value_based_eliminate
 0.27% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.53% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000893    15
48.08% :     0.000429s :      6: func_graph_cloner_run.FuncGraphClonerGraph
14.25% :     0.000127s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.67% :     0.000336s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019044    72
 0.30% :     0.000056s :      1: add_recomputation
 0.48% :     0.000092s :      1: auto_monad
 0.10% :     0.000019s :      1: auto_monad_reorder
 2.19% :     0.000418s :      1: bootstrap
 0.17% :     0.000031s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.39% :     0.000073s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.51% :     0.006381s :      1: jit_opt_a
 1.21% :     0.000231s :      1: jit_opt_after_cconv
 0.28% :     0.000053s :      1: jit_opt_b
 2.47% :     0.000471s :      1: loop_unroll
 3.70% :     0.000705s :      1: mutable_eliminate
 6.09% :     0.001159s :     26: opt.transform.jit_opt_a
 0.26% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.61% :     0.000497s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000058s :      1: py_interpret_to_execute
 0.08% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000018s :      1: remove_dup_value
 2.36% :     0.000450s :      1: renormalize.infer
 1.72% :     0.000327s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000039s :      1: rewriter_after_opt_a
 0.48% :     0.000091s :      1: rewriter_before_opt_a
 0.42% :     0.000080s :      1: symbol_engine_optimizer
39.58% :     0.007537s :      1: type_inference


TotalTime = 0.0211927, [33]
[bootstrap]: 0.00038122
[type_inference]: 0.0047772
[event_method]: 1.674e-05
[auto_monad]: 6.401e-05
[graph_reusing]: 5.97001e-06
[pre_auto_parallel]: 2.77002e-06
[py_interpret_to_execute]: 3.455e-05
[rewriter_before_opt_a]: 7.075e-05
[expand_dump_flag]: 3.14999e-06
[jit_opt_a]: 0.0064259, [2]
    [Cycle 1]: 0.00180356, [27]
        [switch_simplify]: 6.063e-05
        [loop_unroll]: 3.054e-05
        [a_1]: 0.0005944
        [with_stream_mark]: 2.033e-05
        [recompute_prepare]: 8.07003e-06
        [updatestate_depend_eliminate]: 4.73001e-06
        [updatestate_assign_eliminate]: 2.90998e-06
        [updatestate_loads_eliminate]: 3.04999e-06
        [parameter_eliminate]: 1.92001e-06
        [specialize_transform]: 6.61e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 2.04e-06
        [get_grad_eliminate_]: 5.08002e-06
        [merge_forward]: 3.88001e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.43e-05
        [j_node_and_user_rematch]: 9.66e-06
        [meta_fg_expand]: 2.26998e-06
        [replace_old_param]: 9.47001e-06
        [inline_without_move]: 5.05001e-06
        [renormalize]: 0.00075703
        [add_forward_monad_depend]: 1.131e-05
        [auto_monad_grad]: 1.84e-06
        [auto_monad_eliminator]: 1.197e-05
        [cse]: 2.23e-05
        [replace_applicator]: 1.31e-05
    [Cycle 2]: 0.00035497, [27]
        [switch_simplify]: 6.41e-06
        [loop_unroll]: 5.94999e-06
        [a_1]: 0.00010227
        [with_stream_mark]: 1.278e-05
        [recompute_prepare]: 6.06e-06
        [updatestate_depend_eliminate]: 3.13e-06
        [updatestate_assign_eliminate]: 5.66e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [parameter_eliminate]: 1.25999e-06
        [specialize_transform]: 5.18002e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 4.89998e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.70997e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.564e-05
        [j_node_and_user_rematch]: 8.44998e-06
        [meta_fg_expand]: 1.69998e-06
        [replace_old_param]: 8.65001e-06
        [inline_without_move]: 4.82e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 5.86998e-06
        [cse]: 1.219e-05
        [replace_applicator]: 4.98001e-06
[py_interpret_to_execute_after_opt_a]: 1.383e-05
[rewriter_after_opt_a]: 3.73e-05
[convert_after_rewriter]: 1.335e-05
[order_py_execute_after_rewriter]: 5.58002e-06
[mutable_eliminate]: 0.00069188
[jit_opt_b]: 5.493e-05, [1]
    [Cycle 1]: 4.622e-05, [2]
        [frontend_op_eliminate]: 1.836e-05
        [inline_after_opt_a]: 1.583e-05
[cconv]: 2.746e-05
[loop_unroll]: 0.00051615
[jit_opt_after_cconv]: 0.0001505, [1]
    [Cycle 1]: 0.00014327, [11]
        [c_1]: 2.356e-05
        [parameter_eliminate]: 3.48999e-06
        [updatestate_depend_eliminate]: 7.38999e-06
        [updatestate_assign_eliminate]: 2.79999e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [cse]: 2.121e-05
        [call_graph_tuple_transform]: 1.9e-05
        [tuple_list_get_item_eliminator]: 5.40999e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 8.39995e-07
        [switch_simplify]: 5.64e-06
[remove_dup_value]: 1.322e-05
[partial_unused_args_eliminate]: 2.51998e-06
[environ_conv]: 5.08002e-06
[add_recomputation]: 5.035e-05
[cse_after_recomputation]: 2.302e-05, [1]
    [Cycle 1]: 1.631e-05, [1]
        [cse]: 9.92999e-06
[auto_monad_reorder]: 1.464e-05
[get_jit_bprop_graph]: 2.01998e-06
[rewriter_after_jit_bprop_graph]: 7.16999e-06
[opt_after_jit_grad]: 0.00053846
[symbol_engine_optimizer]: 7.415e-05, [1]
    [Cycle 1]: 6.779e-05, [6]
        [build]: 4.27998e-06
        [elim_shapecalc]: 7.85e-06
        [elim_not_effective]: 1.3e-05
        [opt_reshape]: 5.59998e-06
        [fold_const_symbol]: 8.69e-06
        [renormalize]: 2.59985e-07
[validate]: 3.881e-05
[backend_pass]: 1.04998e-06
[task_emit]: 0.00689281
[execute]: 8.45001e-06
Sums
    bootstrap                            :     0.000381s :  2.34%
    type_inference                       :     0.004777s : 29.37%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000064s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.21%
    rewriter_before_opt_a                :     0.000071s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.41%
    jit_opt_a.loop_unroll                :     0.000036s :  0.22%
    jit_opt_a.a_1                        :     0.000697s :  4.28%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000757s :  4.66%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.11%
    jit_opt_a.cse                        :     0.000034s :  0.21%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.09%
    rewriter_after_opt_a                 :     0.000037s :  0.23%
    convert_after_rewriter               :     0.000013s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000692s :  4.25%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000027s :  0.17%
    loop_unroll                          :     0.000516s :  3.17%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000538s :  3.31%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000039s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006893s : 42.38%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000218    23
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.32% :     0.000005s :      3: substitution.graph_param_transform
84.43% :     0.000184s :      4: substitution.inline
 1.71% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.46% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.04% :     0.000004s :      2: substitution.replace_old_param
 5.52% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004710     2
82.33% :     0.003878s :      1: type_inference.infer
17.67% :     0.000832s :      1: type_inference.specialize
------[replace.]   0.000091     6
54.76% :     0.000050s :      4: replace.inline
45.24% :     0.000041s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     6
94.44% :     0.000181s :      4: match.inline
 5.56% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.39% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.05% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.43% :     0.000002s :     10: predicate.addn_zero_filter
 2.21% :     0.000003s :     10: predicate.arithmetic_simplify
 1.16% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.19% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.53% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.33% :     0.000007s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.18% :     0.000001s :      6: predicate.less_batch_normalization
 1.72% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.33% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.93% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     10: predicate.reduce_eliminate
 1.36% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.31% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.22% :     0.000002s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.82% :     0.000010s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.24% :     0.000002s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.36% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000557    12
45.42% :     0.000253s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.63% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.95% :     0.000267s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022979    76
 0.23% :     0.000054s :      1: add_recomputation
 0.29% :     0.000067s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.73% :     0.000398s :      1: bootstrap
 0.13% :     0.000030s :      1: cconv
 0.07% :     0.000016s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.09% :     0.000020s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.98% :     0.006430s :      1: jit_opt_a
 0.67% :     0.000153s :      1: jit_opt_after_cconv
 0.25% :     0.000058s :      1: jit_opt_b
 2.29% :     0.000527s :      1: loop_unroll
 3.06% :     0.000704s :      1: mutable_eliminate
 4.01% :     0.000921s :     26: opt.transform.jit_opt_a
 0.22% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000028s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.39% :     0.000549s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000037s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.84% :     0.000422s :      1: renormalize.infer
 1.42% :     0.000326s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000040s :      1: rewriter_after_opt_a
 0.32% :     0.000074s :      1: rewriter_before_opt_a
 0.33% :     0.000077s :      1: symbol_engine_optimizer
30.07% :     0.006911s :      1: task_emit
20.84% :     0.004789s :      1: type_inference
 0.31% :     0.000071s :      1: validate


TotalTime = 0.0169334, [30]
[bootstrap]: 0.00034035
[type_inference]: 0.00782023
[event_method]: 2.189e-05
[auto_monad]: 8.77e-05
[graph_reusing]: 6.44001e-06
[pre_auto_parallel]: 2.59999e-06
[py_interpret_to_execute]: 6.085e-05
[rewriter_before_opt_a]: 8.931e-05
[expand_dump_flag]: 3.11001e-06
[jit_opt_a]: 0.00594481, [2]
    [Cycle 1]: 0.00200518, [27]
        [switch_simplify]: 0.00013029
        [loop_unroll]: 3.498e-05
        [a_1]: 0.00071298
        [with_stream_mark]: 2.054e-05
        [recompute_prepare]: 7.77998e-06
        [updatestate_depend_eliminate]: 3.78001e-06
        [updatestate_assign_eliminate]: 3.14001e-06
        [updatestate_loads_eliminate]: 2.86999e-06
        [parameter_eliminate]: 2.17999e-06
        [specialize_transform]: 5.96998e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.82001e-06
        [meta_shard_fg_expand]: 2.11e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 3.88001e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.504e-05
        [j_node_and_user_rematch]: 9.54e-06
        [meta_fg_expand]: 2.41998e-06
        [replace_old_param]: 8.99e-06
        [inline_without_move]: 5.65001e-06
        [renormalize]: 0.00076074
        [add_forward_monad_depend]: 6.31e-06
        [auto_monad_grad]: 2.86999e-06
        [auto_monad_eliminator]: 1.601e-05
        [cse]: 2.89e-05
        [replace_applicator]: 1.36e-05
    [Cycle 2]: 0.00032651, [27]
        [switch_simplify]: 5.79999e-06
        [loop_unroll]: 4.68999e-06
        [a_1]: 9.472e-05
        [with_stream_mark]: 1.038e-05
        [recompute_prepare]: 4.92e-06
        [updatestate_depend_eliminate]: 2.81e-06
        [updatestate_assign_eliminate]: 2.38002e-06
        [updatestate_loads_eliminate]: 2.74999e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 5.14e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.07999e-06
        [meta_shard_fg_expand]: 1.66e-06
        [get_grad_eliminate_]: 4.55999e-06
        [merge_forward]: 2.63003e-06
        [cell_reuse_recompute_pass]: 1.65001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.381e-05
        [j_node_and_user_rematch]: 7.99002e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 7.56999e-06
        [inline_without_move]: 4.88001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.40001e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 4.72e-06
        [cse]: 9.74e-06
        [replace_applicator]: 5.10001e-06
[py_interpret_to_execute_after_opt_a]: 1.365e-05
[rewriter_after_opt_a]: 3.586e-05
[convert_after_rewriter]: 7.26001e-06
[order_py_execute_after_rewriter]: 5.08002e-06
[mutable_eliminate]: 0.0007466
[jit_opt_b]: 5.482e-05, [1]
    [Cycle 1]: 4.605e-05, [2]
        [frontend_op_eliminate]: 1.807e-05
        [inline_after_opt_a]: 1.521e-05
[cconv]: 2.641e-05
[loop_unroll]: 0.00051404
[jit_opt_after_cconv]: 0.0001599, [1]
    [Cycle 1]: 0.00015267, [11]
        [c_1]: 2.434e-05
        [parameter_eliminate]: 3.08e-06
        [updatestate_depend_eliminate]: 7.46999e-06
        [updatestate_assign_eliminate]: 2.71e-06
        [updatestate_loads_eliminate]: 2.88e-06
        [cse]: 2.543e-05
        [call_graph_tuple_transform]: 2.176e-05
        [tuple_list_get_item_eliminator]: 5.44998e-06
        [none_parameter_eliminate]: 1.71e-06
        [renormalize]: 6.69999e-07
        [switch_simplify]: 5.44998e-06
[remove_dup_value]: 1.545e-05
[partial_unused_args_eliminate]: 2.46e-06
[environ_conv]: 6.16e-06
[add_recomputation]: 5.176e-05
[cse_after_recomputation]: 2.15e-05, [1]
    [Cycle 1]: 1.569e-05, [1]
        [cse]: 9.60001e-06
[auto_monad_reorder]: 1.62e-05
[get_jit_bprop_graph]: 1.71e-06
[rewriter_after_jit_bprop_graph]: 7.2e-06
[opt_after_jit_grad]: 0.00054538
[symbol_engine_optimizer]: 7.9e-05, [1]
    [Cycle 1]: 7.145e-05, [6]
        [build]: 5.07e-06
        [elim_shapecalc]: 8.52e-06
        [elim_not_effective]: 1.36e-05
        [opt_reshape]: 6.04999e-06
        [fold_const_symbol]: 8.42998e-06
        [renormalize]: 8.2e-07
[validate]: 4.228e-05
Sums
    bootstrap                            :     0.000340s :  2.68%
    type_inference                       :     0.007820s : 61.64%
    event_method                         :     0.000022s :  0.17%
    auto_monad                           :     0.000088s :  0.69%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000061s :  0.48%
    rewriter_before_opt_a                :     0.000089s :  0.70%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000136s :  1.07%
    jit_opt_a.loop_unroll                :     0.000040s :  0.31%
    jit_opt_a.a_1                        :     0.000808s :  6.37%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.13%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.000761s :  6.00%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.16%
    jit_opt_a.cse                        :     0.000039s :  0.30%
    jit_opt_a.replace_applicator         :     0.000019s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.11%
    rewriter_after_opt_a                 :     0.000036s :  0.28%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000747s :  5.88%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000026s :  0.21%
    loop_unroll                          :     0.000514s :  4.05%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000025s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000015s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000052s :  0.41%
    cse_after_recomputation.cse          :     0.000010s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000545s :  4.30%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000042s :  0.33%


Time group info:
------[substitution.]   0.000294    26
 0.69% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000002s :      2: substitution.fold_const_symbol
 2.15% :     0.000006s :      3: substitution.graph_param_transform
84.51% :     0.000248s :      6: substitution.inline
 1.12% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.73% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.46% :     0.000004s :      2: substitution.replace_old_param
 3.27% :     0.000010s :      1: substitution.switch_simplify
 4.54% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007746     2
84.19% :     0.006522s :      1: type_inference.infer
15.81% :     0.001224s :      1: type_inference.specialize
------[replace.]   0.000146     9
47.47% :     0.000070s :      6: replace.inline
28.53% :     0.000042s :      1: replace.switch_simplify
24.00% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000266     9
92.17% :     0.000245s :      6: match.inline
 3.33% :     0.000009s :      1: match.switch_simplify
 4.49% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   801
 1.42% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.64% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     12: predicate.addn_check_dump
 1.23% :     0.000002s :     12: predicate.addn_zero_filter
 2.39% :     0.000003s :     12: predicate.arithmetic_simplify
 1.27% :     0.000002s :     12: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     12: predicate.compare_switch_simplify
 1.51% :     0.000002s :     12: predicate.depend_value_elim
 1.15% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     12: predicate.dict_set_item_eliminator
 1.17% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.12% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.54% :     0.000002s :     12: predicate.environ_get_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.35% :     0.000000s :      3: predicate.graph_param_transform
 5.32% :     0.000007s :     26: predicate.inline
 0.75% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.95% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     17: predicate.load_eliminater
 1.49% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.34% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.90% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     12: predicate.merge_addn
 1.06% :     0.000001s :     12: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.64% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000003s :     17: predicate.partial_eliminate
 1.20% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     12: predicate.reduce_eliminate
 1.56% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.52% :     0.000002s :     20: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     12: predicate.reshape_eliminate
 1.28% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     12: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000002s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.56% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.55% :     0.000004s :     20: predicate.switch_defer_inline
 2.37% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.02% :     0.000011s :     65: predicate.switch_simplify
 1.30% :     0.000002s :     12: predicate.tile_eliminate
 1.21% :     0.000002s :     12: predicate.transpose_eliminate
 1.42% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.64% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.93% :     0.000003s :     12: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.64% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     12: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000906    15
49.35% :     0.000447s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.64% :     0.000096s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.01% :     0.000363s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018839    72
 0.29% :     0.000055s :      1: add_recomputation
 0.49% :     0.000091s :      1: auto_monad
 0.10% :     0.000019s :      1: auto_monad_reorder
 1.90% :     0.000359s :      1: bootstrap
 0.15% :     0.000029s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.13% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.14% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
31.58% :     0.005950s :      1: jit_opt_a
 0.87% :     0.000163s :      1: jit_opt_after_cconv
 0.31% :     0.000058s :      1: jit_opt_b
 2.78% :     0.000523s :      1: loop_unroll
 4.02% :     0.000758s :      1: mutable_eliminate
 5.84% :     0.001100s :     26: opt.transform.jit_opt_a
 0.28% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000027s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.95% :     0.000556s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.34% :     0.000064s :      1: py_interpret_to_execute
 0.09% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000018s :      1: remove_dup_value
 2.13% :     0.000402s :      1: renormalize.infer
 1.86% :     0.000351s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000039s :      1: rewriter_after_opt_a
 0.49% :     0.000092s :      1: rewriter_before_opt_a
 0.43% :     0.000082s :      1: symbol_engine_optimizer
41.59% :     0.007836s :      1: type_inference


TotalTime = 0.0196984, [33]
[bootstrap]: 0.00030607
[type_inference]: 0.00457441
[event_method]: 1.724e-05
[auto_monad]: 6.478e-05
[graph_reusing]: 5.71e-06
[pre_auto_parallel]: 2.32999e-06
[py_interpret_to_execute]: 3.228e-05
[rewriter_before_opt_a]: 6.959e-05
[expand_dump_flag]: 3.40998e-06
[jit_opt_a]: 0.00545722, [2]
    [Cycle 1]: 0.00161639, [27]
        [switch_simplify]: 5.462e-05
        [loop_unroll]: 2.839e-05
        [a_1]: 0.00054744
        [with_stream_mark]: 2.141e-05
        [recompute_prepare]: 8.15999e-06
        [updatestate_depend_eliminate]: 3.38999e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.58003e-06
        [parameter_eliminate]: 1.74e-06
        [specialize_transform]: 6.24001e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 1.77001e-06
        [get_grad_eliminate_]: 5.22999e-06
        [merge_forward]: 4.1e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.5e-05
        [j_node_and_user_rematch]: 9.04e-06
        [meta_fg_expand]: 2.21e-06
        [replace_old_param]: 8.67998e-06
        [inline_without_move]: 5.61e-06
        [renormalize]: 0.00063151
        [add_forward_monad_depend]: 6.07001e-06
        [auto_monad_grad]: 2.31e-06
        [auto_monad_eliminator]: 1.372e-05
        [cse]: 2.697e-05
        [replace_applicator]: 1.546e-05
    [Cycle 2]: 0.00032987, [27]
        [switch_simplify]: 5.37001e-06
        [loop_unroll]: 4.60999e-06
        [a_1]: 8.972e-05
        [with_stream_mark]: 1.122e-05
        [recompute_prepare]: 5.12e-06
        [updatestate_depend_eliminate]: 2.91999e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 5.02999e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 4.80001e-06
        [meta_shard_fg_expand]: 1.35999e-06
        [get_grad_eliminate_]: 5.07999e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 2.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.471e-05
        [j_node_and_user_rematch]: 8.57e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 7.53999e-06
        [inline_without_move]: 6.62002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.48002e-06
        [auto_monad_grad]: 6.30011e-07
        [auto_monad_eliminator]: 5.02e-06
        [cse]: 1.126e-05
        [replace_applicator]: 5.67001e-06
[py_interpret_to_execute_after_opt_a]: 1.2e-05
[rewriter_after_opt_a]: 3.87e-05
[convert_after_rewriter]: 6.88e-06
[order_py_execute_after_rewriter]: 4.85999e-06
[mutable_eliminate]: 0.00106375
[jit_opt_b]: 5.209e-05, [1]
    [Cycle 1]: 4.31e-05, [2]
        [frontend_op_eliminate]: 1.541e-05
        [inline_after_opt_a]: 1.62e-05
[cconv]: 2.679e-05
[loop_unroll]: 0.00050122
[jit_opt_after_cconv]: 0.00014588, [1]
    [Cycle 1]: 0.00013867, [11]
        [c_1]: 2.019e-05
        [parameter_eliminate]: 3.15002e-06
        [updatestate_depend_eliminate]: 6.98e-06
        [updatestate_assign_eliminate]: 2.78998e-06
        [updatestate_loads_eliminate]: 2.59001e-06
        [cse]: 2.042e-05
        [call_graph_tuple_transform]: 2.037e-05
        [tuple_list_get_item_eliminator]: 5.29e-06
        [none_parameter_eliminate]: 1.53002e-06
        [renormalize]: 5.19998e-07
        [switch_simplify]: 5.69999e-06
[remove_dup_value]: 1.54e-05
[partial_unused_args_eliminate]: 2.51e-06
[environ_conv]: 5.42001e-06
[add_recomputation]: 5.028e-05
[cse_after_recomputation]: 2.104e-05, [1]
    [Cycle 1]: 1.581e-05, [1]
        [cse]: 1.004e-05
[auto_monad_reorder]: 1.511e-05
[get_jit_bprop_graph]: 1.31002e-06
[rewriter_after_jit_bprop_graph]: 6.28e-06
[opt_after_jit_grad]: 0.00052372
[symbol_engine_optimizer]: 7.29e-05, [1]
    [Cycle 1]: 6.539e-05, [6]
        [build]: 3.73999e-06
        [elim_shapecalc]: 7.1e-06
        [elim_not_effective]: 1.263e-05
        [opt_reshape]: 5.59e-06
        [fold_const_symbol]: 7.94997e-06
        [renormalize]: 6.09987e-07
[validate]: 3.467e-05
[backend_pass]: 1.00001e-06
[task_emit]: 0.00635435
[execute]: 9.09998e-06
Sums
    bootstrap                            :     0.000306s :  1.97%
    type_inference                       :     0.004574s : 29.38%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000065s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000032s :  0.21%
    rewriter_before_opt_a                :     0.000070s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000060s :  0.39%
    jit_opt_a.loop_unroll                :     0.000033s :  0.21%
    jit_opt_a.a_1                        :     0.000637s :  4.09%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000012s :  0.08%
    jit_opt_a.renormalize                :     0.000632s :  4.06%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000038s :  0.25%
    jit_opt_a.replace_applicator         :     0.000021s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000039s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.001064s :  6.83%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000027s :  0.17%
    loop_unroll                          :     0.000501s :  3.22%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000015s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000524s :  3.36%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006354s : 40.81%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000220    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.47% :     0.000005s :      3: substitution.graph_param_transform
85.35% :     0.000188s :      4: substitution.inline
 1.68% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.95% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.86% :     0.000004s :      2: substitution.replace_old_param
 5.32% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004508     2
82.37% :     0.003714s :      1: type_inference.infer
17.63% :     0.000795s :      1: type_inference.specialize
------[replace.]   0.000067     6
69.61% :     0.000047s :      4: replace.inline
30.39% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000195     6
94.55% :     0.000185s :      4: match.inline
 5.45% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.89% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.65% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.57% :     0.000002s :     10: predicate.addn_zero_filter
 2.05% :     0.000002s :     10: predicate.arithmetic_simplify
 1.11% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.83% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.10% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.45% :     0.000002s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.19% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.44% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.68% :     0.000003s :     15: predicate.load_eliminater
 1.22% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.79% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.78% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 2.18% :     0.000003s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.04% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.42% :     0.000002s :     10: predicate.reduce_eliminate
 1.57% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.26% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.12% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.85% :     0.000009s :     53: predicate.switch_simplify
 1.09% :     0.000001s :     10: predicate.tile_eliminate
 1.36% :     0.000002s :     10: predicate.transpose_eliminate
 1.26% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.55% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.07% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000553    12
50.02% :     0.000277s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.69% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.29% :     0.000240s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021291    76
 0.25% :     0.000053s :      1: add_recomputation
 0.32% :     0.000068s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.52% :     0.000323s :      1: bootstrap
 0.14% :     0.000030s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000003s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.65% :     0.005460s :      1: jit_opt_a
 0.70% :     0.000149s :      1: jit_opt_after_cconv
 0.26% :     0.000055s :      1: jit_opt_b
 2.40% :     0.000510s :      1: loop_unroll
 5.05% :     0.001076s :      1: mutable_eliminate
 3.99% :     0.000851s :     26: opt.transform.jit_opt_a
 0.23% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.50% :     0.000532s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000035s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 1.56% :     0.000333s :      1: renormalize.infer
 1.37% :     0.000291s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000042s :      1: rewriter_after_opt_a
 0.34% :     0.000072s :      1: rewriter_before_opt_a
 0.35% :     0.000076s :      1: symbol_engine_optimizer
29.91% :     0.006369s :      1: task_emit
21.54% :     0.004587s :      1: type_inference
 0.26% :     0.000055s :      1: validate


TotalTime = 0.0266376, [33]
[bootstrap]: 0.00031402
[type_inference]: 0.00492397
[event_method]: 1.907e-05
[auto_monad]: 6.756e-05
[graph_reusing]: 6.25002e-06
[pre_auto_parallel]: 2.64001e-06
[py_interpret_to_execute]: 3.654e-05
[rewriter_before_opt_a]: 7.315e-05
[expand_dump_flag]: 3.13e-06
[jit_opt_a]: 0.00567197, [2]
    [Cycle 1]: 0.00159252, [27]
        [switch_simplify]: 4.934e-05
        [loop_unroll]: 2.981e-05
        [a_1]: 0.00053496
        [with_stream_mark]: 1.711e-05
        [recompute_prepare]: 6.73e-06
        [updatestate_depend_eliminate]: 3.30998e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 5.44e-06
        [updatestate_useless_node_eliminater]: 4.71002e-06
        [accelerated_algorithm]: 5.91e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 5.05001e-06
        [merge_forward]: 2.79001e-06
        [cell_reuse_recompute_pass]: 8.50006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.09e-05
        [j_node_and_user_rematch]: 9.34998e-06
        [meta_fg_expand]: 1.84e-06
        [replace_old_param]: 8.22e-06
        [inline_without_move]: 5.25001e-06
        [renormalize]: 0.00066819
        [add_forward_monad_depend]: 4.75001e-06
        [auto_monad_grad]: 2.48e-06
        [auto_monad_eliminator]: 1.262e-05
        [cse]: 1.818e-05
        [replace_applicator]: 1.198e-05
    [Cycle 2]: 0.00031502, [27]
        [switch_simplify]: 5.55001e-06
        [loop_unroll]: 4.99003e-06
        [a_1]: 9.041e-05
        [with_stream_mark]: 9.56003e-06
        [recompute_prepare]: 4.72998e-06
        [updatestate_depend_eliminate]: 2.31e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 4.89003e-06
        [updatestate_useless_node_eliminater]: 4.72998e-06
        [accelerated_algorithm]: 4.50001e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 4.53001e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.52001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.18e-05
        [j_node_and_user_rematch]: 8.35999e-06
        [meta_fg_expand]: 1.60001e-06
        [replace_old_param]: 8.55001e-06
        [inline_without_move]: 4.65001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.33002e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 5.16002e-06
        [cse]: 1.029e-05
        [replace_applicator]: 4.82e-06
[py_interpret_to_execute_after_opt_a]: 1.043e-05
[rewriter_after_opt_a]: 2.937e-05
[convert_after_rewriter]: 5.86e-06
[order_py_execute_after_rewriter]: 1.058e-05
[mutable_eliminate]: 0.0005611
[jit_opt_b]: 4.888e-05, [1]
    [Cycle 1]: 4.186e-05, [2]
        [frontend_op_eliminate]: 1.605e-05
        [inline_after_opt_a]: 1.467e-05
[cconv]: 2.234e-05
[loop_unroll]: 0.00042312
[jit_opt_after_cconv]: 0.00014004, [1]
    [Cycle 1]: 0.00013361, [11]
        [c_1]: 2.084e-05
        [parameter_eliminate]: 2.98e-06
        [updatestate_depend_eliminate]: 6.11998e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.32001e-06
        [cse]: 2.026e-05
        [call_graph_tuple_transform]: 1.623e-05
        [tuple_list_get_item_eliminator]: 5.31998e-06
        [none_parameter_eliminate]: 1.19e-06
        [renormalize]: 4.60015e-07
        [switch_simplify]: 5.09998e-06
[remove_dup_value]: 1.017e-05
[partial_unused_args_eliminate]: 1.82999e-06
[environ_conv]: 4.33001e-06
[add_recomputation]: 3.774e-05
[cse_after_recomputation]: 2.056e-05, [1]
    [Cycle 1]: 1.516e-05, [1]
        [cse]: 9.61e-06
[auto_monad_reorder]: 1.222e-05
[get_jit_bprop_graph]: 1.69e-06
[rewriter_after_jit_bprop_graph]: 5.49e-06
[opt_after_jit_grad]: 0.00614103
[symbol_engine_optimizer]: 0.00010337, [1]
    [Cycle 1]: 9.233e-05, [6]
        [build]: 7.48e-06
        [elim_shapecalc]: 9.87999e-06
        [elim_not_effective]: 2.498e-05
        [opt_reshape]: 6.98998e-06
        [fold_const_symbol]: 1.031e-05
        [renormalize]: 1.15001e-06
[validate]: 5.105e-05
[backend_pass]: 1.34e-06
[task_emit]: 0.00758627
[execute]: 8.33001e-06
Sums
    bootstrap                            :     0.000314s :  1.42%
    type_inference                       :     0.004924s : 22.19%
    event_method                         :     0.000019s :  0.09%
    auto_monad                           :     0.000068s :  0.30%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000037s :  0.16%
    rewriter_before_opt_a                :     0.000073s :  0.33%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000055s :  0.25%
    jit_opt_a.loop_unroll                :     0.000035s :  0.16%
    jit_opt_a.a_1                        :     0.000625s :  2.82%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000010s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.04%
    jit_opt_a.renormalize                :     0.000668s :  3.01%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.08%
    jit_opt_a.cse                        :     0.000028s :  0.13%
    jit_opt_a.replace_applicator         :     0.000017s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.05%
    rewriter_after_opt_a                 :     0.000029s :  0.13%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000011s :  0.05%
    mutable_eliminate                    :     0.000561s :  2.53%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.07%
    cconv                                :     0.000022s :  0.10%
    loop_unroll                          :     0.000423s :  1.91%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000010s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000038s :  0.17%
    cse_after_recomputation.cse          :     0.000010s :  0.04%
    auto_monad_reorder                   :     0.000012s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.006141s : 27.68%
    symbol_engine_optimizer.build        :     0.000007s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000025s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000051s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007586s : 34.19%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000205    23
 1.22% :     0.000003s :      2: substitution.elim_not_effective
 0.75% :     0.000002s :      2: substitution.fold_const_symbol
 2.01% :     0.000004s :      3: substitution.graph_param_transform
84.77% :     0.000174s :      4: substitution.inline
 1.71% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.03% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.13% :     0.000004s :      2: substitution.replace_old_param
 5.39% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004852     2
80.64% :     0.003913s :      1: type_inference.infer
19.36% :     0.000939s :      1: type_inference.specialize
------[replace.]   0.000072     6
67.34% :     0.000049s :      4: replace.inline
32.66% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000181     6
94.49% :     0.000171s :      4: match.inline
 5.51% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.85% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.92% :     0.000004s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.42% :     0.000003s :     10: predicate.arithmetic_simplify
 1.15% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.47% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.17% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.76% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.72% :     0.000001s :      3: predicate.elim_not_effective
 0.89% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.33% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 4.77% :     0.000006s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000001s :      6: predicate.less_batch_normalization
 1.46% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.33% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.76% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.49% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.62% :     0.000002s :      3: predicate.mutable_eliminate
 0.68% :     0.000001s :      3: predicate.opt_reshape
 2.03% :     0.000002s :     15: predicate.partial_eliminate
 1.39% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.84% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.32% :     0.000002s :     10: predicate.reshape_eliminate
 1.35% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000001s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.66% :     0.000008s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.46% :     0.000002s :     10: predicate.transpose_eliminate
 1.23% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.38% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.79% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000663    12
44.53% :     0.000295s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.80% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.67% :     0.000329s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028274    76
 0.14% :     0.000041s :      1: add_recomputation
 0.25% :     0.000071s :      1: auto_monad
 0.05% :     0.000015s :      1: auto_monad_reorder
 0.02% :     0.000006s :      1: backend_pass
 1.16% :     0.000329s :      1: bootstrap
 0.09% :     0.000025s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.08% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.08% :     0.000023s :      1: event_method
 0.04% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
20.07% :     0.005676s :      1: jit_opt_a
 0.50% :     0.000143s :      1: jit_opt_after_cconv
 0.18% :     0.000051s :      1: jit_opt_b
 1.53% :     0.000431s :      1: loop_unroll
 2.02% :     0.000571s :      1: mutable_eliminate
 2.92% :     0.000825s :     26: opt.transform.jit_opt_a
 0.16% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000025s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000049s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000048s :      4: opt.transform.symbol_engine_opt
21.85% :     0.006179s :      1: opt_after_jit_grad
 0.05% :     0.000013s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000039s :      1: py_interpret_to_execute
 0.05% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000013s :      1: remove_dup_value
 1.28% :     0.000363s :      1: renormalize.infer
 1.06% :     0.000299s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000032s :      1: rewriter_after_opt_a
 0.27% :     0.000076s :      1: rewriter_before_opt_a
 0.38% :     0.000107s :      1: symbol_engine_optimizer
26.88% :     0.007601s :      1: task_emit
17.47% :     0.004939s :      1: type_inference
 0.42% :     0.000119s :      1: validate


TotalTime = 0.0146453, [30]
[bootstrap]: 0.00032114
[type_inference]: 0.00688756
[event_method]: 2.13e-05
[auto_monad]: 8.203e-05
[graph_reusing]: 6.53e-06
[pre_auto_parallel]: 2.74001e-06
[py_interpret_to_execute]: 4.98e-05
[rewriter_before_opt_a]: 8.48e-05
[expand_dump_flag]: 3.28e-06
[jit_opt_a]: 0.00515726, [2]
    [Cycle 1]: 0.00177238, [27]
        [switch_simplify]: 0.00010372
        [loop_unroll]: 3.407e-05
        [a_1]: 0.00071857
        [with_stream_mark]: 1.637e-05
        [recompute_prepare]: 7.45e-06
        [updatestate_depend_eliminate]: 3.48e-06
        [updatestate_assign_eliminate]: 2.97002e-06
        [updatestate_loads_eliminate]: 2.99999e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 6.31e-06
        [updatestate_useless_node_eliminater]: 5.36998e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 2.31e-06
        [get_grad_eliminate_]: 5.24003e-06
        [merge_forward]: 3.48e-06
        [cell_reuse_recompute_pass]: 1.38002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.388e-05
        [j_node_and_user_rematch]: 9.24e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 8.47e-06
        [inline_without_move]: 5.79999e-06
        [renormalize]: 0.00058034
        [add_forward_monad_depend]: 5.24e-06
        [auto_monad_grad]: 1.60999e-06
        [auto_monad_eliminator]: 1.326e-05
        [cse]: 2.805e-05
        [replace_applicator]: 1.12e-05
    [Cycle 2]: 0.0003258, [27]
        [switch_simplify]: 5.67999e-06
        [loop_unroll]: 4.96002e-06
        [a_1]: 9.016e-05
        [with_stream_mark]: 8.98002e-06
        [recompute_prepare]: 5.27999e-06
        [updatestate_depend_eliminate]: 2.69999e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 5.10999e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 5.52999e-06
        [meta_shard_fg_expand]: 1.15999e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.238e-05
        [j_node_and_user_rematch]: 8.36002e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 8.00999e-06
        [inline_without_move]: 5.05001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 4.87e-06
        [cse]: 1.022e-05
        [replace_applicator]: 5.62001e-06
[py_interpret_to_execute_after_opt_a]: 1.02e-05
[rewriter_after_opt_a]: 3.142e-05
[convert_after_rewriter]: 6.71999e-06
[order_py_execute_after_rewriter]: 4.84e-06
[mutable_eliminate]: 0.00051031
[jit_opt_b]: 5.189e-05, [1]
    [Cycle 1]: 4.492e-05, [2]
        [frontend_op_eliminate]: 1.681e-05
        [inline_after_opt_a]: 1.608e-05
[cconv]: 2.586e-05
[loop_unroll]: 0.00041757
[jit_opt_after_cconv]: 0.0001375, [1]
    [Cycle 1]: 0.00013129, [11]
        [c_1]: 2.085e-05
        [parameter_eliminate]: 2.32001e-06
        [updatestate_depend_eliminate]: 5.81e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [cse]: 1.657e-05
        [call_graph_tuple_transform]: 1.886e-05
        [tuple_list_get_item_eliminator]: 5.42999e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 8.10018e-07
        [switch_simplify]: 5.57999e-06
[remove_dup_value]: 1.345e-05
[partial_unused_args_eliminate]: 2.25002e-06
[environ_conv]: 4.85999e-06
[add_recomputation]: 4.723e-05
[cse_after_recomputation]: 2.018e-05, [1]
    [Cycle 1]: 1.472e-05, [1]
        [cse]: 8.60999e-06
[auto_monad_reorder]: 1.503e-05
[get_jit_bprop_graph]: 1.72999e-06
[rewriter_after_jit_bprop_graph]: 5.07999e-06
[opt_after_jit_grad]: 0.00045633
[symbol_engine_optimizer]: 7.388e-05, [1]
    [Cycle 1]: 6.745e-05, [6]
        [build]: 3.63e-06
        [elim_shapecalc]: 7.68001e-06
        [elim_not_effective]: 1.3e-05
        [opt_reshape]: 6.08998e-06
        [fold_const_symbol]: 9.15001e-06
        [renormalize]: 3.9002e-07
[validate]: 3.28e-05
Sums
    bootstrap                            :     0.000321s :  2.92%
    type_inference                       :     0.006888s : 62.52%
    event_method                         :     0.000021s :  0.19%
    auto_monad                           :     0.000082s :  0.74%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000050s :  0.45%
    rewriter_before_opt_a                :     0.000085s :  0.77%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000109s :  0.99%
    jit_opt_a.loop_unroll                :     0.000039s :  0.35%
    jit_opt_a.a_1                        :     0.000809s :  7.34%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000580s :  5.27%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.16%
    jit_opt_a.cse                        :     0.000038s :  0.35%
    jit_opt_a.replace_applicator         :     0.000017s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.09%
    rewriter_after_opt_a                 :     0.000031s :  0.29%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000510s :  4.63%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.15%
    cconv                                :     0.000026s :  0.23%
    loop_unroll                          :     0.000418s :  3.79%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000013s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.43%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000456s :  4.14%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.30%


Time group info:
------[substitution.]   0.000252    26
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.15% :     0.000005s :      3: substitution.graph_param_transform
84.18% :     0.000212s :      6: substitution.inline
 1.07% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.77% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.24% :     0.000003s :      2: substitution.replace_old_param
 3.19% :     0.000008s :      1: substitution.switch_simplify
 4.93% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006823     2
84.67% :     0.005777s :      1: type_inference.infer
15.33% :     0.001046s :      1: type_inference.specialize
------[replace.]   0.000124     9
52.96% :     0.000065s :      6: replace.inline
22.75% :     0.000028s :      1: replace.switch_simplify
24.29% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000228     9
91.78% :     0.000209s :      6: match.inline
 3.23% :     0.000007s :      1: match.switch_simplify
 4.99% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   801
 1.39% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.02% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     12: predicate.addn_check_dump
 1.44% :     0.000002s :     12: predicate.addn_zero_filter
 1.90% :     0.000003s :     12: predicate.arithmetic_simplify
 1.26% :     0.000002s :     12: predicate.cast_eliminate
 0.37% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     12: predicate.compare_switch_simplify
 1.48% :     0.000002s :     12: predicate.depend_value_elim
 1.16% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.74% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.24% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.27% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     12: predicate.environ_get_eliminate
 1.26% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.47% :     0.000007s :     26: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.07% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     17: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.33% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.01% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.45% :     0.000002s :     12: predicate.merge_addn
 1.20% :     0.000002s :     12: predicate.minmaximum_grad
 1.35% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     17: predicate.partial_eliminate
 1.50% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     12: predicate.reduce_eliminate
 1.48% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     20: predicate.replace_applicator
 0.83% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.63% :     0.000002s :     12: predicate.reshape_eliminate
 1.44% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.79% :     0.000004s :     20: predicate.switch_defer_inline
 2.71% :     0.000004s :     20: predicate.switch_layer_defer_inline
 8.19% :     0.000011s :     65: predicate.switch_simplify
 1.24% :     0.000002s :     12: predicate.tile_eliminate
 1.16% :     0.000002s :     12: predicate.transpose_eliminate
 1.52% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.92% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.61% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     12: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000791    15
52.77% :     0.000417s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 9.80% :     0.000078s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.43% :     0.000296s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016360    72
 0.31% :     0.000050s :      1: add_recomputation
 0.52% :     0.000086s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 2.05% :     0.000336s :      1: bootstrap
 0.18% :     0.000029s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
31.54% :     0.005161s :      1: jit_opt_a
 0.86% :     0.000140s :      1: jit_opt_after_cconv
 0.34% :     0.000055s :      1: jit_opt_b
 2.60% :     0.000425s :      1: loop_unroll
 3.17% :     0.000518s :      1: mutable_eliminate
 6.56% :     0.001074s :     26: opt.transform.jit_opt_a
 0.29% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.83% :     0.000464s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.32% :     0.000053s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 1.82% :     0.000298s :      1: renormalize.infer
 1.68% :     0.000275s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000034s :      1: rewriter_after_opt_a
 0.54% :     0.000088s :      1: rewriter_before_opt_a
 0.47% :     0.000077s :      1: symbol_engine_optimizer
42.17% :     0.006900s :      1: type_inference


TotalTime = 0.0165515, [33]
[bootstrap]: 0.000281
[type_inference]: 0.0039425
[event_method]: 1.54e-05
[auto_monad]: 3.887e-05
[graph_reusing]: 3.43e-06
[pre_auto_parallel]: 1.66e-06
[py_interpret_to_execute]: 2.766e-05
[rewriter_before_opt_a]: 7.234e-05
[expand_dump_flag]: 1.99e-06
[jit_opt_a]: 0.0045942, [2]
    [Cycle 1]: 0.00127313, [27]
        [switch_simplify]: 4.351e-05
        [loop_unroll]: 2.829e-05
        [a_1]: 0.00046391
        [with_stream_mark]: 1.125e-05
        [recompute_prepare]: 6.89001e-06
        [updatestate_depend_eliminate]: 2.78998e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 1.95001e-06
        [parameter_eliminate]: 1.14998e-06
        [specialize_transform]: 6.11998e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 5.07999e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.77e-06
        [merge_forward]: 2.70002e-06
        [cell_reuse_recompute_pass]: 9.10019e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.09e-05
        [j_node_and_user_rematch]: 7.95e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 7.35e-06
        [inline_without_move]: 5.07e-06
        [renormalize]: 0.00045311
        [add_forward_monad_depend]: 4.33001e-06
        [auto_monad_grad]: 1.52999e-06
        [auto_monad_eliminator]: 9.40001e-06
        [cse]: 1.656e-05
        [replace_applicator]: 1.1e-05
    [Cycle 2]: 0.00030766, [27]
        [switch_simplify]: 5.38002e-06
        [loop_unroll]: 4.88001e-06
        [a_1]: 8.682e-05
        [with_stream_mark]: 7.88001e-06
        [recompute_prepare]: 5.07999e-06
        [updatestate_depend_eliminate]: 2.43e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 4.95001e-06
        [updatestate_useless_node_eliminater]: 4.70999e-06
        [accelerated_algorithm]: 4.77e-06
        [meta_shard_fg_expand]: 9.5999e-07
        [get_grad_eliminate_]: 4.28999e-06
        [merge_forward]: 2.24001e-06
        [cell_reuse_recompute_pass]: 1.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.18e-05
        [j_node_and_user_rematch]: 7.48e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 6.78e-06
        [inline_without_move]: 4.60999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.05001e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 4.89e-06
        [cse]: 9.27001e-06
        [replace_applicator]: 5.14003e-06
[py_interpret_to_execute_after_opt_a]: 8.23999e-06
[rewriter_after_opt_a]: 2.29e-05
[convert_after_rewriter]: 5.86e-06
[order_py_execute_after_rewriter]: 3.71001e-06
[mutable_eliminate]: 0.00049032
[jit_opt_b]: 4.55e-05, [1]
    [Cycle 1]: 3.959e-05, [2]
        [frontend_op_eliminate]: 1.524e-05
        [inline_after_opt_a]: 1.342e-05
[cconv]: 1.683e-05
[loop_unroll]: 0.00041479
[jit_opt_after_cconv]: 0.0001311, [1]
    [Cycle 1]: 0.00012508, [11]
        [c_1]: 1.959e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 5.17999e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.767e-05
        [call_graph_tuple_transform]: 1.687e-05
        [tuple_list_get_item_eliminator]: 5.09998e-06
        [none_parameter_eliminate]: 8.59989e-07
        [renormalize]: 4.09986e-07
        [switch_simplify]: 5.25001e-06
[remove_dup_value]: 7.60998e-06
[partial_unused_args_eliminate]: 1.59998e-06
[environ_conv]: 3.26999e-06
[add_recomputation]: 3.011e-05
[cse_after_recomputation]: 1.968e-05, [1]
    [Cycle 1]: 1.456e-05, [1]
        [cse]: 9.02999e-06
[auto_monad_reorder]: 9.28002e-06
[get_jit_bprop_graph]: 1.35001e-06
[rewriter_after_jit_bprop_graph]: 4.60001e-06
[opt_after_jit_grad]: 0.00050385
[symbol_engine_optimizer]: 6.885e-05, [1]
    [Cycle 1]: 6.222e-05, [6]
        [build]: 2.76e-06
        [elim_shapecalc]: 6.87002e-06
        [elim_not_effective]: 1.191e-05
        [opt_reshape]: 5.97001e-06
        [fold_const_symbol]: 7.83999e-06
        [renormalize]: 3.69997e-07
[validate]: 2.65e-05
[backend_pass]: 6.69999e-07
[task_emit]: 0.00557652
[execute]: 7.55e-06
Sums
    bootstrap                            :     0.000281s :  2.16%
    type_inference                       :     0.003943s : 30.37%
    event_method                         :     0.000015s :  0.12%
    auto_monad                           :     0.000039s :  0.30%
    graph_reusing                        :     0.000003s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.21%
    rewriter_before_opt_a                :     0.000072s :  0.56%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000049s :  0.38%
    jit_opt_a.loop_unroll                :     0.000033s :  0.26%
    jit_opt_a.a_1                        :     0.000551s :  4.24%
    jit_opt_a.with_stream_mark           :     0.000019s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000015s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000014s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000453s :  3.49%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.11%
    jit_opt_a.cse                        :     0.000026s :  0.20%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000023s :  0.18%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000490s :  3.78%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000013s :  0.10%
    cconv                                :     0.000017s :  0.13%
    loop_unroll                          :     0.000415s :  3.20%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000008s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000003s :  0.03%
    add_recomputation                    :     0.000030s :  0.23%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000009s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000504s :  3.88%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000026s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005577s : 42.96%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000153    23
 0.82% :     0.000001s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.66% :     0.000004s :      3: substitution.graph_param_transform
85.09% :     0.000130s :      4: substitution.inline
 1.62% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.50% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.87% :     0.000003s :      2: substitution.replace_old_param
 4.74% :     0.000007s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003898     2
82.27% :     0.003207s :      1: type_inference.infer
17.73% :     0.000691s :      1: type_inference.specialize
------[replace.]   0.000059     6
65.81% :     0.000039s :      4: replace.inline
34.19% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000134     6
95.40% :     0.000127s :      4: match.inline
 4.60% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000108   691
 1.52% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.11% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000001s :     10: predicate.addn_zero_filter
 1.87% :     0.000002s :     10: predicate.arithmetic_simplify
 1.73% :     0.000002s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.61% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.23% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 4.99% :     0.000005s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.44% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.62% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.55% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.29% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.61% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000002s :     15: predicate.partial_eliminate
 1.23% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     10: predicate.reduce_eliminate
 1.59% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.40% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.14% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.75% :     0.000008s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.27% :     0.000001s :     10: predicate.transpose_eliminate
 1.53% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.46% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000476    12
47.51% :     0.000226s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.49% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.00% :     0.000224s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017848    76
 0.18% :     0.000033s :      1: add_recomputation
 0.23% :     0.000042s :      1: auto_monad
 0.07% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.63% :     0.000291s :      1: bootstrap
 0.11% :     0.000020s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.07% :     0.000013s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000003s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
25.76% :     0.004597s :      1: jit_opt_a
 0.75% :     0.000134s :      1: jit_opt_after_cconv
 0.27% :     0.000048s :      1: jit_opt_b
 2.37% :     0.000422s :      1: loop_unroll
 2.79% :     0.000498s :      1: mutable_eliminate
 4.14% :     0.000739s :     26: opt.transform.jit_opt_a
 0.25% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.87% :     0.000512s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000003s :      1: pre_auto_parallel
 0.17% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000010s :      1: remove_dup_value
 1.26% :     0.000225s :      1: renormalize.infer
 1.25% :     0.000223s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000027s :      1: rewriter_after_opt_a
 0.42% :     0.000075s :      1: rewriter_before_opt_a
 0.40% :     0.000071s :      1: symbol_engine_optimizer
31.32% :     0.005589s :      1: task_emit
22.14% :     0.003952s :      1: type_inference
 0.23% :     0.000041s :      1: validate


TotalTime = 0.817497, [33]
[bootstrap]: 0.00043083
[type_inference]: 0.152499
[event_method]: 2.548e-05
[auto_monad]: 0.00025449
[graph_reusing]: 6.54001e-06
[pre_auto_parallel]: 1.153e-05
[py_interpret_to_execute]: 0.0004164
[rewriter_before_opt_a]: 0.00014455
[expand_dump_flag]: 4.15e-06
[jit_opt_a]: 0.0172613, [2]
    [Cycle 1]: 0.00439268, [27]
        [switch_simplify]: 0.0001059
        [loop_unroll]: 3.765e-05
        [a_1]: 0.00076391
        [with_stream_mark]: 2.882e-05
        [recompute_prepare]: 1.413e-05
        [updatestate_depend_eliminate]: 5.29e-06
        [updatestate_assign_eliminate]: 3.76999e-06
        [updatestate_loads_eliminate]: 2.86e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 7.87998e-06
        [updatestate_useless_node_eliminater]: 6.60002e-06
        [accelerated_algorithm]: 1.005e-05
        [meta_shard_fg_expand]: 1.863e-05
        [get_grad_eliminate_]: 8.63001e-06
        [merge_forward]: 4.48999e-06
        [cell_reuse_recompute_pass]: 1.89999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.575e-05
        [j_node_and_user_rematch]: 1.266e-05
        [meta_fg_expand]: 3.21001e-06
        [replace_old_param]: 1.271e-05
        [inline_without_move]: 8.12e-06
        [renormalize]: 0.00287607
        [add_forward_monad_depend]: 3.296e-05
        [auto_monad_grad]: 2.81e-06
        [auto_monad_eliminator]: 4.16e-05
        [cse]: 3.607e-05
        [replace_applicator]: 2.838e-05
    [Cycle 2]: 0.00047086, [27]
        [switch_simplify]: 8.21002e-06
        [loop_unroll]: 7.29001e-06
        [a_1]: 0.0001524
        [with_stream_mark]: 1.993e-05
        [recompute_prepare]: 8.98002e-06
        [updatestate_depend_eliminate]: 3.93001e-06
        [updatestate_assign_eliminate]: 3.53e-06
        [updatestate_loads_eliminate]: 3.33998e-06
        [parameter_eliminate]: 2.54999e-06
        [specialize_transform]: 6.49999e-06
        [updatestate_useless_node_eliminater]: 6.65002e-06
        [accelerated_algorithm]: 8.2e-06
        [meta_shard_fg_expand]: 2.88e-06
        [get_grad_eliminate_]: 5.92001e-06
        [merge_forward]: 5.10001e-06
        [cell_reuse_recompute_pass]: 5.56e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.727e-05
        [j_node_and_user_rematch]: 1.075e-05
        [meta_fg_expand]: 3.11999e-06
        [replace_old_param]: 1.132e-05
        [inline_without_move]: 6.84001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 2.37001e-06
        [auto_monad_grad]: 1.46998e-06
        [auto_monad_eliminator]: 9.52999e-06
        [cse]: 1.728e-05
        [replace_applicator]: 9.66998e-06
[py_interpret_to_execute_after_opt_a]: 2.358e-05
[rewriter_after_opt_a]: 6.562e-05
[convert_after_rewriter]: 9.10999e-06
[order_py_execute_after_rewriter]: 6.44999e-06
[mutable_eliminate]: 0.00098787
[jit_opt_b]: 7.213e-05, [1]
    [Cycle 1]: 6.226e-05, [2]
        [frontend_op_eliminate]: 2.484e-05
        [inline_after_opt_a]: 2.339e-05
[cconv]: 3.869e-05
[loop_unroll]: 0.0006732
[jit_opt_after_cconv]: 0.0001993, [1]
    [Cycle 1]: 0.00019189, [11]
        [c_1]: 2.814e-05
        [parameter_eliminate]: 5.09998e-06
        [updatestate_depend_eliminate]: 1.057e-05
        [updatestate_assign_eliminate]: 3.30003e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [cse]: 4.087e-05
        [call_graph_tuple_transform]: 2.822e-05
        [tuple_list_get_item_eliminator]: 6.56999e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 9.30013e-07
        [switch_simplify]: 7.36001e-06
[remove_dup_value]: 1.722e-05
[partial_unused_args_eliminate]: 2.68e-06
[environ_conv]: 3.732e-05
[add_recomputation]: 0.00012729
[cse_after_recomputation]: 3.038e-05, [1]
    [Cycle 1]: 2.338e-05, [1]
        [cse]: 1.402e-05
[auto_monad_reorder]: 2.764e-05
[get_jit_bprop_graph]: 2.39999e-06
[rewriter_after_jit_bprop_graph]: 0.00020804
[opt_after_jit_grad]: 0.00063973
[symbol_engine_optimizer]: 9.088e-05, [1]
    [Cycle 1]: 8.318e-05, [6]
        [build]: 6.78e-06
        [elim_shapecalc]: 1.002e-05
        [elim_not_effective]: 1.662e-05
        [opt_reshape]: 7.80998e-06
        [fold_const_symbol]: 1.029e-05
        [renormalize]: 8.89995e-07
[validate]: 0.00013623
[backend_pass]: 1.59998e-06
[task_emit]: 0.642696
[execute]: 7.03998e-06
Sums
    bootstrap                            :     0.000431s :  0.05%
    type_inference                       :     0.152499s : 18.96%
    event_method                         :     0.000025s :  0.00%
    auto_monad                           :     0.000254s :  0.03%
    graph_reusing                        :     0.000007s :  0.00%
    pre_auto_parallel                    :     0.000012s :  0.00%
    py_interpret_to_execute              :     0.000416s :  0.05%
    rewriter_before_opt_a                :     0.000145s :  0.02%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000114s :  0.01%
    jit_opt_a.loop_unroll                :     0.000045s :  0.01%
    jit_opt_a.a_1                        :     0.000916s :  0.11%
    jit_opt_a.with_stream_mark           :     0.000049s :  0.01%
    jit_opt_a.recompute_prepare          :     0.000023s :  0.00%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.00%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.00%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.00%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.00%
    jit_opt_a.specialize_transform       :     0.000014s :  0.00%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000013s :  0.00%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.00%
    jit_opt_a.meta_shard_fg_expand       :     0.000022s :  0.00%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.00%
    jit_opt_a.merge_forward              :     0.000010s :  0.00%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.00%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000043s :  0.01%
    jit_opt_a.j_node_and_user_rematch    :     0.000023s :  0.00%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.00%
    jit_opt_a.replace_old_param          :     0.000024s :  0.00%
    jit_opt_a.inline_without_move        :     0.000015s :  0.00%
    jit_opt_a.renormalize                :     0.002876s :  0.36%
    jit_opt_a.add_forward_monad_depend   :     0.000035s :  0.00%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000051s :  0.01%
    jit_opt_a.cse                        :     0.000053s :  0.01%
    jit_opt_a.replace_applicator         :     0.000038s :  0.00%
    py_interpret_to_execute_after_opt_a  :     0.000024s :  0.00%
    rewriter_after_opt_a                 :     0.000066s :  0.01%
    convert_after_rewriter               :     0.000009s :  0.00%
    order_py_execute_after_rewriter      :     0.000006s :  0.00%
    mutable_eliminate                    :     0.000988s :  0.12%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.00%
    jit_opt_b.inline_after_opt_a         :     0.000023s :  0.00%
    cconv                                :     0.000039s :  0.00%
    loop_unroll                          :     0.000673s :  0.08%
    jit_opt_after_cconv.c_1              :     0.000028s :  0.00%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.00%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000041s :  0.01%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.00%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.00%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.00%
    remove_dup_value                     :     0.000017s :  0.00%
    partial_unused_args_eliminate        :     0.000003s :  0.00%
    environ_conv                         :     0.000037s :  0.00%
    add_recomputation                    :     0.000127s :  0.02%
    cse_after_recomputation.cse          :     0.000014s :  0.00%
    auto_monad_reorder                   :     0.000028s :  0.00%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000208s :  0.03%
    opt_after_jit_grad                   :     0.000640s :  0.08%
    symbol_engine_optimizer.build        :     0.000007s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.00%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.00%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.00%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.00%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000136s :  0.02%
    backend_pass                         :     0.000002s :  0.00%
    task_emit                            :     0.642696s : 79.92%
    execute                              :     0.000007s :  0.00%


Time group info:
------[substitution.]   0.000348    27
 0.68% :     0.000002s :      2: substitution.elim_not_effective
 0.41% :     0.000001s :      2: substitution.fold_const_symbol
 2.02% :     0.000007s :      4: substitution.graph_param_transform
80.92% :     0.000282s :      4: substitution.inline
 1.50% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.90% :     0.000007s :      4: substitution.remove_not_recompute_node
 2.07% :     0.000007s :      2: substitution.replace_old_param
 6.85% :     0.000024s :      3: substitution.tile_eliminate
 3.66% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.152342     2
98.12% :     0.149480s :      1: type_inference.infer
 1.88% :     0.002863s :      1: type_inference.specialize
------[replace.]   0.000099     6
63.17% :     0.000062s :      4: replace.inline
36.83% :     0.000036s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000289     6
96.09% :     0.000278s :      4: match.inline
 3.91% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000162   834
 1.63% :     0.000003s :     12: predicate.accumulaten_eliminater
 1.22% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 0.98% :     0.000002s :     12: predicate.addn_check_dump
 1.24% :     0.000002s :     12: predicate.addn_zero_filter
 2.52% :     0.000004s :     12: predicate.arithmetic_simplify
 1.43% :     0.000002s :     12: predicate.cast_eliminate
 0.41% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.11% :     0.000002s :     12: predicate.compare_switch_simplify
 1.47% :     0.000002s :     12: predicate.depend_value_elim
 0.90% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.13% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     12: predicate.dict_set_item_eliminator
 1.55% :     0.000003s :      4: predicate.dumpgradient_eliminate
 0.32% :     0.000001s :      4: predicate.elim_not_effective
 0.66% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.02% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.01% :     0.000002s :     12: predicate.environ_get_add_eliminate
 0.90% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.05% :     0.000002s :     12: predicate.environ_get_eliminate
 0.87% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      4: predicate.fold_const_symbol
 1.02% :     0.000002s :      8: predicate.get_grad_eliminate
 0.22% :     0.000000s :      4: predicate.graph_param_transform
 5.48% :     0.000009s :     26: predicate.inline
 1.00% :     0.000002s :      8: predicate.inline_without_move
 0.39% :     0.000001s :      8: predicate.j_node_and_user_rematch
 2.14% :     0.000003s :      8: predicate.less_batch_normalization
 1.62% :     0.000003s :     14: predicate.list_to_tuple_eliminator_
 1.52% :     0.000002s :     18: predicate.load_eliminater
 2.35% :     0.000004s :      4: predicate.loop_unroll_after_grad
 3.45% :     0.000006s :     38: predicate.loop_unroll_before_grad
 2.10% :     0.000003s :     16: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000002s :     12: predicate.merge_addn
 1.00% :     0.000002s :     12: predicate.minmaximum_grad
 3.29% :     0.000005s :      4: predicate.mutable_eliminate
 0.62% :     0.000001s :      4: predicate.opt_reshape
 1.82% :     0.000003s :     18: predicate.partial_eliminate
 1.13% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.63% :     0.000003s :     12: predicate.reduce_eliminate
 1.29% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.91% :     0.000003s :     22: predicate.replace_applicator
 0.62% :     0.000001s :      8: predicate.replace_old_param
 0.46% :     0.000001s :      4: predicate.reset_defer_inline
 1.64% :     0.000003s :     12: predicate.reshape_eliminate
 1.12% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 1.06% :     0.000002s :      4: predicate.row_tensor_eliminate
 1.06% :     0.000002s :     12: predicate.same_eliminate
 0.65% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.98% :     0.000002s :      8: predicate.special_op_eliminate
 0.85% :     0.000001s :      8: predicate.specialize_transform
 1.52% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.04% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.15% :     0.000003s :     18: predicate.switch_defer_inline
 1.78% :     0.000003s :     18: predicate.switch_layer_defer_inline
 7.18% :     0.000012s :     60: predicate.switch_simplify
 1.61% :     0.000003s :     12: predicate.tile_eliminate
 1.21% :     0.000002s :     12: predicate.transpose_eliminate
 1.44% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.49% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.78% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.41% :     0.000002s :     18: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000004s :     26: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000003s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      4: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.003060    26
71.86% :     0.002199s :     19: func_graph_cloner_run.FuncGraphClonerGraph
 2.07% :     0.000063s :      1: func_graph_cloner_run.FuncGraphClonerNode
26.07% :     0.000798s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.821769    76
 0.02% :     0.000132s :      1: add_recomputation
 0.03% :     0.000263s :      1: auto_monad
 0.00% :     0.000031s :      1: auto_monad_reorder
 0.00% :     0.000006s :      1: backend_pass
 0.06% :     0.000453s :      1: bootstrap
 0.01% :     0.000041s :      1: cconv
 0.00% :     0.000013s :      1: convert_after_rewriter
 0.00% :     0.000033s :      1: cse_after_recomputation
 0.00% :     0.000041s :      1: environ_conv
 0.00% :     0.000031s :      1: event_method
 0.00% :     0.000010s :      1: execute
 0.00% :     0.000006s :      1: expand_dump_flag
 0.00% :     0.000005s :      1: get_jit_bprop_graph
 0.00% :     0.000010s :      1: graph_reusing
 2.10% :     0.017266s :      1: jit_opt_a
 0.02% :     0.000203s :      1: jit_opt_after_cconv
 0.01% :     0.000076s :      1: jit_opt_b
 0.08% :     0.000685s :      1: loop_unroll
 0.12% :     0.001005s :      1: mutable_eliminate
 0.15% :     0.001243s :     26: opt.transform.jit_opt_a
 0.01% :     0.000067s :      4: opt.transform.jit_opt_after_cconv
 0.00% :     0.000039s :      4: opt.transform.jit_opt_b
 0.00% :     0.000020s :      1: opt.transform.loop_unroll_optimizer
 0.00% :     0.000028s :      1: opt.transform.mutable_eliminate
 0.00% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.00% :     0.000041s :      4: opt.transform.symbol_engine_opt
 0.08% :     0.000654s :      1: opt_after_jit_grad
 0.00% :     0.000009s :      1: order_py_execute_after_rewriter
 0.00% :     0.000005s :      1: partial_unused_args_eliminate
 0.00% :     0.000014s :      1: pre_auto_parallel
 0.05% :     0.000427s :      1: py_interpret_to_execute
 0.00% :     0.000026s :      1: py_interpret_to_execute_after_opt_a
 0.00% :     0.000020s :      1: remove_dup_value
 0.25% :     0.002076s :      1: renormalize.infer
 0.10% :     0.000784s :      1: renormalize.specialize
 0.03% :     0.000212s :      1: rewriter_after_jit_bprop_graph
 0.01% :     0.000071s :      1: rewriter_after_opt_a
 0.02% :     0.000153s :      1: rewriter_before_opt_a
 0.01% :     0.000094s :      1: symbol_engine_optimizer
78.21% :     0.642717s :      1: task_emit
18.56% :     0.152528s :      1: type_inference
 0.02% :     0.000200s :      1: validate


TotalTime = 0.0175445, [33]
[bootstrap]: 0.0002527
[type_inference]: 0.00410253
[event_method]: 1.583e-05
[auto_monad]: 6.139e-05
[graph_reusing]: 5.17e-06
[pre_auto_parallel]: 2.28998e-06
[py_interpret_to_execute]: 3.028e-05
[rewriter_before_opt_a]: 6.752e-05
[expand_dump_flag]: 3.13998e-06
[jit_opt_a]: 0.0049569, [2]
    [Cycle 1]: 0.00142039, [27]
        [switch_simplify]: 4.767e-05
        [loop_unroll]: 2.86e-05
        [a_1]: 0.00049901
        [with_stream_mark]: 1.419e-05
        [recompute_prepare]: 7.70998e-06
        [updatestate_depend_eliminate]: 2.74999e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [parameter_eliminate]: 1.67001e-06
        [specialize_transform]: 5.96e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 5.17999e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 5.04e-06
        [merge_forward]: 3.39001e-06
        [cell_reuse_recompute_pass]: 1.04998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.332e-05
        [j_node_and_user_rematch]: 8.71002e-06
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 8.52998e-06
        [inline_without_move]: 5.12e-06
        [renormalize]: 0.00054426
        [add_forward_monad_depend]: 4.94003e-06
        [auto_monad_grad]: 1.96e-06
        [auto_monad_eliminator]: 1.27e-05
        [cse]: 1.575e-05
        [replace_applicator]: 1.122e-05
    [Cycle 2]: 0.00032296, [27]
        [switch_simplify]: 5.85002e-06
        [loop_unroll]: 5.02999e-06
        [a_1]: 9.07e-05
        [with_stream_mark]: 8.38999e-06
        [recompute_prepare]: 4.69998e-06
        [updatestate_depend_eliminate]: 2.43e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 4.93001e-06
        [updatestate_useless_node_eliminater]: 4.91002e-06
        [accelerated_algorithm]: 1.073e-05
        [meta_shard_fg_expand]: 1.19e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.33998e-06
        [cell_reuse_recompute_pass]: 1.54998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.249e-05
        [j_node_and_user_rematch]: 8.25999e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 7.18998e-06
        [inline_without_move]: 4.70001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.55999e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 4.97e-06
        [cse]: 9.40001e-06
        [replace_applicator]: 5.25001e-06
[py_interpret_to_execute_after_opt_a]: 9.39998e-06
[rewriter_after_opt_a]: 3.013e-05
[convert_after_rewriter]: 5.98998e-06
[order_py_execute_after_rewriter]: 4.82998e-06
[mutable_eliminate]: 0.00051526
[jit_opt_b]: 4.766e-05, [1]
    [Cycle 1]: 4.108e-05, [2]
        [frontend_op_eliminate]: 1.546e-05
        [inline_after_opt_a]: 1.446e-05
[cconv]: 2.337e-05
[loop_unroll]: 0.00042084
[jit_opt_after_cconv]: 0.00013357, [1]
    [Cycle 1]: 0.0001274, [11]
        [c_1]: 2.029e-05
        [parameter_eliminate]: 2.53e-06
        [updatestate_depend_eliminate]: 5.09998e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 1.556e-05
        [call_graph_tuple_transform]: 1.772e-05
        [tuple_list_get_item_eliminator]: 5.10999e-06
        [none_parameter_eliminate]: 1.47999e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 5.72999e-06
[remove_dup_value]: 1.199e-05
[partial_unused_args_eliminate]: 2.02001e-06
[environ_conv]: 5.69e-06
[add_recomputation]: 4.275e-05
[cse_after_recomputation]: 1.953e-05, [1]
    [Cycle 1]: 1.416e-05, [1]
        [cse]: 8.67e-06
[auto_monad_reorder]: 1.41e-05
[get_jit_bprop_graph]: 1.79e-06
[rewriter_after_jit_bprop_graph]: 4.90999e-06
[opt_after_jit_grad]: 0.00045126
[symbol_engine_optimizer]: 6.937e-05, [1]
    [Cycle 1]: 6.355e-05, [6]
        [build]: 3.03e-06
        [elim_shapecalc]: 7.25e-06
        [elim_not_effective]: 1.233e-05
        [opt_reshape]: 5.60001e-06
        [fold_const_symbol]: 8.18999e-06
        [renormalize]: 4.89992e-07
[validate]: 3.031e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00601565
[execute]: 8.69998e-06
Sums
    bootstrap                            :     0.000253s :  1.84%
    type_inference                       :     0.004103s : 29.80%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000061s :  0.45%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.22%
    rewriter_before_opt_a                :     0.000068s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000590s :  4.28%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000544s :  3.95%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000025s :  0.18%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000030s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000515s :  3.74%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000421s :  3.06%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000043s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000451s :  3.28%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000030s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006016s : 43.70%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000174    23
 1.22% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.81% :     0.000005s :      3: substitution.graph_param_transform
82.47% :     0.000144s :      4: substitution.inline
 1.62% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.43% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000003s :      2: substitution.replace_old_param
 6.71% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004047     2
81.54% :     0.003300s :      1: type_inference.infer
18.46% :     0.000747s :      1: type_inference.specialize
------[replace.]   0.000064     6
66.09% :     0.000042s :      4: replace.inline
33.91% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000152     6
93.02% :     0.000141s :      4: match.inline
 6.98% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.15% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.36% :     0.000002s :     10: predicate.addn_check_dump
 1.71% :     0.000002s :     10: predicate.addn_zero_filter
 2.26% :     0.000003s :     10: predicate.arithmetic_simplify
 1.36% :     0.000002s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.73% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.32% :     0.000006s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.68% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.86% :     0.000002s :     15: predicate.load_eliminater
 1.55% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.93% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.22% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.41% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000002s :     15: predicate.partial_eliminate
 1.24% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.97% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.48% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.48% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.59% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.53% :     0.000009s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.33% :     0.000002s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.33% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.83% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.07% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000493    12
48.39% :     0.000239s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.78% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.83% :     0.000226s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018984    76
 0.24% :     0.000046s :      1: add_recomputation
 0.34% :     0.000065s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.38% :     0.000262s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.07% :     0.000013s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.12% :     0.004960s :      1: jit_opt_a
 0.72% :     0.000136s :      1: jit_opt_after_cconv
 0.27% :     0.000050s :      1: jit_opt_b
 2.26% :     0.000428s :      1: loop_unroll
 2.75% :     0.000523s :      1: mutable_eliminate
 4.18% :     0.000793s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.41% :     0.000458s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 1.37% :     0.000261s :      1: renormalize.infer
 1.46% :     0.000277s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000033s :      1: rewriter_after_opt_a
 0.37% :     0.000071s :      1: rewriter_before_opt_a
 0.38% :     0.000072s :      1: symbol_engine_optimizer
31.77% :     0.006031s :      1: task_emit
21.67% :     0.004114s :      1: type_inference
 0.25% :     0.000048s :      1: validate


TotalTime = 0.0169794, [33]
[bootstrap]: 0.00027246
[type_inference]: 0.00427108
[event_method]: 1.717e-05
[auto_monad]: 5.992e-05
[graph_reusing]: 5.47001e-06
[pre_auto_parallel]: 2.36e-06
[py_interpret_to_execute]: 2.96e-05
[rewriter_before_opt_a]: 7.121e-05
[expand_dump_flag]: 2.83e-06
[jit_opt_a]: 0.00496546, [2]
    [Cycle 1]: 0.00138051, [27]
        [switch_simplify]: 4.871e-05
        [loop_unroll]: 2.84e-05
        [a_1]: 0.00049674
        [with_stream_mark]: 1.531e-05
        [recompute_prepare]: 7.31999e-06
        [updatestate_depend_eliminate]: 3.68999e-06
        [updatestate_assign_eliminate]: 3.4e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [parameter_eliminate]: 1.81998e-06
        [specialize_transform]: 5.86e-06
        [updatestate_useless_node_eliminater]: 5.26998e-06
        [accelerated_algorithm]: 5.22999e-06
        [meta_shard_fg_expand]: 1.88002e-06
        [get_grad_eliminate_]: 5.29998e-06
        [merge_forward]: 3.53e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.379e-05
        [j_node_and_user_rematch]: 8.76002e-06
        [meta_fg_expand]: 2.19001e-06
        [replace_old_param]: 8.15e-06
        [inline_without_move]: 4.97999e-06
        [renormalize]: 0.00049424
        [add_forward_monad_depend]: 4.90001e-06
        [auto_monad_grad]: 1.98002e-06
        [auto_monad_eliminator]: 1.291e-05
        [cse]: 2.129e-05
        [replace_applicator]: 1.073e-05
    [Cycle 2]: 0.00030591, [27]
        [switch_simplify]: 5.25001e-06
        [loop_unroll]: 4.68999e-06
        [a_1]: 8.632e-05
        [with_stream_mark]: 7.9e-06
        [recompute_prepare]: 4.83001e-06
        [updatestate_depend_eliminate]: 2.33998e-06
        [updatestate_assign_eliminate]: 1.92001e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 9.69972e-07
        [specialize_transform]: 4.75999e-06
        [updatestate_useless_node_eliminater]: 4.62998e-06
        [accelerated_algorithm]: 5.27001e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 4.73001e-06
        [merge_forward]: 2.31e-06
        [cell_reuse_recompute_pass]: 1.37999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.134e-05
        [j_node_and_user_rematch]: 7.82998e-06
        [meta_fg_expand]: 1.35999e-06
        [replace_old_param]: 6.89999e-06
        [inline_without_move]: 4.48001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.08001e-06
        [auto_monad_grad]: 8.10018e-07
        [auto_monad_eliminator]: 4.84e-06
        [cse]: 9.62999e-06
        [replace_applicator]: 4.87e-06
[py_interpret_to_execute_after_opt_a]: 8.27998e-06
[rewriter_after_opt_a]: 2.828e-05
[convert_after_rewriter]: 6.56e-06
[order_py_execute_after_rewriter]: 5.12999e-06
[mutable_eliminate]: 0.0004723
[jit_opt_b]: 4.549e-05, [1]
    [Cycle 1]: 3.899e-05, [2]
        [frontend_op_eliminate]: 1.443e-05
        [inline_after_opt_a]: 1.391e-05
[cconv]: 2.447e-05
[loop_unroll]: 0.00042365
[jit_opt_after_cconv]: 0.00013046, [1]
    [Cycle 1]: 0.00012429, [11]
        [c_1]: 1.95e-05
        [parameter_eliminate]: 2.32001e-06
        [updatestate_depend_eliminate]: 4.97e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.508e-05
        [call_graph_tuple_transform]: 1.796e-05
        [tuple_list_get_item_eliminator]: 5.22e-06
        [none_parameter_eliminate]: 1.17e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 5.07e-06
[remove_dup_value]: 1.359e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 4.53999e-06
[add_recomputation]: 4.235e-05
[cse_after_recomputation]: 1.87e-05, [1]
    [Cycle 1]: 1.364e-05, [1]
        [cse]: 8.2e-06
[auto_monad_reorder]: 1.36e-05
[get_jit_bprop_graph]: 1.40999e-06
[rewriter_after_jit_bprop_graph]: 4.43999e-06
[opt_after_jit_grad]: 0.00043722
[symbol_engine_optimizer]: 6.956e-05, [1]
    [Cycle 1]: 6.322e-05, [6]
        [build]: 2.99001e-06
        [elim_shapecalc]: 7.23999e-06
        [elim_not_effective]: 1.191e-05
        [opt_reshape]: 5.59e-06
        [fold_const_symbol]: 8.26002e-06
        [renormalize]: 4.89992e-07
[validate]: 3.143e-05
[backend_pass]: 1.17e-06
[task_emit]: 0.00531921
[execute]: 6.58e-06
Sums
    bootstrap                            :     0.000272s :  2.07%
    type_inference                       :     0.004271s : 32.50%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000060s :  0.46%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.23%
    rewriter_before_opt_a                :     0.000071s :  0.54%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.41%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000583s :  4.44%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000494s :  3.76%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.14%
    jit_opt_a.cse                        :     0.000031s :  0.24%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000028s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000472s :  3.59%
    jit_opt_b.frontend_op_eliminate      :     0.000014s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000024s :  0.19%
    loop_unroll                          :     0.000424s :  3.22%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000042s :  0.32%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000437s :  3.33%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005319s : 40.48%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000182    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.85% :     0.000005s :      3: substitution.graph_param_transform
84.12% :     0.000153s :      4: substitution.inline
 1.59% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000003s :      2: substitution.replace_old_param
 5.67% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004214     2
81.90% :     0.003451s :      1: type_inference.infer
18.10% :     0.000763s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.69% :     0.000042s :      4: replace.inline
32.31% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000160     6
94.18% :     0.000150s :      4: match.inline
 5.82% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.33% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.21% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000001s :     10: predicate.addn_zero_filter
 1.74% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.21% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.49% :     0.000001s :      3: predicate.elim_not_effective
 0.72% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.24% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000000s :      3: predicate.graph_param_transform
 5.73% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      6: predicate.less_batch_normalization
 1.81% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 1.35% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.19% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.17% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.34% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.42% :     0.000002s :     10: predicate.reduce_eliminate
 1.78% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000001s :     10: predicate.reshape_eliminate
 1.30% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.47% :     0.000002s :     10: predicate.same_eliminate
 0.64% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.30% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.82% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.34% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.44% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000527    12
48.50% :     0.000256s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.12% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.38% :     0.000234s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018356    76
 0.25% :     0.000045s :      1: add_recomputation
 0.34% :     0.000063s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.54% :     0.000283s :      1: bootstrap
 0.15% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.12% :     0.000022s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.06% :     0.004968s :      1: jit_opt_a
 0.73% :     0.000133s :      1: jit_opt_after_cconv
 0.26% :     0.000048s :      1: jit_opt_b
 2.35% :     0.000432s :      1: loop_unroll
 2.62% :     0.000480s :      1: mutable_eliminate
 4.24% :     0.000779s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.42% :     0.000445s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 1.40% :     0.000258s :      1: renormalize.infer
 1.25% :     0.000230s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000033s :      1: rewriter_after_opt_a
 0.40% :     0.000074s :      1: rewriter_before_opt_a
 0.39% :     0.000072s :      1: symbol_engine_optimizer
29.03% :     0.005328s :      1: task_emit
23.34% :     0.004284s :      1: type_inference
 0.26% :     0.000047s :      1: validate


TotalTime = 0.0192461, [33]
[bootstrap]: 0.00027402
[type_inference]: 0.00435354
[event_method]: 1.708e-05
[auto_monad]: 6.435e-05
[graph_reusing]: 5.26998e-06
[pre_auto_parallel]: 2.29999e-06
[py_interpret_to_execute]: 3.383e-05
[rewriter_before_opt_a]: 7.34e-05
[expand_dump_flag]: 3.47997e-06
[jit_opt_a]: 0.00555367, [2]
    [Cycle 1]: 0.00172393, [27]
        [switch_simplify]: 5.59e-05
        [loop_unroll]: 2.986e-05
        [a_1]: 0.00058371
        [with_stream_mark]: 2.262e-05
        [recompute_prepare]: 7.77e-06
        [updatestate_depend_eliminate]: 3.84002e-06
        [updatestate_assign_eliminate]: 3.41999e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [parameter_eliminate]: 2.34999e-06
        [specialize_transform]: 6.26998e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.90002e-06
        [meta_shard_fg_expand]: 2.19999e-06
        [get_grad_eliminate_]: 5.37001e-06
        [merge_forward]: 4.13999e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.443e-05
        [j_node_and_user_rematch]: 9.37999e-06
        [meta_fg_expand]: 2.34999e-06
        [replace_old_param]: 1.06e-05
        [inline_without_move]: 5.78002e-06
        [renormalize]: 0.0006823
        [add_forward_monad_depend]: 6.38e-06
        [auto_monad_grad]: 2.54999e-06
        [auto_monad_eliminator]: 1.453e-05
        [cse]: 3.099e-05
        [replace_applicator]: 1.359e-05
    [Cycle 2]: 0.0003304, [27]
        [switch_simplify]: 5.44e-06
        [loop_unroll]: 5.02e-06
        [a_1]: 9.478e-05
        [with_stream_mark]: 1.019e-05
        [recompute_prepare]: 5.25999e-06
        [updatestate_depend_eliminate]: 2.66e-06
        [updatestate_assign_eliminate]: 2.47001e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 1.26002e-06
        [specialize_transform]: 4.60999e-06
        [updatestate_useless_node_eliminater]: 5.33002e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.36002e-06
        [get_grad_eliminate_]: 5.02e-06
        [merge_forward]: 2.99001e-06
        [cell_reuse_recompute_pass]: 2.25002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.521e-05
        [j_node_and_user_rematch]: 8.63001e-06
        [meta_fg_expand]: 1.71998e-06
        [replace_old_param]: 7.63001e-06
        [inline_without_move]: 4.63001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.62001e-06
        [auto_monad_grad]: 7.29982e-07
        [auto_monad_eliminator]: 5.28002e-06
        [cse]: 1.055e-05
        [replace_applicator]: 5.20999e-06
[py_interpret_to_execute_after_opt_a]: 1.079e-05
[rewriter_after_opt_a]: 3.646e-05
[convert_after_rewriter]: 6.96001e-06
[order_py_execute_after_rewriter]: 5.75001e-06
[mutable_eliminate]: 0.00064309
[jit_opt_b]: 5.169e-05, [1]
    [Cycle 1]: 4.331e-05, [2]
        [frontend_op_eliminate]: 1.647e-05
        [inline_after_opt_a]: 1.548e-05
[cconv]: 2.718e-05
[loop_unroll]: 0.00043863
[jit_opt_after_cconv]: 0.00014807, [1]
    [Cycle 1]: 0.0001413, [11]
        [c_1]: 2.2e-05
        [parameter_eliminate]: 2.90002e-06
        [updatestate_depend_eliminate]: 6.19001e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [cse]: 2.256e-05
        [call_graph_tuple_transform]: 2.028e-05
        [tuple_list_get_item_eliminator]: 5.25999e-06
        [none_parameter_eliminate]: 1.71e-06
        [renormalize]: 7.79983e-07
        [switch_simplify]: 5.14998e-06
[remove_dup_value]: 1.343e-05
[partial_unused_args_eliminate]: 2.12001e-06
[environ_conv]: 5.98002e-06
[add_recomputation]: 5.114e-05
[cse_after_recomputation]: 2.171e-05, [1]
    [Cycle 1]: 1.655e-05, [1]
        [cse]: 1.012e-05
[auto_monad_reorder]: 1.516e-05
[get_jit_bprop_graph]: 1.91e-06
[rewriter_after_jit_bprop_graph]: 6.17001e-06
[opt_after_jit_grad]: 0.00048161
[symbol_engine_optimizer]: 7.629e-05, [1]
    [Cycle 1]: 6.997e-05, [6]
        [build]: 5.36002e-06
        [elim_shapecalc]: 7.63999e-06
        [elim_not_effective]: 1.364e-05
        [opt_reshape]: 5.87001e-06
        [fold_const_symbol]: 8.67e-06
        [renormalize]: 6.30011e-07
[validate]: 3.764e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.00656463
[execute]: 7.31999e-06
Sums
    bootstrap                            :     0.000274s :  1.81%
    type_inference                       :     0.004354s : 28.81%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000064s :  0.43%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.22%
    rewriter_before_opt_a                :     0.000073s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.41%
    jit_opt_a.loop_unroll                :     0.000035s :  0.23%
    jit_opt_a.a_1                        :     0.000678s :  4.49%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000682s :  4.52%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000042s :  0.27%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000643s :  4.26%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000027s :  0.18%
    loop_unroll                          :     0.000439s :  2.90%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000051s :  0.34%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000482s :  3.19%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006565s : 43.44%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000226    23
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 2.40% :     0.000005s :      3: substitution.graph_param_transform
84.97% :     0.000192s :      4: substitution.inline
 1.76% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.87% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.37% :     0.000005s :      2: substitution.replace_old_param
 5.29% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004289     2
81.03% :     0.003476s :      1: type_inference.infer
18.97% :     0.000814s :      1: type_inference.specialize
------[replace.]   0.000072     6
68.30% :     0.000049s :      4: replace.inline
31.70% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     6
94.59% :     0.000189s :      4: match.inline
 5.41% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   691
 1.29% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.71% :     0.000002s :     10: predicate.addn_zero_filter
 2.45% :     0.000003s :     10: predicate.arithmetic_simplify
 1.32% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.59% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.70% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.22% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.31% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.77% :     0.000007s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.91% :     0.000002s :     15: predicate.load_eliminater
 1.29% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.74% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.34% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 2.14% :     0.000003s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     15: predicate.partial_eliminate
 1.30% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.30% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.24% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.09% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.13% :     0.000004s :     16: predicate.switch_defer_inline
 2.05% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.39% :     0.000010s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.41% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.27% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000572    12
48.09% :     0.000275s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.74% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.17% :     0.000253s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020933    76
 0.26% :     0.000054s :      1: add_recomputation
 0.33% :     0.000068s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.39% :     0.000291s :      1: bootstrap
 0.14% :     0.000030s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.55% :     0.005557s :      1: jit_opt_a
 0.72% :     0.000151s :      1: jit_opt_after_cconv
 0.26% :     0.000055s :      1: jit_opt_b
 2.14% :     0.000447s :      1: loop_unroll
 3.12% :     0.000654s :      1: mutable_eliminate
 4.28% :     0.000896s :     26: opt.transform.jit_opt_a
 0.24% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.35% :     0.000491s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.81% :     0.000378s :      1: renormalize.infer
 1.41% :     0.000295s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000039s :      1: rewriter_after_opt_a
 0.36% :     0.000076s :      1: rewriter_before_opt_a
 0.38% :     0.000079s :      1: symbol_engine_optimizer
31.44% :     0.006581s :      1: task_emit
20.86% :     0.004368s :      1: type_inference
 0.28% :     0.000059s :      1: validate


TotalTime = 0.0187938, [33]
[bootstrap]: 0.00026256
[type_inference]: 0.00418746
[event_method]: 1.716e-05
[auto_monad]: 6.014e-05
[graph_reusing]: 5.19e-06
[pre_auto_parallel]: 2.34999e-06
[py_interpret_to_execute]: 3.047e-05
[rewriter_before_opt_a]: 6.882e-05
[expand_dump_flag]: 2.99999e-06
[jit_opt_a]: 0.00523132, [2]
    [Cycle 1]: 0.00154362, [27]
        [switch_simplify]: 5.188e-05
        [loop_unroll]: 2.81e-05
        [a_1]: 0.00052079
        [with_stream_mark]: 2.045e-05
        [recompute_prepare]: 6.64001e-06
        [updatestate_depend_eliminate]: 3.23998e-06
        [updatestate_assign_eliminate]: 2.72001e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 2.11e-06
        [specialize_transform]: 6.09001e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 5.18002e-06
        [meta_shard_fg_expand]: 1.89999e-06
        [get_grad_eliminate_]: 5.22e-06
        [merge_forward]: 3.53e-06
        [cell_reuse_recompute_pass]: 1.16002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.417e-05
        [j_node_and_user_rematch]: 8.2e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 9.08002e-06
        [inline_without_move]: 5.14e-06
        [renormalize]: 0.0006136
        [add_forward_monad_depend]: 4.79002e-06
        [auto_monad_grad]: 2.39999e-06
        [auto_monad_eliminator]: 1.266e-05
        [cse]: 2.465e-05
        [replace_applicator]: 1.277e-05
    [Cycle 2]: 0.0003152, [27]
        [switch_simplify]: 6.06e-06
        [loop_unroll]: 5.30999e-06
        [a_1]: 8.844e-05
        [with_stream_mark]: 1.029e-05
        [recompute_prepare]: 4.94998e-06
        [updatestate_depend_eliminate]: 2.72001e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 5.09e-06
        [updatestate_useless_node_eliminater]: 4.58001e-06
        [accelerated_algorithm]: 4.47e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.51002e-06
        [merge_forward]: 2.50002e-06
        [cell_reuse_recompute_pass]: 1.45001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.327e-05
        [j_node_and_user_rematch]: 8.03001e-06
        [meta_fg_expand]: 1.72001e-06
        [replace_old_param]: 6.74999e-06
        [inline_without_move]: 4.4e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 6.99976e-07
        [auto_monad_eliminator]: 5.14e-06
        [cse]: 9.96e-06
        [replace_applicator]: 4.72e-06
[py_interpret_to_execute_after_opt_a]: 1.017e-05
[rewriter_after_opt_a]: 3.127e-05
[convert_after_rewriter]: 6.48998e-06
[order_py_execute_after_rewriter]: 5.09003e-06
[mutable_eliminate]: 0.00057585
[jit_opt_b]: 4.786e-05, [1]
    [Cycle 1]: 4.042e-05, [2]
        [frontend_op_eliminate]: 1.432e-05
        [inline_after_opt_a]: 1.46e-05
[cconv]: 2.664e-05
[loop_unroll]: 0.0004235
[jit_opt_after_cconv]: 0.00013901, [1]
    [Cycle 1]: 0.00013267, [11]
        [c_1]: 1.903e-05
        [parameter_eliminate]: 2.64999e-06
        [updatestate_depend_eliminate]: 5.82999e-06
        [updatestate_assign_eliminate]: 2.40002e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 2.118e-05
        [call_graph_tuple_transform]: 1.821e-05
        [tuple_list_get_item_eliminator]: 5.52001e-06
        [none_parameter_eliminate]: 1.20001e-06
        [renormalize]: 5.19998e-07
        [switch_simplify]: 5.10001e-06
[remove_dup_value]: 1.478e-05
[partial_unused_args_eliminate]: 2.12001e-06
[environ_conv]: 4.56002e-06
[add_recomputation]: 4.549e-05
[cse_after_recomputation]: 1.972e-05, [1]
    [Cycle 1]: 1.457e-05, [1]
        [cse]: 9.10001e-06
[auto_monad_reorder]: 1.432e-05
[get_jit_bprop_graph]: 1.87999e-06
[rewriter_after_jit_bprop_graph]: 5.27999e-06
[opt_after_jit_grad]: 0.00047755
[symbol_engine_optimizer]: 7.397e-05, [1]
    [Cycle 1]: 6.739e-05, [6]
        [build]: 4.47998e-06
        [elim_shapecalc]: 7.71999e-06
        [elim_not_effective]: 1.257e-05
        [opt_reshape]: 5.34e-06
        [fold_const_symbol]: 8.28999e-06
        [renormalize]: 5.40022e-07
[validate]: 3.379e-05
[backend_pass]: 1.00999e-06
[task_emit]: 0.00677271
[execute]: 8.55001e-06
Sums
    bootstrap                            :     0.000263s :  1.77%
    type_inference                       :     0.004187s : 28.22%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000060s :  0.41%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.21%
    rewriter_before_opt_a                :     0.000069s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.39%
    jit_opt_a.loop_unroll                :     0.000033s :  0.23%
    jit_opt_a.a_1                        :     0.000609s :  4.11%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000614s :  4.14%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000035s :  0.23%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000031s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000576s :  3.88%
    jit_opt_b.frontend_op_eliminate      :     0.000014s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000027s :  0.18%
    loop_unroll                          :     0.000423s :  2.85%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000045s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000478s :  3.22%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006773s : 45.65%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000205    23
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.62% :     0.000005s :      3: substitution.graph_param_transform
84.42% :     0.000173s :      4: substitution.inline
 1.58% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.29% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.15% :     0.000004s :      2: substitution.replace_old_param
 5.43% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004130     2
81.96% :     0.003385s :      1: type_inference.infer
18.04% :     0.000745s :      1: type_inference.specialize
------[replace.]   0.000060     6
67.06% :     0.000040s :      4: replace.inline
32.94% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000181     6
94.48% :     0.000171s :      4: match.inline
 5.52% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.39% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.53% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.46% :     0.000002s :     10: predicate.addn_zero_filter
 1.80% :     0.000002s :     10: predicate.arithmetic_simplify
 1.33% :     0.000001s :     10: predicate.cast_eliminate
 0.61% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.17% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000000s :      3: predicate.graph_param_transform
 5.74% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.87% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.90% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000000s :      3: predicate.opt_reshape
 2.23% :     0.000002s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.67% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.75% :     0.000009s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.28% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.76% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.48% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000511    12
45.81% :     0.000234s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.78% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.41% :     0.000232s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020324    76
 0.24% :     0.000048s :      1: add_recomputation
 0.31% :     0.000064s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.35% :     0.000274s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
25.76% :     0.005235s :      1: jit_opt_a
 0.70% :     0.000142s :      1: jit_opt_after_cconv
 0.25% :     0.000050s :      1: jit_opt_b
 2.12% :     0.000431s :      1: loop_unroll
 2.88% :     0.000585s :      1: mutable_eliminate
 3.99% :     0.000811s :     26: opt.transform.jit_opt_a
 0.22% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.39% :     0.000487s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.70% :     0.000346s :      1: renormalize.infer
 1.28% :     0.000261s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000035s :      1: rewriter_after_opt_a
 0.35% :     0.000072s :      1: rewriter_before_opt_a
 0.38% :     0.000077s :      1: symbol_engine_optimizer
33.39% :     0.006787s :      1: task_emit
20.66% :     0.004199s :      1: type_inference
 0.26% :     0.000053s :      1: validate


TotalTime = 0.0130545, [30]
[bootstrap]: 0.00025069
[type_inference]: 0.00587602
[event_method]: 1.903e-05
[auto_monad]: 7.42e-05
[graph_reusing]: 5.55001e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 4.68e-05
[rewriter_before_opt_a]: 8.174e-05
[expand_dump_flag]: 2.68e-06
[jit_opt_a]: 0.00480542, [2]
    [Cycle 1]: 0.0015266, [27]
        [switch_simplify]: 9.481e-05
        [loop_unroll]: 3.496e-05
        [a_1]: 0.00062657
        [with_stream_mark]: 1.209e-05
        [recompute_prepare]: 7.06999e-06
        [updatestate_depend_eliminate]: 2.99999e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 5.87999e-06
        [updatestate_useless_node_eliminater]: 5.44e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 1.45999e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 2.31998e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.149e-05
        [j_node_and_user_rematch]: 8.93002e-06
        [meta_fg_expand]: 2.11e-06
        [replace_old_param]: 8.2e-06
        [inline_without_move]: 5.40999e-06
        [renormalize]: 0.0004748
        [add_forward_monad_depend]: 4.77e-06
        [auto_monad_grad]: 1.41002e-06
        [auto_monad_eliminator]: 1.041e-05
        [cse]: 1.725e-05
        [replace_applicator]: 1.094e-05
    [Cycle 2]: 0.00032167, [27]
        [switch_simplify]: 5.34e-06
        [loop_unroll]: 4.81002e-06
        [a_1]: 9.719e-05
        [with_stream_mark]: 8.74e-06
        [recompute_prepare]: 5.27999e-06
        [updatestate_depend_eliminate]: 2.53e-06
        [updatestate_assign_eliminate]: 2.00002e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.09998e-06
        [specialize_transform]: 5.09e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 4.58999e-06
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.79002e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 1.28002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.194e-05
        [j_node_and_user_rematch]: 8.22998e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 6.83e-06
        [inline_without_move]: 4.92e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 5.27001e-06
        [cse]: 1.045e-05
        [replace_applicator]: 4.90999e-06
[py_interpret_to_execute_after_opt_a]: 7.98999e-06
[rewriter_after_opt_a]: 2.65e-05
[convert_after_rewriter]: 6.50002e-06
[order_py_execute_after_rewriter]: 4.48001e-06
[mutable_eliminate]: 0.00046637
[jit_opt_b]: 5.313e-05, [1]
    [Cycle 1]: 4.053e-05, [2]
        [frontend_op_eliminate]: 1.499e-05
        [inline_after_opt_a]: 1.411e-05
[cconv]: 1.637e-05
[loop_unroll]: 0.00040893
[jit_opt_after_cconv]: 0.00013071, [1]
    [Cycle 1]: 0.00012497, [11]
        [c_1]: 2.008e-05
        [parameter_eliminate]: 2.31e-06
        [updatestate_depend_eliminate]: 5.68002e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [cse]: 1.498e-05
        [call_graph_tuple_transform]: 1.661e-05
        [tuple_list_get_item_eliminator]: 5.12999e-06
        [none_parameter_eliminate]: 7.2e-07
        [renormalize]: 3.29979e-07
        [switch_simplify]: 5.25001e-06
[remove_dup_value]: 7.73999e-06
[partial_unused_args_eliminate]: 2.02001e-06
[environ_conv]: 3.93001e-06
[add_recomputation]: 3.396e-05
[cse_after_recomputation]: 1.982e-05, [1]
    [Cycle 1]: 1.472e-05, [1]
        [cse]: 8.89e-06
[auto_monad_reorder]: 1.081e-05
[get_jit_bprop_graph]: 1.54998e-06
[rewriter_after_jit_bprop_graph]: 4.38001e-06
[opt_after_jit_grad]: 0.00043657
[symbol_engine_optimizer]: 6.956e-05, [1]
    [Cycle 1]: 6.37e-05, [6]
        [build]: 2.75002e-06
        [elim_shapecalc]: 7.53e-06
        [elim_not_effective]: 1.223e-05
        [opt_reshape]: 5.44e-06
        [fold_const_symbol]: 8.01001e-06
        [renormalize]: 4.30009e-07
[validate]: 2.609e-05
Sums
    bootstrap                            :     0.000251s :  2.63%
    type_inference                       :     0.005876s : 61.54%
    event_method                         :     0.000019s :  0.20%
    auto_monad                           :     0.000074s :  0.78%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000047s :  0.49%
    rewriter_before_opt_a                :     0.000082s :  0.86%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000100s :  1.05%
    jit_opt_a.loop_unroll                :     0.000040s :  0.42%
    jit_opt_a.a_1                        :     0.000724s :  7.58%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000475s :  4.97%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.16%
    jit_opt_a.cse                        :     0.000028s :  0.29%
    jit_opt_a.replace_applicator         :     0.000016s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.08%
    rewriter_after_opt_a                 :     0.000027s :  0.28%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000466s :  4.88%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.15%
    cconv                                :     0.000016s :  0.17%
    loop_unroll                          :     0.000409s :  4.28%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000008s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000034s :  0.36%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000011s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000437s :  4.57%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000026s :  0.27%


Time group info:
------[substitution.]   0.000223    26
 0.73% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 1.83% :     0.000004s :      3: substitution.graph_param_transform
84.43% :     0.000188s :      6: substitution.inline
 1.33% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.80% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.34% :     0.000003s :      2: substitution.replace_old_param
 3.47% :     0.000008s :      1: substitution.switch_simplify
 4.59% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005825     2
83.61% :     0.004870s :      1: type_inference.infer
16.39% :     0.000955s :      1: type_inference.specialize
------[replace.]   0.000114     9
53.99% :     0.000062s :      6: replace.inline
20.81% :     0.000024s :      1: replace.switch_simplify
25.20% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000201     9
91.97% :     0.000185s :      6: match.inline
 3.49% :     0.000007s :      1: match.switch_simplify
 4.54% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   801
 1.41% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.07% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     12: predicate.addn_check_dump
 1.37% :     0.000002s :     12: predicate.addn_zero_filter
 1.77% :     0.000002s :     12: predicate.arithmetic_simplify
 1.26% :     0.000002s :     12: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000002s :     12: predicate.compare_switch_simplify
 1.44% :     0.000002s :     12: predicate.depend_value_elim
 1.31% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.29% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     12: predicate.environ_get_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.30% :     0.000007s :     26: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     17: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.50% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.08% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.22% :     0.000002s :     12: predicate.merge_addn
 1.16% :     0.000001s :     12: predicate.minmaximum_grad
 1.60% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     17: predicate.partial_eliminate
 1.27% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     12: predicate.reduce_eliminate
 1.73% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     20: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     12: predicate.reshape_eliminate
 1.32% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.63% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     12: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.81% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.73% :     0.000004s :     20: predicate.switch_defer_inline
 2.60% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.73% :     0.000011s :     65: predicate.switch_simplify
 1.22% :     0.000002s :     12: predicate.tile_eliminate
 1.23% :     0.000002s :     12: predicate.transpose_eliminate
 1.37% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.43% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000697    15
48.11% :     0.000335s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.44% :     0.000073s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.44% :     0.000289s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014570    72
 0.25% :     0.000037s :      1: add_recomputation
 0.53% :     0.000078s :      1: auto_monad
 0.09% :     0.000014s :      1: auto_monad_reorder
 1.82% :     0.000265s :      1: bootstrap
 0.13% :     0.000019s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.16% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
33.00% :     0.004808s :      1: jit_opt_a
 0.92% :     0.000134s :      1: jit_opt_after_cconv
 0.38% :     0.000056s :      1: jit_opt_b
 2.85% :     0.000416s :      1: loop_unroll
 3.26% :     0.000474s :      1: mutable_eliminate
 6.69% :     0.000975s :     26: opt.transform.jit_opt_a
 0.30% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000023s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.05% :     0.000444s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.34% :     0.000050s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000010s :      1: remove_dup_value
 1.67% :     0.000244s :      1: renormalize.infer
 1.54% :     0.000225s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000029s :      1: rewriter_after_opt_a
 0.58% :     0.000085s :      1: rewriter_before_opt_a
 0.50% :     0.000072s :      1: symbol_engine_optimizer
40.40% :     0.005886s :      1: type_inference


TotalTime = 0.0183476, [33]
[bootstrap]: 0.00026854
[type_inference]: 0.00434922
[event_method]: 1.777e-05
[auto_monad]: 6.819e-05
[graph_reusing]: 5.66e-06
[pre_auto_parallel]: 2.81e-06
[py_interpret_to_execute]: 3.302e-05
[rewriter_before_opt_a]: 8.076e-05
[expand_dump_flag]: 3.23e-06
[jit_opt_a]: 0.00499505, [2]
    [Cycle 1]: 0.00146625, [27]
        [switch_simplify]: 5.036e-05
        [loop_unroll]: 3.015e-05
        [a_1]: 0.00051206
        [with_stream_mark]: 1.496e-05
        [recompute_prepare]: 7.55998e-06
        [updatestate_depend_eliminate]: 3.35998e-06
        [updatestate_assign_eliminate]: 3.16001e-06
        [updatestate_loads_eliminate]: 2.81e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 6.02999e-06
        [updatestate_useless_node_eliminater]: 5.27999e-06
        [accelerated_algorithm]: 5.87001e-06
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 5.37999e-06
        [merge_forward]: 3.31999e-06
        [cell_reuse_recompute_pass]: 1.43002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.382e-05
        [j_node_and_user_rematch]: 9.54999e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 8.83001e-06
        [inline_without_move]: 5.54e-06
        [renormalize]: 0.00053864
        [add_forward_monad_depend]: 5.62999e-06
        [auto_monad_grad]: 1.60999e-06
        [auto_monad_eliminator]: 1.437e-05
        [cse]: 3.029e-05
        [replace_applicator]: 1.135e-05
    [Cycle 2]: 0.00032931, [27]
        [switch_simplify]: 5.74e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 9.036e-05
        [with_stream_mark]: 1.025e-05
        [recompute_prepare]: 4.89e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [parameter_eliminate]: 1.11002e-06
        [specialize_transform]: 4.93001e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 1.166e-05
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.72e-06
        [merge_forward]: 2.89001e-06
        [cell_reuse_recompute_pass]: 1.47999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.175e-05
        [j_node_and_user_rematch]: 7.77e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 9.04e-06
        [inline_without_move]: 4.63001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 7.10017e-07
        [auto_monad_eliminator]: 5.02e-06
        [cse]: 1.224e-05
        [replace_applicator]: 5.71003e-06
[py_interpret_to_execute_after_opt_a]: 9.03002e-06
[rewriter_after_opt_a]: 3.175e-05
[convert_after_rewriter]: 7.63999e-06
[order_py_execute_after_rewriter]: 5.19998e-06
[mutable_eliminate]: 0.00053454
[jit_opt_b]: 5.004e-05, [1]
    [Cycle 1]: 4.319e-05, [2]
        [frontend_op_eliminate]: 1.664e-05
        [inline_after_opt_a]: 1.491e-05
[cconv]: 2.535e-05
[loop_unroll]: 0.00041981
[jit_opt_after_cconv]: 0.00013859, [1]
    [Cycle 1]: 0.00013171, [11]
        [c_1]: 2.028e-05
        [parameter_eliminate]: 2.81e-06
        [updatestate_depend_eliminate]: 5.68002e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 1.818e-05
        [call_graph_tuple_transform]: 1.82e-05
        [tuple_list_get_item_eliminator]: 4.86997e-06
        [none_parameter_eliminate]: 1.40001e-06
        [renormalize]: 6.40022e-07
        [switch_simplify]: 5.26002e-06
[remove_dup_value]: 1.373e-05
[partial_unused_args_eliminate]: 2.32001e-06
[environ_conv]: 4.53999e-06
[add_recomputation]: 4.858e-05
[cse_after_recomputation]: 1.981e-05, [1]
    [Cycle 1]: 1.464e-05, [1]
        [cse]: 9.09998e-06
[auto_monad_reorder]: 1.482e-05
[get_jit_bprop_graph]: 1.27e-06
[rewriter_after_jit_bprop_graph]: 5.44e-06
[opt_after_jit_grad]: 0.00048723
[symbol_engine_optimizer]: 7.522e-05, [1]
    [Cycle 1]: 6.742e-05, [6]
        [build]: 3.21001e-06
        [elim_shapecalc]: 7.4e-06
        [elim_not_effective]: 1.264e-05
        [opt_reshape]: 6.34001e-06
        [fold_const_symbol]: 8.89e-06
        [renormalize]: 5.29981e-07
[validate]: 3.421e-05
[backend_pass]: 8.99978e-07
[task_emit]: 0.00638919
[execute]: 8.15999e-06
Sums
    bootstrap                            :     0.000269s :  1.85%
    type_inference                       :     0.004349s : 29.90%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000068s :  0.47%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.23%
    rewriter_before_opt_a                :     0.000081s :  0.56%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000602s :  4.14%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000539s :  3.70%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000043s :  0.29%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000032s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000535s :  3.67%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000420s :  2.89%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000487s :  3.35%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006389s : 43.92%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000183    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.97% :     0.000005s :      3: substitution.graph_param_transform
80.95% :     0.000148s :      4: substitution.inline
 1.62% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.31% :     0.000004s :      4: substitution.remove_not_recompute_node
 3.36% :     0.000006s :      2: substitution.replace_old_param
 6.99% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004287     2
81.47% :     0.003492s :      1: type_inference.infer
18.53% :     0.000794s :      1: type_inference.specialize
------[replace.]   0.000066     6
65.50% :     0.000043s :      4: replace.inline
34.50% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000157     6
92.59% :     0.000146s :      4: match.inline
 7.41% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.36% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 2.28% :     0.000003s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.49% :     0.000002s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.29% :     0.000006s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.41% :     0.000002s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.27% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.65% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.26% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.11% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 2.31% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.76% :     0.000009s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.02% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000534    12
50.01% :     0.000267s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.17% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.82% :     0.000240s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019806    76
 0.26% :     0.000052s :      1: add_recomputation
 0.37% :     0.000072s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.45% :     0.000286s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.23% :     0.004998s :      1: jit_opt_a
 0.71% :     0.000141s :      1: jit_opt_after_cconv
 0.26% :     0.000052s :      1: jit_opt_b
 2.16% :     0.000428s :      1: loop_unroll
 2.74% :     0.000543s :      1: mutable_eliminate
 4.12% :     0.000816s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.50% :     0.000495s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.35% :     0.000268s :      1: renormalize.infer
 1.33% :     0.000264s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000035s :      1: rewriter_after_opt_a
 0.42% :     0.000084s :      1: rewriter_before_opt_a
 0.39% :     0.000078s :      1: symbol_engine_optimizer
32.33% :     0.006402s :      1: task_emit
22.02% :     0.004362s :      1: type_inference
 0.27% :     0.000054s :      1: validate


TotalTime = 0.0154204, [33]
[bootstrap]: 0.00025296
[type_inference]: 0.00390231
[event_method]: 1.581e-05
[auto_monad]: 3.967e-05
[graph_reusing]: 3.73001e-06
[pre_auto_parallel]: 1.66e-06
[py_interpret_to_execute]: 2.649e-05
[rewriter_before_opt_a]: 6.045e-05
[expand_dump_flag]: 2.07001e-06
[jit_opt_a]: 0.00443538, [2]
    [Cycle 1]: 0.00121778, [27]
        [switch_simplify]: 4.166e-05
        [loop_unroll]: 2.844e-05
        [a_1]: 0.00044921
        [with_stream_mark]: 1.037e-05
        [recompute_prepare]: 7.08998e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.02998e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 1.19e-06
        [get_grad_eliminate_]: 5.17e-06
        [merge_forward]: 2.29999e-06
        [cell_reuse_recompute_pass]: 8.70001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.028e-05
        [j_node_and_user_rematch]: 8.44002e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 7.41999e-06
        [inline_without_move]: 5.28002e-06
        [renormalize]: 0.00041887
        [add_forward_monad_depend]: 3.88001e-06
        [auto_monad_grad]: 1.54e-06
        [auto_monad_eliminator]: 9.19e-06
        [cse]: 1.422e-05
        [replace_applicator]: 1.043e-05
    [Cycle 2]: 0.00030671, [27]
        [switch_simplify]: 5.63002e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 8.657e-05
        [with_stream_mark]: 7.57998e-06
        [recompute_prepare]: 4.81002e-06
        [updatestate_depend_eliminate]: 2.50002e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.09998e-06
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 4.65999e-06
        [accelerated_algorithm]: 4.73001e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 4.40999e-06
        [merge_forward]: 2.26e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.15e-05
        [j_node_and_user_rematch]: 7.55e-06
        [meta_fg_expand]: 1.32999e-06
        [replace_old_param]: 6.63998e-06
        [inline_without_move]: 9.32999e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 4.68999e-06
        [cse]: 9.47999e-06
        [replace_applicator]: 4.83001e-06
[py_interpret_to_execute_after_opt_a]: 7.83999e-06
[rewriter_after_opt_a]: 2.467e-05
[convert_after_rewriter]: 5.37001e-06
[order_py_execute_after_rewriter]: 4.41002e-06
[mutable_eliminate]: 0.00046201
[jit_opt_b]: 4.51e-05, [1]
    [Cycle 1]: 3.893e-05, [2]
        [frontend_op_eliminate]: 1.479e-05
        [inline_after_opt_a]: 1.351e-05
[cconv]: 1.526e-05
[loop_unroll]: 0.0004273
[jit_opt_after_cconv]: 0.00013093, [1]
    [Cycle 1]: 0.00012487, [11]
        [c_1]: 1.944e-05
        [parameter_eliminate]: 2.81999e-06
        [updatestate_depend_eliminate]: 5.69e-06
        [updatestate_assign_eliminate]: 2.28998e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.587e-05
        [call_graph_tuple_transform]: 1.568e-05
        [tuple_list_get_item_eliminator]: 5.30999e-06
        [none_parameter_eliminate]: 6.90023e-07
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.30999e-06
[remove_dup_value]: 6.97002e-06
[partial_unused_args_eliminate]: 1.83002e-06
[environ_conv]: 3.56001e-06
[add_recomputation]: 3.297e-05
[cse_after_recomputation]: 1.998e-05, [1]
    [Cycle 1]: 1.481e-05, [1]
        [cse]: 8.77e-06
[auto_monad_reorder]: 9.57999e-06
[get_jit_bprop_graph]: 1.44e-06
[rewriter_after_jit_bprop_graph]: 4.58999e-06
[opt_after_jit_grad]: 0.0004569
[symbol_engine_optimizer]: 6.89e-05, [1]
    [Cycle 1]: 6.276e-05, [6]
        [build]: 2.59001e-06
        [elim_shapecalc]: 7.48999e-06
        [elim_not_effective]: 1.17e-05
        [opt_reshape]: 5.19998e-06
        [fold_const_symbol]: 8.22e-06
        [renormalize]: 4.50003e-07
[validate]: 2.469e-05
[backend_pass]: 6.50005e-07
[task_emit]: 0.00476006
[execute]: 4.54002e-06
Sums
    bootstrap                            :     0.000253s :  2.11%
    type_inference                       :     0.003902s : 32.61%
    event_method                         :     0.000016s :  0.13%
    auto_monad                           :     0.000040s :  0.33%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.22%
    rewriter_before_opt_a                :     0.000060s :  0.51%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000047s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.28%
    jit_opt_a.a_1                        :     0.000536s :  4.48%
    jit_opt_a.with_stream_mark           :     0.000018s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000014s :  0.12%
    jit_opt_a.inline_without_move        :     0.000015s :  0.12%
    jit_opt_a.renormalize                :     0.000419s :  3.50%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.12%
    jit_opt_a.cse                        :     0.000024s :  0.20%
    jit_opt_a.replace_applicator         :     0.000015s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.07%
    rewriter_after_opt_a                 :     0.000025s :  0.21%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000462s :  3.86%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000015s :  0.13%
    loop_unroll                          :     0.000427s :  3.57%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000007s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000033s :  0.28%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000010s :  0.08%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000457s :  3.82%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000025s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.004760s : 39.77%
    execute                              :     0.000005s :  0.04%


Time group info:
------[substitution.]   0.000139    23
 1.03% :     0.000001s :      2: substitution.elim_not_effective
 0.90% :     0.000001s :      2: substitution.fold_const_symbol
 2.51% :     0.000003s :      3: substitution.graph_param_transform
83.65% :     0.000117s :      4: substitution.inline
 1.88% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.65% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.99% :     0.000003s :      2: substitution.replace_old_param
 5.41% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003855     2
82.26% :     0.003171s :      1: type_inference.infer
17.74% :     0.000684s :      1: type_inference.specialize
------[replace.]   0.000056     6
66.68% :     0.000038s :      4: replace.inline
33.32% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000121     6
94.69% :     0.000114s :      4: match.inline
 5.31% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000107   691
 1.60% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000001s :     10: predicate.addn_zero_filter
 1.82% :     0.000002s :     10: predicate.arithmetic_simplify
 1.38% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000001s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.57% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.52% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.00% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.15% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.74% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000000s :      3: predicate.opt_reshape
 2.16% :     0.000002s :     15: predicate.partial_eliminate
 1.34% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.80% :     0.000002s :     10: predicate.reduce_eliminate
 1.62% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000001s :     10: predicate.reshape_eliminate
 1.33% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 1.14% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.51% :     0.000008s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.31% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.24% :     0.000003s :     18: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.77% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000466    12
48.07% :     0.000224s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.49% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.43% :     0.000216s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016669    76
 0.22% :     0.000036s :      1: add_recomputation
 0.26% :     0.000043s :      1: auto_monad
 0.07% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.59% :     0.000265s :      1: bootstrap
 0.11% :     0.000018s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.13% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.12% :     0.000019s :      1: event_method
 0.05% :     0.000008s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
26.62% :     0.004438s :      1: jit_opt_a
 0.80% :     0.000134s :      1: jit_opt_after_cconv
 0.29% :     0.000048s :      1: jit_opt_b
 2.61% :     0.000435s :      1: loop_unroll
 2.82% :     0.000470s :      1: mutable_eliminate
 4.36% :     0.000727s :     26: opt.transform.jit_opt_a
 0.25% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000023s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.79% :     0.000464s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000029s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000009s :      1: remove_dup_value
 1.20% :     0.000200s :      1: renormalize.infer
 1.28% :     0.000213s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000027s :      1: rewriter_after_opt_a
 0.38% :     0.000063s :      1: rewriter_before_opt_a
 0.43% :     0.000071s :      1: symbol_engine_optimizer
28.61% :     0.004769s :      1: task_emit
23.47% :     0.003912s :      1: type_inference
 0.23% :     0.000039s :      1: validate


TotalTime = 0.0176905, [33]
[bootstrap]: 0.00026213
[type_inference]: 0.00414148
[event_method]: 1.753e-05
[auto_monad]: 4.783e-05
[graph_reusing]: 4.03999e-06
[pre_auto_parallel]: 1.82001e-06
[py_interpret_to_execute]: 3.163e-05
[rewriter_before_opt_a]: 6.602e-05
[expand_dump_flag]: 2.38002e-06
[jit_opt_a]: 0.00486642, [2]
    [Cycle 1]: 0.00134972, [27]
        [switch_simplify]: 4.659e-05
        [loop_unroll]: 3.075e-05
        [a_1]: 0.0004789
        [with_stream_mark]: 1.331e-05
        [recompute_prepare]: 7.31001e-06
        [updatestate_depend_eliminate]: 3.46001e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 1.35999e-06
        [specialize_transform]: 5.96e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 5.87999e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 5.02999e-06
        [merge_forward]: 2.65002e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.155e-05
        [j_node_and_user_rematch]: 9.42001e-06
        [meta_fg_expand]: 2.07001e-06
        [replace_old_param]: 7.82e-06
        [inline_without_move]: 5.20999e-06
        [renormalize]: 0.00049197
        [add_forward_monad_depend]: 4.77998e-06
        [auto_monad_grad]: 1.23002e-06
        [auto_monad_eliminator]: 1.011e-05
        [cse]: 1.671e-05
        [replace_applicator]: 1.14e-05
    [Cycle 2]: 0.00038987, [27]
        [switch_simplify]: 5.75001e-06
        [loop_unroll]: 5.30001e-06
        [a_1]: 0.00014805
        [with_stream_mark]: 9.40001e-06
        [recompute_prepare]: 5.19e-06
        [updatestate_depend_eliminate]: 2.61e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 1.207e-05
        [meta_shard_fg_expand]: 1.17999e-06
        [get_grad_eliminate_]: 4.70001e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.53002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.255e-05
        [j_node_and_user_rematch]: 7.83001e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 9.46e-06
        [inline_without_move]: 4.82998e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 7.80012e-07
        [auto_monad_eliminator]: 5.34998e-06
        [cse]: 1.146e-05
        [replace_applicator]: 5.86998e-06
[py_interpret_to_execute_after_opt_a]: 8.60999e-06
[rewriter_after_opt_a]: 2.621e-05
[convert_after_rewriter]: 5.94e-06
[order_py_execute_after_rewriter]: 4.13999e-06
[mutable_eliminate]: 0.00048957
[jit_opt_b]: 4.805e-05, [1]
    [Cycle 1]: 4.17e-05, [2]
        [frontend_op_eliminate]: 1.593e-05
        [inline_after_opt_a]: 1.453e-05
[cconv]: 1.619e-05
[loop_unroll]: 0.0004265
[jit_opt_after_cconv]: 0.00013722, [1]
    [Cycle 1]: 0.00013103, [11]
        [c_1]: 2.112e-05
        [parameter_eliminate]: 2.81999e-06
        [updatestate_depend_eliminate]: 5.41998e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.762e-05
        [call_graph_tuple_transform]: 1.649e-05
        [tuple_list_get_item_eliminator]: 5.39e-06
        [none_parameter_eliminate]: 1.07e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 5.52001e-06
[remove_dup_value]: 7.37002e-06
[partial_unused_args_eliminate]: 1.71e-06
[environ_conv]: 3.45e-06
[add_recomputation]: 3.265e-05
[cse_after_recomputation]: 1.969e-05, [1]
    [Cycle 1]: 1.477e-05, [1]
        [cse]: 9.04e-06
[auto_monad_reorder]: 9.24e-06
[get_jit_bprop_graph]: 1.74e-06
[rewriter_after_jit_bprop_graph]: 5.22e-06
[opt_after_jit_grad]: 0.00044999
[symbol_engine_optimizer]: 6.989e-05, [1]
    [Cycle 1]: 6.367e-05, [6]
        [build]: 2.99001e-06
        [elim_shapecalc]: 7.06001e-06
        [elim_not_effective]: 1.221e-05
        [opt_reshape]: 5.71998e-06
        [fold_const_symbol]: 8.11002e-06
        [renormalize]: 5.50004e-07
[validate]: 2.814e-05
[backend_pass]: 1.17999e-06
[task_emit]: 0.00626549
[execute]: 7.93001e-06
Sums
    bootstrap                            :     0.000262s :  1.87%
    type_inference                       :     0.004141s : 29.62%
    event_method                         :     0.000018s :  0.13%
    auto_monad                           :     0.000048s :  0.34%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000032s :  0.23%
    rewriter_before_opt_a                :     0.000066s :  0.47%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000052s :  0.37%
    jit_opt_a.loop_unroll                :     0.000036s :  0.26%
    jit_opt_a.a_1                        :     0.000627s :  4.48%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000492s :  3.52%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000028s :  0.20%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000026s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000490s :  3.50%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000016s :  0.12%
    loop_unroll                          :     0.000427s :  3.05%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000007s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000003s :  0.02%
    add_recomputation                    :     0.000033s :  0.23%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000009s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000450s :  3.22%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000028s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006265s : 44.81%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000155    23
 1.06% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.08% :     0.000003s :      3: substitution.graph_param_transform
82.43% :     0.000128s :      4: substitution.inline
 2.26% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.78% :     0.000004s :      4: substitution.remove_not_recompute_node
 3.57% :     0.000006s :      2: substitution.replace_old_param
 5.18% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004088     2
81.24% :     0.003321s :      1: type_inference.infer
18.76% :     0.000767s :      1: type_inference.specialize
------[replace.]   0.000062     6
65.78% :     0.000041s :      4: replace.inline
34.22% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000132     6
94.85% :     0.000125s :      4: match.inline
 5.15% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.52% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.31% :     0.000002s :     10: predicate.addn_check_dump
 1.68% :     0.000002s :     10: predicate.addn_zero_filter
 2.44% :     0.000003s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.18% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.23% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.60% :     0.000002s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.32% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.30% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.49% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000002s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.97% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.65% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.56% :     0.000002s :     10: predicate.transpose_eliminate
 1.58% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.41% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.82% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000505    12
45.51% :     0.000230s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.25% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.24% :     0.000249s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019121    76
 0.19% :     0.000035s :      1: add_recomputation
 0.27% :     0.000051s :      1: auto_monad
 0.06% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.43% :     0.000274s :      1: bootstrap
 0.10% :     0.000019s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
25.47% :     0.004870s :      1: jit_opt_a
 0.73% :     0.000140s :      1: jit_opt_after_cconv
 0.26% :     0.000051s :      1: jit_opt_b
 2.27% :     0.000435s :      1: loop_unroll
 2.60% :     0.000498s :      1: mutable_eliminate
 4.38% :     0.000837s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.39% :     0.000458s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000010s :      1: remove_dup_value
 1.26% :     0.000242s :      1: renormalize.infer
 1.28% :     0.000244s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000029s :      1: rewriter_after_opt_a
 0.36% :     0.000069s :      1: rewriter_before_opt_a
 0.38% :     0.000072s :      1: symbol_engine_optimizer
32.83% :     0.006278s :      1: task_emit
21.72% :     0.004154s :      1: type_inference
 0.23% :     0.000044s :      1: validate


TotalTime = 0.016867, [33]
[bootstrap]: 0.00027666
[type_inference]: 0.00397328
[event_method]: 1.49e-05
[auto_monad]: 5.008e-05
[graph_reusing]: 4.43001e-06
[pre_auto_parallel]: 1.99e-06
[py_interpret_to_execute]: 2.746e-05
[rewriter_before_opt_a]: 6.37e-05
[expand_dump_flag]: 2.44999e-06
[jit_opt_a]: 0.00484942, [2]
    [Cycle 1]: 0.00144391, [27]
        [switch_simplify]: 4.574e-05
        [loop_unroll]: 2.862e-05
        [a_1]: 0.00047513
        [with_stream_mark]: 1.33e-05
        [recompute_prepare]: 6.99001e-06
        [updatestate_depend_eliminate]: 2.88e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 1.27999e-06
        [specialize_transform]: 6.16e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 5.55001e-06
        [meta_shard_fg_expand]: 1.47999e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 2.94999e-06
        [cell_reuse_recompute_pass]: 1.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.236e-05
        [j_node_and_user_rematch]: 8.99e-06
        [meta_fg_expand]: 2.14999e-06
        [replace_old_param]: 8.28999e-06
        [inline_without_move]: 5.82999e-06
        [renormalize]: 0.00050876
        [add_forward_monad_depend]: 4.68999e-06
        [auto_monad_grad]: 1.53002e-06
        [auto_monad_eliminator]: 1.024e-05
        [cse]: 2.097e-05
        [replace_applicator]: 1.091e-05
    [Cycle 2]: 0.0003178, [27]
        [switch_simplify]: 5.34e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 8.713e-05
        [with_stream_mark]: 8.88002e-06
        [recompute_prepare]: 5.30001e-06
        [updatestate_depend_eliminate]: 2.31e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 9.60019e-07
        [specialize_transform]: 4.72998e-06
        [updatestate_useless_node_eliminater]: 4.53999e-06
        [accelerated_algorithm]: 5.17e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 4.63999e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.283e-05
        [j_node_and_user_rematch]: 7.83001e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 7.60998e-06
        [inline_without_move]: 4.85001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.45001e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 5.61e-06
        [cse]: 9.85002e-06
        [replace_applicator]: 5.39e-06
[py_interpret_to_execute_after_opt_a]: 9.10999e-06
[rewriter_after_opt_a]: 2.99e-05
[convert_after_rewriter]: 6.59001e-06
[order_py_execute_after_rewriter]: 5.61e-06
[mutable_eliminate]: 0.00053715
[jit_opt_b]: 4.999e-05, [1]
    [Cycle 1]: 4.298e-05, [2]
        [frontend_op_eliminate]: 1.578e-05
        [inline_after_opt_a]: 1.523e-05
[cconv]: 2.064e-05
[loop_unroll]: 0.00043257
[jit_opt_after_cconv]: 0.00013927, [1]
    [Cycle 1]: 0.00013275, [11]
        [c_1]: 1.991e-05
        [parameter_eliminate]: 2.72001e-06
        [updatestate_depend_eliminate]: 5.71e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [cse]: 1.888e-05
        [call_graph_tuple_transform]: 1.885e-05
        [tuple_list_get_item_eliminator]: 5.11002e-06
        [none_parameter_eliminate]: 8.70001e-07
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.02e-06
[remove_dup_value]: 1.124e-05
[partial_unused_args_eliminate]: 2.06e-06
[environ_conv]: 4.58999e-06
[add_recomputation]: 3.951e-05
[cse_after_recomputation]: 1.995e-05, [1]
    [Cycle 1]: 1.494e-05, [1]
        [cse]: 9.19e-06
[auto_monad_reorder]: 1.171e-05
[get_jit_bprop_graph]: 2.03002e-06
[rewriter_after_jit_bprop_graph]: 4.91002e-06
[opt_after_jit_grad]: 0.00045116
[symbol_engine_optimizer]: 6.951e-05, [1]
    [Cycle 1]: 6.33e-05, [6]
        [build]: 2.89001e-06
        [elim_shapecalc]: 7.16999e-06
        [elim_not_effective]: 1.181e-05
        [opt_reshape]: 5.34e-06
        [fold_const_symbol]: 8.73001e-06
        [renormalize]: 4.19997e-07
[validate]: 2.829e-05
[backend_pass]: 7.2e-07
[task_emit]: 0.00553589
[execute]: 6.89001e-06
Sums
    bootstrap                            :     0.000277s :  2.11%
    type_inference                       :     0.003973s : 30.29%
    event_method                         :     0.000015s :  0.11%
    auto_monad                           :     0.000050s :  0.38%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000027s :  0.21%
    rewriter_before_opt_a                :     0.000064s :  0.49%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000051s :  0.39%
    jit_opt_a.loop_unroll                :     0.000033s :  0.26%
    jit_opt_a.a_1                        :     0.000562s :  4.29%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.000509s :  3.88%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.12%
    jit_opt_a.cse                        :     0.000031s :  0.23%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000030s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000537s :  4.10%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000021s :  0.16%
    loop_unroll                          :     0.000433s :  3.30%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000011s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000040s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000012s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000451s :  3.44%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005536s : 42.21%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000164    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.99% :     0.000002s :      2: substitution.fold_const_symbol
 2.86% :     0.000005s :      3: substitution.graph_param_transform
82.94% :     0.000136s :      4: substitution.inline
 1.58% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.70% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000003s :      2: substitution.replace_old_param
 5.84% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003923     2
81.59% :     0.003200s :      1: type_inference.infer
18.41% :     0.000722s :      1: type_inference.specialize
------[replace.]   0.000058     6
66.68% :     0.000039s :      4: replace.inline
33.32% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000142     6
94.06% :     0.000134s :      4: match.inline
 5.94% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000108   691
 1.69% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.52% :     0.000002s :     10: predicate.addn_zero_filter
 1.83% :     0.000002s :     10: predicate.arithmetic_simplify
 1.65% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000001s :     10: predicate.depend_value_elim
 1.31% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.22% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.14% :     0.000001s :      6: predicate.get_grad_eliminate
 0.43% :     0.000000s :      3: predicate.graph_param_transform
 5.39% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.44% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000001s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.63% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.27% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000000s :      3: predicate.opt_reshape
 2.17% :     0.000002s :     15: predicate.partial_eliminate
 1.24% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.39% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.32% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.24% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.75% :     0.000008s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.43% :     0.000002s :     10: predicate.transpose_eliminate
 1.32% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.55% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.66% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000487    12
47.87% :     0.000233s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.18% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.95% :     0.000224s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018241    76
 0.23% :     0.000042s :      1: add_recomputation
 0.29% :     0.000053s :      1: auto_monad
 0.08% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.60% :     0.000291s :      1: bootstrap
 0.13% :     0.000023s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000018s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.60% :     0.004852s :      1: jit_opt_a
 0.78% :     0.000143s :      1: jit_opt_after_cconv
 0.29% :     0.000053s :      1: jit_opt_b
 2.41% :     0.000440s :      1: loop_unroll
 2.99% :     0.000545s :      1: mutable_eliminate
 4.16% :     0.000760s :     26: opt.transform.jit_opt_a
 0.25% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.52% :     0.000459s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 1.42% :     0.000259s :      1: renormalize.infer
 1.33% :     0.000243s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000035s :      1: rewriter_after_opt_a
 0.36% :     0.000066s :      1: rewriter_before_opt_a
 0.39% :     0.000072s :      1: symbol_engine_optimizer
30.43% :     0.005551s :      1: task_emit
21.84% :     0.003985s :      1: type_inference
 0.24% :     0.000045s :      1: validate


TotalTime = 0.0169976, [33]
[bootstrap]: 0.00025663
[type_inference]: 0.00404339
[event_method]: 1.515e-05
[auto_monad]: 4.123e-05
[graph_reusing]: 3.41001e-06
[pre_auto_parallel]: 2.02001e-06
[py_interpret_to_execute]: 2.691e-05
[rewriter_before_opt_a]: 6.306e-05
[expand_dump_flag]: 2.21998e-06
[jit_opt_a]: 0.00484142, [2]
    [Cycle 1]: 0.00134898, [27]
        [switch_simplify]: 4.421e-05
        [loop_unroll]: 2.869e-05
        [a_1]: 0.00046431
        [with_stream_mark]: 1.259e-05
        [recompute_prepare]: 7.18998e-06
        [updatestate_depend_eliminate]: 2.86e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.95001e-06
        [parameter_eliminate]: 1.30001e-06
        [specialize_transform]: 5.75001e-06
        [updatestate_useless_node_eliminater]: 4.83001e-06
        [accelerated_algorithm]: 5.65001e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.85001e-06
        [merge_forward]: 2.39001e-06
        [cell_reuse_recompute_pass]: 9.30013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.119e-05
        [j_node_and_user_rematch]: 8.70001e-06
        [meta_fg_expand]: 1.84998e-06
        [replace_old_param]: 8.05e-06
        [inline_without_move]: 5.06002e-06
        [renormalize]: 0.00050764
        [add_forward_monad_depend]: 4.88001e-06
        [auto_monad_grad]: 1.14998e-06
        [auto_monad_eliminator]: 1.014e-05
        [cse]: 1.837e-05
        [replace_applicator]: 1.187e-05
    [Cycle 2]: 0.00032913, [27]
        [switch_simplify]: 5.37001e-06
        [loop_unroll]: 4.70999e-06
        [a_1]: 8.633e-05
        [with_stream_mark]: 9.95002e-06
        [recompute_prepare]: 5.44e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.19998e-06
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 5.10001e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.63001e-06
        [merge_forward]: 2.42001e-06
        [cell_reuse_recompute_pass]: 1.96e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.275e-05
        [j_node_and_user_rematch]: 7.58999e-06
        [meta_fg_expand]: 1.40001e-06
        [replace_old_param]: 7.22002e-06
        [inline_without_move]: 4.58001e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 8.60018e-07
        [auto_monad_eliminator]: 6.06e-06
        [cse]: 1.134e-05
        [replace_applicator]: 1.176e-05
[py_interpret_to_execute_after_opt_a]: 8.55001e-06
[rewriter_after_opt_a]: 2.734e-05
[convert_after_rewriter]: 7.9e-06
[order_py_execute_after_rewriter]: 4.45999e-06
[mutable_eliminate]: 0.00051407
[jit_opt_b]: 4.685e-05, [1]
    [Cycle 1]: 3.972e-05, [2]
        [frontend_op_eliminate]: 1.514e-05
        [inline_after_opt_a]: 1.374e-05
[cconv]: 1.785e-05
[loop_unroll]: 0.00042266
[jit_opt_after_cconv]: 0.00013358, [1]
    [Cycle 1]: 0.00012675, [11]
        [c_1]: 1.989e-05
        [parameter_eliminate]: 2.67001e-06
        [updatestate_depend_eliminate]: 6.07001e-06
        [updatestate_assign_eliminate]: 2.31998e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.713e-05
        [call_graph_tuple_transform]: 1.652e-05
        [tuple_list_get_item_eliminator]: 5.02999e-06
        [none_parameter_eliminate]: 9.10019e-07
        [renormalize]: 4.70027e-07
        [switch_simplify]: 5.00999e-06
[remove_dup_value]: 9.09e-06
[partial_unused_args_eliminate]: 2.04999e-06
[environ_conv]: 3.69002e-06
[add_recomputation]: 3.335e-05
[cse_after_recomputation]: 1.944e-05, [1]
    [Cycle 1]: 1.442e-05, [1]
        [cse]: 8.77999e-06
[auto_monad_reorder]: 9.82999e-06
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 5.14e-06
[opt_after_jit_grad]: 0.00045391
[symbol_engine_optimizer]: 7.84e-05, [1]
    [Cycle 1]: 7.135e-05, [6]
        [build]: 2.89001e-06
        [elim_shapecalc]: 8.97e-06
        [elim_not_effective]: 1.309e-05
        [opt_reshape]: 5.74e-06
        [fold_const_symbol]: 9.20001e-06
        [renormalize]: 4.19997e-07
[validate]: 2.916e-05
[backend_pass]: 7.99977e-07
[task_emit]: 0.00564421
[execute]: 7.03e-06
Sums
    bootstrap                            :     0.000257s :  1.94%
    type_inference                       :     0.004043s : 30.63%
    event_method                         :     0.000015s :  0.11%
    auto_monad                           :     0.000041s :  0.31%
    graph_reusing                        :     0.000003s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000027s :  0.20%
    rewriter_before_opt_a                :     0.000063s :  0.48%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000050s :  0.38%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000551s :  4.17%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000508s :  3.85%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.12%
    jit_opt_a.cse                        :     0.000030s :  0.23%
    jit_opt_a.replace_applicator         :     0.000024s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000027s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000514s :  3.89%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000018s :  0.14%
    loop_unroll                          :     0.000423s :  3.20%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000009s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000033s :  0.25%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000010s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000454s :  3.44%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005644s : 42.75%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000152    23
 1.23% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.53% :     0.000004s :      3: substitution.graph_param_transform
83.01% :     0.000127s :      4: substitution.inline
 1.89% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.56% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.54% :     0.000004s :      2: substitution.replace_old_param
 5.61% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003994     2
82.00% :     0.003275s :      1: type_inference.infer
18.00% :     0.000719s :      1: type_inference.specialize
------[replace.]   0.000058     6
66.73% :     0.000039s :      4: replace.inline
33.27% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000132     6
94.29% :     0.000124s :      4: match.inline
 5.71% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.32% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.81% :     0.000002s :     10: predicate.addn_zero_filter
 2.03% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000001s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.46% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.76% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.30% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.34% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.26% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.07% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.66% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000002s :     15: predicate.partial_eliminate
 1.24% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.84% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.53% :     0.000002s :     10: predicate.reshape_eliminate
 1.49% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.52% :     0.000002s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.31% :     0.000008s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.33% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.77% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000484    12
47.81% :     0.000231s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.42% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.77% :     0.000226s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018322    76
 0.20% :     0.000036s :      1: add_recomputation
 0.24% :     0.000045s :      1: auto_monad
 0.07% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.49% :     0.000273s :      1: bootstrap
 0.11% :     0.000021s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
26.44% :     0.004844s :      1: jit_opt_a
 0.74% :     0.000136s :      1: jit_opt_after_cconv
 0.27% :     0.000049s :      1: jit_opt_b
 2.35% :     0.000431s :      1: loop_unroll
 2.85% :     0.000523s :      1: mutable_eliminate
 4.10% :     0.000751s :     26: opt.transform.jit_opt_a
 0.24% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.52% :     0.000462s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000012s :      1: remove_dup_value
 1.36% :     0.000248s :      1: renormalize.infer
 1.38% :     0.000253s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000030s :      1: rewriter_after_opt_a
 0.36% :     0.000066s :      1: rewriter_before_opt_a
 0.44% :     0.000081s :      1: symbol_engine_optimizer
30.87% :     0.005655s :      1: task_emit
22.13% :     0.004055s :      1: type_inference
 0.25% :     0.000047s :      1: validate


TotalTime = 0.0195024, [33]
[bootstrap]: 0.00027086
[type_inference]: 0.00433951
[event_method]: 1.592e-05
[auto_monad]: 4.242e-05
[graph_reusing]: 3.42002e-06
[pre_auto_parallel]: 1.87001e-06
[py_interpret_to_execute]: 3.039e-05
[rewriter_before_opt_a]: 6.376e-05
[expand_dump_flag]: 2.37001e-06
[jit_opt_a]: 0.00491096, [2]
    [Cycle 1]: 0.00138588, [27]
        [switch_simplify]: 4.659e-05
        [loop_unroll]: 3.034e-05
        [a_1]: 0.00048798
        [with_stream_mark]: 1.354e-05
        [recompute_prepare]: 6.98998e-06
        [updatestate_depend_eliminate]: 2.87002e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.55999e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 1.25001e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 2.63e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.143e-05
        [j_node_and_user_rematch]: 9.40001e-06
        [meta_fg_expand]: 2.05002e-06
        [replace_old_param]: 7.78999e-06
        [inline_without_move]: 5.95002e-06
        [renormalize]: 0.00052191
        [add_forward_monad_depend]: 4.43999e-06
        [auto_monad_grad]: 1.25999e-06
        [auto_monad_eliminator]: 1.016e-05
        [cse]: 1.835e-05
        [replace_applicator]: 1.084e-05
    [Cycle 2]: 0.00031336, [27]
        [switch_simplify]: 5.67999e-06
        [loop_unroll]: 4.94998e-06
        [a_1]: 9.184e-05
        [with_stream_mark]: 8.88002e-06
        [recompute_prepare]: 4.87998e-06
        [updatestate_depend_eliminate]: 2.41e-06
        [updatestate_assign_eliminate]: 2.19999e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 4.92999e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.13002e-06
        [meta_shard_fg_expand]: 1.22999e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 1.99e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.161e-05
        [j_node_and_user_rematch]: 7.67998e-06
        [meta_fg_expand]: 1.55001e-06
        [replace_old_param]: 7.38e-06
        [inline_without_move]: 4.66002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 4.79e-06
        [cse]: 9.51e-06
        [replace_applicator]: 4.84998e-06
[py_interpret_to_execute_after_opt_a]: 8.18001e-06
[rewriter_after_opt_a]: 2.565e-05
[convert_after_rewriter]: 5.52001e-06
[order_py_execute_after_rewriter]: 4.10998e-06
[mutable_eliminate]: 0.00052231
[jit_opt_b]: 4.888e-05, [1]
    [Cycle 1]: 4.23e-05, [2]
        [frontend_op_eliminate]: 1.619e-05
        [inline_after_opt_a]: 1.47e-05
[cconv]: 1.515e-05
[loop_unroll]: 0.00046062
[jit_opt_after_cconv]: 0.00013729, [1]
    [Cycle 1]: 0.00013118, [11]
        [c_1]: 2.257e-05
        [parameter_eliminate]: 2.33998e-06
        [updatestate_depend_eliminate]: 6.01e-06
        [updatestate_assign_eliminate]: 2.39999e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [cse]: 1.762e-05
        [call_graph_tuple_transform]: 1.648e-05
        [tuple_list_get_item_eliminator]: 5.37999e-06
        [none_parameter_eliminate]: 8.39995e-07
        [renormalize]: 5.60016e-07
        [switch_simplify]: 5.48002e-06
[remove_dup_value]: 7.01999e-06
[partial_unused_args_eliminate]: 1.49998e-06
[environ_conv]: 3.48e-06
[add_recomputation]: 3.229e-05
[cse_after_recomputation]: 2.016e-05, [1]
    [Cycle 1]: 1.513e-05, [1]
        [cse]: 9.28002e-06
[auto_monad_reorder]: 9.07001e-06
[get_jit_bprop_graph]: 1.35001e-06
[rewriter_after_jit_bprop_graph]: 5.39e-06
[opt_after_jit_grad]: 0.00047162
[symbol_engine_optimizer]: 7.189e-05, [1]
    [Cycle 1]: 6.605e-05, [6]
        [build]: 2.61999e-06
        [elim_shapecalc]: 7.44002e-06
        [elim_not_effective]: 1.26e-05
        [opt_reshape]: 5.85002e-06
        [fold_const_symbol]: 8.48001e-06
        [renormalize]: 5.10016e-07
[validate]: 2.737e-05
[backend_pass]: 9.79984e-07
[task_emit]: 0.007749
[execute]: 5.50001e-06
Sums
    bootstrap                            :     0.000271s :  1.72%
    type_inference                       :     0.004340s : 27.62%
    event_method                         :     0.000016s :  0.10%
    auto_monad                           :     0.000042s :  0.27%
    graph_reusing                        :     0.000003s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000030s :  0.19%
    rewriter_before_opt_a                :     0.000064s :  0.41%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000052s :  0.33%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000580s :  3.69%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000522s :  3.32%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.10%
    jit_opt_a.cse                        :     0.000028s :  0.18%
    jit_opt_a.replace_applicator         :     0.000016s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000026s :  0.16%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000522s :  3.32%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000015s :  0.10%
    loop_unroll                          :     0.000461s :  2.93%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000007s :  0.04%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000003s :  0.02%
    add_recomputation                    :     0.000032s :  0.21%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000009s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000472s :  3.00%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000027s :  0.17%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007749s : 49.32%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000157    23
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.22% :     0.000003s :      3: substitution.graph_param_transform
84.60% :     0.000133s :      4: substitution.inline
 1.94% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.50% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.97% :     0.000003s :      2: substitution.replace_old_param
 5.21% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004286     2
79.07% :     0.003389s :      1: type_inference.infer
20.93% :     0.000897s :      1: type_inference.specialize
------[replace.]   0.000065     6
65.28% :     0.000042s :      4: replace.inline
34.72% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000137     6
94.91% :     0.000130s :      4: match.inline
 5.09% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000001s :     10: predicate.addn_zero_filter
 2.58% :     0.000003s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.25% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.24% :     0.000006s :     22: predicate.inline
 1.04% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.52% :     0.000002s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.35% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.83% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.17% :     0.000001s :     10: predicate.minmaximum_grad
 1.89% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.58% :     0.000009s :     53: predicate.switch_simplify
 1.32% :     0.000002s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.43% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.26% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.00% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000519    12
46.94% :     0.000243s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.54% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.51% :     0.000246s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020903    76
 0.17% :     0.000035s :      1: add_recomputation
 0.22% :     0.000046s :      1: auto_monad
 0.06% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.36% :     0.000284s :      1: bootstrap
 0.08% :     0.000018s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.09% :     0.000019s :      1: event_method
 0.04% :     0.000009s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
23.51% :     0.004914s :      1: jit_opt_a
 0.67% :     0.000141s :      1: jit_opt_after_cconv
 0.25% :     0.000052s :      1: jit_opt_b
 2.24% :     0.000469s :      1: loop_unroll
 2.54% :     0.000531s :      1: mutable_eliminate
 3.72% :     0.000777s :     26: opt.transform.jit_opt_a
 0.22% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.30% :     0.000481s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000033s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000009s :      1: remove_dup_value
 1.24% :     0.000260s :      1: renormalize.infer
 1.22% :     0.000256s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000029s :      1: rewriter_after_opt_a
 0.32% :     0.000066s :      1: rewriter_before_opt_a
 0.36% :     0.000075s :      1: symbol_engine_optimizer
37.14% :     0.007764s :      1: task_emit
20.81% :     0.004351s :      1: type_inference
 0.21% :     0.000044s :      1: validate


TotalTime = 0.0866435, [33]
[bootstrap]: 0.00038812
[type_inference]: 0.0687971
[event_method]: 1.771e-05
[auto_monad]: 8.659e-05
[graph_reusing]: 5.60001e-06
[pre_auto_parallel]: 2.58e-06
[py_interpret_to_execute]: 0.00026728
[rewriter_before_opt_a]: 8.252e-05
[expand_dump_flag]: 3.45e-06
[jit_opt_a]: 0.00791353, [2]
    [Cycle 1]: 0.00317855, [27]
        [switch_simplify]: 5.278e-05
        [loop_unroll]: 3.333e-05
        [a_1]: 0.00062151
        [with_stream_mark]: 1.881e-05
        [recompute_prepare]: 8.43999e-06
        [updatestate_depend_eliminate]: 3.51999e-06
        [updatestate_assign_eliminate]: 4.32e-06
        [updatestate_loads_eliminate]: 3.07002e-06
        [parameter_eliminate]: 1.60001e-06
        [specialize_transform]: 7.08e-06
        [updatestate_useless_node_eliminater]: 5.92999e-06
        [accelerated_algorithm]: 6.64001e-06
        [meta_shard_fg_expand]: 2.20002e-06
        [get_grad_eliminate_]: 6.58003e-06
        [merge_forward]: 3.58e-06
        [cell_reuse_recompute_pass]: 1.57999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.478e-05
        [j_node_and_user_rematch]: 1.005e-05
        [meta_fg_expand]: 3.15998e-06
        [replace_old_param]: 9.99999e-06
        [inline_without_move]: 5.88002e-06
        [renormalize]: 0.0021155
        [add_forward_monad_depend]: 1.313e-05
        [auto_monad_grad]: 2.39001e-06
        [auto_monad_eliminator]: 1.541e-05
        [cse]: 2.197e-05
        [replace_applicator]: 1.497e-05
    [Cycle 2]: 0.00038364, [27]
        [switch_simplify]: 7.67998e-06
        [loop_unroll]: 6.23e-06
        [a_1]: 0.00013491
        [with_stream_mark]: 1.116e-05
        [recompute_prepare]: 6.64999e-06
        [updatestate_depend_eliminate]: 3.33998e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [parameter_eliminate]: 1.32999e-06
        [specialize_transform]: 6.33e-06
        [updatestate_useless_node_eliminater]: 6.33998e-06
        [accelerated_algorithm]: 5.62001e-06
        [meta_shard_fg_expand]: 1.33002e-06
        [get_grad_eliminate_]: 5.48002e-06
        [merge_forward]: 2.94001e-06
        [cell_reuse_recompute_pass]: 1.99999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.236e-05
        [j_node_and_user_rematch]: 9.49999e-06
        [meta_fg_expand]: 1.94999e-06
        [replace_old_param]: 8.28999e-06
        [inline_without_move]: 5.75001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 8.40024e-07
        [auto_monad_eliminator]: 6.07999e-06
        [cse]: 1.096e-05
        [replace_applicator]: 6.26e-06
[py_interpret_to_execute_after_opt_a]: 1.062e-05
[rewriter_after_opt_a]: 3.502e-05
[convert_after_rewriter]: 7.52998e-06
[order_py_execute_after_rewriter]: 5.40001e-06
[mutable_eliminate]: 0.00056751
[jit_opt_b]: 5.573e-05, [1]
    [Cycle 1]: 4.873e-05, [2]
        [frontend_op_eliminate]: 1.944e-05
        [inline_after_opt_a]: 1.749e-05
[cconv]: 2.323e-05
[loop_unroll]: 0.00044159
[jit_opt_after_cconv]: 0.00014991, [1]
    [Cycle 1]: 0.00014342, [11]
        [c_1]: 2.702e-05
        [parameter_eliminate]: 2.72001e-06
        [updatestate_depend_eliminate]: 5.70001e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.13998e-06
        [cse]: 1.91e-05
        [call_graph_tuple_transform]: 2.054e-05
        [tuple_list_get_item_eliminator]: 6.27001e-06
        [none_parameter_eliminate]: 1.74e-06
        [renormalize]: 4.2998e-07
        [switch_simplify]: 6.17001e-06
[remove_dup_value]: 1.606e-05
[partial_unused_args_eliminate]: 2.06998e-06
[environ_conv]: 9.24e-06
[add_recomputation]: 4.711e-05
[cse_after_recomputation]: 7.287e-05, [1]
    [Cycle 1]: 6.707e-05, [1]
        [cse]: 5.976e-05
[auto_monad_reorder]: 1.744e-05
[get_jit_bprop_graph]: 2.13002e-06
[rewriter_after_jit_bprop_graph]: 5.15001e-06
[opt_after_jit_grad]: 0.00058434
[symbol_engine_optimizer]: 8.532e-05, [1]
    [Cycle 1]: 7.769e-05, [6]
        [build]: 4.32e-06
        [elim_shapecalc]: 9.09998e-06
        [elim_not_effective]: 1.568e-05
        [opt_reshape]: 7.62998e-06
        [fold_const_symbol]: 1.07e-05
        [renormalize]: 6.59988e-07
[validate]: 4.457e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.00667352
[execute]: 8.75001e-06
Sums
    bootstrap                            :     0.000388s :  0.48%
    type_inference                       :     0.068797s : 84.24%
    event_method                         :     0.000018s :  0.02%
    auto_monad                           :     0.000087s :  0.11%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000267s :  0.33%
    rewriter_before_opt_a                :     0.000083s :  0.10%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000060s :  0.07%
    jit_opt_a.loop_unroll                :     0.000040s :  0.05%
    jit_opt_a.a_1                        :     0.000756s :  0.93%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.04%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.00%
    jit_opt_a.specialize_transform       :     0.000013s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.00%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.01%
    jit_opt_a.merge_forward              :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.00%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.03%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.02%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000018s :  0.02%
    jit_opt_a.inline_without_move        :     0.000012s :  0.01%
    jit_opt_a.renormalize                :     0.002116s :  2.59%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.03%
    jit_opt_a.cse                        :     0.000033s :  0.04%
    jit_opt_a.replace_applicator         :     0.000021s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.01%
    rewriter_after_opt_a                 :     0.000035s :  0.04%
    convert_after_rewriter               :     0.000008s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000568s :  0.69%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.02%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.02%
    cconv                                :     0.000023s :  0.03%
    loop_unroll                          :     0.000442s :  0.54%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.03%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000019s :  0.02%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.01%
    remove_dup_value                     :     0.000016s :  0.02%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000009s :  0.01%
    add_recomputation                    :     0.000047s :  0.06%
    cse_after_recomputation.cse          :     0.000060s :  0.07%
    auto_monad_reorder                   :     0.000017s :  0.02%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000584s :  0.72%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.02%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.05%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006674s :  8.17%
    execute                              :     0.000009s :  0.01%


Time group info:
------[substitution.]   0.000220    27
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.60% :     0.000006s :      4: substitution.graph_param_transform
79.83% :     0.000176s :      4: substitution.inline
 1.42% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.96% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.85% :     0.000004s :      2: substitution.replace_old_param
 6.50% :     0.000014s :      3: substitution.tile_eliminate
 4.40% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.068717     2
97.09% :     0.066718s :      1: type_inference.infer
 2.91% :     0.001999s :      1: type_inference.specialize
------[replace.]   0.000072     6
65.30% :     0.000047s :      4: replace.inline
34.70% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000181     6
95.34% :     0.000173s :      4: match.inline
 4.66% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   834
 1.29% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.85% :     0.000003s :      4: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     12: predicate.addn_check_dump
 1.35% :     0.000002s :     12: predicate.addn_zero_filter
 2.27% :     0.000003s :     12: predicate.arithmetic_simplify
 1.16% :     0.000002s :     12: predicate.cast_eliminate
 0.36% :     0.000000s :      4: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     12: predicate.compare_switch_simplify
 1.34% :     0.000002s :     12: predicate.depend_value_elim
 1.03% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.18% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     12: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.37% :     0.000001s :      4: predicate.elim_not_effective
 0.94% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.08% :     0.000001s :     12: predicate.environ_get_eliminate
 1.09% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      4: predicate.fold_const_symbol
 1.18% :     0.000002s :      8: predicate.get_grad_eliminate
 0.31% :     0.000000s :      4: predicate.graph_param_transform
 5.51% :     0.000008s :     26: predicate.inline
 0.92% :     0.000001s :      8: predicate.inline_without_move
 0.48% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      8: predicate.less_batch_normalization
 1.46% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     18: predicate.load_eliminater
 1.44% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.05% :     0.000006s :     38: predicate.loop_unroll_before_grad
 1.96% :     0.000003s :     16: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000002s :     12: predicate.merge_addn
 1.02% :     0.000001s :     12: predicate.minmaximum_grad
 1.35% :     0.000002s :      4: predicate.mutable_eliminate
 0.68% :     0.000001s :      4: predicate.opt_reshape
 2.29% :     0.000003s :     18: predicate.partial_eliminate
 1.40% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     12: predicate.reduce_eliminate
 1.45% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.95% :     0.000003s :     22: predicate.replace_applicator
 0.74% :     0.000001s :      8: predicate.replace_old_param
 0.35% :     0.000000s :      4: predicate.reset_defer_inline
 1.47% :     0.000002s :     12: predicate.reshape_eliminate
 1.40% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     12: predicate.same_eliminate
 0.67% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      8: predicate.special_op_eliminate
 1.01% :     0.000001s :      8: predicate.specialize_transform
 1.41% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     18: predicate.switch_defer_inline
 2.35% :     0.000003s :     18: predicate.switch_layer_defer_inline
 7.37% :     0.000010s :     60: predicate.switch_simplify
 1.75% :     0.000002s :     12: predicate.tile_eliminate
 1.31% :     0.000002s :     12: predicate.transpose_eliminate
 1.38% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     18: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000004s :     26: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     12: predicate.value_based_eliminate
 0.42% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.002226    26
70.88% :     0.001578s :     19: func_graph_cloner_run.FuncGraphClonerGraph
 1.78% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
27.34% :     0.000609s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.089878    76
 0.06% :     0.000050s :      1: add_recomputation
 0.10% :     0.000091s :      1: auto_monad
 0.02% :     0.000020s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.45% :     0.000405s :      1: bootstrap
 0.03% :     0.000026s :      1: cconv
 0.01% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000075s :      1: cse_after_recomputation
 0.01% :     0.000011s :      1: environ_conv
 0.02% :     0.000021s :      1: event_method
 0.01% :     0.000012s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000008s :      1: graph_reusing
 8.81% :     0.007917s :      1: jit_opt_a
 0.17% :     0.000153s :      1: jit_opt_after_cconv
 0.06% :     0.000058s :      1: jit_opt_b
 0.50% :     0.000450s :      1: loop_unroll
 0.64% :     0.000576s :      1: mutable_eliminate
 1.10% :     0.000990s :     26: opt.transform.jit_opt_a
 0.06% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.03% :     0.000030s :      4: opt.transform.jit_opt_b
 0.02% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.04% :     0.000039s :      4: opt.transform.symbol_engine_opt
 0.66% :     0.000595s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000273s :      1: py_interpret_to_execute
 0.01% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000018s :      1: remove_dup_value
 1.64% :     0.001472s :      1: renormalize.infer
 0.71% :     0.000634s :      1: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.04% :     0.000038s :      1: rewriter_after_opt_a
 0.10% :     0.000087s :      1: rewriter_before_opt_a
 0.10% :     0.000088s :      1: symbol_engine_optimizer
 7.44% :     0.006685s :      1: task_emit
76.56% :     0.068810s :      1: type_inference
 0.07% :     0.000067s :      1: validate


TotalTime = 0.0180873, [33]
[bootstrap]: 0.00027274
[type_inference]: 0.00418863
[event_method]: 1.641e-05
[auto_monad]: 6.353e-05
[graph_reusing]: 5.95002e-06
[pre_auto_parallel]: 2.66e-06
[py_interpret_to_execute]: 3.229e-05
[rewriter_before_opt_a]: 7.053e-05
[expand_dump_flag]: 3.16001e-06
[jit_opt_a]: 0.00532259, [2]
    [Cycle 1]: 0.00158568, [27]
        [switch_simplify]: 5.366e-05
        [loop_unroll]: 2.875e-05
        [a_1]: 0.00052797
        [with_stream_mark]: 2.27e-05
        [recompute_prepare]: 8.43999e-06
        [updatestate_depend_eliminate]: 4.08999e-06
        [updatestate_assign_eliminate]: 3.3e-06
        [updatestate_loads_eliminate]: 2.68998e-06
        [parameter_eliminate]: 1.76e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 4.81002e-06
        [accelerated_algorithm]: 5.28002e-06
        [meta_shard_fg_expand]: 2.54999e-06
        [get_grad_eliminate_]: 5.02e-06
        [merge_forward]: 3.6e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.536e-05
        [j_node_and_user_rematch]: 8.79998e-06
        [meta_fg_expand]: 2.06998e-06
        [replace_old_param]: 8.99e-06
        [inline_without_move]: 5.34998e-06
        [renormalize]: 0.00063227
        [add_forward_monad_depend]: 6.02999e-06
        [auto_monad_grad]: 2.91999e-06
        [auto_monad_eliminator]: 1.382e-05
        [cse]: 2.562e-05
        [replace_applicator]: 1.117e-05
    [Cycle 2]: 0.00031442, [27]
        [switch_simplify]: 5.40999e-06
        [loop_unroll]: 4.92999e-06
        [a_1]: 8.818e-05
        [with_stream_mark]: 9.06998e-06
        [recompute_prepare]: 4.94998e-06
        [updatestate_depend_eliminate]: 2.69001e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 4.75001e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.12999e-06
        [meta_shard_fg_expand]: 1.05001e-06
        [get_grad_eliminate_]: 4.43001e-06
        [merge_forward]: 2.78003e-06
        [cell_reuse_recompute_pass]: 1.20999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.314e-05
        [j_node_and_user_rematch]: 7.71999e-06
        [meta_fg_expand]: 1.48002e-06
        [replace_old_param]: 7.36999e-06
        [inline_without_move]: 4.52e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.46002e-06
        [auto_monad_grad]: 6.99976e-07
        [auto_monad_eliminator]: 5.12999e-06
        [cse]: 9.84001e-06
        [replace_applicator]: 5.02999e-06
[py_interpret_to_execute_after_opt_a]: 1.064e-05
[rewriter_after_opt_a]: 3.348e-05
[convert_after_rewriter]: 7.15e-06
[order_py_execute_after_rewriter]: 4.97e-06
[mutable_eliminate]: 0.00055728
[jit_opt_b]: 5.15e-05, [1]
    [Cycle 1]: 4.43e-05, [2]
        [frontend_op_eliminate]: 1.911e-05
        [inline_after_opt_a]: 1.419e-05
[cconv]: 2.535e-05
[loop_unroll]: 0.00047196
[jit_opt_after_cconv]: 0.00013981, [1]
    [Cycle 1]: 0.00013383, [11]
        [c_1]: 2.025e-05
        [parameter_eliminate]: 2.66e-06
        [updatestate_depend_eliminate]: 5.19e-06
        [updatestate_assign_eliminate]: 2.43998e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.805e-05
        [call_graph_tuple_transform]: 1.929e-05
        [tuple_list_get_item_eliminator]: 4.90999e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 3.30008e-07
        [switch_simplify]: 5.25001e-06
[remove_dup_value]: 1.36e-05
[partial_unused_args_eliminate]: 2.51998e-06
[environ_conv]: 5.02e-06
[add_recomputation]: 4.663e-05
[cse_after_recomputation]: 2.04e-05, [1]
    [Cycle 1]: 1.491e-05, [1]
        [cse]: 9.32001e-06
[auto_monad_reorder]: 1.466e-05
[get_jit_bprop_graph]: 1.89e-06
[rewriter_after_jit_bprop_graph]: 5.59e-06
[opt_after_jit_grad]: 0.00046128
[symbol_engine_optimizer]: 7.158e-05, [1]
    [Cycle 1]: 6.529e-05, [6]
        [build]: 3.71999e-06
        [elim_shapecalc]: 7.28999e-06
        [elim_not_effective]: 1.295e-05
        [opt_reshape]: 5.39e-06
        [fold_const_symbol]: 8.54e-06
        [renormalize]: 5.50004e-07
[validate]: 3.524e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.0059422
[execute]: 7.28999e-06
Sums
    bootstrap                            :     0.000273s :  1.94%
    type_inference                       :     0.004189s : 29.74%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000064s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.23%
    rewriter_before_opt_a                :     0.000071s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.42%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000616s :  4.38%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000632s :  4.49%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000035s :  0.25%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.08%
    rewriter_after_opt_a                 :     0.000033s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000557s :  3.96%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.18%
    loop_unroll                          :     0.000472s :  3.35%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000461s :  3.28%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005942s : 42.20%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000206    23
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.66% :     0.000005s :      3: substitution.graph_param_transform
83.57% :     0.000172s :      4: substitution.inline
 1.34% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.44% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.28% :     0.000005s :      2: substitution.replace_old_param
 6.16% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004131     2
81.89% :     0.003383s :      1: type_inference.infer
18.11% :     0.000748s :      1: type_inference.specialize
------[replace.]   0.000065     6
68.36% :     0.000045s :      4: replace.inline
31.64% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000182     6
93.64% :     0.000170s :      4: match.inline
 6.36% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.55% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.12% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 1.81% :     0.000002s :     10: predicate.arithmetic_simplify
 1.31% :     0.000001s :     10: predicate.cast_eliminate
 0.67% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000001s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.30% :     0.000000s :      3: predicate.fold_const_symbol
 1.11% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.61% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 2.15% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.90% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.79% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.16% :     0.000001s :     10: predicate.minmaximum_grad
 1.54% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000000s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.76% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.84% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.64% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.25% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.23% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.70% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.07% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.90% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.28% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000536    12
48.79% :     0.000261s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.60% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.60% :     0.000234s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019653    76
 0.25% :     0.000050s :      1: add_recomputation
 0.34% :     0.000067s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.45% :     0.000285s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.10% :     0.005326s :      1: jit_opt_a
 0.73% :     0.000143s :      1: jit_opt_after_cconv
 0.27% :     0.000054s :      1: jit_opt_b
 2.44% :     0.000480s :      1: loop_unroll
 2.88% :     0.000565s :      1: mutable_eliminate
 4.18% :     0.000822s :     26: opt.transform.jit_opt_a
 0.24% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000027s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.39% :     0.000469s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000035s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.77% :     0.000348s :      1: renormalize.infer
 1.41% :     0.000276s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000037s :      1: rewriter_after_opt_a
 0.37% :     0.000073s :      1: rewriter_before_opt_a
 0.38% :     0.000074s :      1: symbol_engine_optimizer
30.29% :     0.005952s :      1: task_emit
21.38% :     0.004201s :      1: type_inference
 0.28% :     0.000055s :      1: validate


TotalTime = 0.0142361, [30]
[bootstrap]: 0.00027116
[type_inference]: 0.00624975
[event_method]: 2.052e-05
[auto_monad]: 8.267e-05
[graph_reusing]: 6.96001e-06
[pre_auto_parallel]: 2.89001e-06
[py_interpret_to_execute]: 4.261e-05
[rewriter_before_opt_a]: 8.605e-05
[expand_dump_flag]: 3.55e-06
[jit_opt_a]: 0.00536241, [2]
    [Cycle 1]: 0.00182744, [27]
        [switch_simplify]: 0.0001085
        [loop_unroll]: 3.412e-05
        [a_1]: 0.00068593
        [with_stream_mark]: 1.976e-05
        [recompute_prepare]: 6.98998e-06
        [updatestate_depend_eliminate]: 3.61001e-06
        [updatestate_assign_eliminate]: 3.06999e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 5.87999e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 5.85002e-06
        [meta_shard_fg_expand]: 2.05002e-06
        [get_grad_eliminate_]: 5.46998e-06
        [merge_forward]: 3.86001e-06
        [cell_reuse_recompute_pass]: 1.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.605e-05
        [j_node_and_user_rematch]: 9.47001e-06
        [meta_fg_expand]: 2.50002e-06
        [replace_old_param]: 8.35001e-06
        [inline_without_move]: 5.30999e-06
        [renormalize]: 0.00066226
        [add_forward_monad_depend]: 5.25001e-06
        [auto_monad_grad]: 2.54999e-06
        [auto_monad_eliminator]: 1.436e-05
        [cse]: 2.507e-05
        [replace_applicator]: 1.2e-05
    [Cycle 2]: 0.00032343, [27]
        [switch_simplify]: 5.54e-06
        [loop_unroll]: 4.89998e-06
        [a_1]: 8.929e-05
        [with_stream_mark]: 9.95002e-06
        [recompute_prepare]: 5.20001e-06
        [updatestate_depend_eliminate]: 2.34999e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.91998e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 5.07e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.50999e-06
        [merge_forward]: 2.74001e-06
        [cell_reuse_recompute_pass]: 1.19998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.259e-05
        [j_node_and_user_rematch]: 8.11002e-06
        [meta_fg_expand]: 1.39e-06
        [replace_old_param]: 1.44e-05
        [inline_without_move]: 4.67998e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.32999e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.36002e-06
        [cse]: 1.076e-05
        [replace_applicator]: 4.99e-06
[py_interpret_to_execute_after_opt_a]: 1.228e-05
[rewriter_after_opt_a]: 3.344e-05
[convert_after_rewriter]: 7.38e-06
[order_py_execute_after_rewriter]: 5.64e-06
[mutable_eliminate]: 0.00056466
[jit_opt_b]: 4.854e-05, [1]
    [Cycle 1]: 4.123e-05, [2]
        [frontend_op_eliminate]: 1.572e-05
        [inline_after_opt_a]: 1.463e-05
[cconv]: 2.466e-05
[loop_unroll]: 0.00042507
[jit_opt_after_cconv]: 0.00013813, [1]
    [Cycle 1]: 0.00013175, [11]
        [c_1]: 2.066e-05
        [parameter_eliminate]: 2.39999e-06
        [updatestate_depend_eliminate]: 4.95001e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [cse]: 1.682e-05
        [call_graph_tuple_transform]: 1.94e-05
        [tuple_list_get_item_eliminator]: 5.37001e-06
        [none_parameter_eliminate]: 1.56002e-06
        [renormalize]: 6.69999e-07
        [switch_simplify]: 5.41998e-06
[remove_dup_value]: 1.277e-05
[partial_unused_args_eliminate]: 2.51998e-06
[environ_conv]: 5.77001e-06
[add_recomputation]: 4.803e-05
[cse_after_recomputation]: 1.995e-05, [1]
    [Cycle 1]: 1.45e-05, [1]
        [cse]: 8.73001e-06
[auto_monad_reorder]: 1.422e-05
[get_jit_bprop_graph]: 2.62001e-06
[rewriter_after_jit_bprop_graph]: 4.73001e-06
[opt_after_jit_grad]: 0.00047567
[symbol_engine_optimizer]: 7.157e-05, [1]
    [Cycle 1]: 6.518e-05, [6]
        [build]: 3.73999e-06
        [elim_shapecalc]: 7.05e-06
        [elim_not_effective]: 1.225e-05
        [opt_reshape]: 5.51e-06
        [fold_const_symbol]: 8.66002e-06
        [renormalize]: 7.7e-07
[validate]: 3.514e-05
Sums
    bootstrap                            :     0.000271s :  2.59%
    type_inference                       :     0.006250s : 59.69%
    event_method                         :     0.000021s :  0.20%
    auto_monad                           :     0.000083s :  0.79%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000043s :  0.41%
    rewriter_before_opt_a                :     0.000086s :  0.82%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000114s :  1.09%
    jit_opt_a.loop_unroll                :     0.000039s :  0.37%
    jit_opt_a.a_1                        :     0.000775s :  7.40%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000023s :  0.22%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000662s :  6.33%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.19%
    jit_opt_a.cse                        :     0.000036s :  0.34%
    jit_opt_a.replace_applicator         :     0.000017s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.12%
    rewriter_after_opt_a                 :     0.000033s :  0.32%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000565s :  5.39%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.14%
    cconv                                :     0.000025s :  0.24%
    loop_unroll                          :     0.000425s :  4.06%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.12%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000048s :  0.46%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000014s :  0.14%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000476s :  4.54%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000035s :  0.34%


TotalTime = 0.0193317, [33]
[bootstrap]: 0.00028382
[type_inference]: 0.00447984
[event_method]: 1.702e-05
[auto_monad]: 6.103e-05
[graph_reusing]: 5.40001e-06
[pre_auto_parallel]: 2.57001e-06
[py_interpret_to_execute]: 3.374e-05
[rewriter_before_opt_a]: 7.466e-05
[expand_dump_flag]: 3.18e-06
[jit_opt_a]: 0.00529072, [2]
    [Cycle 1]: 0.00150945, [27]
        [switch_simplify]: 4.98e-05
        [loop_unroll]: 3.018e-05
        [a_1]: 0.00049775
        [with_stream_mark]: 1.491e-05
        [recompute_prepare]: 6.94999e-06
        [updatestate_depend_eliminate]: 3.27002e-06
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 6.17001e-06
        [updatestate_useless_node_eliminater]: 5.34e-06
        [accelerated_algorithm]: 4.375e-05
        [meta_shard_fg_expand]: 2.11998e-06
        [get_grad_eliminate_]: 5.52001e-06
        [merge_forward]: 4.13999e-06
        [cell_reuse_recompute_pass]: 1.55001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.399e-05
        [j_node_and_user_rematch]: 8.62998e-06
        [meta_fg_expand]: 2.46e-06
        [replace_old_param]: 9.26998e-06
        [inline_without_move]: 5.47999e-06
        [renormalize]: 0.00056837
        [add_forward_monad_depend]: 5.12e-06
        [auto_monad_grad]: 2.22001e-06
        [auto_monad_eliminator]: 1.247e-05
        [cse]: 2.81e-05
        [replace_applicator]: 1.165e-05
    [Cycle 2]: 0.00032462, [27]
        [switch_simplify]: 5.76e-06
        [loop_unroll]: 4.89e-06
        [a_1]: 8.956e-05
        [with_stream_mark]: 8.65999e-06
        [recompute_prepare]: 5.57999e-06
        [updatestate_depend_eliminate]: 2.59001e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 5.08002e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 5.02999e-06
        [meta_shard_fg_expand]: 1.52999e-06
        [get_grad_eliminate_]: 4.62e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.60999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.195e-05
        [j_node_and_user_rematch]: 7.5e-06
        [meta_fg_expand]: 1.56998e-06
        [replace_old_param]: 7.19001e-06
        [inline_without_move]: 4.85999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 5.02e-06
        [cse]: 1.131e-05
        [replace_applicator]: 5.65001e-06
[py_interpret_to_execute_after_opt_a]: 8.75001e-06
[rewriter_after_opt_a]: 3.256e-05
[convert_after_rewriter]: 7.71999e-06
[order_py_execute_after_rewriter]: 5.20999e-06
[mutable_eliminate]: 0.00058956
[jit_opt_b]: 4.852e-05, [1]
    [Cycle 1]: 4.188e-05, [2]
        [frontend_op_eliminate]: 1.598e-05
        [inline_after_opt_a]: 1.451e-05
[cconv]: 2.355e-05
[loop_unroll]: 0.00042549
[jit_opt_after_cconv]: 0.0001393, [1]
    [Cycle 1]: 0.00013297, [11]
        [c_1]: 2.02e-05
        [parameter_eliminate]: 2.54001e-06
        [updatestate_depend_eliminate]: 5.07e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [cse]: 1.845e-05
        [call_graph_tuple_transform]: 1.805e-05
        [tuple_list_get_item_eliminator]: 5.35999e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 5.51002e-06
[remove_dup_value]: 1.334e-05
[partial_unused_args_eliminate]: 2.22001e-06
[environ_conv]: 4.89e-06
[add_recomputation]: 4.619e-05
[cse_after_recomputation]: 2.01e-05, [1]
    [Cycle 1]: 1.477e-05, [1]
        [cse]: 9.04998e-06
[auto_monad_reorder]: 1.56e-05
[get_jit_bprop_graph]: 1.67999e-06
[rewriter_after_jit_bprop_graph]: 4.82e-06
[opt_after_jit_grad]: 0.00045283
[symbol_engine_optimizer]: 7.113e-05, [1]
    [Cycle 1]: 6.505e-05, [6]
        [build]: 3.27997e-06
        [elim_shapecalc]: 7.57002e-06
        [elim_not_effective]: 1.234e-05
        [opt_reshape]: 5.71e-06
        [fold_const_symbol]: 8.54e-06
        [renormalize]: 6.10016e-07
[validate]: 3.394e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.00693705
[execute]: 8.11002e-06
Sums
    bootstrap                            :     0.000284s :  1.86%
    type_inference                       :     0.004480s : 29.31%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000061s :  0.40%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.22%
    rewriter_before_opt_a                :     0.000075s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.36%
    jit_opt_a.loop_unroll                :     0.000035s :  0.23%
    jit_opt_a.a_1                        :     0.000587s :  3.84%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000049s :  0.32%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000568s :  3.72%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.11%
    jit_opt_a.cse                        :     0.000039s :  0.26%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000590s :  3.86%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000024s :  0.15%
    loop_unroll                          :     0.000425s :  2.78%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000046s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000453s :  2.96%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006937s : 45.38%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000273    26
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000002s :      2: substitution.fold_const_symbol
 2.31% :     0.000006s :      3: substitution.graph_param_transform
84.70% :     0.000231s :      6: substitution.inline
 1.13% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.64% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.35% :     0.000004s :      2: substitution.replace_old_param
 2.81% :     0.000008s :      1: substitution.switch_simplify
 4.71% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006193     2
83.43% :     0.005167s :      1: type_inference.infer
16.57% :     0.001026s :      1: type_inference.specialize
------[replace.]   0.000128     9
50.78% :     0.000065s :      6: replace.inline
24.85% :     0.000032s :      1: replace.switch_simplify
24.37% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000247     9
92.44% :     0.000228s :      6: match.inline
 2.82% :     0.000007s :      1: match.switch_simplify
 4.74% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   801
 1.30% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.00% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.19% :     0.000002s :     12: predicate.addn_check_dump
 1.69% :     0.000002s :     12: predicate.addn_zero_filter
 2.05% :     0.000003s :     12: predicate.arithmetic_simplify
 1.32% :     0.000002s :     12: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.19% :     0.000002s :     12: predicate.compare_switch_simplify
 1.28% :     0.000002s :     12: predicate.depend_value_elim
 1.16% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     12: predicate.environ_get_eliminate
 1.12% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.35% :     0.000007s :     26: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     17: predicate.load_eliminater
 1.05% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000002s :     12: predicate.merge_addn
 1.12% :     0.000001s :     12: predicate.minmaximum_grad
 1.27% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     17: predicate.partial_eliminate
 1.19% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     12: predicate.reduce_eliminate
 1.47% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     20: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.65% :     0.000002s :     12: predicate.reshape_eliminate
 1.27% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     12: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.77% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.86% :     0.000004s :     20: predicate.switch_defer_inline
 2.36% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.75% :     0.000012s :     65: predicate.switch_simplify
 1.26% :     0.000002s :     12: predicate.tile_eliminate
 1.22% :     0.000002s :     12: predicate.transpose_eliminate
 1.47% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.89% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 3.18% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000740    15
48.17% :     0.000356s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.86% :     0.000095s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.97% :     0.000288s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016007    72
 0.32% :     0.000051s :      1: add_recomputation
 0.54% :     0.000086s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.77% :     0.000283s :      1: bootstrap
 0.17% :     0.000027s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.06% :     0.000010s :      1: graph_reusing
33.52% :     0.005365s :      1: jit_opt_a
 0.88% :     0.000141s :      1: jit_opt_after_cconv
 0.32% :     0.000051s :      1: jit_opt_b
 2.70% :     0.000433s :      1: loop_unroll
 3.58% :     0.000573s :      1: mutable_eliminate
 6.56% :     0.001050s :     26: opt.transform.jit_opt_a
 0.30% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.02% :     0.000483s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000045s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000015s :      1: remove_dup_value
 2.27% :     0.000363s :      1: renormalize.infer
 1.83% :     0.000292s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000037s :      1: rewriter_after_opt_a
 0.56% :     0.000089s :      1: rewriter_before_opt_a
 0.46% :     0.000074s :      1: symbol_engine_optimizer
39.11% :     0.006261s :      1: type_inference


Time group info:
------[substitution.]   0.000177    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 3.04% :     0.000005s :      3: substitution.graph_param_transform
82.19% :     0.000146s :      4: substitution.inline
 1.52% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.32% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.40% :     0.000004s :      2: substitution.replace_old_param
 6.72% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004420     2
81.75% :     0.003614s :      1: type_inference.infer
18.25% :     0.000807s :      1: type_inference.specialize
------[replace.]   0.000063     6
66.54% :     0.000042s :      4: replace.inline
33.46% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000154     6
92.99% :     0.000143s :      4: match.inline
 7.01% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.25% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.50% :     0.000002s :     10: predicate.addn_zero_filter
 2.02% :     0.000002s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.68% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.45% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.21% :     0.000006s :     22: predicate.inline
 1.05% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.61% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.23% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.60% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     15: predicate.partial_eliminate
 1.28% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.82% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.70% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.58% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.54% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.71% :     0.000009s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.28% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000639    12
54.24% :     0.000347s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 3.75% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.01% :     0.000268s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020833    76
 0.23% :     0.000049s :      1: add_recomputation
 0.31% :     0.000064s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.44% :     0.000301s :      1: bootstrap
 0.12% :     0.000026s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.41% :     0.005294s :      1: jit_opt_a
 0.68% :     0.000142s :      1: jit_opt_after_cconv
 0.25% :     0.000051s :      1: jit_opt_b
 2.08% :     0.000433s :      1: loop_unroll
 2.87% :     0.000599s :      1: mutable_eliminate
 3.98% :     0.000828s :     26: opt.transform.jit_opt_a
 0.22% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000460s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000036s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.23% :     0.000256s :      1: renormalize.infer
 1.47% :     0.000305s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000036s :      1: rewriter_after_opt_a
 0.37% :     0.000078s :      1: rewriter_before_opt_a
 0.35% :     0.000074s :      1: symbol_engine_optimizer
33.35% :     0.006948s :      1: task_emit
21.56% :     0.004492s :      1: type_inference
 0.25% :     0.000052s :      1: validate


TotalTime = 0.0175308, [33]
[bootstrap]: 0.00026147
[type_inference]: 0.00425787
[event_method]: 1.706e-05
[auto_monad]: 6.026e-05
[graph_reusing]: 5.70001e-06
[pre_auto_parallel]: 2.22001e-06
[py_interpret_to_execute]: 2.995e-05
[rewriter_before_opt_a]: 6.914e-05
[expand_dump_flag]: 3.14001e-06
[jit_opt_a]: 0.00492896, [2]
    [Cycle 1]: 0.00140235, [27]
        [switch_simplify]: 5.226e-05
        [loop_unroll]: 2.899e-05
        [a_1]: 0.0004953
        [with_stream_mark]: 1.653e-05
        [recompute_prepare]: 6.83e-06
        [updatestate_depend_eliminate]: 3.35e-06
        [updatestate_assign_eliminate]: 2.69001e-06
        [updatestate_loads_eliminate]: 2.58003e-06
        [parameter_eliminate]: 2.17001e-06
        [specialize_transform]: 5.59e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.34e-06
        [meta_shard_fg_expand]: 2.27999e-06
        [get_grad_eliminate_]: 5.05999e-06
        [merge_forward]: 3.85e-06
        [cell_reuse_recompute_pass]: 9.99979e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.298e-05
        [j_node_and_user_rematch]: 8.54e-06
        [meta_fg_expand]: 2.64001e-06
        [replace_old_param]: 7.87e-06
        [inline_without_move]: 5.05999e-06
        [renormalize]: 0.00051081
        [add_forward_monad_depend]: 5.17999e-06
        [auto_monad_grad]: 1.76e-06
        [auto_monad_eliminator]: 1.285e-05
        [cse]: 1.947e-05
        [replace_applicator]: 1.11e-05
    [Cycle 2]: 0.00031234, [27]
        [switch_simplify]: 5.79e-06
        [loop_unroll]: 4.70001e-06
        [a_1]: 8.774e-05
        [with_stream_mark]: 8.67e-06
        [recompute_prepare]: 5.03002e-06
        [updatestate_depend_eliminate]: 2.52001e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 4.90999e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 4.81002e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.57998e-06
        [merge_forward]: 2.36998e-06
        [cell_reuse_recompute_pass]: 1.58002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.176e-05
        [j_node_and_user_rematch]: 7.31001e-06
        [meta_fg_expand]: 1.87999e-06
        [replace_old_param]: 6.87002e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 5.01002e-06
        [cse]: 1.045e-05
        [replace_applicator]: 4.81002e-06
[py_interpret_to_execute_after_opt_a]: 8.52e-06
[rewriter_after_opt_a]: 3.131e-05
[convert_after_rewriter]: 6.50002e-06
[order_py_execute_after_rewriter]: 4.76002e-06
[mutable_eliminate]: 0.00050565
[jit_opt_b]: 4.783e-05, [1]
    [Cycle 1]: 4.139e-05, [2]
        [frontend_op_eliminate]: 1.527e-05
        [inline_after_opt_a]: 1.447e-05
[cconv]: 2.424e-05
[loop_unroll]: 0.00040639
[jit_opt_after_cconv]: 0.00013479, [1]
    [Cycle 1]: 0.0001288, [11]
        [c_1]: 2.023e-05
        [parameter_eliminate]: 3.21001e-06
        [updatestate_depend_eliminate]: 5.62999e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 1.98002e-06
        [cse]: 1.679e-05
        [call_graph_tuple_transform]: 1.799e-05
        [tuple_list_get_item_eliminator]: 5.24e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 5.12e-06
[remove_dup_value]: 1.48e-05
[partial_unused_args_eliminate]: 2.05002e-06
[environ_conv]: 4.57e-06
[add_recomputation]: 4.738e-05
[cse_after_recomputation]: 1.979e-05, [1]
    [Cycle 1]: 1.471e-05, [1]
        [cse]: 8.69998e-06
[auto_monad_reorder]: 1.471e-05
[get_jit_bprop_graph]: 1.79e-06
[rewriter_after_jit_bprop_graph]: 5.20999e-06
[opt_after_jit_grad]: 0.00044539
[symbol_engine_optimizer]: 7.448e-05, [1]
    [Cycle 1]: 6.796e-05, [6]
        [build]: 4.90001e-06
        [elim_shapecalc]: 7.91001e-06
        [elim_not_effective]: 1.222e-05
        [opt_reshape]: 5.70001e-06
        [fold_const_symbol]: 8.82999e-06
        [renormalize]: 6.20028e-07
[validate]: 3.541e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00587519
[execute]: 7.92e-06
Sums
    bootstrap                            :     0.000261s :  1.90%
    type_inference                       :     0.004258s : 30.98%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000060s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.22%
    rewriter_before_opt_a                :     0.000069s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.42%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000583s :  4.24%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000511s :  3.72%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000030s :  0.22%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000506s :  3.68%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000406s :  2.96%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000015s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000047s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000445s :  3.24%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005875s : 42.75%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000178    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.80% :     0.000001s :      2: substitution.fold_const_symbol
 2.90% :     0.000005s :      3: substitution.graph_param_transform
82.89% :     0.000148s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.40% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.84% :     0.000003s :      2: substitution.replace_old_param
 6.59% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004198     2
81.96% :     0.003441s :      1: type_inference.infer
18.04% :     0.000757s :      1: type_inference.specialize
------[replace.]   0.000063     6
67.90% :     0.000042s :      4: replace.inline
32.10% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000156     6
93.20% :     0.000145s :      4: match.inline
 6.80% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.30% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.21% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.52% :     0.000002s :     10: predicate.addn_zero_filter
 1.70% :     0.000002s :     10: predicate.arithmetic_simplify
 1.40% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000001s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.15% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.29% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.33% :     0.000006s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.60% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.93% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.65% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.34% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.88% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.71% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.57% :     0.000002s :     10: predicate.same_eliminate
 0.62% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.58% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.85% :     0.000009s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.40% :     0.000002s :     10: predicate.transpose_eliminate
 1.32% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.75% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000480    12
47.93% :     0.000230s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.51% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.56% :     0.000224s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018926    76
 0.27% :     0.000050s :      1: add_recomputation
 0.33% :     0.000063s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.44% :     0.000273s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.06% :     0.004932s :      1: jit_opt_a
 0.73% :     0.000138s :      1: jit_opt_after_cconv
 0.27% :     0.000050s :      1: jit_opt_b
 2.19% :     0.000414s :      1: loop_unroll
 2.71% :     0.000514s :      1: mutable_eliminate
 4.13% :     0.000781s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.39% :     0.000453s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000017s :      1: remove_dup_value
 1.39% :     0.000262s :      1: renormalize.infer
 1.27% :     0.000241s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.38% :     0.000072s :      1: rewriter_before_opt_a
 0.41% :     0.000077s :      1: symbol_engine_optimizer
31.09% :     0.005885s :      1: task_emit
22.57% :     0.004271s :      1: type_inference
 0.28% :     0.000053s :      1: validate


TotalTime = 0.0182261, [33]
[bootstrap]: 0.00027744
[type_inference]: 0.00413247
[event_method]: 1.594e-05
[auto_monad]: 4.792e-05
[graph_reusing]: 4.07998e-06
[pre_auto_parallel]: 1.97001e-06
[py_interpret_to_execute]: 3.028e-05
[rewriter_before_opt_a]: 6.626e-05
[expand_dump_flag]: 2.28998e-06
[jit_opt_a]: 0.00512953, [2]
    [Cycle 1]: 0.00149336, [27]
        [switch_simplify]: 4.555e-05
        [loop_unroll]: 3.048e-05
        [a_1]: 0.00055849
        [with_stream_mark]: 1.255e-05
        [recompute_prepare]: 7.25e-06
        [updatestate_depend_eliminate]: 3.13998e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 1.32e-06
        [specialize_transform]: 5.76998e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 5.25001e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 5.42999e-06
        [merge_forward]: 2.91e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.095e-05
        [j_node_and_user_rematch]: 8.86002e-06
        [meta_fg_expand]: 1.84e-06
        [replace_old_param]: 8.00999e-06
        [inline_without_move]: 3.318e-05
        [renormalize]: 0.0005327
        [add_forward_monad_depend]: 4.30999e-06
        [auto_monad_grad]: 1.29e-06
        [auto_monad_eliminator]: 9.79999e-06
        [cse]: 1.631e-05
        [replace_applicator]: 1.156e-05
    [Cycle 2]: 0.00032166, [27]
        [switch_simplify]: 5.71998e-06
        [loop_unroll]: 4.93001e-06
        [a_1]: 9.171e-05
        [with_stream_mark]: 8.85999e-06
        [recompute_prepare]: 5.21998e-06
        [updatestate_depend_eliminate]: 2.41998e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 4.89998e-06
        [updatestate_useless_node_eliminater]: 4.70999e-06
        [accelerated_algorithm]: 5.29e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.50999e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.50999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.867e-05
        [j_node_and_user_rematch]: 7.88001e-06
        [meta_fg_expand]: 1.65001e-06
        [replace_old_param]: 6.99001e-06
        [inline_without_move]: 4.60001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 6.90023e-07
        [auto_monad_eliminator]: 4.86002e-06
        [cse]: 9.95002e-06
        [replace_applicator]: 7.02997e-06
[py_interpret_to_execute_after_opt_a]: 1.678e-05
[rewriter_after_opt_a]: 2.828e-05
[convert_after_rewriter]: 6.17999e-06
[order_py_execute_after_rewriter]: 4.43001e-06
[mutable_eliminate]: 0.00048378
[jit_opt_b]: 4.739e-05, [1]
    [Cycle 1]: 4.097e-05, [2]
        [frontend_op_eliminate]: 1.586e-05
        [inline_after_opt_a]: 1.443e-05
[cconv]: 1.603e-05
[loop_unroll]: 0.00043849
[jit_opt_after_cconv]: 0.00013354, [1]
    [Cycle 1]: 0.00012714, [11]
        [c_1]: 2.051e-05
        [parameter_eliminate]: 2.26e-06
        [updatestate_depend_eliminate]: 5.05001e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.08998e-06
        [cse]: 1.624e-05
        [call_graph_tuple_transform]: 1.64e-05
        [tuple_list_get_item_eliminator]: 5.45001e-06
        [none_parameter_eliminate]: 1.04e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 5.50001e-06
[remove_dup_value]: 7.21999e-06
[partial_unused_args_eliminate]: 1.66998e-06
[environ_conv]: 3.59002e-06
[add_recomputation]: 3.292e-05
[cse_after_recomputation]: 1.958e-05, [1]
    [Cycle 1]: 1.45e-05, [1]
        [cse]: 8.87e-06
[auto_monad_reorder]: 9.74999e-06
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 4.57e-06
[opt_after_jit_grad]: 0.00049661
[symbol_engine_optimizer]: 7.332e-05, [1]
    [Cycle 1]: 6.7e-05, [6]
        [build]: 2.96999e-06
        [elim_shapecalc]: 7.23999e-06
        [elim_not_effective]: 1.331e-05
        [opt_reshape]: 5.71998e-06
        [fold_const_symbol]: 8.84e-06
        [renormalize]: 5.19998e-07
[validate]: 2.702e-05
[backend_pass]: 7.09988e-07
[task_emit]: 0.00646969
[execute]: 7.88001e-06
Sums
    bootstrap                            :     0.000277s :  1.94%
    type_inference                       :     0.004132s : 28.84%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000048s :  0.33%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000030s :  0.21%
    rewriter_before_opt_a                :     0.000066s :  0.46%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000051s :  0.36%
    jit_opt_a.loop_unroll                :     0.000035s :  0.25%
    jit_opt_a.a_1                        :     0.000650s :  4.54%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000038s :  0.26%
    jit_opt_a.renormalize                :     0.000533s :  3.72%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.10%
    jit_opt_a.cse                        :     0.000026s :  0.18%
    jit_opt_a.replace_applicator         :     0.000019s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.12%
    rewriter_after_opt_a                 :     0.000028s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000484s :  3.38%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000016s :  0.11%
    loop_unroll                          :     0.000438s :  3.06%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000007s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000033s :  0.23%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000010s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000497s :  3.47%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000027s :  0.19%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006470s : 45.14%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000164    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.16% :     0.000004s :      3: substitution.graph_param_transform
80.52% :     0.000132s :      4: substitution.inline
 1.88% :     0.000003s :      4: substitution.j_node_and_user_rematch
 6.46% :     0.000011s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000003s :      2: substitution.replace_old_param
 5.29% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004080     2
81.94% :     0.003343s :      1: type_inference.infer
18.06% :     0.000737s :      1: type_inference.specialize
------[replace.]   0.000134     6
83.19% :     0.000112s :      4: replace.inline
16.81% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000137     6
94.57% :     0.000129s :      4: match.inline
 5.43% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.46% :     0.000002s :     10: predicate.addn_zero_filter
 2.07% :     0.000002s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.22% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 4.94% :     0.000006s :     22: predicate.inline
 1.11% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.32% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.02% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.63% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.46% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000001s :     10: predicate.same_eliminate
 0.60% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.50% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.60% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.65% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.49% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.10% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.90% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000509    12
47.87% :     0.000243s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.13% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.00% :     0.000239s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019748    76
 0.18% :     0.000036s :      1: add_recomputation
 0.26% :     0.000051s :      1: auto_monad
 0.06% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.49% :     0.000294s :      1: bootstrap
 0.09% :     0.000019s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
25.99% :     0.005132s :      1: jit_opt_a
 0.69% :     0.000137s :      1: jit_opt_after_cconv
 0.25% :     0.000050s :      1: jit_opt_b
 2.27% :     0.000447s :      1: loop_unroll
 2.49% :     0.000492s :      1: mutable_eliminate
 4.48% :     0.000884s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.56% :     0.000505s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.10% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000010s :      1: remove_dup_value
 1.33% :     0.000263s :      1: renormalize.infer
 1.33% :     0.000263s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000032s :      1: rewriter_after_opt_a
 0.35% :     0.000069s :      1: rewriter_before_opt_a
 0.38% :     0.000076s :      1: symbol_engine_optimizer
32.82% :     0.006481s :      1: task_emit
20.99% :     0.004144s :      1: type_inference
 0.22% :     0.000044s :      1: validate


TotalTime = 0.0146381, [30]
[bootstrap]: 0.00035926
[type_inference]: 0.00624703
[event_method]: 2.161e-05
[auto_monad]: 8.198e-05
[graph_reusing]: 7.56999e-06
[pre_auto_parallel]: 2.55002e-06
[py_interpret_to_execute]: 5.119e-05
[rewriter_before_opt_a]: 8.826e-05
[expand_dump_flag]: 3.05002e-06
[jit_opt_a]: 0.0056203, [2]
    [Cycle 1]: 0.00185177, [27]
        [switch_simplify]: 0.00012342
        [loop_unroll]: 3.481e-05
        [a_1]: 0.00069415
        [with_stream_mark]: 1.812e-05
        [recompute_prepare]: 9.04998e-06
        [updatestate_depend_eliminate]: 3.65e-06
        [updatestate_assign_eliminate]: 3.11999e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 5.74e-06
        [updatestate_useless_node_eliminater]: 4.83001e-06
        [accelerated_algorithm]: 6.00002e-06
        [meta_shard_fg_expand]: 2.27001e-06
        [get_grad_eliminate_]: 5.24e-06
        [merge_forward]: 4.33999e-06
        [cell_reuse_recompute_pass]: 1.01997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.519e-05
        [j_node_and_user_rematch]: 9.41e-06
        [meta_fg_expand]: 2.21998e-06
        [replace_old_param]: 9.54999e-06
        [inline_without_move]: 5.32001e-06
        [renormalize]: 0.00064704
        [add_forward_monad_depend]: 5.08002e-06
        [auto_monad_grad]: 1.89999e-06
        [auto_monad_eliminator]: 1.376e-05
        [cse]: 2.774e-05
        [replace_applicator]: 1.383e-05
    [Cycle 2]: 0.00036023, [27]
        [switch_simplify]: 5.50001e-06
        [loop_unroll]: 4.84e-06
        [a_1]: 0.00013038
        [with_stream_mark]: 1.015e-05
        [recompute_prepare]: 5.32001e-06
        [updatestate_depend_eliminate]: 2.71e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 5.00999e-06
        [updatestate_useless_node_eliminater]: 4.73001e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 1.22999e-06
        [get_grad_eliminate_]: 4.63999e-06
        [merge_forward]: 2.81e-06
        [cell_reuse_recompute_pass]: 1.59998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.27e-05
        [j_node_and_user_rematch]: 8.45999e-06
        [meta_fg_expand]: 1.55999e-06
        [replace_old_param]: 8.05e-06
        [inline_without_move]: 4.75999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 8.79983e-07
        [auto_monad_eliminator]: 5.37001e-06
        [cse]: 1.043e-05
        [replace_applicator]: 5.15001e-06
[py_interpret_to_execute_after_opt_a]: 8.60001e-06
[rewriter_after_opt_a]: 3.088e-05
[convert_after_rewriter]: 7.77e-06
[order_py_execute_after_rewriter]: 5.34e-06
[mutable_eliminate]: 0.00058658
[jit_opt_b]: 5.662e-05, [1]
    [Cycle 1]: 5.003e-05, [2]
        [frontend_op_eliminate]: 1.558e-05
        [inline_after_opt_a]: 2.282e-05
[cconv]: 2.535e-05
[loop_unroll]: 0.00043331
[jit_opt_after_cconv]: 0.00013936, [1]
    [Cycle 1]: 0.00013313, [11]
        [c_1]: 2.099e-05
        [parameter_eliminate]: 2.69001e-06
        [updatestate_depend_eliminate]: 5.61998e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.724e-05
        [call_graph_tuple_transform]: 1.84e-05
        [tuple_list_get_item_eliminator]: 5.22999e-06
        [none_parameter_eliminate]: 1.77999e-06
        [renormalize]: 9.5999e-07
        [switch_simplify]: 5.52001e-06
[remove_dup_value]: 1.35e-05
[partial_unused_args_eliminate]: 2.19999e-06
[environ_conv]: 5.46e-06
[add_recomputation]: 4.951e-05
[cse_after_recomputation]: 2.102e-05, [1]
    [Cycle 1]: 1.552e-05, [1]
        [cse]: 9.67999e-06
[auto_monad_reorder]: 1.648e-05
[get_jit_bprop_graph]: 1.58002e-06
[rewriter_after_jit_bprop_graph]: 5.06002e-06
[opt_after_jit_grad]: 0.00046549
[symbol_engine_optimizer]: 7.27e-05, [1]
    [Cycle 1]: 6.668e-05, [6]
        [build]: 3.93001e-06
        [elim_shapecalc]: 7.53999e-06
        [elim_not_effective]: 1.243e-05
        [opt_reshape]: 5.56e-06
        [fold_const_symbol]: 8.65999e-06
        [renormalize]: 7.50006e-07
[validate]: 3.57e-05
Sums
    bootstrap                            :     0.000359s :  3.37%
    type_inference                       :     0.006247s : 58.69%
    event_method                         :     0.000022s :  0.20%
    auto_monad                           :     0.000082s :  0.77%
    graph_reusing                        :     0.000008s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000051s :  0.48%
    rewriter_before_opt_a                :     0.000088s :  0.83%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000129s :  1.21%
    jit_opt_a.loop_unroll                :     0.000040s :  0.37%
    jit_opt_a.a_1                        :     0.000825s :  7.75%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.17%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000647s :  6.08%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.18%
    jit_opt_a.cse                        :     0.000038s :  0.36%
    jit_opt_a.replace_applicator         :     0.000019s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.08%
    rewriter_after_opt_a                 :     0.000031s :  0.29%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000587s :  5.51%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000023s :  0.21%
    cconv                                :     0.000025s :  0.24%
    loop_unroll                          :     0.000433s :  4.07%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000014s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000050s :  0.47%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000465s :  4.37%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000036s :  0.34%


Time group info:
------[substitution.]   0.000276    26
 0.66% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000002s :      2: substitution.fold_const_symbol
 1.98% :     0.000005s :      3: substitution.graph_param_transform
83.80% :     0.000231s :      6: substitution.inline
 1.18% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.90% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.91% :     0.000005s :      2: substitution.replace_old_param
 2.94% :     0.000008s :      1: substitution.switch_simplify
 5.05% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006183     2
83.11% :     0.005139s :      1: type_inference.infer
16.89% :     0.001044s :      1: type_inference.specialize
------[replace.]   0.000139     9
50.61% :     0.000070s :      6: replace.inline
25.87% :     0.000036s :      1: replace.switch_simplify
23.51% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000248     9
91.93% :     0.000228s :      6: match.inline
 2.94% :     0.000007s :      1: match.switch_simplify
 5.12% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   801
 1.34% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.19% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     12: predicate.addn_check_dump
 1.39% :     0.000002s :     12: predicate.addn_zero_filter
 1.87% :     0.000002s :     12: predicate.arithmetic_simplify
 1.24% :     0.000002s :     12: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000002s :     12: predicate.compare_switch_simplify
 1.48% :     0.000002s :     12: predicate.depend_value_elim
 1.13% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.64% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     12: predicate.environ_get_eliminate
 1.12% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.76% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 5.49% :     0.000007s :     26: predicate.inline
 0.76% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.75% :     0.000002s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     17: predicate.load_eliminater
 1.33% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.35% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.01% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000002s :     12: predicate.merge_addn
 1.13% :     0.000001s :     12: predicate.minmaximum_grad
 1.30% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     17: predicate.partial_eliminate
 1.27% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     12: predicate.reduce_eliminate
 1.61% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     20: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     12: predicate.reshape_eliminate
 1.27% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.51% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.89% :     0.000004s :     20: predicate.switch_defer_inline
 2.62% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.59% :     0.000011s :     65: predicate.switch_simplify
 1.20% :     0.000002s :     12: predicate.tile_eliminate
 1.30% :     0.000002s :     12: predicate.transpose_eliminate
 1.51% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.95% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.55% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000002s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000747    15
48.13% :     0.000359s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.53% :     0.000086s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.34% :     0.000301s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016454    72
 0.32% :     0.000052s :      1: add_recomputation
 0.59% :     0.000097s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 2.25% :     0.000370s :      1: bootstrap
 0.17% :     0.000028s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000010s :      1: graph_reusing
34.18% :     0.005623s :      1: jit_opt_a
 0.87% :     0.000142s :      1: jit_opt_after_cconv
 0.36% :     0.000059s :      1: jit_opt_b
 2.68% :     0.000441s :      1: loop_unroll
 3.62% :     0.000595s :      1: mutable_eliminate
 6.75% :     0.001111s :     26: opt.transform.jit_opt_a
 0.28% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000032s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.88% :     0.000474s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.33% :     0.000054s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.20% :     0.000363s :      1: renormalize.infer
 1.68% :     0.000277s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000034s :      1: rewriter_after_opt_a
 0.55% :     0.000091s :      1: rewriter_before_opt_a
 0.46% :     0.000075s :      1: symbol_engine_optimizer
38.04% :     0.006259s :      1: type_inference


TotalTime = 0.0176408, [33]
[bootstrap]: 0.00026582
[type_inference]: 0.00429335
[event_method]: 1.745e-05
[auto_monad]: 6.111e-05
[graph_reusing]: 6.32001e-06
[pre_auto_parallel]: 2.48998e-06
[py_interpret_to_execute]: 3.198e-05
[rewriter_before_opt_a]: 6.79e-05
[expand_dump_flag]: 2.86999e-06
[jit_opt_a]: 0.00496127, [2]
    [Cycle 1]: 0.0014688, [27]
        [switch_simplify]: 4.645e-05
        [loop_unroll]: 2.899e-05
        [a_1]: 0.00048538
        [with_stream_mark]: 1.54e-05
        [recompute_prepare]: 6.83e-06
        [updatestate_depend_eliminate]: 3.33e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 5.99e-06
        [updatestate_useless_node_eliminater]: 5.10999e-06
        [accelerated_algorithm]: 5.40001e-06
        [meta_shard_fg_expand]: 1.64e-06
        [get_grad_eliminate_]: 5.10001e-06
        [merge_forward]: 3.04999e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.311e-05
        [j_node_and_user_rematch]: 9.61e-06
        [meta_fg_expand]: 2.02999e-06
        [replace_old_param]: 8.05e-06
        [inline_without_move]: 5.26002e-06
        [renormalize]: 0.00060117
        [add_forward_monad_depend]: 4.21001e-06
        [auto_monad_grad]: 2.22001e-06
        [auto_monad_eliminator]: 1.042e-05
        [cse]: 1.87e-05
        [replace_applicator]: 1.136e-05
    [Cycle 2]: 0.00031231, [27]
        [switch_simplify]: 5.27001e-06
        [loop_unroll]: 4.80001e-06
        [a_1]: 8.842e-05
        [with_stream_mark]: 8.65001e-06
        [recompute_prepare]: 5.13002e-06
        [updatestate_depend_eliminate]: 2.61999e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 5.06002e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.62998e-06
        [merge_forward]: 2.24999e-06
        [cell_reuse_recompute_pass]: 1.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.148e-05
        [j_node_and_user_rematch]: 7.58999e-06
        [meta_fg_expand]: 1.45001e-06
        [replace_old_param]: 6.73998e-06
        [inline_without_move]: 4.65999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 5.13002e-06
        [cse]: 1.079e-05
        [replace_applicator]: 5.54e-06
[py_interpret_to_execute_after_opt_a]: 9.64e-06
[rewriter_after_opt_a]: 3.001e-05
[convert_after_rewriter]: 6.29001e-06
[order_py_execute_after_rewriter]: 4.41002e-06
[mutable_eliminate]: 0.00054233
[jit_opt_b]: 5.437e-05, [1]
    [Cycle 1]: 4.798e-05, [2]
        [frontend_op_eliminate]: 2.109e-05
        [inline_after_opt_a]: 1.492e-05
[cconv]: 2.043e-05
[loop_unroll]: 0.00041721
[jit_opt_after_cconv]: 0.000133, [1]
    [Cycle 1]: 0.00012669, [11]
        [c_1]: 2.032e-05
        [parameter_eliminate]: 2.00002e-06
        [updatestate_depend_eliminate]: 4.97e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.644e-05
        [call_graph_tuple_transform]: 1.674e-05
        [tuple_list_get_item_eliminator]: 5.49998e-06
        [none_parameter_eliminate]: 8.70001e-07
        [renormalize]: 3.80009e-07
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 1.137e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 4.40999e-06
[add_recomputation]: 4.028e-05
[cse_after_recomputation]: 1.952e-05, [1]
    [Cycle 1]: 1.43e-05, [1]
        [cse]: 8.77e-06
[auto_monad_reorder]: 1.227e-05
[get_jit_bprop_graph]: 1.75001e-06
[rewriter_after_jit_bprop_graph]: 4.14997e-06
[opt_after_jit_grad]: 0.00047651
[symbol_engine_optimizer]: 7.083e-05, [1]
    [Cycle 1]: 6.408e-05, [6]
        [build]: 2.81e-06
        [elim_shapecalc]: 7.62002e-06
        [elim_not_effective]: 1.227e-05
        [opt_reshape]: 5.57001e-06
        [fold_const_symbol]: 8.12e-06
        [renormalize]: 4.80009e-07
[validate]: 3.091e-05
[backend_pass]: 6.89994e-07
[task_emit]: 0.00584715
[execute]: 8.28999e-06
Sums
    bootstrap                            :     0.000266s :  1.91%
    type_inference                       :     0.004293s : 30.93%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000061s :  0.44%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.23%
    rewriter_before_opt_a                :     0.000068s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000052s :  0.37%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000574s :  4.13%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000601s :  4.33%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.11%
    jit_opt_a.cse                        :     0.000029s :  0.21%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000030s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000542s :  3.91%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000020s :  0.15%
    loop_unroll                          :     0.000417s :  3.01%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000011s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000040s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000477s :  3.43%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.22%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005847s : 42.12%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000167    23
 1.18% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.83% :     0.000005s :      3: substitution.graph_param_transform
83.00% :     0.000138s :      4: substitution.inline
 1.67% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.33% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.99% :     0.000003s :      2: substitution.replace_old_param
 6.37% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004231     2
81.30% :     0.003440s :      1: type_inference.infer
18.70% :     0.000791s :      1: type_inference.specialize
------[replace.]   0.000062     6
66.32% :     0.000041s :      4: replace.inline
33.68% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000145     6
93.42% :     0.000136s :      4: match.inline
 6.58% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.26% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.28% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 1.94% :     0.000002s :     10: predicate.arithmetic_simplify
 1.64% :     0.000002s :     10: predicate.cast_eliminate
 4.72% :     0.000006s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.51% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.34% :     0.000000s :      3: predicate.graph_param_transform
 4.94% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.44% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.93% :     0.000002s :     15: predicate.load_eliminater
 1.33% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.25% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.02% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.48% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000002s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000001s :     10: predicate.same_eliminate
 0.65% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.32% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.21% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.33% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000515    12
49.85% :     0.000257s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.99% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.16% :     0.000233s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019120    76
 0.22% :     0.000043s :      1: add_recomputation
 0.34% :     0.000064s :      1: auto_monad
 0.08% :     0.000015s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.45% :     0.000276s :      1: bootstrap
 0.12% :     0.000023s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.96% :     0.004964s :      1: jit_opt_a
 0.71% :     0.000136s :      1: jit_opt_after_cconv
 0.30% :     0.000057s :      1: jit_opt_b
 2.23% :     0.000426s :      1: loop_unroll
 2.88% :     0.000551s :      1: mutable_eliminate
 4.03% :     0.000771s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000030s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.54% :     0.000485s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000035s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000014s :      1: remove_dup_value
 1.76% :     0.000337s :      1: renormalize.infer
 1.35% :     0.000257s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000033s :      1: rewriter_after_opt_a
 0.37% :     0.000071s :      1: rewriter_before_opt_a
 0.38% :     0.000073s :      1: symbol_engine_optimizer
30.66% :     0.005862s :      1: task_emit
22.51% :     0.004305s :      1: type_inference
 0.26% :     0.000049s :      1: validate


TotalTime = 0.0198218, [33]
[bootstrap]: 0.00028363
[type_inference]: 0.00434872
[event_method]: 1.711e-05
[auto_monad]: 6.495e-05
[graph_reusing]: 5.99e-06
[pre_auto_parallel]: 2.52001e-06
[py_interpret_to_execute]: 3.642e-05
[rewriter_before_opt_a]: 7.811e-05
[expand_dump_flag]: 3.19001e-06
[jit_opt_a]: 0.0054117, [2]
    [Cycle 1]: 0.00152115, [27]
        [switch_simplify]: 5.397e-05
        [loop_unroll]: 3.048e-05
        [a_1]: 0.0005218
        [with_stream_mark]: 1.8e-05
        [recompute_prepare]: 7.82002e-06
        [updatestate_depend_eliminate]: 3.82998e-06
        [updatestate_assign_eliminate]: 2.79999e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [parameter_eliminate]: 2.16e-06
        [specialize_transform]: 6.29999e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 6.12001e-06
        [meta_shard_fg_expand]: 1.93002e-06
        [get_grad_eliminate_]: 5.42001e-06
        [merge_forward]: 3.61001e-06
        [cell_reuse_recompute_pass]: 1.26997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.552e-05
        [j_node_and_user_rematch]: 9.43002e-06
        [meta_fg_expand]: 2.36e-06
        [replace_old_param]: 8.92999e-06
        [inline_without_move]: 5.93002e-06
        [renormalize]: 0.00054682
        [add_forward_monad_depend]: 5.27001e-06
        [auto_monad_grad]: 2.56e-06
        [auto_monad_eliminator]: 1.424e-05
        [cse]: 3.358e-05
        [replace_applicator]: 1.253e-05
    [Cycle 2]: 0.00033238, [27]
        [switch_simplify]: 5.67001e-06
        [loop_unroll]: 4.77e-06
        [a_1]: 9.222e-05
        [with_stream_mark]: 1.049e-05
        [recompute_prepare]: 5.87999e-06
        [updatestate_depend_eliminate]: 2.39001e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.30001e-06
        [specialize_transform]: 5.12e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 5.09e-06
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.65001e-06
        [merge_forward]: 2.72001e-06
        [cell_reuse_recompute_pass]: 1.66002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.26e-05
        [j_node_and_user_rematch]: 8.10999e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 1.436e-05
        [inline_without_move]: 4.77e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.88002e-06
        [auto_monad_grad]: 1.25001e-06
        [auto_monad_eliminator]: 5.71998e-06
        [cse]: 1.147e-05
        [replace_applicator]: 5.25001e-06
[py_interpret_to_execute_after_opt_a]: 1.032e-05
[rewriter_after_opt_a]: 3.321e-05
[convert_after_rewriter]: 7.19001e-06
[order_py_execute_after_rewriter]: 5.37999e-06
[mutable_eliminate]: 0.00054886
[jit_opt_b]: 5.03e-05, [1]
    [Cycle 1]: 4.326e-05, [2]
        [frontend_op_eliminate]: 1.637e-05
        [inline_after_opt_a]: 1.488e-05
[cconv]: 2.396e-05
[loop_unroll]: 0.00043422
[jit_opt_after_cconv]: 0.00014555, [1]
    [Cycle 1]: 0.00013898, [11]
        [c_1]: 2.214e-05
        [parameter_eliminate]: 2.51998e-06
        [updatestate_depend_eliminate]: 5.77001e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 1.968e-05
        [call_graph_tuple_transform]: 1.988e-05
        [tuple_list_get_item_eliminator]: 5.66e-06
        [none_parameter_eliminate]: 1.35001e-06
        [renormalize]: 6.10016e-07
        [switch_simplify]: 5.42999e-06
[remove_dup_value]: 1.297e-05
[partial_unused_args_eliminate]: 2.78e-06
[environ_conv]: 5.59e-06
[add_recomputation]: 4.601e-05
[cse_after_recomputation]: 2.106e-05, [1]
    [Cycle 1]: 1.528e-05, [1]
        [cse]: 8.77e-06
[auto_monad_reorder]: 1.52e-05
[get_jit_bprop_graph]: 1.82001e-06
[rewriter_after_jit_bprop_graph]: 7.08998e-06
[opt_after_jit_grad]: 0.00048197
[symbol_engine_optimizer]: 7.432e-05, [1]
    [Cycle 1]: 6.785e-05, [6]
        [build]: 4.09002e-06
        [elim_shapecalc]: 7.63999e-06
        [elim_not_effective]: 1.337e-05
        [opt_reshape]: 5.39998e-06
        [fold_const_symbol]: 9.20001e-06
        [renormalize]: 4.50003e-07
[validate]: 3.208e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.007373
[execute]: 9.86e-06
Sums
    bootstrap                            :     0.000284s :  1.82%
    type_inference                       :     0.004349s : 27.86%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000065s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.23%
    rewriter_before_opt_a                :     0.000078s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000060s :  0.38%
    jit_opt_a.loop_unroll                :     0.000035s :  0.23%
    jit_opt_a.a_1                        :     0.000614s :  3.93%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000023s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000547s :  3.50%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000045s :  0.29%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000549s :  3.52%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000024s :  0.15%
    loop_unroll                          :     0.000434s :  2.78%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000046s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.05%
    opt_after_jit_grad                   :     0.000482s :  3.09%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007373s : 47.23%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000197    23
 1.15% :     0.000002s :      2: substitution.elim_not_effective
 0.83% :     0.000002s :      2: substitution.fold_const_symbol
 2.95% :     0.000006s :      3: substitution.graph_param_transform
79.42% :     0.000156s :      4: substitution.inline
 1.82% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.18% :     0.000004s :      4: substitution.remove_not_recompute_node
 5.62% :     0.000011s :      2: substitution.replace_old_param
 6.03% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004285     2
81.60% :     0.003497s :      1: type_inference.infer
18.40% :     0.000788s :      1: type_inference.specialize
------[replace.]   0.000067     6
65.95% :     0.000044s :      4: replace.inline
34.05% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000165     6
93.54% :     0.000154s :      4: match.inline
 6.46% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.26% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.17% :     0.000001s :     10: predicate.addn_zero_filter
 2.23% :     0.000003s :     10: predicate.arithmetic_simplify
 1.51% :     0.000002s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.57% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.41% :     0.000002s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000001s :      3: predicate.graph_param_transform
 6.25% :     0.000008s :     22: predicate.inline
 1.08% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.02% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.61% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     15: predicate.partial_eliminate
 1.49% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.35% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 1.12% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.13% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.93% :     0.000010s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.94% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000531    12
50.25% :     0.000267s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.28% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.47% :     0.000236s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021303    76
 0.23% :     0.000049s :      1: add_recomputation
 0.32% :     0.000069s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.40% :     0.000298s :      1: bootstrap
 0.12% :     0.000027s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000010s :      1: graph_reusing
25.42% :     0.005415s :      1: jit_opt_a
 0.70% :     0.000149s :      1: jit_opt_after_cconv
 0.25% :     0.000053s :      1: jit_opt_b
 2.15% :     0.000457s :      1: loop_unroll
 2.62% :     0.000558s :      1: mutable_eliminate
 3.92% :     0.000834s :     26: opt.transform.jit_opt_a
 0.23% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.30% :     0.000491s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000039s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.23% :     0.000263s :      1: renormalize.infer
 1.29% :     0.000276s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000036s :      1: rewriter_after_opt_a
 0.38% :     0.000081s :      1: rewriter_before_opt_a
 0.36% :     0.000077s :      1: symbol_engine_optimizer
34.70% :     0.007393s :      1: task_emit
20.47% :     0.004361s :      1: type_inference
 0.24% :     0.000051s :      1: validate


TotalTime = 0.0134023, [30]
[bootstrap]: 0.00029715
[type_inference]: 0.00611265
[event_method]: 1.84e-05
[auto_monad]: 5.449e-05
[graph_reusing]: 3.93001e-06
[pre_auto_parallel]: 1.89e-06
[py_interpret_to_execute]: 4.222e-05
[rewriter_before_opt_a]: 7.365e-05
[expand_dump_flag]: 1.71e-06
[jit_opt_a]: 0.00491559, [2]
    [Cycle 1]: 0.00155987, [27]
        [switch_simplify]: 0.00010914
        [loop_unroll]: 3.439e-05
        [a_1]: 0.00061331
        [with_stream_mark]: 1.096e-05
        [recompute_prepare]: 6.86001e-06
        [updatestate_depend_eliminate]: 2.49001e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 2.08002e-06
        [parameter_eliminate]: 9.20001e-07
        [specialize_transform]: 5.76e-06
        [updatestate_useless_node_eliminater]: 5.06002e-06
        [accelerated_algorithm]: 5.82001e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 5.14998e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 7.59988e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.067e-05
        [j_node_and_user_rematch]: 8.95001e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 7.16001e-06
        [inline_without_move]: 5.23002e-06
        [renormalize]: 0.00052908
        [add_forward_monad_depend]: 3.22002e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 7.85e-06
        [cse]: 1.376e-05
        [replace_applicator]: 8.99998e-06
    [Cycle 2]: 0.0003026, [27]
        [switch_simplify]: 5.57999e-06
        [loop_unroll]: 4.90001e-06
        [a_1]: 8.953e-05
        [with_stream_mark]: 6.56999e-06
        [recompute_prepare]: 4.91002e-06
        [updatestate_depend_eliminate]: 2.35002e-06
        [updatestate_assign_eliminate]: 1.86e-06
        [updatestate_loads_eliminate]: 1.75001e-06
        [parameter_eliminate]: 7.09988e-07
        [specialize_transform]: 4.77e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 5.38002e-06
        [meta_shard_fg_expand]: 8.99978e-07
        [get_grad_eliminate_]: 4.40999e-06
        [merge_forward]: 2.09999e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.136e-05
        [j_node_and_user_rematch]: 7.60998e-06
        [meta_fg_expand]: 1.40999e-06
        [replace_old_param]: 6.88e-06
        [inline_without_move]: 4.73001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 9.39996e-07
        [auto_monad_grad]: 5.89993e-07
        [auto_monad_eliminator]: 4.27e-06
        [cse]: 8.55001e-06
        [replace_applicator]: 5.09e-06
[py_interpret_to_execute_after_opt_a]: 6.92002e-06
[rewriter_after_opt_a]: 2.272e-05
[convert_after_rewriter]: 5.57001e-06
[order_py_execute_after_rewriter]: 3.86999e-06
[mutable_eliminate]: 0.00043869
[jit_opt_b]: 4.302e-05, [1]
    [Cycle 1]: 3.787e-05, [2]
        [frontend_op_eliminate]: 1.453e-05
        [inline_after_opt_a]: 1.331e-05
[cconv]: 1.851e-05
[loop_unroll]: 0.00040223
[jit_opt_after_cconv]: 0.0001362, [1]
    [Cycle 1]: 0.00013002, [11]
        [c_1]: 2.016e-05
        [parameter_eliminate]: 1.61998e-06
        [updatestate_depend_eliminate]: 4.05998e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [cse]: 1.316e-05
        [call_graph_tuple_transform]: 1.778e-05
        [tuple_list_get_item_eliminator]: 5.37999e-06
        [none_parameter_eliminate]: 8.39995e-07
        [renormalize]: 4.00003e-07
        [switch_simplify]: 5.79e-06
[remove_dup_value]: 1.22e-05
[partial_unused_args_eliminate]: 2.06e-06
[environ_conv]: 4.58999e-06
[add_recomputation]: 4.437e-05
[cse_after_recomputation]: 1.934e-05, [1]
    [Cycle 1]: 1.434e-05, [1]
        [cse]: 8.64003e-06
[auto_monad_reorder]: 1.416e-05
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 4.75001e-06
[opt_after_jit_grad]: 0.00044378
[symbol_engine_optimizer]: 7e-05, [1]
    [Cycle 1]: 6.348e-05, [6]
        [build]: 3.38e-06
        [elim_shapecalc]: 6.94999e-06
        [elim_not_effective]: 1.223e-05
        [opt_reshape]: 5.47001e-06
        [fold_const_symbol]: 8.10999e-06
        [renormalize]: 6.90023e-07
[validate]: 3.112e-05
Sums
    bootstrap                            :     0.000297s :  3.03%
    type_inference                       :     0.006113s : 62.32%
    event_method                         :     0.000018s :  0.19%
    auto_monad                           :     0.000054s :  0.56%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000042s :  0.43%
    rewriter_before_opt_a                :     0.000074s :  0.75%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000115s :  1.17%
    jit_opt_a.loop_unroll                :     0.000039s :  0.40%
    jit_opt_a.a_1                        :     0.000703s :  7.17%
    jit_opt_a.with_stream_mark           :     0.000018s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000004s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000529s :  5.39%
    jit_opt_a.add_forward_monad_depend   :     0.000004s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000012s :  0.12%
    jit_opt_a.cse                        :     0.000022s :  0.23%
    jit_opt_a.replace_applicator         :     0.000014s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.07%
    rewriter_after_opt_a                 :     0.000023s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000439s :  4.47%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000013s :  0.14%
    cconv                                :     0.000019s :  0.19%
    loop_unroll                          :     0.000402s :  4.10%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000013s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000012s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000044s :  0.45%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000014s :  0.14%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000444s :  4.52%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000031s :  0.32%


Time group info:
------[substitution.]   0.000210    26
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.50% :     0.000005s :      3: substitution.graph_param_transform
84.17% :     0.000176s :      6: substitution.inline
 1.16% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.01% :     0.000004s :      4: substitution.remove_not_recompute_node
 0.94% :     0.000002s :      2: substitution.replace_old_param
 3.91% :     0.000008s :      1: substitution.switch_simplify
 3.89% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006064     2
83.69% :     0.005075s :      1: type_inference.infer
16.31% :     0.000989s :      1: type_inference.specialize
------[replace.]   0.000123     9
49.45% :     0.000061s :      6: replace.inline
26.19% :     0.000032s :      1: replace.switch_simplify
24.37% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000188     9
92.28% :     0.000173s :      6: match.inline
 4.01% :     0.000008s :      1: match.switch_simplify
 3.71% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   801
 1.41% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.05% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.19% :     0.000002s :     12: predicate.addn_check_dump
 1.32% :     0.000002s :     12: predicate.addn_zero_filter
 2.01% :     0.000003s :     12: predicate.arithmetic_simplify
 1.38% :     0.000002s :     12: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.17% :     0.000001s :     12: predicate.compare_switch_simplify
 1.32% :     0.000002s :     12: predicate.depend_value_elim
 1.19% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.46% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.22% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.33% :     0.000002s :     12: predicate.environ_get_eliminate
 1.37% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.50% :     0.000007s :     26: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     17: predicate.load_eliminater
 1.33% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.42% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000001s :     12: predicate.merge_addn
 1.18% :     0.000001s :     12: predicate.minmaximum_grad
 1.19% :     0.000001s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     17: predicate.partial_eliminate
 1.25% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     12: predicate.reduce_eliminate
 1.66% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     20: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.67% :     0.000002s :     12: predicate.reshape_eliminate
 1.41% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.79% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.77% :     0.000003s :     20: predicate.switch_defer_inline
 2.40% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.80% :     0.000011s :     65: predicate.switch_simplify
 1.29% :     0.000002s :     12: predicate.tile_eliminate
 1.33% :     0.000002s :     12: predicate.transpose_eliminate
 1.53% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.25% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.50% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000743    15
46.93% :     0.000349s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 9.59% :     0.000071s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.48% :     0.000323s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014948    72
 0.32% :     0.000047s :      1: add_recomputation
 0.39% :     0.000058s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 2.10% :     0.000314s :      1: bootstrap
 0.14% :     0.000021s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.14% :     0.000021s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
32.91% :     0.004919s :      1: jit_opt_a
 0.93% :     0.000139s :      1: jit_opt_after_cconv
 0.30% :     0.000045s :      1: jit_opt_b
 2.74% :     0.000409s :      1: loop_unroll
 2.98% :     0.000445s :      1: mutable_eliminate
 6.44% :     0.000962s :     26: opt.transform.jit_opt_a
 0.31% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000022s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000010s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.02% :     0.000451s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.30% :     0.000045s :      1: py_interpret_to_execute
 0.06% :     0.000009s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000014s :      1: remove_dup_value
 1.88% :     0.000281s :      1: renormalize.infer
 1.62% :     0.000242s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000026s :      1: rewriter_after_opt_a
 0.51% :     0.000077s :      1: rewriter_before_opt_a
 0.49% :     0.000073s :      1: symbol_engine_optimizer
40.95% :     0.006121s :      1: type_inference


TotalTime = 0.0174285, [33]
[bootstrap]: 0.0002625
[type_inference]: 0.00404735
[event_method]: 1.717e-05
[auto_monad]: 6.16e-05
[graph_reusing]: 5.45001e-06
[pre_auto_parallel]: 2.36998e-06
[py_interpret_to_execute]: 3.16e-05
[rewriter_before_opt_a]: 7.191e-05
[expand_dump_flag]: 3.06001e-06
[jit_opt_a]: 0.0049875, [2]
    [Cycle 1]: 0.00143454, [27]
        [switch_simplify]: 5.223e-05
        [loop_unroll]: 2.905e-05
        [a_1]: 0.00049339
        [with_stream_mark]: 1.431e-05
        [recompute_prepare]: 7.32002e-06
        [updatestate_depend_eliminate]: 3.68999e-06
        [updatestate_assign_eliminate]: 3.06001e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 2.09e-06
        [specialize_transform]: 5.79999e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.31002e-06
        [meta_shard_fg_expand]: 1.92001e-06
        [get_grad_eliminate_]: 5.26998e-06
        [merge_forward]: 3.62002e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.405e-05
        [j_node_and_user_rematch]: 9.11998e-06
        [meta_fg_expand]: 2.63e-06
        [replace_old_param]: 9.12999e-06
        [inline_without_move]: 5.47001e-06
        [renormalize]: 0.00053731
        [add_forward_monad_depend]: 4.55999e-06
        [auto_monad_grad]: 2.08002e-06
        [auto_monad_eliminator]: 1.316e-05
        [cse]: 2.807e-05
        [replace_applicator]: 1.046e-05
    [Cycle 2]: 0.00031477, [27]
        [switch_simplify]: 5.86998e-06
        [loop_unroll]: 4.72e-06
        [a_1]: 8.92e-05
        [with_stream_mark]: 8.57e-06
        [recompute_prepare]: 5.04998e-06
        [updatestate_depend_eliminate]: 2.51998e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 9.20001e-07
        [specialize_transform]: 5.02e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 4.68999e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.45999e-06
        [merge_forward]: 2.32999e-06
        [cell_reuse_recompute_pass]: 1.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.196e-05
        [j_node_and_user_rematch]: 7.73001e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 6.79001e-06
        [inline_without_move]: 4.65999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.05001e-06
        [auto_monad_grad]: 6.50005e-07
        [auto_monad_eliminator]: 4.79e-06
        [cse]: 8.97999e-06
        [replace_applicator]: 5.03002e-06
[py_interpret_to_execute_after_opt_a]: 9.87999e-06
[rewriter_after_opt_a]: 3.341e-05
[convert_after_rewriter]: 6.61999e-06
[order_py_execute_after_rewriter]: 5.31002e-06
[mutable_eliminate]: 0.00052087
[jit_opt_b]: 4.931e-05, [1]
    [Cycle 1]: 4.191e-05, [2]
        [frontend_op_eliminate]: 1.627e-05
        [inline_after_opt_a]: 1.438e-05
[cconv]: 2.33e-05
[loop_unroll]: 0.000403
[jit_opt_after_cconv]: 0.000136, [1]
    [Cycle 1]: 0.00012973, [11]
        [c_1]: 1.994e-05
        [parameter_eliminate]: 2.45002e-06
        [updatestate_depend_eliminate]: 5.38002e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.789e-05
        [call_graph_tuple_transform]: 1.868e-05
        [tuple_list_get_item_eliminator]: 5.12e-06
        [none_parameter_eliminate]: 1.39998e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 5.40001e-06
[remove_dup_value]: 1.249e-05
[partial_unused_args_eliminate]: 2.25002e-06
[environ_conv]: 4.72e-06
[add_recomputation]: 4.386e-05
[cse_after_recomputation]: 1.977e-05, [1]
    [Cycle 1]: 1.435e-05, [1]
        [cse]: 8.25999e-06
[auto_monad_reorder]: 1.454e-05
[get_jit_bprop_graph]: 1.55001e-06
[rewriter_after_jit_bprop_graph]: 4.67e-06
[opt_after_jit_grad]: 0.00044417
[symbol_engine_optimizer]: 7.044e-05, [1]
    [Cycle 1]: 6.342e-05, [6]
        [build]: 3.43999e-06
        [elim_shapecalc]: 7.02997e-06
        [elim_not_effective]: 1.205e-05
        [opt_reshape]: 5.54e-06
        [fold_const_symbol]: 8.27e-06
        [renormalize]: 6.30011e-07
[validate]: 3.135e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00591921
[execute]: 7.36001e-06
Sums
    bootstrap                            :     0.000263s :  1.93%
    type_inference                       :     0.004047s : 29.72%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000062s :  0.45%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.23%
    rewriter_before_opt_a                :     0.000072s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.43%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000583s :  4.28%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000537s :  3.95%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000037s :  0.27%
    jit_opt_a.replace_applicator         :     0.000015s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000521s :  3.82%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000403s :  2.96%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000044s :  0.32%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000444s :  3.26%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000031s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005919s : 43.46%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000175    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.91% :     0.000005s :      3: substitution.graph_param_transform
77.88% :     0.000136s :      4: substitution.inline
 1.67% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.26% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.29% :     0.000004s :      2: substitution.replace_old_param
11.47% :     0.000020s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003991     2
80.02% :     0.003194s :      1: type_inference.infer
19.98% :     0.000797s :      1: type_inference.specialize
------[replace.]   0.000061     6
64.68% :     0.000040s :      4: replace.inline
35.32% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000153     6
87.69% :     0.000134s :      4: match.inline
12.31% :     0.000019s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.43% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.16% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000000s :      3: predicate.graph_param_transform
 5.20% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.95% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.57% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000000s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     10: predicate.reduce_eliminate
 1.56% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.80% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.29% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.10% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.60% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.46% :     0.000010s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.98% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.86% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000534    12
44.37% :     0.000237s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.97% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.66% :     0.000271s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018855    76
 0.25% :     0.000047s :      1: add_recomputation
 0.34% :     0.000065s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.46% :     0.000276s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.47% :     0.004991s :      1: jit_opt_a
 0.73% :     0.000139s :      1: jit_opt_after_cconv
 0.28% :     0.000052s :      1: jit_opt_b
 2.18% :     0.000410s :      1: loop_unroll
 2.80% :     0.000529s :      1: mutable_eliminate
 4.16% :     0.000785s :     26: opt.transform.jit_opt_a
 0.24% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.39% :     0.000452s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.42% :     0.000268s :      1: renormalize.infer
 1.40% :     0.000263s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000037s :      1: rewriter_after_opt_a
 0.40% :     0.000075s :      1: rewriter_before_opt_a
 0.39% :     0.000073s :      1: symbol_engine_optimizer
31.45% :     0.005930s :      1: task_emit
21.52% :     0.004058s :      1: type_inference
 0.26% :     0.000049s :      1: validate


TotalTime = 0.0874058, [33]
[bootstrap]: 0.00033123
[type_inference]: 0.069595
[event_method]: 1.814e-05
[auto_monad]: 9.008e-05
[graph_reusing]: 5.33002e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 0.00027495
[rewriter_before_opt_a]: 8.223e-05
[expand_dump_flag]: 3.23e-06
[jit_opt_a]: 0.00710899, [2]
    [Cycle 1]: 0.00320688, [27]
        [switch_simplify]: 5.516e-05
        [loop_unroll]: 3.401e-05
        [a_1]: 0.00066408
        [with_stream_mark]: 1.653e-05
        [recompute_prepare]: 9.42001e-06
        [updatestate_depend_eliminate]: 3.97e-06
        [updatestate_assign_eliminate]: 2.79999e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 2.06e-06
        [specialize_transform]: 6.74001e-06
        [updatestate_useless_node_eliminater]: 6.48e-06
        [accelerated_algorithm]: 6.45997e-06
        [meta_shard_fg_expand]: 2.45002e-06
        [get_grad_eliminate_]: 6.16e-06
        [merge_forward]: 4.31002e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.634e-05
        [j_node_and_user_rematch]: 9.86e-06
        [meta_fg_expand]: 2.49001e-06
        [replace_old_param]: 9.60001e-06
        [inline_without_move]: 6.34001e-06
        [renormalize]: 0.00210241
        [add_forward_monad_depend]: 6.48e-06
        [auto_monad_grad]: 1.62999e-06
        [auto_monad_eliminator]: 1.594e-05
        [cse]: 2.279e-05
        [replace_applicator]: 1.481e-05
    [Cycle 2]: 0.00041254, [27]
        [switch_simplify]: 6.92997e-06
        [loop_unroll]: 6.43e-06
        [a_1]: 0.00015501
        [with_stream_mark]: 1.252e-05
        [recompute_prepare]: 6.58e-06
        [updatestate_depend_eliminate]: 3.51001e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 6.15002e-06
        [updatestate_useless_node_eliminater]: 6.29001e-06
        [accelerated_algorithm]: 6.14999e-06
        [meta_shard_fg_expand]: 1.36002e-06
        [get_grad_eliminate_]: 5.74e-06
        [merge_forward]: 3.36001e-06
        [cell_reuse_recompute_pass]: 2.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.369e-05
        [j_node_and_user_rematch]: 9.46e-06
        [meta_fg_expand]: 1.93002e-06
        [replace_old_param]: 8.35001e-06
        [inline_without_move]: 5.99e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.29998e-06
        [auto_monad_grad]: 1.04e-06
        [auto_monad_eliminator]: 6.64001e-06
        [cse]: 1.389e-05
        [replace_applicator]: 6.14999e-06
[py_interpret_to_execute_after_opt_a]: 9.75002e-06
[rewriter_after_opt_a]: 3.305e-05
[convert_after_rewriter]: 7.13e-06
[order_py_execute_after_rewriter]: 5.35999e-06
[mutable_eliminate]: 0.0006353
[jit_opt_b]: 5.836e-05, [1]
    [Cycle 1]: 5.125e-05, [2]
        [frontend_op_eliminate]: 2.056e-05
        [inline_after_opt_a]: 1.894e-05
[cconv]: 2.448e-05
[loop_unroll]: 0.00056453
[jit_opt_after_cconv]: 0.00017392, [1]
    [Cycle 1]: 0.00016732, [11]
        [c_1]: 3.134e-05
        [parameter_eliminate]: 3.14001e-06
        [updatestate_depend_eliminate]: 6.73e-06
        [updatestate_assign_eliminate]: 2.67001e-06
        [updatestate_loads_eliminate]: 2.35002e-06
        [cse]: 2.498e-05
        [call_graph_tuple_transform]: 2.264e-05
        [tuple_list_get_item_eliminator]: 7.43e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 1.07998e-06
        [switch_simplify]: 7.31001e-06
[remove_dup_value]: 1.669e-05
[partial_unused_args_eliminate]: 2.49001e-06
[environ_conv]: 8.65001e-06
[add_recomputation]: 4.972e-05
[cse_after_recomputation]: 2.47e-05, [1]
    [Cycle 1]: 1.878e-05, [1]
        [cse]: 1.209e-05
[auto_monad_reorder]: 1.865e-05
[get_jit_bprop_graph]: 1.55001e-06
[rewriter_after_jit_bprop_graph]: 5.71e-06
[opt_after_jit_grad]: 0.00057066
[symbol_engine_optimizer]: 8.93e-05, [1]
    [Cycle 1]: 8.151e-05, [6]
        [build]: 4.05e-06
        [elim_shapecalc]: 8.89e-06
        [elim_not_effective]: 1.632e-05
        [opt_reshape]: 8.40001e-06
        [fold_const_symbol]: 1.232e-05
        [renormalize]: 5.89993e-07
[validate]: 4.824e-05
[backend_pass]: 9.10019e-07
[task_emit]: 0.00726854
[execute]: 8.35999e-06
Sums
    bootstrap                            :     0.000331s :  0.40%
    type_inference                       :     0.069595s : 83.62%
    event_method                         :     0.000018s :  0.02%
    auto_monad                           :     0.000090s :  0.11%
    graph_reusing                        :     0.000005s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.00%
    py_interpret_to_execute              :     0.000275s :  0.33%
    rewriter_before_opt_a                :     0.000082s :  0.10%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000062s :  0.07%
    jit_opt_a.loop_unroll                :     0.000040s :  0.05%
    jit_opt_a.a_1                        :     0.000819s :  0.98%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.03%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.00%
    jit_opt_a.specialize_transform       :     0.000013s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000013s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.00%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.01%
    jit_opt_a.merge_forward              :     0.000008s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.00%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.04%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.02%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000018s :  0.02%
    jit_opt_a.inline_without_move        :     0.000012s :  0.01%
    jit_opt_a.renormalize                :     0.002102s :  2.53%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.03%
    jit_opt_a.cse                        :     0.000037s :  0.04%
    jit_opt_a.replace_applicator         :     0.000021s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.01%
    rewriter_after_opt_a                 :     0.000033s :  0.04%
    convert_after_rewriter               :     0.000007s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000635s :  0.76%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.02%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.02%
    cconv                                :     0.000024s :  0.03%
    loop_unroll                          :     0.000565s :  0.68%
    jit_opt_after_cconv.c_1              :     0.000031s :  0.04%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000025s :  0.03%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.01%
    remove_dup_value                     :     0.000017s :  0.02%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000009s :  0.01%
    add_recomputation                    :     0.000050s :  0.06%
    cse_after_recomputation.cse          :     0.000012s :  0.01%
    auto_monad_reorder                   :     0.000019s :  0.02%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000571s :  0.69%
    symbol_engine_optimizer.build        :     0.000004s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.02%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000048s :  0.06%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007269s :  8.73%
    execute                              :     0.000008s :  0.01%


Time group info:
------[substitution.]   0.000213    27
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.82% :     0.000002s :      2: substitution.fold_const_symbol
 2.97% :     0.000006s :      4: substitution.graph_param_transform
78.15% :     0.000166s :      4: substitution.inline
 1.67% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.24% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.93% :     0.000004s :      2: substitution.replace_old_param
 6.52% :     0.000014s :      3: substitution.tile_eliminate
 4.66% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.069519     2
97.20% :     0.067576s :      1: type_inference.infer
 2.80% :     0.001943s :      1: type_inference.specialize
------[replace.]   0.000074     6
66.81% :     0.000049s :      4: replace.inline
33.19% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000172     6
94.96% :     0.000163s :      4: match.inline
 5.04% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000159   834
 1.17% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.24% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.00% :     0.000002s :     12: predicate.addn_check_dump
 1.15% :     0.000002s :     12: predicate.addn_zero_filter
 1.66% :     0.000003s :     12: predicate.arithmetic_simplify
 1.00% :     0.000002s :     12: predicate.cast_eliminate
 0.35% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.27% :     0.000002s :     12: predicate.compare_switch_simplify
 1.16% :     0.000002s :     12: predicate.depend_value_elim
 1.09% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.03% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.32% :     0.000001s :      4: predicate.elim_not_effective
 0.59% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 0.99% :     0.000002s :     12: predicate.environ_add_const_eliminate
 0.92% :     0.000001s :     12: predicate.environ_get_add_eliminate
 0.90% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.11% :     0.000002s :     12: predicate.environ_get_eliminate
 0.90% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      4: predicate.fold_const_symbol
 0.92% :     0.000001s :      8: predicate.get_grad_eliminate
 0.28% :     0.000000s :      4: predicate.graph_param_transform
 4.44% :     0.000007s :     26: predicate.inline
 0.82% :     0.000001s :      8: predicate.inline_without_move
 0.40% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.09% :     0.000002s :      8: predicate.less_batch_normalization
 1.28% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.63% :     0.000003s :     18: predicate.load_eliminater
 1.49% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.67% :     0.000006s :     38: predicate.loop_unroll_before_grad
 1.62% :     0.000003s :     16: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000002s :     12: predicate.merge_addn
 1.11% :     0.000002s :     12: predicate.minmaximum_grad
 1.24% :     0.000002s :      4: predicate.mutable_eliminate
 0.50% :     0.000001s :      4: predicate.opt_reshape
 1.86% :     0.000003s :     18: predicate.partial_eliminate
 1.11% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.74% :     0.000003s :     12: predicate.reduce_eliminate
 1.22% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.65% :     0.000003s :     22: predicate.replace_applicator
 0.52% :     0.000001s :      8: predicate.replace_old_param
 0.40% :     0.000001s :      4: predicate.reset_defer_inline
 1.35% :     0.000002s :     12: predicate.reshape_eliminate
 1.62% :     0.000003s :     12: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     12: predicate.same_eliminate
 0.62% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.03% :     0.000002s :      8: predicate.special_op_eliminate
 0.86% :     0.000001s :      8: predicate.specialize_transform
13.10% :     0.000021s :     12: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.15% :     0.000003s :     18: predicate.switch_defer_inline
 1.77% :     0.000003s :     18: predicate.switch_layer_defer_inline
 6.14% :     0.000010s :     60: predicate.switch_simplify
 1.51% :     0.000002s :     12: predicate.tile_eliminate
 0.97% :     0.000002s :     12: predicate.transpose_eliminate
 1.19% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 2.96% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.31% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.71% :     0.000003s :     18: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000004s :     26: predicate.updatestate_useless_node_eliminater
 1.31% :     0.000002s :     12: predicate.value_based_eliminate
 0.36% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.53% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.002369    26
70.54% :     0.001671s :     19: func_graph_cloner_run.FuncGraphClonerGraph
 1.17% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
28.29% :     0.000670s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.090661    76
 0.06% :     0.000053s :      1: add_recomputation
 0.10% :     0.000095s :      1: auto_monad
 0.03% :     0.000023s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.38% :     0.000346s :      1: bootstrap
 0.03% :     0.000027s :      1: cconv
 0.01% :     0.000010s :      1: convert_after_rewriter
 0.03% :     0.000027s :      1: cse_after_recomputation
 0.01% :     0.000011s :      1: environ_conv
 0.02% :     0.000022s :      1: event_method
 0.01% :     0.000012s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.00% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000008s :      1: graph_reusing
 7.85% :     0.007112s :      1: jit_opt_a
 0.20% :     0.000177s :      1: jit_opt_after_cconv
 0.07% :     0.000061s :      1: jit_opt_b
 0.63% :     0.000573s :      1: loop_unroll
 0.71% :     0.000645s :      1: mutable_eliminate
 1.17% :     0.001058s :     26: opt.transform.jit_opt_a
 0.07% :     0.000064s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000032s :      4: opt.transform.jit_opt_b
 0.02% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.05% :     0.000042s :      4: opt.transform.symbol_engine_opt
 0.64% :     0.000579s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000281s :      1: py_interpret_to_execute
 0.01% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000020s :      1: remove_dup_value
 1.60% :     0.001452s :      1: renormalize.infer
 0.71% :     0.000640s :      1: renormalize.specialize
 0.01% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.04% :     0.000036s :      1: rewriter_after_opt_a
 0.10% :     0.000087s :      1: rewriter_before_opt_a
 0.10% :     0.000092s :      1: symbol_engine_optimizer
 8.03% :     0.007282s :      1: task_emit
76.78% :     0.069607s :      1: type_inference
 0.08% :     0.000070s :      1: validate


TotalTime = 0.0185228, [33]
[bootstrap]: 0.00027479
[type_inference]: 0.00437645
[event_method]: 1.539e-05
[auto_monad]: 5.663e-05
[graph_reusing]: 4.64998e-06
[pre_auto_parallel]: 1.73002e-06
[py_interpret_to_execute]: 3.11e-05
[rewriter_before_opt_a]: 6.706e-05
[expand_dump_flag]: 2.34001e-06
[jit_opt_a]: 0.00547847, [2]
    [Cycle 1]: 0.0015823, [27]
        [switch_simplify]: 4.701e-05
        [loop_unroll]: 3.075e-05
        [a_1]: 0.00052125
        [with_stream_mark]: 1.496e-05
        [recompute_prepare]: 7.59002e-06
        [updatestate_depend_eliminate]: 2.98e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [parameter_eliminate]: 1.71002e-06
        [specialize_transform]: 5.86998e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 5.40999e-06
        [meta_shard_fg_expand]: 1.79998e-06
        [get_grad_eliminate_]: 4.97e-06
        [merge_forward]: 3.02002e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.312e-05
        [j_node_and_user_rematch]: 9.05001e-06
        [meta_fg_expand]: 2.32001e-06
        [replace_old_param]: 8.40999e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.00059945
        [add_forward_monad_depend]: 4.72e-06
        [auto_monad_grad]: 2.27999e-06
        [auto_monad_eliminator]: 1.769e-05
        [cse]: 2.359e-05
        [replace_applicator]: 1.134e-05
    [Cycle 2]: 0.00036836, [27]
        [switch_simplify]: 5.73997e-06
        [loop_unroll]: 5.66003e-06
        [a_1]: 9.147e-05
        [with_stream_mark]: 9.20001e-06
        [recompute_prepare]: 4.90999e-06
        [updatestate_depend_eliminate]: 2.51e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 4.86002e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 1.24e-06
        [get_grad_eliminate_]: 5.54e-06
        [merge_forward]: 2.90002e-06
        [cell_reuse_recompute_pass]: 2.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.249e-05
        [j_node_and_user_rematch]: 8.33999e-06
        [meta_fg_expand]: 1.55001e-06
        [replace_old_param]: 1.816e-05
        [inline_without_move]: 1.269e-05
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.47999e-06
        [auto_monad_grad]: 9.5999e-07
        [auto_monad_eliminator]: 6.09999e-06
        [cse]: 1.213e-05
        [replace_applicator]: 5.99e-06
[py_interpret_to_execute_after_opt_a]: 1.16e-05
[rewriter_after_opt_a]: 3.594e-05
[convert_after_rewriter]: 1.065e-05
[order_py_execute_after_rewriter]: 6.29001e-06
[mutable_eliminate]: 0.00055322
[jit_opt_b]: 5.427e-05, [1]
    [Cycle 1]: 4.74e-05, [2]
        [frontend_op_eliminate]: 1.779e-05
        [inline_after_opt_a]: 1.676e-05
[cconv]: 2.193e-05
[loop_unroll]: 0.00045068
[jit_opt_after_cconv]: 0.00014566, [1]
    [Cycle 1]: 0.00013939, [11]
        [c_1]: 2.198e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 5.54e-06
        [updatestate_assign_eliminate]: 2.58003e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [cse]: 1.714e-05
        [call_graph_tuple_transform]: 2.061e-05
        [tuple_list_get_item_eliminator]: 6.14001e-06
        [none_parameter_eliminate]: 1.10999e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 5.89e-06
[remove_dup_value]: 1.154e-05
[partial_unused_args_eliminate]: 2.14999e-06
[environ_conv]: 1.481e-05
[add_recomputation]: 4.404e-05
[cse_after_recomputation]: 2.141e-05, [1]
    [Cycle 1]: 1.571e-05, [1]
        [cse]: 9.58002e-06
[auto_monad_reorder]: 1.399e-05
[get_jit_bprop_graph]: 1.92999e-06
[rewriter_after_jit_bprop_graph]: 4.99e-06
[opt_after_jit_grad]: 0.00047867
[symbol_engine_optimizer]: 7.157e-05, [1]
    [Cycle 1]: 6.543e-05, [6]
        [build]: 2.79001e-06
        [elim_shapecalc]: 7.07002e-06
        [elim_not_effective]: 1.231e-05
        [opt_reshape]: 6.38e-06
        [fold_const_symbol]: 8.80999e-06
        [renormalize]: 4.70027e-07
[validate]: 3.087e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00603745
[execute]: 5.57001e-06
Sums
    bootstrap                            :     0.000275s :  1.92%
    type_inference                       :     0.004376s : 30.56%
    event_method                         :     0.000015s :  0.11%
    auto_monad                           :     0.000057s :  0.40%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000031s :  0.22%
    rewriter_before_opt_a                :     0.000067s :  0.47%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.37%
    jit_opt_a.loop_unroll                :     0.000036s :  0.25%
    jit_opt_a.a_1                        :     0.000613s :  4.28%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000027s :  0.19%
    jit_opt_a.inline_without_move        :     0.000018s :  0.13%
    jit_opt_a.renormalize                :     0.000600s :  4.19%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.17%
    jit_opt_a.cse                        :     0.000036s :  0.25%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000036s :  0.25%
    convert_after_rewriter               :     0.000011s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000553s :  3.86%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.12%
    cconv                                :     0.000022s :  0.15%
    loop_unroll                          :     0.000451s :  3.15%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000012s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000015s :  0.10%
    add_recomputation                    :     0.000044s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000479s :  3.34%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006037s : 42.15%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000193    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.93% :     0.000006s :      3: substitution.graph_param_transform
83.89% :     0.000162s :      4: substitution.inline
 1.70% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.11% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.80% :     0.000003s :      2: substitution.replace_old_param
 5.97% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004323     2
81.90% :     0.003541s :      1: type_inference.infer
18.10% :     0.000782s :      1: type_inference.specialize
------[replace.]   0.000067     6
66.26% :     0.000044s :      4: replace.inline
33.74% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000170     6
93.92% :     0.000159s :      4: match.inline
 6.08% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.18% :     0.000003s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.55% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000001s :     10: predicate.compare_switch_simplify
 1.41% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.72% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.36% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.62% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.13% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.22% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.58% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     18: predicate.replace_applicator
 0.85% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.57% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.47% :     0.000009s :     53: predicate.switch_simplify
 1.09% :     0.000001s :     10: predicate.tile_eliminate
 1.31% :     0.000002s :     10: predicate.transpose_eliminate
 1.56% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.41% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000612    12
54.29% :     0.000332s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.53% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.18% :     0.000246s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020070    76
 0.24% :     0.000047s :      1: add_recomputation
 0.30% :     0.000060s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.44% :     0.000288s :      1: bootstrap
 0.12% :     0.000025s :      1: cconv
 0.07% :     0.000013s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.09% :     0.000018s :      1: environ_conv
 0.09% :     0.000019s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
27.31% :     0.005482s :      1: jit_opt_a
 0.74% :     0.000149s :      1: jit_opt_after_cconv
 0.28% :     0.000057s :      1: jit_opt_b
 2.28% :     0.000458s :      1: loop_unroll
 2.80% :     0.000562s :      1: mutable_eliminate
 4.16% :     0.000835s :     26: opt.transform.jit_opt_a
 0.25% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.42% :     0.000486s :      1: opt_after_jit_grad
 0.04% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000034s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000014s :      1: remove_dup_value
 1.46% :     0.000294s :      1: renormalize.infer
 1.49% :     0.000299s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000039s :      1: rewriter_after_opt_a
 0.35% :     0.000070s :      1: rewriter_before_opt_a
 0.37% :     0.000074s :      1: symbol_engine_optimizer
30.14% :     0.006049s :      1: task_emit
21.85% :     0.004386s :      1: type_inference
 0.24% :     0.000048s :      1: validate


TotalTime = 0.015467, [30]
[bootstrap]: 0.0003306
[type_inference]: 0.00732797
[event_method]: 2.148e-05
[auto_monad]: 8.397e-05
[graph_reusing]: 6.10002e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 5.072e-05
[rewriter_before_opt_a]: 8.577e-05
[expand_dump_flag]: 3.02002e-06
[jit_opt_a]: 0.00552488, [2]
    [Cycle 1]: 0.00181764, [27]
        [switch_simplify]: 0.00010315
        [loop_unroll]: 3.451e-05
        [a_1]: 0.00071199
        [with_stream_mark]: 1.746e-05
        [recompute_prepare]: 8.33001e-06
        [updatestate_depend_eliminate]: 3.43e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.62999e-06
        [specialize_transform]: 6.43998e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 1.43002e-06
        [get_grad_eliminate_]: 5.76998e-06
        [merge_forward]: 2.89999e-06
        [cell_reuse_recompute_pass]: 9.79984e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.192e-05
        [j_node_and_user_rematch]: 1.031e-05
        [meta_fg_expand]: 2.16998e-06
        [replace_old_param]: 8.38001e-06
        [inline_without_move]: 5.54e-06
        [renormalize]: 0.00065021
        [add_forward_monad_depend]: 4.89e-06
        [auto_monad_grad]: 1.99999e-06
        [auto_monad_eliminator]: 1.187e-05
        [cse]: 1.816e-05
        [replace_applicator]: 1.038e-05
    [Cycle 2]: 0.00031499, [27]
        [switch_simplify]: 5.44e-06
        [loop_unroll]: 5.02e-06
        [a_1]: 8.783e-05
        [with_stream_mark]: 8.35001e-06
        [recompute_prepare]: 4.80001e-06
        [updatestate_depend_eliminate]: 2.56e-06
        [updatestate_assign_eliminate]: 2.10002e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 5.03002e-06
        [updatestate_useless_node_eliminater]: 4.81002e-06
        [accelerated_algorithm]: 4.86002e-06
        [meta_shard_fg_expand]: 1.05001e-06
        [get_grad_eliminate_]: 4.33999e-06
        [merge_forward]: 2.56e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.233e-05
        [j_node_and_user_rematch]: 8e-06
        [meta_fg_expand]: 1.52999e-06
        [replace_old_param]: 7.2e-06
        [inline_without_move]: 4.59998e-06
        [renormalize]: 3.99887e-08
        [add_forward_monad_depend]: 1.23002e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 4.77e-06
        [cse]: 9.77999e-06
        [replace_applicator]: 4.96002e-06
[py_interpret_to_execute_after_opt_a]: 8.42998e-06
[rewriter_after_opt_a]: 2.849e-05
[convert_after_rewriter]: 5.76e-06
[order_py_execute_after_rewriter]: 5.29e-06
[mutable_eliminate]: 0.00056177
[jit_opt_b]: 4.914e-05, [1]
    [Cycle 1]: 4.22e-05, [2]
        [frontend_op_eliminate]: 1.598e-05
        [inline_after_opt_a]: 1.484e-05
[cconv]: 2.186e-05
[loop_unroll]: 0.00041743
[jit_opt_after_cconv]: 0.00013159, [1]
    [Cycle 1]: 0.00012568, [11]
        [c_1]: 2.051e-05
        [parameter_eliminate]: 1.92001e-06
        [updatestate_depend_eliminate]: 5.07999e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.601e-05
        [call_graph_tuple_transform]: 1.688e-05
        [tuple_list_get_item_eliminator]: 5.19e-06
        [none_parameter_eliminate]: 1.05001e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 5.40999e-06
[remove_dup_value]: 1.067e-05
[partial_unused_args_eliminate]: 1.92999e-06
[environ_conv]: 4.13001e-06
[add_recomputation]: 4.072e-05
[cse_after_recomputation]: 1.959e-05, [1]
    [Cycle 1]: 1.452e-05, [1]
        [cse]: 8.59998e-06
[auto_monad_reorder]: 1.354e-05
[get_jit_bprop_graph]: 1.69e-06
[rewriter_after_jit_bprop_graph]: 4.55001e-06
[opt_after_jit_grad]: 0.00044102
[symbol_engine_optimizer]: 7.07e-05, [1]
    [Cycle 1]: 6.47e-05, [6]
        [build]: 3.17002e-06
        [elim_shapecalc]: 7.34002e-06
        [elim_not_effective]: 1.205e-05
        [opt_reshape]: 5.54e-06
        [fold_const_symbol]: 8.87e-06
        [renormalize]: 4.80009e-07
[validate]: 3.117e-05
Sums
    bootstrap                            :     0.000331s :  2.87%
    type_inference                       :     0.007328s : 63.66%
    event_method                         :     0.000021s :  0.19%
    auto_monad                           :     0.000084s :  0.73%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000051s :  0.44%
    rewriter_before_opt_a                :     0.000086s :  0.75%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000109s :  0.94%
    jit_opt_a.loop_unroll                :     0.000040s :  0.34%
    jit_opt_a.a_1                        :     0.000800s :  6.95%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000650s :  5.65%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.14%
    jit_opt_a.cse                        :     0.000028s :  0.24%
    jit_opt_a.replace_applicator         :     0.000015s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.07%
    rewriter_after_opt_a                 :     0.000028s :  0.25%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000562s :  4.88%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.13%
    cconv                                :     0.000022s :  0.19%
    loop_unroll                          :     0.000417s :  3.63%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000011s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000041s :  0.35%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000441s :  3.83%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.27%


Time group info:
------[substitution.]   0.000248    26
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 1.84% :     0.000005s :      3: substitution.graph_param_transform
84.66% :     0.000210s :      6: substitution.inline
 1.33% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.73% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.22% :     0.000003s :      2: substitution.replace_old_param
 3.06% :     0.000008s :      1: substitution.switch_simplify
 4.89% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007260     2
84.24% :     0.006116s :      1: type_inference.infer
15.76% :     0.001144s :      1: type_inference.specialize
------[replace.]   0.000131     9
50.15% :     0.000066s :      6: replace.inline
23.89% :     0.000031s :      1: replace.switch_simplify
25.96% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000225     9
92.17% :     0.000207s :      6: match.inline
 3.02% :     0.000007s :      1: match.switch_simplify
 4.81% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   801
 1.60% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.25% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.19% :     0.000002s :     12: predicate.addn_check_dump
 1.39% :     0.000002s :     12: predicate.addn_zero_filter
 2.00% :     0.000003s :     12: predicate.arithmetic_simplify
 1.31% :     0.000002s :     12: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.18% :     0.000002s :     12: predicate.compare_switch_simplify
 1.78% :     0.000002s :     12: predicate.depend_value_elim
 1.19% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.24% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     12: predicate.environ_get_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.31% :     0.000007s :     26: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.99% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.88% :     0.000002s :     17: predicate.load_eliminater
 1.01% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.58% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.91% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.37% :     0.000002s :     12: predicate.merge_addn
 1.15% :     0.000002s :     12: predicate.minmaximum_grad
 1.20% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     17: predicate.partial_eliminate
 1.25% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     12: predicate.reduce_eliminate
 1.50% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     20: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.68% :     0.000002s :     12: predicate.reshape_eliminate
 1.41% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     12: predicate.same_eliminate
 0.91% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.81% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.67% :     0.000004s :     20: predicate.switch_defer_inline
 2.37% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.10% :     0.000011s :     65: predicate.switch_simplify
 1.21% :     0.000002s :     12: predicate.tile_eliminate
 1.22% :     0.000002s :     12: predicate.transpose_eliminate
 1.50% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.44% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.11% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     12: predicate.value_based_eliminate
 0.41% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000812    15
48.81% :     0.000396s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 9.79% :     0.000079s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.40% :     0.000336s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017235    72
 0.25% :     0.000044s :      1: add_recomputation
 0.51% :     0.000088s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 2.00% :     0.000345s :      1: bootstrap
 0.14% :     0.000025s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
32.08% :     0.005528s :      1: jit_opt_a
 0.78% :     0.000134s :      1: jit_opt_after_cconv
 0.30% :     0.000052s :      1: jit_opt_b
 2.46% :     0.000424s :      1: loop_unroll
 3.31% :     0.000570s :      1: mutable_eliminate
 6.16% :     0.001062s :     26: opt.transform.jit_opt_a
 0.26% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.60% :     0.000449s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.31% :     0.000054s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000013s :      1: remove_dup_value
 2.00% :     0.000344s :      1: renormalize.infer
 1.73% :     0.000299s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000031s :      1: rewriter_after_opt_a
 0.51% :     0.000089s :      1: rewriter_before_opt_a
 0.42% :     0.000073s :      1: symbol_engine_optimizer
42.59% :     0.007340s :      1: type_inference


TotalTime = 0.0191426, [33]
[bootstrap]: 0.00028918
[type_inference]: 0.00480772
[event_method]: 1.919e-05
[auto_monad]: 6.806e-05
[graph_reusing]: 7.03998e-06
[pre_auto_parallel]: 2.78e-06
[py_interpret_to_execute]: 3.325e-05
[rewriter_before_opt_a]: 7.773e-05
[expand_dump_flag]: 3.78999e-06
[jit_opt_a]: 0.00565391, [2]
    [Cycle 1]: 0.00159346, [27]
        [switch_simplify]: 5.424e-05
        [loop_unroll]: 3.034e-05
        [a_1]: 0.00053497
        [with_stream_mark]: 1.753e-05
        [recompute_prepare]: 8.11002e-06
        [updatestate_depend_eliminate]: 1.728e-05
        [updatestate_assign_eliminate]: 3.58999e-06
        [updatestate_loads_eliminate]: 2.94999e-06
        [parameter_eliminate]: 2.21e-06
        [specialize_transform]: 7.05e-06
        [updatestate_useless_node_eliminater]: 5.35001e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 2.27999e-06
        [get_grad_eliminate_]: 5.19998e-06
        [merge_forward]: 3.98001e-06
        [cell_reuse_recompute_pass]: 1.33002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.59e-05
        [j_node_and_user_rematch]: 1.465e-05
        [meta_fg_expand]: 2.53e-06
        [replace_old_param]: 9.04e-06
        [inline_without_move]: 5.59e-06
        [renormalize]: 0.00060044
        [add_forward_monad_depend]: 5.60001e-06
        [auto_monad_grad]: 2.59999e-06
        [auto_monad_eliminator]: 1.399e-05
        [cse]: 2.637e-05
        [replace_applicator]: 1.323e-05
    [Cycle 2]: 0.0003239, [27]
        [switch_simplify]: 5.44998e-06
        [loop_unroll]: 4.72e-06
        [a_1]: 8.966e-05
        [with_stream_mark]: 9.99999e-06
        [recompute_prepare]: 5.25999e-06
        [updatestate_depend_eliminate]: 2.81999e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [parameter_eliminate]: 1.07998e-06
        [specialize_transform]: 5.19998e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.27999e-06
        [meta_shard_fg_expand]: 1.20001e-06
        [get_grad_eliminate_]: 4.77e-06
        [merge_forward]: 2.73e-06
        [cell_reuse_recompute_pass]: 1.96998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.244e-05
        [j_node_and_user_rematch]: 8.06001e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 7.88999e-06
        [inline_without_move]: 4.77e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 5.05999e-06
        [cse]: 1.046e-05
        [replace_applicator]: 5.09e-06
[py_interpret_to_execute_after_opt_a]: 8.76002e-06
[rewriter_after_opt_a]: 3.226e-05
[convert_after_rewriter]: 7.08e-06
[order_py_execute_after_rewriter]: 5.01002e-06
[mutable_eliminate]: 0.00060179
[jit_opt_b]: 5.224e-05, [1]
    [Cycle 1]: 4.539e-05, [2]
        [frontend_op_eliminate]: 1.764e-05
        [inline_after_opt_a]: 1.597e-05
[cconv]: 2.563e-05
[loop_unroll]: 0.00042286
[jit_opt_after_cconv]: 0.00013628, [1]
    [Cycle 1]: 0.00012963, [11]
        [c_1]: 1.961e-05
        [parameter_eliminate]: 2.74999e-06
        [updatestate_depend_eliminate]: 5.68002e-06
        [updatestate_assign_eliminate]: 2.36998e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [cse]: 1.787e-05
        [call_graph_tuple_transform]: 1.824e-05
        [tuple_list_get_item_eliminator]: 5.22e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 5.49975e-07
        [switch_simplify]: 5.25999e-06
[remove_dup_value]: 1.338e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 4.93001e-06
[add_recomputation]: 4.939e-05
[cse_after_recomputation]: 1.94e-05, [1]
    [Cycle 1]: 1.393e-05, [1]
        [cse]: 8.48001e-06
[auto_monad_reorder]: 1.503e-05
[get_jit_bprop_graph]: 1.64e-06
[rewriter_after_jit_bprop_graph]: 5.14e-06
[opt_after_jit_grad]: 0.00045235
[symbol_engine_optimizer]: 7.209e-05, [1]
    [Cycle 1]: 6.579e-05, [6]
        [build]: 3.72002e-06
        [elim_shapecalc]: 7.41999e-06
        [elim_not_effective]: 1.294e-05
        [opt_reshape]: 5.51e-06
        [fold_const_symbol]: 8.52998e-06
        [renormalize]: 5.19998e-07
[validate]: 3.537e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00602125
[execute]: 7.01999e-06
Sums
    bootstrap                            :     0.000289s :  1.95%
    type_inference                       :     0.004808s : 32.48%
    event_method                         :     0.000019s :  0.13%
    auto_monad                           :     0.000068s :  0.46%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.22%
    rewriter_before_opt_a                :     0.000078s :  0.53%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000060s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000625s :  4.22%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000020s :  0.14%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000023s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000601s :  4.06%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000037s :  0.25%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000032s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000602s :  4.07%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000026s :  0.17%
    loop_unroll                          :     0.000423s :  2.86%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.33%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000452s :  3.06%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006021s : 40.68%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000200    23
 1.06% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.60% :     0.000005s :      3: substitution.graph_param_transform
83.54% :     0.000167s :      4: substitution.inline
 1.42% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.33% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000004s :      2: substitution.replace_old_param
 6.32% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004735     2
82.12% :     0.003888s :      1: type_inference.infer
17.88% :     0.000847s :      1: type_inference.specialize
------[replace.]   0.000066     6
66.92% :     0.000044s :      4: replace.inline
33.08% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000176     6
93.50% :     0.000165s :      4: match.inline
 6.50% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.29% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.11% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.43% :     0.000002s :     10: predicate.addn_zero_filter
 1.87% :     0.000002s :     10: predicate.arithmetic_simplify
 1.19% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.50% :     0.000002s :     10: predicate.environ_get_eliminate
 1.31% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.29% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.64% :     0.000007s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.28% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.50% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.56% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.06% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.79% :     0.000002s :     10: predicate.reshape_eliminate
 1.39% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000001s :     10: predicate.same_eliminate
 0.62% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.61% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.77% :     0.000009s :     53: predicate.switch_simplify
 1.31% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.85% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.88% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000570    12
49.77% :     0.000284s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.13% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.09% :     0.000251s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020692    76
 0.25% :     0.000053s :      1: add_recomputation
 0.35% :     0.000072s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.47% :     0.000303s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000023s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000010s :      1: graph_reusing
27.34% :     0.005657s :      1: jit_opt_a
 0.67% :     0.000139s :      1: jit_opt_after_cconv
 0.27% :     0.000055s :      1: jit_opt_b
 2.08% :     0.000430s :      1: loop_unroll
 2.95% :     0.000610s :      1: mutable_eliminate
 4.07% :     0.000843s :     26: opt.transform.jit_opt_a
 0.22% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.22% :     0.000460s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000036s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.55% :     0.000320s :      1: renormalize.infer
 1.31% :     0.000272s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000035s :      1: rewriter_after_opt_a
 0.39% :     0.000080s :      1: rewriter_before_opt_a
 0.36% :     0.000074s :      1: symbol_engine_optimizer
29.15% :     0.006032s :      1: task_emit
23.32% :     0.004825s :      1: type_inference
 0.27% :     0.000056s :      1: validate


TotalTime = 0.0179497, [33]
[bootstrap]: 0.0002814
[type_inference]: 0.00414667
[event_method]: 1.619e-05
[auto_monad]: 6.056e-05
[graph_reusing]: 4.53001e-06
[pre_auto_parallel]: 1.97999e-06
[py_interpret_to_execute]: 3.053e-05
[rewriter_before_opt_a]: 7.388e-05
[expand_dump_flag]: 2.89001e-06
[jit_opt_a]: 0.0049998, [2]
    [Cycle 1]: 0.00141812, [27]
        [switch_simplify]: 4.751e-05
        [loop_unroll]: 2.989e-05
        [a_1]: 0.00050093
        [with_stream_mark]: 1.395e-05
        [recompute_prepare]: 7.38e-06
        [updatestate_depend_eliminate]: 3.68e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [parameter_eliminate]: 1.21002e-06
        [specialize_transform]: 5.83002e-06
        [updatestate_useless_node_eliminater]: 5.30999e-06
        [accelerated_algorithm]: 5.60001e-06
        [meta_shard_fg_expand]: 1.84998e-06
        [get_grad_eliminate_]: 5.46002e-06
        [merge_forward]: 3.10002e-06
        [cell_reuse_recompute_pass]: 8.80013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.176e-05
        [j_node_and_user_rematch]: 8.87e-06
        [meta_fg_expand]: 2.46e-06
        [replace_old_param]: 8.08001e-06
        [inline_without_move]: 5.19998e-06
        [renormalize]: 0.00054012
        [add_forward_monad_depend]: 4.63001e-06
        [auto_monad_grad]: 1.89999e-06
        [auto_monad_eliminator]: 1.063e-05
        [cse]: 1.797e-05
        [replace_applicator]: 1.247e-05
    [Cycle 2]: 0.00032226, [27]
        [switch_simplify]: 5.57999e-06
        [loop_unroll]: 5.14003e-06
        [a_1]: 9.04e-05
        [with_stream_mark]: 9.08002e-06
        [recompute_prepare]: 4.90999e-06
        [updatestate_depend_eliminate]: 2.53998e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.77999e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.75001e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 4.65001e-06
        [merge_forward]: 2.41998e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.221e-05
        [j_node_and_user_rematch]: 7.72998e-06
        [meta_fg_expand]: 1.77999e-06
        [replace_old_param]: 7.53999e-06
        [inline_without_move]: 4.68001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 7.79983e-07
        [auto_monad_eliminator]: 5.12e-06
        [cse]: 1.042e-05
        [replace_applicator]: 4.83001e-06
[py_interpret_to_execute_after_opt_a]: 9.69e-06
[rewriter_after_opt_a]: 3.007e-05
[convert_after_rewriter]: 6.27001e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.00050854
[jit_opt_b]: 4.933e-05, [1]
    [Cycle 1]: 4.219e-05, [2]
        [frontend_op_eliminate]: 1.644e-05
        [inline_after_opt_a]: 1.437e-05
[cconv]: 2.164e-05
[loop_unroll]: 0.00047978
[jit_opt_after_cconv]: 0.00013663, [1]
    [Cycle 1]: 0.00013016, [11]
        [c_1]: 2.097e-05
        [parameter_eliminate]: 2.64001e-06
        [updatestate_depend_eliminate]: 5.51e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [cse]: 1.711e-05
        [call_graph_tuple_transform]: 1.755e-05
        [tuple_list_get_item_eliminator]: 5.62999e-06
        [none_parameter_eliminate]: 1.22999e-06
        [renormalize]: 3.9002e-07
        [switch_simplify]: 5.32001e-06
[remove_dup_value]: 1.157e-05
[partial_unused_args_eliminate]: 1.86998e-06
[environ_conv]: 4.83001e-06
[add_recomputation]: 4.068e-05
[cse_after_recomputation]: 2.001e-05, [1]
    [Cycle 1]: 1.501e-05, [1]
        [cse]: 9.31e-06
[auto_monad_reorder]: 1.175e-05
[get_jit_bprop_graph]: 1.46998e-06
[rewriter_after_jit_bprop_graph]: 5.35001e-06
[opt_after_jit_grad]: 0.00047918
[symbol_engine_optimizer]: 7.49e-05, [1]
    [Cycle 1]: 6.814e-05, [6]
        [build]: 3.80998e-06
        [elim_shapecalc]: 7.51999e-06
        [elim_not_effective]: 1.322e-05
        [opt_reshape]: 5.64998e-06
        [fold_const_symbol]: 9.00999e-06
        [renormalize]: 3.39991e-07
[validate]: 3.301e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00620837
[execute]: 7.93001e-06
Sums
    bootstrap                            :     0.000281s :  1.99%
    type_inference                       :     0.004147s : 29.38%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000061s :  0.43%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000031s :  0.22%
    rewriter_before_opt_a                :     0.000074s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.38%
    jit_opt_a.loop_unroll                :     0.000035s :  0.25%
    jit_opt_a.a_1                        :     0.000591s :  4.19%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000540s :  3.83%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.11%
    jit_opt_a.cse                        :     0.000028s :  0.20%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000030s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000509s :  3.60%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000022s :  0.15%
    loop_unroll                          :     0.000480s :  3.40%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000041s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000012s :  0.08%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000479s :  3.40%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006208s : 43.99%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000175    23
 1.20% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.63% :     0.000005s :      3: substitution.graph_param_transform
83.59% :     0.000146s :      4: substitution.inline
 1.69% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.37% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.94% :     0.000003s :      2: substitution.replace_old_param
 5.91% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004092     2
81.37% :     0.003330s :      1: type_inference.infer
18.63% :     0.000762s :      1: type_inference.specialize
------[replace.]   0.000065     6
67.36% :     0.000044s :      4: replace.inline
32.64% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000153     6
93.99% :     0.000144s :      4: match.inline
 6.01% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.51% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.17% :     0.000001s :     10: predicate.addn_check_dump
 1.49% :     0.000002s :     10: predicate.addn_zero_filter
 2.33% :     0.000003s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.61% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     10: predicate.compare_switch_simplify
 1.13% :     0.000001s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000001s :     10: predicate.environ_get_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.05% :     0.000006s :     22: predicate.inline
 1.04% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.25% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.26% :     0.000001s :     10: predicate.merge_addn
 1.32% :     0.000002s :     10: predicate.minmaximum_grad
 1.66% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.83% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.73% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.51% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.30% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.28% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.42% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.72% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.16% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000505    12
47.27% :     0.000238s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.53% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.20% :     0.000238s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019385    76
 0.22% :     0.000044s :      1: add_recomputation
 0.33% :     0.000064s :      1: auto_monad
 0.07% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.51% :     0.000293s :      1: bootstrap
 0.12% :     0.000024s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
25.81% :     0.005003s :      1: jit_opt_a
 0.72% :     0.000140s :      1: jit_opt_after_cconv
 0.27% :     0.000052s :      1: jit_opt_b
 2.52% :     0.000488s :      1: loop_unroll
 2.66% :     0.000516s :      1: mutable_eliminate
 4.08% :     0.000791s :     26: opt.transform.jit_opt_a
 0.24% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.52% :     0.000488s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000014s :      1: remove_dup_value
 1.37% :     0.000265s :      1: renormalize.infer
 1.38% :     0.000268s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000033s :      1: rewriter_after_opt_a
 0.40% :     0.000077s :      1: rewriter_before_opt_a
 0.40% :     0.000077s :      1: symbol_engine_optimizer
32.09% :     0.006221s :      1: task_emit
21.44% :     0.004156s :      1: type_inference
 0.27% :     0.000051s :      1: validate


TotalTime = 0.0171598, [33]
[bootstrap]: 0.00026674
[type_inference]: 0.00435783
[event_method]: 1.632e-05
[auto_monad]: 6.332e-05
[graph_reusing]: 5.51002e-06
[pre_auto_parallel]: 2.31998e-06
[py_interpret_to_execute]: 2.991e-05
[rewriter_before_opt_a]: 7.001e-05
[expand_dump_flag]: 2.75002e-06
[jit_opt_a]: 0.00479149, [2]
    [Cycle 1]: 0.00135746, [27]
        [switch_simplify]: 4.424e-05
        [loop_unroll]: 2.986e-05
        [a_1]: 0.00048138
        [with_stream_mark]: 1.219e-05
        [recompute_prepare]: 7.08e-06
        [updatestate_depend_eliminate]: 2.94999e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 5.92999e-06
        [updatestate_useless_node_eliminater]: 5.01997e-06
        [accelerated_algorithm]: 5.69e-06
        [meta_shard_fg_expand]: 1.78002e-06
        [get_grad_eliminate_]: 5.12999e-06
        [merge_forward]: 2.37001e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.156e-05
        [j_node_and_user_rematch]: 9.39e-06
        [meta_fg_expand]: 1.80001e-06
        [replace_old_param]: 8.43999e-06
        [inline_without_move]: 5.22999e-06
        [renormalize]: 0.00049716
        [add_forward_monad_depend]: 4.18001e-06
        [auto_monad_grad]: 1.71e-06
        [auto_monad_eliminator]: 1.077e-05
        [cse]: 1.852e-05
        [replace_applicator]: 1.055e-05
    [Cycle 2]: 0.00031385, [27]
        [switch_simplify]: 5.74999e-06
        [loop_unroll]: 4.95001e-06
        [a_1]: 8.879e-05
        [with_stream_mark]: 8.55999e-06
        [recompute_prepare]: 4.81002e-06
        [updatestate_depend_eliminate]: 2.61999e-06
        [updatestate_assign_eliminate]: 2.14999e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 5.02e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.62998e-06
        [merge_forward]: 2.41998e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.157e-05
        [j_node_and_user_rematch]: 7.68001e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 7.56999e-06
        [inline_without_move]: 4.70001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 5.00001e-06
        [cse]: 9.87999e-06
        [replace_applicator]: 4.94e-06
[py_interpret_to_execute_after_opt_a]: 8.57e-06
[rewriter_after_opt_a]: 2.858e-05
[convert_after_rewriter]: 5.86e-06
[order_py_execute_after_rewriter]: 4.46002e-06
[mutable_eliminate]: 0.00049689
[jit_opt_b]: 5.329e-05, [1]
    [Cycle 1]: 4.663e-05, [2]
        [frontend_op_eliminate]: 1.949e-05
        [inline_after_opt_a]: 1.567e-05
[cconv]: 2.073e-05
[loop_unroll]: 0.00041412
[jit_opt_after_cconv]: 0.00013741, [1]
    [Cycle 1]: 0.00013065, [11]
        [c_1]: 1.995e-05
        [parameter_eliminate]: 2.66e-06
        [updatestate_depend_eliminate]: 5.24e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.655e-05
        [call_graph_tuple_transform]: 1.678e-05
        [tuple_list_get_item_eliminator]: 5.35999e-06
        [none_parameter_eliminate]: 1.12e-06
        [renormalize]: 4.2998e-07
        [switch_simplify]: 6.46999e-06
[remove_dup_value]: 9.79e-06
[partial_unused_args_eliminate]: 2.50002e-06
[environ_conv]: 4.23001e-06
[add_recomputation]: 4.053e-05
[cse_after_recomputation]: 1.925e-05, [1]
    [Cycle 1]: 1.378e-05, [1]
        [cse]: 8.33999e-06
[auto_monad_reorder]: 1.192e-05
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 4.84e-06
[opt_after_jit_grad]: 0.00053463
[symbol_engine_optimizer]: 7.189e-05, [1]
    [Cycle 1]: 6.562e-05, [6]
        [build]: 3.28e-06
        [elim_shapecalc]: 7.63999e-06
        [elim_not_effective]: 1.312e-05
        [opt_reshape]: 5.39e-06
        [fold_const_symbol]: 8.29002e-06
        [renormalize]: 4.59986e-07
[validate]: 2.978e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.0054648
[execute]: 5.69e-06
Sums
    bootstrap                            :     0.000267s :  1.98%
    type_inference                       :     0.004358s : 32.39%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000063s :  0.47%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.22%
    rewriter_before_opt_a                :     0.000070s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000050s :  0.37%
    jit_opt_a.loop_unroll                :     0.000035s :  0.26%
    jit_opt_a.a_1                        :     0.000570s :  4.24%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000497s :  3.70%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.12%
    jit_opt_a.cse                        :     0.000028s :  0.21%
    jit_opt_a.replace_applicator         :     0.000015s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000029s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000497s :  3.69%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.12%
    cconv                                :     0.000021s :  0.15%
    loop_unroll                          :     0.000414s :  3.08%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000010s :  0.07%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000041s :  0.30%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.09%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000535s :  3.97%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000030s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005465s : 40.61%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000159    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.84% :     0.000001s :      2: substitution.fold_const_symbol
 2.56% :     0.000004s :      3: substitution.graph_param_transform
82.14% :     0.000131s :      4: substitution.inline
 2.21% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.32% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.56% :     0.000004s :      2: substitution.replace_old_param
 6.31% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004300     2
82.26% :     0.003537s :      1: type_inference.infer
17.74% :     0.000763s :      1: type_inference.specialize
------[replace.]   0.000061     6
66.50% :     0.000041s :      4: replace.inline
33.50% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000137     6
93.47% :     0.000128s :      4: match.inline
 6.53% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.28% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 2.04% :     0.000002s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000001s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.30% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.71% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.89% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.70% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.51% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000000s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.59% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.85% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.69% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 1.16% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.62% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.53% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000513    12
51.14% :     0.000262s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.07% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.79% :     0.000225s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018531    76
 0.23% :     0.000043s :      1: add_recomputation
 0.36% :     0.000067s :      1: auto_monad
 0.08% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.53% :     0.000283s :      1: bootstrap
 0.13% :     0.000023s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.87% :     0.004794s :      1: jit_opt_a
 0.76% :     0.000140s :      1: jit_opt_after_cconv
 0.30% :     0.000056s :      1: jit_opt_b
 2.27% :     0.000421s :      1: loop_unroll
 2.72% :     0.000505s :      1: mutable_eliminate
 4.13% :     0.000766s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000028s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.93% :     0.000543s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000012s :      1: remove_dup_value
 1.35% :     0.000249s :      1: renormalize.infer
 1.30% :     0.000242s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000032s :      1: rewriter_after_opt_a
 0.39% :     0.000073s :      1: rewriter_before_opt_a
 0.40% :     0.000074s :      1: symbol_engine_optimizer
29.54% :     0.005475s :      1: task_emit
23.59% :     0.004370s :      1: type_inference
 0.25% :     0.000047s :      1: validate


TotalTime = 0.0142617, [30]
[bootstrap]: 0.00025539
[type_inference]: 0.00584645
[event_method]: 1.92e-05
[auto_monad]: 7.961e-05
[graph_reusing]: 5.22999e-06
[pre_auto_parallel]: 2.64999e-06
[py_interpret_to_execute]: 4.467e-05
[rewriter_before_opt_a]: 7.97e-05
[expand_dump_flag]: 2.71e-06
[jit_opt_a]: 0.00558509, [2]
    [Cycle 1]: 0.00195441, [27]
        [switch_simplify]: 0.00013626
        [loop_unroll]: 3.448e-05
        [a_1]: 0.00074029
        [with_stream_mark]: 1.885e-05
        [recompute_prepare]: 7.86001e-06
        [updatestate_depend_eliminate]: 3.00998e-06
        [updatestate_assign_eliminate]: 3.29001e-06
        [updatestate_loads_eliminate]: 2.66999e-06
        [parameter_eliminate]: 2.01e-06
        [specialize_transform]: 5.66e-06
        [updatestate_useless_node_eliminater]: 5.25001e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 5.40001e-06
        [merge_forward]: 3.39001e-06
        [cell_reuse_recompute_pass]: 1.70001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.476e-05
        [j_node_and_user_rematch]: 9.44e-06
        [meta_fg_expand]: 2.62001e-06
        [replace_old_param]: 8.64e-06
        [inline_without_move]: 5.24998e-06
        [renormalize]: 0.00070772
        [add_forward_monad_depend]: 5.40999e-06
        [auto_monad_grad]: 2.56e-06
        [auto_monad_eliminator]: 1.364e-05
        [cse]: 1.76e-05
        [replace_applicator]: 1.186e-05
    [Cycle 2]: 0.00030908, [27]
        [switch_simplify]: 5.77999e-06
        [loop_unroll]: 5.15999e-06
        [a_1]: 8.884e-05
        [with_stream_mark]: 8.60001e-06
        [recompute_prepare]: 5.00001e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 9.30013e-07
        [specialize_transform]: 4.92e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 1.37999e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.27999e-06
        [cell_reuse_recompute_pass]: 1.75001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.245e-05
        [j_node_and_user_rematch]: 7.5e-06
        [meta_fg_expand]: 1.58002e-06
        [replace_old_param]: 7.02002e-06
        [inline_without_move]: 4.68001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.23002e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 4.86997e-06
        [cse]: 9.15999e-06
        [replace_applicator]: 4.72e-06
[py_interpret_to_execute_after_opt_a]: 1.132e-05
[rewriter_after_opt_a]: 3.28e-05
[convert_after_rewriter]: 6.94999e-06
[order_py_execute_after_rewriter]: 5.24998e-06
[mutable_eliminate]: 0.00062651
[jit_opt_b]: 5.821e-05, [1]
    [Cycle 1]: 4.251e-05, [2]
        [frontend_op_eliminate]: 1.618e-05
        [inline_after_opt_a]: 1.53e-05
[cconv]: 2.681e-05
[loop_unroll]: 0.0004815
[jit_opt_after_cconv]: 0.00015389, [1]
    [Cycle 1]: 0.00014636, [11]
        [c_1]: 2.068e-05
        [parameter_eliminate]: 3.7e-06
        [updatestate_depend_eliminate]: 6.34999e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [cse]: 2.268e-05
        [call_graph_tuple_transform]: 2.139e-05
        [tuple_list_get_item_eliminator]: 5.91e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 1.35999e-06
        [switch_simplify]: 5.37001e-06
[remove_dup_value]: 1.531e-05
[partial_unused_args_eliminate]: 2.32999e-06
[environ_conv]: 6.41998e-06
[add_recomputation]: 5.317e-05
[cse_after_recomputation]: 2.295e-05, [1]
    [Cycle 1]: 1.657e-05, [1]
        [cse]: 1.094e-05
[auto_monad_reorder]: 1.533e-05
[get_jit_bprop_graph]: 2.17001e-06
[rewriter_after_jit_bprop_graph]: 7.31001e-06
[opt_after_jit_grad]: 0.00053145
[symbol_engine_optimizer]: 8.205e-05, [1]
    [Cycle 1]: 7.489e-05, [6]
        [build]: 4.99e-06
        [elim_shapecalc]: 8.62998e-06
        [elim_not_effective]: 1.567e-05
        [opt_reshape]: 5.40001e-06
        [fold_const_symbol]: 9.09e-06
        [renormalize]: 9.70002e-07
[validate]: 3.759e-05
Sums
    bootstrap                            :     0.000255s :  2.47%
    type_inference                       :     0.005846s : 56.46%
    event_method                         :     0.000019s :  0.19%
    auto_monad                           :     0.000080s :  0.77%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000045s :  0.43%
    rewriter_before_opt_a                :     0.000080s :  0.77%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000142s :  1.37%
    jit_opt_a.loop_unroll                :     0.000040s :  0.38%
    jit_opt_a.a_1                        :     0.000829s :  8.01%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000708s :  6.83%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.18%
    jit_opt_a.cse                        :     0.000027s :  0.26%
    jit_opt_a.replace_applicator         :     0.000017s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.11%
    rewriter_after_opt_a                 :     0.000033s :  0.32%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000627s :  6.05%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.15%
    cconv                                :     0.000027s :  0.26%
    loop_unroll                          :     0.000481s :  4.65%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000015s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000053s :  0.51%
    cse_after_recomputation.cse          :     0.000011s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000531s :  5.13%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000038s :  0.36%


Time group info:
------[substitution.]   0.000299    26
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 1.82% :     0.000005s :      3: substitution.graph_param_transform
86.27% :     0.000258s :      6: substitution.inline
 1.17% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.62% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.13% :     0.000003s :      2: substitution.replace_old_param
 2.63% :     0.000008s :      1: substitution.switch_simplify
 4.16% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005790     2
83.66% :     0.004844s :      1: type_inference.infer
16.34% :     0.000946s :      1: type_inference.specialize
------[replace.]   0.000158     9
44.35% :     0.000070s :      6: replace.inline
35.10% :     0.000056s :      1: replace.switch_simplify
20.56% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000273     9
93.22% :     0.000255s :      6: match.inline
 2.63% :     0.000007s :      1: match.switch_simplify
 4.15% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000145   801
 1.13% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     12: predicate.addn_check_dump
 1.23% :     0.000002s :     12: predicate.addn_zero_filter
 1.53% :     0.000002s :     12: predicate.arithmetic_simplify
 1.25% :     0.000002s :     12: predicate.cast_eliminate
 0.57% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     12: predicate.compare_switch_simplify
 1.25% :     0.000002s :     12: predicate.depend_value_elim
 1.07% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.12% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.71% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000001s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.08% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.15% :     0.000002s :     12: predicate.environ_get_eliminate
 1.25% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.79% :     0.000001s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 4.94% :     0.000007s :     26: predicate.inline
 0.71% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.04% :     0.000002s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     17: predicate.load_eliminater
 1.22% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.93% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.94% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     12: predicate.merge_addn
 1.00% :     0.000001s :     12: predicate.minmaximum_grad
 1.40% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 1.92% :     0.000003s :     17: predicate.partial_eliminate
 1.23% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     12: predicate.reduce_eliminate
 1.51% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.50% :     0.000002s :     20: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     12: predicate.reshape_eliminate
 1.10% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.66% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     12: predicate.same_eliminate
 0.42% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.81% :     0.000001s :      6: predicate.special_op_eliminate
 0.84% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.31% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000004s :     20: predicate.switch_defer_inline
10.55% :     0.000015s :     20: predicate.switch_layer_defer_inline
 7.46% :     0.000011s :     65: predicate.switch_simplify
 1.20% :     0.000002s :     12: predicate.tile_eliminate
 1.29% :     0.000002s :     12: predicate.transpose_eliminate
 1.42% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.50% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.29% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.55% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.47% :     0.000002s :     12: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.51% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000738    15
46.72% :     0.000345s :      6: func_graph_cloner_run.FuncGraphClonerGraph
14.36% :     0.000106s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.92% :     0.000287s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016157    72
 0.35% :     0.000057s :      1: add_recomputation
 0.52% :     0.000083s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 1.64% :     0.000264s :      1: bootstrap
 0.19% :     0.000030s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
34.59% :     0.005589s :      1: jit_opt_a
 0.97% :     0.000157s :      1: jit_opt_after_cconv
 0.38% :     0.000061s :      1: jit_opt_b
 3.04% :     0.000491s :      1: loop_unroll
 3.93% :     0.000635s :      1: mutable_eliminate
 6.95% :     0.001122s :     26: opt.transform.jit_opt_a
 0.31% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.36% :     0.000543s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000047s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.52% :     0.000407s :      1: renormalize.infer
 1.81% :     0.000293s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000036s :      1: rewriter_after_opt_a
 0.51% :     0.000083s :      1: rewriter_before_opt_a
 0.52% :     0.000085s :      1: symbol_engine_optimizer
36.24% :     0.005856s :      1: type_inference


TotalTime = 0.0201605, [33]
[bootstrap]: 0.00027905
[type_inference]: 0.00422103
[event_method]: 1.703e-05
[auto_monad]: 6.691e-05
[graph_reusing]: 5.61e-06
[pre_auto_parallel]: 2.51998e-06
[py_interpret_to_execute]: 3.183e-05
[rewriter_before_opt_a]: 6.88e-05
[expand_dump_flag]: 3.27002e-06
[jit_opt_a]: 0.00561556, [2]
    [Cycle 1]: 0.0018328, [27]
        [switch_simplify]: 5.545e-05
        [loop_unroll]: 3.356e-05
        [a_1]: 0.00058792
        [with_stream_mark]: 1.915e-05
        [recompute_prepare]: 8.13001e-06
        [updatestate_depend_eliminate]: 3.53e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [parameter_eliminate]: 1.54e-06
        [specialize_transform]: 6.62002e-06
        [updatestate_useless_node_eliminater]: 7.56999e-06
        [accelerated_algorithm]: 7.15e-06
        [meta_shard_fg_expand]: 2.49001e-06
        [get_grad_eliminate_]: 6.78998e-06
        [merge_forward]: 5.14e-06
        [cell_reuse_recompute_pass]: 1.91e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.692e-05
        [j_node_and_user_rematch]: 1.105e-05
        [meta_fg_expand]: 2.48e-06
        [replace_old_param]: 9.02e-06
        [inline_without_move]: 6.21e-06
        [renormalize]: 0.00072246
        [add_forward_monad_depend]: 5.70001e-06
        [auto_monad_grad]: 2.31998e-06
        [auto_monad_eliminator]: 1.464e-05
        [cse]: 2.583e-05
        [replace_applicator]: 1.411e-05
    [Cycle 2]: 0.00035623, [27]
        [switch_simplify]: 7.06001e-06
        [loop_unroll]: 5.10999e-06
        [a_1]: 0.00010033
        [with_stream_mark]: 1.106e-05
        [recompute_prepare]: 5.35001e-06
        [updatestate_depend_eliminate]: 2.96999e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 5.27999e-06
        [updatestate_useless_node_eliminater]: 5.59998e-06
        [accelerated_algorithm]: 5.61e-06
        [meta_shard_fg_expand]: 1.62001e-06
        [get_grad_eliminate_]: 5.23002e-06
        [merge_forward]: 3.39001e-06
        [cell_reuse_recompute_pass]: 1.92001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.355e-05
        [j_node_and_user_rematch]: 8.95001e-06
        [meta_fg_expand]: 1.80001e-06
        [replace_old_param]: 8.06001e-06
        [inline_without_move]: 4.94e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.59e-06
        [auto_monad_grad]: 7.79983e-07
        [auto_monad_eliminator]: 5.53002e-06
        [cse]: 1.163e-05
        [replace_applicator]: 5.65001e-06
[py_interpret_to_execute_after_opt_a]: 1.447e-05
[rewriter_after_opt_a]: 3.595e-05
[convert_after_rewriter]: 8.53001e-06
[order_py_execute_after_rewriter]: 5.70001e-06
[mutable_eliminate]: 0.0006858
[jit_opt_b]: 6.09e-05, [1]
    [Cycle 1]: 5.182e-05, [2]
        [frontend_op_eliminate]: 2e-05
        [inline_after_opt_a]: 1.816e-05
[cconv]: 2.832e-05
[loop_unroll]: 0.00058889
[jit_opt_after_cconv]: 0.00015506, [1]
    [Cycle 1]: 0.00014839, [11]
        [c_1]: 2.115e-05
        [parameter_eliminate]: 3.9e-06
        [updatestate_depend_eliminate]: 6.95998e-06
        [updatestate_assign_eliminate]: 2.69999e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [cse]: 2.326e-05
        [call_graph_tuple_transform]: 2.198e-05
        [tuple_list_get_item_eliminator]: 5.45001e-06
        [none_parameter_eliminate]: 1.32e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 5.98002e-06
[remove_dup_value]: 1.435e-05
[partial_unused_args_eliminate]: 2.34999e-06
[environ_conv]: 6.23e-06
[add_recomputation]: 5.046e-05
[cse_after_recomputation]: 2.228e-05, [1]
    [Cycle 1]: 1.567e-05, [1]
        [cse]: 9.59999e-06
[auto_monad_reorder]: 1.576e-05
[get_jit_bprop_graph]: 1.81e-06
[rewriter_after_jit_bprop_graph]: 8.54002e-06
[opt_after_jit_grad]: 0.00052765
[symbol_engine_optimizer]: 8.252e-05, [1]
    [Cycle 1]: 7.382e-05, [6]
        [build]: 4.48999e-06
        [elim_shapecalc]: 8.23001e-06
        [elim_not_effective]: 1.432e-05
        [opt_reshape]: 6.02999e-06
        [fold_const_symbol]: 9.89999e-06
        [renormalize]: 5.19998e-07
[validate]: 4.079e-05
[backend_pass]: 1.45001e-06
[task_emit]: 0.00727164
[execute]: 8.42e-06
Sums
    bootstrap                            :     0.000279s :  1.74%
    type_inference                       :     0.004221s : 26.36%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000067s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.20%
    rewriter_before_opt_a                :     0.000069s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.39%
    jit_opt_a.loop_unroll                :     0.000039s :  0.24%
    jit_opt_a.a_1                        :     0.000688s :  4.30%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000013s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.08%
    jit_opt_a.merge_forward              :     0.000009s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000723s :  4.51%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000037s :  0.23%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.09%
    rewriter_after_opt_a                 :     0.000036s :  0.22%
    convert_after_rewriter               :     0.000009s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000686s :  4.28%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.11%
    cconv                                :     0.000028s :  0.18%
    loop_unroll                          :     0.000589s :  3.68%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000050s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000528s :  3.30%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007272s : 45.42%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000220    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000002s :      2: substitution.fold_const_symbol
 2.56% :     0.000006s :      3: substitution.graph_param_transform
83.85% :     0.000185s :      4: substitution.inline
 1.83% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.52% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.00% :     0.000004s :      2: substitution.replace_old_param
 5.70% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004159     2
80.17% :     0.003335s :      1: type_inference.infer
19.83% :     0.000825s :      1: type_inference.specialize
------[replace.]   0.000074     6
65.80% :     0.000049s :      4: replace.inline
34.20% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000193     6
94.09% :     0.000182s :      4: match.inline
 5.91% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   691
 1.25% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.30% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.51% :     0.000002s :     10: predicate.addn_zero_filter
 2.24% :     0.000003s :     10: predicate.arithmetic_simplify
 1.23% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.47% :     0.000002s :     10: predicate.depend_value_elim
 1.34% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.20% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.22% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     10: predicate.environ_get_eliminate
 1.51% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.59% :     0.000002s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.26% :     0.000007s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.48% :     0.000002s :      6: predicate.less_batch_normalization
 1.31% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 1.25% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.98% :     0.000003s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.88% :     0.000003s :     10: predicate.reduce_eliminate
 1.30% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.77% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.45% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.31% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.16% :     0.000002s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.68% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.56% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.31% :     0.000010s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.20% :     0.000002s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.31% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.28% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.24% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000549    12
46.25% :     0.000254s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.56% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.19% :     0.000253s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021928    76
 0.25% :     0.000054s :      1: add_recomputation
 0.32% :     0.000071s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.33% :     0.000291s :      1: bootstrap
 0.15% :     0.000032s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.11% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.62% :     0.005619s :      1: jit_opt_a
 0.72% :     0.000158s :      1: jit_opt_after_cconv
 0.29% :     0.000064s :      1: jit_opt_b
 2.73% :     0.000598s :      1: loop_unroll
 3.18% :     0.000698s :      1: mutable_eliminate
 4.20% :     0.000921s :     26: opt.transform.jit_opt_a
 0.23% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000031s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.45% :     0.000538s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000034s :      1: py_interpret_to_execute
 0.08% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.72% :     0.000378s :      1: renormalize.infer
 1.53% :     0.000336s :      1: renormalize.specialize
 0.05% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000039s :      1: rewriter_after_opt_a
 0.33% :     0.000072s :      1: rewriter_before_opt_a
 0.39% :     0.000085s :      1: symbol_engine_optimizer
33.23% :     0.007287s :      1: task_emit
19.31% :     0.004234s :      1: type_inference
 0.29% :     0.000065s :      1: validate


TotalTime = 0.0183554, [33]
[bootstrap]: 0.00025024
[type_inference]: 0.00421014
[event_method]: 1.753e-05
[auto_monad]: 4.582e-05
[graph_reusing]: 4.35e-06
[pre_auto_parallel]: 2.24999e-06
[py_interpret_to_execute]: 3.402e-05
[rewriter_before_opt_a]: 7.324e-05
[expand_dump_flag]: 2.68e-06
[jit_opt_a]: 0.00524947, [2]
    [Cycle 1]: 0.00148993, [27]
        [switch_simplify]: 5.216e-05
        [loop_unroll]: 3.347e-05
        [a_1]: 0.00051213
        [with_stream_mark]: 1.418e-05
        [recompute_prepare]: 7.43e-06
        [updatestate_depend_eliminate]: 3.24001e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 6.29999e-06
        [updatestate_useless_node_eliminater]: 5.94e-06
        [accelerated_algorithm]: 5.81e-06
        [meta_shard_fg_expand]: 1.64e-06
        [get_grad_eliminate_]: 6.37001e-06
        [merge_forward]: 2.89001e-06
        [cell_reuse_recompute_pass]: 7.80012e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.279e-05
        [j_node_and_user_rematch]: 8.85001e-06
        [meta_fg_expand]: 2.05002e-06
        [replace_old_param]: 8.28999e-06
        [inline_without_move]: 6.09001e-06
        [renormalize]: 0.00056889
        [add_forward_monad_depend]: 4.92e-06
        [auto_monad_grad]: 1.72999e-06
        [auto_monad_eliminator]: 1.1e-05
        [cse]: 1.847e-05
        [replace_applicator]: 1.321e-05
    [Cycle 2]: 0.00033689, [27]
        [switch_simplify]: 6.14001e-06
        [loop_unroll]: 5.33002e-06
        [a_1]: 9.821e-05
        [with_stream_mark]: 9.20999e-06
        [recompute_prepare]: 4.75001e-06
        [updatestate_depend_eliminate]: 2.58e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 8.60018e-07
        [specialize_transform]: 4.82e-06
        [updatestate_useless_node_eliminater]: 4.70999e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.19e-06
        [get_grad_eliminate_]: 5.40999e-06
        [merge_forward]: 2.53003e-06
        [cell_reuse_recompute_pass]: 1.83002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.364e-05
        [j_node_and_user_rematch]: 8.90999e-06
        [meta_fg_expand]: 1.78002e-06
        [replace_old_param]: 8.09997e-06
        [inline_without_move]: 5.35999e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 8.90024e-07
        [auto_monad_eliminator]: 5.34998e-06
        [cse]: 1.153e-05
        [replace_applicator]: 6.02999e-06
[py_interpret_to_execute_after_opt_a]: 9.69e-06
[rewriter_after_opt_a]: 2.924e-05
[convert_after_rewriter]: 6.17999e-06
[order_py_execute_after_rewriter]: 1.389e-05
[mutable_eliminate]: 0.00055834
[jit_opt_b]: 5.742e-05, [1]
    [Cycle 1]: 5.021e-05, [2]
        [frontend_op_eliminate]: 2.023e-05
        [inline_after_opt_a]: 1.692e-05
[cconv]: 1.771e-05
[loop_unroll]: 0.00046629
[jit_opt_after_cconv]: 0.00015019, [1]
    [Cycle 1]: 0.00014355, [11]
        [c_1]: 2.377e-05
        [parameter_eliminate]: 2.64999e-06
        [updatestate_depend_eliminate]: 5.74e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.32001e-06
        [cse]: 1.895e-05
        [call_graph_tuple_transform]: 1.888e-05
        [tuple_list_get_item_eliminator]: 6.23002e-06
        [none_parameter_eliminate]: 1.21997e-06
        [renormalize]: 5.59987e-07
        [switch_simplify]: 5.69e-06
[remove_dup_value]: 7.9e-06
[partial_unused_args_eliminate]: 1.76e-06
[environ_conv]: 4.75999e-06
[add_recomputation]: 3.328e-05
[cse_after_recomputation]: 2.139e-05, [1]
    [Cycle 1]: 1.577e-05, [1]
        [cse]: 9.49999e-06
[auto_monad_reorder]: 1.046e-05
[get_jit_bprop_graph]: 1.54998e-06
[rewriter_after_jit_bprop_graph]: 4.97999e-06
[opt_after_jit_grad]: 0.00051939
[symbol_engine_optimizer]: 7.093e-05, [1]
    [Cycle 1]: 6.458e-05, [6]
        [build]: 3.28998e-06
        [elim_shapecalc]: 7.53999e-06
        [elim_not_effective]: 1.17e-05
        [opt_reshape]: 5.61e-06
        [fold_const_symbol]: 8.42e-06
        [renormalize]: 4.80009e-07
[validate]: 2.732e-05
[backend_pass]: 1.20001e-06
[task_emit]: 0.00624083
[execute]: 9.12999e-06
Sums
    bootstrap                            :     0.000250s :  1.75%
    type_inference                       :     0.004210s : 29.42%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000046s :  0.32%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.24%
    rewriter_before_opt_a                :     0.000073s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.41%
    jit_opt_a.loop_unroll                :     0.000039s :  0.27%
    jit_opt_a.a_1                        :     0.000610s :  4.26%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.000569s :  3.98%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.11%
    jit_opt_a.cse                        :     0.000030s :  0.21%
    jit_opt_a.replace_applicator         :     0.000019s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000029s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000014s :  0.10%
    mutable_eliminate                    :     0.000558s :  3.90%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.12%
    cconv                                :     0.000018s :  0.12%
    loop_unroll                          :     0.000466s :  3.26%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.17%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000008s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000033s :  0.23%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000010s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000519s :  3.63%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.19%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006241s : 43.61%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000158    23
 0.87% :     0.000001s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.67% :     0.000004s :      3: substitution.graph_param_transform
83.48% :     0.000132s :      4: substitution.inline
 1.78% :     0.000003s :      4: substitution.j_node_and_user_rematch
 3.01% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.93% :     0.000003s :      2: substitution.replace_old_param
 5.66% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004151     2
80.69% :     0.003350s :      1: type_inference.infer
19.31% :     0.000802s :      1: type_inference.specialize
------[replace.]   0.000066     6
65.40% :     0.000043s :      4: replace.inline
34.60% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000137     6
94.39% :     0.000129s :      4: match.inline
 5.61% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.04% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.24% :     0.000003s :     10: predicate.arithmetic_simplify
 1.36% :     0.000002s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     10: predicate.compare_switch_simplify
 1.50% :     0.000002s :     10: predicate.depend_value_elim
 1.30% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.06% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.24% :     0.000002s :      6: predicate.get_grad_eliminate
 0.52% :     0.000001s :      3: predicate.graph_param_transform
 5.02% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.34% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.29% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.76% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.54% :     0.000002s :     10: predicate.reshape_eliminate
 1.43% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.18% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000002s :     10: predicate.transpose_eliminate
 1.54% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.36% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.77% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000507    12
45.91% :     0.000233s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.30% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.80% :     0.000247s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019859    76
 0.18% :     0.000036s :      1: add_recomputation
 0.25% :     0.000049s :      1: auto_monad
 0.07% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.33% :     0.000264s :      1: bootstrap
 0.10% :     0.000020s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
26.45% :     0.005253s :      1: jit_opt_a
 0.77% :     0.000153s :      1: jit_opt_after_cconv
 0.30% :     0.000060s :      1: jit_opt_b
 2.39% :     0.000475s :      1: loop_unroll
 2.87% :     0.000569s :      1: mutable_eliminate
 4.17% :     0.000828s :     26: opt.transform.jit_opt_a
 0.26% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000030s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.66% :     0.000527s :      1: opt_after_jit_grad
 0.09% :     0.000017s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000010s :      1: remove_dup_value
 1.42% :     0.000283s :      1: renormalize.infer
 1.40% :     0.000279s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000033s :      1: rewriter_after_opt_a
 0.38% :     0.000076s :      1: rewriter_before_opt_a
 0.37% :     0.000073s :      1: symbol_engine_optimizer
31.49% :     0.006254s :      1: task_emit
21.26% :     0.004222s :      1: type_inference
 0.22% :     0.000044s :      1: validate


TotalTime = 0.0147581, [30]
[bootstrap]: 0.00032513
[type_inference]: 0.00619776
[event_method]: 1.942e-05
[auto_monad]: 8.055e-05
[graph_reusing]: 6.26998e-06
[pre_auto_parallel]: 2.39999e-06
[py_interpret_to_execute]: 4.563e-05
[rewriter_before_opt_a]: 8.253e-05
[expand_dump_flag]: 2.68998e-06
[jit_opt_a]: 0.00569312, [2]
    [Cycle 1]: 0.00202697, [27]
        [switch_simplify]: 0.00012898
        [loop_unroll]: 3.537e-05
        [a_1]: 0.00074583
        [with_stream_mark]: 2.066e-05
        [recompute_prepare]: 9.61998e-06
        [updatestate_depend_eliminate]: 3.45e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 1.92001e-06
        [specialize_transform]: 5.98002e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 5.46002e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 4.95999e-06
        [merge_forward]: 3.93001e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.558e-05
        [j_node_and_user_rematch]: 9.27999e-06
        [meta_fg_expand]: 2.18002e-06
        [replace_old_param]: 8.89998e-06
        [inline_without_move]: 5.26002e-06
        [renormalize]: 0.00075049
        [add_forward_monad_depend]: 6.76e-06
        [auto_monad_grad]: 3.06999e-06
        [auto_monad_eliminator]: 1.435e-05
        [cse]: 2.639e-05
        [replace_applicator]: 1.655e-05
    [Cycle 2]: 0.00042778, [27]
        [switch_simplify]: 5.69e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 0.00014823
        [with_stream_mark]: 1.398e-05
        [recompute_prepare]: 5.77999e-06
        [updatestate_depend_eliminate]: 2.59999e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 8.57998e-06
        [parameter_eliminate]: 1.59e-06
        [specialize_transform]: 5.64e-06
        [updatestate_useless_node_eliminater]: 1.17e-05
        [accelerated_algorithm]: 4.63001e-06
        [meta_shard_fg_expand]: 1.43002e-06
        [get_grad_eliminate_]: 4.50999e-06
        [merge_forward]: 3.01001e-06
        [cell_reuse_recompute_pass]: 2.37999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.561e-05
        [j_node_and_user_rematch]: 8.40999e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 8.77999e-06
        [inline_without_move]: 5.19e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.61998e-06
        [auto_monad_grad]: 1.69998e-06
        [auto_monad_eliminator]: 6.61e-06
        [cse]: 1.573e-05
        [replace_applicator]: 7.41001e-06
[py_interpret_to_execute_after_opt_a]: 1.432e-05
[rewriter_after_opt_a]: 3.893e-05
[convert_after_rewriter]: 8.2e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.00071189
[jit_opt_b]: 5.065e-05, [1]
    [Cycle 1]: 4.263e-05, [2]
        [frontend_op_eliminate]: 1.629e-05
        [inline_after_opt_a]: 1.499e-05
[cconv]: 2.826e-05
[loop_unroll]: 0.00044119
[jit_opt_after_cconv]: 0.00015096, [1]
    [Cycle 1]: 0.00014419, [11]
        [c_1]: 2.148e-05
        [parameter_eliminate]: 3.15002e-06
        [updatestate_depend_eliminate]: 6.11998e-06
        [updatestate_assign_eliminate]: 2.64999e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 2.153e-05
        [call_graph_tuple_transform]: 2.007e-05
        [tuple_list_get_item_eliminator]: 6.02001e-06
        [none_parameter_eliminate]: 1.78997e-06
        [renormalize]: 8.49977e-07
        [switch_simplify]: 6.06e-06
[remove_dup_value]: 1.381e-05
[partial_unused_args_eliminate]: 2.32001e-06
[environ_conv]: 5.63002e-06
[add_recomputation]: 4.883e-05
[cse_after_recomputation]: 2.243e-05, [1]
    [Cycle 1]: 1.645e-05, [1]
        [cse]: 9.67999e-06
[auto_monad_reorder]: 1.652e-05
[get_jit_bprop_graph]: 2.22999e-06
[rewriter_after_jit_bprop_graph]: 6.41998e-06
[opt_after_jit_grad]: 0.00046652
[symbol_engine_optimizer]: 7.314e-05, [1]
    [Cycle 1]: 6.605e-05, [6]
        [build]: 3.14001e-06
        [elim_shapecalc]: 7.4e-06
        [elim_not_effective]: 1.348e-05
        [opt_reshape]: 5.24e-06
        [fold_const_symbol]: 8.75999e-06
        [renormalize]: 6.79982e-07
[validate]: 3.418e-05
Sums
    bootstrap                            :     0.000325s :  2.98%
    type_inference                       :     0.006198s : 56.77%
    event_method                         :     0.000019s :  0.18%
    auto_monad                           :     0.000081s :  0.74%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000046s :  0.42%
    rewriter_before_opt_a                :     0.000083s :  0.76%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000135s :  1.23%
    jit_opt_a.loop_unroll                :     0.000040s :  0.37%
    jit_opt_a.a_1                        :     0.000894s :  8.19%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.10%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000017s :  0.15%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000751s :  6.88%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.19%
    jit_opt_a.cse                        :     0.000042s :  0.39%
    jit_opt_a.replace_applicator         :     0.000024s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.13%
    rewriter_after_opt_a                 :     0.000039s :  0.36%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000712s :  6.52%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.14%
    cconv                                :     0.000028s :  0.26%
    loop_unroll                          :     0.000441s :  4.04%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000014s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000049s :  0.45%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000017s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000467s :  4.27%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000034s :  0.31%


Time group info:
------[substitution.]   0.000306    26
 0.58% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 1.84% :     0.000006s :      3: substitution.graph_param_transform
83.97% :     0.000257s :      6: substitution.inline
 1.20% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.47% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.44% :     0.000004s :      2: substitution.replace_old_param
 2.85% :     0.000009s :      1: substitution.switch_simplify
 6.20% :     0.000019s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006139     2
83.86% :     0.005148s :      1: type_inference.infer
16.14% :     0.000991s :      1: type_inference.specialize
------[replace.]   0.000146     9
48.89% :     0.000071s :      6: replace.inline
28.24% :     0.000041s :      1: replace.switch_simplify
22.86% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000278     9
90.90% :     0.000253s :      6: match.inline
 2.87% :     0.000008s :      1: match.switch_simplify
 6.23% :     0.000017s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000208   801
 0.87% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.73% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.72% :     0.000002s :     12: predicate.addn_check_dump
 0.90% :     0.000002s :     12: predicate.addn_zero_filter
 1.39% :     0.000003s :     12: predicate.arithmetic_simplify
 0.87% :     0.000002s :     12: predicate.cast_eliminate
 0.24% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.69% :     0.000001s :     12: predicate.compare_switch_simplify
 0.97% :     0.000002s :     12: predicate.depend_value_elim
 0.72% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 0.79% :     0.000002s :     12: predicate.dict_get_item_eliminator
 0.78% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.46% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.20% :     0.000000s :      3: predicate.elim_not_effective
 0.31% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.75% :     0.000002s :     12: predicate.environ_add_const_eliminate
 0.71% :     0.000001s :     12: predicate.environ_get_add_eliminate
 0.72% :     0.000001s :     12: predicate.environ_get_depend_swap
 0.75% :     0.000002s :     12: predicate.environ_get_eliminate
 0.77% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.13% :     0.000000s :      3: predicate.fold_const_symbol
 0.52% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 3.48% :     0.000007s :     26: predicate.inline
 0.57% :     0.000001s :      6: predicate.inline_without_move
 0.21% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.76% :     0.000002s :      6: predicate.less_batch_normalization
 1.06% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.10% :     0.000002s :     17: predicate.load_eliminater
 0.98% :     0.000002s :      3: predicate.loop_unroll_after_grad
 2.59% :     0.000005s :     40: predicate.loop_unroll_before_grad
 3.68% :     0.000008s :     15: predicate.make_slice_get_slice_eliminator
 0.75% :     0.000002s :     12: predicate.merge_addn
 0.70% :     0.000001s :     12: predicate.minmaximum_grad
 1.00% :     0.000002s :      3: predicate.mutable_eliminate
 0.23% :     0.000000s :      3: predicate.opt_reshape
 1.41% :     0.000003s :     17: predicate.partial_eliminate
 0.73% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.05% :     0.000002s :     12: predicate.reduce_eliminate
 0.91% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.28% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.12% :     0.000002s :     20: predicate.replace_applicator
 0.44% :     0.000001s :      6: predicate.replace_old_param
 0.21% :     0.000000s :      3: predicate.reset_defer_inline
 1.17% :     0.000002s :     12: predicate.reshape_eliminate
 0.82% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.42% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.80% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.73% :     0.000002s :      6: predicate.special_op_eliminate
 0.65% :     0.000001s :      6: predicate.specialize_transform
 0.87% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 0.82% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.22% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 7.00% :     0.000015s :     20: predicate.switch_defer_inline
 1.52% :     0.000003s :     20: predicate.switch_layer_defer_inline
 5.30% :     0.000011s :     65: predicate.switch_simplify
 0.74% :     0.000002s :     12: predicate.tile_eliminate
 0.76% :     0.000002s :     12: predicate.transpose_eliminate
 0.90% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 0.84% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 2.47% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.12% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 0.97% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.04% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
30.27% :     0.000063s :     23: predicate.updatestate_useless_node_eliminater
 0.97% :     0.000002s :     12: predicate.value_based_eliminate
 0.21% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.46% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000754    15
46.76% :     0.000352s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.64% :     0.000103s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.60% :     0.000299s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016770    72
 0.31% :     0.000052s :      1: add_recomputation
 0.50% :     0.000084s :      1: auto_monad
 0.11% :     0.000019s :      1: auto_monad_reorder
 2.02% :     0.000338s :      1: bootstrap
 0.18% :     0.000031s :      1: cconv
 0.07% :     0.000011s :      1: convert_after_rewriter
 0.15% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.13% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.97% :     0.005698s :      1: jit_opt_a
 0.92% :     0.000154s :      1: jit_opt_after_cconv
 0.32% :     0.000054s :      1: jit_opt_b
 2.68% :     0.000449s :      1: loop_unroll
 4.30% :     0.000722s :      1: mutable_eliminate
 7.14% :     0.001198s :     26: opt.transform.jit_opt_a
 0.30% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.83% :     0.000474s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.29% :     0.000048s :      1: py_interpret_to_execute
 0.10% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.45% :     0.000411s :      1: renormalize.infer
 1.97% :     0.000331s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000043s :      1: rewriter_after_opt_a
 0.51% :     0.000085s :      1: rewriter_before_opt_a
 0.45% :     0.000076s :      1: symbol_engine_optimizer
37.01% :     0.006207s :      1: type_inference


TotalTime = 0.0178876, [33]
[bootstrap]: 0.000292
[type_inference]: 0.00429558
[event_method]: 1.48e-05
[auto_monad]: 4.777e-05
[graph_reusing]: 4.17e-06
[pre_auto_parallel]: 2.36998e-06
[py_interpret_to_execute]: 3.249e-05
[rewriter_before_opt_a]: 7.533e-05
[expand_dump_flag]: 2.06e-06
[jit_opt_a]: 0.00492336, [2]
    [Cycle 1]: 0.00142356, [27]
        [switch_simplify]: 4.438e-05
        [loop_unroll]: 2.989e-05
        [a_1]: 0.0004707
        [with_stream_mark]: 1.095e-05
        [recompute_prepare]: 6.47001e-06
        [updatestate_depend_eliminate]: 2.74001e-06
        [updatestate_assign_eliminate]: 2.16003e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 6.38e-06
        [updatestate_useless_node_eliminater]: 5.11997e-06
        [accelerated_algorithm]: 5.34998e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 5.03002e-06
        [merge_forward]: 2.98e-06
        [cell_reuse_recompute_pass]: 8.29983e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.114e-05
        [j_node_and_user_rematch]: 8.38999e-06
        [meta_fg_expand]: 1.89e-06
        [replace_old_param]: 1.396e-05
        [inline_without_move]: 5.52999e-06
        [renormalize]: 0.00057006
        [add_forward_monad_depend]: 4.87998e-06
        [auto_monad_grad]: 1.74e-06
        [auto_monad_eliminator]: 1.273e-05
        [cse]: 2.778e-05
        [replace_applicator]: 1.152e-05
    [Cycle 2]: 0.00033856, [27]
        [switch_simplify]: 5.99e-06
        [loop_unroll]: 4.89998e-06
        [a_1]: 9.13e-05
        [with_stream_mark]: 9.37001e-06
        [recompute_prepare]: 5.34e-06
        [updatestate_depend_eliminate]: 2.86999e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 5.87001e-06
        [updatestate_useless_node_eliminater]: 5.18002e-06
        [accelerated_algorithm]: 5.79999e-06
        [meta_shard_fg_expand]: 1.35999e-06
        [get_grad_eliminate_]: 5.05999e-06
        [merge_forward]: 3.13e-06
        [cell_reuse_recompute_pass]: 2.06998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.275e-05
        [j_node_and_user_rematch]: 9.19e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 7.52998e-06
        [inline_without_move]: 4.89e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.16002e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.68002e-06
        [cse]: 1.136e-05
        [replace_applicator]: 6.04999e-06
[py_interpret_to_execute_after_opt_a]: 9.00999e-06
[rewriter_after_opt_a]: 3.75e-05
[convert_after_rewriter]: 6.75002e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.0005706
[jit_opt_b]: 5.133e-05, [1]
    [Cycle 1]: 4.437e-05, [2]
        [frontend_op_eliminate]: 1.691e-05
        [inline_after_opt_a]: 1.623e-05
[cconv]: 2.624e-05
[loop_unroll]: 0.00044721
[jit_opt_after_cconv]: 0.00013054, [1]
    [Cycle 1]: 0.00012444, [11]
        [c_1]: 2.036e-05
        [parameter_eliminate]: 1.72999e-06
        [updatestate_depend_eliminate]: 4.48999e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 1.478e-05
        [call_graph_tuple_transform]: 1.699e-05
        [tuple_list_get_item_eliminator]: 5.47001e-06
        [none_parameter_eliminate]: 1.20001e-06
        [renormalize]: 5.3001e-07
        [switch_simplify]: 5.39e-06
[remove_dup_value]: 1.424e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 5.17e-06
[add_recomputation]: 4.801e-05
[cse_after_recomputation]: 1.958e-05, [1]
    [Cycle 1]: 1.456e-05, [1]
        [cse]: 8.90001e-06
[auto_monad_reorder]: 1.503e-05
[get_jit_bprop_graph]: 1.67001e-06
[rewriter_after_jit_bprop_graph]: 5.02e-06
[opt_after_jit_grad]: 0.00043734
[symbol_engine_optimizer]: 7.139e-05, [1]
    [Cycle 1]: 6.515e-05, [6]
        [build]: 3.45e-06
        [elim_shapecalc]: 7.48999e-06
        [elim_not_effective]: 1.218e-05
        [opt_reshape]: 5.49e-06
        [fold_const_symbol]: 8.72e-06
        [renormalize]: 7.2e-07
[validate]: 3.488e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00606821
[execute]: 8.37998e-06
Sums
    bootstrap                            :     0.000292s :  2.06%
    type_inference                       :     0.004296s : 30.36%
    event_method                         :     0.000015s :  0.10%
    auto_monad                           :     0.000048s :  0.34%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.23%
    rewriter_before_opt_a                :     0.000075s :  0.53%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000050s :  0.36%
    jit_opt_a.loop_unroll                :     0.000035s :  0.25%
    jit_opt_a.a_1                        :     0.000562s :  3.97%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000570s :  4.03%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000039s :  0.28%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000037s :  0.27%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000571s :  4.03%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000026s :  0.19%
    loop_unroll                          :     0.000447s :  3.16%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000048s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000437s :  3.09%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000035s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006068s : 42.89%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000147    23
 1.34% :     0.000002s :      2: substitution.elim_not_effective
 0.80% :     0.000001s :      2: substitution.fold_const_symbol
 3.47% :     0.000005s :      3: substitution.graph_param_transform
81.93% :     0.000120s :      4: substitution.inline
 2.01% :     0.000003s :      4: substitution.j_node_and_user_rematch
 3.02% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.25% :     0.000003s :      2: substitution.replace_old_param
 5.18% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004243     2
81.74% :     0.003468s :      1: type_inference.infer
18.26% :     0.000775s :      1: type_inference.specialize
------[replace.]   0.000061     6
64.96% :     0.000039s :      4: replace.inline
35.04% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000125     6
94.75% :     0.000118s :      4: match.inline
 5.25% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.24% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.52% :     0.000002s :     10: predicate.addn_zero_filter
 1.88% :     0.000002s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.57% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.20% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 4.74% :     0.000005s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.21% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.63% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.93% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.26% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.73% :     0.000002s :     10: predicate.reduce_eliminate
 1.70% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.84% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.35% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.55% :     0.000002s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.19% :     0.000001s :      6: predicate.special_op_eliminate
 1.14% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.49% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.55% :     0.000003s :     16: predicate.switch_defer_inline
 2.35% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.60% :     0.000009s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.62% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000530    12
45.24% :     0.000240s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.76% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.00% :     0.000260s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019331    76
 0.26% :     0.000051s :      1: add_recomputation
 0.26% :     0.000051s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.59% :     0.000307s :      1: bootstrap
 0.15% :     0.000029s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000018s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
25.49% :     0.004927s :      1: jit_opt_a
 0.69% :     0.000133s :      1: jit_opt_after_cconv
 0.28% :     0.000054s :      1: jit_opt_b
 2.35% :     0.000454s :      1: loop_unroll
 2.98% :     0.000577s :      1: mutable_eliminate
 3.98% :     0.000770s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.29% :     0.000443s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000035s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000017s :      1: remove_dup_value
 1.46% :     0.000283s :      1: renormalize.infer
 1.46% :     0.000282s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000041s :      1: rewriter_after_opt_a
 0.41% :     0.000079s :      1: rewriter_before_opt_a
 0.38% :     0.000074s :      1: symbol_engine_optimizer
31.45% :     0.006080s :      1: task_emit
22.27% :     0.004306s :      1: type_inference
 0.28% :     0.000054s :      1: validate


TotalTime = 0.0870196, [33]
[bootstrap]: 0.00031461
[type_inference]: 0.0705596
[event_method]: 1.752e-05
[auto_monad]: 6.352e-05
[graph_reusing]: 5.92999e-06
[pre_auto_parallel]: 2.32999e-06
[py_interpret_to_execute]: 0.00030829
[rewriter_before_opt_a]: 8.527e-05
[expand_dump_flag]: 4.02e-06
[jit_opt_a]: 0.00730141, [2]
    [Cycle 1]: 0.00323417, [27]
        [switch_simplify]: 6.247e-05
        [loop_unroll]: 3.344e-05
        [a_1]: 0.00059102
        [with_stream_mark]: 1.694e-05
        [recompute_prepare]: 9.10001e-06
        [updatestate_depend_eliminate]: 4.04002e-06
        [updatestate_assign_eliminate]: 3.45e-06
        [updatestate_loads_eliminate]: 3.43e-06
        [parameter_eliminate]: 2.16e-06
        [specialize_transform]: 7.18e-06
        [updatestate_useless_node_eliminater]: 6.54999e-06
        [accelerated_algorithm]: 6.52001e-06
        [meta_shard_fg_expand]: 2.44999e-06
        [get_grad_eliminate_]: 6.54001e-06
        [merge_forward]: 3.91001e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.615e-05
        [j_node_and_user_rematch]: 1.01e-05
        [meta_fg_expand]: 2.34999e-06
        [replace_old_param]: 1.075e-05
        [inline_without_move]: 6.67002e-06
        [renormalize]: 0.00218352
        [add_forward_monad_depend]: 6.38e-06
        [auto_monad_grad]: 1.25999e-06
        [auto_monad_eliminator]: 1.444e-05
        [cse]: 2.201e-05
        [replace_applicator]: 1.385e-05
    [Cycle 2]: 0.00045005, [27]
        [switch_simplify]: 6.87002e-06
        [loop_unroll]: 6.75998e-06
        [a_1]: 0.00017311
        [with_stream_mark]: 1.141e-05
        [recompute_prepare]: 6.68e-06
        [updatestate_depend_eliminate]: 3.31999e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [parameter_eliminate]: 1.09998e-06
        [specialize_transform]: 6.89001e-06
        [updatestate_useless_node_eliminater]: 6.69001e-06
        [accelerated_algorithm]: 6.50002e-06
        [meta_shard_fg_expand]: 1.27999e-06
        [get_grad_eliminate_]: 5.77001e-06
        [merge_forward]: 3.55998e-06
        [cell_reuse_recompute_pass]: 2.98e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.305e-05
        [j_node_and_user_rematch]: 9.19e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 8.95001e-06
        [inline_without_move]: 5.86e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.50001e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 6.94999e-06
        [cse]: 1.282e-05
        [replace_applicator]: 6.76999e-06
[py_interpret_to_execute_after_opt_a]: 1.085e-05
[rewriter_after_opt_a]: 2.835e-05
[convert_after_rewriter]: 5.72001e-06
[order_py_execute_after_rewriter]: 4.55999e-06
[mutable_eliminate]: 0.00055073
[jit_opt_b]: 5.475e-05, [1]
    [Cycle 1]: 4.926e-05, [2]
        [frontend_op_eliminate]: 1.953e-05
        [inline_after_opt_a]: 1.85e-05
[cconv]: 1.73e-05
[loop_unroll]: 0.00041663
[jit_opt_after_cconv]: 0.00014731, [1]
    [Cycle 1]: 0.00014148, [11]
        [c_1]: 2.637e-05
        [parameter_eliminate]: 1.96e-06
        [updatestate_depend_eliminate]: 4.3e-06
        [updatestate_assign_eliminate]: 2.45002e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [cse]: 1.911e-05
        [call_graph_tuple_transform]: 2.053e-05
        [tuple_list_get_item_eliminator]: 6.62002e-06
        [none_parameter_eliminate]: 1.23002e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 6.44999e-06
[remove_dup_value]: 1.018e-05
[partial_unused_args_eliminate]: 1.63002e-06
[environ_conv]: 6.91001e-06
[add_recomputation]: 3.466e-05
[cse_after_recomputation]: 2.334e-05, [1]
    [Cycle 1]: 1.759e-05, [1]
        [cse]: 1.149e-05
[auto_monad_reorder]: 1.023e-05
[get_jit_bprop_graph]: 1.39998e-06
[rewriter_after_jit_bprop_graph]: 4.75001e-06
[opt_after_jit_grad]: 0.00048538
[symbol_engine_optimizer]: 7.839e-05, [1]
    [Cycle 1]: 7.193e-05, [6]
        [build]: 2.53003e-06
        [elim_shapecalc]: 9.00001e-06
        [elim_not_effective]: 1.413e-05
        [opt_reshape]: 7.01001e-06
        [fold_const_symbol]: 1.022e-05
        [renormalize]: 4.30009e-07
[validate]: 3.595e-05
[backend_pass]: 8.00006e-07
[task_emit]: 0.0062312
[execute]: 6.00002e-06
Sums
    bootstrap                            :     0.000315s :  0.38%
    type_inference                       :     0.070560s : 85.25%
    event_method                         :     0.000018s :  0.02%
    auto_monad                           :     0.000064s :  0.08%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.00%
    py_interpret_to_execute              :     0.000308s :  0.37%
    rewriter_before_opt_a                :     0.000085s :  0.10%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000069s :  0.08%
    jit_opt_a.loop_unroll                :     0.000040s :  0.05%
    jit_opt_a.a_1                        :     0.000764s :  0.92%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.03%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.00%
    jit_opt_a.specialize_transform       :     0.000014s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000013s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.00%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.01%
    jit_opt_a.merge_forward              :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.04%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.02%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000020s :  0.02%
    jit_opt_a.inline_without_move        :     0.000013s :  0.02%
    jit_opt_a.renormalize                :     0.002184s :  2.64%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.03%
    jit_opt_a.cse                        :     0.000035s :  0.04%
    jit_opt_a.replace_applicator         :     0.000021s :  0.02%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.01%
    rewriter_after_opt_a                 :     0.000028s :  0.03%
    convert_after_rewriter               :     0.000006s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000551s :  0.67%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.02%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.02%
    cconv                                :     0.000017s :  0.02%
    loop_unroll                          :     0.000417s :  0.50%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.03%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000019s :  0.02%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.02%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.01%
    remove_dup_value                     :     0.000010s :  0.01%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000007s :  0.01%
    add_recomputation                    :     0.000035s :  0.04%
    cse_after_recomputation.cse          :     0.000011s :  0.01%
    auto_monad_reorder                   :     0.000010s :  0.01%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000485s :  0.59%
    symbol_engine_optimizer.build        :     0.000003s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.02%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.04%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006231s :  7.53%
    execute                              :     0.000006s :  0.01%


Time group info:
------[substitution.]   0.000243    27
 0.65% :     0.000002s :      2: substitution.elim_not_effective
 0.42% :     0.000001s :      2: substitution.fold_const_symbol
 1.89% :     0.000005s :      4: substitution.graph_param_transform
69.26% :     0.000168s :      4: substitution.inline
 1.45% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.97% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.83% :     0.000004s :      2: substitution.replace_old_param
18.43% :     0.000045s :      3: substitution.tile_eliminate
 4.10% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.070482     2
97.26% :     0.068547s :      1: type_inference.infer
 2.74% :     0.001935s :      1: type_inference.specialize
------[replace.]   0.000074     6
66.18% :     0.000049s :      4: replace.inline
33.82% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000174     6
94.95% :     0.000165s :      4: match.inline
 5.05% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000141   834
 1.25% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.30% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.09% :     0.000002s :     12: predicate.addn_check_dump
 1.53% :     0.000002s :     12: predicate.addn_zero_filter
 2.35% :     0.000003s :     12: predicate.arithmetic_simplify
 1.35% :     0.000002s :     12: predicate.cast_eliminate
 0.42% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     12: predicate.compare_switch_simplify
 1.26% :     0.000002s :     12: predicate.depend_value_elim
 1.19% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      4: predicate.elim_not_effective
 0.51% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.11% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.14% :     0.000002s :     12: predicate.environ_get_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 1.04% :     0.000001s :      8: predicate.get_grad_eliminate
 0.27% :     0.000000s :      4: predicate.graph_param_transform
 5.13% :     0.000007s :     26: predicate.inline
 0.97% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.39% :     0.000002s :      8: predicate.less_batch_normalization
 1.48% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.79% :     0.000003s :     18: predicate.load_eliminater
 1.28% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.88% :     0.000005s :     38: predicate.loop_unroll_before_grad
 2.07% :     0.000003s :     16: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000002s :     12: predicate.merge_addn
 1.14% :     0.000002s :     12: predicate.minmaximum_grad
 1.41% :     0.000002s :      4: predicate.mutable_eliminate
 0.65% :     0.000001s :      4: predicate.opt_reshape
 2.23% :     0.000003s :     18: predicate.partial_eliminate
 1.18% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     12: predicate.reduce_eliminate
 1.55% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.85% :     0.000003s :     22: predicate.replace_applicator
 0.62% :     0.000001s :      8: predicate.replace_old_param
 0.34% :     0.000000s :      4: predicate.reset_defer_inline
 1.67% :     0.000002s :     12: predicate.reshape_eliminate
 1.33% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.16% :     0.000002s :     12: predicate.same_eliminate
 0.65% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.10% :     0.000002s :      8: predicate.special_op_eliminate
 1.10% :     0.000002s :      8: predicate.specialize_transform
 1.43% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     18: predicate.switch_defer_inline
 2.14% :     0.000003s :     18: predicate.switch_layer_defer_inline
 6.86% :     0.000010s :     60: predicate.switch_simplify
 1.80% :     0.000003s :     12: predicate.tile_eliminate
 1.30% :     0.000002s :     12: predicate.transpose_eliminate
 1.52% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.46% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     18: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000004s :     26: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     12: predicate.value_based_eliminate
 0.38% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.002417    26
70.93% :     0.001714s :     19: func_graph_cloner_run.FuncGraphClonerGraph
 1.08% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
27.99% :     0.000677s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.090342    76
 0.04% :     0.000038s :      1: add_recomputation
 0.07% :     0.000067s :      1: auto_monad
 0.01% :     0.000013s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.36% :     0.000326s :      1: bootstrap
 0.02% :     0.000020s :      1: cconv
 0.01% :     0.000008s :      1: convert_after_rewriter
 0.03% :     0.000026s :      1: cse_after_recomputation
 0.01% :     0.000009s :      1: environ_conv
 0.02% :     0.000021s :      1: event_method
 0.01% :     0.000010s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000008s :      1: graph_reusing
 8.09% :     0.007305s :      1: jit_opt_a
 0.17% :     0.000150s :      1: jit_opt_after_cconv
 0.06% :     0.000058s :      1: jit_opt_b
 0.47% :     0.000423s :      1: loop_unroll
 0.62% :     0.000558s :      1: mutable_eliminate
 1.12% :     0.001013s :     26: opt.transform.jit_opt_a
 0.06% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.03% :     0.000032s :      4: opt.transform.jit_opt_b
 0.01% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.01% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.04% :     0.000037s :      4: opt.transform.symbol_engine_opt
 0.55% :     0.000493s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.00% :     0.000004s :      1: pre_auto_parallel
 0.35% :     0.000314s :      1: py_interpret_to_execute
 0.01% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.01% :     0.000013s :      1: remove_dup_value
 1.62% :     0.001464s :      1: renormalize.infer
 0.79% :     0.000711s :      1: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.03% :     0.000031s :      1: rewriter_after_opt_a
 0.10% :     0.000090s :      1: rewriter_before_opt_a
 0.09% :     0.000081s :      1: symbol_engine_optimizer
 6.91% :     0.006243s :      1: task_emit
78.12% :     0.070572s :      1: type_inference
 0.06% :     0.000054s :      1: validate


TotalTime = 0.0165616, [30]
[bootstrap]: 0.00028961
[type_inference]: 0.0071257
[event_method]: 2.101e-05
[auto_monad]: 8.515e-05
[graph_reusing]: 6.58e-06
[pre_auto_parallel]: 2.50002e-06
[py_interpret_to_execute]: 4.981e-05
[rewriter_before_opt_a]: 8.473e-05
[expand_dump_flag]: 3.21001e-06
[jit_opt_a]: 0.00643362, [2]
    [Cycle 1]: 0.00208585, [27]
        [switch_simplify]: 0.00012932
        [loop_unroll]: 3.464e-05
        [a_1]: 0.00077413
        [with_stream_mark]: 2.276e-05
        [recompute_prepare]: 7.13e-06
        [updatestate_depend_eliminate]: 3.45e-06
        [updatestate_assign_eliminate]: 2.88e-06
        [updatestate_loads_eliminate]: 2.59001e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.62001e-06
        [meta_shard_fg_expand]: 2.40002e-06
        [get_grad_eliminate_]: 5.61e-06
        [merge_forward]: 3.76999e-06
        [cell_reuse_recompute_pass]: 1.10999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.477e-05
        [j_node_and_user_rematch]: 8.94e-06
        [meta_fg_expand]: 2.26998e-06
        [replace_old_param]: 9.29e-06
        [inline_without_move]: 5.89999e-06
        [renormalize]: 0.00078808
        [add_forward_monad_depend]: 7.27002e-06
        [auto_monad_grad]: 2.69999e-06
        [auto_monad_eliminator]: 1.505e-05
        [cse]: 2.697e-05
        [replace_applicator]: 1.454e-05
    [Cycle 2]: 0.0003337, [27]
        [switch_simplify]: 6.24999e-06
        [loop_unroll]: 5.21002e-06
        [a_1]: 9.318e-05
        [with_stream_mark]: 1.091e-05
        [recompute_prepare]: 5.04e-06
        [updatestate_depend_eliminate]: 2.88998e-06
        [updatestate_assign_eliminate]: 2.14999e-06
        [updatestate_loads_eliminate]: 1.95001e-06
        [parameter_eliminate]: 1.17999e-06
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 1.27999e-06
        [get_grad_eliminate_]: 4.50999e-06
        [merge_forward]: 3.21001e-06
        [cell_reuse_recompute_pass]: 1.72999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.404e-05
        [j_node_and_user_rematch]: 8.3e-06
        [meta_fg_expand]: 1.70001e-06
        [replace_old_param]: 7.16001e-06
        [inline_without_move]: 5.36002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.40999e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 5.30999e-06
        [cse]: 1.087e-05
        [replace_applicator]: 5.00001e-06
[py_interpret_to_execute_after_opt_a]: 1.277e-05
[rewriter_after_opt_a]: 3.731e-05
[convert_after_rewriter]: 7.00998e-06
[order_py_execute_after_rewriter]: 5.17e-06
[mutable_eliminate]: 0.00074082
[jit_opt_b]: 5.387e-05, [1]
    [Cycle 1]: 4.492e-05, [2]
        [frontend_op_eliminate]: 1.783e-05
        [inline_after_opt_a]: 1.549e-05
[cconv]: 2.723e-05
[loop_unroll]: 0.00045362
[jit_opt_after_cconv]: 0.00015139, [1]
    [Cycle 1]: 0.00014384, [11]
        [c_1]: 2.146e-05
        [parameter_eliminate]: 3.41001e-06
        [updatestate_depend_eliminate]: 7.06999e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.25002e-06
        [cse]: 2.351e-05
        [call_graph_tuple_transform]: 1.998e-05
        [tuple_list_get_item_eliminator]: 5.59998e-06
        [none_parameter_eliminate]: 1.60001e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 1.396e-05
[partial_unused_args_eliminate]: 2.47001e-06
[environ_conv]: 5.32001e-06
[add_recomputation]: 5.122e-05
[cse_after_recomputation]: 2.15e-05, [1]
    [Cycle 1]: 1.512e-05, [1]
        [cse]: 9.15999e-06
[auto_monad_reorder]: 1.528e-05
[get_jit_bprop_graph]: 1.94e-06
[rewriter_after_jit_bprop_graph]: 6.93e-06
[opt_after_jit_grad]: 0.00050454
[symbol_engine_optimizer]: 8.008e-05, [1]
    [Cycle 1]: 7.326e-05, [6]
        [build]: 5.64e-06
        [elim_shapecalc]: 8.12e-06
        [elim_not_effective]: 1.439e-05
        [opt_reshape]: 5.69999e-06
        [fold_const_symbol]: 9.32001e-06
        [renormalize]: 5.19998e-07
[validate]: 9.811e-05
Sums
    bootstrap                            :     0.000290s :  2.42%
    type_inference                       :     0.007126s : 59.63%
    event_method                         :     0.000021s :  0.18%
    auto_monad                           :     0.000085s :  0.71%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000050s :  0.42%
    rewriter_before_opt_a                :     0.000085s :  0.71%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000136s :  1.13%
    jit_opt_a.loop_unroll                :     0.000040s :  0.33%
    jit_opt_a.a_1                        :     0.000867s :  7.26%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.14%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.000788s :  6.60%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.17%
    jit_opt_a.cse                        :     0.000038s :  0.32%
    jit_opt_a.replace_applicator         :     0.000020s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.11%
    rewriter_after_opt_a                 :     0.000037s :  0.31%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000741s :  6.20%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.13%
    cconv                                :     0.000027s :  0.23%
    loop_unroll                          :     0.000454s :  3.80%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000051s :  0.43%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000015s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000505s :  4.22%
    symbol_engine_optimizer.build        :     0.000006s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000098s :  0.82%


Time group info:
------[substitution.]   0.000320    26
 0.60% :     0.000002s :      2: substitution.elim_not_effective
 0.38% :     0.000001s :      2: substitution.fold_const_symbol
 1.81% :     0.000006s :      3: substitution.graph_param_transform
86.02% :     0.000275s :      6: substitution.inline
 0.93% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.69% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.35% :     0.000004s :      2: substitution.replace_old_param
 2.87% :     0.000009s :      1: substitution.switch_simplify
 4.35% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007060     2
85.04% :     0.006004s :      1: type_inference.infer
14.96% :     0.001056s :      1: type_inference.specialize
------[replace.]   0.000139     9
48.08% :     0.000067s :      6: replace.inline
27.45% :     0.000038s :      1: replace.switch_simplify
24.47% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000293     9
92.80% :     0.000272s :      6: match.inline
 2.86% :     0.000008s :      1: match.switch_simplify
 4.34% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000174   801
 0.96% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.89% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.88% :     0.000002s :     12: predicate.addn_check_dump
 1.11% :     0.000002s :     12: predicate.addn_zero_filter
 1.64% :     0.000003s :     12: predicate.arithmetic_simplify
 1.18% :     0.000002s :     12: predicate.cast_eliminate
 0.30% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.87% :     0.000002s :     12: predicate.compare_switch_simplify
 1.05% :     0.000002s :     12: predicate.depend_value_elim
 0.89% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.00% :     0.000002s :     12: predicate.dict_get_item_eliminator
 0.93% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.71% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000001s :      3: predicate.elim_not_effective
 0.44% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.96% :     0.000002s :     12: predicate.environ_add_const_eliminate
 0.84% :     0.000001s :     12: predicate.environ_get_add_eliminate
 0.90% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.03% :     0.000002s :     12: predicate.environ_get_eliminate
 0.89% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.15% :     0.000000s :      3: predicate.fold_const_symbol
 0.73% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 4.33% :     0.000008s :     26: predicate.inline
 0.66% :     0.000001s :      6: predicate.inline_without_move
 0.24% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000002s :      6: predicate.less_batch_normalization
 1.19% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.27% :     0.000002s :     17: predicate.load_eliminater
 0.94% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.23% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.54% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 0.87% :     0.000002s :     12: predicate.merge_addn
23.42% :     0.000041s :     12: predicate.minmaximum_grad
 1.51% :     0.000003s :      3: predicate.mutable_eliminate
 0.37% :     0.000001s :      3: predicate.opt_reshape
 1.59% :     0.000003s :     17: predicate.partial_eliminate
 0.91% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.38% :     0.000002s :     12: predicate.reduce_eliminate
 1.23% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.19% :     0.000002s :     20: predicate.replace_applicator
 0.33% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.13% :     0.000002s :     12: predicate.reshape_eliminate
 0.99% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.02% :     0.000002s :     12: predicate.same_eliminate
 0.36% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.60% :     0.000001s :      6: predicate.special_op_eliminate
 0.69% :     0.000001s :      6: predicate.specialize_transform
 1.00% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.09% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.28% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.03% :     0.000004s :     20: predicate.switch_defer_inline
 1.74% :     0.000003s :     20: predicate.switch_layer_defer_inline
 6.39% :     0.000011s :     65: predicate.switch_simplify
 1.04% :     0.000002s :     12: predicate.tile_eliminate
 0.99% :     0.000002s :     12: predicate.transpose_eliminate
 1.19% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 0.97% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 2.81% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000003s :     12: predicate.tuple_list_set_item_eliminator
 1.12% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.27% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.02% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.16% :     0.000002s :     12: predicate.value_based_eliminate
 0.29% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.49% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000789    15
44.99% :     0.000355s :      6: func_graph_cloner_run.FuncGraphClonerGraph
15.11% :     0.000119s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.89% :     0.000315s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018510    72
 0.29% :     0.000054s :      1: add_recomputation
 0.48% :     0.000088s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 1.63% :     0.000302s :      1: bootstrap
 0.16% :     0.000030s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.13% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.13% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
34.78% :     0.006437s :      1: jit_opt_a
 0.83% :     0.000154s :      1: jit_opt_after_cconv
 0.31% :     0.000057s :      1: jit_opt_b
 2.50% :     0.000462s :      1: loop_unroll
 4.06% :     0.000751s :      1: mutable_eliminate
 6.27% :     0.001161s :     26: opt.transform.jit_opt_a
 0.26% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.78% :     0.000515s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.28% :     0.000053s :      1: py_interpret_to_execute
 0.08% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 2.37% :     0.000438s :      1: renormalize.infer
 1.84% :     0.000341s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000040s :      1: rewriter_after_opt_a
 0.47% :     0.000087s :      1: rewriter_before_opt_a
 0.45% :     0.000083s :      1: symbol_engine_optimizer
38.57% :     0.007139s :      1: type_inference


TotalTime = 0.0194154, [33]
[bootstrap]: 0.00029504
[type_inference]: 0.00478438
[event_method]: 1.719e-05
[auto_monad]: 6.633e-05
[graph_reusing]: 5.57001e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 3.608e-05
[rewriter_before_opt_a]: 7.202e-05
[expand_dump_flag]: 3.24001e-06
[jit_opt_a]: 0.00559559, [2]
    [Cycle 1]: 0.00166863, [27]
        [switch_simplify]: 5.983e-05
        [loop_unroll]: 2.973e-05
        [a_1]: 0.00053851
        [with_stream_mark]: 1.885e-05
        [recompute_prepare]: 7.78001e-06
        [updatestate_depend_eliminate]: 3.72998e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 2.76999e-06
        [parameter_eliminate]: 1.82999e-06
        [specialize_transform]: 6.16e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 5.23002e-06
        [meta_shard_fg_expand]: 2.16e-06
        [get_grad_eliminate_]: 5.27001e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.498e-05
        [j_node_and_user_rematch]: 8.57e-06
        [meta_fg_expand]: 2.34999e-06
        [replace_old_param]: 9.32999e-06
        [inline_without_move]: 5.45001e-06
        [renormalize]: 0.00068912
        [add_forward_monad_depend]: 6.00002e-06
        [auto_monad_grad]: 2.44999e-06
        [auto_monad_eliminator]: 1.618e-05
        [cse]: 2.543e-05
        [replace_applicator]: 1.303e-05
    [Cycle 2]: 0.0003504, [27]
        [switch_simplify]: 5.66e-06
        [loop_unroll]: 4.99998e-06
        [a_1]: 0.00010358
        [with_stream_mark]: 1.246e-05
        [recompute_prepare]: 5.19e-06
        [updatestate_depend_eliminate]: 3.26999e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 5.00999e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 5.40999e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 4.63001e-06
        [merge_forward]: 3.16001e-06
        [cell_reuse_recompute_pass]: 1.81998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.498e-05
        [j_node_and_user_rematch]: 8.18001e-06
        [meta_fg_expand]: 1.85001e-06
        [replace_old_param]: 7.18e-06
        [inline_without_move]: 4.57e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.77999e-06
        [auto_monad_grad]: 1.22e-06
        [auto_monad_eliminator]: 6.71e-06
        [cse]: 1.11e-05
        [replace_applicator]: 5.50001e-06
[py_interpret_to_execute_after_opt_a]: 2.453e-05
[rewriter_after_opt_a]: 3.752e-05
[convert_after_rewriter]: 6.91999e-06
[order_py_execute_after_rewriter]: 5.22e-06
[mutable_eliminate]: 0.00062908
[jit_opt_b]: 5.559e-05, [1]
    [Cycle 1]: 4.776e-05, [2]
        [frontend_op_eliminate]: 2.072e-05
        [inline_after_opt_a]: 1.514e-05
[cconv]: 2.484e-05
[loop_unroll]: 0.00043101
[jit_opt_after_cconv]: 0.00014068, [1]
    [Cycle 1]: 0.00013351, [11]
        [c_1]: 2.017e-05
        [parameter_eliminate]: 2.96999e-06
        [updatestate_depend_eliminate]: 5.49998e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.92e-05
        [call_graph_tuple_transform]: 1.91e-05
        [tuple_list_get_item_eliminator]: 5.53002e-06
        [none_parameter_eliminate]: 1.32e-06
        [renormalize]: 5.79981e-07
        [switch_simplify]: 5.46998e-06
[remove_dup_value]: 1.422e-05
[partial_unused_args_eliminate]: 2.69999e-06
[environ_conv]: 5.30001e-06
[add_recomputation]: 4.994e-05
[cse_after_recomputation]: 3.003e-05, [1]
    [Cycle 1]: 2.366e-05, [1]
        [cse]: 9.54e-06
[auto_monad_reorder]: 1.581e-05
[get_jit_bprop_graph]: 1.82999e-06
[rewriter_after_jit_bprop_graph]: 6.16998e-06
[opt_after_jit_grad]: 0.00047603
[symbol_engine_optimizer]: 7.496e-05, [1]
    [Cycle 1]: 6.771e-05, [6]
        [build]: 3.95e-06
        [elim_shapecalc]: 7.56999e-06
        [elim_not_effective]: 1.315e-05
        [opt_reshape]: 5.80002e-06
        [fold_const_symbol]: 8.82e-06
        [renormalize]: 8.39995e-07
[validate]: 3.511e-05
[backend_pass]: 1.27e-06
[task_emit]: 0.00627489
[execute]: 7.58999e-06
Sums
    bootstrap                            :     0.000295s :  1.94%
    type_inference                       :     0.004784s : 31.43%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000066s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.24%
    rewriter_before_opt_a                :     0.000072s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.43%
    jit_opt_a.loop_unroll                :     0.000035s :  0.23%
    jit_opt_a.a_1                        :     0.000642s :  4.22%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000689s :  4.53%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.15%
    jit_opt_a.cse                        :     0.000037s :  0.24%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000025s :  0.16%
    rewriter_after_opt_a                 :     0.000038s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000629s :  4.13%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000025s :  0.16%
    loop_unroll                          :     0.000431s :  2.83%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000476s :  3.13%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000035s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006275s : 41.23%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000210    23
 1.07% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.54% :     0.000005s :      3: substitution.graph_param_transform
84.23% :     0.000177s :      4: substitution.inline
 1.61% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.17% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.79% :     0.000004s :      2: substitution.replace_old_param
 6.00% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004722     2
82.49% :     0.003895s :      1: type_inference.infer
17.51% :     0.000827s :      1: type_inference.specialize
------[replace.]   0.000068     6
68.58% :     0.000047s :      4: replace.inline
31.42% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000186     6
93.87% :     0.000175s :      4: match.inline
 6.13% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 2.11% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.60% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.41% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.48% :     0.000001s :      3: predicate.graph_param_transform
 5.61% :     0.000007s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.87% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.93% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.15% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.78% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.42% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.78% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.39% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.90% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.99% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.89% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000552    12
49.04% :     0.000270s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.19% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.77% :     0.000247s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021073    76
 0.25% :     0.000053s :      1: add_recomputation
 0.33% :     0.000070s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.46% :     0.000308s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000032s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.57% :     0.005599s :      1: jit_opt_a
 0.68% :     0.000143s :      1: jit_opt_after_cconv
 0.28% :     0.000058s :      1: jit_opt_b
 2.08% :     0.000439s :      1: loop_unroll
 3.03% :     0.000639s :      1: mutable_eliminate
 4.07% :     0.000858s :     26: opt.transform.jit_opt_a
 0.22% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000029s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.30% :     0.000484s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000039s :      1: py_interpret_to_execute
 0.13% :     0.000027s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.74% :     0.000367s :      1: renormalize.infer
 1.49% :     0.000314s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000041s :      1: rewriter_after_opt_a
 0.35% :     0.000075s :      1: rewriter_before_opt_a
 0.37% :     0.000078s :      1: symbol_engine_optimizer
29.84% :     0.006288s :      1: task_emit
22.75% :     0.004795s :      1: type_inference
 0.26% :     0.000054s :      1: validate


TotalTime = 0.0149971, [30]
[bootstrap]: 0.00028495
[type_inference]: 0.00629978
[event_method]: 2.06e-05
[auto_monad]: 8.324e-05
[graph_reusing]: 6.31e-06
[pre_auto_parallel]: 2.54999e-06
[py_interpret_to_execute]: 4.796e-05
[rewriter_before_opt_a]: 8.338e-05
[expand_dump_flag]: 3.34001e-06
[jit_opt_a]: 0.00574845, [2]
    [Cycle 1]: 0.00200287, [27]
        [switch_simplify]: 0.00012592
        [loop_unroll]: 3.34e-05
        [a_1]: 0.00068525
        [with_stream_mark]: 2.055e-05
        [recompute_prepare]: 7.53e-06
        [updatestate_depend_eliminate]: 3.66999e-06
        [updatestate_assign_eliminate]: 3.28998e-06
        [updatestate_loads_eliminate]: 2.73998e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 5.80002e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 3.784e-05
        [meta_shard_fg_expand]: 2.37999e-06
        [get_grad_eliminate_]: 5.82999e-06
        [merge_forward]: 4.19997e-06
        [cell_reuse_recompute_pass]: 2.08002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.548e-05
        [j_node_and_user_rematch]: 9.27999e-06
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 9.56e-06
        [inline_without_move]: 5.57001e-06
        [renormalize]: 0.00075385
        [add_forward_monad_depend]: 6.28002e-06
        [auto_monad_grad]: 2.53e-06
        [auto_monad_eliminator]: 1.619e-05
        [cse]: 2.55e-05
        [replace_applicator]: 1.518e-05
    [Cycle 2]: 0.00036584, [27]
        [switch_simplify]: 6.34001e-06
        [loop_unroll]: 4.99e-06
        [a_1]: 0.00010119
        [with_stream_mark]: 1.223e-05
        [recompute_prepare]: 5.48002e-06
        [updatestate_depend_eliminate]: 2.71e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.84999e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 5.57001e-06
        [updatestate_useless_node_eliminater]: 5.43002e-06
        [accelerated_algorithm]: 5.75001e-06
        [meta_shard_fg_expand]: 1.54998e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 2.86999e-06
        [cell_reuse_recompute_pass]: 2.39001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.623e-05
        [j_node_and_user_rematch]: 9.91e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 8.77e-06
        [inline_without_move]: 5.26998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.40999e-06
        [auto_monad_grad]: 8.89995e-07
        [auto_monad_eliminator]: 5.46e-06
        [cse]: 1.232e-05
        [replace_applicator]: 5.96e-06
[py_interpret_to_execute_after_opt_a]: 1.216e-05
[rewriter_after_opt_a]: 3.932e-05
[convert_after_rewriter]: 7.51999e-06
[order_py_execute_after_rewriter]: 5.46e-06
[mutable_eliminate]: 0.0007348
[jit_opt_b]: 5.685e-05, [1]
    [Cycle 1]: 4.836e-05, [2]
        [frontend_op_eliminate]: 1.753e-05
        [inline_after_opt_a]: 1.746e-05
[cconv]: 2.81e-05
[loop_unroll]: 0.00049431
[jit_opt_after_cconv]: 0.00014771, [1]
    [Cycle 1]: 0.00014024, [11]
        [c_1]: 2.167e-05
        [parameter_eliminate]: 2.81999e-06
        [updatestate_depend_eliminate]: 6.89999e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.26998e-06
        [cse]: 2.157e-05
        [call_graph_tuple_transform]: 1.92e-05
        [tuple_list_get_item_eliminator]: 5.47999e-06
        [none_parameter_eliminate]: 1.86998e-06
        [renormalize]: 8.50006e-07
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 1.433e-05
[partial_unused_args_eliminate]: 2.62001e-06
[environ_conv]: 5.18002e-06
[add_recomputation]: 4.862e-05
[cse_after_recomputation]: 2.002e-05, [1]
    [Cycle 1]: 1.444e-05, [1]
        [cse]: 9.03002e-06
[auto_monad_reorder]: 1.598e-05
[get_jit_bprop_graph]: 1.84e-06
[rewriter_after_jit_bprop_graph]: 6.12999e-06
[opt_after_jit_grad]: 0.00049572
[symbol_engine_optimizer]: 7.341e-05, [1]
    [Cycle 1]: 6.692e-05, [6]
        [build]: 3.8e-06
        [elim_shapecalc]: 7.77002e-06
        [elim_not_effective]: 1.422e-05
        [opt_reshape]: 5.59e-06
        [fold_const_symbol]: 8.28001e-06
        [renormalize]: 4.69998e-07
[validate]: 3.428e-05
Sums
    bootstrap                            :     0.000285s :  2.59%
    type_inference                       :     0.006300s : 57.28%
    event_method                         :     0.000021s :  0.19%
    auto_monad                           :     0.000083s :  0.76%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000048s :  0.44%
    rewriter_before_opt_a                :     0.000083s :  0.76%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000132s :  1.20%
    jit_opt_a.loop_unroll                :     0.000038s :  0.35%
    jit_opt_a.a_1                        :     0.000786s :  7.15%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000044s :  0.40%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.17%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000754s :  6.85%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.20%
    jit_opt_a.cse                        :     0.000038s :  0.34%
    jit_opt_a.replace_applicator         :     0.000021s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.11%
    rewriter_after_opt_a                 :     0.000039s :  0.36%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000735s :  6.68%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.16%
    cconv                                :     0.000028s :  0.26%
    loop_unroll                          :     0.000494s :  4.49%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.13%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000049s :  0.44%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000496s :  4.51%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.31%


Time group info:
------[substitution.]   0.000278    26
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000002s :      2: substitution.fold_const_symbol
 2.15% :     0.000006s :      3: substitution.graph_param_transform
83.32% :     0.000232s :      6: substitution.inline
 1.42% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.88% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.94% :     0.000005s :      2: substitution.replace_old_param
 3.03% :     0.000008s :      1: substitution.switch_simplify
 4.85% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006233     2
83.35% :     0.005195s :      1: type_inference.infer
16.65% :     0.001038s :      1: type_inference.specialize
------[replace.]   0.000140     9
49.06% :     0.000069s :      6: replace.inline
28.88% :     0.000040s :      1: replace.switch_simplify
22.07% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000248     9
91.93% :     0.000228s :      6: match.inline
 3.06% :     0.000008s :      1: match.switch_simplify
 5.01% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   801
 1.33% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.04% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000002s :     12: predicate.addn_check_dump
 1.38% :     0.000002s :     12: predicate.addn_zero_filter
 2.47% :     0.000003s :     12: predicate.arithmetic_simplify
 1.29% :     0.000002s :     12: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.30% :     0.000002s :     12: predicate.compare_switch_simplify
 1.63% :     0.000002s :     12: predicate.depend_value_elim
 1.13% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.44% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.21% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.11% :     0.000002s :     12: predicate.environ_get_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.34% :     0.000000s :      3: predicate.graph_param_transform
 5.79% :     0.000008s :     26: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     17: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.04% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     12: predicate.merge_addn
 1.09% :     0.000001s :     12: predicate.minmaximum_grad
 2.06% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     17: predicate.partial_eliminate
 1.22% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     12: predicate.reduce_eliminate
 1.44% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.49% :     0.000002s :     20: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.38% :     0.000001s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     12: predicate.reshape_eliminate
 1.23% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     12: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.31% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.65% :     0.000004s :     20: predicate.switch_defer_inline
 2.18% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.01% :     0.000011s :     65: predicate.switch_simplify
 1.25% :     0.000002s :     12: predicate.tile_eliminate
 1.61% :     0.000002s :     12: predicate.transpose_eliminate
 1.38% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.82% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.63% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.47% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.85% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000773    15
49.56% :     0.000383s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.52% :     0.000089s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.92% :     0.000301s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016936    72
 0.30% :     0.000051s :      1: add_recomputation
 0.52% :     0.000088s :      1: auto_monad
 0.11% :     0.000019s :      1: auto_monad_reorder
 1.75% :     0.000297s :      1: bootstrap
 0.19% :     0.000032s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.13% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.97% :     0.005752s :      1: jit_opt_a
 0.89% :     0.000150s :      1: jit_opt_after_cconv
 0.35% :     0.000060s :      1: jit_opt_b
 2.97% :     0.000503s :      1: loop_unroll
 4.40% :     0.000745s :      1: mutable_eliminate
 6.59% :     0.001116s :     26: opt.transform.jit_opt_a
 0.28% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000027s :      4: opt.transform.jit_opt_b
 0.08% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.98% :     0.000504s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.30% :     0.000051s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.40% :     0.000406s :      1: renormalize.infer
 2.01% :     0.000340s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000043s :      1: rewriter_after_opt_a
 0.51% :     0.000086s :      1: rewriter_before_opt_a
 0.45% :     0.000076s :      1: symbol_engine_optimizer
37.29% :     0.006314s :      1: type_inference


TotalTime = 0.0184698, [33]
[bootstrap]: 0.00029022
[type_inference]: 0.00470682
[event_method]: 1.749e-05
[auto_monad]: 6.413e-05
[graph_reusing]: 5.69e-06
[pre_auto_parallel]: 2.37999e-06
[py_interpret_to_execute]: 3.134e-05
[rewriter_before_opt_a]: 6.914e-05
[expand_dump_flag]: 3.09999e-06
[jit_opt_a]: 0.00546966, [2]
    [Cycle 1]: 0.00162193, [27]
        [switch_simplify]: 5.333e-05
        [loop_unroll]: 2.97e-05
        [a_1]: 0.00054777
        [with_stream_mark]: 1.933e-05
        [recompute_prepare]: 8.51002e-06
        [updatestate_depend_eliminate]: 4.01001e-06
        [updatestate_assign_eliminate]: 2.99001e-06
        [updatestate_loads_eliminate]: 2.52001e-06
        [parameter_eliminate]: 2.37999e-06
        [specialize_transform]: 6.24001e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.26002e-06
        [meta_shard_fg_expand]: 2.41998e-06
        [get_grad_eliminate_]: 5.06002e-06
        [merge_forward]: 3.80998e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.353e-05
        [j_node_and_user_rematch]: 9.52001e-06
        [meta_fg_expand]: 2.44001e-06
        [replace_old_param]: 8.07e-06
        [inline_without_move]: 5.25999e-06
        [renormalize]: 0.00065489
        [add_forward_monad_depend]: 6.06998e-06
        [auto_monad_grad]: 2.66e-06
        [auto_monad_eliminator]: 1.513e-05
        [cse]: 1.906e-05
        [replace_applicator]: 1.165e-05
    [Cycle 2]: 0.00032366, [27]
        [switch_simplify]: 5.44e-06
        [loop_unroll]: 5.09e-06
        [a_1]: 9.095e-05
        [with_stream_mark]: 1.051e-05
        [recompute_prepare]: 5.17e-06
        [updatestate_depend_eliminate]: 2.69001e-06
        [updatestate_assign_eliminate]: 2.39999e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.82e-06
        [updatestate_useless_node_eliminater]: 4.89998e-06
        [accelerated_algorithm]: 4.94e-06
        [meta_shard_fg_expand]: 1.39998e-06
        [get_grad_eliminate_]: 4.4e-06
        [merge_forward]: 2.41e-06
        [cell_reuse_recompute_pass]: 1.96003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.334e-05
        [j_node_and_user_rematch]: 7.6e-06
        [meta_fg_expand]: 1.76e-06
        [replace_old_param]: 7.35998e-06
        [inline_without_move]: 4.75001e-06
        [renormalize]: 1.09983e-07
        [add_forward_monad_depend]: 1.37999e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 5.20999e-06
        [cse]: 1.053e-05
        [replace_applicator]: 5.12999e-06
[py_interpret_to_execute_after_opt_a]: 1.096e-05
[rewriter_after_opt_a]: 3.452e-05
[convert_after_rewriter]: 7.23e-06
[order_py_execute_after_rewriter]: 5.51e-06
[mutable_eliminate]: 0.00060049
[jit_opt_b]: 5.02e-05, [1]
    [Cycle 1]: 4.223e-05, [2]
        [frontend_op_eliminate]: 1.603e-05
        [inline_after_opt_a]: 1.494e-05
[cconv]: 2.625e-05
[loop_unroll]: 0.00043184
[jit_opt_after_cconv]: 0.00013983, [1]
    [Cycle 1]: 0.00013291, [11]
        [c_1]: 2.067e-05
        [parameter_eliminate]: 2.88e-06
        [updatestate_depend_eliminate]: 5.87999e-06
        [updatestate_assign_eliminate]: 2.40002e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 1.841e-05
        [call_graph_tuple_transform]: 1.853e-05
        [tuple_list_get_item_eliminator]: 5.33002e-06
        [none_parameter_eliminate]: 1.81e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.42001e-06
[remove_dup_value]: 1.377e-05
[partial_unused_args_eliminate]: 2.24001e-06
[environ_conv]: 5.35999e-06
[add_recomputation]: 4.697e-05
[cse_after_recomputation]: 1.991e-05, [1]
    [Cycle 1]: 1.473e-05, [1]
        [cse]: 9.07999e-06
[auto_monad_reorder]: 1.573e-05
[get_jit_bprop_graph]: 2.07999e-06
[rewriter_after_jit_bprop_graph]: 4.50001e-06
[opt_after_jit_grad]: 0.00047713
[symbol_engine_optimizer]: 7.437e-05, [1]
    [Cycle 1]: 6.753e-05, [6]
        [build]: 3.81001e-06
        [elim_shapecalc]: 7.58999e-06
        [elim_not_effective]: 1.308e-05
        [opt_reshape]: 5.84e-06
        [fold_const_symbol]: 9.17001e-06
        [renormalize]: 4.90021e-07
[validate]: 3.83e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.00558982
[execute]: 9.47001e-06
Sums
    bootstrap                            :     0.000290s :  2.02%
    type_inference                       :     0.004707s : 32.84%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000064s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.22%
    rewriter_before_opt_a                :     0.000069s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.41%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000639s :  4.46%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000655s :  4.57%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.14%
    jit_opt_a.cse                        :     0.000030s :  0.21%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.08%
    rewriter_after_opt_a                 :     0.000035s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000600s :  4.19%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000026s :  0.18%
    loop_unroll                          :     0.000432s :  3.01%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000477s :  3.33%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005590s : 39.00%
    execute                              :     0.000009s :  0.07%


Time group info:
------[substitution.]   0.000203    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.43% :     0.000005s :      3: substitution.graph_param_transform
84.27% :     0.000171s :      4: substitution.inline
 1.50% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.22% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000004s :      2: substitution.replace_old_param
 6.05% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004644     2
83.18% :     0.003863s :      1: type_inference.infer
16.82% :     0.000781s :      1: type_inference.specialize
------[replace.]   0.000065     6
64.39% :     0.000042s :      4: replace.inline
35.61% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000180     6
93.79% :     0.000168s :      4: match.inline
 6.21% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
12.71% :     0.000016s :     10: predicate.accumulaten_eliminater
 0.97% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 1.88% :     0.000002s :     10: predicate.arithmetic_simplify
 1.32% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.05% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.49% :     0.000001s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.03% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     10: predicate.environ_get_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 4.99% :     0.000006s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      6: predicate.less_batch_normalization
 1.34% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.61% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.83% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 0.93% :     0.000001s :     10: predicate.minmaximum_grad
 1.31% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 1.95% :     0.000003s :     15: predicate.partial_eliminate
 1.04% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.23% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.45% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.07% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.80% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.20% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.13% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.56% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.19% :     0.000003s :     16: predicate.switch_defer_inline
 1.91% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.55% :     0.000008s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.27% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.08% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.35% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.33% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.31% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.59% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.52% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000571    12
48.55% :     0.000277s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.06% :     0.000046s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.39% :     0.000248s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020071    76
 0.25% :     0.000050s :      1: add_recomputation
 0.34% :     0.000068s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.53% :     0.000308s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.27% :     0.005473s :      1: jit_opt_a
 0.71% :     0.000143s :      1: jit_opt_after_cconv
 0.26% :     0.000053s :      1: jit_opt_b
 2.19% :     0.000440s :      1: loop_unroll
 3.04% :     0.000610s :      1: mutable_eliminate
 4.21% :     0.000844s :     26: opt.transform.jit_opt_a
 0.23% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.42% :     0.000485s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000034s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.68% :     0.000338s :      1: renormalize.infer
 1.54% :     0.000309s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000037s :      1: rewriter_after_opt_a
 0.36% :     0.000072s :      1: rewriter_before_opt_a
 0.38% :     0.000077s :      1: symbol_engine_optimizer
27.94% :     0.005608s :      1: task_emit
23.51% :     0.004719s :      1: type_inference
 0.28% :     0.000057s :      1: validate


TotalTime = 0.0144846, [30]
[bootstrap]: 0.00030121
[type_inference]: 0.00637626
[event_method]: 2.078e-05
[auto_monad]: 6.141e-05
[graph_reusing]: 4.49002e-06
[pre_auto_parallel]: 2.14999e-06
[py_interpret_to_execute]: 4.712e-05
[rewriter_before_opt_a]: 7.677e-05
[expand_dump_flag]: 2.25002e-06
[jit_opt_a]: 0.00541138, [2]
    [Cycle 1]: 0.00176599, [27]
        [switch_simplify]: 0.00010031
        [loop_unroll]: 3.513e-05
        [a_1]: 0.00068912
        [with_stream_mark]: 1.41e-05
        [recompute_prepare]: 7.7e-06
        [updatestate_depend_eliminate]: 3.41001e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [parameter_eliminate]: 1.16002e-06
        [specialize_transform]: 6.33e-06
        [updatestate_useless_node_eliminater]: 5.26998e-06
        [accelerated_algorithm]: 5.94999e-06
        [meta_shard_fg_expand]: 1.41998e-06
        [get_grad_eliminate_]: 5.90002e-06
        [merge_forward]: 2.49001e-06
        [cell_reuse_recompute_pass]: 1.30001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.22e-05
        [j_node_and_user_rematch]: 9.39998e-06
        [meta_fg_expand]: 1.99999e-06
        [replace_old_param]: 8.13999e-06
        [inline_without_move]: 5.45001e-06
        [renormalize]: 0.0006252
        [add_forward_monad_depend]: 5.42999e-06
        [auto_monad_grad]: 1.66e-06
        [auto_monad_eliminator]: 1.03e-05
        [cse]: 1.636e-05
        [replace_applicator]: 1.232e-05
    [Cycle 2]: 0.00032439, [27]
        [switch_simplify]: 5.81e-06
        [loop_unroll]: 5.10999e-06
        [a_1]: 9.492e-05
        [with_stream_mark]: 9.25001e-06
        [recompute_prepare]: 5.00999e-06
        [updatestate_depend_eliminate]: 2.83998e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 1.14998e-06
        [specialize_transform]: 5.12e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 4.87998e-06
        [meta_shard_fg_expand]: 1.41002e-06
        [get_grad_eliminate_]: 4.88001e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.85001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.348e-05
        [j_node_and_user_rematch]: 8.02e-06
        [meta_fg_expand]: 1.85001e-06
        [replace_old_param]: 7.7e-06
        [inline_without_move]: 5.25999e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 4.90001e-06
        [cse]: 9.40001e-06
        [replace_applicator]: 5.17999e-06
[py_interpret_to_execute_after_opt_a]: 8.52e-06
[rewriter_after_opt_a]: 2.654e-05
[convert_after_rewriter]: 5.46998e-06
[order_py_execute_after_rewriter]: 3.81001e-06
[mutable_eliminate]: 0.00056061
[jit_opt_b]: 6.316e-05, [1]
    [Cycle 1]: 4.433e-05, [2]
        [frontend_op_eliminate]: 1.648e-05
        [inline_after_opt_a]: 1.569e-05
[cconv]: 1.788e-05
[loop_unroll]: 0.00045108
[jit_opt_after_cconv]: 0.00014445, [1]
    [Cycle 1]: 0.00013784, [11]
        [c_1]: 2.32e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 5.94999e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [cse]: 1.911e-05
        [call_graph_tuple_transform]: 1.741e-05
        [tuple_list_get_item_eliminator]: 5.38002e-06
        [none_parameter_eliminate]: 9.89996e-07
        [renormalize]: 6.69999e-07
        [switch_simplify]: 5.82001e-06
[remove_dup_value]: 9.12999e-06
[partial_unused_args_eliminate]: 2.06e-06
[environ_conv]: 3.9e-06
[add_recomputation]: 3.579e-05
[cse_after_recomputation]: 2.158e-05, [1]
    [Cycle 1]: 1.562e-05, [1]
        [cse]: 9.74e-06
[auto_monad_reorder]: 1.013e-05
[get_jit_bprop_graph]: 1.58002e-06
[rewriter_after_jit_bprop_graph]: 4.82e-06
[opt_after_jit_grad]: 0.00054291
[symbol_engine_optimizer]: 7.495e-05, [1]
    [Cycle 1]: 6.878e-05, [6]
        [build]: 3.01999e-06
        [elim_shapecalc]: 8.42e-06
        [elim_not_effective]: 1.323e-05
        [opt_reshape]: 6.34999e-06
        [fold_const_symbol]: 8.64e-06
        [renormalize]: 3.50003e-07
[validate]: 2.888e-05
Sums
    bootstrap                            :     0.000301s :  2.85%
    type_inference                       :     0.006376s : 60.28%
    event_method                         :     0.000021s :  0.20%
    auto_monad                           :     0.000061s :  0.58%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000047s :  0.45%
    rewriter_before_opt_a                :     0.000077s :  0.73%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000106s :  1.00%
    jit_opt_a.loop_unroll                :     0.000040s :  0.38%
    jit_opt_a.a_1                        :     0.000784s :  7.41%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000625s :  5.91%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.14%
    jit_opt_a.cse                        :     0.000026s :  0.24%
    jit_opt_a.replace_applicator         :     0.000017s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.08%
    rewriter_after_opt_a                 :     0.000027s :  0.25%
    convert_after_rewriter               :     0.000005s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000561s :  5.30%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.15%
    cconv                                :     0.000018s :  0.17%
    loop_unroll                          :     0.000451s :  4.26%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000009s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000036s :  0.34%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000010s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000543s :  5.13%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.27%


Time group info:
------[substitution.]   0.000246    26
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.44% :     0.000001s :      2: substitution.fold_const_symbol
 1.68% :     0.000004s :      3: substitution.graph_param_transform
85.83% :     0.000211s :      6: substitution.inline
 1.24% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.75% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.32% :     0.000003s :      2: substitution.replace_old_param
 3.31% :     0.000008s :      1: substitution.switch_simplify
 3.67% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006320     2
82.95% :     0.005242s :      1: type_inference.infer
17.05% :     0.001077s :      1: type_inference.specialize
------[replace.]   0.000134     9
58.45% :     0.000078s :      6: replace.inline
17.33% :     0.000023s :      1: replace.switch_simplify
24.21% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000222     9
93.21% :     0.000207s :      6: match.inline
 3.33% :     0.000007s :      1: match.switch_simplify
 3.47% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   801
 1.38% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.00% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     12: predicate.addn_check_dump
 1.64% :     0.000002s :     12: predicate.addn_zero_filter
 2.21% :     0.000003s :     12: predicate.arithmetic_simplify
 1.29% :     0.000002s :     12: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.21% :     0.000002s :     12: predicate.compare_switch_simplify
 1.42% :     0.000002s :     12: predicate.depend_value_elim
 1.21% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.23% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     12: predicate.environ_get_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.46% :     0.000007s :     26: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     17: predicate.load_eliminater
 1.28% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.20% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.99% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000002s :     12: predicate.merge_addn
 1.15% :     0.000002s :     12: predicate.minmaximum_grad
 1.70% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     17: predicate.partial_eliminate
 1.47% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.87% :     0.000003s :     12: predicate.reduce_eliminate
 1.44% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     20: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     12: predicate.reshape_eliminate
 1.30% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.62% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.73% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.75% :     0.000004s :     20: predicate.switch_defer_inline
 2.42% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.43% :     0.000011s :     65: predicate.switch_simplify
 1.28% :     0.000002s :     12: predicate.tile_eliminate
 1.24% :     0.000002s :     12: predicate.transpose_eliminate
 1.35% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.38% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000002s :     12: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000751    15
47.89% :     0.000360s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.29% :     0.000085s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.82% :     0.000307s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016230    72
 0.24% :     0.000039s :      1: add_recomputation
 0.41% :     0.000066s :      1: auto_monad
 0.08% :     0.000013s :      1: auto_monad_reorder
 1.95% :     0.000317s :      1: bootstrap
 0.13% :     0.000021s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.15% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
33.36% :     0.005415s :      1: jit_opt_a
 0.91% :     0.000148s :      1: jit_opt_after_cconv
 0.41% :     0.000066s :      1: jit_opt_b
 2.83% :     0.000459s :      1: loop_unroll
 3.51% :     0.000569s :      1: mutable_eliminate
 6.46% :     0.001048s :     26: opt.transform.jit_opt_a
 0.30% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.39% :     0.000551s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.31% :     0.000050s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000012s :      1: remove_dup_value
 2.04% :     0.000331s :      1: renormalize.infer
 1.77% :     0.000287s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000030s :      1: rewriter_after_opt_a
 0.49% :     0.000080s :      1: rewriter_before_opt_a
 0.48% :     0.000077s :      1: symbol_engine_optimizer
39.36% :     0.006389s :      1: type_inference


TotalTime = 0.0170015, [33]
[bootstrap]: 0.00025156
[type_inference]: 0.00406432
[event_method]: 1.785e-05
[auto_monad]: 5.396e-05
[graph_reusing]: 5.14e-06
[pre_auto_parallel]: 2.27001e-06
[py_interpret_to_execute]: 3.401e-05
[rewriter_before_opt_a]: 7.019e-05
[expand_dump_flag]: 2.91e-06
[jit_opt_a]: 0.00499332, [2]
    [Cycle 1]: 0.00146258, [27]
        [switch_simplify]: 4.767e-05
        [loop_unroll]: 2.954e-05
        [a_1]: 0.00053028
        [with_stream_mark]: 1.421e-05
        [recompute_prepare]: 7.99997e-06
        [updatestate_depend_eliminate]: 3.02002e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [parameter_eliminate]: 1.27999e-06
        [specialize_transform]: 6.58e-06
        [updatestate_useless_node_eliminater]: 5.22999e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 1.45001e-06
        [get_grad_eliminate_]: 5.25001e-06
        [merge_forward]: 3.03e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.132e-05
        [j_node_and_user_rematch]: 9.32999e-06
        [meta_fg_expand]: 2.32999e-06
        [replace_old_param]: 8.55001e-06
        [inline_without_move]: 5.35001e-06
        [renormalize]: 0.0005445
        [add_forward_monad_depend]: 4.02e-06
        [auto_monad_grad]: 1.92001e-06
        [auto_monad_eliminator]: 1.049e-05
        [cse]: 1.923e-05
        [replace_applicator]: 1.144e-05
    [Cycle 2]: 0.00032311, [27]
        [switch_simplify]: 5.84e-06
        [loop_unroll]: 4.81997e-06
        [a_1]: 9.006e-05
        [with_stream_mark]: 9.54e-06
        [recompute_prepare]: 4.74998e-06
        [updatestate_depend_eliminate]: 2.51e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 5.14e-06
        [updatestate_useless_node_eliminater]: 5.04998e-06
        [accelerated_algorithm]: 5.02e-06
        [meta_shard_fg_expand]: 1.22999e-06
        [get_grad_eliminate_]: 4.4e-06
        [merge_forward]: 2.89999e-06
        [cell_reuse_recompute_pass]: 2.00002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.258e-05
        [j_node_and_user_rematch]: 7.91001e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 7.58001e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 5.58002e-06
        [cse]: 1.133e-05
        [replace_applicator]: 4.93001e-06
[py_interpret_to_execute_after_opt_a]: 8.25e-06
[rewriter_after_opt_a]: 2.989e-05
[convert_after_rewriter]: 6.01e-06
[order_py_execute_after_rewriter]: 5.27001e-06
[mutable_eliminate]: 0.00051131
[jit_opt_b]: 4.806e-05, [1]
    [Cycle 1]: 4.131e-05, [2]
        [frontend_op_eliminate]: 1.566e-05
        [inline_after_opt_a]: 1.458e-05
[cconv]: 1.879e-05
[loop_unroll]: 0.00043303
[jit_opt_after_cconv]: 0.00013521, [1]
    [Cycle 1]: 0.0001288, [11]
        [c_1]: 2.028e-05
        [parameter_eliminate]: 2.32999e-06
        [updatestate_depend_eliminate]: 5.40999e-06
        [updatestate_assign_eliminate]: 2.52001e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 1.608e-05
        [call_graph_tuple_transform]: 1.788e-05
        [tuple_list_get_item_eliminator]: 5.76e-06
        [none_parameter_eliminate]: 9.39996e-07
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.01002e-06
[remove_dup_value]: 1.039e-05
[partial_unused_args_eliminate]: 1.74e-06
[environ_conv]: 3.9e-06
[add_recomputation]: 3.774e-05
[cse_after_recomputation]: 1.992e-05, [1]
    [Cycle 1]: 1.448e-05, [1]
        [cse]: 8.98002e-06
[auto_monad_reorder]: 1.131e-05
[get_jit_bprop_graph]: 1.61002e-06
[rewriter_after_jit_bprop_graph]: 5.61998e-06
[opt_after_jit_grad]: 0.00048544
[symbol_engine_optimizer]: 7.423e-05, [1]
    [Cycle 1]: 6.75e-05, [6]
        [build]: 3.52997e-06
        [elim_shapecalc]: 8.68001e-06
        [elim_not_effective]: 1.281e-05
        [opt_reshape]: 5.57001e-06
        [fold_const_symbol]: 8.50999e-06
        [renormalize]: 5.00004e-07
[validate]: 3.114e-05
[backend_pass]: 8.79983e-07
[task_emit]: 0.00537639
[execute]: 7.4e-06
Sums
    bootstrap                            :     0.000252s :  1.91%
    type_inference                       :     0.004064s : 30.91%
    event_method                         :     0.000018s :  0.14%
    auto_monad                           :     0.000054s :  0.41%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.26%
    rewriter_before_opt_a                :     0.000070s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.26%
    jit_opt_a.a_1                        :     0.000620s :  4.72%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000545s :  4.14%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.12%
    jit_opt_a.cse                        :     0.000031s :  0.23%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000030s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000511s :  3.89%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000019s :  0.14%
    loop_unroll                          :     0.000433s :  3.29%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000010s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000038s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000011s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000485s :  3.69%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000031s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005376s : 40.89%
    execute                              :     0.000007s :  0.06%


Time group info:
------[substitution.]   0.000205    23
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.13% :     0.000004s :      3: substitution.graph_param_transform
69.76% :     0.000143s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.93% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.91% :     0.000004s :      2: substitution.replace_old_param
21.13% :     0.000043s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004008     2
81.77% :     0.003277s :      1: type_inference.infer
18.23% :     0.000730s :      1: type_inference.specialize
------[replace.]   0.000062     6
65.77% :     0.000041s :      4: replace.inline
34.23% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000182     6
76.98% :     0.000140s :      4: match.inline
23.02% :     0.000042s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.10% :     0.000002s :     10: predicate.arithmetic_simplify
 1.50% :     0.000002s :     10: predicate.cast_eliminate
 0.71% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.04% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.27% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.39% :     0.000006s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000001s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.17% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.81% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     15: predicate.partial_eliminate
 1.23% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.91% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.39% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 1.12% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.49% :     0.000008s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.57% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000509    12
48.57% :     0.000247s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.16% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.27% :     0.000231s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018472    76
 0.22% :     0.000041s :      1: add_recomputation
 0.31% :     0.000057s :      1: auto_monad
 0.08% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.42% :     0.000263s :      1: bootstrap
 0.12% :     0.000021s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.12% :     0.000022s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
27.05% :     0.004996s :      1: jit_opt_a
 0.75% :     0.000138s :      1: jit_opt_after_cconv
 0.27% :     0.000051s :      1: jit_opt_b
 2.39% :     0.000441s :      1: loop_unroll
 2.82% :     0.000520s :      1: mutable_eliminate
 4.45% :     0.000822s :     26: opt.transform.jit_opt_a
 0.25% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.67% :     0.000494s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000013s :      1: remove_dup_value
 1.51% :     0.000278s :      1: renormalize.infer
 1.40% :     0.000259s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000033s :      1: rewriter_after_opt_a
 0.40% :     0.000073s :      1: rewriter_before_opt_a
 0.42% :     0.000077s :      1: symbol_engine_optimizer
29.17% :     0.005388s :      1: task_emit
22.38% :     0.004135s :      1: type_inference
 0.26% :     0.000048s :      1: validate


TotalTime = 0.0190627, [33]
[bootstrap]: 0.00029807
[type_inference]: 0.00471506
[event_method]: 1.629e-05
[auto_monad]: 6.212e-05
[graph_reusing]: 5.72001e-06
[pre_auto_parallel]: 2.31e-06
[py_interpret_to_execute]: 3.15e-05
[rewriter_before_opt_a]: 6.91e-05
[expand_dump_flag]: 2.94999e-06
[jit_opt_a]: 0.00563468, [2]
    [Cycle 1]: 0.00153079, [27]
        [switch_simplify]: 5.007e-05
        [loop_unroll]: 3.126e-05
        [a_1]: 0.00051906
        [with_stream_mark]: 1.414e-05
        [recompute_prepare]: 7.06001e-06
        [updatestate_depend_eliminate]: 2.89001e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 6.16e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 5.40999e-06
        [meta_shard_fg_expand]: 1.52001e-06
        [get_grad_eliminate_]: 5.17999e-06
        [merge_forward]: 2.61e-06
        [cell_reuse_recompute_pass]: 1.14998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.249e-05
        [j_node_and_user_rematch]: 9.15001e-06
        [meta_fg_expand]: 2.12001e-06
        [replace_old_param]: 7.88001e-06
        [inline_without_move]: 5.46e-06
        [renormalize]: 0.00058569
        [add_forward_monad_depend]: 4.48001e-06
        [auto_monad_grad]: 1.89e-06
        [auto_monad_eliminator]: 1.058e-05
        [cse]: 1.798e-05
        [replace_applicator]: 1.144e-05
    [Cycle 2]: 0.00032713, [27]
        [switch_simplify]: 5.66e-06
        [loop_unroll]: 4.99e-06
        [a_1]: 9.085e-05
        [with_stream_mark]: 9.62999e-06
        [recompute_prepare]: 5.00001e-06
        [updatestate_depend_eliminate]: 2.31998e-06
        [updatestate_assign_eliminate]: 2.23002e-06
        [updatestate_loads_eliminate]: 1.86998e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 4.95001e-06
        [updatestate_useless_node_eliminater]: 4.86997e-06
        [accelerated_algorithm]: 4.99e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 4.58999e-06
        [merge_forward]: 2.54999e-06
        [cell_reuse_recompute_pass]: 2.00002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.506e-05
        [j_node_and_user_rematch]: 7.80998e-06
        [meta_fg_expand]: 1.63002e-06
        [replace_old_param]: 7.24001e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.92999e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 5.05999e-06
        [cse]: 9.80002e-06
        [replace_applicator]: 7.48e-06
[py_interpret_to_execute_after_opt_a]: 8.50001e-06
[rewriter_after_opt_a]: 2.926e-05
[convert_after_rewriter]: 6.49001e-06
[order_py_execute_after_rewriter]: 4.90001e-06
[mutable_eliminate]: 0.00052664
[jit_opt_b]: 4.886e-05, [1]
    [Cycle 1]: 4.252e-05, [2]
        [frontend_op_eliminate]: 1.605e-05
        [inline_after_opt_a]: 1.518e-05
[cconv]: 2.053e-05
[loop_unroll]: 0.0004312
[jit_opt_after_cconv]: 0.00013569, [1]
    [Cycle 1]: 0.00012926, [11]
        [c_1]: 2.059e-05
        [parameter_eliminate]: 2.10002e-06
        [updatestate_depend_eliminate]: 4.72e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [cse]: 1.83e-05
        [call_graph_tuple_transform]: 1.709e-05
        [tuple_list_get_item_eliminator]: 5.34998e-06
        [none_parameter_eliminate]: 1.00001e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 5.36998e-06
[remove_dup_value]: 9.10999e-06
[partial_unused_args_eliminate]: 1.77001e-06
[environ_conv]: 4.39002e-06
[add_recomputation]: 3.631e-05
[cse_after_recomputation]: 1.995e-05, [1]
    [Cycle 1]: 1.499e-05, [1]
        [cse]: 9.05001e-06
[auto_monad_reorder]: 1.162e-05
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 4.30999e-06
[opt_after_jit_grad]: 0.00050158
[symbol_engine_optimizer]: 7.288e-05, [1]
    [Cycle 1]: 6.649e-05, [6]
        [build]: 2.91999e-06
        [elim_shapecalc]: 8.13001e-06
        [elim_not_effective]: 1.278e-05
        [opt_reshape]: 6.09999e-06
        [fold_const_symbol]: 8.36002e-06
        [renormalize]: 6.30011e-07
[validate]: 2.913e-05
[backend_pass]: 8.49977e-07
[task_emit]: 0.00612691
[execute]: 6.78e-06
Sums
    bootstrap                            :     0.000298s :  2.03%
    type_inference                       :     0.004715s : 32.16%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000062s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.21%
    rewriter_before_opt_a                :     0.000069s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.38%
    jit_opt_a.loop_unroll                :     0.000036s :  0.25%
    jit_opt_a.a_1                        :     0.000610s :  4.16%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000586s :  4.00%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.11%
    jit_opt_a.cse                        :     0.000028s :  0.19%
    jit_opt_a.replace_applicator         :     0.000019s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000029s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000527s :  3.59%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000021s :  0.14%
    loop_unroll                          :     0.000431s :  2.94%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000009s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000036s :  0.25%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.08%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000502s :  3.42%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000029s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006127s : 41.79%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000186    23
 0.78% :     0.000001s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 2.34% :     0.000004s :      3: substitution.graph_param_transform
82.96% :     0.000155s :      4: substitution.inline
 1.43% :     0.000003s :      4: substitution.j_node_and_user_rematch
 4.06% :     0.000008s :      4: substitution.remove_not_recompute_node
 1.76% :     0.000003s :      2: substitution.replace_old_param
 6.15% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004652     2
81.68% :     0.003800s :      1: type_inference.infer
18.32% :     0.000852s :      1: type_inference.specialize
------[replace.]   0.000068     6
66.40% :     0.000045s :      4: replace.inline
33.60% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     6
93.61% :     0.000152s :      4: match.inline
 6.39% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.49% :     0.000002s :     10: predicate.addn_zero_filter
 2.36% :     0.000003s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.18% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.35% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.28% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000001s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.24% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.37% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.17% :     0.000001s :     10: predicate.minmaximum_grad
 1.48% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.41% :     0.000002s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.52% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.61% :     0.000003s :     16: predicate.switch_defer_inline
 2.26% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.84% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.55% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.15% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.14% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000587    12
44.44% :     0.000261s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.96% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.60% :     0.000297s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020571    76
 0.19% :     0.000039s :      1: add_recomputation
 0.32% :     0.000065s :      1: auto_monad
 0.07% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.52% :     0.000312s :      1: bootstrap
 0.11% :     0.000023s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.41% :     0.005638s :      1: jit_opt_a
 0.67% :     0.000139s :      1: jit_opt_after_cconv
 0.25% :     0.000052s :      1: jit_opt_b
 2.13% :     0.000439s :      1: loop_unroll
 2.60% :     0.000534s :      1: mutable_eliminate
 3.98% :     0.000820s :     26: opt.transform.jit_opt_a
 0.22% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.48% :     0.000510s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000034s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000012s :      1: remove_dup_value
 1.40% :     0.000289s :      1: renormalize.infer
 1.41% :     0.000290s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000032s :      1: rewriter_after_opt_a
 0.35% :     0.000072s :      1: rewriter_before_opt_a
 0.37% :     0.000075s :      1: symbol_engine_optimizer
29.85% :     0.006140s :      1: task_emit
22.98% :     0.004726s :      1: type_inference
 0.23% :     0.000047s :      1: validate


TotalTime = 0.0816222, [33]
[bootstrap]: 0.00030987
[type_inference]: 0.0644187
[event_method]: 1.744e-05
[auto_monad]: 6.463e-05
[graph_reusing]: 5.09e-06
[pre_auto_parallel]: 2.66999e-06
[py_interpret_to_execute]: 0.00028061
[rewriter_before_opt_a]: 8.024e-05
[expand_dump_flag]: 3.47002e-06
[jit_opt_a]: 0.00725849, [2]
    [Cycle 1]: 0.00318513, [27]
        [switch_simplify]: 6.06e-05
        [loop_unroll]: 3.292e-05
        [a_1]: 0.00061294
        [with_stream_mark]: 2.129e-05
        [recompute_prepare]: 9.22999e-06
        [updatestate_depend_eliminate]: 3.88999e-06
        [updatestate_assign_eliminate]: 4.23999e-06
        [updatestate_loads_eliminate]: 2.97002e-06
        [parameter_eliminate]: 1.97001e-06
        [specialize_transform]: 7.2e-06
        [updatestate_useless_node_eliminater]: 6.84001e-06
        [accelerated_algorithm]: 6.28998e-06
        [meta_shard_fg_expand]: 2.36e-06
        [get_grad_eliminate_]: 6.78e-06
        [merge_forward]: 3.68e-06
        [cell_reuse_recompute_pass]: 1.30001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.595e-05
        [j_node_and_user_rematch]: 9.47999e-06
        [meta_fg_expand]: 2.81999e-06
        [replace_old_param]: 9.71998e-06
        [inline_without_move]: 6.45002e-06
        [renormalize]: 0.00210285
        [add_forward_monad_depend]: 8.18001e-06
        [auto_monad_grad]: 2.73998e-06
        [auto_monad_eliminator]: 1.739e-05
        [cse]: 2.661e-05
        [replace_applicator]: 1.853e-05
    [Cycle 2]: 0.00039842, [27]
        [switch_simplify]: 7.39002e-06
        [loop_unroll]: 6.46999e-06
        [a_1]: 0.00013863
        [with_stream_mark]: 1.262e-05
        [recompute_prepare]: 6.66999e-06
        [updatestate_depend_eliminate]: 3.33e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.84001e-06
        [parameter_eliminate]: 1.24998e-06
        [specialize_transform]: 5.79e-06
        [updatestate_useless_node_eliminater]: 5.79e-06
        [accelerated_algorithm]: 6.07001e-06
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 5.61e-06
        [merge_forward]: 3.35e-06
        [cell_reuse_recompute_pass]: 3.65998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.352e-05
        [j_node_and_user_rematch]: 8.92e-06
        [meta_fg_expand]: 2.16e-06
        [replace_old_param]: 9.11998e-06
        [inline_without_move]: 5.46998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 6.49999e-06
        [cse]: 1.272e-05
        [replace_applicator]: 6.16998e-06
[py_interpret_to_execute_after_opt_a]: 1.27e-05
[rewriter_after_opt_a]: 3.787e-05
[convert_after_rewriter]: 6.81001e-06
[order_py_execute_after_rewriter]: 5.35999e-06
[mutable_eliminate]: 0.0007056
[jit_opt_b]: 6.015e-05, [1]
    [Cycle 1]: 5.168e-05, [2]
        [frontend_op_eliminate]: 2.09e-05
        [inline_after_opt_a]: 1.82e-05
[cconv]: 2.912e-05
[loop_unroll]: 0.00045044
[jit_opt_after_cconv]: 0.00015556, [1]
    [Cycle 1]: 0.00014891, [11]
        [c_1]: 2.613e-05
        [parameter_eliminate]: 3.16999e-06
        [updatestate_depend_eliminate]: 6.47001e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.98e-06
        [cse]: 2.255e-05
        [call_graph_tuple_transform]: 2.204e-05
        [tuple_list_get_item_eliminator]: 6.49999e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 6.24999e-06
[remove_dup_value]: 1.532e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 8.69e-06
[add_recomputation]: 5.082e-05
[cse_after_recomputation]: 2.426e-05, [1]
    [Cycle 1]: 1.89e-05, [1]
        [cse]: 1.215e-05
[auto_monad_reorder]: 1.54e-05
[get_jit_bprop_graph]: 3.51001e-06
[rewriter_after_jit_bprop_graph]: 5.54e-06
[opt_after_jit_grad]: 0.00049957
[symbol_engine_optimizer]: 0.00014817, [1]
    [Cycle 1]: 0.00014184, [6]
        [build]: 3.76999e-06
        [elim_shapecalc]: 1.031e-05
        [elim_not_effective]: 7.702e-05
        [opt_reshape]: 7.08e-06
        [fold_const_symbol]: 1.126e-05
        [renormalize]: 6.90023e-07
[validate]: 4.861e-05
[backend_pass]: 9.79984e-07
[task_emit]: 0.00668876
[execute]: 8e-06
Sums
    bootstrap                            :     0.000310s :  0.40%
    type_inference                       :     0.064419s : 83.31%
    event_method                         :     0.000017s :  0.02%
    auto_monad                           :     0.000065s :  0.08%
    graph_reusing                        :     0.000005s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000281s :  0.36%
    rewriter_before_opt_a                :     0.000080s :  0.10%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000068s :  0.09%
    jit_opt_a.loop_unroll                :     0.000039s :  0.05%
    jit_opt_a.a_1                        :     0.000752s :  0.97%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.04%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.00%
    jit_opt_a.specialize_transform       :     0.000013s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000013s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.02%
    jit_opt_a.merge_forward              :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.04%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.02%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000019s :  0.02%
    jit_opt_a.inline_without_move        :     0.000012s :  0.02%
    jit_opt_a.renormalize                :     0.002103s :  2.72%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.03%
    jit_opt_a.cse                        :     0.000039s :  0.05%
    jit_opt_a.replace_applicator         :     0.000025s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.02%
    rewriter_after_opt_a                 :     0.000038s :  0.05%
    convert_after_rewriter               :     0.000007s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000706s :  0.91%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.02%
    cconv                                :     0.000029s :  0.04%
    loop_unroll                          :     0.000450s :  0.58%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.03%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.cse              :     0.000023s :  0.03%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.01%
    remove_dup_value                     :     0.000015s :  0.02%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000009s :  0.01%
    add_recomputation                    :     0.000051s :  0.07%
    cse_after_recomputation.cse          :     0.000012s :  0.02%
    auto_monad_reorder                   :     0.000015s :  0.02%
    get_jit_bprop_graph                  :     0.000004s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000500s :  0.65%
    symbol_engine_optimizer.build        :     0.000004s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000077s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000049s :  0.06%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006689s :  8.65%
    execute                              :     0.000008s :  0.01%


Time group info:
------[substitution.]   0.000299    27
20.90% :     0.000062s :      2: substitution.elim_not_effective
 0.44% :     0.000001s :      2: substitution.fold_const_symbol
 1.96% :     0.000006s :      4: substitution.graph_param_transform
64.37% :     0.000192s :      4: substitution.inline
 1.09% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.52% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.47% :     0.000004s :      2: substitution.replace_old_param
 5.00% :     0.000015s :      3: substitution.tile_eliminate
 3.23% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.064337     2
97.06% :     0.062447s :      1: type_inference.infer
 2.94% :     0.001890s :      1: type_inference.specialize
------[replace.]   0.000073     6
67.01% :     0.000049s :      4: replace.inline
32.99% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000198     6
95.69% :     0.000190s :      4: match.inline
 4.31% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000140   834
 1.35% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.47% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.32% :     0.000002s :     12: predicate.addn_check_dump
 1.44% :     0.000002s :     12: predicate.addn_zero_filter
 2.19% :     0.000003s :     12: predicate.arithmetic_simplify
 1.47% :     0.000002s :     12: predicate.cast_eliminate
 0.40% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     12: predicate.compare_switch_simplify
 1.26% :     0.000002s :     12: predicate.depend_value_elim
 1.16% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.15% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.55% :     0.000001s :      4: predicate.elim_not_effective
 0.57% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.17% :     0.000002s :     12: predicate.environ_get_eliminate
 1.08% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      4: predicate.fold_const_symbol
 1.21% :     0.000002s :      8: predicate.get_grad_eliminate
 0.31% :     0.000000s :      4: predicate.graph_param_transform
 5.17% :     0.000007s :     26: predicate.inline
 0.94% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :      8: predicate.less_batch_normalization
 1.36% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     18: predicate.load_eliminater
 1.51% :     0.000002s :      4: predicate.loop_unroll_after_grad
 4.26% :     0.000006s :     38: predicate.loop_unroll_before_grad
 1.88% :     0.000003s :     16: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000002s :     12: predicate.merge_addn
 1.17% :     0.000002s :     12: predicate.minmaximum_grad
 2.04% :     0.000003s :      4: predicate.mutable_eliminate
 0.78% :     0.000001s :      4: predicate.opt_reshape
 2.05% :     0.000003s :     18: predicate.partial_eliminate
 1.44% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     12: predicate.reduce_eliminate
 1.39% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.84% :     0.000003s :     22: predicate.replace_applicator
 0.60% :     0.000001s :      8: predicate.replace_old_param
 0.33% :     0.000000s :      4: predicate.reset_defer_inline
 1.70% :     0.000002s :     12: predicate.reshape_eliminate
 1.22% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     12: predicate.same_eliminate
 0.64% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.17% :     0.000002s :      8: predicate.special_op_eliminate
 0.93% :     0.000001s :      8: predicate.specialize_transform
 1.34% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     18: predicate.switch_defer_inline
 2.04% :     0.000003s :     18: predicate.switch_layer_defer_inline
 7.46% :     0.000010s :     60: predicate.switch_simplify
 1.57% :     0.000002s :     12: predicate.tile_eliminate
 1.36% :     0.000002s :     12: predicate.transpose_eliminate
 1.44% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.63% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     18: predicate.updatestate_pure_node_eliminater
 3.04% :     0.000004s :     26: predicate.updatestate_useless_node_eliminater
 1.40% :     0.000002s :     12: predicate.value_based_eliminate
 0.41% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001943    22
65.84% :     0.001279s :     15: func_graph_cloner_run.FuncGraphClonerGraph
 2.45% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
31.72% :     0.000616s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.084920    76
 0.06% :     0.000054s :      1: add_recomputation
 0.08% :     0.000068s :      1: auto_monad
 0.02% :     0.000018s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.38% :     0.000323s :      1: bootstrap
 0.04% :     0.000032s :      1: cconv
 0.01% :     0.000009s :      1: convert_after_rewriter
 0.03% :     0.000026s :      1: cse_after_recomputation
 0.01% :     0.000011s :      1: environ_conv
 0.02% :     0.000021s :      1: event_method
 0.01% :     0.000011s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000006s :      1: get_jit_bprop_graph
 0.01% :     0.000007s :      1: graph_reusing
 8.55% :     0.007262s :      1: jit_opt_a
 0.19% :     0.000158s :      1: jit_opt_after_cconv
 0.07% :     0.000063s :      1: jit_opt_b
 0.54% :     0.000459s :      1: loop_unroll
 0.84% :     0.000716s :      1: mutable_eliminate
 1.17% :     0.000998s :     26: opt.transform.jit_opt_a
 0.07% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000033s :      4: opt.transform.jit_opt_b
 0.02% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000102s :      4: opt.transform.symbol_engine_opt
 0.60% :     0.000509s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.34% :     0.000286s :      1: py_interpret_to_execute
 0.02% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000018s :      1: remove_dup_value
 1.66% :     0.001411s :      1: renormalize.infer
 0.80% :     0.000681s :      1: renormalize.specialize
 0.01% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.05% :     0.000041s :      1: rewriter_after_opt_a
 0.10% :     0.000085s :      1: rewriter_before_opt_a
 0.18% :     0.000151s :      1: symbol_engine_optimizer
 7.89% :     0.006700s :      1: task_emit
75.87% :     0.064429s :      1: type_inference
 0.08% :     0.000070s :      1: validate


TotalTime = 0.0192813, [33]
[bootstrap]: 0.00030041
[type_inference]: 0.00443666
[event_method]: 1.628e-05
[auto_monad]: 6.348e-05
[graph_reusing]: 5.92001e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 3.44e-05
[rewriter_before_opt_a]: 7.375e-05
[expand_dump_flag]: 3.07002e-06
[jit_opt_a]: 0.00545634, [2]
    [Cycle 1]: 0.0016075, [27]
        [switch_simplify]: 5.579e-05
        [loop_unroll]: 2.943e-05
        [a_1]: 0.00053622
        [with_stream_mark]: 2.031e-05
        [recompute_prepare]: 7.38e-06
        [updatestate_depend_eliminate]: 3.63e-06
        [updatestate_assign_eliminate]: 2.76999e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 2.43e-06
        [specialize_transform]: 6.45002e-06
        [updatestate_useless_node_eliminater]: 5.23002e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 2.43e-06
        [get_grad_eliminate_]: 5.00001e-06
        [merge_forward]: 3.48e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.366e-05
        [j_node_and_user_rematch]: 8.45001e-06
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 8.48001e-06
        [inline_without_move]: 5.26002e-06
        [renormalize]: 0.00064993
        [add_forward_monad_depend]: 6.14001e-06
        [auto_monad_grad]: 2.49001e-06
        [auto_monad_eliminator]: 1.466e-05
        [cse]: 2.448e-05
        [replace_applicator]: 1.329e-05
    [Cycle 2]: 0.00032431, [27]
        [switch_simplify]: 5.94e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 9.016e-05
        [with_stream_mark]: 9.37999e-06
        [recompute_prepare]: 4.87e-06
        [updatestate_depend_eliminate]: 2.37999e-06
        [updatestate_assign_eliminate]: 1.99999e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 9.20001e-07
        [specialize_transform]: 5.00999e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 4.97e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 4.57998e-06
        [merge_forward]: 3.04999e-06
        [cell_reuse_recompute_pass]: 1.76998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.386e-05
        [j_node_and_user_rematch]: 8.84003e-06
        [meta_fg_expand]: 1.76e-06
        [replace_old_param]: 7.63001e-06
        [inline_without_move]: 4.89e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 5.20999e-06
        [cse]: 1.022e-05
        [replace_applicator]: 5.49e-06
[py_interpret_to_execute_after_opt_a]: 1.15e-05
[rewriter_after_opt_a]: 3.425e-05
[convert_after_rewriter]: 6.69999e-06
[order_py_execute_after_rewriter]: 4.82e-06
[mutable_eliminate]: 0.0005918
[jit_opt_b]: 5.112e-05, [1]
    [Cycle 1]: 4.357e-05, [2]
        [frontend_op_eliminate]: 1.735e-05
        [inline_after_opt_a]: 1.472e-05
[cconv]: 2.719e-05
[loop_unroll]: 0.00043913
[jit_opt_after_cconv]: 0.00014532, [1]
    [Cycle 1]: 0.00013801, [11]
        [c_1]: 2.066e-05
        [parameter_eliminate]: 3.41001e-06
        [updatestate_depend_eliminate]: 6.61999e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 2.55002e-06
        [cse]: 1.954e-05
        [call_graph_tuple_transform]: 2.008e-05
        [tuple_list_get_item_eliminator]: 5.42999e-06
        [none_parameter_eliminate]: 1.48002e-06
        [renormalize]: 6.69999e-07
        [switch_simplify]: 5.57999e-06
[remove_dup_value]: 1.469e-05
[partial_unused_args_eliminate]: 2.19001e-06
[environ_conv]: 5.96e-06
[add_recomputation]: 4.961e-05
[cse_after_recomputation]: 1.986e-05, [1]
    [Cycle 1]: 1.462e-05, [1]
        [cse]: 9.04e-06
[auto_monad_reorder]: 1.475e-05
[get_jit_bprop_graph]: 2.38002e-06
[rewriter_after_jit_bprop_graph]: 5.46e-06
[opt_after_jit_grad]: 0.00051622
[symbol_engine_optimizer]: 7.423e-05, [1]
    [Cycle 1]: 6.718e-05, [6]
        [build]: 3.97e-06
        [elim_shapecalc]: 7.75998e-06
        [elim_not_effective]: 1.31e-05
        [opt_reshape]: 5.89e-06
        [fold_const_symbol]: 8.65001e-06
        [renormalize]: 4.19997e-07
[validate]: 3.698e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00663547
[execute]: 8.19002e-06
Sums
    bootstrap                            :     0.000300s :  1.98%
    type_inference                       :     0.004437s : 29.26%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000063s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.23%
    rewriter_before_opt_a                :     0.000074s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000626s :  4.13%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000650s :  4.29%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000035s :  0.23%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000034s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000592s :  3.90%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000027s :  0.18%
    loop_unroll                          :     0.000439s :  2.90%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000015s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000050s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000516s :  3.40%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006635s : 43.77%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000210    23
 1.12% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000002s :      2: substitution.fold_const_symbol
 2.52% :     0.000005s :      3: substitution.graph_param_transform
84.15% :     0.000177s :      4: substitution.inline
 1.51% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.03% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.24% :     0.000005s :      2: substitution.replace_old_param
 5.71% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004375     2
81.56% :     0.003568s :      1: type_inference.infer
18.44% :     0.000807s :      1: type_inference.specialize
------[replace.]   0.000063     6
67.35% :     0.000042s :      4: replace.inline
32.65% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000185     6
94.12% :     0.000174s :      4: match.inline
 5.88% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.22% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.37% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 1.85% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.47% :     0.000001s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.43% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.96% :     0.000001s :      6: predicate.less_batch_normalization
 1.46% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.58% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.11% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.16% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 2.20% :     0.000003s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.35% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.40% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.53% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.19% :     0.000010s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.09% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.43% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 1.05% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000561    12
47.42% :     0.000266s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.77% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.81% :     0.000251s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020877    76
 0.25% :     0.000052s :      1: add_recomputation
 0.32% :     0.000067s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.50% :     0.000313s :      1: bootstrap
 0.14% :     0.000030s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.15% :     0.005460s :      1: jit_opt_a
 0.71% :     0.000148s :      1: jit_opt_after_cconv
 0.26% :     0.000054s :      1: jit_opt_b
 2.15% :     0.000448s :      1: loop_unroll
 2.88% :     0.000601s :      1: mutable_eliminate
 4.01% :     0.000838s :     26: opt.transform.jit_opt_a
 0.23% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.51% :     0.000525s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000037s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.65% :     0.000345s :      1: renormalize.infer
 1.42% :     0.000297s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000037s :      1: rewriter_after_opt_a
 0.37% :     0.000077s :      1: rewriter_before_opt_a
 0.37% :     0.000077s :      1: symbol_engine_optimizer
31.83% :     0.006645s :      1: task_emit
21.32% :     0.004451s :      1: type_inference
 0.27% :     0.000057s :      1: validate


TotalTime = 0.0187451, [33]
[bootstrap]: 0.00028583
[type_inference]: 0.00440477
[event_method]: 1.74e-05
[auto_monad]: 6.348e-05
[graph_reusing]: 5.28002e-06
[pre_auto_parallel]: 2.22999e-06
[py_interpret_to_execute]: 3.164e-05
[rewriter_before_opt_a]: 6.912e-05
[expand_dump_flag]: 3.23998e-06
[jit_opt_a]: 0.00523578, [2]
    [Cycle 1]: 0.00159595, [27]
        [switch_simplify]: 5.063e-05
        [loop_unroll]: 3.028e-05
        [a_1]: 0.00051208
        [with_stream_mark]: 1.597e-05
        [recompute_prepare]: 7.39002e-06
        [updatestate_depend_eliminate]: 2.97002e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.47001e-06
        [parameter_eliminate]: 1.70001e-06
        [specialize_transform]: 5.79e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 5.76998e-06
        [meta_shard_fg_expand]: 2.05002e-06
        [get_grad_eliminate_]: 5.72001e-06
        [merge_forward]: 3.6e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.356e-05
        [j_node_and_user_rematch]: 5.334e-05
        [meta_fg_expand]: 2.61e-06
        [replace_old_param]: 8.86997e-06
        [inline_without_move]: 5.49998e-06
        [renormalize]: 0.00064499
        [add_forward_monad_depend]: 4.55001e-06
        [auto_monad_grad]: 2.29001e-06
        [auto_monad_eliminator]: 1.292e-05
        [cse]: 1.737e-05
        [replace_applicator]: 1.131e-05
    [Cycle 2]: 0.00032621, [27]
        [switch_simplify]: 6.09001e-06
        [loop_unroll]: 5.04e-06
        [a_1]: 9.032e-05
        [with_stream_mark]: 9.20001e-06
        [recompute_prepare]: 5.19998e-06
        [updatestate_depend_eliminate]: 2.71e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 5.07e-06
        [updatestate_useless_node_eliminater]: 4.84998e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.23002e-06
        [get_grad_eliminate_]: 4.75001e-06
        [merge_forward]: 2.84001e-06
        [cell_reuse_recompute_pass]: 1.58002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.203e-05
        [j_node_and_user_rematch]: 8.3e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 7.20003e-06
        [inline_without_move]: 4.91002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 9.60019e-07
        [auto_monad_eliminator]: 4.88001e-06
        [cse]: 1.044e-05
        [replace_applicator]: 5.59e-06
[py_interpret_to_execute_after_opt_a]: 9.52001e-06
[rewriter_after_opt_a]: 3.23e-05
[convert_after_rewriter]: 6.70002e-06
[order_py_execute_after_rewriter]: 4.81002e-06
[mutable_eliminate]: 0.0005065
[jit_opt_b]: 4.989e-05, [1]
    [Cycle 1]: 4.286e-05, [2]
        [frontend_op_eliminate]: 1.612e-05
        [inline_after_opt_a]: 1.521e-05
[cconv]: 2.337e-05
[loop_unroll]: 0.00043094
[jit_opt_after_cconv]: 0.00013836, [1]
    [Cycle 1]: 0.00013189, [11]
        [c_1]: 2.02e-05
        [parameter_eliminate]: 2.51e-06
        [updatestate_depend_eliminate]: 5.51e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.01998e-06
        [cse]: 1.662e-05
        [call_graph_tuple_transform]: 1.902e-05
        [tuple_list_get_item_eliminator]: 5.66e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 5.00004e-07
        [switch_simplify]: 5.32001e-06
[remove_dup_value]: 1.263e-05
[partial_unused_args_eliminate]: 2.76e-06
[environ_conv]: 4.87998e-06
[add_recomputation]: 4.483e-05
[cse_after_recomputation]: 1.997e-05, [1]
    [Cycle 1]: 1.448e-05, [1]
        [cse]: 8.82e-06
[auto_monad_reorder]: 1.402e-05
[get_jit_bprop_graph]: 2.01003e-06
[rewriter_after_jit_bprop_graph]: 5.01002e-06
[opt_after_jit_grad]: 0.00046232
[symbol_engine_optimizer]: 7.198e-05, [1]
    [Cycle 1]: 6.542e-05, [6]
        [build]: 3.28e-06
        [elim_shapecalc]: 7.18e-06
        [elim_not_effective]: 1.261e-05
        [opt_reshape]: 5.82001e-06
        [fold_const_symbol]: 8.76002e-06
        [renormalize]: 4.50003e-07
[validate]: 3.198e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00654928
[execute]: 7.18998e-06
Sums
    bootstrap                            :     0.000286s :  1.93%
    type_inference                       :     0.004405s : 29.69%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000063s :  0.43%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.21%
    rewriter_before_opt_a                :     0.000069s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.38%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000602s :  4.06%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000062s :  0.42%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000645s :  4.35%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000028s :  0.19%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000032s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000506s :  3.41%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000023s :  0.16%
    loop_unroll                          :     0.000431s :  2.91%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000045s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000462s :  3.12%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006549s : 44.15%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000229    23
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000002s :      2: substitution.fold_const_symbol
 2.33% :     0.000005s :      3: substitution.graph_param_transform
67.15% :     0.000154s :      4: substitution.inline
20.24% :     0.000046s :      4: substitution.j_node_and_user_rematch
 1.87% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.60% :     0.000004s :      2: substitution.replace_old_param
 5.25% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004351     2
82.86% :     0.003605s :      1: type_inference.infer
17.14% :     0.000746s :      1: type_inference.specialize
------[replace.]   0.000069     6
67.55% :     0.000046s :      4: replace.inline
32.45% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000163     6
93.23% :     0.000152s :      4: match.inline
 6.77% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.55% :     0.000002s :     10: predicate.addn_zero_filter
 2.27% :     0.000003s :     10: predicate.arithmetic_simplify
 1.37% :     0.000002s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.45% :     0.000006s :     22: predicate.inline
 1.11% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000001s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.44% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.19% :     0.000001s :     10: predicate.minmaximum_grad
 1.36% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.24% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.60% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.16% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.65% :     0.000009s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.51% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000535    12
48.67% :     0.000261s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.15% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.18% :     0.000247s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020343    76
 0.23% :     0.000048s :      1: add_recomputation
 0.33% :     0.000067s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.48% :     0.000300s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.75% :     0.005239s :      1: jit_opt_a
 0.69% :     0.000141s :      1: jit_opt_after_cconv
 0.26% :     0.000052s :      1: jit_opt_b
 2.16% :     0.000440s :      1: loop_unroll
 2.53% :     0.000515s :      1: mutable_eliminate
 4.19% :     0.000852s :     26: opt.transform.jit_opt_a
 0.23% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.31% :     0.000470s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000035s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 1.69% :     0.000344s :      1: renormalize.infer
 1.44% :     0.000293s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000035s :      1: rewriter_after_opt_a
 0.36% :     0.000073s :      1: rewriter_before_opt_a
 0.37% :     0.000074s :      1: symbol_engine_optimizer
32.27% :     0.006564s :      1: task_emit
21.71% :     0.004415s :      1: type_inference
 0.25% :     0.000051s :      1: validate


TotalTime = 0.0151607, [30]
[bootstrap]: 0.00027537
[type_inference]: 0.00664247
[event_method]: 2.275e-05
[auto_monad]: 8.3e-05
[graph_reusing]: 7.37002e-06
[pre_auto_parallel]: 2.74001e-06
[py_interpret_to_execute]: 5.278e-05
[rewriter_before_opt_a]: 8.874e-05
[expand_dump_flag]: 3.14999e-06
[jit_opt_a]: 0.00567174, [2]
    [Cycle 1]: 0.00188978, [27]
        [switch_simplify]: 0.00012064
        [loop_unroll]: 3.537e-05
        [a_1]: 0.00069192
        [with_stream_mark]: 1.804e-05
        [recompute_prepare]: 7.7e-06
        [updatestate_depend_eliminate]: 3.63e-06
        [updatestate_assign_eliminate]: 2.89999e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [parameter_eliminate]: 2.00002e-06
        [specialize_transform]: 6.12999e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 5.64e-06
        [meta_shard_fg_expand]: 2.10002e-06
        [get_grad_eliminate_]: 5.52001e-06
        [merge_forward]: 3.89002e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.458e-05
        [j_node_and_user_rematch]: 8.83001e-06
        [meta_fg_expand]: 2.48998e-06
        [replace_old_param]: 8.60001e-06
        [inline_without_move]: 5.46e-06
        [renormalize]: 0.00068961
        [add_forward_monad_depend]: 5.92999e-06
        [auto_monad_grad]: 2.50997e-06
        [auto_monad_eliminator]: 1.359e-05
        [cse]: 3.008e-05
        [replace_applicator]: 1.088e-05
    [Cycle 2]: 0.00031978, [27]
        [switch_simplify]: 5.49e-06
        [loop_unroll]: 4.95001e-06
        [a_1]: 9.506e-05
        [with_stream_mark]: 9.20999e-06
        [recompute_prepare]: 5.00001e-06
        [updatestate_depend_eliminate]: 2.48002e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.77001e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 5.01002e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 4.99998e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 2.74999e-06
        [cell_reuse_recompute_pass]: 1.51002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.235e-05
        [j_node_and_user_rematch]: 7.86001e-06
        [meta_fg_expand]: 1.63002e-06
        [replace_old_param]: 7.21001e-06
        [inline_without_move]: 4.74e-06
        [renormalize]: 1.10012e-07
        [add_forward_monad_depend]: 1.45999e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 4.53999e-06
        [cse]: 9.93998e-06
        [replace_applicator]: 4.85999e-06
[py_interpret_to_execute_after_opt_a]: 1.104e-05
[rewriter_after_opt_a]: 3.373e-05
[convert_after_rewriter]: 6.94999e-06
[order_py_execute_after_rewriter]: 5.05001e-06
[mutable_eliminate]: 0.00068301
[jit_opt_b]: 5.92e-05, [1]
    [Cycle 1]: 4.184e-05, [2]
        [frontend_op_eliminate]: 1.601e-05
        [inline_after_opt_a]: 1.418e-05
[cconv]: 2.53e-05
[loop_unroll]: 0.00045575
[jit_opt_after_cconv]: 0.0001458, [1]
    [Cycle 1]: 0.0001393, [11]
        [c_1]: 2.109e-05
        [parameter_eliminate]: 2.59001e-06
        [updatestate_depend_eliminate]: 7.43e-06
        [updatestate_assign_eliminate]: 2.64001e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [cse]: 1.905e-05
        [call_graph_tuple_transform]: 1.924e-05
        [tuple_list_get_item_eliminator]: 5.52001e-06
        [none_parameter_eliminate]: 1.49998e-06
        [renormalize]: 5.19998e-07
        [switch_simplify]: 5.59e-06
[remove_dup_value]: 1.469e-05
[partial_unused_args_eliminate]: 2.59001e-06
[environ_conv]: 4.89003e-06
[add_recomputation]: 5.028e-05
[cse_after_recomputation]: 2.102e-05, [1]
    [Cycle 1]: 1.54e-05, [1]
        [cse]: 9.62001e-06
[auto_monad_reorder]: 1.528e-05
[get_jit_bprop_graph]: 1.57999e-06
[rewriter_after_jit_bprop_graph]: 5.67999e-06
[opt_after_jit_grad]: 0.00048512
[symbol_engine_optimizer]: 7.466e-05, [1]
    [Cycle 1]: 6.816e-05, [6]
        [build]: 4.46002e-06
        [elim_shapecalc]: 7.56999e-06
        [elim_not_effective]: 1.253e-05
        [opt_reshape]: 5.56e-06
        [fold_const_symbol]: 8.53001e-06
        [renormalize]: 4.50003e-07
[validate]: 3.66e-05
Sums
    bootstrap                            :     0.000275s :  2.48%
    type_inference                       :     0.006642s : 59.86%
    event_method                         :     0.000023s :  0.21%
    auto_monad                           :     0.000083s :  0.75%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000053s :  0.48%
    rewriter_before_opt_a                :     0.000089s :  0.80%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000126s :  1.14%
    jit_opt_a.loop_unroll                :     0.000040s :  0.36%
    jit_opt_a.a_1                        :     0.000787s :  7.09%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000690s :  6.22%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.16%
    jit_opt_a.cse                        :     0.000040s :  0.36%
    jit_opt_a.replace_applicator         :     0.000016s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.10%
    rewriter_after_opt_a                 :     0.000034s :  0.30%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000683s :  6.16%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.13%
    cconv                                :     0.000025s :  0.23%
    loop_unroll                          :     0.000456s :  4.11%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000015s :  0.13%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000050s :  0.45%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000485s :  4.37%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.33%


Time group info:
------[substitution.]   0.000274    26
 0.73% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 2.07% :     0.000006s :      3: substitution.graph_param_transform
84.86% :     0.000232s :      6: substitution.inline
 0.96% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.65% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.41% :     0.000004s :      2: substitution.replace_old_param
 3.29% :     0.000009s :      1: substitution.switch_simplify
 4.58% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006571     2
82.90% :     0.005447s :      1: type_inference.infer
17.10% :     0.001124s :      1: type_inference.specialize
------[replace.]   0.000135     9
50.02% :     0.000068s :      6: replace.inline
25.68% :     0.000035s :      1: replace.switch_simplify
24.31% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000248     9
92.12% :     0.000229s :      6: match.inline
 3.34% :     0.000008s :      1: match.switch_simplify
 4.54% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   801
 1.34% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.05% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     12: predicate.addn_check_dump
 1.31% :     0.000002s :     12: predicate.addn_zero_filter
 2.04% :     0.000003s :     12: predicate.arithmetic_simplify
 1.20% :     0.000002s :     12: predicate.cast_eliminate
 0.38% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000002s :     12: predicate.compare_switch_simplify
 1.32% :     0.000002s :     12: predicate.depend_value_elim
 1.18% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.38% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.13% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.55% :     0.000002s :     12: predicate.environ_get_eliminate
 1.21% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.17% :     0.000007s :     26: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     17: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.26% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.91% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000002s :     12: predicate.merge_addn
 1.25% :     0.000002s :     12: predicate.minmaximum_grad
 1.41% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     17: predicate.partial_eliminate
 1.18% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.96% :     0.000003s :     12: predicate.reduce_eliminate
 1.60% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     20: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     12: predicate.reshape_eliminate
 1.22% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     12: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.84% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.72% :     0.000004s :     20: predicate.switch_defer_inline
 2.56% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.71% :     0.000012s :     65: predicate.switch_simplify
 1.28% :     0.000002s :     12: predicate.tile_eliminate
 1.28% :     0.000002s :     12: predicate.transpose_eliminate
 1.49% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.45% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     12: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.51% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000777    15
48.17% :     0.000374s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.11% :     0.000086s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.73% :     0.000316s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016972    72
 0.31% :     0.000053s :      1: add_recomputation
 0.51% :     0.000087s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 1.70% :     0.000289s :      1: bootstrap
 0.17% :     0.000028s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.16% :     0.000027s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000010s :      1: graph_reusing
33.44% :     0.005676s :      1: jit_opt_a
 0.88% :     0.000149s :      1: jit_opt_after_cconv
 0.37% :     0.000062s :      1: jit_opt_b
 2.73% :     0.000464s :      1: loop_unroll
 4.08% :     0.000692s :      1: mutable_eliminate
 6.28% :     0.001065s :     26: opt.transform.jit_opt_a
 0.28% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.91% :     0.000494s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.33% :     0.000055s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.21% :     0.000374s :      1: renormalize.infer
 1.81% :     0.000308s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000037s :      1: rewriter_after_opt_a
 0.54% :     0.000092s :      1: rewriter_before_opt_a
 0.46% :     0.000077s :      1: symbol_engine_optimizer
39.25% :     0.006661s :      1: type_inference


TotalTime = 0.0182596, [33]
[bootstrap]: 0.00028762
[type_inference]: 0.00447247
[event_method]: 1.711e-05
[auto_monad]: 5.906e-05
[graph_reusing]: 5.20001e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 3.066e-05
[rewriter_before_opt_a]: 6.644e-05
[expand_dump_flag]: 2.74001e-06
[jit_opt_a]: 0.00532649, [2]
    [Cycle 1]: 0.00164373, [27]
        [switch_simplify]: 4.802e-05
        [loop_unroll]: 3.05e-05
        [a_1]: 0.00055478
        [with_stream_mark]: 1.818e-05
        [recompute_prepare]: 8.35999e-06
        [updatestate_depend_eliminate]: 3.10998e-06
        [updatestate_assign_eliminate]: 2.13998e-06
        [updatestate_loads_eliminate]: 2.74001e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 6.33998e-06
        [updatestate_useless_node_eliminater]: 5.67001e-06
        [accelerated_algorithm]: 5.90002e-06
        [meta_shard_fg_expand]: 2.31e-06
        [get_grad_eliminate_]: 5.60001e-06
        [merge_forward]: 4.07e-06
        [cell_reuse_recompute_pass]: 1.52001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.468e-05
        [j_node_and_user_rematch]: 9.16998e-06
        [meta_fg_expand]: 2.63e-06
        [replace_old_param]: 8.37998e-06
        [inline_without_move]: 5.39e-06
        [renormalize]: 0.00065644
        [add_forward_monad_depend]: 5.77999e-06
        [auto_monad_grad]: 2.61e-06
        [auto_monad_eliminator]: 1.311e-05
        [cse]: 2.804e-05
        [replace_applicator]: 1.463e-05
    [Cycle 2]: 0.00034249, [27]
        [switch_simplify]: 5.81e-06
        [loop_unroll]: 5.20001e-06
        [a_1]: 9.622e-05
        [with_stream_mark]: 1.161e-05
        [recompute_prepare]: 6.12001e-06
        [updatestate_depend_eliminate]: 2.72001e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.40002e-06
        [parameter_eliminate]: 1.47001e-06
        [specialize_transform]: 5.32999e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 5.91998e-06
        [meta_shard_fg_expand]: 1.50999e-06
        [get_grad_eliminate_]: 4.65999e-06
        [merge_forward]: 2.64001e-06
        [cell_reuse_recompute_pass]: 1.94e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.431e-05
        [j_node_and_user_rematch]: 8.02998e-06
        [meta_fg_expand]: 1.83002e-06
        [replace_old_param]: 8.02e-06
        [inline_without_move]: 5.23002e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 9.99979e-07
        [auto_monad_eliminator]: 5.52001e-06
        [cse]: 1.089e-05
        [replace_applicator]: 5.84999e-06
[py_interpret_to_execute_after_opt_a]: 1.235e-05
[rewriter_after_opt_a]: 3.369e-05
[convert_after_rewriter]: 6.65998e-06
[order_py_execute_after_rewriter]: 5.08002e-06
[mutable_eliminate]: 0.00058848
[jit_opt_b]: 5.097e-05, [1]
    [Cycle 1]: 4.4e-05, [2]
        [frontend_op_eliminate]: 1.717e-05
        [inline_after_opt_a]: 1.522e-05
[cconv]: 2.435e-05
[loop_unroll]: 0.00044032
[jit_opt_after_cconv]: 0.00014223, [1]
    [Cycle 1]: 0.00013556, [11]
        [c_1]: 2.216e-05
        [parameter_eliminate]: 2.83e-06
        [updatestate_depend_eliminate]: 5.67999e-06
        [updatestate_assign_eliminate]: 2.47001e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 1.673e-05
        [call_graph_tuple_transform]: 1.927e-05
        [tuple_list_get_item_eliminator]: 5.67001e-06
        [none_parameter_eliminate]: 1.49998e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 5.52001e-06
[remove_dup_value]: 1.23e-05
[partial_unused_args_eliminate]: 2.41998e-06
[environ_conv]: 4.90001e-06
[add_recomputation]: 4.682e-05
[cse_after_recomputation]: 2.093e-05, [1]
    [Cycle 1]: 1.518e-05, [1]
        [cse]: 9.27999e-06
[auto_monad_reorder]: 1.432e-05
[get_jit_bprop_graph]: 2.01e-06
[rewriter_after_jit_bprop_graph]: 6.33002e-06
[opt_after_jit_grad]: 0.00047072
[symbol_engine_optimizer]: 7.531e-05, [1]
    [Cycle 1]: 6.759e-05, [6]
        [build]: 3.20998e-06
        [elim_shapecalc]: 8.03001e-06
        [elim_not_effective]: 1.293e-05
        [opt_reshape]: 5.83002e-06
        [fold_const_symbol]: 8.79e-06
        [renormalize]: 5.40022e-07
[validate]: 3.229e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.0057844
[execute]: 8.37998e-06
Sums
    bootstrap                            :     0.000288s :  2.01%
    type_inference                       :     0.004472s : 31.30%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000059s :  0.41%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.21%
    rewriter_before_opt_a                :     0.000066s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.38%
    jit_opt_a.loop_unroll                :     0.000036s :  0.25%
    jit_opt_a.a_1                        :     0.000651s :  4.56%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000657s :  4.59%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000039s :  0.27%
    jit_opt_a.replace_applicator         :     0.000020s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.09%
    rewriter_after_opt_a                 :     0.000034s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000588s :  4.12%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000440s :  3.08%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000047s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000471s :  3.29%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005784s : 40.48%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000192    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.69% :     0.000005s :      3: substitution.graph_param_transform
83.75% :     0.000161s :      4: substitution.inline
 1.43% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.62% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.14% :     0.000004s :      2: substitution.replace_old_param
 5.71% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004412     2
81.89% :     0.003613s :      1: type_inference.infer
18.11% :     0.000799s :      1: type_inference.specialize
------[replace.]   0.000068     6
68.32% :     0.000046s :      4: replace.inline
31.68% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000168     6
94.17% :     0.000158s :      4: match.inline
 5.83% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 2.09% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000002s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000002s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.24% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000000s :      3: predicate.graph_param_transform
 4.84% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :      6: predicate.less_batch_normalization
 1.65% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.89% :     0.000002s :     15: predicate.load_eliminater
 1.56% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.14% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.50% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000002s :     10: predicate.merge_addn
 1.16% :     0.000001s :     10: predicate.minmaximum_grad
 1.88% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     15: predicate.partial_eliminate
 1.28% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.83% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.40% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.53% :     0.000009s :     53: predicate.switch_simplify
 1.34% :     0.000002s :     10: predicate.tile_eliminate
 1.30% :     0.000002s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.37% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000542    12
48.21% :     0.000261s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.83% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.96% :     0.000249s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019888    76
 0.25% :     0.000050s :      1: add_recomputation
 0.31% :     0.000062s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.52% :     0.000302s :      1: bootstrap
 0.18% :     0.000036s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.80% :     0.005330s :      1: jit_opt_a
 0.73% :     0.000145s :      1: jit_opt_after_cconv
 0.27% :     0.000054s :      1: jit_opt_b
 2.26% :     0.000449s :      1: loop_unroll
 3.00% :     0.000597s :      1: mutable_eliminate
 4.35% :     0.000865s :     26: opt.transform.jit_opt_a
 0.25% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.41% :     0.000479s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 1.70% :     0.000338s :      1: renormalize.infer
 1.56% :     0.000310s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000037s :      1: rewriter_after_opt_a
 0.35% :     0.000069s :      1: rewriter_before_opt_a
 0.39% :     0.000078s :      1: symbol_engine_optimizer
29.16% :     0.005800s :      1: task_emit
22.54% :     0.004484s :      1: type_inference
 0.25% :     0.000051s :      1: validate


TotalTime = 0.0181551, [33]
[bootstrap]: 0.00027803
[type_inference]: 0.00427217
[event_method]: 1.679e-05
[auto_monad]: 5.354e-05
[graph_reusing]: 4.77e-06
[pre_auto_parallel]: 2.27999e-06
[py_interpret_to_execute]: 3.094e-05
[rewriter_before_opt_a]: 6.859e-05
[expand_dump_flag]: 2.43e-06
[jit_opt_a]: 0.00500297, [2]
    [Cycle 1]: 0.00146985, [27]
        [switch_simplify]: 4.723e-05
        [loop_unroll]: 3.089e-05
        [a_1]: 0.00048815
        [with_stream_mark]: 1.4e-05
        [recompute_prepare]: 7.12002e-06
        [updatestate_depend_eliminate]: 2.85002e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.44e-06
        [specialize_transform]: 5.99e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.08002e-06
        [meta_shard_fg_expand]: 1.50999e-06
        [get_grad_eliminate_]: 4.85999e-06
        [merge_forward]: 2.64999e-06
        [cell_reuse_recompute_pass]: 8.70001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.119e-05
        [j_node_and_user_rematch]: 8.17998e-06
        [meta_fg_expand]: 2.10002e-06
        [replace_old_param]: 8.14002e-06
        [inline_without_move]: 5.20999e-06
        [renormalize]: 0.00059852
        [add_forward_monad_depend]: 4.79e-06
        [auto_monad_grad]: 1.97999e-06
        [auto_monad_eliminator]: 1.094e-05
        [cse]: 1.826e-05
        [replace_applicator]: 1.251e-05
    [Cycle 2]: 0.00031976, [27]
        [switch_simplify]: 5.45001e-06
        [loop_unroll]: 4.85999e-06
        [a_1]: 8.995e-05
        [with_stream_mark]: 8.74e-06
        [recompute_prepare]: 4.86997e-06
        [updatestate_depend_eliminate]: 2.51e-06
        [updatestate_assign_eliminate]: 1.92999e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 1.25999e-06
        [specialize_transform]: 4.71002e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 4.68001e-06
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.57998e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 1.71e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.236e-05
        [j_node_and_user_rematch]: 7.58001e-06
        [meta_fg_expand]: 1.73002e-06
        [replace_old_param]: 7.11999e-06
        [inline_without_move]: 1.078e-05
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 5.57001e-06
        [cse]: 1.113e-05
        [replace_applicator]: 5.14003e-06
[py_interpret_to_execute_after_opt_a]: 8.52998e-06
[rewriter_after_opt_a]: 2.957e-05
[convert_after_rewriter]: 5.84999e-06
[order_py_execute_after_rewriter]: 4.43999e-06
[mutable_eliminate]: 0.00054509
[jit_opt_b]: 5.144e-05, [1]
    [Cycle 1]: 4.423e-05, [2]
        [frontend_op_eliminate]: 1.705e-05
        [inline_after_opt_a]: 1.542e-05
[cconv]: 2.259e-05
[loop_unroll]: 0.00043681
[jit_opt_after_cconv]: 0.00014603, [1]
    [Cycle 1]: 0.00013969, [11]
        [c_1]: 2.056e-05
        [parameter_eliminate]: 2.65002e-06
        [updatestate_depend_eliminate]: 6.69999e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [cse]: 2.307e-05
        [call_graph_tuple_transform]: 1.904e-05
        [tuple_list_get_item_eliminator]: 5.59e-06
        [none_parameter_eliminate]: 1.17999e-06
        [renormalize]: 6.89994e-07
        [switch_simplify]: 5.28002e-06
[remove_dup_value]: 1.287e-05
[partial_unused_args_eliminate]: 1.83002e-06
[environ_conv]: 4.74e-06
[add_recomputation]: 4.245e-05
[cse_after_recomputation]: 2.095e-05, [1]
    [Cycle 1]: 1.564e-05, [1]
        [cse]: 9.44e-06
[auto_monad_reorder]: 1.357e-05
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 4.85001e-06
[opt_after_jit_grad]: 0.00047749
[symbol_engine_optimizer]: 7.604e-05, [1]
    [Cycle 1]: 6.808e-05, [6]
        [build]: 4.3e-06
        [elim_shapecalc]: 7.43999e-06
        [elim_not_effective]: 1.406e-05
        [opt_reshape]: 5.45001e-06
        [fold_const_symbol]: 8.53001e-06
        [renormalize]: 9.00007e-07
[validate]: 3.792e-05
[backend_pass]: 8.49977e-07
[task_emit]: 0.00627956
[execute]: 8.27e-06
Sums
    bootstrap                            :     0.000278s :  1.94%
    type_inference                       :     0.004272s : 29.77%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000054s :  0.37%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.22%
    rewriter_before_opt_a                :     0.000069s :  0.48%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.37%
    jit_opt_a.loop_unroll                :     0.000036s :  0.25%
    jit_opt_a.a_1                        :     0.000578s :  4.03%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000016s :  0.11%
    jit_opt_a.renormalize                :     0.000599s :  4.17%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.12%
    jit_opt_a.cse                        :     0.000029s :  0.20%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000030s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000545s :  3.80%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000023s :  0.16%
    loop_unroll                          :     0.000437s :  3.04%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000042s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000477s :  3.33%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000038s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006280s : 43.75%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000164    23
 1.68% :     0.000003s :      2: substitution.elim_not_effective
 0.77% :     0.000001s :      2: substitution.fold_const_symbol
 3.24% :     0.000005s :      3: substitution.graph_param_transform
82.20% :     0.000135s :      4: substitution.inline
 1.40% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.39% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.27% :     0.000004s :      2: substitution.replace_old_param
 6.04% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004214     2
81.77% :     0.003446s :      1: type_inference.infer
18.23% :     0.000768s :      1: type_inference.specialize
------[replace.]   0.000062     6
65.80% :     0.000041s :      4: replace.inline
34.20% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000141     6
93.82% :     0.000132s :      4: match.inline
 6.18% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.52% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.17% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.49% :     0.000001s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.42% :     0.000006s :     22: predicate.inline
 1.02% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.22% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.76% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000002s :     15: predicate.partial_eliminate
 1.24% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.13% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.63% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 2.24% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.65% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.81% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000513    12
48.13% :     0.000247s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.03% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.84% :     0.000240s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019646    76
 0.23% :     0.000045s :      1: add_recomputation
 0.29% :     0.000057s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.50% :     0.000294s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
25.48% :     0.005006s :      1: jit_opt_a
 0.76% :     0.000149s :      1: jit_opt_after_cconv
 0.28% :     0.000054s :      1: jit_opt_b
 2.27% :     0.000445s :      1: loop_unroll
 2.82% :     0.000555s :      1: mutable_eliminate
 3.98% :     0.000782s :     26: opt.transform.jit_opt_a
 0.24% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.47% :     0.000486s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.63% :     0.000321s :      1: renormalize.infer
 1.38% :     0.000271s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000032s :      1: rewriter_after_opt_a
 0.36% :     0.000071s :      1: rewriter_before_opt_a
 0.40% :     0.000079s :      1: symbol_engine_optimizer
32.02% :     0.006291s :      1: task_emit
21.80% :     0.004284s :      1: type_inference
 0.30% :     0.000058s :      1: validate


TotalTime = 0.0147433, [30]
[bootstrap]: 0.00028462
[type_inference]: 0.00623797
[event_method]: 2.077e-05
[auto_monad]: 8.041e-05
[graph_reusing]: 5.79e-06
[pre_auto_parallel]: 2.67001e-06
[py_interpret_to_execute]: 4.42e-05
[rewriter_before_opt_a]: 8.299e-05
[expand_dump_flag]: 3.13e-06
[jit_opt_a]: 0.0057749, [2]
    [Cycle 1]: 0.00183074, [27]
        [switch_simplify]: 0.00011483
        [loop_unroll]: 3.396e-05
        [a_1]: 0.00068163
        [with_stream_mark]: 1.94e-05
        [recompute_prepare]: 8.17e-06
        [updatestate_depend_eliminate]: 3.45998e-06
        [updatestate_assign_eliminate]: 2.77002e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [parameter_eliminate]: 1.49e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.69999e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 5.06997e-06
        [merge_forward]: 3.44001e-06
        [cell_reuse_recompute_pass]: 1.60999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.463e-05
        [j_node_and_user_rematch]: 9.12999e-06
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 8.80001e-06
        [inline_without_move]: 5.42999e-06
        [renormalize]: 0.00066389
        [add_forward_monad_depend]: 6.29001e-06
        [auto_monad_grad]: 2.41998e-06
        [auto_monad_eliminator]: 1.472e-05
        [cse]: 1.887e-05
        [replace_applicator]: 1.351e-05
    [Cycle 2]: 0.00035931, [27]
        [switch_simplify]: 5.33002e-06
        [loop_unroll]: 5.09998e-06
        [a_1]: 8.872e-05
        [with_stream_mark]: 9.39e-06
        [recompute_prepare]: 4.89003e-06
        [updatestate_depend_eliminate]: 2.61e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 9.10019e-07
        [specialize_transform]: 4.92e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 4.87e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.53001e-06
        [merge_forward]: 2.73e-06
        [cell_reuse_recompute_pass]: 1.69e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.334e-05
        [j_node_and_user_rematch]: 7.88001e-06
        [meta_fg_expand]: 2.18998e-06
        [replace_old_param]: 7.7e-06
        [inline_without_move]: 4.70999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.30999e-06
        [auto_monad_grad]: 6.90023e-07
        [auto_monad_eliminator]: 5.37001e-06
        [cse]: 1.037e-05
        [replace_applicator]: 4.84e-06
[py_interpret_to_execute_after_opt_a]: 1.098e-05
[rewriter_after_opt_a]: 3.385e-05
[convert_after_rewriter]: 7.32002e-06
[order_py_execute_after_rewriter]: 5.05001e-06
[mutable_eliminate]: 0.00058913
[jit_opt_b]: 5.096e-05, [1]
    [Cycle 1]: 4.333e-05, [2]
        [frontend_op_eliminate]: 1.629e-05
        [inline_after_opt_a]: 1.509e-05
[cconv]: 2.744e-05
[loop_unroll]: 0.00045543
[jit_opt_after_cconv]: 0.00014078, [1]
    [Cycle 1]: 0.00013424, [11]
        [c_1]: 2.127e-05
        [parameter_eliminate]: 2.73998e-06
        [updatestate_depend_eliminate]: 5.78002e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [cse]: 1.818e-05
        [call_graph_tuple_transform]: 1.853e-05
        [tuple_list_get_item_eliminator]: 5.15999e-06
        [none_parameter_eliminate]: 1.37e-06
        [renormalize]: 6.99976e-07
        [switch_simplify]: 5.52001e-06
[remove_dup_value]: 1.375e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 1.374e-05
[add_recomputation]: 4.72e-05
[cse_after_recomputation]: 2.144e-05, [1]
    [Cycle 1]: 1.582e-05, [1]
        [cse]: 9.71e-06
[auto_monad_reorder]: 1.48e-05
[get_jit_bprop_graph]: 2.06998e-06
[rewriter_after_jit_bprop_graph]: 4.43001e-06
[opt_after_jit_grad]: 0.00049805
[symbol_engine_optimizer]: 7.544e-05, [1]
    [Cycle 1]: 6.807e-05, [6]
        [build]: 3.55e-06
        [elim_shapecalc]: 7.63001e-06
        [elim_not_effective]: 1.297e-05
        [opt_reshape]: 5.66e-06
        [fold_const_symbol]: 8.75999e-06
        [renormalize]: 4.19997e-07
[validate]: 3.341e-05
Sums
    bootstrap                            :     0.000285s :  2.70%
    type_inference                       :     0.006238s : 59.14%
    event_method                         :     0.000021s :  0.20%
    auto_monad                           :     0.000080s :  0.76%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000044s :  0.42%
    rewriter_before_opt_a                :     0.000083s :  0.79%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000120s :  1.14%
    jit_opt_a.loop_unroll                :     0.000039s :  0.37%
    jit_opt_a.a_1                        :     0.000770s :  7.30%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000664s :  6.30%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.19%
    jit_opt_a.cse                        :     0.000029s :  0.28%
    jit_opt_a.replace_applicator         :     0.000018s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.10%
    rewriter_after_opt_a                 :     0.000034s :  0.32%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000589s :  5.59%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.14%
    cconv                                :     0.000027s :  0.26%
    loop_unroll                          :     0.000455s :  4.32%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000014s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000014s :  0.13%
    add_recomputation                    :     0.000047s :  0.45%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.04%
    opt_after_jit_grad                   :     0.000498s :  4.72%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.32%


Time group info:
------[substitution.]   0.000269    26
 0.66% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 1.97% :     0.000005s :      3: substitution.graph_param_transform
84.43% :     0.000228s :      6: substitution.inline
 1.09% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.75% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.34% :     0.000004s :      2: substitution.replace_old_param
 3.06% :     0.000008s :      1: substitution.switch_simplify
 5.21% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006174     2
83.89% :     0.005180s :      1: type_inference.infer
16.11% :     0.000994s :      1: type_inference.specialize
------[replace.]   0.000136     9
49.77% :     0.000068s :      6: replace.inline
27.02% :     0.000037s :      1: replace.switch_simplify
23.21% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000244     9
91.63% :     0.000224s :      6: match.inline
 3.10% :     0.000008s :      1: match.switch_simplify
 5.27% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   801
 1.29% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.54% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     12: predicate.addn_check_dump
 1.33% :     0.000002s :     12: predicate.addn_zero_filter
 2.01% :     0.000003s :     12: predicate.arithmetic_simplify
 1.32% :     0.000002s :     12: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     12: predicate.compare_switch_simplify
 1.57% :     0.000002s :     12: predicate.depend_value_elim
 1.22% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.17% :     0.000002s :     12: predicate.environ_get_eliminate
 1.13% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.29% :     0.000007s :     26: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.95% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     17: predicate.load_eliminater
 1.48% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.26% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.14% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000002s :     12: predicate.merge_addn
 1.21% :     0.000002s :     12: predicate.minmaximum_grad
 1.60% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     17: predicate.partial_eliminate
 1.16% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.81% :     0.000002s :     12: predicate.reduce_eliminate
 1.51% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.68% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     20: predicate.replace_applicator
 0.45% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     12: predicate.reshape_eliminate
 1.21% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     12: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.80% :     0.000004s :     20: predicate.switch_defer_inline
 2.34% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.55% :     0.000011s :     65: predicate.switch_simplify
 1.20% :     0.000002s :     12: predicate.tile_eliminate
 1.47% :     0.000002s :     12: predicate.transpose_eliminate
 1.32% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.81% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000715    15
48.15% :     0.000344s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.40% :     0.000082s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.45% :     0.000289s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016514    72
 0.30% :     0.000050s :      1: add_recomputation
 0.51% :     0.000084s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.79% :     0.000296s :      1: bootstrap
 0.18% :     0.000030s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.10% :     0.000016s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
34.99% :     0.005778s :      1: jit_opt_a
 0.87% :     0.000144s :      1: jit_opt_after_cconv
 0.33% :     0.000054s :      1: jit_opt_b
 2.80% :     0.000463s :      1: loop_unroll
 3.63% :     0.000599s :      1: mutable_eliminate
 6.32% :     0.001044s :     26: opt.transform.jit_opt_a
 0.29% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.06% :     0.000506s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000047s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.19% :     0.000361s :      1: renormalize.infer
 1.78% :     0.000294s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000037s :      1: rewriter_after_opt_a
 0.52% :     0.000086s :      1: rewriter_before_opt_a
 0.47% :     0.000078s :      1: symbol_engine_optimizer
37.84% :     0.006249s :      1: type_inference


TotalTime = 0.0178184, [33]
[bootstrap]: 0.00027401
[type_inference]: 0.00400254
[event_method]: 1.614e-05
[auto_monad]: 4.346e-05
[graph_reusing]: 3.43e-06
[pre_auto_parallel]: 1.87001e-06
[py_interpret_to_execute]: 3.269e-05
[rewriter_before_opt_a]: 6.351e-05
[expand_dump_flag]: 2.51998e-06
[jit_opt_a]: 0.00490125, [2]
    [Cycle 1]: 0.00137404, [27]
        [switch_simplify]: 4.562e-05
        [loop_unroll]: 3.49e-05
        [a_1]: 0.00047234
        [with_stream_mark]: 1.098e-05
        [recompute_prepare]: 7.06001e-06
        [updatestate_depend_eliminate]: 2.83998e-06
        [updatestate_assign_eliminate]: 2.18002e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.97001e-06
        [updatestate_useless_node_eliminater]: 5.43002e-06
        [accelerated_algorithm]: 5.62001e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 5.14998e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 7.99977e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.082e-05
        [j_node_and_user_rematch]: 8.69998e-06
        [meta_fg_expand]: 2.01e-06
        [replace_old_param]: 7.73001e-06
        [inline_without_move]: 5.37999e-06
        [renormalize]: 0.00051786
        [add_forward_monad_depend]: 4.86997e-06
        [auto_monad_grad]: 2.36e-06
        [auto_monad_eliminator]: 1.254e-05
        [cse]: 2.516e-05
        [replace_applicator]: 1.191e-05
    [Cycle 2]: 0.00032988, [27]
        [switch_simplify]: 6.04001e-06
        [loop_unroll]: 5.08002e-06
        [a_1]: 9.504e-05
        [with_stream_mark]: 8.59998e-06
        [recompute_prepare]: 5.34e-06
        [updatestate_depend_eliminate]: 2.39001e-06
        [updatestate_assign_eliminate]: 2.08002e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 5.14e-06
        [updatestate_useless_node_eliminater]: 5.46e-06
        [accelerated_algorithm]: 5.32001e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.63999e-06
        [merge_forward]: 2.37001e-06
        [cell_reuse_recompute_pass]: 2.00002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.186e-05
        [j_node_and_user_rematch]: 7.92003e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 7.35e-06
        [inline_without_move]: 5.01997e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.21002e-06
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 4.85001e-06
        [cse]: 8.86997e-06
        [replace_applicator]: 5.29e-06
[py_interpret_to_execute_after_opt_a]: 8.85001e-06
[rewriter_after_opt_a]: 3.151e-05
[convert_after_rewriter]: 7.47998e-06
[order_py_execute_after_rewriter]: 4.99e-06
[mutable_eliminate]: 0.00049192
[jit_opt_b]: 4.847e-05, [1]
    [Cycle 1]: 4.173e-05, [2]
        [frontend_op_eliminate]: 1.578e-05
        [inline_after_opt_a]: 1.485e-05
[cconv]: 2.433e-05
[loop_unroll]: 0.00041859
[jit_opt_after_cconv]: 0.0001392, [1]
    [Cycle 1]: 0.0001321, [11]
        [c_1]: 2.149e-05
        [parameter_eliminate]: 2.27001e-06
        [updatestate_depend_eliminate]: 5.34998e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [cse]: 1.681e-05
        [call_graph_tuple_transform]: 1.736e-05
        [tuple_list_get_item_eliminator]: 5.57001e-06
        [none_parameter_eliminate]: 1.49998e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 5.69e-06
[remove_dup_value]: 1.5e-05
[partial_unused_args_eliminate]: 2.59999e-06
[environ_conv]: 5.07e-06
[add_recomputation]: 4.677e-05
[cse_after_recomputation]: 2.071e-05, [1]
    [Cycle 1]: 1.556e-05, [1]
        [cse]: 9.73002e-06
[auto_monad_reorder]: 1.47e-05
[get_jit_bprop_graph]: 1.93002e-06
[rewriter_after_jit_bprop_graph]: 5.27999e-06
[opt_after_jit_grad]: 0.00043357
[symbol_engine_optimizer]: 0.00010589, [1]
    [Cycle 1]: 9.995e-05, [6]
        [build]: 3.44001e-06
        [elim_shapecalc]: 8.40001e-06
        [elim_not_effective]: 1.234e-05
        [opt_reshape]: 5.81998e-06
        [fold_const_symbol]: 9.15001e-06
        [renormalize]: 3.69997e-07
[validate]: 3.241e-05
[backend_pass]: 8.80013e-07
[task_emit]: 0.00642193
[execute]: 7.75998e-06
Sums
    bootstrap                            :     0.000274s :  1.96%
    type_inference                       :     0.004003s : 28.59%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000043s :  0.31%
    graph_reusing                        :     0.000003s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000033s :  0.23%
    rewriter_before_opt_a                :     0.000064s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000052s :  0.37%
    jit_opt_a.loop_unroll                :     0.000040s :  0.29%
    jit_opt_a.a_1                        :     0.000567s :  4.05%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000518s :  3.70%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.12%
    jit_opt_a.cse                        :     0.000034s :  0.24%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000032s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000492s :  3.51%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000419s :  2.99%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000015s :  0.11%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000434s :  3.10%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006422s : 45.87%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000138    23
 1.43% :     0.000002s :      2: substitution.elim_not_effective
 0.99% :     0.000001s :      2: substitution.fold_const_symbol
 3.72% :     0.000005s :      3: substitution.graph_param_transform
81.63% :     0.000113s :      4: substitution.inline
 1.83% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.77% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.13% :     0.000003s :      2: substitution.replace_old_param
 5.50% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003953     2
80.85% :     0.003196s :      1: type_inference.infer
19.15% :     0.000757s :      1: type_inference.specialize
------[replace.]   0.000061     6
64.27% :     0.000039s :      4: replace.inline
35.73% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000117     6
94.47% :     0.000110s :      4: match.inline
 5.53% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.71% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.00% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.20% :     0.000001s :     10: predicate.addn_check_dump
 1.71% :     0.000002s :     10: predicate.addn_zero_filter
 1.99% :     0.000002s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000002s :     10: predicate.depend_value_elim
 1.20% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     10: predicate.environ_get_eliminate
 1.34% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 4.66% :     0.000006s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.74% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.16% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.28% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     15: predicate.partial_eliminate
 1.08% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.52% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.60% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.63% :     0.000003s :     16: predicate.switch_defer_inline
 2.24% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.41% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.32% :     0.000002s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.53% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.09% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.84% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.22% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     10: predicate.value_based_eliminate
 0.44% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000501    12
44.99% :     0.000225s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.42% :     0.000022s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.58% :     0.000253s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019211    76
 0.26% :     0.000050s :      1: add_recomputation
 0.24% :     0.000047s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.51% :     0.000289s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
25.53% :     0.004904s :      1: jit_opt_a
 0.74% :     0.000142s :      1: jit_opt_after_cconv
 0.27% :     0.000051s :      1: jit_opt_b
 2.22% :     0.000426s :      1: loop_unroll
 2.60% :     0.000499s :      1: mutable_eliminate
 4.02% :     0.000772s :     26: opt.transform.jit_opt_a
 0.24% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.29% :     0.000440s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000035s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000018s :      1: remove_dup_value
 1.26% :     0.000242s :      1: renormalize.infer
 1.41% :     0.000270s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000037s :      1: rewriter_after_opt_a
 0.35% :     0.000066s :      1: rewriter_before_opt_a
 0.56% :     0.000108s :      1: symbol_engine_optimizer
33.50% :     0.006435s :      1: task_emit
20.89% :     0.004014s :      1: type_inference
 0.26% :     0.000051s :      1: validate


TotalTime = 0.0146748, [30]
[bootstrap]: 0.00028727
[type_inference]: 0.00652649
[event_method]: 2.004e-05
[auto_monad]: 8.306e-05
[graph_reusing]: 5.83997e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 4.756e-05
[rewriter_before_opt_a]: 8.049e-05
[expand_dump_flag]: 3.39001e-06
[jit_opt_a]: 0.0054592, [2]
    [Cycle 1]: 0.00180785, [27]
        [switch_simplify]: 0.00010573
        [loop_unroll]: 3.475e-05
        [a_1]: 0.00066424
        [with_stream_mark]: 1.592e-05
        [recompute_prepare]: 8.12e-06
        [updatestate_depend_eliminate]: 3.68e-06
        [updatestate_assign_eliminate]: 2.15002e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [parameter_eliminate]: 1.39e-06
        [specialize_transform]: 6.49999e-06
        [updatestate_useless_node_eliminater]: 5.07999e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 1.68002e-06
        [get_grad_eliminate_]: 5.37999e-06
        [merge_forward]: 3.21999e-06
        [cell_reuse_recompute_pass]: 1.67001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.346e-05
        [j_node_and_user_rematch]: 8.99e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 8.18999e-06
        [inline_without_move]: 5.25001e-06
        [renormalize]: 0.00067573
        [add_forward_monad_depend]: 5.88002e-06
        [auto_monad_grad]: 2.64001e-06
        [auto_monad_eliminator]: 1.39e-05
        [cse]: 1.939e-05
        [replace_applicator]: 1.337e-05
    [Cycle 2]: 0.00032757, [27]
        [switch_simplify]: 5.50001e-06
        [loop_unroll]: 4.91002e-06
        [a_1]: 8.989e-05
        [with_stream_mark]: 9.87001e-06
        [recompute_prepare]: 5.37999e-06
        [updatestate_depend_eliminate]: 2.53e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 5.12e-06
        [updatestate_useless_node_eliminater]: 1.146e-05
        [accelerated_algorithm]: 4.87e-06
        [meta_shard_fg_expand]: 1.22999e-06
        [get_grad_eliminate_]: 4.38999e-06
        [merge_forward]: 2.27999e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.25e-05
        [j_node_and_user_rematch]: 8.08001e-06
        [meta_fg_expand]: 1.64998e-06
        [replace_old_param]: 9.59e-06
        [inline_without_move]: 4.67998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.45999e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 5.51e-06
        [cse]: 1.022e-05
        [replace_applicator]: 5.39998e-06
[py_interpret_to_execute_after_opt_a]: 1.033e-05
[rewriter_after_opt_a]: 3.037e-05
[convert_after_rewriter]: 6.75998e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.00059082
[jit_opt_b]: 4.968e-05, [1]
    [Cycle 1]: 4.209e-05, [2]
        [frontend_op_eliminate]: 1.585e-05
        [inline_after_opt_a]: 1.47e-05
[cconv]: 2.101e-05
[loop_unroll]: 0.00044198
[jit_opt_after_cconv]: 0.00013674, [1]
    [Cycle 1]: 0.00012992, [11]
        [c_1]: 2.249e-05
        [parameter_eliminate]: 2.56e-06
        [updatestate_depend_eliminate]: 5.32001e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [cse]: 1.476e-05
        [call_graph_tuple_transform]: 1.826e-05
        [tuple_list_get_item_eliminator]: 5.18002e-06
        [none_parameter_eliminate]: 1.33002e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 5.02999e-06
[remove_dup_value]: 1.221e-05
[partial_unused_args_eliminate]: 1.99e-06
[environ_conv]: 4.62e-06
[add_recomputation]: 4.233e-05
[cse_after_recomputation]: 1.891e-05, [1]
    [Cycle 1]: 1.394e-05, [1]
        [cse]: 8.30999e-06
[auto_monad_reorder]: 1.281e-05
[get_jit_bprop_graph]: 1.67001e-06
[rewriter_after_jit_bprop_graph]: 4.67e-06
[opt_after_jit_grad]: 0.00049222
[symbol_engine_optimizer]: 8.049e-05, [1]
    [Cycle 1]: 7.308e-05, [6]
        [build]: 3.56001e-06
        [elim_shapecalc]: 8.00999e-06
        [elim_not_effective]: 1.385e-05
        [opt_reshape]: 6.75002e-06
        [fold_const_symbol]: 9.45001e-06
        [renormalize]: 4.49974e-07
[validate]: 3.187e-05
Sums
    bootstrap                            :     0.000287s :  2.67%
    type_inference                       :     0.006526s : 60.55%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000083s :  0.77%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000048s :  0.44%
    rewriter_before_opt_a                :     0.000080s :  0.75%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000111s :  1.03%
    jit_opt_a.loop_unroll                :     0.000040s :  0.37%
    jit_opt_a.a_1                        :     0.000754s :  7.00%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000017s :  0.15%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000676s :  6.27%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.18%
    jit_opt_a.cse                        :     0.000030s :  0.27%
    jit_opt_a.replace_applicator         :     0.000019s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.10%
    rewriter_after_opt_a                 :     0.000030s :  0.28%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000591s :  5.48%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.14%
    cconv                                :     0.000021s :  0.19%
    loop_unroll                          :     0.000442s :  4.10%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000012s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000042s :  0.39%
    cse_after_recomputation.cse          :     0.000008s :  0.08%
    auto_monad_reorder                   :     0.000013s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000492s :  4.57%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.30%


Time group info:
------[substitution.]   0.000258    26
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 2.07% :     0.000005s :      3: substitution.graph_param_transform
84.09% :     0.000217s :      6: substitution.inline
 1.06% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.68% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.37% :     0.000004s :      2: substitution.replace_old_param
 3.35% :     0.000009s :      1: substitution.switch_simplify
 5.13% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006463     2
83.97% :     0.005427s :      1: type_inference.infer
16.03% :     0.001036s :      1: type_inference.specialize
------[replace.]   0.000135     9
50.83% :     0.000069s :      6: replace.inline
25.89% :     0.000035s :      1: replace.switch_simplify
23.28% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000233     9
91.46% :     0.000213s :      6: match.inline
 3.37% :     0.000008s :      1: match.switch_simplify
 5.18% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   801
 1.28% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.05% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     12: predicate.addn_check_dump
 1.31% :     0.000002s :     12: predicate.addn_zero_filter
 2.55% :     0.000003s :     12: predicate.arithmetic_simplify
 1.31% :     0.000002s :     12: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     12: predicate.compare_switch_simplify
 1.42% :     0.000002s :     12: predicate.depend_value_elim
 1.18% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.71% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.47% :     0.000001s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     12: predicate.environ_get_eliminate
 1.13% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.40% :     0.000001s :      3: predicate.graph_param_transform
 5.07% :     0.000007s :     26: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.02% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 2.20% :     0.000003s :     17: predicate.load_eliminater
 1.61% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.25% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.13% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000002s :     12: predicate.merge_addn
 1.12% :     0.000001s :     12: predicate.minmaximum_grad
 1.08% :     0.000001s :      3: predicate.mutable_eliminate
 0.66% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     17: predicate.partial_eliminate
 1.29% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     12: predicate.reduce_eliminate
 1.49% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     20: predicate.replace_applicator
 0.44% :     0.000001s :      6: predicate.replace_old_param
 0.41% :     0.000001s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     12: predicate.reshape_eliminate
 1.31% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 1.05% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     12: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.67% :     0.000003s :     20: predicate.switch_defer_inline
 2.31% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.31% :     0.000011s :     65: predicate.switch_simplify
 1.27% :     0.000002s :     12: predicate.tile_eliminate
 1.56% :     0.000002s :     12: predicate.transpose_eliminate
 1.39% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.50% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     12: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000734    15
48.31% :     0.000355s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.91% :     0.000080s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.78% :     0.000299s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016440    72
 0.28% :     0.000046s :      1: add_recomputation
 0.53% :     0.000087s :      1: auto_monad
 0.10% :     0.000016s :      1: auto_monad_reorder
 1.82% :     0.000299s :      1: bootstrap
 0.14% :     0.000024s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.14% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
33.23% :     0.005463s :      1: jit_opt_a
 0.85% :     0.000140s :      1: jit_opt_after_cconv
 0.32% :     0.000052s :      1: jit_opt_b
 2.73% :     0.000449s :      1: loop_unroll
 3.64% :     0.000599s :      1: mutable_eliminate
 6.26% :     0.001029s :     26: opt.transform.jit_opt_a
 0.28% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.05% :     0.000501s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000050s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000014s :      1: remove_dup_value
 2.19% :     0.000361s :      1: renormalize.infer
 1.87% :     0.000308s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000033s :      1: rewriter_after_opt_a
 0.51% :     0.000083s :      1: rewriter_before_opt_a
 0.51% :     0.000084s :      1: symbol_engine_optimizer
39.76% :     0.006536s :      1: type_inference


TotalTime = 0.0908304, [33]
[bootstrap]: 0.00031178
[type_inference]: 0.0729421
[event_method]: 1.816e-05
[auto_monad]: 6.907e-05
[graph_reusing]: 5.51e-06
[pre_auto_parallel]: 2.69001e-06
[py_interpret_to_execute]: 0.0002925
[rewriter_before_opt_a]: 8.462e-05
[expand_dump_flag]: 3.70998e-06
[jit_opt_a]: 0.00746054, [2]
    [Cycle 1]: 0.00347696, [27]
        [switch_simplify]: 5.434e-05
        [loop_unroll]: 3.225e-05
        [a_1]: 0.00058559
        [with_stream_mark]: 1.926e-05
        [recompute_prepare]: 1.089e-05
        [updatestate_depend_eliminate]: 4.50999e-06
        [updatestate_assign_eliminate]: 3.33e-06
        [updatestate_loads_eliminate]: 2.95002e-06
        [parameter_eliminate]: 1.97001e-06
        [specialize_transform]: 7.45e-06
        [updatestate_useless_node_eliminater]: 7.08e-06
        [accelerated_algorithm]: 6.61999e-06
        [meta_shard_fg_expand]: 2.24999e-06
        [get_grad_eliminate_]: 7.03e-06
        [merge_forward]: 4.4e-06
        [cell_reuse_recompute_pass]: 1.88002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.579e-05
        [j_node_and_user_rematch]: 9.58002e-06
        [meta_fg_expand]: 2.56e-06
        [replace_old_param]: 9.96998e-06
        [inline_without_move]: 6.54001e-06
        [renormalize]: 0.00239921
        [add_forward_monad_depend]: 7.95998e-06
        [auto_monad_grad]: 2.94001e-06
        [auto_monad_eliminator]: 1.659e-05
        [cse]: 3.138e-05
        [replace_applicator]: 2.124e-05
    [Cycle 2]: 0.00039962, [27]
        [switch_simplify]: 7.98999e-06
        [loop_unroll]: 6.24001e-06
        [a_1]: 0.00013879
        [with_stream_mark]: 1.49e-05
        [recompute_prepare]: 7e-06
        [updatestate_depend_eliminate]: 3.56999e-06
        [updatestate_assign_eliminate]: 2.92002e-06
        [updatestate_loads_eliminate]: 2.69001e-06
        [parameter_eliminate]: 1.65001e-06
        [specialize_transform]: 5.97001e-06
        [updatestate_useless_node_eliminater]: 6.16998e-06
        [accelerated_algorithm]: 5.83002e-06
        [meta_shard_fg_expand]: 2.14e-06
        [get_grad_eliminate_]: 5.34998e-06
        [merge_forward]: 3.43e-06
        [cell_reuse_recompute_pass]: 5.84e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.347e-05
        [j_node_and_user_rematch]: 9.89001e-06
        [meta_fg_expand]: 2.19999e-06
        [replace_old_param]: 9.19998e-06
        [inline_without_move]: 5.98998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.19998e-06
        [auto_monad_grad]: 8.89995e-07
        [auto_monad_eliminator]: 6.35002e-06
        [cse]: 1.147e-05
        [replace_applicator]: 5.97999e-06
[py_interpret_to_execute_after_opt_a]: 1.377e-05
[rewriter_after_opt_a]: 3.545e-05
[convert_after_rewriter]: 7.56999e-06
[order_py_execute_after_rewriter]: 6.02001e-06
[mutable_eliminate]: 0.00068821
[jit_opt_b]: 5.894e-05, [1]
    [Cycle 1]: 5.141e-05, [2]
        [frontend_op_eliminate]: 2.014e-05
        [inline_after_opt_a]: 1.832e-05
[cconv]: 2.444e-05
[loop_unroll]: 0.00047302
[jit_opt_after_cconv]: 0.00021457, [1]
    [Cycle 1]: 0.00020769, [11]
        [c_1]: 2.792e-05
        [parameter_eliminate]: 2.98e-06
        [updatestate_depend_eliminate]: 6.93998e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 3.28e-06
        [cse]: 2.73e-05
        [call_graph_tuple_transform]: 2.494e-05
        [tuple_list_get_item_eliminator]: 6.68998e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 6.49976e-07
        [switch_simplify]: 6.34001e-06
[remove_dup_value]: 1.524e-05
[partial_unused_args_eliminate]: 2.02001e-06
[environ_conv]: 7.47998e-06
[add_recomputation]: 5.018e-05
[cse_after_recomputation]: 2.367e-05, [1]
    [Cycle 1]: 1.799e-05, [1]
        [cse]: 1.238e-05
[auto_monad_reorder]: 1.519e-05
[get_jit_bprop_graph]: 2.53e-06
[rewriter_after_jit_bprop_graph]: 7.3e-06
[opt_after_jit_grad]: 0.00050909
[symbol_engine_optimizer]: 8.349e-05, [1]
    [Cycle 1]: 7.589e-05, [6]
        [build]: 4.50001e-06
        [elim_shapecalc]: 9.47999e-06
        [elim_not_effective]: 1.495e-05
        [opt_reshape]: 6.86001e-06
        [fold_const_symbol]: 1.154e-05
        [renormalize]: 5.49975e-07
[validate]: 4.765e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.00713908
[execute]: 8.05e-06
Sums
    bootstrap                            :     0.000312s :  0.36%
    type_inference                       :     0.072942s : 84.27%
    event_method                         :     0.000018s :  0.02%
    auto_monad                           :     0.000069s :  0.08%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000293s :  0.34%
    rewriter_before_opt_a                :     0.000085s :  0.10%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000062s :  0.07%
    jit_opt_a.loop_unroll                :     0.000038s :  0.04%
    jit_opt_a.a_1                        :     0.000724s :  0.84%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.04%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.00%
    jit_opt_a.specialize_transform       :     0.000013s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000013s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.01%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.01%
    jit_opt_a.merge_forward              :     0.000008s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000008s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.03%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.02%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000019s :  0.02%
    jit_opt_a.inline_without_move        :     0.000013s :  0.01%
    jit_opt_a.renormalize                :     0.002399s :  2.77%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.03%
    jit_opt_a.cse                        :     0.000043s :  0.05%
    jit_opt_a.replace_applicator         :     0.000027s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.02%
    rewriter_after_opt_a                 :     0.000035s :  0.04%
    convert_after_rewriter               :     0.000008s :  0.01%
    order_py_execute_after_rewriter      :     0.000006s :  0.01%
    mutable_eliminate                    :     0.000688s :  0.80%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.02%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.02%
    cconv                                :     0.000024s :  0.03%
    loop_unroll                          :     0.000473s :  0.55%
    jit_opt_after_cconv.c_1              :     0.000028s :  0.03%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.cse              :     0.000027s :  0.03%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.01%
    remove_dup_value                     :     0.000015s :  0.02%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000007s :  0.01%
    add_recomputation                    :     0.000050s :  0.06%
    cse_after_recomputation.cse          :     0.000012s :  0.01%
    auto_monad_reorder                   :     0.000015s :  0.02%
    get_jit_bprop_graph                  :     0.000003s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.01%
    opt_after_jit_grad                   :     0.000509s :  0.59%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.02%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000048s :  0.06%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007139s :  8.25%
    execute                              :     0.000008s :  0.01%


Time group info:
------[substitution.]   0.000214    27
 1.48% :     0.000003s :      2: substitution.elim_not_effective
 0.52% :     0.000001s :      2: substitution.fold_const_symbol
 3.36% :     0.000007s :      4: substitution.graph_param_transform
76.04% :     0.000163s :      4: substitution.inline
 1.68% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.10% :     0.000005s :      2: substitution.replace_old_param
 7.90% :     0.000017s :      3: substitution.tile_eliminate
 4.69% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.072852     2
97.41% :     0.070963s :      1: type_inference.infer
 2.59% :     0.001889s :      1: type_inference.specialize
------[replace.]   0.000072     6
66.61% :     0.000048s :      4: replace.inline
33.39% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000169     6
94.76% :     0.000160s :      4: match.inline
 5.24% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000141   834
 1.28% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.14% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     12: predicate.addn_check_dump
 1.39% :     0.000002s :     12: predicate.addn_zero_filter
 2.23% :     0.000003s :     12: predicate.arithmetic_simplify
 1.48% :     0.000002s :     12: predicate.cast_eliminate
 0.41% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     12: predicate.compare_switch_simplify
 1.11% :     0.000002s :     12: predicate.depend_value_elim
 1.05% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.16% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.75% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.36% :     0.000001s :      4: predicate.elim_not_effective
 0.64% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     12: predicate.environ_get_eliminate
 1.37% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.29% :     0.000000s :      4: predicate.fold_const_symbol
 1.12% :     0.000002s :      8: predicate.get_grad_eliminate
 0.33% :     0.000000s :      4: predicate.graph_param_transform
 5.09% :     0.000007s :     26: predicate.inline
 0.88% :     0.000001s :      8: predicate.inline_without_move
 0.43% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.15% :     0.000002s :      8: predicate.less_batch_normalization
 1.33% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 2.04% :     0.000003s :     18: predicate.load_eliminater
 1.66% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.88% :     0.000005s :     38: predicate.loop_unroll_before_grad
 1.98% :     0.000003s :     16: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     12: predicate.merge_addn
 1.44% :     0.000002s :     12: predicate.minmaximum_grad
 1.13% :     0.000002s :      4: predicate.mutable_eliminate
 0.72% :     0.000001s :      4: predicate.opt_reshape
 2.43% :     0.000003s :     18: predicate.partial_eliminate
 1.32% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.87% :     0.000003s :     12: predicate.reduce_eliminate
 1.49% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.98% :     0.000003s :     22: predicate.replace_applicator
 0.52% :     0.000001s :      8: predicate.replace_old_param
 0.35% :     0.000000s :      4: predicate.reset_defer_inline
 1.53% :     0.000002s :     12: predicate.reshape_eliminate
 1.34% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.17% :     0.000002s :     12: predicate.same_eliminate
 0.68% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      8: predicate.special_op_eliminate
 1.05% :     0.000001s :      8: predicate.specialize_transform
 1.27% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.18% :     0.000003s :     18: predicate.switch_defer_inline
 1.96% :     0.000003s :     18: predicate.switch_layer_defer_inline
 7.18% :     0.000010s :     60: predicate.switch_simplify
 1.52% :     0.000002s :     12: predicate.tile_eliminate
 1.28% :     0.000002s :     12: predicate.transpose_eliminate
 1.37% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.80% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.79% :     0.000003s :     18: predicate.updatestate_pure_node_eliminater
 3.76% :     0.000005s :     26: predicate.updatestate_useless_node_eliminater
 1.42% :     0.000002s :     12: predicate.value_based_eliminate
 0.44% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.002559    30
74.41% :     0.001904s :     23: func_graph_cloner_run.FuncGraphClonerGraph
 1.05% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
24.54% :     0.000628s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.094333    76
 0.06% :     0.000053s :      1: add_recomputation
 0.08% :     0.000072s :      1: auto_monad
 0.02% :     0.000018s :      1: auto_monad_reorder
 0.00% :     0.000003s :      1: backend_pass
 0.35% :     0.000328s :      1: bootstrap
 0.03% :     0.000027s :      1: cconv
 0.01% :     0.000010s :      1: convert_after_rewriter
 0.03% :     0.000026s :      1: cse_after_recomputation
 0.01% :     0.000010s :      1: environ_conv
 0.02% :     0.000022s :      1: event_method
 0.01% :     0.000012s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000008s :      1: graph_reusing
 7.91% :     0.007464s :      1: jit_opt_a
 0.23% :     0.000218s :      1: jit_opt_after_cconv
 0.07% :     0.000061s :      1: jit_opt_b
 0.51% :     0.000483s :      1: loop_unroll
 0.74% :     0.000695s :      1: mutable_eliminate
 1.03% :     0.000970s :     26: opt.transform.jit_opt_a
 0.07% :     0.000061s :      4: opt.transform.jit_opt_after_cconv
 0.03% :     0.000032s :      4: opt.transform.jit_opt_b
 0.02% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.04% :     0.000040s :      4: opt.transform.symbol_engine_opt
 0.55% :     0.000518s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.00% :     0.000005s :      1: pre_auto_parallel
 0.32% :     0.000298s :      1: py_interpret_to_execute
 0.02% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000018s :      1: remove_dup_value
 1.81% :     0.001706s :      1: renormalize.infer
 0.72% :     0.000683s :      1: renormalize.specialize
 0.01% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.04% :     0.000039s :      1: rewriter_after_opt_a
 0.10% :     0.000090s :      1: rewriter_before_opt_a
 0.09% :     0.000086s :      1: symbol_engine_optimizer
 7.58% :     0.007149s :      1: task_emit
77.34% :     0.072957s :      1: type_inference
 0.07% :     0.000069s :      1: validate


TotalTime = 0.0196095, [33]
[bootstrap]: 0.00030178
[type_inference]: 0.00465714
[event_method]: 1.794e-05
[auto_monad]: 6.245e-05
[graph_reusing]: 5.37001e-06
[pre_auto_parallel]: 2.26e-06
[py_interpret_to_execute]: 3.404e-05
[rewriter_before_opt_a]: 6.903e-05
[expand_dump_flag]: 3.04999e-06
[jit_opt_a]: 0.00576313, [2]
    [Cycle 1]: 0.0017987, [27]
        [switch_simplify]: 5.732e-05
        [loop_unroll]: 3.119e-05
        [a_1]: 0.00063997
        [with_stream_mark]: 1.996e-05
        [recompute_prepare]: 8.52998e-06
        [updatestate_depend_eliminate]: 3.7e-06
        [updatestate_assign_eliminate]: 3.22997e-06
        [updatestate_loads_eliminate]: 2.66e-06
        [parameter_eliminate]: 2.09999e-06
        [specialize_transform]: 6.04999e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.81e-06
        [meta_shard_fg_expand]: 2.44001e-06
        [get_grad_eliminate_]: 5.44998e-06
        [merge_forward]: 4.2e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.616e-05
        [j_node_and_user_rematch]: 9.37999e-06
        [meta_fg_expand]: 2.21e-06
        [replace_old_param]: 9.17001e-06
        [inline_without_move]: 5.35999e-06
        [renormalize]: 0.00072441
        [add_forward_monad_depend]: 5.87001e-06
        [auto_monad_grad]: 2.46e-06
        [auto_monad_eliminator]: 1.474e-05
        [cse]: 2.088e-05
        [replace_applicator]: 1.338e-05
    [Cycle 2]: 0.00034079, [27]
        [switch_simplify]: 5.53002e-06
        [loop_unroll]: 5.16002e-06
        [a_1]: 9.27e-05
        [with_stream_mark]: 1.096e-05
        [recompute_prepare]: 5.47999e-06
        [updatestate_depend_eliminate]: 2.99001e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [parameter_eliminate]: 1.57001e-06
        [specialize_transform]: 4.97999e-06
        [updatestate_useless_node_eliminater]: 5.27999e-06
        [accelerated_algorithm]: 5.52001e-06
        [meta_shard_fg_expand]: 1.55999e-06
        [get_grad_eliminate_]: 4.97999e-06
        [merge_forward]: 3.36001e-06
        [cell_reuse_recompute_pass]: 1.77001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.399e-05
        [j_node_and_user_rematch]: 8.58001e-06
        [meta_fg_expand]: 1.76e-06
        [replace_old_param]: 8.37e-06
        [inline_without_move]: 1.172e-05
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.67999e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 5.82001e-06
        [cse]: 1.116e-05
        [replace_applicator]: 5.29998e-06
[py_interpret_to_execute_after_opt_a]: 1.273e-05
[rewriter_after_opt_a]: 3.911e-05
[convert_after_rewriter]: 6.89001e-06
[order_py_execute_after_rewriter]: 5.40999e-06
[mutable_eliminate]: 0.00065776
[jit_opt_b]: 5.152e-05, [1]
    [Cycle 1]: 4.395e-05, [2]
        [frontend_op_eliminate]: 1.677e-05
        [inline_after_opt_a]: 1.559e-05
[cconv]: 2.77e-05
[loop_unroll]: 0.00047539
[jit_opt_after_cconv]: 0.00015337, [1]
    [Cycle 1]: 0.00014583, [11]
        [c_1]: 2.189e-05
        [parameter_eliminate]: 3.28e-06
        [updatestate_depend_eliminate]: 7.95e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [cse]: 2.057e-05
        [call_graph_tuple_transform]: 2.117e-05
        [tuple_list_get_item_eliminator]: 5.59998e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.80002e-06
[remove_dup_value]: 1.526e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 5.26998e-06
[add_recomputation]: 5.046e-05
[cse_after_recomputation]: 2.202e-05, [1]
    [Cycle 1]: 1.574e-05, [1]
        [cse]: 9.10001e-06
[auto_monad_reorder]: 1.599e-05
[get_jit_bprop_graph]: 2.15002e-06
[rewriter_after_jit_bprop_graph]: 6.65002e-06
[opt_after_jit_grad]: 0.00051501
[symbol_engine_optimizer]: 7.89e-05, [1]
    [Cycle 1]: 7.131e-05, [6]
        [build]: 3.79002e-06
        [elim_shapecalc]: 8.22e-06
        [elim_not_effective]: 1.495e-05
        [opt_reshape]: 5.87999e-06
        [fold_const_symbol]: 9.34e-06
        [renormalize]: 4.69998e-07
[validate]: 3.669e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00630838
[execute]: 7.97e-06
Sums
    bootstrap                            :     0.000302s :  1.96%
    type_inference                       :     0.004657s : 30.30%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000062s :  0.41%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000034s :  0.22%
    rewriter_before_opt_a                :     0.000069s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.41%
    jit_opt_a.loop_unroll                :     0.000036s :  0.24%
    jit_opt_a.a_1                        :     0.000733s :  4.77%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000017s :  0.11%
    jit_opt_a.renormalize                :     0.000725s :  4.71%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.13%
    jit_opt_a.cse                        :     0.000032s :  0.21%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000039s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000658s :  4.28%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000028s :  0.18%
    loop_unroll                          :     0.000475s :  3.09%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000015s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000515s :  3.35%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006308s : 41.04%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000299    23
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 1.80% :     0.000005s :      3: substitution.graph_param_transform
88.80% :     0.000265s :      4: substitution.inline
 1.14% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.36% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.71% :     0.000005s :      2: substitution.replace_old_param
 4.05% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004593     2
81.16% :     0.003727s :      1: type_inference.infer
18.84% :     0.000865s :      1: type_inference.specialize
------[replace.]   0.000069     6
67.97% :     0.000047s :      4: replace.inline
32.03% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000274     6
96.00% :     0.000263s :      4: match.inline
 4.00% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.24% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000002s :     10: predicate.addn_zero_filter
 2.34% :     0.000003s :     10: predicate.arithmetic_simplify
 1.66% :     0.000002s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.14% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.21% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.27% :     0.000002s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 6.23% :     0.000008s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.96% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.15% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.07% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.85% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.56% :     0.000003s :     16: predicate.switch_defer_inline
 2.28% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.32% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.06% :     0.000001s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.11% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000584    12
46.32% :     0.000271s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.81% :     0.000046s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.87% :     0.000268s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021415    76
 0.25% :     0.000054s :      1: add_recomputation
 0.31% :     0.000066s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.46% :     0.000313s :      1: bootstrap
 0.14% :     0.000031s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.93% :     0.005766s :      1: jit_opt_a
 0.73% :     0.000156s :      1: jit_opt_after_cconv
 0.25% :     0.000054s :      1: jit_opt_b
 2.26% :     0.000485s :      1: loop_unroll
 3.12% :     0.000668s :      1: mutable_eliminate
 4.48% :     0.000959s :     26: opt.transform.jit_opt_a
 0.24% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.45% :     0.000526s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000037s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 1.81% :     0.000388s :      1: renormalize.infer
 1.54% :     0.000329s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000042s :      1: rewriter_after_opt_a
 0.33% :     0.000072s :      1: rewriter_before_opt_a
 0.38% :     0.000082s :      1: symbol_engine_optimizer
29.52% :     0.006323s :      1: task_emit
21.81% :     0.004670s :      1: type_inference
 0.27% :     0.000058s :      1: validate


TotalTime = 0.0184016, [33]
[bootstrap]: 0.00034371
[type_inference]: 0.00413525
[event_method]: 1.645e-05
[auto_monad]: 6.515e-05
[graph_reusing]: 6.53e-06
[pre_auto_parallel]: 2.43e-06
[py_interpret_to_execute]: 3.483e-05
[rewriter_before_opt_a]: 7.467e-05
[expand_dump_flag]: 3.58e-06
[jit_opt_a]: 0.00517628, [2]
    [Cycle 1]: 0.00155493, [27]
        [switch_simplify]: 6.55e-05
        [loop_unroll]: 3.027e-05
        [a_1]: 0.0005181
        [with_stream_mark]: 1.563e-05
        [recompute_prepare]: 7.15e-06
        [updatestate_depend_eliminate]: 3.83001e-06
        [updatestate_assign_eliminate]: 3.73001e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 5.77999e-06
        [updatestate_useless_node_eliminater]: 5.15999e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 2.37999e-06
        [get_grad_eliminate_]: 5.47999e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 1.20999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.4e-05
        [j_node_and_user_rematch]: 8.77e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 8.74e-06
        [inline_without_move]: 5.32001e-06
        [renormalize]: 0.00060877
        [add_forward_monad_depend]: 5.21998e-06
        [auto_monad_grad]: 2.58e-06
        [auto_monad_eliminator]: 1.474e-05
        [cse]: 2.719e-05
        [replace_applicator]: 1.255e-05
    [Cycle 2]: 0.00034229, [27]
        [switch_simplify]: 6.09001e-06
        [loop_unroll]: 5.10999e-06
        [a_1]: 9.282e-05
        [with_stream_mark]: 9.01002e-06
        [recompute_prepare]: 5.17999e-06
        [updatestate_depend_eliminate]: 2.64999e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 5.34e-06
        [updatestate_useless_node_eliminater]: 4.91002e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.44998e-06
        [get_grad_eliminate_]: 5.43002e-06
        [merge_forward]: 2.53e-06
        [cell_reuse_recompute_pass]: 2.31e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.384e-05
        [j_node_and_user_rematch]: 8.95999e-06
        [meta_fg_expand]: 1.85001e-06
        [replace_old_param]: 8.82e-06
        [inline_without_move]: 5.36998e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.21002e-06
        [auto_monad_grad]: 1.22999e-06
        [auto_monad_eliminator]: 5.26998e-06
        [cse]: 1.122e-05
        [replace_applicator]: 5.90002e-06
[py_interpret_to_execute_after_opt_a]: 1.09e-05
[rewriter_after_opt_a]: 3.463e-05
[convert_after_rewriter]: 7.50998e-06
[order_py_execute_after_rewriter]: 6.45002e-06
[mutable_eliminate]: 0.00055649
[jit_opt_b]: 5.153e-05, [1]
    [Cycle 1]: 4.416e-05, [2]
        [frontend_op_eliminate]: 1.713e-05
        [inline_after_opt_a]: 1.542e-05
[cconv]: 2.563e-05
[loop_unroll]: 0.00043779
[jit_opt_after_cconv]: 0.00014497, [1]
    [Cycle 1]: 0.00013828, [11]
        [c_1]: 2.143e-05
        [parameter_eliminate]: 2.94001e-06
        [updatestate_depend_eliminate]: 5.62001e-06
        [updatestate_assign_eliminate]: 2.63e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 2.021e-05
        [call_graph_tuple_transform]: 1.903e-05
        [tuple_list_get_item_eliminator]: 5.59e-06
        [none_parameter_eliminate]: 1.63002e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.49e-06
[remove_dup_value]: 1.381e-05
[partial_unused_args_eliminate]: 2.59001e-06
[environ_conv]: 5.66e-06
[add_recomputation]: 4.96e-05
[cse_after_recomputation]: 1.975e-05, [1]
    [Cycle 1]: 1.414e-05, [1]
        [cse]: 8.62e-06
[auto_monad_reorder]: 1.562e-05
[get_jit_bprop_graph]: 1.35001e-06
[rewriter_after_jit_bprop_graph]: 5.15001e-06
[opt_after_jit_grad]: 0.00046928
[symbol_engine_optimizer]: 7.26e-05, [1]
    [Cycle 1]: 6.582e-05, [6]
        [build]: 3.33e-06
        [elim_shapecalc]: 7.55e-06
        [elim_not_effective]: 1.272e-05
        [opt_reshape]: 5.82001e-06
        [fold_const_symbol]: 9.17001e-06
        [renormalize]: 3.9002e-07
[validate]: 3.405e-05
[backend_pass]: 1.08001e-06
[task_emit]: 0.00638087
[execute]: 8.42e-06
Sums
    bootstrap                            :     0.000344s :  2.37%
    type_inference                       :     0.004135s : 28.48%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000065s :  0.45%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.24%
    rewriter_before_opt_a                :     0.000075s :  0.51%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000072s :  0.49%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000611s :  4.21%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.08%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000609s :  4.19%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.14%
    jit_opt_a.cse                        :     0.000038s :  0.26%
    jit_opt_a.replace_applicator         :     0.000018s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.08%
    rewriter_after_opt_a                 :     0.000035s :  0.24%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000556s :  3.83%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000026s :  0.18%
    loop_unroll                          :     0.000438s :  3.01%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000050s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000469s :  3.23%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006381s : 43.94%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000187    23
 1.25% :     0.000002s :      2: substitution.elim_not_effective
 0.78% :     0.000001s :      2: substitution.fold_const_symbol
 2.80% :     0.000005s :      3: substitution.graph_param_transform
82.02% :     0.000154s :      4: substitution.inline
 1.75% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.52% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.32% :     0.000004s :      2: substitution.replace_old_param
 6.56% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004074     2
79.81% :     0.003252s :      1: type_inference.infer
20.19% :     0.000823s :      1: type_inference.specialize
------[replace.]   0.000067     6
66.93% :     0.000045s :      4: replace.inline
33.07% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     6
93.18% :     0.000151s :      4: match.inline
 6.82% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.38% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.01% :     0.000002s :     10: predicate.arithmetic_simplify
 1.50% :     0.000002s :     10: predicate.cast_eliminate
 0.58% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000001s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.30% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     10: predicate.environ_get_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.17% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.15% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.73% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.02% :     0.000002s :     15: predicate.load_eliminater
 1.22% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.42% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.81% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.20% :     0.000001s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 1.92% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.83% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.42% :     0.000002s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.05% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.01% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000556    12
44.03% :     0.000245s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.75% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.22% :     0.000285s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019955    76
 0.26% :     0.000053s :      1: add_recomputation
 0.34% :     0.000069s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.79% :     0.000358s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.96% :     0.005180s :      1: jit_opt_a
 0.74% :     0.000148s :      1: jit_opt_after_cconv
 0.27% :     0.000054s :      1: jit_opt_b
 2.24% :     0.000447s :      1: loop_unroll
 2.83% :     0.000565s :      1: mutable_eliminate
 4.19% :     0.000836s :     26: opt.transform.jit_opt_a
 0.24% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.39% :     0.000478s :      1: opt_after_jit_grad
 0.04% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000038s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.38% :     0.000274s :      1: renormalize.infer
 1.64% :     0.000327s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000038s :      1: rewriter_after_opt_a
 0.39% :     0.000078s :      1: rewriter_before_opt_a
 0.38% :     0.000075s :      1: symbol_engine_optimizer
32.03% :     0.006392s :      1: task_emit
20.78% :     0.004146s :      1: type_inference
 0.26% :     0.000052s :      1: validate


TotalTime = 0.0145558, [30]
[bootstrap]: 0.0002468
[type_inference]: 0.00611882
[event_method]: 1.941e-05
[auto_monad]: 7.502e-05
[graph_reusing]: 5.66e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 4.745e-05
[rewriter_before_opt_a]: 8.143e-05
[expand_dump_flag]: 2.96001e-06
[jit_opt_a]: 0.00567439, [2]
    [Cycle 1]: 0.00191118, [27]
        [switch_simplify]: 0.00010689
        [loop_unroll]: 4.898e-05
        [a_1]: 0.00069584
        [with_stream_mark]: 1.471e-05
        [recompute_prepare]: 8.67e-06
        [updatestate_depend_eliminate]: 3.21999e-06
        [updatestate_assign_eliminate]: 2.20002e-06
        [updatestate_loads_eliminate]: 2.48002e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 5.78002e-06
        [updatestate_useless_node_eliminater]: 5.37001e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 3.43999e-06
        [cell_reuse_recompute_pass]: 1.03001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.311e-05
        [j_node_and_user_rematch]: 9.76e-06
        [meta_fg_expand]: 2.09999e-06
        [replace_old_param]: 8.38999e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.00073845
        [add_forward_monad_depend]: 6.52001e-06
        [auto_monad_grad]: 2.43e-06
        [auto_monad_eliminator]: 1.334e-05
        [cse]: 2.208e-05
        [replace_applicator]: 1.231e-05
    [Cycle 2]: 0.00042002, [27]
        [switch_simplify]: 5.52001e-06
        [loop_unroll]: 4.82998e-06
        [a_1]: 9.279e-05
        [with_stream_mark]: 9.99001e-06
        [recompute_prepare]: 5.08002e-06
        [updatestate_depend_eliminate]: 2.54001e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 1.86003e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 5.07e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 4.67e-06
        [meta_shard_fg_expand]: 1.65001e-06
        [get_grad_eliminate_]: 5.00999e-06
        [merge_forward]: 2.64999e-06
        [cell_reuse_recompute_pass]: 1.55999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.447e-05
        [j_node_and_user_rematch]: 9.52999e-06
        [meta_fg_expand]: 2.26998e-06
        [replace_old_param]: 1.242e-05
        [inline_without_move]: 6.16e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 2.26e-06
        [auto_monad_grad]: 1.27e-06
        [auto_monad_eliminator]: 6.77002e-06
        [cse]: 1.33e-05
        [replace_applicator]: 5.59e-06
[py_interpret_to_execute_after_opt_a]: 1.223e-05
[rewriter_after_opt_a]: 3.334e-05
[convert_after_rewriter]: 6.11e-06
[order_py_execute_after_rewriter]: 4.67e-06
[mutable_eliminate]: 0.00067748
[jit_opt_b]: 5.99e-05, [1]
    [Cycle 1]: 5.231e-05, [2]
        [frontend_op_eliminate]: 1.688e-05
        [inline_after_opt_a]: 1.603e-05
[cconv]: 2.527e-05
[loop_unroll]: 0.00047536
[jit_opt_after_cconv]: 0.00013976, [1]
    [Cycle 1]: 0.00013317, [11]
        [c_1]: 2.16e-05
        [parameter_eliminate]: 4.43999e-06
        [updatestate_depend_eliminate]: 4.33001e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 1.883e-05
        [call_graph_tuple_transform]: 1.854e-05
        [tuple_list_get_item_eliminator]: 5.31998e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 5.29981e-07
        [switch_simplify]: 5.27001e-06
[remove_dup_value]: 1.262e-05
[partial_unused_args_eliminate]: 2.19001e-06
[environ_conv]: 5.49e-06
[add_recomputation]: 4.588e-05
[cse_after_recomputation]: 2.045e-05, [1]
    [Cycle 1]: 1.536e-05, [1]
        [cse]: 9.40001e-06
[auto_monad_reorder]: 1.438e-05
[get_jit_bprop_graph]: 2.11e-06
[rewriter_after_jit_bprop_graph]: 5.14e-06
[opt_after_jit_grad]: 0.00047455
[symbol_engine_optimizer]: 7.299e-05, [1]
    [Cycle 1]: 6.661e-05, [6]
        [build]: 3.91999e-06
        [elim_shapecalc]: 7.76001e-06
        [elim_not_effective]: 1.183e-05
        [opt_reshape]: 5.51e-06
        [fold_const_symbol]: 9.17001e-06
        [renormalize]: 4.80009e-07
[validate]: 3.475e-05
Sums
    bootstrap                            :     0.000247s :  2.34%
    type_inference                       :     0.006119s : 57.89%
    event_method                         :     0.000019s :  0.18%
    auto_monad                           :     0.000075s :  0.71%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000047s :  0.45%
    rewriter_before_opt_a                :     0.000081s :  0.77%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000112s :  1.06%
    jit_opt_a.loop_unroll                :     0.000054s :  0.51%
    jit_opt_a.a_1                        :     0.000789s :  7.46%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000021s :  0.20%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000739s :  6.99%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.19%
    jit_opt_a.cse                        :     0.000035s :  0.33%
    jit_opt_a.replace_applicator         :     0.000018s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.12%
    rewriter_after_opt_a                 :     0.000033s :  0.32%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000677s :  6.41%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.15%
    cconv                                :     0.000025s :  0.24%
    loop_unroll                          :     0.000475s :  4.50%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000046s :  0.43%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000014s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000475s :  4.49%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.33%


Time group info:
------[substitution.]   0.000282    26
 0.67% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 1.79% :     0.000005s :      3: substitution.graph_param_transform
85.02% :     0.000239s :      6: substitution.inline
 1.22% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.77% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.62% :     0.000005s :      2: substitution.replace_old_param
 3.05% :     0.000009s :      1: substitution.switch_simplify
 4.38% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006060     2
80.34% :     0.004869s :      1: type_inference.infer
19.66% :     0.001191s :      1: type_inference.specialize
------[replace.]   0.000134     9
49.47% :     0.000067s :      6: replace.inline
26.34% :     0.000035s :      1: replace.switch_simplify
24.19% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000255     9
92.58% :     0.000236s :      6: match.inline
 3.05% :     0.000008s :      1: match.switch_simplify
 4.37% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   801
 1.35% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.02% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     12: predicate.addn_check_dump
 1.31% :     0.000002s :     12: predicate.addn_zero_filter
 2.08% :     0.000003s :     12: predicate.arithmetic_simplify
 1.40% :     0.000002s :     12: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000002s :     12: predicate.compare_switch_simplify
 1.45% :     0.000002s :     12: predicate.depend_value_elim
 1.14% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.73% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     12: predicate.environ_get_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.52% :     0.000007s :     26: predicate.inline
 1.14% :     0.000002s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     17: predicate.load_eliminater
 1.08% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.35% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.93% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000002s :     12: predicate.merge_addn
 1.26% :     0.000002s :     12: predicate.minmaximum_grad
 1.31% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     17: predicate.partial_eliminate
 1.18% :     0.000002s :     12: predicate.print_const_string_wrapper
 2.05% :     0.000003s :     12: predicate.reduce_eliminate
 1.54% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     20: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     12: predicate.reshape_eliminate
 1.34% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     12: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.77% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.97% :     0.000004s :     20: predicate.switch_defer_inline
 2.38% :     0.000003s :     20: predicate.switch_layer_defer_inline
 7.92% :     0.000011s :     65: predicate.switch_simplify
 1.23% :     0.000002s :     12: predicate.tile_eliminate
 1.20% :     0.000002s :     12: predicate.transpose_eliminate
 1.66% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.71% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.66% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     12: predicate.value_based_eliminate
 0.38% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000922    15
37.96% :     0.000350s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.82% :     0.000100s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.22% :     0.000472s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016432    72
 0.30% :     0.000049s :      1: add_recomputation
 0.48% :     0.000078s :      1: auto_monad
 0.10% :     0.000017s :      1: auto_monad_reorder
 1.57% :     0.000258s :      1: bootstrap
 0.17% :     0.000028s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
34.56% :     0.005679s :      1: jit_opt_a
 0.87% :     0.000143s :      1: jit_opt_after_cconv
 0.38% :     0.000063s :      1: jit_opt_b
 2.94% :     0.000483s :      1: loop_unroll
 4.17% :     0.000686s :      1: mutable_eliminate
 6.57% :     0.001080s :     26: opt.transform.jit_opt_a
 0.29% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.93% :     0.000482s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.31% :     0.000051s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000015s :      1: remove_dup_value
 2.57% :     0.000422s :      1: renormalize.infer
 1.87% :     0.000308s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000036s :      1: rewriter_after_opt_a
 0.51% :     0.000084s :      1: rewriter_before_opt_a
 0.46% :     0.000076s :      1: symbol_engine_optimizer
37.30% :     0.006129s :      1: type_inference


TotalTime = 0.0184898, [33]
[bootstrap]: 0.00027092
[type_inference]: 0.00429914
[event_method]: 1.796e-05
[auto_monad]: 6.575e-05
[graph_reusing]: 5.99e-06
[pre_auto_parallel]: 2.76e-06
[py_interpret_to_execute]: 3.329e-05
[rewriter_before_opt_a]: 7.115e-05
[expand_dump_flag]: 3.43e-06
[jit_opt_a]: 0.00516851, [2]
    [Cycle 1]: 0.00147753, [27]
        [switch_simplify]: 5.772e-05
        [loop_unroll]: 3.001e-05
        [a_1]: 0.0005315
        [with_stream_mark]: 1.641e-05
        [recompute_prepare]: 8.08999e-06
        [updatestate_depend_eliminate]: 4.35e-06
        [updatestate_assign_eliminate]: 3.04001e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 5.94999e-06
        [updatestate_useless_node_eliminater]: 4.96997e-06
        [accelerated_algorithm]: 5.97999e-06
        [meta_shard_fg_expand]: 2.69001e-06
        [get_grad_eliminate_]: 6.16998e-06
        [merge_forward]: 3.73999e-06
        [cell_reuse_recompute_pass]: 1.21002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.443e-05
        [j_node_and_user_rematch]: 9.14e-06
        [meta_fg_expand]: 2.46e-06
        [replace_old_param]: 9.17001e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 0.00052252
        [add_forward_monad_depend]: 5.11997e-06
        [auto_monad_grad]: 2.27999e-06
        [auto_monad_eliminator]: 1.33e-05
        [cse]: 2.689e-05
        [replace_applicator]: 1.225e-05
    [Cycle 2]: 0.00032375, [27]
        [switch_simplify]: 6.06998e-06
        [loop_unroll]: 4.99e-06
        [a_1]: 9.296e-05
        [with_stream_mark]: 8.62e-06
        [recompute_prepare]: 5.06002e-06
        [updatestate_depend_eliminate]: 2.88e-06
        [updatestate_assign_eliminate]: 2.00002e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 4.95999e-06
        [updatestate_useless_node_eliminater]: 4.94998e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 4.53001e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.135e-05
        [j_node_and_user_rematch]: 7.77998e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 7.12002e-06
        [inline_without_move]: 4.73001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 4.88001e-06
        [cse]: 1.056e-05
        [replace_applicator]: 5.76e-06
[py_interpret_to_execute_after_opt_a]: 8.80001e-06
[rewriter_after_opt_a]: 3.182e-05
[convert_after_rewriter]: 7.21001e-06
[order_py_execute_after_rewriter]: 4.90001e-06
[mutable_eliminate]: 0.00052204
[jit_opt_b]: 5.112e-05, [1]
    [Cycle 1]: 4.379e-05, [2]
        [frontend_op_eliminate]: 1.73e-05
        [inline_after_opt_a]: 1.447e-05
[cconv]: 2.378e-05
[loop_unroll]: 0.00042672
[jit_opt_after_cconv]: 0.00013664, [1]
    [Cycle 1]: 0.00013014, [11]
        [c_1]: 2.081e-05
        [parameter_eliminate]: 2.06e-06
        [updatestate_depend_eliminate]: 5.00001e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 1.96998e-06
        [cse]: 1.631e-05
        [call_graph_tuple_transform]: 1.756e-05
        [tuple_list_get_item_eliminator]: 5.54998e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.69999e-06
[remove_dup_value]: 1.365e-05
[partial_unused_args_eliminate]: 2.66e-06
[environ_conv]: 5.06002e-06
[add_recomputation]: 4.598e-05
[cse_after_recomputation]: 2.111e-05, [1]
    [Cycle 1]: 1.562e-05, [1]
        [cse]: 9.94001e-06
[auto_monad_reorder]: 1.505e-05
[get_jit_bprop_graph]: 1.49e-06
[rewriter_after_jit_bprop_graph]: 5.06002e-06
[opt_after_jit_grad]: 0.00050917
[symbol_engine_optimizer]: 7.284e-05, [1]
    [Cycle 1]: 6.663e-05, [6]
        [build]: 3.24001e-06
        [elim_shapecalc]: 7.51999e-06
        [elim_not_effective]: 1.284e-05
        [opt_reshape]: 5.79e-06
        [fold_const_symbol]: 8.68001e-06
        [renormalize]: 6.60017e-07
[validate]: 3.378e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.00641642
[execute]: 7.13e-06
Sums
    bootstrap                            :     0.000271s :  1.86%
    type_inference                       :     0.004299s : 29.59%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000066s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.23%
    rewriter_before_opt_a                :     0.000071s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.44%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000624s :  4.30%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000523s :  3.60%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000037s :  0.26%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000032s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000522s :  3.59%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.16%
    loop_unroll                          :     0.000427s :  2.94%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000046s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000509s :  3.50%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006416s : 44.17%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000188    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.79% :     0.000001s :      2: substitution.fold_const_symbol
 2.64% :     0.000005s :      3: substitution.graph_param_transform
82.45% :     0.000155s :      4: substitution.inline
 1.75% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.30% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.17% :     0.000004s :      2: substitution.replace_old_param
 6.92% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004237     2
81.34% :     0.003446s :      1: type_inference.infer
18.66% :     0.000791s :      1: type_inference.specialize
------[replace.]   0.000067     6
65.50% :     0.000044s :      4: replace.inline
34.50% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000164     6
92.76% :     0.000153s :      4: match.inline
 7.24% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.24% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.02% :     0.000002s :     10: predicate.arithmetic_simplify
 1.39% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.37% :     0.000002s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.32% :     0.000007s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.33% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.93% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.80% :     0.000002s :     10: predicate.minmaximum_grad
 1.54% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.24% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.34% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.41% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.62% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.52% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.80% :     0.000010s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.24% :     0.000002s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.16% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.02% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.53% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000527    12
48.45% :     0.000255s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.87% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.68% :     0.000241s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019953    76
 0.24% :     0.000049s :      1: add_recomputation
 0.35% :     0.000069s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.42% :     0.000282s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.92% :     0.005171s :      1: jit_opt_a
 0.70% :     0.000139s :      1: jit_opt_after_cconv
 0.27% :     0.000054s :      1: jit_opt_b
 2.18% :     0.000434s :      1: loop_unroll
 2.66% :     0.000531s :      1: mutable_eliminate
 4.20% :     0.000839s :     26: opt.transform.jit_opt_a
 0.23% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.59% :     0.000518s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.24% :     0.000248s :      1: renormalize.infer
 1.34% :     0.000267s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000035s :      1: rewriter_after_opt_a
 0.37% :     0.000074s :      1: rewriter_before_opt_a
 0.38% :     0.000075s :      1: symbol_engine_optimizer
32.22% :     0.006428s :      1: task_emit
21.61% :     0.004312s :      1: type_inference
 0.27% :     0.000053s :      1: validate


TotalTime = 0.016814, [30]
[bootstrap]: 0.00032233
[type_inference]: 0.00741067
[event_method]: 2.186e-05
[auto_monad]: 8.491e-05
[graph_reusing]: 6.96001e-06
[pre_auto_parallel]: 3.06001e-06
[py_interpret_to_execute]: 5.716e-05
[rewriter_before_opt_a]: 8.825e-05
[expand_dump_flag]: 2.98998e-06
[jit_opt_a]: 0.00621247, [2]
    [Cycle 1]: 0.00218442, [27]
        [switch_simplify]: 0.00013099
        [loop_unroll]: 3.528e-05
        [a_1]: 0.00076102
        [with_stream_mark]: 2.16e-05
        [recompute_prepare]: 1.221e-05
        [updatestate_depend_eliminate]: 4.32e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.83998e-06
        [parameter_eliminate]: 2.74999e-06
        [specialize_transform]: 7.03e-06
        [updatestate_useless_node_eliminater]: 5.09998e-06
        [accelerated_algorithm]: 6.58998e-06
        [meta_shard_fg_expand]: 2.32001e-06
        [get_grad_eliminate_]: 5.15999e-06
        [merge_forward]: 4.08001e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.686e-05
        [j_node_and_user_rematch]: 1.077e-05
        [meta_fg_expand]: 2.96001e-06
        [replace_old_param]: 9.60001e-06
        [inline_without_move]: 5.75001e-06
        [renormalize]: 0.00083801
        [add_forward_monad_depend]: 6.74001e-06
        [auto_monad_grad]: 2.46e-06
        [auto_monad_eliminator]: 1.65e-05
        [cse]: 3.026e-05
        [replace_applicator]: 1.894e-05
    [Cycle 2]: 0.00036088, [27]
        [switch_simplify]: 5.77999e-06
        [loop_unroll]: 4.82e-06
        [a_1]: 9.598e-05
        [with_stream_mark]: 1.538e-05
        [recompute_prepare]: 5.45001e-06
        [updatestate_depend_eliminate]: 3.83001e-06
        [updatestate_assign_eliminate]: 2.87002e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 1.55999e-06
        [specialize_transform]: 5.02999e-06
        [updatestate_useless_node_eliminater]: 4.76002e-06
        [accelerated_algorithm]: 5.24e-06
        [meta_shard_fg_expand]: 1.89999e-06
        [get_grad_eliminate_]: 4.51002e-06
        [merge_forward]: 3.80998e-06
        [cell_reuse_recompute_pass]: 2.48e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.74e-05
        [j_node_and_user_rematch]: 8.76002e-06
        [meta_fg_expand]: 1.84e-06
        [replace_old_param]: 9.15001e-06
        [inline_without_move]: 4.95999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 2.10002e-06
        [auto_monad_grad]: 1.27e-06
        [auto_monad_eliminator]: 6.90998e-06
        [cse]: 1.291e-05
        [replace_applicator]: 6.78e-06
[py_interpret_to_execute_after_opt_a]: 1.484e-05
[rewriter_after_opt_a]: 3.778e-05
[convert_after_rewriter]: 7.35e-06
[order_py_execute_after_rewriter]: 5.19e-06
[mutable_eliminate]: 0.00078399
[jit_opt_b]: 5.699e-05, [1]
    [Cycle 1]: 4.716e-05, [2]
        [frontend_op_eliminate]: 1.74e-05
        [inline_after_opt_a]: 1.714e-05
[cconv]: 2.751e-05
[loop_unroll]: 0.00053181
[jit_opt_after_cconv]: 0.00015849, [1]
    [Cycle 1]: 0.00015112, [11]
        [c_1]: 2.238e-05
        [parameter_eliminate]: 3.68999e-06
        [updatestate_depend_eliminate]: 6.94999e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.64999e-06
        [cse]: 2.661e-05
        [call_graph_tuple_transform]: 2.078e-05
        [tuple_list_get_item_eliminator]: 5.20999e-06
        [none_parameter_eliminate]: 1.48002e-06
        [renormalize]: 8.2e-07
        [switch_simplify]: 5.87001e-06
[remove_dup_value]: 1.48e-05
[partial_unused_args_eliminate]: 2.09e-06
[environ_conv]: 5.44998e-06
[add_recomputation]: 5.567e-05
[cse_after_recomputation]: 2.256e-05, [1]
    [Cycle 1]: 1.592e-05, [1]
        [cse]: 9.73002e-06
[auto_monad_reorder]: 1.646e-05
[get_jit_bprop_graph]: 2.03997e-06
[rewriter_after_jit_bprop_graph]: 8.25e-06
[opt_after_jit_grad]: 0.00053321
[symbol_engine_optimizer]: 8.179e-05, [1]
    [Cycle 1]: 7.463e-05, [6]
        [build]: 5.56e-06
        [elim_shapecalc]: 8.67998e-06
        [elim_not_effective]: 1.454e-05
        [opt_reshape]: 6.68e-06
        [fold_const_symbol]: 8.72e-06
        [renormalize]: 6.50005e-07
[validate]: 4.065e-05
Sums
    bootstrap                            :     0.000322s :  2.58%
    type_inference                       :     0.007411s : 59.40%
    event_method                         :     0.000022s :  0.18%
    auto_monad                           :     0.000085s :  0.68%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000057s :  0.46%
    rewriter_before_opt_a                :     0.000088s :  0.71%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000137s :  1.10%
    jit_opt_a.loop_unroll                :     0.000040s :  0.32%
    jit_opt_a.a_1                        :     0.000857s :  6.87%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000008s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.000838s :  6.72%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.19%
    jit_opt_a.cse                        :     0.000043s :  0.35%
    jit_opt_a.replace_applicator         :     0.000026s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.12%
    rewriter_after_opt_a                 :     0.000038s :  0.30%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000784s :  6.28%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.14%
    cconv                                :     0.000028s :  0.22%
    loop_unroll                          :     0.000532s :  4.26%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000027s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000015s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000056s :  0.45%
    cse_after_recomputation.cse          :     0.000010s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.07%
    opt_after_jit_grad                   :     0.000533s :  4.27%
    symbol_engine_optimizer.build        :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.33%


Time group info:
------[substitution.]   0.000323    26
 0.54% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 1.84% :     0.000006s :      3: substitution.graph_param_transform
85.13% :     0.000275s :      6: substitution.inline
 1.18% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.66% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.90% :     0.000006s :      2: substitution.replace_old_param
 2.84% :     0.000009s :      1: substitution.switch_simplify
 4.47% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007327     2
83.38% :     0.006109s :      1: type_inference.infer
16.62% :     0.001218s :      1: type_inference.specialize
------[replace.]   0.000151     9
49.55% :     0.000075s :      6: replace.inline
25.66% :     0.000039s :      1: replace.switch_simplify
24.79% :     0.000037s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000293     9
92.69% :     0.000271s :      6: match.inline
 2.87% :     0.000008s :      1: match.switch_simplify
 4.44% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000144   801
 1.27% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     12: predicate.addn_check_dump
 1.30% :     0.000002s :     12: predicate.addn_zero_filter
 2.36% :     0.000003s :     12: predicate.arithmetic_simplify
 1.18% :     0.000002s :     12: predicate.cast_eliminate
 0.33% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000002s :     12: predicate.compare_switch_simplify
 1.49% :     0.000002s :     12: predicate.depend_value_elim
 1.08% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.04% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     12: predicate.environ_get_eliminate
 1.03% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.83% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.35% :     0.000008s :     26: predicate.inline
 1.12% :     0.000002s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000002s :      6: predicate.less_batch_normalization
 1.67% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 2.14% :     0.000003s :     17: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.93% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.25% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000002s :     12: predicate.merge_addn
 1.03% :     0.000001s :     12: predicate.minmaximum_grad
 1.74% :     0.000003s :      3: predicate.mutable_eliminate
 0.35% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     17: predicate.partial_eliminate
 1.10% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     12: predicate.reduce_eliminate
 1.65% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.50% :     0.000002s :     20: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.68% :     0.000002s :     12: predicate.reshape_eliminate
 1.40% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.97% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000002s :     12: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.80% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.66% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.07% :     0.000004s :     20: predicate.switch_defer_inline
 2.21% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.54% :     0.000012s :     65: predicate.switch_simplify
 1.12% :     0.000002s :     12: predicate.tile_eliminate
 1.19% :     0.000002s :     12: predicate.transpose_eliminate
 1.55% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.28% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.51% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     12: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000884    15
46.46% :     0.000411s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.32% :     0.000109s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.22% :     0.000365s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018887    72
 0.31% :     0.000059s :      1: add_recomputation
 0.47% :     0.000089s :      1: auto_monad
 0.10% :     0.000019s :      1: auto_monad_reorder
 1.82% :     0.000343s :      1: bootstrap
 0.16% :     0.000030s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.13% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.14% :     0.000027s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000010s :      1: graph_reusing
32.92% :     0.006217s :      1: jit_opt_a
 0.85% :     0.000161s :      1: jit_opt_after_cconv
 0.32% :     0.000060s :      1: jit_opt_b
 2.87% :     0.000542s :      1: loop_unroll
 4.22% :     0.000797s :      1: mutable_eliminate
 6.20% :     0.001170s :     26: opt.transform.jit_opt_a
 0.27% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000027s :      4: opt.transform.jit_opt_b
 0.08% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.87% :     0.000542s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.32% :     0.000060s :      1: py_interpret_to_execute
 0.10% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000018s :      1: remove_dup_value
 2.26% :     0.000426s :      1: renormalize.infer
 2.13% :     0.000402s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000043s :      1: rewriter_after_opt_a
 0.48% :     0.000091s :      1: rewriter_before_opt_a
 0.45% :     0.000085s :      1: symbol_engine_optimizer
39.33% :     0.007428s :      1: type_inference


TotalTime = 0.0199855, [33]
[bootstrap]: 0.00028062
[type_inference]: 0.00411487
[event_method]: 1.67e-05
[auto_monad]: 6.316e-05
[graph_reusing]: 6.11e-06
[pre_auto_parallel]: 2.68e-06
[py_interpret_to_execute]: 3.653e-05
[rewriter_before_opt_a]: 7.697e-05
[expand_dump_flag]: 3.18e-06
[jit_opt_a]: 0.00577108, [2]
    [Cycle 1]: 0.0017268, [27]
        [switch_simplify]: 6.069e-05
        [loop_unroll]: 2.969e-05
        [a_1]: 0.00055754
        [with_stream_mark]: 2.214e-05
        [recompute_prepare]: 8.62e-06
        [updatestate_depend_eliminate]: 3.63999e-06
        [updatestate_assign_eliminate]: 3.66001e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 6.26e-06
        [updatestate_useless_node_eliminater]: 5.09003e-06
        [accelerated_algorithm]: 5.49998e-06
        [meta_shard_fg_expand]: 2.15002e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 3.48999e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.564e-05
        [j_node_and_user_rematch]: 9.10999e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 9.94999e-06
        [inline_without_move]: 5.74e-06
        [renormalize]: 0.00070075
        [add_forward_monad_depend]: 6.59001e-06
        [auto_monad_grad]: 2.71999e-06
        [auto_monad_eliminator]: 1.484e-05
        [cse]: 3.141e-05
        [replace_applicator]: 1.328e-05
    [Cycle 2]: 0.00033078, [27]
        [switch_simplify]: 6.15002e-06
        [loop_unroll]: 4.85999e-06
        [a_1]: 9.129e-05
        [with_stream_mark]: 1.086e-05
        [recompute_prepare]: 5.03002e-06
        [updatestate_depend_eliminate]: 2.73e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.11002e-06
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 5.08002e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.13001e-06
        [get_grad_eliminate_]: 4.84e-06
        [merge_forward]: 2.73e-06
        [cell_reuse_recompute_pass]: 2.39001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.319e-05
        [j_node_and_user_rematch]: 7.98999e-06
        [meta_fg_expand]: 1.81e-06
        [replace_old_param]: 8.05e-06
        [inline_without_move]: 4.48001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 1.19e-06
        [auto_monad_eliminator]: 5.54e-06
        [cse]: 1.164e-05
        [replace_applicator]: 5.40001e-06
[py_interpret_to_execute_after_opt_a]: 1.13e-05
[rewriter_after_opt_a]: 3.584e-05
[convert_after_rewriter]: 7.57002e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00061447
[jit_opt_b]: 5.019e-05, [1]
    [Cycle 1]: 4.253e-05, [2]
        [frontend_op_eliminate]: 1.566e-05
        [inline_after_opt_a]: 1.515e-05
[cconv]: 2.457e-05
[loop_unroll]: 0.00044355
[jit_opt_after_cconv]: 0.00014296, [1]
    [Cycle 1]: 0.00013649, [11]
        [c_1]: 2.215e-05
        [parameter_eliminate]: 2.38002e-06
        [updatestate_depend_eliminate]: 5.62001e-06
        [updatestate_assign_eliminate]: 2.38002e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 2.035e-05
        [call_graph_tuple_transform]: 1.919e-05
        [tuple_list_get_item_eliminator]: 5.31998e-06
        [none_parameter_eliminate]: 1.55001e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.54e-06
[remove_dup_value]: 1.36e-05
[partial_unused_args_eliminate]: 2.23002e-06
[environ_conv]: 5.35001e-06
[add_recomputation]: 8.968e-05
[cse_after_recomputation]: 2.186e-05, [1]
    [Cycle 1]: 1.587e-05, [1]
        [cse]: 9.95002e-06
[auto_monad_reorder]: 1.538e-05
[get_jit_bprop_graph]: 2.43e-06
[rewriter_after_jit_bprop_graph]: 6.20002e-06
[opt_after_jit_grad]: 0.00050627
[symbol_engine_optimizer]: 7.896e-05, [1]
    [Cycle 1]: 7.141e-05, [6]
        [build]: 4.66002e-06
        [elim_shapecalc]: 7.63999e-06
        [elim_not_effective]: 1.434e-05
        [opt_reshape]: 5.79e-06
        [fold_const_symbol]: 9.80002e-06
        [renormalize]: 6.50005e-07
[validate]: 4.056e-05
[backend_pass]: 7.79983e-07
[task_emit]: 0.00728268
[execute]: 8.89e-06
Sums
    bootstrap                            :     0.000281s :  1.80%
    type_inference                       :     0.004115s : 26.32%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000063s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.23%
    rewriter_before_opt_a                :     0.000077s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.43%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000649s :  4.15%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000701s :  4.48%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000043s :  0.28%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000614s :  3.93%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000025s :  0.16%
    loop_unroll                          :     0.000444s :  2.84%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000090s :  0.57%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000506s :  3.24%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.26%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007283s : 46.58%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000207    23
 1.19% :     0.000002s :      2: substitution.elim_not_effective
 0.82% :     0.000002s :      2: substitution.fold_const_symbol
 2.65% :     0.000005s :      3: substitution.graph_param_transform
82.68% :     0.000171s :      4: substitution.inline
 1.46% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.22% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.51% :     0.000005s :      2: substitution.replace_old_param
 6.47% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004055     2
79.75% :     0.003234s :      1: type_inference.infer
20.25% :     0.000821s :      1: type_inference.specialize
------[replace.]   0.000072     6
65.90% :     0.000048s :      4: replace.inline
34.10% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000180     6
93.25% :     0.000168s :      4: match.inline
 6.75% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.21% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.72% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.13% :     0.000003s :     10: predicate.arithmetic_simplify
 1.26% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.14% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.44% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.45% :     0.000002s :     10: predicate.environ_get_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 6.00% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :      6: predicate.less_batch_normalization
 1.96% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.24% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.08% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.82% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.29% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.17% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.27% :     0.000002s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.06% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.48% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000602    12
41.25% :     0.000248s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.65% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
53.10% :     0.000320s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021662    76
 0.43% :     0.000094s :      1: add_recomputation
 0.31% :     0.000067s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.36% :     0.000296s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.66% :     0.005774s :      1: jit_opt_a
 0.67% :     0.000146s :      1: jit_opt_after_cconv
 0.24% :     0.000053s :      1: jit_opt_b
 2.09% :     0.000452s :      1: loop_unroll
 2.88% :     0.000623s :      1: mutable_eliminate
 4.01% :     0.000868s :     26: opt.transform.jit_opt_a
 0.22% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.38% :     0.000516s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000039s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.49% :     0.000323s :      1: renormalize.infer
 1.70% :     0.000369s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000039s :      1: rewriter_after_opt_a
 0.37% :     0.000080s :      1: rewriter_before_opt_a
 0.38% :     0.000082s :      1: symbol_engine_optimizer
33.69% :     0.007297s :      1: task_emit
19.06% :     0.004128s :      1: type_inference
 0.29% :     0.000062s :      1: validate


TotalTime = 0.0214612, [33]
[bootstrap]: 0.00029227
[type_inference]: 0.00449199
[event_method]: 1.658e-05
[auto_monad]: 5.066e-05
[graph_reusing]: 5.13002e-06
[pre_auto_parallel]: 2.20002e-06
[py_interpret_to_execute]: 3.426e-05
[rewriter_before_opt_a]: 6.574e-05
[expand_dump_flag]: 2.63e-06
[jit_opt_a]: 0.00565452, [2]
    [Cycle 1]: 0.0017927, [27]
        [switch_simplify]: 6.088e-05
        [loop_unroll]: 3.025e-05
        [a_1]: 0.00057408
        [with_stream_mark]: 2.209e-05
        [recompute_prepare]: 8.32e-06
        [updatestate_depend_eliminate]: 4.08001e-06
        [updatestate_assign_eliminate]: 2.94001e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 1.85001e-06
        [specialize_transform]: 5.57001e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 5.74999e-06
        [meta_shard_fg_expand]: 2.13998e-06
        [get_grad_eliminate_]: 5.32001e-06
        [merge_forward]: 3.97e-06
        [cell_reuse_recompute_pass]: 1.35001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.461e-05
        [j_node_and_user_rematch]: 9.51e-06
        [meta_fg_expand]: 2.27001e-06
        [replace_old_param]: 8.94e-06
        [inline_without_move]: 5.19e-06
        [renormalize]: 0.00074643
        [add_forward_monad_depend]: 6.26e-06
        [auto_monad_grad]: 2.41e-06
        [auto_monad_eliminator]: 1.687e-05
        [cse]: 2.951e-05
        [replace_applicator]: 1.569e-05
    [Cycle 2]: 0.00034061, [27]
        [switch_simplify]: 6.01e-06
        [loop_unroll]: 4.98001e-06
        [a_1]: 9.191e-05
        [with_stream_mark]: 1.175e-05
        [recompute_prepare]: 4.95999e-06
        [updatestate_depend_eliminate]: 3.12002e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 2.19999e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 4.96002e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.25001e-06
        [get_grad_eliminate_]: 4.72e-06
        [merge_forward]: 2.96001e-06
        [cell_reuse_recompute_pass]: 2.63998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.435e-05
        [j_node_and_user_rematch]: 8.87e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 8.38001e-06
        [inline_without_move]: 4.90001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.49e-06
        [auto_monad_grad]: 9.09989e-07
        [auto_monad_eliminator]: 6.28998e-06
        [cse]: 1.231e-05
        [replace_applicator]: 4.81002e-06
[py_interpret_to_execute_after_opt_a]: 1.244e-05
[rewriter_after_opt_a]: 3.725e-05
[convert_after_rewriter]: 1.846e-05
[order_py_execute_after_rewriter]: 6.17999e-06
[mutable_eliminate]: 0.00062048
[jit_opt_b]: 5.38e-05, [1]
    [Cycle 1]: 4.57e-05, [2]
        [frontend_op_eliminate]: 1.782e-05
        [inline_after_opt_a]: 1.561e-05
[cconv]: 2.433e-05
[loop_unroll]: 0.00046132
[jit_opt_after_cconv]: 0.00014817, [1]
    [Cycle 1]: 0.00014067, [11]
        [c_1]: 2.021e-05
        [parameter_eliminate]: 2.89999e-06
        [updatestate_depend_eliminate]: 6.33998e-06
        [updatestate_assign_eliminate]: 2.30002e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 2.403e-05
        [call_graph_tuple_transform]: 1.955e-05
        [tuple_list_get_item_eliminator]: 5.37999e-06
        [none_parameter_eliminate]: 1.42999e-06
        [renormalize]: 5.3001e-07
        [switch_simplify]: 5.39e-06
[remove_dup_value]: 1.409e-05
[partial_unused_args_eliminate]: 2.57001e-06
[environ_conv]: 5.17999e-06
[add_recomputation]: 5.262e-05
[cse_after_recomputation]: 2.373e-05, [1]
    [Cycle 1]: 1.8e-05, [1]
        [cse]: 1.095e-05
[auto_monad_reorder]: 1.708e-05
[get_jit_bprop_graph]: 2.31e-06
[rewriter_after_jit_bprop_graph]: 7.8e-06
[opt_after_jit_grad]: 0.00057845
[symbol_engine_optimizer]: 7.564e-05, [1]
    [Cycle 1]: 6.709e-05, [6]
        [build]: 4.23001e-06
        [elim_shapecalc]: 8.23001e-06
        [elim_not_effective]: 1.295e-05
        [opt_reshape]: 5.39e-06
        [fold_const_symbol]: 8.35001e-06
        [renormalize]: 6.49976e-07
[validate]: 3.768e-05
[backend_pass]: 1.02e-06
[task_emit]: 0.00842453
[execute]: 8.67e-06
Sums
    bootstrap                            :     0.000292s :  1.69%
    type_inference                       :     0.004492s : 25.99%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000051s :  0.29%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000034s :  0.20%
    rewriter_before_opt_a                :     0.000066s :  0.38%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.20%
    jit_opt_a.a_1                        :     0.000666s :  3.85%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000747s :  4.32%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.13%
    jit_opt_a.cse                        :     0.000042s :  0.24%
    jit_opt_a.replace_applicator         :     0.000021s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.22%
    convert_after_rewriter               :     0.000018s :  0.11%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000620s :  3.59%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000024s :  0.14%
    loop_unroll                          :     0.000461s :  2.67%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000053s :  0.30%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000578s :  3.35%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008425s : 48.75%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000237    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.50% :     0.000006s :      3: substitution.graph_param_transform
85.66% :     0.000203s :      4: substitution.inline
 1.77% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.05% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000005s :      2: substitution.replace_old_param
 4.41% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004432     2
81.90% :     0.003629s :      1: type_inference.infer
18.10% :     0.000802s :      1: type_inference.specialize
------[replace.]   0.000070     6
67.90% :     0.000048s :      4: replace.inline
32.10% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000210     6
95.56% :     0.000201s :      4: match.inline
 4.44% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.62% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.06% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.68% :     0.000002s :     10: predicate.addn_zero_filter
 2.39% :     0.000003s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.79% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.15% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.61% :     0.000007s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.63% :     0.000002s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.72% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 1.90% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.78% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.87% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.25% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.77% :     0.000009s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.59% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.07% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000559    12
44.92% :     0.000251s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.99% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.09% :     0.000258s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023200    76
 0.24% :     0.000056s :      1: add_recomputation
 0.23% :     0.000054s :      1: auto_monad
 0.08% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.33% :     0.000309s :      1: bootstrap
 0.12% :     0.000027s :      1: cconv
 0.10% :     0.000023s :      1: convert_after_rewriter
 0.11% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000020s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
24.39% :     0.005658s :      1: jit_opt_a
 0.65% :     0.000151s :      1: jit_opt_after_cconv
 0.24% :     0.000056s :      1: jit_opt_b
 2.02% :     0.000469s :      1: loop_unroll
 2.71% :     0.000630s :      1: mutable_eliminate
 3.83% :     0.000888s :     26: opt.transform.jit_opt_a
 0.20% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.54% :     0.000588s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.77% :     0.000412s :      1: renormalize.infer
 1.40% :     0.000326s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000040s :      1: rewriter_after_opt_a
 0.30% :     0.000069s :      1: rewriter_before_opt_a
 0.34% :     0.000078s :      1: symbol_engine_optimizer
36.40% :     0.008444s :      1: task_emit
19.42% :     0.004505s :      1: type_inference
 0.26% :     0.000060s :      1: validate


TotalTime = 0.101718, [33]
[bootstrap]: 0.00031918
[type_inference]: 0.0824225
[event_method]: 1.955e-05
[auto_monad]: 7.006e-05
[graph_reusing]: 6.84001e-06
[pre_auto_parallel]: 2.37999e-06
[py_interpret_to_execute]: 0.00037604
[rewriter_before_opt_a]: 9.416e-05
[expand_dump_flag]: 3.76999e-06
[jit_opt_a]: 0.00862105, [2]
    [Cycle 1]: 0.00383401, [27]
        [switch_simplify]: 5.688e-05
        [loop_unroll]: 3.347e-05
        [a_1]: 0.00059313
        [with_stream_mark]: 1.602e-05
        [recompute_prepare]: 9.40001e-06
        [updatestate_depend_eliminate]: 3.16001e-06
        [updatestate_assign_eliminate]: 2.53998e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 7.20003e-06
        [updatestate_useless_node_eliminater]: 6.23998e-06
        [accelerated_algorithm]: 6.61999e-06
        [meta_shard_fg_expand]: 1.39003e-06
        [get_grad_eliminate_]: 6.68e-06
        [merge_forward]: 3.33998e-06
        [cell_reuse_recompute_pass]: 1.35001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.276e-05
        [j_node_and_user_rematch]: 1.024e-05
        [meta_fg_expand]: 1.86e-06
        [replace_old_param]: 9.23002e-06
        [inline_without_move]: 6.41e-06
        [renormalize]: 0.00278557
        [add_forward_monad_depend]: 7.44002e-06
        [auto_monad_grad]: 2.47001e-06
        [auto_monad_eliminator]: 1.621e-05
        [cse]: 2.392e-05
        [replace_applicator]: 2.065e-05
    [Cycle 2]: 0.00042644, [27]
        [switch_simplify]: 7.33999e-06
        [loop_unroll]: 6.23002e-06
        [a_1]: 0.00015359
        [with_stream_mark]: 1.294e-05
        [recompute_prepare]: 6.17001e-06
        [updatestate_depend_eliminate]: 3.18e-06
        [updatestate_assign_eliminate]: 2.94001e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 5.82001e-06
        [updatestate_useless_node_eliminater]: 6.02999e-06
        [accelerated_algorithm]: 6.53998e-06
        [meta_shard_fg_expand]: 2.02999e-06
        [get_grad_eliminate_]: 6.14001e-06
        [merge_forward]: 3.18998e-06
        [cell_reuse_recompute_pass]: 4.75999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.266e-05
        [j_node_and_user_rematch]: 9.61e-06
        [meta_fg_expand]: 2.21998e-06
        [replace_old_param]: 8.92999e-06
        [inline_without_move]: 5.52001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.84e-06
        [auto_monad_grad]: 1.10001e-06
        [auto_monad_eliminator]: 6.59001e-06
        [cse]: 1.314e-05
        [replace_applicator]: 6.44001e-06
[py_interpret_to_execute_after_opt_a]: 1.645e-05
[rewriter_after_opt_a]: 3.834e-05
[convert_after_rewriter]: 7.05998e-06
[order_py_execute_after_rewriter]: 5.02e-06
[mutable_eliminate]: 0.00076622
[jit_opt_b]: 6.934e-05, [1]
    [Cycle 1]: 5.921e-05, [2]
        [frontend_op_eliminate]: 2.713e-05
        [inline_after_opt_a]: 1.969e-05
[cconv]: 2.913e-05
[loop_unroll]: 0.00059813
[jit_opt_after_cconv]: 0.00018428, [1]
    [Cycle 1]: 0.0001765, [11]
        [c_1]: 2.864e-05
        [parameter_eliminate]: 3.56001e-06
        [updatestate_depend_eliminate]: 7.25003e-06
        [updatestate_assign_eliminate]: 2.64001e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [cse]: 3.163e-05
        [call_graph_tuple_transform]: 2.689e-05
        [tuple_list_get_item_eliminator]: 7.8e-06
        [none_parameter_eliminate]: 1.86e-06
        [renormalize]: 6.89994e-07
        [switch_simplify]: 6.81001e-06
[remove_dup_value]: 1.644e-05
[partial_unused_args_eliminate]: 3.56001e-06
[environ_conv]: 8.70001e-06
[add_recomputation]: 5.486e-05
[cse_after_recomputation]: 2.351e-05, [1]
    [Cycle 1]: 1.765e-05, [1]
        [cse]: 1.154e-05
[auto_monad_reorder]: 9.948e-05
[get_jit_bprop_graph]: 2.06e-06
[rewriter_after_jit_bprop_graph]: 1.021e-05
[opt_after_jit_grad]: 0.00067121
[symbol_engine_optimizer]: 9.386e-05, [1]
    [Cycle 1]: 8.173e-05, [6]
        [build]: 4.32e-06
        [elim_shapecalc]: 9.31e-06
        [elim_not_effective]: 1.948e-05
        [opt_reshape]: 9.99999e-06
        [fold_const_symbol]: 9.34e-06
        [renormalize]: 1.11997e-06
[validate]: 0.00023324
[backend_pass]: 9.5999e-07
[task_emit]: 0.00661443
[execute]: 5.71e-06
Sums
    bootstrap                            :     0.000319s :  0.33%
    type_inference                       :     0.082422s : 85.26%
    event_method                         :     0.000020s :  0.02%
    auto_monad                           :     0.000070s :  0.07%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.00%
    py_interpret_to_execute              :     0.000376s :  0.39%
    rewriter_before_opt_a                :     0.000094s :  0.10%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000064s :  0.07%
    jit_opt_a.loop_unroll                :     0.000040s :  0.04%
    jit_opt_a.a_1                        :     0.000747s :  0.77%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.03%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.00%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.00%
    jit_opt_a.specialize_transform       :     0.000013s :  0.01%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.01%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.01%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.00%
    jit_opt_a.get_grad_eliminate_        :     0.000013s :  0.01%
    jit_opt_a.merge_forward              :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.03%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.02%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.00%
    jit_opt_a.replace_old_param          :     0.000018s :  0.02%
    jit_opt_a.inline_without_move        :     0.000012s :  0.01%
    jit_opt_a.renormalize                :     0.002786s :  2.88%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.02%
    jit_opt_a.cse                        :     0.000037s :  0.04%
    jit_opt_a.replace_applicator         :     0.000027s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.02%
    rewriter_after_opt_a                 :     0.000038s :  0.04%
    convert_after_rewriter               :     0.000007s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000766s :  0.79%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.02%
    cconv                                :     0.000029s :  0.03%
    loop_unroll                          :     0.000598s :  0.62%
    jit_opt_after_cconv.c_1              :     0.000029s :  0.03%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000032s :  0.03%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.01%
    remove_dup_value                     :     0.000016s :  0.02%
    partial_unused_args_eliminate        :     0.000004s :  0.00%
    environ_conv                         :     0.000009s :  0.01%
    add_recomputation                    :     0.000055s :  0.06%
    cse_after_recomputation.cse          :     0.000012s :  0.01%
    auto_monad_reorder                   :     0.000099s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.01%
    opt_after_jit_grad                   :     0.000671s :  0.69%
    symbol_engine_optimizer.build        :     0.000004s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.02%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000233s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006614s :  6.84%
    execute                              :     0.000006s :  0.01%


Time group info:
------[substitution.]   0.000222    27
 0.77% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 3.57% :     0.000008s :      4: substitution.graph_param_transform
78.26% :     0.000174s :      4: substitution.inline
 1.65% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.98% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.81% :     0.000004s :      2: substitution.replace_old_param
 8.07% :     0.000018s :      3: substitution.tile_eliminate
 3.36% :     0.000007s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.082330     2
97.21% :     0.080033s :      1: type_inference.infer
 2.79% :     0.002297s :      1: type_inference.specialize
------[replace.]   0.000073     6
67.53% :     0.000050s :      4: replace.inline
32.47% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000177     6
96.48% :     0.000171s :      4: match.inline
 3.52% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000148   834
 1.39% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.76% :     0.000003s :      4: predicate.ad_related_special_op_eliminate
 1.02% :     0.000002s :     12: predicate.addn_check_dump
 1.31% :     0.000002s :     12: predicate.addn_zero_filter
 2.76% :     0.000004s :     12: predicate.arithmetic_simplify
 1.22% :     0.000002s :     12: predicate.cast_eliminate
 0.45% :     0.000001s :      4: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     12: predicate.compare_switch_simplify
 1.46% :     0.000002s :     12: predicate.depend_value_elim
 1.06% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     12: predicate.dict_set_item_eliminator
 1.12% :     0.000002s :      4: predicate.dumpgradient_eliminate
 0.69% :     0.000001s :      4: predicate.elim_not_effective
 0.64% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     12: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     12: predicate.environ_get_eliminate
 1.31% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      4: predicate.fold_const_symbol
 1.13% :     0.000002s :      8: predicate.get_grad_eliminate
 0.25% :     0.000000s :      4: predicate.graph_param_transform
 4.53% :     0.000007s :     26: predicate.inline
 0.96% :     0.000001s :      8: predicate.inline_without_move
 0.40% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      8: predicate.less_batch_normalization
 1.63% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     18: predicate.load_eliminater
 1.57% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.87% :     0.000006s :     38: predicate.loop_unroll_before_grad
 1.72% :     0.000003s :     16: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     12: predicate.merge_addn
 1.00% :     0.000001s :     12: predicate.minmaximum_grad
 2.32% :     0.000003s :      4: predicate.mutable_eliminate
 0.68% :     0.000001s :      4: predicate.opt_reshape
 1.83% :     0.000003s :     18: predicate.partial_eliminate
 1.35% :     0.000002s :     12: predicate.print_const_string_wrapper
 2.02% :     0.000003s :     12: predicate.reduce_eliminate
 1.46% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.93% :     0.000003s :     22: predicate.replace_applicator
 0.51% :     0.000001s :      8: predicate.replace_old_param
 0.31% :     0.000000s :      4: predicate.reset_defer_inline
 1.83% :     0.000003s :     12: predicate.reshape_eliminate
 1.25% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.14% :     0.000002s :     12: predicate.same_eliminate
 0.56% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.25% :     0.000002s :      8: predicate.special_op_eliminate
 0.89% :     0.000001s :      8: predicate.specialize_transform
 1.24% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.19% :     0.000003s :     18: predicate.switch_defer_inline
 2.04% :     0.000003s :     18: predicate.switch_layer_defer_inline
 6.58% :     0.000010s :     60: predicate.switch_simplify
 1.89% :     0.000003s :     12: predicate.tile_eliminate
 1.76% :     0.000003s :     12: predicate.transpose_eliminate
 1.33% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.80% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000003s :     12: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.69% :     0.000003s :     18: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000004s :     26: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     12: predicate.value_based_eliminate
 0.39% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.003206    34
73.03% :     0.002341s :     27: func_graph_cloner_run.FuncGraphClonerGraph
 1.39% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
25.58% :     0.000820s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.105657    76
 0.05% :     0.000058s :      1: add_recomputation
 0.07% :     0.000073s :      1: auto_monad
 0.10% :     0.000103s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.31% :     0.000331s :      1: bootstrap
 0.03% :     0.000032s :      1: cconv
 0.01% :     0.000010s :      1: convert_after_rewriter
 0.02% :     0.000026s :      1: cse_after_recomputation
 0.01% :     0.000011s :      1: environ_conv
 0.02% :     0.000024s :      1: event_method
 0.01% :     0.000009s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000009s :      1: graph_reusing
 8.16% :     0.008625s :      1: jit_opt_a
 0.18% :     0.000187s :      1: jit_opt_after_cconv
 0.07% :     0.000073s :      1: jit_opt_b
 0.58% :     0.000608s :      1: loop_unroll
 0.74% :     0.000778s :      1: mutable_eliminate
 0.94% :     0.000990s :     26: opt.transform.jit_opt_a
 0.06% :     0.000066s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000040s :      4: opt.transform.jit_opt_b
 0.02% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000037s :      1: opt.transform.opt_after_jit_grad
 0.04% :     0.000042s :      4: opt.transform.symbol_engine_opt
 0.65% :     0.000682s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000006s :      1: partial_unused_args_eliminate
 0.00% :     0.000004s :      1: pre_auto_parallel
 0.36% :     0.000382s :      1: py_interpret_to_execute
 0.02% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000019s :      1: remove_dup_value
 1.83% :     0.001932s :      1: renormalize.infer
 0.80% :     0.000844s :      1: renormalize.specialize
 0.01% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.04% :     0.000041s :      1: rewriter_after_opt_a
 0.09% :     0.000100s :      1: rewriter_before_opt_a
 0.09% :     0.000097s :      1: symbol_engine_optimizer
 6.27% :     0.006628s :      1: task_emit
78.03% :     0.082440s :      1: type_inference
 0.24% :     0.000257s :      1: validate


TotalTime = 0.0175192, [30]
[bootstrap]: 0.00032543
[type_inference]: 0.00764592
[event_method]: 2.146e-05
[auto_monad]: 8.786e-05
[graph_reusing]: 6.79001e-06
[pre_auto_parallel]: 3.13e-06
[py_interpret_to_execute]: 5.58e-05
[rewriter_before_opt_a]: 8.795e-05
[expand_dump_flag]: 3.41999e-06
[jit_opt_a]: 0.00650083, [2]
    [Cycle 1]: 0.00227688, [27]
        [switch_simplify]: 0.00012941
        [loop_unroll]: 3.464e-05
        [a_1]: 0.0008174
        [with_stream_mark]: 2.548e-05
        [recompute_prepare]: 1.072e-05
        [updatestate_depend_eliminate]: 4.23999e-06
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 3.04001e-06
        [parameter_eliminate]: 2.34001e-06
        [specialize_transform]: 6.95998e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 5.74e-06
        [meta_shard_fg_expand]: 2.27999e-06
        [get_grad_eliminate_]: 5.88998e-06
        [merge_forward]: 3.81999e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.721e-05
        [j_node_and_user_rematch]: 1.04e-05
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 9.56e-06
        [inline_without_move]: 5.87001e-06
        [renormalize]: 0.00088082
        [add_forward_monad_depend]: 9.26998e-06
        [auto_monad_grad]: 2.79999e-06
        [auto_monad_eliminator]: 2.03e-05
        [cse]: 2.944e-05
        [replace_applicator]: 2.175e-05
    [Cycle 2]: 0.00036884, [27]
        [switch_simplify]: 6.21e-06
        [loop_unroll]: 5.27001e-06
        [a_1]: 9.599e-05
        [with_stream_mark]: 1.596e-05
        [recompute_prepare]: 6.19001e-06
        [updatestate_depend_eliminate]: 3.56999e-06
        [updatestate_assign_eliminate]: 2.63998e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.62001e-06
        [specialize_transform]: 5.85002e-06
        [updatestate_useless_node_eliminater]: 5.56e-06
        [accelerated_algorithm]: 5.72999e-06
        [meta_shard_fg_expand]: 1.50999e-06
        [get_grad_eliminate_]: 4.88001e-06
        [merge_forward]: 3.73999e-06
        [cell_reuse_recompute_pass]: 2.54001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.861e-05
        [j_node_and_user_rematch]: 8.68001e-06
        [meta_fg_expand]: 1.86e-06
        [replace_old_param]: 8.95999e-06
        [inline_without_move]: 4.87998e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.70001e-06
        [auto_monad_grad]: 8.89995e-07
        [auto_monad_eliminator]: 5.80002e-06
        [cse]: 1.409e-05
        [replace_applicator]: 5.79e-06
[py_interpret_to_execute_after_opt_a]: 1.838e-05
[rewriter_after_opt_a]: 4.098e-05
[convert_after_rewriter]: 8.38999e-06
[order_py_execute_after_rewriter]: 5.58002e-06
[mutable_eliminate]: 0.00084342
[jit_opt_b]: 5.625e-05, [1]
    [Cycle 1]: 4.666e-05, [2]
        [frontend_op_eliminate]: 1.755e-05
        [inline_after_opt_a]: 1.645e-05
[cconv]: 3.238e-05
[loop_unroll]: 0.00053928
[jit_opt_after_cconv]: 0.00018702, [1]
    [Cycle 1]: 0.00017698, [11]
        [c_1]: 2.304e-05
        [parameter_eliminate]: 4.27e-06
        [updatestate_depend_eliminate]: 8.07998e-06
        [updatestate_assign_eliminate]: 2.72001e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [cse]: 3.035e-05
        [call_graph_tuple_transform]: 2.385e-05
        [tuple_list_get_item_eliminator]: 5.91998e-06
        [none_parameter_eliminate]: 1.68002e-06
        [renormalize]: 7.79983e-07
        [switch_simplify]: 6.18002e-06
[remove_dup_value]: 1.477e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 5.50001e-06
[add_recomputation]: 6.785e-05
[cse_after_recomputation]: 2.331e-05, [1]
    [Cycle 1]: 1.736e-05, [1]
        [cse]: 1.066e-05
[auto_monad_reorder]: 1.642e-05
[get_jit_bprop_graph]: 2.31998e-06
[rewriter_after_jit_bprop_graph]: 8.06001e-06
[opt_after_jit_grad]: 0.00058172
[symbol_engine_optimizer]: 8.345e-05, [1]
    [Cycle 1]: 7.536e-05, [6]
        [build]: 6.12001e-06
        [elim_shapecalc]: 8.08999e-06
        [elim_not_effective]: 1.493e-05
        [opt_reshape]: 6.67002e-06
        [fold_const_symbol]: 8.89003e-06
        [renormalize]: 8.89995e-07
[validate]: 4.158e-05
Sums
    bootstrap                            :     0.000325s :  2.51%
    type_inference                       :     0.007646s : 58.91%
    event_method                         :     0.000021s :  0.17%
    auto_monad                           :     0.000088s :  0.68%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000056s :  0.43%
    rewriter_before_opt_a                :     0.000088s :  0.68%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000136s :  1.04%
    jit_opt_a.loop_unroll                :     0.000040s :  0.31%
    jit_opt_a.a_1                        :     0.000913s :  7.04%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.08%
    jit_opt_a.merge_forward              :     0.000008s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.14%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.000881s :  6.79%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.20%
    jit_opt_a.cse                        :     0.000044s :  0.34%
    jit_opt_a.replace_applicator         :     0.000028s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.14%
    rewriter_after_opt_a                 :     0.000041s :  0.32%
    convert_after_rewriter               :     0.000008s :  0.06%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000843s :  6.50%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.13%
    cconv                                :     0.000032s :  0.25%
    loop_unroll                          :     0.000539s :  4.16%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000030s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000015s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000068s :  0.52%
    cse_after_recomputation.cse          :     0.000011s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.06%
    opt_after_jit_grad                   :     0.000582s :  4.48%
    symbol_engine_optimizer.build        :     0.000006s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000042s :  0.32%


Time group info:
------[substitution.]   0.000380    26
 0.61% :     0.000002s :      2: substitution.elim_not_effective
 0.35% :     0.000001s :      2: substitution.fold_const_symbol
 1.61% :     0.000006s :      3: substitution.graph_param_transform
87.36% :     0.000332s :      6: substitution.inline
 1.00% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.55% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.36% :     0.000005s :      2: substitution.replace_old_param
 2.43% :     0.000009s :      1: substitution.switch_simplify
 3.73% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007570     2
84.17% :     0.006372s :      1: type_inference.infer
15.83% :     0.001198s :      1: type_inference.specialize
------[replace.]   0.000153     9
49.18% :     0.000075s :      6: replace.inline
24.80% :     0.000038s :      1: replace.switch_simplify
26.01% :     0.000040s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000350     9
93.92% :     0.000328s :      6: match.inline
 2.42% :     0.000008s :      1: match.switch_simplify
 3.67% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000147   801
 1.24% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.28% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     12: predicate.addn_check_dump
 1.26% :     0.000002s :     12: predicate.addn_zero_filter
 2.42% :     0.000004s :     12: predicate.arithmetic_simplify
 1.43% :     0.000002s :     12: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000002s :     12: predicate.compare_switch_simplify
 1.87% :     0.000003s :     12: predicate.depend_value_elim
 1.26% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.56% :     0.000001s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.05% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.05% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     12: predicate.environ_get_eliminate
 1.11% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.17% :     0.000002s :      6: predicate.get_grad_eliminate
 0.43% :     0.000001s :      3: predicate.graph_param_transform
 5.72% :     0.000008s :     26: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000002s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.57% :     0.000002s :     17: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.17% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.17% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000002s :     12: predicate.merge_addn
 1.00% :     0.000001s :     12: predicate.minmaximum_grad
 1.70% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000003s :     17: predicate.partial_eliminate
 1.09% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     12: predicate.reduce_eliminate
 1.51% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000003s :     20: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     12: predicate.reshape_eliminate
 1.27% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000002s :     12: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000002s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.88% :     0.000004s :     20: predicate.switch_defer_inline
 2.46% :     0.000004s :     20: predicate.switch_layer_defer_inline
 8.24% :     0.000012s :     65: predicate.switch_simplify
 1.20% :     0.000002s :     12: predicate.tile_eliminate
 1.13% :     0.000002s :     12: predicate.transpose_eliminate
 1.43% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.24% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.26% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     12: predicate.value_based_eliminate
 0.35% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000929    15
47.47% :     0.000441s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.88% :     0.000120s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.66% :     0.000368s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019698    72
 0.37% :     0.000072s :      1: add_recomputation
 0.47% :     0.000092s :      1: auto_monad
 0.10% :     0.000020s :      1: auto_monad_reorder
 1.72% :     0.000340s :      1: bootstrap
 0.18% :     0.000036s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.13% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.13% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000010s :      1: graph_reusing
33.02% :     0.006505s :      1: jit_opt_a
 0.97% :     0.000191s :      1: jit_opt_after_cconv
 0.30% :     0.000060s :      1: jit_opt_b
 2.79% :     0.000550s :      1: loop_unroll
 4.37% :     0.000861s :      1: mutable_eliminate
 6.23% :     0.001227s :     26: opt.transform.jit_opt_a
 0.28% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000027s :      4: opt.transform.jit_opt_b
 0.08% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.01% :     0.000594s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000059s :      1: py_interpret_to_execute
 0.11% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000017s :      1: remove_dup_value
 2.39% :     0.000470s :      1: renormalize.infer
 2.04% :     0.000402s :      1: renormalize.specialize
 0.05% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000045s :      1: rewriter_after_opt_a
 0.46% :     0.000091s :      1: rewriter_before_opt_a
 0.44% :     0.000086s :      1: symbol_engine_optimizer
38.90% :     0.007663s :      1: type_inference


TotalTime = 0.0203716, [33]
[bootstrap]: 0.00033299
[type_inference]: 0.00490231
[event_method]: 1.804e-05
[auto_monad]: 6.441e-05
[graph_reusing]: 5.66e-06
[pre_auto_parallel]: 2.63e-06
[py_interpret_to_execute]: 3.882e-05
[rewriter_before_opt_a]: 7.164e-05
[expand_dump_flag]: 2.66e-06
[jit_opt_a]: 0.00621005, [2]
    [Cycle 1]: 0.00166861, [27]
        [switch_simplify]: 5.233e-05
        [loop_unroll]: 3.028e-05
        [a_1]: 0.00058177
        [with_stream_mark]: 1.818e-05
        [recompute_prepare]: 8.28001e-06
        [updatestate_depend_eliminate]: 3.38999e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [parameter_eliminate]: 1.26002e-06
        [specialize_transform]: 6.65998e-06
        [updatestate_useless_node_eliminater]: 5.56e-06
        [accelerated_algorithm]: 6.12001e-06
        [meta_shard_fg_expand]: 2.12001e-06
        [get_grad_eliminate_]: 5.75001e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 1.47999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.258e-05
        [j_node_and_user_rematch]: 9.01002e-06
        [meta_fg_expand]: 2.18002e-06
        [replace_old_param]: 8.07e-06
        [inline_without_move]: 5.77001e-06
        [renormalize]: 0.00067291
        [add_forward_monad_depend]: 5.56e-06
        [auto_monad_grad]: 1.93002e-06
        [auto_monad_eliminator]: 1.159e-05
        [cse]: 2.011e-05
        [replace_applicator]: 1.274e-05
    [Cycle 2]: 0.00033915, [27]
        [switch_simplify]: 5.81e-06
        [loop_unroll]: 4.82e-06
        [a_1]: 9.23e-05
        [with_stream_mark]: 1.105e-05
        [recompute_prepare]: 6.33998e-06
        [updatestate_depend_eliminate]: 2.86999e-06
        [updatestate_assign_eliminate]: 2.31998e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.38002e-06
        [specialize_transform]: 4.98001e-06
        [updatestate_useless_node_eliminater]: 5.10999e-06
        [accelerated_algorithm]: 5.37999e-06
        [meta_shard_fg_expand]: 1.57001e-06
        [get_grad_eliminate_]: 4.62e-06
        [merge_forward]: 3.14001e-06
        [cell_reuse_recompute_pass]: 1.98002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.356e-05
        [j_node_and_user_rematch]: 8.51002e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 8.37998e-06
        [inline_without_move]: 4.65999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.50999e-06
        [auto_monad_grad]: 1.22999e-06
        [auto_monad_eliminator]: 6.72002e-06
        [cse]: 1.139e-05
        [replace_applicator]: 5.04998e-06
[py_interpret_to_execute_after_opt_a]: 1.268e-05
[rewriter_after_opt_a]: 3.317e-05
[convert_after_rewriter]: 6.83e-06
[order_py_execute_after_rewriter]: 5.94999e-06
[mutable_eliminate]: 0.00061924
[jit_opt_b]: 5.192e-05, [1]
    [Cycle 1]: 4.494e-05, [2]
        [frontend_op_eliminate]: 1.636e-05
        [inline_after_opt_a]: 1.574e-05
[cconv]: 2.446e-05
[loop_unroll]: 0.00046289
[jit_opt_after_cconv]: 0.0001432, [1]
    [Cycle 1]: 0.0001366, [11]
        [c_1]: 2.122e-05
        [parameter_eliminate]: 3.22002e-06
        [updatestate_depend_eliminate]: 5.89e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.08998e-06
        [cse]: 1.883e-05
        [call_graph_tuple_transform]: 1.988e-05
        [tuple_list_get_item_eliminator]: 5.35999e-06
        [none_parameter_eliminate]: 1.27999e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 5.29e-06
[remove_dup_value]: 1.059e-05
[partial_unused_args_eliminate]: 1.71998e-06
[environ_conv]: 4.87e-06
[add_recomputation]: 3.856e-05
[cse_after_recomputation]: 2.053e-05, [1]
    [Cycle 1]: 1.506e-05, [1]
        [cse]: 8.90999e-06
[auto_monad_reorder]: 1.228e-05
[get_jit_bprop_graph]: 2.06998e-06
[rewriter_after_jit_bprop_graph]: 6.07001e-06
[opt_after_jit_grad]: 0.00050961
[symbol_engine_optimizer]: 7.323e-05, [1]
    [Cycle 1]: 6.613e-05, [6]
        [build]: 3.28998e-06
        [elim_shapecalc]: 7.16999e-06
        [elim_not_effective]: 1.308e-05
        [opt_reshape]: 5.66998e-06
        [fold_const_symbol]: 8.92999e-06
        [renormalize]: 4.49974e-07
[validate]: 3.302e-05
[backend_pass]: 8.99978e-07
[task_emit]: 0.0064393
[execute]: 5.77001e-06
Sums
    bootstrap                            :     0.000333s :  2.14%
    type_inference                       :     0.004902s : 31.53%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000064s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.25%
    rewriter_before_opt_a                :     0.000072s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.37%
    jit_opt_a.loop_unroll                :     0.000035s :  0.23%
    jit_opt_a.a_1                        :     0.000674s :  4.34%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000673s :  4.33%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000032s :  0.20%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000033s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000619s :  3.98%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000024s :  0.16%
    loop_unroll                          :     0.000463s :  2.98%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000011s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000039s :  0.25%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000510s :  3.28%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006439s : 41.42%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000188    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.60% :     0.000005s :      3: substitution.graph_param_transform
82.91% :     0.000156s :      4: substitution.inline
 1.92% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.24% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.18% :     0.000004s :      2: substitution.replace_old_param
 6.43% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004828     2
81.68% :     0.003944s :      1: type_inference.infer
18.32% :     0.000884s :      1: type_inference.specialize
------[replace.]   0.000069     6
63.81% :     0.000044s :      4: replace.inline
36.19% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000164     6
93.34% :     0.000153s :      4: match.inline
 6.66% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.65% :     0.000002s :     10: predicate.addn_zero_filter
 2.31% :     0.000003s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.61% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.33% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.24% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.23% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.92% :     0.000002s :     15: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.29% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.61% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000002s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.67% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.07% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.49% :     0.000001s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.04% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.21% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.35% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.85% :     0.000009s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.55% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000596    12
47.26% :     0.000282s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.01% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.73% :     0.000278s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022042    76
 0.19% :     0.000042s :      1: add_recomputation
 0.31% :     0.000068s :      1: auto_monad
 0.07% :     0.000015s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.57% :     0.000347s :      1: bootstrap
 0.12% :     0.000027s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.04% :     0.000010s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
28.19% :     0.006213s :      1: jit_opt_a
 0.66% :     0.000146s :      1: jit_opt_after_cconv
 0.25% :     0.000055s :      1: jit_opt_b
 2.14% :     0.000472s :      1: loop_unroll
 2.86% :     0.000629s :      1: mutable_eliminate
 4.03% :     0.000889s :     26: opt.transform.jit_opt_a
 0.22% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.36% :     0.000519s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000042s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000013s :      1: remove_dup_value
 1.60% :     0.000354s :      1: renormalize.infer
 1.42% :     0.000312s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000036s :      1: rewriter_after_opt_a
 0.34% :     0.000075s :      1: rewriter_before_opt_a
 0.34% :     0.000076s :      1: symbol_engine_optimizer
29.27% :     0.006451s :      1: task_emit
22.33% :     0.004921s :      1: type_inference
 0.24% :     0.000052s :      1: validate


TotalTime = 0.0156641, [30]
[bootstrap]: 0.000281
[type_inference]: 0.00666062
[event_method]: 2.203e-05
[auto_monad]: 8.219e-05
[graph_reusing]: 6.07001e-06
[pre_auto_parallel]: 2.49001e-06
[py_interpret_to_execute]: 5.286e-05
[rewriter_before_opt_a]: 8.668e-05
[expand_dump_flag]: 3.33998e-06
[jit_opt_a]: 0.00597685, [2]
    [Cycle 1]: 0.00208429, [27]
        [switch_simplify]: 0.00012855
        [loop_unroll]: 3.945e-05
        [a_1]: 0.00077195
        [with_stream_mark]: 1.987e-05
        [recompute_prepare]: 8.60001e-06
        [updatestate_depend_eliminate]: 3.36999e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.27001e-06
        [parameter_eliminate]: 1.37e-06
        [specialize_transform]: 6.97002e-06
        [updatestate_useless_node_eliminater]: 5.24998e-06
        [accelerated_algorithm]: 5.34e-06
        [meta_shard_fg_expand]: 1.87001e-06
        [get_grad_eliminate_]: 5.07e-06
        [merge_forward]: 3.94002e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.365e-05
        [j_node_and_user_rematch]: 9.95002e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 9.93998e-06
        [inline_without_move]: 5.86e-06
        [renormalize]: 0.00078363
        [add_forward_monad_depend]: 6.43e-06
        [auto_monad_grad]: 2.69999e-06
        [auto_monad_eliminator]: 1.575e-05
        [cse]: 2.313e-05
        [replace_applicator]: 1.272e-05
    [Cycle 2]: 0.00034803, [27]
        [switch_simplify]: 6.69999e-06
        [loop_unroll]: 5.37999e-06
        [a_1]: 9.647e-05
        [with_stream_mark]: 1.204e-05
        [recompute_prepare]: 5.74999e-06
        [updatestate_depend_eliminate]: 2.94001e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 5.66e-06
        [updatestate_useless_node_eliminater]: 5.85002e-06
        [accelerated_algorithm]: 5.99e-06
        [meta_shard_fg_expand]: 1.60999e-06
        [get_grad_eliminate_]: 5.40001e-06
        [merge_forward]: 3.46999e-06
        [cell_reuse_recompute_pass]: 1.90001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.625e-05
        [j_node_and_user_rematch]: 9.29e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 8.08999e-06
        [inline_without_move]: 4.58001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.11002e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 5.39e-06
        [cse]: 1.082e-05
        [replace_applicator]: 5.81e-06
[py_interpret_to_execute_after_opt_a]: 1.262e-05
[rewriter_after_opt_a]: 3.461e-05
[convert_after_rewriter]: 7.73001e-06
[order_py_execute_after_rewriter]: 5.17999e-06
[mutable_eliminate]: 0.00073993
[jit_opt_b]: 5.277e-05, [1]
    [Cycle 1]: 4.528e-05, [2]
        [frontend_op_eliminate]: 1.68e-05
        [inline_after_opt_a]: 1.605e-05
[cconv]: 2.666e-05
[loop_unroll]: 0.00050754
[jit_opt_after_cconv]: 0.00015676, [1]
    [Cycle 1]: 0.0001496, [11]
        [c_1]: 2.343e-05
        [parameter_eliminate]: 3.43e-06
        [updatestate_depend_eliminate]: 5.96e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 2.271e-05
        [call_graph_tuple_transform]: 2.091e-05
        [tuple_list_get_item_eliminator]: 6.36e-06
        [none_parameter_eliminate]: 1.60001e-06
        [renormalize]: 3.59985e-07
        [switch_simplify]: 6.51999e-06
[remove_dup_value]: 1.52e-05
[partial_unused_args_eliminate]: 2.32999e-06
[environ_conv]: 5.47001e-06
[add_recomputation]: 5.021e-05
[cse_after_recomputation]: 2.365e-05, [1]
    [Cycle 1]: 1.671e-05, [1]
        [cse]: 1.057e-05
[auto_monad_reorder]: 1.469e-05
[get_jit_bprop_graph]: 2.13998e-06
[rewriter_after_jit_bprop_graph]: 7.88999e-06
[opt_after_jit_grad]: 0.00052031
[symbol_engine_optimizer]: 8.568e-05, [1]
    [Cycle 1]: 7.833e-05, [6]
        [build]: 4.47e-06
        [elim_shapecalc]: 9.29e-06
        [elim_not_effective]: 1.489e-05
        [opt_reshape]: 6.84001e-06
        [fold_const_symbol]: 9.81e-06
        [renormalize]: 7.50006e-07
[validate]: 3.937e-05
Sums
    bootstrap                            :     0.000281s :  2.44%
    type_inference                       :     0.006661s : 57.93%
    event_method                         :     0.000022s :  0.19%
    auto_monad                           :     0.000082s :  0.71%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000053s :  0.46%
    rewriter_before_opt_a                :     0.000087s :  0.75%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000135s :  1.18%
    jit_opt_a.loop_unroll                :     0.000045s :  0.39%
    jit_opt_a.a_1                        :     0.000868s :  7.55%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000784s :  6.82%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.18%
    jit_opt_a.cse                        :     0.000034s :  0.30%
    jit_opt_a.replace_applicator         :     0.000019s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.11%
    rewriter_after_opt_a                 :     0.000035s :  0.30%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000740s :  6.44%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.14%
    cconv                                :     0.000027s :  0.23%
    loop_unroll                          :     0.000508s :  4.41%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.06%
    remove_dup_value                     :     0.000015s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000050s :  0.44%
    cse_after_recomputation.cse          :     0.000011s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.07%
    opt_after_jit_grad                   :     0.000520s :  4.53%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.34%


Time group info:
------[substitution.]   0.000306    26
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 1.77% :     0.000005s :      3: substitution.graph_param_transform
84.68% :     0.000259s :      6: substitution.inline
 1.26% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.07% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.39% :     0.000004s :      2: substitution.replace_old_param
 3.26% :     0.000010s :      1: substitution.switch_simplify
 4.40% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006595     2
82.76% :     0.005458s :      1: type_inference.infer
17.24% :     0.001137s :      1: type_inference.specialize
------[replace.]   0.000149     9
50.12% :     0.000075s :      6: replace.inline
25.20% :     0.000038s :      1: replace.switch_simplify
24.68% :     0.000037s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000276     9
92.30% :     0.000255s :      6: match.inline
 3.27% :     0.000009s :      1: match.switch_simplify
 4.43% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000150   801
 1.41% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000002s :     12: predicate.addn_check_dump
 1.56% :     0.000002s :     12: predicate.addn_zero_filter
 1.90% :     0.000003s :     12: predicate.arithmetic_simplify
 1.34% :     0.000002s :     12: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000002s :     12: predicate.compare_switch_simplify
 1.24% :     0.000002s :     12: predicate.depend_value_elim
 1.14% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.78% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.08% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     12: predicate.environ_get_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.17% :     0.000008s :     26: predicate.inline
 0.78% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.67% :     0.000003s :     14: predicate.list_to_tuple_eliminator_
 1.67% :     0.000003s :     17: predicate.load_eliminater
 1.72% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.54% :     0.000007s :     40: predicate.loop_unroll_before_grad
 2.37% :     0.000004s :     15: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000002s :     12: predicate.merge_addn
 1.19% :     0.000002s :     12: predicate.minmaximum_grad
 1.17% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 1.98% :     0.000003s :     17: predicate.partial_eliminate
 1.35% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.78% :     0.000003s :     12: predicate.reduce_eliminate
 1.40% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.39% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000003s :     20: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.75% :     0.000003s :     12: predicate.reshape_eliminate
 1.39% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000002s :     12: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.32% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.01% :     0.000005s :     20: predicate.switch_defer_inline
 2.46% :     0.000004s :     20: predicate.switch_layer_defer_inline
 8.82% :     0.000013s :     65: predicate.switch_simplify
 1.25% :     0.000002s :     12: predicate.tile_eliminate
 1.22% :     0.000002s :     12: predicate.transpose_eliminate
 1.46% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.38% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.67% :     0.000003s :     17: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000003s :     12: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000814    15
46.25% :     0.000377s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.95% :     0.000105s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.81% :     0.000332s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017688    72
 0.30% :     0.000053s :      1: add_recomputation
 0.49% :     0.000086s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 1.68% :     0.000296s :      1: bootstrap
 0.17% :     0.000029s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.81% :     0.005981s :      1: jit_opt_a
 0.91% :     0.000161s :      1: jit_opt_after_cconv
 0.31% :     0.000055s :      1: jit_opt_b
 2.92% :     0.000516s :      1: loop_unroll
 4.24% :     0.000750s :      1: mutable_eliminate
 6.64% :     0.001175s :     26: opt.transform.jit_opt_a
 0.30% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.99% :     0.000530s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000056s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000018s :      1: remove_dup_value
 2.46% :     0.000435s :      1: renormalize.infer
 1.92% :     0.000340s :      1: renormalize.specialize
 0.06% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000038s :      1: rewriter_after_opt_a
 0.51% :     0.000090s :      1: rewriter_before_opt_a
 0.50% :     0.000089s :      1: symbol_engine_optimizer
37.73% :     0.006674s :      1: type_inference


TotalTime = 0.0300717, [33]
[bootstrap]: 0.00032611
[type_inference]: 0.00488851
[event_method]: 1.71e-05
[auto_monad]: 6.231e-05
[graph_reusing]: 5.65001e-06
[pre_auto_parallel]: 2.59999e-06
[py_interpret_to_execute]: 3.515e-05
[rewriter_before_opt_a]: 7.41e-05
[expand_dump_flag]: 3.15002e-06
[jit_opt_a]: 0.00550439, [2]
    [Cycle 1]: 0.0016488, [27]
        [switch_simplify]: 5.427e-05
        [loop_unroll]: 2.955e-05
        [a_1]: 0.0005557
        [with_stream_mark]: 1.971e-05
        [recompute_prepare]: 8.32e-06
        [updatestate_depend_eliminate]: 3.55e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 2.36998e-06
        [specialize_transform]: 2.246e-05
        [updatestate_useless_node_eliminater]: 5.39e-06
        [accelerated_algorithm]: 5.18002e-06
        [meta_shard_fg_expand]: 2.11e-06
        [get_grad_eliminate_]: 5.18002e-06
        [merge_forward]: 3.56001e-06
        [cell_reuse_recompute_pass]: 1.32999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.495e-05
        [j_node_and_user_rematch]: 8.60999e-06
        [meta_fg_expand]: 2.19001e-06
        [replace_old_param]: 8.27998e-06
        [inline_without_move]: 5.15999e-06
        [renormalize]: 0.0005684
        [add_forward_monad_depend]: 6.16e-06
        [auto_monad_grad]: 2.29001e-06
        [auto_monad_eliminator]: 6.757e-05
        [cse]: 3.22e-05
        [replace_applicator]: 1.394e-05
    [Cycle 2]: 0.00032996, [27]
        [switch_simplify]: 6.02001e-06
        [loop_unroll]: 4.80999e-06
        [a_1]: 9.361e-05
        [with_stream_mark]: 1.016e-05
        [recompute_prepare]: 5.20999e-06
        [updatestate_depend_eliminate]: 3.87998e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 4.87e-06
        [meta_shard_fg_expand]: 1.35999e-06
        [get_grad_eliminate_]: 4.45e-06
        [merge_forward]: 2.60002e-06
        [cell_reuse_recompute_pass]: 2.01e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.347e-05
        [j_node_and_user_rematch]: 8.09997e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 7.15e-06
        [inline_without_move]: 4.57998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.82999e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 5.39e-06
        [cse]: 1.094e-05
        [replace_applicator]: 5.14e-06
[py_interpret_to_execute_after_opt_a]: 1.219e-05
[rewriter_after_opt_a]: 3.51e-05
[convert_after_rewriter]: 6.66999e-06
[order_py_execute_after_rewriter]: 5.02999e-06
[mutable_eliminate]: 0.00056435
[jit_opt_b]: 4.956e-05, [1]
    [Cycle 1]: 4.254e-05, [2]
        [frontend_op_eliminate]: 1.571e-05
        [inline_after_opt_a]: 1.483e-05
[cconv]: 2.571e-05
[loop_unroll]: 0.00042603
[jit_opt_after_cconv]: 0.00014048, [1]
    [Cycle 1]: 0.00013349, [11]
        [c_1]: 2.065e-05
        [parameter_eliminate]: 2.66999e-06
        [updatestate_depend_eliminate]: 5.06997e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 2.126e-05
        [call_graph_tuple_transform]: 1.786e-05
        [tuple_list_get_item_eliminator]: 5.40001e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 6.60017e-07
        [switch_simplify]: 5.24998e-06
[remove_dup_value]: 1.366e-05
[partial_unused_args_eliminate]: 2.32001e-06
[environ_conv]: 5.36998e-06
[add_recomputation]: 4.927e-05
[cse_after_recomputation]: 2.053e-05, [1]
    [Cycle 1]: 1.494e-05, [1]
        [cse]: 9.10999e-06
[auto_monad_reorder]: 1.5e-05
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 5.80002e-06
[opt_after_jit_grad]: 0.0004638
[symbol_engine_optimizer]: 7.311e-05, [1]
    [Cycle 1]: 6.569e-05, [6]
        [build]: 3.31001e-06
        [elim_shapecalc]: 7.33e-06
        [elim_not_effective]: 1.264e-05
        [opt_reshape]: 5.37999e-06
        [fold_const_symbol]: 8.60999e-06
        [renormalize]: 1.10999e-06
[validate]: 3.631e-05
[backend_pass]: 1.16002e-06
[task_emit]: 0.0169647
[execute]: 8.97e-06
Sums
    bootstrap                            :     0.000326s :  1.26%
    type_inference                       :     0.004889s : 18.88%
    event_method                         :     0.000017s :  0.07%
    auto_monad                           :     0.000062s :  0.24%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000035s :  0.14%
    rewriter_before_opt_a                :     0.000074s :  0.29%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000060s :  0.23%
    jit_opt_a.loop_unroll                :     0.000034s :  0.13%
    jit_opt_a.a_1                        :     0.000649s :  2.51%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000027s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000015s :  0.06%
    jit_opt_a.inline_without_move        :     0.000010s :  0.04%
    jit_opt_a.renormalize                :     0.000568s :  2.20%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000073s :  0.28%
    jit_opt_a.cse                        :     0.000043s :  0.17%
    jit_opt_a.replace_applicator         :     0.000019s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.05%
    rewriter_after_opt_a                 :     0.000035s :  0.14%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000564s :  2.18%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.06%
    cconv                                :     0.000026s :  0.10%
    loop_unroll                          :     0.000426s :  1.65%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000014s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000049s :  0.19%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000015s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000464s :  1.79%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.016965s : 65.53%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000194    23
 1.17% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.82% :     0.000005s :      3: substitution.graph_param_transform
83.39% :     0.000161s :      4: substitution.inline
 1.51% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.30% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.97% :     0.000004s :      2: substitution.replace_old_param
 6.17% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004823     2
83.00% :     0.004003s :      1: type_inference.infer
17.00% :     0.000820s :      1: type_inference.specialize
------[replace.]   0.000066     6
66.90% :     0.000044s :      4: replace.inline
33.10% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000170     6
93.60% :     0.000159s :      4: match.inline
 6.40% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.20% :     0.000001s :     10: predicate.addn_check_dump
 1.46% :     0.000002s :     10: predicate.addn_zero_filter
 2.33% :     0.000003s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.71% :     0.000002s :     10: predicate.depend_value_elim
 1.27% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.34% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.91% :     0.000001s :      6: predicate.less_batch_normalization
 2.12% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.97% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.73% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.27% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.18% :     0.000001s :     10: predicate.minmaximum_grad
 1.63% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.47% :     0.000003s :     15: predicate.partial_eliminate
 1.23% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.61% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.44% :     0.000001s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.48% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.84% :     0.000001s :      6: predicate.special_op_eliminate
 1.12% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 2.43% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.54% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.31% :     0.000002s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.51% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000550    12
50.76% :     0.000279s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.39% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.84% :     0.000241s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.031608    76
 0.17% :     0.000052s :      1: add_recomputation
 0.21% :     0.000066s :      1: auto_monad
 0.06% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.10% :     0.000349s :      1: bootstrap
 0.09% :     0.000028s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.07% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.07% :     0.000021s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
17.42% :     0.005508s :      1: jit_opt_a
 0.45% :     0.000143s :      1: jit_opt_after_cconv
 0.17% :     0.000052s :      1: jit_opt_b
 1.37% :     0.000434s :      1: loop_unroll
 1.81% :     0.000573s :      1: mutable_eliminate
 2.77% :     0.000874s :     26: opt.transform.jit_opt_a
 0.15% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000024s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.49% :     0.000472s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000038s :      1: py_interpret_to_execute
 0.05% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000016s :      1: remove_dup_value
 0.91% :     0.000289s :      1: renormalize.infer
 0.86% :     0.000271s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000038s :      1: rewriter_after_opt_a
 0.24% :     0.000077s :      1: rewriter_before_opt_a
 0.24% :     0.000076s :      1: symbol_engine_optimizer
53.76% :     0.016991s :      1: task_emit
15.50% :     0.004900s :      1: type_inference
 0.19% :     0.000059s :      1: validate


TotalTime = 0.0168287, [30]
[bootstrap]: 0.00032383
[type_inference]: 0.00706334
[event_method]: 2.498e-05
[auto_monad]: 8.993e-05
[graph_reusing]: 7.09001e-06
[pre_auto_parallel]: 2.82002e-06
[py_interpret_to_execute]: 5.993e-05
[rewriter_before_opt_a]: 9.214e-05
[expand_dump_flag]: 3.33e-06
[jit_opt_a]: 0.00637284, [2]
    [Cycle 1]: 0.00211091, [27]
        [switch_simplify]: 0.00013759
        [loop_unroll]: 3.506e-05
        [a_1]: 0.00076268
        [with_stream_mark]: 2.161e-05
        [recompute_prepare]: 9.19998e-06
        [updatestate_depend_eliminate]: 4.28999e-06
        [updatestate_assign_eliminate]: 3.2e-06
        [updatestate_loads_eliminate]: 2.51998e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 6.17001e-06
        [updatestate_useless_node_eliminater]: 5.37001e-06
        [accelerated_algorithm]: 6.91999e-06
        [meta_shard_fg_expand]: 2.24999e-06
        [get_grad_eliminate_]: 5.33002e-06
        [merge_forward]: 3.51001e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.611e-05
        [j_node_and_user_rematch]: 9.25001e-06
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 9.77999e-06
        [inline_without_move]: 5.29998e-06
        [renormalize]: 0.00076796
        [add_forward_monad_depend]: 7.22002e-06
        [auto_monad_grad]: 2.78e-06
        [auto_monad_eliminator]: 1.643e-05
        [cse]: 3.281e-05
        [replace_applicator]: 1.676e-05
    [Cycle 2]: 0.00034371, [27]
        [switch_simplify]: 6.40002e-06
        [loop_unroll]: 5.14e-06
        [a_1]: 9.511e-05
        [with_stream_mark]: 1.321e-05
        [recompute_prepare]: 5.39e-06
        [updatestate_depend_eliminate]: 3.58999e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 1.49e-06
        [specialize_transform]: 5.17e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 5.94e-06
        [meta_shard_fg_expand]: 1.36998e-06
        [get_grad_eliminate_]: 4.67998e-06
        [merge_forward]: 3.29001e-06
        [cell_reuse_recompute_pass]: 2.16e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.404e-05
        [j_node_and_user_rematch]: 8.46002e-06
        [meta_fg_expand]: 1.71002e-06
        [replace_old_param]: 7.8e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 8.79983e-07
        [auto_monad_eliminator]: 5.20999e-06
        [cse]: 1.201e-05
        [replace_applicator]: 5.71998e-06
[py_interpret_to_execute_after_opt_a]: 1.357e-05
[rewriter_after_opt_a]: 4.014e-05
[convert_after_rewriter]: 7.35998e-06
[order_py_execute_after_rewriter]: 5.30999e-06
[mutable_eliminate]: 0.00088053
[jit_opt_b]: 7.509e-05, [1]
    [Cycle 1]: 6.493e-05, [2]
        [frontend_op_eliminate]: 2.054e-05
        [inline_after_opt_a]: 1.916e-05
[cconv]: 3.266e-05
[loop_unroll]: 0.00058126
[jit_opt_after_cconv]: 0.0001645, [1]
    [Cycle 1]: 0.00015652, [11]
        [c_1]: 2.163e-05
        [parameter_eliminate]: 3.48e-06
        [updatestate_depend_eliminate]: 6.48e-06
        [updatestate_assign_eliminate]: 3.14999e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [cse]: 2.86e-05
        [call_graph_tuple_transform]: 2.254e-05
        [tuple_list_get_item_eliminator]: 5.54e-06
        [none_parameter_eliminate]: 2.22001e-06
        [renormalize]: 9.09989e-07
        [switch_simplify]: 6.69999e-06
[remove_dup_value]: 1.592e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 5.10001e-06
[add_recomputation]: 5.312e-05
[cse_after_recomputation]: 2.413e-05, [1]
    [Cycle 1]: 1.763e-05, [1]
        [cse]: 1.121e-05
[auto_monad_reorder]: 1.494e-05
[get_jit_bprop_graph]: 1.96e-06
[rewriter_after_jit_bprop_graph]: 7.34002e-06
[opt_after_jit_grad]: 0.00052856
[symbol_engine_optimizer]: 7.97e-05, [1]
    [Cycle 1]: 7.206e-05, [6]
        [build]: 5.44e-06
        [elim_shapecalc]: 7.50998e-06
        [elim_not_effective]: 1.348e-05
        [opt_reshape]: 5.75001e-06
        [fold_const_symbol]: 9.29998e-06
        [renormalize]: 8.09989e-07
[validate]: 4.131e-05
Sums
    bootstrap                            :     0.000324s :  2.65%
    type_inference                       :     0.007063s : 57.82%
    event_method                         :     0.000025s :  0.20%
    auto_monad                           :     0.000090s :  0.74%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000060s :  0.49%
    rewriter_before_opt_a                :     0.000092s :  0.75%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000144s :  1.18%
    jit_opt_a.loop_unroll                :     0.000040s :  0.33%
    jit_opt_a.a_1                        :     0.000858s :  7.02%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000768s :  6.29%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.18%
    jit_opt_a.cse                        :     0.000045s :  0.37%
    jit_opt_a.replace_applicator         :     0.000022s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.11%
    rewriter_after_opt_a                 :     0.000040s :  0.33%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000881s :  7.21%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.16%
    cconv                                :     0.000033s :  0.27%
    loop_unroll                          :     0.000581s :  4.76%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000029s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.05%
    remove_dup_value                     :     0.000016s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000053s :  0.43%
    cse_after_recomputation.cse          :     0.000011s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000529s :  4.33%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.34%


Time group info:
------[substitution.]   0.000328    26
 0.61% :     0.000002s :      2: substitution.elim_not_effective
 0.41% :     0.000001s :      2: substitution.fold_const_symbol
 1.99% :     0.000007s :      3: substitution.graph_param_transform
85.27% :     0.000280s :      6: substitution.inline
 1.13% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.72% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.58% :     0.000005s :      2: substitution.replace_old_param
 3.10% :     0.000010s :      1: substitution.switch_simplify
 4.19% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006978     2
83.08% :     0.005798s :      1: type_inference.infer
16.92% :     0.001181s :      1: type_inference.specialize
------[replace.]   0.000153     9
49.08% :     0.000075s :      6: replace.inline
27.60% :     0.000042s :      1: replace.switch_simplify
23.31% :     0.000036s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000298     9
92.77% :     0.000276s :      6: match.inline
 3.02% :     0.000009s :      1: match.switch_simplify
 4.21% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000147   801
 1.20% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.65% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     12: predicate.addn_check_dump
 1.25% :     0.000002s :     12: predicate.addn_zero_filter
 2.29% :     0.000003s :     12: predicate.arithmetic_simplify
 1.19% :     0.000002s :     12: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     12: predicate.compare_switch_simplify
 1.71% :     0.000003s :     12: predicate.depend_value_elim
 1.08% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.18% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     12: predicate.dict_set_item_eliminator
 1.17% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.24% :     0.000000s :      3: predicate.elim_not_effective
 0.46% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.03% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.06% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.10% :     0.000002s :     12: predicate.environ_get_eliminate
 1.12% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.79% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.18% :     0.000008s :     26: predicate.inline
 0.77% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.76% :     0.000003s :      6: predicate.less_batch_normalization
 2.18% :     0.000003s :     14: predicate.list_to_tuple_eliminator_
 1.59% :     0.000002s :     17: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.95% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.16% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000002s :     12: predicate.merge_addn
 1.10% :     0.000002s :     12: predicate.minmaximum_grad
 2.07% :     0.000003s :      3: predicate.mutable_eliminate
 0.39% :     0.000001s :      3: predicate.opt_reshape
 2.03% :     0.000003s :     17: predicate.partial_eliminate
 1.15% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.74% :     0.000003s :     12: predicate.reduce_eliminate
 1.77% :     0.000003s :     14: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.47% :     0.000002s :     20: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.20% :     0.000000s :      3: predicate.reset_defer_inline
 1.24% :     0.000002s :     12: predicate.reshape_eliminate
 1.54% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 0.84% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.90% :     0.000004s :     20: predicate.switch_defer_inline
 2.14% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.48% :     0.000012s :     65: predicate.switch_simplify
 1.17% :     0.000002s :     12: predicate.tile_eliminate
 1.20% :     0.000002s :     12: predicate.transpose_eliminate
 1.45% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.44% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.65% :     0.000007s :     20: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000003s :     12: predicate.tuple_list_set_item_eliminator
 1.32% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.42% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.42% :     0.000002s :     12: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000866    15
46.98% :     0.000407s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.18% :     0.000114s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.84% :     0.000345s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018835    72
 0.30% :     0.000057s :      1: add_recomputation
 0.50% :     0.000094s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 1.82% :     0.000343s :      1: bootstrap
 0.19% :     0.000037s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.16% :     0.000029s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000010s :      1: graph_reusing
33.86% :     0.006378s :      1: jit_opt_a
 0.89% :     0.000168s :      1: jit_opt_after_cconv
 0.42% :     0.000079s :      1: jit_opt_b
 3.14% :     0.000592s :      1: loop_unroll
 4.76% :     0.000896s :      1: mutable_eliminate
 6.21% :     0.001169s :     26: opt.transform.jit_opt_a
 0.28% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000031s :      4: opt.transform.jit_opt_b
 0.09% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.87% :     0.000541s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.34% :     0.000064s :      1: py_interpret_to_execute
 0.09% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000019s :      1: remove_dup_value
 2.16% :     0.000406s :      1: renormalize.infer
 1.87% :     0.000353s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000044s :      1: rewriter_after_opt_a
 0.51% :     0.000096s :      1: rewriter_before_opt_a
 0.44% :     0.000083s :      1: symbol_engine_optimizer
37.61% :     0.007084s :      1: type_inference


TotalTime = 0.0179102, [33]
[bootstrap]: 0.00028209
[type_inference]: 0.00408736
[event_method]: 1.62e-05
[auto_monad]: 6.307e-05
[graph_reusing]: 6.00002e-06
[pre_auto_parallel]: 2.60997e-06
[py_interpret_to_execute]: 3.17e-05
[rewriter_before_opt_a]: 7.307e-05
[expand_dump_flag]: 2.98e-06
[jit_opt_a]: 0.00514938, [2]
    [Cycle 1]: 0.00159525, [27]
        [switch_simplify]: 5.017e-05
        [loop_unroll]: 3.056e-05
        [a_1]: 0.00054948
        [with_stream_mark]: 1.544e-05
        [recompute_prepare]: 6.93e-06
        [updatestate_depend_eliminate]: 3.8e-06
        [updatestate_assign_eliminate]: 3.03998e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [parameter_eliminate]: 2.06998e-06
        [specialize_transform]: 5.66e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 5.22999e-06
        [meta_shard_fg_expand]: 2.37999e-06
        [get_grad_eliminate_]: 5.02e-06
        [merge_forward]: 3.71001e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.408e-05
        [j_node_and_user_rematch]: 8.75001e-06
        [meta_fg_expand]: 2.43002e-06
        [replace_old_param]: 8.79e-06
        [inline_without_move]: 5.39e-06
        [renormalize]: 0.00060064
        [add_forward_monad_depend]: 5.48002e-06
        [auto_monad_grad]: 1.97001e-06
        [auto_monad_eliminator]: 1.314e-05
        [cse]: 2.717e-05
        [replace_applicator]: 1.214e-05
    [Cycle 2]: 0.00031833, [27]
        [switch_simplify]: 5.91e-06
        [loop_unroll]: 4.86002e-06
        [a_1]: 8.967e-05
        [with_stream_mark]: 8.48001e-06
        [recompute_prepare]: 4.84e-06
        [updatestate_depend_eliminate]: 2.52001e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 4.95001e-06
        [updatestate_useless_node_eliminater]: 4.94998e-06
        [accelerated_algorithm]: 5.02e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.53001e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 1.81e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.231e-05
        [j_node_and_user_rematch]: 7.95998e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 7.61001e-06
        [inline_without_move]: 4.60001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 5.39998e-06
        [cse]: 1.047e-05
        [replace_applicator]: 5.10999e-06
[py_interpret_to_execute_after_opt_a]: 8.52e-06
[rewriter_after_opt_a]: 3.077e-05
[convert_after_rewriter]: 6.28e-06
[order_py_execute_after_rewriter]: 5.10999e-06
[mutable_eliminate]: 0.00048598
[jit_opt_b]: 4.817e-05, [1]
    [Cycle 1]: 4.155e-05, [2]
        [frontend_op_eliminate]: 1.568e-05
        [inline_after_opt_a]: 1.464e-05
[cconv]: 2.422e-05
[loop_unroll]: 0.00040332
[jit_opt_after_cconv]: 0.00013437, [1]
    [Cycle 1]: 0.00012815, [11]
        [c_1]: 1.974e-05
        [parameter_eliminate]: 2.42001e-06
        [updatestate_depend_eliminate]: 5.17e-06
        [updatestate_assign_eliminate]: 2.33002e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 1.668e-05
        [call_graph_tuple_transform]: 1.801e-05
        [tuple_list_get_item_eliminator]: 5.39e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 6.60017e-07
        [switch_simplify]: 5.34998e-06
[remove_dup_value]: 1.404e-05
[partial_unused_args_eliminate]: 2.40002e-06
[environ_conv]: 5.09e-06
[add_recomputation]: 4.427e-05
[cse_after_recomputation]: 1.954e-05, [1]
    [Cycle 1]: 1.436e-05, [1]
        [cse]: 8.97e-06
[auto_monad_reorder]: 1.401e-05
[get_jit_bprop_graph]: 1.50001e-06
[rewriter_after_jit_bprop_graph]: 4.83001e-06
[opt_after_jit_grad]: 0.00043962
[symbol_engine_optimizer]: 7.041e-05, [1]
    [Cycle 1]: 6.406e-05, [6]
        [build]: 2.79001e-06
        [elim_shapecalc]: 7.51001e-06
        [elim_not_effective]: 1.168e-05
        [opt_reshape]: 5.54e-06
        [fold_const_symbol]: 8.89e-06
        [renormalize]: 5.70028e-07
[validate]: 3.115e-05
[backend_pass]: 8.10018e-07
[task_emit]: 0.00621303
[execute]: 7.02002e-06
Sums
    bootstrap                            :     0.000282s :  2.01%
    type_inference                       :     0.004087s : 29.07%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000063s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.23%
    rewriter_before_opt_a                :     0.000073s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.25%
    jit_opt_a.a_1                        :     0.000639s :  4.55%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000601s :  4.27%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000038s :  0.27%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000486s :  3.46%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000403s :  2.87%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000044s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000440s :  3.13%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000031s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006213s : 44.19%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000200    23
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.75% :     0.000005s :      3: substitution.graph_param_transform
84.09% :     0.000168s :      4: substitution.inline
 1.39% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.18% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.29% :     0.000005s :      2: substitution.replace_old_param
 5.81% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004029     2
81.26% :     0.003274s :      1: type_inference.infer
18.74% :     0.000755s :      1: type_inference.specialize
------[replace.]   0.000065     6
67.46% :     0.000044s :      4: replace.inline
32.54% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000176     6
94.01% :     0.000166s :      4: match.inline
 5.99% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000001s :     10: predicate.addn_zero_filter
 2.21% :     0.000003s :     10: predicate.arithmetic_simplify
 1.31% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.73% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.29% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.32% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.36% :     0.000002s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000000s :      3: predicate.opt_reshape
 2.42% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.71% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.61% :     0.000003s :     16: predicate.switch_defer_inline
 2.24% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.75% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.92% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.93% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000652    12
37.24% :     0.000243s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.34% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
54.42% :     0.000355s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019455    76
 0.24% :     0.000047s :      1: add_recomputation
 0.34% :     0.000066s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.53% :     0.000297s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.48% :     0.005152s :      1: jit_opt_a
 0.71% :     0.000138s :      1: jit_opt_after_cconv
 0.26% :     0.000051s :      1: jit_opt_b
 2.11% :     0.000411s :      1: loop_unroll
 2.54% :     0.000495s :      1: mutable_eliminate
 4.33% :     0.000842s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.30% :     0.000448s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000035s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.16% :     0.000225s :      1: renormalize.infer
 1.89% :     0.000368s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000034s :      1: rewriter_after_opt_a
 0.39% :     0.000076s :      1: rewriter_before_opt_a
 0.37% :     0.000073s :      1: symbol_engine_optimizer
31.99% :     0.006224s :      1: task_emit
21.07% :     0.004100s :      1: type_inference
 0.25% :     0.000049s :      1: validate


TotalTime = 0.0834524, [33]
[bootstrap]: 0.00033841
[type_inference]: 0.0734575
[event_method]: 1.981e-05
[auto_monad]: 6.921e-05
[graph_reusing]: 5.57999e-06
[pre_auto_parallel]: 2.39999e-06
[py_interpret_to_execute]: 0.00027802
[rewriter_before_opt_a]: 8.642e-05
[expand_dump_flag]: 3.88001e-06
[jit_opt_a]: 0.00690811, [2]
    [Cycle 1]: 0.00298241, [27]
        [switch_simplify]: 6.275e-05
        [loop_unroll]: 3.49e-05
        [a_1]: 0.00065522
        [with_stream_mark]: 2.037e-05
        [recompute_prepare]: 8.59e-06
        [updatestate_depend_eliminate]: 4.08001e-06
        [updatestate_assign_eliminate]: 2.79999e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [parameter_eliminate]: 2.16e-06
        [specialize_transform]: 6.71e-06
        [updatestate_useless_node_eliminater]: 4.74998e-06
        [accelerated_algorithm]: 4.99e-06
        [meta_shard_fg_expand]: 2.02001e-06
        [get_grad_eliminate_]: 4.62e-06
        [merge_forward]: 3.20998e-06
        [cell_reuse_recompute_pass]: 1.83002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.148e-05
        [j_node_and_user_rematch]: 7.34002e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 9.46e-06
        [inline_without_move]: 4.66002e-06
        [renormalize]: 0.00185029
        [add_forward_monad_depend]: 7.08e-06
        [auto_monad_grad]: 2.54001e-06
        [auto_monad_eliminator]: 1.483e-05
        [cse]: 2.952e-05
        [replace_applicator]: 1.612e-05
    [Cycle 2]: 0.00028289, [27]
        [switch_simplify]: 4.90001e-06
        [loop_unroll]: 4.27e-06
        [a_1]: 6.648e-05
        [with_stream_mark]: 1.165e-05
        [recompute_prepare]: 4.37e-06
        [updatestate_depend_eliminate]: 2.69999e-06
        [updatestate_assign_eliminate]: 1.65001e-06
        [updatestate_loads_eliminate]: 1.64e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 4.33999e-06
        [updatestate_useless_node_eliminater]: 3.9e-06
        [accelerated_algorithm]: 4.08999e-06
        [meta_shard_fg_expand]: 1.62001e-06
        [get_grad_eliminate_]: 3.70998e-06
        [merge_forward]: 2.66999e-06
        [cell_reuse_recompute_pass]: 4.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.305e-05
        [j_node_and_user_rematch]: 6.39001e-06
        [meta_fg_expand]: 1.31002e-06
        [replace_old_param]: 7.38999e-06
        [inline_without_move]: 3.77002e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.79998e-06
        [auto_monad_grad]: 8.90024e-07
        [auto_monad_eliminator]: 3.71001e-06
        [cse]: 7.92003e-06
        [replace_applicator]: 4.56002e-06
[py_interpret_to_execute_after_opt_a]: 1.204e-05
[rewriter_after_opt_a]: 3.071e-05
[convert_after_rewriter]: 6.31e-06
[order_py_execute_after_rewriter]: 4.4e-06
[mutable_eliminate]: 0.00065936
[jit_opt_b]: 4.42e-05, [1]
    [Cycle 1]: 3.704e-05, [2]
        [frontend_op_eliminate]: 1.289e-05
        [inline_after_opt_a]: 1.238e-05
[cconv]: 2.394e-05
[loop_unroll]: 0.00043815
[jit_opt_after_cconv]: 0.00012463, [1]
    [Cycle 1]: 0.00011878, [11]
        [c_1]: 1.581e-05
        [parameter_eliminate]: 3.00002e-06
        [updatestate_depend_eliminate]: 4.70001e-06
        [updatestate_assign_eliminate]: 1.63002e-06
        [updatestate_loads_eliminate]: 1.49998e-06
        [cse]: 1.708e-05
        [call_graph_tuple_transform]: 1.564e-05
        [tuple_list_get_item_eliminator]: 4.27e-06
        [none_parameter_eliminate]: 1.91e-06
        [renormalize]: 5.99975e-07
        [switch_simplify]: 4.33001e-06
[remove_dup_value]: 1.303e-05
[partial_unused_args_eliminate]: 2.51e-06
[environ_conv]: 4.62e-06
[add_recomputation]: 3.808e-05
[cse_after_recomputation]: 1.76e-05, [1]
    [Cycle 1]: 1.231e-05, [1]
        [cse]: 6.63e-06
[auto_monad_reorder]: 1.326e-05
[get_jit_bprop_graph]: 1.87001e-06
[rewriter_after_jit_bprop_graph]: 4.88001e-06
[opt_after_jit_grad]: 0.00048055
[symbol_engine_optimizer]: 6.898e-05, [1]
    [Cycle 1]: 6.213e-05, [6]
        [build]: 3.95e-06
        [elim_shapecalc]: 7.71001e-06
        [elim_not_effective]: 1.149e-05
        [opt_reshape]: 5.02999e-06
        [fold_const_symbol]: 6.44999e-06
        [renormalize]: 3.89991e-07
[validate]: 7.698e-05
[backend_pass]: 1.00999e-06
[task_emit]: 2.269e-05
[execute]: 1.25999e-06
Sums
    bootstrap                            :     0.000338s :  0.43%
    type_inference                       :     0.073458s : 92.77%
    event_method                         :     0.000020s :  0.03%
    auto_monad                           :     0.000069s :  0.09%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.00%
    py_interpret_to_execute              :     0.000278s :  0.35%
    rewriter_before_opt_a                :     0.000086s :  0.11%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000068s :  0.09%
    jit_opt_a.loop_unroll                :     0.000039s :  0.05%
    jit_opt_a.a_1                        :     0.000722s :  0.91%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.04%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.00%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.00%
    jit_opt_a.specialize_transform       :     0.000011s :  0.01%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.01%
    jit_opt_a.accelerated_algorithm      :     0.000009s :  0.01%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.00%
    jit_opt_a.get_grad_eliminate_        :     0.000008s :  0.01%
    jit_opt_a.merge_forward              :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.03%
    jit_opt_a.j_node_and_user_rematch    :     0.000014s :  0.02%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.00%
    jit_opt_a.replace_old_param          :     0.000017s :  0.02%
    jit_opt_a.inline_without_move        :     0.000008s :  0.01%
    jit_opt_a.renormalize                :     0.001850s :  2.34%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.02%
    jit_opt_a.cse                        :     0.000037s :  0.05%
    jit_opt_a.replace_applicator         :     0.000021s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.02%
    rewriter_after_opt_a                 :     0.000031s :  0.04%
    convert_after_rewriter               :     0.000006s :  0.01%
    order_py_execute_after_rewriter      :     0.000004s :  0.01%
    mutable_eliminate                    :     0.000659s :  0.83%
    jit_opt_b.frontend_op_eliminate      :     0.000013s :  0.02%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.02%
    cconv                                :     0.000024s :  0.03%
    loop_unroll                          :     0.000438s :  0.55%
    jit_opt_after_cconv.c_1              :     0.000016s :  0.02%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.cse              :     0.000017s :  0.02%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.02%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.01%
    remove_dup_value                     :     0.000013s :  0.02%
    partial_unused_args_eliminate        :     0.000003s :  0.00%
    environ_conv                         :     0.000005s :  0.01%
    add_recomputation                    :     0.000038s :  0.05%
    cse_after_recomputation.cse          :     0.000007s :  0.01%
    auto_monad_reorder                   :     0.000013s :  0.02%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000481s :  0.61%
    symbol_engine_optimizer.build        :     0.000004s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.01%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000077s :  0.10%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000023s :  0.03%
    execute                              :     0.000001s :  0.00%


Time group info:
------[substitution.]   0.000238    17
 0.68% :     0.000002s :      1: substitution.elim_not_effective
 0.47% :     0.000001s :      1: substitution.fold_const_symbol
 1.83% :     0.000004s :      2: substitution.graph_param_transform
82.65% :     0.000196s :      4: substitution.inline
 1.17% :     0.000003s :      2: substitution.j_node_and_user_rematch
 1.48% :     0.000004s :      2: substitution.remove_not_recompute_node
 2.21% :     0.000005s :      2: substitution.replace_old_param
 5.08% :     0.000012s :      1: substitution.tile_eliminate
 4.43% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.073371     2
97.44% :     0.071494s :      1: type_inference.infer
 2.56% :     0.001878s :      1: type_inference.specialize
------[replace.]   0.000079     7
62.52% :     0.000049s :      4: replace.inline
 5.96% :     0.000005s :      1: replace.tile_eliminate
31.52% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000214     7
90.42% :     0.000194s :      4: match.inline
 5.16% :     0.000011s :      1: match.tile_eliminate
 4.42% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   677
 1.25% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.21% :     0.000001s :      2: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 2.84% :     0.000004s :     10: predicate.arithmetic_simplify
 1.68% :     0.000002s :     11: predicate.cast_eliminate
 0.48% :     0.000001s :      2: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 2.04% :     0.000003s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     11: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     11: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     11: predicate.dict_set_item_eliminator
 1.06% :     0.000001s :      2: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      2: predicate.elim_not_effective
 0.66% :     0.000001s :      2: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     11: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     11: predicate.environ_get_add_eliminate
 1.18% :     0.000001s :     11: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     11: predicate.environ_get_eliminate
 1.16% :     0.000001s :     11: predicate.environ_get_set_eliminate
 0.17% :     0.000000s :      2: predicate.fold_const_symbol
 0.83% :     0.000001s :      4: predicate.get_grad_eliminate
 0.21% :     0.000000s :      2: predicate.graph_param_transform
 5.69% :     0.000007s :     21: predicate.inline
 0.63% :     0.000001s :      4: predicate.inline_without_move
 0.28% :     0.000000s :      4: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :      4: predicate.less_batch_normalization
 1.51% :     0.000002s :     13: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 1.34% :     0.000002s :      2: predicate.loop_unroll_after_grad
 4.54% :     0.000006s :     36: predicate.loop_unroll_before_grad
 1.72% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 1.26% :     0.000002s :      2: predicate.mutable_eliminate
 0.65% :     0.000001s :      2: predicate.opt_reshape
 1.98% :     0.000002s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.42% :     0.000003s :     11: predicate.reduce_eliminate
 1.63% :     0.000002s :     13: predicate.redundant_stop_gradient_eliminater
 0.37% :     0.000000s :      4: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     17: predicate.replace_applicator
 1.00% :     0.000001s :      4: predicate.replace_old_param
 0.25% :     0.000000s :      2: predicate.reset_defer_inline
 1.90% :     0.000002s :     11: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.56% :     0.000001s :      2: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.39% :     0.000000s :      4: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      4: predicate.special_op_eliminate
 0.72% :     0.000001s :      4: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.25% :     0.000000s :      2: predicate.switch_call_monad_eliminater
 2.64% :     0.000003s :     17: predicate.switch_defer_inline
 2.06% :     0.000003s :     17: predicate.switch_layer_defer_inline
 7.62% :     0.000009s :     55: predicate.switch_simplify
 1.50% :     0.000002s :     11: predicate.tile_eliminate
 1.24% :     0.000002s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     11: predicate.tuple_list_convert_item_index_to_positive
 1.80% :     0.000002s :     11: predicate.tuple_list_get_item_depend_reorder
 3.98% :     0.000005s :     17: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     11: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     13: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.41% :     0.000003s :     19: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.21% :     0.000000s :      2: predicate.virtual_view_grad_eliminate
 0.48% :     0.000001s :      2: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.002533    26
74.29% :     0.001881s :     19: func_graph_cloner_run.FuncGraphClonerGraph
 1.78% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
23.93% :     0.000606s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.086316    76
 0.05% :     0.000041s :      1: add_recomputation
 0.08% :     0.000072s :      1: auto_monad
 0.02% :     0.000016s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.41% :     0.000353s :      1: bootstrap
 0.03% :     0.000026s :      1: cconv
 0.01% :     0.000009s :      1: convert_after_rewriter
 0.02% :     0.000020s :      1: cse_after_recomputation
 0.01% :     0.000007s :      1: environ_conv
 0.03% :     0.000024s :      1: event_method
 0.00% :     0.000003s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000008s :      1: graph_reusing
 8.01% :     0.006912s :      1: jit_opt_a
 0.15% :     0.000127s :      1: jit_opt_after_cconv
 0.05% :     0.000047s :      1: jit_opt_b
 0.52% :     0.000446s :      1: loop_unroll
 0.77% :     0.000668s :      1: mutable_eliminate
 1.08% :     0.000932s :     26: opt.transform.jit_opt_a
 0.04% :     0.000037s :      4: opt.transform.jit_opt_after_cconv
 0.02% :     0.000019s :      4: opt.transform.jit_opt_b
 0.01% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.01% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.02% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.03% :     0.000028s :      4: opt.transform.symbol_engine_opt
 0.57% :     0.000489s :      1: opt_after_jit_grad
 0.01% :     0.000006s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.00% :     0.000004s :      1: pre_auto_parallel
 0.33% :     0.000284s :      1: py_interpret_to_execute
 0.02% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000015s :      1: remove_dup_value
 1.47% :     0.001272s :      1: renormalize.infer
 0.66% :     0.000569s :      1: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.04% :     0.000034s :      1: rewriter_after_opt_a
 0.11% :     0.000091s :      1: rewriter_before_opt_a
 0.08% :     0.000072s :      1: symbol_engine_optimizer
 0.03% :     0.000027s :      1: task_emit
85.12% :     0.073476s :      1: type_inference
 0.11% :     0.000099s :      1: validate


TotalTime = 0.0147796, [30]
[bootstrap]: 0.00029695
[type_inference]: 0.00650335
[event_method]: 2.184e-05
[auto_monad]: 8.437e-05
[graph_reusing]: 6.62002e-06
[pre_auto_parallel]: 2.60002e-06
[py_interpret_to_execute]: 4.998e-05
[rewriter_before_opt_a]: 8.409e-05
[expand_dump_flag]: 2.96001e-06
[jit_opt_a]: 0.00572005, [2]
    [Cycle 1]: 0.00176194, [27]
        [switch_simplify]: 0.00011153
        [loop_unroll]: 3.487e-05
        [a_1]: 0.00066549
        [with_stream_mark]: 1.651e-05
        [recompute_prepare]: 7.51999e-06
        [updatestate_depend_eliminate]: 3.89002e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 2.24001e-06
        [specialize_transform]: 6.11e-06
        [updatestate_useless_node_eliminater]: 5.40001e-06
        [accelerated_algorithm]: 5.47999e-06
        [meta_shard_fg_expand]: 2.04e-06
        [get_grad_eliminate_]: 5.53002e-06
        [merge_forward]: 4.03999e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.387e-05
        [j_node_and_user_rematch]: 9.22001e-06
        [meta_fg_expand]: 2.44001e-06
        [replace_old_param]: 9.58997e-06
        [inline_without_move]: 5.32999e-06
        [renormalize]: 0.00061715
        [add_forward_monad_depend]: 4.95999e-06
        [auto_monad_grad]: 2.41e-06
        [auto_monad_eliminator]: 1.306e-05
        [cse]: 2.714e-05
        [replace_applicator]: 1.148e-05
    [Cycle 2]: 0.0003498, [27]
        [switch_simplify]: 5.54e-06
        [loop_unroll]: 4.77998e-06
        [a_1]: 8.851e-05
        [with_stream_mark]: 9.84001e-06
        [recompute_prepare]: 6.01e-06
        [updatestate_depend_eliminate]: 2.70997e-06
        [updatestate_assign_eliminate]: 2.03002e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [parameter_eliminate]: 9.10019e-07
        [specialize_transform]: 5.20999e-06
        [updatestate_useless_node_eliminater]: 5.08002e-06
        [accelerated_algorithm]: 4.84e-06
        [meta_shard_fg_expand]: 1.07998e-06
        [get_grad_eliminate_]: 4.74998e-06
        [merge_forward]: 2.61999e-06
        [cell_reuse_recompute_pass]: 1.50001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.192e-05
        [j_node_and_user_rematch]: 7.82e-06
        [meta_fg_expand]: 1.53002e-06
        [replace_old_param]: 7.03998e-06
        [inline_without_move]: 4.83001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 1.04003e-06
        [auto_monad_eliminator]: 5.08002e-06
        [cse]: 1.062e-05
        [replace_applicator]: 4.95001e-06
[py_interpret_to_execute_after_opt_a]: 9.48002e-06
[rewriter_after_opt_a]: 3.234e-05
[convert_after_rewriter]: 6.98998e-06
[order_py_execute_after_rewriter]: 4.94e-06
[mutable_eliminate]: 0.00047543
[jit_opt_b]: 4.871e-05, [1]
    [Cycle 1]: 4.205e-05, [2]
        [frontend_op_eliminate]: 1.653e-05
        [inline_after_opt_a]: 1.434e-05
[cconv]: 2.387e-05
[loop_unroll]: 0.00041658
[jit_opt_after_cconv]: 0.00013789, [1]
    [Cycle 1]: 0.0001317, [11]
        [c_1]: 2.117e-05
        [parameter_eliminate]: 2.44999e-06
        [updatestate_depend_eliminate]: 5.39998e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.663e-05
        [call_graph_tuple_transform]: 1.822e-05
        [tuple_list_get_item_eliminator]: 5.33002e-06
        [none_parameter_eliminate]: 1.41998e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.39e-06
[remove_dup_value]: 1.435e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 4.77998e-06
[add_recomputation]: 4.635e-05
[cse_after_recomputation]: 2.119e-05, [1]
    [Cycle 1]: 1.559e-05, [1]
        [cse]: 9.70002e-06
[auto_monad_reorder]: 1.467e-05
[get_jit_bprop_graph]: 1.50999e-06
[rewriter_after_jit_bprop_graph]: 5.04e-06
[opt_after_jit_grad]: 0.00045488
[symbol_engine_optimizer]: 7.498e-05, [1]
    [Cycle 1]: 6.871e-05, [6]
        [build]: 4.22998e-06
        [elim_shapecalc]: 7.73001e-06
        [elim_not_effective]: 1.296e-05
        [opt_reshape]: 5.97999e-06
        [fold_const_symbol]: 8.71997e-06
        [renormalize]: 4.50003e-07
[validate]: 3.393e-05
Sums
    bootstrap                            :     0.000297s :  2.81%
    type_inference                       :     0.006503s : 61.59%
    event_method                         :     0.000022s :  0.21%
    auto_monad                           :     0.000084s :  0.80%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000050s :  0.47%
    rewriter_before_opt_a                :     0.000084s :  0.80%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000117s :  1.11%
    jit_opt_a.loop_unroll                :     0.000040s :  0.38%
    jit_opt_a.a_1                        :     0.000754s :  7.14%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000617s :  5.85%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.17%
    jit_opt_a.cse                        :     0.000038s :  0.36%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.09%
    rewriter_after_opt_a                 :     0.000032s :  0.31%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000475s :  4.50%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.14%
    cconv                                :     0.000024s :  0.23%
    loop_unroll                          :     0.000417s :  3.95%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000046s :  0.44%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000455s :  4.31%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.32%


Time group info:
------[substitution.]   0.000257    26
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 2.06% :     0.000005s :      3: substitution.graph_param_transform
84.66% :     0.000218s :      6: substitution.inline
 1.28% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.69% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.66% :     0.000004s :      2: substitution.replace_old_param
 2.79% :     0.000007s :      1: substitution.switch_simplify
 4.66% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006434     2
83.63% :     0.005381s :      1: type_inference.infer
16.37% :     0.001053s :      1: type_inference.specialize
------[replace.]   0.000131     9
50.30% :     0.000066s :      6: replace.inline
26.35% :     0.000035s :      1: replace.switch_simplify
23.35% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000232     9
92.55% :     0.000214s :      6: match.inline
 2.76% :     0.000006s :      1: match.switch_simplify
 4.69% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   801
 1.33% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.22% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     12: predicate.addn_check_dump
 1.41% :     0.000002s :     12: predicate.addn_zero_filter
 1.89% :     0.000002s :     12: predicate.arithmetic_simplify
 1.25% :     0.000002s :     12: predicate.cast_eliminate
 0.71% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     12: predicate.compare_switch_simplify
 1.45% :     0.000002s :     12: predicate.depend_value_elim
 1.23% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     12: predicate.environ_get_eliminate
 1.31% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.88% :     0.000008s :     26: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.71% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     17: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.24% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.09% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     12: predicate.merge_addn
 1.23% :     0.000002s :     12: predicate.minmaximum_grad
 1.31% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     17: predicate.partial_eliminate
 1.34% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     12: predicate.reduce_eliminate
 1.47% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     20: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.62% :     0.000002s :     12: predicate.reshape_eliminate
 1.22% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     12: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.80% :     0.000004s :     20: predicate.switch_defer_inline
 2.36% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.23% :     0.000011s :     65: predicate.switch_simplify
 1.20% :     0.000002s :     12: predicate.tile_eliminate
 1.25% :     0.000002s :     12: predicate.transpose_eliminate
 1.59% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.56% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.47% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.67% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     12: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000754    15
49.14% :     0.000370s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.93% :     0.000082s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.93% :     0.000301s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016471    72
 0.30% :     0.000049s :      1: add_recomputation
 0.54% :     0.000088s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.89% :     0.000311s :      1: bootstrap
 0.16% :     0.000027s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.16% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
34.75% :     0.005724s :      1: jit_opt_a
 0.85% :     0.000141s :      1: jit_opt_after_cconv
 0.31% :     0.000051s :      1: jit_opt_b
 2.58% :     0.000424s :      1: loop_unroll
 2.93% :     0.000483s :      1: mutable_eliminate
 6.22% :     0.001025s :     26: opt.transform.jit_opt_a
 0.28% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.81% :     0.000463s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.32% :     0.000053s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.10% :     0.000345s :      1: renormalize.infer
 1.60% :     0.000264s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000035s :      1: rewriter_after_opt_a
 0.53% :     0.000087s :      1: rewriter_before_opt_a
 0.47% :     0.000078s :      1: symbol_engine_optimizer
39.59% :     0.006521s :      1: type_inference


TotalTime = 0.0219886, [33]
[bootstrap]: 0.0004566
[type_inference]: 0.00569246
[event_method]: 1.925e-05
[auto_monad]: 6.596e-05
[graph_reusing]: 6.06e-06
[pre_auto_parallel]: 2.41998e-06
[py_interpret_to_execute]: 4.109e-05
[rewriter_before_opt_a]: 7.974e-05
[expand_dump_flag]: 3.05998e-06
[jit_opt_a]: 0.00637073, [2]
    [Cycle 1]: 0.00196703, [27]
        [switch_simplify]: 7.056e-05
        [loop_unroll]: 2.931e-05
        [a_1]: 0.00058766
        [with_stream_mark]: 2.335e-05
        [recompute_prepare]: 7.08e-06
        [updatestate_depend_eliminate]: 3.56001e-06
        [updatestate_assign_eliminate]: 3.43e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [parameter_eliminate]: 2.50002e-06
        [specialize_transform]: 6.09001e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.74e-06
        [meta_shard_fg_expand]: 2.33002e-06
        [get_grad_eliminate_]: 5.35001e-06
        [merge_forward]: 3.66001e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.46e-05
        [j_node_and_user_rematch]: 9.38002e-06
        [meta_fg_expand]: 2.37001e-06
        [replace_old_param]: 1.019e-05
        [inline_without_move]: 5.19e-06
        [renormalize]: 0.00091829
        [add_forward_monad_depend]: 6.18998e-06
        [auto_monad_grad]: 2.63998e-06
        [auto_monad_eliminator]: 1.449e-05
        [cse]: 3.138e-05
        [replace_applicator]: 1.235e-05
    [Cycle 2]: 0.0003309, [27]
        [switch_simplify]: 5.89e-06
        [loop_unroll]: 4.92999e-06
        [a_1]: 9.105e-05
        [with_stream_mark]: 1.042e-05
        [recompute_prepare]: 4.97e-06
        [updatestate_depend_eliminate]: 2.84999e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.80001e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 5.00001e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 5.25001e-06
        [meta_shard_fg_expand]: 1.19998e-06
        [get_grad_eliminate_]: 4.65001e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 1.88997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.222e-05
        [j_node_and_user_rematch]: 7.38999e-06
        [meta_fg_expand]: 1.94e-06
        [replace_old_param]: 1.491e-05
        [inline_without_move]: 4.77998e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.42e-06
        [auto_monad_grad]: 1.24998e-06
        [auto_monad_eliminator]: 5.67001e-06
        [cse]: 1.233e-05
        [replace_applicator]: 5.54e-06
[py_interpret_to_execute_after_opt_a]: 1.346e-05
[rewriter_after_opt_a]: 4.012e-05
[convert_after_rewriter]: 7.6e-06
[order_py_execute_after_rewriter]: 7.9e-06
[mutable_eliminate]: 0.00068339
[jit_opt_b]: 5.276e-05, [1]
    [Cycle 1]: 4.475e-05, [2]
        [frontend_op_eliminate]: 1.817e-05
        [inline_after_opt_a]: 1.518e-05
[cconv]: 3.01e-05
[loop_unroll]: 0.00044141
[jit_opt_after_cconv]: 0.00014277, [1]
    [Cycle 1]: 0.00013636, [11]
        [c_1]: 1.992e-05
        [parameter_eliminate]: 3.31001e-06
        [updatestate_depend_eliminate]: 5.99999e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 2.017e-05
        [call_graph_tuple_transform]: 1.94e-05
        [tuple_list_get_item_eliminator]: 5.52001e-06
        [none_parameter_eliminate]: 1.81e-06
        [renormalize]: 8.00006e-07
        [switch_simplify]: 5.20999e-06
[remove_dup_value]: 1.536e-05
[partial_unused_args_eliminate]: 2.69001e-06
[environ_conv]: 5.81998e-06
[add_recomputation]: 5.406e-05
[cse_after_recomputation]: 1.986e-05, [1]
    [Cycle 1]: 1.466e-05, [1]
        [cse]: 8.89998e-06
[auto_monad_reorder]: 1.541e-05
[get_jit_bprop_graph]: 1.99e-06
[rewriter_after_jit_bprop_graph]: 5.29e-06
[opt_after_jit_grad]: 0.00049889
[symbol_engine_optimizer]: 8.09e-05, [1]
    [Cycle 1]: 7.168e-05, [6]
        [build]: 3.92998e-06
        [elim_shapecalc]: 8.33001e-06
        [elim_not_effective]: 1.48e-05
        [opt_reshape]: 5.96e-06
        [fold_const_symbol]: 8.94998e-06
        [renormalize]: 1.19998e-06
[validate]: 4.735e-05
[backend_pass]: 1.19998e-06
[task_emit]: 0.006829
[execute]: 9.86998e-06
Sums
    bootstrap                            :     0.000457s :  2.65%
    type_inference                       :     0.005692s : 32.99%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000066s :  0.38%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.24%
    rewriter_before_opt_a                :     0.000080s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.44%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000679s :  3.93%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000025s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000918s :  5.32%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000044s :  0.25%
    jit_opt_a.replace_applicator         :     0.000018s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000040s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000008s :  0.05%
    mutable_eliminate                    :     0.000683s :  3.96%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000030s :  0.17%
    loop_unroll                          :     0.000441s :  2.56%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000054s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000499s :  2.89%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000047s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006829s : 39.58%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000248    23
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000002s :      2: substitution.fold_const_symbol
 2.25% :     0.000006s :      3: substitution.graph_param_transform
85.79% :     0.000213s :      4: substitution.inline
 1.31% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.82% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.17% :     0.000005s :      2: substitution.replace_old_param
 5.27% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005608     2
83.14% :     0.004662s :      1: type_inference.infer
16.86% :     0.000946s :      1: type_inference.specialize
------[replace.]   0.000075     6
68.16% :     0.000051s :      4: replace.inline
31.84% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000223     6
94.62% :     0.000211s :      4: match.inline
 5.38% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.39% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.62% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.16% :     0.000001s :     10: predicate.addn_zero_filter
 2.30% :     0.000003s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.62% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.49% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.24% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.50% :     0.000001s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.48% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.31% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.74% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.67% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.22% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 2.06% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.58% :     0.000003s :     15: predicate.partial_eliminate
 1.32% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.62% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.05% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.60% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.27% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.61% :     0.000009s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.07% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.95% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000705    12
48.82% :     0.000344s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.02% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.16% :     0.000311s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023913    76
 0.24% :     0.000057s :      1: add_recomputation
 0.29% :     0.000069s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 2.00% :     0.000478s :      1: bootstrap
 0.14% :     0.000033s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000025s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.66% :     0.006375s :      1: jit_opt_a
 0.61% :     0.000145s :      1: jit_opt_after_cconv
 0.23% :     0.000056s :      1: jit_opt_b
 1.88% :     0.000450s :      1: loop_unroll
 2.90% :     0.000693s :      1: mutable_eliminate
 3.78% :     0.000905s :     26: opt.transform.jit_opt_a
 0.20% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.13% :     0.000510s :      1: opt_after_jit_grad
 0.04% :     0.000010s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000044s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 1.98% :     0.000473s :      1: renormalize.infer
 1.83% :     0.000437s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000043s :      1: rewriter_after_opt_a
 0.35% :     0.000083s :      1: rewriter_before_opt_a
 0.35% :     0.000084s :      1: symbol_engine_optimizer
28.65% :     0.006851s :      1: task_emit
23.88% :     0.005711s :      1: type_inference
 0.30% :     0.000072s :      1: validate


TotalTime = 0.0153886, [30]
[bootstrap]: 0.00030261
[type_inference]: 0.00660279
[event_method]: 2.401e-05
[auto_monad]: 8.733e-05
[graph_reusing]: 6.78e-06
[pre_auto_parallel]: 2.83e-06
[py_interpret_to_execute]: 5.469e-05
[rewriter_before_opt_a]: 8.588e-05
[expand_dump_flag]: 3.18998e-06
[jit_opt_a]: 0.00588193, [2]
    [Cycle 1]: 0.00205851, [27]
        [switch_simplify]: 0.00012834
        [loop_unroll]: 3.524e-05
        [a_1]: 0.00073281
        [with_stream_mark]: 2.083e-05
        [recompute_prepare]: 8.69003e-06
        [updatestate_depend_eliminate]: 3.83999e-06
        [updatestate_assign_eliminate]: 2.76999e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 2.51998e-06
        [specialize_transform]: 5.97001e-06
        [updatestate_useless_node_eliminater]: 5.35999e-06
        [accelerated_algorithm]: 5.37999e-06
        [meta_shard_fg_expand]: 2.31998e-06
        [get_grad_eliminate_]: 5.35999e-06
        [merge_forward]: 3.83001e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.399e-05
        [j_node_and_user_rematch]: 8.87e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 9.84999e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.00078281
        [add_forward_monad_depend]: 6.16e-06
        [auto_monad_grad]: 2.86e-06
        [auto_monad_eliminator]: 1.557e-05
        [cse]: 3.257e-05
        [replace_applicator]: 1.422e-05
    [Cycle 2]: 0.00032914, [27]
        [switch_simplify]: 6.66999e-06
        [loop_unroll]: 5.16998e-06
        [a_1]: 9.511e-05
        [with_stream_mark]: 1.065e-05
        [recompute_prepare]: 5.20999e-06
        [updatestate_depend_eliminate]: 2.79001e-06
        [updatestate_assign_eliminate]: 2.58998e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [parameter_eliminate]: 1.35001e-06
        [specialize_transform]: 5.06002e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 4.67e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.67e-06
        [merge_forward]: 2.79999e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.419e-05
        [j_node_and_user_rematch]: 7.87003e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 7.51001e-06
        [inline_without_move]: 4.84e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 9.89996e-07
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 5.18002e-06
        [cse]: 1.056e-05
        [replace_applicator]: 5.17e-06
[py_interpret_to_execute_after_opt_a]: 1.288e-05
[rewriter_after_opt_a]: 3.864e-05
[convert_after_rewriter]: 7.05e-06
[order_py_execute_after_rewriter]: 4.97e-06
[mutable_eliminate]: 0.00064259
[jit_opt_b]: 6.085e-05, [1]
    [Cycle 1]: 5.273e-05, [2]
        [frontend_op_eliminate]: 1.593e-05
        [inline_after_opt_a]: 1.645e-05
[cconv]: 2.784e-05
[loop_unroll]: 0.00049469
[jit_opt_after_cconv]: 0.00014971, [1]
    [Cycle 1]: 0.00014249, [11]
        [c_1]: 2.142e-05
        [parameter_eliminate]: 3.53999e-06
        [updatestate_depend_eliminate]: 7.45e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [cse]: 2.051e-05
        [call_graph_tuple_transform]: 2.097e-05
        [tuple_list_get_item_eliminator]: 5.51998e-06
        [none_parameter_eliminate]: 1.72001e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 1.494e-05
[partial_unused_args_eliminate]: 2.11998e-06
[environ_conv]: 5.43002e-06
[add_recomputation]: 5.144e-05
[cse_after_recomputation]: 2.085e-05, [1]
    [Cycle 1]: 1.545e-05, [1]
        [cse]: 9.31e-06
[auto_monad_reorder]: 1.559e-05
[get_jit_bprop_graph]: 1.97001e-06
[rewriter_after_jit_bprop_graph]: 6.24001e-06
[opt_after_jit_grad]: 0.00048128
[symbol_engine_optimizer]: 7.359e-05, [1]
    [Cycle 1]: 6.724e-05, [6]
        [build]: 3.61999e-06
        [elim_shapecalc]: 7.46999e-06
        [elim_not_effective]: 1.293e-05
        [opt_reshape]: 5.81e-06
        [fold_const_symbol]: 8.35001e-06
        [renormalize]: 7.00005e-07
[validate]: 3.609e-05
Sums
    bootstrap                            :     0.000303s :  2.69%
    type_inference                       :     0.006603s : 58.62%
    event_method                         :     0.000024s :  0.21%
    auto_monad                           :     0.000087s :  0.78%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000055s :  0.49%
    rewriter_before_opt_a                :     0.000086s :  0.76%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000135s :  1.20%
    jit_opt_a.loop_unroll                :     0.000040s :  0.36%
    jit_opt_a.a_1                        :     0.000828s :  7.35%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000783s :  6.95%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.18%
    jit_opt_a.cse                        :     0.000043s :  0.38%
    jit_opt_a.replace_applicator         :     0.000019s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.11%
    rewriter_after_opt_a                 :     0.000039s :  0.34%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000643s :  5.71%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.15%
    cconv                                :     0.000028s :  0.25%
    loop_unroll                          :     0.000495s :  4.39%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000015s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000051s :  0.46%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000481s :  4.27%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000036s :  0.32%


Time group info:
------[substitution.]   0.000303    26
 0.58% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 1.82% :     0.000006s :      3: substitution.graph_param_transform
85.62% :     0.000259s :      6: substitution.inline
 0.95% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.51% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.66% :     0.000005s :      2: substitution.replace_old_param
 2.81% :     0.000009s :      1: substitution.switch_simplify
 4.57% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006525     2
82.00% :     0.005350s :      1: type_inference.infer
18.00% :     0.001174s :      1: type_inference.specialize
------[replace.]   0.000144     9
50.50% :     0.000073s :      6: replace.inline
26.87% :     0.000039s :      1: replace.switch_simplify
22.63% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000276     9
92.75% :     0.000256s :      6: match.inline
 2.68% :     0.000007s :      1: match.switch_simplify
 4.58% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   801
 1.23% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.06% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     12: predicate.addn_check_dump
 1.49% :     0.000002s :     12: predicate.addn_zero_filter
 2.30% :     0.000003s :     12: predicate.arithmetic_simplify
 1.39% :     0.000002s :     12: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000002s :     12: predicate.compare_switch_simplify
 1.38% :     0.000002s :     12: predicate.depend_value_elim
 1.11% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.39% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.71% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.21% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.39% :     0.000002s :     12: predicate.environ_get_eliminate
 1.13% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.58% :     0.000001s :      3: predicate.graph_param_transform
 5.33% :     0.000007s :     26: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     17: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.06% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     12: predicate.merge_addn
 1.13% :     0.000002s :     12: predicate.minmaximum_grad
 2.02% :     0.000003s :      3: predicate.mutable_eliminate
 0.39% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000003s :     17: predicate.partial_eliminate
 1.46% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.79% :     0.000002s :     12: predicate.reduce_eliminate
 1.56% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.39% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     20: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     12: predicate.reshape_eliminate
 1.24% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     12: predicate.same_eliminate
 0.60% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.78% :     0.000004s :     20: predicate.switch_defer_inline
 2.27% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.60% :     0.000012s :     65: predicate.switch_simplify
 1.18% :     0.000002s :     12: predicate.tile_eliminate
 1.20% :     0.000002s :     12: predicate.transpose_eliminate
 1.62% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.96% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.65% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000816    15
45.65% :     0.000373s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.61% :     0.000111s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.74% :     0.000333s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017349    72
 0.32% :     0.000055s :      1: add_recomputation
 0.52% :     0.000091s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 1.85% :     0.000322s :      1: bootstrap
 0.18% :     0.000031s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.13% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.17% :     0.000029s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.93% :     0.005886s :      1: jit_opt_a
 0.88% :     0.000153s :      1: jit_opt_after_cconv
 0.37% :     0.000064s :      1: jit_opt_b
 2.90% :     0.000504s :      1: loop_unroll
 3.76% :     0.000653s :      1: mutable_eliminate
 6.45% :     0.001120s :     26: opt.transform.jit_opt_a
 0.29% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.82% :     0.000490s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.33% :     0.000058s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000018s :      1: remove_dup_value
 2.57% :     0.000447s :      1: renormalize.infer
 1.88% :     0.000326s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000042s :      1: rewriter_after_opt_a
 0.51% :     0.000089s :      1: rewriter_before_opt_a
 0.44% :     0.000076s :      1: symbol_engine_optimizer
38.17% :     0.006622s :      1: type_inference


TotalTime = 0.0257071, [33]
[bootstrap]: 0.0002882
[type_inference]: 0.00650761
[event_method]: 1.924e-05
[auto_monad]: 6.683e-05
[graph_reusing]: 5.93998e-06
[pre_auto_parallel]: 2.38002e-06
[py_interpret_to_execute]: 5.025e-05
[rewriter_before_opt_a]: 7.215e-05
[expand_dump_flag]: 3.32002e-06
[jit_opt_a]: 0.005913, [2]
    [Cycle 1]: 0.00183342, [27]
        [switch_simplify]: 5.143e-05
        [loop_unroll]: 3.253e-05
        [a_1]: 0.00055041
        [with_stream_mark]: 2.154e-05
        [recompute_prepare]: 1.009e-05
        [updatestate_depend_eliminate]: 4.32e-06
        [updatestate_assign_eliminate]: 2.78998e-06
        [updatestate_loads_eliminate]: 2.36998e-06
        [parameter_eliminate]: 2.03997e-06
        [specialize_transform]: 5.81e-06
        [updatestate_useless_node_eliminater]: 5.44998e-06
        [accelerated_algorithm]: 5.92999e-06
        [meta_shard_fg_expand]: 2.59999e-06
        [get_grad_eliminate_]: 5.97999e-06
        [merge_forward]: 4.08001e-06
        [cell_reuse_recompute_pass]: 1.71e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.479e-05
        [j_node_and_user_rematch]: 1.008e-05
        [meta_fg_expand]: 2.21e-06
        [replace_old_param]: 9.64e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.00082547
        [add_forward_monad_depend]: 6.54999e-06
        [auto_monad_grad]: 2.39001e-06
        [auto_monad_eliminator]: 1.564e-05
        [cse]: 2.833e-05
        [replace_applicator]: 1.468e-05
    [Cycle 2]: 0.00034758, [27]
        [switch_simplify]: 6.33e-06
        [loop_unroll]: 4.85999e-06
        [a_1]: 9.293e-05
        [with_stream_mark]: 1.261e-05
        [recompute_prepare]: 5.30001e-06
        [updatestate_depend_eliminate]: 3.11001e-06
        [updatestate_assign_eliminate]: 2.55002e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [parameter_eliminate]: 1.18001e-06
        [specialize_transform]: 5.02999e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.07999e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 4.67e-06
        [merge_forward]: 3.12002e-06
        [cell_reuse_recompute_pass]: 1.94e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.439e-05
        [j_node_and_user_rematch]: 8e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 8.21002e-06
        [inline_without_move]: 4.85001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 1.09e-06
        [auto_monad_eliminator]: 6.32001e-06
        [cse]: 1.295e-05
        [replace_applicator]: 6.36998e-06
[py_interpret_to_execute_after_opt_a]: 1.341e-05
[rewriter_after_opt_a]: 3.929e-05
[convert_after_rewriter]: 9.68002e-06
[order_py_execute_after_rewriter]: 5.37999e-06
[mutable_eliminate]: 0.0007848
[jit_opt_b]: 5.255e-05, [1]
    [Cycle 1]: 4.443e-05, [2]
        [frontend_op_eliminate]: 1.68e-05
        [inline_after_opt_a]: 1.57e-05
[cconv]: 2.863e-05
[loop_unroll]: 0.00055163
[jit_opt_after_cconv]: 0.00016211, [1]
    [Cycle 1]: 0.00015459, [11]
        [c_1]: 2.247e-05
        [parameter_eliminate]: 3.8e-06
        [updatestate_depend_eliminate]: 6.86001e-06
        [updatestate_assign_eliminate]: 2.89001e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [cse]: 2.678e-05
        [call_graph_tuple_transform]: 2.138e-05
        [tuple_list_get_item_eliminator]: 5.91e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 7.59988e-07
        [switch_simplify]: 5.99e-06
[remove_dup_value]: 1.374e-05
[partial_unused_args_eliminate]: 2.22999e-06
[environ_conv]: 6.07001e-06
[add_recomputation]: 5.225e-05
[cse_after_recomputation]: 2.365e-05, [1]
    [Cycle 1]: 1.762e-05, [1]
        [cse]: 1.089e-05
[auto_monad_reorder]: 1.54e-05
[get_jit_bprop_graph]: 2.09e-06
[rewriter_after_jit_bprop_graph]: 8.33999e-06
[opt_after_jit_grad]: 0.00059557
[symbol_engine_optimizer]: 7.907e-05, [1]
    [Cycle 1]: 7.095e-05, [6]
        [build]: 5.44e-06
        [elim_shapecalc]: 7.63001e-06
        [elim_not_effective]: 1.386e-05
        [opt_reshape]: 5.52001e-06
        [fold_const_symbol]: 8.83001e-06
        [renormalize]: 1.32e-06
[validate]: 4.127e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.0100478
[execute]: 8.42e-06
Sums
    bootstrap                            :     0.000288s :  1.35%
    type_inference                       :     0.006508s : 30.56%
    event_method                         :     0.000019s :  0.09%
    auto_monad                           :     0.000067s :  0.31%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000050s :  0.24%
    rewriter_before_opt_a                :     0.000072s :  0.34%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.27%
    jit_opt_a.loop_unroll                :     0.000037s :  0.18%
    jit_opt_a.a_1                        :     0.000643s :  3.02%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000826s :  3.88%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.10%
    jit_opt_a.cse                        :     0.000041s :  0.19%
    jit_opt_a.replace_applicator         :     0.000021s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.06%
    rewriter_after_opt_a                 :     0.000039s :  0.18%
    convert_after_rewriter               :     0.000010s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000785s :  3.69%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.07%
    cconv                                :     0.000029s :  0.13%
    loop_unroll                          :     0.000552s :  2.59%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000052s :  0.25%
    cse_after_recomputation.cse          :     0.000011s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000596s :  2.80%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.19%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010048s : 47.18%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000205    23
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.61% :     0.000005s :      3: substitution.graph_param_transform
82.93% :     0.000170s :      4: substitution.inline
 1.90% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.22% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.59% :     0.000005s :      2: substitution.replace_old_param
 6.26% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006437     2
74.38% :     0.004788s :      1: type_inference.infer
25.62% :     0.001649s :      1: type_inference.specialize
------[replace.]   0.000073     6
64.07% :     0.000047s :      4: replace.inline
35.93% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000179     6
93.49% :     0.000168s :      4: match.inline
 6.51% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.94% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 3.01% :     0.000004s :     10: predicate.arithmetic_simplify
 1.39% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.48% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.59% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.32% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.44% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000001s :     10: predicate.environ_get_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.52% :     0.000002s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.11% :     0.000006s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :      6: predicate.less_batch_normalization
 1.33% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.84% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 2.41% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.59% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.52% :     0.000002s :     18: predicate.replace_applicator
 0.86% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.19% :     0.000001s :     10: predicate.reshape_eliminate
 1.30% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.23% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.08% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 1.93% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.21% :     0.000010s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.35% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001281    12
20.67% :     0.000265s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 2.43% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
76.90% :     0.000985s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027515    76
 0.20% :     0.000055s :      1: add_recomputation
 0.26% :     0.000070s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.12% :     0.000308s :      1: bootstrap
 0.11% :     0.000032s :      1: cconv
 0.04% :     0.000012s :      1: convert_after_rewriter
 0.09% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.08% :     0.000023s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
21.50% :     0.005917s :      1: jit_opt_a
 0.60% :     0.000165s :      1: jit_opt_after_cconv
 0.20% :     0.000056s :      1: jit_opt_b
 2.06% :     0.000568s :      1: loop_unroll
 2.90% :     0.000798s :      1: mutable_eliminate
 3.14% :     0.000864s :     26: opt.transform.jit_opt_a
 0.19% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000607s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000054s :      1: py_interpret_to_execute
 0.06% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000016s :      1: remove_dup_value
 1.71% :     0.000472s :      1: renormalize.infer
 1.25% :     0.000344s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000043s :      1: rewriter_after_opt_a
 0.27% :     0.000075s :      1: rewriter_before_opt_a
 0.30% :     0.000082s :      1: symbol_engine_optimizer
36.57% :     0.010063s :      1: task_emit
23.71% :     0.006525s :      1: type_inference
 0.23% :     0.000065s :      1: validate


TotalTime = 0.0160781, [30]
[bootstrap]: 0.0003083
[type_inference]: 0.00682292
[event_method]: 2.208e-05
[auto_monad]: 8.539e-05
[graph_reusing]: 6.43998e-06
[pre_auto_parallel]: 2.75997e-06
[py_interpret_to_execute]: 5.848e-05
[rewriter_before_opt_a]: 8.733e-05
[expand_dump_flag]: 3.44001e-06
[jit_opt_a]: 0.0061188, [2]
    [Cycle 1]: 0.00206922, [27]
        [switch_simplify]: 0.00013283
        [loop_unroll]: 3.596e-05
        [a_1]: 0.00073133
        [with_stream_mark]: 1.889e-05
        [recompute_prepare]: 8.72e-06
        [updatestate_depend_eliminate]: 4.18999e-06
        [updatestate_assign_eliminate]: 3.73001e-06
        [updatestate_loads_eliminate]: 2.62001e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 5.40999e-06
        [accelerated_algorithm]: 6.34999e-06
        [meta_shard_fg_expand]: 2.36998e-06
        [get_grad_eliminate_]: 5.61e-06
        [merge_forward]: 4.1e-06
        [cell_reuse_recompute_pass]: 1.00999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.53e-05
        [j_node_and_user_rematch]: 1.015e-05
        [meta_fg_expand]: 2.47001e-06
        [replace_old_param]: 9.42999e-06
        [inline_without_move]: 5.39998e-06
        [renormalize]: 0.00078613
        [add_forward_monad_depend]: 6.61999e-06
        [auto_monad_grad]: 2.53003e-06
        [auto_monad_eliminator]: 1.467e-05
        [cse]: 3.204e-05
        [replace_applicator]: 1.488e-05
    [Cycle 2]: 0.0003485, [27]
        [switch_simplify]: 5.90002e-06
        [loop_unroll]: 5.32999e-06
        [a_1]: 9.605e-05
        [with_stream_mark]: 1.087e-05
        [recompute_prepare]: 5.23002e-06
        [updatestate_depend_eliminate]: 2.86e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 1.28002e-06
        [specialize_transform]: 4.95001e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.84e-06
        [meta_shard_fg_expand]: 1.69e-06
        [get_grad_eliminate_]: 5.18002e-06
        [merge_forward]: 3.88999e-06
        [cell_reuse_recompute_pass]: 2.71999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.477e-05
        [j_node_and_user_rematch]: 8.46002e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 7.50998e-06
        [inline_without_move]: 5.21002e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.83002e-06
        [auto_monad_grad]: 1.33002e-06
        [auto_monad_eliminator]: 5.27001e-06
        [cse]: 1.11e-05
        [replace_applicator]: 5.89e-06
[py_interpret_to_execute_after_opt_a]: 1.234e-05
[rewriter_after_opt_a]: 3.728e-05
[convert_after_rewriter]: 7.16001e-06
[order_py_execute_after_rewriter]: 5.07e-06
[mutable_eliminate]: 0.00071821
[jit_opt_b]: 5.714e-05, [1]
    [Cycle 1]: 4.867e-05, [2]
        [frontend_op_eliminate]: 1.885e-05
        [inline_after_opt_a]: 1.69e-05
[cconv]: 2.912e-05
[loop_unroll]: 0.00050141
[jit_opt_after_cconv]: 0.00021425, [1]
    [Cycle 1]: 0.00020625, [11]
        [c_1]: 2.253e-05
        [parameter_eliminate]: 2.91e-06
        [updatestate_depend_eliminate]: 5.231e-05
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [cse]: 2.521e-05
        [call_graph_tuple_transform]: 2.542e-05
        [tuple_list_get_item_eliminator]: 5.99999e-06
        [none_parameter_eliminate]: 1.91e-06
        [renormalize]: 8.30012e-07
        [switch_simplify]: 6.02999e-06
[remove_dup_value]: 1.573e-05
[partial_unused_args_eliminate]: 2.58e-06
[environ_conv]: 6.76e-06
[add_recomputation]: 5.527e-05
[cse_after_recomputation]: 2.337e-05, [1]
    [Cycle 1]: 1.676e-05, [1]
        [cse]: 9.90002e-06
[auto_monad_reorder]: 1.571e-05
[get_jit_bprop_graph]: 2.38002e-06
[rewriter_after_jit_bprop_graph]: 6.61999e-06
[opt_after_jit_grad]: 0.00053679
[symbol_engine_optimizer]: 8.166e-05, [1]
    [Cycle 1]: 7.35e-05, [6]
        [build]: 4.92999e-06
        [elim_shapecalc]: 8.06001e-06
        [elim_not_effective]: 1.486e-05
        [opt_reshape]: 6.59999e-06
        [fold_const_symbol]: 9.22001e-06
        [renormalize]: 6.90023e-07
[validate]: 3.918e-05
Sums
    bootstrap                            :     0.000308s :  2.63%
    type_inference                       :     0.006823s : 58.20%
    event_method                         :     0.000022s :  0.19%
    auto_monad                           :     0.000085s :  0.73%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000058s :  0.50%
    rewriter_before_opt_a                :     0.000087s :  0.74%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000139s :  1.18%
    jit_opt_a.loop_unroll                :     0.000041s :  0.35%
    jit_opt_a.a_1                        :     0.000827s :  7.06%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.09%
    jit_opt_a.merge_forward              :     0.000008s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.14%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.000786s :  6.71%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.17%
    jit_opt_a.cse                        :     0.000043s :  0.37%
    jit_opt_a.replace_applicator         :     0.000021s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.11%
    rewriter_after_opt_a                 :     0.000037s :  0.32%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000718s :  6.13%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.14%
    cconv                                :     0.000029s :  0.25%
    loop_unroll                          :     0.000501s :  4.28%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000052s :  0.45%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000025s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000016s :  0.13%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000007s :  0.06%
    add_recomputation                    :     0.000055s :  0.47%
    cse_after_recomputation.cse          :     0.000010s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000537s :  4.58%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.33%


Time group info:
------[substitution.]   0.000307    26
 0.80% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.19% :     0.000007s :      3: substitution.graph_param_transform
85.05% :     0.000261s :      6: substitution.inline
 1.17% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.59% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.41% :     0.000004s :      2: substitution.replace_old_param
 3.01% :     0.000009s :      1: substitution.switch_simplify
 4.30% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006747     2
82.89% :     0.005593s :      1: type_inference.infer
17.11% :     0.001154s :      1: type_inference.specialize
------[replace.]   0.000147     9
48.37% :     0.000071s :      6: replace.inline
26.73% :     0.000039s :      1: replace.switch_simplify
24.90% :     0.000037s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000278     9
92.73% :     0.000258s :      6: match.inline
 3.06% :     0.000009s :      1: match.switch_simplify
 4.22% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000141   801
 1.27% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     12: predicate.addn_check_dump
 1.29% :     0.000002s :     12: predicate.addn_zero_filter
 2.31% :     0.000003s :     12: predicate.arithmetic_simplify
 1.46% :     0.000002s :     12: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     12: predicate.compare_switch_simplify
 1.54% :     0.000002s :     12: predicate.depend_value_elim
 1.05% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.56% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.66% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.09% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.07% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     12: predicate.environ_get_eliminate
 1.13% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.15% :     0.000002s :      6: predicate.get_grad_eliminate
 0.36% :     0.000001s :      3: predicate.graph_param_transform
 5.26% :     0.000007s :     26: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000002s :      6: predicate.less_batch_normalization
 1.71% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     17: predicate.load_eliminater
 1.18% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.90% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000002s :     12: predicate.merge_addn
 1.05% :     0.000001s :     12: predicate.minmaximum_grad
 1.48% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.02% :     0.000003s :     17: predicate.partial_eliminate
 1.13% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     12: predicate.reduce_eliminate
 1.46% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     20: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.57% :     0.000002s :     12: predicate.reshape_eliminate
 1.28% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 0.86% :     0.000001s :      6: predicate.specialize_transform
 1.66% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.59% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.66% :     0.000004s :     20: predicate.switch_defer_inline
 2.21% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.84% :     0.000012s :     65: predicate.switch_simplify
 1.50% :     0.000002s :     12: predicate.tile_eliminate
 1.53% :     0.000002s :     12: predicate.transpose_eliminate
 1.46% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.85% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.48% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000844    15
46.59% :     0.000393s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.83% :     0.000108s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.58% :     0.000342s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018061    72
 0.32% :     0.000058s :      1: add_recomputation
 0.49% :     0.000089s :      1: auto_monad
 0.10% :     0.000019s :      1: auto_monad_reorder
 1.80% :     0.000326s :      1: bootstrap
 0.18% :     0.000032s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000010s :      1: environ_conv
 0.15% :     0.000027s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.90% :     0.006123s :      1: jit_opt_a
 1.21% :     0.000218s :      1: jit_opt_after_cconv
 0.33% :     0.000060s :      1: jit_opt_b
 2.83% :     0.000511s :      1: loop_unroll
 4.04% :     0.000729s :      1: mutable_eliminate
 6.26% :     0.001131s :     26: opt.transform.jit_opt_a
 0.31% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000028s :      4: opt.transform.jit_opt_b
 0.08% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.02% :     0.000546s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.34% :     0.000062s :      1: py_interpret_to_execute
 0.08% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000018s :      1: remove_dup_value
 2.22% :     0.000401s :      1: renormalize.infer
 2.09% :     0.000377s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000040s :      1: rewriter_after_opt_a
 0.50% :     0.000090s :      1: rewriter_before_opt_a
 0.47% :     0.000085s :      1: symbol_engine_optimizer
37.87% :     0.006840s :      1: type_inference


TotalTime = 0.0909547, [33]
[bootstrap]: 0.00028466
[type_inference]: 0.0716452
[event_method]: 1.76e-05
[auto_monad]: 7.022e-05
[graph_reusing]: 6.12999e-06
[pre_auto_parallel]: 2.37001e-06
[py_interpret_to_execute]: 0.00029322
[rewriter_before_opt_a]: 8.743e-05
[expand_dump_flag]: 3.93001e-06
[jit_opt_a]: 0.00816352, [2]
    [Cycle 1]: 0.00372255, [27]
        [switch_simplify]: 6.132e-05
        [loop_unroll]: 3.157e-05
        [a_1]: 0.0006173
        [with_stream_mark]: 2.473e-05
        [recompute_prepare]: 1.219e-05
        [updatestate_depend_eliminate]: 4.60001e-06
        [updatestate_assign_eliminate]: 3.04001e-06
        [updatestate_loads_eliminate]: 2.88998e-06
        [parameter_eliminate]: 2.32001e-06
        [specialize_transform]: 8.38001e-06
        [updatestate_useless_node_eliminater]: 6.39001e-06
        [accelerated_algorithm]: 6.55002e-06
        [meta_shard_fg_expand]: 2.14e-06
        [get_grad_eliminate_]: 7.36001e-06
        [merge_forward]: 4.18001e-06
        [cell_reuse_recompute_pass]: 2.70002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.736e-05
        [j_node_and_user_rematch]: 9.64e-06
        [meta_fg_expand]: 2.25002e-06
        [replace_old_param]: 1.194e-05
        [inline_without_move]: 6.51e-06
        [renormalize]: 0.00256184
        [add_forward_monad_depend]: 8.70999e-06
        [auto_monad_grad]: 2.73003e-06
        [auto_monad_eliminator]: 2.169e-05
        [cse]: 3.284e-05
        [replace_applicator]: 2.541e-05
    [Cycle 2]: 0.00041061, [27]
        [switch_simplify]: 6.98e-06
        [loop_unroll]: 6.05002e-06
        [a_1]: 0.00013625
        [with_stream_mark]: 1.802e-05
        [recompute_prepare]: 6.79999e-06
        [updatestate_depend_eliminate]: 3.76001e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 2.19999e-06
        [specialize_transform]: 5.87999e-06
        [updatestate_useless_node_eliminater]: 6.67002e-06
        [accelerated_algorithm]: 6.19999e-06
        [meta_shard_fg_expand]: 1.82999e-06
        [get_grad_eliminate_]: 5.69999e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 4.4e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.553e-05
        [j_node_and_user_rematch]: 9.57999e-06
        [meta_fg_expand]: 1.98002e-06
        [replace_old_param]: 9.81e-06
        [inline_without_move]: 5.38002e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.59e-06
        [auto_monad_grad]: 1.40999e-06
        [auto_monad_eliminator]: 7.78001e-06
        [cse]: 1.363e-05
        [replace_applicator]: 5.66003e-06
[py_interpret_to_execute_after_opt_a]: 1.684e-05
[rewriter_after_opt_a]: 4.23e-05
[convert_after_rewriter]: 7e-06
[order_py_execute_after_rewriter]: 5.94999e-06
[mutable_eliminate]: 0.00077197
[jit_opt_b]: 6.042e-05, [1]
    [Cycle 1]: 5.078e-05, [2]
        [frontend_op_eliminate]: 2.033e-05
        [inline_after_opt_a]: 1.879e-05
[cconv]: 3.232e-05
[loop_unroll]: 0.00055929
[jit_opt_after_cconv]: 0.00017723, [1]
    [Cycle 1]: 0.000169, [11]
        [c_1]: 2.585e-05
        [parameter_eliminate]: 5.05999e-06
        [updatestate_depend_eliminate]: 8.77e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.39001e-06
        [cse]: 3.348e-05
        [call_graph_tuple_transform]: 2.351e-05
        [tuple_list_get_item_eliminator]: 6.71999e-06
        [none_parameter_eliminate]: 1.65001e-06
        [renormalize]: 7.29982e-07
        [switch_simplify]: 7.13998e-06
[remove_dup_value]: 1.667e-05
[partial_unused_args_eliminate]: 3.54002e-06
[environ_conv]: 9.89001e-06
[add_recomputation]: 5.79e-05
[cse_after_recomputation]: 2.607e-05, [1]
    [Cycle 1]: 1.954e-05, [1]
        [cse]: 1.213e-05
[auto_monad_reorder]: 1.513e-05
[get_jit_bprop_graph]: 2.29999e-06
[rewriter_after_jit_bprop_graph]: 8.54e-06
[opt_after_jit_grad]: 0.0005627
[symbol_engine_optimizer]: 0.0001262, [1]
    [Cycle 1]: 7.264e-05, [6]
        [build]: 4.17998e-06
        [elim_shapecalc]: 9.39998e-06
        [elim_not_effective]: 1.486e-05
        [opt_reshape]: 6.68998e-06
        [fold_const_symbol]: 9.19e-06
        [renormalize]: 6.19999e-07
[validate]: 5.411e-05
[backend_pass]: 1.50999e-06
[task_emit]: 0.00760126
[execute]: 7.9e-06
Sums
    bootstrap                            :     0.000285s :  0.33%
    type_inference                       :     0.071645s : 83.12%
    event_method                         :     0.000018s :  0.02%
    auto_monad                           :     0.000070s :  0.08%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.00%
    py_interpret_to_execute              :     0.000293s :  0.34%
    rewriter_before_opt_a                :     0.000087s :  0.10%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000068s :  0.08%
    jit_opt_a.loop_unroll                :     0.000038s :  0.04%
    jit_opt_a.a_1                        :     0.000754s :  0.87%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.05%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000014s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000013s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.01%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.00%
    jit_opt_a.get_grad_eliminate_        :     0.000013s :  0.02%
    jit_opt_a.merge_forward              :     0.000008s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.04%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.02%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.00%
    jit_opt_a.replace_old_param          :     0.000022s :  0.03%
    jit_opt_a.inline_without_move        :     0.000012s :  0.01%
    jit_opt_a.renormalize                :     0.002562s :  2.97%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000029s :  0.03%
    jit_opt_a.cse                        :     0.000046s :  0.05%
    jit_opt_a.replace_applicator         :     0.000031s :  0.04%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.02%
    rewriter_after_opt_a                 :     0.000042s :  0.05%
    convert_after_rewriter               :     0.000007s :  0.01%
    order_py_execute_after_rewriter      :     0.000006s :  0.01%
    mutable_eliminate                    :     0.000772s :  0.90%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.02%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.02%
    cconv                                :     0.000032s :  0.04%
    loop_unroll                          :     0.000559s :  0.65%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.03%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000033s :  0.04%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.01%
    remove_dup_value                     :     0.000017s :  0.02%
    partial_unused_args_eliminate        :     0.000004s :  0.00%
    environ_conv                         :     0.000010s :  0.01%
    add_recomputation                    :     0.000058s :  0.07%
    cse_after_recomputation.cse          :     0.000012s :  0.01%
    auto_monad_reorder                   :     0.000015s :  0.02%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.01%
    opt_after_jit_grad                   :     0.000563s :  0.65%
    symbol_engine_optimizer.build        :     0.000004s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.02%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000054s :  0.06%
    backend_pass                         :     0.000002s :  0.00%
    task_emit                            :     0.007601s :  8.82%
    execute                              :     0.000008s :  0.01%


Time group info:
------[substitution.]   0.000245    27
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.68% :     0.000007s :      4: substitution.graph_param_transform
77.96% :     0.000191s :      4: substitution.inline
 1.34% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.07% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.04% :     0.000005s :      2: substitution.replace_old_param
 8.13% :     0.000020s :      3: substitution.tile_eliminate
 4.47% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.071557     2
97.28% :     0.069611s :      1: type_inference.infer
 2.72% :     0.001946s :      1: type_inference.specialize
------[replace.]   0.000079     6
66.98% :     0.000053s :      4: replace.inline
33.02% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000199     6
95.05% :     0.000189s :      4: match.inline
 4.95% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000144   834
 1.40% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.59% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     12: predicate.addn_check_dump
 1.21% :     0.000002s :     12: predicate.addn_zero_filter
 2.87% :     0.000004s :     12: predicate.arithmetic_simplify
 1.77% :     0.000003s :     12: predicate.cast_eliminate
 0.44% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     12: predicate.compare_switch_simplify
 1.68% :     0.000002s :     12: predicate.depend_value_elim
 0.97% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.10% :     0.000002s :     12: predicate.dict_set_item_eliminator
 1.13% :     0.000002s :      4: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      4: predicate.elim_not_effective
 0.87% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000002s :     12: predicate.environ_add_const_eliminate
 0.95% :     0.000001s :     12: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     12: predicate.environ_get_eliminate
 0.98% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      4: predicate.fold_const_symbol
 1.18% :     0.000002s :      8: predicate.get_grad_eliminate
 0.39% :     0.000001s :      4: predicate.graph_param_transform
 5.33% :     0.000008s :     26: predicate.inline
 0.88% :     0.000001s :      8: predicate.inline_without_move
 0.44% :     0.000001s :      8: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      8: predicate.less_batch_normalization
 1.48% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.74% :     0.000003s :     18: predicate.load_eliminater
 1.66% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.36% :     0.000005s :     38: predicate.loop_unroll_before_grad
 1.93% :     0.000003s :     16: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000002s :     12: predicate.merge_addn
 0.95% :     0.000001s :     12: predicate.minmaximum_grad
 2.39% :     0.000003s :      4: predicate.mutable_eliminate
 0.71% :     0.000001s :      4: predicate.opt_reshape
 2.09% :     0.000003s :     18: predicate.partial_eliminate
 1.02% :     0.000001s :     12: predicate.print_const_string_wrapper
 1.97% :     0.000003s :     12: predicate.reduce_eliminate
 1.38% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      8: predicate.remove_not_recompute_node
 1.99% :     0.000003s :     22: predicate.replace_applicator
 0.72% :     0.000001s :      8: predicate.replace_old_param
 0.49% :     0.000001s :      4: predicate.reset_defer_inline
 1.22% :     0.000002s :     12: predicate.reshape_eliminate
 1.22% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.11% :     0.000002s :     12: predicate.same_eliminate
 0.74% :     0.000001s :      8: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      8: predicate.special_op_eliminate
 0.93% :     0.000001s :      8: predicate.specialize_transform
 1.39% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.06% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.32% :     0.000003s :     18: predicate.switch_defer_inline
 1.86% :     0.000003s :     18: predicate.switch_layer_defer_inline
 6.71% :     0.000010s :     60: predicate.switch_simplify
 1.61% :     0.000002s :     12: predicate.tile_eliminate
 1.31% :     0.000002s :     12: predicate.transpose_eliminate
 1.45% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.48% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000003s :     12: predicate.tuple_list_set_item_eliminator
 1.29% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     18: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000004s :     26: predicate.updatestate_useless_node_eliminater
 1.31% :     0.000002s :     12: predicate.value_based_eliminate
 0.40% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.002566    30
73.41% :     0.001884s :     23: func_graph_cloner_run.FuncGraphClonerGraph
 1.38% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
25.21% :     0.000647s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.094675    76
 0.06% :     0.000061s :      1: add_recomputation
 0.08% :     0.000073s :      1: auto_monad
 0.02% :     0.000018s :      1: auto_monad_reorder
 0.00% :     0.000005s :      1: backend_pass
 0.31% :     0.000296s :      1: bootstrap
 0.04% :     0.000036s :      1: cconv
 0.01% :     0.000010s :      1: convert_after_rewriter
 0.03% :     0.000028s :      1: cse_after_recomputation
 0.01% :     0.000012s :      1: environ_conv
 0.02% :     0.000021s :      1: event_method
 0.01% :     0.000012s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.00% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000008s :      1: graph_reusing
 8.63% :     0.008167s :      1: jit_opt_a
 0.19% :     0.000180s :      1: jit_opt_after_cconv
 0.07% :     0.000064s :      1: jit_opt_b
 0.60% :     0.000570s :      1: loop_unroll
 0.83% :     0.000784s :      1: mutable_eliminate
 1.07% :     0.001013s :     26: opt.transform.jit_opt_a
 0.06% :     0.000060s :      4: opt.transform.jit_opt_after_cconv
 0.03% :     0.000032s :      4: opt.transform.jit_opt_b
 0.02% :     0.000020s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.04% :     0.000036s :      4: opt.transform.symbol_engine_opt
 0.60% :     0.000573s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.00% :     0.000004s :      1: pre_auto_parallel
 0.32% :     0.000301s :      1: py_interpret_to_execute
 0.02% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000019s :      1: remove_dup_value
 1.93% :     0.001824s :      1: renormalize.infer
 0.77% :     0.000726s :      1: renormalize.specialize
 0.01% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.05% :     0.000047s :      1: rewriter_after_opt_a
 0.10% :     0.000093s :      1: rewriter_before_opt_a
 0.14% :     0.000129s :      1: symbol_engine_optimizer
 8.04% :     0.007614s :      1: task_emit
75.69% :     0.071658s :      1: type_inference
 0.08% :     0.000077s :      1: validate


TotalTime = 0.024142, [33]
[bootstrap]: 0.00039668
[type_inference]: 0.00572792
[event_method]: 1.83e-05
[auto_monad]: 7.302e-05
[graph_reusing]: 6.18002e-06
[pre_auto_parallel]: 2.26998e-06
[py_interpret_to_execute]: 4.744e-05
[rewriter_before_opt_a]: 7.399e-05
[expand_dump_flag]: 3.53e-06
[jit_opt_a]: 0.00709427, [2]
    [Cycle 1]: 0.00232111, [27]
        [switch_simplify]: 7.369e-05
        [loop_unroll]: 3.569e-05
        [a_1]: 0.00068917
        [with_stream_mark]: 2.427e-05
        [recompute_prepare]: 1.031e-05
        [updatestate_depend_eliminate]: 3.76999e-06
        [updatestate_assign_eliminate]: 3.06001e-06
        [updatestate_loads_eliminate]: 2.99999e-06
        [parameter_eliminate]: 2.41998e-06
        [specialize_transform]: 7.56999e-06
        [updatestate_useless_node_eliminater]: 5.64998e-06
        [accelerated_algorithm]: 7.06001e-06
        [meta_shard_fg_expand]: 2.93e-06
        [get_grad_eliminate_]: 6.15002e-06
        [merge_forward]: 4.45999e-06
        [cell_reuse_recompute_pass]: 1.52001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.881e-05
        [j_node_and_user_rematch]: 1.169e-05
        [meta_fg_expand]: 2.76e-06
        [replace_old_param]: 1.154e-05
        [inline_without_move]: 5.20001e-06
        [renormalize]: 0.00109241
        [add_forward_monad_depend]: 1.146e-05
        [auto_monad_grad]: 2.74999e-06
        [auto_monad_eliminator]: 1.822e-05
        [cse]: 2.348e-05
        [replace_applicator]: 2.314e-05
    [Cycle 2]: 0.00038306, [27]
        [switch_simplify]: 6.69999e-06
        [loop_unroll]: 5.22e-06
        [a_1]: 0.00011016
        [with_stream_mark]: 1.398e-05
        [recompute_prepare]: 5.24e-06
        [updatestate_depend_eliminate]: 3.38e-06
        [updatestate_assign_eliminate]: 2.39999e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [parameter_eliminate]: 1.76998e-06
        [specialize_transform]: 4.78001e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 6.16e-06
        [meta_shard_fg_expand]: 1.93997e-06
        [get_grad_eliminate_]: 5.32001e-06
        [merge_forward]: 3.99002e-06
        [cell_reuse_recompute_pass]: 2.70002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.853e-05
        [j_node_and_user_rematch]: 9.46e-06
        [meta_fg_expand]: 2.53e-06
        [replace_old_param]: 9.06998e-06
        [inline_without_move]: 4.90999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.88002e-06
        [auto_monad_grad]: 1.45001e-06
        [auto_monad_eliminator]: 6.38998e-06
        [cse]: 1.31e-05
        [replace_applicator]: 5.66e-06
[py_interpret_to_execute_after_opt_a]: 1.773e-05
[rewriter_after_opt_a]: 4.109e-05
[convert_after_rewriter]: 7.48e-06
[order_py_execute_after_rewriter]: 5.66e-06
[mutable_eliminate]: 0.00093756
[jit_opt_b]: 6.702e-05, [1]
    [Cycle 1]: 5.721e-05, [2]
        [frontend_op_eliminate]: 2.547e-05
        [inline_after_opt_a]: 1.819e-05
[cconv]: 3.52e-05
[loop_unroll]: 0.00054124
[jit_opt_after_cconv]: 0.00016124, [1]
    [Cycle 1]: 0.00015306, [11]
        [c_1]: 2.129e-05
        [parameter_eliminate]: 4.19002e-06
        [updatestate_depend_eliminate]: 6.64001e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.26998e-06
        [cse]: 2.749e-05
        [call_graph_tuple_transform]: 2.215e-05
        [tuple_list_get_item_eliminator]: 5.64e-06
        [none_parameter_eliminate]: 2.11998e-06
        [renormalize]: 9.20001e-07
        [switch_simplify]: 5.50001e-06
[remove_dup_value]: 1.597e-05
[partial_unused_args_eliminate]: 2.54001e-06
[environ_conv]: 6.36e-06
[add_recomputation]: 5.889e-05
[cse_after_recomputation]: 2.325e-05, [1]
    [Cycle 1]: 1.715e-05, [1]
        [cse]: 1.008e-05
[auto_monad_reorder]: 1.71e-05
[get_jit_bprop_graph]: 2.58998e-06
[rewriter_after_jit_bprop_graph]: 8.71997e-06
[opt_after_jit_grad]: 0.00060164
[symbol_engine_optimizer]: 8.362e-05, [1]
    [Cycle 1]: 7.483e-05, [6]
        [build]: 4.53001e-06
        [elim_shapecalc]: 9.04e-06
        [elim_not_effective]: 1.504e-05
        [opt_reshape]: 6.52001e-06
        [fold_const_symbol]: 9.51e-06
        [renormalize]: 1.10999e-06
[validate]: 4.638e-05
[backend_pass]: 8.80013e-07
[task_emit]: 0.0077631
[execute]: 8.77999e-06
Sums
    bootstrap                            :     0.000397s :  2.09%
    type_inference                       :     0.005728s : 30.11%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000073s :  0.38%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000047s :  0.25%
    rewriter_before_opt_a                :     0.000074s :  0.39%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000080s :  0.42%
    jit_opt_a.loop_unroll                :     0.000041s :  0.22%
    jit_opt_a.a_1                        :     0.000799s :  4.20%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000021s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.001092s :  5.74%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.13%
    jit_opt_a.cse                        :     0.000037s :  0.19%
    jit_opt_a.replace_applicator         :     0.000029s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.09%
    rewriter_after_opt_a                 :     0.000041s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000938s :  4.93%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.10%
    cconv                                :     0.000035s :  0.19%
    loop_unroll                          :     0.000541s :  2.84%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000059s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000602s :  3.16%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000046s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007763s : 40.80%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000299    23
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000002s :      2: substitution.fold_const_symbol
 2.29% :     0.000007s :      3: substitution.graph_param_transform
85.10% :     0.000254s :      4: substitution.inline
 1.72% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.95% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.16% :     0.000006s :      2: substitution.replace_old_param
 5.49% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005646     2
80.91% :     0.004569s :      1: type_inference.infer
19.09% :     0.001078s :      1: type_inference.specialize
------[replace.]   0.000091     6
66.16% :     0.000060s :      4: replace.inline
33.84% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000266     6
94.27% :     0.000251s :      4: match.inline
 5.73% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000146   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.86% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 3.17% :     0.000005s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.34% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.88% :     0.000001s :     10: predicate.compare_switch_simplify
 2.22% :     0.000003s :     10: predicate.depend_value_elim
 0.94% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.34% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      3: predicate.elim_not_effective
 0.44% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.89% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000002s :     10: predicate.environ_get_eliminate
 1.21% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 4.85% :     0.000007s :     22: predicate.inline
 0.70% :     0.000001s :      6: predicate.inline_without_move
 0.29% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.50% :     0.000002s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.42% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.11% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 1.87% :     0.000003s :      3: predicate.mutable_eliminate
 0.58% :     0.000001s :      3: predicate.opt_reshape
 1.89% :     0.000003s :     15: predicate.partial_eliminate
 1.35% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.16% :     0.000003s :     10: predicate.reduce_eliminate
 2.07% :     0.000003s :     12: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000001s :      3: predicate.reset_defer_inline
 1.85% :     0.000003s :     10: predicate.reshape_eliminate
 1.24% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.50% :     0.000002s :     10: predicate.same_eliminate
 0.42% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.37% :     0.000002s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.85% :     0.000004s :     16: predicate.switch_defer_inline
 2.14% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.07% :     0.000010s :     53: predicate.switch_simplify
 1.23% :     0.000002s :     10: predicate.tile_eliminate
 1.22% :     0.000002s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.99% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.68% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.33% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.19% :     0.000005s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.27% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000760    12
43.68% :     0.000332s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.16% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.16% :     0.000373s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026418    76
 0.24% :     0.000063s :      1: add_recomputation
 0.29% :     0.000077s :      1: auto_monad
 0.08% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.57% :     0.000415s :      1: bootstrap
 0.14% :     0.000038s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.08% :     0.000022s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.87% :     0.007099s :      1: jit_opt_a
 0.62% :     0.000164s :      1: jit_opt_after_cconv
 0.26% :     0.000070s :      1: jit_opt_b
 2.09% :     0.000551s :      1: loop_unroll
 3.60% :     0.000952s :      1: mutable_eliminate
 4.03% :     0.001066s :     26: opt.transform.jit_opt_a
 0.19% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000036s :      4: opt.transform.jit_opt_b
 0.06% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.32% :     0.000613s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000050s :      1: py_interpret_to_execute
 0.08% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 2.17% :     0.000574s :      1: renormalize.infer
 1.92% :     0.000506s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000045s :      1: rewriter_after_opt_a
 0.29% :     0.000077s :      1: rewriter_before_opt_a
 0.33% :     0.000086s :      1: symbol_engine_optimizer
29.46% :     0.007784s :      1: task_emit
21.74% :     0.005745s :      1: type_inference
 0.26% :     0.000069s :      1: validate


TotalTime = 0.0181411, [30]
[bootstrap]: 0.00037123
[type_inference]: 0.0080561
[event_method]: 2.157e-05
[auto_monad]: 8.898e-05
[graph_reusing]: 7.33999e-06
[pre_auto_parallel]: 3.01001e-06
[py_interpret_to_execute]: 6.047e-05
[rewriter_before_opt_a]: 8.369e-05
[expand_dump_flag]: 2.68e-06
[jit_opt_a]: 0.00663986, [2]
    [Cycle 1]: 0.00225957, [27]
        [switch_simplify]: 0.0001544
        [loop_unroll]: 3.576e-05
        [a_1]: 0.00087743
        [with_stream_mark]: 2.136e-05
        [recompute_prepare]: 8.80999e-06
        [updatestate_depend_eliminate]: 3.88001e-06
        [updatestate_assign_eliminate]: 3.04001e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 2.15002e-06
        [specialize_transform]: 6.93e-06
        [updatestate_useless_node_eliminater]: 5.01002e-06
        [accelerated_algorithm]: 5.95002e-06
        [meta_shard_fg_expand]: 2.78e-06
        [get_grad_eliminate_]: 5.66e-06
        [merge_forward]: 3.80998e-06
        [cell_reuse_recompute_pass]: 1.56002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.593e-05
        [j_node_and_user_rematch]: 1.039e-05
        [meta_fg_expand]: 2.21998e-06
        [replace_old_param]: 9.59e-06
        [inline_without_move]: 5.54998e-06
        [renormalize]: 0.00079506
        [add_forward_monad_depend]: 6.93e-06
        [auto_monad_grad]: 2.54999e-06
        [auto_monad_eliminator]: 1.647e-05
        [cse]: 3.155e-05
        [replace_applicator]: 1.677e-05
    [Cycle 2]: 0.00034502, [27]
        [switch_simplify]: 6.29001e-06
        [loop_unroll]: 5.04e-06
        [a_1]: 9.454e-05
        [with_stream_mark]: 1.338e-05
        [recompute_prepare]: 5.82001e-06
        [updatestate_depend_eliminate]: 3.38e-06
        [updatestate_assign_eliminate]: 2.49999e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.35001e-06
        [specialize_transform]: 5.09e-06
        [updatestate_useless_node_eliminater]: 5.32999e-06
        [accelerated_algorithm]: 5.12e-06
        [meta_shard_fg_expand]: 2.07001e-06
        [get_grad_eliminate_]: 4.79998e-06
        [merge_forward]: 4.02e-06
        [cell_reuse_recompute_pass]: 2.63e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.639e-05
        [j_node_and_user_rematch]: 8.3e-06
        [meta_fg_expand]: 1.91998e-06
        [replace_old_param]: 8.15e-06
        [inline_without_move]: 4.80001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.40001e-06
        [auto_monad_grad]: 8.90024e-07
        [auto_monad_eliminator]: 5.51e-06
        [cse]: 1.153e-05
        [replace_applicator]: 5.72001e-06
[py_interpret_to_execute_after_opt_a]: 1.708e-05
[rewriter_after_opt_a]: 3.989e-05
[convert_after_rewriter]: 7.51999e-06
[order_py_execute_after_rewriter]: 5.12999e-06
[mutable_eliminate]: 0.00084098
[jit_opt_b]: 7.378e-05, [1]
    [Cycle 1]: 4.674e-05, [2]
        [frontend_op_eliminate]: 1.809e-05
        [inline_after_opt_a]: 1.562e-05
[cconv]: 3.109e-05
[loop_unroll]: 0.0005524
[jit_opt_after_cconv]: 0.00016814, [1]
    [Cycle 1]: 0.00015805, [11]
        [c_1]: 2.425e-05
        [parameter_eliminate]: 3.41001e-06
        [updatestate_depend_eliminate]: 6.41e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [cse]: 2.865e-05
        [call_graph_tuple_transform]: 2.428e-05
        [tuple_list_get_item_eliminator]: 5.84e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 7.50006e-07
        [switch_simplify]: 6.17999e-06
[remove_dup_value]: 1.655e-05
[partial_unused_args_eliminate]: 2.71999e-06
[environ_conv]: 6.49999e-06
[add_recomputation]: 5.881e-05
[cse_after_recomputation]: 2.359e-05, [1]
    [Cycle 1]: 1.692e-05, [1]
        [cse]: 1.075e-05
[auto_monad_reorder]: 1.665e-05
[get_jit_bprop_graph]: 2.57001e-06
[rewriter_after_jit_bprop_graph]: 1.155e-05
[opt_after_jit_grad]: 0.00060631
[symbol_engine_optimizer]: 8.354e-05, [1]
    [Cycle 1]: 7.572e-05, [6]
        [build]: 5.44e-06
        [elim_shapecalc]: 8.61002e-06
        [elim_not_effective]: 1.627e-05
        [opt_reshape]: 6.14999e-06
        [fold_const_symbol]: 8.89e-06
        [renormalize]: 6.80011e-07
[validate]: 4.53e-05
Sums
    bootstrap                            :     0.000371s :  2.76%
    type_inference                       :     0.008056s : 59.96%
    event_method                         :     0.000022s :  0.16%
    auto_monad                           :     0.000089s :  0.66%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000060s :  0.45%
    rewriter_before_opt_a                :     0.000084s :  0.62%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000161s :  1.20%
    jit_opt_a.loop_unroll                :     0.000041s :  0.30%
    jit_opt_a.a_1                        :     0.000972s :  7.23%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000008s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.13%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000795s :  5.92%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.16%
    jit_opt_a.cse                        :     0.000043s :  0.32%
    jit_opt_a.replace_applicator         :     0.000022s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.13%
    rewriter_after_opt_a                 :     0.000040s :  0.30%
    convert_after_rewriter               :     0.000008s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000841s :  6.26%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.12%
    cconv                                :     0.000031s :  0.23%
    loop_unroll                          :     0.000552s :  4.11%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000029s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000017s :  0.12%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000059s :  0.44%
    cse_after_recomputation.cse          :     0.000011s :  0.08%
    auto_monad_reorder                   :     0.000017s :  0.12%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000012s :  0.09%
    opt_after_jit_grad                   :     0.000606s :  4.51%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000045s :  0.34%


Time group info:
------[substitution.]   0.000433    26
 0.51% :     0.000002s :      2: substitution.elim_not_effective
 0.41% :     0.000002s :      2: substitution.fold_const_symbol
 1.50% :     0.000007s :      3: substitution.graph_param_transform
88.66% :     0.000384s :      6: substitution.inline
 0.88% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.31% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.09% :     0.000005s :      2: substitution.replace_old_param
 2.44% :     0.000011s :      1: substitution.switch_simplify
 3.21% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007976     2
84.58% :     0.006746s :      1: type_inference.infer
15.42% :     0.001230s :      1: type_inference.specialize
------[replace.]   0.000172     9
49.39% :     0.000085s :      6: replace.inline
29.24% :     0.000050s :      1: replace.switch_simplify
21.37% :     0.000037s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000403     9
94.51% :     0.000380s :      6: match.inline
 2.37% :     0.000010s :      1: match.switch_simplify
 3.12% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000148   801
 1.30% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.30% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     12: predicate.addn_check_dump
 1.34% :     0.000002s :     12: predicate.addn_zero_filter
 2.29% :     0.000003s :     12: predicate.arithmetic_simplify
 1.40% :     0.000002s :     12: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     12: predicate.compare_switch_simplify
 1.51% :     0.000002s :     12: predicate.depend_value_elim
 1.16% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     12: predicate.dict_set_item_eliminator
 1.08% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.04% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.01% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     12: predicate.environ_get_eliminate
 1.05% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000002s :      6: predicate.get_grad_eliminate
 0.34% :     0.000001s :      3: predicate.graph_param_transform
 5.88% :     0.000009s :     26: predicate.inline
 0.72% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000002s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     17: predicate.load_eliminater
 1.33% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.82% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.04% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000002s :     12: predicate.merge_addn
 1.07% :     0.000002s :     12: predicate.minmaximum_grad
 1.85% :     0.000003s :      3: predicate.mutable_eliminate
 0.36% :     0.000001s :      3: predicate.opt_reshape
 2.37% :     0.000004s :     17: predicate.partial_eliminate
 1.10% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     12: predicate.reduce_eliminate
 1.29% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.25% :     0.000003s :     20: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     12: predicate.reshape_eliminate
 1.11% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     12: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000002s :      6: predicate.special_op_eliminate
 0.79% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.66% :     0.000004s :     20: predicate.switch_defer_inline
 2.17% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.98% :     0.000013s :     65: predicate.switch_simplify
 1.19% :     0.000002s :     12: predicate.tile_eliminate
 1.15% :     0.000002s :     12: predicate.transpose_eliminate
 1.64% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.44% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 2.14% :     0.000003s :     12: predicate.tuple_list_set_item_eliminator
 1.32% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.47% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     12: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001039    15
43.36% :     0.000450s :      6: func_graph_cloner_run.FuncGraphClonerGraph
19.90% :     0.000207s :      1: func_graph_cloner_run.FuncGraphClonerNode
36.74% :     0.000382s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020293    72
 0.31% :     0.000063s :      1: add_recomputation
 0.46% :     0.000093s :      1: auto_monad
 0.10% :     0.000019s :      1: auto_monad_reorder
 1.92% :     0.000389s :      1: bootstrap
 0.17% :     0.000034s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.13% :     0.000026s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.13% :     0.000025s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000010s :      1: graph_reusing
32.74% :     0.006644s :      1: jit_opt_a
 0.84% :     0.000171s :      1: jit_opt_after_cconv
 0.38% :     0.000077s :      1: jit_opt_b
 2.77% :     0.000562s :      1: loop_unroll
 4.20% :     0.000852s :      1: mutable_eliminate
 6.40% :     0.001299s :     26: opt.transform.jit_opt_a
 0.28% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000027s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.04% :     0.000618s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000063s :      1: py_interpret_to_execute
 0.10% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000019s :      1: remove_dup_value
 2.03% :     0.000411s :      1: renormalize.infer
 1.85% :     0.000375s :      1: renormalize.specialize
 0.07% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000044s :      1: rewriter_after_opt_a
 0.43% :     0.000087s :      1: rewriter_before_opt_a
 0.43% :     0.000086s :      1: symbol_engine_optimizer
39.78% :     0.008072s :      1: type_inference


TotalTime = 0.0213043, [33]
[bootstrap]: 0.00039757
[type_inference]: 0.00497032
[event_method]: 2.001e-05
[auto_monad]: 6.843e-05
[graph_reusing]: 5.61e-06
[pre_auto_parallel]: 2.78998e-06
[py_interpret_to_execute]: 3.782e-05
[rewriter_before_opt_a]: 7.35e-05
[expand_dump_flag]: 3.24001e-06
[jit_opt_a]: 0.00620435, [2]
    [Cycle 1]: 0.00175629, [27]
        [switch_simplify]: 5.922e-05
        [loop_unroll]: 3.077e-05
        [a_1]: 0.00056584
        [with_stream_mark]: 2.349e-05
        [recompute_prepare]: 7.27997e-06
        [updatestate_depend_eliminate]: 4.51002e-06
        [updatestate_assign_eliminate]: 3.19001e-06
        [updatestate_loads_eliminate]: 2.98e-06
        [parameter_eliminate]: 1.92001e-06
        [specialize_transform]: 6.10002e-06
        [updatestate_useless_node_eliminater]: 5.09998e-06
        [accelerated_algorithm]: 5.27999e-06
        [meta_shard_fg_expand]: 2.09e-06
        [get_grad_eliminate_]: 5.05001e-06
        [merge_forward]: 4.04002e-06
        [cell_reuse_recompute_pass]: 1.33002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.442e-05
        [j_node_and_user_rematch]: 9.71998e-06
        [meta_fg_expand]: 2.31998e-06
        [replace_old_param]: 9.52001e-06
        [inline_without_move]: 5.27999e-06
        [renormalize]: 0.00073294
        [add_forward_monad_depend]: 6.25002e-06
        [auto_monad_grad]: 2.66999e-06
        [auto_monad_eliminator]: 1.544e-05
        [cse]: 3.163e-05
        [replace_applicator]: 1.262e-05
    [Cycle 2]: 0.00037518, [27]
        [switch_simplify]: 7.38e-06
        [loop_unroll]: 5.01997e-06
        [a_1]: 0.00012018
        [with_stream_mark]: 1.142e-05
        [recompute_prepare]: 5.69e-06
        [updatestate_depend_eliminate]: 3.70998e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [parameter_eliminate]: 1.30999e-06
        [specialize_transform]: 4.98001e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 4.99e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 4.60999e-06
        [merge_forward]: 2.88e-06
        [cell_reuse_recompute_pass]: 2.96999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.484e-05
        [j_node_and_user_rematch]: 8.69e-06
        [meta_fg_expand]: 1.92999e-06
        [replace_old_param]: 8.16002e-06
        [inline_without_move]: 4.94998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.80001e-06
        [auto_monad_grad]: 1.39e-06
        [auto_monad_eliminator]: 6.07999e-06
        [cse]: 1.455e-05
        [replace_applicator]: 5.88002e-06
[py_interpret_to_execute_after_opt_a]: 1.305e-05
[rewriter_after_opt_a]: 3.895e-05
[convert_after_rewriter]: 8.60001e-06
[order_py_execute_after_rewriter]: 5.56e-06
[mutable_eliminate]: 0.0006932
[jit_opt_b]: 5.135e-05, [1]
    [Cycle 1]: 4.321e-05, [2]
        [frontend_op_eliminate]: 1.646e-05
        [inline_after_opt_a]: 1.435e-05
[cconv]: 3.148e-05
[loop_unroll]: 0.00050731
[jit_opt_after_cconv]: 0.00015845, [1]
    [Cycle 1]: 0.00015004, [11]
        [c_1]: 2.152e-05
        [parameter_eliminate]: 3.58e-06
        [updatestate_depend_eliminate]: 8.04002e-06
        [updatestate_assign_eliminate]: 2.69999e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [cse]: 2.528e-05
        [call_graph_tuple_transform]: 2.064e-05
        [tuple_list_get_item_eliminator]: 5.55001e-06
        [none_parameter_eliminate]: 1.90001e-06
        [renormalize]: 1.05001e-06
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 1.343e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 5.89e-06
[add_recomputation]: 5.373e-05
[cse_after_recomputation]: 2.298e-05, [1]
    [Cycle 1]: 1.696e-05, [1]
        [cse]: 1.027e-05
[auto_monad_reorder]: 1.638e-05
[get_jit_bprop_graph]: 1.96003e-06
[rewriter_after_jit_bprop_graph]: 7.56999e-06
[opt_after_jit_grad]: 0.00050142
[symbol_engine_optimizer]: 7.4e-05, [1]
    [Cycle 1]: 6.685e-05, [6]
        [build]: 3.91999e-06
        [elim_shapecalc]: 7.18998e-06
        [elim_not_effective]: 1.384e-05
        [opt_reshape]: 5.54e-06
        [fold_const_symbol]: 8.82e-06
        [renormalize]: 6.60017e-07
[validate]: 3.958e-05
[backend_pass]: 8.79983e-07
[task_emit]: 0.00704077
[execute]: 1.014e-05
Sums
    bootstrap                            :     0.000398s :  2.40%
    type_inference                       :     0.004970s : 29.99%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000068s :  0.41%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.23%
    rewriter_before_opt_a                :     0.000073s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.40%
    jit_opt_a.loop_unroll                :     0.000036s :  0.22%
    jit_opt_a.a_1                        :     0.000686s :  4.14%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000733s :  4.42%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.13%
    jit_opt_a.cse                        :     0.000046s :  0.28%
    jit_opt_a.replace_applicator         :     0.000019s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000039s :  0.23%
    convert_after_rewriter               :     0.000009s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000693s :  4.18%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000031s :  0.19%
    loop_unroll                          :     0.000507s :  3.06%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000054s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000501s :  3.03%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007041s : 42.48%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000237    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.03% :     0.000005s :      3: substitution.graph_param_transform
85.44% :     0.000202s :      4: substitution.inline
 1.58% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.88% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.18% :     0.000005s :      2: substitution.replace_old_param
 5.40% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004894     2
82.52% :     0.004039s :      1: type_inference.infer
17.48% :     0.000855s :      1: type_inference.specialize
------[replace.]   0.000069     6
67.99% :     0.000047s :      4: replace.inline
32.01% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000212     6
94.48% :     0.000200s :      4: match.inline
 5.52% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.18% :     0.000001s :     10: predicate.addn_zero_filter
 3.01% :     0.000004s :     10: predicate.arithmetic_simplify
 1.19% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.18% :     0.000001s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.37% :     0.000007s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.93% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.14% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 2.37% :     0.000003s :      3: predicate.mutable_eliminate
 0.62% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.32% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.97% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.06% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.25% :     0.000010s :     53: predicate.switch_simplify
 1.29% :     0.000002s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.00% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.37% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.86% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000635    12
50.53% :     0.000321s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.01% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.46% :     0.000263s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023053    76
 0.25% :     0.000057s :      1: add_recomputation
 0.31% :     0.000072s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.81% :     0.000417s :      1: bootstrap
 0.15% :     0.000034s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000024s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.93% :     0.006208s :      1: jit_opt_a
 0.70% :     0.000161s :      1: jit_opt_after_cconv
 0.23% :     0.000054s :      1: jit_opt_b
 2.25% :     0.000518s :      1: loop_unroll
 3.05% :     0.000704s :      1: mutable_eliminate
 3.93% :     0.000906s :     26: opt.transform.jit_opt_a
 0.22% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.22% :     0.000511s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000041s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.74% :     0.000401s :      1: renormalize.infer
 1.40% :     0.000322s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000042s :      1: rewriter_after_opt_a
 0.33% :     0.000077s :      1: rewriter_before_opt_a
 0.33% :     0.000076s :      1: symbol_engine_optimizer
30.62% :     0.007058s :      1: task_emit
21.64% :     0.004989s :      1: type_inference
 0.27% :     0.000062s :      1: validate


TotalTime = 0.015517, [30]
[bootstrap]: 0.0003264
[type_inference]: 0.00720549
[event_method]: 2.074e-05
[auto_monad]: 8.288e-05
[graph_reusing]: 6.09001e-06
[pre_auto_parallel]: 2.68998e-06
[py_interpret_to_execute]: 5.005e-05
[rewriter_before_opt_a]: 8.631e-05
[expand_dump_flag]: 3.04001e-06
[jit_opt_a]: 0.00551788, [2]
    [Cycle 1]: 0.00188194, [27]
        [switch_simplify]: 0.00011648
        [loop_unroll]: 4.088e-05
        [a_1]: 0.00075014
        [with_stream_mark]: 1.772e-05
        [recompute_prepare]: 8.50999e-06
        [updatestate_depend_eliminate]: 4.23001e-06
        [updatestate_assign_eliminate]: 3.72998e-06
        [updatestate_loads_eliminate]: 3.01001e-06
        [parameter_eliminate]: 2.18002e-06
        [specialize_transform]: 7.45998e-06
        [updatestate_useless_node_eliminater]: 6.07001e-06
        [accelerated_algorithm]: 6.35002e-06
        [meta_shard_fg_expand]: 2.27999e-06
        [get_grad_eliminate_]: 6.16e-06
        [merge_forward]: 3.58999e-06
        [cell_reuse_recompute_pass]: 1.52001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.668e-05
        [j_node_and_user_rematch]: 9.67001e-06
        [meta_fg_expand]: 2.51e-06
        [replace_old_param]: 9.36998e-06
        [inline_without_move]: 5.40001e-06
        [renormalize]: 0.00061397
        [add_forward_monad_depend]: 4.89e-06
        [auto_monad_grad]: 2.22001e-06
        [auto_monad_eliminator]: 1.388e-05
        [cse]: 2.874e-05
        [replace_applicator]: 1.086e-05
    [Cycle 2]: 0.00032333, [27]
        [switch_simplify]: 5.94e-06
        [loop_unroll]: 5.15999e-06
        [a_1]: 9.067e-05
        [with_stream_mark]: 8.20999e-06
        [recompute_prepare]: 5.48002e-06
        [updatestate_depend_eliminate]: 2.56e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 5.54e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 4.84e-06
        [meta_shard_fg_expand]: 1.14998e-06
        [get_grad_eliminate_]: 4.72e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.34998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.238e-05
        [j_node_and_user_rematch]: 8.07e-06
        [meta_fg_expand]: 1.69998e-06
        [replace_old_param]: 8.08999e-06
        [inline_without_move]: 5.00001e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.47999e-06
        [auto_monad_grad]: 8.10018e-07
        [auto_monad_eliminator]: 5.34e-06
        [cse]: 9.95002e-06
        [replace_applicator]: 5.88998e-06
[py_interpret_to_execute_after_opt_a]: 1.019e-05
[rewriter_after_opt_a]: 3.261e-05
[convert_after_rewriter]: 6.92002e-06
[order_py_execute_after_rewriter]: 4.86002e-06
[mutable_eliminate]: 0.00052853
[jit_opt_b]: 5.379e-05, [1]
    [Cycle 1]: 4.663e-05, [2]
        [frontend_op_eliminate]: 1.946e-05
        [inline_after_opt_a]: 1.543e-05
[cconv]: 3.437e-05
[loop_unroll]: 0.00047931
[jit_opt_after_cconv]: 0.0001629, [1]
    [Cycle 1]: 0.0001556, [11]
        [c_1]: 2.513e-05
        [parameter_eliminate]: 2.64999e-06
        [updatestate_depend_eliminate]: 6.43998e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [cse]: 1.877e-05
        [call_graph_tuple_transform]: 2.196e-05
        [tuple_list_get_item_eliminator]: 6.78e-06
        [none_parameter_eliminate]: 1.86e-06
        [renormalize]: 5.60016e-07
        [switch_simplify]: 7.01001e-06
[remove_dup_value]: 1.606e-05
[partial_unused_args_eliminate]: 3.08e-06
[environ_conv]: 5.47001e-06
[add_recomputation]: 4.861e-05
[cse_after_recomputation]: 2.221e-05, [1]
    [Cycle 1]: 1.639e-05, [1]
        [cse]: 1.053e-05
[auto_monad_reorder]: 1.544e-05
[get_jit_bprop_graph]: 2.02999e-06
[rewriter_after_jit_bprop_graph]: 5.19e-06
[opt_after_jit_grad]: 0.00049192
[symbol_engine_optimizer]: 7.412e-05, [1]
    [Cycle 1]: 6.81e-05, [6]
        [build]: 3.8e-06
        [elim_shapecalc]: 7.83001e-06
        [elim_not_effective]: 1.324e-05
        [opt_reshape]: 5.90002e-06
        [fold_const_symbol]: 8.67998e-06
        [renormalize]: 4.19997e-07
[validate]: 3.283e-05
Sums
    bootstrap                            :     0.000326s :  2.82%
    type_inference                       :     0.007205s : 62.18%
    event_method                         :     0.000021s :  0.18%
    auto_monad                           :     0.000083s :  0.72%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000050s :  0.43%
    rewriter_before_opt_a                :     0.000086s :  0.74%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000122s :  1.06%
    jit_opt_a.loop_unroll                :     0.000046s :  0.40%
    jit_opt_a.a_1                        :     0.000841s :  7.26%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000614s :  5.30%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.17%
    jit_opt_a.cse                        :     0.000039s :  0.33%
    jit_opt_a.replace_applicator         :     0.000017s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.09%
    rewriter_after_opt_a                 :     0.000033s :  0.28%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000529s :  4.56%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.13%
    cconv                                :     0.000034s :  0.30%
    loop_unroll                          :     0.000479s :  4.14%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.06%
    remove_dup_value                     :     0.000016s :  0.14%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000049s :  0.42%
    cse_after_recomputation.cse          :     0.000011s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000492s :  4.24%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.28%


Time group info:
------[substitution.]   0.000273    26
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.14% :     0.000006s :      3: substitution.graph_param_transform
84.03% :     0.000229s :      6: substitution.inline
 1.04% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.83% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.56% :     0.000004s :      2: substitution.replace_old_param
 3.13% :     0.000009s :      1: substitution.switch_simplify
 5.05% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007136     2
85.38% :     0.006093s :      1: type_inference.infer
14.62% :     0.001043s :      1: type_inference.specialize
------[replace.]   0.000139     9
50.62% :     0.000070s :      6: replace.inline
24.77% :     0.000034s :      1: replace.switch_simplify
24.61% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000245     9
91.88% :     0.000225s :      6: match.inline
 3.10% :     0.000008s :      1: match.switch_simplify
 5.02% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000149   801
 1.42% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.04% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000002s :     12: predicate.addn_check_dump
 1.38% :     0.000002s :     12: predicate.addn_zero_filter
 2.02% :     0.000003s :     12: predicate.arithmetic_simplify
 1.34% :     0.000002s :     12: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.18% :     0.000002s :     12: predicate.compare_switch_simplify
 1.41% :     0.000002s :     12: predicate.depend_value_elim
 1.24% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.70% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.75% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     12: predicate.environ_get_eliminate
 1.22% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.19% :     0.000000s :      3: predicate.graph_param_transform
 5.39% :     0.000008s :     26: predicate.inline
 0.77% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.91% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.91% :     0.000003s :     17: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.50% :     0.000007s :     40: predicate.loop_unroll_before_grad
 1.87% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.26% :     0.000002s :     12: predicate.merge_addn
 1.14% :     0.000002s :     12: predicate.minmaximum_grad
 1.78% :     0.000003s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     17: predicate.partial_eliminate
 1.18% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.89% :     0.000003s :     12: predicate.reduce_eliminate
 1.46% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     20: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     12: predicate.reshape_eliminate
 1.28% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.79% :     0.000004s :     20: predicate.switch_defer_inline
 2.55% :     0.000004s :     20: predicate.switch_layer_defer_inline
 8.16% :     0.000012s :     65: predicate.switch_simplify
 1.26% :     0.000002s :     12: predicate.tile_eliminate
 1.22% :     0.000002s :     12: predicate.transpose_eliminate
 1.45% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.76% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.78% :     0.000003s :     17: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     12: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000821    15
53.97% :     0.000443s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.02% :     0.000082s :      1: func_graph_cloner_run.FuncGraphClonerNode
36.00% :     0.000296s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017332    72
 0.30% :     0.000051s :      1: add_recomputation
 0.50% :     0.000087s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 2.01% :     0.000348s :      1: bootstrap
 0.22% :     0.000037s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
31.86% :     0.005521s :      1: jit_opt_a
 0.96% :     0.000166s :      1: jit_opt_after_cconv
 0.33% :     0.000057s :      1: jit_opt_b
 2.81% :     0.000488s :      1: loop_unroll
 3.10% :     0.000538s :      1: mutable_eliminate
 6.53% :     0.001131s :     26: opt.transform.jit_opt_a
 0.33% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000029s :      4: opt.transform.jit_opt_b
 0.08% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.89% :     0.000500s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000053s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000019s :      1: remove_dup_value
 1.89% :     0.000327s :      1: renormalize.infer
 1.61% :     0.000279s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000036s :      1: rewriter_after_opt_a
 0.52% :     0.000089s :      1: rewriter_before_opt_a
 0.44% :     0.000077s :      1: symbol_engine_optimizer
41.67% :     0.007221s :      1: type_inference


TotalTime = 0.0212239, [33]
[bootstrap]: 0.00032472
[type_inference]: 0.0049117
[event_method]: 1.855e-05
[auto_monad]: 6.945e-05
[graph_reusing]: 6.51e-06
[pre_auto_parallel]: 2.79001e-06
[py_interpret_to_execute]: 4.035e-05
[rewriter_before_opt_a]: 8.166e-05
[expand_dump_flag]: 3.32997e-06
[jit_opt_a]: 0.00662426, [2]
    [Cycle 1]: 0.00209249, [27]
        [switch_simplify]: 7.124e-05
        [loop_unroll]: 3.074e-05
        [a_1]: 0.00061898
        [with_stream_mark]: 2.635e-05
        [recompute_prepare]: 7.98999e-06
        [updatestate_depend_eliminate]: 4.82e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.94999e-06
        [parameter_eliminate]: 2.27999e-06
        [specialize_transform]: 6.07999e-06
        [updatestate_useless_node_eliminater]: 5.40999e-06
        [accelerated_algorithm]: 6.19999e-06
        [meta_shard_fg_expand]: 2.49001e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 4.03001e-06
        [cell_reuse_recompute_pass]: 1.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.548e-05
        [j_node_and_user_rematch]: 1.005e-05
        [meta_fg_expand]: 2.34999e-06
        [replace_old_param]: 1.017e-05
        [inline_without_move]: 5.99e-06
        [renormalize]: 0.00098903
        [add_forward_monad_depend]: 6.12999e-06
        [auto_monad_grad]: 2.34999e-06
        [auto_monad_eliminator]: 1.556e-05
        [cse]: 3.225e-05
        [replace_applicator]: 1.374e-05
    [Cycle 2]: 0.00034144, [27]
        [switch_simplify]: 5.61998e-06
        [loop_unroll]: 6.42001e-06
        [a_1]: 9.925e-05
        [with_stream_mark]: 1.068e-05
        [recompute_prepare]: 5.19003e-06
        [updatestate_depend_eliminate]: 3.04999e-06
        [updatestate_assign_eliminate]: 2.23002e-06
        [updatestate_loads_eliminate]: 2.06003e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 5.25001e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.17e-06
        [meta_shard_fg_expand]: 1.47999e-06
        [get_grad_eliminate_]: 4.67e-06
        [merge_forward]: 2.70997e-06
        [cell_reuse_recompute_pass]: 2.01e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.367e-05
        [j_node_and_user_rematch]: 7.82e-06
        [meta_fg_expand]: 1.34998e-06
        [replace_old_param]: 8.24998e-06
        [inline_without_move]: 4.57998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 8.60018e-07
        [auto_monad_eliminator]: 5.21002e-06
        [cse]: 1.195e-05
        [replace_applicator]: 4.85001e-06
[py_interpret_to_execute_after_opt_a]: 1.265e-05
[rewriter_after_opt_a]: 4.093e-05
[convert_after_rewriter]: 7.25e-06
[order_py_execute_after_rewriter]: 4.77998e-06
[mutable_eliminate]: 0.00073087
[jit_opt_b]: 5.326e-05, [1]
    [Cycle 1]: 4.561e-05, [2]
        [frontend_op_eliminate]: 1.876e-05
        [inline_after_opt_a]: 1.519e-05
[cconv]: 2.848e-05
[loop_unroll]: 0.00043752
[jit_opt_after_cconv]: 0.00014611, [1]
    [Cycle 1]: 0.00013971, [11]
        [c_1]: 2.165e-05
        [parameter_eliminate]: 2.76e-06
        [updatestate_depend_eliminate]: 6.32001e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [cse]: 2.114e-05
        [call_graph_tuple_transform]: 2.016e-05
        [tuple_list_get_item_eliminator]: 5.39998e-06
        [none_parameter_eliminate]: 1.38002e-06
        [renormalize]: 8.50006e-07
        [switch_simplify]: 5.32999e-06
[remove_dup_value]: 1.516e-05
[partial_unused_args_eliminate]: 2.32001e-06
[environ_conv]: 5.81998e-06
[add_recomputation]: 5.381e-05
[cse_after_recomputation]: 2.098e-05, [1]
    [Cycle 1]: 1.495e-05, [1]
        [cse]: 9.06002e-06
[auto_monad_reorder]: 1.646e-05
[get_jit_bprop_graph]: 2.63e-06
[rewriter_after_jit_bprop_graph]: 7.26999e-06
[opt_after_jit_grad]: 0.00046292
[symbol_engine_optimizer]: 7.505e-05, [1]
    [Cycle 1]: 6.782e-05, [6]
        [build]: 3.82002e-06
        [elim_shapecalc]: 8.33999e-06
        [elim_not_effective]: 1.313e-05
        [opt_reshape]: 5.46998e-06
        [fold_const_symbol]: 9.18002e-06
        [renormalize]: 7.60017e-07
[validate]: 4.22e-05
[backend_pass]: 1.00001e-06
[task_emit]: 0.00673578
[execute]: 1.02e-05
Sums
    bootstrap                            :     0.000325s :  1.98%
    type_inference                       :     0.004912s : 30.00%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000069s :  0.42%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.25%
    rewriter_before_opt_a                :     0.000082s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000077s :  0.47%
    jit_opt_a.loop_unroll                :     0.000037s :  0.23%
    jit_opt_a.a_1                        :     0.000718s :  4.39%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000989s :  6.04%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.13%
    jit_opt_a.cse                        :     0.000044s :  0.27%
    jit_opt_a.replace_applicator         :     0.000019s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000041s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000731s :  4.46%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000028s :  0.17%
    loop_unroll                          :     0.000438s :  2.67%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000054s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000463s :  2.83%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006736s : 41.14%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000271    23
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000002s :      2: substitution.fold_const_symbol
 2.13% :     0.000006s :      3: substitution.graph_param_transform
85.94% :     0.000233s :      4: substitution.inline
 1.38% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.93% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.19% :     0.000006s :      2: substitution.replace_old_param
 4.95% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004840     2
80.94% :     0.003917s :      1: type_inference.infer
19.06% :     0.000922s :      1: type_inference.specialize
------[replace.]   0.000077     6
66.66% :     0.000051s :      4: replace.inline
33.34% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000242     6
94.95% :     0.000230s :      4: match.inline
 5.05% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.75% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 2.01% :     0.000002s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.66% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.61% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.44% :     0.000002s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 6.13% :     0.000008s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.33% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.86% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.69% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.53% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.00% :     0.000002s :     15: predicate.partial_eliminate
 1.42% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.58% :     0.000002s :     10: predicate.reshape_eliminate
 1.51% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.05% :     0.000010s :     53: predicate.switch_simplify
 1.48% :     0.000002s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.23% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.38% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.67% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.88% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000689    12
46.91% :     0.000323s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.01% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.08% :     0.000311s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023263    76
 0.24% :     0.000057s :      1: add_recomputation
 0.31% :     0.000073s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.48% :     0.000345s :      1: bootstrap
 0.13% :     0.000031s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
28.49% :     0.006628s :      1: jit_opt_a
 0.64% :     0.000149s :      1: jit_opt_after_cconv
 0.24% :     0.000056s :      1: jit_opt_b
 1.91% :     0.000444s :      1: loop_unroll
 3.19% :     0.000741s :      1: mutable_eliminate
 4.09% :     0.000953s :     26: opt.transform.jit_opt_a
 0.21% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.03% :     0.000471s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000043s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 2.49% :     0.000579s :      1: renormalize.infer
 1.72% :     0.000401s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000044s :      1: rewriter_after_opt_a
 0.36% :     0.000085s :      1: rewriter_before_opt_a
 0.33% :     0.000078s :      1: symbol_engine_optimizer
29.04% :     0.006755s :      1: task_emit
21.18% :     0.004926s :      1: type_inference
 0.28% :     0.000064s :      1: validate


TotalTime = 0.0142731, [30]
[bootstrap]: 0.00027457
[type_inference]: 0.00613698
[event_method]: 2.101e-05
[auto_monad]: 8.361e-05
[graph_reusing]: 7.67998e-06
[pre_auto_parallel]: 3.13998e-06
[py_interpret_to_execute]: 5.005e-05
[rewriter_before_opt_a]: 8.985e-05
[expand_dump_flag]: 3.65e-06
[jit_opt_a]: 0.00555596, [2]
    [Cycle 1]: 0.0018163, [27]
        [switch_simplify]: 0.00011383
        [loop_unroll]: 3.564e-05
        [a_1]: 0.00067983
        [with_stream_mark]: 1.759e-05
        [recompute_prepare]: 7.32002e-06
        [updatestate_depend_eliminate]: 3.55e-06
        [updatestate_assign_eliminate]: 3.4e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 6.02999e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.86e-06
        [meta_shard_fg_expand]: 2.14999e-06
        [get_grad_eliminate_]: 5.24e-06
        [merge_forward]: 4.18001e-06
        [cell_reuse_recompute_pass]: 9.40025e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.473e-05
        [j_node_and_user_rematch]: 9.74e-06
        [meta_fg_expand]: 2.48e-06
        [replace_old_param]: 9.74e-06
        [inline_without_move]: 5.61e-06
        [renormalize]: 0.00064086
        [add_forward_monad_depend]: 5.17e-06
        [auto_monad_grad]: 2.37001e-06
        [auto_monad_eliminator]: 1.485e-05
        [cse]: 2.947e-05
        [replace_applicator]: 1.179e-05
    [Cycle 2]: 0.00031655, [27]
        [switch_simplify]: 5.59e-06
        [loop_unroll]: 5.04e-06
        [a_1]: 8.995e-05
        [with_stream_mark]: 9.91e-06
        [recompute_prepare]: 5.01002e-06
        [updatestate_depend_eliminate]: 2.62001e-06
        [updatestate_assign_eliminate]: 2.15002e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 5.26998e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 5.07e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.63999e-06
        [merge_forward]: 2.58e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.263e-05
        [j_node_and_user_rematch]: 8.39002e-06
        [meta_fg_expand]: 1.81e-06
        [replace_old_param]: 7.10998e-06
        [inline_without_move]: 4.83001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 4.74e-06
        [cse]: 9.32001e-06
        [replace_applicator]: 5.01997e-06
[py_interpret_to_execute_after_opt_a]: 1.034e-05
[rewriter_after_opt_a]: 3.414e-05
[convert_after_rewriter]: 6.88e-06
[order_py_execute_after_rewriter]: 5.05999e-06
[mutable_eliminate]: 0.00049109
[jit_opt_b]: 4.773e-05, [1]
    [Cycle 1]: 4.139e-05, [2]
        [frontend_op_eliminate]: 1.565e-05
        [inline_after_opt_a]: 1.447e-05
[cconv]: 2.456e-05
[loop_unroll]: 0.000422
[jit_opt_after_cconv]: 0.00013893, [1]
    [Cycle 1]: 0.0001324, [11]
        [c_1]: 2.249e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 5.36002e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [cse]: 1.591e-05
        [call_graph_tuple_transform]: 1.848e-05
        [tuple_list_get_item_eliminator]: 5.30999e-06
        [none_parameter_eliminate]: 1.82001e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.46998e-06
[remove_dup_value]: 1.44e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 5.53002e-06
[add_recomputation]: 4.864e-05
[cse_after_recomputation]: 2.034e-05, [1]
    [Cycle 1]: 1.49e-05, [1]
        [cse]: 9.05999e-06
[auto_monad_reorder]: 1.489e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 5.40999e-06
[opt_after_jit_grad]: 0.00046858
[symbol_engine_optimizer]: 7.253e-05, [1]
    [Cycle 1]: 6.634e-05, [6]
        [build]: 4.10998e-06
        [elim_shapecalc]: 7.31001e-06
        [elim_not_effective]: 1.269e-05
        [opt_reshape]: 5.56e-06
        [fold_const_symbol]: 8.70999e-06
        [renormalize]: 4.59986e-07
[validate]: 3.267e-05
Sums
    bootstrap                            :     0.000275s :  2.67%
    type_inference                       :     0.006137s : 59.78%
    event_method                         :     0.000021s :  0.20%
    auto_monad                           :     0.000084s :  0.81%
    graph_reusing                        :     0.000008s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000050s :  0.49%
    rewriter_before_opt_a                :     0.000090s :  0.88%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000119s :  1.16%
    jit_opt_a.loop_unroll                :     0.000041s :  0.40%
    jit_opt_a.a_1                        :     0.000770s :  7.50%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000641s :  6.24%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.19%
    jit_opt_a.cse                        :     0.000039s :  0.38%
    jit_opt_a.replace_applicator         :     0.000017s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.10%
    rewriter_after_opt_a                 :     0.000034s :  0.33%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000491s :  4.78%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.14%
    cconv                                :     0.000025s :  0.24%
    loop_unroll                          :     0.000422s :  4.11%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000049s :  0.47%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000469s :  4.56%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.32%


Time group info:
------[substitution.]   0.000266    26
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.05% :     0.000005s :      3: substitution.graph_param_transform
84.38% :     0.000225s :      6: substitution.inline
 1.22% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.65% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.47% :     0.000004s :      2: substitution.replace_old_param
 3.08% :     0.000008s :      1: substitution.switch_simplify
 4.86% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006073     2
83.63% :     0.005079s :      1: type_inference.infer
16.37% :     0.000994s :      1: type_inference.specialize
------[replace.]   0.000133     9
51.12% :     0.000068s :      6: replace.inline
25.40% :     0.000034s :      1: replace.switch_simplify
23.48% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000240     9
92.02% :     0.000221s :      6: match.inline
 3.07% :     0.000007s :      1: match.switch_simplify
 4.91% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   801
 1.37% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.13% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     12: predicate.addn_check_dump
 1.44% :     0.000002s :     12: predicate.addn_zero_filter
 2.19% :     0.000003s :     12: predicate.arithmetic_simplify
 1.29% :     0.000002s :     12: predicate.cast_eliminate
 0.38% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     12: predicate.compare_switch_simplify
 1.34% :     0.000002s :     12: predicate.depend_value_elim
 1.16% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.38% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     12: predicate.environ_get_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.63% :     0.000007s :     26: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     17: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.35% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000002s :     12: predicate.merge_addn
 1.13% :     0.000002s :     12: predicate.minmaximum_grad
 1.44% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     17: predicate.partial_eliminate
 1.18% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.82% :     0.000002s :     12: predicate.reduce_eliminate
 1.48% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     20: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     12: predicate.reshape_eliminate
 1.23% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     12: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.72% :     0.000004s :     20: predicate.switch_defer_inline
 2.32% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.35% :     0.000011s :     65: predicate.switch_simplify
 1.39% :     0.000002s :     12: predicate.tile_eliminate
 1.20% :     0.000002s :     12: predicate.transpose_eliminate
 1.63% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.86% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.54% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000776    15
52.27% :     0.000406s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.19% :     0.000087s :      1: func_graph_cloner_run.FuncGraphClonerNode
36.54% :     0.000284s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016006    72
 0.32% :     0.000051s :      1: add_recomputation
 0.54% :     0.000087s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.82% :     0.000291s :      1: bootstrap
 0.17% :     0.000027s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000025s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.07% :     0.000011s :      1: graph_reusing
34.73% :     0.005559s :      1: jit_opt_a
 0.89% :     0.000142s :      1: jit_opt_after_cconv
 0.32% :     0.000051s :      1: jit_opt_b
 2.69% :     0.000431s :      1: loop_unroll
 3.12% :     0.000500s :      1: mutable_eliminate
 6.53% :     0.001045s :     26: opt.transform.jit_opt_a
 0.30% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.98% :     0.000477s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.33% :     0.000053s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.25% :     0.000359s :      1: renormalize.infer
 1.71% :     0.000274s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000037s :      1: rewriter_after_opt_a
 0.58% :     0.000093s :      1: rewriter_before_opt_a
 0.47% :     0.000075s :      1: symbol_engine_optimizer
38.42% :     0.006149s :      1: type_inference


TotalTime = 0.0284424, [33]
[bootstrap]: 0.00028769
[type_inference]: 0.00497197
[event_method]: 1.994e-05
[auto_monad]: 6.875e-05
[graph_reusing]: 5.93002e-06
[pre_auto_parallel]: 2.88e-06
[py_interpret_to_execute]: 4.512e-05
[rewriter_before_opt_a]: 7.58e-05
[expand_dump_flag]: 3.61001e-06
[jit_opt_a]: 0.0144215, [2]
    [Cycle 1]: 0.00172493, [27]
        [switch_simplify]: 5.844e-05
        [loop_unroll]: 3.159e-05
        [a_1]: 0.00055549
        [with_stream_mark]: 2.175e-05
        [recompute_prepare]: 7.15e-06
        [updatestate_depend_eliminate]: 3.75e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.43998e-06
        [parameter_eliminate]: 1.86998e-06
        [specialize_transform]: 5.99999e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 2.30002e-06
        [get_grad_eliminate_]: 5.25001e-06
        [merge_forward]: 3.46999e-06
        [cell_reuse_recompute_pass]: 1.94999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.384e-05
        [j_node_and_user_rematch]: 9.07001e-06
        [meta_fg_expand]: 2.17999e-06
        [replace_old_param]: 8.85001e-06
        [inline_without_move]: 5.30001e-06
        [renormalize]: 0.00072934
        [add_forward_monad_depend]: 5.19e-06
        [auto_monad_grad]: 2.34001e-06
        [auto_monad_eliminator]: 1.451e-05
        [cse]: 3.026e-05
        [replace_applicator]: 1.195e-05
    [Cycle 2]: 0.00032371, [27]
        [switch_simplify]: 6.01998e-06
        [loop_unroll]: 4.96002e-06
        [a_1]: 9.238e-05
        [with_stream_mark]: 1.041e-05
        [recompute_prepare]: 5.20999e-06
        [updatestate_depend_eliminate]: 3.26001e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.12999e-06
        [specialize_transform]: 4.69998e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.32001e-06
        [meta_shard_fg_expand]: 1.35999e-06
        [get_grad_eliminate_]: 4.61002e-06
        [merge_forward]: 2.62001e-06
        [cell_reuse_recompute_pass]: 2.02999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.36e-05
        [j_node_and_user_rematch]: 8.03001e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 8.01001e-06
        [inline_without_move]: 4.84998e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 1.10001e-06
        [auto_monad_eliminator]: 5.30999e-06
        [cse]: 1.127e-05
        [replace_applicator]: 4.87e-06
[py_interpret_to_execute_after_opt_a]: 1.162e-05
[rewriter_after_opt_a]: 3.556e-05
[convert_after_rewriter]: 8.67e-06
[order_py_execute_after_rewriter]: 4.95001e-06
[mutable_eliminate]: 0.00058134
[jit_opt_b]: 5.555e-05, [1]
    [Cycle 1]: 4.862e-05, [2]
        [frontend_op_eliminate]: 2.075e-05
        [inline_after_opt_a]: 1.588e-05
[cconv]: 2.601e-05
[loop_unroll]: 0.00043493
[jit_opt_after_cconv]: 0.00014207, [1]
    [Cycle 1]: 0.00013524, [11]
        [c_1]: 2.094e-05
        [parameter_eliminate]: 2.79999e-06
        [updatestate_depend_eliminate]: 5.30001e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.25002e-06
        [cse]: 1.924e-05
        [call_graph_tuple_transform]: 1.818e-05
        [tuple_list_get_item_eliminator]: 5.93002e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 6.01998e-06
[remove_dup_value]: 1.304e-05
[partial_unused_args_eliminate]: 2.10002e-06
[environ_conv]: 5.13002e-06
[add_recomputation]: 5.105e-05
[cse_after_recomputation]: 2.034e-05, [1]
    [Cycle 1]: 1.462e-05, [1]
        [cse]: 8.85999e-06
[auto_monad_reorder]: 1.537e-05
[get_jit_bprop_graph]: 1.97999e-06
[rewriter_after_jit_bprop_graph]: 6.12999e-06
[opt_after_jit_grad]: 0.00049088
[symbol_engine_optimizer]: 7.416e-05, [1]
    [Cycle 1]: 6.723e-05, [6]
        [build]: 3.52997e-06
        [elim_shapecalc]: 8.11002e-06
        [elim_not_effective]: 1.267e-05
        [opt_reshape]: 6.12999e-06
        [fold_const_symbol]: 8.72e-06
        [renormalize]: 8.2e-07
[validate]: 3.757e-05
[backend_pass]: 1.02e-06
[task_emit]: 0.00630556
[execute]: 8.25e-06
Sums
    bootstrap                            :     0.000288s :  1.86%
    type_inference                       :     0.004972s : 32.17%
    event_method                         :     0.000020s :  0.13%
    auto_monad                           :     0.000069s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000045s :  0.29%
    rewriter_before_opt_a                :     0.000076s :  0.49%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.42%
    jit_opt_a.loop_unroll                :     0.000037s :  0.24%
    jit_opt_a.a_1                        :     0.000648s :  4.19%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000729s :  4.72%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000042s :  0.27%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000036s :  0.23%
    convert_after_rewriter               :     0.000009s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000581s :  3.76%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000026s :  0.17%
    loop_unroll                          :     0.000435s :  2.81%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000051s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000491s :  3.18%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000038s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006306s : 40.80%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000214    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.33% :     0.000005s :      3: substitution.graph_param_transform
84.28% :     0.000181s :      4: substitution.inline
 1.52% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.28% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000004s :      2: substitution.replace_old_param
 5.93% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004890     2
81.20% :     0.003971s :      1: type_inference.infer
18.80% :     0.000919s :      1: type_inference.specialize
------[replace.]   0.000073     6
67.17% :     0.000049s :      4: replace.inline
32.83% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000190     6
93.86% :     0.000178s :      4: match.inline
 6.14% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 2.52% :     0.000003s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.13% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.64% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.74% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.36% :     0.000002s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.61% :     0.000007s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.76% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.35% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 1.79% :     0.000002s :      3: predicate.mutable_eliminate
 0.68% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.91% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.25% :     0.000002s :     10: predicate.reshape_eliminate
 1.32% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.23% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.57% :     0.000009s :     53: predicate.switch_simplify
 1.08% :     0.000001s :     10: predicate.tile_eliminate
 1.50% :     0.000002s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.71% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.81% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000599    12
47.53% :     0.000285s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.11% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.36% :     0.000278s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.030143    76
 0.18% :     0.000054s :      1: add_recomputation
 0.24% :     0.000072s :      1: auto_monad
 0.06% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.00% :     0.000301s :      1: bootstrap
 0.09% :     0.000029s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.07% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000025s :      1: event_method
 0.04% :     0.000012s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
47.86% :     0.014425s :      1: jit_opt_a
 0.48% :     0.000145s :      1: jit_opt_after_cconv
 0.19% :     0.000058s :      1: jit_opt_b
 1.47% :     0.000444s :      1: loop_unroll
 1.96% :     0.000590s :      1: mutable_eliminate
 2.86% :     0.000862s :     26: opt.transform.jit_opt_a
 0.16% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000029s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.66% :     0.000499s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000006s :      1: pre_auto_parallel
 0.16% :     0.000048s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000015s :      1: remove_dup_value
 1.28% :     0.000385s :      1: renormalize.infer
 1.12% :     0.000336s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000039s :      1: rewriter_after_opt_a
 0.26% :     0.000079s :      1: rewriter_before_opt_a
 0.25% :     0.000077s :      1: symbol_engine_optimizer
20.97% :     0.006322s :      1: task_emit
16.56% :     0.004991s :      1: type_inference
 0.19% :     0.000057s :      1: validate


TotalTime = 0.0135924, [30]
[bootstrap]: 0.00025775
[type_inference]: 0.00589518
[event_method]: 2.098e-05
[auto_monad]: 7.918e-05
[graph_reusing]: 6.33998e-06
[pre_auto_parallel]: 3.09999e-06
[py_interpret_to_execute]: 4.76e-05
[rewriter_before_opt_a]: 8.532e-05
[expand_dump_flag]: 3.31001e-06
[jit_opt_a]: 0.00516834, [2]
    [Cycle 1]: 0.00177142, [27]
        [switch_simplify]: 0.0001147
        [loop_unroll]: 3.519e-05
        [a_1]: 0.00067691
        [with_stream_mark]: 1.79e-05
        [recompute_prepare]: 7.46001e-06
        [updatestate_depend_eliminate]: 3.64002e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.51998e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 5.86e-06
        [updatestate_useless_node_eliminater]: 5.08002e-06
        [accelerated_algorithm]: 5.45001e-06
        [meta_shard_fg_expand]: 2.11e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 3.92998e-06
        [cell_reuse_recompute_pass]: 1.54e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.53e-05
        [j_node_and_user_rematch]: 9.02999e-06
        [meta_fg_expand]: 2.38002e-06
        [replace_old_param]: 8.87e-06
        [inline_without_move]: 5.54e-06
        [renormalize]: 0.00059309
        [add_forward_monad_depend]: 4.90001e-06
        [auto_monad_grad]: 2.37999e-06
        [auto_monad_eliminator]: 1.343e-05
        [cse]: 2.829e-05
        [replace_applicator]: 1.177e-05
    [Cycle 2]: 0.00031536, [27]
        [switch_simplify]: 5.44e-06
        [loop_unroll]: 4.82e-06
        [a_1]: 8.999e-05
        [with_stream_mark]: 8.96998e-06
        [recompute_prepare]: 5.22999e-06
        [updatestate_depend_eliminate]: 2.59001e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 5.30001e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 4.83001e-06
        [meta_shard_fg_expand]: 1.34998e-06
        [get_grad_eliminate_]: 4.94998e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 1.44998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.232e-05
        [j_node_and_user_rematch]: 7.53e-06
        [meta_fg_expand]: 1.81e-06
        [replace_old_param]: 6.86001e-06
        [inline_without_move]: 4.80999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.32999e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 5.00001e-06
        [cse]: 1.066e-05
        [replace_applicator]: 5.05001e-06
[py_interpret_to_execute_after_opt_a]: 8.87e-06
[rewriter_after_opt_a]: 3.416e-05
[convert_after_rewriter]: 7.63999e-06
[order_py_execute_after_rewriter]: 5.20999e-06
[mutable_eliminate]: 0.00049698
[jit_opt_b]: 4.91e-05, [1]
    [Cycle 1]: 4.267e-05, [2]
        [frontend_op_eliminate]: 1.631e-05
        [inline_after_opt_a]: 1.538e-05
[cconv]: 2.455e-05
[loop_unroll]: 0.00042582
[jit_opt_after_cconv]: 0.00013966, [1]
    [Cycle 1]: 0.00013355, [11]
        [c_1]: 2.215e-05
        [parameter_eliminate]: 2.26998e-06
        [updatestate_depend_eliminate]: 6.27001e-06
        [updatestate_assign_eliminate]: 2.41998e-06
        [updatestate_loads_eliminate]: 2.01998e-06
        [cse]: 1.889e-05
        [call_graph_tuple_transform]: 1.734e-05
        [tuple_list_get_item_eliminator]: 5.34e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.36002e-06
[remove_dup_value]: 1.419e-05
[partial_unused_args_eliminate]: 2.43e-06
[environ_conv]: 5.51e-06
[add_recomputation]: 5.022e-05
[cse_after_recomputation]: 2.059e-05, [1]
    [Cycle 1]: 1.509e-05, [1]
        [cse]: 9.53002e-06
[auto_monad_reorder]: 1.461e-05
[get_jit_bprop_graph]: 1.64e-06
[rewriter_after_jit_bprop_graph]: 5.04e-06
[opt_after_jit_grad]: 0.00044889
[symbol_engine_optimizer]: 7.156e-05, [1]
    [Cycle 1]: 6.546e-05, [6]
        [build]: 3.31999e-06
        [elim_shapecalc]: 7.55e-06
        [elim_not_effective]: 1.217e-05
        [opt_reshape]: 5.37001e-06
        [fold_const_symbol]: 8.33001e-06
        [renormalize]: 5.19998e-07
[validate]: 3.307e-05
Sums
    bootstrap                            :     0.000258s :  2.60%
    type_inference                       :     0.005895s : 59.36%
    event_method                         :     0.000021s :  0.21%
    auto_monad                           :     0.000079s :  0.80%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000048s :  0.48%
    rewriter_before_opt_a                :     0.000085s :  0.86%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000120s :  1.21%
    jit_opt_a.loop_unroll                :     0.000040s :  0.40%
    jit_opt_a.a_1                        :     0.000767s :  7.72%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000593s :  5.97%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.19%
    jit_opt_a.cse                        :     0.000039s :  0.39%
    jit_opt_a.replace_applicator         :     0.000017s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.09%
    rewriter_after_opt_a                 :     0.000034s :  0.34%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000497s :  5.00%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.15%
    cconv                                :     0.000025s :  0.25%
    loop_unroll                          :     0.000426s :  4.29%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000050s :  0.51%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000449s :  4.52%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000033s :  0.33%


Time group info:
------[substitution.]   0.000265    26
 0.68% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 1.90% :     0.000005s :      3: substitution.graph_param_transform
84.99% :     0.000225s :      6: substitution.inline
 1.12% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.63% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.43% :     0.000004s :      2: substitution.replace_old_param
 3.08% :     0.000008s :      1: substitution.switch_simplify
 4.68% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005839     2
83.39% :     0.004869s :      1: type_inference.infer
16.61% :     0.000970s :      1: type_inference.specialize
------[replace.]   0.000132     9
50.00% :     0.000066s :      6: replace.inline
26.06% :     0.000034s :      1: replace.switch_simplify
23.94% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000240     9
92.29% :     0.000221s :      6: match.inline
 3.07% :     0.000007s :      1: match.switch_simplify
 4.64% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   801
 1.23% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.11% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     12: predicate.addn_check_dump
 1.62% :     0.000002s :     12: predicate.addn_zero_filter
 1.95% :     0.000003s :     12: predicate.arithmetic_simplify
 1.25% :     0.000002s :     12: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     12: predicate.compare_switch_simplify
 1.37% :     0.000002s :     12: predicate.depend_value_elim
 1.23% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     12: predicate.environ_get_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.51% :     0.000007s :     26: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     17: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.35% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.18% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     12: predicate.merge_addn
 1.18% :     0.000002s :     12: predicate.minmaximum_grad
 1.51% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     17: predicate.partial_eliminate
 1.18% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     12: predicate.reduce_eliminate
 1.55% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     20: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     12: predicate.reshape_eliminate
 1.22% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.66% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     12: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.76% :     0.000004s :     20: predicate.switch_defer_inline
 2.36% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.91% :     0.000012s :     65: predicate.switch_simplify
 1.24% :     0.000002s :     12: predicate.tile_eliminate
 1.18% :     0.000002s :     12: predicate.transpose_eliminate
 1.58% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.70% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     12: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000732    15
50.08% :     0.000367s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.73% :     0.000086s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.19% :     0.000280s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015281    72
 0.35% :     0.000053s :      1: add_recomputation
 0.54% :     0.000083s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.80% :     0.000275s :      1: bootstrap
 0.18% :     0.000028s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
33.84% :     0.005172s :      1: jit_opt_a
 0.93% :     0.000143s :      1: jit_opt_after_cconv
 0.34% :     0.000052s :      1: jit_opt_b
 2.83% :     0.000433s :      1: loop_unroll
 3.30% :     0.000505s :      1: mutable_eliminate
 6.80% :     0.001039s :     26: opt.transform.jit_opt_a
 0.31% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.99% :     0.000457s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.33% :     0.000050s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 2.10% :     0.000321s :      1: renormalize.infer
 1.73% :     0.000265s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000038s :      1: rewriter_after_opt_a
 0.58% :     0.000088s :      1: rewriter_before_opt_a
 0.49% :     0.000074s :      1: symbol_engine_optimizer
38.65% :     0.005906s :      1: type_inference


TotalTime = 0.0210256, [33]
[bootstrap]: 0.00034714
[type_inference]: 0.00575094
[event_method]: 1.829e-05
[auto_monad]: 6.674e-05
[graph_reusing]: 6.06e-06
[pre_auto_parallel]: 2.57001e-06
[py_interpret_to_execute]: 3.859e-05
[rewriter_before_opt_a]: 7.467e-05
[expand_dump_flag]: 3.36999e-06
[jit_opt_a]: 0.00615954, [2]
    [Cycle 1]: 0.00181041, [27]
        [switch_simplify]: 6.281e-05
        [loop_unroll]: 2.897e-05
        [a_1]: 0.00058715
        [with_stream_mark]: 2.562e-05
        [recompute_prepare]: 7.36001e-06
        [updatestate_depend_eliminate]: 4.02e-06
        [updatestate_assign_eliminate]: 3.01999e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [parameter_eliminate]: 1.95001e-06
        [specialize_transform]: 6.24999e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 2.06998e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 3.80998e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.493e-05
        [j_node_and_user_rematch]: 8.97e-06
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 9.00999e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 0.00074901
        [add_forward_monad_depend]: 5.72999e-06
        [auto_monad_grad]: 2.34999e-06
        [auto_monad_eliminator]: 1.528e-05
        [cse]: 2.921e-05
        [replace_applicator]: 1.495e-05
    [Cycle 2]: 0.00035721, [27]
        [switch_simplify]: 6.49999e-06
        [loop_unroll]: 5.56998e-06
        [a_1]: 0.00010422
        [with_stream_mark]: 1.312e-05
        [recompute_prepare]: 5.44998e-06
        [updatestate_depend_eliminate]: 3.48999e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.06997e-06
        [specialize_transform]: 4.86997e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.07e-06
        [meta_shard_fg_expand]: 1.35001e-06
        [get_grad_eliminate_]: 4.69002e-06
        [merge_forward]: 3.16001e-06
        [cell_reuse_recompute_pass]: 2.83e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.543e-05
        [j_node_and_user_rematch]: 8.37e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 8.37e-06
        [inline_without_move]: 1.208e-05
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.60001e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 6.29001e-06
        [cse]: 1.234e-05
        [replace_applicator]: 5.15999e-06
[py_interpret_to_execute_after_opt_a]: 1.517e-05
[rewriter_after_opt_a]: 3.638e-05
[convert_after_rewriter]: 7.80998e-06
[order_py_execute_after_rewriter]: 5.12999e-06
[mutable_eliminate]: 0.00067228
[jit_opt_b]: 5.249e-05, [1]
    [Cycle 1]: 4.418e-05, [2]
        [frontend_op_eliminate]: 1.673e-05
        [inline_after_opt_a]: 1.542e-05
[cconv]: 2.784e-05
[loop_unroll]: 0.00043388
[jit_opt_after_cconv]: 0.00014352, [1]
    [Cycle 1]: 0.00013729, [11]
        [c_1]: 2.139e-05
        [parameter_eliminate]: 2.87002e-06
        [updatestate_depend_eliminate]: 6.47001e-06
        [updatestate_assign_eliminate]: 2.36998e-06
        [updatestate_loads_eliminate]: 1.96998e-06
        [cse]: 2.034e-05
        [call_graph_tuple_transform]: 2.009e-05
        [tuple_list_get_item_eliminator]: 5.34998e-06
        [none_parameter_eliminate]: 1.86998e-06
        [renormalize]: 6.39993e-07
        [switch_simplify]: 5.39e-06
[remove_dup_value]: 1.363e-05
[partial_unused_args_eliminate]: 2.02999e-06
[environ_conv]: 5.56e-06
[add_recomputation]: 5.047e-05
[cse_after_recomputation]: 2.176e-05, [1]
    [Cycle 1]: 1.597e-05, [1]
        [cse]: 9.82001e-06
[auto_monad_reorder]: 1.44e-05
[get_jit_bprop_graph]: 1.89999e-06
[rewriter_after_jit_bprop_graph]: 6.61e-06
[opt_after_jit_grad]: 0.00046723
[symbol_engine_optimizer]: 7.347e-05, [1]
    [Cycle 1]: 6.641e-05, [6]
        [build]: 3.54002e-06
        [elim_shapecalc]: 7.33e-06
        [elim_not_effective]: 1.314e-05
        [opt_reshape]: 5.68002e-06
        [fold_const_symbol]: 8.47998e-06
        [renormalize]: 5.3001e-07
[validate]: 3.831e-05
[backend_pass]: 9.49978e-07
[task_emit]: 0.00624147
[execute]: 8.94e-06
Sums
    bootstrap                            :     0.000347s :  2.12%
    type_inference                       :     0.005751s : 35.11%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000067s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.24%
    rewriter_before_opt_a                :     0.000075s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000069s :  0.42%
    jit_opt_a.loop_unroll                :     0.000035s :  0.21%
    jit_opt_a.a_1                        :     0.000691s :  4.22%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000017s :  0.10%
    jit_opt_a.renormalize                :     0.000749s :  4.57%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.13%
    jit_opt_a.cse                        :     0.000042s :  0.25%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.09%
    rewriter_after_opt_a                 :     0.000036s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000672s :  4.10%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000028s :  0.17%
    loop_unroll                          :     0.000434s :  2.65%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000050s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000467s :  2.85%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006241s : 38.11%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000250    23
 0.73% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.57% :     0.000006s :      3: substitution.graph_param_transform
86.09% :     0.000215s :      4: substitution.inline
 1.36% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.86% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.14% :     0.000005s :      2: substitution.replace_old_param
 4.69% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005676     2
84.61% :     0.004803s :      1: type_inference.infer
15.39% :     0.000873s :      1: type_inference.specialize
------[replace.]   0.000071     6
68.38% :     0.000049s :      4: replace.inline
31.62% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000223     6
95.27% :     0.000212s :      4: match.inline
 4.73% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.47% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.15% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.78% :     0.000002s :     10: predicate.addn_zero_filter
 2.53% :     0.000003s :     10: predicate.arithmetic_simplify
 1.30% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.11% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.98% :     0.000007s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.75% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.36% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 2.02% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.67% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.82% :     0.000002s :     18: predicate.replace_applicator
 0.95% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.23% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.90% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.65% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.09% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.99% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.47% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000639    12
47.99% :     0.000307s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.45% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.57% :     0.000278s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022798    76
 0.23% :     0.000053s :      1: add_recomputation
 0.31% :     0.000070s :      1: auto_monad
 0.07% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.63% :     0.000371s :      1: bootstrap
 0.13% :     0.000030s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.03% :     0.006163s :      1: jit_opt_a
 0.64% :     0.000146s :      1: jit_opt_after_cconv
 0.24% :     0.000055s :      1: jit_opt_b
 1.94% :     0.000443s :      1: loop_unroll
 2.99% :     0.000682s :      1: mutable_eliminate
 4.04% :     0.000921s :     26: opt.transform.jit_opt_a
 0.22% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.09% :     0.000475s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000041s :      1: py_interpret_to_execute
 0.08% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.78% :     0.000406s :      1: renormalize.infer
 1.47% :     0.000334s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000039s :      1: rewriter_after_opt_a
 0.34% :     0.000078s :      1: rewriter_before_opt_a
 0.33% :     0.000076s :      1: symbol_engine_optimizer
27.45% :     0.006257s :      1: task_emit
25.31% :     0.005770s :      1: type_inference
 0.26% :     0.000059s :      1: validate


TotalTime = 0.0246425, [33]
[bootstrap]: 0.00030951
[type_inference]: 0.00552521
[event_method]: 1.773e-05
[auto_monad]: 6.177e-05
[graph_reusing]: 6.02999e-06
[pre_auto_parallel]: 2.27001e-06
[py_interpret_to_execute]: 3.792e-05
[rewriter_before_opt_a]: 7.043e-05
[expand_dump_flag]: 2.92002e-06
[jit_opt_a]: 0.00830227, [2]
    [Cycle 1]: 0.00217709, [27]
        [switch_simplify]: 6.943e-05
        [loop_unroll]: 2.969e-05
        [a_1]: 0.00063571
        [with_stream_mark]: 2.177e-05
        [recompute_prepare]: 8.57e-06
        [updatestate_depend_eliminate]: 3.75998e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 6.26998e-06
        [updatestate_useless_node_eliminater]: 5.44998e-06
        [accelerated_algorithm]: 5.59998e-06
        [meta_shard_fg_expand]: 2.12001e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 4.02998e-06
        [cell_reuse_recompute_pass]: 1.38002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.452e-05
        [j_node_and_user_rematch]: 1.014e-05
        [meta_fg_expand]: 2.21e-06
        [replace_old_param]: 9.20001e-06
        [inline_without_move]: 5.60001e-06
        [renormalize]: 0.00107483
        [add_forward_monad_depend]: 7.3e-06
        [auto_monad_grad]: 2.42001e-06
        [auto_monad_eliminator]: 1.606e-05
        [cse]: 2.095e-05
        [replace_applicator]: 1.64e-05
    [Cycle 2]: 0.00036575, [27]
        [switch_simplify]: 5.97999e-06
        [loop_unroll]: 4.95001e-06
        [a_1]: 0.00010059
        [with_stream_mark]: 1.177e-05
        [recompute_prepare]: 5.86e-06
        [updatestate_depend_eliminate]: 3.11001e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.21003e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 5.77999e-06
        [updatestate_useless_node_eliminater]: 4.91002e-06
        [accelerated_algorithm]: 6.86999e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 4.60999e-06
        [merge_forward]: 3.25e-06
        [cell_reuse_recompute_pass]: 1.84e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.555e-05
        [j_node_and_user_rematch]: 9.82001e-06
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 9.89999e-06
        [inline_without_move]: 5.29998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 1.35999e-06
        [auto_monad_eliminator]: 7.34002e-06
        [cse]: 1.117e-05
        [replace_applicator]: 5.45001e-06
[py_interpret_to_execute_after_opt_a]: 1.448e-05
[rewriter_after_opt_a]: 4.027e-05
[convert_after_rewriter]: 1.09e-05
[order_py_execute_after_rewriter]: 6.26998e-06
[mutable_eliminate]: 0.00082144
[jit_opt_b]: 6.424e-05, [1]
    [Cycle 1]: 5.504e-05, [2]
        [frontend_op_eliminate]: 2.345e-05
        [inline_after_opt_a]: 1.773e-05
[cconv]: 3.198e-05
[loop_unroll]: 0.00050685
[jit_opt_after_cconv]: 0.00015672, [1]
    [Cycle 1]: 0.00015019, [11]
        [c_1]: 2.212e-05
        [parameter_eliminate]: 2.70002e-06
        [updatestate_depend_eliminate]: 6.33e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [cse]: 2.626e-05
        [call_graph_tuple_transform]: 2.221e-05
        [tuple_list_get_item_eliminator]: 5.05999e-06
        [none_parameter_eliminate]: 1.71e-06
        [renormalize]: 6.39993e-07
        [switch_simplify]: 5.59998e-06
[remove_dup_value]: 1.551e-05
[partial_unused_args_eliminate]: 2.56998e-06
[environ_conv]: 6.10002e-06
[add_recomputation]: 5.341e-05
[cse_after_recomputation]: 2.343e-05, [1]
    [Cycle 1]: 1.742e-05, [1]
        [cse]: 1.049e-05
[auto_monad_reorder]: 1.699e-05
[get_jit_bprop_graph]: 2.10002e-06
[rewriter_after_jit_bprop_graph]: 8.43999e-06
[opt_after_jit_grad]: 0.00054265
[symbol_engine_optimizer]: 8.173e-05, [1]
    [Cycle 1]: 7.316e-05, [6]
        [build]: 5.29e-06
        [elim_shapecalc]: 8.46002e-06
        [elim_not_effective]: 1.439e-05
        [opt_reshape]: 5.84e-06
        [fold_const_symbol]: 9.71998e-06
        [renormalize]: 7.29982e-07
[validate]: 4.707e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00760076
[execute]: 1.056e-05
Sums
    bootstrap                            :     0.000310s :  1.70%
    type_inference                       :     0.005525s : 30.38%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000062s :  0.34%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000038s :  0.21%
    rewriter_before_opt_a                :     0.000070s :  0.39%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000075s :  0.41%
    jit_opt_a.loop_unroll                :     0.000035s :  0.19%
    jit_opt_a.a_1                        :     0.000736s :  4.05%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001075s :  5.91%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.13%
    jit_opt_a.cse                        :     0.000032s :  0.18%
    jit_opt_a.replace_applicator         :     0.000022s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000040s :  0.22%
    convert_after_rewriter               :     0.000011s :  0.06%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000821s :  4.52%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.10%
    cconv                                :     0.000032s :  0.18%
    loop_unroll                          :     0.000507s :  2.79%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000053s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000543s :  2.98%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000047s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007601s : 41.79%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000280    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 2.10% :     0.000006s :      3: substitution.graph_param_transform
85.97% :     0.000240s :      4: substitution.inline
 1.65% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.98% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.34% :     0.000007s :      2: substitution.replace_old_param
 4.68% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005460     2
83.90% :     0.004581s :      1: type_inference.infer
16.10% :     0.000879s :      1: type_inference.specialize
------[replace.]   0.000080     6
69.32% :     0.000056s :      4: replace.inline
30.68% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000250     6
95.24% :     0.000238s :      4: match.inline
 4.76% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.61% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 1.95% :     0.000003s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.75% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.32% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.34% :     0.000002s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.11% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.44% :     0.000007s :     22: predicate.inline
 0.78% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.07% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.49% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.84% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 2.23% :     0.000003s :      3: predicate.mutable_eliminate
 0.75% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.34% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.36% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.50% :     0.000002s :     18: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.88% :     0.000002s :     10: predicate.reshape_eliminate
 1.49% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.15% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.32% :     0.000002s :      6: predicate.special_op_eliminate
 0.82% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.24% :     0.000003s :     16: predicate.switch_defer_inline
 2.00% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.12% :     0.000009s :     53: predicate.switch_simplify
 1.32% :     0.000002s :     10: predicate.tile_eliminate
 1.56% :     0.000002s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.50% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000683    12
44.39% :     0.000303s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.90% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.71% :     0.000326s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026807    76
 0.21% :     0.000057s :      1: add_recomputation
 0.24% :     0.000065s :      1: auto_monad
 0.08% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.21% :     0.000324s :      1: bootstrap
 0.13% :     0.000035s :      1: cconv
 0.05% :     0.000013s :      1: convert_after_rewriter
 0.10% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.08% :     0.000021s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.05% :     0.000013s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
30.98% :     0.008306s :      1: jit_opt_a
 0.60% :     0.000160s :      1: jit_opt_after_cconv
 0.25% :     0.000067s :      1: jit_opt_b
 1.92% :     0.000516s :      1: loop_unroll
 3.11% :     0.000835s :      1: mutable_eliminate
 3.64% :     0.000977s :     26: opt.transform.jit_opt_a
 0.19% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000033s :      4: opt.transform.jit_opt_b
 0.05% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.06% :     0.000553s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000041s :      1: py_interpret_to_execute
 0.06% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.20% :     0.000589s :      1: renormalize.infer
 1.77% :     0.000476s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000044s :      1: rewriter_after_opt_a
 0.28% :     0.000074s :      1: rewriter_before_opt_a
 0.32% :     0.000085s :      1: symbol_engine_optimizer
28.44% :     0.007623s :      1: task_emit
20.66% :     0.005537s :      1: type_inference
 0.26% :     0.000070s :      1: validate


TotalTime = 0.0153458, [30]
[bootstrap]: 0.0002544
[type_inference]: 0.00657312
[event_method]: 2.295e-05
[auto_monad]: 8.509e-05
[graph_reusing]: 6.64999e-06
[pre_auto_parallel]: 3.08e-06
[py_interpret_to_execute]: 4.996e-05
[rewriter_before_opt_a]: 8.692e-05
[expand_dump_flag]: 3.3e-06
[jit_opt_a]: 0.00585763, [2]
    [Cycle 1]: 0.00200362, [27]
        [switch_simplify]: 0.00012447
        [loop_unroll]: 3.504e-05
        [a_1]: 0.00075361
        [with_stream_mark]: 2.185e-05
        [recompute_prepare]: 8.15e-06
        [updatestate_depend_eliminate]: 4.00998e-06
        [updatestate_assign_eliminate]: 3.38999e-06
        [updatestate_loads_eliminate]: 2.91e-06
        [parameter_eliminate]: 2.34001e-06
        [specialize_transform]: 6.11e-06
        [updatestate_useless_node_eliminater]: 5.11997e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 2.07001e-06
        [get_grad_eliminate_]: 5.72999e-06
        [merge_forward]: 3.9e-06
        [cell_reuse_recompute_pass]: 1.29998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.546e-05
        [j_node_and_user_rematch]: 9.19e-06
        [meta_fg_expand]: 2.83e-06
        [replace_old_param]: 8.76002e-06
        [inline_without_move]: 5.49e-06
        [renormalize]: 0.0007224
        [add_forward_monad_depend]: 5.71e-06
        [auto_monad_grad]: 2.55002e-06
        [auto_monad_eliminator]: 1.556e-05
        [cse]: 3.144e-05
        [replace_applicator]: 1.39e-05
    [Cycle 2]: 0.00032828, [27]
        [switch_simplify]: 5.76e-06
        [loop_unroll]: 4.75001e-06
        [a_1]: 9.102e-05
        [with_stream_mark]: 1.135e-05
        [recompute_prepare]: 6.14001e-06
        [updatestate_depend_eliminate]: 3.31001e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 5.49e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 4.94e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 4.40999e-06
        [merge_forward]: 3.21001e-06
        [cell_reuse_recompute_pass]: 1.19998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.368e-05
        [j_node_and_user_rematch]: 8.23999e-06
        [meta_fg_expand]: 2.07999e-06
        [replace_old_param]: 7.8e-06
        [inline_without_move]: 5.02e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 5.98998e-06
        [cse]: 1.061e-05
        [replace_applicator]: 5.28002e-06
[py_interpret_to_execute_after_opt_a]: 1.232e-05
[rewriter_after_opt_a]: 3.624e-05
[convert_after_rewriter]: 7.03e-06
[order_py_execute_after_rewriter]: 5.52999e-06
[mutable_eliminate]: 0.00061194
[jit_opt_b]: 6.373e-05, [1]
    [Cycle 1]: 5.548e-05, [2]
        [frontend_op_eliminate]: 1.777e-05
        [inline_after_opt_a]: 2.565e-05
[cconv]: 2.854e-05
[loop_unroll]: 0.0004865
[jit_opt_after_cconv]: 0.00015639, [1]
    [Cycle 1]: 0.00014841, [11]
        [c_1]: 2.213e-05
        [parameter_eliminate]: 4.02002e-06
        [updatestate_depend_eliminate]: 7.06999e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 2.547e-05
        [call_graph_tuple_transform]: 2.06e-05
        [tuple_list_get_item_eliminator]: 5.10001e-06
        [none_parameter_eliminate]: 1.66002e-06
        [renormalize]: 2.9002e-07
        [switch_simplify]: 5.32999e-06
[remove_dup_value]: 1.511e-05
[partial_unused_args_eliminate]: 2.26998e-06
[environ_conv]: 5.84999e-06
[add_recomputation]: 5.351e-05
[cse_after_recomputation]: 2.394e-05, [1]
    [Cycle 1]: 1.784e-05, [1]
        [cse]: 1.038e-05
[auto_monad_reorder]: 1.532e-05
[get_jit_bprop_graph]: 1.96998e-06
[rewriter_after_jit_bprop_graph]: 6.89001e-06
[opt_after_jit_grad]: 0.00054279
[symbol_engine_optimizer]: 9.019e-05, [1]
    [Cycle 1]: 8.23e-05, [6]
        [build]: 5.53002e-06
        [elim_shapecalc]: 7.9e-06
        [elim_not_effective]: 1.421e-05
        [opt_reshape]: 5.61998e-06
        [fold_const_symbol]: 8.93002e-06
        [renormalize]: 6.19999e-07
[validate]: 4.66e-05
Sums
    bootstrap                            :     0.000254s :  2.27%
    type_inference                       :     0.006573s : 58.74%
    event_method                         :     0.000023s :  0.21%
    auto_monad                           :     0.000085s :  0.76%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000050s :  0.45%
    rewriter_before_opt_a                :     0.000087s :  0.78%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000130s :  1.16%
    jit_opt_a.loop_unroll                :     0.000040s :  0.36%
    jit_opt_a.a_1                        :     0.000845s :  7.55%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.000722s :  6.46%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.19%
    jit_opt_a.cse                        :     0.000042s :  0.38%
    jit_opt_a.replace_applicator         :     0.000019s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.11%
    rewriter_after_opt_a                 :     0.000036s :  0.32%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000612s :  5.47%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.23%
    cconv                                :     0.000029s :  0.26%
    loop_unroll                          :     0.000487s :  4.35%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000025s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000015s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000054s :  0.48%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000543s :  4.85%
    symbol_engine_optimizer.build        :     0.000006s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000047s :  0.42%


Time group info:
------[substitution.]   0.000339    26
 0.59% :     0.000002s :      2: substitution.elim_not_effective
 0.42% :     0.000001s :      2: substitution.fold_const_symbol
 1.57% :     0.000005s :      3: substitution.graph_param_transform
87.31% :     0.000296s :      6: substitution.inline
 1.06% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.41% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.34% :     0.000005s :      2: substitution.replace_old_param
 2.26% :     0.000008s :      1: substitution.switch_simplify
 4.03% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006502     2
81.61% :     0.005307s :      1: type_inference.infer
18.39% :     0.001195s :      1: type_inference.specialize
------[replace.]   0.000141     9
48.73% :     0.000069s :      6: replace.inline
29.13% :     0.000041s :      1: replace.switch_simplify
22.14% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000312     9
93.79% :     0.000293s :      6: match.inline
 2.22% :     0.000007s :      1: match.switch_simplify
 3.99% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000140   801
 1.29% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.75% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     12: predicate.addn_check_dump
 1.40% :     0.000002s :     12: predicate.addn_zero_filter
 1.69% :     0.000002s :     12: predicate.arithmetic_simplify
 1.19% :     0.000002s :     12: predicate.cast_eliminate
 0.36% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000002s :     12: predicate.compare_switch_simplify
 1.30% :     0.000002s :     12: predicate.depend_value_elim
 1.09% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.17% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.12% :     0.000002s :     12: predicate.dict_set_item_eliminator
 1.40% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000001s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.53% :     0.000002s :     12: predicate.environ_get_eliminate
 1.35% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.56% :     0.000008s :     26: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000002s :      6: predicate.less_batch_normalization
 1.37% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.59% :     0.000002s :     17: predicate.load_eliminater
 2.53% :     0.000004s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.89% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     12: predicate.merge_addn
 1.12% :     0.000002s :     12: predicate.minmaximum_grad
 2.39% :     0.000003s :      3: predicate.mutable_eliminate
 0.37% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000003s :     17: predicate.partial_eliminate
 1.24% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     12: predicate.reduce_eliminate
 1.39% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.55% :     0.000002s :     20: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000001s :      3: predicate.reset_defer_inline
 1.27% :     0.000002s :     12: predicate.reshape_eliminate
 1.17% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     12: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 0.84% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.61% :     0.000004s :     20: predicate.switch_defer_inline
 2.25% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.57% :     0.000012s :     65: predicate.switch_simplify
 1.18% :     0.000002s :     12: predicate.tile_eliminate
 1.13% :     0.000002s :     12: predicate.transpose_eliminate
 1.42% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.93% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.57% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     12: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000901    15
41.82% :     0.000377s :      6: func_graph_cloner_run.FuncGraphClonerGraph
16.50% :     0.000149s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.68% :     0.000376s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017268    72
 0.33% :     0.000057s :      1: add_recomputation
 0.51% :     0.000089s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 1.56% :     0.000270s :      1: bootstrap
 0.18% :     0.000031s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000028s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000010s :      1: graph_reusing
33.94% :     0.005861s :      1: jit_opt_a
 0.92% :     0.000159s :      1: jit_opt_after_cconv
 0.39% :     0.000067s :      1: jit_opt_b
 2.87% :     0.000496s :      1: loop_unroll
 3.62% :     0.000626s :      1: mutable_eliminate
 6.57% :     0.001135s :     26: opt.transform.jit_opt_a
 0.28% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000037s :      4: opt.transform.jit_opt_b
 0.10% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.21% :     0.000555s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000053s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000018s :      1: remove_dup_value
 2.30% :     0.000397s :      1: renormalize.infer
 1.83% :     0.000315s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000039s :      1: rewriter_after_opt_a
 0.52% :     0.000090s :      1: rewriter_before_opt_a
 0.54% :     0.000093s :      1: symbol_engine_optimizer
38.15% :     0.006588s :      1: type_inference


TotalTime = 0.017506, [33]
[bootstrap]: 0.00026303
[type_inference]: 0.00419649
[event_method]: 1.641e-05
[auto_monad]: 6.291e-05
[graph_reusing]: 5.66e-06
[pre_auto_parallel]: 2.37999e-06
[py_interpret_to_execute]: 3.479e-05
[rewriter_before_opt_a]: 7.187e-05
[expand_dump_flag]: 2.99999e-06
[jit_opt_a]: 0.00492051, [2]
    [Cycle 1]: 0.00140733, [27]
        [switch_simplify]: 5.144e-05
        [loop_unroll]: 2.837e-05
        [a_1]: 0.00048813
        [with_stream_mark]: 1.482e-05
        [recompute_prepare]: 6.74001e-06
        [updatestate_depend_eliminate]: 3.6e-06
        [updatestate_assign_eliminate]: 3.6e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [parameter_eliminate]: 2.44999e-06
        [specialize_transform]: 5.93002e-06
        [updatestate_useless_node_eliminater]: 4.91997e-06
        [accelerated_algorithm]: 5.08002e-06
        [meta_shard_fg_expand]: 1.83002e-06
        [get_grad_eliminate_]: 5.52001e-06
        [merge_forward]: 3.85e-06
        [cell_reuse_recompute_pass]: 1.65001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.342e-05
        [j_node_and_user_rematch]: 8.62998e-06
        [meta_fg_expand]: 2.10002e-06
        [replace_old_param]: 8.69e-06
        [inline_without_move]: 4.87998e-06
        [renormalize]: 0.00051816
        [add_forward_monad_depend]: 5.12e-06
        [auto_monad_grad]: 1.84e-06
        [auto_monad_eliminator]: 1.272e-05
        [cse]: 2.798e-05
        [replace_applicator]: 1.145e-05
    [Cycle 2]: 0.0003147, [27]
        [switch_simplify]: 5.38002e-06
        [loop_unroll]: 4.81002e-06
        [a_1]: 8.918e-05
        [with_stream_mark]: 8.17998e-06
        [recompute_prepare]: 4.87e-06
        [updatestate_depend_eliminate]: 2.51e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.86998e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 4.65001e-06
        [updatestate_useless_node_eliminater]: 4.68999e-06
        [accelerated_algorithm]: 5.14e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.16998e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.216e-05
        [j_node_and_user_rematch]: 7.63001e-06
        [meta_fg_expand]: 1.68002e-06
        [replace_old_param]: 1.25e-05
        [inline_without_move]: 4.62e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 5.23002e-06
        [cse]: 1.007e-05
        [replace_applicator]: 5.09e-06
[py_interpret_to_execute_after_opt_a]: 9.89001e-06
[rewriter_after_opt_a]: 3.186e-05
[convert_after_rewriter]: 7.13998e-06
[order_py_execute_after_rewriter]: 5.54e-06
[mutable_eliminate]: 0.00046904
[jit_opt_b]: 4.696e-05, [1]
    [Cycle 1]: 4.056e-05, [2]
        [frontend_op_eliminate]: 1.551e-05
        [inline_after_opt_a]: 1.383e-05
[cconv]: 2.545e-05
[loop_unroll]: 0.00047302
[jit_opt_after_cconv]: 0.00013319, [1]
    [Cycle 1]: 0.00012682, [11]
        [c_1]: 2.01e-05
        [parameter_eliminate]: 2.33998e-06
        [updatestate_depend_eliminate]: 5.11002e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.632e-05
        [call_graph_tuple_transform]: 1.693e-05
        [tuple_list_get_item_eliminator]: 5.42001e-06
        [none_parameter_eliminate]: 1.57999e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 5.20999e-06
[remove_dup_value]: 1.262e-05
[partial_unused_args_eliminate]: 2.05002e-06
[environ_conv]: 5.22e-06
[add_recomputation]: 4.62e-05
[cse_after_recomputation]: 1.94e-05, [1]
    [Cycle 1]: 1.427e-05, [1]
        [cse]: 8.67998e-06
[auto_monad_reorder]: 1.385e-05
[get_jit_bprop_graph]: 1.39e-06
[rewriter_after_jit_bprop_graph]: 4.82e-06
[opt_after_jit_grad]: 0.0004471
[symbol_engine_optimizer]: 6.811e-05, [1]
    [Cycle 1]: 6.21e-05, [6]
        [build]: 3.46999e-06
        [elim_shapecalc]: 6.94999e-06
        [elim_not_effective]: 1.085e-05
        [opt_reshape]: 5.59e-06
        [fold_const_symbol]: 8.42998e-06
        [renormalize]: 3.59985e-07
[validate]: 3.239e-05
[backend_pass]: 1.12999e-06
[task_emit]: 0.00587478
[execute]: 7.94002e-06
Sums
    bootstrap                            :     0.000263s :  1.92%
    type_inference                       :     0.004196s : 30.57%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000063s :  0.46%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.25%
    rewriter_before_opt_a                :     0.000072s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.41%
    jit_opt_a.loop_unroll                :     0.000033s :  0.24%
    jit_opt_a.a_1                        :     0.000577s :  4.20%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000021s :  0.15%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000518s :  3.77%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000038s :  0.28%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000469s :  3.42%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.19%
    loop_unroll                          :     0.000473s :  3.45%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000046s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000447s :  3.26%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005875s : 42.79%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000171    23
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.87% :     0.000001s :      2: substitution.fold_const_symbol
 2.90% :     0.000005s :      3: substitution.graph_param_transform
81.57% :     0.000139s :      4: substitution.inline
 1.71% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.42% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.14% :     0.000004s :      2: substitution.replace_old_param
 7.34% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004136     2
81.41% :     0.003367s :      1: type_inference.infer
18.59% :     0.000769s :      1: type_inference.specialize
------[replace.]   0.000063     6
68.35% :     0.000043s :      4: replace.inline
31.65% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000148     6
92.35% :     0.000137s :      4: match.inline
 7.65% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.21% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.08% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.51% :     0.000002s :     10: predicate.addn_zero_filter
 2.10% :     0.000002s :     10: predicate.arithmetic_simplify
 1.39% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000001s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.61% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.30% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000000s :      3: predicate.graph_param_transform
 5.44% :     0.000006s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.17% :     0.000002s :     15: predicate.load_eliminater
 1.56% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.14% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.16% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000002s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.66% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.49% :     0.000008s :     53: predicate.switch_simplify
 1.31% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.08% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000539    12
49.72% :     0.000268s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.01% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.27% :     0.000244s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018902    76
 0.26% :     0.000049s :      1: add_recomputation
 0.35% :     0.000066s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.48% :     0.000279s :      1: bootstrap
 0.15% :     0.000028s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.07% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.05% :     0.004923s :      1: jit_opt_a
 0.72% :     0.000136s :      1: jit_opt_after_cconv
 0.26% :     0.000049s :      1: jit_opt_b
 2.55% :     0.000482s :      1: loop_unroll
 2.53% :     0.000478s :      1: mutable_eliminate
 4.13% :     0.000781s :     26: opt.transform.jit_opt_a
 0.23% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000018s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.41% :     0.000455s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000037s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.37% :     0.000259s :      1: renormalize.infer
 1.33% :     0.000252s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000035s :      1: rewriter_after_opt_a
 0.39% :     0.000074s :      1: rewriter_before_opt_a
 0.37% :     0.000071s :      1: symbol_engine_optimizer
31.15% :     0.005888s :      1: task_emit
22.27% :     0.004209s :      1: type_inference
 0.27% :     0.000051s :      1: validate


TotalTime = 0.0240189, [33]
[bootstrap]: 0.00032786
[type_inference]: 0.00497822
[event_method]: 1.641e-05
[auto_monad]: 6.7e-05
[graph_reusing]: 5.35001e-06
[pre_auto_parallel]: 2.91e-06
[py_interpret_to_execute]: 3.848e-05
[rewriter_before_opt_a]: 7.368e-05
[expand_dump_flag]: 2.96001e-06
[jit_opt_a]: 0.00752777, [2]
    [Cycle 1]: 0.00183486, [27]
        [switch_simplify]: 6.312e-05
        [loop_unroll]: 2.967e-05
        [a_1]: 0.00057496
        [with_stream_mark]: 2.253e-05
        [recompute_prepare]: 7.54002e-06
        [updatestate_depend_eliminate]: 3.6e-06
        [updatestate_assign_eliminate]: 2.94001e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 1.77999e-06
        [specialize_transform]: 6.16998e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.26002e-06
        [meta_shard_fg_expand]: 1.94999e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 3.73999e-06
        [cell_reuse_recompute_pass]: 9.69972e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.402e-05
        [j_node_and_user_rematch]: 9.44e-06
        [meta_fg_expand]: 2.46e-06
        [replace_old_param]: 9.60001e-06
        [inline_without_move]: 5.61998e-06
        [renormalize]: 0.00082347
        [add_forward_monad_depend]: 5.93002e-06
        [auto_monad_grad]: 3.06001e-06
        [auto_monad_eliminator]: 1.498e-05
        [cse]: 2.951e-05
        [replace_applicator]: 1.162e-05
    [Cycle 2]: 0.00031117, [27]
        [switch_simplify]: 5.49998e-06
        [loop_unroll]: 4.58999e-06
        [a_1]: 9.003e-05
        [with_stream_mark]: 8.83001e-06
        [recompute_prepare]: 4.69002e-06
        [updatestate_depend_eliminate]: 2.35002e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 8.89995e-07
        [specialize_transform]: 4.97999e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 4.68999e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.44002e-06
        [merge_forward]: 2.37001e-06
        [cell_reuse_recompute_pass]: 1.75001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.228e-05
        [j_node_and_user_rematch]: 8.13001e-06
        [meta_fg_expand]: 1.57999e-06
        [replace_old_param]: 6.84999e-06
        [inline_without_move]: 4.67e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 7.29982e-07
        [auto_monad_eliminator]: 5.05001e-06
        [cse]: 9.72001e-06
        [replace_applicator]: 4.85001e-06
[py_interpret_to_execute_after_opt_a]: 1.118e-05
[rewriter_after_opt_a]: 3.773e-05
[convert_after_rewriter]: 7.87998e-06
[order_py_execute_after_rewriter]: 4.91002e-06
[mutable_eliminate]: 0.00053626
[jit_opt_b]: 4.727e-05, [1]
    [Cycle 1]: 4.059e-05, [2]
        [frontend_op_eliminate]: 1.553e-05
        [inline_after_opt_a]: 1.381e-05
[cconv]: 2.608e-05
[loop_unroll]: 0.00044026
[jit_opt_after_cconv]: 0.00013451, [1]
    [Cycle 1]: 0.00012798, [11]
        [c_1]: 2.018e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 5.09998e-06
        [updatestate_assign_eliminate]: 2.19999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.792e-05
        [call_graph_tuple_transform]: 1.696e-05
        [tuple_list_get_item_eliminator]: 5.01997e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 7.80012e-07
        [switch_simplify]: 5.15999e-06
[remove_dup_value]: 1.463e-05
[partial_unused_args_eliminate]: 2.76e-06
[environ_conv]: 5.01997e-06
[add_recomputation]: 5.032e-05
[cse_after_recomputation]: 2.073e-05, [1]
    [Cycle 1]: 1.538e-05, [1]
        [cse]: 9.87999e-06
[auto_monad_reorder]: 1.545e-05
[get_jit_bprop_graph]: 2.14e-06
[rewriter_after_jit_bprop_graph]: 4.74e-06
[opt_after_jit_grad]: 0.00045882
[symbol_engine_optimizer]: 7.128e-05, [1]
    [Cycle 1]: 6.504e-05, [6]
        [build]: 3.33998e-06
        [elim_shapecalc]: 6.86001e-06
        [elim_not_effective]: 1.304e-05
        [opt_reshape]: 5.41998e-06
        [fold_const_symbol]: 8.83001e-06
        [renormalize]: 4.09986e-07
[validate]: 3.668e-05
[backend_pass]: 1.07e-06
[task_emit]: 0.0088198
[execute]: 9.76e-06
Sums
    bootstrap                            :     0.000328s :  1.82%
    type_inference                       :     0.004978s : 27.62%
    event_method                         :     0.000016s :  0.09%
    auto_monad                           :     0.000067s :  0.37%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.21%
    rewriter_before_opt_a                :     0.000074s :  0.41%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000069s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.19%
    jit_opt_a.a_1                        :     0.000665s :  3.69%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000824s :  4.57%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.11%
    jit_opt_a.cse                        :     0.000039s :  0.22%
    jit_opt_a.replace_applicator         :     0.000016s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000038s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000536s :  2.98%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.08%
    cconv                                :     0.000026s :  0.14%
    loop_unroll                          :     0.000440s :  2.44%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.28%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000459s :  2.55%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008820s : 48.94%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000242    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000002s :      2: substitution.fold_const_symbol
 2.08% :     0.000005s :      3: substitution.graph_param_transform
85.94% :     0.000208s :      4: substitution.inline
 1.29% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.98% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000005s :      2: substitution.replace_old_param
 5.26% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004908     2
81.77% :     0.004013s :      1: type_inference.infer
18.23% :     0.000895s :      1: type_inference.specialize
------[replace.]   0.000071     6
68.47% :     0.000049s :      4: replace.inline
31.53% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000217     6
94.63% :     0.000205s :      4: match.inline
 5.37% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.27% :     0.000003s :     10: predicate.arithmetic_simplify
 1.40% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000001s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.32% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.28% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000001s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.63% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.89% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.39% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.74% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.34% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.66% :     0.000002s :     10: predicate.reshape_eliminate
 1.36% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.18% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.14% :     0.000009s :     53: predicate.switch_simplify
 1.44% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.72% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.53% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000669    12
48.94% :     0.000328s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.07% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.99% :     0.000295s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025812    76
 0.21% :     0.000053s :      1: add_recomputation
 0.27% :     0.000070s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.35% :     0.000349s :      1: bootstrap
 0.11% :     0.000029s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.08% :     0.000021s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
29.18% :     0.007531s :      1: jit_opt_a
 0.53% :     0.000137s :      1: jit_opt_after_cconv
 0.19% :     0.000050s :      1: jit_opt_b
 1.74% :     0.000449s :      1: loop_unroll
 2.11% :     0.000545s :      1: mutable_eliminate
 3.41% :     0.000880s :     26: opt.transform.jit_opt_a
 0.17% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.81% :     0.000467s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000041s :      1: py_interpret_to_execute
 0.05% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.82% :     0.000470s :      1: renormalize.infer
 1.34% :     0.000345s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000040s :      1: rewriter_after_opt_a
 0.30% :     0.000076s :      1: rewriter_before_opt_a
 0.29% :     0.000074s :      1: symbol_engine_optimizer
34.25% :     0.008840s :      1: task_emit
19.35% :     0.004994s :      1: type_inference
 0.22% :     0.000058s :      1: validate


TotalTime = 0.0181431, [33]
[bootstrap]: 0.0002701
[type_inference]: 0.00421797
[event_method]: 1.765e-05
[auto_monad]: 6.38e-05
[graph_reusing]: 5.88002e-06
[pre_auto_parallel]: 2.44001e-06
[py_interpret_to_execute]: 3.328e-05
[rewriter_before_opt_a]: 7.164e-05
[expand_dump_flag]: 3.41999e-06
[jit_opt_a]: 0.00539592, [2]
    [Cycle 1]: 0.00166801, [27]
        [switch_simplify]: 5.899e-05
        [loop_unroll]: 3.156e-05
        [a_1]: 0.00058257
        [with_stream_mark]: 2.144e-05
        [recompute_prepare]: 8.52e-06
        [updatestate_depend_eliminate]: 4.11001e-06
        [updatestate_assign_eliminate]: 3.46999e-06
        [updatestate_loads_eliminate]: 2.69001e-06
        [parameter_eliminate]: 1.96003e-06
        [specialize_transform]: 6.37001e-06
        [updatestate_useless_node_eliminater]: 5.74e-06
        [accelerated_algorithm]: 6.07001e-06
        [meta_shard_fg_expand]: 2.22001e-06
        [get_grad_eliminate_]: 5.59e-06
        [merge_forward]: 3.86999e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.432e-05
        [j_node_and_user_rematch]: 9.32001e-06
        [meta_fg_expand]: 2.02001e-06
        [replace_old_param]: 9.37001e-06
        [inline_without_move]: 5.30999e-06
        [renormalize]: 0.00062081
        [add_forward_monad_depend]: 5.76e-06
        [auto_monad_grad]: 2.49001e-06
        [auto_monad_eliminator]: 1.368e-05
        [cse]: 3.169e-05
        [replace_applicator]: 1.306e-05
    [Cycle 2]: 0.0003463, [27]
        [switch_simplify]: 6.74999e-06
        [loop_unroll]: 5.62001e-06
        [a_1]: 9.965e-05
        [with_stream_mark]: 9.07001e-06
        [recompute_prepare]: 5.76998e-06
        [updatestate_depend_eliminate]: 2.65002e-06
        [updatestate_assign_eliminate]: 2.10002e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 1.34998e-06
        [specialize_transform]: 5.20001e-06
        [updatestate_useless_node_eliminater]: 5.30001e-06
        [accelerated_algorithm]: 5.32999e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.99e-06
        [merge_forward]: 2.27999e-06
        [cell_reuse_recompute_pass]: 2.56998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.484e-05
        [j_node_and_user_rematch]: 8.87999e-06
        [meta_fg_expand]: 1.56002e-06
        [replace_old_param]: 7.53999e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.37999e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 5.54e-06
        [cse]: 1.068e-05
        [replace_applicator]: 5.29998e-06
[py_interpret_to_execute_after_opt_a]: 1.198e-05
[rewriter_after_opt_a]: 3.62e-05
[convert_after_rewriter]: 9.44998e-06
[order_py_execute_after_rewriter]: 6.14999e-06
[mutable_eliminate]: 0.00057716
[jit_opt_b]: 5.116e-05, [1]
    [Cycle 1]: 4.407e-05, [2]
        [frontend_op_eliminate]: 1.67e-05
        [inline_after_opt_a]: 1.523e-05
[cconv]: 2.662e-05
[loop_unroll]: 0.00043861
[jit_opt_after_cconv]: 0.00013457, [1]
    [Cycle 1]: 0.00012784, [11]
        [c_1]: 2.008e-05
        [parameter_eliminate]: 2.53998e-06
        [updatestate_depend_eliminate]: 5.10001e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.11998e-06
        [cse]: 1.709e-05
        [call_graph_tuple_transform]: 1.739e-05
        [tuple_list_get_item_eliminator]: 5.25999e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 5.31998e-06
[remove_dup_value]: 1.321e-05
[partial_unused_args_eliminate]: 2.23998e-06
[environ_conv]: 4.95999e-06
[add_recomputation]: 4.932e-05
[cse_after_recomputation]: 1.934e-05, [1]
    [Cycle 1]: 1.39e-05, [1]
        [cse]: 8.51002e-06
[auto_monad_reorder]: 1.473e-05
[get_jit_bprop_graph]: 2.02001e-06
[rewriter_after_jit_bprop_graph]: 5.13002e-06
[opt_after_jit_grad]: 0.00045137
[symbol_engine_optimizer]: 7.195e-05, [1]
    [Cycle 1]: 6.578e-05, [6]
        [build]: 3.36999e-06
        [elim_shapecalc]: 7.35e-06
        [elim_not_effective]: 1.249e-05
        [opt_reshape]: 5.58002e-06
        [fold_const_symbol]: 9.18002e-06
        [renormalize]: 4.69998e-07
[validate]: 3.584e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00589885
[execute]: 8.28001e-06
Sums
    bootstrap                            :     0.000270s :  1.91%
    type_inference                       :     0.004218s : 29.84%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000064s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.24%
    rewriter_before_opt_a                :     0.000072s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.47%
    jit_opt_a.loop_unroll                :     0.000037s :  0.26%
    jit_opt_a.a_1                        :     0.000682s :  4.83%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000621s :  4.39%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.14%
    jit_opt_a.cse                        :     0.000042s :  0.30%
    jit_opt_a.replace_applicator         :     0.000018s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000036s :  0.26%
    convert_after_rewriter               :     0.000009s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000577s :  4.08%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000027s :  0.19%
    loop_unroll                          :     0.000439s :  3.10%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000049s :  0.35%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000451s :  3.19%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005899s : 41.73%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000246    23
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 1.91% :     0.000005s :      3: substitution.graph_param_transform
86.60% :     0.000213s :      4: substitution.inline
 1.44% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.16% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.55% :     0.000004s :      2: substitution.replace_old_param
 4.98% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004150     2
80.49% :     0.003340s :      1: type_inference.infer
19.51% :     0.000809s :      1: type_inference.specialize
------[replace.]   0.000067     6
67.50% :     0.000045s :      4: replace.inline
32.50% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000221     6
94.95% :     0.000210s :      4: match.inline
 5.05% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.27% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 2.07% :     0.000002s :     10: predicate.arithmetic_simplify
 1.19% :     0.000001s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.66% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.56% :     0.000007s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.07% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.63% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000002s :     15: predicate.partial_eliminate
 1.32% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.95% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.61% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.33% :     0.000010s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.25% :     0.000001s :     10: predicate.transpose_eliminate
 1.69% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.08% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000550    12
50.57% :     0.000278s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.56% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.87% :     0.000241s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019772    76
 0.26% :     0.000052s :      1: add_recomputation
 0.34% :     0.000067s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.42% :     0.000281s :      1: bootstrap
 0.15% :     0.000030s :      1: cconv
 0.06% :     0.000012s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.31% :     0.005399s :      1: jit_opt_a
 0.69% :     0.000137s :      1: jit_opt_after_cconv
 0.27% :     0.000054s :      1: jit_opt_b
 2.26% :     0.000447s :      1: loop_unroll
 2.97% :     0.000587s :      1: mutable_eliminate
 4.58% :     0.000906s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.33% :     0.000460s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000036s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.61% :     0.000318s :      1: renormalize.infer
 1.49% :     0.000295s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000039s :      1: rewriter_after_opt_a
 0.38% :     0.000074s :      1: rewriter_before_opt_a
 0.38% :     0.000074s :      1: symbol_engine_optimizer
29.91% :     0.005914s :      1: task_emit
21.40% :     0.004232s :      1: type_inference
 0.28% :     0.000055s :      1: validate


TotalTime = 0.0189994, [30]
[bootstrap]: 0.00037797
[type_inference]: 0.00879897
[event_method]: 2.408e-05
[auto_monad]: 8.879e-05
[graph_reusing]: 6.49001e-06
[pre_auto_parallel]: 2.59001e-06
[py_interpret_to_execute]: 9.274e-05
[rewriter_before_opt_a]: 8.858e-05
[expand_dump_flag]: 3.48999e-06
[jit_opt_a]: 0.00670369, [2]
    [Cycle 1]: 0.00240321, [27]
        [switch_simplify]: 0.00014769
        [loop_unroll]: 3.585e-05
        [a_1]: 0.00090817
        [with_stream_mark]: 2.668e-05
        [recompute_prepare]: 9.87999e-06
        [updatestate_depend_eliminate]: 4.42e-06
        [updatestate_assign_eliminate]: 3.03e-06
        [updatestate_loads_eliminate]: 2.98998e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 6.67002e-06
        [updatestate_useless_node_eliminater]: 5.62999e-06
        [accelerated_algorithm]: 6.51999e-06
        [meta_shard_fg_expand]: 2.69999e-06
        [get_grad_eliminate_]: 5.81e-06
        [merge_forward]: 3.98001e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.542e-05
        [j_node_and_user_rematch]: 1.147e-05
        [meta_fg_expand]: 2.87002e-06
        [replace_old_param]: 1.087e-05
        [inline_without_move]: 6.54999e-06
        [renormalize]: 0.00089566
        [add_forward_monad_depend]: 6.86001e-06
        [auto_monad_grad]: 3.19001e-06
        [auto_monad_eliminator]: 1.766e-05
        [cse]: 3.231e-05
        [replace_applicator]: 1.525e-05
    [Cycle 2]: 0.00035047, [27]
        [switch_simplify]: 5.94e-06
        [loop_unroll]: 5.24e-06
        [a_1]: 9.936e-05
        [with_stream_mark]: 1.244e-05
        [recompute_prepare]: 5.45001e-06
        [updatestate_depend_eliminate]: 3.03e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 1.51002e-06
        [specialize_transform]: 5.76e-06
        [updatestate_useless_node_eliminater]: 5.10001e-06
        [accelerated_algorithm]: 4.98001e-06
        [meta_shard_fg_expand]: 1.93002e-06
        [get_grad_eliminate_]: 4.95999e-06
        [merge_forward]: 3.6e-06
        [cell_reuse_recompute_pass]: 1.57999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.649e-05
        [j_node_and_user_rematch]: 9.20999e-06
        [meta_fg_expand]: 1.92001e-06
        [replace_old_param]: 8.33001e-06
        [inline_without_move]: 5.02999e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.55999e-06
        [auto_monad_grad]: 1.34998e-06
        [auto_monad_eliminator]: 6.42001e-06
        [cse]: 1.236e-05
        [replace_applicator]: 5.49998e-06
[py_interpret_to_execute_after_opt_a]: 1.499e-05
[rewriter_after_opt_a]: 3.973e-05
[convert_after_rewriter]: 8.63001e-06
[order_py_execute_after_rewriter]: 5.04e-06
[mutable_eliminate]: 0.00088189
[jit_opt_b]: 6.946e-05, [1]
    [Cycle 1]: 5.931e-05, [2]
        [frontend_op_eliminate]: 1.75e-05
        [inline_after_opt_a]: 1.816e-05
[cconv]: 2.897e-05
[loop_unroll]: 0.0005259
[jit_opt_after_cconv]: 0.00016961, [1]
    [Cycle 1]: 0.00016097, [11]
        [c_1]: 2.35e-05
        [parameter_eliminate]: 3.76999e-06
        [updatestate_depend_eliminate]: 7.43e-06
        [updatestate_assign_eliminate]: 2.74999e-06
        [updatestate_loads_eliminate]: 2.98e-06
        [cse]: 2.695e-05
        [call_graph_tuple_transform]: 2.454e-05
        [tuple_list_get_item_eliminator]: 5.89e-06
        [none_parameter_eliminate]: 2.04e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 6.79999e-06
[remove_dup_value]: 1.801e-05
[partial_unused_args_eliminate]: 2.63e-06
[environ_conv]: 6.61e-06
[add_recomputation]: 5.718e-05
[cse_after_recomputation]: 2.459e-05, [1]
    [Cycle 1]: 1.791e-05, [1]
        [cse]: 1.199e-05
[auto_monad_reorder]: 1.661e-05
[get_jit_bprop_graph]: 2.36e-06
[rewriter_after_jit_bprop_graph]: 8.62e-06
[opt_after_jit_grad]: 0.00058204
[symbol_engine_optimizer]: 8.536e-05, [1]
    [Cycle 1]: 7.767e-05, [6]
        [build]: 6.60002e-06
        [elim_shapecalc]: 9.29998e-06
        [elim_not_effective]: 1.486e-05
        [opt_reshape]: 6.21e-06
        [fold_const_symbol]: 8.79e-06
        [renormalize]: 7.89994e-07
[validate]: 4.294e-05
Sums
    bootstrap                            :     0.000378s :  2.63%
    type_inference                       :     0.008799s : 61.31%
    event_method                         :     0.000024s :  0.17%
    auto_monad                           :     0.000089s :  0.62%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000093s :  0.65%
    rewriter_before_opt_a                :     0.000089s :  0.62%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000154s :  1.07%
    jit_opt_a.loop_unroll                :     0.000041s :  0.29%
    jit_opt_a.a_1                        :     0.001008s :  7.02%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.08%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.13%
    jit_opt_a.inline_without_move        :     0.000012s :  0.08%
    jit_opt_a.renormalize                :     0.000896s :  6.24%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.17%
    jit_opt_a.cse                        :     0.000045s :  0.31%
    jit_opt_a.replace_applicator         :     0.000021s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.10%
    rewriter_after_opt_a                 :     0.000040s :  0.28%
    convert_after_rewriter               :     0.000009s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000882s :  6.14%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.13%
    cconv                                :     0.000029s :  0.20%
    loop_unroll                          :     0.000526s :  3.66%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000027s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.05%
    remove_dup_value                     :     0.000018s :  0.13%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000007s :  0.05%
    add_recomputation                    :     0.000057s :  0.40%
    cse_after_recomputation.cse          :     0.000012s :  0.08%
    auto_monad_reorder                   :     0.000017s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.06%
    opt_after_jit_grad                   :     0.000582s :  4.06%
    symbol_engine_optimizer.build        :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.30%


Time group info:
------[substitution.]   0.000448    26
 0.44% :     0.000002s :      2: substitution.elim_not_effective
 0.30% :     0.000001s :      2: substitution.fold_const_symbol
 1.33% :     0.000006s :      3: substitution.graph_param_transform
89.15% :     0.000400s :      6: substitution.inline
 1.01% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.05% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.17% :     0.000005s :      2: substitution.replace_old_param
 2.21% :     0.000010s :      1: substitution.switch_simplify
 3.33% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008719     2
84.39% :     0.007358s :      1: type_inference.infer
15.61% :     0.001361s :      1: type_inference.specialize
------[replace.]   0.000167     9
48.42% :     0.000081s :      6: replace.inline
30.53% :     0.000051s :      1: replace.switch_simplify
21.05% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000418     9
94.53% :     0.000395s :      6: match.inline
 2.20% :     0.000009s :      1: match.switch_simplify
 3.27% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000151   801
 1.21% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.50% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000002s :     12: predicate.addn_check_dump
 1.23% :     0.000002s :     12: predicate.addn_zero_filter
 2.44% :     0.000004s :     12: predicate.arithmetic_simplify
 1.46% :     0.000002s :     12: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000002s :     12: predicate.compare_switch_simplify
 2.29% :     0.000003s :     12: predicate.depend_value_elim
 1.15% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000001s :      3: predicate.elim_not_effective
 0.75% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.05% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     12: predicate.environ_get_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.17% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.70% :     0.000009s :     26: predicate.inline
 0.77% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.66% :     0.000003s :     17: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.91% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000002s :     12: predicate.merge_addn
 1.05% :     0.000002s :     12: predicate.minmaximum_grad
 1.90% :     0.000003s :      3: predicate.mutable_eliminate
 0.38% :     0.000001s :      3: predicate.opt_reshape
 2.01% :     0.000003s :     17: predicate.partial_eliminate
 1.37% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.79% :     0.000003s :     12: predicate.reduce_eliminate
 1.47% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.44% :     0.000002s :     20: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000002s :     12: predicate.reshape_eliminate
 1.17% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.97% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     12: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.72% :     0.000001s :      6: predicate.special_op_eliminate
 1.22% :     0.000002s :      6: predicate.specialize_transform
 1.50% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.34% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.56% :     0.000004s :     20: predicate.switch_defer_inline
 2.14% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.34% :     0.000013s :     65: predicate.switch_simplify
 1.36% :     0.000002s :     12: predicate.tile_eliminate
 1.22% :     0.000002s :     12: predicate.transpose_eliminate
 1.36% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001092    15
45.44% :     0.000496s :      6: func_graph_cloner_run.FuncGraphClonerGraph
20.04% :     0.000219s :      1: func_graph_cloner_run.FuncGraphClonerNode
34.52% :     0.000377s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021289    72
 0.28% :     0.000060s :      1: add_recomputation
 0.43% :     0.000093s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 1.90% :     0.000405s :      1: bootstrap
 0.15% :     0.000032s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.13% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.14% :     0.000029s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
31.51% :     0.006708s :      1: jit_opt_a
 0.81% :     0.000173s :      1: jit_opt_after_cconv
 0.34% :     0.000072s :      1: jit_opt_b
 2.52% :     0.000537s :      1: loop_unroll
 4.21% :     0.000896s :      1: mutable_eliminate
 6.27% :     0.001334s :     26: opt.transform.jit_opt_a
 0.26% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000027s :      4: opt.transform.jit_opt_b
 0.07% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.80% :     0.000596s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.45% :     0.000096s :      1: py_interpret_to_execute
 0.08% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000021s :      1: remove_dup_value
 2.40% :     0.000511s :      1: renormalize.infer
 1.76% :     0.000375s :      1: renormalize.specialize
 0.05% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000043s :      1: rewriter_after_opt_a
 0.43% :     0.000092s :      1: rewriter_before_opt_a
 0.41% :     0.000088s :      1: symbol_engine_optimizer
41.43% :     0.008819s :      1: type_inference


TotalTime = 0.0172114, [33]
[bootstrap]: 0.00029154
[type_inference]: 0.00416316
[event_method]: 1.54e-05
[auto_monad]: 4.082e-05
[graph_reusing]: 3.41001e-06
[pre_auto_parallel]: 1.73002e-06
[py_interpret_to_execute]: 2.725e-05
[rewriter_before_opt_a]: 6.053e-05
[expand_dump_flag]: 1.92001e-06
[jit_opt_a]: 0.00469711, [2]
    [Cycle 1]: 0.00138966, [27]
        [switch_simplify]: 4.379e-05
        [loop_unroll]: 2.905e-05
        [a_1]: 0.00047843
        [with_stream_mark]: 1.503e-05
        [recompute_prepare]: 7.08e-06
        [updatestate_depend_eliminate]: 3.37997e-06
        [updatestate_assign_eliminate]: 2.74999e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 5.73002e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.42999e-06
        [meta_shard_fg_expand]: 1.86998e-06
        [get_grad_eliminate_]: 5.04e-06
        [merge_forward]: 3.25e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.386e-05
        [j_node_and_user_rematch]: 9.05999e-06
        [meta_fg_expand]: 2.11998e-06
        [replace_old_param]: 8.89e-06
        [inline_without_move]: 4.97e-06
        [renormalize]: 0.00052589
        [add_forward_monad_depend]: 5.04e-06
        [auto_monad_grad]: 1.81998e-06
        [auto_monad_eliminator]: 1.236e-05
        [cse]: 2.632e-05
        [replace_applicator]: 1.134e-05
    [Cycle 2]: 0.00031817, [27]
        [switch_simplify]: 5.38002e-06
        [loop_unroll]: 5.35999e-06
        [a_1]: 9.093e-05
        [with_stream_mark]: 8.33001e-06
        [recompute_prepare]: 5.29e-06
        [updatestate_depend_eliminate]: 2.56e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 5.19e-06
        [updatestate_useless_node_eliminater]: 4.92999e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.48999e-06
        [merge_forward]: 2.25002e-06
        [cell_reuse_recompute_pass]: 1.72999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.21e-05
        [j_node_and_user_rematch]: 7.38e-06
        [meta_fg_expand]: 1.36002e-06
        [replace_old_param]: 7.01001e-06
        [inline_without_move]: 4.82e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.08001e-06
        [auto_monad_grad]: 8.79983e-07
        [auto_monad_eliminator]: 5.09e-06
        [cse]: 1.001e-05
        [replace_applicator]: 5.04e-06
[py_interpret_to_execute_after_opt_a]: 8.05e-06
[rewriter_after_opt_a]: 3.162e-05
[convert_after_rewriter]: 7.3e-06
[order_py_execute_after_rewriter]: 5.62999e-06
[mutable_eliminate]: 0.0004788
[jit_opt_b]: 4.899e-05, [1]
    [Cycle 1]: 4.248e-05, [2]
        [frontend_op_eliminate]: 1.599e-05
        [inline_after_opt_a]: 1.543e-05
[cconv]: 2.461e-05
[loop_unroll]: 0.00041586
[jit_opt_after_cconv]: 0.00013395, [1]
    [Cycle 1]: 0.00012782, [11]
        [c_1]: 2.008e-05
        [parameter_eliminate]: 2.47001e-06
        [updatestate_depend_eliminate]: 5.12e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.08002e-06
        [cse]: 1.687e-05
        [call_graph_tuple_transform]: 1.709e-05
        [tuple_list_get_item_eliminator]: 5.35999e-06
        [none_parameter_eliminate]: 1.72001e-06
        [renormalize]: 5.19998e-07
        [switch_simplify]: 5.24998e-06
[remove_dup_value]: 1.352e-05
[partial_unused_args_eliminate]: 2.17001e-06
[environ_conv]: 4.99e-06
[add_recomputation]: 4.652e-05
[cse_after_recomputation]: 2.043e-05, [1]
    [Cycle 1]: 1.491e-05, [1]
        [cse]: 9.17999e-06
[auto_monad_reorder]: 1.404e-05
[get_jit_bprop_graph]: 1.74e-06
[rewriter_after_jit_bprop_graph]: 4.88001e-06
[opt_after_jit_grad]: 0.00055049
[symbol_engine_optimizer]: 7.706e-05, [1]
    [Cycle 1]: 7.038e-05, [6]
        [build]: 3.16001e-06
        [elim_shapecalc]: 6.84001e-06
        [elim_not_effective]: 1.271e-05
        [opt_reshape]: 5.45001e-06
        [fold_const_symbol]: 8.67998e-06
        [renormalize]: 5.99975e-07
[validate]: 3.155e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.00580109
[execute]: 7.24001e-06
Sums
    bootstrap                            :     0.000292s :  2.14%
    type_inference                       :     0.004163s : 30.50%
    event_method                         :     0.000015s :  0.11%
    auto_monad                           :     0.000041s :  0.30%
    graph_reusing                        :     0.000003s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000027s :  0.20%
    rewriter_before_opt_a                :     0.000061s :  0.44%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000049s :  0.36%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000569s :  4.17%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000526s :  3.85%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000036s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000032s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000479s :  3.51%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000025s :  0.18%
    loop_unroll                          :     0.000416s :  3.05%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000550s :  4.03%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005801s : 42.50%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000155    23
 1.39% :     0.000002s :      2: substitution.elim_not_effective
 0.98% :     0.000002s :      2: substitution.fold_const_symbol
 3.12% :     0.000005s :      3: substitution.graph_param_transform
79.81% :     0.000124s :      4: substitution.inline
 1.85% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.76% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.39% :     0.000004s :      2: substitution.replace_old_param
 7.70% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004115     2
81.96% :     0.003373s :      1: type_inference.infer
18.04% :     0.000742s :      1: type_inference.specialize
------[replace.]   0.000074     6
52.04% :     0.000038s :      4: replace.inline
47.96% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000132     6
91.81% :     0.000122s :      4: match.inline
 8.19% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.33% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     10: predicate.addn_check_dump
 1.44% :     0.000002s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.33% :     0.000001s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000001s :     10: predicate.depend_value_elim
 1.28% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.54% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.39% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.33% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.49% :     0.000002s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.89% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000000s :      3: predicate.opt_reshape
 2.18% :     0.000002s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.86% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.33% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.58% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.34% :     0.000008s :     53: predicate.switch_simplify
 1.30% :     0.000001s :     10: predicate.tile_eliminate
 1.25% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.62% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.77% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000524    12
48.86% :     0.000256s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.21% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.93% :     0.000241s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018608    76
 0.27% :     0.000049s :      1: add_recomputation
 0.24% :     0.000044s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.64% :     0.000306s :      1: bootstrap
 0.15% :     0.000027s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
25.26% :     0.004700s :      1: jit_opt_a
 0.74% :     0.000137s :      1: jit_opt_after_cconv
 0.28% :     0.000052s :      1: jit_opt_b
 2.28% :     0.000424s :      1: loop_unroll
 2.62% :     0.000487s :      1: mutable_eliminate
 4.11% :     0.000764s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.01% :     0.000559s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 1.54% :     0.000286s :      1: renormalize.infer
 1.25% :     0.000233s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000035s :      1: rewriter_after_opt_a
 0.34% :     0.000063s :      1: rewriter_before_opt_a
 0.43% :     0.000080s :      1: symbol_engine_optimizer
31.23% :     0.005812s :      1: task_emit
22.43% :     0.004173s :      1: type_inference
 0.27% :     0.000050s :      1: validate


TotalTime = 0.0261149, [33]
[bootstrap]: 0.00065841
[type_inference]: 0.00732937
[event_method]: 1.878e-05
[auto_monad]: 7.032e-05
[graph_reusing]: 6.39999e-06
[pre_auto_parallel]: 2.64999e-06
[py_interpret_to_execute]: 3.991e-05
[rewriter_before_opt_a]: 7.695e-05
[expand_dump_flag]: 3.23998e-06
[jit_opt_a]: 0.00710149, [2]
    [Cycle 1]: 0.00190908, [27]
        [switch_simplify]: 6.167e-05
        [loop_unroll]: 3.031e-05
        [a_1]: 0.00059155
        [with_stream_mark]: 2.373e-05
        [recompute_prepare]: 9.72999e-06
        [updatestate_depend_eliminate]: 4.13999e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 5.78002e-06
        [updatestate_useless_node_eliminater]: 5.49998e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 2.34999e-06
        [get_grad_eliminate_]: 5.70001e-06
        [merge_forward]: 3.95e-06
        [cell_reuse_recompute_pass]: 1.85001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.638e-05
        [j_node_and_user_rematch]: 9.64999e-06
        [meta_fg_expand]: 2.34999e-06
        [replace_old_param]: 1.089e-05
        [inline_without_move]: 5.82001e-06
        [renormalize]: 0.00080521
        [add_forward_monad_depend]: 1.361e-05
        [auto_monad_grad]: 2.77002e-06
        [auto_monad_eliminator]: 1.854e-05
        [cse]: 3.45e-05
        [replace_applicator]: 1.471e-05
    [Cycle 2]: 0.00035809, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 4.99e-06
        [a_1]: 9.012e-05
        [with_stream_mark]: 1.297e-05
        [recompute_prepare]: 6.34001e-06
        [updatestate_depend_eliminate]: 2.78e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.42e-06
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 5.37001e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 3.06001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.573e-05
        [j_node_and_user_rematch]: 9.36e-06
        [meta_fg_expand]: 2.21e-06
        [replace_old_param]: 7.47002e-06
        [inline_without_move]: 4.98001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 2.37999e-06
        [auto_monad_grad]: 1.15999e-06
        [auto_monad_eliminator]: 9.76e-06
        [cse]: 1.42e-05
        [replace_applicator]: 6.34001e-06
[py_interpret_to_execute_after_opt_a]: 1.364e-05
[rewriter_after_opt_a]: 3.867e-05
[convert_after_rewriter]: 7.32002e-06
[order_py_execute_after_rewriter]: 5.17e-06
[mutable_eliminate]: 0.00071567
[jit_opt_b]: 5.199e-05, [1]
    [Cycle 1]: 4.466e-05, [2]
        [frontend_op_eliminate]: 1.793e-05
        [inline_after_opt_a]: 1.491e-05
[cconv]: 2.873e-05
[loop_unroll]: 0.00058562
[jit_opt_after_cconv]: 0.00016614, [1]
    [Cycle 1]: 0.00015726, [11]
        [c_1]: 2.617e-05
        [parameter_eliminate]: 4.11001e-06
        [updatestate_depend_eliminate]: 7.71999e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [cse]: 2.394e-05
        [call_graph_tuple_transform]: 2.105e-05
        [tuple_list_get_item_eliminator]: 6.49001e-06
        [none_parameter_eliminate]: 2.14999e-06
        [renormalize]: 7.09988e-07
        [switch_simplify]: 6.84999e-06
[remove_dup_value]: 1.403e-05
[partial_unused_args_eliminate]: 1.91998e-06
[environ_conv]: 5.27001e-06
[add_recomputation]: 6.542e-05
[cse_after_recomputation]: 2.919e-05, [1]
    [Cycle 1]: 2.188e-05, [1]
        [cse]: 1.291e-05
[auto_monad_reorder]: 1.636e-05
[get_jit_bprop_graph]: 2.19001e-06
[rewriter_after_jit_bprop_graph]: 6.12001e-06
[opt_after_jit_grad]: 0.00056503
[symbol_engine_optimizer]: 7.763e-05, [1]
    [Cycle 1]: 7.068e-05, [6]
        [build]: 5.45001e-06
        [elim_shapecalc]: 8.15999e-06
        [elim_not_effective]: 1.366e-05
        [opt_reshape]: 5.58002e-06
        [fold_const_symbol]: 8.86002e-06
        [renormalize]: 4.89992e-07
[validate]: 4.322e-05
[backend_pass]: 8.90024e-07
[task_emit]: 0.00809837
[execute]: 8.76002e-06
Sums
    bootstrap                            :     0.000658s :  3.20%
    type_inference                       :     0.007329s : 35.68%
    event_method                         :     0.000019s :  0.09%
    auto_monad                           :     0.000070s :  0.34%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000040s :  0.19%
    rewriter_before_opt_a                :     0.000077s :  0.37%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.33%
    jit_opt_a.loop_unroll                :     0.000035s :  0.17%
    jit_opt_a.a_1                        :     0.000682s :  3.32%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.000805s :  3.92%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.14%
    jit_opt_a.cse                        :     0.000049s :  0.24%
    jit_opt_a.replace_applicator         :     0.000021s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.07%
    rewriter_after_opt_a                 :     0.000039s :  0.19%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000716s :  3.48%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.07%
    cconv                                :     0.000029s :  0.14%
    loop_unroll                          :     0.000586s :  2.85%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000014s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000065s :  0.32%
    cse_after_recomputation.cse          :     0.000013s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000565s :  2.75%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000043s :  0.21%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008098s : 39.42%
    execute                              :     0.000009s :  0.04%


Time group info:
------[substitution.]   0.000242    23
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.24% :     0.000005s :      3: substitution.graph_param_transform
85.15% :     0.000206s :      4: substitution.inline
 1.36% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.32% :     0.000006s :      2: substitution.replace_old_param
 5.38% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007239     2
84.99% :     0.006152s :      1: type_inference.infer
15.01% :     0.001087s :      1: type_inference.specialize
------[replace.]   0.000079     6
65.11% :     0.000052s :      4: replace.inline
34.89% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000215     6
94.49% :     0.000203s :      4: match.inline
 5.51% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.55% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.42% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000002s :     10: predicate.addn_zero_filter
 2.57% :     0.000003s :     10: predicate.arithmetic_simplify
 1.12% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     10: predicate.compare_switch_simplify
 1.65% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.09% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.63% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.06% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.83% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.49% :     0.000007s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000002s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 2.14% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.69% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.40% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.63% :     0.000003s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000003s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.33% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.51% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.36% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.05% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.10% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.07% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.57% :     0.000005s :     16: predicate.switch_defer_inline
 1.93% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.48% :     0.000011s :     53: predicate.switch_simplify
 1.08% :     0.000001s :     10: predicate.tile_eliminate
 1.04% :     0.000001s :     10: predicate.transpose_eliminate
 1.12% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.97% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.26% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.13% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.30% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000947    12
54.98% :     0.000521s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.91% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.11% :     0.000380s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027943    76
 0.25% :     0.000070s :      1: add_recomputation
 0.27% :     0.000074s :      1: auto_monad
 0.07% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 2.44% :     0.000681s :      1: bootstrap
 0.11% :     0.000031s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000023s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
25.42% :     0.007105s :      1: jit_opt_a
 0.61% :     0.000170s :      1: jit_opt_after_cconv
 0.20% :     0.000055s :      1: jit_opt_b
 2.14% :     0.000599s :      1: loop_unroll
 2.61% :     0.000729s :      1: mutable_eliminate
 3.26% :     0.000912s :     26: opt.transform.jit_opt_a
 0.20% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.07% :     0.000578s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.15% :     0.000043s :      1: py_interpret_to_execute
 0.06% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000016s :      1: remove_dup_value
 1.52% :     0.000424s :      1: renormalize.infer
 1.33% :     0.000372s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000043s :      1: rewriter_after_opt_a
 0.29% :     0.000080s :      1: rewriter_before_opt_a
 0.29% :     0.000080s :      1: symbol_engine_optimizer
29.04% :     0.008114s :      1: task_emit
26.30% :     0.007349s :      1: type_inference
 0.24% :     0.000068s :      1: validate


TotalTime = 0.02483, [33]
[bootstrap]: 0.00032808
[type_inference]: 0.00658954
[event_method]: 1.831e-05
[auto_monad]: 6.733e-05
[graph_reusing]: 5.89999e-06
[pre_auto_parallel]: 2.84001e-06
[py_interpret_to_execute]: 4.247e-05
[rewriter_before_opt_a]: 7.204e-05
[expand_dump_flag]: 3.35e-06
[jit_opt_a]: 0.00854979, [2]
    [Cycle 1]: 0.00204561, [27]
        [switch_simplify]: 6.789e-05
        [loop_unroll]: 2.825e-05
        [a_1]: 0.00063722
        [with_stream_mark]: 2.589e-05
        [recompute_prepare]: 8.94003e-06
        [updatestate_depend_eliminate]: 4.13999e-06
        [updatestate_assign_eliminate]: 3.61001e-06
        [updatestate_loads_eliminate]: 2.89999e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 6.17999e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 2.58003e-06
        [get_grad_eliminate_]: 5.73002e-06
        [merge_forward]: 3.58e-06
        [cell_reuse_recompute_pass]: 1.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.611e-05
        [j_node_and_user_rematch]: 9.71998e-06
        [meta_fg_expand]: 2.24999e-06
        [replace_old_param]: 1.132e-05
        [inline_without_move]: 6.33e-06
        [renormalize]: 0.00093035
        [add_forward_monad_depend]: 6.78e-06
        [auto_monad_grad]: 2.56998e-06
        [auto_monad_eliminator]: 1.71e-05
        [cse]: 3.12e-05
        [replace_applicator]: 1.493e-05
    [Cycle 2]: 0.00034452, [27]
        [switch_simplify]: 6.54999e-06
        [loop_unroll]: 6.41e-06
        [a_1]: 9.91e-05
        [with_stream_mark]: 1.134e-05
        [recompute_prepare]: 4.89e-06
        [updatestate_depend_eliminate]: 2.66e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 5.15001e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 1.59e-06
        [get_grad_eliminate_]: 4.45999e-06
        [merge_forward]: 2.99999e-06
        [cell_reuse_recompute_pass]: 2.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.332e-05
        [j_node_and_user_rematch]: 8.92e-06
        [meta_fg_expand]: 1.93002e-06
        [replace_old_param]: 8.29002e-06
        [inline_without_move]: 4.99e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 5.43997e-06
        [cse]: 1.1e-05
        [replace_applicator]: 4.79e-06
[py_interpret_to_execute_after_opt_a]: 1.14e-05
[rewriter_after_opt_a]: 3.842e-05
[convert_after_rewriter]: 7.15e-06
[order_py_execute_after_rewriter]: 4.72e-06
[mutable_eliminate]: 0.00074466
[jit_opt_b]: 5.087e-05, [1]
    [Cycle 1]: 4.359e-05, [2]
        [frontend_op_eliminate]: 1.683e-05
        [inline_after_opt_a]: 1.537e-05
[cconv]: 2.709e-05
[loop_unroll]: 0.00042937
[jit_opt_after_cconv]: 0.00014098, [1]
    [Cycle 1]: 0.00013442, [11]
        [c_1]: 2.126e-05
        [parameter_eliminate]: 2.93e-06
        [updatestate_depend_eliminate]: 6.43998e-06
        [updatestate_assign_eliminate]: 2.49999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [cse]: 1.793e-05
        [call_graph_tuple_transform]: 1.935e-05
        [tuple_list_get_item_eliminator]: 5.44998e-06
        [none_parameter_eliminate]: 1.87999e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 5.36002e-06
[remove_dup_value]: 1.539e-05
[partial_unused_args_eliminate]: 2.20002e-06
[environ_conv]: 5.35001e-06
[add_recomputation]: 4.995e-05
[cse_after_recomputation]: 2.139e-05, [1]
    [Cycle 1]: 1.56e-05, [1]
        [cse]: 9.15001e-06
[auto_monad_reorder]: 1.527e-05
[get_jit_bprop_graph]: 2.53e-06
[rewriter_after_jit_bprop_graph]: 5.52001e-06
[opt_after_jit_grad]: 0.00049442
[symbol_engine_optimizer]: 7.594e-05, [1]
    [Cycle 1]: 6.951e-05, [6]
        [build]: 4.59998e-06
        [elim_shapecalc]: 8e-06
        [elim_not_effective]: 1.311e-05
        [opt_reshape]: 5.94999e-06
        [fold_const_symbol]: 8.93002e-06
        [renormalize]: 8.29983e-07
[validate]: 4.18e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.006718
[execute]: 1.088e-05
Sums
    bootstrap                            :     0.000328s :  1.82%
    type_inference                       :     0.006590s : 36.59%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000067s :  0.37%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000042s :  0.24%
    rewriter_before_opt_a                :     0.000072s :  0.40%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.41%
    jit_opt_a.loop_unroll                :     0.000035s :  0.19%
    jit_opt_a.a_1                        :     0.000736s :  4.09%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000930s :  5.17%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.13%
    jit_opt_a.cse                        :     0.000042s :  0.23%
    jit_opt_a.replace_applicator         :     0.000020s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000038s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000745s :  4.14%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000027s :  0.15%
    loop_unroll                          :     0.000429s :  2.38%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.28%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.08%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000494s :  2.75%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.23%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006718s : 37.31%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000267    23
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000002s :      2: substitution.fold_const_symbol
 2.18% :     0.000006s :      3: substitution.graph_param_transform
85.43% :     0.000228s :      4: substitution.inline
 1.45% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.98% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.24% :     0.000006s :      2: substitution.replace_old_param
 5.23% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006515     2
84.67% :     0.005516s :      1: type_inference.infer
15.33% :     0.000998s :      1: type_inference.specialize
------[replace.]   0.000077     6
68.54% :     0.000053s :      4: replace.inline
31.46% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000238     6
94.60% :     0.000226s :      4: match.inline
 5.40% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.54% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.38% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 2.42% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.41% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.42% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.38% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.79% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.25% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.79% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.92% :     0.000008s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.20% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.46% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.02% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 2.04% :     0.000003s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     15: predicate.partial_eliminate
 1.34% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.61% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.50% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.68% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.19% :     0.000003s :     16: predicate.switch_defer_inline
 2.01% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.25% :     0.000009s :     53: predicate.switch_simplify
 1.37% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.07% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.37% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.79% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.65% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.45% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000735    12
49.11% :     0.000361s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.61% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.27% :     0.000318s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026811    76
 0.20% :     0.000053s :      1: add_recomputation
 0.26% :     0.000071s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.30% :     0.000348s :      1: bootstrap
 0.11% :     0.000030s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000023s :      1: event_method
 0.06% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
31.90% :     0.008553s :      1: jit_opt_a
 0.54% :     0.000144s :      1: jit_opt_after_cconv
 0.20% :     0.000053s :      1: jit_opt_b
 1.63% :     0.000437s :      1: loop_unroll
 2.81% :     0.000754s :      1: mutable_eliminate
 3.62% :     0.000970s :     26: opt.transform.jit_opt_a
 0.18% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000026s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.88% :     0.000504s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000045s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 1.93% :     0.000517s :      1: renormalize.infer
 1.45% :     0.000389s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000041s :      1: rewriter_after_opt_a
 0.28% :     0.000075s :      1: rewriter_before_opt_a
 0.29% :     0.000078s :      1: symbol_engine_optimizer
25.14% :     0.006742s :      1: task_emit
24.64% :     0.006607s :      1: type_inference
 0.24% :     0.000066s :      1: validate


TotalTime = 0.0138474, [30]
[bootstrap]: 0.00048975
[type_inference]: 0.00637426
[event_method]: 1.998e-05
[auto_monad]: 6.027e-05
[graph_reusing]: 4.63999e-06
[pre_auto_parallel]: 2.16e-06
[py_interpret_to_execute]: 4.445e-05
[rewriter_before_opt_a]: 8.282e-05
[expand_dump_flag]: 2.21e-06
[jit_opt_a]: 0.00481941, [2]
    [Cycle 1]: 0.00155929, [27]
        [switch_simplify]: 9.806e-05
        [loop_unroll]: 3.655e-05
        [a_1]: 0.00065073
        [with_stream_mark]: 1.196e-05
        [recompute_prepare]: 7.41001e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.18001e-06
        [specialize_transform]: 5.86e-06
        [updatestate_useless_node_eliminater]: 5.20001e-06
        [accelerated_algorithm]: 5.82999e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 5.27999e-06
        [merge_forward]: 2.98998e-06
        [cell_reuse_recompute_pass]: 1.10999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.204e-05
        [j_node_and_user_rematch]: 8.88002e-06
        [meta_fg_expand]: 1.91003e-06
        [replace_old_param]: 7.61999e-06
        [inline_without_move]: 5.60001e-06
        [renormalize]: 0.00047594
        [add_forward_monad_depend]: 4.41002e-06
        [auto_monad_grad]: 1.38002e-06
        [auto_monad_eliminator]: 9.51998e-06
        [cse]: 1.589e-05
        [replace_applicator]: 1.058e-05
    [Cycle 2]: 0.00031459, [27]
        [switch_simplify]: 5.56e-06
        [loop_unroll]: 5.02e-06
        [a_1]: 8.953e-05
        [with_stream_mark]: 8.35999e-06
        [recompute_prepare]: 5.24998e-06
        [updatestate_depend_eliminate]: 2.81e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 5.15999e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 4.62e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.42001e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.266e-05
        [j_node_and_user_rematch]: 7.83999e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 7.3e-06
        [inline_without_move]: 4.69998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 5.12999e-06
        [cse]: 9.38002e-06
        [replace_applicator]: 4.89998e-06
[py_interpret_to_execute_after_opt_a]: 7.48999e-06
[rewriter_after_opt_a]: 2.468e-05
[convert_after_rewriter]: 5.73002e-06
[order_py_execute_after_rewriter]: 3.94002e-06
[mutable_eliminate]: 0.0004608
[jit_opt_b]: 5.502e-05, [1]
    [Cycle 1]: 4.153e-05, [2]
        [frontend_op_eliminate]: 1.58e-05
        [inline_after_opt_a]: 1.453e-05
[cconv]: 1.614e-05
[loop_unroll]: 0.00040851
[jit_opt_after_cconv]: 0.0001372, [1]
    [Cycle 1]: 0.0001309, [11]
        [c_1]: 2.129e-05
        [parameter_eliminate]: 3.01999e-06
        [updatestate_depend_eliminate]: 5.32001e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.724e-05
        [call_graph_tuple_transform]: 1.708e-05
        [tuple_list_get_item_eliminator]: 5.30999e-06
        [none_parameter_eliminate]: 8.89995e-07
        [renormalize]: 4.90021e-07
        [switch_simplify]: 5.36002e-06
[remove_dup_value]: 6.31e-06
[partial_unused_args_eliminate]: 1.59998e-06
[environ_conv]: 3.56001e-06
[add_recomputation]: 3.19e-05
[cse_after_recomputation]: 2.038e-05, [1]
    [Cycle 1]: 1.516e-05, [1]
        [cse]: 9.17999e-06
[auto_monad_reorder]: 9.22999e-06
[get_jit_bprop_graph]: 1.53002e-06
[rewriter_after_jit_bprop_graph]: 4.56002e-06
[opt_after_jit_grad]: 0.0004626
[symbol_engine_optimizer]: 7.234e-05, [1]
    [Cycle 1]: 6.623e-05, [6]
        [build]: 2.69999e-06
        [elim_shapecalc]: 7.62002e-06
        [elim_not_effective]: 1.258e-05
        [opt_reshape]: 5.71e-06
        [fold_const_symbol]: 8.57998e-06
        [renormalize]: 5.90022e-07
[validate]: 4.937e-05
Sums
    bootstrap                            :     0.000490s :  4.74%
    type_inference                       :     0.006374s : 61.69%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000060s :  0.58%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000044s :  0.43%
    rewriter_before_opt_a                :     0.000083s :  0.80%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000104s :  1.00%
    jit_opt_a.loop_unroll                :     0.000042s :  0.40%
    jit_opt_a.a_1                        :     0.000740s :  7.16%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000476s :  4.61%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.14%
    jit_opt_a.cse                        :     0.000025s :  0.24%
    jit_opt_a.replace_applicator         :     0.000015s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.07%
    rewriter_after_opt_a                 :     0.000025s :  0.24%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000461s :  4.46%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.14%
    cconv                                :     0.000016s :  0.16%
    loop_unroll                          :     0.000409s :  3.95%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000006s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000032s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000009s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000463s :  4.48%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000049s :  0.48%


Time group info:
------[substitution.]   0.000226    26
 0.73% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 1.55% :     0.000003s :      3: substitution.graph_param_transform
85.88% :     0.000194s :      6: substitution.inline
 1.15% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.90% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.11% :     0.000003s :      2: substitution.replace_old_param
 3.44% :     0.000008s :      1: substitution.switch_simplify
 3.59% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006321     2
85.15% :     0.005382s :      1: type_inference.infer
14.85% :     0.000939s :      1: type_inference.specialize
------[replace.]   0.000117     9
55.94% :     0.000065s :      6: replace.inline
18.27% :     0.000021s :      1: replace.switch_simplify
25.79% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000204     9
93.21% :     0.000190s :      6: match.inline
 3.43% :     0.000007s :      1: match.switch_simplify
 3.36% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   801
 1.53% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.42% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.24% :     0.000002s :     12: predicate.addn_check_dump
 1.39% :     0.000002s :     12: predicate.addn_zero_filter
 1.94% :     0.000003s :     12: predicate.arithmetic_simplify
 1.28% :     0.000002s :     12: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.22% :     0.000002s :     12: predicate.compare_switch_simplify
 1.48% :     0.000002s :     12: predicate.depend_value_elim
 1.28% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.26% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.22% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     12: predicate.environ_get_eliminate
 1.22% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.81% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 5.17% :     0.000007s :     26: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     17: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.43% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.22% :     0.000002s :     12: predicate.merge_addn
 1.27% :     0.000002s :     12: predicate.minmaximum_grad
 1.42% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     17: predicate.partial_eliminate
 1.23% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     12: predicate.reduce_eliminate
 1.51% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     20: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.58% :     0.000002s :     12: predicate.reshape_eliminate
 1.29% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.77% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.55% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.39% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.82% :     0.000004s :     20: predicate.switch_defer_inline
 2.43% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.29% :     0.000011s :     65: predicate.switch_simplify
 1.25% :     0.000002s :     12: predicate.tile_eliminate
 1.45% :     0.000002s :     12: predicate.transpose_eliminate
 1.59% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.44% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000738    15
52.02% :     0.000384s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.53% :     0.000078s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.44% :     0.000276s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015358    72
 0.23% :     0.000035s :      1: add_recomputation
 0.42% :     0.000064s :      1: auto_monad
 0.08% :     0.000012s :      1: auto_monad_reorder
 3.32% :     0.000510s :      1: bootstrap
 0.12% :     0.000019s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
31.40% :     0.004822s :      1: jit_opt_a
 0.91% :     0.000140s :      1: jit_opt_after_cconv
 0.38% :     0.000058s :      1: jit_opt_b
 2.71% :     0.000416s :      1: loop_unroll
 3.05% :     0.000469s :      1: mutable_eliminate
 6.46% :     0.000992s :     26: opt.transform.jit_opt_a
 0.30% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.07% :     0.000471s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.31% :     0.000047s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000009s :      1: remove_dup_value
 1.56% :     0.000240s :      1: renormalize.infer
 1.50% :     0.000230s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000028s :      1: rewriter_after_opt_a
 0.56% :     0.000086s :      1: rewriter_before_opt_a
 0.49% :     0.000075s :      1: symbol_engine_optimizer
41.57% :     0.006384s :      1: type_inference


TotalTime = 0.0255905, [33]
[bootstrap]: 0.00027202
[type_inference]: 0.00437012
[event_method]: 1.699e-05
[auto_monad]: 6.189e-05
[graph_reusing]: 5.42001e-06
[pre_auto_parallel]: 2.58e-06
[py_interpret_to_execute]: 3.442e-05
[rewriter_before_opt_a]: 7.116e-05
[expand_dump_flag]: 3.61001e-06
[jit_opt_a]: 0.00575705, [2]
    [Cycle 1]: 0.00138457, [27]
        [switch_simplify]: 5.626e-05
        [loop_unroll]: 2.98e-05
        [a_1]: 0.000496
        [with_stream_mark]: 1.566e-05
        [recompute_prepare]: 7.02002e-06
        [updatestate_depend_eliminate]: 3.71999e-06
        [updatestate_assign_eliminate]: 2.84999e-06
        [updatestate_loads_eliminate]: 2.76e-06
        [parameter_eliminate]: 1.93002e-06
        [specialize_transform]: 5.49998e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 4.92e-06
        [meta_shard_fg_expand]: 1.96e-06
        [get_grad_eliminate_]: 4.85999e-06
        [merge_forward]: 3.96001e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.368e-05
        [j_node_and_user_rematch]: 8.59e-06
        [meta_fg_expand]: 2.36e-06
        [replace_old_param]: 8.30999e-06
        [inline_without_move]: 5.25999e-06
        [renormalize]: 0.00048506
        [add_forward_monad_depend]: 4.92e-06
        [auto_monad_grad]: 2.53998e-06
        [auto_monad_eliminator]: 1.314e-05
        [cse]: 2.677e-05
        [replace_applicator]: 1.153e-05
    [Cycle 2]: 0.00031993, [27]
        [switch_simplify]: 5.61003e-06
        [loop_unroll]: 4.72998e-06
        [a_1]: 9.34e-05
        [with_stream_mark]: 9.37001e-06
        [recompute_prepare]: 5.29998e-06
        [updatestate_depend_eliminate]: 2.37999e-06
        [updatestate_assign_eliminate]: 1.91e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 4.75999e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 4.66002e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 4.35999e-06
        [merge_forward]: 2.24001e-06
        [cell_reuse_recompute_pass]: 1.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.166e-05
        [j_node_and_user_rematch]: 7.99002e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 7.46001e-06
        [inline_without_move]: 5.16002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 9.10019e-07
        [auto_monad_eliminator]: 5.23002e-06
        [cse]: 9.86998e-06
        [replace_applicator]: 4.76002e-06
[py_interpret_to_execute_after_opt_a]: 8.42998e-06
[rewriter_after_opt_a]: 3.178e-05
[convert_after_rewriter]: 6.96001e-06
[order_py_execute_after_rewriter]: 4.77e-06
[mutable_eliminate]: 0.00048503
[jit_opt_b]: 4.755e-05, [1]
    [Cycle 1]: 4.107e-05, [2]
        [frontend_op_eliminate]: 1.551e-05
        [inline_after_opt_a]: 1.446e-05
[cconv]: 2.487e-05
[loop_unroll]: 0.00043731
[jit_opt_after_cconv]: 0.00013312, [1]
    [Cycle 1]: 0.00012686, [11]
        [c_1]: 2.095e-05
        [parameter_eliminate]: 2.54001e-06
        [updatestate_depend_eliminate]: 5.02e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.681e-05
        [call_graph_tuple_transform]: 1.751e-05
        [tuple_list_get_item_eliminator]: 5.23002e-06
        [none_parameter_eliminate]: 1.45001e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.07999e-06
[remove_dup_value]: 1.373e-05
[partial_unused_args_eliminate]: 2.26998e-06
[environ_conv]: 4.82e-06
[add_recomputation]: 4.51e-05
[cse_after_recomputation]: 1.911e-05, [1]
    [Cycle 1]: 1.395e-05, [1]
        [cse]: 8.67e-06
[auto_monad_reorder]: 1.396e-05
[get_jit_bprop_graph]: 1.37e-06
[rewriter_after_jit_bprop_graph]: 4.94e-06
[opt_after_jit_grad]: 0.00043972
[symbol_engine_optimizer]: 6.971e-05, [1]
    [Cycle 1]: 6.374e-05, [6]
        [build]: 3.11001e-06
        [elim_shapecalc]: 7.01001e-06
        [elim_not_effective]: 1.236e-05
        [opt_reshape]: 5.46002e-06
        [fold_const_symbol]: 8.72998e-06
        [renormalize]: 5.10016e-07
[validate]: 3.287e-05
[backend_pass]: 8.80013e-07
[task_emit]: 0.0129611
[execute]: 8.29002e-06
Sums
    bootstrap                            :     0.000272s :  1.30%
    type_inference                       :     0.004370s : 20.86%
    event_method                         :     0.000017s :  0.08%
    auto_monad                           :     0.000062s :  0.30%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000034s :  0.16%
    rewriter_before_opt_a                :     0.000071s :  0.34%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.30%
    jit_opt_a.loop_unroll                :     0.000035s :  0.16%
    jit_opt_a.a_1                        :     0.000589s :  2.81%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000010s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.04%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000485s :  2.32%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.09%
    jit_opt_a.cse                        :     0.000037s :  0.17%
    jit_opt_a.replace_applicator         :     0.000016s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.04%
    rewriter_after_opt_a                 :     0.000032s :  0.15%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000485s :  2.31%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.07%
    cconv                                :     0.000025s :  0.12%
    loop_unroll                          :     0.000437s :  2.09%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000014s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000045s :  0.22%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000014s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000440s :  2.10%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000033s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012961s : 61.86%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000176    23
 1.15% :     0.000002s :      2: substitution.elim_not_effective
 0.85% :     0.000001s :      2: substitution.fold_const_symbol
 2.78% :     0.000005s :      3: substitution.graph_param_transform
82.61% :     0.000145s :      4: substitution.inline
 1.48% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.44% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.05% :     0.000004s :      2: substitution.replace_old_param
 6.64% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004308     2
81.87% :     0.003527s :      1: type_inference.infer
18.13% :     0.000781s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.71% :     0.000042s :      4: replace.inline
32.29% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000153     6
93.18% :     0.000143s :      4: match.inline
 6.82% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.24% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.19% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000001s :     10: predicate.addn_zero_filter
 1.97% :     0.000002s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.37% :     0.000006s :     22: predicate.inline
 1.05% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.19% :     0.000002s :     15: predicate.load_eliminater
 1.60% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.17% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.15% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.60% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000000s :      3: predicate.opt_reshape
 2.23% :     0.000002s :     15: predicate.partial_eliminate
 1.31% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.81% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.32% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.51% :     0.000010s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.30% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.91% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.79% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.41% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000531    12
51.89% :     0.000275s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.93% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.18% :     0.000229s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026969    76
 0.18% :     0.000048s :      1: add_recomputation
 0.24% :     0.000065s :      1: auto_monad
 0.06% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.07% :     0.000289s :      1: bootstrap
 0.10% :     0.000027s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.08% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.08% :     0.000021s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
21.36% :     0.005759s :      1: jit_opt_a
 0.50% :     0.000136s :      1: jit_opt_after_cconv
 0.19% :     0.000050s :      1: jit_opt_b
 1.65% :     0.000446s :      1: loop_unroll
 1.83% :     0.000494s :      1: mutable_eliminate
 2.94% :     0.000794s :     26: opt.transform.jit_opt_a
 0.17% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000024s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000030s :      4: opt.transform.symbol_engine_opt
 1.66% :     0.000448s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000037s :      1: py_interpret_to_execute
 0.04% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000016s :      1: remove_dup_value
 0.89% :     0.000240s :      1: renormalize.infer
 0.89% :     0.000239s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000035s :      1: rewriter_after_opt_a
 0.27% :     0.000074s :      1: rewriter_before_opt_a
 0.27% :     0.000072s :      1: symbol_engine_optimizer
48.13% :     0.012979s :      1: task_emit
16.25% :     0.004383s :      1: type_inference
 0.19% :     0.000052s :      1: validate


TotalTime = 0.0176473, [33]
[bootstrap]: 0.00025184
[type_inference]: 0.00402898
[event_method]: 1.599e-05
[auto_monad]: 6.313e-05
[graph_reusing]: 5.61003e-06
[pre_auto_parallel]: 2.68998e-06
[py_interpret_to_execute]: 3.31e-05
[rewriter_before_opt_a]: 7.003e-05
[expand_dump_flag]: 2.80002e-06
[jit_opt_a]: 0.00483649, [2]
    [Cycle 1]: 0.00146279, [27]
        [switch_simplify]: 4.931e-05
        [loop_unroll]: 2.911e-05
        [a_1]: 0.00050568
        [with_stream_mark]: 1.602e-05
        [recompute_prepare]: 7.78001e-06
        [updatestate_depend_eliminate]: 3.55998e-06
        [updatestate_assign_eliminate]: 2.96999e-06
        [updatestate_loads_eliminate]: 2.93e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 6.58998e-06
        [updatestate_useless_node_eliminater]: 5.50001e-06
        [accelerated_algorithm]: 5.81e-06
        [meta_shard_fg_expand]: 1.96e-06
        [get_grad_eliminate_]: 5.54998e-06
        [merge_forward]: 3.48999e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.404e-05
        [j_node_and_user_rematch]: 8.79e-06
        [meta_fg_expand]: 2.44001e-06
        [replace_old_param]: 8.50001e-06
        [inline_without_move]: 5.25999e-06
        [renormalize]: 0.00050322
        [add_forward_monad_depend]: 4.4e-06
        [auto_monad_grad]: 1.91e-06
        [auto_monad_eliminator]: 1.223e-05
        [cse]: 2.708e-05
        [replace_applicator]: 1.092e-05
    [Cycle 2]: 0.00030908, [27]
        [switch_simplify]: 5.57001e-06
        [loop_unroll]: 4.81002e-06
        [a_1]: 9.281e-05
        [with_stream_mark]: 7.92e-06
        [recompute_prepare]: 5.09e-06
        [updatestate_depend_eliminate]: 2.66999e-06
        [updatestate_assign_eliminate]: 1.91e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 9.00007e-07
        [specialize_transform]: 4.87e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 4.86002e-06
        [meta_shard_fg_expand]: 9.39996e-07
        [get_grad_eliminate_]: 4.47998e-06
        [merge_forward]: 2.27001e-06
        [cell_reuse_recompute_pass]: 1.46998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.146e-05
        [j_node_and_user_rematch]: 7.79002e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 6.89999e-06
        [inline_without_move]: 4.80001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 4.32e-06
        [cse]: 9.19e-06
        [replace_applicator]: 4.90999e-06
[py_interpret_to_execute_after_opt_a]: 8.00999e-06
[rewriter_after_opt_a]: 2.941e-05
[convert_after_rewriter]: 6.43e-06
[order_py_execute_after_rewriter]: 4.78001e-06
[mutable_eliminate]: 0.00046233
[jit_opt_b]: 4.836e-05, [1]
    [Cycle 1]: 4.191e-05, [2]
        [frontend_op_eliminate]: 1.562e-05
        [inline_after_opt_a]: 1.55e-05
[cconv]: 2.255e-05
[loop_unroll]: 0.00041927
[jit_opt_after_cconv]: 0.00013292, [1]
    [Cycle 1]: 0.00012669, [11]
        [c_1]: 2.093e-05
        [parameter_eliminate]: 2.59999e-06
        [updatestate_depend_eliminate]: 5.20001e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 1.96003e-06
        [cse]: 1.532e-05
        [call_graph_tuple_transform]: 1.717e-05
        [tuple_list_get_item_eliminator]: 5.47999e-06
        [none_parameter_eliminate]: 1.43002e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 6.01998e-06
[remove_dup_value]: 1.203e-05
[partial_unused_args_eliminate]: 2.24001e-06
[environ_conv]: 4.60999e-06
[add_recomputation]: 4.546e-05
[cse_after_recomputation]: 3.039e-05, [1]
    [Cycle 1]: 2.49e-05, [1]
        [cse]: 9.09e-06
[auto_monad_reorder]: 1.429e-05
[get_jit_bprop_graph]: 1.49e-06
[rewriter_after_jit_bprop_graph]: 5.72001e-06
[opt_after_jit_grad]: 0.00046608
[symbol_engine_optimizer]: 7.34e-05, [1]
    [Cycle 1]: 6.713e-05, [6]
        [build]: 3.79002e-06
        [elim_shapecalc]: 7.54002e-06
        [elim_not_effective]: 1.283e-05
        [opt_reshape]: 5.61998e-06
        [fold_const_symbol]: 9.00999e-06
        [renormalize]: 4.59986e-07
[validate]: 3.412e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.00631948
[execute]: 8.88002e-06
Sums
    bootstrap                            :     0.000252s :  1.81%
    type_inference                       :     0.004029s : 28.88%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000063s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.24%
    rewriter_before_opt_a                :     0.000070s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000598s :  4.29%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000503s :  3.61%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.12%
    jit_opt_a.cse                        :     0.000036s :  0.26%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000029s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000462s :  3.31%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000023s :  0.16%
    loop_unroll                          :     0.000419s :  3.01%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000045s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000466s :  3.34%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006319s : 45.30%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000179    23
 1.21% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.56% :     0.000005s :      3: substitution.graph_param_transform
82.76% :     0.000148s :      4: substitution.inline
 1.55% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.48% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.91% :     0.000003s :      2: substitution.replace_old_param
 6.85% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003970     2
79.96% :     0.003174s :      1: type_inference.infer
20.04% :     0.000796s :      1: type_inference.specialize
------[replace.]   0.000064     6
66.75% :     0.000043s :      4: replace.inline
33.25% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000156     6
93.11% :     0.000145s :      4: match.inline
 6.89% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.32% :     0.000002s :     10: predicate.addn_check_dump
 1.44% :     0.000002s :     10: predicate.addn_zero_filter
 1.91% :     0.000002s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.18% :     0.000001s :     10: predicate.compare_switch_simplify
 1.53% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.17% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.70% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.18% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.18% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000001s :      6: predicate.less_batch_normalization
 1.73% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.64% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.92% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.29% :     0.000002s :     10: predicate.merge_addn
 1.19% :     0.000001s :     10: predicate.minmaximum_grad
 1.67% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000002s :     15: predicate.partial_eliminate
 1.31% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.63% :     0.000002s :     10: predicate.reshape_eliminate
 1.35% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.51% :     0.000009s :     53: predicate.switch_simplify
 1.34% :     0.000002s :     10: predicate.tile_eliminate
 1.26% :     0.000001s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.92% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000514    12
45.90% :     0.000236s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.11% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.00% :     0.000252s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019057    76
 0.25% :     0.000048s :      1: add_recomputation
 0.35% :     0.000067s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.38% :     0.000263s :      1: bootstrap
 0.13% :     0.000025s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000033s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.39% :     0.004839s :      1: jit_opt_a
 0.71% :     0.000135s :      1: jit_opt_after_cconv
 0.27% :     0.000051s :      1: jit_opt_b
 2.24% :     0.000427s :      1: loop_unroll
 2.47% :     0.000470s :      1: mutable_eliminate
 4.19% :     0.000799s :     26: opt.transform.jit_opt_a
 0.24% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.49% :     0.000475s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000036s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 1.32% :     0.000252s :      1: renormalize.infer
 1.28% :     0.000244s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000032s :      1: rewriter_after_opt_a
 0.38% :     0.000073s :      1: rewriter_before_opt_a
 0.40% :     0.000076s :      1: symbol_engine_optimizer
33.24% :     0.006335s :      1: task_emit
21.20% :     0.004040s :      1: type_inference
 0.27% :     0.000052s :      1: validate


TotalTime = 0.0165593, [33]
[bootstrap]: 0.00025778
[type_inference]: 0.00408383
[event_method]: 1.641e-05
[auto_monad]: 5.708e-05
[graph_reusing]: 5.29e-06
[pre_auto_parallel]: 2.11998e-06
[py_interpret_to_execute]: 2.774e-05
[rewriter_before_opt_a]: 6.486e-05
[expand_dump_flag]: 2.81999e-06
[jit_opt_a]: 0.00463726, [2]
    [Cycle 1]: 0.00132739, [27]
        [switch_simplify]: 4.722e-05
        [loop_unroll]: 2.85e-05
        [a_1]: 0.00047498
        [with_stream_mark]: 1.469e-05
        [recompute_prepare]: 6.66e-06
        [updatestate_depend_eliminate]: 3.36001e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [parameter_eliminate]: 1.94999e-06
        [specialize_transform]: 5.71998e-06
        [updatestate_useless_node_eliminater]: 4.74998e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 5.12999e-06
        [merge_forward]: 3.26001e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.353e-05
        [j_node_and_user_rematch]: 8.60001e-06
        [meta_fg_expand]: 2.16e-06
        [replace_old_param]: 8.36002e-06
        [inline_without_move]: 5.05999e-06
        [renormalize]: 0.00047328
        [add_forward_monad_depend]: 5.51998e-06
        [auto_monad_grad]: 2.17001e-06
        [auto_monad_eliminator]: 1.241e-05
        [cse]: 1.862e-05
        [replace_applicator]: 1.109e-05
    [Cycle 2]: 0.00033311, [27]
        [switch_simplify]: 6.17001e-06
        [loop_unroll]: 5.05999e-06
        [a_1]: 9.202e-05
        [with_stream_mark]: 8.87e-06
        [recompute_prepare]: 4.67e-06
        [updatestate_depend_eliminate]: 2.40002e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 5.07e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 5.00999e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.70999e-06
        [merge_forward]: 2.58998e-06
        [cell_reuse_recompute_pass]: 1.43002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.13e-05
        [j_node_and_user_rematch]: 8.15999e-06
        [meta_fg_expand]: 1.37999e-06
        [replace_old_param]: 6.69999e-06
        [inline_without_move]: 4.51002e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 5.58002e-06
        [cse]: 9.90002e-06
        [replace_applicator]: 5.30999e-06
[py_interpret_to_execute_after_opt_a]: 9.84001e-06
[rewriter_after_opt_a]: 3.234e-05
[convert_after_rewriter]: 6.63e-06
[order_py_execute_after_rewriter]: 5.35001e-06
[mutable_eliminate]: 0.00048482
[jit_opt_b]: 4.519e-05, [1]
    [Cycle 1]: 3.876e-05, [2]
        [frontend_op_eliminate]: 1.482e-05
        [inline_after_opt_a]: 1.316e-05
[cconv]: 2.268e-05
[loop_unroll]: 0.00042125
[jit_opt_after_cconv]: 0.00013103, [1]
    [Cycle 1]: 0.00012498, [11]
        [c_1]: 1.998e-05
        [parameter_eliminate]: 2.37999e-06
        [updatestate_depend_eliminate]: 5.07999e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [cse]: 1.593e-05
        [call_graph_tuple_transform]: 1.723e-05
        [tuple_list_get_item_eliminator]: 4.88001e-06
        [none_parameter_eliminate]: 1.37999e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 5.19998e-06
[remove_dup_value]: 1.302e-05
[partial_unused_args_eliminate]: 1.97001e-06
[environ_conv]: 4.79e-06
[add_recomputation]: 4.023e-05
[cse_after_recomputation]: 1.913e-05, [1]
    [Cycle 1]: 1.41e-05, [1]
        [cse]: 8.75001e-06
[auto_monad_reorder]: 1.463e-05
[get_jit_bprop_graph]: 1.49e-06
[rewriter_after_jit_bprop_graph]: 4.60999e-06
[opt_after_jit_grad]: 0.00045082
[symbol_engine_optimizer]: 6.974e-05, [1]
    [Cycle 1]: 6.414e-05, [6]
        [build]: 3.18e-06
        [elim_shapecalc]: 7.55e-06
        [elim_not_effective]: 1.226e-05
        [opt_reshape]: 5.37999e-06
        [fold_const_symbol]: 8.13999e-06
        [renormalize]: 4.19997e-07
[validate]: 3.238e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.00540159
[execute]: 7.38e-06
Sums
    bootstrap                            :     0.000258s :  1.98%
    type_inference                       :     0.004084s : 31.42%
    event_method                         :     0.000016s :  0.13%
    auto_monad                           :     0.000057s :  0.44%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.21%
    rewriter_before_opt_a                :     0.000065s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.26%
    jit_opt_a.a_1                        :     0.000567s :  4.36%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000473s :  3.64%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.14%
    jit_opt_a.cse                        :     0.000029s :  0.22%
    jit_opt_a.replace_applicator         :     0.000016s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.08%
    rewriter_after_opt_a                 :     0.000032s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000485s :  3.73%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000013s :  0.10%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000421s :  3.24%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000040s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000451s :  3.47%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005402s : 41.55%
    execute                              :     0.000007s :  0.06%


Time group info:
------[substitution.]   0.000162    23
 1.19% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 3.09% :     0.000005s :      3: substitution.graph_param_transform
81.85% :     0.000133s :      4: substitution.inline
 1.99% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.46% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000003s :      2: substitution.replace_old_param
 6.64% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004029     2
81.86% :     0.003298s :      1: type_inference.infer
18.14% :     0.000731s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.70% :     0.000042s :      4: replace.inline
32.30% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000140     6
93.10% :     0.000130s :      4: match.inline
 6.90% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.35% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 2.26% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.45% :     0.000002s :     10: predicate.depend_value_elim
 1.26% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.12% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.78% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.29% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.12% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.25% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.79% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.08% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.66% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000002s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.82% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000001s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.72% :     0.000008s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.55% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000482    12
48.66% :     0.000234s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.50% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.83% :     0.000221s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017897    76
 0.24% :     0.000043s :      1: add_recomputation
 0.34% :     0.000060s :      1: auto_monad
 0.10% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.51% :     0.000270s :      1: bootstrap
 0.14% :     0.000025s :      1: cconv
 0.08% :     0.000015s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
25.93% :     0.004640s :      1: jit_opt_a
 0.75% :     0.000134s :      1: jit_opt_after_cconv
 0.27% :     0.000048s :      1: jit_opt_b
 2.40% :     0.000429s :      1: loop_unroll
 2.76% :     0.000493s :      1: mutable_eliminate
 4.26% :     0.000763s :     26: opt.transform.jit_opt_a
 0.25% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000022s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.56% :     0.000459s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000030s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000015s :      1: remove_dup_value
 1.39% :     0.000250s :      1: renormalize.infer
 1.21% :     0.000217s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000035s :      1: rewriter_after_opt_a
 0.38% :     0.000068s :      1: rewriter_before_opt_a
 0.40% :     0.000072s :      1: symbol_engine_optimizer
30.25% :     0.005414s :      1: task_emit
22.88% :     0.004094s :      1: type_inference
 0.29% :     0.000051s :      1: validate


TotalTime = 0.0200874, [33]
[bootstrap]: 0.00033973
[type_inference]: 0.00504744
[event_method]: 1.647e-05
[auto_monad]: 6.073e-05
[graph_reusing]: 5.49998e-06
[pre_auto_parallel]: 2.39001e-06
[py_interpret_to_execute]: 3.215e-05
[rewriter_before_opt_a]: 7e-05
[expand_dump_flag]: 2.95998e-06
[jit_opt_a]: 0.00541682, [2]
    [Cycle 1]: 0.00179091, [27]
        [switch_simplify]: 5.659e-05
        [loop_unroll]: 3.009e-05
        [a_1]: 0.00055279
        [with_stream_mark]: 1.835e-05
        [recompute_prepare]: 6.71e-06
        [updatestate_depend_eliminate]: 3.86001e-06
        [updatestate_assign_eliminate]: 3.04001e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [parameter_eliminate]: 1.74e-06
        [specialize_transform]: 6.01e-06
        [updatestate_useless_node_eliminater]: 5.35999e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 2.31998e-06
        [get_grad_eliminate_]: 5.04e-06
        [merge_forward]: 3.52002e-06
        [cell_reuse_recompute_pass]: 1.38002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.393e-05
        [j_node_and_user_rematch]: 9.10999e-06
        [meta_fg_expand]: 2.46e-06
        [replace_old_param]: 9.23002e-06
        [inline_without_move]: 5.03002e-06
        [renormalize]: 0.0007605
        [add_forward_monad_depend]: 7.17002e-06
        [auto_monad_grad]: 3.14999e-06
        [auto_monad_eliminator]: 1.6e-05
        [cse]: 3.333e-05
        [replace_applicator]: 1.505e-05
    [Cycle 2]: 0.00034232, [27]
        [switch_simplify]: 5.60001e-06
        [loop_unroll]: 4.73001e-06
        [a_1]: 9.144e-05
        [with_stream_mark]: 1.111e-05
        [recompute_prepare]: 5.61e-06
        [updatestate_depend_eliminate]: 2.66999e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.66e-06
        [specialize_transform]: 5.40999e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 4.92e-06
        [meta_shard_fg_expand]: 1.74998e-06
        [get_grad_eliminate_]: 5.25999e-06
        [merge_forward]: 3.85e-06
        [cell_reuse_recompute_pass]: 3.28e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.398e-05
        [j_node_and_user_rematch]: 8.38001e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 7.56001e-06
        [inline_without_move]: 4.97e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 9.5999e-07
        [auto_monad_eliminator]: 6.38e-06
        [cse]: 1.451e-05
        [replace_applicator]: 5.40999e-06
[py_interpret_to_execute_after_opt_a]: 1.355e-05
[rewriter_after_opt_a]: 3.864e-05
[convert_after_rewriter]: 7.15998e-06
[order_py_execute_after_rewriter]: 5.06002e-06
[mutable_eliminate]: 0.00070843
[jit_opt_b]: 5.194e-05, [1]
    [Cycle 1]: 4.354e-05, [2]
        [frontend_op_eliminate]: 1.627e-05
        [inline_after_opt_a]: 1.498e-05
[cconv]: 3.171e-05
[loop_unroll]: 0.00057108
[jit_opt_after_cconv]: 0.00015022, [1]
    [Cycle 1]: 0.00014266, [11]
        [c_1]: 2.071e-05
        [parameter_eliminate]: 4e-06
        [updatestate_depend_eliminate]: 7.15e-06
        [updatestate_assign_eliminate]: 3.16001e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [cse]: 2.466e-05
        [call_graph_tuple_transform]: 1.879e-05
        [tuple_list_get_item_eliminator]: 5.07999e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 7.10017e-07
        [switch_simplify]: 5.19e-06
[remove_dup_value]: 1.384e-05
[partial_unused_args_eliminate]: 2.09e-06
[environ_conv]: 1.277e-05
[add_recomputation]: 5.588e-05
[cse_after_recomputation]: 2.302e-05, [1]
    [Cycle 1]: 1.69e-05, [1]
        [cse]: 9.82999e-06
[auto_monad_reorder]: 2.71e-05
[get_jit_bprop_graph]: 1.97001e-06
[rewriter_after_jit_bprop_graph]: 7.49002e-06
[opt_after_jit_grad]: 0.00053304
[symbol_engine_optimizer]: 7.787e-05, [1]
    [Cycle 1]: 7.103e-05, [6]
        [build]: 5.64998e-06
        [elim_shapecalc]: 8.20999e-06
        [elim_not_effective]: 1.292e-05
        [opt_reshape]: 5.39e-06
        [fold_const_symbol]: 9.67999e-06
        [renormalize]: 8.50006e-07
[validate]: 4.119e-05
[backend_pass]: 1.12e-06
[task_emit]: 0.00647979
[execute]: 8.80001e-06
Sums
    bootstrap                            :     0.000340s :  2.11%
    type_inference                       :     0.005047s : 31.32%
    event_method                         :     0.000016s :  0.10%
    auto_monad                           :     0.000061s :  0.38%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000032s :  0.20%
    rewriter_before_opt_a                :     0.000070s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000644s :  4.00%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000761s :  4.72%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.14%
    jit_opt_a.cse                        :     0.000048s :  0.30%
    jit_opt_a.replace_applicator         :     0.000020s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000039s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000708s :  4.40%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000032s :  0.20%
    loop_unroll                          :     0.000571s :  3.54%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000013s :  0.08%
    add_recomputation                    :     0.000056s :  0.35%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000027s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.05%
    opt_after_jit_grad                   :     0.000533s :  3.31%
    symbol_engine_optimizer.build        :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006480s : 40.21%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000213    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.77% :     0.000002s :      2: substitution.fold_const_symbol
 2.42% :     0.000005s :      3: substitution.graph_param_transform
84.00% :     0.000179s :      4: substitution.inline
 1.37% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.13% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.29% :     0.000005s :      2: substitution.replace_old_param
 6.14% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004980     2
82.85% :     0.004126s :      1: type_inference.infer
17.15% :     0.000854s :      1: type_inference.specialize
------[replace.]   0.000067     6
68.21% :     0.000046s :      4: replace.inline
31.79% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000189     6
93.62% :     0.000177s :      4: match.inline
 6.38% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.29% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.36% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000002s :     10: predicate.addn_zero_filter
 2.40% :     0.000003s :     10: predicate.arithmetic_simplify
 1.22% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.51% :     0.000002s :     10: predicate.depend_value_elim
 1.22% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.11% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 2.02% :     0.000003s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.93% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.04% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.04% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.21% :     0.000002s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 5.33% :     0.000007s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000002s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.97% :     0.000003s :     15: predicate.load_eliminater
 2.31% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 2.98% :     0.000004s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.26% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.66% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     18: predicate.replace_applicator
 0.85% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.07% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.13% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.07% :     0.000010s :     53: predicate.switch_simplify
 1.05% :     0.000001s :     10: predicate.tile_eliminate
 1.30% :     0.000002s :     10: predicate.transpose_eliminate
 1.26% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.57% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.39% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.35% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.64% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.36% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000667    12
44.92% :     0.000300s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.97% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.11% :     0.000328s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021821    76
 0.27% :     0.000060s :      1: add_recomputation
 0.29% :     0.000064s :      1: auto_monad
 0.14% :     0.000030s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.67% :     0.000365s :      1: bootstrap
 0.16% :     0.000034s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000025s :      1: cse_after_recomputation
 0.07% :     0.000015s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
24.84% :     0.005420s :      1: jit_opt_a
 0.70% :     0.000153s :      1: jit_opt_after_cconv
 0.25% :     0.000055s :      1: jit_opt_b
 2.67% :     0.000583s :      1: loop_unroll
 3.30% :     0.000721s :      1: mutable_eliminate
 3.94% :     0.000860s :     26: opt.transform.jit_opt_a
 0.21% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.49% :     0.000543s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000035s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.88% :     0.000410s :      1: renormalize.infer
 1.56% :     0.000341s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000043s :      1: rewriter_after_opt_a
 0.33% :     0.000073s :      1: rewriter_before_opt_a
 0.37% :     0.000081s :      1: symbol_engine_optimizer
29.76% :     0.006495s :      1: task_emit
23.20% :     0.005062s :      1: type_inference
 0.31% :     0.000067s :      1: validate


TotalTime = 0.0241792, [33]
[bootstrap]: 0.00037363
[type_inference]: 0.00572309
[event_method]: 2.118e-05
[auto_monad]: 7.245e-05
[graph_reusing]: 6.06998e-06
[pre_auto_parallel]: 3.14001e-06
[py_interpret_to_execute]: 5.141e-05
[rewriter_before_opt_a]: 8.051e-05
[expand_dump_flag]: 4.26001e-06
[jit_opt_a]: 0.00705363, [2]
    [Cycle 1]: 0.00206917, [27]
        [switch_simplify]: 7.202e-05
        [loop_unroll]: 3.003e-05
        [a_1]: 0.00061751
        [with_stream_mark]: 2.19e-05
        [recompute_prepare]: 8.91002e-06
        [updatestate_depend_eliminate]: 4.12998e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 2.65002e-06
        [parameter_eliminate]: 2.26e-06
        [specialize_transform]: 6.30002e-06
        [updatestate_useless_node_eliminater]: 5.52999e-06
        [accelerated_algorithm]: 7.08998e-06
        [meta_shard_fg_expand]: 2.74999e-06
        [get_grad_eliminate_]: 5.74e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 1.79e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.435e-05
        [j_node_and_user_rematch]: 9.67001e-06
        [meta_fg_expand]: 2.17999e-06
        [replace_old_param]: 1.001e-05
        [inline_without_move]: 6.29999e-06
        [renormalize]: 0.00095232
        [add_forward_monad_depend]: 6.58e-06
        [auto_monad_grad]: 2.49001e-06
        [auto_monad_eliminator]: 1.812e-05
        [cse]: 3.25e-05
        [replace_applicator]: 1.714e-05
    [Cycle 2]: 0.000362, [27]
        [switch_simplify]: 5.71e-06
        [loop_unroll]: 5.29e-06
        [a_1]: 9.709e-05
        [with_stream_mark]: 1.28e-05
        [recompute_prepare]: 4.74998e-06
        [updatestate_depend_eliminate]: 3.22997e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 5.02999e-06
        [updatestate_useless_node_eliminater]: 4.74998e-06
        [accelerated_algorithm]: 6.16e-06
        [meta_shard_fg_expand]: 1.57001e-06
        [get_grad_eliminate_]: 4.82e-06
        [merge_forward]: 2.44001e-06
        [cell_reuse_recompute_pass]: 2.36998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.534e-05
        [j_node_and_user_rematch]: 7.97e-06
        [meta_fg_expand]: 1.77999e-06
        [replace_old_param]: 2.184e-05
        [inline_without_move]: 5.45001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.16002e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 6.02999e-06
        [cse]: 1.238e-05
        [replace_applicator]: 4.97999e-06
[py_interpret_to_execute_after_opt_a]: 1.437e-05
[rewriter_after_opt_a]: 3.991e-05
[convert_after_rewriter]: 7.56001e-06
[order_py_execute_after_rewriter]: 4.73001e-06
[mutable_eliminate]: 0.00088879
[jit_opt_b]: 6.419e-05, [1]
    [Cycle 1]: 5.461e-05, [2]
        [frontend_op_eliminate]: 2.087e-05
        [inline_after_opt_a]: 1.835e-05
[cconv]: 3.467e-05
[loop_unroll]: 0.00064182
[jit_opt_after_cconv]: 0.00018471, [1]
    [Cycle 1]: 0.00017654, [11]
        [c_1]: 2.54e-05
        [parameter_eliminate]: 5.09e-06
        [updatestate_depend_eliminate]: 7.95e-06
        [updatestate_assign_eliminate]: 3.57002e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [cse]: 3.581e-05
        [call_graph_tuple_transform]: 2.621e-05
        [tuple_list_get_item_eliminator]: 6.11e-06
        [none_parameter_eliminate]: 1.72001e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 5.68002e-06
[remove_dup_value]: 1.608e-05
[partial_unused_args_eliminate]: 2.53e-06
[environ_conv]: 6.11e-06
[add_recomputation]: 5.679e-05
[cse_after_recomputation]: 2.52e-05, [1]
    [Cycle 1]: 1.762e-05, [1]
        [cse]: 9.92001e-06
[auto_monad_reorder]: 1.695e-05
[get_jit_bprop_graph]: 2.54001e-06
[rewriter_after_jit_bprop_graph]: 1.151e-05
[opt_after_jit_grad]: 0.00068315
[symbol_engine_optimizer]: 8.653e-05, [1]
    [Cycle 1]: 7.706e-05, [6]
        [build]: 4.74998e-06
        [elim_shapecalc]: 8.72e-06
        [elim_not_effective]: 1.66e-05
        [opt_reshape]: 6.95998e-06
        [fold_const_symbol]: 1.002e-05
        [renormalize]: 1.54e-06
[validate]: 4.511e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00767964
[execute]: 9.20999e-06
Sums
    bootstrap                            :     0.000374s :  1.98%
    type_inference                       :     0.005723s : 30.40%
    event_method                         :     0.000021s :  0.11%
    auto_monad                           :     0.000072s :  0.38%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000051s :  0.27%
    rewriter_before_opt_a                :     0.000081s :  0.43%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000078s :  0.41%
    jit_opt_a.loop_unroll                :     0.000035s :  0.19%
    jit_opt_a.a_1                        :     0.000715s :  3.80%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000032s :  0.17%
    jit_opt_a.inline_without_move        :     0.000012s :  0.06%
    jit_opt_a.renormalize                :     0.000952s :  5.06%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.13%
    jit_opt_a.cse                        :     0.000045s :  0.24%
    jit_opt_a.replace_applicator         :     0.000022s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000040s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000889s :  4.72%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.10%
    cconv                                :     0.000035s :  0.18%
    loop_unroll                          :     0.000642s :  3.41%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000036s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000057s :  0.30%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000012s :  0.06%
    opt_after_jit_grad                   :     0.000683s :  3.63%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000002s :  0.01%
    validate                             :     0.000045s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007680s : 40.80%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000250    23
 1.05% :     0.000003s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 2.49% :     0.000006s :      3: substitution.graph_param_transform
84.64% :     0.000212s :      4: substitution.inline
 1.46% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.07% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.19% :     0.000005s :      2: substitution.replace_old_param
 5.58% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005631     2
78.68% :     0.004430s :      1: type_inference.infer
21.32% :     0.001201s :      1: type_inference.specialize
------[replace.]   0.000087     6
67.80% :     0.000059s :      4: replace.inline
32.20% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000222     6
94.18% :     0.000209s :      4: match.inline
 5.82% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.69% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.13% :     0.000002s :     10: predicate.addn_zero_filter
 2.37% :     0.000003s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.91% :     0.000001s :     10: predicate.compare_switch_simplify
 1.54% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.71% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.24% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.32% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 6.18% :     0.000008s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 2.44% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.47% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.31% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.93% :     0.000001s :     10: predicate.merge_addn
 1.23% :     0.000002s :     10: predicate.minmaximum_grad
 1.96% :     0.000003s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.00% :     0.000003s :     15: predicate.partial_eliminate
 1.28% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.23% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.53% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.71% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.06% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000002s :     10: predicate.same_eliminate
 0.63% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.33% :     0.000002s :      6: predicate.special_op_eliminate
 0.85% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 1.75% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.97% :     0.000009s :     53: predicate.switch_simplify
 1.44% :     0.000002s :     10: predicate.tile_eliminate
 1.00% :     0.000001s :     10: predicate.transpose_eliminate
 1.24% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.44% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.00% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.39% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.37% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000715    12
42.55% :     0.000304s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.97% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
52.48% :     0.000375s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026231    76
 0.23% :     0.000061s :      1: add_recomputation
 0.29% :     0.000077s :      1: auto_monad
 0.08% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.50% :     0.000394s :      1: bootstrap
 0.14% :     0.000038s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000028s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000026s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.03% :     0.000007s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.91% :     0.007058s :      1: jit_opt_a
 0.72% :     0.000189s :      1: jit_opt_after_cconv
 0.26% :     0.000067s :      1: jit_opt_b
 2.49% :     0.000654s :      1: loop_unroll
 3.44% :     0.000903s :      1: mutable_eliminate
 3.69% :     0.000968s :     26: opt.transform.jit_opt_a
 0.23% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000032s :      4: opt.transform.jit_opt_b
 0.07% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000023s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000039s :      4: opt.transform.symbol_engine_opt
 2.66% :     0.000697s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000054s :      1: py_interpret_to_execute
 0.06% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 1.81% :     0.000476s :      1: renormalize.infer
 1.77% :     0.000465s :      1: renormalize.specialize
 0.05% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000043s :      1: rewriter_after_opt_a
 0.32% :     0.000084s :      1: rewriter_before_opt_a
 0.34% :     0.000089s :      1: symbol_engine_optimizer
29.36% :     0.007700s :      1: task_emit
21.90% :     0.005744s :      1: type_inference
 0.26% :     0.000068s :      1: validate


TotalTime = 0.0184656, [33]
[bootstrap]: 0.00030114
[type_inference]: 0.00495375
[event_method]: 1.788e-05
[auto_monad]: 6.691e-05
[graph_reusing]: 5.25999e-06
[pre_auto_parallel]: 2.29999e-06
[py_interpret_to_execute]: 3.808e-05
[rewriter_before_opt_a]: 7e-05
[expand_dump_flag]: 3.46999e-06
[jit_opt_a]: 0.00536928, [2]
    [Cycle 1]: 0.00163341, [27]
        [switch_simplify]: 4.95e-05
        [loop_unroll]: 2.956e-05
        [a_1]: 0.00057324
        [with_stream_mark]: 1.685e-05
        [recompute_prepare]: 7.03e-06
        [updatestate_depend_eliminate]: 3.22002e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 5.79999e-06
        [updatestate_useless_node_eliminater]: 5.24998e-06
        [accelerated_algorithm]: 5.78002e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 5.32001e-06
        [merge_forward]: 3.51001e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.339e-05
        [j_node_and_user_rematch]: 8.55001e-06
        [meta_fg_expand]: 2.17999e-06
        [replace_old_param]: 7.87998e-06
        [inline_without_move]: 5.55001e-06
        [renormalize]: 0.00066904
        [add_forward_monad_depend]: 5.15001e-06
        [auto_monad_grad]: 2.11e-06
        [auto_monad_eliminator]: 1.133e-05
        [cse]: 2.127e-05
        [replace_applicator]: 1.104e-05
    [Cycle 2]: 0.00030807, [27]
        [switch_simplify]: 5.42001e-06
        [loop_unroll]: 4.68999e-06
        [a_1]: 8.989e-05
        [with_stream_mark]: 8.45001e-06
        [recompute_prepare]: 4.93001e-06
        [updatestate_depend_eliminate]: 2.41e-06
        [updatestate_assign_eliminate]: 2.06998e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 4.80001e-06
        [meta_shard_fg_expand]: 1.05001e-06
        [get_grad_eliminate_]: 4.38001e-06
        [merge_forward]: 2.22001e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.2e-05
        [j_node_and_user_rematch]: 7.38999e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 6.93e-06
        [inline_without_move]: 4.52998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 4.96997e-06
        [cse]: 9.34e-06
        [replace_applicator]: 4.79e-06
[py_interpret_to_execute_after_opt_a]: 9.99001e-06
[rewriter_after_opt_a]: 3.21e-05
[convert_after_rewriter]: 6.21e-06
[order_py_execute_after_rewriter]: 4.92e-06
[mutable_eliminate]: 0.00051685
[jit_opt_b]: 4.714e-05, [1]
    [Cycle 1]: 4.06e-05, [2]
        [frontend_op_eliminate]: 1.547e-05
        [inline_after_opt_a]: 1.451e-05
[cconv]: 3.331e-05
[loop_unroll]: 0.00043731
[jit_opt_after_cconv]: 0.00012945, [1]
    [Cycle 1]: 0.00012331, [11]
        [c_1]: 2.011e-05
        [parameter_eliminate]: 2.54999e-06
        [updatestate_depend_eliminate]: 4.49998e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [cse]: 1.5e-05
        [call_graph_tuple_transform]: 1.706e-05
        [tuple_list_get_item_eliminator]: 5.09e-06
        [none_parameter_eliminate]: 1.19e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 5.42999e-06
[remove_dup_value]: 1.229e-05
[partial_unused_args_eliminate]: 2.39001e-06
[environ_conv]: 4.94e-06
[add_recomputation]: 4.075e-05
[cse_after_recomputation]: 1.962e-05, [1]
    [Cycle 1]: 1.44e-05, [1]
        [cse]: 8.79e-06
[auto_monad_reorder]: 1.349e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 5.74e-06
[opt_after_jit_grad]: 0.00045466
[symbol_engine_optimizer]: 7.098e-05, [1]
    [Cycle 1]: 6.47e-05, [6]
        [build]: 2.68e-06
        [elim_shapecalc]: 7.58001e-06
        [elim_not_effective]: 1.28e-05
        [opt_reshape]: 5.66e-06
        [fold_const_symbol]: 8.74998e-06
        [renormalize]: 3.69997e-07
[validate]: 3.065e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00556773
[execute]: 6.48e-06
Sums
    bootstrap                            :     0.000301s :  2.08%
    type_inference                       :     0.004954s : 34.25%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000067s :  0.46%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.26%
    rewriter_before_opt_a                :     0.000070s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000663s :  4.59%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000669s :  4.63%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.11%
    jit_opt_a.cse                        :     0.000031s :  0.21%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000517s :  3.57%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000033s :  0.23%
    loop_unroll                          :     0.000437s :  3.02%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000041s :  0.28%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000013s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000455s :  3.14%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005568s : 38.50%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000240    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000002s :      2: substitution.fold_const_symbol
 2.17% :     0.000005s :      3: substitution.graph_param_transform
87.28% :     0.000210s :      4: substitution.inline
 1.20% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.74% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.30% :     0.000003s :      2: substitution.replace_old_param
 4.87% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004888     2
81.92% :     0.004004s :      1: type_inference.infer
18.08% :     0.000884s :      1: type_inference.specialize
------[replace.]   0.000066     6
67.16% :     0.000044s :      4: replace.inline
32.84% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000218     6
95.14% :     0.000207s :      4: match.inline
 4.86% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.28% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 2.11% :     0.000002s :     10: predicate.arithmetic_simplify
 1.19% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.50% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.65% :     0.000002s :      6: predicate.less_batch_normalization
 1.88% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.51% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.15% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.44% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.66% :     0.000002s :     10: predicate.reshape_eliminate
 1.32% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.12% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.95% :     0.000009s :     53: predicate.switch_simplify
 1.32% :     0.000002s :     10: predicate.tile_eliminate
 1.26% :     0.000001s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.35% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.32% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000595    12
47.63% :     0.000284s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.99% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.37% :     0.000276s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020097    76
 0.22% :     0.000044s :      1: add_recomputation
 0.35% :     0.000071s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.58% :     0.000318s :      1: bootstrap
 0.18% :     0.000036s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.73% :     0.005372s :      1: jit_opt_a
 0.66% :     0.000132s :      1: jit_opt_after_cconv
 0.25% :     0.000050s :      1: jit_opt_b
 2.21% :     0.000445s :      1: loop_unroll
 2.61% :     0.000525s :      1: mutable_eliminate
 4.28% :     0.000861s :     26: opt.transform.jit_opt_a
 0.22% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.30% :     0.000462s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000041s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 1.86% :     0.000373s :      1: renormalize.infer
 1.43% :     0.000288s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000035s :      1: rewriter_after_opt_a
 0.36% :     0.000073s :      1: rewriter_before_opt_a
 0.37% :     0.000073s :      1: symbol_engine_optimizer
27.77% :     0.005581s :      1: task_emit
24.71% :     0.004966s :      1: type_inference
 0.25% :     0.000049s :      1: validate


TotalTime = 0.0176095, [33]
[bootstrap]: 0.00030962
[type_inference]: 0.00428697
[event_method]: 1.877e-05
[auto_monad]: 6.522e-05
[graph_reusing]: 6.42001e-06
[pre_auto_parallel]: 2.34999e-06
[py_interpret_to_execute]: 3.23e-05
[rewriter_before_opt_a]: 7.318e-05
[expand_dump_flag]: 3.19001e-06
[jit_opt_a]: 0.00508093, [2]
    [Cycle 1]: 0.00148942, [27]
        [switch_simplify]: 7.529e-05
        [loop_unroll]: 2.936e-05
        [a_1]: 0.00054275
        [with_stream_mark]: 1.608e-05
        [recompute_prepare]: 6.75998e-06
        [updatestate_depend_eliminate]: 3.33e-06
        [updatestate_assign_eliminate]: 2.76999e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [parameter_eliminate]: 2.16998e-06
        [specialize_transform]: 5.49e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 5.24998e-06
        [meta_shard_fg_expand]: 2.32001e-06
        [get_grad_eliminate_]: 5.20001e-06
        [merge_forward]: 3.23998e-06
        [cell_reuse_recompute_pass]: 1.02998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.404e-05
        [j_node_and_user_rematch]: 8.79e-06
        [meta_fg_expand]: 2.17999e-06
        [replace_old_param]: 8.28001e-06
        [inline_without_move]: 5.04e-06
        [renormalize]: 0.00052264
        [add_forward_monad_depend]: 5.04e-06
        [auto_monad_grad]: 2.34001e-06
        [auto_monad_eliminator]: 1.31e-05
        [cse]: 2.758e-05
        [replace_applicator]: 1.014e-05
    [Cycle 2]: 0.00031583, [27]
        [switch_simplify]: 5.24e-06
        [loop_unroll]: 4.69002e-06
        [a_1]: 8.61e-05
        [with_stream_mark]: 8.57e-06
        [recompute_prepare]: 4.89e-06
        [updatestate_depend_eliminate]: 2.49001e-06
        [updatestate_assign_eliminate]: 1.91e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 8.39995e-07
        [specialize_transform]: 4.88001e-06
        [updatestate_useless_node_eliminater]: 4.52998e-06
        [accelerated_algorithm]: 4.61002e-06
        [meta_shard_fg_expand]: 9.70002e-07
        [get_grad_eliminate_]: 4.37e-06
        [merge_forward]: 2.16003e-06
        [cell_reuse_recompute_pass]: 1.45001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.147e-05
        [j_node_and_user_rematch]: 7.33e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 6.62002e-06
        [inline_without_move]: 4.52e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.03001e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 5.54e-06
        [cse]: 1.097e-05
        [replace_applicator]: 5.13002e-06
[py_interpret_to_execute_after_opt_a]: 7.64002e-06
[rewriter_after_opt_a]: 3.278e-05
[convert_after_rewriter]: 6.44999e-06
[order_py_execute_after_rewriter]: 4.72e-06
[mutable_eliminate]: 0.00047819
[jit_opt_b]: 4.728e-05, [1]
    [Cycle 1]: 4.096e-05, [2]
        [frontend_op_eliminate]: 1.516e-05
        [inline_after_opt_a]: 1.462e-05
[cconv]: 2.313e-05
[loop_unroll]: 0.00039635
[jit_opt_after_cconv]: 0.00012959, [1]
    [Cycle 1]: 0.0001235, [11]
        [c_1]: 1.988e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 5.12e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.573e-05
        [call_graph_tuple_transform]: 1.689e-05
        [tuple_list_get_item_eliminator]: 4.79998e-06
        [none_parameter_eliminate]: 1.48002e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 5.00001e-06
[remove_dup_value]: 1.3e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 5.61e-06
[add_recomputation]: 4.655e-05
[cse_after_recomputation]: 1.965e-05, [1]
    [Cycle 1]: 1.411e-05, [1]
        [cse]: 8.85999e-06
[auto_monad_reorder]: 1.478e-05
[get_jit_bprop_graph]: 1.60999e-06
[rewriter_after_jit_bprop_graph]: 4.68001e-06
[opt_after_jit_grad]: 0.00043796
[symbol_engine_optimizer]: 6.825e-05, [1]
    [Cycle 1]: 6.251e-05, [6]
        [build]: 3.00998e-06
        [elim_shapecalc]: 7.06999e-06
        [elim_not_effective]: 1.172e-05
        [opt_reshape]: 5.80002e-06
        [fold_const_symbol]: 7.92e-06
        [renormalize]: 4.20026e-07
[validate]: 3.155e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.00576407
[execute]: 7.18e-06
Sums
    bootstrap                            :     0.000310s :  2.25%
    type_inference                       :     0.004287s : 31.18%
    event_method                         :     0.000019s :  0.14%
    auto_monad                           :     0.000065s :  0.47%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.23%
    rewriter_before_opt_a                :     0.000073s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000081s :  0.59%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000629s :  4.57%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000523s :  3.80%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.14%
    jit_opt_a.cse                        :     0.000039s :  0.28%
    jit_opt_a.replace_applicator         :     0.000015s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.24%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000478s :  3.48%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000396s :  2.88%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000047s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000438s :  3.19%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005764s : 41.92%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000192    23
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.58% :     0.000005s :      3: substitution.graph_param_transform
84.53% :     0.000163s :      4: substitution.inline
 1.46% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.10% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.77% :     0.000003s :      2: substitution.replace_old_param
 6.02% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004219     2
78.49% :     0.003311s :      1: type_inference.infer
21.51% :     0.000908s :      1: type_inference.specialize
------[replace.]   0.000062     6
68.84% :     0.000043s :      4: replace.inline
31.16% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000171     6
93.89% :     0.000160s :      4: match.inline
 6.11% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.35% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 2.29% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000001s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.46% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.04% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.52% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.15% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.69% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.37% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000001s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.51% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.97% :     0.000009s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.43% :     0.000002s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.98% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.53% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000517    12
48.52% :     0.000251s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.53% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.95% :     0.000232s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019078    76
 0.26% :     0.000049s :      1: add_recomputation
 0.36% :     0.000069s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.71% :     0.000326s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.12% :     0.000023s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
26.65% :     0.005083s :      1: jit_opt_a
 0.69% :     0.000132s :      1: jit_opt_after_cconv
 0.26% :     0.000050s :      1: jit_opt_b
 2.12% :     0.000404s :      1: loop_unroll
 2.55% :     0.000486s :      1: mutable_eliminate
 4.45% :     0.000849s :     26: opt.transform.jit_opt_a
 0.23% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.34% :     0.000446s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000035s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.43% :     0.000273s :      1: renormalize.infer
 1.27% :     0.000242s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000036s :      1: rewriter_after_opt_a
 0.40% :     0.000076s :      1: rewriter_before_opt_a
 0.37% :     0.000071s :      1: symbol_engine_optimizer
30.27% :     0.005775s :      1: task_emit
22.56% :     0.004304s :      1: type_inference
 0.26% :     0.000049s :      1: validate


TotalTime = 0.0171142, [33]
[bootstrap]: 0.00025946
[type_inference]: 0.00395228
[event_method]: 1.628e-05
[auto_monad]: 4.261e-05
[graph_reusing]: 3.58999e-06
[pre_auto_parallel]: 1.72999e-06
[py_interpret_to_execute]: 2.945e-05
[rewriter_before_opt_a]: 6.36e-05
[expand_dump_flag]: 1.97001e-06
[jit_opt_a]: 0.00475161, [2]
    [Cycle 1]: 0.00138875, [27]
        [switch_simplify]: 4.695e-05
        [loop_unroll]: 3.121e-05
        [a_1]: 0.0004968
        [with_stream_mark]: 1.255e-05
        [recompute_prepare]: 7.31001e-06
        [updatestate_depend_eliminate]: 3.16001e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.95001e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 6.24999e-06
        [updatestate_useless_node_eliminater]: 4.96002e-06
        [accelerated_algorithm]: 6.38e-06
        [meta_shard_fg_expand]: 1.85001e-06
        [get_grad_eliminate_]: 6.04999e-06
        [merge_forward]: 2.75002e-06
        [cell_reuse_recompute_pass]: 1.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.27e-05
        [j_node_and_user_rematch]: 9.79e-06
        [meta_fg_expand]: 1.97001e-06
        [replace_old_param]: 8.48001e-06
        [inline_without_move]: 5.62001e-06
        [renormalize]: 0.00049159
        [add_forward_monad_depend]: 4.54002e-06
        [auto_monad_grad]: 1.27e-06
        [auto_monad_eliminator]: 9.87001e-06
        [cse]: 1.767e-05
        [replace_applicator]: 1.222e-05
    [Cycle 2]: 0.00033359, [27]
        [switch_simplify]: 6.41e-06
        [loop_unroll]: 5.24e-06
        [a_1]: 9.784e-05
        [with_stream_mark]: 9.86998e-06
        [recompute_prepare]: 5.39e-06
        [updatestate_depend_eliminate]: 2.72001e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 2.03002e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 5.51002e-06
        [updatestate_useless_node_eliminater]: 5.44998e-06
        [accelerated_algorithm]: 5.24e-06
        [meta_shard_fg_expand]: 1.32999e-06
        [get_grad_eliminate_]: 5.15001e-06
        [merge_forward]: 2.56e-06
        [cell_reuse_recompute_pass]: 1.97001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.26e-05
        [j_node_and_user_rematch]: 8.57998e-06
        [meta_fg_expand]: 1.52999e-06
        [replace_old_param]: 6.94001e-06
        [inline_without_move]: 4.61002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.64e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 4.74e-06
        [cse]: 9.97999e-06
        [replace_applicator]: 5.39e-06
[py_interpret_to_execute_after_opt_a]: 9.02e-06
[rewriter_after_opt_a]: 2.743e-05
[convert_after_rewriter]: 5.97001e-06
[order_py_execute_after_rewriter]: 4.36002e-06
[mutable_eliminate]: 0.000523
[jit_opt_b]: 4.721e-05, [1]
    [Cycle 1]: 4.146e-05, [2]
        [frontend_op_eliminate]: 1.596e-05
        [inline_after_opt_a]: 1.428e-05
[cconv]: 1.706e-05
[loop_unroll]: 0.00042519
[jit_opt_after_cconv]: 0.00013718, [1]
    [Cycle 1]: 0.00013136, [11]
        [c_1]: 2.088e-05
        [parameter_eliminate]: 2.56998e-06
        [updatestate_depend_eliminate]: 5.44e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [cse]: 1.83e-05
        [call_graph_tuple_transform]: 1.68e-05
        [tuple_list_get_item_eliminator]: 5.59e-06
        [none_parameter_eliminate]: 8.2e-07
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.37999e-06
[remove_dup_value]: 1.448e-05
[partial_unused_args_eliminate]: 1.42e-06
[environ_conv]: 3.3e-06
[add_recomputation]: 3.36e-05
[cse_after_recomputation]: 2.094e-05, [1]
    [Cycle 1]: 1.532e-05, [1]
        [cse]: 9.41e-06
[auto_monad_reorder]: 8.57998e-06
[get_jit_bprop_graph]: 1.40999e-06
[rewriter_after_jit_bprop_graph]: 4.89e-06
[opt_after_jit_grad]: 0.00044922
[symbol_engine_optimizer]: 6.792e-05, [1]
    [Cycle 1]: 6.241e-05, [6]
        [build]: 2.36e-06
        [elim_shapecalc]: 7.16999e-06
        [elim_not_effective]: 1.157e-05
        [opt_reshape]: 5.30001e-06
        [fold_const_symbol]: 8.10999e-06
        [renormalize]: 4.80009e-07
[validate]: 2.459e-05
[backend_pass]: 1.28002e-06
[task_emit]: 0.00598683
[execute]: 7.65e-06
Sums
    bootstrap                            :     0.000259s :  1.92%
    type_inference                       :     0.003952s : 29.27%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000043s :  0.32%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000029s :  0.22%
    rewriter_before_opt_a                :     0.000064s :  0.47%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000053s :  0.40%
    jit_opt_a.loop_unroll                :     0.000036s :  0.27%
    jit_opt_a.a_1                        :     0.000595s :  4.40%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000492s :  3.64%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000028s :  0.20%
    jit_opt_a.replace_applicator         :     0.000018s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000027s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000523s :  3.87%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000017s :  0.13%
    loop_unroll                          :     0.000425s :  3.15%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.11%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000003s :  0.02%
    add_recomputation                    :     0.000034s :  0.25%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000009s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000449s :  3.33%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000025s :  0.18%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005987s : 44.34%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000152    23
 0.89% :     0.000001s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.29% :     0.000003s :      3: substitution.graph_param_transform
83.76% :     0.000128s :      4: substitution.inline
 2.28% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.75% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.00% :     0.000003s :      2: substitution.replace_old_param
 5.37% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003904     2
80.59% :     0.003146s :      1: type_inference.infer
19.41% :     0.000758s :      1: type_inference.specialize
------[replace.]   0.000061     6
66.12% :     0.000040s :      4: replace.inline
33.88% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000132     6
94.75% :     0.000125s :      4: match.inline
 5.25% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.66% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.64% :     0.000002s :     10: predicate.addn_zero_filter
 2.41% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.23% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.10% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.76% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.10% :     0.000003s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.22% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.62% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.07% :     0.000002s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.83% :     0.000002s :     10: predicate.reshape_eliminate
 1.38% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.76% :     0.000009s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.35% :     0.000002s :     10: predicate.transpose_eliminate
 1.60% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.19% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.80% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.39% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000506    12
45.66% :     0.000231s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.32% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.02% :     0.000248s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018515    76
 0.20% :     0.000037s :      1: add_recomputation
 0.25% :     0.000046s :      1: auto_monad
 0.06% :     0.000011s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.46% :     0.000270s :      1: bootstrap
 0.11% :     0.000020s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000005s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
25.68% :     0.004754s :      1: jit_opt_a
 0.76% :     0.000140s :      1: jit_opt_after_cconv
 0.27% :     0.000050s :      1: jit_opt_b
 2.34% :     0.000433s :      1: loop_unroll
 2.87% :     0.000531s :      1: mutable_eliminate
 4.34% :     0.000803s :     26: opt.transform.jit_opt_a
 0.25% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.47% :     0.000457s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000017s :      1: remove_dup_value
 1.31% :     0.000243s :      1: renormalize.infer
 1.31% :     0.000243s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000031s :      1: rewriter_after_opt_a
 0.36% :     0.000066s :      1: rewriter_before_opt_a
 0.38% :     0.000070s :      1: symbol_engine_optimizer
32.39% :     0.005996s :      1: task_emit
21.40% :     0.003962s :      1: type_inference
 0.22% :     0.000041s :      1: validate


TotalTime = 0.0290645, [33]
[bootstrap]: 0.00032377
[type_inference]: 0.00500523
[event_method]: 1.731e-05
[auto_monad]: 6.185e-05
[graph_reusing]: 5.63002e-06
[pre_auto_parallel]: 2.66e-06
[py_interpret_to_execute]: 3.993e-05
[rewriter_before_opt_a]: 6.981e-05
[expand_dump_flag]: 2.66e-06
[jit_opt_a]: 0.00847432, [2]
    [Cycle 1]: 0.00212426, [27]
        [switch_simplify]: 5.942e-05
        [loop_unroll]: 2.904e-05
        [a_1]: 0.00060444
        [with_stream_mark]: 1.973e-05
        [recompute_prepare]: 8.74e-06
        [updatestate_depend_eliminate]: 3.38e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.50999e-06
        [specialize_transform]: 6.83e-06
        [updatestate_useless_node_eliminater]: 5.90002e-06
        [accelerated_algorithm]: 5.81998e-06
        [meta_shard_fg_expand]: 1.92999e-06
        [get_grad_eliminate_]: 5.24e-06
        [merge_forward]: 3.36999e-06
        [cell_reuse_recompute_pass]: 2.16e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.292e-05
        [j_node_and_user_rematch]: 1.055e-05
        [meta_fg_expand]: 1.94e-06
        [replace_old_param]: 8.84e-06
        [inline_without_move]: 5.40001e-06
        [renormalize]: 0.0010638
        [add_forward_monad_depend]: 7.38e-06
        [auto_monad_grad]: 3.04001e-06
        [auto_monad_eliminator]: 1.686e-05
        [cse]: 3.055e-05
        [replace_applicator]: 1.927e-05
    [Cycle 2]: 0.00038896, [27]
        [switch_simplify]: 6.22001e-06
        [loop_unroll]: 5.42001e-06
        [a_1]: 0.00010764
        [with_stream_mark]: 1.342e-05
        [recompute_prepare]: 6.61e-06
        [updatestate_depend_eliminate]: 3.13998e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 6.07999e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 1.35999e-06
        [get_grad_eliminate_]: 4.98001e-06
        [merge_forward]: 3.75e-06
        [cell_reuse_recompute_pass]: 3.65998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.675e-05
        [j_node_and_user_rematch]: 9.12001e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 1.007e-05
        [inline_without_move]: 5.57001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 2.58003e-06
        [auto_monad_grad]: 1.09e-06
        [auto_monad_eliminator]: 7.36001e-06
        [cse]: 1.253e-05
        [replace_applicator]: 6.20002e-06
[py_interpret_to_execute_after_opt_a]: 1.72e-05
[rewriter_after_opt_a]: 4.223e-05
[convert_after_rewriter]: 7.65e-06
[order_py_execute_after_rewriter]: 5.56e-06
[mutable_eliminate]: 0.00456862
[jit_opt_b]: 9.106e-05, [1]
    [Cycle 1]: 7.816e-05, [2]
        [frontend_op_eliminate]: 4.101e-05
        [inline_after_opt_a]: 2.23e-05
[cconv]: 4.393e-05
[loop_unroll]: 0.00062711
[jit_opt_after_cconv]: 0.00020975, [1]
    [Cycle 1]: 0.00020214, [11]
        [c_1]: 2.588e-05
        [parameter_eliminate]: 7.4e-06
        [updatestate_depend_eliminate]: 1.155e-05
        [updatestate_assign_eliminate]: 3.86999e-06
        [updatestate_loads_eliminate]: 2.89001e-06
        [cse]: 4.718e-05
        [call_graph_tuple_transform]: 2.891e-05
        [tuple_list_get_item_eliminator]: 5.92001e-06
        [none_parameter_eliminate]: 1.84998e-06
        [renormalize]: 8.09989e-07
        [switch_simplify]: 6.77002e-06
[remove_dup_value]: 1.791e-05
[partial_unused_args_eliminate]: 3.11999e-06
[environ_conv]: 7.68001e-06
[add_recomputation]: 6.033e-05
[cse_after_recomputation]: 2.485e-05, [1]
    [Cycle 1]: 1.696e-05, [1]
        [cse]: 9.60001e-06
[auto_monad_reorder]: 1.78e-05
[get_jit_bprop_graph]: 2.56998e-06
[rewriter_after_jit_bprop_graph]: 1.286e-05
[opt_after_jit_grad]: 0.00062749
[symbol_engine_optimizer]: 9.545e-05, [1]
    [Cycle 1]: 8.509e-05, [6]
        [build]: 5.92001e-06
        [elim_shapecalc]: 9.17001e-06
        [elim_not_effective]: 1.81e-05
        [opt_reshape]: 7.93999e-06
        [fold_const_symbol]: 1.134e-05
        [renormalize]: 1.17e-06
[validate]: 5.142e-05
[backend_pass]: 9.49978e-07
[task_emit]: 0.00825781
[execute]: 1.122e-05
Sums
    bootstrap                            :     0.000324s :  1.45%
    type_inference                       :     0.005005s : 22.37%
    event_method                         :     0.000017s :  0.08%
    auto_monad                           :     0.000062s :  0.28%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000040s :  0.18%
    rewriter_before_opt_a                :     0.000070s :  0.31%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000066s :  0.29%
    jit_opt_a.loop_unroll                :     0.000034s :  0.15%
    jit_opt_a.a_1                        :     0.000712s :  3.18%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.08%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.001064s :  4.75%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.11%
    jit_opt_a.cse                        :     0.000043s :  0.19%
    jit_opt_a.replace_applicator         :     0.000025s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.08%
    rewriter_after_opt_a                 :     0.000042s :  0.19%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.004569s : 20.42%
    jit_opt_b.frontend_op_eliminate      :     0.000041s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000022s :  0.10%
    cconv                                :     0.000044s :  0.20%
    loop_unroll                          :     0.000627s :  2.80%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000047s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000029s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000018s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000008s :  0.03%
    add_recomputation                    :     0.000060s :  0.27%
    cse_after_recomputation.cse          :     0.000010s :  0.04%
    auto_monad_reorder                   :     0.000018s :  0.08%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000013s :  0.06%
    opt_after_jit_grad                   :     0.000627s :  2.80%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000051s :  0.23%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008258s : 36.90%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000243    23
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.94% :     0.000002s :      2: substitution.fold_const_symbol
 2.66% :     0.000006s :      3: substitution.graph_param_transform
83.71% :     0.000204s :      4: substitution.inline
 1.57% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.20% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.57% :     0.000006s :      2: substitution.replace_old_param
 5.39% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004934     2
80.60% :     0.003977s :      1: type_inference.infer
19.40% :     0.000957s :      1: type_inference.specialize
------[replace.]   0.000083     6
66.45% :     0.000055s :      4: replace.inline
33.55% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000213     6
94.38% :     0.000201s :      4: match.inline
 5.62% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000150   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.71% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 2.06% :     0.000003s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.91% :     0.000001s :     10: predicate.compare_switch_simplify
 1.53% :     0.000002s :     10: predicate.depend_value_elim
 0.94% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.52% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.08% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000001s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.60% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000002s :     10: predicate.environ_get_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.75% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 4.20% :     0.000006s :     22: predicate.inline
 0.74% :     0.000001s :      6: predicate.inline_without_move
 0.29% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000002s :      6: predicate.less_batch_normalization
 1.14% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.29% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.05% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.16% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.91% :     0.000001s :     10: predicate.merge_addn
 0.95% :     0.000001s :     10: predicate.minmaximum_grad
11.98% :     0.000018s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 1.77% :     0.000003s :     15: predicate.partial_eliminate
 1.08% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.36% :     0.000002s :     10: predicate.reduce_eliminate
 1.23% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.48% :     0.000002s :     18: predicate.replace_applicator
 0.45% :     0.000001s :      6: predicate.replace_old_param
 0.52% :     0.000001s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.03% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.75% :     0.000003s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 0.83% :     0.000001s :      6: predicate.specialize_transform
 1.12% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.31% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.16% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.05% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000002s :     10: predicate.tile_eliminate
 1.06% :     0.000002s :     10: predicate.transpose_eliminate
 1.25% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 2.75% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.31% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.51% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.89% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000703    12
45.09% :     0.000317s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.44% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.47% :     0.000341s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.031263    76
 0.20% :     0.000064s :      1: add_recomputation
 0.21% :     0.000065s :      1: auto_monad
 0.07% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.08% :     0.000338s :      1: bootstrap
 0.15% :     0.000047s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000028s :      1: cse_after_recomputation
 0.04% :     0.000012s :      1: environ_conv
 0.07% :     0.000022s :      1: event_method
 0.05% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
27.12% :     0.008478s :      1: jit_opt_a
 0.68% :     0.000213s :      1: jit_opt_after_cconv
 0.30% :     0.000095s :      1: jit_opt_b
 2.04% :     0.000639s :      1: loop_unroll
14.69% :     0.004593s :      1: mutable_eliminate
 3.04% :     0.000950s :     26: opt.transform.jit_opt_a
 0.20% :     0.000064s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000053s :      4: opt.transform.jit_opt_b
 0.05% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.15% :     0.000047s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000042s :      4: opt.transform.symbol_engine_opt
 2.04% :     0.000639s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000043s :      1: py_interpret_to_execute
 0.06% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000021s :      1: remove_dup_value
 1.80% :     0.000563s :      1: renormalize.infer
 1.57% :     0.000490s :      1: renormalize.specialize
 0.05% :     0.000016s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000045s :      1: rewriter_after_opt_a
 0.23% :     0.000073s :      1: rewriter_before_opt_a
 0.31% :     0.000098s :      1: symbol_engine_optimizer
26.49% :     0.008280s :      1: task_emit
16.06% :     0.005020s :      1: type_inference
 0.24% :     0.000075s :      1: validate


TotalTime = 0.0242517, [33]
[bootstrap]: 0.00029148
[type_inference]: 0.00454356
[event_method]: 1.67e-05
[auto_monad]: 6.241e-05
[graph_reusing]: 5.08002e-06
[pre_auto_parallel]: 2.73e-06
[py_interpret_to_execute]: 3.049e-05
[rewriter_before_opt_a]: 7.175e-05
[expand_dump_flag]: 3.09999e-06
[jit_opt_a]: 0.0109036, [2]
    [Cycle 1]: 0.00178048, [27]
        [switch_simplify]: 5.906e-05
        [loop_unroll]: 2.849e-05
        [a_1]: 0.00065026
        [with_stream_mark]: 2.146e-05
        [recompute_prepare]: 7.52998e-06
        [updatestate_depend_eliminate]: 3.91999e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 5.64998e-06
        [updatestate_useless_node_eliminater]: 5.25999e-06
        [accelerated_algorithm]: 5.10001e-06
        [meta_shard_fg_expand]: 1.99999e-06
        [get_grad_eliminate_]: 5.47001e-06
        [merge_forward]: 3.24001e-06
        [cell_reuse_recompute_pass]: 1.19998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.424e-05
        [j_node_and_user_rematch]: 8.67e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 9.21002e-06
        [inline_without_move]: 5.37999e-06
        [renormalize]: 0.00069555
        [add_forward_monad_depend]: 5.40999e-06
        [auto_monad_grad]: 2.19999e-06
        [auto_monad_eliminator]: 1.421e-05
        [cse]: 2.683e-05
        [replace_applicator]: 1.171e-05
    [Cycle 2]: 0.00031625, [27]
        [switch_simplify]: 5.56e-06
        [loop_unroll]: 4.76002e-06
        [a_1]: 8.897e-05
        [with_stream_mark]: 9.31e-06
        [recompute_prepare]: 4.95001e-06
        [updatestate_depend_eliminate]: 2.68e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 4.91997e-06
        [updatestate_useless_node_eliminater]: 4.92999e-06
        [accelerated_algorithm]: 4.82e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.49002e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 2.14999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.283e-05
        [j_node_and_user_rematch]: 8.02e-06
        [meta_fg_expand]: 1.60001e-06
        [replace_old_param]: 7.02002e-06
        [inline_without_move]: 9.23002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 4.93001e-06
        [cse]: 9.51e-06
        [replace_applicator]: 5.05001e-06
[py_interpret_to_execute_after_opt_a]: 1.098e-05
[rewriter_after_opt_a]: 3.968e-05
[convert_after_rewriter]: 6.99001e-06
[order_py_execute_after_rewriter]: 5.05001e-06
[mutable_eliminate]: 0.00057298
[jit_opt_b]: 4.798e-05, [1]
    [Cycle 1]: 4.095e-05, [2]
        [frontend_op_eliminate]: 1.59e-05
        [inline_after_opt_a]: 1.404e-05
[cconv]: 2.645e-05
[loop_unroll]: 0.00042529
[jit_opt_after_cconv]: 0.00013901, [1]
    [Cycle 1]: 0.00013226, [11]
        [c_1]: 2.011e-05
        [parameter_eliminate]: 2.89999e-06
        [updatestate_depend_eliminate]: 6.28e-06
        [updatestate_assign_eliminate]: 2.28998e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.892e-05
        [call_graph_tuple_transform]: 1.765e-05
        [tuple_list_get_item_eliminator]: 5.05999e-06
        [none_parameter_eliminate]: 1.81998e-06
        [renormalize]: 6.29982e-07
        [switch_simplify]: 5.25001e-06
[remove_dup_value]: 1.31e-05
[partial_unused_args_eliminate]: 2.07999e-06
[environ_conv]: 5.14e-06
[add_recomputation]: 4.835e-05
[cse_after_recomputation]: 2.026e-05, [1]
    [Cycle 1]: 1.45e-05, [1]
        [cse]: 8.79e-06
[auto_monad_reorder]: 1.422e-05
[get_jit_bprop_graph]: 1.90001e-06
[rewriter_after_jit_bprop_graph]: 5.95002e-06
[opt_after_jit_grad]: 0.00046022
[symbol_engine_optimizer]: 6.992e-05, [1]
    [Cycle 1]: 6.384e-05, [6]
        [build]: 3.09999e-06
        [elim_shapecalc]: 7.31001e-06
        [elim_not_effective]: 1.242e-05
        [opt_reshape]: 5.22e-06
        [fold_const_symbol]: 8.42998e-06
        [renormalize]: 4.50003e-07
[validate]: 3.473e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00618167
[execute]: 7.63001e-06
Sums
    bootstrap                            :     0.000291s :  1.96%
    type_inference                       :     0.004544s : 30.58%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000062s :  0.42%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.21%
    rewriter_before_opt_a                :     0.000072s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.43%
    jit_opt_a.loop_unroll                :     0.000033s :  0.22%
    jit_opt_a.a_1                        :     0.000739s :  4.98%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000015s :  0.10%
    jit_opt_a.renormalize                :     0.000696s :  4.68%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000036s :  0.24%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000040s :  0.27%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000573s :  3.86%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000026s :  0.18%
    loop_unroll                          :     0.000425s :  2.86%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000048s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000460s :  3.10%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006182s : 41.61%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000325    23
 0.55% :     0.000002s :      2: substitution.elim_not_effective
 0.40% :     0.000001s :      2: substitution.fold_const_symbol
 1.54% :     0.000005s :      3: substitution.graph_param_transform
90.04% :     0.000293s :      4: substitution.inline
 1.00% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.27% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.45% :     0.000005s :      2: substitution.replace_old_param
 3.74% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004483     2
82.58% :     0.003702s :      1: type_inference.infer
17.42% :     0.000781s :      1: type_inference.specialize
------[replace.]   0.000067     6
69.21% :     0.000046s :      4: replace.inline
30.79% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000301     6
96.31% :     0.000290s :      4: match.inline
 3.69% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.33% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.54% :     0.000002s :     10: predicate.addn_zero_filter
 2.05% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.06% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.50% :     0.000001s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.87% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.97% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.96% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 2.22% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.72% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.65% :     0.000009s :     53: predicate.switch_simplify
 1.32% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.87% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.16% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000570    12
48.49% :     0.000276s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.94% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.57% :     0.000248s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025997    76
 0.20% :     0.000051s :      1: add_recomputation
 0.25% :     0.000066s :      1: auto_monad
 0.06% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.17% :     0.000305s :      1: bootstrap
 0.11% :     0.000029s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000020s :      1: event_method
 0.04% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
41.95% :     0.010907s :      1: jit_opt_a
 0.54% :     0.000142s :      1: jit_opt_after_cconv
 0.19% :     0.000050s :      1: jit_opt_b
 1.67% :     0.000433s :      1: loop_unroll
 2.24% :     0.000581s :      1: mutable_eliminate
 3.66% :     0.000951s :     26: opt.transform.jit_opt_a
 0.17% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000024s :      4: opt.transform.jit_opt_b
 0.04% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000030s :      4: opt.transform.symbol_engine_opt
 1.80% :     0.000468s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000033s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000015s :      1: remove_dup_value
 1.45% :     0.000378s :      1: renormalize.infer
 1.19% :     0.000309s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000043s :      1: rewriter_after_opt_a
 0.29% :     0.000075s :      1: rewriter_before_opt_a
 0.28% :     0.000072s :      1: symbol_engine_optimizer
23.82% :     0.006192s :      1: task_emit
17.52% :     0.004555s :      1: type_inference
 0.21% :     0.000055s :      1: validate


TotalTime = 0.0180874, [33]
[bootstrap]: 0.00028287
[type_inference]: 0.00490827
[event_method]: 1.703e-05
[auto_monad]: 6.221e-05
[graph_reusing]: 5.71998e-06
[pre_auto_parallel]: 2.39999e-06
[py_interpret_to_execute]: 3.513e-05
[rewriter_before_opt_a]: 7.231e-05
[expand_dump_flag]: 3.16999e-06
[jit_opt_a]: 0.00484172, [2]
    [Cycle 1]: 0.00136683, [27]
        [switch_simplify]: 5.082e-05
        [loop_unroll]: 2.901e-05
        [a_1]: 0.00049011
        [with_stream_mark]: 1.422e-05
        [recompute_prepare]: 7.05998e-06
        [updatestate_depend_eliminate]: 3.11999e-06
        [updatestate_assign_eliminate]: 3.33e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 1.76e-06
        [specialize_transform]: 6.36e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.07e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 4.82e-06
        [merge_forward]: 3.45e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.337e-05
        [j_node_and_user_rematch]: 8.09002e-06
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 8.38999e-06
        [inline_without_move]: 4.95999e-06
        [renormalize]: 0.00048614
        [add_forward_monad_depend]: 4.93001e-06
        [auto_monad_grad]: 1.82001e-06
        [auto_monad_eliminator]: 1.265e-05
        [cse]: 2.347e-05
        [replace_applicator]: 1.079e-05
    [Cycle 2]: 0.00033652, [27]
        [switch_simplify]: 5.67999e-06
        [loop_unroll]: 4.71997e-06
        [a_1]: 8.94e-05
        [with_stream_mark]: 7.65e-06
        [recompute_prepare]: 4.74e-06
        [updatestate_depend_eliminate]: 2.43e-06
        [updatestate_assign_eliminate]: 1.98997e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 8.89995e-07
        [specialize_transform]: 4.65001e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 4.48999e-06
        [meta_shard_fg_expand]: 1.15999e-06
        [get_grad_eliminate_]: 4.33001e-06
        [merge_forward]: 2.28998e-06
        [cell_reuse_recompute_pass]: 1.89999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.142e-05
        [j_node_and_user_rematch]: 7.89002e-06
        [meta_fg_expand]: 1.47999e-06
        [replace_old_param]: 6.54001e-06
        [inline_without_move]: 4.38001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.44998e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.44998e-06
        [cse]: 1.022e-05
        [replace_applicator]: 5.27999e-06
[py_interpret_to_execute_after_opt_a]: 6.75002e-06
[rewriter_after_opt_a]: 2.989e-05
[convert_after_rewriter]: 6.33e-06
[order_py_execute_after_rewriter]: 4.84e-06
[mutable_eliminate]: 0.00047296
[jit_opt_b]: 4.665e-05, [1]
    [Cycle 1]: 4.015e-05, [2]
        [frontend_op_eliminate]: 1.542e-05
        [inline_after_opt_a]: 1.425e-05
[cconv]: 2.102e-05
[loop_unroll]: 0.00043169
[jit_opt_after_cconv]: 0.00013579, [1]
    [Cycle 1]: 0.00012963, [11]
        [c_1]: 2.169e-05
        [parameter_eliminate]: 1.97999e-06
        [updatestate_depend_eliminate]: 4.99e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [cse]: 1.555e-05
        [call_graph_tuple_transform]: 1.765e-05
        [tuple_list_get_item_eliminator]: 5.69e-06
        [none_parameter_eliminate]: 1.24998e-06
        [renormalize]: 5.00004e-07
        [switch_simplify]: 5.86e-06
[remove_dup_value]: 1.27e-05
[partial_unused_args_eliminate]: 2.11e-06
[environ_conv]: 4.80999e-06
[add_recomputation]: 4.348e-05
[cse_after_recomputation]: 2.071e-05, [1]
    [Cycle 1]: 1.525e-05, [1]
        [cse]: 9.49e-06
[auto_monad_reorder]: 1.429e-05
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 5.17999e-06
[opt_after_jit_grad]: 0.0004687
[symbol_engine_optimizer]: 7.289e-05, [1]
    [Cycle 1]: 6.675e-05, [6]
        [build]: 3.12002e-06
        [elim_shapecalc]: 7.42002e-06
        [elim_not_effective]: 1.249e-05
        [opt_reshape]: 5.89e-06
        [fold_const_symbol]: 9.14998e-06
        [renormalize]: 4.09986e-07
[validate]: 2.941e-05
[backend_pass]: 8.29983e-07
[task_emit]: 0.00584019
[execute]: 7.03998e-06
Sums
    bootstrap                            :     0.000283s :  1.97%
    type_inference                       :     0.004908s : 34.19%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000062s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.24%
    rewriter_before_opt_a                :     0.000072s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000580s :  4.04%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000486s :  3.39%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000034s :  0.23%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.05%
    rewriter_after_opt_a                 :     0.000030s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000473s :  3.29%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000021s :  0.15%
    loop_unroll                          :     0.000432s :  3.01%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000043s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000469s :  3.27%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005840s : 40.69%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000168    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.88% :     0.000001s :      2: substitution.fold_const_symbol
 2.95% :     0.000005s :      3: substitution.graph_param_transform
82.05% :     0.000138s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.60% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.97% :     0.000003s :      2: substitution.replace_old_param
 6.90% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004843     2
83.50% :     0.004044s :      1: type_inference.infer
16.50% :     0.000799s :      1: type_inference.specialize
------[replace.]   0.000064     6
66.80% :     0.000043s :      4: replace.inline
33.20% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000146     6
92.79% :     0.000135s :      4: match.inline
 7.21% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.29% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     10: predicate.addn_check_dump
 1.54% :     0.000002s :     10: predicate.addn_zero_filter
 2.17% :     0.000002s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.19% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.44% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.41% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.30% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.11% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.04% :     0.000002s :     15: predicate.load_eliminater
 1.48% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.38% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     15: predicate.partial_eliminate
 1.26% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.56% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.77% :     0.000002s :     10: predicate.reshape_eliminate
 1.38% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.97% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.44% :     0.000009s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.60% :     0.000002s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.85% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 2.00% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.45% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000565    12
53.93% :     0.000304s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 3.80% :     0.000021s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.27% :     0.000239s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019457    76
 0.24% :     0.000046s :      1: add_recomputation
 0.34% :     0.000065s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.53% :     0.000298s :      1: bootstrap
 0.12% :     0.000024s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
24.90% :     0.004845s :      1: jit_opt_a
 0.71% :     0.000138s :      1: jit_opt_after_cconv
 0.25% :     0.000049s :      1: jit_opt_b
 2.26% :     0.000439s :      1: loop_unroll
 2.47% :     0.000481s :      1: mutable_eliminate
 3.99% :     0.000776s :     26: opt.transform.jit_opt_a
 0.24% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.44% :     0.000475s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000038s :      1: py_interpret_to_execute
 0.05% :     0.000009s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.28% :     0.000249s :      1: renormalize.infer
 1.19% :     0.000231s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000033s :      1: rewriter_after_opt_a
 0.39% :     0.000075s :      1: rewriter_before_opt_a
 0.39% :     0.000075s :      1: symbol_engine_optimizer
30.06% :     0.005848s :      1: task_emit
25.29% :     0.004922s :      1: type_inference
 0.24% :     0.000046s :      1: validate


TotalTime = 0.0128099, [30]
[bootstrap]: 0.00025484
[type_inference]: 0.00556958
[event_method]: 1.974e-05
[auto_monad]: 6.115e-05
[graph_reusing]: 4.42e-06
[pre_auto_parallel]: 2.04e-06
[py_interpret_to_execute]: 4.745e-05
[rewriter_before_opt_a]: 8.124e-05
[expand_dump_flag]: 2.45002e-06
[jit_opt_a]: 0.004836, [2]
    [Cycle 1]: 0.00159722, [27]
        [switch_simplify]: 8.756e-05
        [loop_unroll]: 3.666e-05
        [a_1]: 0.00066304
        [with_stream_mark]: 1.085e-05
        [recompute_prepare]: 7.11001e-06
        [updatestate_depend_eliminate]: 2.93e-06
        [updatestate_assign_eliminate]: 2.10002e-06
        [updatestate_loads_eliminate]: 1.91003e-06
        [parameter_eliminate]: 1.31002e-06
        [specialize_transform]: 5.84999e-06
        [updatestate_useless_node_eliminater]: 5.17999e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 5.34998e-06
        [merge_forward]: 2.73e-06
        [cell_reuse_recompute_pass]: 9.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.131e-05
        [j_node_and_user_rematch]: 8.92e-06
        [meta_fg_expand]: 2.01003e-06
        [replace_old_param]: 7.64002e-06
        [inline_without_move]: 5.35999e-06
        [renormalize]: 0.00053103
        [add_forward_monad_depend]: 4.07e-06
        [auto_monad_grad]: 1.22999e-06
        [auto_monad_eliminator]: 8.92e-06
        [cse]: 1.347e-05
        [replace_applicator]: 1.008e-05
    [Cycle 2]: 0.00032856, [27]
        [switch_simplify]: 5.87999e-06
        [loop_unroll]: 5.32999e-06
        [a_1]: 9.297e-05
        [with_stream_mark]: 8.37e-06
        [recompute_prepare]: 5.11002e-06
        [updatestate_depend_eliminate]: 2.58e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 5.10001e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 5.49998e-06
        [meta_shard_fg_expand]: 1.16002e-06
        [get_grad_eliminate_]: 4.59002e-06
        [merge_forward]: 2.43998e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.14e-05
        [j_node_and_user_rematch]: 7.97e-06
        [meta_fg_expand]: 1.63002e-06
        [replace_old_param]: 7.18e-06
        [inline_without_move]: 1.375e-05
        [renormalize]: 1.10012e-07
        [add_forward_monad_depend]: 1.40001e-06
        [auto_monad_grad]: 9.80013e-07
        [auto_monad_eliminator]: 5.17e-06
        [cse]: 9.96e-06
        [replace_applicator]: 5.10999e-06
[py_interpret_to_execute_after_opt_a]: 7.7e-06
[rewriter_after_opt_a]: 2.412e-05
[convert_after_rewriter]: 5.22999e-06
[order_py_execute_after_rewriter]: 4.17e-06
[mutable_eliminate]: 0.00046561
[jit_opt_b]: 4.84e-05, [1]
    [Cycle 1]: 4.259e-05, [2]
        [frontend_op_eliminate]: 1.648e-05
        [inline_after_opt_a]: 1.525e-05
[cconv]: 1.594e-05
[loop_unroll]: 0.00041617
[jit_opt_after_cconv]: 0.0001319, [1]
    [Cycle 1]: 0.00012566, [11]
        [c_1]: 2.134e-05
        [parameter_eliminate]: 2.06998e-06
        [updatestate_depend_eliminate]: 4.2e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [cse]: 1.516e-05
        [call_graph_tuple_transform]: 1.576e-05
        [tuple_list_get_item_eliminator]: 5.42999e-06
        [none_parameter_eliminate]: 9.50007e-07
        [renormalize]: 5.99975e-07
        [switch_simplify]: 5.67001e-06
[remove_dup_value]: 6.54001e-06
[partial_unused_args_eliminate]: 1.66e-06
[environ_conv]: 3.56001e-06
[add_recomputation]: 2.975e-05
[cse_after_recomputation]: 1.989e-05, [1]
    [Cycle 1]: 1.486e-05, [1]
        [cse]: 9.24998e-06
[auto_monad_reorder]: 9.17001e-06
[get_jit_bprop_graph]: 1.51002e-06
[rewriter_after_jit_bprop_graph]: 3.863e-05
[opt_after_jit_grad]: 0.00045888
[symbol_engine_optimizer]: 6.993e-05, [1]
    [Cycle 1]: 6.364e-05, [6]
        [build]: 2.43998e-06
        [elim_shapecalc]: 7.78001e-06
        [elim_not_effective]: 1.142e-05
        [opt_reshape]: 5.81e-06
        [fold_const_symbol]: 8.28999e-06
        [renormalize]: 6.00005e-07
[validate]: 2.546e-05
Sums
    bootstrap                            :     0.000255s :  2.72%
    type_inference                       :     0.005570s : 59.45%
    event_method                         :     0.000020s :  0.21%
    auto_monad                           :     0.000061s :  0.65%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000047s :  0.51%
    rewriter_before_opt_a                :     0.000081s :  0.87%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000093s :  1.00%
    jit_opt_a.loop_unroll                :     0.000042s :  0.45%
    jit_opt_a.a_1                        :     0.000756s :  8.07%
    jit_opt_a.with_stream_mark           :     0.000019s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000005s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.16%
    jit_opt_a.inline_without_move        :     0.000019s :  0.20%
    jit_opt_a.renormalize                :     0.000531s :  5.67%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.15%
    jit_opt_a.cse                        :     0.000023s :  0.25%
    jit_opt_a.replace_applicator         :     0.000015s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.08%
    rewriter_after_opt_a                 :     0.000024s :  0.26%
    convert_after_rewriter               :     0.000005s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000466s :  4.97%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000016s :  0.17%
    loop_unroll                          :     0.000416s :  4.44%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000007s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000030s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000009s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000039s :  0.41%
    opt_after_jit_grad                   :     0.000459s :  4.90%
    symbol_engine_optimizer.build        :     0.000002s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000025s :  0.27%


Time group info:
------[substitution.]   0.000221    26
 0.58% :     0.000001s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 1.59% :     0.000003s :      3: substitution.graph_param_transform
86.06% :     0.000190s :      6: substitution.inline
 1.20% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.80% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.22% :     0.000003s :      2: substitution.replace_old_param
 3.15% :     0.000007s :      1: substitution.switch_simplify
 3.88% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005521     2
82.37% :     0.004547s :      1: type_inference.infer
17.63% :     0.000973s :      1: type_inference.specialize
------[replace.]   0.000119     9
55.69% :     0.000066s :      6: replace.inline
17.67% :     0.000021s :      1: replace.switch_simplify
26.64% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     9
93.24% :     0.000186s :      6: match.inline
 3.08% :     0.000006s :      1: match.switch_simplify
 3.68% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   801
 1.40% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.99% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.20% :     0.000002s :     12: predicate.addn_check_dump
 1.46% :     0.000002s :     12: predicate.addn_zero_filter
 1.96% :     0.000003s :     12: predicate.arithmetic_simplify
 1.33% :     0.000002s :     12: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000002s :     12: predicate.compare_switch_simplify
 1.40% :     0.000002s :     12: predicate.depend_value_elim
 1.36% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.53% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.62% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.46% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.33% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     12: predicate.environ_get_eliminate
 1.24% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 4.92% :     0.000007s :     26: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.97% :     0.000001s :      6: predicate.less_batch_normalization
 1.77% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     17: predicate.load_eliminater
 1.09% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.60% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.01% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000002s :     12: predicate.merge_addn
 1.23% :     0.000002s :     12: predicate.minmaximum_grad
 1.14% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     17: predicate.partial_eliminate
 1.29% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.74% :     0.000002s :     12: predicate.reduce_eliminate
 1.55% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     20: predicate.replace_applicator
 0.43% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.64% :     0.000002s :     12: predicate.reshape_eliminate
 1.32% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     12: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.80% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.68% :     0.000004s :     20: predicate.switch_defer_inline
 2.37% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.43% :     0.000011s :     65: predicate.switch_simplify
 1.48% :     0.000002s :     12: predicate.tile_eliminate
 1.52% :     0.000002s :     12: predicate.transpose_eliminate
 1.58% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.35% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.84% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000690    15
46.85% :     0.000323s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.71% :     0.000074s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.44% :     0.000293s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014417    72
 0.23% :     0.000033s :      1: add_recomputation
 0.45% :     0.000065s :      1: auto_monad
 0.08% :     0.000012s :      1: auto_monad_reorder
 1.84% :     0.000265s :      1: bootstrap
 0.13% :     0.000019s :      1: cconv
 0.05% :     0.000007s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.16% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
33.56% :     0.004839s :      1: jit_opt_a
 0.94% :     0.000135s :      1: jit_opt_after_cconv
 0.35% :     0.000051s :      1: jit_opt_b
 2.94% :     0.000423s :      1: loop_unroll
 3.28% :     0.000472s :      1: mutable_eliminate
 7.01% :     0.001010s :     26: opt.transform.jit_opt_a
 0.31% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.23% :     0.000465s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.35% :     0.000050s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000009s :      1: remove_dup_value
 1.83% :     0.000264s :      1: renormalize.infer
 1.81% :     0.000261s :      1: renormalize.specialize
 0.29% :     0.000041s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000027s :      1: rewriter_after_opt_a
 0.58% :     0.000084s :      1: rewriter_before_opt_a
 0.50% :     0.000073s :      1: symbol_engine_optimizer
38.70% :     0.005579s :      1: type_inference


TotalTime = 0.0217262, [33]
[bootstrap]: 0.00033898
[type_inference]: 0.00500605
[event_method]: 1.888e-05
[auto_monad]: 5.9e-05
[graph_reusing]: 4.37998e-06
[pre_auto_parallel]: 2.24001e-06
[py_interpret_to_execute]: 4.617e-05
[rewriter_before_opt_a]: 7.355e-05
[expand_dump_flag]: 3.13e-06
[jit_opt_a]: 0.0066373, [2]
    [Cycle 1]: 0.00205054, [27]
        [switch_simplify]: 6.326e-05
        [loop_unroll]: 2.906e-05
        [a_1]: 0.00059731
        [with_stream_mark]: 1.903e-05
        [recompute_prepare]: 7.83001e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 1.23002e-06
        [specialize_transform]: 6.17001e-06
        [updatestate_useless_node_eliminater]: 5.08002e-06
        [accelerated_algorithm]: 6.27001e-06
        [meta_shard_fg_expand]: 2.15002e-06
        [get_grad_eliminate_]: 5.12999e-06
        [merge_forward]: 4.08001e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.413e-05
        [j_node_and_user_rematch]: 1.001e-05
        [meta_fg_expand]: 2.49999e-06
        [replace_old_param]: 1.06e-05
        [inline_without_move]: 5.69999e-06
        [renormalize]: 0.00099911
        [add_forward_monad_depend]: 7.00002e-06
        [auto_monad_grad]: 2.04999e-06
        [auto_monad_eliminator]: 1.789e-05
        [cse]: 2.618e-05
        [replace_applicator]: 1.565e-05
    [Cycle 2]: 0.00036158, [27]
        [switch_simplify]: 5.67999e-06
        [loop_unroll]: 5.32001e-06
        [a_1]: 0.00010017
        [with_stream_mark]: 1.208e-05
        [recompute_prepare]: 6.14999e-06
        [updatestate_depend_eliminate]: 2.98998e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 6.09999e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 1.77999e-06
        [get_grad_eliminate_]: 4.89e-06
        [merge_forward]: 3.83001e-06
        [cell_reuse_recompute_pass]: 3.33998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.467e-05
        [j_node_and_user_rematch]: 8.69e-06
        [meta_fg_expand]: 2.26998e-06
        [replace_old_param]: 7.93001e-06
        [inline_without_move]: 1.152e-05
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 2.18998e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 6.43e-06
        [cse]: 1.1e-05
        [replace_applicator]: 5.25999e-06
[py_interpret_to_execute_after_opt_a]: 1.292e-05
[rewriter_after_opt_a]: 4.046e-05
[convert_after_rewriter]: 8.32998e-06
[order_py_execute_after_rewriter]: 5.58002e-06
[mutable_eliminate]: 0.00081981
[jit_opt_b]: 5.937e-05, [1]
    [Cycle 1]: 5.152e-05, [2]
        [frontend_op_eliminate]: 1.966e-05
        [inline_after_opt_a]: 1.806e-05
[cconv]: 3.194e-05
[loop_unroll]: 0.00052191
[jit_opt_after_cconv]: 0.00016502, [1]
    [Cycle 1]: 0.00015687, [11]
        [c_1]: 2.334e-05
        [parameter_eliminate]: 4e-06
        [updatestate_depend_eliminate]: 6.91999e-06
        [updatestate_assign_eliminate]: 3.71999e-06
        [updatestate_loads_eliminate]: 2.96999e-06
        [cse]: 2.885e-05
        [call_graph_tuple_transform]: 2.065e-05
        [tuple_list_get_item_eliminator]: 5.57001e-06
        [none_parameter_eliminate]: 1.42999e-06
        [renormalize]: 9.99979e-07
        [switch_simplify]: 5.52001e-06
[remove_dup_value]: 1.427e-05
[partial_unused_args_eliminate]: 2.28002e-06
[environ_conv]: 6.13998e-06
[add_recomputation]: 5.243e-05
[cse_after_recomputation]: 2.259e-05, [1]
    [Cycle 1]: 1.68e-05, [1]
        [cse]: 9.43002e-06
[auto_monad_reorder]: 1.529e-05
[get_jit_bprop_graph]: 2.32999e-06
[rewriter_after_jit_bprop_graph]: 8.58001e-06
[opt_after_jit_grad]: 0.00055341
[symbol_engine_optimizer]: 8.255e-05, [1]
    [Cycle 1]: 7.392e-05, [6]
        [build]: 4.80999e-06
        [elim_shapecalc]: 7.9e-06
        [elim_not_effective]: 1.587e-05
        [opt_reshape]: 5.72999e-06
        [fold_const_symbol]: 8.95001e-06
        [renormalize]: 9.90025e-07
[validate]: 4.416e-05
[backend_pass]: 1.04e-06
[task_emit]: 0.00682602
[execute]: 1.021e-05
Sums
    bootstrap                            :     0.000339s :  2.01%
    type_inference                       :     0.005006s : 29.74%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000059s :  0.35%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000046s :  0.27%
    rewriter_before_opt_a                :     0.000074s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000069s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000697s :  4.14%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000017s :  0.10%
    jit_opt_a.renormalize                :     0.000999s :  5.94%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.14%
    jit_opt_a.cse                        :     0.000037s :  0.22%
    jit_opt_a.replace_applicator         :     0.000021s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000040s :  0.24%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000820s :  4.87%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.11%
    cconv                                :     0.000032s :  0.19%
    loop_unroll                          :     0.000522s :  3.10%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000029s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000052s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000553s :  3.29%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006826s : 40.56%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000242    23
 1.19% :     0.000003s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.31% :     0.000006s :      3: substitution.graph_param_transform
84.95% :     0.000206s :      4: substitution.inline
 1.49% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.06% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.24% :     0.000005s :      2: substitution.replace_old_param
 5.26% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004932     2
80.59% :     0.003974s :      1: type_inference.infer
19.41% :     0.000957s :      1: type_inference.specialize
------[replace.]   0.000085     6
67.44% :     0.000057s :      4: replace.inline
32.56% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000215     6
94.58% :     0.000203s :      4: match.inline
 5.42% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.32% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.70% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.46% :     0.000002s :     10: predicate.addn_zero_filter
 2.18% :     0.000003s :     10: predicate.arithmetic_simplify
 1.37% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.48% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.09% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.70% :     0.000001s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.66% :     0.000002s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 4.97% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.56% :     0.000002s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.56% :     0.000002s :     15: predicate.load_eliminater
 1.72% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.47% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.49% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 2.51% :     0.000003s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.04% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.27% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.45% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.47% :     0.000001s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.05% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.06% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.35% :     0.000002s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.70% :     0.000003s :     16: predicate.switch_defer_inline
 1.81% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.98% :     0.000009s :     53: predicate.switch_simplify
 1.56% :     0.000002s :     10: predicate.tile_eliminate
 1.20% :     0.000002s :     10: predicate.transpose_eliminate
 1.59% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.07% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.63% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 2.18% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.48% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000716    12
41.53% :     0.000297s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.97% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
53.50% :     0.000383s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023773    76
 0.23% :     0.000055s :      1: add_recomputation
 0.26% :     0.000062s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.50% :     0.000358s :      1: bootstrap
 0.15% :     0.000035s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
27.94% :     0.006641s :      1: jit_opt_a
 0.71% :     0.000168s :      1: jit_opt_after_cconv
 0.26% :     0.000062s :      1: jit_opt_b
 2.23% :     0.000531s :      1: loop_unroll
 3.49% :     0.000831s :      1: mutable_eliminate
 3.93% :     0.000935s :     26: opt.transform.jit_opt_a
 0.22% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000030s :      4: opt.transform.jit_opt_b
 0.06% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.37% :     0.000564s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000049s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.13% :     0.000506s :      1: renormalize.infer
 2.03% :     0.000484s :      1: renormalize.specialize
 0.05% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000045s :      1: rewriter_after_opt_a
 0.32% :     0.000076s :      1: rewriter_before_opt_a
 0.36% :     0.000086s :      1: symbol_engine_optimizer
28.79% :     0.006844s :      1: task_emit
21.13% :     0.005023s :      1: type_inference
 0.27% :     0.000064s :      1: validate


TotalTime = 0.017035, [33]
[bootstrap]: 0.00028125
[type_inference]: 0.00406567
[event_method]: 1.62e-05
[auto_monad]: 5.913e-05
[graph_reusing]: 5.67001e-06
[pre_auto_parallel]: 2.66e-06
[py_interpret_to_execute]: 2.986e-05
[rewriter_before_opt_a]: 7.023e-05
[expand_dump_flag]: 2.94001e-06
[jit_opt_a]: 0.00472841, [2]
    [Cycle 1]: 0.00140746, [27]
        [switch_simplify]: 5.047e-05
        [loop_unroll]: 2.857e-05
        [a_1]: 0.00047149
        [with_stream_mark]: 1.448e-05
        [recompute_prepare]: 7.27002e-06
        [updatestate_depend_eliminate]: 3.85998e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [parameter_eliminate]: 1.77001e-06
        [specialize_transform]: 5.57001e-06
        [updatestate_useless_node_eliminater]: 4.71002e-06
        [accelerated_algorithm]: 5.15001e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 5.17e-06
        [merge_forward]: 3.60003e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.28e-05
        [j_node_and_user_rematch]: 8.33001e-06
        [meta_fg_expand]: 2.22999e-06
        [replace_old_param]: 9.44998e-06
        [inline_without_move]: 5.32999e-06
        [renormalize]: 0.00047815
        [add_forward_monad_depend]: 5.50001e-06
        [auto_monad_grad]: 1.84e-06
        [auto_monad_eliminator]: 1.385e-05
        [cse]: 2.603e-05
        [replace_applicator]: 1.142e-05
    [Cycle 2]: 0.00031071, [27]
        [switch_simplify]: 6.09001e-06
        [loop_unroll]: 4.68001e-06
        [a_1]: 8.694e-05
        [with_stream_mark]: 8.23001e-06
        [recompute_prepare]: 5.14998e-06
        [updatestate_depend_eliminate]: 2.35002e-06
        [updatestate_assign_eliminate]: 2.00002e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [parameter_eliminate]: 1.01997e-06
        [specialize_transform]: 4.57e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 5.16002e-06
        [meta_shard_fg_expand]: 1.05001e-06
        [get_grad_eliminate_]: 4.38001e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.55001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.186e-05
        [j_node_and_user_rematch]: 7.33e-06
        [meta_fg_expand]: 1.42999e-06
        [replace_old_param]: 7.52002e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 4.93001e-06
        [cse]: 9.86e-06
        [replace_applicator]: 4.90001e-06
[py_interpret_to_execute_after_opt_a]: 7.73999e-06
[rewriter_after_opt_a]: 3.13e-05
[convert_after_rewriter]: 6.89999e-06
[order_py_execute_after_rewriter]: 4.57998e-06
[mutable_eliminate]: 0.00046794
[jit_opt_b]: 4.843e-05, [1]
    [Cycle 1]: 4.182e-05, [2]
        [frontend_op_eliminate]: 1.56e-05
        [inline_after_opt_a]: 1.521e-05
[cconv]: 2.397e-05
[loop_unroll]: 0.00040716
[jit_opt_after_cconv]: 0.00013015, [1]
    [Cycle 1]: 0.00012361, [11]
        [c_1]: 1.933e-05
        [parameter_eliminate]: 2.52001e-06
        [updatestate_depend_eliminate]: 4.97999e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 1.98002e-06
        [cse]: 1.542e-05
        [call_graph_tuple_transform]: 1.686e-05
        [tuple_list_get_item_eliminator]: 5.00001e-06
        [none_parameter_eliminate]: 1.33002e-06
        [renormalize]: 5.99975e-07
        [switch_simplify]: 4.86002e-06
[remove_dup_value]: 1.287e-05
[partial_unused_args_eliminate]: 2.42001e-06
[environ_conv]: 4.65001e-06
[add_recomputation]: 4.518e-05
[cse_after_recomputation]: 1.858e-05, [1]
    [Cycle 1]: 1.344e-05, [1]
        [cse]: 8.13999e-06
[auto_monad_reorder]: 1.397e-05
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 4.75001e-06
[opt_after_jit_grad]: 0.00044233
[symbol_engine_optimizer]: 6.792e-05, [1]
    [Cycle 1]: 6.188e-05, [6]
        [build]: 3.2e-06
        [elim_shapecalc]: 7.03e-06
        [elim_not_effective]: 1.164e-05
        [opt_reshape]: 5.20001e-06
        [fold_const_symbol]: 8.03999e-06
        [renormalize]: 5.60016e-07
[validate]: 2.984e-05
[backend_pass]: 1.35001e-06
[task_emit]: 0.00581696
[execute]: 7.77002e-06
Sums
    bootstrap                            :     0.000281s :  2.10%
    type_inference                       :     0.004066s : 30.34%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000059s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.22%
    rewriter_before_opt_a                :     0.000070s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.42%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000558s :  4.17%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.13%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000478s :  3.57%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.14%
    jit_opt_a.cse                        :     0.000036s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000468s :  3.49%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000407s :  3.04%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000045s :  0.34%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000442s :  3.30%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000030s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005817s : 43.42%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000166    23
 1.10% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000001s :      2: substitution.fold_const_symbol
 2.95% :     0.000005s :      3: substitution.graph_param_transform
81.27% :     0.000135s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.62% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.52% :     0.000004s :      2: substitution.replace_old_param
 7.24% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004011     2
81.54% :     0.003270s :      1: type_inference.infer
18.46% :     0.000741s :      1: type_inference.specialize
------[replace.]   0.000058     6
66.70% :     0.000039s :      4: replace.inline
33.30% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000143     6
92.44% :     0.000132s :      4: match.inline
 7.56% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000108   691
 1.29% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.25% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.74% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.15% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.45% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.38% :     0.000003s :     15: predicate.partial_eliminate
 1.23% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.91% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 1.20% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000001s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.52% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.14% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.78% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.34% :     0.000001s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000494    12
49.00% :     0.000242s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.31% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.69% :     0.000226s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018371    76
 0.26% :     0.000048s :      1: add_recomputation
 0.34% :     0.000062s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.61% :     0.000295s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.75% :     0.004731s :      1: jit_opt_a
 0.72% :     0.000133s :      1: jit_opt_after_cconv
 0.28% :     0.000051s :      1: jit_opt_b
 2.26% :     0.000415s :      1: loop_unroll
 2.59% :     0.000476s :      1: mutable_eliminate
 4.12% :     0.000757s :     26: opt.transform.jit_opt_a
 0.23% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.45% :     0.000450s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000032s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.30% :     0.000238s :      1: renormalize.infer
 1.27% :     0.000233s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000035s :      1: rewriter_after_opt_a
 0.40% :     0.000073s :      1: rewriter_before_opt_a
 0.38% :     0.000070s :      1: symbol_engine_optimizer
31.72% :     0.005828s :      1: task_emit
22.20% :     0.004078s :      1: type_inference
 0.25% :     0.000047s :      1: validate


TotalTime = 0.0179626, [33]
[bootstrap]: 0.00027309
[type_inference]: 0.00462068
[event_method]: 1.764e-05
[auto_monad]: 6.383e-05
[graph_reusing]: 5.87001e-06
[pre_auto_parallel]: 2.92002e-06
[py_interpret_to_execute]: 3.667e-05
[rewriter_before_opt_a]: 7.869e-05
[expand_dump_flag]: 3.35e-06
[jit_opt_a]: 0.00507121, [2]
    [Cycle 1]: 0.001529, [27]
        [switch_simplify]: 5.248e-05
        [loop_unroll]: 3.191e-05
        [a_1]: 0.00059497
        [with_stream_mark]: 1.69e-05
        [recompute_prepare]: 7.93001e-06
        [updatestate_depend_eliminate]: 3.35998e-06
        [updatestate_assign_eliminate]: 3.16999e-06
        [updatestate_loads_eliminate]: 2.99999e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 6.01e-06
        [updatestate_useless_node_eliminater]: 5.42999e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 2.12999e-06
        [get_grad_eliminate_]: 5.37001e-06
        [merge_forward]: 3.60998e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.56e-05
        [j_node_and_user_rematch]: 9.81e-06
        [meta_fg_expand]: 2.21998e-06
        [replace_old_param]: 9.33002e-06
        [inline_without_move]: 5.90002e-06
        [renormalize]: 0.00051485
        [add_forward_monad_depend]: 4.79e-06
        [auto_monad_grad]: 2.19001e-06
        [auto_monad_eliminator]: 1.297e-05
        [cse]: 2.317e-05
        [replace_applicator]: 1.231e-05
    [Cycle 2]: 0.00032673, [27]
        [switch_simplify]: 6.04999e-06
        [loop_unroll]: 5.57001e-06
        [a_1]: 9.472e-05
        [with_stream_mark]: 8.82999e-06
        [recompute_prepare]: 5.10001e-06
        [updatestate_depend_eliminate]: 2.82002e-06
        [updatestate_assign_eliminate]: 2.23002e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 5.34998e-06
        [updatestate_useless_node_eliminater]: 5.23002e-06
        [accelerated_algorithm]: 5.21998e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.91997e-06
        [merge_forward]: 2.36998e-06
        [cell_reuse_recompute_pass]: 1.49998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.264e-05
        [j_node_and_user_rematch]: 8.2e-06
        [meta_fg_expand]: 1.39e-06
        [replace_old_param]: 7.26999e-06
        [inline_without_move]: 4.89e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 4.84e-06
        [cse]: 9.56998e-06
        [replace_applicator]: 5.25999e-06
[py_interpret_to_execute_after_opt_a]: 7.77e-06
[rewriter_after_opt_a]: 3.076e-05
[convert_after_rewriter]: 8.18001e-06
[order_py_execute_after_rewriter]: 5.46998e-06
[mutable_eliminate]: 0.00050216
[jit_opt_b]: 5.134e-05, [1]
    [Cycle 1]: 4.451e-05, [2]
        [frontend_op_eliminate]: 1.715e-05
        [inline_after_opt_a]: 1.505e-05
[cconv]: 2.326e-05
[loop_unroll]: 0.00042369
[jit_opt_after_cconv]: 0.00014414, [1]
    [Cycle 1]: 0.00013778, [11]
        [c_1]: 1.993e-05
        [parameter_eliminate]: 2.63e-06
        [updatestate_depend_eliminate]: 5.55001e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [cse]: 1.636e-05
        [call_graph_tuple_transform]: 1.797e-05
        [tuple_list_get_item_eliminator]: 5.32001e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 5.14998e-06
[remove_dup_value]: 1.303e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 5.14998e-06
[add_recomputation]: 4.373e-05
[cse_after_recomputation]: 2.077e-05, [1]
    [Cycle 1]: 1.548e-05, [1]
        [cse]: 9.50001e-06
[auto_monad_reorder]: 1.369e-05
[get_jit_bprop_graph]: 1.54e-06
[rewriter_after_jit_bprop_graph]: 4.97999e-06
[opt_after_jit_grad]: 0.00044823
[symbol_engine_optimizer]: 6.916e-05, [1]
    [Cycle 1]: 6.334e-05, [6]
        [build]: 3.25002e-06
        [elim_shapecalc]: 6.79999e-06
        [elim_not_effective]: 1.123e-05
        [opt_reshape]: 5.59998e-06
        [fold_const_symbol]: 8.48999e-06
        [renormalize]: 3.99974e-07
[validate]: 2.794e-05
[backend_pass]: 8.00006e-07
[task_emit]: 0.00576365
[execute]: 6.74999e-06
Sums
    bootstrap                            :     0.000273s :  1.93%
    type_inference                       :     0.004621s : 32.63%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000064s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.26%
    rewriter_before_opt_a                :     0.000079s :  0.56%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.41%
    jit_opt_a.loop_unroll                :     0.000037s :  0.26%
    jit_opt_a.a_1                        :     0.000690s :  4.87%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.000515s :  3.64%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000033s :  0.23%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000031s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000502s :  3.55%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000023s :  0.16%
    loop_unroll                          :     0.000424s :  2.99%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000044s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000448s :  3.17%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005764s : 40.70%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000184    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.74% :     0.000005s :      3: substitution.graph_param_transform
83.33% :     0.000154s :      4: substitution.inline
 1.69% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.52% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.71% :     0.000003s :      2: substitution.replace_old_param
 6.30% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004562     2
81.66% :     0.003725s :      1: type_inference.infer
18.34% :     0.000837s :      1: type_inference.specialize
------[replace.]   0.000065     6
67.45% :     0.000044s :      4: replace.inline
32.55% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     6
93.51% :     0.000151s :      4: match.inline
 6.49% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.67% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.04% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.08% :     0.000003s :     10: predicate.arithmetic_simplify
 1.43% :     0.000002s :     10: predicate.cast_eliminate
 0.55% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.32% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     10: predicate.environ_get_eliminate
 1.26% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.31% :     0.000007s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.85% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.46% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.81% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.50% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.40% :     0.000001s :      3: predicate.reset_defer_inline
 1.58% :     0.000002s :     10: predicate.reshape_eliminate
 1.81% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.67% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.27% :     0.000002s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.20% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.71% :     0.000010s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.27% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.53% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000566    12
48.35% :     0.000274s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.97% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.68% :     0.000264s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019486    76
 0.24% :     0.000046s :      1: add_recomputation
 0.34% :     0.000067s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.46% :     0.000284s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.04% :     0.005074s :      1: jit_opt_a
 0.75% :     0.000147s :      1: jit_opt_after_cconv
 0.28% :     0.000054s :      1: jit_opt_b
 2.21% :     0.000431s :      1: loop_unroll
 2.62% :     0.000510s :      1: mutable_eliminate
 4.64% :     0.000904s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.34% :     0.000456s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000039s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.32% :     0.000257s :      1: renormalize.infer
 1.29% :     0.000251s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000034s :      1: rewriter_after_opt_a
 0.42% :     0.000082s :      1: rewriter_before_opt_a
 0.37% :     0.000072s :      1: symbol_engine_optimizer
29.62% :     0.005772s :      1: task_emit
23.77% :     0.004633s :      1: type_inference
 0.22% :     0.000044s :      1: validate


TotalTime = 0.0131105, [30]
[bootstrap]: 0.00026169
[type_inference]: 0.0059141
[event_method]: 1.859e-05
[auto_monad]: 6.049e-05
[graph_reusing]: 4.14002e-06
[pre_auto_parallel]: 1.97999e-06
[py_interpret_to_execute]: 4.699e-05
[rewriter_before_opt_a]: 7.59e-05
[expand_dump_flag]: 2.19001e-06
[jit_opt_a]: 0.00483586, [2]
    [Cycle 1]: 0.00157994, [27]
        [switch_simplify]: 9.065e-05
        [loop_unroll]: 3.653e-05
        [a_1]: 0.0006513
        [with_stream_mark]: 1.106e-05
        [recompute_prepare]: 8e-06
        [updatestate_depend_eliminate]: 2.99999e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 6.37001e-06
        [updatestate_useless_node_eliminater]: 5.39998e-06
        [accelerated_algorithm]: 5.77001e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 5.27001e-06
        [merge_forward]: 2.61999e-06
        [cell_reuse_recompute_pass]: 1.30001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.163e-05
        [j_node_and_user_rematch]: 9.17999e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 7.82e-06
        [inline_without_move]: 5.54998e-06
        [renormalize]: 0.00051315
        [add_forward_monad_depend]: 4.35e-06
        [auto_monad_grad]: 1.19e-06
        [auto_monad_eliminator]: 9.59999e-06
        [cse]: 1.442e-05
        [replace_applicator]: 1.128e-05
    [Cycle 2]: 0.00033635, [27]
        [switch_simplify]: 5.66003e-06
        [loop_unroll]: 5.24e-06
        [a_1]: 9.54e-05
        [with_stream_mark]: 8.66002e-06
        [recompute_prepare]: 5.16002e-06
        [updatestate_depend_eliminate]: 2.54001e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 9.30013e-07
        [specialize_transform]: 5.54e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.62999e-06
        [meta_shard_fg_expand]: 1.19e-06
        [get_grad_eliminate_]: 4.71002e-06
        [merge_forward]: 2.58e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.222e-05
        [j_node_and_user_rematch]: 8.01001e-06
        [meta_fg_expand]: 1.56998e-06
        [replace_old_param]: 7.58999e-06
        [inline_without_move]: 4.93001e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 7.99977e-07
        [auto_monad_eliminator]: 5.94999e-06
        [cse]: 1.124e-05
        [replace_applicator]: 5.35001e-06
[py_interpret_to_execute_after_opt_a]: 8.30999e-06
[rewriter_after_opt_a]: 2.417e-05
[convert_after_rewriter]: 5.64e-06
[order_py_execute_after_rewriter]: 4.53001e-06
[mutable_eliminate]: 0.00046577
[jit_opt_b]: 5.114e-05, [1]
    [Cycle 1]: 4.51e-05, [2]
        [frontend_op_eliminate]: 1.76e-05
        [inline_after_opt_a]: 1.591e-05
[cconv]: 1.645e-05
[loop_unroll]: 0.00040542
[jit_opt_after_cconv]: 0.00013842, [1]
    [Cycle 1]: 0.00013137, [11]
        [c_1]: 2.256e-05
        [parameter_eliminate]: 2.53003e-06
        [updatestate_depend_eliminate]: 5.30999e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 1.609e-05
        [call_graph_tuple_transform]: 1.651e-05
        [tuple_list_get_item_eliminator]: 5.72001e-06
        [none_parameter_eliminate]: 9.00007e-07
        [renormalize]: 9.39996e-07
        [switch_simplify]: 5.71e-06
[remove_dup_value]: 6.89001e-06
[partial_unused_args_eliminate]: 1.66002e-06
[environ_conv]: 4.42e-06
[add_recomputation]: 3.143e-05
[cse_after_recomputation]: 2.009e-05, [1]
    [Cycle 1]: 1.479e-05, [1]
        [cse]: 9.05999e-06
[auto_monad_reorder]: 9.49e-06
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 4.73001e-06
[opt_after_jit_grad]: 0.00044225
[symbol_engine_optimizer]: 7.107e-05, [1]
    [Cycle 1]: 6.479e-05, [6]
        [build]: 2.40002e-06
        [elim_shapecalc]: 7.88001e-06
        [elim_not_effective]: 1.218e-05
        [opt_reshape]: 5.85002e-06
        [fold_const_symbol]: 8.65001e-06
        [renormalize]: 5.19998e-07
[validate]: 2.5e-05
Sums
    bootstrap                            :     0.000262s :  2.71%
    type_inference                       :     0.005914s : 61.34%
    event_method                         :     0.000019s :  0.19%
    auto_monad                           :     0.000060s :  0.63%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000047s :  0.49%
    rewriter_before_opt_a                :     0.000076s :  0.79%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000096s :  1.00%
    jit_opt_a.loop_unroll                :     0.000042s :  0.43%
    jit_opt_a.a_1                        :     0.000747s :  7.74%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000513s :  5.32%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.16%
    jit_opt_a.cse                        :     0.000026s :  0.27%
    jit_opt_a.replace_applicator         :     0.000017s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.09%
    rewriter_after_opt_a                 :     0.000024s :  0.25%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000466s :  4.83%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.17%
    cconv                                :     0.000016s :  0.17%
    loop_unroll                          :     0.000405s :  4.20%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000007s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000031s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000009s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000442s :  4.59%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000025s :  0.26%


Time group info:
------[substitution.]   0.000221    26
 0.66% :     0.000001s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 1.67% :     0.000004s :      3: substitution.graph_param_transform
85.85% :     0.000190s :      6: substitution.inline
 1.16% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.84% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.29% :     0.000003s :      2: substitution.replace_old_param
 3.22% :     0.000007s :      1: substitution.switch_simplify
 3.73% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005865     2
84.28% :     0.004944s :      1: type_inference.infer
15.72% :     0.000922s :      1: type_inference.specialize
------[replace.]   0.000115     9
55.72% :     0.000064s :      6: replace.inline
18.32% :     0.000021s :      1: replace.switch_simplify
25.96% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     9
93.28% :     0.000186s :      6: match.inline
 3.19% :     0.000006s :      1: match.switch_simplify
 3.54% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   801
 1.42% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.03% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.21% :     0.000002s :     12: predicate.addn_check_dump
 1.38% :     0.000002s :     12: predicate.addn_zero_filter
 2.02% :     0.000003s :     12: predicate.arithmetic_simplify
 1.35% :     0.000002s :     12: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.18% :     0.000002s :     12: predicate.compare_switch_simplify
 1.47% :     0.000002s :     12: predicate.depend_value_elim
 1.28% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.70% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.82% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     12: predicate.environ_get_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.80% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.46% :     0.000007s :     26: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     17: predicate.load_eliminater
 1.25% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.49% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.88% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.22% :     0.000002s :     12: predicate.merge_addn
 1.23% :     0.000002s :     12: predicate.minmaximum_grad
 1.46% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     17: predicate.partial_eliminate
 1.22% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.77% :     0.000002s :     12: predicate.reduce_eliminate
 1.45% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     20: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     12: predicate.reshape_eliminate
 1.30% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.60% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     12: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.58% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.84% :     0.000004s :     20: predicate.switch_defer_inline
 2.41% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.26% :     0.000011s :     65: predicate.switch_simplify
 1.28% :     0.000002s :     12: predicate.tile_eliminate
 1.28% :     0.000002s :     12: predicate.transpose_eliminate
 1.56% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.37% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.19% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000002s :     12: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000719    15
48.80% :     0.000351s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.57% :     0.000076s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.64% :     0.000292s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014698    72
 0.23% :     0.000034s :      1: add_recomputation
 0.43% :     0.000064s :      1: auto_monad
 0.08% :     0.000012s :      1: auto_monad_reorder
 1.86% :     0.000273s :      1: bootstrap
 0.13% :     0.000019s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000022s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
32.92% :     0.004839s :      1: jit_opt_a
 0.96% :     0.000141s :      1: jit_opt_after_cconv
 0.37% :     0.000054s :      1: jit_opt_b
 2.82% :     0.000414s :      1: loop_unroll
 3.22% :     0.000474s :      1: mutable_eliminate
 6.80% :     0.001000s :     26: opt.transform.jit_opt_a
 0.32% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.06% :     0.000450s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.34% :     0.000050s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000009s :      1: remove_dup_value
 1.67% :     0.000245s :      1: renormalize.infer
 1.78% :     0.000262s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000027s :      1: rewriter_after_opt_a
 0.54% :     0.000079s :      1: rewriter_before_opt_a
 0.50% :     0.000074s :      1: symbol_engine_optimizer
40.30% :     0.005923s :      1: type_inference


TotalTime = 0.0174306, [33]
[bootstrap]: 0.00027324
[type_inference]: 0.00427951
[event_method]: 1.623e-05
[auto_monad]: 6.081e-05
[graph_reusing]: 5.49e-06
[pre_auto_parallel]: 2.52001e-06
[py_interpret_to_execute]: 3.135e-05
[rewriter_before_opt_a]: 7.056e-05
[expand_dump_flag]: 3.15998e-06
[jit_opt_a]: 0.00484231, [2]
    [Cycle 1]: 0.00140284, [27]
        [switch_simplify]: 4.934e-05
        [loop_unroll]: 2.804e-05
        [a_1]: 0.00049494
        [with_stream_mark]: 1.729e-05
        [recompute_prepare]: 7.20998e-06
        [updatestate_depend_eliminate]: 3.46001e-06
        [updatestate_assign_eliminate]: 2.84001e-06
        [updatestate_loads_eliminate]: 2.74001e-06
        [parameter_eliminate]: 1.87001e-06
        [specialize_transform]: 5.67999e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 4.94e-06
        [meta_shard_fg_expand]: 2.02001e-06
        [get_grad_eliminate_]: 4.87998e-06
        [merge_forward]: 3.41001e-06
        [cell_reuse_recompute_pass]: 1.10999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.385e-05
        [j_node_and_user_rematch]: 8.43999e-06
        [meta_fg_expand]: 2.16e-06
        [replace_old_param]: 8.21002e-06
        [inline_without_move]: 5.14e-06
        [renormalize]: 0.00051623
        [add_forward_monad_depend]: 4.90999e-06
        [auto_monad_grad]: 2.24999e-06
        [auto_monad_eliminator]: 1.269e-05
        [cse]: 2.519e-05
        [replace_applicator]: 1.131e-05
    [Cycle 2]: 0.0003104, [27]
        [switch_simplify]: 5.24e-06
        [loop_unroll]: 4.65001e-06
        [a_1]: 8.687e-05
        [with_stream_mark]: 8.47e-06
        [recompute_prepare]: 5.08002e-06
        [updatestate_depend_eliminate]: 2.39001e-06
        [updatestate_assign_eliminate]: 2.23998e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.72e-06
        [updatestate_useless_node_eliminater]: 4.73001e-06
        [accelerated_algorithm]: 5.14e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.66002e-06
        [merge_forward]: 2.39001e-06
        [cell_reuse_recompute_pass]: 1.53002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.169e-05
        [j_node_and_user_rematch]: 7.55998e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 1.168e-05
        [inline_without_move]: 4.60001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 5.23002e-06
        [cse]: 9.91998e-06
        [replace_applicator]: 4.74e-06
[py_interpret_to_execute_after_opt_a]: 8.80001e-06
[rewriter_after_opt_a]: 2.988e-05
[convert_after_rewriter]: 6.98e-06
[order_py_execute_after_rewriter]: 5.00999e-06
[mutable_eliminate]: 0.00049027
[jit_opt_b]: 4.664e-05, [1]
    [Cycle 1]: 4.036e-05, [2]
        [frontend_op_eliminate]: 1.571e-05
        [inline_after_opt_a]: 1.399e-05
[cconv]: 2.367e-05
[loop_unroll]: 0.00041312
[jit_opt_after_cconv]: 0.0002298, [1]
    [Cycle 1]: 0.00022352, [11]
        [c_1]: 2.04e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 4.97e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [cse]: 1.587e-05
        [call_graph_tuple_transform]: 0.00011253
        [tuple_list_get_item_eliminator]: 5.37999e-06
        [none_parameter_eliminate]: 1.87999e-06
        [renormalize]: 3.50003e-07
        [switch_simplify]: 5.27999e-06
[remove_dup_value]: 1.358e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 4.97999e-06
[add_recomputation]: 4.783e-05
[cse_after_recomputation]: 2.088e-05, [1]
    [Cycle 1]: 1.526e-05, [1]
        [cse]: 9.89999e-06
[auto_monad_reorder]: 1.45e-05
[get_jit_bprop_graph]: 1.67001e-06
[rewriter_after_jit_bprop_graph]: 4.32e-06
[opt_after_jit_grad]: 0.00046255
[symbol_engine_optimizer]: 7.066e-05, [1]
    [Cycle 1]: 6.428e-05, [6]
        [build]: 2.96001e-06
        [elim_shapecalc]: 7.5e-06
        [elim_not_effective]: 1.266e-05
        [opt_reshape]: 5.47001e-06
        [fold_const_symbol]: 8.37e-06
        [renormalize]: 4.09986e-07
[validate]: 3.071e-05
[backend_pass]: 1.05999e-06
[task_emit]: 0.00573037
[execute]: 8.21002e-06
Sums
    bootstrap                            :     0.000273s :  1.99%
    type_inference                       :     0.004280s : 31.15%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000061s :  0.44%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.23%
    rewriter_before_opt_a                :     0.000071s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.24%
    jit_opt_a.a_1                        :     0.000582s :  4.23%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000020s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000516s :  3.76%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000035s :  0.26%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000030s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000490s :  3.57%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000413s :  3.01%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000113s :  0.82%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000048s :  0.35%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000463s :  3.37%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005730s : 41.71%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000281    23
 0.69% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
34.99% :     0.000098s :      3: substitution.graph_param_transform
54.30% :     0.000153s :      4: substitution.inline
 0.97% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.46% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.87% :     0.000008s :      2: substitution.replace_old_param
 4.24% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004222     2
81.79% :     0.003453s :      1: type_inference.infer
18.21% :     0.000769s :      1: type_inference.specialize
------[replace.]   0.000061     6
67.64% :     0.000041s :      4: replace.inline
32.36% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000161     6
93.21% :     0.000150s :      4: match.inline
 6.79% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.26% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.46% :     0.000002s :     10: predicate.addn_zero_filter
 2.11% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000001s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.63% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.10% :     0.000001s :      6: predicate.get_grad_eliminate
 0.36% :     0.000000s :      3: predicate.graph_param_transform
 5.40% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.73% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.34% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.02% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.54% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.59% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000001s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.31% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.56% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.20% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.69% :     0.000008s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.39% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.20% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000523    12
49.78% :     0.000260s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.81% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.41% :     0.000232s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018924    76
 0.27% :     0.000051s :      1: add_recomputation
 0.34% :     0.000064s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.51% :     0.000286s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.60% :     0.004845s :      1: jit_opt_a
 1.23% :     0.000232s :      1: jit_opt_after_cconv
 0.26% :     0.000049s :      1: jit_opt_b
 2.22% :     0.000421s :      1: loop_unroll
 2.63% :     0.000498s :      1: mutable_eliminate
 4.13% :     0.000781s :     26: opt.transform.jit_opt_a
 0.74% :     0.000140s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.49% :     0.000470s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.38% :     0.000262s :      1: renormalize.infer
 1.30% :     0.000247s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000033s :      1: rewriter_after_opt_a
 0.39% :     0.000074s :      1: rewriter_before_opt_a
 0.39% :     0.000073s :      1: symbol_engine_optimizer
30.34% :     0.005742s :      1: task_emit
22.68% :     0.004292s :      1: type_inference
 0.26% :     0.000049s :      1: validate


TotalTime = 0.0225724, [33]
[bootstrap]: 0.00038849
[type_inference]: 0.00645498
[event_method]: 1.962e-05
[auto_monad]: 7.123e-05
[graph_reusing]: 7.10002e-06
[pre_auto_parallel]: 3.33998e-06
[py_interpret_to_execute]: 4.31e-05
[rewriter_before_opt_a]: 0.00010364
[expand_dump_flag]: 3.63e-06
[jit_opt_a]: 0.00690287, [2]
    [Cycle 1]: 0.00199801, [27]
        [switch_simplify]: 7.754e-05
        [loop_unroll]: 2.958e-05
        [a_1]: 0.0006188
        [with_stream_mark]: 2.627e-05
        [recompute_prepare]: 7.55e-06
        [updatestate_depend_eliminate]: 4.68999e-06
        [updatestate_assign_eliminate]: 3.17002e-06
        [updatestate_loads_eliminate]: 2.85002e-06
        [parameter_eliminate]: 1.94999e-06
        [specialize_transform]: 6.16e-06
        [updatestate_useless_node_eliminater]: 5.13002e-06
        [accelerated_algorithm]: 5.44998e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 5.37999e-06
        [merge_forward]: 3.85e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.515e-05
        [j_node_and_user_rematch]: 9.46003e-06
        [meta_fg_expand]: 2.44999e-06
        [replace_old_param]: 9.29e-06
        [inline_without_move]: 5.39998e-06
        [renormalize]: 0.0009114
        [add_forward_monad_depend]: 5.49e-06
        [auto_monad_grad]: 2.52001e-06
        [auto_monad_eliminator]: 1.409e-05
        [cse]: 3.04e-05
        [replace_applicator]: 1.231e-05
    [Cycle 2]: 0.00031737, [27]
        [switch_simplify]: 5.54998e-06
        [loop_unroll]: 4.73001e-06
        [a_1]: 8.964e-05
        [with_stream_mark]: 9.84999e-06
        [recompute_prepare]: 4.85001e-06
        [updatestate_depend_eliminate]: 2.64999e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 4.82e-06
        [updatestate_useless_node_eliminater]: 4.79998e-06
        [accelerated_algorithm]: 4.87998e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 4.55001e-06
        [merge_forward]: 2.49001e-06
        [cell_reuse_recompute_pass]: 1.85001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.208e-05
        [j_node_and_user_rematch]: 7.98001e-06
        [meta_fg_expand]: 1.55001e-06
        [replace_old_param]: 6.98e-06
        [inline_without_move]: 4.61002e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 1.00999e-06
        [auto_monad_eliminator]: 5.02e-06
        [cse]: 1.006e-05
        [replace_applicator]: 5.15999e-06
[py_interpret_to_execute_after_opt_a]: 1.135e-05
[rewriter_after_opt_a]: 3.838e-05
[convert_after_rewriter]: 7.21999e-06
[order_py_execute_after_rewriter]: 4.92999e-06
[mutable_eliminate]: 0.00061394
[jit_opt_b]: 4.857e-05, [1]
    [Cycle 1]: 4.162e-05, [2]
        [frontend_op_eliminate]: 1.599e-05
        [inline_after_opt_a]: 1.443e-05
[cconv]: 2.713e-05
[loop_unroll]: 0.00048605
[jit_opt_after_cconv]: 0.00013682, [1]
    [Cycle 1]: 0.00012991, [11]
        [c_1]: 2.047e-05
        [parameter_eliminate]: 2.43998e-06
        [updatestate_depend_eliminate]: 5.40001e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [cse]: 1.785e-05
        [call_graph_tuple_transform]: 1.9e-05
        [tuple_list_get_item_eliminator]: 5.28002e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.13002e-06
[remove_dup_value]: 1.536e-05
[partial_unused_args_eliminate]: 2.39999e-06
[environ_conv]: 5.92999e-06
[add_recomputation]: 4.885e-05
[cse_after_recomputation]: 1.985e-05, [1]
    [Cycle 1]: 1.404e-05, [1]
        [cse]: 8.60001e-06
[auto_monad_reorder]: 1.644e-05
[get_jit_bprop_graph]: 2.19999e-06
[rewriter_after_jit_bprop_graph]: 4.52998e-06
[opt_after_jit_grad]: 0.00044954
[symbol_engine_optimizer]: 7.206e-05, [1]
    [Cycle 1]: 6.579e-05, [6]
        [build]: 3.08998e-06
        [elim_shapecalc]: 8.37e-06
        [elim_not_effective]: 1.282e-05
        [opt_reshape]: 5.47999e-06
        [fold_const_symbol]: 8.62e-06
        [renormalize]: 8.00006e-07
[validate]: 3.701e-05
[backend_pass]: 9.60019e-07
[task_emit]: 0.00629299
[execute]: 8.37e-06
Sums
    bootstrap                            :     0.000388s :  2.24%
    type_inference                       :     0.006455s : 37.20%
    event_method                         :     0.000020s :  0.11%
    auto_monad                           :     0.000071s :  0.41%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.25%
    rewriter_before_opt_a                :     0.000104s :  0.60%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000083s :  0.48%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000708s :  4.08%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000911s :  5.25%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.11%
    jit_opt_a.cse                        :     0.000040s :  0.23%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000614s :  3.54%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.08%
    cconv                                :     0.000027s :  0.16%
    loop_unroll                          :     0.000486s :  2.80%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000049s :  0.28%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000450s :  2.59%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006293s : 36.27%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000271    23
 0.90% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000002s :      2: substitution.fold_const_symbol
 2.14% :     0.000006s :      3: substitution.graph_param_transform
86.66% :     0.000235s :      4: substitution.inline
 1.28% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.87% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.61% :     0.000004s :      2: substitution.replace_old_param
 4.95% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006366     2
83.60% :     0.005322s :      1: type_inference.infer
16.40% :     0.001044s :      1: type_inference.specialize
------[replace.]   0.000075     6
68.54% :     0.000051s :      4: replace.inline
31.46% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000245     6
94.96% :     0.000232s :      4: match.inline
 5.04% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 1.77% :     0.000002s :     10: predicate.arithmetic_simplify
 1.32% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.18% :     0.000001s :     10: predicate.depend_value_elim
 1.38% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.62% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.77% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.48% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.70% :     0.000002s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.13% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.54% :     0.000007s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.65% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.16% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.91% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     15: predicate.partial_eliminate
 1.37% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.36% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.49% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.05% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.51% :     0.000003s :     16: predicate.switch_defer_inline
 1.97% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.54% :     0.000010s :     53: predicate.switch_simplify
 1.50% :     0.000002s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.08% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.66% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.65% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000766    12
49.17% :     0.000377s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.03% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.80% :     0.000336s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024510    76
 0.21% :     0.000052s :      1: add_recomputation
 0.30% :     0.000075s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.67% :     0.000408s :      1: bootstrap
 0.12% :     0.000030s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
28.18% :     0.006907s :      1: jit_opt_a
 0.57% :     0.000140s :      1: jit_opt_after_cconv
 0.21% :     0.000051s :      1: jit_opt_b
 2.02% :     0.000494s :      1: loop_unroll
 2.54% :     0.000622s :      1: mutable_eliminate
 3.83% :     0.000938s :     26: opt.transform.jit_opt_a
 0.19% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.87% :     0.000457s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000046s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.10% :     0.000516s :      1: renormalize.infer
 1.58% :     0.000386s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000041s :      1: rewriter_after_opt_a
 0.44% :     0.000108s :      1: rewriter_before_opt_a
 0.30% :     0.000075s :      1: symbol_engine_optimizer
25.75% :     0.006311s :      1: task_emit
26.41% :     0.006473s :      1: type_inference
 0.23% :     0.000057s :      1: validate


TotalTime = 0.0170821, [33]
[bootstrap]: 0.00026213
[type_inference]: 0.00426536
[event_method]: 1.59e-05
[auto_monad]: 6.147e-05
[graph_reusing]: 5.51e-06
[pre_auto_parallel]: 2.28002e-06
[py_interpret_to_execute]: 3.44e-05
[rewriter_before_opt_a]: 7.143e-05
[expand_dump_flag]: 3.31001e-06
[jit_opt_a]: 0.00473083, [2]
    [Cycle 1]: 0.00138153, [27]
        [switch_simplify]: 4.948e-05
        [loop_unroll]: 2.949e-05
        [a_1]: 0.00049343
        [with_stream_mark]: 1.483e-05
        [recompute_prepare]: 7.07002e-06
        [updatestate_depend_eliminate]: 3.45998e-06
        [updatestate_assign_eliminate]: 2.67001e-06
        [updatestate_loads_eliminate]: 2.59001e-06
        [parameter_eliminate]: 1.81998e-06
        [specialize_transform]: 5.69e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 1.90001e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 3.45003e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.409e-05
        [j_node_and_user_rematch]: 8.75001e-06
        [meta_fg_expand]: 2.19999e-06
        [replace_old_param]: 8.37e-06
        [inline_without_move]: 5.09e-06
        [renormalize]: 0.00049981
        [add_forward_monad_depend]: 5.17999e-06
        [auto_monad_grad]: 2.16e-06
        [auto_monad_eliminator]: 1.361e-05
        [cse]: 2.405e-05
        [replace_applicator]: 1.098e-05
    [Cycle 2]: 0.00030846, [27]
        [switch_simplify]: 5.88998e-06
        [loop_unroll]: 4.64002e-06
        [a_1]: 8.958e-05
        [with_stream_mark]: 8.45999e-06
        [recompute_prepare]: 4.87e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 1.04003e-06
        [specialize_transform]: 5.04998e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 4.72998e-06
        [meta_shard_fg_expand]: 1.30001e-06
        [get_grad_eliminate_]: 4.55999e-06
        [merge_forward]: 2.34999e-06
        [cell_reuse_recompute_pass]: 1.55001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.155e-05
        [j_node_and_user_rematch]: 7.93999e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 7.08998e-06
        [inline_without_move]: 4.87998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 5.90022e-07
        [auto_monad_eliminator]: 5.06002e-06
        [cse]: 9.84999e-06
        [replace_applicator]: 4.92e-06
[py_interpret_to_execute_after_opt_a]: 8.47e-06
[rewriter_after_opt_a]: 3.076e-05
[convert_after_rewriter]: 7.3e-06
[order_py_execute_after_rewriter]: 4.69002e-06
[mutable_eliminate]: 0.00047462
[jit_opt_b]: 4.72e-05, [1]
    [Cycle 1]: 4.115e-05, [2]
        [frontend_op_eliminate]: 1.567e-05
        [inline_after_opt_a]: 1.436e-05
[cconv]: 2.257e-05
[loop_unroll]: 0.00041916
[jit_opt_after_cconv]: 0.00013222, [1]
    [Cycle 1]: 0.0001259, [11]
        [c_1]: 2.078e-05
        [parameter_eliminate]: 2.06e-06
        [updatestate_depend_eliminate]: 4.57e-06
        [updatestate_assign_eliminate]: 2.15002e-06
        [updatestate_loads_eliminate]: 2.06003e-06
        [cse]: 1.582e-05
        [call_graph_tuple_transform]: 1.732e-05
        [tuple_list_get_item_eliminator]: 5.53002e-06
        [none_parameter_eliminate]: 1.40999e-06
        [renormalize]: 5.40022e-07
        [switch_simplify]: 5.19e-06
[remove_dup_value]: 1.302e-05
[partial_unused_args_eliminate]: 2.49001e-06
[environ_conv]: 4.42e-06
[add_recomputation]: 4.52e-05
[cse_after_recomputation]: 2.034e-05, [1]
    [Cycle 1]: 1.502e-05, [1]
        [cse]: 9.00001e-06
[auto_monad_reorder]: 1.498e-05
[get_jit_bprop_graph]: 1.79998e-06
[rewriter_after_jit_bprop_graph]: 4.2e-06
[opt_after_jit_grad]: 0.00044832
[symbol_engine_optimizer]: 6.912e-05, [1]
    [Cycle 1]: 6.363e-05, [6]
        [build]: 3.38e-06
        [elim_shapecalc]: 6.98998e-06
        [elim_not_effective]: 1.24e-05
        [opt_reshape]: 5.59e-06
        [fold_const_symbol]: 8.17e-06
        [renormalize]: 3.29979e-07
[validate]: 2.783e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00565556
[execute]: 6.74001e-06
Sums
    bootstrap                            :     0.000262s :  1.94%
    type_inference                       :     0.004265s : 31.61%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000061s :  0.46%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.25%
    rewriter_before_opt_a                :     0.000071s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000583s :  4.32%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000500s :  3.70%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.14%
    jit_opt_a.cse                        :     0.000034s :  0.25%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000475s :  3.52%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000419s :  3.11%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000045s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000448s :  3.32%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005656s : 41.91%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000174    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000001s :      2: substitution.fold_const_symbol
 2.94% :     0.000005s :      3: substitution.graph_param_transform
82.98% :     0.000144s :      4: substitution.inline
 1.81% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.44% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000003s :      2: substitution.replace_old_param
 6.17% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004211     2
81.18% :     0.003419s :      1: type_inference.infer
18.82% :     0.000792s :      1: type_inference.specialize
------[replace.]   0.000062     6
65.51% :     0.000040s :      4: replace.inline
34.49% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000151     6
93.65% :     0.000142s :      4: match.inline
 6.35% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.27% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.27% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000001s :     10: predicate.addn_zero_filter
 1.84% :     0.000002s :     10: predicate.arithmetic_simplify
 1.37% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.42% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.71% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000001s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.95% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.53% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.69% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.34% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.20% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.51% :     0.000003s :     16: predicate.switch_defer_inline
 2.27% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.58% :     0.000009s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.11% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.82% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000533    12
47.33% :     0.000252s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.78% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.90% :     0.000255s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018466    76
 0.26% :     0.000048s :      1: add_recomputation
 0.35% :     0.000065s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.48% :     0.000273s :      1: bootstrap
 0.14% :     0.000025s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.63% :     0.004734s :      1: jit_opt_a
 0.73% :     0.000135s :      1: jit_opt_after_cconv
 0.27% :     0.000050s :      1: jit_opt_b
 2.31% :     0.000426s :      1: loop_unroll
 2.61% :     0.000483s :      1: mutable_eliminate
 4.23% :     0.000781s :     26: opt.transform.jit_opt_a
 0.25% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.47% :     0.000456s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.35% :     0.000250s :      1: renormalize.infer
 1.32% :     0.000243s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.40% :     0.000074s :      1: rewriter_before_opt_a
 0.39% :     0.000071s :      1: symbol_engine_optimizer
30.67% :     0.005664s :      1: task_emit
23.16% :     0.004276s :      1: type_inference
 0.24% :     0.000044s :      1: validate


TotalTime = 0.012712, [30]
[bootstrap]: 0.0002324
[type_inference]: 0.00557402
[event_method]: 1.909e-05
[auto_monad]: 5.996e-05
[graph_reusing]: 4.2e-06
[pre_auto_parallel]: 1.91e-06
[py_interpret_to_execute]: 4.475e-05
[rewriter_before_opt_a]: 7.591e-05
[expand_dump_flag]: 2.14e-06
[jit_opt_a]: 0.00478492, [2]
    [Cycle 1]: 0.00156206, [27]
        [switch_simplify]: 0.00010033
        [loop_unroll]: 3.558e-05
        [a_1]: 0.0006246
        [with_stream_mark]: 1.169e-05
        [recompute_prepare]: 6.81001e-06
        [updatestate_depend_eliminate]: 2.84001e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 5.86e-06
        [updatestate_useless_node_eliminater]: 5.39998e-06
        [accelerated_algorithm]: 5.29e-06
        [meta_shard_fg_expand]: 1.60999e-06
        [get_grad_eliminate_]: 5.30001e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 9.29984e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.065e-05
        [j_node_and_user_rematch]: 8.28999e-06
        [meta_fg_expand]: 1.82001e-06
        [replace_old_param]: 7.52998e-06
        [inline_without_move]: 5.76e-06
        [renormalize]: 0.00051222
        [add_forward_monad_depend]: 4.50001e-06
        [auto_monad_grad]: 1.15001e-06
        [auto_monad_eliminator]: 1.026e-05
        [cse]: 1.553e-05
        [replace_applicator]: 1.132e-05
    [Cycle 2]: 0.00034182, [27]
        [switch_simplify]: 6.04001e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 9.084e-05
        [with_stream_mark]: 8.43999e-06
        [recompute_prepare]: 5.05999e-06
        [updatestate_depend_eliminate]: 2.79999e-06
        [updatestate_assign_eliminate]: 2.06998e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.02999e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.29998e-06
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.60999e-06
        [merge_forward]: 2.39001e-06
        [cell_reuse_recompute_pass]: 1.39998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.135e-05
        [j_node_and_user_rematch]: 7.99997e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 2.021e-05
        [inline_without_move]: 4.74e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.52999e-06
        [cse]: 1.136e-05
        [replace_applicator]: 5.12e-06
[py_interpret_to_execute_after_opt_a]: 8.35001e-06
[rewriter_after_opt_a]: 2.452e-05
[convert_after_rewriter]: 5.47001e-06
[order_py_execute_after_rewriter]: 4.77e-06
[mutable_eliminate]: 0.00047735
[jit_opt_b]: 4.998e-05, [1]
    [Cycle 1]: 4.332e-05, [2]
        [frontend_op_eliminate]: 1.712e-05
        [inline_after_opt_a]: 1.452e-05
[cconv]: 1.61e-05
[loop_unroll]: 0.00041409
[jit_opt_after_cconv]: 0.00013357, [1]
    [Cycle 1]: 0.00012756, [11]
        [c_1]: 2.077e-05
        [parameter_eliminate]: 2.27001e-06
        [updatestate_depend_eliminate]: 5.19e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [cse]: 1.596e-05
        [call_graph_tuple_transform]: 1.68e-05
        [tuple_list_get_item_eliminator]: 5.59e-06
        [none_parameter_eliminate]: 9.79984e-07
        [renormalize]: 3.99974e-07
        [switch_simplify]: 5.52999e-06
[remove_dup_value]: 6.46e-06
[partial_unused_args_eliminate]: 1.49998e-06
[environ_conv]: 1.52e-05
[add_recomputation]: 3.178e-05
[cse_after_recomputation]: 1.992e-05, [1]
    [Cycle 1]: 1.476e-05, [1]
        [cse]: 9.26998e-06
[auto_monad_reorder]: 9.25999e-06
[get_jit_bprop_graph]: 1.57001e-06
[rewriter_after_jit_bprop_graph]: 4.60999e-06
[opt_after_jit_grad]: 0.00044537
[symbol_engine_optimizer]: 7.029e-05, [1]
    [Cycle 1]: 6.413e-05, [6]
        [build]: 2.99001e-06
        [elim_shapecalc]: 7.36001e-06
        [elim_not_effective]: 1.25e-05
        [opt_reshape]: 5.47001e-06
        [fold_const_symbol]: 7.97998e-06
        [renormalize]: 5.8001e-07
[validate]: 2.4e-05
Sums
    bootstrap                            :     0.000232s :  2.50%
    type_inference                       :     0.005574s : 59.99%
    event_method                         :     0.000019s :  0.21%
    auto_monad                           :     0.000060s :  0.65%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000045s :  0.48%
    rewriter_before_opt_a                :     0.000076s :  0.82%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000106s :  1.14%
    jit_opt_a.loop_unroll                :     0.000041s :  0.44%
    jit_opt_a.a_1                        :     0.000715s :  7.70%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.04%
    jit_opt_a.replace_old_param          :     0.000028s :  0.30%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000512s :  5.51%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.17%
    jit_opt_a.cse                        :     0.000027s :  0.29%
    jit_opt_a.replace_applicator         :     0.000016s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.09%
    rewriter_after_opt_a                 :     0.000025s :  0.26%
    convert_after_rewriter               :     0.000005s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000477s :  5.14%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000016s :  0.17%
    loop_unroll                          :     0.000414s :  4.46%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000006s :  0.07%
    partial_unused_args_eliminate        :     0.000001s :  0.02%
    environ_conv                         :     0.000015s :  0.16%
    add_recomputation                    :     0.000032s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000009s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000445s :  4.79%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000024s :  0.26%


Time group info:
------[substitution.]   0.000220    26
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 1.67% :     0.000004s :      3: substitution.graph_param_transform
82.45% :     0.000181s :      6: substitution.inline
 1.19% :     0.000003s :      4: substitution.j_node_and_user_rematch
 5.49% :     0.000012s :      4: substitution.remove_not_recompute_node
 1.37% :     0.000003s :      2: substitution.replace_old_param
 3.06% :     0.000007s :      1: substitution.switch_simplify
 3.64% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005527     2
83.42% :     0.004611s :      1: type_inference.infer
16.58% :     0.000916s :      1: type_inference.specialize
------[replace.]   0.000113     9
55.37% :     0.000062s :      6: replace.inline
18.41% :     0.000021s :      1: replace.switch_simplify
26.22% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000191     9
93.34% :     0.000178s :      6: match.inline
 3.09% :     0.000006s :      1: match.switch_simplify
 3.57% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   801
 1.34% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.37% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.23% :     0.000002s :     12: predicate.addn_check_dump
 1.42% :     0.000002s :     12: predicate.addn_zero_filter
 2.13% :     0.000003s :     12: predicate.arithmetic_simplify
 1.40% :     0.000002s :     12: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.19% :     0.000002s :     12: predicate.compare_switch_simplify
 1.63% :     0.000002s :     12: predicate.depend_value_elim
 1.26% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.24% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     12: predicate.environ_get_eliminate
 1.15% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 4.92% :     0.000006s :     26: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     17: predicate.load_eliminater
 1.32% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.35% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.08% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     12: predicate.merge_addn
 1.18% :     0.000002s :     12: predicate.minmaximum_grad
 1.45% :     0.000002s :      3: predicate.mutable_eliminate
 0.37% :     0.000000s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     17: predicate.partial_eliminate
 1.23% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     12: predicate.reduce_eliminate
 1.46% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     20: predicate.replace_applicator
 0.44% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     12: predicate.reshape_eliminate
 1.28% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     12: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.86% :     0.000004s :     20: predicate.switch_defer_inline
 2.28% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.30% :     0.000011s :     65: predicate.switch_simplify
 1.34% :     0.000002s :     12: predicate.tile_eliminate
 1.34% :     0.000002s :     12: predicate.transpose_eliminate
 1.70% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.17% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     12: predicate.value_based_eliminate
 0.27% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000673    15
48.74% :     0.000328s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.72% :     0.000072s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.53% :     0.000273s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014303    72
 0.24% :     0.000035s :      1: add_recomputation
 0.44% :     0.000063s :      1: auto_monad
 0.08% :     0.000012s :      1: auto_monad_reorder
 1.69% :     0.000242s :      1: bootstrap
 0.13% :     0.000019s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.16% :     0.000022s :      1: cse_after_recomputation
 0.13% :     0.000018s :      1: environ_conv
 0.16% :     0.000022s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
33.47% :     0.004788s :      1: jit_opt_a
 0.95% :     0.000136s :      1: jit_opt_after_cconv
 0.37% :     0.000053s :      1: jit_opt_b
 2.95% :     0.000422s :      1: loop_unroll
 3.39% :     0.000485s :      1: mutable_eliminate
 6.95% :     0.000993s :     26: opt.transform.jit_opt_a
 0.32% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.17% :     0.000024s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.17% :     0.000453s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.33% :     0.000048s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000009s :      1: remove_dup_value
 1.76% :     0.000252s :      1: renormalize.infer
 1.78% :     0.000254s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000030s :      1: rewriter_after_opt_a
 0.55% :     0.000079s :      1: rewriter_before_opt_a
 0.51% :     0.000073s :      1: symbol_engine_optimizer
39.03% :     0.005583s :      1: type_inference


TotalTime = 0.0163573, [33]
[bootstrap]: 0.00025299
[type_inference]: 0.00399855
[event_method]: 1.634e-05
[auto_monad]: 5.668e-05
[graph_reusing]: 4.70001e-06
[pre_auto_parallel]: 2.09e-06
[py_interpret_to_execute]: 3.149e-05
[rewriter_before_opt_a]: 6.531e-05
[expand_dump_flag]: 2.59001e-06
[jit_opt_a]: 0.00463928, [2]
    [Cycle 1]: 0.00135941, [27]
        [switch_simplify]: 4.635e-05
        [loop_unroll]: 3.013e-05
        [a_1]: 0.00049641
        [with_stream_mark]: 1.336e-05
        [recompute_prepare]: 6.83e-06
        [updatestate_depend_eliminate]: 3.08e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 2.25002e-06
        [parameter_eliminate]: 1.26002e-06
        [specialize_transform]: 5.66e-06
        [updatestate_useless_node_eliminater]: 5.29998e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.65001e-06
        [get_grad_eliminate_]: 5.05001e-06
        [merge_forward]: 3.17002e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.246e-05
        [j_node_and_user_rematch]: 8.41002e-06
        [meta_fg_expand]: 2.17001e-06
        [replace_old_param]: 7.95e-06
        [inline_without_move]: 5.42999e-06
        [renormalize]: 0.00049064
        [add_forward_monad_depend]: 4.42e-06
        [auto_monad_grad]: 2.09e-06
        [auto_monad_eliminator]: 1.058e-05
        [cse]: 1.927e-05
        [replace_applicator]: 1.067e-05
    [Cycle 2]: 0.00030864, [27]
        [switch_simplify]: 5.54998e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 9.057e-05
        [with_stream_mark]: 8.65001e-06
        [recompute_prepare]: 4.63001e-06
        [updatestate_depend_eliminate]: 2.47001e-06
        [updatestate_assign_eliminate]: 1.98002e-06
        [updatestate_loads_eliminate]: 1.83002e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 5.12e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 4.65001e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.68999e-06
        [merge_forward]: 2.32999e-06
        [cell_reuse_recompute_pass]: 1.45001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.196e-05
        [j_node_and_user_rematch]: 7.48999e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 6.73e-06
        [inline_without_move]: 4.64998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 4.70999e-06
        [cse]: 9.51003e-06
        [replace_applicator]: 4.84e-06
[py_interpret_to_execute_after_opt_a]: 7.56001e-06
[rewriter_after_opt_a]: 2.836e-05
[convert_after_rewriter]: 6.10002e-06
[order_py_execute_after_rewriter]: 4.76997e-06
[mutable_eliminate]: 0.00046436
[jit_opt_b]: 4.696e-05, [1]
    [Cycle 1]: 4.04e-05, [2]
        [frontend_op_eliminate]: 1.516e-05
        [inline_after_opt_a]: 1.389e-05
[cconv]: 1.603e-05
[loop_unroll]: 0.00042692
[jit_opt_after_cconv]: 0.00013101, [1]
    [Cycle 1]: 0.00012466, [11]
        [c_1]: 2.028e-05
        [parameter_eliminate]: 2.37001e-06
        [updatestate_depend_eliminate]: 5.30999e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 1.574e-05
        [call_graph_tuple_transform]: 1.574e-05
        [tuple_list_get_item_eliminator]: 5.40999e-06
        [none_parameter_eliminate]: 1.14e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 5.29e-06
[remove_dup_value]: 1.084e-05
[partial_unused_args_eliminate]: 1.52999e-06
[environ_conv]: 4.59002e-06
[add_recomputation]: 3.774e-05
[cse_after_recomputation]: 1.926e-05, [1]
    [Cycle 1]: 1.435e-05, [1]
        [cse]: 8.77999e-06
[auto_monad_reorder]: 1.047e-05
[get_jit_bprop_graph]: 1.55999e-06
[rewriter_after_jit_bprop_graph]: 4.70999e-06
[opt_after_jit_grad]: 0.00045224
[symbol_engine_optimizer]: 6.742e-05, [1]
    [Cycle 1]: 6.15e-05, [6]
        [build]: 2.86999e-06
        [elim_shapecalc]: 6.92002e-06
        [elim_not_effective]: 1.064e-05
        [opt_reshape]: 5.49998e-06
        [fold_const_symbol]: 8.34002e-06
        [renormalize]: 4.19997e-07
[validate]: 2.635e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00535065
[execute]: 6.50002e-06
Sums
    bootstrap                            :     0.000253s :  1.97%
    type_inference                       :     0.003999s : 31.13%
    event_method                         :     0.000016s :  0.13%
    auto_monad                           :     0.000057s :  0.44%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.25%
    rewriter_before_opt_a                :     0.000065s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000052s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.27%
    jit_opt_a.a_1                        :     0.000587s :  4.57%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000491s :  3.82%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.12%
    jit_opt_a.cse                        :     0.000029s :  0.22%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000028s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000464s :  3.62%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000016s :  0.12%
    loop_unroll                          :     0.000427s :  3.32%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000011s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000038s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000010s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000452s :  3.52%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000026s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005351s : 41.66%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000177    23
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.77% :     0.000001s :      2: substitution.fold_const_symbol
 2.21% :     0.000004s :      3: substitution.graph_param_transform
85.32% :     0.000151s :      4: substitution.inline
 1.55% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.21% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.77% :     0.000003s :      2: substitution.replace_old_param
 5.30% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003949     2
80.81% :     0.003191s :      1: type_inference.infer
19.19% :     0.000758s :      1: type_inference.specialize
------[replace.]   0.000061     6
66.82% :     0.000041s :      4: replace.inline
33.18% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000157     6
94.68% :     0.000149s :      4: match.inline
 5.32% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.29% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000001s :     10: predicate.addn_zero_filter
 2.04% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.15% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.35% :     0.000002s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.02% :     0.000006s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.89% :     0.000002s :     15: predicate.load_eliminater
 1.78% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.51% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 1.56% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.68% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.66% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.75% :     0.000003s :     16: predicate.switch_defer_inline
 2.23% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.67% :     0.000009s :     53: predicate.switch_simplify
 1.30% :     0.000001s :     10: predicate.tile_eliminate
 1.37% :     0.000002s :     10: predicate.transpose_eliminate
 1.51% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.10% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.32% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.77% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000487    12
46.77% :     0.000228s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.42% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.81% :     0.000233s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017733    76
 0.23% :     0.000041s :      1: add_recomputation
 0.34% :     0.000060s :      1: auto_monad
 0.07% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.48% :     0.000262s :      1: bootstrap
 0.10% :     0.000019s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.18% :     0.004642s :      1: jit_opt_a
 0.75% :     0.000134s :      1: jit_opt_after_cconv
 0.28% :     0.000050s :      1: jit_opt_b
 2.45% :     0.000434s :      1: loop_unroll
 2.66% :     0.000471s :      1: mutable_eliminate
 4.41% :     0.000782s :     26: opt.transform.jit_opt_a
 0.25% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000028s :      4: opt.transform.symbol_engine_opt
 2.60% :     0.000461s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000013s :      1: remove_dup_value
 1.45% :     0.000257s :      1: renormalize.infer
 1.28% :     0.000228s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000031s :      1: rewriter_after_opt_a
 0.38% :     0.000068s :      1: rewriter_before_opt_a
 0.39% :     0.000070s :      1: symbol_engine_optimizer
30.23% :     0.005360s :      1: task_emit
22.61% :     0.004009s :      1: type_inference
 0.23% :     0.000040s :      1: validate


TotalTime = 0.021024, [33]
[bootstrap]: 0.00033402
[type_inference]: 0.00487509
[event_method]: 1.917e-05
[auto_monad]: 6.584e-05
[graph_reusing]: 6.46e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 3.738e-05
[rewriter_before_opt_a]: 7.275e-05
[expand_dump_flag]: 3.58e-06
[jit_opt_a]: 0.00609011, [2]
    [Cycle 1]: 0.00205605, [27]
        [switch_simplify]: 6.05e-05
        [loop_unroll]: 2.89e-05
        [a_1]: 0.00076385
        [with_stream_mark]: 2.147e-05
        [recompute_prepare]: 7.66999e-06
        [updatestate_depend_eliminate]: 3.81001e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.81999e-06
        [parameter_eliminate]: 2.12001e-06
        [specialize_transform]: 6.21998e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 5.47999e-06
        [meta_shard_fg_expand]: 2.11e-06
        [get_grad_eliminate_]: 5.52001e-06
        [merge_forward]: 3.41001e-06
        [cell_reuse_recompute_pass]: 1.52001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.54e-05
        [j_node_and_user_rematch]: 9.07001e-06
        [meta_fg_expand]: 2.41e-06
        [replace_old_param]: 9.19e-06
        [inline_without_move]: 5.51e-06
        [renormalize]: 0.00083407
        [add_forward_monad_depend]: 5.35999e-06
        [auto_monad_grad]: 2.42001e-06
        [auto_monad_eliminator]: 1.363e-05
        [cse]: 2.953e-05
        [replace_applicator]: 1.148e-05
    [Cycle 2]: 0.00031588, [27]
        [switch_simplify]: 5.60001e-06
        [loop_unroll]: 4.50999e-06
        [a_1]: 8.953e-05
        [with_stream_mark]: 8.39998e-06
        [recompute_prepare]: 5.32001e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 2.11998e-06
        [updatestate_loads_eliminate]: 1.80001e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 4.86002e-06
        [updatestate_useless_node_eliminater]: 4.61002e-06
        [accelerated_algorithm]: 4.89003e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.29002e-06
        [merge_forward]: 2.98e-06
        [cell_reuse_recompute_pass]: 1.65001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.168e-05
        [j_node_and_user_rematch]: 7.73001e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 1.289e-05
        [inline_without_move]: 4.80001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.30999e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 4.99998e-06
        [cse]: 1.043e-05
        [replace_applicator]: 4.75001e-06
[py_interpret_to_execute_after_opt_a]: 1.13e-05
[rewriter_after_opt_a]: 3.578e-05
[convert_after_rewriter]: 6.81999e-06
[order_py_execute_after_rewriter]: 4.99998e-06
[mutable_eliminate]: 0.00059553
[jit_opt_b]: 4.912e-05, [1]
    [Cycle 1]: 4.197e-05, [2]
        [frontend_op_eliminate]: 1.643e-05
        [inline_after_opt_a]: 1.443e-05
[cconv]: 2.548e-05
[loop_unroll]: 0.00042155
[jit_opt_after_cconv]: 0.00013776, [1]
    [Cycle 1]: 0.00013149, [11]
        [c_1]: 1.991e-05
        [parameter_eliminate]: 2.61999e-06
        [updatestate_depend_eliminate]: 5.34e-06
        [updatestate_assign_eliminate]: 2.47001e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [cse]: 1.766e-05
        [call_graph_tuple_transform]: 1.86e-05
        [tuple_list_get_item_eliminator]: 5.24e-06
        [none_parameter_eliminate]: 1.45001e-06
        [renormalize]: 8.59989e-07
        [switch_simplify]: 5.43002e-06
[remove_dup_value]: 1.324e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 5.63002e-06
[add_recomputation]: 4.921e-05
[cse_after_recomputation]: 2.056e-05, [1]
    [Cycle 1]: 1.475e-05, [1]
        [cse]: 9.11998e-06
[auto_monad_reorder]: 1.516e-05
[get_jit_bprop_graph]: 1.86e-06
[rewriter_after_jit_bprop_graph]: 5.64e-06
[opt_after_jit_grad]: 0.00046068
[symbol_engine_optimizer]: 7.208e-05, [1]
    [Cycle 1]: 6.553e-05, [6]
        [build]: 2.98e-06
        [elim_shapecalc]: 7.14001e-06
        [elim_not_effective]: 1.338e-05
        [opt_reshape]: 5.24998e-06
        [fold_const_symbol]: 8.83001e-06
        [renormalize]: 1.12e-06
[validate]: 3.676e-05
[backend_pass]: 1.02e-06
[task_emit]: 0.00731886
[execute]: 8.99998e-06
Sums
    bootstrap                            :     0.000334s :  2.00%
    type_inference                       :     0.004875s : 29.25%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000066s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000037s :  0.22%
    rewriter_before_opt_a                :     0.000073s :  0.44%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.20%
    jit_opt_a.a_1                        :     0.000853s :  5.12%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.13%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000834s :  5.01%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.11%
    jit_opt_a.cse                        :     0.000040s :  0.24%
    jit_opt_a.replace_applicator         :     0.000016s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000596s :  3.57%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000025s :  0.15%
    loop_unroll                          :     0.000422s :  2.53%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000049s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000461s :  2.76%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000037s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007319s : 43.92%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000410    23
 0.56% :     0.000002s :      2: substitution.elim_not_effective
 0.30% :     0.000001s :      2: substitution.fold_const_symbol
 1.34% :     0.000006s :      3: substitution.graph_param_transform
90.05% :     0.000370s :      4: substitution.inline
 0.80% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.20% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.49% :     0.000010s :      2: substitution.replace_old_param
 3.24% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004803     2
81.21% :     0.003901s :      1: type_inference.infer
18.79% :     0.000903s :      1: type_inference.specialize
------[replace.]   0.000083     6
69.87% :     0.000058s :      4: replace.inline
30.13% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000378     6
96.81% :     0.000366s :      4: match.inline
 3.19% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.68% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.28% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 1.95% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.44% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.42% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.24% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.38% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.66% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.66% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.93% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.27% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.56% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.01% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.54% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.71% :     0.000009s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.31% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.04% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.56% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.43% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000689    12
47.72% :     0.000329s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.39% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.89% :     0.000309s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023018    76
 0.23% :     0.000052s :      1: add_recomputation
 0.30% :     0.000069s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.55% :     0.000356s :      1: bootstrap
 0.12% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.47% :     0.006094s :      1: jit_opt_a
 0.61% :     0.000140s :      1: jit_opt_after_cconv
 0.22% :     0.000052s :      1: jit_opt_b
 1.87% :     0.000430s :      1: loop_unroll
 2.62% :     0.000604s :      1: mutable_eliminate
 4.64% :     0.001069s :     26: opt.transform.jit_opt_a
 0.20% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.03% :     0.000468s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000040s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.02% :     0.000464s :      1: renormalize.infer
 1.57% :     0.000362s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000039s :      1: rewriter_after_opt_a
 0.33% :     0.000076s :      1: rewriter_before_opt_a
 0.32% :     0.000075s :      1: symbol_engine_optimizer
31.88% :     0.007337s :      1: task_emit
21.24% :     0.004890s :      1: type_inference
 0.24% :     0.000056s :      1: validate


TotalTime = 0.0127145, [30]
[bootstrap]: 0.00023081
[type_inference]: 0.00556854
[event_method]: 1.861e-05
[auto_monad]: 5.713e-05
[graph_reusing]: 3.93001e-06
[pre_auto_parallel]: 1.87999e-06
[py_interpret_to_execute]: 4.566e-05
[rewriter_before_opt_a]: 7.517e-05
[expand_dump_flag]: 1.97001e-06
[jit_opt_a]: 0.00479721, [2]
    [Cycle 1]: 0.00158978, [27]
        [switch_simplify]: 8.967e-05
        [loop_unroll]: 3.496e-05
        [a_1]: 0.00064687
        [with_stream_mark]: 1.211e-05
        [recompute_prepare]: 7.18e-06
        [updatestate_depend_eliminate]: 2.96001e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 6.05002e-06
        [updatestate_useless_node_eliminater]: 5.26002e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.44998e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 2.51e-06
        [cell_reuse_recompute_pass]: 9.60019e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.211e-05
        [j_node_and_user_rematch]: 8.67998e-06
        [meta_fg_expand]: 1.72999e-06
        [replace_old_param]: 7.51001e-06
        [inline_without_move]: 5.10999e-06
        [renormalize]: 0.00052565
        [add_forward_monad_depend]: 4.72e-06
        [auto_monad_grad]: 1.37e-06
        [auto_monad_eliminator]: 9.47001e-06
        [cse]: 1.568e-05
        [replace_applicator]: 1.079e-05
    [Cycle 2]: 0.00031049, [27]
        [switch_simplify]: 5.71e-06
        [loop_unroll]: 5.02e-06
        [a_1]: 9.026e-05
        [with_stream_mark]: 8.17e-06
        [recompute_prepare]: 5.21002e-06
        [updatestate_depend_eliminate]: 2.58e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 4.86002e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 4.68001e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.42001e-06
        [cell_reuse_recompute_pass]: 1.37999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.212e-05
        [j_node_and_user_rematch]: 7.71999e-06
        [meta_fg_expand]: 1.45999e-06
        [replace_old_param]: 6.83e-06
        [inline_without_move]: 4.75001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 7.79983e-07
        [auto_monad_eliminator]: 4.87e-06
        [cse]: 9.39e-06
        [replace_applicator]: 4.90999e-06
[py_interpret_to_execute_after_opt_a]: 7.77e-06
[rewriter_after_opt_a]: 2.388e-05
[convert_after_rewriter]: 5.59e-06
[order_py_execute_after_rewriter]: 4.28999e-06
[mutable_eliminate]: 0.00046674
[jit_opt_b]: 5.702e-05, [1]
    [Cycle 1]: 5.061e-05, [2]
        [frontend_op_eliminate]: 1.63e-05
        [inline_after_opt_a]: 2.35e-05
[cconv]: 2.734e-05
[loop_unroll]: 0.00041083
[jit_opt_after_cconv]: 0.00013482, [1]
    [Cycle 1]: 0.00012877, [11]
        [c_1]: 2.141e-05
        [parameter_eliminate]: 2.66999e-06
        [updatestate_depend_eliminate]: 5.44e-06
        [updatestate_assign_eliminate]: 2.30002e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.629e-05
        [call_graph_tuple_transform]: 1.617e-05
        [tuple_list_get_item_eliminator]: 5.59998e-06
        [none_parameter_eliminate]: 9.89996e-07
        [renormalize]: 4.89992e-07
        [switch_simplify]: 5.67001e-06
[remove_dup_value]: 7.28999e-06
[partial_unused_args_eliminate]: 1.55999e-06
[environ_conv]: 3.53e-06
[add_recomputation]: 3.245e-05
[cse_after_recomputation]: 2.086e-05, [1]
    [Cycle 1]: 1.557e-05, [1]
        [cse]: 9.96e-06
[auto_monad_reorder]: 8.97e-06
[get_jit_bprop_graph]: 1.56002e-06
[rewriter_after_jit_bprop_graph]: 4.84998e-06
[opt_after_jit_grad]: 0.00045409
[symbol_engine_optimizer]: 6.802e-05, [1]
    [Cycle 1]: 6.242e-05, [6]
        [build]: 2.56998e-06
        [elim_shapecalc]: 6.96999e-06
        [elim_not_effective]: 1.145e-05
        [opt_reshape]: 5.67001e-06
        [fold_const_symbol]: 7.99002e-06
        [renormalize]: 5.8001e-07
[validate]: 2.429e-05
Sums
    bootstrap                            :     0.000231s :  2.49%
    type_inference                       :     0.005569s : 60.00%
    event_method                         :     0.000019s :  0.20%
    auto_monad                           :     0.000057s :  0.62%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000046s :  0.49%
    rewriter_before_opt_a                :     0.000075s :  0.81%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000095s :  1.03%
    jit_opt_a.loop_unroll                :     0.000040s :  0.43%
    jit_opt_a.a_1                        :     0.000737s :  7.94%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000526s :  5.66%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.15%
    jit_opt_a.cse                        :     0.000025s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.08%
    rewriter_after_opt_a                 :     0.000024s :  0.26%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000467s :  5.03%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.25%
    cconv                                :     0.000027s :  0.29%
    loop_unroll                          :     0.000411s :  4.43%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000007s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000032s :  0.35%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000009s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000454s :  4.89%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000024s :  0.26%


Time group info:
------[substitution.]   0.000234    26
 0.57% :     0.000001s :      2: substitution.elim_not_effective
 0.39% :     0.000001s :      2: substitution.fold_const_symbol
 1.44% :     0.000003s :      3: substitution.graph_param_transform
87.28% :     0.000205s :      6: substitution.inline
 1.09% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.68% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.17% :     0.000003s :      2: substitution.replace_old_param
 3.01% :     0.000007s :      1: substitution.switch_simplify
 3.38% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005524     2
83.53% :     0.004614s :      1: type_inference.infer
16.47% :     0.000910s :      1: type_inference.specialize
------[replace.]   0.000115     9
56.33% :     0.000065s :      6: replace.inline
18.23% :     0.000021s :      1: replace.switch_simplify
25.44% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000214     9
93.93% :     0.000201s :      6: match.inline
 2.94% :     0.000006s :      1: match.switch_simplify
 3.13% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   801
 1.27% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.13% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     12: predicate.addn_check_dump
 1.39% :     0.000002s :     12: predicate.addn_zero_filter
 1.90% :     0.000002s :     12: predicate.arithmetic_simplify
 1.46% :     0.000002s :     12: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     12: predicate.compare_switch_simplify
 1.60% :     0.000002s :     12: predicate.depend_value_elim
 1.19% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.53% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.70% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.33% :     0.000002s :     12: predicate.environ_get_eliminate
 1.26% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.43% :     0.000007s :     26: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.98% :     0.000001s :      6: predicate.less_batch_normalization
 1.69% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     17: predicate.load_eliminater
 1.23% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.39% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.06% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     12: predicate.merge_addn
 1.43% :     0.000002s :     12: predicate.minmaximum_grad
 1.44% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     17: predicate.partial_eliminate
 1.27% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     12: predicate.reduce_eliminate
 1.40% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     20: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.38% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     12: predicate.reshape_eliminate
 1.29% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     12: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.75% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.65% :     0.000003s :     20: predicate.switch_defer_inline
 2.30% :     0.000003s :     20: predicate.switch_layer_defer_inline
 7.93% :     0.000010s :     65: predicate.switch_simplify
 1.47% :     0.000002s :     12: predicate.tile_eliminate
 1.45% :     0.000002s :     12: predicate.transpose_eliminate
 1.50% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.37% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.27% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     12: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000721    15
47.58% :     0.000343s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.43% :     0.000075s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.99% :     0.000303s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014298    72
 0.25% :     0.000036s :      1: add_recomputation
 0.42% :     0.000060s :      1: auto_monad
 0.08% :     0.000011s :      1: auto_monad_reorder
 1.69% :     0.000241s :      1: bootstrap
 0.21% :     0.000031s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.16% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.16% :     0.000022s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
33.57% :     0.004800s :      1: jit_opt_a
 0.96% :     0.000138s :      1: jit_opt_after_cconv
 0.42% :     0.000060s :      1: jit_opt_b
 2.93% :     0.000418s :      1: loop_unroll
 3.32% :     0.000475s :      1: mutable_eliminate
 6.85% :     0.000980s :     26: opt.transform.jit_opt_a
 0.32% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.23% :     0.000033s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000029s :      4: opt.transform.symbol_engine_opt
 3.23% :     0.000462s :      1: opt_after_jit_grad
 0.05% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.34% :     0.000048s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000010s :      1: remove_dup_value
 1.76% :     0.000251s :      1: renormalize.infer
 1.87% :     0.000268s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000027s :      1: rewriter_after_opt_a
 0.55% :     0.000078s :      1: rewriter_before_opt_a
 0.49% :     0.000070s :      1: symbol_engine_optimizer
39.01% :     0.005577s :      1: type_inference


TotalTime = 0.0168115, [33]
[bootstrap]: 0.00025365
[type_inference]: 0.00397639
[event_method]: 1.541e-05
[auto_monad]: 5.537e-05
[graph_reusing]: 4.22e-06
[pre_auto_parallel]: 1.99e-06
[py_interpret_to_execute]: 2.727e-05
[rewriter_before_opt_a]: 6.609e-05
[expand_dump_flag]: 2.42001e-06
[jit_opt_a]: 0.00465945, [2]
    [Cycle 1]: 0.00137789, [27]
        [switch_simplify]: 4.365e-05
        [loop_unroll]: 2.846e-05
        [a_1]: 0.00047937
        [with_stream_mark]: 1.344e-05
        [recompute_prepare]: 6.94999e-06
        [updatestate_depend_eliminate]: 2.89999e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 1.18001e-06
        [specialize_transform]: 5.69e-06
        [updatestate_useless_node_eliminater]: 4.64002e-06
        [accelerated_algorithm]: 5.27001e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 5.05001e-06
        [merge_forward]: 2.63998e-06
        [cell_reuse_recompute_pass]: 9.10019e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.197e-05
        [j_node_and_user_rematch]: 8.69e-06
        [meta_fg_expand]: 2.17001e-06
        [replace_old_param]: 7.88999e-06
        [inline_without_move]: 5.09003e-06
        [renormalize]: 0.00053699
        [add_forward_monad_depend]: 4.64002e-06
        [auto_monad_grad]: 1.49e-06
        [auto_monad_eliminator]: 1.032e-05
        [cse]: 1.557e-05
        [replace_applicator]: 1.074e-05
    [Cycle 2]: 0.0003124, [27]
        [switch_simplify]: 5.30999e-06
        [loop_unroll]: 4.94003e-06
        [a_1]: 8.829e-05
        [with_stream_mark]: 8.62998e-06
        [recompute_prepare]: 4.97999e-06
        [updatestate_depend_eliminate]: 2.49999e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.85999e-06
        [updatestate_useless_node_eliminater]: 4.58999e-06
        [accelerated_algorithm]: 5.09998e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.35e-06
        [merge_forward]: 2.31e-06
        [cell_reuse_recompute_pass]: 1.50001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.239e-05
        [j_node_and_user_rematch]: 7.47002e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 6.98998e-06
        [inline_without_move]: 4.49998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.28002e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 4.60999e-06
        [cse]: 9.99001e-06
        [replace_applicator]: 4.78001e-06
[py_interpret_to_execute_after_opt_a]: 8.13001e-06
[rewriter_after_opt_a]: 2.71e-05
[convert_after_rewriter]: 6.11998e-06
[order_py_execute_after_rewriter]: 4.81002e-06
[mutable_eliminate]: 0.00047027
[jit_opt_b]: 5.124e-05, [1]
    [Cycle 1]: 4.452e-05, [2]
        [frontend_op_eliminate]: 1.756e-05
        [inline_after_opt_a]: 1.542e-05
[cconv]: 1.912e-05
[loop_unroll]: 0.00040538
[jit_opt_after_cconv]: 0.00013009, [1]
    [Cycle 1]: 0.0001241, [11]
        [c_1]: 2.073e-05
        [parameter_eliminate]: 2.27001e-06
        [updatestate_depend_eliminate]: 5.05999e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [cse]: 1.523e-05
        [call_graph_tuple_transform]: 1.588e-05
        [tuple_list_get_item_eliminator]: 5.99999e-06
        [none_parameter_eliminate]: 1.14e-06
        [renormalize]: 4.90021e-07
        [switch_simplify]: 5.12999e-06
[remove_dup_value]: 9.02999e-06
[partial_unused_args_eliminate]: 2.03002e-06
[environ_conv]: 3.77998e-06
[add_recomputation]: 3.749e-05
[cse_after_recomputation]: 1.931e-05, [1]
    [Cycle 1]: 1.437e-05, [1]
        [cse]: 8.45001e-06
[auto_monad_reorder]: 1.158e-05
[get_jit_bprop_graph]: 1.39e-06
[rewriter_after_jit_bprop_graph]: 4.14002e-06
[opt_after_jit_grad]: 0.00043591
[symbol_engine_optimizer]: 6.912e-05, [1]
    [Cycle 1]: 6.283e-05, [6]
        [build]: 2.79999e-06
        [elim_shapecalc]: 7.25998e-06
        [elim_not_effective]: 1.147e-05
        [opt_reshape]: 5.41002e-06
        [fold_const_symbol]: 8.25e-06
        [renormalize]: 5.10016e-07
[validate]: 2.857e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00582504
[execute]: 5.35001e-06
Sums
    bootstrap                            :     0.000254s :  1.91%
    type_inference                       :     0.003976s : 29.94%
    event_method                         :     0.000015s :  0.12%
    auto_monad                           :     0.000055s :  0.42%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000027s :  0.21%
    rewriter_before_opt_a                :     0.000066s :  0.50%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000049s :  0.37%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000568s :  4.27%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000537s :  4.04%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000026s :  0.19%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000027s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000470s :  3.54%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000019s :  0.14%
    loop_unroll                          :     0.000405s :  3.05%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000009s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000037s :  0.28%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.09%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000436s :  3.28%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000029s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005825s : 43.86%
    execute                              :     0.000005s :  0.04%


Time group info:
------[substitution.]   0.000159    23
 0.89% :     0.000001s :      2: substitution.elim_not_effective
 0.89% :     0.000001s :      2: substitution.fold_const_symbol
 2.54% :     0.000004s :      3: substitution.graph_param_transform
82.60% :     0.000131s :      4: substitution.inline
 1.74% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.63% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.14% :     0.000003s :      2: substitution.replace_old_param
 6.58% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003928     2
81.37% :     0.003196s :      1: type_inference.infer
18.63% :     0.000732s :      1: type_inference.specialize
------[replace.]   0.000061     6
66.06% :     0.000040s :      4: replace.inline
33.94% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000138     6
93.19% :     0.000129s :      4: match.inline
 6.81% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.61% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.51% :     0.000002s :     10: predicate.addn_zero_filter
 1.89% :     0.000002s :     10: predicate.arithmetic_simplify
 1.32% :     0.000001s :     10: predicate.cast_eliminate
 0.71% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000001s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.52% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.28% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.87% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.87% :     0.000002s :     15: predicate.load_eliminater
 1.22% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.25% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.29% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 2.06% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000002s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.44% :     0.000002s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.21% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.43% :     0.000008s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.25% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.27% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.86% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000487    12
47.12% :     0.000230s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.51% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.37% :     0.000231s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018212    76
 0.22% :     0.000040s :      1: add_recomputation
 0.32% :     0.000058s :      1: auto_monad
 0.08% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.46% :     0.000265s :      1: bootstrap
 0.12% :     0.000022s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
25.60% :     0.004662s :      1: jit_opt_a
 0.73% :     0.000133s :      1: jit_opt_after_cconv
 0.30% :     0.000054s :      1: jit_opt_b
 2.27% :     0.000413s :      1: loop_unroll
 2.63% :     0.000479s :      1: mutable_eliminate
 4.16% :     0.000758s :     26: opt.transform.jit_opt_a
 0.25% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.44% :     0.000444s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000011s :      1: remove_dup_value
 1.60% :     0.000292s :      1: renormalize.infer
 1.31% :     0.000239s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000030s :      1: rewriter_after_opt_a
 0.38% :     0.000069s :      1: rewriter_before_opt_a
 0.39% :     0.000072s :      1: symbol_engine_optimizer
32.05% :     0.005836s :      1: task_emit
21.89% :     0.003987s :      1: type_inference
 0.24% :     0.000044s :      1: validate


TotalTime = 0.0172144, [33]
[bootstrap]: 0.00024337
[type_inference]: 0.0039652
[event_method]: 1.571e-05
[auto_monad]: 5.483e-05
[graph_reusing]: 4.57e-06
[pre_auto_parallel]: 2.14e-06
[py_interpret_to_execute]: 3.308e-05
[rewriter_before_opt_a]: 6.604e-05
[expand_dump_flag]: 2.88e-06
[jit_opt_a]: 0.00491877, [2]
    [Cycle 1]: 0.00150601, [27]
        [switch_simplify]: 5.294e-05
        [loop_unroll]: 3.103e-05
        [a_1]: 0.00050639
        [with_stream_mark]: 1.531e-05
        [recompute_prepare]: 6.79001e-06
        [updatestate_depend_eliminate]: 3.48e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [parameter_eliminate]: 2.01003e-06
        [specialize_transform]: 5.52999e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 5.87001e-06
        [meta_shard_fg_expand]: 1.93002e-06
        [get_grad_eliminate_]: 5.15001e-06
        [merge_forward]: 2.98e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.316e-05
        [j_node_and_user_rematch]: 8.69998e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 8.08999e-06
        [inline_without_move]: 5.09998e-06
        [renormalize]: 0.00054128
        [add_forward_monad_depend]: 4.52998e-06
        [auto_monad_grad]: 2.16e-06
        [auto_monad_eliminator]: 1.344e-05
        [cse]: 2.015e-05
        [replace_applicator]: 1.046e-05
    [Cycle 2]: 0.00030828, [27]
        [switch_simplify]: 5.54e-06
        [loop_unroll]: 4.75999e-06
        [a_1]: 8.947e-05
        [with_stream_mark]: 8.27e-06
        [recompute_prepare]: 4.90999e-06
        [updatestate_depend_eliminate]: 2.49999e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 4.89003e-06
        [updatestate_useless_node_eliminater]: 4.49998e-06
        [accelerated_algorithm]: 4.89e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 2.34999e-06
        [cell_reuse_recompute_pass]: 1.94e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.221e-05
        [j_node_and_user_rematch]: 7.78001e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 7.01001e-06
        [inline_without_move]: 4.67e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.18001e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 4.85001e-06
        [cse]: 9.24998e-06
        [replace_applicator]: 4.79e-06
[py_interpret_to_execute_after_opt_a]: 9.57001e-06
[rewriter_after_opt_a]: 3.072e-05
[convert_after_rewriter]: 6.11e-06
[order_py_execute_after_rewriter]: 4.75999e-06
[mutable_eliminate]: 0.00050098
[jit_opt_b]: 4.675e-05, [1]
    [Cycle 1]: 3.998e-05, [2]
        [frontend_op_eliminate]: 1.476e-05
        [inline_after_opt_a]: 1.44e-05
[cconv]: 2.312e-05
[loop_unroll]: 0.00042153
[jit_opt_after_cconv]: 0.00014173, [1]
    [Cycle 1]: 0.00013537, [11]
        [c_1]: 2.061e-05
        [parameter_eliminate]: 2.68998e-06
        [updatestate_depend_eliminate]: 5.32999e-06
        [updatestate_assign_eliminate]: 2.43002e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 1.635e-05
        [call_graph_tuple_transform]: 1.709e-05
        [tuple_list_get_item_eliminator]: 5.32001e-06
        [none_parameter_eliminate]: 1.93002e-06
        [renormalize]: 4.7998e-07
        [switch_simplify]: 5.24e-06
[remove_dup_value]: 1.253e-05
[partial_unused_args_eliminate]: 2.04e-06
[environ_conv]: 4.99998e-06
[add_recomputation]: 4.286e-05
[cse_after_recomputation]: 2.002e-05, [1]
    [Cycle 1]: 1.443e-05, [1]
        [cse]: 8.69998e-06
[auto_monad_reorder]: 1.346e-05
[get_jit_bprop_graph]: 1.55001e-06
[rewriter_after_jit_bprop_graph]: 4.62e-06
[opt_after_jit_grad]: 0.00046342
[symbol_engine_optimizer]: 6.972e-05, [1]
    [Cycle 1]: 6.357e-05, [6]
        [build]: 3.33e-06
        [elim_shapecalc]: 7.21999e-06
        [elim_not_effective]: 1.178e-05
        [opt_reshape]: 5.40001e-06
        [fold_const_symbol]: 8.35001e-06
        [renormalize]: 4.89992e-07
[validate]: 3.182e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00587759
[execute]: 7.92e-06
Sums
    bootstrap                            :     0.000243s :  1.81%
    type_inference                       :     0.003965s : 29.41%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000055s :  0.41%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.25%
    rewriter_before_opt_a                :     0.000066s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.43%
    jit_opt_a.loop_unroll                :     0.000036s :  0.27%
    jit_opt_a.a_1                        :     0.000596s :  4.42%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000541s :  4.02%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.14%
    jit_opt_a.cse                        :     0.000029s :  0.22%
    jit_opt_a.replace_applicator         :     0.000015s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000031s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000501s :  3.72%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000422s :  3.13%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000043s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000013s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000463s :  3.44%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005878s : 43.60%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000188    23
 0.95% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.68% :     0.000005s :      3: substitution.graph_param_transform
84.21% :     0.000159s :      4: substitution.inline
 1.66% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.24% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.87% :     0.000004s :      2: substitution.replace_old_param
 5.68% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003918     2
81.33% :     0.003187s :      1: type_inference.infer
18.67% :     0.000731s :      1: type_inference.specialize
------[replace.]   0.000061     6
66.64% :     0.000041s :      4: replace.inline
33.36% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000166     6
94.20% :     0.000156s :      4: match.inline
 5.80% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.84% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.04% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.46% :     0.000002s :     10: predicate.depend_value_elim
 1.20% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.18% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.26% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.59% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.24% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.04% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.82% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.45% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.87% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.40% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.77% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.02% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000499    12
46.46% :     0.000232s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.08% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.46% :     0.000232s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018658    76
 0.24% :     0.000046s :      1: add_recomputation
 0.31% :     0.000058s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.35% :     0.000252s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.38% :     0.004921s :      1: jit_opt_a
 0.77% :     0.000145s :      1: jit_opt_after_cconv
 0.26% :     0.000049s :      1: jit_opt_b
 2.30% :     0.000429s :      1: loop_unroll
 2.73% :     0.000509s :      1: mutable_eliminate
 4.28% :     0.000798s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.52% :     0.000471s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.55% :     0.000289s :      1: renormalize.infer
 1.32% :     0.000246s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.37% :     0.000069s :      1: rewriter_before_opt_a
 0.39% :     0.000072s :      1: symbol_engine_optimizer
31.56% :     0.005889s :      1: task_emit
21.30% :     0.003975s :      1: type_inference
 0.27% :     0.000050s :      1: validate


TotalTime = 0.0205618, [33]
[bootstrap]: 0.00030823
[type_inference]: 0.00470488
[event_method]: 1.741e-05
[auto_monad]: 6.788e-05
[graph_reusing]: 5.81998e-06
[pre_auto_parallel]: 2.69001e-06
[py_interpret_to_execute]: 3.512e-05
[rewriter_before_opt_a]: 7.253e-05
[expand_dump_flag]: 3.25e-06
[jit_opt_a]: 0.00628652, [2]
    [Cycle 1]: 0.00174133, [27]
        [switch_simplify]: 5.879e-05
        [loop_unroll]: 3.002e-05
        [a_1]: 0.00054743
        [with_stream_mark]: 1.934e-05
        [recompute_prepare]: 6.78998e-06
        [updatestate_depend_eliminate]: 3.81999e-06
        [updatestate_assign_eliminate]: 3.14001e-06
        [updatestate_loads_eliminate]: 3.01999e-06
        [parameter_eliminate]: 1.92001e-06
        [specialize_transform]: 6.30002e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.87001e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 3.59002e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.456e-05
        [j_node_and_user_rematch]: 9.05999e-06
        [meta_fg_expand]: 2.70002e-06
        [replace_old_param]: 8.87e-06
        [inline_without_move]: 5.20001e-06
        [renormalize]: 0.00076041
        [add_forward_monad_depend]: 5.51e-06
        [auto_monad_grad]: 2.43e-06
        [auto_monad_eliminator]: 1.342e-05
        [cse]: 2.999e-05
        [replace_applicator]: 1.282e-05
    [Cycle 2]: 0.00036209, [27]
        [switch_simplify]: 5.58002e-06
        [loop_unroll]: 4.85999e-06
        [a_1]: 8.863e-05
        [with_stream_mark]: 8.20999e-06
        [recompute_prepare]: 4.88001e-06
        [updatestate_depend_eliminate]: 2.41e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 4.78001e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 5.15999e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.30999e-06
        [merge_forward]: 2.31e-06
        [cell_reuse_recompute_pass]: 2.06e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.189e-05
        [j_node_and_user_rematch]: 7.61001e-06
        [meta_fg_expand]: 1.95001e-06
        [replace_old_param]: 6.86001e-06
        [inline_without_move]: 4.679e-05
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.74e-06
        [cse]: 1.183e-05
        [replace_applicator]: 6.12999e-06
[py_interpret_to_execute_after_opt_a]: 1.231e-05
[rewriter_after_opt_a]: 3.945e-05
[convert_after_rewriter]: 8.27998e-06
[order_py_execute_after_rewriter]: 5.92999e-06
[mutable_eliminate]: 0.00060791
[jit_opt_b]: 5.205e-05, [1]
    [Cycle 1]: 4.461e-05, [2]
        [frontend_op_eliminate]: 1.66e-05
        [inline_after_opt_a]: 1.647e-05
[cconv]: 2.759e-05
[loop_unroll]: 0.00044389
[jit_opt_after_cconv]: 0.00014212, [1]
    [Cycle 1]: 0.00013504, [11]
        [c_1]: 2.071e-05
        [parameter_eliminate]: 2.41998e-06
        [updatestate_depend_eliminate]: 5.20999e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 2.027e-05
        [call_graph_tuple_transform]: 1.913e-05
        [tuple_list_get_item_eliminator]: 6.26e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 5.79981e-07
        [switch_simplify]: 5.40999e-06
[remove_dup_value]: 1.471e-05
[partial_unused_args_eliminate]: 2.47001e-06
[environ_conv]: 5.12e-06
[add_recomputation]: 5.384e-05
[cse_after_recomputation]: 2.141e-05, [1]
    [Cycle 1]: 1.565e-05, [1]
        [cse]: 9.42999e-06
[auto_monad_reorder]: 1.612e-05
[get_jit_bprop_graph]: 2.07999e-06
[rewriter_after_jit_bprop_graph]: 5.17999e-06
[opt_after_jit_grad]: 0.00047524
[symbol_engine_optimizer]: 7.747e-05, [1]
    [Cycle 1]: 7.076e-05, [6]
        [build]: 3.66999e-06
        [elim_shapecalc]: 8.37e-06
        [elim_not_effective]: 1.24e-05
        [opt_reshape]: 6.19001e-06
        [fold_const_symbol]: 9.89001e-06
        [renormalize]: 4.20026e-07
[validate]: 4.086e-05
[backend_pass]: 1.31998e-06
[task_emit]: 0.00676923
[execute]: 9.55001e-06
Sums
    bootstrap                            :     0.000308s :  1.96%
    type_inference                       :     0.004705s : 29.89%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000068s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.22%
    rewriter_before_opt_a                :     0.000073s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.41%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000636s :  4.04%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000052s :  0.33%
    jit_opt_a.renormalize                :     0.000760s :  4.83%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000042s :  0.27%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000039s :  0.25%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000608s :  3.86%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000028s :  0.18%
    loop_unroll                          :     0.000444s :  2.82%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000054s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000475s :  3.02%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000041s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006769s : 43.00%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000213    23
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.69% :     0.000006s :      3: substitution.graph_param_transform
84.99% :     0.000181s :      4: substitution.inline
 1.40% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.03% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.67% :     0.000004s :      2: substitution.replace_old_param
 5.66% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004635     2
81.26% :     0.003766s :      1: type_inference.infer
18.74% :     0.000868s :      1: type_inference.specialize
------[replace.]   0.000070     6
66.70% :     0.000047s :      4: replace.inline
33.30% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000189     6
94.23% :     0.000178s :      4: match.inline
 5.77% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.58% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.21% :     0.000001s :     10: predicate.addn_zero_filter
 2.25% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.60% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.16% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.62% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 2.17% :     0.000003s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 1.98% :     0.000002s :     15: predicate.partial_eliminate
 1.39% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.41% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.52% :     0.000002s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.23% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.85% :     0.000009s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.43% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.47% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000657    12
51.48% :     0.000338s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.89% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.63% :     0.000287s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022312    76
 0.26% :     0.000057s :      1: add_recomputation
 0.32% :     0.000071s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.48% :     0.000331s :      1: bootstrap
 0.14% :     0.000031s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.11% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
28.19% :     0.006290s :      1: jit_opt_a
 0.65% :     0.000145s :      1: jit_opt_after_cconv
 0.25% :     0.000055s :      1: jit_opt_b
 2.02% :     0.000451s :      1: loop_unroll
 2.76% :     0.000617s :      1: mutable_eliminate
 3.99% :     0.000889s :     26: opt.transform.jit_opt_a
 0.21% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.17% :     0.000483s :      1: opt_after_jit_grad
 0.04% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000038s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.75% :     0.000391s :      1: renormalize.infer
 1.62% :     0.000361s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000043s :      1: rewriter_after_opt_a
 0.34% :     0.000075s :      1: rewriter_before_opt_a
 0.36% :     0.000080s :      1: symbol_engine_optimizer
30.43% :     0.006789s :      1: task_emit
21.15% :     0.004720s :      1: type_inference
 0.28% :     0.000063s :      1: validate


TotalTime = 0.0147512, [30]
[bootstrap]: 0.00026171
[type_inference]: 0.00668214
[event_method]: 2.278e-05
[auto_monad]: 7.231e-05
[graph_reusing]: 5.54998e-06
[pre_auto_parallel]: 2.34001e-06
[py_interpret_to_execute]: 6.306e-05
[rewriter_before_opt_a]: 8.825e-05
[expand_dump_flag]: 3.18e-06
[jit_opt_a]: 0.0055197, [2]
    [Cycle 1]: 0.00171466, [27]
        [switch_simplify]: 9.397e-05
        [loop_unroll]: 5.588e-05
        [a_1]: 0.0006534
        [with_stream_mark]: 1.294e-05
        [recompute_prepare]: 7.08e-06
        [updatestate_depend_eliminate]: 2.89999e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 1.07998e-06
        [specialize_transform]: 6.14001e-06
        [updatestate_useless_node_eliminater]: 5.26002e-06
        [accelerated_algorithm]: 5.98002e-06
        [meta_shard_fg_expand]: 1.57999e-06
        [get_grad_eliminate_]: 5.55001e-06
        [merge_forward]: 2.50002e-06
        [cell_reuse_recompute_pass]: 8.70001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.919e-05
        [j_node_and_user_rematch]: 8.66002e-06
        [meta_fg_expand]: 1.95001e-06
        [replace_old_param]: 8.45001e-06
        [inline_without_move]: 5.40999e-06
        [renormalize]: 0.00058108
        [add_forward_monad_depend]: 5.10001e-06
        [auto_monad_grad]: 1.39998e-06
        [auto_monad_eliminator]: 1.064e-05
        [cse]: 3.43e-05
        [replace_applicator]: 1.132e-05
    [Cycle 2]: 0.00039381, [27]
        [switch_simplify]: 5.42999e-06
        [loop_unroll]: 5.09e-06
        [a_1]: 0.00010995
        [with_stream_mark]: 9.12001e-06
        [recompute_prepare]: 5.20001e-06
        [updatestate_depend_eliminate]: 2.44001e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 5.27001e-06
        [updatestate_useless_node_eliminater]: 5.32999e-06
        [accelerated_algorithm]: 4.72998e-06
        [meta_shard_fg_expand]: 1.14998e-06
        [get_grad_eliminate_]: 4.55001e-06
        [merge_forward]: 2.48002e-06
        [cell_reuse_recompute_pass]: 1.30001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.206e-05
        [j_node_and_user_rematch]: 7.61999e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 7.04001e-06
        [inline_without_move]: 4.73001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 6.90023e-07
        [auto_monad_eliminator]: 5.04003e-06
        [cse]: 6.686e-05
        [replace_applicator]: 5.57001e-06
[py_interpret_to_execute_after_opt_a]: 8.38999e-06
[rewriter_after_opt_a]: 2.595e-05
[convert_after_rewriter]: 5.61e-06
[order_py_execute_after_rewriter]: 4.12e-06
[mutable_eliminate]: 0.00047869
[jit_opt_b]: 4.794e-05, [1]
    [Cycle 1]: 4.141e-05, [2]
        [frontend_op_eliminate]: 1.575e-05
        [inline_after_opt_a]: 1.453e-05
[cconv]: 1.58e-05
[loop_unroll]: 0.00045594
[jit_opt_after_cconv]: 0.00013415, [1]
    [Cycle 1]: 0.00012806, [11]
        [c_1]: 2.022e-05
        [parameter_eliminate]: 2.54999e-06
        [updatestate_depend_eliminate]: 5.47999e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [cse]: 1.68e-05
        [call_graph_tuple_transform]: 1.627e-05
        [tuple_list_get_item_eliminator]: 5.35999e-06
        [none_parameter_eliminate]: 1.09003e-06
        [renormalize]: 2.79979e-07
        [switch_simplify]: 5.75001e-06
[remove_dup_value]: 7.33999e-06
[partial_unused_args_eliminate]: 1.75001e-06
[environ_conv]: 3.83999e-06
[add_recomputation]: 3.397e-05
[cse_after_recomputation]: 1.982e-05, [1]
    [Cycle 1]: 1.465e-05, [1]
        [cse]: 8.90999e-06
[auto_monad_reorder]: 9.22001e-06
[get_jit_bprop_graph]: 1.49e-06
[rewriter_after_jit_bprop_graph]: 5.21002e-06
[opt_after_jit_grad]: 0.0005033
[symbol_engine_optimizer]: 7.165e-05, [1]
    [Cycle 1]: 6.564e-05, [6]
        [build]: 2.81e-06
        [elim_shapecalc]: 7.95e-06
        [elim_not_effective]: 1.231e-05
        [opt_reshape]: 5.69999e-06
        [fold_const_symbol]: 7.89002e-06
        [renormalize]: 4.39992e-07
[validate]: 2.57e-05
Sums
    bootstrap                            :     0.000262s :  2.43%
    type_inference                       :     0.006682s : 62.05%
    event_method                         :     0.000023s :  0.21%
    auto_monad                           :     0.000072s :  0.67%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000063s :  0.59%
    rewriter_before_opt_a                :     0.000088s :  0.82%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000099s :  0.92%
    jit_opt_a.loop_unroll                :     0.000061s :  0.57%
    jit_opt_a.a_1                        :     0.000763s :  7.09%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000581s :  5.40%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.15%
    jit_opt_a.cse                        :     0.000101s :  0.94%
    jit_opt_a.replace_applicator         :     0.000017s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.08%
    rewriter_after_opt_a                 :     0.000026s :  0.24%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000479s :  4.45%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.13%
    cconv                                :     0.000016s :  0.15%
    loop_unroll                          :     0.000456s :  4.23%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000007s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000034s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000009s :  0.09%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000503s :  4.67%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000026s :  0.24%


Time group info:
------[substitution.]   0.000233    26
 0.60% :     0.000001s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 1.77% :     0.000004s :      3: substitution.graph_param_transform
85.66% :     0.000200s :      6: substitution.inline
 1.17% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.76% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.26% :     0.000003s :      2: substitution.replace_old_param
 3.60% :     0.000008s :      1: substitution.switch_simplify
 3.75% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006617     2
83.00% :     0.005492s :      1: type_inference.infer
17.00% :     0.001125s :      1: type_inference.specialize
------[replace.]   0.000118     9
54.89% :     0.000065s :      6: replace.inline
19.52% :     0.000023s :      1: replace.switch_simplify
25.59% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000212     9
92.80% :     0.000196s :      6: match.inline
 3.61% :     0.000008s :      1: match.switch_simplify
 3.59% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   801
 1.46% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.16% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.19% :     0.000002s :     12: predicate.addn_check_dump
 1.42% :     0.000002s :     12: predicate.addn_zero_filter
 2.05% :     0.000003s :     12: predicate.arithmetic_simplify
 1.34% :     0.000002s :     12: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000002s :     12: predicate.compare_switch_simplify
 1.41% :     0.000002s :     12: predicate.depend_value_elim
 1.40% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.42% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     12: predicate.environ_get_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 5.04% :     0.000007s :     26: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.02% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     17: predicate.load_eliminater
 1.56% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.19% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.99% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000002s :     12: predicate.merge_addn
 1.14% :     0.000002s :     12: predicate.minmaximum_grad
 1.44% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     17: predicate.partial_eliminate
 1.21% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.94% :     0.000003s :     12: predicate.reduce_eliminate
 1.56% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     20: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     12: predicate.reshape_eliminate
 1.25% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     12: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.80% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.66% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.81% :     0.000004s :     20: predicate.switch_defer_inline
 2.64% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.00% :     0.000011s :     65: predicate.switch_simplify
 1.26% :     0.000002s :     12: predicate.tile_eliminate
 1.27% :     0.000002s :     12: predicate.transpose_eliminate
 1.46% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.19% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.69% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.52% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000748    15
47.41% :     0.000355s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.68% :     0.000080s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.91% :     0.000314s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016458    72
 0.22% :     0.000037s :      1: add_recomputation
 0.46% :     0.000076s :      1: auto_monad
 0.07% :     0.000012s :      1: auto_monad_reorder
 1.66% :     0.000273s :      1: bootstrap
 0.11% :     0.000018s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.13% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.16% :     0.000027s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
33.56% :     0.005523s :      1: jit_opt_a
 0.83% :     0.000137s :      1: jit_opt_after_cconv
 0.31% :     0.000050s :      1: jit_opt_b
 2.82% :     0.000464s :      1: loop_unroll
 2.96% :     0.000486s :      1: mutable_eliminate
 6.34% :     0.001044s :     26: opt.transform.jit_opt_a
 0.27% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.23% :     0.000038s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.15% :     0.000518s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.40% :     0.000066s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000010s :      1: remove_dup_value
 1.81% :     0.000298s :      1: renormalize.infer
 1.68% :     0.000277s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000029s :      1: rewriter_after_opt_a
 0.56% :     0.000092s :      1: rewriter_before_opt_a
 0.45% :     0.000074s :      1: symbol_engine_optimizer
40.71% :     0.006700s :      1: type_inference


TotalTime = 0.0159965, [33]
[bootstrap]: 0.00026368
[type_inference]: 0.00396867
[event_method]: 1.582e-05
[auto_monad]: 5.368e-05
[graph_reusing]: 5.14e-06
[pre_auto_parallel]: 2.02001e-06
[py_interpret_to_execute]: 2.938e-05
[rewriter_before_opt_a]: 6.486e-05
[expand_dump_flag]: 2.55002e-06
[jit_opt_a]: 0.00456674, [2]
    [Cycle 1]: 0.00133872, [27]
        [switch_simplify]: 4.469e-05
        [loop_unroll]: 2.966e-05
        [a_1]: 0.00048442
        [with_stream_mark]: 1.362e-05
        [recompute_prepare]: 6.61e-06
        [updatestate_depend_eliminate]: 2.96001e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 6.11e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 5.36998e-06
        [meta_shard_fg_expand]: 1.31002e-06
        [get_grad_eliminate_]: 4.86002e-06
        [merge_forward]: 2.49001e-06
        [cell_reuse_recompute_pass]: 9.5999e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.152e-05
        [j_node_and_user_rematch]: 8.40999e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 8.07e-06
        [inline_without_move]: 5.26002e-06
        [renormalize]: 0.00049561
        [add_forward_monad_depend]: 4.51002e-06
        [auto_monad_grad]: 1.64e-06
        [auto_monad_eliminator]: 9.57999e-06
        [cse]: 1.502e-05
        [replace_applicator]: 1.082e-05
    [Cycle 2]: 0.00030787, [27]
        [switch_simplify]: 5.22e-06
        [loop_unroll]: 4.70999e-06
        [a_1]: 8.778e-05
        [with_stream_mark]: 7.77e-06
        [recompute_prepare]: 4.74002e-06
        [updatestate_depend_eliminate]: 2.43e-06
        [updatestate_assign_eliminate]: 2.08002e-06
        [updatestate_loads_eliminate]: 1.83002e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 4.71002e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 4.54998e-06
        [meta_shard_fg_expand]: 1.02998e-06
        [get_grad_eliminate_]: 4.45e-06
        [merge_forward]: 2.32999e-06
        [cell_reuse_recompute_pass]: 1.61002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.279e-05
        [j_node_and_user_rematch]: 8.67e-06
        [meta_fg_expand]: 1.60001e-06
        [replace_old_param]: 7.3e-06
        [inline_without_move]: 5.22999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.42e-06
        [auto_monad_grad]: 7.99977e-07
        [auto_monad_eliminator]: 5.02e-06
        [cse]: 8.99e-06
        [replace_applicator]: 4.72998e-06
[py_interpret_to_execute_after_opt_a]: 8.20999e-06
[rewriter_after_opt_a]: 2.561e-05
[convert_after_rewriter]: 6.16998e-06
[order_py_execute_after_rewriter]: 4.74e-06
[mutable_eliminate]: 0.00047819
[jit_opt_b]: 5.457e-05, [1]
    [Cycle 1]: 4.82e-05, [2]
        [frontend_op_eliminate]: 1.525e-05
        [inline_after_opt_a]: 1.491e-05
[cconv]: 1.851e-05
[loop_unroll]: 0.00042138
[jit_opt_after_cconv]: 0.00013377, [1]
    [Cycle 1]: 0.00012754, [11]
        [c_1]: 2.012e-05
        [parameter_eliminate]: 2.54001e-06
        [updatestate_depend_eliminate]: 5.27999e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.664e-05
        [call_graph_tuple_transform]: 1.625e-05
        [tuple_list_get_item_eliminator]: 5.44e-06
        [none_parameter_eliminate]: 1.05001e-06
        [renormalize]: 5.29981e-07
        [switch_simplify]: 5.52999e-06
[remove_dup_value]: 8.97e-06
[partial_unused_args_eliminate]: 1.72001e-06
[environ_conv]: 4.13001e-06
[add_recomputation]: 3.403e-05
[cse_after_recomputation]: 1.897e-05, [1]
    [Cycle 1]: 1.387e-05, [1]
        [cse]: 8.45001e-06
[auto_monad_reorder]: 1.07e-05
[get_jit_bprop_graph]: 1.62001e-06
[rewriter_after_jit_bprop_graph]: 4.57998e-06
[opt_after_jit_grad]: 0.00045331
[symbol_engine_optimizer]: 6.92e-05, [1]
    [Cycle 1]: 6.352e-05, [6]
        [build]: 2.68e-06
        [elim_shapecalc]: 7.72002e-06
        [elim_not_effective]: 1.203e-05
        [opt_reshape]: 5.35999e-06
        [fold_const_symbol]: 8.50999e-06
        [renormalize]: 4.30009e-07
[validate]: 2.778e-05
[backend_pass]: 1.02998e-06
[task_emit]: 0.00506608
[execute]: 5.23002e-06
Sums
    bootstrap                            :     0.000264s :  2.11%
    type_inference                       :     0.003969s : 31.69%
    event_method                         :     0.000016s :  0.13%
    auto_monad                           :     0.000054s :  0.43%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.23%
    rewriter_before_opt_a                :     0.000065s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000050s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.27%
    jit_opt_a.a_1                        :     0.000572s :  4.57%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000496s :  3.96%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.12%
    jit_opt_a.cse                        :     0.000024s :  0.19%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.07%
    rewriter_after_opt_a                 :     0.000026s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000478s :  3.82%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000019s :  0.15%
    loop_unroll                          :     0.000421s :  3.36%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000009s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000034s :  0.27%
    cse_after_recomputation.cse          :     0.000008s :  0.07%
    auto_monad_reorder                   :     0.000011s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000453s :  3.62%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005066s : 40.45%
    execute                              :     0.000005s :  0.04%


Time group info:
------[substitution.]   0.000162    23
 1.06% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.43% :     0.000004s :      3: substitution.graph_param_transform
83.89% :     0.000136s :      4: substitution.inline
 1.75% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.56% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.92% :     0.000003s :      2: substitution.replace_old_param
 5.76% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003920     2
81.92% :     0.003211s :      1: type_inference.infer
18.08% :     0.000709s :      1: type_inference.specialize
------[replace.]   0.000060     6
65.86% :     0.000039s :      4: replace.inline
34.14% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000142     6
94.24% :     0.000134s :      4: match.inline
 5.76% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.57% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.19% :     0.000001s :     10: predicate.addn_check_dump
 1.50% :     0.000002s :     10: predicate.addn_zero_filter
 2.05% :     0.000002s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.18% :     0.000001s :     10: predicate.compare_switch_simplify
 1.70% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.09% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.09% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.60% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.30% :     0.000002s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.50% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     10: predicate.reduce_eliminate
 1.67% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.57% :     0.000002s :     10: predicate.reshape_eliminate
 1.38% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.97% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.64% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 2.53% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.57% :     0.000009s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.35% :     0.000002s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.24% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000487    12
47.15% :     0.000230s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.01% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.84% :     0.000228s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017365    76
 0.21% :     0.000037s :      1: add_recomputation
 0.33% :     0.000057s :      1: auto_monad
 0.08% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.58% :     0.000274s :      1: bootstrap
 0.12% :     0.000021s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.05% :     0.000008s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.31% :     0.004569s :      1: jit_opt_a
 0.79% :     0.000136s :      1: jit_opt_after_cconv
 0.33% :     0.000057s :      1: jit_opt_b
 2.47% :     0.000430s :      1: loop_unroll
 2.80% :     0.000486s :      1: mutable_eliminate
 4.41% :     0.000766s :     26: opt.transform.jit_opt_a
 0.25% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.66% :     0.000461s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000011s :      1: remove_dup_value
 1.42% :     0.000247s :      1: renormalize.infer
 1.39% :     0.000242s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000028s :      1: rewriter_after_opt_a
 0.39% :     0.000067s :      1: rewriter_before_opt_a
 0.41% :     0.000072s :      1: symbol_engine_optimizer
29.23% :     0.005075s :      1: task_emit
22.91% :     0.003978s :      1: type_inference
 0.25% :     0.000043s :      1: validate


TotalTime = 0.017366, [33]
[bootstrap]: 0.00024416
[type_inference]: 0.00402591
[event_method]: 1.739e-05
[auto_monad]: 6.156e-05
[graph_reusing]: 5.84e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 3.05e-05
[rewriter_before_opt_a]: 6.996e-05
[expand_dump_flag]: 3.35998e-06
[jit_opt_a]: 0.00501122, [2]
    [Cycle 1]: 0.00151501, [27]
        [switch_simplify]: 5.908e-05
        [loop_unroll]: 3.143e-05
        [a_1]: 0.00051365
        [with_stream_mark]: 1.522e-05
        [recompute_prepare]: 6.88e-06
        [updatestate_depend_eliminate]: 3.21001e-06
        [updatestate_assign_eliminate]: 2.81999e-06
        [updatestate_loads_eliminate]: 2.70997e-06
        [parameter_eliminate]: 1.87001e-06
        [specialize_transform]: 5.69e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.29e-06
        [meta_shard_fg_expand]: 1.86003e-06
        [get_grad_eliminate_]: 5.06997e-06
        [merge_forward]: 3.51001e-06
        [cell_reuse_recompute_pass]: 1.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.787e-05
        [j_node_and_user_rematch]: 8.34998e-06
        [meta_fg_expand]: 2.12001e-06
        [replace_old_param]: 8.07e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.0005831
        [add_forward_monad_depend]: 5.30001e-06
        [auto_monad_grad]: 2.34001e-06
        [auto_monad_eliminator]: 1.329e-05
        [cse]: 2.697e-05
        [replace_applicator]: 1.116e-05
    [Cycle 2]: 0.00030839, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 4.80999e-06
        [a_1]: 8.753e-05
        [with_stream_mark]: 8.54e-06
        [recompute_prepare]: 4.80001e-06
        [updatestate_depend_eliminate]: 2.62001e-06
        [updatestate_assign_eliminate]: 2.15002e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 9.99979e-07
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 5.30001e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 4.47998e-06
        [merge_forward]: 2.71e-06
        [cell_reuse_recompute_pass]: 1.69998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.2e-05
        [j_node_and_user_rematch]: 7.45e-06
        [meta_fg_expand]: 1.55999e-06
        [replace_old_param]: 7.6e-06
        [inline_without_move]: 4.73001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 8.10018e-07
        [auto_monad_eliminator]: 4.83001e-06
        [cse]: 9.41003e-06
        [replace_applicator]: 4.95001e-06
[py_interpret_to_execute_after_opt_a]: 8.13001e-06
[rewriter_after_opt_a]: 2.952e-05
[convert_after_rewriter]: 6.73998e-06
[order_py_execute_after_rewriter]: 4.79e-06
[mutable_eliminate]: 0.00047159
[jit_opt_b]: 4.778e-05, [1]
    [Cycle 1]: 4.071e-05, [2]
        [frontend_op_eliminate]: 1.547e-05
        [inline_after_opt_a]: 1.422e-05
[cconv]: 2.399e-05
[loop_unroll]: 0.0004182
[jit_opt_after_cconv]: 0.00013381, [1]
    [Cycle 1]: 0.00012747, [11]
        [c_1]: 1.997e-05
        [parameter_eliminate]: 2.49001e-06
        [updatestate_depend_eliminate]: 5.19e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.655e-05
        [call_graph_tuple_transform]: 1.794e-05
        [tuple_list_get_item_eliminator]: 5.05001e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.22999e-06
[remove_dup_value]: 1.305e-05
[partial_unused_args_eliminate]: 2.14999e-06
[environ_conv]: 5.34e-06
[add_recomputation]: 4.554e-05
[cse_after_recomputation]: 1.995e-05, [1]
    [Cycle 1]: 1.467e-05, [1]
        [cse]: 8.99e-06
[auto_monad_reorder]: 1.442e-05
[get_jit_bprop_graph]: 1.55999e-06
[rewriter_after_jit_bprop_graph]: 4.99e-06
[opt_after_jit_grad]: 0.00044228
[symbol_engine_optimizer]: 7.067e-05, [1]
    [Cycle 1]: 6.41e-05, [6]
        [build]: 3.18e-06
        [elim_shapecalc]: 7.33e-06
        [elim_not_effective]: 1.326e-05
        [opt_reshape]: 5.09e-06
        [fold_const_symbol]: 8.08001e-06
        [renormalize]: 5.3001e-07
[validate]: 3.024e-05
[backend_pass]: 9.10019e-07
[task_emit]: 0.00591683
[execute]: 6.96001e-06
Sums
    bootstrap                            :     0.000244s :  1.79%
    type_inference                       :     0.004026s : 29.58%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000062s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.22%
    rewriter_before_opt_a                :     0.000070s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.48%
    jit_opt_a.loop_unroll                :     0.000036s :  0.27%
    jit_opt_a.a_1                        :     0.000601s :  4.42%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000583s :  4.28%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000036s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000030s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000472s :  3.46%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000418s :  3.07%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000046s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000442s :  3.25%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000030s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005917s : 43.47%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000185    23
 1.20% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000001s :      2: substitution.fold_const_symbol
 2.85% :     0.000005s :      3: substitution.graph_param_transform
83.07% :     0.000154s :      4: substitution.inline
 1.39% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.44% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.91% :     0.000004s :      2: substitution.replace_old_param
 6.40% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003971     2
80.52% :     0.003197s :      1: type_inference.infer
19.48% :     0.000774s :      1: type_inference.specialize
------[replace.]   0.000063     6
67.23% :     0.000042s :      4: replace.inline
32.77% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     6
93.42% :     0.000152s :      4: match.inline
 6.58% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.28% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 2.10% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.42% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.30% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.63% :     0.000006s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.56% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.54% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.23% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.76% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.51% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.45% :     0.000010s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.48% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000498    12
46.19% :     0.000230s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.67% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.14% :     0.000240s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018869    76
 0.26% :     0.000048s :      1: add_recomputation
 0.34% :     0.000065s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.36% :     0.000257s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.57% :     0.005014s :      1: jit_opt_a
 0.72% :     0.000137s :      1: jit_opt_after_cconv
 0.27% :     0.000050s :      1: jit_opt_b
 2.26% :     0.000426s :      1: loop_unroll
 2.54% :     0.000480s :      1: mutable_eliminate
 4.32% :     0.000815s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.38% :     0.000450s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.66% :     0.000313s :      1: renormalize.infer
 1.39% :     0.000263s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000032s :      1: rewriter_after_opt_a
 0.38% :     0.000073s :      1: rewriter_before_opt_a
 0.39% :     0.000073s :      1: symbol_engine_optimizer
31.42% :     0.005928s :      1: task_emit
21.40% :     0.004038s :      1: type_inference
 0.25% :     0.000047s :      1: validate


TotalTime = 0.013174, [30]
[bootstrap]: 0.00024463
[type_inference]: 0.00584193
[event_method]: 1.983e-05
[auto_monad]: 6.581e-05
[graph_reusing]: 4.79e-06
[pre_auto_parallel]: 2.04999e-06
[py_interpret_to_execute]: 4.707e-05
[rewriter_before_opt_a]: 7.841e-05
[expand_dump_flag]: 2.51e-06
[jit_opt_a]: 0.00493029, [2]
    [Cycle 1]: 0.00157544, [27]
        [switch_simplify]: 8.73e-05
        [loop_unroll]: 3.344e-05
        [a_1]: 0.00062951
        [with_stream_mark]: 1.113e-05
        [recompute_prepare]: 7.25e-06
        [updatestate_depend_eliminate]: 2.71e-06
        [updatestate_assign_eliminate]: 2.18998e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 5.95002e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 5.39998e-06
        [meta_shard_fg_expand]: 1.45999e-06
        [get_grad_eliminate_]: 5.42999e-06
        [merge_forward]: 2.39999e-06
        [cell_reuse_recompute_pass]: 9.89996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.156e-05
        [j_node_and_user_rematch]: 8.77e-06
        [meta_fg_expand]: 1.77001e-06
        [replace_old_param]: 7.34002e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 0.00052318
        [add_forward_monad_depend]: 4.70001e-06
        [auto_monad_grad]: 1.40999e-06
        [auto_monad_eliminator]: 1.11e-05
        [cse]: 1.765e-05
        [replace_applicator]: 1.131e-05
    [Cycle 2]: 0.00032145, [27]
        [switch_simplify]: 5.37999e-06
        [loop_unroll]: 5.11002e-06
        [a_1]: 9.159e-05
        [with_stream_mark]: 8.17998e-06
        [recompute_prepare]: 4.95001e-06
        [updatestate_depend_eliminate]: 2.68e-06
        [updatestate_assign_eliminate]: 2.03002e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 9.09989e-07
        [specialize_transform]: 4.98001e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.00999e-06
        [meta_shard_fg_expand]: 1.19e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.31e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.969e-05
        [j_node_and_user_rematch]: 7.64002e-06
        [meta_fg_expand]: 1.50999e-06
        [replace_old_param]: 6.99001e-06
        [inline_without_move]: 4.73001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 4.94998e-06
        [cse]: 1.004e-05
        [replace_applicator]: 5.29e-06
[py_interpret_to_execute_after_opt_a]: 8.99998e-06
[rewriter_after_opt_a]: 2.719e-05
[convert_after_rewriter]: 5.37001e-06
[order_py_execute_after_rewriter]: 4.67998e-06
[mutable_eliminate]: 0.00047863
[jit_opt_b]: 4.751e-05, [1]
    [Cycle 1]: 4.157e-05, [2]
        [frontend_op_eliminate]: 1.536e-05
        [inline_after_opt_a]: 1.416e-05
[cconv]: 1.539e-05
[loop_unroll]: 0.00042162
[jit_opt_after_cconv]: 0.00013229, [1]
    [Cycle 1]: 0.00012629, [11]
        [c_1]: 2.1e-05
        [parameter_eliminate]: 2.50002e-06
        [updatestate_depend_eliminate]: 5.47001e-06
        [updatestate_assign_eliminate]: 2.23002e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.58e-05
        [call_graph_tuple_transform]: 1.581e-05
        [tuple_list_get_item_eliminator]: 5.07999e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 5.37999e-06
[remove_dup_value]: 6.83998e-06
[partial_unused_args_eliminate]: 1.60999e-06
[environ_conv]: 4.22e-06
[add_recomputation]: 3.328e-05
[cse_after_recomputation]: 1.92e-05, [1]
    [Cycle 1]: 1.397e-05, [1]
        [cse]: 8.70999e-06
[auto_monad_reorder]: 8.97999e-06
[get_jit_bprop_graph]: 1.44e-06
[rewriter_after_jit_bprop_graph]: 4.36002e-06
[opt_after_jit_grad]: 0.00046702
[symbol_engine_optimizer]: 7.088e-05, [1]
    [Cycle 1]: 6.444e-05, [6]
        [build]: 2.60002e-06
        [elim_shapecalc]: 7.33e-06
        [elim_not_effective]: 1.231e-05
        [opt_reshape]: 5.76e-06
        [fold_const_symbol]: 8.23999e-06
        [renormalize]: 4.89992e-07
[validate]: 2.44e-05
Sums
    bootstrap                            :     0.000245s :  2.55%
    type_inference                       :     0.005842s : 60.91%
    event_method                         :     0.000020s :  0.21%
    auto_monad                           :     0.000066s :  0.69%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000047s :  0.49%
    rewriter_before_opt_a                :     0.000078s :  0.82%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000093s :  0.97%
    jit_opt_a.loop_unroll                :     0.000039s :  0.40%
    jit_opt_a.a_1                        :     0.000721s :  7.52%
    jit_opt_a.with_stream_mark           :     0.000019s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000523s :  5.46%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.17%
    jit_opt_a.cse                        :     0.000028s :  0.29%
    jit_opt_a.replace_applicator         :     0.000017s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.09%
    rewriter_after_opt_a                 :     0.000027s :  0.28%
    convert_after_rewriter               :     0.000005s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000479s :  4.99%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.15%
    cconv                                :     0.000015s :  0.16%
    loop_unroll                          :     0.000422s :  4.40%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000007s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000033s :  0.35%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000009s :  0.09%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000467s :  4.87%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000024s :  0.25%


Time group info:
------[substitution.]   0.000221    26
 0.69% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 1.52% :     0.000003s :      3: substitution.graph_param_transform
82.23% :     0.000182s :      6: substitution.inline
 1.24% :     0.000003s :      4: substitution.j_node_and_user_rematch
 5.42% :     0.000012s :      4: substitution.remove_not_recompute_node
 1.31% :     0.000003s :      2: substitution.replace_old_param
 3.38% :     0.000007s :      1: substitution.switch_simplify
 3.76% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005790     2
83.43% :     0.004831s :      1: type_inference.infer
16.57% :     0.000959s :      1: type_inference.specialize
------[replace.]   0.000113     9
56.10% :     0.000063s :      6: replace.inline
18.10% :     0.000020s :      1: replace.switch_simplify
25.80% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000193     9
92.78% :     0.000179s :      6: match.inline
 3.49% :     0.000007s :      1: match.switch_simplify
 3.74% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   801
 1.32% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.18% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.21% :     0.000002s :     12: predicate.addn_check_dump
 1.42% :     0.000002s :     12: predicate.addn_zero_filter
 1.84% :     0.000002s :     12: predicate.arithmetic_simplify
 1.38% :     0.000002s :     12: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     12: predicate.compare_switch_simplify
 1.52% :     0.000002s :     12: predicate.depend_value_elim
 1.26% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.74% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     12: predicate.environ_get_eliminate
 1.43% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.31% :     0.000007s :     26: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.89% :     0.000001s :      6: predicate.less_batch_normalization
 1.76% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     17: predicate.load_eliminater
 1.14% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000002s :     12: predicate.merge_addn
 1.22% :     0.000002s :     12: predicate.minmaximum_grad
 1.56% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     17: predicate.partial_eliminate
 1.26% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.82% :     0.000002s :     12: predicate.reduce_eliminate
 1.60% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     20: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     12: predicate.reshape_eliminate
 1.27% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.45% :     0.000002s :     12: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.72% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.75% :     0.000004s :     20: predicate.switch_defer_inline
 2.44% :     0.000003s :     20: predicate.switch_layer_defer_inline
 7.89% :     0.000010s :     65: predicate.switch_simplify
 1.25% :     0.000002s :     12: predicate.tile_eliminate
 1.42% :     0.000002s :     12: predicate.transpose_eliminate
 1.45% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.43% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     12: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000692    15
49.10% :     0.000340s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 9.67% :     0.000067s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.24% :     0.000285s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014734    72
 0.25% :     0.000036s :      1: add_recomputation
 0.47% :     0.000069s :      1: auto_monad
 0.08% :     0.000011s :      1: auto_monad_reorder
 1.72% :     0.000253s :      1: bootstrap
 0.12% :     0.000018s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
33.48% :     0.004933s :      1: jit_opt_a
 0.92% :     0.000135s :      1: jit_opt_after_cconv
 0.34% :     0.000050s :      1: jit_opt_b
 2.92% :     0.000430s :      1: loop_unroll
 3.30% :     0.000487s :      1: mutable_eliminate
 6.59% :     0.000971s :     26: opt.transform.jit_opt_a
 0.30% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.22% :     0.000475s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.34% :     0.000050s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000009s :      1: remove_dup_value
 1.70% :     0.000250s :      1: renormalize.infer
 1.81% :     0.000267s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000030s :      1: rewriter_after_opt_a
 0.55% :     0.000081s :      1: rewriter_before_opt_a
 0.50% :     0.000073s :      1: symbol_engine_optimizer
39.73% :     0.005854s :      1: type_inference


TotalTime = 0.0153001, [33]
[bootstrap]: 0.00025694
[type_inference]: 0.00388855
[event_method]: 1.393e-05
[auto_monad]: 4.136e-05
[graph_reusing]: 3.4e-06
[pre_auto_parallel]: 1.79e-06
[py_interpret_to_execute]: 2.803e-05
[rewriter_before_opt_a]: 6.045e-05
[expand_dump_flag]: 2.10002e-06
[jit_opt_a]: 0.00437081, [2]
    [Cycle 1]: 0.00121256, [27]
        [switch_simplify]: 3.959e-05
        [loop_unroll]: 2.802e-05
        [a_1]: 0.00045139
        [with_stream_mark]: 1.003e-05
        [recompute_prepare]: 6.61e-06
        [updatestate_depend_eliminate]: 2.61999e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 5.61e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 5.08002e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.97999e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.137e-05
        [j_node_and_user_rematch]: 8.39998e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 7.23999e-06
        [inline_without_move]: 4.97e-06
        [renormalize]: 0.00042606
        [add_forward_monad_depend]: 3.95998e-06
        [auto_monad_grad]: 1.14003e-06
        [auto_monad_eliminator]: 8.45999e-06
        [cse]: 1.482e-05
        [replace_applicator]: 1.04e-05
    [Cycle 2]: 0.00030363, [27]
        [switch_simplify]: 5.44998e-06
        [loop_unroll]: 4.75999e-06
        [a_1]: 8.91e-05
        [with_stream_mark]: 8.15e-06
        [recompute_prepare]: 4.90999e-06
        [updatestate_depend_eliminate]: 2.39999e-06
        [updatestate_assign_eliminate]: 2.01998e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 4.81002e-06
        [updatestate_useless_node_eliminater]: 4.46002e-06
        [accelerated_algorithm]: 4.58001e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 4.63001e-06
        [merge_forward]: 2.20002e-06
        [cell_reuse_recompute_pass]: 1.52999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.137e-05
        [j_node_and_user_rematch]: 7.38e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 6.73e-06
        [inline_without_move]: 4.47998e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 4.56002e-06
        [cse]: 9.44e-06
        [replace_applicator]: 5.10001e-06
[py_interpret_to_execute_after_opt_a]: 7.39002e-06
[rewriter_after_opt_a]: 2.274e-05
[convert_after_rewriter]: 5.15001e-06
[order_py_execute_after_rewriter]: 3.7e-06
[mutable_eliminate]: 0.00047226
[jit_opt_b]: 4.631e-05, [1]
    [Cycle 1]: 4.03e-05, [2]
        [frontend_op_eliminate]: 1.493e-05
        [inline_after_opt_a]: 1.41e-05
[cconv]: 1.49e-05
[loop_unroll]: 0.00042849
[jit_opt_after_cconv]: 0.00012692, [1]
    [Cycle 1]: 0.00012104, [11]
        [c_1]: 1.937e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 4.37998e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.58e-05
        [call_graph_tuple_transform]: 1.524e-05
        [tuple_list_get_item_eliminator]: 5.19998e-06
        [none_parameter_eliminate]: 7.59988e-07
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.27999e-06
[remove_dup_value]: 6.32001e-06
[partial_unused_args_eliminate]: 1.49e-06
[environ_conv]: 3.35e-06
[add_recomputation]: 2.839e-05
[cse_after_recomputation]: 1.983e-05, [1]
    [Cycle 1]: 1.496e-05, [1]
        [cse]: 8.94e-06
[auto_monad_reorder]: 8.93002e-06
[get_jit_bprop_graph]: 1.52001e-06
[rewriter_after_jit_bprop_graph]: 4.56002e-06
[opt_after_jit_grad]: 0.00045756
[symbol_engine_optimizer]: 6.529e-05, [1]
    [Cycle 1]: 5.986e-05, [6]
        [build]: 2.00002e-06
        [elim_shapecalc]: 6.72002e-06
        [elim_not_effective]: 1.117e-05
        [opt_reshape]: 5.27999e-06
        [fold_const_symbol]: 7.85e-06
        [renormalize]: 6.00005e-07
[validate]: 2.3e-05
[backend_pass]: 8.09989e-07
[task_emit]: 0.00472463
[execute]: 4.89998e-06
Sums
    bootstrap                            :     0.000257s :  2.15%
    type_inference                       :     0.003889s : 32.61%
    event_method                         :     0.000014s :  0.12%
    auto_monad                           :     0.000041s :  0.35%
    graph_reusing                        :     0.000003s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.24%
    rewriter_before_opt_a                :     0.000060s :  0.51%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000045s :  0.38%
    jit_opt_a.loop_unroll                :     0.000033s :  0.27%
    jit_opt_a.a_1                        :     0.000540s :  4.53%
    jit_opt_a.with_stream_mark           :     0.000018s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.12%
    jit_opt_a.inline_without_move        :     0.000009s :  0.08%
    jit_opt_a.renormalize                :     0.000426s :  3.57%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.11%
    jit_opt_a.cse                        :     0.000024s :  0.20%
    jit_opt_a.replace_applicator         :     0.000016s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.06%
    rewriter_after_opt_a                 :     0.000023s :  0.19%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000472s :  3.96%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.12%
    cconv                                :     0.000015s :  0.12%
    loop_unroll                          :     0.000428s :  3.59%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000006s :  0.05%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000003s :  0.03%
    add_recomputation                    :     0.000028s :  0.24%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000009s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000458s :  3.84%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000023s :  0.19%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.004725s : 39.62%
    execute                              :     0.000005s :  0.04%


Time group info:
------[substitution.]   0.000140    23
 0.97% :     0.000001s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.51% :     0.000004s :      3: substitution.graph_param_transform
84.19% :     0.000118s :      4: substitution.inline
 1.90% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.70% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.74% :     0.000002s :      2: substitution.replace_old_param
 5.36% :     0.000007s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003849     2
82.02% :     0.003157s :      1: type_inference.infer
17.98% :     0.000692s :      1: type_inference.specialize
------[replace.]   0.000058     6
66.23% :     0.000039s :      4: replace.inline
33.77% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000122     6
94.77% :     0.000115s :      4: match.inline
 5.23% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.36% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.36% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.22% :     0.000001s :     10: predicate.addn_check_dump
 1.54% :     0.000002s :     10: predicate.addn_zero_filter
 1.82% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.30% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.22% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.33% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.32% :     0.000001s :     10: predicate.environ_get_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 4.80% :     0.000005s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.02% :     0.000002s :     15: predicate.load_eliminater
 1.34% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.00% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.93% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.35% :     0.000001s :     10: predicate.merge_addn
 1.21% :     0.000001s :     10: predicate.minmaximum_grad
 1.29% :     0.000001s :      3: predicate.mutable_eliminate
 0.44% :     0.000000s :      3: predicate.opt_reshape
 2.23% :     0.000002s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.65% :     0.000002s :     10: predicate.reshape_eliminate
 1.34% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 2.25% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.35% :     0.000008s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.42% :     0.000002s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.42% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.77% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.30% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.84% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000469    12
48.06% :     0.000226s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.33% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.60% :     0.000219s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016550    76
 0.19% :     0.000031s :      1: add_recomputation
 0.27% :     0.000044s :      1: auto_monad
 0.07% :     0.000011s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.60% :     0.000265s :      1: bootstrap
 0.11% :     0.000018s :      1: cconv
 0.04% :     0.000007s :      1: convert_after_rewriter
 0.13% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000005s :      1: environ_conv
 0.10% :     0.000017s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.03% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000005s :      1: graph_reusing
26.43% :     0.004373s :      1: jit_opt_a
 0.78% :     0.000129s :      1: jit_opt_after_cconv
 0.30% :     0.000049s :      1: jit_opt_b
 2.63% :     0.000436s :      1: loop_unroll
 2.90% :     0.000479s :      1: mutable_eliminate
 4.37% :     0.000724s :     26: opt.transform.jit_opt_a
 0.25% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000023s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000028s :      4: opt.transform.symbol_engine_opt
 2.81% :     0.000465s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000031s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000009s :      1: remove_dup_value
 1.29% :     0.000214s :      1: renormalize.infer
 1.25% :     0.000206s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000025s :      1: rewriter_after_opt_a
 0.38% :     0.000063s :      1: rewriter_before_opt_a
 0.41% :     0.000068s :      1: symbol_engine_optimizer
28.61% :     0.004734s :      1: task_emit
23.54% :     0.003896s :      1: type_inference
 0.23% :     0.000038s :      1: validate


TotalTime = 0.0285515, [33]
[bootstrap]: 0.00035381
[type_inference]: 0.0113483
[event_method]: 1.928e-05
[auto_monad]: 0.00010162
[graph_reusing]: 6.81999e-06
[pre_auto_parallel]: 2.88998e-06
[py_interpret_to_execute]: 4.213e-05
[rewriter_before_opt_a]: 7.668e-05
[expand_dump_flag]: 3.03998e-06
[jit_opt_a]: 0.00702967, [2]
    [Cycle 1]: 0.0021624, [27]
        [switch_simplify]: 7.368e-05
        [loop_unroll]: 5.031e-05
        [a_1]: 0.00060414
        [with_stream_mark]: 2.452e-05
        [recompute_prepare]: 8.32998e-06
        [updatestate_depend_eliminate]: 3.73001e-06
        [updatestate_assign_eliminate]: 3.66001e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [parameter_eliminate]: 2.41998e-06
        [specialize_transform]: 6.43998e-06
        [updatestate_useless_node_eliminater]: 5.57999e-06
        [accelerated_algorithm]: 6.07999e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 5.91e-06
        [merge_forward]: 3.52002e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.482e-05
        [j_node_and_user_rematch]: 9.20999e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 9.95002e-06
        [inline_without_move]: 5.32999e-06
        [renormalize]: 0.00106839
        [add_forward_monad_depend]: 6.09999e-06
        [auto_monad_grad]: 2.44001e-06
        [auto_monad_eliminator]: 1.669e-05
        [cse]: 3.17e-05
        [replace_applicator]: 1.39e-05
    [Cycle 2]: 0.00033923, [27]
        [switch_simplify]: 6.10002e-06
        [loop_unroll]: 4.82e-06
        [a_1]: 9.331e-05
        [with_stream_mark]: 1.143e-05
        [recompute_prepare]: 5.51e-06
        [updatestate_depend_eliminate]: 3.55e-06
        [updatestate_assign_eliminate]: 2.31998e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.07998e-06
        [specialize_transform]: 5.24e-06
        [updatestate_useless_node_eliminater]: 5.30999e-06
        [accelerated_algorithm]: 5.56e-06
        [meta_shard_fg_expand]: 1.45001e-06
        [get_grad_eliminate_]: 4.84e-06
        [merge_forward]: 2.98e-06
        [cell_reuse_recompute_pass]: 2.44001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.46e-05
        [j_node_and_user_rematch]: 7.62002e-06
        [meta_fg_expand]: 1.70001e-06
        [replace_old_param]: 7.31001e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 5.91998e-06
        [cse]: 1.152e-05
        [replace_applicator]: 5.20999e-06
[py_interpret_to_execute_after_opt_a]: 1.197e-05
[rewriter_after_opt_a]: 4.299e-05
[convert_after_rewriter]: 7.12002e-06
[order_py_execute_after_rewriter]: 5.79999e-06
[mutable_eliminate]: 0.00073308
[jit_opt_b]: 5.068e-05, [1]
    [Cycle 1]: 4.329e-05, [2]
        [frontend_op_eliminate]: 1.654e-05
        [inline_after_opt_a]: 1.586e-05
[cconv]: 2.714e-05
[loop_unroll]: 0.00042525
[jit_opt_after_cconv]: 0.00014344, [1]
    [Cycle 1]: 0.00013711, [11]
        [c_1]: 2.118e-05
        [parameter_eliminate]: 3.01999e-06
        [updatestate_depend_eliminate]: 5.99e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [cse]: 1.91e-05
        [call_graph_tuple_transform]: 1.883e-05
        [tuple_list_get_item_eliminator]: 5.76003e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 5.62001e-06
[remove_dup_value]: 1.492e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 5.12999e-06
[add_recomputation]: 5.602e-05
[cse_after_recomputation]: 2.1e-05, [1]
    [Cycle 1]: 1.556e-05, [1]
        [cse]: 9.17999e-06
[auto_monad_reorder]: 1.572e-05
[get_jit_bprop_graph]: 1.71e-06
[rewriter_after_jit_bprop_graph]: 5.54998e-06
[opt_after_jit_grad]: 0.00046264
[symbol_engine_optimizer]: 7.515e-05, [1]
    [Cycle 1]: 6.769e-05, [6]
        [build]: 3.38e-06
        [elim_shapecalc]: 7.48999e-06
        [elim_not_effective]: 1.265e-05
        [opt_reshape]: 6.21e-06
        [fold_const_symbol]: 8.74998e-06
        [renormalize]: 8.50006e-07
[validate]: 4.604e-05
[backend_pass]: 1.00001e-06
[task_emit]: 0.00717809
[execute]: 8.74e-06
Sums
    bootstrap                            :     0.000354s :  1.51%
    type_inference                       :     0.011348s : 48.54%
    event_method                         :     0.000019s :  0.08%
    auto_monad                           :     0.000102s :  0.43%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.18%
    rewriter_before_opt_a                :     0.000077s :  0.33%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000080s :  0.34%
    jit_opt_a.loop_unroll                :     0.000055s :  0.24%
    jit_opt_a.a_1                        :     0.000697s :  2.98%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.07%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.001068s :  4.57%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.10%
    jit_opt_a.cse                        :     0.000043s :  0.18%
    jit_opt_a.replace_applicator         :     0.000019s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.05%
    rewriter_after_opt_a                 :     0.000043s :  0.18%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000733s :  3.14%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.07%
    cconv                                :     0.000027s :  0.12%
    loop_unroll                          :     0.000425s :  1.82%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000015s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000056s :  0.24%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000463s :  1.98%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.20%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007178s : 30.71%
    execute                              :     0.000009s :  0.04%


Time group info:
------[substitution.]   0.000255    23
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 2.19% :     0.000006s :      3: substitution.graph_param_transform
86.21% :     0.000220s :      4: substitution.inline
 1.34% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.99% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.84% :     0.000005s :      2: substitution.replace_old_param
 5.18% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011220     2
91.23% :     0.010236s :      1: type_inference.infer
 8.77% :     0.000984s :      1: type_inference.specialize
------[replace.]   0.000082     6
67.25% :     0.000055s :      4: replace.inline
32.75% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000229     6
94.74% :     0.000217s :      4: match.inline
 5.26% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.28% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.49% :     0.000002s :     10: predicate.addn_zero_filter
 2.23% :     0.000003s :     10: predicate.arithmetic_simplify
 1.14% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.19% :     0.000001s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.11% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.29% :     0.000002s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.29% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.62% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.29% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.39% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.15% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.50% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.83% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000001s :     10: predicate.same_eliminate
 0.77% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.64% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.97% :     0.000009s :     53: predicate.switch_simplify
 1.31% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.24% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.19% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.54% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000789    12
49.95% :     0.000394s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.32% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.73% :     0.000345s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.030669    76
 0.19% :     0.000059s :      1: add_recomputation
 0.35% :     0.000106s :      1: auto_monad
 0.06% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.23% :     0.000376s :      1: bootstrap
 0.10% :     0.000030s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000024s :      1: event_method
 0.04% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
22.93% :     0.007033s :      1: jit_opt_a
 0.48% :     0.000147s :      1: jit_opt_after_cconv
 0.17% :     0.000053s :      1: jit_opt_b
 1.41% :     0.000434s :      1: loop_unroll
 2.42% :     0.000743s :      1: mutable_eliminate
 3.11% :     0.000954s :     26: opt.transform.jit_opt_a
 0.16% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000026s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.54% :     0.000471s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.15% :     0.000045s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 2.01% :     0.000616s :      1: renormalize.infer
 1.44% :     0.000443s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000049s :      1: rewriter_after_opt_a
 0.26% :     0.000080s :      1: rewriter_before_opt_a
 0.25% :     0.000078s :      1: symbol_engine_optimizer
23.45% :     0.007193s :      1: task_emit
37.06% :     0.011367s :      1: type_inference
 0.21% :     0.000066s :      1: validate


TotalTime = 0.013421, [30]
[bootstrap]: 0.00030176
[type_inference]: 0.00619915
[event_method]: 1.813e-05
[auto_monad]: 5.765e-05
[graph_reusing]: 3.46999e-06
[pre_auto_parallel]: 2.04999e-06
[py_interpret_to_execute]: 4.284e-05
[rewriter_before_opt_a]: 7.022e-05
[expand_dump_flag]: 1.336e-05
[jit_opt_a]: 0.00485489, [2]
    [Cycle 1]: 0.00153144, [27]
        [switch_simplify]: 0.00010734
        [loop_unroll]: 3.352e-05
        [a_1]: 0.00060294
        [with_stream_mark]: 9.86998e-06
        [recompute_prepare]: 7.08998e-06
        [updatestate_depend_eliminate]: 2.68e-06
        [updatestate_assign_eliminate]: 2.14999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 5.72001e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 5.17e-06
        [meta_shard_fg_expand]: 1.22999e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 2.29999e-06
        [cell_reuse_recompute_pass]: 8.89995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.04e-05
        [j_node_and_user_rematch]: 8.26002e-06
        [meta_fg_expand]: 1.64e-06
        [replace_old_param]: 7.35998e-06
        [inline_without_move]: 4.99998e-06
        [renormalize]: 0.00052098
        [add_forward_monad_depend]: 3.14999e-06
        [auto_monad_grad]: 1.00999e-06
        [auto_monad_eliminator]: 7.71001e-06
        [cse]: 1.23e-05
        [replace_applicator]: 8.54e-06
    [Cycle 2]: 0.00029789, [27]
        [switch_simplify]: 5.12999e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 8.87e-05
        [with_stream_mark]: 6.66e-06
        [recompute_prepare]: 4.87e-06
        [updatestate_depend_eliminate]: 2.27001e-06
        [updatestate_assign_eliminate]: 1.90001e-06
        [updatestate_loads_eliminate]: 1.80001e-06
        [parameter_eliminate]: 6.90023e-07
        [specialize_transform]: 4.62998e-06
        [updatestate_useless_node_eliminater]: 4.72998e-06
        [accelerated_algorithm]: 4.58001e-06
        [meta_shard_fg_expand]: 8.79983e-07
        [get_grad_eliminate_]: 4.63001e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 9.30013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.046e-05
        [j_node_and_user_rematch]: 7.6e-06
        [meta_fg_expand]: 1.39998e-06
        [replace_old_param]: 6.41e-06
        [inline_without_move]: 4.87998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 9.39996e-07
        [auto_monad_grad]: 5.8001e-07
        [auto_monad_eliminator]: 4.18001e-06
        [cse]: 8.03999e-06
        [replace_applicator]: 4.57e-06
[py_interpret_to_execute_after_opt_a]: 6.39999e-06
[rewriter_after_opt_a]: 2.175e-05
[convert_after_rewriter]: 5.25001e-06
[order_py_execute_after_rewriter]: 3.93001e-06
[mutable_eliminate]: 0.00043659
[jit_opt_b]: 4.442e-05, [1]
    [Cycle 1]: 3.935e-05, [2]
        [frontend_op_eliminate]: 1.488e-05
        [inline_after_opt_a]: 1.352e-05
[cconv]: 1.108e-05
[loop_unroll]: 0.00039014
[jit_opt_after_cconv]: 0.00013184, [1]
    [Cycle 1]: 0.00012609, [11]
        [c_1]: 2.058e-05
        [parameter_eliminate]: 1.50001e-06
        [updatestate_depend_eliminate]: 4.18999e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [cse]: 1.251e-05
        [call_graph_tuple_transform]: 1.442e-05
        [tuple_list_get_item_eliminator]: 5.19e-06
        [none_parameter_eliminate]: 8.89995e-07
        [renormalize]: 4.00003e-07
        [switch_simplify]: 5.40999e-06
[remove_dup_value]: 1.345e-05
[partial_unused_args_eliminate]: 2.58e-06
[environ_conv]: 4.47e-06
[add_recomputation]: 4.543e-05
[cse_after_recomputation]: 1.896e-05, [1]
    [Cycle 1]: 1.382e-05, [1]
        [cse]: 8.42e-06
[auto_monad_reorder]: 1.349e-05
[get_jit_bprop_graph]: 1.21002e-06
[rewriter_after_jit_bprop_graph]: 5.00001e-06
[opt_after_jit_grad]: 0.00044554
[symbol_engine_optimizer]: 6.908e-05, [1]
    [Cycle 1]: 6.322e-05, [6]
        [build]: 2.98e-06
        [elim_shapecalc]: 6.93998e-06
        [elim_not_effective]: 1.169e-05
        [opt_reshape]: 5.54e-06
        [fold_const_symbol]: 8.05e-06
        [renormalize]: 5.40022e-07
[validate]: 3.023e-05
Sums
    bootstrap                            :     0.000302s :  3.06%
    type_inference                       :     0.006199s : 62.90%
    event_method                         :     0.000018s :  0.18%
    auto_monad                           :     0.000058s :  0.58%
    graph_reusing                        :     0.000003s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.43%
    rewriter_before_opt_a                :     0.000070s :  0.71%
    expand_dump_flag                     :     0.000013s :  0.14%
    jit_opt_a.switch_simplify            :     0.000112s :  1.14%
    jit_opt_a.loop_unroll                :     0.000038s :  0.39%
    jit_opt_a.a_1                        :     0.000692s :  7.02%
    jit_opt_a.with_stream_mark           :     0.000017s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000021s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000521s :  5.29%
    jit_opt_a.add_forward_monad_depend   :     0.000004s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000012s :  0.12%
    jit_opt_a.cse                        :     0.000020s :  0.21%
    jit_opt_a.replace_applicator         :     0.000013s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000006s :  0.06%
    rewriter_after_opt_a                 :     0.000022s :  0.22%
    convert_after_rewriter               :     0.000005s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000437s :  4.43%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.14%
    cconv                                :     0.000011s :  0.11%
    loop_unroll                          :     0.000390s :  3.96%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000013s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.14%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000045s :  0.46%
    cse_after_recomputation.cse          :     0.000008s :  0.09%
    auto_monad_reorder                   :     0.000013s :  0.14%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000446s :  4.52%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000030s :  0.31%


Time group info:
------[substitution.]   0.000205    26
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 1.54% :     0.000003s :      3: substitution.graph_param_transform
85.00% :     0.000174s :      6: substitution.inline
 1.24% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.07% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.20% :     0.000002s :      2: substitution.replace_old_param
 4.07% :     0.000008s :      1: substitution.switch_simplify
 3.48% :     0.000007s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006151     2
84.48% :     0.005196s :      1: type_inference.infer
15.52% :     0.000955s :      1: type_inference.specialize
------[replace.]   0.000121     9
49.18% :     0.000059s :      6: replace.inline
26.38% :     0.000032s :      1: replace.switch_simplify
24.44% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000184     9
92.66% :     0.000171s :      6: match.inline
 4.08% :     0.000008s :      1: match.switch_simplify
 3.25% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   801
 1.53% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.11% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000001s :     12: predicate.addn_check_dump
 1.43% :     0.000002s :     12: predicate.addn_zero_filter
 1.87% :     0.000002s :     12: predicate.arithmetic_simplify
 1.41% :     0.000002s :     12: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.20% :     0.000001s :     12: predicate.compare_switch_simplify
 1.58% :     0.000002s :     12: predicate.depend_value_elim
 1.22% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.25% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.24% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.23% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     12: predicate.environ_get_eliminate
 1.24% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.83% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.08% :     0.000006s :     26: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.98% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     17: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.55% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.29% :     0.000002s :     12: predicate.merge_addn
 1.21% :     0.000002s :     12: predicate.minmaximum_grad
 1.44% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000003s :     17: predicate.partial_eliminate
 1.31% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     12: predicate.reduce_eliminate
 1.49% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     20: predicate.replace_applicator
 0.43% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     12: predicate.reshape_eliminate
 1.27% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.57% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.84% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.58% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.88% :     0.000004s :     20: predicate.switch_defer_inline
 2.38% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.42% :     0.000010s :     65: predicate.switch_simplify
 1.26% :     0.000002s :     12: predicate.tile_eliminate
 1.49% :     0.000002s :     12: predicate.transpose_eliminate
 1.47% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.44% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.27% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.78% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     12: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000715    15
52.52% :     0.000376s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 9.78% :     0.000070s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.70% :     0.000270s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014938    72
 0.32% :     0.000048s :      1: add_recomputation
 0.41% :     0.000061s :      1: auto_monad
 0.11% :     0.000016s :      1: auto_monad_reorder
 2.14% :     0.000320s :      1: bootstrap
 0.09% :     0.000014s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.14% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.14% :     0.000021s :      1: event_method
 0.11% :     0.000016s :      1: expand_dump_flag
 0.02% :     0.000003s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
32.52% :     0.004858s :      1: jit_opt_a
 0.90% :     0.000134s :      1: jit_opt_after_cconv
 0.31% :     0.000047s :      1: jit_opt_b
 2.66% :     0.000397s :      1: loop_unroll
 2.96% :     0.000443s :      1: mutable_eliminate
 6.32% :     0.000944s :     26: opt.transform.jit_opt_a
 0.28% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000023s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000029s :      4: opt.transform.symbol_engine_opt
 3.03% :     0.000453s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.30% :     0.000046s :      1: py_interpret_to_execute
 0.06% :     0.000009s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000016s :      1: remove_dup_value
 1.90% :     0.000284s :      1: renormalize.infer
 1.55% :     0.000231s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000025s :      1: rewriter_after_opt_a
 0.49% :     0.000073s :      1: rewriter_before_opt_a
 0.48% :     0.000071s :      1: symbol_engine_optimizer
41.56% :     0.006208s :      1: type_inference


TotalTime = 0.01683, [33]
[bootstrap]: 0.00024891
[type_inference]: 0.00393207
[event_method]: 1.62e-05
[auto_monad]: 5.777e-05
[graph_reusing]: 5.71e-06
[pre_auto_parallel]: 2.74999e-06
[py_interpret_to_execute]: 3.19e-05
[rewriter_before_opt_a]: 7.077e-05
[expand_dump_flag]: 3.7e-06
[jit_opt_a]: 0.00478804, [2]
    [Cycle 1]: 0.00142448, [27]
        [switch_simplify]: 5.011e-05
        [loop_unroll]: 2.79e-05
        [a_1]: 0.00047497
        [with_stream_mark]: 1.574e-05
        [recompute_prepare]: 6.54001e-06
        [updatestate_depend_eliminate]: 3.51999e-06
        [updatestate_assign_eliminate]: 3.27002e-06
        [updatestate_loads_eliminate]: 2.82002e-06
        [parameter_eliminate]: 1.83002e-06
        [specialize_transform]: 5.56e-06
        [updatestate_useless_node_eliminater]: 4.79998e-06
        [accelerated_algorithm]: 5.33002e-06
        [meta_shard_fg_expand]: 1.84998e-06
        [get_grad_eliminate_]: 5.04998e-06
        [merge_forward]: 3.81001e-06
        [cell_reuse_recompute_pass]: 1.18001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.33e-05
        [j_node_and_user_rematch]: 8.47e-06
        [meta_fg_expand]: 1.98997e-06
        [replace_old_param]: 8.40999e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.00055437
        [add_forward_monad_depend]: 5.20999e-06
        [auto_monad_grad]: 2.07001e-06
        [auto_monad_eliminator]: 1.22e-05
        [cse]: 2.809e-05
        [replace_applicator]: 1.132e-05
    [Cycle 2]: 0.00030806, [27]
        [switch_simplify]: 5.40001e-06
        [loop_unroll]: 4.70999e-06
        [a_1]: 8.828e-05
        [with_stream_mark]: 7.91001e-06
        [recompute_prepare]: 4.70001e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 2.11998e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 4.81002e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 1.03001e-06
        [get_grad_eliminate_]: 4.36002e-06
        [merge_forward]: 2.29999e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.185e-05
        [j_node_and_user_rematch]: 7.7e-06
        [meta_fg_expand]: 1.39e-06
        [replace_old_param]: 6.57002e-06
        [inline_without_move]: 4.33001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.00999e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 5.22999e-06
        [cse]: 8.90999e-06
        [replace_applicator]: 4.80999e-06
[py_interpret_to_execute_after_opt_a]: 7.83001e-06
[rewriter_after_opt_a]: 3.086e-05
[convert_after_rewriter]: 6.47001e-06
[order_py_execute_after_rewriter]: 4.63999e-06
[mutable_eliminate]: 0.0004839
[jit_opt_b]: 4.696e-05, [1]
    [Cycle 1]: 4.04e-05, [2]
        [frontend_op_eliminate]: 1.526e-05
        [inline_after_opt_a]: 1.457e-05
[cconv]: 2.391e-05
[loop_unroll]: 0.00040767
[jit_opt_after_cconv]: 0.00013199, [1]
    [Cycle 1]: 0.00012573, [11]
        [c_1]: 1.91e-05
        [parameter_eliminate]: 2.64999e-06
        [updatestate_depend_eliminate]: 5.89e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [cse]: 1.595e-05
        [call_graph_tuple_transform]: 1.782e-05
        [tuple_list_get_item_eliminator]: 5.50001e-06
        [none_parameter_eliminate]: 1.30999e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 5.15001e-06
[remove_dup_value]: 1.4e-05
[partial_unused_args_eliminate]: 2.39999e-06
[environ_conv]: 4.82998e-06
[add_recomputation]: 4.528e-05
[cse_after_recomputation]: 1.849e-05, [1]
    [Cycle 1]: 1.358e-05, [1]
        [cse]: 7.98001e-06
[auto_monad_reorder]: 1.404e-05
[get_jit_bprop_graph]: 1.43002e-06
[rewriter_after_jit_bprop_graph]: 4.87e-06
[opt_after_jit_grad]: 0.00044599
[symbol_engine_optimizer]: 6.868e-05, [1]
    [Cycle 1]: 6.238e-05, [6]
        [build]: 2.98e-06
        [elim_shapecalc]: 6.78e-06
        [elim_not_effective]: 1.224e-05
        [opt_reshape]: 5.18002e-06
        [fold_const_symbol]: 8.57e-06
        [renormalize]: 3.69997e-07
[validate]: 3.053e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.0056984
[execute]: 7.55e-06
Sums
    bootstrap                            :     0.000249s :  1.88%
    type_inference                       :     0.003932s : 29.75%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000058s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.24%
    rewriter_before_opt_a                :     0.000071s :  0.54%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000056s :  0.42%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000563s :  4.26%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000554s :  4.19%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000037s :  0.28%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000484s :  3.66%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000408s :  3.08%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000045s :  0.34%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000446s :  3.37%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005698s : 43.11%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000168    23
 1.16% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000001s :      2: substitution.fold_const_symbol
 2.99% :     0.000005s :      3: substitution.graph_param_transform
81.82% :     0.000137s :      4: substitution.inline
 1.81% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.57% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.13% :     0.000004s :      2: substitution.replace_old_param
 6.75% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003882     2
82.93% :     0.003219s :      1: type_inference.infer
17.07% :     0.000663s :      1: type_inference.specialize
------[replace.]   0.000059     6
67.36% :     0.000040s :      4: replace.inline
32.64% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000145     6
92.93% :     0.000135s :      4: match.inline
 7.07% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000001s :     10: predicate.addn_zero_filter
 1.81% :     0.000002s :     10: predicate.arithmetic_simplify
 1.42% :     0.000002s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000001s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.50% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.11% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.46% :     0.000006s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.50% :     0.000002s :      6: predicate.less_batch_normalization
 2.04% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.87% :     0.000002s :     15: predicate.load_eliminater
 1.49% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.93% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.69% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.77% :     0.000002s :     10: predicate.reshape_eliminate
 1.33% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.61% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.54% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.84% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.84% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000477    12
48.77% :     0.000233s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.91% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.31% :     0.000216s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018246    76
 0.26% :     0.000048s :      1: add_recomputation
 0.33% :     0.000061s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.43% :     0.000261s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.11% :     0.000020s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000003s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.26% :     0.004791s :      1: jit_opt_a
 0.74% :     0.000134s :      1: jit_opt_after_cconv
 0.27% :     0.000049s :      1: jit_opt_b
 2.27% :     0.000414s :      1: loop_unroll
 2.70% :     0.000492s :      1: mutable_eliminate
 4.16% :     0.000760s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.49% :     0.000454s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000034s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 1.66% :     0.000303s :      1: renormalize.infer
 1.34% :     0.000245s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.40% :     0.000073s :      1: rewriter_before_opt_a
 0.39% :     0.000071s :      1: symbol_engine_optimizer
31.29% :     0.005709s :      1: task_emit
21.61% :     0.003943s :      1: type_inference
 0.26% :     0.000048s :      1: validate


TotalTime = 0.0201958, [33]
[bootstrap]: 0.00033164
[type_inference]: 0.00467538
[event_method]: 1.766e-05
[auto_monad]: 6.563e-05
[graph_reusing]: 5.01002e-06
[pre_auto_parallel]: 2.39001e-06
[py_interpret_to_execute]: 3.428e-05
[rewriter_before_opt_a]: 6.983e-05
[expand_dump_flag]: 2.95002e-06
[jit_opt_a]: 0.00572855, [2]
    [Cycle 1]: 0.0016936, [27]
        [switch_simplify]: 5.623e-05
        [loop_unroll]: 2.923e-05
        [a_1]: 0.0005392
        [with_stream_mark]: 1.686e-05
        [recompute_prepare]: 7.13e-06
        [updatestate_depend_eliminate]: 3.02002e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.67001e-06
        [parameter_eliminate]: 1.76998e-06
        [specialize_transform]: 5.96e-06
        [updatestate_useless_node_eliminater]: 5.39e-06
        [accelerated_algorithm]: 6.06e-06
        [meta_shard_fg_expand]: 2.12001e-06
        [get_grad_eliminate_]: 5.35001e-06
        [merge_forward]: 3.56001e-06
        [cell_reuse_recompute_pass]: 1.21002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.582e-05
        [j_node_and_user_rematch]: 9.81e-06
        [meta_fg_expand]: 2.14e-06
        [replace_old_param]: 8.69e-06
        [inline_without_move]: 5.55001e-06
        [renormalize]: 0.00073494
        [add_forward_monad_depend]: 5.27999e-06
        [auto_monad_grad]: 2.46998e-06
        [auto_monad_eliminator]: 1.399e-05
        [cse]: 2.003e-05
        [replace_applicator]: 1.241e-05
    [Cycle 2]: 0.00031705, [27]
        [switch_simplify]: 5.47999e-06
        [loop_unroll]: 4.84e-06
        [a_1]: 8.977e-05
        [with_stream_mark]: 9.91998e-06
        [recompute_prepare]: 4.94e-06
        [updatestate_depend_eliminate]: 2.48e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [parameter_eliminate]: 1.18001e-06
        [specialize_transform]: 4.85001e-06
        [updatestate_useless_node_eliminater]: 4.65999e-06
        [accelerated_algorithm]: 5.00001e-06
        [meta_shard_fg_expand]: 9.80013e-07
        [get_grad_eliminate_]: 4.36002e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.81e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.156e-05
        [j_node_and_user_rematch]: 7.43e-06
        [meta_fg_expand]: 1.63002e-06
        [replace_old_param]: 7.56001e-06
        [inline_without_move]: 4.80999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 4.98001e-06
        [cse]: 1.009e-05
        [replace_applicator]: 4.84e-06
[py_interpret_to_execute_after_opt_a]: 1.016e-05
[rewriter_after_opt_a]: 3.235e-05
[convert_after_rewriter]: 7.11001e-06
[order_py_execute_after_rewriter]: 8.03999e-06
[mutable_eliminate]: 0.00058454
[jit_opt_b]: 5.036e-05, [1]
    [Cycle 1]: 4.315e-05, [2]
        [frontend_op_eliminate]: 1.734e-05
        [inline_after_opt_a]: 1.472e-05
[cconv]: 2.757e-05
[loop_unroll]: 0.00043554
[jit_opt_after_cconv]: 0.00014416, [1]
    [Cycle 1]: 0.00013711, [11]
        [c_1]: 2.18e-05
        [parameter_eliminate]: 2.58003e-06
        [updatestate_depend_eliminate]: 4.58999e-06
        [updatestate_assign_eliminate]: 2.20002e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [cse]: 2.099e-05
        [call_graph_tuple_transform]: 1.905e-05
        [tuple_list_get_item_eliminator]: 5.76998e-06
        [none_parameter_eliminate]: 1.33002e-06
        [renormalize]: 7.50006e-07
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 1.415e-05
[partial_unused_args_eliminate]: 2.68998e-06
[environ_conv]: 5.06002e-06
[add_recomputation]: 4.978e-05
[cse_after_recomputation]: 2.194e-05, [1]
    [Cycle 1]: 1.632e-05, [1]
        [cse]: 9.99001e-06
[auto_monad_reorder]: 1.67e-05
[get_jit_bprop_graph]: 2.35002e-06
[rewriter_after_jit_bprop_graph]: 5.93998e-06
[opt_after_jit_grad]: 0.00048523
[symbol_engine_optimizer]: 7.442e-05, [1]
    [Cycle 1]: 6.738e-05, [6]
        [build]: 3.78999e-06
        [elim_shapecalc]: 7.21999e-06
        [elim_not_effective]: 1.233e-05
        [opt_reshape]: 6.28e-06
        [fold_const_symbol]: 8.93002e-06
        [renormalize]: 7.00005e-07
[validate]: 3.827e-05
[backend_pass]: 1.10999e-06
[task_emit]: 0.00703104
[execute]: 9.28997e-06
Sums
    bootstrap                            :     0.000332s :  2.09%
    type_inference                       :     0.004675s : 29.48%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000066s :  0.41%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.22%
    rewriter_before_opt_a                :     0.000070s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.21%
    jit_opt_a.a_1                        :     0.000629s :  3.97%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000735s :  4.63%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000030s :  0.19%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000032s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000008s :  0.05%
    mutable_eliminate                    :     0.000585s :  3.69%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000028s :  0.17%
    loop_unroll                          :     0.000436s :  2.75%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000485s :  3.06%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007031s : 44.33%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000205    23
 0.94% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.73% :     0.000006s :      3: substitution.graph_param_transform
83.87% :     0.000172s :      4: substitution.inline
 1.29% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.34% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.23% :     0.000005s :      2: substitution.replace_old_param
 5.90% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004606     2
80.81% :     0.003722s :      1: type_inference.infer
19.19% :     0.000884s :      1: type_inference.specialize
------[replace.]   0.000072     6
66.11% :     0.000047s :      4: replace.inline
33.89% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000181     6
93.88% :     0.000169s :      4: match.inline
 6.12% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.79% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000001s :     10: predicate.addn_zero_filter
 2.10% :     0.000002s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.42% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.10% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.40% :     0.000002s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.05% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.02% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.23% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000002s :     15: predicate.partial_eliminate
 1.70% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.46% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.53% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.30% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.57% :     0.000009s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.06% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.00% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000002s :     10: predicate.value_based_eliminate
 0.43% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000612    12
47.71% :     0.000292s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.80% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.49% :     0.000291s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021870    76
 0.24% :     0.000053s :      1: add_recomputation
 0.31% :     0.000069s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.58% :     0.000346s :      1: bootstrap
 0.14% :     0.000030s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
26.21% :     0.005732s :      1: jit_opt_a
 0.67% :     0.000147s :      1: jit_opt_after_cconv
 0.24% :     0.000053s :      1: jit_opt_b
 2.03% :     0.000443s :      1: loop_unroll
 2.71% :     0.000593s :      1: mutable_eliminate
 3.84% :     0.000839s :     26: opt.transform.jit_opt_a
 0.22% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.26% :     0.000494s :      1: opt_after_jit_grad
 0.05% :     0.000010s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.73% :     0.000379s :      1: renormalize.infer
 1.59% :     0.000347s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000035s :      1: rewriter_after_opt_a
 0.33% :     0.000072s :      1: rewriter_before_opt_a
 0.35% :     0.000077s :      1: symbol_engine_optimizer
32.24% :     0.007051s :      1: task_emit
21.43% :     0.004687s :      1: type_inference
 0.27% :     0.000058s :      1: validate


TotalTime = 0.0126639, [30]
[bootstrap]: 0.00024063
[type_inference]: 0.00555214
[event_method]: 2.007e-05
[auto_monad]: 8.538e-05
[graph_reusing]: 4.08001e-06
[pre_auto_parallel]: 2.01e-06
[py_interpret_to_execute]: 4.426e-05
[rewriter_before_opt_a]: 7.579e-05
[expand_dump_flag]: 2.59999e-06
[jit_opt_a]: 0.0047485, [2]
    [Cycle 1]: 0.00157932, [27]
        [switch_simplify]: 9.354e-05
        [loop_unroll]: 3.569e-05
        [a_1]: 0.00062524
        [with_stream_mark]: 1.19e-05
        [recompute_prepare]: 7.43999e-06
        [updatestate_depend_eliminate]: 2.99001e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 1.07998e-06
        [specialize_transform]: 5.89999e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 5.05001e-06
        [merge_forward]: 2.54001e-06
        [cell_reuse_recompute_pass]: 9.10019e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.112e-05
        [j_node_and_user_rematch]: 5.555e-05
        [meta_fg_expand]: 1.76998e-06
        [replace_old_param]: 7.8e-06
        [inline_without_move]: 5.35999e-06
        [renormalize]: 0.00048142
        [add_forward_monad_depend]: 4.40999e-06
        [auto_monad_grad]: 1.44998e-06
        [auto_monad_eliminator]: 9.86003e-06
        [cse]: 1.477e-05
        [replace_applicator]: 1.114e-05
    [Cycle 2]: 0.00031928, [27]
        [switch_simplify]: 5.76998e-06
        [loop_unroll]: 5.20001e-06
        [a_1]: 8.886e-05
        [with_stream_mark]: 8.18999e-06
        [recompute_prepare]: 4.84e-06
        [updatestate_depend_eliminate]: 2.75002e-06
        [updatestate_assign_eliminate]: 2.11998e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 9.09989e-07
        [specialize_transform]: 4.91002e-06
        [updatestate_useless_node_eliminater]: 4.79998e-06
        [accelerated_algorithm]: 5.01002e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 4.79e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.237e-05
        [j_node_and_user_rematch]: 7.41999e-06
        [meta_fg_expand]: 1.45999e-06
        [replace_old_param]: 6.71e-06
        [inline_without_move]: 4.69002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.26997e-06
        [auto_monad_grad]: 9.10019e-07
        [auto_monad_eliminator]: 5.10999e-06
        [cse]: 9.63002e-06
        [replace_applicator]: 5.64e-06
[py_interpret_to_execute_after_opt_a]: 8.17e-06
[rewriter_after_opt_a]: 2.466e-05
[convert_after_rewriter]: 5.51e-06
[order_py_execute_after_rewriter]: 4.47e-06
[mutable_eliminate]: 0.00046931
[jit_opt_b]: 4.729e-05, [1]
    [Cycle 1]: 4.139e-05, [2]
        [frontend_op_eliminate]: 1.561e-05
        [inline_after_opt_a]: 1.448e-05
[cconv]: 1.596e-05
[loop_unroll]: 0.00041574
[jit_opt_after_cconv]: 0.00013106, [1]
    [Cycle 1]: 0.00012503, [11]
        [c_1]: 2.082e-05
        [parameter_eliminate]: 2.34999e-06
        [updatestate_depend_eliminate]: 5.19e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.03002e-06
        [cse]: 1.564e-05
        [call_graph_tuple_transform]: 1.559e-05
        [tuple_list_get_item_eliminator]: 5.29998e-06
        [none_parameter_eliminate]: 8.2e-07
        [renormalize]: 3.39991e-07
        [switch_simplify]: 5.34998e-06
[remove_dup_value]: 7.37002e-06
[partial_unused_args_eliminate]: 1.52999e-06
[environ_conv]: 3.4e-06
[add_recomputation]: 3.244e-05
[cse_after_recomputation]: 1.994e-05, [1]
    [Cycle 1]: 1.451e-05, [1]
        [cse]: 8.92999e-06
[auto_monad_reorder]: 9.20001e-06
[get_jit_bprop_graph]: 1.46002e-06
[rewriter_after_jit_bprop_graph]: 4.60001e-06
[opt_after_jit_grad]: 0.00044532
[symbol_engine_optimizer]: 6.918e-05, [1]
    [Cycle 1]: 6.365e-05, [6]
        [build]: 2.73e-06
        [elim_shapecalc]: 7.18998e-06
        [elim_not_effective]: 1.219e-05
        [opt_reshape]: 5.56e-06
        [fold_const_symbol]: 8.3e-06
        [renormalize]: 4.39992e-07
[validate]: 2.428e-05
Sums
    bootstrap                            :     0.000241s :  2.60%
    type_inference                       :     0.005552s : 59.93%
    event_method                         :     0.000020s :  0.22%
    auto_monad                           :     0.000085s :  0.92%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000044s :  0.48%
    rewriter_before_opt_a                :     0.000076s :  0.82%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000099s :  1.07%
    jit_opt_a.loop_unroll                :     0.000041s :  0.44%
    jit_opt_a.a_1                        :     0.000714s :  7.71%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000063s :  0.68%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000481s :  5.20%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.16%
    jit_opt_a.cse                        :     0.000024s :  0.26%
    jit_opt_a.replace_applicator         :     0.000017s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.09%
    rewriter_after_opt_a                 :     0.000025s :  0.27%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000469s :  5.07%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.16%
    cconv                                :     0.000016s :  0.17%
    loop_unroll                          :     0.000416s :  4.49%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000007s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000003s :  0.04%
    add_recomputation                    :     0.000032s :  0.35%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000009s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000445s :  4.81%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000024s :  0.26%


Time group info:
------[substitution.]   0.000267    26
 0.58% :     0.000002s :      2: substitution.elim_not_effective
 0.42% :     0.000001s :      2: substitution.fold_const_symbol
 1.22% :     0.000003s :      3: substitution.graph_param_transform
70.89% :     0.000189s :      6: substitution.inline
18.15% :     0.000048s :      4: substitution.j_node_and_user_rematch
 1.57% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.01% :     0.000003s :      2: substitution.replace_old_param
 2.98% :     0.000008s :      1: substitution.switch_simplify
 3.19% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005504     2
83.47% :     0.004594s :      1: type_inference.infer
16.53% :     0.000910s :      1: type_inference.specialize
------[replace.]   0.000113     9
55.08% :     0.000062s :      6: replace.inline
19.86% :     0.000023s :      1: replace.switch_simplify
25.06% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     9
92.89% :     0.000186s :      6: match.inline
 3.57% :     0.000007s :      1: match.switch_simplify
 3.54% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   801
 1.37% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.19% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.17% :     0.000001s :     12: predicate.addn_check_dump
 1.38% :     0.000002s :     12: predicate.addn_zero_filter
 2.30% :     0.000003s :     12: predicate.arithmetic_simplify
 1.37% :     0.000002s :     12: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.20% :     0.000002s :     12: predicate.compare_switch_simplify
 1.52% :     0.000002s :     12: predicate.depend_value_elim
 1.16% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.50% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.33% :     0.000002s :     12: predicate.environ_get_eliminate
 1.21% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.23% :     0.000007s :     26: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     17: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.37% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.05% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     12: predicate.merge_addn
 1.18% :     0.000002s :     12: predicate.minmaximum_grad
 1.49% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     17: predicate.partial_eliminate
 1.33% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.93% :     0.000002s :     12: predicate.reduce_eliminate
 1.52% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     20: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     12: predicate.reshape_eliminate
 1.27% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.57% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.49% :     0.000002s :     12: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.71% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.73% :     0.000003s :     20: predicate.switch_defer_inline
 2.32% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.44% :     0.000011s :     65: predicate.switch_simplify
 1.36% :     0.000002s :     12: predicate.tile_eliminate
 1.30% :     0.000002s :     12: predicate.transpose_eliminate
 1.55% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.23% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.48% :     0.000002s :     12: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000664    15
48.22% :     0.000320s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.01% :     0.000073s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.77% :     0.000271s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014226    72
 0.25% :     0.000035s :      1: add_recomputation
 0.63% :     0.000090s :      1: auto_monad
 0.08% :     0.000012s :      1: auto_monad_reorder
 1.77% :     0.000252s :      1: bootstrap
 0.13% :     0.000018s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.16% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
33.40% :     0.004751s :      1: jit_opt_a
 0.94% :     0.000134s :      1: jit_opt_after_cconv
 0.35% :     0.000050s :      1: jit_opt_b
 2.98% :     0.000424s :      1: loop_unroll
 3.36% :     0.000478s :      1: mutable_eliminate
 7.10% :     0.001010s :     26: opt.transform.jit_opt_a
 0.30% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.18% :     0.000453s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.33% :     0.000047s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000010s :      1: remove_dup_value
 1.66% :     0.000236s :      1: renormalize.infer
 1.68% :     0.000239s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000028s :      1: rewriter_after_opt_a
 0.55% :     0.000079s :      1: rewriter_before_opt_a
 0.50% :     0.000072s :      1: symbol_engine_optimizer
39.10% :     0.005563s :      1: type_inference


TotalTime = 0.016424, [33]
[bootstrap]: 0.0002634
[type_inference]: 0.00411833
[event_method]: 1.573e-05
[auto_monad]: 6.092e-05
[graph_reusing]: 5.47001e-06
[pre_auto_parallel]: 2.52001e-06
[py_interpret_to_execute]: 3.075e-05
[rewriter_before_opt_a]: 6.782e-05
[expand_dump_flag]: 2.88e-06
[jit_opt_a]: 0.00466731, [2]
    [Cycle 1]: 0.00134602, [27]
        [switch_simplify]: 4.846e-05
        [loop_unroll]: 2.843e-05
        [a_1]: 0.00048167
        [with_stream_mark]: 1.367e-05
        [recompute_prepare]: 7.11999e-06
        [updatestate_depend_eliminate]: 3.32002e-06
        [updatestate_assign_eliminate]: 2.84001e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 5.74e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.34998e-06
        [meta_shard_fg_expand]: 1.90001e-06
        [get_grad_eliminate_]: 5.04998e-06
        [merge_forward]: 4.02e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.375e-05
        [j_node_and_user_rematch]: 8.48001e-06
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 7.92e-06
        [inline_without_move]: 5.19998e-06
        [renormalize]: 0.00045476
        [add_forward_monad_depend]: 4.95001e-06
        [auto_monad_grad]: 2.37999e-06
        [auto_monad_eliminator]: 1.259e-05
        [cse]: 1.633e-05
        [replace_applicator]: 4.025e-05
    [Cycle 2]: 0.00031196, [27]
        [switch_simplify]: 6.01e-06
        [loop_unroll]: 4.99e-06
        [a_1]: 8.809e-05
        [with_stream_mark]: 7.87003e-06
        [recompute_prepare]: 4.91002e-06
        [updatestate_depend_eliminate]: 2.41998e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 9.00007e-07
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.70999e-06
        [accelerated_algorithm]: 5.05999e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 4.28999e-06
        [merge_forward]: 2.35002e-06
        [cell_reuse_recompute_pass]: 1.82001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.161e-05
        [j_node_and_user_rematch]: 7.56999e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 7.35e-06
        [inline_without_move]: 4.45e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 9.79984e-07
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 5.23002e-06
        [cse]: 1.049e-05
        [replace_applicator]: 4.92e-06
[py_interpret_to_execute_after_opt_a]: 7.18e-06
[rewriter_after_opt_a]: 2.786e-05
[convert_after_rewriter]: 6.42001e-06
[order_py_execute_after_rewriter]: 4.85001e-06
[mutable_eliminate]: 0.00046698
[jit_opt_b]: 4.726e-05, [1]
    [Cycle 1]: 4.109e-05, [2]
        [frontend_op_eliminate]: 1.559e-05
        [inline_after_opt_a]: 1.41e-05
[cconv]: 2.187e-05
[loop_unroll]: 0.00041236
[jit_opt_after_cconv]: 0.00013066, [1]
    [Cycle 1]: 0.00012462, [11]
        [c_1]: 1.971e-05
        [parameter_eliminate]: 2.07999e-06
        [updatestate_depend_eliminate]: 4.62e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.536e-05
        [call_graph_tuple_transform]: 1.754e-05
        [tuple_list_get_item_eliminator]: 5.29e-06
        [none_parameter_eliminate]: 1.37999e-06
        [renormalize]: 5.10016e-07
        [switch_simplify]: 5.22999e-06
[remove_dup_value]: 1.201e-05
[partial_unused_args_eliminate]: 2.39001e-06
[environ_conv]: 4.79e-06
[add_recomputation]: 4.099e-05
[cse_after_recomputation]: 1.898e-05, [1]
    [Cycle 1]: 1.391e-05, [1]
        [cse]: 8.21002e-06
[auto_monad_reorder]: 1.381e-05
[get_jit_bprop_graph]: 1.39e-06
[rewriter_after_jit_bprop_graph]: 4.79e-06
[opt_after_jit_grad]: 0.00044335
[symbol_engine_optimizer]: 6.934e-05, [1]
    [Cycle 1]: 6.298e-05, [6]
        [build]: 3.42002e-06
        [elim_shapecalc]: 7.54002e-06
        [elim_not_effective]: 1.162e-05
        [opt_reshape]: 5.48002e-06
        [fold_const_symbol]: 8.13001e-06
        [renormalize]: 4.99975e-07
[validate]: 2.857e-05
[backend_pass]: 1.00001e-06
[task_emit]: 0.00525059
[execute]: 6.11e-06
Sums
    bootstrap                            :     0.000263s :  2.05%
    type_inference                       :     0.004118s : 32.02%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000061s :  0.47%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.24%
    rewriter_before_opt_a                :     0.000068s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.42%
    jit_opt_a.loop_unroll                :     0.000033s :  0.26%
    jit_opt_a.a_1                        :     0.000570s :  4.43%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000455s :  3.54%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.14%
    jit_opt_a.cse                        :     0.000027s :  0.21%
    jit_opt_a.replace_applicator         :     0.000045s :  0.35%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.06%
    rewriter_after_opt_a                 :     0.000028s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000467s :  3.63%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000022s :  0.17%
    loop_unroll                          :     0.000412s :  3.21%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000041s :  0.32%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000443s :  3.45%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005251s : 40.83%
    execute                              :     0.000006s :  0.05%


Time group info:
------[substitution.]   0.000170    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000001s :      2: substitution.fold_const_symbol
 3.05% :     0.000005s :      3: substitution.graph_param_transform
82.28% :     0.000140s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.63% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.26% :     0.000004s :      2: substitution.replace_old_param
 6.44% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004066     2
81.93% :     0.003332s :      1: type_inference.infer
18.07% :     0.000735s :      1: type_inference.specialize
------[replace.]   0.000059     6
66.30% :     0.000039s :      4: replace.inline
33.70% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000147     6
93.27% :     0.000138s :      4: match.inline
 6.73% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.52% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.19% :     0.000001s :     10: predicate.addn_check_dump
 1.48% :     0.000002s :     10: predicate.addn_zero_filter
 1.72% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.22% :     0.000001s :     10: predicate.compare_switch_simplify
 1.61% :     0.000002s :     10: predicate.depend_value_elim
 1.20% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.34% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.22% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.43% :     0.000000s :      3: predicate.graph_param_transform
 5.46% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.93% :     0.000002s :     15: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.04% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.77% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.17% :     0.000001s :     10: predicate.minmaximum_grad
 1.47% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.40% :     0.000003s :     15: predicate.partial_eliminate
 1.26% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     10: predicate.reduce_eliminate
 1.59% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.79% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.56% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.57% :     0.000008s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.28% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.80% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.39% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.11% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000497    12
50.32% :     0.000250s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.40% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.29% :     0.000220s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017776    76
 0.25% :     0.000044s :      1: add_recomputation
 0.36% :     0.000064s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.55% :     0.000275s :      1: bootstrap
 0.14% :     0.000024s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.27% :     0.004670s :      1: jit_opt_a
 0.75% :     0.000133s :      1: jit_opt_after_cconv
 0.28% :     0.000050s :      1: jit_opt_b
 2.36% :     0.000420s :      1: loop_unroll
 2.67% :     0.000475s :      1: mutable_eliminate
 4.47% :     0.000795s :     26: opt.transform.jit_opt_a
 0.25% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.54% :     0.000452s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000033s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 1.27% :     0.000226s :      1: renormalize.infer
 1.25% :     0.000222s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000031s :      1: rewriter_after_opt_a
 0.40% :     0.000071s :      1: rewriter_before_opt_a
 0.40% :     0.000072s :      1: symbol_engine_optimizer
29.59% :     0.005259s :      1: task_emit
23.22% :     0.004127s :      1: type_inference
 0.25% :     0.000044s :      1: validate


TotalTime = 0.0170237, [33]
[bootstrap]: 0.00026693
[type_inference]: 0.00408113
[event_method]: 1.715e-05
[auto_monad]: 6.055e-05
[graph_reusing]: 6.31998e-06
[pre_auto_parallel]: 2.37999e-06
[py_interpret_to_execute]: 3.136e-05
[rewriter_before_opt_a]: 7.218e-05
[expand_dump_flag]: 3.01001e-06
[jit_opt_a]: 0.00474503, [2]
    [Cycle 1]: 0.00138371, [27]
        [switch_simplify]: 4.923e-05
        [loop_unroll]: 2.92e-05
        [a_1]: 0.00049528
        [with_stream_mark]: 1.533e-05
        [recompute_prepare]: 6.97002e-06
        [updatestate_depend_eliminate]: 3.23e-06
        [updatestate_assign_eliminate]: 2.71999e-06
        [updatestate_loads_eliminate]: 2.89999e-06
        [parameter_eliminate]: 2.20002e-06
        [specialize_transform]: 5.65001e-06
        [updatestate_useless_node_eliminater]: 4.79002e-06
        [accelerated_algorithm]: 5.14e-06
        [meta_shard_fg_expand]: 2.10002e-06
        [get_grad_eliminate_]: 5.17999e-06
        [merge_forward]: 3.61999e-06
        [cell_reuse_recompute_pass]: 1.31998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.422e-05
        [j_node_and_user_rematch]: 8.77999e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 3.107e-05
        [inline_without_move]: 5.71e-06
        [renormalize]: 0.00047042
        [add_forward_monad_depend]: 4.58999e-06
        [auto_monad_grad]: 1.95001e-06
        [auto_monad_eliminator]: 1.248e-05
        [cse]: 2.517e-05
        [replace_applicator]: 1.1e-05
    [Cycle 2]: 0.00030902, [27]
        [switch_simplify]: 5.49e-06
        [loop_unroll]: 4.80001e-06
        [a_1]: 8.712e-05
        [with_stream_mark]: 8.27e-06
        [recompute_prepare]: 4.85001e-06
        [updatestate_depend_eliminate]: 2.53998e-06
        [updatestate_assign_eliminate]: 1.92001e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 4.92999e-06
        [updatestate_useless_node_eliminater]: 4.63999e-06
        [accelerated_algorithm]: 5.05999e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.27e-06
        [merge_forward]: 2.16e-06
        [cell_reuse_recompute_pass]: 1.37999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.172e-05
        [j_node_and_user_rematch]: 7.63999e-06
        [meta_fg_expand]: 1.37e-06
        [replace_old_param]: 7.11001e-06
        [inline_without_move]: 4.58001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 9.70002e-07
        [auto_monad_grad]: 8.40024e-07
        [auto_monad_eliminator]: 4.75999e-06
        [cse]: 9.74e-06
        [replace_applicator]: 4.90999e-06
[py_interpret_to_execute_after_opt_a]: 8.50001e-06
[rewriter_after_opt_a]: 3.098e-05
[convert_after_rewriter]: 6.71999e-06
[order_py_execute_after_rewriter]: 5.59e-06
[mutable_eliminate]: 0.00047179
[jit_opt_b]: 4.71e-05, [1]
    [Cycle 1]: 4.046e-05, [2]
        [frontend_op_eliminate]: 1.527e-05
        [inline_after_opt_a]: 1.467e-05
[cconv]: 2.33e-05
[loop_unroll]: 0.0004085
[jit_opt_after_cconv]: 0.00013389, [1]
    [Cycle 1]: 0.00012752, [11]
        [c_1]: 1.977e-05
        [parameter_eliminate]: 2.35002e-06
        [updatestate_depend_eliminate]: 5.28002e-06
        [updatestate_assign_eliminate]: 2.19999e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [cse]: 1.707e-05
        [call_graph_tuple_transform]: 1.739e-05
        [tuple_list_get_item_eliminator]: 5.28002e-06
        [none_parameter_eliminate]: 1.82999e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.09e-06
[remove_dup_value]: 1.358e-05
[partial_unused_args_eliminate]: 2.07001e-06
[environ_conv]: 4.74998e-06
[add_recomputation]: 4.659e-05
[cse_after_recomputation]: 1.968e-05, [1]
    [Cycle 1]: 1.457e-05, [1]
        [cse]: 8.89e-06
[auto_monad_reorder]: 1.476e-05
[get_jit_bprop_graph]: 1.44e-06
[rewriter_after_jit_bprop_graph]: 4.75999e-06
[opt_after_jit_grad]: 0.00044824
[symbol_engine_optimizer]: 6.863e-05, [1]
    [Cycle 1]: 6.238e-05, [6]
        [build]: 2.98e-06
        [elim_shapecalc]: 6.88e-06
        [elim_not_effective]: 1.17e-05
        [opt_reshape]: 5.52999e-06
        [fold_const_symbol]: 8.20999e-06
        [renormalize]: 5.09986e-07
[validate]: 3.05e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.0057123
[execute]: 7.48999e-06
Sums
    bootstrap                            :     0.000267s :  2.00%
    type_inference                       :     0.004081s : 30.56%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000061s :  0.45%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.23%
    rewriter_before_opt_a                :     0.000072s :  0.54%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000582s :  4.36%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000038s :  0.29%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000471s :  3.52%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000035s :  0.26%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000472s :  3.53%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000408s :  3.06%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.35%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000448s :  3.36%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000030s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005712s : 42.77%
    execute                              :     0.000007s :  0.06%


Time group info:
------[substitution.]   0.000198    23
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.39% :     0.000005s :      3: substitution.graph_param_transform
73.26% :     0.000145s :      4: substitution.inline
 1.49% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.14% :     0.000004s :      4: substitution.remove_not_recompute_node
13.19% :     0.000026s :      2: substitution.replace_old_param
 6.00% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004023     2
81.03% :     0.003260s :      1: type_inference.infer
18.97% :     0.000763s :      1: type_inference.specialize
------[replace.]   0.000063     6
68.34% :     0.000043s :      4: replace.inline
31.66% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000153     6
93.06% :     0.000143s :      4: match.inline
 6.94% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.62% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 2.12% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.56% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.72% :     0.000001s :      3: predicate.graph_param_transform
 5.40% :     0.000006s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.33% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.49% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.26% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.57% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.87% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.26% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.86% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000494    12
48.32% :     0.000239s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.19% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.49% :     0.000230s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018399    76
 0.27% :     0.000049s :      1: add_recomputation
 0.61% :     0.000112s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.55% :     0.000284s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
25.80% :     0.004748s :      1: jit_opt_a
 0.74% :     0.000137s :      1: jit_opt_after_cconv
 0.27% :     0.000050s :      1: jit_opt_b
 2.26% :     0.000416s :      1: loop_unroll
 2.61% :     0.000480s :      1: mutable_eliminate
 4.36% :     0.000803s :     26: opt.transform.jit_opt_a
 0.24% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.47% :     0.000455s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 1.27% :     0.000234s :      1: renormalize.infer
 1.25% :     0.000230s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.41% :     0.000075s :      1: rewriter_before_opt_a
 0.39% :     0.000071s :      1: symbol_engine_optimizer
31.10% :     0.005723s :      1: task_emit
22.25% :     0.004093s :      1: type_inference
 0.26% :     0.000048s :      1: validate


TotalTime = 0.0156354, [33]
[bootstrap]: 0.00024081
[type_inference]: 0.00387568
[event_method]: 1.542e-05
[auto_monad]: 5.035e-05
[graph_reusing]: 4.07998e-06
[pre_auto_parallel]: 2.01e-06
[py_interpret_to_execute]: 2.999e-05
[rewriter_before_opt_a]: 6.25e-05
[expand_dump_flag]: 2.38998e-06
[jit_opt_a]: 0.00450485, [2]
    [Cycle 1]: 0.0012648, [27]
        [switch_simplify]: 4.268e-05
        [loop_unroll]: 2.861e-05
        [a_1]: 0.00046848
        [with_stream_mark]: 1.196e-05
        [recompute_prepare]: 7.03e-06
        [updatestate_depend_eliminate]: 3.09001e-06
        [updatestate_assign_eliminate]: 2.51998e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 1.37e-06
        [specialize_transform]: 5.67001e-06
        [updatestate_useless_node_eliminater]: 4.73001e-06
        [accelerated_algorithm]: 5.05999e-06
        [meta_shard_fg_expand]: 1.55999e-06
        [get_grad_eliminate_]: 5.13002e-06
        [merge_forward]: 2.46998e-06
        [cell_reuse_recompute_pass]: 9.30013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.13e-05
        [j_node_and_user_rematch]: 8.28999e-06
        [meta_fg_expand]: 1.79998e-06
        [replace_old_param]: 8e-06
        [inline_without_move]: 4.87998e-06
        [renormalize]: 0.00043975
        [add_forward_monad_depend]: 4.06001e-06
        [auto_monad_grad]: 1.45999e-06
        [auto_monad_eliminator]: 1.076e-05
        [cse]: 1.663e-05
        [replace_applicator]: 1.05e-05
    [Cycle 2]: 0.00030261, [27]
        [switch_simplify]: 5.30001e-06
        [loop_unroll]: 4.67998e-06
        [a_1]: 8.823e-05
        [with_stream_mark]: 7.77e-06
        [recompute_prepare]: 4.89e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 1.94999e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.43002e-06
        [specialize_transform]: 4.85001e-06
        [updatestate_useless_node_eliminater]: 4.68001e-06
        [accelerated_algorithm]: 5.03002e-06
        [meta_shard_fg_expand]: 1.21002e-06
        [get_grad_eliminate_]: 4.37998e-06
        [merge_forward]: 2.29999e-06
        [cell_reuse_recompute_pass]: 1.69998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.096e-05
        [j_node_and_user_rematch]: 7.13998e-06
        [meta_fg_expand]: 1.39003e-06
        [replace_old_param]: 7.28e-06
        [inline_without_move]: 4.94e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 4.42e-06
        [cse]: 9.37999e-06
        [replace_applicator]: 4.80001e-06
[py_interpret_to_execute_after_opt_a]: 7.26001e-06
[rewriter_after_opt_a]: 2.413e-05
[convert_after_rewriter]: 6.36e-06
[order_py_execute_after_rewriter]: 4.15e-06
[mutable_eliminate]: 0.00047428
[jit_opt_b]: 4.746e-05, [1]
    [Cycle 1]: 4.131e-05, [2]
        [frontend_op_eliminate]: 1.552e-05
        [inline_after_opt_a]: 1.471e-05
[cconv]: 1.708e-05
[loop_unroll]: 0.00041303
[jit_opt_after_cconv]: 0.00012918, [1]
    [Cycle 1]: 0.00012341, [11]
        [c_1]: 2.066e-05
        [parameter_eliminate]: 2.39999e-06
        [updatestate_depend_eliminate]: 4.68001e-06
        [updatestate_assign_eliminate]: 2.28998e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.529e-05
        [call_graph_tuple_transform]: 1.595e-05
        [tuple_list_get_item_eliminator]: 5.12e-06
        [none_parameter_eliminate]: 1.45001e-06
        [renormalize]: 5.00004e-07
        [switch_simplify]: 5.07999e-06
[remove_dup_value]: 8.90999e-06
[partial_unused_args_eliminate]: 1.55001e-06
[environ_conv]: 3.8e-06
[add_recomputation]: 3.321e-05
[cse_after_recomputation]: 1.821e-05, [1]
    [Cycle 1]: 1.342e-05, [1]
        [cse]: 8.07998e-06
[auto_monad_reorder]: 1.164e-05
[get_jit_bprop_graph]: 1.41002e-06
[rewriter_after_jit_bprop_graph]: 4.64998e-06
[opt_after_jit_grad]: 0.00046402
[symbol_engine_optimizer]: 6.911e-05, [1]
    [Cycle 1]: 6.283e-05, [6]
        [build]: 2.86999e-06
        [elim_shapecalc]: 7.41001e-06
        [elim_not_effective]: 1.179e-05
        [opt_reshape]: 5.52001e-06
        [fold_const_symbol]: 8.33001e-06
        [renormalize]: 4.69998e-07
[validate]: 2.51e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.00492058
[execute]: 5.15999e-06
Sums
    bootstrap                            :     0.000241s :  1.98%
    type_inference                       :     0.003876s : 31.86%
    event_method                         :     0.000015s :  0.13%
    auto_monad                           :     0.000050s :  0.41%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.25%
    rewriter_before_opt_a                :     0.000063s :  0.51%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000048s :  0.39%
    jit_opt_a.loop_unroll                :     0.000033s :  0.27%
    jit_opt_a.a_1                        :     0.000557s :  4.58%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000015s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.13%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000440s :  3.61%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.12%
    jit_opt_a.cse                        :     0.000026s :  0.21%
    jit_opt_a.replace_applicator         :     0.000015s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.06%
    rewriter_after_opt_a                 :     0.000024s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000474s :  3.90%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000017s :  0.14%
    loop_unroll                          :     0.000413s :  3.39%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.17%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000009s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000033s :  0.27%
    cse_after_recomputation.cse          :     0.000008s :  0.07%
    auto_monad_reorder                   :     0.000012s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000464s :  3.81%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000025s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.004921s : 40.44%
    execute                              :     0.000005s :  0.04%


Time group info:
------[substitution.]   0.000156    23
 1.14% :     0.000002s :      2: substitution.elim_not_effective
 0.79% :     0.000001s :      2: substitution.fold_const_symbol
 2.45% :     0.000004s :      3: substitution.graph_param_transform
82.67% :     0.000129s :      4: substitution.inline
 1.84% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.56% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.31% :     0.000004s :      2: substitution.replace_old_param
 6.24% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003831     2
81.01% :     0.003103s :      1: type_inference.infer
18.99% :     0.000727s :      1: type_inference.specialize
------[replace.]   0.000060     6
66.11% :     0.000039s :      4: replace.inline
33.89% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000135     6
93.59% :     0.000127s :      4: match.inline
 6.41% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.58% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.30% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 1.84% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000001s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.27% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.37% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.43% :     0.000000s :      3: predicate.graph_param_transform
 5.40% :     0.000006s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.99% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.61% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.09% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.54% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.56% :     0.000003s :     15: predicate.partial_eliminate
 1.29% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.10% :     0.000002s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.32% :     0.000001s :     10: predicate.reshape_eliminate
 1.29% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.76% :     0.000008s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.58% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.45% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.38% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.80% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000467    12
46.88% :     0.000219s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.50% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.61% :     0.000222s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016925    76
 0.21% :     0.000036s :      1: add_recomputation
 0.32% :     0.000054s :      1: auto_monad
 0.08% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.48% :     0.000250s :      1: bootstrap
 0.12% :     0.000020s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000020s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.05% :     0.000008s :      1: execute
 0.03% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
26.63% :     0.004507s :      1: jit_opt_a
 0.78% :     0.000132s :      1: jit_opt_after_cconv
 0.30% :     0.000050s :      1: jit_opt_b
 2.48% :     0.000420s :      1: loop_unroll
 2.84% :     0.000481s :      1: mutable_eliminate
 4.40% :     0.000744s :     26: opt.transform.jit_opt_a
 0.26% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.78% :     0.000471s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000009s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000011s :      1: remove_dup_value
 1.27% :     0.000216s :      1: renormalize.infer
 1.29% :     0.000218s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000027s :      1: rewriter_after_opt_a
 0.39% :     0.000065s :      1: rewriter_before_opt_a
 0.42% :     0.000072s :      1: symbol_engine_optimizer
29.12% :     0.004929s :      1: task_emit
22.96% :     0.003886s :      1: type_inference
 0.23% :     0.000039s :      1: validate


TotalTime = 0.012607, [30]
[bootstrap]: 0.00023743
[type_inference]: 0.00553002
[event_method]: 1.885e-05
[auto_monad]: 5.895e-05
[graph_reusing]: 4.04002e-06
[pre_auto_parallel]: 2.01e-06
[py_interpret_to_execute]: 4.225e-05
[rewriter_before_opt_a]: 7.579e-05
[expand_dump_flag]: 2.22999e-06
[jit_opt_a]: 0.00472728, [2]
    [Cycle 1]: 0.0015676, [27]
        [switch_simplify]: 8.804e-05
        [loop_unroll]: 3.281e-05
        [a_1]: 0.00067403
        [with_stream_mark]: 1.274e-05
        [recompute_prepare]: 7.06999e-06
        [updatestate_depend_eliminate]: 2.83998e-06
        [updatestate_assign_eliminate]: 2.03002e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 5.72001e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 4.89003e-06
        [merge_forward]: 2.64999e-06
        [cell_reuse_recompute_pass]: 8.79983e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.139e-05
        [j_node_and_user_rematch]: 8.69998e-06
        [meta_fg_expand]: 1.71002e-06
        [replace_old_param]: 7.38999e-06
        [inline_without_move]: 5.15001e-06
        [renormalize]: 0.00047986
        [add_forward_monad_depend]: 4.77e-06
        [auto_monad_grad]: 1.53002e-06
        [auto_monad_eliminator]: 1.016e-05
        [cse]: 1.464e-05
        [replace_applicator]: 1.103e-05
    [Cycle 2]: 0.00030296, [27]
        [switch_simplify]: 5.15999e-06
        [loop_unroll]: 4.85001e-06
        [a_1]: 8.837e-05
        [with_stream_mark]: 7.88001e-06
        [recompute_prepare]: 4.80001e-06
        [updatestate_depend_eliminate]: 2.40002e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.71002e-06
        [updatestate_useless_node_eliminater]: 4.64002e-06
        [accelerated_algorithm]: 4.57e-06
        [meta_shard_fg_expand]: 1.07998e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.17001e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.215e-05
        [j_node_and_user_rematch]: 7.41999e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 6.79999e-06
        [inline_without_move]: 4.76002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 7.79983e-07
        [auto_monad_eliminator]: 4.85001e-06
        [cse]: 9.51e-06
        [replace_applicator]: 4.51002e-06
[py_interpret_to_execute_after_opt_a]: 9.07999e-06
[rewriter_after_opt_a]: 2.388e-05
[convert_after_rewriter]: 5.30999e-06
[order_py_execute_after_rewriter]: 4.67e-06
[mutable_eliminate]: 0.00047644
[jit_opt_b]: 4.6e-05, [1]
    [Cycle 1]: 3.989e-05, [2]
        [frontend_op_eliminate]: 1.517e-05
        [inline_after_opt_a]: 1.394e-05
[cconv]: 1.833e-05
[loop_unroll]: 0.00041269
[jit_opt_after_cconv]: 0.00013183, [1]
    [Cycle 1]: 0.00012573, [11]
        [c_1]: 2.132e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 5.16002e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [cse]: 1.492e-05
        [call_graph_tuple_transform]: 1.777e-05
        [tuple_list_get_item_eliminator]: 5.24998e-06
        [none_parameter_eliminate]: 8.70001e-07
        [renormalize]: 2.00002e-07
        [switch_simplify]: 5.20999e-06
[remove_dup_value]: 9.87001e-06
[partial_unused_args_eliminate]: 1.59998e-06
[environ_conv]: 4.75001e-06
[add_recomputation]: 3.6e-05
[cse_after_recomputation]: 1.999e-05, [1]
    [Cycle 1]: 1.473e-05, [1]
        [cse]: 8.94e-06
[auto_monad_reorder]: 1.065e-05
[get_jit_bprop_graph]: 1.52001e-06
[rewriter_after_jit_bprop_graph]: 4.72998e-06
[opt_after_jit_grad]: 0.00044365
[symbol_engine_optimizer]: 6.757e-05, [1]
    [Cycle 1]: 6.17e-05, [6]
        [build]: 2.66e-06
        [elim_shapecalc]: 6.78e-06
        [elim_not_effective]: 1.153e-05
        [opt_reshape]: 5.14e-06
        [fold_const_symbol]: 8e-06
        [renormalize]: 4.19997e-07
[validate]: 2.546e-05
Sums
    bootstrap                            :     0.000237s :  2.58%
    type_inference                       :     0.005530s : 60.04%
    event_method                         :     0.000019s :  0.20%
    auto_monad                           :     0.000059s :  0.64%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000042s :  0.46%
    rewriter_before_opt_a                :     0.000076s :  0.82%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000093s :  1.01%
    jit_opt_a.loop_unroll                :     0.000038s :  0.41%
    jit_opt_a.a_1                        :     0.000762s :  8.28%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000480s :  5.21%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.16%
    jit_opt_a.cse                        :     0.000024s :  0.26%
    jit_opt_a.replace_applicator         :     0.000016s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.10%
    rewriter_after_opt_a                 :     0.000024s :  0.26%
    convert_after_rewriter               :     0.000005s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000476s :  5.17%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.15%
    cconv                                :     0.000018s :  0.20%
    loop_unroll                          :     0.000413s :  4.48%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000010s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000036s :  0.39%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000011s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000444s :  4.82%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000025s :  0.28%


Time group info:
------[substitution.]   0.000268    26
 0.56% :     0.000002s :      2: substitution.elim_not_effective
 0.37% :     0.000001s :      2: substitution.fold_const_symbol
 1.85% :     0.000005s :      3: substitution.graph_param_transform
87.10% :     0.000234s :      6: substitution.inline
 1.06% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.60% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.01% :     0.000003s :      2: substitution.replace_old_param
 2.76% :     0.000007s :      1: substitution.switch_simplify
 3.68% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005485     2
83.62% :     0.004586s :      1: type_inference.infer
16.38% :     0.000899s :      1: type_inference.specialize
------[replace.]   0.000113     9
54.93% :     0.000062s :      6: replace.inline
19.90% :     0.000023s :      1: replace.switch_simplify
25.16% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000246     9
93.80% :     0.000230s :      6: match.inline
 2.68% :     0.000007s :      1: match.switch_simplify
 3.53% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   801
 1.48% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.49% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.17% :     0.000001s :     12: predicate.addn_check_dump
 1.39% :     0.000002s :     12: predicate.addn_zero_filter
 1.82% :     0.000002s :     12: predicate.arithmetic_simplify
 1.26% :     0.000002s :     12: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.18% :     0.000001s :     12: predicate.compare_switch_simplify
 1.46% :     0.000002s :     12: predicate.depend_value_elim
 1.21% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.19% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.36% :     0.000002s :     12: predicate.environ_get_eliminate
 1.30% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.47% :     0.000007s :     26: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     17: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.23% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000001s :     12: predicate.merge_addn
 1.15% :     0.000001s :     12: predicate.minmaximum_grad
 1.67% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     17: predicate.partial_eliminate
 1.22% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     12: predicate.reduce_eliminate
 1.46% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     20: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     12: predicate.reshape_eliminate
 1.30% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.61% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     12: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.77% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.71% :     0.000003s :     20: predicate.switch_defer_inline
 2.29% :     0.000003s :     20: predicate.switch_layer_defer_inline
 7.94% :     0.000010s :     65: predicate.switch_simplify
 1.42% :     0.000002s :     12: predicate.tile_eliminate
 1.23% :     0.000002s :     12: predicate.transpose_eliminate
 1.54% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.29% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.79% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     12: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000700    15
44.87% :     0.000314s :      6: func_graph_cloner_run.FuncGraphClonerGraph
17.42% :     0.000122s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.71% :     0.000264s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014148    72
 0.27% :     0.000039s :      1: add_recomputation
 0.44% :     0.000063s :      1: auto_monad
 0.09% :     0.000013s :      1: auto_monad_reorder
 1.76% :     0.000249s :      1: bootstrap
 0.15% :     0.000021s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.16% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000022s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000006s :      1: graph_reusing
33.43% :     0.004730s :      1: jit_opt_a
 0.95% :     0.000134s :      1: jit_opt_after_cconv
 0.34% :     0.000048s :      1: jit_opt_b
 2.97% :     0.000420s :      1: loop_unroll
 3.43% :     0.000485s :      1: mutable_eliminate
 7.07% :     0.001000s :     26: opt.transform.jit_opt_a
 0.32% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000023s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000029s :      4: opt.transform.symbol_engine_opt
 3.19% :     0.000451s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.32% :     0.000045s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000012s :      1: remove_dup_value
 1.75% :     0.000247s :      1: renormalize.infer
 1.60% :     0.000226s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000027s :      1: rewriter_after_opt_a
 0.56% :     0.000079s :      1: rewriter_before_opt_a
 0.49% :     0.000070s :      1: symbol_engine_optimizer
39.16% :     0.005540s :      1: type_inference


TotalTime = 0.0251075, [33]
[bootstrap]: 0.00042315
[type_inference]: 0.00887998
[event_method]: 2.382e-05
[auto_monad]: 7.571e-05
[graph_reusing]: 6.23e-06
[pre_auto_parallel]: 3.7e-06
[py_interpret_to_execute]: 5.058e-05
[rewriter_before_opt_a]: 8.148e-05
[expand_dump_flag]: 3.31001e-06
[jit_opt_a]: 0.00627834, [2]
    [Cycle 1]: 0.00184934, [27]
        [switch_simplify]: 5.913e-05
        [loop_unroll]: 2.984e-05
        [a_1]: 0.00055701
        [with_stream_mark]: 2.011e-05
        [recompute_prepare]: 7.31999e-06
        [updatestate_depend_eliminate]: 3.14999e-06
        [updatestate_assign_eliminate]: 2.80997e-06
        [updatestate_loads_eliminate]: 2.65997e-06
        [parameter_eliminate]: 1.88002e-06
        [specialize_transform]: 5.92999e-06
        [updatestate_useless_node_eliminater]: 5.10999e-06
        [accelerated_algorithm]: 5.66e-06
        [meta_shard_fg_expand]: 2.06003e-06
        [get_grad_eliminate_]: 5.32999e-06
        [merge_forward]: 4.04002e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.438e-05
        [j_node_and_user_rematch]: 9.46998e-06
        [meta_fg_expand]: 2.64999e-06
        [replace_old_param]: 9.89001e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 0.00086255
        [add_forward_monad_depend]: 5.56998e-06
        [auto_monad_grad]: 2.58003e-06
        [auto_monad_eliminator]: 1.314e-05
        [cse]: 2.838e-05
        [replace_applicator]: 1.197e-05
    [Cycle 2]: 0.0003147, [27]
        [switch_simplify]: 5.84999e-06
        [loop_unroll]: 5.05999e-06
        [a_1]: 9.09e-05
        [with_stream_mark]: 9.47001e-06
        [recompute_prepare]: 5.07e-06
        [updatestate_depend_eliminate]: 2.88e-06
        [updatestate_assign_eliminate]: 2.06998e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 4.90999e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 4.75999e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.57e-06
        [merge_forward]: 2.63998e-06
        [cell_reuse_recompute_pass]: 2.32001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.179e-05
        [j_node_and_user_rematch]: 7.43e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 6.96001e-06
        [inline_without_move]: 4.98001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 5.17e-06
        [cse]: 1.01e-05
        [replace_applicator]: 4.85999e-06
[py_interpret_to_execute_after_opt_a]: 9.92999e-06
[rewriter_after_opt_a]: 3.615e-05
[convert_after_rewriter]: 7.53e-06
[order_py_execute_after_rewriter]: 5.05001e-06
[mutable_eliminate]: 0.00061341
[jit_opt_b]: 4.973e-05, [1]
    [Cycle 1]: 4.282e-05, [2]
        [frontend_op_eliminate]: 1.641e-05
        [inline_after_opt_a]: 1.424e-05
[cconv]: 3.585e-05
[loop_unroll]: 0.00042723
[jit_opt_after_cconv]: 0.00014062, [1]
    [Cycle 1]: 0.0001343, [11]
        [c_1]: 2.14e-05
        [parameter_eliminate]: 2.47001e-06
        [updatestate_depend_eliminate]: 6.17999e-06
        [updatestate_assign_eliminate]: 2.47001e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 1.811e-05
        [call_graph_tuple_transform]: 1.812e-05
        [tuple_list_get_item_eliminator]: 5.64998e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 7.00005e-07
        [switch_simplify]: 5.26998e-06
[remove_dup_value]: 1.45e-05
[partial_unused_args_eliminate]: 3.23e-06
[environ_conv]: 5.64998e-06
[add_recomputation]: 4.698e-05
[cse_after_recomputation]: 2.087e-05, [1]
    [Cycle 1]: 1.511e-05, [1]
        [cse]: 9.42001e-06
[auto_monad_reorder]: 1.602e-05
[get_jit_bprop_graph]: 2.08002e-06
[rewriter_after_jit_bprop_graph]: 5.27001e-06
[opt_after_jit_grad]: 0.00045852
[symbol_engine_optimizer]: 7.4e-05, [1]
    [Cycle 1]: 6.773e-05, [6]
        [build]: 4.13999e-06
        [elim_shapecalc]: 8.05e-06
        [elim_not_effective]: 1.265e-05
        [opt_reshape]: 5.78002e-06
        [fold_const_symbol]: 8.80999e-06
        [renormalize]: 7.30011e-07
[validate]: 3.82e-05
[backend_pass]: 9.10019e-07
[task_emit]: 0.00702063
[execute]: 9.29e-06
Sums
    bootstrap                            :     0.000423s :  2.08%
    type_inference                       :     0.008880s : 43.63%
    event_method                         :     0.000024s :  0.12%
    auto_monad                           :     0.000076s :  0.37%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000004s :  0.02%
    py_interpret_to_execute              :     0.000051s :  0.25%
    rewriter_before_opt_a                :     0.000081s :  0.40%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.32%
    jit_opt_a.loop_unroll                :     0.000035s :  0.17%
    jit_opt_a.a_1                        :     0.000648s :  3.18%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000863s :  4.24%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.09%
    jit_opt_a.cse                        :     0.000038s :  0.19%
    jit_opt_a.replace_applicator         :     0.000017s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.05%
    rewriter_after_opt_a                 :     0.000036s :  0.18%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000613s :  3.01%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.07%
    cconv                                :     0.000036s :  0.18%
    loop_unroll                          :     0.000427s :  2.10%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.07%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000047s :  0.23%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000459s :  2.25%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.19%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007021s : 34.49%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000217    23
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.55% :     0.000006s :      3: substitution.graph_param_transform
84.54% :     0.000184s :      4: substitution.inline
 1.49% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.18% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.81% :     0.000004s :      2: substitution.replace_old_param
 5.89% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008776     2
85.09% :     0.007468s :      1: type_inference.infer
14.91% :     0.001308s :      1: type_inference.specialize
------[replace.]   0.000074     6
67.97% :     0.000051s :      4: replace.inline
32.03% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000193     6
94.00% :     0.000181s :      4: match.inline
 6.00% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 2.50% :     0.000003s :     10: predicate.arithmetic_simplify
 1.32% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000001s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.56% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.36% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.92% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.35% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.51% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.83% :     0.000009s :     53: predicate.switch_simplify
 1.40% :     0.000002s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.67% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000759    12
46.51% :     0.000353s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.07% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.42% :     0.000375s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026919    76
 0.18% :     0.000050s :      1: add_recomputation
 0.30% :     0.000080s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 1.66% :     0.000448s :      1: bootstrap
 0.14% :     0.000039s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.11% :     0.000030s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
23.34% :     0.006282s :      1: jit_opt_a
 0.53% :     0.000143s :      1: jit_opt_after_cconv
 0.19% :     0.000052s :      1: jit_opt_b
 1.62% :     0.000435s :      1: loop_unroll
 2.31% :     0.000623s :      1: mutable_eliminate
 3.20% :     0.000861s :     26: opt.transform.jit_opt_a
 0.17% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000024s :      4: opt.transform.jit_opt_b
 0.04% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.73% :     0.000467s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000006s :      1: pre_auto_parallel
 0.20% :     0.000054s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 1.56% :     0.000421s :      1: renormalize.infer
 1.61% :     0.000434s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000039s :      1: rewriter_after_opt_a
 0.31% :     0.000084s :      1: rewriter_before_opt_a
 0.29% :     0.000077s :      1: symbol_engine_optimizer
26.15% :     0.007039s :      1: task_emit
33.09% :     0.008907s :      1: type_inference
 0.22% :     0.000058s :      1: validate


TotalTime = 0.016619, [30]
[bootstrap]: 0.00026579
[type_inference]: 0.00590058
[event_method]: 1.923e-05
[auto_monad]: 7.683e-05
[graph_reusing]: 5.88998e-06
[pre_auto_parallel]: 2.39001e-06
[py_interpret_to_execute]: 4.356e-05
[rewriter_before_opt_a]: 8.455e-05
[expand_dump_flag]: 2.97002e-06
[jit_opt_a]: 0.00830402, [2]
    [Cycle 1]: 0.00152113, [27]
        [switch_simplify]: 8.783e-05
        [loop_unroll]: 3.301e-05
        [a_1]: 0.00061719
        [with_stream_mark]: 1.224e-05
        [recompute_prepare]: 6.64999e-06
        [updatestate_depend_eliminate]: 2.93e-06
        [updatestate_assign_eliminate]: 2.13998e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 5.59e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.08002e-06
        [meta_shard_fg_expand]: 1.49998e-06
        [get_grad_eliminate_]: 5.03002e-06
        [merge_forward]: 2.84999e-06
        [cell_reuse_recompute_pass]: 9.80013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.104e-05
        [j_node_and_user_rematch]: 8.75999e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 7.46001e-06
        [inline_without_move]: 5.11997e-06
        [renormalize]: 0.00049551
        [add_forward_monad_depend]: 4.10998e-06
        [auto_monad_grad]: 1.19998e-06
        [auto_monad_eliminator]: 9.68002e-06
        [cse]: 1.541e-05
        [replace_applicator]: 1.044e-05
    [Cycle 2]: 0.00030453, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 5.01002e-06
        [a_1]: 8.769e-05
        [with_stream_mark]: 7.78999e-06
        [recompute_prepare]: 4.74998e-06
        [updatestate_depend_eliminate]: 2.44001e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 4.78001e-06
        [updatestate_useless_node_eliminater]: 4.83001e-06
        [accelerated_algorithm]: 4.67998e-06
        [meta_shard_fg_expand]: 1.03001e-06
        [get_grad_eliminate_]: 4.34002e-06
        [merge_forward]: 2.42001e-06
        [cell_reuse_recompute_pass]: 1.30001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.202e-05
        [j_node_and_user_rematch]: 7.63999e-06
        [meta_fg_expand]: 1.47999e-06
        [replace_old_param]: 6.97002e-06
        [inline_without_move]: 4.72998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 4.43999e-06
        [cse]: 9.02999e-06
        [replace_applicator]: 5.25999e-06
[py_interpret_to_execute_after_opt_a]: 7.73999e-06
[rewriter_after_opt_a]: 2.367e-05
[convert_after_rewriter]: 5.48997e-06
[order_py_execute_after_rewriter]: 4.11001e-06
[mutable_eliminate]: 0.00049593
[jit_opt_b]: 4.655e-05, [1]
    [Cycle 1]: 4.078e-05, [2]
        [frontend_op_eliminate]: 1.536e-05
        [inline_after_opt_a]: 1.399e-05
[cconv]: 1.641e-05
[loop_unroll]: 0.00040972
[jit_opt_after_cconv]: 0.00012922, [1]
    [Cycle 1]: 0.00012332, [11]
        [c_1]: 2.005e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 4.92e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 1.98997e-06
        [cse]: 1.533e-05
        [call_graph_tuple_transform]: 1.543e-05
        [tuple_list_get_item_eliminator]: 5.02999e-06
        [none_parameter_eliminate]: 1.06002e-06
        [renormalize]: 6.80011e-07
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 7.43999e-06
[partial_unused_args_eliminate]: 1.54e-06
[environ_conv]: 3.61001e-06
[add_recomputation]: 3.101e-05
[cse_after_recomputation]: 2.004e-05, [1]
    [Cycle 1]: 1.502e-05, [1]
        [cse]: 9.32001e-06
[auto_monad_reorder]: 9.57999e-06
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 4.84998e-06
[opt_after_jit_grad]: 0.00044213
[symbol_engine_optimizer]: 6.751e-05, [1]
    [Cycle 1]: 6.171e-05, [6]
        [build]: 2.47001e-06
        [elim_shapecalc]: 7.23e-06
        [elim_not_effective]: 1.164e-05
        [opt_reshape]: 5.30999e-06
        [fold_const_symbol]: 7.71999e-06
        [renormalize]: 4.99975e-07
[validate]: 2.355e-05
Sums
    bootstrap                            :     0.000266s :  2.77%
    type_inference                       :     0.005901s : 61.50%
    event_method                         :     0.000019s :  0.20%
    auto_monad                           :     0.000077s :  0.80%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000044s :  0.45%
    rewriter_before_opt_a                :     0.000085s :  0.88%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000093s :  0.97%
    jit_opt_a.loop_unroll                :     0.000038s :  0.40%
    jit_opt_a.a_1                        :     0.000705s :  7.35%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000496s :  5.17%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.15%
    jit_opt_a.cse                        :     0.000024s :  0.25%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.08%
    rewriter_after_opt_a                 :     0.000024s :  0.25%
    convert_after_rewriter               :     0.000005s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000496s :  5.17%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.15%
    cconv                                :     0.000016s :  0.17%
    loop_unroll                          :     0.000410s :  4.27%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000007s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000031s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000010s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000442s :  4.61%
    symbol_engine_optimizer.build        :     0.000002s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000024s :  0.25%


Time group info:
------[substitution.]   0.000219    26
 0.68% :     0.000001s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 1.56% :     0.000003s :      3: substitution.graph_param_transform
85.78% :     0.000188s :      6: substitution.inline
 1.24% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.96% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.26% :     0.000003s :      2: substitution.replace_old_param
 3.44% :     0.000008s :      1: substitution.switch_simplify
 3.62% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005845     2
83.86% :     0.004901s :      1: type_inference.infer
16.14% :     0.000943s :      1: type_inference.specialize
------[replace.]   0.000111     9
55.37% :     0.000061s :      6: replace.inline
18.97% :     0.000021s :      1: replace.switch_simplify
25.66% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000198     9
93.16% :     0.000184s :      6: match.inline
 3.41% :     0.000007s :      1: match.switch_simplify
 3.43% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   801
 1.56% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.08% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.19% :     0.000001s :     12: predicate.addn_check_dump
 1.37% :     0.000002s :     12: predicate.addn_zero_filter
 2.01% :     0.000002s :     12: predicate.arithmetic_simplify
 1.37% :     0.000002s :     12: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.19% :     0.000001s :     12: predicate.compare_switch_simplify
 1.43% :     0.000002s :     12: predicate.depend_value_elim
 1.18% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.38% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.19% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.45% :     0.000002s :     12: predicate.environ_get_eliminate
 1.22% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.47% :     0.000007s :     26: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.95% :     0.000001s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     17: predicate.load_eliminater
 1.51% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.49% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.93% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000001s :     12: predicate.merge_addn
 1.20% :     0.000001s :     12: predicate.minmaximum_grad
 1.42% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000000s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     17: predicate.partial_eliminate
 1.23% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     12: predicate.reduce_eliminate
 1.54% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     20: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     12: predicate.reshape_eliminate
 1.31% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.62% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     12: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.70% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.73% :     0.000003s :     20: predicate.switch_defer_inline
 2.47% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.25% :     0.000010s :     65: predicate.switch_simplify
 1.31% :     0.000002s :     12: predicate.tile_eliminate
 1.42% :     0.000002s :     12: predicate.transpose_eliminate
 1.38% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.20% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000676    15
49.64% :     0.000336s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.96% :     0.000074s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.40% :     0.000267s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018111    72
 0.19% :     0.000034s :      1: add_recomputation
 0.44% :     0.000080s :      1: auto_monad
 0.07% :     0.000012s :      1: auto_monad_reorder
 1.54% :     0.000279s :      1: bootstrap
 0.11% :     0.000019s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.12% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
45.87% :     0.008307s :      1: jit_opt_a
 0.73% :     0.000132s :      1: jit_opt_after_cconv
 0.27% :     0.000049s :      1: jit_opt_b
 2.30% :     0.000417s :      1: loop_unroll
 2.79% :     0.000504s :      1: mutable_eliminate
 5.20% :     0.000943s :     26: opt.transform.jit_opt_a
 0.23% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.48% :     0.000450s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000046s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000010s :      1: remove_dup_value
 1.40% :     0.000253s :      1: renormalize.infer
 1.31% :     0.000237s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000027s :      1: rewriter_after_opt_a
 0.48% :     0.000087s :      1: rewriter_before_opt_a
 0.39% :     0.000070s :      1: symbol_engine_optimizer
32.64% :     0.005912s :      1: type_inference


TotalTime = 0.016014, [33]
[bootstrap]: 0.00025317
[type_inference]: 0.00390485
[event_method]: 1.395e-05
[auto_monad]: 4.068e-05
[graph_reusing]: 3.21001e-06
[pre_auto_parallel]: 1.67001e-06
[py_interpret_to_execute]: 2.484e-05
[rewriter_before_opt_a]: 6.14e-05
[expand_dump_flag]: 2.14e-06
[jit_opt_a]: 0.00462747, [2]
    [Cycle 1]: 0.00132646, [27]
        [switch_simplify]: 4.75e-05
        [loop_unroll]: 2.826e-05
        [a_1]: 0.00047812
        [with_stream_mark]: 1.494e-05
        [recompute_prepare]: 6.58998e-06
        [updatestate_depend_eliminate]: 3.22002e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 2.15002e-06
        [specialize_transform]: 6.07001e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 2.08002e-06
        [get_grad_eliminate_]: 4.95999e-06
        [merge_forward]: 3.43e-06
        [cell_reuse_recompute_pass]: 1.67001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.318e-05
        [j_node_and_user_rematch]: 8.41002e-06
        [meta_fg_expand]: 2.19999e-06
        [replace_old_param]: 7.97003e-06
        [inline_without_move]: 5.02e-06
        [renormalize]: 0.00047251
        [add_forward_monad_depend]: 4.67998e-06
        [auto_monad_grad]: 2.06e-06
        [auto_monad_eliminator]: 1.238e-05
        [cse]: 1.678e-05
        [replace_applicator]: 1.102e-05
    [Cycle 2]: 0.00031067, [27]
        [switch_simplify]: 5.34e-06
        [loop_unroll]: 4.74e-06
        [a_1]: 8.708e-05
        [with_stream_mark]: 8.55999e-06
        [recompute_prepare]: 4.87998e-06
        [updatestate_depend_eliminate]: 2.39999e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 4.83001e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 4.98001e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.19001e-06
        [cell_reuse_recompute_pass]: 1.97999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.171e-05
        [j_node_and_user_rematch]: 7.63999e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 7.32002e-06
        [inline_without_move]: 4.80001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 4.63999e-06
        [cse]: 1.017e-05
        [replace_applicator]: 4.89e-06
[py_interpret_to_execute_after_opt_a]: 8.59e-06
[rewriter_after_opt_a]: 2.979e-05
[convert_after_rewriter]: 6.28998e-06
[order_py_execute_after_rewriter]: 4.95001e-06
[mutable_eliminate]: 0.00045561
[jit_opt_b]: 4.576e-05, [1]
    [Cycle 1]: 3.897e-05, [2]
        [frontend_op_eliminate]: 1.459e-05
        [inline_after_opt_a]: 1.374e-05
[cconv]: 2.219e-05
[loop_unroll]: 0.00039492
[jit_opt_after_cconv]: 0.00012938, [1]
    [Cycle 1]: 0.00012351, [11]
        [c_1]: 1.959e-05
        [parameter_eliminate]: 2.46998e-06
        [updatestate_depend_eliminate]: 4.89e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [cse]: 1.578e-05
        [call_graph_tuple_transform]: 1.632e-05
        [tuple_list_get_item_eliminator]: 5.01002e-06
        [none_parameter_eliminate]: 1.27e-06
        [renormalize]: 4.7998e-07
        [switch_simplify]: 5.14e-06
[remove_dup_value]: 1.285e-05
[partial_unused_args_eliminate]: 2.34999e-06
[environ_conv]: 4.49002e-06
[add_recomputation]: 4.056e-05
[cse_after_recomputation]: 1.87e-05, [1]
    [Cycle 1]: 1.352e-05, [1]
        [cse]: 8.08001e-06
[auto_monad_reorder]: 1.365e-05
[get_jit_bprop_graph]: 1.47999e-06
[rewriter_after_jit_bprop_graph]: 4.67e-06
[opt_after_jit_grad]: 0.00046731
[symbol_engine_optimizer]: 6.912e-05, [1]
    [Cycle 1]: 6.316e-05, [6]
        [build]: 2.98e-06
        [elim_shapecalc]: 7.21001e-06
        [elim_not_effective]: 1.235e-05
        [opt_reshape]: 5.37999e-06
        [fold_const_symbol]: 8e-06
        [renormalize]: 5.10016e-07
[validate]: 2.843e-05
[backend_pass]: 9.79984e-07
[task_emit]: 0.00514944
[execute]: 6.61e-06
Sums
    bootstrap                            :     0.000253s :  2.03%
    type_inference                       :     0.003905s : 31.30%
    event_method                         :     0.000014s :  0.11%
    auto_monad                           :     0.000041s :  0.33%
    graph_reusing                        :     0.000003s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000025s :  0.20%
    rewriter_before_opt_a                :     0.000061s :  0.49%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.42%
    jit_opt_a.loop_unroll                :     0.000033s :  0.26%
    jit_opt_a.a_1                        :     0.000565s :  4.53%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.08%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000473s :  3.79%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.14%
    jit_opt_a.cse                        :     0.000027s :  0.22%
    jit_opt_a.replace_applicator         :     0.000016s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000030s :  0.24%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000456s :  3.65%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000022s :  0.18%
    loop_unroll                          :     0.000395s :  3.17%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000041s :  0.33%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000467s :  3.75%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000028s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005149s : 41.27%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000170    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.90% :     0.000005s :      3: substitution.graph_param_transform
82.57% :     0.000140s :      4: substitution.inline
 1.71% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.33% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.22% :     0.000004s :      2: substitution.replace_old_param
 6.46% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003858     2
82.09% :     0.003167s :      1: type_inference.infer
17.91% :     0.000691s :      1: type_inference.specialize
------[replace.]   0.000059     6
66.32% :     0.000039s :      4: replace.inline
33.68% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000148     6
93.29% :     0.000138s :      4: match.inline
 6.71% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000108   691
 1.38% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.15% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000001s :     10: predicate.addn_zero_filter
 1.93% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.28% :     0.000006s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.18% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.40% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.51% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000002s :     15: predicate.partial_eliminate
 1.28% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.36% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.61% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.01% :     0.000009s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.26% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.81% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.78% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000485    12
48.71% :     0.000236s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.59% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.70% :     0.000222s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017344    76
 0.25% :     0.000043s :      1: add_recomputation
 0.25% :     0.000044s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.53% :     0.000265s :      1: bootstrap
 0.14% :     0.000025s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000017s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000005s :      1: graph_reusing
26.69% :     0.004630s :      1: jit_opt_a
 0.76% :     0.000132s :      1: jit_opt_after_cconv
 0.28% :     0.000048s :      1: jit_opt_b
 2.32% :     0.000403s :      1: loop_unroll
 2.67% :     0.000463s :      1: mutable_eliminate
 4.37% :     0.000757s :     26: opt.transform.jit_opt_a
 0.25% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.74% :     0.000475s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000027s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000015s :      1: remove_dup_value
 1.39% :     0.000241s :      1: renormalize.infer
 1.29% :     0.000224s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000033s :      1: rewriter_after_opt_a
 0.37% :     0.000064s :      1: rewriter_before_opt_a
 0.41% :     0.000071s :      1: symbol_engine_optimizer
29.74% :     0.005157s :      1: task_emit
22.56% :     0.003914s :      1: type_inference
 0.26% :     0.000045s :      1: validate


TotalTime = 0.0138416, [30]
[bootstrap]: 0.00051372
[type_inference]: 0.00638965
[event_method]: 1.948e-05
[auto_monad]: 7.29e-05
[graph_reusing]: 6.33e-06
[pre_auto_parallel]: 2.41998e-06
[py_interpret_to_execute]: 4.263e-05
[rewriter_before_opt_a]: 8.024e-05
[expand_dump_flag]: 3.04999e-06
[jit_opt_a]: 0.00478472, [2]
    [Cycle 1]: 0.00152315, [27]
        [switch_simplify]: 9.831e-05
        [loop_unroll]: 3.326e-05
        [a_1]: 0.00062393
        [with_stream_mark]: 1.388e-05
        [recompute_prepare]: 7.23e-06
        [updatestate_depend_eliminate]: 3.23e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 5.59e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.24e-06
        [meta_shard_fg_expand]: 1.76e-06
        [get_grad_eliminate_]: 5.24998e-06
        [merge_forward]: 3.08e-06
        [cell_reuse_recompute_pass]: 8.90024e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.312e-05
        [j_node_and_user_rematch]: 8.89e-06
        [meta_fg_expand]: 1.99e-06
        [replace_old_param]: 8.40999e-06
        [inline_without_move]: 5.12999e-06
        [renormalize]: 0.00046642
        [add_forward_monad_depend]: 5.07e-06
        [auto_monad_grad]: 1.64e-06
        [auto_monad_eliminator]: 1.17e-05
        [cse]: 1.838e-05
        [replace_applicator]: 1.05e-05
    [Cycle 2]: 0.0003144, [27]
        [switch_simplify]: 5.75001e-06
        [loop_unroll]: 4.82e-06
        [a_1]: 9.511e-05
        [with_stream_mark]: 8.75001e-06
        [recompute_prepare]: 4.84003e-06
        [updatestate_depend_eliminate]: 2.39999e-06
        [updatestate_assign_eliminate]: 2.05002e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 4.80001e-06
        [updatestate_useless_node_eliminater]: 4.62e-06
        [accelerated_algorithm]: 5.06002e-06
        [meta_shard_fg_expand]: 1.02e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.71999e-06
        [cell_reuse_recompute_pass]: 1.26997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.213e-05
        [j_node_and_user_rematch]: 7.7e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 6.86999e-06
        [inline_without_move]: 4.65999e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.05001e-06
        [auto_monad_grad]: 7.10017e-07
        [auto_monad_eliminator]: 4.78001e-06
        [cse]: 1.036e-05
        [replace_applicator]: 4.79e-06
[py_interpret_to_execute_after_opt_a]: 7.49002e-06
[rewriter_after_opt_a]: 2.669e-05
[convert_after_rewriter]: 6.54001e-06
[order_py_execute_after_rewriter]: 4.52998e-06
[mutable_eliminate]: 0.00047913
[jit_opt_b]: 4.627e-05, [1]
    [Cycle 1]: 4.015e-05, [2]
        [frontend_op_eliminate]: 1.498e-05
        [inline_after_opt_a]: 1.427e-05
[cconv]: 2.257e-05
[loop_unroll]: 0.00041814
[jit_opt_after_cconv]: 0.00012868, [1]
    [Cycle 1]: 0.00012285, [11]
        [c_1]: 2.127e-05
        [parameter_eliminate]: 2.54001e-06
        [updatestate_depend_eliminate]: 4.74002e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 1.95001e-06
        [cse]: 1.486e-05
        [call_graph_tuple_transform]: 1.582e-05
        [tuple_list_get_item_eliminator]: 5.05999e-06
        [none_parameter_eliminate]: 1.10001e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.00001e-06
[remove_dup_value]: 1.135e-05
[partial_unused_args_eliminate]: 1.71e-06
[environ_conv]: 4.77e-06
[add_recomputation]: 4.02e-05
[cse_after_recomputation]: 1.99e-05, [1]
    [Cycle 1]: 1.448e-05, [1]
        [cse]: 8.85001e-06
[auto_monad_reorder]: 1.28e-05
[get_jit_bprop_graph]: 1.37999e-06
[rewriter_after_jit_bprop_graph]: 4.72e-06
[opt_after_jit_grad]: 0.00044486
[symbol_engine_optimizer]: 6.827e-05, [1]
    [Cycle 1]: 6.244e-05, [6]
        [build]: 2.66e-06
        [elim_shapecalc]: 7.45e-06
        [elim_not_effective]: 1.174e-05
        [opt_reshape]: 5.59e-06
        [fold_const_symbol]: 8e-06
        [renormalize]: 5.20027e-07
[validate]: 2.834e-05
Sums
    bootstrap                            :     0.000514s :  4.96%
    type_inference                       :     0.006390s : 61.65%
    event_method                         :     0.000019s :  0.19%
    auto_monad                           :     0.000073s :  0.70%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.41%
    rewriter_before_opt_a                :     0.000080s :  0.77%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000104s :  1.00%
    jit_opt_a.loop_unroll                :     0.000038s :  0.37%
    jit_opt_a.a_1                        :     0.000719s :  6.94%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000466s :  4.50%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.16%
    jit_opt_a.cse                        :     0.000029s :  0.28%
    jit_opt_a.replace_applicator         :     0.000015s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.07%
    rewriter_after_opt_a                 :     0.000027s :  0.26%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000479s :  4.62%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.14%
    cconv                                :     0.000023s :  0.22%
    loop_unroll                          :     0.000418s :  4.03%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000011s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000040s :  0.39%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000013s :  0.12%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000445s :  4.29%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000028s :  0.27%


Time group info:
------[substitution.]   0.000228    26
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 1.98% :     0.000005s :      3: substitution.graph_param_transform
84.08% :     0.000192s :      6: substitution.inline
 1.16% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.86% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.55% :     0.000004s :      2: substitution.replace_old_param
 3.08% :     0.000007s :      1: substitution.switch_simplify
 4.96% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006338     2
85.41% :     0.005413s :      1: type_inference.infer
14.59% :     0.000925s :      1: type_inference.specialize
------[replace.]   0.000116     9
51.78% :     0.000060s :      6: replace.inline
24.21% :     0.000028s :      1: replace.switch_simplify
24.01% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000205     9
91.97% :     0.000189s :      6: match.inline
 3.01% :     0.000006s :      1: match.switch_simplify
 5.02% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   801
 1.30% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.92% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     12: predicate.addn_check_dump
 1.29% :     0.000002s :     12: predicate.addn_zero_filter
 1.83% :     0.000002s :     12: predicate.arithmetic_simplify
 1.36% :     0.000002s :     12: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     12: predicate.compare_switch_simplify
 1.44% :     0.000002s :     12: predicate.depend_value_elim
 1.16% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.69% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_depend_swap
 6.63% :     0.000009s :     12: predicate.environ_get_eliminate
 1.11% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.76% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.36% :     0.000007s :     26: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000002s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     17: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.05% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.98% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     12: predicate.merge_addn
 1.18% :     0.000002s :     12: predicate.minmaximum_grad
 1.34% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     17: predicate.partial_eliminate
 1.17% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     12: predicate.reduce_eliminate
 1.41% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     20: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     12: predicate.reshape_eliminate
 1.18% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.59% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.53% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.53% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.55% :     0.000003s :     20: predicate.switch_defer_inline
 2.19% :     0.000003s :     20: predicate.switch_layer_defer_inline
 7.79% :     0.000010s :     65: predicate.switch_simplify
 1.20% :     0.000002s :     12: predicate.tile_eliminate
 1.16% :     0.000002s :     12: predicate.transpose_eliminate
 1.30% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.18% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.52% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000712    15
52.54% :     0.000374s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.00% :     0.000071s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.46% :     0.000267s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015340    72
 0.28% :     0.000043s :      1: add_recomputation
 0.50% :     0.000076s :      1: auto_monad
 0.10% :     0.000015s :      1: auto_monad_reorder
 3.43% :     0.000527s :      1: bootstrap
 0.16% :     0.000025s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
31.21% :     0.004787s :      1: jit_opt_a
 0.86% :     0.000131s :      1: jit_opt_after_cconv
 0.32% :     0.000049s :      1: jit_opt_b
 2.78% :     0.000426s :      1: loop_unroll
 3.18% :     0.000487s :      1: mutable_eliminate
 6.32% :     0.000970s :     26: opt.transform.jit_opt_a
 0.29% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000023s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.95% :     0.000452s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.30% :     0.000045s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000014s :      1: remove_dup_value
 1.54% :     0.000237s :      1: renormalize.infer
 1.45% :     0.000223s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000029s :      1: rewriter_after_opt_a
 0.54% :     0.000083s :      1: rewriter_before_opt_a
 0.46% :     0.000071s :      1: symbol_engine_optimizer
41.72% :     0.006400s :      1: type_inference


TotalTime = 0.0211966, [33]
[bootstrap]: 0.00034017
[type_inference]: 0.00529338
[event_method]: 1.941e-05
[auto_monad]: 6.768e-05
[graph_reusing]: 6.53998e-06
[pre_auto_parallel]: 2.72001e-06
[py_interpret_to_execute]: 4.107e-05
[rewriter_before_opt_a]: 7.859e-05
[expand_dump_flag]: 3.40998e-06
[jit_opt_a]: 0.00627756, [2]
    [Cycle 1]: 0.00184102, [27]
        [switch_simplify]: 6.068e-05
        [loop_unroll]: 3.009e-05
        [a_1]: 0.00058106
        [with_stream_mark]: 2.089e-05
        [recompute_prepare]: 7.63001e-06
        [updatestate_depend_eliminate]: 4.38001e-06
        [updatestate_assign_eliminate]: 2.91999e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 2.24001e-06
        [specialize_transform]: 5.78002e-06
        [updatestate_useless_node_eliminater]: 4.96002e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 2.07001e-06
        [get_grad_eliminate_]: 5.89e-06
        [merge_forward]: 3.73001e-06
        [cell_reuse_recompute_pass]: 1.46002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.488e-05
        [j_node_and_user_rematch]: 9.44e-06
        [meta_fg_expand]: 2.48e-06
        [replace_old_param]: 8.79e-06
        [inline_without_move]: 5.39e-06
        [renormalize]: 0.00080618
        [add_forward_monad_depend]: 5.73997e-06
        [auto_monad_grad]: 2.69999e-06
        [auto_monad_eliminator]: 1.526e-05
        [cse]: 3.28e-05
        [replace_applicator]: 1.577e-05
    [Cycle 2]: 0.00055222, [27]
        [switch_simplify]: 6.19001e-06
        [loop_unroll]: 4.67e-05
        [a_1]: 9.359e-05
        [with_stream_mark]: 1.211e-05
        [recompute_prepare]: 5.13002e-06
        [updatestate_depend_eliminate]: 2.89999e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 4.88001e-06
        [updatestate_useless_node_eliminater]: 5.33002e-06
        [accelerated_algorithm]: 5.10001e-06
        [meta_shard_fg_expand]: 1.53002e-06
        [get_grad_eliminate_]: 5.21002e-06
        [merge_forward]: 5.64e-06
        [cell_reuse_recompute_pass]: 3.13e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.948e-05
        [j_node_and_user_rematch]: 9.22999e-06
        [meta_fg_expand]: 2.58e-06
        [replace_old_param]: 8.35001e-06
        [inline_without_move]: 5.57999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.70002e-06
        [auto_monad_grad]: 1.07e-06
        [auto_monad_eliminator]: 7.30998e-06
        [cse]: 1.767e-05
        [replace_applicator]: 5.32999e-06
[py_interpret_to_execute_after_opt_a]: 1.258e-05
[rewriter_after_opt_a]: 3.827e-05
[convert_after_rewriter]: 7.25e-06
[order_py_execute_after_rewriter]: 5.17e-06
[mutable_eliminate]: 0.00068755
[jit_opt_b]: 5.466e-05, [1]
    [Cycle 1]: 4.69e-05, [2]
        [frontend_op_eliminate]: 1.828e-05
        [inline_after_opt_a]: 1.666e-05
[cconv]: 2.838e-05
[loop_unroll]: 0.00049307
[jit_opt_after_cconv]: 0.00015524, [1]
    [Cycle 1]: 0.00014787, [11]
        [c_1]: 2.27e-05
        [parameter_eliminate]: 3.32002e-06
        [updatestate_depend_eliminate]: 6.98e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [cse]: 2.41e-05
        [call_graph_tuple_transform]: 2.009e-05
        [tuple_list_get_item_eliminator]: 5.63002e-06
        [none_parameter_eliminate]: 1.76003e-06
        [renormalize]: 1.07e-06
        [switch_simplify]: 6.32001e-06
[remove_dup_value]: 1.622e-05
[partial_unused_args_eliminate]: 2.48e-06
[environ_conv]: 6.13002e-06
[add_recomputation]: 5.232e-05
[cse_after_recomputation]: 2.212e-05, [1]
    [Cycle 1]: 1.617e-05, [1]
        [cse]: 9.89001e-06
[auto_monad_reorder]: 1.641e-05
[get_jit_bprop_graph]: 2.81999e-06
[rewriter_after_jit_bprop_graph]: 6.53e-06
[opt_after_jit_grad]: 0.00050186
[symbol_engine_optimizer]: 7.887e-05, [1]
    [Cycle 1]: 7.154e-05, [6]
        [build]: 4.02e-06
        [elim_shapecalc]: 7.97e-06
        [elim_not_effective]: 1.422e-05
        [opt_reshape]: 6.67002e-06
        [fold_const_symbol]: 1.02e-05
        [renormalize]: 8.60018e-07
[validate]: 4.369e-05
[backend_pass]: 9.79984e-07
[task_emit]: 0.00660079
[execute]: 8.13999e-06
Sums
    bootstrap                            :     0.000340s :  2.06%
    type_inference                       :     0.005293s : 32.06%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000068s :  0.41%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000041s :  0.25%
    rewriter_before_opt_a                :     0.000079s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.40%
    jit_opt_a.loop_unroll                :     0.000077s :  0.47%
    jit_opt_a.a_1                        :     0.000675s :  4.09%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000009s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000806s :  4.88%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.14%
    jit_opt_a.cse                        :     0.000050s :  0.31%
    jit_opt_a.replace_applicator         :     0.000021s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000688s :  4.16%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.10%
    cconv                                :     0.000028s :  0.17%
    loop_unroll                          :     0.000493s :  2.99%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000016s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000052s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000502s :  3.04%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006601s : 39.97%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000237    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.88% :     0.000002s :      2: substitution.fold_const_symbol
 2.35% :     0.000006s :      3: substitution.graph_param_transform
84.45% :     0.000200s :      4: substitution.inline
 1.67% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.24% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.95% :     0.000005s :      2: substitution.replace_old_param
 5.62% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005220     2
81.93% :     0.004277s :      1: type_inference.infer
18.07% :     0.000943s :      1: type_inference.specialize
------[replace.]   0.000078     6
67.51% :     0.000053s :      4: replace.inline
32.49% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000210     6
94.20% :     0.000198s :      4: match.inline
 5.80% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.60% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.23% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 2.08% :     0.000003s :     10: predicate.arithmetic_simplify
 1.15% :     0.000001s :     10: predicate.cast_eliminate
 0.86% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.24% :     0.000002s :     10: predicate.depend_value_elim
 1.44% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.17% :     0.000001s :      6: predicate.get_grad_eliminate
 0.34% :     0.000000s :      3: predicate.graph_param_transform
 5.73% :     0.000007s :     22: predicate.inline
 1.10% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.04% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.67% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.53% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.56% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.29% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.86% :     0.000010s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.48% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.96% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000935    12
63.29% :     0.000592s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 3.24% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
33.47% :     0.000313s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023054    76
 0.24% :     0.000055s :      1: add_recomputation
 0.31% :     0.000071s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.56% :     0.000359s :      1: bootstrap
 0.14% :     0.000031s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
27.25% :     0.006282s :      1: jit_opt_a
 0.69% :     0.000158s :      1: jit_opt_after_cconv
 0.25% :     0.000057s :      1: jit_opt_b
 2.18% :     0.000502s :      1: loop_unroll
 3.03% :     0.000698s :      1: mutable_eliminate
 4.10% :     0.000944s :     26: opt.transform.jit_opt_a
 0.22% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000028s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000510s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000044s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000019s :      1: remove_dup_value
 1.80% :     0.000414s :      1: renormalize.infer
 1.66% :     0.000382s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000041s :      1: rewriter_after_opt_a
 0.35% :     0.000081s :      1: rewriter_before_opt_a
 0.35% :     0.000081s :      1: symbol_engine_optimizer
28.71% :     0.006620s :      1: task_emit
23.03% :     0.005309s :      1: type_inference
 0.28% :     0.000065s :      1: validate


TotalTime = 0.0140318, [30]
[bootstrap]: 0.00025923
[type_inference]: 0.00629145
[event_method]: 1.953e-05
[auto_monad]: 8.129e-05
[graph_reusing]: 6.15002e-06
[pre_auto_parallel]: 2.84001e-06
[py_interpret_to_execute]: 4.731e-05
[rewriter_before_opt_a]: 8.482e-05
[expand_dump_flag]: 3.40998e-06
[jit_opt_a]: 0.00520776, [2]
    [Cycle 1]: 0.00170739, [27]
        [switch_simplify]: 0.00010959
        [loop_unroll]: 3.322e-05
        [a_1]: 0.00065219
        [with_stream_mark]: 1.728e-05
        [recompute_prepare]: 6.76e-06
        [updatestate_depend_eliminate]: 3.53e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [parameter_eliminate]: 1.72999e-06
        [specialize_transform]: 6.11e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.91998e-06
        [get_grad_eliminate_]: 5.00999e-06
        [merge_forward]: 3.75998e-06
        [cell_reuse_recompute_pass]: 1.12999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.356e-05
        [j_node_and_user_rematch]: 9.20001e-06
        [meta_fg_expand]: 2.06998e-06
        [replace_old_param]: 8.85999e-06
        [inline_without_move]: 5.27001e-06
        [renormalize]: 0.00059091
        [add_forward_monad_depend]: 5.44998e-06
        [auto_monad_grad]: 2.29999e-06
        [auto_monad_eliminator]: 1.409e-05
        [cse]: 1.815e-05
        [replace_applicator]: 1.032e-05
    [Cycle 2]: 0.00031106, [27]
        [switch_simplify]: 5.30999e-06
        [loop_unroll]: 4.81002e-06
        [a_1]: 8.665e-05
        [with_stream_mark]: 8.32998e-06
        [recompute_prepare]: 4.77998e-06
        [updatestate_depend_eliminate]: 2.56e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 4.90999e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 5.00999e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.32e-06
        [merge_forward]: 2.41998e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.175e-05
        [j_node_and_user_rematch]: 7.17002e-06
        [meta_fg_expand]: 1.45999e-06
        [replace_old_param]: 7.22002e-06
        [inline_without_move]: 4.52e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.41998e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 4.95001e-06
        [cse]: 1.015e-05
        [replace_applicator]: 5.04e-06
[py_interpret_to_execute_after_opt_a]: 1.062e-05
[rewriter_after_opt_a]: 3.168e-05
[convert_after_rewriter]: 7.23999e-06
[order_py_execute_after_rewriter]: 5.35001e-06
[mutable_eliminate]: 0.00052408
[jit_opt_b]: 4.926e-05, [1]
    [Cycle 1]: 4.257e-05, [2]
        [frontend_op_eliminate]: 1.641e-05
        [inline_after_opt_a]: 1.433e-05
[cconv]: 2.379e-05
[loop_unroll]: 0.00042083
[jit_opt_after_cconv]: 0.00013203, [1]
    [Cycle 1]: 0.00012592, [11]
        [c_1]: 2.013e-05
        [parameter_eliminate]: 2.49999e-06
        [updatestate_depend_eliminate]: 5.35999e-06
        [updatestate_assign_eliminate]: 2.40002e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [cse]: 1.504e-05
        [call_graph_tuple_transform]: 1.769e-05
        [tuple_list_get_item_eliminator]: 5.25001e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 5.24e-06
[remove_dup_value]: 1.29e-05
[partial_unused_args_eliminate]: 1.98002e-06
[environ_conv]: 4.94998e-06
[add_recomputation]: 4.548e-05
[cse_after_recomputation]: 2.006e-05, [1]
    [Cycle 1]: 1.463e-05, [1]
        [cse]: 8.94e-06
[auto_monad_reorder]: 1.484e-05
[get_jit_bprop_graph]: 1.62999e-06
[rewriter_after_jit_bprop_graph]: 5.20001e-06
[opt_after_jit_grad]: 0.00045438
[symbol_engine_optimizer]: 7.054e-05, [1]
    [Cycle 1]: 6.424e-05, [6]
        [build]: 3.04001e-06
        [elim_shapecalc]: 6.88e-06
        [elim_not_effective]: 1.231e-05
        [opt_reshape]: 5.62001e-06
        [fold_const_symbol]: 8.50001e-06
        [renormalize]: 4.60015e-07
[validate]: 3.072e-05
Sums
    bootstrap                            :     0.000259s :  2.52%
    type_inference                       :     0.006291s : 61.20%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000081s :  0.79%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000047s :  0.46%
    rewriter_before_opt_a                :     0.000085s :  0.83%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000115s :  1.12%
    jit_opt_a.loop_unroll                :     0.000038s :  0.37%
    jit_opt_a.a_1                        :     0.000739s :  7.19%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000591s :  5.75%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.19%
    jit_opt_a.cse                        :     0.000028s :  0.28%
    jit_opt_a.replace_applicator         :     0.000015s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.10%
    rewriter_after_opt_a                 :     0.000032s :  0.31%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000524s :  5.10%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.14%
    cconv                                :     0.000024s :  0.23%
    loop_unroll                          :     0.000421s :  4.09%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000045s :  0.44%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000454s :  4.42%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.30%


Time group info:
------[substitution.]   0.000256    26
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.01% :     0.000005s :      3: substitution.graph_param_transform
84.38% :     0.000216s :      6: substitution.inline
 1.13% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.67% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.55% :     0.000004s :      2: substitution.replace_old_param
 3.17% :     0.000008s :      1: substitution.switch_simplify
 4.85% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006224     2
83.91% :     0.005223s :      1: type_inference.infer
16.09% :     0.001002s :      1: type_inference.specialize
------[replace.]   0.000121     9
50.42% :     0.000061s :      6: replace.inline
25.89% :     0.000031s :      1: replace.switch_simplify
23.69% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000231     9
91.94% :     0.000213s :      6: match.inline
 3.18% :     0.000007s :      1: match.switch_simplify
 4.89% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   801
 1.39% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.11% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     12: predicate.addn_check_dump
 1.39% :     0.000002s :     12: predicate.addn_zero_filter
 1.74% :     0.000002s :     12: predicate.arithmetic_simplify
 1.63% :     0.000002s :     12: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.18% :     0.000002s :     12: predicate.compare_switch_simplify
 1.64% :     0.000002s :     12: predicate.depend_value_elim
 1.16% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.33% :     0.000002s :     12: predicate.environ_get_eliminate
 1.15% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.33% :     0.000007s :     26: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.02% :     0.000001s :      6: predicate.less_batch_normalization
 1.75% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     17: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.33% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     12: predicate.merge_addn
 1.14% :     0.000001s :     12: predicate.minmaximum_grad
 1.42% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     17: predicate.partial_eliminate
 1.23% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     12: predicate.reduce_eliminate
 1.51% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     20: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     12: predicate.reshape_eliminate
 1.23% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     12: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.76% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.67% :     0.000003s :     20: predicate.switch_defer_inline
 2.34% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.55% :     0.000011s :     65: predicate.switch_simplify
 1.34% :     0.000002s :     12: predicate.tile_eliminate
 1.25% :     0.000002s :     12: predicate.transpose_eliminate
 1.59% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.55% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     12: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000814    15
54.29% :     0.000442s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.60% :     0.000086s :      1: func_graph_cloner_run.FuncGraphClonerNode
35.11% :     0.000286s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015683    72
 0.31% :     0.000048s :      1: add_recomputation
 0.54% :     0.000085s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.73% :     0.000271s :      1: bootstrap
 0.17% :     0.000027s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.23% :     0.005211s :      1: jit_opt_a
 0.86% :     0.000135s :      1: jit_opt_after_cconv
 0.33% :     0.000052s :      1: jit_opt_b
 2.73% :     0.000428s :      1: loop_unroll
 3.39% :     0.000532s :      1: mutable_eliminate
 6.38% :     0.001000s :     26: opt.transform.jit_opt_a
 0.29% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.94% :     0.000462s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.32% :     0.000050s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000015s :      1: remove_dup_value
 2.03% :     0.000318s :      1: renormalize.infer
 1.70% :     0.000266s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000035s :      1: rewriter_after_opt_a
 0.56% :     0.000088s :      1: rewriter_before_opt_a
 0.47% :     0.000073s :      1: symbol_engine_optimizer
40.22% :     0.006308s :      1: type_inference


TotalTime = 0.0166049, [33]
[bootstrap]: 0.00024616
[type_inference]: 0.00396456
[event_method]: 1.529e-05
[auto_monad]: 5.738e-05
[graph_reusing]: 4.97e-06
[pre_auto_parallel]: 2.11e-06
[py_interpret_to_execute]: 2.584e-05
[rewriter_before_opt_a]: 6.429e-05
[expand_dump_flag]: 2.94999e-06
[jit_opt_a]: 0.00465473, [2]
    [Cycle 1]: 0.00134547, [27]
        [switch_simplify]: 4.815e-05
        [loop_unroll]: 2.86e-05
        [a_1]: 0.0004848
        [with_stream_mark]: 1.423e-05
        [recompute_prepare]: 6.88e-06
        [updatestate_depend_eliminate]: 3.09001e-06
        [updatestate_assign_eliminate]: 2.31998e-06
        [updatestate_loads_eliminate]: 2.68998e-06
        [parameter_eliminate]: 1.59e-06
        [specialize_transform]: 5.70001e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 5.48002e-06
        [meta_shard_fg_expand]: 1.55999e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 2.91e-06
        [cell_reuse_recompute_pass]: 9.30013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.339e-05
        [j_node_and_user_rematch]: 8.72998e-06
        [meta_fg_expand]: 1.82001e-06
        [replace_old_param]: 8.69998e-06
        [inline_without_move]: 5.25999e-06
        [renormalize]: 0.0004795
        [add_forward_monad_depend]: 4.89e-06
        [auto_monad_grad]: 1.88002e-06
        [auto_monad_eliminator]: 1.23e-05
        [cse]: 1.817e-05
        [replace_applicator]: 1.112e-05
    [Cycle 2]: 0.00031614, [27]
        [switch_simplify]: 5.59998e-06
        [loop_unroll]: 4.95001e-06
        [a_1]: 9.067e-05
        [with_stream_mark]: 8.32998e-06
        [recompute_prepare]: 4.85001e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 1.96998e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 1.02998e-06
        [specialize_transform]: 5.59e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 1.20001e-06
        [get_grad_eliminate_]: 4.80999e-06
        [merge_forward]: 2.24001e-06
        [cell_reuse_recompute_pass]: 1.55001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.144e-05
        [j_node_and_user_rematch]: 7.36001e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 7.1e-06
        [inline_without_move]: 4.48999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 5.25999e-06
        [cse]: 9.89999e-06
        [replace_applicator]: 4.74e-06
[py_interpret_to_execute_after_opt_a]: 9.08002e-06
[rewriter_after_opt_a]: 2.959e-05
[convert_after_rewriter]: 6.14001e-06
[order_py_execute_after_rewriter]: 5.32001e-06
[mutable_eliminate]: 0.00048601
[jit_opt_b]: 4.647e-05, [1]
    [Cycle 1]: 4.006e-05, [2]
        [frontend_op_eliminate]: 1.492e-05
        [inline_after_opt_a]: 1.421e-05
[cconv]: 2.294e-05
[loop_unroll]: 0.00040576
[jit_opt_after_cconv]: 0.00013024, [1]
    [Cycle 1]: 0.00012388, [11]
        [c_1]: 1.987e-05
        [parameter_eliminate]: 2.34001e-06
        [updatestate_depend_eliminate]: 4.94e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [cse]: 1.594e-05
        [call_graph_tuple_transform]: 1.657e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.12e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.25999e-06
[remove_dup_value]: 1.187e-05
[partial_unused_args_eliminate]: 2.12001e-06
[environ_conv]: 4.52e-06
[add_recomputation]: 4.131e-05
[cse_after_recomputation]: 1.916e-05, [1]
    [Cycle 1]: 1.396e-05, [1]
        [cse]: 8.32998e-06
[auto_monad_reorder]: 1.232e-05
[get_jit_bprop_graph]: 1.85001e-06
[rewriter_after_jit_bprop_graph]: 4.63999e-06
[opt_after_jit_grad]: 0.00051081
[symbol_engine_optimizer]: 7.489e-05, [1]
    [Cycle 1]: 6.849e-05, [6]
        [build]: 3.48e-06
        [elim_shapecalc]: 7.87998e-06
        [elim_not_effective]: 1.32e-05
        [opt_reshape]: 6.04999e-06
        [fold_const_symbol]: 9.15999e-06
        [renormalize]: 4.60015e-07
[validate]: 3.148e-05
[backend_pass]: 1.12e-06
[task_emit]: 0.00553406
[execute]: 6.64001e-06
Sums
    bootstrap                            :     0.000246s :  1.89%
    type_inference                       :     0.003965s : 30.38%
    event_method                         :     0.000015s :  0.12%
    auto_monad                           :     0.000057s :  0.44%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.20%
    rewriter_before_opt_a                :     0.000064s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.26%
    jit_opt_a.a_1                        :     0.000575s :  4.41%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000480s :  3.67%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000028s :  0.22%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000030s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000486s :  3.72%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000023s :  0.18%
    loop_unroll                          :     0.000406s :  3.11%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000041s :  0.32%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000511s :  3.91%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005534s : 42.40%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000168    23
 1.16% :     0.000002s :      2: substitution.elim_not_effective
 0.86% :     0.000001s :      2: substitution.fold_const_symbol
 2.70% :     0.000005s :      3: substitution.graph_param_transform
81.58% :     0.000137s :      4: substitution.inline
 1.77% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.87% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.10% :     0.000004s :      2: substitution.replace_old_param
 6.95% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003914     2
81.99% :     0.003209s :      1: type_inference.infer
18.01% :     0.000705s :      1: type_inference.specialize
------[replace.]   0.000058     6
66.03% :     0.000039s :      4: replace.inline
33.97% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000145     6
92.71% :     0.000135s :      4: match.inline
 7.29% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 2.19% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000001s :     10: predicate.depend_value_elim
 1.24% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.13% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.19% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.88% :     0.000002s :     15: predicate.load_eliminater
 1.65% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.35% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.83% :     0.000002s :     18: predicate.replace_applicator
 0.80% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.75% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 1.12% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.83% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.60% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.42% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000515    12
52.37% :     0.000269s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.10% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.54% :     0.000219s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017967    76
 0.25% :     0.000044s :      1: add_recomputation
 0.34% :     0.000061s :      1: auto_monad
 0.08% :     0.000015s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.43% :     0.000257s :      1: bootstrap
 0.14% :     0.000025s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
25.92% :     0.004657s :      1: jit_opt_a
 0.74% :     0.000133s :      1: jit_opt_after_cconv
 0.27% :     0.000049s :      1: jit_opt_b
 2.30% :     0.000413s :      1: loop_unroll
 2.75% :     0.000494s :      1: mutable_eliminate
 4.31% :     0.000774s :     26: opt.transform.jit_opt_a
 0.24% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.89% :     0.000519s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000028s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 1.34% :     0.000241s :      1: renormalize.infer
 1.29% :     0.000231s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000034s :      1: rewriter_after_opt_a
 0.37% :     0.000067s :      1: rewriter_before_opt_a
 0.43% :     0.000078s :      1: symbol_engine_optimizer
30.85% :     0.005543s :      1: task_emit
22.12% :     0.003975s :      1: type_inference
 0.28% :     0.000050s :      1: validate


TotalTime = 0.0217554, [33]
[bootstrap]: 0.0004598
[type_inference]: 0.00507951
[event_method]: 1.954e-05
[auto_monad]: 6.898e-05
[graph_reusing]: 5.64998e-06
[pre_auto_parallel]: 2.70997e-06
[py_interpret_to_execute]: 3.982e-05
[rewriter_before_opt_a]: 7.684e-05
[expand_dump_flag]: 3.30998e-06
[jit_opt_a]: 0.00682886, [2]
    [Cycle 1]: 0.00200083, [27]
        [switch_simplify]: 6.954e-05
        [loop_unroll]: 3.108e-05
        [a_1]: 0.00059604
        [with_stream_mark]: 2.007e-05
        [recompute_prepare]: 9.37001e-06
        [updatestate_depend_eliminate]: 4.38999e-06
        [updatestate_assign_eliminate]: 3.59002e-06
        [updatestate_loads_eliminate]: 3.61001e-06
        [parameter_eliminate]: 2.17001e-06
        [specialize_transform]: 7.05998e-06
        [updatestate_useless_node_eliminater]: 5.59e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 2.69999e-06
        [get_grad_eliminate_]: 6.26e-06
        [merge_forward]: 4.37e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.619e-05
        [j_node_and_user_rematch]: 1.165e-05
        [meta_fg_expand]: 2.76999e-06
        [replace_old_param]: 1.029e-05
        [inline_without_move]: 5.86e-06
        [renormalize]: 0.00091143
        [add_forward_monad_depend]: 1.228e-05
        [auto_monad_grad]: 2.35002e-06
        [auto_monad_eliminator]: 1.762e-05
        [cse]: 2.99e-05
        [replace_applicator]: 1.49e-05
    [Cycle 2]: 0.0003607, [27]
        [switch_simplify]: 6.11e-06
        [loop_unroll]: 5.32001e-06
        [a_1]: 0.00010464
        [with_stream_mark]: 1.251e-05
        [recompute_prepare]: 5.94e-06
        [updatestate_depend_eliminate]: 2.91999e-06
        [updatestate_assign_eliminate]: 2.06998e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 5.66e-06
        [updatestate_useless_node_eliminater]: 6.11998e-06
        [accelerated_algorithm]: 4.68001e-06
        [meta_shard_fg_expand]: 1.44998e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 3.48999e-06
        [cell_reuse_recompute_pass]: 2.96001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.64e-05
        [j_node_and_user_rematch]: 1.045e-05
        [meta_fg_expand]: 2.29001e-06
        [replace_old_param]: 8.45001e-06
        [inline_without_move]: 5.62001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 1.02e-06
        [auto_monad_eliminator]: 5.79999e-06
        [cse]: 1.12e-05
        [replace_applicator]: 5.29e-06
[py_interpret_to_execute_after_opt_a]: 1.067e-05
[rewriter_after_opt_a]: 5.556e-05
[convert_after_rewriter]: 7.77e-06
[order_py_execute_after_rewriter]: 5.24e-06
[mutable_eliminate]: 0.00071146
[jit_opt_b]: 5.457e-05, [1]
    [Cycle 1]: 4.666e-05, [2]
        [frontend_op_eliminate]: 1.843e-05
        [inline_after_opt_a]: 1.579e-05
[cconv]: 3.02e-05
[loop_unroll]: 0.00045107
[jit_opt_after_cconv]: 0.0001662, [1]
    [Cycle 1]: 0.00015922, [11]
        [c_1]: 2.157e-05
        [parameter_eliminate]: 3.72998e-06
        [updatestate_depend_eliminate]: 1.947e-05
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [cse]: 2.209e-05
        [call_graph_tuple_transform]: 2.166e-05
        [tuple_list_get_item_eliminator]: 5.70001e-06
        [none_parameter_eliminate]: 1.51002e-06
        [renormalize]: 7.09988e-07
        [switch_simplify]: 5.58002e-06
[remove_dup_value]: 1.53e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 6.17999e-06
[add_recomputation]: 5.131e-05
[cse_after_recomputation]: 2.098e-05, [1]
    [Cycle 1]: 1.523e-05, [1]
        [cse]: 9.19e-06
[auto_monad_reorder]: 1.717e-05
[get_jit_bprop_graph]: 2.43e-06
[rewriter_after_jit_bprop_graph]: 5.71e-06
[opt_after_jit_grad]: 0.00047511
[symbol_engine_optimizer]: 7.532e-05, [1]
    [Cycle 1]: 6.909e-05, [6]
        [build]: 4.17e-06
        [elim_shapecalc]: 7.5e-06
        [elim_not_effective]: 1.339e-05
        [opt_reshape]: 5.91998e-06
        [fold_const_symbol]: 9.76e-06
        [renormalize]: 8.79983e-07
[validate]: 4.367e-05
[backend_pass]: 8.80013e-07
[task_emit]: 0.00673106
[execute]: 8.79e-06
Sums
    bootstrap                            :     0.000460s :  2.77%
    type_inference                       :     0.005080s : 30.55%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000069s :  0.41%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.24%
    rewriter_before_opt_a                :     0.000077s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.45%
    jit_opt_a.loop_unroll                :     0.000036s :  0.22%
    jit_opt_a.a_1                        :     0.000701s :  4.21%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000022s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000911s :  5.48%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.14%
    jit_opt_a.cse                        :     0.000041s :  0.25%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000056s :  0.33%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000711s :  4.28%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000030s :  0.18%
    loop_unroll                          :     0.000451s :  2.71%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000019s :  0.12%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000051s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000475s :  2.86%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006731s : 40.48%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000229    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 3.12% :     0.000007s :      3: substitution.graph_param_transform
82.74% :     0.000189s :      4: substitution.inline
 2.18% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.46% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000004s :      2: substitution.replace_old_param
 6.10% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005003     2
81.30% :     0.004067s :      1: type_inference.infer
18.70% :     0.000936s :      1: type_inference.specialize
------[replace.]   0.000080     6
65.98% :     0.000053s :      4: replace.inline
34.02% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     6
93.53% :     0.000187s :      4: match.inline
 6.47% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.75% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.24% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 2.49% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000002s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.46% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.08% :     0.000001s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.64% :     0.000007s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.74% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.24% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.68% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.32% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000002s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 2.49% :     0.000003s :      3: predicate.mutable_eliminate
 0.37% :     0.000000s :      3: predicate.opt_reshape
 2.07% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.32% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.77% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.48% :     0.000002s :     18: predicate.replace_applicator
 0.45% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000001s :     10: predicate.same_eliminate
 0.69% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.14% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.41% :     0.000010s :     53: predicate.switch_simplify
 1.53% :     0.000002s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.61% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.46% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.83% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000668    12
46.97% :     0.000314s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.63% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.40% :     0.000323s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023717    76
 0.23% :     0.000054s :      1: add_recomputation
 0.31% :     0.000072s :      1: auto_monad
 0.08% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 2.02% :     0.000479s :      1: bootstrap
 0.14% :     0.000033s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
28.81% :     0.006834s :      1: jit_opt_a
 0.71% :     0.000169s :      1: jit_opt_after_cconv
 0.24% :     0.000058s :      1: jit_opt_b
 1.94% :     0.000460s :      1: loop_unroll
 3.04% :     0.000721s :      1: mutable_eliminate
 3.99% :     0.000947s :     26: opt.transform.jit_opt_a
 0.22% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.04% :     0.000483s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000043s :      1: py_interpret_to_execute
 0.05% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 2.01% :     0.000477s :      1: renormalize.infer
 1.79% :     0.000425s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000059s :      1: rewriter_after_opt_a
 0.34% :     0.000080s :      1: rewriter_before_opt_a
 0.33% :     0.000078s :      1: symbol_engine_optimizer
28.44% :     0.006746s :      1: task_emit
21.48% :     0.005095s :      1: type_inference
 0.27% :     0.000065s :      1: validate


TotalTime = 0.0230657, [33]
[bootstrap]: 0.00025005
[type_inference]: 0.00756169
[event_method]: 1.748e-05
[auto_monad]: 6.519e-05
[graph_reusing]: 6.80002e-06
[pre_auto_parallel]: 3.18e-06
[py_interpret_to_execute]: 3.231e-05
[rewriter_before_opt_a]: 7.508e-05
[expand_dump_flag]: 3.05998e-06
[jit_opt_a]: 0.00704273, [2]
    [Cycle 1]: 0.00163004, [27]
        [switch_simplify]: 5.371e-05
        [loop_unroll]: 3.013e-05
        [a_1]: 0.00061569
        [with_stream_mark]: 1.847e-05
        [recompute_prepare]: 7.63001e-06
        [updatestate_depend_eliminate]: 4.87998e-06
        [updatestate_assign_eliminate]: 3.04001e-06
        [updatestate_loads_eliminate]: 2.52001e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 1.38e-05
        [updatestate_useless_node_eliminater]: 5.44998e-06
        [accelerated_algorithm]: 5.87001e-06
        [meta_shard_fg_expand]: 2.04999e-06
        [get_grad_eliminate_]: 5.56002e-06
        [merge_forward]: 3.93001e-06
        [cell_reuse_recompute_pass]: 1.24998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.49e-05
        [j_node_and_user_rematch]: 9.82999e-06
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 9.45001e-06
        [inline_without_move]: 5.15999e-06
        [renormalize]: 0.00058275
        [add_forward_monad_depend]: 5.04e-06
        [auto_monad_grad]: 2.44001e-06
        [auto_monad_eliminator]: 1.332e-05
        [cse]: 2.704e-05
        [replace_applicator]: 1.123e-05
    [Cycle 2]: 0.00032406, [27]
        [switch_simplify]: 5.74e-06
        [loop_unroll]: 5.24e-06
        [a_1]: 9.399e-05
        [with_stream_mark]: 8.54e-06
        [recompute_prepare]: 5.20001e-06
        [updatestate_depend_eliminate]: 2.63e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 5.02999e-06
        [updatestate_useless_node_eliminater]: 5.10001e-06
        [accelerated_algorithm]: 5.29998e-06
        [meta_shard_fg_expand]: 1.17999e-06
        [get_grad_eliminate_]: 4.53001e-06
        [merge_forward]: 2.35002e-06
        [cell_reuse_recompute_pass]: 1.84998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.18e-05
        [j_node_and_user_rematch]: 7.77e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 1.212e-05
        [inline_without_move]: 4.79e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 4.62998e-06
        [cse]: 1.036e-05
        [replace_applicator]: 5.02e-06
[py_interpret_to_execute_after_opt_a]: 8.78001e-06
[rewriter_after_opt_a]: 3.347e-05
[convert_after_rewriter]: 6.53003e-06
[order_py_execute_after_rewriter]: 4.92e-06
[mutable_eliminate]: 0.00055086
[jit_opt_b]: 4.953e-05, [1]
    [Cycle 1]: 4.255e-05, [2]
        [frontend_op_eliminate]: 1.615e-05
        [inline_after_opt_a]: 1.467e-05
[cconv]: 2.542e-05
[loop_unroll]: 0.00041033
[jit_opt_after_cconv]: 0.00014606, [1]
    [Cycle 1]: 0.00014002, [11]
        [c_1]: 2.058e-05
        [parameter_eliminate]: 2.47001e-06
        [updatestate_depend_eliminate]: 5.37999e-06
        [updatestate_assign_eliminate]: 2.39999e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [cse]: 2.352e-05
        [call_graph_tuple_transform]: 1.948e-05
        [tuple_list_get_item_eliminator]: 5.52001e-06
        [none_parameter_eliminate]: 1.89e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.45001e-06
[remove_dup_value]: 1.297e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 4.74e-06
[add_recomputation]: 4.665e-05
[cse_after_recomputation]: 2.591e-05, [1]
    [Cycle 1]: 1.532e-05, [1]
        [cse]: 9.50001e-06
[auto_monad_reorder]: 1.467e-05
[get_jit_bprop_graph]: 1.82001e-06
[rewriter_after_jit_bprop_graph]: 5.05999e-06
[opt_after_jit_grad]: 0.00045624
[symbol_engine_optimizer]: 7.18e-05, [1]
    [Cycle 1]: 6.549e-05, [6]
        [build]: 3.28998e-06
        [elim_shapecalc]: 7.02002e-06
        [elim_not_effective]: 1.244e-05
        [opt_reshape]: 5.54998e-06
        [fold_const_symbol]: 8.78001e-06
        [renormalize]: 6.99976e-07
[validate]: 3.295e-05
[backend_pass]: 1.23002e-06
[task_emit]: 0.00590764
[execute]: 7.61999e-06
Sums
    bootstrap                            :     0.000250s :  1.44%
    type_inference                       :     0.007562s : 43.49%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000065s :  0.37%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.19%
    rewriter_before_opt_a                :     0.000075s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.34%
    jit_opt_a.loop_unroll                :     0.000035s :  0.20%
    jit_opt_a.a_1                        :     0.000710s :  4.08%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000019s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000583s :  3.35%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.10%
    jit_opt_a.cse                        :     0.000037s :  0.22%
    jit_opt_a.replace_applicator         :     0.000016s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.05%
    rewriter_after_opt_a                 :     0.000033s :  0.19%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000551s :  3.17%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000025s :  0.15%
    loop_unroll                          :     0.000410s :  2.36%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000047s :  0.27%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000456s :  2.62%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000033s :  0.19%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005908s : 33.98%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000204    23
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.87% :     0.000006s :      3: substitution.graph_param_transform
80.37% :     0.000164s :      4: substitution.inline
 1.58% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.24% :     0.000005s :      4: substitution.remove_not_recompute_node
 4.27% :     0.000009s :      2: substitution.replace_old_param
 7.14% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007498     2
88.69% :     0.006650s :      1: type_inference.infer
11.31% :     0.000848s :      1: type_inference.specialize
------[replace.]   0.000071     6
62.95% :     0.000044s :      4: replace.inline
37.05% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000174     6
92.55% :     0.000161s :      4: match.inline
 7.45% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.12% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.22% :     0.000002s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 1.97% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.17% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.55% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.35% :     0.000002s :     10: predicate.environ_get_eliminate
 1.23% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.29% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.73% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.28% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.23% :     0.000002s :     10: predicate.minmaximum_grad
 1.66% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.24% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.33% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.52% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.79% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.28% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.88% :     0.000010s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.08% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.11% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000541    12
46.65% :     0.000252s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.64% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.70% :     0.000253s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024690    76
 0.20% :     0.000049s :      1: add_recomputation
 0.28% :     0.000069s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.06% :     0.000262s :      1: bootstrap
 0.11% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000028s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
28.54% :     0.007045s :      1: jit_opt_a
 0.60% :     0.000149s :      1: jit_opt_after_cconv
 0.21% :     0.000052s :      1: jit_opt_b
 1.69% :     0.000418s :      1: loop_unroll
 2.26% :     0.000559s :      1: mutable_eliminate
 3.77% :     0.000932s :     26: opt.transform.jit_opt_a
 0.19% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.88% :     0.000464s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000035s :      1: py_interpret_to_execute
 0.04% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000015s :      1: remove_dup_value
 1.23% :     0.000305s :      1: renormalize.infer
 1.10% :     0.000271s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000036s :      1: rewriter_after_opt_a
 0.32% :     0.000078s :      1: rewriter_before_opt_a
 0.30% :     0.000074s :      1: symbol_engine_optimizer
23.97% :     0.005918s :      1: task_emit
30.68% :     0.007576s :      1: type_inference
 0.21% :     0.000051s :      1: validate


TotalTime = 0.0139714, [30]
[bootstrap]: 0.00025934
[type_inference]: 0.00612609
[event_method]: 2.105e-05
[auto_monad]: 8.142e-05
[graph_reusing]: 6.00002e-06
[pre_auto_parallel]: 2.74999e-06
[py_interpret_to_execute]: 4.687e-05
[rewriter_before_opt_a]: 8.211e-05
[expand_dump_flag]: 2.85998e-06
[jit_opt_a]: 0.00525201, [2]
    [Cycle 1]: 0.00170999, [27]
        [switch_simplify]: 0.00010908
        [loop_unroll]: 3.49e-05
        [a_1]: 0.00065419
        [with_stream_mark]: 1.691e-05
        [recompute_prepare]: 7.10998e-06
        [updatestate_depend_eliminate]: 3.48e-06
        [updatestate_assign_eliminate]: 2.86e-06
        [updatestate_loads_eliminate]: 2.76999e-06
        [parameter_eliminate]: 1.94999e-06
        [specialize_transform]: 5.74e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.29998e-06
        [meta_shard_fg_expand]: 2.11998e-06
        [get_grad_eliminate_]: 5.03002e-06
        [merge_forward]: 3.38999e-06
        [cell_reuse_recompute_pass]: 1.11002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.424e-05
        [j_node_and_user_rematch]: 9.51998e-06
        [meta_fg_expand]: 2.20002e-06
        [replace_old_param]: 8.12e-06
        [inline_without_move]: 5.54998e-06
        [renormalize]: 0.0005905
        [add_forward_monad_depend]: 5.52001e-06
        [auto_monad_grad]: 2.26998e-06
        [auto_monad_eliminator]: 1.357e-05
        [cse]: 2.052e-05
        [replace_applicator]: 1.1e-05
    [Cycle 2]: 0.000307, [27]
        [switch_simplify]: 5.20001e-06
        [loop_unroll]: 4.89998e-06
        [a_1]: 8.852e-05
        [with_stream_mark]: 8e-06
        [recompute_prepare]: 5.29998e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.76e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 4.69998e-06
        [updatestate_useless_node_eliminater]: 4.74998e-06
        [accelerated_algorithm]: 4.64998e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 2.27001e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.23e-05
        [j_node_and_user_rematch]: 8.02e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 6.87002e-06
        [inline_without_move]: 4.66002e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 4.52e-06
        [cse]: 9.49999e-06
        [replace_applicator]: 4.98001e-06
[py_interpret_to_execute_after_opt_a]: 9.91e-06
[rewriter_after_opt_a]: 3.066e-05
[convert_after_rewriter]: 6.74999e-06
[order_py_execute_after_rewriter]: 5.02999e-06
[mutable_eliminate]: 0.00051718
[jit_opt_b]: 4.772e-05, [1]
    [Cycle 1]: 4.124e-05, [2]
        [frontend_op_eliminate]: 1.646e-05
        [inline_after_opt_a]: 1.374e-05
[cconv]: 2.319e-05
[loop_unroll]: 0.00049064
[jit_opt_after_cconv]: 0.00013441, [1]
    [Cycle 1]: 0.00012804, [11]
        [c_1]: 2.094e-05
        [parameter_eliminate]: 2.71999e-06
        [updatestate_depend_eliminate]: 5.12e-06
        [updatestate_assign_eliminate]: 2.33998e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [cse]: 1.668e-05
        [call_graph_tuple_transform]: 1.788e-05
        [tuple_list_get_item_eliminator]: 5.00999e-06
        [none_parameter_eliminate]: 1.35001e-06
        [renormalize]: 4.49974e-07
        [switch_simplify]: 4.97e-06
[remove_dup_value]: 1.369e-05
[partial_unused_args_eliminate]: 2.48e-06
[environ_conv]: 5.49e-06
[add_recomputation]: 4.539e-05
[cse_after_recomputation]: 2.045e-05, [1]
    [Cycle 1]: 1.476e-05, [1]
        [cse]: 9.10001e-06
[auto_monad_reorder]: 1.466e-05
[get_jit_bprop_graph]: 2.06e-06
[rewriter_after_jit_bprop_graph]: 5.05999e-06
[opt_after_jit_grad]: 0.00045739
[symbol_engine_optimizer]: 7.048e-05, [1]
    [Cycle 1]: 6.45e-05, [6]
        [build]: 2.76999e-06
        [elim_shapecalc]: 7.5e-06
        [elim_not_effective]: 1.256e-05
        [opt_reshape]: 5.40001e-06
        [fold_const_symbol]: 8.39998e-06
        [renormalize]: 4.19997e-07
[validate]: 3.13e-05
Sums
    bootstrap                            :     0.000259s :  2.55%
    type_inference                       :     0.006126s : 60.14%
    event_method                         :     0.000021s :  0.21%
    auto_monad                           :     0.000081s :  0.80%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000047s :  0.46%
    rewriter_before_opt_a                :     0.000082s :  0.81%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000114s :  1.12%
    jit_opt_a.loop_unroll                :     0.000040s :  0.39%
    jit_opt_a.a_1                        :     0.000743s :  7.29%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000010s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000591s :  5.80%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.18%
    jit_opt_a.cse                        :     0.000030s :  0.29%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.10%
    rewriter_after_opt_a                 :     0.000031s :  0.30%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000517s :  5.08%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.13%
    cconv                                :     0.000023s :  0.23%
    loop_unroll                          :     0.000491s :  4.82%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000045s :  0.45%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000457s :  4.49%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.31%


Time group info:
------[substitution.]   0.000251    26
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.21% :     0.000006s :      3: substitution.graph_param_transform
83.79% :     0.000210s :      6: substitution.inline
 1.33% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.67% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.33% :     0.000003s :      2: substitution.replace_old_param
 3.42% :     0.000009s :      1: substitution.switch_simplify
 4.98% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006068     2
83.59% :     0.005072s :      1: type_inference.infer
16.41% :     0.000996s :      1: type_inference.specialize
------[replace.]   0.000122     9
50.61% :     0.000062s :      6: replace.inline
24.99% :     0.000030s :      1: replace.switch_simplify
24.40% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000225     9
91.53% :     0.000206s :      6: match.inline
 3.46% :     0.000008s :      1: match.switch_simplify
 5.01% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   801
 1.43% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.15% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.22% :     0.000002s :     12: predicate.addn_check_dump
 1.39% :     0.000002s :     12: predicate.addn_zero_filter
 2.01% :     0.000003s :     12: predicate.arithmetic_simplify
 1.29% :     0.000002s :     12: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.20% :     0.000002s :     12: predicate.compare_switch_simplify
 1.59% :     0.000002s :     12: predicate.depend_value_elim
 1.19% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     12: predicate.environ_get_eliminate
 1.13% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.35% :     0.000007s :     26: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.94% :     0.000001s :      6: predicate.less_batch_normalization
 1.65% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     17: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.27% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.07% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.29% :     0.000002s :     12: predicate.merge_addn
 1.19% :     0.000002s :     12: predicate.minmaximum_grad
 1.39% :     0.000002s :      3: predicate.mutable_eliminate
 0.36% :     0.000000s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     17: predicate.partial_eliminate
 1.27% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     12: predicate.reduce_eliminate
 1.49% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     20: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     12: predicate.reshape_eliminate
 1.27% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     12: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.36% :     0.000002s :      6: predicate.special_op_eliminate
 0.86% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.60% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.60% :     0.000003s :     20: predicate.switch_defer_inline
 2.26% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.20% :     0.000011s :     65: predicate.switch_simplify
 1.42% :     0.000002s :     12: predicate.tile_eliminate
 1.22% :     0.000002s :     12: predicate.transpose_eliminate
 1.44% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.70% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000724    15
49.60% :     0.000359s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.78% :     0.000078s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.62% :     0.000287s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015630    72
 0.31% :     0.000048s :      1: add_recomputation
 0.54% :     0.000085s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.75% :     0.000274s :      1: bootstrap
 0.16% :     0.000026s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
33.62% :     0.005255s :      1: jit_opt_a
 0.88% :     0.000137s :      1: jit_opt_after_cconv
 0.32% :     0.000050s :      1: jit_opt_b
 3.19% :     0.000499s :      1: loop_unroll
 3.36% :     0.000525s :      1: mutable_eliminate
 6.44% :     0.001007s :     26: opt.transform.jit_opt_a
 0.29% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.98% :     0.000465s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.32% :     0.000049s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 1.99% :     0.000311s :      1: renormalize.infer
 1.74% :     0.000272s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000034s :      1: rewriter_after_opt_a
 0.54% :     0.000085s :      1: rewriter_before_opt_a
 0.47% :     0.000073s :      1: symbol_engine_optimizer
39.27% :     0.006138s :      1: type_inference


TotalTime = 0.0162627, [33]
[bootstrap]: 0.00025784
[type_inference]: 0.00405606
[event_method]: 1.507e-05
[auto_monad]: 5.502e-05
[graph_reusing]: 4.60999e-06
[pre_auto_parallel]: 2.12999e-06
[py_interpret_to_execute]: 2.629e-05
[rewriter_before_opt_a]: 6.521e-05
[expand_dump_flag]: 3.12002e-06
[jit_opt_a]: 0.0046838, [2]
    [Cycle 1]: 0.0013921, [27]
        [switch_simplify]: 4.382e-05
        [loop_unroll]: 2.837e-05
        [a_1]: 0.00048142
        [with_stream_mark]: 1.34e-05
        [recompute_prepare]: 6.89999e-06
        [updatestate_depend_eliminate]: 3.14001e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.13002e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 5.90002e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 6.02001e-06
        [meta_shard_fg_expand]: 1.82999e-06
        [get_grad_eliminate_]: 5.50001e-06
        [merge_forward]: 3.01999e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.169e-05
        [j_node_and_user_rematch]: 8.24998e-06
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 8.40001e-06
        [inline_without_move]: 5.35001e-06
        [renormalize]: 0.00052909
        [add_forward_monad_depend]: 4.92e-06
        [auto_monad_grad]: 2.54001e-06
        [auto_monad_eliminator]: 1.223e-05
        [cse]: 2.058e-05
        [replace_applicator]: 1.195e-05
    [Cycle 2]: 0.00031129, [27]
        [switch_simplify]: 5.20999e-06
        [loop_unroll]: 4.46002e-06
        [a_1]: 8.899e-05
        [with_stream_mark]: 8.37e-06
        [recompute_prepare]: 4.77e-06
        [updatestate_depend_eliminate]: 2.58e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 4.82998e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.14e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.43001e-06
        [merge_forward]: 2.34999e-06
        [cell_reuse_recompute_pass]: 2.06998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.196e-05
        [j_node_and_user_rematch]: 7.63001e-06
        [meta_fg_expand]: 1.39e-06
        [replace_old_param]: 7.53e-06
        [inline_without_move]: 4.47e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 8.80013e-07
        [auto_monad_eliminator]: 4.86002e-06
        [cse]: 9.10001e-06
        [replace_applicator]: 4.86002e-06
[py_interpret_to_execute_after_opt_a]: 8.50001e-06
[rewriter_after_opt_a]: 2.981e-05
[convert_after_rewriter]: 6.21e-06
[order_py_execute_after_rewriter]: 5.05001e-06
[mutable_eliminate]: 0.00049326
[jit_opt_b]: 4.623e-05, [1]
    [Cycle 1]: 3.977e-05, [2]
        [frontend_op_eliminate]: 1.521e-05
        [inline_after_opt_a]: 1.387e-05
[cconv]: 2.303e-05
[loop_unroll]: 0.00040371
[jit_opt_after_cconv]: 0.00013444, [1]
    [Cycle 1]: 0.00012793, [11]
        [c_1]: 2.027e-05
        [parameter_eliminate]: 2.49999e-06
        [updatestate_depend_eliminate]: 5.20999e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.637e-05
        [call_graph_tuple_transform]: 1.817e-05
        [tuple_list_get_item_eliminator]: 5.10999e-06
        [none_parameter_eliminate]: 1.45001e-06
        [renormalize]: 4.90021e-07
        [switch_simplify]: 5.35001e-06
[remove_dup_value]: 1.171e-05
[partial_unused_args_eliminate]: 1.99e-06
[environ_conv]: 4.81002e-06
[add_recomputation]: 3.781e-05
[cse_after_recomputation]: 1.915e-05, [1]
    [Cycle 1]: 1.403e-05, [1]
        [cse]: 8.47e-06
[auto_monad_reorder]: 1.163e-05
[get_jit_bprop_graph]: 1.74998e-06
[rewriter_after_jit_bprop_graph]: 5.28002e-06
[opt_after_jit_grad]: 0.00043896
[symbol_engine_optimizer]: 7.009e-05, [1]
    [Cycle 1]: 6.359e-05, [6]
        [build]: 2.81e-06
        [elim_shapecalc]: 7.28e-06
        [elim_not_effective]: 1.214e-05
        [opt_reshape]: 5.15001e-06
        [fold_const_symbol]: 8.77e-06
        [renormalize]: 6.69999e-07
[validate]: 2.945e-05
[backend_pass]: 8.2e-07
[task_emit]: 0.00513583
[execute]: 5.58997e-06
Sums
    bootstrap                            :     0.000258s :  2.03%
    type_inference                       :     0.004056s : 31.88%
    event_method                         :     0.000015s :  0.12%
    auto_monad                           :     0.000055s :  0.43%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.21%
    rewriter_before_opt_a                :     0.000065s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000049s :  0.39%
    jit_opt_a.loop_unroll                :     0.000033s :  0.26%
    jit_opt_a.a_1                        :     0.000570s :  4.48%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.13%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000529s :  4.16%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000030s :  0.23%
    jit_opt_a.replace_applicator         :     0.000017s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000030s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000493s :  3.88%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000023s :  0.18%
    loop_unroll                          :     0.000404s :  3.17%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000038s :  0.30%
    cse_after_recomputation.cse          :     0.000008s :  0.07%
    auto_monad_reorder                   :     0.000012s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000439s :  3.45%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000029s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005136s : 40.37%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000173    23
 1.06% :     0.000002s :      2: substitution.elim_not_effective
 0.79% :     0.000001s :      2: substitution.fold_const_symbol
 3.11% :     0.000005s :      3: substitution.graph_param_transform
82.53% :     0.000143s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.22% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.25% :     0.000004s :      2: substitution.replace_old_param
 6.47% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004003     2
82.42% :     0.003300s :      1: type_inference.infer
17.58% :     0.000704s :      1: type_inference.specialize
------[replace.]   0.000060     6
65.59% :     0.000039s :      4: replace.inline
34.41% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000151     6
93.25% :     0.000141s :      4: match.inline
 6.75% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.34% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.38% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.77% :     0.000002s :     10: predicate.addn_zero_filter
 1.95% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000001s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.50% :     0.000002s :     10: predicate.depend_value_elim
 1.25% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.57% :     0.000001s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.01% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.51% :     0.000002s :      6: predicate.less_batch_normalization
 1.96% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.72% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.02% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.70% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000000s :      3: predicate.opt_reshape
 2.15% :     0.000002s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.83% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.49% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.59% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.61% :     0.000008s :     53: predicate.switch_simplify
 1.34% :     0.000001s :     10: predicate.tile_eliminate
 1.32% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.22% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000486    12
47.97% :     0.000233s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.27% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.76% :     0.000222s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017662    76
 0.23% :     0.000041s :      1: add_recomputation
 0.33% :     0.000058s :      1: auto_monad
 0.08% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.53% :     0.000270s :      1: bootstrap
 0.15% :     0.000026s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000018s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.53% :     0.004686s :      1: jit_opt_a
 0.78% :     0.000137s :      1: jit_opt_after_cconv
 0.28% :     0.000049s :      1: jit_opt_b
 2.33% :     0.000411s :      1: loop_unroll
 2.84% :     0.000501s :      1: mutable_eliminate
 4.32% :     0.000764s :     26: opt.transform.jit_opt_a
 0.26% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.53% :     0.000446s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000029s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 1.57% :     0.000276s :      1: renormalize.infer
 1.39% :     0.000246s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000032s :      1: rewriter_after_opt_a
 0.38% :     0.000068s :      1: rewriter_before_opt_a
 0.41% :     0.000073s :      1: symbol_engine_optimizer
29.14% :     0.005147s :      1: task_emit
23.02% :     0.004066s :      1: type_inference
 0.27% :     0.000047s :      1: validate


TotalTime = 0.0130964, [30]
[bootstrap]: 0.00024489
[type_inference]: 0.00576809
[event_method]: 1.887e-05
[auto_monad]: 6.755e-05
[graph_reusing]: 5.30999e-06
[pre_auto_parallel]: 2.31998e-06
[py_interpret_to_execute]: 3.926e-05
[rewriter_before_opt_a]: 7.666e-05
[expand_dump_flag]: 2.64001e-06
[jit_opt_a]: 0.00485555, [2]
    [Cycle 1]: 0.001627, [27]
        [switch_simplify]: 9.374e-05
        [loop_unroll]: 3.368e-05
        [a_1]: 0.00066506
        [with_stream_mark]: 1.443e-05
        [recompute_prepare]: 7.87e-06
        [updatestate_depend_eliminate]: 3.07002e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.44998e-06
        [specialize_transform]: 6.24001e-06
        [updatestate_useless_node_eliminater]: 5.15999e-06
        [accelerated_algorithm]: 5.47999e-06
        [meta_shard_fg_expand]: 1.70001e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 2.86999e-06
        [cell_reuse_recompute_pass]: 1.07998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.149e-05
        [j_node_and_user_rematch]: 9.28002e-06
        [meta_fg_expand]: 2.08002e-06
        [replace_old_param]: 8e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 0.00051959
        [add_forward_monad_depend]: 4.84e-06
        [auto_monad_grad]: 1.57999e-06
        [auto_monad_eliminator]: 1.274e-05
        [cse]: 2.131e-05
        [replace_applicator]: 1.275e-05
    [Cycle 2]: 0.00032436, [27]
        [switch_simplify]: 5.88002e-06
        [loop_unroll]: 4.82e-06
        [a_1]: 9.106e-05
        [with_stream_mark]: 9.89999e-06
        [recompute_prepare]: 5.27001e-06
        [updatestate_depend_eliminate]: 2.66e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 4.85999e-06
        [updatestate_useless_node_eliminater]: 5.54998e-06
        [accelerated_algorithm]: 4.70001e-06
        [meta_shard_fg_expand]: 1.19e-06
        [get_grad_eliminate_]: 4.80001e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 1.33002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.253e-05
        [j_node_and_user_rematch]: 8.42e-06
        [meta_fg_expand]: 1.55001e-06
        [replace_old_param]: 6.89999e-06
        [inline_without_move]: 4.82e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.23002e-06
        [auto_monad_grad]: 7.40023e-07
        [auto_monad_eliminator]: 5.02e-06
        [cse]: 1.052e-05
        [replace_applicator]: 5.19998e-06
[py_interpret_to_execute_after_opt_a]: 9.20999e-06
[rewriter_after_opt_a]: 3.15e-05
[convert_after_rewriter]: 6.94001e-06
[order_py_execute_after_rewriter]: 4.75001e-06
[mutable_eliminate]: 0.00052761
[jit_opt_b]: 4.743e-05, [1]
    [Cycle 1]: 4.059e-05, [2]
        [frontend_op_eliminate]: 1.508e-05
        [inline_after_opt_a]: 1.399e-05
[cconv]: 2.124e-05
[loop_unroll]: 0.00043375
[jit_opt_after_cconv]: 0.00013496, [1]
    [Cycle 1]: 0.00012874, [11]
        [c_1]: 2.033e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 5.96998e-06
        [updatestate_assign_eliminate]: 2.49999e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [cse]: 1.733e-05
        [call_graph_tuple_transform]: 1.705e-05
        [tuple_list_get_item_eliminator]: 4.97e-06
        [none_parameter_eliminate]: 1.62001e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.29e-06
[remove_dup_value]: 1.324e-05
[partial_unused_args_eliminate]: 2.13998e-06
[environ_conv]: 5.60001e-06
[add_recomputation]: 4.13e-05
[cse_after_recomputation]: 2.059e-05, [1]
    [Cycle 1]: 1.497e-05, [1]
        [cse]: 9.34e-06
[auto_monad_reorder]: 1.4e-05
[get_jit_bprop_graph]: 1.55999e-06
[rewriter_after_jit_bprop_graph]: 5.35999e-06
[opt_after_jit_grad]: 0.00044082
[symbol_engine_optimizer]: 7.17e-05, [1]
    [Cycle 1]: 6.547e-05, [6]
        [build]: 3.47997e-06
        [elim_shapecalc]: 7.33999e-06
        [elim_not_effective]: 1.251e-05
        [opt_reshape]: 5.44e-06
        [fold_const_symbol]: 8.42e-06
        [renormalize]: 9.20001e-07
[validate]: 3.043e-05
Sums
    bootstrap                            :     0.000245s :  2.54%
    type_inference                       :     0.005768s : 59.84%
    event_method                         :     0.000019s :  0.20%
    auto_monad                           :     0.000068s :  0.70%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.41%
    rewriter_before_opt_a                :     0.000077s :  0.80%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000100s :  1.03%
    jit_opt_a.loop_unroll                :     0.000039s :  0.40%
    jit_opt_a.a_1                        :     0.000756s :  7.84%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000520s :  5.39%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.18%
    jit_opt_a.cse                        :     0.000032s :  0.33%
    jit_opt_a.replace_applicator         :     0.000018s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.10%
    rewriter_after_opt_a                 :     0.000032s :  0.33%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000528s :  5.47%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.15%
    cconv                                :     0.000021s :  0.22%
    loop_unroll                          :     0.000434s :  4.50%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000041s :  0.43%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000014s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000441s :  4.57%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000030s :  0.32%


Time group info:
------[substitution.]   0.000231    26
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.12% :     0.000005s :      3: substitution.graph_param_transform
83.93% :     0.000194s :      6: substitution.inline
 1.44% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.84% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.28% :     0.000003s :      2: substitution.replace_old_param
 3.12% :     0.000007s :      1: substitution.switch_simplify
 4.98% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005717     2
82.95% :     0.004742s :      1: type_inference.infer
17.05% :     0.000975s :      1: type_inference.specialize
------[replace.]   0.000142     9
62.22% :     0.000088s :      6: replace.inline
17.86% :     0.000025s :      1: replace.switch_simplify
19.92% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000207     9
91.92% :     0.000190s :      6: match.inline
 3.13% :     0.000006s :      1: match.switch_simplify
 4.95% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   801
 1.37% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.10% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.23% :     0.000002s :     12: predicate.addn_check_dump
 1.65% :     0.000002s :     12: predicate.addn_zero_filter
 1.94% :     0.000002s :     12: predicate.arithmetic_simplify
 1.31% :     0.000002s :     12: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.30% :     0.000002s :     12: predicate.compare_switch_simplify
 1.43% :     0.000002s :     12: predicate.depend_value_elim
 1.24% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.74% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.27% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     12: predicate.environ_get_eliminate
 1.27% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.56% :     0.000007s :     26: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.92% :     0.000002s :     17: predicate.load_eliminater
 1.20% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.19% :     0.000005s :     40: predicate.loop_unroll_before_grad
 2.02% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.24% :     0.000002s :     12: predicate.merge_addn
 1.23% :     0.000002s :     12: predicate.minmaximum_grad
 1.42% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     17: predicate.partial_eliminate
 1.30% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     12: predicate.reduce_eliminate
 1.60% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     20: predicate.replace_applicator
 0.42% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     12: predicate.reshape_eliminate
 1.33% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     12: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.73% :     0.000003s :     20: predicate.switch_defer_inline
 2.43% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.10% :     0.000010s :     65: predicate.switch_simplify
 1.24% :     0.000002s :     12: predicate.tile_eliminate
 1.27% :     0.000002s :     12: predicate.transpose_eliminate
 1.37% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.41% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.78% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     12: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000673    15
48.55% :     0.000327s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.15% :     0.000075s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.31% :     0.000271s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014684    72
 0.30% :     0.000044s :      1: add_recomputation
 0.49% :     0.000071s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.75% :     0.000257s :      1: bootstrap
 0.16% :     0.000024s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000022s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
33.09% :     0.004859s :      1: jit_opt_a
 0.94% :     0.000137s :      1: jit_opt_after_cconv
 0.34% :     0.000050s :      1: jit_opt_b
 3.01% :     0.000442s :      1: loop_unroll
 3.65% :     0.000536s :      1: mutable_eliminate
 6.87% :     0.001008s :     26: opt.transform.jit_opt_a
 0.30% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000023s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.05% :     0.000449s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.29% :     0.000042s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000016s :      1: remove_dup_value
 1.86% :     0.000274s :      1: renormalize.infer
 1.63% :     0.000240s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000035s :      1: rewriter_after_opt_a
 0.54% :     0.000080s :      1: rewriter_before_opt_a
 0.50% :     0.000074s :      1: symbol_engine_optimizer
39.35% :     0.005778s :      1: type_inference


TotalTime = 0.0226568, [33]
[bootstrap]: 0.00037208
[type_inference]: 0.00534478
[event_method]: 1.751e-05
[auto_monad]: 6.949e-05
[graph_reusing]: 6.17999e-06
[pre_auto_parallel]: 3.06001e-06
[py_interpret_to_execute]: 3.801e-05
[rewriter_before_opt_a]: 7.182e-05
[expand_dump_flag]: 3.46001e-06
[jit_opt_a]: 0.0064174, [2]
    [Cycle 1]: 0.00208399, [27]
        [switch_simplify]: 6.571e-05
        [loop_unroll]: 3.132e-05
        [a_1]: 0.00060811
        [with_stream_mark]: 2.004e-05
        [recompute_prepare]: 1.044e-05
        [updatestate_depend_eliminate]: 3.68e-06
        [updatestate_assign_eliminate]: 3.12002e-06
        [updatestate_loads_eliminate]: 2.89001e-06
        [parameter_eliminate]: 2.04999e-06
        [specialize_transform]: 5.79e-06
        [updatestate_useless_node_eliminater]: 5.35999e-06
        [accelerated_algorithm]: 6.02999e-06
        [meta_shard_fg_expand]: 2.96001e-06
        [get_grad_eliminate_]: 5.69e-06
        [merge_forward]: 3.63e-06
        [cell_reuse_recompute_pass]: 1.71e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.45e-05
        [j_node_and_user_rematch]: 9.32001e-06
        [meta_fg_expand]: 2.76e-06
        [replace_old_param]: 9.17001e-06
        [inline_without_move]: 5.62999e-06
        [renormalize]: 0.00101542
        [add_forward_monad_depend]: 6.51999e-06
        [auto_monad_grad]: 2.09999e-06
        [auto_monad_eliminator]: 1.498e-05
        [cse]: 2.685e-05
        [replace_applicator]: 1.404e-05
    [Cycle 2]: 0.00033156, [27]
        [switch_simplify]: 6.06e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 9.608e-05
        [with_stream_mark]: 1.286e-05
        [recompute_prepare]: 4.97e-06
        [updatestate_depend_eliminate]: 2.61999e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 1.23002e-06
        [specialize_transform]: 5.04998e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 5.17999e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 4.51002e-06
        [merge_forward]: 2.98998e-06
        [cell_reuse_recompute_pass]: 2.18002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.383e-05
        [j_node_and_user_rematch]: 8.18001e-06
        [meta_fg_expand]: 1.59998e-06
        [replace_old_param]: 8.03999e-06
        [inline_without_move]: 4.82e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.35999e-06
        [auto_monad_grad]: 1.00999e-06
        [auto_monad_eliminator]: 4.69998e-06
        [cse]: 1.047e-05
        [replace_applicator]: 4.79e-06
[py_interpret_to_execute_after_opt_a]: 1.055e-05
[rewriter_after_opt_a]: 3.551e-05
[convert_after_rewriter]: 7.36001e-06
[order_py_execute_after_rewriter]: 5.15999e-06
[mutable_eliminate]: 0.00073583
[jit_opt_b]: 5.011e-05, [1]
    [Cycle 1]: 4.279e-05, [2]
        [frontend_op_eliminate]: 1.612e-05
        [inline_after_opt_a]: 1.485e-05
[cconv]: 2.743e-05
[loop_unroll]: 0.0004632
[jit_opt_after_cconv]: 0.00014767, [1]
    [Cycle 1]: 0.00014056, [11]
        [c_1]: 2.142e-05
        [parameter_eliminate]: 3.56001e-06
        [updatestate_depend_eliminate]: 5.57999e-06
        [updatestate_assign_eliminate]: 2.49999e-06
        [updatestate_loads_eliminate]: 2.11998e-06
        [cse]: 2.083e-05
        [call_graph_tuple_transform]: 2.021e-05
        [tuple_list_get_item_eliminator]: 5.65001e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 6.80011e-07
        [switch_simplify]: 5.44e-06
[remove_dup_value]: 1.432e-05
[partial_unused_args_eliminate]: 2.73e-06
[environ_conv]: 5.44e-06
[add_recomputation]: 4.851e-05
[cse_after_recomputation]: 2.225e-05, [1]
    [Cycle 1]: 1.635e-05, [1]
        [cse]: 9.95002e-06
[auto_monad_reorder]: 1.591e-05
[get_jit_bprop_graph]: 2.36e-06
[rewriter_after_jit_bprop_graph]: 6.71999e-06
[opt_after_jit_grad]: 0.00052978
[symbol_engine_optimizer]: 8.623e-05, [1]
    [Cycle 1]: 7.856e-05, [6]
        [build]: 5.74999e-06
        [elim_shapecalc]: 9.59e-06
        [elim_not_effective]: 1.688e-05
        [opt_reshape]: 6.07001e-06
        [fold_const_symbol]: 1.019e-05
        [renormalize]: 1.03001e-06
[validate]: 4.466e-05
[backend_pass]: 1.04e-06
[task_emit]: 0.00782667
[execute]: 9.49999e-06
Sums
    bootstrap                            :     0.000372s :  2.07%
    type_inference                       :     0.005345s : 29.67%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000069s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.21%
    rewriter_before_opt_a                :     0.000072s :  0.40%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000072s :  0.40%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000704s :  3.91%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001015s :  5.64%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.11%
    jit_opt_a.cse                        :     0.000037s :  0.21%
    jit_opt_a.replace_applicator         :     0.000019s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000036s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000736s :  4.08%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000027s :  0.15%
    loop_unroll                          :     0.000463s :  2.57%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.27%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000530s :  2.94%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000045s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007827s : 43.44%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000251    23
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.95% :     0.000002s :      2: substitution.fold_const_symbol
 2.37% :     0.000006s :      3: substitution.graph_param_transform
84.88% :     0.000213s :      4: substitution.inline
 1.51% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.01% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000005s :      2: substitution.replace_old_param
 5.34% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005273     2
82.15% :     0.004332s :      1: type_inference.infer
17.85% :     0.000941s :      1: type_inference.specialize
------[replace.]   0.000084     6
65.84% :     0.000055s :      4: replace.inline
34.16% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000222     6
94.48% :     0.000210s :      4: match.inline
 5.52% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 2.28% :     0.000003s :     10: predicate.accumulaten_eliminater
 1.88% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.12% :     0.000001s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000002s :     10: predicate.depend_value_elim
 1.69% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.51% :     0.000001s :      3: predicate.elim_not_effective
 0.77% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     10: predicate.environ_get_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.57% :     0.000007s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      6: predicate.less_batch_normalization
 1.35% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.39% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.18% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.74% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.01% :     0.000002s :     15: predicate.partial_eliminate
 1.01% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     10: predicate.reduce_eliminate
 1.33% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.69% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.09% :     0.000001s :     10: predicate.same_eliminate
 0.60% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.57% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.63% :     0.000010s :     53: predicate.switch_simplify
 1.32% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.51% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.73% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000716    12
47.54% :     0.000340s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.18% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.27% :     0.000331s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024717    76
 0.21% :     0.000052s :      1: add_recomputation
 0.30% :     0.000073s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.57% :     0.000389s :      1: bootstrap
 0.12% :     0.000030s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
25.98% :     0.006422s :      1: jit_opt_a
 0.61% :     0.000150s :      1: jit_opt_after_cconv
 0.21% :     0.000053s :      1: jit_opt_b
 1.91% :     0.000472s :      1: loop_unroll
 3.01% :     0.000745s :      1: mutable_eliminate
 3.78% :     0.000934s :     26: opt.transform.jit_opt_a
 0.20% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000038s :      4: opt.transform.symbol_engine_opt
 2.18% :     0.000540s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000041s :      1: py_interpret_to_execute
 0.05% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.39% :     0.000590s :      1: renormalize.infer
 1.68% :     0.000416s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000038s :      1: rewriter_after_opt_a
 0.30% :     0.000075s :      1: rewriter_before_opt_a
 0.36% :     0.000089s :      1: symbol_engine_optimizer
31.73% :     0.007842s :      1: task_emit
21.67% :     0.005357s :      1: type_inference
 0.27% :     0.000067s :      1: validate


TotalTime = 0.0158454, [33]
[bootstrap]: 0.00023711
[type_inference]: 0.00384474
[event_method]: 1.479e-05
[auto_monad]: 4.222e-05
[graph_reusing]: 4.07e-06
[pre_auto_parallel]: 1.89e-06
[py_interpret_to_execute]: 2.568e-05
[rewriter_before_opt_a]: 6.11e-05
[expand_dump_flag]: 2.03997e-06
[jit_opt_a]: 0.00482237, [2]
    [Cycle 1]: 0.00145201, [27]
        [switch_simplify]: 4.414e-05
        [loop_unroll]: 2.805e-05
        [a_1]: 0.00047073
        [with_stream_mark]: 1.416e-05
        [recompute_prepare]: 6.76e-06
        [updatestate_depend_eliminate]: 2.84001e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 1.42e-06
        [specialize_transform]: 5.74999e-06
        [updatestate_useless_node_eliminater]: 4.80999e-06
        [accelerated_algorithm]: 5.20001e-06
        [meta_shard_fg_expand]: 1.40999e-06
        [get_grad_eliminate_]: 4.99e-06
        [merge_forward]: 2.48998e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.1e-05
        [j_node_and_user_rematch]: 8.38999e-06
        [meta_fg_expand]: 1.71002e-06
        [replace_old_param]: 7.6e-06
        [inline_without_move]: 5.07e-06
        [renormalize]: 0.00062049
        [add_forward_monad_depend]: 4.62e-06
        [auto_monad_grad]: 1.69998e-06
        [auto_monad_eliminator]: 1.136e-05
        [cse]: 1.542e-05
        [replace_applicator]: 1.152e-05
    [Cycle 2]: 0.00031324, [27]
        [switch_simplify]: 5.59e-06
        [loop_unroll]: 4.95001e-06
        [a_1]: 8.764e-05
        [with_stream_mark]: 8.89998e-06
        [recompute_prepare]: 4.65001e-06
        [updatestate_depend_eliminate]: 2.63003e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.92999e-06
        [updatestate_useless_node_eliminater]: 4.78001e-06
        [accelerated_algorithm]: 5.35001e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.43001e-06
        [merge_forward]: 2.21998e-06
        [cell_reuse_recompute_pass]: 2.09999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.193e-05
        [j_node_and_user_rematch]: 7.41001e-06
        [meta_fg_expand]: 1.45999e-06
        [replace_old_param]: 6.86999e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.08001e-06
        [auto_monad_grad]: 6.29982e-07
        [auto_monad_eliminator]: 4.96002e-06
        [cse]: 9.96e-06
        [replace_applicator]: 5.04e-06
[py_interpret_to_execute_after_opt_a]: 9.24e-06
[rewriter_after_opt_a]: 2.76e-05
[convert_after_rewriter]: 5.63002e-06
[order_py_execute_after_rewriter]: 4.58001e-06
[mutable_eliminate]: 0.00050307
[jit_opt_b]: 4.709e-05, [1]
    [Cycle 1]: 4.065e-05, [2]
        [frontend_op_eliminate]: 1.515e-05
        [inline_after_opt_a]: 1.401e-05
[cconv]: 1.757e-05
[loop_unroll]: 0.00041618
[jit_opt_after_cconv]: 0.00013236, [1]
    [Cycle 1]: 0.00012625, [11]
        [c_1]: 2.004e-05
        [parameter_eliminate]: 2.34001e-06
        [updatestate_depend_eliminate]: 5.22e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.65e-05
        [call_graph_tuple_transform]: 1.682e-05
        [tuple_list_get_item_eliminator]: 4.90999e-06
        [none_parameter_eliminate]: 1.72999e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 5.16998e-06
[remove_dup_value]: 9.64999e-06
[partial_unused_args_eliminate]: 1.92999e-06
[environ_conv]: 4.17e-06
[add_recomputation]: 3.587e-05
[cse_after_recomputation]: 1.973e-05, [1]
    [Cycle 1]: 1.461e-05, [1]
        [cse]: 8.71002e-06
[auto_monad_reorder]: 1.11e-05
[get_jit_bprop_graph]: 1.42e-06
[rewriter_after_jit_bprop_graph]: 4.09002e-06
[opt_after_jit_grad]: 0.00044399
[symbol_engine_optimizer]: 7.234e-05, [1]
    [Cycle 1]: 6.596e-05, [6]
        [build]: 2.93e-06
        [elim_shapecalc]: 7.5e-06
        [elim_not_effective]: 1.265e-05
        [opt_reshape]: 5.74999e-06
        [fold_const_symbol]: 8.62998e-06
        [renormalize]: 5.39992e-07
[validate]: 2.942e-05
[backend_pass]: 7.99977e-07
[task_emit]: 0.00481993
[execute]: 4.80999e-06
Sums
    bootstrap                            :     0.000237s :  1.94%
    type_inference                       :     0.003845s : 31.43%
    event_method                         :     0.000015s :  0.12%
    auto_monad                           :     0.000042s :  0.35%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.21%
    rewriter_before_opt_a                :     0.000061s :  0.50%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000050s :  0.41%
    jit_opt_a.loop_unroll                :     0.000033s :  0.27%
    jit_opt_a.a_1                        :     0.000558s :  4.56%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000621s :  5.07%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.13%
    jit_opt_a.cse                        :     0.000025s :  0.21%
    jit_opt_a.replace_applicator         :     0.000017s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.08%
    rewriter_after_opt_a                 :     0.000028s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000503s :  4.11%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000018s :  0.14%
    loop_unroll                          :     0.000416s :  3.40%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000010s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000036s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000011s :  0.09%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000444s :  3.63%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000029s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.004820s : 39.40%
    execute                              :     0.000005s :  0.04%


Time group info:
------[substitution.]   0.000161    23
 1.20% :     0.000002s :      2: substitution.elim_not_effective
 0.80% :     0.000001s :      2: substitution.fold_const_symbol
 2.89% :     0.000005s :      3: substitution.graph_param_transform
83.37% :     0.000134s :      4: substitution.inline
 1.51% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.47% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.86% :     0.000003s :      2: substitution.replace_old_param
 5.91% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003800     2
82.07% :     0.003119s :      1: type_inference.infer
17.93% :     0.000681s :      1: type_inference.specialize
------[replace.]   0.000059     6
68.06% :     0.000040s :      4: replace.inline
31.94% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000140     6
94.03% :     0.000132s :      4: match.inline
 5.97% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.48% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.07% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000001s :     10: predicate.addn_zero_filter
 1.87% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.86% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.17% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.25% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.20% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.65% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.17% :     0.000001s :     10: predicate.minmaximum_grad
 1.79% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000002s :     15: predicate.partial_eliminate
 1.26% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.96% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.66% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.60% :     0.000008s :     53: predicate.switch_simplify
 1.28% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.33% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.99% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000464    12
45.99% :     0.000213s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.86% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.16% :     0.000223s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017321    76
 0.22% :     0.000039s :      1: add_recomputation
 0.26% :     0.000045s :      1: auto_monad
 0.08% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.42% :     0.000245s :      1: bootstrap
 0.12% :     0.000020s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.13% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000018s :      1: event_method
 0.05% :     0.000008s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
27.86% :     0.004825s :      1: jit_opt_a
 0.78% :     0.000135s :      1: jit_opt_after_cconv
 0.29% :     0.000049s :      1: jit_opt_b
 2.45% :     0.000424s :      1: loop_unroll
 2.95% :     0.000512s :      1: mutable_eliminate
 4.32% :     0.000749s :     26: opt.transform.jit_opt_a
 0.25% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000023s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.61% :     0.000452s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000028s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000012s :      1: remove_dup_value
 1.65% :     0.000286s :      1: renormalize.infer
 1.90% :     0.000329s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000030s :      1: rewriter_after_opt_a
 0.37% :     0.000064s :      1: rewriter_before_opt_a
 0.43% :     0.000075s :      1: symbol_engine_optimizer
27.89% :     0.004830s :      1: task_emit
22.25% :     0.003854s :      1: type_inference
 0.28% :     0.000048s :      1: validate


TotalTime = 0.0182179, [33]
[bootstrap]: 0.00058333
[type_inference]: 0.00429184
[event_method]: 1.705e-05
[auto_monad]: 6.21e-05
[graph_reusing]: 6.16e-06
[pre_auto_parallel]: 2.59999e-06
[py_interpret_to_execute]: 3.125e-05
[rewriter_before_opt_a]: 8.053e-05
[expand_dump_flag]: 3.7e-06
[jit_opt_a]: 0.00503803, [2]
    [Cycle 1]: 0.00143448, [27]
        [switch_simplify]: 5.43e-05
        [loop_unroll]: 3.091e-05
        [a_1]: 0.00051317
        [with_stream_mark]: 1.697e-05
        [recompute_prepare]: 7.38e-06
        [updatestate_depend_eliminate]: 3.68e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.75001e-06
        [specialize_transform]: 6.43998e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 5.63002e-06
        [meta_shard_fg_expand]: 1.92999e-06
        [get_grad_eliminate_]: 5.13002e-06
        [merge_forward]: 3.6e-06
        [cell_reuse_recompute_pass]: 1.31002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.389e-05
        [j_node_and_user_rematch]: 8.97999e-06
        [meta_fg_expand]: 2.44999e-06
        [replace_old_param]: 8.79e-06
        [inline_without_move]: 5.35001e-06
        [renormalize]: 0.00051313
        [add_forward_monad_depend]: 4.91997e-06
        [auto_monad_grad]: 2.41e-06
        [auto_monad_eliminator]: 1.242e-05
        [cse]: 2.552e-05
        [replace_applicator]: 1.12e-05
    [Cycle 2]: 0.00031455, [27]
        [switch_simplify]: 5.52001e-06
        [loop_unroll]: 5.21002e-06
        [a_1]: 8.756e-05
        [with_stream_mark]: 8.47e-06
        [recompute_prepare]: 4.95001e-06
        [updatestate_depend_eliminate]: 2.71e-06
        [updatestate_assign_eliminate]: 1.98002e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 4.80001e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.10999e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.50999e-06
        [merge_forward]: 2.53e-06
        [cell_reuse_recompute_pass]: 1.61998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.182e-05
        [j_node_and_user_rematch]: 7.66999e-06
        [meta_fg_expand]: 1.39e-06
        [replace_old_param]: 7.5e-06
        [inline_without_move]: 4.55001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 5.22999e-06
        [cse]: 9.96e-06
        [replace_applicator]: 5.24e-06
[py_interpret_to_execute_after_opt_a]: 8.87999e-06
[rewriter_after_opt_a]: 3.189e-05
[convert_after_rewriter]: 6.70002e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.00047765
[jit_opt_b]: 4.938e-05, [1]
    [Cycle 1]: 4.278e-05, [2]
        [frontend_op_eliminate]: 1.603e-05
        [inline_after_opt_a]: 1.518e-05
[cconv]: 2.454e-05
[loop_unroll]: 0.00040483
[jit_opt_after_cconv]: 0.00014852, [1]
    [Cycle 1]: 0.0001424, [11]
        [c_1]: 2.03e-05
        [parameter_eliminate]: 2.37999e-06
        [updatestate_depend_eliminate]: 5.32999e-06
        [updatestate_assign_eliminate]: 1.345e-05
        [updatestate_loads_eliminate]: 2.19001e-06
        [cse]: 1.642e-05
        [call_graph_tuple_transform]: 1.82e-05
        [tuple_list_get_item_eliminator]: 5.45001e-06
        [none_parameter_eliminate]: 1.67001e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.42001e-06
[remove_dup_value]: 1.273e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 5.00999e-06
[add_recomputation]: 4.677e-05
[cse_after_recomputation]: 1.894e-05, [1]
    [Cycle 1]: 1.351e-05, [1]
        [cse]: 7.98001e-06
[auto_monad_reorder]: 1.501e-05
[get_jit_bprop_graph]: 1.66002e-06
[rewriter_after_jit_bprop_graph]: 5.32999e-06
[opt_after_jit_grad]: 0.00046919
[symbol_engine_optimizer]: 7.095e-05, [1]
    [Cycle 1]: 6.436e-05, [6]
        [build]: 3.13e-06
        [elim_shapecalc]: 7.75e-06
        [elim_not_effective]: 1.208e-05
        [opt_reshape]: 5.52001e-06
        [fold_const_symbol]: 8.10999e-06
        [renormalize]: 5.00004e-07
[validate]: 3.187e-05
[backend_pass]: 9.60019e-07
[task_emit]: 0.00606651
[execute]: 7.66999e-06
Sums
    bootstrap                            :     0.000583s :  4.07%
    type_inference                       :     0.004292s : 29.93%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000062s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.22%
    rewriter_before_opt_a                :     0.000081s :  0.56%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000060s :  0.42%
    jit_opt_a.loop_unroll                :     0.000036s :  0.25%
    jit_opt_a.a_1                        :     0.000601s :  4.19%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000513s :  3.58%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000035s :  0.25%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000032s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000478s :  3.33%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000405s :  2.82%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000013s :  0.09%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000047s :  0.33%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000469s :  3.27%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006067s : 42.30%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000184    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.82% :     0.000005s :      3: substitution.graph_param_transform
83.61% :     0.000154s :      4: substitution.inline
 1.47% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.14% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.92% :     0.000004s :      2: substitution.replace_old_param
 6.31% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004229     2
81.16% :     0.003432s :      1: type_inference.infer
18.84% :     0.000797s :      1: type_inference.specialize
------[replace.]   0.000067     6
68.23% :     0.000045s :      4: replace.inline
31.77% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     6
93.56% :     0.000152s :      4: match.inline
 6.44% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.29% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.60% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 1.95% :     0.000002s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000001s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.52% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.46% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.22% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.66% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.52% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.37% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000001s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.76% :     0.000003s :     16: predicate.switch_defer_inline
 2.60% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.49% :     0.000010s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.02% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.48% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000510    12
48.39% :     0.000247s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.88% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.73% :     0.000233s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019643    76
 0.25% :     0.000050s :      1: add_recomputation
 0.33% :     0.000066s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 3.06% :     0.000600s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.66% :     0.005041s :      1: jit_opt_a
 0.77% :     0.000151s :      1: jit_opt_after_cconv
 0.26% :     0.000052s :      1: jit_opt_b
 2.10% :     0.000413s :      1: loop_unroll
 2.48% :     0.000487s :      1: mutable_eliminate
 4.11% :     0.000808s :     26: opt.transform.jit_opt_a
 0.24% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.43% :     0.000477s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.32% :     0.000259s :      1: renormalize.infer
 1.26% :     0.000248s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000035s :      1: rewriter_after_opt_a
 0.42% :     0.000083s :      1: rewriter_before_opt_a
 0.37% :     0.000073s :      1: symbol_engine_optimizer
30.95% :     0.006080s :      1: task_emit
21.91% :     0.004304s :      1: type_inference
 0.25% :     0.000050s :      1: validate


TotalTime = 0.0124702, [30]
[bootstrap]: 0.00023909
[type_inference]: 0.00552303
[event_method]: 1.92e-05
[auto_monad]: 5.728e-05
[graph_reusing]: 4.07e-06
[pre_auto_parallel]: 1.91e-06
[py_interpret_to_execute]: 4.037e-05
[rewriter_before_opt_a]: 7.438e-05
[expand_dump_flag]: 1.92001e-06
[jit_opt_a]: 0.00460623, [2]
    [Cycle 1]: 0.00146699, [27]
        [switch_simplify]: 8.628e-05
        [loop_unroll]: 3.33e-05
        [a_1]: 0.00060512
        [with_stream_mark]: 1.233e-05
        [recompute_prepare]: 6.93e-06
        [updatestate_depend_eliminate]: 2.96001e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 5.44998e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 5.04e-06
        [merge_forward]: 2.27999e-06
        [cell_reuse_recompute_pass]: 8.40024e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.089e-05
        [j_node_and_user_rematch]: 8.72e-06
        [meta_fg_expand]: 2.27999e-06
        [replace_old_param]: 7.4e-06
        [inline_without_move]: 5.30999e-06
        [renormalize]: 0.00045546
        [add_forward_monad_depend]: 4.13001e-06
        [auto_monad_grad]: 1.27999e-06
        [auto_monad_eliminator]: 9.44e-06
        [cse]: 1.464e-05
        [replace_applicator]: 1.087e-05
    [Cycle 2]: 0.00030072, [27]
        [switch_simplify]: 5.54998e-06
        [loop_unroll]: 4.61002e-06
        [a_1]: 8.671e-05
        [with_stream_mark]: 7.97e-06
        [recompute_prepare]: 4.86997e-06
        [updatestate_depend_eliminate]: 2.37999e-06
        [updatestate_assign_eliminate]: 1.89999e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.74e-06
        [updatestate_useless_node_eliminater]: 4.76002e-06
        [accelerated_algorithm]: 4.75999e-06
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.30999e-06
        [merge_forward]: 2.38002e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.211e-05
        [j_node_and_user_rematch]: 7.53999e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 6.93e-06
        [inline_without_move]: 4.65999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 6.50005e-07
        [auto_monad_eliminator]: 4.4e-06
        [cse]: 8.99e-06
        [replace_applicator]: 5.08002e-06
[py_interpret_to_execute_after_opt_a]: 7.93001e-06
[rewriter_after_opt_a]: 2.583e-05
[convert_after_rewriter]: 5.17e-06
[order_py_execute_after_rewriter]: 3.81999e-06
[mutable_eliminate]: 0.00045988
[jit_opt_b]: 5.187e-05, [1]
    [Cycle 1]: 3.966e-05, [2]
        [frontend_op_eliminate]: 1.451e-05
        [inline_after_opt_a]: 1.376e-05
[cconv]: 1.457e-05
[loop_unroll]: 0.00045141
[jit_opt_after_cconv]: 0.0001293, [1]
    [Cycle 1]: 0.00012305, [11]
        [c_1]: 2.035e-05
        [parameter_eliminate]: 2.59999e-06
        [updatestate_depend_eliminate]: 5.12e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [cse]: 1.528e-05
        [call_graph_tuple_transform]: 1.559e-05
        [tuple_list_get_item_eliminator]: 5.15999e-06
        [none_parameter_eliminate]: 9.10019e-07
        [renormalize]: 6.40022e-07
        [switch_simplify]: 5.21002e-06
[remove_dup_value]: 6.35002e-06
[partial_unused_args_eliminate]: 1.62999e-06
[environ_conv]: 3.77998e-06
[add_recomputation]: 2.97e-05
[cse_after_recomputation]: 1.978e-05, [1]
    [Cycle 1]: 1.463e-05, [1]
        [cse]: 9.00001e-06
[auto_monad_reorder]: 8.82999e-06
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 4.72e-06
[opt_after_jit_grad]: 0.00043839
[symbol_engine_optimizer]: 6.88e-05, [1]
    [Cycle 1]: 6.28e-05, [6]
        [build]: 2.59001e-06
        [elim_shapecalc]: 7.17002e-06
        [elim_not_effective]: 1.169e-05
        [opt_reshape]: 5.48002e-06
        [fold_const_symbol]: 7.65e-06
        [renormalize]: 4.69998e-07
[validate]: 2.436e-05
Sums
    bootstrap                            :     0.000239s :  2.63%
    type_inference                       :     0.005523s : 60.71%
    event_method                         :     0.000019s :  0.21%
    auto_monad                           :     0.000057s :  0.63%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.44%
    rewriter_before_opt_a                :     0.000074s :  0.82%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000092s :  1.01%
    jit_opt_a.loop_unroll                :     0.000038s :  0.42%
    jit_opt_a.a_1                        :     0.000692s :  7.60%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000014s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000456s :  5.01%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.15%
    jit_opt_a.cse                        :     0.000024s :  0.26%
    jit_opt_a.replace_applicator         :     0.000016s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.09%
    rewriter_after_opt_a                 :     0.000026s :  0.28%
    convert_after_rewriter               :     0.000005s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000460s :  5.05%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.15%
    cconv                                :     0.000015s :  0.16%
    loop_unroll                          :     0.000451s :  4.96%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000006s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000030s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000009s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000438s :  4.82%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000024s :  0.27%


Time group info:
------[substitution.]   0.000210    26
 0.65% :     0.000001s :      2: substitution.elim_not_effective
 0.42% :     0.000001s :      2: substitution.fold_const_symbol
 1.54% :     0.000003s :      3: substitution.graph_param_transform
85.98% :     0.000181s :      6: substitution.inline
 1.22% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.94% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.28% :     0.000003s :      2: substitution.replace_old_param
 3.33% :     0.000007s :      1: substitution.switch_simplify
 3.63% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005476     2
83.64% :     0.004581s :      1: type_inference.infer
16.36% :     0.000896s :      1: type_inference.specialize
------[replace.]   0.000107     9
55.16% :     0.000059s :      6: replace.inline
19.10% :     0.000020s :      1: replace.switch_simplify
25.74% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000190     9
93.22% :     0.000178s :      6: match.inline
 3.28% :     0.000006s :      1: match.switch_simplify
 3.50% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   801
 1.43% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.20% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000001s :     12: predicate.addn_check_dump
 1.44% :     0.000002s :     12: predicate.addn_zero_filter
 1.86% :     0.000002s :     12: predicate.arithmetic_simplify
 1.31% :     0.000002s :     12: predicate.cast_eliminate
 0.35% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.20% :     0.000001s :     12: predicate.compare_switch_simplify
 1.38% :     0.000002s :     12: predicate.depend_value_elim
 1.19% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.50% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.22% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     12: predicate.environ_get_eliminate
 1.21% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.16% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.36% :     0.000007s :     26: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :      6: predicate.less_batch_normalization
 1.64% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     17: predicate.load_eliminater
 1.30% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.30% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.83% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.25% :     0.000002s :     12: predicate.merge_addn
 1.22% :     0.000001s :     12: predicate.minmaximum_grad
 1.56% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     17: predicate.partial_eliminate
 1.24% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     12: predicate.reduce_eliminate
 1.58% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     20: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     12: predicate.reshape_eliminate
 1.31% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     12: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.75% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.81% :     0.000003s :     20: predicate.switch_defer_inline
 2.38% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.21% :     0.000010s :     65: predicate.switch_simplify
 1.23% :     0.000002s :     12: predicate.tile_eliminate
 1.29% :     0.000002s :     12: predicate.transpose_eliminate
 1.43% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.28% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000002s :     12: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000654    15
47.92% :     0.000313s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.00% :     0.000072s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.08% :     0.000268s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013921    72
 0.23% :     0.000033s :      1: add_recomputation
 0.44% :     0.000061s :      1: auto_monad
 0.08% :     0.000011s :      1: auto_monad_reorder
 1.81% :     0.000252s :      1: bootstrap
 0.12% :     0.000017s :      1: cconv
 0.05% :     0.000007s :      1: convert_after_rewriter
 0.16% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.16% :     0.000023s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
33.11% :     0.004609s :      1: jit_opt_a
 0.95% :     0.000132s :      1: jit_opt_after_cconv
 0.39% :     0.000054s :      1: jit_opt_b
 3.30% :     0.000459s :      1: loop_unroll
 3.36% :     0.000468s :      1: mutable_eliminate
 6.67% :     0.000929s :     26: opt.transform.jit_opt_a
 0.31% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000022s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000029s :      4: opt.transform.symbol_engine_opt
 3.21% :     0.000446s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.31% :     0.000043s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000009s :      1: remove_dup_value
 1.69% :     0.000235s :      1: renormalize.infer
 1.54% :     0.000215s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000029s :      1: rewriter_after_opt_a
 0.56% :     0.000077s :      1: rewriter_before_opt_a
 0.51% :     0.000071s :      1: symbol_engine_optimizer
39.75% :     0.005533s :      1: type_inference


TotalTime = 0.0151244, [33]
[bootstrap]: 0.00023519
[type_inference]: 0.00377516
[event_method]: 1.388e-05
[auto_monad]: 4.608e-05
[graph_reusing]: 4.17e-06
[pre_auto_parallel]: 1.94e-06
[py_interpret_to_execute]: 2.768e-05
[rewriter_before_opt_a]: 6.39e-05
[expand_dump_flag]: 2.81999e-06
[jit_opt_a]: 0.00444597, [2]
    [Cycle 1]: 0.00122993, [27]
        [switch_simplify]: 3.933e-05
        [loop_unroll]: 2.755e-05
        [a_1]: 0.00044956
        [with_stream_mark]: 1.069e-05
        [recompute_prepare]: 6.38e-06
        [updatestate_depend_eliminate]: 2.64001e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.22999e-06
        [updatestate_useless_node_eliminater]: 4.73001e-06
        [accelerated_algorithm]: 5.40999e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 5.05999e-06
        [merge_forward]: 2.29999e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.031e-05
        [j_node_and_user_rematch]: 8.07e-06
        [meta_fg_expand]: 1.70001e-06
        [replace_old_param]: 7.41999e-06
        [inline_without_move]: 4.86002e-06
        [renormalize]: 0.00044482
        [add_forward_monad_depend]: 4.08001e-06
        [auto_monad_grad]: 1.09e-06
        [auto_monad_eliminator]: 9.02e-06
        [cse]: 1.503e-05
        [replace_applicator]: 1.02e-05
    [Cycle 2]: 0.00030294, [27]
        [switch_simplify]: 5.10001e-06
        [loop_unroll]: 4.79e-06
        [a_1]: 8.628e-05
        [with_stream_mark]: 7.73001e-06
        [recompute_prepare]: 4.72e-06
        [updatestate_depend_eliminate]: 2.27999e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.67e-06
        [updatestate_useless_node_eliminater]: 4.57e-06
        [accelerated_algorithm]: 4.90999e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.30999e-06
        [merge_forward]: 2.16e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.14e-05
        [j_node_and_user_rematch]: 7.21001e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 6.74999e-06
        [inline_without_move]: 4.48001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 4.75999e-06
        [cse]: 8.85001e-06
        [replace_applicator]: 4.86002e-06
[py_interpret_to_execute_after_opt_a]: 7.37002e-06
[rewriter_after_opt_a]: 2.449e-05
[convert_after_rewriter]: 5.32001e-06
[order_py_execute_after_rewriter]: 4.17e-06
[mutable_eliminate]: 0.00046118
[jit_opt_b]: 4.68e-05, [1]
    [Cycle 1]: 4.103e-05, [2]
        [frontend_op_eliminate]: 1.537e-05
        [inline_after_opt_a]: 1.405e-05
[cconv]: 1.558e-05
[loop_unroll]: 0.00040527
[jit_opt_after_cconv]: 0.00012537, [1]
    [Cycle 1]: 0.00011957, [11]
        [c_1]: 1.897e-05
        [parameter_eliminate]: 2.66e-06
        [updatestate_depend_eliminate]: 4.21001e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [cse]: 1.493e-05
        [call_graph_tuple_transform]: 1.534e-05
        [tuple_list_get_item_eliminator]: 5.10999e-06
        [none_parameter_eliminate]: 8.2e-07
        [renormalize]: 4.39992e-07
        [switch_simplify]: 4.97e-06
[remove_dup_value]: 6.96001e-06
[partial_unused_args_eliminate]: 1.52001e-06
[environ_conv]: 3.42002e-06
[add_recomputation]: 7.015e-05
[cse_after_recomputation]: 2.2e-05, [1]
    [Cycle 1]: 1.602e-05, [1]
        [cse]: 1.013e-05
[auto_monad_reorder]: 9.12001e-06
[get_jit_bprop_graph]: 1.43002e-06
[rewriter_after_jit_bprop_graph]: 4.43999e-06
[opt_after_jit_grad]: 0.00044344
[symbol_engine_optimizer]: 6.649e-05, [1]
    [Cycle 1]: 6.08e-05, [6]
        [build]: 2.19001e-06
        [elim_shapecalc]: 7.19001e-06
        [elim_not_effective]: 1.163e-05
        [opt_reshape]: 5.32999e-06
        [fold_const_symbol]: 7.81001e-06
        [renormalize]: 5.90022e-07
[validate]: 2.305e-05
[backend_pass]: 6.80011e-07
[task_emit]: 0.0046006
[execute]: 4.60999e-06
Sums
    bootstrap                            :     0.000235s :  2.01%
    type_inference                       :     0.003775s : 32.31%
    event_method                         :     0.000014s :  0.12%
    auto_monad                           :     0.000046s :  0.39%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.24%
    rewriter_before_opt_a                :     0.000064s :  0.55%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000044s :  0.38%
    jit_opt_a.loop_unroll                :     0.000032s :  0.28%
    jit_opt_a.a_1                        :     0.000536s :  4.59%
    jit_opt_a.with_stream_mark           :     0.000018s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.08%
    jit_opt_a.merge_forward              :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000015s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.12%
    jit_opt_a.inline_without_move        :     0.000009s :  0.08%
    jit_opt_a.renormalize                :     0.000445s :  3.81%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.12%
    jit_opt_a.cse                        :     0.000024s :  0.20%
    jit_opt_a.replace_applicator         :     0.000015s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.06%
    rewriter_after_opt_a                 :     0.000024s :  0.21%
    convert_after_rewriter               :     0.000005s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000461s :  3.95%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.12%
    cconv                                :     0.000016s :  0.13%
    loop_unroll                          :     0.000405s :  3.47%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000007s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000003s :  0.03%
    add_recomputation                    :     0.000070s :  0.60%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000009s :  0.08%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.04%
    opt_after_jit_grad                   :     0.000443s :  3.80%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000023s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.004601s : 39.38%
    execute                              :     0.000005s :  0.04%


Time group info:
------[substitution.]   0.000145    23
 1.20% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.60% :     0.000004s :      3: substitution.graph_param_transform
82.74% :     0.000120s :      4: substitution.inline
 1.81% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.75% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000003s :      2: substitution.replace_old_param
 6.10% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003730     2
81.43% :     0.003037s :      1: type_inference.infer
18.57% :     0.000693s :      1: type_inference.specialize
------[replace.]   0.000059     6
66.25% :     0.000039s :      4: replace.inline
33.75% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000125     6
93.91% :     0.000117s :      4: match.inline
 6.09% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000105   691
 1.49% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000001s :     10: predicate.addn_zero_filter
 2.22% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.18% :     0.000001s :     10: predicate.compare_switch_simplify
 1.45% :     0.000002s :     10: predicate.depend_value_elim
 1.29% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.39% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.41% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.09% :     0.000005s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.88% :     0.000002s :     15: predicate.load_eliminater
 1.48% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.11% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.08% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.43% :     0.000001s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000002s :     15: predicate.partial_eliminate
 1.28% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     10: predicate.reduce_eliminate
 1.63% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.40% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000001s :     10: predicate.reshape_eliminate
 1.32% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.29% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.62% :     0.000008s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.28% :     0.000001s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.38% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.79% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000466    12
46.67% :     0.000217s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.39% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.94% :     0.000223s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016388    76
 0.45% :     0.000074s :      1: add_recomputation
 0.30% :     0.000049s :      1: auto_monad
 0.07% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.48% :     0.000243s :      1: bootstrap
 0.11% :     0.000018s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.15% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.10% :     0.000017s :      1: event_method
 0.05% :     0.000008s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
27.14% :     0.004448s :      1: jit_opt_a
 0.78% :     0.000128s :      1: jit_opt_after_cconv
 0.30% :     0.000049s :      1: jit_opt_b
 2.52% :     0.000413s :      1: loop_unroll
 2.86% :     0.000469s :      1: mutable_eliminate
 4.37% :     0.000717s :     26: opt.transform.jit_opt_a
 0.25% :     0.000041s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.75% :     0.000450s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000009s :      1: remove_dup_value
 1.35% :     0.000221s :      1: renormalize.infer
 1.33% :     0.000218s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000027s :      1: rewriter_after_opt_a
 0.41% :     0.000067s :      1: rewriter_before_opt_a
 0.42% :     0.000069s :      1: symbol_engine_optimizer
28.12% :     0.004609s :      1: task_emit
23.08% :     0.003783s :      1: type_inference
 0.23% :     0.000038s :      1: validate


TotalTime = 0.0125598, [30]
[bootstrap]: 0.00023074
[type_inference]: 0.00555577
[event_method]: 1.892e-05
[auto_monad]: 7.894e-05
[graph_reusing]: 4.38001e-06
[pre_auto_parallel]: 1.99e-06
[py_interpret_to_execute]: 3.82e-05
[rewriter_before_opt_a]: 7.411e-05
[expand_dump_flag]: 2.14e-06
[jit_opt_a]: 0.00468212, [2]
    [Cycle 1]: 0.0014937, [27]
        [switch_simplify]: 8.522e-05
        [loop_unroll]: 3.294e-05
        [a_1]: 0.00060596
        [with_stream_mark]: 1.18e-05
        [recompute_prepare]: 6.53998e-06
        [updatestate_depend_eliminate]: 2.95998e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 5.66e-06
        [updatestate_useless_node_eliminater]: 4.78001e-06
        [accelerated_algorithm]: 5.19998e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 8.39995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.07e-05
        [j_node_and_user_rematch]: 8.11002e-06
        [meta_fg_expand]: 1.84e-06
        [replace_old_param]: 7.31001e-06
        [inline_without_move]: 5.25001e-06
        [renormalize]: 0.00048564
        [add_forward_monad_depend]: 4.83001e-06
        [auto_monad_grad]: 1.39e-06
        [auto_monad_eliminator]: 9.44998e-06
        [cse]: 1.463e-05
        [replace_applicator]: 1.009e-05
    [Cycle 2]: 0.00030493, [27]
        [switch_simplify]: 5.20999e-06
        [loop_unroll]: 4.65001e-06
        [a_1]: 8.775e-05
        [with_stream_mark]: 8.14997e-06
        [recompute_prepare]: 4.88001e-06
        [updatestate_depend_eliminate]: 2.51998e-06
        [updatestate_assign_eliminate]: 1.97999e-06
        [updatestate_loads_eliminate]: 1.76998e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 4.94e-06
        [updatestate_useless_node_eliminater]: 4.74998e-06
        [accelerated_algorithm]: 4.97e-06
        [meta_shard_fg_expand]: 1.02e-06
        [get_grad_eliminate_]: 4.55001e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 1.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.187e-05
        [j_node_and_user_rematch]: 7.28999e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 7.11999e-06
        [inline_without_move]: 4.52e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 4.75999e-06
        [cse]: 8.79e-06
        [replace_applicator]: 4.85001e-06
[py_interpret_to_execute_after_opt_a]: 7.21999e-06
[rewriter_after_opt_a]: 2.273e-05
[convert_after_rewriter]: 5.35001e-06
[order_py_execute_after_rewriter]: 3.75998e-06
[mutable_eliminate]: 0.00046053
[jit_opt_b]: 4.695e-05, [1]
    [Cycle 1]: 4.089e-05, [2]
        [frontend_op_eliminate]: 1.56e-05
        [inline_after_opt_a]: 1.424e-05
[cconv]: 1.602e-05
[loop_unroll]: 0.00042001
[jit_opt_after_cconv]: 0.00012965, [1]
    [Cycle 1]: 0.00012369, [11]
        [c_1]: 2e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 5.61e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [cse]: 1.517e-05
        [call_graph_tuple_transform]: 1.596e-05
        [tuple_list_get_item_eliminator]: 4.95001e-06
        [none_parameter_eliminate]: 9.5999e-07
        [renormalize]: 3.19997e-07
        [switch_simplify]: 5.09e-06
[remove_dup_value]: 6.53998e-06
[partial_unused_args_eliminate]: 1.29998e-06
[environ_conv]: 4.46002e-06
[add_recomputation]: 3.189e-05
[cse_after_recomputation]: 1.957e-05, [1]
    [Cycle 1]: 1.45e-05, [1]
        [cse]: 8.84e-06
[auto_monad_reorder]: 8.55001e-06
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 4.63999e-06
[opt_after_jit_grad]: 0.00044516
[symbol_engine_optimizer]: 6.825e-05, [1]
    [Cycle 1]: 6.235e-05, [6]
        [build]: 2.27999e-06
        [elim_shapecalc]: 7.2e-06
        [elim_not_effective]: 1.187e-05
        [opt_reshape]: 5.35999e-06
        [fold_const_symbol]: 8.14002e-06
        [renormalize]: 4.89992e-07
[validate]: 2.349e-05
Sums
    bootstrap                            :     0.000231s :  2.52%
    type_inference                       :     0.005556s : 60.74%
    event_method                         :     0.000019s :  0.21%
    auto_monad                           :     0.000079s :  0.86%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.42%
    rewriter_before_opt_a                :     0.000074s :  0.81%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000090s :  0.99%
    jit_opt_a.loop_unroll                :     0.000038s :  0.41%
    jit_opt_a.a_1                        :     0.000694s :  7.58%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000015s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.04%
    jit_opt_a.replace_old_param          :     0.000014s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000486s :  5.31%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.16%
    jit_opt_a.cse                        :     0.000023s :  0.26%
    jit_opt_a.replace_applicator         :     0.000015s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.08%
    rewriter_after_opt_a                 :     0.000023s :  0.25%
    convert_after_rewriter               :     0.000005s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000461s :  5.03%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.16%
    cconv                                :     0.000016s :  0.18%
    loop_unroll                          :     0.000420s :  4.59%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000007s :  0.07%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000032s :  0.35%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000009s :  0.09%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000445s :  4.87%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000023s :  0.26%


Time group info:
------[substitution.]   0.000211    26
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 1.92% :     0.000004s :      3: substitution.graph_param_transform
85.35% :     0.000180s :      6: substitution.inline
 1.18% :     0.000002s :      4: substitution.j_node_and_user_rematch
 1.91% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.25% :     0.000003s :      2: substitution.replace_old_param
 3.36% :     0.000007s :      1: substitution.switch_simplify
 3.76% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005510     2
84.06% :     0.004631s :      1: type_inference.infer
15.94% :     0.000878s :      1: type_inference.specialize
------[replace.]   0.000107     9
56.03% :     0.000060s :      6: replace.inline
18.70% :     0.000020s :      1: replace.switch_simplify
25.27% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000190     9
93.08% :     0.000177s :      6: match.inline
 3.31% :     0.000006s :      1: match.switch_simplify
 3.61% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   801
 1.47% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.03% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.18% :     0.000001s :     12: predicate.addn_check_dump
 1.49% :     0.000002s :     12: predicate.addn_zero_filter
 1.96% :     0.000002s :     12: predicate.arithmetic_simplify
 1.41% :     0.000002s :     12: predicate.cast_eliminate
 0.35% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.19% :     0.000001s :     12: predicate.compare_switch_simplify
 1.39% :     0.000002s :     12: predicate.depend_value_elim
 1.25% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.54% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.34% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.33% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.21% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.20% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     12: predicate.environ_get_eliminate
 1.27% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.38% :     0.000007s :     26: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     17: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.29% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.27% :     0.000002s :     12: predicate.merge_addn
 1.19% :     0.000001s :     12: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.39% :     0.000000s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     17: predicate.partial_eliminate
 1.31% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     12: predicate.reduce_eliminate
 1.53% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     20: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     12: predicate.reshape_eliminate
 1.32% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     12: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.76% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.81% :     0.000003s :     20: predicate.switch_defer_inline
 2.43% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.09% :     0.000010s :     65: predicate.switch_simplify
 1.30% :     0.000002s :     12: predicate.tile_eliminate
 1.37% :     0.000002s :     12: predicate.transpose_eliminate
 1.41% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.17% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     12: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000648    15
49.35% :     0.000320s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.13% :     0.000072s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.51% :     0.000256s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014043    72
 0.25% :     0.000035s :      1: add_recomputation
 0.59% :     0.000083s :      1: auto_monad
 0.08% :     0.000011s :      1: auto_monad_reorder
 1.72% :     0.000242s :      1: bootstrap
 0.13% :     0.000019s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000006s :      1: environ_conv
 0.16% :     0.000022s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
33.36% :     0.004685s :      1: jit_opt_a
 0.94% :     0.000132s :      1: jit_opt_after_cconv
 0.35% :     0.000050s :      1: jit_opt_b
 3.05% :     0.000428s :      1: loop_unroll
 3.33% :     0.000468s :      1: mutable_eliminate
 6.61% :     0.000928s :     26: opt.transform.jit_opt_a
 0.30% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.23% :     0.000453s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.29% :     0.000041s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000009s :      1: remove_dup_value
 1.89% :     0.000266s :      1: renormalize.infer
 1.53% :     0.000215s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000026s :      1: rewriter_after_opt_a
 0.55% :     0.000077s :      1: rewriter_before_opt_a
 0.50% :     0.000071s :      1: symbol_engine_optimizer
39.64% :     0.005567s :      1: type_inference


TotalTime = 0.0234002, [33]
[bootstrap]: 0.00126003
[type_inference]: 0.00547616
[event_method]: 1.688e-05
[auto_monad]: 6.621e-05
[graph_reusing]: 5.85002e-06
[pre_auto_parallel]: 3.16999e-06
[py_interpret_to_execute]: 3.685e-05
[rewriter_before_opt_a]: 7.499e-05
[expand_dump_flag]: 3.28e-06
[jit_opt_a]: 0.006654, [2]
    [Cycle 1]: 0.00220359, [27]
        [switch_simplify]: 7.249e-05
        [loop_unroll]: 3.171e-05
        [a_1]: 0.00065019
        [with_stream_mark]: 8.672e-05
        [recompute_prepare]: 8.75999e-06
        [updatestate_depend_eliminate]: 4.48001e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 2.78998e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 6.41e-06
        [updatestate_useless_node_eliminater]: 5.91e-06
        [accelerated_algorithm]: 6.59001e-06
        [meta_shard_fg_expand]: 2.29001e-06
        [get_grad_eliminate_]: 5.73002e-06
        [merge_forward]: 3.68999e-06
        [cell_reuse_recompute_pass]: 1.50999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.589e-05
        [j_node_and_user_rematch]: 9.89001e-06
        [meta_fg_expand]: 2.17999e-06
        [replace_old_param]: 1.044e-05
        [inline_without_move]: 5.30001e-06
        [renormalize]: 0.00099075
        [add_forward_monad_depend]: 7.08998e-06
        [auto_monad_grad]: 2.73998e-06
        [auto_monad_eliminator]: 1.647e-05
        [cse]: 3.17e-05
        [replace_applicator]: 1.617e-05
    [Cycle 2]: 0.00034213, [27]
        [switch_simplify]: 5.66e-06
        [loop_unroll]: 5.99e-06
        [a_1]: 9.568e-05
        [with_stream_mark]: 1.133e-05
        [recompute_prepare]: 4.82998e-06
        [updatestate_depend_eliminate]: 2.97002e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 5.09e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 5.43002e-06
        [meta_shard_fg_expand]: 1.64e-06
        [get_grad_eliminate_]: 4.94998e-06
        [merge_forward]: 3.14999e-06
        [cell_reuse_recompute_pass]: 3.23e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.394e-05
        [j_node_and_user_rematch]: 9.17001e-06
        [meta_fg_expand]: 1.69998e-06
        [replace_old_param]: 8.1e-06
        [inline_without_move]: 4.70999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.36002e-06
        [auto_monad_grad]: 1.03001e-06
        [auto_monad_eliminator]: 5.17e-06
        [cse]: 1.058e-05
        [replace_applicator]: 5.25001e-06
[py_interpret_to_execute_after_opt_a]: 1.38e-05
[rewriter_after_opt_a]: 4.059e-05
[convert_after_rewriter]: 7.31001e-06
[order_py_execute_after_rewriter]: 5.07e-06
[mutable_eliminate]: 0.00079594
[jit_opt_b]: 5.227e-05, [1]
    [Cycle 1]: 4.428e-05, [2]
        [frontend_op_eliminate]: 1.745e-05
        [inline_after_opt_a]: 1.5e-05
[cconv]: 2.835e-05
[loop_unroll]: 0.000464
[jit_opt_after_cconv]: 0.00015053, [1]
    [Cycle 1]: 0.00014361, [11]
        [c_1]: 2.035e-05
        [parameter_eliminate]: 3.3e-06
        [updatestate_depend_eliminate]: 6.44001e-06
        [updatestate_assign_eliminate]: 2.74001e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [cse]: 2.321e-05
        [call_graph_tuple_transform]: 2.102e-05
        [tuple_list_get_item_eliminator]: 5.29e-06
        [none_parameter_eliminate]: 1.58002e-06
        [renormalize]: 1.09e-06
        [switch_simplify]: 5.59e-06
[remove_dup_value]: 1.632e-05
[partial_unused_args_eliminate]: 2.78998e-06
[environ_conv]: 6.29999e-06
[add_recomputation]: 5.37e-05
[cse_after_recomputation]: 2.146e-05, [1]
    [Cycle 1]: 1.588e-05, [1]
        [cse]: 9.57999e-06
[auto_monad_reorder]: 1.643e-05
[get_jit_bprop_graph]: 2.36e-06
[rewriter_after_jit_bprop_graph]: 6.53003e-06
[opt_after_jit_grad]: 0.00067481
[symbol_engine_optimizer]: 8.431e-05, [1]
    [Cycle 1]: 7.553e-05, [6]
        [build]: 5.16998e-06
        [elim_shapecalc]: 8.57e-06
        [elim_not_effective]: 1.53e-05
        [opt_reshape]: 6.77002e-06
        [fold_const_symbol]: 9.87001e-06
        [renormalize]: 1.11997e-06
[validate]: 4.782e-05
[backend_pass]: 1.93997e-06
[task_emit]: 0.00706723
[execute]: 9.41e-06
Sums
    bootstrap                            :     0.001260s :  6.77%
    type_inference                       :     0.005476s : 29.43%
    event_method                         :     0.000017s :  0.09%
    auto_monad                           :     0.000066s :  0.36%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.20%
    rewriter_before_opt_a                :     0.000075s :  0.40%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000078s :  0.42%
    jit_opt_a.loop_unroll                :     0.000038s :  0.20%
    jit_opt_a.a_1                        :     0.000746s :  4.01%
    jit_opt_a.with_stream_mark           :     0.000098s :  0.53%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000991s :  5.32%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.12%
    jit_opt_a.cse                        :     0.000042s :  0.23%
    jit_opt_a.replace_applicator         :     0.000021s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.07%
    rewriter_after_opt_a                 :     0.000041s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000796s :  4.28%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000028s :  0.15%
    loop_unroll                          :     0.000464s :  2.49%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000054s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000675s :  3.63%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000048s :  0.26%
    backend_pass                         :     0.000002s :  0.01%
    task_emit                            :     0.007067s : 37.98%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000291    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000002s :      2: substitution.fold_const_symbol
 2.07% :     0.000006s :      3: substitution.graph_param_transform
86.57% :     0.000252s :      4: substitution.inline
 1.47% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.65% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000006s :      2: substitution.replace_old_param
 4.75% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005404     2
82.63% :     0.004465s :      1: type_inference.infer
17.37% :     0.000939s :      1: type_inference.specialize
------[replace.]   0.000084     6
67.16% :     0.000056s :      4: replace.inline
32.84% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000262     6
95.11% :     0.000249s :      4: match.inline
 4.89% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.48% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.69% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.19% :     0.000002s :     10: predicate.addn_zero_filter
 2.20% :     0.000003s :     10: predicate.arithmetic_simplify
 1.37% :     0.000002s :     10: predicate.cast_eliminate
 0.35% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.90% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.71% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.20% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.77% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.34% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.22% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.34% :     0.000002s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.13% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.26% :     0.000007s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.70% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.59% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.19% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.91% :     0.000001s :     10: predicate.merge_addn
 0.95% :     0.000001s :     10: predicate.minmaximum_grad
 1.63% :     0.000002s :      3: predicate.mutable_eliminate
 0.76% :     0.000001s :      3: predicate.opt_reshape
 1.97% :     0.000003s :     15: predicate.partial_eliminate
 1.41% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.85% :     0.000002s :     10: predicate.reduce_eliminate
 1.33% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000002s :     10: predicate.reshape_eliminate
 1.32% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.04% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.32% :     0.000003s :     16: predicate.switch_defer_inline
 1.91% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.72% :     0.000010s :     53: predicate.switch_simplify
 1.45% :     0.000002s :     10: predicate.tile_eliminate
 1.24% :     0.000002s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.82% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.81% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.41% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000002s :     10: predicate.value_based_eliminate
 0.48% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000741    12
48.61% :     0.000360s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.27% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.12% :     0.000327s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025476    76
 0.22% :     0.000057s :      1: add_recomputation
 0.27% :     0.000069s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 5.03% :     0.001282s :      1: bootstrap
 0.12% :     0.000031s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.08% :     0.000021s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.14% :     0.006659s :      1: jit_opt_a
 0.60% :     0.000153s :      1: jit_opt_after_cconv
 0.22% :     0.000055s :      1: jit_opt_b
 1.85% :     0.000472s :      1: loop_unroll
 3.16% :     0.000806s :      1: mutable_eliminate
 3.87% :     0.000986s :     26: opt.transform.jit_opt_a
 0.19% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.69% :     0.000685s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.15% :     0.000039s :      1: py_interpret_to_execute
 0.06% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 2.05% :     0.000523s :      1: renormalize.infer
 1.80% :     0.000458s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000043s :      1: rewriter_after_opt_a
 0.31% :     0.000078s :      1: rewriter_before_opt_a
 0.34% :     0.000087s :      1: symbol_engine_optimizer
27.81% :     0.007086s :      1: task_emit
21.55% :     0.005490s :      1: type_inference
 0.28% :     0.000072s :      1: validate


TotalTime = 0.0166826, [33]
[bootstrap]: 0.00026504
[type_inference]: 0.00411586
[event_method]: 1.58e-05
[auto_monad]: 6.048e-05
[graph_reusing]: 6.01e-06
[pre_auto_parallel]: 2.43998e-06
[py_interpret_to_execute]: 2.932e-05
[rewriter_before_opt_a]: 6.917e-05
[expand_dump_flag]: 3.46999e-06
[jit_opt_a]: 0.00464977, [2]
    [Cycle 1]: 0.00135308, [27]
        [switch_simplify]: 4.752e-05
        [loop_unroll]: 2.904e-05
        [a_1]: 0.00049478
        [with_stream_mark]: 1.468e-05
        [recompute_prepare]: 6.28002e-06
        [updatestate_depend_eliminate]: 3.22002e-06
        [updatestate_assign_eliminate]: 3.04999e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 5.85002e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.59998e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 5.18002e-06
        [merge_forward]: 3.61999e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.388e-05
        [j_node_and_user_rematch]: 8.79e-06
        [meta_fg_expand]: 1.98002e-06
        [replace_old_param]: 8.19998e-06
        [inline_without_move]: 5.50001e-06
        [renormalize]: 0.00046612
        [add_forward_monad_depend]: 5.04e-06
        [auto_monad_grad]: 1.64e-06
        [auto_monad_eliminator]: 1.245e-05
        [cse]: 2.62e-05
        [replace_applicator]: 1.159e-05
    [Cycle 2]: 0.00032125, [27]
        [switch_simplify]: 5.59998e-06
        [loop_unroll]: 4.90001e-06
        [a_1]: 9.438e-05
        [with_stream_mark]: 7.85e-06
        [recompute_prepare]: 4.55001e-06
        [updatestate_depend_eliminate]: 2.36998e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 4.77e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 4.94003e-06
        [meta_shard_fg_expand]: 1.01002e-06
        [get_grad_eliminate_]: 4.50999e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.56002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.312e-05
        [j_node_and_user_rematch]: 8.01001e-06
        [meta_fg_expand]: 1.53002e-06
        [replace_old_param]: 7.04001e-06
        [inline_without_move]: 5.00001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.03001e-06
        [auto_monad_grad]: 7.10017e-07
        [auto_monad_eliminator]: 4.93001e-06
        [cse]: 9.29e-06
        [replace_applicator]: 5.14e-06
[py_interpret_to_execute_after_opt_a]: 7.7e-06
[rewriter_after_opt_a]: 3.126e-05
[convert_after_rewriter]: 7.06001e-06
[order_py_execute_after_rewriter]: 5.16998e-06
[mutable_eliminate]: 0.00047063
[jit_opt_b]: 4.604e-05, [1]
    [Cycle 1]: 3.974e-05, [2]
        [frontend_op_eliminate]: 1.502e-05
        [inline_after_opt_a]: 1.369e-05
[cconv]: 2.31e-05
[loop_unroll]: 0.00039766
[jit_opt_after_cconv]: 0.00013074, [1]
    [Cycle 1]: 0.00012455, [11]
        [c_1]: 1.983e-05
        [parameter_eliminate]: 2.39001e-06
        [updatestate_depend_eliminate]: 4.79e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [cse]: 1.558e-05
        [call_graph_tuple_transform]: 1.735e-05
        [tuple_list_get_item_eliminator]: 5.23002e-06
        [none_parameter_eliminate]: 1.24e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.05999e-06
[remove_dup_value]: 1.321e-05
[partial_unused_args_eliminate]: 2.32001e-06
[environ_conv]: 4.89e-06
[add_recomputation]: 4.348e-05
[cse_after_recomputation]: 1.772e-05, [1]
    [Cycle 1]: 1.289e-05, [1]
        [cse]: 7.48e-06
[auto_monad_reorder]: 1.407e-05
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 5.05999e-06
[opt_after_jit_grad]: 0.00043602
[symbol_engine_optimizer]: 6.821e-05, [1]
    [Cycle 1]: 6.245e-05, [6]
        [build]: 3.06001e-06
        [elim_shapecalc]: 6.73e-06
        [elim_not_effective]: 1.195e-05
        [opt_reshape]: 5.30001e-06
        [fold_const_symbol]: 8.30999e-06
        [renormalize]: 4.50003e-07
[validate]: 2.993e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00553084
[execute]: 7.5e-06
Sums
    bootstrap                            :     0.000265s :  2.02%
    type_inference                       :     0.004116s : 31.31%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000060s :  0.46%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.22%
    rewriter_before_opt_a                :     0.000069s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000053s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.26%
    jit_opt_a.a_1                        :     0.000589s :  4.48%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.000466s :  3.55%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000035s :  0.27%
    jit_opt_a.replace_applicator         :     0.000017s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000471s :  3.58%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000023s :  0.18%
    loop_unroll                          :     0.000398s :  3.03%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000043s :  0.33%
    cse_after_recomputation.cse          :     0.000007s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000436s :  3.32%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000030s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005531s : 42.08%
    execute                              :     0.000007s :  0.06%


Time group info:
------[substitution.]   0.000171    23
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000001s :      2: substitution.fold_const_symbol
 2.93% :     0.000005s :      3: substitution.graph_param_transform
81.90% :     0.000140s :      4: substitution.inline
 1.88% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.98% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.93% :     0.000003s :      2: substitution.replace_old_param
 6.59% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004060     2
81.95% :     0.003327s :      1: type_inference.infer
18.05% :     0.000733s :      1: type_inference.specialize
------[replace.]   0.000060     6
67.39% :     0.000040s :      4: replace.inline
32.61% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000148     6
93.10% :     0.000137s :      4: match.inline
 6.90% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.47% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000001s :     10: predicate.addn_zero_filter
 1.86% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000001s :     10: predicate.compare_switch_simplify
 1.49% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.23% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.24% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.24% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.07% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.02% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.19% :     0.000001s :     10: predicate.minmaximum_grad
 1.51% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     15: predicate.partial_eliminate
 1.32% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.65% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.88% :     0.000002s :     10: predicate.reshape_eliminate
 1.36% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.53% :     0.000003s :     16: predicate.switch_defer_inline
 2.26% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.97% :     0.000009s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.34% :     0.000002s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.41% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000497    12
50.51% :     0.000251s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.20% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.29% :     0.000220s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018035    76
 0.26% :     0.000046s :      1: add_recomputation
 0.35% :     0.000064s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.55% :     0.000280s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000020s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
25.80% :     0.004652s :      1: jit_opt_a
 0.74% :     0.000133s :      1: jit_opt_after_cconv
 0.27% :     0.000049s :      1: jit_opt_b
 2.24% :     0.000404s :      1: loop_unroll
 2.65% :     0.000478s :      1: mutable_eliminate
 4.36% :     0.000786s :     26: opt.transform.jit_opt_a
 0.25% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.46% :     0.000443s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 1.30% :     0.000234s :      1: renormalize.infer
 1.25% :     0.000226s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000034s :      1: rewriter_after_opt_a
 0.40% :     0.000072s :      1: rewriter_before_opt_a
 0.39% :     0.000071s :      1: symbol_engine_optimizer
30.72% :     0.005539s :      1: task_emit
22.88% :     0.004127s :      1: type_inference
 0.25% :     0.000046s :      1: validate


TotalTime = 0.0330607, [33]
[bootstrap]: 0.00025881
[type_inference]: 0.00407716
[event_method]: 1.687e-05
[auto_monad]: 6.3e-05
[graph_reusing]: 5.15999e-06
[pre_auto_parallel]: 2.54999e-06
[py_interpret_to_execute]: 2.828e-05
[rewriter_before_opt_a]: 6.784e-05
[expand_dump_flag]: 2.94999e-06
[jit_opt_a]: 0.00495505, [2]
    [Cycle 1]: 0.00151542, [27]
        [switch_simplify]: 5.132e-05
        [loop_unroll]: 3.018e-05
        [a_1]: 0.0005184
        [with_stream_mark]: 1.681e-05
        [recompute_prepare]: 7.48e-06
        [updatestate_depend_eliminate]: 3.56999e-06
        [updatestate_assign_eliminate]: 2.66999e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [parameter_eliminate]: 1.72001e-06
        [specialize_transform]: 5.61e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.93997e-06
        [get_grad_eliminate_]: 5.38002e-06
        [merge_forward]: 3.25e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.226e-05
        [j_node_and_user_rematch]: 8.78001e-06
        [meta_fg_expand]: 2.19001e-06
        [replace_old_param]: 8.95999e-06
        [inline_without_move]: 5.39998e-06
        [renormalize]: 0.00059899
        [add_forward_monad_depend]: 4.99e-06
        [auto_monad_grad]: 2.59001e-06
        [auto_monad_eliminator]: 1.337e-05
        [cse]: 1.959e-05
        [replace_applicator]: 1.259e-05
    [Cycle 2]: 0.00032142, [27]
        [switch_simplify]: 5.79e-06
        [loop_unroll]: 4.72998e-06
        [a_1]: 8.875e-05
        [with_stream_mark]: 9.44e-06
        [recompute_prepare]: 5.14998e-06
        [updatestate_depend_eliminate]: 2.79001e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 1.11002e-06
        [specialize_transform]: 5.04e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 4.95001e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.55001e-06
        [merge_forward]: 2.35002e-06
        [cell_reuse_recompute_pass]: 1.71e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.222e-05
        [j_node_and_user_rematch]: 7.82e-06
        [meta_fg_expand]: 1.64998e-06
        [replace_old_param]: 7.73999e-06
        [inline_without_move]: 4.77e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 5.19e-06
        [cse]: 1.007e-05
        [replace_applicator]: 5.39998e-06
[py_interpret_to_execute_after_opt_a]: 8.72998e-06
[rewriter_after_opt_a]: 3.142e-05
[convert_after_rewriter]: 6.56999e-06
[order_py_execute_after_rewriter]: 4.67e-06
[mutable_eliminate]: 0.00050933
[jit_opt_b]: 4.904e-05, [1]
    [Cycle 1]: 4.235e-05, [2]
        [frontend_op_eliminate]: 1.577e-05
        [inline_after_opt_a]: 1.509e-05
[cconv]: 2.472e-05
[loop_unroll]: 0.00040754
[jit_opt_after_cconv]: 0.00013542, [1]
    [Cycle 1]: 0.00012952, [11]
        [c_1]: 2.074e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 5.38002e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [cse]: 1.613e-05
        [call_graph_tuple_transform]: 1.816e-05
        [tuple_list_get_item_eliminator]: 5.34e-06
        [none_parameter_eliminate]: 1.34e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 5.20999e-06
[remove_dup_value]: 1.214e-05
[partial_unused_args_eliminate]: 2.39999e-06
[environ_conv]: 4.82998e-06
[add_recomputation]: 4.082e-05
[cse_after_recomputation]: 2.034e-05, [1]
    [Cycle 1]: 1.478e-05, [1]
        [cse]: 8.75001e-06
[auto_monad_reorder]: 1.37e-05
[get_jit_bprop_graph]: 2.43e-06
[rewriter_after_jit_bprop_graph]: 4.22e-06
[opt_after_jit_grad]: 0.0004657
[symbol_engine_optimizer]: 7.169e-05, [1]
    [Cycle 1]: 6.529e-05, [6]
        [build]: 3.5e-06
        [elim_shapecalc]: 7.63001e-06
        [elim_not_effective]: 1.281e-05
        [opt_reshape]: 5.61e-06
        [fold_const_symbol]: 8.24998e-06
        [renormalize]: 4.69998e-07
[validate]: 3.181e-05
[backend_pass]: 8.59989e-07
[task_emit]: 0.0215466
[execute]: 8.30999e-06
Sums
    bootstrap                            :     0.000259s :  0.88%
    type_inference                       :     0.004077s : 13.89%
    event_method                         :     0.000017s :  0.06%
    auto_monad                           :     0.000063s :  0.21%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.10%
    rewriter_before_opt_a                :     0.000068s :  0.23%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000057s :  0.19%
    jit_opt_a.loop_unroll                :     0.000035s :  0.12%
    jit_opt_a.a_1                        :     0.000607s :  2.07%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000017s :  0.06%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.000599s :  2.04%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.06%
    jit_opt_a.cse                        :     0.000030s :  0.10%
    jit_opt_a.replace_applicator         :     0.000018s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.03%
    rewriter_after_opt_a                 :     0.000031s :  0.11%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000509s :  1.73%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.05%
    cconv                                :     0.000025s :  0.08%
    loop_unroll                          :     0.000408s :  1.39%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000012s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000041s :  0.14%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000014s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.01%
    opt_after_jit_grad                   :     0.000466s :  1.59%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.021547s : 73.38%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000193    23
 1.28% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.74% :     0.000005s :      3: substitution.graph_param_transform
83.14% :     0.000160s :      4: substitution.inline
 1.61% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.18% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.13% :     0.000004s :      2: substitution.replace_old_param
 6.28% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004019     2
81.67% :     0.003282s :      1: type_inference.infer
18.33% :     0.000737s :      1: type_inference.specialize
------[replace.]   0.000066     6
66.23% :     0.000044s :      4: replace.inline
33.77% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000169     6
93.43% :     0.000158s :      4: match.inline
 6.57% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.23% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.37% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 2.45% :     0.000003s :     10: predicate.arithmetic_simplify
 1.39% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.47% :     0.000002s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.46% :     0.000006s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :      6: predicate.less_batch_normalization
 1.86% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.00% :     0.000002s :     15: predicate.load_eliminater
 1.49% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.35% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.62% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000000s :      3: predicate.opt_reshape
 2.12% :     0.000002s :     15: predicate.partial_eliminate
 1.09% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.95% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.60% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.60% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.30% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.40% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.99% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000514    12
47.36% :     0.000243s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.63% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.00% :     0.000236s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.034581    76
 0.13% :     0.000044s :      1: add_recomputation
 0.19% :     0.000067s :      1: auto_monad
 0.05% :     0.000016s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.79% :     0.000273s :      1: bootstrap
 0.08% :     0.000027s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.06% :     0.000022s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.06% :     0.000020s :      1: event_method
 0.04% :     0.000013s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
14.34% :     0.004958s :      1: jit_opt_a
 0.40% :     0.000138s :      1: jit_opt_after_cconv
 0.15% :     0.000052s :      1: jit_opt_b
 1.20% :     0.000415s :      1: loop_unroll
 1.50% :     0.000518s :      1: mutable_eliminate
 2.35% :     0.000813s :     26: opt.transform.jit_opt_a
 0.13% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000025s :      4: opt.transform.jit_opt_b
 0.03% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.37% :     0.000474s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.09% :     0.000031s :      1: py_interpret_to_execute
 0.03% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000015s :      1: remove_dup_value
 0.92% :     0.000317s :      1: renormalize.infer
 0.79% :     0.000274s :      1: renormalize.specialize
 0.02% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000034s :      1: rewriter_after_opt_a
 0.20% :     0.000071s :      1: rewriter_before_opt_a
 0.21% :     0.000074s :      1: symbol_engine_optimizer
62.34% :     0.021559s :      1: task_emit
11.82% :     0.004089s :      1: type_inference
 0.15% :     0.000051s :      1: validate


TotalTime = 0.0130046, [30]
[bootstrap]: 0.00027381
[type_inference]: 0.00582153
[event_method]: 1.871e-05
[auto_monad]: 6.004e-05
[graph_reusing]: 4.12e-06
[pre_auto_parallel]: 1.94e-06
[py_interpret_to_execute]: 3.831e-05
[rewriter_before_opt_a]: 7.406e-05
[expand_dump_flag]: 2.07999e-06
[jit_opt_a]: 0.00475879, [2]
    [Cycle 1]: 0.00152244, [27]
        [switch_simplify]: 8.727e-05
        [loop_unroll]: 3.319e-05
        [a_1]: 0.00060862
        [with_stream_mark]: 1.386e-05
        [recompute_prepare]: 7.59002e-06
        [updatestate_depend_eliminate]: 3.13e-06
        [updatestate_assign_eliminate]: 2.01998e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.32e-06
        [specialize_transform]: 6.44001e-06
        [updatestate_useless_node_eliminater]: 5.32999e-06
        [accelerated_algorithm]: 6.06e-06
        [meta_shard_fg_expand]: 1.26002e-06
        [get_grad_eliminate_]: 5.10001e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 8.60018e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.106e-05
        [j_node_and_user_rematch]: 8.55001e-06
        [meta_fg_expand]: 1.77001e-06
        [replace_old_param]: 8.04002e-06
        [inline_without_move]: 5.47001e-06
        [renormalize]: 0.00049697
        [add_forward_monad_depend]: 4.27e-06
        [auto_monad_grad]: 1.23002e-06
        [auto_monad_eliminator]: 9.42001e-06
        [cse]: 1.619e-05
        [replace_applicator]: 1.088e-05
    [Cycle 2]: 0.00030247, [27]
        [switch_simplify]: 5.50001e-06
        [loop_unroll]: 4.80001e-06
        [a_1]: 8.71e-05
        [with_stream_mark]: 8.25e-06
        [recompute_prepare]: 4.99998e-06
        [updatestate_depend_eliminate]: 2.52001e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 4.77e-06
        [updatestate_useless_node_eliminater]: 4.62e-06
        [accelerated_algorithm]: 4.75001e-06
        [meta_shard_fg_expand]: 1.08001e-06
        [get_grad_eliminate_]: 4.4e-06
        [merge_forward]: 2.53e-06
        [cell_reuse_recompute_pass]: 1.30001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.129e-05
        [j_node_and_user_rematch]: 7.4e-06
        [meta_fg_expand]: 1.35001e-06
        [replace_old_param]: 6.71e-06
        [inline_without_move]: 4.67998e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 4.47003e-06
        [cse]: 9.14e-06
        [replace_applicator]: 4.68001e-06
[py_interpret_to_execute_after_opt_a]: 7.63001e-06
[rewriter_after_opt_a]: 2.353e-05
[convert_after_rewriter]: 5.27001e-06
[order_py_execute_after_rewriter]: 3.68999e-06
[mutable_eliminate]: 0.00047044
[jit_opt_b]: 4.528e-05, [1]
    [Cycle 1]: 3.932e-05, [2]
        [frontend_op_eliminate]: 1.454e-05
        [inline_after_opt_a]: 1.387e-05
[cconv]: 1.507e-05
[loop_unroll]: 0.00041737
[jit_opt_after_cconv]: 0.00012786, [1]
    [Cycle 1]: 0.00012203, [11]
        [c_1]: 1.98e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 5.19e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.506e-05
        [call_graph_tuple_transform]: 1.529e-05
        [tuple_list_get_item_eliminator]: 4.99e-06
        [none_parameter_eliminate]: 9.70002e-07
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.07e-06
[remove_dup_value]: 6.95002e-06
[partial_unused_args_eliminate]: 1.46002e-06
[environ_conv]: 3.11001e-06
[add_recomputation]: 3.061e-05
[cse_after_recomputation]: 1.945e-05, [1]
    [Cycle 1]: 1.434e-05, [1]
        [cse]: 8.66002e-06
[auto_monad_reorder]: 9.36e-06
[get_jit_bprop_graph]: 1.49e-06
[rewriter_after_jit_bprop_graph]: 5.14e-06
[opt_after_jit_grad]: 0.00051392
[symbol_engine_optimizer]: 6.762e-05, [1]
    [Cycle 1]: 6.188e-05, [6]
        [build]: 2.71e-06
        [elim_shapecalc]: 6.93e-06
        [elim_not_effective]: 1.158e-05
        [opt_reshape]: 5.54e-06
        [fold_const_symbol]: 7.83999e-06
        [renormalize]: 4.80009e-07
[validate]: 2.309e-05
Sums
    bootstrap                            :     0.000274s :  2.87%
    type_inference                       :     0.005822s : 61.08%
    event_method                         :     0.000019s :  0.20%
    auto_monad                           :     0.000060s :  0.63%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.40%
    rewriter_before_opt_a                :     0.000074s :  0.78%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000093s :  0.97%
    jit_opt_a.loop_unroll                :     0.000038s :  0.40%
    jit_opt_a.a_1                        :     0.000696s :  7.30%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000497s :  5.21%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.15%
    jit_opt_a.cse                        :     0.000025s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.08%
    rewriter_after_opt_a                 :     0.000024s :  0.25%
    convert_after_rewriter               :     0.000005s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000470s :  4.94%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.15%
    cconv                                :     0.000015s :  0.16%
    loop_unroll                          :     0.000417s :  4.38%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000007s :  0.07%
    partial_unused_args_eliminate        :     0.000001s :  0.02%
    environ_conv                         :     0.000003s :  0.03%
    add_recomputation                    :     0.000031s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000009s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000514s :  5.39%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000023s :  0.24%


Time group info:
------[substitution.]   0.000215    26
 0.63% :     0.000001s :      2: substitution.elim_not_effective
 0.40% :     0.000001s :      2: substitution.fold_const_symbol
 1.57% :     0.000003s :      3: substitution.graph_param_transform
86.07% :     0.000185s :      6: substitution.inline
 1.17% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.85% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.34% :     0.000003s :      2: substitution.replace_old_param
 3.06% :     0.000007s :      1: substitution.switch_simplify
 3.91% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005771     2
84.64% :     0.004885s :      1: type_inference.infer
15.36% :     0.000886s :      1: type_inference.specialize
------[replace.]   0.000108     9
53.93% :     0.000058s :      6: replace.inline
19.75% :     0.000021s :      1: replace.switch_simplify
26.32% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000194     9
93.33% :     0.000181s :      6: match.inline
 2.97% :     0.000006s :      1: match.switch_simplify
 3.69% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   801
 1.42% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.19% :     0.000001s :     12: predicate.addn_check_dump
 1.44% :     0.000002s :     12: predicate.addn_zero_filter
 1.71% :     0.000002s :     12: predicate.arithmetic_simplify
 1.29% :     0.000002s :     12: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.18% :     0.000001s :     12: predicate.compare_switch_simplify
 1.77% :     0.000002s :     12: predicate.depend_value_elim
 1.21% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.41% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.36% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.25% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     12: predicate.environ_get_eliminate
 1.23% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.42% :     0.000007s :     26: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     17: predicate.load_eliminater
 1.33% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.19% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.98% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.22% :     0.000002s :     12: predicate.merge_addn
 1.23% :     0.000002s :     12: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     17: predicate.partial_eliminate
 1.26% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     12: predicate.reduce_eliminate
 1.60% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     20: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     12: predicate.reshape_eliminate
 1.33% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.60% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.71% :     0.000003s :     20: predicate.switch_defer_inline
 2.45% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.46% :     0.000010s :     65: predicate.switch_simplify
 1.28% :     0.000002s :     12: predicate.tile_eliminate
 1.29% :     0.000002s :     12: predicate.transpose_eliminate
 1.45% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.21% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     12: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000702    15
51.40% :     0.000361s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.42% :     0.000073s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.17% :     0.000268s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014504    72
 0.23% :     0.000033s :      1: add_recomputation
 0.44% :     0.000063s :      1: auto_monad
 0.08% :     0.000012s :      1: auto_monad_reorder
 2.01% :     0.000292s :      1: bootstrap
 0.12% :     0.000018s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.15% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000005s :      1: environ_conv
 0.15% :     0.000022s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
32.83% :     0.004761s :      1: jit_opt_a
 0.90% :     0.000131s :      1: jit_opt_after_cconv
 0.33% :     0.000048s :      1: jit_opt_b
 2.93% :     0.000425s :      1: loop_unroll
 3.30% :     0.000478s :      1: mutable_eliminate
 6.45% :     0.000936s :     26: opt.transform.jit_opt_a
 0.29% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000023s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000029s :      4: opt.transform.symbol_engine_opt
 3.60% :     0.000522s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.28% :     0.000041s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000009s :      1: remove_dup_value
 1.81% :     0.000263s :      1: renormalize.infer
 1.57% :     0.000228s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000026s :      1: rewriter_after_opt_a
 0.53% :     0.000077s :      1: rewriter_before_opt_a
 0.48% :     0.000070s :      1: symbol_engine_optimizer
40.21% :     0.005832s :      1: type_inference


TotalTime = 0.0213843, [33]
[bootstrap]: 0.00039908
[type_inference]: 0.00497084
[event_method]: 1.813e-05
[auto_monad]: 6.793e-05
[graph_reusing]: 6.43e-06
[pre_auto_parallel]: 2.74001e-06
[py_interpret_to_execute]: 4.049e-05
[rewriter_before_opt_a]: 7.368e-05
[expand_dump_flag]: 3.20998e-06
[jit_opt_a]: 0.00593725, [2]
    [Cycle 1]: 0.00179308, [27]
        [switch_simplify]: 6.473e-05
        [loop_unroll]: 2.97e-05
        [a_1]: 0.00055676
        [with_stream_mark]: 1.905e-05
        [recompute_prepare]: 7.83999e-06
        [updatestate_depend_eliminate]: 3.90998e-06
        [updatestate_assign_eliminate]: 3.41999e-06
        [updatestate_loads_eliminate]: 2.41998e-06
        [parameter_eliminate]: 1.75001e-06
        [specialize_transform]: 5.72001e-06
        [updatestate_useless_node_eliminater]: 4.83001e-06
        [accelerated_algorithm]: 5.53002e-06
        [meta_shard_fg_expand]: 2.14999e-06
        [get_grad_eliminate_]: 5.06997e-06
        [merge_forward]: 3.68999e-06
        [cell_reuse_recompute_pass]: 1.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.373e-05
        [j_node_and_user_rematch]: 9.46e-06
        [meta_fg_expand]: 2.34999e-06
        [replace_old_param]: 9.67001e-06
        [inline_without_move]: 5.49e-06
        [renormalize]: 0.00078546
        [add_forward_monad_depend]: 6.02001e-06
        [auto_monad_grad]: 2.81999e-06
        [auto_monad_eliminator]: 1.455e-05
        [cse]: 3.059e-05
        [replace_applicator]: 1.3e-05
    [Cycle 2]: 0.00033566, [27]
        [switch_simplify]: 6.12999e-06
        [loop_unroll]: 5.20999e-06
        [a_1]: 9.146e-05
        [with_stream_mark]: 1.075e-05
        [recompute_prepare]: 4.74998e-06
        [updatestate_depend_eliminate]: 2.46998e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 2.31998e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 4.94e-06
        [updatestate_useless_node_eliminater]: 4.62998e-06
        [accelerated_algorithm]: 5.37999e-06
        [meta_shard_fg_expand]: 1.57001e-06
        [get_grad_eliminate_]: 4.42998e-06
        [merge_forward]: 2.59001e-06
        [cell_reuse_recompute_pass]: 2.02999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.334e-05
        [j_node_and_user_rematch]: 7.98001e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 8.28999e-06
        [inline_without_move]: 4.59002e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.52001e-06
        [auto_monad_grad]: 8.89995e-07
        [auto_monad_eliminator]: 6.21e-06
        [cse]: 1.241e-05
        [replace_applicator]: 5.28002e-06
[py_interpret_to_execute_after_opt_a]: 1.208e-05
[rewriter_after_opt_a]: 3.857e-05
[convert_after_rewriter]: 6.93e-06
[order_py_execute_after_rewriter]: 4.91002e-06
[mutable_eliminate]: 0.00076573
[jit_opt_b]: 5.362e-05, [1]
    [Cycle 1]: 4.546e-05, [2]
        [frontend_op_eliminate]: 1.835e-05
        [inline_after_opt_a]: 1.53e-05
[cconv]: 2.936e-05
[loop_unroll]: 0.00046751
[jit_opt_after_cconv]: 0.00015289, [1]
    [Cycle 1]: 0.00014544, [11]
        [c_1]: 2.098e-05
        [parameter_eliminate]: 3.35e-06
        [updatestate_depend_eliminate]: 6.19001e-06
        [updatestate_assign_eliminate]: 2.53998e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [cse]: 2.466e-05
        [call_graph_tuple_transform]: 2.006e-05
        [tuple_list_get_item_eliminator]: 5.38002e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 9.90025e-07
        [switch_simplify]: 5.76e-06
[remove_dup_value]: 1.565e-05
[partial_unused_args_eliminate]: 2.89001e-06
[environ_conv]: 5.97001e-06
[add_recomputation]: 5.252e-05
[cse_after_recomputation]: 2.291e-05, [1]
    [Cycle 1]: 1.72e-05, [1]
        [cse]: 9.53002e-06
[auto_monad_reorder]: 1.641e-05
[get_jit_bprop_graph]: 2.15002e-06
[rewriter_after_jit_bprop_graph]: 6.83998e-06
[opt_after_jit_grad]: 0.00049997
[symbol_engine_optimizer]: 7.919e-05, [1]
    [Cycle 1]: 7.1e-05, [6]
        [build]: 4.87998e-06
        [elim_shapecalc]: 7.63001e-06
        [elim_not_effective]: 1.474e-05
        [opt_reshape]: 5.51e-06
        [fold_const_symbol]: 9.30001e-06
        [renormalize]: 8.10018e-07
[validate]: 4.12e-05
[backend_pass]: 1.04998e-06
[task_emit]: 0.00735601
[execute]: 8.77e-06
Sums
    bootstrap                            :     0.000399s :  2.36%
    type_inference                       :     0.004971s : 29.38%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000068s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.24%
    rewriter_before_opt_a                :     0.000074s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000071s :  0.42%
    jit_opt_a.loop_unroll                :     0.000035s :  0.21%
    jit_opt_a.a_1                        :     0.000648s :  3.83%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000786s :  4.64%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.12%
    jit_opt_a.cse                        :     0.000043s :  0.25%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000039s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000766s :  4.53%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000029s :  0.17%
    loop_unroll                          :     0.000468s :  2.76%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000053s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000500s :  2.95%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007356s : 43.47%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000219    23
 1.37% :     0.000003s :      2: substitution.elim_not_effective
 0.70% :     0.000002s :      2: substitution.fold_const_symbol
 2.58% :     0.000006s :      3: substitution.graph_param_transform
83.25% :     0.000182s :      4: substitution.inline
 1.67% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.84% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.41% :     0.000005s :      2: substitution.replace_old_param
 6.18% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004892     2
81.07% :     0.003966s :      1: type_inference.infer
18.93% :     0.000926s :      1: type_inference.specialize
------[replace.]   0.000074     6
67.17% :     0.000049s :      4: replace.inline
32.83% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     6
93.53% :     0.000180s :      4: match.inline
 6.47% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.82% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.21% :     0.000001s :     10: predicate.addn_zero_filter
 2.37% :     0.000003s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.23% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.36% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.28% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.85% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.73% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.37% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.74% :     0.000009s :     53: predicate.switch_simplify
 1.32% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.98% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.96% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.91% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000639    12
48.94% :     0.000313s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.59% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.47% :     0.000297s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023144    76
 0.24% :     0.000056s :      1: add_recomputation
 0.31% :     0.000071s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.81% :     0.000419s :      1: bootstrap
 0.14% :     0.000032s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.67% :     0.005942s :      1: jit_opt_a
 0.67% :     0.000156s :      1: jit_opt_after_cconv
 0.24% :     0.000057s :      1: jit_opt_b
 2.06% :     0.000477s :      1: loop_unroll
 3.36% :     0.000778s :      1: mutable_eliminate
 3.75% :     0.000869s :     26: opt.transform.jit_opt_a
 0.21% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.20% :     0.000508s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000043s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 1.77% :     0.000410s :      1: renormalize.infer
 1.59% :     0.000367s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000042s :      1: rewriter_after_opt_a
 0.33% :     0.000077s :      1: rewriter_before_opt_a
 0.36% :     0.000082s :      1: symbol_engine_optimizer
31.85% :     0.007371s :      1: task_emit
21.54% :     0.004986s :      1: type_inference
 0.27% :     0.000062s :      1: validate


TotalTime = 0.0167407, [33]
[bootstrap]: 0.00026636
[type_inference]: 0.0039134
[event_method]: 1.479e-05
[auto_monad]: 5.476e-05
[graph_reusing]: 4.37998e-06
[pre_auto_parallel]: 2.29999e-06
[py_interpret_to_execute]: 2.634e-05
[rewriter_before_opt_a]: 6.673e-05
[expand_dump_flag]: 2.37999e-06
[jit_opt_a]: 0.00480887, [2]
    [Cycle 1]: 0.00146156, [27]
        [switch_simplify]: 4.685e-05
        [loop_unroll]: 2.856e-05
        [a_1]: 0.00049582
        [with_stream_mark]: 1.488e-05
        [recompute_prepare]: 7.53e-06
        [updatestate_depend_eliminate]: 2.98998e-06
        [updatestate_assign_eliminate]: 2.91999e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [parameter_eliminate]: 1.67001e-06
        [specialize_transform]: 5.86e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 5.82999e-06
        [meta_shard_fg_expand]: 1.92999e-06
        [get_grad_eliminate_]: 5.35999e-06
        [merge_forward]: 3.46999e-06
        [cell_reuse_recompute_pass]: 9.5999e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.346e-05
        [j_node_and_user_rematch]: 8.71002e-06
        [meta_fg_expand]: 2.41e-06
        [replace_old_param]: 8.58001e-06
        [inline_without_move]: 5.19e-06
        [renormalize]: 0.0005859
        [add_forward_monad_depend]: 5.30999e-06
        [auto_monad_grad]: 2.64999e-06
        [auto_monad_eliminator]: 1.255e-05
        [cse]: 1.711e-05
        [replace_applicator]: 1.282e-05
    [Cycle 2]: 0.00031202, [27]
        [switch_simplify]: 5.22999e-06
        [loop_unroll]: 4.74002e-06
        [a_1]: 8.814e-05
        [with_stream_mark]: 8.72e-06
        [recompute_prepare]: 4.98001e-06
        [updatestate_depend_eliminate]: 2.48002e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 4.85999e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 5.19e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 4.33999e-06
        [merge_forward]: 2.56e-06
        [cell_reuse_recompute_pass]: 1.78002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.157e-05
        [j_node_and_user_rematch]: 7.85e-06
        [meta_fg_expand]: 1.61998e-06
        [replace_old_param]: 7.17002e-06
        [inline_without_move]: 4.58001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.35001e-06
        [auto_monad_grad]: 1.15999e-06
        [auto_monad_eliminator]: 5.22e-06
        [cse]: 9.44e-06
        [replace_applicator]: 9.51998e-06
[py_interpret_to_execute_after_opt_a]: 9.75002e-06
[rewriter_after_opt_a]: 3.252e-05
[convert_after_rewriter]: 6.13998e-06
[order_py_execute_after_rewriter]: 5.00001e-06
[mutable_eliminate]: 0.00053743
[jit_opt_b]: 4.86e-05, [1]
    [Cycle 1]: 4.157e-05, [2]
        [frontend_op_eliminate]: 1.599e-05
        [inline_after_opt_a]: 1.43e-05
[cconv]: 2.38e-05
[loop_unroll]: 0.00040901
[jit_opt_after_cconv]: 0.00013303, [1]
    [Cycle 1]: 0.00012664, [11]
        [c_1]: 1.974e-05
        [parameter_eliminate]: 2.34001e-06
        [updatestate_depend_eliminate]: 5.36002e-06
        [updatestate_assign_eliminate]: 2.36998e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.693e-05
        [call_graph_tuple_transform]: 1.731e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.19003e-06
        [renormalize]: 9.70002e-07
        [switch_simplify]: 5.39e-06
[remove_dup_value]: 1.214e-05
[partial_unused_args_eliminate]: 1.84e-06
[environ_conv]: 4.94e-06
[add_recomputation]: 4.368e-05
[cse_after_recomputation]: 1.9e-05, [1]
    [Cycle 1]: 1.393e-05, [1]
        [cse]: 8.42998e-06
[auto_monad_reorder]: 1.432e-05
[get_jit_bprop_graph]: 1.83002e-06
[rewriter_after_jit_bprop_graph]: 4.57e-06
[opt_after_jit_grad]: 0.00049973
[symbol_engine_optimizer]: 7.362e-05, [1]
    [Cycle 1]: 6.696e-05, [6]
        [build]: 3.36001e-06
        [elim_shapecalc]: 7.72002e-06
        [elim_not_effective]: 1.333e-05
        [opt_reshape]: 5.39e-06
        [fold_const_symbol]: 8.38001e-06
        [renormalize]: 8.00006e-07
[validate]: 3.444e-05
[backend_pass]: 1.21002e-06
[task_emit]: 0.00548984
[execute]: 5.69e-06
Sums
    bootstrap                            :     0.000266s :  2.02%
    type_inference                       :     0.003913s : 29.75%
    event_method                         :     0.000015s :  0.11%
    auto_monad                           :     0.000055s :  0.42%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.20%
    rewriter_before_opt_a                :     0.000067s :  0.51%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000052s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000584s :  4.44%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000586s :  4.45%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.14%
    jit_opt_a.cse                        :     0.000027s :  0.20%
    jit_opt_a.replace_applicator         :     0.000022s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.25%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000537s :  4.09%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000409s :  3.11%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000044s :  0.33%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000500s :  3.80%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000034s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005490s : 41.73%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000182    23
 1.18% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000001s :      2: substitution.fold_const_symbol
 2.50% :     0.000005s :      3: substitution.graph_param_transform
82.91% :     0.000151s :      4: substitution.inline
 1.63% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.26% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.56% :     0.000005s :      2: substitution.replace_old_param
 6.21% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003863     2
81.83% :     0.003161s :      1: type_inference.infer
18.17% :     0.000702s :      1: type_inference.specialize
------[replace.]   0.000062     6
66.41% :     0.000041s :      4: replace.inline
33.59% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000159     6
93.69% :     0.000149s :      4: match.inline
 6.31% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.39% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 2.11% :     0.000002s :     10: predicate.arithmetic_simplify
 1.32% :     0.000001s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.41% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.04% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.42% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.13% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.82% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.34% :     0.000001s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 2.04% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.34% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.01% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.54% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000001s :     10: predicate.same_eliminate
 0.93% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.53% :     0.000008s :     53: predicate.switch_simplify
 1.28% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.35% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000485    12
46.61% :     0.000226s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.61% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.78% :     0.000227s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018223    76
 0.25% :     0.000046s :      1: add_recomputation
 0.32% :     0.000058s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.51% :     0.000274s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000018s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
26.40% :     0.004812s :      1: jit_opt_a
 0.74% :     0.000136s :      1: jit_opt_after_cconv
 0.28% :     0.000052s :      1: jit_opt_b
 2.29% :     0.000417s :      1: loop_unroll
 3.00% :     0.000546s :      1: mutable_eliminate
 4.32% :     0.000788s :     26: opt.transform.jit_opt_a
 0.24% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.79% :     0.000508s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000029s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 1.72% :     0.000314s :      1: renormalize.infer
 1.45% :     0.000264s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000035s :      1: rewriter_after_opt_a
 0.38% :     0.000069s :      1: rewriter_before_opt_a
 0.42% :     0.000076s :      1: symbol_engine_optimizer
30.19% :     0.005501s :      1: task_emit
21.53% :     0.003923s :      1: type_inference
 0.29% :     0.000053s :      1: validate


TotalTime = 0.012934, [30]
[bootstrap]: 0.00024695
[type_inference]: 0.0058521
[event_method]: 1.857e-05
[auto_monad]: 5.748e-05
[graph_reusing]: 3.83999e-06
[pre_auto_parallel]: 1.82999e-06
[py_interpret_to_execute]: 3.95e-05
[rewriter_before_opt_a]: 7.438e-05
[expand_dump_flag]: 2.10002e-06
[jit_opt_a]: 0.00471238, [2]
    [Cycle 1]: 0.00155287, [27]
        [switch_simplify]: 9.674e-05
        [loop_unroll]: 3.322e-05
        [a_1]: 0.00065779
        [with_stream_mark]: 1.154e-05
        [recompute_prepare]: 6.62002e-06
        [updatestate_depend_eliminate]: 2.89001e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.82001e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.30001e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 5.36998e-06
        [merge_forward]: 2.66e-06
        [cell_reuse_recompute_pass]: 9.30013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.091e-05
        [j_node_and_user_rematch]: 8.53001e-06
        [meta_fg_expand]: 1.77001e-06
        [replace_old_param]: 7.43999e-06
        [inline_without_move]: 5.24e-06
        [renormalize]: 0.000473
        [add_forward_monad_depend]: 4.55001e-06
        [auto_monad_grad]: 1.42e-06
        [auto_monad_eliminator]: 9.52999e-06
        [cse]: 1.559e-05
        [replace_applicator]: 1.114e-05
    [Cycle 2]: 0.00030135, [27]
        [switch_simplify]: 5.50001e-06
        [loop_unroll]: 4.75001e-06
        [a_1]: 8.687e-05
        [with_stream_mark]: 7.91001e-06
        [recompute_prepare]: 4.69002e-06
        [updatestate_depend_eliminate]: 2.39999e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 4.52998e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.49998e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.147e-05
        [j_node_and_user_rematch]: 7.45998e-06
        [meta_fg_expand]: 1.41998e-06
        [replace_old_param]: 6.83e-06
        [inline_without_move]: 4.63999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.14003e-06
        [auto_monad_grad]: 8.10018e-07
        [auto_monad_eliminator]: 4.74e-06
        [cse]: 9.04e-06
        [replace_applicator]: 4.92e-06
[py_interpret_to_execute_after_opt_a]: 7.33e-06
[rewriter_after_opt_a]: 2.884e-05
[convert_after_rewriter]: 6.17001e-06
[order_py_execute_after_rewriter]: 4.93001e-06
[mutable_eliminate]: 0.00046259
[jit_opt_b]: 5.32e-05, [1]
    [Cycle 1]: 4.048e-05, [2]
        [frontend_op_eliminate]: 1.537e-05
        [inline_after_opt_a]: 1.428e-05
[cconv]: 2.268e-05
[loop_unroll]: 0.00041156
[jit_opt_after_cconv]: 0.00013165, [1]
    [Cycle 1]: 0.00012522, [11]
        [c_1]: 2.054e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 5.10999e-06
        [updatestate_assign_eliminate]: 2.30002e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [cse]: 1.487e-05
        [call_graph_tuple_transform]: 1.721e-05
        [tuple_list_get_item_eliminator]: 5.12e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 4.7998e-07
        [switch_simplify]: 5.29e-06
[remove_dup_value]: 1.277e-05
[partial_unused_args_eliminate]: 2.38998e-06
[environ_conv]: 4.85001e-06
[add_recomputation]: 4.467e-05
[cse_after_recomputation]: 1.983e-05, [1]
    [Cycle 1]: 1.455e-05, [1]
        [cse]: 9.00001e-06
[auto_monad_reorder]: 1.381e-05
[get_jit_bprop_graph]: 1.44e-06
[rewriter_after_jit_bprop_graph]: 5.34e-06
[opt_after_jit_grad]: 0.0004459
[symbol_engine_optimizer]: 6.997e-05, [1]
    [Cycle 1]: 6.422e-05, [6]
        [build]: 3.26999e-06
        [elim_shapecalc]: 7.92998e-06
        [elim_not_effective]: 1.174e-05
        [opt_reshape]: 5.39998e-06
        [fold_const_symbol]: 7.95998e-06
        [renormalize]: 5.60016e-07
[validate]: 2.905e-05
Sums
    bootstrap                            :     0.000247s :  2.59%
    type_inference                       :     0.005852s : 61.37%
    event_method                         :     0.000019s :  0.19%
    auto_monad                           :     0.000057s :  0.60%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.41%
    rewriter_before_opt_a                :     0.000074s :  0.78%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000102s :  1.07%
    jit_opt_a.loop_unroll                :     0.000038s :  0.40%
    jit_opt_a.a_1                        :     0.000745s :  7.81%
    jit_opt_a.with_stream_mark           :     0.000019s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000473s :  4.96%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.15%
    jit_opt_a.cse                        :     0.000025s :  0.26%
    jit_opt_a.replace_applicator         :     0.000016s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.08%
    rewriter_after_opt_a                 :     0.000029s :  0.30%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000463s :  4.85%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.15%
    cconv                                :     0.000023s :  0.24%
    loop_unroll                          :     0.000412s :  4.32%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000013s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000045s :  0.47%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000014s :  0.14%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000446s :  4.68%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000029s :  0.30%


Time group info:
------[substitution.]   0.000214    26
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.14% :     0.000005s :      3: substitution.graph_param_transform
84.76% :     0.000181s :      6: substitution.inline
 1.16% :     0.000002s :      4: substitution.j_node_and_user_rematch
 1.88% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.34% :     0.000003s :      2: substitution.replace_old_param
 3.31% :     0.000007s :      1: substitution.switch_simplify
 3.97% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005804     2
83.74% :     0.004861s :      1: type_inference.infer
16.26% :     0.000944s :      1: type_inference.specialize
------[replace.]   0.000164     9
64.98% :     0.000107s :      6: replace.inline
18.13% :     0.000030s :      1: replace.switch_simplify
16.89% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     9
92.86% :     0.000178s :      6: match.inline
 3.32% :     0.000006s :      1: match.switch_simplify
 3.83% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   801
 1.51% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.04% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.21% :     0.000002s :     12: predicate.addn_check_dump
 1.40% :     0.000002s :     12: predicate.addn_zero_filter
 1.80% :     0.000002s :     12: predicate.arithmetic_simplify
 1.34% :     0.000002s :     12: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.22% :     0.000002s :     12: predicate.compare_switch_simplify
 1.49% :     0.000002s :     12: predicate.depend_value_elim
 1.17% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     12: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.27% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     12: predicate.environ_get_eliminate
 1.18% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.36% :     0.000007s :     26: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     17: predicate.load_eliminater
 1.17% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.40% :     0.000005s :     40: predicate.loop_unroll_before_grad
 2.29% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.22% :     0.000002s :     12: predicate.merge_addn
 1.18% :     0.000001s :     12: predicate.minmaximum_grad
 1.47% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     17: predicate.partial_eliminate
 1.34% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     12: predicate.reduce_eliminate
 1.52% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     20: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     12: predicate.reshape_eliminate
 1.29% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     12: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.76% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.85% :     0.000004s :     20: predicate.switch_defer_inline
 2.42% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.42% :     0.000010s :     65: predicate.switch_simplify
 1.22% :     0.000002s :     12: predicate.tile_eliminate
 1.26% :     0.000002s :     12: predicate.transpose_eliminate
 1.47% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.13% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.81% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     12: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000684    15
50.10% :     0.000343s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.47% :     0.000072s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.43% :     0.000270s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014450    72
 0.33% :     0.000048s :      1: add_recomputation
 0.42% :     0.000061s :      1: auto_monad
 0.11% :     0.000016s :      1: auto_monad_reorder
 1.79% :     0.000259s :      1: bootstrap
 0.17% :     0.000025s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000022s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
32.63% :     0.004715s :      1: jit_opt_a
 0.93% :     0.000134s :      1: jit_opt_after_cconv
 0.39% :     0.000056s :      1: jit_opt_b
 2.90% :     0.000418s :      1: loop_unroll
 3.26% :     0.000471s :      1: mutable_eliminate
 6.86% :     0.000991s :     26: opt.transform.jit_opt_a
 0.31% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.14% :     0.000453s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.29% :     0.000042s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000015s :      1: remove_dup_value
 1.63% :     0.000236s :      1: renormalize.infer
 1.51% :     0.000219s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000032s :      1: rewriter_after_opt_a
 0.53% :     0.000077s :      1: rewriter_before_opt_a
 0.50% :     0.000072s :      1: symbol_engine_optimizer
40.57% :     0.005863s :      1: type_inference


TotalTime = 0.0150148, [33]
[bootstrap]: 0.0002349
[type_inference]: 0.00375906
[event_method]: 1.471e-05
[auto_monad]: 4.153e-05
[graph_reusing]: 4.2e-06
[pre_auto_parallel]: 1.69e-06
[py_interpret_to_execute]: 2.551e-05
[rewriter_before_opt_a]: 6.04e-05
[expand_dump_flag]: 1.92001e-06
[jit_opt_a]: 0.00433222, [2]
    [Cycle 1]: 0.00123996, [27]
        [switch_simplify]: 4.164e-05
        [loop_unroll]: 2.873e-05
        [a_1]: 0.00046478
        [with_stream_mark]: 1.148e-05
        [recompute_prepare]: 7.28999e-06
        [updatestate_depend_eliminate]: 2.93e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 5.62999e-06
        [updatestate_useless_node_eliminater]: 5.12999e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 5.26002e-06
        [merge_forward]: 2.36998e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.072e-05
        [j_node_and_user_rematch]: 7.97e-06
        [meta_fg_expand]: 1.90001e-06
        [replace_old_param]: 7.36001e-06
        [inline_without_move]: 4.91002e-06
        [renormalize]: 0.00042221
        [add_forward_monad_depend]: 3.95998e-06
        [auto_monad_grad]: 1.42999e-06
        [auto_monad_eliminator]: 9.88002e-06
        [cse]: 1.413e-05
        [replace_applicator]: 1.161e-05
    [Cycle 2]: 0.00030483, [27]
        [switch_simplify]: 5.41002e-06
        [loop_unroll]: 4.94e-06
        [a_1]: 8.644e-05
        [with_stream_mark]: 7.75e-06
        [recompute_prepare]: 4.90999e-06
        [updatestate_depend_eliminate]: 2.23998e-06
        [updatestate_assign_eliminate]: 2.18998e-06
        [updatestate_loads_eliminate]: 1.78997e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 4.77e-06
        [updatestate_useless_node_eliminater]: 4.68999e-06
        [accelerated_algorithm]: 4.85001e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.30002e-06
        [cell_reuse_recompute_pass]: 1.55001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.138e-05
        [j_node_and_user_rematch]: 7.42002e-06
        [meta_fg_expand]: 1.39998e-06
        [replace_old_param]: 6.72002e-06
        [inline_without_move]: 4.46002e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 5.19e-06
        [cse]: 9.32001e-06
        [replace_applicator]: 5.00001e-06
[py_interpret_to_execute_after_opt_a]: 7.43999e-06
[rewriter_after_opt_a]: 2.257e-05
[convert_after_rewriter]: 7.43e-06
[order_py_execute_after_rewriter]: 3.76999e-06
[mutable_eliminate]: 0.00046171
[jit_opt_b]: 4.59e-05, [1]
    [Cycle 1]: 3.975e-05, [2]
        [frontend_op_eliminate]: 1.508e-05
        [inline_after_opt_a]: 1.37e-05
[cconv]: 1.529e-05
[loop_unroll]: 0.00040757
[jit_opt_after_cconv]: 0.00012807, [1]
    [Cycle 1]: 0.00012186, [11]
        [c_1]: 2.002e-05
        [parameter_eliminate]: 2.46998e-06
        [updatestate_depend_eliminate]: 5.07e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 1.96998e-06
        [cse]: 1.501e-05
        [call_graph_tuple_transform]: 1.542e-05
        [tuple_list_get_item_eliminator]: 5.12e-06
        [none_parameter_eliminate]: 8.30012e-07
        [renormalize]: 4.60015e-07
        [switch_simplify]: 5.00001e-06
[remove_dup_value]: 5.76e-06
[partial_unused_args_eliminate]: 1.42e-06
[environ_conv]: 3.06001e-06
[add_recomputation]: 2.855e-05
[cse_after_recomputation]: 1.865e-05, [1]
    [Cycle 1]: 1.361e-05, [1]
        [cse]: 8.13001e-06
[auto_monad_reorder]: 8.59e-06
[get_jit_bprop_graph]: 1.48002e-06
[rewriter_after_jit_bprop_graph]: 4.27998e-06
[opt_after_jit_grad]: 0.00044311
[symbol_engine_optimizer]: 6.605e-05, [1]
    [Cycle 1]: 6.018e-05, [6]
        [build]: 2.69999e-06
        [elim_shapecalc]: 7e-06
        [elim_not_effective]: 1.11e-05
        [opt_reshape]: 5.19998e-06
        [fold_const_symbol]: 7.63999e-06
        [renormalize]: 3.7998e-07
[validate]: 2.435e-05
[backend_pass]: 7.2e-07
[task_emit]: 0.00466788
[execute]: 4.32998e-06
Sums
    bootstrap                            :     0.000235s :  2.01%
    type_inference                       :     0.003759s : 32.17%
    event_method                         :     0.000015s :  0.13%
    auto_monad                           :     0.000042s :  0.36%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.22%
    rewriter_before_opt_a                :     0.000060s :  0.52%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000047s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.29%
    jit_opt_a.a_1                        :     0.000551s :  4.72%
    jit_opt_a.with_stream_mark           :     0.000019s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000015s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.12%
    jit_opt_a.inline_without_move        :     0.000009s :  0.08%
    jit_opt_a.renormalize                :     0.000422s :  3.61%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.13%
    jit_opt_a.cse                        :     0.000023s :  0.20%
    jit_opt_a.replace_applicator         :     0.000017s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.06%
    rewriter_after_opt_a                 :     0.000023s :  0.19%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000462s :  3.95%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.12%
    cconv                                :     0.000015s :  0.13%
    loop_unroll                          :     0.000408s :  3.49%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.17%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000006s :  0.05%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000003s :  0.03%
    add_recomputation                    :     0.000029s :  0.24%
    cse_after_recomputation.cse          :     0.000008s :  0.07%
    auto_monad_reorder                   :     0.000009s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.04%
    opt_after_jit_grad                   :     0.000443s :  3.79%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000024s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.004668s : 39.95%
    execute                              :     0.000004s :  0.04%


Time group info:
------[substitution.]   0.000140    23
 0.91% :     0.000001s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.55% :     0.000004s :      3: substitution.graph_param_transform
84.08% :     0.000118s :      4: substitution.inline
 1.76% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.69% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.13% :     0.000003s :      2: substitution.replace_old_param
 5.23% :     0.000007s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003717     2
80.96% :     0.003009s :      1: type_inference.infer
19.04% :     0.000708s :      1: type_inference.specialize
------[replace.]   0.000056     6
65.97% :     0.000037s :      4: replace.inline
34.03% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000122     6
94.89% :     0.000115s :      4: match.inline
 5.11% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000001s :     10: predicate.addn_check_dump
 1.49% :     0.000002s :     10: predicate.addn_zero_filter
 1.84% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.75% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.33% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.35% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.14% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.14% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.94% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000002s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.79% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.66% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.57% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.25% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.34% :     0.000008s :     53: predicate.switch_simplify
 1.46% :     0.000002s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.41% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000446    12
46.74% :     0.000208s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.44% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.82% :     0.000213s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016281    76
 0.19% :     0.000031s :      1: add_recomputation
 0.28% :     0.000045s :      1: auto_monad
 0.07% :     0.000011s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.51% :     0.000246s :      1: bootstrap
 0.11% :     0.000018s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.13% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000005s :      1: environ_conv
 0.11% :     0.000018s :      1: event_method
 0.05% :     0.000008s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
26.62% :     0.004334s :      1: jit_opt_a
 0.80% :     0.000131s :      1: jit_opt_after_cconv
 0.30% :     0.000048s :      1: jit_opt_b
 2.55% :     0.000416s :      1: loop_unroll
 2.88% :     0.000470s :      1: mutable_eliminate
 4.54% :     0.000738s :     26: opt.transform.jit_opt_a
 0.26% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000023s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000028s :      4: opt.transform.symbol_engine_opt
 2.77% :     0.000451s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000028s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000008s :      1: remove_dup_value
 1.27% :     0.000207s :      1: renormalize.infer
 1.29% :     0.000211s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000025s :      1: rewriter_after_opt_a
 0.39% :     0.000063s :      1: rewriter_before_opt_a
 0.42% :     0.000069s :      1: symbol_engine_optimizer
28.72% :     0.004676s :      1: task_emit
23.15% :     0.003769s :      1: type_inference
 0.24% :     0.000040s :      1: validate


TotalTime = 0.0172937, [33]
[bootstrap]: 0.00026173
[type_inference]: 0.00432364
[event_method]: 1.81e-05
[auto_monad]: 6.298e-05
[graph_reusing]: 5.62001e-06
[pre_auto_parallel]: 2.84999e-06
[py_interpret_to_execute]: 3.274e-05
[rewriter_before_opt_a]: 7.131e-05
[expand_dump_flag]: 3.03e-06
[jit_opt_a]: 0.00475523, [2]
    [Cycle 1]: 0.00136223, [27]
        [switch_simplify]: 5.025e-05
        [loop_unroll]: 2.972e-05
        [a_1]: 0.0004906
        [with_stream_mark]: 1.541e-05
        [recompute_prepare]: 7.05e-06
        [updatestate_depend_eliminate]: 3.65003e-06
        [updatestate_assign_eliminate]: 2.84999e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 5.99e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 5.42999e-06
        [meta_shard_fg_expand]: 2.04e-06
        [get_grad_eliminate_]: 5.17999e-06
        [merge_forward]: 4.17998e-06
        [cell_reuse_recompute_pass]: 1.35001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.38e-05
        [j_node_and_user_rematch]: 9.27001e-06
        [meta_fg_expand]: 2.07999e-06
        [replace_old_param]: 8.22e-06
        [inline_without_move]: 5.27001e-06
        [renormalize]: 0.00047207
        [add_forward_monad_depend]: 4.83001e-06
        [auto_monad_grad]: 2.12001e-06
        [auto_monad_eliminator]: 1.303e-05
        [cse]: 2.538e-05
        [replace_applicator]: 1.221e-05
    [Cycle 2]: 0.00030695, [27]
        [switch_simplify]: 5.44998e-06
        [loop_unroll]: 4.75001e-06
        [a_1]: 9.025e-05
        [with_stream_mark]: 8.80001e-06
        [recompute_prepare]: 4.89e-06
        [updatestate_depend_eliminate]: 2.53998e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 9.09989e-07
        [specialize_transform]: 4.73001e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 4.58999e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.25e-06
        [merge_forward]: 2.24001e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.191e-05
        [j_node_and_user_rematch]: 7.88999e-06
        [meta_fg_expand]: 1.55999e-06
        [replace_old_param]: 6.89999e-06
        [inline_without_move]: 4.55001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 9.89996e-07
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 4.62998e-06
        [cse]: 9.54999e-06
        [replace_applicator]: 4.74e-06
[py_interpret_to_execute_after_opt_a]: 7.44002e-06
[rewriter_after_opt_a]: 3.249e-05
[convert_after_rewriter]: 6.61e-06
[order_py_execute_after_rewriter]: 4.90001e-06
[mutable_eliminate]: 0.00050383
[jit_opt_b]: 5.016e-05, [1]
    [Cycle 1]: 4.348e-05, [2]
        [frontend_op_eliminate]: 1.68e-05
        [inline_after_opt_a]: 1.516e-05
[cconv]: 2.294e-05
[loop_unroll]: 0.0004134
[jit_opt_after_cconv]: 0.00013716, [1]
    [Cycle 1]: 0.00013057, [11]
        [c_1]: 2.118e-05
        [parameter_eliminate]: 2.49001e-06
        [updatestate_depend_eliminate]: 5.51e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.695e-05
        [call_graph_tuple_transform]: 1.777e-05
        [tuple_list_get_item_eliminator]: 5.29e-06
        [none_parameter_eliminate]: 1.57999e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 5.39e-06
[remove_dup_value]: 1.223e-05
[partial_unused_args_eliminate]: 2.11e-06
[environ_conv]: 5.62001e-06
[add_recomputation]: 4.633e-05
[cse_after_recomputation]: 2.003e-05, [1]
    [Cycle 1]: 1.476e-05, [1]
        [cse]: 8.82e-06
[auto_monad_reorder]: 1.416e-05
[get_jit_bprop_graph]: 1.83997e-06
[rewriter_after_jit_bprop_graph]: 4.70999e-06
[opt_after_jit_grad]: 0.00044249
[symbol_engine_optimizer]: 7.141e-05, [1]
    [Cycle 1]: 6.518e-05, [6]
        [build]: 3.13e-06
        [elim_shapecalc]: 7.36999e-06
        [elim_not_effective]: 1.266e-05
        [opt_reshape]: 5.42999e-06
        [fold_const_symbol]: 9.02999e-06
        [renormalize]: 4.00003e-07
[validate]: 3.303e-05
[backend_pass]: 1.02998e-06
[task_emit]: 0.0057314
[execute]: 7.62002e-06
Sums
    bootstrap                            :     0.000262s :  1.92%
    type_inference                       :     0.004324s : 31.71%
    event_method                         :     0.000018s :  0.13%
    auto_monad                           :     0.000063s :  0.46%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.24%
    rewriter_before_opt_a                :     0.000071s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000581s :  4.26%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000472s :  3.46%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000035s :  0.26%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.05%
    rewriter_after_opt_a                 :     0.000032s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000504s :  3.69%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000413s :  3.03%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000046s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000442s :  3.24%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005731s : 42.03%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000170    23
 1.27% :     0.000002s :      2: substitution.elim_not_effective
 0.83% :     0.000001s :      2: substitution.fold_const_symbol
 3.01% :     0.000005s :      3: substitution.graph_param_transform
81.13% :     0.000138s :      4: substitution.inline
 1.96% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.45% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.15% :     0.000004s :      2: substitution.replace_old_param
 7.21% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004262     2
81.75% :     0.003484s :      1: type_inference.infer
18.25% :     0.000778s :      1: type_inference.specialize
------[replace.]   0.000062     6
66.38% :     0.000041s :      4: replace.inline
33.62% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000146     6
92.45% :     0.000135s :      4: match.inline
 7.55% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.26% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.37% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000001s :     10: predicate.addn_zero_filter
 2.55% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.65% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.30% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.33% :     0.000000s :      3: predicate.graph_param_transform
 6.08% :     0.000007s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.44% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :      6: predicate.less_batch_normalization
 1.77% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.55% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.85% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.67% :     0.000009s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.84% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.84% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.90% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000528    12
50.21% :     0.000265s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.14% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.66% :     0.000236s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018657    76
 0.26% :     0.000049s :      1: add_recomputation
 0.35% :     0.000066s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.49% :     0.000278s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.12% :     0.000022s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.50% :     0.004758s :      1: jit_opt_a
 0.75% :     0.000140s :      1: jit_opt_after_cconv
 0.28% :     0.000053s :      1: jit_opt_b
 2.26% :     0.000421s :      1: loop_unroll
 2.75% :     0.000513s :      1: mutable_eliminate
 4.19% :     0.000782s :     26: opt.transform.jit_opt_a
 0.25% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.42% :     0.000451s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000035s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 1.29% :     0.000240s :      1: renormalize.infer
 1.20% :     0.000225s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000035s :      1: rewriter_after_opt_a
 0.40% :     0.000074s :      1: rewriter_before_opt_a
 0.40% :     0.000074s :      1: symbol_engine_optimizer
30.77% :     0.005741s :      1: task_emit
23.24% :     0.004337s :      1: type_inference
 0.28% :     0.000052s :      1: validate


TotalTime = 0.0125835, [30]
[bootstrap]: 0.00023235
[type_inference]: 0.00561137
[event_method]: 1.904e-05
[auto_monad]: 5.935e-05
[graph_reusing]: 4.26001e-06
[pre_auto_parallel]: 2.08002e-06
[py_interpret_to_execute]: 3.93e-05
[rewriter_before_opt_a]: 7.373e-05
[expand_dump_flag]: 1.97001e-06
[jit_opt_a]: 0.00462638, [2]
    [Cycle 1]: 0.00148223, [27]
        [switch_simplify]: 8.577e-05
        [loop_unroll]: 3.234e-05
        [a_1]: 0.00060487
        [with_stream_mark]: 1.18e-05
        [recompute_prepare]: 6.83e-06
        [updatestate_depend_eliminate]: 2.99001e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.86003e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 5.69e-06
        [updatestate_useless_node_eliminater]: 5.35999e-06
        [accelerated_algorithm]: 5.84e-06
        [meta_shard_fg_expand]: 1.30001e-06
        [get_grad_eliminate_]: 5.12999e-06
        [merge_forward]: 2.68998e-06
        [cell_reuse_recompute_pass]: 1.00999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.106e-05
        [j_node_and_user_rematch]: 9.09e-06
        [meta_fg_expand]: 1.72999e-06
        [replace_old_param]: 7.23999e-06
        [inline_without_move]: 5.22999e-06
        [renormalize]: 0.00047209
        [add_forward_monad_depend]: 4.27998e-06
        [auto_monad_grad]: 1.32e-06
        [auto_monad_eliminator]: 9.67001e-06
        [cse]: 1.451e-05
        [replace_applicator]: 1.098e-05
    [Cycle 2]: 0.0003131, [27]
        [switch_simplify]: 5.30999e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 9.377e-05
        [with_stream_mark]: 8.63001e-06
        [recompute_prepare]: 5.10001e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 2.13002e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.82998e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 4.63999e-06
        [meta_shard_fg_expand]: 1.02e-06
        [get_grad_eliminate_]: 4.53999e-06
        [merge_forward]: 2.34999e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.169e-05
        [j_node_and_user_rematch]: 7.43e-06
        [meta_fg_expand]: 1.41998e-06
        [replace_old_param]: 7.1e-06
        [inline_without_move]: 4.53999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.11002e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 4.83001e-06
        [cse]: 1.037e-05
        [replace_applicator]: 4.68999e-06
[py_interpret_to_execute_after_opt_a]: 9.44e-06
[rewriter_after_opt_a]: 2.406e-05
[convert_after_rewriter]: 5.22e-06
[order_py_execute_after_rewriter]: 4e-06
[mutable_eliminate]: 0.00046599
[jit_opt_b]: 6.767e-05, [1]
    [Cycle 1]: 6.155e-05, [2]
        [frontend_op_eliminate]: 2.238e-05
        [inline_after_opt_a]: 1.509e-05
[cconv]: 1.727e-05
[loop_unroll]: 0.00042803
[jit_opt_after_cconv]: 0.00012874, [1]
    [Cycle 1]: 0.00012287, [11]
        [c_1]: 2.007e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 5.27999e-06
        [updatestate_assign_eliminate]: 2.20002e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [cse]: 1.524e-05
        [call_graph_tuple_transform]: 1.56e-05
        [tuple_list_get_item_eliminator]: 4.95001e-06
        [none_parameter_eliminate]: 7.00005e-07
        [renormalize]: 4.09986e-07
        [switch_simplify]: 5.35999e-06
[remove_dup_value]: 7.77e-06
[partial_unused_args_eliminate]: 1.75001e-06
[environ_conv]: 3.35998e-06
[add_recomputation]: 3.119e-05
[cse_after_recomputation]: 1.99e-05, [1]
    [Cycle 1]: 1.457e-05, [1]
        [cse]: 8.99e-06
[auto_monad_reorder]: 9.39998e-06
[get_jit_bprop_graph]: 1.39e-06
[rewriter_after_jit_bprop_graph]: 4.48001e-06
[opt_after_jit_grad]: 0.00044384
[symbol_engine_optimizer]: 6.743e-05, [1]
    [Cycle 1]: 6.16e-05, [6]
        [build]: 3.03e-06
        [elim_shapecalc]: 7.13e-06
        [elim_not_effective]: 1.133e-05
        [opt_reshape]: 5.24e-06
        [fold_const_symbol]: 7.74002e-06
        [renormalize]: 3.9002e-07
[validate]: 2.531e-05
Sums
    bootstrap                            :     0.000232s :  2.52%
    type_inference                       :     0.005611s : 60.94%
    event_method                         :     0.000019s :  0.21%
    auto_monad                           :     0.000059s :  0.64%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.43%
    rewriter_before_opt_a                :     0.000074s :  0.80%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000091s :  0.99%
    jit_opt_a.loop_unroll                :     0.000037s :  0.41%
    jit_opt_a.a_1                        :     0.000699s :  7.59%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000472s :  5.13%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.16%
    jit_opt_a.cse                        :     0.000025s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.10%
    rewriter_after_opt_a                 :     0.000024s :  0.26%
    convert_after_rewriter               :     0.000005s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000466s :  5.06%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.24%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000017s :  0.19%
    loop_unroll                          :     0.000428s :  4.65%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000008s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000003s :  0.04%
    add_recomputation                    :     0.000031s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000009s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000444s :  4.82%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000025s :  0.27%


Time group info:
------[substitution.]   0.000211    26
 0.69% :     0.000001s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 1.62% :     0.000003s :      3: substitution.graph_param_transform
85.77% :     0.000181s :      6: substitution.inline
 1.20% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.95% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.40% :     0.000003s :      2: substitution.replace_old_param
 3.28% :     0.000007s :      1: substitution.switch_simplify
 3.66% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005567     2
83.66% :     0.004657s :      1: type_inference.infer
16.34% :     0.000910s :      1: type_inference.specialize
------[replace.]   0.000107     9
55.78% :     0.000060s :      6: replace.inline
19.01% :     0.000020s :      1: replace.switch_simplify
25.21% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000191     9
93.21% :     0.000178s :      6: match.inline
 3.26% :     0.000006s :      1: match.switch_simplify
 3.53% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   801
 1.42% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.21% :     0.000001s :     12: predicate.addn_check_dump
 1.52% :     0.000002s :     12: predicate.addn_zero_filter
 1.96% :     0.000002s :     12: predicate.arithmetic_simplify
 1.47% :     0.000002s :     12: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.20% :     0.000001s :     12: predicate.compare_switch_simplify
 1.41% :     0.000002s :     12: predicate.depend_value_elim
 1.28% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     12: predicate.environ_get_eliminate
 1.18% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.32% :     0.000007s :     26: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.02% :     0.000001s :      6: predicate.less_batch_normalization
 1.66% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     17: predicate.load_eliminater
 1.24% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.47% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000001s :     12: predicate.merge_addn
 1.22% :     0.000001s :     12: predicate.minmaximum_grad
 1.46% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     17: predicate.partial_eliminate
 1.26% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     12: predicate.reduce_eliminate
 1.54% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     20: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.38% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     12: predicate.reshape_eliminate
 1.30% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     12: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.81% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.89% :     0.000004s :     20: predicate.switch_defer_inline
 2.42% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.22% :     0.000010s :     65: predicate.switch_simplify
 1.26% :     0.000002s :     12: predicate.tile_eliminate
 1.24% :     0.000002s :     12: predicate.transpose_eliminate
 1.51% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.22% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     12: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000671    15
49.28% :     0.000331s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.81% :     0.000073s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.91% :     0.000268s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014062    72
 0.24% :     0.000034s :      1: add_recomputation
 0.45% :     0.000063s :      1: auto_monad
 0.08% :     0.000012s :      1: auto_monad_reorder
 1.73% :     0.000243s :      1: bootstrap
 0.14% :     0.000020s :      1: cconv
 0.05% :     0.000007s :      1: convert_after_rewriter
 0.16% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.16% :     0.000022s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
32.92% :     0.004629s :      1: jit_opt_a
 0.93% :     0.000131s :      1: jit_opt_after_cconv
 0.50% :     0.000070s :      1: jit_opt_b
 3.10% :     0.000436s :      1: loop_unroll
 3.37% :     0.000475s :      1: mutable_eliminate
 6.65% :     0.000936s :     26: opt.transform.jit_opt_a
 0.30% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000031s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000028s :      4: opt.transform.symbol_engine_opt
 3.22% :     0.000452s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.30% :     0.000042s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000010s :      1: remove_dup_value
 1.77% :     0.000249s :      1: renormalize.infer
 1.54% :     0.000217s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000027s :      1: rewriter_after_opt_a
 0.55% :     0.000077s :      1: rewriter_before_opt_a
 0.50% :     0.000070s :      1: symbol_engine_optimizer
39.98% :     0.005622s :      1: type_inference


TotalTime = 0.0168507, [33]
[bootstrap]: 0.00025574
[type_inference]: 0.00412707
[event_method]: 1.614e-05
[auto_monad]: 6.108e-05
[graph_reusing]: 5.30001e-06
[pre_auto_parallel]: 2.37001e-06
[py_interpret_to_execute]: 2.834e-05
[rewriter_before_opt_a]: 7.149e-05
[expand_dump_flag]: 3.3e-06
[jit_opt_a]: 0.00458853, [2]
    [Cycle 1]: 0.00128009, [27]
        [switch_simplify]: 4.795e-05
        [loop_unroll]: 2.798e-05
        [a_1]: 0.00047805
        [with_stream_mark]: 1.479e-05
        [recompute_prepare]: 7.12002e-06
        [updatestate_depend_eliminate]: 3.25e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.38002e-06
        [parameter_eliminate]: 1.80001e-06
        [specialize_transform]: 5.58002e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 5.04998e-06
        [meta_shard_fg_expand]: 2.19999e-06
        [get_grad_eliminate_]: 5.24998e-06
        [merge_forward]: 3.52002e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.333e-05
        [j_node_and_user_rematch]: 8.18999e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 8.32e-06
        [inline_without_move]: 5.05001e-06
        [renormalize]: 0.00042157
        [add_forward_monad_depend]: 4.84003e-06
        [auto_monad_grad]: 1.75001e-06
        [auto_monad_eliminator]: 1.25e-05
        [cse]: 2.388e-05
        [replace_applicator]: 1.028e-05
    [Cycle 2]: 0.00029789, [27]
        [switch_simplify]: 5.45001e-06
        [loop_unroll]: 4.82998e-06
        [a_1]: 8.614e-05
        [with_stream_mark]: 7.68001e-06
        [recompute_prepare]: 4.68999e-06
        [updatestate_depend_eliminate]: 2.41e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 9.30013e-07
        [specialize_transform]: 4.77e-06
        [updatestate_useless_node_eliminater]: 4.51002e-06
        [accelerated_algorithm]: 4.57e-06
        [meta_shard_fg_expand]: 1.00999e-06
        [get_grad_eliminate_]: 4.38999e-06
        [merge_forward]: 2.55002e-06
        [cell_reuse_recompute_pass]: 1.48002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.156e-05
        [j_node_and_user_rematch]: 7.11999e-06
        [meta_fg_expand]: 1.42e-06
        [replace_old_param]: 6.51999e-06
        [inline_without_move]: 4.42e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 6.59988e-07
        [auto_monad_eliminator]: 4.70999e-06
        [cse]: 8.85001e-06
        [replace_applicator]: 4.85999e-06
[py_interpret_to_execute_after_opt_a]: 7.27002e-06
[rewriter_after_opt_a]: 2.999e-05
[convert_after_rewriter]: 6.28e-06
[order_py_execute_after_rewriter]: 4.75999e-06
[mutable_eliminate]: 0.0004596
[jit_opt_b]: 4.595e-05, [1]
    [Cycle 1]: 3.986e-05, [2]
        [frontend_op_eliminate]: 1.461e-05
        [inline_after_opt_a]: 1.443e-05
[cconv]: 2.983e-05
[loop_unroll]: 0.00044567
[jit_opt_after_cconv]: 0.00013037, [1]
    [Cycle 1]: 0.00012416, [11]
        [c_1]: 1.963e-05
        [parameter_eliminate]: 2.41998e-06
        [updatestate_depend_eliminate]: 5.35001e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 2.13998e-06
        [cse]: 1.596e-05
        [call_graph_tuple_transform]: 1.654e-05
        [tuple_list_get_item_eliminator]: 4.95001e-06
        [none_parameter_eliminate]: 1.55001e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 5.17e-06
[remove_dup_value]: 1.131e-05
[partial_unused_args_eliminate]: 2.46e-06
[environ_conv]: 4.50001e-06
[add_recomputation]: 4.469e-05
[cse_after_recomputation]: 1.932e-05, [1]
    [Cycle 1]: 1.403e-05, [1]
        [cse]: 8.44002e-06
[auto_monad_reorder]: 1.432e-05
[get_jit_bprop_graph]: 1.51998e-06
[rewriter_after_jit_bprop_graph]: 4.51002e-06
[opt_after_jit_grad]: 0.00044336
[symbol_engine_optimizer]: 6.977e-05, [1]
    [Cycle 1]: 6.364e-05, [6]
        [build]: 3.42002e-06
        [elim_shapecalc]: 7.37002e-06
        [elim_not_effective]: 1.225e-05
        [opt_reshape]: 5.26002e-06
        [fold_const_symbol]: 8.05e-06
        [renormalize]: 3.80009e-07
[validate]: 2.938e-05
[backend_pass]: 9.49978e-07
[task_emit]: 0.00571106
[execute]: 7.1e-06
Sums
    bootstrap                            :     0.000256s :  1.92%
    type_inference                       :     0.004127s : 31.04%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000061s :  0.46%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.21%
    rewriter_before_opt_a                :     0.000071s :  0.54%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000564s :  4.24%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000015s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000422s :  3.17%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000033s :  0.25%
    jit_opt_a.replace_applicator         :     0.000015s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.05%
    rewriter_after_opt_a                 :     0.000030s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000460s :  3.46%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000030s :  0.22%
    loop_unroll                          :     0.000446s :  3.35%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000011s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000045s :  0.34%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000443s :  3.33%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005711s : 42.95%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000170    23
 1.14% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.72% :     0.000005s :      3: substitution.graph_param_transform
82.28% :     0.000140s :      4: substitution.inline
 1.65% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.50% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000003s :      2: substitution.replace_old_param
 6.96% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004071     2
81.92% :     0.003335s :      1: type_inference.infer
18.08% :     0.000736s :      1: type_inference.specialize
------[replace.]   0.000062     6
68.61% :     0.000042s :      4: replace.inline
31.39% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000149     6
92.71% :     0.000138s :      4: match.inline
 7.29% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000106   691
 1.33% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000001s :     10: predicate.addn_zero_filter
 1.86% :     0.000002s :     10: predicate.arithmetic_simplify
 1.33% :     0.000001s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000001s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.15% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.13% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.42% :     0.000006s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.48% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.32% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.50% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000000s :      3: predicate.opt_reshape
 2.34% :     0.000002s :     15: predicate.partial_eliminate
 1.29% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.39% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.47% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.55% :     0.000003s :     16: predicate.switch_defer_inline
 2.20% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.21% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.28% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.74% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.80% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000493    12
50.83% :     0.000251s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.11% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.06% :     0.000217s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018130    76
 0.26% :     0.000047s :      1: add_recomputation
 0.35% :     0.000064s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.49% :     0.000270s :      1: bootstrap
 0.18% :     0.000033s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
25.32% :     0.004591s :      1: jit_opt_a
 0.73% :     0.000133s :      1: jit_opt_after_cconv
 0.27% :     0.000048s :      1: jit_opt_b
 2.50% :     0.000453s :      1: loop_unroll
 2.58% :     0.000467s :      1: mutable_eliminate
 4.17% :     0.000755s :     26: opt.transform.jit_opt_a
 0.24% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.49% :     0.000451s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000031s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000013s :      1: remove_dup_value
 1.18% :     0.000215s :      1: renormalize.infer
 1.11% :     0.000201s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000033s :      1: rewriter_after_opt_a
 0.41% :     0.000074s :      1: rewriter_before_opt_a
 0.40% :     0.000072s :      1: symbol_engine_optimizer
31.56% :     0.005721s :      1: task_emit
22.82% :     0.004137s :      1: type_inference
 0.25% :     0.000046s :      1: validate


TotalTime = 0.0194543, [33]
[bootstrap]: 0.00034029
[type_inference]: 0.00481339
[event_method]: 1.686e-05
[auto_monad]: 4.714e-05
[graph_reusing]: 4.67998e-06
[pre_auto_parallel]: 1.70001e-06
[py_interpret_to_execute]: 3.348e-05
[rewriter_before_opt_a]: 6.554e-05
[expand_dump_flag]: 2.91e-06
[jit_opt_a]: 0.00532095, [2]
    [Cycle 1]: 0.00157693, [27]
        [switch_simplify]: 4.883e-05
        [loop_unroll]: 2.9e-05
        [a_1]: 0.00049283
        [with_stream_mark]: 1.427e-05
        [recompute_prepare]: 7.1e-06
        [updatestate_depend_eliminate]: 3.13e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.01998e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 5.76998e-06
        [updatestate_useless_node_eliminater]: 4.79002e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 5.30001e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.151e-05
        [j_node_and_user_rematch]: 8.92999e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 7.57002e-06
        [inline_without_move]: 5.04e-06
        [renormalize]: 0.00069139
        [add_forward_monad_depend]: 5.88002e-06
        [auto_monad_grad]: 2.14e-06
        [auto_monad_eliminator]: 1.32e-05
        [cse]: 2.841e-05
        [replace_applicator]: 1.189e-05
    [Cycle 2]: 0.00031525, [27]
        [switch_simplify]: 5.66003e-06
        [loop_unroll]: 4.73001e-06
        [a_1]: 8.871e-05
        [with_stream_mark]: 9.14e-06
        [recompute_prepare]: 5.20999e-06
        [updatestate_depend_eliminate]: 2.52001e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 4.92999e-06
        [updatestate_useless_node_eliminater]: 4.63001e-06
        [accelerated_algorithm]: 4.95001e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.39998e-06
        [merge_forward]: 2.17999e-06
        [cell_reuse_recompute_pass]: 1.77001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.162e-05
        [j_node_and_user_rematch]: 7.40998e-06
        [meta_fg_expand]: 1.50999e-06
        [replace_old_param]: 6.99001e-06
        [inline_without_move]: 4.58001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.35999e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 5.07e-06
        [cse]: 1.022e-05
        [replace_applicator]: 4.90999e-06
[py_interpret_to_execute_after_opt_a]: 9.83002e-06
[rewriter_after_opt_a]: 3.501e-05
[convert_after_rewriter]: 7.25e-06
[order_py_execute_after_rewriter]: 7.82e-06
[mutable_eliminate]: 0.00056365
[jit_opt_b]: 5.002e-05, [1]
    [Cycle 1]: 4.301e-05, [2]
        [frontend_op_eliminate]: 1.681e-05
        [inline_after_opt_a]: 1.486e-05
[cconv]: 2.584e-05
[loop_unroll]: 0.00042294
[jit_opt_after_cconv]: 0.00013864, [1]
    [Cycle 1]: 0.00013228, [11]
        [c_1]: 2.125e-05
        [parameter_eliminate]: 2.34001e-06
        [updatestate_depend_eliminate]: 5.29e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 2.03002e-06
        [cse]: 1.752e-05
        [call_graph_tuple_transform]: 1.944e-05
        [tuple_list_get_item_eliminator]: 5.20001e-06
        [none_parameter_eliminate]: 1.96003e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.88998e-06
[remove_dup_value]: 1.422e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 5.20999e-06
[add_recomputation]: 4.837e-05
[cse_after_recomputation]: 2.027e-05, [1]
    [Cycle 1]: 1.495e-05, [1]
        [cse]: 9.30001e-06
[auto_monad_reorder]: 1.619e-05
[get_jit_bprop_graph]: 1.87999e-06
[rewriter_after_jit_bprop_graph]: 5.86e-06
[opt_after_jit_grad]: 0.00049688
[symbol_engine_optimizer]: 8.058e-05, [1]
    [Cycle 1]: 7.244e-05, [6]
        [build]: 4.07e-06
        [elim_shapecalc]: 7.93001e-06
        [elim_not_effective]: 1.345e-05
        [opt_reshape]: 5.89e-06
        [fold_const_symbol]: 1.009e-05
        [renormalize]: 9.30013e-07
[validate]: 4.202e-05
[backend_pass]: 8.99978e-07
[task_emit]: 0.00659199
[execute]: 9.29e-06
Sums
    bootstrap                            :     0.000340s :  2.21%
    type_inference                       :     0.004813s : 31.23%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000047s :  0.31%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000033s :  0.22%
    rewriter_before_opt_a                :     0.000066s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.35%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000582s :  3.77%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000691s :  4.49%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000039s :  0.25%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000035s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000008s :  0.05%
    mutable_eliminate                    :     0.000564s :  3.66%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000026s :  0.17%
    loop_unroll                          :     0.000423s :  2.74%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000048s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000497s :  3.22%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000042s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006592s : 42.76%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000170    23
 1.18% :     0.000002s :      2: substitution.elim_not_effective
 0.88% :     0.000001s :      2: substitution.fold_const_symbol
 3.61% :     0.000006s :      3: substitution.graph_param_transform
83.32% :     0.000141s :      4: substitution.inline
 1.64% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.45% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.92% :     0.000003s :      2: substitution.replace_old_param
 5.00% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004751     2
80.09% :     0.003805s :      1: type_inference.infer
19.91% :     0.000946s :      1: type_inference.specialize
------[replace.]   0.000064     6
64.72% :     0.000041s :      4: replace.inline
35.28% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000146     6
94.91% :     0.000139s :      4: match.inline
 5.09% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.55% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.67% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 2.04% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.38% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.70% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.81% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.65% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.05% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.47% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.79% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000002s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.54% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.93% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.50% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.32% :     0.000008s :     53: predicate.switch_simplify
 1.38% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.11% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.93% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000677    12
43.43% :     0.000294s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.30% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
52.27% :     0.000354s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021027    76
 0.24% :     0.000051s :      1: add_recomputation
 0.24% :     0.000050s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.70% :     0.000358s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
25.32% :     0.005325s :      1: jit_opt_a
 0.67% :     0.000142s :      1: jit_opt_after_cconv
 0.25% :     0.000053s :      1: jit_opt_b
 2.05% :     0.000431s :      1: loop_unroll
 2.72% :     0.000572s :      1: mutable_eliminate
 3.70% :     0.000778s :     26: opt.transform.jit_opt_a
 0.23% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.40% :     0.000506s :      1: opt_after_jit_grad
 0.05% :     0.000010s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.69% :     0.000355s :      1: renormalize.infer
 1.57% :     0.000330s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000038s :      1: rewriter_after_opt_a
 0.32% :     0.000068s :      1: rewriter_before_opt_a
 0.40% :     0.000083s :      1: symbol_engine_optimizer
31.44% :     0.006611s :      1: task_emit
22.95% :     0.004826s :      1: type_inference
 0.30% :     0.000062s :      1: validate


TotalTime = 0.017467, [33]
[bootstrap]: 0.00026996
[type_inference]: 0.00427541
[event_method]: 1.795e-05
[auto_monad]: 6.4e-05
[graph_reusing]: 5.79999e-06
[pre_auto_parallel]: 2.39001e-06
[py_interpret_to_execute]: 3.157e-05
[rewriter_before_opt_a]: 7.699e-05
[expand_dump_flag]: 3.36001e-06
[jit_opt_a]: 0.00488536, [2]
    [Cycle 1]: 0.00143493, [27]
        [switch_simplify]: 5.316e-05
        [loop_unroll]: 3.176e-05
        [a_1]: 0.00050518
        [with_stream_mark]: 1.58e-05
        [recompute_prepare]: 7.31999e-06
        [updatestate_depend_eliminate]: 3.43e-06
        [updatestate_assign_eliminate]: 3.45998e-06
        [updatestate_loads_eliminate]: 2.64999e-06
        [parameter_eliminate]: 2.12999e-06
        [specialize_transform]: 5.97001e-06
        [updatestate_useless_node_eliminater]: 5.23002e-06
        [accelerated_algorithm]: 5.40999e-06
        [meta_shard_fg_expand]: 2.43e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 3.66001e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.433e-05
        [j_node_and_user_rematch]: 8.90001e-06
        [meta_fg_expand]: 2.27001e-06
        [replace_old_param]: 8.74e-06
        [inline_without_move]: 5.14e-06
        [renormalize]: 0.00052496
        [add_forward_monad_depend]: 5.04e-06
        [auto_monad_grad]: 2.05002e-06
        [auto_monad_eliminator]: 1.28e-05
        [cse]: 2.546e-05
        [replace_applicator]: 1.172e-05
    [Cycle 2]: 0.00031696, [27]
        [switch_simplify]: 5.76998e-06
        [loop_unroll]: 4.82e-06
        [a_1]: 9.018e-05
        [with_stream_mark]: 8.75999e-06
        [recompute_prepare]: 4.95999e-06
        [updatestate_depend_eliminate]: 2.50997e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 5.03002e-06
        [updatestate_useless_node_eliminater]: 5.06002e-06
        [accelerated_algorithm]: 4.80999e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.35e-06
        [merge_forward]: 2.41e-06
        [cell_reuse_recompute_pass]: 1.47999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.166e-05
        [j_node_and_user_rematch]: 7.46999e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 6.74001e-06
        [inline_without_move]: 4.58001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.11002e-06
        [auto_monad_grad]: 1.18001e-06
        [auto_monad_eliminator]: 5.04e-06
        [cse]: 1.034e-05
        [replace_applicator]: 5.13002e-06
[py_interpret_to_execute_after_opt_a]: 9.17001e-06
[rewriter_after_opt_a]: 3.067e-05
[convert_after_rewriter]: 7.16999e-06
[order_py_execute_after_rewriter]: 7.05e-06
[mutable_eliminate]: 0.00047817
[jit_opt_b]: 4.851e-05, [1]
    [Cycle 1]: 4.174e-05, [2]
        [frontend_op_eliminate]: 1.614e-05
        [inline_after_opt_a]: 1.457e-05
[cconv]: 2.34e-05
[loop_unroll]: 0.00041059
[jit_opt_after_cconv]: 0.00013474, [1]
    [Cycle 1]: 0.00012835, [11]
        [c_1]: 2.079e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 5.26002e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 2.13002e-06
        [cse]: 1.614e-05
        [call_graph_tuple_transform]: 1.787e-05
        [tuple_list_get_item_eliminator]: 5.37999e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 5.24e-06
[remove_dup_value]: 1.235e-05
[partial_unused_args_eliminate]: 2.06998e-06
[environ_conv]: 5.07e-06
[add_recomputation]: 4.502e-05
[cse_after_recomputation]: 1.986e-05, [1]
    [Cycle 1]: 1.419e-05, [1]
        [cse]: 8.75001e-06
[auto_monad_reorder]: 1.504e-05
[get_jit_bprop_graph]: 1.95001e-06
[rewriter_after_jit_bprop_graph]: 4.25e-06
[opt_after_jit_grad]: 0.00045535
[symbol_engine_optimizer]: 7.711e-05, [1]
    [Cycle 1]: 7.115e-05, [6]
        [build]: 8.67e-06
        [elim_shapecalc]: 7.19001e-06
        [elim_not_effective]: 1.228e-05
        [opt_reshape]: 5.40999e-06
        [fold_const_symbol]: 8.25e-06
        [renormalize]: 5.19998e-07
[validate]: 3.34e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.00579339
[execute]: 7.73001e-06
Sums
    bootstrap                            :     0.000270s :  1.97%
    type_inference                       :     0.004275s : 31.14%
    event_method                         :     0.000018s :  0.13%
    auto_monad                           :     0.000064s :  0.47%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.23%
    rewriter_before_opt_a                :     0.000077s :  0.56%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.43%
    jit_opt_a.loop_unroll                :     0.000037s :  0.27%
    jit_opt_a.a_1                        :     0.000595s :  4.34%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000525s :  3.82%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000036s :  0.26%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000031s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000007s :  0.05%
    mutable_eliminate                    :     0.000478s :  3.48%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000411s :  2.99%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000045s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000455s :  3.32%
    symbol_engine_optimizer.build        :     0.000009s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000033s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005793s : 42.20%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000180    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.97% :     0.000005s :      3: substitution.graph_param_transform
82.36% :     0.000148s :      4: substitution.inline
 1.55% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.50% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000004s :      2: substitution.replace_old_param
 6.83% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004213     2
80.94% :     0.003410s :      1: type_inference.infer
19.06% :     0.000803s :      1: type_inference.specialize
------[replace.]   0.000064     6
67.93% :     0.000044s :      4: replace.inline
32.07% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000157     6
92.89% :     0.000146s :      4: match.inline
 7.11% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.43% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.66% :     0.000002s :     10: predicate.addn_zero_filter
 2.02% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000001s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.81% :     0.000007s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000001s :      6: predicate.less_batch_normalization
 1.78% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.29% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000000s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.82% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.51% :     0.000002s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.92% :     0.000009s :     53: predicate.switch_simplify
 1.10% :     0.000001s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.11% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.44% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000546    12
51.19% :     0.000279s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.92% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.88% :     0.000239s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018897    76
 0.25% :     0.000048s :      1: add_recomputation
 0.36% :     0.000067s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.51% :     0.000285s :      1: bootstrap
 0.27% :     0.000051s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.12% :     0.000022s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.87% :     0.004888s :      1: jit_opt_a
 0.73% :     0.000137s :      1: jit_opt_after_cconv
 0.27% :     0.000051s :      1: jit_opt_b
 2.22% :     0.000419s :      1: loop_unroll
 2.57% :     0.000486s :      1: mutable_eliminate
 4.24% :     0.000801s :     26: opt.transform.jit_opt_a
 0.24% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.45% :     0.000463s :      1: opt_after_jit_grad
 0.05% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.39% :     0.000263s :      1: renormalize.infer
 1.35% :     0.000254s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.43% :     0.000080s :      1: rewriter_before_opt_a
 0.42% :     0.000080s :      1: symbol_engine_optimizer
30.73% :     0.005806s :      1: task_emit
22.69% :     0.004289s :      1: type_inference
 0.27% :     0.000052s :      1: validate


TotalTime = 0.0123822, [30]
[bootstrap]: 0.00024105
[type_inference]: 0.00550221
[event_method]: 1.728e-05
[auto_monad]: 5.298e-05
[graph_reusing]: 3.43e-06
[pre_auto_parallel]: 1.86e-06
[py_interpret_to_execute]: 3.831e-05
[rewriter_before_opt_a]: 7.069e-05
[expand_dump_flag]: 1.74998e-06
[jit_opt_a]: 0.00461335, [2]
    [Cycle 1]: 0.00145261, [27]
        [switch_simplify]: 9.038e-05
        [loop_unroll]: 3.428e-05
        [a_1]: 0.00059768
        [with_stream_mark]: 9.48002e-06
        [recompute_prepare]: 6.19001e-06
        [updatestate_depend_eliminate]: 2.61e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 8.59989e-07
        [specialize_transform]: 6.24001e-06
        [updatestate_useless_node_eliminater]: 5.72001e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 1.05001e-06
        [get_grad_eliminate_]: 5.61e-06
        [merge_forward]: 2.26e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.174e-05
        [j_node_and_user_rematch]: 8.64e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 7.13e-06
        [inline_without_move]: 5.07999e-06
        [renormalize]: 0.00044864
        [add_forward_monad_depend]: 4.06001e-06
        [auto_monad_grad]: 1.14003e-06
        [auto_monad_eliminator]: 8.82999e-06
        [cse]: 1.518e-05
        [replace_applicator]: 1.042e-05
    [Cycle 2]: 0.0002995, [27]
        [switch_simplify]: 5.41998e-06
        [loop_unroll]: 4.80001e-06
        [a_1]: 8.728e-05
        [with_stream_mark]: 7.71999e-06
        [recompute_prepare]: 4.65999e-06
        [updatestate_depend_eliminate]: 2.36998e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.79998e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 5.00001e-06
        [updatestate_useless_node_eliminater]: 4.60001e-06
        [accelerated_algorithm]: 4.48999e-06
        [meta_shard_fg_expand]: 1.06002e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.31998e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.14e-05
        [j_node_and_user_rematch]: 7.68001e-06
        [meta_fg_expand]: 1.46002e-06
        [replace_old_param]: 6.86001e-06
        [inline_without_move]: 4.56002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 9.5999e-07
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 5.04003e-06
        [cse]: 8.30999e-06
        [replace_applicator]: 4.75001e-06
[py_interpret_to_execute_after_opt_a]: 7.16999e-06
[rewriter_after_opt_a]: 2.383e-05
[convert_after_rewriter]: 5.12999e-06
[order_py_execute_after_rewriter]: 3.94002e-06
[mutable_eliminate]: 0.00046489
[jit_opt_b]: 5.557e-05, [1]
    [Cycle 1]: 4.936e-05, [2]
        [frontend_op_eliminate]: 2.338e-05
        [inline_after_opt_a]: 1.481e-05
[cconv]: 1.546e-05
[loop_unroll]: 0.00039328
[jit_opt_after_cconv]: 0.0001287, [1]
    [Cycle 1]: 0.00012268, [11]
        [c_1]: 2.007e-05
        [parameter_eliminate]: 2.29001e-06
        [updatestate_depend_eliminate]: 4.97999e-06
        [updatestate_assign_eliminate]: 2.30002e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [cse]: 1.575e-05
        [call_graph_tuple_transform]: 1.535e-05
        [tuple_list_get_item_eliminator]: 5.09e-06
        [none_parameter_eliminate]: 9.60019e-07
        [renormalize]: 4.89992e-07
        [switch_simplify]: 5.29e-06
[remove_dup_value]: 7.13e-06
[partial_unused_args_eliminate]: 1.48002e-06
[environ_conv]: 3.67002e-06
[add_recomputation]: 3.712e-05
[cse_after_recomputation]: 1.908e-05, [1]
    [Cycle 1]: 1.441e-05, [1]
        [cse]: 8.66002e-06
[auto_monad_reorder]: 8.96998e-06
[get_jit_bprop_graph]: 1.44998e-06
[rewriter_after_jit_bprop_graph]: 4.53999e-06
[opt_after_jit_grad]: 0.00042744
[symbol_engine_optimizer]: 6.695e-05, [1]
    [Cycle 1]: 6.109e-05, [6]
        [build]: 2.32999e-06
        [elim_shapecalc]: 7.06001e-06
        [elim_not_effective]: 1.213e-05
        [opt_reshape]: 5.12e-06
        [fold_const_symbol]: 7.66999e-06
        [renormalize]: 4.30009e-07
[validate]: 2.318e-05
Sums
    bootstrap                            :     0.000241s :  2.68%
    type_inference                       :     0.005502s : 61.12%
    event_method                         :     0.000017s :  0.19%
    auto_monad                           :     0.000053s :  0.59%
    graph_reusing                        :     0.000003s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.43%
    rewriter_before_opt_a                :     0.000071s :  0.79%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000096s :  1.06%
    jit_opt_a.loop_unroll                :     0.000039s :  0.43%
    jit_opt_a.a_1                        :     0.000685s :  7.61%
    jit_opt_a.with_stream_mark           :     0.000017s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000449s :  4.98%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.15%
    jit_opt_a.cse                        :     0.000023s :  0.26%
    jit_opt_a.replace_applicator         :     0.000015s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.08%
    rewriter_after_opt_a                 :     0.000024s :  0.26%
    convert_after_rewriter               :     0.000005s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000465s :  5.16%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.26%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000015s :  0.17%
    loop_unroll                          :     0.000393s :  4.37%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000007s :  0.08%
    partial_unused_args_eliminate        :     0.000001s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000037s :  0.41%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000009s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000427s :  4.75%
    symbol_engine_optimizer.build        :     0.000002s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000023s :  0.26%


Time group info:
------[substitution.]   0.000196    26
 0.74% :     0.000001s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 1.79% :     0.000003s :      3: substitution.graph_param_transform
85.30% :     0.000167s :      6: substitution.inline
 1.21% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.06% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.25% :     0.000002s :      2: substitution.replace_old_param
 3.43% :     0.000007s :      1: substitution.switch_simplify
 3.74% :     0.000007s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005461     2
84.23% :     0.004600s :      1: type_inference.infer
15.77% :     0.000861s :      1: type_inference.specialize
------[replace.]   0.000107     9
53.50% :     0.000057s :      6: replace.inline
20.71% :     0.000022s :      1: replace.switch_simplify
25.79% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000176     9
93.03% :     0.000164s :      6: match.inline
 3.38% :     0.000006s :      1: match.switch_simplify
 3.59% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   801
 1.48% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.21% :     0.000002s :     12: predicate.addn_check_dump
 1.41% :     0.000002s :     12: predicate.addn_zero_filter
 1.75% :     0.000002s :     12: predicate.arithmetic_simplify
 1.51% :     0.000002s :     12: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.18% :     0.000001s :     12: predicate.compare_switch_simplify
 1.41% :     0.000002s :     12: predicate.depend_value_elim
 1.21% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     12: predicate.environ_get_eliminate
 1.27% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.22% :     0.000007s :     26: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     17: predicate.load_eliminater
 1.20% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.39% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.27% :     0.000002s :     12: predicate.merge_addn
 1.17% :     0.000001s :     12: predicate.minmaximum_grad
 1.42% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     17: predicate.partial_eliminate
 1.38% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.73% :     0.000002s :     12: predicate.reduce_eliminate
 1.54% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     20: predicate.replace_applicator
 0.45% :     0.000001s :      6: predicate.replace_old_param
 0.21% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     12: predicate.reshape_eliminate
 1.43% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.62% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.49% :     0.000002s :     12: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.32% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.77% :     0.000003s :     20: predicate.switch_defer_inline
 2.45% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.65% :     0.000011s :     65: predicate.switch_simplify
 1.43% :     0.000002s :     12: predicate.tile_eliminate
 1.32% :     0.000002s :     12: predicate.transpose_eliminate
 1.40% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.13% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.79% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000652    15
49.28% :     0.000321s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.59% :     0.000069s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.13% :     0.000262s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013833    72
 0.29% :     0.000040s :      1: add_recomputation
 0.41% :     0.000056s :      1: auto_monad
 0.08% :     0.000012s :      1: auto_monad_reorder
 1.83% :     0.000253s :      1: bootstrap
 0.13% :     0.000018s :      1: cconv
 0.05% :     0.000007s :      1: convert_after_rewriter
 0.15% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
33.37% :     0.004616s :      1: jit_opt_a
 0.95% :     0.000131s :      1: jit_opt_after_cconv
 0.42% :     0.000058s :      1: jit_opt_b
 2.89% :     0.000400s :      1: loop_unroll
 3.42% :     0.000473s :      1: mutable_eliminate
 6.71% :     0.000928s :     26: opt.transform.jit_opt_a
 0.30% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.23% :     0.000032s :      4: opt.transform.jit_opt_b
 0.08% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000029s :      4: opt.transform.symbol_engine_opt
 3.14% :     0.000434s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.29% :     0.000041s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000009s :      1: remove_dup_value
 1.60% :     0.000222s :      1: renormalize.infer
 1.60% :     0.000222s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000027s :      1: rewriter_after_opt_a
 0.53% :     0.000073s :      1: rewriter_before_opt_a
 0.50% :     0.000069s :      1: symbol_engine_optimizer
39.83% :     0.005510s :      1: type_inference


TotalTime = 0.0168595, [33]
[bootstrap]: 0.00026198
[type_inference]: 0.004198
[event_method]: 1.603e-05
[auto_monad]: 6.038e-05
[graph_reusing]: 5.77999e-06
[pre_auto_parallel]: 2.48002e-06
[py_interpret_to_execute]: 2.936e-05
[rewriter_before_opt_a]: 6.883e-05
[expand_dump_flag]: 2.93998e-06
[jit_opt_a]: 0.00466469, [2]
    [Cycle 1]: 0.00134508, [27]
        [switch_simplify]: 4.861e-05
        [loop_unroll]: 2.859e-05
        [a_1]: 0.00048324
        [with_stream_mark]: 1.35e-05
        [recompute_prepare]: 6.89999e-06
        [updatestate_depend_eliminate]: 3.58e-06
        [updatestate_assign_eliminate]: 3.18998e-06
        [updatestate_loads_eliminate]: 2.64001e-06
        [parameter_eliminate]: 1.77999e-06
        [specialize_transform]: 5.47001e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.29e-06
        [meta_shard_fg_expand]: 1.87001e-06
        [get_grad_eliminate_]: 5.30001e-06
        [merge_forward]: 4e-06
        [cell_reuse_recompute_pass]: 1.05999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.333e-05
        [j_node_and_user_rematch]: 8.59e-06
        [meta_fg_expand]: 2.05002e-06
        [replace_old_param]: 8.17e-06
        [inline_without_move]: 5.10999e-06
        [renormalize]: 0.00045636
        [add_forward_monad_depend]: 5.30001e-06
        [auto_monad_grad]: 2.37999e-06
        [auto_monad_eliminator]: 1.211e-05
        [cse]: 2.814e-05
        [replace_applicator]: 1.104e-05
    [Cycle 2]: 0.00031809, [27]
        [switch_simplify]: 5.51e-06
        [loop_unroll]: 4.64998e-06
        [a_1]: 9.277e-05
        [with_stream_mark]: 7.38999e-06
        [recompute_prepare]: 5.19e-06
        [updatestate_depend_eliminate]: 2.34001e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 8.79983e-07
        [specialize_transform]: 4.59998e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 4.89998e-06
        [meta_shard_fg_expand]: 1.05001e-06
        [get_grad_eliminate_]: 4.60999e-06
        [merge_forward]: 2.29999e-06
        [cell_reuse_recompute_pass]: 1.97001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.297e-05
        [j_node_and_user_rematch]: 8.2e-06
        [meta_fg_expand]: 1.34e-06
        [replace_old_param]: 6.84999e-06
        [inline_without_move]: 4.58999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 7.80012e-07
        [auto_monad_eliminator]: 4.99e-06
        [cse]: 1.021e-05
        [replace_applicator]: 4.90001e-06
[py_interpret_to_execute_after_opt_a]: 7.75e-06
[rewriter_after_opt_a]: 3.168e-05
[convert_after_rewriter]: 8.74e-06
[order_py_execute_after_rewriter]: 5.92001e-06
[mutable_eliminate]: 0.00046419
[jit_opt_b]: 4.767e-05, [1]
    [Cycle 1]: 4.112e-05, [2]
        [frontend_op_eliminate]: 1.515e-05
        [inline_after_opt_a]: 1.438e-05
[cconv]: 2.305e-05
[loop_unroll]: 0.00041362
[jit_opt_after_cconv]: 0.000132, [1]
    [Cycle 1]: 0.00012573, [11]
        [c_1]: 1.923e-05
        [parameter_eliminate]: 2.46e-06
        [updatestate_depend_eliminate]: 5.02e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 1.607e-05
        [call_graph_tuple_transform]: 1.764e-05
        [tuple_list_get_item_eliminator]: 4.94998e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.05999e-06
[remove_dup_value]: 1.266e-05
[partial_unused_args_eliminate]: 2.24999e-06
[environ_conv]: 4.88001e-06
[add_recomputation]: 4.376e-05
[cse_after_recomputation]: 1.984e-05, [1]
    [Cycle 1]: 1.473e-05, [1]
        [cse]: 8.95999e-06
[auto_monad_reorder]: 1.461e-05
[get_jit_bprop_graph]: 2.16e-06
[rewriter_after_jit_bprop_graph]: 4.55001e-06
[opt_after_jit_grad]: 0.00043366
[symbol_engine_optimizer]: 6.841e-05, [1]
    [Cycle 1]: 6.243e-05, [6]
        [build]: 3.25e-06
        [elim_shapecalc]: 7.07002e-06
        [elim_not_effective]: 1.184e-05
        [opt_reshape]: 5.17e-06
        [fold_const_symbol]: 8.32e-06
        [renormalize]: 3.69997e-07
[validate]: 3.058e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00559662
[execute]: 6.40002e-06
Sums
    bootstrap                            :     0.000262s :  1.97%
    type_inference                       :     0.004198s : 31.61%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000060s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.22%
    rewriter_before_opt_a                :     0.000069s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.41%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000576s :  4.34%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000456s :  3.44%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000038s :  0.29%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000032s :  0.24%
    convert_after_rewriter               :     0.000009s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000464s :  3.50%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000414s :  3.11%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000044s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000434s :  3.27%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005597s : 42.14%
    execute                              :     0.000006s :  0.05%


Time group info:
------[substitution.]   0.000175    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.78% :     0.000001s :      2: substitution.fold_const_symbol
 2.93% :     0.000005s :      3: substitution.graph_param_transform
82.42% :     0.000144s :      4: substitution.inline
 1.86% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.44% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.99% :     0.000003s :      2: substitution.replace_old_param
 6.49% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004137     2
81.44% :     0.003369s :      1: type_inference.infer
18.56% :     0.000768s :      1: type_inference.specialize
------[replace.]   0.000060     6
67.61% :     0.000040s :      4: replace.inline
32.39% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000152     6
93.22% :     0.000142s :      4: match.inline
 6.78% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.35% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.20% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.82% :     0.000002s :     10: predicate.addn_zero_filter
 1.95% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000001s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.15% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.09% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.29% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.39% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.79% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.82% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.23% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.74% :     0.000008s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.35% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.04% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000529    12
48.55% :     0.000257s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.84% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.61% :     0.000247s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018191    76
 0.26% :     0.000047s :      1: add_recomputation
 0.35% :     0.000064s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.53% :     0.000279s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
25.66% :     0.004667s :      1: jit_opt_a
 0.74% :     0.000135s :      1: jit_opt_after_cconv
 0.28% :     0.000050s :      1: jit_opt_b
 2.32% :     0.000421s :      1: loop_unroll
 2.59% :     0.000472s :      1: mutable_eliminate
 4.24% :     0.000772s :     26: opt.transform.jit_opt_a
 0.24% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.42% :     0.000441s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.27% :     0.000231s :      1: renormalize.infer
 1.20% :     0.000218s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000035s :      1: rewriter_after_opt_a
 0.39% :     0.000071s :      1: rewriter_before_opt_a
 0.39% :     0.000071s :      1: symbol_engine_optimizer
30.81% :     0.005605s :      1: task_emit
23.14% :     0.004209s :      1: type_inference
 0.26% :     0.000048s :      1: validate


TotalTime = 0.016101, [33]
[bootstrap]: 0.00025888
[type_inference]: 0.00414213
[event_method]: 1.684e-05
[auto_monad]: 5.024e-05
[graph_reusing]: 4.04002e-06
[pre_auto_parallel]: 1.95001e-06
[py_interpret_to_execute]: 2.849e-05
[rewriter_before_opt_a]: 6.379e-05
[expand_dump_flag]: 2.64001e-06
[jit_opt_a]: 0.00463172, [2]
    [Cycle 1]: 0.00133793, [27]
        [switch_simplify]: 4.662e-05
        [loop_unroll]: 2.882e-05
        [a_1]: 0.00048304
        [with_stream_mark]: 1.317e-05
        [recompute_prepare]: 6.93e-06
        [updatestate_depend_eliminate]: 2.96999e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 5.59e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 5.27001e-06
        [meta_shard_fg_expand]: 1.39998e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 2.51998e-06
        [cell_reuse_recompute_pass]: 9.20001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.07e-05
        [j_node_and_user_rematch]: 8.40999e-06
        [meta_fg_expand]: 1.99e-06
        [replace_old_param]: 7.65e-06
        [inline_without_move]: 5.25999e-06
        [renormalize]: 0.00046641
        [add_forward_monad_depend]: 4.69002e-06
        [auto_monad_grad]: 1.93002e-06
        [auto_monad_eliminator]: 1.006e-05
        [cse]: 1.639e-05
        [replace_applicator]: 1.094e-05
    [Cycle 2]: 0.00032301, [27]
        [switch_simplify]: 5.11997e-06
        [loop_unroll]: 4.60001e-06
        [a_1]: 8.858e-05
        [with_stream_mark]: 9.49999e-06
        [recompute_prepare]: 5.37001e-06
        [updatestate_depend_eliminate]: 2.51998e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 5.05001e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 5.10001e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.47998e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 1.74e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.135e-05
        [j_node_and_user_rematch]: 7.7e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 7.05e-06
        [inline_without_move]: 4.55001e-06
        [renormalize]: 3.99887e-08
        [add_forward_monad_depend]: 1.24998e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 5.66e-06
        [cse]: 1.043e-05
        [replace_applicator]: 4.90001e-06
[py_interpret_to_execute_after_opt_a]: 8.06001e-06
[rewriter_after_opt_a]: 2.652e-05
[convert_after_rewriter]: 6.56999e-06
[order_py_execute_after_rewriter]: 4.35e-06
[mutable_eliminate]: 0.00047347
[jit_opt_b]: 4.731e-05, [1]
    [Cycle 1]: 4.102e-05, [2]
        [frontend_op_eliminate]: 1.58e-05
        [inline_after_opt_a]: 1.419e-05
[cconv]: 1.906e-05
[loop_unroll]: 0.00040457
[jit_opt_after_cconv]: 0.00013078, [1]
    [Cycle 1]: 0.00012478, [11]
        [c_1]: 1.975e-05
        [parameter_eliminate]: 2.53e-06
        [updatestate_depend_eliminate]: 5.01002e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.19999e-06
        [cse]: 1.559e-05
        [call_graph_tuple_transform]: 1.613e-05
        [tuple_list_get_item_eliminator]: 5.49e-06
        [none_parameter_eliminate]: 1.34e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.30999e-06
[remove_dup_value]: 7.44002e-06
[partial_unused_args_eliminate]: 1.69998e-06
[environ_conv]: 3.94997e-06
[add_recomputation]: 3.596e-05
[cse_after_recomputation]: 1.887e-05, [1]
    [Cycle 1]: 1.377e-05, [1]
        [cse]: 8.33001e-06
[auto_monad_reorder]: 1.091e-05
[get_jit_bprop_graph]: 1.60001e-06
[rewriter_after_jit_bprop_graph]: 4.54998e-06
[opt_after_jit_grad]: 0.0004448
[symbol_engine_optimizer]: 6.88e-05, [1]
    [Cycle 1]: 6.295e-05, [6]
        [build]: 2.96001e-06
        [elim_shapecalc]: 7.08e-06
        [elim_not_effective]: 1.237e-05
        [opt_reshape]: 5.44998e-06
        [fold_const_symbol]: 8.18999e-06
        [renormalize]: 5.59987e-07
[validate]: 2.713e-05
[backend_pass]: 7.50006e-07
[task_emit]: 0.00497396
[execute]: 5.79999e-06
Sums
    bootstrap                            :     0.000259s :  2.06%
    type_inference                       :     0.004142s : 33.04%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000050s :  0.40%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.23%
    rewriter_before_opt_a                :     0.000064s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000052s :  0.41%
    jit_opt_a.loop_unroll                :     0.000033s :  0.27%
    jit_opt_a.a_1                        :     0.000572s :  4.56%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000466s :  3.72%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.13%
    jit_opt_a.cse                        :     0.000027s :  0.21%
    jit_opt_a.replace_applicator         :     0.000016s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000027s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000473s :  3.78%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000019s :  0.15%
    loop_unroll                          :     0.000405s :  3.23%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000007s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000036s :  0.29%
    cse_after_recomputation.cse          :     0.000008s :  0.07%
    auto_monad_reorder                   :     0.000011s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000445s :  3.55%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000027s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.004974s : 39.67%
    execute                              :     0.000006s :  0.05%


Time group info:
------[substitution.]   0.000163    23
 1.07% :     0.000002s :      2: substitution.elim_not_effective
 0.80% :     0.000001s :      2: substitution.fold_const_symbol
 2.26% :     0.000004s :      3: substitution.graph_param_transform
83.97% :     0.000137s :      4: substitution.inline
 1.75% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.44% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000003s :      2: substitution.replace_old_param
 5.68% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004090     2
82.16% :     0.003360s :      1: type_inference.infer
17.84% :     0.000730s :      1: type_inference.specialize
------[replace.]   0.000062     6
66.49% :     0.000041s :      4: replace.inline
33.51% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000143     6
94.31% :     0.000134s :      4: match.inline
 5.69% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.34% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.15% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.15% :     0.000002s :     10: predicate.arithmetic_simplify
 1.45% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000001s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.65% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.34% :     0.000001s :     10: predicate.environ_get_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.26% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.38% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000002s :     15: predicate.partial_eliminate
 1.25% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.87% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.66% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.29% :     0.000008s :     53: predicate.switch_simplify
 1.29% :     0.000001s :     10: predicate.tile_eliminate
 1.33% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.43% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000511    12
46.82% :     0.000239s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.09% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.09% :     0.000241s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017434    76
 0.22% :     0.000039s :      1: add_recomputation
 0.31% :     0.000053s :      1: auto_monad
 0.08% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.56% :     0.000272s :      1: bootstrap
 0.13% :     0.000022s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.12% :     0.000020s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
26.58% :     0.004635s :      1: jit_opt_a
 0.77% :     0.000134s :      1: jit_opt_after_cconv
 0.29% :     0.000050s :      1: jit_opt_b
 2.36% :     0.000412s :      1: loop_unroll
 2.76% :     0.000482s :      1: mutable_eliminate
 4.38% :     0.000763s :     26: opt.transform.jit_opt_a
 0.25% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.60% :     0.000453s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000031s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000010s :      1: remove_dup_value
 1.27% :     0.000222s :      1: renormalize.infer
 1.37% :     0.000238s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000032s :      1: rewriter_after_opt_a
 0.38% :     0.000067s :      1: rewriter_before_opt_a
 0.41% :     0.000071s :      1: symbol_engine_optimizer
28.58% :     0.004983s :      1: task_emit
23.82% :     0.004154s :      1: type_inference
 0.25% :     0.000043s :      1: validate


TotalTime = 0.0131126, [30]
[bootstrap]: 0.00025504
[type_inference]: 0.00581599
[event_method]: 1.884e-05
[auto_monad]: 5.834e-05
[graph_reusing]: 3.95e-06
[pre_auto_parallel]: 1.90001e-06
[py_interpret_to_execute]: 4.058e-05
[rewriter_before_opt_a]: 7.464e-05
[expand_dump_flag]: 1.80001e-06
[jit_opt_a]: 0.00487223, [2]
    [Cycle 1]: 0.00158475, [27]
        [switch_simplify]: 8.998e-05
        [loop_unroll]: 3.379e-05
        [a_1]: 0.00066807
        [with_stream_mark]: 1.208e-05
        [recompute_prepare]: 7.13e-06
        [updatestate_depend_eliminate]: 2.77002e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 6.00002e-06
        [updatestate_useless_node_eliminater]: 4.76002e-06
        [accelerated_algorithm]: 5.66998e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 2.39999e-06
        [cell_reuse_recompute_pass]: 8.70001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.159e-05
        [j_node_and_user_rematch]: 9.72001e-06
        [meta_fg_expand]: 1.72001e-06
        [replace_old_param]: 8.02e-06
        [inline_without_move]: 5.32001e-06
        [renormalize]: 0.00049838
        [add_forward_monad_depend]: 3.85e-06
        [auto_monad_grad]: 1.17999e-06
        [auto_monad_eliminator]: 9.42001e-06
        [cse]: 1.623e-05
        [replace_applicator]: 1.122e-05
    [Cycle 2]: 0.00030403, [27]
        [switch_simplify]: 5.81e-06
        [loop_unroll]: 4.75999e-06
        [a_1]: 8.804e-05
        [with_stream_mark]: 7.88999e-06
        [recompute_prepare]: 4.95999e-06
        [updatestate_depend_eliminate]: 2.27999e-06
        [updatestate_assign_eliminate]: 1.91998e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 4.87e-06
        [updatestate_useless_node_eliminater]: 4.60001e-06
        [accelerated_algorithm]: 4.65999e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.65001e-06
        [merge_forward]: 2.40002e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.181e-05
        [j_node_and_user_rematch]: 8.06001e-06
        [meta_fg_expand]: 1.39e-06
        [replace_old_param]: 6.80002e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 4.75999e-06
        [cse]: 9.66e-06
        [replace_applicator]: 4.85999e-06
[py_interpret_to_execute_after_opt_a]: 7.79997e-06
[rewriter_after_opt_a]: 2.373e-05
[convert_after_rewriter]: 5.39e-06
[order_py_execute_after_rewriter]: 3.86999e-06
[mutable_eliminate]: 0.00047955
[jit_opt_b]: 4.897e-05, [1]
    [Cycle 1]: 4.26e-05, [2]
        [frontend_op_eliminate]: 1.621e-05
        [inline_after_opt_a]: 1.518e-05
[cconv]: 1.538e-05
[loop_unroll]: 0.00043281
[jit_opt_after_cconv]: 0.00014791, [1]
    [Cycle 1]: 0.00014067, [11]
        [c_1]: 2.298e-05
        [parameter_eliminate]: 2.59001e-06
        [updatestate_depend_eliminate]: 5.55001e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 2.11998e-06
        [cse]: 1.661e-05
        [call_graph_tuple_transform]: 2.526e-05
        [tuple_list_get_item_eliminator]: 5.66e-06
        [none_parameter_eliminate]: 1.10999e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 5.29e-06
[remove_dup_value]: 6.91999e-06
[partial_unused_args_eliminate]: 1.44e-06
[environ_conv]: 3.3e-06
[add_recomputation]: 3.278e-05
[cse_after_recomputation]: 2.112e-05, [1]
    [Cycle 1]: 1.582e-05, [1]
        [cse]: 9.92999e-06
[auto_monad_reorder]: 8.62998e-06
[get_jit_bprop_graph]: 1.39998e-06
[rewriter_after_jit_bprop_graph]: 5.19e-06
[opt_after_jit_grad]: 0.0004764
[symbol_engine_optimizer]: 7.031e-05, [1]
    [Cycle 1]: 6.428e-05, [6]
        [build]: 2.79001e-06
        [elim_shapecalc]: 7.46999e-06
        [elim_not_effective]: 1.226e-05
        [opt_reshape]: 5.52999e-06
        [fold_const_symbol]: 8.10999e-06
        [renormalize]: 5.50004e-07
[validate]: 2.353e-05
Sums
    bootstrap                            :     0.000255s :  2.66%
    type_inference                       :     0.005816s : 60.68%
    event_method                         :     0.000019s :  0.20%
    auto_monad                           :     0.000058s :  0.61%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000041s :  0.42%
    rewriter_before_opt_a                :     0.000075s :  0.78%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000096s :  1.00%
    jit_opt_a.loop_unroll                :     0.000039s :  0.40%
    jit_opt_a.a_1                        :     0.000756s :  7.89%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000498s :  5.20%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.15%
    jit_opt_a.cse                        :     0.000026s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.08%
    rewriter_after_opt_a                 :     0.000024s :  0.25%
    convert_after_rewriter               :     0.000005s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000480s :  5.00%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000015s :  0.16%
    loop_unroll                          :     0.000433s :  4.52%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.26%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000007s :  0.07%
    partial_unused_args_eliminate        :     0.000001s :  0.02%
    environ_conv                         :     0.000003s :  0.03%
    add_recomputation                    :     0.000033s :  0.34%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000009s :  0.09%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000476s :  4.97%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000024s :  0.25%


Time group info:
------[substitution.]   0.000252    26
 0.52% :     0.000001s :      2: substitution.elim_not_effective
 0.38% :     0.000001s :      2: substitution.fold_const_symbol
 1.47% :     0.000004s :      3: substitution.graph_param_transform
87.67% :     0.000221s :      6: substitution.inline
 1.21% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.71% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.14% :     0.000003s :      2: substitution.replace_old_param
 2.84% :     0.000007s :      1: substitution.switch_simplify
 3.06% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005769     2
82.75% :     0.004774s :      1: type_inference.infer
17.25% :     0.000995s :      1: type_inference.specialize
------[replace.]   0.000114     9
56.53% :     0.000065s :      6: replace.inline
18.44% :     0.000021s :      1: replace.switch_simplify
25.04% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000231     9
94.34% :     0.000218s :      6: match.inline
 2.80% :     0.000006s :      1: match.switch_simplify
 2.86% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   801
 1.40% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.24% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     12: predicate.addn_check_dump
 1.44% :     0.000002s :     12: predicate.addn_zero_filter
 1.95% :     0.000002s :     12: predicate.arithmetic_simplify
 1.24% :     0.000002s :     12: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.20% :     0.000002s :     12: predicate.compare_switch_simplify
 1.42% :     0.000002s :     12: predicate.depend_value_elim
 1.22% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     12: predicate.environ_get_eliminate
 1.17% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.20% :     0.000007s :     26: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :      6: predicate.less_batch_normalization
 1.76% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     17: predicate.load_eliminater
 1.60% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.29% :     0.000005s :     40: predicate.loop_unroll_before_grad
 2.09% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000002s :     12: predicate.merge_addn
 1.18% :     0.000001s :     12: predicate.minmaximum_grad
 1.57% :     0.000002s :      3: predicate.mutable_eliminate
 0.37% :     0.000000s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     17: predicate.partial_eliminate
 1.30% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     12: predicate.reduce_eliminate
 1.53% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     20: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     12: predicate.reshape_eliminate
 1.28% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     12: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.72% :     0.000003s :     20: predicate.switch_defer_inline
 2.43% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.32% :     0.000011s :     65: predicate.switch_simplify
 1.34% :     0.000002s :     12: predicate.tile_eliminate
 1.22% :     0.000002s :     12: predicate.transpose_eliminate
 1.46% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.26% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     12: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000735    15
43.65% :     0.000321s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.35% :     0.000091s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.00% :     0.000324s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014698    72
 0.24% :     0.000035s :      1: add_recomputation
 0.42% :     0.000062s :      1: auto_monad
 0.08% :     0.000011s :      1: auto_monad_reorder
 1.83% :     0.000269s :      1: bootstrap
 0.12% :     0.000018s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.16% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000005s :      1: environ_conv
 0.15% :     0.000022s :      1: event_method
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
33.17% :     0.004875s :      1: jit_opt_a
 1.03% :     0.000151s :      1: jit_opt_after_cconv
 0.35% :     0.000052s :      1: jit_opt_b
 3.00% :     0.000440s :      1: loop_unroll
 3.32% :     0.000488s :      1: mutable_eliminate
 6.81% :     0.001001s :     26: opt.transform.jit_opt_a
 0.38% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.30% :     0.000484s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.29% :     0.000043s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000009s :      1: remove_dup_value
 1.80% :     0.000264s :      1: renormalize.infer
 1.55% :     0.000229s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000027s :      1: rewriter_after_opt_a
 0.53% :     0.000077s :      1: rewriter_before_opt_a
 0.49% :     0.000073s :      1: symbol_engine_optimizer
39.65% :     0.005828s :      1: type_inference


TotalTime = 0.0176461, [33]
[bootstrap]: 0.00025645
[type_inference]: 0.00448596
[event_method]: 1.685e-05
[auto_monad]: 6.192e-05
[graph_reusing]: 6.51999e-06
[pre_auto_parallel]: 2.51998e-06
[py_interpret_to_execute]: 3.088e-05
[rewriter_before_opt_a]: 7.05e-05
[expand_dump_flag]: 3.56001e-06
[jit_opt_a]: 0.0048803, [2]
    [Cycle 1]: 0.00142626, [27]
        [switch_simplify]: 4.978e-05
        [loop_unroll]: 2.811e-05
        [a_1]: 0.00049103
        [with_stream_mark]: 1.59e-05
        [recompute_prepare]: 7.24001e-06
        [updatestate_depend_eliminate]: 3.36001e-06
        [updatestate_assign_eliminate]: 3.11999e-06
        [updatestate_loads_eliminate]: 2.56998e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 6.00002e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 5.69e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 5.59e-06
        [merge_forward]: 4.01001e-06
        [cell_reuse_recompute_pass]: 1.36998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.388e-05
        [j_node_and_user_rematch]: 9.26998e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 8.43001e-06
        [inline_without_move]: 5.74e-06
        [renormalize]: 0.00052737
        [add_forward_monad_depend]: 5.42001e-06
        [auto_monad_grad]: 1.64e-06
        [auto_monad_eliminator]: 1.317e-05
        [cse]: 2.837e-05
        [replace_applicator]: 1.112e-05
    [Cycle 2]: 0.00030848, [27]
        [switch_simplify]: 5.34e-06
        [loop_unroll]: 4.63001e-06
        [a_1]: 8.728e-05
        [with_stream_mark]: 8.56002e-06
        [recompute_prepare]: 4.87e-06
        [updatestate_depend_eliminate]: 2.46998e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 4.77e-06
        [updatestate_useless_node_eliminater]: 4.60999e-06
        [accelerated_algorithm]: 4.81002e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.38001e-06
        [merge_forward]: 2.29999e-06
        [cell_reuse_recompute_pass]: 1.55001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.227e-05
        [j_node_and_user_rematch]: 7.53e-06
        [meta_fg_expand]: 1.51002e-06
        [replace_old_param]: 6.93e-06
        [inline_without_move]: 4.74e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 9.20001e-07
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 5.04998e-06
        [cse]: 9.84001e-06
        [replace_applicator]: 4.79e-06
[py_interpret_to_execute_after_opt_a]: 7.5e-06
[rewriter_after_opt_a]: 3.302e-05
[convert_after_rewriter]: 7.45e-06
[order_py_execute_after_rewriter]: 5.25999e-06
[mutable_eliminate]: 0.00051571
[jit_opt_b]: 4.995e-05, [1]
    [Cycle 1]: 4.312e-05, [2]
        [frontend_op_eliminate]: 1.613e-05
        [inline_after_opt_a]: 1.513e-05
[cconv]: 2.431e-05
[loop_unroll]: 0.00042807
[jit_opt_after_cconv]: 0.00013314, [1]
    [Cycle 1]: 0.00012702, [11]
        [c_1]: 1.953e-05
        [parameter_eliminate]: 2.51998e-06
        [updatestate_depend_eliminate]: 5.27001e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.704e-05
        [call_graph_tuple_transform]: 1.711e-05
        [tuple_list_get_item_eliminator]: 4.93001e-06
        [none_parameter_eliminate]: 2.44001e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 4.90999e-06
[remove_dup_value]: 1.25e-05
[partial_unused_args_eliminate]: 2.24001e-06
[environ_conv]: 5.04e-06
[add_recomputation]: 4.66e-05
[cse_after_recomputation]: 1.93e-05, [1]
    [Cycle 1]: 1.41e-05, [1]
        [cse]: 8.47e-06
[auto_monad_reorder]: 1.494e-05
[get_jit_bprop_graph]: 1.55001e-06
[rewriter_after_jit_bprop_graph]: 5.07e-06
[opt_after_jit_grad]: 0.00044568
[symbol_engine_optimizer]: 6.97e-05, [1]
    [Cycle 1]: 6.363e-05, [6]
        [build]: 3.29001e-06
        [elim_shapecalc]: 7.5e-06
        [elim_not_effective]: 1.197e-05
        [opt_reshape]: 5.25999e-06
        [fold_const_symbol]: 8.12998e-06
        [renormalize]: 5.50004e-07
[validate]: 3.118e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.0057906
[execute]: 7.35998e-06
Sums
    bootstrap                            :     0.000256s :  1.84%
    type_inference                       :     0.004486s : 32.20%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000062s :  0.44%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.22%
    rewriter_before_opt_a                :     0.000071s :  0.51%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000055s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.24%
    jit_opt_a.a_1                        :     0.000578s :  4.15%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000527s :  3.79%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000038s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.05%
    rewriter_after_opt_a                 :     0.000033s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000516s :  3.70%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000428s :  3.07%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.33%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000446s :  3.20%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000031s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005791s : 41.57%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000176    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.80% :     0.000005s :      3: substitution.graph_param_transform
82.37% :     0.000145s :      4: substitution.inline
 1.75% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.27% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.94% :     0.000003s :      2: substitution.replace_old_param
 7.07% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004422     2
82.40% :     0.003644s :      1: type_inference.infer
17.60% :     0.000778s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.74% :     0.000042s :      4: replace.inline
32.26% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000154     6
92.60% :     0.000143s :      4: match.inline
 7.40% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.29% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.47% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.54% :     0.000002s :     10: predicate.addn_zero_filter
 1.77% :     0.000002s :     10: predicate.arithmetic_simplify
 1.41% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.14% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.42% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.66% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.74% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.88% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 2.07% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.79% :     0.000001s :      6: predicate.replace_old_param
 0.40% :     0.000000s :      3: predicate.reset_defer_inline
 1.53% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.71% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.93% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000528    12
52.28% :     0.000276s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.89% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.84% :     0.000226s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019055    76
 0.26% :     0.000049s :      1: add_recomputation
 0.34% :     0.000065s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.42% :     0.000271s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
25.62% :     0.004883s :      1: jit_opt_a
 0.71% :     0.000136s :      1: jit_opt_after_cconv
 0.28% :     0.000053s :      1: jit_opt_b
 2.29% :     0.000436s :      1: loop_unroll
 2.75% :     0.000525s :      1: mutable_eliminate
 4.07% :     0.000776s :     26: opt.transform.jit_opt_a
 0.23% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.38% :     0.000453s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000033s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.49% :     0.000285s :      1: renormalize.infer
 1.23% :     0.000235s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000036s :      1: rewriter_after_opt_a
 0.38% :     0.000073s :      1: rewriter_before_opt_a
 0.38% :     0.000072s :      1: symbol_engine_optimizer
30.43% :     0.005799s :      1: task_emit
23.60% :     0.004497s :      1: type_inference
 0.26% :     0.000049s :      1: validate


TotalTime = 0.0172433, [33]
[bootstrap]: 0.00026473
[type_inference]: 0.00423851
[event_method]: 1.683e-05
[auto_monad]: 6.321e-05
[graph_reusing]: 5.31998e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 3.127e-05
[rewriter_before_opt_a]: 6.677e-05
[expand_dump_flag]: 2.97002e-06
[jit_opt_a]: 0.00492635, [2]
    [Cycle 1]: 0.0014818, [27]
        [switch_simplify]: 4.945e-05
        [loop_unroll]: 2.903e-05
        [a_1]: 0.00052506
        [with_stream_mark]: 1.973e-05
        [recompute_prepare]: 8.02998e-06
        [updatestate_depend_eliminate]: 3.78001e-06
        [updatestate_assign_eliminate]: 2.71999e-06
        [updatestate_loads_eliminate]: 2.63003e-06
        [parameter_eliminate]: 1.65001e-06
        [specialize_transform]: 6.13002e-06
        [updatestate_useless_node_eliminater]: 5.30001e-06
        [accelerated_algorithm]: 5.67001e-06
        [meta_shard_fg_expand]: 1.94999e-06
        [get_grad_eliminate_]: 5.74999e-06
        [merge_forward]: 3.76999e-06
        [cell_reuse_recompute_pass]: 1.08001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.49e-05
        [j_node_and_user_rematch]: 9.37001e-06
        [meta_fg_expand]: 2.04999e-06
        [replace_old_param]: 8.99e-06
        [inline_without_move]: 5.42001e-06
        [renormalize]: 0.00055383
        [add_forward_monad_depend]: 5.00001e-06
        [auto_monad_grad]: 2.56998e-06
        [auto_monad_eliminator]: 1.269e-05
        [cse]: 2.146e-05
        [replace_applicator]: 1.054e-05
    [Cycle 2]: 0.00031419, [27]
        [switch_simplify]: 5.61e-06
        [loop_unroll]: 4.72998e-06
        [a_1]: 9.012e-05
        [with_stream_mark]: 8.21002e-06
        [recompute_prepare]: 4.94998e-06
        [updatestate_depend_eliminate]: 2.69999e-06
        [updatestate_assign_eliminate]: 1.86e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.12999e-06
        [specialize_transform]: 4.77998e-06
        [updatestate_useless_node_eliminater]: 4.58999e-06
        [accelerated_algorithm]: 5.12e-06
        [meta_shard_fg_expand]: 1.28002e-06
        [get_grad_eliminate_]: 4.22e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.34998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.186e-05
        [j_node_and_user_rematch]: 7.48e-06
        [meta_fg_expand]: 1.45001e-06
        [replace_old_param]: 6.76999e-06
        [inline_without_move]: 4.82e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.30999e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 5.39e-06
        [cse]: 9.89001e-06
        [replace_applicator]: 4.93001e-06
[py_interpret_to_execute_after_opt_a]: 9.54e-06
[rewriter_after_opt_a]: 3.077e-05
[convert_after_rewriter]: 7.28999e-06
[order_py_execute_after_rewriter]: 4.82e-06
[mutable_eliminate]: 0.00048484
[jit_opt_b]: 4.566e-05, [1]
    [Cycle 1]: 3.913e-05, [2]
        [frontend_op_eliminate]: 1.469e-05
        [inline_after_opt_a]: 1.367e-05
[cconv]: 2.431e-05
[loop_unroll]: 0.00041152
[jit_opt_after_cconv]: 0.00013608, [1]
    [Cycle 1]: 0.0001297, [11]
        [c_1]: 2.129e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 5.02e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.63e-05
        [call_graph_tuple_transform]: 1.806e-05
        [tuple_list_get_item_eliminator]: 5.14998e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 5.34998e-06
[remove_dup_value]: 1.374e-05
[partial_unused_args_eliminate]: 2.54001e-06
[environ_conv]: 4.43999e-06
[add_recomputation]: 4.294e-05
[cse_after_recomputation]: 2.041e-05, [1]
    [Cycle 1]: 1.513e-05, [1]
        [cse]: 9.13002e-06
[auto_monad_reorder]: 1.495e-05
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 5.17e-06
[opt_after_jit_grad]: 0.00044168
[symbol_engine_optimizer]: 6.985e-05, [1]
    [Cycle 1]: 6.377e-05, [6]
        [build]: 2.78003e-06
        [elim_shapecalc]: 7.31999e-06
        [elim_not_effective]: 1.19e-05
        [opt_reshape]: 5.45001e-06
        [fold_const_symbol]: 8.13001e-06
        [renormalize]: 5.8001e-07
[validate]: 3.031e-05
[backend_pass]: 1.09e-06
[task_emit]: 0.00563656
[execute]: 7.11001e-06
Sums
    bootstrap                            :     0.000265s :  1.96%
    type_inference                       :     0.004239s : 31.31%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000063s :  0.47%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.23%
    rewriter_before_opt_a                :     0.000067s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000615s :  4.54%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000554s :  4.09%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000031s :  0.23%
    jit_opt_a.replace_applicator         :     0.000015s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000031s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000485s :  3.58%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000412s :  3.04%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000043s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000442s :  3.26%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000030s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005637s : 41.63%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000185    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.61% :     0.000005s :      3: substitution.graph_param_transform
83.31% :     0.000154s :      4: substitution.inline
 1.62% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.49% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.05% :     0.000004s :      2: substitution.replace_old_param
 6.39% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004180     2
81.04% :     0.003388s :      1: type_inference.infer
18.96% :     0.000793s :      1: type_inference.specialize
------[replace.]   0.000062     6
64.85% :     0.000040s :      4: replace.inline
35.15% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     6
93.43% :     0.000152s :      4: match.inline
 6.57% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
11.25% :     0.000014s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000002s :     10: predicate.addn_zero_filter
 1.85% :     0.000002s :     10: predicate.arithmetic_simplify
 1.10% :     0.000001s :     10: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.19% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.46% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.01% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000001s :      6: predicate.j_node_and_user_rematch
 0.99% :     0.000001s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.55% :     0.000002s :     15: predicate.load_eliminater
 1.24% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.54% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.77% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 1.56% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     15: predicate.partial_eliminate
 1.31% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.82% :     0.000002s :     10: predicate.reduce_eliminate
 1.32% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.53% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.24% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.63% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000002s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.18% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.21% :     0.000003s :     16: predicate.switch_defer_inline
 1.93% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.49% :     0.000008s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.25% :     0.000002s :     10: predicate.transpose_eliminate
 1.19% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.41% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.34% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.29% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.47% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.48% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000539    12
50.32% :     0.000271s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.20% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.48% :     0.000234s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018717    76
 0.24% :     0.000046s :      1: add_recomputation
 0.36% :     0.000067s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.48% :     0.000278s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.34% :     0.004929s :      1: jit_opt_a
 0.74% :     0.000139s :      1: jit_opt_after_cconv
 0.26% :     0.000048s :      1: jit_opt_b
 2.24% :     0.000420s :      1: loop_unroll
 2.63% :     0.000493s :      1: mutable_eliminate
 4.36% :     0.000816s :     26: opt.transform.jit_opt_a
 0.25% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.40% :     0.000449s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 1.56% :     0.000291s :      1: renormalize.infer
 1.36% :     0.000255s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.37% :     0.000069s :      1: rewriter_before_opt_a
 0.39% :     0.000072s :      1: symbol_engine_optimizer
30.18% :     0.005648s :      1: task_emit
22.71% :     0.004251s :      1: type_inference
 0.26% :     0.000049s :      1: validate


TotalTime = 0.0182119, [33]
[bootstrap]: 0.00036984
[type_inference]: 0.00462488
[event_method]: 1.67e-05
[auto_monad]: 6.236e-05
[graph_reusing]: 5.99999e-06
[pre_auto_parallel]: 2.41998e-06
[py_interpret_to_execute]: 2.969e-05
[rewriter_before_opt_a]: 7.456e-05
[expand_dump_flag]: 3.03998e-06
[jit_opt_a]: 0.0051603, [2]
    [Cycle 1]: 0.00145807, [27]
        [switch_simplify]: 5.009e-05
        [loop_unroll]: 2.929e-05
        [a_1]: 0.00049932
        [with_stream_mark]: 1.796e-05
        [recompute_prepare]: 7.01999e-06
        [updatestate_depend_eliminate]: 3.88999e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 2.50002e-06
        [parameter_eliminate]: 1.83002e-06
        [specialize_transform]: 5.76e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 5.03002e-06
        [meta_shard_fg_expand]: 2.61e-06
        [get_grad_eliminate_]: 5.17e-06
        [merge_forward]: 4.05998e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.396e-05
        [j_node_and_user_rematch]: 8.89e-06
        [meta_fg_expand]: 2.56e-06
        [replace_old_param]: 8.42998e-06
        [inline_without_move]: 5.14e-06
        [renormalize]: 0.00055346
        [add_forward_monad_depend]: 5.38002e-06
        [auto_monad_grad]: 2.15002e-06
        [auto_monad_eliminator]: 1.306e-05
        [cse]: 2.702e-05
        [replace_applicator]: 1.22e-05
    [Cycle 2]: 0.00030839, [27]
        [switch_simplify]: 5.41002e-06
        [loop_unroll]: 5.04e-06
        [a_1]: 8.707e-05
        [with_stream_mark]: 8.76002e-06
        [recompute_prepare]: 4.89998e-06
        [updatestate_depend_eliminate]: 2.37001e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 4.72e-06
        [updatestate_useless_node_eliminater]: 4.77998e-06
        [accelerated_algorithm]: 5.06002e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 4.51002e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.242e-05
        [j_node_and_user_rematch]: 7.48e-06
        [meta_fg_expand]: 1.37999e-06
        [replace_old_param]: 6.84001e-06
        [inline_without_move]: 4.50001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 4.72e-06
        [cse]: 9.19998e-06
        [replace_applicator]: 4.97e-06
[py_interpret_to_execute_after_opt_a]: 7.98999e-06
[rewriter_after_opt_a]: 3.525e-05
[convert_after_rewriter]: 6.66e-06
[order_py_execute_after_rewriter]: 5.61e-06
[mutable_eliminate]: 0.00048871
[jit_opt_b]: 4.71e-05, [1]
    [Cycle 1]: 4.033e-05, [2]
        [frontend_op_eliminate]: 1.508e-05
        [inline_after_opt_a]: 1.398e-05
[cconv]: 2.387e-05
[loop_unroll]: 0.00041208
[jit_opt_after_cconv]: 0.00013755, [1]
    [Cycle 1]: 0.00013124, [11]
        [c_1]: 2.214e-05
        [parameter_eliminate]: 2.47001e-06
        [updatestate_depend_eliminate]: 5.17999e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.693e-05
        [call_graph_tuple_transform]: 1.799e-05
        [tuple_list_get_item_eliminator]: 5.25001e-06
        [none_parameter_eliminate]: 1.32999e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 4.96997e-06
[remove_dup_value]: 1.38e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 4.58001e-06
[add_recomputation]: 5.01e-05
[cse_after_recomputation]: 1.912e-05, [1]
    [Cycle 1]: 1.383e-05, [1]
        [cse]: 8.32e-06
[auto_monad_reorder]: 1.465e-05
[get_jit_bprop_graph]: 2.10002e-06
[rewriter_after_jit_bprop_graph]: 4.72e-06
[opt_after_jit_grad]: 0.00044792
[symbol_engine_optimizer]: 7.03e-05, [1]
    [Cycle 1]: 6.39e-05, [6]
        [build]: 3.62002e-06
        [elim_shapecalc]: 7.71001e-06
        [elim_not_effective]: 1.21e-05
        [opt_reshape]: 5.10999e-06
        [fold_const_symbol]: 8.29002e-06
        [renormalize]: 5.10016e-07
[validate]: 3.024e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.00580674
[execute]: 6.33998e-06
Sums
    bootstrap                            :     0.000370s :  2.60%
    type_inference                       :     0.004625s : 32.57%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000062s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.21%
    rewriter_before_opt_a                :     0.000075s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000586s :  4.13%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000554s :  3.90%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000036s :  0.26%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000035s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000489s :  3.44%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000412s :  2.90%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.35%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000448s :  3.15%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000030s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005807s : 40.89%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000183    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.81% :     0.000005s :      3: substitution.graph_param_transform
82.73% :     0.000151s :      4: substitution.inline
 1.68% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.48% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.97% :     0.000004s :      2: substitution.replace_old_param
 6.65% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004560     2
82.93% :     0.003782s :      1: type_inference.infer
17.07% :     0.000779s :      1: type_inference.specialize
------[replace.]   0.000064     6
67.82% :     0.000044s :      4: replace.inline
32.18% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000160     6
93.11% :     0.000149s :      4: match.inline
 6.89% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.24% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.74% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 3.01% :     0.000003s :     10: predicate.arithmetic_simplify
 1.50% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.62% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 1.20% :     0.000001s :      6: predicate.get_grad_eliminate
 0.34% :     0.000000s :      3: predicate.graph_param_transform
 5.38% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.29% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.69% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000000s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.76% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.80% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.25% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.65% :     0.000009s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000548    12
53.28% :     0.000292s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.03% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.69% :     0.000229s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019655    76
 0.27% :     0.000053s :      1: add_recomputation
 0.33% :     0.000066s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.97% :     0.000388s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.27% :     0.005163s :      1: jit_opt_a
 0.71% :     0.000140s :      1: jit_opt_after_cconv
 0.25% :     0.000050s :      1: jit_opt_b
 2.34% :     0.000459s :      1: loop_unroll
 2.53% :     0.000497s :      1: mutable_eliminate
 4.00% :     0.000786s :     26: opt.transform.jit_opt_a
 0.24% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.32% :     0.000456s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000032s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.44% :     0.000283s :      1: renormalize.infer
 1.34% :     0.000263s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000040s :      1: rewriter_after_opt_a
 0.40% :     0.000078s :      1: rewriter_before_opt_a
 0.37% :     0.000073s :      1: symbol_engine_optimizer
29.59% :     0.005815s :      1: task_emit
23.60% :     0.004639s :      1: type_inference
 0.24% :     0.000048s :      1: validate


TotalTime = 0.0132423, [30]
[bootstrap]: 0.00025339
[type_inference]: 0.00595161
[event_method]: 2.021e-05
[auto_monad]: 7.437e-05
[graph_reusing]: 5.41002e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 4.278e-05
[rewriter_before_opt_a]: 8.069e-05
[expand_dump_flag]: 2.84999e-06
[jit_opt_a]: 0.00489128, [2]
    [Cycle 1]: 0.00155458, [27]
        [switch_simplify]: 9.676e-05
        [loop_unroll]: 3.461e-05
        [a_1]: 0.00065315
        [with_stream_mark]: 1.164e-05
        [recompute_prepare]: 6.76999e-06
        [updatestate_depend_eliminate]: 3.01001e-06
        [updatestate_assign_eliminate]: 2.30002e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 5.86998e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.45001e-06
        [meta_shard_fg_expand]: 1.35001e-06
        [get_grad_eliminate_]: 5.61e-06
        [merge_forward]: 2.81e-06
        [cell_reuse_recompute_pass]: 8.10018e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.174e-05
        [j_node_and_user_rematch]: 9.39998e-06
        [meta_fg_expand]: 2.10002e-06
        [replace_old_param]: 7.49002e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 0.00047713
        [add_forward_monad_depend]: 4.15999e-06
        [auto_monad_grad]: 2.14999e-06
        [auto_monad_eliminator]: 1.005e-05
        [cse]: 1.512e-05
        [replace_applicator]: 1.102e-05
    [Cycle 2]: 0.0003059, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 4.80001e-06
        [a_1]: 8.899e-05
        [with_stream_mark]: 8.47e-06
        [recompute_prepare]: 5.17e-06
        [updatestate_depend_eliminate]: 2.49001e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 1.07998e-06
        [specialize_transform]: 5.10001e-06
        [updatestate_useless_node_eliminater]: 4.58001e-06
        [accelerated_algorithm]: 4.58999e-06
        [meta_shard_fg_expand]: 1.15999e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.27001e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.238e-05
        [j_node_and_user_rematch]: 7.66999e-06
        [meta_fg_expand]: 1.50999e-06
        [replace_old_param]: 7.00002e-06
        [inline_without_move]: 4.58001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 6.00005e-07
        [auto_monad_eliminator]: 4.94e-06
        [cse]: 9.00999e-06
        [replace_applicator]: 4.85001e-06
[py_interpret_to_execute_after_opt_a]: 7.36999e-06
[rewriter_after_opt_a]: 2.489e-05
[convert_after_rewriter]: 5.77999e-06
[order_py_execute_after_rewriter]: 4.61002e-06
[mutable_eliminate]: 0.00046458
[jit_opt_b]: 4.91e-05, [1]
    [Cycle 1]: 4.295e-05, [2]
        [frontend_op_eliminate]: 1.613e-05
        [inline_after_opt_a]: 1.535e-05
[cconv]: 1.788e-05
[loop_unroll]: 0.00041724
[jit_opt_after_cconv]: 0.00013488, [1]
    [Cycle 1]: 0.00012841, [11]
        [c_1]: 2.087e-05
        [parameter_eliminate]: 2.44001e-06
        [updatestate_depend_eliminate]: 5.42001e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.629e-05
        [call_graph_tuple_transform]: 1.73e-05
        [tuple_list_get_item_eliminator]: 5.37999e-06
        [none_parameter_eliminate]: 1.02998e-06
        [renormalize]: 3.30008e-07
        [switch_simplify]: 5.27001e-06
[remove_dup_value]: 8.03001e-06
[partial_unused_args_eliminate]: 2.09e-06
[environ_conv]: 3.93001e-06
[add_recomputation]: 3.56e-05
[cse_after_recomputation]: 1.996e-05, [1]
    [Cycle 1]: 1.46e-05, [1]
        [cse]: 9.09e-06
[auto_monad_reorder]: 1.849e-05
[get_jit_bprop_graph]: 1.64e-06
[rewriter_after_jit_bprop_graph]: 5.34e-06
[opt_after_jit_grad]: 0.00044978
[symbol_engine_optimizer]: 6.996e-05, [1]
    [Cycle 1]: 6.392e-05, [6]
        [build]: 2.99999e-06
        [elim_shapecalc]: 7.23e-06
        [elim_not_effective]: 1.262e-05
        [opt_reshape]: 5.79e-06
        [fold_const_symbol]: 8.03999e-06
        [renormalize]: 4.30009e-07
[validate]: 2.726e-05
Sums
    bootstrap                            :     0.000253s :  2.62%
    type_inference                       :     0.005952s : 61.51%
    event_method                         :     0.000020s :  0.21%
    auto_monad                           :     0.000074s :  0.77%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.44%
    rewriter_before_opt_a                :     0.000081s :  0.83%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000102s :  1.06%
    jit_opt_a.loop_unroll                :     0.000039s :  0.41%
    jit_opt_a.a_1                        :     0.000742s :  7.67%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000014s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000477s :  4.93%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.15%
    jit_opt_a.cse                        :     0.000024s :  0.25%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.08%
    rewriter_after_opt_a                 :     0.000025s :  0.26%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000465s :  4.80%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000018s :  0.18%
    loop_unroll                          :     0.000417s :  4.31%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000008s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000036s :  0.37%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000018s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000450s :  4.65%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.28%


Time group info:
------[substitution.]   0.000238    26
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 1.55% :     0.000004s :      3: substitution.graph_param_transform
85.66% :     0.000204s :      6: substitution.inline
 1.33% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.84% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.04% :     0.000002s :      2: substitution.replace_old_param
 3.35% :     0.000008s :      1: substitution.switch_simplify
 3.95% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005899     2
83.48% :     0.004924s :      1: type_inference.infer
16.52% :     0.000975s :      1: type_inference.specialize
------[replace.]   0.000123     9
55.91% :     0.000068s :      6: replace.inline
20.15% :     0.000025s :      1: replace.switch_simplify
23.93% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000216     9
92.90% :     0.000201s :      6: match.inline
 3.32% :     0.000007s :      1: match.switch_simplify
 3.78% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   801
 1.47% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     12: predicate.addn_check_dump
 1.43% :     0.000002s :     12: predicate.addn_zero_filter
 1.95% :     0.000002s :     12: predicate.arithmetic_simplify
 1.27% :     0.000002s :     12: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.18% :     0.000001s :     12: predicate.compare_switch_simplify
 1.49% :     0.000002s :     12: predicate.depend_value_elim
 1.17% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.43% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     12: predicate.environ_get_eliminate
 1.17% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.51% :     0.000007s :     26: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.95% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     17: predicate.load_eliminater
 1.27% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.43% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.93% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000002s :     12: predicate.merge_addn
 1.17% :     0.000001s :     12: predicate.minmaximum_grad
 1.42% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     17: predicate.partial_eliminate
 1.18% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     12: predicate.reduce_eliminate
 1.61% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     20: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     12: predicate.reshape_eliminate
 1.62% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     12: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.53% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.72% :     0.000003s :     20: predicate.switch_defer_inline
 2.32% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.36% :     0.000011s :     65: predicate.switch_simplify
 1.23% :     0.000002s :     12: predicate.tile_eliminate
 1.24% :     0.000002s :     12: predicate.transpose_eliminate
 1.61% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.18% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000717    15
47.25% :     0.000339s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.00% :     0.000086s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.74% :     0.000292s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014782    72
 0.26% :     0.000038s :      1: add_recomputation
 0.53% :     0.000078s :      1: auto_monad
 0.15% :     0.000022s :      1: auto_monad_reorder
 1.80% :     0.000266s :      1: bootstrap
 0.14% :     0.000020s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
33.11% :     0.004894s :      1: jit_opt_a
 0.93% :     0.000137s :      1: jit_opt_after_cconv
 0.35% :     0.000052s :      1: jit_opt_b
 2.87% :     0.000425s :      1: loop_unroll
 3.20% :     0.000473s :      1: mutable_eliminate
 6.72% :     0.000994s :     26: opt.transform.jit_opt_a
 0.30% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.09% :     0.000457s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.31% :     0.000045s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000010s :      1: remove_dup_value
 1.65% :     0.000245s :      1: renormalize.infer
 1.53% :     0.000226s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000028s :      1: rewriter_after_opt_a
 0.56% :     0.000083s :      1: rewriter_before_opt_a
 0.49% :     0.000072s :      1: symbol_engine_optimizer
40.34% :     0.005962s :      1: type_inference


TotalTime = 0.0256815, [33]
[bootstrap]: 0.00034684
[type_inference]: 0.0067794
[event_method]: 2.02e-05
[auto_monad]: 8.386e-05
[graph_reusing]: 6.49999e-06
[pre_auto_parallel]: 2.79999e-06
[py_interpret_to_execute]: 5.964e-05
[rewriter_before_opt_a]: 8.038e-05
[expand_dump_flag]: 2.83e-06
[jit_opt_a]: 0.00807622, [2]
    [Cycle 1]: 0.00365361, [27]
        [switch_simplify]: 5.3e-05
        [loop_unroll]: 2.898e-05
        [a_1]: 0.00228094
        [with_stream_mark]: 2.022e-05
        [recompute_prepare]: 9.38002e-06
        [updatestate_depend_eliminate]: 3.35e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [parameter_eliminate]: 1.76998e-06
        [specialize_transform]: 6.21e-06
        [updatestate_useless_node_eliminater]: 4.91002e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 2.44999e-06
        [get_grad_eliminate_]: 5.30001e-06
        [merge_forward]: 3.15998e-06
        [cell_reuse_recompute_pass]: 1.84e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.297e-05
        [j_node_and_user_rematch]: 1.043e-05
        [meta_fg_expand]: 2.53003e-06
        [replace_old_param]: 9.04e-06
        [inline_without_move]: 5.66e-06
        [renormalize]: 0.00093701
        [add_forward_monad_depend]: 5.04e-06
        [auto_monad_grad]: 2.64001e-06
        [auto_monad_eliminator]: 1.385e-05
        [cse]: 2.71e-05
        [replace_applicator]: 1.23e-05
    [Cycle 2]: 0.00035676, [27]
        [switch_simplify]: 5.18002e-06
        [loop_unroll]: 5.04e-06
        [a_1]: 0.00011713
        [with_stream_mark]: 1.009e-05
        [recompute_prepare]: 5.20999e-06
        [updatestate_depend_eliminate]: 2.64999e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 4.90001e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 5.57001e-06
        [meta_shard_fg_expand]: 1.52001e-06
        [get_grad_eliminate_]: 4.38001e-06
        [merge_forward]: 2.68998e-06
        [cell_reuse_recompute_pass]: 2.21e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.273e-05
        [j_node_and_user_rematch]: 7.56999e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 7.8e-06
        [inline_without_move]: 5.15999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 1.01002e-06
        [auto_monad_eliminator]: 5.77001e-06
        [cse]: 1.207e-05
        [replace_applicator]: 5.61e-06
[py_interpret_to_execute_after_opt_a]: 1.18e-05
[rewriter_after_opt_a]: 4.112e-05
[convert_after_rewriter]: 7.91001e-06
[order_py_execute_after_rewriter]: 5.27001e-06
[mutable_eliminate]: 0.00078868
[jit_opt_b]: 5.131e-05, [1]
    [Cycle 1]: 4.388e-05, [2]
        [frontend_op_eliminate]: 1.73e-05
        [inline_after_opt_a]: 1.538e-05
[cconv]: 2.71e-05
[loop_unroll]: 0.00043682
[jit_opt_after_cconv]: 0.00014234, [1]
    [Cycle 1]: 0.00013483, [11]
        [c_1]: 2.11e-05
        [parameter_eliminate]: 2.89001e-06
        [updatestate_depend_eliminate]: 5.19998e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 1.883e-05
        [call_graph_tuple_transform]: 1.897e-05
        [tuple_list_get_item_eliminator]: 5.69999e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 5.20999e-06
[remove_dup_value]: 1.445e-05
[partial_unused_args_eliminate]: 2.16003e-06
[environ_conv]: 5.75001e-06
[add_recomputation]: 9.392e-05
[cse_after_recomputation]: 2.278e-05, [1]
    [Cycle 1]: 1.661e-05, [1]
        [cse]: 1.065e-05
[auto_monad_reorder]: 1.408e-05
[get_jit_bprop_graph]: 2.36e-06
[rewriter_after_jit_bprop_graph]: 5.74e-06
[opt_after_jit_grad]: 0.00047432
[symbol_engine_optimizer]: 7.559e-05, [1]
    [Cycle 1]: 6.897e-05, [6]
        [build]: 4.30999e-06
        [elim_shapecalc]: 7.91001e-06
        [elim_not_effective]: 1.315e-05
        [opt_reshape]: 5.91e-06
        [fold_const_symbol]: 9.02999e-06
        [renormalize]: 4.80009e-07
[validate]: 4.161e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.00770023
[execute]: 8.01001e-06
Sums
    bootstrap                            :     0.000347s :  1.66%
    type_inference                       :     0.006779s : 32.37%
    event_method                         :     0.000020s :  0.10%
    auto_monad                           :     0.000084s :  0.40%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000060s :  0.28%
    rewriter_before_opt_a                :     0.000080s :  0.38%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000058s :  0.28%
    jit_opt_a.loop_unroll                :     0.000034s :  0.16%
    jit_opt_a.a_1                        :     0.002398s : 11.45%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.08%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.000937s :  4.47%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.09%
    jit_opt_a.cse                        :     0.000039s :  0.19%
    jit_opt_a.replace_applicator         :     0.000018s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.06%
    rewriter_after_opt_a                 :     0.000041s :  0.20%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000789s :  3.77%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.07%
    cconv                                :     0.000027s :  0.13%
    loop_unroll                          :     0.000437s :  2.09%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000014s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000094s :  0.45%
    cse_after_recomputation.cse          :     0.000011s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000474s :  2.27%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000042s :  0.20%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007700s : 36.77%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000248    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.22% :     0.000006s :      3: substitution.graph_param_transform
84.95% :     0.000211s :      4: substitution.inline
 1.59% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.76% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.58% :     0.000004s :      2: substitution.replace_old_param
 6.56% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006682     2
60.30% :     0.004029s :      1: type_inference.infer
39.70% :     0.002653s :      1: type_inference.specialize
------[replace.]   0.000094     6
53.41% :     0.000050s :      4: replace.inline
46.59% :     0.000044s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000223     6
93.59% :     0.000209s :      4: match.inline
 6.41% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.001781   691
 0.11% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.09% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.07% :     0.000001s :     10: predicate.addn_check_dump
 0.08% :     0.000001s :     10: predicate.addn_zero_filter
 0.13% :     0.000002s :     10: predicate.arithmetic_simplify
 0.08% :     0.000002s :     10: predicate.cast_eliminate
 0.03% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.07% :     0.000001s :     10: predicate.compare_switch_simplify
 0.09% :     0.000002s :     10: predicate.depend_value_elim
 0.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 0.10% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.08% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.02% :     0.000000s :      3: predicate.elim_not_effective
 0.04% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.08% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.07% :     0.000001s :     10: predicate.environ_get_eliminate
 0.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.01% :     0.000000s :      3: predicate.fold_const_symbol
 0.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.02% :     0.000000s :      3: predicate.graph_param_transform
 0.36% :     0.000006s :     22: predicate.inline
 0.06% :     0.000001s :      6: predicate.inline_without_move
 0.02% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.11% :     0.000002s :      6: predicate.less_batch_normalization
 0.10% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 0.12% :     0.000002s :     15: predicate.load_eliminater
 0.09% :     0.000002s :      3: predicate.loop_unroll_after_grad
 0.26% :     0.000005s :     34: predicate.loop_unroll_before_grad
 0.15% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.07% :     0.000001s :     10: predicate.merge_addn
 0.07% :     0.000001s :     10: predicate.minmaximum_grad
 0.11% :     0.000002s :      3: predicate.mutable_eliminate
 0.03% :     0.000001s :      3: predicate.opt_reshape
 0.16% :     0.000003s :     15: predicate.partial_eliminate
 0.09% :     0.000002s :     10: predicate.print_const_string_wrapper
 0.13% :     0.000002s :     10: predicate.reduce_eliminate
 0.10% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.04% :     0.000001s :      6: predicate.remove_not_recompute_node
 0.12% :     0.000002s :     18: predicate.replace_applicator
 0.04% :     0.000001s :      6: predicate.replace_old_param
 0.02% :     0.000000s :      3: predicate.reset_defer_inline
 0.09% :     0.000002s :     10: predicate.reshape_eliminate
 0.08% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.07% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.10% :     0.000002s :     10: predicate.same_eliminate
 0.05% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.08% :     0.000001s :      6: predicate.special_op_eliminate
 0.07% :     0.000001s :      6: predicate.specialize_transform
 0.09% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 0.09% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.03% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 0.16% :     0.000003s :     16: predicate.switch_defer_inline
 0.15% :     0.000003s :     16: predicate.switch_layer_defer_inline
 0.46% :     0.000008s :     53: predicate.switch_simplify
 0.08% :     0.000001s :     10: predicate.tile_eliminate
 0.09% :     0.000002s :     10: predicate.transpose_eliminate
 0.09% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.08% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 0.23% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 0.67% :     0.000012s :     10: predicate.tuple_list_set_item_eliminator
 0.12% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 0.14% :     0.000003s :     15: predicate.updatestate_pure_node_eliminater
92.93% :     0.001655s :     21: predicate.updatestate_useless_node_eliminater
 0.12% :     0.000002s :     10: predicate.value_based_eliminate
 0.02% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.04% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000684    12
45.62% :     0.000312s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.24% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.14% :     0.000315s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.029307    76
 0.33% :     0.000097s :      1: add_recomputation
 0.30% :     0.000088s :      1: auto_monad
 0.06% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.26% :     0.000368s :      1: bootstrap
 0.10% :     0.000030s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000024s :      1: event_method
 0.04% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
27.57% :     0.008080s :      1: jit_opt_a
 0.50% :     0.000145s :      1: jit_opt_after_cconv
 0.18% :     0.000054s :      1: jit_opt_b
 1.52% :     0.000445s :      1: loop_unroll
 2.73% :     0.000799s :      1: mutable_eliminate
 8.90% :     0.002609s :     26: opt.transform.jit_opt_a
 0.16% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000026s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.65% :     0.000482s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000062s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 1.79% :     0.000525s :      1: renormalize.infer
 1.38% :     0.000403s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000045s :      1: rewriter_after_opt_a
 0.28% :     0.000083s :      1: rewriter_before_opt_a
 0.27% :     0.000078s :      1: symbol_engine_optimizer
26.34% :     0.007720s :      1: task_emit
23.20% :     0.006800s :      1: type_inference
 0.22% :     0.000063s :      1: validate


TotalTime = 0.0135694, [30]
[bootstrap]: 0.00024752
[type_inference]: 0.00592434
[event_method]: 1.962e-05
[auto_monad]: 6.043e-05
[graph_reusing]: 4e-06
[pre_auto_parallel]: 1.89999e-06
[py_interpret_to_execute]: 4.54e-05
[rewriter_before_opt_a]: 7.55e-05
[expand_dump_flag]: 1.97999e-06
[jit_opt_a]: 0.00507164, [2]
    [Cycle 1]: 0.00165009, [27]
        [switch_simplify]: 8.456e-05
        [loop_unroll]: 3.423e-05
        [a_1]: 0.00068206
        [with_stream_mark]: 9.06002e-06
        [recompute_prepare]: 6.44001e-06
        [updatestate_depend_eliminate]: 2.73998e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 1.77999e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 5.96e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.96e-06
        [meta_shard_fg_expand]: 1.26002e-06
        [get_grad_eliminate_]: 5.83002e-06
        [merge_forward]: 2.22001e-06
        [cell_reuse_recompute_pass]: 8.50006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.152e-05
        [j_node_and_user_rematch]: 9.85002e-06
        [meta_fg_expand]: 1.81e-06
        [replace_old_param]: 8.68001e-06
        [inline_without_move]: 5.74e-06
        [renormalize]: 0.00056581
        [add_forward_monad_depend]: 4.84e-06
        [auto_monad_grad]: 1.10001e-06
        [auto_monad_eliminator]: 9.90002e-06
        [cse]: 1.612e-05
        [replace_applicator]: 1.002e-05
    [Cycle 2]: 0.00031456, [27]
        [switch_simplify]: 5.74e-06
        [loop_unroll]: 5.20999e-06
        [a_1]: 9.156e-05
        [with_stream_mark]: 9.19e-06
        [recompute_prepare]: 5.14e-06
        [updatestate_depend_eliminate]: 2.68998e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.79998e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 5.19e-06
        [updatestate_useless_node_eliminater]: 4.99003e-06
        [accelerated_algorithm]: 4.89998e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 5.05001e-06
        [merge_forward]: 2.98e-06
        [cell_reuse_recompute_pass]: 1.65001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.238e-05
        [j_node_and_user_rematch]: 7.48e-06
        [meta_fg_expand]: 1.55999e-06
        [replace_old_param]: 6.73e-06
        [inline_without_move]: 4.72998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 6.60017e-07
        [auto_monad_eliminator]: 4.75001e-06
        [cse]: 9.97001e-06
        [replace_applicator]: 4.74998e-06
[py_interpret_to_execute_after_opt_a]: 8.31002e-06
[rewriter_after_opt_a]: 2.454e-05
[convert_after_rewriter]: 2.952e-05
[order_py_execute_after_rewriter]: 5.09e-06
[mutable_eliminate]: 0.000585
[jit_opt_b]: 4.972e-05, [1]
    [Cycle 1]: 4.361e-05, [2]
        [frontend_op_eliminate]: 1.706e-05
        [inline_after_opt_a]: 1.528e-05
[cconv]: 1.693e-05
[loop_unroll]: 0.00045825
[jit_opt_after_cconv]: 0.00013758, [1]
    [Cycle 1]: 0.00013115, [11]
        [c_1]: 2.313e-05
        [parameter_eliminate]: 1.96998e-06
        [updatestate_depend_eliminate]: 4.85001e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.26998e-06
        [cse]: 1.63e-05
        [call_graph_tuple_transform]: 1.654e-05
        [tuple_list_get_item_eliminator]: 5.29998e-06
        [none_parameter_eliminate]: 9.99979e-07
        [renormalize]: 7.00005e-07
        [switch_simplify]: 5.81998e-06
[remove_dup_value]: 8.62998e-06
[partial_unused_args_eliminate]: 1.63002e-06
[environ_conv]: 3.93999e-06
[add_recomputation]: 3.155e-05
[cse_after_recomputation]: 2.11e-05, [1]
    [Cycle 1]: 1.529e-05, [1]
        [cse]: 8.88002e-06
[auto_monad_reorder]: 1.021e-05
[get_jit_bprop_graph]: 1.40999e-06
[rewriter_after_jit_bprop_graph]: 5.04e-06
[opt_after_jit_grad]: 0.00045549
[symbol_engine_optimizer]: 8.86e-05, [1]
    [Cycle 1]: 8.236e-05, [6]
        [build]: 2.69999e-06
        [elim_shapecalc]: 8.82999e-06
        [elim_not_effective]: 1.171e-05
        [opt_reshape]: 5.69999e-06
        [fold_const_symbol]: 8.03999e-06
        [renormalize]: 4.99975e-07
[validate]: 2.451e-05
Sums
    bootstrap                            :     0.000248s :  2.50%
    type_inference                       :     0.005924s : 59.78%
    event_method                         :     0.000020s :  0.20%
    auto_monad                           :     0.000060s :  0.61%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000045s :  0.46%
    rewriter_before_opt_a                :     0.000076s :  0.76%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000090s :  0.91%
    jit_opt_a.loop_unroll                :     0.000039s :  0.40%
    jit_opt_a.a_1                        :     0.000774s :  7.81%
    jit_opt_a.with_stream_mark           :     0.000018s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000566s :  5.71%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.15%
    jit_opt_a.cse                        :     0.000026s :  0.26%
    jit_opt_a.replace_applicator         :     0.000015s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.08%
    rewriter_after_opt_a                 :     0.000025s :  0.25%
    convert_after_rewriter               :     0.000030s :  0.30%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000585s :  5.90%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.15%
    cconv                                :     0.000017s :  0.17%
    loop_unroll                          :     0.000458s :  4.62%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000009s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000032s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000010s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000455s :  4.60%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000025s :  0.25%


TotalTime = 0.0170118, [33]
[bootstrap]: 0.00026856
[type_inference]: 0.0041425
[event_method]: 1.562e-05
[auto_monad]: 6.012e-05
[graph_reusing]: 4.86002e-06
[pre_auto_parallel]: 2.23002e-06
[py_interpret_to_execute]: 2.896e-05
[rewriter_before_opt_a]: 6.842e-05
[expand_dump_flag]: 3.34001e-06
[jit_opt_a]: 0.00486218, [2]
    [Cycle 1]: 0.00142121, [27]
        [switch_simplify]: 4.961e-05
        [loop_unroll]: 2.825e-05
        [a_1]: 0.00049294
        [with_stream_mark]: 1.776e-05
        [recompute_prepare]: 7.31001e-06
        [updatestate_depend_eliminate]: 3.51001e-06
        [updatestate_assign_eliminate]: 3.55e-06
        [updatestate_loads_eliminate]: 2.69001e-06
        [parameter_eliminate]: 2.05002e-06
        [specialize_transform]: 6.23998e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 5.24998e-06
        [meta_shard_fg_expand]: 2.13002e-06
        [get_grad_eliminate_]: 5.05001e-06
        [merge_forward]: 3.51999e-06
        [cell_reuse_recompute_pass]: 1.07998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.34e-05
        [j_node_and_user_rematch]: 9.44e-06
        [meta_fg_expand]: 2.02001e-06
        [replace_old_param]: 8.25e-06
        [inline_without_move]: 5.17999e-06
        [renormalize]: 0.00053991
        [add_forward_monad_depend]: 5.32001e-06
        [auto_monad_grad]: 2.04999e-06
        [auto_monad_eliminator]: 1.209e-05
        [cse]: 1.856e-05
        [replace_applicator]: 1.13e-05
    [Cycle 2]: 0.00031247, [27]
        [switch_simplify]: 5.52999e-06
        [loop_unroll]: 4.70999e-06
        [a_1]: 8.669e-05
        [with_stream_mark]: 8.97e-06
        [recompute_prepare]: 5.15001e-06
        [updatestate_depend_eliminate]: 2.61e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.32e-06
        [specialize_transform]: 5.07999e-06
        [updatestate_useless_node_eliminater]: 5.04998e-06
        [accelerated_algorithm]: 5.01002e-06
        [meta_shard_fg_expand]: 1.00999e-06
        [get_grad_eliminate_]: 4.50999e-06
        [merge_forward]: 2.39999e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.196e-05
        [j_node_and_user_rematch]: 7.86001e-06
        [meta_fg_expand]: 1.57999e-06
        [replace_old_param]: 6.85998e-06
        [inline_without_move]: 4.55999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 6.79982e-07
        [auto_monad_eliminator]: 5.14e-06
        [cse]: 9.72999e-06
        [replace_applicator]: 5.07e-06
[py_interpret_to_execute_after_opt_a]: 8.11002e-06
[rewriter_after_opt_a]: 3.037e-05
[convert_after_rewriter]: 6.53998e-06
[order_py_execute_after_rewriter]: 4.47e-06
[mutable_eliminate]: 0.00048188
[jit_opt_b]: 4.715e-05, [1]
    [Cycle 1]: 4.038e-05, [2]
        [frontend_op_eliminate]: 1.511e-05
        [inline_after_opt_a]: 1.433e-05
[cconv]: 2.415e-05
[loop_unroll]: 0.00041386
[jit_opt_after_cconv]: 0.00013164, [1]
    [Cycle 1]: 0.00012552, [11]
        [c_1]: 1.987e-05
        [parameter_eliminate]: 2.83e-06
        [updatestate_depend_eliminate]: 4.89e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [cse]: 1.647e-05
        [call_graph_tuple_transform]: 1.719e-05
        [tuple_list_get_item_eliminator]: 5.10001e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 2.89991e-07
        [switch_simplify]: 5.21002e-06
[remove_dup_value]: 1.259e-05
[partial_unused_args_eliminate]: 2.21998e-06
[environ_conv]: 5.00999e-06
[add_recomputation]: 4.371e-05
[cse_after_recomputation]: 1.957e-05, [1]
    [Cycle 1]: 1.45e-05, [1]
        [cse]: 8.87999e-06
[auto_monad_reorder]: 1.374e-05
[get_jit_bprop_graph]: 1.85001e-06
[rewriter_after_jit_bprop_graph]: 5.02999e-06
[opt_after_jit_grad]: 0.00050109
[symbol_engine_optimizer]: 6.957e-05, [1]
    [Cycle 1]: 6.288e-05, [6]
        [build]: 2.95002e-06
        [elim_shap
Time group info:
------[substitution.]   0.000231    26
 0.60% :     0.000001s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 1.73% :     0.000004s :      3: substitution.graph_param_transform
85.12% :     0.000197s :      6: substitution.inline
 1.34% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.93% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.17% :     0.000003s :      2: substitution.replace_old_param
 3.52% :     0.000008s :      1: substitution.switch_simplify
 4.15% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005877     2
82.65% :     0.004857s :      1: type_inference.infer
17.35% :     0.001020s :      1: type_inference.specialize
------[replace.]   0.000121     9
57.56% :     0.000069s :      6: replace.inline
15.53% :     0.000019s :      1: replace.switch_simplify
26.91% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000209     9
92.57% :     0.000194s :      6: match.inline
 3.49% :     0.000007s :      1: match.switch_simplify
 3.93% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   801
 1.61% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.04% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000002s :     12: predicate.addn_check_dump
 1.51% :     0.000002s :     12: predicate.addn_zero_filter
 2.02% :     0.000003s :     12: predicate.arithmetic_simplify
 1.32% :     0.000002s :     12: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.18% :     0.000002s :     12: predicate.compare_switch_simplify
 1.35% :     0.000002s :     12: predicate.depend_value_elim
 1.15% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.39% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.61% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     12: predicate.environ_get_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.63% :     0.000007s :     26: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     17: predicate.load_eliminater
 1.18% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.39% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.93% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000002s :     12: predicate.merge_addn
 1.17% :     0.000002s :     12: predicate.minmaximum_grad
 1.28% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     17: predicate.partial_eliminate
 1.25% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     12: predicate.reduce_eliminate
 1.56% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     20: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_decalc]: 7.10998e-06
        [elim_not_effective]: 1.177e-05
        [opt_reshape]: 5.52999e-06
        [fold_const_symbol]: 8.42e-06
        [renormalize]: 7.00005e-07
[validate]: 3.252e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00551271
[execute]: 7.81001e-06
Sums
    bootstrap                            :     0.000269s :  2.02%
    type_inference                       :     0.004142s : 31.11%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000060s :  0.45%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.22%
    rewriter_before_opt_a                :     0.000068s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000055s :  0.41%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000580s :  4.35%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000540s :  4.05%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000028s :  0.21%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000030s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000482s :  3.62%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000414s :  3.11%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value           efer_inline
 1.55% :     0.000002s :     12: predicate.reshape_eliminate
 1.34% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.59% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.44% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 1.18% :     0.000002s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.00% :     0.000004s :     20: predicate.switch_defer_inline
 2.49% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.30% :     0.000011s :     65: predicate.switch_simplify
 1.28% :     0.000002s :     12: predicate.tile_eliminate
 1.42% :     0.000002s :     12: predicate.transpose_eliminate
 1.55% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.24% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     12: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000760    15
46.15% :     0.000351s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.45% :     0.000079s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.40% :     0.000330s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015229    72
 0.23% :     0.000034s :      1: add_recomputation
 0.42% :     0.000064s :      1: auto_monad
 0.08% :     0.000013s :      1: auto_monad_reorder
 1.70% :     0.000259s :      1: bootstrap
 0.13% :     0.000020s :      1: cconv
 0.21% :     0.000032s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
33.32% :     0.005074s :      1: jit_opt_a
 0.92% :     0.000140s :      1: jit_opt_after_cconv
 0.34% :     0.000052s :      1: jit_opt_b
 3.06% :     0.000466s :      1: loop_unroll
 3.91% :     0.000595s :      1: mutable_eliminate
 6.68% :     0.001017s :     26: opt.transform.jit_opt_a
 0.31% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.04% :     0.000463s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.32% :     0.000048s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000011s :      1: remove_dup_value
 1.92% :     0.000292s :      1: renormalize.infer
 1.76% :     0.000268s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000027s :      1: rewriter_after_opt_a
 0.51% :     0.000078s :      1: rewriter_before_opt_a
          :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000044s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000501s :  3.76%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000033s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005513s : 41.40%
    execute                              :     0.000008s :  0.06%

 0.60% :     0.000091s :      1: symbol_engine_optimizer
38.97% :     0.005935s :      1: type_inference


Time group info:
------[substitution.]   0.000180    23
 0.94% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.84% :     0.000005s :      3: substitution.graph_param_transform
83.21% :     0.000149s :      4: substitution.inline
 1.89% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.38% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000003s :      2: substitution.replace_old_param
 6.16% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004083     2
81.66% :     0.003334s :      1: type_inference.infer
18.34% :     0.000749s :      1: type_inference.specialize
------[replace.]   0.000061     6
65.45% :     0.000040s :      4: replace.inline
34.55% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000157     6
93.62% :     0.000147s :      4: match.inline
 6.38% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.32% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.23% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 2.07% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.69% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000001s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.08% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 6.45% :     0.000007s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.07% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.11% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.32% :     0.000001s :      3: predicate.mutable_eliminate
 0.44% :     0.000000s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 1.57% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.38% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.52% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.52% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.20% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000008s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.88% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000512    12
47.80% :     0.000245s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.46% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.73% :     0.000234s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018428    76
 0.25% :     0.000046s :      1: add_recomputation
 0.34% :     0.000063s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.52% :     0.000280s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.07% :     0.000013s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.40% :     0.004865s :      1: jit_opt_a
 0.73% :     0.000134s :      1: jit_opt_after_cconv
 0.27% :     0.000050s :      1: jit_opt_b
 2.29% :     0.000421s :      1: loop_unroll
 2.66% :     0.000490s :      1: mutable_eliminate
 4.22% :     0.000778s :     26: opt.transform.jit_opt_a
 0.24% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.76% :     0.000509s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000031s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.51% :     0.000279s :      1: renormalize.infer
 1.37% :     0.000253s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000033s :      1: rewriter_after_opt_a
 0.39% :     0.000071s :      1: rewriter_before_opt_a
 0.39% :     0.000072s :      1: symbol_engine_optimizer
29.99% :     0.005526s :      1: task_emit
22.54% :     0.004154s :      1: type_inference
 0.28% :     0.000052s :      1: validate


TotalTime = 0.0162573, [33]
[bootstrap]: 0.00024135
[type_inference]: 0.00388862
[event_method]: 1.516e-05
[auto_monad]: 5.232e-05
[graph_reusing]: 5.05999e-06
[pre_auto_parallel]: 2.14e-06
[py_interpret_to_execute]: 2.619e-05
[rewriter_before_opt_a]: 6.337e-05
[expand_dump_flag]: 2.58e-06
[jit_opt_a]: 0.00465976, [2]
    [Cycle 1]: 0.00138295, [27]
        [switch_simplify]: 5.874e-05
        [loop_unroll]: 2.826e-05
        [a_1]: 0.00048574
        [with_stream_mark]: 1.33e-05
        [recompute_prepare]: 7.1e-06
        [updatestate_depend_eliminate]: 2.99999e-06
        [updatestate_assign_eliminate]: 9.05001e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [parameter_eliminate]: 1.39998e-06
        [specialize_transform]: 6.23e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 5.33002e-06
        [meta_shard_fg_expand]: 1.72999e-06
        [get_grad_eliminate_]: 5.08002e-06
        [merge_forward]: 2.73e-06
        [cell_reuse_recompute_pass]: 8.99978e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.357e-05
        [j_node_and_user_rematch]: 8.57e-06
        [meta_fg_expand]: 1.99999e-06
        [replace_old_param]: 8.28999e-06
        [inline_without_move]: 5.01002e-06
        [renormalize]: 0.00050868
        [add_forward_monad_depend]: 3.83001e-06
        [auto_monad_grad]: 1.45001e-06
        [auto_monad_eliminator]: 1.145e-05
        [cse]: 1.788e-05
        [replace_applicator]: 9.92001e-06
    [Cycle 2]: 0.00030421, [27]
        [switch_simplify]: 5.29e-06
        [loop_unroll]: 4.70001e-06
        [a_1]: 8.609e-05
        [with_stream_mark]: 7.79002e-06
        [recompute_prepare]: 4.74e-06
        [updatestate_depend_eliminate]: 2.35002e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 1.18001e-06
        [specialize_transform]: 4.85999e-06
        [updatestate_useless_node_eliminater]: 4.86997e-06
        [accelerated_algorithm]: 5.11002e-06
        [meta_shard_fg_expand]: 1.11002e-06
        [get_grad_eliminate_]: 4.4e-06
        [merge_forward]: 2.66e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.059e-05
        [j_node_and_user_rematch]: 7.21999e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 6.94001e-06
        [inline_without_move]: 4.65001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 5.00999e-06
        [cse]: 8.92e-06
        [replace_applicator]: 6.92002e-06
[py_interpret_to_execute_after_opt_a]: 8.03001e-06
[rewriter_after_opt_a]: 2.919e-05
[convert_after_rewriter]: 6.34001e-06
[order_py_execute_after_rewriter]: 4.42e-06
[mutable_eliminate]: 0.00049499
[jit_opt_b]: 4.787e-05, [1]
    [Cycle 1]: 4.111e-05, [2]
        [frontend_op_eliminate]: 1.527e-05
        [inline_after_opt_a]: 1.463e-05
[cconv]: 2.207e-05
[loop_unroll]: 0.00041882
[jit_opt_after_cconv]: 0.00012278, [1]
    [Cycle 1]: 0.00011712, [11]
        [c_1]: 1.863e-05
        [parameter_eliminate]: 1.56002e-06
        [updatestate_depend_eliminate]: 4.00998e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [cse]: 1.29e-05
        [call_graph_tuple_transform]: 1.59e-05
        [tuple_list_get_item_eliminator]: 4.95999e-06
        [none_parameter_eliminate]: 1.27e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 5.09998e-06
[remove_dup_value]: 1.198e-05
[partial_unused_args_eliminate]: 2.22999e-06
[environ_conv]: 4.80999e-06
[add_recomputation]: 4.186e-05
[cse_after_recomputation]: 1.948e-05, [1]
    [Cycle 1]: 1.415e-05, [1]
        [cse]: 8.2e-06
[auto_monad_reorder]: 1.288e-05
[get_jit_bprop_graph]: 1.56002e-06
[rewriter_after_jit_bprop_graph]: 4.67998e-06
[opt_after_jit_grad]: 0.0004605
[symbol_engine_optimizer]: 7.272e-05, [1]
    [Cycle 1]: 6.623e-05, [6]
        [build]: 3.33e-06
        [elim_shapecalc]: 7.2e-06
        [elim_not_effective]: 1.314e-05
        [opt_reshape]: 5.17e-06
        [fold_const_symbol]: 8.75001e-06
        [renormalize]: 5.50004e-07
[validate]: 3.132e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00531275
[execute]: 5.42001e-06
Sums
    bootstrap                            :     0.000241s :  1.89%
    type_inference                       :     0.003889s : 30.53%
    event_method                         :     0.000015s :  0.12%
    auto_monad                           :     0.000052s :  0.41%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.21%
    rewriter_before_opt_a                :     0.000063s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.50%
    jit_opt_a.loop_unroll                :     0.000033s :  0.26%
    jit_opt_a.a_1                        :     0.000572s :  4.49%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.09%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000509s :  3.99%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.13%
    jit_opt_a.cse                        :     0.000027s :  0.21%
    jit_opt_a.replace_applicator         :     0.000017s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000029s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000495s :  3.89%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000022s :  0.17%
    loop_unroll                          :     0.000419s :  3.29%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000013s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000042s :  0.33%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000013s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000461s :  3.62%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000031s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005313s : 41.71%
    execute                              :     0.000005s :  0.04%


Time group info:
------[substitution.]   0.000178    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000001s :      2: substitution.fold_const_symbol
 2.85% :     0.000005s :      3: substitution.graph_param_transform
78.61% :     0.000140s :      4: substitution.inline
 1.66% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.37% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.84% :     0.000003s :      2: substitution.replace_old_param
10.89% :     0.000019s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003842     2
81.70% :     0.003138s :      1: type_inference.infer
18.30% :     0.000703s :      1: type_inference.specialize
------[replace.]   0.000059     6
64.86% :     0.000038s :      4: replace.inline
35.14% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000156     6
88.33% :     0.000138s :      4: match.inline
11.67% :     0.000018s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.27% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 1.68% :     0.000002s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.19% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 4.87% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.34% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 0.99% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.67% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.58% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.70% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 1.97% :     0.000002s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.32% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.63% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.08% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.20% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
16.16% :     0.000019s :     53: predicate.switch_simplify
 1.08% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.27% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.03% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.42% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000480    12
47.70% :     0.000229s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.85% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.45% :     0.000223s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017648    76
 0.25% :     0.000045s :      1: add_recomputation
 0.31% :     0.000055s :      1: auto_monad
 0.09% :     0.000015s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.43% :     0.000252s :      1: bootstrap
 0.14% :     0.000025s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.42% :     0.004662s :      1: jit_opt_a
 0.71% :     0.000125s :      1: jit_opt_after_cconv
 0.29% :     0.000050s :      1: jit_opt_b
 2.41% :     0.000425s :      1: loop_unroll
 2.85% :     0.000502s :      1: mutable_eliminate
 4.41% :     0.000779s :     26: opt.transform.jit_opt_a
 0.24% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.05% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.66% :     0.000469s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000029s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 1.53% :     0.000269s :      1: renormalize.infer
 1.32% :     0.000233s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000032s :      1: rewriter_after_opt_a
 0.37% :     0.000066s :      1: rewriter_before_opt_a
 0.43% :     0.000075s :      1: symbol_engine_optimizer
30.15% :     0.005322s :      1: task_emit
22.09% :     0.003899s :      1: type_inference
 0.28% :     0.000049s :      1: validate


TotalTime = 0.0161424, [33]
[bootstrap]: 0.00027216
[type_inference]: 0.00399475
[event_method]: 1.444e-05
[auto_monad]: 4.121e-05
[graph_reusing]: 3.28e-06
[pre_auto_parallel]: 1.60001e-06
[py_interpret_to_execute]: 3.537e-05
[rewriter_before_opt_a]: 6.845e-05
[expand_dump_flag]: 2.38998e-06
[jit_opt_a]: 0.00464821, [2]
    [Cycle 1]: 0.00131691, [27]
        [switch_simplify]: 4.309e-05
        [loop_unroll]: 4.613e-05
        [a_1]: 0.00045849
        [with_stream_mark]: 1.188e-05
        [recompute_prepare]: 6.84999e-06
        [updatestate_depend_eliminate]: 3.06001e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.26997e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.08002e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.72e-06
        [merge_forward]: 2.40002e-06
        [cell_reuse_recompute_pass]: 7.80012e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.048e-05
        [j_node_and_user_rematch]: 8.43001e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 7.85998e-06
        [inline_without_move]: 4.89e-06
        [renormalize]: 0.0004779
        [add_forward_monad_depend]: 3.73001e-06
        [auto_monad_grad]: 1.20999e-06
        [auto_monad_eliminator]: 9.72999e-06
        [cse]: 1.818e-05
        [replace_applicator]: 1.289e-05
    [Cycle 2]: 0.00033, [27]
        [switch_simplify]: 6.20002e-06
        [loop_unroll]: 5.58002e-06
        [a_1]: 9.459e-05
        [with_stream_mark]: 8.28001e-06
        [recompute_prepare]: 5.17e-06
        [updatestate_depend_eliminate]: 2.49001e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 1.40001e-06
        [specialize_transform]: 5.02999e-06
        [updatestate_useless_node_eliminater]: 4.94998e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.07998e-06
        [get_grad_eliminate_]: 4.74e-06
        [merge_forward]: 2.32001e-06
        [cell_reuse_recompute_pass]: 1.55999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.212e-05
        [j_node_and_user_rematch]: 8.05999e-06
        [meta_fg_expand]: 1.45999e-06
        [replace_old_param]: 6.88998e-06
        [inline_without_move]: 4.62998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.33002e-06
        [auto_monad_grad]: 7.99977e-07
        [auto_monad_eliminator]: 6.04999e-06
        [cse]: 1.036e-05
        [replace_applicator]: 4.94998e-06
[py_interpret_to_execute_after_opt_a]: 8.50001e-06
[rewriter_after_opt_a]: 2.619e-05
[convert_after_rewriter]: 5.29998e-06
[order_py_execute_after_rewriter]: 3.83999e-06
[mutable_eliminate]: 0.00049274
[jit_opt_b]: 6.897e-05, [1]
    [Cycle 1]: 4.082e-05, [2]
        [frontend_op_eliminate]: 1.578e-05
        [inline_after_opt_a]: 1.409e-05
[cconv]: 1.66e-05
[loop_unroll]: 0.00043512
[jit_opt_after_cconv]: 0.00012999, [1]
    [Cycle 1]: 0.00012391, [11]
        [c_1]: 1.983e-05
        [parameter_eliminate]: 2.94999e-06
        [updatestate_depend_eliminate]: 5.25999e-06
        [updatestate_assign_eliminate]: 2.31998e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [cse]: 1.584e-05
        [call_graph_tuple_transform]: 1.512e-05
        [tuple_list_get_item_eliminator]: 5.20999e-06
        [none_parameter_eliminate]: 9.30013e-07
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.10001e-06
[remove_dup_value]: 6.56999e-06
[partial_unused_args_eliminate]: 1.43002e-06
[environ_conv]: 3.61999e-06
[add_recomputation]: 3.058e-05
[cse_after_recomputation]: 1.954e-05, [1]
    [Cycle 1]: 1.45e-05, [1]
        [cse]: 8.62e-06
[auto_monad_reorder]: 8.95001e-06
[get_jit_bprop_graph]: 1.43002e-06
[rewriter_after_jit_bprop_graph]: 4.57998e-06
[opt_after_jit_grad]: 0.0004616
[symbol_engine_optimizer]: 7.04e-05, [1]
    [Cycle 1]: 6.398e-05, [6]
        [build]: 2.66e-06
        [elim_shapecalc]: 7.16001e-06
        [elim_not_effective]: 1.29e-05
        [opt_reshape]: 5.30999e-06
        [fold_const_symbol]: 7.85e-06
        [renormalize]: 4.39992e-07
[validate]: 2.441e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00506625
[execute]: 7.23e-06
Sums
    bootstrap                            :     0.000272s :  2.17%
    type_inference                       :     0.003995s : 31.82%
    event_method                         :     0.000014s :  0.12%
    auto_monad                           :     0.000041s :  0.33%
    graph_reusing                        :     0.000003s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000035s :  0.28%
    rewriter_before_opt_a                :     0.000068s :  0.55%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000049s :  0.39%
    jit_opt_a.loop_unroll                :     0.000052s :  0.41%
    jit_opt_a.a_1                        :     0.000553s :  4.41%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000478s :  3.81%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.13%
    jit_opt_a.cse                        :     0.000029s :  0.23%
    jit_opt_a.replace_applicator         :     0.000018s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000026s :  0.21%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000493s :  3.92%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000017s :  0.13%
    loop_unroll                          :     0.000435s :  3.47%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000007s :  0.05%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000031s :  0.24%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000009s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000462s :  3.68%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000024s :  0.19%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005066s : 40.35%
    execute                              :     0.000007s :  0.06%


Time group info:
------[substitution.]   0.000148    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.24% :     0.000003s :      3: substitution.graph_param_transform
83.66% :     0.000124s :      4: substitution.inline
 1.93% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.64% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.39% :     0.000004s :      2: substitution.replace_old_param
 5.37% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003946     2
80.62% :     0.003182s :      1: type_inference.infer
19.38% :     0.000765s :      1: type_inference.specialize
------[replace.]   0.000059     6
67.21% :     0.000040s :      4: replace.inline
32.79% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000128     6
94.66% :     0.000122s :      4: match.inline
 5.34% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.57% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 1.84% :     0.000002s :     10: predicate.arithmetic_simplify
 1.37% :     0.000002s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.44% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.00% :     0.000005s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.87% :     0.000002s :     15: predicate.load_eliminater
 1.67% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.28% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.02% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.63% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000002s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.33% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 1.12% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.49% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.72% :     0.000008s :     53: predicate.switch_simplify
 1.44% :     0.000002s :     10: predicate.tile_eliminate
 1.33% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.38% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000477    12
47.62% :     0.000227s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.43% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.95% :     0.000224s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017492    76
 0.19% :     0.000033s :      1: add_recomputation
 0.25% :     0.000044s :      1: auto_monad
 0.07% :     0.000011s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.62% :     0.000284s :      1: bootstrap
 0.11% :     0.000019s :      1: cconv
 0.04% :     0.000007s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.10% :     0.000018s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000005s :      1: graph_reusing
26.59% :     0.004651s :      1: jit_opt_a
 0.76% :     0.000133s :      1: jit_opt_after_cconv
 0.41% :     0.000072s :      1: jit_opt_b
 2.53% :     0.000443s :      1: loop_unroll
 2.87% :     0.000501s :      1: mutable_eliminate
 4.37% :     0.000764s :     26: opt.transform.jit_opt_a
 0.24% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.68% :     0.000469s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.22% :     0.000038s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000009s :      1: remove_dup_value
 1.37% :     0.000239s :      1: renormalize.infer
 1.33% :     0.000233s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000029s :      1: rewriter_after_opt_a
 0.41% :     0.000071s :      1: rewriter_before_opt_a
 0.42% :     0.000073s :      1: symbol_engine_optimizer
29.01% :     0.005075s :      1: task_emit
22.90% :     0.004005s :      1: type_inference
 0.23% :     0.000040s :      1: validate


TotalTime = 0.0191264, [33]
[bootstrap]: 0.00025406
[type_inference]: 0.0042
[event_method]: 1.629e-05
[auto_monad]: 6.368e-05
[graph_reusing]: 5.59e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 3.31e-05
[rewriter_before_opt_a]: 7.237e-05
[expand_dump_flag]: 3.28e-06
[jit_opt_a]: 0.00529685, [2]
    [Cycle 1]: 0.00149078, [27]
        [switch_simplify]: 5.085e-05
        [loop_unroll]: 2.875e-05
        [a_1]: 0.0005236
        [with_stream_mark]: 1.727e-05
        [recompute_prepare]: 7.18e-06
        [updatestate_depend_eliminate]: 3.73001e-06
        [updatestate_assign_eliminate]: 3.5e-06
        [updatestate_loads_eliminate]: 2.52001e-06
        [parameter_eliminate]: 2.32001e-06
        [specialize_transform]: 5.60001e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 1.89999e-06
        [get_grad_eliminate_]: 5.05999e-06
        [merge_forward]: 3.55003e-06
        [cell_reuse_recompute_pass]: 1.06002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.376e-05
        [j_node_and_user_rematch]: 8.59e-06
        [meta_fg_expand]: 2.09999e-06
        [replace_old_param]: 7.88999e-06
        [inline_without_move]: 5.10001e-06
        [renormalize]: 0.00055393
        [add_forward_monad_depend]: 5.75001e-06
        [auto_monad_grad]: 2.31e-06
        [auto_monad_eliminator]: 1.329e-05
        [cse]: 2.765e-05
        [replace_applicator]: 1.129e-05
    [Cycle 2]: 0.00035072, [27]
        [switch_simplify]: 5.71e-06
        [loop_unroll]: 4.68999e-06
        [a_1]: 0.00010083
        [with_stream_mark]: 9.25001e-06
        [recompute_prepare]: 5.32001e-06
        [updatestate_depend_eliminate]: 2.49999e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 5.02999e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.01997e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.47001e-06
        [cell_reuse_recompute_pass]: 2.13002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.677e-05
        [j_node_and_user_rematch]: 7.87e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 7.77e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 6.60017e-07
        [auto_monad_eliminator]: 5.77999e-06
        [cse]: 1.16e-05
        [replace_applicator]: 8.38999e-06
[py_interpret_to_execute_after_opt_a]: 9.40001e-06
[rewriter_after_opt_a]: 3.509e-05
[convert_after_rewriter]: 7.4e-06
[order_py_execute_after_rewriter]: 4.77e-06
[mutable_eliminate]: 0.00053343
[jit_opt_b]: 4.839e-05, [1]
    [Cycle 1]: 4.164e-05, [2]
        [frontend_op_eliminate]: 1.569e-05
        [inline_after_opt_a]: 1.477e-05
[cconv]: 2.639e-05
[loop_unroll]: 0.00043126
[jit_opt_after_cconv]: 0.00014563, [1]
    [Cycle 1]: 0.00013878, [11]
        [c_1]: 2.257e-05
        [parameter_eliminate]: 2.63e-06
        [updatestate_depend_eliminate]: 5.54e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.13998e-06
        [cse]: 1.884e-05
        [call_graph_tuple_transform]: 1.804e-05
        [tuple_list_get_item_eliminator]: 5.82001e-06
        [none_parameter_eliminate]: 1.48002e-06
        [renormalize]: 5.10016e-07
        [switch_simplify]: 5.30001e-06
[remove_dup_value]: 1.361e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 5.20999e-06
[add_recomputation]: 5.004e-05
[cse_after_recomputation]: 2.013e-05, [1]
    [Cycle 1]: 1.461e-05, [1]
        [cse]: 8.90001e-06
[auto_monad_reorder]: 1.396e-05
[get_jit_bprop_graph]: 1.61998e-06
[rewriter_after_jit_bprop_graph]: 5.19e-06
[opt_after_jit_grad]: 0.00048194
[symbol_engine_optimizer]: 7.247e-05, [1]
    [Cycle 1]: 6.566e-05, [6]
        [build]: 4.03999e-06
        [elim_shapecalc]: 7.43e-06
        [elim_not_effective]: 1.259e-05
        [opt_reshape]: 5.64e-06
        [fold_const_symbol]: 8.72e-06
        [renormalize]: 3.00002e-07
[validate]: 3.328e-05
[backend_pass]: 8.30012e-07
[task_emit]: 0.00704379
[execute]: 8.59002e-06
Sums
    bootstrap                            :     0.000254s :  1.69%
    type_inference                       :     0.004200s : 27.86%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000064s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.22%
    rewriter_before_opt_a                :     0.000072s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.38%
    jit_opt_a.loop_unroll                :     0.000033s :  0.22%
    jit_opt_a.a_1                        :     0.000624s :  4.14%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000041s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000554s :  3.67%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000039s :  0.26%
    jit_opt_a.replace_applicator         :     0.000020s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000035s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000533s :  3.54%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000026s :  0.18%
    loop_unroll                          :     0.000431s :  2.86%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000482s :  3.20%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007044s : 46.72%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000203    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.71% :     0.000006s :      3: substitution.graph_param_transform
77.36% :     0.000157s :      4: substitution.inline
 1.49% :     0.000003s :      4: substitution.j_node_and_user_rematch
 8.73% :     0.000018s :      4: substitution.remove_not_recompute_node
 1.99% :     0.000004s :      2: substitution.replace_old_param
 6.00% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004140     2
81.23% :     0.003362s :      1: type_inference.infer
18.77% :     0.000777s :      1: type_inference.specialize
------[replace.]   0.000064     6
68.09% :     0.000044s :      4: replace.inline
31.91% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000166     6
93.29% :     0.000155s :      4: match.inline
 6.71% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.28% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.28% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 2.13% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000001s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.90% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.34% :     0.000002s :     10: predicate.environ_get_eliminate
 1.26% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.42% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.70% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.67% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.93% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.67% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.15% :     0.000009s :     53: predicate.switch_simplify
 1.40% :     0.000002s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.62% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.18% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000505    12
47.52% :     0.000240s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.59% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.89% :     0.000237s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020630    76
 0.26% :     0.000053s :      1: add_recomputation
 0.32% :     0.000067s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.30% :     0.000268s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.69% :     0.005300s :      1: jit_opt_a
 0.72% :     0.000148s :      1: jit_opt_after_cconv
 0.25% :     0.000051s :      1: jit_opt_b
 2.14% :     0.000441s :      1: loop_unroll
 2.63% :     0.000542s :      1: mutable_eliminate
 4.08% :     0.000841s :     26: opt.transform.jit_opt_a
 0.23% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.37% :     0.000490s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.35% :     0.000278s :      1: renormalize.infer
 1.30% :     0.000269s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000038s :      1: rewriter_after_opt_a
 0.36% :     0.000075s :      1: rewriter_before_opt_a
 0.36% :     0.000075s :      1: symbol_engine_optimizer
34.20% :     0.007056s :      1: task_emit
20.42% :     0.004213s :      1: type_inference
 0.25% :     0.000052s :      1: validate


TotalTime = 0.0146742, [30]
[bootstrap]: 0.00025566
[type_inference]: 0.00642387
[event_method]: 2.013e-05
[auto_monad]: 7.591e-05
[graph_reusing]: 5.49e-06
[pre_auto_parallel]: 2.73e-06
[py_interpret_to_execute]: 4.983e-05
[rewriter_before_opt_a]: 8.167e-05
[expand_dump_flag]: 2.73e-06
[jit_opt_a]: 0.00565066, [2]
    [Cycle 1]: 0.00198599, [27]
        [switch_simplify]: 0.00010404
        [loop_unroll]: 3.397e-05
        [a_1]: 0.0007782
        [with_stream_mark]: 1.625e-05
        [recompute_prepare]: 7.28e-06
        [updatestate_depend_eliminate]: 3.68e-06
        [updatestate_assign_eliminate]: 2.63e-06
        [updatestate_loads_eliminate]: 2.74001e-06
        [parameter_eliminate]: 1.78002e-06
        [specialize_transform]: 6.07999e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 6.01003e-06
        [meta_shard_fg_expand]: 2.10002e-06
        [get_grad_eliminate_]: 5.57999e-06
        [merge_forward]: 3.54002e-06
        [cell_reuse_recompute_pass]: 1.05999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.387e-05
        [j_node_and_user_rematch]: 1.024e-05
        [meta_fg_expand]: 2.59999e-06
        [replace_old_param]: 8.71002e-06
        [inline_without_move]: 5.39e-06
        [renormalize]: 0.00074452
        [add_forward_monad_depend]: 5.25001e-06
        [auto_monad_grad]: 2.53e-06
        [auto_monad_eliminator]: 1.307e-05
        [cse]: 1.913e-05
        [replace_applicator]: 1.129e-05
    [Cycle 2]: 0.00031016, [27]
        [switch_simplify]: 5.72001e-06
        [loop_unroll]: 4.84e-06
        [a_1]: 9.081e-05
        [with_stream_mark]: 8.10999e-06
        [recompute_prepare]: 5.15999e-06
        [updatestate_depend_eliminate]: 2.49999e-06
        [updatestate_assign_eliminate]: 1.99999e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 5.01997e-06
        [updatestate_useless_node_eliminater]: 5.04998e-06
        [accelerated_algorithm]: 4.80001e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.33001e-06
        [merge_forward]: 2.27999e-06
        [cell_reuse_recompute_pass]: 1.57999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.195e-05
        [j_node_and_user_rematch]: 8.48999e-06
        [meta_fg_expand]: 1.72999e-06
        [replace_old_param]: 7.25e-06
        [inline_without_move]: 4.77e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 4.67e-06
        [cse]: 9.20999e-06
        [replace_applicator]: 4.95001e-06
[py_interpret_to_execute_after_opt_a]: 2.94e-05
[rewriter_after_opt_a]: 3.215e-05
[convert_after_rewriter]: 6.07001e-06
[order_py_execute_after_rewriter]: 4.67e-06
[mutable_eliminate]: 0.00054436
[jit_opt_b]: 4.93e-05, [1]
    [Cycle 1]: 4.293e-05, [2]
        [frontend_op_eliminate]: 1.59e-05
        [inline_after_opt_a]: 1.511e-05
[cconv]: 2.323e-05
[loop_unroll]: 0.00043674
[jit_opt_after_cconv]: 0.00013618, [1]
    [Cycle 1]: 0.00012978, [11]
        [c_1]: 2.125e-05
        [parameter_eliminate]: 2.71999e-06
        [updatestate_depend_eliminate]: 5.34e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.632e-05
        [call_graph_tuple_transform]: 1.734e-05
        [tuple_list_get_item_eliminator]: 5.17e-06
        [none_parameter_eliminate]: 1.41998e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 5.29e-06
[remove_dup_value]: 1.41e-05
[partial_unused_args_eliminate]: 2.11998e-06
[environ_conv]: 4.52e-06
[add_recomputation]: 4.514e-05
[cse_after_recomputation]: 2.03e-05, [1]
    [Cycle 1]: 1.525e-05, [1]
        [cse]: 9.52999e-06
[auto_monad_reorder]: 1.476e-05
[get_jit_bprop_graph]: 1.82001e-06
[rewriter_after_jit_bprop_graph]: 6.02999e-06
[opt_after_jit_grad]: 0.00046747
[symbol_engine_optimizer]: 7.288e-05, [1]
    [Cycle 1]: 6.693e-05, [6]
        [build]: 3.44001e-06
        [elim_shapecalc]: 7.51001e-06
        [elim_not_effective]: 1.374e-05
        [opt_reshape]: 5.71e-06
        [fold_const_symbol]: 8.85999e-06
        [renormalize]: 4.59986e-07
[validate]: 3.079e-05
Sums
    bootstrap                            :     0.000256s :  2.38%
    type_inference                       :     0.006424s : 59.70%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000076s :  0.71%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000050s :  0.46%
    rewriter_before_opt_a                :     0.000082s :  0.76%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000110s :  1.02%
    jit_opt_a.loop_unroll                :     0.000039s :  0.36%
    jit_opt_a.a_1                        :     0.000869s :  8.08%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000745s :  6.92%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.16%
    jit_opt_a.cse                        :     0.000028s :  0.26%
    jit_opt_a.replace_applicator         :     0.000016s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000029s :  0.27%
    rewriter_after_opt_a                 :     0.000032s :  0.30%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000544s :  5.06%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.14%
    cconv                                :     0.000023s :  0.22%
    loop_unroll                          :     0.000437s :  4.06%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000045s :  0.42%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000467s :  4.34%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.29%


Time group info:
------[substitution.]   0.000315    26
 0.65% :     0.000002s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 1.44% :     0.000005s :      3: substitution.graph_param_transform
87.37% :     0.000276s :      6: substitution.inline
 1.19% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.47% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.11% :     0.000004s :      2: substitution.replace_old_param
 2.35% :     0.000007s :      1: substitution.switch_simplify
 3.98% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006370     2
82.08% :     0.005229s :      1: type_inference.infer
17.92% :     0.001141s :      1: type_inference.specialize
------[replace.]   0.000137     9
53.29% :     0.000073s :      6: replace.inline
21.04% :     0.000029s :      1: replace.switch_simplify
25.68% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000290     9
93.82% :     0.000272s :      6: match.inline
 2.29% :     0.000007s :      1: match.switch_simplify
 3.89% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   801
 1.37% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.16% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     12: predicate.addn_check_dump
 1.36% :     0.000002s :     12: predicate.addn_zero_filter
 2.04% :     0.000003s :     12: predicate.arithmetic_simplify
 1.26% :     0.000002s :     12: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     12: predicate.compare_switch_simplify
 1.30% :     0.000002s :     12: predicate.depend_value_elim
 1.14% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     12: predicate.environ_get_eliminate
 1.24% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.81% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.55% :     0.000007s :     26: predicate.inline
 0.77% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.82% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     17: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.28% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.80% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000002s :     12: predicate.merge_addn
 1.22% :     0.000002s :     12: predicate.minmaximum_grad
 1.75% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     17: predicate.partial_eliminate
 1.36% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     12: predicate.reduce_eliminate
 1.62% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     20: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     12: predicate.reshape_eliminate
 1.54% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000002s :     12: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.77% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.52% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.98% :     0.000004s :     20: predicate.switch_defer_inline
 2.44% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.25% :     0.000011s :     65: predicate.switch_simplify
 1.29% :     0.000002s :     12: predicate.tile_eliminate
 1.22% :     0.000002s :     12: predicate.transpose_eliminate
 1.45% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.35% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000002s :     12: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000828    15
43.63% :     0.000361s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.83% :     0.000106s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.55% :     0.000361s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016607    72
 0.29% :     0.000048s :      1: add_recomputation
 0.48% :     0.000079s :      1: auto_monad
 0.10% :     0.000017s :      1: auto_monad_reorder
 1.61% :     0.000267s :      1: bootstrap
 0.16% :     0.000026s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.13% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.14% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
34.04% :     0.005654s :      1: jit_opt_a
 0.84% :     0.000139s :      1: jit_opt_after_cconv
 0.31% :     0.000052s :      1: jit_opt_b
 2.68% :     0.000445s :      1: loop_unroll
 3.33% :     0.000553s :      1: mutable_eliminate
 6.82% :     0.001132s :     26: opt.transform.jit_opt_a
 0.28% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.86% :     0.000475s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.32% :     0.000053s :      1: py_interpret_to_execute
 0.19% :     0.000032s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.34% :     0.000389s :      1: renormalize.infer
 2.10% :     0.000348s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000035s :      1: rewriter_after_opt_a
 0.51% :     0.000085s :      1: rewriter_before_opt_a
 0.45% :     0.000075s :      1: symbol_engine_optimizer
38.75% :     0.006435s :      1: type_inference


TotalTime = 0.0278833, [33]
[bootstrap]: 0.0003664
[type_inference]: 0.00507117
[event_method]: 1.583e-05
[auto_monad]: 6.606e-05
[graph_reusing]: 6.17999e-06
[pre_auto_parallel]: 2.57001e-06
[py_interpret_to_execute]: 3.898e-05
[rewriter_before_opt_a]: 7.44e-05
[expand_dump_flag]: 3.18e-06
[jit_opt_a]: 0.0131898, [2]
    [Cycle 1]: 0.00188679, [27]
        [switch_simplify]: 6.092e-05
        [loop_unroll]: 2.849e-05
        [a_1]: 0.00061401
        [with_stream_mark]: 1.965e-05
        [recompute_prepare]: 7.19001e-06
        [updatestate_depend_eliminate]: 3.31999e-06
        [updatestate_assign_eliminate]: 3.13e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [parameter_eliminate]: 2.08002e-06
        [specialize_transform]: 6.19001e-06
        [updatestate_useless_node_eliminater]: 5.33002e-06
        [accelerated_algorithm]: 5.34e-06
        [meta_shard_fg_expand]: 2.03997e-06
        [get_grad_eliminate_]: 5.49e-06
        [merge_forward]: 3.75e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.451e-05
        [j_node_and_user_rematch]: 9.69e-06
        [meta_fg_expand]: 2.39999e-06
        [replace_old_param]: 9.31e-06
        [inline_without_move]: 5.25999e-06
        [renormalize]: 0.00084255
        [add_forward_monad_depend]: 5.65001e-06
        [auto_monad_grad]: 2.68e-06
        [auto_monad_eliminator]: 1.367e-05
        [cse]: 2.746e-05
        [replace_applicator]: 1.182e-05
    [Cycle 2]: 0.00031559, [27]
        [switch_simplify]: 5.30999e-06
        [loop_unroll]: 4.80001e-06
        [a_1]: 8.852e-05
        [with_stream_mark]: 9.60001e-06
        [recompute_prepare]: 5.05001e-06
        [updatestate_depend_eliminate]: 2.59999e-06
        [updatestate_assign_eliminate]: 1.94999e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 9.40025e-07
        [specialize_transform]: 4.92e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 4.71002e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.45999e-06
        [merge_forward]: 2.26e-06
        [cell_reuse_recompute_pass]: 1.95001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.316e-05
        [j_node_and_user_rematch]: 7.62002e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 7.01001e-06
        [inline_without_move]: 4.55001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 4.72e-06
        [cse]: 1.021e-05
        [replace_applicator]: 4.74e-06
[py_interpret_to_execute_after_opt_a]: 1.1e-05
[rewriter_after_opt_a]: 3.659e-05
[convert_after_rewriter]: 6.94001e-06
[order_py_execute_after_rewriter]: 5.04e-06
[mutable_eliminate]: 0.00053067
[jit_opt_b]: 4.822e-05, [1]
    [Cycle 1]: 4.153e-05, [2]
        [frontend_op_eliminate]: 1.614e-05
        [inline_after_opt_a]: 1.434e-05
[cconv]: 2.533e-05
[loop_unroll]: 0.00042421
[jit_opt_after_cconv]: 0.00014015, [1]
    [Cycle 1]: 0.00013353, [11]
        [c_1]: 2.039e-05
        [parameter_eliminate]: 2.56e-06
        [updatestate_depend_eliminate]: 5.61e-06
        [updatestate_assign_eliminate]: 2.38002e-06
        [updatestate_loads_eliminate]: 1.98002e-06
        [cse]: 1.928e-05
        [call_graph_tuple_transform]: 1.84e-05
        [tuple_list_get_item_eliminator]: 5.35999e-06
        [none_parameter_eliminate]: 1.72001e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 5.56e-06
[remove_dup_value]: 1.488e-05
[partial_unused_args_eliminate]: 2.68e-06
[environ_conv]: 5.63002e-06
[add_recomputation]: 5.058e-05
[cse_after_recomputation]: 2.201e-05, [1]
    [Cycle 1]: 1.665e-05, [1]
        [cse]: 1.047e-05
[auto_monad_reorder]: 1.504e-05
[get_jit_bprop_graph]: 2.14999e-06
[rewriter_after_jit_bprop_graph]: 5.09998e-06
[opt_after_jit_grad]: 0.00046665
[symbol_engine_optimizer]: 7.318e-05, [1]
    [Cycle 1]: 6.678e-05, [6]
        [build]: 3.91999e-06
        [elim_shapecalc]: 7.55e-06
        [elim_not_effective]: 1.329e-05
        [opt_reshape]: 5.37001e-06
        [fold_const_symbol]: 8.30999e-06
        [renormalize]: 5.50004e-07
[validate]: 4.109e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.0069034
[execute]: 9.79e-06
Sums
    bootstrap                            :     0.000366s :  2.25%
    type_inference                       :     0.005071s : 31.14%
    event_method                         :     0.000016s :  0.10%
    auto_monad                           :     0.000066s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.24%
    rewriter_before_opt_a                :     0.000074s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.41%
    jit_opt_a.loop_unroll                :     0.000033s :  0.20%
    jit_opt_a.a_1                        :     0.000703s :  4.31%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000843s :  5.17%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.11%
    jit_opt_a.cse                        :     0.000038s :  0.23%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000531s :  3.26%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000025s :  0.16%
    loop_unroll                          :     0.000424s :  2.61%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000051s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000467s :  2.87%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006903s : 42.40%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000278    23
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 1.96% :     0.000005s :      3: substitution.graph_param_transform
88.07% :     0.000244s :      4: substitution.inline
 1.06% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.77% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.52% :     0.000004s :      2: substitution.replace_old_param
 4.33% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005002     2
80.50% :     0.004027s :      1: type_inference.infer
19.50% :     0.000975s :      1: type_inference.specialize
------[replace.]   0.000073     6
68.33% :     0.000050s :      4: replace.inline
31.67% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000253     6
95.68% :     0.000242s :      4: match.inline
 4.32% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.65% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.83% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 1.93% :     0.000002s :     10: predicate.arithmetic_simplify
 1.32% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.67% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.24% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.49% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.88% :     0.000002s :     15: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.83% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.13% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.53% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.56% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.23% :     0.000008s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.26% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.60% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.53% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000667    12
46.96% :     0.000313s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.79% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.25% :     0.000302s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.029735    76
 0.18% :     0.000053s :      1: add_recomputation
 0.23% :     0.000069s :      1: auto_monad
 0.06% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.28% :     0.000381s :      1: bootstrap
 0.09% :     0.000028s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.07% :     0.000020s :      1: event_method
 0.05% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
44.37% :     0.013195s :      1: jit_opt_a
 0.48% :     0.000143s :      1: jit_opt_after_cconv
 0.17% :     0.000051s :      1: jit_opt_b
 1.45% :     0.000432s :      1: loop_unroll
 1.81% :     0.000540s :      1: mutable_eliminate
 3.07% :     0.000914s :     26: opt.transform.jit_opt_a
 0.16% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000025s :      4: opt.transform.jit_opt_b
 0.04% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.60% :     0.000475s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000042s :      1: py_interpret_to_execute
 0.04% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 1.53% :     0.000454s :      1: renormalize.infer
 1.28% :     0.000380s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000039s :      1: rewriter_after_opt_a
 0.26% :     0.000078s :      1: rewriter_before_opt_a
 0.25% :     0.000076s :      1: symbol_engine_optimizer
23.29% :     0.006925s :      1: task_emit
17.09% :     0.005083s :      1: type_inference
 0.20% :     0.000060s :      1: validate


TotalTime = 0.0189951, [33]
[bootstrap]: 0.00027563
[type_inference]: 0.00422847
[event_method]: 1.662e-05
[auto_monad]: 6.239e-05
[graph_reusing]: 5.46e-06
[pre_auto_parallel]: 2.27001e-06
[py_interpret_to_execute]: 3.154e-05
[rewriter_before_opt_a]: 6.949e-05
[expand_dump_flag]: 2.91e-06
[jit_opt_a]: 0.00565025, [2]
    [Cycle 1]: 0.00155111, [27]
        [switch_simplify]: 5.08e-05
        [loop_unroll]: 2.893e-05
        [a_1]: 0.00049553
        [with_stream_mark]: 1.779e-05
        [recompute_prepare]: 6.79999e-06
        [updatestate_depend_eliminate]: 3.61001e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 2.47001e-06
        [parameter_eliminate]: 1.91998e-06
        [specialize_transform]: 5.67999e-06
        [updatestate_useless_node_eliminater]: 5.14998e-06
        [accelerated_algorithm]: 5.25001e-06
        [meta_shard_fg_expand]: 2.21998e-06
        [get_grad_eliminate_]: 4.82e-06
        [merge_forward]: 3.58e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.369e-05
        [j_node_and_user_rematch]: 8.55001e-06
        [meta_fg_expand]: 2.02999e-06
        [replace_old_param]: 9.00001e-06
        [inline_without_move]: 5.17999e-06
        [renormalize]: 0.00065763
        [add_forward_monad_depend]: 4.94e-06
        [auto_monad_grad]: 2.54999e-06
        [auto_monad_eliminator]: 1.312e-05
        [cse]: 1.933e-05
        [replace_applicator]: 1.165e-05
    [Cycle 2]: 0.00030848, [27]
        [switch_simplify]: 5.63002e-06
        [loop_unroll]: 4.85999e-06
        [a_1]: 8.869e-05
        [with_stream_mark]: 8.43999e-06
        [recompute_prepare]: 5.09e-06
        [updatestate_depend_eliminate]: 2.43998e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.76998e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 4.73001e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 4.50001e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.25999e-06
        [merge_forward]: 2.51e-06
        [cell_reuse_recompute_pass]: 1.46002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.225e-05
        [j_node_and_user_rematch]: 7.19001e-06
        [meta_fg_expand]: 1.72999e-06
        [replace_old_param]: 7.08e-06
        [inline_without_move]: 4.67e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.47999e-06
        [auto_monad_grad]: 9.40025e-07
        [auto_monad_eliminator]: 4.85001e-06
        [cse]: 9.59999e-06
        [replace_applicator]: 4.74e-06
[py_interpret_to_execute_after_opt_a]: 9.47999e-06
[rewriter_after_opt_a]: 3.253e-05
[convert_after_rewriter]: 7.11999e-06
[order_py_execute_after_rewriter]: 5.32999e-06
[mutable_eliminate]: 0.00070951
[jit_opt_b]: 4.781e-05, [1]
    [Cycle 1]: 4.054e-05, [2]
        [frontend_op_eliminate]: 1.539e-05
        [inline_after_opt_a]: 1.389e-05
[cconv]: 2.606e-05
[loop_unroll]: 0.00045362
[jit_opt_after_cconv]: 0.00013565, [1]
    [Cycle 1]: 0.00012926, [11]
        [c_1]: 2.035e-05
        [parameter_eliminate]: 2.62001e-06
        [updatestate_depend_eliminate]: 5.64e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 1.69e-05
        [call_graph_tuple_transform]: 1.843e-05
        [tuple_list_get_item_eliminator]: 5.15999e-06
        [none_parameter_eliminate]: 1.64998e-06
        [renormalize]: 5.19998e-07
        [switch_simplify]: 5.25999e-06
[remove_dup_value]: 1.336e-05
[partial_unused_args_eliminate]: 2.33998e-06
[environ_conv]: 5.59e-06
[add_recomputation]: 4.629e-05
[cse_after_recomputation]: 1.93e-05, [1]
    [Cycle 1]: 1.404e-05, [1]
        [cse]: 8.53001e-06
[auto_monad_reorder]: 1.511e-05
[get_jit_bprop_graph]: 2.00002e-06
[rewriter_after_jit_bprop_graph]: 5.99999e-06
[opt_after_jit_grad]: 0.00046092
[symbol_engine_optimizer]: 7.165e-05, [1]
    [Cycle 1]: 6.512e-05, [6]
        [build]: 3.31999e-06
        [elim_shapecalc]: 7.7e-06
        [elim_not_effective]: 1.249e-05
        [opt_reshape]: 5.27001e-06
        [fold_const_symbol]: 8.77999e-06
        [renormalize]: 5.29981e-07
[validate]: 3.354e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00635195
[execute]: 7.8e-06
Sums
    bootstrap                            :     0.000276s :  1.88%
    type_inference                       :     0.004228s : 28.91%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000062s :  0.43%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.22%
    rewriter_before_opt_a                :     0.000069s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000584s :  3.99%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000658s :  4.50%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000029s :  0.20%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000710s :  4.85%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000026s :  0.18%
    loop_unroll                          :     0.000454s :  3.10%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000046s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000461s :  3.15%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006352s : 43.43%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000182    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.86% :     0.000005s :      3: substitution.graph_param_transform
82.95% :     0.000151s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.34% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.35% :     0.000004s :      2: substitution.replace_old_param
 6.19% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004169     2
81.76% :     0.003408s :      1: type_inference.infer
18.24% :     0.000760s :      1: type_inference.specialize
------[replace.]   0.000060     6
66.72% :     0.000040s :      4: replace.inline
33.28% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000158     6
93.59% :     0.000148s :      4: match.inline
 6.41% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.22% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.54% :     0.000002s :     10: predicate.addn_zero_filter
 1.83% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000001s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.08% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.35% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.07% :     0.000001s :      6: predicate.less_batch_normalization
 1.65% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.20% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 2.08% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.47% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.77% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.86% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.24% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.74% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.87% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.04% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000520    12
48.26% :     0.000251s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.05% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.69% :     0.000238s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020539    76
 0.24% :     0.000049s :      1: add_recomputation
 0.32% :     0.000066s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.40% :     0.000287s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.52% :     0.005653s :      1: jit_opt_a
 0.67% :     0.000139s :      1: jit_opt_after_cconv
 0.25% :     0.000050s :      1: jit_opt_b
 2.25% :     0.000462s :      1: loop_unroll
 3.50% :     0.000720s :      1: mutable_eliminate
 3.81% :     0.000782s :     26: opt.transform.jit_opt_a
 0.22% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.28% :     0.000469s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.89% :     0.000388s :      1: renormalize.infer
 1.28% :     0.000262s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000036s :      1: rewriter_after_opt_a
 0.35% :     0.000072s :      1: rewriter_before_opt_a
 0.36% :     0.000074s :      1: symbol_engine_optimizer
31.00% :     0.006367s :      1: task_emit
20.64% :     0.004240s :      1: type_inference
 0.26% :     0.000053s :      1: validate


TotalTime = 0.0234329, [33]
[bootstrap]: 0.00025484
[type_inference]: 0.00425485
[event_method]: 1.594e-05
[auto_monad]: 6.209e-05
[graph_reusing]: 5.99e-06
[pre_auto_parallel]: 2.37999e-06
[py_interpret_to_execute]: 3.086e-05
[rewriter_before_opt_a]: 0.00011051
[expand_dump_flag]: 3.01001e-06
[jit_opt_a]: 0.00516103, [2]
    [Cycle 1]: 0.0015981, [27]
        [switch_simplify]: 5.625e-05
        [loop_unroll]: 2.978e-05
        [a_1]: 0.00052415
        [with_stream_mark]: 2.147e-05
        [recompute_prepare]: 8.38999e-06
        [updatestate_depend_eliminate]: 3.65e-06
        [updatestate_assign_eliminate]: 3.17002e-06
        [updatestate_loads_eliminate]: 2.70002e-06
        [parameter_eliminate]: 2.09999e-06
        [specialize_transform]: 5.35999e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 5.57999e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 5.09998e-06
        [merge_forward]: 3.36001e-06
        [cell_reuse_recompute_pass]: 2.01998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.51e-05
        [j_node_and_user_rematch]: 8.59e-06
        [meta_fg_expand]: 2.25002e-06
        [replace_old_param]: 8.80999e-06
        [inline_without_move]: 4.95999e-06
        [renormalize]: 0.00061805
        [add_forward_monad_depend]: 8.49002e-06
        [auto_monad_grad]: 2.98e-06
        [auto_monad_eliminator]: 1.513e-05
        [cse]: 2.812e-05
        [replace_applicator]: 1.573e-05
    [Cycle 2]: 0.00031404, [27]
        [switch_simplify]: 5.22e-06
        [loop_unroll]: 4.93001e-06
        [a_1]: 8.752e-05
        [with_stream_mark]: 9.85002e-06
        [recompute_prepare]: 4.83001e-06
        [updatestate_depend_eliminate]: 2.79001e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 1.78002e-06
        [parameter_eliminate]: 9.69972e-07
        [specialize_transform]: 4.80999e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 5.44e-06
        [meta_shard_fg_expand]: 1.45001e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.53e-06
        [cell_reuse_recompute_pass]: 3.4e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.367e-05
        [j_node_and_user_rematch]: 7.46999e-06
        [meta_fg_expand]: 1.40001e-06
        [replace_old_param]: 6.71e-06
        [inline_without_move]: 5.02e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.40999e-06
        [auto_monad_grad]: 6.39993e-07
        [auto_monad_eliminator]: 4.98001e-06
        [cse]: 9.99999e-06
        [replace_applicator]: 5.03002e-06
[py_interpret_to_execute_after_opt_a]: 1.177e-05
[rewriter_after_opt_a]: 3.608e-05
[convert_after_rewriter]: 6.53e-06
[order_py_execute_after_rewriter]: 4.74e-06
[mutable_eliminate]: 0.0006433
[jit_opt_b]: 4.701e-05, [1]
    [Cycle 1]: 3.99e-05, [2]
        [frontend_op_eliminate]: 1.492e-05
        [inline_after_opt_a]: 1.372e-05
[cconv]: 2.477e-05
[loop_unroll]: 0.00042047
[jit_opt_after_cconv]: 0.00013791, [1]
    [Cycle 1]: 0.0001314, [11]
        [c_1]: 2.022e-05
        [parameter_eliminate]: 2.79001e-06
        [updatestate_depend_eliminate]: 5.22999e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [cse]: 1.851e-05
        [call_graph_tuple_transform]: 1.899e-05
        [tuple_list_get_item_eliminator]: 5.21002e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 5.12e-06
[remove_dup_value]: 1.319e-05
[partial_unused_args_eliminate]: 2.32999e-06
[environ_conv]: 4.63001e-06
[add_recomputation]: 4.61e-05
[cse_after_recomputation]: 1.964e-05, [1]
    [Cycle 1]: 1.447e-05, [1]
        [cse]: 8.94003e-06
[auto_monad_reorder]: 1.451e-05
[get_jit_bprop_graph]: 1.82999e-06
[rewriter_after_jit_bprop_graph]: 5.40001e-06
[opt_after_jit_grad]: 0.0004408
[symbol_engine_optimizer]: 6.8e-05, [1]
    [Cycle 1]: 6.207e-05, [6]
        [build]: 3.38999e-06
        [elim_shapecalc]: 6.79001e-06
        [elim_not_effective]: 1.18e-05
        [opt_reshape]: 5.39e-06
        [fold_const_symbol]: 8.07e-06
        [renormalize]: 4.30009e-07
[validate]: 3.4e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.0113578
[execute]: 6.85002e-06
Sums
    bootstrap                            :     0.000255s :  1.30%
    type_inference                       :     0.004255s : 21.73%
    event_method                         :     0.000016s :  0.08%
    auto_monad                           :     0.000062s :  0.32%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000031s :  0.16%
    rewriter_before_opt_a                :     0.000111s :  0.56%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.31%
    jit_opt_a.loop_unroll                :     0.000035s :  0.18%
    jit_opt_a.a_1                        :     0.000612s :  3.12%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000618s :  3.16%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.10%
    jit_opt_a.cse                        :     0.000038s :  0.19%
    jit_opt_a.replace_applicator         :     0.000021s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.06%
    rewriter_after_opt_a                 :     0.000036s :  0.18%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000643s :  3.28%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.07%
    cconv                                :     0.000025s :  0.13%
    loop_unroll                          :     0.000420s :  2.15%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000046s :  0.24%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000441s :  2.25%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.17%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011358s : 57.99%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000206    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.47% :     0.000005s :      3: substitution.graph_param_transform
84.89% :     0.000175s :      4: substitution.inline
 1.37% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.46% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.68% :     0.000003s :      2: substitution.replace_old_param
 5.68% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004193     2
82.05% :     0.003441s :      1: type_inference.infer
17.95% :     0.000753s :      1: type_inference.specialize
------[replace.]   0.000063     6
66.48% :     0.000042s :      4: replace.inline
33.52% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000183     6
94.22% :     0.000173s :      4: match.inline
 5.78% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.21% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.27% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.71% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.15% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.52% :     0.000001s :      3: predicate.graph_param_transform
 5.63% :     0.000006s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.38% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.78% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 2.11% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.37% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.98% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.28% :     0.000001s :     10: predicate.reshape_eliminate
 1.32% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 1.21% :     0.000001s :      6: predicate.specialize_transform
 1.25% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.88% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.89% :     0.000009s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.31% :     0.000002s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000510    12
47.72% :     0.000243s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.44% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.84% :     0.000234s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024977    76
 0.20% :     0.000049s :      1: add_recomputation
 0.26% :     0.000065s :      1: auto_monad
 0.07% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.07% :     0.000266s :      1: bootstrap
 0.11% :     0.000027s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.08% :     0.000021s :      1: event_method
 0.04% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
20.68% :     0.005165s :      1: jit_opt_a
 0.56% :     0.000141s :      1: jit_opt_after_cconv
 0.20% :     0.000050s :      1: jit_opt_b
 1.71% :     0.000428s :      1: loop_unroll
 2.61% :     0.000652s :      1: mutable_eliminate
 3.30% :     0.000823s :     26: opt.transform.jit_opt_a
 0.19% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000023s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000029s :      4: opt.transform.symbol_engine_opt
 1.80% :     0.000449s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.13% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000015s :      1: remove_dup_value
 1.35% :     0.000338s :      1: renormalize.infer
 1.09% :     0.000273s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000039s :      1: rewriter_after_opt_a
 0.46% :     0.000114s :      1: rewriter_before_opt_a
 0.28% :     0.000070s :      1: symbol_engine_optimizer
45.52% :     0.011370s :      1: task_emit
17.09% :     0.004269s :      1: type_inference
 0.21% :     0.000053s :      1: validate


TotalTime = 0.0231243, [30]
[bootstrap]: 0.00028835
[type_inference]: 0.00740681
[event_method]: 2.103e-05
[auto_monad]: 8.489e-05
[graph_reusing]: 5.81998e-06
[pre_auto_parallel]: 2.44999e-06
[py_interpret_to_execute]: 5.804e-05
[rewriter_before_opt_a]: 8.633e-05
[expand_dump_flag]: 3.03e-06
[jit_opt_a]: 0.0126625, [2]
    [Cycle 1]: 0.00823878, [27]
        [switch_simplify]: 0.00015088
        [loop_unroll]: 3.641e-05
        [a_1]: 0.00630629
        [with_stream_mark]: 3.462e-05
        [recompute_prepare]: 1.384e-05
        [updatestate_depend_eliminate]: 4.44002e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 3.04001e-06
        [parameter_eliminate]: 2.70002e-06
        [specialize_transform]: 7.46999e-06
        [updatestate_useless_node_eliminater]: 6.65002e-06
        [accelerated_algorithm]: 8.17e-06
        [meta_shard_fg_expand]: 2.372e-05
        [get_grad_eliminate_]: 6.71e-06
        [merge_forward]: 4.32998e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.736e-05
        [j_node_and_user_rematch]: 1.223e-05
        [meta_fg_expand]: 3.48999e-06
        [replace_old_param]: 1.104e-05
        [inline_without_move]: 6.19001e-06
        [renormalize]: 0.00121735
        [add_forward_monad_depend]: 5.87999e-06
        [auto_monad_grad]: 2.45002e-06
        [auto_monad_eliminator]: 1.433e-05
        [cse]: 3.125e-05
        [replace_applicator]: 1.237e-05
    [Cycle 2]: 0.00031878, [27]
        [switch_simplify]: 6.19999e-06
        [loop_unroll]: 5.22e-06
        [a_1]: 9.231e-05
        [with_stream_mark]: 9.14998e-06
        [recompute_prepare]: 5.03002e-06
        [updatestate_depend_eliminate]: 2.71999e-06
        [updatestate_assign_eliminate]: 2.28998e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 5.09003e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 4.72e-06
        [meta_shard_fg_expand]: 9.99979e-07
        [get_grad_eliminate_]: 4.62e-06
        [merge_forward]: 2.44999e-06
        [cell_reuse_recompute_pass]: 1.38002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.311e-05
        [j_node_and_user_rematch]: 7.93001e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 7.53999e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.00999e-06
        [auto_monad_grad]: 7.29982e-07
        [auto_monad_eliminator]: 5.00999e-06
        [cse]: 9.35001e-06
        [replace_applicator]: 5.29e-06
[py_interpret_to_execute_after_opt_a]: 1.179e-05
[rewriter_after_opt_a]: 3.61e-05
[convert_after_rewriter]: 7.06001e-06
[order_py_execute_after_rewriter]: 5.25001e-06
[mutable_eliminate]: 0.00088837
[jit_opt_b]: 6.644e-05, [1]
    [Cycle 1]: 5.78e-05, [2]
        [frontend_op_eliminate]: 1.86e-05
        [inline_after_opt_a]: 2.65e-05
[cconv]: 2.715e-05
[loop_unroll]: 0.00044063
[jit_opt_after_cconv]: 0.00014795, [1]
    [Cycle 1]: 0.00014089, [11]
        [c_1]: 2.289e-05
        [parameter_eliminate]: 2.61e-06
        [updatestate_depend_eliminate]: 5.25001e-06
        [updatestate_assign_eliminate]: 2.36998e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [cse]: 1.989e-05
        [call_graph_tuple_transform]: 2.008e-05
        [tuple_list_get_item_eliminator]: 5.70001e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 6.14001e-06
[remove_dup_value]: 1.436e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 4.85001e-06
[add_recomputation]: 5.137e-05
[cse_after_recomputation]: 2.211e-05, [1]
    [Cycle 1]: 1.579e-05, [1]
        [cse]: 9.82999e-06
[auto_monad_reorder]: 1.529e-05
[get_jit_bprop_graph]: 2.22001e-06
[rewriter_after_jit_bprop_graph]: 5.59998e-06
[opt_after_jit_grad]: 0.00047307
[symbol_engine_optimizer]: 7.836e-05, [1]
    [Cycle 1]: 7.162e-05, [6]
        [build]: 3.63e-06
        [elim_shapecalc]: 8.30999e-06
        [elim_not_effective]: 1.335e-05
        [opt_reshape]: 6.04999e-06
        [fold_const_symbol]: 8.85001e-06
        [renormalize]: 6.79982e-07
[validate]: 3.714e-05
Sums
    bootstrap                            :     0.000288s :  1.57%
    type_inference                       :     0.007407s : 40.43%
    event_method                         :     0.000021s :  0.11%
    auto_monad                           :     0.000085s :  0.46%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000058s :  0.32%
    rewriter_before_opt_a                :     0.000086s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000157s :  0.86%
    jit_opt_a.loop_unroll                :     0.000042s :  0.23%
    jit_opt_a.a_1                        :     0.006399s : 34.93%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000025s :  0.13%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001217s :  6.65%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.11%
    jit_opt_a.cse                        :     0.000041s :  0.22%
    jit_opt_a.replace_applicator         :     0.000018s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.06%
    rewriter_after_opt_a                 :     0.000036s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000888s :  4.85%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.14%
    cconv                                :     0.000027s :  0.15%
    loop_unroll                          :     0.000441s :  2.41%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000051s :  0.28%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000473s :  2.58%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.20%


Time group info:
------[substitution.]   0.000532    26
 0.43% :     0.000002s :      2: substitution.elim_not_effective
 0.22% :     0.000001s :      2: substitution.fold_const_symbol
 1.15% :     0.000006s :      3: substitution.graph_param_transform
89.04% :     0.000473s :      6: substitution.inline
 0.69% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.00% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.02% :     0.000005s :      2: substitution.replace_old_param
 1.87% :     0.000010s :      1: substitution.switch_simplify
 4.57% :     0.000024s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007340     2
82.35% :     0.006045s :      1: type_inference.infer
17.65% :     0.001295s :      1: type_inference.specialize
------[replace.]   0.005265     9
97.70% :     0.005144s :      6: replace.inline
 0.82% :     0.000043s :      1: replace.switch_simplify
 1.48% :     0.000078s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000501     9
93.75% :     0.000469s :      6: match.inline
 1.84% :     0.000009s :      1: match.switch_simplify
 4.41% :     0.000022s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000277   801
 1.12% :     0.000003s :     12: predicate.accumulaten_eliminater
 0.70% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.54% :     0.000002s :     12: predicate.addn_check_dump
 7.18% :     0.000020s :     12: predicate.addn_zero_filter
 2.96% :     0.000008s :     12: predicate.arithmetic_simplify
 1.22% :     0.000003s :     12: predicate.cast_eliminate
 0.23% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.57% :     0.000002s :     12: predicate.compare_switch_simplify
 6.40% :     0.000018s :     12: predicate.depend_value_elim
 0.60% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.04% :     0.000003s :     12: predicate.dict_get_item_eliminator
 0.66% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.49% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.15% :     0.000000s :      3: predicate.elim_not_effective
 0.29% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.75% :     0.000002s :     12: predicate.environ_add_const_eliminate
 0.69% :     0.000002s :     12: predicate.environ_get_add_eliminate
 0.65% :     0.000002s :     12: predicate.environ_get_depend_swap
 0.86% :     0.000002s :     12: predicate.environ_get_eliminate
 0.63% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      3: predicate.fold_const_symbol
 0.45% :     0.000001s :      6: predicate.get_grad_eliminate
 0.10% :     0.000000s :      3: predicate.graph_param_transform
 3.76% :     0.000010s :     26: predicate.inline
 0.40% :     0.000001s :      6: predicate.inline_without_move
 0.16% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.98% :     0.000003s :      6: predicate.less_batch_normalization
 1.68% :     0.000005s :     14: predicate.list_to_tuple_eliminator_
 6.90% :     0.000019s :     17: predicate.load_eliminater
 0.80% :     0.000002s :      3: predicate.loop_unroll_after_grad
 2.34% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.32% :     0.000004s :     15: predicate.make_slice_get_slice_eliminator
 0.58% :     0.000002s :     12: predicate.merge_addn
 0.67% :     0.000002s :     12: predicate.minmaximum_grad
 0.64% :     0.000002s :      3: predicate.mutable_eliminate
 0.21% :     0.000001s :      3: predicate.opt_reshape
 1.19% :     0.000003s :     17: predicate.partial_eliminate
 1.07% :     0.000003s :     12: predicate.print_const_string_wrapper
 1.81% :     0.000005s :     12: predicate.reduce_eliminate
 1.13% :     0.000003s :     14: predicate.redundant_stop_gradient_eliminater
 0.23% :     0.000001s :      6: predicate.remove_not_recompute_node
 0.77% :     0.000002s :     20: predicate.replace_applicator
 0.30% :     0.000001s :      6: predicate.replace_old_param
 0.13% :     0.000000s :      3: predicate.reset_defer_inline
 1.07% :     0.000003s :     12: predicate.reshape_eliminate
 1.02% :     0.000003s :     12: predicate.row_tensor_add_zeros_like
 0.43% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.93% :     0.000003s :     12: predicate.same_eliminate
 0.23% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.67% :     0.000002s :      6: predicate.special_op_eliminate
 0.49% :     0.000001s :      6: predicate.specialize_transform
 0.80% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 0.87% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.17% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 4.36% :     0.000012s :     20: predicate.switch_defer_inline
15.24% :     0.000042s :     20: predicate.switch_layer_defer_inline
 5.26% :     0.000015s :     65: predicate.switch_simplify
 0.86% :     0.000002s :     12: predicate.tile_eliminate
 0.84% :     0.000002s :     12: predicate.transpose_eliminate
 0.95% :     0.000003s :     12: predicate.tuple_list_convert_item_index_to_positive
 0.99% :     0.000003s :     12: predicate.tuple_list_get_item_depend_reorder
 2.73% :     0.000008s :     20: predicate.tuple_list_get_item_eliminator
 1.39% :     0.000004s :     12: predicate.tuple_list_set_item_eliminator
 1.11% :     0.000003s :     14: predicate.tuple_to_list_eliminator_
 0.80% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 1.34% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000004s :     12: predicate.value_based_eliminate
 0.17% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.27% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001050    15
43.93% :     0.000461s :      6: func_graph_cloner_run.FuncGraphClonerGraph
16.25% :     0.000171s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.82% :     0.000418s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.031140    72
 0.17% :     0.000054s :      1: add_recomputation
 0.28% :     0.000089s :      1: auto_monad
 0.06% :     0.000018s :      1: auto_monad_reorder
 0.98% :     0.000306s :      1: bootstrap
 0.10% :     0.000030s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.08% :     0.000024s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
40.68% :     0.012667s :      1: jit_opt_a
 0.48% :     0.000151s :      1: jit_opt_after_cconv
 0.22% :     0.000069s :      1: jit_opt_b
 1.44% :     0.000448s :      1: loop_unroll
 2.88% :     0.000897s :      1: mutable_eliminate
21.60% :     0.006727s :     26: opt.transform.jit_opt_a
 0.16% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000038s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.55% :     0.000481s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000061s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000017s :      1: remove_dup_value
 2.40% :     0.000748s :      1: renormalize.infer
 1.48% :     0.000461s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000039s :      1: rewriter_after_opt_a
 0.29% :     0.000089s :      1: rewriter_before_opt_a
 0.26% :     0.000081s :      1: symbol_engine_optimizer
23.82% :     0.007417s :      1: type_inference


TotalTime = 0.0174381, [33]
[bootstrap]: 0.00030677
[type_inference]: 0.00422893
[event_method]: 1.566e-05
[auto_monad]: 7.087e-05
[graph_reusing]: 5.27999e-06
[pre_auto_parallel]: 2.88e-06
[py_interpret_to_execute]: 2.785e-05
[rewriter_before_opt_a]: 6.655e-05
[expand_dump_flag]: 3.45e-06
[jit_opt_a]: 0.00507292, [2]
    [Cycle 1]: 0.00138596, [27]
        [switch_simplify]: 4.531e-05
        [loop_unroll]: 2.782e-05
        [a_1]: 0.00048648
        [with_stream_mark]: 1.626e-05
        [recompute_prepare]: 7.03998e-06
        [updatestate_depend_eliminate]: 3.18e-06
        [updatestate_assign_eliminate]: 3.00002e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [parameter_eliminate]: 1.76e-06
        [specialize_transform]: 5.39998e-06
        [updatestate_useless_node_eliminater]: 4.65999e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 2.16e-06
        [get_grad_eliminate_]: 5.01997e-06
        [merge_forward]: 3.25e-06
        [cell_reuse_recompute_pass]: 1.31998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.358e-05
        [j_node_and_user_rematch]: 8.65001e-06
        [meta_fg_expand]: 2.22001e-06
        [replace_old_param]: 7.81001e-06
        [inline_without_move]: 5.35999e-06
        [renormalize]: 0.00051448
        [add_forward_monad_depend]: 8.26002e-06
        [auto_monad_grad]: 2.06998e-06
        [auto_monad_eliminator]: 1.241e-05
        [cse]: 1.972e-05
        [replace_applicator]: 1.08e-05
    [Cycle 2]: 0.00044095, [27]
        [switch_simplify]: 5.31002e-06
        [loop_unroll]: 4.77998e-06
        [a_1]: 0.00020809
        [with_stream_mark]: 9.22999e-06
        [recompute_prepare]: 5.44998e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.94003e-06
        [accelerated_algorithm]: 5.07999e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.65999e-06
        [merge_forward]: 2.10002e-06
        [cell_reuse_recompute_pass]: 1.93002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.238e-05
        [j_node_and_user_rematch]: 7.84002e-06
        [meta_fg_expand]: 1.42e-06
        [replace_old_param]: 1.198e-05
        [inline_without_move]: 4.92999e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 9.29984e-07
        [auto_monad_eliminator]: 6.21e-06
        [cse]: 1.175e-05
        [replace_applicator]: 4.93001e-06
[py_interpret_to_execute_after_opt_a]: 9.86e-06
[rewriter_after_opt_a]: 3.073e-05
[convert_after_rewriter]: 6.29001e-06
[order_py_execute_after_rewriter]: 5.20001e-06
[mutable_eliminate]: 0.00050412
[jit_opt_b]: 4.682e-05, [1]
    [Cycle 1]: 3.966e-05, [2]
        [frontend_op_eliminate]: 1.495e-05
        [inline_after_opt_a]: 1.376e-05
[cconv]: 2.401e-05
[loop_unroll]: 0.00040517
[jit_opt_after_cconv]: 0.00013153, [1]
    [Cycle 1]: 0.00012487, [11]
        [c_1]: 1.954e-05
        [parameter_eliminate]: 2.34999e-06
        [updatestate_depend_eliminate]: 4.94998e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.525e-05
        [call_graph_tuple_transform]: 1.732e-05
        [tuple_list_get_item_eliminator]: 5.20999e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 5.15001e-06
[remove_dup_value]: 1.36e-05
[partial_unused_args_eliminate]: 2.43e-06
[environ_conv]: 5.31002e-06
[add_recomputation]: 4.371e-05
[cse_after_recomputation]: 1.897e-05, [1]
    [Cycle 1]: 1.396e-05, [1]
        [cse]: 8.23999e-06
[auto_monad_reorder]: 1.42e-05
[get_jit_bprop_graph]: 1.77001e-06
[rewriter_after_jit_bprop_graph]: 4.91002e-06
[opt_after_jit_grad]: 0.00043825
[symbol_engine_optimizer]: 7.03e-05, [1]
    [Cycle 1]: 6.393e-05, [6]
        [build]: 3.29001e-06
        [elim_shapecalc]: 7.13e-06
        [elim_not_effective]: 1.189e-05
        [opt_reshape]: 5.69e-06
        [fold_const_symbol]: 8.47998e-06
        [renormalize]: 7.50006e-07
[validate]: 3.064e-05
[backend_pass]: 9.79984e-07
[task_emit]: 0.00564921
[execute]: 7.64002e-06
Sums
    bootstrap                            :     0.000307s :  2.25%
    type_inference                       :     0.004229s : 31.02%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000071s :  0.52%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.20%
    rewriter_before_opt_a                :     0.000067s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000051s :  0.37%
    jit_opt_a.loop_unroll                :     0.000033s :  0.24%
    jit_opt_a.a_1                        :     0.000695s :  5.10%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000020s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000515s :  3.77%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.14%
    jit_opt_a.cse                        :     0.000031s :  0.23%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000031s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000504s :  3.70%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000405s :  2.97%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000044s :  0.32%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000438s :  3.22%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000031s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005649s : 41.44%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000181    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.92% :     0.000002s :      2: substitution.fold_const_symbol
 2.82% :     0.000005s :      3: substitution.graph_param_transform
82.68% :     0.000149s :      4: substitution.inline
 1.70% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.46% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.91% :     0.000003s :      2: substitution.replace_old_param
 6.54% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004172     2
82.57% :     0.003445s :      1: type_inference.infer
17.43% :     0.000727s :      1: type_inference.specialize
------[replace.]   0.000057     6
66.85% :     0.000038s :      4: replace.inline
33.15% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000158     6
93.22% :     0.000147s :      4: match.inline
 6.78% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000108   691
 1.49% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000001s :     10: predicate.addn_zero_filter
 1.72% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     10: predicate.compare_switch_simplify
 1.42% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.23% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.45% :     0.000006s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.37% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.91% :     0.000002s :     15: predicate.load_eliminater
 1.51% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.17% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.40% :     0.000003s :     15: predicate.partial_eliminate
 1.23% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.56% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.60% :     0.000003s :     16: predicate.switch_defer_inline
 2.24% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.79% :     0.000008s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.25% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.77% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000499    12
49.21% :     0.000246s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.96% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.83% :     0.000224s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018947    76
 0.24% :     0.000046s :      1: add_recomputation
 0.39% :     0.000074s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.69% :     0.000320s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.79% :     0.005076s :      1: jit_opt_a
 0.71% :     0.000134s :      1: jit_opt_after_cconv
 0.26% :     0.000049s :      1: jit_opt_b
 2.18% :     0.000413s :      1: loop_unroll
 2.70% :     0.000512s :      1: mutable_eliminate
 4.71% :     0.000892s :     26: opt.transform.jit_opt_a
 0.23% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.35% :     0.000446s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000031s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.41% :     0.000268s :      1: renormalize.infer
 1.26% :     0.000240s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000033s :      1: rewriter_after_opt_a
 0.37% :     0.000069s :      1: rewriter_before_opt_a
 0.38% :     0.000073s :      1: symbol_engine_optimizer
29.88% :     0.005661s :      1: task_emit
22.38% :     0.004241s :      1: type_inference
 0.25% :     0.000048s :      1: validate


TotalTime = 0.0178444, [30]
[bootstrap]: 0.00029745
[type_inference]: 0.00795934
[event_method]: 2.67e-05
[auto_monad]: 9.106e-05
[graph_reusing]: 6.72002e-06
[pre_auto_parallel]: 2.88e-06
[py_interpret_to_execute]: 6.356e-05
[rewriter_before_opt_a]: 0.00010774
[expand_dump_flag]: 3.88001e-06
[jit_opt_a]: 0.00671923, [2]
    [Cycle 1]: 0.00254081, [27]
        [switch_simplify]: 0.00014433
        [loop_unroll]: 3.722e-05
        [a_1]: 0.00087776
        [with_stream_mark]: 2.425e-05
        [recompute_prepare]: 8.54998e-06
        [updatestate_depend_eliminate]: 4.65999e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 2.84999e-06
        [parameter_eliminate]: 2.26998e-06
        [specialize_transform]: 6.66e-06
        [updatestate_useless_node_eliminater]: 5.85002e-06
        [accelerated_algorithm]: 6.57002e-06
        [meta_shard_fg_expand]: 2.76e-06
        [get_grad_eliminate_]: 5.75001e-06
        [merge_forward]: 4.25e-06
        [cell_reuse_recompute_pass]: 1.35001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.654e-05
        [j_node_and_user_rematch]: 1.053e-05
        [meta_fg_expand]: 2.68e-06
        [replace_old_param]: 1.156e-05
        [inline_without_move]: 5.73002e-06
        [renormalize]: 0.00108072
        [add_forward_monad_depend]: 8.27998e-06
        [auto_monad_grad]: 2.76e-06
        [auto_monad_eliminator]: 1.584e-05
        [cse]: 2.981e-05
        [replace_applicator]: 1.446e-05
    [Cycle 2]: 0.0003441, [27]
        [switch_simplify]: 6.41e-06
        [loop_unroll]: 5.57999e-06
        [a_1]: 9.719e-05
        [with_stream_mark]: 1.155e-05
        [recompute_prepare]: 5.69999e-06
        [updatestate_depend_eliminate]: 2.93e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 1.18001e-06
        [specialize_transform]: 5.29e-06
        [updatestate_useless_node_eliminater]: 5.92001e-06
        [accelerated_algorithm]: 5.32001e-06
        [meta_shard_fg_expand]: 1.50001e-06
        [get_grad_eliminate_]: 4.69002e-06
        [merge_forward]: 3.5e-06
        [cell_reuse_recompute_pass]: 1.78002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.696e-05
        [j_node_and_user_rematch]: 8.80999e-06
        [meta_fg_expand]: 1.81e-06
        [replace_old_param]: 8.31002e-06
        [inline_without_move]: 4.97e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 9.5999e-07
        [auto_monad_eliminator]: 5.14003e-06
        [cse]: 1.06e-05
        [replace_applicator]: 5.92001e-06
[py_interpret_to_execute_after_opt_a]: 1.114e-05
[rewriter_after_opt_a]: 3.821e-05
[convert_after_rewriter]: 8.05e-06
[order_py_execute_after_rewriter]: 4.90001e-06
[mutable_eliminate]: 0.00075403
[jit_opt_b]: 6.023e-05, [1]
    [Cycle 1]: 5.173e-05, [2]
        [frontend_op_eliminate]: 2.021e-05
        [inline_after_opt_a]: 1.758e-05
[cconv]: 2.953e-05
[loop_unroll]: 0.0005423
[jit_opt_after_cconv]: 0.00016411, [1]
    [Cycle 1]: 0.0001551, [11]
        [c_1]: 2.584e-05
        [parameter_eliminate]: 4.42e-06
        [updatestate_depend_eliminate]: 7.03998e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [cse]: 2.576e-05
        [call_graph_tuple_transform]: 2.143e-05
        [tuple_list_get_item_eliminator]: 5.35999e-06
        [none_parameter_eliminate]: 1.67001e-06
        [renormalize]: 7.39994e-07
        [switch_simplify]: 6.14999e-06
[remove_dup_value]: 1.523e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 5.27001e-06
[add_recomputation]: 5.002e-05
[cse_after_recomputation]: 2.471e-05, [1]
    [Cycle 1]: 1.735e-05, [1]
        [cse]: 1.108e-05
[auto_monad_reorder]: 1.583e-05
[get_jit_bprop_graph]: 2.34999e-06
[rewriter_after_jit_bprop_graph]: 6.58998e-06
[opt_after_jit_grad]: 0.00051767
[symbol_engine_optimizer]: 8.33e-05, [1]
    [Cycle 1]: 7.645e-05, [6]
        [build]: 4.97e-06
        [elim_shapecalc]: 8.32998e-06
        [elim_not_effective]: 1.512e-05
        [opt_reshape]: 6.69001e-06
        [fold_const_symbol]: 9.91e-06
        [renormalize]: 7.60017e-07
[validate]: 3.923e-05
Sums
    bootstrap                            :     0.000297s :  2.23%
    type_inference                       :     0.007959s : 59.55%
    event_method                         :     0.000027s :  0.20%
    auto_monad                           :     0.000091s :  0.68%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000064s :  0.48%
    rewriter_before_opt_a                :     0.000108s :  0.81%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000151s :  1.13%
    jit_opt_a.loop_unroll                :     0.000043s :  0.32%
    jit_opt_a.a_1                        :     0.000975s :  7.29%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000008s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000020s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.001081s :  8.09%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.16%
    jit_opt_a.cse                        :     0.000040s :  0.30%
    jit_opt_a.replace_applicator         :     0.000020s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.29%
    convert_after_rewriter               :     0.000008s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000754s :  5.64%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.13%
    cconv                                :     0.000030s :  0.22%
    loop_unroll                          :     0.000542s :  4.06%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000026s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000015s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000050s :  0.37%
    cse_after_recomputation.cse          :     0.000011s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.05%
    opt_after_jit_grad                   :     0.000518s :  3.87%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.29%


Time group info:
------[substitution.]   0.000407    26
 0.61% :     0.000002s :      2: substitution.elim_not_effective
 0.30% :     0.000001s :      2: substitution.fold_const_symbol
 1.46% :     0.000006s :      3: substitution.graph_param_transform
87.58% :     0.000356s :      6: substitution.inline
 0.78% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.58% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.37% :     0.000006s :      2: substitution.replace_old_param
 2.38% :     0.000010s :      1: substitution.switch_simplify
 3.94% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007873     2
78.36% :     0.006170s :      1: type_inference.infer
21.64% :     0.001704s :      1: type_inference.specialize
------[replace.]   0.000173     9
49.90% :     0.000086s :      6: replace.inline
25.89% :     0.000045s :      1: replace.switch_simplify
24.21% :     0.000042s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000376     9
93.70% :     0.000352s :      6: match.inline
 2.36% :     0.000009s :      1: match.switch_simplify
 3.95% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000151   801
 1.21% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.37% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000002s :     12: predicate.addn_check_dump
 1.32% :     0.000002s :     12: predicate.addn_zero_filter
 2.36% :     0.000004s :     12: predicate.arithmetic_simplify
 1.26% :     0.000002s :     12: predicate.cast_eliminate
 0.59% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000002s :     12: predicate.compare_switch_simplify
 1.27% :     0.000002s :     12: predicate.depend_value_elim
 1.02% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.17% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.11% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.40% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.05% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.04% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.07% :     0.000002s :     12: predicate.environ_get_eliminate
 1.05% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.76% :     0.000001s :      6: predicate.get_grad_eliminate
 0.36% :     0.000001s :      3: predicate.graph_param_transform
 5.83% :     0.000009s :     26: predicate.inline
 0.73% :     0.000001s :      6: predicate.inline_without_move
 0.29% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.87% :     0.000003s :     14: predicate.list_to_tuple_eliminator_
 2.19% :     0.000003s :     17: predicate.load_eliminater
 2.24% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.19% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.02% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000002s :     12: predicate.merge_addn
 1.05% :     0.000002s :     12: predicate.minmaximum_grad
 2.17% :     0.000003s :      3: predicate.mutable_eliminate
 0.39% :     0.000001s :      3: predicate.opt_reshape
 2.37% :     0.000004s :     17: predicate.partial_eliminate
 1.19% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     12: predicate.reduce_eliminate
 1.64% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.52% :     0.000002s :     20: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.23% :     0.000002s :     12: predicate.reshape_eliminate
 1.19% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     12: predicate.same_eliminate
 0.42% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000002s :      6: predicate.special_op_eliminate
 0.79% :     0.000001s :      6: predicate.specialize_transform
 1.51% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.34% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.56% :     0.000005s :     20: predicate.switch_defer_inline
 2.18% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.46% :     0.000013s :     65: predicate.switch_simplify
 1.15% :     0.000002s :     12: predicate.tile_eliminate
 1.48% :     0.000002s :     12: predicate.transpose_eliminate
 1.38% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.71% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.52% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     12: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001127    15
40.63% :     0.000458s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.73% :     0.000155s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.63% :     0.000514s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020300    72
 0.26% :     0.000053s :      1: add_recomputation
 0.47% :     0.000095s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 1.54% :     0.000312s :      1: bootstrap
 0.16% :     0.000033s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.13% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.16% :     0.000032s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.12% :     0.006723s :      1: jit_opt_a
 0.82% :     0.000167s :      1: jit_opt_after_cconv
 0.31% :     0.000064s :      1: jit_opt_b
 2.72% :     0.000552s :      1: loop_unroll
 3.77% :     0.000765s :      1: mutable_eliminate
 6.40% :     0.001300s :     26: opt.transform.jit_opt_a
 0.27% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000029s :      4: opt.transform.jit_opt_b
 0.08% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.60% :     0.000528s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.33% :     0.000068s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000018s :      1: remove_dup_value
 3.31% :     0.000673s :      1: renormalize.infer
 1.96% :     0.000398s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000041s :      1: rewriter_after_opt_a
 0.55% :     0.000113s :      1: rewriter_before_opt_a
 0.42% :     0.000086s :      1: symbol_engine_optimizer
39.30% :     0.007977s :      1: type_inference


TotalTime = 0.0189522, [33]
[bootstrap]: 0.00090054
[type_inference]: 0.00429496
[event_method]: 1.652e-05
[auto_monad]: 6.039e-05
[graph_reusing]: 6.24999e-06
[pre_auto_parallel]: 2.49001e-06
[py_interpret_to_execute]: 3.198e-05
[rewriter_before_opt_a]: 7.13e-05
[expand_dump_flag]: 3.46001e-06
[jit_opt_a]: 0.00531501, [2]
    [Cycle 1]: 0.00162403, [27]
        [switch_simplify]: 5.803e-05
        [loop_unroll]: 3.009e-05
        [a_1]: 0.0005452
        [with_stream_mark]: 2.03e-05
        [recompute_prepare]: 7.68001e-06
        [updatestate_depend_eliminate]: 3.65998e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.97999e-06
        [specialize_transform]: 6.00002e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 5.55001e-06
        [meta_shard_fg_expand]: 2.02001e-06
        [get_grad_eliminate_]: 5.76e-06
        [merge_forward]: 3.9e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.511e-05
        [j_node_and_user_rematch]: 1.015e-05
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 9.61003e-06
        [inline_without_move]: 5.86e-06
        [renormalize]: 0.00063933
        [add_forward_monad_depend]: 5.23002e-06
        [auto_monad_grad]: 2.35002e-06
        [auto_monad_eliminator]: 1.306e-05
        [cse]: 2.952e-05
        [replace_applicator]: 1.209e-05
    [Cycle 2]: 0.00032336, [27]
        [switch_simplify]: 5.39998e-06
        [loop_unroll]: 5.32001e-06
        [a_1]: 8.91e-05
        [with_stream_mark]: 9.42999e-06
        [recompute_prepare]: 5.00001e-06
        [updatestate_depend_eliminate]: 2.41e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.03002e-06
        [meta_shard_fg_expand]: 1.02e-06
        [get_grad_eliminate_]: 4.56002e-06
        [merge_forward]: 2.41998e-06
        [cell_reuse_recompute_pass]: 2.69999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.295e-05
        [j_node_and_user_rematch]: 8.36002e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 7.93999e-06
        [inline_without_move]: 5.34998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.21997e-06
        [auto_monad_grad]: 1.07998e-06
        [auto_monad_eliminator]: 5.39e-06
        [cse]: 1.025e-05
        [replace_applicator]: 4.87e-06
[py_interpret_to_execute_after_opt_a]: 1.371e-05
[rewriter_after_opt_a]: 3.503e-05
[convert_after_rewriter]: 6.54001e-06
[order_py_execute_after_rewriter]: 5.89e-06
[mutable_eliminate]: 0.00057056
[jit_opt_b]: 5.515e-05, [1]
    [Cycle 1]: 4.849e-05, [2]
        [frontend_op_eliminate]: 2.049e-05
        [inline_after_opt_a]: 1.606e-05
[cconv]: 2.524e-05
[loop_unroll]: 0.00044014
[jit_opt_after_cconv]: 0.0001747, [1]
    [Cycle 1]: 0.00016818, [11]
        [c_1]: 2.048e-05
        [parameter_eliminate]: 2.60002e-06
        [updatestate_depend_eliminate]: 5.67001e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [cse]: 1.867e-05
        [call_graph_tuple_transform]: 2.186e-05
        [tuple_list_get_item_eliminator]: 5.72999e-06
        [none_parameter_eliminate]: 1.83002e-06
        [renormalize]: 4.10015e-07
        [switch_simplify]: 5.22999e-06
[remove_dup_value]: 1.362e-05
[partial_unused_args_eliminate]: 2.33998e-06
[environ_conv]: 5.41002e-06
[add_recomputation]: 4.918e-05
[cse_after_recomputation]: 2.2e-05, [1]
    [Cycle 1]: 1.576e-05, [1]
        [cse]: 1.025e-05
[auto_monad_reorder]: 1.558e-05
[get_jit_bprop_graph]: 1.57999e-06
[rewriter_after_jit_bprop_graph]: 5.40999e-06
[opt_after_jit_grad]: 0.0004737
[symbol_engine_optimizer]: 7.222e-05, [1]
    [Cycle 1]: 6.576e-05, [6]
        [build]: 3.08e-06
        [elim_shapecalc]: 7.15e-06
        [elim_not_effective]: 1.286e-05
        [opt_reshape]: 5.54998e-06
        [fold_const_symbol]: 8.64e-06
        [renormalize]: 5.00004e-07
[validate]: 3.586e-05
[backend_pass]: 8.2e-07
[task_emit]: 0.00602267
[execute]: 7.71001e-06
Sums
    bootstrap                            :     0.000901s :  6.03%
    type_inference                       :     0.004295s : 28.74%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000060s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.21%
    rewriter_before_opt_a                :     0.000071s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.42%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000634s :  4.24%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.000639s :  4.28%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000040s :  0.27%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.09%
    rewriter_after_opt_a                 :     0.000035s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000571s :  3.82%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000440s :  2.95%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000049s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000474s :  3.17%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006023s : 40.31%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000213    23
 0.90% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.59% :     0.000006s :      3: substitution.graph_param_transform
84.01% :     0.000179s :      4: substitution.inline
 1.62% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.35% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.75% :     0.000004s :      2: substitution.replace_old_param
 6.13% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004231     2
81.72% :     0.003458s :      1: type_inference.infer
18.28% :     0.000773s :      1: type_inference.specialize
------[replace.]   0.000067     6
68.15% :     0.000046s :      4: replace.inline
31.85% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000188     6
93.63% :     0.000176s :      4: match.inline
 6.37% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.22% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.46% :     0.000002s :     10: predicate.addn_zero_filter
 2.39% :     0.000003s :     10: predicate.arithmetic_simplify
 1.28% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.33% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.09% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.49% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.45% :     0.000002s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.34% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.15% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.46% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.13% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.23% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     16: predicate.switch_defer_inline
 2.20% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.72% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.26% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.30% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.97% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000540    12
47.59% :     0.000257s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.76% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.65% :     0.000246s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020551    76
 0.25% :     0.000052s :      1: add_recomputation
 0.31% :     0.000064s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 4.45% :     0.000914s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.88% :     0.005318s :      1: jit_opt_a
 0.87% :     0.000178s :      1: jit_opt_after_cconv
 0.28% :     0.000058s :      1: jit_opt_b
 2.18% :     0.000448s :      1: loop_unroll
 2.82% :     0.000579s :      1: mutable_eliminate
 4.14% :     0.000851s :     26: opt.transform.jit_opt_a
 0.24% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000030s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.34% :     0.000482s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000034s :      1: py_interpret_to_execute
 0.08% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.59% :     0.000326s :      1: renormalize.infer
 1.49% :     0.000306s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000038s :      1: rewriter_after_opt_a
 0.36% :     0.000074s :      1: rewriter_before_opt_a
 0.36% :     0.000075s :      1: symbol_engine_optimizer
29.39% :     0.006039s :      1: task_emit
20.96% :     0.004308s :      1: type_inference
 0.28% :     0.000057s :      1: validate


TotalTime = 0.0176696, [33]
[bootstrap]: 0.00029525
[type_inference]: 0.00429081
[event_method]: 1.652e-05
[auto_monad]: 5.068e-05
[graph_reusing]: 4.82e-06
[pre_auto_parallel]: 2.11e-06
[py_interpret_to_execute]: 2.862e-05
[rewriter_before_opt_a]: 7.056e-05
[expand_dump_flag]: 2.50002e-06
[jit_opt_a]: 0.00541757, [2]
    [Cycle 1]: 0.00158733, [27]
        [switch_simplify]: 4.977e-05
        [loop_unroll]: 2.946e-05
        [a_1]: 0.00052299
        [with_stream_mark]: 1.489e-05
        [recompute_prepare]: 7.38e-06
        [updatestate_depend_eliminate]: 3.03e-06
        [updatestate_assign_eliminate]: 2.20002e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 1.22999e-06
        [specialize_transform]: 5.94999e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 4.95999e-06
        [meta_shard_fg_expand]: 1.79e-06
        [get_grad_eliminate_]: 5.00001e-06
        [merge_forward]: 2.96001e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.154e-05
        [j_node_and_user_rematch]: 8.55001e-06
        [meta_fg_expand]: 1.82999e-06
        [replace_old_param]: 8.22e-06
        [inline_without_move]: 5.27001e-06
        [renormalize]: 0.00067311
        [add_forward_monad_depend]: 5.93002e-06
        [auto_monad_grad]: 2.32999e-06
        [auto_monad_eliminator]: 1.246e-05
        [cse]: 1.981e-05
        [replace_applicator]: 1.281e-05
    [Cycle 2]: 0.00033302, [27]
        [switch_simplify]: 5.42001e-06
        [loop_unroll]: 4.70999e-06
        [a_1]: 8.729e-05
        [with_stream_mark]: 9.81998e-06
        [recompute_prepare]: 5.37999e-06
        [updatestate_depend_eliminate]: 2.49001e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 1.30001e-06
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 4.61002e-06
        [meta_shard_fg_expand]: 1.00999e-06
        [get_grad_eliminate_]: 4.89e-06
        [merge_forward]: 2.71e-06
        [cell_reuse_recompute_pass]: 1.51998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.336e-05
        [j_node_and_user_rematch]: 7.82998e-06
        [meta_fg_expand]: 1.55999e-06
        [replace_old_param]: 7.76001e-06
        [inline_without_move]: 9.37999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.49e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 5.69e-06
        [cse]: 1.116e-05
        [replace_applicator]: 5.03002e-06
[py_interpret_to_execute_after_opt_a]: 1.142e-05
[rewriter_after_opt_a]: 3.561e-05
[convert_after_rewriter]: 6.51e-06
[order_py_execute_after_rewriter]: 4.57e-06
[mutable_eliminate]: 0.00058821
[jit_opt_b]: 4.878e-05, [1]
    [Cycle 1]: 4.167e-05, [2]
        [frontend_op_eliminate]: 1.568e-05
        [inline_after_opt_a]: 1.445e-05
[cconv]: 2.571e-05
[loop_unroll]: 0.00043219
[jit_opt_after_cconv]: 0.00013752, [1]
    [Cycle 1]: 0.00013097, [11]
        [c_1]: 2.012e-05
        [parameter_eliminate]: 2.91999e-06
        [updatestate_depend_eliminate]: 5.32999e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [cse]: 1.88e-05
        [call_graph_tuple_transform]: 1.847e-05
        [tuple_list_get_item_eliminator]: 4.96002e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 5.39992e-07
        [switch_simplify]: 5.05001e-06
[remove_dup_value]: 1.121e-05
[partial_unused_args_eliminate]: 1.91e-06
[environ_conv]: 4.49002e-06
[add_recomputation]: 4.62e-05
[cse_after_recomputation]: 1.982e-05, [1]
    [Cycle 1]: 1.484e-05, [1]
        [cse]: 9.30001e-06
[auto_monad_reorder]: 1.328e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 6.31e-06
[opt_after_jit_grad]: 0.00046807
[symbol_engine_optimizer]: 7.343e-05, [1]
    [Cycle 1]: 6.664e-05, [6]
        [build]: 3.8e-06
        [elim_shapecalc]: 7.09001e-06
        [elim_not_effective]: 1.249e-05
        [opt_reshape]: 5.79999e-06
        [fold_const_symbol]: 8.35999e-06
        [renormalize]: 5.89993e-07
[validate]: 3.743e-05
[backend_pass]: 6.59988e-07
[task_emit]: 0.00532851
[execute]: 5.62001e-06
Sums
    bootstrap                            :     0.000295s :  2.17%
    type_inference                       :     0.004291s : 31.59%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000051s :  0.37%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.21%
    rewriter_before_opt_a                :     0.000071s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000610s :  4.49%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000015s :  0.11%
    jit_opt_a.renormalize                :     0.000673s :  4.96%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000031s :  0.23%
    jit_opt_a.replace_applicator         :     0.000018s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.08%
    rewriter_after_opt_a                 :     0.000036s :  0.26%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000588s :  4.33%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000026s :  0.19%
    loop_unroll                          :     0.000432s :  3.18%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000011s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000046s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000013s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000468s :  3.45%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.28%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005329s : 39.23%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000206    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.46% :     0.000005s :      3: substitution.graph_param_transform
84.75% :     0.000175s :      4: substitution.inline
 1.48% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.01% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.79% :     0.000004s :      2: substitution.replace_old_param
 5.92% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004231     2
82.19% :     0.003477s :      1: type_inference.infer
17.81% :     0.000753s :      1: type_inference.specialize
------[replace.]   0.000066     6
67.62% :     0.000045s :      4: replace.inline
32.38% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000184     6
93.91% :     0.000173s :      4: match.inline
 6.09% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.55% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 1.64% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.67% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.47% :     0.000001s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.13% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.04% :     0.000001s :      6: predicate.less_batch_normalization
 1.64% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.01% :     0.000002s :     15: predicate.load_eliminater
 1.77% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.24% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.82% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000002s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.26% :     0.000001s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.67% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.41% :     0.000008s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.31% :     0.000001s :     10: predicate.transpose_eliminate
 1.55% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.46% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000002s :     10: predicate.value_based_eliminate
 0.43% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000520    12
45.41% :     0.000236s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.02% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.57% :     0.000242s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019268    76
 0.25% :     0.000049s :      1: add_recomputation
 0.28% :     0.000054s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.61% :     0.000310s :      1: bootstrap
 0.15% :     0.000028s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
28.13% :     0.005420s :      1: jit_opt_a
 0.73% :     0.000140s :      1: jit_opt_after_cconv
 0.27% :     0.000051s :      1: jit_opt_b
 2.28% :     0.000439s :      1: loop_unroll
 3.10% :     0.000597s :      1: mutable_eliminate
 4.24% :     0.000817s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.48% :     0.000477s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000031s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000014s :      1: remove_dup_value
 1.95% :     0.000376s :      1: renormalize.infer
 1.50% :     0.000289s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000038s :      1: rewriter_after_opt_a
 0.38% :     0.000073s :      1: rewriter_before_opt_a
 0.39% :     0.000076s :      1: symbol_engine_optimizer
27.72% :     0.005341s :      1: task_emit
22.33% :     0.004302s :      1: type_inference
 0.30% :     0.000057s :      1: validate


TotalTime = 0.0203631, [33]
[bootstrap]: 0.00033199
[type_inference]: 0.00526112
[event_method]: 1.896e-05
[auto_monad]: 6.779e-05
[graph_reusing]: 5.95002e-06
[pre_auto_parallel]: 2.54999e-06
[py_interpret_to_execute]: 3.862e-05
[rewriter_before_opt_a]: 7.558e-05
[expand_dump_flag]: 3.18e-06
[jit_opt_a]: 0.00589661, [2]
    [Cycle 1]: 0.00170566, [27]
        [switch_simplify]: 5.762e-05
        [loop_unroll]: 2.918e-05
        [a_1]: 0.00053068
        [with_stream_mark]: 1.931e-05
        [recompute_prepare]: 8.17998e-06
        [updatestate_depend_eliminate]: 3.97002e-06
        [updatestate_assign_eliminate]: 3.28998e-06
        [updatestate_loads_eliminate]: 2.64001e-06
        [parameter_eliminate]: 2.28002e-06
        [specialize_transform]: 6.53e-06
        [updatestate_useless_node_eliminater]: 5.67999e-06
        [accelerated_algorithm]: 6.14999e-06
        [meta_shard_fg_expand]: 2.17999e-06
        [get_grad_eliminate_]: 5.55001e-06
        [merge_forward]: 4.08001e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.438e-05
        [j_node_and_user_rematch]: 9.12999e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 9.12001e-06
        [inline_without_move]: 5.25001e-06
        [renormalize]: 0.00072581
        [add_forward_monad_depend]: 5.96998e-06
        [auto_monad_grad]: 2.22001e-06
        [auto_monad_eliminator]: 1.347e-05
        [cse]: 3.106e-05
        [replace_applicator]: 1.1e-05
    [Cycle 2]: 0.0003112, [27]
        [switch_simplify]: 5.42999e-06
        [loop_unroll]: 5.04998e-06
        [a_1]: 8.924e-05
        [with_stream_mark]: 8.77e-06
        [recompute_prepare]: 4.94998e-06
        [updatestate_depend_eliminate]: 2.68003e-06
        [updatestate_assign_eliminate]: 1.98997e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 9.90025e-07
        [specialize_transform]: 4.75999e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.05999e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 4.51002e-06
        [merge_forward]: 2.27001e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.284e-05
        [j_node_and_user_rematch]: 7.75e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 7.28999e-06
        [inline_without_move]: 4.63999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.40999e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.08002e-06
        [cse]: 1.002e-05
        [replace_applicator]: 4.74e-06
[py_interpret_to_execute_after_opt_a]: 1.004e-05
[rewriter_after_opt_a]: 3.699e-05
[convert_after_rewriter]: 7.71001e-06
[order_py_execute_after_rewriter]: 5.07999e-06
[mutable_eliminate]: 0.00055557
[jit_opt_b]: 4.979e-05, [1]
    [Cycle 1]: 4.256e-05, [2]
        [frontend_op_eliminate]: 1.633e-05
        [inline_after_opt_a]: 1.471e-05
[cconv]: 2.579e-05
[loop_unroll]: 0.00042193
[jit_opt_after_cconv]: 0.00013994, [1]
    [Cycle 1]: 0.00013345, [11]
        [c_1]: 2.026e-05
        [parameter_eliminate]: 3.06999e-06
        [updatestate_depend_eliminate]: 5.44e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [cse]: 1.899e-05
        [call_graph_tuple_transform]: 1.867e-05
        [tuple_list_get_item_eliminator]: 5.35001e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 8.59989e-07
        [switch_simplify]: 4.97e-06
[remove_dup_value]: 1.503e-05
[partial_unused_args_eliminate]: 2.24999e-06
[environ_conv]: 5.35001e-06
[add_recomputation]: 5.094e-05
[cse_after_recomputation]: 1.966e-05, [1]
    [Cycle 1]: 1.455e-05, [1]
        [cse]: 8.99998e-06
[auto_monad_reorder]: 1.598e-05
[get_jit_bprop_graph]: 1.96e-06
[rewriter_after_jit_bprop_graph]: 4.95001e-06
[opt_after_jit_grad]: 0.00046107
[symbol_engine_optimizer]: 7.235e-05, [1]
    [Cycle 1]: 6.597e-05, [6]
        [build]: 3.68e-06
        [elim_shapecalc]: 7.51999e-06
        [elim_not_effective]: 1.246e-05
        [opt_reshape]: 5.59e-06
        [fold_const_symbol]: 9.27999e-06
        [renormalize]: 6.30011e-07
[validate]: 3.769e-05
[backend_pass]: 1.02e-06
[task_emit]: 0.00649407
[execute]: 9.90002e-06
Sums
    bootstrap                            :     0.000332s :  2.09%
    type_inference                       :     0.005261s : 33.19%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000068s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.24%
    rewriter_before_opt_a                :     0.000076s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000620s :  3.91%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000726s :  4.58%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000041s :  0.26%
    jit_opt_a.replace_applicator         :     0.000016s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000037s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000556s :  3.50%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000026s :  0.16%
    loop_unroll                          :     0.000422s :  2.66%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000051s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000461s :  2.91%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006494s : 40.97%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000202    23
 1.14% :     0.000002s :      2: substitution.elim_not_effective
 0.83% :     0.000002s :      2: substitution.fold_const_symbol
 2.74% :     0.000006s :      3: substitution.graph_param_transform
83.30% :     0.000168s :      4: substitution.inline
 1.30% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.25% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.14% :     0.000004s :      2: substitution.replace_old_param
 6.30% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005187     2
81.18% :     0.004211s :      1: type_inference.infer
18.82% :     0.000976s :      1: type_inference.specialize
------[replace.]   0.000068     6
68.21% :     0.000046s :      4: replace.inline
31.79% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000178     6
93.41% :     0.000166s :      4: match.inline
 6.59% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.81% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.59% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000001s :     10: predicate.addn_zero_filter
 2.10% :     0.000002s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.47% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.65% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000001s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.58% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.37% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 2.20% :     0.000003s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.32% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.81% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.31% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.38% :     0.000008s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.88% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000002s :     10: predicate.value_based_eliminate
 0.45% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000683    12
49.77% :     0.000340s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.35% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.88% :     0.000313s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022017    76
 0.24% :     0.000054s :      1: add_recomputation
 0.32% :     0.000071s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.59% :     0.000350s :      1: bootstrap
 0.13% :     0.000029s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000024s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.80% :     0.005900s :      1: jit_opt_a
 0.65% :     0.000143s :      1: jit_opt_after_cconv
 0.24% :     0.000053s :      1: jit_opt_b
 1.96% :     0.000431s :      1: loop_unroll
 2.56% :     0.000564s :      1: mutable_eliminate
 3.77% :     0.000831s :     26: opt.transform.jit_opt_a
 0.21% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.13% :     0.000469s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000041s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.70% :     0.000373s :      1: renormalize.infer
 1.57% :     0.000345s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000040s :      1: rewriter_after_opt_a
 0.36% :     0.000078s :      1: rewriter_before_opt_a
 0.34% :     0.000075s :      1: symbol_engine_optimizer
29.58% :     0.006514s :      1: task_emit
23.97% :     0.005278s :      1: type_inference
 0.26% :     0.000058s :      1: validate


TotalTime = 0.0188675, [33]
[bootstrap]: 0.0003134
[type_inference]: 0.0046519
[event_method]: 1.592e-05
[auto_monad]: 5.978e-05
[graph_reusing]: 5.59998e-06
[pre_auto_parallel]: 2.78e-06
[py_interpret_to_execute]: 2.999e-05
[rewriter_before_opt_a]: 7.194e-05
[expand_dump_flag]: 2.89001e-06
[jit_opt_a]: 0.00514722, [2]
    [Cycle 1]: 0.00156267, [27]
        [switch_simplify]: 5.026e-05
        [loop_unroll]: 2.859e-05
        [a_1]: 0.00050289
        [with_stream_mark]: 1.799e-05
        [recompute_prepare]: 8.51002e-06
        [updatestate_depend_eliminate]: 3.71001e-06
        [updatestate_assign_eliminate]: 3.03e-06
        [updatestate_loads_eliminate]: 2.61999e-06
        [parameter_eliminate]: 2.18002e-06
        [specialize_transform]: 5.86e-06
        [updatestate_useless_node_eliminater]: 5.12999e-06
        [accelerated_algorithm]: 4.94998e-06
        [meta_shard_fg_expand]: 2.31e-06
        [get_grad_eliminate_]: 5.27001e-06
        [merge_forward]: 3.76999e-06
        [cell_reuse_recompute_pass]: 1.87999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.453e-05
        [j_node_and_user_rematch]: 9.04998e-06
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 9.40001e-06
        [inline_without_move]: 5.22e-06
        [renormalize]: 0.0006272
        [add_forward_monad_depend]: 5.55001e-06
        [auto_monad_grad]: 2.46998e-06
        [auto_monad_eliminator]: 1.61e-05
        [cse]: 2.807e-05
        [replace_applicator]: 1.343e-05
    [Cycle 2]: 0.00032395, [27]
        [switch_simplify]: 5.56e-06
        [loop_unroll]: 4.90999e-06
        [a_1]: 8.795e-05
        [with_stream_mark]: 9.51998e-06
        [recompute_prepare]: 5.10001e-06
        [updatestate_depend_eliminate]: 2.74999e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [parameter_eliminate]: 1.11997e-06
        [specialize_transform]: 4.87998e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 4.88001e-06
        [meta_shard_fg_expand]: 1.13001e-06
        [get_grad_eliminate_]: 4.27e-06
        [merge_forward]: 2.67001e-06
        [cell_reuse_recompute_pass]: 2.16e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.41e-05
        [j_node_and_user_rematch]: 7.85998e-06
        [meta_fg_expand]: 1.42e-06
        [replace_old_param]: 7.78999e-06
        [inline_without_move]: 9.37001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 1.14998e-06
        [auto_monad_eliminator]: 5.79999e-06
        [cse]: 1.154e-05
        [replace_applicator]: 5.62999e-06
[py_interpret_to_execute_after_opt_a]: 1.172e-05
[rewriter_after_opt_a]: 3.689e-05
[convert_after_rewriter]: 7.28e-06
[order_py_execute_after_rewriter]: 5.50001e-06
[mutable_eliminate]: 0.00059019
[jit_opt_b]: 4.829e-05, [1]
    [Cycle 1]: 4.088e-05, [2]
        [frontend_op_eliminate]: 1.524e-05
        [inline_after_opt_a]: 1.397e-05
[cconv]: 2.58e-05
[loop_unroll]: 0.00054161
[jit_opt_after_cconv]: 0.00014649, [1]
    [Cycle 1]: 0.00013946, [11]
        [c_1]: 2.055e-05
        [parameter_eliminate]: 2.92002e-06
        [updatestate_depend_eliminate]: 5.24e-06
        [updatestate_assign_eliminate]: 2.23002e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 1.936e-05
        [call_graph_tuple_transform]: 1.877e-05
        [tuple_list_get_item_eliminator]: 4.95001e-06
        [none_parameter_eliminate]: 1.35999e-06
        [renormalize]: 5.3001e-07
        [switch_simplify]: 5.47999e-06
[remove_dup_value]: 1.435e-05
[partial_unused_args_eliminate]: 2.20002e-06
[environ_conv]: 5.44e-06
[add_recomputation]: 4.853e-05
[cse_after_recomputation]: 2.021e-05, [1]
    [Cycle 1]: 1.485e-05, [1]
        [cse]: 9.05999e-06
[auto_monad_reorder]: 1.536e-05
[get_jit_bprop_graph]: 2.14999e-06
[rewriter_after_jit_bprop_graph]: 4.75999e-06
[opt_after_jit_grad]: 0.00050557
[symbol_engine_optimizer]: 7.258e-05, [1]
    [Cycle 1]: 6.53e-05, [6]
        [build]: 4.65999e-06
        [elim_shapecalc]: 7.18e-06
        [elim_not_effective]: 1.255e-05
        [opt_reshape]: 5.34e-06
        [fold_const_symbol]: 7.73001e-06
        [renormalize]: 8.70001e-07
[validate]: 3.521e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00620968
[execute]: 8.32e-06
Sums
    bootstrap                            :     0.000313s :  2.09%
    type_inference                       :     0.004652s : 31.05%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000060s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.20%
    rewriter_before_opt_a                :     0.000072s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.37%
    jit_opt_a.loop_unroll                :     0.000033s :  0.22%
    jit_opt_a.a_1                        :     0.000591s :  3.94%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000015s :  0.10%
    jit_opt_a.renormalize                :     0.000627s :  4.19%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.15%
    jit_opt_a.cse                        :     0.000040s :  0.26%
    jit_opt_a.replace_applicator         :     0.000019s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000037s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000590s :  3.94%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000026s :  0.17%
    loop_unroll                          :     0.000542s :  3.62%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000049s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000506s :  3.37%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000035s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006210s : 41.45%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000189    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 3.06% :     0.000006s :      3: substitution.graph_param_transform
82.33% :     0.000156s :      4: substitution.inline
 1.67% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.71% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.56% :     0.000005s :      2: substitution.replace_old_param
 6.07% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004588     2
83.14% :     0.003815s :      1: type_inference.infer
16.86% :     0.000773s :      1: type_inference.specialize
------[replace.]   0.000064     6
67.60% :     0.000043s :      4: replace.inline
32.40% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000164     6
93.67% :     0.000153s :      4: match.inline
 6.33% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.75% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000001s :     10: predicate.addn_zero_filter
 1.84% :     0.000002s :     10: predicate.arithmetic_simplify
 1.37% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.65% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.06% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.59% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.04% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.73% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.21% :     0.000001s :     10: predicate.minmaximum_grad
 1.94% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000000s :      3: predicate.opt_reshape
 2.19% :     0.000002s :     15: predicate.partial_eliminate
 1.49% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.99% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.28% :     0.000001s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.51% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.85% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.82% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.08% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000543    12
51.52% :     0.000280s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.18% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.29% :     0.000235s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020398    76
 0.25% :     0.000051s :      1: add_recomputation
 0.31% :     0.000063s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.63% :     0.000333s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.25% :     0.005151s :      1: jit_opt_a
 0.73% :     0.000149s :      1: jit_opt_after_cconv
 0.25% :     0.000051s :      1: jit_opt_b
 2.70% :     0.000550s :      1: loop_unroll
 2.95% :     0.000601s :      1: mutable_eliminate
 3.92% :     0.000800s :     26: opt.transform.jit_opt_a
 0.23% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000023s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.53% :     0.000515s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000032s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.65% :     0.000336s :      1: renormalize.infer
 1.39% :     0.000283s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000040s :      1: rewriter_after_opt_a
 0.37% :     0.000075s :      1: rewriter_before_opt_a
 0.37% :     0.000075s :      1: symbol_engine_optimizer
30.51% :     0.006224s :      1: task_emit
22.88% :     0.004667s :      1: type_inference
 0.28% :     0.000057s :      1: validate


TotalTime = 0.0194759, [33]
[bootstrap]: 0.00029769
[type_inference]: 0.00478685
[event_method]: 1.799e-05
[auto_monad]: 6.555e-05
[graph_reusing]: 5.98002e-06
[pre_auto_parallel]: 2.34999e-06
[py_interpret_to_execute]: 3.512e-05
[rewriter_before_opt_a]: 7.525e-05
[expand_dump_flag]: 3.71001e-06
[jit_opt_a]: 0.00568453, [2]
    [Cycle 1]: 0.00160183, [27]
        [switch_simplify]: 6.001e-05
        [loop_unroll]: 3.106e-05
        [a_1]: 0.00053912
        [with_stream_mark]: 1.83e-05
        [recompute_prepare]: 7.05002e-06
        [updatestate_depend_eliminate]: 4.08001e-06
        [updatestate_assign_eliminate]: 3.2e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 7.26001e-06
        [updatestate_useless_node_eliminater]: 5.47001e-06
        [accelerated_algorithm]: 6.56e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 5.74e-06
        [merge_forward]: 3.88001e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.431e-05
        [j_node_and_user_rematch]: 9.69e-06
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 9.11002e-06
        [inline_without_move]: 5.66003e-06
        [renormalize]: 0.00062714
        [add_forward_monad_depend]: 4.95999e-06
        [auto_monad_grad]: 2.39999e-06
        [auto_monad_eliminator]: 1.478e-05
        [cse]: 2.514e-05
        [replace_applicator]: 1.235e-05
    [Cycle 2]: 0.0003369, [27]
        [switch_simplify]: 5.69e-06
        [loop_unroll]: 5.20001e-06
        [a_1]: 8.975e-05
        [with_stream_mark]: 2.211e-05
        [recompute_prepare]: 5.94999e-06
        [updatestate_depend_eliminate]: 2.82002e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 5.59e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.05999e-06
        [meta_shard_fg_expand]: 1.38002e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 2.51e-06
        [cell_reuse_recompute_pass]: 2.20002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.388e-05
        [j_node_and_user_rematch]: 7.95e-06
        [meta_fg_expand]: 2.00002e-06
        [replace_old_param]: 7.61999e-06
        [inline_without_move]: 4.74e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 1.17e-06
        [auto_monad_eliminator]: 5.54998e-06
        [cse]: 1.149e-05
        [replace_applicator]: 4.72e-06
[py_interpret_to_execute_after_opt_a]: 1.08e-05
[rewriter_after_opt_a]: 3.515e-05
[convert_after_rewriter]: 7.23e-06
[order_py_execute_after_rewriter]: 5.32999e-06
[mutable_eliminate]: 0.00059289
[jit_opt_b]: 4.779e-05, [1]
    [Cycle 1]: 4.16e-05, [2]
        [frontend_op_eliminate]: 1.63e-05
        [inline_after_opt_a]: 1.394e-05
[cconv]: 2.497e-05
[loop_unroll]: 0.00043724
[jit_opt_after_cconv]: 0.0001408, [1]
    [Cycle 1]: 0.00013415, [11]
        [c_1]: 2.121e-05
        [parameter_eliminate]: 2.69001e-06
        [updatestate_depend_eliminate]: 5.54e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.856e-05
        [call_graph_tuple_transform]: 1.919e-05
        [tuple_list_get_item_eliminator]: 5.24e-06
        [none_parameter_eliminate]: 1.72001e-06
        [renormalize]: 9.39996e-07
        [switch_simplify]: 5.32999e-06
[remove_dup_value]: 1.382e-05
[partial_unused_args_eliminate]: 2.48e-06
[environ_conv]: 4.87e-06
[add_recomputation]: 4.973e-05
[cse_after_recomputation]: 2.039e-05, [1]
    [Cycle 1]: 1.503e-05, [1]
        [cse]: 9.37001e-06
[auto_monad_reorder]: 1.543e-05
[get_jit_bprop_graph]: 2.03002e-06
[rewriter_after_jit_bprop_graph]: 5.11997e-06
[opt_after_jit_grad]: 0.00046826
[symbol_engine_optimizer]: 7.393e-05, [1]
    [Cycle 1]: 6.735e-05, [6]
        [build]: 3.75e-06
        [elim_shapecalc]: 7.93999e-06
        [elim_not_effective]: 1.29e-05
        [opt_reshape]: 5.46998e-06
        [fold_const_symbol]: 8.35001e-06
        [renormalize]: 4.19997e-07
[validate]: 3.604e-05
[backend_pass]: 8.10018e-07
[task_emit]: 0.0062952
[execute]: 8.34998e-06
Sums
    bootstrap                            :     0.000298s :  1.97%
    type_inference                       :     0.004787s : 31.66%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000066s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.23%
    rewriter_before_opt_a                :     0.000075s :  0.50%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.43%
    jit_opt_a.loop_unroll                :     0.000036s :  0.24%
    jit_opt_a.a_1                        :     0.000629s :  4.16%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000627s :  4.15%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000037s :  0.24%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000593s :  3.92%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000437s :  2.89%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000468s :  3.10%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006295s : 41.63%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000197    23
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.78% :     0.000005s :      3: substitution.graph_param_transform
83.28% :     0.000164s :      4: substitution.inline
 1.67% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.28% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000004s :      2: substitution.replace_old_param
 6.32% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004720     2
81.01% :     0.003824s :      1: type_inference.infer
18.99% :     0.000896s :      1: type_inference.specialize
------[replace.]   0.000066     6
65.47% :     0.000043s :      4: replace.inline
34.53% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000172     6
93.50% :     0.000161s :      4: match.inline
 6.50% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.55% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.51% :     0.000002s :     10: predicate.addn_zero_filter
 2.07% :     0.000002s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.57% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.13% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.40% :     0.000000s :      3: predicate.graph_param_transform
 5.19% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.56% :     0.000002s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.43% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.76% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.65% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.09% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.71% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.62% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.21% :     0.000001s :      6: predicate.special_op_eliminate
 1.14% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.66% :     0.000003s :     16: predicate.switch_defer_inline
 2.20% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.01% :     0.000010s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.03% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.16% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000593    12
45.53% :     0.000270s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.17% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.30% :     0.000292s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021055    76
 0.25% :     0.000053s :      1: add_recomputation
 0.33% :     0.000069s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.47% :     0.000310s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.01% :     0.005688s :      1: jit_opt_a
 0.68% :     0.000143s :      1: jit_opt_after_cconv
 0.24% :     0.000050s :      1: jit_opt_b
 2.11% :     0.000445s :      1: loop_unroll
 2.85% :     0.000601s :      1: mutable_eliminate
 4.04% :     0.000850s :     26: opt.transform.jit_opt_a
 0.22% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.26% :     0.000476s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000038s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.51% :     0.000319s :      1: renormalize.infer
 1.43% :     0.000301s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000038s :      1: rewriter_after_opt_a
 0.37% :     0.000078s :      1: rewriter_before_opt_a
 0.36% :     0.000076s :      1: symbol_engine_optimizer
29.97% :     0.006310s :      1: task_emit
22.84% :     0.004810s :      1: type_inference
 0.27% :     0.000056s :      1: validate


TotalTime = 0.0173785, [30]
[bootstrap]: 0.00028729
[type_inference]: 0.00762999
[event_method]: 2.213e-05
[auto_monad]: 8.71e-05
[graph_reusing]: 5.71003e-06
[pre_auto_parallel]: 2.73e-06
[py_interpret_to_execute]: 6.968e-05
[rewriter_before_opt_a]: 8.891e-05
[expand_dump_flag]: 3.2e-06
[jit_opt_a]: 0.00666721, [2]
    [Cycle 1]: 0.00232645, [27]
        [switch_simplify]: 0.00014869
        [loop_unroll]: 3.501e-05
        [a_1]: 0.00079034
        [with_stream_mark]: 2.051e-05
        [recompute_prepare]: 7.04001e-06
        [updatestate_depend_eliminate]: 3.71001e-06
        [updatestate_assign_eliminate]: 2.98998e-06
        [updatestate_loads_eliminate]: 2.50002e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 6.83e-06
        [updatestate_useless_node_eliminater]: 5.56e-06
        [accelerated_algorithm]: 5.80002e-06
        [meta_shard_fg_expand]: 2.44001e-06
        [get_grad_eliminate_]: 5.37001e-06
        [merge_forward]: 3.85998e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.472e-05
        [j_node_and_user_rematch]: 9.92001e-06
        [meta_fg_expand]: 2.29999e-06
        [replace_old_param]: 9.55001e-06
        [inline_without_move]: 5.71e-06
        [renormalize]: 0.00098696
        [add_forward_monad_depend]: 6.63e-06
        [auto_monad_grad]: 2.54001e-06
        [auto_monad_eliminator]: 1.49e-05
        [cse]: 2.442e-05
        [replace_applicator]: 1.576e-05
    [Cycle 2]: 0.00034218, [27]
        [switch_simplify]: 6.02999e-06
        [loop_unroll]: 5.25999e-06
        [a_1]: 9.955e-05
        [with_stream_mark]: 1.285e-05
        [recompute_prepare]: 5.25001e-06
        [updatestate_depend_eliminate]: 3.01999e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 5.05999e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.33002e-06
        [meta_shard_fg_expand]: 1.37999e-06
        [get_grad_eliminate_]: 4.65999e-06
        [merge_forward]: 2.66e-06
        [cell_reuse_recompute_pass]: 2.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.488e-05
        [j_node_and_user_rematch]: 8.89e-06
        [meta_fg_expand]: 1.99e-06
        [replace_old_param]: 8.17e-06
        [inline_without_move]: 4.95001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.18001e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 5.02e-06
        [cse]: 1.169e-05
        [replace_applicator]: 5.17e-06
[py_interpret_to_execute_after_opt_a]: 1.262e-05
[rewriter_after_opt_a]: 3.662e-05
[convert_after_rewriter]: 7.88001e-06
[order_py_execute_after_rewriter]: 4.90001e-06
[mutable_eliminate]: 0.00074258
[jit_opt_b]: 5.455e-05, [1]
    [Cycle 1]: 4.661e-05, [2]
        [frontend_op_eliminate]: 1.794e-05
        [inline_after_opt_a]: 1.519e-05
[cconv]: 3.132e-05
[loop_unroll]: 0.00050484
[jit_opt_after_cconv]: 0.00015649, [1]
    [Cycle 1]: 0.0001492, [11]
        [c_1]: 2.244e-05
        [parameter_eliminate]: 4.16001e-06
        [updatestate_depend_eliminate]: 7.98999e-06
        [updatestate_assign_eliminate]: 2.72001e-06
        [updatestate_loads_eliminate]: 2.26998e-06
        [cse]: 2.372e-05
        [call_graph_tuple_transform]: 2.028e-05
        [tuple_list_get_item_eliminator]: 5.88998e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 8.39995e-07
        [switch_simplify]: 5.88002e-06
[remove_dup_value]: 1.506e-05
[partial_unused_args_eliminate]: 2.34001e-06
[environ_conv]: 6.34999e-06
[add_recomputation]: 5.163e-05
[cse_after_recomputation]: 2.436e-05, [1]
    [Cycle 1]: 1.802e-05, [1]
        [cse]: 1.148e-05
[auto_monad_reorder]: 1.57e-05
[get_jit_bprop_graph]: 2.26e-06
[rewriter_after_jit_bprop_graph]: 7.43e-06
[opt_after_jit_grad]: 0.00052302
[symbol_engine_optimizer]: 8.42e-05, [1]
    [Cycle 1]: 7.703e-05, [6]
        [build]: 4.55999e-06
        [elim_shapecalc]: 8.79e-06
        [elim_not_effective]: 1.507e-05
        [opt_reshape]: 6.33002e-06
        [fold_const_symbol]: 1.064e-05
        [renormalize]: 3.80009e-07
[validate]: 4.16e-05
Sums
    bootstrap                            :     0.000287s :  2.25%
    type_inference                       :     0.007630s : 59.82%
    event_method                         :     0.000022s :  0.17%
    auto_monad                           :     0.000087s :  0.68%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000070s :  0.55%
    rewriter_before_opt_a                :     0.000089s :  0.70%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000155s :  1.21%
    jit_opt_a.loop_unroll                :     0.000040s :  0.32%
    jit_opt_a.a_1                        :     0.000890s :  6.98%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.14%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.000987s :  7.74%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.16%
    jit_opt_a.cse                        :     0.000036s :  0.28%
    jit_opt_a.replace_applicator         :     0.000021s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.10%
    rewriter_after_opt_a                 :     0.000037s :  0.29%
    convert_after_rewriter               :     0.000008s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000743s :  5.82%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000031s :  0.25%
    loop_unroll                          :     0.000505s :  3.96%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000015s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000052s :  0.40%
    cse_after_recomputation.cse          :     0.000011s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000523s :  4.10%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000042s :  0.33%


Time group info:
------[substitution.]   0.000342    26
 0.56% :     0.000002s :      2: substitution.elim_not_effective
 0.44% :     0.000002s :      2: substitution.fold_const_symbol
 1.62% :     0.000006s :      3: substitution.graph_param_transform
86.40% :     0.000296s :      6: substitution.inline
 1.00% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.39% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.33% :     0.000005s :      2: substitution.replace_old_param
 2.98% :     0.000010s :      1: substitution.switch_simplify
 4.27% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007555     2
82.16% :     0.006207s :      1: type_inference.infer
17.84% :     0.001348s :      1: type_inference.specialize
------[replace.]   0.000164     9
47.53% :     0.000078s :      6: replace.inline
27.48% :     0.000045s :      1: replace.switch_simplify
25.00% :     0.000041s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000315     9
92.77% :     0.000292s :      6: match.inline
 2.99% :     0.000009s :      1: match.switch_simplify
 4.25% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000143   801
 1.39% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     12: predicate.addn_check_dump
 1.46% :     0.000002s :     12: predicate.addn_zero_filter
 2.27% :     0.000003s :     12: predicate.arithmetic_simplify
 1.78% :     0.000003s :     12: predicate.cast_eliminate
 0.37% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000002s :     12: predicate.compare_switch_simplify
 1.32% :     0.000002s :     12: predicate.depend_value_elim
 1.27% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.50% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.46% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.08% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     12: predicate.environ_get_eliminate
 1.08% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.40% :     0.000008s :     26: predicate.inline
 0.75% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :      6: predicate.less_batch_normalization
 1.78% :     0.000003s :     14: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     17: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.04% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.71% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000002s :     12: predicate.merge_addn
 1.13% :     0.000002s :     12: predicate.minmaximum_grad
 1.61% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.04% :     0.000003s :     17: predicate.partial_eliminate
 1.26% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     12: predicate.reduce_eliminate
 1.46% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     20: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     12: predicate.reshape_eliminate
 1.37% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     12: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.84% :     0.000001s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.53% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.73% :     0.000004s :     20: predicate.switch_defer_inline
 2.16% :     0.000003s :     20: predicate.switch_layer_defer_inline
 7.96% :     0.000011s :     65: predicate.switch_simplify
 1.42% :     0.000002s :     12: predicate.tile_eliminate
 1.11% :     0.000002s :     12: predicate.transpose_eliminate
 1.67% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.16% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.90% :     0.000003s :     12: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.66% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000003s :     12: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001011    15
45.12% :     0.000456s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.13% :     0.000123s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.75% :     0.000432s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019631    72
 0.28% :     0.000055s :      1: add_recomputation
 0.46% :     0.000090s :      1: auto_monad
 0.10% :     0.000019s :      1: auto_monad_reorder
 1.55% :     0.000304s :      1: bootstrap
 0.17% :     0.000034s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.13% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
33.99% :     0.006672s :      1: jit_opt_a
 0.81% :     0.000159s :      1: jit_opt_after_cconv
 0.29% :     0.000057s :      1: jit_opt_b
 2.62% :     0.000515s :      1: loop_unroll
 3.84% :     0.000754s :      1: mutable_eliminate
 6.16% :     0.001209s :     26: opt.transform.jit_opt_a
 0.26% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.71% :     0.000532s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.37% :     0.000072s :      1: py_interpret_to_execute
 0.08% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000018s :      1: remove_dup_value
 2.82% :     0.000553s :      1: renormalize.infer
 2.16% :     0.000424s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000040s :      1: rewriter_after_opt_a
 0.47% :     0.000092s :      1: rewriter_before_opt_a
 0.45% :     0.000087s :      1: symbol_engine_optimizer
38.95% :     0.007646s :      1: type_inference


TotalTime = 0.0212884, [33]
[bootstrap]: 0.00032941
[type_inference]: 0.00496834
[event_method]: 1.604e-05
[auto_monad]: 4.255e-05
[graph_reusing]: 3.44001e-06
[pre_auto_parallel]: 1.89e-06
[py_interpret_to_execute]: 3.365e-05
[rewriter_before_opt_a]: 6.393e-05
[expand_dump_flag]: 2.36e-06
[jit_opt_a]: 0.00754177, [2]
    [Cycle 1]: 0.00353938, [27]
        [switch_simplify]: 5.275e-05
        [loop_unroll]: 2.861e-05
        [a_1]: 0.0005259
        [with_stream_mark]: 1.605e-05
        [recompute_prepare]: 7.25e-06
        [updatestate_depend_eliminate]: 2.64999e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 5.96e-06
        [updatestate_useless_node_eliminater]: 5.61e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 1.40999e-06
        [get_grad_eliminate_]: 5.27001e-06
        [merge_forward]: 2.64001e-06
        [cell_reuse_recompute_pass]: 1.11002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.195e-05
        [j_node_and_user_rematch]: 9.19e-06
        [meta_fg_expand]: 1.92001e-06
        [replace_old_param]: 8.17998e-06
        [inline_without_move]: 5.74e-06
        [renormalize]: 0.0025988
        [add_forward_monad_depend]: 6.44001e-06
        [auto_monad_grad]: 1.18001e-06
        [auto_monad_eliminator]: 1.468e-05
        [cse]: 2.033e-05
        [replace_applicator]: 1.672e-05
    [Cycle 2]: 0.00034905, [27]
        [switch_simplify]: 6.43e-06
        [loop_unroll]: 5.20999e-06
        [a_1]: 0.00010259
        [with_stream_mark]: 1.064e-05
        [recompute_prepare]: 4.89e-06
        [updatestate_depend_eliminate]: 2.61999e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 5.69999e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 6.84001e-06
        [meta_shard_fg_expand]: 1.32999e-06
        [get_grad_eliminate_]: 4.89e-06
        [merge_forward]: 2.79999e-06
        [cell_reuse_recompute_pass]: 3.20002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.389e-05
        [j_node_and_user_rematch]: 7.85e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 8.62e-06
        [inline_without_move]: 4.53001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.62999e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 5.39e-06
        [cse]: 1.112e-05
        [replace_applicator]: 4.87998e-06
[py_interpret_to_execute_after_opt_a]: 1.213e-05
[rewriter_after_opt_a]: 3.189e-05
[convert_after_rewriter]: 5.55001e-06
[order_py_execute_after_rewriter]: 3.9e-06
[mutable_eliminate]: 0.00072847
[jit_opt_b]: 5.583e-05, [1]
    [Cycle 1]: 4.911e-05, [2]
        [frontend_op_eliminate]: 2.236e-05
        [inline_after_opt_a]: 1.508e-05
[cconv]: 1.846e-05
[loop_unroll]: 0.00043314
[jit_opt_after_cconv]: 0.00014204, [1]
    [Cycle 1]: 0.00013599, [11]
        [c_1]: 2.078e-05
        [parameter_eliminate]: 3.2e-06
        [updatestate_depend_eliminate]: 4.68999e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [cse]: 2.072e-05
        [call_graph_tuple_transform]: 1.82e-05
        [tuple_list_get_item_eliminator]: 5.44e-06
        [none_parameter_eliminate]: 1.17e-06
        [renormalize]: 2.20025e-07
        [switch_simplify]: 5.67001e-06
[remove_dup_value]: 8.03999e-06
[partial_unused_args_eliminate]: 1.96003e-06
[environ_conv]: 4.41002e-06
[add_recomputation]: 3.493e-05
[cse_after_recomputation]: 3.647e-05, [1]
    [Cycle 1]: 1.585e-05, [1]
        [cse]: 8.99e-06
[auto_monad_reorder]: 1.091e-05
[get_jit_bprop_graph]: 1.89e-06
[rewriter_after_jit_bprop_graph]: 6.59999e-06
[opt_after_jit_grad]: 0.00051989
[symbol_engine_optimizer]: 7.858e-05, [1]
    [Cycle 1]: 7.17e-05, [6]
        [build]: 4.06001e-06
        [elim_shapecalc]: 8.13999e-06
        [elim_not_effective]: 1.479e-05
        [opt_reshape]: 6.65998e-06
        [fold_const_symbol]: 9.00001e-06
        [renormalize]: 6.29982e-07
[validate]: 3.15e-05
[backend_pass]: 6.89994e-07
[task_emit]: 0.00591123
[execute]: 8.92e-06
Sums
    bootstrap                            :     0.000329s :  1.94%
    type_inference                       :     0.004968s : 29.23%
    event_method                         :     0.000016s :  0.09%
    auto_monad                           :     0.000043s :  0.25%
    graph_reusing                        :     0.000003s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000034s :  0.20%
    rewriter_before_opt_a                :     0.000064s :  0.38%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000059s :  0.35%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000628s :  3.70%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.002599s : 15.29%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000031s :  0.19%
    jit_opt_a.replace_applicator         :     0.000022s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000728s :  4.29%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000018s :  0.11%
    loop_unroll                          :     0.000433s :  2.55%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000008s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000035s :  0.21%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000011s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000520s :  3.06%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.19%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005911s : 34.77%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000182    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.54% :     0.000005s :      3: substitution.graph_param_transform
83.63% :     0.000152s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.73% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.47% :     0.000005s :      2: substitution.replace_old_param
 5.40% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004907     2
82.48% :     0.004048s :      1: type_inference.infer
17.52% :     0.000860s :      1: type_inference.specialize
------[replace.]   0.000069     6
66.58% :     0.000046s :      4: replace.inline
33.42% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000158     6
94.49% :     0.000150s :      4: match.inline
 5.51% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.51% :     0.000002s :     10: predicate.addn_zero_filter
 2.47% :     0.000003s :     10: predicate.arithmetic_simplify
 1.72% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000002s :     10: predicate.depend_value_elim
 1.22% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.55% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.77% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.37% :     0.000002s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.23% :     0.000002s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 4.73% :     0.000006s :     22: predicate.inline
 1.08% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.47% :     0.000002s :      6: predicate.less_batch_normalization
 1.95% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.79% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 1.93% :     0.000002s :     15: predicate.partial_eliminate
 1.25% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.00% :     0.000002s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.42% :     0.000003s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.58% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.19% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.39% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.21% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.82% :     0.000008s :     53: predicate.switch_simplify
 1.50% :     0.000002s :     10: predicate.tile_eliminate
 1.22% :     0.000002s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.06% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.31% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.96% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.35% :     0.000003s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000665    12
43.94% :     0.000292s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.87% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.18% :     0.000340s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024837    76
 0.15% :     0.000038s :      1: add_recomputation
 0.18% :     0.000046s :      1: auto_monad
 0.06% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.40% :     0.000347s :      1: bootstrap
 0.09% :     0.000021s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.16% :     0.000039s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000020s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000006s :      1: graph_reusing
30.38% :     0.007547s :      1: jit_opt_a
 0.58% :     0.000144s :      1: jit_opt_after_cconv
 0.23% :     0.000058s :      1: jit_opt_b
 1.77% :     0.000441s :      1: loop_unroll
 2.97% :     0.000737s :      1: mutable_eliminate
 3.40% :     0.000844s :     26: opt.transform.jit_opt_a
 0.19% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000030s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.12% :     0.000527s :      1: opt_after_jit_grad
 0.02% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000010s :      1: remove_dup_value
 8.56% :     0.002127s :      1: renormalize.infer
 1.86% :     0.000462s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000035s :      1: rewriter_after_opt_a
 0.27% :     0.000067s :      1: rewriter_before_opt_a
 0.33% :     0.000081s :      1: symbol_engine_optimizer
23.86% :     0.005926s :      1: task_emit
20.05% :     0.004979s :      1: type_inference
 0.20% :     0.000049s :      1: validate


TotalTime = 0.0191248, [33]
[bootstrap]: 0.00029008
[type_inference]: 0.00463139
[event_method]: 1.72e-05
[auto_monad]: 6.363e-05
[graph_reusing]: 6.27001e-06
[pre_auto_parallel]: 2.39001e-06
[py_interpret_to_execute]: 3.087e-05
[rewriter_before_opt_a]: 6.918e-05
[expand_dump_flag]: 2.96001e-06
[jit_opt_a]: 0.0055013, [2]
    [Cycle 1]: 0.00160774, [27]
        [switch_simplify]: 5.891e-05
        [loop_unroll]: 3.086e-05
        [a_1]: 0.0005226
        [with_stream_mark]: 1.737e-05
        [recompute_prepare]: 8.42998e-06
        [updatestate_depend_eliminate]: 3.53e-06
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 2.67001e-06
        [parameter_eliminate]: 2.24001e-06
        [specialize_transform]: 5.92999e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.58002e-06
        [meta_shard_fg_expand]: 2.32001e-06
        [get_grad_eliminate_]: 5.20001e-06
        [merge_forward]: 3.86999e-06
        [cell_reuse_recompute_pass]: 1.71998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.527e-05
        [j_node_and_user_rematch]: 8.49002e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 8.64e-06
        [inline_without_move]: 5.14e-06
        [renormalize]: 0.0006367
        [add_forward_monad_depend]: 6.89999e-06
        [auto_monad_grad]: 3.11001e-06
        [auto_monad_eliminator]: 1.562e-05
        [cse]: 2.562e-05
        [replace_applicator]: 1.519e-05
    [Cycle 2]: 0.00032487, [27]
        [switch_simplify]: 5.71003e-06
        [loop_unroll]: 4.75001e-06
        [a_1]: 8.819e-05
        [with_stream_mark]: 9.19e-06
        [recompute_prepare]: 5.15999e-06
        [updatestate_depend_eliminate]: 2.56e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 4.79e-06
        [updatestate_useless_node_eliminater]: 5.06997e-06
        [accelerated_algorithm]: 5.12e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 3.14001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.309e-05
        [j_node_and_user_rematch]: 7.56001e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 7.61001e-06
        [inline_without_move]: 4.46002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.45999e-06
        [auto_monad_grad]: 1.04003e-06
        [auto_monad_eliminator]: 7.3e-06
        [cse]: 1.18e-05
        [replace_applicator]: 5.26002e-06
[py_interpret_to_execute_after_opt_a]: 1.286e-05
[rewriter_after_opt_a]: 3.498e-05
[convert_after_rewriter]: 6.96001e-06
[order_py_execute_after_rewriter]: 5.32001e-06
[mutable_eliminate]: 0.00061791
[jit_opt_b]: 4.757e-05, [1]
    [Cycle 1]: 4.04e-05, [2]
        [frontend_op_eliminate]: 1.549e-05
        [inline_after_opt_a]: 1.424e-05
[cconv]: 2.622e-05
[loop_unroll]: 0.00046376
[jit_opt_after_cconv]: 0.00014175, [1]
    [Cycle 1]: 0.00013462, [11]
        [c_1]: 1.955e-05
        [parameter_eliminate]: 3.29001e-06
        [updatestate_depend_eliminate]: 6.16e-06
        [updatestate_assign_eliminate]: 2.49999e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [cse]: 1.944e-05
        [call_graph_tuple_transform]: 1.842e-05
        [tuple_list_get_item_eliminator]: 5.00001e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 8.2e-07
        [switch_simplify]: 4.79998e-06
[remove_dup_value]: 1.39e-05
[partial_unused_args_eliminate]: 2.11e-06
[environ_conv]: 4.65001e-06
[add_recomputation]: 4.99e-05
[cse_after_recomputation]: 2.184e-05, [1]
    [Cycle 1]: 1.601e-05, [1]
        [cse]: 9.74e-06
[auto_monad_reorder]: 1.583e-05
[get_jit_bprop_graph]: 1.42999e-06
[rewriter_after_jit_bprop_graph]: 6.23998e-06
[opt_after_jit_grad]: 0.00051664
[symbol_engine_optimizer]: 7.314e-05, [1]
    [Cycle 1]: 6.61e-05, [6]
        [build]: 3.76001e-06
        [elim_shapecalc]: 7.58001e-06
        [elim_not_effective]: 1.321e-05
        [opt_reshape]: 5.36002e-06
        [fold_const_symbol]: 8.32e-06
        [renormalize]: 5.50004e-07
[validate]: 3.422e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00620692
[execute]: 8.56002e-06
Sums
    bootstrap                            :     0.000290s :  1.94%
    type_inference                       :     0.004631s : 31.01%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000064s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.21%
    rewriter_before_opt_a                :     0.000069s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.43%
    jit_opt_a.loop_unroll                :     0.000036s :  0.24%
    jit_opt_a.a_1                        :     0.000611s :  4.09%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000637s :  4.26%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.15%
    jit_opt_a.cse                        :     0.000037s :  0.25%
    jit_opt_a.replace_applicator         :     0.000020s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.09%
    rewriter_after_opt_a                 :     0.000035s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000618s :  4.14%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000026s :  0.18%
    loop_unroll                          :     0.000464s :  3.11%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000016s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000517s :  3.46%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006207s : 41.56%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000197    23
 1.13% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.68% :     0.000005s :      3: substitution.graph_param_transform
83.45% :     0.000164s :      4: substitution.inline
 1.52% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.36% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000004s :      2: substitution.replace_old_param
 6.09% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004565     2
83.15% :     0.003796s :      1: type_inference.infer
16.85% :     0.000769s :      1: type_inference.specialize
------[replace.]   0.000065     6
67.33% :     0.000044s :      4: replace.inline
32.67% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000173     6
93.72% :     0.000162s :      4: match.inline
 6.28% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.74% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.30% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 2.11% :     0.000002s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.76% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.65% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.00% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.07% :     0.000001s :      6: predicate.less_batch_normalization
 1.37% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.56% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.17% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.36% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.31% :     0.000002s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 2.01% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.77% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.26% :     0.000001s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.55% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.82% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.23% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.89% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000533    12
51.08% :     0.000272s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.43% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.49% :     0.000232s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020699    76
 0.26% :     0.000053s :      1: add_recomputation
 0.32% :     0.000067s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.46% :     0.000302s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.59% :     0.005504s :      1: jit_opt_a
 0.70% :     0.000144s :      1: jit_opt_after_cconv
 0.24% :     0.000050s :      1: jit_opt_b
 2.29% :     0.000474s :      1: loop_unroll
 3.03% :     0.000628s :      1: mutable_eliminate
 4.00% :     0.000828s :     26: opt.transform.jit_opt_a
 0.22% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000023s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.54% :     0.000526s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000034s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.62% :     0.000335s :      1: renormalize.infer
 1.42% :     0.000294s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000038s :      1: rewriter_after_opt_a
 0.35% :     0.000072s :      1: rewriter_before_opt_a
 0.37% :     0.000076s :      1: symbol_engine_optimizer
30.05% :     0.006220s :      1: task_emit
22.46% :     0.004648s :      1: type_inference
 0.26% :     0.000054s :      1: validate


TotalTime = 0.0187167, [33]
[bootstrap]: 0.00030041
[type_inference]: 0.00470714
[event_method]: 1.761e-05
[auto_monad]: 5.498e-05
[graph_reusing]: 4.65999e-06
[pre_auto_parallel]: 1.97001e-06
[py_interpret_to_execute]: 3.289e-05
[rewriter_before_opt_a]: 6.765e-05
[expand_dump_flag]: 2.37001e-06
[jit_opt_a]: 0.0055439, [2]
    [Cycle 1]: 0.00172597, [27]
        [switch_simplify]: 5.53e-05
        [loop_unroll]: 3.088e-05
        [a_1]: 0.00057401
        [with_stream_mark]: 1.482e-05
        [recompute_prepare]: 7.17002e-06
        [updatestate_depend_eliminate]: 3.63e-06
        [updatestate_assign_eliminate]: 2.56998e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 6.46e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 6.17001e-06
        [meta_shard_fg_expand]: 1.30001e-06
        [get_grad_eliminate_]: 5.52999e-06
        [merge_forward]: 2.79999e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.141e-05
        [j_node_and_user_rematch]: 9.57001e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 8.27998e-06
        [inline_without_move]: 5.39998e-06
        [renormalize]: 0.00073589
        [add_forward_monad_depend]: 5.87999e-06
        [auto_monad_grad]: 2.29001e-06
        [auto_monad_eliminator]: 1.321e-05
        [cse]: 2.289e-05
        [replace_applicator]: 1.291e-05
    [Cycle 2]: 0.00036539, [27]
        [switch_simplify]: 6.17999e-06
        [loop_unroll]: 5.22e-06
        [a_1]: 9.844e-05
        [with_stream_mark]: 1.24e-05
        [recompute_prepare]: 5.49e-06
        [updatestate_depend_eliminate]: 2.66999e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.16002e-06
        [specialize_transform]: 5.67001e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 5.74e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 5.35999e-06
        [merge_forward]: 2.77002e-06
        [cell_reuse_recompute_pass]: 2.49999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.583e-05
        [j_node_and_user_rematch]: 8.48999e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 9.87999e-06
        [inline_without_move]: 5.42999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 1.54e-06
        [auto_monad_eliminator]: 6.52001e-06
        [cse]: 1.327e-05
        [replace_applicator]: 7.05e-06
[py_interpret_to_execute_after_opt_a]: 1.183e-05
[rewriter_after_opt_a]: 3.359e-05
[convert_after_rewriter]: 6.19001e-06
[order_py_execute_after_rewriter]: 5.19e-06
[mutable_eliminate]: 0.00061008
[jit_opt_b]: 5.16e-05, [1]
    [Cycle 1]: 4.508e-05, [2]
        [frontend_op_eliminate]: 1.765e-05
        [inline_after_opt_a]: 1.575e-05
[cconv]: 2.317e-05
[loop_unroll]: 0.00044648
[jit_opt_after_cconv]: 0.00014439, [1]
    [Cycle 1]: 0.00013766, [11]
        [c_1]: 2.222e-05
        [parameter_eliminate]: 2.99999e-06
        [updatestate_depend_eliminate]: 5.25999e-06
        [updatestate_assign_eliminate]: 3.26001e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 2.072e-05
        [call_graph_tuple_transform]: 1.764e-05
        [tuple_list_get_item_eliminator]: 5.51e-06
        [none_parameter_eliminate]: 1.34998e-06
        [renormalize]: 4.40021e-07
        [switch_simplify]: 5.77001e-06
[remove_dup_value]: 9.42001e-06
[partial_unused_args_eliminate]: 1.99e-06
[environ_conv]: 4.34002e-06
[add_recomputation]: 3.885e-05
[cse_after_recomputation]: 2.062e-05, [1]
    [Cycle 1]: 1.527e-05, [1]
        [cse]: 9.20999e-06
[auto_monad_reorder]: 1.177e-05
[get_jit_bprop_graph]: 1.60001e-06
[rewriter_after_jit_bprop_graph]: 5.34998e-06
[opt_after_jit_grad]: 0.00047466
[symbol_engine_optimizer]: 7.431e-05, [1]
    [Cycle 1]: 6.809e-05, [6]
        [build]: 3.54002e-06
        [elim_shapecalc]: 8.20999e-06
        [elim_not_effective]: 1.3e-05
        [opt_reshape]: 6.17001e-06
        [fold_const_symbol]: 8.48999e-06
        [renormalize]: 5.50004e-07
[validate]: 4.834e-05
[backend_pass]: 7.00005e-07
[task_emit]: 0.0057406
[execute]: 9.10001e-06
Sums
    bootstrap                            :     0.000300s :  2.05%
    type_inference                       :     0.004707s : 32.18%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000055s :  0.38%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000033s :  0.22%
    rewriter_before_opt_a                :     0.000068s :  0.46%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.42%
    jit_opt_a.loop_unroll                :     0.000036s :  0.25%
    jit_opt_a.a_1                        :     0.000672s :  4.60%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000736s :  5.03%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000036s :  0.25%
    jit_opt_a.replace_applicator         :     0.000020s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000034s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000610s :  4.17%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000023s :  0.16%
    loop_unroll                          :     0.000446s :  3.05%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000009s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000039s :  0.27%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000475s :  3.24%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000048s :  0.33%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005741s : 39.24%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000196    23
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.22% :     0.000004s :      3: substitution.graph_param_transform
84.66% :     0.000166s :      4: substitution.inline
 1.94% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.35% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.21% :     0.000004s :      2: substitution.replace_old_param
 5.11% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004646     2
80.44% :     0.003738s :      1: type_inference.infer
19.56% :     0.000909s :      1: type_inference.specialize
------[replace.]   0.000071     6
67.96% :     0.000048s :      4: replace.inline
32.04% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000172     6
94.85% :     0.000163s :      4: match.inline
 5.15% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000150   691
 1.21% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.99% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.87% :     0.000001s :     10: predicate.addn_check_dump
 0.99% :     0.000001s :     10: predicate.addn_zero_filter
 1.67% :     0.000003s :     10: predicate.arithmetic_simplify
 1.12% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.88% :     0.000001s :     10: predicate.compare_switch_simplify
 1.06% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.08% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.98% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.64% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.94% :     0.000001s :     10: predicate.environ_get_eliminate
 0.87% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.77% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 4.79% :     0.000007s :     22: predicate.inline
 1.07% :     0.000002s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.49% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.23% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.75% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.89% :     0.000001s :     10: predicate.merge_addn
 0.91% :     0.000001s :     10: predicate.minmaximum_grad
 1.18% :     0.000002s :      3: predicate.mutable_eliminate
 0.39% :     0.000001s :      3: predicate.opt_reshape
 1.69% :     0.000003s :     15: predicate.partial_eliminate
19.28% :     0.000029s :     10: predicate.print_const_string_wrapper
 1.33% :     0.000002s :     10: predicate.reduce_eliminate
 1.13% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.31% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.40% :     0.000001s :      3: predicate.reset_defer_inline
 1.03% :     0.000002s :     10: predicate.reshape_eliminate
 0.99% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.59% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.00% :     0.000002s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000002s :      6: predicate.specialize_transform
 1.05% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.34% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 1.91% :     0.000003s :     16: predicate.switch_defer_inline
 1.93% :     0.000003s :     16: predicate.switch_layer_defer_inline
 5.78% :     0.000009s :     53: predicate.switch_simplify
 0.97% :     0.000001s :     10: predicate.tile_eliminate
 1.12% :     0.000002s :     10: predicate.transpose_eliminate
 1.12% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.00% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 2.75% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.23% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.30% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.44% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.34% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000614    12
45.32% :     0.000278s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.24% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.44% :     0.000304s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020449    76
 0.20% :     0.000042s :      1: add_recomputation
 0.28% :     0.000058s :      1: auto_monad
 0.07% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.54% :     0.000316s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
27.13% :     0.005548s :      1: jit_opt_a
 0.72% :     0.000147s :      1: jit_opt_after_cconv
 0.26% :     0.000054s :      1: jit_opt_b
 2.22% :     0.000454s :      1: loop_unroll
 3.03% :     0.000620s :      1: mutable_eliminate
 4.38% :     0.000896s :     26: opt.transform.jit_opt_a
 0.23% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.36% :     0.000483s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000036s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000012s :      1: remove_dup_value
 1.77% :     0.000363s :      1: renormalize.infer
 1.79% :     0.000366s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000037s :      1: rewriter_after_opt_a
 0.34% :     0.000070s :      1: rewriter_before_opt_a
 0.38% :     0.000077s :      1: symbol_engine_optimizer
28.16% :     0.005758s :      1: task_emit
23.08% :     0.004719s :      1: type_inference
 0.34% :     0.000069s :      1: validate


TotalTime = 0.0213453, [33]
[bootstrap]: 0.0003361
[type_inference]: 0.00515656
[event_method]: 2.031e-05
[auto_monad]: 7.045e-05
[graph_reusing]: 6.83e-06
[pre_auto_parallel]: 2.54999e-06
[py_interpret_to_execute]: 4.06e-05
[rewriter_before_opt_a]: 9.543e-05
[expand_dump_flag]: 3.24001e-06
[jit_opt_a]: 0.00635043, [2]
    [Cycle 1]: 0.00194717, [27]
        [switch_simplify]: 6.818e-05
        [loop_unroll]: 3.298e-05
        [a_1]: 0.00061747
        [with_stream_mark]: 2.204e-05
        [recompute_prepare]: 9.54e-06
        [updatestate_depend_eliminate]: 3.95e-06
        [updatestate_assign_eliminate]: 3.83001e-06
        [updatestate_loads_eliminate]: 3.03e-06
        [parameter_eliminate]: 2.16e-06
        [specialize_transform]: 6.28998e-06
        [updatestate_useless_node_eliminater]: 5.34998e-06
        [accelerated_algorithm]: 6.28e-06
        [meta_shard_fg_expand]: 3.06999e-06
        [get_grad_eliminate_]: 5.34998e-06
        [merge_forward]: 3.81999e-06
        [cell_reuse_recompute_pass]: 2.53e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.586e-05
        [j_node_and_user_rematch]: 9.73002e-06
        [meta_fg_expand]: 2.64001e-06
        [replace_old_param]: 9.76998e-06
        [inline_without_move]: 5.15999e-06
        [renormalize]: 0.00085056
        [add_forward_monad_depend]: 6.96001e-06
        [auto_monad_grad]: 2.24999e-06
        [auto_monad_eliminator]: 1.457e-05
        [cse]: 3.096e-05
        [replace_applicator]: 1.631e-05
    [Cycle 2]: 0.00034749, [27]
        [switch_simplify]: 6.31e-06
        [loop_unroll]: 4.99998e-06
        [a_1]: 9.31e-05
        [with_stream_mark]: 1.192e-05
        [recompute_prepare]: 5.91e-06
        [updatestate_depend_eliminate]: 2.60002e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.23002e-06
        [specialize_transform]: 5.35999e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.43002e-06
        [meta_shard_fg_expand]: 1.76998e-06
        [get_grad_eliminate_]: 4.58999e-06
        [merge_forward]: 2.94001e-06
        [cell_reuse_recompute_pass]: 3.26001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.465e-05
        [j_node_and_user_rematch]: 8.88002e-06
        [meta_fg_expand]: 1.76e-06
        [replace_old_param]: 8.61002e-06
        [inline_without_move]: 5.67999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.21998e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 5.69e-06
        [cse]: 1.268e-05
        [replace_applicator]: 5.20001e-06
[py_interpret_to_execute_after_opt_a]: 1.335e-05
[rewriter_after_opt_a]: 3.839e-05
[convert_after_rewriter]: 7.41001e-06
[order_py_execute_after_rewriter]: 5.88002e-06
[mutable_eliminate]: 0.00085477
[jit_opt_b]: 5.955e-05, [1]
    [Cycle 1]: 5.029e-05, [2]
        [frontend_op_eliminate]: 1.939e-05
        [inline_after_opt_a]: 1.776e-05
[cconv]: 2.699e-05
[loop_unroll]: 0.00049472
[jit_opt_after_cconv]: 0.00018568, [1]
    [Cycle 1]: 0.00017837, [11]
        [c_1]: 2.454e-05
        [parameter_eliminate]: 5.09e-06
        [updatestate_depend_eliminate]: 7.83001e-06
        [updatestate_assign_eliminate]: 3.04001e-06
        [updatestate_loads_eliminate]: 2.32001e-06
        [cse]: 2.99e-05
        [call_graph_tuple_transform]: 2.494e-05
        [tuple_list_get_item_eliminator]: 5.29e-06
        [none_parameter_eliminate]: 1.91e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 5.25001e-06
[remove_dup_value]: 1.617e-05
[partial_unused_args_eliminate]: 2.41998e-06
[environ_conv]: 6.36998e-06
[add_recomputation]: 5.679e-05
[cse_after_recomputation]: 2.268e-05, [1]
    [Cycle 1]: 1.583e-05, [1]
        [cse]: 9.77999e-06
[auto_monad_reorder]: 1.784e-05
[get_jit_bprop_graph]: 2.63998e-06
[rewriter_after_jit_bprop_graph]: 1.006e-05
[opt_after_jit_grad]: 0.00057841
[symbol_engine_optimizer]: 8.3e-05, [1]
    [Cycle 1]: 7.552e-05, [6]
        [build]: 4.67e-06
        [elim_shapecalc]: 8.15e-06
        [elim_not_effective]: 1.557e-05
        [opt_reshape]: 7.08e-06
        [fold_const_symbol]: 9.35001e-06
        [renormalize]: 1.16002e-06
[validate]: 4.796e-05
[backend_pass]: 1.13001e-06
[task_emit]: 0.00648684
[execute]: 8.05999e-06
Sums
    bootstrap                            :     0.000336s :  2.03%
    type_inference                       :     0.005157s : 31.07%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000070s :  0.42%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000041s :  0.24%
    rewriter_before_opt_a                :     0.000095s :  0.57%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.45%
    jit_opt_a.loop_unroll                :     0.000038s :  0.23%
    jit_opt_a.a_1                        :     0.000711s :  4.28%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000851s :  5.13%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000044s :  0.26%
    jit_opt_a.replace_applicator         :     0.000022s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000855s :  5.15%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.11%
    cconv                                :     0.000027s :  0.16%
    loop_unroll                          :     0.000495s :  2.98%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000030s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000057s :  0.34%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000018s :  0.11%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.06%
    opt_after_jit_grad                   :     0.000578s :  3.48%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000048s :  0.29%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006487s : 39.08%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000254    23
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000002s :      2: substitution.fold_const_symbol
 2.53% :     0.000006s :      3: substitution.graph_param_transform
84.19% :     0.000214s :      4: substitution.inline
 1.41% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.92% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.40% :     0.000006s :      2: substitution.replace_old_param
 5.92% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005083     2
82.09% :     0.004173s :      1: type_inference.infer
17.91% :     0.000910s :      1: type_inference.specialize
------[replace.]   0.000080     6
64.02% :     0.000051s :      4: replace.inline
35.98% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000225     6
93.76% :     0.000211s :      4: match.inline
 6.24% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.32% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.87% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.21% :     0.000002s :     10: predicate.addn_zero_filter
 3.02% :     0.000004s :     10: predicate.arithmetic_simplify
 1.19% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.59% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.26% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.41% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.02% :     0.000001s :     10: predicate.environ_get_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.83% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.38% :     0.000007s :     22: predicate.inline
 1.27% :     0.000002s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.52% :     0.000002s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.49% :     0.000002s :     15: predicate.load_eliminater
 1.79% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.80% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.61% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.15% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 1.95% :     0.000003s :     15: predicate.partial_eliminate
 1.02% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.38% :     0.000002s :     10: predicate.reduce_eliminate
 1.23% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.03% :     0.000003s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.13% :     0.000001s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.06% :     0.000001s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.16% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.62% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.98% :     0.000010s :     53: predicate.switch_simplify
 1.58% :     0.000002s :     10: predicate.tile_eliminate
 1.07% :     0.000001s :     10: predicate.transpose_eliminate
 1.52% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.48% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.22% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.37% :     0.000003s :     10: predicate.value_based_eliminate
 0.48% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000653    12
48.31% :     0.000316s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.50% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.20% :     0.000302s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023269    76
 0.26% :     0.000060s :      1: add_recomputation
 0.32% :     0.000074s :      1: auto_monad
 0.09% :     0.000021s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.53% :     0.000357s :      1: bootstrap
 0.13% :     0.000031s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.11% :     0.000026s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
27.31% :     0.006354s :      1: jit_opt_a
 0.81% :     0.000189s :      1: jit_opt_after_cconv
 0.27% :     0.000063s :      1: jit_opt_b
 2.17% :     0.000504s :      1: loop_unroll
 3.73% :     0.000867s :      1: mutable_eliminate
 4.08% :     0.000949s :     26: opt.transform.jit_opt_a
 0.24% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000029s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.53% :     0.000588s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000043s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000019s :      1: remove_dup_value
 1.95% :     0.000453s :      1: renormalize.infer
 1.67% :     0.000389s :      1: renormalize.specialize
 0.05% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000042s :      1: rewriter_after_opt_a
 0.42% :     0.000099s :      1: rewriter_before_opt_a
 0.37% :     0.000086s :      1: symbol_engine_optimizer
27.94% :     0.006501s :      1: task_emit
22.23% :     0.005173s :      1: type_inference
 0.31% :     0.000072s :      1: validate


TotalTime = 0.0178679, [33]
[bootstrap]: 0.00032748
[type_inference]: 0.00476465
[event_method]: 1.651e-05
[auto_monad]: 6.043e-05
[graph_reusing]: 5.99e-06
[pre_auto_parallel]: 2.51e-06
[py_interpret_to_execute]: 3.014e-05
[rewriter_before_opt_a]: 7.082e-05
[expand_dump_flag]: 2.88e-06
[jit_opt_a]: 0.00483854, [2]
    [Cycle 1]: 0.00139042, [27]
        [switch_simplify]: 4.98e-05
        [loop_unroll]: 2.756e-05
        [a_1]: 0.00048457
        [with_stream_mark]: 1.67e-05
        [recompute_prepare]: 7.01999e-06
        [updatestate_depend_eliminate]: 3.35e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.87001e-06
        [specialize_transform]: 6.17001e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.25001e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 4.97e-06
        [merge_forward]: 3.96001e-06
        [cell_reuse_recompute_pass]: 1.20999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.317e-05
        [j_node_and_user_rematch]: 8.23999e-06
        [meta_fg_expand]: 2.01e-06
        [replace_old_param]: 8.45001e-06
        [inline_without_move]: 5.11002e-06
        [renormalize]: 0.00051032
        [add_forward_monad_depend]: 5.16998e-06
        [auto_monad_grad]: 2.22999e-06
        [auto_monad_eliminator]: 1.221e-05
        [cse]: 2.5e-05
        [replace_applicator]: 1.139e-05
    [Cycle 2]: 0.00030019, [27]
        [switch_simplify]: 5.21998e-06
        [loop_unroll]: 4.66002e-06
        [a_1]: 8.608e-05
        [with_stream_mark]: 7.73999e-06
        [recompute_prepare]: 4.84e-06
        [updatestate_depend_eliminate]: 2.37001e-06
        [updatestate_assign_eliminate]: 1.98997e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 4.92e-06
        [updatestate_useless_node_eliminater]: 4.78001e-06
        [accelerated_algorithm]: 4.56002e-06
        [meta_shard_fg_expand]: 9.99979e-07
        [get_grad_eliminate_]: 4.45e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 2.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.178e-05
        [j_node_and_user_rematch]: 7.85e-06
        [meta_fg_expand]: 1.70001e-06
        [replace_old_param]: 6.84001e-06
        [inline_without_move]: 4.58001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.02998e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 4.81002e-06
        [cse]: 8.85999e-06
        [replace_applicator]: 4.87e-06
[py_interpret_to_execute_after_opt_a]: 8.3e-06
[rewriter_after_opt_a]: 3.041e-05
[convert_after_rewriter]: 6.42001e-06
[order_py_execute_after_rewriter]: 4.64002e-06
[mutable_eliminate]: 0.00048532
[jit_opt_b]: 4.619e-05, [1]
    [Cycle 1]: 3.931e-05, [2]
        [frontend_op_eliminate]: 1.496e-05
        [inline_after_opt_a]: 1.385e-05
[cconv]: 2.414e-05
[loop_unroll]: 0.00040969
[jit_opt_after_cconv]: 0.00013238, [1]
    [Cycle 1]: 0.00012586, [11]
        [c_1]: 2.002e-05
        [parameter_eliminate]: 2.57001e-06
        [updatestate_depend_eliminate]: 4.68999e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 1.662e-05
        [call_graph_tuple_transform]: 1.705e-05
        [tuple_list_get_item_eliminator]: 5.21998e-06
        [none_parameter_eliminate]: 1.43002e-06
        [renormalize]: 5.10016e-07
        [switch_simplify]: 4.94e-06
[remove_dup_value]: 1.303e-05
[partial_unused_args_eliminate]: 2.04e-06
[environ_conv]: 4.67e-06
[add_recomputation]: 4.494e-05
[cse_after_recomputation]: 1.932e-05, [1]
    [Cycle 1]: 1.436e-05, [1]
        [cse]: 8.47998e-06
[auto_monad_reorder]: 1.445e-05
[get_jit_bprop_graph]: 2.17001e-06
[rewriter_after_jit_bprop_graph]: 4.90999e-06
[opt_after_jit_grad]: 0.00043533
[symbol_engine_optimizer]: 6.841e-05, [1]
    [Cycle 1]: 6.266e-05, [6]
        [build]: 3.01999e-06
        [elim_shapecalc]: 7.64002e-06
        [elim_not_effective]: 1.16e-05
        [opt_reshape]: 5.44998e-06
        [fold_const_symbol]: 8.05e-06
        [renormalize]: 3.69997e-07
[validate]: 2.99e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00576906
[execute]: 7.8e-06
Sums
    bootstrap                            :     0.000327s :  2.31%
    type_inference                       :     0.004765s : 33.67%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000060s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.21%
    rewriter_before_opt_a                :     0.000071s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.39%
    jit_opt_a.loop_unroll                :     0.000032s :  0.23%
    jit_opt_a.a_1                        :     0.000571s :  4.03%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000510s :  3.61%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.12%
    jit_opt_a.cse                        :     0.000034s :  0.24%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000030s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000485s :  3.43%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000410s :  2.89%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000045s :  0.32%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000435s :  3.08%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000030s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005769s : 40.77%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000173    23
 1.21% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.97% :     0.000005s :      3: substitution.graph_param_transform
81.96% :     0.000142s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.43% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.11% :     0.000004s :      2: substitution.replace_old_param
 7.05% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004700     2
84.02% :     0.003949s :      1: type_inference.infer
15.98% :     0.000751s :      1: type_inference.specialize
------[replace.]   0.000060     6
67.52% :     0.000040s :      4: replace.inline
32.48% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000151     6
92.63% :     0.000140s :      4: match.inline
 7.37% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.73% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 1.67% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.51% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.41% :     0.000000s :      3: predicate.graph_param_transform
 5.19% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.66% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.99% :     0.000002s :     15: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.47% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000000s :      3: predicate.opt_reshape
 2.34% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.81% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 1.20% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.23% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.74% :     0.000008s :     53: predicate.switch_simplify
 1.24% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.16% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000544    12
54.59% :     0.000297s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.80% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.62% :     0.000221s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019242    76
 0.25% :     0.000048s :      1: add_recomputation
 0.33% :     0.000064s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.80% :     0.000346s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.16% :     0.004841s :      1: jit_opt_a
 0.70% :     0.000135s :      1: jit_opt_after_cconv
 0.26% :     0.000049s :      1: jit_opt_b
 2.17% :     0.000418s :      1: loop_unroll
 2.56% :     0.000493s :      1: mutable_eliminate
 3.98% :     0.000766s :     26: opt.transform.jit_opt_a
 0.23% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.30% :     0.000443s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.42% :     0.000274s :      1: renormalize.infer
 1.19% :     0.000229s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000033s :      1: rewriter_after_opt_a
 0.38% :     0.000073s :      1: rewriter_before_opt_a
 0.37% :     0.000071s :      1: symbol_engine_optimizer
30.04% :     0.005780s :      1: task_emit
24.84% :     0.004780s :      1: type_inference
 0.24% :     0.000047s :      1: validate


TotalTime = 0.014536, [30]
[bootstrap]: 0.00025951
[type_inference]: 0.00645437
[event_method]: 1.982e-05
[auto_monad]: 6.467e-05
[graph_reusing]: 3.65e-06
[pre_auto_parallel]: 1.95001e-06
[py_interpret_to_execute]: 5.282e-05
[rewriter_before_opt_a]: 7.394e-05
[expand_dump_flag]: 1.78002e-06
[jit_opt_a]: 0.00550842, [2]
    [Cycle 1]: 0.00182725, [27]
        [switch_simplify]: 0.00010699
        [loop_unroll]: 3.49e-05
        [a_1]: 0.00066935
        [with_stream_mark]: 1.011e-05
        [recompute_prepare]: 6.84999e-06
        [updatestate_depend_eliminate]: 2.64999e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 9.10019e-07
        [specialize_transform]: 5.93002e-06
        [updatestate_useless_node_eliminater]: 5.76e-06
        [accelerated_algorithm]: 6.23e-06
        [meta_shard_fg_expand]: 1.15999e-06
        [get_grad_eliminate_]: 5.72999e-06
        [merge_forward]: 2.31e-06
        [cell_reuse_recompute_pass]: 9.89996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.105e-05
        [j_node_and_user_rematch]: 9.32001e-06
        [meta_fg_expand]: 1.82999e-06
        [replace_old_param]: 8.72e-06
        [inline_without_move]: 5.84e-06
        [renormalize]: 0.00073434
        [add_forward_monad_depend]: 3.10002e-06
        [auto_monad_grad]: 1.12999e-06
        [auto_monad_eliminator]: 8.12e-06
        [cse]: 1.439e-05
        [replace_applicator]: 1.027e-05
    [Cycle 2]: 0.00031117, [27]
        [switch_simplify]: 5.61e-06
        [loop_unroll]: 5.07999e-06
        [a_1]: 9.182e-05
        [with_stream_mark]: 7.58999e-06
        [recompute_prepare]: 5.18002e-06
        [updatestate_depend_eliminate]: 2.30002e-06
        [updatestate_assign_eliminate]: 2.05002e-06
        [updatestate_loads_eliminate]: 1.72001e-06
        [parameter_eliminate]: 7.99977e-07
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 4.85001e-06
        [meta_shard_fg_expand]: 9.70002e-07
        [get_grad_eliminate_]: 4.84e-06
        [merge_forward]: 2.18998e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.1e-05
        [j_node_and_user_rematch]: 8.12e-06
        [meta_fg_expand]: 1.53002e-06
        [replace_old_param]: 6.75998e-06
        [inline_without_move]: 4.97e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.01997e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 4.1e-06
        [cse]: 9.84999e-06
        [replace_applicator]: 5.07999e-06
[py_interpret_to_execute_after_opt_a]: 1.155e-05
[rewriter_after_opt_a]: 3.355e-05
[convert_after_rewriter]: 6.71999e-06
[order_py_execute_after_rewriter]: 5.25999e-06
[mutable_eliminate]: 0.0005403
[jit_opt_b]: 5.02e-05, [1]
    [Cycle 1]: 4.277e-05, [2]
        [frontend_op_eliminate]: 1.674e-05
        [inline_after_opt_a]: 1.547e-05
[cconv]: 2.18e-05
[loop_unroll]: 0.00041594
[jit_opt_after_cconv]: 0.00013971, [1]
    [Cycle 1]: 0.00013363, [11]
        [c_1]: 2.108e-05
        [parameter_eliminate]: 2.88e-06
        [updatestate_depend_eliminate]: 6.33e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 1.721e-05
        [call_graph_tuple_transform]: 1.811e-05
        [tuple_list_get_item_eliminator]: 5.39e-06
        [none_parameter_eliminate]: 1.15001e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 5.61003e-06
[remove_dup_value]: 1.326e-05
[partial_unused_args_eliminate]: 1.96998e-06
[environ_conv]: 4.75001e-06
[add_recomputation]: 4.408e-05
[cse_after_recomputation]: 1.97e-05, [1]
    [Cycle 1]: 1.447e-05, [1]
        [cse]: 8.89e-06
[auto_monad_reorder]: 1.422e-05
[get_jit_bprop_graph]: 1.49e-06
[rewriter_after_jit_bprop_graph]: 5.25999e-06
[opt_after_jit_grad]: 0.00048609
[symbol_engine_optimizer]: 7.242e-05, [1]
    [Cycle 1]: 6.588e-05, [6]
        [build]: 2.77002e-06
        [elim_shapecalc]: 7.84002e-06
        [elim_not_effective]: 1.301e-05
        [opt_reshape]: 5.47001e-06
        [fold_const_symbol]: 9.22001e-06
        [renormalize]: 6.19999e-07
[validate]: 3.418e-05
Sums
    bootstrap                            :     0.000260s :  2.45%
    type_inference                       :     0.006454s : 60.85%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000065s :  0.61%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000053s :  0.50%
    rewriter_before_opt_a                :     0.000074s :  0.70%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000113s :  1.06%
    jit_opt_a.loop_unroll                :     0.000040s :  0.38%
    jit_opt_a.a_1                        :     0.000761s :  7.18%
    jit_opt_a.with_stream_mark           :     0.000018s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.10%
    jit_opt_a.merge_forward              :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000734s :  6.92%
    jit_opt_a.add_forward_monad_depend   :     0.000004s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000012s :  0.12%
    jit_opt_a.cse                        :     0.000024s :  0.23%
    jit_opt_a.replace_applicator         :     0.000015s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.11%
    rewriter_after_opt_a                 :     0.000034s :  0.32%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000540s :  5.09%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.15%
    cconv                                :     0.000022s :  0.21%
    loop_unroll                          :     0.000416s :  3.92%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000013s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000044s :  0.42%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000014s :  0.13%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000486s :  4.58%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000034s :  0.32%


Time group info:
------[substitution.]   0.000249    26
 1.01% :     0.000003s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 2.09% :     0.000005s :      3: substitution.graph_param_transform
86.04% :     0.000214s :      6: substitution.inline
 1.10% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.65% :     0.000004s :      4: substitution.remove_not_recompute_node
 0.95% :     0.000002s :      2: substitution.replace_old_param
 3.39% :     0.000008s :      1: substitution.switch_simplify
 3.25% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006405     2
80.69% :     0.005168s :      1: type_inference.infer
19.31% :     0.001237s :      1: type_inference.specialize
------[replace.]   0.000131     9
49.79% :     0.000065s :      6: replace.inline
24.92% :     0.000033s :      1: replace.switch_simplify
25.29% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000225     9
93.54% :     0.000211s :      6: match.inline
 3.39% :     0.000008s :      1: match.switch_simplify
 3.08% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   801
 1.37% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.12% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     12: predicate.addn_check_dump
 1.59% :     0.000002s :     12: predicate.addn_zero_filter
 2.14% :     0.000003s :     12: predicate.arithmetic_simplify
 1.37% :     0.000002s :     12: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     12: predicate.compare_switch_simplify
 1.39% :     0.000002s :     12: predicate.depend_value_elim
 1.18% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.71% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.47% :     0.000002s :     12: predicate.environ_get_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 4.88% :     0.000006s :     26: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000002s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     17: predicate.load_eliminater
 1.65% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.56% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.97% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000002s :     12: predicate.merge_addn
 1.15% :     0.000002s :     12: predicate.minmaximum_grad
 1.25% :     0.000002s :      3: predicate.mutable_eliminate
 0.39% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     17: predicate.partial_eliminate
 1.21% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.80% :     0.000002s :     12: predicate.reduce_eliminate
 1.44% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     20: predicate.replace_applicator
 0.44% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     12: predicate.reshape_eliminate
 1.58% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.73% :     0.000004s :     20: predicate.switch_defer_inline
 2.36% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.18% :     0.000011s :     65: predicate.switch_simplify
 1.32% :     0.000002s :     12: predicate.tile_eliminate
 1.35% :     0.000002s :     12: predicate.transpose_eliminate
 1.51% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.26% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     12: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.88% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000932    15
46.40% :     0.000432s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 9.89% :     0.000092s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.71% :     0.000407s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016341    72
 0.29% :     0.000047s :      1: add_recomputation
 0.42% :     0.000068s :      1: auto_monad
 0.10% :     0.000017s :      1: auto_monad_reorder
 1.68% :     0.000274s :      1: bootstrap
 0.15% :     0.000025s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.13% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
33.74% :     0.005513s :      1: jit_opt_a
 0.87% :     0.000142s :      1: jit_opt_after_cconv
 0.32% :     0.000053s :      1: jit_opt_b
 2.59% :     0.000423s :      1: loop_unroll
 3.35% :     0.000548s :      1: mutable_eliminate
 6.28% :     0.001026s :     26: opt.transform.jit_opt_a
 0.28% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.02% :     0.000493s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.34% :     0.000055s :      1: py_interpret_to_execute
 0.08% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.30% :     0.000375s :      1: renormalize.infer
 2.16% :     0.000352s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000036s :      1: rewriter_after_opt_a
 0.47% :     0.000077s :      1: rewriter_before_opt_a
 0.46% :     0.000075s :      1: symbol_engine_optimizer
39.57% :     0.006466s :      1: type_inference


TotalTime = 0.0203863, [33]
[bootstrap]: 0.00030692
[type_inference]: 0.00502441
[event_method]: 1.713e-05
[auto_monad]: 6.432e-05
[graph_reusing]: 5.56e-06
[pre_auto_parallel]: 2.62001e-06
[py_interpret_to_execute]: 3.777e-05
[rewriter_before_opt_a]: 7.223e-05
[expand_dump_flag]: 3.06001e-06
[jit_opt_a]: 0.00578712, [2]
    [Cycle 1]: 0.0018628, [27]
        [switch_simplify]: 6.237e-05
        [loop_unroll]: 3.035e-05
        [a_1]: 0.00057061
        [with_stream_mark]: 2.129e-05
        [recompute_prepare]: 8.77999e-06
        [updatestate_depend_eliminate]: 3.53999e-06
        [updatestate_assign_eliminate]: 3.91999e-06
        [updatestate_loads_eliminate]: 2.66e-06
        [parameter_eliminate]: 2.38998e-06
        [specialize_transform]: 8.28001e-06
        [updatestate_useless_node_eliminater]: 5.51e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 2.17001e-06
        [get_grad_eliminate_]: 5.64e-06
        [merge_forward]: 4e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.561e-05
        [j_node_and_user_rematch]: 9.90002e-06
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 9.34e-06
        [inline_without_move]: 1.215e-05
        [renormalize]: 0.00076241
        [add_forward_monad_depend]: 6.04001e-06
        [auto_monad_grad]: 3.18e-06
        [auto_monad_eliminator]: 1.461e-05
        [cse]: 2.931e-05
        [replace_applicator]: 1.385e-05
    [Cycle 2]: 0.00033903, [27]
        [switch_simplify]: 5.52001e-06
        [loop_unroll]: 5.17e-06
        [a_1]: 9.379e-05
        [with_stream_mark]: 1.079e-05
        [recompute_prepare]: 5.44e-06
        [updatestate_depend_eliminate]: 2.57001e-06
        [updatestate_assign_eliminate]: 2.10002e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 5.37999e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.65001e-06
        [meta_shard_fg_expand]: 1.64e-06
        [get_grad_eliminate_]: 4.60999e-06
        [merge_forward]: 2.97002e-06
        [cell_reuse_recompute_pass]: 2.29999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.366e-05
        [j_node_and_user_rematch]: 8.35001e-06
        [meta_fg_expand]: 1.94e-06
        [replace_old_param]: 7.6e-06
        [inline_without_move]: 4.46002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.73002e-06
        [auto_monad_grad]: 1.10001e-06
        [auto_monad_eliminator]: 5.79e-06
        [cse]: 1.184e-05
        [replace_applicator]: 5.49e-06
[py_interpret_to_execute_after_opt_a]: 1.099e-05
[rewriter_after_opt_a]: 3.69e-05
[convert_after_rewriter]: 7.07002e-06
[order_py_execute_after_rewriter]: 4.99e-06
[mutable_eliminate]: 0.00059551
[jit_opt_b]: 5.158e-05, [1]
    [Cycle 1]: 4.448e-05, [2]
        [frontend_op_eliminate]: 1.735e-05
        [inline_after_opt_a]: 1.543e-05
[cconv]: 2.562e-05
[loop_unroll]: 0.00045908
[jit_opt_after_cconv]: 0.00014426, [1]
    [Cycle 1]: 0.00013817, [11]
        [c_1]: 2.161e-05
        [parameter_eliminate]: 2.66999e-06
        [updatestate_depend_eliminate]: 5.87999e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.11998e-06
        [cse]: 2.027e-05
        [call_graph_tuple_transform]: 1.979e-05
        [tuple_list_get_item_eliminator]: 5.44998e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 6.89994e-07
        [switch_simplify]: 5.44e-06
[remove_dup_value]: 1.351e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 5.61998e-06
[add_recomputation]: 5.057e-05
[cse_after_recomputation]: 2.125e-05, [1]
    [Cycle 1]: 1.578e-05, [1]
        [cse]: 9.52999e-06
[auto_monad_reorder]: 1.492e-05
[get_jit_bprop_graph]: 2.02001e-06
[rewriter_after_jit_bprop_graph]: 6.18998e-06
[opt_after_jit_grad]: 0.00048978
[symbol_engine_optimizer]: 7.856e-05, [1]
    [Cycle 1]: 7.146e-05, [6]
        [build]: 4.52998e-06
        [elim_shapecalc]: 7.97998e-06
        [elim_not_effective]: 1.464e-05
        [opt_reshape]: 5.87999e-06
        [fold_const_symbol]: 9.76e-06
        [renormalize]: 4.19997e-07
[validate]: 3.861e-05
[backend_pass]: 8.00006e-07
[task_emit]: 0.00678258
[execute]: 8.67e-06
Sums
    bootstrap                            :     0.000307s :  1.91%
    type_inference                       :     0.005024s : 31.21%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000064s :  0.40%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.23%
    rewriter_before_opt_a                :     0.000072s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.42%
    jit_opt_a.loop_unroll                :     0.000036s :  0.22%
    jit_opt_a.a_1                        :     0.000664s :  4.13%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000014s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000017s :  0.10%
    jit_opt_a.renormalize                :     0.000762s :  4.74%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000041s :  0.26%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000596s :  3.70%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000026s :  0.16%
    loop_unroll                          :     0.000459s :  2.85%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000051s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000490s :  3.04%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000039s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006783s : 42.13%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000225    23
 1.19% :     0.000003s :      2: substitution.elim_not_effective
 0.75% :     0.000002s :      2: substitution.fold_const_symbol
 2.52% :     0.000006s :      3: substitution.graph_param_transform
83.76% :     0.000189s :      4: substitution.inline
 1.39% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.08% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.97% :     0.000004s :      2: substitution.replace_old_param
 6.36% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004954     2
81.35% :     0.004030s :      1: type_inference.infer
18.65% :     0.000924s :      1: type_inference.specialize
------[replace.]   0.000075     6
67.01% :     0.000050s :      4: replace.inline
32.99% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000199     6
93.38% :     0.000186s :      4: match.inline
 6.62% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.42% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.72% :     0.000003s :     10: predicate.arithmetic_simplify
 1.39% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000002s :     10: predicate.depend_value_elim
 0.98% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.73% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.10% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000001s :      3: predicate.graph_param_transform
 6.00% :     0.000007s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.56% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.44% :     0.000002s :      6: predicate.less_batch_normalization
 2.26% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 2.01% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.23% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.73% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.28% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.52% :     0.000002s :     18: predicate.replace_applicator
 0.85% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.27% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.25% :     0.000002s :      6: predicate.special_op_eliminate
 1.30% :     0.000002s :      6: predicate.specialize_transform
 1.25% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.09% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.23% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.40% :     0.000009s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.23% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.06% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.96% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000668    12
48.36% :     0.000323s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.43% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.21% :     0.000309s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022151    76
 0.24% :     0.000054s :      1: add_recomputation
 0.31% :     0.000068s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.48% :     0.000327s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.14% :     0.005791s :      1: jit_opt_a
 0.66% :     0.000147s :      1: jit_opt_after_cconv
 0.24% :     0.000054s :      1: jit_opt_b
 2.11% :     0.000467s :      1: loop_unroll
 2.73% :     0.000604s :      1: mutable_eliminate
 4.06% :     0.000900s :     26: opt.transform.jit_opt_a
 0.22% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.25% :     0.000498s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000040s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.77% :     0.000393s :      1: renormalize.infer
 1.63% :     0.000360s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000040s :      1: rewriter_after_opt_a
 0.34% :     0.000075s :      1: rewriter_before_opt_a
 0.37% :     0.000082s :      1: symbol_engine_optimizer
30.70% :     0.006800s :      1: task_emit
22.74% :     0.005037s :      1: type_inference
 0.26% :     0.000058s :      1: validate


TotalTime = 0.0215014, [33]
[bootstrap]: 0.00031321
[type_inference]: 0.00542784
[event_method]: 1.72e-05
[auto_monad]: 6.869e-05
[graph_reusing]: 5.99e-06
[pre_auto_parallel]: 2.49001e-06
[py_interpret_to_execute]: 3.767e-05
[rewriter_before_opt_a]: 7.269e-05
[expand_dump_flag]: 3.25e-06
[jit_opt_a]: 0.00558569, [2]
    [Cycle 1]: 0.00163784, [27]
        [switch_simplify]: 5.571e-05
        [loop_unroll]: 2.886e-05
        [a_1]: 0.00050354
        [with_stream_mark]: 1.725e-05
        [recompute_prepare]: 7.05e-06
        [updatestate_depend_eliminate]: 3.71999e-06
        [updatestate_assign_eliminate]: 3.26999e-06
        [updatestate_loads_eliminate]: 2.64001e-06
        [parameter_eliminate]: 1.94999e-06
        [specialize_transform]: 6.06998e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 2.19001e-06
        [get_grad_eliminate_]: 5.62001e-06
        [merge_forward]: 3.75e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.419e-05
        [j_node_and_user_rematch]: 9.29998e-06
        [meta_fg_expand]: 2.29999e-06
        [replace_old_param]: 8.57e-06
        [inline_without_move]: 5.56e-06
        [renormalize]: 0.00070843
        [add_forward_monad_depend]: 5.40001e-06
        [auto_monad_grad]: 2.58998e-06
        [auto_monad_eliminator]: 1.405e-05
        [cse]: 2.911e-05
        [replace_applicator]: 1.395e-05
    [Cycle 2]: 0.00032039, [27]
        [switch_simplify]: 5.61998e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 8.856e-05
        [with_stream_mark]: 9.11002e-06
        [recompute_prepare]: 5.65001e-06
        [updatestate_depend_eliminate]: 3.06001e-06
        [updatestate_assign_eliminate]: 2.36998e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 5.02e-06
        [updatestate_useless_node_eliminater]: 4.77998e-06
        [accelerated_algorithm]: 5.28002e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 4.25e-06
        [merge_forward]: 3.21001e-06
        [cell_reuse_recompute_pass]: 1.71002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.268e-05
        [j_node_and_user_rematch]: 7.61999e-06
        [meta_fg_expand]: 1.65001e-06
        [replace_old_param]: 6.94999e-06
        [inline_without_move]: 4.51002e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.64e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 5.10999e-06
        [cse]: 1.071e-05
        [replace_applicator]: 5.32001e-06
[py_interpret_to_execute_after_opt_a]: 1.081e-05
[rewriter_after_opt_a]: 3.661e-05
[convert_after_rewriter]: 7.71999e-06
[order_py_execute_after_rewriter]: 5.29e-06
[mutable_eliminate]: 0.00058906
[jit_opt_b]: 4.84e-05, [1]
    [Cycle 1]: 4.182e-05, [2]
        [frontend_op_eliminate]: 1.603e-05
        [inline_after_opt_a]: 1.446e-05
[cconv]: 2.622e-05
[loop_unroll]: 0.00042748
[jit_opt_after_cconv]: 0.00013995, [1]
    [Cycle 1]: 0.00013363, [11]
        [c_1]: 2.046e-05
        [parameter_eliminate]: 2.37999e-06
        [updatestate_depend_eliminate]: 5.61e-06
        [updatestate_assign_eliminate]: 2.31998e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [cse]: 1.945e-05
        [call_graph_tuple_transform]: 1.861e-05
        [tuple_list_get_item_eliminator]: 5.23002e-06
        [none_parameter_eliminate]: 1.45001e-06
        [renormalize]: 8.49977e-07
        [switch_simplify]: 5.47999e-06
[remove_dup_value]: 1.393e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 5.97001e-06
[add_recomputation]: 4.752e-05
[cse_after_recomputation]: 2.062e-05, [1]
    [Cycle 1]: 1.493e-05, [1]
        [cse]: 9.01002e-06
[auto_monad_reorder]: 1.493e-05
[get_jit_bprop_graph]: 2.14e-06
[rewriter_after_jit_bprop_graph]: 4.63001e-06
[opt_after_jit_grad]: 0.00184689
[symbol_engine_optimizer]: 8.293e-05, [1]
    [Cycle 1]: 7.469e-05, [6]
        [build]: 5.09e-06
        [elim_shapecalc]: 8.25999e-06
        [elim_not_effective]: 1.455e-05
        [opt_reshape]: 6.51e-06
        [fold_const_symbol]: 8.60001e-06
        [renormalize]: 1.10999e-06
[validate]: 4.302e-05
[backend_pass]: 1.26997e-06
[task_emit]: 0.00635911
[execute]: 8.75001e-06
Sums
    bootstrap                            :     0.000313s :  1.82%
    type_inference                       :     0.005428s : 31.48%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000069s :  0.40%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000038s :  0.22%
    rewriter_before_opt_a                :     0.000073s :  0.42%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.36%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000592s :  3.43%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000709s :  4.11%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.11%
    jit_opt_a.cse                        :     0.000040s :  0.23%
    jit_opt_a.replace_applicator         :     0.000019s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000037s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000589s :  3.42%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.08%
    cconv                                :     0.000026s :  0.15%
    loop_unroll                          :     0.000427s :  2.48%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000048s :  0.28%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.001847s : 10.71%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006359s : 36.88%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000183    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 3.15% :     0.000006s :      3: substitution.graph_param_transform
81.95% :     0.000150s :      4: substitution.inline
 1.72% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.32% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000004s :      2: substitution.replace_old_param
 7.06% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005354     2
81.40% :     0.004358s :      1: type_inference.infer
18.60% :     0.000996s :      1: type_inference.specialize
------[replace.]   0.000062     6
66.20% :     0.000041s :      4: replace.inline
33.80% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000160     6
92.53% :     0.000148s :      4: match.inline
 7.47% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.02% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000001s :     10: predicate.addn_zero_filter
 1.92% :     0.000002s :     10: predicate.arithmetic_simplify
 1.49% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000001s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.39% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.28% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.52% :     0.000001s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.20% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.17% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.67% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.92% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.24% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.18% :     0.000001s :     10: predicate.minmaximum_grad
 1.52% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.45% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.03% :     0.000002s :     18: predicate.replace_applicator
 0.45% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.93% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.54% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.30% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.47% :     0.000009s :     53: predicate.switch_simplify
 1.34% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.70% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.43% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000674    12
44.84% :     0.000302s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.56% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.60% :     0.000341s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023121    76
 0.22% :     0.000050s :      1: add_recomputation
 0.31% :     0.000072s :      1: auto_monad
 0.07% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.43% :     0.000330s :      1: bootstrap
 0.12% :     0.000029s :      1: cconv
 0.05% :     0.000012s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
24.17% :     0.005589s :      1: jit_opt_a
 0.62% :     0.000143s :      1: jit_opt_after_cconv
 0.22% :     0.000051s :      1: jit_opt_b
 1.88% :     0.000435s :      1: loop_unroll
 2.59% :     0.000598s :      1: mutable_eliminate
 3.46% :     0.000801s :     26: opt.transform.jit_opt_a
 0.20% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000035s :      4: opt.transform.symbol_engine_opt
 8.05% :     0.001861s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000041s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.56% :     0.000361s :      1: renormalize.infer
 1.47% :     0.000340s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000040s :      1: rewriter_after_opt_a
 0.33% :     0.000076s :      1: rewriter_before_opt_a
 0.37% :     0.000086s :      1: symbol_engine_optimizer
27.58% :     0.006377s :      1: task_emit
23.55% :     0.005445s :      1: type_inference
 0.28% :     0.000065s :      1: validate


TotalTime = 0.0178303, [33]
[bootstrap]: 0.00026276
[type_inference]: 0.00443578
[event_method]: 1.701e-05
[auto_monad]: 6.367e-05
[graph_reusing]: 5.37001e-06
[pre_auto_parallel]: 2.39001e-06
[py_interpret_to_execute]: 3.045e-05
[rewriter_before_opt_a]: 7.723e-05
[expand_dump_flag]: 3.17002e-06
[jit_opt_a]: 0.00521029, [2]
    [Cycle 1]: 0.00150536, [27]
        [switch_simplify]: 5.219e-05
        [loop_unroll]: 2.779e-05
        [a_1]: 0.00052357
        [with_stream_mark]: 1.922e-05
        [recompute_prepare]: 6.63998e-06
        [updatestate_depend_eliminate]: 3.52002e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.70997e-06
        [parameter_eliminate]: 2.04999e-06
        [specialize_transform]: 5.74e-06
        [updatestate_useless_node_eliminater]: 5.06002e-06
        [accelerated_algorithm]: 5.39998e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 4.86002e-06
        [merge_forward]: 3.28998e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.358e-05
        [j_node_and_user_rematch]: 8.50999e-06
        [meta_fg_expand]: 2.46e-06
        [replace_old_param]: 8.53001e-06
        [inline_without_move]: 5.04e-06
        [renormalize]: 0.00057808
        [add_forward_monad_depend]: 5.42001e-06
        [auto_monad_grad]: 2.26e-06
        [auto_monad_eliminator]: 1.307e-05
        [cse]: 2.642e-05
        [replace_applicator]: 1.095e-05
    [Cycle 2]: 0.0003181, [27]
        [switch_simplify]: 5.29e-06
        [loop_unroll]: 4.84e-06
        [a_1]: 9.348e-05
        [with_stream_mark]: 8.65999e-06
        [recompute_prepare]: 5.30999e-06
        [updatestate_depend_eliminate]: 2.44001e-06
        [updatestate_assign_eliminate]: 1.98997e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 4.82e-06
        [updatestate_useless_node_eliminater]: 4.62998e-06
        [accelerated_algorithm]: 4.84e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.35e-06
        [merge_forward]: 2.21e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.19e-05
        [j_node_and_user_rematch]: 7.51999e-06
        [meta_fg_expand]: 1.35999e-06
        [replace_old_param]: 1.18e-05
        [inline_without_move]: 4.45999e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.52001e-06
        [auto_monad_grad]: 7.49977e-07
        [auto_monad_eliminator]: 4.89998e-06
        [cse]: 1.112e-05
        [replace_applicator]: 4.99003e-06
[py_interpret_to_execute_after_opt_a]: 9.76e-06
[rewriter_after_opt_a]: 3.353e-05
[convert_after_rewriter]: 7.01001e-06
[order_py_execute_after_rewriter]: 5.35999e-06
[mutable_eliminate]: 0.00050391
[jit_opt_b]: 4.755e-05, [1]
    [Cycle 1]: 4.079e-05, [2]
        [frontend_op_eliminate]: 1.558e-05
        [inline_after_opt_a]: 1.415e-05
[cconv]: 2.484e-05
[loop_unroll]: 0.00040191
[jit_opt_after_cconv]: 0.00013168, [1]
    [Cycle 1]: 0.00012548, [11]
        [c_1]: 1.97e-05
        [parameter_eliminate]: 2.28002e-06
        [updatestate_depend_eliminate]: 5.06997e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [cse]: 1.592e-05
        [call_graph_tuple_transform]: 1.721e-05
        [tuple_list_get_item_eliminator]: 5.01997e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 5.19e-06
[remove_dup_value]: 1.327e-05
[partial_unused_args_eliminate]: 2.07999e-06
[environ_conv]: 4.55001e-06
[add_recomputation]: 4.627e-05
[cse_after_recomputation]: 1.916e-05, [1]
    [Cycle 1]: 1.39e-05, [1]
        [cse]: 8.38999e-06
[auto_monad_reorder]: 1.413e-05
[get_jit_bprop_graph]: 1.95001e-06
[rewriter_after_jit_bprop_graph]: 4.75001e-06
[opt_after_jit_grad]: 0.00043623
[symbol_engine_optimizer]: 6.883e-05, [1]
    [Cycle 1]: 6.27e-05, [6]
        [build]: 3.28998e-06
        [elim_shapecalc]: 6.83998e-06
        [elim_not_effective]: 1.167e-05
        [opt_reshape]: 5.39e-06
        [fold_const_symbol]: 8.74e-06
        [renormalize]: 4.00003e-07
[validate]: 3.046e-05
[backend_pass]: 9.10019e-07
[task_emit]: 0.00568692
[execute]: 6.98998e-06
Sums
    bootstrap                            :     0.000263s :  1.90%
    type_inference                       :     0.004436s : 32.06%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000064s :  0.46%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.22%
    rewriter_before_opt_a                :     0.000077s :  0.56%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.42%
    jit_opt_a.loop_unroll                :     0.000033s :  0.24%
    jit_opt_a.a_1                        :     0.000617s :  4.46%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000020s :  0.15%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000578s :  4.18%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000038s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000504s :  3.64%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.18%
    loop_unroll                          :     0.000402s :  2.91%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000046s :  0.33%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000436s :  3.15%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000030s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005687s : 41.11%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000208    23
 0.80% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000002s :      2: substitution.fold_const_symbol
 2.44% :     0.000005s :      3: substitution.graph_param_transform
82.52% :     0.000171s :      4: substitution.inline
 1.40% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.88% :     0.000004s :      4: substitution.remove_not_recompute_node
 4.25% :     0.000009s :      2: substitution.replace_old_param
 5.98% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004370     2
81.00% :     0.003540s :      1: type_inference.infer
19.00% :     0.000830s :      1: type_inference.specialize
------[replace.]   0.000068     6
72.17% :     0.000049s :      4: replace.inline
27.83% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000180     6
93.71% :     0.000169s :      4: match.inline
 6.29% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 1.71% :     0.000002s :     10: predicate.arithmetic_simplify
 1.38% :     0.000002s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.45% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000000s :      3: predicate.graph_param_transform
 6.16% :     0.000007s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.73% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.69% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.41% :     0.000003s :     15: predicate.partial_eliminate
 1.39% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.79% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.99% :     0.000009s :     53: predicate.switch_simplify
 1.32% :     0.000001s :     10: predicate.tile_eliminate
 1.32% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.89% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000542    12
49.97% :     0.000271s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.53% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.51% :     0.000230s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019325    76
 0.25% :     0.000049s :      1: add_recomputation
 0.35% :     0.000067s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.61% :     0.000312s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.98% :     0.005213s :      1: jit_opt_a
 0.70% :     0.000134s :      1: jit_opt_after_cconv
 0.26% :     0.000050s :      1: jit_opt_b
 2.12% :     0.000410s :      1: loop_unroll
 2.65% :     0.000512s :      1: mutable_eliminate
 4.24% :     0.000819s :     26: opt.transform.jit_opt_a
 0.23% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.30% :     0.000444s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.65% :     0.000319s :      1: renormalize.infer
 1.31% :     0.000252s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000037s :      1: rewriter_after_opt_a
 0.42% :     0.000080s :      1: rewriter_before_opt_a
 0.37% :     0.000071s :      1: symbol_engine_optimizer
29.49% :     0.005699s :      1: task_emit
23.02% :     0.004449s :      1: type_inference
 0.26% :     0.000050s :      1: validate


TotalTime = 0.0149638, [30]
[bootstrap]: 0.00026108
[type_inference]: 0.00670363
[event_method]: 2.118e-05
[auto_monad]: 8.692e-05
[graph_reusing]: 6.93e-06
[pre_auto_parallel]: 2.91999e-06
[py_interpret_to_execute]: 6.049e-05
[rewriter_before_opt_a]: 9.269e-05
[expand_dump_flag]: 3.37002e-06
[jit_opt_a]: 0.00568918, [2]
    [Cycle 1]: 0.00201298, [27]
        [switch_simplify]: 0.00012109
        [loop_unroll]: 3.612e-05
        [a_1]: 0.00075096
        [with_stream_mark]: 1.743e-05
        [recompute_prepare]: 7.18e-06
        [updatestate_depend_eliminate]: 3.45e-06
        [updatestate_assign_eliminate]: 3.36999e-06
        [updatestate_loads_eliminate]: 2.66e-06
        [parameter_eliminate]: 1.95001e-06
        [specialize_transform]: 6.53e-06
        [updatestate_useless_node_eliminater]: 5.25999e-06
        [accelerated_algorithm]: 5.61e-06
        [meta_shard_fg_expand]: 2.21998e-06
        [get_grad_eliminate_]: 5.67999e-06
        [merge_forward]: 3.61001e-06
        [cell_reuse_recompute_pass]: 1.43002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.458e-05
        [j_node_and_user_rematch]: 1e-05
        [meta_fg_expand]: 2.33002e-06
        [replace_old_param]: 8.85001e-06
        [inline_without_move]: 5.82001e-06
        [renormalize]: 0.00071071
        [add_forward_monad_depend]: 4.85999e-06
        [auto_monad_grad]: 2.07999e-06
        [auto_monad_eliminator]: 1.301e-05
        [cse]: 2.855e-05
        [replace_applicator]: 1.156e-05
    [Cycle 2]: 0.00032918, [27]
        [switch_simplify]: 5.71e-06
        [loop_unroll]: 4.88001e-06
        [a_1]: 9.136e-05
        [with_stream_mark]: 8.18999e-06
        [recompute_prepare]: 5.14e-06
        [updatestate_depend_eliminate]: 2.51e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 5.07e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 5.63002e-06
        [meta_shard_fg_expand]: 1.27999e-06
        [get_grad_eliminate_]: 4.68001e-06
        [merge_forward]: 2.44999e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.192e-05
        [j_node_and_user_rematch]: 7.88999e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 7.59002e-06
        [inline_without_move]: 4.86002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 5.66e-06
        [cse]: 1.086e-05
        [replace_applicator]: 5.40999e-06
[py_interpret_to_execute_after_opt_a]: 8.43999e-06
[rewriter_after_opt_a]: 3.216e-05
[convert_after_rewriter]: 6.64001e-06
[order_py_execute_after_rewriter]: 5.61e-06
[mutable_eliminate]: 0.00051767
[jit_opt_b]: 5.105e-05, [1]
    [Cycle 1]: 4.373e-05, [2]
        [frontend_op_eliminate]: 1.679e-05
        [inline_after_opt_a]: 1.48e-05
[cconv]: 2.513e-05
[loop_unroll]: 0.00041718
[jit_opt_after_cconv]: 0.00014043, [1]
    [Cycle 1]: 0.00013371, [11]
        [c_1]: 2.346e-05
        [parameter_eliminate]: 2.11998e-06
        [updatestate_depend_eliminate]: 4.74e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [cse]: 1.791e-05
        [call_graph_tuple_transform]: 1.749e-05
        [tuple_list_get_item_eliminator]: 5.40001e-06
        [none_parameter_eliminate]: 1.74998e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.67999e-06
[remove_dup_value]: 1.314e-05
[partial_unused_args_eliminate]: 2.50002e-06
[environ_conv]: 5.19e-06
[add_recomputation]: 4.586e-05
[cse_after_recomputation]: 2.022e-05, [1]
    [Cycle 1]: 1.44e-05, [1]
        [cse]: 8.83001e-06
[auto_monad_reorder]: 1.324e-05
[get_jit_bprop_graph]: 1.49e-06
[rewriter_after_jit_bprop_graph]: 4.35e-06
[opt_after_jit_grad]: 0.00045167
[symbol_engine_optimizer]: 7.281e-05, [1]
    [Cycle 1]: 6.602e-05, [6]
        [build]: 3.82002e-06
        [elim_shapecalc]: 7.2e-06
        [elim_not_effective]: 1.208e-05
        [opt_reshape]: 6.05002e-06
        [fold_const_symbol]: 8.70001e-06
        [renormalize]: 4.30009e-07
[validate]: 3.221e-05
Sums
    bootstrap                            :     0.000261s :  2.38%
    type_inference                       :     0.006704s : 61.07%
    event_method                         :     0.000021s :  0.19%
    auto_monad                           :     0.000087s :  0.79%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000060s :  0.55%
    rewriter_before_opt_a                :     0.000093s :  0.84%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000127s :  1.16%
    jit_opt_a.loop_unroll                :     0.000041s :  0.37%
    jit_opt_a.a_1                        :     0.000842s :  7.67%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000711s :  6.47%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.17%
    jit_opt_a.cse                        :     0.000039s :  0.36%
    jit_opt_a.replace_applicator         :     0.000017s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.08%
    rewriter_after_opt_a                 :     0.000032s :  0.29%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000518s :  4.72%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.13%
    cconv                                :     0.000025s :  0.23%
    loop_unroll                          :     0.000417s :  3.80%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000013s :  0.12%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000046s :  0.42%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000013s :  0.12%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.04%
    opt_after_jit_grad                   :     0.000452s :  4.11%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.29%


Time group info:
------[substitution.]   0.000298    26
 0.63% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 1.61% :     0.000005s :      3: substitution.graph_param_transform
86.55% :     0.000258s :      6: substitution.inline
 1.05% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.45% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.15% :     0.000003s :      2: substitution.replace_old_param
 2.79% :     0.000008s :      1: substitution.switch_simplify
 4.30% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006636     2
82.26% :     0.005458s :      1: type_inference.infer
17.74% :     0.001177s :      1: type_inference.specialize
------[replace.]   0.000158     9
49.47% :     0.000078s :      6: replace.inline
22.75% :     0.000036s :      1: replace.switch_simplify
27.78% :     0.000044s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000273     9
93.05% :     0.000254s :      6: match.inline
 2.74% :     0.000007s :      1: match.switch_simplify
 4.21% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   801
 1.37% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.93% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     12: predicate.addn_check_dump
 1.31% :     0.000002s :     12: predicate.addn_zero_filter
 2.19% :     0.000003s :     12: predicate.arithmetic_simplify
 1.32% :     0.000002s :     12: predicate.cast_eliminate
 0.37% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.42% :     0.000002s :     12: predicate.compare_switch_simplify
 1.47% :     0.000002s :     12: predicate.depend_value_elim
 1.12% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.71% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.45% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     12: predicate.environ_get_eliminate
 1.09% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.41% :     0.000001s :      3: predicate.graph_param_transform
 5.56% :     0.000007s :     26: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     17: predicate.load_eliminater
 1.13% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.43% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.87% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000002s :     12: predicate.merge_addn
 1.17% :     0.000002s :     12: predicate.minmaximum_grad
 1.49% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     17: predicate.partial_eliminate
 1.23% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     12: predicate.reduce_eliminate
 1.42% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.92% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     20: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     12: predicate.reshape_eliminate
 1.27% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.60% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     12: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.77% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.53% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.77% :     0.000004s :     20: predicate.switch_defer_inline
 2.38% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.89% :     0.000012s :     65: predicate.switch_simplify
 1.23% :     0.000002s :     12: predicate.tile_eliminate
 1.17% :     0.000002s :     12: predicate.transpose_eliminate
 1.57% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.94% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.87% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000916    15
50.80% :     0.000466s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.06% :     0.000101s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.14% :     0.000349s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016855    72
 0.29% :     0.000049s :      1: add_recomputation
 0.54% :     0.000091s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 1.62% :     0.000273s :      1: bootstrap
 0.16% :     0.000028s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
33.77% :     0.005693s :      1: jit_opt_a
 0.85% :     0.000143s :      1: jit_opt_after_cconv
 0.32% :     0.000054s :      1: jit_opt_b
 2.52% :     0.000425s :      1: loop_unroll
 3.12% :     0.000527s :      1: mutable_eliminate
 6.68% :     0.001126s :     26: opt.transform.jit_opt_a
 0.29% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.72% :     0.000459s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.38% :     0.000064s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 2.22% :     0.000373s :      1: renormalize.infer
 1.96% :     0.000330s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000035s :      1: rewriter_after_opt_a
 0.57% :     0.000096s :      1: rewriter_before_opt_a
 0.45% :     0.000075s :      1: symbol_engine_optimizer
39.85% :     0.006716s :      1: type_inference


TotalTime = 0.0163146, [33]
[bootstrap]: 0.00025814
[type_inference]: 0.00389785
[event_method]: 1.474e-05
[auto_monad]: 4.009e-05
[graph_reusing]: 3.59002e-06
[pre_auto_parallel]: 1.60001e-06
[py_interpret_to_execute]: 2.55e-05
[rewriter_before_opt_a]: 5.975e-05
[expand_dump_flag]: 1.72999e-06
[jit_opt_a]: 0.00446063, [2]
    [Cycle 1]: 0.00128718, [27]
        [switch_simplify]: 4.25e-05
        [loop_unroll]: 2.801e-05
        [a_1]: 0.00044613
        [with_stream_mark]: 1.143e-05
        [recompute_prepare]: 6.69001e-06
        [updatestate_depend_eliminate]: 3.00998e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 4.71002e-06
        [accelerated_algorithm]: 4.99998e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.79002e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.12999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.011e-05
        [j_node_and_user_rematch]: 8.3e-06
        [meta_fg_expand]: 1.85001e-06
        [replace_old_param]: 7.92e-06
        [inline_without_move]: 5.07e-06
        [renormalize]: 0.00048759
        [add_forward_monad_depend]: 4.15e-06
        [auto_monad_grad]: 1.23002e-06
        [auto_monad_eliminator]: 9.14e-06
        [cse]: 1.566e-05
        [replace_applicator]: 1.088e-05
    [Cycle 2]: 0.00030524, [27]
        [switch_simplify]: 5.13002e-06
        [loop_unroll]: 4.90999e-06
        [a_1]: 8.635e-05
        [with_stream_mark]: 8e-06
        [recompute_prepare]: 4.90001e-06
        [updatestate_depend_eliminate]: 2.51e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 5.09e-06
        [updatestate_useless_node_eliminater]: 4.70999e-06
        [accelerated_algorithm]: 5.01002e-06
        [meta_shard_fg_expand]: 1.07998e-06
        [get_grad_eliminate_]: 4.32e-06
        [merge_forward]: 2.37001e-06
        [cell_reuse_recompute_pass]: 1.52001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.362e-05
        [j_node_and_user_rematch]: 7.39002e-06
        [meta_fg_expand]: 1.50999e-06
        [replace_old_param]: 6.74001e-06
        [inline_without_move]: 4.44998e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.11002e-06
        [auto_monad_grad]: 6.09987e-07
        [auto_monad_eliminator]: 4.63999e-06
        [cse]: 9.14998e-06
        [replace_applicator]: 4.98001e-06
[py_interpret_to_execute_after_opt_a]: 7.92998e-06
[rewriter_after_opt_a]: 2.464e-05
[convert_after_rewriter]: 5.76998e-06
[order_py_execute_after_rewriter]: 4.41002e-06
[mutable_eliminate]: 0.0004648
[jit_opt_b]: 4.627e-05, [1]
    [Cycle 1]: 4.047e-05, [2]
        [frontend_op_eliminate]: 1.524e-05
        [inline_after_opt_a]: 1.401e-05
[cconv]: 1.505e-05
[loop_unroll]: 0.00040363
[jit_opt_after_cconv]: 0.0001393, [1]
    [Cycle 1]: 0.00013317, [11]
        [c_1]: 1.962e-05
        [parameter_eliminate]: 2.41998e-06
        [updatestate_depend_eliminate]: 5.07e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [cse]: 1.593e-05
        [call_graph_tuple_transform]: 2.391e-05
        [tuple_list_get_item_eliminator]: 5.30999e-06
        [none_parameter_eliminate]: 1.07998e-06
        [renormalize]: 4.10015e-07
        [switch_simplify]: 5.39e-06
[remove_dup_value]: 6.92002e-06
[partial_unused_args_eliminate]: 1.72999e-06
[environ_conv]: 3.31999e-06
[add_recomputation]: 3.049e-05
[cse_after_recomputation]: 1.846e-05, [1]
    [Cycle 1]: 1.336e-05, [1]
        [cse]: 8.1e-06
[auto_monad_reorder]: 8.79998e-06
[get_jit_bprop_graph]: 1.47999e-06
[rewriter_after_jit_bprop_graph]: 4.37e-06
[opt_after_jit_grad]: 0.00044197
[symbol_engine_optimizer]: 6.719e-05, [1]
    [Cycle 1]: 6.127e-05, [6]
        [build]: 2.64999e-06
        [elim_shapecalc]: 7.36999e-06
        [elim_not_effective]: 1.144e-05
        [opt_reshape]: 5.14998e-06
        [fold_const_symbol]: 7.98999e-06
        [renormalize]: 3.60014e-07
[validate]: 2.379e-05
[backend_pass]: 8.90024e-07
[task_emit]: 0.00564831
[execute]: 8.97e-06
Sums
    bootstrap                            :     0.000258s :  2.00%
    type_inference                       :     0.003898s : 30.24%
    event_method                         :     0.000015s :  0.11%
    auto_monad                           :     0.000040s :  0.31%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000025s :  0.20%
    rewriter_before_opt_a                :     0.000060s :  0.46%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000048s :  0.37%
    jit_opt_a.loop_unroll                :     0.000033s :  0.26%
    jit_opt_a.a_1                        :     0.000532s :  4.13%
    jit_opt_a.with_stream_mark           :     0.000019s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000488s :  3.78%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.11%
    jit_opt_a.cse                        :     0.000025s :  0.19%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000025s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000465s :  3.61%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000015s :  0.12%
    loop_unroll                          :     0.000404s :  3.13%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000007s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000003s :  0.03%
    add_recomputation                    :     0.000030s :  0.24%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000009s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000442s :  3.43%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000024s :  0.18%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005648s : 43.81%
    execute                              :     0.000009s :  0.07%


Time group info:
------[substitution.]   0.000142    23
 0.90% :     0.000001s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.65% :     0.000004s :      3: substitution.graph_param_transform
82.80% :     0.000117s :      4: substitution.inline
 1.77% :     0.000003s :      4: substitution.j_node_and_user_rematch
 3.82% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.25% :     0.000003s :      2: substitution.replace_old_param
 5.12% :     0.000007s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003851     2
82.13% :     0.003163s :      1: type_inference.infer
17.87% :     0.000688s :      1: type_inference.specialize
------[replace.]   0.000057     6
66.56% :     0.000038s :      4: replace.inline
33.44% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000121     6
94.84% :     0.000115s :      4: match.inline
 5.16% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000105   691
 1.28% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     10: predicate.addn_check_dump
 1.50% :     0.000002s :     10: predicate.addn_zero_filter
 1.92% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.47% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.17% :     0.000005s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.19% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.18% :     0.000001s :     10: predicate.minmaximum_grad
 1.90% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000002s :     15: predicate.partial_eliminate
 1.26% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.51% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.75% :     0.000008s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.27% :     0.000001s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.61% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000455    12
47.22% :     0.000215s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.48% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.30% :     0.000215s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017630    76
 0.19% :     0.000033s :      1: add_recomputation
 0.25% :     0.000043s :      1: auto_monad
 0.06% :     0.000011s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.55% :     0.000274s :      1: bootstrap
 0.10% :     0.000018s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000020s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.10% :     0.000018s :      1: event_method
 0.07% :     0.000012s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
25.31% :     0.004463s :      1: jit_opt_a
 0.81% :     0.000142s :      1: jit_opt_after_cconv
 0.28% :     0.000049s :      1: jit_opt_b
 2.33% :     0.000412s :      1: loop_unroll
 2.68% :     0.000473s :      1: mutable_eliminate
 4.08% :     0.000720s :     26: opt.transform.jit_opt_a
 0.29% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.55% :     0.000450s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000028s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000009s :      1: remove_dup_value
 1.52% :     0.000269s :      1: renormalize.infer
 1.21% :     0.000214s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000027s :      1: rewriter_after_opt_a
 0.35% :     0.000063s :      1: rewriter_before_opt_a
 0.39% :     0.000070s :      1: symbol_engine_optimizer
32.09% :     0.005658s :      1: task_emit
22.17% :     0.003908s :      1: type_inference
 0.22% :     0.000039s :      1: validate


TotalTime = 0.0186018, [33]
[bootstrap]: 0.00032834
[type_inference]: 0.00482697
[event_method]: 1.777e-05
[auto_monad]: 6.419e-05
[graph_reusing]: 5.72999e-06
[pre_auto_parallel]: 2.22999e-06
[py_interpret_to_execute]: 3.868e-05
[rewriter_before_opt_a]: 7.288e-05
[expand_dump_flag]: 3.66001e-06
[jit_opt_a]: 0.00562076, [2]
    [Cycle 1]: 0.00178159, [27]
        [switch_simplify]: 5.497e-05
        [loop_unroll]: 3.431e-05
        [a_1]: 0.00052935
        [with_stream_mark]: 1.424e-05
        [recompute_prepare]: 7.04001e-06
        [updatestate_depend_eliminate]: 2.98e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.50999e-06
        [specialize_transform]: 6.12001e-06
        [updatestate_useless_node_eliminater]: 5.30999e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.24e-06
        [get_grad_eliminate_]: 5.51002e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.203e-05
        [j_node_and_user_rematch]: 9.05999e-06
        [meta_fg_expand]: 1.84998e-06
        [replace_old_param]: 7.78001e-06
        [inline_without_move]: 5.30999e-06
        [renormalize]: 0.00077315
        [add_forward_monad_depend]: 4.55999e-06
        [auto_monad_grad]: 1.32e-06
        [auto_monad_eliminator]: 1.155e-05
        [cse]: 1.855e-05
        [replace_applicator]: 1.158e-05
    [Cycle 2]: 0.00031477, [27]
        [switch_simplify]: 5.37001e-06
        [loop_unroll]: 4.91002e-06
        [a_1]: 8.918e-05
        [with_stream_mark]: 9.67001e-06
        [recompute_prepare]: 4.90001e-06
        [updatestate_depend_eliminate]: 2.36e-06
        [updatestate_assign_eliminate]: 1.82999e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 4.89998e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 4.97e-06
        [meta_shard_fg_expand]: 1.13001e-06
        [get_grad_eliminate_]: 4.31002e-06
        [merge_forward]: 2.23998e-06
        [cell_reuse_recompute_pass]: 1.26002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.2e-05
        [j_node_and_user_rematch]: 7.39002e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 7.20998e-06
        [inline_without_move]: 4.67e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 5.02e-06
        [cse]: 9.86998e-06
        [replace_applicator]: 4.86002e-06
[py_interpret_to_execute_after_opt_a]: 8.28999e-06
[rewriter_after_opt_a]: 3.101e-05
[convert_after_rewriter]: 5.84e-06
[order_py_execute_after_rewriter]: 4.45e-06
[mutable_eliminate]: 0.0005603
[jit_opt_b]: 4.789e-05, [1]
    [Cycle 1]: 4.15e-05, [2]
        [frontend_op_eliminate]: 1.614e-05
        [inline_after_opt_a]: 1.462e-05
[cconv]: 1.755e-05
[loop_unroll]: 0.0004197
[jit_opt_after_cconv]: 0.00013414, [1]
    [Cycle 1]: 0.000128, [11]
        [c_1]: 1.944e-05
        [parameter_eliminate]: 2.58e-06
        [updatestate_depend_eliminate]: 5.66998e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [cse]: 1.75e-05
        [call_graph_tuple_transform]: 1.806e-05
        [tuple_list_get_item_eliminator]: 5.82999e-06
        [none_parameter_eliminate]: 9.00007e-07
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.02e-06
[remove_dup_value]: 8.62e-06
[partial_unused_args_eliminate]: 2.06e-06
[environ_conv]: 3.96001e-06
[add_recomputation]: 3.266e-05
[cse_after_recomputation]: 1.871e-05, [1]
    [Cycle 1]: 1.366e-05, [1]
        [cse]: 8.28001e-06
[auto_monad_reorder]: 1.06e-05
[get_jit_bprop_graph]: 1.47999e-06
[rewriter_after_jit_bprop_graph]: 4.68001e-06
[opt_after_jit_grad]: 0.00045355
[symbol_engine_optimizer]: 6.855e-05, [1]
    [Cycle 1]: 6.229e-05, [6]
        [build]: 3.00002e-06
        [elim_shapecalc]: 7.10998e-06
        [elim_not_effective]: 1.189e-05
        [opt_reshape]: 5.29998e-06
        [fold_const_symbol]: 8.08999e-06
        [renormalize]: 6.10016e-07
[validate]: 3.003e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.00551597
[execute]: 7.78001e-06
Sums
    bootstrap                            :     0.000328s :  2.29%
    type_inference                       :     0.004827s : 33.60%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000064s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.27%
    rewriter_before_opt_a                :     0.000073s :  0.51%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000060s :  0.42%
    jit_opt_a.loop_unroll                :     0.000039s :  0.27%
    jit_opt_a.a_1                        :     0.000619s :  4.31%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000773s :  5.38%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.12%
    jit_opt_a.cse                        :     0.000028s :  0.20%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000560s :  3.90%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000018s :  0.12%
    loop_unroll                          :     0.000420s :  2.92%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000009s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000033s :  0.23%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000011s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000454s :  3.16%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000030s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005516s : 38.40%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000195    23
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.21% :     0.000004s :      3: substitution.graph_param_transform
85.49% :     0.000167s :      4: substitution.inline
 1.42% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.21% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.71% :     0.000003s :      2: substitution.replace_old_param
 5.64% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004756     2
80.48% :     0.003828s :      1: type_inference.infer
19.52% :     0.000928s :      1: type_inference.specialize
------[replace.]   0.000069     6
65.98% :     0.000045s :      4: replace.inline
34.02% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000174     6
94.34% :     0.000165s :      4: match.inline
 5.66% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.48% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.06% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.65% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.33% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.29% :     0.000001s :     10: predicate.environ_get_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.24% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.37% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.40% :     0.000002s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.08% :     0.000002s :     15: predicate.load_eliminater
 1.32% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.57% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.22% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.56% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000002s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.74% :     0.000002s :     10: predicate.reduce_eliminate
 1.61% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.32% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.88% :     0.000009s :     53: predicate.switch_simplify
 1.30% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.28% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.35% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000616    12
48.59% :     0.000299s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.36% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.06% :     0.000284s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020259    76
 0.18% :     0.000036s :      1: add_recomputation
 0.33% :     0.000067s :      1: auto_monad
 0.06% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.68% :     0.000341s :      1: bootstrap
 0.10% :     0.000020s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.10% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.76% :     0.005624s :      1: jit_opt_a
 0.68% :     0.000137s :      1: jit_opt_after_cconv
 0.25% :     0.000051s :      1: jit_opt_b
 2.11% :     0.000427s :      1: loop_unroll
 2.81% :     0.000568s :      1: mutable_eliminate
 4.09% :     0.000828s :     26: opt.transform.jit_opt_a
 0.22% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.28% :     0.000461s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000041s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000011s :      1: remove_dup_value
 2.10% :     0.000425s :      1: renormalize.infer
 1.68% :     0.000340s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000034s :      1: rewriter_after_opt_a
 0.37% :     0.000076s :      1: rewriter_before_opt_a
 0.35% :     0.000071s :      1: symbol_engine_optimizer
27.29% :     0.005529s :      1: task_emit
23.90% :     0.004842s :      1: type_inference
 0.24% :     0.000048s :      1: validate


TotalTime = 0.0213516, [33]
[bootstrap]: 0.00030974
[type_inference]: 0.00560325
[event_method]: 1.962e-05
[auto_monad]: 6.699e-05
[graph_reusing]: 5.67999e-06
[pre_auto_parallel]: 3.25002e-06
[py_interpret_to_execute]: 4.324e-05
[rewriter_before_opt_a]: 7.568e-05
[expand_dump_flag]: 3.24001e-06
[jit_opt_a]: 0.00633238, [2]
    [Cycle 1]: 0.00196754, [27]
        [switch_simplify]: 7.407e-05
        [loop_unroll]: 3.114e-05
        [a_1]: 0.00061118
        [with_stream_mark]: 2.33e-05
        [recompute_prepare]: 7.61001e-06
        [updatestate_depend_eliminate]: 4.12003e-06
        [updatestate_assign_eliminate]: 3.18e-06
        [updatestate_loads_eliminate]: 2.41998e-06
        [parameter_eliminate]: 2.27001e-06
        [specialize_transform]: 6.23e-06
        [updatestate_useless_node_eliminater]: 5.42001e-06
        [accelerated_algorithm]: 6.01e-06
        [meta_shard_fg_expand]: 2.27999e-06
        [get_grad_eliminate_]: 5.47001e-06
        [merge_forward]: 4.26001e-06
        [cell_reuse_recompute_pass]: 1.09998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.417e-05
        [j_node_and_user_rematch]: 1.024e-05
        [meta_fg_expand]: 2.49001e-06
        [replace_old_param]: 9.29e-06
        [inline_without_move]: 6.35002e-06
        [renormalize]: 0.00087657
        [add_forward_monad_depend]: 6.46e-06
        [auto_monad_grad]: 2.88e-06
        [auto_monad_eliminator]: 1.463e-05
        [cse]: 3.035e-05
        [replace_applicator]: 1.301e-05
    [Cycle 2]: 0.00032429, [27]
        [switch_simplify]: 6.28998e-06
        [loop_unroll]: 4.85001e-06
        [a_1]: 9.025e-05
        [with_stream_mark]: 9.44998e-06
        [recompute_prepare]: 4.88001e-06
        [updatestate_depend_eliminate]: 2.79999e-06
        [updatestate_assign_eliminate]: 2.05002e-06
        [updatestate_loads_eliminate]: 1.91998e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 4.80999e-06
        [updatestate_useless_node_eliminater]: 5.02999e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 4.74002e-06
        [merge_forward]: 2.70002e-06
        [cell_reuse_recompute_pass]: 2.21e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.229e-05
        [j_node_and_user_rematch]: 8.33001e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 7.07997e-06
        [inline_without_move]: 4.65999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 5.14e-06
        [cse]: 1.075e-05
        [replace_applicator]: 5.03002e-06
[py_interpret_to_execute_after_opt_a]: 1.187e-05
[rewriter_after_opt_a]: 3.582e-05
[convert_after_rewriter]: 7.51001e-06
[order_py_execute_after_rewriter]: 4.88001e-06
[mutable_eliminate]: 0.00069677
[jit_opt_b]: 5.568e-05, [1]
    [Cycle 1]: 4.739e-05, [2]
        [frontend_op_eliminate]: 2.09e-05
        [inline_after_opt_a]: 1.533e-05
[cconv]: 2.559e-05
[loop_unroll]: 0.0004425
[jit_opt_after_cconv]: 0.00014218, [1]
    [Cycle 1]: 0.00013559, [11]
        [c_1]: 2.173e-05
        [parameter_eliminate]: 2.76999e-06
        [updatestate_depend_eliminate]: 5.54e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [cse]: 1.899e-05
        [call_graph_tuple_transform]: 1.92e-05
        [tuple_list_get_item_eliminator]: 5.56998e-06
        [none_parameter_eliminate]: 1.86003e-06
        [renormalize]: 2.9002e-07
        [switch_simplify]: 5.25001e-06
[remove_dup_value]: 1.373e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 5.02e-06
[add_recomputation]: 4.803e-05
[cse_after_recomputation]: 1.998e-05, [1]
    [Cycle 1]: 1.469e-05, [1]
        [cse]: 8.96002e-06
[auto_monad_reorder]: 1.501e-05
[get_jit_bprop_graph]: 2.10002e-06
[rewriter_after_jit_bprop_graph]: 5.27001e-06
[opt_after_jit_grad]: 0.00047293
[symbol_engine_optimizer]: 7.95e-05, [1]
    [Cycle 1]: 7.208e-05, [6]
        [build]: 4.33999e-06
        [elim_shapecalc]: 7.90998e-06
        [elim_not_effective]: 1.343e-05
        [opt_reshape]: 5.98002e-06
        [fold_const_symbol]: 9.36002e-06
        [renormalize]: 5.09986e-07
[validate]: 3.826e-05
[backend_pass]: 8.80013e-07
[task_emit]: 0.00653989
[execute]: 9.39e-06
Sums
    bootstrap                            :     0.000310s :  1.86%
    type_inference                       :     0.005603s : 33.62%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000067s :  0.40%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.26%
    rewriter_before_opt_a                :     0.000076s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000080s :  0.48%
    jit_opt_a.loop_unroll                :     0.000036s :  0.22%
    jit_opt_a.a_1                        :     0.000701s :  4.21%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000877s :  5.26%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000041s :  0.25%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000697s :  4.18%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000026s :  0.15%
    loop_unroll                          :     0.000442s :  2.66%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000048s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000473s :  2.84%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006540s : 39.25%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000258    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 1.99% :     0.000005s :      3: substitution.graph_param_transform
86.48% :     0.000223s :      4: substitution.inline
 1.43% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.79% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.60% :     0.000004s :      2: substitution.replace_old_param
 5.49% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005524     2
74.90% :     0.004137s :      1: type_inference.infer
25.10% :     0.001386s :      1: type_inference.specialize
------[replace.]   0.000078     6
67.09% :     0.000053s :      4: replace.inline
32.91% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000234     6
94.41% :     0.000221s :      4: match.inline
 5.59% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.20% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.04% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.18% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.13% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 6.15% :     0.000007s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.34% :     0.000002s :      6: predicate.less_batch_normalization
 1.68% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.60% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.78% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.75% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.57% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     10: predicate.same_eliminate
 0.62% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.13% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.50% :     0.000010s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.44% :     0.000002s :     10: predicate.transpose_eliminate
 1.63% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.91% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001021    12
29.80% :     0.000304s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.03% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
65.17% :     0.000665s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023269    76
 0.22% :     0.000051s :      1: add_recomputation
 0.30% :     0.000071s :      1: auto_monad
 0.07% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.40% :     0.000325s :      1: bootstrap
 0.12% :     0.000028s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
27.23% :     0.006335s :      1: jit_opt_a
 0.62% :     0.000145s :      1: jit_opt_after_cconv
 0.25% :     0.000058s :      1: jit_opt_b
 1.94% :     0.000452s :      1: loop_unroll
 3.03% :     0.000706s :      1: mutable_eliminate
 4.02% :     0.000935s :     26: opt.transform.jit_opt_a
 0.21% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000030s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.07% :     0.000482s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000006s :      1: pre_auto_parallel
 0.20% :     0.000046s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.06% :     0.000479s :      1: renormalize.infer
 1.67% :     0.000389s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000039s :      1: rewriter_after_opt_a
 0.34% :     0.000078s :      1: rewriter_before_opt_a
 0.35% :     0.000082s :      1: symbol_engine_optimizer
28.18% :     0.006558s :      1: task_emit
24.15% :     0.005621s :      1: type_inference
 0.26% :     0.000060s :      1: validate


TotalTime = 0.0160553, [30]
[bootstrap]: 0.00031357
[type_inference]: 0.00683314
[event_method]: 2.166e-05
[auto_monad]: 8.913e-05
[graph_reusing]: 6.63e-06
[pre_auto_parallel]: 2.56998e-06
[py_interpret_to_execute]: 5.695e-05
[rewriter_before_opt_a]: 8.741e-05
[expand_dump_flag]: 3.16001e-06
[jit_opt_a]: 0.00612505, [2]
    [Cycle 1]: 0.00228574, [27]
        [switch_simplify]: 0.00011913
        [loop_unroll]: 3.488e-05
        [a_1]: 0.00077423
        [with_stream_mark]: 1.834e-05
        [recompute_prepare]: 6.89001e-06
        [updatestate_depend_eliminate]: 3.3e-06
        [updatestate_assign_eliminate]: 3.68e-06
        [updatestate_loads_eliminate]: 2.221e-05
        [parameter_eliminate]: 2.17999e-06
        [specialize_transform]: 6.86999e-06
        [updatestate_useless_node_eliminater]: 5.44998e-06
        [accelerated_algorithm]: 5.82999e-06
        [meta_shard_fg_expand]: 2.29001e-06
        [get_grad_eliminate_]: 5.17e-06
        [merge_forward]: 3.90998e-06
        [cell_reuse_recompute_pass]: 2e-05
        [cell_reuse_handle_not_recompute_node_pass]: 1.553e-05
        [j_node_and_user_rematch]: 9.54e-06
        [meta_fg_expand]: 2.66999e-06
        [replace_old_param]: 9.44998e-06
        [inline_without_move]: 5.36998e-06
        [renormalize]: 0.00094881
        [add_forward_monad_depend]: 6.51e-06
        [auto_monad_grad]: 2.26e-06
        [auto_monad_eliminator]: 1.617e-05
        [cse]: 3.146e-05
        [replace_applicator]: 1.709e-05
    [Cycle 2]: 0.00033506, [27]
        [switch_simplify]: 5.34e-06
        [loop_unroll]: 5.08002e-06
        [a_1]: 9.728e-05
        [with_stream_mark]: 1.113e-05
        [recompute_prepare]: 5.40001e-06
        [updatestate_depend_eliminate]: 2.81e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [parameter_eliminate]: 1.39e-06
        [specialize_transform]: 5.57001e-06
        [updatestate_useless_node_eliminater]: 5.18002e-06
        [accelerated_algorithm]: 5.37999e-06
        [meta_shard_fg_expand]: 1.52001e-06
        [get_grad_eliminate_]: 4.85999e-06
        [merge_forward]: 2.79001e-06
        [cell_reuse_recompute_pass]: 1.41998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.501e-05
        [j_node_and_user_rematch]: 8.17e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 8.18999e-06
        [inline_without_move]: 5.49e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.30999e-06
        [auto_monad_grad]: 6.50005e-07
        [auto_monad_eliminator]: 5.09e-06
        [cse]: 1.004e-05
        [replace_applicator]: 5.66998e-06
[py_interpret_to_execute_after_opt_a]: 1.176e-05
[rewriter_after_opt_a]: 3.634e-05
[convert_after_rewriter]: 7.6e-06
[order_py_execute_after_rewriter]: 5.35999e-06
[mutable_eliminate]: 0.00081282
[jit_opt_b]: 6.789e-05, [1]
    [Cycle 1]: 5.956e-05, [2]
        [frontend_op_eliminate]: 1.829e-05
        [inline_after_opt_a]: 2.926e-05
[cconv]: 2.587e-05
[loop_unroll]: 0.00048099
[jit_opt_after_cconv]: 0.00014454, [1]
    [Cycle 1]: 0.00013749, [11]
        [c_1]: 2.234e-05
        [parameter_eliminate]: 2.53998e-06
        [updatestate_depend_eliminate]: 5.49e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [cse]: 1.916e-05
        [call_graph_tuple_transform]: 1.894e-05
        [tuple_list_get_item_eliminator]: 5.28002e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 7.80012e-07
        [switch_simplify]: 5.39998e-06
[remove_dup_value]: 1.413e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 4.92999e-06
[add_recomputation]: 4.711e-05
[cse_after_recomputation]: 2.183e-05, [1]
    [Cycle 1]: 1.609e-05, [1]
        [cse]: 1.023e-05
[auto_monad_reorder]: 1.637e-05
[get_jit_bprop_graph]: 2.13002e-06
[rewriter_after_jit_bprop_graph]: 5.87001e-06
[opt_after_jit_grad]: 0.00052819
[symbol_engine_optimizer]: 7.381e-05, [1]
    [Cycle 1]: 6.775e-05, [6]
        [build]: 3.9e-06
        [elim_shapecalc]: 7.82e-06
        [elim_not_effective]: 1.286e-05
        [opt_reshape]: 5.79e-06
        [fold_const_symbol]: 8.39002e-06
        [renormalize]: 4.40021e-07
[validate]: 3.424e-05
Sums
    bootstrap                            :     0.000314s :  2.62%
    type_inference                       :     0.006833s : 57.16%
    event_method                         :     0.000022s :  0.18%
    auto_monad                           :     0.000089s :  0.75%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000057s :  0.48%
    rewriter_before_opt_a                :     0.000087s :  0.73%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000124s :  1.04%
    jit_opt_a.loop_unroll                :     0.000040s :  0.33%
    jit_opt_a.a_1                        :     0.000872s :  7.29%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000024s :  0.20%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000021s :  0.18%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.000949s :  7.94%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.18%
    jit_opt_a.cse                        :     0.000041s :  0.35%
    jit_opt_a.replace_applicator         :     0.000023s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.10%
    rewriter_after_opt_a                 :     0.000036s :  0.30%
    convert_after_rewriter               :     0.000008s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000813s :  6.80%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000029s :  0.24%
    cconv                                :     0.000026s :  0.22%
    loop_unroll                          :     0.000481s :  4.02%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.39%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000528s :  4.42%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.29%


Time group info:
------[substitution.]   0.000336    26
 0.57% :     0.000002s :      2: substitution.elim_not_effective
 0.38% :     0.000001s :      2: substitution.fold_const_symbol
 1.66% :     0.000006s :      3: substitution.graph_param_transform
79.72% :     0.000268s :      6: substitution.inline
 0.88% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.57% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.32% :     0.000004s :      2: substitution.replace_old_param
 2.98% :     0.000010s :      1: substitution.switch_simplify
10.92% :     0.000037s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006763     2
81.51% :     0.005513s :      1: type_inference.infer
18.49% :     0.001250s :      1: type_inference.specialize
------[replace.]   0.000142     9
49.93% :     0.000071s :      6: replace.inline
24.68% :     0.000035s :      1: replace.switch_simplify
25.39% :     0.000036s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000308     9
85.56% :     0.000264s :      6: match.inline
 2.99% :     0.000009s :      1: match.switch_simplify
11.45% :     0.000035s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000141   801
 1.31% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.13% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.33% :     0.000002s :     12: predicate.addn_check_dump
 1.47% :     0.000002s :     12: predicate.addn_zero_filter
 2.36% :     0.000003s :     12: predicate.arithmetic_simplify
 1.25% :     0.000002s :     12: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     12: predicate.compare_switch_simplify
 1.43% :     0.000002s :     12: predicate.depend_value_elim
 1.17% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.50% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.66% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.09% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.11% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.38% :     0.000002s :     12: predicate.environ_get_eliminate
 1.35% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.17% :     0.000007s :     26: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.99% :     0.000001s :      6: predicate.less_batch_normalization
 1.70% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     17: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.17% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.19% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     12: predicate.merge_addn
 1.26% :     0.000002s :     12: predicate.minmaximum_grad
 1.50% :     0.000002s :      3: predicate.mutable_eliminate
 0.38% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     17: predicate.partial_eliminate
 1.19% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     12: predicate.reduce_eliminate
 1.87% :     0.000003s :     14: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     20: predicate.replace_applicator
 0.97% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     12: predicate.reshape_eliminate
 1.17% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.60% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     12: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.26% :     0.000002s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.53% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.56% :     0.000004s :     20: predicate.switch_defer_inline
 2.30% :     0.000003s :     20: predicate.switch_layer_defer_inline
 7.89% :     0.000011s :     65: predicate.switch_simplify
 1.21% :     0.000002s :     12: predicate.tile_eliminate
 1.15% :     0.000002s :     12: predicate.transpose_eliminate
 1.48% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.41% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.22% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     12: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.94% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000928    15
45.05% :     0.000418s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.86% :     0.000119s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.09% :     0.000391s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018239    72
 0.27% :     0.000050s :      1: add_recomputation
 0.51% :     0.000093s :      1: auto_monad
 0.10% :     0.000019s :      1: auto_monad_reorder
 1.82% :     0.000332s :      1: bootstrap
 0.16% :     0.000029s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.13% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.14% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.61% :     0.006130s :      1: jit_opt_a
 0.81% :     0.000147s :      1: jit_opt_after_cconv
 0.39% :     0.000071s :      1: jit_opt_b
 2.68% :     0.000489s :      1: loop_unroll
 4.50% :     0.000821s :      1: mutable_eliminate
 6.36% :     0.001160s :     26: opt.transform.jit_opt_a
 0.26% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.23% :     0.000041s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.94% :     0.000537s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.33% :     0.000060s :      1: py_interpret_to_execute
 0.08% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000017s :      1: remove_dup_value
 2.58% :     0.000470s :      1: renormalize.infer
 2.57% :     0.000469s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000039s :      1: rewriter_after_opt_a
 0.50% :     0.000090s :      1: rewriter_before_opt_a
 0.42% :     0.000077s :      1: symbol_engine_optimizer
37.53% :     0.006845s :      1: type_inference


TotalTime = 0.0163742, [33]
[bootstrap]: 0.00024259
[type_inference]: 0.00380327
[event_method]: 1.501e-05
[auto_monad]: 3.899e-05
[graph_reusing]: 3.63999e-06
[pre_auto_parallel]: 1.74e-06
[py_interpret_to_execute]: 2.523e-05
[rewriter_before_opt_a]: 6.245e-05
[expand_dump_flag]: 2.15002e-06
[jit_opt_a]: 0.00464987, [2]
    [Cycle 1]: 0.00146414, [27]
        [switch_simplify]: 4.268e-05
        [loop_unroll]: 2.85e-05
        [a_1]: 0.00046075
        [with_stream_mark]: 1.689e-05
        [recompute_prepare]: 7.14001e-06
        [updatestate_depend_eliminate]: 3.55e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.30002e-06
        [parameter_eliminate]: 1.73002e-06
        [specialize_transform]: 5.61998e-06
        [updatestate_useless_node_eliminater]: 4.72e-06
        [accelerated_algorithm]: 5.12999e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 4.90999e-06
        [merge_forward]: 3.48999e-06
        [cell_reuse_recompute_pass]: 1.57999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.318e-05
        [j_node_and_user_rematch]: 8.08999e-06
        [meta_fg_expand]: 2.10002e-06
        [replace_old_param]: 8.59e-06
        [inline_without_move]: 5.22999e-06
        [renormalize]: 0.00062328
        [add_forward_monad_depend]: 5.70001e-06
        [auto_monad_grad]: 2.14999e-06
        [auto_monad_eliminator]: 1.319e-05
        [cse]: 2.278e-05
        [replace_applicator]: 1.11e-05
    [Cycle 2]: 0.00030822, [27]
        [switch_simplify]: 5.35001e-06
        [loop_unroll]: 4.64998e-06
        [a_1]: 8.733e-05
        [with_stream_mark]: 8.38999e-06
        [recompute_prepare]: 4.75999e-06
        [updatestate_depend_eliminate]: 2.34999e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.79998e-06
        [parameter_eliminate]: 1.06002e-06
        [specialize_transform]: 4.78001e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 4.81997e-06
        [meta_shard_fg_expand]: 1.01002e-06
        [get_grad_eliminate_]: 4.48999e-06
        [merge_forward]: 2.31998e-06
        [cell_reuse_recompute_pass]: 1.84e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.221e-05
        [j_node_and_user_rematch]: 7.58001e-06
        [meta_fg_expand]: 1.41002e-06
        [replace_old_param]: 6.69999e-06
        [inline_without_move]: 4.54002e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 6.59988e-07
        [auto_monad_eliminator]: 4.68001e-06
        [cse]: 9.41003e-06
        [replace_applicator]: 5.14e-06
[py_interpret_to_execute_after_opt_a]: 8.60001e-06
[rewriter_after_opt_a]: 3.221e-05
[convert_after_rewriter]: 7.31001e-06
[order_py_execute_after_rewriter]: 4.76002e-06
[mutable_eliminate]: 0.00050033
[jit_opt_b]: 4.754e-05, [1]
    [Cycle 1]: 4.07e-05, [2]
        [frontend_op_eliminate]: 1.562e-05
        [inline_after_opt_a]: 1.418e-05
[cconv]: 2.394e-05
[loop_unroll]: 0.0003979
[jit_opt_after_cconv]: 0.00013186, [1]
    [Cycle 1]: 0.00012583, [11]
        [c_1]: 2.007e-05
        [parameter_eliminate]: 2.23998e-06
        [updatestate_depend_eliminate]: 5.17e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 1.656e-05
        [call_graph_tuple_transform]: 1.693e-05
        [tuple_list_get_item_eliminator]: 5.04003e-06
        [none_parameter_eliminate]: 1.62001e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 4.94e-06
[remove_dup_value]: 1.354e-05
[partial_unused_args_eliminate]: 2.04e-06
[environ_conv]: 4.66002e-06
[add_recomputation]: 4.672e-05
[cse_after_recomputation]: 1.995e-05, [1]
    [Cycle 1]: 1.454e-05, [1]
        [cse]: 8.81997e-06
[auto_monad_reorder]: 1.44e-05
[get_jit_bprop_graph]: 1.77999e-06
[rewriter_after_jit_bprop_graph]: 4.34002e-06
[opt_after_jit_grad]: 0.00042734
[symbol_engine_optimizer]: 7.105e-05, [1]
    [Cycle 1]: 6.501e-05, [6]
        [build]: 3.29001e-06
        [elim_shapecalc]: 7.80998e-06
        [elim_not_effective]: 1.29e-05
        [opt_reshape]: 5.64998e-06
        [fold_const_symbol]: 7.97e-06
        [renormalize]: 5.19998e-07
[validate]: 3.297e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00556283
[execute]: 7.67002e-06
Sums
    bootstrap                            :     0.000243s :  1.87%
    type_inference                       :     0.003803s : 29.37%
    event_method                         :     0.000015s :  0.12%
    auto_monad                           :     0.000039s :  0.30%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000025s :  0.19%
    rewriter_before_opt_a                :     0.000062s :  0.48%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000048s :  0.37%
    jit_opt_a.loop_unroll                :     0.000033s :  0.26%
    jit_opt_a.a_1                        :     0.000548s :  4.23%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000623s :  4.81%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.14%
    jit_opt_a.cse                        :     0.000032s :  0.25%
    jit_opt_a.replace_applicator         :     0.000016s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000500s :  3.86%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000398s :  3.07%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.36%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000427s :  3.30%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000033s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005563s : 42.96%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000156    23
 1.33% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000001s :      2: substitution.fold_const_symbol
 3.06% :     0.000005s :      3: substitution.graph_param_transform
79.96% :     0.000125s :      4: substitution.inline
 1.76% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.80% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.49% :     0.000004s :      2: substitution.replace_old_param
 7.84% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003757     2
81.66% :     0.003068s :      1: type_inference.infer
18.34% :     0.000689s :      1: type_inference.specialize
------[replace.]   0.000057     6
66.71% :     0.000038s :      4: replace.inline
33.29% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000134     6
91.61% :     0.000122s :      4: match.inline
 8.39% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000108   691
 1.36% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.44% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     10: predicate.addn_check_dump
 1.66% :     0.000002s :     10: predicate.addn_zero_filter
 2.03% :     0.000002s :     10: predicate.arithmetic_simplify
 1.37% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000001s :     10: predicate.depend_value_elim
 1.33% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.32% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.19% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.90% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.19% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.83% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.87% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000002s :     15: predicate.partial_eliminate
 1.24% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.82% :     0.000002s :     18: predicate.replace_applicator
 0.82% :     0.000001s :      6: predicate.replace_old_param
 0.43% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.66% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.33% :     0.000008s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.54% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.79% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000490    12
48.30% :     0.000237s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.90% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.80% :     0.000224s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017839    76
 0.28% :     0.000050s :      1: add_recomputation
 0.24% :     0.000042s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.42% :     0.000254s :      1: bootstrap
 0.15% :     0.000026s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000018s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
26.08% :     0.004653s :      1: jit_opt_a
 0.75% :     0.000135s :      1: jit_opt_after_cconv
 0.28% :     0.000050s :      1: jit_opt_b
 2.27% :     0.000405s :      1: loop_unroll
 2.85% :     0.000508s :      1: mutable_eliminate
 4.14% :     0.000738s :     26: opt.transform.jit_opt_a
 0.25% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.43% :     0.000434s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000028s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 2.04% :     0.000363s :      1: renormalize.infer
 1.41% :     0.000252s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000035s :      1: rewriter_after_opt_a
 0.36% :     0.000065s :      1: rewriter_before_opt_a
 0.41% :     0.000073s :      1: symbol_engine_optimizer
31.23% :     0.005571s :      1: task_emit
21.38% :     0.003813s :      1: type_inference
 0.29% :     0.000052s :      1: validate


TotalTime = 0.0196325, [33]
[bootstrap]: 0.00031399
[type_inference]: 0.00444456
[event_method]: 1.702e-05
[auto_monad]: 6.737e-05
[graph_reusing]: 6.66e-06
[pre_auto_parallel]: 2.69999e-06
[py_interpret_to_execute]: 3.388e-05
[rewriter_before_opt_a]: 9.848e-05
[expand_dump_flag]: 3.95e-06
[jit_opt_a]: 0.00559677, [2]
    [Cycle 1]: 0.00175315, [27]
        [switch_simplify]: 5.664e-05
        [loop_unroll]: 2.848e-05
        [a_1]: 0.00058279
        [with_stream_mark]: 1.76e-05
        [recompute_prepare]: 7.15e-06
        [updatestate_depend_eliminate]: 3.51001e-06
        [updatestate_assign_eliminate]: 3.16999e-06
        [updatestate_loads_eliminate]: 2.94001e-06
        [parameter_eliminate]: 2.06e-06
        [specialize_transform]: 5.80002e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.61998e-06
        [meta_shard_fg_expand]: 2.15002e-06
        [get_grad_eliminate_]: 5.22e-06
        [merge_forward]: 4.00998e-06
        [cell_reuse_recompute_pass]: 1.02998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.468e-05
        [j_node_and_user_rematch]: 9.96998e-06
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 9.17001e-06
        [inline_without_move]: 5.11997e-06
        [renormalize]: 0.00073703
        [add_forward_monad_depend]: 5.75001e-06
        [auto_monad_grad]: 2.56998e-06
        [auto_monad_eliminator]: 1.455e-05
        [cse]: 2.987e-05
        [replace_applicator]: 1.439e-05
    [Cycle 2]: 0.00032597, [27]
        [switch_simplify]: 5.76e-06
        [loop_unroll]: 4.84e-06
        [a_1]: 8.969e-05
        [with_stream_mark]: 9.49999e-06
        [recompute_prepare]: 5.35999e-06
        [updatestate_depend_eliminate]: 2.66e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 5.09e-06
        [updatestate_useless_node_eliminater]: 4.73001e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 1.22999e-06
        [get_grad_eliminate_]: 4.38001e-06
        [merge_forward]: 2.96001e-06
        [cell_reuse_recompute_pass]: 1.80001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.188e-05
        [j_node_and_user_rematch]: 7.91001e-06
        [meta_fg_expand]: 1.56998e-06
        [replace_old_param]: 7.28e-06
        [inline_without_move]: 4.63999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.97001e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 5.37001e-06
        [cse]: 1.112e-05
        [replace_applicator]: 4.95001e-06
[py_interpret_to_execute_after_opt_a]: 9.90002e-06
[rewriter_after_opt_a]: 3.755e-05
[convert_after_rewriter]: 7.02997e-06
[order_py_execute_after_rewriter]: 4.82e-06
[mutable_eliminate]: 0.00054575
[jit_opt_b]: 5.286e-05, [1]
    [Cycle 1]: 4.619e-05, [2]
        [frontend_op_eliminate]: 1.998e-05
        [inline_after_opt_a]: 1.483e-05
[cconv]: 2.44e-05
[loop_unroll]: 0.00041906
[jit_opt_after_cconv]: 0.00013776, [1]
    [Cycle 1]: 0.00013138, [11]
        [c_1]: 2.139e-05
        [parameter_eliminate]: 2.13998e-06
        [updatestate_depend_eliminate]: 4.89e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 1.839e-05
        [call_graph_tuple_transform]: 1.842e-05
        [tuple_list_get_item_eliminator]: 5.25999e-06
        [none_parameter_eliminate]: 1.32e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 5.02999e-06
[remove_dup_value]: 1.454e-05
[partial_unused_args_eliminate]: 2.58998e-06
[environ_conv]: 5.30001e-06
[add_recomputation]: 4.94e-05
[cse_after_recomputation]: 1.985e-05, [1]
    [Cycle 1]: 1.459e-05, [1]
        [cse]: 9.15999e-06
[auto_monad_reorder]: 1.487e-05
[get_jit_bprop_graph]: 2.04999e-06
[rewriter_after_jit_bprop_graph]: 4.97e-06
[opt_after_jit_grad]: 0.00052814
[symbol_engine_optimizer]: 7.641e-05, [1]
    [Cycle 1]: 6.801e-05, [6]
        [build]: 4.94e-06
        [elim_shapecalc]: 7.45998e-06
        [elim_not_effective]: 1.277e-05
        [opt_reshape]: 6.04999e-06
        [fold_const_symbol]: 8.80999e-06
        [renormalize]: 1.05001e-06
[validate]: 4.336e-05
[backend_pass]: 1.07e-06
[task_emit]: 0.00681656
[execute]: 9.70002e-06
Sums
    bootstrap                            :     0.000314s :  2.03%
    type_inference                       :     0.004445s : 28.71%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000067s :  0.44%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.22%
    rewriter_before_opt_a                :     0.000098s :  0.64%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000062s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.22%
    jit_opt_a.a_1                        :     0.000672s :  4.34%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000737s :  4.76%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000041s :  0.26%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000038s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000546s :  3.53%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000024s :  0.16%
    loop_unroll                          :     0.000419s :  2.71%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000528s :  3.41%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.28%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006817s : 44.03%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000255    23
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000002s :      2: substitution.fold_const_symbol
 2.10% :     0.000005s :      3: substitution.graph_param_transform
86.82% :     0.000221s :      4: substitution.inline
 1.46% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.84% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.72% :     0.000004s :      2: substitution.replace_old_param
 4.59% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004371     2
80.24% :     0.003507s :      1: type_inference.infer
19.76% :     0.000864s :      1: type_inference.specialize
------[replace.]   0.000070     6
66.67% :     0.000047s :      4: replace.inline
33.33% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000230     6
95.35% :     0.000219s :      4: match.inline
 4.65% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.72% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.61% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 1.88% :     0.000002s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.55% :     0.000001s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.33% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.38% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.60% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.13% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.39% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.90% :     0.000009s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.66% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000623    12
43.14% :     0.000269s :      5: func_graph_cloner_run.FuncGraphClonerGraph
12.46% :     0.000078s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.40% :     0.000276s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021360    76
 0.24% :     0.000052s :      1: add_recomputation
 0.33% :     0.000071s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.57% :     0.000335s :      1: bootstrap
 0.13% :     0.000027s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.22% :     0.005600s :      1: jit_opt_a
 0.66% :     0.000140s :      1: jit_opt_after_cconv
 0.26% :     0.000055s :      1: jit_opt_b
 2.00% :     0.000426s :      1: loop_unroll
 2.60% :     0.000555s :      1: mutable_eliminate
 4.14% :     0.000883s :     26: opt.transform.jit_opt_a
 0.22% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000028s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.52% :     0.000539s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.83% :     0.000391s :      1: renormalize.infer
 1.59% :     0.000340s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000041s :      1: rewriter_after_opt_a
 0.48% :     0.000102s :      1: rewriter_before_opt_a
 0.37% :     0.000079s :      1: symbol_engine_optimizer
32.01% :     0.006837s :      1: task_emit
20.87% :     0.004457s :      1: type_inference
 0.31% :     0.000066s :      1: validate


TotalTime = 0.0151617, [30]
[bootstrap]: 0.00028157
[type_inference]: 0.00675064
[event_method]: 2.099e-05
[auto_monad]: 8.293e-05
[graph_reusing]: 6.37001e-06
[pre_auto_parallel]: 2.82002e-06
[py_interpret_to_execute]: 6.159e-05
[rewriter_before_opt_a]: 8.966e-05
[expand_dump_flag]: 3.18e-06
[jit_opt_a]: 0.00580189, [2]
    [Cycle 1]: 0.00198976, [27]
        [switch_simplify]: 0.00012015
        [loop_unroll]: 4.647e-05
        [a_1]: 0.00074002
        [with_stream_mark]: 1.802e-05
        [recompute_prepare]: 7.13998e-06
        [updatestate_depend_eliminate]: 3.55e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.83e-06
        [parameter_eliminate]: 2.04999e-06
        [specialize_transform]: 6.26998e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 6.17999e-06
        [meta_shard_fg_expand]: 2.23998e-06
        [get_grad_eliminate_]: 5.36998e-06
        [merge_forward]: 3.56001e-06
        [cell_reuse_recompute_pass]: 1.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.52e-05
        [j_node_and_user_rematch]: 9.03002e-06
        [meta_fg_expand]: 2.84999e-06
        [replace_old_param]: 8.54e-06
        [inline_without_move]: 5.44998e-06
        [renormalize]: 0.00074526
        [add_forward_monad_depend]: 5.47999e-06
        [auto_monad_grad]: 1.91e-06
        [auto_monad_eliminator]: 1.313e-05
        [cse]: 2.577e-05
        [replace_applicator]: 1.13e-05
    [Cycle 2]: 0.00031423, [27]
        [switch_simplify]: 5.52999e-06
        [loop_unroll]: 4.82998e-06
        [a_1]: 9.047e-05
        [with_stream_mark]: 8.64998e-06
        [recompute_prepare]: 5.14e-06
        [updatestate_depend_eliminate]: 2.47001e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 5.25001e-06
        [updatestate_useless_node_eliminater]: 4.76002e-06
        [accelerated_algorithm]: 4.85001e-06
        [meta_shard_fg_expand]: 1.14998e-06
        [get_grad_eliminate_]: 4.69998e-06
        [merge_forward]: 2.89001e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.232e-05
        [j_node_and_user_rematch]: 7.93001e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 7.11001e-06
        [inline_without_move]: 4.97999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 4.85999e-06
        [cse]: 1.055e-05
        [replace_applicator]: 5.04e-06
[py_interpret_to_execute_after_opt_a]: 9.52001e-06
[rewriter_after_opt_a]: 3.312e-05
[convert_after_rewriter]: 6.61e-06
[order_py_execute_after_rewriter]: 5.74e-06
[mutable_eliminate]: 0.00050399
[jit_opt_b]: 5.004e-05, [1]
    [Cycle 1]: 4.346e-05, [2]
        [frontend_op_eliminate]: 1.623e-05
        [inline_after_opt_a]: 1.542e-05
[cconv]: 2.409e-05
[loop_unroll]: 0.00043984
[jit_opt_after_cconv]: 0.00013996, [1]
    [Cycle 1]: 0.00013314, [11]
        [c_1]: 2.248e-05
        [parameter_eliminate]: 2.44999e-06
        [updatestate_depend_eliminate]: 5.42001e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 1.678e-05
        [call_graph_tuple_transform]: 1.778e-05
        [tuple_list_get_item_eliminator]: 5.27999e-06
        [none_parameter_eliminate]: 1.62001e-06
        [renormalize]: 3.80009e-07
        [switch_simplify]: 5.47999e-06
[remove_dup_value]: 1.413e-05
[partial_unused_args_eliminate]: 2.66e-06
[environ_conv]: 5.05001e-06
[add_recomputation]: 5.486e-05
[cse_after_recomputation]: 2.202e-05, [1]
    [Cycle 1]: 1.607e-05, [1]
        [cse]: 1.001e-05
[auto_monad_reorder]: 1.512e-05
[get_jit_bprop_graph]: 1.52999e-06
[rewriter_after_jit_bprop_graph]: 4.78001e-06
[opt_after_jit_grad]: 0.000463
[symbol_engine_optimizer]: 7.312e-05, [1]
    [Cycle 1]: 6.715e-05, [6]
        [build]: 3.04999e-06
        [elim_shapecalc]: 7.85e-06
        [elim_not_effective]: 1.303e-05
        [opt_reshape]: 5.42999e-06
        [fold_const_symbol]: 9.29e-06
        [renormalize]: 3.9002e-07
[validate]: 3.196e-05
Sums
    bootstrap                            :     0.000282s :  2.54%
    type_inference                       :     0.006751s : 60.82%
    event_method                         :     0.000021s :  0.19%
    auto_monad                           :     0.000083s :  0.75%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000062s :  0.55%
    rewriter_before_opt_a                :     0.000090s :  0.81%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000126s :  1.13%
    jit_opt_a.loop_unroll                :     0.000051s :  0.46%
    jit_opt_a.a_1                        :     0.000830s :  7.48%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000745s :  6.72%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.16%
    jit_opt_a.cse                        :     0.000036s :  0.33%
    jit_opt_a.replace_applicator         :     0.000016s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.09%
    rewriter_after_opt_a                 :     0.000033s :  0.30%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000504s :  4.54%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.14%
    cconv                                :     0.000024s :  0.22%
    loop_unroll                          :     0.000440s :  3.96%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.13%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000055s :  0.49%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000463s :  4.17%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.29%


Time group info:
------[substitution.]   0.000302    26
 0.66% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 1.70% :     0.000005s :      3: substitution.graph_param_transform
86.79% :     0.000262s :      6: substitution.inline
 0.96% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.53% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.13% :     0.000003s :      2: substitution.replace_old_param
 2.57% :     0.000008s :      1: substitution.switch_simplify
 4.20% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006687     2
82.77% :     0.005535s :      1: type_inference.infer
17.23% :     0.001152s :      1: type_inference.specialize
------[replace.]   0.000147     9
52.18% :     0.000077s :      6: replace.inline
24.79% :     0.000036s :      1: replace.switch_simplify
23.03% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000277     9
93.34% :     0.000259s :      6: match.inline
 2.49% :     0.000007s :      1: match.switch_simplify
 4.17% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   801
 1.37% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     12: predicate.addn_check_dump
 1.42% :     0.000002s :     12: predicate.addn_zero_filter
 1.92% :     0.000003s :     12: predicate.arithmetic_simplify
 1.40% :     0.000002s :     12: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000002s :     12: predicate.compare_switch_simplify
 1.45% :     0.000002s :     12: predicate.depend_value_elim
 1.15% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.49% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.74% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.38% :     0.000002s :     12: predicate.environ_get_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.40% :     0.000007s :     26: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     17: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.53% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.93% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000002s :     12: predicate.merge_addn
 1.18% :     0.000002s :     12: predicate.minmaximum_grad
 1.42% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     17: predicate.partial_eliminate
 1.22% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     12: predicate.reduce_eliminate
 1.45% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     20: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     12: predicate.reshape_eliminate
 1.22% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.91% :     0.000004s :     20: predicate.switch_defer_inline
 2.33% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.54% :     0.000011s :     65: predicate.switch_simplify
 1.26% :     0.000002s :     12: predicate.tile_eliminate
 1.38% :     0.000002s :     12: predicate.transpose_eliminate
 1.51% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.76% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     12: predicate.value_based_eliminate
 0.37% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000903    15
48.47% :     0.000438s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.25% :     0.000111s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.27% :     0.000355s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017091    72
 0.34% :     0.000058s :      1: add_recomputation
 0.51% :     0.000086s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 1.72% :     0.000293s :      1: bootstrap
 0.16% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.96% :     0.005805s :      1: jit_opt_a
 0.83% :     0.000143s :      1: jit_opt_after_cconv
 0.31% :     0.000053s :      1: jit_opt_b
 2.61% :     0.000447s :      1: loop_unroll
 2.99% :     0.000512s :      1: mutable_eliminate
 6.56% :     0.001121s :     26: opt.transform.jit_opt_a
 0.28% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.75% :     0.000471s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.38% :     0.000064s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.52% :     0.000431s :      1: renormalize.infer
 1.80% :     0.000307s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000036s :      1: rewriter_after_opt_a
 0.54% :     0.000092s :      1: rewriter_before_opt_a
 0.44% :     0.000076s :      1: symbol_engine_optimizer
39.57% :     0.006763s :      1: type_inference


TotalTime = 0.0218315, [33]
[bootstrap]: 0.0003609
[type_inference]: 0.00473675
[event_method]: 1.842e-05
[auto_monad]: 6.748e-05
[graph_reusing]: 5.73002e-06
[pre_auto_parallel]: 2.62001e-06
[py_interpret_to_execute]: 3.503e-05
[rewriter_before_opt_a]: 7.431e-05
[expand_dump_flag]: 3.36001e-06
[jit_opt_a]: 0.00570112, [2]
    [Cycle 1]: 0.0017909, [27]
        [switch_simplify]: 5.912e-05
        [loop_unroll]: 3.08e-05
        [a_1]: 0.00055722
        [with_stream_mark]: 2.318e-05
        [recompute_prepare]: 8.03999e-06
        [updatestate_depend_eliminate]: 4.07e-06
        [updatestate_assign_eliminate]: 3.32002e-06
        [updatestate_loads_eliminate]: 2.50002e-06
        [parameter_eliminate]: 2.31e-06
        [specialize_transform]: 6.46999e-06
        [updatestate_useless_node_eliminater]: 5.08002e-06
        [accelerated_algorithm]: 6.02999e-06
        [meta_shard_fg_expand]: 2.43e-06
        [get_grad_eliminate_]: 5.45001e-06
        [merge_forward]: 3.76999e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.458e-05
        [j_node_and_user_rematch]: 9.56e-06
        [meta_fg_expand]: 2.19001e-06
        [replace_old_param]: 9.47001e-06
        [inline_without_move]: 5.42001e-06
        [renormalize]: 0.00079382
        [add_forward_monad_depend]: 5.71e-06
        [auto_monad_grad]: 2.52001e-06
        [auto_monad_eliminator]: 1.342e-05
        [cse]: 3.08e-05
        [replace_applicator]: 1.209e-05
    [Cycle 2]: 0.00032645, [27]
        [switch_simplify]: 6.21e-06
        [loop_unroll]: 5.14e-06
        [a_1]: 9.08e-05
        [with_stream_mark]: 1.022e-05
        [recompute_prepare]: 5.17e-06
        [updatestate_depend_eliminate]: 2.65002e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.27999e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.19e-06
        [meta_shard_fg_expand]: 1.39998e-06
        [get_grad_eliminate_]: 4.47998e-06
        [merge_forward]: 2.75002e-06
        [cell_reuse_recompute_pass]: 2.16998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.277e-05
        [j_node_and_user_rematch]: 8.2e-06
        [meta_fg_expand]: 1.84e-06
        [replace_old_param]: 7.21001e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.11002e-06
        [auto_monad_grad]: 9.30013e-07
        [auto_monad_eliminator]: 4.89e-06
        [cse]: 1.013e-05
        [replace_applicator]: 5.11997e-06
[py_interpret_to_execute_after_opt_a]: 1.027e-05
[rewriter_after_opt_a]: 3.985e-05
[convert_after_rewriter]: 6.96999e-06
[order_py_execute_after_rewriter]: 5.32001e-06
[mutable_eliminate]: 0.00054175
[jit_opt_b]: 5.572e-05, [1]
    [Cycle 1]: 4.847e-05, [2]
        [frontend_op_eliminate]: 2.1e-05
        [inline_after_opt_a]: 1.593e-05
[cconv]: 2.598e-05
[loop_unroll]: 0.00044017
[jit_opt_after_cconv]: 0.00014341, [1]
    [Cycle 1]: 0.00013681, [11]
        [c_1]: 2.195e-05
        [parameter_eliminate]: 2.69001e-06
        [updatestate_depend_eliminate]: 5.74e-06
        [updatestate_assign_eliminate]: 2.38998e-06
        [updatestate_loads_eliminate]: 2.21998e-06
        [cse]: 1.976e-05
        [call_graph_tuple_transform]: 1.814e-05
        [tuple_list_get_item_eliminator]: 5.27001e-06
        [none_parameter_eliminate]: 1.65001e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 5.35001e-06
[remove_dup_value]: 1.417e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 5.82999e-06
[add_recomputation]: 5.451e-05
[cse_after_recomputation]: 2.088e-05, [1]
    [Cycle 1]: 1.531e-05, [1]
        [cse]: 9.22001e-06
[auto_monad_reorder]: 1.565e-05
[get_jit_bprop_graph]: 2.04999e-06
[rewriter_after_jit_bprop_graph]: 5.61998e-06
[opt_after_jit_grad]: 0.00048967
[symbol_engine_optimizer]: 7.591e-05, [1]
    [Cycle 1]: 6.873e-05, [6]
        [build]: 3.63999e-06
        [elim_shapecalc]: 8.13999e-06
        [elim_not_effective]: 1.327e-05
        [opt_reshape]: 5.77001e-06
        [fold_const_symbol]: 9.60001e-06
        [renormalize]: 7.09988e-07
[validate]: 3.69e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.00861807
[execute]: 8.76997e-06
Sums
    bootstrap                            :     0.000361s :  2.05%
    type_inference                       :     0.004737s : 26.87%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000067s :  0.38%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000035s :  0.20%
    rewriter_before_opt_a                :     0.000074s :  0.42%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.37%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000648s :  3.68%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000794s :  4.50%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.10%
    jit_opt_a.cse                        :     0.000041s :  0.23%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000040s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000542s :  3.07%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000026s :  0.15%
    loop_unroll                          :     0.000440s :  2.50%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000055s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000490s :  2.78%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008618s : 48.88%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000221    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.27% :     0.000005s :      3: substitution.graph_param_transform
85.45% :     0.000189s :      4: substitution.inline
 1.49% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.90% :     0.000004s :      2: substitution.replace_old_param
 5.42% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004670     2
81.72% :     0.003816s :      1: type_inference.infer
18.28% :     0.000854s :      1: type_inference.specialize
------[replace.]   0.000072     6
67.61% :     0.000048s :      4: replace.inline
32.39% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000197     6
94.49% :     0.000186s :      4: match.inline
 5.51% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.18% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.39% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.49% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.22% :     0.000001s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.63% :     0.000002s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 6.20% :     0.000007s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.37% :     0.000002s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.03% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.79% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     15: predicate.partial_eliminate
 1.23% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.33% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.27% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.30% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.69% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.00% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.87% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000626    12
45.14% :     0.000283s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.56% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.30% :     0.000302s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023601    76
 0.24% :     0.000057s :      1: add_recomputation
 0.30% :     0.000071s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.61% :     0.000381s :      1: bootstrap
 0.12% :     0.000029s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
24.17% :     0.005704s :      1: jit_opt_a
 0.62% :     0.000147s :      1: jit_opt_after_cconv
 0.25% :     0.000058s :      1: jit_opt_b
 1.90% :     0.000448s :      1: loop_unroll
 2.34% :     0.000551s :      1: mutable_eliminate
 3.66% :     0.000865s :     26: opt.transform.jit_opt_a
 0.20% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000030s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.11% :     0.000497s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000038s :      1: py_interpret_to_execute
 0.05% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.67% :     0.000395s :      1: renormalize.infer
 1.66% :     0.000391s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000043s :      1: rewriter_after_opt_a
 0.33% :     0.000077s :      1: rewriter_before_opt_a
 0.33% :     0.000078s :      1: symbol_engine_optimizer
36.59% :     0.008634s :      1: task_emit
20.12% :     0.004749s :      1: type_inference
 0.25% :     0.000059s :      1: validate


TotalTime = 0.0211202, [33]
[bootstrap]: 0.00037146
[type_inference]: 0.00529123
[event_method]: 1.742e-05
[auto_monad]: 0.00011581
[graph_reusing]: 5.64e-06
[pre_auto_parallel]: 2.51e-06
[py_interpret_to_execute]: 3.636e-05
[rewriter_before_opt_a]: 7.229e-05
[expand_dump_flag]: 3.65e-06
[jit_opt_a]: 0.00599953, [2]
    [Cycle 1]: 0.00196486, [27]
        [switch_simplify]: 5.709e-05
        [loop_unroll]: 2.847e-05
        [a_1]: 0.00060189
        [with_stream_mark]: 1.992e-05
        [recompute_prepare]: 7.98001e-06
        [updatestate_depend_eliminate]: 3.7e-06
        [updatestate_assign_eliminate]: 3.38e-06
        [updatestate_loads_eliminate]: 2.98e-06
        [parameter_eliminate]: 2.41e-06
        [specialize_transform]: 6.98998e-06
        [updatestate_useless_node_eliminater]: 5.37001e-06
        [accelerated_algorithm]: 6.17001e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 5.99e-06
        [merge_forward]: 4.20999e-06
        [cell_reuse_recompute_pass]: 1.21002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.578e-05
        [j_node_and_user_rematch]: 1.091e-05
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 9.31998e-06
        [inline_without_move]: 5.79e-06
        [renormalize]: 0.00089093
        [add_forward_monad_depend]: 6.56e-06
        [auto_monad_grad]: 2.31e-06
        [auto_monad_eliminator]: 1.674e-05
        [cse]: 2.837e-05
        [replace_applicator]: 1.709e-05
    [Cycle 2]: 0.00036223, [27]
        [switch_simplify]: 5.97001e-06
        [loop_unroll]: 5.30999e-06
        [a_1]: 0.000103
        [with_stream_mark]: 1.196e-05
        [recompute_prepare]: 6.03998e-06
        [updatestate_depend_eliminate]: 2.89999e-06
        [updatestate_assign_eliminate]: 2.49999e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 6.06e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 5.31998e-06
        [meta_shard_fg_expand]: 1.70001e-06
        [get_grad_eliminate_]: 5.02999e-06
        [merge_forward]: 3.51001e-06
        [cell_reuse_recompute_pass]: 2.63e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.567e-05
        [j_node_and_user_rematch]: 8.71997e-06
        [meta_fg_expand]: 2.38998e-06
        [replace_old_param]: 8.3e-06
        [inline_without_move]: 5.35001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.19998e-06
        [auto_monad_grad]: 9.80013e-07
        [auto_monad_eliminator]: 6.67002e-06
        [cse]: 1.192e-05
        [replace_applicator]: 6.28e-06
[py_interpret_to_execute_after_opt_a]: 1.381e-05
[rewriter_after_opt_a]: 4.111e-05
[convert_after_rewriter]: 8.11002e-06
[order_py_execute_after_rewriter]: 5.77001e-06
[mutable_eliminate]: 0.000779
[jit_opt_b]: 6.57e-05, [1]
    [Cycle 1]: 5.554e-05, [2]
        [frontend_op_eliminate]: 2.368e-05
        [inline_after_opt_a]: 1.839e-05
[cconv]: 2.906e-05
[loop_unroll]: 0.00047905
[jit_opt_after_cconv]: 0.00015268, [1]
    [Cycle 1]: 0.00014613, [11]
        [c_1]: 2.136e-05
        [parameter_eliminate]: 3.89002e-06
        [updatestate_depend_eliminate]: 7.71001e-06
        [updatestate_assign_eliminate]: 2.74001e-06
        [updatestate_loads_eliminate]: 2.11998e-06
        [cse]: 2.301e-05
        [call_graph_tuple_transform]: 2.165e-05
        [tuple_list_get_item_eliminator]: 5.19e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 9.60019e-07
        [switch_simplify]: 5.14e-06
[remove_dup_value]: 1.556e-05
[partial_unused_args_eliminate]: 2.84001e-06
[environ_conv]: 6.44001e-06
[add_recomputation]: 5.165e-05
[cse_after_recomputation]: 2.109e-05, [1]
    [Cycle 1]: 1.495e-05, [1]
        [cse]: 8.90999e-06
[auto_monad_reorder]: 1.745e-05
[get_jit_bprop_graph]: 2.74001e-06
[rewriter_after_jit_bprop_graph]: 6.967e-05
[opt_after_jit_grad]: 0.00050524
[symbol_engine_optimizer]: 7.775e-05, [1]
    [Cycle 1]: 7.036e-05, [6]
        [build]: 3.78001e-06
        [elim_shapecalc]: 8.3e-06
        [elim_not_effective]: 1.426e-05
        [opt_reshape]: 5.89999e-06
        [fold_const_symbol]: 9.17999e-06
        [renormalize]: 1.00001e-06
[validate]: 4.269e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00658989
[execute]: 9.02e-06
Sums
    bootstrap                            :     0.000371s :  2.21%
    type_inference                       :     0.005291s : 31.53%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000116s :  0.69%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000036s :  0.22%
    rewriter_before_opt_a                :     0.000072s :  0.43%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000705s :  4.20%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000891s :  5.31%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.14%
    jit_opt_a.cse                        :     0.000040s :  0.24%
    jit_opt_a.replace_applicator         :     0.000023s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000041s :  0.25%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000779s :  4.64%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.11%
    cconv                                :     0.000029s :  0.17%
    loop_unroll                          :     0.000479s :  2.86%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000052s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000070s :  0.42%
    opt_after_jit_grad                   :     0.000505s :  3.01%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006590s : 39.27%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000220    23
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.93% :     0.000002s :      2: substitution.fold_const_symbol
 3.18% :     0.000007s :      3: substitution.graph_param_transform
82.87% :     0.000183s :      4: substitution.inline
 1.99% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.25% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.18% :     0.000005s :      2: substitution.replace_old_param
 5.64% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005217     2
82.28% :     0.004293s :      1: type_inference.infer
17.72% :     0.000924s :      1: type_inference.specialize
------[replace.]   0.000069     6
64.69% :     0.000045s :      4: replace.inline
35.31% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     6
94.03% :     0.000180s :      4: match.inline
 5.97% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   691
 1.81% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.41% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 2.12% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000002s :     10: predicate.cast_eliminate
 0.57% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.75% :     0.000002s :     10: predicate.depend_value_elim
 1.23% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000002s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.82% :     0.000008s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.72% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.17% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.72% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.19% :     0.000002s :     10: predicate.minmaximum_grad
 1.98% :     0.000003s :      3: predicate.mutable_eliminate
 0.38% :     0.000001s :      3: predicate.opt_reshape
 2.03% :     0.000003s :     15: predicate.partial_eliminate
 1.45% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.84% :     0.000002s :     10: predicate.reduce_eliminate
 1.34% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.27% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.05% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.56% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.49% :     0.000002s :      6: predicate.specialize_transform
 1.24% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.60% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.71% :     0.000004s :     16: predicate.switch_defer_inline
 2.25% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.97% :     0.000009s :     53: predicate.switch_simplify
 1.41% :     0.000002s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.22% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.84% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.33% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000681    12
48.52% :     0.000330s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.70% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.78% :     0.000312s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023058    76
 0.24% :     0.000055s :      1: add_recomputation
 0.52% :     0.000120s :      1: auto_monad
 0.09% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.67% :     0.000384s :      1: bootstrap
 0.14% :     0.000032s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.04% :     0.006004s :      1: jit_opt_a
 0.67% :     0.000155s :      1: jit_opt_after_cconv
 0.30% :     0.000068s :      1: jit_opt_b
 2.12% :     0.000488s :      1: loop_unroll
 3.43% :     0.000791s :      1: mutable_eliminate
 4.05% :     0.000933s :     26: opt.transform.jit_opt_a
 0.22% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000033s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.23% :     0.000515s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000039s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 2.02% :     0.000465s :      1: renormalize.infer
 1.80% :     0.000416s :      1: renormalize.specialize
 0.32% :     0.000075s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000045s :      1: rewriter_after_opt_a
 0.33% :     0.000075s :      1: rewriter_before_opt_a
 0.35% :     0.000080s :      1: symbol_engine_optimizer
28.66% :     0.006609s :      1: task_emit
23.00% :     0.005303s :      1: type_inference
 0.27% :     0.000063s :      1: validate


TotalTime = 0.016989, [30]
[bootstrap]: 0.00028528
[type_inference]: 0.00773895
[event_method]: 2.608e-05
[auto_monad]: 9.214e-05
[graph_reusing]: 7.14001e-06
[pre_auto_parallel]: 2.76e-06
[py_interpret_to_execute]: 8.122e-05
[rewriter_before_opt_a]: 9.674e-05
[expand_dump_flag]: 3.53e-06
[jit_opt_a]: 0.00645651, [2]
    [Cycle 1]: 0.00226968, [27]
        [switch_simplify]: 0.00014183
        [loop_unroll]: 3.655e-05
        [a_1]: 0.00080012
        [with_stream_mark]: 2.138e-05
        [recompute_prepare]: 7.38999e-06
        [updatestate_depend_eliminate]: 3.88001e-06
        [updatestate_assign_eliminate]: 3.14999e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 5.13002e-06
        [accelerated_algorithm]: 5.74999e-06
        [meta_shard_fg_expand]: 2.20002e-06
        [get_grad_eliminate_]: 5.58997e-06
        [merge_forward]: 3.31001e-06
        [cell_reuse_recompute_pass]: 1.10999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.464e-05
        [j_node_and_user_rematch]: 9.34e-06
        [meta_fg_expand]: 2.91e-06
        [replace_old_param]: 9.20999e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 0.00093322
        [add_forward_monad_depend]: 6.11e-06
        [auto_monad_grad]: 2.23002e-06
        [auto_monad_eliminator]: 1.337e-05
        [cse]: 2.715e-05
        [replace_applicator]: 1.379e-05
    [Cycle 2]: 0.00032142, [27]
        [switch_simplify]: 6.09999e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 9.321e-05
        [with_stream_mark]: 9.69999e-06
        [recompute_prepare]: 5.06997e-06
        [updatestate_depend_eliminate]: 2.66e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 5.10001e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 4.80001e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.78001e-06
        [merge_forward]: 2.42001e-06
        [cell_reuse_recompute_pass]: 2.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.291e-05
        [j_node_and_user_rematch]: 7.94002e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 7.16999e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 9.60019e-07
        [auto_monad_eliminator]: 4.68001e-06
        [cse]: 1.005e-05
        [replace_applicator]: 5.33002e-06
[py_interpret_to_execute_after_opt_a]: 1.147e-05
[rewriter_after_opt_a]: 3.519e-05
[convert_after_rewriter]: 6.78998e-06
[order_py_execute_after_rewriter]: 4.97e-06
[mutable_eliminate]: 0.00056905
[jit_opt_b]: 5.077e-05, [1]
    [Cycle 1]: 4.385e-05, [2]
        [frontend_op_eliminate]: 1.664e-05
        [inline_after_opt_a]: 1.532e-05
[cconv]: 2.568e-05
[loop_unroll]: 0.0004547
[jit_opt_after_cconv]: 0.00014717, [1]
    [Cycle 1]: 0.00014044, [11]
        [c_1]: 2.229e-05
        [parameter_eliminate]: 2.98e-06
        [updatestate_depend_eliminate]: 5.86e-06
        [updatestate_assign_eliminate]: 2.60997e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.954e-05
        [call_graph_tuple_transform]: 1.976e-05
        [tuple_list_get_item_eliminator]: 5.22999e-06
        [none_parameter_eliminate]: 1.53002e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 5.82001e-06
[remove_dup_value]: 1.464e-05
[partial_unused_args_eliminate]: 2.32999e-06
[environ_conv]: 5.42001e-06
[add_recomputation]: 4.565e-05
[cse_after_recomputation]: 2.197e-05, [1]
    [Cycle 1]: 1.588e-05, [1]
        [cse]: 9.87999e-06
[auto_monad_reorder]: 1.515e-05
[get_jit_bprop_graph]: 2.39999e-06
[rewriter_after_jit_bprop_graph]: 5.91e-06
[opt_after_jit_grad]: 0.00048018
[symbol_engine_optimizer]: 7.729e-05, [1]
    [Cycle 1]: 7.098e-05, [6]
        [build]: 3.49001e-06
        [elim_shapecalc]: 8.10999e-06
        [elim_not_effective]: 1.437e-05
        [opt_reshape]: 6.17999e-06
        [fold_const_symbol]: 8.68001e-06
        [renormalize]: 5.19998e-07
[validate]: 3.792e-05
Sums
    bootstrap                            :     0.000285s :  2.28%
    type_inference                       :     0.007739s : 61.83%
    event_method                         :     0.000026s :  0.21%
    auto_monad                           :     0.000092s :  0.74%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000081s :  0.65%
    rewriter_before_opt_a                :     0.000097s :  0.77%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000148s :  1.18%
    jit_opt_a.loop_unroll                :     0.000042s :  0.33%
    jit_opt_a.a_1                        :     0.000893s :  7.14%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.13%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000933s :  7.46%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.14%
    jit_opt_a.cse                        :     0.000037s :  0.30%
    jit_opt_a.replace_applicator         :     0.000019s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.09%
    rewriter_after_opt_a                 :     0.000035s :  0.28%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000569s :  4.55%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000026s :  0.21%
    loop_unroll                          :     0.000455s :  3.63%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000015s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000046s :  0.36%
    cse_after_recomputation.cse          :     0.000010s :  0.08%
    auto_monad_reorder                   :     0.000015s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000480s :  3.84%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.30%


Time group info:
------[substitution.]   0.000353    26
 0.80% :     0.000003s :      2: substitution.elim_not_effective
 0.41% :     0.000001s :      2: substitution.fold_const_symbol
 1.72% :     0.000006s :      3: substitution.graph_param_transform
87.31% :     0.000308s :      6: substitution.inline
 0.82% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.40% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.09% :     0.000004s :      2: substitution.replace_old_param
 2.59% :     0.000009s :      1: substitution.switch_simplify
 3.86% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007648     2
78.61% :     0.006012s :      1: type_inference.infer
21.39% :     0.001636s :      1: type_inference.specialize
------[replace.]   0.000161     9
48.58% :     0.000078s :      6: replace.inline
26.66% :     0.000043s :      1: replace.switch_simplify
24.75% :     0.000040s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000325     9
93.68% :     0.000304s :      6: match.inline
 2.50% :     0.000008s :      1: match.switch_simplify
 3.82% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   801
 1.28% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.19% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     12: predicate.addn_check_dump
 1.49% :     0.000002s :     12: predicate.addn_zero_filter
 2.10% :     0.000003s :     12: predicate.arithmetic_simplify
 1.19% :     0.000002s :     12: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.29% :     0.000002s :     12: predicate.compare_switch_simplify
 1.35% :     0.000002s :     12: predicate.depend_value_elim
 1.20% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.32% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.70% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000001s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.11% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.36% :     0.000002s :     12: predicate.environ_get_eliminate
 1.10% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.71% :     0.000008s :     26: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     17: predicate.load_eliminater
 1.28% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.29% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.31% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     12: predicate.merge_addn
 1.17% :     0.000002s :     12: predicate.minmaximum_grad
 1.42% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     17: predicate.partial_eliminate
 1.23% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     12: predicate.reduce_eliminate
 1.41% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.84% :     0.000003s :     20: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     12: predicate.reshape_eliminate
 1.23% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.59% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.63% :     0.000004s :     20: predicate.switch_defer_inline
 2.24% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.88% :     0.000012s :     65: predicate.switch_simplify
 1.15% :     0.000002s :     12: predicate.tile_eliminate
 1.20% :     0.000002s :     12: predicate.transpose_eliminate
 1.40% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     12: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.51% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001063    15
44.82% :     0.000476s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.41% :     0.000132s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.77% :     0.000455s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019174    72
 0.25% :     0.000049s :      1: add_recomputation
 0.50% :     0.000096s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 1.57% :     0.000300s :      1: bootstrap
 0.15% :     0.000028s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.13% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.16% :     0.000031s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000010s :      1: graph_reusing
33.69% :     0.006460s :      1: jit_opt_a
 0.78% :     0.000150s :      1: jit_opt_after_cconv
 0.28% :     0.000053s :      1: jit_opt_b
 2.41% :     0.000462s :      1: loop_unroll
 3.01% :     0.000577s :      1: mutable_eliminate
 6.26% :     0.001199s :     26: opt.transform.jit_opt_a
 0.26% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.55% :     0.000489s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.44% :     0.000084s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000017s :      1: remove_dup_value
 2.55% :     0.000489s :      1: renormalize.infer
 2.27% :     0.000435s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000038s :      1: rewriter_after_opt_a
 0.52% :     0.000100s :      1: rewriter_before_opt_a
 0.42% :     0.000080s :      1: symbol_engine_optimizer
40.47% :     0.007760s :      1: type_inference


TotalTime = 0.0176472, [33]
[bootstrap]: 0.00025978
[type_inference]: 0.00407381
[event_method]: 1.472e-05
[auto_monad]: 4.035e-05
[graph_reusing]: 3.31001e-06
[pre_auto_parallel]: 1.66e-06
[py_interpret_to_execute]: 2.679e-05
[rewriter_before_opt_a]: 6.267e-05
[expand_dump_flag]: 2.03002e-06
[jit_opt_a]: 0.00486364, [2]
    [Cycle 1]: 0.00146106, [27]
        [switch_simplify]: 5.062e-05
        [loop_unroll]: 2.906e-05
        [a_1]: 0.00051675
        [with_stream_mark]: 1.425e-05
        [recompute_prepare]: 7.11999e-06
        [updatestate_depend_eliminate]: 3.21001e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 5.67999e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.60001e-06
        [meta_shard_fg_expand]: 1.15999e-06
        [get_grad_eliminate_]: 5.17999e-06
        [merge_forward]: 2.47001e-06
        [cell_reuse_recompute_pass]: 1.02998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.116e-05
        [j_node_and_user_rematch]: 9.22001e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 8.40999e-06
        [inline_without_move]: 5.40001e-06
        [renormalize]: 0.00056364
        [add_forward_monad_depend]: 4.45999e-06
        [auto_monad_grad]: 1.29e-06
        [auto_monad_eliminator]: 9.74e-06
        [cse]: 1.625e-05
        [replace_applicator]: 1.108e-05
    [Cycle 2]: 0.00031257, [27]
        [switch_simplify]: 5.77001e-06
        [loop_unroll]: 5.21002e-06
        [a_1]: 8.865e-05
        [with_stream_mark]: 8.82e-06
        [recompute_prepare]: 4.87998e-06
        [updatestate_depend_eliminate]: 2.37001e-06
        [updatestate_assign_eliminate]: 1.99999e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 4.78001e-06
        [updatestate_useless_node_eliminater]: 4.70999e-06
        [accelerated_algorithm]: 4.79e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.53001e-06
        [merge_forward]: 2.31e-06
        [cell_reuse_recompute_pass]: 2.35002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.215e-05
        [j_node_and_user_rematch]: 7.8e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 6.59001e-06
        [inline_without_move]: 4.53001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 4.87e-06
        [cse]: 9.32001e-06
        [replace_applicator]: 4.78001e-06
[py_interpret_to_execute_after_opt_a]: 8.37e-06
[rewriter_after_opt_a]: 2.655e-05
[convert_after_rewriter]: 6.11e-06
[order_py_execute_after_rewriter]: 4.07998e-06
[mutable_eliminate]: 0.00058855
[jit_opt_b]: 4.76e-05, [1]
    [Cycle 1]: 4.082e-05, [2]
        [frontend_op_eliminate]: 1.567e-05
        [inline_after_opt_a]: 1.392e-05
[cconv]: 1.676e-05
[loop_unroll]: 0.000444
[jit_opt_after_cconv]: 0.00013877, [1]
    [Cycle 1]: 0.00013301, [11]
        [c_1]: 2.049e-05
        [parameter_eliminate]: 3.41999e-06
        [updatestate_depend_eliminate]: 5.57001e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [cse]: 1.943e-05
        [call_graph_tuple_transform]: 1.874e-05
        [tuple_list_get_item_eliminator]: 5.13002e-06
        [none_parameter_eliminate]: 1.47999e-06
        [renormalize]: 5.50004e-07
        [switch_simplify]: 5.12999e-06
[remove_dup_value]: 1.394e-05
[partial_unused_args_eliminate]: 2.40002e-06
[environ_conv]: 5.99e-06
[add_recomputation]: 4.93e-05
[cse_after_recomputation]: 2.011e-05, [1]
    [Cycle 1]: 1.478e-05, [1]
        [cse]: 8.80001e-06
[auto_monad_reorder]: 1.493e-05
[get_jit_bprop_graph]: 1.99999e-06
[rewriter_after_jit_bprop_graph]: 6.81001e-06
[opt_after_jit_grad]: 0.00050631
[symbol_engine_optimizer]: 7.526e-05, [1]
    [Cycle 1]: 6.75e-05, [6]
        [build]: 4.62998e-06
        [elim_shapecalc]: 7.26001e-06
        [elim_not_effective]: 1.375e-05
        [opt_reshape]: 5.82001e-06
        [fold_const_symbol]: 8.25e-06
        [renormalize]: 5.90022e-07
[validate]: 3.96e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.00608045
[execute]: 9.09e-06
Sums
    bootstrap                            :     0.000260s :  1.86%
    type_inference                       :     0.004074s : 29.17%
    event_method                         :     0.000015s :  0.11%
    auto_monad                           :     0.000040s :  0.29%
    graph_reusing                        :     0.000003s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000027s :  0.19%
    rewriter_before_opt_a                :     0.000063s :  0.45%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000056s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000605s :  4.33%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000564s :  4.04%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.10%
    jit_opt_a.cse                        :     0.000026s :  0.18%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000027s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000589s :  4.21%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000017s :  0.12%
    loop_unroll                          :     0.000444s :  3.18%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000049s :  0.35%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.05%
    opt_after_jit_grad                   :     0.000506s :  3.62%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.28%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006080s : 43.53%
    execute                              :     0.000009s :  0.07%


Time group info:
------[substitution.]   0.000199    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.63% :     0.000005s :      3: substitution.graph_param_transform
86.00% :     0.000171s :      4: substitution.inline
 1.69% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.95% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.71% :     0.000003s :      2: substitution.replace_old_param
 4.40% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004023     2
81.00% :     0.003259s :      1: type_inference.infer
19.00% :     0.000764s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.08% :     0.000041s :      4: replace.inline
32.92% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000144     6
94.77% :     0.000136s :      4: match.inline
 5.23% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.28% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.59% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 2.00% :     0.000002s :     10: predicate.arithmetic_simplify
 1.38% :     0.000002s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.49% :     0.000002s :     10: predicate.depend_value_elim
 1.40% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.55% :     0.000002s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.09% :     0.000006s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.59% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 2.02% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000000s :      3: predicate.opt_reshape
 2.04% :     0.000002s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.05% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.61% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.65% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.13% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.77% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.88% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000500    12
48.27% :     0.000241s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.09% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.65% :     0.000228s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019128    76
 0.27% :     0.000052s :      1: add_recomputation
 0.23% :     0.000043s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.41% :     0.000270s :      1: bootstrap
 0.10% :     0.000019s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.09% :     0.000018s :      1: event_method
 0.08% :     0.000015s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
25.44% :     0.004866s :      1: jit_opt_a
 0.74% :     0.000141s :      1: jit_opt_after_cconv
 0.26% :     0.000050s :      1: jit_opt_b
 2.36% :     0.000452s :      1: loop_unroll
 3.12% :     0.000597s :      1: mutable_eliminate
 4.21% :     0.000805s :     26: opt.transform.jit_opt_a
 0.24% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.70% :     0.000517s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000029s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 1.57% :     0.000301s :      1: renormalize.infer
 1.35% :     0.000258s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000031s :      1: rewriter_after_opt_a
 0.34% :     0.000065s :      1: rewriter_before_opt_a
 0.41% :     0.000078s :      1: symbol_engine_optimizer
31.88% :     0.006098s :      1: task_emit
21.35% :     0.004084s :      1: type_inference
 0.32% :     0.000061s :      1: validate


TotalTime = 0.0209539, [33]
[bootstrap]: 0.00032718
[type_inference]: 0.00539273
[event_method]: 1.812e-05
[auto_monad]: 6.71e-05
[graph_reusing]: 6.40997e-06
[pre_auto_parallel]: 2.52001e-06
[py_interpret_to_execute]: 3.907e-05
[rewriter_before_opt_a]: 7.357e-05
[expand_dump_flag]: 3.02002e-06
[jit_opt_a]: 0.00612927, [2]
    [Cycle 1]: 0.0018241, [27]
        [switch_simplify]: 6.815e-05
        [loop_unroll]: 3.025e-05
        [a_1]: 0.00056764
        [with_stream_mark]: 2.012e-05
        [recompute_prepare]: 7.4e-06
        [updatestate_depend_eliminate]: 3.66001e-06
        [updatestate_assign_eliminate]: 2.85998e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [parameter_eliminate]: 2.02999e-06
        [specialize_transform]: 5.82001e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 5.99e-06
        [meta_shard_fg_expand]: 2.39001e-06
        [get_grad_eliminate_]: 5.29998e-06
        [merge_forward]: 3.61001e-06
        [cell_reuse_recompute_pass]: 1.50999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.439e-05
        [j_node_and_user_rematch]: 9.33002e-06
        [meta_fg_expand]: 2.29001e-06
        [replace_old_param]: 9.72001e-06
        [inline_without_move]: 5.89e-06
        [renormalize]: 0.00080346
        [add_forward_monad_depend]: 5.74e-06
        [auto_monad_grad]: 2.19999e-06
        [auto_monad_eliminator]: 1.415e-05
        [cse]: 2.912e-05
        [replace_applicator]: 1.369e-05
    [Cycle 2]: 0.00032007, [27]
        [switch_simplify]: 5.81e-06
        [loop_unroll]: 4.79e-06
        [a_1]: 8.934e-05
        [with_stream_mark]: 9.22999e-06
        [recompute_prepare]: 4.95999e-06
        [updatestate_depend_eliminate]: 2.64999e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 4.88001e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 4.92e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 4.39002e-06
        [merge_forward]: 2.63e-06
        [cell_reuse_recompute_pass]: 2.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.311e-05
        [j_node_and_user_rematch]: 8.39998e-06
        [meta_fg_expand]: 1.45001e-06
        [replace_old_param]: 7.85998e-06
        [inline_without_move]: 4.65001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.41002e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 5.02e-06
        [cse]: 9.66e-06
        [replace_applicator]: 5.17999e-06
[py_interpret_to_execute_after_opt_a]: 1.136e-05
[rewriter_after_opt_a]: 3.895e-05
[convert_after_rewriter]: 7.33e-06
[order_py_execute_after_rewriter]: 5.37001e-06
[mutable_eliminate]: 0.0006764
[jit_opt_b]: 5.469e-05, [1]
    [Cycle 1]: 4.641e-05, [2]
        [frontend_op_eliminate]: 2.009e-05
        [inline_after_opt_a]: 1.501e-05
[cconv]: 2.692e-05
[loop_unroll]: 0.0005711
[jit_opt_after_cconv]: 0.00014232, [1]
    [Cycle 1]: 0.0001353, [11]
        [c_1]: 2.073e-05
        [parameter_eliminate]: 2.44999e-06
        [updatestate_depend_eliminate]: 7.05998e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [cse]: 1.979e-05
        [call_graph_tuple_transform]: 1.933e-05
        [tuple_list_get_item_eliminator]: 5.79999e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 8.40024e-07
        [switch_simplify]: 5.27999e-06
[remove_dup_value]: 1.502e-05
[partial_unused_args_eliminate]: 2.55002e-06
[environ_conv]: 5.54e-06
[add_recomputation]: 5.166e-05
[cse_after_recomputation]: 2.002e-05, [1]
    [Cycle 1]: 1.38e-05, [1]
        [cse]: 7.99002e-06
[auto_monad_reorder]: 1.517e-05
[get_jit_bprop_graph]: 2.09999e-06
[rewriter_after_jit_bprop_graph]: 6.53998e-06
[opt_after_jit_grad]: 0.00049472
[symbol_engine_optimizer]: 7.723e-05, [1]
    [Cycle 1]: 6.968e-05, [6]
        [build]: 3.66999e-06
        [elim_shapecalc]: 7.92e-06
        [elim_not_effective]: 1.413e-05
        [opt_reshape]: 5.96e-06
        [fold_const_symbol]: 9.19e-06
        [renormalize]: 9.09989e-07
[validate]: 4.144e-05
[backend_pass]: 9.60019e-07
[task_emit]: 0.00639721
[execute]: 9.20999e-06
Sums
    bootstrap                            :     0.000327s :  2.00%
    type_inference                       :     0.005393s : 33.02%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000067s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.24%
    rewriter_before_opt_a                :     0.000074s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.45%
    jit_opt_a.loop_unroll                :     0.000035s :  0.21%
    jit_opt_a.a_1                        :     0.000657s :  4.02%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000804s :  4.92%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000039s :  0.24%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000039s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000676s :  4.14%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000027s :  0.16%
    loop_unroll                          :     0.000571s :  3.50%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000052s :  0.32%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000495s :  3.03%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006397s : 39.17%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000229    23
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.65% :     0.000006s :      3: substitution.graph_param_transform
83.79% :     0.000192s :      4: substitution.inline
 1.68% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.04% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000005s :      2: substitution.replace_old_param
 6.26% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005311     2
82.40% :     0.004376s :      1: type_inference.infer
17.60% :     0.000935s :      1: type_inference.specialize
------[replace.]   0.000074     6
66.89% :     0.000050s :      4: replace.inline
33.11% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000203     6
93.49% :     0.000189s :      4: match.inline
 6.51% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000001s :     10: predicate.addn_zero_filter
 2.05% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.52% :     0.000002s :     10: predicate.depend_value_elim
 1.22% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.60% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.81% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.32% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.88% :     0.000007s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.86% :     0.000002s :     15: predicate.load_eliminater
 1.62% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.46% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.83% :     0.000002s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.71% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.57% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.63% :     0.000009s :     53: predicate.switch_simplify
 1.34% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.11% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.69% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000661    12
50.76% :     0.000336s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.06% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.18% :     0.000292s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022743    76
 0.24% :     0.000054s :      1: add_recomputation
 0.31% :     0.000071s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.52% :     0.000346s :      1: bootstrap
 0.13% :     0.000030s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.97% :     0.006133s :      1: jit_opt_a
 0.64% :     0.000145s :      1: jit_opt_after_cconv
 0.25% :     0.000057s :      1: jit_opt_b
 2.56% :     0.000581s :      1: loop_unroll
 3.02% :     0.000686s :      1: mutable_eliminate
 3.88% :     0.000882s :     26: opt.transform.jit_opt_a
 0.21% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000029s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000504s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000042s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.85% :     0.000420s :      1: renormalize.infer
 1.65% :     0.000375s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000042s :      1: rewriter_after_opt_a
 0.34% :     0.000077s :      1: rewriter_before_opt_a
 0.35% :     0.000080s :      1: symbol_engine_optimizer
28.21% :     0.006416s :      1: task_emit
23.78% :     0.005409s :      1: type_inference
 0.27% :     0.000061s :      1: validate


TotalTime = 0.018015, [33]
[bootstrap]: 0.00027889
[type_inference]: 0.00468899
[event_method]: 1.716e-05
[auto_monad]: 6.111e-05
[graph_reusing]: 5.87999e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 2.812e-05
[rewriter_before_opt_a]: 6.925e-05
[expand_dump_flag]: 3.09999e-06
[jit_opt_a]: 0.00491652, [2]
    [Cycle 1]: 0.00142083, [27]
        [switch_simplify]: 5.055e-05
        [loop_unroll]: 2.842e-05
        [a_1]: 0.00049725
        [with_stream_mark]: 1.637e-05
        [recompute_prepare]: 6.79001e-06
        [updatestate_depend_eliminate]: 3.33998e-06
        [updatestate_assign_eliminate]: 2.64999e-06
        [updatestate_loads_eliminate]: 2.59001e-06
        [parameter_eliminate]: 1.79998e-06
        [specialize_transform]: 5.83002e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 2.14e-06
        [get_grad_eliminate_]: 5.05999e-06
        [merge_forward]: 3.40003e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.376e-05
        [j_node_and_user_rematch]: 8.57e-06
        [meta_fg_expand]: 2.27999e-06
        [replace_old_param]: 7.69002e-06
        [inline_without_move]: 4.94e-06
        [renormalize]: 0.00053046
        [add_forward_monad_depend]: 5.05999e-06
        [auto_monad_grad]: 2.18998e-06
        [auto_monad_eliminator]: 1.258e-05
        [cse]: 2.525e-05
        [replace_applicator]: 1.116e-05
    [Cycle 2]: 0.00030804, [27]
        [switch_simplify]: 5.17999e-06
        [loop_unroll]: 4.79e-06
        [a_1]: 8.675e-05
        [with_stream_mark]: 8.62e-06
        [recompute_prepare]: 5.02999e-06
        [updatestate_depend_eliminate]: 2.36e-06
        [updatestate_assign_eliminate]: 1.99999e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 4.72e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 5.14998e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.18001e-06
        [merge_forward]: 2.19001e-06
        [cell_reuse_recompute_pass]: 1.36998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.188e-05
        [j_node_and_user_rematch]: 8.02998e-06
        [meta_fg_expand]: 1.52999e-06
        [replace_old_param]: 6.91999e-06
        [inline_without_move]: 4.51002e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 5.14998e-06
        [cse]: 9.28002e-06
        [replace_applicator]: 5.30999e-06
[py_interpret_to_execute_after_opt_a]: 8.88002e-06
[rewriter_after_opt_a]: 3.195e-05
[convert_after_rewriter]: 7.28e-06
[order_py_execute_after_rewriter]: 4.70999e-06
[mutable_eliminate]: 0.00051217
[jit_opt_b]: 4.657e-05, [1]
    [Cycle 1]: 4.02e-05, [2]
        [frontend_op_eliminate]: 1.522e-05
        [inline_after_opt_a]: 1.395e-05
[cconv]: 2.47e-05
[loop_unroll]: 0.00040164
[jit_opt_after_cconv]: 0.00013543, [1]
    [Cycle 1]: 0.00012905, [11]
        [c_1]: 1.99e-05
        [parameter_eliminate]: 2.76e-06
        [updatestate_depend_eliminate]: 5.55001e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [cse]: 1.627e-05
        [call_graph_tuple_transform]: 1.933e-05
        [tuple_list_get_item_eliminator]: 5.10001e-06
        [none_parameter_eliminate]: 1.34e-06
        [renormalize]: 6.89994e-07
        [switch_simplify]: 5.36002e-06
[remove_dup_value]: 1.318e-05
[partial_unused_args_eliminate]: 2.71e-06
[environ_conv]: 5.04998e-06
[add_recomputation]: 4.635e-05
[cse_after_recomputation]: 1.935e-05, [1]
    [Cycle 1]: 1.415e-05, [1]
        [cse]: 8.25e-06
[auto_monad_reorder]: 1.389e-05
[get_jit_bprop_graph]: 1.65001e-06
[rewriter_after_jit_bprop_graph]: 5.46e-06
[opt_after_jit_grad]: 0.00043664
[symbol_engine_optimizer]: 7.096e-05, [1]
    [Cycle 1]: 6.447e-05, [6]
        [build]: 3.95e-06
        [elim_shapecalc]: 8e-06
        [elim_not_effective]: 1.157e-05
        [opt_reshape]: 5.37999e-06
        [fold_const_symbol]: 8.22e-06
        [renormalize]: 5.89993e-07
[validate]: 3.246e-05
[backend_pass]: 8.29983e-07
[task_emit]: 0.0059277
[execute]: 6.65998e-06
Sums
    bootstrap                            :     0.000279s :  1.96%
    type_inference                       :     0.004689s : 32.90%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000061s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.20%
    rewriter_before_opt_a                :     0.000069s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.39%
    jit_opt_a.loop_unroll                :     0.000033s :  0.23%
    jit_opt_a.a_1                        :     0.000584s :  4.10%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000531s :  3.72%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000035s :  0.24%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000032s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000512s :  3.59%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000402s :  2.82%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000046s :  0.33%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000437s :  3.06%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005928s : 41.59%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000187    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.86% :     0.000002s :      2: substitution.fold_const_symbol
 2.80% :     0.000005s :      3: substitution.graph_param_transform
83.66% :     0.000156s :      4: substitution.inline
 1.45% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.99% :     0.000004s :      2: substitution.replace_old_param
 6.02% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004627     2
83.87% :     0.003881s :      1: type_inference.infer
16.13% :     0.000746s :      1: type_inference.specialize
------[replace.]   0.000061     6
68.14% :     0.000042s :      4: replace.inline
31.86% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000164     6
93.76% :     0.000154s :      4: match.inline
 6.24% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.39% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.42% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 2.36% :     0.000003s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000001s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.48% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.67% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.09% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.65% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000001s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.55% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.12% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.69% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.34% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000545    12
53.25% :     0.000290s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.65% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.09% :     0.000224s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019427    76
 0.25% :     0.000049s :      1: add_recomputation
 0.33% :     0.000064s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.53% :     0.000298s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.32% :     0.004919s :      1: jit_opt_a
 0.71% :     0.000138s :      1: jit_opt_after_cconv
 0.25% :     0.000049s :      1: jit_opt_b
 2.11% :     0.000409s :      1: loop_unroll
 2.68% :     0.000521s :      1: mutable_eliminate
 4.02% :     0.000781s :     26: opt.transform.jit_opt_a
 0.24% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.29% :     0.000444s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000031s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.44% :     0.000280s :      1: renormalize.infer
 1.25% :     0.000242s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000035s :      1: rewriter_after_opt_a
 0.37% :     0.000072s :      1: rewriter_before_opt_a
 0.38% :     0.000073s :      1: symbol_engine_optimizer
30.56% :     0.005937s :      1: task_emit
24.21% :     0.004703s :      1: type_inference
 0.26% :     0.000051s :      1: validate


TotalTime = 0.0184692, [33]
[bootstrap]: 0.00028879
[type_inference]: 0.00450531
[event_method]: 1.735e-05
[auto_monad]: 5.557e-05
[graph_reusing]: 5.09e-06
[pre_auto_parallel]: 1.87999e-06
[py_interpret_to_execute]: 3.685e-05
[rewriter_before_opt_a]: 6.989e-05
[expand_dump_flag]: 2.53e-06
[jit_opt_a]: 0.00533377, [2]
    [Cycle 1]: 0.00169268, [27]
        [switch_simplify]: 4.996e-05
        [loop_unroll]: 3.07e-05
        [a_1]: 0.00054386
        [with_stream_mark]: 1.675e-05
        [recompute_prepare]: 8.42e-06
        [updatestate_depend_eliminate]: 3.58e-06
        [updatestate_assign_eliminate]: 2.61999e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [parameter_eliminate]: 1.19998e-06
        [specialize_transform]: 7.35e-06
        [updatestate_useless_node_eliminater]: 6.02001e-06
        [accelerated_algorithm]: 7.11001e-06
        [meta_shard_fg_expand]: 1.88002e-06
        [get_grad_eliminate_]: 5.92001e-06
        [merge_forward]: 2.54999e-06
        [cell_reuse_recompute_pass]: 1.34998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.234e-05
        [j_node_and_user_rematch]: 1.029e-05
        [meta_fg_expand]: 2.21998e-06
        [replace_old_param]: 9.27001e-06
        [inline_without_move]: 6.54001e-06
        [renormalize]: 0.0007291
        [add_forward_monad_depend]: 5.89999e-06
        [auto_monad_grad]: 1.81e-06
        [auto_monad_eliminator]: 1.216e-05
        [cse]: 2.192e-05
        [replace_applicator]: 1.264e-05
    [Cycle 2]: 0.00033749, [27]
        [switch_simplify]: 6.48e-06
        [loop_unroll]: 6.12999e-06
        [a_1]: 9.889e-05
        [with_stream_mark]: 1.025e-05
        [recompute_prepare]: 5.14998e-06
        [updatestate_depend_eliminate]: 2.41e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 5.04e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.15001e-06
        [meta_shard_fg_expand]: 1.45001e-06
        [get_grad_eliminate_]: 4.78001e-06
        [merge_forward]: 2.54999e-06
        [cell_reuse_recompute_pass]: 1.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.255e-05
        [j_node_and_user_rematch]: 8.31002e-06
        [meta_fg_expand]: 2.03002e-06
        [replace_old_param]: 1.38e-05
        [inline_without_move]: 4.90999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 4.99e-06
        [cse]: 1.101e-05
        [replace_applicator]: 5.00999e-06
[py_interpret_to_execute_after_opt_a]: 9.39e-06
[rewriter_after_opt_a]: 3.041e-05
[convert_after_rewriter]: 6.17999e-06
[order_py_execute_after_rewriter]: 4.58999e-06
[mutable_eliminate]: 0.00056868
[jit_opt_b]: 4.961e-05, [1]
    [Cycle 1]: 4.307e-05, [2]
        [frontend_op_eliminate]: 1.636e-05
        [inline_after_opt_a]: 1.508e-05
[cconv]: 2.097e-05
[loop_unroll]: 0.00042285
[jit_opt_after_cconv]: 0.0001368, [1]
    [Cycle 1]: 0.00013072, [11]
        [c_1]: 2.099e-05
        [parameter_eliminate]: 2.39999e-06
        [updatestate_depend_eliminate]: 5.29e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.717e-05
        [call_graph_tuple_transform]: 1.82e-05
        [tuple_list_get_item_eliminator]: 5.54e-06
        [none_parameter_eliminate]: 1.19e-06
        [renormalize]: 3.9002e-07
        [switch_simplify]: 5.66e-06
[remove_dup_value]: 9.49e-06
[partial_unused_args_eliminate]: 2.16998e-06
[environ_conv]: 4.03001e-06
[add_recomputation]: 3.811e-05
[cse_after_recomputation]: 2.038e-05, [1]
    [Cycle 1]: 1.525e-05, [1]
        [cse]: 9.32001e-06
[auto_monad_reorder]: 1.193e-05
[get_jit_bprop_graph]: 1.55999e-06
[rewriter_after_jit_bprop_graph]: 4.70999e-06
[opt_after_jit_grad]: 0.00046729
[symbol_engine_optimizer]: 7.521e-05, [1]
    [Cycle 1]: 6.876e-05, [6]
        [build]: 3.8e-06
        [elim_shapecalc]: 7.24001e-06
        [elim_not_effective]: 1.312e-05
        [opt_reshape]: 5.79e-06
        [fold_const_symbol]: 9.80002e-06
        [renormalize]: 5.19998e-07
[validate]: 3.382e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00602078
[execute]: 9.87999e-06
Sums
    bootstrap                            :     0.000289s :  1.98%
    type_inference                       :     0.004505s : 30.96%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000056s :  0.38%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000037s :  0.25%
    rewriter_before_opt_a                :     0.000070s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.39%
    jit_opt_a.loop_unroll                :     0.000037s :  0.25%
    jit_opt_a.a_1                        :     0.000643s :  4.42%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000023s :  0.16%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.000729s :  5.01%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.12%
    jit_opt_a.cse                        :     0.000033s :  0.23%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000030s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000569s :  3.91%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000021s :  0.14%
    loop_unroll                          :     0.000423s :  2.91%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000009s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000038s :  0.26%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000467s :  3.21%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006021s : 41.37%
    execute                              :     0.000010s :  0.07%


Time group info:
------[substitution.]   0.000193    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 1.05% :     0.000002s :      2: substitution.fold_const_symbol
 2.50% :     0.000005s :      3: substitution.graph_param_transform
81.01% :     0.000156s :      4: substitution.inline
 2.09% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.13% :     0.000004s :      4: substitution.remove_not_recompute_node
 4.97% :     0.000010s :      2: substitution.replace_old_param
 5.16% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004443     2
80.17% :     0.003562s :      1: type_inference.infer
19.83% :     0.000881s :      1: type_inference.specialize
------[replace.]   0.000072     6
65.03% :     0.000047s :      4: replace.inline
34.97% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000163     6
94.54% :     0.000154s :      4: match.inline
 5.46% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.30% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.15% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.96% :     0.000004s :     10: predicate.arithmetic_simplify
 1.42% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.26% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.24% :     0.000007s :     22: predicate.inline
 1.02% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.34% :     0.000002s :      6: predicate.less_batch_normalization
 1.66% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.55% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 2.50% :     0.000003s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.59% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.65% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.92% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.22% :     0.000002s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.12% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.42% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.61% :     0.000010s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.31% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000604    12
45.55% :     0.000275s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.87% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.59% :     0.000299s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020166    76
 0.20% :     0.000041s :      1: add_recomputation
 0.29% :     0.000059s :      1: auto_monad
 0.07% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.51% :     0.000304s :      1: bootstrap
 0.12% :     0.000024s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.08% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.46% :     0.005337s :      1: jit_opt_a
 0.69% :     0.000140s :      1: jit_opt_after_cconv
 0.26% :     0.000052s :      1: jit_opt_b
 2.14% :     0.000431s :      1: loop_unroll
 2.86% :     0.000577s :      1: mutable_eliminate
 4.29% :     0.000866s :     26: opt.transform.jit_opt_a
 0.23% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.36% :     0.000475s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000040s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000012s :      1: remove_dup_value
 1.83% :     0.000369s :      1: renormalize.infer
 1.75% :     0.000353s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000033s :      1: rewriter_after_opt_a
 0.36% :     0.000073s :      1: rewriter_before_opt_a
 0.39% :     0.000078s :      1: symbol_engine_optimizer
29.94% :     0.006037s :      1: task_emit
22.40% :     0.004517s :      1: type_inference
 0.26% :     0.000053s :      1: validate


TotalTime = 0.0206918, [33]
[bootstrap]: 0.0003228
[type_inference]: 0.00507947
[event_method]: 1.774e-05
[auto_monad]: 6.688e-05
[graph_reusing]: 5.71e-06
[pre_auto_parallel]: 2.52001e-06
[py_interpret_to_execute]: 3.301e-05
[rewriter_before_opt_a]: 7.052e-05
[expand_dump_flag]: 3.11001e-06
[jit_opt_a]: 0.00603609, [2]
    [Cycle 1]: 0.00192933, [27]
        [switch_simplify]: 5.278e-05
        [loop_unroll]: 2.941e-05
        [a_1]: 0.00056673
        [with_stream_mark]: 2.163e-05
        [recompute_prepare]: 8.03999e-06
        [updatestate_depend_eliminate]: 3.63e-06
        [updatestate_assign_eliminate]: 2.47001e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [parameter_eliminate]: 1.87001e-06
        [specialize_transform]: 6.54999e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 5.67999e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 3.58e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.443e-05
        [j_node_and_user_rematch]: 9.79999e-06
        [meta_fg_expand]: 2.21e-06
        [replace_old_param]: 8.85999e-06
        [inline_without_move]: 5.69e-06
        [renormalize]: 0.00093334
        [add_forward_monad_depend]: 6.69999e-06
        [auto_monad_grad]: 2.36e-06
        [auto_monad_eliminator]: 1.595e-05
        [cse]: 2.789e-05
        [replace_applicator]: 1.466e-05
    [Cycle 2]: 0.00034312, [27]
        [switch_simplify]: 5.49998e-06
        [loop_unroll]: 5.17999e-06
        [a_1]: 9.205e-05
        [with_stream_mark]: 1.064e-05
        [recompute_prepare]: 5.12e-06
        [updatestate_depend_eliminate]: 3.43e-06
        [updatestate_assign_eliminate]: 2.18998e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.32e-06
        [specialize_transform]: 4.80999e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 5.40001e-06
        [meta_shard_fg_expand]: 1.55999e-06
        [get_grad_eliminate_]: 4.38999e-06
        [merge_forward]: 3.06001e-06
        [cell_reuse_recompute_pass]: 2.28998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.399e-05
        [j_node_and_user_rematch]: 8.08001e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 7.8e-06
        [inline_without_move]: 4.60999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 1.00999e-06
        [auto_monad_eliminator]: 5.86e-06
        [cse]: 1.238e-05
        [replace_applicator]: 5.34998e-06
[py_interpret_to_execute_after_opt_a]: 1.297e-05
[rewriter_after_opt_a]: 3.891e-05
[convert_after_rewriter]: 7.93001e-06
[order_py_execute_after_rewriter]: 5.18002e-06
[mutable_eliminate]: 0.00073598
[jit_opt_b]: 5.926e-05, [1]
    [Cycle 1]: 5.028e-05, [2]
        [frontend_op_eliminate]: 2.164e-05
        [inline_after_opt_a]: 1.606e-05
[cconv]: 2.745e-05
[loop_unroll]: 0.00051488
[jit_opt_after_cconv]: 0.00014942, [1]
    [Cycle 1]: 0.0001424, [11]
        [c_1]: 2.197e-05
        [parameter_eliminate]: 2.99999e-06
        [updatestate_depend_eliminate]: 6.07999e-06
        [updatestate_assign_eliminate]: 2.45002e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 2.194e-05
        [call_graph_tuple_transform]: 1.941e-05
        [tuple_list_get_item_eliminator]: 5.17e-06
        [none_parameter_eliminate]: 2.02999e-06
        [renormalize]: 7.39994e-07
        [switch_simplify]: 5.63002e-06
[remove_dup_value]: 1.516e-05
[partial_unused_args_eliminate]: 2.71999e-06
[environ_conv]: 6.26e-06
[add_recomputation]: 5.317e-05
[cse_after_recomputation]: 2.239e-05, [1]
    [Cycle 1]: 1.585e-05, [1]
        [cse]: 9.74999e-06
[auto_monad_reorder]: 1.571e-05
[get_jit_bprop_graph]: 2.01998e-06
[rewriter_after_jit_bprop_graph]: 6.44999e-06
[opt_after_jit_grad]: 0.00049186
[symbol_engine_optimizer]: 7.883e-05, [1]
    [Cycle 1]: 7.056e-05, [6]
        [build]: 4.11001e-06
        [elim_shapecalc]: 8.63001e-06
        [elim_not_effective]: 1.331e-05
        [opt_reshape]: 5.84999e-06
        [fold_const_symbol]: 9.44e-06
        [renormalize]: 1.12e-06
[validate]: 4.145e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.0065336
[execute]: 9.07999e-06
Sums
    bootstrap                            :     0.000323s :  1.98%
    type_inference                       :     0.005079s : 31.20%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000067s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.20%
    rewriter_before_opt_a                :     0.000071s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.36%
    jit_opt_a.loop_unroll                :     0.000035s :  0.21%
    jit_opt_a.a_1                        :     0.000659s :  4.05%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000933s :  5.73%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.13%
    jit_opt_a.cse                        :     0.000040s :  0.25%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000039s :  0.24%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000736s :  4.52%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000027s :  0.17%
    loop_unroll                          :     0.000515s :  3.16%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000053s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000492s :  3.02%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006534s : 40.13%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000232    23
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.95% :     0.000002s :      2: substitution.fold_const_symbol
 2.50% :     0.000006s :      3: substitution.graph_param_transform
84.80% :     0.000197s :      4: substitution.inline
 1.67% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.84% :     0.000004s :      2: substitution.replace_old_param
 5.46% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005012     2
80.96% :     0.004058s :      1: type_inference.infer
19.04% :     0.000955s :      1: type_inference.specialize
------[replace.]   0.000074     6
68.03% :     0.000050s :      4: replace.inline
31.97% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000206     6
94.41% :     0.000194s :      4: match.inline
 5.59% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.60% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.38% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000001s :     10: predicate.addn_zero_filter
 1.94% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000002s :     10: predicate.cast_eliminate
 0.57% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.22% :     0.000001s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.38% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.68% :     0.000007s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.41% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.47% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000002s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.49% :     0.000001s :      3: predicate.reset_defer_inline
 1.28% :     0.000001s :     10: predicate.reshape_eliminate
 1.31% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.03% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.61% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 2.40% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.08% :     0.000008s :     53: predicate.switch_simplify
 1.31% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.56% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.79% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.53% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000694    12
47.42% :     0.000329s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.78% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.80% :     0.000318s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022600    76
 0.25% :     0.000056s :      1: add_recomputation
 0.31% :     0.000070s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.51% :     0.000342s :      1: bootstrap
 0.13% :     0.000030s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.73% :     0.006040s :      1: jit_opt_a
 0.67% :     0.000153s :      1: jit_opt_after_cconv
 0.27% :     0.000062s :      1: jit_opt_b
 2.31% :     0.000523s :      1: loop_unroll
 3.30% :     0.000746s :      1: mutable_eliminate
 3.86% :     0.000872s :     26: opt.transform.jit_opt_a
 0.22% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000031s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.22% :     0.000501s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000036s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 2.28% :     0.000516s :      1: renormalize.infer
 1.80% :     0.000408s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000050s :      1: rewriter_after_opt_a
 0.32% :     0.000073s :      1: rewriter_before_opt_a
 0.36% :     0.000081s :      1: symbol_engine_optimizer
28.98% :     0.006550s :      1: task_emit
22.53% :     0.005091s :      1: type_inference
 0.28% :     0.000062s :      1: validate


TotalTime = 0.0182708, [33]
[bootstrap]: 0.00025421
[type_inference]: 0.00414099
[event_method]: 1.603e-05
[auto_monad]: 6.467e-05
[graph_reusing]: 5.66e-06
[pre_auto_parallel]: 2.47001e-06
[py_interpret_to_execute]: 3.13e-05
[rewriter_before_opt_a]: 7.123e-05
[expand_dump_flag]: 3.20002e-06
[jit_opt_a]: 0.00519824, [2]
    [Cycle 1]: 0.00146888, [27]
        [switch_simplify]: 5.211e-05
        [loop_unroll]: 2.884e-05
        [a_1]: 0.00048568
        [with_stream_mark]: 1.885e-05
        [recompute_prepare]: 7.93001e-06
        [updatestate_depend_eliminate]: 3.48e-06
        [updatestate_assign_eliminate]: 3.38999e-06
        [updatestate_loads_eliminate]: 2.43998e-06
        [parameter_eliminate]: 2.03002e-06
        [specialize_transform]: 5.82999e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.56e-06
        [meta_shard_fg_expand]: 2.19001e-06
        [get_grad_eliminate_]: 5.79e-06
        [merge_forward]: 3.83999e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.453e-05
        [j_node_and_user_rematch]: 9.38002e-06
        [meta_fg_expand]: 2.06998e-06
        [replace_old_param]: 7.65e-06
        [inline_without_move]: 5.79e-06
        [renormalize]: 0.00056352
        [add_forward_monad_depend]: 5.99999e-06
        [auto_monad_grad]: 2.98e-06
        [auto_monad_eliminator]: 1.627e-05
        [cse]: 2.446e-05
        [replace_applicator]: 1.311e-05
    [Cycle 2]: 0.00034416, [27]
        [switch_simplify]: 5.72001e-06
        [loop_unroll]: 5.17e-06
        [a_1]: 8.966e-05
        [with_stream_mark]: 9.46e-06
        [recompute_prepare]: 1.864e-05
        [updatestate_depend_eliminate]: 3.04001e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 1.74e-06
        [parameter_eliminate]: 1.56998e-06
        [specialize_transform]: 5.12e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.29e-06
        [meta_shard_fg_expand]: 1.45999e-06
        [get_grad_eliminate_]: 4.69998e-06
        [merge_forward]: 3.35e-06
        [cell_reuse_recompute_pass]: 2.48e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.363e-05
        [j_node_and_user_rematch]: 8.05e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 1.311e-05
        [inline_without_move]: 4.77998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.81e-06
        [auto_monad_grad]: 1.55001e-06
        [auto_monad_eliminator]: 5.72001e-06
        [cse]: 1.162e-05
        [replace_applicator]: 5.86e-06
[py_interpret_to_execute_after_opt_a]: 1.012e-05
[rewriter_after_opt_a]: 3.117e-05
[convert_after_rewriter]: 6.64001e-06
[order_py_execute_after_rewriter]: 5.57001e-06
[mutable_eliminate]: 0.00055581
[jit_opt_b]: 5.071e-05, [1]
    [Cycle 1]: 4.319e-05, [2]
        [frontend_op_eliminate]: 1.571e-05
        [inline_after_opt_a]: 1.567e-05
[cconv]: 2.857e-05
[loop_unroll]: 0.00045211
[jit_opt_after_cconv]: 0.00014306, [1]
    [Cycle 1]: 0.0001362, [11]
        [c_1]: 2.01e-05
        [parameter_eliminate]: 3.94002e-06
        [updatestate_depend_eliminate]: 6.63998e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.987e-05
        [call_graph_tuple_transform]: 1.905e-05
        [tuple_list_get_item_eliminator]: 5.00999e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 5.50004e-07
        [switch_simplify]: 5.09998e-06
[remove_dup_value]: 1.423e-05
[partial_unused_args_eliminate]: 2.22001e-06
[environ_conv]: 5.20001e-06
[add_recomputation]: 4.686e-05
[cse_after_recomputation]: 2.013e-05, [1]
    [Cycle 1]: 1.457e-05, [1]
        [cse]: 8.52998e-06
[auto_monad_reorder]: 1.472e-05
[get_jit_bprop_graph]: 1.90001e-06
[rewriter_after_jit_bprop_graph]: 7.41001e-06
[opt_after_jit_grad]: 0.0004689
[symbol_engine_optimizer]: 7.228e-05, [1]
    [Cycle 1]: 6.547e-05, [6]
        [build]: 4.33001e-06
        [elim_shapecalc]: 7.85e-06
        [elim_not_effective]: 1.238e-05
        [opt_reshape]: 5.14e-06
        [fold_const_symbol]: 8.22e-06
        [renormalize]: 6.89994e-07
[validate]: 3.53e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00632394
[execute]: 7.9e-06
Sums
    bootstrap                            :     0.000254s :  1.78%
    type_inference                       :     0.004141s : 28.94%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000065s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.22%
    rewriter_before_opt_a                :     0.000071s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000575s :  4.02%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000027s :  0.19%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000021s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000564s :  3.94%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.15%
    jit_opt_a.cse                        :     0.000036s :  0.25%
    jit_opt_a.replace_applicator         :     0.000019s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000031s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000556s :  3.89%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000029s :  0.20%
    loop_unroll                          :     0.000452s :  3.16%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.05%
    opt_after_jit_grad                   :     0.000469s :  3.28%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006324s : 44.20%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000174    23
 1.20% :     0.000002s :      2: substitution.elim_not_effective
 0.83% :     0.000001s :      2: substitution.fold_const_symbol
 3.34% :     0.000006s :      3: substitution.graph_param_transform
80.51% :     0.000140s :      4: substitution.inline
 2.14% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.65% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.90% :     0.000003s :      2: substitution.replace_old_param
 7.44% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004082     2
81.20% :     0.003315s :      1: type_inference.infer
18.80% :     0.000767s :      1: type_inference.specialize
------[replace.]   0.000062     6
65.56% :     0.000040s :      4: replace.inline
34.44% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000150     6
92.08% :     0.000138s :      4: match.inline
 7.92% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.28% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.11% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.79% :     0.000002s :     10: predicate.addn_zero_filter
 2.08% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.52% :     0.000001s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.70% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000001s :      6: predicate.less_batch_normalization
 1.64% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.60% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.23% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.90% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000000s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.54% :     0.000001s :      3: predicate.reset_defer_inline
 1.53% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.13% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.60% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.17% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.31% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.54% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.00% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000487    12
47.40% :     0.000231s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.55% :     0.000022s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.04% :     0.000234s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019738    76
 0.25% :     0.000050s :      1: add_recomputation
 0.34% :     0.000068s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.33% :     0.000263s :      1: bootstrap
 0.16% :     0.000031s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.35% :     0.005201s :      1: jit_opt_a
 0.74% :     0.000146s :      1: jit_opt_after_cconv
 0.27% :     0.000053s :      1: jit_opt_b
 2.34% :     0.000462s :      1: loop_unroll
 2.87% :     0.000567s :      1: mutable_eliminate
 4.00% :     0.000790s :     26: opt.transform.jit_opt_a
 0.23% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.42% :     0.000477s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.57% :     0.000310s :      1: renormalize.infer
 1.24% :     0.000245s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000034s :      1: rewriter_after_opt_a
 0.38% :     0.000074s :      1: rewriter_before_opt_a
 0.38% :     0.000075s :      1: symbol_engine_optimizer
32.09% :     0.006335s :      1: task_emit
21.03% :     0.004150s :      1: type_inference
 0.28% :     0.000055s :      1: validate


TotalTime = 0.0168028, [30]
[bootstrap]: 0.00028753
[type_inference]: 0.00719926
[event_method]: 2.111e-05
[auto_monad]: 8.58e-05
[graph_reusing]: 6.33e-06
[pre_auto_parallel]: 2.79001e-06
[py_interpret_to_execute]: 6.625e-05
[rewriter_before_opt_a]: 9.25e-05
[expand_dump_flag]: 3.16001e-06
[jit_opt_a]: 0.00641066, [2]
    [Cycle 1]: 0.00237326, [27]
        [switch_simplify]: 0.00013813
        [loop_unroll]: 3.629e-05
        [a_1]: 0.00084505
        [with_stream_mark]: 1.933e-05
        [recompute_prepare]: 8.08001e-06
        [updatestate_depend_eliminate]: 2.88e-06
        [updatestate_assign_eliminate]: 2.43002e-06
        [updatestate_loads_eliminate]: 2.59001e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 6.16998e-06
        [updatestate_useless_node_eliminater]: 5.35001e-06
        [accelerated_algorithm]: 6.01003e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 5.40999e-06
        [merge_forward]: 3.53999e-06
        [cell_reuse_recompute_pass]: 1.07998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.422e-05
        [j_node_and_user_rematch]: 9.31e-06
        [meta_fg_expand]: 2.42001e-06
        [replace_old_param]: 9.19e-06
        [inline_without_move]: 5.19e-06
        [renormalize]: 0.00100256
        [add_forward_monad_depend]: 7.06999e-06
        [auto_monad_grad]: 2.72001e-06
        [auto_monad_eliminator]: 1.541e-05
        [cse]: 2.727e-05
        [replace_applicator]: 1.567e-05
    [Cycle 2]: 0.00036725, [27]
        [switch_simplify]: 6.44999e-06
        [loop_unroll]: 6.24999e-06
        [a_1]: 9.985e-05
        [with_stream_mark]: 1.227e-05
        [recompute_prepare]: 6.09001e-06
        [updatestate_depend_eliminate]: 2.91e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [parameter_eliminate]: 1.56002e-06
        [specialize_transform]: 5.14e-06
        [updatestate_useless_node_eliminater]: 5.47999e-06
        [accelerated_algorithm]: 6.14001e-06
        [meta_shard_fg_expand]: 2.48e-06
        [get_grad_eliminate_]: 4.93001e-06
        [merge_forward]: 3.48999e-06
        [cell_reuse_recompute_pass]: 2.58998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.622e-05
        [j_node_and_user_rematch]: 9.89999e-06
        [meta_fg_expand]: 2.41e-06
        [replace_old_param]: 9.56998e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.54e-06
        [auto_monad_grad]: 1.14998e-06
        [auto_monad_eliminator]: 5.87001e-06
        [cse]: 1.296e-05
        [replace_applicator]: 6.89001e-06
[py_interpret_to_execute_after_opt_a]: 1.555e-05
[rewriter_after_opt_a]: 3.89e-05
[convert_after_rewriter]: 6.88e-06
[order_py_execute_after_rewriter]: 5.32999e-06
[mutable_eliminate]: 0.00085802
[jit_opt_b]: 5.888e-05, [1]
    [Cycle 1]: 4.856e-05, [2]
        [frontend_op_eliminate]: 1.886e-05
        [inline_after_opt_a]: 1.622e-05
[cconv]: 3.066e-05
[loop_unroll]: 0.00051522
[jit_opt_after_cconv]: 0.00015229, [1]
    [Cycle 1]: 0.00014527, [11]
        [c_1]: 2.243e-05
        [parameter_eliminate]: 3.01001e-06
        [updatestate_depend_eliminate]: 5.84e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.33002e-06
        [cse]: 2.201e-05
        [call_graph_tuple_transform]: 2.052e-05
        [tuple_list_get_item_eliminator]: 5.74e-06
        [none_parameter_eliminate]: 1.53002e-06
        [renormalize]: 7.89994e-07
        [switch_simplify]: 5.64e-06
[remove_dup_value]: 1.662e-05
[partial_unused_args_eliminate]: 2.45002e-06
[environ_conv]: 5.39e-06
[add_recomputation]: 4.745e-05
[cse_after_recomputation]: 2.425e-05, [1]
    [Cycle 1]: 1.809e-05, [1]
        [cse]: 1.18e-05
[auto_monad_reorder]: 1.578e-05
[get_jit_bprop_graph]: 2.38002e-06
[rewriter_after_jit_bprop_graph]: 6.16998e-06
[opt_after_jit_grad]: 0.00051061
[symbol_engine_optimizer]: 9.344e-05, [1]
    [Cycle 1]: 8.627e-05, [6]
        [build]: 3.73001e-06
        [elim_shapecalc]: 8.99998e-06
        [elim_not_effective]: 1.569e-05
        [opt_reshape]: 7.48e-06
        [fold_const_symbol]: 9.54e-06
        [renormalize]: 7.00005e-07
[validate]: 4.38e-05
Sums
    bootstrap                            :     0.000288s :  2.30%
    type_inference                       :     0.007199s : 57.55%
    event_method                         :     0.000021s :  0.17%
    auto_monad                           :     0.000086s :  0.69%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000066s :  0.53%
    rewriter_before_opt_a                :     0.000093s :  0.74%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000145s :  1.16%
    jit_opt_a.loop_unroll                :     0.000043s :  0.34%
    jit_opt_a.a_1                        :     0.000945s :  7.55%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.001003s :  8.01%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.17%
    jit_opt_a.cse                        :     0.000040s :  0.32%
    jit_opt_a.replace_applicator         :     0.000023s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.12%
    rewriter_after_opt_a                 :     0.000039s :  0.31%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000858s :  6.86%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.13%
    cconv                                :     0.000031s :  0.25%
    loop_unroll                          :     0.000515s :  4.12%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000017s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.38%
    cse_after_recomputation.cse          :     0.000012s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000511s :  4.08%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.35%


Time group info:
------[substitution.]   0.000379    26
 0.48% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000002s :      2: substitution.fold_const_symbol
 1.49% :     0.000006s :      3: substitution.graph_param_transform
87.27% :     0.000331s :      6: substitution.inline
 1.08% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.40% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.39% :     0.000005s :      2: substitution.replace_old_param
 2.62% :     0.000010s :      1: substitution.switch_simplify
 3.78% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007131     2
80.70% :     0.005755s :      1: type_inference.infer
19.30% :     0.001376s :      1: type_inference.specialize
------[replace.]   0.000178     9
46.63% :     0.000083s :      6: replace.inline
25.10% :     0.000045s :      1: replace.switch_simplify
28.26% :     0.000050s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000349     9
93.63% :     0.000327s :      6: match.inline
 2.65% :     0.000009s :      1: match.switch_simplify
 3.73% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000141   801
 1.23% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.25% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000002s :     12: predicate.addn_check_dump
 1.39% :     0.000002s :     12: predicate.addn_zero_filter
 2.32% :     0.000003s :     12: predicate.arithmetic_simplify
 1.38% :     0.000002s :     12: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000002s :     12: predicate.compare_switch_simplify
 1.41% :     0.000002s :     12: predicate.depend_value_elim
 1.14% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.18% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.61% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.56% :     0.000001s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.08% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     12: predicate.environ_get_eliminate
 1.08% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.08% :     0.000002s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 4.96% :     0.000007s :     26: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000002s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.95% :     0.000003s :     17: predicate.load_eliminater
 1.04% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.93% :     0.000007s :     40: predicate.loop_unroll_before_grad
 1.78% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000002s :     12: predicate.merge_addn
 1.34% :     0.000002s :     12: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.38% :     0.000001s :      3: predicate.opt_reshape
 1.99% :     0.000003s :     17: predicate.partial_eliminate
 1.42% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.77% :     0.000002s :     12: predicate.reduce_eliminate
 1.36% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     20: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     12: predicate.reshape_eliminate
 1.23% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     12: predicate.same_eliminate
 0.64% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.84% :     0.000003s :     12: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.59% :     0.000004s :     20: predicate.switch_defer_inline
 2.25% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.65% :     0.000012s :     65: predicate.switch_simplify
 1.31% :     0.000002s :     12: predicate.tile_eliminate
 1.18% :     0.000002s :     12: predicate.transpose_eliminate
 1.51% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.49% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.60% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.56% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     12: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001047    15
47.40% :     0.000496s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.50% :     0.000141s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.10% :     0.000410s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019121    72
 0.27% :     0.000051s :      1: add_recomputation
 0.47% :     0.000089s :      1: auto_monad
 0.10% :     0.000019s :      1: auto_monad_reorder
 1.56% :     0.000299s :      1: bootstrap
 0.18% :     0.000034s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.13% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.56% :     0.006417s :      1: jit_opt_a
 0.82% :     0.000156s :      1: jit_opt_after_cconv
 0.32% :     0.000061s :      1: jit_opt_b
 2.74% :     0.000524s :      1: loop_unroll
 4.54% :     0.000868s :      1: mutable_eliminate
 6.59% :     0.001261s :     26: opt.transform.jit_opt_a
 0.26% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000028s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000038s :      4: opt.transform.symbol_engine_opt
 2.71% :     0.000518s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.37% :     0.000070s :      1: py_interpret_to_execute
 0.09% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000019s :      1: remove_dup_value
 2.96% :     0.000567s :      1: renormalize.infer
 2.22% :     0.000425s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000042s :      1: rewriter_after_opt_a
 0.50% :     0.000095s :      1: rewriter_before_opt_a
 0.50% :     0.000096s :      1: symbol_engine_optimizer
37.70% :     0.007209s :      1: type_inference


TotalTime = 0.0210598, [33]
[bootstrap]: 0.00030405
[type_inference]: 0.00495709
[event_method]: 2.072e-05
[auto_monad]: 7.082e-05
[graph_reusing]: 6.31e-06
[pre_auto_parallel]: 3.50003e-06
[py_interpret_to_execute]: 4.121e-05
[rewriter_before_opt_a]: 7.554e-05
[expand_dump_flag]: 3.76999e-06
[jit_opt_a]: 0.0063379, [2]
    [Cycle 1]: 0.00207405, [27]
        [switch_simplify]: 6.312e-05
        [loop_unroll]: 3.006e-05
        [a_1]: 0.00063644
        [with_stream_mark]: 2.741e-05
        [recompute_prepare]: 9.54999e-06
        [updatestate_depend_eliminate]: 4.72998e-06
        [updatestate_assign_eliminate]: 3.21001e-06
        [updatestate_loads_eliminate]: 2.87002e-06
        [parameter_eliminate]: 2.09e-06
        [specialize_transform]: 7.21001e-06
        [updatestate_useless_node_eliminater]: 5.48997e-06
        [accelerated_algorithm]: 5.99999e-06
        [meta_shard_fg_expand]: 2.16998e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 4.27e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.572e-05
        [j_node_and_user_rematch]: 1.081e-05
        [meta_fg_expand]: 2.86e-06
        [replace_old_param]: 1.077e-05
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.00094289
        [add_forward_monad_depend]: 6.74999e-06
        [auto_monad_grad]: 2.85002e-06
        [auto_monad_eliminator]: 1.774e-05
        [cse]: 3.114e-05
        [replace_applicator]: 1.516e-05
    [Cycle 2]: 0.00034351, [27]
        [switch_simplify]: 5.65001e-06
        [loop_unroll]: 5.84e-06
        [a_1]: 9.38e-05
        [with_stream_mark]: 1.274e-05
        [recompute_prepare]: 5.64998e-06
        [updatestate_depend_eliminate]: 3.35e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.08998e-06
        [parameter_eliminate]: 1.37e-06
        [specialize_transform]: 4.75001e-06
        [updatestate_useless_node_eliminater]: 6.08998e-06
        [accelerated_algorithm]: 5.34e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.53999e-06
        [merge_forward]: 3.04999e-06
        [cell_reuse_recompute_pass]: 2.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.401e-05
        [j_node_and_user_rematch]: 8.52e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 8.06001e-06
        [inline_without_move]: 4.70999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 9.29984e-07
        [auto_monad_eliminator]: 5.39e-06
        [cse]: 1.115e-05
        [replace_applicator]: 5.28002e-06
[py_interpret_to_execute_after_opt_a]: 1.225e-05
[rewriter_after_opt_a]: 3.693e-05
[convert_after_rewriter]: 6.91999e-06
[order_py_execute_after_rewriter]: 5.08002e-06
[mutable_eliminate]: 0.00066506
[jit_opt_b]: 6.733e-05, [1]
    [Cycle 1]: 4.732e-05, [2]
        [frontend_op_eliminate]: 1.868e-05
        [inline_after_opt_a]: 1.621e-05
[cconv]: 3.146e-05
[loop_unroll]: 0.00048917
[jit_opt_after_cconv]: 0.00015119, [1]
    [Cycle 1]: 0.00014376, [11]
        [c_1]: 2.224e-05
        [parameter_eliminate]: 3.26999e-06
        [updatestate_depend_eliminate]: 7.46999e-06
        [updatestate_assign_eliminate]: 2.53998e-06
        [updatestate_loads_eliminate]: 2.34999e-06
        [cse]: 2.108e-05
        [call_graph_tuple_transform]: 2.094e-05
        [tuple_list_get_item_eliminator]: 5.60001e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 7.59988e-07
        [switch_simplify]: 5.42999e-06
[remove_dup_value]: 1.441e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 5.97999e-06
[add_recomputation]: 5.282e-05
[cse_after_recomputation]: 2.185e-05, [1]
    [Cycle 1]: 1.621e-05, [1]
        [cse]: 9.47999e-06
[auto_monad_reorder]: 1.596e-05
[get_jit_bprop_graph]: 1.84998e-06
[rewriter_after_jit_bprop_graph]: 6.31998e-06
[opt_after_jit_grad]: 0.00056176
[symbol_engine_optimizer]: 7.489e-05, [1]
    [Cycle 1]: 6.822e-05, [6]
        [build]: 4.87e-06
        [elim_shapecalc]: 7.35e-06
        [elim_not_effective]: 1.363e-05
        [opt_reshape]: 5.78997e-06
        [fold_const_symbol]: 8.85999e-06
        [renormalize]: 4.99975e-07
[validate]: 4.021e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.0067364
[execute]: 9.92999e-06
Sums
    bootstrap                            :     0.000304s :  1.85%
    type_inference                       :     0.004957s : 30.13%
    event_method                         :     0.000021s :  0.13%
    auto_monad                           :     0.000071s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000004s :  0.02%
    py_interpret_to_execute              :     0.000041s :  0.25%
    rewriter_before_opt_a                :     0.000076s :  0.46%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000069s :  0.42%
    jit_opt_a.loop_unroll                :     0.000036s :  0.22%
    jit_opt_a.a_1                        :     0.000730s :  4.44%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000943s :  5.73%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.14%
    jit_opt_a.cse                        :     0.000042s :  0.26%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000665s :  4.04%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000031s :  0.19%
    loop_unroll                          :     0.000489s :  2.97%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000053s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000562s :  3.41%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000040s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006736s : 40.95%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000260    23
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.22% :     0.000006s :      3: substitution.graph_param_transform
85.79% :     0.000223s :      4: substitution.inline
 1.31% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.90% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.24% :     0.000006s :      2: substitution.replace_old_param
 5.33% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004876     2
80.14% :     0.003907s :      1: type_inference.infer
19.86% :     0.000969s :      1: type_inference.specialize
------[replace.]   0.000081     6
64.86% :     0.000053s :      4: replace.inline
35.14% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000232     6
94.73% :     0.000220s :      4: match.inline
 5.27% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.56% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000002s :     10: predicate.addn_zero_filter
 2.79% :     0.000004s :     10: predicate.arithmetic_simplify
 1.23% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 2.13% :     0.000003s :     10: predicate.depend_value_elim
 0.99% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.06% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.49% :     0.000007s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000002s :      6: predicate.less_batch_normalization
 1.36% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.57% :     0.000002s :     15: predicate.load_eliminater
 1.84% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.86% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.12% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 2.27% :     0.000003s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.74% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.25% :     0.000002s :     10: predicate.reshape_eliminate
 1.44% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.08% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.22% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.10% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.94% :     0.000004s :     16: predicate.switch_defer_inline
 2.08% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.62% :     0.000010s :     53: predicate.switch_simplify
 1.08% :     0.000001s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.55% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.44% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.42% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000701    12
45.97% :     0.000322s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.89% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.14% :     0.000323s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023074    76
 0.24% :     0.000056s :      1: add_recomputation
 0.32% :     0.000075s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.40% :     0.000323s :      1: bootstrap
 0.15% :     0.000034s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.11% :     0.000026s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
27.49% :     0.006342s :      1: jit_opt_a
 0.67% :     0.000154s :      1: jit_opt_after_cconv
 0.31% :     0.000071s :      1: jit_opt_b
 2.16% :     0.000498s :      1: loop_unroll
 2.93% :     0.000677s :      1: mutable_eliminate
 4.17% :     0.000962s :     26: opt.transform.jit_opt_a
 0.22% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000028s :      4: opt.transform.jit_opt_b
 0.06% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.47% :     0.000570s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000044s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.23% :     0.000514s :      1: renormalize.infer
 1.81% :     0.000418s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000040s :      1: rewriter_after_opt_a
 0.34% :     0.000079s :      1: rewriter_before_opt_a
 0.34% :     0.000078s :      1: symbol_engine_optimizer
29.28% :     0.006756s :      1: task_emit
21.56% :     0.004974s :      1: type_inference
 0.27% :     0.000062s :      1: validate


TotalTime = 0.0210998, [33]
[bootstrap]: 0.0003215
[type_inference]: 0.00474984
[event_method]: 1.975e-05
[auto_monad]: 6.714e-05
[graph_reusing]: 6.31e-06
[pre_auto_parallel]: 2.51e-06
[py_interpret_to_execute]: 3.989e-05
[rewriter_before_opt_a]: 7.449e-05
[expand_dump_flag]: 3.14001e-06
[jit_opt_a]: 0.00631513, [2]
    [Cycle 1]: 0.00193008, [27]
        [switch_simplify]: 6.997e-05
        [loop_unroll]: 3.066e-05
        [a_1]: 0.000585
        [with_stream_mark]: 2.125e-05
        [recompute_prepare]: 9.51e-06
        [updatestate_depend_eliminate]: 4.35e-06
        [updatestate_assign_eliminate]: 2.76e-06
        [updatestate_loads_eliminate]: 2.94001e-06
        [parameter_eliminate]: 1.81998e-06
        [specialize_transform]: 6.04999e-06
        [updatestate_useless_node_eliminater]: 5.42999e-06
        [accelerated_algorithm]: 5.36998e-06
        [meta_shard_fg_expand]: 2.76e-06
        [get_grad_eliminate_]: 5.39998e-06
        [merge_forward]: 3.88001e-06
        [cell_reuse_recompute_pass]: 1.58002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.591e-05
        [j_node_and_user_rematch]: 9.47999e-06
        [meta_fg_expand]: 3.01999e-06
        [replace_old_param]: 9.99001e-06
        [inline_without_move]: 5.27999e-06
        [renormalize]: 0.00086677
        [add_forward_monad_depend]: 6.64001e-06
        [auto_monad_grad]: 2.35002e-06
        [auto_monad_eliminator]: 1.613e-05
        [cse]: 3.325e-05
        [replace_applicator]: 1.539e-05
    [Cycle 2]: 0.00033462, [27]
        [switch_simplify]: 6.17001e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 9.53e-05
        [with_stream_mark]: 1.248e-05
        [recompute_prepare]: 4.97999e-06
        [updatestate_depend_eliminate]: 3.14999e-06
        [updatestate_assign_eliminate]: 1.97999e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [parameter_eliminate]: 1.17999e-06
        [specialize_transform]: 5.03002e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 1.69998e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.66999e-06
        [cell_reuse_recompute_pass]: 1.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.425e-05
        [j_node_and_user_rematch]: 7.60998e-06
        [meta_fg_expand]: 1.98002e-06
        [replace_old_param]: 7.09001e-06
        [inline_without_move]: 4.61002e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.60001e-06
        [auto_monad_grad]: 9.09989e-07
        [auto_monad_eliminator]: 5.43002e-06
        [cse]: 1.086e-05
        [replace_applicator]: 5.07e-06
[py_interpret_to_execute_after_opt_a]: 1.386e-05
[rewriter_after_opt_a]: 8.697e-05
[convert_after_rewriter]: 7.87e-06
[order_py_execute_after_rewriter]: 5.22e-06
[mutable_eliminate]: 0.00076756
[jit_opt_b]: 5.288e-05, [1]
    [Cycle 1]: 4.472e-05, [2]
        [frontend_op_eliminate]: 1.722e-05
        [inline_after_opt_a]: 1.516e-05
[cconv]: 2.99e-05
[loop_unroll]: 0.0004561
[jit_opt_after_cconv]: 0.00014782, [1]
    [Cycle 1]: 0.00014094, [11]
        [c_1]: 2.139e-05
        [parameter_eliminate]: 2.96001e-06
        [updatestate_depend_eliminate]: 5.74999e-06
        [updatestate_assign_eliminate]: 2.71999e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 2.276e-05
        [call_graph_tuple_transform]: 1.997e-05
        [tuple_list_get_item_eliminator]: 5.49e-06
        [none_parameter_eliminate]: 1.54998e-06
        [renormalize]: 9.09989e-07
        [switch_simplify]: 5.35999e-06
[remove_dup_value]: 1.588e-05
[partial_unused_args_eliminate]: 2.39001e-06
[environ_conv]: 5.95002e-06
[add_recomputation]: 5.2e-05
[cse_after_recomputation]: 2.272e-05, [1]
    [Cycle 1]: 1.575e-05, [1]
        [cse]: 9.66e-06
[auto_monad_reorder]: 1.587e-05
[get_jit_bprop_graph]: 2.75002e-06
[rewriter_after_jit_bprop_graph]: 5.94999e-06
[opt_after_jit_grad]: 0.00047114
[symbol_engine_optimizer]: 7.434e-05, [1]
    [Cycle 1]: 6.781e-05, [6]
        [build]: 3.97998e-06
        [elim_shapecalc]: 7.90998e-06
        [elim_not_effective]: 1.311e-05
        [opt_reshape]: 5.67999e-06
        [fold_const_symbol]: 9.07999e-06
        [renormalize]: 7.39994e-07
[validate]: 3.993e-05
[backend_pass]: 8.69972e-07
[task_emit]: 0.00699624
[execute]: 1.005e-05
Sums
    bootstrap                            :     0.000321s :  1.96%
    type_inference                       :     0.004750s : 28.95%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000067s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.24%
    rewriter_before_opt_a                :     0.000074s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.46%
    jit_opt_a.loop_unroll                :     0.000036s :  0.22%
    jit_opt_a.a_1                        :     0.000680s :  4.15%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000867s :  5.28%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.13%
    jit_opt_a.cse                        :     0.000044s :  0.27%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000087s :  0.53%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000768s :  4.68%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000030s :  0.18%
    loop_unroll                          :     0.000456s :  2.78%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000052s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000471s :  2.87%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006996s : 42.65%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000238    23
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000002s :      2: substitution.fold_const_symbol
 2.53% :     0.000006s :      3: substitution.graph_param_transform
84.25% :     0.000200s :      4: substitution.inline
 1.32% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.44% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.07% :     0.000005s :      2: substitution.replace_old_param
 5.85% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004672     2
80.22% :     0.003747s :      1: type_inference.infer
19.78% :     0.000924s :      1: type_inference.specialize
------[replace.]   0.000076     6
64.07% :     0.000049s :      4: replace.inline
35.93% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000211     6
93.90% :     0.000198s :      4: match.inline
 6.10% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.72% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000002s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.57% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.38% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.07% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.15% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.77% :     0.000007s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      6: predicate.less_batch_normalization
 1.78% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.62% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.05% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.35% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.46% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.53% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.19% :     0.000001s :     10: predicate.reshape_eliminate
 1.30% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.67% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.65% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.24% :     0.000009s :     53: predicate.switch_simplify
 1.39% :     0.000002s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.08% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.52% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.02% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000648    12
45.00% :     0.000291s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.24% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.76% :     0.000309s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022979    76
 0.24% :     0.000056s :      1: add_recomputation
 0.31% :     0.000070s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.46% :     0.000335s :      1: bootstrap
 0.14% :     0.000032s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.11% :     0.000024s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.50% :     0.006319s :      1: jit_opt_a
 0.65% :     0.000150s :      1: jit_opt_after_cconv
 0.24% :     0.000056s :      1: jit_opt_b
 2.02% :     0.000464s :      1: loop_unroll
 3.38% :     0.000777s :      1: mutable_eliminate
 3.97% :     0.000913s :     26: opt.transform.jit_opt_a
 0.21% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.08% :     0.000479s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000043s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 1.96% :     0.000450s :      1: renormalize.infer
 1.78% :     0.000408s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.39% :     0.000090s :      1: rewriter_after_opt_a
 0.34% :     0.000078s :      1: rewriter_before_opt_a
 0.33% :     0.000077s :      1: symbol_engine_optimizer
30.54% :     0.007017s :      1: task_emit
20.73% :     0.004764s :      1: type_inference
 0.27% :     0.000061s :      1: validate


TotalTime = 0.0205726, [33]
[bootstrap]: 0.0003093
[type_inference]: 0.00524786
[event_method]: 1.771e-05
[auto_monad]: 6.51e-05
[graph_reusing]: 6.51e-06
[pre_auto_parallel]: 2.43e-06
[py_interpret_to_execute]: 2.985e-05
[rewriter_before_opt_a]: 7.191e-05
[expand_dump_flag]: 3.01999e-06
[jit_opt_a]: 0.00588878, [2]
    [Cycle 1]: 0.00182277, [27]
        [switch_simplify]: 5.566e-05
        [loop_unroll]: 2.913e-05
        [a_1]: 0.00057919
        [with_stream_mark]: 2.495e-05
        [recompute_prepare]: 9.24e-06
        [updatestate_depend_eliminate]: 3.62002e-06
        [updatestate_assign_eliminate]: 3.14999e-06
        [updatestate_loads_eliminate]: 2.82002e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 6.38003e-06
        [updatestate_useless_node_eliminater]: 5.01002e-06
        [accelerated_algorithm]: 5.15001e-06
        [meta_shard_fg_expand]: 2.43002e-06
        [get_grad_eliminate_]: 5.32999e-06
        [merge_forward]: 4.22998e-06
        [cell_reuse_recompute_pass]: 1.87001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.41e-05
        [j_node_and_user_rematch]: 1.008e-05
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 9.73998e-06
        [inline_without_move]: 5.27001e-06
        [renormalize]: 0.00077659
        [add_forward_monad_depend]: 6.54999e-06
        [auto_monad_grad]: 2.47001e-06
        [auto_monad_eliminator]: 1.681e-05
        [cse]: 2.823e-05
        [replace_applicator]: 1.805e-05
    [Cycle 2]: 0.00034963, [27]
        [switch_simplify]: 5.00999e-06
        [loop_unroll]: 4.72e-06
        [a_1]: 9.011e-05
        [with_stream_mark]: 1.392e-05
        [recompute_prepare]: 4.74e-06
        [updatestate_depend_eliminate]: 2.81e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 4.82998e-06
        [updatestate_useless_node_eliminater]: 5.30001e-06
        [accelerated_algorithm]: 5.32001e-06
        [meta_shard_fg_expand]: 2.39001e-06
        [get_grad_eliminate_]: 4.89e-06
        [merge_forward]: 2.83e-06
        [cell_reuse_recompute_pass]: 2.46e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.526e-05
        [j_node_and_user_rematch]: 8.55001e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 8.77e-06
        [inline_without_move]: 4.92e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.67001e-06
        [auto_monad_grad]: 1.29e-06
        [auto_monad_eliminator]: 7.23e-06
        [cse]: 1.362e-05
        [replace_applicator]: 7.35003e-06
[py_interpret_to_execute_after_opt_a]: 1.62e-05
[rewriter_after_opt_a]: 3.491e-05
[convert_after_rewriter]: 7.33e-06
[order_py_execute_after_rewriter]: 5.50001e-06
[mutable_eliminate]: 0.00070395
[jit_opt_b]: 5.061e-05, [1]
    [Cycle 1]: 4.316e-05, [2]
        [frontend_op_eliminate]: 1.606e-05
        [inline_after_opt_a]: 1.461e-05
[cconv]: 2.731e-05
[loop_unroll]: 0.00045501
[jit_opt_after_cconv]: 0.0001434, [1]
    [Cycle 1]: 0.00013625, [11]
        [c_1]: 2.072e-05
        [parameter_eliminate]: 3.70998e-06
        [updatestate_depend_eliminate]: 6.86001e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 1.919e-05
        [call_graph_tuple_transform]: 1.938e-05
        [tuple_list_get_item_eliminator]: 5.02999e-06
        [none_parameter_eliminate]: 1.67001e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 5.30999e-06
[remove_dup_value]: 1.283e-05
[partial_unused_args_eliminate]: 2.04e-06
[environ_conv]: 5.70001e-06
[add_recomputation]: 4.726e-05
[cse_after_recomputation]: 1.945e-05, [1]
    [Cycle 1]: 1.373e-05, [1]
        [cse]: 8.03999e-06
[auto_monad_reorder]: 1.483e-05
[get_jit_bprop_graph]: 2.27999e-06
[rewriter_after_jit_bprop_graph]: 5.54e-06
[opt_after_jit_grad]: 0.00045747
[symbol_engine_optimizer]: 7.894e-05, [1]
    [Cycle 1]: 6.385e-05, [6]
        [build]: 3.34001e-06
        [elim_shapecalc]: 7.46001e-06
        [elim_not_effective]: 1.181e-05
        [opt_reshape]: 5.51e-06
        [fold_const_symbol]: 8.44998e-06
        [renormalize]: 6.29982e-07
[validate]: 3.538e-05
[backend_pass]: 8.29983e-07
[task_emit]: 0.00658719
[execute]: 9.00001e-06
Sums
    bootstrap                            :     0.000309s :  1.91%
    type_inference                       :     0.005248s : 32.39%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000065s :  0.40%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000030s :  0.18%
    rewriter_before_opt_a                :     0.000072s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.37%
    jit_opt_a.loop_unroll                :     0.000034s :  0.21%
    jit_opt_a.a_1                        :     0.000669s :  4.13%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000777s :  4.79%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.15%
    jit_opt_a.cse                        :     0.000042s :  0.26%
    jit_opt_a.replace_applicator         :     0.000025s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.10%
    rewriter_after_opt_a                 :     0.000035s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000704s :  4.34%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000027s :  0.17%
    loop_unroll                          :     0.000455s :  2.81%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000047s :  0.29%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000457s :  2.82%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006587s : 40.65%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000228    23
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.74% :     0.000006s :      3: substitution.graph_param_transform
83.87% :     0.000191s :      4: substitution.inline
 1.83% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.96% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.65% :     0.000006s :      2: substitution.replace_old_param
 5.44% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005177     2
83.64% :     0.004330s :      1: type_inference.infer
16.36% :     0.000847s :      1: type_inference.specialize
------[replace.]   0.000074     6
63.06% :     0.000047s :      4: replace.inline
36.94% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000199     6
94.38% :     0.000188s :      4: match.inline
 5.62% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.19% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.15% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000001s :     10: predicate.addn_zero_filter
 2.29% :     0.000003s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.39% :     0.000002s :     10: predicate.compare_switch_simplify
 1.95% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.27% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 6.47% :     0.000008s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.34% :     0.000002s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.69% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.93% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.48% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 1.99% :     0.000002s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.67% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.94% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.14% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.23% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.67% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.95% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.33% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.25% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.33% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000637    12
49.73% :     0.000317s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.28% :     0.000046s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.98% :     0.000274s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022339    76
 0.22% :     0.000050s :      1: add_recomputation
 0.31% :     0.000069s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.45% :     0.000325s :      1: bootstrap
 0.13% :     0.000030s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.38% :     0.005892s :      1: jit_opt_a
 0.65% :     0.000146s :      1: jit_opt_after_cconv
 0.24% :     0.000053s :      1: jit_opt_b
 2.07% :     0.000463s :      1: loop_unroll
 3.20% :     0.000716s :      1: mutable_eliminate
 3.98% :     0.000889s :     26: opt.transform.jit_opt_a
 0.21% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.08% :     0.000466s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000033s :      1: py_interpret_to_execute
 0.08% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 1.87% :     0.000418s :      1: renormalize.infer
 1.56% :     0.000349s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000038s :      1: rewriter_after_opt_a
 0.34% :     0.000075s :      1: rewriter_before_opt_a
 0.36% :     0.000081s :      1: symbol_engine_optimizer
29.57% :     0.006605s :      1: task_emit
23.57% :     0.005264s :      1: type_inference
 0.25% :     0.000056s :      1: validate


TotalTime = 0.0179346, [30]
[bootstrap]: 0.0002962
[type_inference]: 0.00872038
[event_method]: 2.664e-05
[auto_monad]: 8.964e-05
[graph_reusing]: 7.21001e-06
[pre_auto_parallel]: 2.83e-06
[py_interpret_to_execute]: 6.871e-05
[rewriter_before_opt_a]: 9.577e-05
[expand_dump_flag]: 3.49001e-06
[jit_opt_a]: 0.0063167, [2]
    [Cycle 1]: 0.00238372, [27]
        [switch_simplify]: 0.00012633
        [loop_unroll]: 3.623e-05
        [a_1]: 0.00075057
        [with_stream_mark]: 2.002e-05
        [recompute_prepare]: 8.1e-06
        [updatestate_depend_eliminate]: 3.5e-06
        [updatestate_assign_eliminate]: 3.07002e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [parameter_eliminate]: 1.87001e-06
        [specialize_transform]: 6.46999e-06
        [updatestate_useless_node_eliminater]: 5.10001e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 2.40002e-06
        [get_grad_eliminate_]: 5.56e-06
        [merge_forward]: 3.58999e-06
        [cell_reuse_recompute_pass]: 1.11002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.61e-05
        [j_node_and_user_rematch]: 9.84001e-06
        [meta_fg_expand]: 2.29999e-06
        [replace_old_param]: 9.97999e-06
        [inline_without_move]: 5.71e-06
        [renormalize]: 0.00111066
        [add_forward_monad_depend]: 5.42001e-06
        [auto_monad_grad]: 2.55997e-06
        [auto_monad_eliminator]: 1.404e-05
        [cse]: 2.907e-05
        [replace_applicator]: 1.341e-05
    [Cycle 2]: 0.00032303, [27]
        [switch_simplify]: 5.92999e-06
        [loop_unroll]: 5.32001e-06
        [a_1]: 9.332e-05
        [with_stream_mark]: 1.042e-05
        [recompute_prepare]: 5.12e-06
        [updatestate_depend_eliminate]: 2.91999e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 1.98002e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.02999e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 4.83001e-06
        [meta_shard_fg_expand]: 1.31002e-06
        [get_grad_eliminate_]: 4.73001e-06
        [merge_forward]: 2.99999e-06
        [cell_reuse_recompute_pass]: 1.64998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.293e-05
        [j_node_and_user_rematch]: 8.23999e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 6.76999e-06
        [inline_without_move]: 4.89e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.54e-06
        [auto_monad_grad]: 1.00999e-06
        [auto_monad_eliminator]: 5.11997e-06
        [cse]: 9.65002e-06
        [replace_applicator]: 5.20999e-06
[py_interpret_to_execute_after_opt_a]: 1.072e-05
[rewriter_after_opt_a]: 3.441e-05
[convert_after_rewriter]: 7.78001e-06
[order_py_execute_after_rewriter]: 4.80999e-06
[mutable_eliminate]: 0.00065475
[jit_opt_b]: 5.123e-05, [1]
    [Cycle 1]: 4.401e-05, [2]
        [frontend_op_eliminate]: 1.745e-05
        [inline_after_opt_a]: 1.456e-05
[cconv]: 2.425e-05
[loop_unroll]: 0.00047778
[jit_opt_after_cconv]: 0.00014765, [1]
    [Cycle 1]: 0.00014041, [11]
        [c_1]: 2.247e-05
        [parameter_eliminate]: 2.12001e-06
        [updatestate_depend_eliminate]: 6.24001e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.968e-05
        [call_graph_tuple_transform]: 1.93e-05
        [tuple_list_get_item_eliminator]: 5.60001e-06
        [none_parameter_eliminate]: 1.72001e-06
        [renormalize]: 6.60017e-07
        [switch_simplify]: 6.30002e-06
[remove_dup_value]: 1.418e-05
[partial_unused_args_eliminate]: 2.83998e-06
[environ_conv]: 6.24999e-06
[add_recomputation]: 5.033e-05
[cse_after_recomputation]: 2.128e-05, [1]
    [Cycle 1]: 1.576e-05, [1]
        [cse]: 9.81e-06
[auto_monad_reorder]: 1.459e-05
[get_jit_bprop_graph]: 1.54998e-06
[rewriter_after_jit_bprop_graph]: 6.63e-06
[opt_after_jit_grad]: 0.00047209
[symbol_engine_optimizer]: 7.809e-05, [1]
    [Cycle 1]: 7.127e-05, [6]
        [build]: 4.28001e-06
        [elim_shapecalc]: 7.99997e-06
        [elim_not_effective]: 1.48e-05
        [opt_reshape]: 6.17999e-06
        [fold_const_symbol]: 9.14998e-06
        [renormalize]: 5.10016e-07
[validate]: 3.987e-05
Sums
    bootstrap                            :     0.000296s :  2.16%
    type_inference                       :     0.008720s : 63.56%
    event_method                         :     0.000027s :  0.19%
    auto_monad                           :     0.000090s :  0.65%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000069s :  0.50%
    rewriter_before_opt_a                :     0.000096s :  0.70%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000132s :  0.96%
    jit_opt_a.loop_unroll                :     0.000042s :  0.30%
    jit_opt_a.a_1                        :     0.000844s :  6.15%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.001111s :  8.10%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.14%
    jit_opt_a.cse                        :     0.000039s :  0.28%
    jit_opt_a.replace_applicator         :     0.000019s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.08%
    rewriter_after_opt_a                 :     0.000034s :  0.25%
    convert_after_rewriter               :     0.000008s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000655s :  4.77%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000478s :  3.48%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000050s :  0.37%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.05%
    opt_after_jit_grad                   :     0.000472s :  3.44%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.29%


Time group info:
------[substitution.]   0.000314    26
 0.61% :     0.000002s :      2: substitution.elim_not_effective
 0.39% :     0.000001s :      2: substitution.fold_const_symbol
 1.78% :     0.000006s :      3: substitution.graph_param_transform
86.40% :     0.000271s :      6: substitution.inline
 1.19% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.53% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.20% :     0.000004s :      2: substitution.replace_old_param
 2.49% :     0.000008s :      1: substitution.switch_simplify
 4.41% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008638     2
78.57% :     0.006787s :      1: type_inference.infer
21.43% :     0.001851s :      1: type_inference.specialize
------[replace.]   0.000151     9
49.50% :     0.000075s :      6: replace.inline
25.54% :     0.000039s :      1: replace.switch_simplify
24.97% :     0.000038s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000287     9
93.19% :     0.000267s :      6: match.inline
 2.44% :     0.000007s :      1: match.switch_simplify
 4.37% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   801
 1.28% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.99% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000002s :     12: predicate.addn_check_dump
 1.50% :     0.000002s :     12: predicate.addn_zero_filter
 2.19% :     0.000003s :     12: predicate.arithmetic_simplify
 1.47% :     0.000002s :     12: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     12: predicate.compare_switch_simplify
 1.52% :     0.000002s :     12: predicate.depend_value_elim
 1.10% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.24% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.42% :     0.000002s :     12: predicate.environ_get_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000001s :      3: predicate.graph_param_transform
 5.35% :     0.000007s :     26: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.04% :     0.000001s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     17: predicate.load_eliminater
 1.10% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.31% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.45% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     12: predicate.merge_addn
 1.07% :     0.000001s :     12: predicate.minmaximum_grad
 1.31% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     17: predicate.partial_eliminate
 1.16% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.90% :     0.000003s :     12: predicate.reduce_eliminate
 1.51% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     20: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.20% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     12: predicate.reshape_eliminate
 1.23% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     12: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.99% :     0.000004s :     20: predicate.switch_defer_inline
 2.24% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.33% :     0.000011s :     65: predicate.switch_simplify
 1.27% :     0.000002s :     12: predicate.tile_eliminate
 1.42% :     0.000002s :     12: predicate.transpose_eliminate
 1.40% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.18% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.56% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     12: predicate.value_based_eliminate
 0.37% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001199    15
45.08% :     0.000540s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 8.62% :     0.000103s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.30% :     0.000555s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020217    72
 0.26% :     0.000053s :      1: add_recomputation
 0.46% :     0.000094s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 1.51% :     0.000305s :      1: bootstrap
 0.13% :     0.000027s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.15% :     0.000030s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000010s :      1: graph_reusing
31.27% :     0.006321s :      1: jit_opt_a
 0.75% :     0.000152s :      1: jit_opt_after_cconv
 0.27% :     0.000054s :      1: jit_opt_b
 2.40% :     0.000486s :      1: loop_unroll
 3.28% :     0.000664s :      1: mutable_eliminate
 5.62% :     0.001135s :     26: opt.transform.jit_opt_a
 0.24% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.37% :     0.000480s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.35% :     0.000071s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 3.22% :     0.000650s :      1: renormalize.infer
 2.24% :     0.000452s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000037s :      1: rewriter_after_opt_a
 0.49% :     0.000099s :      1: rewriter_before_opt_a
 0.40% :     0.000081s :      1: symbol_engine_optimizer
43.22% :     0.008737s :      1: type_inference


TotalTime = 0.0212924, [33]
[bootstrap]: 0.00032831
[type_inference]: 0.00553907
[event_method]: 1.989e-05
[auto_monad]: 6.785e-05
[graph_reusing]: 6.11e-06
[pre_auto_parallel]: 2.94999e-06
[py_interpret_to_execute]: 3.939e-05
[rewriter_before_opt_a]: 7.423e-05
[expand_dump_flag]: 3.25e-06
[jit_opt_a]: 0.00601548, [2]
    [Cycle 1]: 0.00212228, [27]
        [switch_simplify]: 5.81e-05
        [loop_unroll]: 3.026e-05
        [a_1]: 0.00083529
        [with_stream_mark]: 2.453e-05
        [recompute_prepare]: 9.11998e-06
        [updatestate_depend_eliminate]: 4.67998e-06
        [updatestate_assign_eliminate]: 3.46001e-06
        [updatestate_loads_eliminate]: 2.58003e-06
        [parameter_eliminate]: 2.08002e-06
        [specialize_transform]: 6.62002e-06
        [updatestate_useless_node_eliminater]: 5.42999e-06
        [accelerated_algorithm]: 6.12999e-06
        [meta_shard_fg_expand]: 2.71999e-06
        [get_grad_eliminate_]: 5.52999e-06
        [merge_forward]: 3.75e-06
        [cell_reuse_recompute_pass]: 1.77001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.497e-05
        [j_node_and_user_rematch]: 1.038e-05
        [meta_fg_expand]: 2.40997e-06
        [replace_old_param]: 9.77999e-06
        [inline_without_move]: 5.82001e-06
        [renormalize]: 0.0008254
        [add_forward_monad_depend]: 6.02999e-06
        [auto_monad_grad]: 2.31998e-06
        [auto_monad_eliminator]: 1.612e-05
        [cse]: 2.678e-05
        [replace_applicator]: 1.504e-05
    [Cycle 2]: 0.00033562, [27]
        [switch_simplify]: 5.96e-06
        [loop_unroll]: 5.02e-06
        [a_1]: 9.431e-05
        [with_stream_mark]: 1.053e-05
        [recompute_prepare]: 4.92999e-06
        [updatestate_depend_eliminate]: 2.76e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.06003e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 5.13002e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 5.27999e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.50999e-06
        [merge_forward]: 2.29999e-06
        [cell_reuse_recompute_pass]: 2.01998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.314e-05
        [j_node_and_user_rematch]: 8.56002e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 7.88999e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.60001e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 6.35002e-06
        [cse]: 1.143e-05
        [replace_applicator]: 5.32999e-06
[py_interpret_to_execute_after_opt_a]: 1.218e-05
[rewriter_after_opt_a]: 3.411e-05
[convert_after_rewriter]: 7.01001e-06
[order_py_execute_after_rewriter]: 5.13002e-06
[mutable_eliminate]: 0.00060816
[jit_opt_b]: 5.303e-05, [1]
    [Cycle 1]: 4.527e-05, [2]
        [frontend_op_eliminate]: 1.747e-05
        [inline_after_opt_a]: 1.593e-05
[cconv]: 2.879e-05
[loop_unroll]: 0.00043939
[jit_opt_after_cconv]: 0.00014738, [1]
    [Cycle 1]: 0.00014067, [11]
        [c_1]: 2.17e-05
        [parameter_eliminate]: 2.96999e-06
        [updatestate_depend_eliminate]: 6.19001e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [cse]: 2.161e-05
        [call_graph_tuple_transform]: 2.019e-05
        [tuple_list_get_item_eliminator]: 5.54998e-06
        [none_parameter_eliminate]: 1.40999e-06
        [renormalize]: 8.10018e-07
        [switch_simplify]: 5.61e-06
[remove_dup_value]: 1.361e-05
[partial_unused_args_eliminate]: 2.67001e-06
[environ_conv]: 4.95999e-06
[add_recomputation]: 5.027e-05
[cse_after_recomputation]: 2.027e-05, [1]
    [Cycle 1]: 1.516e-05, [1]
        [cse]: 9.19e-06
[auto_monad_reorder]: 1.541e-05
[get_jit_bprop_graph]: 2.01e-06
[rewriter_after_jit_bprop_graph]: 5.28002e-06
[opt_after_jit_grad]: 0.00048861
[symbol_engine_optimizer]: 9.292e-05, [1]
    [Cycle 1]: 8.511e-05, [6]
        [build]: 5.13002e-06
        [elim_shapecalc]: 2.162e-05
        [elim_not_effective]: 1.389e-05
        [opt_reshape]: 5.49998e-06
        [fold_const_symbol]: 8.74e-06
        [renormalize]: 4.19997e-07
[validate]: 3.896e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00687606
[execute]: 7.85e-06
Sums
    bootstrap                            :     0.000328s :  1.92%
    type_inference                       :     0.005539s : 32.46%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000068s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.23%
    rewriter_before_opt_a                :     0.000074s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.38%
    jit_opt_a.loop_unroll                :     0.000035s :  0.21%
    jit_opt_a.a_1                        :     0.000930s :  5.45%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000825s :  4.84%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.13%
    jit_opt_a.cse                        :     0.000038s :  0.22%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000608s :  3.56%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000029s :  0.17%
    loop_unroll                          :     0.000439s :  2.57%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000489s :  2.86%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000022s :  0.13%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000039s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006876s : 40.30%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000228    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.49% :     0.000006s :      3: substitution.graph_param_transform
83.66% :     0.000191s :      4: substitution.inline
 1.76% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.16% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.38% :     0.000005s :      2: substitution.replace_old_param
 6.10% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005467     2
82.59% :     0.004516s :      1: type_inference.infer
17.41% :     0.000952s :      1: type_inference.specialize
------[replace.]   0.000082     6
57.99% :     0.000047s :      4: replace.inline
42.01% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     6
93.85% :     0.000188s :      4: match.inline
 6.15% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 3.07% :     0.000004s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.56% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.15% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.71% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.62% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.48% :     0.000007s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.16% :     0.000003s :     15: predicate.load_eliminater
 1.89% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.76% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.10% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 2.35% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.42% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.40% :     0.000002s :     10: predicate.reduce_eliminate
 1.69% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.22% :     0.000002s :     10: predicate.reshape_eliminate
 1.10% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.09% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.70% :     0.000003s :     16: predicate.switch_defer_inline
 1.88% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.43% :     0.000010s :     53: predicate.switch_simplify
 1.50% :     0.000002s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.30% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.46% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.09% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000725    12
50.21% :     0.000364s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.57% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.21% :     0.000321s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023387    76
 0.23% :     0.000053s :      1: add_recomputation
 0.36% :     0.000084s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.49% :     0.000348s :      1: bootstrap
 0.13% :     0.000031s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.74% :     0.006020s :      1: jit_opt_a
 0.64% :     0.000150s :      1: jit_opt_after_cconv
 0.24% :     0.000056s :      1: jit_opt_b
 1.92% :     0.000448s :      1: loop_unroll
 2.64% :     0.000619s :      1: mutable_eliminate
 4.93% :     0.001153s :     26: opt.transform.jit_opt_a
 0.21% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000046s :      4: opt.transform.symbol_engine_opt
 2.13% :     0.000498s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000042s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.90% :     0.000444s :      1: renormalize.infer
 1.60% :     0.000373s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000037s :      1: rewriter_after_opt_a
 0.33% :     0.000077s :      1: rewriter_before_opt_a
 0.41% :     0.000096s :      1: symbol_engine_optimizer
29.48% :     0.006894s :      1: task_emit
23.76% :     0.005556s :      1: type_inference
 0.26% :     0.000061s :      1: validate


TotalTime = 0.0223216, [33]
[bootstrap]: 0.00033905
[type_inference]: 0.00564729
[event_method]: 1.93e-05
[auto_monad]: 6.742e-05
[graph_reusing]: 6.21998e-06
[pre_auto_parallel]: 3.01999e-06
[py_interpret_to_execute]: 4.492e-05
[rewriter_before_opt_a]: 7.641e-05
[expand_dump_flag]: 3.29001e-06
[jit_opt_a]: 0.00647408, [2]
    [Cycle 1]: 0.00192596, [27]
        [switch_simplify]: 6.74e-05
        [loop_unroll]: 2.953e-05
        [a_1]: 0.00062093
        [with_stream_mark]: 2.296e-05
        [recompute_prepare]: 8.58001e-06
        [updatestate_depend_eliminate]: 3.88001e-06
        [updatestate_assign_eliminate]: 3.73001e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 2.02999e-06
        [specialize_transform]: 5.93002e-06
        [updatestate_useless_node_eliminater]: 4.81002e-06
        [accelerated_algorithm]: 5.44998e-06
        [meta_shard_fg_expand]: 2.53e-06
        [get_grad_eliminate_]: 5.78997e-06
        [merge_forward]: 3.65e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.496e-05
        [j_node_and_user_rematch]: 8.92999e-06
        [meta_fg_expand]: 2.51998e-06
        [replace_old_param]: 9.74e-06
        [inline_without_move]: 5.61998e-06
        [renormalize]: 0.00083372
        [add_forward_monad_depend]: 5.64e-06
        [auto_monad_grad]: 2.52001e-06
        [auto_monad_eliminator]: 1.511e-05
        [cse]: 3.34e-05
        [replace_applicator]: 1.544e-05
    [Cycle 2]: 0.0003281, [27]
        [switch_simplify]: 5.83002e-06
        [loop_unroll]: 4.96997e-06
        [a_1]: 9.273e-05
        [with_stream_mark]: 1.089e-05
        [recompute_prepare]: 5.05001e-06
        [updatestate_depend_eliminate]: 2.70997e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 2.69999e-06
        [parameter_eliminate]: 1.07998e-06
        [specialize_transform]: 4.78001e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 5.05001e-06
        [meta_shard_fg_expand]: 1.59998e-06
        [get_grad_eliminate_]: 5.05999e-06
        [merge_forward]: 2.89999e-06
        [cell_reuse_recompute_pass]: 1.86998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.321e-05
        [j_node_and_user_rematch]: 7.9e-06
        [meta_fg_expand]: 1.56002e-06
        [replace_old_param]: 7.65998e-06
        [inline_without_move]: 4.55999e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.11997e-06
        [auto_monad_grad]: 9.30013e-07
        [auto_monad_eliminator]: 5.81e-06
        [cse]: 1.081e-05
        [replace_applicator]: 4.97e-06
[py_interpret_to_execute_after_opt_a]: 1.202e-05
[rewriter_after_opt_a]: 4.026e-05
[convert_after_rewriter]: 7.32002e-06
[order_py_execute_after_rewriter]: 5.25999e-06
[mutable_eliminate]: 0.00082552
[jit_opt_b]: 5.597e-05, [1]
    [Cycle 1]: 4.76e-05, [2]
        [frontend_op_eliminate]: 1.973e-05
        [inline_after_opt_a]: 1.589e-05
[cconv]: 2.758e-05
[loop_unroll]: 0.00071492
[jit_opt_after_cconv]: 0.00015948, [1]
    [Cycle 1]: 0.00015196, [11]
        [c_1]: 2.455e-05
        [parameter_eliminate]: 3.23e-06
        [updatestate_depend_eliminate]: 7.00998e-06
        [updatestate_assign_eliminate]: 2.73998e-06
        [updatestate_loads_eliminate]: 2.77002e-06
        [cse]: 2.605e-05
        [call_graph_tuple_transform]: 2.08e-05
        [tuple_list_get_item_eliminator]: 5.29998e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 1.02e-06
        [switch_simplify]: 6.18998e-06
[remove_dup_value]: 1.584e-05
[partial_unused_args_eliminate]: 2.04999e-06
[environ_conv]: 6.41e-06
[add_recomputation]: 5.367e-05
[cse_after_recomputation]: 2.169e-05, [1]
    [Cycle 1]: 1.604e-05, [1]
        [cse]: 1.028e-05
[auto_monad_reorder]: 1.568e-05
[get_jit_bprop_graph]: 2.63998e-06
[rewriter_after_jit_bprop_graph]: 7.66999e-06
[opt_after_jit_grad]: 0.00053786
[symbol_engine_optimizer]: 8.078e-05, [1]
    [Cycle 1]: 7.247e-05, [6]
        [build]: 3.43e-06
        [elim_shapecalc]: 8.54e-06
        [elim_not_effective]: 1.499e-05
        [opt_reshape]: 6.53e-06
        [fold_const_symbol]: 9.32001e-06
        [renormalize]: 8.2e-07
[validate]: 4.506e-05
[backend_pass]: 8.99978e-07
[task_emit]: 0.00675417
[execute]: 9.47999e-06
Sums
    bootstrap                            :     0.000339s :  1.94%
    type_inference                       :     0.005647s : 32.39%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000067s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000045s :  0.26%
    rewriter_before_opt_a                :     0.000076s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000073s :  0.42%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000714s :  4.09%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000834s :  4.78%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.12%
    jit_opt_a.cse                        :     0.000044s :  0.25%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000040s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000826s :  4.74%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000028s :  0.16%
    loop_unroll                          :     0.000715s :  4.10%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000026s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000054s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000538s :  3.09%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006754s : 38.74%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000262    23
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.79% :     0.000002s :      2: substitution.fold_const_symbol
 2.13% :     0.000006s :      3: substitution.graph_param_transform
85.75% :     0.000224s :      4: substitution.inline
 1.18% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.13% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.02% :     0.000005s :      2: substitution.replace_old_param
 5.14% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005564     2
81.00% :     0.004507s :      1: type_inference.infer
19.00% :     0.001057s :      1: type_inference.specialize
------[replace.]   0.000077     6
66.99% :     0.000052s :      4: replace.inline
33.01% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000234     6
94.73% :     0.000221s :      4: match.inline
 5.27% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.39% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.84% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.50% :     0.000002s :     10: predicate.addn_zero_filter
 2.32% :     0.000003s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.82% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.34% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.43% :     0.000002s :     10: predicate.environ_get_eliminate
 1.19% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.11% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 5.45% :     0.000007s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.96% :     0.000001s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.03% :     0.000003s :     15: predicate.load_eliminater
 1.67% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.79% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.30% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 1.84% :     0.000002s :      3: predicate.mutable_eliminate
 0.38% :     0.000000s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 1.09% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.41% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.79% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.53% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.46% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.09% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 1.96% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.29% :     0.000009s :     53: predicate.switch_simplify
 1.37% :     0.000002s :     10: predicate.tile_eliminate
 1.24% :     0.000002s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.04% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.77% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.28% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.87% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.58% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000744    12
48.20% :     0.000359s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.17% :     0.000046s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.63% :     0.000340s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024210    76
 0.23% :     0.000057s :      1: add_recomputation
 0.29% :     0.000071s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.48% :     0.000358s :      1: bootstrap
 0.13% :     0.000030s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.76% :     0.006479s :      1: jit_opt_a
 0.67% :     0.000162s :      1: jit_opt_after_cconv
 0.24% :     0.000059s :      1: jit_opt_b
 3.00% :     0.000726s :      1: loop_unroll
 3.45% :     0.000836s :      1: mutable_eliminate
 3.88% :     0.000940s :     26: opt.transform.jit_opt_a
 0.22% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000029s :      4: opt.transform.jit_opt_b
 0.06% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.26% :     0.000548s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000048s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 1.83% :     0.000442s :      1: renormalize.infer
 1.58% :     0.000383s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000043s :      1: rewriter_after_opt_a
 0.33% :     0.000079s :      1: rewriter_before_opt_a
 0.34% :     0.000083s :      1: symbol_engine_optimizer
27.98% :     0.006775s :      1: task_emit
23.40% :     0.005666s :      1: type_inference
 0.27% :     0.000066s :      1: validate


TotalTime = 0.0194713, [33]
[bootstrap]: 0.00033782
[type_inference]: 0.004804
[event_method]: 1.756e-05
[auto_monad]: 6.571e-05
[graph_reusing]: 5.68002e-06
[pre_auto_parallel]: 2.39001e-06
[py_interpret_to_execute]: 3.259e-05
[rewriter_before_opt_a]: 7.349e-05
[expand_dump_flag]: 3.41001e-06
[jit_opt_a]: 0.00550069, [2]
    [Cycle 1]: 0.00159975, [27]
        [switch_simplify]: 5.548e-05
        [loop_unroll]: 2.882e-05
        [a_1]: 0.00052653
        [with_stream_mark]: 2.076e-05
        [recompute_prepare]: 8.37e-06
        [updatestate_depend_eliminate]: 3.56999e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [parameter_eliminate]: 2.14999e-06
        [specialize_transform]: 5.56e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 5.60001e-06
        [meta_shard_fg_expand]: 2.00002e-06
        [get_grad_eliminate_]: 4.97999e-06
        [merge_forward]: 4.82e-06
        [cell_reuse_recompute_pass]: 1.45001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.508e-05
        [j_node_and_user_rematch]: 8.71002e-06
        [meta_fg_expand]: 2.16e-06
        [replace_old_param]: 9.46e-06
        [inline_without_move]: 5.04e-06
        [renormalize]: 0.00063695
        [add_forward_monad_depend]: 5.87999e-06
        [auto_monad_grad]: 2.26e-06
        [auto_monad_eliminator]: 1.388e-05
        [cse]: 2.336e-05
        [replace_applicator]: 1.3e-05
    [Cycle 2]: 0.00033839, [27]
        [switch_simplify]: 5.46e-06
        [loop_unroll]: 5.10999e-06
        [a_1]: 8.701e-05
        [with_stream_mark]: 9.91e-06
        [recompute_prepare]: 5.24e-06
        [updatestate_depend_eliminate]: 3.43e-06
        [updatestate_assign_eliminate]: 2.01998e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 4.79998e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 4.70001e-06
        [meta_shard_fg_expand]: 1.19e-06
        [get_grad_eliminate_]: 4.51002e-06
        [merge_forward]: 2.70997e-06
        [cell_reuse_recompute_pass]: 2.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.559e-05
        [j_node_and_user_rematch]: 8.45001e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 8.86997e-06
        [inline_without_move]: 5.47999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 8.80013e-07
        [auto_monad_eliminator]: 8.1e-06
        [cse]: 1.383e-05
        [replace_applicator]: 6.17001e-06
[py_interpret_to_execute_after_opt_a]: 1.299e-05
[rewriter_after_opt_a]: 3.404e-05
[convert_after_rewriter]: 6.23e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.0005827
[jit_opt_b]: 4.92e-05, [1]
    [Cycle 1]: 4.106e-05, [2]
        [frontend_op_eliminate]: 1.56e-05
        [inline_after_opt_a]: 1.488e-05
[cconv]: 2.721e-05
[loop_unroll]: 0.00047798
[jit_opt_after_cconv]: 0.00014064, [1]
    [Cycle 1]: 0.00013416, [11]
        [c_1]: 2.006e-05
        [parameter_eliminate]: 3.33e-06
        [updatestate_depend_eliminate]: 6.17001e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.26998e-06
        [cse]: 1.918e-05
        [call_graph_tuple_transform]: 1.884e-05
        [tuple_list_get_item_eliminator]: 5.51e-06
        [none_parameter_eliminate]: 1.36002e-06
        [renormalize]: 5.60016e-07
        [switch_simplify]: 5.12e-06
[remove_dup_value]: 1.408e-05
[partial_unused_args_eliminate]: 2.06e-06
[environ_conv]: 5.13002e-06
[add_recomputation]: 4.775e-05
[cse_after_recomputation]: 2.102e-05, [1]
    [Cycle 1]: 1.512e-05, [1]
        [cse]: 8.98002e-06
[auto_monad_reorder]: 1.53e-05
[get_jit_bprop_graph]: 1.78002e-06
[rewriter_after_jit_bprop_graph]: 5.31002e-06
[opt_after_jit_grad]: 0.00051863
[symbol_engine_optimizer]: 7.641e-05, [1]
    [Cycle 1]: 6.825e-05, [6]
        [build]: 4.08001e-06
        [elim_shapecalc]: 8.09002e-06
        [elim_not_effective]: 1.376e-05
        [opt_reshape]: 5.53002e-06
        [fold_const_symbol]: 8.17e-06
        [renormalize]: 3.10014e-07
[validate]: 3.715e-05
[backend_pass]: 8.60018e-07
[task_emit]: 0.00634776
[execute]: 8.60001e-06
Sums
    bootstrap                            :     0.000338s :  2.21%
    type_inference                       :     0.004804s : 31.42%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000066s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.21%
    rewriter_before_opt_a                :     0.000073s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000614s :  4.01%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000637s :  4.17%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.14%
    jit_opt_a.cse                        :     0.000037s :  0.24%
    jit_opt_a.replace_applicator         :     0.000019s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000034s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000583s :  3.81%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000027s :  0.18%
    loop_unroll                          :     0.000478s :  3.13%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000048s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000519s :  3.39%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006348s : 41.52%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000211    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.59% :     0.000005s :      3: substitution.graph_param_transform
83.92% :     0.000177s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.30% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.47% :     0.000005s :      2: substitution.replace_old_param
 5.54% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004739     2
82.27% :     0.003899s :      1: type_inference.infer
17.73% :     0.000841s :      1: type_inference.specialize
------[replace.]   0.000064     6
65.54% :     0.000042s :      4: replace.inline
34.46% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000184     6
94.26% :     0.000174s :      4: match.inline
 5.74% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.29% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.84% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000001s :     10: predicate.addn_zero_filter
 1.81% :     0.000002s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.29% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.84% :     0.000007s :     22: predicate.inline
 1.05% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.60% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.30% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.94% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.23% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     10: predicate.reshape_eliminate
 1.29% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.46% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.22% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.96% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.32% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.85% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000564    12
46.79% :     0.000264s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.87% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.34% :     0.000256s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021049    76
 0.24% :     0.000051s :      1: add_recomputation
 0.33% :     0.000069s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.66% :     0.000350s :      1: bootstrap
 0.14% :     0.000030s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.14% :     0.005503s :      1: jit_opt_a
 0.68% :     0.000143s :      1: jit_opt_after_cconv
 0.25% :     0.000052s :      1: jit_opt_b
 2.31% :     0.000486s :      1: loop_unroll
 2.82% :     0.000593s :      1: mutable_eliminate
 3.93% :     0.000828s :     26: opt.transform.jit_opt_a
 0.22% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.51% :     0.000529s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000035s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.66% :     0.000349s :      1: renormalize.infer
 1.32% :     0.000279s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000037s :      1: rewriter_after_opt_a
 0.36% :     0.000076s :      1: rewriter_before_opt_a
 0.38% :     0.000079s :      1: symbol_engine_optimizer
30.21% :     0.006359s :      1: task_emit
22.89% :     0.004818s :      1: type_inference
 0.28% :     0.000059s :      1: validate


TotalTime = 0.0221025, [33]
[bootstrap]: 0.00034412
[type_inference]: 0.00542975
[event_method]: 2.053e-05
[auto_monad]: 6.77e-05
[graph_reusing]: 5.87001e-06
[pre_auto_parallel]: 3.26999e-06
[py_interpret_to_execute]: 4.211e-05
[rewriter_before_opt_a]: 7.786e-05
[expand_dump_flag]: 3.16999e-06
[jit_opt_a]: 0.00645872, [2]
    [Cycle 1]: 0.0020785, [27]
        [switch_simplify]: 0.00016363
        [loop_unroll]: 3.781e-05
        [a_1]: 0.00061069
        [with_stream_mark]: 2.18e-05
        [recompute_prepare]: 7.52002e-06
        [updatestate_depend_eliminate]: 3.90998e-06
        [updatestate_assign_eliminate]: 3.13998e-06
        [updatestate_loads_eliminate]: 2.75997e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 6.10002e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 5.57001e-06
        [meta_shard_fg_expand]: 2.56e-06
        [get_grad_eliminate_]: 5.47999e-06
        [merge_forward]: 4.03001e-06
        [cell_reuse_recompute_pass]: 1.26002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.45e-05
        [j_node_and_user_rematch]: 1.026e-05
        [meta_fg_expand]: 2.93003e-06
        [replace_old_param]: 8.52e-06
        [inline_without_move]: 5.50001e-06
        [renormalize]: 0.00089577
        [add_forward_monad_depend]: 6.54999e-06
        [auto_monad_grad]: 2.89001e-06
        [auto_monad_eliminator]: 1.367e-05
        [cse]: 3.238e-05
        [replace_applicator]: 1.339e-05
    [Cycle 2]: 0.00034027, [27]
        [switch_simplify]: 6.01e-06
        [loop_unroll]: 5.28002e-06
        [a_1]: 9.411e-05
        [with_stream_mark]: 1.059e-05
        [recompute_prepare]: 5.04003e-06
        [updatestate_depend_eliminate]: 2.96001e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 1.96998e-06
        [parameter_eliminate]: 1.27999e-06
        [specialize_transform]: 5.27999e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 1.60999e-06
        [get_grad_eliminate_]: 4.90999e-06
        [merge_forward]: 2.51998e-06
        [cell_reuse_recompute_pass]: 2.37001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.334e-05
        [j_node_and_user_rematch]: 9.08002e-06
        [meta_fg_expand]: 1.74998e-06
        [replace_old_param]: 8.23999e-06
        [inline_without_move]: 4.71002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 5.93002e-06
        [cse]: 1.175e-05
        [replace_applicator]: 5.20001e-06
[py_interpret_to_execute_after_opt_a]: 1.037e-05
[rewriter_after_opt_a]: 3.588e-05
[convert_after_rewriter]: 1.013e-05
[order_py_execute_after_rewriter]: 4.77e-06
[mutable_eliminate]: 0.00067985
[jit_opt_b]: 5.292e-05, [1]
    [Cycle 1]: 4.518e-05, [2]
        [frontend_op_eliminate]: 1.81e-05
        [inline_after_opt_a]: 1.562e-05
[cconv]: 2.769e-05
[loop_unroll]: 0.00050492
[jit_opt_after_cconv]: 0.00015806, [1]
    [Cycle 1]: 0.00015003, [11]
        [c_1]: 2.344e-05
        [parameter_eliminate]: 3.01999e-06
        [updatestate_depend_eliminate]: 5.99e-06
        [updatestate_assign_eliminate]: 2.68003e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [cse]: 2.393e-05
        [call_graph_tuple_transform]: 2.149e-05
        [tuple_list_get_item_eliminator]: 5.59e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 4.99975e-07
        [switch_simplify]: 5.76998e-06
[remove_dup_value]: 1.463e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 6.02001e-06
[add_recomputation]: 5.336e-05
[cse_after_recomputation]: 4.068e-05, [1]
    [Cycle 1]: 3.376e-05, [1]
        [cse]: 1.051e-05
[auto_monad_reorder]: 1.622e-05
[get_jit_bprop_graph]: 2.34001e-06
[rewriter_after_jit_bprop_graph]: 7.38e-06
[opt_after_jit_grad]: 0.0005391
[symbol_engine_optimizer]: 8.585e-05, [1]
    [Cycle 1]: 7.691e-05, [6]
        [build]: 4.77e-06
        [elim_shapecalc]: 8.47e-06
        [elim_not_effective]: 1.571e-05
        [opt_reshape]: 6.81999e-06
        [fold_const_symbol]: 1.065e-05
        [renormalize]: 4.50003e-07
[validate]: 4.277e-05
[backend_pass]: 1.25001e-06
[task_emit]: 0.00711259
[execute]: 9.07001e-06
Sums
    bootstrap                            :     0.000344s :  1.98%
    type_inference                       :     0.005430s : 31.26%
    event_method                         :     0.000021s :  0.12%
    auto_monad                           :     0.000068s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000042s :  0.24%
    rewriter_before_opt_a                :     0.000078s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000170s :  0.98%
    jit_opt_a.loop_unroll                :     0.000043s :  0.25%
    jit_opt_a.a_1                        :     0.000705s :  4.06%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000896s :  5.16%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.11%
    jit_opt_a.cse                        :     0.000044s :  0.25%
    jit_opt_a.replace_applicator         :     0.000019s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000036s :  0.21%
    convert_after_rewriter               :     0.000010s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000680s :  3.91%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000028s :  0.16%
    loop_unroll                          :     0.000505s :  2.91%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000053s :  0.31%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000539s :  3.10%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000043s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007113s : 40.94%
    execute                              :     0.000009s :  0.05%


TotalTime = 0.0174176, [30]
[bootstrap]: 0.00031729
[type_inference]: 0.00762449
[event_method]: 2.278e-05
[auto_monad]: 8.823e-05
[graph_reusing]: 7.49002e-06
[pre_auto_parallel]: 3.27002e-06
[py_interpret_to_execute]: 6.798e-05
[rewriter_before_opt_a]: 9.146e-05
[expand_dump_flag]: 3.38e-06
[jit_opt_a]: 0.00665285, [2]
    [Cycle 1]: 0.00250453, [27]
        [switch_simplify]: 0.00013516
        [loop_unroll]: 3.704e-05
        [a_1]: 0.00077936
        [with_stream_mark]: 2.063e-05
        [recompute_prepare]: 9.82999e-06
        [updatestate_depend_eliminate]: 4.66002e-06
        [updatestate_assign_eliminate]: 3.35003e-06
        [updatestate_loads_eliminate]: 2.48998e-06
        [parameter_eliminate]: 1.97999e-06
        [specialize_transform]: 7.06001e-06
        [updatestate_useless_node_eliminater]: 5.81e-06
        [accelerated_algorithm]: 7.19001e-06
        [meta_shard_fg_expand]: 3.02002e-06
        [get_grad_eliminate_]: 5.94999e-06
        [merge_forward]: 4.84998e-06
        [cell_reuse_recompute_pass]: 1.41002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.568e-05
        [j_node_and_user_rematch]: 1.002e-05
        [meta_fg_expand]: 2.48e-06
        [replace_old_param]: 9.89001e-06
        [inline_without_move]: 5.59e-06
        [renormalize]: 0.00114447
        [add_forward_monad_depend]: 8.02e-06
        [auto_monad_grad]: 2.71999e-06
        [auto_monad_eliminator]: 1.827e-05
        [cse]: 3.104e-05
        [replace_applicator]: 1.768e-05
    [Cycle 2]: 0.00036547, [27]
        [switch_simplify]: 6.21e-06
        [loop_unroll]: 5.15001e-06
        [a_1]: 9.412e-05
        [with_stream_mark]: 1.413e-05
        [recompute_prepare]: 6.81999e-06
        [updatestate_depend_eliminate]: 3.68999e-06
        [updatestate_assign_eliminate]: 2.72001e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 5.10001e-06
        [updatestate_useless_node_eliminater]: 5.52001e-06
        [accelerated_algorithm]: 6.02001e-06
        [meta_shard_fg_expand]: 1.72999e-06
        [get_grad_eliminate_]: 4.84998e-06
        [merge_forward]: 3.75003e-06
        [cell_reuse_recompute_pass]: 2.36998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.861e-05
        [j_node_and_user_rematch]: 9.21998e-06
        [meta_fg_expand]: 1.84998e-06
        [replace_old_param]: 7.92003e-06
        [inline_without_move]: 4.95999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.54001e-06
        [auto_monad_grad]: 1.49e-06
        [auto_monad_eliminator]: 6.86001e-06
        [cse]: 1.451e-05
        [replace_applicator]: 5.55001e-06
[py_interpret_to_execute_after_opt_a]: 1.443e-05
[rewriter_after_opt_a]: 3.786e-05
[convert_after_rewriter]: 7.35e-06
[order_py_execute_after_rewriter]: 4.97999e-06
[mutable_eliminate]: 0.00072262
[jit_opt_b]: 5.643e-05, [1]
    [Cycle 1]: 4.745e-05, [2]
        [frontend_op_eliminate]: 1.723e-05
        [inline_after_opt_a]: 1.562e-05
[cconv]: 2.732e-05
[loop_unroll]: 0.00050734
[jit_opt_after_cconv]: 0.00016438, [1]
    [Cycle 1]: 0.00015634, [11]
        [c_1]: 2.5e-05
        [parameter_eliminate]: 3.54002e-06
        [updatestate_depend_eliminate]: 7.06999e-06
        [updatestate_assign_eliminate]: 2.30002e-06
        [updatestate_loads_eliminate]: 2.35002e-06
        [cse]: 2.261e-05
        [call_graph_tuple_transform]: 2.253e-05
        [tuple_list_get_item_eliminator]: 6.17001e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 5.59987e-07
        [switch_simplify]: 6.56e-06
[remove_dup_value]: 1.659e-05
[partial_unused_args_eliminate]: 2.44999e-06
[environ_conv]: 6.12001e-06
[add_recomputation]: 5.454e-05
[cse_after_recomputation]: 2.49e-05, [1]
    [Cycle 1]: 1.929e-05, [1]
        [cse]: 1.104e-05
[auto_monad_reorder]: 1.537e-05
[get_jit_bprop_graph]: 2.22999e-06
[rewriter_after_jit_bprop_graph]: 7.76001e-06
[opt_after_jit_grad]: 0.00053028
[symbol_engine_optimizer]: 8.198e-05, [1]
    [Cycle 1]: 7.49e-05, [6]
        [build]: 4.93001e-06
        [elim_shapecalc]: 7.75e-06
        [elim_not_effective]: 1.623e-05
        [opt_reshape]: 5.87999e-06
        [fold_const_symbol]: 8.33999e-06
        [renormalize]: 5.19998e-07
[validate]: 6.217e-05
Sums
    bootstrap                            :     0.000317s :  2.45%
    type_inference                       :     0.007624s : 58.79%
    event_method                         :     0.000023s :  0.18%
    auto_monad                           :     0.000088s :  0.68%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000068s :  0.52%
    rewriter_before_opt_a                :     0.000091s :  0.71%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000141s :  1.09%
    jit_opt_a.loop_unroll                :     0.000042s :  0.33%
    jit_opt_a.a_1                        :     0.000873s :  6.74%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.08%
    jit_opt_a.merge_forward              :     0.000009s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.14%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.001145s :  8.83%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.19%
    jit_opt_a.cse                        :     0.000046s :  0.35%
    jit_opt_a.replace_applicator         :     0.000023s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.11%
    rewriter_after_opt_a                 :     0.000038s :  0.29%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000723s :  5.57%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.12%
    cconv                                :     0.000027s :  0.21%
    loop_unroll                          :     0.000507s :  3.91%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.05%
    remove_dup_value                     :     0.000017s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000055s :  0.42%
    cse_after_recomputation.cse          :     0.000011s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.06%
    opt_after_jit_grad                   :     0.000530s :  4.09%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000062s :  0.48%


Time group info:
------[substitution.]   0.000247    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.83% :     0.000002s :      2: substitution.fold_const_symbol
 2.58% :     0.000006s :      3: substitution.graph_param_transform
84.87% :     0.000210s :      4: substitution.inline
 1.52% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.83% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.54% :     0.000004s :      2: substitution.replace_old_param
 5.84% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005340     2
80.85% :     0.004317s :      1: type_inference.infer
19.15% :     0.001023s :      1: type_inference.specialize
------[replace.]   0.000082     6
66.74% :     0.000055s :      4: replace.inline
33.26% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000221     6
93.97% :     0.000207s :      4: match.inline
 6.03% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.36% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000002s :     10: predicate.addn_zero_filter
 2.64% :     0.000003s :     10: predicate.arithmetic_simplify
 1.36% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.19% :     0.000002s :     10: predicate.depend_value_elim
 0.98% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.15% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.03% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.62% :     0.000007s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.37% :     0.000002s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.87% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.99% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.19% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 1.88% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.51% :     0.000002s :     18: predicate.replace_applicator
 1.04% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.05% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.25% :     0.000002s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 1.99% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.58% :     0.000010s :     53: predicate.switch_simplify
 1.54% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.38% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000723    12
47.89% :     0.000346s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.19% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.92% :     0.000339s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024143    76
 0.23% :     0.000057s :      1: add_recomputation
 0.30% :     0.000072s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.50% :     0.000362s :      1: bootstrap
 0.13% :     0.000031s :      1: cconv
 0.05% :     0.000013s :      1: convert_after_rewriter
 0.18% :     0.000043s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000025s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.77% :     0.006463s :      1: jit_opt_a
 0.67% :     0.000161s :      1: jit_opt_after_cconv
 0.23% :     0.000056s :      1: jit_opt_b
 2.13% :     0.000514s :      1: loop_unroll
 2.86% :     0.000690s :      1: mutable_eliminate
 4.28% :     0.001034s :     26: opt.transform.jit_opt_a
 0.22% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000038s :      4: opt.transform.symbol_engine_opt
 2.27% :     0.000549s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000045s :      1: py_interpret_to_execute
 0.05% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.93% :     0.000467s :      1: renormalize.infer
 1.73% :     0.000419s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000039s :      1: rewriter_after_opt_a
 0.34% :     0.000081s :      1: rewriter_before_opt_a
 0.37% :     0.000088s :      1: symbol_engine_optimizer
29.53% :     0.007129s :      1: task_emit
22.57% :     0.005449s :      1: type_inference
 0.27% :     0.000065s :      1: validate


Time group info:
------[substitution.]   0.000312    26
 1.01% :     0.000003s :      2: substitution.elim_not_effective
 0.40% :     0.000001s :      2: substitution.fold_const_symbol
 1.99% :     0.000006s :      3: substitution.graph_param_transform
84.52% :     0.000264s :      6: substitution.inline
 1.28% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.70% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.57% :     0.000005s :      2: substitution.replace_old_param
 2.64% :     0.000008s :      1: substitution.switch_simplify
 4.89% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007551     2
81.17% :     0.006129s :      1: type_inference.infer
18.83% :     0.001422s :      1: type_inference.specialize
------[replace.]   0.000176     9
49.24% :     0.000086s :      6: replace.inline
23.96% :     0.000042s :      1: replace.switch_simplify
26.80% :     0.000047s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000281     9
92.51% :     0.000260s :      6: match.inline
 2.65% :     0.000007s :      1: match.switch_simplify
 4.84% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000147   801
 1.19% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.73% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     12: predicate.addn_check_dump
 1.27% :     0.000002s :     12: predicate.addn_zero_filter
 2.60% :     0.000004s :     12: predicate.arithmetic_simplify
 1.13% :     0.000002s :     12: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     12: predicate.compare_switch_simplify
 1.89% :     0.000003s :     12: predicate.depend_value_elim
 1.47% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.55% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      3: predicate.elim_not_effective
 0.44% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     12: predicate.environ_get_depend_swap
 2.25% :     0.000003s :     12: predicate.environ_get_eliminate
 1.06% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.58% :     0.000008s :     26: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :      6: predicate.less_batch_normalization
 1.30% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.54% :     0.000002s :     17: predicate.load_eliminater
 1.56% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.19% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.38% :     0.000004s :     15: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000002s :     12: predicate.merge_addn
 1.06% :     0.000002s :     12: predicate.minmaximum_grad
 1.74% :     0.000003s :      3: predicate.mutable_eliminate
 0.33% :     0.000000s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     17: predicate.partial_eliminate
 1.11% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.78% :     0.000003s :     12: predicate.reduce_eliminate
 1.33% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.49% :     0.000002s :     20: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.21% :     0.000002s :     12: predicate.reshape_eliminate
 1.10% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.66% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     12: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000002s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.72% :     0.000004s :     20: predicate.switch_defer_inline
 2.14% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.48% :     0.000012s :     65: predicate.switch_simplify
 1.23% :     0.000002s :     12: predicate.tile_eliminate
 1.15% :     0.000002s :     12: predicate.transpose_eliminate
 1.42% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.78% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.65% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.85% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001077    15
47.79% :     0.000515s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 8.25% :     0.000089s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.96% :     0.000473s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019785    72
 0.29% :     0.000058s :      1: add_recomputation
 0.47% :     0.000092s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 1.67% :     0.000331s :      1: bootstrap
 0.15% :     0.000030s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000028s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.13% :     0.000026s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000010s :      1: graph_reusing
33.65% :     0.006658s :      1: jit_opt_a
 0.85% :     0.000167s :      1: jit_opt_after_cconv
 0.30% :     0.000059s :      1: jit_opt_b
 2.62% :     0.000519s :      1: loop_unroll
 3.71% :     0.000735s :      1: mutable_eliminate
 6.02% :     0.001192s :     26: opt.transform.jit_opt_a
 0.28% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.72% :     0.000537s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.36% :     0.000071s :      1: py_interpret_to_execute
 0.09% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000019s :      1: remove_dup_value
 3.48% :     0.000689s :      1: renormalize.infer
 2.24% :     0.000443s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000043s :      1: rewriter_after_opt_a
 0.48% :     0.000096s :      1: rewriter_before_opt_a
 0.43% :     0.000084s :      1: symbol_engine_optimizer
38.59% :     0.007636s :      1: type_inference


TotalTime = 0.0222331, [33]
[bootstrap]: 0.00034974
[type_inference]: 0.00527804
[event_method]: 1.709e-05
[auto_monad]: 7.022e-05
[graph_reusing]: 6.41e-06
[pre_auto_parallel]: 2.26e-06
[py_interpret_to_execute]: 4.444e-05
[rewriter_before_opt_a]: 7.358e-05
[expand_dump_flag]: 3.53e-06
[jit_opt_a]: 0.00697243, [2]
    [Cycle 1]: 0.00207823, [27]
        [switch_simplify]: 6.119e-05
        [loop_unroll]: 2.976e-05
        [a_1]: 0.00061357
        [with_stream_mark]: 2.085e-05
        [recompute_prepare]: 7.38e-06
        [updatestate_depend_eliminate]: 2.95002e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 6.21e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 6.24999e-06
        [meta_shard_fg_expand]: 2.16e-06
        [get_grad_eliminate_]: 5.64998e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 1.01002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.284e-05
        [j_node_and_user_rematch]: 9.13002e-06
        [meta_fg_expand]: 2.48e-06
        [replace_old_param]: 9.92999e-06
        [inline_without_move]: 5.30999e-06
        [renormalize]: 0.00100124
        [add_forward_monad_depend]: 8.71002e-06
        [auto_monad_grad]: 2.83e-06
        [auto_monad_eliminator]: 2.031e-05
        [cse]: 3.029e-05
        [replace_applicator]: 1.797e-05
    [Cycle 2]: 0.00042544, [27]
        [switch_simplify]: 6.60002e-06
        [loop_unroll]: 5.16002e-06
        [a_1]: 0.00010441
        [with_stream_mark]: 1.526e-05
        [recompute_prepare]: 5.39e-06
        [updatestate_depend_eliminate]: 3.26001e-06
        [updatestate_assign_eliminate]: 2.73998e-06
        [updatestate_loads_eliminate]: 2.58003e-06
        [parameter_eliminate]: 2.07999e-06
        [specialize_transform]: 5.00999e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 7.30998e-06
        [meta_shard_fg_expand]: 2.62001e-06
        [get_grad_eliminate_]: 4.55001e-06
        [merge_forward]: 3.66999e-06
        [cell_reuse_recompute_pass]: 3.41999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.057e-05
        [j_node_and_user_rematch]: 1.127e-05
        [meta_fg_expand]: 2.09999e-06
        [replace_old_param]: 1.547e-05
        [inline_without_move]: 6.66e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.66e-06
        [auto_monad_grad]: 1.32e-06
        [auto_monad_eliminator]: 7.46001e-06
        [cse]: 1.567e-05
        [replace_applicator]: 6.69999e-06
[py_interpret_to_execute_after_opt_a]: 1.568e-05
[rewriter_after_opt_a]: 4.254e-05
[convert_after_rewriter]: 9.19e-06
[order_py_execute_after_rewriter]: 5.66e-06
[mutable_eliminate]: 0.00094954
[jit_opt_b]: 6.39e-05, [1]
    [Cycle 1]: 5.339e-05, [2]
        [frontend_op_eliminate]: 2.178e-05
        [inline_after_opt_a]: 1.847e-05
[cconv]: 3.719e-05
[loop_unroll]: 0.00063148
[jit_opt_after_cconv]: 0.00018059, [1]
    [Cycle 1]: 0.00017136, [11]
        [c_1]: 2.28e-05
        [parameter_eliminate]: 6.09999e-06
        [updatestate_depend_eliminate]: 8.08999e-06
        [updatestate_assign_eliminate]: 2.64999e-06
        [updatestate_loads_eliminate]: 2.69001e-06
        [cse]: 3.048e-05
        [call_graph_tuple_transform]: 2.846e-05
        [tuple_list_get_item_eliminator]: 6.66e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 1.00001e-06
        [switch_simplify]: 5.66e-06
[remove_dup_value]: 1.692e-05
[partial_unused_args_eliminate]: 2.77002e-06
[environ_conv]: 6.43e-06
[add_recomputation]: 5.434e-05
[cse_after_recomputation]: 2.462e-05, [1]
    [Cycle 1]: 1.72e-05, [1]
        [cse]: 1.003e-05
[auto_monad_reorder]: 1.555e-05
[get_jit_bprop_graph]: 2.59001e-06
[rewriter_after_jit_bprop_graph]: 1.126e-05
[opt_after_jit_grad]: 0.00065737
[symbol_engine_optimizer]: 8.775e-05, [1]
    [Cycle 1]: 7.75e-05, [6]
        [build]: 5.00001e-06
        [elim_shapecalc]: 9.17999e-06
        [elim_not_effective]: 1.567e-05
        [opt_reshape]: 6.17999e-06
        [fold_const_symbol]: 9.51e-06
        [renormalize]: 1.07e-06
[validate]: 4.776e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00629155
[execute]: 1.012e-05
Sums
    bootstrap                            :     0.000350s :  2.05%
    type_inference                       :     0.005278s : 30.99%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000070s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.26%
    rewriter_before_opt_a                :     0.000074s :  0.43%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.21%
    jit_opt_a.a_1                        :     0.000718s :  4.22%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000014s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000025s :  0.15%
    jit_opt_a.inline_without_move        :     0.000012s :  0.07%
    jit_opt_a.renormalize                :     0.001001s :  5.88%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.16%
    jit_opt_a.cse                        :     0.000046s :  0.27%
    jit_opt_a.replace_applicator         :     0.000025s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.09%
    rewriter_after_opt_a                 :     0.000043s :  0.25%
    convert_after_rewriter               :     0.000009s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000950s :  5.58%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.11%
    cconv                                :     0.000037s :  0.22%
    loop_unroll                          :     0.000631s :  3.71%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000030s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000054s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.07%
    opt_after_jit_grad                   :     0.000657s :  3.86%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000048s :  0.28%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006292s : 36.95%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000277    23
 0.97% :     0.000003s :      2: substitution.elim_not_effective
 0.52% :     0.000001s :      2: substitution.fold_const_symbol
 2.58% :     0.000007s :      3: substitution.graph_param_transform
83.48% :     0.000231s :      4: substitution.inline
 1.86% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.97% :     0.000005s :      4: substitution.remove_not_recompute_node
 4.20% :     0.000012s :      2: substitution.replace_old_param
 4.43% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005205     2
80.36% :     0.004183s :      1: type_inference.infer
19.64% :     0.001022s :      1: type_inference.specialize
------[replace.]   0.000079     6
68.79% :     0.000055s :      4: replace.inline
31.21% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000239     6
95.30% :     0.000228s :      4: match.inline
 4.70% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.62% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.50% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 2.14% :     0.000003s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.22% :     0.000002s :     10: predicate.depend_value_elim
 0.99% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.09% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.18% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.73% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.37% :     0.000002s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.30% :     0.000002s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.49% :     0.000007s :     22: predicate.inline
 1.30% :     0.000002s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.49% :     0.000002s :      6: predicate.less_batch_normalization
 1.26% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 2.11% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.76% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 2.14% :     0.000003s :      3: predicate.mutable_eliminate
 0.39% :     0.000000s :      3: predicate.opt_reshape
 2.04% :     0.000003s :     15: predicate.partial_eliminate
 1.32% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.68% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.35% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000001s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.79% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.50% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.21% :     0.000009s :     53: predicate.switch_simplify
 1.69% :     0.000002s :     10: predicate.tile_eliminate
 1.24% :     0.000002s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.42% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.96% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.91% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000720    12
45.67% :     0.000329s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.90% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.43% :     0.000334s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024339    76
 0.24% :     0.000058s :      1: add_recomputation
 0.30% :     0.000074s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.50% :     0.000366s :      1: bootstrap
 0.16% :     0.000040s :      1: cconv
 0.05% :     0.000012s :      1: convert_after_rewriter
 0.11% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
28.66% :     0.006977s :      1: jit_opt_a
 0.75% :     0.000183s :      1: jit_opt_after_cconv
 0.28% :     0.000067s :      1: jit_opt_b
 2.64% :     0.000643s :      1: loop_unroll
 3.97% :     0.000966s :      1: mutable_eliminate
 3.96% :     0.000965s :     26: opt.transform.jit_opt_a
 0.24% :     0.000060s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000032s :      4: opt.transform.jit_opt_b
 0.08% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000023s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.75% :     0.000670s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000048s :      1: py_interpret_to_execute
 0.08% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000021s :      1: remove_dup_value
 2.23% :     0.000542s :      1: renormalize.infer
 1.85% :     0.000450s :      1: renormalize.specialize
 0.06% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000047s :      1: rewriter_after_opt_a
 0.32% :     0.000077s :      1: rewriter_before_opt_a
 0.37% :     0.000091s :      1: symbol_engine_optimizer
25.93% :     0.006311s :      1: task_emit
21.75% :     0.005294s :      1: type_inference
 0.30% :     0.000074s :      1: validate


TotalTime = 0.0183254, [33]
[bootstrap]: 0.00030972
[type_inference]: 0.00475769
[event_method]: 1.692e-05
[auto_monad]: 6.705e-05
[graph_reusing]: 6.39001e-06
[pre_auto_parallel]: 2.59001e-06
[py_interpret_to_execute]: 2.993e-05
[rewriter_before_opt_a]: 6.822e-05
[expand_dump_flag]: 2.97002e-06
[jit_opt_a]: 0.00550348, [2]
    [Cycle 1]: 0.00159591, [27]
        [switch_simplify]: 5.161e-05
        [loop_unroll]: 3.948e-05
        [a_1]: 0.00051813
        [with_stream_mark]: 1.869e-05
        [recompute_prepare]: 6.92002e-06
        [updatestate_depend_eliminate]: 3.03003e-06
        [updatestate_assign_eliminate]: 3.15998e-06
        [updatestate_loads_eliminate]: 2.34999e-06
        [parameter_eliminate]: 1.76e-06
        [specialize_transform]: 5.69999e-06
        [updatestate_useless_node_eliminater]: 4.65999e-06
        [accelerated_algorithm]: 5.46998e-06
        [meta_shard_fg_expand]: 1.76998e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 4.16001e-06
        [cell_reuse_recompute_pass]: 1.69998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.376e-05
        [j_node_and_user_rematch]: 8.77e-06
        [meta_fg_expand]: 2.11e-06
        [replace_old_param]: 8.97e-06
        [inline_without_move]: 5.09998e-06
        [renormalize]: 0.00067295
        [add_forward_monad_depend]: 5.15001e-06
        [auto_monad_grad]: 2.19001e-06
        [auto_monad_eliminator]: 1.313e-05
        [cse]: 1.754e-05
        [replace_applicator]: 1.042e-05
    [Cycle 2]: 0.00031274, [27]
        [switch_simplify]: 6.21e-06
        [loop_unroll]: 4.57e-06
        [a_1]: 8.756e-05
        [with_stream_mark]: 9.57001e-06
        [recompute_prepare]: 4.59002e-06
        [updatestate_depend_eliminate]: 2.66999e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 4.84003e-06
        [updatestate_useless_node_eliminater]: 4.61002e-06
        [accelerated_algorithm]: 5.02999e-06
        [meta_shard_fg_expand]: 1.01997e-06
        [get_grad_eliminate_]: 4.22e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 2.54999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.223e-05
        [j_node_and_user_rematch]: 7.29001e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 7.01001e-06
        [inline_without_move]: 4.45999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 5.19e-06
        [cse]: 9.76998e-06
        [replace_applicator]: 5.10001e-06
[py_interpret_to_execute_after_opt_a]: 1.015e-05
[rewriter_after_opt_a]: 3.191e-05
[convert_after_rewriter]: 7.40998e-06
[order_py_execute_after_rewriter]: 4.90999e-06
[mutable_eliminate]: 0.00054414
[jit_opt_b]: 4.821e-05, [1]
    [Cycle 1]: 4.155e-05, [2]
        [frontend_op_eliminate]: 1.593e-05
        [inline_after_opt_a]: 1.393e-05
[cconv]: 2.486e-05
[loop_unroll]: 0.00042172
[jit_opt_after_cconv]: 0.0001333, [1]
    [Cycle 1]: 0.00012651, [11]
        [c_1]: 1.916e-05
        [parameter_eliminate]: 2.59001e-06
        [updatestate_depend_eliminate]: 5.06002e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 1.745e-05
        [call_graph_tuple_transform]: 1.769e-05
        [tuple_list_get_item_eliminator]: 5.04e-06
        [none_parameter_eliminate]: 1.38002e-06
        [renormalize]: 7.30011e-07
        [switch_simplify]: 5.11997e-06
[remove_dup_value]: 1.308e-05
[partial_unused_args_eliminate]: 2.19001e-06
[environ_conv]: 5.17e-06
[add_recomputation]: 4.356e-05
[cse_after_recomputation]: 2.005e-05, [1]
    [Cycle 1]: 1.499e-05, [1]
        [cse]: 8.54e-06
[auto_monad_reorder]: 1.392e-05
[get_jit_bprop_graph]: 1.96e-06
[rewriter_after_jit_bprop_graph]: 5.52001e-06
[opt_after_jit_grad]: 0.00046244
[symbol_engine_optimizer]: 7.277e-05, [1]
    [Cycle 1]: 6.588e-05, [6]
        [build]: 3.88001e-06
        [elim_shapecalc]: 7.45e-06
        [elim_not_effective]: 1.266e-05
        [opt_reshape]: 5.34e-06
        [fold_const_symbol]: 8.3e-06
        [renormalize]: 6.40022e-07
[validate]: 3.372e-05
[backend_pass]: 8.60018e-07
[task_emit]: 0.00547552
[execute]: 5.59e-06
Sums
    bootstrap                            :     0.000310s :  2.19%
    type_inference                       :     0.004758s : 33.60%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000067s :  0.47%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.21%
    rewriter_before_opt_a                :     0.000068s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.41%
    jit_opt_a.loop_unroll                :     0.000044s :  0.31%
    jit_opt_a.a_1                        :     0.000606s :  4.28%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000673s :  4.75%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000027s :  0.19%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000544s :  3.84%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.18%
    loop_unroll                          :     0.000422s :  2.98%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000044s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000462s :  3.27%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005476s : 38.67%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000202    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.61% :     0.000005s :      3: substitution.graph_param_transform
84.02% :     0.000170s :      4: substitution.inline
 1.50% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.31% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.02% :     0.000004s :      2: substitution.replace_old_param
 5.92% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004690     2
81.98% :     0.003845s :      1: type_inference.infer
18.02% :     0.000845s :      1: type_inference.specialize
------[replace.]   0.000061     6
67.01% :     0.000041s :      4: replace.inline
32.99% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000179     6
93.88% :     0.000168s :      4: match.inline
 6.12% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.62% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.53% :     0.000002s :     10: predicate.addn_zero_filter
 2.00% :     0.000002s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.19% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.11% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.51% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.41% :     0.000002s :      6: predicate.less_batch_normalization
 1.90% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.87% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.92% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.08% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.98% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000002s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000001s :     10: predicate.same_eliminate
 0.60% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.52% :     0.000002s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.53% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     16: predicate.switch_defer_inline
 2.39% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.94% :     0.000009s :     53: predicate.switch_simplify
 1.30% :     0.000001s :     10: predicate.tile_eliminate
 1.35% :     0.000002s :     10: predicate.transpose_eliminate
 1.27% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.54% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.83% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000571    12
47.76% :     0.000273s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.24% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.00% :     0.000257s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019917    76
 0.23% :     0.000046s :      1: add_recomputation
 0.35% :     0.000071s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.61% :     0.000321s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.04% :     0.000009s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
27.65% :     0.005506s :      1: jit_opt_a
 0.68% :     0.000136s :      1: jit_opt_after_cconv
 0.25% :     0.000051s :      1: jit_opt_b
 2.15% :     0.000429s :      1: loop_unroll
 2.78% :     0.000554s :      1: mutable_eliminate
 4.09% :     0.000815s :     26: opt.transform.jit_opt_a
 0.22% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.36% :     0.000470s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.94% :     0.000387s :      1: renormalize.infer
 1.40% :     0.000279s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000035s :      1: rewriter_after_opt_a
 0.36% :     0.000071s :      1: rewriter_before_opt_a
 0.38% :     0.000076s :      1: symbol_engine_optimizer
27.54% :     0.005485s :      1: task_emit
23.95% :     0.004770s :      1: type_inference
 0.27% :     0.000054s :      1: validate


TotalTime = 0.0152362, [30]
[bootstrap]: 0.0002835
[type_inference]: 0.00651231
[event_method]: 2.056e-05
[auto_monad]: 8.258e-05
[graph_reusing]: 7.04001e-06
[pre_auto_parallel]: 2.78e-06
[py_interpret_to_execute]: 5.534e-05
[rewriter_before_opt_a]: 8.838e-05
[expand_dump_flag]: 3.16999e-06
[jit_opt_a]: 0.0057695, [2]
    [Cycle 1]: 0.00211275, [27]
        [switch_simplify]: 0.00012953
        [loop_unroll]: 3.565e-05
        [a_1]: 0.00075952
        [with_stream_mark]: 1.749e-05
        [recompute_prepare]: 9.03002e-06
        [updatestate_depend_eliminate]: 3.48e-06
        [updatestate_assign_eliminate]: 3.06001e-06
        [updatestate_loads_eliminate]: 2.67001e-06
        [parameter_eliminate]: 2.16e-06
        [specialize_transform]: 6.86999e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 5.49998e-06
        [meta_shard_fg_expand]: 2.12999e-06
        [get_grad_eliminate_]: 5.42001e-06
        [merge_forward]: 3.81999e-06
        [cell_reuse_recompute_pass]: 1.76998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.537e-05
        [j_node_and_user_rematch]: 9.17001e-06
        [meta_fg_expand]: 3.04999e-06
        [replace_old_param]: 9.05999e-06
        [inline_without_move]: 5.94e-06
        [renormalize]: 0.00081525
        [add_forward_monad_depend]: 5.94999e-06
        [auto_monad_grad]: 2.32001e-06
        [auto_monad_eliminator]: 1.627e-05
        [cse]: 3.159e-05
        [replace_applicator]: 1.232e-05
    [Cycle 2]: 0.00033935, [27]
        [switch_simplify]: 6.07999e-06
        [loop_unroll]: 5.39e-06
        [a_1]: 9.654e-05
        [with_stream_mark]: 1.023e-05
        [recompute_prepare]: 6.23e-06
        [updatestate_depend_eliminate]: 2.94999e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 1.25999e-06
        [specialize_transform]: 5.52001e-06
        [updatestate_useless_node_eliminater]: 5.44e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.40001e-06
        [get_grad_eliminate_]: 4.73001e-06
        [merge_forward]: 4.12e-06
        [cell_reuse_recompute_pass]: 1.90001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.426e-05
        [j_node_and_user_rematch]: 8.79998e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 7.70998e-06
        [inline_without_move]: 4.95001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.60999e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 5.67999e-06
        [cse]: 1.252e-05
        [replace_applicator]: 5.07e-06
[py_interpret_to_execute_after_opt_a]: 1.148e-05
[rewriter_after_opt_a]: 3.522e-05
[convert_after_rewriter]: 7.97e-06
[order_py_execute_after_rewriter]: 5.64e-06
[mutable_eliminate]: 0.00070927
[jit_opt_b]: 7.078e-05, [1]
    [Cycle 1]: 6.086e-05, [2]
        [frontend_op_eliminate]: 1.708e-05
        [inline_after_opt_a]: 1.81e-05
[cconv]: 2.618e-05
[loop_unroll]: 0.00049607
[jit_opt_after_cconv]: 0.00015524, [1]
    [Cycle 1]: 0.00014836, [11]
        [c_1]: 2.398e-05
        [parameter_eliminate]: 2.61e-06
        [updatestate_depend_eliminate]: 6.54001e-06
        [updatestate_assign_eliminate]: 2.28998e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [cse]: 2.193e-05
        [call_graph_tuple_transform]: 2.266e-05
        [tuple_list_get_item_eliminator]: 5.21998e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 7.50006e-07
        [switch_simplify]: 6.37001e-06
[remove_dup_value]: 1.492e-05
[partial_unused_args_eliminate]: 3.13e-06
[environ_conv]: 5.19e-06
[add_recomputation]: 5.021e-05
[cse_after_recomputation]: 2.263e-05, [1]
    [Cycle 1]: 1.713e-05, [1]
        [cse]: 1.146e-05
[auto_monad_reorder]: 1.527e-05
[get_jit_bprop_graph]: 1.55001e-06
[rewriter_after_jit_bprop_graph]: 6.64999e-06
[opt_after_jit_grad]: 0.00047986
[symbol_engine_optimizer]: 7.951e-05, [1]
    [Cycle 1]: 7.248e-05, [6]
        [build]: 4.62998e-06
        [elim_shapecalc]: 9.15999e-06
        [elim_not_effective]: 1.249e-05
        [opt_reshape]: 6.19001e-06
        [fold_const_symbol]: 9.20999e-06
        [renormalize]: 4.90021e-07
[validate]: 3.901e-05
Sums
    bootstrap                            :     0.000283s :  2.51%
    type_inference                       :     0.006512s : 57.67%
    event_method                         :     0.000021s :  0.18%
    auto_monad                           :     0.000083s :  0.73%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000055s :  0.49%
    rewriter_before_opt_a                :     0.000088s :  0.78%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000136s :  1.20%
    jit_opt_a.loop_unroll                :     0.000041s :  0.36%
    jit_opt_a.a_1                        :     0.000856s :  7.58%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000008s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000815s :  7.22%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.19%
    jit_opt_a.cse                        :     0.000044s :  0.39%
    jit_opt_a.replace_applicator         :     0.000017s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.10%
    rewriter_after_opt_a                 :     0.000035s :  0.31%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000709s :  6.28%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.16%
    cconv                                :     0.000026s :  0.23%
    loop_unroll                          :     0.000496s :  4.39%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000015s :  0.13%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000050s :  0.44%
    cse_after_recomputation.cse          :     0.000011s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000480s :  4.25%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000039s :  0.35%


Time group info:
------[substitution.]   0.000320    26
 0.54% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000002s :      2: substitution.fold_const_symbol
 2.09% :     0.000007s :      3: substitution.graph_param_transform
85.56% :     0.000274s :      6: substitution.inline
 1.21% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.76% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.35% :     0.000004s :      2: substitution.replace_old_param
 2.84% :     0.000009s :      1: substitution.switch_simplify
 4.16% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006450     2
82.47% :     0.005319s :      1: type_inference.infer
17.53% :     0.001131s :      1: type_inference.specialize
------[replace.]   0.000158     9
49.35% :     0.000078s :      6: replace.inline
26.05% :     0.000041s :      1: replace.switch_simplify
24.61% :     0.000039s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000291     9
93.02% :     0.000270s :      6: match.inline
 2.81% :     0.000008s :      1: match.switch_simplify
 4.16% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000140   801
 1.23% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.12% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000002s :     12: predicate.addn_check_dump
 1.34% :     0.000002s :     12: predicate.addn_zero_filter
 2.31% :     0.000003s :     12: predicate.arithmetic_simplify
 1.26% :     0.000002s :     12: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000002s :     12: predicate.compare_switch_simplify
 1.68% :     0.000002s :     12: predicate.depend_value_elim
 1.15% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.60% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.42% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.08% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     12: predicate.environ_get_eliminate
 1.07% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.80% :     0.000001s :      6: predicate.get_grad_eliminate
 0.35% :     0.000000s :      3: predicate.graph_param_transform
 5.63% :     0.000008s :     26: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.02% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.81% :     0.000003s :     17: predicate.load_eliminater
 1.19% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.36% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.05% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     12: predicate.merge_addn
 1.26% :     0.000002s :     12: predicate.minmaximum_grad
 1.34% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     17: predicate.partial_eliminate
 1.40% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.83% :     0.000003s :     12: predicate.reduce_eliminate
 1.62% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.55% :     0.000002s :     20: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     12: predicate.reshape_eliminate
 1.23% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.56% :     0.000002s :     12: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 0.86% :     0.000001s :      6: predicate.specialize_transform
 1.64% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.63% :     0.000004s :     20: predicate.switch_defer_inline
 2.30% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.43% :     0.000012s :     65: predicate.switch_simplify
 1.37% :     0.000002s :     12: predicate.tile_eliminate
 1.18% :     0.000002s :     12: predicate.transpose_eliminate
 1.41% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.01% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.64% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000003s :     12: predicate.value_based_eliminate
 0.26% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000862    15
48.12% :     0.000415s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.28% :     0.000106s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.60% :     0.000341s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017263    72
 0.31% :     0.000053s :      1: add_recomputation
 0.50% :     0.000086s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 1.72% :     0.000297s :      1: bootstrap
 0.17% :     0.000029s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.14% :     0.000024s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000010s :      1: graph_reusing
33.45% :     0.005774s :      1: jit_opt_a
 0.92% :     0.000158s :      1: jit_opt_after_cconv
 0.42% :     0.000073s :      1: jit_opt_b
 2.92% :     0.000505s :      1: loop_unroll
 4.16% :     0.000718s :      1: mutable_eliminate
 6.68% :     0.001154s :     26: opt.transform.jit_opt_a
 0.32% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.82% :     0.000487s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.34% :     0.000058s :      1: py_interpret_to_execute
 0.08% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000018s :      1: remove_dup_value
 2.56% :     0.000442s :      1: renormalize.infer
 2.11% :     0.000365s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000039s :      1: rewriter_after_opt_a
 0.53% :     0.000091s :      1: rewriter_before_opt_a
 0.47% :     0.000082s :      1: symbol_engine_optimizer
37.80% :     0.006525s :      1: type_inference


TotalTime = 0.0212509, [33]
[bootstrap]: 0.00036171
[type_inference]: 0.00578165
[event_method]: 2.206e-05
[auto_monad]: 7.111e-05
[graph_reusing]: 6.21e-06
[pre_auto_parallel]: 2.64999e-06
[py_interpret_to_execute]: 4.881e-05
[rewriter_before_opt_a]: 7.954e-05
[expand_dump_flag]: 3.09999e-06
[jit_opt_a]: 0.00596562, [2]
    [Cycle 1]: 0.00189852, [27]
        [switch_simplify]: 6.045e-05
        [loop_unroll]: 2.968e-05
        [a_1]: 0.00057924
        [with_stream_mark]: 2.182e-05
        [recompute_prepare]: 8.25999e-06
        [updatestate_depend_eliminate]: 4.17e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.53003e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 6.32001e-06
        [updatestate_useless_node_eliminater]: 5.26002e-06
        [accelerated_algorithm]: 5.86003e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 5.67001e-06
        [merge_forward]: 3.68e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.48e-05
        [j_node_and_user_rematch]: 9.57001e-06
        [meta_fg_expand]: 2.36998e-06
        [replace_old_param]: 1.05e-05
        [inline_without_move]: 5.33002e-06
        [renormalize]: 0.00087623
        [add_forward_monad_depend]: 5.74999e-06
        [auto_monad_grad]: 2.76e-06
        [auto_monad_eliminator]: 1.324e-05
        [cse]: 2.806e-05
        [replace_applicator]: 1.25e-05
    [Cycle 2]: 0.00032698, [27]
        [switch_simplify]: 5.73997e-06
        [loop_unroll]: 5.13002e-06
        [a_1]: 9.009e-05
        [with_stream_mark]: 9.33002e-06
        [recompute_prepare]: 4.91002e-06
        [updatestate_depend_eliminate]: 2.49999e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.07998e-06
        [specialize_transform]: 4.76002e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 5.29e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.89e-06
        [merge_forward]: 2.54001e-06
        [cell_reuse_recompute_pass]: 1.96998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.275e-05
        [j_node_and_user_rematch]: 7.44002e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 7.97998e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.33002e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.25999e-06
        [cse]: 1.028e-05
        [replace_applicator]: 4.99e-06
[py_interpret_to_execute_after_opt_a]: 1.042e-05
[rewriter_after_opt_a]: 3.443e-05
[convert_after_rewriter]: 6.64001e-06
[order_py_execute_after_rewriter]: 4.73001e-06
[mutable_eliminate]: 0.00059179
[jit_opt_b]: 5.117e-05, [1]
    [Cycle 1]: 4.442e-05, [2]
        [frontend_op_eliminate]: 1.727e-05
        [inline_after_opt_a]: 1.571e-05
[cconv]: 2.558e-05
[loop_unroll]: 0.00050588
[jit_opt_after_cconv]: 0.00015638, [1]
    [Cycle 1]: 0.00014905, [11]
        [c_1]: 2.433e-05
        [parameter_eliminate]: 2.76e-06
        [updatestate_depend_eliminate]: 6.21e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 2.081e-05
        [call_graph_tuple_transform]: 2.242e-05
        [tuple_list_get_item_eliminator]: 5.84e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 2.9002e-07
        [switch_simplify]: 5.52001e-06
[remove_dup_value]: 1.408e-05
[partial_unused_args_eliminate]: 2.93e-06
[environ_conv]: 5.40999e-06
[add_recomputation]: 5.381e-05
[cse_after_recomputation]: 2.145e-05, [1]
    [Cycle 1]: 1.536e-05, [1]
        [cse]: 9.20001e-06
[auto_monad_reorder]: 1.508e-05
[get_jit_bprop_graph]: 2.43998e-06
[rewriter_after_jit_bprop_graph]: 5.59e-06
[opt_after_jit_grad]: 0.00051968
[symbol_engine_optimizer]: 8.393e-05, [1]
    [Cycle 1]: 7.573e-05, [6]
        [build]: 4.28001e-06
        [elim_shapecalc]: 8.52998e-06
        [elim_not_effective]: 1.441e-05
        [opt_reshape]: 6.74001e-06
        [fold_const_symbol]: 9.93002e-06
        [renormalize]: 5.69999e-07
[validate]: 4.225e-05
[backend_pass]: 1.19e-06
[task_emit]: 0.00651485
[execute]: 8.78001e-06
Sums
    bootstrap                            :     0.000362s :  2.15%
    type_inference                       :     0.005782s : 34.31%
    event_method                         :     0.000022s :  0.13%
    auto_monad                           :     0.000071s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000049s :  0.29%
    rewriter_before_opt_a                :     0.000080s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.21%
    jit_opt_a.a_1                        :     0.000669s :  3.97%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000876s :  5.20%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.11%
    jit_opt_a.cse                        :     0.000038s :  0.23%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000034s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000592s :  3.51%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000026s :  0.15%
    loop_unroll                          :     0.000506s :  3.00%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000054s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000520s :  3.08%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006515s : 38.66%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000239    23
 0.80% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.68% :     0.000006s :      3: substitution.graph_param_transform
85.03% :     0.000203s :      4: substitution.inline
 1.34% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.86% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.11% :     0.000005s :      2: substitution.replace_old_param
 5.59% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005697     2
80.06% :     0.004561s :      1: type_inference.infer
19.94% :     0.001136s :      1: type_inference.specialize
------[replace.]   0.000073     6
67.17% :     0.000049s :      4: replace.inline
32.83% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000213     6
94.26% :     0.000200s :      4: match.inline
 5.74% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.19% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.64% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000001s :     10: predicate.addn_zero_filter
 2.13% :     0.000003s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.17% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.76% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.27% :     0.000002s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.76% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.24% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.87% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.70% :     0.000003s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.81% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.34% :     0.000002s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.73% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.07% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.27% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     10: predicate.value_based_eliminate
 0.44% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000780    12
48.15% :     0.000376s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.16% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.69% :     0.000356s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023120    76
 0.25% :     0.000057s :      1: add_recomputation
 0.32% :     0.000075s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.65% :     0.000380s :      1: bootstrap
 0.12% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.12% :     0.000028s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.82% :     0.005969s :      1: jit_opt_a
 0.69% :     0.000159s :      1: jit_opt_after_cconv
 0.23% :     0.000054s :      1: jit_opt_b
 2.23% :     0.000515s :      1: loop_unroll
 2.59% :     0.000599s :      1: mutable_eliminate
 3.84% :     0.000888s :     26: opt.transform.jit_opt_a
 0.23% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.28% :     0.000528s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000052s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.06% :     0.000476s :      1: renormalize.infer
 1.69% :     0.000391s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000037s :      1: rewriter_after_opt_a
 0.36% :     0.000084s :      1: rewriter_before_opt_a
 0.38% :     0.000087s :      1: symbol_engine_optimizer
28.24% :     0.006529s :      1: task_emit
25.09% :     0.005801s :      1: type_inference
 0.28% :     0.000066s :      1: validate


TotalTime = 0.0215539, [33]
[bootstrap]: 0.00036483
[type_inference]: 0.00537931
[event_method]: 1.999e-05
[auto_monad]: 6.718e-05
[graph_reusing]: 5.92001e-06
[pre_auto_parallel]: 2.69001e-06
[py_interpret_to_execute]: 4.546e-05
[rewriter_before_opt_a]: 8.28e-05
[expand_dump_flag]: 3.81999e-06
[jit_opt_a]: 0.00615107, [2]
    [Cycle 1]: 0.00174898, [27]
        [switch_simplify]: 6.225e-05
        [loop_unroll]: 2.842e-05
        [a_1]: 0.00055074
        [with_stream_mark]: 1.961e-05
        [recompute_prepare]: 7.85e-06
        [updatestate_depend_eliminate]: 3.91001e-06
        [updatestate_assign_eliminate]: 3.06999e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [parameter_eliminate]: 2.19999e-06
        [specialize_transform]: 6.19999e-06
        [updatestate_useless_node_eliminater]: 5.55001e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 2.41e-06
        [get_grad_eliminate_]: 5.49998e-06
        [merge_forward]: 3.93999e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.511e-05
        [j_node_and_user_rematch]: 8.84e-06
        [meta_fg_expand]: 2.37001e-06
        [replace_old_param]: 9.36002e-06
        [inline_without_move]: 5.70001e-06
        [renormalize]: 0.00075306
        [add_forward_monad_depend]: 5.74e-06
        [auto_monad_grad]: 2.21e-06
        [auto_monad_eliminator]: 1.344e-05
        [cse]: 2.638e-05
        [replace_applicator]: 1.387e-05
    [Cycle 2]: 0.00039376, [27]
        [switch_simplify]: 5.49998e-06
        [loop_unroll]: 4.65999e-06
        [a_1]: 9.35e-05
        [with_stream_mark]: 1.149e-05
        [recompute_prepare]: 5.15001e-06
        [updatestate_depend_eliminate]: 3.11001e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.81998e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 5.24e-06
        [updatestate_useless_node_eliminater]: 5.09998e-06
        [accelerated_algorithm]: 5.23002e-06
        [meta_shard_fg_expand]: 1.02e-06
        [get_grad_eliminate_]: 4.40999e-06
        [merge_forward]: 2.33998e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.436e-05
        [j_node_and_user_rematch]: 8.38999e-06
        [meta_fg_expand]: 1.56002e-06
        [replace_old_param]: 8.40001e-06
        [inline_without_move]: 4.90999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 8.99978e-07
        [auto_monad_eliminator]: 5.12e-06
        [cse]: 6.828e-05
        [replace_applicator]: 6.32001e-06
[py_interpret_to_execute_after_opt_a]: 1.115e-05
[rewriter_after_opt_a]: 3.863e-05
[convert_after_rewriter]: 7.51999e-06
[order_py_execute_after_rewriter]: 5.82001e-06
[mutable_eliminate]: 0.00069339
[jit_opt_b]: 5.458e-05, [1]
    [Cycle 1]: 4.665e-05, [2]
        [frontend_op_eliminate]: 1.778e-05
        [inline_after_opt_a]: 1.625e-05
[cconv]: 2.66e-05
[loop_unroll]: 0.00047455
[jit_opt_after_cconv]: 0.00015481, [1]
    [Cycle 1]: 0.00014737, [11]
        [c_1]: 2.35e-05
        [parameter_eliminate]: 2.84001e-06
        [updatestate_depend_eliminate]: 6.08002e-06
        [updatestate_assign_eliminate]: 3.16001e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [cse]: 2.349e-05
        [call_graph_tuple_transform]: 1.996e-05
        [tuple_list_get_item_eliminator]: 5.91998e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 6.47001e-06
[remove_dup_value]: 1.489e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 5.85002e-06
[add_recomputation]: 5.256e-05
[cse_after_recomputation]: 2.363e-05, [1]
    [Cycle 1]: 1.72e-05, [1]
        [cse]: 9.65002e-06
[auto_monad_reorder]: 1.649e-05
[get_jit_bprop_graph]: 2.15002e-06
[rewriter_after_jit_bprop_graph]: 6.24001e-06
[opt_after_jit_grad]: 0.00051366
[symbol_engine_optimizer]: 7.803e-05, [1]
    [Cycle 1]: 7.095e-05, [6]
        [build]: 3.83001e-06
        [elim_shapecalc]: 7.77998e-06
        [elim_not_effective]: 1.321e-05
        [opt_reshape]: 6.31e-06
        [fold_const_symbol]: 1.028e-05
        [renormalize]: 7.7e-07
[validate]: 4.18e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00696217
[execute]: 8.31002e-06
Sums
    bootstrap                            :     0.000365s :  2.16%
    type_inference                       :     0.005379s : 31.87%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000067s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000045s :  0.27%
    rewriter_before_opt_a                :     0.000083s :  0.49%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.20%
    jit_opt_a.a_1                        :     0.000644s :  3.82%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000753s :  4.46%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.11%
    jit_opt_a.cse                        :     0.000095s :  0.56%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000039s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000693s :  4.11%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000027s :  0.16%
    loop_unroll                          :     0.000475s :  2.81%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000053s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000514s :  3.04%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006962s : 41.25%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000218    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.83% :     0.000002s :      2: substitution.fold_const_symbol
 2.92% :     0.000006s :      3: substitution.graph_param_transform
83.15% :     0.000182s :      4: substitution.inline
 1.35% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.35% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.35% :     0.000005s :      2: substitution.replace_old_param
 6.20% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005300     2
81.90% :     0.004341s :      1: type_inference.infer
18.10% :     0.000959s :      1: type_inference.specialize
------[replace.]   0.000074     6
68.21% :     0.000050s :      4: replace.inline
31.79% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000191     6
93.55% :     0.000179s :      4: match.inline
 6.45% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.57% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.54% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 2.33% :     0.000003s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000001s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.59% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.25% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.63% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.12% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.35% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.91% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.78% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.35% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.73% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.26% :     0.000001s :     10: predicate.reshape_eliminate
 1.58% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.60% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 1.15% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.78% :     0.000009s :     53: predicate.switch_simplify
 1.44% :     0.000002s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.28% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.81% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.66% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.45% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000656    12
49.42% :     0.000324s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.67% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.92% :     0.000301s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023280    76
 0.24% :     0.000055s :      1: add_recomputation
 0.30% :     0.000071s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.64% :     0.000383s :      1: bootstrap
 0.13% :     0.000029s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000026s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.11% :     0.000025s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.44% :     0.006155s :      1: jit_opt_a
 0.68% :     0.000158s :      1: jit_opt_after_cconv
 0.24% :     0.000057s :      1: jit_opt_b
 2.07% :     0.000483s :      1: loop_unroll
 3.02% :     0.000703s :      1: mutable_eliminate
 3.73% :     0.000867s :     26: opt.transform.jit_opt_a
 0.22% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.24% :     0.000522s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.21% :     0.000048s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 1.66% :     0.000386s :      1: renormalize.infer
 1.54% :     0.000358s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000042s :      1: rewriter_after_opt_a
 0.37% :     0.000086s :      1: rewriter_before_opt_a
 0.35% :     0.000081s :      1: symbol_engine_optimizer
30.00% :     0.006984s :      1: task_emit
23.19% :     0.005398s :      1: type_inference
 0.28% :     0.000065s :      1: validate


TotalTime = 0.0200121, [33]
[bootstrap]: 0.00026551
[type_inference]: 0.00447418
[event_method]: 1.782e-05
[auto_monad]: 6.456e-05
[graph_reusing]: 5.82001e-06
[pre_auto_parallel]: 2.59999e-06
[py_interpret_to_execute]: 3.295e-05
[rewriter_before_opt_a]: 6.796e-05
[expand_dump_flag]: 2.89999e-06
[jit_opt_a]: 0.00571676, [2]
    [Cycle 1]: 0.00177835, [27]
        [switch_simplify]: 5.393e-05
        [loop_unroll]: 3.143e-05
        [a_1]: 0.00058022
        [with_stream_mark]: 2.081e-05
        [recompute_prepare]: 8.62998e-06
        [updatestate_depend_eliminate]: 3.56999e-06
        [updatestate_assign_eliminate]: 2.54001e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [parameter_eliminate]: 1.67001e-06
        [specialize_transform]: 6.83e-06
        [updatestate_useless_node_eliminater]: 5.52001e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 1.99999e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 3.60003e-06
        [cell_reuse_recompute_pass]: 2.09999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.636e-05
        [j_node_and_user_rematch]: 9.07001e-06
        [meta_fg_expand]: 2.02999e-06
        [replace_old_param]: 9.59e-06
        [inline_without_move]: 5.59998e-06
        [renormalize]: 0.00073945
        [add_forward_monad_depend]: 6.40002e-06
        [auto_monad_grad]: 2.86e-06
        [auto_monad_eliminator]: 1.486e-05
        [cse]: 2.268e-05
        [replace_applicator]: 1.577e-05
    [Cycle 2]: 0.00034935, [27]
        [switch_simplify]: 5.95002e-06
        [loop_unroll]: 5.54e-06
        [a_1]: 9.412e-05
        [with_stream_mark]: 1.233e-05
        [recompute_prepare]: 5.79e-06
        [updatestate_depend_eliminate]: 2.69001e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 5.54e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.69e-06
        [meta_shard_fg_expand]: 1.54998e-06
        [get_grad_eliminate_]: 4.74e-06
        [merge_forward]: 3.11001e-06
        [cell_reuse_recompute_pass]: 1.91e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.463e-05
        [j_node_and_user_rematch]: 9.15001e-06
        [meta_fg_expand]: 1.72001e-06
        [replace_old_param]: 9.80002e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.40999e-06
        [auto_monad_grad]: 1.01002e-06
        [auto_monad_eliminator]: 6.44999e-06
        [cse]: 1.055e-05
        [replace_applicator]: 6.69999e-06
[py_interpret_to_execute_after_opt_a]: 1.651e-05
[rewriter_after_opt_a]: 3.709e-05
[convert_after_rewriter]: 8.02e-06
[order_py_execute_after_rewriter]: 5.24e-06
[mutable_eliminate]: 0.00070151
[jit_opt_b]: 4.967e-05, [1]
    [Cycle 1]: 4.274e-05, [2]
        [frontend_op_eliminate]: 1.622e-05
        [inline_after_opt_a]: 1.545e-05
[cconv]: 2.488e-05
[loop_unroll]: 0.00044545
[jit_opt_after_cconv]: 0.00013855, [1]
    [Cycle 1]: 0.00013218, [11]
        [c_1]: 1.994e-05
        [parameter_eliminate]: 3.23998e-06
        [updatestate_depend_eliminate]: 5.12e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [cse]: 1.813e-05
        [call_graph_tuple_transform]: 1.842e-05
        [tuple_list_get_item_eliminator]: 5.16002e-06
        [none_parameter_eliminate]: 1.32999e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 5.48997e-06
[remove_dup_value]: 1.215e-05
[partial_unused_args_eliminate]: 2.10002e-06
[environ_conv]: 4.60999e-06
[add_recomputation]: 4.229e-05
[cse_after_recomputation]: 2.189e-05, [1]
    [Cycle 1]: 1.591e-05, [1]
        [cse]: 8.47e-06
[auto_monad_reorder]: 1.204e-05
[get_jit_bprop_graph]: 1.78002e-06
[rewriter_after_jit_bprop_graph]: 7.04001e-06
[opt_after_jit_grad]: 0.000531
[symbol_engine_optimizer]: 7.365e-05, [1]
    [Cycle 1]: 6.594e-05, [6]
        [build]: 3.58e-06
        [elim_shapecalc]: 7.43e-06
        [elim_not_effective]: 1.229e-05
        [opt_reshape]: 5.22e-06
        [fold_const_symbol]: 8.42998e-06
        [renormalize]: 4.00003e-07
[validate]: 3.286e-05
[backend_pass]: 7.89994e-07
[task_emit]: 0.00697517
[execute]: 9.80002e-06
Sums
    bootstrap                            :     0.000266s :  1.68%
    type_inference                       :     0.004474s : 28.37%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000065s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.21%
    rewriter_before_opt_a                :     0.000068s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000060s :  0.38%
    jit_opt_a.loop_unroll                :     0.000037s :  0.23%
    jit_opt_a.a_1                        :     0.000674s :  4.28%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000740s :  4.69%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.14%
    jit_opt_a.cse                        :     0.000033s :  0.21%
    jit_opt_a.replace_applicator         :     0.000022s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.10%
    rewriter_after_opt_a                 :     0.000037s :  0.24%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000702s :  4.45%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000025s :  0.16%
    loop_unroll                          :     0.000445s :  2.82%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000012s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000042s :  0.27%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000012s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000531s :  3.37%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006975s : 44.23%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000228    23
 0.68% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.16% :     0.000005s :      3: substitution.graph_param_transform
85.08% :     0.000194s :      4: substitution.inline
 1.42% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.53% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.95% :     0.000004s :      2: substitution.replace_old_param
 5.57% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004409     2
81.71% :     0.003603s :      1: type_inference.infer
18.29% :     0.000806s :      1: type_inference.specialize
------[replace.]   0.000074     6
63.39% :     0.000047s :      4: replace.inline
36.61% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000203     6
94.29% :     0.000191s :      4: match.inline
 5.71% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.29% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.42% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.52% :     0.000002s :     10: predicate.addn_zero_filter
 1.91% :     0.000002s :     10: predicate.arithmetic_simplify
 1.45% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.73% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.47% :     0.000001s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.73% :     0.000007s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.45% :     0.000002s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.23% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.68% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.23% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.42% :     0.000002s :     10: predicate.reduce_eliminate
 1.85% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.67% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.38% :     0.000000s :      3: predicate.reset_defer_inline
 1.84% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.10% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 1.15% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.65% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.74% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.27% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.08% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.90% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000548    12
44.70% :     0.000245s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.57% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.73% :     0.000262s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021750    76
 0.21% :     0.000045s :      1: add_recomputation
 0.31% :     0.000068s :      1: auto_monad
 0.07% :     0.000015s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.27% :     0.000277s :      1: bootstrap
 0.13% :     0.000027s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.11% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.30% :     0.005720s :      1: jit_opt_a
 0.65% :     0.000141s :      1: jit_opt_after_cconv
 0.24% :     0.000052s :      1: jit_opt_b
 2.09% :     0.000454s :      1: loop_unroll
 3.27% :     0.000711s :      1: mutable_eliminate
 4.14% :     0.000901s :     26: opt.transform.jit_opt_a
 0.21% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.48% :     0.000540s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000036s :      1: py_interpret_to_execute
 0.09% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000014s :      1: remove_dup_value
 1.93% :     0.000421s :      1: renormalize.infer
 1.42% :     0.000308s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000041s :      1: rewriter_after_opt_a
 0.32% :     0.000071s :      1: rewriter_before_opt_a
 0.35% :     0.000076s :      1: symbol_engine_optimizer
32.16% :     0.006995s :      1: task_emit
20.64% :     0.004488s :      1: type_inference
 0.25% :     0.000054s :      1: validate


TotalTime = 0.0158611, [30]
[bootstrap]: 0.00035517
[type_inference]: 0.00755042
[event_method]: 2.169e-05
[auto_monad]: 6.094e-05
[graph_reusing]: 3.78001e-06
[pre_auto_parallel]: 1.85001e-06
[py_interpret_to_execute]: 5.513e-05
[rewriter_before_opt_a]: 7.514e-05
[expand_dump_flag]: 1.76e-06
[jit_opt_a]: 0.00556223, [2]
    [Cycle 1]: 0.00186035, [27]
        [switch_simplify]: 0.00011625
        [loop_unroll]: 3.6e-05
        [a_1]: 0.00070075
        [with_stream_mark]: 1.195e-05
        [recompute_prepare]: 6.74999e-06
        [updatestate_depend_eliminate]: 2.81999e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 6.23002e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 5.64e-06
        [meta_shard_fg_expand]: 1.26002e-06
        [get_grad_eliminate_]: 5.43002e-06
        [merge_forward]: 2.69001e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.198e-05
        [j_node_and_user_rematch]: 9.67999e-06
        [meta_fg_expand]: 1.92001e-06
        [replace_old_param]: 7.93001e-06
        [inline_without_move]: 5.40999e-06
        [renormalize]: 0.00071954
        [add_forward_monad_depend]: 4.94e-06
        [auto_monad_grad]: 1.25001e-06
        [auto_monad_eliminator]: 8.59e-06
        [cse]: 1.449e-05
        [replace_applicator]: 1.127e-05
    [Cycle 2]: 0.00034905, [27]
        [switch_simplify]: 5.92999e-06
        [loop_unroll]: 5.29e-06
        [a_1]: 9.494e-05
        [with_stream_mark]: 8.65001e-06
        [recompute_prepare]: 6.15002e-06
        [updatestate_depend_eliminate]: 2.87002e-06
        [updatestate_assign_eliminate]: 2.64999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 9.00007e-07
        [specialize_transform]: 5.46998e-06
        [updatestate_useless_node_eliminater]: 5.49e-06
        [accelerated_algorithm]: 4.85999e-06
        [meta_shard_fg_expand]: 1.29998e-06
        [get_grad_eliminate_]: 4.87e-06
        [merge_forward]: 3.14999e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.386e-05
        [j_node_and_user_rematch]: 8.22e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 9.19e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.70001e-06
        [auto_monad_grad]: 1.25001e-06
        [auto_monad_eliminator]: 7.83999e-06
        [cse]: 1.344e-05
        [replace_applicator]: 6.61999e-06
[py_interpret_to_execute_after_opt_a]: 9.67001e-06
[rewriter_after_opt_a]: 2.72e-05
[convert_after_rewriter]: 6.96999e-06
[order_py_execute_after_rewriter]: 4.00998e-06
[mutable_eliminate]: 0.0005431
[jit_opt_b]: 6.37e-05, [1]
    [Cycle 1]: 4.523e-05, [2]
        [frontend_op_eliminate]: 1.848e-05
        [inline_after_opt_a]: 1.518e-05
[cconv]: 1.851e-05
[loop_unroll]: 0.00048336
[jit_opt_after_cconv]: 0.00014454, [1]
    [Cycle 1]: 0.00013809, [11]
        [c_1]: 2.282e-05
        [parameter_eliminate]: 1.76e-06
        [updatestate_depend_eliminate]: 5.09e-06
        [updatestate_assign_eliminate]: 3.01999e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [cse]: 1.518e-05
        [call_graph_tuple_transform]: 1.908e-05
        [tuple_list_get_item_eliminator]: 5.57001e-06
        [none_parameter_eliminate]: 8.60018e-07
        [renormalize]: 4.50003e-07
        [switch_simplify]: 6.46e-06
[remove_dup_value]: 7.6e-06
[partial_unused_args_eliminate]: 1.87001e-06
[environ_conv]: 3.8e-06
[add_recomputation]: 3.406e-05
[cse_after_recomputation]: 2.543e-05, [1]
    [Cycle 1]: 1.941e-05, [1]
        [cse]: 1.046e-05
[auto_monad_reorder]: 1.065e-05
[get_jit_bprop_graph]: 1.54998e-06
[rewriter_after_jit_bprop_graph]: 5.92999e-06
[opt_after_jit_grad]: 0.00049738
[symbol_engine_optimizer]: 7.649e-05, [1]
    [Cycle 1]: 7.017e-05, [6]
        [build]: 2.56e-06
        [elim_shapecalc]: 8.11002e-06
        [elim_not_effective]: 1.403e-05
        [opt_reshape]: 5.67999e-06
        [fold_const_symbol]: 9.34998e-06
        [renormalize]: 5.10016e-07
[validate]: 3.013e-05
Sums
    bootstrap                            :     0.000355s :  2.98%
    type_inference                       :     0.007550s : 63.41%
    event_method                         :     0.000022s :  0.18%
    auto_monad                           :     0.000061s :  0.51%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000055s :  0.46%
    rewriter_before_opt_a                :     0.000075s :  0.63%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000122s :  1.03%
    jit_opt_a.loop_unroll                :     0.000041s :  0.35%
    jit_opt_a.a_1                        :     0.000796s :  6.68%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.14%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.000720s :  6.04%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.14%
    jit_opt_a.cse                        :     0.000028s :  0.23%
    jit_opt_a.replace_applicator         :     0.000018s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.08%
    rewriter_after_opt_a                 :     0.000027s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000543s :  4.56%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.13%
    cconv                                :     0.000019s :  0.16%
    loop_unroll                          :     0.000483s :  4.06%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000008s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000034s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000011s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000497s :  4.18%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000030s :  0.25%


Time group info:
------[substitution.]   0.000263    26
 0.67% :     0.000002s :      2: substitution.elim_not_effective
 0.42% :     0.000001s :      2: substitution.fold_const_symbol
 1.69% :     0.000004s :      3: substitution.graph_param_transform
86.67% :     0.000228s :      6: substitution.inline
 1.04% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.86% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.29% :     0.000003s :      2: substitution.replace_old_param
 3.18% :     0.000008s :      1: substitution.switch_simplify
 3.19% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007482     2
83.12% :     0.006218s :      1: type_inference.infer
16.88% :     0.001263s :      1: type_inference.specialize
------[replace.]   0.000148     9
47.11% :     0.000070s :      6: replace.inline
26.17% :     0.000039s :      1: replace.switch_simplify
26.72% :     0.000040s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000239     9
93.88% :     0.000224s :      6: match.inline
 3.09% :     0.000007s :      1: match.switch_simplify
 3.03% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   801
 1.27% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.46% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     12: predicate.addn_check_dump
 1.47% :     0.000002s :     12: predicate.addn_zero_filter
 2.06% :     0.000003s :     12: predicate.arithmetic_simplify
 1.26% :     0.000002s :     12: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.33% :     0.000002s :     12: predicate.compare_switch_simplify
 1.74% :     0.000002s :     12: predicate.depend_value_elim
 1.22% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.58% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.65% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.43% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.11% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.33% :     0.000002s :     12: predicate.environ_get_eliminate
 1.26% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.37% :     0.000007s :     26: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.65% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     17: predicate.load_eliminater
 1.17% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.34% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.92% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000002s :     12: predicate.merge_addn
 1.17% :     0.000002s :     12: predicate.minmaximum_grad
 2.27% :     0.000003s :      3: predicate.mutable_eliminate
 0.36% :     0.000000s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     17: predicate.partial_eliminate
 1.16% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     12: predicate.reduce_eliminate
 1.41% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     20: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.41% :     0.000001s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     12: predicate.reshape_eliminate
 1.25% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.93% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     12: predicate.same_eliminate
 0.62% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.52% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     20: predicate.switch_defer_inline
 2.26% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.20% :     0.000011s :     65: predicate.switch_simplify
 1.29% :     0.000002s :     12: predicate.tile_eliminate
 1.30% :     0.000002s :     12: predicate.transpose_eliminate
 1.54% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.05% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.63% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000932    15
52.41% :     0.000489s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.27% :     0.000096s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.32% :     0.000348s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017724    72
 0.21% :     0.000038s :      1: add_recomputation
 0.36% :     0.000064s :      1: auto_monad
 0.08% :     0.000014s :      1: auto_monad_reorder
 2.11% :     0.000374s :      1: bootstrap
 0.12% :     0.000021s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000028s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.14% :     0.000025s :      1: event_method
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
31.40% :     0.005566s :      1: jit_opt_a
 0.83% :     0.000147s :      1: jit_opt_after_cconv
 0.37% :     0.000066s :      1: jit_opt_b
 2.77% :     0.000491s :      1: loop_unroll
 3.11% :     0.000552s :      1: mutable_eliminate
 6.08% :     0.001079s :     26: opt.transform.jit_opt_a
 0.28% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000027s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.85% :     0.000505s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.33% :     0.000058s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000010s :      1: remove_dup_value
 2.29% :     0.000406s :      1: renormalize.infer
 1.73% :     0.000306s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000030s :      1: rewriter_after_opt_a
 0.44% :     0.000078s :      1: rewriter_before_opt_a
 0.45% :     0.000079s :      1: symbol_engine_optimizer
42.68% :     0.007564s :      1: type_inference


TotalTime = 0.0179117, [33]
[bootstrap]: 0.00031943
[type_inference]: 0.0044454
[event_method]: 1.749e-05
[auto_monad]: 5.543e-05
[graph_reusing]: 5.62999e-06
[pre_auto_parallel]: 2.54001e-06
[py_interpret_to_execute]: 3.188e-05
[rewriter_before_opt_a]: 6.825e-05
[expand_dump_flag]: 2.81e-06
[jit_opt_a]: 0.00526092, [2]
    [Cycle 1]: 0.0015621, [27]
        [switch_simplify]: 4.88e-05
        [loop_unroll]: 2.824e-05
        [a_1]: 0.00050152
        [with_stream_mark]: 1.387e-05
        [recompute_prepare]: 7.08e-06
        [updatestate_depend_eliminate]: 2.74999e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.26002e-06
        [specialize_transform]: 5.87999e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.24998e-06
        [meta_shard_fg_expand]: 1.81e-06
        [get_grad_eliminate_]: 5.38002e-06
        [merge_forward]: 2.54001e-06
        [cell_reuse_recompute_pass]: 1.37999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.073e-05
        [j_node_and_user_rematch]: 8.81002e-06
        [meta_fg_expand]: 1.81998e-06
        [replace_old_param]: 8.93002e-06
        [inline_without_move]: 5.46e-06
        [renormalize]: 0.00067173
        [add_forward_monad_depend]: 5.57999e-06
        [auto_monad_grad]: 2.57001e-06
        [auto_monad_eliminator]: 1.233e-05
        [cse]: 1.954e-05
        [replace_applicator]: 1.256e-05
    [Cycle 2]: 0.0003083, [27]
        [switch_simplify]: 5.66e-06
        [loop_unroll]: 4.78001e-06
        [a_1]: 8.831e-05
        [with_stream_mark]: 1.029e-05
        [recompute_prepare]: 4.80001e-06
        [updatestate_depend_eliminate]: 2.25002e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 1.86003e-06
        [parameter_eliminate]: 1.09998e-06
        [specialize_transform]: 4.76997e-06
        [updatestate_useless_node_eliminater]: 4.60999e-06
        [accelerated_algorithm]: 4.38999e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.50999e-06
        [merge_forward]: 2.51998e-06
        [cell_reuse_recompute_pass]: 2.98e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.184e-05
        [j_node_and_user_rematch]: 7.8e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 7.26001e-06
        [inline_without_move]: 4.60999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.00999e-06
        [auto_monad_grad]: 5.59987e-07
        [auto_monad_eliminator]: 5.22e-06
        [cse]: 8.33001e-06
        [replace_applicator]: 5.10001e-06
[py_interpret_to_execute_after_opt_a]: 1.064e-05
[rewriter_after_opt_a]: 3.146e-05
[convert_after_rewriter]: 6.94999e-06
[order_py_execute_after_rewriter]: 4.72998e-06
[mutable_eliminate]: 0.00059575
[jit_opt_b]: 5.45e-05, [1]
    [Cycle 1]: 4.772e-05, [2]
        [frontend_op_eliminate]: 1.592e-05
        [inline_after_opt_a]: 1.984e-05
[cconv]: 2.626e-05
[loop_unroll]: 0.00044061
[jit_opt_after_cconv]: 0.00014637, [1]
    [Cycle 1]: 0.00014017, [11]
        [c_1]: 2.032e-05
        [parameter_eliminate]: 3.88999e-06
        [updatestate_depend_eliminate]: 5.81e-06
        [updatestate_assign_eliminate]: 2.43998e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 2.05e-05
        [call_graph_tuple_transform]: 1.959e-05
        [tuple_list_get_item_eliminator]: 5.02e-06
        [none_parameter_eliminate]: 1.74e-06
        [renormalize]: 5.10016e-07
        [switch_simplify]: 5.30999e-06
[remove_dup_value]: 1.133e-05
[partial_unused_args_eliminate]: 2.11e-06
[environ_conv]: 4.25e-06
[add_recomputation]: 4.407e-05
[cse_after_recomputation]: 2.126e-05, [1]
    [Cycle 1]: 1.581e-05, [1]
        [cse]: 9.27999e-06
[auto_monad_reorder]: 1.399e-05
[get_jit_bprop_graph]: 1.88997e-06
[rewriter_after_jit_bprop_graph]: 5.57999e-06
[opt_after_jit_grad]: 0.00048014
[symbol_engine_optimizer]: 7.219e-05, [1]
    [Cycle 1]: 6.622e-05, [6]
        [build]: 4.35999e-06
        [elim_shapecalc]: 7.17002e-06
        [elim_not_effective]: 1.223e-05
        [opt_reshape]: 5.99e-06
        [fold_const_symbol]: 8.48999e-06
        [renormalize]: 9.80013e-07
[validate]: 3.203e-05
[backend_pass]: 1.00001e-06
[task_emit]: 0.00548813
[execute]: 6.38998e-06
Sums
    bootstrap                            :     0.000319s :  2.29%
    type_inference                       :     0.004445s : 31.93%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000055s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.23%
    rewriter_before_opt_a                :     0.000068s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.39%
    jit_opt_a.loop_unroll                :     0.000033s :  0.24%
    jit_opt_a.a_1                        :     0.000590s :  4.24%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000672s :  4.83%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000028s :  0.20%
    jit_opt_a.replace_applicator         :     0.000018s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.08%
    rewriter_after_opt_a                 :     0.000031s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000596s :  4.28%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.14%
    cconv                                :     0.000026s :  0.19%
    loop_unroll                          :     0.000441s :  3.16%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000011s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000044s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000480s :  3.45%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005488s : 39.42%
    execute                              :     0.000006s :  0.05%


Time group info:
------[substitution.]   0.000180    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.79% :     0.000001s :      2: substitution.fold_const_symbol
 2.72% :     0.000005s :      3: substitution.graph_param_transform
82.45% :     0.000148s :      4: substitution.inline
 1.93% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.11% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.53% :     0.000005s :      2: substitution.replace_old_param
 6.39% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004382     2
83.70% :     0.003667s :      1: type_inference.infer
16.30% :     0.000714s :      1: type_inference.specialize
------[replace.]   0.000069     6
70.42% :     0.000048s :      4: replace.inline
29.58% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000156     6
93.36% :     0.000146s :      4: match.inline
 6.64% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.28% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.07% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 1.80% :     0.000002s :     10: predicate.arithmetic_simplify
 1.41% :     0.000002s :     10: predicate.cast_eliminate
 0.57% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.63% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.91% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.53% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.69% :     0.000001s :      3: predicate.graph_param_transform
 5.31% :     0.000006s :     22: predicate.inline
 1.07% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000001s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.90% :     0.000002s :     15: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.04% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.21% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.42% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000000s :      3: predicate.opt_reshape
 2.09% :     0.000002s :     15: predicate.partial_eliminate
 1.32% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.97% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000001s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.79% :     0.000009s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.54% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.42% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000532    12
48.38% :     0.000257s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.22% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.41% :     0.000236s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019480    76
 0.24% :     0.000048s :      1: add_recomputation
 0.30% :     0.000059s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.74% :     0.000339s :      1: bootstrap
 0.15% :     0.000030s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.02% :     0.005264s :      1: jit_opt_a
 0.77% :     0.000149s :      1: jit_opt_after_cconv
 0.29% :     0.000057s :      1: jit_opt_b
 2.30% :     0.000449s :      1: loop_unroll
 3.10% :     0.000604s :      1: mutable_eliminate
 4.05% :     0.000788s :     26: opt.transform.jit_opt_a
 0.24% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000030s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.51% :     0.000488s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000035s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000014s :      1: remove_dup_value
 1.97% :     0.000383s :      1: renormalize.infer
 1.44% :     0.000281s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000035s :      1: rewriter_after_opt_a
 0.36% :     0.000071s :      1: rewriter_before_opt_a
 0.38% :     0.000074s :      1: symbol_engine_optimizer
28.24% :     0.005502s :      1: task_emit
22.89% :     0.004459s :      1: type_inference
 0.27% :     0.000053s :      1: validate


TotalTime = 0.0220939, [33]
[bootstrap]: 0.00036922
[type_inference]: 0.00524872
[event_method]: 1.808e-05
[auto_monad]: 5.155e-05
[graph_reusing]: 4.19002e-06
[pre_auto_parallel]: 2.09999e-06
[py_interpret_to_execute]: 3.506e-05
[rewriter_before_opt_a]: 6.764e-05
[expand_dump_flag]: 2.21998e-06
[jit_opt_a]: 0.0073754, [2]
    [Cycle 1]: 0.002936, [27]
        [switch_simplify]: 6.011e-05
        [loop_unroll]: 2.944e-05
        [a_1]: 0.00058551
        [with_stream_mark]: 1.758e-05
        [recompute_prepare]: 1.003e-05
        [updatestate_depend_eliminate]: 3.56999e-06
        [updatestate_assign_eliminate]: 2.88e-06
        [updatestate_loads_eliminate]: 3.05002e-06
        [parameter_eliminate]: 2.02001e-06
        [specialize_transform]: 7.08e-06
        [updatestate_useless_node_eliminater]: 5.59998e-06
        [accelerated_algorithm]: 6.46999e-06
        [meta_shard_fg_expand]: 2.46e-06
        [get_grad_eliminate_]: 6.41e-06
        [merge_forward]: 3.98001e-06
        [cell_reuse_recompute_pass]: 1.53002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.579e-05
        [j_node_and_user_rematch]: 1.01e-05
        [meta_fg_expand]: 2.41998e-06
        [replace_old_param]: 1.032e-05
        [inline_without_move]: 5.67999e-06
        [renormalize]: 0.00189267
        [add_forward_monad_depend]: 6.83e-06
        [auto_monad_grad]: 1.45001e-06
        [auto_monad_eliminator]: 1.467e-05
        [cse]: 2.074e-05
        [replace_applicator]: 1.852e-05
    [Cycle 2]: 0.00033683, [27]
        [switch_simplify]: 5.77999e-06
        [loop_unroll]: 5.00999e-06
        [a_1]: 0.00010464
        [with_stream_mark]: 1.121e-05
        [recompute_prepare]: 5.02e-06
        [updatestate_depend_eliminate]: 2.73e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 5.09998e-06
        [updatestate_useless_node_eliminater]: 4.60001e-06
        [accelerated_algorithm]: 4.75001e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 5.02e-06
        [merge_forward]: 2.72001e-06
        [cell_reuse_recompute_pass]: 3.33998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.277e-05
        [j_node_and_user_rematch]: 8.90001e-06
        [meta_fg_expand]: 2.36e-06
        [replace_old_param]: 7.46001e-06
        [inline_without_move]: 5.02e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.04e-06
        [auto_monad_grad]: 6.70028e-07
        [auto_monad_eliminator]: 5.00999e-06
        [cse]: 9.84001e-06
        [replace_applicator]: 4.70999e-06
[py_interpret_to_execute_after_opt_a]: 1.183e-05
[rewriter_after_opt_a]: 2.788e-05
[convert_after_rewriter]: 5.94999e-06
[order_py_execute_after_rewriter]: 4.05998e-06
[mutable_eliminate]: 0.0006952
[jit_opt_b]: 4.811e-05, [1]
    [Cycle 1]: 4.222e-05, [2]
        [frontend_op_eliminate]: 1.664e-05
        [inline_after_opt_a]: 1.409e-05
[cconv]: 1.843e-05
[loop_unroll]: 0.0004582
[jit_opt_after_cconv]: 0.00013922, [1]
    [Cycle 1]: 0.00013248, [11]
        [c_1]: 2.134e-05
        [parameter_eliminate]: 3.01001e-06
        [updatestate_depend_eliminate]: 5.05999e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.796e-05
        [call_graph_tuple_transform]: 1.747e-05
        [tuple_list_get_item_eliminator]: 5.57001e-06
        [none_parameter_eliminate]: 1.12999e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 5.82001e-06
[remove_dup_value]: 8.53001e-06
[partial_unused_args_eliminate]: 1.92999e-06
[environ_conv]: 3.28e-06
[add_recomputation]: 3.486e-05
[cse_after_recomputation]: 2.181e-05, [1]
    [Cycle 1]: 1.637e-05, [1]
        [cse]: 9.91e-06
[auto_monad_reorder]: 1.169e-05
[get_jit_bprop_graph]: 1.73002e-06
[rewriter_after_jit_bprop_graph]: 6.17999e-06
[opt_after_jit_grad]: 0.00049391
[symbol_engine_optimizer]: 7.65e-05, [1]
    [Cycle 1]: 6.933e-05, [6]
        [build]: 3.45e-06
        [elim_shapecalc]: 7.71999e-06
        [elim_not_effective]: 1.421e-05
        [opt_reshape]: 5.45001e-06
        [fold_const_symbol]: 8.72998e-06
        [renormalize]: 6.39993e-07
[validate]: 3.515e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.00659834
[execute]: 9.03002e-06
Sums
    bootstrap                            :     0.000369s :  2.13%
    type_inference                       :     0.005249s : 30.23%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000052s :  0.30%
    graph_reusing                        :     0.000004s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000035s :  0.20%
    rewriter_before_opt_a                :     0.000068s :  0.39%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000066s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000690s :  3.98%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001893s : 10.90%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.11%
    jit_opt_a.cse                        :     0.000031s :  0.18%
    jit_opt_a.replace_applicator         :     0.000023s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000028s :  0.16%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000695s :  4.00%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.08%
    cconv                                :     0.000018s :  0.11%
    loop_unroll                          :     0.000458s :  2.64%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000009s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000003s :  0.02%
    add_recomputation                    :     0.000035s :  0.20%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000494s :  2.85%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006598s : 38.01%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000238    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000001s :      2: substitution.fold_const_symbol
 1.79% :     0.000004s :      3: substitution.graph_param_transform
85.89% :     0.000205s :      4: substitution.inline
 1.69% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.04% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000005s :      2: substitution.replace_old_param
 5.00% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005176     2
81.38% :     0.004212s :      1: type_inference.infer
18.62% :     0.000964s :      1: type_inference.specialize
------[replace.]   0.000078     6
64.67% :     0.000051s :      4: replace.inline
35.33% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000212     6
94.85% :     0.000201s :      4: match.inline
 5.15% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.70% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000002s :     10: predicate.addn_zero_filter
 2.44% :     0.000003s :     10: predicate.arithmetic_simplify
 1.36% :     0.000002s :     10: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.65% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.33% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.37% :     0.000002s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.14% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 1.19% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.71% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.67% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 1.23% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.28% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.65% :     0.000002s :     10: predicate.reshape_eliminate
 1.36% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.11% :     0.000001s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 2.31% :     0.000003s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.71% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.15% :     0.000009s :     53: predicate.switch_simplify
 1.48% :     0.000002s :     10: predicate.tile_eliminate
 1.38% :     0.000002s :     10: predicate.transpose_eliminate
 1.27% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.26% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.16% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.73% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.98% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000763    12
44.43% :     0.000339s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.44% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.13% :     0.000375s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025001    76
 0.15% :     0.000038s :      1: add_recomputation
 0.22% :     0.000055s :      1: auto_monad
 0.06% :     0.000014s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.55% :     0.000389s :      1: bootstrap
 0.09% :     0.000021s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000005s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
29.52% :     0.007380s :      1: jit_opt_a
 0.57% :     0.000142s :      1: jit_opt_after_cconv
 0.20% :     0.000051s :      1: jit_opt_b
 1.86% :     0.000466s :      1: loop_unroll
 2.82% :     0.000704s :      1: mutable_eliminate
 3.68% :     0.000921s :     26: opt.transform.jit_opt_a
 0.19% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.04% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.01% :     0.000502s :      1: opt_after_jit_grad
 0.02% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000038s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000011s :      1: remove_dup_value
 5.76% :     0.001441s :      1: renormalize.infer
 1.76% :     0.000440s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000031s :      1: rewriter_after_opt_a
 0.28% :     0.000070s :      1: rewriter_before_opt_a
 0.32% :     0.000080s :      1: symbol_engine_optimizer
26.44% :     0.006611s :      1: task_emit
21.06% :     0.005265s :      1: type_inference
 0.21% :     0.000054s :      1: validate


TotalTime = 0.0131108, [30]
[bootstrap]: 0.00027384
[type_inference]: 0.00577993
[event_method]: 1.904e-05
[auto_monad]: 6.026e-05
[graph_reusing]: 4.28999e-06
[pre_auto_parallel]: 2.02001e-06
[py_interpret_to_execute]: 5.379e-05
[rewriter_before_opt_a]: 7.791e-05
[expand_dump_flag]: 2.16e-06
[jit_opt_a]: 0.00488333, [2]
    [Cycle 1]: 0.00163816, [27]
        [switch_simplify]: 9.211e-05
        [loop_unroll]: 3.564e-05
        [a_1]: 0.00067797
        [with_stream_mark]: 1.069e-05
        [recompute_prepare]: 7.4e-06
        [updatestate_depend_eliminate]: 2.99001e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 6.01e-06
        [updatestate_useless_node_eliminater]: 5.33002e-06
        [accelerated_algorithm]: 5.57999e-06
        [meta_shard_fg_expand]: 1.40999e-06
        [get_grad_eliminate_]: 5.44e-06
        [merge_forward]: 2.32999e-06
        [cell_reuse_recompute_pass]: 1.28002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.068e-05
        [j_node_and_user_rematch]: 8.84003e-06
        [meta_fg_expand]: 2.02999e-06
        [replace_old_param]: 7.85998e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 0.00054765
        [add_forward_monad_depend]: 4.23999e-06
        [auto_monad_grad]: 1.23002e-06
        [auto_monad_eliminator]: 9.19998e-06
        [cse]: 1.478e-05
        [replace_applicator]: 9.81e-06
    [Cycle 2]: 0.00031328, [27]
        [switch_simplify]: 5.27999e-06
        [loop_unroll]: 4.96002e-06
        [a_1]: 9.173e-05
        [with_stream_mark]: 7.53999e-06
        [recompute_prepare]: 4.78001e-06
        [updatestate_depend_eliminate]: 2.39001e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 5.14e-06
        [updatestate_useless_node_eliminater]: 5.06997e-06
        [accelerated_algorithm]: 4.93001e-06
        [meta_shard_fg_expand]: 1.24e-06
        [get_grad_eliminate_]: 4.72998e-06
        [merge_forward]: 2.34999e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.161e-05
        [j_node_and_user_rematch]: 7.97e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 7.36999e-06
        [inline_without_move]: 4.89e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 5.02e-06
        [cse]: 9.08002e-06
        [replace_applicator]: 4.89e-06
[py_interpret_to_execute_after_opt_a]: 7.58001e-06
[rewriter_after_opt_a]: 2.28e-05
[convert_after_rewriter]: 6.20002e-06
[order_py_execute_after_rewriter]: 3.98001e-06
[mutable_eliminate]: 0.00048032
[jit_opt_b]: 5.756e-05, [1]
    [Cycle 1]: 4.212e-05, [2]
        [frontend_op_eliminate]: 1.611e-05
        [inline_after_opt_a]: 1.492e-05
[cconv]: 1.596e-05
[loop_unroll]: 0.0004222
[jit_opt_after_cconv]: 0.00013551, [1]
    [Cycle 1]: 0.00012973, [11]
        [c_1]: 2.299e-05
        [parameter_eliminate]: 2.40002e-06
        [updatestate_depend_eliminate]: 4.39002e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.555e-05
        [call_graph_tuple_transform]: 1.634e-05
        [tuple_list_get_item_eliminator]: 5.44e-06
        [none_parameter_eliminate]: 1.09003e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 5.47001e-06
[remove_dup_value]: 8.38999e-06
[partial_unused_args_eliminate]: 1.77001e-06
[environ_conv]: 3.91001e-06
[add_recomputation]: 3.353e-05
[cse_after_recomputation]: 2.157e-05, [1]
    [Cycle 1]: 1.587e-05, [1]
        [cse]: 9.20999e-06
[auto_monad_reorder]: 9.51e-06
[get_jit_bprop_graph]: 1.93002e-06
[rewriter_after_jit_bprop_graph]: 4.4e-06
[opt_after_jit_grad]: 0.000458
[symbol_engine_optimizer]: 7.31e-05, [1]
    [Cycle 1]: 6.727e-05, [6]
        [build]: 3.80998e-06
        [elim_shapecalc]: 8.00999e-06
        [elim_not_effective]: 1.198e-05
        [opt_reshape]: 5.52999e-06
        [fold_const_symbol]: 7.98001e-06
        [renormalize]: 1.11002e-06
[validate]: 2.758e-05
Sums
    bootstrap                            :     0.000274s :  2.85%
    type_inference                       :     0.005780s : 60.08%
    event_method                         :     0.000019s :  0.20%
    auto_monad                           :     0.000060s :  0.63%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000054s :  0.56%
    rewriter_before_opt_a                :     0.000078s :  0.81%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000097s :  1.01%
    jit_opt_a.loop_unroll                :     0.000041s :  0.42%
    jit_opt_a.a_1                        :     0.000770s :  8.00%
    jit_opt_a.with_stream_mark           :     0.000018s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000548s :  5.69%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.15%
    jit_opt_a.cse                        :     0.000024s :  0.25%
    jit_opt_a.replace_applicator         :     0.000015s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.08%
    rewriter_after_opt_a                 :     0.000023s :  0.24%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000480s :  4.99%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000016s :  0.17%
    loop_unroll                          :     0.000422s :  4.39%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000008s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000034s :  0.35%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000010s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000458s :  4.76%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000028s :  0.29%


Time group info:
------[substitution.]   0.000236    26
 0.80% :     0.000002s :      2: substitution.elim_not_effective
 0.38% :     0.000001s :      2: substitution.fold_const_symbol
 1.59% :     0.000004s :      3: substitution.graph_param_transform
86.35% :     0.000204s :      6: substitution.inline
 1.12% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.61% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.22% :     0.000003s :      2: substitution.replace_old_param
 3.16% :     0.000007s :      1: substitution.switch_simplify
 3.78% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005727     2
82.51% :     0.004725s :      1: type_inference.infer
17.49% :     0.001002s :      1: type_inference.specialize
------[replace.]   0.000133     9
59.02% :     0.000079s :      6: replace.inline
17.27% :     0.000023s :      1: replace.switch_simplify
23.71% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000215     9
93.32% :     0.000200s :      6: match.inline
 3.10% :     0.000007s :      1: match.switch_simplify
 3.58% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   801
 1.37% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.08% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.24% :     0.000002s :     12: predicate.addn_check_dump
 1.55% :     0.000002s :     12: predicate.addn_zero_filter
 1.83% :     0.000003s :     12: predicate.arithmetic_simplify
 1.40% :     0.000002s :     12: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     12: predicate.compare_switch_simplify
 1.55% :     0.000002s :     12: predicate.depend_value_elim
 1.24% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.45% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.40% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.69% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.22% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.39% :     0.000002s :     12: predicate.environ_get_eliminate
 1.26% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 4.87% :     0.000007s :     26: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.99% :     0.000001s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.85% :     0.000003s :     17: predicate.load_eliminater
 1.08% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.26% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.40% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.30% :     0.000002s :     12: predicate.merge_addn
 1.30% :     0.000002s :     12: predicate.minmaximum_grad
 1.29% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.01% :     0.000003s :     17: predicate.partial_eliminate
 1.39% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     12: predicate.reduce_eliminate
 1.60% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     20: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     12: predicate.reshape_eliminate
 1.29% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     12: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.84% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.71% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.86% :     0.000004s :     20: predicate.switch_defer_inline
 2.46% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.09% :     0.000011s :     65: predicate.switch_simplify
 1.41% :     0.000002s :     12: predicate.tile_eliminate
 1.32% :     0.000002s :     12: predicate.transpose_eliminate
 1.57% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.14% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.77% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     12: predicate.value_based_eliminate
 0.39% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000717    15
48.18% :     0.000345s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.42% :     0.000082s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.40% :     0.000290s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014741    72
 0.25% :     0.000036s :      1: add_recomputation
 0.43% :     0.000064s :      1: auto_monad
 0.08% :     0.000012s :      1: auto_monad_reorder
 1.96% :     0.000288s :      1: bootstrap
 0.13% :     0.000019s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
33.15% :     0.004887s :      1: jit_opt_a
 0.94% :     0.000138s :      1: jit_opt_after_cconv
 0.41% :     0.000060s :      1: jit_opt_b
 2.91% :     0.000429s :      1: loop_unroll
 3.31% :     0.000488s :      1: mutable_eliminate
 6.90% :     0.001017s :     26: opt.transform.jit_opt_a
 0.32% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.16% :     0.000465s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.38% :     0.000057s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000011s :      1: remove_dup_value
 1.94% :     0.000286s :      1: renormalize.infer
 1.74% :     0.000256s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000026s :      1: rewriter_after_opt_a
 0.55% :     0.000081s :      1: rewriter_before_opt_a
 0.52% :     0.000076s :      1: symbol_engine_optimizer
39.29% :     0.005791s :      1: type_inference


TotalTime = 0.0203796, [33]
[bootstrap]: 0.00038345
[type_inference]: 0.0047334
[event_method]: 1.686e-05
[auto_monad]: 6.594e-05
[graph_reusing]: 5.74999e-06
[pre_auto_parallel]: 2.33998e-06
[py_interpret_to_execute]: 3.979e-05
[rewriter_before_opt_a]: 7.055e-05
[expand_dump_flag]: 3.30998e-06
[jit_opt_a]: 0.00589624, [2]
    [Cycle 1]: 0.00194776, [27]
        [switch_simplify]: 6.452e-05
        [loop_unroll]: 2.98e-05
        [a_1]: 0.00058255
        [with_stream_mark]: 2.244e-05
        [recompute_prepare]: 8.01001e-06
        [updatestate_depend_eliminate]: 3.74002e-06
        [updatestate_assign_eliminate]: 3.19001e-06
        [updatestate_loads_eliminate]: 2.91e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 6.24001e-06
        [updatestate_useless_node_eliminater]: 5.25999e-06
        [accelerated_algorithm]: 5.63997e-06
        [meta_shard_fg_expand]: 1.95001e-06
        [get_grad_eliminate_]: 5.46e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.423e-05
        [j_node_and_user_rematch]: 9.66e-06
        [meta_fg_expand]: 3.06999e-06
        [replace_old_param]: 9.39e-06
        [inline_without_move]: 5.73002e-06
        [renormalize]: 0.00090545
        [add_forward_monad_depend]: 6.09999e-06
        [auto_monad_grad]: 2.14e-06
        [auto_monad_eliminator]: 1.484e-05
        [cse]: 2.977e-05
        [replace_applicator]: 1.452e-05
    [Cycle 2]: 0.0003329, [27]
        [switch_simplify]: 6.01998e-06
        [loop_unroll]: 5.09e-06
        [a_1]: 9.5e-05
        [with_stream_mark]: 1.147e-05
        [recompute_prepare]: 5.29e-06
        [updatestate_depend_eliminate]: 2.67001e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.19999e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 5.07e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 4.91997e-06
        [meta_shard_fg_expand]: 1.74e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.55002e-06
        [cell_reuse_recompute_pass]: 2.25002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.35e-05
        [j_node_and_user_rematch]: 8.55001e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 7.53e-06
        [inline_without_move]: 4.82998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.07998e-06
        [auto_monad_grad]: 1.05001e-06
        [auto_monad_eliminator]: 4.74002e-06
        [cse]: 1.177e-05
        [replace_applicator]: 4.90999e-06
[py_interpret_to_execute_after_opt_a]: 1.246e-05
[rewriter_after_opt_a]: 3.634e-05
[convert_after_rewriter]: 8.55999e-06
[order_py_execute_after_rewriter]: 4.86002e-06
[mutable_eliminate]: 0.0006645
[jit_opt_b]: 5.104e-05, [1]
    [Cycle 1]: 4.381e-05, [2]
        [frontend_op_eliminate]: 1.68e-05
        [inline_after_opt_a]: 1.492e-05
[cconv]: 2.883e-05
[loop_unroll]: 0.00053663
[jit_opt_after_cconv]: 0.00015201, [1]
    [Cycle 1]: 0.00014501, [11]
        [c_1]: 2.15e-05
        [parameter_eliminate]: 3.18e-06
        [updatestate_depend_eliminate]: 5.92001e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.55002e-06
        [cse]: 2.076e-05
        [call_graph_tuple_transform]: 2.072e-05
        [tuple_list_get_item_eliminator]: 5.81003e-06
        [none_parameter_eliminate]: 1.49998e-06
        [renormalize]: 8.39995e-07
        [switch_simplify]: 5.77999e-06
[remove_dup_value]: 1.439e-05
[partial_unused_args_eliminate]: 2.11e-06
[environ_conv]: 5.52001e-06
[add_recomputation]: 5.022e-05
[cse_after_recomputation]: 2.414e-05, [1]
    [Cycle 1]: 1.788e-05, [1]
        [cse]: 1.106e-05
[auto_monad_reorder]: 1.595e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 7.06999e-06
[opt_after_jit_grad]: 0.0004976
[symbol_engine_optimizer]: 7.735e-05, [1]
    [Cycle 1]: 7.091e-05, [6]
        [build]: 4.83001e-06
        [elim_shapecalc]: 8.09002e-06
        [elim_not_effective]: 1.384e-05
        [opt_reshape]: 5.44e-06
        [fold_const_symbol]: 9.42999e-06
        [renormalize]: 5.19998e-07
[validate]: 4.148e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.00670165
[execute]: 9.19e-06
Sums
    bootstrap                            :     0.000383s :  2.38%
    type_inference                       :     0.004733s : 29.37%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000066s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000040s :  0.25%
    rewriter_before_opt_a                :     0.000071s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000071s :  0.44%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000678s :  4.20%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000906s :  5.62%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000042s :  0.26%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000036s :  0.23%
    convert_after_rewriter               :     0.000009s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000664s :  4.12%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000029s :  0.18%
    loop_unroll                          :     0.000537s :  3.33%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000050s :  0.31%
    cse_after_recomputation.cse          :     0.000011s :  0.07%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000498s :  3.09%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006702s : 41.58%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000225    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.72% :     0.000006s :      3: substitution.graph_param_transform
83.81% :     0.000188s :      4: substitution.inline
 1.67% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.14% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.92% :     0.000004s :      2: substitution.replace_old_param
 6.16% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004661     2
79.22% :     0.003692s :      1: type_inference.infer
20.78% :     0.000969s :      1: type_inference.specialize
------[replace.]   0.000080     6
67.92% :     0.000055s :      4: replace.inline
32.08% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000199     6
93.61% :     0.000186s :      4: match.inline
 6.39% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.38% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.16% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.45% :     0.000001s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.35% :     0.000000s :      3: predicate.graph_param_transform
 5.21% :     0.000006s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000001s :      6: predicate.less_batch_normalization
 2.07% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.59% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.89% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000002s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     10: predicate.reduce_eliminate
 1.66% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.60% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.93% :     0.000010s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.67% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.13% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.05% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000648    12
46.55% :     0.000301s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.05% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.40% :     0.000313s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022292    76
 0.24% :     0.000053s :      1: add_recomputation
 0.31% :     0.000069s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.78% :     0.000397s :      1: bootstrap
 0.14% :     0.000032s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.12% :     0.000027s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.46% :     0.005899s :      1: jit_opt_a
 0.70% :     0.000155s :      1: jit_opt_after_cconv
 0.24% :     0.000054s :      1: jit_opt_b
 2.45% :     0.000547s :      1: loop_unroll
 3.03% :     0.000675s :      1: mutable_eliminate
 4.05% :     0.000903s :     26: opt.transform.jit_opt_a
 0.22% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.27% :     0.000506s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000043s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 2.20% :     0.000490s :      1: renormalize.infer
 1.82% :     0.000406s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000040s :      1: rewriter_after_opt_a
 0.33% :     0.000074s :      1: rewriter_before_opt_a
 0.36% :     0.000080s :      1: symbol_engine_optimizer
30.15% :     0.006720s :      1: task_emit
21.29% :     0.004746s :      1: type_inference
 0.29% :     0.000064s :      1: validate


TotalTime = 0.0159178, [33]
[bootstrap]: 0.00021183
[type_inference]: 0.0038864
[event_method]: 1.528e-05
[auto_monad]: 4.062e-05
[graph_reusing]: 3.55998e-06
[pre_auto_parallel]: 1.70001e-06
[py_interpret_to_execute]: 2.825e-05
[rewriter_before_opt_a]: 6.05e-05
[expand_dump_flag]: 2.56e-06
[jit_opt_a]: 0.00462883, [2]
    [Cycle 1]: 0.00133287, [27]
        [switch_simplify]: 4.45e-05
        [loop_unroll]: 2.906e-05
        [a_1]: 0.00047293
        [with_stream_mark]: 1.158e-05
        [recompute_prepare]: 7.05e-06
        [updatestate_depend_eliminate]: 2.85998e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 5.90002e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 1.18001e-06
        [get_grad_eliminate_]: 5.22999e-06
        [merge_forward]: 2.36998e-06
        [cell_reuse_recompute_pass]: 1.13001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.11e-05
        [j_node_and_user_rematch]: 8.95999e-06
        [meta_fg_expand]: 1.75001e-06
        [replace_old_param]: 7.87e-06
        [inline_without_move]: 5.96998e-06
        [renormalize]: 0.00048389
        [add_forward_monad_depend]: 3.85998e-06
        [auto_monad_grad]: 1.39e-06
        [auto_monad_eliminator]: 9.84001e-06
        [cse]: 1.738e-05
        [replace_applicator]: 1.115e-05
    [Cycle 2]: 0.00032998, [27]
        [switch_simplify]: 5.62999e-06
        [loop_unroll]: 4.79002e-06
        [a_1]: 9.345e-05
        [with_stream_mark]: 7.83001e-06
        [recompute_prepare]: 5.02e-06
        [updatestate_depend_eliminate]: 2.36e-06
        [updatestate_assign_eliminate]: 2.01998e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 5.07999e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 5.14e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.53e-06
        [cell_reuse_recompute_pass]: 1.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.238e-05
        [j_node_and_user_rematch]: 7.98999e-06
        [meta_fg_expand]: 1.39e-06
        [replace_old_param]: 7.49002e-06
        [inline_without_move]: 4.70001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 9.70002e-07
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 4.94e-06
        [cse]: 1.03e-05
        [replace_applicator]: 5.71998e-06
[py_interpret_to_execute_after_opt_a]: 8.39002e-06
[rewriter_after_opt_a]: 2.564e-05
[convert_after_rewriter]: 6.02001e-06
[order_py_execute_after_rewriter]: 4.39002e-06
[mutable_eliminate]: 0.00045854
[jit_opt_b]: 4.587e-05, [1]
    [Cycle 1]: 4.001e-05, [2]
        [frontend_op_eliminate]: 1.464e-05
        [inline_after_opt_a]: 1.458e-05
[cconv]: 1.432e-05
[loop_unroll]: 0.0004225
[jit_opt_after_cconv]: 0.00012951, [1]
    [Cycle 1]: 0.00012372, [11]
        [c_1]: 2.031e-05
        [parameter_eliminate]: 1.88002e-06
        [updatestate_depend_eliminate]: 5.00999e-06
        [updatestate_assign_eliminate]: 2.61999e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [cse]: 1.584e-05
        [call_graph_tuple_transform]: 1.57e-05
        [tuple_list_get_item_eliminator]: 5.18002e-06
        [none_parameter_eliminate]: 7.2e-07
        [renormalize]: 3.19997e-07
        [switch_simplify]: 5.10001e-06
[remove_dup_value]: 6.10002e-06
[partial_unused_args_eliminate]: 1.94e-06
[environ_conv]: 3.41999e-06
[add_recomputation]: 2.968e-05
[cse_after_recomputation]: 1.863e-05, [1]
    [Cycle 1]: 1.377e-05, [1]
        [cse]: 8.38999e-06
[auto_monad_reorder]: 8.74e-06
[get_jit_bprop_graph]: 1.29e-06
[rewriter_after_jit_bprop_graph]: 5.07e-06
[opt_after_jit_grad]: 0.00043772
[symbol_engine_optimizer]: 6.815e-05, [1]
    [Cycle 1]: 6.215e-05, [6]
        [build]: 2.32999e-06
        [elim_shapecalc]: 6.99001e-06
        [elim_not_effective]: 1.145e-05
        [opt_reshape]: 5.50001e-06
        [fold_const_symbol]: 8.38999e-06
        [renormalize]: 4.49974e-07
[validate]: 2.449e-05
[backend_pass]: 7.50006e-07
[task_emit]: 0.00515231
[execute]: 6.91001e-06
Sums
    bootstrap                            :     0.000212s :  1.71%
    type_inference                       :     0.003886s : 31.38%
    event_method                         :     0.000015s :  0.12%
    auto_monad                           :     0.000041s :  0.33%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.23%
    rewriter_before_opt_a                :     0.000060s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000050s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.27%
    jit_opt_a.a_1                        :     0.000566s :  4.57%
    jit_opt_a.with_stream_mark           :     0.000019s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.000484s :  3.91%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.12%
    jit_opt_a.cse                        :     0.000028s :  0.22%
    jit_opt_a.replace_applicator         :     0.000017s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.07%
    rewriter_after_opt_a                 :     0.000026s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000459s :  3.70%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000014s :  0.12%
    loop_unroll                          :     0.000422s :  3.41%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000006s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000003s :  0.03%
    add_recomputation                    :     0.000030s :  0.24%
    cse_after_recomputation.cse          :     0.000008s :  0.07%
    auto_monad_reorder                   :     0.000009s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000438s :  3.53%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000024s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005152s : 41.60%
    execute                              :     0.000007s :  0.06%


Time group info:
------[substitution.]   0.000149    23
 1.11% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.40% :     0.000004s :      3: substitution.graph_param_transform
84.01% :     0.000125s :      4: substitution.inline
 1.94% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.61% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.94% :     0.000003s :      2: substitution.replace_old_param
 5.38% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003839     2
81.58% :     0.003132s :      1: type_inference.infer
18.42% :     0.000707s :      1: type_inference.specialize
------[replace.]   0.000059     6
65.55% :     0.000039s :      4: replace.inline
34.45% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000129     6
94.71% :     0.000122s :      4: match.inline
 5.29% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.33% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 1.89% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.19% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000001s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.23% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000001s :     10: predicate.environ_get_eliminate
 1.26% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.09% :     0.000006s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000001s :      6: predicate.less_batch_normalization
 1.65% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.92% :     0.000002s :     15: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.23% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.16% :     0.000001s :     10: predicate.minmaximum_grad
 1.38% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000002s :     15: predicate.partial_eliminate
 1.30% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.74% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 2.31% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.88% :     0.000009s :     53: predicate.switch_simplify
 1.30% :     0.000001s :     10: predicate.tile_eliminate
 1.30% :     0.000001s :     10: predicate.transpose_eliminate
 1.61% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.99% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.86% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000464    12
45.84% :     0.000213s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.75% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.41% :     0.000225s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017266    76
 0.19% :     0.000032s :      1: add_recomputation
 0.26% :     0.000044s :      1: auto_monad
 0.06% :     0.000011s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.27% :     0.000220s :      1: bootstrap
 0.10% :     0.000017s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.03% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000003s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
26.83% :     0.004632s :      1: jit_opt_a
 0.77% :     0.000132s :      1: jit_opt_after_cconv
 0.28% :     0.000048s :      1: jit_opt_b
 2.49% :     0.000429s :      1: loop_unroll
 2.69% :     0.000465s :      1: mutable_eliminate
 4.42% :     0.000763s :     26: opt.transform.jit_opt_a
 0.25% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000010s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.58% :     0.000445s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000031s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000008s :      1: remove_dup_value
 1.43% :     0.000247s :      1: renormalize.infer
 1.34% :     0.000231s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000028s :      1: rewriter_after_opt_a
 0.37% :     0.000063s :      1: rewriter_before_opt_a
 0.41% :     0.000070s :      1: symbol_engine_optimizer
29.90% :     0.005162s :      1: task_emit
22.58% :     0.003898s :      1: type_inference
 0.23% :     0.000040s :      1: validate


TotalTime = 0.014924, [30]
[bootstrap]: 0.00024395
[type_inference]: 0.00611183
[event_method]: 2.132e-05
[auto_monad]: 7.627e-05
[graph_reusing]: 5.77999e-06
[pre_auto_parallel]: 2.24001e-06
[py_interpret_to_execute]: 6.513e-05
[rewriter_before_opt_a]: 8.176e-05
[expand_dump_flag]: 2.03997e-06
[jit_opt_a]: 0.00583556, [2]
    [Cycle 1]: 0.00197069, [27]
        [switch_simplify]: 0.00010801
        [loop_unroll]: 3.569e-05
        [a_1]: 0.00071993
        [with_stream_mark]: 1.468e-05
        [recompute_prepare]: 7.46001e-06
        [updatestate_depend_eliminate]: 3.29001e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 6.48e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.82999e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 5.48002e-06
        [merge_forward]: 3.04001e-06
        [cell_reuse_recompute_pass]: 1.06002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.285e-05
        [j_node_and_user_rematch]: 9.20999e-06
        [meta_fg_expand]: 2.71e-06
        [replace_old_param]: 8.17e-06
        [inline_without_move]: 5.70001e-06
        [renormalize]: 0.00076387
        [add_forward_monad_depend]: 4.93001e-06
        [auto_monad_grad]: 2.31e-06
        [auto_monad_eliminator]: 1.441e-05
        [cse]: 1.951e-05
        [replace_applicator]: 1.345e-05
    [Cycle 2]: 0.00033104, [27]
        [switch_simplify]: 5.59998e-06
        [loop_unroll]: 4.87998e-06
        [a_1]: 9.543e-05
        [with_stream_mark]: 1.038e-05
        [recompute_prepare]: 5.12e-06
        [updatestate_depend_eliminate]: 2.57001e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 5.17e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 1.48002e-06
        [get_grad_eliminate_]: 4.63999e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 1.43002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.492e-05
        [j_node_and_user_rematch]: 8.12e-06
        [meta_fg_expand]: 1.95001e-06
        [replace_old_param]: 8.09997e-06
        [inline_without_move]: 4.92999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.54e-06
        [auto_monad_grad]: 1.33002e-06
        [auto_monad_eliminator]: 5.27999e-06
        [cse]: 1.004e-05
        [replace_applicator]: 5.03002e-06
[py_interpret_to_execute_after_opt_a]: 1.233e-05
[rewriter_after_opt_a]: 3.254e-05
[convert_after_rewriter]: 6.65002e-06
[order_py_execute_after_rewriter]: 5.52001e-06
[mutable_eliminate]: 0.00066567
[jit_opt_b]: 6.968e-05, [1]
    [Cycle 1]: 6.146e-05, [2]
        [frontend_op_eliminate]: 1.842e-05
        [inline_after_opt_a]: 3.113e-05
[cconv]: 2.817e-05
[loop_unroll]: 0.00052325
[jit_opt_after_cconv]: 0.00016084, [1]
    [Cycle 1]: 0.00015316, [11]
        [c_1]: 2.359e-05
        [parameter_eliminate]: 3.85e-06
        [updatestate_depend_eliminate]: 7.24001e-06
        [updatestate_assign_eliminate]: 2.70002e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [cse]: 2.488e-05
        [call_graph_tuple_transform]: 2.136e-05
        [tuple_list_get_item_eliminator]: 6.12999e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 7.89994e-07
        [switch_simplify]: 6.07999e-06
[remove_dup_value]: 1.35e-05
[partial_unused_args_eliminate]: 2.48e-06
[environ_conv]: 6.11e-06
[add_recomputation]: 4.732e-05
[cse_after_recomputation]: 2.239e-05, [1]
    [Cycle 1]: 1.652e-05, [1]
        [cse]: 1.049e-05
[auto_monad_reorder]: 1.501e-05
[get_jit_bprop_graph]: 2.41e-06
[rewriter_after_jit_bprop_graph]: 7.87e-06
[opt_after_jit_grad]: 0.00056173
[symbol_engine_optimizer]: 7.977e-05, [1]
    [Cycle 1]: 7.277e-05, [6]
        [build]: 3.58e-06
        [elim_shapecalc]: 8.12998e-06
        [elim_not_effective]: 1.48e-05
        [opt_reshape]: 6.46e-06
        [fold_const_symbol]: 9.64e-06
        [renormalize]: 1.00001e-06
[validate]: 3.936e-05
Sums
    bootstrap                            :     0.000244s :  2.26%
    type_inference                       :     0.006112s : 56.69%
    event_method                         :     0.000021s :  0.20%
    auto_monad                           :     0.000076s :  0.71%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000065s :  0.60%
    rewriter_before_opt_a                :     0.000082s :  0.76%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000114s :  1.05%
    jit_opt_a.loop_unroll                :     0.000041s :  0.38%
    jit_opt_a.a_1                        :     0.000815s :  7.56%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000764s :  7.09%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.18%
    jit_opt_a.cse                        :     0.000030s :  0.27%
    jit_opt_a.replace_applicator         :     0.000018s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.11%
    rewriter_after_opt_a                 :     0.000033s :  0.30%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000666s :  6.17%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000031s :  0.29%
    cconv                                :     0.000028s :  0.26%
    loop_unroll                          :     0.000523s :  4.85%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000025s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000013s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000047s :  0.44%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.07%
    opt_after_jit_grad                   :     0.000562s :  5.21%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.37%


Time group info:
------[substitution.]   0.000275    26
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 1.97% :     0.000005s :      3: substitution.graph_param_transform
85.23% :     0.000235s :      6: substitution.inline
 1.01% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.71% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.35% :     0.000004s :      2: substitution.replace_old_param
 3.36% :     0.000009s :      1: substitution.switch_simplify
 4.08% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006050     2
82.23% :     0.004975s :      1: type_inference.infer
17.77% :     0.001075s :      1: type_inference.specialize
------[replace.]   0.000137     9
51.82% :     0.000071s :      6: replace.inline
22.51% :     0.000031s :      1: replace.switch_simplify
25.67% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000249     9
92.68% :     0.000231s :      6: match.inline
 3.36% :     0.000008s :      1: match.switch_simplify
 3.97% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000142   801
 1.38% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000002s :     12: predicate.addn_check_dump
 1.39% :     0.000002s :     12: predicate.addn_zero_filter
 2.20% :     0.000003s :     12: predicate.arithmetic_simplify
 1.23% :     0.000002s :     12: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000002s :     12: predicate.compare_switch_simplify
 1.69% :     0.000002s :     12: predicate.depend_value_elim
 1.28% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.15% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.11% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     12: predicate.environ_get_eliminate
 1.13% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.79% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.34% :     0.000008s :     26: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.77% :     0.000003s :     17: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.20% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.09% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000002s :     12: predicate.merge_addn
 1.21% :     0.000002s :     12: predicate.minmaximum_grad
 1.53% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     17: predicate.partial_eliminate
 1.25% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.87% :     0.000003s :     12: predicate.reduce_eliminate
 1.56% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.47% :     0.000002s :     20: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.61% :     0.000002s :     12: predicate.reshape_eliminate
 1.19% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.45% :     0.000002s :     12: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.61% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.67% :     0.000004s :     20: predicate.switch_defer_inline
 2.35% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.03% :     0.000011s :     65: predicate.switch_simplify
 1.35% :     0.000002s :     12: predicate.tile_eliminate
 1.43% :     0.000002s :     12: predicate.transpose_eliminate
 1.66% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.60% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.69% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.01% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000774    15
46.25% :     0.000358s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.14% :     0.000094s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.61% :     0.000322s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016860    72
 0.30% :     0.000050s :      1: add_recomputation
 0.48% :     0.000080s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 1.51% :     0.000255s :      1: bootstrap
 0.19% :     0.000031s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.02% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
34.63% :     0.005839s :      1: jit_opt_a
 0.97% :     0.000164s :      1: jit_opt_after_cconv
 0.43% :     0.000073s :      1: jit_opt_b
 3.15% :     0.000532s :      1: loop_unroll
 4.01% :     0.000676s :      1: mutable_eliminate
 6.46% :     0.001089s :     26: opt.transform.jit_opt_a
 0.32% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.25% :     0.000042s :      4: opt.transform.jit_opt_b
 0.09% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.39% :     0.000571s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.40% :     0.000068s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.56% :     0.000432s :      1: renormalize.infer
 1.92% :     0.000324s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000036s :      1: rewriter_after_opt_a
 0.50% :     0.000085s :      1: rewriter_before_opt_a
 0.49% :     0.000083s :      1: symbol_engine_optimizer
36.33% :     0.006125s :      1: type_inference


TotalTime = 0.0182855, [33]
[bootstrap]: 0.00028685
[type_inference]: 0.00436716
[event_method]: 1.702e-05
[auto_monad]: 6.274e-05
[graph_reusing]: 5.67999e-06
[pre_auto_parallel]: 2.55002e-06
[py_interpret_to_execute]: 3.06e-05
[rewriter_before_opt_a]: 7.129e-05
[expand_dump_flag]: 3.08998e-06
[jit_opt_a]: 0.00504677, [2]
    [Cycle 1]: 0.00154736, [27]
        [switch_simplify]: 5.372e-05
        [loop_unroll]: 3.043e-05
        [a_1]: 0.00051402
        [with_stream_mark]: 1.633e-05
        [recompute_prepare]: 7.39002e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [parameter_eliminate]: 1.77999e-06
        [specialize_transform]: 5.82001e-06
        [updatestate_useless_node_eliminater]: 5.14998e-06
        [accelerated_algorithm]: 5.34e-06
        [meta_shard_fg_expand]: 2.16e-06
        [get_grad_eliminate_]: 5.32001e-06
        [merge_forward]: 3.51001e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.42e-05
        [j_node_and_user_rematch]: 9.09e-06
        [meta_fg_expand]: 2.11998e-06
        [replace_old_param]: 9.12999e-06
        [inline_without_move]: 5.35999e-06
        [renormalize]: 0.0006173
        [add_forward_monad_depend]: 5.21002e-06
        [auto_monad_grad]: 1.89999e-06
        [auto_monad_eliminator]: 1.308e-05
        [cse]: 2.737e-05
        [replace_applicator]: 1.213e-05
    [Cycle 2]: 0.00031538, [27]
        [switch_simplify]: 5.46e-06
        [loop_unroll]: 4.95999e-06
        [a_1]: 8.831e-05
        [with_stream_mark]: 8.87e-06
        [recompute_prepare]: 4.70999e-06
        [updatestate_depend_eliminate]: 2.56e-06
        [updatestate_assign_eliminate]: 2.06998e-06
        [updatestate_loads_eliminate]: 1.83997e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.89998e-06
        [updatestate_useless_node_eliminater]: 4.77998e-06
        [accelerated_algorithm]: 5.14998e-06
        [meta_shard_fg_expand]: 1.16002e-06
        [get_grad_eliminate_]: 4.30999e-06
        [merge_forward]: 2.32001e-06
        [cell_reuse_recompute_pass]: 2.11e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.196e-05
        [j_node_and_user_rematch]: 7.85e-06
        [meta_fg_expand]: 1.82001e-06
        [replace_old_param]: 6.96001e-06
        [inline_without_move]: 4.95999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.49e-06
        [auto_monad_grad]: 6.50005e-07
        [auto_monad_eliminator]: 5.27001e-06
        [cse]: 9.56998e-06
        [replace_applicator]: 5.22e-06
[py_interpret_to_execute_after_opt_a]: 9.58002e-06
[rewriter_after_opt_a]: 3.359e-05
[convert_after_rewriter]: 6.66e-06
[order_py_execute_after_rewriter]: 5.62001e-06
[mutable_eliminate]: 0.0005578
[jit_opt_b]: 4.773e-05, [1]
    [Cycle 1]: 4.118e-05, [2]
        [frontend_op_eliminate]: 1.602e-05
        [inline_after_opt_a]: 1.437e-05
[cconv]: 2.281e-05
[loop_unroll]: 0.0004278
[jit_opt_after_cconv]: 0.0001379, [1]
    [Cycle 1]: 0.00013174, [11]
        [c_1]: 2.057e-05
        [parameter_eliminate]: 2.89999e-06
        [updatestate_depend_eliminate]: 5.54998e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.772e-05
        [call_graph_tuple_transform]: 1.828e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.35999e-06
        [renormalize]: 6.40022e-07
        [switch_simplify]: 5.19e-06
[remove_dup_value]: 1.38e-05
[partial_unused_args_eliminate]: 2.91999e-06
[environ_conv]: 5.27999e-06
[add_recomputation]: 4.805e-05
[cse_after_recomputation]: 1.926e-05, [1]
    [Cycle 1]: 1.385e-05, [1]
        [cse]: 8.2e-06
[auto_monad_reorder]: 1.404e-05
[get_jit_bprop_graph]: 2.01998e-06
[rewriter_after_jit_bprop_graph]: 4.80999e-06
[opt_after_jit_grad]: 0.00045998
[symbol_engine_optimizer]: 7.369e-05, [1]
    [Cycle 1]: 6.707e-05, [6]
        [build]: 3.73001e-06
        [elim_shapecalc]: 7.26001e-06
        [elim_not_effective]: 1.312e-05
        [opt_reshape]: 5.91e-06
        [fold_const_symbol]: 8.53001e-06
        [renormalize]: 4.40021e-07
[validate]: 3.739e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00626634
[execute]: 7.82e-06
Sums
    bootstrap                            :     0.000287s :  1.98%
    type_inference                       :     0.004367s : 30.09%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000063s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.21%
    rewriter_before_opt_a                :     0.000071s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.41%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000602s :  4.15%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000617s :  4.25%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000037s :  0.25%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000558s :  3.84%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000023s :  0.16%
    loop_unroll                          :     0.000428s :  2.95%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000048s :  0.33%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000460s :  3.17%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006266s : 43.18%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000191    23
 1.16% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.64% :     0.000005s :      3: substitution.graph_param_transform
83.53% :     0.000160s :      4: substitution.inline
 1.52% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000004s :      2: substitution.replace_old_param
 6.14% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004306     2
81.24% :     0.003498s :      1: type_inference.infer
18.76% :     0.000808s :      1: type_inference.specialize
------[replace.]   0.000065     6
66.76% :     0.000044s :      4: replace.inline
33.24% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000168     6
93.65% :     0.000157s :      4: match.inline
 6.35% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.32% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.39% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000001s :     10: predicate.addn_zero_filter
 2.10% :     0.000002s :     10: predicate.arithmetic_simplify
 1.12% :     0.000001s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000001s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.19% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.43% :     0.000000s :      3: predicate.graph_param_transform
 5.56% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000001s :      6: predicate.less_batch_normalization
 1.85% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.93% :     0.000002s :     15: predicate.load_eliminater
 1.76% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.20% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.42% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.29% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.85% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.84% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.01% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000528    12
49.25% :     0.000260s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.59% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.16% :     0.000238s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019817    76
 0.26% :     0.000051s :      1: add_recomputation
 0.33% :     0.000066s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.53% :     0.000303s :      1: bootstrap
 0.13% :     0.000025s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.48% :     0.005049s :      1: jit_opt_a
 0.71% :     0.000141s :      1: jit_opt_after_cconv
 0.25% :     0.000050s :      1: jit_opt_b
 2.20% :     0.000435s :      1: loop_unroll
 2.86% :     0.000567s :      1: mutable_eliminate
 4.08% :     0.000808s :     26: opt.transform.jit_opt_a
 0.23% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.36% :     0.000468s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.75% :     0.000347s :      1: renormalize.infer
 1.33% :     0.000263s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000039s :      1: rewriter_after_opt_a
 0.37% :     0.000074s :      1: rewriter_before_opt_a
 0.38% :     0.000076s :      1: symbol_engine_optimizer
31.68% :     0.006279s :      1: task_emit
22.09% :     0.004377s :      1: type_inference
 0.29% :     0.000058s :      1: validate


TotalTime = 0.0240576, [33]
[bootstrap]: 0.00037176
[type_inference]: 0.00569223
[event_method]: 2.457e-05
[auto_monad]: 7.394e-05
[graph_reusing]: 6.75998e-06
[pre_auto_parallel]: 2.97002e-06
[py_interpret_to_execute]: 5.424e-05
[rewriter_before_opt_a]: 8.294e-05
[expand_dump_flag]: 3.14001e-06
[jit_opt_a]: 0.00768789, [2]
    [Cycle 1]: 0.002262, [27]
        [switch_simplify]: 0.00010936
        [loop_unroll]: 2.946e-05
        [a_1]: 0.00078917
        [with_stream_mark]: 2.691e-05
        [recompute_prepare]: 9.34998e-06
        [updatestate_depend_eliminate]: 4.43999e-06
        [updatestate_assign_eliminate]: 3.02002e-06
        [updatestate_loads_eliminate]: 2.41998e-06
        [parameter_eliminate]: 2.01e-06
        [specialize_transform]: 6.23e-06
        [updatestate_useless_node_eliminater]: 5.23002e-06
        [accelerated_algorithm]: 5.82001e-06
        [meta_shard_fg_expand]: 2.32999e-06
        [get_grad_eliminate_]: 5.64e-06
        [merge_forward]: 4.13001e-06
        [cell_reuse_recompute_pass]: 1.80001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.675e-05
        [j_node_and_user_rematch]: 1.057e-05
        [meta_fg_expand]: 2.73998e-06
        [replace_old_param]: 1.04e-05
        [inline_without_move]: 5.25999e-06
        [renormalize]: 0.00094013
        [add_forward_monad_depend]: 6.53e-06
        [auto_monad_grad]: 2.63998e-06
        [auto_monad_eliminator]: 1.644e-05
        [cse]: 3.074e-05
        [replace_applicator]: 1.63e-05
    [Cycle 2]: 0.00032349, [27]
        [switch_simplify]: 5.57001e-06
        [loop_unroll]: 4.41002e-06
        [a_1]: 9.234e-05
        [with_stream_mark]: 1.067e-05
        [recompute_prepare]: 4.95999e-06
        [updatestate_depend_eliminate]: 2.63998e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 5.20999e-06
        [updatestate_useless_node_eliminater]: 4.53999e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 1.40999e-06
        [get_grad_eliminate_]: 4.68999e-06
        [merge_forward]: 2.99001e-06
        [cell_reuse_recompute_pass]: 2.82002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.245e-05
        [j_node_and_user_rematch]: 8.06001e-06
        [meta_fg_expand]: 1.69998e-06
        [replace_old_param]: 7.31999e-06
        [inline_without_move]: 4.55001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.56998e-06
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 5.02e-06
        [cse]: 1.06e-05
        [replace_applicator]: 4.99998e-06
[py_interpret_to_execute_after_opt_a]: 1.406e-05
[rewriter_after_opt_a]: 3.778e-05
[convert_after_rewriter]: 7.59002e-06
[order_py_execute_after_rewriter]: 4.81002e-06
[mutable_eliminate]: 0.00080174
[jit_opt_b]: 5.305e-05, [1]
    [Cycle 1]: 4.523e-05, [2]
        [frontend_op_eliminate]: 1.811e-05
        [inline_after_opt_a]: 1.525e-05
[cconv]: 3.201e-05
[loop_unroll]: 0.00045895
[jit_opt_after_cconv]: 0.00017082, [1]
    [Cycle 1]: 0.00016323, [11]
        [c_1]: 3.503e-05
        [parameter_eliminate]: 3.92998e-06
        [updatestate_depend_eliminate]: 7.56999e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.56998e-06
        [cse]: 2.387e-05
        [call_graph_tuple_transform]: 2.035e-05
        [tuple_list_get_item_eliminator]: 5.50001e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 8.30012e-07
        [switch_simplify]: 5.77999e-06
[remove_dup_value]: 1.585e-05
[partial_unused_args_eliminate]: 2.66999e-06
[environ_conv]: 6.28e-06
[add_recomputation]: 5.617e-05
[cse_after_recomputation]: 2.138e-05, [1]
    [Cycle 1]: 1.582e-05, [1]
        [cse]: 9.31002e-06
[auto_monad_reorder]: 1.576e-05
[get_jit_bprop_graph]: 2.20002e-06
[rewriter_after_jit_bprop_graph]: 8.37e-06
[opt_after_jit_grad]: 0.00055548
[symbol_engine_optimizer]: 8.653e-05, [1]
    [Cycle 1]: 7.773e-05, [6]
        [build]: 6.12001e-06
        [elim_shapecalc]: 9.50001e-06
        [elim_not_effective]: 1.496e-05
        [opt_reshape]: 6.36e-06
        [fold_const_symbol]: 1.022e-05
        [renormalize]: 9.5999e-07
[validate]: 4.896e-05
[backend_pass]: 8.2e-07
[task_emit]: 0.00739443
[execute]: 1.076e-05
Sums
    bootstrap                            :     0.000372s :  2.04%
    type_inference                       :     0.005692s : 31.17%
    event_method                         :     0.000025s :  0.13%
    auto_monad                           :     0.000074s :  0.40%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000054s :  0.30%
    rewriter_before_opt_a                :     0.000083s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000115s :  0.63%
    jit_opt_a.loop_unroll                :     0.000034s :  0.19%
    jit_opt_a.a_1                        :     0.000882s :  4.83%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000940s :  5.15%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.12%
    jit_opt_a.cse                        :     0.000041s :  0.23%
    jit_opt_a.replace_applicator         :     0.000021s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000802s :  4.39%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000032s :  0.18%
    loop_unroll                          :     0.000459s :  2.51%
    jit_opt_after_cconv.c_1              :     0.000035s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000056s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000555s :  3.04%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000049s :  0.27%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007394s : 40.49%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000274    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000002s :      2: substitution.fold_const_symbol
 2.05% :     0.000006s :      3: substitution.graph_param_transform
85.91% :     0.000235s :      4: substitution.inline
 1.65% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.67% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.10% :     0.000006s :      2: substitution.replace_old_param
 5.06% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005599     2
81.24% :     0.004549s :      1: type_inference.infer
18.76% :     0.001051s :      1: type_inference.specialize
------[replace.]   0.000080     6
67.42% :     0.000054s :      4: replace.inline
32.58% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000245     6
94.90% :     0.000233s :      4: match.inline
 5.10% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.91% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.46% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000002s :     10: predicate.addn_zero_filter
 2.29% :     0.000003s :     10: predicate.arithmetic_simplify
 1.16% :     0.000001s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.77% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.34% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.53% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     10: predicate.environ_get_depend_swap
 2.10% :     0.000003s :     10: predicate.environ_get_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.28% :     0.000002s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.29% :     0.000007s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.55% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.83% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.13% :     0.000003s :      3: predicate.mutable_eliminate
 0.64% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     15: predicate.partial_eliminate
 1.02% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.36% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.60% :     0.000002s :     10: predicate.reshape_eliminate
 1.39% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.04% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.08% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.18% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.27% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.38% :     0.000009s :     53: predicate.switch_simplify
 1.94% :     0.000002s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.73% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.59% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000773    12
47.99% :     0.000371s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.85% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.16% :     0.000349s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026271    76
 0.23% :     0.000059s :      1: add_recomputation
 0.30% :     0.000078s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.49% :     0.000391s :      1: bootstrap
 0.13% :     0.000035s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.11% :     0.000029s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
29.28% :     0.007692s :      1: jit_opt_a
 0.66% :     0.000174s :      1: jit_opt_after_cconv
 0.21% :     0.000056s :      1: jit_opt_b
 1.78% :     0.000467s :      1: loop_unroll
 3.09% :     0.000812s :      1: mutable_eliminate
 4.39% :     0.001154s :     26: opt.transform.jit_opt_a
 0.24% :     0.000063s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.16% :     0.000568s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000058s :      1: py_interpret_to_execute
 0.06% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 1.97% :     0.000517s :      1: renormalize.infer
 1.57% :     0.000413s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000041s :      1: rewriter_after_opt_a
 0.33% :     0.000087s :      1: rewriter_before_opt_a
 0.34% :     0.000089s :      1: symbol_engine_optimizer
28.23% :     0.007417s :      1: task_emit
21.76% :     0.005716s :      1: type_inference
 0.29% :     0.000075s :      1: validate


TotalTime = 0.0230325, [33]
[bootstrap]: 0.00028541
[type_inference]: 0.00438251
[event_method]: 1.644e-05
[auto_monad]: 6.102e-05
[graph_reusing]: 6.02001e-06
[pre_auto_parallel]: 2.63998e-06
[py_interpret_to_execute]: 3.014e-05
[rewriter_before_opt_a]: 7.141e-05
[expand_dump_flag]: 3.51999e-06
[jit_opt_a]: 0.00482649, [2]
    [Cycle 1]: 0.00143699, [27]
        [switch_simplify]: 4.9e-05
        [loop_unroll]: 2.954e-05
        [a_1]: 0.00050211
        [with_stream_mark]: 1.536e-05
        [recompute_prepare]: 7.01999e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.68998e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 5.99e-06
        [updatestate_useless_node_eliminater]: 5.13002e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 1.91003e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 3.86999e-06
        [cell_reuse_recompute_pass]: 1.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.396e-05
        [j_node_and_user_rematch]: 8.87e-06
        [meta_fg_expand]: 2.17001e-06
        [replace_old_param]: 8.27e-06
        [inline_without_move]: 5.17e-06
        [renormalize]: 0.00053171
        [add_forward_monad_depend]: 5.09e-06
        [auto_monad_grad]: 2.41998e-06
        [auto_monad_eliminator]: 1.329e-05
        [cse]: 2.65e-05
        [replace_applicator]: 1.123e-05
    [Cycle 2]: 0.00031641, [27]
        [switch_simplify]: 5.61e-06
        [loop_unroll]: 4.85001e-06
        [a_1]: 8.79e-05
        [with_stream_mark]: 8.54998e-06
        [recompute_prepare]: 5.32001e-06
        [updatestate_depend_eliminate]: 2.49001e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 4.87998e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.22999e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.51e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.224e-05
        [j_node_and_user_rematch]: 7.91001e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 7.38999e-06
        [inline_without_move]: 4.90001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.28002e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 5.22e-06
        [cse]: 1.02e-05
        [replace_applicator]: 5.86998e-06
[py_interpret_to_execute_after_opt_a]: 8.45001e-06
[rewriter_after_opt_a]: 3.264e-05
[convert_after_rewriter]: 6.64999e-06
[order_py_execute_after_rewriter]: 4.47e-06
[mutable_eliminate]: 0.00049242
[jit_opt_b]: 4.882e-05, [1]
    [Cycle 1]: 4.244e-05, [2]
        [frontend_op_eliminate]: 1.63e-05
        [inline_after_opt_a]: 1.46e-05
[cconv]: 2.275e-05
[loop_unroll]: 0.00041353
[jit_opt_after_cconv]: 0.00013494, [1]
    [Cycle 1]: 0.00012894, [11]
        [c_1]: 2.019e-05
        [parameter_eliminate]: 2.63003e-06
        [updatestate_depend_eliminate]: 5.08002e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 1.581e-05
        [call_graph_tuple_transform]: 1.804e-05
        [tuple_list_get_item_eliminator]: 5.32999e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 5.19998e-07
        [switch_simplify]: 5.39998e-06
[remove_dup_value]: 1.346e-05
[partial_unused_args_eliminate]: 2.27001e-06
[environ_conv]: 4.42998e-06
[add_recomputation]: 4.451e-05
[cse_after_recomputation]: 2.043e-05, [1]
    [Cycle 1]: 1.528e-05, [1]
        [cse]: 9.44998e-06
[auto_monad_reorder]: 0.00515866
[get_jit_bprop_graph]: 2.55002e-06
[rewriter_after_jit_bprop_graph]: 9.49e-06
[opt_after_jit_grad]: 0.00063853
[symbol_engine_optimizer]: 7.9e-05, [1]
    [Cycle 1]: 7.255e-05, [6]
        [build]: 3.89002e-06
        [elim_shapecalc]: 8.27e-06
        [elim_not_effective]: 1.644e-05
        [opt_reshape]: 6.22001e-06
        [fold_const_symbol]: 8.89e-06
        [renormalize]: 6.09987e-07
[validate]: 3.795e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.00597116
[execute]: 7.93001e-06
Sums
    bootstrap                            :     0.000285s :  1.47%
    type_inference                       :     0.004383s : 22.63%
    event_method                         :     0.000016s :  0.08%
    auto_monad                           :     0.000061s :  0.32%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000030s :  0.16%
    rewriter_before_opt_a                :     0.000071s :  0.37%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.28%
    jit_opt_a.loop_unroll                :     0.000034s :  0.18%
    jit_opt_a.a_1                        :     0.000590s :  3.05%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000532s :  2.75%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.10%
    jit_opt_a.cse                        :     0.000037s :  0.19%
    jit_opt_a.replace_applicator         :     0.000017s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.04%
    rewriter_after_opt_a                 :     0.000033s :  0.17%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000492s :  2.54%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000023s :  0.12%
    loop_unroll                          :     0.000414s :  2.14%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000045s :  0.23%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.005159s : 26.63%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000639s :  3.30%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005971s : 30.83%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000183    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.93% :     0.000002s :      2: substitution.fold_const_symbol
 2.97% :     0.000005s :      3: substitution.graph_param_transform
82.77% :     0.000151s :      4: substitution.inline
 1.68% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.34% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.90% :     0.000003s :      2: substitution.replace_old_param
 6.40% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004322     2
82.23% :     0.003554s :      1: type_inference.infer
17.77% :     0.000768s :      1: type_inference.specialize
------[replace.]   0.000064     6
66.48% :     0.000042s :      4: replace.inline
33.52% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000159     6
93.35% :     0.000149s :      4: match.inline
 6.65% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.31% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.00% :     0.000002s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000001s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.11% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000000s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.78% :     0.000007s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.66% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.76% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.88% :     0.000002s :      3: predicate.mutable_eliminate
 0.64% :     0.000001s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.84% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000001s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.87% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.91% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000520    12
50.55% :     0.000263s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.22% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.23% :     0.000230s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024470    76
 0.19% :     0.000047s :      1: add_recomputation
 0.26% :     0.000064s :      1: auto_monad
21.13% :     0.005171s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.22% :     0.000299s :      1: bootstrap
 0.10% :     0.000025s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.08% :     0.000020s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000006s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
19.74% :     0.004829s :      1: jit_opt_a
 0.56% :     0.000138s :      1: jit_opt_after_cconv
 0.21% :     0.000051s :      1: jit_opt_b
 1.72% :     0.000420s :      1: loop_unroll
 2.04% :     0.000500s :      1: mutable_eliminate
 3.23% :     0.000791s :     26: opt.transform.jit_opt_a
 0.19% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.65% :     0.000647s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.13% :     0.000033s :      1: py_interpret_to_execute
 0.04% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000016s :      1: remove_dup_value
 1.10% :     0.000269s :      1: renormalize.infer
 1.05% :     0.000256s :      1: renormalize.specialize
 0.05% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000035s :      1: rewriter_after_opt_a
 0.30% :     0.000074s :      1: rewriter_before_opt_a
 0.33% :     0.000082s :      1: symbol_engine_optimizer
24.44% :     0.005981s :      1: task_emit
17.95% :     0.004393s :      1: type_inference
 0.25% :     0.000062s :      1: validate


TotalTime = 0.0142611, [30]
[bootstrap]: 0.00027489
[type_inference]: 0.00617469
[event_method]: 1.957e-05
[auto_monad]: 7.555e-05
[graph_reusing]: 5.47999e-06
[pre_auto_parallel]: 2.87002e-06
[py_interpret_to_execute]: 5.243e-05
[rewriter_before_opt_a]: 8.399e-05
[expand_dump_flag]: 3.25998e-06
[jit_opt_a]: 0.00547085, [2]
    [Cycle 1]: 0.00192405, [27]
        [switch_simplify]: 0.00011413
        [loop_unroll]: 4.007e-05
        [a_1]: 0.00074384
        [with_stream_mark]: 1.566e-05
        [recompute_prepare]: 7.2e-06
        [updatestate_depend_eliminate]: 3.19001e-06
        [updatestate_assign_eliminate]: 2.58998e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 6.36e-06
        [updatestate_useless_node_eliminater]: 5.66998e-06
        [accelerated_algorithm]: 6.29001e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 5.44e-06
        [merge_forward]: 4.25999e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.313e-05
        [j_node_and_user_rematch]: 9.01998e-06
        [meta_fg_expand]: 2.29999e-06
        [replace_old_param]: 9.27001e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 0.00070401
        [add_forward_monad_depend]: 4.95001e-06
        [auto_monad_grad]: 2.44999e-06
        [auto_monad_eliminator]: 1.339e-05
        [cse]: 1.781e-05
        [replace_applicator]: 1.192e-05
    [Cycle 2]: 0.00031583, [27]
        [switch_simplify]: 5.95002e-06
        [loop_unroll]: 5.11997e-06
        [a_1]: 9.027e-05
        [with_stream_mark]: 8.85999e-06
        [recompute_prepare]: 5.20999e-06
        [updatestate_depend_eliminate]: 2.61999e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 4.92999e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 1.18001e-06
        [get_grad_eliminate_]: 4.79e-06
        [merge_forward]: 3.08e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.272e-05
        [j_node_and_user_rematch]: 8.02e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 7.18e-06
        [inline_without_move]: 4.72998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 4.81002e-06
        [cse]: 1.004e-05
        [replace_applicator]: 5.25999e-06
[py_interpret_to_execute_after_opt_a]: 1.035e-05
[rewriter_after_opt_a]: 3.21e-05
[convert_after_rewriter]: 6.34001e-06
[order_py_execute_after_rewriter]: 5.32999e-06
[mutable_eliminate]: 0.00054328
[jit_opt_b]: 6.123e-05, [1]
    [Cycle 1]: 5.495e-05, [2]
        [frontend_op_eliminate]: 1.798e-05
        [inline_after_opt_a]: 2.497e-05
[cconv]: 2.28e-05
[loop_unroll]: 0.00041872
[jit_opt_after_cconv]: 0.00013781, [1]
    [Cycle 1]: 0.00013151, [11]
        [c_1]: 2.079e-05
        [parameter_eliminate]: 2.79001e-06
        [updatestate_depend_eliminate]: 5.22e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.582e-05
        [call_graph_tuple_transform]: 1.872e-05
        [tuple_list_get_item_eliminator]: 5.57001e-06
        [none_parameter_eliminate]: 1.60001e-06
        [renormalize]: 4.20026e-07
        [switch_simplify]: 5.87999e-06
[remove_dup_value]: 1.172e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 4.60999e-06
[add_recomputation]: 4.012e-05
[cse_after_recomputation]: 2.107e-05, [1]
    [Cycle 1]: 1.54e-05, [1]
        [cse]: 9.74e-06
[auto_monad_reorder]: 2.184e-05
[get_jit_bprop_graph]: 1.91e-06
[rewriter_after_jit_bprop_graph]: 4.97e-06
[opt_after_jit_grad]: 0.00047132
[symbol_engine_optimizer]: 7.475e-05, [1]
    [Cycle 1]: 6.874e-05, [6]
        [build]: 3.98999e-06
        [elim_shapecalc]: 7.89002e-06
        [elim_not_effective]: 1.332e-05
        [opt_reshape]: 5.75001e-06
        [fold_const_symbol]: 8.64e-06
        [renormalize]: 6.39993e-07
[validate]: 3.134e-05
Sums
    bootstrap                            :     0.000275s :  2.63%
    type_inference                       :     0.006175s : 59.06%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000076s :  0.72%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000052s :  0.50%
    rewriter_before_opt_a                :     0.000084s :  0.80%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000120s :  1.15%
    jit_opt_a.loop_unroll                :     0.000045s :  0.43%
    jit_opt_a.a_1                        :     0.000834s :  7.98%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000704s :  6.73%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.17%
    jit_opt_a.cse                        :     0.000028s :  0.27%
    jit_opt_a.replace_applicator         :     0.000017s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.10%
    rewriter_after_opt_a                 :     0.000032s :  0.31%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000543s :  5.20%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.24%
    cconv                                :     0.000023s :  0.22%
    loop_unroll                          :     0.000419s :  4.00%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000012s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000040s :  0.38%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000022s :  0.21%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000471s :  4.51%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000031s :  0.30%


Time group info:
------[substitution.]   0.000280    26
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.41% :     0.000001s :      2: substitution.fold_const_symbol
 1.80% :     0.000005s :      3: substitution.graph_param_transform
85.88% :     0.000241s :      6: substitution.inline
 1.05% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.52% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.46% :     0.000004s :      2: substitution.replace_old_param
 2.88% :     0.000008s :      1: substitution.switch_simplify
 4.28% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006117     2
82.69% :     0.005058s :      1: type_inference.infer
17.31% :     0.001059s :      1: type_inference.specialize
------[replace.]   0.000137     9
51.66% :     0.000071s :      6: replace.inline
23.79% :     0.000033s :      1: replace.switch_simplify
24.54% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000255     9
92.97% :     0.000237s :      6: match.inline
 2.84% :     0.000007s :      1: match.switch_simplify
 4.20% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000141   801
 1.31% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000002s :     12: predicate.addn_check_dump
 1.47% :     0.000002s :     12: predicate.addn_zero_filter
 2.19% :     0.000003s :     12: predicate.arithmetic_simplify
 1.40% :     0.000002s :     12: predicate.cast_eliminate
 0.64% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     12: predicate.compare_switch_simplify
 1.47% :     0.000002s :     12: predicate.depend_value_elim
 1.25% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.39% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.13% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     12: predicate.environ_get_eliminate
 1.27% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.19% :     0.000007s :     26: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000002s :      6: predicate.less_batch_normalization
 1.65% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     17: predicate.load_eliminater
 1.22% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.53% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.92% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000002s :     12: predicate.merge_addn
 1.09% :     0.000002s :     12: predicate.minmaximum_grad
 1.51% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 1.99% :     0.000003s :     17: predicate.partial_eliminate
 1.13% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     12: predicate.reduce_eliminate
 1.41% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     20: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     12: predicate.reshape_eliminate
 1.31% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.63% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.27% :     0.000002s :      6: predicate.special_op_eliminate
 0.84% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.08% :     0.000004s :     20: predicate.switch_defer_inline
 2.43% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.85% :     0.000013s :     65: predicate.switch_simplify
 1.39% :     0.000002s :     12: predicate.tile_eliminate
 1.12% :     0.000002s :     12: predicate.transpose_eliminate
 1.61% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.17% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.64% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     12: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000788    15
47.93% :     0.000378s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.41% :     0.000098s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.66% :     0.000312s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016155    72
 0.27% :     0.000043s :      1: add_recomputation
 0.49% :     0.000079s :      1: auto_monad
 0.16% :     0.000025s :      1: auto_monad_reorder
 1.77% :     0.000286s :      1: bootstrap
 0.16% :     0.000025s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.10% :     0.000016s :      1: graph_reusing
33.89% :     0.005474s :      1: jit_opt_a
 0.87% :     0.000141s :      1: jit_opt_after_cconv
 0.40% :     0.000064s :      1: jit_opt_b
 2.64% :     0.000426s :      1: loop_unroll
 3.41% :     0.000551s :      1: mutable_eliminate
 6.90% :     0.001115s :     26: opt.transform.jit_opt_a
 0.29% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.23% :     0.000037s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.96% :     0.000479s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.34% :     0.000055s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000014s :      1: remove_dup_value
 2.46% :     0.000398s :      1: renormalize.infer
 1.85% :     0.000299s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000035s :      1: rewriter_after_opt_a
 0.54% :     0.000087s :      1: rewriter_before_opt_a
 0.48% :     0.000078s :      1: symbol_engine_optimizer
38.33% :     0.006192s :      1: type_inference


TotalTime = 0.023214, [33]
[bootstrap]: 0.00034183
[type_inference]: 0.0049873
[event_method]: 2.03e-05
[auto_monad]: 7.333e-05
[graph_reusing]: 6.50002e-06
[pre_auto_parallel]: 3.14001e-06
[py_interpret_to_execute]: 4.927e-05
[rewriter_before_opt_a]: 8.056e-05
[expand_dump_flag]: 3.35e-06
[jit_opt_a]: 0.00739469, [2]
    [Cycle 1]: 0.00216327, [27]
        [switch_simplify]: 7.289e-05
        [loop_unroll]: 2.992e-05
        [a_1]: 0.00064477
        [with_stream_mark]: 2.482e-05
        [recompute_prepare]: 9.44998e-06
        [updatestate_depend_eliminate]: 4.70999e-06
        [updatestate_assign_eliminate]: 2.88e-06
        [updatestate_loads_eliminate]: 2.78e-06
        [parameter_eliminate]: 2.11e-06
        [specialize_transform]: 8.17e-06
        [updatestate_useless_node_eliminater]: 6.14001e-06
        [accelerated_algorithm]: 6.42001e-06
        [meta_shard_fg_expand]: 3.16001e-06
        [get_grad_eliminate_]: 7.35998e-06
        [merge_forward]: 4.22e-06
        [cell_reuse_recompute_pass]: 1.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.708e-05
        [j_node_and_user_rematch]: 1.14e-05
        [meta_fg_expand]: 2.51e-06
        [replace_old_param]: 1.146e-05
        [inline_without_move]: 6.14999e-06
        [renormalize]: 0.00099494
        [add_forward_monad_depend]: 8.45999e-06
        [auto_monad_grad]: 2.80002e-06
        [auto_monad_eliminator]: 1.769e-05
        [cse]: 3.146e-05
        [replace_applicator]: 1.783e-05
    [Cycle 2]: 0.00044891, [27]
        [switch_simplify]: 5.72001e-06
        [loop_unroll]: 5.37001e-06
        [a_1]: 0.00016118
        [with_stream_mark]: 1.532e-05
        [recompute_prepare]: 6.40997e-06
        [updatestate_depend_eliminate]: 3.85e-06
        [updatestate_assign_eliminate]: 2.71999e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [parameter_eliminate]: 1.92001e-06
        [specialize_transform]: 6.44999e-06
        [updatestate_useless_node_eliminater]: 5.35999e-06
        [accelerated_algorithm]: 6.73e-06
        [meta_shard_fg_expand]: 2.35002e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 3.14001e-06
        [cell_reuse_recompute_pass]: 4.28999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.767e-05
        [j_node_and_user_rematch]: 8.62998e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 8.70001e-06
        [inline_without_move]: 5.35999e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.74998e-06
        [auto_monad_grad]: 1.48002e-06
        [auto_monad_eliminator]: 6.73998e-06
        [cse]: 1.665e-05
        [replace_applicator]: 1.549e-05
[py_interpret_to_execute_after_opt_a]: 2.798e-05
[rewriter_after_opt_a]: 4.838e-05
[convert_after_rewriter]: 8.50001e-06
[order_py_execute_after_rewriter]: 5.25001e-06
[mutable_eliminate]: 0.00091453
[jit_opt_b]: 7.238e-05, [1]
    [Cycle 1]: 6.013e-05, [2]
        [frontend_op_eliminate]: 2.799e-05
        [inline_after_opt_a]: 1.8e-05
[cconv]: 3.766e-05
[loop_unroll]: 0.00063295
[jit_opt_after_cconv]: 0.00017806, [1]
    [Cycle 1]: 0.00016975, [11]
        [c_1]: 2.492e-05
        [parameter_eliminate]: 5.32999e-06
        [updatestate_depend_eliminate]: 9.71e-06
        [updatestate_assign_eliminate]: 2.71999e-06
        [updatestate_loads_eliminate]: 2.40002e-06
        [cse]: 3.029e-05
        [call_graph_tuple_transform]: 2.282e-05
        [tuple_list_get_item_eliminator]: 6.49001e-06
        [none_parameter_eliminate]: 1.75001e-06
        [renormalize]: 1.05001e-06
        [switch_simplify]: 5.52999e-06
[remove_dup_value]: 1.814e-05
[partial_unused_args_eliminate]: 2.81999e-06
[environ_conv]: 7.06999e-06
[add_recomputation]: 5.899e-05
[cse_after_recomputation]: 2.583e-05, [1]
    [Cycle 1]: 1.793e-05, [1]
        [cse]: 1.002e-05
[auto_monad_reorder]: 1.588e-05
[get_jit_bprop_graph]: 2.43e-06
[rewriter_after_jit_bprop_graph]: 9.69e-06
[opt_after_jit_grad]: 0.0006437
[symbol_engine_optimizer]: 8.444e-05, [1]
    [Cycle 1]: 7.547e-05, [6]
        [build]: 4.80001e-06
        [elim_shapecalc]: 8.15999e-06
        [elim_not_effective]: 1.636e-05
        [opt_reshape]: 6.92002e-06
        [fold_const_symbol]: 7.76001e-06
        [renormalize]: 1.04e-06
[validate]: 4.731e-05
[backend_pass]: 8.30012e-07
[task_emit]: 0.00713984
[execute]: 8.32998e-06
Sums
    bootstrap                            :     0.000342s :  1.93%
    type_inference                       :     0.004987s : 28.20%
    event_method                         :     0.000020s :  0.11%
    auto_monad                           :     0.000073s :  0.41%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000049s :  0.28%
    rewriter_before_opt_a                :     0.000081s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000079s :  0.44%
    jit_opt_a.loop_unroll                :     0.000035s :  0.20%
    jit_opt_a.a_1                        :     0.000806s :  4.56%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000015s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.11%
    jit_opt_a.inline_without_move        :     0.000012s :  0.07%
    jit_opt_a.renormalize                :     0.000995s :  5.63%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.14%
    jit_opt_a.cse                        :     0.000048s :  0.27%
    jit_opt_a.replace_applicator         :     0.000033s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000028s :  0.16%
    rewriter_after_opt_a                 :     0.000048s :  0.27%
    convert_after_rewriter               :     0.000009s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000915s :  5.17%
    jit_opt_b.frontend_op_eliminate      :     0.000028s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.10%
    cconv                                :     0.000038s :  0.21%
    loop_unroll                          :     0.000633s :  3.58%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000030s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000018s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000059s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.05%
    opt_after_jit_grad                   :     0.000644s :  3.64%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000047s :  0.27%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007140s : 40.37%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000273    23
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.19% :     0.000006s :      3: substitution.graph_param_transform
85.46% :     0.000233s :      4: substitution.inline
 1.58% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.12% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000005s :      2: substitution.replace_old_param
 5.32% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004896     2
79.70% :     0.003902s :      1: type_inference.infer
20.30% :     0.000994s :      1: type_inference.specialize
------[replace.]   0.000084     6
66.33% :     0.000056s :      4: replace.inline
33.67% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000244     6
94.49% :     0.000231s :      4: match.inline
 5.51% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   691
 1.63% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.87% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.57% :     0.000002s :     10: predicate.addn_zero_filter
 2.44% :     0.000003s :     10: predicate.arithmetic_simplify
 1.27% :     0.000002s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.04% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      3: predicate.elim_not_effective
 0.79% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.00% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.54% :     0.000008s :     22: predicate.inline
 1.07% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.04% :     0.000001s :      6: predicate.less_batch_normalization
 1.31% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 2.19% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.35% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.84% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000002s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 2.56% :     0.000004s :      3: predicate.mutable_eliminate
 0.31% :     0.000000s :      3: predicate.opt_reshape
 1.92% :     0.000003s :     15: predicate.partial_eliminate
 1.31% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.07% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 1.98% :     0.000003s :     16: predicate.switch_defer_inline
 2.44% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.91% :     0.000010s :     53: predicate.switch_simplify
 1.76% :     0.000002s :     10: predicate.tile_eliminate
 1.65% :     0.000002s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.18% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 2.01% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.61% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.34% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.03% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000702    12
44.29% :     0.000311s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.02% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.69% :     0.000335s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025416    76
 0.25% :     0.000063s :      1: add_recomputation
 0.31% :     0.000078s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.41% :     0.000359s :      1: bootstrap
 0.16% :     0.000041s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.11% :     0.000028s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.10% :     0.000025s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000006s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
29.11% :     0.007399s :      1: jit_opt_a
 0.71% :     0.000181s :      1: jit_opt_after_cconv
 0.30% :     0.000075s :      1: jit_opt_b
 2.54% :     0.000646s :      1: loop_unroll
 3.66% :     0.000931s :      1: mutable_eliminate
 4.21% :     0.001071s :     26: opt.transform.jit_opt_a
 0.22% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000038s :      4: opt.transform.jit_opt_b
 0.07% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000025s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.59% :     0.000657s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000052s :      1: py_interpret_to_execute
 0.12% :     0.000031s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000021s :      1: remove_dup_value
 2.07% :     0.000527s :      1: renormalize.infer
 1.80% :     0.000458s :      1: renormalize.specialize
 0.05% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000052s :      1: rewriter_after_opt_a
 0.33% :     0.000084s :      1: rewriter_before_opt_a
 0.34% :     0.000087s :      1: symbol_engine_optimizer
28.17% :     0.007160s :      1: task_emit
19.70% :     0.005007s :      1: type_inference
 0.28% :     0.000070s :      1: validate


TotalTime = 0.0157814, [30]
[bootstrap]: 0.00033024
[type_inference]: 0.00723629
[event_method]: 2.065e-05
[auto_monad]: 6.976e-05
[graph_reusing]: 5.47999e-06
[pre_auto_parallel]: 2.06e-06
[py_interpret_to_execute]: 6.251e-05
[rewriter_before_opt_a]: 9.343e-05
[expand_dump_flag]: 3.04999e-06
[jit_opt_a]: 0.00572953, [2]
    [Cycle 1]: 0.00194972, [27]
        [switch_simplify]: 9.874e-05
        [loop_unroll]: 6.231e-05
        [a_1]: 0.000723
        [with_stream_mark]: 1.593e-05
        [recompute_prepare]: 7.26001e-06
        [updatestate_depend_eliminate]: 3.35003e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [parameter_eliminate]: 1.35001e-06
        [specialize_transform]: 6.36e-06
        [updatestate_useless_node_eliminater]: 5.07999e-06
        [accelerated_algorithm]: 6.66e-06
        [meta_shard_fg_expand]: 2.22999e-06
        [get_grad_eliminate_]: 5.45001e-06
        [merge_forward]: 2.83e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.282e-05
        [j_node_and_user_rematch]: 9.82999e-06
        [meta_fg_expand]: 2.13002e-06
        [replace_old_param]: 8.72e-06
        [inline_without_move]: 5.72001e-06
        [renormalize]: 0.00074338
        [add_forward_monad_depend]: 5.37999e-06
        [auto_monad_grad]: 1.54998e-06
        [auto_monad_eliminator]: 9.37999e-06
        [cse]: 1.677e-05
        [replace_applicator]: 1.417e-05
    [Cycle 2]: 0.00032875, [27]
        [switch_simplify]: 6.53998e-06
        [loop_unroll]: 5.14e-06
        [a_1]: 9.403e-05
        [with_stream_mark]: 9.44e-06
        [recompute_prepare]: 5.09998e-06
        [updatestate_depend_eliminate]: 2.81e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.09003e-06
        [specialize_transform]: 5.38002e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 5.19e-06
        [meta_shard_fg_expand]: 1.33002e-06
        [get_grad_eliminate_]: 5.07e-06
        [merge_forward]: 2.64999e-06
        [cell_reuse_recompute_pass]: 2.41998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.503e-05
        [j_node_and_user_rematch]: 7.87998e-06
        [meta_fg_expand]: 1.84e-06
        [replace_old_param]: 7.33e-06
        [inline_without_move]: 4.95001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.59998e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 4.67998e-06
        [cse]: 1.043e-05
        [replace_applicator]: 5.07e-06
[py_interpret_to_execute_after_opt_a]: 1.07e-05
[rewriter_after_opt_a]: 2.828e-05
[convert_after_rewriter]: 5.83002e-06
[order_py_execute_after_rewriter]: 4.16001e-06
[mutable_eliminate]: 0.00058001
[jit_opt_b]: 5.219e-05, [1]
    [Cycle 1]: 4.519e-05, [2]
        [frontend_op_eliminate]: 1.656e-05
        [inline_after_opt_a]: 1.612e-05
[cconv]: 1.827e-05
[loop_unroll]: 0.00047366
[jit_opt_after_cconv]: 0.00014564, [1]
    [Cycle 1]: 0.0001392, [11]
        [c_1]: 2.378e-05
        [parameter_eliminate]: 3.97998e-06
        [updatestate_depend_eliminate]: 5.19998e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [cse]: 1.899e-05
        [call_graph_tuple_transform]: 1.752e-05
        [tuple_list_get_item_eliminator]: 5.56e-06
        [none_parameter_eliminate]: 1.20999e-06
        [renormalize]: 3.60014e-07
        [switch_simplify]: 6.07999e-06
[remove_dup_value]: 9.10001e-06
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 3.97e-06
[add_recomputation]: 3.431e-05
[cse_after_recomputation]: 2.107e-05, [1]
    [Cycle 1]: 1.565e-05, [1]
        [cse]: 9.96e-06
[auto_monad_reorder]: 1.079e-05
[get_jit_bprop_graph]: 1.82001e-06
[rewriter_after_jit_bprop_graph]: 5.67001e-06
[opt_after_jit_grad]: 0.00053049
[symbol_engine_optimizer]: 7.614e-05, [1]
    [Cycle 1]: 6.94e-05, [6]
        [build]: 2.82002e-06
        [elim_shapecalc]: 8.67e-06
        [elim_not_effective]: 1.356e-05
        [opt_reshape]: 6.16e-06
        [fold_const_symbol]: 8.84e-06
        [renormalize]: 7.39994e-07
[validate]: 3.064e-05
Sums
    bootstrap                            :     0.000330s :  2.81%
    type_inference                       :     0.007236s : 61.66%
    event_method                         :     0.000021s :  0.18%
    auto_monad                           :     0.000070s :  0.59%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000063s :  0.53%
    rewriter_before_opt_a                :     0.000093s :  0.80%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000105s :  0.90%
    jit_opt_a.loop_unroll                :     0.000067s :  0.57%
    jit_opt_a.a_1                        :     0.000817s :  6.96%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.09%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.14%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.000743s :  6.34%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.12%
    jit_opt_a.cse                        :     0.000027s :  0.23%
    jit_opt_a.replace_applicator         :     0.000019s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.09%
    rewriter_after_opt_a                 :     0.000028s :  0.24%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000580s :  4.94%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.14%
    cconv                                :     0.000018s :  0.16%
    loop_unroll                          :     0.000474s :  4.04%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000009s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000034s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.08%
    auto_monad_reorder                   :     0.000011s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000530s :  4.52%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000031s :  0.26%


Time group info:
------[substitution.]   0.000277    26
 0.49% :     0.000001s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 1.46% :     0.000004s :      3: substitution.graph_param_transform
87.09% :     0.000241s :      6: substitution.inline
 1.09% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.65% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.10% :     0.000003s :      2: substitution.replace_old_param
 3.24% :     0.000009s :      1: substitution.switch_simplify
 3.34% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007082     2
83.68% :     0.005926s :      1: type_inference.infer
16.32% :     0.001156s :      1: type_inference.specialize
------[replace.]   0.000138     9
56.51% :     0.000078s :      6: replace.inline
16.19% :     0.000022s :      1: replace.switch_simplify
27.30% :     0.000038s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000253     9
93.82% :     0.000238s :      6: match.inline
 3.18% :     0.000008s :      1: match.switch_simplify
 3.01% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   801
 1.55% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.03% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000002s :     12: predicate.addn_check_dump
 1.49% :     0.000002s :     12: predicate.addn_zero_filter
 2.07% :     0.000003s :     12: predicate.arithmetic_simplify
 1.33% :     0.000002s :     12: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     12: predicate.compare_switch_simplify
 1.65% :     0.000002s :     12: predicate.depend_value_elim
 1.36% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.76% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     12: predicate.environ_get_eliminate
 1.11% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.17% :     0.000002s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.34% :     0.000007s :     26: predicate.inline
 1.03% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000002s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     17: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.66% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.92% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     12: predicate.merge_addn
 1.11% :     0.000002s :     12: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     17: predicate.partial_eliminate
 1.40% :     0.000002s :     12: predicate.print_const_string_wrapper
 2.11% :     0.000003s :     12: predicate.reduce_eliminate
 1.56% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.40% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.06% :     0.000003s :     20: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     12: predicate.reshape_eliminate
 1.22% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.63% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     12: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.97% :     0.000004s :     20: predicate.switch_defer_inline
 2.40% :     0.000003s :     20: predicate.switch_layer_defer_inline
 7.53% :     0.000010s :     65: predicate.switch_simplify
 1.27% :     0.000002s :     12: predicate.tile_eliminate
 1.25% :     0.000002s :     12: predicate.transpose_eliminate
 1.43% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.50% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.51% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     12: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000932    15
50.32% :     0.000469s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.23% :     0.000105s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.45% :     0.000358s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017698    72
 0.21% :     0.000037s :      1: add_recomputation
 0.42% :     0.000074s :      1: auto_monad
 0.08% :     0.000014s :      1: auto_monad_reorder
 1.98% :     0.000350s :      1: bootstrap
 0.12% :     0.000021s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.13% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.14% :     0.000024s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
32.40% :     0.005735s :      1: jit_opt_a
 0.84% :     0.000149s :      1: jit_opt_after_cconv
 0.31% :     0.000055s :      1: jit_opt_b
 2.73% :     0.000483s :      1: loop_unroll
 3.32% :     0.000588s :      1: mutable_eliminate
 6.28% :     0.001111s :     26: opt.transform.jit_opt_a
 0.28% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.05% :     0.000540s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.37% :     0.000066s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000012s :      1: remove_dup_value
 2.24% :     0.000396s :      1: renormalize.infer
 1.92% :     0.000340s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000032s :      1: rewriter_after_opt_a
 0.55% :     0.000098s :      1: rewriter_before_opt_a
 0.45% :     0.000079s :      1: symbol_engine_optimizer
40.96% :     0.007250s :      1: type_inference


TotalTime = 0.0263953, [33]
[bootstrap]: 0.00073259
[type_inference]: 0.00728486
[event_method]: 2e-05
[auto_monad]: 7.374e-05
[graph_reusing]: 6.53e-06
[pre_auto_parallel]: 3.12002e-06
[py_interpret_to_execute]: 5.138e-05
[rewriter_before_opt_a]: 7.767e-05
[expand_dump_flag]: 3.81999e-06
[jit_opt_a]: 0.00741194, [2]
    [Cycle 1]: 0.0028961, [27]
        [switch_simplify]: 7.372e-05
        [loop_unroll]: 3.155e-05
        [a_1]: 0.00144298
        [with_stream_mark]: 2.472e-05
        [recompute_prepare]: 1.018e-05
        [updatestate_depend_eliminate]: 3.93001e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [parameter_eliminate]: 2.01e-06
        [specialize_transform]: 6.81999e-06
        [updatestate_useless_node_eliminater]: 5.44e-06
        [accelerated_algorithm]: 6.42001e-06
        [meta_shard_fg_expand]: 2.71e-06
        [get_grad_eliminate_]: 5.69e-06
        [merge_forward]: 4e-06
        [cell_reuse_recompute_pass]: 2.46998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.531e-05
        [j_node_and_user_rematch]: 1.006e-05
        [meta_fg_expand]: 2.99999e-06
        [replace_old_param]: 1.051e-05
        [inline_without_move]: 6.07001e-06
        [renormalize]: 0.0009245
        [add_forward_monad_depend]: 6.33e-06
        [auto_monad_grad]: 2.56e-06
        [auto_monad_eliminator]: 1.806e-05
        [cse]: 3.419e-05
        [replace_applicator]: 1.626e-05
    [Cycle 2]: 0.00036058, [27]
        [switch_simplify]: 6.06998e-06
        [loop_unroll]: 5.26002e-06
        [a_1]: 0.00010856
        [with_stream_mark]: 1.381e-05
        [recompute_prepare]: 5.52999e-06
        [updatestate_depend_eliminate]: 3.75e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.35002e-06
        [parameter_eliminate]: 1.43002e-06
        [specialize_transform]: 5.52001e-06
        [updatestate_useless_node_eliminater]: 5.10999e-06
        [accelerated_algorithm]: 4.88001e-06
        [meta_shard_fg_expand]: 1.79e-06
        [get_grad_eliminate_]: 4.57998e-06
        [merge_forward]: 3.23e-06
        [cell_reuse_recompute_pass]: 3.01999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.558e-05
        [j_node_and_user_rematch]: 9.08002e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 9.05999e-06
        [inline_without_move]: 4.87998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.27999e-06
        [auto_monad_grad]: 1.17e-06
        [auto_monad_eliminator]: 6.03998e-06
        [cse]: 1.242e-05
        [replace_applicator]: 5.05001e-06
[py_interpret_to_execute_after_opt_a]: 1.296e-05
[rewriter_after_opt_a]: 3.78e-05
[convert_after_rewriter]: 6.80998e-06
[order_py_execute_after_rewriter]: 4.82998e-06
[mutable_eliminate]: 0.00086335
[jit_opt_b]: 5.989e-05, [1]
    [Cycle 1]: 4.968e-05, [2]
        [frontend_op_eliminate]: 2.012e-05
        [inline_after_opt_a]: 1.717e-05
[cconv]: 3.087e-05
[loop_unroll]: 0.00059
[jit_opt_after_cconv]: 0.00016587, [1]
    [Cycle 1]: 0.00015712, [11]
        [c_1]: 2.502e-05
        [parameter_eliminate]: 2.74999e-06
        [updatestate_depend_eliminate]: 7.77998e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 3.12002e-06
        [cse]: 2.752e-05
        [call_graph_tuple_transform]: 2.069e-05
        [tuple_list_get_item_eliminator]: 5.65001e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 9.09989e-07
        [switch_simplify]: 5.82999e-06
[remove_dup_value]: 1.335e-05
[partial_unused_args_eliminate]: 2.29001e-06
[environ_conv]: 5.46e-06
[add_recomputation]: 5.118e-05
[cse_after_recomputation]: 2.437e-05, [1]
    [Cycle 1]: 1.731e-05, [1]
        [cse]: 1.07e-05
[auto_monad_reorder]: 1.574e-05
[get_jit_bprop_graph]: 1.62001e-06
[rewriter_after_jit_bprop_graph]: 7.38e-06
[opt_after_jit_grad]: 0.00053229
[symbol_engine_optimizer]: 7.993e-05, [1]
    [Cycle 1]: 7.338e-05, [6]
        [build]: 4.79e-06
        [elim_shapecalc]: 8.10999e-06
        [elim_not_effective]: 1.493e-05
        [opt_reshape]: 6.63e-06
        [fold_const_symbol]: 9.32001e-06
        [renormalize]: 5.69999e-07
[validate]: 4.62e-05
[backend_pass]: 1.14e-06
[task_emit]: 0.00791995
[execute]: 8.01001e-06
Sums
    bootstrap                            :     0.000733s :  3.40%
    type_inference                       :     0.007285s : 33.85%
    event_method                         :     0.000020s :  0.09%
    auto_monad                           :     0.000074s :  0.34%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000051s :  0.24%
    rewriter_before_opt_a                :     0.000078s :  0.36%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000080s :  0.37%
    jit_opt_a.loop_unroll                :     0.000037s :  0.17%
    jit_opt_a.a_1                        :     0.001552s :  7.21%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.000925s :  4.30%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.11%
    jit_opt_a.cse                        :     0.000047s :  0.22%
    jit_opt_a.replace_applicator         :     0.000021s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.06%
    rewriter_after_opt_a                 :     0.000038s :  0.18%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000863s :  4.01%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.08%
    cconv                                :     0.000031s :  0.14%
    loop_unroll                          :     0.000590s :  2.74%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000013s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000051s :  0.24%
    cse_after_recomputation.cse          :     0.000011s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.03%
    opt_after_jit_grad                   :     0.000532s :  2.47%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007920s : 36.80%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000292    23
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 1.88% :     0.000005s :      3: substitution.graph_param_transform
85.65% :     0.000250s :      4: substitution.inline
 1.64% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.80% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000006s :      2: substitution.replace_old_param
 5.72% :     0.000017s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007189     2
84.03% :     0.006041s :      1: type_inference.infer
15.97% :     0.001148s :      1: type_inference.specialize
------[replace.]   0.000114     6
66.04% :     0.000075s :      4: replace.inline
33.96% :     0.000039s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000261     6
94.20% :     0.000246s :      4: match.inline
 5.80% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000144   691
 1.13% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.91% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.87% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000002s :     10: predicate.addn_zero_filter
 2.85% :     0.000004s :     10: predicate.arithmetic_simplify
 1.45% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.86% :     0.000001s :     10: predicate.compare_switch_simplify
 1.52% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.69% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000001s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.49% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     10: predicate.environ_get_eliminate
 1.10% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.76% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000001s :      3: predicate.graph_param_transform
 5.40% :     0.000008s :     22: predicate.inline
 0.75% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000002s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.95% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.54% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.67% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 0.89% :     0.000001s :     10: predicate.merge_addn
 0.87% :     0.000001s :     10: predicate.minmaximum_grad
 2.20% :     0.000003s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 1.91% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.63% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.54% :     0.000002s :     18: predicate.replace_applicator
 1.00% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.67% :     0.000002s :     10: predicate.reshape_eliminate
 1.02% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.08% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000002s :     10: predicate.same_eliminate
 0.40% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.20% :     0.000002s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.50% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 2.94% :     0.000004s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.71% :     0.000004s :     16: predicate.switch_defer_inline
 1.78% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.07% :     0.000010s :     53: predicate.switch_simplify
 1.12% :     0.000002s :     10: predicate.tile_eliminate
 1.23% :     0.000002s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.10% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 2.13% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.61% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000858    12
47.49% :     0.000407s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.25% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.26% :     0.000397s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.029225    76
 0.19% :     0.000054s :      1: add_recomputation
 0.27% :     0.000078s :      1: auto_monad
 0.06% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 2.59% :     0.000756s :      1: bootstrap
 0.11% :     0.000034s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000027s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000025s :      1: event_method
 0.04% :     0.000012s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
25.38% :     0.007417s :      1: jit_opt_a
 0.58% :     0.000169s :      1: jit_opt_after_cconv
 0.22% :     0.000063s :      1: jit_opt_b
 2.05% :     0.000599s :      1: loop_unroll
 3.00% :     0.000877s :      1: mutable_eliminate
 6.15% :     0.001796s :     26: opt.transform.jit_opt_a
 0.18% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000030s :      4: opt.transform.jit_opt_b
 0.05% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.85% :     0.000541s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000054s :      1: py_interpret_to_execute
 0.05% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000016s :      1: remove_dup_value
 1.64% :     0.000481s :      1: renormalize.infer
 1.48% :     0.000433s :      1: renormalize.specialize
 0.03% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000041s :      1: rewriter_after_opt_a
 0.28% :     0.000081s :      1: rewriter_before_opt_a
 0.28% :     0.000083s :      1: symbol_engine_optimizer
27.15% :     0.007935s :      1: task_emit
25.01% :     0.007309s :      1: type_inference
 0.23% :     0.000068s :      1: validate


TotalTime = 0.0196528, [33]
[bootstrap]: 0.00027692
[type_inference]: 0.00459501
[event_method]: 1.747e-05
[auto_monad]: 6.311e-05
[graph_reusing]: 5.82999e-06
[pre_auto_parallel]: 2.42001e-06
[py_interpret_to_execute]: 3.075e-05
[rewriter_before_opt_a]: 6.824e-05
[expand_dump_flag]: 3.69002e-06
[jit_opt_a]: 0.00599035, [2]
    [Cycle 1]: 0.0018068, [27]
        [switch_simplify]: 6.157e-05
        [loop_unroll]: 3.049e-05
        [a_1]: 0.00059137
        [with_stream_mark]: 2.216e-05
        [recompute_prepare]: 7.31001e-06
        [updatestate_depend_eliminate]: 3.52997e-06
        [updatestate_assign_eliminate]: 2.84999e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 5.97999e-06
        [updatestate_useless_node_eliminater]: 5.07999e-06
        [accelerated_algorithm]: 5.27001e-06
        [meta_shard_fg_expand]: 2.11e-06
        [get_grad_eliminate_]: 3.111e-05
        [merge_forward]: 3.66999e-06
        [cell_reuse_recompute_pass]: 1.50999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.411e-05
        [j_node_and_user_rematch]: 9.52001e-06
        [meta_fg_expand]: 2.36e-06
        [replace_old_param]: 9.49e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.00075895
        [add_forward_monad_depend]: 4.74998e-06
        [auto_monad_grad]: 2.56e-06
        [auto_monad_eliminator]: 1.384e-05
        [cse]: 2.072e-05
        [replace_applicator]: 1.046e-05
    [Cycle 2]: 0.00032553, [27]
        [switch_simplify]: 5.62999e-06
        [loop_unroll]: 5.05001e-06
        [a_1]: 9.167e-05
        [with_stream_mark]: 9.61998e-06
        [recompute_prepare]: 5.32001e-06
        [updatestate_depend_eliminate]: 2.31998e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.77001e-06
        [parameter_eliminate]: 8.09989e-07
        [specialize_transform]: 5.00001e-06
        [updatestate_useless_node_eliminater]: 4.72998e-06
        [accelerated_algorithm]: 5.72001e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.54002e-06
        [merge_forward]: 2.43998e-06
        [cell_reuse_recompute_pass]: 2.24001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.247e-05
        [j_node_and_user_rematch]: 7.87e-06
        [meta_fg_expand]: 1.52999e-06
        [replace_old_param]: 7.1e-06
        [inline_without_move]: 4.84998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 6.19999e-06
        [cse]: 1.074e-05
        [replace_applicator]: 5.67001e-06
[py_interpret_to_execute_after_opt_a]: 1.386e-05
[rewriter_after_opt_a]: 3.667e-05
[convert_after_rewriter]: 6.75998e-06
[order_py_execute_after_rewriter]: 7.9e-06
[mutable_eliminate]: 0.00069335
[jit_opt_b]: 4.906e-05, [1]
    [Cycle 1]: 4.139e-05, [2]
        [frontend_op_eliminate]: 1.627e-05
        [inline_after_opt_a]: 1.395e-05
[cconv]: 2.738e-05
[loop_unroll]: 0.00046443
[jit_opt_after_cconv]: 0.00014425, [1]
    [Cycle 1]: 0.00013729, [11]
        [c_1]: 2.117e-05
        [parameter_eliminate]: 2.79001e-06
        [updatestate_depend_eliminate]: 5.46998e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 1.942e-05
        [call_graph_tuple_transform]: 2.008e-05
        [tuple_list_get_item_eliminator]: 5.72999e-06
        [none_parameter_eliminate]: 1.54998e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 5.33002e-06
[remove_dup_value]: 1.534e-05
[partial_unused_args_eliminate]: 2.45002e-06
[environ_conv]: 6.81001e-06
[add_recomputation]: 5.071e-05
[cse_after_recomputation]: 2.148e-05, [1]
    [Cycle 1]: 1.509e-05, [1]
        [cse]: 9.51e-06
[auto_monad_reorder]: 1.592e-05
[get_jit_bprop_graph]: 1.90001e-06
[rewriter_after_jit_bprop_graph]: 7.06999e-06
[opt_after_jit_grad]: 0.00044411
[symbol_engine_optimizer]: 7.287e-05, [1]
    [Cycle 1]: 6.642e-05, [6]
        [build]: 3.97e-06
        [elim_shapecalc]: 7.19001e-06
        [elim_not_effective]: 1.277e-05
        [opt_reshape]: 5.64e-06
        [fold_const_symbol]: 8.93002e-06
        [renormalize]: 7.2e-07
[validate]: 3.732e-05
[backend_pass]: 1.24e-06
[task_emit]: 0.00626312
[execute]: 1.059e-05
Sums
    bootstrap                            :     0.000277s :  1.82%
    type_inference                       :     0.004595s : 30.28%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000063s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.20%
    rewriter_before_opt_a                :     0.000068s :  0.45%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.44%
    jit_opt_a.loop_unroll                :     0.000036s :  0.23%
    jit_opt_a.a_1                        :     0.000683s :  4.50%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000036s :  0.23%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000759s :  5.00%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000031s :  0.21%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.09%
    rewriter_after_opt_a                 :     0.000037s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000008s :  0.05%
    mutable_eliminate                    :     0.000693s :  4.57%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000027s :  0.18%
    loop_unroll                          :     0.000464s :  3.06%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000015s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000051s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.05%
    opt_after_jit_grad                   :     0.000444s :  2.93%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006263s : 41.27%
    execute                              :     0.000011s :  0.07%


Time group info:
------[substitution.]   0.000245    23
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000002s :      2: substitution.fold_const_symbol
 2.37% :     0.000006s :      3: substitution.graph_param_transform
86.17% :     0.000211s :      4: substitution.inline
 1.27% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.94% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.69% :     0.000004s :      2: substitution.replace_old_param
 5.09% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004530     2
81.93% :     0.003711s :      1: type_inference.infer
18.07% :     0.000818s :      1: type_inference.specialize
------[replace.]   0.000071     6
67.41% :     0.000048s :      4: replace.inline
32.59% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000220     6
94.81% :     0.000208s :      4: match.inline
 5.19% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.24% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.37% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 2.03% :     0.000002s :     10: predicate.arithmetic_simplify
 1.40% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.11% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.54% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.49% :     0.000002s :      6: predicate.less_batch_normalization
 1.65% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.07% :     0.000002s :     15: predicate.load_eliminater
 2.41% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.49% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.32% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.10% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.31% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.01% :     0.000010s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.08% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000570    12
46.63% :     0.000266s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.71% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.66% :     0.000255s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021439    76
 0.25% :     0.000054s :      1: add_recomputation
 0.31% :     0.000066s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.36% :     0.000292s :      1: bootstrap
 0.14% :     0.000030s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.08% :     0.000016s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.96% :     0.005994s :      1: jit_opt_a
 0.69% :     0.000148s :      1: jit_opt_after_cconv
 0.24% :     0.000052s :      1: jit_opt_b
 2.21% :     0.000473s :      1: loop_unroll
 3.27% :     0.000700s :      1: mutable_eliminate
 4.31% :     0.000925s :     26: opt.transform.jit_opt_a
 0.23% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.10% :     0.000451s :      1: opt_after_jit_grad
 0.05% :     0.000010s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000033s :      1: py_interpret_to_execute
 0.08% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 2.03% :     0.000435s :      1: renormalize.infer
 1.47% :     0.000316s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000040s :      1: rewriter_after_opt_a
 0.33% :     0.000071s :      1: rewriter_before_opt_a
 0.35% :     0.000075s :      1: symbol_engine_optimizer
29.32% :     0.006285s :      1: task_emit
21.49% :     0.004607s :      1: type_inference
 0.28% :     0.000060s :      1: validate


TotalTime = 0.0242593, [33]
[bootstrap]: 0.000316
[type_inference]: 0.00552643
[event_method]: 2.056e-05
[auto_monad]: 7.196e-05
[graph_reusing]: 5.57001e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 4.693e-05
[rewriter_before_opt_a]: 7.674e-05
[expand_dump_flag]: 3.53e-06
[jit_opt_a]: 0.00796261, [2]
    [Cycle 1]: 0.00220901, [27]
        [switch_simplify]: 8.43e-05
        [loop_unroll]: 3.243e-05
        [a_1]: 0.00067324
        [with_stream_mark]: 2.482e-05
        [recompute_prepare]: 9.97999e-06
        [updatestate_depend_eliminate]: 3.53999e-06
        [updatestate_assign_eliminate]: 3.14999e-06
        [updatestate_loads_eliminate]: 2.78e-06
        [parameter_eliminate]: 2.70002e-06
        [specialize_transform]: 7.85998e-06
        [updatestate_useless_node_eliminater]: 5.69999e-06
        [accelerated_algorithm]: 6.93e-06
        [meta_shard_fg_expand]: 3.24001e-06
        [get_grad_eliminate_]: 7.13e-06
        [merge_forward]: 4.4e-06
        [cell_reuse_recompute_pass]: 1.79998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.716e-05
        [j_node_and_user_rematch]: 1.109e-05
        [meta_fg_expand]: 4e-06
        [replace_old_param]: 1.118e-05
        [inline_without_move]: 5.53997e-06
        [renormalize]: 0.00098981
        [add_forward_monad_depend]: 8.58001e-06
        [auto_monad_grad]: 2.46e-06
        [auto_monad_eliminator]: 1.648e-05
        [cse]: 2.918e-05
        [replace_applicator]: 2.054e-05
    [Cycle 2]: 0.00035735, [27]
        [switch_simplify]: 5.39e-06
        [loop_unroll]: 5.48002e-06
        [a_1]: 9.88e-05
        [with_stream_mark]: 1.322e-05
        [recompute_prepare]: 5.24998e-06
        [updatestate_depend_eliminate]: 3.01001e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.66e-06
        [specialize_transform]: 5.74999e-06
        [updatestate_useless_node_eliminater]: 5.74999e-06
        [accelerated_algorithm]: 6.41998e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 5.24998e-06
        [merge_forward]: 3.11001e-06
        [cell_reuse_recompute_pass]: 2.59999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.397e-05
        [j_node_and_user_rematch]: 8.75001e-06
        [meta_fg_expand]: 1.55999e-06
        [replace_old_param]: 1.165e-05
        [inline_without_move]: 5.46998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.84e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 6.51e-06
        [cse]: 1.102e-05
        [replace_applicator]: 5.47999e-06
[py_interpret_to_execute_after_opt_a]: 1.591e-05
[rewriter_after_opt_a]: 3.945e-05
[convert_after_rewriter]: 7.43e-06
[order_py_execute_after_rewriter]: 6.11e-06
[mutable_eliminate]: 0.00094521
[jit_opt_b]: 6.013e-05, [1]
    [Cycle 1]: 5.045e-05, [2]
        [frontend_op_eliminate]: 2.09e-05
        [inline_after_opt_a]: 1.667e-05
[cconv]: 3.082e-05
[loop_unroll]: 0.00057967
[jit_opt_after_cconv]: 0.00016889, [1]
    [Cycle 1]: 0.00016017, [11]
        [c_1]: 2.453e-05
        [parameter_eliminate]: 5.44e-06
        [updatestate_depend_eliminate]: 8.04002e-06
        [updatestate_assign_eliminate]: 2.55002e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [cse]: 2.802e-05
        [call_graph_tuple_transform]: 2.218e-05
        [tuple_list_get_item_eliminator]: 5.71003e-06
        [none_parameter_eliminate]: 1.74e-06
        [renormalize]: 1.07998e-06
        [switch_simplify]: 5.44e-06
[remove_dup_value]: 1.715e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 6.79001e-06
[add_recomputation]: 5.848e-05
[cse_after_recomputation]: 2.475e-05, [1]
    [Cycle 1]: 1.768e-05, [1]
        [cse]: 1.055e-05
[auto_monad_reorder]: 1.774e-05
[get_jit_bprop_graph]: 2.78e-06
[rewriter_after_jit_bprop_graph]: 8.03999e-06
[opt_after_jit_grad]: 0.00058963
[symbol_engine_optimizer]: 8.493e-05, [1]
    [Cycle 1]: 7.575e-05, [6]
        [build]: 4.59998e-06
        [elim_shapecalc]: 8.82e-06
        [elim_not_effective]: 1.555e-05
        [opt_reshape]: 6.43e-06
        [fold_const_symbol]: 9.94001e-06
        [renormalize]: 1.07998e-06
[validate]: 4.838e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.0072683
[execute]: 9.74999e-06
Sums
    bootstrap                            :     0.000316s :  1.74%
    type_inference                       :     0.005526s : 30.45%
    event_method                         :     0.000021s :  0.11%
    auto_monad                           :     0.000072s :  0.40%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000047s :  0.26%
    rewriter_before_opt_a                :     0.000077s :  0.42%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000090s :  0.49%
    jit_opt_a.loop_unroll                :     0.000038s :  0.21%
    jit_opt_a.a_1                        :     0.000772s :  4.25%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000014s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.07%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.03%
    jit_opt_a.replace_old_param          :     0.000023s :  0.13%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000990s :  5.45%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.13%
    jit_opt_a.cse                        :     0.000040s :  0.22%
    jit_opt_a.replace_applicator         :     0.000026s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.09%
    rewriter_after_opt_a                 :     0.000039s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000945s :  5.21%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000031s :  0.17%
    loop_unroll                          :     0.000580s :  3.19%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000017s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000058s :  0.32%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000018s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000590s :  3.25%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000048s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007268s : 40.04%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000293    23
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 2.11% :     0.000006s :      3: substitution.graph_param_transform
84.96% :     0.000249s :      4: substitution.inline
 1.30% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.87% :     0.000005s :      4: substitution.remove_not_recompute_node
 3.18% :     0.000009s :      2: substitution.replace_old_param
 5.35% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005447     2
81.94% :     0.004463s :      1: type_inference.infer
18.06% :     0.000983s :      1: type_inference.specialize
------[replace.]   0.000087     6
68.63% :     0.000059s :      4: replace.inline
31.37% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000261     6
94.43% :     0.000246s :      4: match.inline
 5.57% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   691
 1.86% :     0.000003s :     10: predicate.accumulaten_eliminater
 1.39% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.48% :     0.000002s :     10: predicate.addn_zero_filter
 2.37% :     0.000003s :     10: predicate.arithmetic_simplify
 1.12% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.90% :     0.000001s :     10: predicate.compare_switch_simplify
 1.16% :     0.000002s :     10: predicate.depend_value_elim
 0.92% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.86% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.97% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 1.16% :     0.000002s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.07% :     0.000007s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.80% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.62% :     0.000002s :     15: predicate.load_eliminater
 2.06% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.74% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000002s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 1.88% :     0.000003s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 1.84% :     0.000003s :     15: predicate.partial_eliminate
 1.01% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.23% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.05% :     0.000001s :     10: predicate.same_eliminate
 0.69% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.14% :     0.000002s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.11% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000003s :     16: predicate.switch_layer_defer_inline
10.63% :     0.000015s :     53: predicate.switch_simplify
 1.54% :     0.000002s :     10: predicate.tile_eliminate
 1.00% :     0.000001s :     10: predicate.transpose_eliminate
 1.28% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.44% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.06% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.41% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.58% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.92% :     0.000003s :     10: predicate.value_based_eliminate
 0.27% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000764    12
46.87% :     0.000358s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.38% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.75% :     0.000350s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026418    76
 0.23% :     0.000061s :      1: add_recomputation
 0.29% :     0.000075s :      1: auto_monad
 0.08% :     0.000021s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.24% :     0.000327s :      1: bootstrap
 0.13% :     0.000034s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.09% :     0.000025s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
30.16% :     0.007967s :      1: jit_opt_a
 0.65% :     0.000172s :      1: jit_opt_after_cconv
 0.24% :     0.000063s :      1: jit_opt_b
 2.24% :     0.000591s :      1: loop_unroll
 3.63% :     0.000960s :      1: mutable_eliminate
 3.94% :     0.001042s :     26: opt.transform.jit_opt_a
 0.20% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000030s :      4: opt.transform.jit_opt_b
 0.07% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.28% :     0.000601s :      1: opt_after_jit_grad
 0.03% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000050s :      1: py_interpret_to_execute
 0.07% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000020s :      1: remove_dup_value
 1.93% :     0.000510s :      1: renormalize.infer
 1.78% :     0.000471s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000042s :      1: rewriter_after_opt_a
 0.30% :     0.000080s :      1: rewriter_before_opt_a
 0.33% :     0.000087s :      1: symbol_engine_optimizer
27.58% :     0.007287s :      1: task_emit
20.98% :     0.005542s :      1: type_inference
 0.28% :     0.000073s :      1: validate


TotalTime = 0.0164996, [30]
[bootstrap]: 0.00030824
[type_inference]: 0.0073768
[event_method]: 2.45e-05
[auto_monad]: 0.00010083
[graph_reusing]: 7.38999e-06
[pre_auto_parallel]: 2.37001e-06
[py_interpret_to_execute]: 6.87e-05
[rewriter_before_opt_a]: 9.317e-05
[expand_dump_flag]: 3.19001e-06
[jit_opt_a]: 0.00624954, [2]
    [Cycle 1]: 0.00223259, [27]
        [switch_simplify]: 0.00013954
        [loop_unroll]: 3.898e-05
        [a_1]: 0.00086768
        [with_stream_mark]: 1.756e-05
        [recompute_prepare]: 8.75001e-06
        [updatestate_depend_eliminate]: 3.37002e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [parameter_eliminate]: 1.42999e-06
        [specialize_transform]: 7.33999e-06
        [updatestate_useless_node_eliminater]: 5.94999e-06
        [accelerated_algorithm]: 7e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 5.66e-06
        [merge_forward]: 3.41001e-06
        [cell_reuse_recompute_pass]: 1.55001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.452e-05
        [j_node_and_user_rematch]: 1.274e-05
        [meta_fg_expand]: 2.38998e-06
        [replace_old_param]: 1.02e-05
        [inline_without_move]: 6.42001e-06
        [renormalize]: 0.00082992
        [add_forward_monad_depend]: 5.27999e-06
        [auto_monad_grad]: 2.78e-06
        [auto_monad_eliminator]: 1.204e-05
        [cse]: 1.59e-05
        [replace_applicator]: 1.416e-05
    [Cycle 2]: 0.00035411, [27]
        [switch_simplify]: 6.59001e-06
        [loop_unroll]: 5.44e-06
        [a_1]: 0.00010465
        [with_stream_mark]: 1.106e-05
        [recompute_prepare]: 6.43e-06
        [updatestate_depend_eliminate]: 2.73998e-06
        [updatestate_assign_eliminate]: 2.51998e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 5.44998e-06
        [updatestate_useless_node_eliminater]: 5.69999e-06
        [accelerated_algorithm]: 5.80002e-06
        [meta_shard_fg_expand]: 1.22999e-06
        [get_grad_eliminate_]: 5.01002e-06
        [merge_forward]: 2.52001e-06
        [cell_reuse_recompute_pass]: 2.39001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.305e-05
        [j_node_and_user_rematch]: 8.76997e-06
        [meta_fg_expand]: 1.77999e-06
        [replace_old_param]: 8.70999e-06
        [inline_without_move]: 5.35999e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 4.87e-06
        [cse]: 1.043e-05
        [replace_applicator]: 5.35999e-06
[py_interpret_to_execute_after_opt_a]: 1.178e-05
[rewriter_after_opt_a]: 3.233e-05
[convert_after_rewriter]: 6.81001e-06
[order_py_execute_after_rewriter]: 5.41002e-06
[mutable_eliminate]: 0.00060862
[jit_opt_b]: 5.598e-05, [1]
    [Cycle 1]: 4.885e-05, [2]
        [frontend_op_eliminate]: 1.895e-05
        [inline_after_opt_a]: 1.715e-05
[cconv]: 2.268e-05
[loop_unroll]: 0.00046889
[jit_opt_after_cconv]: 0.00015262, [1]
    [Cycle 1]: 0.00014493, [11]
        [c_1]: 2.347e-05
        [parameter_eliminate]: 2.63e-06
        [updatestate_depend_eliminate]: 5.20001e-06
        [updatestate_assign_eliminate]: 2.59001e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [cse]: 1.905e-05
        [call_graph_tuple_transform]: 1.976e-05
        [tuple_list_get_item_eliminator]: 6.76e-06
        [none_parameter_eliminate]: 1.15001e-06
        [renormalize]: 4.70027e-07
        [switch_simplify]: 6.41e-06
[remove_dup_value]: 1.067e-05
[partial_unused_args_eliminate]: 2.40002e-06
[environ_conv]: 4.72998e-06
[add_recomputation]: 3.689e-05
[cse_after_recomputation]: 3.92e-05, [1]
    [Cycle 1]: 1.848e-05, [1]
        [cse]: 1.165e-05
[auto_monad_reorder]: 1.247e-05
[get_jit_bprop_graph]: 1.50999e-06
[rewriter_after_jit_bprop_graph]: 5.50001e-06
[opt_after_jit_grad]: 0.00048099
[symbol_engine_optimizer]: 7.743e-05, [1]
    [Cycle 1]: 7.134e-05, [6]
        [build]: 3.60998e-06
        [elim_shapecalc]: 8.63001e-06
        [elim_not_effective]: 1.332e-05
        [opt_reshape]: 5.79999e-06
        [fold_const_symbol]: 9.14e-06
        [renormalize]: 6.80011e-07
[validate]: 3.152e-05
Sums
    bootstrap                            :     0.000308s :  2.53%
    type_inference                       :     0.007377s : 60.57%
    event_method                         :     0.000025s :  0.20%
    auto_monad                           :     0.000101s :  0.83%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000069s :  0.56%
    rewriter_before_opt_a                :     0.000093s :  0.76%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000146s :  1.20%
    jit_opt_a.loop_unroll                :     0.000044s :  0.36%
    jit_opt_a.a_1                        :     0.000972s :  7.98%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000022s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.16%
    jit_opt_a.inline_without_move        :     0.000012s :  0.10%
    jit_opt_a.renormalize                :     0.000830s :  6.82%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.14%
    jit_opt_a.cse                        :     0.000026s :  0.22%
    jit_opt_a.replace_applicator         :     0.000020s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.10%
    rewriter_after_opt_a                 :     0.000032s :  0.27%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000609s :  5.00%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.14%
    cconv                                :     0.000023s :  0.19%
    loop_unroll                          :     0.000469s :  3.85%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000011s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000037s :  0.30%
    cse_after_recomputation.cse          :     0.000012s :  0.10%
    auto_monad_reorder                   :     0.000012s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000481s :  3.95%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000032s :  0.26%


Time group info:
------[substitution.]   0.000345    26
 0.48% :     0.000002s :      2: substitution.elim_not_effective
 0.36% :     0.000001s :      2: substitution.fold_const_symbol
 1.36% :     0.000005s :      3: substitution.graph_param_transform
87.52% :     0.000302s :      6: substitution.inline
 1.23% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.46% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.24% :     0.000004s :      2: substitution.replace_old_param
 2.70% :     0.000009s :      1: substitution.switch_simplify
 3.66% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007299     2
82.22% :     0.006001s :      1: type_inference.infer
17.78% :     0.001297s :      1: type_inference.specialize
------[replace.]   0.000193     9
48.38% :     0.000094s :      6: replace.inline
26.85% :     0.000052s :      1: replace.switch_simplify
24.77% :     0.000048s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000318     9
93.79% :     0.000298s :      6: match.inline
 2.67% :     0.000008s :      1: match.switch_simplify
 3.54% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000157   801
 1.38% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.90% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000002s :     12: predicate.addn_check_dump
 1.69% :     0.000003s :     12: predicate.addn_zero_filter
 3.22% :     0.000005s :     12: predicate.arithmetic_simplify
 1.20% :     0.000002s :     12: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000002s :     12: predicate.compare_switch_simplify
 1.55% :     0.000002s :     12: predicate.depend_value_elim
 1.17% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.11% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.62% :     0.000003s :     12: predicate.environ_add_const_eliminate
 1.04% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.13% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.15% :     0.000002s :     12: predicate.environ_get_eliminate
 1.13% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.77% :     0.000001s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 5.07% :     0.000008s :     26: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.86% :     0.000003s :     14: predicate.list_to_tuple_eliminator_
 1.57% :     0.000002s :     17: predicate.load_eliminater
 1.19% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.00% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     12: predicate.merge_addn
 1.11% :     0.000002s :     12: predicate.minmaximum_grad
 1.81% :     0.000003s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 1.95% :     0.000003s :     17: predicate.partial_eliminate
 1.03% :     0.000002s :     12: predicate.print_const_string_wrapper
 2.24% :     0.000004s :     12: predicate.reduce_eliminate
 1.53% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.36% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.51% :     0.000002s :     20: predicate.replace_applicator
 0.41% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     12: predicate.reshape_eliminate
 1.31% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     12: predicate.same_eliminate
 0.62% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 0.85% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.13% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.33% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.06% :     0.000005s :     20: predicate.switch_defer_inline
 2.40% :     0.000004s :     20: predicate.switch_layer_defer_inline
 8.42% :     0.000013s :     65: predicate.switch_simplify
 1.17% :     0.000002s :     12: predicate.tile_eliminate
 1.10% :     0.000002s :     12: predicate.transpose_eliminate
 2.63% :     0.000004s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.62% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.46% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     12: predicate.value_based_eliminate
 0.24% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000923    15
46.10% :     0.000426s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.96% :     0.000120s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.94% :     0.000378s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018680    72
 0.21% :     0.000040s :      1: add_recomputation
 0.57% :     0.000107s :      1: auto_monad
 0.08% :     0.000015s :      1: auto_monad_reorder
 1.72% :     0.000322s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.22% :     0.000042s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000029s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000010s :      1: graph_reusing
33.48% :     0.006254s :      1: jit_opt_a
 0.83% :     0.000156s :      1: jit_opt_after_cconv
 0.32% :     0.000059s :      1: jit_opt_b
 2.56% :     0.000477s :      1: loop_unroll
 3.31% :     0.000618s :      1: mutable_eliminate
 6.93% :     0.001295s :     26: opt.transform.jit_opt_a
 0.28% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000029s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.62% :     0.000489s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.39% :     0.000073s :      1: py_interpret_to_execute
 0.08% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000013s :      1: remove_dup_value
 2.43% :     0.000455s :      1: renormalize.infer
 1.96% :     0.000365s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000036s :      1: rewriter_after_opt_a
 0.52% :     0.000098s :      1: rewriter_before_opt_a
 0.43% :     0.000080s :      1: symbol_engine_optimizer
39.58% :     0.007394s :      1: type_inference


TotalTime = 0.0198101, [33]
[bootstrap]: 0.0003214
[type_inference]: 0.00526147
[event_method]: 1.966e-05
[auto_monad]: 6.596e-05
[graph_reusing]: 6.19001e-06
[pre_auto_parallel]: 2.31998e-06
[py_interpret_to_execute]: 4.407e-05
[rewriter_before_opt_a]: 7.593e-05
[expand_dump_flag]: 3.75e-06
[jit_opt_a]: 0.00598538, [2]
    [Cycle 1]: 0.00180976, [27]
        [switch_simplify]: 5.944e-05
        [loop_unroll]: 2.993e-05
        [a_1]: 0.00057695
        [with_stream_mark]: 1.945e-05
        [recompute_prepare]: 8.00999e-06
        [updatestate_depend_eliminate]: 3.52002e-06
        [updatestate_assign_eliminate]: 2.59999e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [parameter_eliminate]: 2.26998e-06
        [specialize_transform]: 5.83002e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.70001e-06
        [meta_shard_fg_expand]: 2.41998e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 3.58e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.417e-05
        [j_node_and_user_rematch]: 9.84999e-06
        [meta_fg_expand]: 2.71e-06
        [replace_old_param]: 8.63001e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.00078892
        [add_forward_monad_depend]: 6.53e-06
        [auto_monad_grad]: 2.44999e-06
        [auto_monad_eliminator]: 1.445e-05
        [cse]: 2.662e-05
        [replace_applicator]: 1.356e-05
    [Cycle 2]: 0.00034311, [27]
        [switch_simplify]: 5.93002e-06
        [loop_unroll]: 5.72001e-06
        [a_1]: 9.611e-05
        [with_stream_mark]: 1.186e-05
        [recompute_prepare]: 4.96002e-06
        [updatestate_depend_eliminate]: 2.72001e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [parameter_eliminate]: 1.50001e-06
        [specialize_transform]: 4.93001e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.27999e-06
        [meta_shard_fg_expand]: 1.40001e-06
        [get_grad_eliminate_]: 4.66002e-06
        [merge_forward]: 2.88e-06
        [cell_reuse_recompute_pass]: 2.56e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.416e-05
        [j_node_and_user_rematch]: 8.52e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 8.15999e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.50001e-06
        [auto_monad_grad]: 1.38002e-06
        [auto_monad_eliminator]: 5.56e-06
        [cse]: 1.185e-05
        [replace_applicator]: 5.13002e-06
[py_interpret_to_execute_after_opt_a]: 1.402e-05
[rewriter_after_opt_a]: 3.681e-05
[convert_after_rewriter]: 7.21999e-06
[order_py_execute_after_rewriter]: 4.95001e-06
[mutable_eliminate]: 0.00062461
[jit_opt_b]: 5.331e-05, [1]
    [Cycle 1]: 4.501e-05, [2]
        [frontend_op_eliminate]: 1.725e-05
        [inline_after_opt_a]: 1.586e-05
[cconv]: 4.384e-05
[loop_unroll]: 0.00054556
[jit_opt_after_cconv]: 0.00016603, [1]
    [Cycle 1]: 0.00015776, [11]
        [c_1]: 2.423e-05
        [parameter_eliminate]: 3.36999e-06
        [updatestate_depend_eliminate]: 7.16001e-06
        [updatestate_assign_eliminate]: 2.69999e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [cse]: 2.448e-05
        [call_graph_tuple_transform]: 2.216e-05
        [tuple_list_get_item_eliminator]: 6.39001e-06
        [none_parameter_eliminate]: 1.80001e-06
        [renormalize]: 7.60017e-07
        [switch_simplify]: 6.16e-06
[remove_dup_value]: 1.397e-05
[partial_unused_args_eliminate]: 2.48e-06
[environ_conv]: 5.64e-06
[add_recomputation]: 5.406e-05
[cse_after_recomputation]: 2.457e-05, [1]
    [Cycle 1]: 1.809e-05, [1]
        [cse]: 1.121e-05
[auto_monad_reorder]: 1.649e-05
[get_jit_bprop_graph]: 1.97001e-06
[rewriter_after_jit_bprop_graph]: 7.76001e-06
[opt_after_jit_grad]: 0.00057542
[symbol_engine_optimizer]: 8.224e-05, [1]
    [Cycle 1]: 7.353e-05, [6]
        [build]: 4.25e-06
        [elim_shapecalc]: 9.04e-06
        [elim_not_effective]: 1.537e-05
        [opt_reshape]: 6.07001e-06
        [fold_const_symbol]: 8.60001e-06
        [renormalize]: 7.7e-07
[validate]: 3.971e-05
[backend_pass]: 1.04e-06
[task_emit]: 0.0054825
[execute]: 5.59e-06
Sums
    bootstrap                            :     0.000321s :  2.10%
    type_inference                       :     0.005261s : 34.34%
    event_method                         :     0.000020s :  0.13%
    auto_monad                           :     0.000066s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000044s :  0.29%
    rewriter_before_opt_a                :     0.000076s :  0.50%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.43%
    jit_opt_a.loop_unroll                :     0.000036s :  0.23%
    jit_opt_a.a_1                        :     0.000673s :  4.39%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000789s :  5.15%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000038s :  0.25%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.09%
    rewriter_after_opt_a                 :     0.000037s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000625s :  4.08%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000044s :  0.29%
    loop_unroll                          :     0.000546s :  3.56%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000054s :  0.35%
    cse_after_recomputation.cse          :     0.000011s :  0.07%
    auto_monad_reorder                   :     0.000016s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000575s :  3.76%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005482s : 35.78%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000238    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.56% :     0.000006s :      3: substitution.graph_param_transform
84.89% :     0.000202s :      4: substitution.inline
 1.80% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.95% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.77% :     0.000004s :      2: substitution.replace_old_param
 5.58% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005185     2
81.02% :     0.004201s :      1: type_inference.infer
18.98% :     0.000984s :      1: type_inference.specialize
------[replace.]   0.000072     6
67.66% :     0.000048s :      4: replace.inline
32.34% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000212     6
94.28% :     0.000200s :      4: match.inline
 5.72% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.63% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 2.44% :     0.000003s :     10: predicate.arithmetic_simplify
 1.55% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.41% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.14% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.53% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000002s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.93% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.05% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.16% :     0.000001s :     10: predicate.minmaximum_grad
 1.54% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.19% :     0.000001s :     10: predicate.reshape_eliminate
 1.35% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 1.91% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.67% :     0.000010s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.61% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.71% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.97% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.08% :     0.000003s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000653    12
46.06% :     0.000301s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.32% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.62% :     0.000304s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021616    76
 0.27% :     0.000057s :      1: add_recomputation
 0.32% :     0.000070s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.56% :     0.000337s :      1: bootstrap
 0.22% :     0.000048s :      1: cconv
 0.06% :     0.000013s :      1: convert_after_rewriter
 0.12% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.11% :     0.000024s :      1: event_method
 0.04% :     0.000009s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
27.70% :     0.005988s :      1: jit_opt_a
 0.78% :     0.000169s :      1: jit_opt_after_cconv
 0.26% :     0.000056s :      1: jit_opt_b
 2.57% :     0.000556s :      1: loop_unroll
 2.94% :     0.000635s :      1: mutable_eliminate
 4.12% :     0.000891s :     26: opt.transform.jit_opt_a
 0.25% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.70% :     0.000584s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.22% :     0.000047s :      1: py_interpret_to_execute
 0.08% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.96% :     0.000425s :      1: renormalize.infer
 1.64% :     0.000354s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000040s :      1: rewriter_after_opt_a
 0.37% :     0.000079s :      1: rewriter_before_opt_a
 0.39% :     0.000085s :      1: symbol_engine_optimizer
25.41% :     0.005492s :      1: task_emit
24.42% :     0.005279s :      1: type_inference
 0.30% :     0.000064s :      1: validate


TotalTime = 0.0232383, [33]
[bootstrap]: 0.00030532
[type_inference]: 0.00560039
[event_method]: 1.677e-05
[auto_monad]: 4.711e-05
[graph_reusing]: 3.53e-06
[pre_auto_parallel]: 1.69998e-06
[py_interpret_to_execute]: 4.119e-05
[rewriter_before_opt_a]: 6.507e-05
[expand_dump_flag]: 2.03002e-06
[jit_opt_a]: 0.00594227, [2]
    [Cycle 1]: 0.00195026, [27]
        [switch_simplify]: 6.258e-05
        [loop_unroll]: 2.981e-05
        [a_1]: 0.00058762
        [with_stream_mark]: 2.464e-05
        [recompute_prepare]: 6.64001e-06
        [updatestate_depend_eliminate]: 3.78999e-06
        [updatestate_assign_eliminate]: 2.80997e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 1.74998e-06
        [specialize_transform]: 6.11998e-06
        [updatestate_useless_node_eliminater]: 6.01e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 2.27001e-06
        [get_grad_eliminate_]: 5.27001e-06
        [merge_forward]: 3.40998e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.456e-05
        [j_node_and_user_rematch]: 9.10001e-06
        [meta_fg_expand]: 2.35002e-06
        [replace_old_param]: 2.234e-05
        [inline_without_move]: 5.79e-06
        [renormalize]: 0.00090085
        [add_forward_monad_depend]: 5.15999e-06
        [auto_monad_grad]: 2.76999e-06
        [auto_monad_eliminator]: 1.379e-05
        [cse]: 3.16e-05
        [replace_applicator]: 1.197e-05
    [Cycle 2]: 0.00033686, [27]
        [switch_simplify]: 5.79e-06
        [loop_unroll]: 4.89e-06
        [a_1]: 9.241e-05
        [with_stream_mark]: 9.76e-06
        [recompute_prepare]: 5.20001e-06
        [updatestate_depend_eliminate]: 2.63e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 4.80999e-06
        [updatestate_useless_node_eliminater]: 4.72e-06
        [accelerated_algorithm]: 5.61e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.85999e-06
        [merge_forward]: 2.74999e-06
        [cell_reuse_recompute_pass]: 1.91003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.245e-05
        [j_node_and_user_rematch]: 8.47998e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 7.00002e-06
        [inline_without_move]: 4.97e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 5.39e-06
        [cse]: 1.245e-05
        [replace_applicator]: 6.16e-06
[py_interpret_to_execute_after_opt_a]: 1.328e-05
[rewriter_after_opt_a]: 3.594e-05
[convert_after_rewriter]: 6.83998e-06
[order_py_execute_after_rewriter]: 4.91002e-06
[mutable_eliminate]: 0.00068311
[jit_opt_b]: 5.175e-05, [1]
    [Cycle 1]: 4.425e-05, [2]
        [frontend_op_eliminate]: 1.71e-05
        [inline_after_opt_a]: 1.471e-05
[cconv]: 2.899e-05
[loop_unroll]: 0.00044352
[jit_opt_after_cconv]: 0.00014958, [1]
    [Cycle 1]: 0.00014256, [11]
        [c_1]: 2.149e-05
        [parameter_eliminate]: 2.99001e-06
        [updatestate_depend_eliminate]: 5.87999e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [cse]: 2.276e-05
        [call_graph_tuple_transform]: 2.006e-05
        [tuple_list_get_item_eliminator]: 5.21998e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 6.89994e-07
        [switch_simplify]: 5.54e-06
[remove_dup_value]: 1.401e-05
[partial_unused_args_eliminate]: 2.24001e-06
[environ_conv]: 5.67999e-06
[add_recomputation]: 5.276e-05
[cse_after_recomputation]: 2.206e-05, [1]
    [Cycle 1]: 1.594e-05, [1]
        [cse]: 9.73002e-06
[auto_monad_reorder]: 1.536e-05
[get_jit_bprop_graph]: 2.22001e-06
[rewriter_after_jit_bprop_graph]: 6.31e-06
[opt_after_jit_grad]: 0.00048117
[symbol_engine_optimizer]: 7.602e-05, [1]
    [Cycle 1]: 6.92e-05, [6]
        [build]: 4.21001e-06
        [elim_shapecalc]: 7.76001e-06
        [elim_not_effective]: 1.359e-05
        [opt_reshape]: 5.69999e-06
        [fold_const_symbol]: 9.00001e-06
        [renormalize]: 7.09988e-07
[validate]: 3.873e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00884295
[execute]: 1.09e-05
Sums
    bootstrap                            :     0.000305s :  1.61%
    type_inference                       :     0.005600s : 29.58%
    event_method                         :     0.000017s :  0.09%
    auto_monad                           :     0.000047s :  0.25%
    graph_reusing                        :     0.000004s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.22%
    rewriter_before_opt_a                :     0.000065s :  0.34%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000068s :  0.36%
    jit_opt_a.loop_unroll                :     0.000035s :  0.18%
    jit_opt_a.a_1                        :     0.000680s :  3.59%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000029s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000901s :  4.76%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.10%
    jit_opt_a.cse                        :     0.000044s :  0.23%
    jit_opt_a.replace_applicator         :     0.000018s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.19%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000683s :  3.61%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000029s :  0.15%
    loop_unroll                          :     0.000444s :  2.34%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000053s :  0.28%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000481s :  2.54%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.20%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008843s : 46.71%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000259    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 2.28% :     0.000006s :      3: substitution.graph_param_transform
82.16% :     0.000213s :      4: substitution.inline
 1.38% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.76% :     0.000005s :      4: substitution.remove_not_recompute_node
 6.47% :     0.000017s :      2: substitution.replace_old_param
 4.60% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005530     2
81.85% :     0.004526s :      1: type_inference.infer
18.15% :     0.001004s :      1: type_inference.specialize
------[replace.]   0.000076     6
68.79% :     0.000052s :      4: replace.inline
31.21% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000221     6
95.10% :     0.000210s :      4: match.inline
 4.90% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.16% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 2.91% :     0.000004s :     10: predicate.arithmetic_simplify
 1.24% :     0.000002s :     10: predicate.cast_eliminate
 0.67% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.51% :     0.000002s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.19% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.38% :     0.000002s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000001s :      3: predicate.graph_param_transform
 5.56% :     0.000007s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.51% :     0.000002s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.88% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.87% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 1.84% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.36% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.84% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.27% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.31% :     0.000002s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.57% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.01% :     0.000010s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.04% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000690    12
42.94% :     0.000296s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.28% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.78% :     0.000343s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025151    76
 0.22% :     0.000056s :      1: add_recomputation
 0.20% :     0.000050s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.28% :     0.000323s :      1: bootstrap
 0.13% :     0.000032s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000020s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000006s :      1: graph_reusing
23.64% :     0.005946s :      1: jit_opt_a
 0.61% :     0.000153s :      1: jit_opt_after_cconv
 0.22% :     0.000055s :      1: jit_opt_b
 1.80% :     0.000453s :      1: loop_unroll
 2.76% :     0.000694s :      1: mutable_eliminate
 3.63% :     0.000912s :     26: opt.transform.jit_opt_a
 0.19% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.95% :     0.000489s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000044s :      1: py_interpret_to_execute
 0.06% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.02% :     0.000507s :      1: renormalize.infer
 1.54% :     0.000386s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000039s :      1: rewriter_after_opt_a
 0.27% :     0.000068s :      1: rewriter_before_opt_a
 0.31% :     0.000079s :      1: symbol_engine_optimizer
35.24% :     0.008863s :      1: task_emit
22.32% :     0.005614s :      1: type_inference
 0.24% :     0.000061s :      1: validate


TotalTime = 0.02245, [33]
[bootstrap]: 0.0003562
[type_inference]: 0.00592316
[event_method]: 2.26e-05
[auto_monad]: 7.231e-05
[graph_reusing]: 8.30999e-06
[pre_auto_parallel]: 2.67001e-06
[py_interpret_to_execute]: 4.921e-05
[rewriter_before_opt_a]: 8.537e-05
[expand_dump_flag]: 3.74002e-06
[jit_opt_a]: 0.00622074, [2]
    [Cycle 1]: 0.00191664, [27]
        [switch_simplify]: 6.623e-05
        [loop_unroll]: 2.967e-05
        [a_1]: 0.00059848
        [with_stream_mark]: 1.923e-05
        [recompute_prepare]: 8.92e-06
        [updatestate_depend_eliminate]: 3.7e-06
        [updatestate_assign_eliminate]: 3.05998e-06
        [updatestate_loads_eliminate]: 2.98998e-06
        [parameter_eliminate]: 2.15002e-06
        [specialize_transform]: 6.84001e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 2.90998e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 4.17003e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.578e-05
        [j_node_and_user_rematch]: 9.76e-06
        [meta_fg_expand]: 2.20002e-06
        [replace_old_param]: 9.80002e-06
        [inline_without_move]: 5.67999e-06
        [renormalize]: 0.00085008
        [add_forward_monad_depend]: 5.54e-06
        [auto_monad_grad]: 2.51e-06
        [auto_monad_eliminator]: 1.488e-05
        [cse]: 2.914e-05
        [replace_applicator]: 1.632e-05
    [Cycle 2]: 0.0003333, [27]
        [switch_simplify]: 5.72001e-06
        [loop_unroll]: 4.95999e-06
        [a_1]: 9.253e-05
        [with_stream_mark]: 1.122e-05
        [recompute_prepare]: 4.76002e-06
        [updatestate_depend_eliminate]: 2.61e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 4.93001e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 5.20001e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 4.53999e-06
        [merge_forward]: 2.66e-06
        [cell_reuse_recompute_pass]: 1.75001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.451e-05
        [j_node_and_user_rematch]: 8.55001e-06
        [meta_fg_expand]: 2.09999e-06
        [replace_old_param]: 1.365e-05
        [inline_without_move]: 5.15999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.52999e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 5.07999e-06
        [cse]: 1.057e-05
        [replace_applicator]: 5.48002e-06
[py_interpret_to_execute_after_opt_a]: 1.28e-05
[rewriter_after_opt_a]: 3.671e-05
[convert_after_rewriter]: 7.3e-06
[order_py_execute_after_rewriter]: 4.92e-06
[mutable_eliminate]: 0.00072176
[jit_opt_b]: 5.171e-05, [1]
    [Cycle 1]: 4.393e-05, [2]
        [frontend_op_eliminate]: 1.679e-05
        [inline_after_opt_a]: 1.49e-05
[cconv]: 2.852e-05
[loop_unroll]: 0.0004827
[jit_opt_after_cconv]: 0.00015162, [1]
    [Cycle 1]: 0.00014499, [11]
        [c_1]: 2.124e-05
        [parameter_eliminate]: 4.02002e-06
        [updatestate_depend_eliminate]: 6.22001e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.52001e-06
        [cse]: 2.383e-05
        [call_graph_tuple_transform]: 1.946e-05
        [tuple_list_get_item_eliminator]: 5.19e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 9.50007e-07
        [switch_simplify]: 5.49e-06
[remove_dup_value]: 1.571e-05
[partial_unused_args_eliminate]: 2.21998e-06
[environ_conv]: 5.11997e-06
[add_recomputation]: 5.132e-05
[cse_after_recomputation]: 2.238e-05, [1]
    [Cycle 1]: 1.619e-05, [1]
        [cse]: 1.011e-05
[auto_monad_reorder]: 1.57e-05
[get_jit_bprop_graph]: 2.29999e-06
[rewriter_after_jit_bprop_graph]: 6.40002e-06
[opt_after_jit_grad]: 0.000493
[symbol_engine_optimizer]: 7.586e-05, [1]
    [Cycle 1]: 6.862e-05, [6]
        [build]: 4.16001e-06
        [elim_shapecalc]: 7.66999e-06
        [elim_not_effective]: 1.317e-05
        [opt_reshape]: 5.39e-06
        [fold_const_symbol]: 9.24e-06
        [renormalize]: 1.12999e-06
[validate]: 4.143e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00722962
[execute]: 9.69e-06
Sums
    bootstrap                            :     0.000356s :  2.00%
    type_inference                       :     0.005923s : 33.25%
    event_method                         :     0.000023s :  0.13%
    auto_monad                           :     0.000072s :  0.41%
    graph_reusing                        :     0.000008s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000049s :  0.28%
    rewriter_before_opt_a                :     0.000085s :  0.48%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000072s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.19%
    jit_opt_a.a_1                        :     0.000691s :  3.88%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000023s :  0.13%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000850s :  4.77%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.11%
    jit_opt_a.cse                        :     0.000040s :  0.22%
    jit_opt_a.replace_applicator         :     0.000022s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000722s :  4.05%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000029s :  0.16%
    loop_unroll                          :     0.000483s :  2.71%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000051s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000493s :  2.77%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007230s : 40.58%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000254    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.33% :     0.000006s :      3: substitution.graph_param_transform
83.13% :     0.000212s :      4: substitution.inline
 1.48% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.92% :     0.000005s :      4: substitution.remove_not_recompute_node
 4.29% :     0.000011s :      2: substitution.replace_old_param
 5.45% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005830     2
81.71% :     0.004764s :      1: type_inference.infer
18.29% :     0.001066s :      1: type_inference.specialize
------[replace.]   0.000078     6
65.20% :     0.000051s :      4: replace.inline
34.80% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000222     6
94.29% :     0.000209s :      4: match.inline
 5.71% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.52% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.20% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 2.40% :     0.000003s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000001s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.84% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.17% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.87% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.21% :     0.000006s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.90% :     0.000002s :     15: predicate.load_eliminater
 1.28% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.65% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000000s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.08% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.34% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.39% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.68% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.39% :     0.000002s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.25% :     0.000003s :     16: predicate.switch_defer_inline
 1.99% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.51% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.31% :     0.000002s :     10: predicate.transpose_eliminate
 1.89% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.93% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000767    12
47.45% :     0.000364s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.04% :     0.000046s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.51% :     0.000357s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024322    76
 0.22% :     0.000055s :      1: add_recomputation
 0.31% :     0.000076s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.55% :     0.000378s :      1: bootstrap
 0.13% :     0.000031s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.11% :     0.000027s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000011s :      1: graph_reusing
25.59% :     0.006225s :      1: jit_opt_a
 0.64% :     0.000155s :      1: jit_opt_after_cconv
 0.22% :     0.000054s :      1: jit_opt_b
 2.02% :     0.000492s :      1: loop_unroll
 3.01% :     0.000731s :      1: mutable_eliminate
 3.81% :     0.000927s :     26: opt.transform.jit_opt_a
 0.20% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.06% :     0.000502s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000052s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 1.82% :     0.000443s :      1: renormalize.infer
 1.63% :     0.000397s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000040s :      1: rewriter_after_opt_a
 0.37% :     0.000090s :      1: rewriter_before_opt_a
 0.32% :     0.000078s :      1: symbol_engine_optimizer
29.80% :     0.007248s :      1: task_emit
24.43% :     0.005943s :      1: type_inference
 0.26% :     0.000063s :      1: validate


TotalTime = 0.0172055, [33]
[bootstrap]: 0.00026794
[type_inference]: 0.00427634
[event_method]: 1.624e-05
[auto_monad]: 6.275e-05
[graph_reusing]: 5.55001e-06
[pre_auto_parallel]: 2.55997e-06
[py_interpret_to_execute]: 3.187e-05
[rewriter_before_opt_a]: 7.258e-05
[expand_dump_flag]: 2.99999e-06
[jit_opt_a]: 0.00478819, [2]
    [Cycle 1]: 0.00139303, [27]
        [switch_simplify]: 5.025e-05
        [loop_unroll]: 2.918e-05
        [a_1]: 0.0005035
        [with_stream_mark]: 1.499e-05
        [recompute_prepare]: 6.86999e-06
        [updatestate_depend_eliminate]: 3.84002e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.30002e-06
        [parameter_eliminate]: 1.97999e-06
        [specialize_transform]: 6.06e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.17e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 5.02e-06
        [merge_forward]: 3.82998e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.404e-05
        [j_node_and_user_rematch]: 8.58001e-06
        [meta_fg_expand]: 2.29001e-06
        [replace_old_param]: 7.75e-06
        [inline_without_move]: 4.94e-06
        [renormalize]: 0.00049726
        [add_forward_monad_depend]: 5.25999e-06
        [auto_monad_grad]: 1.81e-06
        [auto_monad_eliminator]: 1.337e-05
        [cse]: 2.53e-05
        [replace_applicator]: 1.115e-05
    [Cycle 2]: 0.00032421, [27]
        [switch_simplify]: 5.50001e-06
        [loop_unroll]: 4.74002e-06
        [a_1]: 0.00010301
        [with_stream_mark]: 9.02999e-06
        [recompute_prepare]: 5.06002e-06
        [updatestate_depend_eliminate]: 2.61e-06
        [updatestate_assign_eliminate]: 2.14999e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 9.60019e-07
        [specialize_transform]: 4.82e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 4.43001e-06
        [meta_shard_fg_expand]: 1.66e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.29001e-06
        [cell_reuse_recompute_pass]: 1.44003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.216e-05
        [j_node_and_user_rematch]: 7.52002e-06
        [meta_fg_expand]: 1.51002e-06
        [replace_old_param]: 7.06001e-06
        [inline_without_move]: 4.61002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 9.49978e-07
        [auto_monad_eliminator]: 5.44998e-06
        [cse]: 1.019e-05
        [replace_applicator]: 4.75999e-06
[py_interpret_to_execute_after_opt_a]: 9.07999e-06
[rewriter_after_opt_a]: 3.007e-05
[convert_after_rewriter]: 6.27001e-06
[order_py_execute_after_rewriter]: 4.99e-06
[mutable_eliminate]: 0.00047673
[jit_opt_b]: 4.614e-05, [1]
    [Cycle 1]: 4.009e-05, [2]
        [frontend_op_eliminate]: 1.487e-05
        [inline_after_opt_a]: 1.471e-05
[cconv]: 2.302e-05
[loop_unroll]: 0.00042525
[jit_opt_after_cconv]: 0.00013475, [1]
    [Cycle 1]: 0.00012836, [11]
        [c_1]: 2.146e-05
        [parameter_eliminate]: 2.47001e-06
        [updatestate_depend_eliminate]: 5.42001e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.707e-05
        [call_graph_tuple_transform]: 1.669e-05
        [tuple_list_get_item_eliminator]: 5.12e-06
        [none_parameter_eliminate]: 1.74998e-06
        [renormalize]: 4.2998e-07
        [switch_simplify]: 5.12999e-06
[remove_dup_value]: 1.347e-05
[partial_unused_args_eliminate]: 2.19999e-06
[environ_conv]: 5.41002e-06
[add_recomputation]: 4.45e-05
[cse_after_recomputation]: 2.034e-05, [1]
    [Cycle 1]: 1.454e-05, [1]
        [cse]: 9.05001e-06
[auto_monad_reorder]: 1.431e-05
[get_jit_bprop_graph]: 1.44998e-06
[rewriter_after_jit_bprop_graph]: 5.02999e-06
[opt_after_jit_grad]: 0.00045166
[symbol_engine_optimizer]: 7.283e-05, [1]
    [Cycle 1]: 6.69e-05, [6]
        [build]: 3.2e-06
        [elim_shapecalc]: 7.7e-06
        [elim_not_effective]: 1.266e-05
        [opt_reshape]: 7.30998e-06
        [fold_const_symbol]: 8.64e-06
        [renormalize]: 4.09986e-07
[validate]: 3.064e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.00568604
[execute]: 7.16001e-06
Sums
    bootstrap                            :     0.000268s :  1.97%
    type_inference                       :     0.004276s : 31.48%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000063s :  0.46%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.23%
    rewriter_before_opt_a                :     0.000073s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000607s :  4.46%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000497s :  3.66%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.14%
    jit_opt_a.cse                        :     0.000035s :  0.26%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000030s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000477s :  3.51%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000425s :  3.13%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000045s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000452s :  3.32%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005686s : 41.85%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000178    23
 1.17% :     0.000002s :      2: substitution.elim_not_effective
 0.77% :     0.000001s :      2: substitution.fold_const_symbol
 2.63% :     0.000005s :      3: substitution.graph_param_transform
83.06% :     0.000148s :      4: substitution.inline
 1.70% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.41% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.82% :     0.000003s :      2: substitution.replace_old_param
 6.42% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004215     2
81.87% :     0.003451s :      1: type_inference.infer
18.13% :     0.000764s :      1: type_inference.specialize
------[replace.]   0.000063     6
67.06% :     0.000042s :      4: replace.inline
32.94% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000156     6
93.41% :     0.000146s :      4: match.inline
 6.59% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.20% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000001s :     10: predicate.addn_zero_filter
 1.76% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000001s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.49% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.61% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.17% :     0.000001s :     10: predicate.minmaximum_grad
 1.90% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.54% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.89% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.90% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.73% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.58% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.00% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000517    12
50.18% :     0.000260s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.19% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.63% :     0.000231s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018614    76
 0.25% :     0.000047s :      1: add_recomputation
 0.35% :     0.000066s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.51% :     0.000280s :      1: bootstrap
 0.14% :     0.000025s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.74% :     0.004791s :      1: jit_opt_a
 0.74% :     0.000137s :      1: jit_opt_after_cconv
 0.26% :     0.000048s :      1: jit_opt_b
 2.32% :     0.000433s :      1: loop_unroll
 2.60% :     0.000484s :      1: mutable_eliminate
 4.32% :     0.000803s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.46% :     0.000459s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.38% :     0.000257s :      1: renormalize.infer
 1.25% :     0.000233s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000033s :      1: rewriter_after_opt_a
 0.40% :     0.000075s :      1: rewriter_before_opt_a
 0.40% :     0.000075s :      1: symbol_engine_optimizer
30.60% :     0.005695s :      1: task_emit
23.03% :     0.004287s :      1: type_inference
 0.26% :     0.000049s :      1: validate


TotalTime = 0.0138113, [30]
[bootstrap]: 0.00025151
[type_inference]: 0.00605445
[event_method]: 1.893e-05
[auto_monad]: 7.327e-05
[graph_reusing]: 4.90001e-06
[pre_auto_parallel]: 2.15002e-06
[py_interpret_to_execute]: 5.815e-05
[rewriter_before_opt_a]: 8.062e-05
[expand_dump_flag]: 2.74999e-06
[jit_opt_a]: 0.00523178, [2]
    [Cycle 1]: 0.0017835, [27]
        [switch_simplify]: 0.00010013
        [loop_unroll]: 3.537e-05
        [a_1]: 0.00067758
        [with_stream_mark]: 1.332e-05
        [recompute_prepare]: 7.78001e-06
        [updatestate_depend_eliminate]: 3.11999e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.37e-06
        [specialize_transform]: 6.66999e-06
        [updatestate_useless_node_eliminater]: 5.69999e-06
        [accelerated_algorithm]: 5.96e-06
        [meta_shard_fg_expand]: 1.64e-06
        [get_grad_eliminate_]: 5.51998e-06
        [merge_forward]: 2.69001e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.282e-05
        [j_node_and_user_rematch]: 9.89999e-06
        [meta_fg_expand]: 2.32999e-06
        [replace_old_param]: 8.79e-06
        [inline_without_move]: 5.66998e-06
        [renormalize]: 0.00065824
        [add_forward_monad_depend]: 4.71002e-06
        [auto_monad_grad]: 1.59e-06
        [auto_monad_eliminator]: 1.066e-05
        [cse]: 1.61e-05
        [replace_applicator]: 1.196e-05
    [Cycle 2]: 0.00031616, [27]
        [switch_simplify]: 5.80002e-06
        [loop_unroll]: 5.32001e-06
        [a_1]: 9.329e-05
        [with_stream_mark]: 8.19002e-06
        [recompute_prepare]: 5.22999e-06
        [updatestate_depend_eliminate]: 2.59001e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 8.50006e-07
        [specialize_transform]: 4.92e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 4.79e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.75999e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.19998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.209e-05
        [j_node_and_user_rematch]: 7.80998e-06
        [meta_fg_expand]: 1.93997e-06
        [replace_old_param]: 7.45e-06
        [inline_without_move]: 4.95999e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 9.80013e-07
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 4.64002e-06
        [cse]: 9.84001e-06
        [replace_applicator]: 5.39998e-06
[py_interpret_to_execute_after_opt_a]: 8.67e-06
[rewriter_after_opt_a]: 2.758e-05
[convert_after_rewriter]: 6.02999e-06
[order_py_execute_after_rewriter]: 4.80999e-06
[mutable_eliminate]: 0.00051139
[jit_opt_b]: 6.1e-05, [1]
    [Cycle 1]: 5.391e-05, [2]
        [frontend_op_eliminate]: 1.536e-05
        [inline_after_opt_a]: 2.724e-05
[cconv]: 2.09e-05
[loop_unroll]: 0.00042621
[jit_opt_after_cconv]: 0.00013896, [1]
    [Cycle 1]: 0.00013224, [11]
        [c_1]: 2.231e-05
        [parameter_eliminate]: 2.55002e-06
        [updatestate_depend_eliminate]: 5.72999e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [cse]: 1.646e-05
        [call_graph_tuple_transform]: 1.782e-05
        [tuple_list_get_item_eliminator]: 5.51e-06
        [none_parameter_eliminate]: 1.29e-06
        [renormalize]: 4.40021e-07
        [switch_simplify]: 5.62001e-06
[remove_dup_value]: 1.146e-05
[partial_unused_args_eliminate]: 1.79998e-06
[environ_conv]: 4.33001e-06
[add_recomputation]: 3.714e-05
[cse_after_recomputation]: 2.121e-05, [1]
    [Cycle 1]: 1.557e-05, [1]
        [cse]: 9.74e-06
[auto_monad_reorder]: 1.24e-05
[get_jit_bprop_graph]: 1.37999e-06
[rewriter_after_jit_bprop_graph]: 5.04e-06
[opt_after_jit_grad]: 0.00046588
[symbol_engine_optimizer]: 7.453e-05, [1]
    [Cycle 1]: 6.837e-05, [6]
        [build]: 3.48e-06
        [elim_shapecalc]: 8.09002e-06
        [elim_not_effective]: 1.316e-05
        [opt_reshape]: 5.59e-06
        [fold_const_symbol]: 8.85999e-06
        [renormalize]: 4.59986e-07
[validate]: 3.065e-05
Sums
    bootstrap                            :     0.000252s :  2.49%
    type_inference                       :     0.006054s : 59.84%
    event_method                         :     0.000019s :  0.19%
    auto_monad                           :     0.000073s :  0.72%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000058s :  0.57%
    rewriter_before_opt_a                :     0.000081s :  0.80%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000106s :  1.05%
    jit_opt_a.loop_unroll                :     0.000041s :  0.40%
    jit_opt_a.a_1                        :     0.000771s :  7.62%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.16%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000658s :  6.51%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.15%
    jit_opt_a.cse                        :     0.000026s :  0.26%
    jit_opt_a.replace_applicator         :     0.000017s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.09%
    rewriter_after_opt_a                 :     0.000028s :  0.27%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000511s :  5.05%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.27%
    cconv                                :     0.000021s :  0.21%
    loop_unroll                          :     0.000426s :  4.21%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000011s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000037s :  0.37%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000012s :  0.12%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000466s :  4.60%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.30%


Time group info:
------[substitution.]   0.000254    26
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 1.92% :     0.000005s :      3: substitution.graph_param_transform
84.75% :     0.000215s :      6: substitution.inline
 1.03% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.88% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.38% :     0.000003s :      2: substitution.replace_old_param
 3.30% :     0.000008s :      1: substitution.switch_simplify
 4.36% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006001     2
81.21% :     0.004874s :      1: type_inference.infer
18.79% :     0.001128s :      1: type_inference.specialize
------[replace.]   0.000128     9
53.56% :     0.000068s :      6: replace.inline
21.59% :     0.000028s :      1: replace.switch_simplify
24.85% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000229     9
92.46% :     0.000211s :      6: match.inline
 3.30% :     0.000008s :      1: match.switch_simplify
 4.25% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   801
 1.23% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.08% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000002s :     12: predicate.addn_check_dump
 1.41% :     0.000002s :     12: predicate.addn_zero_filter
 2.16% :     0.000003s :     12: predicate.arithmetic_simplify
 1.45% :     0.000002s :     12: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.18% :     0.000002s :     12: predicate.compare_switch_simplify
 1.49% :     0.000002s :     12: predicate.depend_value_elim
 1.16% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.51% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     12: predicate.environ_get_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.02% :     0.000007s :     26: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.04% :     0.000001s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     17: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.17% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.29% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000002s :     12: predicate.merge_addn
 1.10% :     0.000002s :     12: predicate.minmaximum_grad
 1.50% :     0.000002s :      3: predicate.mutable_eliminate
 0.38% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     17: predicate.partial_eliminate
 1.23% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.84% :     0.000003s :     12: predicate.reduce_eliminate
 1.43% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     20: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     12: predicate.reshape_eliminate
 1.30% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.62% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.59% :     0.000002s :     12: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.78% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.01% :     0.000004s :     20: predicate.switch_defer_inline
 2.33% :     0.000003s :     20: predicate.switch_layer_defer_inline
 7.85% :     0.000011s :     65: predicate.switch_simplify
 1.33% :     0.000002s :     12: predicate.tile_eliminate
 1.30% :     0.000002s :     12: predicate.transpose_eliminate
 1.50% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.55% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.68% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000003s :     12: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.52% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000792    15
43.10% :     0.000341s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.05% :     0.000087s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.85% :     0.000363s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015581    72
 0.26% :     0.000040s :      1: add_recomputation
 0.49% :     0.000077s :      1: auto_monad
 0.10% :     0.000015s :      1: auto_monad_reorder
 1.70% :     0.000264s :      1: bootstrap
 0.15% :     0.000024s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.15% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.14% :     0.000022s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
33.61% :     0.005237s :      1: jit_opt_a
 0.91% :     0.000142s :      1: jit_opt_after_cconv
 0.41% :     0.000064s :      1: jit_opt_b
 2.78% :     0.000434s :      1: loop_unroll
 3.34% :     0.000520s :      1: mutable_eliminate
 6.64% :     0.001034s :     26: opt.transform.jit_opt_a
 0.31% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.23% :     0.000036s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.04% :     0.000474s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.39% :     0.000061s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000014s :      1: remove_dup_value
 2.28% :     0.000355s :      1: renormalize.infer
 1.90% :     0.000296s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000030s :      1: rewriter_after_opt_a
 0.54% :     0.000084s :      1: rewriter_before_opt_a
 0.50% :     0.000077s :      1: symbol_engine_optimizer
38.93% :     0.006066s :      1: type_inference


TotalTime = 0.0214846, [33]
[bootstrap]: 0.0003183
[type_inference]: 0.00495996
[event_method]: 1.933e-05
[auto_monad]: 6.641e-05
[graph_reusing]: 6.02999e-06
[pre_auto_parallel]: 2.65002e-06
[py_interpret_to_execute]: 4.009e-05
[rewriter_before_opt_a]: 7.445e-05
[expand_dump_flag]: 3.31001e-06
[jit_opt_a]: 0.00612478, [2]
    [Cycle 1]: 0.00187353, [27]
        [switch_simplify]: 6.244e-05
        [loop_unroll]: 2.887e-05
        [a_1]: 0.00056804
        [with_stream_mark]: 2.339e-05
        [recompute_prepare]: 8.13001e-06
        [updatestate_depend_eliminate]: 4.22003e-06
        [updatestate_assign_eliminate]: 2.85998e-06
        [updatestate_loads_eliminate]: 2.99001e-06
        [parameter_eliminate]: 2.06e-06
        [specialize_transform]: 6.12001e-06
        [updatestate_useless_node_eliminater]: 5.39e-06
        [accelerated_algorithm]: 5.34e-06
        [meta_shard_fg_expand]: 2.29999e-06
        [get_grad_eliminate_]: 5.94e-06
        [merge_forward]: 4.13001e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.484e-05
        [j_node_and_user_rematch]: 1.008e-05
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 9.90002e-06
        [inline_without_move]: 5.47001e-06
        [renormalize]: 0.00082741
        [add_forward_monad_depend]: 7.18998e-06
        [auto_monad_grad]: 2.33998e-06
        [auto_monad_eliminator]: 1.782e-05
        [cse]: 3.23e-05
        [replace_applicator]: 1.546e-05
    [Cycle 2]: 0.00033922, [27]
        [switch_simplify]: 5.67999e-06
        [loop_unroll]: 4.88001e-06
        [a_1]: 9.531e-05
        [with_stream_mark]: 1.475e-05
        [recompute_prepare]: 4.88001e-06
        [updatestate_depend_eliminate]: 3.33998e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [parameter_eliminate]: 1.39e-06
        [specialize_transform]: 5.12e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 4.83001e-06
        [meta_shard_fg_expand]: 1.70001e-06
        [get_grad_eliminate_]: 4.48001e-06
        [merge_forward]: 3.36999e-06
        [cell_reuse_recompute_pass]: 2.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.484e-05
        [j_node_and_user_rematch]: 8.18999e-06
        [meta_fg_expand]: 2.17999e-06
        [replace_old_param]: 8.44002e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 9.80013e-07
        [auto_monad_eliminator]: 6.23002e-06
        [cse]: 1.11e-05
        [replace_applicator]: 5.60001e-06
[py_interpret_to_execute_after_opt_a]: 3.032e-05
[rewriter_after_opt_a]: 3.678e-05
[convert_after_rewriter]: 8.90001e-06
[order_py_execute_after_rewriter]: 5.06997e-06
[mutable_eliminate]: 0.00074843
[jit_opt_b]: 5.418e-05, [1]
    [Cycle 1]: 4.631e-05, [2]
        [frontend_op_eliminate]: 1.769e-05
        [inline_after_opt_a]: 1.655e-05
[cconv]: 2.917e-05
[loop_unroll]: 0.0004923
[jit_opt_after_cconv]: 0.00015328, [1]
    [Cycle 1]: 0.00014586, [11]
        [c_1]: 2.085e-05
        [parameter_eliminate]: 3.58e-06
        [updatestate_depend_eliminate]: 6.21e-06
        [updatestate_assign_eliminate]: 2.65002e-06
        [updatestate_loads_eliminate]: 2.52001e-06
        [cse]: 2.61e-05
        [call_graph_tuple_transform]: 1.949e-05
        [tuple_list_get_item_eliminator]: 5.63997e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 7.29982e-07
        [switch_simplify]: 5.67999e-06
[remove_dup_value]: 1.434e-05
[partial_unused_args_eliminate]: 2.08998e-06
[environ_conv]: 5.46e-06
[add_recomputation]: 5.211e-05
[cse_after_recomputation]: 2.448e-05, [1]
    [Cycle 1]: 1.824e-05, [1]
        [cse]: 1.138e-05
[auto_monad_reorder]: 1.533e-05
[get_jit_bprop_graph]: 2.43e-06
[rewriter_after_jit_bprop_graph]: 7.21999e-06
[opt_after_jit_grad]: 0.00050747
[symbol_engine_optimizer]: 8.202e-05, [1]
    [Cycle 1]: 7.453e-05, [6]
        [build]: 4.90999e-06
        [elim_shapecalc]: 8.94e-06
        [elim_not_effective]: 1.549e-05
        [opt_reshape]: 6.38e-06
        [fold_const_symbol]: 9.84001e-06
        [renormalize]: 5.19998e-07
[validate]: 4.427e-05
[backend_pass]: 1.02998e-06
[task_emit]: 0.00730885
[execute]: 9.92999e-06
Sums
    bootstrap                            :     0.000318s :  1.88%
    type_inference                       :     0.004960s : 29.34%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000066s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.24%
    rewriter_before_opt_a                :     0.000074s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000663s :  3.92%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000827s :  4.90%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.14%
    jit_opt_a.cse                        :     0.000043s :  0.26%
    jit_opt_a.replace_applicator         :     0.000021s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000030s :  0.18%
    rewriter_after_opt_a                 :     0.000037s :  0.22%
    convert_after_rewriter               :     0.000009s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000748s :  4.43%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.10%
    cconv                                :     0.000029s :  0.17%
    loop_unroll                          :     0.000492s :  2.91%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000052s :  0.31%
    cse_after_recomputation.cse          :     0.000011s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000507s :  3.00%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007309s : 43.24%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000225    23
 1.11% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.48% :     0.000006s :      3: substitution.graph_param_transform
83.99% :     0.000189s :      4: substitution.inline
 1.61% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.96% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.33% :     0.000005s :      2: substitution.replace_old_param
 5.91% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004884     2
81.61% :     0.003985s :      1: type_inference.infer
18.39% :     0.000898s :      1: type_inference.specialize
------[replace.]   0.000076     6
63.62% :     0.000049s :      4: replace.inline
36.38% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000198     6
93.88% :     0.000186s :      4: match.inline
 6.12% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.83% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 2.79% :     0.000004s :     10: predicate.arithmetic_simplify
 1.27% :     0.000002s :     10: predicate.cast_eliminate
 0.56% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.22% :     0.000002s :     10: predicate.depend_value_elim
 0.97% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.05% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.74% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     10: predicate.environ_get_eliminate
 1.24% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000001s :      3: predicate.graph_param_transform
 5.15% :     0.000007s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.54% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.56% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.68% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 2.40% :     0.000003s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000003s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.77% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.89% :     0.000002s :     18: predicate.replace_applicator
 0.83% :     0.000001s :      6: predicate.replace_old_param
 0.67% :     0.000001s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.06% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.17% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 1.84% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.91% :     0.000010s :     53: predicate.switch_simplify
 1.68% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.05% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.12% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.48% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.64% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000653    12
50.58% :     0.000330s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.34% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.08% :     0.000288s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023306    76
 0.24% :     0.000055s :      1: add_recomputation
 0.30% :     0.000070s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.45% :     0.000338s :      1: bootstrap
 0.14% :     0.000032s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.12% :     0.000027s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.30% :     0.006129s :      1: jit_opt_a
 0.67% :     0.000156s :      1: jit_opt_after_cconv
 0.25% :     0.000058s :      1: jit_opt_b
 2.15% :     0.000501s :      1: loop_unroll
 3.26% :     0.000759s :      1: mutable_eliminate
 3.80% :     0.000886s :     26: opt.transform.jit_opt_a
 0.21% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.22% :     0.000517s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000043s :      1: py_interpret_to_execute
 0.15% :     0.000035s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.93% :     0.000450s :      1: renormalize.infer
 1.58% :     0.000369s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000040s :      1: rewriter_after_opt_a
 0.33% :     0.000078s :      1: rewriter_before_opt_a
 0.36% :     0.000085s :      1: symbol_engine_optimizer
31.45% :     0.007329s :      1: task_emit
21.35% :     0.004977s :      1: type_inference
 0.29% :     0.000068s :      1: validate


TotalTime = 0.022393, [33]
[bootstrap]: 0.00097235
[type_inference]: 0.00535
[event_method]: 2.163e-05
[auto_monad]: 7.164e-05
[graph_reusing]: 6.19001e-06
[pre_auto_parallel]: 3.26001e-06
[py_interpret_to_execute]: 4.557e-05
[rewriter_before_opt_a]: 7.666e-05
[expand_dump_flag]: 3.45998e-06
[jit_opt_a]: 0.00632068, [2]
    [Cycle 1]: 0.00195292, [27]
        [switch_simplify]: 6.552e-05
        [loop_unroll]: 2.911e-05
        [a_1]: 0.00060311
        [with_stream_mark]: 2.447e-05
        [recompute_prepare]: 7.71001e-06
        [updatestate_depend_eliminate]: 3.70998e-06
        [updatestate_assign_eliminate]: 3.11999e-06
        [updatestate_loads_eliminate]: 2.69999e-06
        [parameter_eliminate]: 2.07999e-06
        [specialize_transform]: 6.53e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 2.67001e-06
        [get_grad_eliminate_]: 5.40999e-06
        [merge_forward]: 3.75e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.481e-05
        [j_node_and_user_rematch]: 9.54e-06
        [meta_fg_expand]: 2.26998e-06
        [replace_old_param]: 9.59e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.00088988
        [add_forward_monad_depend]: 6.43e-06
        [auto_monad_grad]: 2.59999e-06
        [auto_monad_eliminator]: 1.408e-05
        [cse]: 3.223e-05
        [replace_applicator]: 1.348e-05
    [Cycle 2]: 0.0003247, [27]
        [switch_simplify]: 5.42001e-06
        [loop_unroll]: 4.76002e-06
        [a_1]: 9.151e-05
        [with_stream_mark]: 9.81998e-06
        [recompute_prepare]: 5.05001e-06
        [updatestate_depend_eliminate]: 2.61999e-06
        [updatestate_assign_eliminate]: 1.89e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.14003e-06
        [specialize_transform]: 4.72e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 4.97999e-06
        [meta_shard_fg_expand]: 1.18001e-06
        [get_grad_eliminate_]: 4.43001e-06
        [merge_forward]: 2.53998e-06
        [cell_reuse_recompute_pass]: 2.22001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.277e-05
        [j_node_and_user_rematch]: 7.98999e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 7.33e-06
        [inline_without_move]: 4.63001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 6.30011e-07
        [auto_monad_eliminator]: 6.26998e-06
        [cse]: 1.044e-05
        [replace_applicator]: 4.99003e-06
[py_interpret_to_execute_after_opt_a]: 1.252e-05
[rewriter_after_opt_a]: 3.736e-05
[convert_after_rewriter]: 7.33999e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00068426
[jit_opt_b]: 5.088e-05, [1]
    [Cycle 1]: 4.361e-05, [2]
        [frontend_op_eliminate]: 1.758e-05
        [inline_after_opt_a]: 1.466e-05
[cconv]: 2.726e-05
[loop_unroll]: 0.00049155
[jit_opt_after_cconv]: 0.0001492, [1]
    [Cycle 1]: 0.00014137, [11]
        [c_1]: 2.051e-05
        [parameter_eliminate]: 3.31999e-06
        [updatestate_depend_eliminate]: 5.99999e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.08002e-06
        [cse]: 2.382e-05
        [call_graph_tuple_transform]: 1.912e-05
        [tuple_list_get_item_eliminator]: 5.27001e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 1.07e-06
        [switch_simplify]: 5.49e-06
[remove_dup_value]: 1.495e-05
[partial_unused_args_eliminate]: 2.32999e-06
[environ_conv]: 6.06e-06
[add_recomputation]: 5.088e-05
[cse_after_recomputation]: 2.26e-05, [1]
    [Cycle 1]: 1.613e-05, [1]
        [cse]: 9.52001e-06
[auto_monad_reorder]: 1.649e-05
[get_jit_bprop_graph]: 2.69999e-06
[rewriter_after_jit_bprop_graph]: 5.64998e-06
[opt_after_jit_grad]: 0.00052606
[symbol_engine_optimizer]: 7.942e-05, [1]
    [Cycle 1]: 7.189e-05, [6]
        [build]: 5.46998e-06
        [elim_shapecalc]: 8.32998e-06
        [elim_not_effective]: 1.442e-05
        [opt_reshape]: 6.44999e-06
        [fold_const_symbol]: 9.12001e-06
        [renormalize]: 9.5999e-07
[validate]: 4.281e-05
[backend_pass]: 8.30012e-07
[task_emit]: 0.00704081
[execute]: 9.38002e-06
Sums
    bootstrap                            :     0.000972s :  5.50%
    type_inference                       :     0.005350s : 30.24%
    event_method                         :     0.000022s :  0.12%
    auto_monad                           :     0.000072s :  0.40%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000046s :  0.26%
    rewriter_before_opt_a                :     0.000077s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000071s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.19%
    jit_opt_a.a_1                        :     0.000695s :  3.93%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000890s :  5.03%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000043s :  0.24%
    jit_opt_a.replace_applicator         :     0.000018s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000684s :  3.87%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000027s :  0.15%
    loop_unroll                          :     0.000492s :  2.78%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000051s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000526s :  2.97%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007041s : 39.80%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000257    23
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000002s :      2: substitution.fold_const_symbol
 2.12% :     0.000005s :      3: substitution.graph_param_transform
85.95% :     0.000221s :      4: substitution.inline
 1.12% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.15% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000005s :      2: substitution.replace_old_param
 5.21% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005260     2
80.18% :     0.004218s :      1: type_inference.infer
19.82% :     0.001042s :      1: type_inference.specialize
------[replace.]   0.000076     6
67.57% :     0.000052s :      4: replace.inline
32.43% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000230     6
94.66% :     0.000218s :      4: match.inline
 5.34% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.67% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000001s :     10: predicate.addn_zero_filter
 2.31% :     0.000003s :     10: predicate.arithmetic_simplify
 1.57% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.54% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.06% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.71% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.48% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.40% :     0.000002s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 6.09% :     0.000007s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.83% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.36% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000000s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.92% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.29% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.10% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.31% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.42% :     0.000009s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.94% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.91% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000729    12
48.66% :     0.000355s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.37% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.97% :     0.000320s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024296    76
 0.22% :     0.000054s :      1: add_recomputation
 0.31% :     0.000075s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 4.08% :     0.000992s :      1: bootstrap
 0.12% :     0.000030s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.11% :     0.000026s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.03% :     0.006325s :      1: jit_opt_a
 0.62% :     0.000152s :      1: jit_opt_after_cconv
 0.22% :     0.000054s :      1: jit_opt_b
 2.06% :     0.000500s :      1: loop_unroll
 2.85% :     0.000694s :      1: mutable_eliminate
 3.77% :     0.000915s :     26: opt.transform.jit_opt_a
 0.19% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000536s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000048s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.04% :     0.000495s :      1: renormalize.infer
 1.58% :     0.000384s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000040s :      1: rewriter_after_opt_a
 0.33% :     0.000080s :      1: rewriter_before_opt_a
 0.34% :     0.000083s :      1: symbol_engine_optimizer
29.06% :     0.007061s :      1: task_emit
22.11% :     0.005372s :      1: type_inference
 0.27% :     0.000066s :      1: validate


TotalTime = 0.0183959, [33]
[bootstrap]: 0.00027893
[type_inference]: 0.00449874
[event_method]: 2.488e-05
[auto_monad]: 6.299e-05
[graph_reusing]: 5.27999e-06
[pre_auto_parallel]: 2.37001e-06
[py_interpret_to_execute]: 3.162e-05
[rewriter_before_opt_a]: 7.076e-05
[expand_dump_flag]: 2.86999e-06
[jit_opt_a]: 0.00527248, [2]
    [Cycle 1]: 0.00154066, [27]
        [switch_simplify]: 4.991e-05
        [loop_unroll]: 2.879e-05
        [a_1]: 0.00052706
        [with_stream_mark]: 1.661e-05
        [recompute_prepare]: 6.90002e-06
        [updatestate_depend_eliminate]: 3.63999e-06
        [updatestate_assign_eliminate]: 3.26001e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 5.96e-06
        [updatestate_useless_node_eliminater]: 5.79e-06
        [accelerated_algorithm]: 5.66e-06
        [meta_shard_fg_expand]: 2.23998e-06
        [get_grad_eliminate_]: 5.32999e-06
        [merge_forward]: 3.86999e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.401e-05
        [j_node_and_user_rematch]: 9.20001e-06
        [meta_fg_expand]: 2.25002e-06
        [replace_old_param]: 8.77e-06
        [inline_without_move]: 5.64e-06
        [renormalize]: 0.0006015
        [add_forward_monad_depend]: 4.72e-06
        [auto_monad_grad]: 1.99999e-06
        [auto_monad_eliminator]: 1.284e-05
        [cse]: 2.742e-05
        [replace_applicator]: 1.116e-05
    [Cycle 2]: 0.00031305, [27]
        [switch_simplify]: 5.67001e-06
        [loop_unroll]: 4.64998e-06
        [a_1]: 8.714e-05
        [with_stream_mark]: 8.69e-06
        [recompute_prepare]: 4.98001e-06
        [updatestate_depend_eliminate]: 2.59001e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 4.99e-06
        [updatestate_useless_node_eliminater]: 4.57e-06
        [accelerated_algorithm]: 4.84003e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 4.53001e-06
        [merge_forward]: 2.14999e-06
        [cell_reuse_recompute_pass]: 1.68002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.17e-05
        [j_node_and_user_rematch]: 7.75e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 6.88e-06
        [inline_without_move]: 4.60001e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 6.09987e-07
        [auto_monad_eliminator]: 5.35999e-06
        [cse]: 9.81e-06
        [replace_applicator]: 5.02e-06
[py_interpret_to_execute_after_opt_a]: 8.28001e-06
[rewriter_after_opt_a]: 3.325e-05
[convert_after_rewriter]: 7.36001e-06
[order_py_execute_after_rewriter]: 5.17e-06
[mutable_eliminate]: 0.00049229
[jit_opt_b]: 4.897e-05, [1]
    [Cycle 1]: 4.166e-05, [2]
        [frontend_op_eliminate]: 1.62e-05
        [inline_after_opt_a]: 1.404e-05
[cconv]: 2.322e-05
[loop_unroll]: 0.00041168
[jit_opt_after_cconv]: 0.00013338, [1]
    [Cycle 1]: 0.00012707, [11]
        [c_1]: 2.026e-05
        [parameter_eliminate]: 2.51998e-06
        [updatestate_depend_eliminate]: 5.46e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.628e-05
        [call_graph_tuple_transform]: 1.72e-05
        [tuple_list_get_item_eliminator]: 5.10001e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 4.60015e-07
        [switch_simplify]: 5.35999e-06
[remove_dup_value]: 1.258e-05
[partial_unused_args_eliminate]: 1.97999e-06
[environ_conv]: 4.80001e-06
[add_recomputation]: 4.707e-05
[cse_after_recomputation]: 2.024e-05, [1]
    [Cycle 1]: 1.484e-05, [1]
        [cse]: 9.05999e-06
[auto_monad_reorder]: 1.471e-05
[get_jit_bprop_graph]: 1.50001e-06
[rewriter_after_jit_bprop_graph]: 5.02999e-06
[opt_after_jit_grad]: 0.00059443
[symbol_engine_optimizer]: 7.178e-05, [1]
    [Cycle 1]: 6.512e-05, [6]
        [build]: 3.53e-06
        [elim_shapecalc]: 7.79997e-06
        [elim_not_effective]: 1.249e-05
        [opt_reshape]: 5.64e-06
        [fold_const_symbol]: 8.33001e-06
        [renormalize]: 4.30009e-07
[validate]: 3.275e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.00598538
[execute]: 6.68e-06
Sums
    bootstrap                            :     0.000279s :  1.94%
    type_inference                       :     0.004499s : 31.26%
    event_method                         :     0.000025s :  0.17%
    auto_monad                           :     0.000063s :  0.44%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.22%
    rewriter_before_opt_a                :     0.000071s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.39%
    jit_opt_a.loop_unroll                :     0.000033s :  0.23%
    jit_opt_a.a_1                        :     0.000614s :  4.27%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000602s :  4.18%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000037s :  0.26%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000492s :  3.42%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000023s :  0.16%
    loop_unroll                          :     0.000412s :  2.86%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000047s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000594s :  4.13%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005985s : 41.59%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000184    23
 1.18% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.62% :     0.000005s :      3: substitution.graph_param_transform
83.15% :     0.000153s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.58% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.71% :     0.000003s :      2: substitution.replace_old_param
 6.48% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004440     2
81.69% :     0.003627s :      1: type_inference.infer
18.31% :     0.000813s :      1: type_inference.specialize
------[replace.]   0.000065     6
65.84% :     0.000043s :      4: replace.inline
34.16% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000161     6
93.30% :     0.000150s :      4: match.inline
 6.70% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     10: predicate.addn_check_dump
 1.44% :     0.000002s :     10: predicate.addn_zero_filter
 2.05% :     0.000002s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.41% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.88% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.62% :     0.000007s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.78% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.82% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000000s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.82% :     0.000002s :     10: predicate.reduce_eliminate
 1.66% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.58% :     0.000002s :     10: predicate.reshape_eliminate
 1.47% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.30% :     0.000003s :     16: predicate.switch_defer_inline
 2.25% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.40% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.57% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.91% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000564    12
46.70% :     0.000263s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.97% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.33% :     0.000273s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019914    76
 0.25% :     0.000050s :      1: add_recomputation
 0.33% :     0.000066s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.47% :     0.000293s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000029s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.49% :     0.005275s :      1: jit_opt_a
 0.68% :     0.000136s :      1: jit_opt_after_cconv
 0.26% :     0.000052s :      1: jit_opt_b
 2.11% :     0.000419s :      1: loop_unroll
 2.51% :     0.000500s :      1: mutable_eliminate
 4.09% :     0.000815s :     26: opt.transform.jit_opt_a
 0.22% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.03% :     0.000603s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000034s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.57% :     0.000312s :      1: renormalize.infer
 1.41% :     0.000281s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000036s :      1: rewriter_after_opt_a
 0.37% :     0.000073s :      1: rewriter_before_opt_a
 0.37% :     0.000074s :      1: symbol_engine_optimizer
30.11% :     0.005996s :      1: task_emit
22.65% :     0.004510s :      1: type_inference
 0.25% :     0.000050s :      1: validate


TotalTime = 0.0140813, [30]
[bootstrap]: 0.00026116
[type_inference]: 0.00620696
[event_method]: 2.005e-05
[auto_monad]: 7.431e-05
[graph_reusing]: 5.42999e-06
[pre_auto_parallel]: 2.32999e-06
[py_interpret_to_execute]: 5.396e-05
[rewriter_before_opt_a]: 8.038e-05
[expand_dump_flag]: 3.76999e-06
[jit_opt_a]: 0.00532191, [2]
    [Cycle 1]: 0.00185101, [27]
        [switch_simplify]: 0.00010106
        [loop_unroll]: 3.498e-05
        [a_1]: 0.0006894
        [with_stream_mark]: 1.411e-05
        [recompute_prepare]: 6.99001e-06
        [updatestate_depend_eliminate]: 3.14001e-06
        [updatestate_assign_eliminate]: 2.62001e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 2.09999e-06
        [specialize_transform]: 6.14999e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 5.56e-06
        [meta_shard_fg_expand]: 1.90001e-06
        [get_grad_eliminate_]: 5.26002e-06
        [merge_forward]: 3.01999e-06
        [cell_reuse_recompute_pass]: 7.80012e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.246e-05
        [j_node_and_user_rematch]: 9.46e-06
        [meta_fg_expand]: 2.21e-06
        [replace_old_param]: 7.77e-06
        [inline_without_move]: 5.49998e-06
        [renormalize]: 0.00071434
        [add_forward_monad_depend]: 4.35999e-06
        [auto_monad_grad]: 1.91e-06
        [auto_monad_eliminator]: 1.283e-05
        [cse]: 1.752e-05
        [replace_applicator]: 1.222e-05
    [Cycle 2]: 0.00031623, [27]
        [switch_simplify]: 5.56e-06
        [loop_unroll]: 4.85999e-06
        [a_1]: 9.262e-05
        [with_stream_mark]: 8.53001e-06
        [recompute_prepare]: 5.17999e-06
        [updatestate_depend_eliminate]: 2.41e-06
        [updatestate_assign_eliminate]: 2.11003e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 5.09998e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 5.12999e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 4.94e-06
        [merge_forward]: 2.37001e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.189e-05
        [j_node_and_user_rematch]: 8.13999e-06
        [meta_fg_expand]: 1.56998e-06
        [replace_old_param]: 7.11999e-06
        [inline_without_move]: 4.94e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.00001e-06
        [auto_monad_grad]: 7.10017e-07
        [auto_monad_eliminator]: 5.10999e-06
        [cse]: 9.94999e-06
        [replace_applicator]: 5.09e-06
[py_interpret_to_execute_after_opt_a]: 9.62999e-06
[rewriter_after_opt_a]: 3.015e-05
[convert_after_rewriter]: 7.41999e-06
[order_py_execute_after_rewriter]: 4.58999e-06
[mutable_eliminate]: 0.00051422
[jit_opt_b]: 5.94e-05, [1]
    [Cycle 1]: 5.281e-05, [2]
        [frontend_op_eliminate]: 1.621e-05
        [inline_after_opt_a]: 2.533e-05
[cconv]: 2.459e-05
[loop_unroll]: 0.00042417
[jit_opt_after_cconv]: 0.00014148, [1]
    [Cycle 1]: 0.00013501, [11]
        [c_1]: 2.24e-05
        [parameter_eliminate]: 2.68e-06
        [updatestate_depend_eliminate]: 5.19e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.708e-05
        [call_graph_tuple_transform]: 1.925e-05
        [tuple_list_get_item_eliminator]: 5.46998e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 5.92001e-06
[remove_dup_value]: 1.291e-05
[partial_unused_args_eliminate]: 2.29001e-06
[environ_conv]: 4.74998e-06
[add_recomputation]: 4.248e-05
[cse_after_recomputation]: 2.289e-05, [1]
    [Cycle 1]: 1.734e-05, [1]
        [cse]: 1.132e-05
[auto_monad_reorder]: 1.414e-05
[get_jit_bprop_graph]: 1.97999e-06
[rewriter_after_jit_bprop_graph]: 5.75001e-06
[opt_after_jit_grad]: 0.00045855
[symbol_engine_optimizer]: 7.343e-05, [1]
    [Cycle 1]: 6.684e-05, [6]
        [build]: 3.59002e-06
        [elim_shapecalc]: 7.46001e-06
        [elim_not_effective]: 1.251e-05
        [opt_reshape]: 5.74999e-06
        [fold_const_symbol]: 9.04998e-06
        [renormalize]: 4.10015e-07
[validate]: 3.227e-05
Sums
    bootstrap                            :     0.000261s :  2.52%
    type_inference                       :     0.006207s : 59.89%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000074s :  0.72%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000054s :  0.52%
    rewriter_before_opt_a                :     0.000080s :  0.78%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000107s :  1.03%
    jit_opt_a.loop_unroll                :     0.000040s :  0.38%
    jit_opt_a.a_1                        :     0.000782s :  7.55%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000714s :  6.89%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.17%
    jit_opt_a.cse                        :     0.000027s :  0.27%
    jit_opt_a.replace_applicator         :     0.000017s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.09%
    rewriter_after_opt_a                 :     0.000030s :  0.29%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000514s :  4.96%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.24%
    cconv                                :     0.000025s :  0.24%
    loop_unroll                          :     0.000424s :  4.09%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000013s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000042s :  0.41%
    cse_after_recomputation.cse          :     0.000011s :  0.11%
    auto_monad_reorder                   :     0.000014s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000459s :  4.42%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.31%


Time group info:
------[substitution.]   0.000270    26
 0.68% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 1.79% :     0.000005s :      3: substitution.graph_param_transform
85.84% :     0.000232s :      6: substitution.inline
 1.15% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.52% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.13% :     0.000003s :      2: substitution.replace_old_param
 3.18% :     0.000009s :      1: substitution.switch_simplify
 4.25% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006151     2
80.80% :     0.004969s :      1: type_inference.infer
19.20% :     0.001181s :      1: type_inference.specialize
------[replace.]   0.000126     9
54.26% :     0.000068s :      6: replace.inline
20.66% :     0.000026s :      1: replace.switch_simplify
25.08% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000246     9
92.64% :     0.000228s :      6: match.inline
 3.18% :     0.000008s :      1: match.switch_simplify
 4.18% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   801
 1.34% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     12: predicate.addn_check_dump
 1.44% :     0.000002s :     12: predicate.addn_zero_filter
 2.19% :     0.000003s :     12: predicate.arithmetic_simplify
 1.34% :     0.000002s :     12: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     12: predicate.compare_switch_simplify
 1.45% :     0.000002s :     12: predicate.depend_value_elim
 1.16% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     12: predicate.environ_get_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.51% :     0.000001s :      3: predicate.graph_param_transform
 5.38% :     0.000007s :     26: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.76% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     17: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.19% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.10% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     12: predicate.merge_addn
 1.18% :     0.000002s :     12: predicate.minmaximum_grad
 1.42% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     17: predicate.partial_eliminate
 1.33% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.83% :     0.000002s :     12: predicate.reduce_eliminate
 1.50% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     20: predicate.replace_applicator
 0.45% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     12: predicate.reshape_eliminate
 1.25% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     12: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.84% :     0.000004s :     20: predicate.switch_defer_inline
 2.31% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.08% :     0.000011s :     65: predicate.switch_simplify
 1.28% :     0.000002s :     12: predicate.tile_eliminate
 1.31% :     0.000002s :     12: predicate.transpose_eliminate
 1.60% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.46% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.83% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     12: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000841    15
44.65% :     0.000376s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.90% :     0.000092s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.45% :     0.000374s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015910    72
 0.28% :     0.000045s :      1: add_recomputation
 0.49% :     0.000078s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.72% :     0.000274s :      1: bootstrap
 0.17% :     0.000028s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
33.47% :     0.005325s :      1: jit_opt_a
 0.91% :     0.000145s :      1: jit_opt_after_cconv
 0.39% :     0.000062s :      1: jit_opt_b
 2.72% :     0.000432s :      1: loop_unroll
 3.29% :     0.000523s :      1: mutable_eliminate
 6.55% :     0.001042s :     26: opt.transform.jit_opt_a
 0.31% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000035s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.93% :     0.000467s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.36% :     0.000057s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000015s :      1: remove_dup_value
 2.31% :     0.000367s :      1: renormalize.infer
 2.13% :     0.000339s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000033s :      1: rewriter_after_opt_a
 0.52% :     0.000083s :      1: rewriter_before_opt_a
 0.48% :     0.000076s :      1: symbol_engine_optimizer
39.09% :     0.006219s :      1: type_inference


TotalTime = 0.0176383, [33]
[bootstrap]: 0.000262
[type_inference]: 0.0042949
[event_method]: 1.625e-05
[auto_monad]: 6.134e-05
[graph_reusing]: 4.75999e-06
[pre_auto_parallel]: 2.45997e-06
[py_interpret_to_execute]: 3.035e-05
[rewriter_before_opt_a]: 6.812e-05
[expand_dump_flag]: 3.26001e-06
[jit_opt_a]: 0.00483035, [2]
    [Cycle 1]: 0.0013558, [27]
        [switch_simplify]: 4.77e-05
        [loop_unroll]: 2.872e-05
        [a_1]: 0.00048534
        [with_stream_mark]: 1.522e-05
        [recompute_prepare]: 7.83001e-06
        [updatestate_depend_eliminate]: 3.29001e-06
        [updatestate_assign_eliminate]: 2.64999e-06
        [updatestate_loads_eliminate]: 2.52001e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 5.69e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.56e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 5.19998e-06
        [merge_forward]: 3.23998e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.339e-05
        [j_node_and_user_rematch]: 8.71002e-06
        [meta_fg_expand]: 2.17001e-06
        [replace_old_param]: 7.8e-06
        [inline_without_move]: 5.16002e-06
        [renormalize]: 0.00048004
        [add_forward_monad_depend]: 4.85999e-06
        [auto_monad_grad]: 1.87999e-06
        [auto_monad_eliminator]: 1.25e-05
        [cse]: 2.42e-05
        [replace_applicator]: 1.122e-05
    [Cycle 2]: 0.00033825, [27]
        [switch_simplify]: 5.35999e-06
        [loop_unroll]: 5.37001e-06
        [a_1]: 8.809e-05
        [with_stream_mark]: 8.1e-06
        [recompute_prepare]: 4.97e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 5.05001e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 1.25001e-06
        [get_grad_eliminate_]: 4.25e-06
        [merge_forward]: 2.58e-06
        [cell_reuse_recompute_pass]: 1.47999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.198e-05
        [j_node_and_user_rematch]: 7.73999e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 1.27e-05
        [inline_without_move]: 4.63999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.38002e-06
        [auto_monad_grad]: 8.49977e-07
        [auto_monad_eliminator]: 5.52001e-06
        [cse]: 1.125e-05
        [replace_applicator]: 5.06002e-06
[py_interpret_to_execute_after_opt_a]: 2.387e-05
[rewriter_after_opt_a]: 3.142e-05
[convert_after_rewriter]: 7.18998e-06
[order_py_execute_after_rewriter]: 5.54e-06
[mutable_eliminate]: 0.00050667
[jit_opt_b]: 4.739e-05, [1]
    [Cycle 1]: 4.107e-05, [2]
        [frontend_op_eliminate]: 1.513e-05
        [inline_after_opt_a]: 1.486e-05
[cconv]: 2.303e-05
[loop_unroll]: 0.00046727
[jit_opt_after_cconv]: 0.00013879, [1]
    [Cycle 1]: 0.00013224, [11]
        [c_1]: 2.06e-05
        [parameter_eliminate]: 3.03e-06
        [updatestate_depend_eliminate]: 5.47001e-06
        [updatestate_assign_eliminate]: 2.63e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.834e-05
        [call_graph_tuple_transform]: 1.831e-05
        [tuple_list_get_item_eliminator]: 5.29e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 5.21998e-06
[remove_dup_value]: 1.323e-05
[partial_unused_args_eliminate]: 2.19001e-06
[environ_conv]: 4.59998e-06
[add_recomputation]: 4.458e-05
[cse_after_recomputation]: 1.951e-05, [1]
    [Cycle 1]: 1.458e-05, [1]
        [cse]: 8.97e-06
[auto_monad_reorder]: 1.45e-05
[get_jit_bprop_graph]: 1.39e-06
[rewriter_after_jit_bprop_graph]: 4.88001e-06
[opt_after_jit_grad]: 0.00044377
[symbol_engine_optimizer]: 7.131e-05, [1]
    [Cycle 1]: 6.493e-05, [6]
        [build]: 3.8e-06
        [elim_shapecalc]: 7.64002e-06
        [elim_not_effective]: 1.25e-05
        [opt_reshape]: 5.37999e-06
        [fold_const_symbol]: 8.65999e-06
        [renormalize]: 4.30009e-07
[validate]: 3.3e-05
[backend_pass]: 1.14e-06
[task_emit]: 0.00597747
[execute]: 6.47001e-06
Sums
    bootstrap                            :     0.000262s :  1.88%
    type_inference                       :     0.004295s : 30.86%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000061s :  0.44%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.22%
    rewriter_before_opt_a                :     0.000068s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000573s :  4.12%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000020s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000480s :  3.45%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000035s :  0.25%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000024s :  0.17%
    rewriter_after_opt_a                 :     0.000031s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000507s :  3.64%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000467s :  3.36%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000045s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000444s :  3.19%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005977s : 42.95%
    execute                              :     0.000006s :  0.05%


Time group info:
------[substitution.]   0.000178    23
 1.06% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 3.12% :     0.000006s :      3: substitution.graph_param_transform
79.31% :     0.000141s :      4: substitution.inline
 1.67% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.53% :     0.000004s :      4: substitution.remove_not_recompute_node
 5.08% :     0.000009s :      2: substitution.replace_old_param
 6.50% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004239     2
82.53% :     0.003498s :      1: type_inference.infer
17.47% :     0.000741s :      1: type_inference.specialize
------[replace.]   0.000060     6
65.89% :     0.000040s :      4: replace.inline
34.11% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000149     6
93.02% :     0.000138s :      4: match.inline
 6.98% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.33% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 2.00% :     0.000002s :     10: predicate.arithmetic_simplify
 1.36% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.43% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.74% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.68% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.19% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.99% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000000s :      3: predicate.opt_reshape
 2.34% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.78% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.62% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.51% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.72% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.96% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000500    12
49.58% :     0.000248s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.18% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.23% :     0.000226s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019001    76
 0.25% :     0.000047s :      1: add_recomputation
 0.34% :     0.000065s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.44% :     0.000274s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
25.43% :     0.004833s :      1: jit_opt_a
 0.74% :     0.000142s :      1: jit_opt_after_cconv
 0.26% :     0.000050s :      1: jit_opt_b
 2.51% :     0.000476s :      1: loop_unroll
 2.71% :     0.000515s :      1: mutable_eliminate
 4.08% :     0.000776s :     26: opt.transform.jit_opt_a
 0.24% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.38% :     0.000451s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.14% :     0.000026s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.27% :     0.000242s :      1: renormalize.infer
 1.22% :     0.000232s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.37% :     0.000071s :      1: rewriter_before_opt_a
 0.39% :     0.000074s :      1: symbol_engine_optimizer
31.51% :     0.005987s :      1: task_emit
22.66% :     0.004305s :      1: type_inference
 0.27% :     0.000052s :      1: validate


TotalTime = 0.022713, [33]
[bootstrap]: 0.00037521
[type_inference]: 0.00558493
[event_method]: 2.083e-05
[auto_monad]: 7.182e-05
[graph_reusing]: 6.58e-06
[pre_auto_parallel]: 2.88e-06
[py_interpret_to_execute]: 5.016e-05
[rewriter_before_opt_a]: 8.013e-05
[expand_dump_flag]: 3.79002e-06
[jit_opt_a]: 0.00648568, [2]
    [Cycle 1]: 0.00186285, [27]
        [switch_simplify]: 6.726e-05
        [loop_unroll]: 3.024e-05
        [a_1]: 0.00060595
        [with_stream_mark]: 2.064e-05
        [recompute_prepare]: 8.48001e-06
        [updatestate_depend_eliminate]: 4.15e-06
        [updatestate_assign_eliminate]: 2.97002e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 2.29001e-06
        [specialize_transform]: 6.19001e-06
        [updatestate_useless_node_eliminater]: 5.67001e-06
        [accelerated_algorithm]: 5.82001e-06
        [meta_shard_fg_expand]: 2.48002e-06
        [get_grad_eliminate_]: 5.86998e-06
        [merge_forward]: 3.83001e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.632e-05
        [j_node_and_user_rematch]: 9.81e-06
        [meta_fg_expand]: 2.95002e-06
        [replace_old_param]: 9.05001e-06
        [inline_without_move]: 5.49e-06
        [renormalize]: 0.00079762
        [add_forward_monad_depend]: 5.39e-06
        [auto_monad_grad]: 1.97999e-06
        [auto_monad_eliminator]: 1.236e-05
        [cse]: 2.773e-05
        [replace_applicator]: 1.276e-05
    [Cycle 2]: 0.00032939, [27]
        [switch_simplify]: 5.53997e-06
        [loop_unroll]: 5.07e-06
        [a_1]: 9.303e-05
        [with_stream_mark]: 1e-05
        [recompute_prepare]: 4.70999e-06
        [updatestate_depend_eliminate]: 2.97002e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.75001e-06
        [updatestate_useless_node_eliminater]: 4.76002e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.18001e-06
        [get_grad_eliminate_]: 4.89e-06
        [merge_forward]: 2.76e-06
        [cell_reuse_recompute_pass]: 1.94e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.286e-05
        [j_node_and_user_rematch]: 8.79e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 7.40998e-06
        [inline_without_move]: 4.99e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 5.32001e-06
        [cse]: 1.1e-05
        [replace_applicator]: 5.60001e-06
[py_interpret_to_execute_after_opt_a]: 1.17e-05
[rewriter_after_opt_a]: 3.597e-05
[convert_after_rewriter]: 6.81999e-06
[order_py_execute_after_rewriter]: 7.32002e-06
[mutable_eliminate]: 0.00082567
[jit_opt_b]: 5.584e-05, [1]
    [Cycle 1]: 4.682e-05, [2]
        [frontend_op_eliminate]: 1.779e-05
        [inline_after_opt_a]: 1.608e-05
[cconv]: 3.109e-05
[loop_unroll]: 0.00048437
[jit_opt_after_cconv]: 0.00016575, [1]
    [Cycle 1]: 0.00015752, [11]
        [c_1]: 2.231e-05
        [parameter_eliminate]: 4.49998e-06
        [updatestate_depend_eliminate]: 8.03001e-06
        [updatestate_assign_eliminate]: 3.26999e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [cse]: 2.869e-05
        [call_graph_tuple_transform]: 2.251e-05
        [tuple_list_get_item_eliminator]: 5.29998e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 1.10999e-06
        [switch_simplify]: 6.40002e-06
[remove_dup_value]: 1.554e-05
[partial_unused_args_eliminate]: 2.54001e-06
[environ_conv]: 5.99999e-06
[add_recomputation]: 5.267e-05
[cse_after_recomputation]: 2.197e-05, [1]
    [Cycle 1]: 1.625e-05, [1]
        [cse]: 1.043e-05
[auto_monad_reorder]: 1.691e-05
[get_jit_bprop_graph]: 2.41e-06
[rewriter_after_jit_bprop_graph]: 7.79002e-06
[opt_after_jit_grad]: 0.00050349
[symbol_engine_optimizer]: 8.094e-05, [1]
    [Cycle 1]: 7.358e-05, [6]
        [build]: 5.13002e-06
        [elim_shapecalc]: 7.97e-06
        [elim_not_effective]: 1.465e-05
        [opt_reshape]: 6.40002e-06
        [fold_const_symbol]: 9.39998e-06
        [renormalize]: 6.99976e-07
[validate]: 4.559e-05
[backend_pass]: 9.10019e-07
[task_emit]: 0.00739767
[execute]: 1.056e-05
Sums
    bootstrap                            :     0.000375s :  2.11%
    type_inference                       :     0.005585s : 31.47%
    event_method                         :     0.000021s :  0.12%
    auto_monad                           :     0.000072s :  0.40%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000050s :  0.28%
    rewriter_before_opt_a                :     0.000080s :  0.45%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000073s :  0.41%
    jit_opt_a.loop_unroll                :     0.000035s :  0.20%
    jit_opt_a.a_1                        :     0.000699s :  3.94%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000798s :  4.50%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.10%
    jit_opt_a.cse                        :     0.000039s :  0.22%
    jit_opt_a.replace_applicator         :     0.000018s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000007s :  0.04%
    mutable_eliminate                    :     0.000826s :  4.65%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000031s :  0.18%
    loop_unroll                          :     0.000484s :  2.73%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000053s :  0.30%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000503s :  2.84%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007398s : 41.69%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000254    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000002s :      2: substitution.fold_const_symbol
 2.23% :     0.000006s :      3: substitution.graph_param_transform
86.20% :     0.000219s :      4: substitution.inline
 1.27% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.17% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.61% :     0.000004s :      2: substitution.replace_old_param
 5.11% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005496     2
80.57% :     0.004428s :      1: type_inference.infer
19.43% :     0.001068s :      1: type_inference.specialize
------[replace.]   0.000079     6
66.84% :     0.000053s :      4: replace.inline
33.16% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000228     6
94.80% :     0.000217s :      4: match.inline
 5.20% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.15% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000002s :     10: predicate.addn_zero_filter
 2.09% :     0.000003s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.31% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.09% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.55% :     0.000002s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.76% :     0.000007s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.35% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.61% :     0.000002s :     15: predicate.load_eliminater
 1.71% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.21% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.35% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 2.33% :     0.000003s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     15: predicate.partial_eliminate
 1.03% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.33% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.54% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.47% :     0.000001s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 1.19% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.00% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.80% :     0.000010s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.04% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.11% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.08% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.05% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.48% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000712    12
48.57% :     0.000346s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.22% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.21% :     0.000322s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024543    76
 0.23% :     0.000056s :      1: add_recomputation
 0.31% :     0.000076s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.61% :     0.000396s :      1: bootstrap
 0.14% :     0.000034s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000025s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.44% :     0.006490s :      1: jit_opt_a
 0.69% :     0.000169s :      1: jit_opt_after_cconv
 0.24% :     0.000059s :      1: jit_opt_b
 2.01% :     0.000493s :      1: loop_unroll
 3.41% :     0.000838s :      1: mutable_eliminate
 3.77% :     0.000926s :     26: opt.transform.jit_opt_a
 0.22% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.08% :     0.000511s :      1: opt_after_jit_grad
 0.04% :     0.000010s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000054s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 1.74% :     0.000426s :      1: renormalize.infer
 1.48% :     0.000363s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000039s :      1: rewriter_after_opt_a
 0.34% :     0.000084s :      1: rewriter_before_opt_a
 0.34% :     0.000083s :      1: symbol_engine_optimizer
30.22% :     0.007418s :      1: task_emit
22.84% :     0.005607s :      1: type_inference
 0.27% :     0.000067s :      1: validate


TotalTime = 0.0156656, [30]
[bootstrap]: 0.0002646
[type_inference]: 0.00651636
[event_method]: 2.229e-05
[auto_monad]: 8.329e-05
[graph_reusing]: 6.39001e-06
[pre_auto_parallel]: 2.47001e-06
[py_interpret_to_execute]: 6.652e-05
[rewriter_before_opt_a]: 8.63e-05
[expand_dump_flag]: 4.05e-06
[jit_opt_a]: 0.00619193, [2]
    [Cycle 1]: 0.00216669, [27]
        [switch_simplify]: 0.00013819
        [loop_unroll]: 3.517e-05
        [a_1]: 0.00076549
        [with_stream_mark]: 1.884e-05
        [recompute_prepare]: 7.23e-06
        [updatestate_depend_eliminate]: 3.66001e-06
        [updatestate_assign_eliminate]: 2.92002e-06
        [updatestate_loads_eliminate]: 2.75002e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 5.89999e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 6.26e-06
        [meta_shard_fg_expand]: 2.18998e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 3.63999e-06
        [cell_reuse_recompute_pass]: 1.31998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.413e-05
        [j_node_and_user_rematch]: 8.82e-06
        [meta_fg_expand]: 2.19999e-06
        [replace_old_param]: 8.80999e-06
        [inline_without_move]: 5.26002e-06
        [renormalize]: 0.00087628
        [add_forward_monad_depend]: 5.71998e-06
        [auto_monad_grad]: 2.66e-06
        [auto_monad_eliminator]: 1.435e-05
        [cse]: 2.622e-05
        [replace_applicator]: 1.296e-05
    [Cycle 2]: 0.00032834, [27]
        [switch_simplify]: 6.14001e-06
        [loop_unroll]: 5.20999e-06
        [a_1]: 9.46e-05
        [with_stream_mark]: 1.087e-05
        [recompute_prepare]: 5.09e-06
        [updatestate_depend_eliminate]: 2.58e-06
        [updatestate_assign_eliminate]: 2.11998e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [parameter_eliminate]: 1.46998e-06
        [specialize_transform]: 5.52999e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.10999e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 4.65001e-06
        [merge_forward]: 3.53e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.273e-05
        [j_node_and_user_rematch]: 7.8e-06
        [meta_fg_expand]: 2.09999e-06
        [replace_old_param]: 7.63999e-06
        [inline_without_move]: 5.19e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 6.60017e-07
        [auto_monad_eliminator]: 5.56e-06
        [cse]: 1.08e-05
        [replace_applicator]: 5.21002e-06
[py_interpret_to_execute_after_opt_a]: 1.251e-05
[rewriter_after_opt_a]: 3.483e-05
[convert_after_rewriter]: 7.04001e-06
[order_py_execute_after_rewriter]: 4.82e-06
[mutable_eliminate]: 0.00068533
[jit_opt_b]: 7.061e-05, [1]
    [Cycle 1]: 6.205e-05, [2]
        [frontend_op_eliminate]: 3.24e-05
        [inline_after_opt_a]: 1.688e-05
[cconv]: 2.867e-05
[loop_unroll]: 0.00050845
[jit_opt_after_cconv]: 0.00014847, [1]
    [Cycle 1]: 0.00014181, [11]
        [c_1]: 2.23e-05
        [parameter_eliminate]: 3.49001e-06
        [updatestate_depend_eliminate]: 5.99999e-06
        [updatestate_assign_eliminate]: 2.54001e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [cse]: 2.013e-05
        [call_graph_tuple_transform]: 2e-05
        [tuple_list_get_item_eliminator]: 5.72999e-06
        [none_parameter_eliminate]: 1.62001e-06
        [renormalize]: 7.80012e-07
        [switch_simplify]: 5.44e-06
[remove_dup_value]: 1.509e-05
[partial_unused_args_eliminate]: 2.49001e-06
[environ_conv]: 5.77999e-06
[add_recomputation]: 5.043e-05
[cse_after_recomputation]: 2.273e-05, [1]
    [Cycle 1]: 1.667e-05, [1]
        [cse]: 1.023e-05
[auto_monad_reorder]: 1.579e-05
[get_jit_bprop_graph]: 2.06e-06
[rewriter_after_jit_bprop_graph]: 7.36999e-06
[opt_after_jit_grad]: 0.00049982
[symbol_engine_optimizer]: 7.989e-05, [1]
    [Cycle 1]: 7.333e-05, [6]
        [build]: 3.74002e-06
        [elim_shapecalc]: 8.51002e-06
        [elim_not_effective]: 1.461e-05
        [opt_reshape]: 6.71e-06
        [fold_const_symbol]: 8.38999e-06
        [renormalize]: 6.89994e-07
[validate]: 4.205e-05
Sums
    bootstrap                            :     0.000265s :  2.33%
    type_inference                       :     0.006516s : 57.33%
    event_method                         :     0.000022s :  0.20%
    auto_monad                           :     0.000083s :  0.73%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000067s :  0.59%
    rewriter_before_opt_a                :     0.000086s :  0.76%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000144s :  1.27%
    jit_opt_a.loop_unroll                :     0.000040s :  0.36%
    jit_opt_a.a_1                        :     0.000860s :  7.57%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000876s :  7.71%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.18%
    jit_opt_a.cse                        :     0.000037s :  0.33%
    jit_opt_a.replace_applicator         :     0.000018s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.11%
    rewriter_after_opt_a                 :     0.000035s :  0.31%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000685s :  6.03%
    jit_opt_b.frontend_op_eliminate      :     0.000032s :  0.29%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.15%
    cconv                                :     0.000029s :  0.25%
    loop_unroll                          :     0.000508s :  4.47%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000015s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000050s :  0.44%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000500s :  4.40%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000042s :  0.37%


Time group info:
------[substitution.]   0.000328    26
 0.66% :     0.000002s :      2: substitution.elim_not_effective
 0.41% :     0.000001s :      2: substitution.fold_const_symbol
 1.71% :     0.000006s :      3: substitution.graph_param_transform
86.98% :     0.000286s :      6: substitution.inline
 0.88% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.36% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.18% :     0.000004s :      2: substitution.replace_old_param
 2.99% :     0.000010s :      1: substitution.switch_simplify
 3.82% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006447     2
81.02% :     0.005223s :      1: type_inference.infer
18.98% :     0.001224s :      1: type_inference.specialize
------[replace.]   0.000151     9
48.23% :     0.000073s :      6: replace.inline
27.70% :     0.000042s :      1: replace.switch_simplify
24.07% :     0.000036s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000302     9
93.31% :     0.000282s :      6: match.inline
 2.99% :     0.000009s :      1: match.switch_simplify
 3.71% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000142   801
 1.36% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.28% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000002s :     12: predicate.addn_check_dump
 1.47% :     0.000002s :     12: predicate.addn_zero_filter
 2.16% :     0.000003s :     12: predicate.arithmetic_simplify
 1.43% :     0.000002s :     12: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000002s :     12: predicate.compare_switch_simplify
 1.90% :     0.000003s :     12: predicate.depend_value_elim
 1.19% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.07% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.07% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     12: predicate.environ_get_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 6.28% :     0.000009s :     26: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000002s :      6: predicate.less_batch_normalization
 1.67% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     17: predicate.load_eliminater
 1.48% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.94% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     12: predicate.merge_addn
 1.13% :     0.000002s :     12: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     17: predicate.partial_eliminate
 1.19% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.78% :     0.000003s :     12: predicate.reduce_eliminate
 1.49% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     20: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     12: predicate.reshape_eliminate
 1.31% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.79% :     0.000001s :      6: predicate.special_op_eliminate
 0.85% :     0.000001s :      6: predicate.specialize_transform
 1.52% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.56% :     0.000004s :     20: predicate.switch_defer_inline
 2.21% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.18% :     0.000012s :     65: predicate.switch_simplify
 1.28% :     0.000002s :     12: predicate.tile_eliminate
 1.19% :     0.000002s :     12: predicate.transpose_eliminate
 1.73% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.84% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.47% :     0.000002s :     12: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000910    15
48.77% :     0.000444s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.13% :     0.000120s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.10% :     0.000347s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017771    72
 0.30% :     0.000053s :      1: add_recomputation
 0.49% :     0.000087s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 1.55% :     0.000276s :      1: bootstrap
 0.18% :     0.000032s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000026s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
34.87% :     0.006198s :      1: jit_opt_a
 0.85% :     0.000152s :      1: jit_opt_after_cconv
 0.41% :     0.000073s :      1: jit_opt_b
 2.91% :     0.000517s :      1: loop_unroll
 3.92% :     0.000696s :      1: mutable_eliminate
 6.52% :     0.001159s :     26: opt.transform.jit_opt_a
 0.28% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.23% :     0.000042s :      4: opt.transform.jit_opt_b
 0.08% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.86% :     0.000509s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.40% :     0.000070s :      1: py_interpret_to_execute
 0.08% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000018s :      1: remove_dup_value
 2.86% :     0.000509s :      1: renormalize.infer
 2.02% :     0.000359s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000038s :      1: rewriter_after_opt_a
 0.50% :     0.000089s :      1: rewriter_before_opt_a
 0.47% :     0.000083s :      1: symbol_engine_optimizer
36.74% :     0.006529s :      1: type_inference


TotalTime = 0.0221441, [33]
[bootstrap]: 0.00042099
[type_inference]: 0.00585977
[event_method]: 1.82e-05
[auto_monad]: 7.113e-05
[graph_reusing]: 6.89001e-06
[pre_auto_parallel]: 2.46998e-06
[py_interpret_to_execute]: 4.654e-05
[rewriter_before_opt_a]: 7.677e-05
[expand_dump_flag]: 2.99999e-06
[jit_opt_a]: 0.00656785, [2]
    [Cycle 1]: 0.00227307, [27]
        [switch_simplify]: 6.229e-05
        [loop_unroll]: 3.024e-05
        [a_1]: 0.00092462
        [with_stream_mark]: 2.346e-05
        [recompute_prepare]: 8.50001e-06
        [updatestate_depend_eliminate]: 4.72e-06
        [updatestate_assign_eliminate]: 3.31001e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 1.92999e-06
        [specialize_transform]: 6.02999e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.81e-06
        [meta_shard_fg_expand]: 2.12999e-06
        [get_grad_eliminate_]: 5.45001e-06
        [merge_forward]: 4.40999e-06
        [cell_reuse_recompute_pass]: 1.28002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.459e-05
        [j_node_and_user_rematch]: 1.017e-05
        [meta_fg_expand]: 2.32999e-06
        [replace_old_param]: 1.121e-05
        [inline_without_move]: 5.42999e-06
        [renormalize]: 0.00087241
        [add_forward_monad_depend]: 6.26998e-06
        [auto_monad_grad]: 2.39001e-06
        [auto_monad_eliminator]: 1.997e-05
        [cse]: 2.785e-05
        [replace_applicator]: 1.4e-05
    [Cycle 2]: 0.00033598, [27]
        [switch_simplify]: 6.06998e-06
        [loop_unroll]: 4.80001e-06
        [a_1]: 9.125e-05
        [with_stream_mark]: 1.231e-05
        [recompute_prepare]: 4.85001e-06
        [updatestate_depend_eliminate]: 3.3e-06
        [updatestate_assign_eliminate]: 2.33998e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.12999e-06
        [specialize_transform]: 5.04e-06
        [updatestate_useless_node_eliminater]: 4.73001e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 1.23002e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.32001e-06
        [cell_reuse_recompute_pass]: 1.89e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.507e-05
        [j_node_and_user_rematch]: 7.97e-06
        [meta_fg_expand]: 1.40999e-06
        [replace_old_param]: 7.85998e-06
        [inline_without_move]: 5.04e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.26002e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 5.54e-06
        [cse]: 1.167e-05
        [replace_applicator]: 4.97e-06
[py_interpret_to_execute_after_opt_a]: 1.26e-05
[rewriter_after_opt_a]: 3.578e-05
[convert_after_rewriter]: 7.31001e-06
[order_py_execute_after_rewriter]: 5.22e-06
[mutable_eliminate]: 0.0007115
[jit_opt_b]: 5.006e-05, [1]
    [Cycle 1]: 4.331e-05, [2]
        [frontend_op_eliminate]: 1.643e-05
        [inline_after_opt_a]: 1.51e-05
[cconv]: 2.748e-05
[loop_unroll]: 0.00043045
[jit_opt_after_cconv]: 0.00014268, [1]
    [Cycle 1]: 0.00013645, [11]
        [c_1]: 2.019e-05
        [parameter_eliminate]: 2.77002e-06
        [updatestate_depend_eliminate]: 7.06999e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 2.121e-05
        [call_graph_tuple_transform]: 1.915e-05
        [tuple_list_get_item_eliminator]: 5.04e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 4.10015e-07
        [switch_simplify]: 5.36002e-06
[remove_dup_value]: 1.324e-05
[partial_unused_args_eliminate]: 2.14999e-06
[environ_conv]: 5.24e-06
[add_recomputation]: 5.018e-05
[cse_after_recomputation]: 2.031e-05, [1]
    [Cycle 1]: 1.505e-05, [1]
        [cse]: 9.51003e-06
[auto_monad_reorder]: 1.518e-05
[get_jit_bprop_graph]: 1.93002e-06
[rewriter_after_jit_bprop_graph]: 5.91998e-06
[opt_after_jit_grad]: 0.00054256
[symbol_engine_optimizer]: 7.412e-05, [1]
    [Cycle 1]: 6.72e-05, [6]
        [build]: 3.71001e-06
        [elim_shapecalc]: 7.63999e-06
        [elim_not_effective]: 1.27e-05
        [opt_reshape]: 5.72999e-06
        [fold_const_symbol]: 9.24e-06
        [renormalize]: 4.69998e-07
[validate]: 4.026e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00665001
[execute]: 1.032e-05
Sums
    bootstrap                            :     0.000421s :  2.40%
    type_inference                       :     0.005860s : 33.42%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000071s :  0.41%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000047s :  0.27%
    rewriter_before_opt_a                :     0.000077s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.20%
    jit_opt_a.a_1                        :     0.001016s :  5.79%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000872s :  4.98%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.15%
    jit_opt_a.cse                        :     0.000040s :  0.23%
    jit_opt_a.replace_applicator         :     0.000019s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000711s :  4.06%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000027s :  0.16%
    loop_unroll                          :     0.000430s :  2.45%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000543s :  3.09%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000040s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006650s : 37.92%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000252    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000002s :      2: substitution.fold_const_symbol
 2.37% :     0.000006s :      3: substitution.graph_param_transform
84.82% :     0.000214s :      4: substitution.inline
 1.67% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.80% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.55% :     0.000006s :      2: substitution.replace_old_param
 5.44% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005775     2
83.34% :     0.004813s :      1: type_inference.infer
16.66% :     0.000962s :      1: type_inference.specialize
------[replace.]   0.000092     6
65.88% :     0.000061s :      4: replace.inline
34.12% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000223     6
94.44% :     0.000211s :      4: match.inline
 5.56% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.13% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.57% :     0.000002s :     10: predicate.addn_zero_filter
 4.54% :     0.000006s :     10: predicate.arithmetic_simplify
 1.17% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.91% :     0.000001s :     10: predicate.compare_switch_simplify
 1.80% :     0.000002s :     10: predicate.depend_value_elim
 1.30% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.77% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.24% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.01% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     10: predicate.environ_get_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.18% :     0.000007s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :      6: predicate.less_batch_normalization
 1.37% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.57% :     0.000002s :     15: predicate.load_eliminater
 1.58% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.64% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.71% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 1.17% :     0.000002s :     10: predicate.minmaximum_grad
 1.79% :     0.000002s :      3: predicate.mutable_eliminate
 0.39% :     0.000001s :      3: predicate.opt_reshape
 2.00% :     0.000003s :     15: predicate.partial_eliminate
 1.01% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.52% :     0.000003s :     10: predicate.reduce_eliminate
 1.56% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.55% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.82% :     0.000002s :     10: predicate.reshape_eliminate
 1.80% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.09% :     0.000001s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.15% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.32% :     0.000003s :     16: predicate.switch_defer_inline
 1.93% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.36% :     0.000010s :     53: predicate.switch_simplify
 1.05% :     0.000001s :     10: predicate.tile_eliminate
 1.28% :     0.000002s :     10: predicate.transpose_eliminate
 1.90% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 2.00% :     0.000003s :     10: predicate.tuple_list_get_item_depend_reorder
 3.66% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.29% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000708    12
49.94% :     0.000353s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.68% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.39% :     0.000300s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024353    76
 0.22% :     0.000053s :      1: add_recomputation
 0.31% :     0.000075s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.80% :     0.000438s :      1: bootstrap
 0.12% :     0.000030s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000023s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.99% :     0.006572s :      1: jit_opt_a
 0.60% :     0.000146s :      1: jit_opt_after_cconv
 0.22% :     0.000053s :      1: jit_opt_b
 1.80% :     0.000438s :      1: loop_unroll
 2.95% :     0.000719s :      1: mutable_eliminate
 5.09% :     0.001240s :     26: opt.transform.jit_opt_a
 0.19% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.26% :     0.000551s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000050s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000016s :      1: remove_dup_value
 2.01% :     0.000489s :      1: renormalize.infer
 1.53% :     0.000373s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000039s :      1: rewriter_after_opt_a
 0.33% :     0.000080s :      1: rewriter_before_opt_a
 0.32% :     0.000077s :      1: symbol_engine_optimizer
27.39% :     0.006669s :      1: task_emit
24.13% :     0.005876s :      1: type_inference
 0.25% :     0.000061s :      1: validate


TotalTime = 0.0194854, [33]
[bootstrap]: 0.00030297
[type_inference]: 0.00471413
[event_method]: 1.667e-05
[auto_monad]: 6.14e-05
[graph_reusing]: 5.45001e-06
[pre_auto_parallel]: 2.53998e-06
[py_interpret_to_execute]: 3.227e-05
[rewriter_before_opt_a]: 7.024e-05
[expand_dump_flag]: 2.99001e-06
[jit_opt_a]: 0.00580147, [2]
    [Cycle 1]: 0.00170031, [27]
        [switch_simplify]: 6.023e-05
        [loop_unroll]: 3.174e-05
        [a_1]: 0.00056528
        [with_stream_mark]: 2.151e-05
        [recompute_prepare]: 7.68999e-06
        [updatestate_depend_eliminate]: 4.13001e-06
        [updatestate_assign_eliminate]: 3.81001e-06
        [updatestate_loads_eliminate]: 2.65002e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 6.31998e-06
        [updatestate_useless_node_eliminater]: 5.48997e-06
        [accelerated_algorithm]: 5.59998e-06
        [meta_shard_fg_expand]: 2.31998e-06
        [get_grad_eliminate_]: 5.84999e-06
        [merge_forward]: 3.58999e-06
        [cell_reuse_recompute_pass]: 1.58002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.592e-05
        [j_node_and_user_rematch]: 9.71e-06
        [meta_fg_expand]: 2.24999e-06
        [replace_old_param]: 8.63001e-06
        [inline_without_move]: 5.83002e-06
        [renormalize]: 0.00067666
        [add_forward_monad_depend]: 6.32001e-06
        [auto_monad_grad]: 2.44001e-06
        [auto_monad_eliminator]: 1.476e-05
        [cse]: 2.866e-05
        [replace_applicator]: 1.416e-05
    [Cycle 2]: 0.00034827, [27]
        [switch_simplify]: 6.17001e-06
        [loop_unroll]: 5.25001e-06
        [a_1]: 9.345e-05
        [with_stream_mark]: 1.002e-05
        [recompute_prepare]: 4.78001e-06
        [updatestate_depend_eliminate]: 2.37999e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 2.657e-05
        [updatestate_useless_node_eliminater]: 5.20001e-06
        [accelerated_algorithm]: 4.67e-06
        [meta_shard_fg_expand]: 1.26002e-06
        [get_grad_eliminate_]: 4.53999e-06
        [merge_forward]: 2.76e-06
        [cell_reuse_recompute_pass]: 2.84001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.303e-05
        [j_node_and_user_rematch]: 7.52002e-06
        [meta_fg_expand]: 1.53002e-06
        [replace_old_param]: 6.89999e-06
        [inline_without_move]: 1.021e-05
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.40001e-06
        [auto_monad_grad]: 6.60017e-07
        [auto_monad_eliminator]: 5.39998e-06
        [cse]: 1.076e-05
        [replace_applicator]: 5.05001e-06
[py_interpret_to_execute_after_opt_a]: 1.185e-05
[rewriter_after_opt_a]: 3.568e-05
[convert_after_rewriter]: 7.05002e-06
[order_py_execute_after_rewriter]: 5.37001e-06
[mutable_eliminate]: 0.00057731
[jit_opt_b]: 7.659e-05, [1]
    [Cycle 1]: 6.837e-05, [2]
        [frontend_op_eliminate]: 1.558e-05
        [inline_after_opt_a]: 1.424e-05
[cconv]: 2.659e-05
[loop_unroll]: 0.00042827
[jit_opt_after_cconv]: 0.00015552, [1]
    [Cycle 1]: 0.00014877, [11]
        [c_1]: 1.967e-05
        [parameter_eliminate]: 2.83e-06
        [updatestate_depend_eliminate]: 5.79e-06
        [updatestate_assign_eliminate]: 2.36998e-06
        [updatestate_loads_eliminate]: 2.06003e-06
        [cse]: 1.933e-05
        [call_graph_tuple_transform]: 3.237e-05
        [tuple_list_get_item_eliminator]: 5.61e-06
        [none_parameter_eliminate]: 2.30002e-06
        [renormalize]: 6.39993e-07
        [switch_simplify]: 5.02999e-06
[remove_dup_value]: 1.446e-05
[partial_unused_args_eliminate]: 2.02001e-06
[environ_conv]: 4.89e-06
[add_recomputation]: 4.721e-05
[cse_after_recomputation]: 2.072e-05, [1]
    [Cycle 1]: 1.526e-05, [1]
        [cse]: 9.47001e-06
[auto_monad_reorder]: 1.498e-05
[get_jit_bprop_graph]: 1.94999e-06
[rewriter_after_jit_bprop_graph]: 5.90002e-06
[opt_after_jit_grad]: 0.00046441
[symbol_engine_optimizer]: 7.235e-05, [1]
    [Cycle 1]: 6.558e-05, [6]
        [build]: 3.71001e-06
        [elim_shapecalc]: 7.12002e-06
        [elim_not_effective]: 1.291e-05
        [opt_reshape]: 5.44e-06
        [fold_const_symbol]: 8.50999e-06
        [renormalize]: 4.89992e-07
[validate]: 3.613e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00628108
[execute]: 6.84999e-06
Sums
    bootstrap                            :     0.000303s :  2.01%
    type_inference                       :     0.004714s : 31.20%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000061s :  0.41%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.21%
    rewriter_before_opt_a                :     0.000070s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.44%
    jit_opt_a.loop_unroll                :     0.000037s :  0.24%
    jit_opt_a.a_1                        :     0.000659s :  4.36%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000033s :  0.22%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000016s :  0.11%
    jit_opt_a.renormalize                :     0.000677s :  4.48%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000039s :  0.26%
    jit_opt_a.replace_applicator         :     0.000019s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000036s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000577s :  3.82%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000027s :  0.18%
    loop_unroll                          :     0.000428s :  2.83%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000032s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000047s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000464s :  3.07%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006281s : 41.57%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000220    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.72% :     0.000006s :      3: substitution.graph_param_transform
84.96% :     0.000187s :      4: substitution.inline
 1.39% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.21% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.55% :     0.000003s :      2: substitution.replace_old_param
 5.57% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004622     2
81.56% :     0.003769s :      1: type_inference.infer
18.44% :     0.000852s :      1: type_inference.specialize
------[replace.]   0.000066     6
67.72% :     0.000044s :      4: replace.inline
32.28% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000196     6
94.29% :     0.000184s :      4: match.inline
 5.71% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.54% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.37% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.93% :     0.000002s :     10: predicate.addn_zero_filter
 1.84% :     0.000002s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.77% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.17% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000000s :      3: predicate.graph_param_transform
 5.39% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.02% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.54% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.05% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.78% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.63% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.96% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.22% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.15% :     0.000010s :     53: predicate.switch_simplify
 1.22% :     0.000002s :     10: predicate.tile_eliminate
 1.46% :     0.000002s :     10: predicate.transpose_eliminate
 1.25% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.93% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000613    12
48.48% :     0.000297s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.72% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.79% :     0.000268s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021190    76
 0.24% :     0.000050s :      1: add_recomputation
 0.31% :     0.000065s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.48% :     0.000314s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.39% :     0.005804s :      1: jit_opt_a
 0.75% :     0.000158s :      1: jit_opt_after_cconv
 0.38% :     0.000079s :      1: jit_opt_b
 2.06% :     0.000436s :      1: loop_unroll
 2.76% :     0.000585s :      1: mutable_eliminate
 4.27% :     0.000905s :     26: opt.transform.jit_opt_a
 0.28% :     0.000060s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.23% :     0.000472s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000035s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.79% :     0.000380s :      1: renormalize.infer
 1.37% :     0.000290s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000039s :      1: rewriter_after_opt_a
 0.35% :     0.000073s :      1: rewriter_before_opt_a
 0.35% :     0.000075s :      1: symbol_engine_optimizer
29.69% :     0.006291s :      1: task_emit
22.32% :     0.004729s :      1: type_inference
 0.26% :     0.000055s :      1: validate


TotalTime = 0.0212263, [33]
[bootstrap]: 0.00041968
[type_inference]: 0.00547014
[event_method]: 1.819e-05
[auto_monad]: 6.647e-05
[graph_reusing]: 5.89e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 4.12e-05
[rewriter_before_opt_a]: 7.467e-05
[expand_dump_flag]: 3.33e-06
[jit_opt_a]: 0.00616397, [2]
    [Cycle 1]: 0.00178346, [27]
        [switch_simplify]: 6.031e-05
        [loop_unroll]: 2.929e-05
        [a_1]: 0.00059016
        [with_stream_mark]: 1.974e-05
        [recompute_prepare]: 7.28999e-06
        [updatestate_depend_eliminate]: 4.49002e-06
        [updatestate_assign_eliminate]: 3.14999e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [parameter_eliminate]: 1.87001e-06
        [specialize_transform]: 6.34001e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.28002e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 5.50001e-06
        [merge_forward]: 3.88999e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.492e-05
        [j_node_and_user_rematch]: 9.37001e-06
        [meta_fg_expand]: 2.43e-06
        [replace_old_param]: 9.12999e-06
        [inline_without_move]: 5.23002e-06
        [renormalize]: 0.00075402
        [add_forward_monad_depend]: 5.46e-06
        [auto_monad_grad]: 2.29001e-06
        [auto_monad_eliminator]: 1.473e-05
        [cse]: 3.053e-05
        [replace_applicator]: 1.208e-05
    [Cycle 2]: 0.00031617, [27]
        [switch_simplify]: 5.57001e-06
        [loop_unroll]: 5.04e-06
        [a_1]: 9.15e-05
        [with_stream_mark]: 9.62999e-06
        [recompute_prepare]: 5.22e-06
        [updatestate_depend_eliminate]: 2.84999e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 4.82998e-06
        [updatestate_useless_node_eliminater]: 4.68001e-06
        [accelerated_algorithm]: 4.84e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 4.45999e-06
        [merge_forward]: 2.54001e-06
        [cell_reuse_recompute_pass]: 2.05002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.263e-05
        [j_node_and_user_rematch]: 7.33e-06
        [meta_fg_expand]: 1.65001e-06
        [replace_old_param]: 6.94001e-06
        [inline_without_move]: 4.74002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 4.80999e-06
        [cse]: 1.03e-05
        [replace_applicator]: 5.05001e-06
[py_interpret_to_execute_after_opt_a]: 1.097e-05
[rewriter_after_opt_a]: 3.642e-05
[convert_after_rewriter]: 6.80002e-06
[order_py_execute_after_rewriter]: 4.92e-06
[mutable_eliminate]: 0.00059351
[jit_opt_b]: 4.975e-05, [1]
    [Cycle 1]: 4.251e-05, [2]
        [frontend_op_eliminate]: 1.705e-05
        [inline_after_opt_a]: 1.46e-05
[cconv]: 2.57e-05
[loop_unroll]: 0.0004249
[jit_opt_after_cconv]: 0.00014487, [1]
    [Cycle 1]: 0.00013816, [11]
        [c_1]: 2.238e-05
        [parameter_eliminate]: 2.55002e-06
        [updatestate_depend_eliminate]: 5.42001e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.18998e-06
        [cse]: 2.053e-05
        [call_graph_tuple_transform]: 1.824e-05
        [tuple_list_get_item_eliminator]: 5.22999e-06
        [none_parameter_eliminate]: 2.04e-06
        [renormalize]: 1.07e-06
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 1.451e-05
[partial_unused_args_eliminate]: 2.44001e-06
[environ_conv]: 5.27001e-06
[add_recomputation]: 4.954e-05
[cse_after_recomputation]: 2.165e-05, [1]
    [Cycle 1]: 1.611e-05, [1]
        [cse]: 1.027e-05
[auto_monad_reorder]: 1.639e-05
[get_jit_bprop_graph]: 2.46998e-06
[rewriter_after_jit_bprop_graph]: 5.32001e-06
[opt_after_jit_grad]: 0.00046646
[symbol_engine_optimizer]: 7.32e-05, [1]
    [Cycle 1]: 6.737e-05, [6]
        [build]: 3.69002e-06
        [elim_shapecalc]: 7.59002e-06
        [elim_not_effective]: 1.367e-05
        [opt_reshape]: 5.34e-06
        [fold_const_symbol]: 9.35001e-06
        [renormalize]: 5.3001e-07
[validate]: 4.034e-05
[backend_pass]: 8.2e-07
[task_emit]: 0.00673888
[execute]: 9.34e-06
Sums
    bootstrap                            :     0.000420s :  2.54%
    type_inference                       :     0.005470s : 33.07%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000066s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.25%
    rewriter_before_opt_a                :     0.000075s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.21%
    jit_opt_a.a_1                        :     0.000682s :  4.12%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000754s :  4.56%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000041s :  0.25%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000594s :  3.59%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000026s :  0.16%
    loop_unroll                          :     0.000425s :  2.57%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.30%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000466s :  2.82%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006739s : 40.74%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000220    23
 1.10% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.43% :     0.000005s :      3: substitution.graph_param_transform
84.30% :     0.000185s :      4: substitution.inline
 1.30% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.34% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.91% :     0.000004s :      2: substitution.replace_old_param
 5.96% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005396     2
82.35% :     0.004443s :      1: type_inference.infer
17.65% :     0.000952s :      1: type_inference.specialize
------[replace.]   0.000074     6
67.48% :     0.000050s :      4: replace.inline
32.52% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000195     6
93.84% :     0.000183s :      4: match.inline
 6.16% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.73% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.13% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.63% :     0.000007s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.18% :     0.000001s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.89% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.62% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.54% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.88% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.15% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.18% :     0.000001s :      6: predicate.special_op_eliminate
 1.21% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.00% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.61% :     0.000009s :     53: predicate.switch_simplify
 1.34% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.44% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000697    12
52.28% :     0.000364s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.36% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.36% :     0.000302s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022972    76
 0.23% :     0.000052s :      1: add_recomputation
 0.30% :     0.000070s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.93% :     0.000443s :      1: bootstrap
 0.12% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.85% :     0.006167s :      1: jit_opt_a
 0.64% :     0.000148s :      1: jit_opt_after_cconv
 0.23% :     0.000052s :      1: jit_opt_b
 1.88% :     0.000433s :      1: loop_unroll
 2.62% :     0.000602s :      1: mutable_eliminate
 3.89% :     0.000894s :     26: opt.transform.jit_opt_a
 0.21% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.07% :     0.000475s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000044s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.73% :     0.000396s :      1: renormalize.infer
 1.52% :     0.000349s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000039s :      1: rewriter_after_opt_a
 0.34% :     0.000078s :      1: rewriter_before_opt_a
 0.33% :     0.000076s :      1: symbol_engine_optimizer
29.41% :     0.006755s :      1: task_emit
23.89% :     0.005487s :      1: type_inference
 0.26% :     0.000059s :      1: validate


TotalTime = 0.015719, [30]
[bootstrap]: 0.00031686
[type_inference]: 0.00688285
[event_method]: 2.082e-05
[auto_monad]: 8.475e-05
[graph_reusing]: 5.51e-06
[pre_auto_parallel]: 2.42001e-06
[py_interpret_to_execute]: 5.248e-05
[rewriter_before_opt_a]: 8.625e-05
[expand_dump_flag]: 3.10002e-06
[jit_opt_a]: 0.00577788, [2]
    [Cycle 1]: 0.00201334, [27]
        [switch_simplify]: 0.00011859
        [loop_unroll]: 3.75e-05
        [a_1]: 0.00071813
        [with_stream_mark]: 1.813e-05
        [recompute_prepare]: 8.18999e-06
        [updatestate_depend_eliminate]: 3.75e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 2.64999e-06
        [parameter_eliminate]: 1.54e-06
        [specialize_transform]: 6.99001e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 5.82999e-06
        [meta_shard_fg_expand]: 2.72001e-06
        [get_grad_eliminate_]: 5.62999e-06
        [merge_forward]: 3.6e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.436e-05
        [j_node_and_user_rematch]: 1.017e-05
        [meta_fg_expand]: 2.71e-06
        [replace_old_param]: 9.94001e-06
        [inline_without_move]: 6.02999e-06
        [renormalize]: 0.0007891
        [add_forward_monad_depend]: 6.41998e-06
        [auto_monad_grad]: 2.59001e-06
        [auto_monad_eliminator]: 1.504e-05
        [cse]: 2.415e-05
        [replace_applicator]: 1.301e-05
    [Cycle 2]: 0.0003321, [27]
        [switch_simplify]: 6.17001e-06
        [loop_unroll]: 4.85999e-06
        [a_1]: 9.274e-05
        [with_stream_mark]: 1.058e-05
        [recompute_prepare]: 5.37999e-06
        [updatestate_depend_eliminate]: 2.84999e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 1.37e-06
        [specialize_transform]: 4.88001e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 4.82e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 4.79e-06
        [merge_forward]: 2.84999e-06
        [cell_reuse_recompute_pass]: 1.86e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.503e-05
        [j_node_and_user_rematch]: 8.04002e-06
        [meta_fg_expand]: 1.74998e-06
        [replace_old_param]: 6.76e-06
        [inline_without_move]: 5.07e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.71002e-06
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 5.67001e-06
        [cse]: 1.081e-05
        [replace_applicator]: 5.46e-06
[py_interpret_to_execute_after_opt_a]: 1.222e-05
[rewriter_after_opt_a]: 3.538e-05
[convert_after_rewriter]: 8.08001e-06
[order_py_execute_after_rewriter]: 5.25999e-06
[mutable_eliminate]: 0.00063921
[jit_opt_b]: 6.927e-05, [1]
    [Cycle 1]: 6.148e-05, [2]
        [frontend_op_eliminate]: 2.878e-05
        [inline_after_opt_a]: 1.748e-05
[cconv]: 2.721e-05
[loop_unroll]: 0.00054748
[jit_opt_after_cconv]: 0.00015586, [1]
    [Cycle 1]: 0.00014839, [11]
        [c_1]: 2.376e-05
        [parameter_eliminate]: 3.35998e-06
        [updatestate_depend_eliminate]: 6.06e-06
        [updatestate_assign_eliminate]: 3.06001e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 2.364e-05
        [call_graph_tuple_transform]: 2.067e-05
        [tuple_list_get_item_eliminator]: 5.56e-06
        [none_parameter_eliminate]: 1.37999e-06
        [renormalize]: 8.00006e-07
        [switch_simplify]: 5.84e-06
[remove_dup_value]: 1.385e-05
[partial_unused_args_eliminate]: 2.67001e-06
[environ_conv]: 4.60001e-06
[add_recomputation]: 5.333e-05
[cse_after_recomputation]: 2.188e-05, [1]
    [Cycle 1]: 1.613e-05, [1]
        [cse]: 1.002e-05
[auto_monad_reorder]: 1.518e-05
[get_jit_bprop_graph]: 1.62001e-06
[rewriter_after_jit_bprop_graph]: 6.67002e-06
[opt_after_jit_grad]: 0.00055798
[symbol_engine_optimizer]: 8.554e-05, [1]
    [Cycle 1]: 7.766e-05, [6]
        [build]: 4.43001e-06
        [elim_shapecalc]: 8.48999e-06
        [elim_not_effective]: 1.805e-05
        [opt_reshape]: 5.29e-06
        [fold_const_symbol]: 1.038e-05
        [renormalize]: 6.89994e-07
[validate]: 4.349e-05
Sums
    bootstrap                            :     0.000317s :  2.71%
    type_inference                       :     0.006883s : 58.93%
    event_method                         :     0.000021s :  0.18%
    auto_monad                           :     0.000085s :  0.73%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000052s :  0.45%
    rewriter_before_opt_a                :     0.000086s :  0.74%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000125s :  1.07%
    jit_opt_a.loop_unroll                :     0.000042s :  0.36%
    jit_opt_a.a_1                        :     0.000811s :  6.94%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.14%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000789s :  6.76%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.18%
    jit_opt_a.cse                        :     0.000035s :  0.30%
    jit_opt_a.replace_applicator         :     0.000018s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.10%
    rewriter_after_opt_a                 :     0.000035s :  0.30%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000639s :  5.47%
    jit_opt_b.frontend_op_eliminate      :     0.000029s :  0.25%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.15%
    cconv                                :     0.000027s :  0.23%
    loop_unroll                          :     0.000547s :  4.69%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000014s :  0.12%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000053s :  0.46%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000558s :  4.78%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.37%


Time group info:
------[substitution.]   0.000273    26
 1.04% :     0.000003s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 1.93% :     0.000005s :      3: substitution.graph_param_transform
84.30% :     0.000231s :      6: substitution.inline
 0.99% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.95% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.44% :     0.000004s :      2: substitution.replace_old_param
 3.34% :     0.000009s :      1: substitution.switch_simplify
 4.57% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006820     2
82.67% :     0.005638s :      1: type_inference.infer
17.33% :     0.001182s :      1: type_inference.specialize
------[replace.]   0.000141     9
51.81% :     0.000073s :      6: replace.inline
24.59% :     0.000035s :      1: replace.switch_simplify
23.60% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000246     9
92.11% :     0.000227s :      6: match.inline
 3.32% :     0.000008s :      1: match.switch_simplify
 4.57% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000146   801
 1.57% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.77% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000002s :     12: predicate.addn_check_dump
 1.35% :     0.000002s :     12: predicate.addn_zero_filter
 2.49% :     0.000004s :     12: predicate.arithmetic_simplify
 1.41% :     0.000002s :     12: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000002s :     12: predicate.compare_switch_simplify
 1.59% :     0.000002s :     12: predicate.depend_value_elim
 1.19% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.39% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.08% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.03% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.11% :     0.000002s :     12: predicate.environ_get_eliminate
 1.09% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.36% :     0.000001s :      3: predicate.graph_param_transform
 5.55% :     0.000008s :     26: predicate.inline
 0.78% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.95% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     17: predicate.load_eliminater
 1.83% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.19% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.28% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000002s :     12: predicate.merge_addn
 1.05% :     0.000002s :     12: predicate.minmaximum_grad
 1.91% :     0.000003s :      3: predicate.mutable_eliminate
 0.35% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     17: predicate.partial_eliminate
 1.15% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     12: predicate.reduce_eliminate
 1.45% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     20: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     12: predicate.reshape_eliminate
 1.24% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     12: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.76% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000004s :     20: predicate.switch_defer_inline
 2.32% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.71% :     0.000013s :     65: predicate.switch_simplify
 1.21% :     0.000002s :     12: predicate.tile_eliminate
 1.17% :     0.000002s :     12: predicate.transpose_eliminate
 1.50% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.62% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.59% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     12: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.49% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000879    15
50.53% :     0.000444s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 9.81% :     0.000086s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.66% :     0.000349s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017681    72
 0.32% :     0.000056s :      1: add_recomputation
 0.50% :     0.000088s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 1.88% :     0.000332s :      1: bootstrap
 0.17% :     0.000030s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.14% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
32.70% :     0.005782s :      1: jit_opt_a
 0.90% :     0.000158s :      1: jit_opt_after_cconv
 0.41% :     0.000072s :      1: jit_opt_b
 3.14% :     0.000556s :      1: loop_unroll
 3.67% :     0.000649s :      1: mutable_eliminate
 6.21% :     0.001098s :     26: opt.transform.jit_opt_a
 0.29% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000038s :      4: opt.transform.jit_opt_b
 0.09% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000038s :      4: opt.transform.symbol_engine_opt
 3.21% :     0.000567s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.31% :     0.000055s :      1: py_interpret_to_execute
 0.08% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 2.39% :     0.000422s :      1: renormalize.infer
 2.03% :     0.000358s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000039s :      1: rewriter_after_opt_a
 0.51% :     0.000090s :      1: rewriter_before_opt_a
 0.50% :     0.000089s :      1: symbol_engine_optimizer
38.99% :     0.006894s :      1: type_inference


TotalTime = 0.0196986, [33]
[bootstrap]: 0.00029188
[type_inference]: 0.00454406
[event_method]: 1.762e-05
[auto_monad]: 6.331e-05
[graph_reusing]: 6.18002e-06
[pre_auto_parallel]: 2.68e-06
[py_interpret_to_execute]: 3.412e-05
[rewriter_before_opt_a]: 7.31e-05
[expand_dump_flag]: 2.90002e-06
[jit_opt_a]: 0.00589548, [2]
    [Cycle 1]: 0.0021247, [27]
        [switch_simplify]: 5.732e-05
        [loop_unroll]: 3.104e-05
        [a_1]: 0.00055571
        [with_stream_mark]: 1.972e-05
        [recompute_prepare]: 7.91001e-06
        [updatestate_depend_eliminate]: 4.13001e-06
        [updatestate_assign_eliminate]: 3.02002e-06
        [updatestate_loads_eliminate]: 3.10998e-06
        [parameter_eliminate]: 1.93997e-06
        [specialize_transform]: 6.29001e-06
        [updatestate_useless_node_eliminater]: 5.68997e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 2.48e-06
        [get_grad_eliminate_]: 5.62999e-06
        [merge_forward]: 3.70998e-06
        [cell_reuse_recompute_pass]: 1.35001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.6e-05
        [j_node_and_user_rematch]: 1e-05
        [meta_fg_expand]: 2.65002e-06
        [replace_old_param]: 1.022e-05
        [inline_without_move]: 5.51998e-06
        [renormalize]: 0.00111051
        [add_forward_monad_depend]: 6.60002e-06
        [auto_monad_grad]: 2.40002e-06
        [auto_monad_eliminator]: 1.605e-05
        [cse]: 2.791e-05
        [replace_applicator]: 1.487e-05
    [Cycle 2]: 0.00035452, [27]
        [switch_simplify]: 6.33e-06
        [loop_unroll]: 5.22e-06
        [a_1]: 0.00010057
        [with_stream_mark]: 1.025e-05
        [recompute_prepare]: 5.63002e-06
        [updatestate_depend_eliminate]: 2.73e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [parameter_eliminate]: 1.22999e-06
        [specialize_transform]: 5.77001e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 6.03002e-06
        [meta_shard_fg_expand]: 1.30001e-06
        [get_grad_eliminate_]: 5.33002e-06
        [merge_forward]: 3.08e-06
        [cell_reuse_recompute_pass]: 2.42001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.439e-05
        [j_node_and_user_rematch]: 9.07999e-06
        [meta_fg_expand]: 1.50999e-06
        [replace_old_param]: 8.23001e-06
        [inline_without_move]: 4.94e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 1.00999e-06
        [auto_monad_eliminator]: 5.37001e-06
        [cse]: 1.051e-05
        [replace_applicator]: 5.94e-06
[py_interpret_to_execute_after_opt_a]: 1.22e-05
[rewriter_after_opt_a]: 3.728e-05
[convert_after_rewriter]: 7.18998e-06
[order_py_execute_after_rewriter]: 4.78001e-06
[mutable_eliminate]: 0.00055699
[jit_opt_b]: 5.009e-05, [1]
    [Cycle 1]: 4.341e-05, [2]
        [frontend_op_eliminate]: 1.674e-05
        [inline_after_opt_a]: 1.469e-05
[cconv]: 2.647e-05
[loop_unroll]: 0.00041821
[jit_opt_after_cconv]: 0.00013855, [1]
    [Cycle 1]: 0.00013217, [11]
        [c_1]: 2.034e-05
        [parameter_eliminate]: 3.05998e-06
        [updatestate_depend_eliminate]: 6.46e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.865e-05
        [call_graph_tuple_transform]: 1.842e-05
        [tuple_list_get_item_eliminator]: 4.94e-06
        [none_parameter_eliminate]: 1.59998e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.82999e-06
[remove_dup_value]: 1.42e-05
[partial_unused_args_eliminate]: 2.44001e-06
[environ_conv]: 4.77e-06
[add_recomputation]: 4.936e-05
[cse_after_recomputation]: 1.906e-05, [1]
    [Cycle 1]: 1.345e-05, [1]
        [cse]: 8.07e-06
[auto_monad_reorder]: 1.441e-05
[get_jit_bprop_graph]: 2.09999e-06
[rewriter_after_jit_bprop_graph]: 5.22e-06
[opt_after_jit_grad]: 0.00045074
[symbol_engine_optimizer]: 7.415e-05, [1]
    [Cycle 1]: 6.723e-05, [6]
        [build]: 4.62e-06
        [elim_shapecalc]: 7.08e-06
        [elim_not_effective]: 1.378e-05
        [opt_reshape]: 5.97999e-06
        [fold_const_symbol]: 8.38999e-06
        [renormalize]: 5.19998e-07
[validate]: 3.782e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00662576
[execute]: 7.56999e-06
Sums
    bootstrap                            :     0.000292s :  1.87%
    type_inference                       :     0.004544s : 29.05%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000063s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.22%
    rewriter_before_opt_a                :     0.000073s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.41%
    jit_opt_a.loop_unroll                :     0.000036s :  0.23%
    jit_opt_a.a_1                        :     0.000656s :  4.20%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.001111s :  7.10%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.14%
    jit_opt_a.cse                        :     0.000038s :  0.25%
    jit_opt_a.replace_applicator         :     0.000021s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000037s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000557s :  3.56%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000026s :  0.17%
    loop_unroll                          :     0.000418s :  2.67%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.32%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000451s :  2.88%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006626s : 42.35%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000217    23
 1.28% :     0.000003s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.46% :     0.000005s :      3: substitution.graph_param_transform
83.68% :     0.000182s :      4: substitution.inline
 1.88% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.31% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.18% :     0.000005s :      2: substitution.replace_old_param
 5.66% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004476     2
82.14% :     0.003676s :      1: type_inference.infer
17.86% :     0.000800s :      1: type_inference.specialize
------[replace.]   0.000068     6
67.90% :     0.000046s :      4: replace.inline
32.10% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000191     6
94.15% :     0.000180s :      4: match.inline
 5.85% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.30% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.24% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 2.50% :     0.000003s :     10: predicate.arithmetic_simplify
 1.29% :     0.000002s :     10: predicate.cast_eliminate
 0.64% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.50% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.43% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.44% :     0.000002s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.12% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.78% :     0.000007s :     22: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :      6: predicate.less_batch_normalization
 2.04% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.62% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.75% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 1.23% :     0.000002s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.62% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.84% :     0.000010s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000002s :     10: predicate.transpose_eliminate
 1.21% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.11% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000962    12
31.27% :     0.000301s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.15% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
64.58% :     0.000621s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021784    76
 0.24% :     0.000052s :      1: add_recomputation
 0.31% :     0.000067s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.42% :     0.000308s :      1: bootstrap
 0.13% :     0.000029s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
27.08% :     0.005899s :      1: jit_opt_a
 0.65% :     0.000142s :      1: jit_opt_after_cconv
 0.24% :     0.000053s :      1: jit_opt_b
 1.96% :     0.000426s :      1: loop_unroll
 2.60% :     0.000566s :      1: mutable_eliminate
 4.05% :     0.000883s :     26: opt.transform.jit_opt_a
 0.21% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.10% :     0.000458s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000037s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.75% :     0.000381s :      1: renormalize.infer
 3.30% :     0.000719s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000040s :      1: rewriter_after_opt_a
 0.35% :     0.000076s :      1: rewriter_before_opt_a
 0.35% :     0.000077s :      1: symbol_engine_optimizer
30.50% :     0.006643s :      1: task_emit
20.92% :     0.004558s :      1: type_inference
 0.26% :     0.000057s :      1: validate


TotalTime = 0.0201331, [33]
[bootstrap]: 0.0002774
[type_inference]: 0.00480646
[event_method]: 1.831e-05
[auto_monad]: 6.511e-05
[graph_reusing]: 4.95999e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 3.395e-05
[rewriter_before_opt_a]: 6.953e-05
[expand_dump_flag]: 3.21999e-06
[jit_opt_a]: 0.00594589, [2]
    [Cycle 1]: 0.00180113, [27]
        [switch_simplify]: 5.7e-05
        [loop_unroll]: 2.912e-05
        [a_1]: 0.00056878
        [with_stream_mark]: 2.176e-05
        [recompute_prepare]: 8.07998e-06
        [updatestate_depend_eliminate]: 3.53e-06
        [updatestate_assign_eliminate]: 2.99001e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 5.84999e-06
        [updatestate_useless_node_eliminater]: 5.44998e-06
        [accelerated_algorithm]: 5.34998e-06
        [meta_shard_fg_expand]: 2.17999e-06
        [get_grad_eliminate_]: 5.13002e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 2.48002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.635e-05
        [j_node_and_user_rematch]: 9.20001e-06
        [meta_fg_expand]: 2.07999e-06
        [replace_old_param]: 9.14e-06
        [inline_without_move]: 5.19e-06
        [renormalize]: 0.00078594
        [add_forward_monad_depend]: 7.38e-06
        [auto_monad_grad]: 2.20002e-06
        [auto_monad_eliminator]: 1.408e-05
        [cse]: 2.761e-05
        [replace_applicator]: 1.389e-05
    [Cycle 2]: 0.0003682, [27]
        [switch_simplify]: 6.53e-06
        [loop_unroll]: 4.67e-06
        [a_1]: 0.00010979
        [with_stream_mark]: 1.136e-05
        [recompute_prepare]: 6.21e-06
        [updatestate_depend_eliminate]: 3.57997e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 5.01002e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.00001e-06
        [meta_shard_fg_expand]: 1.30001e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 3.20998e-06
        [cell_reuse_recompute_pass]: 3.68e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.504e-05
        [j_node_and_user_rematch]: 8.14002e-06
        [meta_fg_expand]: 1.79998e-06
        [replace_old_param]: 6.91001e-06
        [inline_without_move]: 5.20999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.22999e-06
        [auto_monad_grad]: 1.00999e-06
        [auto_monad_eliminator]: 7.18e-06
        [cse]: 1.287e-05
        [replace_applicator]: 6.14999e-06
[py_interpret_to_execute_after_opt_a]: 1.3e-05
[rewriter_after_opt_a]: 3.68e-05
[convert_after_rewriter]: 7.68001e-06
[order_py_execute_after_rewriter]: 5.49e-06
[mutable_eliminate]: 0.0007979
[jit_opt_b]: 5.223e-05, [1]
    [Cycle 1]: 4.415e-05, [2]
        [frontend_op_eliminate]: 1.655e-05
        [inline_after_opt_a]: 1.544e-05
[cconv]: 2.773e-05
[loop_unroll]: 0.00050627
[jit_opt_after_cconv]: 0.0001824, [1]
    [Cycle 1]: 0.00017452, [11]
        [c_1]: 2.176e-05
        [parameter_eliminate]: 4.05e-06
        [updatestate_depend_eliminate]: 5.91998e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [cse]: 2.108e-05
        [call_graph_tuple_transform]: 2.175e-05
        [tuple_list_get_item_eliminator]: 5.82001e-06
        [none_parameter_eliminate]: 1.70001e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 5.27001e-06
[remove_dup_value]: 1.398e-05
[partial_unused_args_eliminate]: 2.07001e-06
[environ_conv]: 5.61e-06
[add_recomputation]: 4.704e-05
[cse_after_recomputation]: 2.183e-05, [1]
    [Cycle 1]: 1.517e-05, [1]
        [cse]: 9.09e-06
[auto_monad_reorder]: 1.468e-05
[get_jit_bprop_graph]: 2.11e-06
[rewriter_after_jit_bprop_graph]: 7.5e-06
[opt_after_jit_grad]: 0.0005573
[symbol_engine_optimizer]: 7.612e-05, [1]
    [Cycle 1]: 6.865e-05, [6]
        [build]: 4.28001e-06
        [elim_shapecalc]: 8.07998e-06
        [elim_not_effective]: 1.352e-05
        [opt_reshape]: 5.24e-06
        [fold_const_symbol]: 8.48001e-06
        [renormalize]: 9.39996e-07
[validate]: 4.129e-05
[backend_pass]: 7.30011e-07
[task_emit]: 0.00627999
[execute]: 8.18999e-06
Sums
    bootstrap                            :     0.000277s :  1.77%
    type_inference                       :     0.004806s : 30.63%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000065s :  0.41%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.22%
    rewriter_before_opt_a                :     0.000070s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000679s :  4.32%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000786s :  5.01%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.14%
    jit_opt_a.cse                        :     0.000040s :  0.26%
    jit_opt_a.replace_applicator         :     0.000020s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000037s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000798s :  5.08%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000028s :  0.18%
    loop_unroll                          :     0.000506s :  3.23%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000047s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.05%
    opt_after_jit_grad                   :     0.000557s :  3.55%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.26%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006280s : 40.02%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000235    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.44% :     0.000006s :      3: substitution.graph_param_transform
85.25% :     0.000200s :      4: substitution.inline
 1.65% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.18% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000005s :      2: substitution.replace_old_param
 5.11% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004740     2
80.30% :     0.003806s :      1: type_inference.infer
19.70% :     0.000934s :      1: type_inference.specialize
------[replace.]   0.000070     6
68.25% :     0.000048s :      4: replace.inline
31.75% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000209     6
94.78% :     0.000198s :      4: match.inline
 5.22% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.55% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.43% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.44% :     0.000002s :     10: predicate.addn_zero_filter
 2.20% :     0.000003s :     10: predicate.arithmetic_simplify
 1.54% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.45% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.73% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.70% :     0.000001s :      3: predicate.graph_param_transform
 5.09% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.82% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.90% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.54% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 2.03% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.41% :     0.000002s :     10: predicate.reduce_eliminate
 1.71% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.84% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.84% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.44% :     0.000001s :      3: predicate.reset_defer_inline
 1.32% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.75% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.13% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.59% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.55% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.30% :     0.000002s :     10: predicate.transpose_eliminate
 1.66% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.50% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000572    12
44.22% :     0.000253s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.38% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.40% :     0.000271s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021934    76
 0.23% :     0.000050s :      1: add_recomputation
 0.31% :     0.000069s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.32% :     0.000289s :      1: bootstrap
 0.14% :     0.000031s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
27.12% :     0.005949s :      1: jit_opt_a
 0.85% :     0.000186s :      1: jit_opt_after_cconv
 0.25% :     0.000055s :      1: jit_opt_b
 2.35% :     0.000516s :      1: loop_unroll
 3.69% :     0.000809s :      1: mutable_eliminate
 4.09% :     0.000897s :     26: opt.transform.jit_opt_a
 0.23% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.59% :     0.000567s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000037s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.92% :     0.000422s :      1: renormalize.infer
 1.62% :     0.000355s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000040s :      1: rewriter_after_opt_a
 0.33% :     0.000072s :      1: rewriter_before_opt_a
 0.36% :     0.000079s :      1: symbol_engine_optimizer
28.70% :     0.006295s :      1: task_emit
21.97% :     0.004818s :      1: type_inference
 0.28% :     0.000062s :      1: validate


TotalTime = 0.0148233, [30]
[bootstrap]: 0.00028754
[type_inference]: 0.00660038
[event_method]: 1.966e-05
[auto_monad]: 8.392e-05
[graph_reusing]: 6.01998e-06
[pre_auto_parallel]: 2.44001e-06
[py_interpret_to_execute]: 5.541e-05
[rewriter_before_opt_a]: 8.667e-05
[expand_dump_flag]: 2.86999e-06
[jit_opt_a]: 0.00549326, [2]
    [Cycle 1]: 0.0018899, [27]
        [switch_simplify]: 0.00011191
        [loop_unroll]: 3.418e-05
        [a_1]: 0.00069762
        [with_stream_mark]: 1.549e-05
        [recompute_prepare]: 6.74999e-06
        [updatestate_depend_eliminate]: 2.86e-06
        [updatestate_assign_eliminate]: 2.53998e-06
        [updatestate_loads_eliminate]: 2.77002e-06
        [parameter_eliminate]: 1.66e-06
        [specialize_transform]: 6.09001e-06
        [updatestate_useless_node_eliminater]: 5.19998e-06
        [accelerated_algorithm]: 5.84e-06
        [meta_shard_fg_expand]: 2.37999e-06
        [get_grad_eliminate_]: 5.47001e-06
        [merge_forward]: 3.31999e-06
        [cell_reuse_recompute_pass]: 1.09003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.444e-05
        [j_node_and_user_rematch]: 9.18997e-06
        [meta_fg_expand]: 2.48998e-06
        [replace_old_param]: 9.54e-06
        [inline_without_move]: 5.50001e-06
        [renormalize]: 0.00072821
        [add_forward_monad_depend]: 4.77e-06
        [auto_monad_grad]: 1.35999e-06
        [auto_monad_eliminator]: 9.92999e-06
        [cse]: 1.737e-05
        [replace_applicator]: 1.357e-05
    [Cycle 2]: 0.00032979, [27]
        [switch_simplify]: 7.13e-06
        [loop_unroll]: 5.49e-06
        [a_1]: 0.00010252
        [with_stream_mark]: 8.95999e-06
        [recompute_prepare]: 4.63999e-06
        [updatestate_depend_eliminate]: 2.47001e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.91998e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 4.92e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 4.83001e-06
        [meta_shard_fg_expand]: 1.22999e-06
        [get_grad_eliminate_]: 4.76002e-06
        [merge_forward]: 2.27999e-06
        [cell_reuse_recompute_pass]: 1.42e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.254e-05
        [j_node_and_user_rematch]: 7.87998e-06
        [meta_fg_expand]: 2.00002e-06
        [replace_old_param]: 7.20003e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 4.80001e-06
        [cse]: 9.60001e-06
        [replace_applicator]: 5.07e-06
[py_interpret_to_execute_after_opt_a]: 8.55001e-06
[rewriter_after_opt_a]: 2.487e-05
[convert_after_rewriter]: 5.70001e-06
[order_py_execute_after_rewriter]: 3.94002e-06
[mutable_eliminate]: 0.00056364
[jit_opt_b]: 5.004e-05, [1]
    [Cycle 1]: 4.305e-05, [2]
        [frontend_op_eliminate]: 1.721e-05
        [inline_after_opt_a]: 1.497e-05
[cconv]: 2.332e-05
[loop_unroll]: 0.00045438
[jit_opt_after_cconv]: 0.00014072, [1]
    [Cycle 1]: 0.0001341, [11]
        [c_1]: 2.166e-05
        [parameter_eliminate]: 2.74999e-06
        [updatestate_depend_eliminate]: 5.71e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.722e-05
        [call_graph_tuple_transform]: 1.752e-05
        [tuple_list_get_item_eliminator]: 5.32999e-06
        [none_parameter_eliminate]: 1.62001e-06
        [renormalize]: 6.89994e-07
        [switch_simplify]: 6.38003e-06
[remove_dup_value]: 1.438e-05
[partial_unused_args_eliminate]: 2.24001e-06
[environ_conv]: 5.00999e-06
[add_recomputation]: 4.41e-05
[cse_after_recomputation]: 2.16e-05, [1]
    [Cycle 1]: 1.6e-05, [1]
        [cse]: 1.029e-05
[auto_monad_reorder]: 1.117e-05
[get_jit_bprop_graph]: 2.36998e-06
[rewriter_after_jit_bprop_graph]: 5.60001e-06
[opt_after_jit_grad]: 0.00052925
[symbol_engine_optimizer]: 7.494e-05, [1]
    [Cycle 1]: 6.791e-05, [6]
        [build]: 3.16999e-06
        [elim_shapecalc]: 8.06001e-06
        [elim_not_effective]: 1.275e-05
        [opt_reshape]: 5.60001e-06
        [fold_const_symbol]: 8.60999e-06
        [renormalize]: 1.32999e-06
[validate]: 2.804e-05
Sums
    bootstrap                            :     0.000288s :  2.62%
    type_inference                       :     0.006600s : 60.14%
    event_method                         :     0.000020s :  0.18%
    auto_monad                           :     0.000084s :  0.76%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000055s :  0.50%
    rewriter_before_opt_a                :     0.000087s :  0.79%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000119s :  1.08%
    jit_opt_a.loop_unroll                :     0.000040s :  0.36%
    jit_opt_a.a_1                        :     0.000800s :  7.29%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000728s :  6.64%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.13%
    jit_opt_a.cse                        :     0.000027s :  0.25%
    jit_opt_a.replace_applicator         :     0.000019s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.08%
    rewriter_after_opt_a                 :     0.000025s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000564s :  5.14%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.14%
    cconv                                :     0.000023s :  0.21%
    loop_unroll                          :     0.000454s :  4.14%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000014s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000044s :  0.40%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000011s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000529s :  4.82%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000028s :  0.26%


Time group info:
------[substitution.]   0.000275    26
 0.54% :     0.000001s :      2: substitution.elim_not_effective
 0.32% :     0.000001s :      2: substitution.fold_const_symbol
 1.69% :     0.000005s :      3: substitution.graph_param_transform
85.85% :     0.000236s :      6: substitution.inline
 1.10% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.54% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.31% :     0.000004s :      2: substitution.replace_old_param
 3.05% :     0.000008s :      1: substitution.switch_simplify
 4.61% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006541     2
82.95% :     0.005425s :      1: type_inference.infer
17.05% :     0.001115s :      1: type_inference.specialize
------[replace.]   0.000136     9
50.92% :     0.000069s :      6: replace.inline
24.98% :     0.000034s :      1: replace.switch_simplify
24.10% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000251     9
92.46% :     0.000232s :      6: match.inline
 3.03% :     0.000008s :      1: match.switch_simplify
 4.52% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   801
 1.71% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.99% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     12: predicate.addn_check_dump
 1.44% :     0.000002s :     12: predicate.addn_zero_filter
 2.12% :     0.000003s :     12: predicate.arithmetic_simplify
 1.44% :     0.000002s :     12: predicate.cast_eliminate
 0.37% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     12: predicate.compare_switch_simplify
 1.47% :     0.000002s :     12: predicate.depend_value_elim
 1.19% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.16% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.11% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.55% :     0.000002s :     12: predicate.environ_get_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.19% :     0.000007s :     26: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000002s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.92% :     0.000003s :     17: predicate.load_eliminater
 1.30% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000002s :     12: predicate.merge_addn
 1.22% :     0.000002s :     12: predicate.minmaximum_grad
 1.35% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     17: predicate.partial_eliminate
 1.25% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     12: predicate.reduce_eliminate
 1.47% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     20: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.66% :     0.000002s :     12: predicate.reshape_eliminate
 1.31% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     12: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 0.86% :     0.000001s :      6: predicate.specialize_transform
 1.62% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.84% :     0.000004s :     20: predicate.switch_defer_inline
 2.41% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.02% :     0.000011s :     65: predicate.switch_simplify
 1.30% :     0.000002s :     12: predicate.tile_eliminate
 1.35% :     0.000002s :     12: predicate.transpose_eliminate
 1.57% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.80% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000002s :     12: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.53% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000844    15
47.63% :     0.000402s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.26% :     0.000095s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.11% :     0.000347s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016681    72
 0.28% :     0.000047s :      1: add_recomputation
 0.52% :     0.000088s :      1: auto_monad
 0.08% :     0.000014s :      1: auto_monad_reorder
 1.81% :     0.000301s :      1: bootstrap
 0.16% :     0.000026s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.14% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
32.95% :     0.005497s :      1: jit_opt_a
 0.86% :     0.000143s :      1: jit_opt_after_cconv
 0.32% :     0.000053s :      1: jit_opt_b
 2.77% :     0.000462s :      1: loop_unroll
 3.43% :     0.000572s :      1: mutable_eliminate
 6.43% :     0.001073s :     26: opt.transform.jit_opt_a
 0.28% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.23% :     0.000538s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.35% :     0.000058s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.34% :     0.000390s :      1: renormalize.infer
 1.98% :     0.000331s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000028s :      1: rewriter_after_opt_a
 0.54% :     0.000090s :      1: rewriter_before_opt_a
 0.47% :     0.000078s :      1: symbol_engine_optimizer
39.63% :     0.006611s :      1: type_inference


TotalTime = 0.0201718, [33]
[bootstrap]: 0.00029383
[type_inference]: 0.00449454
[event_method]: 1.685e-05
[auto_monad]: 6.186e-05
[graph_reusing]: 5.32001e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 3.381e-05
[rewriter_before_opt_a]: 7.079e-05
[expand_dump_flag]: 3.48999e-06
[jit_opt_a]: 0.0063174, [2]
    [Cycle 1]: 0.00268273, [27]
        [switch_simplify]: 5.218e-05
        [loop_unroll]: 2.888e-05
        [a_1]: 0.00050678
        [with_stream_mark]: 1.792e-05
        [recompute_prepare]: 7.73999e-06
        [updatestate_depend_eliminate]: 3.66001e-06
        [updatestate_assign_eliminate]: 2.86e-06
        [updatestate_loads_eliminate]: 2.58003e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 5.71e-06
        [updatestate_useless_node_eliminater]: 4.92999e-06
        [accelerated_algorithm]: 6.01e-06
        [meta_shard_fg_expand]: 1.91998e-06
        [get_grad_eliminate_]: 5.50001e-06
        [merge_forward]: 3.59002e-06
        [cell_reuse_recompute_pass]: 1.71e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.44e-05
        [j_node_and_user_rematch]: 8.84e-06
        [meta_fg_expand]: 2.12999e-06
        [replace_old_param]: 8.16002e-06
        [inline_without_move]: 5.11997e-06
        [renormalize]: 0.00175371
        [add_forward_monad_depend]: 5.48002e-06
        [auto_monad_grad]: 2.21998e-06
        [auto_monad_eliminator]: 1.366e-05
        [cse]: 2.822e-05
        [replace_applicator]: 1.323e-05
    [Cycle 2]: 0.00032325, [27]
        [switch_simplify]: 6.21e-06
        [loop_unroll]: 5.07e-06
        [a_1]: 9.117e-05
        [with_stream_mark]: 1.023e-05
        [recompute_prepare]: 5.08002e-06
        [updatestate_depend_eliminate]: 2.83e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.17999e-06
        [specialize_transform]: 4.85999e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 5.26998e-06
        [meta_shard_fg_expand]: 1.20001e-06
        [get_grad_eliminate_]: 4.62998e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 2.96999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.179e-05
        [j_node_and_user_rematch]: 7.9e-06
        [meta_fg_expand]: 1.60001e-06
        [replace_old_param]: 7.13e-06
        [inline_without_move]: 4.52998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.02e-06
        [auto_monad_grad]: 8.80013e-07
        [auto_monad_eliminator]: 5.08002e-06
        [cse]: 1.06e-05
        [replace_applicator]: 4.74998e-06
[py_interpret_to_execute_after_opt_a]: 9.61e-06
[rewriter_after_opt_a]: 3.259e-05
[convert_after_rewriter]: 6.60997e-06
[order_py_execute_after_rewriter]: 4.95001e-06
[mutable_eliminate]: 0.00057944
[jit_opt_b]: 4.717e-05, [1]
    [Cycle 1]: 4.033e-05, [2]
        [frontend_op_eliminate]: 1.562e-05
        [inline_after_opt_a]: 1.374e-05
[cconv]: 2.436e-05
[loop_unroll]: 0.00042046
[jit_opt_after_cconv]: 0.00013582, [1]
    [Cycle 1]: 0.00012958, [11]
        [c_1]: 2.13e-05
        [parameter_eliminate]: 1.85001e-06
        [updatestate_depend_eliminate]: 4.42e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [cse]: 1.804e-05
        [call_graph_tuple_transform]: 1.792e-05
        [tuple_list_get_item_eliminator]: 5.29998e-06
        [none_parameter_eliminate]: 1.50999e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.22e-06
[remove_dup_value]: 1.285e-05
[partial_unused_args_eliminate]: 2.37001e-06
[environ_conv]: 4.83001e-06
[add_recomputation]: 5.013e-05
[cse_after_recomputation]: 1.959e-05, [1]
    [Cycle 1]: 1.446e-05, [1]
        [cse]: 8.84e-06
[auto_monad_reorder]: 1.443e-05
[get_jit_bprop_graph]: 1.59998e-06
[rewriter_after_jit_bprop_graph]: 4.79e-06
[opt_after_jit_grad]: 0.00045476
[symbol_engine_optimizer]: 7.287e-05, [1]
    [Cycle 1]: 6.617e-05, [6]
        [build]: 4.29997e-06
        [elim_shapecalc]: 7.31001e-06
        [elim_not_effective]: 1.209e-05
        [opt_reshape]: 5.86998e-06
        [fold_const_symbol]: 8.82e-06
        [renormalize]: 4.50003e-07
[validate]: 3.696e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00672209
[execute]: 9.41998e-06
Sums
    bootstrap                            :     0.000294s :  1.81%
    type_inference                       :     0.004495s : 27.66%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000062s :  0.38%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.21%
    rewriter_before_opt_a                :     0.000071s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.36%
    jit_opt_a.loop_unroll                :     0.000034s :  0.21%
    jit_opt_a.a_1                        :     0.000598s :  3.68%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001754s : 10.79%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000039s :  0.24%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000579s :  3.57%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.08%
    cconv                                :     0.000024s :  0.15%
    loop_unroll                          :     0.000420s :  2.59%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000455s :  2.80%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006722s : 41.37%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000187    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.78% :     0.000001s :      2: substitution.fold_const_symbol
 2.81% :     0.000005s :      3: substitution.graph_param_transform
83.14% :     0.000155s :      4: substitution.inline
 1.88% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.15% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.99% :     0.000004s :      2: substitution.replace_old_param
 6.25% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004427     2
81.60% :     0.003613s :      1: type_inference.infer
18.40% :     0.000815s :      1: type_inference.specialize
------[replace.]   0.000065     6
67.25% :     0.000044s :      4: replace.inline
32.75% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000164     6
93.54% :     0.000153s :      4: match.inline
 6.46% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.23% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 2.33% :     0.000003s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.44% :     0.000002s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.92% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.45% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.14% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.30% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.67% :     0.000002s :      6: predicate.less_batch_normalization
 1.65% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.15% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.16% :     0.000001s :     10: predicate.minmaximum_grad
 1.41% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000002s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.63% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.86% :     0.000009s :     53: predicate.switch_simplify
 1.32% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.98% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000569    12
50.02% :     0.000285s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.78% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.20% :     0.000257s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022819    76
 0.23% :     0.000053s :      1: add_recomputation
 0.29% :     0.000065s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.36% :     0.000311s :      1: bootstrap
 0.12% :     0.000027s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
27.70% :     0.006321s :      1: jit_opt_a
 0.61% :     0.000138s :      1: jit_opt_after_cconv
 0.22% :     0.000050s :      1: jit_opt_b
 1.88% :     0.000428s :      1: loop_unroll
 2.58% :     0.000588s :      1: mutable_eliminate
 3.52% :     0.000803s :     26: opt.transform.jit_opt_a
 0.20% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.02% :     0.000462s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000037s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 6.26% :     0.001430s :      1: renormalize.infer
 1.38% :     0.000316s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000035s :      1: rewriter_after_opt_a
 0.32% :     0.000074s :      1: rewriter_before_opt_a
 0.33% :     0.000075s :      1: symbol_engine_optimizer
29.53% :     0.006739s :      1: task_emit
19.75% :     0.004508s :      1: type_inference
 0.25% :     0.000056s :      1: validate


TotalTime = 0.0173907, [33]
[bootstrap]: 0.00027595
[type_inference]: 0.00409257
[event_method]: 1.564e-05
[auto_monad]: 4.13e-05
[graph_reusing]: 3.66001e-06
[pre_auto_parallel]: 1.79998e-06
[py_interpret_to_execute]: 2.803e-05
[rewriter_before_opt_a]: 6.006e-05
[expand_dump_flag]: 1.96e-06
[jit_opt_a]: 0.00469141, [2]
    [Cycle 1]: 0.00137469, [27]
        [switch_simplify]: 4.304e-05
        [loop_unroll]: 2.844e-05
        [a_1]: 0.00045597
        [with_stream_mark]: 1.177e-05
        [recompute_prepare]: 6.68e-06
        [updatestate_depend_eliminate]: 2.67001e-06
        [updatestate_assign_eliminate]: 2.15002e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 1.06002e-06
        [specialize_transform]: 5.82001e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.10001e-06
        [meta_shard_fg_expand]: 1.19e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 2.29999e-06
        [cell_reuse_recompute_pass]: 7.90023e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.095e-05
        [j_node_and_user_rematch]: 7.73999e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 7.46001e-06
        [inline_without_move]: 4.89998e-06
        [renormalize]: 0.00055
        [add_forward_monad_depend]: 4.92e-06
        [auto_monad_grad]: 2.21e-06
        [auto_monad_eliminator]: 1.21e-05
        [cse]: 1.953e-05
        [replace_applicator]: 1.232e-05
    [Cycle 2]: 0.00033942, [27]
        [switch_simplify]: 5.59e-06
        [loop_unroll]: 5.32999e-06
        [a_1]: 0.00010595
        [with_stream_mark]: 1.016e-05
        [recompute_prepare]: 6.26e-06
        [updatestate_depend_eliminate]: 2.48e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 5.35999e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 4.91002e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.54998e-06
        [merge_forward]: 2.24001e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.175e-05
        [j_node_and_user_rematch]: 7.57002e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 6.63998e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.29998e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 5.29e-06
        [cse]: 1.023e-05
        [replace_applicator]: 5.14e-06
[py_interpret_to_execute_after_opt_a]: 9.19998e-06
[rewriter_after_opt_a]: 3.078e-05
[convert_after_rewriter]: 6.38e-06
[order_py_execute_after_rewriter]: 5.74e-06
[mutable_eliminate]: 0.00063614
[jit_opt_b]: 4.867e-05, [1]
    [Cycle 1]: 4.181e-05, [2]
        [frontend_op_eliminate]: 1.573e-05
        [inline_after_opt_a]: 1.475e-05
[cconv]: 1.784e-05
[loop_unroll]: 0.00045274
[jit_opt_after_cconv]: 0.0001357, [1]
    [Cycle 1]: 0.00012955, [11]
        [c_1]: 2.106e-05
        [parameter_eliminate]: 2.83998e-06
        [updatestate_depend_eliminate]: 5.47001e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.751e-05
        [call_graph_tuple_transform]: 1.678e-05
        [tuple_list_get_item_eliminator]: 5.24e-06
        [none_parameter_eliminate]: 1.01002e-06
        [renormalize]: 3.30008e-07
        [switch_simplify]: 5.55001e-06
[remove_dup_value]: 6.59001e-06
[partial_unused_args_eliminate]: 1.73002e-06
[environ_conv]: 5.24e-06
[add_recomputation]: 4.393e-05
[cse_after_recomputation]: 1.96e-05, [1]
    [Cycle 1]: 1.43e-05, [1]
        [cse]: 8.67e-06
[auto_monad_reorder]: 1.42e-05
[get_jit_bprop_graph]: 2.41e-06
[rewriter_after_jit_bprop_graph]: 6.46999e-06
[opt_after_jit_grad]: 0.0004865
[symbol_engine_optimizer]: 7.198e-05, [1]
    [Cycle 1]: 6.522e-05, [6]
        [build]: 3.89002e-06
        [elim_shapecalc]: 7.5e-06
        [elim_not_effective]: 1.237e-05
        [opt_reshape]: 5.71e-06
        [fold_const_symbol]: 8.38999e-06
        [renormalize]: 5.19998e-07
[validate]: 3.426e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00594126
[execute]: 8.97e-06
Sums
    bootstrap                            :     0.000276s :  2.00%
    type_inference                       :     0.004093s : 29.61%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000041s :  0.30%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.20%
    rewriter_before_opt_a                :     0.000060s :  0.43%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000049s :  0.35%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000562s :  4.07%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000015s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000014s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000550s :  3.98%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000030s :  0.22%
    jit_opt_a.replace_applicator         :     0.000017s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000031s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000636s :  4.60%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000018s :  0.13%
    loop_unroll                          :     0.000453s :  3.28%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000007s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000044s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000486s :  3.52%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005941s : 42.98%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000146    23
 1.25% :     0.000002s :      2: substitution.elim_not_effective
 0.92% :     0.000001s :      2: substitution.fold_const_symbol
 2.45% :     0.000004s :      3: substitution.graph_param_transform
84.16% :     0.000123s :      4: substitution.inline
 1.66% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.61% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.91% :     0.000003s :      2: substitution.replace_old_param
 5.04% :     0.000007s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004044     2
82.29% :     0.003327s :      1: type_inference.infer
17.71% :     0.000716s :      1: type_inference.specialize
------[replace.]   0.000058     6
65.65% :     0.000038s :      4: replace.inline
34.35% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000127     6
95.05% :     0.000121s :      4: match.inline
 4.95% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.81% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.49% :     0.000002s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.25% :     0.000006s :     22: predicate.inline
 1.02% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.58% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 2.10% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000000s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.49% :     0.000001s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.63% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.60% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 1.13% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.72% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.52% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.07% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000496    12
47.31% :     0.000235s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.48% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.20% :     0.000234s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018803    76
 0.25% :     0.000047s :      1: add_recomputation
 0.24% :     0.000045s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.53% :     0.000287s :      1: bootstrap
 0.11% :     0.000021s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
24.96% :     0.004694s :      1: jit_opt_a
 0.74% :     0.000139s :      1: jit_opt_after_cconv
 0.27% :     0.000051s :      1: jit_opt_b
 2.46% :     0.000462s :      1: loop_unroll
 3.43% :     0.000645s :      1: mutable_eliminate
 4.01% :     0.000754s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.63% :     0.000495s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000031s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000009s :      1: remove_dup_value
 1.53% :     0.000288s :      1: renormalize.infer
 1.36% :     0.000256s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.33% :     0.000063s :      1: rewriter_before_opt_a
 0.40% :     0.000075s :      1: symbol_engine_optimizer
31.69% :     0.005958s :      1: task_emit
21.82% :     0.004103s :      1: type_inference
 0.29% :     0.000054s :      1: validate


TotalTime = 0.0175127, [33]
[bootstrap]: 0.00027148
[type_inference]: 0.00441422
[event_method]: 1.555e-05
[auto_monad]: 5.321e-05
[graph_reusing]: 4.90001e-06
[pre_auto_parallel]: 2.41998e-06
[py_interpret_to_execute]: 3.306e-05
[rewriter_before_opt_a]: 6.901e-05
[expand_dump_flag]: 2.52001e-06
[jit_opt_a]: 0.00519678, [2]
    [Cycle 1]: 0.00163087, [27]
        [switch_simplify]: 4.941e-05
        [loop_unroll]: 2.867e-05
        [a_1]: 0.00050305
        [with_stream_mark]: 1.577e-05
        [recompute_prepare]: 0.00010533
        [updatestate_depend_eliminate]: 3.41999e-06
        [updatestate_assign_eliminate]: 2.31998e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.90001e-06
        [specialize_transform]: 6.16e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.52999e-06
        [meta_shard_fg_expand]: 1.54e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 3.41001e-06
        [cell_reuse_recompute_pass]: 1.29998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.192e-05
        [j_node_and_user_rematch]: 8.70001e-06
        [meta_fg_expand]: 1.89999e-06
        [replace_old_param]: 8.75999e-06
        [inline_without_move]: 5.25001e-06
        [renormalize]: 0.00062901
        [add_forward_monad_depend]: 4.99e-06
        [auto_monad_grad]: 2.45002e-06
        [auto_monad_eliminator]: 1.229e-05
        [cse]: 2.123e-05
        [replace_applicator]: 1.247e-05
    [Cycle 2]: 0.00031563, [27]
        [switch_simplify]: 5.90002e-06
        [loop_unroll]: 5.03002e-06
        [a_1]: 8.84e-05
        [with_stream_mark]: 9.00999e-06
        [recompute_prepare]: 4.89998e-06
        [updatestate_depend_eliminate]: 2.74001e-06
        [updatestate_assign_eliminate]: 2.10002e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 4.81002e-06
        [updatestate_useless_node_eliminater]: 4.66002e-06
        [accelerated_algorithm]: 4.67e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 4.35e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 2.88998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.195e-05
        [j_node_and_user_rematch]: 7.73001e-06
        [meta_fg_expand]: 1.48002e-06
        [replace_old_param]: 7.77e-06
        [inline_without_move]: 4.55999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 4.90001e-06
        [cse]: 9.94001e-06
        [replace_applicator]: 4.93001e-06
[py_interpret_to_execute_after_opt_a]: 1.032e-05
[rewriter_after_opt_a]: 3.139e-05
[convert_after_rewriter]: 6.54999e-06
[order_py_execute_after_rewriter]: 4.80001e-06
[mutable_eliminate]: 0.00059315
[jit_opt_b]: 4.869e-05, [1]
    [Cycle 1]: 4.173e-05, [2]
        [frontend_op_eliminate]: 1.629e-05
        [inline_after_opt_a]: 1.406e-05
[cconv]: 2.608e-05
[loop_unroll]: 0.00042965
[jit_opt_after_cconv]: 0.00013915, [1]
    [Cycle 1]: 0.00013257, [11]
        [c_1]: 2.019e-05
        [parameter_eliminate]: 2.73e-06
        [updatestate_depend_eliminate]: 5.25999e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [cse]: 2.077e-05
        [call_graph_tuple_transform]: 1.807e-05
        [tuple_list_get_item_eliminator]: 5.17999e-06
        [none_parameter_eliminate]: 1.37e-06
        [renormalize]: 4.10015e-07
        [switch_simplify]: 5.15001e-06
[remove_dup_value]: 1.135e-05
[partial_unused_args_eliminate]: 2.31998e-06
[environ_conv]: 4.30999e-06
[add_recomputation]: 5.225e-05
[cse_after_recomputation]: 2.219e-05, [1]
    [Cycle 1]: 1.657e-05, [1]
        [cse]: 1.002e-05
[auto_monad_reorder]: 1.181e-05
[get_jit_bprop_graph]: 1.86e-06
[rewriter_after_jit_bprop_graph]: 4.85999e-06
[opt_after_jit_grad]: 0.00047993
[symbol_engine_optimizer]: 7.183e-05, [1]
    [Cycle 1]: 6.566e-05, [6]
        [build]: 3.81999e-06
        [elim_shapecalc]: 8.19002e-06
        [elim_not_effective]: 1.29e-05
        [opt_reshape]: 5.32001e-06
        [fold_const_symbol]: 8.27e-06
        [renormalize]: 5.19998e-07
[validate]: 3.198e-05
[backend_pass]: 1.02e-06
[task_emit]: 0.00526437
[execute]: 4.95001e-06
Sums
    bootstrap                            :     0.000271s :  1.99%
    type_inference                       :     0.004414s : 32.31%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000053s :  0.39%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.24%
    rewriter_before_opt_a                :     0.000069s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000591s :  4.33%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000110s :  0.81%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000629s :  4.60%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000031s :  0.23%
    jit_opt_a.replace_applicator         :     0.000017s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.08%
    rewriter_after_opt_a                 :     0.000031s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000593s :  4.34%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000026s :  0.19%
    loop_unroll                          :     0.000430s :  3.14%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000011s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000052s :  0.38%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000012s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000480s :  3.51%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005264s : 38.53%
    execute                              :     0.000005s :  0.04%


Time group info:
------[substitution.]   0.000186    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.72% :     0.000005s :      3: substitution.graph_param_transform
82.93% :     0.000155s :      4: substitution.inline
 1.70% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.16% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.69% :     0.000005s :      2: substitution.replace_old_param
 6.11% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004358     2
81.81% :     0.003565s :      1: type_inference.infer
18.19% :     0.000793s :      1: type_inference.specialize
------[replace.]   0.000063     6
66.33% :     0.000042s :      4: replace.inline
33.67% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000163     6
93.65% :     0.000152s :      4: match.inline
 6.35% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.31% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.36% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.49% :     0.000002s :     10: predicate.addn_zero_filter
 2.01% :     0.000002s :     10: predicate.arithmetic_simplify
 1.33% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.86% :     0.000002s :     10: predicate.depend_value_elim
 1.38% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.45% :     0.000002s :     10: predicate.environ_get_eliminate
 1.23% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 4.79% :     0.000005s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 1.70% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.24% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 2.05% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000002s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.35% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.54% :     0.000008s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.26% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.00% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000534    12
47.31% :     0.000252s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.68% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.01% :     0.000251s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019125    76
 0.29% :     0.000055s :      1: add_recomputation
 0.29% :     0.000056s :      1: auto_monad
 0.07% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.49% :     0.000285s :      1: bootstrap
 0.15% :     0.000029s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.04% :     0.000009s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
27.19% :     0.005199s :      1: jit_opt_a
 0.74% :     0.000142s :      1: jit_opt_after_cconv
 0.27% :     0.000051s :      1: jit_opt_b
 2.29% :     0.000439s :      1: loop_unroll
 3.15% :     0.000603s :      1: mutable_eliminate
 4.66% :     0.000891s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.56% :     0.000489s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000036s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000014s :      1: remove_dup_value
 1.71% :     0.000328s :      1: renormalize.infer
 1.54% :     0.000294s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.38% :     0.000073s :      1: rewriter_before_opt_a
 0.39% :     0.000074s :      1: symbol_engine_optimizer
27.57% :     0.005274s :      1: task_emit
23.15% :     0.004427s :      1: type_inference
 0.26% :     0.000050s :      1: validate


TotalTime = 0.0202789, [33]
[bootstrap]: 0.00035929
[type_inference]: 0.00521691
[event_method]: 1.884e-05
[auto_monad]: 5.539e-05
[graph_reusing]: 4.58001e-06
[pre_auto_parallel]: 2.27999e-06
[py_interpret_to_execute]: 4.348e-05
[rewriter_before_opt_a]: 6.851e-05
[expand_dump_flag]: 3.11001e-06
[jit_opt_a]: 0.00607214, [2]
    [Cycle 1]: 0.00195591, [27]
        [switch_simplify]: 5.215e-05
        [loop_unroll]: 3.009e-05
        [a_1]: 0.00052533
        [with_stream_mark]: 1.389e-05
        [recompute_prepare]: 7.06001e-06
        [updatestate_depend_eliminate]: 2.96001e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 6.02999e-06
        [updatestate_useless_node_eliminater]: 5.57999e-06
        [accelerated_algorithm]: 5.74e-06
        [meta_shard_fg_expand]: 1.40999e-06
        [get_grad_eliminate_]: 5.14998e-06
        [merge_forward]: 2.63e-06
        [cell_reuse_recompute_pass]: 1.00999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.164e-05
        [j_node_and_user_rematch]: 8.94e-06
        [meta_fg_expand]: 1.77001e-06
        [replace_old_param]: 8.12e-06
        [inline_without_move]: 5.69e-06
        [renormalize]: 0.0010232
        [add_forward_monad_depend]: 6.31e-06
        [auto_monad_grad]: 1.66e-06
        [auto_monad_eliminator]: 1.283e-05
        [cse]: 2.036e-05
        [replace_applicator]: 1.374e-05
    [Cycle 2]: 0.00032875, [27]
        [switch_simplify]: 5.54e-06
        [loop_unroll]: 4.82998e-06
        [a_1]: 9.39e-05
        [with_stream_mark]: 1.06e-05
        [recompute_prepare]: 4.89e-06
        [updatestate_depend_eliminate]: 2.44001e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 4.87e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 2.03002e-06
        [get_grad_eliminate_]: 4.69002e-06
        [merge_forward]: 2.35997e-06
        [cell_reuse_recompute_pass]: 2.93e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.238e-05
        [j_node_and_user_rematch]: 8e-06
        [meta_fg_expand]: 1.43002e-06
        [replace_old_param]: 7.71999e-06
        [inline_without_move]: 4.94e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.21002e-06
        [auto_monad_grad]: 7.79983e-07
        [auto_monad_eliminator]: 4.88001e-06
        [cse]: 1.084e-05
        [replace_applicator]: 4.86997e-06
[py_interpret_to_execute_after_opt_a]: 1.075e-05
[rewriter_after_opt_a]: 3.181e-05
[convert_after_rewriter]: 6.66e-06
[order_py_execute_after_rewriter]: 4.55999e-06
[mutable_eliminate]: 0.00065826
[jit_opt_b]: 5.197e-05, [1]
    [Cycle 1]: 4.508e-05, [2]
        [frontend_op_eliminate]: 1.774e-05
        [inline_after_opt_a]: 1.573e-05
[cconv]: 2.519e-05
[loop_unroll]: 0.00046523
[jit_opt_after_cconv]: 0.00015082, [1]
    [Cycle 1]: 0.00014414, [11]
        [c_1]: 2.173e-05
        [parameter_eliminate]: 4.12e-06
        [updatestate_depend_eliminate]: 6.69999e-06
        [updatestate_assign_eliminate]: 2.90998e-06
        [updatestate_loads_eliminate]: 2.39001e-06
        [cse]: 2.248e-05
        [call_graph_tuple_transform]: 1.942e-05
        [tuple_list_get_item_eliminator]: 5.39e-06
        [none_parameter_eliminate]: 1.17e-06
        [renormalize]: 9.60019e-07
        [switch_simplify]: 5.29e-06
[remove_dup_value]: 1.045e-05
[partial_unused_args_eliminate]: 2.52001e-06
[environ_conv]: 4.79e-06
[add_recomputation]: 4.216e-05
[cse_after_recomputation]: 2.235e-05, [1]
    [Cycle 1]: 1.624e-05, [1]
        [cse]: 1.012e-05
[auto_monad_reorder]: 1.161e-05
[get_jit_bprop_graph]: 1.74e-06
[rewriter_after_jit_bprop_graph]: 6.62002e-06
[opt_after_jit_grad]: 0.00049112
[symbol_engine_optimizer]: 7.436e-05, [1]
    [Cycle 1]: 6.766e-05, [6]
        [build]: 3.68999e-06
        [elim_shapecalc]: 7.62998e-06
        [elim_not_effective]: 1.318e-05
        [opt_reshape]: 5.48002e-06
        [fold_const_symbol]: 8.67e-06
        [renormalize]: 4.59986e-07
[validate]: 3.22e-05
[backend_pass]: 8.40024e-07
[task_emit]: 0.0060925
[execute]: 1.051e-05
Sums
    bootstrap                            :     0.000359s :  2.27%
    type_inference                       :     0.005217s : 32.92%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000055s :  0.35%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.27%
    rewriter_before_opt_a                :     0.000069s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.36%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000619s :  3.91%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.001023s :  6.46%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.11%
    jit_opt_a.cse                        :     0.000031s :  0.20%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000658s :  4.15%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000025s :  0.16%
    loop_unroll                          :     0.000465s :  2.94%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000010s :  0.07%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000042s :  0.27%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000491s :  3.10%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006092s : 38.45%
    execute                              :     0.000011s :  0.07%


Time group info:
------[substitution.]   0.000189    23
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 2.73% :     0.000005s :      3: substitution.graph_param_transform
84.58% :     0.000160s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.43% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.25% :     0.000004s :      2: substitution.replace_old_param
 4.87% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005146     2
81.45% :     0.004191s :      1: type_inference.infer
18.55% :     0.000955s :      1: type_inference.specialize
------[replace.]   0.000072     6
65.98% :     0.000047s :      4: replace.inline
34.02% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000166     6
95.12% :     0.000158s :      4: match.inline
 4.88% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.54% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.21% :     0.000001s :     10: predicate.addn_zero_filter
 2.32% :     0.000003s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.44% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.56% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.58% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.25% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.25% :     0.000006s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.88% :     0.000002s :     15: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.10% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.84% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.08% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.39% :     0.000003s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.61% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.05% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      6: predicate.special_op_eliminate
 1.13% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.71% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.81% :     0.000008s :     53: predicate.switch_simplify
 1.32% :     0.000002s :     10: predicate.tile_eliminate
 1.26% :     0.000001s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.58% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.89% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.51% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000717    12
45.21% :     0.000324s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.42% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.37% :     0.000361s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022226    76
 0.20% :     0.000045s :      1: add_recomputation
 0.26% :     0.000059s :      1: auto_monad
 0.06% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.70% :     0.000379s :      1: bootstrap
 0.12% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
27.34% :     0.006076s :      1: jit_opt_a
 0.69% :     0.000154s :      1: jit_opt_after_cconv
 0.25% :     0.000055s :      1: jit_opt_b
 2.13% :     0.000473s :      1: loop_unroll
 3.00% :     0.000668s :      1: mutable_eliminate
 3.72% :     0.000827s :     26: opt.transform.jit_opt_a
 0.22% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.25% :     0.000499s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.21% :     0.000047s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000013s :      1: remove_dup_value
 2.74% :     0.000608s :      1: renormalize.infer
 1.82% :     0.000405s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000035s :      1: rewriter_after_opt_a
 0.32% :     0.000072s :      1: rewriter_before_opt_a
 0.35% :     0.000077s :      1: symbol_engine_optimizer
27.51% :     0.006114s :      1: task_emit
23.55% :     0.005234s :      1: type_inference
 0.23% :     0.000052s :      1: validate


TotalTime = 0.0146656, [30]
[bootstrap]: 0.00030908
[type_inference]: 0.00683864
[event_method]: 2.039e-05
[auto_monad]: 8.283e-05
[graph_reusing]: 6.36e-06
[pre_auto_parallel]: 2.37001e-06
[py_interpret_to_execute]: 5.016e-05
[rewriter_before_opt_a]: 8.266e-05
[expand_dump_flag]: 3.02002e-06
[jit_opt_a]: 0.00527867, [2]
    [Cycle 1]: 0.00184276, [27]
        [switch_simplify]: 0.00014761
        [loop_unroll]: 3.573e-05
        [a_1]: 0.00068225
        [with_stream_mark]: 1.551e-05
        [recompute_prepare]: 7.07002e-06
        [updatestate_depend_eliminate]: 3.25e-06
        [updatestate_assign_eliminate]: 2.62001e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 1.60001e-06
        [specialize_transform]: 6.04999e-06
        [updatestate_useless_node_eliminater]: 5.13002e-06
        [accelerated_algorithm]: 5.60001e-06
        [meta_shard_fg_expand]: 1.91998e-06
        [get_grad_eliminate_]: 5.22e-06
        [merge_forward]: 3.35e-06
        [cell_reuse_recompute_pass]: 1.19998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.352e-05
        [j_node_and_user_rematch]: 9.29e-06
        [meta_fg_expand]: 2.41998e-06
        [replace_old_param]: 8.89e-06
        [inline_without_move]: 5.55001e-06
        [renormalize]: 0.00066322
        [add_forward_monad_depend]: 5.15001e-06
        [auto_monad_grad]: 2.24001e-06
        [auto_monad_eliminator]: 1.244e-05
        [cse]: 1.863e-05
        [replace_applicator]: 1.126e-05
    [Cycle 2]: 0.00030985, [27]
        [switch_simplify]: 6.06998e-06
        [loop_unroll]: 5.02e-06
        [a_1]: 9.091e-05
        [with_stream_mark]: 8.08001e-06
        [recompute_prepare]: 5.12e-06
        [updatestate_depend_eliminate]: 2.45002e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 4.98001e-06
        [updatestate_useless_node_eliminater]: 4.89998e-06
        [accelerated_algorithm]: 4.95999e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.42998e-06
        [merge_forward]: 2.39001e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.181e-05
        [j_node_and_user_rematch]: 7.84002e-06
        [meta_fg_expand]: 1.60001e-06
        [replace_old_param]: 6.88e-06
        [inline_without_move]: 4.70999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 4.72e-06
        [cse]: 9.35001e-06
        [replace_applicator]: 4.85001e-06
[py_interpret_to_execute_after_opt_a]: 8.72e-06
[rewriter_after_opt_a]: 3.091e-05
[convert_after_rewriter]: 6.83e-06
[order_py_execute_after_rewriter]: 4.90001e-06
[mutable_eliminate]: 0.00047969
[jit_opt_b]: 4.927e-05, [1]
    [Cycle 1]: 4.271e-05, [2]
        [frontend_op_eliminate]: 1.706e-05
        [inline_after_opt_a]: 1.449e-05
[cconv]: 2.311e-05
[loop_unroll]: 0.00042595
[jit_opt_after_cconv]: 0.00013645, [1]
    [Cycle 1]: 0.0001301, [11]
        [c_1]: 2.088e-05
        [parameter_eliminate]: 2.32001e-06
        [updatestate_depend_eliminate]: 5.17999e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [cse]: 1.564e-05
        [call_graph_tuple_transform]: 1.758e-05
        [tuple_list_get_item_eliminator]: 5.20001e-06
        [none_parameter_eliminate]: 1.25999e-06
        [renormalize]: 3.80009e-07
        [switch_simplify]: 6.15002e-06
[remove_dup_value]: 1.253e-05
[partial_unused_args_eliminate]: 2.74001e-06
[environ_conv]: 5.09e-06
[add_recomputation]: 4.426e-05
[cse_after_recomputation]: 2.093e-05, [1]
    [Cycle 1]: 1.531e-05, [1]
        [cse]: 9.34e-06
[auto_monad_reorder]: 1.447e-05
[get_jit_bprop_graph]: 1.86e-06
[rewriter_after_jit_bprop_graph]: 4.68999e-06
[opt_after_jit_grad]: 0.00045292
[symbol_engine_optimizer]: 7.03e-05, [1]
    [Cycle 1]: 6.402e-05, [6]
        [build]: 2.96001e-06
        [elim_shapecalc]: 7.39002e-06
        [elim_not_effective]: 1.174e-05
        [opt_reshape]: 5.42999e-06
        [fold_const_symbol]: 8.27e-06
        [renormalize]: 4.30009e-07
[validate]: 3.171e-05
Sums
    bootstrap                            :     0.000309s :  2.81%
    type_inference                       :     0.006839s : 62.27%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000083s :  0.75%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000050s :  0.46%
    rewriter_before_opt_a                :     0.000083s :  0.75%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000154s :  1.40%
    jit_opt_a.loop_unroll                :     0.000041s :  0.37%
    jit_opt_a.a_1                        :     0.000773s :  7.04%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000663s :  6.04%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.16%
    jit_opt_a.cse                        :     0.000028s :  0.25%
    jit_opt_a.replace_applicator         :     0.000016s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.08%
    rewriter_after_opt_a                 :     0.000031s :  0.28%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000480s :  4.37%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.13%
    cconv                                :     0.000023s :  0.21%
    loop_unroll                          :     0.000426s :  3.88%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000013s :  0.11%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000044s :  0.40%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000014s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000453s :  4.12%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.29%


Time group info:
------[substitution.]   0.000268    26
 0.69% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 1.89% :     0.000005s :      3: substitution.graph_param_transform
85.41% :     0.000229s :      6: substitution.inline
 1.09% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.57% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.21% :     0.000003s :      2: substitution.replace_old_param
 2.97% :     0.000008s :      1: substitution.switch_simplify
 4.71% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006775     2
82.91% :     0.005618s :      1: type_inference.infer
17.09% :     0.001158s :      1: type_inference.specialize
------[replace.]   0.000130     9
51.14% :     0.000066s :      6: replace.inline
25.14% :     0.000033s :      1: replace.switch_simplify
23.73% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000244     9
92.35% :     0.000225s :      6: match.inline
 2.95% :     0.000007s :      1: match.switch_simplify
 4.70% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   801
 1.31% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.16% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     12: predicate.addn_check_dump
 1.43% :     0.000002s :     12: predicate.addn_zero_filter
 2.03% :     0.000003s :     12: predicate.arithmetic_simplify
 1.42% :     0.000002s :     12: predicate.cast_eliminate
 0.68% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     12: predicate.compare_switch_simplify
 1.45% :     0.000002s :     12: predicate.depend_value_elim
 1.13% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.15% :     0.000002s :     12: predicate.environ_get_eliminate
 1.26% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.36% :     0.000000s :      3: predicate.graph_param_transform
 5.50% :     0.000007s :     26: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.02% :     0.000001s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     17: predicate.load_eliminater
 1.21% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.31% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.96% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000002s :     12: predicate.merge_addn
 1.14% :     0.000002s :     12: predicate.minmaximum_grad
 1.31% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     17: predicate.partial_eliminate
 1.18% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     12: predicate.reduce_eliminate
 1.44% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     20: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     12: predicate.reshape_eliminate
 1.35% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     12: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.38% :     0.000002s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.69% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.72% :     0.000004s :     20: predicate.switch_defer_inline
 2.37% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.91% :     0.000012s :     65: predicate.switch_simplify
 1.20% :     0.000002s :     12: predicate.tile_eliminate
 1.36% :     0.000002s :     12: predicate.transpose_eliminate
 1.54% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.33% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.53% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000868    15
51.32% :     0.000446s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.81% :     0.000094s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.87% :     0.000329s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016464    72
 0.29% :     0.000047s :      1: add_recomputation
 0.52% :     0.000086s :      1: auto_monad
 0.10% :     0.000017s :      1: auto_monad_reorder
 1.97% :     0.000324s :      1: bootstrap
 0.16% :     0.000026s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
32.08% :     0.005282s :      1: jit_opt_a
 0.85% :     0.000139s :      1: jit_opt_after_cconv
 0.32% :     0.000052s :      1: jit_opt_b
 2.63% :     0.000434s :      1: loop_unroll
 2.96% :     0.000488s :      1: mutable_eliminate
 6.55% :     0.001079s :     26: opt.transform.jit_opt_a
 0.28% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.80% :     0.000461s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.32% :     0.000053s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000015s :      1: remove_dup_value
 2.26% :     0.000372s :      1: renormalize.infer
 1.72% :     0.000284s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000034s :      1: rewriter_after_opt_a
 0.52% :     0.000085s :      1: rewriter_before_opt_a
 0.44% :     0.000073s :      1: symbol_engine_optimizer
41.62% :     0.006852s :      1: type_inference


TotalTime = 0.0207588, [33]
[bootstrap]: 0.00032888
[type_inference]: 0.00556834
[event_method]: 1.897e-05
[auto_monad]: 6.51e-05
[graph_reusing]: 6.16e-06
[pre_auto_parallel]: 2.66e-06
[py_interpret_to_execute]: 3.607e-05
[rewriter_before_opt_a]: 7.243e-05
[expand_dump_flag]: 3.28e-06
[jit_opt_a]: 0.0054799, [2]
    [Cycle 1]: 0.00166155, [27]
        [switch_simplify]: 5.555e-05
        [loop_unroll]: 2.945e-05
        [a_1]: 0.0005419
        [with_stream_mark]: 2.128e-05
        [recompute_prepare]: 8.18999e-06
        [updatestate_depend_eliminate]: 3.85e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.66999e-06
        [parameter_eliminate]: 2.46998e-06
        [specialize_transform]: 6.19999e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 5.36002e-06
        [merge_forward]: 3.62002e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.431e-05
        [j_node_and_user_rematch]: 8.66002e-06
        [meta_fg_expand]: 2.61e-06
        [replace_old_param]: 8.96002e-06
        [inline_without_move]: 5.12e-06
        [renormalize]: 0.00068579
        [add_forward_monad_depend]: 5.87999e-06
        [auto_monad_grad]: 2.59001e-06
        [auto_monad_eliminator]: 1.366e-05
        [cse]: 2.955e-05
        [replace_applicator]: 1.281e-05
    [Cycle 2]: 0.00031757, [27]
        [switch_simplify]: 5.29e-06
        [loop_unroll]: 4.74998e-06
        [a_1]: 8.858e-05
        [with_stream_mark]: 1.022e-05
        [recompute_prepare]: 5.12999e-06
        [updatestate_depend_eliminate]: 3.33e-06
        [updatestate_assign_eliminate]: 2.06003e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.21997e-06
        [specialize_transform]: 5.00999e-06
        [updatestate_useless_node_eliminater]: 5.44e-06
        [accelerated_algorithm]: 4.90999e-06
        [meta_shard_fg_expand]: 1.25001e-06
        [get_grad_eliminate_]: 4.41002e-06
        [merge_forward]: 2.17001e-06
        [cell_reuse_recompute_pass]: 1.33002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.242e-05
        [j_node_and_user_rematch]: 7.71999e-06
        [meta_fg_expand]: 1.42e-06
        [replace_old_param]: 7.21001e-06
        [inline_without_move]: 4.55999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 1.11997e-06
        [auto_monad_eliminator]: 4.67998e-06
        [cse]: 1.036e-05
        [replace_applicator]: 5.11002e-06
[py_interpret_to_execute_after_opt_a]: 1.097e-05
[rewriter_after_opt_a]: 3.525e-05
[convert_after_rewriter]: 7.05e-06
[order_py_execute_after_rewriter]: 4.99e-06
[mutable_eliminate]: 0.0005884
[jit_opt_b]: 5.202e-05, [1]
    [Cycle 1]: 4.373e-05, [2]
        [frontend_op_eliminate]: 1.7e-05
        [inline_after_opt_a]: 1.458e-05
[cconv]: 2.737e-05
[loop_unroll]: 0.00043224
[jit_opt_after_cconv]: 0.00017834, [1]
    [Cycle 1]: 0.00017123, [11]
        [c_1]: 4.581e-05
        [parameter_eliminate]: 3.29001e-06
        [updatestate_depend_eliminate]: 6.11998e-06
        [updatestate_assign_eliminate]: 2.60002e-06
        [updatestate_loads_eliminate]: 2.23002e-06
        [cse]: 1.993e-05
        [call_graph_tuple_transform]: 2.083e-05
        [tuple_list_get_item_eliminator]: 6.01998e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 6.11998e-06
[remove_dup_value]: 1.381e-05
[partial_unused_args_eliminate]: 2.70002e-06
[environ_conv]: 5.69999e-06
[add_recomputation]: 5.211e-05
[cse_after_recomputation]: 2.121e-05, [1]
    [Cycle 1]: 1.559e-05, [1]
        [cse]: 9.22001e-06
[auto_monad_reorder]: 1.489e-05
[get_jit_bprop_graph]: 1.89999e-06
[rewriter_after_jit_bprop_graph]: 6.04001e-06
[opt_after_jit_grad]: 0.00049822
[symbol_engine_optimizer]: 8.031e-05, [1]
    [Cycle 1]: 7.336e-05, [6]
        [build]: 3.26999e-06
        [elim_shapecalc]: 9.03002e-06
        [elim_not_effective]: 1.437e-05
        [opt_reshape]: 6.36e-06
        [fold_const_symbol]: 9.14e-06
        [renormalize]: 5.50004e-07
[validate]: 3.99e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.0068756
[execute]: 8.52e-06
Sums
    bootstrap                            :     0.000329s :  1.98%
    type_inference                       :     0.005568s : 33.51%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000065s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.22%
    rewriter_before_opt_a                :     0.000072s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.37%
    jit_opt_a.loop_unroll                :     0.000034s :  0.21%
    jit_opt_a.a_1                        :     0.000630s :  3.79%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000686s :  4.13%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.11%
    jit_opt_a.cse                        :     0.000040s :  0.24%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000588s :  3.54%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000027s :  0.16%
    loop_unroll                          :     0.000432s :  2.60%
    jit_opt_after_cconv.c_1              :     0.000046s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000052s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000498s :  3.00%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006876s : 41.38%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000218    23
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000002s :      2: substitution.fold_const_symbol
 2.48% :     0.000005s :      3: substitution.graph_param_transform
84.83% :     0.000185s :      4: substitution.inline
 1.38% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.91% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000004s :      2: substitution.replace_old_param
 5.75% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005493     2
85.03% :     0.004670s :      1: type_inference.infer
14.97% :     0.000822s :      1: type_inference.specialize
------[replace.]   0.000066     6
67.86% :     0.000045s :      4: replace.inline
32.14% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000194     6
94.13% :     0.000183s :      4: match.inline
 5.87% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.20% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000001s :     10: predicate.addn_zero_filter
 2.58% :     0.000003s :     10: predicate.arithmetic_simplify
 1.28% :     0.000002s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.17% :     0.000001s :     10: predicate.compare_switch_simplify
 1.51% :     0.000002s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.14% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.86% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.47% :     0.000002s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.51% :     0.000001s :      3: predicate.graph_param_transform
 5.76% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      6: predicate.less_batch_normalization
 1.35% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.12% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.13% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 2.38% :     0.000003s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000002s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.67% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.88% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.39% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.20% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.12% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.67% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.27% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.91% :     0.000009s :     53: predicate.switch_simplify
 1.06% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.98% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.92% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.91% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000614    12
52.29% :     0.000321s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.66% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.05% :     0.000252s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022414    76
 0.25% :     0.000055s :      1: add_recomputation
 0.31% :     0.000068s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.54% :     0.000345s :      1: bootstrap
 0.13% :     0.000030s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000024s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
24.46% :     0.005483s :      1: jit_opt_a
 0.81% :     0.000181s :      1: jit_opt_after_cconv
 0.25% :     0.000055s :      1: jit_opt_b
 1.96% :     0.000440s :      1: loop_unroll
 2.67% :     0.000598s :      1: mutable_eliminate
 3.74% :     0.000838s :     26: opt.transform.jit_opt_a
 0.33% :     0.000074s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.27% :     0.000508s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000039s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.67% :     0.000373s :      1: renormalize.infer
 1.36% :     0.000305s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000038s :      1: rewriter_after_opt_a
 0.34% :     0.000075s :      1: rewriter_before_opt_a
 0.37% :     0.000083s :      1: symbol_engine_optimizer
30.75% :     0.006893s :      1: task_emit
24.93% :     0.005588s :      1: type_inference
 0.28% :     0.000062s :      1: validate


TotalTime = 0.0141457, [30]
[bootstrap]: 0.00025429
[type_inference]: 0.00600721
[event_method]: 2.123e-05
[auto_monad]: 0.00011254
[graph_reusing]: 6.29001e-06
[pre_auto_parallel]: 2.53e-06
[py_interpret_to_execute]: 5.491e-05
[rewriter_before_opt_a]: 8.554e-05
[expand_dump_flag]: 2.93e-06
[jit_opt_a]: 0.00544165, [2]
    [Cycle 1]: 0.00203263, [27]
        [switch_simplify]: 0.00010788
        [loop_unroll]: 3.588e-05
        [a_1]: 0.00075817
        [with_stream_mark]: 1.907e-05
        [recompute_prepare]: 7.56999e-06
        [updatestate_depend_eliminate]: 3.43999e-06
        [updatestate_assign_eliminate]: 3.18e-06
        [updatestate_loads_eliminate]: 2.78998e-06
        [parameter_eliminate]: 2.41998e-06
        [specialize_transform]: 6.59999e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 5.86e-06
        [meta_shard_fg_expand]: 2.43998e-06
        [get_grad_eliminate_]: 5.61e-06
        [merge_forward]: 3.78001e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.389e-05
        [j_node_and_user_rematch]: 9.05001e-06
        [meta_fg_expand]: 2.51e-06
        [replace_old_param]: 8.64003e-06
        [inline_without_move]: 5.19e-06
        [renormalize]: 0.00078622
        [add_forward_monad_depend]: 5.47001e-06
        [auto_monad_grad]: 2.26998e-06
        [auto_monad_eliminator]: 1.313e-05
        [cse]: 2.581e-05
        [replace_applicator]: 1.232e-05
    [Cycle 2]: 0.00031855, [27]
        [switch_simplify]: 6.39999e-06
        [loop_unroll]: 4.97999e-06
        [a_1]: 9.181e-05
        [with_stream_mark]: 9.09e-06
        [recompute_prepare]: 5.28002e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 5.04e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 4.74e-06
        [meta_shard_fg_expand]: 1.18001e-06
        [get_grad_eliminate_]: 4.69998e-06
        [merge_forward]: 2.53e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.245e-05
        [j_node_and_user_rematch]: 7.83001e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 7.05e-06
        [inline_without_move]: 4.84e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.37999e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 5.22e-06
        [cse]: 1.089e-05
        [replace_applicator]: 5.14e-06
[py_interpret_to_execute_after_opt_a]: 1.044e-05
[rewriter_after_opt_a]: 3.654e-05
[convert_after_rewriter]: 7.25003e-06
[order_py_execute_after_rewriter]: 5.74999e-06
[mutable_eliminate]: 0.00055762
[jit_opt_b]: 5.237e-05, [1]
    [Cycle 1]: 4.462e-05, [2]
        [frontend_op_eliminate]: 1.763e-05
        [inline_after_opt_a]: 1.435e-05
[cconv]: 2.494e-05
[loop_unroll]: 0.0004552
[jit_opt_after_cconv]: 0.00014959, [1]
    [Cycle 1]: 0.00014307, [11]
        [c_1]: 2.05e-05
        [parameter_eliminate]: 2.71999e-06
        [updatestate_depend_eliminate]: 5.50001e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.40002e-06
        [cse]: 1.748e-05
        [call_graph_tuple_transform]: 1.799e-05
        [tuple_list_get_item_eliminator]: 5.89999e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 7.2e-07
        [switch_simplify]: 5.51e-06
[remove_dup_value]: 1.316e-05
[partial_unused_args_eliminate]: 2.63998e-06
[environ_conv]: 5.47001e-06
[add_recomputation]: 4.677e-05
[cse_after_recomputation]: 2.177e-05, [1]
    [Cycle 1]: 1.596e-05, [1]
        [cse]: 9.79e-06
[auto_monad_reorder]: 1.511e-05
[get_jit_bprop_graph]: 1.74e-06
[rewriter_after_jit_bprop_graph]: 5.03002e-06
[opt_after_jit_grad]: 0.00046582
[symbol_engine_optimizer]: 7.368e-05, [1]
    [Cycle 1]: 6.761e-05, [6]
        [build]: 3.26999e-06
        [elim_shapecalc]: 7.93001e-06
        [elim_not_effective]: 1.306e-05
        [opt_reshape]: 6.23e-06
        [fold_const_symbol]: 8.64e-06
        [renormalize]: 3.80009e-07
[validate]: 3.529e-05
Sums
    bootstrap                            :     0.000254s :  2.43%
    type_inference                       :     0.006007s : 57.41%
    event_method                         :     0.000021s :  0.20%
    auto_monad                           :     0.000113s :  1.08%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000055s :  0.52%
    rewriter_before_opt_a                :     0.000086s :  0.82%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000114s :  1.09%
    jit_opt_a.loop_unroll                :     0.000041s :  0.39%
    jit_opt_a.a_1                        :     0.000850s :  8.12%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000786s :  7.51%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.18%
    jit_opt_a.cse                        :     0.000037s :  0.35%
    jit_opt_a.replace_applicator         :     0.000017s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.10%
    rewriter_after_opt_a                 :     0.000037s :  0.35%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000558s :  5.33%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.14%
    cconv                                :     0.000025s :  0.24%
    loop_unroll                          :     0.000455s :  4.35%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000013s :  0.13%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000047s :  0.45%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000466s :  4.45%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.34%


Time group info:
------[substitution.]   0.000275    26
 0.77% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 1.79% :     0.000005s :      3: substitution.graph_param_transform
85.67% :     0.000236s :      6: substitution.inline
 0.99% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.60% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.32% :     0.000004s :      2: substitution.replace_old_param
 2.78% :     0.000008s :      1: substitution.switch_simplify
 4.62% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005944     2
82.82% :     0.004923s :      1: type_inference.infer
17.18% :     0.001021s :      1: type_inference.specialize
------[replace.]   0.000130     9
50.38% :     0.000066s :      6: replace.inline
22.94% :     0.000030s :      1: replace.switch_simplify
26.69% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000250     9
92.62% :     0.000232s :      6: match.inline
 2.73% :     0.000007s :      1: match.switch_simplify
 4.65% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   801
 1.51% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.27% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     12: predicate.addn_check_dump
 1.34% :     0.000002s :     12: predicate.addn_zero_filter
 2.20% :     0.000003s :     12: predicate.arithmetic_simplify
 1.46% :     0.000002s :     12: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     12: predicate.compare_switch_simplify
 1.46% :     0.000002s :     12: predicate.depend_value_elim
 1.16% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     12: predicate.environ_get_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.28% :     0.000007s :     26: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.04% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     17: predicate.load_eliminater
 1.65% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.25% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.85% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     12: predicate.merge_addn
 1.13% :     0.000002s :     12: predicate.minmaximum_grad
 1.81% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     17: predicate.partial_eliminate
 1.18% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     12: predicate.reduce_eliminate
 1.43% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     20: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.68% :     0.000002s :     12: predicate.reshape_eliminate
 1.40% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     12: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.78% :     0.000001s :      6: predicate.special_op_eliminate
 0.86% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.70% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.67% :     0.000004s :     20: predicate.switch_defer_inline
 2.46% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.49% :     0.000012s :     65: predicate.switch_simplify
 1.33% :     0.000002s :     12: predicate.tile_eliminate
 1.16% :     0.000002s :     12: predicate.transpose_eliminate
 1.44% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.53% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.55% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.90% :     0.000003s :     12: predicate.value_based_eliminate
 0.38% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.87% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000782    15
48.49% :     0.000379s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.18% :     0.000095s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.33% :     0.000307s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016112    72
 0.31% :     0.000050s :      1: add_recomputation
 0.72% :     0.000117s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 1.65% :     0.000266s :      1: bootstrap
 0.17% :     0.000028s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
33.80% :     0.005446s :      1: jit_opt_a
 0.95% :     0.000152s :      1: jit_opt_after_cconv
 0.34% :     0.000055s :      1: jit_opt_b
 2.88% :     0.000464s :      1: loop_unroll
 3.53% :     0.000568s :      1: mutable_eliminate
 6.95% :     0.001119s :     26: opt.transform.jit_opt_a
 0.29% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.95% :     0.000475s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.36% :     0.000058s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.76% :     0.000444s :      1: renormalize.infer
 2.07% :     0.000334s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000041s :      1: rewriter_after_opt_a
 0.55% :     0.000088s :      1: rewriter_before_opt_a
 0.47% :     0.000076s :      1: symbol_engine_optimizer
37.36% :     0.006020s :      1: type_inference


TotalTime = 0.0167251, [33]
[bootstrap]: 0.0002919
[type_inference]: 0.00400733
[event_method]: 1.474e-05
[auto_monad]: 3.912e-05
[graph_reusing]: 3.48e-06
[pre_auto_parallel]: 1.69e-06
[py_interpret_to_execute]: 2.636e-05
[rewriter_before_opt_a]: 6.143e-05
[expand_dump_flag]: 1.98997e-06
[jit_opt_a]: 0.00459178, [2]
    [Cycle 1]: 0.00128679, [27]
        [switch_simplify]: 4.465e-05
        [loop_unroll]: 2.819e-05
        [a_1]: 0.00046387
        [with_stream_mark]: 1.204e-05
        [recompute_prepare]: 6.86999e-06
        [updatestate_depend_eliminate]: 3.21001e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 5.67999e-06
        [updatestate_useless_node_eliminater]: 4.92999e-06
        [accelerated_algorithm]: 5.34998e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 5.24e-06
        [merge_forward]: 2.72001e-06
        [cell_reuse_recompute_pass]: 9.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.081e-05
        [j_node_and_user_rematch]: 8.71997e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 7.44002e-06
        [inline_without_move]: 5.12999e-06
        [renormalize]: 0.00046199
        [add_forward_monad_depend]: 4.78001e-06
        [auto_monad_grad]: 1.38002e-06
        [auto_monad_eliminator]: 9.50001e-06
        [cse]: 1.658e-05
        [replace_applicator]: 1.127e-05
    [Cycle 2]: 0.00030751, [27]
        [switch_simplify]: 4.97e-06
        [loop_unroll]: 4.47998e-06
        [a_1]: 8.588e-05
        [with_stream_mark]: 8.67e-06
        [recompute_prepare]: 4.57998e-06
        [updatestate_depend_eliminate]: 2.93998e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 4.80001e-06
        [updatestate_useless_node_eliminater]: 4.77998e-06
        [accelerated_algorithm]: 5.02e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 4.34002e-06
        [merge_forward]: 2.41998e-06
        [cell_reuse_recompute_pass]: 1.67001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.299e-05
        [j_node_and_user_rematch]: 7.73999e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 7.08e-06
        [inline_without_move]: 4.37e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 9.39996e-07
        [auto_monad_grad]: 6.19999e-07
        [auto_monad_eliminator]: 5.00001e-06
        [cse]: 9.89001e-06
        [replace_applicator]: 5.02e-06
[py_interpret_to_execute_after_opt_a]: 7.92e-06
[rewriter_after_opt_a]: 2.367e-05
[convert_after_rewriter]: 7.17002e-06
[order_py_execute_after_rewriter]: 4.28999e-06
[mutable_eliminate]: 0.00047807
[jit_opt_b]: 4.574e-05, [1]
    [Cycle 1]: 3.965e-05, [2]
        [frontend_op_eliminate]: 1.538e-05
        [inline_after_opt_a]: 1.391e-05
[cconv]: 1.555e-05
[loop_unroll]: 0.00041067
[jit_opt_after_cconv]: 0.00012939, [1]
    [Cycle 1]: 0.00012347, [11]
        [c_1]: 1.924e-05
        [parameter_eliminate]: 2.76e-06
        [updatestate_depend_eliminate]: 5.24e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.683e-05
        [call_graph_tuple_transform]: 1.547e-05
        [tuple_list_get_item_eliminator]: 5.15999e-06
        [none_parameter_eliminate]: 7.80012e-07
        [renormalize]: 4.2998e-07
        [switch_simplify]: 5.22e-06
[remove_dup_value]: 6.21e-06
[partial_unused_args_eliminate]: 1.44998e-06
[environ_conv]: 3.60998e-06
[add_recomputation]: 3.174e-05
[cse_after_recomputation]: 1.992e-05, [1]
    [Cycle 1]: 1.463e-05, [1]
        [cse]: 9.10999e-06
[auto_monad_reorder]: 8.57e-06
[get_jit_bprop_graph]: 1.35999e-06
[rewriter_after_jit_bprop_graph]: 4.44998e-06
[opt_after_jit_grad]: 0.00044603
[symbol_engine_optimizer]: 6.784e-05, [1]
    [Cycle 1]: 6.172e-05, [6]
        [build]: 2.78e-06
        [elim_shapecalc]: 6.91999e-06
        [elim_not_effective]: 1.16e-05
        [opt_reshape]: 5.53002e-06
        [fold_const_symbol]: 7.85e-06
        [renormalize]: 4.89992e-07
[validate]: 2.837e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00576884
[execute]: 7.18e-06
Sums
    bootstrap                            :     0.000292s :  2.21%
    type_inference                       :     0.004007s : 30.41%
    event_method                         :     0.000015s :  0.11%
    auto_monad                           :     0.000039s :  0.30%
    graph_reusing                        :     0.000003s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.20%
    rewriter_before_opt_a                :     0.000061s :  0.47%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000050s :  0.38%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000550s :  4.17%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000462s :  3.51%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000026s :  0.20%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000024s :  0.18%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000478s :  3.63%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000016s :  0.12%
    loop_unroll                          :     0.000411s :  3.12%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000006s :  0.05%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000032s :  0.24%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000009s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000446s :  3.38%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005769s : 43.77%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000148    23
 0.96% :     0.000001s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.25% :     0.000003s :      3: substitution.graph_param_transform
83.49% :     0.000124s :      4: substitution.inline
 1.98% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.82% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.02% :     0.000003s :      2: substitution.replace_old_param
 5.83% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003959     2
82.40% :     0.003263s :      1: type_inference.infer
17.60% :     0.000697s :      1: type_inference.specialize
------[replace.]   0.000059     6
66.15% :     0.000039s :      4: replace.inline
33.85% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000129     6
94.16% :     0.000121s :      4: match.inline
 5.84% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000108   691
 1.31% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.37% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000001s :     10: predicate.addn_zero_filter
 1.89% :     0.000002s :     10: predicate.arithmetic_simplify
 1.31% :     0.000001s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.50% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.30% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.04% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.20% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.81% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.14% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.77% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.70% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000002s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.82% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.60% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.51% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.50% :     0.000008s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.30% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.34% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.45% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.09% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000486    12
48.86% :     0.000238s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.39% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.75% :     0.000222s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018032    76
 0.19% :     0.000035s :      1: add_recomputation
 0.23% :     0.000042s :      1: auto_monad
 0.06% :     0.000011s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.69% :     0.000304s :      1: bootstrap
 0.10% :     0.000018s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.10% :     0.000018s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
25.48% :     0.004594s :      1: jit_opt_a
 0.73% :     0.000132s :      1: jit_opt_after_cconv
 0.27% :     0.000048s :      1: jit_opt_b
 2.32% :     0.000418s :      1: loop_unroll
 2.70% :     0.000486s :      1: mutable_eliminate
 4.11% :     0.000741s :     26: opt.transform.jit_opt_a
 0.23% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.52% :     0.000454s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000029s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000008s :      1: remove_dup_value
 1.31% :     0.000237s :      1: renormalize.infer
 1.22% :     0.000220s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000027s :      1: rewriter_after_opt_a
 0.36% :     0.000064s :      1: rewriter_before_opt_a
 0.39% :     0.000070s :      1: symbol_engine_optimizer
32.05% :     0.005779s :      1: task_emit
22.28% :     0.004018s :      1: type_inference
 0.24% :     0.000044s :      1: validate


TotalTime = 0.0183624, [33]
[bootstrap]: 0.00031668
[type_inference]: 0.00476119
[event_method]: 1.859e-05
[auto_monad]: 5.077e-05
[graph_reusing]: 3.93001e-06
[pre_auto_parallel]: 1.96e-06
[py_interpret_to_execute]: 3.348e-05
[rewriter_before_opt_a]: 6.762e-05
[expand_dump_flag]: 2.56e-06
[jit_opt_a]: 0.00499565, [2]
    [Cycle 1]: 0.00147629, [27]
        [switch_simplify]: 4.708e-05
        [loop_unroll]: 3e-05
        [a_1]: 0.00054433
        [with_stream_mark]: 1.465e-05
        [recompute_prepare]: 7.60998e-06
        [updatestate_depend_eliminate]: 3.31999e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 6.04999e-06
        [updatestate_useless_node_eliminater]: 4.99998e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 5.11002e-06
        [merge_forward]: 2.88e-06
        [cell_reuse_recompute_pass]: 8.70001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.168e-05
        [j_node_and_user_rematch]: 8.65999e-06
        [meta_fg_expand]: 1.77001e-06
        [replace_old_param]: 7.59002e-06
        [inline_without_move]: 5.15999e-06
        [renormalize]: 0.00055102
        [add_forward_monad_depend]: 4.67e-06
        [auto_monad_grad]: 1.44e-06
        [auto_monad_eliminator]: 1.054e-05
        [cse]: 2.11e-05
        [replace_applicator]: 1.184e-05
    [Cycle 2]: 0.00032159, [27]
        [switch_simplify]: 5.59e-06
        [loop_unroll]: 5.15001e-06
        [a_1]: 9.176e-05
        [with_stream_mark]: 9.69999e-06
        [recompute_prepare]: 5.05001e-06
        [updatestate_depend_eliminate]: 2.71e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.02999e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.61002e-06
        [merge_forward]: 2.32001e-06
        [cell_reuse_recompute_pass]: 1.76998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.211e-05
        [j_node_and_user_rematch]: 8.03999e-06
        [meta_fg_expand]: 1.46002e-06
        [replace_old_param]: 7.51999e-06
        [inline_without_move]: 4.74e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 5.04e-06
        [cse]: 1.025e-05
        [replace_applicator]: 5.87999e-06
[py_interpret_to_execute_after_opt_a]: 8.48999e-06
[rewriter_after_opt_a]: 2.799e-05
[convert_after_rewriter]: 5.92999e-06
[order_py_execute_after_rewriter]: 4.20999e-06
[mutable_eliminate]: 0.00053846
[jit_opt_b]: 4.888e-05, [1]
    [Cycle 1]: 4.237e-05, [2]
        [frontend_op_eliminate]: 1.597e-05
        [inline_after_opt_a]: 1.47e-05
[cconv]: 1.947e-05
[loop_unroll]: 0.00041236
[jit_opt_after_cconv]: 0.00013871, [1]
    [Cycle 1]: 0.00013225, [11]
        [c_1]: 2.087e-05
        [parameter_eliminate]: 2.81999e-06
        [updatestate_depend_eliminate]: 5.95002e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [cse]: 1.888e-05
        [call_graph_tuple_transform]: 1.751e-05
        [tuple_list_get_item_eliminator]: 5.42001e-06
        [none_parameter_eliminate]: 1.17e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 5.39998e-06
[remove_dup_value]: 7.98999e-06
[partial_unused_args_eliminate]: 1.58002e-06
[environ_conv]: 4.70001e-06
[add_recomputation]: 3.548e-05
[cse_after_recomputation]: 1.936e-05, [1]
    [Cycle 1]: 1.431e-05, [1]
        [cse]: 8.87e-06
[auto_monad_reorder]: 1.08e-05
[get_jit_bprop_graph]: 1.57001e-06
[rewriter_after_jit_bprop_graph]: 5.02e-06
[opt_after_jit_grad]: 0.00045215
[symbol_engine_optimizer]: 7.379e-05, [1]
    [Cycle 1]: 6.712e-05, [6]
        [build]: 2.67001e-06
        [elim_shapecalc]: 7.44002e-06
        [elim_not_effective]: 1.327e-05
        [opt_reshape]: 5.71e-06
        [fold_const_symbol]: 8.99e-06
        [renormalize]: 4.50003e-07
[validate]: 3.132e-05
[backend_pass]: 6.80011e-07
[task_emit]: 0.00605293
[execute]: 7.98001e-06
Sums
    bootstrap                            :     0.000317s :  2.17%
    type_inference                       :     0.004761s : 32.68%
    event_method                         :     0.000019s :  0.13%
    auto_monad                           :     0.000051s :  0.35%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000033s :  0.23%
    rewriter_before_opt_a                :     0.000068s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.36%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000636s :  4.37%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000551s :  3.78%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.11%
    jit_opt_a.cse                        :     0.000031s :  0.22%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000028s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000538s :  3.70%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000019s :  0.13%
    loop_unroll                          :     0.000412s :  2.83%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000008s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000035s :  0.24%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000011s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000452s :  3.10%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.21%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006053s : 41.55%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000182    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.00% :     0.000004s :      3: substitution.graph_param_transform
85.62% :     0.000156s :      4: substitution.inline
 1.91% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.28% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000003s :      2: substitution.replace_old_param
 4.67% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004699     2
83.45% :     0.003921s :      1: type_inference.infer
16.55% :     0.000778s :      1: type_inference.specialize
------[replace.]   0.000063     6
65.43% :     0.000041s :      4: replace.inline
34.57% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000161     6
95.39% :     0.000153s :      4: match.inline
 4.61% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.27% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 2.13% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000001s :     10: predicate.depend_value_elim
 1.26% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.74% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.51% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.87% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.40% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.82% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.77% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 1.15% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.58% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.27% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.62% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.92% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.11% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000576    12
50.83% :     0.000293s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.88% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.29% :     0.000238s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019850    76
 0.19% :     0.000038s :      1: add_recomputation
 0.27% :     0.000054s :      1: auto_monad
 0.07% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.70% :     0.000337s :      1: bootstrap
 0.11% :     0.000022s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
25.18% :     0.004999s :      1: jit_opt_a
 0.71% :     0.000142s :      1: jit_opt_after_cconv
 0.26% :     0.000051s :      1: jit_opt_b
 2.12% :     0.000421s :      1: loop_unroll
 2.75% :     0.000547s :      1: mutable_eliminate
 4.21% :     0.000836s :     26: opt.transform.jit_opt_a
 0.23% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.32% :     0.000461s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000036s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000010s :      1: remove_dup_value
 1.43% :     0.000283s :      1: renormalize.infer
 1.32% :     0.000262s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000031s :      1: rewriter_after_opt_a
 0.36% :     0.000071s :      1: rewriter_before_opt_a
 0.39% :     0.000076s :      1: symbol_engine_optimizer
30.56% :     0.006067s :      1: task_emit
24.05% :     0.004774s :      1: type_inference
 0.25% :     0.000049s :      1: validate


TotalTime = 0.0132233, [30]
[bootstrap]: 0.00026722
[type_inference]: 0.00579971
[event_method]: 1.985e-05
[auto_monad]: 8.024e-05
[graph_reusing]: 6.36998e-06
[pre_auto_parallel]: 2.68998e-06
[py_interpret_to_execute]: 5.078e-05
[rewriter_before_opt_a]: 8.141e-05
[expand_dump_flag]: 2.81e-06
[jit_opt_a]: 0.00492891, [2]
    [Cycle 1]: 0.00160289, [27]
        [switch_simplify]: 9.904e-05
        [loop_unroll]: 3.56e-05
        [a_1]: 0.00066423
        [with_stream_mark]: 1.332e-05
        [recompute_prepare]: 7.28e-06
        [updatestate_depend_eliminate]: 3.13e-06
        [updatestate_assign_eliminate]: 2.59999e-06
        [updatestate_loads_eliminate]: 2.18998e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 6.00002e-06
        [updatestate_useless_node_eliminater]: 5.30001e-06
        [accelerated_algorithm]: 6.29001e-06
        [meta_shard_fg_expand]: 1.85001e-06
        [get_grad_eliminate_]: 5.40001e-06
        [merge_forward]: 2.93e-06
        [cell_reuse_recompute_pass]: 1.02998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.308e-05
        [j_node_and_user_rematch]: 9.94999e-06
        [meta_fg_expand]: 2.27001e-06
        [replace_old_param]: 8.47998e-06
        [inline_without_move]: 5.25999e-06
        [renormalize]: 0.00049461
        [add_forward_monad_depend]: 4.22e-06
        [auto_monad_grad]: 1.57999e-06
        [auto_monad_eliminator]: 1.151e-05
        [cse]: 1.626e-05
        [replace_applicator]: 1.138e-05
    [Cycle 2]: 0.00031344, [27]
        [switch_simplify]: 5.59998e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 9.191e-05
        [with_stream_mark]: 8.18001e-06
        [recompute_prepare]: 5.04e-06
        [updatestate_depend_eliminate]: 2.41e-06
        [updatestate_assign_eliminate]: 2.05002e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [parameter_eliminate]: 9.00007e-07
        [specialize_transform]: 4.82998e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 4.82e-06
        [meta_shard_fg_expand]: 1.14003e-06
        [get_grad_eliminate_]: 4.56002e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 1.20999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.183e-05
        [j_node_and_user_rematch]: 7.71001e-06
        [meta_fg_expand]: 1.59998e-06
        [replace_old_param]: 7.03e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.02e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 5.02999e-06
        [cse]: 9.31e-06
        [replace_applicator]: 5.14e-06
[py_interpret_to_execute_after_opt_a]: 7.40998e-06
[rewriter_after_opt_a]: 2.655e-05
[convert_after_rewriter]: 6.06e-06
[order_py_execute_after_rewriter]: 5.08002e-06
[mutable_eliminate]: 0.00049398
[jit_opt_b]: 5.89e-05, [1]
    [Cycle 1]: 4.404e-05, [2]
        [frontend_op_eliminate]: 1.699e-05
        [inline_after_opt_a]: 1.527e-05
[cconv]: 2.065e-05
[loop_unroll]: 0.00042318
[jit_opt_after_cconv]: 0.000138, [1]
    [Cycle 1]: 0.00013178, [11]
        [c_1]: 2.125e-05
        [parameter_eliminate]: 2.66e-06
        [updatestate_depend_eliminate]: 5.24e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.13002e-06
        [cse]: 1.666e-05
        [call_graph_tuple_transform]: 1.753e-05
        [tuple_list_get_item_eliminator]: 5.44e-06
        [none_parameter_eliminate]: 1.19e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 5.57999e-06
[remove_dup_value]: 1.097e-05
[partial_unused_args_eliminate]: 2.02999e-06
[environ_conv]: 4.50001e-06
[add_recomputation]: 3.604e-05
[cse_after_recomputation]: 2.013e-05, [1]
    [Cycle 1]: 1.494e-05, [1]
        [cse]: 9.31e-06
[auto_monad_reorder]: 1.149e-05
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 5.12999e-06
[opt_after_jit_grad]: 0.00045391
[symbol_engine_optimizer]: 7.147e-05, [1]
    [Cycle 1]: 6.549e-05, [6]
        [build]: 3.13e-06
        [elim_shapecalc]: 7.78001e-06
        [elim_not_effective]: 1.24e-05
        [opt_reshape]: 5.74e-06
        [fold_const_symbol]: 8.39002e-06
        [renormalize]: 4.80009e-07
[validate]: 2.802e-05
Sums
    bootstrap                            :     0.000267s :  2.77%
    type_inference                       :     0.005800s : 60.13%
    event_method                         :     0.000020s :  0.21%
    auto_monad                           :     0.000080s :  0.83%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000051s :  0.53%
    rewriter_before_opt_a                :     0.000081s :  0.84%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000105s :  1.08%
    jit_opt_a.loop_unroll                :     0.000040s :  0.42%
    jit_opt_a.a_1                        :     0.000756s :  7.84%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000495s :  5.13%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.17%
    jit_opt_a.cse                        :     0.000026s :  0.27%
    jit_opt_a.replace_applicator         :     0.000017s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.08%
    rewriter_after_opt_a                 :     0.000027s :  0.28%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000494s :  5.12%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000021s :  0.21%
    loop_unroll                          :     0.000423s :  4.39%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000011s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000036s :  0.37%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000011s :  0.12%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000454s :  4.71%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.29%


Time group info:
------[substitution.]   0.000241    26
 0.80% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 1.78% :     0.000004s :      3: substitution.graph_param_transform
84.70% :     0.000204s :      6: substitution.inline
 1.24% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.83% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.31% :     0.000003s :      2: substitution.replace_old_param
 3.23% :     0.000008s :      1: substitution.switch_simplify
 4.62% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005746     2
83.22% :     0.004782s :      1: type_inference.infer
16.78% :     0.000964s :      1: type_inference.specialize
------[replace.]   0.000122     9
53.98% :     0.000066s :      6: replace.inline
21.38% :     0.000026s :      1: replace.switch_simplify
24.64% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000218     9
92.24% :     0.000201s :      6: match.inline
 3.19% :     0.000007s :      1: match.switch_simplify
 4.57% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   801
 1.36% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.05% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     12: predicate.addn_check_dump
 1.43% :     0.000002s :     12: predicate.addn_zero_filter
 2.06% :     0.000003s :     12: predicate.arithmetic_simplify
 1.31% :     0.000002s :     12: predicate.cast_eliminate
 0.38% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     12: predicate.compare_switch_simplify
 1.43% :     0.000002s :     12: predicate.depend_value_elim
 1.14% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.44% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.33% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.27% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.27% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     12: predicate.environ_get_eliminate
 1.23% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.18% :     0.000007s :     26: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     17: predicate.load_eliminater
 1.07% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.58% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.83% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     12: predicate.merge_addn
 1.21% :     0.000002s :     12: predicate.minmaximum_grad
 1.39% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.37% :     0.000003s :     17: predicate.partial_eliminate
 1.21% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     12: predicate.reduce_eliminate
 1.46% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     20: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     12: predicate.reshape_eliminate
 1.32% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.62% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.49% :     0.000002s :     12: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.51% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.86% :     0.000004s :     20: predicate.switch_defer_inline
 2.33% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.44% :     0.000011s :     65: predicate.switch_simplify
 1.31% :     0.000002s :     12: predicate.tile_eliminate
 1.23% :     0.000002s :     12: predicate.transpose_eliminate
 1.52% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.37% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.66% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000002s :     12: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000712    15
49.01% :     0.000349s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.10% :     0.000079s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.90% :     0.000284s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014799    72
 0.26% :     0.000039s :      1: add_recomputation
 0.57% :     0.000084s :      1: auto_monad
 0.10% :     0.000014s :      1: auto_monad_reorder
 1.89% :     0.000279s :      1: bootstrap
 0.16% :     0.000023s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
33.32% :     0.004932s :      1: jit_opt_a
 0.95% :     0.000141s :      1: jit_opt_after_cconv
 0.42% :     0.000062s :      1: jit_opt_b
 2.91% :     0.000431s :      1: loop_unroll
 3.40% :     0.000503s :      1: mutable_eliminate
 6.85% :     0.001014s :     26: opt.transform.jit_opt_a
 0.31% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.12% :     0.000462s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.36% :     0.000054s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000013s :      1: remove_dup_value
 1.67% :     0.000248s :      1: renormalize.infer
 1.63% :     0.000241s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000029s :      1: rewriter_after_opt_a
 0.57% :     0.000084s :      1: rewriter_before_opt_a
 0.50% :     0.000074s :      1: symbol_engine_optimizer
39.27% :     0.005812s :      1: type_inference


TotalTime = 0.0152939, [33]
[bootstrap]: 0.00024335
[type_inference]: 0.00381608
[event_method]: 1.519e-05
[auto_monad]: 4.202e-05
[graph_reusing]: 3.81999e-06
[pre_auto_parallel]: 1.86e-06
[py_interpret_to_execute]: 2.578e-05
[rewriter_before_opt_a]: 6.188e-05
[expand_dump_flag]: 2.46e-06
[jit_opt_a]: 0.00444629, [2]
    [Cycle 1]: 0.00129841, [27]
        [switch_simplify]: 4.281e-05
        [loop_unroll]: 2.848e-05
        [a_1]: 0.0004577
        [with_stream_mark]: 1.247e-05
        [recompute_prepare]: 6.88998e-06
        [updatestate_depend_eliminate]: 3.06001e-06
        [updatestate_assign_eliminate]: 2.20002e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.74999e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 5.32001e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 5.15001e-06
        [merge_forward]: 2.52001e-06
        [cell_reuse_recompute_pass]: 1.05999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.051e-05
        [j_node_and_user_rematch]: 8.38001e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 7.61001e-06
        [inline_without_move]: 4.94e-06
        [renormalize]: 0.00047405
        [add_forward_monad_depend]: 4.14002e-06
        [auto_monad_grad]: 1.79e-06
        [auto_monad_eliminator]: 1.022e-05
        [cse]: 1.599e-05
        [replace_applicator]: 1.053e-05
    [Cycle 2]: 0.00030717, [27]
        [switch_simplify]: 5.50001e-06
        [loop_unroll]: 4.70001e-06
        [a_1]: 8.7e-05
        [with_stream_mark]: 8.53001e-06
        [recompute_prepare]: 4.78001e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 1.92001e-06
        [updatestate_loads_eliminate]: 1.86003e-06
        [parameter_eliminate]: 8.90024e-07
        [specialize_transform]: 4.72e-06
        [updatestate_useless_node_eliminater]: 4.78001e-06
        [accelerated_algorithm]: 4.80999e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.54998e-06
        [merge_forward]: 2.24999e-06
        [cell_reuse_recompute_pass]: 1.52999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.098e-05
        [j_node_and_user_rematch]: 7.3e-06
        [meta_fg_expand]: 1.44998e-06
        [replace_old_param]: 6.59999e-06
        [inline_without_move]: 4.53999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 4.80001e-06
        [cse]: 9.64999e-06
        [replace_applicator]: 4.67e-06
[py_interpret_to_execute_after_opt_a]: 8.17e-06
[rewriter_after_opt_a]: 2.661e-05
[convert_after_rewriter]: 6.17001e-06
[order_py_execute_after_rewriter]: 1.191e-05
[mutable_eliminate]: 0.00050287
[jit_opt_b]: 4.69e-05, [1]
    [Cycle 1]: 4.099e-05, [2]
        [frontend_op_eliminate]: 1.592e-05
        [inline_after_opt_a]: 1.363e-05
[cconv]: 1.693e-05
[loop_unroll]: 0.00040015
[jit_opt_after_cconv]: 0.0001294, [1]
    [Cycle 1]: 0.00012359, [11]
        [c_1]: 1.999e-05
        [parameter_eliminate]: 2.22999e-06
        [updatestate_depend_eliminate]: 4.71002e-06
        [updatestate_assign_eliminate]: 2.43998e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [cse]: 1.603e-05
        [call_graph_tuple_transform]: 1.605e-05
        [tuple_list_get_item_eliminator]: 4.96997e-06
        [none_parameter_eliminate]: 9.10019e-07
        [renormalize]: 4.90021e-07
        [switch_simplify]: 5.10001e-06
[remove_dup_value]: 7.63001e-06
[partial_unused_args_eliminate]: 2.47001e-06
[environ_conv]: 4.18001e-06
[add_recomputation]: 3.204e-05
[cse_after_recomputation]: 1.893e-05, [1]
    [Cycle 1]: 1.403e-05, [1]
        [cse]: 8.59e-06
[auto_monad_reorder]: 8.82e-06
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 4.34002e-06
[opt_after_jit_grad]: 0.00043949
[symbol_engine_optimizer]: 6.852e-05, [1]
    [Cycle 1]: 6.267e-05, [6]
        [build]: 2.55997e-06
        [elim_shapecalc]: 7.07002e-06
        [elim_not_effective]: 1.187e-05
        [opt_reshape]: 5.14998e-06
        [fold_const_symbol]: 8.23001e-06
        [renormalize]: 5.69999e-07
[validate]: 2.435e-05
[backend_pass]: 7.00005e-07
[task_emit]: 0.00470405
[execute]: 5.17e-06
Sums
    bootstrap                            :     0.000243s :  2.05%
    type_inference                       :     0.003816s : 32.08%
    event_method                         :     0.000015s :  0.13%
    auto_monad                           :     0.000042s :  0.35%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.22%
    rewriter_before_opt_a                :     0.000062s :  0.52%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000048s :  0.41%
    jit_opt_a.loop_unroll                :     0.000033s :  0.28%
    jit_opt_a.a_1                        :     0.000545s :  4.58%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000021s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.12%
    jit_opt_a.inline_without_move        :     0.000009s :  0.08%
    jit_opt_a.renormalize                :     0.000474s :  3.99%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.13%
    jit_opt_a.cse                        :     0.000026s :  0.22%
    jit_opt_a.replace_applicator         :     0.000015s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.07%
    rewriter_after_opt_a                 :     0.000027s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000012s :  0.10%
    mutable_eliminate                    :     0.000503s :  4.23%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000017s :  0.14%
    loop_unroll                          :     0.000400s :  3.36%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.17%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000008s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000032s :  0.27%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000009s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.04%
    opt_after_jit_grad                   :     0.000439s :  3.69%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000024s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.004704s : 39.55%
    execute                              :     0.000005s :  0.04%


Time group info:
------[substitution.]   0.000147    23
 1.26% :     0.000002s :      2: substitution.elim_not_effective
 0.78% :     0.000001s :      2: substitution.fold_const_symbol
 2.56% :     0.000004s :      3: substitution.graph_param_transform
83.15% :     0.000122s :      4: substitution.inline
 1.77% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.61% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.86% :     0.000003s :      2: substitution.replace_old_param
 6.02% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003771     2
81.87% :     0.003087s :      1: type_inference.infer
18.13% :     0.000684s :      1: type_inference.specialize
------[replace.]   0.000059     6
65.63% :     0.000038s :      4: replace.inline
34.37% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000128     6
93.88% :     0.000120s :      4: match.inline
 6.12% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000107   691
 1.36% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.54% :     0.000002s :     10: predicate.addn_check_dump
 1.44% :     0.000002s :     10: predicate.addn_zero_filter
 1.87% :     0.000002s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.42% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.18% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.30% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.76% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.86% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.58% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000002s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.63% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.85% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 2.24% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.61% :     0.000008s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.18% :     0.000003s :     18: predicate.tuple_list_get_item_eliminator
 1.86% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.78% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000461    12
47.47% :     0.000219s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.65% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.88% :     0.000216s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016598    76
 0.21% :     0.000035s :      1: add_recomputation
 0.27% :     0.000045s :      1: auto_monad
 0.07% :     0.000011s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.53% :     0.000254s :      1: bootstrap
 0.12% :     0.000020s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000018s :      1: event_method
 0.05% :     0.000008s :      1: execute
 0.03% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
26.80% :     0.004449s :      1: jit_opt_a
 0.79% :     0.000132s :      1: jit_opt_after_cconv
 0.30% :     0.000049s :      1: jit_opt_b
 2.45% :     0.000407s :      1: loop_unroll
 3.08% :     0.000511s :      1: mutable_eliminate
 4.40% :     0.000730s :     26: opt.transform.jit_opt_a
 0.26% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.69% :     0.000447s :      1: opt_after_jit_grad
 0.09% :     0.000015s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000028s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000010s :      1: remove_dup_value
 1.43% :     0.000237s :      1: renormalize.infer
 1.39% :     0.000230s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000030s :      1: rewriter_after_opt_a
 0.39% :     0.000065s :      1: rewriter_before_opt_a
 0.43% :     0.000071s :      1: symbol_engine_optimizer
28.39% :     0.004712s :      1: task_emit
23.05% :     0.003827s :      1: type_inference
 0.24% :     0.000039s :      1: validate


TotalTime = 0.0144048, [30]
[bootstrap]: 0.00025388
[type_inference]: 0.0060648
[event_method]: 2.002e-05
[auto_monad]: 8.192e-05
[graph_reusing]: 6.48998e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 5.396e-05
[rewriter_before_opt_a]: 8.438e-05
[expand_dump_flag]: 3.16999e-06
[jit_opt_a]: 0.00569411, [2]
    [Cycle 1]: 0.0020495, [27]
        [switch_simplify]: 0.00011405
        [loop_unroll]: 3.58e-05
        [a_1]: 0.00076726
        [with_stream_mark]: 1.752e-05
        [recompute_prepare]: 7.78001e-06
        [updatestate_depend_eliminate]: 3.34001e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [parameter_eliminate]: 1.75001e-06
        [specialize_transform]: 7e-06
        [updatestate_useless_node_eliminater]: 5.45001e-06
        [accelerated_algorithm]: 5.69e-06
        [meta_shard_fg_expand]: 2.29999e-06
        [get_grad_eliminate_]: 5.63002e-06
        [merge_forward]: 3.12002e-06
        [cell_reuse_recompute_pass]: 1.05999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.512e-05
        [j_node_and_user_rematch]: 1.022e-05
        [meta_fg_expand]: 2.54001e-06
        [replace_old_param]: 1.041e-05
        [inline_without_move]: 5.74e-06
        [renormalize]: 0.00078661
        [add_forward_monad_depend]: 5.68997e-06
        [auto_monad_grad]: 2.24001e-06
        [auto_monad_eliminator]: 1.272e-05
        [cse]: 2.376e-05
        [replace_applicator]: 1.225e-05
    [Cycle 2]: 0.00032897, [27]
        [switch_simplify]: 5.74e-06
        [loop_unroll]: 5.12e-06
        [a_1]: 9.285e-05
        [with_stream_mark]: 9.49e-06
        [recompute_prepare]: 5.38002e-06
        [updatestate_depend_eliminate]: 2.79999e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 9.99979e-07
        [specialize_transform]: 5.11002e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 5.17999e-06
        [meta_shard_fg_expand]: 1.16997e-06
        [get_grad_eliminate_]: 4.69998e-06
        [merge_forward]: 2.55002e-06
        [cell_reuse_recompute_pass]: 1.70001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.255e-05
        [j_node_and_user_rematch]: 7.98001e-06
        [meta_fg_expand]: 1.79998e-06
        [replace_old_param]: 7.55e-06
        [inline_without_move]: 4.81997e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.64e-06
        [auto_monad_grad]: 7.90023e-07
        [auto_monad_eliminator]: 5.12999e-06
        [cse]: 1.063e-05
        [replace_applicator]: 5.12e-06
[py_interpret_to_execute_after_opt_a]: 1.256e-05
[rewriter_after_opt_a]: 3.545e-05
[convert_after_rewriter]: 7.23e-06
[order_py_execute_after_rewriter]: 5.00001e-06
[mutable_eliminate]: 0.00056932
[jit_opt_b]: 5.085e-05, [1]
    [Cycle 1]: 4.357e-05, [2]
        [frontend_op_eliminate]: 1.674e-05
        [inline_after_opt_a]: 1.511e-05
[cconv]: 2.539e-05
[loop_unroll]: 0.00042848
[jit_opt_after_cconv]: 0.00015561, [1]
    [Cycle 1]: 0.00014894, [11]
        [c_1]: 2.154e-05
        [parameter_eliminate]: 2.44999e-06
        [updatestate_depend_eliminate]: 5.37001e-06
        [updatestate_assign_eliminate]: 2.41998e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 1.74e-05
        [call_graph_tuple_transform]: 1.975e-05
        [tuple_list_get_item_eliminator]: 5.66998e-06
        [none_parameter_eliminate]: 1.48002e-06
        [renormalize]: 5.50004e-07
        [switch_simplify]: 1.928e-05
[remove_dup_value]: 1.432e-05
[partial_unused_args_eliminate]: 2.16998e-06
[environ_conv]: 5.42001e-06
[add_recomputation]: 4.749e-05
[cse_after_recomputation]: 2.116e-05, [1]
    [Cycle 1]: 1.58e-05, [1]
        [cse]: 9.94001e-06
[auto_monad_reorder]: 1.472e-05
[get_jit_bprop_graph]: 2.29001e-06
[rewriter_after_jit_bprop_graph]: 5.46e-06
[opt_after_jit_grad]: 0.00046503
[symbol_engine_optimizer]: 7.343e-05, [1]
    [Cycle 1]: 6.698e-05, [6]
        [build]: 3.25e-06
        [elim_shapecalc]: 8.08001e-06
        [elim_not_effective]: 1.263e-05
        [opt_reshape]: 6.19001e-06
        [fold_const_symbol]: 8.77999e-06
        [renormalize]: 4.10015e-07
[validate]: 3.505e-05
Sums
    bootstrap                            :     0.000254s :  2.42%
    type_inference                       :     0.006065s : 57.73%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000082s :  0.78%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000054s :  0.51%
    rewriter_before_opt_a                :     0.000084s :  0.80%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000120s :  1.14%
    jit_opt_a.loop_unroll                :     0.000041s :  0.39%
    jit_opt_a.a_1                        :     0.000860s :  8.19%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.17%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000787s :  7.49%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.17%
    jit_opt_a.cse                        :     0.000034s :  0.33%
    jit_opt_a.replace_applicator         :     0.000017s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.12%
    rewriter_after_opt_a                 :     0.000035s :  0.34%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000569s :  5.42%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.14%
    cconv                                :     0.000025s :  0.24%
    loop_unroll                          :     0.000428s :  4.08%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000019s :  0.18%
    remove_dup_value                     :     0.000014s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000047s :  0.45%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000465s :  4.43%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.33%


Time group info:
------[substitution.]   0.000315    26
 0.60% :     0.000002s :      2: substitution.elim_not_effective
 0.42% :     0.000001s :      2: substitution.fold_const_symbol
 1.93% :     0.000006s :      3: substitution.graph_param_transform
86.54% :     0.000273s :      6: substitution.inline
 0.87% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.44% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.40% :     0.000004s :      2: substitution.replace_old_param
 2.40% :     0.000008s :      1: substitution.switch_simplify
 4.41% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006008     2
80.98% :     0.004865s :      1: type_inference.infer
19.02% :     0.001143s :      1: type_inference.specialize
------[replace.]   0.000143     9
51.47% :     0.000074s :      6: replace.inline
23.23% :     0.000033s :      1: replace.switch_simplify
25.31% :     0.000036s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000289     9
93.27% :     0.000270s :      6: match.inline
 2.34% :     0.000007s :      1: match.switch_simplify
 4.39% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   801
 1.43% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.87% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     12: predicate.addn_check_dump
 1.30% :     0.000002s :     12: predicate.addn_zero_filter
 2.33% :     0.000003s :     12: predicate.arithmetic_simplify
 1.28% :     0.000002s :     12: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000002s :     12: predicate.compare_switch_simplify
 1.47% :     0.000002s :     12: predicate.depend_value_elim
 1.16% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.42% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.70% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.24% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.33% :     0.000002s :     12: predicate.environ_get_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.52% :     0.000007s :     26: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.95% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     17: predicate.load_eliminater
 1.48% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.54% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.94% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000002s :     12: predicate.merge_addn
 1.21% :     0.000002s :     12: predicate.minmaximum_grad
 1.47% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.37% :     0.000003s :     17: predicate.partial_eliminate
 1.13% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     12: predicate.reduce_eliminate
 1.44% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.55% :     0.000002s :     20: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     12: predicate.reshape_eliminate
 1.22% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     12: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.93% :     0.000003s :     12: predicate.split_environ_get_set_with_tuple_value
 1.39% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.00% :     0.000004s :     20: predicate.switch_defer_inline
 2.25% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.34% :     0.000011s :     65: predicate.switch_simplify
 1.35% :     0.000002s :     12: predicate.tile_eliminate
 1.26% :     0.000002s :     12: predicate.transpose_eliminate
 1.72% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.68% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     12: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000839    15
42.46% :     0.000356s :      6: func_graph_cloner_run.FuncGraphClonerGraph
14.51% :     0.000122s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.03% :     0.000361s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016390    72
 0.31% :     0.000050s :      1: add_recomputation
 0.52% :     0.000085s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.62% :     0.000266s :      1: bootstrap
 0.17% :     0.000028s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
34.76% :     0.005698s :      1: jit_opt_a
 0.97% :     0.000158s :      1: jit_opt_after_cconv
 0.33% :     0.000053s :      1: jit_opt_b
 2.66% :     0.000436s :      1: loop_unroll
 3.52% :     0.000578s :      1: mutable_eliminate
 6.95% :     0.001140s :     26: opt.transform.jit_opt_a
 0.30% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.89% :     0.000473s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.35% :     0.000057s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.60% :     0.000426s :      1: renormalize.infer
 2.15% :     0.000352s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000038s :      1: rewriter_after_opt_a
 0.53% :     0.000087s :      1: rewriter_before_opt_a
 0.47% :     0.000076s :      1: symbol_engine_optimizer
37.08% :     0.006077s :      1: type_inference


TotalTime = 0.018122, [33]
[bootstrap]: 0.00026988
[type_inference]: 0.00425591
[event_method]: 1.87e-05
[auto_monad]: 5.713e-05
[graph_reusing]: 4.92e-06
[pre_auto_parallel]: 2.34999e-06
[py_interpret_to_execute]: 3.618e-05
[rewriter_before_opt_a]: 6.861e-05
[expand_dump_flag]: 2.84999e-06
[jit_opt_a]: 0.00528181, [2]
    [Cycle 1]: 0.00153238, [27]
        [switch_simplify]: 4.727e-05
        [loop_unroll]: 2.986e-05
        [a_1]: 0.00050691
        [with_stream_mark]: 1.449e-05
        [recompute_prepare]: 6.89001e-06
        [updatestate_depend_eliminate]: 3.25e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.27999e-06
        [specialize_transform]: 6.01998e-06
        [updatestate_useless_node_eliminater]: 5.40999e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 5.67999e-06
        [merge_forward]: 3.08e-06
        [cell_reuse_recompute_pass]: 9.19972e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.108e-05
        [j_node_and_user_rematch]: 9.04998e-06
        [meta_fg_expand]: 1.89999e-06
        [replace_old_param]: 8.09002e-06
        [inline_without_move]: 5.14e-06
        [renormalize]: 0.0006531
        [add_forward_monad_depend]: 4.29002e-06
        [auto_monad_grad]: 1.57001e-06
        [auto_monad_eliminator]: 9.61e-06
        [cse]: 1.559e-05
        [replace_applicator]: 1.053e-05
    [Cycle 2]: 0.00032161, [27]
        [switch_simplify]: 5.49e-06
        [loop_unroll]: 4.97999e-06
        [a_1]: 8.867e-05
        [with_stream_mark]: 8.69e-06
        [recompute_prepare]: 4.80001e-06
        [updatestate_depend_eliminate]: 2.69999e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.98002e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 5.64998e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 5.34e-06
        [meta_shard_fg_expand]: 1.28002e-06
        [get_grad_eliminate_]: 5.25999e-06
        [merge_forward]: 2.58998e-06
        [cell_reuse_recompute_pass]: 1.64998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.232e-05
        [j_node_and_user_rematch]: 7.71999e-06
        [meta_fg_expand]: 1.55999e-06
        [replace_old_param]: 7.28e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 4.95999e-06
        [cse]: 9.78002e-06
        [replace_applicator]: 5.67001e-06
[py_interpret_to_execute_after_opt_a]: 9.07001e-06
[rewriter_after_opt_a]: 2.871e-05
[convert_after_rewriter]: 5.82001e-06
[order_py_execute_after_rewriter]: 6.46e-06
[mutable_eliminate]: 0.00053265
[jit_opt_b]: 4.951e-05, [1]
    [Cycle 1]: 4.327e-05, [2]
        [frontend_op_eliminate]: 1.678e-05
        [inline_after_opt_a]: 1.516e-05
[cconv]: 1.721e-05
[loop_unroll]: 0.00042018
[jit_opt_after_cconv]: 0.00013473, [1]
    [Cycle 1]: 0.00012868, [11]
        [c_1]: 2.105e-05
        [parameter_eliminate]: 2.46e-06
        [updatestate_depend_eliminate]: 4.78001e-06
        [updatestate_assign_eliminate]: 2.38002e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.786e-05
        [call_graph_tuple_transform]: 1.622e-05
        [tuple_list_get_item_eliminator]: 5.07e-06
        [none_parameter_eliminate]: 8.59989e-07
        [renormalize]: 5.69999e-07
        [switch_simplify]: 5.42999e-06
[remove_dup_value]: 7.95e-06
[partial_unused_args_eliminate]: 1.67999e-06
[environ_conv]: 3.75e-06
[add_recomputation]: 3.376e-05
[cse_after_recomputation]: 1.939e-05, [1]
    [Cycle 1]: 1.437e-05, [1]
        [cse]: 9.02999e-06
[auto_monad_reorder]: 1.161e-05
[get_jit_bprop_graph]: 1.54e-06
[rewriter_after_jit_bprop_graph]: 4.79e-06
[opt_after_jit_grad]: 0.0004563
[symbol_engine_optimizer]: 7.112e-05, [1]
    [Cycle 1]: 6.514e-05, [6]
        [build]: 2.76e-06
        [elim_shapecalc]: 7.78001e-06
        [elim_not_effective]: 1.227e-05
        [opt_reshape]: 5.54e-06
        [fold_const_symbol]: 8.22998e-06
        [renormalize]: 7.89994e-07
[validate]: 2.958e-05
[backend_pass]: 8.30012e-07
[task_emit]: 0.00607221
[execute]: 7.68999e-06
Sums
    bootstrap                            :     0.000270s :  1.91%
    type_inference                       :     0.004256s : 30.18%
    event_method                         :     0.000019s :  0.13%
    auto_monad                           :     0.000057s :  0.41%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.26%
    rewriter_before_opt_a                :     0.000069s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.37%
    jit_opt_a.loop_unroll                :     0.000035s :  0.25%
    jit_opt_a.a_1                        :     0.000596s :  4.22%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.08%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000653s :  4.63%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.10%
    jit_opt_a.cse                        :     0.000025s :  0.18%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000029s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000533s :  3.78%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000017s :  0.12%
    loop_unroll                          :     0.000420s :  2.98%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000008s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000034s :  0.24%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000456s :  3.24%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000030s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006072s : 43.06%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000181    23
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000001s :      2: substitution.fold_const_symbol
 1.97% :     0.000004s :      3: substitution.graph_param_transform
85.37% :     0.000155s :      4: substitution.inline
 1.59% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.32% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.12% :     0.000004s :      2: substitution.replace_old_param
 5.19% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004191     2
80.96% :     0.003393s :      1: type_inference.infer
19.04% :     0.000798s :      1: type_inference.specialize
------[replace.]   0.000065     6
67.12% :     0.000044s :      4: replace.inline
32.88% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000161     6
94.83% :     0.000153s :      4: match.inline
 5.17% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.31% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.58% :     0.000002s :     10: predicate.addn_zero_filter
 2.24% :     0.000003s :     10: predicate.arithmetic_simplify
 1.12% :     0.000001s :     10: predicate.cast_eliminate
 0.56% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000001s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.47% :     0.000001s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.25% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.14% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000001s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.55% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.21% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.74% :     0.000002s :     10: predicate.reduce_eliminate
 1.64% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000001s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.52% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.18% :     0.000001s :      6: predicate.special_op_eliminate
 1.14% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.59% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.58% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.96% :     0.000009s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.57% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.34% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.04% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000568    12
42.53% :     0.000242s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.54% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.93% :     0.000295s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019670    76
 0.19% :     0.000037s :      1: add_recomputation
 0.31% :     0.000060s :      1: auto_monad
 0.07% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.45% :     0.000285s :      1: bootstrap
 0.10% :     0.000020s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.87% :     0.005285s :      1: jit_opt_a
 0.70% :     0.000138s :      1: jit_opt_after_cconv
 0.26% :     0.000052s :      1: jit_opt_b
 2.17% :     0.000428s :      1: loop_unroll
 2.75% :     0.000540s :      1: mutable_eliminate
 4.05% :     0.000796s :     26: opt.transform.jit_opt_a
 0.23% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.36% :     0.000464s :      1: opt_after_jit_grad
 0.04% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000039s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000010s :      1: remove_dup_value
 1.59% :     0.000312s :      1: renormalize.infer
 1.70% :     0.000335s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000032s :      1: rewriter_after_opt_a
 0.36% :     0.000072s :      1: rewriter_before_opt_a
 0.37% :     0.000074s :      1: symbol_engine_optimizer
30.96% :     0.006089s :      1: task_emit
21.71% :     0.004270s :      1: type_inference
 0.24% :     0.000047s :      1: validate


TotalTime = 0.0136978, [30]
[bootstrap]: 0.00024437
[type_inference]: 0.00588019
[event_method]: 1.96e-05
[auto_monad]: 6.364e-05
[graph_reusing]: 4.50999e-06
[pre_auto_parallel]: 2.14999e-06
[py_interpret_to_execute]: 5.085e-05
[rewriter_before_opt_a]: 8.189e-05
[expand_dump_flag]: 2.60002e-06
[jit_opt_a]: 0.00520793, [2]
    [Cycle 1]: 0.00180633, [27]
        [switch_simplify]: 9.8e-05
        [loop_unroll]: 3.908e-05
        [a_1]: 0.00068241
        [with_stream_mark]: 1.476e-05
        [recompute_prepare]: 7.24001e-06
        [updatestate_depend_eliminate]: 3.11999e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [parameter_eliminate]: 1.78002e-06
        [specialize_transform]: 6.99001e-06
        [updatestate_useless_node_eliminater]: 5.62001e-06
        [accelerated_algorithm]: 5.75001e-06
        [meta_shard_fg_expand]: 1.64e-06
        [get_grad_eliminate_]: 5.55001e-06
        [merge_forward]: 3.01001e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.271e-05
        [j_node_and_user_rematch]: 1.02e-05
        [meta_fg_expand]: 2.54001e-06
        [replace_old_param]: 8.73001e-06
        [inline_without_move]: 5.48002e-06
        [renormalize]: 0.00066511
        [add_forward_monad_depend]: 4.94998e-06
        [auto_monad_grad]: 2.04e-06
        [auto_monad_eliminator]: 1.212e-05
        [cse]: 2.004e-05
        [replace_applicator]: 1.223e-05
    [Cycle 2]: 0.00032655, [27]
        [switch_simplify]: 5.69e-06
        [loop_unroll]: 5.69999e-06
        [a_1]: 9.19e-05
        [with_stream_mark]: 8.94e-06
        [recompute_prepare]: 4.81002e-06
        [updatestate_depend_eliminate]: 2.48e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 5.65001e-06
        [updatestate_useless_node_eliminater]: 5.47999e-06
        [accelerated_algorithm]: 5.56002e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 5.30999e-06
        [merge_forward]: 2.69001e-06
        [cell_reuse_recompute_pass]: 1.55999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.482e-05
        [j_node_and_user_rematch]: 8.61002e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 7.16001e-06
        [inline_without_move]: 5.19e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 7.40023e-07
        [auto_monad_eliminator]: 4.67e-06
        [cse]: 9.60001e-06
        [replace_applicator]: 5.12e-06
[py_interpret_to_execute_after_opt_a]: 8.90999e-06
[rewriter_after_opt_a]: 3.041e-05
[convert_after_rewriter]: 6.54999e-06
[order_py_execute_after_rewriter]: 4.77e-06
[mutable_eliminate]: 0.00054588
[jit_opt_b]: 6.453e-05, [1]
    [Cycle 1]: 5.702e-05, [2]
        [frontend_op_eliminate]: 1.779e-05
        [inline_after_opt_a]: 2.653e-05
[cconv]: 2.154e-05
[loop_unroll]: 0.00044669
[jit_opt_after_cconv]: 0.00014094, [1]
    [Cycle 1]: 0.0001345, [11]
        [c_1]: 2.154e-05
        [parameter_eliminate]: 2.68e-06
        [updatestate_depend_eliminate]: 5.51002e-06
        [updatestate_assign_eliminate]: 2.33998e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.733e-05
        [call_graph_tuple_transform]: 1.958e-05
        [tuple_list_get_item_eliminator]: 5.40001e-06
        [none_parameter_eliminate]: 1.20999e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 5.69999e-06
[remove_dup_value]: 1.244e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 4.31002e-06
[add_recomputation]: 4.247e-05
[cse_after_recomputation]: 2.103e-05, [1]
    [Cycle 1]: 1.553e-05, [1]
        [cse]: 9.82999e-06
[auto_monad_reorder]: 3.717e-05
[get_jit_bprop_graph]: 1.90001e-06
[rewriter_after_jit_bprop_graph]: 5.17e-06
[opt_after_jit_grad]: 0.00047269
[symbol_engine_optimizer]: 7.333e-05, [1]
    [Cycle 1]: 6.696e-05, [6]
        [build]: 3.2e-06
        [elim_shapecalc]: 7.57998e-06
        [elim_not_effective]: 1.346e-05
        [opt_reshape]: 5.57001e-06
        [fold_const_symbol]: 8.33001e-06
        [renormalize]: 3.89991e-07
[validate]: 3.347e-05
Sums
    bootstrap                            :     0.000244s :  2.43%
    type_inference                       :     0.005880s : 58.51%
    event_method                         :     0.000020s :  0.20%
    auto_monad                           :     0.000064s :  0.63%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000051s :  0.51%
    rewriter_before_opt_a                :     0.000082s :  0.81%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000104s :  1.03%
    jit_opt_a.loop_unroll                :     0.000045s :  0.45%
    jit_opt_a.a_1                        :     0.000774s :  7.70%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.16%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000665s :  6.62%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.17%
    jit_opt_a.cse                        :     0.000030s :  0.29%
    jit_opt_a.replace_applicator         :     0.000017s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.09%
    rewriter_after_opt_a                 :     0.000030s :  0.30%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000546s :  5.43%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.26%
    cconv                                :     0.000022s :  0.21%
    loop_unroll                          :     0.000447s :  4.44%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000012s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000042s :  0.42%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000037s :  0.37%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000473s :  4.70%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.33%


Time group info:
------[substitution.]   0.000251    26
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 1.99% :     0.000005s :      3: substitution.graph_param_transform
85.21% :     0.000214s :      6: substitution.inline
 1.34% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.77% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.17% :     0.000003s :      2: substitution.replace_old_param
 3.01% :     0.000008s :      1: substitution.switch_simplify
 4.15% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005823     2
81.94% :     0.004771s :      1: type_inference.infer
18.06% :     0.001052s :      1: type_inference.specialize
------[replace.]   0.000121     9
55.21% :     0.000067s :      6: replace.inline
19.54% :     0.000024s :      1: replace.switch_simplify
25.25% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000226     9
92.93% :     0.000210s :      6: match.inline
 2.96% :     0.000007s :      1: match.switch_simplify
 4.11% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   801
 1.37% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.93% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000002s :     12: predicate.addn_check_dump
 1.31% :     0.000002s :     12: predicate.addn_zero_filter
 2.05% :     0.000003s :     12: predicate.arithmetic_simplify
 1.50% :     0.000002s :     12: predicate.cast_eliminate
 0.66% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     12: predicate.compare_switch_simplify
 1.42% :     0.000002s :     12: predicate.depend_value_elim
 1.23% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.40% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.72% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.11% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     12: predicate.environ_get_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 5.27% :     0.000007s :     26: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000002s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.90% :     0.000003s :     17: predicate.load_eliminater
 1.27% :     0.000002s :      3: predicate.loop_unroll_after_grad
 5.13% :     0.000007s :     40: predicate.loop_unroll_before_grad
 1.95% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     12: predicate.merge_addn
 1.25% :     0.000002s :     12: predicate.minmaximum_grad
 1.46% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 1.92% :     0.000003s :     17: predicate.partial_eliminate
 1.27% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.82% :     0.000003s :     12: predicate.reduce_eliminate
 1.50% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.52% :     0.000002s :     20: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     12: predicate.reshape_eliminate
 1.22% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.13% :     0.000002s :      6: predicate.specialize_transform
 1.90% :     0.000003s :     12: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.91% :     0.000004s :     20: predicate.switch_defer_inline
 2.26% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.44% :     0.000012s :     65: predicate.switch_simplify
 1.19% :     0.000002s :     12: predicate.tile_eliminate
 1.38% :     0.000002s :     12: predicate.transpose_eliminate
 1.50% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.12% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000779    15
44.47% :     0.000346s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.60% :     0.000090s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.92% :     0.000342s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015479    72
 0.29% :     0.000045s :      1: add_recomputation
 0.43% :     0.000067s :      1: auto_monad
 0.26% :     0.000040s :      1: auto_monad_reorder
 1.65% :     0.000256s :      1: bootstrap
 0.16% :     0.000025s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
33.67% :     0.005212s :      1: jit_opt_a
 0.93% :     0.000144s :      1: jit_opt_after_cconv
 0.44% :     0.000067s :      1: jit_opt_b
 2.94% :     0.000455s :      1: loop_unroll
 3.59% :     0.000555s :      1: mutable_eliminate
 6.75% :     0.001044s :     26: opt.transform.jit_opt_a
 0.31% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.24% :     0.000037s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.10% :     0.000481s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.35% :     0.000054s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000015s :      1: remove_dup_value
 2.29% :     0.000354s :      1: renormalize.infer
 1.96% :     0.000303s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000033s :      1: rewriter_after_opt_a
 0.55% :     0.000085s :      1: rewriter_before_opt_a
 0.49% :     0.000076s :      1: symbol_engine_optimizer
38.06% :     0.005891s :      1: type_inference


TotalTime = 0.0192473, [33]
[bootstrap]: 0.00025081
[type_inference]: 0.00385916
[event_method]: 1.477e-05
[auto_monad]: 4.634e-05
[graph_reusing]: 4.70999e-06
[pre_auto_parallel]: 2.17001e-06
[py_interpret_to_execute]: 2.777e-05
[rewriter_before_opt_a]: 6.512e-05
[expand_dump_flag]: 2.40002e-06
[jit_opt_a]: 0.00437723, [2]
    [Cycle 1]: 0.00122036, [27]
        [switch_simplify]: 3.997e-05
        [loop_unroll]: 2.832e-05
        [a_1]: 0.00044453
        [with_stream_mark]: 1.028e-05
        [recompute_prepare]: 6.36998e-06
        [updatestate_depend_eliminate]: 2.58e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.98997e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 5.49998e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 1.16002e-06
        [get_grad_eliminate_]: 5.05001e-06
        [merge_forward]: 2.58e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.023e-05
        [j_node_and_user_rematch]: 8.03999e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 7.85e-06
        [inline_without_move]: 5.04e-06
        [renormalize]: 0.00043733
        [add_forward_monad_depend]: 3.86999e-06
        [auto_monad_grad]: 1.17e-06
        [auto_monad_eliminator]: 8.17003e-06
        [cse]: 1.517e-05
        [replace_applicator]: 1.004e-05
    [Cycle 2]: 0.00030679, [27]
        [switch_simplify]: 5.36002e-06
        [loop_unroll]: 4.58999e-06
        [a_1]: 8.773e-05
        [with_stream_mark]: 7.98999e-06
        [recompute_prepare]: 4.60001e-06
        [updatestate_depend_eliminate]: 2.36998e-06
        [updatestate_assign_eliminate]: 1.98002e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 4.67e-06
        [updatestate_useless_node_eliminater]: 4.71002e-06
        [accelerated_algorithm]: 4.84e-06
        [meta_shard_fg_expand]: 1.07998e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.20002e-06
        [cell_reuse_recompute_pass]: 1.21002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.13e-05
        [j_node_and_user_rematch]: 7.46001e-06
        [meta_fg_expand]: 1.50999e-06
        [replace_old_param]: 6.61999e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.17999e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 4.80001e-06
        [cse]: 9.54999e-06
        [replace_applicator]: 4.89003e-06
[py_interpret_to_execute_after_opt_a]: 7.28999e-06
[rewriter_after_opt_a]: 2.702e-05
[convert_after_rewriter]: 5.99e-06
[order_py_execute_after_rewriter]: 4.70001e-06
[mutable_eliminate]: 0.00045566
[jit_opt_b]: 4.537e-05, [1]
    [Cycle 1]: 3.913e-05, [2]
        [frontend_op_eliminate]: 1.491e-05
        [inline_after_opt_a]: 1.366e-05
[cconv]: 1.598e-05
[loop_unroll]: 0.00040098
[jit_opt_after_cconv]: 0.00012676, [1]
    [Cycle 1]: 0.00012124, [11]
        [c_1]: 1.996e-05
        [parameter_eliminate]: 2.49999e-06
        [updatestate_depend_eliminate]: 4.58999e-06
        [updatestate_assign_eliminate]: 2.39999e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 1.551e-05
        [call_graph_tuple_transform]: 1.514e-05
        [tuple_list_get_item_eliminator]: 4.99e-06
        [none_parameter_eliminate]: 9.60019e-07
        [renormalize]: 6.69999e-07
        [switch_simplify]: 4.92e-06
[remove_dup_value]: 6.66e-06
[partial_unused_args_eliminate]: 1.64998e-06
[environ_conv]: 3.34001e-06
[add_recomputation]: 2.997e-05
[cse_after_recomputation]: 1.898e-05, [1]
    [Cycle 1]: 1.412e-05, [1]
        [cse]: 8.62e-06
[auto_monad_reorder]: 8.89e-06
[get_jit_bprop_graph]: 1.32e-06
[rewriter_after_jit_bprop_graph]: 4.18001e-06
[opt_after_jit_grad]: 0.00043226
[symbol_engine_optimizer]: 6.586e-05, [1]
    [Cycle 1]: 5.981e-05, [6]
        [build]: 2.07999e-06
        [elim_shapecalc]: 6.86001e-06
        [elim_not_effective]: 1.108e-05
        [opt_reshape]: 5.47999e-06
        [fold_const_symbol]: 7.85e-06
        [renormalize]: 4.50003e-07
[validate]: 2.425e-05
[backend_pass]: 6.50005e-07
[task_emit]: 0.00873392
[execute]: 6.43e-06
Sums
    bootstrap                            :     0.000251s :  1.58%
    type_inference                       :     0.003859s : 24.34%
    event_method                         :     0.000015s :  0.09%
    auto_monad                           :     0.000046s :  0.29%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.18%
    rewriter_before_opt_a                :     0.000065s :  0.41%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000045s :  0.29%
    jit_opt_a.loop_unroll                :     0.000033s :  0.21%
    jit_opt_a.a_1                        :     0.000532s :  3.36%
    jit_opt_a.with_stream_mark           :     0.000018s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000010s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000014s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000437s :  2.76%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.08%
    jit_opt_a.cse                        :     0.000025s :  0.16%
    jit_opt_a.replace_applicator         :     0.000015s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.05%
    rewriter_after_opt_a                 :     0.000027s :  0.17%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000456s :  2.87%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000016s :  0.10%
    loop_unroll                          :     0.000401s :  2.53%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000007s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000003s :  0.02%
    add_recomputation                    :     0.000030s :  0.19%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000009s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000432s :  2.73%
    symbol_engine_optimizer.build        :     0.000002s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000024s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008734s : 55.09%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000139    23
 0.98% :     0.000001s :      2: substitution.elim_not_effective
 0.78% :     0.000001s :      2: substitution.fold_const_symbol
 2.48% :     0.000003s :      3: substitution.graph_param_transform
84.11% :     0.000117s :      4: substitution.inline
 1.80% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.76% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.90% :     0.000003s :      2: substitution.replace_old_param
 5.20% :     0.000007s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003814     2
81.74% :     0.003117s :      1: type_inference.infer
18.26% :     0.000696s :      1: type_inference.specialize
------[replace.]   0.000057     6
66.39% :     0.000038s :      4: replace.inline
33.61% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000121     6
94.85% :     0.000115s :      4: match.inline
 5.15% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000105   691
 1.34% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000001s :     10: predicate.addn_zero_filter
 1.91% :     0.000002s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.18% :     0.000001s :     10: predicate.compare_switch_simplify
 1.49% :     0.000002s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.32% :     0.000006s :     22: predicate.inline
 1.02% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.86% :     0.000002s :     15: predicate.load_eliminater
 1.49% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.24% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000001s :     10: predicate.merge_addn
 1.16% :     0.000001s :     10: predicate.minmaximum_grad
 1.52% :     0.000002s :      3: predicate.mutable_eliminate
 0.60% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000002s :     15: predicate.partial_eliminate
 1.26% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.82% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000001s :     10: predicate.reshape_eliminate
 1.31% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.44% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.14% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 2.44% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.46% :     0.000008s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.28% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.45% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.79% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000464    12
48.78% :     0.000227s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.43% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.78% :     0.000213s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020488    76
 0.16% :     0.000033s :      1: add_recomputation
 0.24% :     0.000049s :      1: auto_monad
 0.06% :     0.000011s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.27% :     0.000260s :      1: bootstrap
 0.09% :     0.000019s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.10% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.09% :     0.000018s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000003s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
21.38% :     0.004380s :      1: jit_opt_a
 0.63% :     0.000129s :      1: jit_opt_after_cconv
 0.23% :     0.000048s :      1: jit_opt_b
 1.99% :     0.000408s :      1: loop_unroll
 2.26% :     0.000462s :      1: mutable_eliminate
 3.49% :     0.000715s :     26: opt.transform.jit_opt_a
 0.20% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000028s :      4: opt.transform.symbol_engine_opt
 2.14% :     0.000439s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000030s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000009s :      1: remove_dup_value
 1.07% :     0.000219s :      1: renormalize.infer
 1.04% :     0.000213s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000030s :      1: rewriter_after_opt_a
 0.33% :     0.000068s :      1: rewriter_before_opt_a
 0.33% :     0.000068s :      1: symbol_engine_optimizer
42.68% :     0.008744s :      1: task_emit
18.87% :     0.003867s :      1: type_inference
 0.19% :     0.000039s :      1: validate


TotalTime = 0.0269481, [33]
[bootstrap]: 0.00038415
[type_inference]: 0.00542156
[event_method]: 1.941e-05
[auto_monad]: 7.035e-05
[graph_reusing]: 6.36e-06
[pre_auto_parallel]: 2.73998e-06
[py_interpret_to_execute]: 4.286e-05
[rewriter_before_opt_a]: 7.513e-05
[expand_dump_flag]: 3.55e-06
[jit_opt_a]: 0.00810693, [2]
    [Cycle 1]: 0.00221199, [27]
        [switch_simplify]: 7.779e-05
        [loop_unroll]: 3.198e-05
        [a_1]: 0.00064823
        [with_stream_mark]: 2.476e-05
        [recompute_prepare]: 1.199e-05
        [updatestate_depend_eliminate]: 4.80001e-06
        [updatestate_assign_eliminate]: 3.19001e-06
        [updatestate_loads_eliminate]: 3.04001e-06
        [parameter_eliminate]: 2.51e-06
        [specialize_transform]: 7.14001e-06
        [updatestate_useless_node_eliminater]: 5.94e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 3.06001e-06
        [get_grad_eliminate_]: 6.76e-06
        [merge_forward]: 4.48999e-06
        [cell_reuse_recompute_pass]: 2.07001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.755e-05
        [j_node_and_user_rematch]: 1.24e-05
        [meta_fg_expand]: 3.5e-06
        [replace_old_param]: 1.183e-05
        [inline_without_move]: 5.84999e-06
        [renormalize]: 0.00101205
        [add_forward_monad_depend]: 8.17998e-06
        [auto_monad_grad]: 2.53003e-06
        [auto_monad_eliminator]: 1.813e-05
        [cse]: 3.367e-05
        [replace_applicator]: 2.218e-05
    [Cycle 2]: 0.00039394, [27]
        [switch_simplify]: 6.33e-06
        [loop_unroll]: 6.15002e-06
        [a_1]: 0.00010525
        [with_stream_mark]: 1.594e-05
        [recompute_prepare]: 5.62999e-06
        [updatestate_depend_eliminate]: 4.08001e-06
        [updatestate_assign_eliminate]: 2.86e-06
        [updatestate_loads_eliminate]: 2.26998e-06
        [parameter_eliminate]: 1.53002e-06
        [specialize_transform]: 5.29e-06
        [updatestate_useless_node_eliminater]: 5.15999e-06
        [accelerated_algorithm]: 6.39999e-06
        [meta_shard_fg_expand]: 1.90001e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 4.15e-06
        [cell_reuse_recompute_pass]: 3.05002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.881e-05
        [j_node_and_user_rematch]: 9.41003e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 9.59999e-06
        [inline_without_move]: 5.36002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.84e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 7.11999e-06
        [cse]: 1.387e-05
        [replace_applicator]: 5.97001e-06
[py_interpret_to_execute_after_opt_a]: 1.53e-05
[rewriter_after_opt_a]: 4.539e-05
[convert_after_rewriter]: 7.90998e-06
[order_py_execute_after_rewriter]: 5.54998e-06
[mutable_eliminate]: 0.00087855
[jit_opt_b]: 6.224e-05, [1]
    [Cycle 1]: 5.303e-05, [2]
        [frontend_op_eliminate]: 2.015e-05
        [inline_after_opt_a]: 1.967e-05
[cconv]: 3.51e-05
[loop_unroll]: 0.00059401
[jit_opt_after_cconv]: 0.00018534, [1]
    [Cycle 1]: 0.00017713, [11]
        [c_1]: 2.614e-05
        [parameter_eliminate]: 5.11997e-06
        [updatestate_depend_eliminate]: 8.52e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 3.06001e-06
        [cse]: 3.45e-05
        [call_graph_tuple_transform]: 2.45e-05
        [tuple_list_get_item_eliminator]: 6.54999e-06
        [none_parameter_eliminate]: 1.89e-06
        [renormalize]: 1.21002e-06
        [switch_simplify]: 5.78002e-06
[remove_dup_value]: 1.711e-05
[partial_unused_args_eliminate]: 2.51e-06
[environ_conv]: 7.17002e-06
[add_recomputation]: 6.059e-05
[cse_after_recomputation]: 2.718e-05, [1]
    [Cycle 1]: 1.932e-05, [1]
        [cse]: 1.16e-05
[auto_monad_reorder]: 1.782e-05
[get_jit_bprop_graph]: 2.33002e-06
[rewriter_after_jit_bprop_graph]: 8.69e-06
[opt_after_jit_grad]: 0.00062656
[symbol_engine_optimizer]: 8.829e-05, [1]
    [Cycle 1]: 7.797e-05, [6]
        [build]: 6.04001e-06
        [elim_shapecalc]: 9.21998e-06
        [elim_not_effective]: 1.609e-05
        [opt_reshape]: 6.61e-06
        [fold_const_symbol]: 1.02e-05
        [renormalize]: 9.29984e-07
[validate]: 4.945e-05
[backend_pass]: 1.34e-06
[task_emit]: 0.00980957
[execute]: 8.97999e-06
Sums
    bootstrap                            :     0.000384s :  1.86%
    type_inference                       :     0.005422s : 26.21%
    event_method                         :     0.000019s :  0.09%
    auto_monad                           :     0.000070s :  0.34%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.21%
    rewriter_before_opt_a                :     0.000075s :  0.36%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000084s :  0.41%
    jit_opt_a.loop_unroll                :     0.000038s :  0.18%
    jit_opt_a.a_1                        :     0.000753s :  3.64%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.06%
    jit_opt_a.merge_forward              :     0.000009s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000022s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.001012s :  4.89%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.12%
    jit_opt_a.cse                        :     0.000048s :  0.23%
    jit_opt_a.replace_applicator         :     0.000028s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.07%
    rewriter_after_opt_a                 :     0.000045s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000879s :  4.25%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.10%
    cconv                                :     0.000035s :  0.17%
    loop_unroll                          :     0.000594s :  2.87%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000035s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000061s :  0.29%
    cse_after_recomputation.cse          :     0.000012s :  0.06%
    auto_monad_reorder                   :     0.000018s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.04%
    opt_after_jit_grad                   :     0.000627s :  3.03%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000049s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.009810s : 47.42%
    execute                              :     0.000009s :  0.04%


Time group info:
------[substitution.]   0.000264    23
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.84% :     0.000002s :      2: substitution.fold_const_symbol
 2.49% :     0.000007s :      3: substitution.graph_param_transform
83.01% :     0.000219s :      4: substitution.inline
 1.91% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.21% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.45% :     0.000006s :      2: substitution.replace_old_param
 6.18% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005335     2
81.23% :     0.004334s :      1: type_inference.infer
18.77% :     0.001001s :      1: type_inference.specialize
------[replace.]   0.000094     6
61.22% :     0.000058s :      4: replace.inline
38.78% :     0.000036s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000230     6
93.57% :     0.000216s :      4: match.inline
 6.43% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000142   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.59% :     0.000004s :      3: predicate.ad_related_special_op_eliminate
 0.87% :     0.000001s :     10: predicate.addn_check_dump
 1.60% :     0.000002s :     10: predicate.addn_zero_filter
 1.95% :     0.000003s :     10: predicate.arithmetic_simplify
 1.10% :     0.000002s :     10: predicate.cast_eliminate
 0.35% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.88% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.27% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.26% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.67% :     0.000001s :      3: predicate.elim_not_effective
 0.80% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.86% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.66% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.65% :     0.000002s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 1.48% :     0.000002s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 5.58% :     0.000008s :     22: predicate.inline
 0.76% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000002s :      6: predicate.less_batch_normalization
 1.32% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.42% :     0.000002s :     15: predicate.load_eliminater
 1.83% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.70% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.35% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.88% :     0.000001s :     10: predicate.merge_addn
 0.89% :     0.000001s :     10: predicate.minmaximum_grad
 1.84% :     0.000003s :      3: predicate.mutable_eliminate
 0.33% :     0.000000s :      3: predicate.opt_reshape
 1.80% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     10: predicate.reduce_eliminate
 1.19% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.70% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.27% :     0.000002s :     10: predicate.reshape_eliminate
 1.08% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.37% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000002s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000002s :      6: predicate.specialize_transform
 1.87% :     0.000003s :     10: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.25% :     0.000003s :     16: predicate.switch_defer_inline
 1.65% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.74% :     0.000011s :     53: predicate.switch_simplify
 1.38% :     0.000002s :     10: predicate.tile_eliminate
 1.32% :     0.000002s :     10: predicate.transpose_eliminate
 1.69% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.95% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.20% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.83% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.36% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.44% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000003s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.54% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000709    12
47.89% :     0.000340s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.09% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.02% :     0.000333s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.029116    76
 0.22% :     0.000065s :      1: add_recomputation
 0.25% :     0.000074s :      1: auto_monad
 0.07% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.39% :     0.000405s :      1: bootstrap
 0.13% :     0.000038s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000030s :      1: cse_after_recomputation
 0.04% :     0.000011s :      1: environ_conv
 0.08% :     0.000024s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
27.86% :     0.008112s :      1: jit_opt_a
 0.65% :     0.000188s :      1: jit_opt_after_cconv
 0.22% :     0.000065s :      1: jit_opt_b
 2.07% :     0.000604s :      1: loop_unroll
 3.06% :     0.000891s :      1: mutable_eliminate
 3.51% :     0.001021s :     26: opt.transform.jit_opt_a
 0.20% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000032s :      4: opt.transform.jit_opt_b
 0.06% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000025s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000039s :      4: opt.transform.symbol_engine_opt
 2.20% :     0.000639s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000046s :      1: py_interpret_to_execute
 0.06% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 1.70% :     0.000494s :      1: renormalize.infer
 1.74% :     0.000506s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000051s :      1: rewriter_after_opt_a
 0.27% :     0.000078s :      1: rewriter_before_opt_a
 0.31% :     0.000091s :      1: symbol_engine_optimizer
33.77% :     0.009831s :      1: task_emit
18.68% :     0.005440s :      1: type_inference
 0.25% :     0.000073s :      1: validate


TotalTime = 0.0201315, [33]
[bootstrap]: 0.00029803
[type_inference]: 0.00462533
[event_method]: 1.795e-05
[auto_monad]: 6.376e-05
[graph_reusing]: 5.40999e-06
[pre_auto_parallel]: 3.01001e-06
[py_interpret_to_execute]: 3.997e-05
[rewriter_before_opt_a]: 7.317e-05
[expand_dump_flag]: 3.13e-06
[jit_opt_a]: 0.00572862, [2]
    [Cycle 1]: 0.00183433, [27]
        [switch_simplify]: 6.206e-05
        [loop_unroll]: 3.118e-05
        [a_1]: 0.00058923
        [with_stream_mark]: 2.345e-05
        [recompute_prepare]: 7.97e-06
        [updatestate_depend_eliminate]: 4.2e-06
        [updatestate_assign_eliminate]: 2.85002e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [parameter_eliminate]: 2.02999e-06
        [specialize_transform]: 6.42001e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 5.40999e-06
        [meta_shard_fg_expand]: 2.33002e-06
        [get_grad_eliminate_]: 5.19e-06
        [merge_forward]: 4.40999e-06
        [cell_reuse_recompute_pass]: 1.78002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.406e-05
        [j_node_and_user_rematch]: 1.002e-05
        [meta_fg_expand]: 2.38002e-06
        [replace_old_param]: 1.003e-05
        [inline_without_move]: 5.34998e-06
        [renormalize]: 0.00077347
        [add_forward_monad_depend]: 6.66e-06
        [auto_monad_grad]: 2.44001e-06
        [auto_monad_eliminator]: 1.635e-05
        [cse]: 3.191e-05
        [replace_applicator]: 1.406e-05
    [Cycle 2]: 0.00036167, [27]
        [switch_simplify]: 6.54999e-06
        [loop_unroll]: 5.15001e-06
        [a_1]: 0.00010547
        [with_stream_mark]: 1.202e-05
        [recompute_prepare]: 5.29003e-06
        [updatestate_depend_eliminate]: 3.16001e-06
        [updatestate_assign_eliminate]: 2.15002e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 5.22999e-06
        [updatestate_useless_node_eliminater]: 5.91998e-06
        [accelerated_algorithm]: 5.92999e-06
        [meta_shard_fg_expand]: 1.52999e-06
        [get_grad_eliminate_]: 5.42001e-06
        [merge_forward]: 3.45e-06
        [cell_reuse_recompute_pass]: 2.69999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.588e-05
        [j_node_and_user_rematch]: 8.50999e-06
        [meta_fg_expand]: 1.63002e-06
        [replace_old_param]: 8.72e-06
        [inline_without_move]: 4.84e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.30999e-06
        [auto_monad_grad]: 1.18001e-06
        [auto_monad_eliminator]: 6.29001e-06
        [cse]: 1.221e-05
        [replace_applicator]: 5.42001e-06
[py_interpret_to_execute_after_opt_a]: 1.215e-05
[rewriter_after_opt_a]: 4.017e-05
[convert_after_rewriter]: 6.91001e-06
[order_py_execute_after_rewriter]: 7.21001e-06
[mutable_eliminate]: 0.00067104
[jit_opt_b]: 5.388e-05, [1]
    [Cycle 1]: 4.609e-05, [2]
        [frontend_op_eliminate]: 1.719e-05
        [inline_after_opt_a]: 1.624e-05
[cconv]: 2.966e-05
[loop_unroll]: 0.00052926
[jit_opt_after_cconv]: 0.00015671, [1]
    [Cycle 1]: 0.00014945, [11]
        [c_1]: 2.097e-05
        [parameter_eliminate]: 4.28001e-06
        [updatestate_depend_eliminate]: 7.08e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [cse]: 2.489e-05
        [call_graph_tuple_transform]: 2.163e-05
        [tuple_list_get_item_eliminator]: 5.96998e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 7.39994e-07
        [switch_simplify]: 6.26e-06
[remove_dup_value]: 1.442e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 5.57001e-06
[add_recomputation]: 5.243e-05
[cse_after_recomputation]: 2.208e-05, [1]
    [Cycle 1]: 1.551e-05, [1]
        [cse]: 9.52999e-06
[auto_monad_reorder]: 1.534e-05
[get_jit_bprop_graph]: 2.08998e-06
[rewriter_after_jit_bprop_graph]: 6.13998e-06
[opt_after_jit_grad]: 0.00050356
[symbol_engine_optimizer]: 7.881e-05, [1]
    [Cycle 1]: 7.15e-05, [6]
        [build]: 5.46002e-06
        [elim_shapecalc]: 7.77e-06
        [elim_not_effective]: 1.263e-05
        [opt_reshape]: 6.34001e-06
        [fold_const_symbol]: 8.86997e-06
        [renormalize]: 5.00004e-07
[validate]: 3.886e-05
[backend_pass]: 9.49978e-07
[task_emit]: 0.00675154
[execute]: 8.72e-06
Sums
    bootstrap                            :     0.000298s :  1.88%
    type_inference                       :     0.004625s : 29.10%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000064s :  0.40%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.25%
    rewriter_before_opt_a                :     0.000073s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000069s :  0.43%
    jit_opt_a.loop_unroll                :     0.000036s :  0.23%
    jit_opt_a.a_1                        :     0.000695s :  4.37%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000774s :  4.87%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.14%
    jit_opt_a.cse                        :     0.000044s :  0.28%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000040s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000007s :  0.05%
    mutable_eliminate                    :     0.000671s :  4.22%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000030s :  0.19%
    loop_unroll                          :     0.000529s :  3.33%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000052s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000504s :  3.17%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006752s : 42.48%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000241    23
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.45% :     0.000006s :      3: substitution.graph_param_transform
84.24% :     0.000203s :      4: substitution.inline
 1.49% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.09% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.40% :     0.000006s :      2: substitution.replace_old_param
 5.89% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004553     2
81.74% :     0.003721s :      1: type_inference.infer
18.26% :     0.000831s :      1: type_inference.specialize
------[replace.]   0.000076     6
68.44% :     0.000052s :      4: replace.inline
31.56% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000213     6
93.93% :     0.000200s :      4: match.inline
 6.07% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.15% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.42% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 2.16% :     0.000003s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 0.97% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.14% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.33% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.48% :     0.000001s :      3: predicate.graph_param_transform
 5.85% :     0.000008s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.69% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.79% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.88% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.80% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 2.84% :     0.000004s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.02% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.31% :     0.000003s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.63% :     0.000001s :      3: predicate.reset_defer_inline
 1.23% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.55% :     0.000003s :     16: predicate.switch_defer_inline
 1.96% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.38% :     0.000011s :     53: predicate.switch_simplify
 1.42% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.66% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.96% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.33% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.43% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.65% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.36% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000593    12
48.33% :     0.000287s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.18% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.48% :     0.000258s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021901    76
 0.26% :     0.000056s :      1: add_recomputation
 0.31% :     0.000067s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.43% :     0.000314s :      1: bootstrap
 0.15% :     0.000033s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.18% :     0.005733s :      1: jit_opt_a
 0.73% :     0.000160s :      1: jit_opt_after_cconv
 0.26% :     0.000057s :      1: jit_opt_b
 2.46% :     0.000540s :      1: loop_unroll
 3.12% :     0.000683s :      1: mutable_eliminate
 4.22% :     0.000924s :     26: opt.transform.jit_opt_a
 0.23% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.34% :     0.000512s :      1: opt_after_jit_grad
 0.04% :     0.000010s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000043s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.89% :     0.000414s :      1: renormalize.infer
 1.61% :     0.000352s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000044s :      1: rewriter_after_opt_a
 0.35% :     0.000076s :      1: rewriter_before_opt_a
 0.37% :     0.000081s :      1: symbol_engine_optimizer
30.91% :     0.006770s :      1: task_emit
21.20% :     0.004643s :      1: type_inference
 0.28% :     0.000060s :      1: validate


TotalTime = 0.0141535, [30]
[bootstrap]: 0.00026511
[type_inference]: 0.00626461
[event_method]: 1.988e-05
[auto_monad]: 7.732e-05
[graph_reusing]: 5.57999e-06
[pre_auto_parallel]: 2.53e-06
[py_interpret_to_execute]: 5.072e-05
[rewriter_before_opt_a]: 8.268e-05
[expand_dump_flag]: 2.76e-06
[jit_opt_a]: 0.00521325, [2]
    [Cycle 1]: 0.00180095, [27]
        [switch_simplify]: 0.0001016
        [loop_unroll]: 3.444e-05
        [a_1]: 0.00068329
        [with_stream_mark]: 1.395e-05
        [recompute_prepare]: 8.2e-06
        [updatestate_depend_eliminate]: 3.06999e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 6.47001e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 5.59998e-06
        [meta_shard_fg_expand]: 2.46e-06
        [get_grad_eliminate_]: 5.08002e-06
        [merge_forward]: 2.60002e-06
        [cell_reuse_recompute_pass]: 8.89995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.739e-05
        [j_node_and_user_rematch]: 9.66998e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 9.04998e-06
        [inline_without_move]: 5.54e-06
        [renormalize]: 0.00065321
        [add_forward_monad_depend]: 5.11002e-06
        [auto_monad_grad]: 1.91998e-06
        [auto_monad_eliminator]: 1.148e-05
        [cse]: 1.808e-05
        [replace_applicator]: 1.25e-05
    [Cycle 2]: 0.00033394, [27]
        [switch_simplify]: 5.96998e-06
        [loop_unroll]: 5.19998e-06
        [a_1]: 9.267e-05
        [with_stream_mark]: 9.62001e-06
        [recompute_prepare]: 5.10999e-06
        [updatestate_depend_eliminate]: 2.55002e-06
        [updatestate_assign_eliminate]: 2.01998e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.66e-06
        [specialize_transform]: 5.01002e-06
        [updatestate_useless_node_eliminater]: 1.342e-05
        [accelerated_algorithm]: 4.90001e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.71999e-06
        [cell_reuse_recompute_pass]: 1.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.298e-05
        [j_node_and_user_rematch]: 8.13999e-06
        [meta_fg_expand]: 1.86998e-06
        [replace_old_param]: 7.25e-06
        [inline_without_move]: 4.82e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.52001e-06
        [auto_monad_grad]: 9.09989e-07
        [auto_monad_eliminator]: 5.17999e-06
        [cse]: 1.101e-05
        [replace_applicator]: 5.12e-06
[py_interpret_to_execute_after_opt_a]: 8.97e-06
[rewriter_after_opt_a]: 3.049e-05
[convert_after_rewriter]: 6.17999e-06
[order_py_execute_after_rewriter]: 5.52001e-06
[mutable_eliminate]: 0.00055657
[jit_opt_b]: 5.053e-05, [1]
    [Cycle 1]: 4.404e-05, [2]
        [frontend_op_eliminate]: 1.732e-05
        [inline_after_opt_a]: 1.54e-05
[cconv]: 2.041e-05
[loop_unroll]: 0.00042982
[jit_opt_after_cconv]: 0.00015162, [1]
    [Cycle 1]: 0.00014489, [11]
        [c_1]: 2.123e-05
        [parameter_eliminate]: 2.54001e-06
        [updatestate_depend_eliminate]: 4.95001e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 1.344e-05
        [cse]: 1.736e-05
        [call_graph_tuple_transform]: 1.841e-05
        [tuple_list_get_item_eliminator]: 5.45001e-06
        [none_parameter_eliminate]: 1.37999e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 5.54e-06
[remove_dup_value]: 1.014e-05
[partial_unused_args_eliminate]: 1.91e-06
[environ_conv]: 4.80001e-06
[add_recomputation]: 3.911e-05
[cse_after_recomputation]: 2.092e-05, [1]
    [Cycle 1]: 1.532e-05, [1]
        [cse]: 9.39998e-06
[auto_monad_reorder]: 1.278e-05
[get_jit_bprop_graph]: 1.69e-06
[rewriter_after_jit_bprop_graph]: 5.01002e-06
[opt_after_jit_grad]: 0.00054275
[symbol_engine_optimizer]: 7.536e-05, [1]
    [Cycle 1]: 6.919e-05, [6]
        [build]: 3.93001e-06
        [elim_shapecalc]: 7.78999e-06
        [elim_not_effective]: 1.353e-05
        [opt_reshape]: 5.64e-06
        [fold_const_symbol]: 8.70999e-06
        [renormalize]: 4.90021e-07
[validate]: 3.198e-05
Sums
    bootstrap                            :     0.000265s :  2.53%
    type_inference                       :     0.006265s : 59.67%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000077s :  0.74%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000051s :  0.48%
    rewriter_before_opt_a                :     0.000083s :  0.79%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000108s :  1.02%
    jit_opt_a.loop_unroll                :     0.000040s :  0.38%
    jit_opt_a.a_1                        :     0.000776s :  7.39%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000019s :  0.18%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000653s :  6.22%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.16%
    jit_opt_a.cse                        :     0.000029s :  0.28%
    jit_opt_a.replace_applicator         :     0.000018s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.09%
    rewriter_after_opt_a                 :     0.000030s :  0.29%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000557s :  5.30%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.15%
    cconv                                :     0.000020s :  0.19%
    loop_unroll                          :     0.000430s :  4.09%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000013s :  0.13%
    jit_opt_after_cconv.cse              :     0.000017s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000010s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000039s :  0.37%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000013s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000543s :  5.17%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.30%


Time group info:
------[substitution.]   0.000264    26
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 1.86% :     0.000005s :      3: substitution.graph_param_transform
83.66% :     0.000221s :      6: substitution.inline
 1.21% :     0.000003s :      4: substitution.j_node_and_user_rematch
 3.43% :     0.000009s :      4: substitution.remove_not_recompute_node
 1.27% :     0.000003s :      2: substitution.replace_old_param
 3.16% :     0.000008s :      1: substitution.switch_simplify
 4.16% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006210     2
81.75% :     0.005077s :      1: type_inference.infer
18.25% :     0.001133s :      1: type_inference.specialize
------[replace.]   0.000130     9
54.54% :     0.000071s :      6: replace.inline
21.31% :     0.000028s :      1: replace.switch_simplify
24.15% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000235     9
92.68% :     0.000217s :      6: match.inline
 3.15% :     0.000007s :      1: match.switch_simplify
 4.17% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   801
 1.38% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.46% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     12: predicate.addn_check_dump
 1.35% :     0.000002s :     12: predicate.addn_zero_filter
 2.13% :     0.000003s :     12: predicate.arithmetic_simplify
 1.32% :     0.000002s :     12: predicate.cast_eliminate
 0.38% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000002s :     12: predicate.compare_switch_simplify
 1.39% :     0.000002s :     12: predicate.depend_value_elim
 1.16% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     12: predicate.environ_get_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 4.97% :     0.000007s :     26: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.02% :     0.000001s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     17: predicate.load_eliminater
 1.28% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.49% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.00% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     12: predicate.merge_addn
 1.16% :     0.000002s :     12: predicate.minmaximum_grad
 1.71% :     0.000002s :      3: predicate.mutable_eliminate
 0.39% :     0.000001s :      3: predicate.opt_reshape
 2.00% :     0.000003s :     17: predicate.partial_eliminate
 1.31% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     12: predicate.reduce_eliminate
 1.56% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     20: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     12: predicate.reshape_eliminate
 1.29% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.23% :     0.000002s :      6: predicate.special_op_eliminate
 0.83% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.81% :     0.000004s :     20: predicate.switch_defer_inline
 2.39% :     0.000003s :     20: predicate.switch_layer_defer_inline
 7.85% :     0.000011s :     65: predicate.switch_simplify
 1.26% :     0.000002s :     12: predicate.tile_eliminate
 1.23% :     0.000002s :     12: predicate.transpose_eliminate
 1.42% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.41% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.42% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000788    15
46.13% :     0.000364s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.51% :     0.000091s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.36% :     0.000334s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015926    72
 0.26% :     0.000042s :      1: add_recomputation
 0.51% :     0.000081s :      1: auto_monad
 0.10% :     0.000016s :      1: auto_monad_reorder
 1.75% :     0.000279s :      1: bootstrap
 0.15% :     0.000023s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
32.75% :     0.005216s :      1: jit_opt_a
 0.97% :     0.000155s :      1: jit_opt_after_cconv
 0.34% :     0.000053s :      1: jit_opt_b
 2.75% :     0.000438s :      1: loop_unroll
 3.55% :     0.000565s :      1: mutable_eliminate
 6.62% :     0.001054s :     26: opt.transform.jit_opt_a
 0.30% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.47% :     0.000552s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.34% :     0.000053s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000013s :      1: remove_dup_value
 2.18% :     0.000347s :      1: renormalize.infer
 1.88% :     0.000299s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000033s :      1: rewriter_after_opt_a
 0.54% :     0.000086s :      1: rewriter_before_opt_a
 0.49% :     0.000078s :      1: symbol_engine_optimizer
39.41% :     0.006276s :      1: type_inference


TotalTime = 0.0241013, [33]
[bootstrap]: 0.00034161
[type_inference]: 0.00606842
[event_method]: 1.903e-05
[auto_monad]: 6.948e-05
[graph_reusing]: 5.43002e-06
[pre_auto_parallel]: 2.56e-06
[py_interpret_to_execute]: 4.909e-05
[rewriter_before_opt_a]: 7.48e-05
[expand_dump_flag]: 2.88998e-06
[jit_opt_a]: 0.00626112, [2]
    [Cycle 1]: 0.00174485, [27]
        [switch_simplify]: 5.861e-05
        [loop_unroll]: 2.893e-05
        [a_1]: 0.00055447
        [with_stream_mark]: 1.226e-05
        [recompute_prepare]: 6.94999e-06
        [updatestate_depend_eliminate]: 2.81e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.69e-06
        [parameter_eliminate]: 7.7e-07
        [specialize_transform]: 5.68002e-06
        [updatestate_useless_node_eliminater]: 4.97999e-06
        [accelerated_algorithm]: 5.62999e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 5.77999e-06
        [merge_forward]: 2.25002e-06
        [cell_reuse_recompute_pass]: 7.50006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.109e-05
        [j_node_and_user_rematch]: 9.15999e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 8.27998e-06
        [inline_without_move]: 5.68002e-06
        [renormalize]: 0.00079839
        [add_forward_monad_depend]: 4.99e-06
        [auto_monad_grad]: 1.78002e-06
        [auto_monad_eliminator]: 1.051e-05
        [cse]: 1.971e-05
        [replace_applicator]: 1.164e-05
    [Cycle 2]: 0.00032148, [27]
        [switch_simplify]: 5.76e-06
        [loop_unroll]: 4.82e-06
        [a_1]: 9.176e-05
        [with_stream_mark]: 9.67001e-06
        [recompute_prepare]: 5.50001e-06
        [updatestate_depend_eliminate]: 2.68e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 4.94e-06
        [updatestate_useless_node_eliminater]: 4.56002e-06
        [accelerated_algorithm]: 4.97999e-06
        [meta_shard_fg_expand]: 1.02e-06
        [get_grad_eliminate_]: 4.48001e-06
        [merge_forward]: 2.47001e-06
        [cell_reuse_recompute_pass]: 2.36e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.238e-05
        [j_node_and_user_rematch]: 7.66999e-06
        [meta_fg_expand]: 1.69998e-06
        [replace_old_param]: 6.83e-06
        [inline_without_move]: 4.53001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 1.00001e-06
        [auto_monad_eliminator]: 5.85002e-06
        [cse]: 1.045e-05
        [replace_applicator]: 4.86002e-06
[py_interpret_to_execute_after_opt_a]: 8.3e-06
[rewriter_after_opt_a]: 2.976e-05
[convert_after_rewriter]: 6.24999e-06
[order_py_execute_after_rewriter]: 4.26001e-06
[mutable_eliminate]: 0.00063868
[jit_opt_b]: 4.738e-05, [1]
    [Cycle 1]: 4.129e-05, [2]
        [frontend_op_eliminate]: 1.593e-05
        [inline_after_opt_a]: 1.427e-05
[cconv]: 1.764e-05
[loop_unroll]: 0.00043132
[jit_opt_after_cconv]: 0.00013399, [1]
    [Cycle 1]: 0.00012789, [11]
        [c_1]: 2.043e-05
        [parameter_eliminate]: 2.79999e-06
        [updatestate_depend_eliminate]: 4.57e-06
        [updatestate_assign_eliminate]: 2.19999e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.679e-05
        [call_graph_tuple_transform]: 1.734e-05
        [tuple_list_get_item_eliminator]: 5.20999e-06
        [none_parameter_eliminate]: 1.41998e-06
        [renormalize]: 8.99978e-07
        [switch_simplify]: 5.32999e-06
[remove_dup_value]: 8.2e-06
[partial_unused_args_eliminate]: 1.97999e-06
[environ_conv]: 4.15999e-06
[add_recomputation]: 3.417e-05
[cse_after_recomputation]: 2.002e-05, [1]
    [Cycle 1]: 1.474e-05, [1]
        [cse]: 8.57998e-06
[auto_monad_reorder]: 1.029e-05
[get_jit_bprop_graph]: 1.41002e-06
[rewriter_after_jit_bprop_graph]: 5.34998e-06
[opt_after_jit_grad]: 0.00325284
[symbol_engine_optimizer]: 9.59e-05, [1]
    [Cycle 1]: 8.517e-05, [6]
        [build]: 5.11002e-06
        [elim_shapecalc]: 9.27001e-06
        [elim_not_effective]: 1.931e-05
        [opt_reshape]: 7.30998e-06
        [fold_const_symbol]: 9.85002e-06
        [renormalize]: 1.64998e-06
[validate]: 4.783e-05
[backend_pass]: 1.74e-06
[task_emit]: 0.00617641
[execute]: 8.90001e-06
Sums
    bootstrap                            :     0.000342s :  1.77%
    type_inference                       :     0.006068s : 31.48%
    event_method                         :     0.000019s :  0.10%
    auto_monad                           :     0.000069s :  0.36%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000049s :  0.25%
    rewriter_before_opt_a                :     0.000075s :  0.39%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000064s :  0.33%
    jit_opt_a.loop_unroll                :     0.000034s :  0.18%
    jit_opt_a.a_1                        :     0.000646s :  3.35%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000798s :  4.14%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.08%
    jit_opt_a.cse                        :     0.000030s :  0.16%
    jit_opt_a.replace_applicator         :     0.000017s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.04%
    rewriter_after_opt_a                 :     0.000030s :  0.15%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000639s :  3.31%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.07%
    cconv                                :     0.000018s :  0.09%
    loop_unroll                          :     0.000431s :  2.24%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000008s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000034s :  0.18%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000010s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.003253s : 16.87%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000002s :  0.01%
    validate                             :     0.000048s :  0.25%
    backend_pass                         :     0.000002s :  0.01%
    task_emit                            :     0.006176s : 32.04%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000215    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.87% :     0.000002s :      2: substitution.fold_const_symbol
 2.10% :     0.000005s :      3: substitution.graph_param_transform
85.72% :     0.000184s :      4: substitution.inline
 1.35% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.98% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.49% :     0.000003s :      2: substitution.replace_old_param
 5.43% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005983     2
69.01% :     0.004129s :      1: type_inference.infer
30.99% :     0.001854s :      1: type_inference.specialize
------[replace.]   0.000074     6
67.33% :     0.000050s :      4: replace.inline
32.67% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     6
94.47% :     0.000182s :      4: match.inline
 5.53% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.54% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.74% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000001s :     10: predicate.addn_zero_filter
 2.30% :     0.000003s :     10: predicate.arithmetic_simplify
 1.41% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.45% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.50% :     0.000001s :      3: predicate.elim_not_effective
 1.04% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 4.80% :     0.000006s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.31% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.80% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.18% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.70% :     0.000001s :      3: predicate.opt_reshape
 2.02% :     0.000002s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.52% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.43% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.17% :     0.000008s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.87% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.06% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.66% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000002s :     10: predicate.value_based_eliminate
 0.63% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000659    12
46.16% :     0.000304s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.94% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.90% :     0.000322s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025883    76
 0.14% :     0.000037s :      1: add_recomputation
 0.28% :     0.000073s :      1: auto_monad
 0.05% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 1.37% :     0.000355s :      1: bootstrap
 0.08% :     0.000020s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.09% :     0.000025s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
24.20% :     0.006265s :      1: jit_opt_a
 0.53% :     0.000137s :      1: jit_opt_after_cconv
 0.19% :     0.000050s :      1: jit_opt_b
 1.69% :     0.000438s :      1: loop_unroll
 2.50% :     0.000648s :      1: mutable_eliminate
 3.30% :     0.000855s :     26: opt.transform.jit_opt_a
 0.17% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000041s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000042s :      4: opt.transform.symbol_engine_opt
12.64% :     0.003271s :      1: opt_after_jit_grad
 0.02% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000052s :      1: py_interpret_to_execute
 0.04% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000011s :      1: remove_dup_value
 1.69% :     0.000439s :      1: renormalize.infer
 1.36% :     0.000352s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000033s :      1: rewriter_after_opt_a
 0.30% :     0.000078s :      1: rewriter_before_opt_a
 0.38% :     0.000099s :      1: symbol_engine_optimizer
23.92% :     0.006192s :      1: task_emit
23.51% :     0.006086s :      1: type_inference
 0.28% :     0.000073s :      1: validate


TotalTime = 0.0171276, [33]
[bootstrap]: 0.00026679
[type_inference]: 0.00438216
[event_method]: 1.602e-05
[auto_monad]: 6.025e-05
[graph_reusing]: 5.64998e-06
[pre_auto_parallel]: 2.43e-06
[py_interpret_to_execute]: 2.873e-05
[rewriter_before_opt_a]: 7.007e-05
[expand_dump_flag]: 2.96999e-06
[jit_opt_a]: 0.00478341, [2]
    [Cycle 1]: 0.00139836, [27]
        [switch_simplify]: 4.879e-05
        [loop_unroll]: 2.797e-05
        [a_1]: 0.00048882
        [with_stream_mark]: 1.52e-05
        [recompute_prepare]: 7.18998e-06
        [updatestate_depend_eliminate]: 3.55e-06
        [updatestate_assign_eliminate]: 3.5e-06
        [updatestate_loads_eliminate]: 2.47001e-06
        [parameter_eliminate]: 2.71999e-06
        [specialize_transform]: 5.65001e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.25001e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 5.02e-06
        [merge_forward]: 3.98999e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.323e-05
        [j_node_and_user_rematch]: 8.87999e-06
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 8.17e-06
        [inline_without_move]: 5.10999e-06
        [renormalize]: 0.0005156
        [add_forward_monad_depend]: 4.68999e-06
        [auto_monad_grad]: 1.50999e-06
        [auto_monad_eliminator]: 1.215e-05
        [cse]: 2.831e-05
        [replace_applicator]: 1.025e-05
    [Cycle 2]: 0.0003036, [27]
        [switch_simplify]: 5.20001e-06
        [loop_unroll]: 4.65001e-06
        [a_1]: 8.684e-05
        [with_stream_mark]: 8.47e-06
        [recompute_prepare]: 4.89e-06
        [updatestate_depend_eliminate]: 2.49999e-06
        [updatestate_assign_eliminate]: 1.93002e-06
        [updatestate_loads_eliminate]: 1.91998e-06
        [parameter_eliminate]: 9.99979e-07
        [specialize_transform]: 4.55999e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 4.89998e-06
        [meta_shard_fg_expand]: 1.05001e-06
        [get_grad_eliminate_]: 4.35999e-06
        [merge_forward]: 2.25002e-06
        [cell_reuse_recompute_pass]: 1.66998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.165e-05
        [j_node_and_user_rematch]: 7.43e-06
        [meta_fg_expand]: 1.38002e-06
        [replace_old_param]: 7.13e-06
        [inline_without_move]: 4.46002e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.02998e-06
        [auto_monad_grad]: 6.19999e-07
        [auto_monad_eliminator]: 4.13999e-06
        [cse]: 8.69e-06
        [replace_applicator]: 4.75999e-06
[py_interpret_to_execute_after_opt_a]: 7.69002e-06
[rewriter_after_opt_a]: 3.275e-05
[convert_after_rewriter]: 6.45997e-06
[order_py_execute_after_rewriter]: 5.25001e-06
[mutable_eliminate]: 0.00046971
[jit_opt_b]: 4.671e-05, [1]
    [Cycle 1]: 4.019e-05, [2]
        [frontend_op_eliminate]: 1.488e-05
        [inline_after_opt_a]: 1.427e-05
[cconv]: 2.252e-05
[loop_unroll]: 0.00039742
[jit_opt_after_cconv]: 0.00013138, [1]
    [Cycle 1]: 0.00012539, [11]
        [c_1]: 1.996e-05
        [parameter_eliminate]: 2.39001e-06
        [updatestate_depend_eliminate]: 5.14e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 1.511e-05
        [call_graph_tuple_transform]: 1.71e-05
        [tuple_list_get_item_eliminator]: 5.10001e-06
        [none_parameter_eliminate]: 1.94e-06
        [renormalize]: 5.00004e-07
        [switch_simplify]: 5.20999e-06
[remove_dup_value]: 1.227e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 4.88001e-06
[add_recomputation]: 4.444e-05
[cse_after_recomputation]: 1.867e-05, [1]
    [Cycle 1]: 1.363e-05, [1]
        [cse]: 8.25e-06
[auto_monad_reorder]: 1.366e-05
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 4.70999e-06
[opt_after_jit_grad]: 0.00048459
[symbol_engine_optimizer]: 7.099e-05, [1]
    [Cycle 1]: 6.484e-05, [6]
        [build]: 3.55e-06
        [elim_shapecalc]: 7.8e-06
        [elim_not_effective]: 1.188e-05
        [opt_reshape]: 5.43002e-06
        [fold_const_symbol]: 8.15999e-06
        [renormalize]: 3.60014e-07
[validate]: 3.083e-05
[backend_pass]: 8.30012e-07
[task_emit]: 0.00551874
[execute]: 7.50998e-06
Sums
    bootstrap                            :     0.000267s :  1.98%
    type_inference                       :     0.004382s : 32.50%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000060s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.21%
    rewriter_before_opt_a                :     0.000070s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.24%
    jit_opt_a.a_1                        :     0.000576s :  4.27%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000516s :  3.82%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.12%
    jit_opt_a.cse                        :     0.000037s :  0.27%
    jit_opt_a.replace_applicator         :     0.000015s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.24%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000470s :  3.48%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000397s :  2.95%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000044s :  0.33%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000485s :  3.59%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005519s : 40.92%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000175    23
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.94% :     0.000005s :      3: substitution.graph_param_transform
82.38% :     0.000144s :      4: substitution.inline
 1.65% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.43% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.19% :     0.000004s :      2: substitution.replace_old_param
 6.64% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004320     2
82.28% :     0.003554s :      1: type_inference.infer
17.72% :     0.000766s :      1: type_inference.specialize
------[replace.]   0.000059     6
67.22% :     0.000040s :      4: replace.inline
32.78% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000152     6
93.08% :     0.000141s :      4: match.inline
 6.92% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.33% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.20% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000001s :     10: predicate.addn_zero_filter
 1.81% :     0.000002s :     10: predicate.arithmetic_simplify
 1.35% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.49% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.30% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.35% :     0.000000s :      3: predicate.graph_param_transform
 5.40% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.14% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.49% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000000s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.35% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.66% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.49% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.65% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 2.26% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.99% :     0.000009s :     53: predicate.switch_simplify
 1.28% :     0.000001s :     10: predicate.tile_eliminate
 1.31% :     0.000001s :     10: predicate.transpose_eliminate
 1.53% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.85% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.34% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.30% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000521    12
51.38% :     0.000268s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.19% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.42% :     0.000226s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018512    76
 0.26% :     0.000047s :      1: add_recomputation
 0.34% :     0.000063s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.53% :     0.000283s :      1: bootstrap
 0.14% :     0.000025s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.85% :     0.004786s :      1: jit_opt_a
 0.72% :     0.000134s :      1: jit_opt_after_cconv
 0.27% :     0.000049s :      1: jit_opt_b
 2.18% :     0.000404s :      1: loop_unroll
 2.57% :     0.000477s :      1: mutable_eliminate
 4.16% :     0.000770s :     26: opt.transform.jit_opt_a
 0.24% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.66% :     0.000492s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000031s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.43% :     0.000266s :      1: renormalize.infer
 1.31% :     0.000243s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000035s :      1: rewriter_after_opt_a
 0.39% :     0.000073s :      1: rewriter_before_opt_a
 0.40% :     0.000074s :      1: symbol_engine_optimizer
29.86% :     0.005528s :      1: task_emit
23.74% :     0.004394s :      1: type_inference
 0.26% :     0.000049s :      1: validate


TotalTime = 0.0187557, [33]
[bootstrap]: 0.00029158
[type_inference]: 0.0043427
[event_method]: 1.786e-05
[auto_monad]: 6.197e-05
[graph_reusing]: 5.61998e-06
[pre_auto_parallel]: 2.54001e-06
[py_interpret_to_execute]: 3.262e-05
[rewriter_before_opt_a]: 0.0001356
[expand_dump_flag]: 3.33e-06
[jit_opt_a]: 0.00542524, [2]
    [Cycle 1]: 0.00178586, [27]
        [switch_simplify]: 5.22e-05
        [loop_unroll]: 2.984e-05
        [a_1]: 0.00077791
        [with_stream_mark]: 2.073e-05
        [recompute_prepare]: 8.54002e-06
        [updatestate_depend_eliminate]: 3.78001e-06
        [updatestate_assign_eliminate]: 2.75002e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 6.01e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 2.17999e-06
        [get_grad_eliminate_]: 5.00999e-06
        [merge_forward]: 3.97e-06
        [cell_reuse_recompute_pass]: 1.89e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.577e-05
        [j_node_and_user_rematch]: 8.74e-06
        [meta_fg_expand]: 2.24999e-06
        [replace_old_param]: 9.26002e-06
        [inline_without_move]: 5.24e-06
        [renormalize]: 0.00057067
        [add_forward_monad_depend]: 4.74002e-06
        [auto_monad_grad]: 2.32999e-06
        [auto_monad_eliminator]: 1.375e-05
        [cse]: 2.754e-05
        [replace_applicator]: 1.265e-05
    [Cycle 2]: 0.00032734, [27]
        [switch_simplify]: 6.60002e-06
        [loop_unroll]: 5.18002e-06
        [a_1]: 9.289e-05
        [with_stream_mark]: 8.65999e-06
        [recompute_prepare]: 5.68002e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.18001e-06
        [specialize_transform]: 5.19e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 5.25001e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.47998e-06
        [merge_forward]: 2.56e-06
        [cell_reuse_recompute_pass]: 1.72999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.119e-05
        [j_node_and_user_rematch]: 7.63001e-06
        [meta_fg_expand]: 1.66998e-06
        [replace_old_param]: 7.14001e-06
        [inline_without_move]: 4.84e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.34998e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 5.62999e-06
        [cse]: 9.94999e-06
        [replace_applicator]: 4.97999e-06
[py_interpret_to_execute_after_opt_a]: 9.82001e-06
[rewriter_after_opt_a]: 3.563e-05
[convert_after_rewriter]: 7.7e-06
[order_py_execute_after_rewriter]: 5.77001e-06
[mutable_eliminate]: 0.00053843
[jit_opt_b]: 5.016e-05, [1]
    [Cycle 1]: 4.347e-05, [2]
        [frontend_op_eliminate]: 1.663e-05
        [inline_after_opt_a]: 1.587e-05
[cconv]: 2.448e-05
[loop_unroll]: 0.00042031
[jit_opt_after_cconv]: 0.00013814, [1]
    [Cycle 1]: 0.00013182, [11]
        [c_1]: 2.116e-05
        [parameter_eliminate]: 2.62001e-06
        [updatestate_depend_eliminate]: 5.35999e-06
        [updatestate_assign_eliminate]: 2.63e-06
        [updatestate_loads_eliminate]: 2.34999e-06
        [cse]: 1.616e-05
        [call_graph_tuple_transform]: 1.775e-05
        [tuple_list_get_item_eliminator]: 5.57001e-06
        [none_parameter_eliminate]: 1.44003e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.24998e-06
[remove_dup_value]: 1.341e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 4.94e-06
[add_recomputation]: 4.769e-05
[cse_after_recomputation]: 1.995e-05, [1]
    [Cycle 1]: 1.47e-05, [1]
        [cse]: 8.99e-06
[auto_monad_reorder]: 1.467e-05
[get_jit_bprop_graph]: 1.66998e-06
[rewriter_after_jit_bprop_graph]: 4.72998e-06
[opt_after_jit_grad]: 0.00045967
[symbol_engine_optimizer]: 7.934e-05, [1]
    [Cycle 1]: 7.222e-05, [6]
        [build]: 4.02e-06
        [elim_shapecalc]: 8.44998e-06
        [elim_not_effective]: 1.423e-05
        [opt_reshape]: 5.68002e-06
        [fold_const_symbol]: 8.79e-06
        [renormalize]: 4.89992e-07
[validate]: 3.543e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.00630935
[execute]: 7.98001e-06
Sums
    bootstrap                            :     0.000292s :  1.97%
    type_inference                       :     0.004343s : 29.32%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000062s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.22%
    rewriter_before_opt_a                :     0.000136s :  0.92%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000871s :  5.88%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000571s :  3.85%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000037s :  0.25%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.24%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000538s :  3.63%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000420s :  2.84%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000048s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000460s :  3.10%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006309s : 42.59%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000185    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.67% :     0.000005s :      3: substitution.graph_param_transform
83.12% :     0.000154s :      4: substitution.inline
 1.48% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.29% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.35% :     0.000004s :      2: substitution.replace_old_param
 6.36% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004273     2
80.90% :     0.003457s :      1: type_inference.infer
19.10% :     0.000816s :      1: type_inference.specialize
------[replace.]   0.000064     6
67.58% :     0.000043s :      4: replace.inline
32.42% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     6
93.44% :     0.000152s :      4: match.inline
 6.56% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.23% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 2.23% :     0.000003s :     10: predicate.arithmetic_simplify
 1.37% :     0.000002s :     10: predicate.cast_eliminate
 0.68% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000001s :     10: predicate.depend_value_elim
 1.20% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000000s :      3: predicate.graph_param_transform
 5.85% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000002s :      6: predicate.less_batch_normalization
 1.36% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.23% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.05% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.81% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.84% :     0.000002s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 1.31% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.12% :     0.000003s :     18: predicate.replace_applicator
 0.93% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.24% :     0.000001s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.27% :     0.000002s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.83% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.58% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.65% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.86% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.14% :     0.000003s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000530    12
47.33% :     0.000251s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.39% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.28% :     0.000251s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020505    76
 0.25% :     0.000051s :      1: add_recomputation
 0.32% :     0.000065s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.51% :     0.000309s :      1: bootstrap
 0.13% :     0.000027s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000023s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.48% :     0.005429s :      1: jit_opt_a
 0.69% :     0.000141s :      1: jit_opt_after_cconv
 0.26% :     0.000053s :      1: jit_opt_b
 2.09% :     0.000429s :      1: loop_unroll
 2.67% :     0.000547s :      1: mutable_eliminate
 5.26% :     0.001079s :     26: opt.transform.jit_opt_a
 0.23% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.28% :     0.000468s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000035s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.37% :     0.000282s :      1: renormalize.infer
 1.37% :     0.000281s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000039s :      1: rewriter_after_opt_a
 0.68% :     0.000139s :      1: rewriter_before_opt_a
 0.40% :     0.000082s :      1: symbol_engine_optimizer
30.83% :     0.006322s :      1: task_emit
21.25% :     0.004357s :      1: type_inference
 0.28% :     0.000056s :      1: validate


TotalTime = 0.0174022, [33]
[bootstrap]: 0.00026607
[type_inference]: 0.00417359
[event_method]: 1.673e-05
[auto_monad]: 6.051e-05
[graph_reusing]: 6.02999e-06
[pre_auto_parallel]: 3.16001e-06
[py_interpret_to_execute]: 3.772e-05
[rewriter_before_opt_a]: 7.356e-05
[expand_dump_flag]: 2.91e-06
[jit_opt_a]: 0.00490992, [2]
    [Cycle 1]: 0.00146429, [27]
        [switch_simplify]: 4.97e-05
        [loop_unroll]: 2.86e-05
        [a_1]: 0.00049639
        [with_stream_mark]: 1.55e-05
        [recompute_prepare]: 7.36001e-06
        [updatestate_depend_eliminate]: 3.16999e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.74999e-06
        [parameter_eliminate]: 1.84998e-06
        [specialize_transform]: 5.76e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 5.09998e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 4.85999e-06
        [merge_forward]: 4.03999e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.349e-05
        [j_node_and_user_rematch]: 8.89e-06
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 8.17998e-06
        [inline_without_move]: 5.22e-06
        [renormalize]: 0.00056375
        [add_forward_monad_depend]: 5.57999e-06
        [auto_monad_grad]: 1.90001e-06
        [auto_monad_eliminator]: 1.486e-05
        [cse]: 2.745e-05
        [replace_applicator]: 1.188e-05
    [Cycle 2]: 0.00031135, [27]
        [switch_simplify]: 5.44e-06
        [loop_unroll]: 4.86997e-06
        [a_1]: 8.775e-05
        [with_stream_mark]: 8.85999e-06
        [recompute_prepare]: 4.59002e-06
        [updatestate_depend_eliminate]: 2.53998e-06
        [updatestate_assign_eliminate]: 2.11003e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 4.81002e-06
        [updatestate_useless_node_eliminater]: 4.71002e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 1.04998e-06
        [get_grad_eliminate_]: 4.25999e-06
        [merge_forward]: 2.31e-06
        [cell_reuse_recompute_pass]: 1.74e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.181e-05
        [j_node_and_user_rematch]: 7.6e-06
        [meta_fg_expand]: 1.53002e-06
        [replace_old_param]: 6.67002e-06
        [inline_without_move]: 4.77e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 5.00999e-06
        [cse]: 9.46e-06
        [replace_applicator]: 5.02e-06
[py_interpret_to_execute_after_opt_a]: 1.035e-05
[rewriter_after_opt_a]: 3.454e-05
[convert_after_rewriter]: 6.85998e-06
[order_py_execute_after_rewriter]: 5.12999e-06
[mutable_eliminate]: 0.00054531
[jit_opt_b]: 4.761e-05, [1]
    [Cycle 1]: 4.077e-05, [2]
        [frontend_op_eliminate]: 1.534e-05
        [inline_after_opt_a]: 1.405e-05
[cconv]: 2.398e-05
[loop_unroll]: 0.00041344
[jit_opt_after_cconv]: 0.00013113, [1]
    [Cycle 1]: 0.00012497, [11]
        [c_1]: 1.922e-05
        [parameter_eliminate]: 2.41998e-06
        [updatestate_depend_eliminate]: 5.10001e-06
        [updatestate_assign_eliminate]: 2.13002e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [cse]: 1.626e-05
        [call_graph_tuple_transform]: 1.737e-05
        [tuple_list_get_item_eliminator]: 4.89e-06
        [none_parameter_eliminate]: 1.47999e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 5.32999e-06
[remove_dup_value]: 1.367e-05
[partial_unused_args_eliminate]: 2.04999e-06
[environ_conv]: 4.82e-06
[add_recomputation]: 4.669e-05
[cse_after_recomputation]: 1.909e-05, [1]
    [Cycle 1]: 1.391e-05, [1]
        [cse]: 8.27e-06
[auto_monad_reorder]: 1.435e-05
[get_jit_bprop_graph]: 1.60999e-06
[rewriter_after_jit_bprop_graph]: 4.70999e-06
[opt_after_jit_grad]: 0.00046379
[symbol_engine_optimizer]: 7.222e-05, [1]
    [Cycle 1]: 6.559e-05, [6]
        [build]: 3.9e-06
        [elim_shapecalc]: 8.18999e-06
        [elim_not_effective]: 1.231e-05
        [opt_reshape]: 5.46998e-06
        [fold_const_symbol]: 7.87003e-06
        [renormalize]: 1.60013e-07
[validate]: 3.214e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.005769
[execute]: 7.52002e-06
Sums
    bootstrap                            :     0.000266s :  1.94%
    type_inference                       :     0.004174s : 30.49%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000061s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.28%
    rewriter_before_opt_a                :     0.000074s :  0.54%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.24%
    jit_opt_a.a_1                        :     0.000584s :  4.27%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000564s :  4.12%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.15%
    jit_opt_a.cse                        :     0.000037s :  0.27%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.08%
    rewriter_after_opt_a                 :     0.000035s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000545s :  3.98%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000413s :  3.02%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.34%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000464s :  3.39%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005769s : 42.14%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000182    23
 1.06% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.88% :     0.000005s :      3: substitution.graph_param_transform
82.73% :     0.000151s :      4: substitution.inline
 1.92% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.29% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.99% :     0.000004s :      2: substitution.replace_old_param
 6.44% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004117     2
81.82% :     0.003368s :      1: type_inference.infer
18.18% :     0.000748s :      1: type_inference.specialize
------[replace.]   0.000062     6
68.05% :     0.000042s :      4: replace.inline
31.95% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000158     6
93.29% :     0.000148s :      4: match.inline
 6.71% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.30% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 1.95% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.56% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.76% :     0.000006s :     22: predicate.inline
 1.14% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.04% :     0.000001s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.87% :     0.000002s :     15: predicate.load_eliminater
 1.51% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.21% :     0.000001s :     10: predicate.minmaximum_grad
 1.92% :     0.000002s :      3: predicate.mutable_eliminate
 0.68% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.19% :     0.000009s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.77% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000510    12
50.39% :     0.000257s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.20% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.41% :     0.000227s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018849    76
 0.26% :     0.000050s :      1: add_recomputation
 0.34% :     0.000064s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.50% :     0.000282s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
26.06% :     0.004912s :      1: jit_opt_a
 0.71% :     0.000134s :      1: jit_opt_after_cconv
 0.27% :     0.000050s :      1: jit_opt_b
 2.23% :     0.000421s :      1: loop_unroll
 2.94% :     0.000554s :      1: mutable_eliminate
 4.14% :     0.000781s :     26: opt.transform.jit_opt_a
 0.23% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.50% :     0.000471s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000041s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.58% :     0.000298s :      1: renormalize.infer
 1.37% :     0.000258s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000038s :      1: rewriter_after_opt_a
 0.41% :     0.000076s :      1: rewriter_before_opt_a
 0.40% :     0.000075s :      1: symbol_engine_optimizer
30.67% :     0.005781s :      1: task_emit
22.20% :     0.004185s :      1: type_inference
 0.27% :     0.000051s :      1: validate


TotalTime = 0.0246269, [33]
[bootstrap]: 0.0003642
[type_inference]: 0.0064145
[event_method]: 2.12e-05
[auto_monad]: 7.111e-05
[graph_reusing]: 5.89e-06
[pre_auto_parallel]: 2.61999e-06
[py_interpret_to_execute]: 4.485e-05
[rewriter_before_opt_a]: 7.999e-05
[expand_dump_flag]: 3.43999e-06
[jit_opt_a]: 0.0075328, [2]
    [Cycle 1]: 0.00197696, [27]
        [switch_simplify]: 6.699e-05
        [loop_unroll]: 2.89e-05
        [a_1]: 0.00059636
        [with_stream_mark]: 2.432e-05
        [recompute_prepare]: 8.11002e-06
        [updatestate_depend_eliminate]: 4.15e-06
        [updatestate_assign_eliminate]: 3.26001e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 2.38002e-06
        [specialize_transform]: 6.04001e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 5.27999e-06
        [meta_shard_fg_expand]: 2.17001e-06
        [get_grad_eliminate_]: 5.02e-06
        [merge_forward]: 3.66999e-06
        [cell_reuse_recompute_pass]: 1.28002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.444e-05
        [j_node_and_user_rematch]: 8.85999e-06
        [meta_fg_expand]: 2.43998e-06
        [replace_old_param]: 1.015e-05
        [inline_without_move]: 5.05001e-06
        [renormalize]: 0.000916
        [add_forward_monad_depend]: 7.13e-06
        [auto_monad_grad]: 2.43e-06
        [auto_monad_eliminator]: 1.538e-05
        [cse]: 2.91e-05
        [replace_applicator]: 1.545e-05
    [Cycle 2]: 0.00033639, [27]
        [switch_simplify]: 5.92999e-06
        [loop_unroll]: 5.11002e-06
        [a_1]: 9.243e-05
        [with_stream_mark]: 1.176e-05
        [recompute_prepare]: 5.28002e-06
        [updatestate_depend_eliminate]: 2.94999e-06
        [updatestate_assign_eliminate]: 2.49999e-06
        [updatestate_loads_eliminate]: 2.18002e-06
        [parameter_eliminate]: 1.57001e-06
        [specialize_transform]: 4.78001e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.56e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 4.39998e-06
        [merge_forward]: 2.71e-06
        [cell_reuse_recompute_pass]: 2.06e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.968e-05
        [j_node_and_user_rematch]: 7.98001e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 7.1e-06
        [inline_without_move]: 4.76002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 8.90024e-07
        [auto_monad_eliminator]: 5.37001e-06
        [cse]: 1.189e-05
        [replace_applicator]: 7.52002e-06
[py_interpret_to_execute_after_opt_a]: 1.296e-05
[rewriter_after_opt_a]: 3.919e-05
[convert_after_rewriter]: 7.73999e-06
[order_py_execute_after_rewriter]: 4.79e-06
[mutable_eliminate]: 0.00071968
[jit_opt_b]: 4.992e-05, [1]
    [Cycle 1]: 4.221e-05, [2]
        [frontend_op_eliminate]: 1.674e-05
        [inline_after_opt_a]: 1.457e-05
[cconv]: 2.855e-05
[loop_unroll]: 0.00044422
[jit_opt_after_cconv]: 0.00014415, [1]
    [Cycle 1]: 0.0001379, [11]
        [c_1]: 2.065e-05
        [parameter_eliminate]: 3.01001e-06
        [updatestate_depend_eliminate]: 5.47001e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [cse]: 2.05e-05
        [call_graph_tuple_transform]: 1.952e-05
        [tuple_list_get_item_eliminator]: 5.35999e-06
        [none_parameter_eliminate]: 1.81e-06
        [renormalize]: 9.70002e-07
        [switch_simplify]: 5.80002e-06
[remove_dup_value]: 1.697e-05
[partial_unused_args_eliminate]: 2.44999e-06
[environ_conv]: 5.54998e-06
[add_recomputation]: 5.063e-05
[cse_after_recomputation]: 2.033e-05, [1]
    [Cycle 1]: 1.467e-05, [1]
        [cse]: 8.88002e-06
[auto_monad_reorder]: 1.645e-05
[get_jit_bprop_graph]: 2.07001e-06
[rewriter_after_jit_bprop_graph]: 8.39998e-06
[opt_after_jit_grad]: 0.00047211
[symbol_engine_optimizer]: 7.731e-05, [1]
    [Cycle 1]: 6.976e-05, [6]
        [build]: 4.52998e-06
        [elim_shapecalc]: 8.79e-06
        [elim_not_effective]: 1.304e-05
        [opt_reshape]: 6.76e-06
        [fold_const_symbol]: 8.95001e-06
        [renormalize]: 1.09e-06
[validate]: 4.31e-05
[backend_pass]: 8.80013e-07
[task_emit]: 0.00767994
[execute]: 1.08e-05
Sums
    bootstrap                            :     0.000364s :  1.94%
    type_inference                       :     0.006415s : 34.19%
    event_method                         :     0.000021s :  0.11%
    auto_monad                           :     0.000071s :  0.38%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000045s :  0.24%
    rewriter_before_opt_a                :     0.000080s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000073s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.18%
    jit_opt_a.a_1                        :     0.000689s :  3.67%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000916s :  4.88%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.11%
    jit_opt_a.cse                        :     0.000041s :  0.22%
    jit_opt_a.replace_applicator         :     0.000023s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000039s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000720s :  3.84%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000029s :  0.15%
    loop_unroll                          :     0.000444s :  2.37%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000051s :  0.27%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000472s :  2.52%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.23%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007680s : 40.94%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000257    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000001s :      2: substitution.fold_const_symbol
 2.35% :     0.000006s :      3: substitution.graph_param_transform
84.07% :     0.000216s :      4: substitution.inline
 1.13% :     0.000003s :      4: substitution.j_node_and_user_rematch
 4.05% :     0.000010s :      4: substitution.remove_not_recompute_node
 2.00% :     0.000005s :      2: substitution.replace_old_param
 5.14% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006313     2
84.02% :     0.005304s :      1: type_inference.infer
15.98% :     0.001009s :      1: type_inference.specialize
------[replace.]   0.000073     6
66.36% :     0.000048s :      4: replace.inline
33.64% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000226     6
94.58% :     0.000213s :      4: match.inline
 5.42% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.57% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000001s :     10: predicate.addn_zero_filter
 2.18% :     0.000003s :     10: predicate.arithmetic_simplify
 1.29% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000001s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.94% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.46% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.25% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.88% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.04% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.45% :     0.000002s :      3: predicate.mutable_eliminate
 0.75% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.57% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.21% :     0.000000s :      3: predicate.reset_defer_inline
 1.32% :     0.000002s :     10: predicate.reshape_eliminate
 1.53% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.51% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.66% :     0.000009s :     53: predicate.switch_simplify
 1.29% :     0.000002s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.79% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000755    12
48.85% :     0.000369s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.74% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.41% :     0.000335s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026563    76
 0.20% :     0.000053s :      1: add_recomputation
 0.28% :     0.000075s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.45% :     0.000385s :      1: bootstrap
 0.12% :     0.000031s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000025s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
28.37% :     0.007537s :      1: jit_opt_a
 0.55% :     0.000147s :      1: jit_opt_after_cconv
 0.20% :     0.000053s :      1: jit_opt_b
 1.70% :     0.000452s :      1: loop_unroll
 2.75% :     0.000730s :      1: mutable_eliminate
 3.46% :     0.000920s :     26: opt.transform.jit_opt_a
 0.18% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.80% :     0.000479s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000047s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 1.92% :     0.000510s :      1: renormalize.infer
 1.50% :     0.000398s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000042s :      1: rewriter_after_opt_a
 0.31% :     0.000083s :      1: rewriter_before_opt_a
 0.30% :     0.000080s :      1: symbol_engine_optimizer
28.99% :     0.007702s :      1: task_emit
24.22% :     0.006433s :      1: type_inference
 0.24% :     0.000064s :      1: validate


TotalTime = 0.0162363, [30]
[bootstrap]: 0.00027948
[type_inference]: 0.00695056
[event_method]: 2.333e-05
[auto_monad]: 8.419e-05
[graph_reusing]: 6.77002e-06
[pre_auto_parallel]: 2.81e-06
[py_interpret_to_execute]: 6.289e-05
[rewriter_before_opt_a]: 9.637e-05
[expand_dump_flag]: 3.41999e-06
[jit_opt_a]: 0.00654349, [2]
    [Cycle 1]: 0.00245017, [27]
        [switch_simplify]: 0.00015463
        [loop_unroll]: 3.556e-05
        [a_1]: 0.00095657
        [with_stream_mark]: 2.553e-05
        [recompute_prepare]: 8.82e-06
        [updatestate_depend_eliminate]: 3.95998e-06
        [updatestate_assign_eliminate]: 2.89999e-06
        [updatestate_loads_eliminate]: 2.64999e-06
        [parameter_eliminate]: 2.19001e-06
        [specialize_transform]: 6.83e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 6.43e-06
        [meta_shard_fg_expand]: 3.30998e-06
        [get_grad_eliminate_]: 6.68003e-06
        [merge_forward]: 3.53e-06
        [cell_reuse_recompute_pass]: 1.46998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.648e-05
        [j_node_and_user_rematch]: 1.135e-05
        [meta_fg_expand]: 2.34999e-06
        [replace_old_param]: 1.029e-05
        [inline_without_move]: 5.81e-06
        [renormalize]: 0.00090337
        [add_forward_monad_depend]: 1.265e-05
        [auto_monad_grad]: 2.39001e-06
        [auto_monad_eliminator]: 1.508e-05
        [cse]: 2.762e-05
        [replace_applicator]: 1.299e-05
    [Cycle 2]: 0.00032274, [27]
        [switch_simplify]: 6.37001e-06
        [loop_unroll]: 4.97999e-06
        [a_1]: 9.219e-05
        [with_stream_mark]: 1.03e-05
        [recompute_prepare]: 5.44998e-06
        [updatestate_depend_eliminate]: 2.91e-06
        [updatestate_assign_eliminate]: 2.64001e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 5.12e-06
        [updatestate_useless_node_eliminater]: 4.72e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.21997e-06
        [get_grad_eliminate_]: 4.63999e-06
        [merge_forward]: 2.83e-06
        [cell_reuse_recompute_pass]: 2.06998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.35e-05
        [j_node_and_user_rematch]: 7.99002e-06
        [meta_fg_expand]: 1.48002e-06
        [replace_old_param]: 7.33e-06
        [inline_without_move]: 4.97e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.19003e-06
        [auto_monad_grad]: 6.79982e-07
        [auto_monad_eliminator]: 4.62998e-06
        [cse]: 1.013e-05
        [replace_applicator]: 4.85001e-06
[py_interpret_to_execute_after_opt_a]: 1.053e-05
[rewriter_after_opt_a]: 3.548e-05
[convert_after_rewriter]: 6.82002e-06
[order_py_execute_after_rewriter]: 5.09998e-06
[mutable_eliminate]: 0.00061645
[jit_opt_b]: 5.115e-05, [1]
    [Cycle 1]: 4.337e-05, [2]
        [frontend_op_eliminate]: 1.721e-05
        [inline_after_opt_a]: 1.518e-05
[cconv]: 2.478e-05
[loop_unroll]: 0.00043014
[jit_opt_after_cconv]: 0.00013936, [1]
    [Cycle 1]: 0.00013267, [11]
        [c_1]: 2.152e-05
        [parameter_eliminate]: 2.40002e-06
        [updatestate_depend_eliminate]: 5.30001e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.772e-05
        [call_graph_tuple_transform]: 1.833e-05
        [tuple_list_get_item_eliminator]: 5.17999e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 5.37999e-06
[remove_dup_value]: 1.321e-05
[partial_unused_args_eliminate]: 2.02999e-06
[environ_conv]: 4.94998e-06
[add_recomputation]: 4.546e-05
[cse_after_recomputation]: 2.302e-05, [1]
    [Cycle 1]: 1.647e-05, [1]
        [cse]: 1.012e-05
[auto_monad_reorder]: 1.558e-05
[get_jit_bprop_graph]: 2.24001e-06
[rewriter_after_jit_bprop_graph]: 4.94998e-06
[opt_after_jit_grad]: 0.00044789
[symbol_engine_optimizer]: 8.975e-05, [1]
    [Cycle 1]: 8.393e-05, [6]
        [build]: 3.41999e-06
        [elim_shapecalc]: 2.374e-05
        [elim_not_effective]: 1.33e-05
        [opt_reshape]: 5.76e-06
        [fold_const_symbol]: 8.54e-06
        [renormalize]: 4.30009e-07
[validate]: 3.402e-05
Sums
    bootstrap                            :     0.000279s :  2.36%
    type_inference                       :     0.006951s : 58.67%
    event_method                         :     0.000023s :  0.20%
    auto_monad                           :     0.000084s :  0.71%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000063s :  0.53%
    rewriter_before_opt_a                :     0.000096s :  0.81%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000161s :  1.36%
    jit_opt_a.loop_unroll                :     0.000041s :  0.34%
    jit_opt_a.a_1                        :     0.001049s :  8.85%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.000903s :  7.63%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.12%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.17%
    jit_opt_a.cse                        :     0.000038s :  0.32%
    jit_opt_a.replace_applicator         :     0.000018s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.09%
    rewriter_after_opt_a                 :     0.000035s :  0.30%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000616s :  5.20%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.13%
    cconv                                :     0.000025s :  0.21%
    loop_unroll                          :     0.000430s :  3.63%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000045s :  0.38%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000448s :  3.78%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000024s :  0.20%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.29%


Time group info:
------[substitution.]   0.000464    26
 0.45% :     0.000002s :      2: substitution.elim_not_effective
 0.28% :     0.000001s :      2: substitution.fold_const_symbol
 1.12% :     0.000005s :      3: substitution.graph_param_transform
89.26% :     0.000414s :      6: substitution.inline
 0.95% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.03% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.01% :     0.000005s :      2: substitution.replace_old_param
 2.50% :     0.000012s :      1: substitution.switch_simplify
 3.41% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006880     2
81.23% :     0.005588s :      1: type_inference.infer
18.77% :     0.001291s :      1: type_inference.specialize
------[replace.]   0.000186     9
50.21% :     0.000093s :      6: replace.inline
25.15% :     0.000047s :      1: replace.switch_simplify
24.64% :     0.000046s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000385     9
93.50% :     0.000360s :      6: match.inline
 2.76% :     0.000011s :      1: match.switch_simplify
 3.74% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000143   801
 1.82% :     0.000003s :     12: predicate.accumulaten_eliminater
 1.10% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     12: predicate.addn_check_dump
 1.46% :     0.000002s :     12: predicate.addn_zero_filter
 2.58% :     0.000004s :     12: predicate.arithmetic_simplify
 1.22% :     0.000002s :     12: predicate.cast_eliminate
 0.37% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000002s :     12: predicate.compare_switch_simplify
 1.42% :     0.000002s :     12: predicate.depend_value_elim
 1.32% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.52% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.65% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.24% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.78% :     0.000003s :     12: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     12: predicate.environ_get_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.31% :     0.000002s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 5.77% :     0.000008s :     26: predicate.inline
 0.78% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.40% :     0.000002s :      6: predicate.less_batch_normalization
 1.79% :     0.000003s :     14: predicate.list_to_tuple_eliminator_
 1.56% :     0.000002s :     17: predicate.load_eliminater
 1.17% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.69% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000002s :     12: predicate.merge_addn
 1.09% :     0.000002s :     12: predicate.minmaximum_grad
 1.39% :     0.000002s :      3: predicate.mutable_eliminate
 0.37% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     17: predicate.partial_eliminate
 1.16% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     12: predicate.reduce_eliminate
 1.58% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     20: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     12: predicate.reshape_eliminate
 1.15% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.55% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     12: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.76% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.31% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.63% :     0.000004s :     20: predicate.switch_defer_inline
 2.52% :     0.000004s :     20: predicate.switch_layer_defer_inline
 8.99% :     0.000013s :     65: predicate.switch_simplify
 1.22% :     0.000002s :     12: predicate.tile_eliminate
 1.17% :     0.000002s :     12: predicate.transpose_eliminate
 1.38% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.05% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.75% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.56% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.38% :     0.000002s :     12: predicate.value_based_eliminate
 0.27% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.97% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000959    15
43.27% :     0.000415s :      6: func_graph_cloner_run.FuncGraphClonerGraph
16.87% :     0.000162s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.87% :     0.000382s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018560    72
 0.26% :     0.000049s :      1: add_recomputation
 0.48% :     0.000088s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 1.58% :     0.000293s :      1: bootstrap
 0.15% :     0.000028s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000029s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
35.27% :     0.006547s :      1: jit_opt_a
 0.76% :     0.000142s :      1: jit_opt_after_cconv
 0.29% :     0.000054s :      1: jit_opt_b
 2.36% :     0.000438s :      1: loop_unroll
 3.37% :     0.000625s :      1: mutable_eliminate
 7.41% :     0.001375s :     26: opt.transform.jit_opt_a
 0.25% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.46% :     0.000456s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.36% :     0.000066s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 2.83% :     0.000526s :      1: renormalize.infer
 1.98% :     0.000368s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000038s :      1: rewriter_after_opt_a
 0.54% :     0.000100s :      1: rewriter_before_opt_a
 0.50% :     0.000092s :      1: symbol_engine_optimizer
37.54% :     0.006967s :      1: type_inference


TotalTime = 0.0202558, [33]
[bootstrap]: 0.00029499
[type_inference]: 0.00474544
[event_method]: 1.744e-05
[auto_monad]: 6.278e-05
[graph_reusing]: 5.89e-06
[pre_auto_parallel]: 2.62001e-06
[py_interpret_to_execute]: 3.598e-05
[rewriter_before_opt_a]: 7.801e-05
[expand_dump_flag]: 3.54002e-06
[jit_opt_a]: 0.00642949, [2]
    [Cycle 1]: 0.00265985, [27]
        [switch_simplify]: 5.491e-05
        [loop_unroll]: 4.809e-05
        [a_1]: 0.0007084
        [with_stream_mark]: 3.234e-05
        [recompute_prepare]: 8.48001e-06
        [updatestate_depend_eliminate]: 3.93999e-06
        [updatestate_assign_eliminate]: 3.11001e-06
        [updatestate_loads_eliminate]: 2.69999e-06
        [parameter_eliminate]: 1.98002e-06
        [specialize_transform]: 6.38e-06
        [updatestate_useless_node_eliminater]: 4.94998e-06
        [accelerated_algorithm]: 5.31998e-06
        [meta_shard_fg_expand]: 1.99999e-06
        [get_grad_eliminate_]: 4.99e-06
        [merge_forward]: 4.12e-06
        [cell_reuse_recompute_pass]: 1.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.536e-05
        [j_node_and_user_rematch]: 9.27001e-06
        [meta_fg_expand]: 2.18998e-06
        [replace_old_param]: 8.61002e-06
        [inline_without_move]: 5.25001e-06
        [renormalize]: 0.00146546
        [add_forward_monad_depend]: 6.76e-06
        [auto_monad_grad]: 2.66999e-06
        [auto_monad_eliminator]: 1.547e-05
        [cse]: 2.874e-05
        [replace_applicator]: 1.4e-05
    [Cycle 2]: 0.00034241, [27]
        [switch_simplify]: 6.00002e-06
        [loop_unroll]: 5.37001e-06
        [a_1]: 9.631e-05
        [with_stream_mark]: 1.086e-05
        [recompute_prepare]: 5.42999e-06
        [updatestate_depend_eliminate]: 3.14001e-06
        [updatestate_assign_eliminate]: 2.41998e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.27999e-06
        [specialize_transform]: 5.32999e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.82001e-06
        [get_grad_eliminate_]: 4.62e-06
        [merge_forward]: 3.39001e-06
        [cell_reuse_recompute_pass]: 2.37001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.401e-05
        [j_node_and_user_rematch]: 8.23999e-06
        [meta_fg_expand]: 1.97001e-06
        [replace_old_param]: 1.425e-05
        [inline_without_move]: 5.56998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.49e-06
        [auto_monad_grad]: 1.44998e-06
        [auto_monad_eliminator]: 4.85001e-06
        [cse]: 1.069e-05
        [replace_applicator]: 5.00001e-06
[py_interpret_to_execute_after_opt_a]: 1.328e-05
[rewriter_after_opt_a]: 3.689e-05
[convert_after_rewriter]: 7.73999e-06
[order_py_execute_after_rewriter]: 5.79e-06
[mutable_eliminate]: 0.00065621
[jit_opt_b]: 5.113e-05, [1]
    [Cycle 1]: 4.327e-05, [2]
        [frontend_op_eliminate]: 1.679e-05
        [inline_after_opt_a]: 1.503e-05
[cconv]: 2.705e-05
[loop_unroll]: 0.00042524
[jit_opt_after_cconv]: 0.00014, [1]
    [Cycle 1]: 0.00013334, [11]
        [c_1]: 2.14e-05
        [parameter_eliminate]: 2.71e-06
        [updatestate_depend_eliminate]: 5.49e-06
        [updatestate_assign_eliminate]: 2.60002e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.818e-05
        [call_graph_tuple_transform]: 1.815e-05
        [tuple_list_get_item_eliminator]: 5.52999e-06
        [none_parameter_eliminate]: 1.54998e-06
        [renormalize]: 5.39992e-07
        [switch_simplify]: 5.52999e-06
[remove_dup_value]: 1.416e-05
[partial_unused_args_eliminate]: 2.24999e-06
[environ_conv]: 5.34e-06
[add_recomputation]: 5.026e-05
[cse_after_recomputation]: 2.046e-05, [1]
    [Cycle 1]: 1.51e-05, [1]
        [cse]: 9.29e-06
[auto_monad_reorder]: 1.45e-05
[get_jit_bprop_graph]: 2.10002e-06
[rewriter_after_jit_bprop_graph]: 5.17999e-06
[opt_after_jit_grad]: 0.00045638
[symbol_engine_optimizer]: 7.537e-05, [1]
    [Cycle 1]: 6.794e-05, [6]
        [build]: 4.23001e-06
        [elim_shapecalc]: 7.58999e-06
        [elim_not_effective]: 1.316e-05
        [opt_reshape]: 6.16998e-06
        [fold_const_symbol]: 8.60999e-06
        [renormalize]: 5.8001e-07
[validate]: 7.4e-05
[backend_pass]: 8.2e-07
[task_emit]: 0.00628213
[execute]: 7.46999e-06
Sums
    bootstrap                            :     0.000295s :  1.82%
    type_inference                       :     0.004745s : 29.30%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000063s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.22%
    rewriter_before_opt_a                :     0.000078s :  0.48%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.38%
    jit_opt_a.loop_unroll                :     0.000053s :  0.33%
    jit_opt_a.a_1                        :     0.000805s :  4.97%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000023s :  0.14%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.001466s :  9.05%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000039s :  0.24%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000037s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000656s :  4.05%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000027s :  0.17%
    loop_unroll                          :     0.000425s :  2.63%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000456s :  2.82%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000074s :  0.46%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006282s : 38.79%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000275    23
 0.68% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000002s :      2: substitution.fold_const_symbol
 1.85% :     0.000005s :      3: substitution.graph_param_transform
80.50% :     0.000222s :      4: substitution.inline
 1.32% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.59% :     0.000004s :      4: substitution.remove_not_recompute_node
 3.97% :     0.000011s :      2: substitution.replace_old_param
 9.49% :     0.000026s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004681     2
81.77% :     0.003827s :      1: type_inference.infer
18.23% :     0.000853s :      1: type_inference.specialize
------[replace.]   0.000102     6
77.07% :     0.000079s :      4: replace.inline
22.93% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000244     6
89.77% :     0.000219s :      4: match.inline
10.23% :     0.000025s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000150   691
 1.05% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.82% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.83% :     0.000001s :     10: predicate.addn_check_dump
 1.07% :     0.000002s :     10: predicate.addn_zero_filter
 2.23% :     0.000003s :     10: predicate.arithmetic_simplify
 1.12% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.83% :     0.000001s :     10: predicate.compare_switch_simplify
 1.00% :     0.000001s :     10: predicate.depend_value_elim
 0.87% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.05% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.98% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.25% :     0.000000s :      3: predicate.elim_not_effective
 0.42% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.96% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000002s :     10: predicate.environ_get_depend_swap
 0.97% :     0.000001s :     10: predicate.environ_get_eliminate
 0.89% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.79% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 4.52% :     0.000007s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.29% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.84% :     0.000001s :      6: predicate.less_batch_normalization
 1.37% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.43% :     0.000002s :     15: predicate.load_eliminater
 1.11% :     0.000002s :      3: predicate.loop_unroll_after_grad
14.09% :     0.000021s :     34: predicate.loop_unroll_before_grad
 1.70% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.83% :     0.000001s :     10: predicate.merge_addn
 0.83% :     0.000001s :     10: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 1.70% :     0.000003s :     15: predicate.partial_eliminate
 0.99% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.25% :     0.000002s :     10: predicate.reduce_eliminate
 1.19% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.41% :     0.000002s :     18: predicate.replace_applicator
 0.37% :     0.000001s :      6: predicate.replace_old_param
 0.21% :     0.000000s :      3: predicate.reset_defer_inline
 1.05% :     0.000002s :     10: predicate.reshape_eliminate
 0.91% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.99% :     0.000001s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 0.82% :     0.000001s :      6: predicate.specialize_transform
 1.10% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.00% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.33% :     0.000001s :      3: predicate.switch_call_monad_eliminater
10.89% :     0.000016s :     16: predicate.switch_defer_inline
 1.77% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.13% :     0.000009s :     53: predicate.switch_simplify
 0.95% :     0.000001s :     10: predicate.tile_eliminate
 1.09% :     0.000002s :     10: predicate.transpose_eliminate
 1.17% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.03% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.07% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.27% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.22% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.25% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.29% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.53% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000606    12
47.12% :     0.000286s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.35% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.53% :     0.000270s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022862    76
 0.23% :     0.000053s :      1: add_recomputation
 0.29% :     0.000066s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.36% :     0.000312s :      1: bootstrap
 0.13% :     0.000030s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
28.14% :     0.006433s :      1: jit_opt_a
 0.63% :     0.000143s :      1: jit_opt_after_cconv
 0.23% :     0.000054s :      1: jit_opt_b
 1.90% :     0.000433s :      1: loop_unroll
 2.91% :     0.000666s :      1: mutable_eliminate
 4.57% :     0.001044s :     26: opt.transform.jit_opt_a
 0.20% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.03% :     0.000465s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000039s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 4.90% :     0.001120s :      1: renormalize.infer
 1.47% :     0.000335s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000040s :      1: rewriter_after_opt_a
 0.35% :     0.000081s :      1: rewriter_before_opt_a
 0.34% :     0.000078s :      1: symbol_engine_optimizer
27.54% :     0.006296s :      1: task_emit
20.81% :     0.004758s :      1: type_inference
 0.42% :     0.000096s :      1: validate


TotalTime = 0.017899, [33]
[bootstrap]: 0.0002537
[type_inference]: 0.00387589
[event_method]: 1.493e-05
[auto_monad]: 4.111e-05
[graph_reusing]: 4.45999e-06
[pre_auto_parallel]: 2.44999e-06
[py_interpret_to_execute]: 2.964e-05
[rewriter_before_opt_a]: 6.739e-05
[expand_dump_flag]: 3.21001e-06
[jit_opt_a]: 0.00503189, [2]
    [Cycle 1]: 0.00149025, [27]
        [switch_simplify]: 5.3e-05
        [loop_unroll]: 2.949e-05
        [a_1]: 0.00051418
        [with_stream_mark]: 1.727e-05
        [recompute_prepare]: 6.79999e-06
        [updatestate_depend_eliminate]: 3.47002e-06
        [updatestate_assign_eliminate]: 2.80997e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 1.88002e-06
        [specialize_transform]: 5.81998e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.24e-06
        [meta_shard_fg_expand]: 2.12001e-06
        [get_grad_eliminate_]: 4.79998e-06
        [merge_forward]: 4.05998e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.223e-05
        [j_node_and_user_rematch]: 8.24002e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 7.88001e-06
        [inline_without_move]: 5.06002e-06
        [renormalize]: 0.0005777
        [add_forward_monad_depend]: 5.23002e-06
        [auto_monad_grad]: 1.76e-06
        [auto_monad_eliminator]: 1.322e-05
        [cse]: 1.723e-05
        [replace_applicator]: 1.161e-05
    [Cycle 2]: 0.0003092, [27]
        [switch_simplify]: 5.34e-06
        [loop_unroll]: 4.78001e-06
        [a_1]: 8.623e-05
        [with_stream_mark]: 8.28999e-06
        [recompute_prepare]: 5.06002e-06
        [updatestate_depend_eliminate]: 2.36e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.98002e-06
        [parameter_eliminate]: 9.30013e-07
        [specialize_transform]: 4.77e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 4.88001e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 4.55001e-06
        [merge_forward]: 2.50002e-06
        [cell_reuse_recompute_pass]: 1.84e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.152e-05
        [j_node_and_user_rematch]: 7.41001e-06
        [meta_fg_expand]: 1.44998e-06
        [replace_old_param]: 7.13998e-06
        [inline_without_move]: 4.50999e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.14003e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 5.14998e-06
        [cse]: 1.009e-05
        [replace_applicator]: 4.81002e-06
[py_interpret_to_execute_after_opt_a]: 1.02e-05
[rewriter_after_opt_a]: 3.14e-05
[convert_after_rewriter]: 6.66999e-06
[order_py_execute_after_rewriter]: 5.19e-06
[mutable_eliminate]: 0.00052795
[jit_opt_b]: 4.747e-05, [1]
    [Cycle 1]: 4.041e-05, [2]
        [frontend_op_eliminate]: 1.553e-05
        [inline_after_opt_a]: 1.388e-05
[cconv]: 2.419e-05
[loop_unroll]: 0.00040843
[jit_opt_after_cconv]: 0.00014763, [1]
    [Cycle 1]: 0.00013008, [11]
        [c_1]: 2.032e-05
        [parameter_eliminate]: 2.55002e-06
        [updatestate_depend_eliminate]: 5.16002e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [cse]: 1.694e-05
        [call_graph_tuple_transform]: 1.747e-05
        [tuple_list_get_item_eliminator]: 5.03002e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.54e-06
[remove_dup_value]: 1.268e-05
[partial_unused_args_eliminate]: 2.24999e-06
[environ_conv]: 5.00999e-06
[add_recomputation]: 4.463e-05
[cse_after_recomputation]: 1.979e-05, [1]
    [Cycle 1]: 1.428e-05, [1]
        [cse]: 8.06001e-06
[auto_monad_reorder]: 1.374e-05
[get_jit_bprop_graph]: 2.04999e-06
[rewriter_after_jit_bprop_graph]: 4.97999e-06
[opt_after_jit_grad]: 0.00044962
[symbol_engine_optimizer]: 6.891e-05, [1]
    [Cycle 1]: 6.274e-05, [6]
        [build]: 3.11001e-06
        [elim_shapecalc]: 7.15e-06
        [elim_not_effective]: 1.214e-05
        [opt_reshape]: 4.99e-06
        [fold_const_symbol]: 8.15e-06
        [renormalize]: 5.00004e-07
[validate]: 3.397e-05
[backend_pass]: 1.08001e-06
[task_emit]: 0.00651394
[execute]: 9.37001e-06
Sums
    bootstrap                            :     0.000254s :  1.80%
    type_inference                       :     0.003876s : 27.55%
    event_method                         :     0.000015s :  0.11%
    auto_monad                           :     0.000041s :  0.29%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.21%
    rewriter_before_opt_a                :     0.000067s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000600s :  4.27%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000578s :  4.11%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000027s :  0.19%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000031s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000528s :  3.75%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000408s :  2.90%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000045s :  0.32%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000450s :  3.20%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006514s : 46.30%
    execute                              :     0.000009s :  0.07%


Time group info:
------[substitution.]   0.000197    23
 0.95% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.57% :     0.000005s :      3: substitution.graph_param_transform
84.65% :     0.000167s :      4: substitution.inline
 1.38% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.03% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.75% :     0.000003s :      2: substitution.replace_old_param
 6.08% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003828     2
82.08% :     0.003142s :      1: type_inference.infer
17.92% :     0.000686s :      1: type_inference.specialize
------[replace.]   0.000063     6
68.57% :     0.000044s :      4: replace.inline
31.43% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000175     6
93.79% :     0.000164s :      4: match.inline
 6.21% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.32% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 2.04% :     0.000002s :     10: predicate.addn_zero_filter
 1.74% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.47% :     0.000002s :     10: predicate.depend_value_elim
 1.22% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.39% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.65% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000001s :      6: predicate.less_batch_normalization
 1.90% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.14% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.68% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.40% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.89% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.61% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.99% :     0.000009s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.43% :     0.000002s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000497    12
47.05% :     0.000234s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.37% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.57% :     0.000227s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019375    76
 0.25% :     0.000047s :      1: add_recomputation
 0.23% :     0.000044s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.36% :     0.000264s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.09% :     0.000018s :      1: event_method
 0.08% :     0.000015s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
25.99% :     0.005035s :      1: jit_opt_a
 0.78% :     0.000150s :      1: jit_opt_after_cconv
 0.26% :     0.000050s :      1: jit_opt_b
 2.15% :     0.000416s :      1: loop_unroll
 2.77% :     0.000536s :      1: mutable_eliminate
 4.12% :     0.000799s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.36% :     0.000457s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.66% :     0.000321s :      1: renormalize.infer
 1.28% :     0.000249s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.36% :     0.000070s :      1: rewriter_before_opt_a
 0.37% :     0.000071s :      1: symbol_engine_optimizer
33.71% :     0.006531s :      1: task_emit
20.06% :     0.003886s :      1: type_inference
 0.28% :     0.000053s :      1: validate


TotalTime = 0.0272866, [33]
[bootstrap]: 0.00040099
[type_inference]: 0.00575899
[event_method]: 2.2e-05
[auto_monad]: 7.531e-05
[graph_reusing]: 6.09001e-06
[pre_auto_parallel]: 2.68e-06
[py_interpret_to_execute]: 5.234e-05
[rewriter_before_opt_a]: 8.411e-05
[expand_dump_flag]: 3.23e-06
[jit_opt_a]: 0.0089477, [2]
    [Cycle 1]: 0.00236729, [27]
        [switch_simplify]: 8.2e-05
        [loop_unroll]: 3.2e-05
        [a_1]: 0.00071795
        [with_stream_mark]: 2.944e-05
        [recompute_prepare]: 1.103e-05
        [updatestate_depend_eliminate]: 4.19997e-06
        [updatestate_assign_eliminate]: 3.53e-06
        [updatestate_loads_eliminate]: 2.66999e-06
        [parameter_eliminate]: 2.47001e-06
        [specialize_transform]: 7.77998e-06
        [updatestate_useless_node_eliminater]: 6.04001e-06
        [accelerated_algorithm]: 6.40002e-06
        [meta_shard_fg_expand]: 2.96001e-06
        [get_grad_eliminate_]: 5.77001e-06
        [merge_forward]: 4.07e-06
        [cell_reuse_recompute_pass]: 2.70002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.05e-05
        [j_node_and_user_rematch]: 9.90002e-06
        [meta_fg_expand]: 2.61e-06
        [replace_old_param]: 1.183e-05
        [inline_without_move]: 6.26e-06
        [renormalize]: 0.00109328
        [add_forward_monad_depend]: 8.74998e-06
        [auto_monad_grad]: 2.46998e-06
        [auto_monad_eliminator]: 1.934e-05
        [cse]: 3.106e-05
        [replace_applicator]: 2.158e-05
    [Cycle 2]: 0.00039953, [27]
        [switch_simplify]: 6.78998e-06
        [loop_unroll]: 5.61e-06
        [a_1]: 9.682e-05
        [with_stream_mark]: 1.435e-05
        [recompute_prepare]: 7.35998e-06
        [updatestate_depend_eliminate]: 2.93e-06
        [updatestate_assign_eliminate]: 3.15998e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 2.63e-06
        [specialize_transform]: 5.46e-06
        [updatestate_useless_node_eliminater]: 5.84e-06
        [accelerated_algorithm]: 6.39001e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 4.44002e-06
        [merge_forward]: 4.03001e-06
        [cell_reuse_recompute_pass]: 3.88001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.908e-05
        [j_node_and_user_rematch]: 1.007e-05
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 9.99001e-06
        [inline_without_move]: 6.31998e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 2.06e-06
        [auto_monad_grad]: 2.04e-06
        [auto_monad_eliminator]: 8.89e-06
        [cse]: 1.475e-05
        [replace_applicator]: 7.65998e-06
[py_interpret_to_execute_after_opt_a]: 1.726e-05
[rewriter_after_opt_a]: 4.675e-05
[convert_after_rewriter]: 8.13001e-06
[order_py_execute_after_rewriter]: 6.23e-06
[mutable_eliminate]: 0.00093865
[jit_opt_b]: 8.409e-05, [1]
    [Cycle 1]: 7.501e-05, [2]
        [frontend_op_eliminate]: 4.016e-05
        [inline_after_opt_a]: 1.917e-05
[cconv]: 3.866e-05
[loop_unroll]: 0.00058027
[jit_opt_after_cconv]: 0.00017235, [1]
    [Cycle 1]: 0.00016549, [11]
        [c_1]: 2.246e-05
        [parameter_eliminate]: 5.59e-06
        [updatestate_depend_eliminate]: 1.081e-05
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.99999e-06
        [cse]: 2.903e-05
        [call_graph_tuple_transform]: 2.296e-05
        [tuple_list_get_item_eliminator]: 5.72999e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 1.00999e-06
        [switch_simplify]: 6.26e-06
[remove_dup_value]: 1.472e-05
[partial_unused_args_eliminate]: 2.51e-06
[environ_conv]: 6.46e-06
[add_recomputation]: 5.756e-05
[cse_after_recomputation]: 2.168e-05, [1]
    [Cycle 1]: 1.592e-05, [1]
        [cse]: 9.64e-06
[auto_monad_reorder]: 1.657e-05
[get_jit_bprop_graph]: 2.43e-06
[rewriter_after_jit_bprop_graph]: 9.00999e-06
[opt_after_jit_grad]: 0.00049324
[symbol_engine_optimizer]: 7.951e-05, [1]
    [Cycle 1]: 7.198e-05, [6]
        [build]: 4.67e-06
        [elim_shapecalc]: 8.08001e-06
        [elim_not_effective]: 1.376e-05
        [opt_reshape]: 6.31e-06
        [fold_const_symbol]: 9.73002e-06
        [renormalize]: 1.72001e-06
[validate]: 4.492e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.0090258
[execute]: 1.08e-05
Sums
    bootstrap                            :     0.000401s :  1.97%
    type_inference                       :     0.005759s : 28.29%
    event_method                         :     0.000022s :  0.11%
    auto_monad                           :     0.000075s :  0.37%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000052s :  0.26%
    rewriter_before_opt_a                :     0.000084s :  0.41%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000089s :  0.44%
    jit_opt_a.loop_unroll                :     0.000038s :  0.18%
    jit_opt_a.a_1                        :     0.000815s :  4.00%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.11%
    jit_opt_a.inline_without_move        :     0.000013s :  0.06%
    jit_opt_a.renormalize                :     0.001093s :  5.37%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.14%
    jit_opt_a.cse                        :     0.000046s :  0.23%
    jit_opt_a.replace_applicator         :     0.000029s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.08%
    rewriter_after_opt_a                 :     0.000047s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000939s :  4.61%
    jit_opt_b.frontend_op_eliminate      :     0.000040s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.09%
    cconv                                :     0.000039s :  0.19%
    loop_unroll                          :     0.000580s :  2.85%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.07%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000058s :  0.28%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.04%
    opt_after_jit_grad                   :     0.000493s :  2.42%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000002s :  0.01%
    validate                             :     0.000045s :  0.22%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009026s : 44.34%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000319    23
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000002s :      2: substitution.fold_const_symbol
 1.98% :     0.000006s :      3: substitution.graph_param_transform
86.17% :     0.000275s :      4: substitution.inline
 1.33% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.28% :     0.000007s :      4: substitution.remove_not_recompute_node
 2.34% :     0.000007s :      2: substitution.replace_old_param
 4.60% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005671     2
82.60% :     0.004684s :      1: type_inference.infer
17.40% :     0.000987s :      1: type_inference.specialize
------[replace.]   0.000097     6
63.41% :     0.000062s :      4: replace.inline
36.59% :     0.000036s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000285     6
95.24% :     0.000271s :      4: match.inline
 4.76% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000143   691
 1.21% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.15% :     0.000002s :     10: predicate.addn_zero_filter
 2.68% :     0.000004s :     10: predicate.arithmetic_simplify
 1.14% :     0.000002s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.89% :     0.000001s :     10: predicate.compare_switch_simplify
 1.95% :     0.000003s :     10: predicate.depend_value_elim
 1.59% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.10% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.00% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.03% :     0.000001s :     10: predicate.environ_get_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.17% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.19% :     0.000000s :      3: predicate.graph_param_transform
 5.73% :     0.000008s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000003s :     15: predicate.load_eliminater
 1.63% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.85% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.88% :     0.000001s :     10: predicate.merge_addn
 1.36% :     0.000002s :     10: predicate.minmaximum_grad
 2.64% :     0.000004s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 1.65% :     0.000002s :     15: predicate.partial_eliminate
 1.30% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.42% :     0.000002s :     10: predicate.reduce_eliminate
 1.18% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.54% :     0.000002s :     18: predicate.replace_applicator
 0.87% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 2.22% :     0.000003s :     10: predicate.reshape_eliminate
 1.00% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.09% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.61% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 0.96% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.72% :     0.000004s :     16: predicate.switch_defer_inline
 1.90% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.97% :     0.000011s :     53: predicate.switch_simplify
 1.25% :     0.000002s :     10: predicate.tile_eliminate
 1.01% :     0.000001s :     10: predicate.transpose_eliminate
 1.07% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.00% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 3.12% :     0.000004s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.46% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.41% :     0.000002s :     10: predicate.value_based_eliminate
 0.27% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000829    12
50.36% :     0.000418s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.42% :     0.000062s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.22% :     0.000350s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.029609    76
 0.21% :     0.000061s :      1: add_recomputation
 0.27% :     0.000080s :      1: auto_monad
 0.06% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.42% :     0.000421s :      1: bootstrap
 0.14% :     0.000042s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.08% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.09% :     0.000027s :      1: event_method
 0.06% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
30.24% :     0.008953s :      1: jit_opt_a
 0.59% :     0.000176s :      1: jit_opt_after_cconv
 0.29% :     0.000087s :      1: jit_opt_b
 2.00% :     0.000591s :      1: loop_unroll
 3.22% :     0.000952s :      1: mutable_eliminate
 3.69% :     0.001093s :     26: opt.transform.jit_opt_a
 0.18% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000051s :      4: opt.transform.jit_opt_b
 0.06% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000025s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.69% :     0.000501s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000055s :      1: py_interpret_to_execute
 0.07% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 1.99% :     0.000588s :      1: renormalize.infer
 1.66% :     0.000492s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000051s :      1: rewriter_after_opt_a
 0.29% :     0.000087s :      1: rewriter_before_opt_a
 0.28% :     0.000082s :      1: symbol_engine_optimizer
30.56% :     0.009048s :      1: task_emit
19.51% :     0.005778s :      1: type_inference
 0.23% :     0.000068s :      1: validate


TotalTime = 0.0191851, [33]
[bootstrap]: 0.00040548
[type_inference]: 0.00443773
[event_method]: 1.823e-05
[auto_monad]: 6.126e-05
[graph_reusing]: 6.17001e-06
[pre_auto_parallel]: 2.46998e-06
[py_interpret_to_execute]: 3.546e-05
[rewriter_before_opt_a]: 7.317e-05
[expand_dump_flag]: 2.86999e-06
[jit_opt_a]: 0.00550048, [2]
    [Cycle 1]: 0.00168312, [27]
        [switch_simplify]: 5.857e-05
        [loop_unroll]: 3.156e-05
        [a_1]: 0.00055231
        [with_stream_mark]: 2.272e-05
        [recompute_prepare]: 7.7e-06
        [updatestate_depend_eliminate]: 3.64002e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.96999e-06
        [parameter_eliminate]: 1.95001e-06
        [specialize_transform]: 6.38e-06
        [updatestate_useless_node_eliminater]: 5.27999e-06
        [accelerated_algorithm]: 6.29999e-06
        [meta_shard_fg_expand]: 2.32999e-06
        [get_grad_eliminate_]: 6.08998e-06
        [merge_forward]: 3.48999e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.398e-05
        [j_node_and_user_rematch]: 9.39e-06
        [meta_fg_expand]: 2.24999e-06
        [replace_old_param]: 9.23002e-06
        [inline_without_move]: 5.54e-06
        [renormalize]: 0.00068331
        [add_forward_monad_depend]: 5.81e-06
        [auto_monad_grad]: 2.46e-06
        [auto_monad_eliminator]: 1.311e-05
        [cse]: 2.799e-05
        [replace_applicator]: 1.218e-05
    [Cycle 2]: 0.00031198, [27]
        [switch_simplify]: 5.37001e-06
        [loop_unroll]: 4.84e-06
        [a_1]: 8.93e-05
        [with_stream_mark]: 9.31998e-06
        [recompute_prepare]: 5.12e-06
        [updatestate_depend_eliminate]: 2.73998e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 9.09989e-07
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 4.95999e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 4.40999e-06
        [merge_forward]: 2.38998e-06
        [cell_reuse_recompute_pass]: 1.47999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.253e-05
        [j_node_and_user_rematch]: 7.65e-06
        [meta_fg_expand]: 1.73002e-06
        [replace_old_param]: 6.78998e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.39998e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 5.09998e-06
        [cse]: 1.016e-05
        [replace_applicator]: 5.04e-06
[py_interpret_to_execute_after_opt_a]: 9.41e-06
[rewriter_after_opt_a]: 3.44e-05
[convert_after_rewriter]: 6.78e-06
[order_py_execute_after_rewriter]: 4.84e-06
[mutable_eliminate]: 0.00056245
[jit_opt_b]: 5.017e-05, [1]
    [Cycle 1]: 4.338e-05, [2]
        [frontend_op_eliminate]: 1.57e-05
        [inline_after_opt_a]: 1.615e-05
[cconv]: 2.647e-05
[loop_unroll]: 0.00050635
[jit_opt_after_cconv]: 0.00014104, [1]
    [Cycle 1]: 0.0001345, [11]
        [c_1]: 2.034e-05
        [parameter_eliminate]: 2.54999e-06
        [updatestate_depend_eliminate]: 6.38998e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [cse]: 1.975e-05
        [call_graph_tuple_transform]: 1.829e-05
        [tuple_list_get_item_eliminator]: 5.62001e-06
        [none_parameter_eliminate]: 1.70001e-06
        [renormalize]: 5.60016e-07
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 1.368e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 5.04e-06
[add_recomputation]: 4.914e-05
[cse_after_recomputation]: 2.072e-05, [1]
    [Cycle 1]: 1.501e-05, [1]
        [cse]: 9.12999e-06
[auto_monad_reorder]: 1.516e-05
[get_jit_bprop_graph]: 1.91e-06
[rewriter_after_jit_bprop_graph]: 6.00002e-06
[opt_after_jit_grad]: 0.00046726
[symbol_engine_optimizer]: 7.399e-05, [1]
    [Cycle 1]: 6.766e-05, [6]
        [build]: 4.54998e-06
        [elim_shapecalc]: 7.63999e-06
        [elim_not_effective]: 1.289e-05
        [opt_reshape]: 5.62999e-06
        [fold_const_symbol]: 8.70999e-06
        [renormalize]: 3.69997e-07
[validate]: 3.914e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.00637481
[execute]: 8.84e-06
Sums
    bootstrap                            :     0.000405s :  2.69%
    type_inference                       :     0.004438s : 29.49%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000061s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.24%
    rewriter_before_opt_a                :     0.000073s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.42%
    jit_opt_a.loop_unroll                :     0.000036s :  0.24%
    jit_opt_a.a_1                        :     0.000642s :  4.26%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000683s :  4.54%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000038s :  0.25%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000034s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000562s :  3.74%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000026s :  0.18%
    loop_unroll                          :     0.000506s :  3.37%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000467s :  3.11%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000039s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006375s : 42.37%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000213    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.39% :     0.000005s :      3: substitution.graph_param_transform
85.12% :     0.000182s :      4: substitution.inline
 1.48% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.04% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.65% :     0.000004s :      2: substitution.replace_old_param
 5.83% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004368     2
81.34% :     0.003553s :      1: type_inference.infer
18.66% :     0.000815s :      1: type_inference.specialize
------[replace.]   0.000070     6
67.00% :     0.000047s :      4: replace.inline
33.00% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000190     6
94.07% :     0.000179s :      4: match.inline
 5.93% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.52% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 1.90% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.70% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.29% :     0.000002s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.92% :     0.000007s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.66% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.71% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.68% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.42% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.50% :     0.000001s :      3: predicate.reset_defer_inline
 1.28% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.99% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.79% :     0.000009s :     53: predicate.switch_simplify
 1.39% :     0.000002s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.08% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.07% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000559    12
47.57% :     0.000266s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.17% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.25% :     0.000253s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020823    76
 0.25% :     0.000052s :      1: add_recomputation
 0.31% :     0.000065s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 2.08% :     0.000433s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.43% :     0.005504s :      1: jit_opt_a
 0.69% :     0.000144s :      1: jit_opt_after_cconv
 0.25% :     0.000053s :      1: jit_opt_b
 2.48% :     0.000515s :      1: loop_unroll
 2.75% :     0.000572s :      1: mutable_eliminate
 4.11% :     0.000855s :     26: opt.transform.jit_opt_a
 0.22% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.28% :     0.000475s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000038s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.86% :     0.000387s :      1: renormalize.infer
 1.39% :     0.000288s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000037s :      1: rewriter_after_opt_a
 0.37% :     0.000076s :      1: rewriter_before_opt_a
 0.37% :     0.000076s :      1: symbol_engine_optimizer
30.69% :     0.006391s :      1: task_emit
21.39% :     0.004453s :      1: type_inference
 0.29% :     0.000061s :      1: validate


TotalTime = 0.0189805, [33]
[bootstrap]: 0.00030283
[type_inference]: 0.00511855
[event_method]: 1.736e-05
[auto_monad]: 6.216e-05
[graph_reusing]: 5.82001e-06
[pre_auto_parallel]: 2.36e-06
[py_interpret_to_execute]: 3.169e-05
[rewriter_before_opt_a]: 7.266e-05
[expand_dump_flag]: 2.94001e-06
[jit_opt_a]: 0.00513256, [2]
    [Cycle 1]: 0.00146945, [27]
        [switch_simplify]: 5.161e-05
        [loop_unroll]: 2.819e-05
        [a_1]: 0.00052064
        [with_stream_mark]: 1.903e-05
        [recompute_prepare]: 7.38e-06
        [updatestate_depend_eliminate]: 3.73001e-06
        [updatestate_assign_eliminate]: 3.33e-06
        [updatestate_loads_eliminate]: 2.77002e-06
        [parameter_eliminate]: 1.97001e-06
        [specialize_transform]: 6.03002e-06
        [updatestate_useless_node_eliminater]: 5.20001e-06
        [accelerated_algorithm]: 5.28002e-06
        [meta_shard_fg_expand]: 2.41998e-06
        [get_grad_eliminate_]: 5.54998e-06
        [merge_forward]: 3.63999e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.454e-05
        [j_node_and_user_rematch]: 9.22001e-06
        [meta_fg_expand]: 2.32999e-06
        [replace_old_param]: 8.18999e-06
        [inline_without_move]: 5.09998e-06
        [renormalize]: 0.00053554
        [add_forward_monad_depend]: 5.91e-06
        [auto_monad_grad]: 2.66999e-06
        [auto_monad_eliminator]: 1.344e-05
        [cse]: 2.527e-05
        [replace_applicator]: 1.038e-05
    [Cycle 2]: 0.00031375, [27]
        [switch_simplify]: 5.29998e-06
        [loop_unroll]: 4.62998e-06
        [a_1]: 8.717e-05
        [with_stream_mark]: 9.37999e-06
        [recompute_prepare]: 4.63999e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.86998e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.85001e-06
        [updatestate_useless_node_eliminater]: 4.68999e-06
        [accelerated_algorithm]: 5.08002e-06
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.89999e-06
        [cell_reuse_recompute_pass]: 2.06e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.255e-05
        [j_node_and_user_rematch]: 7.82e-06
        [meta_fg_expand]: 1.44998e-06
        [replace_old_param]: 7.73999e-06
        [inline_without_move]: 4.54998e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.05999e-06
        [auto_monad_grad]: 1.03001e-06
        [auto_monad_eliminator]: 5.39e-06
        [cse]: 9.07001e-06
        [replace_applicator]: 5.05001e-06
[py_interpret_to_execute_after_opt_a]: 9.32999e-06
[rewriter_after_opt_a]: 3.382e-05
[convert_after_rewriter]: 6.91999e-06
[order_py_execute_after_rewriter]: 4.71002e-06
[mutable_eliminate]: 0.00071851
[jit_opt_b]: 5.354e-05, [1]
    [Cycle 1]: 4.605e-05, [2]
        [frontend_op_eliminate]: 1.865e-05
        [inline_after_opt_a]: 1.543e-05
[cconv]: 2.388e-05
[loop_unroll]: 0.0004436
[jit_opt_after_cconv]: 0.00015015, [1]
    [Cycle 1]: 0.00014307, [11]
        [c_1]: 2.13e-05
        [parameter_eliminate]: 2.78998e-06
        [updatestate_depend_eliminate]: 6.11e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [cse]: 1.698e-05
        [call_graph_tuple_transform]: 1.866e-05
        [tuple_list_get_item_eliminator]: 5.83002e-06
        [none_parameter_eliminate]: 1.81003e-06
        [renormalize]: 4.90021e-07
        [switch_simplify]: 5.45001e-06
[remove_dup_value]: 1.361e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 5.30999e-06
[add_recomputation]: 4.736e-05
[cse_after_recomputation]: 1.991e-05, [1]
    [Cycle 1]: 1.41e-05, [1]
        [cse]: 8.40001e-06
[auto_monad_reorder]: 1.467e-05
[get_jit_bprop_graph]: 1.81e-06
[rewriter_after_jit_bprop_graph]: 5.51e-06
[opt_after_jit_grad]: 0.00046409
[symbol_engine_optimizer]: 7.217e-05, [1]
    [Cycle 1]: 6.578e-05, [6]
        [build]: 3.65e-06
        [elim_shapecalc]: 7.15998e-06
        [elim_not_effective]: 1.257e-05
        [opt_reshape]: 5.69e-06
        [fold_const_symbol]: 8.92e-06
        [renormalize]: 4.20026e-07
[validate]: 2.998e-05
[backend_pass]: 1.32e-06
[task_emit]: 0.0059055
[execute]: 7.97998e-06
Sums
    bootstrap                            :     0.000303s :  2.02%
    type_inference                       :     0.005119s : 34.07%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000062s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.21%
    rewriter_before_opt_a                :     0.000073s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.38%
    jit_opt_a.loop_unroll                :     0.000033s :  0.22%
    jit_opt_a.a_1                        :     0.000608s :  4.05%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000536s :  3.57%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000034s :  0.23%
    jit_opt_a.replace_applicator         :     0.000015s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000034s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000719s :  4.78%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000024s :  0.16%
    loop_unroll                          :     0.000444s :  2.95%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.32%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000464s :  3.09%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000030s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005906s : 39.31%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000207    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.49% :     0.000005s :      3: substitution.graph_param_transform
84.48% :     0.000175s :      4: substitution.inline
 1.35% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.20% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.98% :     0.000004s :      2: substitution.replace_old_param
 5.79% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005049     2
81.51% :     0.004115s :      1: type_inference.infer
18.49% :     0.000934s :      1: type_inference.specialize
------[replace.]   0.000062     6
68.25% :     0.000042s :      4: replace.inline
31.75% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000183     6
94.08% :     0.000172s :      4: match.inline
 5.92% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.15% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 1.67% :     0.000002s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.08% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.81% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.62% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.88% :     0.000002s :     15: predicate.load_eliminater
 1.51% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.02% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.28% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.96% :     0.000002s :      3: predicate.mutable_eliminate
 0.57% :     0.000001s :      3: predicate.opt_reshape
 2.53% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.25% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.92% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.95% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000590    12
51.64% :     0.000305s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.67% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.70% :     0.000246s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020432    76
 0.25% :     0.000050s :      1: add_recomputation
 0.32% :     0.000065s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.58% :     0.000323s :      1: bootstrap
 0.13% :     0.000027s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.13% :     0.005135s :      1: jit_opt_a
 0.75% :     0.000153s :      1: jit_opt_after_cconv
 0.27% :     0.000056s :      1: jit_opt_b
 2.21% :     0.000452s :      1: loop_unroll
 3.57% :     0.000729s :      1: mutable_eliminate
 3.96% :     0.000809s :     26: opt.transform.jit_opt_a
 0.24% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000028s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.31% :     0.000471s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.36% :     0.000279s :      1: renormalize.infer
 1.22% :     0.000249s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000037s :      1: rewriter_after_opt_a
 0.37% :     0.000076s :      1: rewriter_before_opt_a
 0.37% :     0.000075s :      1: symbol_engine_optimizer
28.96% :     0.005916s :      1: task_emit
25.13% :     0.005134s :      1: type_inference
 0.23% :     0.000048s :      1: validate


TotalTime = 0.0150826, [30]
[bootstrap]: 0.00029427
[type_inference]: 0.00655588
[event_method]: 2.139e-05
[auto_monad]: 6.659e-05
[graph_reusing]: 5.05001e-06
[pre_auto_parallel]: 2.10002e-06
[py_interpret_to_execute]: 6.035e-05
[rewriter_before_opt_a]: 7.929e-05
[expand_dump_flag]: 2.48e-06
[jit_opt_a]: 0.00570978, [2]
    [Cycle 1]: 0.0021278, [27]
        [switch_simplify]: 9.917e-05
        [loop_unroll]: 3.433e-05
        [a_1]: 0.00066739
        [with_stream_mark]: 1.233e-05
        [recompute_prepare]: 6.94001e-06
        [updatestate_depend_eliminate]: 2.88e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.01997e-06
        [specialize_transform]: 6.06998e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 5.82001e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 5.74999e-06
        [merge_forward]: 2.64001e-06
        [cell_reuse_recompute_pass]: 1.43002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.161e-05
        [j_node_and_user_rematch]: 9.02e-06
        [meta_fg_expand]: 1.81998e-06
        [replace_old_param]: 9.02999e-06
        [inline_without_move]: 5.53002e-06
        [renormalize]: 0.00102983
        [add_forward_monad_depend]: 4.63999e-06
        [auto_monad_grad]: 1.23002e-06
        [auto_monad_eliminator]: 8.77e-06
        [cse]: 1.52e-05
        [replace_applicator]: 9.71e-06
    [Cycle 2]: 0.00032568, [27]
        [switch_simplify]: 5.50001e-06
        [loop_unroll]: 5.00999e-06
        [a_1]: 9.127e-05
        [with_stream_mark]: 9.24e-06
        [recompute_prepare]: 5.63002e-06
        [updatestate_depend_eliminate]: 2.52001e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 5.06002e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 4.95001e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.50999e-06
        [merge_forward]: 2.21e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.442e-05
        [j_node_and_user_rematch]: 9.07999e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 8.03999e-06
        [inline_without_move]: 5.24e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.23002e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 5.00001e-06
        [cse]: 1.023e-05
        [replace_applicator]: 5.62001e-06
[py_interpret_to_execute_after_opt_a]: 1.059e-05
[rewriter_after_opt_a]: 2.991e-05
[convert_after_rewriter]: 6.57002e-06
[order_py_execute_after_rewriter]: 4.65001e-06
[mutable_eliminate]: 0.00054796
[jit_opt_b]: 4.737e-05, [1]
    [Cycle 1]: 4.164e-05, [2]
        [frontend_op_eliminate]: 1.591e-05
        [inline_after_opt_a]: 1.5e-05
[cconv]: 1.907e-05
[loop_unroll]: 0.00046795
[jit_opt_after_cconv]: 0.00015425, [1]
    [Cycle 1]: 0.00014796, [11]
        [c_1]: 2.393e-05
        [parameter_eliminate]: 2.32999e-06
        [updatestate_depend_eliminate]: 5.12e-06
        [updatestate_assign_eliminate]: 2.57001e-06
        [updatestate_loads_eliminate]: 2.30997e-06
        [cse]: 2.198e-05
        [call_graph_tuple_transform]: 2.053e-05
        [tuple_list_get_item_eliminator]: 5.91e-06
        [none_parameter_eliminate]: 1.14e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 6.68e-06
[remove_dup_value]: 1.192e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 5.15001e-06
[add_recomputation]: 4.034e-05
[cse_after_recomputation]: 2.484e-05, [1]
    [Cycle 1]: 1.847e-05, [1]
        [cse]: 1.187e-05
[auto_monad_reorder]: 1.219e-05
[get_jit_bprop_graph]: 2.17001e-06
[rewriter_after_jit_bprop_graph]: 6.78e-06
[opt_after_jit_grad]: 0.00057848
[symbol_engine_optimizer]: 8.825e-05, [1]
    [Cycle 1]: 8.093e-05, [6]
        [build]: 4.40999e-06
        [elim_shapecalc]: 9.36e-06
        [elim_not_effective]: 1.709e-05
        [opt_reshape]: 7.43999e-06
        [fold_const_symbol]: 9.94001e-06
        [renormalize]: 4.00003e-07
[validate]: 4.076e-05
Sums
    bootstrap                            :     0.000294s :  2.62%
    type_inference                       :     0.006556s : 58.37%
    event_method                         :     0.000021s :  0.19%
    auto_monad                           :     0.000067s :  0.59%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000060s :  0.54%
    rewriter_before_opt_a                :     0.000079s :  0.71%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000105s :  0.93%
    jit_opt_a.loop_unroll                :     0.000039s :  0.35%
    jit_opt_a.a_1                        :     0.000759s :  6.75%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.001030s :  9.17%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.12%
    jit_opt_a.cse                        :     0.000025s :  0.23%
    jit_opt_a.replace_applicator         :     0.000015s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.09%
    rewriter_after_opt_a                 :     0.000030s :  0.27%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000548s :  4.88%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.13%
    cconv                                :     0.000019s :  0.17%
    loop_unroll                          :     0.000468s :  4.17%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.06%
    remove_dup_value                     :     0.000012s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000040s :  0.36%
    cse_after_recomputation.cse          :     0.000012s :  0.11%
    auto_monad_reorder                   :     0.000012s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000578s :  5.15%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000041s :  0.36%


Time group info:
------[substitution.]   0.000255    26
 1.10% :     0.000003s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 1.97% :     0.000005s :      3: substitution.graph_param_transform
85.23% :     0.000217s :      6: substitution.inline
 1.13% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.70% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.37% :     0.000003s :      2: substitution.replace_old_param
 3.34% :     0.000009s :      1: substitution.switch_simplify
 3.72% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006497     2
82.73% :     0.005375s :      1: type_inference.infer
17.27% :     0.001122s :      1: type_inference.specialize
------[replace.]   0.000126     9
53.12% :     0.000067s :      6: replace.inline
19.33% :     0.000024s :      1: replace.switch_simplify
27.56% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000230     9
93.01% :     0.000214s :      6: match.inline
 3.34% :     0.000008s :      1: match.switch_simplify
 3.65% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   801
 1.37% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.61% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     12: predicate.addn_check_dump
 1.34% :     0.000002s :     12: predicate.addn_zero_filter
 1.93% :     0.000003s :     12: predicate.arithmetic_simplify
 1.34% :     0.000002s :     12: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     12: predicate.compare_switch_simplify
 1.50% :     0.000002s :     12: predicate.depend_value_elim
 1.15% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.97% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.13% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.43% :     0.000002s :     12: predicate.environ_get_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 4.90% :     0.000006s :     26: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :      6: predicate.less_batch_normalization
 1.67% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     17: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.95% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     12: predicate.merge_addn
 1.12% :     0.000001s :     12: predicate.minmaximum_grad
 1.29% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     17: predicate.partial_eliminate
 1.14% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     12: predicate.reduce_eliminate
 1.54% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     20: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.20% :     0.000000s :      3: predicate.reset_defer_inline
 1.92% :     0.000003s :     12: predicate.reshape_eliminate
 1.24% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     12: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.62% :     0.000003s :     20: predicate.switch_defer_inline
 2.50% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.19% :     0.000011s :     65: predicate.switch_simplify
 1.24% :     0.000002s :     12: predicate.tile_eliminate
 1.24% :     0.000002s :     12: predicate.transpose_eliminate
 1.49% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.39% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000002s :     12: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000865    15
46.59% :     0.000403s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 9.86% :     0.000085s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.55% :     0.000377s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017188    72
 0.25% :     0.000043s :      1: add_recomputation
 0.41% :     0.000071s :      1: auto_monad
 0.09% :     0.000015s :      1: auto_monad_reorder
 1.80% :     0.000309s :      1: bootstrap
 0.13% :     0.000022s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.14% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
33.24% :     0.005714s :      1: jit_opt_a
 0.92% :     0.000157s :      1: jit_opt_after_cconv
 0.29% :     0.000050s :      1: jit_opt_b
 2.76% :     0.000475s :      1: loop_unroll
 3.23% :     0.000556s :      1: mutable_eliminate
 5.93% :     0.001020s :     26: opt.transform.jit_opt_a
 0.31% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000039s :      4: opt.transform.symbol_engine_opt
 3.44% :     0.000591s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.37% :     0.000063s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 3.25% :     0.000559s :      1: renormalize.infer
 2.70% :     0.000464s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000033s :      1: rewriter_after_opt_a
 0.48% :     0.000082s :      1: rewriter_before_opt_a
 0.53% :     0.000091s :      1: symbol_engine_optimizer
38.23% :     0.006570s :      1: type_inference


TotalTime = 0.0191655, [33]
[bootstrap]: 0.00029858
[type_inference]: 0.00502228
[event_method]: 1.873e-05
[auto_monad]: 6.299e-05
[graph_reusing]: 5.88002e-06
[pre_auto_parallel]: 2.37999e-06
[py_interpret_to_execute]: 3.082e-05
[rewriter_before_opt_a]: 7.852e-05
[expand_dump_flag]: 2.98998e-06
[jit_opt_a]: 0.00546228, [2]
    [Cycle 1]: 0.00157072, [27]
        [switch_simplify]: 5.349e-05
        [loop_unroll]: 2.934e-05
        [a_1]: 0.00053955
        [with_stream_mark]: 1.988e-05
        [recompute_prepare]: 7.3e-06
        [updatestate_depend_eliminate]: 3.51999e-06
        [updatestate_assign_eliminate]: 2.61999e-06
        [updatestate_loads_eliminate]: 2.55002e-06
        [parameter_eliminate]: 1.74e-06
        [specialize_transform]: 5.74e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 5.64e-06
        [meta_shard_fg_expand]: 1.89999e-06
        [get_grad_eliminate_]: 5.07999e-06
        [merge_forward]: 3.61999e-06
        [cell_reuse_recompute_pass]: 1.14998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.368e-05
        [j_node_and_user_rematch]: 8.45999e-06
        [meta_fg_expand]: 1.95001e-06
        [replace_old_param]: 8.87e-06
        [inline_without_move]: 5.11002e-06
        [renormalize]: 0.00061886
        [add_forward_monad_depend]: 5.15001e-06
        [auto_monad_grad]: 2.73e-06
        [auto_monad_eliminator]: 1.278e-05
        [cse]: 2.829e-05
        [replace_applicator]: 1.118e-05
    [Cycle 2]: 0.00036315, [27]
        [switch_simplify]: 5.84999e-06
        [loop_unroll]: 5.01997e-06
        [a_1]: 0.0001316
        [with_stream_mark]: 9.07999e-06
        [recompute_prepare]: 5.15999e-06
        [updatestate_depend_eliminate]: 2.61e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 9.20001e-07
        [specialize_transform]: 4.91002e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 4.91002e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.48002e-06
        [cell_reuse_recompute_pass]: 1.67999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.243e-05
        [j_node_and_user_rematch]: 7.61999e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 6.76999e-06
        [inline_without_move]: 4.73001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.27999e-06
        [auto_monad_grad]: 1.10999e-06
        [auto_monad_eliminator]: 5.70001e-06
        [cse]: 1.158e-05
        [replace_applicator]: 4.82e-06
[py_interpret_to_execute_after_opt_a]: 9.51998e-06
[rewriter_after_opt_a]: 4.071e-05
[convert_after_rewriter]: 7.1e-06
[order_py_execute_after_rewriter]: 5.12999e-06
[mutable_eliminate]: 0.00051189
[jit_opt_b]: 4.952e-05, [1]
    [Cycle 1]: 4.27e-05, [2]
        [frontend_op_eliminate]: 1.607e-05
        [inline_after_opt_a]: 1.511e-05
[cconv]: 2.511e-05
[loop_unroll]: 0.00041261
[jit_opt_after_cconv]: 0.00013361, [1]
    [Cycle 1]: 0.00012761, [11]
        [c_1]: 1.952e-05
        [parameter_eliminate]: 2.63998e-06
        [updatestate_depend_eliminate]: 5.12e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 1.701e-05
        [call_graph_tuple_transform]: 1.743e-05
        [tuple_list_get_item_eliminator]: 5.02999e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 6.39993e-07
        [switch_simplify]: 5.17e-06
[remove_dup_value]: 1.251e-05
[partial_unused_args_eliminate]: 2.08002e-06
[environ_conv]: 5.19003e-06
[add_recomputation]: 4.706e-05
[cse_after_recomputation]: 2.02e-05, [1]
    [Cycle 1]: 1.44e-05, [1]
        [cse]: 8.60999e-06
[auto_monad_reorder]: 1.456e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 5.77999e-06
[opt_after_jit_grad]: 0.00044449
[symbol_engine_optimizer]: 7.027e-05, [1]
    [Cycle 1]: 6.426e-05, [6]
        [build]: 3.65998e-06
        [elim_shapecalc]: 7.43999e-06
        [elim_not_effective]: 1.231e-05
        [opt_reshape]: 5.40999e-06
        [fold_const_symbol]: 8.40001e-06
        [renormalize]: 7.2e-07
[validate]: 3.42e-05
[backend_pass]: 1.09e-06
[task_emit]: 0.00610461
[execute]: 8.69e-06
Sums
    bootstrap                            :     0.000299s :  1.99%
    type_inference                       :     0.005022s : 33.43%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000063s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.21%
    rewriter_before_opt_a                :     0.000079s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000671s :  4.47%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000619s :  4.12%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000040s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000041s :  0.27%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000512s :  3.41%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000413s :  2.75%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000047s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000444s :  2.96%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006105s : 40.63%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000224    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.34% :     0.000005s :      3: substitution.graph_param_transform
85.78% :     0.000192s :      4: substitution.inline
 1.37% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.91% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000004s :      2: substitution.replace_old_param
 5.29% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004950     2
83.08% :     0.004113s :      1: type_inference.infer
16.92% :     0.000838s :      1: type_inference.specialize
------[replace.]   0.000065     6
68.02% :     0.000044s :      4: replace.inline
31.98% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     6
94.65% :     0.000190s :      4: match.inline
 5.35% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000153   691
 0.95% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.12% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.80% :     0.000001s :     10: predicate.addn_check_dump
 0.97% :     0.000001s :     10: predicate.addn_zero_filter
 1.38% :     0.000002s :     10: predicate.arithmetic_simplify
 1.10% :     0.000002s :     10: predicate.cast_eliminate
 0.35% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.80% :     0.000001s :     10: predicate.compare_switch_simplify
 0.95% :     0.000001s :     10: predicate.depend_value_elim
 0.80% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 0.91% :     0.000001s :     10: predicate.dict_get_item_eliminator
 0.83% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.70% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.25% :     0.000000s :      3: predicate.elim_not_effective
 0.38% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.85% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.80% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.79% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.86% :     0.000001s :     10: predicate.environ_get_eliminate
 0.82% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.17% :     0.000000s :      3: predicate.fold_const_symbol
 0.65% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 4.08% :     0.000006s :     22: predicate.inline
 0.65% :     0.000001s :      6: predicate.inline_without_move
 0.27% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.74% :     0.000001s :      6: predicate.less_batch_normalization
 1.13% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.29% :     0.000002s :     15: predicate.load_eliminater
 1.09% :     0.000002s :      3: predicate.loop_unroll_after_grad
 2.94% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.40% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.82% :     0.000001s :     10: predicate.merge_addn
 0.80% :     0.000001s :     10: predicate.minmaximum_grad
 1.57% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 1.74% :     0.000003s :     15: predicate.partial_eliminate
 0.85% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.01% :     0.000002s :     10: predicate.reduce_eliminate
 1.10% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.27% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.38% :     0.000001s :      3: predicate.reset_defer_inline
 0.99% :     0.000002s :     10: predicate.reshape_eliminate
 0.93% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.60% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.88% :     0.000001s :     10: predicate.same_eliminate
 0.40% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.67% :     0.000001s :      6: predicate.special_op_eliminate
 0.79% :     0.000001s :      6: predicate.specialize_transform
 0.91% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 0.99% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.31% :     0.000000s :      3: predicate.switch_call_monad_eliminater
29.81% :     0.000046s :     16: predicate.switch_defer_inline
 1.57% :     0.000002s :     16: predicate.switch_layer_defer_inline
 5.71% :     0.000009s :     53: predicate.switch_simplify
 0.86% :     0.000001s :     10: predicate.tile_eliminate
 0.87% :     0.000001s :     10: predicate.transpose_eliminate
 0.97% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.86% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 2.75% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.03% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.03% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.26% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.10% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.02% :     0.000002s :     10: predicate.value_based_eliminate
 0.27% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.46% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000633    12
51.65% :     0.000327s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.84% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.51% :     0.000250s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020760    76
 0.24% :     0.000051s :      1: add_recomputation
 0.32% :     0.000067s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.51% :     0.000313s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000023s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.32% :     0.005465s :      1: jit_opt_a
 0.66% :     0.000136s :      1: jit_opt_after_cconv
 0.25% :     0.000052s :      1: jit_opt_b
 2.02% :     0.000420s :      1: loop_unroll
 2.51% :     0.000521s :      1: mutable_eliminate
 4.21% :     0.000873s :     26: opt.transform.jit_opt_a
 0.21% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.18% :     0.000452s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 1.67% :     0.000348s :      1: renormalize.infer
 1.27% :     0.000264s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000046s :      1: rewriter_after_opt_a
 0.39% :     0.000082s :      1: rewriter_before_opt_a
 0.35% :     0.000073s :      1: symbol_engine_optimizer
29.48% :     0.006119s :      1: task_emit
24.27% :     0.005039s :      1: type_inference
 0.32% :     0.000067s :      1: validate


TotalTime = 0.0191956, [33]
[bootstrap]: 0.00032003
[type_inference]: 0.00454273
[event_method]: 1.722e-05
[auto_monad]: 6.555e-05
[graph_reusing]: 6.26e-06
[pre_auto_parallel]: 2.49999e-06
[py_interpret_to_execute]: 3.671e-05
[rewriter_before_opt_a]: 7.315e-05
[expand_dump_flag]: 3.61001e-06
[jit_opt_a]: 0.00537743, [2]
    [Cycle 1]: 0.00164111, [27]
        [switch_simplify]: 5.806e-05
        [loop_unroll]: 3.228e-05
        [a_1]: 0.00054589
        [with_stream_mark]: 2.192e-05
        [recompute_prepare]: 7.38e-06
        [updatestate_depend_eliminate]: 3.97e-06
        [updatestate_assign_eliminate]: 3.16001e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 6.40002e-06
        [updatestate_useless_node_eliminater]: 4.80999e-06
        [accelerated_algorithm]: 5.42999e-06
        [meta_shard_fg_expand]: 2.25002e-06
        [get_grad_eliminate_]: 5.35999e-06
        [merge_forward]: 3.45e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.394e-05
        [j_node_and_user_rematch]: 9.02e-06
        [meta_fg_expand]: 2.39999e-06
        [replace_old_param]: 9.25999e-06
        [inline_without_move]: 5.45001e-06
        [renormalize]: 0.00065994
        [add_forward_monad_depend]: 5.30999e-06
        [auto_monad_grad]: 2.37999e-06
        [auto_monad_eliminator]: 1.296e-05
        [cse]: 2.819e-05
        [replace_applicator]: 1.223e-05
    [Cycle 2]: 0.00032557, [27]
        [switch_simplify]: 5.52999e-06
        [loop_unroll]: 5.07e-06
        [a_1]: 9.159e-05
        [with_stream_mark]: 9.62001e-06
        [recompute_prepare]: 5.02999e-06
        [updatestate_depend_eliminate]: 2.54999e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.26002e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 4.99e-06
        [meta_shard_fg_expand]: 1.07998e-06
        [get_grad_eliminate_]: 4.49002e-06
        [merge_forward]: 2.66999e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.297e-05
        [j_node_and_user_rematch]: 7.86001e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 7.13e-06
        [inline_without_move]: 4.85999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.33002e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 4.92e-06
        [cse]: 9.90002e-06
        [replace_applicator]: 5.22999e-06
[py_interpret_to_execute_after_opt_a]: 1.005e-05
[rewriter_after_opt_a]: 3.327e-05
[convert_after_rewriter]: 6.91001e-06
[order_py_execute_after_rewriter]: 4.91002e-06
[mutable_eliminate]: 0.00055205
[jit_opt_b]: 5.306e-05, [1]
    [Cycle 1]: 4.507e-05, [2]
        [frontend_op_eliminate]: 1.725e-05
        [inline_after_opt_a]: 1.505e-05
[cconv]: 2.622e-05
[loop_unroll]: 0.00050831
[jit_opt_after_cconv]: 0.00015073, [1]
    [Cycle 1]: 0.00014312, [11]
        [c_1]: 2.152e-05
        [parameter_eliminate]: 2.99999e-06
        [updatestate_depend_eliminate]: 6.79999e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 2.302e-05
        [call_graph_tuple_transform]: 1.986e-05
        [tuple_list_get_item_eliminator]: 5.42001e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 6.79982e-07
        [switch_simplify]: 5.81e-06
[remove_dup_value]: 1.352e-05
[partial_unused_args_eliminate]: 2.19001e-06
[environ_conv]: 5.70001e-06
[add_recomputation]: 5.199e-05
[cse_after_recomputation]: 2.284e-05, [1]
    [Cycle 1]: 1.663e-05, [1]
        [cse]: 1.042e-05
[auto_monad_reorder]: 1.544e-05
[get_jit_bprop_graph]: 1.94e-06
[rewriter_after_jit_bprop_graph]: 7e-06
[opt_after_jit_grad]: 0.00049549
[symbol_engine_optimizer]: 7.601e-05, [1]
    [Cycle 1]: 6.816e-05, [6]
        [build]: 3.65e-06
        [elim_shapecalc]: 8e-06
        [elim_not_effective]: 1.355e-05
        [opt_reshape]: 5.69e-06
        [fold_const_symbol]: 8.98002e-06
        [renormalize]: 5.3001e-07
[validate]: 3.865e-05
[backend_pass]: 8.2e-07
[task_emit]: 0.00645557
[execute]: 9.56003e-06
Sums
    bootstrap                            :     0.000320s :  2.11%
    type_inference                       :     0.004543s : 29.97%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000066s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.24%
    rewriter_before_opt_a                :     0.000073s :  0.48%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.42%
    jit_opt_a.loop_unroll                :     0.000037s :  0.25%
    jit_opt_a.a_1                        :     0.000637s :  4.21%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000660s :  4.35%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000038s :  0.25%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000552s :  3.64%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000026s :  0.17%
    loop_unroll                          :     0.000508s :  3.35%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000052s :  0.34%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.05%
    opt_after_jit_grad                   :     0.000495s :  3.27%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006456s : 42.59%
    execute                              :     0.000010s :  0.06%


TotalTime = 0.0241861, [33]
[bootstrap]: 0.0004126
[type_inference]: 0.00663248
[event_method]: 2.301e-05
[auto_monad]: 7.728e-05
[graph_reusing]: 6.74001e-06
[pre_auto_parallel]: 2.86e-06
[py_interpret_to_execute]: 5.343e-05
[rewriter_before_opt_a]: 9.066e-05
[expand_dump_flag]: 4.15e-06
[jit_opt_a]: 0.00700321, [2]
    [Cycle 1]: 0.00205297, [27]
        [switch_simplify]: 7.075e-05
        [loop_unroll]: 2.936e-05
        [a_1]: 0.00068919
        [with_stream_mark]: 2.701e-05
        [recompute_prepare]: 8.32e-06
        [updatestate_depend_eliminate]: 3.80998e-06
        [updatestate_assign_eliminate]: 3.03998e-06
        [updatestate_loads_eliminate]: 2.64999e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 5.96e-06
        [updatestate_useless_node_eliminater]: 4.89998e-06
        [accelerated_algorithm]: 5.57001e-06
        [meta_shard_fg_expand]: 2.89001e-06
        [get_grad_eliminate_]: 5.34998e-06
        [merge_forward]: 4.25999e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.37e-05
        [j_node_and_user_rematch]: 9.02999e-06
        [meta_fg_expand]: 2.16e-06
        [replace_old_param]: 8.60001e-06
        [inline_without_move]: 5.33002e-06
        [renormalize]: 0.00089402
        [add_forward_monad_depend]: 5.35001e-06
        [auto_monad_grad]: 2.65002e-06
        [auto_monad_eliminator]: 1.439e-05
        [cse]: 3.159e-05
        [replace_applicator]: 1.503e-05
    [Cycle 2]: 0.00032754, [27]
        [switch_simplify]: 5.42001e-06
        [loop_unroll]: 5.17999e-06
        [a_1]: 9.07e-05
        [with_stream_mark]: 1.103e-05
        [recompute_prepare]: 5.12e-06
        [updatestate_depend_eliminate]: 2.54999e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.17999e-06
        [specialize_transform]: 4.85001e-06
        [updatestate_useless_node_eliminater]: 4.66002e-06
        [accelerated_algorithm]: 5.06002e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 4.62e-06
        [merge_forward]: 2.57001e-06
        [cell_reuse_recompute_pass]: 1.21002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.41e-05
        [j_node_and_user_rematch]: 7.73999e-06
        [meta_fg_expand]: 1.77001e-06
        [replace_old_param]: 6.58e-06
        [inline_without_move]: 4.48999e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 7.10017e-07
        [auto_monad_eliminator]: 5.58002e-06
        [cse]: 1.162e-05
        [replace_applicator]: 5.44e-06
[py_interpret_to_execute_after_opt_a]: 1.245e-05
[rewriter_after_opt_a]: 3.934e-05
[convert_after_rewriter]: 7.45998e-06
[order_py_execute_after_rewriter]: 4.67e-06
[mutable_eliminate]: 0.00074339
[jit_opt_b]: 5.381e-05, [1]
    [Cycle 1]: 4.524e-05, [2]
        [frontend_op_eliminate]: 1.75e-05
        [inline_after_opt_a]: 1.595e-05
[cconv]: 2.974e-05
[loop_unroll]: 0.00046336
[jit_opt_after_cconv]: 0.00015735, [1]
    [Cycle 1]: 0.00014969, [11]
        [c_1]: 2.239e-05
        [parameter_eliminate]: 3.61001e-06
        [updatestate_depend_eliminate]: 6.44001e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 2.595e-05
        [call_graph_tuple_transform]: 2.208e-05
        [tuple_list_get_item_eliminator]: 5.67999e-06
        [none_parameter_eliminate]: 1.76e-06
        [renormalize]: 9.49978e-07
        [switch_simplify]: 5.61e-06
[remove_dup_value]: 1.687e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 5.27001e-06
[add_recomputation]: 5.517e-05
[cse_after_recomputation]: 2.322e-05, [1]
    [Cycle 1]: 1.664e-05, [1]
        [cse]: 1.002e-05
[auto_monad_reorder]: 1.592e-05
[get_jit_bprop_graph]: 2.48e-06
[rewriter_after_jit_bprop_graph]: 7.95e-06
[opt_after_jit_grad]: 0.00054939
[symbol_engine_optimizer]: 7.964e-05, [1]
    [Cycle 1]: 7.152e-05, [6]
        [build]: 4.74998e-06
        [elim_shapecalc]: 8.20999e-06
        [elim_not_effective]: 1.401e-05
        [opt_reshape]: 5.72999e-06
        [fold_const_symbol]: 1.013e-05
        [renormalize]: 4.19997e-07
[validate]: 4.676e-05
[backend_pass]: 1.00001e-06
[task_emit]: 0.00729505
[execute]: 1.236e-05
Sums
    bootstrap                            :     0.000413s :  2.19%
    type_inference                       :     0.006632s : 35.14%
    event_method                         :     0.000023s :  0.12%
    auto_monad                           :     0.000077s :  0.41%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000053s :  0.28%
    rewriter_before_opt_a                :     0.000091s :  0.48%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.18%
    jit_opt_a.a_1                        :     0.000780s :  4.13%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000894s :  4.74%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.11%
    jit_opt_a.cse                        :     0.000043s :  0.23%
    jit_opt_a.replace_applicator         :     0.000020s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000039s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000743s :  3.94%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.000030s :  0.16%
    loop_unroll                          :     0.000463s :  2.45%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000055s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000549s :  2.91%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000047s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007295s : 38.65%
    execute                              :     0.000012s :  0.07%


Time group info:
------[substitution.]   0.000211    23
 1.11% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000002s :      2: substitution.fold_const_symbol
 2.52% :     0.000005s :      3: substitution.graph_param_transform
84.19% :     0.000177s :      4: substitution.inline
 1.40% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.15% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.98% :     0.000004s :      2: substitution.replace_old_param
 5.91% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004475     2
81.63% :     0.003653s :      1: type_inference.infer
18.37% :     0.000822s :      1: type_inference.specialize
------[replace.]   0.000069     6
66.92% :     0.000046s :      4: replace.inline
33.08% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000186     6
93.95% :     0.000175s :      4: match.inline
 6.05% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.43% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 1.93% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000001s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.74% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.33% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.30% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 6.15% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.37% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.26% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.41% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.27% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000000s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.41% :     0.000000s :      3: predicate.reset_defer_inline
 1.56% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.19% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.16% :     0.000001s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.72% :     0.000009s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.10% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.47% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000560    12
47.50% :     0.000266s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.14% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.36% :     0.000254s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020815    76
 0.26% :     0.000055s :      1: add_recomputation
 0.33% :     0.000069s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.62% :     0.000338s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.85% :     0.005381s :      1: jit_opt_a
 0.74% :     0.000154s :      1: jit_opt_after_cconv
 0.27% :     0.000055s :      1: jit_opt_b
 2.49% :     0.000518s :      1: loop_unroll
 2.71% :     0.000564s :      1: mutable_eliminate
 4.09% :     0.000852s :     26: opt.transform.jit_opt_a
 0.23% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.42% :     0.000504s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000040s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.68% :     0.000351s :      1: renormalize.infer
 1.45% :     0.000302s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000037s :      1: rewriter_after_opt_a
 0.36% :     0.000076s :      1: rewriter_before_opt_a
 0.38% :     0.000079s :      1: symbol_engine_optimizer
31.08% :     0.006469s :      1: task_emit
21.89% :     0.004557s :      1: type_inference
 0.29% :     0.000060s :      1: validate


Time group info:
------[substitution.]   0.000271    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.21% :     0.000006s :      3: substitution.graph_param_transform
87.61% :     0.000237s :      4: substitution.inline
 1.19% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.61% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.37% :     0.000004s :      2: substitution.replace_old_param
 4.78% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006530     2
82.44% :     0.005384s :      1: type_inference.infer
17.56% :     0.001147s :      1: type_inference.specialize
------[replace.]   0.000079     6
67.41% :     0.000053s :      4: replace.inline
32.59% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000246     6
95.17% :     0.000234s :      4: match.inline
 4.83% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.61% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.55% :     0.000002s :     10: predicate.addn_zero_filter
 2.01% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.16% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.73% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.55% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.14% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.48% :     0.000007s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.78% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.37% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.21% :     0.000001s :     10: predicate.minmaximum_grad
 2.22% :     0.000003s :      3: predicate.mutable_eliminate
 0.40% :     0.000000s :      3: predicate.opt_reshape
 2.06% :     0.000002s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.53% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.93% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.38% :     0.000009s :     53: predicate.switch_simplify
 1.31% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.26% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.08% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.05% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.89% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.15% :     0.000003s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.03% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000828    12
49.87% :     0.000413s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.97% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.16% :     0.000357s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026191    76
 0.22% :     0.000058s :      1: add_recomputation
 0.31% :     0.000082s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.66% :     0.000436s :      1: bootstrap
 0.12% :     0.000033s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.11% :     0.000028s :      1: event_method
 0.07% :     0.000019s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
26.76% :     0.007008s :      1: jit_opt_a
 0.61% :     0.000160s :      1: jit_opt_after_cconv
 0.22% :     0.000057s :      1: jit_opt_b
 1.80% :     0.000472s :      1: loop_unroll
 2.88% :     0.000754s :      1: mutable_eliminate
 3.84% :     0.001006s :     26: opt.transform.jit_opt_a
 0.20% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000027s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.14% :     0.000559s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000057s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 1.87% :     0.000491s :      1: renormalize.infer
 1.51% :     0.000395s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000044s :      1: rewriter_after_opt_a
 0.36% :     0.000094s :      1: rewriter_before_opt_a
 0.31% :     0.000082s :      1: symbol_engine_optimizer
27.94% :     0.007319s :      1: task_emit
25.42% :     0.006657s :      1: type_inference
 0.27% :     0.000070s :      1: validate


TotalTime = 0.018713, [33]
[bootstrap]: 0.00027721
[type_inference]: 0.0044368
[event_method]: 1.612e-05
[auto_monad]: 6.249e-05
[graph_reusing]: 5.89999e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 2.977e-05
[rewriter_before_opt_a]: 6.941e-05
[expand_dump_flag]: 3.53999e-06
[jit_opt_a]: 0.00531849, [2]
    [Cycle 1]: 0.00161399, [27]
        [switch_simplify]: 5.29e-05
        [loop_unroll]: 2.811e-05
        [a_1]: 0.00053544
        [with_stream_mark]: 1.905e-05
        [recompute_prepare]: 7.48e-06
        [updatestate_depend_eliminate]: 3.52002e-06
        [updatestate_assign_eliminate]: 3.13e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 5.88002e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 5.03002e-06
        [merge_forward]: 3.33e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.378e-05
        [j_node_and_user_rematch]: 8.45999e-06
        [meta_fg_expand]: 2.26998e-06
        [replace_old_param]: 7.88999e-06
        [inline_without_move]: 4.90001e-06
        [renormalize]: 0.00066553
        [add_forward_monad_depend]: 5.86e-06
        [auto_monad_grad]: 2.69001e-06
        [auto_monad_eliminator]: 1.445e-05
        [cse]: 2.831e-05
        [replace_applicator]: 1.289e-05
    [Cycle 2]: 0.00031274, [27]
        [switch_simplify]: 5.48002e-06
        [loop_unroll]: 4.79002e-06
        [a_1]: 8.786e-05
        [with_stream_mark]: 9.99001e-06
        [recompute_prepare]: 4.62998e-06
        [updatestate_depend_eliminate]: 2.58003e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 4.55001e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 4.90001e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.47998e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 2.03997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.266e-05
        [j_node_and_user_rematch]: 8.18999e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 6.79001e-06
        [inline_without_move]: 4.50001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 5.10001e-06
        [cse]: 9.93998e-06
        [replace_applicator]: 4.82e-06
[py_interpret_to_execute_after_opt_a]: 1.069e-05
[rewriter_after_opt_a]: 5.327e-05
[convert_after_rewriter]: 7.11999e-06
[order_py_execute_after_rewriter]: 5.03002e-06
[mutable_eliminate]: 0.00057085
[jit_opt_b]: 4.811e-05, [1]
    [Cycle 1]: 4.093e-05, [2]
        [frontend_op_eliminate]: 1.563e-05
        [inline_after_opt_a]: 1.425e-05
[cconv]: 2.486e-05
[loop_unroll]: 0.00041685
[jit_opt_after_cconv]: 0.00024028, [1]
    [Cycle 1]: 0.00023343, [11]
        [c_1]: 2.002e-05
        [parameter_eliminate]: 2.51e-06
        [updatestate_depend_eliminate]: 6.44999e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 1.794e-05
        [call_graph_tuple_transform]: 2.122e-05
        [tuple_list_get_item_eliminator]: 6.02001e-06
        [none_parameter_eliminate]: 2.19999e-06
        [renormalize]: 3.59985e-07
        [switch_simplify]: 5.12e-06
[remove_dup_value]: 1.436e-05
[partial_unused_args_eliminate]: 2.59999e-06
[environ_conv]: 6.02001e-06
[add_recomputation]: 4.988e-05
[cse_after_recomputation]: 2.282e-05, [1]
    [Cycle 1]: 1.665e-05, [1]
        [cse]: 1.086e-05
[auto_monad_reorder]: 1.462e-05
[get_jit_bprop_graph]: 2.14e-06
[rewriter_after_jit_bprop_graph]: 6.05002e-06
[opt_after_jit_grad]: 0.00048847
[symbol_engine_optimizer]: 7.189e-05, [1]
    [Cycle 1]: 6.546e-05, [6]
        [build]: 3.88001e-06
        [elim_shapecalc]: 7.46001e-06
        [elim_not_effective]: 1.181e-05
        [opt_reshape]: 5.62001e-06
        [fold_const_symbol]: 8.38999e-06
        [renormalize]: 9.79984e-07
[validate]: 3.598e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00619801
[execute]: 8.32e-06
Sums
    bootstrap                            :     0.000277s :  1.89%
    type_inference                       :     0.004437s : 30.32%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000062s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.20%
    rewriter_before_opt_a                :     0.000069s :  0.47%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.22%
    jit_opt_a.a_1                        :     0.000623s :  4.26%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000009s :  0.06%
    jit_opt_a.renormalize                :     0.000666s :  4.55%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000038s :  0.26%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000053s :  0.36%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000571s :  3.90%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000417s :  2.85%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000050s :  0.34%
    cse_after_recomputation.cse          :     0.000011s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000488s :  3.34%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000036s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006198s : 42.35%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000212    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.80% :     0.000006s :      3: substitution.graph_param_transform
84.33% :     0.000179s :      4: substitution.inline
 1.70% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.07% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.66% :     0.000004s :      2: substitution.replace_old_param
 5.98% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004372     2
82.57% :     0.003610s :      1: type_inference.infer
17.43% :     0.000762s :      1: type_inference.specialize
------[replace.]   0.000064     6
68.99% :     0.000044s :      4: replace.inline
31.01% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000188     6
93.83% :     0.000177s :      4: match.inline
 6.17% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.28% :     0.000001s :     10: predicate.accumulaten_eliminater
 2.08% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 1.85% :     0.000002s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.42% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.19% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.42% :     0.000002s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.77% :     0.000007s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      6: predicate.less_batch_normalization
 1.36% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.60% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.17% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 2.14% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000000s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.94% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.49% :     0.000001s :      3: predicate.reset_defer_inline
 1.27% :     0.000001s :     10: predicate.reshape_eliminate
 1.34% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.84% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.22% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.32% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.57% :     0.000009s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.26% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.09% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.44% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.91% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000548    12
47.98% :     0.000263s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.06% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.96% :     0.000241s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020307    76
 0.26% :     0.000053s :      1: add_recomputation
 0.32% :     0.000066s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.42% :     0.000289s :      1: bootstrap
 0.13% :     0.000027s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.21% :     0.005321s :      1: jit_opt_a
 1.20% :     0.000243s :      1: jit_opt_after_cconv
 0.25% :     0.000051s :      1: jit_opt_b
 2.09% :     0.000425s :      1: loop_unroll
 2.85% :     0.000579s :      1: mutable_eliminate
 4.06% :     0.000824s :     26: opt.transform.jit_opt_a
 0.24% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.44% :     0.000496s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.80% :     0.000366s :      1: renormalize.infer
 1.43% :     0.000291s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000057s :      1: rewriter_after_opt_a
 0.35% :     0.000072s :      1: rewriter_before_opt_a
 0.37% :     0.000074s :      1: symbol_engine_optimizer
30.59% :     0.006213s :      1: task_emit
21.92% :     0.004451s :      1: type_inference
 0.28% :     0.000057s :      1: validate


TotalTime = 0.0185274, [33]
[bootstrap]: 0.00033651
[type_inference]: 0.00446826
[event_method]: 1.692e-05
[auto_monad]: 6.581e-05
[graph_reusing]: 5.91998e-06
[pre_auto_parallel]: 2.80997e-06
[py_interpret_to_execute]: 3.564e-05
[rewriter_before_opt_a]: 7.503e-05
[expand_dump_flag]: 3.04999e-06
[jit_opt_a]: 0.00540881, [2]
    [Cycle 1]: 0.00160849, [27]
        [switch_simplify]: 5.573e-05
        [loop_unroll]: 3.146e-05
        [a_1]: 0.00052712
        [with_stream_mark]: 1.757e-05
        [recompute_prepare]: 7.82002e-06
        [updatestate_depend_eliminate]: 3.36999e-06
        [updatestate_assign_eliminate]: 3.38e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [parameter_eliminate]: 1.78002e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.85002e-06
        [meta_shard_fg_expand]: 2.27999e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 3.65e-06
        [cell_reuse_recompute_pass]: 1.00999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.466e-05
        [j_node_and_user_rematch]: 9.91e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 8.68001e-06
        [inline_without_move]: 5.88002e-06
        [renormalize]: 0.00066414
        [add_forward_monad_depend]: 5.34e-06
        [auto_monad_grad]: 2.99999e-06
        [auto_monad_eliminator]: 1.31e-05
        [cse]: 2.271e-05
        [replace_applicator]: 1.267e-05
    [Cycle 2]: 0.00032925, [27]
        [switch_simplify]: 6.34999e-06
        [loop_unroll]: 5.06002e-06
        [a_1]: 9.116e-05
        [with_stream_mark]: 9.46998e-06
        [recompute_prepare]: 4.95999e-06
        [updatestate_depend_eliminate]: 2.54001e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 5.28002e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 5.15001e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 4.55999e-06
        [merge_forward]: 2.49001e-06
        [cell_reuse_recompute_pass]: 2.24001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.276e-05
        [j_node_and_user_rematch]: 7.82e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 7.48e-06
        [inline_without_move]: 4.80001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.50001e-06
        [auto_monad_grad]: 1.05999e-06
        [auto_monad_eliminator]: 5.46e-06
        [cse]: 1.16e-05
        [replace_applicator]: 5.17e-06
[py_interpret_to_execute_after_opt_a]: 9.73998e-06
[rewriter_after_opt_a]: 3.358e-05
[convert_after_rewriter]: 7.52998e-06
[order_py_execute_after_rewriter]: 8.32e-06
[mutable_eliminate]: 0.00053218
[jit_opt_b]: 5.05e-05, [1]
    [Cycle 1]: 4.364e-05, [2]
        [frontend_op_eliminate]: 1.675e-05
        [inline_after_opt_a]: 1.521e-05
[cconv]: 2.612e-05
[loop_unroll]: 0.00042402
[jit_opt_after_cconv]: 0.00013967, [1]
    [Cycle 1]: 0.00013293, [11]
        [c_1]: 2.184e-05
        [parameter_eliminate]: 2.54999e-06
        [updatestate_depend_eliminate]: 5.37999e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 1.769e-05
        [call_graph_tuple_transform]: 1.818e-05
        [tuple_list_get_item_eliminator]: 5.54e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 6.80011e-07
        [switch_simplify]: 5.54998e-06
[remove_dup_value]: 1.28e-05
[partial_unused_args_eliminate]: 2.23002e-06
[environ_conv]: 5.40999e-06
[add_recomputation]: 4.814e-05
[cse_after_recomputation]: 2.056e-05, [1]
    [Cycle 1]: 1.508e-05, [1]
        [cse]: 9.60001e-06
[auto_monad_reorder]: 1.539e-05
[get_jit_bprop_graph]: 1.94e-06
[rewriter_after_jit_bprop_graph]: 5.24998e-06
[opt_after_jit_grad]: 0.00051573
[symbol_engine_optimizer]: 7.333e-05, [1]
    [Cycle 1]: 6.671e-05, [6]
        [build]: 3.21999e-06
        [elim_shapecalc]: 7.68999e-06
        [elim_not_effective]: 1.298e-05
        [opt_reshape]: 5.82999e-06
        [fold_const_symbol]: 8.81997e-06
        [renormalize]: 5.10016e-07
[validate]: 3.496e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00593404
[execute]: 7.08998e-06
Sums
    bootstrap                            :     0.000337s :  2.33%
    type_inference                       :     0.004468s : 30.92%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000066s :  0.46%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.25%
    rewriter_before_opt_a                :     0.000075s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.43%
    jit_opt_a.loop_unroll                :     0.000037s :  0.25%
    jit_opt_a.a_1                        :     0.000618s :  4.28%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000664s :  4.60%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000034s :  0.24%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000008s :  0.06%
    mutable_eliminate                    :     0.000532s :  3.68%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000026s :  0.18%
    loop_unroll                          :     0.000424s :  2.93%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000048s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000516s :  3.57%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005934s : 41.06%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000193    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.86% :     0.000002s :      2: substitution.fold_const_symbol
 2.55% :     0.000005s :      3: substitution.graph_param_transform
82.84% :     0.000160s :      4: substitution.inline
 1.86% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.35% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000004s :      2: substitution.replace_old_param
 6.61% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004400     2
78.91% :     0.003471s :      1: type_inference.infer
21.09% :     0.000928s :      1: type_inference.specialize
------[replace.]   0.000068     6
65.25% :     0.000044s :      4: replace.inline
34.75% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000169     6
93.14% :     0.000157s :      4: match.inline
 6.86% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.39% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.56% :     0.000002s :     10: predicate.addn_zero_filter
 1.96% :     0.000002s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.19% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.19% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000000s :      3: predicate.graph_param_transform
 5.30% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.68% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.22% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.22% :     0.000001s :     10: predicate.minmaximum_grad
 1.47% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      6: predicate.special_op_eliminate
 1.13% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.99% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.62% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.13% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.66% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000581    12
45.25% :     0.000263s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.54% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.22% :     0.000286s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020128    76
 0.25% :     0.000051s :      1: add_recomputation
 0.34% :     0.000069s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.75% :     0.000353s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.89% :     0.005413s :      1: jit_opt_a
 0.71% :     0.000142s :      1: jit_opt_after_cconv
 0.26% :     0.000053s :      1: jit_opt_b
 2.15% :     0.000433s :      1: loop_unroll
 2.69% :     0.000541s :      1: mutable_eliminate
 4.13% :     0.000832s :     26: opt.transform.jit_opt_a
 0.24% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.61% :     0.000525s :      1: opt_after_jit_grad
 0.05% :     0.000011s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000038s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.59% :     0.000319s :      1: renormalize.infer
 1.67% :     0.000337s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000037s :      1: rewriter_after_opt_a
 0.39% :     0.000078s :      1: rewriter_before_opt_a
 0.38% :     0.000076s :      1: symbol_engine_optimizer
29.55% :     0.005948s :      1: task_emit
22.27% :     0.004482s :      1: type_inference
 0.27% :     0.000053s :      1: validate


TotalTime = 0.0171314, [30]
[bootstrap]: 0.0003085
[type_inference]: 0.00750523
[event_method]: 2.262e-05
[auto_monad]: 8.245e-05
[graph_reusing]: 6.10002e-06
[pre_auto_parallel]: 2.78003e-06
[py_interpret_to_execute]: 6.9e-05
[rewriter_before_opt_a]: 8.739e-05
[expand_dump_flag]: 3.00998e-06
[jit_opt_a]: 0.00649149, [2]
    [Cycle 1]: 0.00239147, [27]
        [switch_simplify]: 0.00012994
        [loop_unroll]: 3.535e-05
        [a_1]: 0.00077478
        [with_stream_mark]: 2.281e-05
        [recompute_prepare]: 8.08001e-06
        [updatestate_depend_eliminate]: 3.85e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 2.24001e-06
        [specialize_transform]: 6.38e-06
        [updatestate_useless_node_eliminater]: 5.46e-06
        [accelerated_algorithm]: 5.69e-06
        [meta_shard_fg_expand]: 2.09999e-06
        [get_grad_eliminate_]: 6.43998e-06
        [merge_forward]: 3.85e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.856e-05
        [j_node_and_user_rematch]: 1.053e-05
        [meta_fg_expand]: 2.36e-06
        [replace_old_param]: 1.085e-05
        [inline_without_move]: 6.51e-06
        [renormalize]: 0.00107085
        [add_forward_monad_depend]: 6.23998e-06
        [auto_monad_grad]: 2.60002e-06
        [auto_monad_eliminator]: 1.652e-05
        [cse]: 2.423e-05
        [replace_applicator]: 1.472e-05
    [Cycle 2]: 0.00033164, [27]
        [switch_simplify]: 5.97001e-06
        [loop_unroll]: 5.34e-06
        [a_1]: 9.547e-05
        [with_stream_mark]: 1.083e-05
        [recompute_prepare]: 4.91002e-06
        [updatestate_depend_eliminate]: 2.66999e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 4.70001e-06
        [updatestate_useless_node_eliminater]: 5.61e-06
        [accelerated_algorithm]: 5.19e-06
        [meta_shard_fg_expand]: 2.00002e-06
        [get_grad_eliminate_]: 4.60001e-06
        [merge_forward]: 2.83e-06
        [cell_reuse_recompute_pass]: 2.20002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.572e-05
        [j_node_and_user_rematch]: 8.96002e-06
        [meta_fg_expand]: 1.87999e-06
        [replace_old_param]: 7.99002e-06
        [inline_without_move]: 5.22999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 5.00999e-06
        [cse]: 1.008e-05
        [replace_applicator]: 5.11997e-06
[py_interpret_to_execute_after_opt_a]: 1.246e-05
[rewriter_after_opt_a]: 3.579e-05
[convert_after_rewriter]: 6.98998e-06
[order_py_execute_after_rewriter]: 5.29998e-06
[mutable_eliminate]: 0.00070991
[jit_opt_b]: 6.005e-05, [1]
    [Cycle 1]: 5.024e-05, [2]
        [frontend_op_eliminate]: 2.005e-05
        [inline_after_opt_a]: 1.654e-05
[cconv]: 2.928e-05
[loop_unroll]: 0.00054144
[jit_opt_after_cconv]: 0.00016464, [1]
    [Cycle 1]: 0.00015586, [11]
        [c_1]: 2.636e-05
        [parameter_eliminate]: 3.7e-06
        [updatestate_depend_eliminate]: 7.8e-06
        [updatestate_assign_eliminate]: 3.12002e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [cse]: 2.243e-05
        [call_graph_tuple_transform]: 2.166e-05
        [tuple_list_get_item_eliminator]: 5.30001e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 6.33e-06
[remove_dup_value]: 1.41e-05
[partial_unused_args_eliminate]: 2.39001e-06
[environ_conv]: 6.09001e-06
[add_recomputation]: 5.279e-05
[cse_after_recomputation]: 2.267e-05, [1]
    [Cycle 1]: 1.739e-05, [1]
        [cse]: 1.104e-05
[auto_monad_reorder]: 1.508e-05
[get_jit_bprop_graph]: 1.97999e-06
[rewriter_after_jit_bprop_graph]: 8.37e-06
[opt_after_jit_grad]: 0.00054066
[symbol_engine_optimizer]: 8.107e-05, [1]
    [Cycle 1]: 7.281e-05, [6]
        [build]: 4.43999e-06
        [elim_shapecalc]: 8.47e-06
        [elim_not_effective]: 1.39e-05
        [opt_reshape]: 6.01003e-06
        [fold_const_symbol]: 8.82e-06
        [renormalize]: 6.80011e-07
[validate]: 4.013e-05
Sums
    bootstrap                            :     0.000308s :  2.43%
    type_inference                       :     0.007505s : 59.01%
    event_method                         :     0.000023s :  0.18%
    auto_monad                           :     0.000082s :  0.65%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000069s :  0.54%
    rewriter_before_opt_a                :     0.000087s :  0.69%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000136s :  1.07%
    jit_opt_a.loop_unroll                :     0.000041s :  0.32%
    jit_opt_a.a_1                        :     0.000870s :  6.84%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.15%
    jit_opt_a.inline_without_move        :     0.000012s :  0.09%
    jit_opt_a.renormalize                :     0.001071s :  8.42%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.17%
    jit_opt_a.cse                        :     0.000034s :  0.27%
    jit_opt_a.replace_applicator         :     0.000020s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.10%
    rewriter_after_opt_a                 :     0.000036s :  0.28%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000710s :  5.58%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.13%
    cconv                                :     0.000029s :  0.23%
    loop_unroll                          :     0.000541s :  4.26%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000014s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000053s :  0.42%
    cse_after_recomputation.cse          :     0.000011s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.07%
    opt_after_jit_grad                   :     0.000541s :  4.25%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.32%


Time group info:
------[substitution.]   0.000340    26
 0.55% :     0.000002s :      2: substitution.elim_not_effective
 0.37% :     0.000001s :      2: substitution.fold_const_symbol
 1.65% :     0.000006s :      3: substitution.graph_param_transform
86.79% :     0.000295s :      6: substitution.inline
 1.36% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.63% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.38% :     0.000005s :      2: substitution.replace_old_param
 2.35% :     0.000008s :      1: substitution.switch_simplify
 3.91% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007427     2
81.36% :     0.006043s :      1: type_inference.infer
18.64% :     0.001384s :      1: type_inference.specialize
------[replace.]   0.000149     9
50.69% :     0.000076s :      6: replace.inline
24.89% :     0.000037s :      1: replace.switch_simplify
24.42% :     0.000036s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000311     9
93.78% :     0.000292s :      6: match.inline
 2.31% :     0.000007s :      1: match.switch_simplify
 3.90% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000145   801
 1.48% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.59% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     12: predicate.addn_check_dump
 1.60% :     0.000002s :     12: predicate.addn_zero_filter
 2.09% :     0.000003s :     12: predicate.arithmetic_simplify
 1.29% :     0.000002s :     12: predicate.cast_eliminate
 0.62% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     12: predicate.compare_switch_simplify
 1.38% :     0.000002s :     12: predicate.depend_value_elim
 1.22% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.14% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.82% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.12% :     0.000002s :     12: predicate.environ_get_eliminate
 1.05% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 6.18% :     0.000009s :     26: predicate.inline
 0.77% :     0.000001s :      6: predicate.inline_without_move
 0.29% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.93% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.81% :     0.000003s :     17: predicate.load_eliminater
 1.91% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.29% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.54% :     0.000004s :     15: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     12: predicate.merge_addn
 1.13% :     0.000002s :     12: predicate.minmaximum_grad
 2.18% :     0.000003s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 1.95% :     0.000003s :     17: predicate.partial_eliminate
 1.11% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     12: predicate.reduce_eliminate
 1.33% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.53% :     0.000002s :     20: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.45% :     0.000001s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     12: predicate.reshape_eliminate
 1.17% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.19% :     0.000002s :      6: predicate.special_op_eliminate
 1.05% :     0.000002s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.56% :     0.000004s :     20: predicate.switch_defer_inline
 2.17% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.13% :     0.000012s :     65: predicate.switch_simplify
 1.15% :     0.000002s :     12: predicate.tile_eliminate
 1.16% :     0.000002s :     12: predicate.transpose_eliminate
 1.33% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.66% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001042    15
45.13% :     0.000470s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.70% :     0.000132s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.17% :     0.000439s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019452    72
 0.29% :     0.000057s :      1: add_recomputation
 0.44% :     0.000086s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 1.68% :     0.000327s :      1: bootstrap
 0.16% :     0.000032s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.14% :     0.000028s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
33.39% :     0.006496s :      1: jit_opt_a
 0.87% :     0.000168s :      1: jit_opt_after_cconv
 0.32% :     0.000063s :      1: jit_opt_b
 2.85% :     0.000554s :      1: loop_unroll
 3.72% :     0.000724s :      1: mutable_eliminate
 6.05% :     0.001177s :     26: opt.transform.jit_opt_a
 0.29% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000030s :      4: opt.transform.jit_opt_b
 0.08% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.84% :     0.000552s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.37% :     0.000072s :      1: py_interpret_to_execute
 0.08% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000017s :      1: remove_dup_value
 3.25% :     0.000633s :      1: renormalize.infer
 2.20% :     0.000428s :      1: renormalize.specialize
 0.05% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000039s :      1: rewriter_after_opt_a
 0.46% :     0.000090s :      1: rewriter_before_opt_a
 0.43% :     0.000084s :      1: symbol_engine_optimizer
38.67% :     0.007521s :      1: type_inference


TotalTime = 0.0328457, [33]
[bootstrap]: 0.00041578
[type_inference]: 0.00792447
[event_method]: 2.278e-05
[auto_monad]: 7.089e-05
[graph_reusing]: 6.24001e-06
[pre_auto_parallel]: 2.73e-06
[py_interpret_to_execute]: 4.909e-05
[rewriter_before_opt_a]: 7.716e-05
[expand_dump_flag]: 3.28e-06
[jit_opt_a]: 0.0121843, [2]
    [Cycle 1]: 0.00409634, [27]
        [switch_simplify]: 6.765e-05
        [loop_unroll]: 2.955e-05
        [a_1]: 0.0006216
        [with_stream_mark]: 2.57e-05
        [recompute_prepare]: 1.017e-05
        [updatestate_depend_eliminate]: 4.52e-06
        [updatestate_assign_eliminate]: 3.13e-06
        [updatestate_loads_eliminate]: 2.58998e-06
        [parameter_eliminate]: 1.81003e-06
        [specialize_transform]: 6.38e-06
        [updatestate_useless_node_eliminater]: 5.12999e-06
        [accelerated_algorithm]: 6.31e-06
        [meta_shard_fg_expand]: 2.77002e-06
        [get_grad_eliminate_]: 6.03998e-06
        [merge_forward]: 4.57e-06
        [cell_reuse_recompute_pass]: 1.55001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.588e-05
        [j_node_and_user_rematch]: 1.012e-05
        [meta_fg_expand]: 2.32999e-06
        [replace_old_param]: 1.16e-05
        [inline_without_move]: 5.56e-06
        [renormalize]: 0.00291872
        [add_forward_monad_depend]: 1.121e-05
        [auto_monad_grad]: 2.91e-06
        [auto_monad_eliminator]: 2.366e-05
        [cse]: 3.464e-05
        [replace_applicator]: 2.809e-05
    [Cycle 2]: 0.00038767, [27]
        [switch_simplify]: 6.54999e-06
        [loop_unroll]: 5.20999e-06
        [a_1]: 0.00010902
        [with_stream_mark]: 1.721e-05
        [recompute_prepare]: 5.87999e-06
        [updatestate_depend_eliminate]: 3.39001e-06
        [updatestate_assign_eliminate]: 2.86999e-06
        [updatestate_loads_eliminate]: 2.43002e-06
        [parameter_eliminate]: 1.93002e-06
        [specialize_transform]: 5.19e-06
        [updatestate_useless_node_eliminater]: 5.47001e-06
        [accelerated_algorithm]: 6.76999e-06
        [meta_shard_fg_expand]: 2.39001e-06
        [get_grad_eliminate_]: 5.77001e-06
        [merge_forward]: 3.57002e-06
        [cell_reuse_recompute_pass]: 4.89e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.085e-05
        [j_node_and_user_rematch]: 1.032e-05
        [meta_fg_expand]: 2.59001e-06
        [replace_old_param]: 9.35001e-06
        [inline_without_move]: 5.19998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.44998e-06
        [auto_monad_grad]: 9.49978e-07
        [auto_monad_eliminator]: 6.21998e-06
        [cse]: 1.22e-05
        [replace_applicator]: 5.12999e-06
[py_interpret_to_execute_after_opt_a]: 1.983e-05
[rewriter_after_opt_a]: 4.246e-05
[convert_after_rewriter]: 7.46999e-06
[order_py_execute_after_rewriter]: 5.04e-06
[mutable_eliminate]: 0.00093759
[jit_opt_b]: 6.005e-05, [1]
    [Cycle 1]: 5.027e-05, [2]
        [frontend_op_eliminate]: 1.916e-05
        [inline_after_opt_a]: 1.874e-05
[cconv]: 3.368e-05
[loop_unroll]: 0.00051787
[jit_opt_after_cconv]: 0.00017716, [1]
    [Cycle 1]: 0.00016895, [11]
        [c_1]: 2.35e-05
        [parameter_eliminate]: 4.3e-06
        [updatestate_depend_eliminate]: 8.83001e-06
        [updatestate_assign_eliminate]: 3.09001e-06
        [updatestate_loads_eliminate]: 2.94999e-06
        [cse]: 3.51e-05
        [call_graph_tuple_transform]: 2.471e-05
        [tuple_list_get_item_eliminator]: 6.13002e-06
        [none_parameter_eliminate]: 1.55001e-06
        [renormalize]: 1.57001e-06
        [switch_simplify]: 6.11e-06
[remove_dup_value]: 1.687e-05
[partial_unused_args_eliminate]: 2.31e-06
[environ_conv]: 5.75001e-06
[add_recomputation]: 6.216e-05
[cse_after_recomputation]: 2.185e-05, [1]
    [Cycle 1]: 1.575e-05, [1]
        [cse]: 9.39e-06
[auto_monad_reorder]: 1.633e-05
[get_jit_bprop_graph]: 2.57001e-06
[rewriter_after_jit_bprop_graph]: 1.063e-05
[opt_after_jit_grad]: 0.00054461
[symbol_engine_optimizer]: 8.237e-05, [1]
    [Cycle 1]: 7.444e-05, [6]
        [build]: 5.84e-06
        [elim_shapecalc]: 8.87e-06
        [elim_not_effective]: 1.489e-05
        [opt_reshape]: 6.12001e-06
        [fold_const_symbol]: 9.77999e-06
        [renormalize]: 1.12999e-06
[validate]: 4.953e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00920862
[execute]: 1.046e-05
Sums
    bootstrap                            :     0.000416s :  1.70%
    type_inference                       :     0.007924s : 32.47%
    event_method                         :     0.000023s :  0.09%
    auto_monad                           :     0.000071s :  0.29%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000049s :  0.20%
    rewriter_before_opt_a                :     0.000077s :  0.32%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000074s :  0.30%
    jit_opt_a.loop_unroll                :     0.000035s :  0.14%
    jit_opt_a.a_1                        :     0.000731s :  2.99%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.04%
    jit_opt_a.renormalize                :     0.002919s : 11.96%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000030s :  0.12%
    jit_opt_a.cse                        :     0.000047s :  0.19%
    jit_opt_a.replace_applicator         :     0.000033s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000020s :  0.08%
    rewriter_after_opt_a                 :     0.000042s :  0.17%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000938s :  3.84%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.08%
    cconv                                :     0.000034s :  0.14%
    loop_unroll                          :     0.000518s :  2.12%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000035s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000002s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000062s :  0.25%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.07%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.04%
    opt_after_jit_grad                   :     0.000545s :  2.23%
    symbol_engine_optimizer.build        :     0.000006s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000050s :  0.20%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009209s : 37.73%
    execute                              :     0.000010s :  0.04%


Time group info:
------[substitution.]   0.000265    23
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000002s :      2: substitution.fold_const_symbol
 2.56% :     0.000007s :      3: substitution.graph_param_transform
83.37% :     0.000221s :      4: substitution.inline
 2.03% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.27% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.84% :     0.000008s :      2: substitution.replace_old_param
 5.39% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007832     2
85.44% :     0.006692s :      1: type_inference.infer
14.56% :     0.001140s :      1: type_inference.specialize
------[replace.]   0.000084     6
66.56% :     0.000056s :      4: replace.inline
33.44% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000231     6
94.27% :     0.000218s :      4: match.inline
 5.73% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.65% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 3.28% :     0.000004s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.92% :     0.000001s :     10: predicate.compare_switch_simplify
 1.18% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.56% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.21% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.83% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.96% :     0.000001s :     10: predicate.environ_get_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.96% :     0.000008s :     22: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.68% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.68% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.55% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.91% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 2.15% :     0.000003s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 1.96% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.86% :     0.000001s :      6: predicate.replace_old_param
 0.66% :     0.000001s :      3: predicate.reset_defer_inline
 1.63% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.32% :     0.000002s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.63% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 1.81% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.82% :     0.000009s :     53: predicate.switch_simplify
 1.31% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.96% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.17% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.93% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.39% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.23% :     0.000003s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000898    12
46.60% :     0.000418s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.67% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.73% :     0.000437s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.036864    76
 0.18% :     0.000066s :      1: add_recomputation
 0.20% :     0.000075s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.18% :     0.000437s :      1: bootstrap
 0.10% :     0.000037s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.07% :     0.000027s :      1: event_method
 0.04% :     0.000016s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
33.06% :     0.012189s :      1: jit_opt_a
 0.49% :     0.000180s :      1: jit_opt_after_cconv
 0.17% :     0.000063s :      1: jit_opt_b
 1.43% :     0.000527s :      1: loop_unroll
 2.58% :     0.000952s :      1: mutable_eliminate
 2.68% :     0.000989s :     26: opt.transform.jit_opt_a
 0.15% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000029s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.50% :     0.000554s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000052s :      1: py_interpret_to_execute
 0.06% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 6.18% :     0.002279s :      1: renormalize.infer
 1.69% :     0.000622s :      1: renormalize.specialize
 0.03% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000046s :      1: rewriter_after_opt_a
 0.22% :     0.000080s :      1: rewriter_before_opt_a
 0.23% :     0.000085s :      1: symbol_engine_optimizer
25.04% :     0.009231s :      1: task_emit
21.56% :     0.007948s :      1: type_inference
 0.20% :     0.000072s :      1: validate


TotalTime = 0.0208542, [33]
[bootstrap]: 0.00029344
[type_inference]: 0.00459693
[event_method]: 1.682e-05
[auto_monad]: 6.471e-05
[graph_reusing]: 5.91e-06
[pre_auto_parallel]: 2.69001e-06
[py_interpret_to_execute]: 3.124e-05
[rewriter_before_opt_a]: 7.112e-05
[expand_dump_flag]: 3.02002e-06
[jit_opt_a]: 0.00523844, [2]
    [Cycle 1]: 0.00154008, [27]
        [switch_simplify]: 5.345e-05
        [loop_unroll]: 2.852e-05
        [a_1]: 0.00049198
        [with_stream_mark]: 1.815e-05
        [recompute_prepare]: 7.16001e-06
        [updatestate_depend_eliminate]: 3.14999e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.69001e-06
        [parameter_eliminate]: 1.88002e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 5.32001e-06
        [meta_shard_fg_expand]: 1.511e-05
        [get_grad_eliminate_]: 6.12999e-06
        [merge_forward]: 4.15e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.456e-05
        [j_node_and_user_rematch]: 9.01002e-06
        [meta_fg_expand]: 2.17999e-06
        [replace_old_param]: 9.68002e-06
        [inline_without_move]: 5.15999e-06
        [renormalize]: 0.00057561
        [add_forward_monad_depend]: 6.61e-06
        [auto_monad_grad]: 2.47001e-06
        [auto_monad_eliminator]: 1.368e-05
        [cse]: 2.501e-05
        [replace_applicator]: 1.518e-05
    [Cycle 2]: 0.00033534, [27]
        [switch_simplify]: 5.32001e-06
        [loop_unroll]: 5.02e-06
        [a_1]: 9.734e-05
        [with_stream_mark]: 1.03e-05
        [recompute_prepare]: 5.64998e-06
        [updatestate_depend_eliminate]: 3.08998e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.06002e-06
        [specialize_transform]: 5.23002e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 4.68999e-06
        [meta_shard_fg_expand]: 1.50999e-06
        [get_grad_eliminate_]: 4.52998e-06
        [merge_forward]: 2.73e-06
        [cell_reuse_recompute_pass]: 2.36e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.357e-05
        [j_node_and_user_rematch]: 8.52998e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 7.51999e-06
        [inline_without_move]: 4.50001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.62999e-06
        [auto_monad_grad]: 1.27e-06
        [auto_monad_eliminator]: 5.52001e-06
        [cse]: 1.031e-05
        [replace_applicator]: 5.29e-06
[py_interpret_to_execute_after_opt_a]: 1.161e-05
[rewriter_after_opt_a]: 3.572e-05
[convert_after_rewriter]: 6.78998e-06
[order_py_execute_after_rewriter]: 4.52998e-06
[mutable_eliminate]: 0.0005856
[jit_opt_b]: 5.706e-05, [1]
    [Cycle 1]: 4.894e-05, [2]
        [frontend_op_eliminate]: 1.676e-05
        [inline_after_opt_a]: 1.905e-05
[cconv]: 2.579e-05
[loop_unroll]: 0.00044572
[jit_opt_after_cconv]: 0.00014185, [1]
    [Cycle 1]: 0.00013591, [11]
        [c_1]: 2.035e-05
        [parameter_eliminate]: 3.42002e-06
        [updatestate_depend_eliminate]: 6.92002e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [cse]: 2.017e-05
        [call_graph_tuple_transform]: 1.769e-05
        [tuple_list_get_item_eliminator]: 5.09998e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 6.79982e-07
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 1.456e-05
[partial_unused_args_eliminate]: 2.53e-06
[environ_conv]: 4.85999e-06
[add_recomputation]: 4.763e-05
[cse_after_recomputation]: 2.002e-05, [1]
    [Cycle 1]: 1.44e-05, [1]
        [cse]: 8.21002e-06
[auto_monad_reorder]: 1.436e-05
[get_jit_bprop_graph]: 1.55001e-06
[rewriter_after_jit_bprop_graph]: 5.44e-06
[opt_after_jit_grad]: 0.00048563
[symbol_engine_optimizer]: 8.296e-05, [1]
    [Cycle 1]: 7.519e-05, [6]
        [build]: 3.78001e-06
        [elim_shapecalc]: 7.95e-06
        [elim_not_effective]: 1.341e-05
        [opt_reshape]: 5.54998e-06
        [fold_const_symbol]: 8.55001e-06
        [renormalize]: 6.99976e-07
[validate]: 3.543e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00828622
[execute]: 7.83001e-06
Sums
    bootstrap                            :     0.000293s :  1.74%
    type_inference                       :     0.004597s : 27.32%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000065s :  0.38%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.19%
    rewriter_before_opt_a                :     0.000071s :  0.42%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.35%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000589s :  3.50%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000017s :  0.10%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000576s :  3.42%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.11%
    jit_opt_a.cse                        :     0.000035s :  0.21%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000586s :  3.48%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.11%
    cconv                                :     0.000026s :  0.15%
    loop_unroll                          :     0.000446s :  2.65%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000048s :  0.28%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000486s :  2.89%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008286s : 49.24%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000182    23
 1.16% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.71% :     0.000005s :      3: substitution.graph_param_transform
81.19% :     0.000148s :      4: substitution.inline
 2.04% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.57% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.48% :     0.000005s :      2: substitution.replace_old_param
 7.12% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004536     2
82.62% :     0.003748s :      1: type_inference.infer
17.38% :     0.000789s :      1: type_inference.specialize
------[replace.]   0.000060     6
68.01% :     0.000041s :      4: replace.inline
31.99% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000157     6
92.43% :     0.000145s :      4: match.inline
 7.57% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 2.14% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.44% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.71% :     0.000002s :     10: predicate.addn_zero_filter
 1.83% :     0.000002s :     10: predicate.arithmetic_simplify
 1.37% :     0.000002s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000001s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.75% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.82% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.73% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000002s :     15: predicate.partial_eliminate
 1.23% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     10: predicate.reduce_eliminate
 1.61% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.32% :     0.000001s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.97% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.25% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.45% :     0.000008s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.99% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000542    12
52.17% :     0.000283s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.03% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.79% :     0.000232s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022339    76
 0.23% :     0.000051s :      1: add_recomputation
 0.30% :     0.000068s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.35% :     0.000302s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
23.53% :     0.005256s :      1: jit_opt_a
 0.65% :     0.000145s :      1: jit_opt_after_cconv
 0.27% :     0.000060s :      1: jit_opt_b
 2.04% :     0.000455s :      1: loop_unroll
 2.71% :     0.000606s :      1: mutable_eliminate
 3.59% :     0.000801s :     26: opt.transform.jit_opt_a
 0.20% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000494s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.15% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.34% :     0.000299s :      1: renormalize.infer
 1.20% :     0.000269s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000041s :      1: rewriter_after_opt_a
 0.33% :     0.000074s :      1: rewriter_before_opt_a
 0.38% :     0.000086s :      1: symbol_engine_optimizer
37.14% :     0.008296s :      1: task_emit
20.63% :     0.004608s :      1: type_inference
 0.25% :     0.000057s :      1: validate


TotalTime = 0.0146564, [30]
[bootstrap]: 0.00032398
[type_inference]: 0.00646812
[event_method]: 2.153e-05
[auto_monad]: 8.079e-05
[graph_reusing]: 6.68e-06
[pre_auto_parallel]: 2.91999e-06
[py_interpret_to_execute]: 5.272e-05
[rewriter_before_opt_a]: 8.276e-05
[expand_dump_flag]: 3.11001e-06
[jit_opt_a]: 0.00560377, [2]
    [Cycle 1]: 0.00193121, [27]
        [switch_simplify]: 0.00010771
        [loop_unroll]: 3.445e-05
        [a_1]: 0.00069692
        [with_stream_mark]: 1.617e-05
        [recompute_prepare]: 6.88e-06
        [updatestate_depend_eliminate]: 3.43e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.47001e-06
        [parameter_eliminate]: 1.42e-06
        [specialize_transform]: 6.09999e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 2.49001e-06
        [get_grad_eliminate_]: 5.35999e-06
        [merge_forward]: 3.19001e-06
        [cell_reuse_recompute_pass]: 1.26002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.339e-05
        [j_node_and_user_rematch]: 8.79e-06
        [meta_fg_expand]: 1.99999e-06
        [replace_old_param]: 8.85999e-06
        [inline_without_move]: 5.56998e-06
        [renormalize]: 0.00067835
        [add_forward_monad_depend]: 5.24e-06
        [auto_monad_grad]: 2.17001e-06
        [auto_monad_eliminator]: 1.383e-05
        [cse]: 9.257e-05
        [replace_applicator]: 1.272e-05
    [Cycle 2]: 0.00034603, [27]
        [switch_simplify]: 5.76e-06
        [loop_unroll]: 4.97999e-06
        [a_1]: 9.164e-05
        [with_stream_mark]: 1.108e-05
        [recompute_prepare]: 5.58002e-06
        [updatestate_depend_eliminate]: 2.77002e-06
        [updatestate_assign_eliminate]: 2.69001e-06
        [updatestate_loads_eliminate]: 1.76e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 5.12e-06
        [updatestate_useless_node_eliminater]: 4.94998e-06
        [accelerated_algorithm]: 5.65001e-06
        [meta_shard_fg_expand]: 1.52001e-06
        [get_grad_eliminate_]: 4.74002e-06
        [merge_forward]: 2.39001e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.413e-05
        [j_node_and_user_rematch]: 7.78999e-06
        [meta_fg_expand]: 1.66002e-06
        [replace_old_param]: 7.53999e-06
        [inline_without_move]: 4.74e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.73002e-06
        [auto_monad_grad]: 1.10999e-06
        [auto_monad_eliminator]: 6.43e-06
        [cse]: 1.157e-05
        [replace_applicator]: 5.71e-06
[py_interpret_to_execute_after_opt_a]: 1.083e-05
[rewriter_after_opt_a]: 3.047e-05
[convert_after_rewriter]: 6.79001e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.00053526
[jit_opt_b]: 5.053e-05, [1]
    [Cycle 1]: 4.322e-05, [2]
        [frontend_op_eliminate]: 1.655e-05
        [inline_after_opt_a]: 1.468e-05
[cconv]: 2.375e-05
[loop_unroll]: 0.00041027
[jit_opt_after_cconv]: 0.00013034, [1]
    [Cycle 1]: 0.00012371, [11]
        [c_1]: 2.078e-05
        [parameter_eliminate]: 1.91e-06
        [updatestate_depend_eliminate]: 4.07e-06
        [updatestate_assign_eliminate]: 2.14999e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [cse]: 1.579e-05
        [call_graph_tuple_transform]: 1.736e-05
        [tuple_list_get_item_eliminator]: 4.99998e-06
        [none_parameter_eliminate]: 1.14e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.42001e-06
[remove_dup_value]: 1.212e-05
[partial_unused_args_eliminate]: 1.96e-06
[environ_conv]: 5.25999e-06
[add_recomputation]: 4.025e-05
[cse_after_recomputation]: 2.005e-05, [1]
    [Cycle 1]: 1.488e-05, [1]
        [cse]: 8.91002e-06
[auto_monad_reorder]: 1.367e-05
[get_jit_bprop_graph]: 1.76e-06
[rewriter_after_jit_bprop_graph]: 4.60001e-06
[opt_after_jit_grad]: 0.00044192
[symbol_engine_optimizer]: 7.172e-05, [1]
    [Cycle 1]: 6.477e-05, [6]
        [build]: 2.85002e-06
        [elim_shapecalc]: 6.89999e-06
        [elim_not_effective]: 1.292e-05
        [opt_reshape]: 5.64e-06
        [fold_const_symbol]: 8.18001e-06
        [renormalize]: 4.19997e-07
[validate]: 3.115e-05
Sums
    bootstrap                            :     0.000324s :  3.02%
    type_inference                       :     0.006468s : 60.22%
    event_method                         :     0.000022s :  0.20%
    auto_monad                           :     0.000081s :  0.75%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000053s :  0.49%
    rewriter_before_opt_a                :     0.000083s :  0.77%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000113s :  1.06%
    jit_opt_a.loop_unroll                :     0.000039s :  0.37%
    jit_opt_a.a_1                        :     0.000789s :  7.34%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000678s :  6.32%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.19%
    jit_opt_a.cse                        :     0.000104s :  0.97%
    jit_opt_a.replace_applicator         :     0.000018s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.10%
    rewriter_after_opt_a                 :     0.000030s :  0.28%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000535s :  4.98%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.14%
    cconv                                :     0.000024s :  0.22%
    loop_unroll                          :     0.000410s :  3.82%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000012s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000040s :  0.37%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000014s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000442s :  4.11%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.29%


Time group info:
------[substitution.]   0.000281    26
 0.77% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 1.57% :     0.000004s :      3: substitution.graph_param_transform
82.48% :     0.000232s :      6: substitution.inline
 0.96% :     0.000003s :      4: substitution.j_node_and_user_rematch
 5.17% :     0.000015s :      4: substitution.remove_not_recompute_node
 1.46% :     0.000004s :      2: substitution.replace_old_param
 2.80% :     0.000008s :      1: substitution.switch_simplify
 4.32% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006409     2
82.07% :     0.005260s :      1: type_inference.infer
17.93% :     0.001149s :      1: type_inference.specialize
------[replace.]   0.000128     9
53.10% :     0.000068s :      6: replace.inline
22.99% :     0.000029s :      1: replace.switch_simplify
23.91% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000247     9
92.67% :     0.000229s :      6: match.inline
 2.88% :     0.000007s :      1: match.switch_simplify
 4.45% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   801
 1.40% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     12: predicate.addn_check_dump
 1.68% :     0.000002s :     12: predicate.addn_zero_filter
 1.98% :     0.000003s :     12: predicate.arithmetic_simplify
 1.35% :     0.000002s :     12: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.24% :     0.000002s :     12: predicate.compare_switch_simplify
 1.63% :     0.000002s :     12: predicate.depend_value_elim
 1.24% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.67% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.48% :     0.000002s :     12: predicate.environ_get_eliminate
 1.14% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.37% :     0.000007s :     26: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.98% :     0.000003s :     17: predicate.load_eliminater
 1.12% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.33% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.20% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000002s :     12: predicate.merge_addn
 1.12% :     0.000001s :     12: predicate.minmaximum_grad
 1.37% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     17: predicate.partial_eliminate
 1.26% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     12: predicate.reduce_eliminate
 1.53% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     20: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.69% :     0.000002s :     12: predicate.reshape_eliminate
 1.27% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     12: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.84% :     0.000001s :      6: predicate.special_op_eliminate
 0.86% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.85% :     0.000004s :     20: predicate.switch_defer_inline
 2.36% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.03% :     0.000011s :     65: predicate.switch_simplify
 1.26% :     0.000002s :     12: predicate.tile_eliminate
 1.27% :     0.000002s :     12: predicate.transpose_eliminate
 1.39% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.42% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000876    15
46.77% :     0.000409s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.27% :     0.000099s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.96% :     0.000367s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016454    72
 0.26% :     0.000043s :      1: add_recomputation
 0.51% :     0.000084s :      1: auto_monad
 0.10% :     0.000016s :      1: auto_monad_reorder
 2.04% :     0.000336s :      1: bootstrap
 0.16% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
34.08% :     0.005607s :      1: jit_opt_a
 0.81% :     0.000133s :      1: jit_opt_after_cconv
 0.32% :     0.000053s :      1: jit_opt_b
 2.54% :     0.000418s :      1: loop_unroll
 3.31% :     0.000544s :      1: mutable_eliminate
 6.49% :     0.001067s :     26: opt.transform.jit_opt_a
 0.27% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.74% :     0.000450s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.34% :     0.000055s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000014s :      1: remove_dup_value
 2.03% :     0.000334s :      1: renormalize.infer
 2.04% :     0.000336s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000033s :      1: rewriter_after_opt_a
 0.52% :     0.000085s :      1: rewriter_before_opt_a
 0.45% :     0.000074s :      1: symbol_engine_optimizer
39.38% :     0.006480s :      1: type_inference


TotalTime = 0.0186301, [33]
[bootstrap]: 0.00029067
[type_inference]: 0.00444936
[event_method]: 1.697e-05
[auto_monad]: 5.425e-05
[graph_reusing]: 4.67998e-06
[pre_auto_parallel]: 2.09e-06
[py_interpret_to_execute]: 3.456e-05
[rewriter_before_opt_a]: 6.884e-05
[expand_dump_flag]: 2.84001e-06
[jit_opt_a]: 0.00546805, [2]
    [Cycle 1]: 0.00165867, [27]
        [switch_simplify]: 5.063e-05
        [loop_unroll]: 3.051e-05
        [a_1]: 0.00053381
        [with_stream_mark]: 1.693e-05
        [recompute_prepare]: 7.15e-06
        [updatestate_depend_eliminate]: 3.2e-06
        [updatestate_assign_eliminate]: 2.08002e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.33002e-06
        [specialize_transform]: 5.64998e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 1.29998e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 2.69999e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.061e-05
        [j_node_and_user_rematch]: 9.09e-06
        [meta_fg_expand]: 1.93002e-06
        [replace_old_param]: 8.17e-06
        [inline_without_move]: 5.15001e-06
        [renormalize]: 0.00073217
        [add_forward_monad_depend]: 4.83001e-06
        [auto_monad_grad]: 2.04e-06
        [auto_monad_eliminator]: 1.117e-05
        [cse]: 2.013e-05
        [replace_applicator]: 1.278e-05
    [Cycle 2]: 0.00032426, [27]
        [switch_simplify]: 5.61e-06
        [loop_unroll]: 4.82e-06
        [a_1]: 8.897e-05
        [with_stream_mark]: 9.97001e-06
        [recompute_prepare]: 5.10999e-06
        [updatestate_depend_eliminate]: 2.51998e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 5.17e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.17999e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 4.57e-06
        [merge_forward]: 2.65002e-06
        [cell_reuse_recompute_pass]: 2.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.233e-05
        [j_node_and_user_rematch]: 7.85e-06
        [meta_fg_expand]: 1.81e-06
        [replace_old_param]: 7.47002e-06
        [inline_without_move]: 4.63001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 5.19e-06
        [cse]: 1.114e-05
        [replace_applicator]: 5.12e-06
[py_interpret_to_execute_after_opt_a]: 9.92999e-06
[rewriter_after_opt_a]: 2.917e-05
[convert_after_rewriter]: 6.19001e-06
[order_py_execute_after_rewriter]: 4.96002e-06
[mutable_eliminate]: 0.00062107
[jit_opt_b]: 5.172e-05, [1]
    [Cycle 1]: 4.503e-05, [2]
        [frontend_op_eliminate]: 1.644e-05
        [inline_after_opt_a]: 1.515e-05
[cconv]: 2.114e-05
[loop_unroll]: 0.00045152
[jit_opt_after_cconv]: 0.00014646, [1]
    [Cycle 1]: 0.00014019, [11]
        [c_1]: 2.149e-05
        [parameter_eliminate]: 2.84999e-06
        [updatestate_depend_eliminate]: 5.89e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 2.076e-05
        [call_graph_tuple_transform]: 1.977e-05
        [tuple_list_get_item_eliminator]: 5.49e-06
        [none_parameter_eliminate]: 1.24e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 5.67001e-06
[remove_dup_value]: 1.072e-05
[partial_unused_args_eliminate]: 1.86e-06
[environ_conv]: 5.02999e-06
[add_recomputation]: 4.018e-05
[cse_after_recomputation]: 2.053e-05, [1]
    [Cycle 1]: 1.46e-05, [1]
        [cse]: 8.97999e-06
[auto_monad_reorder]: 1.185e-05
[get_jit_bprop_graph]: 1.81e-06
[rewriter_after_jit_bprop_graph]: 6.48e-06
[opt_after_jit_grad]: 0.00048525
[symbol_engine_optimizer]: 7.584e-05, [1]
    [Cycle 1]: 6.869e-05, [6]
        [build]: 3.43e-06
        [elim_shapecalc]: 8.48001e-06
        [elim_not_effective]: 1.306e-05
        [opt_reshape]: 6.16e-06
        [fold_const_symbol]: 8.75001e-06
        [renormalize]: 4.2998e-07
[validate]: 3.522e-05
[backend_pass]: 1.00001e-06
[task_emit]: 0.00598451
[execute]: 9.05999e-06
Sums
    bootstrap                            :     0.000291s :  2.00%
    type_inference                       :     0.004449s : 30.63%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000054s :  0.37%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000035s :  0.24%
    rewriter_before_opt_a                :     0.000069s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000623s :  4.29%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000732s :  5.04%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.11%
    jit_opt_a.cse                        :     0.000031s :  0.22%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000029s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000621s :  4.28%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000021s :  0.15%
    loop_unroll                          :     0.000452s :  3.11%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000011s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000040s :  0.28%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000485s :  3.34%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005985s : 41.19%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000201    23
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.44% :     0.000005s :      3: substitution.graph_param_transform
85.45% :     0.000172s :      4: substitution.inline
 1.59% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.05% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.84% :     0.000004s :      2: substitution.replace_old_param
 5.08% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004387     2
80.16% :     0.003517s :      1: type_inference.infer
19.84% :     0.000870s :      1: type_inference.specialize
------[replace.]   0.000069     6
67.65% :     0.000047s :      4: replace.inline
32.35% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000179     6
94.97% :     0.000170s :      4: match.inline
 5.03% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.23% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 2.21% :     0.000003s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.48% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.09% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.55% :     0.000001s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.19% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.68% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.17% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.21% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.29% :     0.000002s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000002s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000001s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.07% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.19% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.01% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.71% :     0.000009s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.41% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.26% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000614    12
46.80% :     0.000287s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.71% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.48% :     0.000292s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020299    76
 0.21% :     0.000043s :      1: add_recomputation
 0.28% :     0.000057s :      1: auto_monad
 0.07% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.52% :     0.000309s :      1: bootstrap
 0.12% :     0.000024s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
26.96% :     0.005472s :      1: jit_opt_a
 0.73% :     0.000149s :      1: jit_opt_after_cconv
 0.27% :     0.000054s :      1: jit_opt_b
 2.27% :     0.000460s :      1: loop_unroll
 3.11% :     0.000631s :      1: mutable_eliminate
 4.07% :     0.000827s :     26: opt.transform.jit_opt_a
 0.24% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.43% :     0.000494s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000013s :      1: remove_dup_value
 1.85% :     0.000376s :      1: renormalize.infer
 1.72% :     0.000348s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000032s :      1: rewriter_after_opt_a
 0.35% :     0.000072s :      1: rewriter_before_opt_a
 0.39% :     0.000078s :      1: symbol_engine_optimizer
29.58% :     0.006004s :      1: task_emit
21.98% :     0.004462s :      1: type_inference
 0.27% :     0.000054s :      1: validate


TotalTime = 0.0166261, [33]
[bootstrap]: 0.0002471
[type_inference]: 0.00440525
[event_method]: 1.521e-05
[auto_monad]: 5.533e-05
[graph_reusing]: 4.90999e-06
[pre_auto_parallel]: 2.13998e-06
[py_interpret_to_execute]: 2.669e-05
[rewriter_before_opt_a]: 6.496e-05
[expand_dump_flag]: 2.94001e-06
[jit_opt_a]: 0.00477314, [2]
    [Cycle 1]: 0.00129251, [27]
        [switch_simplify]: 4.317e-05
        [loop_unroll]: 2.79e-05
        [a_1]: 0.00046607
        [with_stream_mark]: 1.328e-05
        [recompute_prepare]: 6.76999e-06
        [updatestate_depend_eliminate]: 2.91e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 1.37e-06
        [specialize_transform]: 5.56e-06
        [updatestate_useless_node_eliminater]: 4.76002e-06
        [accelerated_algorithm]: 5.14e-06
        [meta_shard_fg_expand]: 1.81003e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 3.33e-06
        [cell_reuse_recompute_pass]: 1.07998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.284e-05
        [j_node_and_user_rematch]: 8.40001e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 7.65998e-06
        [inline_without_move]: 5.07999e-06
        [renormalize]: 0.00046749
        [add_forward_monad_depend]: 4.76002e-06
        [auto_monad_grad]: 1.81998e-06
        [auto_monad_eliminator]: 1.09e-05
        [cse]: 1.562e-05
        [replace_applicator]: 1.09e-05
    [Cycle 2]: 0.00031352, [27]
        [switch_simplify]: 5.43002e-06
        [loop_unroll]: 4.62e-06
        [a_1]: 8.654e-05
        [with_stream_mark]: 7.82002e-06
        [recompute_prepare]: 4.47998e-06
        [updatestate_depend_eliminate]: 2.39001e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 1.04998e-06
        [specialize_transform]: 4.75001e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 4.55001e-06
        [meta_shard_fg_expand]: 1.19e-06
        [get_grad_eliminate_]: 4.43001e-06
        [merge_forward]: 2.22001e-06
        [cell_reuse_recompute_pass]: 1.57999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.151e-05
        [j_node_and_user_rematch]: 7.63999e-06
        [meta_fg_expand]: 1.45999e-06
        [replace_old_param]: 6.73e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.40001e-06
        [auto_monad_grad]: 1.02e-06
        [auto_monad_eliminator]: 5.04998e-06
        [cse]: 9.71998e-06
        [replace_applicator]: 4.63001e-06
[py_interpret_to_execute_after_opt_a]: 7.77e-06
[rewriter_after_opt_a]: 2.691e-05
[convert_after_rewriter]: 6.20002e-06
[order_py_execute_after_rewriter]: 4.04997e-06
[mutable_eliminate]: 0.00046399
[jit_opt_b]: 4.515e-05, [1]
    [Cycle 1]: 3.902e-05, [2]
        [frontend_op_eliminate]: 1.501e-05
        [inline_after_opt_a]: 1.355e-05
[cconv]: 2.547e-05
[loop_unroll]: 0.00040241
[jit_opt_after_cconv]: 0.00013149, [1]
    [Cycle 1]: 0.00012535, [11]
        [c_1]: 2.044e-05
        [parameter_eliminate]: 2.74999e-06
        [updatestate_depend_eliminate]: 5.22e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [cse]: 1.587e-05
        [call_graph_tuple_transform]: 1.644e-05
        [tuple_list_get_item_eliminator]: 5.02e-06
        [none_parameter_eliminate]: 1.19998e-06
        [renormalize]: 5.39992e-07
        [switch_simplify]: 5.17999e-06
[remove_dup_value]: 1.029e-05
[partial_unused_args_eliminate]: 1.89e-06
[environ_conv]: 3.76999e-06
[add_recomputation]: 3.645e-05
[cse_after_recomputation]: 1.928e-05, [1]
    [Cycle 1]: 1.435e-05, [1]
        [cse]: 8.91002e-06
[auto_monad_reorder]: 1.217e-05
[get_jit_bprop_graph]: 1.47999e-06
[rewriter_after_jit_bprop_graph]: 4.95999e-06
[opt_after_jit_grad]: 0.00044417
[symbol_engine_optimizer]: 6.878e-05, [1]
    [Cycle 1]: 6.29e-05, [6]
        [build]: 3.18e-06
        [elim_shapecalc]: 6.88e-06
        [elim_not_effective]: 1.207e-05
        [opt_reshape]: 5.29e-06
        [fold_const_symbol]: 8.05e-06
        [renormalize]: 5.10016e-07
[validate]: 2.801e-05
[backend_pass]: 8.2e-07
[task_emit]: 0.00512166
[execute]: 6.72002e-06
Sums
    bootstrap                            :     0.000247s :  1.91%
    type_inference                       :     0.004405s : 34.10%
    event_method                         :     0.000015s :  0.12%
    auto_monad                           :     0.000055s :  0.43%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000027s :  0.21%
    rewriter_before_opt_a                :     0.000065s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000049s :  0.38%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000553s :  4.28%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000468s :  3.62%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.12%
    jit_opt_a.cse                        :     0.000025s :  0.20%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000027s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000464s :  3.59%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.20%
    loop_unroll                          :     0.000402s :  3.12%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000010s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000036s :  0.28%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000012s :  0.09%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000444s :  3.44%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000028s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005122s : 39.65%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000160    23
 1.18% :     0.000002s :      2: substitution.elim_not_effective
 0.83% :     0.000001s :      2: substitution.fold_const_symbol
 2.74% :     0.000004s :      3: substitution.graph_param_transform
82.20% :     0.000131s :      4: substitution.inline
 1.74% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.74% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.89% :     0.000003s :      2: substitution.replace_old_param
 6.69% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004357     2
83.84% :     0.003653s :      1: type_inference.infer
16.16% :     0.000704s :      1: type_inference.specialize
------[replace.]   0.000059     6
66.96% :     0.000039s :      4: replace.inline
33.04% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000139     6
93.05% :     0.000129s :      4: match.inline
 6.95% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000107   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000001s :     10: predicate.addn_zero_filter
 1.73% :     0.000002s :     10: predicate.arithmetic_simplify
 1.37% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000001s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.57% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.88% :     0.000002s :     15: predicate.load_eliminater
 1.58% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.26% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.20% :     0.000001s :     10: predicate.minmaximum_grad
 1.40% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.36% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.83% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.38% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.50% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 2.26% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.78% :     0.000008s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.51% :     0.000002s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.51% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000473    12
48.43% :     0.000229s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.17% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.40% :     0.000220s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017936    76
 0.22% :     0.000039s :      1: add_recomputation
 0.33% :     0.000058s :      1: auto_monad
 0.08% :     0.000015s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.42% :     0.000255s :      1: bootstrap
 0.16% :     0.000028s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.63% :     0.004776s :      1: jit_opt_a
 0.75% :     0.000134s :      1: jit_opt_after_cconv
 0.26% :     0.000048s :      1: jit_opt_b
 2.29% :     0.000410s :      1: loop_unroll
 2.62% :     0.000470s :      1: mutable_eliminate
 4.12% :     0.000739s :     26: opt.transform.jit_opt_a
 0.25% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.52% :     0.000453s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000013s :      1: remove_dup_value
 1.31% :     0.000236s :      1: renormalize.infer
 1.26% :     0.000226s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000030s :      1: rewriter_after_opt_a
 0.38% :     0.000068s :      1: rewriter_before_opt_a
 0.40% :     0.000071s :      1: symbol_engine_optimizer
28.61% :     0.005131s :      1: task_emit
24.61% :     0.004413s :      1: type_inference
 0.24% :     0.000043s :      1: validate


TotalTime = 0.0280258, [33]
[bootstrap]: 0.00041268
[type_inference]: 0.0109928
[event_method]: 1.773e-05
[auto_monad]: 6.894e-05
[graph_reusing]: 5.17e-06
[pre_auto_parallel]: 2.48002e-06
[py_interpret_to_execute]: 4.601e-05
[rewriter_before_opt_a]: 7.047e-05
[expand_dump_flag]: 3.60998e-06
[jit_opt_a]: 0.0072977, [2]
    [Cycle 1]: 0.00279136, [27]
        [switch_simplify]: 0.00076301
        [loop_unroll]: 3.153e-05
        [a_1]: 0.0006011
        [with_stream_mark]: 2.106e-05
        [recompute_prepare]: 7.1e-06
        [updatestate_depend_eliminate]: 3.79002e-06
        [updatestate_assign_eliminate]: 3.33e-06
        [updatestate_loads_eliminate]: 2.64001e-06
        [parameter_eliminate]: 2.26998e-06
        [specialize_transform]: 6.33e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 5.76998e-06
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 5.64e-06
        [merge_forward]: 3.51999e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.391e-05
        [j_node_and_user_rematch]: 9.36998e-06
        [meta_fg_expand]: 2.72001e-06
        [replace_old_param]: 1.199e-05
        [inline_without_move]: 5.74999e-06
        [renormalize]: 0.00097887
        [add_forward_monad_depend]: 6.81001e-06
        [auto_monad_grad]: 2.79999e-06
        [auto_monad_eliminator]: 1.388e-05
        [cse]: 2.307e-05
        [replace_applicator]: 1.342e-05
    [Cycle 2]: 0.00033025, [27]
        [switch_simplify]: 5.81998e-06
        [loop_unroll]: 5.30001e-06
        [a_1]: 9.266e-05
        [with_stream_mark]: 1.085e-05
        [recompute_prepare]: 5.09e-06
        [updatestate_depend_eliminate]: 2.84001e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 1.19998e-06
        [specialize_transform]: 4.78001e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.14e-06
        [meta_shard_fg_expand]: 1.46002e-06
        [get_grad_eliminate_]: 4.32998e-06
        [merge_forward]: 2.64999e-06
        [cell_reuse_recompute_pass]: 2.73998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.398e-05
        [j_node_and_user_rematch]: 7.82e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 7.09001e-06
        [inline_without_move]: 5.24e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.32999e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 6.09001e-06
        [cse]: 1.009e-05
        [replace_applicator]: 5.07e-06
[py_interpret_to_execute_after_opt_a]: 1.37e-05
[rewriter_after_opt_a]: 3.722e-05
[convert_after_rewriter]: 8.06001e-06
[order_py_execute_after_rewriter]: 5.36002e-06
[mutable_eliminate]: 0.00080685
[jit_opt_b]: 5.071e-05, [1]
    [Cycle 1]: 4.353e-05, [2]
        [frontend_op_eliminate]: 1.68e-05
        [inline_after_opt_a]: 1.506e-05
[cconv]: 2.823e-05
[loop_unroll]: 0.00045692
[jit_opt_after_cconv]: 0.00015037, [1]
    [Cycle 1]: 0.00014343, [11]
        [c_1]: 2.247e-05
        [parameter_eliminate]: 3.69002e-06
        [updatestate_depend_eliminate]: 6.71999e-06
        [updatestate_assign_eliminate]: 2.45997e-06
        [updatestate_loads_eliminate]: 2.28002e-06
        [cse]: 2.155e-05
        [call_graph_tuple_transform]: 2.184e-05
        [tuple_list_get_item_eliminator]: 5.20001e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 2.60014e-07
        [switch_simplify]: 5.46e-06
[remove_dup_value]: 1.587e-05
[partial_unused_args_eliminate]: 2.89999e-06
[environ_conv]: 6.50997e-06
[add_recomputation]: 5.198e-05
[cse_after_recomputation]: 2.009e-05, [1]
    [Cycle 1]: 1.43e-05, [1]
        [cse]: 8.59002e-06
[auto_monad_reorder]: 1.575e-05
[get_jit_bprop_graph]: 2.16e-06
[rewriter_after_jit_bprop_graph]: 6.10002e-06
[opt_after_jit_grad]: 0.00048351
[symbol_engine_optimizer]: 7.458e-05, [1]
    [Cycle 1]: 6.803e-05, [6]
        [build]: 4.06001e-06
        [elim_shapecalc]: 7.66001e-06
        [elim_not_effective]: 1.296e-05
        [opt_reshape]: 5.72999e-06
        [fold_const_symbol]: 9.39998e-06
        [renormalize]: 8.50006e-07
[validate]: 4.279e-05
[backend_pass]: 1.04e-06
[task_emit]: 0.00659048
[execute]: 1.102e-05
Sums
    bootstrap                            :     0.000413s :  1.78%
    type_inference                       :     0.010993s : 47.50%
    event_method                         :     0.000018s :  0.08%
    auto_monad                           :     0.000069s :  0.30%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000046s :  0.20%
    rewriter_before_opt_a                :     0.000070s :  0.30%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000769s :  3.32%
    jit_opt_a.loop_unroll                :     0.000037s :  0.16%
    jit_opt_a.a_1                        :     0.000694s :  3.00%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.08%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.000979s :  4.23%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.09%
    jit_opt_a.cse                        :     0.000033s :  0.14%
    jit_opt_a.replace_applicator         :     0.000018s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.06%
    rewriter_after_opt_a                 :     0.000037s :  0.16%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000807s :  3.49%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.07%
    cconv                                :     0.000028s :  0.12%
    loop_unroll                          :     0.000457s :  1.97%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000016s :  0.07%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000052s :  0.22%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000484s :  2.09%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.18%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006590s : 28.48%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000261    23
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.46% :     0.000006s :      3: substitution.graph_param_transform
86.11% :     0.000225s :      4: substitution.inline
 1.25% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.78% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.83% :     0.000005s :      2: substitution.replace_old_param
 5.23% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010907     2
89.31% :     0.009741s :      1: type_inference.infer
10.69% :     0.001166s :      1: type_inference.specialize
------[replace.]   0.000077     6
68.10% :     0.000052s :      4: replace.inline
31.90% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000235     6
94.66% :     0.000222s :      4: match.inline
 5.34% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.82% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.30% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 2.15% :     0.000003s :     10: predicate.arithmetic_simplify
 1.63% :     0.000002s :     10: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.42% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.39% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.02% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.87% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 1.99% :     0.000002s :     15: predicate.partial_eliminate
 1.25% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.82% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.27% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.14% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.00% :     0.000002s :     16: predicate.switch_layer_defer_inline
 9.73% :     0.000012s :     53: predicate.switch_simplify
 1.63% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.91% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000802    12
43.72% :     0.000351s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.61% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.67% :     0.000399s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.030722    76
 0.18% :     0.000055s :      1: add_recomputation
 0.23% :     0.000072s :      1: auto_monad
 0.06% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.40% :     0.000429s :      1: bootstrap
 0.10% :     0.000031s :      1: cconv
 0.04% :     0.000013s :      1: convert_after_rewriter
 0.07% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.07% :     0.000022s :      1: event_method
 0.06% :     0.000018s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
23.77% :     0.007302s :      1: jit_opt_a
 0.50% :     0.000153s :      1: jit_opt_after_cconv
 0.17% :     0.000053s :      1: jit_opt_b
 1.51% :     0.000465s :      1: loop_unroll
 2.65% :     0.000816s :      1: mutable_eliminate
 5.26% :     0.001616s :     26: opt.transform.jit_opt_a
 0.17% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000025s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.60% :     0.000492s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000049s :      1: py_interpret_to_execute
 0.05% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000019s :      1: remove_dup_value
 1.75% :     0.000538s :      1: renormalize.infer
 1.41% :     0.000432s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000040s :      1: rewriter_after_opt_a
 0.24% :     0.000073s :      1: rewriter_before_opt_a
 0.25% :     0.000077s :      1: symbol_engine_optimizer
21.52% :     0.006612s :      1: task_emit
35.84% :     0.011010s :      1: type_inference
 0.21% :     0.000066s :      1: validate


TotalTime = 0.0149982, [30]
[bootstrap]: 0.00026211
[type_inference]: 0.00686724
[event_method]: 2.147e-05
[auto_monad]: 8.705e-05
[graph_reusing]: 6.73e-06
[pre_auto_parallel]: 2.93e-06
[py_interpret_to_execute]: 6.765e-05
[rewriter_before_opt_a]: 8.63e-05
[expand_dump_flag]: 3.07002e-06
[jit_opt_a]: 0.00558917, [2]
    [Cycle 1]: 0.00199532, [27]
        [switch_simplify]: 0.0001088
        [loop_unroll]: 3.391e-05
        [a_1]: 0.00072581
        [with_stream_mark]: 1.604e-05
        [recompute_prepare]: 7.78001e-06
        [updatestate_depend_eliminate]: 3.41001e-06
        [updatestate_assign_eliminate]: 3.6e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 6.27001e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 5.94999e-06
        [meta_shard_fg_expand]: 2.17001e-06
        [get_grad_eliminate_]: 5.52999e-06
        [merge_forward]: 3.5e-06
        [cell_reuse_recompute_pass]: 9.89996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.475e-05
        [j_node_and_user_rematch]: 9.91998e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 9.34e-06
        [inline_without_move]: 5.49e-06
        [renormalize]: 0.00079271
        [add_forward_monad_depend]: 5.15001e-06
        [auto_monad_grad]: 1.97001e-06
        [auto_monad_eliminator]: 1.279e-05
        [cse]: 2.449e-05
        [replace_applicator]: 1.212e-05
    [Cycle 2]: 0.0003092, [27]
        [switch_simplify]: 5.27001e-06
        [loop_unroll]: 4.72e-06
        [a_1]: 8.759e-05
        [with_stream_mark]: 1.005e-05
        [recompute_prepare]: 4.75999e-06
        [updatestate_depend_eliminate]: 2.66e-06
        [updatestate_assign_eliminate]: 2.18002e-06
        [updatestate_loads_eliminate]: 2.01998e-06
        [parameter_eliminate]: 8.39995e-07
        [specialize_transform]: 4.82998e-06
        [updatestate_useless_node_eliminater]: 4.63001e-06
        [accelerated_algorithm]: 4.69998e-06
        [meta_shard_fg_expand]: 1.08001e-06
        [get_grad_eliminate_]: 4.40999e-06
        [merge_forward]: 2.71e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.239e-05
        [j_node_and_user_rematch]: 7.78001e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 7.05998e-06
        [inline_without_move]: 4.69998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.05999e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 5.14e-06
        [cse]: 9.76998e-06
        [replace_applicator]: 4.97e-06
[py_interpret_to_execute_after_opt_a]: 9.74e-06
[rewriter_after_opt_a]: 3.24e-05
[convert_after_rewriter]: 6.37001e-06
[order_py_execute_after_rewriter]: 5.56e-06
[mutable_eliminate]: 0.00051749
[jit_opt_b]: 4.714e-05, [1]
    [Cycle 1]: 4.103e-05, [2]
        [frontend_op_eliminate]: 1.572e-05
        [inline_after_opt_a]: 1.463e-05
[cconv]: 2.322e-05
[loop_unroll]: 0.00040642
[jit_opt_after_cconv]: 0.00014654, [1]
    [Cycle 1]: 0.00014051, [11]
        [c_1]: 2.106e-05
        [parameter_eliminate]: 2.42001e-06
        [updatestate_depend_eliminate]: 5.34e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 1.664e-05
        [call_graph_tuple_transform]: 1.818e-05
        [tuple_list_get_item_eliminator]: 5.34e-06
        [none_parameter_eliminate]: 1.43002e-06
        [renormalize]: 6.80011e-07
        [switch_simplify]: 5.30999e-06
[remove_dup_value]: 1.297e-05
[partial_unused_args_eliminate]: 2.44999e-06
[environ_conv]: 5.15999e-06
[add_recomputation]: 4.609e-05
[cse_after_recomputation]: 1.98e-05, [1]
    [Cycle 1]: 1.461e-05, [1]
        [cse]: 9.07001e-06
[auto_monad_reorder]: 1.462e-05
[get_jit_bprop_graph]: 1.99999e-06
[rewriter_after_jit_bprop_graph]: 5.12999e-06
[opt_after_jit_grad]: 0.00042599
[symbol_engine_optimizer]: 7.089e-05, [1]
    [Cycle 1]: 6.503e-05, [6]
        [build]: 2.84001e-06
        [elim_shapecalc]: 7.31999e-06
        [elim_not_effective]: 1.279e-05
        [opt_reshape]: 5.38002e-06
        [fold_const_symbol]: 8.3e-06
        [renormalize]: 4.30009e-07
[validate]: 3.495e-05
Sums
    bootstrap                            :     0.000262s :  2.35%
    type_inference                       :     0.006867s : 61.67%
    event_method                         :     0.000021s :  0.19%
    auto_monad                           :     0.000087s :  0.78%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000068s :  0.61%
    rewriter_before_opt_a                :     0.000086s :  0.77%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000114s :  1.02%
    jit_opt_a.loop_unroll                :     0.000039s :  0.35%
    jit_opt_a.a_1                        :     0.000813s :  7.30%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000793s :  7.12%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.16%
    jit_opt_a.cse                        :     0.000034s :  0.31%
    jit_opt_a.replace_applicator         :     0.000017s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.09%
    rewriter_after_opt_a                 :     0.000032s :  0.29%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000517s :  4.65%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.13%
    cconv                                :     0.000023s :  0.21%
    loop_unroll                          :     0.000406s :  3.65%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000046s :  0.41%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000015s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000426s :  3.83%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.31%


Time group info:
------[substitution.]   0.000271    26
 0.69% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000001s :      2: substitution.fold_const_symbol
 1.99% :     0.000005s :      3: substitution.graph_param_transform
84.48% :     0.000229s :      6: substitution.inline
 1.38% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.74% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.32% :     0.000004s :      2: substitution.replace_old_param
 2.98% :     0.000008s :      1: substitution.switch_simplify
 4.89% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006791     2
80.83% :     0.005489s :      1: type_inference.infer
19.17% :     0.001302s :      1: type_inference.specialize
------[replace.]   0.000141     9
52.79% :     0.000075s :      6: replace.inline
21.63% :     0.000031s :      1: replace.switch_simplify
25.58% :     0.000036s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000245     9
92.13% :     0.000226s :      6: match.inline
 2.97% :     0.000007s :      1: match.switch_simplify
 4.91% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   801
 1.41% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.98% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     12: predicate.addn_check_dump
 1.46% :     0.000002s :     12: predicate.addn_zero_filter
 2.05% :     0.000003s :     12: predicate.arithmetic_simplify
 1.40% :     0.000002s :     12: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000001s :     12: predicate.compare_switch_simplify
 1.48% :     0.000002s :     12: predicate.depend_value_elim
 1.25% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.24% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     12: predicate.environ_get_eliminate
 1.14% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.79% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.45% :     0.000007s :     26: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     17: predicate.load_eliminater
 1.17% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.30% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.80% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.28% :     0.000002s :     12: predicate.merge_addn
 1.15% :     0.000001s :     12: predicate.minmaximum_grad
 1.49% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     17: predicate.partial_eliminate
 1.32% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     12: predicate.reduce_eliminate
 1.49% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     20: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     12: predicate.reshape_eliminate
 1.25% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     12: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.76% :     0.000001s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.82% :     0.000004s :     20: predicate.switch_defer_inline
 2.48% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.37% :     0.000011s :     65: predicate.switch_simplify
 1.39% :     0.000002s :     12: predicate.tile_eliminate
 1.26% :     0.000002s :     12: predicate.transpose_eliminate
 1.44% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.42% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.64% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000002s :     12: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000904    15
45.93% :     0.000415s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 9.62% :     0.000087s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.45% :     0.000402s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016920    72
 0.29% :     0.000049s :      1: add_recomputation
 0.53% :     0.000090s :      1: auto_monad
 0.10% :     0.000017s :      1: auto_monad_reorder
 1.63% :     0.000275s :      1: bootstrap
 0.16% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
33.06% :     0.005593s :      1: jit_opt_a
 0.88% :     0.000149s :      1: jit_opt_after_cconv
 0.29% :     0.000050s :      1: jit_opt_b
 2.45% :     0.000414s :      1: loop_unroll
 3.11% :     0.000526s :      1: mutable_eliminate
 6.39% :     0.001081s :     26: opt.transform.jit_opt_a
 0.27% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.57% :     0.000434s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.42% :     0.000071s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000015s :      1: remove_dup_value
 2.57% :     0.000436s :      1: renormalize.infer
 2.06% :     0.000348s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000035s :      1: rewriter_after_opt_a
 0.53% :     0.000089s :      1: rewriter_before_opt_a
 0.43% :     0.000073s :      1: symbol_engine_optimizer
40.68% :     0.006883s :      1: type_inference


TotalTime = 0.0191126, [33]
[bootstrap]: 0.00029392
[type_inference]: 0.00460258
[event_method]: 1.737e-05
[auto_monad]: 6.542e-05
[graph_reusing]: 5.94e-06
[pre_auto_parallel]: 2.48998e-06
[py_interpret_to_execute]: 3.752e-05
[rewriter_before_opt_a]: 7.81e-05
[expand_dump_flag]: 3.18998e-06
[jit_opt_a]: 0.00569529, [2]
    [Cycle 1]: 0.0018213, [27]
        [switch_simplify]: 6.13e-05
        [loop_unroll]: 2.928e-05
        [a_1]: 0.00058151
        [with_stream_mark]: 2.276e-05
        [recompute_prepare]: 6.73e-06
        [updatestate_depend_eliminate]: 4.27998e-06
        [updatestate_assign_eliminate]: 3.35e-06
        [updatestate_loads_eliminate]: 2.78e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 6.44999e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 1.91003e-06
        [get_grad_eliminate_]: 5.05001e-06
        [merge_forward]: 3.46999e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.402e-05
        [j_node_and_user_rematch]: 8.70999e-06
        [meta_fg_expand]: 2.04999e-06
        [replace_old_param]: 9.01002e-06
        [inline_without_move]: 5.81e-06
        [renormalize]: 0.00079892
        [add_forward_monad_depend]: 5.65001e-06
        [auto_monad_grad]: 2.58e-06
        [auto_monad_eliminator]: 1.449e-05
        [cse]: 2.837e-05
        [replace_applicator]: 1.321e-05
    [Cycle 2]: 0.00033328, [27]
        [switch_simplify]: 5.88998e-06
        [loop_unroll]: 5.57001e-06
        [a_1]: 9.257e-05
        [with_stream_mark]: 9.86998e-06
        [recompute_prepare]: 5.20999e-06
        [updatestate_depend_eliminate]: 2.93e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.52999e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 4.53001e-06
        [merge_forward]: 2.89001e-06
        [cell_reuse_recompute_pass]: 1.78002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.194e-05
        [j_node_and_user_rematch]: 7.96001e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 7.33e-06
        [inline_without_move]: 1.227e-05
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.59e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 5.63997e-06
        [cse]: 1.119e-05
        [replace_applicator]: 4.77e-06
[py_interpret_to_execute_after_opt_a]: 1.076e-05
[rewriter_after_opt_a]: 3.705e-05
[convert_after_rewriter]: 8.06001e-06
[order_py_execute_after_rewriter]: 5.70001e-06
[mutable_eliminate]: 0.00061644
[jit_opt_b]: 5.014e-05, [1]
    [Cycle 1]: 4.246e-05, [2]
        [frontend_op_eliminate]: 1.686e-05
        [inline_after_opt_a]: 1.449e-05
[cconv]: 2.523e-05
[loop_unroll]: 0.00048889
[jit_opt_after_cconv]: 0.00014426, [1]
    [Cycle 1]: 0.00013752, [11]
        [c_1]: 2.16e-05
        [parameter_eliminate]: 2.70997e-06
        [updatestate_depend_eliminate]: 5.91e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 1.96998e-06
        [cse]: 1.993e-05
        [call_graph_tuple_transform]: 1.874e-05
        [tuple_list_get_item_eliminator]: 5.72999e-06
        [none_parameter_eliminate]: 1.48002e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.61003e-06
[remove_dup_value]: 1.412e-05
[partial_unused_args_eliminate]: 2.52001e-06
[environ_conv]: 5.10999e-06
[add_recomputation]: 4.906e-05
[cse_after_recomputation]: 2.042e-05, [1]
    [Cycle 1]: 1.508e-05, [1]
        [cse]: 9.12001e-06
[auto_monad_reorder]: 1.465e-05
[get_jit_bprop_graph]: 2.74001e-06
[rewriter_after_jit_bprop_graph]: 5.35999e-06
[opt_after_jit_grad]: 0.00047897
[symbol_engine_optimizer]: 7.272e-05, [1]
    [Cycle 1]: 6.604e-05, [6]
        [build]: 3.83999e-06
        [elim_shapecalc]: 7.83001e-06
        [elim_not_effective]: 1.217e-05
        [opt_reshape]: 5.73002e-06
        [fold_const_symbol]: 8.43001e-06
        [renormalize]: 4.19997e-07
[validate]: 3.601e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00600557
[execute]: 7.92998e-06
Sums
    bootstrap                            :     0.000294s :  1.97%
    type_inference                       :     0.004603s : 30.78%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000065s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.25%
    rewriter_before_opt_a                :     0.000078s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.45%
    jit_opt_a.loop_unroll                :     0.000035s :  0.23%
    jit_opt_a.a_1                        :     0.000674s :  4.51%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000018s :  0.12%
    jit_opt_a.renormalize                :     0.000799s :  5.34%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000040s :  0.26%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.25%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000616s :  4.12%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000489s :  3.27%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000479s :  3.20%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006006s : 40.17%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000241    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.09% :     0.000005s :      3: substitution.graph_param_transform
86.30% :     0.000208s :      4: substitution.inline
 1.33% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.76% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.85% :     0.000004s :      2: substitution.replace_old_param
 5.35% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004537     2
81.02% :     0.003675s :      1: type_inference.infer
18.98% :     0.000861s :      1: type_inference.specialize
------[replace.]   0.000075     6
70.04% :     0.000052s :      4: replace.inline
29.96% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000217     6
94.57% :     0.000206s :      4: match.inline
 5.43% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.52% :     0.000003s :     10: predicate.arithmetic_simplify
 1.49% :     0.000002s :     10: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.40% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 4.93% :     0.000006s :     22: predicate.inline
 1.05% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000001s :      6: predicate.less_batch_normalization
 1.85% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.23% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.96% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.20% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.85% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.85% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.59% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.07% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.23% :     0.000010s :     53: predicate.switch_simplify
 1.24% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.59% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.95% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000618    12
46.84% :     0.000289s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.75% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.41% :     0.000280s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020909    76
 0.25% :     0.000052s :      1: add_recomputation
 0.33% :     0.000069s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.48% :     0.000310s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.26% :     0.005699s :      1: jit_opt_a
 0.70% :     0.000147s :      1: jit_opt_after_cconv
 0.25% :     0.000053s :      1: jit_opt_b
 2.38% :     0.000498s :      1: loop_unroll
 2.99% :     0.000626s :      1: mutable_eliminate
 4.29% :     0.000896s :     26: opt.transform.jit_opt_a
 0.23% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.33% :     0.000488s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000040s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 2.12% :     0.000442s :      1: renormalize.infer
 1.66% :     0.000348s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000040s :      1: rewriter_after_opt_a
 0.39% :     0.000081s :      1: rewriter_before_opt_a
 0.36% :     0.000076s :      1: symbol_engine_optimizer
28.80% :     0.006022s :      1: task_emit
22.08% :     0.004616s :      1: type_inference
 0.28% :     0.000058s :      1: validate


TotalTime = 0.0156435, [33]
[bootstrap]: 0.00027091
[type_inference]: 0.00403569
[event_method]: 1.413e-05
[auto_monad]: 5.551e-05
[graph_reusing]: 5.56e-06
[pre_auto_parallel]: 2.18002e-06
[py_interpret_to_execute]: 3.052e-05
[rewriter_before_opt_a]: 6.499e-05
[expand_dump_flag]: 2.95998e-06
[jit_opt_a]: 0.00457422, [2]
    [Cycle 1]: 0.00131409, [27]
        [switch_simplify]: 3.886e-05
        [loop_unroll]: 2.758e-05
        [a_1]: 0.00049914
        [with_stream_mark]: 1.257e-05
        [recompute_prepare]: 6.57002e-06
        [updatestate_depend_eliminate]: 2.77002e-06
        [updatestate_assign_eliminate]: 2.13998e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 5.81e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 5.32001e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 4.82e-06
        [merge_forward]: 2.67001e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.06e-05
        [j_node_and_user_rematch]: 8.13999e-06
        [meta_fg_expand]: 1.82999e-06
        [replace_old_param]: 7.25003e-06
        [inline_without_move]: 5.24e-06
        [renormalize]: 0.00047365
        [add_forward_monad_depend]: 4.32003e-06
        [auto_monad_grad]: 1.59e-06
        [auto_monad_eliminator]: 8.99e-06
        [cse]: 1.401e-05
        [replace_applicator]: 1.056e-05
    [Cycle 2]: 0.00030237, [27]
        [switch_simplify]: 5.09e-06
        [loop_unroll]: 4.90999e-06
        [a_1]: 8.582e-05
        [with_stream_mark]: 7.58999e-06
        [recompute_prepare]: 4.38001e-06
        [updatestate_depend_eliminate]: 2.34001e-06
        [updatestate_assign_eliminate]: 2.08002e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 5.14998e-06
        [updatestate_useless_node_eliminater]: 4.65999e-06
        [accelerated_algorithm]: 4.83001e-06
        [meta_shard_fg_expand]: 1.03001e-06
        [get_grad_eliminate_]: 4.37e-06
        [merge_forward]: 2.16e-06
        [cell_reuse_recompute_pass]: 1.55001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.148e-05
        [j_node_and_user_rematch]: 7.3e-06
        [meta_fg_expand]: 1.35001e-06
        [replace_old_param]: 1.074e-05
        [inline_without_move]: 4.47e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 6.90023e-07
        [auto_monad_eliminator]: 5.26998e-06
        [cse]: 8.82e-06
        [replace_applicator]: 4.82e-06
[py_interpret_to_execute_after_opt_a]: 8.28999e-06
[rewriter_after_opt_a]: 2.49e-05
[convert_after_rewriter]: 5.18002e-06
[order_py_execute_after_rewriter]: 4.42e-06
[mutable_eliminate]: 0.00049161
[jit_opt_b]: 4.612e-05, [1]
    [Cycle 1]: 4.033e-05, [2]
        [frontend_op_eliminate]: 1.499e-05
        [inline_after_opt_a]: 1.456e-05
[cconv]: 1.964e-05
[loop_unroll]: 0.00039986
[jit_opt_after_cconv]: 0.00012804, [1]
    [Cycle 1]: 0.00012206, [11]
        [c_1]: 1.973e-05
        [parameter_eliminate]: 2.75997e-06
        [updatestate_depend_eliminate]: 4.05998e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [cse]: 1.58e-05
        [call_graph_tuple_transform]: 1.568e-05
        [tuple_list_get_item_eliminator]: 4.89998e-06
        [none_parameter_eliminate]: 9.90025e-07
        [renormalize]: 4.89992e-07
        [switch_simplify]: 5.35999e-06
[remove_dup_value]: 8.45999e-06
[partial_unused_args_eliminate]: 1.80001e-06
[environ_conv]: 3.86999e-06
[add_recomputation]: 3.403e-05
[cse_after_recomputation]: 1.833e-05, [1]
    [Cycle 1]: 1.356e-05, [1]
        [cse]: 8.05e-06
[auto_monad_reorder]: 1.026e-05
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 4.22e-06
[opt_after_jit_grad]: 0.00043015
[symbol_engine_optimizer]: 6.845e-05, [1]
    [Cycle 1]: 6.236e-05, [6]
        [build]: 2.26e-06
        [elim_shapecalc]: 6.93998e-06
        [elim_not_effective]: 1.214e-05
        [opt_reshape]: 5.19e-06
        [fold_const_symbol]: 8.56002e-06
        [renormalize]: 4.80009e-07
[validate]: 2.607e-05
[backend_pass]: 6.69999e-07
[task_emit]: 0.00469309
[execute]: 4.46002e-06
Sums
    bootstrap                            :     0.000271s :  2.23%
    type_inference                       :     0.004036s : 33.18%
    event_method                         :     0.000014s :  0.12%
    auto_monad                           :     0.000056s :  0.46%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.25%
    rewriter_before_opt_a                :     0.000065s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000044s :  0.36%
    jit_opt_a.loop_unroll                :     0.000032s :  0.27%
    jit_opt_a.a_1                        :     0.000585s :  4.81%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000015s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000474s :  3.90%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.12%
    jit_opt_a.cse                        :     0.000023s :  0.19%
    jit_opt_a.replace_applicator         :     0.000015s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.07%
    rewriter_after_opt_a                 :     0.000025s :  0.20%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000492s :  4.04%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000020s :  0.16%
    loop_unroll                          :     0.000400s :  3.29%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000008s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000034s :  0.28%
    cse_after_recomputation.cse          :     0.000008s :  0.07%
    auto_monad_reorder                   :     0.000010s :  0.08%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000430s :  3.54%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000026s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.004693s : 38.59%
    execute                              :     0.000004s :  0.04%


Time group info:
------[substitution.]   0.000198    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 1.98% :     0.000004s :      3: substitution.graph_param_transform
85.53% :     0.000169s :      4: substitution.inline
 1.24% :     0.000002s :      4: substitution.j_node_and_user_rematch
 1.99% :     0.000004s :      4: substitution.remove_not_recompute_node
 3.35% :     0.000007s :      2: substitution.replace_old_param
 4.48% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003988     2
80.78% :     0.003221s :      1: type_inference.infer
19.22% :     0.000766s :      1: type_inference.specialize
------[replace.]   0.000058     6
67.29% :     0.000039s :      4: replace.inline
32.71% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000175     6
95.54% :     0.000167s :      4: match.inline
 4.46% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000107   691
 1.40% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.05% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.74% :     0.000002s :     10: predicate.addn_zero_filter
 1.66% :     0.000002s :     10: predicate.arithmetic_simplify
 1.39% :     0.000001s :     10: predicate.cast_eliminate
 0.74% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.60% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.01% :     0.000005s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.86% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.20% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.19% :     0.000001s :     10: predicate.minmaximum_grad
 2.02% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000002s :     15: predicate.partial_eliminate
 1.23% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.90% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.37% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.67% :     0.000008s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.30% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.41% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.10% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.47% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000483    12
48.37% :     0.000234s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.67% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.95% :     0.000222s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016992    76
 0.22% :     0.000037s :      1: add_recomputation
 0.34% :     0.000058s :      1: auto_monad
 0.07% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.65% :     0.000280s :      1: bootstrap
 0.13% :     0.000022s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000020s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.10% :     0.000017s :      1: event_method
 0.05% :     0.000008s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
26.94% :     0.004577s :      1: jit_opt_a
 0.77% :     0.000131s :      1: jit_opt_after_cconv
 0.29% :     0.000049s :      1: jit_opt_b
 2.39% :     0.000406s :      1: loop_unroll
 2.93% :     0.000498s :      1: mutable_eliminate
 4.54% :     0.000771s :     26: opt.transform.jit_opt_a
 0.25% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000023s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.57% :     0.000437s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000011s :      1: remove_dup_value
 1.44% :     0.000245s :      1: renormalize.infer
 1.31% :     0.000223s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000028s :      1: rewriter_after_opt_a
 0.40% :     0.000068s :      1: rewriter_before_opt_a
 0.42% :     0.000071s :      1: symbol_engine_optimizer
27.67% :     0.004702s :      1: task_emit
23.79% :     0.004043s :      1: type_inference
 0.25% :     0.000042s :      1: validate


TotalTime = 0.014068, [30]
[bootstrap]: 0.00026551
[type_inference]: 0.0064375
[event_method]: 2.113e-05
[auto_monad]: 8.143e-05
[graph_reusing]: 5.51e-06
[pre_auto_parallel]: 2.49999e-06
[py_interpret_to_execute]: 4.966e-05
[rewriter_before_opt_a]: 8.204e-05
[expand_dump_flag]: 2.89001e-06
[jit_opt_a]: 0.00522036, [2]
    [Cycle 1]: 0.00177405, [27]
        [switch_simplify]: 0.00010615
        [loop_unroll]: 3.359e-05
        [a_1]: 0.00064615
        [with_stream_mark]: 9.57999e-06
        [recompute_prepare]: 6.79999e-06
        [updatestate_depend_eliminate]: 2.69001e-06
        [updatestate_assign_eliminate]: 2.06998e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 6.24999e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 6.06998e-06
        [meta_shard_fg_expand]: 1.22999e-06
        [get_grad_eliminate_]: 5.99e-06
        [merge_forward]: 2.58e-06
        [cell_reuse_recompute_pass]: 8.30012e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.141e-05
        [j_node_and_user_rematch]: 9.91998e-06
        [meta_fg_expand]: 1.91998e-06
        [replace_old_param]: 8.37e-06
        [inline_without_move]: 5.86e-06
        [renormalize]: 0.00070639
        [add_forward_monad_depend]: 3.85998e-06
        [auto_monad_grad]: 1.03001e-06
        [auto_monad_eliminator]: 7.70998e-06
        [cse]: 1.343e-05
        [replace_applicator]: 9.77999e-06
    [Cycle 2]: 0.00032182, [27]
        [switch_simplify]: 5.46e-06
        [loop_unroll]: 5.17e-06
        [a_1]: 9.118e-05
        [with_stream_mark]: 7.65e-06
        [recompute_prepare]: 4.89e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 9.00007e-07
        [specialize_transform]: 4.98001e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.17999e-06
        [meta_shard_fg_expand]: 1.25001e-06
        [get_grad_eliminate_]: 4.48001e-06
        [merge_forward]: 2.33002e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.174e-05
        [j_node_and_user_rematch]: 7.95998e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 7.08998e-06
        [inline_without_move]: 4.94998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.14998e-06
        [auto_monad_grad]: 6.50005e-07
        [auto_monad_eliminator]: 5.04e-06
        [cse]: 1.017e-05
        [replace_applicator]: 4.79e-06
[py_interpret_to_execute_after_opt_a]: 7.82e-06
[rewriter_after_opt_a]: 2.4e-05
[convert_after_rewriter]: 5.24e-06
[order_py_execute_after_rewriter]: 3.92998e-06
[mutable_eliminate]: 0.00048602
[jit_opt_b]: 4.74e-05, [1]
    [Cycle 1]: 4.14e-05, [2]
        [frontend_op_eliminate]: 1.62e-05
        [inline_after_opt_a]: 1.406e-05
[cconv]: 1.561e-05
[loop_unroll]: 0.00040633
[jit_opt_after_cconv]: 0.00013011, [1]
    [Cycle 1]: 0.00012394, [11]
        [c_1]: 2.073e-05
        [parameter_eliminate]: 1.92999e-06
        [updatestate_depend_eliminate]: 4.55999e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.49e-05
        [call_graph_tuple_transform]: 1.664e-05
        [tuple_list_get_item_eliminator]: 5.25999e-06
        [none_parameter_eliminate]: 9.40025e-07
        [renormalize]: 3.59985e-07
        [switch_simplify]: 5.68002e-06
[remove_dup_value]: 6.74999e-06
[partial_unused_args_eliminate]: 1.50001e-06
[environ_conv]: 3.38e-06
[add_recomputation]: 3.028e-05
[cse_after_recomputation]: 1.893e-05, [1]
    [Cycle 1]: 1.393e-05, [1]
        [cse]: 8.24002e-06
[auto_monad_reorder]: 9.20999e-06
[get_jit_bprop_graph]: 1.43002e-06
[rewriter_after_jit_bprop_graph]: 4.15e-06
[opt_after_jit_grad]: 0.00044545
[symbol_engine_optimizer]: 6.916e-05, [1]
    [Cycle 1]: 6.229e-05, [6]
        [build]: 2.32001e-06
        [elim_shapecalc]: 6.95998e-06
        [elim_not_effective]: 1.234e-05
        [opt_reshape]: 5.47999e-06
        [fold_const_symbol]: 8.07e-06
        [renormalize]: 4.50003e-07
[validate]: 2.54e-05
Sums
    bootstrap                            :     0.000266s :  2.55%
    type_inference                       :     0.006438s : 61.94%
    event_method                         :     0.000021s :  0.20%
    auto_monad                           :     0.000081s :  0.78%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000050s :  0.48%
    rewriter_before_opt_a                :     0.000082s :  0.79%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000112s :  1.07%
    jit_opt_a.loop_unroll                :     0.000039s :  0.37%
    jit_opt_a.a_1                        :     0.000737s :  7.09%
    jit_opt_a.with_stream_mark           :     0.000017s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000706s :  6.80%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.12%
    jit_opt_a.cse                        :     0.000024s :  0.23%
    jit_opt_a.replace_applicator         :     0.000015s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.08%
    rewriter_after_opt_a                 :     0.000024s :  0.23%
    convert_after_rewriter               :     0.000005s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000486s :  4.68%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.14%
    cconv                                :     0.000016s :  0.15%
    loop_unroll                          :     0.000406s :  3.91%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000007s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000003s :  0.03%
    add_recomputation                    :     0.000030s :  0.29%
    cse_after_recomputation.cse          :     0.000008s :  0.08%
    auto_monad_reorder                   :     0.000009s :  0.09%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.04%
    opt_after_jit_grad                   :     0.000445s :  4.29%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000025s :  0.24%


Time group info:
------[substitution.]   0.000234    26
 0.58% :     0.000001s :      2: substitution.elim_not_effective
 0.39% :     0.000001s :      2: substitution.fold_const_symbol
 1.63% :     0.000004s :      3: substitution.graph_param_transform
86.15% :     0.000201s :      6: substitution.inline
 1.37% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.83% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.08% :     0.000003s :      2: substitution.replace_old_param
 3.40% :     0.000008s :      1: substitution.switch_simplify
 3.59% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006380     2
81.36% :     0.005191s :      1: type_inference.infer
18.64% :     0.001189s :      1: type_inference.specialize
------[replace.]   0.000125     9
51.26% :     0.000064s :      6: replace.inline
24.89% :     0.000031s :      1: replace.switch_simplify
23.84% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000213     9
93.26% :     0.000198s :      6: match.inline
 3.36% :     0.000007s :      1: match.switch_simplify
 3.39% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   801
 1.40% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.00% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.17% :     0.000001s :     12: predicate.addn_check_dump
 1.58% :     0.000002s :     12: predicate.addn_zero_filter
 1.85% :     0.000002s :     12: predicate.arithmetic_simplify
 1.41% :     0.000002s :     12: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.18% :     0.000001s :     12: predicate.compare_switch_simplify
 1.63% :     0.000002s :     12: predicate.depend_value_elim
 1.24% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.65% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     12: predicate.environ_get_eliminate
 1.18% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 4.92% :     0.000006s :     26: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.18% :     0.000001s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.87% :     0.000002s :     17: predicate.load_eliminater
 1.05% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.21% :     0.000005s :     40: predicate.loop_unroll_before_grad
 2.01% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000002s :     12: predicate.merge_addn
 1.22% :     0.000002s :     12: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000003s :     17: predicate.partial_eliminate
 1.24% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.93% :     0.000002s :     12: predicate.reduce_eliminate
 1.52% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     20: predicate.replace_applicator
 0.45% :     0.000001s :      6: predicate.replace_old_param
 0.42% :     0.000001s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     12: predicate.reshape_eliminate
 1.33% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     12: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.73% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.61% :     0.000003s :     20: predicate.switch_defer_inline
 2.46% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.75% :     0.000011s :     65: predicate.switch_simplify
 1.37% :     0.000002s :     12: predicate.tile_eliminate
 1.30% :     0.000002s :     12: predicate.transpose_eliminate
 1.49% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 2.95% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.92% :     0.000002s :     12: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000846    15
45.58% :     0.000385s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.86% :     0.000092s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.56% :     0.000368s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015834    72
 0.21% :     0.000033s :      1: add_recomputation
 0.54% :     0.000085s :      1: auto_monad
 0.07% :     0.000012s :      1: auto_monad_reorder
 1.75% :     0.000277s :      1: bootstrap
 0.11% :     0.000018s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.13% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.16% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
33.00% :     0.005225s :      1: jit_opt_a
 0.84% :     0.000133s :      1: jit_opt_after_cconv
 0.32% :     0.000050s :      1: jit_opt_b
 2.61% :     0.000413s :      1: loop_unroll
 3.12% :     0.000494s :      1: mutable_eliminate
 6.31% :     0.000999s :     26: opt.transform.jit_opt_a
 0.28% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000018s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.86% :     0.000452s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.33% :     0.000052s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000009s :      1: remove_dup_value
 2.22% :     0.000352s :      1: renormalize.infer
 2.20% :     0.000348s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000027s :      1: rewriter_after_opt_a
 0.54% :     0.000085s :      1: rewriter_before_opt_a
 0.45% :     0.000071s :      1: symbol_engine_optimizer
40.73% :     0.006449s :      1: type_inference


TotalTime = 0.0160855, [33]
[bootstrap]: 0.00024293
[type_inference]: 0.00381004
[event_method]: 1.405e-05
[auto_monad]: 3.557e-05
[graph_reusing]: 2.97002e-06
[pre_auto_parallel]: 1.61998e-06
[py_interpret_to_execute]: 2.512e-05
[rewriter_before_opt_a]: 7.087e-05
[expand_dump_flag]: 1.94999e-06
[jit_opt_a]: 0.00462683, [2]
    [Cycle 1]: 0.0013491, [27]
        [switch_simplify]: 4.091e-05
        [loop_unroll]: 2.75e-05
        [a_1]: 0.00046629
        [with_stream_mark]: 1.523e-05
        [recompute_prepare]: 6.71999e-06
        [updatestate_depend_eliminate]: 3.50998e-06
        [updatestate_assign_eliminate]: 3.26001e-06
        [updatestate_loads_eliminate]: 2.36998e-06
        [parameter_eliminate]: 2.14999e-06
        [specialize_transform]: 5.97001e-06
        [updatestate_useless_node_eliminater]: 4.84998e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 1.77999e-06
        [get_grad_eliminate_]: 4.81997e-06
        [merge_forward]: 3.51999e-06
        [cell_reuse_recompute_pass]: 9.79984e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.334e-05
        [j_node_and_user_rematch]: 7.88999e-06
        [meta_fg_expand]: 1.92001e-06
        [replace_old_param]: 7.81001e-06
        [inline_without_move]: 5.10001e-06
        [renormalize]: 0.00050892
        [add_forward_monad_depend]: 4.25999e-06
        [auto_monad_grad]: 2.01003e-06
        [auto_monad_eliminator]: 1.174e-05
        [cse]: 1.646e-05
        [replace_applicator]: 9.53002e-06
    [Cycle 2]: 0.0003151, [27]
        [switch_simplify]: 5.17999e-06
        [loop_unroll]: 4.71002e-06
        [a_1]: 8.726e-05
        [with_stream_mark]: 8.48999e-06
        [recompute_prepare]: 5.32001e-06
        [updatestate_depend_eliminate]: 2.53998e-06
        [updatestate_assign_eliminate]: 2.10002e-06
        [updatestate_loads_eliminate]: 1.76e-06
        [parameter_eliminate]: 9.10019e-07
        [specialize_transform]: 4.98001e-06
        [updatestate_useless_node_eliminater]: 4.67998e-06
        [accelerated_algorithm]: 4.95001e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 4.37998e-06
        [merge_forward]: 2.64001e-06
        [cell_reuse_recompute_pass]: 2.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.195e-05
        [j_node_and_user_rematch]: 7.77e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 7.25e-06
        [inline_without_move]: 4.60999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.64998e-06
        [auto_monad_grad]: 9.99979e-07
        [auto_monad_eliminator]: 4.84e-06
        [cse]: 9.76e-06
        [replace_applicator]: 5.42001e-06
[py_interpret_to_execute_after_opt_a]: 1.002e-05
[rewriter_after_opt_a]: 3.278e-05
[convert_after_rewriter]: 6.73998e-06
[order_py_execute_after_rewriter]: 5.04e-06
[mutable_eliminate]: 0.00052249
[jit_opt_b]: 4.631e-05, [1]
    [Cycle 1]: 3.967e-05, [2]
        [frontend_op_eliminate]: 1.486e-05
        [inline_after_opt_a]: 1.364e-05
[cconv]: 2.365e-05
[loop_unroll]: 0.00037505
[jit_opt_after_cconv]: 0.00012803, [1]
    [Cycle 1]: 0.0001227, [11]
        [c_1]: 1.856e-05
        [parameter_eliminate]: 2.85002e-06
        [updatestate_depend_eliminate]: 5.22e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.526e-05
        [call_graph_tuple_transform]: 1.741e-05
        [tuple_list_get_item_eliminator]: 4.85001e-06
        [none_parameter_eliminate]: 1.49998e-06
        [renormalize]: 3.80009e-07
        [switch_simplify]: 5.00999e-06
[remove_dup_value]: 1.227e-05
[partial_unused_args_eliminate]: 2.26998e-06
[environ_conv]: 5.04e-06
[add_recomputation]: 4.192e-05
[cse_after_recomputation]: 1.947e-05, [1]
    [Cycle 1]: 1.422e-05, [1]
        [cse]: 8.23999e-06
[auto_monad_reorder]: 1.327e-05
[get_jit_bprop_graph]: 1.66e-06
[rewriter_after_jit_bprop_graph]: 4.80999e-06
[opt_after_jit_grad]: 0.00040936
[symbol_engine_optimizer]: 6.735e-05, [1]
    [Cycle 1]: 6.195e-05, [6]
        [build]: 2.91e-06
        [elim_shapecalc]: 6.48003e-06
        [elim_not_effective]: 1.186e-05
        [opt_reshape]: 5.40001e-06
        [fold_const_symbol]: 8.23999e-06
        [renormalize]: 3.80009e-07
[validate]: 3.098e-05
[backend_pass]: 8.69972e-07
[task_emit]: 0.00533134
[execute]: 5.51998e-06
Sums
    bootstrap                            :     0.000243s :  1.93%
    type_inference                       :     0.003810s : 30.31%
    event_method                         :     0.000014s :  0.11%
    auto_monad                           :     0.000036s :  0.28%
    graph_reusing                        :     0.000003s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000025s :  0.20%
    rewriter_before_opt_a                :     0.000071s :  0.56%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000046s :  0.37%
    jit_opt_a.loop_unroll                :     0.000032s :  0.26%
    jit_opt_a.a_1                        :     0.000554s :  4.40%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000509s :  4.05%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000026s :  0.21%
    jit_opt_a.replace_applicator         :     0.000015s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.08%
    rewriter_after_opt_a                 :     0.000033s :  0.26%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000522s :  4.16%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000024s :  0.19%
    loop_unroll                          :     0.000375s :  2.98%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000042s :  0.33%
    cse_after_recomputation.cse          :     0.000008s :  0.07%
    auto_monad_reorder                   :     0.000013s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000409s :  3.26%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005331s : 42.41%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000164    23
 1.24% :     0.000002s :      2: substitution.elim_not_effective
 0.77% :     0.000001s :      2: substitution.fold_const_symbol
 3.30% :     0.000005s :      3: substitution.graph_param_transform
81.07% :     0.000133s :      4: substitution.inline
 1.55% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.91% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.29% :     0.000004s :      2: substitution.replace_old_param
 6.87% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003766     2
82.59% :     0.003111s :      1: type_inference.infer
17.41% :     0.000656s :      1: type_inference.specialize
------[replace.]   0.000057     6
66.07% :     0.000038s :      4: replace.inline
33.93% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000141     6
92.78% :     0.000131s :      4: match.inline
 7.22% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000107   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.17% :     0.000001s :     10: predicate.addn_check_dump
 1.62% :     0.000002s :     10: predicate.addn_zero_filter
 1.77% :     0.000002s :     10: predicate.arithmetic_simplify
 1.64% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.17% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000001s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.49% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.33% :     0.000001s :      6: predicate.less_batch_normalization
 1.70% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.03% :     0.000002s :     15: predicate.load_eliminater
 1.17% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000001s :     10: predicate.merge_addn
 1.17% :     0.000001s :     10: predicate.minmaximum_grad
 1.50% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.54% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.72% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.76% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.23% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.56% :     0.000008s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.63% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.41% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000459    12
47.54% :     0.000218s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.69% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.76% :     0.000210s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017429    76
 0.26% :     0.000045s :      1: add_recomputation
 0.22% :     0.000039s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.45% :     0.000253s :      1: bootstrap
 0.15% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000017s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000005s :      1: graph_reusing
26.56% :     0.004629s :      1: jit_opt_a
 0.75% :     0.000130s :      1: jit_opt_after_cconv
 0.28% :     0.000049s :      1: jit_opt_b
 2.18% :     0.000381s :      1: loop_unroll
 3.03% :     0.000529s :      1: mutable_eliminate
 4.25% :     0.000741s :     26: opt.transform.jit_opt_a
 0.25% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.05% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000017s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.38% :     0.000415s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000028s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 1.56% :     0.000271s :      1: renormalize.infer
 1.33% :     0.000231s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000036s :      1: rewriter_after_opt_a
 0.42% :     0.000074s :      1: rewriter_before_opt_a
 0.40% :     0.000069s :      1: symbol_engine_optimizer
30.64% :     0.005340s :      1: task_emit
21.90% :     0.003817s :      1: type_inference
 0.28% :     0.000048s :      1: validate


TotalTime = 0.0182187, [33]
[bootstrap]: 0.00032485
[type_inference]: 0.00444047
[event_method]: 1.527e-05
[auto_monad]: 4.368e-05
[graph_reusing]: 3.36001e-06
[pre_auto_parallel]: 1.87001e-06
[py_interpret_to_execute]: 3.24e-05
[rewriter_before_opt_a]: 6.325e-05
[expand_dump_flag]: 2.11998e-06
[jit_opt_a]: 0.00504724, [2]
    [Cycle 1]: 0.00156213, [27]
        [switch_simplify]: 5.34e-05
        [loop_unroll]: 2.84e-05
        [a_1]: 0.00050964
        [with_stream_mark]: 1.69e-05
        [recompute_prepare]: 6.98e-06
        [updatestate_depend_eliminate]: 3.3e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.70002e-06
        [parameter_eliminate]: 2.07999e-06
        [specialize_transform]: 5.89e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 5.37001e-06
        [merge_forward]: 3.40998e-06
        [cell_reuse_recompute_pass]: 1.35001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.341e-05
        [j_node_and_user_rematch]: 8.33999e-06
        [meta_fg_expand]: 2.29999e-06
        [replace_old_param]: 8.53001e-06
        [inline_without_move]: 5.61998e-06
        [renormalize]: 0.00064777
        [add_forward_monad_depend]: 5.02999e-06
        [auto_monad_grad]: 2.15002e-06
        [auto_monad_eliminator]: 1.269e-05
        [cse]: 2.978e-05
        [replace_applicator]: 1.186e-05
    [Cycle 2]: 0.00032223, [27]
        [switch_simplify]: 5.44e-06
        [loop_unroll]: 4.85999e-06
        [a_1]: 8.919e-05
        [with_stream_mark]: 8.3e-06
        [recompute_prepare]: 5.15001e-06
        [updatestate_depend_eliminate]: 2.92002e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 4.92999e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 4.89e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.57e-06
        [merge_forward]: 1.138e-05
        [cell_reuse_recompute_pass]: 1.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.195e-05
        [j_node_and_user_rematch]: 7.66001e-06
        [meta_fg_expand]: 1.47999e-06
        [replace_old_param]: 7.51999e-06
        [inline_without_move]: 4.70001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 5.47001e-06
        [cse]: 1.116e-05
        [replace_applicator]: 4.87e-06
[py_interpret_to_execute_after_opt_a]: 1.037e-05
[rewriter_after_opt_a]: 3.646e-05
[convert_after_rewriter]: 7.3e-06
[order_py_execute_after_rewriter]: 5.32999e-06
[mutable_eliminate]: 0.0005453
[jit_opt_b]: 4.639e-05, [1]
    [Cycle 1]: 3.933e-05, [2]
        [frontend_op_eliminate]: 1.563e-05
        [inline_after_opt_a]: 1.358e-05
[cconv]: 2.41e-05
[loop_unroll]: 0.00043003
[jit_opt_after_cconv]: 0.00015469, [1]
    [Cycle 1]: 0.00014827, [11]
        [c_1]: 1.954e-05
        [parameter_eliminate]: 2.24999e-06
        [updatestate_depend_eliminate]: 4.57e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 1.425e-05
        [cse]: 1.95e-05
        [call_graph_tuple_transform]: 1.919e-05
        [tuple_list_get_item_eliminator]: 6.02001e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.71e-06
[remove_dup_value]: 1.378e-05
[partial_unused_args_eliminate]: 2.26998e-06
[environ_conv]: 4.87e-06
[add_recomputation]: 5.005e-05
[cse_after_recomputation]: 2.059e-05, [1]
    [Cycle 1]: 1.522e-05, [1]
        [cse]: 9.44e-06
[auto_monad_reorder]: 1.519e-05
[get_jit_bprop_graph]: 2.12999e-06
[rewriter_after_jit_bprop_graph]: 6.25002e-06
[opt_after_jit_grad]: 0.00046163
[symbol_engine_optimizer]: 7.04e-05, [1]
    [Cycle 1]: 6.471e-05, [6]
        [build]: 3.53999e-06
        [elim_shapecalc]: 7.78999e-06
        [elim_not_effective]: 1.23e-05
        [opt_reshape]: 5.35001e-06
        [fold_const_symbol]: 8.69003e-06
        [renormalize]: 4.50003e-07
[validate]: 3.509e-05
[backend_pass]: 1.00999e-06
[task_emit]: 0.00609034
[execute]: 9.42999e-06
Sums
    bootstrap                            :     0.000325s :  2.25%
    type_inference                       :     0.004440s : 30.70%
    event_method                         :     0.000015s :  0.11%
    auto_monad                           :     0.000044s :  0.30%
    graph_reusing                        :     0.000003s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000032s :  0.22%
    rewriter_before_opt_a                :     0.000063s :  0.44%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000059s :  0.41%
    jit_opt_a.loop_unroll                :     0.000033s :  0.23%
    jit_opt_a.a_1                        :     0.000599s :  4.14%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000015s :  0.10%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000648s :  4.48%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000041s :  0.28%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000545s :  3.77%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000430s :  2.97%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000014s :  0.10%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.35%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000462s :  3.19%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006090s : 42.11%
    execute                              :     0.000009s :  0.07%


Time group info:
------[substitution.]   0.000190    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 3.09% :     0.000006s :      3: substitution.graph_param_transform
82.90% :     0.000157s :      4: substitution.inline
 1.53% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.46% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.91% :     0.000004s :      2: substitution.replace_old_param
 6.45% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004388     2
83.25% :     0.003653s :      1: type_inference.infer
16.75% :     0.000735s :      1: type_inference.specialize
------[replace.]   0.000065     6
66.39% :     0.000043s :      4: replace.inline
33.61% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000166     6
93.30% :     0.000155s :      4: match.inline
 6.70% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 2.03% :     0.000002s :     10: predicate.arithmetic_simplify
 1.50% :     0.000002s :     10: predicate.cast_eliminate
 0.69% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000001s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.17% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.90% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.29% :     0.000001s :     10: predicate.minmaximum_grad
 1.29% :     0.000001s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.34% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.63% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.53% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.37% :     0.000008s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.36% :     0.000002s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.23% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000569    12
51.86% :     0.000295s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.27% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.87% :     0.000244s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019759    76
 0.27% :     0.000053s :      1: add_recomputation
 0.24% :     0.000047s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.74% :     0.000345s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.09% :     0.000019s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
25.56% :     0.005050s :      1: jit_opt_a
 0.80% :     0.000158s :      1: jit_opt_after_cconv
 0.25% :     0.000049s :      1: jit_opt_b
 2.22% :     0.000438s :      1: loop_unroll
 2.80% :     0.000553s :      1: mutable_eliminate
 4.06% :     0.000802s :     26: opt.transform.jit_opt_a
 0.24% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000010s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.37% :     0.000469s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000035s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.81% :     0.000357s :      1: renormalize.infer
 1.43% :     0.000283s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000039s :      1: rewriter_after_opt_a
 0.33% :     0.000066s :      1: rewriter_before_opt_a
 0.37% :     0.000073s :      1: symbol_engine_optimizer
30.91% :     0.006107s :      1: task_emit
22.53% :     0.004451s :      1: type_inference
 0.28% :     0.000056s :      1: validate


TotalTime = 0.0235011, [33]
[bootstrap]: 0.00032545
[type_inference]: 0.00495429
[event_method]: 1.744e-05
[auto_monad]: 6.643e-05
[graph_reusing]: 5.67001e-06
[pre_auto_parallel]: 2.54999e-06
[py_interpret_to_execute]: 3.818e-05
[rewriter_before_opt_a]: 7.494e-05
[expand_dump_flag]: 3.36001e-06
[jit_opt_a]: 0.00747521, [2]
    [Cycle 1]: 0.00202381, [27]
        [switch_simplify]: 6.895e-05
        [loop_unroll]: 2.875e-05
        [a_1]: 0.00058064
        [with_stream_mark]: 2.353e-05
        [recompute_prepare]: 7.26999e-06
        [updatestate_depend_eliminate]: 3.61999e-06
        [updatestate_assign_eliminate]: 3.69002e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 2.21e-06
        [specialize_transform]: 6.65002e-06
        [updatestate_useless_node_eliminater]: 5.24998e-06
        [accelerated_algorithm]: 5.40999e-06
        [meta_shard_fg_expand]: 2.06998e-06
        [get_grad_eliminate_]: 5.20001e-06
        [merge_forward]: 3.73001e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.533e-05
        [j_node_and_user_rematch]: 9.64e-06
        [meta_fg_expand]: 2.17001e-06
        [replace_old_param]: 9.91e-06
        [inline_without_move]: 5.66998e-06
        [renormalize]: 0.00096801
        [add_forward_monad_depend]: 6.58e-06
        [auto_monad_grad]: 3.36999e-06
        [auto_monad_eliminator]: 1.627e-05
        [cse]: 3.368e-05
        [replace_applicator]: 1.559e-05
    [Cycle 2]: 0.00033199, [27]
        [switch_simplify]: 6.04999e-06
        [loop_unroll]: 5.05001e-06
        [a_1]: 9.401e-05
        [with_stream_mark]: 1.26e-05
        [recompute_prepare]: 5.18002e-06
        [updatestate_depend_eliminate]: 2.66e-06
        [updatestate_assign_eliminate]: 1.98002e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.09998e-06
        [specialize_transform]: 5.13002e-06
        [updatestate_useless_node_eliminater]: 4.45999e-06
        [accelerated_algorithm]: 4.63001e-06
        [meta_shard_fg_expand]: 1.56002e-06
        [get_grad_eliminate_]: 4.48001e-06
        [merge_forward]: 3.41999e-06
        [cell_reuse_recompute_pass]: 1.90001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.324e-05
        [j_node_and_user_rematch]: 9.47999e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 8.06001e-06
        [inline_without_move]: 5.11002e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 5.56002e-06
        [cse]: 1.127e-05
        [replace_applicator]: 5.14e-06
[py_interpret_to_execute_after_opt_a]: 1.326e-05
[rewriter_after_opt_a]: 3.896e-05
[convert_after_rewriter]: 7.84002e-06
[order_py_execute_after_rewriter]: 5.20001e-06
[mutable_eliminate]: 0.00080983
[jit_opt_b]: 5.143e-05, [1]
    [Cycle 1]: 4.367e-05, [2]
        [frontend_op_eliminate]: 1.663e-05
        [inline_after_opt_a]: 1.48e-05
[cconv]: 3.176e-05
[loop_unroll]: 0.00050535
[jit_opt_after_cconv]: 0.00021986, [1]
    [Cycle 1]: 0.00021301, [11]
        [c_1]: 2.214e-05
        [parameter_eliminate]: 4.33999e-06
        [updatestate_depend_eliminate]: 8.92e-06
        [updatestate_assign_eliminate]: 2.94001e-06
        [updatestate_loads_eliminate]: 2.84999e-06
        [cse]: 2.672e-05
        [call_graph_tuple_transform]: 2.467e-05
        [tuple_list_get_item_eliminator]: 5.56998e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 1.22e-06
        [switch_simplify]: 5.74e-06
[remove_dup_value]: 1.674e-05
[partial_unused_args_eliminate]: 3.11999e-06
[environ_conv]: 6.70002e-06
[add_recomputation]: 5.651e-05
[cse_after_recomputation]: 2.312e-05, [1]
    [Cycle 1]: 1.714e-05, [1]
        [cse]: 1.098e-05
[auto_monad_reorder]: 1.688e-05
[get_jit_bprop_graph]: 1.90001e-06
[rewriter_after_jit_bprop_graph]: 7.05e-06
[opt_after_jit_grad]: 0.00050922
[symbol_engine_optimizer]: 7.871e-05, [1]
    [Cycle 1]: 7.223e-05, [6]
        [build]: 3.95e-06
        [elim_shapecalc]: 8.33999e-06
        [elim_not_effective]: 1.365e-05
        [opt_reshape]: 6.31998e-06
        [fold_const_symbol]: 9.57999e-06
        [renormalize]: 1.07e-06
[validate]: 4.155e-05
[backend_pass]: 1.24998e-06
[task_emit]: 0.00784963
[execute]: 9.42001e-06
Sums
    bootstrap                            :     0.000325s :  1.84%
    type_inference                       :     0.004954s : 28.04%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000066s :  0.38%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000038s :  0.22%
    rewriter_before_opt_a                :     0.000075s :  0.42%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000075s :  0.42%
    jit_opt_a.loop_unroll                :     0.000034s :  0.19%
    jit_opt_a.a_1                        :     0.000675s :  3.82%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000968s :  5.48%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.12%
    jit_opt_a.cse                        :     0.000045s :  0.25%
    jit_opt_a.replace_applicator         :     0.000021s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000039s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000810s :  4.58%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000032s :  0.18%
    loop_unroll                          :     0.000505s :  2.86%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000027s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000057s :  0.32%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000509s :  2.88%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000042s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007850s : 44.43%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000242    23
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.81% :     0.000002s :      2: substitution.fold_const_symbol
 3.14% :     0.000008s :      3: substitution.graph_param_transform
83.94% :     0.000203s :      4: substitution.inline
 1.53% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.05% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.21% :     0.000005s :      2: substitution.replace_old_param
 5.51% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004876     2
80.85% :     0.003942s :      1: type_inference.infer
19.15% :     0.000934s :      1: type_inference.specialize
------[replace.]   0.000080     6
69.51% :     0.000056s :      4: replace.inline
30.49% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000213     6
94.28% :     0.000201s :      4: match.inline
 5.72% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.56% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.57% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 1.89% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000001s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.21% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.61% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.07% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.36% :     0.000006s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000001s :      6: predicate.less_batch_normalization
 1.72% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.87% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.29% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.76% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.06% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     18: predicate.replace_applicator
 0.87% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.26% :     0.000002s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.67% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.18% :     0.000003s :     16: predicate.switch_defer_inline
 2.00% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.37% :     0.000010s :     53: predicate.switch_simplify
 1.24% :     0.000001s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.21% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.92% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000671    12
46.94% :     0.000315s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.18% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.87% :     0.000321s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025486    76
 0.23% :     0.000059s :      1: add_recomputation
 0.27% :     0.000070s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.34% :     0.000342s :      1: bootstrap
 0.14% :     0.000035s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
29.35% :     0.007480s :      1: jit_opt_a
 0.87% :     0.000223s :      1: jit_opt_after_cconv
 0.21% :     0.000054s :      1: jit_opt_b
 2.02% :     0.000515s :      1: loop_unroll
 3.23% :     0.000823s :      1: mutable_eliminate
 3.55% :     0.000905s :     26: opt.transform.jit_opt_a
 0.22% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.03% :     0.000518s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000041s :      1: py_interpret_to_execute
 0.06% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000019s :      1: remove_dup_value
 2.05% :     0.000523s :      1: renormalize.infer
 1.71% :     0.000435s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000043s :      1: rewriter_after_opt_a
 0.31% :     0.000078s :      1: rewriter_before_opt_a
 0.32% :     0.000081s :      1: symbol_engine_optimizer
30.88% :     0.007870s :      1: task_emit
19.50% :     0.004970s :      1: type_inference
 0.25% :     0.000064s :      1: validate


TotalTime = 0.0151179, [30]
[bootstrap]: 0.00028266
[type_inference]: 0.00696852
[event_method]: 1.912e-05
[auto_monad]: 5.948e-05
[graph_reusing]: 4.68001e-06
[pre_auto_parallel]: 2.01998e-06
[py_interpret_to_execute]: 6.799e-05
[rewriter_before_opt_a]: 7.671e-05
[expand_dump_flag]: 1.94999e-06
[jit_opt_a]: 0.00566742, [2]
    [Cycle 1]: 0.00209847, [27]
        [switch_simplify]: 0.00011751
        [loop_unroll]: 3.424e-05
        [a_1]: 0.00074837
        [with_stream_mark]: 1.259e-05
        [recompute_prepare]: 6.80002e-06
        [updatestate_depend_eliminate]: 2.66e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 1.14003e-06
        [specialize_transform]: 6.09001e-06
        [updatestate_useless_node_eliminater]: 5.56e-06
        [accelerated_algorithm]: 6.78e-06
        [meta_shard_fg_expand]: 1.28002e-06
        [get_grad_eliminate_]: 6.06e-06
        [merge_forward]: 2.82002e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.223e-05
        [j_node_and_user_rematch]: 1.017e-05
        [meta_fg_expand]: 1.83002e-06
        [replace_old_param]: 8.71002e-06
        [inline_without_move]: 5.95002e-06
        [renormalize]: 0.00090123
        [add_forward_monad_depend]: 4.1e-06
        [auto_monad_grad]: 1.07e-06
        [auto_monad_eliminator]: 8.63001e-06
        [cse]: 1.419e-05
        [replace_applicator]: 1.032e-05
    [Cycle 2]: 0.00032066, [27]
        [switch_simplify]: 5.81e-06
        [loop_unroll]: 5.12e-06
        [a_1]: 9.339e-05
        [with_stream_mark]: 9.39e-06
        [recompute_prepare]: 5.29e-06
        [updatestate_depend_eliminate]: 2.41e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.92999e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 4.95001e-06
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.70001e-06
        [merge_forward]: 2.56e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.227e-05
        [j_node_and_user_rematch]: 7.79997e-06
        [meta_fg_expand]: 1.44998e-06
        [replace_old_param]: 7.45e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.35999e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 4.82998e-06
        [cse]: 1.065e-05
        [replace_applicator]: 4.92e-06
[py_interpret_to_execute_after_opt_a]: 9.44e-06
[rewriter_after_opt_a]: 2.531e-05
[convert_after_rewriter]: 5.66e-06
[order_py_execute_after_rewriter]: 4.67e-06
[mutable_eliminate]: 0.00051856
[jit_opt_b]: 4.862e-05, [1]
    [Cycle 1]: 4.24e-05, [2]
        [frontend_op_eliminate]: 1.608e-05
        [inline_after_opt_a]: 1.56e-05
[cconv]: 1.564e-05
[loop_unroll]: 0.00041989
[jit_opt_after_cconv]: 0.00013214, [1]
    [Cycle 1]: 0.00012573, [11]
        [c_1]: 2.196e-05
        [parameter_eliminate]: 1.81e-06
        [updatestate_depend_eliminate]: 4.86002e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.23998e-06
        [cse]: 1.403e-05
        [call_graph_tuple_transform]: 1.623e-05
        [tuple_list_get_item_eliminator]: 5.70001e-06
        [none_parameter_eliminate]: 8.40024e-07
        [renormalize]: 4.59986e-07
        [switch_simplify]: 5.55001e-06
[remove_dup_value]: 7.51999e-06
[partial_unused_args_eliminate]: 1.49e-06
[environ_conv]: 3.33e-06
[add_recomputation]: 3.086e-05
[cse_after_recomputation]: 2.124e-05, [1]
    [Cycle 1]: 1.58e-05, [1]
        [cse]: 1.004e-05
[auto_monad_reorder]: 9.14998e-06
[get_jit_bprop_graph]: 1.44998e-06
[rewriter_after_jit_bprop_graph]: 4.57998e-06
[opt_after_jit_grad]: 0.00043309
[symbol_engine_optimizer]: 8.669e-05, [1]
    [Cycle 1]: 6.713e-05, [6]
        [build]: 2.69001e-06
        [elim_shapecalc]: 7.65998e-06
        [elim_not_effective]: 1.184e-05
        [opt_reshape]: 5.43002e-06
        [fold_const_symbol]: 8.84e-06
        [renormalize]: 3.99974e-07
[validate]: 2.775e-05
Sums
    bootstrap                            :     0.000283s :  2.50%
    type_inference                       :     0.006969s : 61.68%
    event_method                         :     0.000019s :  0.17%
    auto_monad                           :     0.000059s :  0.53%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000068s :  0.60%
    rewriter_before_opt_a                :     0.000077s :  0.68%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000123s :  1.09%
    jit_opt_a.loop_unroll                :     0.000039s :  0.35%
    jit_opt_a.a_1                        :     0.000842s :  7.45%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.14%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000901s :  7.98%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.12%
    jit_opt_a.cse                        :     0.000025s :  0.22%
    jit_opt_a.replace_applicator         :     0.000015s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.08%
    rewriter_after_opt_a                 :     0.000025s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000519s :  4.59%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.14%
    cconv                                :     0.000016s :  0.14%
    loop_unroll                          :     0.000420s :  3.72%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000014s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000008s :  0.07%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000003s :  0.03%
    add_recomputation                    :     0.000031s :  0.27%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000009s :  0.08%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000433s :  3.83%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.25%


Time group info:
------[substitution.]   0.000267    26
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.35% :     0.000001s :      2: substitution.fold_const_symbol
 1.34% :     0.000004s :      3: substitution.graph_param_transform
87.47% :     0.000233s :      6: substitution.inline
 1.00% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.59% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.06% :     0.000003s :      2: substitution.replace_old_param
 3.02% :     0.000008s :      1: substitution.switch_simplify
 3.40% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006913     2
82.14% :     0.005679s :      1: type_inference.infer
17.86% :     0.001234s :      1: type_inference.specialize
------[replace.]   0.000146     9
48.11% :     0.000070s :      6: replace.inline
26.96% :     0.000039s :      1: replace.switch_simplify
24.93% :     0.000036s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000245     9
93.80% :     0.000230s :      6: match.inline
 3.01% :     0.000007s :      1: match.switch_simplify
 3.20% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   801
 1.34% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.99% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     12: predicate.addn_check_dump
 1.49% :     0.000002s :     12: predicate.addn_zero_filter
 2.49% :     0.000003s :     12: predicate.arithmetic_simplify
 1.27% :     0.000002s :     12: predicate.cast_eliminate
 0.30% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     12: predicate.compare_switch_simplify
 1.52% :     0.000002s :     12: predicate.depend_value_elim
 1.26% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.47% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.24% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.34% :     0.000002s :     12: predicate.environ_get_eliminate
 1.22% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 4.89% :     0.000006s :     26: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.18% :     0.000002s :      6: predicate.less_batch_normalization
 1.73% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     17: predicate.load_eliminater
 1.04% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000005s :     40: predicate.loop_unroll_before_grad
 2.20% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000002s :     12: predicate.merge_addn
 1.18% :     0.000002s :     12: predicate.minmaximum_grad
 1.41% :     0.000002s :      3: predicate.mutable_eliminate
 0.38% :     0.000000s :      3: predicate.opt_reshape
 2.05% :     0.000003s :     17: predicate.partial_eliminate
 1.26% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     12: predicate.reduce_eliminate
 1.69% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.54% :     0.000002s :     20: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.68% :     0.000002s :     12: predicate.reshape_eliminate
 1.28% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.61% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.53% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.55% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.05% :     0.000004s :     20: predicate.switch_defer_inline
 2.45% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.11% :     0.000011s :     65: predicate.switch_simplify
 1.50% :     0.000002s :     12: predicate.tile_eliminate
 1.21% :     0.000002s :     12: predicate.transpose_eliminate
 1.49% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.35% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.85% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     12: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000925    15
45.38% :     0.000420s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.53% :     0.000107s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.09% :     0.000399s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017193    72
 0.20% :     0.000034s :      1: add_recomputation
 0.37% :     0.000063s :      1: auto_monad
 0.07% :     0.000012s :      1: auto_monad_reorder
 1.74% :     0.000299s :      1: bootstrap
 0.11% :     0.000018s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000005s :      1: environ_conv
 0.13% :     0.000023s :      1: event_method
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
32.98% :     0.005671s :      1: jit_opt_a
 0.78% :     0.000135s :      1: jit_opt_after_cconv
 0.30% :     0.000051s :      1: jit_opt_b
 2.48% :     0.000427s :      1: loop_unroll
 3.06% :     0.000526s :      1: mutable_eliminate
 6.51% :     0.001120s :     26: opt.transform.jit_opt_a
 0.26% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000018s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.56% :     0.000440s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.41% :     0.000071s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000010s :      1: remove_dup_value
 3.01% :     0.000518s :      1: renormalize.infer
 2.18% :     0.000376s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000029s :      1: rewriter_after_opt_a
 0.46% :     0.000079s :      1: rewriter_before_opt_a
 0.52% :     0.000090s :      1: symbol_engine_optimizer
40.59% :     0.006978s :      1: type_inference


TotalTime = 0.0177563, [33]
[bootstrap]: 0.00026357
[type_inference]: 0.0043536
[event_method]: 1.685e-05
[auto_monad]: 6.473e-05
[graph_reusing]: 5.01002e-06
[pre_auto_parallel]: 2.38002e-06
[py_interpret_to_execute]: 3.267e-05
[rewriter_before_opt_a]: 6.876e-05
[expand_dump_flag]: 2.66999e-06
[jit_opt_a]: 0.00548828, [2]
    [Cycle 1]: 0.00169859, [27]
        [switch_simplify]: 5.432e-05
        [loop_unroll]: 2.816e-05
        [a_1]: 0.00058656
        [with_stream_mark]: 1.806e-05
        [recompute_prepare]: 6.94001e-06
        [updatestate_depend_eliminate]: 3.21999e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 2.62001e-06
        [parameter_eliminate]: 1.45001e-06
        [specialize_transform]: 6.07999e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 1.84998e-06
        [get_grad_eliminate_]: 5.00999e-06
        [merge_forward]: 3.55998e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.318e-05
        [j_node_and_user_rematch]: 8.38001e-06
        [meta_fg_expand]: 2.36e-06
        [replace_old_param]: 8.25e-06
        [inline_without_move]: 5.20001e-06
        [renormalize]: 0.00069992
        [add_forward_monad_depend]: 4.85999e-06
        [auto_monad_grad]: 2.56998e-06
        [auto_monad_eliminator]: 1.314e-05
        [cse]: 2.573e-05
        [replace_applicator]: 1.238e-05
    [Cycle 2]: 0.00031115, [27]
        [switch_simplify]: 5.37001e-06
        [loop_unroll]: 4.87998e-06
        [a_1]: 9.046e-05
        [with_stream_mark]: 9.46e-06
        [recompute_prepare]: 4.82998e-06
        [updatestate_depend_eliminate]: 2.63e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 2.13002e-06
        [parameter_eliminate]: 8.2e-07
        [specialize_transform]: 4.71002e-06
        [updatestate_useless_node_eliminater]: 4.55001e-06
        [accelerated_algorithm]: 4.87e-06
        [meta_shard_fg_expand]: 1.32999e-06
        [get_grad_eliminate_]: 4.53999e-06
        [merge_forward]: 2.30002e-06
        [cell_reuse_recompute_pass]: 1.76998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.163e-05
        [j_node_and_user_rematch]: 7.94002e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 6.79999e-06
        [inline_without_move]: 4.52998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.42e-06
        [auto_monad_grad]: 1.07998e-06
        [auto_monad_eliminator]: 4.97e-06
        [cse]: 1.032e-05
        [replace_applicator]: 4.79998e-06
[py_interpret_to_execute_after_opt_a]: 1.136e-05
[rewriter_after_opt_a]: 3.428e-05
[convert_after_rewriter]: 6.38003e-06
[order_py_execute_after_rewriter]: 4.74e-06
[mutable_eliminate]: 0.00060699
[jit_opt_b]: 5.674e-05, [1]
    [Cycle 1]: 3.996e-05, [2]
        [frontend_op_eliminate]: 1.529e-05
        [inline_after_opt_a]: 1.385e-05
[cconv]: 2.462e-05
[loop_unroll]: 0.00042592
[jit_opt_after_cconv]: 0.00013327, [1]
    [Cycle 1]: 0.00012736, [11]
        [c_1]: 1.952e-05
        [parameter_eliminate]: 2.89999e-06
        [updatestate_depend_eliminate]: 5.12999e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [cse]: 1.756e-05
        [call_graph_tuple_transform]: 1.828e-05
        [tuple_list_get_item_eliminator]: 4.99e-06
        [none_parameter_eliminate]: 1.29e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 5.14e-06
[remove_dup_value]: 1.285e-05
[partial_unused_args_eliminate]: 2.63e-06
[environ_conv]: 4.66002e-06
[add_recomputation]: 4.39e-05
[cse_after_recomputation]: 1.946e-05, [1]
    [Cycle 1]: 1.416e-05, [1]
        [cse]: 8.74e-06
[auto_monad_reorder]: 1.438e-05
[get_jit_bprop_graph]: 2.17001e-06
[rewriter_after_jit_bprop_graph]: 4.75001e-06
[opt_after_jit_grad]: 0.00043692
[symbol_engine_optimizer]: 6.834e-05, [1]
    [Cycle 1]: 6.297e-05, [6]
        [build]: 3.56999e-06
        [elim_shapecalc]: 6.94999e-06
        [elim_not_effective]: 1.186e-05
        [opt_reshape]: 5.41998e-06
        [fold_const_symbol]: 8.21002e-06
        [renormalize]: 3.7998e-07
[validate]: 3.201e-05
[backend_pass]: 1.17999e-06
[task_emit]: 0.00532217
[execute]: 6.74001e-06
Sums
    bootstrap                            :     0.000264s :  1.92%
    type_inference                       :     0.004354s : 31.80%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000065s :  0.47%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.24%
    rewriter_before_opt_a                :     0.000069s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000060s :  0.44%
    jit_opt_a.loop_unroll                :     0.000033s :  0.24%
    jit_opt_a.a_1                        :     0.000677s :  4.94%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000700s :  5.11%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000036s :  0.26%
    jit_opt_a.replace_applicator         :     0.000017s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.08%
    rewriter_after_opt_a                 :     0.000034s :  0.25%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000607s :  4.43%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.18%
    loop_unroll                          :     0.000426s :  3.11%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000044s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000437s :  3.19%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005322s : 38.87%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000249    23
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 2.24% :     0.000006s :      3: substitution.graph_param_transform
87.84% :     0.000219s :      4: substitution.inline
 1.09% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.75% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.38% :     0.000003s :      2: substitution.replace_old_param
 4.46% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004292     2
81.29% :     0.003489s :      1: type_inference.infer
18.71% :     0.000803s :      1: type_inference.specialize
------[replace.]   0.000066     6
67.70% :     0.000045s :      4: replace.inline
32.30% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000227     6
95.52% :     0.000216s :      4: match.inline
 4.48% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.27% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.46% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.74% :     0.000002s :     10: predicate.addn_zero_filter
 2.18% :     0.000002s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.47% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.48% :     0.000002s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.19% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.99% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.87% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.16% :     0.000001s :     10: predicate.minmaximum_grad
 1.25% :     0.000001s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.28% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.70% :     0.000009s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.92% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.37% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.37% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.34% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.83% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000538    12
45.03% :     0.000242s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.94% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.03% :     0.000253s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019428    76
 0.24% :     0.000047s :      1: add_recomputation
 0.35% :     0.000068s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.42% :     0.000276s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
28.27% :     0.005492s :      1: jit_opt_a
 0.70% :     0.000136s :      1: jit_opt_after_cconv
 0.31% :     0.000059s :      1: jit_opt_b
 2.22% :     0.000432s :      1: loop_unroll
 3.16% :     0.000614s :      1: mutable_eliminate
 4.52% :     0.000879s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.28% :     0.000443s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000035s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 2.03% :     0.000394s :      1: renormalize.infer
 1.54% :     0.000299s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000037s :      1: rewriter_after_opt_a
 0.37% :     0.000071s :      1: rewriter_before_opt_a
 0.36% :     0.000071s :      1: symbol_engine_optimizer
27.46% :     0.005334s :      1: task_emit
22.48% :     0.004367s :      1: type_inference
 0.26% :     0.000051s :      1: validate


TotalTime = 0.0198998, [33]
[bootstrap]: 0.00028913
[type_inference]: 0.00458573
[event_method]: 1.986e-05
[auto_monad]: 6.276e-05
[graph_reusing]: 6.00002e-06
[pre_auto_parallel]: 2.51998e-06
[py_interpret_to_execute]: 4.299e-05
[rewriter_before_opt_a]: 8.154e-05
[expand_dump_flag]: 3.26999e-06
[jit_opt_a]: 0.00563917, [2]
    [Cycle 1]: 0.00174492, [27]
        [switch_simplify]: 5.74e-05
        [loop_unroll]: 2.941e-05
        [a_1]: 0.00055549
        [with_stream_mark]: 2.088e-05
        [recompute_prepare]: 7.84002e-06
        [updatestate_depend_eliminate]: 3.65e-06
        [updatestate_assign_eliminate]: 3.03998e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 7.47002e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 1.96e-06
        [get_grad_eliminate_]: 5.35999e-06
        [merge_forward]: 4.43001e-06
        [cell_reuse_recompute_pass]: 1.16002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.52e-05
        [j_node_and_user_rematch]: 9.29998e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 1.301e-05
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.00071757
        [add_forward_monad_depend]: 6.06e-06
        [auto_monad_grad]: 2.98e-06
        [auto_monad_eliminator]: 1.558e-05
        [cse]: 2.889e-05
        [replace_applicator]: 1.434e-05
    [Cycle 2]: 0.00034883, [27]
        [switch_simplify]: 6.03002e-06
        [loop_unroll]: 5.02999e-06
        [a_1]: 9.376e-05
        [with_stream_mark]: 1.012e-05
        [recompute_prepare]: 5.52001e-06
        [updatestate_depend_eliminate]: 2.93e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 1.24998e-06
        [specialize_transform]: 4.92e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 5.43002e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 4.81002e-06
        [merge_forward]: 3.43999e-06
        [cell_reuse_recompute_pass]: 2.48e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.423e-05
        [j_node_and_user_rematch]: 9.19e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 7.82e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 1.08001e-06
        [auto_monad_eliminator]: 6.45002e-06
        [cse]: 1.19e-05
        [replace_applicator]: 6.19001e-06
[py_interpret_to_execute_after_opt_a]: 1.303e-05
[rewriter_after_opt_a]: 3.931e-05
[convert_after_rewriter]: 7.18e-06
[order_py_execute_after_rewriter]: 8.12998e-06
[mutable_eliminate]: 0.00076803
[jit_opt_b]: 5.436e-05, [1]
    [Cycle 1]: 4.547e-05, [2]
        [frontend_op_eliminate]: 1.785e-05
        [inline_after_opt_a]: 1.495e-05
[cconv]: 2.871e-05
[loop_unroll]: 0.00050639
[jit_opt_after_cconv]: 0.00015511, [1]
    [Cycle 1]: 0.00014819, [11]
        [c_1]: 2.058e-05
        [parameter_eliminate]: 3.46999e-06
        [updatestate_depend_eliminate]: 8.35999e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 2.277e-05
        [call_graph_tuple_transform]: 2.091e-05
        [tuple_list_get_item_eliminator]: 5.61e-06
        [none_parameter_eliminate]: 1.65001e-06
        [renormalize]: 6.39993e-07
        [switch_simplify]: 5.82001e-06
[remove_dup_value]: 1.375e-05
[partial_unused_args_eliminate]: 2.07001e-06
[environ_conv]: 5.67001e-06
[add_recomputation]: 5.208e-05
[cse_after_recomputation]: 2.462e-05, [1]
    [Cycle 1]: 1.859e-05, [1]
        [cse]: 1.037e-05
[auto_monad_reorder]: 1.575e-05
[get_jit_bprop_graph]: 2.31e-06
[rewriter_after_jit_bprop_graph]: 7.80998e-06
[opt_after_jit_grad]: 0.00053071
[symbol_engine_optimizer]: 7.935e-05, [1]
    [Cycle 1]: 7.142e-05, [6]
        [build]: 4.45999e-06
        [elim_shapecalc]: 9.09e-06
        [elim_not_effective]: 1.486e-05
        [opt_reshape]: 5.62001e-06
        [fold_const_symbol]: 8.62e-06
        [renormalize]: 3.60014e-07
[validate]: 4.206e-05
[backend_pass]: 1.20999e-06
[task_emit]: 0.0065617
[execute]: 9.00999e-06
Sums
    bootstrap                            :     0.000289s :  1.85%
    type_inference                       :     0.004586s : 29.29%
    event_method                         :     0.000020s :  0.13%
    auto_monad                           :     0.000063s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.27%
    rewriter_before_opt_a                :     0.000082s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000649s :  4.15%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.13%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000718s :  4.58%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.14%
    jit_opt_a.cse                        :     0.000041s :  0.26%
    jit_opt_a.replace_applicator         :     0.000021s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000039s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000008s :  0.05%
    mutable_eliminate                    :     0.000768s :  4.91%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000029s :  0.18%
    loop_unroll                          :     0.000506s :  3.23%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000052s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000531s :  3.39%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000042s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006562s : 41.92%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000217    23
 1.20% :     0.000003s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.68% :     0.000006s :      3: substitution.graph_param_transform
82.97% :     0.000180s :      4: substitution.inline
 1.72% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.14% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.51% :     0.000005s :      2: substitution.replace_old_param
 6.17% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004513     2
80.73% :     0.003643s :      1: type_inference.infer
19.27% :     0.000869s :      1: type_inference.specialize
------[replace.]   0.000081     6
69.66% :     0.000056s :      4: replace.inline
30.34% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000189     6
93.55% :     0.000177s :      4: match.inline
 6.45% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.22% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 2.42% :     0.000003s :     10: predicate.arithmetic_simplify
 1.42% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.22% :     0.000001s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.16% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.17% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.94% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.51% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 6.22% :     0.000008s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000002s :      6: predicate.less_batch_normalization
 2.12% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.29% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.00% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.16% :     0.000001s :     10: predicate.minmaximum_grad
 2.10% :     0.000003s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 1.02% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.25% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.15% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.15% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.49% :     0.000009s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.26% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000593    12
50.47% :     0.000299s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.84% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.70% :     0.000259s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021602    76
 0.26% :     0.000056s :      1: add_recomputation
 0.31% :     0.000066s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.42% :     0.000307s :      1: bootstrap
 0.15% :     0.000031s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000024s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.12% :     0.005643s :      1: jit_opt_a
 0.73% :     0.000158s :      1: jit_opt_after_cconv
 0.26% :     0.000057s :      1: jit_opt_b
 2.39% :     0.000516s :      1: loop_unroll
 3.63% :     0.000783s :      1: mutable_eliminate
 4.04% :     0.000873s :     26: opt.transform.jit_opt_a
 0.23% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.52% :     0.000544s :      1: opt_after_jit_grad
 0.05% :     0.000010s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000046s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.68% :     0.000364s :      1: renormalize.infer
 1.59% :     0.000343s :      1: renormalize.specialize
 0.05% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000044s :      1: rewriter_after_opt_a
 0.40% :     0.000085s :      1: rewriter_before_opt_a
 0.38% :     0.000082s :      1: symbol_engine_optimizer
30.45% :     0.006579s :      1: task_emit
21.31% :     0.004603s :      1: type_inference
 0.30% :     0.000064s :      1: validate


TotalTime = 0.0213808, [33]
[bootstrap]: 0.00031697
[type_inference]: 0.00481358
[event_method]: 1.758e-05
[auto_monad]: 6.467e-05
[graph_reusing]: 5.67999e-06
[pre_auto_parallel]: 2.81999e-06
[py_interpret_to_execute]: 3.668e-05
[rewriter_before_opt_a]: 7.21e-05
[expand_dump_flag]: 3.51001e-06
[jit_opt_a]: 0.00604722, [2]
    [Cycle 1]: 0.00202559, [27]
        [switch_simplify]: 5.74e-05
        [loop_unroll]: 2.936e-05
        [a_1]: 0.00064046
        [with_stream_mark]: 2.674e-05
        [recompute_prepare]: 8.44998e-06
        [updatestate_depend_eliminate]: 4.55001e-06
        [updatestate_assign_eliminate]: 3.55e-06
        [updatestate_loads_eliminate]: 2.96001e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 7.10998e-06
        [updatestate_useless_node_eliminater]: 5.31998e-06
        [accelerated_algorithm]: 6.81001e-06
        [meta_shard_fg_expand]: 2.21998e-06
        [get_grad_eliminate_]: 5.33002e-06
        [merge_forward]: 4.28999e-06
        [cell_reuse_recompute_pass]: 1.65001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.508e-05
        [j_node_and_user_rematch]: 8.78001e-06
        [meta_fg_expand]: 2.51998e-06
        [replace_old_param]: 9.81998e-06
        [inline_without_move]: 6.60997e-06
        [renormalize]: 0.00091052
        [add_forward_monad_depend]: 6.91001e-06
        [auto_monad_grad]: 2.98998e-06
        [auto_monad_eliminator]: 1.49e-05
        [cse]: 2.979e-05
        [replace_applicator]: 1.503e-05
    [Cycle 2]: 0.00033074, [27]
        [switch_simplify]: 5.73997e-06
        [loop_unroll]: 4.90999e-06
        [a_1]: 9.416e-05
        [with_stream_mark]: 1.106e-05
        [recompute_prepare]: 5.32001e-06
        [updatestate_depend_eliminate]: 2.76e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 1.74998e-06
        [parameter_eliminate]: 1.18001e-06
        [specialize_transform]: 4.87998e-06
        [updatestate_useless_node_eliminater]: 4.80999e-06
        [accelerated_algorithm]: 5.49998e-06
        [meta_shard_fg_expand]: 1.02998e-06
        [get_grad_eliminate_]: 4.35999e-06
        [merge_forward]: 3.11999e-06
        [cell_reuse_recompute_pass]: 2.59001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.479e-05
        [j_node_and_user_rematch]: 8.17998e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 7.77e-06
        [inline_without_move]: 4.93001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 7.49977e-07
        [auto_monad_eliminator]: 5.49e-06
        [cse]: 1.038e-05
        [replace_applicator]: 4.90001e-06
[py_interpret_to_execute_after_opt_a]: 1.395e-05
[rewriter_after_opt_a]: 3.891e-05
[convert_after_rewriter]: 7.35e-06
[order_py_execute_after_rewriter]: 8.82999e-06
[mutable_eliminate]: 0.00081566
[jit_opt_b]: 5.549e-05, [1]
    [Cycle 1]: 4.623e-05, [2]
        [frontend_op_eliminate]: 1.865e-05
        [inline_after_opt_a]: 1.541e-05
[cconv]: 2.926e-05
[loop_unroll]: 0.00050613
[jit_opt_after_cconv]: 0.00016241, [1]
    [Cycle 1]: 0.00015516, [11]
        [c_1]: 2.141e-05
        [parameter_eliminate]: 3.84002e-06
        [updatestate_depend_eliminate]: 7.72998e-06
        [updatestate_assign_eliminate]: 2.69001e-06
        [updatestate_loads_eliminate]: 2.88e-06
        [cse]: 2.765e-05
        [call_graph_tuple_transform]: 2.241e-05
        [tuple_list_get_item_eliminator]: 5.57999e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 1.29e-06
        [switch_simplify]: 5.25999e-06
[remove_dup_value]: 1.579e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 6.04001e-06
[add_recomputation]: 5.608e-05
[cse_after_recomputation]: 2.438e-05, [1]
    [Cycle 1]: 1.779e-05, [1]
        [cse]: 1.124e-05
[auto_monad_reorder]: 1.693e-05
[get_jit_bprop_graph]: 2.12001e-06
[rewriter_after_jit_bprop_graph]: 9.49e-06
[opt_after_jit_grad]: 0.0006979
[symbol_engine_optimizer]: 9.298e-05, [1]
    [Cycle 1]: 8.329e-05, [6]
        [build]: 5.54e-06
        [elim_shapecalc]: 9.23002e-06
        [elim_not_effective]: 1.681e-05
        [opt_reshape]: 6.44999e-06
        [fold_const_symbol]: 1.022e-05
        [renormalize]: 1.52001e-06
[validate]: 5.116e-05
[backend_pass]: 9.40025e-07
[task_emit]: 0.00709369
[execute]: 1.119e-05
Sums
    bootstrap                            :     0.000317s :  1.87%
    type_inference                       :     0.004814s : 28.38%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000065s :  0.38%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.22%
    rewriter_before_opt_a                :     0.000072s :  0.43%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.37%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000735s :  4.33%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000012s :  0.07%
    jit_opt_a.renormalize                :     0.000911s :  5.37%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000040s :  0.24%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000039s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000009s :  0.05%
    mutable_eliminate                    :     0.000816s :  4.81%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000029s :  0.17%
    loop_unroll                          :     0.000506s :  2.98%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000028s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000056s :  0.33%
    cse_after_recomputation.cse          :     0.000011s :  0.07%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.06%
    opt_after_jit_grad                   :     0.000698s :  4.11%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000002s :  0.01%
    validate                             :     0.000051s :  0.30%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007094s : 41.82%
    execute                              :     0.000011s :  0.07%


Time group info:
------[substitution.]   0.000279    23
 0.73% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000002s :      2: substitution.fold_const_symbol
 2.13% :     0.000006s :      3: substitution.graph_param_transform
86.23% :     0.000241s :      4: substitution.inline
 1.29% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.81% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.82% :     0.000005s :      2: substitution.replace_old_param
 5.24% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004736     2
77.78% :     0.003683s :      1: type_inference.infer
22.22% :     0.001052s :      1: type_inference.specialize
------[replace.]   0.000079     6
66.84% :     0.000053s :      4: replace.inline
33.16% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000251     6
94.59% :     0.000238s :      4: match.inline
 5.41% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.04% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 2.15% :     0.000003s :     10: predicate.arithmetic_simplify
 1.28% :     0.000002s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.23% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.62% :     0.000001s :      3: predicate.elim_not_effective
 1.01% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.40% :     0.000002s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.07% :     0.000006s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.62% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.52% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.28% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 2.61% :     0.000003s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.95% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.55% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.59% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.73% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.84% :     0.000009s :     53: predicate.switch_simplify
 1.50% :     0.000002s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.28% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.04% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.87% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     10: predicate.value_based_eliminate
 0.55% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000658    12
43.85% :     0.000289s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.52% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.63% :     0.000320s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023367    76
 0.25% :     0.000059s :      1: add_recomputation
 0.29% :     0.000068s :      1: auto_monad
 0.08% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.44% :     0.000337s :      1: bootstrap
 0.14% :     0.000032s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000026s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
25.90% :     0.006051s :      1: jit_opt_a
 0.71% :     0.000165s :      1: jit_opt_after_cconv
 0.25% :     0.000058s :      1: jit_opt_b
 2.21% :     0.000516s :      1: loop_unroll
 3.54% :     0.000827s :      1: mutable_eliminate
 4.09% :     0.000955s :     26: opt.transform.jit_opt_a
 0.22% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000039s :      4: opt.transform.symbol_engine_opt
 3.05% :     0.000713s :      1: opt_after_jit_grad
 0.05% :     0.000011s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000039s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 2.00% :     0.000468s :      1: renormalize.infer
 1.85% :     0.000433s :      1: renormalize.specialize
 0.05% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000042s :      1: rewriter_after_opt_a
 0.50% :     0.000116s :      1: rewriter_before_opt_a
 0.41% :     0.000096s :      1: symbol_engine_optimizer
30.43% :     0.007110s :      1: task_emit
20.67% :     0.004829s :      1: type_inference
 0.34% :     0.000080s :      1: validate


TotalTime = 0.0162095, [33]
[bootstrap]: 0.00028422
[type_inference]: 0.00396137
[event_method]: 1.419e-05
[auto_monad]: 4.481e-05
[graph_reusing]: 4.17998e-06
[pre_auto_parallel]: 1.87001e-06
[py_interpret_to_execute]: 2.757e-05
[rewriter_before_opt_a]: 6.285e-05
[expand_dump_flag]: 2.81e-06
[jit_opt_a]: 0.00468635, [2]
    [Cycle 1]: 0.0013777, [27]
        [switch_simplify]: 4.41e-05
        [loop_unroll]: 2.851e-05
        [a_1]: 0.00048084
        [with_stream_mark]: 1.529e-05
        [recompute_prepare]: 7.63999e-06
        [updatestate_depend_eliminate]: 2.96001e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 5.65001e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 5.27999e-06
        [meta_shard_fg_expand]: 1.76e-06
        [get_grad_eliminate_]: 5.33002e-06
        [merge_forward]: 2.99999e-06
        [cell_reuse_recompute_pass]: 1.60999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.203e-05
        [j_node_and_user_rematch]: 8.73001e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 8.37e-06
        [inline_without_move]: 4.97e-06
        [renormalize]: 0.00051604
        [add_forward_monad_depend]: 4.52e-06
        [auto_monad_grad]: 1.40999e-06
        [auto_monad_eliminator]: 1.243e-05
        [cse]: 2.171e-05
        [replace_applicator]: 1.381e-05
    [Cycle 2]: 0.00031688, [27]
        [switch_simplify]: 5.39e-06
        [loop_unroll]: 5.04e-06
        [a_1]: 8.688e-05
        [with_stream_mark]: 1.011e-05
        [recompute_prepare]: 4.77998e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 4.95999e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 4.90001e-06
        [meta_shard_fg_expand]: 1.46998e-06
        [get_grad_eliminate_]: 4.29002e-06
        [merge_forward]: 2.96999e-06
        [cell_reuse_recompute_pass]: 2.96001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.151e-05
        [j_node_and_user_rematch]: 8.37e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 7.24001e-06
        [inline_without_move]: 4.58001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 9.90025e-07
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 5.20999e-06
        [cse]: 1.035e-05
        [replace_applicator]: 4.90999e-06
[py_interpret_to_execute_after_opt_a]: 7.41999e-06
[rewriter_after_opt_a]: 3.06e-05
[convert_after_rewriter]: 6.34001e-06
[order_py_execute_after_rewriter]: 4.28001e-06
[mutable_eliminate]: 0.00050001
[jit_opt_b]: 4.843e-05, [1]
    [Cycle 1]: 4.163e-05, [2]
        [frontend_op_eliminate]: 1.554e-05
        [inline_after_opt_a]: 1.429e-05
[cconv]: 2.12e-05
[loop_unroll]: 0.00041882
[jit_opt_after_cconv]: 0.00016087, [1]
    [Cycle 1]: 0.00015482, [11]
        [c_1]: 1.991e-05
        [parameter_eliminate]: 2.63e-06
        [updatestate_depend_eliminate]: 5.47001e-06
        [updatestate_assign_eliminate]: 2.30002e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.715e-05
        [call_graph_tuple_transform]: 1.712e-05
        [tuple_list_get_item_eliminator]: 5.20999e-06
        [none_parameter_eliminate]: 1.37999e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 4.99e-06
[remove_dup_value]: 1.027e-05
[partial_unused_args_eliminate]: 2.24999e-06
[environ_conv]: 3.93001e-06
[add_recomputation]: 3.854e-05
[cse_after_recomputation]: 1.898e-05, [1]
    [Cycle 1]: 1.385e-05, [1]
        [cse]: 8.48999e-06
[auto_monad_reorder]: 1.197e-05
[get_jit_bprop_graph]: 1.48002e-06
[rewriter_after_jit_bprop_graph]: 5.05001e-06
[opt_after_jit_grad]: 0.00044375
[symbol_engine_optimizer]: 6.807e-05, [1]
    [Cycle 1]: 6.212e-05, [6]
        [build]: 2.59001e-06
        [elim_shapecalc]: 7.12002e-06
        [elim_not_effective]: 1.146e-05
        [opt_reshape]: 5.35001e-06
        [fold_const_symbol]: 8.20999e-06
        [renormalize]: 6.80011e-07
[validate]: 3.055e-05
[backend_pass]: 1.49998e-06
[task_emit]: 0.00511525
[execute]: 5.00999e-06
Sums
    bootstrap                            :     0.000284s :  2.25%
    type_inference                       :     0.003961s : 31.35%
    event_method                         :     0.000014s :  0.11%
    auto_monad                           :     0.000045s :  0.35%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.22%
    rewriter_before_opt_a                :     0.000063s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000049s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.27%
    jit_opt_a.a_1                        :     0.000568s :  4.49%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000516s :  4.09%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.14%
    jit_opt_a.cse                        :     0.000032s :  0.25%
    jit_opt_a.replace_applicator         :     0.000019s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.24%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000500s :  3.96%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000021s :  0.17%
    loop_unroll                          :     0.000419s :  3.32%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000010s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000039s :  0.31%
    cse_after_recomputation.cse          :     0.000008s :  0.07%
    auto_monad_reorder                   :     0.000012s :  0.09%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000444s :  3.51%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000031s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005115s : 40.49%
    execute                              :     0.000005s :  0.04%


Time group info:
------[substitution.]   0.000173    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.78% :     0.000001s :      2: substitution.fold_const_symbol
 2.71% :     0.000005s :      3: substitution.graph_param_transform
82.81% :     0.000143s :      4: substitution.inline
 2.08% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.21% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.18% :     0.000004s :      2: substitution.replace_old_param
 6.21% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003911     2
82.63% :     0.003232s :      1: type_inference.infer
17.37% :     0.000679s :      1: type_inference.specialize
------[replace.]   0.000058     6
67.02% :     0.000039s :      4: replace.inline
32.98% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000151     6
93.59% :     0.000141s :      4: match.inline
 6.41% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000108   691
 1.36% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000001s :     10: predicate.addn_zero_filter
 1.72% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     10: predicate.compare_switch_simplify
 1.48% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.35% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.73% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.60% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.71% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.02% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.36% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.47% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000002s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.68% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.00% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000001s :     10: predicate.reshape_eliminate
 1.30% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      6: predicate.special_op_eliminate
 1.19% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.57% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.63% :     0.000008s :     53: predicate.switch_simplify
 1.24% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.62% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.14% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.97% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000478    12
50.02% :     0.000239s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.29% :     0.000021s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.69% :     0.000218s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017594    76
 0.24% :     0.000041s :      1: add_recomputation
 0.27% :     0.000048s :      1: auto_monad
 0.08% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.68% :     0.000296s :      1: bootstrap
 0.14% :     0.000024s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.10% :     0.000017s :      1: event_method
 0.04% :     0.000008s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
26.65% :     0.004689s :      1: jit_opt_a
 0.93% :     0.000164s :      1: jit_opt_after_cconv
 0.29% :     0.000051s :      1: jit_opt_b
 2.43% :     0.000427s :      1: loop_unroll
 2.89% :     0.000509s :      1: mutable_eliminate
 4.34% :     0.000764s :     26: opt.transform.jit_opt_a
 0.25% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.57% :     0.000452s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000030s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000013s :      1: remove_dup_value
 1.55% :     0.000273s :      1: renormalize.infer
 1.34% :     0.000235s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000034s :      1: rewriter_after_opt_a
 0.37% :     0.000066s :      1: rewriter_before_opt_a
 0.40% :     0.000070s :      1: symbol_engine_optimizer
29.12% :     0.005124s :      1: task_emit
22.57% :     0.003970s :      1: type_inference
 0.27% :     0.000047s :      1: validate


TotalTime = 0.0157255, [30]
[bootstrap]: 0.0002722
[type_inference]: 0.00673183
[event_method]: 2.224e-05
[auto_monad]: 0.00013139
[graph_reusing]: 5.77001e-06
[pre_auto_parallel]: 2.23998e-06
[py_interpret_to_execute]: 7.499e-05
[rewriter_before_opt_a]: 8.911e-05
[expand_dump_flag]: 2.72001e-06
[jit_opt_a]: 0.00613057, [2]
    [Cycle 1]: 0.00227017, [27]
        [switch_simplify]: 0.00012455
        [loop_unroll]: 3.414e-05
        [a_1]: 0.00077809
        [with_stream_mark]: 1.657e-05
        [recompute_prepare]: 6.92002e-06
        [updatestate_depend_eliminate]: 2.88e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 1.62001e-06
        [specialize_transform]: 6.44001e-06
        [updatestate_useless_node_eliminater]: 5.66e-06
        [accelerated_algorithm]: 6.12001e-06
        [meta_shard_fg_expand]: 1.78002e-06
        [get_grad_eliminate_]: 5.66998e-06
        [merge_forward]: 3.58e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.265e-05
        [j_node_and_user_rematch]: 1.048e-05
        [meta_fg_expand]: 1.94e-06
        [replace_old_param]: 9.49e-06
        [inline_without_move]: 6.17999e-06
        [renormalize]: 0.00101338
        [add_forward_monad_depend]: 5.54e-06
        [auto_monad_grad]: 2.91e-06
        [auto_monad_eliminator]: 1.165e-05
        [cse]: 1.496e-05
        [replace_applicator]: 1.202e-05
    [Cycle 2]: 0.00032523, [27]
        [switch_simplify]: 5.64998e-06
        [loop_unroll]: 5.24e-06
        [a_1]: 9.247e-05
        [with_stream_mark]: 9.27999e-06
        [recompute_prepare]: 4.91002e-06
        [updatestate_depend_eliminate]: 2.48e-06
        [updatestate_assign_eliminate]: 2.48002e-06
        [updatestate_loads_eliminate]: 1.91998e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 4.75999e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 5.44e-06
        [meta_shard_fg_expand]: 1.32999e-06
        [get_grad_eliminate_]: 4.43001e-06
        [merge_forward]: 2.58e-06
        [cell_reuse_recompute_pass]: 2.10002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.249e-05
        [j_node_and_user_rematch]: 7.58001e-06
        [meta_fg_expand]: 1.52999e-06
        [replace_old_param]: 7.91001e-06
        [inline_without_move]: 4.60999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.40001e-06
        [auto_monad_grad]: 1.02998e-06
        [auto_monad_eliminator]: 4.85001e-06
        [cse]: 1.003e-05
        [replace_applicator]: 4.75999e-06
[py_interpret_to_execute_after_opt_a]: 1.22e-05
[rewriter_after_opt_a]: 3.812e-05
[convert_after_rewriter]: 1.031e-05
[order_py_execute_after_rewriter]: 5.96e-06
[mutable_eliminate]: 0.00061265
[jit_opt_b]: 5.121e-05, [1]
    [Cycle 1]: 4.422e-05, [2]
        [frontend_op_eliminate]: 1.794e-05
        [inline_after_opt_a]: 1.469e-05
[cconv]: 2.623e-05
[loop_unroll]: 0.00042852
[jit_opt_after_cconv]: 0.00013972, [1]
    [Cycle 1]: 0.00013339, [11]
        [c_1]: 2.161e-05
        [parameter_eliminate]: 2.88e-06
        [updatestate_depend_eliminate]: 5.09998e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.749e-05
        [call_graph_tuple_transform]: 1.935e-05
        [tuple_list_get_item_eliminator]: 5.52001e-06
        [none_parameter_eliminate]: 1.39998e-06
        [renormalize]: 9.89996e-07
        [switch_simplify]: 5.37999e-06
[remove_dup_value]: 1.302e-05
[partial_unused_args_eliminate]: 2.02001e-06
[environ_conv]: 4.90001e-06
[add_recomputation]: 9.329e-05
[cse_after_recomputation]: 2.218e-05, [1]
    [Cycle 1]: 1.626e-05, [1]
        [cse]: 9.85002e-06
[auto_monad_reorder]: 1.514e-05
[get_jit_bprop_graph]: 1.99999e-06
[rewriter_after_jit_bprop_graph]: 5.87001e-06
[opt_after_jit_grad]: 0.00047144
[symbol_engine_optimizer]: 7.73e-05, [1]
    [Cycle 1]: 6.976e-05, [6]
        [build]: 4.27998e-06
        [elim_shapecalc]: 8.15e-06
        [elim_not_effective]: 1.389e-05
        [opt_reshape]: 5.76003e-06
        [fold_const_symbol]: 9.71e-06
        [renormalize]: 4.39992e-07
[validate]: 3.744e-05
Sums
    bootstrap                            :     0.000272s :  2.35%
    type_inference                       :     0.006732s : 58.09%
    event_method                         :     0.000022s :  0.19%
    auto_monad                           :     0.000131s :  1.13%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000075s :  0.65%
    rewriter_before_opt_a                :     0.000089s :  0.77%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000130s :  1.12%
    jit_opt_a.loop_unroll                :     0.000039s :  0.34%
    jit_opt_a.a_1                        :     0.000871s :  7.51%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.001013s :  8.74%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.14%
    jit_opt_a.cse                        :     0.000025s :  0.22%
    jit_opt_a.replace_applicator         :     0.000017s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.11%
    rewriter_after_opt_a                 :     0.000038s :  0.33%
    convert_after_rewriter               :     0.000010s :  0.09%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000613s :  5.29%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.13%
    cconv                                :     0.000026s :  0.23%
    loop_unroll                          :     0.000429s :  3.70%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000093s :  0.80%
    cse_after_recomputation.cse          :     0.000010s :  0.08%
    auto_monad_reorder                   :     0.000015s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000471s :  4.07%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.32%


Time group info:
------[substitution.]   0.000336    26
 0.55% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000002s :      2: substitution.fold_const_symbol
 1.77% :     0.000006s :      3: substitution.graph_param_transform
86.79% :     0.000291s :      6: substitution.inline
 0.82% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.42% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.27% :     0.000004s :      2: substitution.replace_old_param
 2.65% :     0.000009s :      1: substitution.switch_simplify
 4.12% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006666     2
81.09% :     0.005405s :      1: type_inference.infer
18.91% :     0.001261s :      1: type_inference.specialize
------[replace.]   0.000154     9
50.31% :     0.000078s :      6: replace.inline
25.06% :     0.000039s :      1: replace.switch_simplify
24.63% :     0.000038s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000309     9
93.24% :     0.000288s :      6: match.inline
 2.64% :     0.000008s :      1: match.switch_simplify
 4.12% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   801
 1.40% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000002s :     12: predicate.addn_check_dump
 1.38% :     0.000002s :     12: predicate.addn_zero_filter
 1.94% :     0.000003s :     12: predicate.arithmetic_simplify
 1.35% :     0.000002s :     12: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.33% :     0.000002s :     12: predicate.compare_switch_simplify
 1.47% :     0.000002s :     12: predicate.depend_value_elim
 1.14% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.68% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.25% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     12: predicate.environ_get_eliminate
 1.27% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.40% :     0.000001s :      3: predicate.graph_param_transform
 5.16% :     0.000007s :     26: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      6: predicate.less_batch_normalization
 1.65% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.90% :     0.000002s :     17: predicate.load_eliminater
 1.18% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.30% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.07% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     12: predicate.merge_addn
 1.16% :     0.000001s :     12: predicate.minmaximum_grad
 1.16% :     0.000001s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     17: predicate.partial_eliminate
 1.28% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     12: predicate.reduce_eliminate
 1.53% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     20: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     12: predicate.reshape_eliminate
 1.20% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.50% :     0.000002s :     12: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.75% :     0.000004s :     20: predicate.switch_defer_inline
 2.32% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.12% :     0.000011s :     65: predicate.switch_simplify
 1.22% :     0.000002s :     12: predicate.tile_eliminate
 1.22% :     0.000002s :     12: predicate.transpose_eliminate
 1.50% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.56% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     12: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000967    15
47.08% :     0.000455s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.69% :     0.000132s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.23% :     0.000379s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017950    72
 0.54% :     0.000097s :      1: add_recomputation
 0.75% :     0.000135s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 1.73% :     0.000311s :      1: bootstrap
 0.16% :     0.000029s :      1: cconv
 0.07% :     0.000013s :      1: convert_after_rewriter
 0.14% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.14% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
34.18% :     0.006135s :      1: jit_opt_a
 0.80% :     0.000143s :      1: jit_opt_after_cconv
 0.30% :     0.000054s :      1: jit_opt_b
 2.42% :     0.000435s :      1: loop_unroll
 3.45% :     0.000620s :      1: mutable_eliminate
 6.45% :     0.001158s :     26: opt.transform.jit_opt_a
 0.27% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.67% :     0.000480s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.43% :     0.000078s :      1: py_interpret_to_execute
 0.08% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000015s :      1: remove_dup_value
 3.52% :     0.000631s :      1: renormalize.infer
 2.09% :     0.000375s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000041s :      1: rewriter_after_opt_a
 0.51% :     0.000092s :      1: rewriter_before_opt_a
 0.45% :     0.000080s :      1: symbol_engine_optimizer
37.57% :     0.006744s :      1: type_inference


TotalTime = 0.0193782, [33]
[bootstrap]: 0.00029034
[type_inference]: 0.00456436
[event_method]: 1.831e-05
[auto_monad]: 6.987e-05
[graph_reusing]: 6.00002e-06
[pre_auto_parallel]: 2.09999e-06
[py_interpret_to_execute]: 3.314e-05
[rewriter_before_opt_a]: 6.991e-05
[expand_dump_flag]: 3.73001e-06
[jit_opt_a]: 0.00557658, [2]
    [Cycle 1]: 0.00170032, [27]
        [switch_simplify]: 4.943e-05
        [loop_unroll]: 2.957e-05
        [a_1]: 0.00053779
        [with_stream_mark]: 2.074e-05
        [recompute_prepare]: 7.01001e-06
        [updatestate_depend_eliminate]: 3.33e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [parameter_eliminate]: 1.86998e-06
        [specialize_transform]: 5.52001e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 2.58e-06
        [get_grad_eliminate_]: 5.35999e-06
        [merge_forward]: 3.56001e-06
        [cell_reuse_recompute_pass]: 1.05999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.387e-05
        [j_node_and_user_rematch]: 9.41e-06
        [meta_fg_expand]: 2.06998e-06
        [replace_old_param]: 8.75999e-06
        [inline_without_move]: 5.05999e-06
        [renormalize]: 0.00074281
        [add_forward_monad_depend]: 5.56998e-06
        [auto_monad_grad]: 2.88e-06
        [auto_monad_eliminator]: 1.42e-05
        [cse]: 2.733e-05
        [replace_applicator]: 1.263e-05
    [Cycle 2]: 0.00034082, [27]
        [switch_simplify]: 5.91e-06
        [loop_unroll]: 5.07e-06
        [a_1]: 9.095e-05
        [with_stream_mark]: 1.082e-05
        [recompute_prepare]: 5.29e-06
        [updatestate_depend_eliminate]: 2.78998e-06
        [updatestate_assign_eliminate]: 2.05002e-06
        [updatestate_loads_eliminate]: 1.88997e-06
        [parameter_eliminate]: 1.44e-06
        [specialize_transform]: 5.19e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 5.05999e-06
        [meta_shard_fg_expand]: 1.31002e-06
        [get_grad_eliminate_]: 4.71002e-06
        [merge_forward]: 2.66999e-06
        [cell_reuse_recompute_pass]: 2.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.38e-05
        [j_node_and_user_rematch]: 7.83001e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 7.50998e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.42e-06
        [auto_monad_grad]: 1.05999e-06
        [auto_monad_eliminator]: 5.48002e-06
        [cse]: 1.294e-05
        [replace_applicator]: 4.92e-06
[py_interpret_to_execute_after_opt_a]: 1.223e-05
[rewriter_after_opt_a]: 3.577e-05
[convert_after_rewriter]: 7.3e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00065283
[jit_opt_b]: 5.045e-05, [1]
    [Cycle 1]: 4.303e-05, [2]
        [frontend_op_eliminate]: 1.665e-05
        [inline_after_opt_a]: 1.442e-05
[cconv]: 2.924e-05
[loop_unroll]: 0.00044688
[jit_opt_after_cconv]: 0.00014906, [1]
    [Cycle 1]: 0.00014242, [11]
        [c_1]: 2.123e-05
        [parameter_eliminate]: 3.81999e-06
        [updatestate_depend_eliminate]: 6.38003e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.03002e-06
        [cse]: 2.291e-05
        [call_graph_tuple_transform]: 2.012e-05
        [tuple_list_get_item_eliminator]: 5.49e-06
        [none_parameter_eliminate]: 1.67999e-06
        [renormalize]: 5.59987e-07
        [switch_simplify]: 5.16998e-06
[remove_dup_value]: 1.396e-05
[partial_unused_args_eliminate]: 2.19999e-06
[environ_conv]: 6.63998e-06
[add_recomputation]: 5.067e-05
[cse_after_recomputation]: 2.199e-05, [1]
    [Cycle 1]: 1.629e-05, [1]
        [cse]: 9.69e-06
[auto_monad_reorder]: 1.577e-05
[get_jit_bprop_graph]: 1.96998e-06
[rewriter_after_jit_bprop_graph]: 5.60001e-06
[opt_after_jit_grad]: 0.00049841
[symbol_engine_optimizer]: 7.51e-05, [1]
    [Cycle 1]: 6.815e-05, [6]
        [build]: 4.45999e-06
        [elim_shapecalc]: 8.2e-06
        [elim_not_effective]: 1.337e-05
        [opt_reshape]: 5.39998e-06
        [fold_const_symbol]: 8.94998e-06
        [renormalize]: 7.10017e-07
[validate]: 4.383e-05
[backend_pass]: 8.79983e-07
[task_emit]: 0.00640661
[execute]: 6.14001e-06
Sums
    bootstrap                            :     0.000290s :  1.91%
    type_inference                       :     0.004564s : 30.00%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000070s :  0.46%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000033s :  0.22%
    rewriter_before_opt_a                :     0.000070s :  0.46%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.36%
    jit_opt_a.loop_unroll                :     0.000035s :  0.23%
    jit_opt_a.a_1                        :     0.000629s :  4.13%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000743s :  4.88%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000040s :  0.26%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000036s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000653s :  4.29%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000029s :  0.19%
    loop_unroll                          :     0.000447s :  2.94%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000051s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000498s :  3.28%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.29%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006407s : 42.11%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000216    23
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.66% :     0.000006s :      3: substitution.graph_param_transform
84.26% :     0.000182s :      4: substitution.inline
 1.71% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.09% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.15% :     0.000005s :      2: substitution.replace_old_param
 5.51% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004499     2
82.66% :     0.003719s :      1: type_inference.infer
17.34% :     0.000780s :      1: type_inference.specialize
------[replace.]   0.000068     6
67.72% :     0.000046s :      4: replace.inline
32.28% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000190     6
94.30% :     0.000179s :      4: match.inline
 5.70% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000001s :     10: predicate.addn_zero_filter
 2.32% :     0.000003s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.51% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.94% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000000s :      3: predicate.graph_param_transform
 5.39% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000001s :      6: predicate.less_batch_normalization
 1.76% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.64% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 2.03% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.35% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.79% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.64% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.43% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.79% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.77% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.39% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000646    12
53.04% :     0.000343s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.09% :     0.000046s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.87% :     0.000258s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021062    76
 0.25% :     0.000054s :      1: add_recomputation
 0.35% :     0.000073s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 1.45% :     0.000305s :      1: bootstrap
 0.15% :     0.000032s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.50% :     0.005581s :      1: jit_opt_a
 0.72% :     0.000152s :      1: jit_opt_after_cconv
 0.25% :     0.000053s :      1: jit_opt_b
 2.16% :     0.000456s :      1: loop_unroll
 3.15% :     0.000663s :      1: mutable_eliminate
 3.96% :     0.000835s :     26: opt.transform.jit_opt_a
 0.23% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.41% :     0.000508s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000036s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.85% :     0.000390s :      1: renormalize.infer
 1.64% :     0.000345s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000039s :      1: rewriter_after_opt_a
 0.35% :     0.000073s :      1: rewriter_before_opt_a
 0.37% :     0.000078s :      1: symbol_engine_optimizer
30.48% :     0.006421s :      1: task_emit
21.73% :     0.004578s :      1: type_inference
 0.31% :     0.000066s :      1: validate


TotalTime = 0.0223669, [33]
[bootstrap]: 0.00032923
[type_inference]: 0.00515838
[event_method]: 1.926e-05
[auto_monad]: 0.00012163
[graph_reusing]: 6.62002e-06
[pre_auto_parallel]: 3.21001e-06
[py_interpret_to_execute]: 4.591e-05
[rewriter_before_opt_a]: 8.001e-05
[expand_dump_flag]: 3.91001e-06
[jit_opt_a]: 0.0066379, [2]
    [Cycle 1]: 0.00208353, [27]
        [switch_simplify]: 7.361e-05
        [loop_unroll]: 2.911e-05
        [a_1]: 0.00065374
        [with_stream_mark]: 2.468e-05
        [recompute_prepare]: 8.79e-06
        [updatestate_depend_eliminate]: 3.86999e-06
        [updatestate_assign_eliminate]: 3.21001e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [parameter_eliminate]: 2.66e-06
        [specialize_transform]: 5.80002e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.52999e-06
        [meta_shard_fg_expand]: 2.44001e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 4.14002e-06
        [cell_reuse_recompute_pass]: 1.42e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.586e-05
        [j_node_and_user_rematch]: 9.90002e-06
        [meta_fg_expand]: 2.48002e-06
        [replace_old_param]: 9.94999e-06
        [inline_without_move]: 5.61e-06
        [renormalize]: 0.00094408
        [add_forward_monad_depend]: 7.18998e-06
        [auto_monad_grad]: 2.49001e-06
        [auto_monad_eliminator]: 1.618e-05
        [cse]: 2.993e-05
        [replace_applicator]: 1.545e-05
    [Cycle 2]: 0.0003457, [27]
        [switch_simplify]: 5.54998e-06
        [loop_unroll]: 4.99e-06
        [a_1]: 0.00010545
        [with_stream_mark]: 1.227e-05
        [recompute_prepare]: 5.32999e-06
        [updatestate_depend_eliminate]: 2.98e-06
        [updatestate_assign_eliminate]: 2.13002e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 4.77998e-06
        [updatestate_useless_node_eliminater]: 4.49998e-06
        [accelerated_algorithm]: 5.29e-06
        [meta_shard_fg_expand]: 1.72999e-06
        [get_grad_eliminate_]: 4.48001e-06
        [merge_forward]: 3.06001e-06
        [cell_reuse_recompute_pass]: 2.68e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.433e-05
        [j_node_and_user_rematch]: 8.23001e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 7.05e-06
        [inline_without_move]: 4.66002e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.47001e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 5.56e-06
        [cse]: 1.105e-05
        [replace_applicator]: 4.85999e-06
[py_interpret_to_execute_after_opt_a]: 1.314e-05
[rewriter_after_opt_a]: 3.847e-05
[convert_after_rewriter]: 7.97003e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00079619
[jit_opt_b]: 5.206e-05, [1]
    [Cycle 1]: 4.438e-05, [2]
        [frontend_op_eliminate]: 1.711e-05
        [inline_after_opt_a]: 1.568e-05
[cconv]: 3.113e-05
[loop_unroll]: 0.0010405
[jit_opt_after_cconv]: 0.00017189, [1]
    [Cycle 1]: 0.00016368, [11]
        [c_1]: 2.285e-05
        [parameter_eliminate]: 5.19998e-06
        [updatestate_depend_eliminate]: 8.89e-06
        [updatestate_assign_eliminate]: 2.89999e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [cse]: 2.951e-05
        [call_graph_tuple_transform]: 2.593e-05
        [tuple_list_get_item_eliminator]: 5.47001e-06
        [none_parameter_eliminate]: 2.01e-06
        [renormalize]: 8.70001e-07
        [switch_simplify]: 5.55001e-06
[remove_dup_value]: 1.443e-05
[partial_unused_args_eliminate]: 2.85998e-06
[environ_conv]: 6.49001e-06
[add_recomputation]: 5.573e-05
[cse_after_recomputation]: 2.164e-05, [1]
    [Cycle 1]: 1.594e-05, [1]
        [cse]: 1.036e-05
[auto_monad_reorder]: 1.624e-05
[get_jit_bprop_graph]: 2.29001e-06
[rewriter_after_jit_bprop_graph]: 8.74e-06
[opt_after_jit_grad]: 0.00051294
[symbol_engine_optimizer]: 8.145e-05, [1]
    [Cycle 1]: 7.312e-05, [6]
        [build]: 4.61002e-06
        [elim_shapecalc]: 7.91001e-06
        [elim_not_effective]: 1.42e-05
        [opt_reshape]: 6.01998e-06
        [fold_const_symbol]: 1.025e-05
        [renormalize]: 4.39992e-07
[validate]: 4.748e-05
[backend_pass]: 1.29e-06
[task_emit]: 0.00678834
[execute]: 8.77e-06
Sums
    bootstrap                            :     0.000329s :  1.88%
    type_inference                       :     0.005158s : 29.50%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000122s :  0.70%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000046s :  0.26%
    rewriter_before_opt_a                :     0.000080s :  0.46%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000079s :  0.45%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000759s :  4.34%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000944s :  5.40%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.12%
    jit_opt_a.cse                        :     0.000041s :  0.23%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000796s :  4.55%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000031s :  0.18%
    loop_unroll                          :     0.001041s :  5.95%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000030s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000056s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000513s :  2.93%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000047s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006788s : 38.83%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000285    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.79% :     0.000002s :      2: substitution.fold_const_symbol
 2.21% :     0.000006s :      3: substitution.graph_param_transform
86.53% :     0.000247s :      4: substitution.inline
 1.21% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.81% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.68% :     0.000005s :      2: substitution.replace_old_param
 5.02% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005076     2
80.80% :     0.004102s :      1: type_inference.infer
19.20% :     0.000975s :      1: type_inference.specialize
------[replace.]   0.000084     6
65.27% :     0.000055s :      4: replace.inline
34.73% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000257     6
94.87% :     0.000244s :      4: match.inline
 5.13% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   691
 1.48% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.39% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.82% :     0.000002s :     10: predicate.addn_zero_filter
 2.20% :     0.000003s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      3: predicate.elim_not_effective
 0.44% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.76% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 6.12% :     0.000008s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000002s :      6: predicate.less_batch_normalization
 1.25% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.77% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.48% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.34% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 1.70% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 1.78% :     0.000002s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.40% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.43% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 2.30% :     0.000003s :     10: predicate.row_tensor_add_zeros_like
 1.12% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.78% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.76% :     0.000001s :      6: predicate.special_op_eliminate
 0.86% :     0.000001s :      6: predicate.specialize_transform
 1.99% :     0.000003s :     10: predicate.split_environ_get_set_with_tuple_value
 2.23% :     0.000003s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.98% :     0.000004s :     16: predicate.switch_defer_inline
 1.80% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.14% :     0.000010s :     53: predicate.switch_simplify
 1.20% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.23% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.08% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.82% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.81% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.43% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.99% :     0.000003s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000708    12
45.68% :     0.000323s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.16% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.16% :     0.000327s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024412    76
 0.24% :     0.000059s :      1: add_recomputation
 0.52% :     0.000126s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.41% :     0.000344s :      1: bootstrap
 0.14% :     0.000034s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
27.21% :     0.006642s :      1: jit_opt_a
 0.72% :     0.000175s :      1: jit_opt_after_cconv
 0.23% :     0.000055s :      1: jit_opt_b
 4.32% :     0.001055s :      1: loop_unroll
 3.30% :     0.000807s :      1: mutable_eliminate
 4.05% :     0.000990s :     26: opt.transform.jit_opt_a
 0.23% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.14% :     0.000522s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000049s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.12% :     0.000518s :      1: renormalize.infer
 1.71% :     0.000417s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000041s :      1: rewriter_after_opt_a
 0.34% :     0.000084s :      1: rewriter_before_opt_a
 0.34% :     0.000084s :      1: symbol_engine_optimizer
27.89% :     0.006808s :      1: task_emit
21.20% :     0.005176s :      1: type_inference
 0.29% :     0.000070s :      1: validate


TotalTime = 0.0177474, [33]
[bootstrap]: 0.00028738
[type_inference]: 0.00472068
[event_method]: 1.725e-05
[auto_monad]: 6.291e-05
[graph_reusing]: 5.61998e-06
[pre_auto_parallel]: 2.43e-06
[py_interpret_to_execute]: 2.711e-05
[rewriter_before_opt_a]: 7.023e-05
[expand_dump_flag]: 3.04001e-06
[jit_opt_a]: 0.00483286, [2]
    [Cycle 1]: 0.00144297, [27]
        [switch_simplify]: 4.907e-05
        [loop_unroll]: 2.852e-05
        [a_1]: 0.00049651
        [with_stream_mark]: 1.576e-05
        [recompute_prepare]: 7.30998e-06
        [updatestate_depend_eliminate]: 3.21001e-06
        [updatestate_assign_eliminate]: 4.00998e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 5.57001e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 5.19e-06
        [meta_shard_fg_expand]: 2.19001e-06
        [get_grad_eliminate_]: 4.85001e-06
        [merge_forward]: 3.68e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.34e-05
        [j_node_and_user_rematch]: 8.50999e-06
        [meta_fg_expand]: 2.12999e-06
        [replace_old_param]: 8.88002e-06
        [inline_without_move]: 4.92e-06
        [renormalize]: 0.00054408
        [add_forward_monad_depend]: 4.99998e-06
        [auto_monad_grad]: 1.72001e-06
        [auto_monad_eliminator]: 1.427e-05
        [cse]: 2.853e-05
        [replace_applicator]: 1.316e-05
    [Cycle 2]: 0.0003211, [27]
        [switch_simplify]: 5.59e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 8.726e-05
        [with_stream_mark]: 9.64999e-06
        [recompute_prepare]: 5.59998e-06
        [updatestate_depend_eliminate]: 2.55002e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.57001e-06
        [specialize_transform]: 4.82998e-06
        [updatestate_useless_node_eliminater]: 4.72998e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 1.40001e-06
        [get_grad_eliminate_]: 4.27e-06
        [merge_forward]: 2.56e-06
        [cell_reuse_recompute_pass]: 2.19999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.212e-05
        [j_node_and_user_rematch]: 7.61999e-06
        [meta_fg_expand]: 1.76998e-06
        [replace_old_param]: 7.31001e-06
        [inline_without_move]: 4.48001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.28002e-06
        [auto_monad_grad]: 1.19e-06
        [auto_monad_eliminator]: 5.74999e-06
        [cse]: 1.099e-05
        [replace_applicator]: 4.90001e-06
[py_interpret_to_execute_after_opt_a]: 1.033e-05
[rewriter_after_opt_a]: 3.258e-05
[convert_after_rewriter]: 7.11001e-06
[order_py_execute_after_rewriter]: 7.4e-06
[mutable_eliminate]: 0.00053139
[jit_opt_b]: 4.849e-05, [1]
    [Cycle 1]: 4.122e-05, [2]
        [frontend_op_eliminate]: 1.493e-05
        [inline_after_opt_a]: 1.459e-05
[cconv]: 4.145e-05
[loop_unroll]: 0.00043343
[jit_opt_after_cconv]: 0.00013736, [1]
    [Cycle 1]: 0.00013109, [11]
        [c_1]: 1.928e-05
        [parameter_eliminate]: 2.89999e-06
        [updatestate_depend_eliminate]: 6.48998e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [cse]: 1.985e-05
        [call_graph_tuple_transform]: 1.758e-05
        [tuple_list_get_item_eliminator]: 4.85001e-06
        [none_parameter_eliminate]: 2.21e-06
        [renormalize]: 7.09988e-07
        [switch_simplify]: 4.84e-06
[remove_dup_value]: 1.424e-05
[partial_unused_args_eliminate]: 2.22001e-06
[environ_conv]: 5.47001e-06
[add_recomputation]: 4.716e-05
[cse_after_recomputation]: 1.977e-05, [1]
    [Cycle 1]: 1.452e-05, [1]
        [cse]: 8.94e-06
[auto_monad_reorder]: 1.395e-05
[get_jit_bprop_graph]: 2.07999e-06
[rewriter_after_jit_bprop_graph]: 5.35999e-06
[opt_after_jit_grad]: 0.00043491
[symbol_engine_optimizer]: 6.928e-05, [1]
    [Cycle 1]: 6.286e-05, [6]
        [build]: 3.13998e-06
        [elim_shapecalc]: 7.35998e-06
        [elim_not_effective]: 1.196e-05
        [opt_reshape]: 5.15999e-06
        [fold_const_symbol]: 8.14002e-06
        [renormalize]: 5.79981e-07
[validate]: 3.729e-05
[backend_pass]: 7.50006e-07
[task_emit]: 0.00562202
[execute]: 7.07002e-06
Sums
    bootstrap                            :     0.000287s :  2.04%
    type_inference                       :     0.004721s : 33.49%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000063s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000027s :  0.19%
    rewriter_before_opt_a                :     0.000070s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.39%
    jit_opt_a.loop_unroll                :     0.000033s :  0.24%
    jit_opt_a.a_1                        :     0.000584s :  4.14%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000544s :  3.86%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.14%
    jit_opt_a.cse                        :     0.000040s :  0.28%
    jit_opt_a.replace_applicator         :     0.000018s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000007s :  0.05%
    mutable_eliminate                    :     0.000531s :  3.77%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000041s :  0.29%
    loop_unroll                          :     0.000433s :  3.07%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000435s :  3.09%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005622s : 39.88%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000182    23
 1.21% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.62% :     0.000005s :      3: substitution.graph_param_transform
82.88% :     0.000150s :      4: substitution.inline
 1.54% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.25% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.33% :     0.000004s :      2: substitution.replace_old_param
 6.47% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004657     2
83.52% :     0.003890s :      1: type_inference.infer
16.48% :     0.000767s :      1: type_inference.specialize
------[replace.]   0.000063     6
67.34% :     0.000043s :      4: replace.inline
32.66% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000159     6
93.28% :     0.000148s :      4: match.inline
 6.72% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.28% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000001s :     10: predicate.addn_zero_filter
 1.91% :     0.000002s :     10: predicate.arithmetic_simplify
 1.34% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000001s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.72% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.02% :     0.000001s :      6: predicate.less_batch_normalization
 1.65% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.70% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.11% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.40% :     0.000002s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 2.03% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000000s :      3: predicate.opt_reshape
 2.42% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.37% :     0.000002s :     10: predicate.reduce_eliminate
 1.63% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.76% :     0.000001s :      3: predicate.reset_defer_inline
 1.54% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.69% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.73% :     0.000009s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.49% :     0.000002s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.88% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000581    12
54.70% :     0.000318s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.84% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.46% :     0.000235s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019179    76
 0.26% :     0.000050s :      1: add_recomputation
 0.35% :     0.000066s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.58% :     0.000303s :      1: bootstrap
 0.23% :     0.000045s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.21% :     0.004836s :      1: jit_opt_a
 0.73% :     0.000140s :      1: jit_opt_after_cconv
 0.27% :     0.000051s :      1: jit_opt_b
 2.30% :     0.000441s :      1: loop_unroll
 2.82% :     0.000541s :      1: mutable_eliminate
 4.08% :     0.000783s :     26: opt.transform.jit_opt_a
 0.23% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.31% :     0.000443s :      1: opt_after_jit_grad
 0.05% :     0.000010s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000030s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000017s :      1: remove_dup_value
 1.49% :     0.000286s :      1: renormalize.infer
 1.31% :     0.000251s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000036s :      1: rewriter_after_opt_a
 0.38% :     0.000073s :      1: rewriter_before_opt_a
 0.37% :     0.000072s :      1: symbol_engine_optimizer
29.36% :     0.005631s :      1: task_emit
24.69% :     0.004736s :      1: type_inference
 0.30% :     0.000057s :      1: validate


TotalTime = 0.0151511, [30]
[bootstrap]: 0.00026203
[type_inference]: 0.00650191
[event_method]: 2.021e-05
[auto_monad]: 7.918e-05
[graph_reusing]: 6.39001e-06
[pre_auto_parallel]: 2.71999e-06
[py_interpret_to_execute]: 5.191e-05
[rewriter_before_opt_a]: 8.542e-05
[expand_dump_flag]: 3.21999e-06
[jit_opt_a]: 0.0059743, [2]
    [Cycle 1]: 0.00233267, [27]
        [switch_simplify]: 0.00012676
        [loop_unroll]: 8.01e-05
        [a_1]: 0.00074794
        [with_stream_mark]: 1.83e-05
        [recompute_prepare]: 7.25998e-06
        [updatestate_depend_eliminate]: 3.58e-06
        [updatestate_assign_eliminate]: 2.89999e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [parameter_eliminate]: 2.07999e-06
        [specialize_transform]: 5.99999e-06
        [updatestate_useless_node_eliminater]: 5.01002e-06
        [accelerated_algorithm]: 5.87001e-06
        [meta_shard_fg_expand]: 2.42001e-06
        [get_grad_eliminate_]: 5.46e-06
        [merge_forward]: 3.85e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.52e-05
        [j_node_and_user_rematch]: 9.89001e-06
        [meta_fg_expand]: 2.40002e-06
        [replace_old_param]: 1.041e-05
        [inline_without_move]: 5.90002e-06
        [renormalize]: 0.00102276
        [add_forward_monad_depend]: 5.05999e-06
        [auto_monad_grad]: 2.06e-06
        [auto_monad_eliminator]: 1.299e-05
        [cse]: 2.922e-05
        [replace_applicator]: 1.449e-05
    [Cycle 2]: 0.0003189, [27]
        [switch_simplify]: 5.78997e-06
        [loop_unroll]: 5.02e-06
        [a_1]: 9.422e-05
        [with_stream_mark]: 9.51e-06
        [recompute_prepare]: 5.06002e-06
        [updatestate_depend_eliminate]: 2.73e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 4.81002e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.00999e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 4.43001e-06
        [merge_forward]: 3.04999e-06
        [cell_reuse_recompute_pass]: 1.53002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.21e-05
        [j_node_and_user_rematch]: 7.53999e-06
        [meta_fg_expand]: 2.03997e-06
        [replace_old_param]: 7.21001e-06
        [inline_without_move]: 4.95001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.04998e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 5.39e-06
        [cse]: 9.74999e-06
        [replace_applicator]: 4.62e-06
[py_interpret_to_execute_after_opt_a]: 1.098e-05
[rewriter_after_opt_a]: 3.396e-05
[convert_after_rewriter]: 7.10998e-06
[order_py_execute_after_rewriter]: 5.24e-06
[mutable_eliminate]: 0.00061396
[jit_opt_b]: 5.108e-05, [1]
    [Cycle 1]: 4.373e-05, [2]
        [frontend_op_eliminate]: 1.788e-05
        [inline_after_opt_a]: 1.428e-05
[cconv]: 2.4e-05
[loop_unroll]: 0.000412
[jit_opt_after_cconv]: 0.00013669, [1]
    [Cycle 1]: 0.00012982, [11]
        [c_1]: 2.141e-05
        [parameter_eliminate]: 2.02999e-06
        [updatestate_depend_eliminate]: 5.40999e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [cse]: 1.504e-05
        [call_graph_tuple_transform]: 1.831e-05
        [tuple_list_get_item_eliminator]: 5.13002e-06
        [none_parameter_eliminate]: 1.45001e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 5.46002e-06
[remove_dup_value]: 1.406e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 4.97e-06
[add_recomputation]: 4.839e-05
[cse_after_recomputation]: 2.201e-05, [1]
    [Cycle 1]: 1.626e-05, [1]
        [cse]: 1.001e-05
[auto_monad_reorder]: 1.476e-05
[get_jit_bprop_graph]: 1.59e-06
[rewriter_after_jit_bprop_graph]: 5.74999e-06
[opt_after_jit_grad]: 0.00047921
[symbol_engine_optimizer]: 7.149e-05, [1]
    [Cycle 1]: 6.532e-05, [6]
        [build]: 3.26001e-06
        [elim_shapecalc]: 8.02e-06
        [elim_not_effective]: 1.203e-05
        [opt_reshape]: 5.64e-06
        [fold_const_symbol]: 8.46997e-06
        [renormalize]: 4.2998e-07
[validate]: 3.385e-05
Sums
    bootstrap                            :     0.000262s :  2.33%
    type_inference                       :     0.006502s : 57.82%
    event_method                         :     0.000020s :  0.18%
    auto_monad                           :     0.000079s :  0.70%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000052s :  0.46%
    rewriter_before_opt_a                :     0.000085s :  0.76%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000133s :  1.18%
    jit_opt_a.loop_unroll                :     0.000085s :  0.76%
    jit_opt_a.a_1                        :     0.000842s :  7.49%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.16%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.001023s :  9.10%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.16%
    jit_opt_a.cse                        :     0.000039s :  0.35%
    jit_opt_a.replace_applicator         :     0.000019s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.10%
    rewriter_after_opt_a                 :     0.000034s :  0.30%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000614s :  5.46%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.13%
    cconv                                :     0.000024s :  0.21%
    loop_unroll                          :     0.000412s :  3.66%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000048s :  0.43%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000479s :  4.26%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.30%


Time group info:
------[substitution.]   0.000307    26
 0.59% :     0.000002s :      2: substitution.elim_not_effective
 0.39% :     0.000001s :      2: substitution.fold_const_symbol
 1.88% :     0.000006s :      3: substitution.graph_param_transform
86.17% :     0.000265s :      6: substitution.inline
 0.88% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.68% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.30% :     0.000004s :      2: substitution.replace_old_param
 2.74% :     0.000008s :      1: substitution.switch_simplify
 4.37% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006445     2
80.55% :     0.005192s :      1: type_inference.infer
19.45% :     0.001253s :      1: type_inference.specialize
------[replace.]   0.000155     9
52.21% :     0.000081s :      6: replace.inline
25.12% :     0.000039s :      1: replace.switch_simplify
22.67% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000281     9
92.93% :     0.000261s :      6: match.inline
 2.70% :     0.000008s :      1: match.switch_simplify
 4.38% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   801
 1.37% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.26% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     12: predicate.addn_check_dump
 1.35% :     0.000002s :     12: predicate.addn_zero_filter
 2.17% :     0.000003s :     12: predicate.arithmetic_simplify
 1.49% :     0.000002s :     12: predicate.cast_eliminate
 0.56% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000002s :     12: predicate.compare_switch_simplify
 1.43% :     0.000002s :     12: predicate.depend_value_elim
 1.31% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.63% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.74% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.36% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     12: predicate.environ_get_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.26% :     0.000007s :     26: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.18% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     17: predicate.load_eliminater
 1.24% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.31% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.10% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000002s :     12: predicate.merge_addn
 1.17% :     0.000002s :     12: predicate.minmaximum_grad
 1.08% :     0.000001s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     17: predicate.partial_eliminate
 1.19% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     12: predicate.reduce_eliminate
 1.40% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     20: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     12: predicate.reshape_eliminate
 1.26% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.77% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.71% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.83% :     0.000004s :     20: predicate.switch_defer_inline
 2.47% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.21% :     0.000011s :     65: predicate.switch_simplify
 1.28% :     0.000002s :     12: predicate.tile_eliminate
 1.49% :     0.000002s :     12: predicate.transpose_eliminate
 1.49% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.46% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.58% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000002s :     12: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000926    15
41.42% :     0.000383s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.75% :     0.000109s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.83% :     0.000434s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017395    72
 0.30% :     0.000052s :      1: add_recomputation
 0.48% :     0.000083s :      1: auto_monad
 0.10% :     0.000017s :      1: auto_monad_reorder
 1.57% :     0.000273s :      1: bootstrap
 0.15% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.14% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
34.37% :     0.005979s :      1: jit_opt_a
 0.80% :     0.000139s :      1: jit_opt_after_cconv
 0.31% :     0.000054s :      1: jit_opt_b
 2.41% :     0.000419s :      1: loop_unroll
 3.58% :     0.000622s :      1: mutable_eliminate
 6.76% :     0.001176s :     26: opt.transform.jit_opt_a
 0.27% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.81% :     0.000488s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000055s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 3.22% :     0.000561s :      1: renormalize.infer
 2.61% :     0.000454s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000037s :      1: rewriter_after_opt_a
 0.51% :     0.000088s :      1: rewriter_before_opt_a
 0.43% :     0.000074s :      1: symbol_engine_optimizer
37.44% :     0.006513s :      1: type_inference


TotalTime = 0.0198463, [33]
[bootstrap]: 0.00034973
[type_inference]: 0.00479749
[event_method]: 1.792e-05
[auto_monad]: 6.749e-05
[graph_reusing]: 6.06e-06
[pre_auto_parallel]: 2.37999e-06
[py_interpret_to_execute]: 4.094e-05
[rewriter_before_opt_a]: 7.818e-05
[expand_dump_flag]: 3.26001e-06
[jit_opt_a]: 0.00554875, [2]
    [Cycle 1]: 0.00169063, [27]
        [switch_simplify]: 5.71e-05
        [loop_unroll]: 2.975e-05
        [a_1]: 0.00052291
        [with_stream_mark]: 2.082e-05
        [recompute_prepare]: 7.45e-06
        [updatestate_depend_eliminate]: 3.51001e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.69999e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 5.66998e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.57001e-06
        [meta_shard_fg_expand]: 2.11e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 3.65998e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.34e-05
        [j_node_and_user_rematch]: 8.60001e-06
        [meta_fg_expand]: 2.11e-06
        [replace_old_param]: 5.63e-05
        [inline_without_move]: 6.42001e-06
        [renormalize]: 0.00069315
        [add_forward_monad_depend]: 5.39998e-06
        [auto_monad_grad]: 2.71e-06
        [auto_monad_eliminator]: 1.435e-05
        [cse]: 2.977e-05
        [replace_applicator]: 1.21e-05
    [Cycle 2]: 0.00032282, [27]
        [switch_simplify]: 5.97001e-06
        [loop_unroll]: 5.25001e-06
        [a_1]: 9.096e-05
        [with_stream_mark]: 1.096e-05
        [recompute_prepare]: 5.08002e-06
        [updatestate_depend_eliminate]: 2.54001e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 4.94e-06
        [updatestate_useless_node_eliminater]: 5.34998e-06
        [accelerated_algorithm]: 4.95999e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.16001e-06
        [merge_forward]: 2.64999e-06
        [cell_reuse_recompute_pass]: 1.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.257e-05
        [j_node_and_user_rematch]: 7.65998e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 1.252e-05
        [inline_without_move]: 4.74e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 1.24e-06
        [auto_monad_eliminator]: 5.20001e-06
        [cse]: 1.037e-05
        [replace_applicator]: 4.89e-06
[py_interpret_to_execute_after_opt_a]: 1.288e-05
[rewriter_after_opt_a]: 3.546e-05
[convert_after_rewriter]: 7.56999e-06
[order_py_execute_after_rewriter]: 5.44998e-06
[mutable_eliminate]: 0.00063133
[jit_opt_b]: 5.321e-05, [1]
    [Cycle 1]: 4.575e-05, [2]
        [frontend_op_eliminate]: 1.778e-05
        [inline_after_opt_a]: 1.549e-05
[cconv]: 2.635e-05
[loop_unroll]: 0.00047402
[jit_opt_after_cconv]: 0.00015215, [1]
    [Cycle 1]: 0.00014466, [11]
        [c_1]: 2.09e-05
        [parameter_eliminate]: 3.78001e-06
        [updatestate_depend_eliminate]: 6.43e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [cse]: 2.284e-05
        [call_graph_tuple_transform]: 2.116e-05
        [tuple_list_get_item_eliminator]: 6.09001e-06
        [none_parameter_eliminate]: 1.59998e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 5.44e-06
[remove_dup_value]: 1.339e-05
[partial_unused_args_eliminate]: 2.39001e-06
[environ_conv]: 5.27999e-06
[add_recomputation]: 4.977e-05
[cse_after_recomputation]: 2.189e-05, [1]
    [Cycle 1]: 1.567e-05, [1]
        [cse]: 9.22001e-06
[auto_monad_reorder]: 1.492e-05
[get_jit_bprop_graph]: 2.21e-06
[rewriter_after_jit_bprop_graph]: 6.52001e-06
[opt_after_jit_grad]: 0.00051281
[symbol_engine_optimizer]: 7.764e-05, [1]
    [Cycle 1]: 6.967e-05, [6]
        [build]: 4.46002e-06
        [elim_shapecalc]: 8.24998e-06
        [elim_not_effective]: 1.385e-05
        [opt_reshape]: 6.22001e-06
        [fold_const_symbol]: 9.12999e-06
        [renormalize]: 4.80009e-07
[validate]: 4.058e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.00655306
[execute]: 8.39998e-06
Sums
    bootstrap                            :     0.000350s :  2.23%
    type_inference                       :     0.004797s : 30.60%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000067s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000041s :  0.26%
    rewriter_before_opt_a                :     0.000078s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000614s :  3.91%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000069s :  0.44%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000693s :  4.42%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000040s :  0.26%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000035s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000631s :  4.03%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000026s :  0.17%
    loop_unroll                          :     0.000474s :  3.02%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000513s :  3.27%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000041s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006553s : 41.79%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000199    23
 1.10% :     0.000002s :      2: substitution.elim_not_effective
 0.84% :     0.000002s :      2: substitution.fold_const_symbol
 2.81% :     0.000006s :      3: substitution.graph_param_transform
83.61% :     0.000166s :      4: substitution.inline
 1.47% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.20% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.10% :     0.000004s :      2: substitution.replace_old_param
 5.87% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004724     2
82.55% :     0.003900s :      1: type_inference.infer
17.45% :     0.000825s :      1: type_inference.specialize
------[replace.]   0.000069     6
68.34% :     0.000047s :      4: replace.inline
31.66% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000174     6
93.93% :     0.000164s :      4: match.inline
 6.07% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.62% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000001s :     10: predicate.addn_zero_filter
 2.20% :     0.000003s :     10: predicate.arithmetic_simplify
 1.47% :     0.000002s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.17% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.73% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.87% :     0.000007s :     22: predicate.inline
 1.14% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.99% :     0.000001s :      6: predicate.less_batch_normalization
 1.62% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.51% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.92% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 2.03% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.87% :     0.000002s :     10: predicate.reduce_eliminate
 1.34% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.87% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.40% :     0.000002s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.32% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.83% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.51% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.22% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000563    12
49.39% :     0.000278s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.30% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.31% :     0.000255s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021524    76
 0.25% :     0.000053s :      1: add_recomputation
 0.33% :     0.000072s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.73% :     0.000372s :      1: bootstrap
 0.13% :     0.000029s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.80% :     0.005552s :      1: jit_opt_a
 0.72% :     0.000155s :      1: jit_opt_after_cconv
 0.26% :     0.000056s :      1: jit_opt_b
 2.24% :     0.000483s :      1: loop_unroll
 2.98% :     0.000642s :      1: mutable_eliminate
 4.07% :     0.000877s :     26: opt.transform.jit_opt_a
 0.23% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.42% :     0.000521s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.21% :     0.000044s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.68% :     0.000361s :      1: renormalize.infer
 1.51% :     0.000325s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000038s :      1: rewriter_after_opt_a
 0.38% :     0.000081s :      1: rewriter_before_opt_a
 0.37% :     0.000080s :      1: symbol_engine_optimizer
30.54% :     0.006574s :      1: task_emit
22.36% :     0.004814s :      1: type_inference
 0.29% :     0.000062s :      1: validate


TotalTime = 0.0164968, [33]
[bootstrap]: 0.00027656
[type_inference]: 0.00434943
[event_method]: 1.684e-05
[auto_monad]: 6.147e-05
[graph_reusing]: 4.97999e-06
[pre_auto_parallel]: 2.29999e-06
[py_interpret_to_execute]: 2.595e-05
[rewriter_before_opt_a]: 6.325e-05
[expand_dump_flag]: 3.06001e-06
[jit_opt_a]: 0.0046922, [2]
    [Cycle 1]: 0.00135563, [27]
        [switch_simplify]: 4.52e-05
        [loop_unroll]: 2.848e-05
        [a_1]: 0.0004909
        [with_stream_mark]: 1.554e-05
        [recompute_prepare]: 6.71e-06
        [updatestate_depend_eliminate]: 2.65002e-06
        [updatestate_assign_eliminate]: 2.89001e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [parameter_eliminate]: 1.38002e-06
        [specialize_transform]: 5.52001e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 5.00999e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 4.92999e-06
        [merge_forward]: 3.33e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.253e-05
        [j_node_and_user_rematch]: 8.90999e-06
        [meta_fg_expand]: 2.01998e-06
        [replace_old_param]: 8.35999e-06
        [inline_without_move]: 4.84e-06
        [renormalize]: 0.00049463
        [add_forward_monad_depend]: 3.88999e-06
        [auto_monad_grad]: 2.12001e-06
        [auto_monad_eliminator]: 1.085e-05
        [cse]: 1.954e-05
        [replace_applicator]: 1.091e-05
    [Cycle 2]: 0.00036483, [27]
        [switch_simplify]: 5.25999e-06
        [loop_unroll]: 4.58999e-06
        [a_1]: 8.748e-05
        [with_stream_mark]: 8.37998e-06
        [recompute_prepare]: 4.92999e-06
        [updatestate_depend_eliminate]: 2.37999e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 9.09989e-07
        [specialize_transform]: 4.79998e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 4.67e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 6.108e-05
        [merge_forward]: 2.95998e-06
        [cell_reuse_recompute_pass]: 1.89e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.178e-05
        [j_node_and_user_rematch]: 7.7e-06
        [meta_fg_expand]: 1.48002e-06
        [replace_old_param]: 6.74999e-06
        [inline_without_move]: 4.55001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 8.79983e-07
        [auto_monad_eliminator]: 5.10001e-06
        [cse]: 1.073e-05
        [replace_applicator]: 4.97e-06
[py_interpret_to_execute_after_opt_a]: 8.92e-06
[rewriter_after_opt_a]: 2.987e-05
[convert_after_rewriter]: 6.14999e-06
[order_py_execute_after_rewriter]: 4.54002e-06
[mutable_eliminate]: 0.00047935
[jit_opt_b]: 4.551e-05, [1]
    [Cycle 1]: 3.953e-05, [2]
        [frontend_op_eliminate]: 1.511e-05
        [inline_after_opt_a]: 1.409e-05
[cconv]: 1.956e-05
[loop_unroll]: 0.00041036
[jit_opt_after_cconv]: 0.00012706, [1]
    [Cycle 1]: 0.00012126, [11]
        [c_1]: 1.971e-05
        [parameter_eliminate]: 1.97001e-06
        [updatestate_depend_eliminate]: 4.35999e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 1.54e-05
        [call_graph_tuple_transform]: 1.672e-05
        [tuple_list_get_item_eliminator]: 4.90999e-06
        [none_parameter_eliminate]: 1.12e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 5.04998e-06
[remove_dup_value]: 1.143e-05
[partial_unused_args_eliminate]: 2.15002e-06
[environ_conv]: 4.95999e-06
[add_recomputation]: 4.174e-05
[cse_after_recomputation]: 1.931e-05, [1]
    [Cycle 1]: 1.428e-05, [1]
        [cse]: 8.80999e-06
[auto_monad_reorder]: 1.286e-05
[get_jit_bprop_graph]: 1.79e-06
[rewriter_after_jit_bprop_graph]: 4.13999e-06
[opt_after_jit_grad]: 0.00044049
[symbol_engine_optimizer]: 6.691e-05, [1]
    [Cycle 1]: 6.149e-05, [6]
        [build]: 2.65002e-06
        [elim_shapecalc]: 6.83e-06
        [elim_not_effective]: 1.114e-05
        [opt_reshape]: 5.47999e-06
        [fold_const_symbol]: 8.2e-06
        [renormalize]: 4.59986e-07
[validate]: 2.764e-05
[backend_pass]: 6.79982e-07
[task_emit]: 0.00506596
[execute]: 5.22999e-06
Sums
    bootstrap                            :     0.000277s :  2.13%
    type_inference                       :     0.004349s : 33.50%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000061s :  0.47%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.20%
    rewriter_before_opt_a                :     0.000063s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000050s :  0.39%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000578s :  4.45%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000066s :  0.51%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000495s :  3.81%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.12%
    jit_opt_a.cse                        :     0.000030s :  0.23%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000030s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000479s :  3.69%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000020s :  0.15%
    loop_unroll                          :     0.000410s :  3.16%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000011s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000042s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000013s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000440s :  3.39%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005066s : 39.02%
    execute                              :     0.000005s :  0.04%


Time group info:
------[substitution.]   0.000173    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.99% :     0.000005s :      3: substitution.graph_param_transform
83.14% :     0.000144s :      4: substitution.inline
 1.41% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.42% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.82% :     0.000003s :      2: substitution.replace_old_param
 6.50% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004295     2
82.91% :     0.003561s :      1: type_inference.infer
17.09% :     0.000734s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.85% :     0.000042s :      4: replace.inline
32.15% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000152     6
93.26% :     0.000141s :      4: match.inline
 6.74% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 1.87% :     0.000002s :     10: predicate.arithmetic_simplify
 1.63% :     0.000002s :     10: predicate.cast_eliminate
 0.71% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.73% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.30% :     0.000001s :     10: predicate.environ_get_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.47% :     0.000001s :      3: predicate.graph_param_transform
 5.65% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      6: predicate.less_batch_normalization
 1.67% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.88% :     0.000002s :     15: predicate.load_eliminater
 1.21% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.39% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000002s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.57% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.84% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.64% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000008s :     53: predicate.switch_simplify
 1.56% :     0.000002s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.32% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.80% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.08% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000499    12
50.86% :     0.000254s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.09% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.04% :     0.000220s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017916    76
 0.25% :     0.000045s :      1: add_recomputation
 0.36% :     0.000065s :      1: auto_monad
 0.09% :     0.000015s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.64% :     0.000293s :      1: bootstrap
 0.12% :     0.000022s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.05% :     0.000008s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.21% :     0.004695s :      1: jit_opt_a
 0.72% :     0.000130s :      1: jit_opt_after_cconv
 0.27% :     0.000048s :      1: jit_opt_b
 2.33% :     0.000417s :      1: loop_unroll
 2.72% :     0.000487s :      1: mutable_eliminate
 4.61% :     0.000826s :     26: opt.transform.jit_opt_a
 0.24% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.50% :     0.000447s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000028s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 1.45% :     0.000260s :      1: renormalize.infer
 1.28% :     0.000229s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000033s :      1: rewriter_after_opt_a
 0.37% :     0.000066s :      1: rewriter_before_opt_a
 0.39% :     0.000070s :      1: symbol_engine_optimizer
28.32% :     0.005074s :      1: task_emit
24.32% :     0.004357s :      1: type_inference
 0.25% :     0.000044s :      1: validate


TotalTime = 0.0228481, [33]
[bootstrap]: 0.00039713
[type_inference]: 0.00563784
[event_method]: 1.934e-05
[auto_monad]: 6.786e-05
[graph_reusing]: 5.81e-06
[pre_auto_parallel]: 3.07002e-06
[py_interpret_to_execute]: 4.127e-05
[rewriter_before_opt_a]: 7.697e-05
[expand_dump_flag]: 3.13e-06
[jit_opt_a]: 0.00685902, [2]
    [Cycle 1]: 0.00223207, [27]
        [switch_simplify]: 6.965e-05
        [loop_unroll]: 2.997e-05
        [a_1]: 0.00069253
        [with_stream_mark]: 2.302e-05
        [recompute_prepare]: 1.14e-05
        [updatestate_depend_eliminate]: 4.99e-06
        [updatestate_assign_eliminate]: 3.43e-06
        [updatestate_loads_eliminate]: 2.61999e-06
        [parameter_eliminate]: 2.30002e-06
        [specialize_transform]: 6.35002e-06
        [updatestate_useless_node_eliminater]: 5.58002e-06
        [accelerated_algorithm]: 6.82002e-06
        [meta_shard_fg_expand]: 2.54001e-06
        [get_grad_eliminate_]: 6.21998e-06
        [merge_forward]: 4.52e-06
        [cell_reuse_recompute_pass]: 1.94e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.608e-05
        [j_node_and_user_rematch]: 1.038e-05
        [meta_fg_expand]: 3.2e-06
        [replace_old_param]: 1.188e-05
        [inline_without_move]: 6.16e-06
        [renormalize]: 0.00100295
        [add_forward_monad_depend]: 8.65999e-06
        [auto_monad_grad]: 3.14001e-06
        [auto_monad_eliminator]: 1.838e-05
        [cse]: 3.353e-05
        [replace_applicator]: 2.313e-05
    [Cycle 2]: 0.00036181, [27]
        [switch_simplify]: 7.06999e-06
        [loop_unroll]: 5.10999e-06
        [a_1]: 9.923e-05
        [with_stream_mark]: 1.353e-05
        [recompute_prepare]: 5.22e-06
        [updatestate_depend_eliminate]: 3.28998e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [parameter_eliminate]: 1.71e-06
        [specialize_transform]: 5.07e-06
        [updatestate_useless_node_eliminater]: 5.10999e-06
        [accelerated_algorithm]: 6.36e-06
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 4.68999e-06
        [merge_forward]: 4.63999e-06
        [cell_reuse_recompute_pass]: 2.29999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.675e-05
        [j_node_and_user_rematch]: 8.92e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 8.69e-06
        [inline_without_move]: 5.06997e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.49e-06
        [auto_monad_grad]: 1.32e-06
        [auto_monad_eliminator]: 6.89001e-06
        [cse]: 1.366e-05
        [replace_applicator]: 5.15001e-06
[py_interpret_to_execute_after_opt_a]: 1.758e-05
[rewriter_after_opt_a]: 3.968e-05
[convert_after_rewriter]: 7.09001e-06
[order_py_execute_after_rewriter]: 4.91002e-06
[mutable_eliminate]: 0.00088582
[jit_opt_b]: 5.761e-05, [1]
    [Cycle 1]: 4.831e-05, [2]
        [frontend_op_eliminate]: 1.793e-05
        [inline_after_opt_a]: 1.688e-05
[cconv]: 3.208e-05
[loop_unroll]: 0.00054328
[jit_opt_after_cconv]: 0.00017152, [1]
    [Cycle 1]: 0.00016301, [11]
        [c_1]: 2.3e-05
        [parameter_eliminate]: 4.53001e-06
        [updatestate_depend_eliminate]: 8.37e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 3.01999e-06
        [cse]: 3.312e-05
        [call_graph_tuple_transform]: 2.282e-05
        [tuple_list_get_item_eliminator]: 5.25001e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 1.45999e-06
        [switch_simplify]: 5.97001e-06
[remove_dup_value]: 1.691e-05
[partial_unused_args_eliminate]: 2.34999e-06
[environ_conv]: 5.99e-06
[add_recomputation]: 5.849e-05
[cse_after_recomputation]: 2.288e-05, [1]
    [Cycle 1]: 1.604e-05, [1]
        [cse]: 1.012e-05
[auto_monad_reorder]: 1.74e-05
[get_jit_bprop_graph]: 2.14e-06
[rewriter_after_jit_bprop_graph]: 1.009e-05
[opt_after_jit_grad]: 0.00056044
[symbol_engine_optimizer]: 8.043e-05, [1]
    [Cycle 1]: 7.277e-05, [6]
        [build]: 5.17e-06
        [elim_shapecalc]: 8.50001e-06
        [elim_not_effective]: 1.314e-05
        [opt_reshape]: 6.14999e-06
        [fold_const_symbol]: 1.002e-05
        [renormalize]: 1.00999e-06
[validate]: 4.16e-05
[backend_pass]: 8.79983e-07
[task_emit]: 0.00690752
[execute]: 9.32001e-06
Sums
    bootstrap                            :     0.000397s :  2.22%
    type_inference                       :     0.005638s : 31.55%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000068s :  0.38%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000041s :  0.23%
    rewriter_before_opt_a                :     0.000077s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000077s :  0.43%
    jit_opt_a.loop_unroll                :     0.000035s :  0.20%
    jit_opt_a.a_1                        :     0.000792s :  4.43%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000009s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000021s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001003s :  5.61%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.14%
    jit_opt_a.cse                        :     0.000047s :  0.26%
    jit_opt_a.replace_applicator         :     0.000028s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.10%
    rewriter_after_opt_a                 :     0.000040s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000886s :  4.96%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000032s :  0.18%
    loop_unroll                          :     0.000543s :  3.04%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000033s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000058s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.06%
    opt_after_jit_grad                   :     0.000560s :  3.14%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000042s :  0.23%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006908s : 38.66%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000263    23
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.84% :     0.000002s :      2: substitution.fold_const_symbol
 2.39% :     0.000006s :      3: substitution.graph_param_transform
83.48% :     0.000220s :      4: substitution.inline
 1.80% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.17% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.70% :     0.000007s :      2: substitution.replace_old_param
 5.85% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005557     2
81.28% :     0.004517s :      1: type_inference.infer
18.72% :     0.001040s :      1: type_inference.specialize
------[replace.]   0.000114     6
72.06% :     0.000082s :      4: replace.inline
27.94% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000232     6
93.79% :     0.000217s :      4: match.inline
 6.21% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.48% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.71% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000002s :     10: predicate.addn_zero_filter
 2.66% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.15% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.46% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000001s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.71% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     10: predicate.environ_get_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.15% :     0.000002s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.34% :     0.000007s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.77% :     0.000002s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.43% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.44% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.94% :     0.000001s :     10: predicate.merge_addn
 0.92% :     0.000001s :     10: predicate.minmaximum_grad
 1.66% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.01% :     0.000003s :     15: predicate.partial_eliminate
 1.47% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.89% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.67% :     0.000002s :     10: predicate.reshape_eliminate
 1.06% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.05% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     10: predicate.same_eliminate
 0.96% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.03% :     0.000004s :     16: predicate.switch_defer_inline
 1.82% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.80% :     0.000009s :     53: predicate.switch_simplify
 1.41% :     0.000002s :     10: predicate.tile_eliminate
 1.24% :     0.000002s :     10: predicate.transpose_eliminate
 1.21% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.24% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 2.39% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.15% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000738    12
45.77% :     0.000338s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.90% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.33% :     0.000364s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025006    76
 0.25% :     0.000062s :      1: add_recomputation
 0.28% :     0.000071s :      1: auto_monad
 0.08% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.67% :     0.000417s :      1: bootstrap
 0.14% :     0.000035s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000025s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
27.45% :     0.006864s :      1: jit_opt_a
 0.70% :     0.000174s :      1: jit_opt_after_cconv
 0.24% :     0.000060s :      1: jit_opt_b
 2.21% :     0.000553s :      1: loop_unroll
 3.60% :     0.000900s :      1: mutable_eliminate
 4.17% :     0.001043s :     26: opt.transform.jit_opt_a
 0.21% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.27% :     0.000568s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000045s :      1: py_interpret_to_execute
 0.08% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000019s :      1: remove_dup_value
 2.06% :     0.000516s :      1: renormalize.infer
 1.90% :     0.000476s :      1: renormalize.specialize
 0.05% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000042s :      1: rewriter_after_opt_a
 0.32% :     0.000080s :      1: rewriter_before_opt_a
 0.33% :     0.000083s :      1: symbol_engine_optimizer
27.69% :     0.006925s :      1: task_emit
22.62% :     0.005657s :      1: type_inference
 0.25% :     0.000064s :      1: validate


TotalTime = 0.0161169, [33]
[bootstrap]: 0.00024913
[type_inference]: 0.00399281
[event_method]: 1.591e-05
[auto_monad]: 6.113e-05
[graph_reusing]: 4.68001e-06
[pre_auto_parallel]: 2.01e-06
[py_interpret_to_execute]: 2.86e-05
[rewriter_before_opt_a]: 6.608e-05
[expand_dump_flag]: 2.71e-06
[jit_opt_a]: 0.00453562, [2]
    [Cycle 1]: 0.00132692, [27]
        [switch_simplify]: 4.604e-05
        [loop_unroll]: 2.841e-05
        [a_1]: 0.00047707
        [with_stream_mark]: 1.445e-05
        [recompute_prepare]: 7.26999e-06
        [updatestate_depend_eliminate]: 3.95e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 1.49e-06
        [specialize_transform]: 5.71e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 5.08002e-06
        [meta_shard_fg_expand]: 1.84998e-06
        [get_grad_eliminate_]: 4.72e-06
        [merge_forward]: 3.31999e-06
        [cell_reuse_recompute_pass]: 9.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.317e-05
        [j_node_and_user_rematch]: 8.35999e-06
        [meta_fg_expand]: 2.05002e-06
        [replace_old_param]: 7.6e-06
        [inline_without_move]: 5.05999e-06
        [renormalize]: 0.00047686
        [add_forward_monad_depend]: 4.92999e-06
        [auto_monad_grad]: 1.89e-06
        [auto_monad_eliminator]: 1.208e-05
        [cse]: 2.029e-05
        [replace_applicator]: 1.085e-05
    [Cycle 2]: 0.00030553, [27]
        [switch_simplify]: 5.29e-06
        [loop_unroll]: 4.73001e-06
        [a_1]: 8.668e-05
        [with_stream_mark]: 7.88999e-06
        [recompute_prepare]: 4.96002e-06
        [updatestate_depend_eliminate]: 2.34001e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 4.79e-06
        [updatestate_useless_node_eliminater]: 4.62e-06
        [accelerated_algorithm]: 4.92e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.23001e-06
        [merge_forward]: 2.16e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.163e-05
        [j_node_and_user_rematch]: 7.56001e-06
        [meta_fg_expand]: 1.40001e-06
        [replace_old_param]: 6.95998e-06
        [inline_without_move]: 4.55001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 6.29982e-07
        [auto_monad_eliminator]: 5.08002e-06
        [cse]: 9.45001e-06
        [replace_applicator]: 5.15001e-06
[py_interpret_to_execute_after_opt_a]: 7.52002e-06
[rewriter_after_opt_a]: 2.874e-05
[convert_after_rewriter]: 6.08002e-06
[order_py_execute_after_rewriter]: 5.32999e-06
[mutable_eliminate]: 0.00048973
[jit_opt_b]: 4.817e-05, [1]
    [Cycle 1]: 4.182e-05, [2]
        [frontend_op_eliminate]: 1.562e-05
        [inline_after_opt_a]: 1.496e-05
[cconv]: 2.257e-05
[loop_unroll]: 0.00040084
[jit_opt_after_cconv]: 0.00014062, [1]
    [Cycle 1]: 0.0001342, [11]
        [c_1]: 2.135e-05
        [parameter_eliminate]: 2.47001e-06
        [updatestate_depend_eliminate]: 5.37001e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.38002e-06
        [cse]: 1.619e-05
        [call_graph_tuple_transform]: 1.889e-05
        [tuple_list_get_item_eliminator]: 5.79e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 4.7998e-07
        [switch_simplify]: 5.16998e-06
[remove_dup_value]: 1.131e-05
[partial_unused_args_eliminate]: 1.74e-06
[environ_conv]: 4.34997e-06
[add_recomputation]: 4.334e-05
[cse_after_recomputation]: 2.089e-05, [1]
    [Cycle 1]: 1.509e-05, [1]
        [cse]: 9.00001e-06
[auto_monad_reorder]: 1.282e-05
[get_jit_bprop_graph]: 1.43002e-06
[rewriter_after_jit_bprop_graph]: 4.99e-06
[opt_after_jit_grad]: 0.00043308
[symbol_engine_optimizer]: 6.944e-05, [1]
    [Cycle 1]: 6.288e-05, [6]
        [build]: 2.83e-06
        [elim_shapecalc]: 7.12002e-06
        [elim_not_effective]: 1.247e-05
        [opt_reshape]: 5.31998e-06
        [fold_const_symbol]: 8.02998e-06
        [renormalize]: 4.69998e-07
[validate]: 2.626e-05
[backend_pass]: 8.79983e-07
[task_emit]: 0.00520519
[execute]: 7.2e-06
Sums
    bootstrap                            :     0.000249s :  1.97%
    type_inference                       :     0.003993s : 31.54%
    event_method                         :     0.000016s :  0.13%
    auto_monad                           :     0.000061s :  0.48%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.23%
    rewriter_before_opt_a                :     0.000066s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000051s :  0.41%
    jit_opt_a.loop_unroll                :     0.000033s :  0.26%
    jit_opt_a.a_1                        :     0.000564s :  4.45%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000477s :  3.77%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.14%
    jit_opt_a.cse                        :     0.000030s :  0.23%
    jit_opt_a.replace_applicator         :     0.000016s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000029s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000490s :  3.87%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000023s :  0.18%
    loop_unroll                          :     0.000401s :  3.17%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.17%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000011s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000043s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000013s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000433s :  3.42%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000026s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005205s : 41.11%
    execute                              :     0.000007s :  0.06%


Time group info:
------[substitution.]   0.000169    23
 1.22% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.98% :     0.000005s :      3: substitution.graph_param_transform
82.74% :     0.000139s :      4: substitution.inline
 1.61% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.32% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000003s :      2: substitution.replace_old_param
 6.48% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003943     2
81.83% :     0.003227s :      1: type_inference.infer
18.17% :     0.000716s :      1: type_inference.specialize
------[replace.]   0.000058     6
66.67% :     0.000039s :      4: replace.inline
33.33% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000147     6
93.34% :     0.000137s :      4: match.inline
 6.66% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.36% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.28% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000001s :     10: predicate.addn_zero_filter
 1.76% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.17% :     0.000001s :     10: predicate.compare_switch_simplify
 1.50% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.46% :     0.000001s :      3: predicate.graph_param_transform
 5.42% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.93% :     0.000002s :     15: predicate.load_eliminater
 1.87% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.27% :     0.000001s :     10: predicate.minmaximum_grad
 1.66% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.28% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.85% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.46% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.62% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.68% :     0.000008s :     53: predicate.switch_simplify
 1.38% :     0.000001s :     10: predicate.tile_eliminate
 1.30% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.45% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.78% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.47% :     0.000002s :     10: predicate.value_based_eliminate
 0.43% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000484    12
48.74% :     0.000236s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.61% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.66% :     0.000221s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017457    76
 0.27% :     0.000047s :      1: add_recomputation
 0.37% :     0.000064s :      1: auto_monad
 0.09% :     0.000015s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.48% :     0.000259s :      1: bootstrap
 0.15% :     0.000025s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.13% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.00% :     0.004538s :      1: jit_opt_a
 0.82% :     0.000143s :      1: jit_opt_after_cconv
 0.29% :     0.000051s :      1: jit_opt_b
 2.34% :     0.000408s :      1: loop_unroll
 2.85% :     0.000498s :      1: mutable_eliminate
 4.32% :     0.000754s :     26: opt.transform.jit_opt_a
 0.27% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.52% :     0.000440s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000031s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 1.38% :     0.000241s :      1: renormalize.infer
 1.31% :     0.000229s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000031s :      1: rewriter_after_opt_a
 0.39% :     0.000069s :      1: rewriter_before_opt_a
 0.41% :     0.000072s :      1: symbol_engine_optimizer
29.87% :     0.005215s :      1: task_emit
22.93% :     0.004003s :      1: type_inference
 0.25% :     0.000043s :      1: validate


TotalTime = 0.01992, [33]
[bootstrap]: 0.00033641
[type_inference]: 0.00500378
[event_method]: 1.709e-05
[auto_monad]: 6.631e-05
[graph_reusing]: 5.56998e-06
[pre_auto_parallel]: 2.38998e-06
[py_interpret_to_execute]: 3.587e-05
[rewriter_before_opt_a]: 7.237e-05
[expand_dump_flag]: 3.14999e-06
[jit_opt_a]: 0.00590961, [2]
    [Cycle 1]: 0.00192459, [27]
        [switch_simplify]: 5.579e-05
        [loop_unroll]: 2.909e-05
        [a_1]: 0.00056422
        [with_stream_mark]: 1.991e-05
        [recompute_prepare]: 7.48999e-06
        [updatestate_depend_eliminate]: 3.02002e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [parameter_eliminate]: 1.40001e-06
        [specialize_transform]: 6.28998e-06
        [updatestate_useless_node_eliminater]: 5.46002e-06
        [accelerated_algorithm]: 5.77001e-06
        [meta_shard_fg_expand]: 1.84e-06
        [get_grad_eliminate_]: 5.46e-06
        [merge_forward]: 3.51001e-06
        [cell_reuse_recompute_pass]: 1.35001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.432e-05
        [j_node_and_user_rematch]: 1.046e-05
        [meta_fg_expand]: 2.29999e-06
        [replace_old_param]: 9.19e-06
        [inline_without_move]: 5.22e-06
        [renormalize]: 0.00093261
        [add_forward_monad_depend]: 5.67999e-06
        [auto_monad_grad]: 2.98998e-06
        [auto_monad_eliminator]: 1.42e-05
        [cse]: 2.759e-05
        [replace_applicator]: 1.399e-05
    [Cycle 2]: 0.00032976, [27]
        [switch_simplify]: 5.29998e-06
        [loop_unroll]: 4.60999e-06
        [a_1]: 9.374e-05
        [with_stream_mark]: 9.67999e-06
        [recompute_prepare]: 5.19998e-06
        [updatestate_depend_eliminate]: 2.63998e-06
        [updatestate_assign_eliminate]: 1.94999e-06
        [updatestate_loads_eliminate]: 1.81998e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 4.77e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.17e-06
        [meta_shard_fg_expand]: 1.24998e-06
        [get_grad_eliminate_]: 4.53001e-06
        [merge_forward]: 2.59001e-06
        [cell_reuse_recompute_pass]: 2.48998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.39e-05
        [j_node_and_user_rematch]: 8.08001e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 7.02002e-06
        [inline_without_move]: 4.53001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.59e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 5.34998e-06
        [cse]: 1.054e-05
        [replace_applicator]: 1.158e-05
[py_interpret_to_execute_after_opt_a]: 1.186e-05
[rewriter_after_opt_a]: 3.476e-05
[convert_after_rewriter]: 7.19001e-06
[order_py_execute_after_rewriter]: 4.73001e-06
[mutable_eliminate]: 0.00071089
[jit_opt_b]: 5.025e-05, [1]
    [Cycle 1]: 4.294e-05, [2]
        [frontend_op_eliminate]: 1.685e-05
        [inline_after_opt_a]: 1.47e-05
[cconv]: 2.631e-05
[loop_unroll]: 0.00043562
[jit_opt_after_cconv]: 0.00014146, [1]
    [Cycle 1]: 0.00013536, [11]
        [c_1]: 2.108e-05
        [parameter_eliminate]: 2.84001e-06
        [updatestate_depend_eliminate]: 5.52001e-06
        [updatestate_assign_eliminate]: 2.54001e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [cse]: 1.984e-05
        [call_graph_tuple_transform]: 1.859e-05
        [tuple_list_get_item_eliminator]: 5.13002e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 2.59985e-07
        [switch_simplify]: 5.38002e-06
[remove_dup_value]: 1.457e-05
[partial_unused_args_eliminate]: 2.48002e-06
[environ_conv]: 5.00001e-06
[add_recomputation]: 4.81e-05
[cse_after_recomputation]: 2.007e-05, [1]
    [Cycle 1]: 1.435e-05, [1]
        [cse]: 8.87e-06
[auto_monad_reorder]: 1.426e-05
[get_jit_bprop_graph]: 2.02999e-06
[rewriter_after_jit_bprop_graph]: 5.05001e-06
[opt_after_jit_grad]: 0.00046597
[symbol_engine_optimizer]: 7.157e-05, [1]
    [Cycle 1]: 6.449e-05, [6]
        [build]: 3.24001e-06
        [elim_shapecalc]: 7.20998e-06
        [elim_not_effective]: 1.308e-05
        [opt_reshape]: 5.30001e-06
        [fold_const_symbol]: 8.23999e-06
        [renormalize]: 5.39992e-07
[validate]: 3.907e-05
[backend_pass]: 1.17e-06
[task_emit]: 0.00613715
[execute]: 8.13001e-06
Sums
    bootstrap                            :     0.000336s :  2.15%
    type_inference                       :     0.005004s : 31.97%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000066s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.23%
    rewriter_before_opt_a                :     0.000072s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000658s :  4.20%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000933s :  5.96%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000038s :  0.24%
    jit_opt_a.replace_applicator         :     0.000026s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000035s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000711s :  4.54%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000026s :  0.17%
    loop_unroll                          :     0.000436s :  2.78%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000048s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000466s :  2.98%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006137s : 39.21%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000225    23
 1.13% :     0.000003s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 2.32% :     0.000005s :      3: substitution.graph_param_transform
85.09% :     0.000192s :      4: substitution.inline
 1.52% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.02% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.78% :     0.000004s :      2: substitution.replace_old_param
 5.66% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004934     2
82.40% :     0.004066s :      1: type_inference.infer
17.60% :     0.000868s :      1: type_inference.specialize
------[replace.]   0.000075     6
67.72% :     0.000051s :      4: replace.inline
32.28% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000201     6
94.28% :     0.000189s :      4: match.inline
 5.72% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.62% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 2.17% :     0.000003s :     10: predicate.arithmetic_simplify
 1.29% :     0.000001s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000001s :     10: predicate.depend_value_elim
 1.24% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.64% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.24% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.33% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.43% :     0.000002s :     10: predicate.environ_get_eliminate
 1.43% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.51% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.35% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.89% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.04% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.62% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000000s :      3: predicate.opt_reshape
 2.04% :     0.000002s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.24% :     0.000001s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.01% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.19% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.59% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.51% :     0.000009s :     53: predicate.switch_simplify
 1.42% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.54% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.46% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.33% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.85% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.93% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000657    12
48.09% :     0.000316s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.90% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.00% :     0.000302s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021828    76
 0.24% :     0.000051s :      1: add_recomputation
 0.32% :     0.000070s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.62% :     0.000354s :      1: bootstrap
 0.13% :     0.000029s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.09% :     0.005913s :      1: jit_opt_a
 0.66% :     0.000144s :      1: jit_opt_after_cconv
 0.24% :     0.000053s :      1: jit_opt_b
 2.03% :     0.000443s :      1: loop_unroll
 3.29% :     0.000719s :      1: mutable_eliminate
 4.03% :     0.000880s :     26: opt.transform.jit_opt_a
 0.21% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.17% :     0.000474s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000039s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 2.36% :     0.000516s :      1: renormalize.infer
 1.87% :     0.000407s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000037s :      1: rewriter_after_opt_a
 0.34% :     0.000075s :      1: rewriter_before_opt_a
 0.34% :     0.000075s :      1: symbol_engine_optimizer
28.20% :     0.006155s :      1: task_emit
22.98% :     0.005016s :      1: type_inference
 0.27% :     0.000059s :      1: validate


TotalTime = 0.0144171, [30]
[bootstrap]: 0.00023601
[type_inference]: 0.00612573
[event_method]: 2.045e-05
[auto_monad]: 6.814e-05
[graph_reusing]: 4.68001e-06
[pre_auto_parallel]: 2.21003e-06
[py_interpret_to_execute]: 5.106e-05
[rewriter_before_opt_a]: 7.862e-05
[expand_dump_flag]: 2.17999e-06
[jit_opt_a]: 0.0056137, [2]
    [Cycle 1]: 0.00219435, [27]
        [switch_simplify]: 9.506e-05
        [loop_unroll]: 3.555e-05
        [a_1]: 0.00070783
        [with_stream_mark]: 1.401e-05
        [recompute_prepare]: 9.64e-06
        [updatestate_depend_eliminate]: 3.61999e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.42e-06
        [specialize_transform]: 8.53001e-06
        [updatestate_useless_node_eliminater]: 6.40002e-06
        [accelerated_algorithm]: 6.99001e-06
        [meta_shard_fg_expand]: 1.77999e-06
        [get_grad_eliminate_]: 6.01e-06
        [merge_forward]: 3.08e-06
        [cell_reuse_recompute_pass]: 1.59998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.379e-05
        [j_node_and_user_rematch]: 1.059e-05
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 1.016e-05
        [inline_without_move]: 6.79999e-06
        [renormalize]: 0.00101443
        [add_forward_monad_depend]: 5.09e-06
        [auto_monad_grad]: 2.34999e-06
        [auto_monad_eliminator]: 1.089e-05
        [cse]: 1.839e-05
        [replace_applicator]: 1.405e-05
    [Cycle 2]: 0.00032752, [27]
        [switch_simplify]: 6.41e-06
        [loop_unroll]: 4.89e-06
        [a_1]: 9.168e-05
        [with_stream_mark]: 9.05999e-06
        [recompute_prepare]: 5.37001e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 2.21003e-06
        [updatestate_loads_eliminate]: 2.11998e-06
        [parameter_eliminate]: 1.18001e-06
        [specialize_transform]: 5.02e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.44e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 4.33999e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.248e-05
        [j_node_and_user_rematch]: 8.02998e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 7.5e-06
        [inline_without_move]: 1.246e-05
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.01997e-06
        [auto_monad_grad]: 7.29982e-07
        [auto_monad_eliminator]: 4.90999e-06
        [cse]: 1.135e-05
        [replace_applicator]: 5.04e-06
[py_interpret_to_execute_after_opt_a]: 9.57001e-06
[rewriter_after_opt_a]: 2.829e-05
[convert_after_rewriter]: 5.59e-06
[order_py_execute_after_rewriter]: 4.82e-06
[mutable_eliminate]: 0.00057261
[jit_opt_b]: 5.63e-05, [1]
    [Cycle 1]: 4.811e-05, [2]
        [frontend_op_eliminate]: 1.817e-05
        [inline_after_opt_a]: 1.648e-05
[cconv]: 2.386e-05
[loop_unroll]: 0.0004389
[jit_opt_after_cconv]: 0.00014377, [1]
    [Cycle 1]: 0.00013728, [11]
        [c_1]: 2.207e-05
        [parameter_eliminate]: 3.14999e-06
        [updatestate_depend_eliminate]: 5.72999e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [cse]: 1.907e-05
        [call_graph_tuple_transform]: 1.781e-05
        [tuple_list_get_item_eliminator]: 5.73997e-06
        [none_parameter_eliminate]: 1.00999e-06
        [renormalize]: 6.89994e-07
        [switch_simplify]: 5.76998e-06
[remove_dup_value]: 1.001e-05
[partial_unused_args_eliminate]: 1.79e-06
[environ_conv]: 4.28001e-06
[add_recomputation]: 4.209e-05
[cse_after_recomputation]: 2.337e-05, [1]
    [Cycle 1]: 1.78e-05, [1]
        [cse]: 1.051e-05
[auto_monad_reorder]: 1.285e-05
[get_jit_bprop_graph]: 1.60999e-06
[rewriter_after_jit_bprop_graph]: 5.59e-06
[opt_after_jit_grad]: 0.00053814
[symbol_engine_optimizer]: 7.717e-05, [1]
    [Cycle 1]: 6.968e-05, [6]
        [build]: 2.31998e-06
        [elim_shapecalc]: 8.23999e-06
        [elim_not_effective]: 1.475e-05
        [opt_reshape]: 5.66e-06
        [fold_const_symbol]: 8.38001e-06
        [renormalize]: 1.16997e-06
[validate]: 3.155e-05
Sums
    bootstrap                            :     0.000236s :  2.20%
    type_inference                       :     0.006126s : 57.14%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000068s :  0.64%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000051s :  0.48%
    rewriter_before_opt_a                :     0.000079s :  0.73%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000101s :  0.95%
    jit_opt_a.loop_unroll                :     0.000040s :  0.38%
    jit_opt_a.a_1                        :     0.000800s :  7.46%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000014s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.16%
    jit_opt_a.inline_without_move        :     0.000019s :  0.18%
    jit_opt_a.renormalize                :     0.001014s :  9.46%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.15%
    jit_opt_a.cse                        :     0.000030s :  0.28%
    jit_opt_a.replace_applicator         :     0.000019s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.09%
    rewriter_after_opt_a                 :     0.000028s :  0.26%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000573s :  5.34%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.15%
    cconv                                :     0.000024s :  0.22%
    loop_unroll                          :     0.000439s :  4.09%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000010s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000042s :  0.39%
    cse_after_recomputation.cse          :     0.000011s :  0.10%
    auto_monad_reorder                   :     0.000013s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000538s :  5.02%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000032s :  0.29%


Time group info:
------[substitution.]   0.000265    26
 0.77% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 1.68% :     0.000004s :      3: substitution.graph_param_transform
85.33% :     0.000226s :      6: substitution.inline
 1.08% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.71% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.57% :     0.000004s :      2: substitution.replace_old_param
 3.04% :     0.000008s :      1: substitution.switch_simplify
 4.33% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006055     2
81.20% :     0.004917s :      1: type_inference.infer
18.80% :     0.001138s :      1: type_inference.specialize
------[replace.]   0.000131     9
52.33% :     0.000068s :      6: replace.inline
18.36% :     0.000024s :      1: replace.switch_simplify
29.31% :     0.000038s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000240     9
92.86% :     0.000223s :      6: match.inline
 3.03% :     0.000007s :      1: match.switch_simplify
 4.10% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   801
 1.30% :     0.000002s :     12: predicate.accumulaten_eliminater
 2.02% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000002s :     12: predicate.addn_check_dump
 1.40% :     0.000002s :     12: predicate.addn_zero_filter
 1.84% :     0.000003s :     12: predicate.arithmetic_simplify
 1.17% :     0.000002s :     12: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000002s :     12: predicate.compare_switch_simplify
 1.42% :     0.000002s :     12: predicate.depend_value_elim
 1.11% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.60% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.49% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.70% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     12: predicate.environ_get_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.21% :     0.000002s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 5.03% :     0.000007s :     26: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.73% :     0.000002s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     17: predicate.load_eliminater
 1.06% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.20% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.97% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     12: predicate.merge_addn
 1.11% :     0.000002s :     12: predicate.minmaximum_grad
 1.49% :     0.000002s :      3: predicate.mutable_eliminate
 0.37% :     0.000001s :      3: predicate.opt_reshape
 1.93% :     0.000003s :     17: predicate.partial_eliminate
 1.17% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     12: predicate.reduce_eliminate
 1.70% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.48% :     0.000002s :     20: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     12: predicate.reshape_eliminate
 1.18% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     12: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.53% :     0.000005s :     20: predicate.switch_defer_inline
 2.46% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.21% :     0.000011s :     65: predicate.switch_simplify
 1.19% :     0.000002s :     12: predicate.tile_eliminate
 1.25% :     0.000002s :     12: predicate.transpose_eliminate
 1.36% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.14% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 3.21% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     12: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.84% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000879    15
43.91% :     0.000386s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.26% :     0.000090s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.83% :     0.000403s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016566    72
 0.28% :     0.000047s :      1: add_recomputation
 0.43% :     0.000072s :      1: auto_monad
 0.10% :     0.000016s :      1: auto_monad_reorder
 1.48% :     0.000245s :      1: bootstrap
 0.16% :     0.000027s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.15% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
33.91% :     0.005617s :      1: jit_opt_a
 0.89% :     0.000147s :      1: jit_opt_after_cconv
 0.36% :     0.000059s :      1: jit_opt_b
 2.69% :     0.000446s :      1: loop_unroll
 3.52% :     0.000582s :      1: mutable_eliminate
 6.50% :     0.001077s :     26: opt.transform.jit_opt_a
 0.29% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000028s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.32% :     0.000550s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.32% :     0.000054s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000013s :      1: remove_dup_value
 3.10% :     0.000513s :      1: renormalize.infer
 2.98% :     0.000493s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000031s :      1: rewriter_after_opt_a
 0.49% :     0.000082s :      1: rewriter_before_opt_a
 0.48% :     0.000080s :      1: symbol_engine_optimizer
37.06% :     0.006140s :      1: type_inference


TotalTime = 0.0236784, [33]
[bootstrap]: 0.00040083
[type_inference]: 0.00598261
[event_method]: 1.778e-05
[auto_monad]: 7.067e-05
[graph_reusing]: 6.10002e-06
[pre_auto_parallel]: 2.66e-06
[py_interpret_to_execute]: 4.279e-05
[rewriter_before_opt_a]: 7.945e-05
[expand_dump_flag]: 3.31001e-06
[jit_opt_a]: 0.00659204, [2]
    [Cycle 1]: 0.0019682, [27]
        [switch_simplify]: 6.277e-05
        [loop_unroll]: 3.147e-05
        [a_1]: 0.00059398
        [with_stream_mark]: 2.633e-05
        [recompute_prepare]: 1.271e-05
        [updatestate_depend_eliminate]: 4.36002e-06
        [updatestate_assign_eliminate]: 3.49001e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 2.46998e-06
        [specialize_transform]: 6.72002e-06
        [updatestate_useless_node_eliminater]: 6.28002e-06
        [accelerated_algorithm]: 5.69999e-06
        [meta_shard_fg_expand]: 2.37001e-06
        [get_grad_eliminate_]: 5.57001e-06
        [merge_forward]: 3.56001e-06
        [cell_reuse_recompute_pass]: 1.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 3.266e-05
        [j_node_and_user_rematch]: 9.69999e-06
        [meta_fg_expand]: 2.41e-06
        [replace_old_param]: 1.015e-05
        [inline_without_move]: 5.40001e-06
        [renormalize]: 0.00082441
        [add_forward_monad_depend]: 8.92e-06
        [auto_monad_grad]: 2.90002e-06
        [auto_monad_eliminator]: 2.117e-05
        [cse]: 3.318e-05
        [replace_applicator]: 2.02e-05
    [Cycle 2]: 0.00038665, [27]
        [switch_simplify]: 6.04001e-06
        [loop_unroll]: 4.97999e-06
        [a_1]: 9.329e-05
        [with_stream_mark]: 1.688e-05
        [recompute_prepare]: 6.07001e-06
        [updatestate_depend_eliminate]: 3.58e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 1.81003e-06
        [specialize_transform]: 6.10002e-06
        [updatestate_useless_node_eliminater]: 5.42999e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 2.37999e-06
        [get_grad_eliminate_]: 4.95001e-06
        [merge_forward]: 4.55001e-06
        [cell_reuse_recompute_pass]: 3.53e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.009e-05
        [j_node_and_user_rematch]: 1.009e-05
        [meta_fg_expand]: 2.41e-06
        [replace_old_param]: 8.87e-06
        [inline_without_move]: 1.076e-05
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.79e-06
        [auto_monad_grad]: 1.26002e-06
        [auto_monad_eliminator]: 1.057e-05
        [cse]: 1.493e-05
        [replace_applicator]: 6.63e-06
[py_interpret_to_execute_after_opt_a]: 1.891e-05
[rewriter_after_opt_a]: 4.6e-05
[convert_after_rewriter]: 8.28999e-06
[order_py_execute_after_rewriter]: 5.39e-06
[mutable_eliminate]: 0.00087707
[jit_opt_b]: 6.192e-05, [1]
    [Cycle 1]: 5.132e-05, [2]
        [frontend_op_eliminate]: 1.967e-05
        [inline_after_opt_a]: 1.82e-05
[cconv]: 3.729e-05
[loop_unroll]: 0.00056509
[jit_opt_after_cconv]: 0.00016978, [1]
    [Cycle 1]: 0.00016182, [11]
        [c_1]: 2.246e-05
        [parameter_eliminate]: 3.46999e-06
        [updatestate_depend_eliminate]: 6.90002e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [cse]: 3.022e-05
        [call_graph_tuple_transform]: 2.546e-05
        [tuple_list_get_item_eliminator]: 5.31002e-06
        [none_parameter_eliminate]: 1.29998e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 5.54e-06
[remove_dup_value]: 1.581e-05
[partial_unused_args_eliminate]: 2.10002e-06
[environ_conv]: 6.08998e-06
[add_recomputation]: 5.545e-05
[cse_after_recomputation]: 2.822e-05, [1]
    [Cycle 1]: 1.946e-05, [1]
        [cse]: 1.109e-05
[auto_monad_reorder]: 1.778e-05
[get_jit_bprop_graph]: 2.59999e-06
[rewriter_after_jit_bprop_graph]: 7.75e-06
[opt_after_jit_grad]: 0.00064006
[symbol_engine_optimizer]: 8.989e-05, [1]
    [Cycle 1]: 8.104e-05, [6]
        [build]: 5.49e-06
        [elim_shapecalc]: 9.42999e-06
        [el
TotalTime = 0.0180135, [33]
[bootstrap]: 0.00024959
[type_inference]: 0.00411076
[event_method]: 1.739e-05
[auto_monad]: 6.45e-05
[graph_reusing]: 5.71998e-06
[pre_auto_parallel]: 2.62001e-06
[py_interpret_to_execute]: 3.078e-05
[rewriter_before_opt_a]: 6.989e-05
[expand_dump_flag]: 3.23998e-06
[jit_opt_a]: 0.00495156, [2]
    [Cycle 1]: 0.00150889, [27]
        [switch_simplify]: 5.27e-05
        [loop_unroll]: 3.51e-05
        [a_1]: 0.00052579
        [with_stream_mark]: 1.546e-05
        [recompute_prepare]: 7.57002e-06
        [updatestate_depend_eliminate]: 3.58e-06
        [updatestate_assign_eliminate]: 3.43999e-06
        [updatestate_loads_eliminate]: 2.79001e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 5.86e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 2.07001e-06
        [get_grad_eliminate_]: 5.56e-06
        [merge_forward]: 3.70998e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.462e-05
        [j_node_and_user_rematch]: 9.38002e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 9.10001e-06
        [inline_without_move]: 5.72001e-06
        [renormalize]: 0.00055449
        [add_forward_monad_depend]: 4.95001e-06
        [auto_monad_grad]: 2.12001e-06
        [auto_monad_eliminator]: 1.268e-05
        [cse]: 3.008e-05
        [replace_applicator]: 1.163e-05
    [Cycle 2]: 0.00031897, [27]
        [switch_simplify]: 5.97001e-06
        [loop_unroll]: 5.00001e-06
        [a_1]: 9.188e-05
        [with_stream_mark]: 8.28001e-06
        [recompute_prepare]: 5.25001e-06
        [updatestate_depend_eliminate]: 2.41998e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 5.17999e-06
        [updatestate_useless_node_eliminater]: 4.79998e-06
        [accelerated_algorithm]: 5.24e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.57e-06
        [merge_forward]: 2.27001e-06
        [cell_reuse_recompute_pass]: 1.62999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.247e-05
        [j_node_and_user_rematch]: 8.08001e-06
        [meta_fg_expand]: 1.74998e-06
        [replace_old_param]: 7.23e-06
        [inline_without_move]: 4.90999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 9.70002e-07
        [auto_monad_grad]: 7.90023e-07
        [auto_monad_eliminator]: 5.39e-06
        [cse]: 1.035e-05
        [replace_applicator]: 4.99998e-06
[py_interpret_to_execute_after_opt_a]: 8.88002e-06
[rewriter_after_opt_a]: 3.373e-05
[convert_after_rewriter]: 6.29001e-06
[order_py_execute_after_rewriter]: 6.69001e-06
[mutable_eliminate]: 0.00044818
[jit_opt_b]: 5.79e-05, [1]
    [Cycle 1]: 5.024e-05, [2]
        [frontend_op_eliminate]: 1.981e-05
        [inline_after_opt_a]: 1.599e-05
[cconv]: 2.503e-05
[loop_unroll]: 0.00045651
[jit_opt_after_cconv]: 0.0001398, [1]
    [Cycle 1]: 0.00013325, [11]
        [c_1]: 2.174e-05
        [parameter_eliminate]: 2.12999e-06
        [updatestate_depend_eliminate]: 5.64e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [cse]: 1.87e-05
        [call_graph_tuple_transform]: 1.707e-05
        [tuple_list_get_item_eliminator]: 5.32999e-06
        [none_parameter_eliminate]: 1.37999e-06
        [renormalize]: 4.7998e-07
        [switch_simplify]: 5.67999e-06
[remove_dup_value]: 1.256e-05
[partial_unused_args_eliminate]: 1.98002e-06
[environ_conv]: 5.13002e-06
[add_recomputation]: 4.943e-05
[cse_after_recomputation]: 2.347e-05, [1]
    [Cycle 1]: 1.787e-05, [1]
        [cse]: 1.036e-05
[auto_monad_reorder]: 1.525e-05
[get_jit_bprop_graph]: 1.86998e-06
[rewriter_after_jit_bprop_graph]: 6.44999e-06
[opt_after_jit_grad]: 0.00052235
[symbol_engine_optimizer]: 7.688e-05, [1]
    [Cycle 1]: 6.933e-05, [6]
        [build]: 4.57998e-06
        [elim_shapecalc]:im_not_effective]: 1.601e-05
        [opt_reshape]: 6.48e-06
        [fold_const_symbol]: 1.005e-05
        [renormalize]: 1.05999e-06
[validate]: 4.888e-05
[backend_pass]: 1.05999e-06
[task_emit]: 0.00748971
[execute]: 9.12999e-06
Sums
    bootstrap                            :     0.000401s :  2.15%
    type_inference                       :     0.005983s : 32.05%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000071s :  0.38%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.23%
    rewriter_before_opt_a                :     0.000079s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000069s :  0.37%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000687s :  3.68%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000053s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.10%
    jit_opt_a.inline_without_move        :     0.000016s :  0.09%
    jit_opt_a.renormalize                :     0.000824s :  4.42%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000032s :  0.17%
    jit_opt_a.cse                        :     0.000048s :  0.26%
    jit_opt_a.replace_applicator         :     0.000027s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.10%
    rewriter_after_opt_a                 :     0.000046s :  0.25%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000877s :  4.70%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.10%
    cconv                                :     0.000037s :  0.20%
    loop_unroll                          :     0.000565s :  3.03%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000030s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.08 7.41001e-06
        [elim_not_effective]: 1.276e-05
        [opt_reshape]: 6.17001e-06
        [fold_const_symbol]: 8.80001e-06
        [renormalize]: 4.49974e-07
[validate]: 3.647e-05
[backend_pass]: 1.52001e-06
[task_emit]: 0.00628207
[execute]: 9.17001e-06
Sums
    bootstrap                            :     0.000250s :  1.76%
    type_inference                       :     0.004111s : 28.96%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000064s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.22%
    rewriter_before_opt_a                :     0.000070s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.41%
    jit_opt_a.loop_unroll                :     0.000040s :  0.28%
    jit_opt_a.a_1                        :     0.000618s :  4.35%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000555s :  3.91%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000040s :  0.28%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000034s :  0.24%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000007s :  0.05%
    mutable_eliminate                    :     0.000448s :  3.16%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000025s :  0.18%
    loop_unroll                          :     0.000457s :  3.22%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value               %
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000055s :  0.30%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000018s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000640s :  3.43%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000049s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007490s : 40.13%
    execute                              :     0.000009s :  0.05%

      :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000049s :  0.35%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000522s :  3.68%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.26%
    backend_pass                         :     0.000002s :  0.01%
    task_emit                            :     0.006282s : 44.26%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000189    23
 0.90% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.71% :     0.000005s :      3: substitution.graph_param_transform
83.15% :     0.000157s :      4: substitution.inline
 1.61% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.12% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.16% :     0.000004s :      2: substitution.replace_old_param
 6.67% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004049     2
81.30% :     0.003292s :      1: type_inference.infer
18.70% :     0.000757s :      1: type_inference.specialize
------[replace.]   0.000066     6
68.78% :     0.000045s :      4: replace.inline
31.22% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000166     6
93.13% :     0.000154s :      4: match.inline
 6.87% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 1.93% :     0.000002s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.14% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.33% :     0.000002s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.40% :     0.000000s :      3: predicate.graph_param_transform
 5.38% :     0.000006s :     22: predicate.inline
 1.12% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000002s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.01% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.74% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.46% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3:
Time group info:
------[substitution.]   0.000240    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.83% :     0.000007s :      3: substitution.graph_param_transform
82.78% :     0.000199s :      4: substitution.inline
 1.83% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.50% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.38% :     0.000006s :      2: substitution.replace_old_param
 6.29% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005899     2
83.85% :     0.004947s :      1: type_inference.infer
16.15% :     0.000953s :      1: type_inference.specialize
------[replace.]   0.000082     6
65.49% :     0.000054s :      4: replace.inline
34.51% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000209     6
93.53% :     0.000196s :      4: match.inline
 6.47% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.21% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000002s :     10: predicate.addn_zero_filter
 2.45% :     0.000003s :     10: predicate.arithmetic_simplify
 1.08% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000002s :     10: predicate.depend_value_elim
 0.98% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.04% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.18% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.84% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.03% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.00% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 7.07% :     0.000009s :     22: predicate.inline
 1.12% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000002s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.53% :     0.000002s :     15: predicate.load_eliminater
 1.71% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.71% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.93% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 2.47% :     0.000003s :      3: predicate.mutable_eliminate
 0.66% :     0.000001s :      3: predicate.opt_reshape
 1.97% :     0.000003s :     15: predicate.partial_eliminate
 1.02% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.76% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.42% :     0.000001s :      3: predicate.reset_defer_inline
 1.10% :     0.000001s :     10: predicate.reshape_eliminate
 1.05% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.18% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.86% :     0.000009s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.69% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.04% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000504    12
48.86% :     0.000246s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.84% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.30% :     0.000228s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019511    76
 0.27% :     0.000052s :      1: add_recomputation
 0.35% :     0.000068s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.35% :     0.000262s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.07% :     0.000013s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.39% :     0.004954s :      1: jit_opt_a
 0.73% :     0.000142s :      1: jit_opt_after_cconv
 0.31% :     0.000061s :      1: jit_opt_b
 2.38% :     0.000464s :      1: loop_unroll
 2.79% :     0.000544s :      1: mutable_eliminate
 4.26% :     0.000831s :     26: opt.transform.jit_opt_a
 0.24% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000029s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.73% :     0.000533s :      1: opt_after_jit_grad
 0.05% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.49% :     0.000291s :      1: renormalize.infer
 1.32% :     0.000257s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000037s :      1: rewriter_after_opt_a
 0.37% :     0.000073s :      1: rewriter_before_opt_a
 0.41% :     0.000080s :      1: symbol_engine_optimizer
32.27% :     0.006296s :       predicate.row_tensor_eliminate
 1.22% :     0.000002s :     10: predicate.same_eliminate
 0.93% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.16% :     0.000002s :      6: predicate.special_op_eliminate
 1.15% :     0.000002s :      6: predicate.specialize_transform
 1.06% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.12% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 1.78% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.72% :     0.000009s :     53: predicate.switch_simplify
 1.71% :     0.000002s :     10: predicate.tile_eliminate
 1.30% :     0.000002s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.05% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.56% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.44% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.99% :     0.000003s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000732    12
53.12% :     0.000389s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.04% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.84% :     0.000306s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025575    76
 0.24% :     0.000060s :      1: add_recomputation
 0.29% :     0.000075s :      1: auto_monad
 0.08% :     0.000021s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.66% :     0.000425s :      1: bootstrap
 0.16% :     0.000041s :      1: cconv
 0.05% :     0.000012s :      1: convert_after_rewriter
 0.12% :     0.000031s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.08% :     0.000022s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
25.79% :     0.006596s :      1: jit_opt_a
 0.68% :     0.000173s :      1: jit_opt_after_cconv
 0.25% :     0.000065s :      1: jit_opt_b
 2.28% :     0.000582s :      1: loop_unroll
 3.51% :     0.000897s :      1: mutable_eliminate
 3.69% :     0.000943s :     26: opt.transform.jit_opt_a
 0.21% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000029s :      4: opt.transform.jit_opt_b
 0.07% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000038s :      4: opt.transform.symbol_engine_opt
 2.56% :     0.000653s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000047s :      1: py_interpret_to_execute
 0.09% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 1.70% :     0.000435s :      1: renormalize.infer
 1.48% :     0.000378s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000050s :      1: rewriter_after_opt_a
 0.33% :     0.000084s :      1: rewriter_before_opt_a
 0.36% :     0.000093s :      1: symbol_engine_optimizer
29.37% :     0.007511s :      1: task_emit
21.14% :     0.004124s :      1: type_inference
 0.30% :     0.000059s :      1: validate

1: task_emit
23.46% :     0.006000s :      1: type_inference
 0.29% :     0.000074s :      1: validate


TotalTime = 0.0148821, [30]
[bootstrap]: 0.00023126
[type_inference]: 0.00638963
[event_method]: 2.154e-05
[auto_monad]: 7.255e-05
[graph_reusing]: 5.69e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 5.437e-05
[rewriter_before_opt_a]: 8.193e-05
[expand_dump_flag]: 3.06001e-06
[jit_opt_a]: 0.0058367, [2]
    [Cycle 1]: 0.00218979, [27]
        [switch_simplify]: 0.00010179
        [loop_unroll]: 3.332e-05
        [a_1]: 0.00070288
        [with_stream_mark]: 1.286e-05
        [recompute_prepare]: 6.64999e-06
        [updatestate_depend_eliminate]: 2.54001e-06
        [updatestate_assign_eliminate]: 2.15002e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 6.79001e-06
        [updatestate_useless_node_eliminater]: 5.54998e-06
        [accelerated_algorithm]: 5.97999e-06
        [meta_shard_fg_expand]: 1.40999e-06
        [get_grad_eliminate_]: 5.55001e-06
        [merge_forward]: 2.91e-06
        [cell_reuse_recompute_pass]: 9.30013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.35e-05
        [j_node_and_user_rematch]: 9.72999e-06
        [meta_fg_expand]: 2.56e-06
        [replace_old_param]: 8.83001e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 0.0010401
        [add_forward_monad_depend]: 5.49e-06
        [auto_monad_grad]: 2.08998e-06
        [auto_monad_eliminator]: 1.139e-05
        [cse]: 1.867e-05
        [replace_applicator]: 1.306e-05
    [Cycle 2]: 0.00035859, [27]
        [switch_simplify]: 3.81e-05
        [loop_unroll]: 5.49998e-06
        [a_1]: 9.153e-05
        [with_stream_mark]: 9.72001e-06
        [recompute_prepare]: 5.39e-06
        [updatestate_depend_eliminate]: 2.76e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 5.30001e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 4.95001e-06
        [meta_shard_fg_expand]: 1.11002e-06
        [get_grad_eliminate_]: 4.74e-06
        [merge_forward]: 2.81e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.345e-05
        [j_node_and_user_rematch]: 7.96001e-06
        [meta_fg_expand]: 1.76e-06
        [replace_old_param]: 7.33e-06
        [inline_without_move]: 4.65001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 5.07999e-06
        [cse]: 1.123e-05
        [replace_applicator]: 5.05999e-06
[py_interpret_to_execute_after_opt_a]: 1.037e-05
[rewriter_after_opt_a]: 3.077e-05
[convert_after_rewriter]: 6.46e-06
[order_py_execute_after_rewriter]: 5.10999e-06
[mutable_eliminate]: 0.00061737
[jit_opt_b]: 4.922e-05, [1]
    [Cycle 1]: 4.353e-05, [2]
        [frontend_op_eliminate]: 1.62e-05
        [inline_after_opt_a]: 1.542e-05
[cconv]: 2.485e-05
[loop_unroll]: 0.00044486
[jit_opt_after_cconv]: 0.00013768, [1]
    [Cycle 1]: 0.00013105, [11]
        [c_1]: 2.144e-05
        [parameter_eliminate]: 3.01999e-06
        [updatestate_depend_eliminate]: 4.46002e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [cse]: 1.556e-05
        [call_graph_tuple_transform]: 1.855e-05
        [tuple_list_get_item_eliminator]: 5.22e-06
        [none_parameter_eliminate]: 1.00001e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.65001e-06
[remove_dup_value]: 1.161e-05
[partial_unused_args_eliminate]: 2.07001e-06
[environ_conv]: 3.81999e-06
[add_recomputation]: 4.442e-05
[cse_after_recomputation]: 2.198e-05, [1]
    [Cycle 1]: 1.643e-05, [1]
        [cse]: 1.024e-05
[auto_monad_reorder]: 1.237e-05
[get_jit_bprop_graph]: 2.12001e-06
[rewriter_after_jit_bprop_graph]: 5.91998e-06
[opt_after_jit_grad]: 0.00047244
[symbol_engine_optimizer]: 7.567e-05, [1]
    [Cycle 1]: 6.902e-05, [6]
        [build]: 3.32002e-06
        [elim_shapecalc]: 7.34002e-06
        [elim_not_effective]: 1.295e-05
        [opt_reshape]: 6.82002e-06
        [fold_const_symbol]: 1.001e-05
        [renormalize]: 4.50003e-07
[validate]: 3.421e-05
Sums
    bootstrap                            :     0.000231s :  2.10%
    type_inference                       :     0.006390s : 57.98%
    event_method                         :     0.000022s :  0.20%
    auto_monad                           :     0.000073s :  0.66%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000054s :  0.49%
    rewriter_before_opt_a                :     0.000082s :  0.74%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000140s :  1.27%
    jit_opt_a.loop_unroll                :     0.000039s :  0.35%
    jit_opt_a.a_1                        :     0.000794s :  7.21%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.001040s :  9.44%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.15%
    jit_opt_a.cse                        :     0.000030s :  0.27%
    jit_opt_a.replace_applicator         :     0.000018s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.09%
    rewriter_after_opt_a                 :     0.000031s :  0.28%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000617s :  5.60%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.14%
    cconv                                :     0.000025s :  0.23%
    loop_unroll                          :     0.000445s :  4.04%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000012s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000044s :  0.40%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000012s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000472s :  4.29%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.31%


Time group info:
------[substitution.]   0.000284    26
 0.59% :     0.000002s :      2: substitution.elim_not_effective
 0.77% :     0.000002s :      2: substitution.fold_const_symbol
 1.80% :     0.000005s :      3: substitution.graph_param_transform
85.56% :     0.000243s :      6: substitution.inline
 1.26% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.84% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.26% :     0.000004s :      2: substitution.replace_old_param
 3.10% :     0.000009s :      1: substitution.switch_simplify
 3.82% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006326     2
78.98% :     0.004996s :      1: type_inference.infer
21.02% :     0.001330s :      1: type_inference.specialize
------[replace.]   0.000140     9
49.94% :     0.000070s :      6: replace.inline
21.62% :     0.000030s :      1: replace.switch_simplify
28.44% :     0.000040s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000257     9
93.19% :     0.000239s :      6: match.inline
 3.14% :     0.000008s :      1: match.switch_simplify
 3.67% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   801
 1.54% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.11% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     12: predicate.addn_check_dump
 1.44% :     0.000002s :     12: predicate.addn_zero_filter
 1.99% :     0.000003s :     12: predicate.arithmetic_simplify
 1.38% :     0.000002s :     12: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.19% :     0.000002s :     12: predicate.compare_switch_simplify
 1.44% :     0.000002s :     12: predicate.depend_value_elim
 1.41% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.32% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     12: predicate.environ_get_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 5.00% :     0.000007s :     26: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.44% :     0.000002s :      6: predicate.less_batch_normalization
 1.69% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     17: predicate.load_eliminater
 1.05% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.11% :     0.000005s :     40: predicate.loop_unroll_before_grad
 2.05% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.28% :     0.000002s :     12: predicate.merge_addn
 1.16% :     0.000002s :     12: predicate.minmaximum_grad
 1.74% :     0.000002s :      3: predicate.mutable_eliminate
 0.38% :     0.000000s :      3: predicate.opt_reshape
 1.94% :     0.000003s :     17: predicate.partial_eliminate
 1.51% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     12: predicate.reduce_eliminate
 1.60% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     20: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.20% :     0.000000s :      3: predicate.reset_defer_inline
 1.79% :     0.000002s :     12: predicate.reshape_eliminate
 1.28% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     12: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.81% :     0.000001s :      6: predicate.special_op_eliminate
 0.85% :     0.000001s :      6: predicate.specialize_transform
 1.64% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.73% :     0.000004s :     20: predicate.switch_defer_inline
 2.32% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.52% :     0.000011s :     65: predicate.switch_simplify
 1.37% :     0.000002s :     12: predicate.tile_eliminate
 1.24% :     0.000002s :     12: predicate.transpose_eliminate
 1.32% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.41% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.20% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 2.07% :     0.000003s :     12: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000865    15
41.16% :     0.000356s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.86% :     0.000103s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.98% :     0.000406s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017072    72
 0.28% :     0.000048s :      1: add_recomputation
 0.45% :     0.000076s :      1: auto_monad
 0.09% :     0.000015s :      1: auto_monad_reorder
 1.42% :     0.000243s :      1: bootstrap
 0.16% :     0.000028s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
34.22% :     0.005841s :      1: jit_opt_a
 0.82% :     0.000141s :      1: jit_opt_after_cconv
 0.31% :     0.000052s :      1: jit_opt_b
 2.66% :     0.000454s :      1: loop_unroll
 3.66% :     0.000624s :      1: mutable_eliminate
 6.41% :     0.001094s :     26: opt.transform.jit_opt_a
 0.28% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.81% :     0.000480s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.33% :     0.000057s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 3.55% :     0.000606s :      1: renormalize.infer
 2.49% :     0.000425s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000034s :      1: rewriter_after_opt_a
 0.50% :     0.000085s :      1: rewriter_before_opt_a
 0.46% :     0.000078s :      1: symbol_engine_optimizer
37.52% :     0.006405s :      1: type_inference


TotalTime = 0.0180444, [33]
[bootstrap]: 0.00025031
[type_inference]: 0.00425777
[event_method]: 1.66e-05
[auto_monad]: 6.634e-05
[graph_reusing]: 6.44999e-06
[pre_auto_parallel]: 2.42001e-06
[py_interpret_to_execute]: 3.316e-05
[rewriter_before_opt_a]: 8.314e-05
[expand_dump_flag]: 3.78001e-06
[jit_opt_a]: 0.00515556, [2]
    [Cycle 1]: 0.00162329, [27]
        [switch_simplify]: 5.333e-05
        [loop_unroll]: 3.145e-05
        [a_1]: 0.00051954
        [with_stream_mark]: 1.6e-05
        [recompute_prepare]: 7.38e-06
        [updatestate_depend_eliminate]: 3.26999e-06
        [updatestate_assign_eliminate]: 2.81999e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [parameter_eliminate]: 2.09e-06
        [specialize_transform]: 5.86998e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.67001e-06
        [meta_shard_fg_expand]: 1.97001e-06
        [get_grad_eliminate_]: 5.40001e-06
        [merge_forward]: 3.4e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.459e-05
        [j_node_and_user_rematch]: 9.01998e-06
        [meta_fg_expand]: 2.11e-06
        [replace_old_param]: 8.97999e-06
        [inline_without_move]: 5.28002e-06
        [renormalize]: 0.00067865
        [add_forward_monad_depend]: 6.39001e-06
        [auto_monad_grad]: 1.82001e-06
        [auto_monad_eliminator]: 1.468e-05
        [cse]: 3.034e-05
        [replace_applicator]: 1.287e-05
    [Cycle 2]: 0.00031954, [27]
        [switch_simplify]: 5.96e-06
        [loop_unroll]: 5.14e-06
        [a_1]: 8.975e-05
        [with_stream_mark]: 8.80001e-06
        [recompute_prepare]: 5.24e-06
        [updatestate_depend_eliminate]: 2.85002e-06
        [updatestate_assign_eliminate]: 2.01998e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 5.07999e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 5.11002e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.33002e-06
        [cell_reuse_recompute_pass]: 2.09999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.304e-05
        [j_node_and_user_rematch]: 7.61001e-06
        [meta_fg_expand]: 1.55999e-06
        [replace_old_param]: 7.93999e-06
        [inline_without_move]: 4.71002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 8.10018e-07
        [auto_monad_eliminator]: 5.12e-06
        [cse]: 9.39998e-06
        [replace_applicator]: 5.04e-06
[py_interpret_to_execute_after_opt_a]: 9.42999e-06
[rewriter_after_opt_a]: 3.274e-05
[convert_after_rewriter]: 6.74999e-06
[order_py_execute_after_rewriter]: 6.03002e-06
[mutable_eliminate]: 0.00048732
[jit_opt_b]: 4.805e-05, [1]
    [Cycle 1]: 4.137e-05, [2]
        [frontend_op_eliminate]: 1.583e-05
        [inline_after_opt_a]: 1.442e-05
[cconv]: 2.479e-05
[loop_unroll]: 0.00039647
[jit_opt_after_cconv]: 0.0001413, [1]
    [Cycle 1]: 0.000135, [11]
        [c_1]: 2.004e-05
        [parameter_eliminate]: 2.34001e-06
        [updatestate_depend_eliminate]: 5.22999e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.03002e-06
        [cse]: 2.037e-05
        [call_graph_tuple_transform]: 1.938e-05
        [tuple_list_get_item_eliminator]: 5.67001e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.37999e-06
[remove_dup_value]: 1.307e-05
[partial_unused_args_eliminate]: 2.66e-06
[environ_conv]: 4.87e-06
[add_recomputation]: 4.983e-05
[cse_after_recomputation]: 2.172e-05, [1]
    [Cycle 1]: 1.616e-05, [1]
        [cse]: 9.64e-06
[auto_monad_reorder]: 1.512e-05
[get_jit_bprop_graph]: 2.13998e-06
[rewriter_after_jit_bprop_graph]: 5.24998e-06
[opt_after_jit_grad]: 0.00046633
[symbol_engine_optimizer]: 7.132e-05, [1]
    [Cycle 1]: 6.44e-05, [6]
        [build]: 3.27002e-06
        [elim_shapecalc]: 7.27002e-06
        [elim_not_effective]: 1.188e-05
        [opt_reshape]: 5.48002e-06
        [fold_const_symbol]: 8.48001e-06
        [renormalize]: 2.80008e-07
[validate]: 3.351e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00612671
[execute]: 7.65e-06
Sums
    bootstrap                            :     0.000250s :  1.76%
    type_inference                       :     0.004258s : 29.92%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000066s :  0.47%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.23%
    rewriter_before_opt_a                :     0.000083s :  0.58%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000059s :  0.42%
    jit_opt_a.loop_unroll                :     0.000037s :  0.26%
    jit_opt_a.a_1                        :     0.000609s :  4.28%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000679s :  4.77%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.14%
    jit_opt_a.cse                        :     0.000040s :  0.28%
    jit_opt_a.replace_applicator         :     0.000018s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000487s :  3.42%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000396s :  2.79%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.35%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000466s :  3.28%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006127s : 43.06%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000195    23
 0.95% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 3.17% :     0.000006s :      3: substitution.graph_param_transform
82.62% :     0.000161s :      4: substitution.inline
 1.49% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.28% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.36% :     0.000005s :      2: substitution.replace_old_param
 6.44% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004194     2
81.74% :     0.003428s :      1: type_inference.infer
18.26% :     0.000766s :      1: type_inference.specialize
------[replace.]   0.000067     6
66.10% :     0.000044s :      4: replace.inline
33.90% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000170     6
93.38% :     0.000159s :      4: match.inline
 6.62% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.57% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.12% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 1.88% :     0.000002s :     10: predicate.arithmetic_simplify
 1.13% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.49% :     0.000002s :     10: predicate.depend_value_elim
 1.20% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.11% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.41% :     0.000000s :      3: predicate.graph_param_transform
 5.66% :     0.000007s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.40% :     0.000002s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.62% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.73% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.77% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000002s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.64% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.15% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.81% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000527    12
50.38% :     0.000265s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.87% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.75% :     0.000230s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019646    76
 0.27% :     0.000052s :      1: add_recomputation
 0.35% :     0.000070s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.36% :     0.000268s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.26% :     0.005159s :      1: jit_opt_a
 0.73% :     0.000144s :      1: jit_opt_after_cconv
 0.26% :     0.000051s :      1: jit_opt_b
 2.06% :     0.000405s :      1: loop_unroll
 2.53% :     0.000496s :      1: mutable_eliminate
 4.18% :     0.000820s :     26: opt.transform.jit_opt_a
 0.24% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.41% :     0.000473s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.73% :     0.000340s :      1: renormalize.infer
 1.69% :     0.000332s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000036s :      1: rewriter_after_opt_a
 0.44% :     0.000086s :      1: rewriter_before_opt_a
 0.37% :     0.000074s :      1: symbol_engine_optimizer
31.24% :     0.006137s :      1: task_emit
21.75% :     0.004272s :      1: type_inference
 0.28% :     0.000054s :      1: validate


TotalTime = 0.0213583, [33]
[bootstrap]: 0.00035078
[type_inference]: 0.00516789
[event_method]: 1.782e-05
[auto_monad]: 6.968e-05
[graph_reusing]: 5.69e-06
[pre_auto_parallel]: 2.66999e-06
[py_interpret_to_execute]: 3.806e-05
[rewriter_before_opt_a]: 7.328e-05
[expand_dump_flag]: 4.04002e-06
[jit_opt_a]: 0.00677465, [2]
    [Cycle 1]: 0.00226464, [27]
        [switch_simplify]: 6.708e-05
        [loop_unroll]: 2.919e-05
        [a_1]: 0.00061136
        [with_stream_mark]: 2.617e-05
        [recompute_prepare]: 7.95998e-06
        [updatestate_depend_eliminate]: 4.67e-06
        [updatestate_assign_eliminate]: 3.58999e-06
        [updatestate_loads_eliminate]: 2.78e-06
        [parameter_eliminate]: 2.56e-06
        [specialize_transform]: 6.41e-06
        [updatestate_useless_node_eliminater]: 5.38002e-06
        [accelerated_algorithm]: 6.56999e-06
        [meta_shard_fg_expand]: 1.96998e-06
        [get_grad_eliminate_]: 6.14001e-06
        [merge_forward]: 4.43001e-06
        [cell_reuse_recompute_pass]: 1.93002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.707e-05
        [j_node_and_user_rematch]: 1.111e-05
        [meta_fg_expand]: 2.80002e-06
        [replace_old_param]: 1.18e-05
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.00112323
        [add_forward_monad_depend]: 1.122e-05
        [auto_monad_grad]: 2.97002e-06
        [auto_monad_eliminator]: 2.352e-05
        [cse]: 3.271e-05
        [replace_applicator]: 2.449e-05
    [Cycle 2]: 0.00039323, [27]
        [switch_simplify]: 6.94999e-06
        [loop_unroll]: 5.27001e-06
        [a_1]: 0.00010965
        [with_stream_mark]: 1.597e-05
        [recompute_prepare]: 6.71e-06
        [updatestate_depend_eliminate]: 3.65e-06
        [updatestate_assign_eliminate]: 3.36001e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 5.24998e-06
        [updatestate_useless_node_eliminater]: 5.54998e-06
        [accelerated_algorithm]: 5.92001e-06
        [meta_shard_fg_expand]: 2.46e-06
        [get_grad_eliminate_]: 5.15001e-06
        [merge_forward]: 3.95e-06
        [cell_reuse_recompute_pass]: 3.97e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.028e-05
        [j_node_and_user_rematch]: 9.17999e-06
        [meta_fg_expand]: 2.56e-06
        [replace_old_param]: 8.88002e-06
        [inline_without_move]: 4.95001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 2.07001e-06
        [auto_monad_grad]: 1.48002e-06
        [auto_monad_eliminator]: 6.24999e-06
        [cse]: 1.236e-05
        [replace_applicator]: 6.09999e-06
[py_interpret_to_execute_after_opt_a]: 1.735e-05
[rewriter_after_opt_a]: 4.135e-05
[convert_after_rewriter]: 7.7e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.00091718
[jit_opt_b]: 8.881e-05, [1]
    [Cycle 1]: 7.894e-05, [2]
        [frontend_op_eliminate]: 4.493e-05
        [inline_after_opt_a]: 1.755e-05
[cconv]: 3.804e-05
[loop_unroll]: 0.00046896
[jit_opt_after_cconv]: 0.00015512, [1]
    [Cycle 1]: 0.00014748, [11]
        [c_1]: 2.289e-05
        [parameter_eliminate]: 4.09002e-06
        [updatestate_depend_eliminate]: 7.77002e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [cse]: 2.528e-05
        [call_graph_tuple_transform]: 2.069e-05
        [tuple_list_get_item_eliminator]: 5.10001e-06
        [none_parameter_eliminate]: 1.76998e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 5.13002e-06
[remove_dup_value]: 1.618e-05
[partial_unused_args_eliminate]: 2.21998e-06
[environ_conv]: 5.72999e-06
[add_recomputation]: 5.288e-05
[cse_after_recomputation]: 2.178e-05, [1]
    [Cycle 1]: 1.59e-05, [1]
        [cse]: 9.49e-06
[auto_monad_reorder]: 1.591e-05
[get_jit_bprop_graph]: 2.54001e-06
[rewriter_after_jit_bprop_graph]: 7.01001e-06
[opt_after_jit_grad]: 0.0004667
[symbol_engine_optimizer]: 7.469e-05, [1]
    [Cycle 1]: 6.698e-05, [6]
        [build]: 3.64002e-06
        [elim_shapecalc]: 7.43999e-06
        [elim_not_effective]: 1.307e-05
        [opt_reshape]: 5.61998e-06
        [fold_const_symbol]: 8.99e-06
        [renormalize]: 9.09989e-07
[validate]: 4.038e-05
[backend_pass]: 1.47001e-06
[task_emit]: 0.00612988
[execute]: 7.05e-06
Sums
    bootstrap                            :     0.000351s :  2.13%
    type_inference                       :     0.005168s : 31.32%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000070s :  0.42%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.23%
    rewriter_before_opt_a                :     0.000073s :  0.44%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.45%
    jit_opt_a.loop_unroll                :     0.000034s :  0.21%
    jit_opt_a.a_1                        :     0.000721s :  4.37%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000021s :  0.13%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001123s :  6.81%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000030s :  0.18%
    jit_opt_a.cse                        :     0.000045s :  0.27%
    jit_opt_a.replace_applicator         :     0.000031s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.11%
    rewriter_after_opt_a                 :     0.000041s :  0.25%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000917s :  5.56%
    jit_opt_b.frontend_op_eliminate      :     0.000045s :  0.27%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.11%
    cconv                                :     0.000038s :  0.23%
    loop_unroll                          :     0.000469s :  2.84%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000053s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000467s :  2.83%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006130s : 37.15%
    execute                              :     0.000007s :  0.04%


Time group info:
------[substitution.]   0.000266    23
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 2.32% :     0.000006s :      3: substitution.graph_param_transform
84.38% :     0.000225s :      4: substitution.inline
 1.95% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.24% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.66% :     0.000007s :      2: substitution.replace_old_param
 5.16% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005095     2
81.61% :     0.004158s :      1: type_inference.infer
18.39% :     0.000937s :      1: type_inference.specialize
------[replace.]   0.000079     6
65.87% :     0.000052s :      4: replace.inline
34.13% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000234     6
94.64% :     0.000222s :      4: match.inline
 5.36% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 2.18% :     0.000003s :     10: predicate.arithmetic_simplify
 1.37% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.26% :     0.000002s :     10: predicate.compare_switch_simplify
 1.14% :     0.000001s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.99% :     0.000003s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.18% :     0.000002s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.83% :     0.000007s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.66% :     0.000002s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.28% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.51% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.76% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 0.95% :     0.000001s :     10: predicate.minmaximum_grad
 2.16% :     0.000003s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 1.91% :     0.000002s :     15: predicate.partial_eliminate
 1.29% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     10: predicate.reshape_eliminate
 1.08% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.08% :     0.000001s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 0.86% :     0.000001s :      6: predicate.specialize_transform
 1.53% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.63% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.21% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.94% :     0.000009s :     53: predicate.switch_simplify
 1.46% :     0.000002s :     10: predicate.tile_eliminate
 1.45% :     0.000002s :     10: predicate.transpose_eliminate
 1.71% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.28% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.66% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000718    12
42.78% :     0.000307s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.82% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.40% :     0.000355s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023591    76
 0.24% :     0.000056s :      1: add_recomputation
 0.31% :     0.000073s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.55% :     0.000366s :      1: bootstrap
 0.17% :     0.000041s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
28.73% :     0.006779s :      1: jit_opt_a
 0.67% :     0.000158s :      1: jit_opt_after_cconv
 0.39% :     0.000091s :      1: jit_opt_b
 2.02% :     0.000477s :      1: loop_unroll
 3.94% :     0.000929s :      1: mutable_eliminate
 4.13% :     0.000974s :     26: opt.transform.jit_opt_a
 0.21% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.23% :     0.000055s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.01% :     0.000475s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000041s :      1: py_interpret_to_execute
 0.08% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000019s :      1: remove_dup_value
 2.52% :     0.000595s :      1: renormalize.infer
 2.19% :     0.000516s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000044s :      1: rewriter_after_opt_a
 0.32% :     0.000076s :      1: rewriter_before_opt_a
 0.33% :     0.000077s :      1: symbol_engine_optimizer
26.04% :     0.006142s :      1: task_emit
22.22% :     0.005242s :      1: type_inference
 0.27% :     0.000063s :      1: validate


TotalTime = 0.0193373, [33]
[bootstrap]: 0.00030293
[type_inference]: 0.00470043
[event_method]: 1.809e-05
[auto_monad]: 6.409e-05
[graph_reusing]: 5.62999e-06
[pre_auto_parallel]: 2.36e-06
[py_interpret_to_execute]: 3.119e-05
[rewriter_before_opt_a]: 7.176e-05
[expand_dump_flag]: 3.06001e-06
[jit_opt_a]: 0.00516039, [2]
    [Cycle 1]: 0.00153574, [27]
        [switch_simplify]: 5.455e-05
        [loop_unroll]: 2.935e-05
        [a_1]: 0.00051734
        [with_stream_mark]: 1.908e-05
        [recompute_prepare]: 9.07001e-06
        [updatestate_depend_eliminate]: 3.95e-06
        [updatestate_assign_eliminate]: 2.87002e-06
        [updatestate_loads_eliminate]: 3.04001e-06
        [parameter_eliminate]: 2.17999e-06
        [specialize_transform]: 7.53e-06
        [updatestate_useless_node_eliminater]: 5.10001e-06
        [accelerated_algorithm]: 5.45001e-06
        [meta_shard_fg_expand]: 2.34001e-06
        [get_grad_eliminate_]: 5.48002e-06
        [merge_forward]: 3.92002e-06
        [cell_reuse_recompute_pass]: 2.26e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.607e-05
        [j_node_and_user_rematch]: 9.34e-06
        [meta_fg_expand]: 2.61e-06
        [replace_old_param]: 9.00999e-06
        [inline_without_move]: 6.24999e-06
        [renormalize]: 0.00057149
        [add_forward_monad_depend]: 5.14998e-06
        [auto_monad_grad]: 2.43e-06
        [auto_monad_eliminator]: 1.358e-05
        [cse]: 2.674e-05
        [replace_applicator]: 1.24e-05
    [Cycle 2]: 0.00034803, [27]
        [switch_simplify]: 5.69e-06
        [loop_unroll]: 4.84e-06
        [a_1]: 8.96e-05
        [with_stream_mark]: 1.001e-05
        [recompute_prepare]: 5.25001e-06
        [updatestate_depend_eliminate]: 2.59999e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 5.04e-06
        [updatestate_useless_node_eliminater]: 4.74998e-06
        [accelerated_algorithm]: 4.92e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.79e-06
        [merge_forward]: 2.87002e-06
        [cell_reuse_recompute_pass]: 2.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.487e-05
        [j_node_and_user_rematch]: 9.04e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 7.42002e-06
        [inline_without_move]: 5.20999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.57999e-06
        [auto_monad_grad]: 1.07e-06
        [auto_monad_eliminator]: 6.74999e-06
        [cse]: 1.347e-05
        [replace_applicator]: 6.41e-06
[py_interpret_to_execute_after_opt_a]: 1.106e-05
[rewriter_after_opt_a]: 3.505e-05
[convert_after_rewriter]: 8.75999e-06
[order_py_execute_after_rewriter]: 6.71e-06
[mutable_eliminate]: 0.00054162
[jit_opt_b]: 5.109e-05, [1]
    [Cycle 1]: 4.355e-05, [2]
        [frontend_op_eliminate]: 1.659e-05
        [inline_after_opt_a]: 1.461e-05
[cconv]: 2.391e-05
[loop_unroll]: 0.00048224
[jit_opt_after_cconv]: 0.0001472, [1]
    [Cycle 1]: 0.00014049, [11]
        [c_1]: 2.056e-05
        [parameter_eliminate]: 3.13998e-06
        [updatestate_depend_eliminate]: 6.26e-06
        [updatestate_assign_eliminate]: 2.69999e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [cse]: 2.005e-05
        [call_graph_tuple_transform]: 1.808e-05
        [tuple_list_get_item_eliminator]: 5.61e-06
        [none_parameter_eliminate]: 1.40999e-06
        [renormalize]: 7.00005e-07
        [switch_simplify]: 5.89e-06
[remove_dup_value]: 1.343e-05
[partial_unused_args_eliminate]: 2.29001e-06
[environ_conv]: 4.97e-06
[add_recomputation]: 5.18e-05
[cse_after_recomputation]: 2.474e-05, [1]
    [Cycle 1]: 1.897e-05, [1]
        [cse]: 1.199e-05
[auto_monad_reorder]: 1.604e-05
[get_jit_bprop_graph]: 1.66e-06
[rewriter_after_jit_bprop_graph]: 6.60997e-06
[opt_after_jit_grad]: 0.0005323
[symbol_engine_optimizer]: 7.249e-05, [1]
    [Cycle 1]: 6.584e-05, [6]
        [build]: 4.05998e-06
        [elim_shapecalc]: 7.83001e-06
        [elim_not_effective]: 1.198e-05
        [opt_reshape]: 5.34e-06
        [fold_const_symbol]: 8.70999e-06
        [renormalize]: 6.09987e-07
[validate]: 3.594e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.00669327
[execute]: 8.48001e-06
Sums
    bootstrap                            :     0.000303s :  1.96%
    type_inference                       :     0.004700s : 30.48%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000064s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.20%
    rewriter_before_opt_a                :     0.000072s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000060s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000607s :  3.94%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000041s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000572s :  3.71%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000040s :  0.26%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.23%
    convert_after_rewriter               :     0.000009s :  0.06%
    order_py_execute_after_rewriter      :     0.000007s :  0.04%
    mutable_eliminate                    :     0.000542s :  3.51%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000024s :  0.16%
    loop_unroll                          :     0.000482s :  3.13%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000052s :  0.34%
    cse_after_recomputation.cse          :     0.000012s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000532s :  3.45%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006693s : 43.41%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000187    23
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.75% :     0.000005s :      3: substitution.graph_param_transform
81.52% :     0.000153s :      4: substitution.inline
 1.92% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.76% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.41% :     0.000005s :      2: substitution.replace_old_param
 6.97% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004635     2
82.42% :     0.003821s :      1: type_inference.infer
17.58% :     0.000815s :      1: type_inference.specialize
------[replace.]   0.000067     6
64.95% :     0.000043s :      4: replace.inline
35.05% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     6
92.87% :     0.000150s :      4: match.inline
 7.13% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.23% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 2.53% :     0.000003s :     10: predicate.arithmetic_simplify
 1.51% :     0.000002s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.62% :     0.000002s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.48% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.71% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.83% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 5.23% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000001s :      6: predicate.less_batch_normalization
 1.37% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.75% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.87% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.23% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.13% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.85% :     0.000003s :     16: predicate.switch_defer_inline
 2.23% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.48% :     0.000010s :     53: predicate.switch_simplify
 1.08% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.36% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.85% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000568    12
52.25% :     0.000297s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.44% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.31% :     0.000246s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020856    76
 0.27% :     0.000056s :      1: add_recomputation
 0.32% :     0.000068s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.55% :     0.000324s :      1: bootstrap
 0.13% :     0.000027s :      1: cconv
 0.06% :     0.000012s :      1: convert_after_rewriter
 0.13% :     0.000027s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
24.76% :     0.005163s :      1: jit_opt_a
 0.72% :     0.000150s :      1: jit_opt_after_cconv
 0.26% :     0.000054s :      1: jit_opt_b
 2.36% :     0.000493s :      1: loop_unroll
 2.65% :     0.000552s :      1: mutable_eliminate
 4.01% :     0.000836s :     26: opt.transform.jit_opt_a
 0.22% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.58% :     0.000539s :      1: opt_after_jit_grad
 0.04% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000034s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.42% :     0.000296s :      1: renormalize.infer
 1.28% :     0.000268s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000039s :      1: rewriter_after_opt_a
 0.36% :     0.000075s :      1: rewriter_before_opt_a
 0.36% :     0.000075s :      1: symbol_engine_optimizer
32.15% :     0.006705s :      1: task_emit
22.61% :     0.004716s :      1: type_inference
 0.28% :     0.000058s :      1: validate


TotalTime = 0.0167671, [30]
[bootstrap]: 0.00027877
[type_inference]: 0.00690103
[event_method]: 2.185e-05
[auto_monad]: 8.456e-05
[graph_reusing]: 5.82999e-06
[pre_auto_parallel]: 2.24001e-06
[py_interpret_to_execute]: 8.31e-05
[rewriter_before_opt_a]: 8.864e-05
[expand_dump_flag]: 3.56999e-06
[jit_opt_a]: 0.00651314, [2]
    [Cycle 1]: 0.00243885, [27]
        [switch_simplify]: 0.00012717
        [loop_unroll]: 3.549e-05
        [a_1]: 0.00081267
        [with_stream_mark]: 2.165e-05
        [recompute_prepare]: 1.012e-05
        [updatestate_depend_eliminate]: 3.81999e-06
        [updatestate_assign_eliminate]: 2.88e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [parameter_eliminate]: 2.70002e-06
        [specialize_transform]: 8.21002e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 6.53e-06
        [meta_shard_fg_expand]: 2.59999e-06
        [get_grad_eliminate_]: 6.02001e-06
        [merge_forward]: 3.93999e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.694e-05
        [j_node_and_user_rematch]: 1.246e-05
        [meta_fg_expand]: 4.39002e-06
        [replace_old_param]: 9.40001e-06
        [inline_without_move]: 6.24001e-06
        [renormalize]: 0.00107514
        [add_forward_monad_depend]: 5.99e-06
        [auto_monad_grad]: 2.59999e-06
        [auto_monad_eliminator]: 1.493e-05
        [cse]: 2.169e-05
        [replace_applicator]: 1.583e-05
    [Cycle 2]: 0.00038005, [27]
        [switch_simplify]: 5.86e-06
        [loop_unroll]: 5.56e-06
        [a_1]: 0.00010191
        [with_stream_mark]: 1.281e-05
        [recompute_prepare]: 5.46e-06
        [updatestate_depend_eliminate]: 2.86999e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [parameter_eliminate]: 1.32999e-06
        [specialize_transform]: 5.37001e-06
        [updatestate_useless_node_eliminater]: 6.06998e-06
        [accelerated_algorithm]: 6.21e-06
        [meta_shard_fg_expand]: 1.60001e-06
        [get_grad_eliminate_]: 4.95001e-06
        [merge_forward]: 2.94001e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.695e-05
        [j_node_and_user_rematch]: 1.002e-05
        [meta_fg_expand]: 2.56e-06
        [replace_old_param]: 9.95002e-06
        [inline_without_move]: 6.16e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.89e-06
        [auto_monad_grad]: 1.21002e-06
        [auto_monad_eliminator]: 6.14999e-06
        [cse]: 1.117e-05
        [replace_applicator]: 9.32001e-06
[py_interpret_to_execute_after_opt_a]: 1.237e-05
[rewriter_after_opt_a]: 3.847e-05
[convert_after_rewriter]: 9.46e-06
[order_py_execute_after_rewriter]: 5.81e-06
[mutable_eliminate]: 0.00077891
[jit_opt_b]: 6.685e-05, [1]
    [Cycle 1]: 5.692e-05, [2]
        [frontend_op_eliminate]: 2.056e-05
        [inline_after_opt_a]: 2.055e-05
[cconv]: 3.01e-05
[loop_unroll]: 0.00059728
[jit_opt_after_cconv]: 0.00018317, [1]
    [Cycle 1]: 0.00017452, [11]
        [c_1]: 2.583e-05
        [parameter_eliminate]: 5.24e-06
        [updatestate_depend_eliminate]: 8.05999e-06
        [updatestate_assign_eliminate]: 3.41001e-06
        [updatestate_loads_eliminate]: 2.86e-06
        [cse]: 2.689e-05
        [call_graph_tuple_transform]: 2.632e-05
        [tuple_list_get_item_eliminator]: 6.07999e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 1.29e-06
        [switch_simplify]: 6.07999e-06
[remove_dup_value]: 1.432e-05
[partial_unused_args_eliminate]: 2.62001e-06
[environ_conv]: 6.36998e-06
[add_recomputation]: 5.547e-05
[cse_after_recomputation]: 2.948e-05, [1]
    [Cycle 1]: 2.192e-05, [1]
        [cse]: 1.106e-05
[auto_monad_reorder]: 1.534e-05
[get_jit_bprop_graph]: 2.86e-06
[rewriter_after_jit_bprop_graph]: 7.23999e-06
[opt_after_jit_grad]: 0.0006001
[symbol_engine_optimizer]: 8.478e-05, [1]
    [Cycle 1]: 7.703e-05, [6]
        [build]: 3.59002e-06
        [elim_shapecalc]: 8.96998e-06
        [elim_not_effective]: 1.469e-05
        [opt_reshape]: 7.46001e-06
        [fold_const_symbol]: 1.036e-05
        [renormalize]: 8.29983e-07
[validate]: 3.972e-05
Sums
    bootstrap                            :     0.000279s :  2.25%
    type_inference                       :     0.006901s : 55.74%
    event_method                         :     0.000022s :  0.18%
    auto_monad                           :     0.000085s :  0.68%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000083s :  0.67%
    rewriter_before_opt_a                :     0.000089s :  0.72%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000133s :  1.07%
    jit_opt_a.loop_unroll                :     0.000041s :  0.33%
    jit_opt_a.a_1                        :     0.000915s :  7.39%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000014s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000022s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.06%
    jit_opt_a.replace_old_param          :     0.000019s :  0.16%
    jit_opt_a.inline_without_move        :     0.000012s :  0.10%
    jit_opt_a.renormalize                :     0.001075s :  8.68%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.17%
    jit_opt_a.cse                        :     0.000033s :  0.27%
    jit_opt_a.replace_applicator         :     0.000025s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.10%
    rewriter_after_opt_a                 :     0.000038s :  0.31%
    convert_after_rewriter               :     0.000009s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000779s :  6.29%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000021s :  0.17%
    cconv                                :     0.000030s :  0.24%
    loop_unroll                          :     0.000597s :  4.82%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000027s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000014s :  0.12%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000055s :  0.45%
    cse_after_recomputation.cse          :     0.000011s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.12%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000600s :  4.85%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.32%


Time group info:
------[substitution.]   0.000339    26
 0.53% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000002s :      2: substitution.fold_const_symbol
 2.13% :     0.000007s :      3: substitution.graph_param_transform
85.75% :     0.000291s :      6: substitution.inline
 1.39% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.44% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.26% :     0.000004s :      2: substitution.replace_old_param
 2.62% :     0.000009s :      1: substitution.switch_simplify
 4.24% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006833     2
81.52% :     0.005571s :      1: type_inference.infer
18.48% :     0.001263s :      1: type_inference.specialize
------[replace.]   0.000179     9
48.70% :     0.000087s :      6: replace.inline
22.59% :     0.000040s :      1: replace.switch_simplify
28.71% :     0.000051s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000308     9
93.16% :     0.000287s :      6: match.inline
 2.66% :     0.000008s :      1: match.switch_simplify
 4.18% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000152   801
 1.23% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.71% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     12: predicate.addn_check_dump
 1.30% :     0.000002s :     12: predicate.addn_zero_filter
 2.12% :     0.000003s :     12: predicate.arithmetic_simplify
 1.06% :     0.000002s :     12: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000002s :     12: predicate.compare_switch_simplify
 1.22% :     0.000002s :     12: predicate.depend_value_elim
 1.04% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.03% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.02% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.24% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.05% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.02% :     0.000002s :     12: predicate.environ_get_eliminate
 0.97% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 4.61% :     0.000007s :     26: predicate.inline
 0.76% :     0.000001s :      6: predicate.inline_without_move
 0.29% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.69% :     0.000003s :      6: predicate.less_batch_normalization
 1.75% :     0.000003s :     14: predicate.list_to_tuple_eliminator_
 1.53% :     0.000002s :     17: predicate.load_eliminater
 2.48% :     0.000004s :      3: predicate.loop_unroll_after_grad
 4.78% :     0.000007s :     40: predicate.loop_unroll_before_grad
 1.99% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000002s :     12: predicate.merge_addn
 0.97% :     0.000001s :     12: predicate.minmaximum_grad
 2.22% :     0.000003s :      3: predicate.mutable_eliminate
 0.63% :     0.000001s :      3: predicate.opt_reshape
 1.88% :     0.000003s :     17: predicate.partial_eliminate
 1.11% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.66% :     0.000003s :     12: predicate.reduce_eliminate
 2.00% :     0.000003s :     14: predicate.redundant_stop_gradient_eliminater
 0.39% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.45% :     0.000002s :     20: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 2.11% :     0.000003s :     12: predicate.reshape_eliminate
 1.13% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000002s :     12: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.68% :     0.000001s :      6: predicate.special_op_eliminate
 0.74% :     0.000001s :      6: predicate.specialize_transform
 1.22% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.08% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.30% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.56% :     0.000005s :     20: predicate.switch_defer_inline
 2.03% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.22% :     0.000012s :     65: predicate.switch_simplify
 1.09% :     0.000002s :     12: predicate.tile_eliminate
 1.45% :     0.000002s :     12: predicate.transpose_eliminate
 1.75% :     0.000003s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.03% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.52% :     0.000007s :     20: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.95% :     0.000003s :     14: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.53% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     12: predicate.value_based_eliminate
 0.34% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.53% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000979    15
43.96% :     0.000430s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.41% :     0.000121s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.63% :     0.000427s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019159    72
 0.32% :     0.000061s :      1: add_recomputation
 0.46% :     0.000089s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 1.51% :     0.000289s :      1: bootstrap
 0.18% :     0.000034s :      1: cconv
 0.07% :     0.000013s :      1: convert_after_rewriter
 0.17% :     0.000032s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.13% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000006s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
34.02% :     0.006517s :      1: jit_opt_a
 0.98% :     0.000187s :      1: jit_opt_after_cconv
 0.37% :     0.000070s :      1: jit_opt_b
 3.19% :     0.000612s :      1: loop_unroll
 4.13% :     0.000792s :      1: mutable_eliminate
 6.43% :     0.001232s :     26: opt.transform.jit_opt_a
 0.31% :     0.000060s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000033s :      4: opt.transform.jit_opt_b
 0.09% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000038s :      4: opt.transform.symbol_engine_opt
 3.21% :     0.000615s :      1: opt_after_jit_grad
 0.05% :     0.000009s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.45% :     0.000086s :      1: py_interpret_to_execute
 0.08% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000017s :      1: remove_dup_value
 3.30% :     0.000633s :      1: renormalize.infer
 2.25% :     0.000431s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000043s :      1: rewriter_after_opt_a
 0.48% :     0.000093s :      1: rewriter_before_opt_a
 0.46% :     0.000088s :      1: symbol_engine_optimizer
36.09% :     0.006914s :      1: type_inference


TotalTime = 0.027021, [33]
[bootstrap]: 0.0003494
[type_inference]: 0.00961666
[event_method]: 1.888e-05
[auto_monad]: 7.008e-05
[graph_reusing]: 6.24999e-06
[pre_auto_parallel]: 2.74001e-06
[py_interpret_to_execute]: 5.083e-05
[rewriter_before_opt_a]: 7.919e-05
[expand_dump_flag]: 3.04001e-06
[jit_opt_a]: 0.00739461, [2]
    [Cycle 1]: 0.00301541, [27]
        [switch_simplify]: 6.689e-05
        [loop_unroll]: 2.94e-05
        [a_1]: 0.00155552
        [with_stream_mark]: 2.579e-05
        [recompute_prepare]: 9.78002e-06
        [updatestate_depend_eliminate]: 4.17998e-06
        [updatestate_assign_eliminate]: 3.53999e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 2.51e-06
        [specialize_transform]: 5.94e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 6.96001e-06
        [meta_shard_fg_expand]: 2.69001e-06
        [get_grad_eliminate_]: 5.44998e-06
        [merge_forward]: 4.18999e-06
        [cell_reuse_recompute_pass]: 1.50001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.537e-05
        [j_node_and_user_rematch]: 9.94001e-06
        [meta_fg_expand]: 2.33002e-06
        [replace_old_param]: 1.019e-05
        [inline_without_move]: 5.54e-06
        [renormalize]: 0.00097169
        [add_forward_monad_depend]: 7.42002e-06
        [auto_monad_grad]: 2.91e-06
        [auto_monad_eliminator]: 1.498e-05
        [cse]: 2.927e-05
        [replace_applicator]: 1.447e-05
    [Cycle 2]: 0.00032913, [27]
        [switch_simplify]: 5.71e-06
        [loop_unroll]: 5.46e-06
        [a_1]: 9.293e-05
        [with_stream_mark]: 9.89001e-06
        [recompute_prepare]: 5.25001e-06
        [updatestate_depend_eliminate]: 2.59999e-06
        [updatestate_assign_eliminate]: 1.92999e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.21002e-06
        [specialize_transform]: 5.01002e-06
        [updatestate_useless_node_eliminater]: 4.99998e-06
        [accelerated_algorithm]: 5.01002e-06
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.54998e-06
        [merge_forward]: 2.98998e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.435e-05
        [j_node_and_user_rematch]: 8.37e-06
        [meta_fg_expand]: 1.45999e-06
        [replace_old_param]: 1.485e-05
        [inline_without_move]: 4.63999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.80001e-06
        [auto_monad_grad]: 8.90024e-07
        [auto_monad_eliminator]: 4.74e-06
        [cse]: 1.005e-05
        [replace_applicator]: 4.87e-06
[py_interpret_to_execute_after_opt_a]: 1.179e-05
[rewriter_after_opt_a]: 3.68e-05
[convert_after_rewriter]: 7.11999e-06
[order_py_execute_after_rewriter]: 7.43e-06
[mutable_eliminate]: 0.00077942
[jit_opt_b]: 5.148e-05, [1]
    [Cycle 1]: 4.384e-05, [2]
        [frontend_op_eliminate]: 1.686e-05
        [inline_after_opt_a]: 1.505e-05
[cconv]: 2.824e-05
[loop_unroll]: 0.0004411
[jit_opt_after_cconv]: 0.00014049, [1]
    [Cycle 1]: 0.00013426, [11]
        [c_1]: 2.063e-05
        [parameter_eliminate]: 2.53e-06
        [updatestate_depend_eliminate]: 5.64998e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.985e-05
        [call_graph_tuple_transform]: 1.923e-05
        [tuple_list_get_item_eliminator]: 5.07e-06
        [none_parameter_eliminate]: 1.63002e-06
        [renormalize]: 7.89994e-07
        [switch_simplify]: 5.15001e-06
[remove_dup_value]: 1.45e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 5.36002e-06
[add_recomputation]: 5.27e-05
[cse_after_recomputation]: 2.076e-05, [1]
    [Cycle 1]: 1.507e-05, [1]
        [cse]: 9.49e-06
[auto_monad_reorder]: 1.689e-05
[get_jit_bprop_graph]: 2.23002e-06
[rewriter_after_jit_bprop_graph]: 6.36e-06
[opt_after_jit_grad]: 0.00047146
[symbol_engine_optimizer]: 7.533e-05, [1]
    [Cycle 1]: 6.825e-05, [6]
        [build]: 3.90998e-06
        [elim_shapecalc]: 7.7e-06
        [elim_not_effective]: 1.268e-05
        [opt_reshape]: 5.64998e-06
        [fold_const_symbol]: 9.31002e-06
        [renormalize]: 1.12e-06
[validate]: 4.102e-05
[backend_pass]: 1.10999e-06
[task_emit]: 0.00698265
[execute]: 8.67e-06
Sums
    bootstrap                            :     0.000349s :  1.57%
    type_inference                       :     0.009617s : 43.09%
    event_method                         :     0.000019s :  0.08%
    auto_monad                           :     0.000070s :  0.31%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000051s :  0.23%
    rewriter_before_opt_a                :     0.000079s :  0.35%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000073s :  0.33%
    jit_opt_a.loop_unroll                :     0.000035s :  0.16%
    jit_opt_a.a_1                        :     0.001648s :  7.39%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000025s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000972s :  4.35%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.09%
    jit_opt_a.cse                        :     0.000039s :  0.18%
    jit_opt_a.replace_applicator         :     0.000019s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.05%
    rewriter_after_opt_a                 :     0.000037s :  0.16%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.03%
    mutable_eliminate                    :     0.000779s :  3.49%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.07%
    cconv                                :     0.000028s :  0.13%
    loop_unroll                          :     0.000441s :  1.98%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000015s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000053s :  0.24%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000017s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000471s :  2.11%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.18%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006983s : 31.29%
    execute                              :     0.000009s :  0.04%


Time group info:
------[substitution.]   0.000290    23
 0.68% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 2.10% :     0.000006s :      3: substitution.graph_param_transform
86.79% :     0.000251s :      4: substitution.inline
 1.42% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.66% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.70% :     0.000005s :      2: substitution.replace_old_param
 5.20% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009523     2
41.89% :     0.003989s :      1: type_inference.infer
58.11% :     0.005534s :      1: type_inference.specialize
------[replace.]   0.000965     6
97.05% :     0.000937s :      4: replace.inline
 2.95% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000263     6
94.79% :     0.000249s :      4: match.inline
 5.21% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.76% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.23% :     0.000002s :     10: predicate.addn_zero_filter
 2.47% :     0.000003s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.17% :     0.000001s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.39% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.55% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.12% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.63% :     0.000007s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.55% :     0.000002s :      6: predicate.less_batch_normalization
 1.66% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.72% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.08% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.91% :     0.000002s :      3: predicate.mutable_eliminate
 0.58% :     0.000001s :      3: predicate.opt_reshape
 1.92% :     0.000002s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.72% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.01% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.75% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.10% :     0.000004s :     16: predicate.switch_defer_inline
 1.94% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.44% :     0.000009s :     53: predicate.switch_simplify
 1.29% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.53% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 2.26% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.89% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000687    12
42.60% :     0.000293s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.64% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.76% :     0.000342s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.029971    76
 0.19% :     0.000055s :      1: add_recomputation
 0.25% :     0.000074s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.23% :     0.000368s :      1: bootstrap
 0.10% :     0.000031s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000024s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
24.69% :     0.007398s :      1: jit_opt_a
 0.48% :     0.000143s :      1: jit_opt_after_cconv
 0.18% :     0.000054s :      1: jit_opt_b
 1.50% :     0.000449s :      1: loop_unroll
 2.63% :     0.000788s :      1: mutable_eliminate
 6.29% :     0.001884s :     26: opt.transform.jit_opt_a
 0.16% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000026s :      4: opt.transform.jit_opt_b
 0.04% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.60% :     0.000479s :      1: opt_after_jit_grad
 0.03% :     0.000010s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000054s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 1.82% :     0.000547s :      1: renormalize.infer
 1.38% :     0.000415s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000040s :      1: rewriter_after_opt_a
 0.28% :     0.000082s :      1: rewriter_before_opt_a
 0.26% :     0.000078s :      1: symbol_engine_optimizer
23.36% :     0.007001s :      1: task_emit
32.16% :     0.009638s :      1: type_inference
 0.21% :     0.000063s :      1: validate


TotalTime = 0.0186465, [33]
[bootstrap]: 0.00031021
[type_inference]: 0.00485301
[event_method]: 3.989e-05
[auto_monad]: 6.613e-05
[graph_reusing]: 6.35002e-06
[pre_auto_parallel]: 2.26e-06
[py_interpret_to_execute]: 3.305e-05
[rewriter_before_opt_a]: 9.091e-05
[expand_dump_flag]: 3.5e-06
[jit_opt_a]: 0.00503512, [2]
    [Cycle 1]: 0.00149167, [27]
        [switch_simplify]: 7.129e-05
        [loop_unroll]: 4.448e-05
        [a_1]: 0.00052596
        [with_stream_mark]: 1.6e-05
        [recompute_prepare]: 6.94999e-06
        [updatestate_depend_eliminate]: 3.33e-06
        [updatestate_assign_eliminate]: 3.16001e-06
        [updatestate_loads_eliminate]: 2.55002e-06
        [parameter_eliminate]: 1.87001e-06
        [specialize_transform]: 6.05002e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.15001e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 4.96002e-06
        [merge_forward]: 3.63e-06
        [cell_reuse_recompute_pass]: 1.11002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.393e-05
        [j_node_and_user_rematch]: 8.74998e-06
        [meta_fg_expand]: 2.56998e-06
        [replace_old_param]: 9.00001e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.00052733
        [add_forward_monad_depend]: 5.14e-06
        [auto_monad_grad]: 2.01e-06
        [auto_monad_eliminator]: 1.315e-05
        [cse]: 2.563e-05
        [replace_applicator]: 1.164e-05
    [Cycle 2]: 0.00033194, [27]
        [switch_simplify]: 5.93002e-06
        [loop_unroll]: 4.89e-06
        [a_1]: 9.09e-05
        [with_stream_mark]: 8.67e-06
        [recompute_prepare]: 4.87e-06
        [updatestate_depend_eliminate]: 2.61999e-06
        [updatestate_assign_eliminate]: 2.03997e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 5.12e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 5.32001e-06
        [meta_shard_fg_expand]: 1.02998e-06
        [get_grad_eliminate_]: 4.65999e-06
        [merge_forward]: 2.58e-06
        [cell_reuse_recompute_pass]: 1.94e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.216e-05
        [j_node_and_user_rematch]: 7.77e-06
        [meta_fg_expand]: 1.70001e-06
        [replace_old_param]: 7e-06
        [inline_without_move]: 4.67e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 4.65999e-06
        [cse]: 2.139e-05
        [replace_applicator]: 5.96e-06
[py_interpret_to_execute_after_opt_a]: 8.33001e-06
[rewriter_after_opt_a]: 3.387e-05
[convert_after_rewriter]: 8.32e-06
[order_py_execute_after_rewriter]: 5.20001e-06
[mutable_eliminate]: 0.00051133
[jit_opt_b]: 5.907e-05, [1]
    [Cycle 1]: 5.097e-05, [2]
        [frontend_op_eliminate]: 2.146e-05
        [inline_after_opt_a]: 1.688e-05
[cconv]: 2.41e-05
[loop_unroll]: 0.00048265
[jit_opt_after_cconv]: 0.0001458, [1]
    [Cycle 1]: 0.00013859, [11]
        [c_1]: 2.202e-05
        [parameter_eliminate]: 2.73998e-06
        [updatestate_depend_eliminate]: 5.45001e-06
        [updatestate_assign_eliminate]: 2.36998e-06
        [updatestate_loads_eliminate]: 2.31998e-06
        [cse]: 1.727e-05
        [call_graph_tuple_transform]: 1.972e-05
        [tuple_list_get_item_eliminator]: 5.78002e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 4.60015e-07
        [switch_simplify]: 5.74999e-06
[remove_dup_value]: 1.303e-05
[partial_unused_args_eliminate]: 2.32001e-06
[environ_conv]: 4.92999e-06
[add_recomputation]: 4.862e-05
[cse_after_recomputation]: 2.141e-05, [1]
    [Cycle 1]: 1.577e-05, [1]
        [cse]: 9.86998e-06
[auto_monad_reorder]: 1.593e-05
[get_jit_bprop_graph]: 1.59e-06
[rewriter_after_jit_bprop_graph]: 5.50001e-06
[opt_after_jit_grad]: 0.00052597
[symbol_engine_optimizer]: 7.767e-05, [1]
    [Cycle 1]: 7.089e-05, [6]
        [build]: 3.43e-06
        [elim_shapecalc]: 8.40001e-06
        [elim_not_effective]: 1.37e-05
        [opt_reshape]: 5.94e-06
        [fold_const_symbol]: 9.32001e-06
        [renormalize]: 5.50004e-07
[validate]: 3.133e-05
[backend_pass]: 8.59989e-07
[task_emit]: 0.00595915
[execute]: 7.08998e-06
Sums
    bootstrap                            :     0.000310s :  2.09%
    type_inference                       :     0.004853s : 32.76%
    event_method                         :     0.000040s :  0.27%
    auto_monad                           :     0.000066s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.22%
    rewriter_before_opt_a                :     0.000091s :  0.61%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000077s :  0.52%
    jit_opt_a.loop_unroll                :     0.000049s :  0.33%
    jit_opt_a.a_1                        :     0.000617s :  4.16%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000527s :  3.56%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000047s :  0.32%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000034s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000511s :  3.45%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.11%
    cconv                                :     0.000024s :  0.16%
    loop_unroll                          :     0.000483s :  3.26%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000016s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000526s :  3.55%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000031s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005959s : 40.23%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000201    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.69% :     0.000005s :      3: substitution.graph_param_transform
84.10% :     0.000169s :      4: substitution.inline
 1.42% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.13% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.78% :     0.000004s :      2: substitution.replace_old_param
 6.15% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004786     2
82.62% :     0.003954s :      1: type_inference.infer
17.38% :     0.000832s :      1: type_inference.specialize
------[replace.]   0.000066     6
68.06% :     0.000045s :      4: replace.inline
31.94% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000177     6
93.81% :     0.000166s :      4: match.inline
 6.19% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.38% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 1.84% :     0.000002s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.25% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.12% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000000s :      3: predicate.elim_not_effective
 0.75% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.60% :     0.000007s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.80% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.02% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 2.15% :     0.000003s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.93% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.81% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.58% :     0.000009s :     53: predicate.switch_simplify
 1.31% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.71% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000553    12
48.97% :     0.000271s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.21% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.82% :     0.000253s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020154    76
 0.26% :     0.000052s :      1: add_recomputation
 0.34% :     0.000069s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.63% :     0.000329s :      1: bootstrap
 0.13% :     0.000027s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.22% :     0.000045s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.13% :     0.000027s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.00% :     0.005038s :      1: jit_opt_a
 0.74% :     0.000149s :      1: jit_opt_after_cconv
 0.31% :     0.000062s :      1: jit_opt_b
 2.43% :     0.000491s :      1: loop_unroll
 2.58% :     0.000520s :      1: mutable_eliminate
 4.25% :     0.000856s :     26: opt.transform.jit_opt_a
 0.25% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000031s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.66% :     0.000535s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000036s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.33% :     0.000268s :      1: renormalize.infer
 1.25% :     0.000253s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000037s :      1: rewriter_after_opt_a
 0.47% :     0.000094s :      1: rewriter_before_opt_a
 0.40% :     0.000081s :      1: symbol_engine_optimizer
29.62% :     0.005969s :      1: task_emit
24.14% :     0.004865s :      1: type_inference
 0.24% :     0.000049s :      1: validate


TotalTime = 0.0140361, [30]
[bootstrap]: 0.00025163
[type_inference]: 0.00595042
[event_method]: 1.975e-05
[auto_monad]: 7.263e-05
[graph_reusing]: 5.07999e-06
[pre_auto_parallel]: 1.97001e-06
[py_interpret_to_execute]: 4.225e-05
[rewriter_before_opt_a]: 8.146e-05
[expand_dump_flag]: 2.89999e-06
[jit_opt_a]: 0.00539791, [2]
    [Cycle 1]: 0.00206828, [27]
        [switch_simplify]: 0.00010774
        [loop_unroll]: 3.417e-05
        [a_1]: 0.00069992
        [with_stream_mark]: 1.72e-05
        [recompute_prepare]: 7.01001e-06
        [updatestate_depend_eliminate]: 3.46001e-06
        [updatestate_assign_eliminate]: 3.25e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [parameter_eliminate]: 2.37999e-06
        [specialize_transform]: 6.32001e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 5.35001e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 5.19998e-06
        [merge_forward]: 3.80998e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.322e-05
        [j_node_and_user_rematch]: 9.92999e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 9.24998e-06
        [inline_without_move]: 5.40999e-06
        [renormalize]: 0.0008931
        [add_forward_monad_depend]: 5.23002e-06
        [auto_monad_grad]: 3.18998e-06
        [auto_monad_eliminator]: 1.472e-05
        [cse]: 2.312e-05
        [replace_applicator]: 1.172e-05
    [Cycle 2]: 0.00032246, [27]
        [switch_simplify]: 5.84999e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 8.913e-05
        [with_stream_mark]: 1.011e-05
        [recompute_prepare]: 4.87e-06
        [updatestate_depend_eliminate]: 2.58e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 1.14998e-06
        [specialize_transform]: 4.79e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 5.34e-06
        [meta_shard_fg_expand]: 1.41998e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 2.92002e-06
        [cell_reuse_recompute_pass]: 1.81e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.921e-05
        [j_node_and_user_rematch]: 7.88999e-06
        [meta_fg_expand]: 1.69998e-06
        [replace_old_param]: 7.42002e-06
        [inline_without_move]: 5.00001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 4.50001e-06
        [cse]: 1.08e-05
        [replace_applicator]: 5.46e-06
[py_interpret_to_execute_after_opt_a]: 1.077e-05
[rewriter_after_opt_a]: 3.636e-05
[convert_after_rewriter]: 7.19001e-06
[order_py_execute_after_rewriter]: 4.82e-06
[mutable_eliminate]: 0.00062146
[jit_opt_b]: 5.261e-05, [1]
    [Cycle 1]: 4.482e-05, [2]
        [frontend_op_eliminate]: 1.826e-05
        [inline_after_opt_a]: 1.583e-05
[cconv]: 2.901e-05
[loop_unroll]: 0.00043748
[jit_opt_after_cconv]: 0.00014265, [1]
    [Cycle 1]: 0.00013616, [11]
        [c_1]: 2.098e-05
        [parameter_eliminate]: 2.78998e-06
        [updatestate_depend_eliminate]: 6.26998e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [cse]: 1.897e-05
        [call_graph_tuple_transform]: 1.957e-05
        [tuple_list_get_item_eliminator]: 5.30001e-06
        [none_parameter_eliminate]: 1.46998e-06
        [renormalize]: 9.50007e-07
        [switch_simplify]: 5.14e-06
[remove_dup_value]: 1.455e-05
[partial_unused_args_eliminate]: 2.51998e-06
[environ_conv]: 5.82999e-06
[add_recomputation]: 4.649e-05
[cse_after_recomputation]: 2.098e-05, [1]
    [Cycle 1]: 1.515e-05, [1]
        [cse]: 9.19e-06
[auto_monad_reorder]: 1.555e-05
[get_jit_bprop_graph]: 2.17001e-06
[rewriter_after_jit_bprop_graph]: 5.69999e-06
[opt_after_jit_grad]: 0.00046661
[symbol_engine_optimizer]: 7.731e-05, [1]
    [Cycle 1]: 6.991e-05, [6]
        [build]: 5.42001e-06
        [elim_shapecalc]: 8.01001e-06
        [elim_not_effective]: 1.407e-05
        [opt_reshape]: 5.89e-06
        [fold_const_symbol]: 8.22998e-06
        [renormalize]: 5.8001e-07
[validate]: 4.188e-05
Sums
    bootstrap                            :     0.000252s :  2.41%
    type_inference                       :     0.005950s : 56.90%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000073s :  0.69%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000042s :  0.40%
    rewriter_before_opt_a                :     0.000081s :  0.78%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000114s :  1.09%
    jit_opt_a.loop_unroll                :     0.000039s :  0.37%
    jit_opt_a.a_1                        :     0.000789s :  7.55%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000893s :  8.54%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.18%
    jit_opt_a.cse                        :     0.000034s :  0.32%
    jit_opt_a.replace_applicator         :     0.000017s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.10%
    rewriter_after_opt_a                 :     0.000036s :  0.35%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000621s :  5.94%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.15%
    cconv                                :     0.000029s :  0.28%
    loop_unroll                          :     0.000437s :  4.18%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000015s :  0.14%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000046s :  0.44%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000467s :  4.46%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000042s :  0.40%


Time group info:
------[substitution.]   0.000294    26
 1.00% :     0.000003s :      2: substitution.elim_not_effective
 0.40% :     0.000001s :      2: substitution.fold_const_symbol
 1.93% :     0.000006s :      3: substitution.graph_param_transform
83.37% :     0.000245s :      6: substitution.inline
 1.15% :     0.000003s :      4: substitution.j_node_and_user_rematch
 3.48% :     0.000010s :      4: substitution.remove_not_recompute_node
 1.43% :     0.000004s :      2: substitution.replace_old_param
 3.03% :     0.000009s :      1: substitution.switch_simplify
 4.21% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005900     2
83.80% :     0.004945s :      1: type_inference.infer
16.20% :     0.000956s :      1: type_inference.specialize
------[replace.]   0.000132     9
51.20% :     0.000068s :      6: replace.inline
22.87% :     0.000030s :      1: replace.switch_simplify
25.93% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000261     9
92.65% :     0.000242s :      6: match.inline
 3.12% :     0.000008s :      1: match.switch_simplify
 4.23% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   801
 1.39% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.26% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     12: predicate.addn_check_dump
 1.42% :     0.000002s :     12: predicate.addn_zero_filter
 1.85% :     0.000002s :     12: predicate.arithmetic_simplify
 1.28% :     0.000002s :     12: predicate.cast_eliminate
 0.33% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     12: predicate.compare_switch_simplify
 1.37% :     0.000002s :     12: predicate.depend_value_elim
 1.16% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     12: predicate.environ_get_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.03% :     0.000007s :     26: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000002s :      6: predicate.less_batch_normalization
 1.72% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     17: predicate.load_eliminater
 1.87% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000005s :     40: predicate.loop_unroll_before_grad
 2.17% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000002s :     12: predicate.merge_addn
 1.16% :     0.000002s :     12: predicate.minmaximum_grad
 1.65% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     17: predicate.partial_eliminate
 1.18% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     12: predicate.reduce_eliminate
 1.52% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.54% :     0.000002s :     20: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.62% :     0.000002s :     12: predicate.reshape_eliminate
 1.22% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.93% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.76% :     0.000001s :      6: predicate.special_op_eliminate
 1.23% :     0.000002s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.92% :     0.000004s :     20: predicate.switch_defer_inline
 2.36% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.28% :     0.000011s :     65: predicate.switch_simplify
 1.20% :     0.000002s :     12: predicate.tile_eliminate
 1.24% :     0.000002s :     12: predicate.transpose_eliminate
 1.32% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.86% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000773    15
47.68% :     0.000369s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.01% :     0.000101s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.31% :     0.000304s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016035    72
 0.31% :     0.000049s :      1: add_recomputation
 0.48% :     0.000076s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 1.64% :     0.000263s :      1: bootstrap
 0.20% :     0.000032s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
33.69% :     0.005402s :      1: jit_opt_a
 0.91% :     0.000145s :      1: jit_opt_after_cconv
 0.34% :     0.000055s :      1: jit_opt_b
 2.77% :     0.000445s :      1: loop_unroll
 3.93% :     0.000630s :      1: mutable_eliminate
 6.62% :     0.001062s :     26: opt.transform.jit_opt_a
 0.30% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000027s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.96% :     0.000475s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.28% :     0.000045s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 3.36% :     0.000538s :      1: renormalize.infer
 2.16% :     0.000346s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000039s :      1: rewriter_after_opt_a
 0.52% :     0.000084s :      1: rewriter_before_opt_a
 0.50% :     0.000080s :      1: symbol_engine_optimizer
37.18% :     0.005962s :      1: type_inference


TotalTime = 0.0204239, [33]
[bootstrap]: 0.00029172
[type_inference]: 0.00507703
[event_method]: 1.749e-05
[auto_monad]: 6.857e-05
[graph_reusing]: 6.75998e-06
[pre_auto_parallel]: 3.08e-06
[py_interpret_to_execute]: 3.384e-05
[rewriter_before_opt_a]: 7.181e-05
[expand_dump_flag]: 3.6e-06
[jit_opt_a]: 0.00623799, [2]
    [Cycle 1]: 0.00202278, [27]
        [switch_simplify]: 6.877e-05
        [loop_unroll]: 6.56e-05
        [a_1]: 0.0006961
        [with_stream_mark]: 2.716e-05
        [recompute_prepare]: 1.061e-05
        [updatestate_depend_eliminate]: 4.66002e-06
        [updatestate_assign_eliminate]: 3.43e-06
        [updatestate_loads_eliminate]: 2.72001e-06
        [parameter_eliminate]: 2.31e-06
        [specialize_transform]: 7.38999e-06
        [updatestate_useless_node_eliminater]: 5.86998e-06
        [accelerated_algorithm]: 7.3e-06
        [meta_shard_fg_expand]: 2.12999e-06
        [get_grad_eliminate_]: 5.67001e-06
        [merge_forward]: 3.92998e-06
        [cell_reuse_recompute_pass]: 2.54999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.698e-05
        [j_node_and_user_rematch]: 1.034e-05
        [meta_fg_expand]: 3.21001e-06
        [replace_old_param]: 1.086e-05
        [inline_without_move]: 6.40002e-06
        [renormalize]: 0.00078199
        [add_forward_monad_depend]: 7e-06
        [auto_monad_grad]: 2.66e-06
        [auto_monad_eliminator]: 1.553e-05
        [cse]: 2.889e-05
        [replace_applicator]: 1.367e-05
    [Cycle 2]: 0.00035347, [27]
        [switch_simplify]: 5.76e-06
        [loop_unroll]: 5.24998e-06
        [a_1]: 0.00010871
        [with_stream_mark]: 1.102e-05
        [recompute_prepare]: 5.32999e-06
        [updatestate_depend_eliminate]: 2.88e-06
        [updatestate_assign_eliminate]: 2.10002e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.19998e-06
        [specialize_transform]: 4.66002e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.53002e-06
        [get_grad_eliminate_]: 4.70001e-06
        [merge_forward]: 2.27999e-06
        [cell_reuse_recompute_pass]: 2.66999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.271e-05
        [j_node_and_user_rematch]: 8.67e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 8.2e-06
        [inline_without_move]: 5.67999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 8.79983e-07
        [auto_monad_eliminator]: 5.42001e-06
        [cse]: 1.038e-05
        [replace_applicator]: 5.15999e-06
[py_interpret_to_execute_after_opt_a]: 1.107e-05
[rewriter_after_opt_a]: 3.71e-05
[convert_after_rewriter]: 8.10999e-06
[order_py_execute_after_rewriter]: 5.46e-06
[mutable_eliminate]: 0.00062721
[jit_opt_b]: 5.028e-05, [1]
    [Cycle 1]: 4.299e-05, [2]
        [frontend_op_eliminate]: 1.656e-05
        [inline_after_opt_a]: 1.501e-05
[cconv]: 2.559e-05
[loop_unroll]: 0.00042444
[jit_opt_after_cconv]: 0.00013526, [1]
    [Cycle 1]: 0.00012906, [11]
        [c_1]: 2.038e-05
        [parameter_eliminate]: 2.46998e-06
        [updatestate_depend_eliminate]: 5.14e-06
        [updatestate_assign_eliminate]: 2.36998e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.687e-05
        [call_graph_tuple_transform]: 1.803e-05
        [tuple_list_get_item_eliminator]: 5.32001e-06
        [none_parameter_eliminate]: 1.63997e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 5.47999e-06
[remove_dup_value]: 1.282e-05
[partial_unused_args_eliminate]: 2.41998e-06
[environ_conv]: 4.69998e-06
[add_recomputation]: 5.148e-05
[cse_after_recomputation]: 2.13e-05, [1]
    [Cycle 1]: 1.61e-05, [1]
        [cse]: 9.81e-06
[auto_monad_reorder]: 1.419e-05
[get_jit_bprop_graph]: 2.19001e-06
[rewriter_after_jit_bprop_graph]: 4.91997e-06
[opt_after_jit_grad]: 0.00048679
[symbol_engine_optimizer]: 7.777e-05, [1]
    [Cycle 1]: 7.064e-05, [6]
        [build]: 3.63e-06
        [elim_shapecalc]: 7.98001e-06
        [elim_not_effective]: 1.384e-05
        [opt_reshape]: 5.83997e-06
        [fold_const_symbol]: 9.17999e-06
        [renormalize]: 5.50004e-07
[validate]: 3.573e-05
[backend_pass]: 1.00999e-06
[task_emit]: 0.00636331
[execute]: 1.037e-05
Sums
    bootstrap                            :     0.000292s :  1.83%
    type_inference                       :     0.005077s : 31.91%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000069s :  0.43%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.21%
    rewriter_before_opt_a                :     0.000072s :  0.45%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000075s :  0.47%
    jit_opt_a.loop_unroll                :     0.000071s :  0.45%
    jit_opt_a.a_1                        :     0.000805s :  5.06%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.12%
    jit_opt_a.inline_without_move        :     0.000012s :  0.08%
    jit_opt_a.renormalize                :     0.000782s :  4.92%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.13%
    jit_opt_a.cse                        :     0.000039s :  0.25%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000627s :  3.94%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000026s :  0.16%
    loop_unroll                          :     0.000424s :  2.67%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000051s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000487s :  3.06%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006363s : 40.00%
    execute                              :     0.000010s :  0.07%


Time group info:
------[substitution.]   0.000282    23
 0.67% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 1.93% :     0.000005s :      3: substitution.graph_param_transform
87.15% :     0.000246s :      4: substitution.inline
 1.39% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.59% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.80% :     0.000005s :      2: substitution.replace_old_param
 5.01% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005007     2
83.34% :     0.004173s :      1: type_inference.infer
16.66% :     0.000834s :      1: type_inference.specialize
------[replace.]   0.000084     6
64.65% :     0.000054s :      4: replace.inline
35.35% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000256     6
95.00% :     0.000243s :      4: match.inline
 5.00% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000155   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.11% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.88% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 2.14% :     0.000003s :     10: predicate.arithmetic_simplify
 1.07% :     0.000002s :     10: predicate.cast_eliminate
 0.32% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.92% :     0.000001s :     10: predicate.compare_switch_simplify
 1.96% :     0.000003s :     10: predicate.depend_value_elim
 1.07% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.13% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.73% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.98% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.05% :     0.000002s :     10: predicate.environ_get_eliminate
 0.99% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.77% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 4.68% :     0.000007s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.56% :     0.000002s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000003s :     15: predicate.load_eliminater
 1.08% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.44% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.84% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.88% :     0.000001s :     10: predicate.merge_addn
 0.92% :     0.000001s :     10: predicate.minmaximum_grad
 1.38% :     0.000002s :      3: predicate.mutable_eliminate
 0.34% :     0.000001s :      3: predicate.opt_reshape
 1.95% :     0.000003s :     15: predicate.partial_eliminate
 0.95% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.19% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.37% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.47% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.11% :     0.000002s :     10: predicate.reshape_eliminate
12.30% :     0.000019s :     10: predicate.row_tensor_add_zeros_like
 0.59% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.98% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000002s :      6: predicate.special_op_eliminate
 0.85% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 0.99% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.33% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.86% :     0.000004s :     16: predicate.switch_defer_inline
 1.90% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.93% :     0.000012s :     53: predicate.switch_simplify
 0.98% :     0.000002s :     10: predicate.tile_eliminate
 0.93% :     0.000001s :     10: predicate.transpose_eliminate
 1.19% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.42% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.29% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.31% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.41% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.52% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.21% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.52% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000606    12
47.89% :     0.000290s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.55% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.56% :     0.000264s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022366    76
 0.24% :     0.000054s :      1: add_recomputation
 0.32% :     0.000072s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.36% :     0.000303s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
27.90% :     0.006241s :      1: jit_opt_a
 0.62% :     0.000138s :      1: jit_opt_after_cconv
 0.24% :     0.000053s :      1: jit_opt_b
 1.93% :     0.000432s :      1: loop_unroll
 2.84% :     0.000636s :      1: mutable_eliminate
 4.83% :     0.001080s :     26: opt.transform.jit_opt_a
 0.21% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000495s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 1.88% :     0.000421s :      1: renormalize.infer
 1.45% :     0.000325s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000040s :      1: rewriter_after_opt_a
 0.34% :     0.000075s :      1: rewriter_before_opt_a
 0.36% :     0.000081s :      1: symbol_engine_optimizer
28.53% :     0.006381s :      1: task_emit
22.77% :     0.005093s :      1: type_inference
 0.25% :     0.000055s :      1: validate


TotalTime = 0.0199923, [33]
[bootstrap]: 0.00036811
[type_inference]: 0.00486325
[event_method]: 1.797e-05
[auto_monad]: 6.584e-05
[graph_reusing]: 5.89e-06
[pre_auto_parallel]: 2.66999e-06
[py_interpret_to_execute]: 3.777e-05
[rewriter_before_opt_a]: 7.705e-05
[expand_dump_flag]: 3.21001e-06
[jit_opt_a]: 0.00568051, [2]
    [Cycle 1]: 0.00183157, [27]
        [switch_simplify]: 5.477e-05
        [loop_unroll]: 2.877e-05
        [a_1]: 0.00054753
        [with_stream_mark]: 2.059e-05
        [recompute_prepare]: 7.37002e-06
        [updatestate_depend_eliminate]: 3.82998e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.47001e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 5.46002e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 5.95002e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 5.34998e-06
        [merge_forward]: 3.41999e-06
        [cell_reuse_recompute_pass]: 1.19998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.422e-05
        [j_node_and_user_rematch]: 8.77999e-06
        [meta_fg_expand]: 2.07001e-06
        [replace_old_param]: 9.17001e-06
        [inline_without_move]: 5.17e-06
        [renormalize]: 0.0008562
        [add_forward_monad_depend]: 5.37999e-06
        [auto_monad_grad]: 2.28998e-06
        [auto_monad_eliminator]: 1.402e-05
        [cse]: 2.869e-05
        [replace_applicator]: 1.189e-05
    [Cycle 2]: 0.00036185, [27]
        [switch_simplify]: 5.32999e-06
        [loop_unroll]: 4.72e-06
        [a_1]: 0.00012936
        [with_stream_mark]: 1.033e-05
        [recompute_prepare]: 5.19e-06
        [updatestate_depend_eliminate]: 2.93e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 4.96002e-06
        [updatestate_useless_node_eliminater]: 4.62e-06
        [accelerated_algorithm]: 5.12999e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 4.48001e-06
        [merge_forward]: 2.81999e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.209e-05
        [j_node_and_user_rematch]: 7.55e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 7.13e-06
        [inline_without_move]: 4.60999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 1.25001e-06
        [auto_monad_eliminator]: 5.05001e-06
        [cse]: 1.079e-05
        [replace_applicator]: 5.17e-06
[py_interpret_to_execute_after_opt_a]: 9.32999e-06
[rewriter_after_opt_a]: 3.359e-05
[convert_after_rewriter]: 7.28e-06
[order_py_execute_after_rewriter]: 5.42999e-06
[mutable_eliminate]: 0.00054436
[jit_opt_b]: 4.847e-05, [1]
    [Cycle 1]: 4.07e-05, [2]
        [frontend_op_eliminate]: 1.502e-05
        [inline_after_opt_a]: 1.432e-05
[cconv]: 2.601e-05
[loop_unroll]: 0.00042312
[jit_opt_after_cconv]: 0.00013865, [1]
    [Cycle 1]: 0.00013192, [11]
        [c_1]: 2.112e-05
        [parameter_eliminate]: 3.09001e-06
        [updatestate_depend_eliminate]: 6.16e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.81e-05
        [call_graph_tuple_transform]: 1.783e-05
        [tuple_list_get_item_eliminator]: 5.06997e-06
        [none_parameter_eliminate]: 1.46998e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 5.00999e-06
[remove_dup_value]: 1.311e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 4.92e-06
[add_recomputation]: 4.868e-05
[cse_after_recomputation]: 1.983e-05, [1]
    [Cycle 1]: 1.437e-05, [1]
        [cse]: 8.48999e-06
[auto_monad_reorder]: 1.482e-05
[get_jit_bprop_graph]: 1.84998e-06
[rewriter_after_jit_bprop_graph]: 6.16998e-06
[opt_after_jit_grad]: 0.00045348
[symbol_engine_optimizer]: 6.9e-05, [1]
    [Cycle 1]: 6.216e-05, [6]
        [build]: 3.23998e-06
        [elim_shapecalc]: 7.03e-06
        [elim_not_effective]: 1.146e-05
        [opt_reshape]: 5.53002e-06
        [fold_const_symbol]: 8.38999e-06
        [renormalize]: 4.80009e-07
[validate]: 3.669e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.00673151
[execute]: 8.87e-06
Sums
    bootstrap                            :     0.000368s :  2.32%
    type_inference                       :     0.004863s : 30.64%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000066s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.24%
    rewriter_before_opt_a                :     0.000077s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000060s :  0.38%
    jit_opt_a.loop_unroll                :     0.000033s :  0.21%
    jit_opt_a.a_1                        :     0.000677s :  4.26%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000856s :  5.40%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000039s :  0.25%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000034s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000544s :  3.43%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000026s :  0.16%
    loop_unroll                          :     0.000423s :  2.67%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.31%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000453s :  2.86%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006732s : 42.41%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000220    23
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.47% :     0.000005s :      3: substitution.graph_param_transform
85.07% :     0.000187s :      4: substitution.inline
 1.37% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.96% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.02% :     0.000004s :      2: substitution.replace_old_param
 5.67% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004791     2
75.53% :     0.003619s :      1: type_inference.infer
24.47% :     0.001172s :      1: type_inference.specialize
------[replace.]   0.000067     6
68.29% :     0.000046s :      4: replace.inline
31.71% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000196     6
94.17% :     0.000185s :      4: match.inline
 5.83% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.30% :     0.000001s :     10: predicate.accumulaten_eliminater
 0.99% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.83% :     0.000002s :     10: predicate.addn_zero_filter
 2.55% :     0.000003s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000001s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.30% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.33% :     0.000002s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 6.20% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.07% :     0.000001s :      6: predicate.less_batch_normalization
 1.69% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.92% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.32% :     0.000001s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     15: predicate.partial_eliminate
 1.45% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.92% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.53% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.23% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.15% :     0.000008s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.66% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000883    12
28.71% :     0.000254s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.20% :     0.000046s :      1: func_graph_cloner_run.FuncGraphClonerNode
66.09% :     0.000584s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021823    76
 0.24% :     0.000052s :      1: add_recomputation
 0.32% :     0.000069s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.77% :     0.000387s :      1: bootstrap
 0.13% :     0.000029s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.05% :     0.005684s :      1: jit_opt_a
 0.65% :     0.000141s :      1: jit_opt_after_cconv
 0.23% :     0.000051s :      1: jit_opt_b
 1.98% :     0.000432s :      1: loop_unroll
 2.53% :     0.000553s :      1: mutable_eliminate
 4.04% :     0.000882s :     26: opt.transform.jit_opt_a
 0.21% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000018s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.11% :     0.000461s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000041s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 2.45% :     0.000534s :      1: renormalize.infer
 1.44% :     0.000314s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000038s :      1: rewriter_after_opt_a
 0.37% :     0.000080s :      1: rewriter_before_opt_a
 0.33% :     0.000071s :      1: symbol_engine_optimizer
30.92% :     0.006748s :      1: task_emit
22.42% :     0.004893s :      1: type_inference
 0.25% :     0.000055s :      1: validate


TotalTime = 0.0271048, [33]
[bootstrap]: 0.00033317
[type_inference]: 0.00971814
[event_method]: 2.373e-05
[auto_monad]: 7.647e-05
[graph_reusing]: 6.06998e-06
[pre_auto_parallel]: 2.98e-06
[py_interpret_to_execute]: 4.766e-05
[rewriter_before_opt_a]: 0.00010267
[expand_dump_flag]: 3.24001e-06
[jit_opt_a]: 0.00683418, [2]
    [Cycle 1]: 0.00206991, [27]
        [switch_simplify]: 6.413e-05
        [loop_unroll]: 3.015e-05
        [a_1]: 0.0006205
        [with_stream_mark]: 2.392e-05
        [recompute_prepare]: 8.12998e-06
        [updatestate_depend_eliminate]: 4.27998e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.50002e-06
        [parameter_eliminate]: 2.44001e-06
        [specialize_transform]: 6.12001e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.96e-06
        [meta_shard_fg_expand]: 2.48e-06
        [get_grad_eliminate_]: 5.84e-06
        [merge_forward]: 3.48e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.578e-05
        [j_node_and_user_rematch]: 1.003e-05
        [meta_fg_expand]: 2.51998e-06
        [replace_old_param]: 9.99999e-06
        [inline_without_move]: 5.66e-06
        [renormalize]: 0.00097507
        [add_forward_monad_depend]: 6.68e-06
        [auto_monad_grad]: 2.81999e-06
        [auto_monad_eliminator]: 1.597e-05
        [cse]: 3.029e-05
        [replace_applicator]: 1.51e-05
    [Cycle 2]: 0.00036479, [27]
        [switch_simplify]: 2.015e-05
        [loop_unroll]: 5.86e-06
        [a_1]: 9.585e-05
        [with_stream_mark]: 1.14e-05
        [recompute_prepare]: 5.19998e-06
        [updatestate_depend_eliminate]: 2.97002e-06
        [updatestate_assign_eliminate]: 2.43998e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.81998e-06
        [specialize_transform]: 5.32001e-06
        [updatestate_useless_node_eliminater]: 5.33002e-06
        [accelerated_algorithm]: 5.86998e-06
        [meta_shard_fg_expand]: 1.69e-06
        [get_grad_eliminate_]: 4.86997e-06
        [merge_forward]: 2.58e-06
        [cell_reuse_recompute_pass]: 2.20002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.354e-05
        [j_node_and_user_rematch]: 8.97999e-06
        [meta_fg_expand]: 1.99e-06
        [replace_old_param]: 8.01001e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 5.82001e-06
        [cse]: 1.416e-05
        [replace_applicator]: 5.00999e-06
[py_interpret_to_execute_after_opt_a]: 1.429e-05
[rewriter_after_opt_a]: 3.887e-05
[convert_after_rewriter]: 8.26002e-06
[order_py_execute_after_rewriter]: 5.39e-06
[mutable_eliminate]: 0.0007522
[jit_opt_b]: 5.664e-05, [1]
    [Cycle 1]: 4.868e-05, [2]
        [frontend_op_eliminate]: 1.854e-05
        [inline_after_opt_a]: 1.633e-05
[cconv]: 3.531e-05
[loop_unroll]: 0.00051547
[jit_opt_after_cconv]: 0.00016544, [1]
    [Cycle 1]: 0.00015881, [11]
        [c_1]: 2.324e-05
        [parameter_eliminate]: 4.45e-06
        [updatestate_depend_eliminate]: 9.34e-06
        [updatestate_assign_eliminate]: 2.58998e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [cse]: 2.603e-05
        [call_graph_tuple_transform]: 2.383e-05
        [tuple_list_get_item_eliminator]: 6.12999e-06
        [none_parameter_eliminate]: 1.51002e-06
        [renormalize]: 1.08001e-06
        [switch_simplify]: 5.91e-06
[remove_dup_value]: 1.474e-05
[partial_unused_args_eliminate]: 2.27001e-06
[environ_conv]: 5.94e-06
[add_recomputation]: 5.623e-05
[cse_after_recomputation]: 2.244e-05, [1]
    [Cycle 1]: 1.644e-05, [1]
        [cse]: 9.74e-06
[auto_monad_reorder]: 1.742e-05
[get_jit_bprop_graph]: 2.79999e-06
[rewriter_after_jit_bprop_graph]: 8.33001e-06
[opt_after_jit_grad]: 0.0005595
[symbol_engine_optimizer]: 8.07e-05, [1]
    [Cycle 1]: 7.254e-05, [6]
        [build]: 4.33001e-06
        [elim_shapecalc]: 8.22e-06
        [elim_not_effective]: 1.409e-05
        [opt_reshape]: 6.38998e-06
        [fold_const_symbol]: 9.41998e-06
        [renormalize]: 1.30001e-06
[validate]: 4.343e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00728502
[execute]: 1.099e-05
Sums
    bootstrap                            :     0.000333s :  1.51%
    type_inference                       :     0.009718s : 44.16%
    event_method                         :     0.000024s :  0.11%
    auto_monad                           :     0.000076s :  0.35%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000048s :  0.22%
    rewriter_before_opt_a                :     0.000103s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000084s :  0.38%
    jit_opt_a.loop_unroll                :     0.000036s :  0.16%
    jit_opt_a.a_1                        :     0.000716s :  3.26%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000975s :  4.43%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.10%
    jit_opt_a.cse                        :     0.000044s :  0.20%
    jit_opt_a.replace_applicator         :     0.000020s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.06%
    rewriter_after_opt_a                 :     0.000039s :  0.18%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000752s :  3.42%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.07%
    cconv                                :     0.000035s :  0.16%
    loop_unroll                          :     0.000515s :  2.34%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000056s :  0.26%
    cse_after_recomputation.cse          :     0.000010s :  0.04%
    auto_monad_reorder                   :     0.000017s :  0.08%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000560s :  2.54%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.20%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007285s : 33.11%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000267    23
 0.97% :     0.000003s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 2.50% :     0.000007s :      3: substitution.graph_param_transform
85.67% :     0.000229s :      4: substitution.inline
 1.45% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.96% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.76% :     0.000005s :      2: substitution.replace_old_param
 5.21% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009619     2
87.01% :     0.008369s :      1: type_inference.infer
12.99% :     0.001250s :      1: type_inference.specialize
------[replace.]   0.000082     6
68.52% :     0.000056s :      4: replace.inline
31.48% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000239     6
94.76% :     0.000226s :      4: match.inline
 5.24% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.44% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.39% :     0.000003s :     10: predicate.arithmetic_simplify
 1.19% :     0.000002s :     10: predicate.cast_eliminate
 0.56% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.14% :     0.000001s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.48% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.50% :     0.000001s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.41% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.08% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.40% :     0.000001s :      3: predicate.graph_param_transform
 5.54% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      6: predicate.less_batch_normalization
 1.34% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 1.74% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.25% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.98% :     0.000004s :      3: predicate.mutable_eliminate
 0.58% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000003s :     15: predicate.partial_eliminate
 1.04% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 2.29% :     0.000003s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.32% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.88% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.95% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000824    12
41.19% :     0.000340s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.60% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
52.21% :     0.000430s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.029156    76
 0.20% :     0.000059s :      1: add_recomputation
 0.28% :     0.000081s :      1: auto_monad
 0.07% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 1.21% :     0.000354s :      1: bootstrap
 0.13% :     0.000038s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.10% :     0.000029s :      1: event_method
 0.05% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
23.45% :     0.006838s :      1: jit_opt_a
 0.58% :     0.000169s :      1: jit_opt_after_cconv
 0.20% :     0.000060s :      1: jit_opt_b
 1.80% :     0.000524s :      1: loop_unroll
 2.62% :     0.000765s :      1: mutable_eliminate
 3.29% :     0.000960s :     26: opt.transform.jit_opt_a
 0.19% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.95% :     0.000570s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000051s :      1: py_interpret_to_execute
 0.06% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000018s :      1: remove_dup_value
 1.93% :     0.000562s :      1: renormalize.infer
 1.39% :     0.000404s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000043s :      1: rewriter_after_opt_a
 0.37% :     0.000106s :      1: rewriter_before_opt_a
 0.28% :     0.000083s :      1: symbol_engine_optimizer
25.06% :     0.007307s :      1: task_emit
33.41% :     0.009742s :      1: type_inference
 0.23% :     0.000067s :      1: validate


TotalTime = 0.0147489, [30]
[bootstrap]: 0.00031348
[type_inference]: 0.00668624
[event_method]: 2.045e-05
[auto_monad]: 7.963e-05
[graph_reusing]: 5.74e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 4.428e-05
[rewriter_before_opt_a]: 7.945e-05
[expand_dump_flag]: 2.84999e-06
[jit_opt_a]: 0.00557749, [2]
    [Cycle 1]: 0.0017625, [27]
        [switch_simplify]: 0.00010916
        [loop_unroll]: 3.331e-05
        [a_1]: 0.00067727
        [with_stream_mark]: 1.891e-05
        [recompute_prepare]: 7.22002e-06
        [updatestate_depend_eliminate]: 3.34001e-06
        [updatestate_assign_eliminate]: 3.25e-06
        [updatestate_loads_eliminate]: 2.48998e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 6.05002e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 1.74e-06
        [get_grad_eliminate_]: 5.25999e-06
        [merge_forward]: 3.31999e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.219e-05
        [j_node_and_user_rematch]: 9.22999e-06
        [meta_fg_expand]: 2.10002e-06
        [replace_old_param]: 7.85e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 0.00062108
        [add_forward_monad_depend]: 5.01997e-06
        [auto_monad_grad]: 2.42001e-06
        [auto_monad_eliminator]: 1.255e-05
        [cse]: 2.092e-05
        [replace_applicator]: 1.094e-05
    [Cycle 2]: 0.00030776, [27]
        [switch_simplify]: 5.30999e-06
        [loop_unroll]: 4.83001e-06
        [a_1]: 9.239e-05
        [with_stream_mark]: 8.33001e-06
        [recompute_prepare]: 5.04998e-06
        [updatestate_depend_eliminate]: 2.34999e-06
        [updatestate_assign_eliminate]: 1.97999e-06
        [updatestate_loads_eliminate]: 1.91998e-06
        [parameter_eliminate]: 9.49978e-07
        [specialize_transform]: 4.85999e-06
        [updatestate_useless_node_eliminater]: 4.72998e-06
        [accelerated_algorithm]: 4.82e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.52998e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.42e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.187e-05
        [j_node_and_user_rematch]: 7.61999e-06
        [meta_fg_expand]: 1.55001e-06
        [replace_old_param]: 6.68e-06
        [inline_without_move]: 4.65999e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.00001e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 4.90999e-06
        [cse]: 8.77e-06
        [replace_applicator]: 4.82e-06
[py_interpret_to_execute_after_opt_a]: 9.83002e-06
[rewriter_after_opt_a]: 3.144e-05
[convert_after_rewriter]: 6.33e-06
[order_py_execute_after_rewriter]: 5.15999e-06
[mutable_eliminate]: 0.00049688
[jit_opt_b]: 4.604e-05, [1]
    [Cycle 1]: 3.958e-05, [2]
        [frontend_op_eliminate]: 1.456e-05
        [inline_after_opt_a]: 1.417e-05
[cconv]: 2.819e-05
[loop_unroll]: 0.00040244
[jit_opt_after_cconv]: 0.00013138, [1]
    [Cycle 1]: 0.00012508, [11]
        [c_1]: 2.03e-05
        [parameter_eliminate]: 2.51998e-06
        [updatestate_depend_eliminate]: 5.54e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 1.503e-05
        [call_graph_tuple_transform]: 1.685e-05
        [tuple_list_get_item_eliminator]: 5.00001e-06
        [none_parameter_eliminate]: 1.27999e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.09e-06
[remove_dup_value]: 1.146e-05
[partial_unused_args_eliminate]: 2.10002e-06
[environ_conv]: 4.31002e-06
[add_recomputation]: 3.921e-05
[cse_after_recomputation]: 1.927e-05, [1]
    [Cycle 1]: 1.415e-05, [1]
        [cse]: 8.60999e-06
[auto_monad_reorder]: 1.259e-05
[get_jit_bprop_graph]: 1.87999e-06
[rewriter_after_jit_bprop_graph]: 4.59998e-06
[opt_after_jit_grad]: 0.00043297
[symbol_engine_optimizer]: 6.746e-05, [1]
    [Cycle 1]: 6.176e-05, [6]
        [build]: 2.66999e-06
        [elim_shapeca
TotalTime = 0.0191028, [33]
[bootstrap]: 0.00030931
[type_inference]: 0.00450429
[event_method]: 1.782e-05
[auto_monad]: 6.504e-05
[graph_reusing]: 5.53997e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 3.369e-05
[rewriter_before_opt_a]: 7.327e-05
[expand_dump_flag]: 3.35e-06
[jit_opt_a]: 0.00546921, [2]
    [Cycle 1]: 0.00163181, [27]
        [switch_simplify]: 5.657e-05
        [loop_unroll]: 3.136e-05
        [a_1]: 0.00055321
        [with_stream_mark]: 1.81e-05
        [recompute_prepare]: 7.25e-06
        [updatestate_depend_eliminate]: 3.58e-06
        [updatestate_assign_eliminate]: 3.09001e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [parameter_eliminate]: 2.09999e-06
        [specialize_transform]: 6.14001e-06
        [updatestate_useless_node_eliminater]: 5.04998e-06
        [accelerated_algorithm]: 5.21002e-06
        [meta_shard_fg_expand]: 2.02001e-06
        [get_grad_eliminate_]: 4.89998e-06
        [merge_forward]: 4.07e-06
        [cell_reuse_recompute_pass]: 1.28002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.466e-05
        [j_node_and_user_rematch]: 8.79e-06
        [meta_fg_expand]: 2.49001e-06
        [replace_old_param]: 8.75999e-06
        [inline_without_move]: 5.04003e-06
        [renormalize]: 0.00062911
        [add_forward_monad_depend]: 5.47999e-06
        [auto_monad_grad]: 2.54999e-06
        [auto_monad_eliminator]: 1.338e-05
        [cse]: 2.884e-05
        [replace_applicator]: 1.219e-05
    [Cycle 2]: 0.00032925, [27]
        [switch_simplify]: 6.14001e-06
        [loop_unroll]: 4.87998e-06
        [a_1]: 9.038e-05
        [with_stream_mark]: 9.31e-06
        [recompute_prepare]: 5.19e-06
        [updatestate_depend_eliminate]: 2.64999e-06
        [updatestate_assign_eliminate]: 1.99999e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 5.04e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 5.18002e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.27999e-06
        [cell_reuse_recompute_pass]: 2.06998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.309e-05
        [j_node_and_user_rematch]: 8.38001e-06
        [meta_fg_expand]: 1.82999e-06
        [replace_old_param]: 7.64002e-06
        [inline_without_move]: 4.92e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.49998e-06
        [auto_monad_grad]: 1.15001e-06
        [auto_monad_eliminator]: 5.39e-06
        [cse]: 1.073e-05
        [replace_applicator]: 5.61003e-06
[py_interpret_to_execute_after_opt_a]: 1.079e-05
[rewriter_after_opt_a]: 3.721e-05
[convert_after_rewriter]: 7.11999e-06
[order_py_execute_after_rewriter]: 5.85002e-06
[mutable_eliminate]: 0.00062243
[jit_opt_b]: 5.418e-05, [1]
    [Cycle 1]: 4.555e-05, [2]
        [frontend_op_eliminate]: 1.723e-05
        [inline_after_opt_a]: 1.633e-05
[cconv]: 2.695e-05
[loop_unroll]: 0.00047102
[jit_opt_after_cconv]: 0.00018838, [1]
    [Cycle 1]: 0.00018147, [11]
        [c_1]: 2.218e-05
        [parameter_eliminate]: 3.586e-05
        [updatestate_depend_eliminate]: 7.84997e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [cse]: 2.34e-05
        [call_graph_tuple_transform]: 2.151e-05
        [tuple_list_get_item_eliminator]: 5.61e-06
        [none_parameter_eliminate]: 1.54998e-06
        [renormalize]: 7.99977e-07
        [switch_simplify]: 5.39998e-06
[remove_dup_value]: 1.38e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 5.76998e-06
[add_recomputation]: 5.13e-05
[cse_after_recomputation]: 2.079e-05, [1]
    [Cycle 1]: 1.505e-05, [1]
        [cse]: 8.97999e-06
[auto_monad_reorder]: 1.525e-05
[get_jit_bprop_graph]: 1.76e-06
[rewriter_after_jit_bprop_graph]: 6.26e-06
[opt_after_jit_grad]: 0.00049083
[symbol_engine_optimizer]: 7.462e-05, [1]
    [Cycle 1]: 6.762e-05, [6]
        [build]: 4.16001e-06
        [elim_shapecalc]: 8.33001e-0lc]: 7.11999e-06
        [elim_not_effective]: 1.158e-05
        [opt_reshape]: 5.42001e-06
        [fold_const_symbol]: 7.92e-06
        [renormalize]: 4.60015e-07
[validate]: 2.81e-05
Sums
    bootstrap                            :     0.000313s :  2.93%
    type_inference                       :     0.006686s : 62.52%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000080s :  0.74%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000044s :  0.41%
    rewriter_before_opt_a                :     0.000079s :  0.74%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000114s :  1.07%
    jit_opt_a.loop_unroll                :     0.000038s :  0.36%
    jit_opt_a.a_1                        :     0.000770s :  7.20%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000621s :  5.81%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.16%
    jit_opt_a.cse                        :     0.000030s :  0.28%
    jit_opt_a.replace_applicator         :     0.000016s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.09%
    rewriter_after_opt_a                 :     0.000031s :  0.29%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000497s :  4.65%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.13%
    cconv                                :     0.000028s :  0.26%
    loop_unroll                          :     0.000402s :  3.76%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000011s :  0.11%
    partial_unused_args_eliminate        :  6
        [elim_not_effective]: 1.322e-05
        [opt_reshape]: 5.76e-06
        [fold_const_symbol]: 8.48999e-06
        [renormalize]: 3.69997e-07
[validate]: 3.737e-05
[backend_pass]: 1.45999e-06
[task_emit]: 0.00623933
[execute]: 7.60998e-06
Sums
    bootstrap                            :     0.000309s :  2.07%
    type_inference                       :     0.004504s : 30.17%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000065s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.23%
    rewriter_before_opt_a                :     0.000073s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.42%
    jit_opt_a.loop_unroll                :     0.000036s :  0.24%
    jit_opt_a.a_1                        :     0.000644s :  4.31%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000629s :  4.21%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000040s :  0.27%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000622s :  4.17%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000027s :  0.18%
    loop_unroll                          :     0.000471s :  3.16%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000036s :  0.24%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.   0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000039s :  0.37%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000013s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000433s :  4.05%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.26%

000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000051s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000491s :  3.29%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006239s : 41.80%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000261    26
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 1.69% :     0.000004s :      3: substitution.graph_param_transform
85.84% :     0.000224s :      6: substitution.inline
 1.21% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.63% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.17% :     0.000003s :      2: substitution.replace_old_param
 2.84% :     0.000007s :      1: substitution.switch_simplify
 4.43% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006628     2
84.28% :     0.005586s :      1: type_inference.infer
15.72% :     0.001042s :      1: type_inference.specialize
------[replace.]   0.000128     9
49.31% :     0.000063s :      6: replace.inline
27.86% :     0.000036s :      1: replace.switch_simplify
22.83% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000238     9
92.84% :     0.000221s :      6: match.inline
 2.83% :     0.000007s :      1: match.switch_simplify
 4.34% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   801
 1.35% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.02% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.20% :     0.000001s :     12: predicate.addn_check_dump
 1.56% :     0.000002s :     12: predicate.addn_zero_filter
 1.88% :     0.000002s :     12: predicate.arithmetic_simplify
 1.27% :     0.000002s :     12: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.19% :     0.000001s :     12: predicate.compare_switch_simplify
 1.68% :     0.000002s :     12: predicate.depend_value_elim
 1.16% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     12: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.25% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.23% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     12: predicate.environ_get_eliminate
 1.18% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 5.70% :     0.000007s :     26: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.64% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     17: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.22% :     0.000005s :     40: predicate.loop_unroll_before_grad
 2.12% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     12: predicate.merge_addn
 1.17% :     0.000001s :     12: predicate.minmaximum_grad
 1.41% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     17: predicate.partial_eliminate
 1.26% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.86% :     0.000002s :     12: predicate.reduce_eliminate
 1.55% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     20: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     12: predicate.reshape_eliminate
 1.35% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     12: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.72% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.82% :     0.000004s :     20: predicate.switch_defer_inline
 2.44% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.40% :     0.000010s :     65: predicate.switch_simplify
 1.23% :     0.000002s :     12: predicate.tile_eliminate
 1.24% :     0.000002s :     12: predicate.transpose_eliminate
 1.40% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.32% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.77% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     12: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000846    15
54.93% :     0.000464s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.73% :     0.000091s :      1: func_graph_cloner_run.FuncGraphClonerNode
34.34% :     0.000290s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016459    72
 0.25% :     0.000042s :      1: add_recomputation
 0.51% :     0.000083s :      1: auto_monad
 0.09% :     0.000015s :      1: auto_monad_reorder
 2.01% :     0.000331s :      1: bootstrap
 0.19% :     0.000031s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
33.90% :     0.005580s :      1: jit_opt_a
 0.81% :     0.000134s :      1: jit_opt_after_cconv
 0.30% :     0.000049s :      1: jit_opt_b
 2.49% :     0.000410s :      1: loop_unroll
 3.07% :     0.000505s :      1: mutable_eliminate
 6.26% :     0.001030s :     26: opt.transform.jit_opt_a
 0.27% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000023s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.67% :     0.000440s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.28% :     0.000047s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 2.17% :     0.000357s :      1: renormalize.infer
 1.56% :     0.000257s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000034s :      1: rewriter_after_opt_a
 0.50% :     0.000082s :      1: rewriter_before_opt_a
 0.42% :     0.000070s :      1: symbol_engine_optimizer
40.70% :     0.006698s :      1: type_inference


Time group info:
------[substitution.]   0.000220    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.58% :     0.000006s :      3: substitution.graph_param_transform
84.48% :     0.000186s :      4: substitution.inline
 1.51% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.20% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.95% :     0.000004s :      2: substitution.replace_old_param
 5.63% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004442     2
81.04% :     0.003600s :      1: type_inference.infer
18.96% :     0.000842s :      1: type_inference.specialize
------[replace.]   0.000069     6
68.50% :     0.000047s :      4: replace.inline
31.50% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000194     6
94.23% :     0.000183s :      4: match.inline
 5.77% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.30% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.48% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000001s :     10: predicate.addn_zero_filter
 1.95% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.33% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.34% :     0.000000s :      3: predicate.graph_param_transform
 5.62% :     0.000007s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.02% :     0.000001s :      6: predicate.less_batch_normalization
 1.69% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.85% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.21% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.20% :     0.000001s :     10: predicate.minmaximum_grad
 1.85% :     0.000002s :      3: predicate.mutable_eliminate
 0.66% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.80% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.36% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.64% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.56% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.12% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.95% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000558    12
49.43% :     0.000276s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.51% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.07% :     0.000252s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020702    76
 0.26% :     0.000054s :      1: add_recomputation
 0.33% :     0.000068s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 1.59% :     0.000329s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.06% :     0.000012s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.43% :     0.005472s :      1: jit_opt_a
 0.92% :     0.000191s :      1: jit_opt_after_cconv
 0.28% :     0.000057s :      1: jit_opt_b
 2.32% :     0.000481s :      1: loop_unroll
 3.06% :     0.000633s :      1: mutable_eliminate
 4.14% :     0.000856s :     26: opt.transform.jit_opt_a
 0.24% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.41% :     0.000499s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.56% :     0.000323s :      1: renormalize.infer
 1.45% :     0.000299s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000040s :      1: rewriter_after_opt_a
 0.37% :     0.000076s :      1: rewriter_before_opt_a
 0.37% :     0.000077s :      1: symbol_engine_optimizer
30.19% :     0.006249s :      1: task_emit
21.81% :     0.004516s :      1: type_inference
 0.42% :     0.000086s :      1: validate


TotalTime = 0.0249785, [33]
[bootstrap]: 0.00036968
[type_inference]: 0.00594662
[event_method]: 2.137e-05
[auto_monad]: 7.026e-05
[graph_reusing]: 7.03e-06
[pre_auto_parallel]: 3.51001e-06
[py_interpret_to_execute]: 4.528e-05
[rewriter_before_opt_a]: 7.834e-05
[expand_dump_flag]: 3.81999e-06
[jit_opt_a]: 0.00777482, [2]
    [Cycle 1]: 0.00224918, [27]
        [switch_simplify]: 7.33e-05
        [loop_unroll]: 3.098e-05
        [a_1]: 0.00068219
        [with_stream_mark]: 2.582e-05
        [recompute_prepare]: 8.04002e-06
        [updatestate_depend_eliminate]: 4.27003e-06
        [updatestate_assign_eliminate]: 3.29001e-06
        [updatestate_loads_eliminate]: 2.87002e-06
        [parameter_eliminate]: 2.17999e-06
        [specialize_transform]: 5.81998e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 6.01998e-06
        [meta_shard_fg_expand]: 2.41e-06
        [get_grad_eliminate_]: 5.99e-06
        [merge_forward]: 4.3e-06
        [cell_reuse_recompute_pass]: 1.53002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.443e-05
        [j_node_and_user_rematch]: 1.054e-05
        [meta_fg_expand]: 2.36998e-06
        [replace_old_param]: 1.065e-05
        [inline_without_move]: 5.77999e-06
        [renormalize]: 0.00102217
        [add_forward_monad_depend]: 8.97e-06
        [auto_monad_grad]: 2.79001e-06
        [auto_monad_eliminator]: 1.811e-05
        [cse]: 3.206e-05
        [replace_applicator]: 1.853e-05
    [Cycle 2]: 0.00035626, [27]
        [switch_simplify]: 6.36e-06
        [loop_unroll]: 5.71e-06
        [a_1]: 9.777e-05
        [with_stream_mark]: 1.339e-05
        [recompute_prepare]: 4.85999e-06
        [updatestate_depend_eliminate]: 2.91e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.68002e-06
        [specialize_transform]: 5.56e-06
        [updatestate_useless_node_eliminater]: 5.27999e-06
        [accelerated_algorithm]: 5.79999e-06
        [meta_shard_fg_expand]: 1.77001e-06
        [get_grad_eliminate_]: 4.57998e-06
        [merge_forward]: 3.08e-06
        [cell_reuse_recompute_pass]: 2.55002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.588e-05
        [j_node_and_user_rematch]: 8.40999e-06
        [meta_fg_expand]: 2.18998e-06
        [replace_old_param]: 7.66999e-06
        [inline_without_move]: 4.92e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 5.62001e-06
        [cse]: 1.285e-05
        [replace_applicator]: 6.52001e-06
[py_interpret_to_execute_after_opt_a]: 1.448e-05
[rewriter_after_opt_a]: 4.035e-05
[convert_after_rewriter]: 1e-05
[order_py_execute_after_rewriter]: 5.22999e-06
[mutable_eliminate]: 0.00084297
[jit_opt_b]: 5.569e-05, [1]
    [Cycle 1]: 4.811e-05, [2]
        [frontend_op_eliminate]: 1.857e-05
        [inline_after_opt_a]: 1.609e-05
[cconv]: 3.404e-05
[loop_unroll]: 0.00050964
[jit_opt_after_cconv]: 0.00016221, [1]
    [Cycle 1]: 0.00015435, [11]
        [c_1]: 2.163e-05
        [parameter_eliminate]: 4.59998e-06
        [updatestate_depend_eliminate]: 7.63001e-06
        [updatestate_assign_eliminate]: 2.52001e-06
        [updatestate_loads_eliminate]: 2.35002e-06
        [cse]: 2.557e-05
        [call_graph_tuple_transform]: 2.339e-05
        [tuple_list_get_item_eliminator]: 5.43002e-06
        [none_parameter_eliminate]: 1.84998e-06
        [renormalize]: 1.42e-06
        [switch_simplify]: 5.90002e-06
[remove_dup_value]: 1.571e-05
[partial_unused_args_eliminate]: 2.34999e-06
[environ_conv]: 5.17e-06
[add_recomputation]: 5.512e-05
[cse_after_recomputation]: 2.196e-05, [1]
    [Cycle 1]: 1.602e-05, [1]
        [cse]: 9.71e-06
[auto_monad_reorder]: 1.658e-05
[get_jit_bprop_graph]: 2.29001e-06
[rewriter_after_jit_bprop_graph]: 7.82002e-06
[opt_after_jit_grad]: 0.00052553
[symbol_engine_optimizer]: 7.879e-05, [1]
    [Cycle 1]: 7.11e-05, [6]
        [build]: 4e-06
        [elim_shapecalc]: 7.78999e-06
        [elim_not_effective]: 1.328e-05
        [opt_reshape]: 6.29999e-06
        [fold_const_symbol]: 9.99999e-06
        [renormalize]: 9.39996e-07
[validate]: 4.427e-05
[backend_pass]: 8.59989e-07
[task_emit]: 0.00795086
[execute]: 1.09e-05
Sums
    bootstrap                            :     0.000370s :  1.94%
    type_inference                       :     0.005947s : 31.18%
    event_method                         :     0.000021s :  0.11%
    auto_monad                           :     0.000070s :  0.37%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000004s :  0.02%
    py_interpret_to_execute              :     0.000045s :  0.24%
    rewriter_before_opt_a                :     0.000078s :  0.41%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000080s :  0.42%
    jit_opt_a.loop_unroll                :     0.000037s :  0.19%
    jit_opt_a.a_1                        :     0.000780s :  4.09%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001022s :  5.36%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.12%
    jit_opt_a.cse                        :     0.000045s :  0.24%
    jit_opt_a.replace_applicator         :     0.000025s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000040s :  0.21%
    convert_after_rewriter               :     0.000010s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000843s :  4.42%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.000034s :  0.18%
    loop_unroll                          :     0.000510s :  2.67%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000055s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000526s :  2.76%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.23%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007951s : 41.69%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000295    23
 0.64% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000002s :      2: substitution.fold_const_symbol
 2.09% :     0.000006s :      3: substitution.graph_param_transform
86.38% :     0.000255s :      4: substitution.inline
 1.59% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.60% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000006s :      2: substitution.replace_old_param
 4.99% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005857     2
82.15% :     0.004811s :      1: type_inference.infer
17.85% :     0.001046s :      1: type_inference.specialize
------[replace.]   0.000089     6
68.13% :     0.000060s :      4: replace.inline
31.87% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000265     6
94.98% :     0.000252s :      4: match.inline
 5.02% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.60% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.30% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.20% :     0.000002s :     10: predicate.addn_zero_filter
 1.99% :     0.000003s :     10: predicate.arithmetic_simplify
 1.22% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.52% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.83% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 5.48% :     0.000007s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.45% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.27% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.06% :     0.000003s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.04% :     0.000003s :     15: predicate.partial_eliminate
 1.00% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.37% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.85% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.58% :     0.000002s :     10: predicate.reshape_eliminate
 1.11% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.22% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.18% :     0.000002s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.55% :     0.000003s :     16: predicate.switch_defer_inline
 1.93% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.12% :     0.000010s :     53: predicate.switch_simplify
 1.53% :     0.000002s :     10: predicate.tile_eliminate
 1.47% :     0.000002s :     10: predicate.transpose_eliminate
 1.69% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.13% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.44% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.55% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.84% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000733    12
44.34% :     0.000325s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.84% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.82% :     0.000350s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027130    76
 0.22% :     0.000059s :      1: add_recomputation
 0.27% :     0.000074s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.45% :     0.000393s :      1: bootstrap
 0.14% :     0.000037s :      1: cconv
 0.05% :     0.000012s :      1: convert_after_rewriter
 0.09% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000026s :      1: event_method
 0.06% :     0.000017s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
28.67% :     0.007779s :      1: jit_opt_a
 0.61% :     0.000165s :      1: jit_opt_after_cconv
 0.22% :     0.000059s :      1: jit_opt_b
 1.91% :     0.000518s :      1: loop_unroll
 3.16% :     0.000856s :      1: mutable_eliminate
 3.77% :     0.001022s :     26: opt.transform.jit_opt_a
 0.20% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.97% :     0.000535s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000048s :      1: py_interpret_to_execute
 0.06% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.03% :     0.000551s :      1: renormalize.infer
 1.70% :     0.000461s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000043s :      1: rewriter_after_opt_a
 0.30% :     0.000082s :      1: rewriter_before_opt_a
 0.30% :     0.000081s :      1: symbol_engine_optimizer
29.38% :     0.007972s :      1: task_emit
22.00% :     0.005967s :      1: type_inference
 0.24% :     0.000066s :      1: validate


TotalTime = 0.0176505, [33]
[bootstrap]: 0.00026765
[type_inference]: 0.00434673
[event_method]: 1.606e-05
[auto_monad]: 6.257e-05
[graph_reusing]: 5.46e-06
[pre_auto_parallel]: 2.37999e-06
[py_interpret_to_execute]: 3.353e-05
[rewriter_before_opt_a]: 7.053e-05
[expand_dump_flag]: 1.738e-05
[jit_opt_a]: 0.00500156, [2]
    [Cycle 1]: 0.00147644, [27]
        [switch_simplify]: 4.943e-05
        [loop_unroll]: 2.919e-05
        [a_1]: 0.00050505
        [with_stream_mark]: 1.583e-05
        [recompute_prepare]: 6.84999e-06
        [updatestate_depend_eliminate]: 3.63e-06
        [updatestate_assign_eliminate]: 3.00002e-06
        [updatestate_loads_eliminate]: 2.59001e-06
        [parameter_eliminate]: 1.60001e-06
        [specialize_transform]: 5.92001e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.24998e-06
        [meta_shard_fg_expand]: 2.34001e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 3.46999e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.329e-05
        [j_node_and_user_rematch]: 8.65001e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 8.47998e-06
        [inline_without_move]: 5.13002e-06
        [renormalize]: 0.0005442
        [add_forward_monad_depend]: 4.90999e-06
        [auto_monad_grad]: 2.35002e-06
        [auto_monad_eliminator]: 1.339e-05
        [cse]: 2.534e-05
        [replace_applicator]: 1.049e-05
    [Cycle 2]: 0.00033489, [27]
        [switch_simplify]: 5.48002e-06
        [loop_unroll]: 4.89e-06
        [a_1]: 8.647e-05
        [with_stream_mark]: 8.05999e-06
        [recompute_prepare]: 4.99998e-06
        [updatestate_depend_eliminate]: 2.74999e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 5.46e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 5.10999e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 4.23999e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.50999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.167e-05
        [j_node_and_user_rematch]: 7.7e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 6.94999e-06
        [inline_without_move]: 4.50001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.32999e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 5.00999e-06
        [cse]: 1.069e-05
        [replace_applicator]: 4.99998e-06
[py_interpret_to_execute_after_opt_a]: 8.37998e-06
[rewriter_after_opt_a]: 3.012e-05
[convert_after_rewriter]: 7.18998e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00047436
[jit_opt_b]: 7.316e-05, [1]
    [Cycle 1]: 6.601e-05, [2]
        [frontend_op_eliminate]: 3.824e-05
        [inline_after_opt_a]: 1.484e-05
[cconv]: 2.321e-05
[loop_unroll]: 0.0004461
[jit_opt_after_cconv]: 0.00013458, [1]
    [Cycle 1]: 0.00012814, [11]
        [c_1]: 2.077e-05
        [parameter_eliminate]: 2.52001e-06
        [updatestate_depend_eliminate]: 5.52001e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 1.686e-05
        [call_graph_tuple_transform]: 1.74e-05
        [tuple_list_get_item_eliminator]: 5.13002e-06
        [none_parameter_eliminate]: 1.07998e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 5.42999e-06
[remove_dup_value]: 1.283e-05
[partial_unused_args_eliminate]: 2.09e-06
[environ_conv]: 5.22e-06
[add_recomputation]: 4.382e-05
[cse_after_recomputation]: 1.928e-05, [1]
    [Cycle 1]: 1.414e-05, [1]
        [cse]: 8.55001e-06
[auto_monad_reorder]: 1.405e-05
[get_jit_bprop_graph]: 1.52001e-06
[rewriter_after_jit_bprop_graph]: 4.67e-06
[opt_after_jit_grad]: 0.00045226
[symbol_engine_optimizer]: 7.214e-05, [1]
    [Cycle 1]: 6.594e-05, [6]
        [build]: 2.96999e-06
        [elim_shapecalc]: 7.55e-06
        [elim_not_effective]: 1.243e-05
        [opt_reshape]: 5.47001e-06
        [fold_const_symbol]: 8.64998e-06
        [renormalize]: 3.9002e-07
[validate]: 3.014e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00578781
[execute]: 7e-06
Sums
    bootstrap                            :     0.000268s :  1.93%
    type_inference                       :     0.004347s : 31.41%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000063s :  0.45%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.24%
    rewriter_before_opt_a                :     0.000071s :  0.51%
    expand_dump_flag                     :     0.000017s :  0.13%
    jit_opt_a.switch_simplify            :     0.000055s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000592s :  4.27%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000544s :  3.93%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000036s :  0.26%
    jit_opt_a.replace_applicator         :     0.000015s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000030s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000474s :  3.43%
    jit_opt_b.frontend_op_eliminate      :     0.000038s :  0.28%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000446s :  3.22%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000044s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000452s :  3.27%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000030s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005788s : 41.82%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000183    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.83% :     0.000002s :      2: substitution.fold_const_symbol
 2.61% :     0.000005s :      3: substitution.graph_param_transform
83.41% :     0.000152s :      4: substitution.inline
 1.53% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.31% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000004s :      2: substitution.replace_old_param
 6.17% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004291     2
81.20% :     0.003484s :      1: type_inference.infer
18.80% :     0.000807s :      1: type_inference.specialize
------[replace.]   0.000064     6
65.70% :     0.000042s :      4: replace.inline
34.30% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000160     6
93.64% :     0.000150s :      4: match.inline
 6.36% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.26% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 1.90% :     0.000002s :     10: predicate.arithmetic_simplify
 1.44% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.51% :     0.000002s :     10: predicate.depend_value_elim
 1.28% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.31% :     0.000006s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.76% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.07% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.20% :     0.000001s :     10: predicate.minmaximum_grad
 1.94% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.23% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.19% :     0.000001s :      6: predicate.special_op_eliminate
 1.14% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.70% :     0.000009s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.22% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000522    12
49.68% :     0.000259s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.33% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.99% :     0.000235s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019111    76
 0.25% :     0.000047s :      1: add_recomputation
 0.34% :     0.000066s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.46% :     0.000279s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.10% :     0.000020s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.18% :     0.005004s :      1: jit_opt_a
 0.72% :     0.000137s :      1: jit_opt_after_cconv
 0.40% :     0.000076s :      1: jit_opt_b
 2.38% :     0.000454s :      1: loop_unroll
 2.53% :     0.000483s :      1: mutable_eliminate
 4.13% :     0.000789s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.24% :     0.000046s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.41% :     0.000460s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.52% :     0.000290s :      1: renormalize.infer
 1.29% :     0.000247s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000033s :      1: rewriter_after_opt_a
 0.38% :     0.000073s :      1: rewriter_before_opt_a
 0.39% :     0.000075s :      1: symbol_engine_optimizer
30.34% :     0.005798s :      1: task_emit
22.80% :     0.004357s :      1: type_inference
 0.24% :     0.000047s :      1: validate


TotalTime = 0.0215135, [33]
[bootstrap]: 0.0003251
[type_inference]: 0.00570591
[event_method]: 1.792e-05
[auto_monad]: 6.741e-05
[graph_reusing]: 5.91e-06
[pre_auto_parallel]: 2.53e-06
[py_interpret_to_execute]: 3.878e-05
[rewriter_before_opt_a]: 7.386e-05
[expand_dump_flag]: 3.18e-06
[jit_opt_a]: 0.0061015, [2]
    [Cycle 1]: 0.00194862, [27]
        [switch_simplify]: 5.983e-05
        [loop_unroll]: 2.906e-05
        [a_1]: 0.00054354
        [with_stream_mark]: 2.192e-05
        [recompute_prepare]: 8.32e-06
        [updatestate_depend_eliminate]: 3.93999e-06
        [updatestate_assign_eliminate]: 3.06999e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 5.92999e-06
        [updatestate_useless_node_eliminater]: 5.22999e-06
        [accelerated_algorithm]: 5.65001e-06
        [meta_shard_fg_expand]: 2.18002e-06
        [get_grad_eliminate_]: 5.31002e-06
        [merge_forward]: 3.98001e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.464e-05
        [j_node_and_user_rematch]: 8.97e-06
        [meta_fg_expand]: 2.21998e-06
        [replace_old_param]: 9.22001e-06
        [inline_without_move]: 5.48002e-06
        [renormalize]: 0.00071107
        [add_forward_monad_depend]: 5.67999e-06
        [auto_monad_grad]: 2.32999e-06
        [auto_monad_eliminator]: 1.489e-05
        [cse]: 3.029e-05
        [replace_applicator]: 0.00024309
    [Cycle 2]: 0.00037071, [27]
        [switch_simplify]: 9.02999e-06
        [loop_unroll]: 5.46998e-06
        [a_1]: 9.235e-05
        [with_stream_mark]: 1.761e-05
        [recompute_prepare]: 5.53002e-06
        [updatestate_depend_eliminate]: 3.85998e-06
        [updatestate_assign_eliminate]: 1.92001e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.28002e-06
        [specialize_transform]: 5.39e-06
        [updatestate_useless_node_eliminater]: 4.68001e-06
        [accelerated_algorithm]: 5.32999e-06
        [meta_shard_fg_expand]: 1.60001e-06
        [get_grad_eliminate_]: 4.79998e-06
        [merge_forward]: 2.97002e-06
        [cell_reuse_recompute_pass]: 2.21e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.678e-05
        [j_node_and_user_rematch]: 8.69003e-06
        [meta_fg_expand]: 2.17001e-06
        [replace_old_param]: 7.88999e-06
        [inline_without_move]: 4.68001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.64001e-06
        [auto_monad_grad]: 1.49998e-06
        [auto_monad_eliminator]: 7.20998e-06
        [cse]: 1.984e-05
        [replace_applicator]: 5.52999e-06
[py_interpret_to_execute_after_opt_a]: 1.501e-05
[rewriter_after_opt_a]: 3.782e-05
[convert_after_rewriter]: 7.61999e-06
[order_py_execute_after_rewriter]: 5.04e-06
[mutable_eliminate]: 0.00075653
[jit_opt_b]: 6.099e-05, [1]
    [Cycle 1]: 5.136e-05, [2]
        [frontend_op_eliminate]: 2.111e-05
        [inline_after_opt_a]: 1.706e-05
[cconv]: 3.173e-05
[loop_unroll]: 0.00049311
[jit_opt_after_cconv]: 0.00015431, [1]
    [Cycle 1]: 0.00014717, [11]
        [c_1]: 2.281e-05
        [parameter_eliminate]: 3.4e-06
        [updatestate_depend_eliminate]: 6.30002e-06
        [updatestate_assign_eliminate]: 2.40002e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [cse]: 2.637e-05
        [call_graph_tuple_transform]: 1.997e-05
        [tuple_list_get_item_eliminator]: 5.27999e-06
        [none_parameter_eliminate]: 1.62001e-06
        [renormalize]: 8.70001e-07
        [switch_simplify]: 5.61e-06
[remove_dup_value]: 1.461e-05
[partial_unused_args_eliminate]: 2.48998e-06
[environ_conv]: 5.55001e-06
[add_recomputation]: 5.682e-05
[cse_after_recomputation]: 2.217e-05, [1]
    [Cycle 1]: 1.618e-05, [1]
        [cse]: 9.91998e-06
[auto_monad_reorder]: 1.551e-05
[get_jit_bprop_graph]: 2.23998e-06
[rewriter_after_jit_bprop_graph]: 7.92998e-06
[opt_after_jit_grad]: 0.00052485
[symbol_engine_optimizer]: 7.886e-05, [1]
    [Cycle 1]: 7.091e-05, [6]
        [build]: 4.52e-06
        [elim_shapecalc]: 8.54e-06
        [elim_not_effective]: 1.333e-05
        [opt_reshape]: 6.17999e-06
        [fold_const_symbol]: 9.57001e-06
        [renormalize]: 6.00005e-07
[validate]: 4.218e-05
[backend_pass]: 1.25001e-06
[task_emit]: 0.00658977
[execute]: 9.71998e-06
Sums
    bootstrap                            :     0.000325s :  1.91%
    type_inference                       :     0.005706s : 33.48%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000067s :  0.40%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.23%
    rewriter_before_opt_a                :     0.000074s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000069s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.20%
    jit_opt_a.a_1                        :     0.000636s :  3.73%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000711s :  4.17%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.13%
    jit_opt_a.cse                        :     0.000050s :  0.29%
    jit_opt_a.replace_applicator         :     0.000249s :  1.46%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.09%
    rewriter_after_opt_a                 :     0.000038s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000757s :  4.44%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.10%
    cconv                                :     0.000032s :  0.19%
    loop_unroll                          :     0.000493s :  2.89%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000026s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000057s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000525s :  3.08%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006590s : 38.67%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000212    23
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000002s :      2: substitution.fold_const_symbol
 2.75% :     0.000006s :      3: substitution.graph_param_transform
83.84% :     0.000178s :      4: substitution.inline
 1.73% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.29% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.20% :     0.000005s :      2: substitution.replace_old_param
 5.42% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005631     2
84.03% :     0.004731s :      1: type_inference.infer
15.97% :     0.000899s :      1: type_inference.specialize
------[replace.]   0.000071     6
68.40% :     0.000048s :      4: replace.inline
31.60% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000185     6
94.37% :     0.000175s :      4: match.inline
 5.63% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.26% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.53% :     0.000002s :     10: predicate.addn_zero_filter
 2.55% :     0.000003s :     10: predicate.arithmetic_simplify
 1.29% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.54% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.11% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.39% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.45% :     0.000001s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.91% :     0.000007s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.37% :     0.000002s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.63% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.42% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 1.89% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     15: predicate.partial_eliminate
 1.08% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.78% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.28% :     0.000003s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.27% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.21% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.21% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.12% :     0.000010s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000002s :     10: predicate.transpose_eliminate
 1.74% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.16% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.34% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.54% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000633    12
50.52% :     0.000320s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.28% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.20% :     0.000280s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023427    76
 0.26% :     0.000060s :      1: add_recomputation
 0.30% :     0.000071s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.48% :     0.000347s :      1: bootstrap
 0.15% :     0.000035s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.06% :     0.006106s :      1: jit_opt_a
 0.67% :     0.000157s :      1: jit_opt_after_cconv
 0.27% :     0.000064s :      1: jit_opt_b
 2.14% :     0.000502s :      1: loop_unroll
 3.28% :     0.000768s :      1: mutable_eliminate
 4.65% :     0.001089s :     26: opt.transform.jit_opt_a
 0.21% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000031s :      4: opt.transform.jit_opt_b
 0.07% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.29% :     0.000536s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000042s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.53% :     0.000358s :      1: renormalize.infer
 1.47% :     0.000345s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000042s :      1: rewriter_after_opt_a
 0.33% :     0.000077s :      1: rewriter_before_opt_a
 0.35% :     0.000082s :      1: symbol_engine_optimizer
28.21% :     0.006609s :      1: task_emit
24.43% :     0.005724s :      1: type_inference
 0.28% :     0.000066s :      1: validate


TotalTime = 0.0131009, [30]
[bootstrap]: 0.00025138
[type_inference]: 0.0056882
[event_method]: 1.965e-05
[auto_monad]: 7.195e-05
[graph_reusing]: 5.20999e-06
[pre_auto_parallel]: 2.17999e-06
[py_interpret_to_execute]: 4.168e-05
[rewriter_before_opt_a]: 7.825e-05
[expand_dump_flag]: 3.06999e-06
[jit_opt_a]: 0.00493512, [2]
    [Cycle 1]: 0.00165153, [27]
        [switch_simplify]: 9.898e-05
        [loop_unroll]: 3.376e-05
        [a_1]: 0.00065684
        [with_stream_mark]: 1.413e-05
        [recompute_prepare]: 7.4e-06
        [updatestate_depend_eliminate]: 3.50998e-06
        [updatestate_assign_eliminate]: 2.55002e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [parameter_eliminate]: 1.71e-06
        [specialize_transform]: 5.90002e-06
        [updatestate_useless_node_eliminater]: 4.91997e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 1.89999e-06
        [get_grad_eliminate_]: 5.02e-06
        [merge_forward]: 3.31999e-06
        [cell_reuse_recompute_pass]: 8.50006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.32e-05
        [j_node_and_user_rematch]: 8.34002e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 8.51002e-06
        [inline_without_move]: 5.34998e-06
        [renormalize]: 0.0005527
        [add_forward_monad_depend]: 4.90001e-06
        [auto_monad_grad]: 2.26998e-06
        [auto_monad_eliminator]: 1.194e-05
        [cse]: 1.822e-05
        [replace_applicator]: 1.189e-05
    [Cycle 2]: 0.00030968, [27]
        [switch_simplify]: 5.87999e-06
        [loop_unroll]: 4.84e-06
        [a_1]: 8.757e-05
        [with_stream_mark]: 7.96001e-06
        [recompute_prepare]: 4.62998e-06
        [updatestate_depend_eliminate]: 2.31e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.74002e-06
        [updatestate_useless_node_eliminater]: 4.57e-06
        [accelerated_algorithm]: 4.74e-06
        [meta_shard_fg_expand]: 1.40001e-06
        [get_grad_eliminate_]: 4.47998e-06
        [merge_forward]: 2.48998e-06
        [cell_reuse_recompute_pass]: 1.39003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.308e-05
        [j_node_and_user_rematch]: 7.82998e-06
        [meta_fg_expand]: 1.57999e-06
        [replace_old_param]: 7.73999e-06
        [inline_without_move]: 5.20001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.26002e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 4.43999e-06
        [cse]: 9.82001e-06
        [replace_applicator]: 4.88001e-06
[py_interpret_to_execute_after_opt_a]: 8.46002e-06
[rewriter_after_opt_a]: 3.181e-05
[convert_after_rewriter]: 7.06001e-06
[order_py_execute_after_rewriter]: 5.02e-06
[mutable_eliminate]: 0.00047387
[jit_opt_b]: 4.685e-05, [1]
    [Cycle 1]: 4.025e-05, [2]
        [frontend_op_eliminate]: 1.526e-05
        [inline_after_opt_a]: 1.408e-05
[cconv]: 7.739e-05
[loop_unroll]: 0.00042357
[jit_opt_after_cconv]: 0.0001348, [1]
    [Cycle 1]: 0.00012838, [11]
        [c_1]: 2.055e-05
        [parameter_eliminate]: 2.66e-06
        [updatestate_depend_eliminate]: 5.45001e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [cse]: 1.622e-05
        [call_graph_tuple_transform]: 1.705e-05
        [tuple_list_get_item_eliminator]: 5.02999e-06
        [none_parameter_eliminate]: 1.73997e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 5.24e-06
[remove_dup_value]: 1.223e-05
[partial_unused_args_eliminate]: 2.02001e-06
[environ_conv]: 4.78001e-06
[add_recomputation]: 3.972e-05
[cse_after_recomputation]: 1.958e-05, [1]
    [Cycle 1]: 1.416e-05, [1]
        [cse]: 8.70999e-06
[auto_monad_reorder]: 1.422e-05
[get_jit_bprop_graph]: 1.49e-06
[rewriter_after_jit_bprop_graph]: 4.92999e-06
[opt_after_jit_grad]: 0.00044068
[symbol_engine_optimizer]: 6.995e-05, [1]
    [Cycle 1]: 6.413e-05, [6]
        [build]: 3.30998e-06
        [elim_shapecalc]: 7.1e-06
        [elim_not_effective]: 1.226e-05
        [opt_reshape]: 5.49e-06
        [fold_const_symbol]: 8.1e-06
        [renormalize]: 4.69998e-07
[validate]: 2.84e-05
Sums
    bootstrap                            :     0.000251s :  2.62%
    type_inference                       :     0.005688s : 59.39%
    event_method                         :     0.000020s :  0.21%
    auto_monad                           :     0.000072s :  0.75%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000042s :  0.44%
    rewriter_before_opt_a                :     0.000078s :  0.82%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000105s :  1.09%
    jit_opt_a.loop_unroll                :     0.000039s :  0.40%
    jit_opt_a.a_1                        :     0.000744s :  7.77%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.17%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000553s :  5.77%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.17%
    jit_opt_a.cse                        :     0.000028s :  0.29%
    jit_opt_a.replace_applicator         :     0.000017s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.09%
    rewriter_after_opt_a                 :     0.000032s :  0.33%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000474s :  4.95%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.15%
    cconv                                :     0.000077s :  0.81%
    loop_unroll                          :     0.000424s :  4.42%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000012s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000040s :  0.41%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000014s :  0.15%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000441s :  4.60%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.30%


Time group info:
------[substitution.]   0.000241    26
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 1.87% :     0.000005s :      3: substitution.graph_param_transform
84.88% :     0.000205s :      6: substitution.inline
 1.10% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.88% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.40% :     0.000003s :      2: substitution.replace_old_param
 2.97% :     0.000007s :      1: substitution.switch_simplify
 4.53% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005636     2
83.67% :     0.004715s :      1: type_inference.infer
16.33% :     0.000921s :      1: type_inference.specialize
------[replace.]   0.000120     9
52.04% :     0.000062s :      6: replace.inline
23.33% :     0.000028s :      1: replace.switch_simplify
24.63% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000217     9
92.54% :     0.000201s :      6: match.inline
 2.96% :     0.000006s :      1: match.switch_simplify
 4.50% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   801
 1.43% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.07% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.21% :     0.000002s :     12: predicate.addn_check_dump
 1.48% :     0.000002s :     12: predicate.addn_zero_filter
 1.90% :     0.000002s :     12: predicate.arithmetic_simplify
 1.38% :     0.000002s :     12: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.21% :     0.000002s :     12: predicate.compare_switch_simplify
 1.49% :     0.000002s :     12: predicate.depend_value_elim
 1.23% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.46% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.25% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.23% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.40% :     0.000002s :     12: predicate.environ_get_eliminate
 1.21% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.41% :     0.000007s :     26: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.64% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     17: predicate.load_eliminater
 1.23% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.23% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000002s :     12: predicate.merge_addn
 1.24% :     0.000002s :     12: predicate.minmaximum_grad
 1.47% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     17: predicate.partial_eliminate
 1.32% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     12: predicate.reduce_eliminate
 1.50% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     20: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     12: predicate.reshape_eliminate
 1.35% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     12: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.75% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.60% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.86% :     0.000004s :     20: predicate.switch_defer_inline
 2.51% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.24% :     0.000011s :     65: predicate.switch_simplify
 1.27% :     0.000002s :     12: predicate.tile_eliminate
 1.28% :     0.000002s :     12: predicate.transpose_eliminate
 1.44% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.66% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     12: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000753    15
52.63% :     0.000396s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.46% :     0.000079s :      1: func_graph_cloner_run.FuncGraphClonerNode
36.91% :     0.000278s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014714    72
 0.29% :     0.000043s :      1: add_recomputation
 0.51% :     0.000076s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.81% :     0.000267s :      1: bootstrap
 0.55% :     0.000081s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
33.56% :     0.004939s :      1: jit_opt_a
 0.93% :     0.000138s :      1: jit_opt_after_cconv
 0.34% :     0.000049s :      1: jit_opt_b
 2.94% :     0.000432s :      1: loop_unroll
 3.27% :     0.000482s :      1: mutable_eliminate
 6.79% :     0.000999s :     26: opt.transform.jit_opt_a
 0.30% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000023s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.05% :     0.000449s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.30% :     0.000044s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000014s :      1: remove_dup_value
 2.01% :     0.000295s :      1: renormalize.infer
 1.70% :     0.000251s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000035s :      1: rewriter_after_opt_a
 0.55% :     0.000081s :      1: rewriter_before_opt_a
 0.49% :     0.000072s :      1: symbol_engine_optimizer
38.73% :     0.005699s :      1: type_inference


TotalTime = 0.0170327, [33]
[bootstrap]: 0.00025082
[type_inference]: 0.00422425
[event_method]: 1.704e-05
[auto_monad]: 5.301e-05
[graph_reusing]: 5.09e-06
[pre_auto_parallel]: 2.44001e-06
[py_interpret_to_execute]: 2.934e-05
[rewriter_before_opt_a]: 6.898e-05
[expand_dump_flag]: 3.58e-06
[jit_opt_a]: 0.00473378, [2]
    [Cycle 1]: 0.00141958, [27]
        [switch_simplify]: 4.601e-05
        [loop_unroll]: 2.97e-05
        [a_1]: 0.00053773
        [with_stream_mark]: 1.453e-05
        [recompute_prepare]: 7.57998e-06
        [updatestate_depend_eliminate]: 3.20002e-06
        [updatestate_assign_eliminate]: 2.77002e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.59998e-06
        [specialize_transform]: 5.97001e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 1.81e-06
        [get_grad_eliminate_]: 5.30999e-06
        [merge_forward]: 3.00998e-06
        [cell_reuse_recompute_pass]: 1.12999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.322e-05
        [j_node_and_user_rematch]: 8.59e-06
        [meta_fg_expand]: 2.46e-06
        [replace_old_param]: 8.13001e-06
        [inline_without_move]: 5.17e-06
        [renormalize]: 0.00050341
        [add_forward_monad_depend]: 4.23001e-06
        [auto_monad_grad]: 1.96e-06
        [auto_monad_eliminator]: 1.322e-05
        [cse]: 1.801e-05
        [replace_applicator]: 1.043e-05
    [Cycle 2]: 0.00032096, [27]
        [switch_simplify]: 5.74e-06
        [loop_unroll]: 4.91997e-06
        [a_1]: 9.063e-05
        [with_stream_mark]: 8.02e-06
        [recompute_prepare]: 4.97e-06
        [updatestate_depend_eliminate]: 2.56e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 5.05001e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.12e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.45999e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.15e-05
        [j_node_and_user_rematch]: 7.98001e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 6.76999e-06
        [inline_without_move]: 4.71002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.54e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 5.22e-06
        [cse]: 1.048e-05
        [replace_applicator]: 5.64e-06
[py_interpret_to_execute_after_opt_a]: 8.03999e-06
[rewriter_after_opt_a]: 2.929e-05
[convert_after_rewriter]: 6.33e-06
[order_py_execute_after_rewriter]: 4.67e-06
[mutable_eliminate]: 0.00051353
[jit_opt_b]: 4.956e-05, [1]
    [Cycle 1]: 4.305e-05, [2]
        [frontend_op_eliminate]: 1.628e-05
        [inline_after_opt_a]: 1.486e-05
[cconv]: 2.182e-05
[loop_unroll]: 0.00041203
[jit_opt_after_cconv]: 0.00013632, [1]
    [Cycle 1]: 0.00013007, [11]
        [c_1]: 2.084e-05
        [parameter_eliminate]: 2.47001e-06
        [updatestate_depend_eliminate]: 5.05001e-06
        [updatestate_assign_eliminate]: 2.51998e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [cse]: 1.635e-05
        [call_graph_tuple_transform]: 1.84e-05
        [tuple_list_get_item_eliminator]: 5.23002e-06
        [none_parameter_eliminate]: 1.23002e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.45001e-06
[remove_dup_value]: 1.074e-05
[partial_unused_args_eliminate]: 1.97999e-06
[environ_conv]: 4.68001e-06
[add_recomputation]: 4.235e-05
[cse_after_recomputation]: 2.016e-05, [1]
    [Cycle 1]: 1.464e-05, [1]
        [cse]: 9.02999e-06
[auto_monad_reorder]: 1.258e-05
[get_jit_bprop_graph]: 1.52001e-06
[rewriter_after_jit_bprop_graph]: 4.95999e-06
[opt_after_jit_grad]: 0.0004474
[symbol_engine_optimizer]: 0.00011284, [1]
    [Cycle 1]: 0.0001066, [6]
        [build]: 2.96001e-06
        [elim_shapecalc]: 4.671e-05
        [elim_not_effective]: 1.279e-05
        [opt_reshape]: 5.86998e-06
        [fold_const_symbol]: 8.70999e-06
        [renormalize]: 3.89991e-07
[validate]: 2.899e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00560057
[execute]: 6.63e-06
Sums
    bootstrap                            :     0.000251s :  1.86%
    type_inference                       :     0.004224s : 31.35%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000053s :  0.39%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.22%
    rewriter_before_opt_a                :     0.000069s :  0.51%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000052s :  0.38%
    jit_opt_a.loop_unroll                :     0.000035s :  0.26%
    jit_opt_a.a_1                        :     0.000628s :  4.66%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000503s :  3.74%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.14%
    jit_opt_a.cse                        :     0.000028s :  0.21%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000029s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000514s :  3.81%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000022s :  0.16%
    loop_unroll                          :     0.000412s :  3.06%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000011s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000042s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000013s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000447s :  3.32%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000047s :  0.35%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005601s : 41.56%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000172    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000001s :      2: substitution.fold_const_symbol
 2.78% :     0.000005s :      3: substitution.graph_param_transform
83.54% :     0.000144s :      4: substitution.inline
 1.58% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.41% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.79% :     0.000003s :      2: substitution.replace_old_param
 6.11% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004173     2
80.20% :     0.003347s :      1: type_inference.infer
19.80% :     0.000826s :      1: type_inference.specialize
------[replace.]   0.000063     6
64.32% :     0.000040s :      4: replace.inline
35.68% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000151     6
93.91% :     0.000141s :      4: match.inline
 6.09% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.21% :     0.000001s :     10: predicate.addn_zero_filter
 1.94% :     0.000002s :     10: predicate.arithmetic_simplify
 1.38% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.50% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.86% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 4.98% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.45% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.79% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.93% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.51% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000000s :      3: predicate.opt_reshape
 2.18% :     0.000002s :     15: predicate.partial_eliminate
 1.53% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.82% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.65% :     0.000009s :     53: predicate.switch_simplify
 1.34% :     0.000002s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.50% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000492    12
46.70% :     0.000230s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.90% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.40% :     0.000233s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018509    76
 0.24% :     0.000045s :      1: add_recomputation
 0.30% :     0.000056s :      1: auto_monad
 0.08% :     0.000015s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.41% :     0.000261s :      1: bootstrap
 0.13% :     0.000025s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
25.59% :     0.004736s :      1: jit_opt_a
 0.75% :     0.000139s :      1: jit_opt_after_cconv
 0.28% :     0.000052s :      1: jit_opt_b
 2.27% :     0.000420s :      1: loop_unroll
 2.82% :     0.000522s :      1: mutable_eliminate
 4.46% :     0.000825s :     26: opt.transform.jit_opt_a
 0.25% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.38% :     0.000071s :      4: opt.transform.symbol_engine_opt
 2.46% :     0.000455s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000013s :      1: remove_dup_value
 1.36% :     0.000252s :      1: renormalize.infer
 1.32% :     0.000244s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000032s :      1: rewriter_after_opt_a
 0.39% :     0.000072s :      1: rewriter_before_opt_a
 0.62% :     0.000116s :      1: symbol_engine_optimizer
30.31% :     0.005610s :      1: task_emit
22.88% :     0.004234s :      1: type_inference
 0.24% :     0.000045s :      1: validate


TotalTime = 0.0231505, [33]
[bootstrap]: 0.00034425
[type_inference]: 0.00588587
[event_method]: 1.976e-05
[auto_monad]: 6.977e-05
[graph_reusing]: 6.31e-06
[pre_auto_parallel]: 2.79999e-06
[py_interpret_to_execute]: 4.274e-05
[rewriter_before_opt_a]: 8.011e-05
[expand_dump_flag]: 3.50998e-06
[jit_opt_a]: 0.00646612, [2]
    [Cycle 1]: 0.00212645, [27]
        [switch_simplify]: 6.56e-05
        [loop_unroll]: 3.092e-05
        [a_1]: 0.0007886
        [with_stream_mark]: 2.889e-05
        [recompute_prepare]: 1.116e-05
        [updatestate_depend_eliminate]: 4.23001e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 2.83003e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 6.16998e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 6.05002e-06
        [meta_shard_fg_expand]: 2.49001e-06
        [get_grad_eliminate_]: 5.77999e-06
        [merge_forward]: 3.68e-06
        [cell_reuse_recompute_pass]: 1.55001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.735e-05
        [j_node_and_user_rematch]: 9.73998e-06
        [meta_fg_expand]: 2.46998e-06
        [replace_old_param]: 1.107e-05
        [inline_without_move]: 5.66e-06
        [renormalize]: 0.00085233
        [add_forward_monad_depend]: 6.17001e-06
        [auto_monad_grad]: 2.39001e-06
        [auto_monad_eliminator]: 1.485e-05
        [cse]: 3.041e-05
        [replace_applicator]: 1.277e-05
    [Cycle 2]: 0.0003474, [27]
        [switch_simplify]: 5.67999e-06
        [loop_unroll]: 5.25001e-06
        [a_1]: 9.421e-05
        [with_stream_mark]: 9.54e-06
        [recompute_prepare]: 5.14998e-06
        [updatestate_depend_eliminate]: 2.86e-06
        [updatestate_assign_eliminate]: 1.98002e-06
        [updatestate_loads_eliminate]: 2.20002e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 5.25999e-06
        [updatestate_useless_node_eliminater]: 5.12999e-06
        [accelerated_algorithm]: 5.32999e-06
        [meta_shard_fg_expand]: 1.34998e-06
        [get_grad_eliminate_]: 4.38001e-06
        [merge_forward]: 2.22999e-06
        [cell_reuse_recompute_pass]: 2.07001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.267e-05
        [j_node_and_user_rematch]: 8.18001e-06
        [meta_fg_expand]: 1.57999e-06
        [replace_old_param]: 7.3e-06
        [inline_without_move]: 4.96997e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.39998e-06
        [auto_monad_grad]: 9.29984e-07
        [auto_monad_eliminator]: 5.82001e-06
        [cse]: 2.505e-05
        [replace_applicator]: 5.79e-06
[py_interpret_to_execute_after_opt_a]: 1.149e-05
[rewriter_after_opt_a]: 3.791e-05
[convert_after_rewriter]: 7.80998e-06
[order_py_execute_after_rewriter]: 5.73002e-06
[mutable_eliminate]: 0.00065581
[jit_opt_b]: 4.994e-05, [1]
    [Cycle 1]: 4.266e-05, [2]
        [frontend_op_eliminate]: 1.654e-05
        [inline_after_opt_a]: 1.46e-05
[cconv]: 2.667e-05
[loop_unroll]: 0.00044655
[jit_opt_after_cconv]: 0.0001439, [1]
    [Cycle 1]: 0.00013722, [11]
        [c_1]: 2.115e-05
        [parameter_eliminate]: 2.88e-06
        [updatestate_depend_eliminate]: 6.19999e-06
        [updatestate_assign_eliminate]: 2.52001e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 1.993e-05
        [call_graph_tuple_transform]: 1.902e-05
        [tuple_list_get_item_eliminator]: 5.40999e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 7.80012e-07
        [switch_simplify]: 5.09e-06
[remove_dup_value]: 1.45e-05
[partial_unused_args_eliminate]: 2.19001e-06
[environ_conv]: 5.84e-06
[add_recomputation]: 4.993e-05
[cse_after_recomputation]: 2.018e-05, [1]
    [Cycle 1]: 1.442e-05, [1]
        [cse]: 8.88002e-06
[auto_monad_reorder]: 1.754e-05
[get_jit_bprop_graph]: 2.32001e-06
[rewriter_after_jit_bprop_graph]: 5.24998e-06
[opt_after_jit_grad]: 0.00145502
[symbol_engine_optimizer]: 7.939e-05, [1]
    [Cycle 1]: 7.189e-05, [6]
        [build]: 4.95001e-06
        [elim_shapecalc]: 7.96001e-06
        [elim_not_effective]: 1.48e-05
        [opt_reshape]: 5.74999e-06
        [fold_const_symbol]: 8.97e-06
        [renormalize]: 1.07e-06
[validate]: 4.275e-05
[backend_pass]: 1.28002e-06
[task_emit]: 0.00690215
[execute]: 1.028e-05
Sums
    bootstrap                            :     0.000344s :  1.86%
    type_inference                       :     0.005886s : 31.84%
    event_method                         :     0.000020s :  0.11%
    auto_monad                           :     0.000070s :  0.38%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.23%
    rewriter_before_opt_a                :     0.000080s :  0.43%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000071s :  0.39%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000883s :  4.78%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000852s :  4.61%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.11%
    jit_opt_a.cse                        :     0.000055s :  0.30%
    jit_opt_a.replace_applicator         :     0.000019s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000038s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000656s :  3.55%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000027s :  0.14%
    loop_unroll                          :     0.000447s :  2.42%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000050s :  0.27%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000018s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.001455s :  7.87%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006902s : 37.33%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000232    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000001s :      2: substitution.fold_const_symbol
 2.46% :     0.000006s :      3: substitution.graph_param_transform
84.81% :     0.000196s :      4: substitution.inline
 1.52% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.90% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.19% :     0.000005s :      2: substitution.replace_old_param
 5.75% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005799     2
82.16% :     0.004765s :      1: type_inference.infer
17.84% :     0.001034s :      1: type_inference.specialize
------[replace.]   0.000083     6
67.01% :     0.000056s :      4: replace.inline
32.99% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000206     6
94.18% :     0.000194s :      4: match.inline
 5.82% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.54% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.48% :     0.000003s :     10: predicate.arithmetic_simplify
 1.10% :     0.000001s :     10: predicate.cast_eliminate
 0.56% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.13% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.08% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.36% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000000s :      3: predicate.graph_param_transform
 5.42% :     0.000007s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.36% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 2.07% :     0.000003s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.06% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.71% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.35% :     0.000002s :      6: predicate.specialize_transform
 1.51% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.64% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.53% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.57% :     0.000009s :     53: predicate.switch_simplify
 1.39% :     0.000002s :     10: predicate.tile_eliminate
 1.23% :     0.000002s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.02% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000700    12
50.12% :     0.000351s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.74% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.13% :     0.000316s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025217    76
 0.21% :     0.000053s :      1: add_recomputation
 0.29% :     0.000073s :      1: auto_monad
 0.08% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.44% :     0.000364s :      1: bootstrap
 0.12% :     0.000030s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
25.66% :     0.006470s :      1: jit_opt_a
 0.58% :     0.000147s :      1: jit_opt_after_cconv
 0.21% :     0.000053s :      1: jit_opt_b
 1.80% :     0.000455s :      1: loop_unroll
 2.64% :     0.000665s :      1: mutable_eliminate
 4.42% :     0.001114s :     26: opt.transform.jit_opt_a
 0.19% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000034s :      4: opt.transform.symbol_engine_opt
 5.82% :     0.001467s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000046s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.76% :     0.000443s :      1: renormalize.infer
 1.59% :     0.000401s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000041s :      1: rewriter_after_opt_a
 0.33% :     0.000084s :      1: rewriter_before_opt_a
 0.33% :     0.000082s :      1: symbol_engine_optimizer
27.44% :     0.006920s :      1: task_emit
23.42% :     0.005905s :      1: type_inference
 0.26% :     0.000066s :      1: validate


TotalTime = 0.0133745, [30]
[bootstrap]: 0.00024375
[type_inference]: 0.00582455
[event_method]: 2.007e-05
[auto_monad]: 7.773e-05
[graph_reusing]: 6.02999e-06
[pre_auto_parallel]: 2.37001e-06
[py_interpret_to_execute]: 4.34e-05
[rewriter_before_opt_a]: 8.364e-05
[expand_dump_flag]: 2.96001e-06
[jit_opt_a]: 0.00505716, [2]
    [Cycle 1]: 0.00174378, [27]
        [switch_simplify]: 0.00010894
        [loop_unroll]: 3.322e-05
        [a_1]: 0.00066603
        [with_stream_mark]: 1.721e-05
        [recompute_prepare]: 7.1e-06
        [updatestate_depend_eliminate]: 3.19001e-06
        [updatestate_assign_eliminate]: 3.26999e-06
        [updatestate_loads_eliminate]: 2.81999e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 5.92999e-06
        [updatestate_useless_node_eliminater]: 4.70999e-06
        [accelerated_algorithm]: 5.66e-06
        [meta_shard_fg_expand]: 2.18002e-06
        [get_grad_eliminate_]: 5.14e-06
        [merge_forward]: 3.45e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.378e-05
        [j_node_and_user_rematch]: 8.84998e-06
        [meta_fg_expand]: 2.24999e-06
        [replace_old_param]: 8.62998e-06
        [inline_without_move]: 5.31998e-06
        [renormalize]: 0.00060345
        [add_forward_monad_depend]: 5.42001e-06
        [auto_monad_grad]: 2.37999e-06
        [auto_monad_eliminator]: 1.368e-05
        [cse]: 2.58e-05
        [replace_applicator]: 1.176e-05
    [Cycle 2]: 0.000324, [27]
        [switch_simplify]: 5.59e-06
        [loop_unroll]: 4.93001e-06
        [a_1]: 9.54e-05
        [with_stream_mark]: 8.70001e-06
        [recompute_prepare]: 5.14998e-06
        [updatestate_depend_eliminate]: 2.51998e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 4.94e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 1.18001e-06
        [get_grad_eliminate_]: 4.85999e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 1.57999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.423e-05
        [j_node_and_user_rematch]: 8.23001e-06
        [meta_fg_expand]: 1.52999e-06
        [replace_old_param]: 7.51999e-06
        [inline_without_move]: 5.13002e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 4.69002e-06
        [cse]: 9.81e-06
        [replace_applicator]: 4.99998e-06
[py_interpret_to_execute_after_opt_a]: 8.33999e-06
[rewriter_after_opt_a]: 3.134e-05
[convert_after_rewriter]: 7.12002e-06
[order_py_execute_after_rewriter]: 5.00001e-06
[mutable_eliminate]: 0.00050207
[jit_opt_b]: 4.557e-05, [1]
    [Cycle 1]: 3.905e-05, [2]
        [frontend_op_eliminate]: 1.476e-05
        [inline_after_opt_a]: 1.345e-05
[cconv]: 2.487e-05
[loop_unroll]: 0.00044795
[jit_opt_after_cconv]: 0.00013262, [1]
    [Cycle 1]: 0.00012631, [11]
        [c_1]: 2.026e-05
        [parameter_eliminate]: 2.58e-06
        [updatestate_depend_eliminate]: 4.93001e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [cse]: 1.576e-05
        [call_graph_tuple_transform]: 1.745e-05
        [tuple_list_get_item_eliminator]: 5.05001e-06
        [none_parameter_eliminate]: 1.57999e-06
        [renormalize]: 4.2998e-07
        [switch_simplify]: 5.24e-06
[remove_dup_value]: 1.338e-05
[partial_unused_args_eliminate]: 2.34001e-06
[environ_conv]: 4.80999e-06
[add_recomputation]: 4.348e-05
[cse_after_recomputation]: 2.001e-05, [1]
    [Cycle 1]: 1.439e-05, [1]
        [cse]: 8.93002e-06
[auto_monad_reorder]: 1.392e-05
[get_jit_bprop_graph]: 1.64e-06
[rewriter_after_jit_bprop_graph]: 5.01002e-06
[opt_after_jit_grad]: 0.00043931
[symbol_engine_optimizer]: 6.969e-05, [1]
    [Cycle 1]: 6.35e-05, [6]
        [build]: 3.35e-06
        [elim_shapecalc]: 7.19001e-06
        [elim_not_effective]: 1.184e-05
        [opt_reshape]: 5.27001e-06
        [fold_const_symbol]: 8.50999e-06
        [renormalize]: 4.09986e-07
[validate]: 3.124e-05
Sums
    bootstrap                            :     0.000244s :  2.48%
    type_inference                       :     0.005825s : 59.31%
    event_method                         :     0.000020s :  0.20%
    auto_monad                           :     0.000078s :  0.79%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.44%
    rewriter_before_opt_a                :     0.000084s :  0.85%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000115s :  1.17%
    jit_opt_a.loop_unroll                :     0.000038s :  0.39%
    jit_opt_a.a_1                        :     0.000761s :  7.75%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000604s :  6.15%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.19%
    jit_opt_a.cse                        :     0.000036s :  0.36%
    jit_opt_a.replace_applicator         :     0.000017s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.08%
    rewriter_after_opt_a                 :     0.000031s :  0.32%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000502s :  5.11%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000013s :  0.14%
    cconv                                :     0.000025s :  0.25%
    loop_unroll                          :     0.000448s :  4.56%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000043s :  0.44%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000014s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000439s :  4.47%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.32%


Time group info:
------[substitution.]   0.000264    26
 0.67% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 1.81% :     0.000005s :      3: substitution.graph_param_transform
84.63% :     0.000224s :      6: substitution.inline
 1.14% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.94% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.42% :     0.000004s :      2: substitution.replace_old_param
 3.33% :     0.000009s :      1: substitution.switch_simplify
 4.61% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005769     2
83.30% :     0.004805s :      1: type_inference.infer
16.70% :     0.000964s :      1: type_inference.specialize
------[replace.]   0.000123     9
52.31% :     0.000064s :      6: replace.inline
25.01% :     0.000031s :      1: replace.switch_simplify
22.68% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000239     9
92.00% :     0.000220s :      6: match.inline
 3.36% :     0.000008s :      1: match.switch_simplify
 4.63% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   801
 1.34% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.06% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000002s :     12: predicate.addn_check_dump
 1.75% :     0.000002s :     12: predicate.addn_zero_filter
 1.83% :     0.000002s :     12: predicate.arithmetic_simplify
 1.30% :     0.000002s :     12: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.24% :     0.000002s :     12: predicate.compare_switch_simplify
 1.48% :     0.000002s :     12: predicate.depend_value_elim
 1.20% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     12: predicate.environ_get_eliminate
 1.42% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.33% :     0.000000s :      3: predicate.graph_param_transform
 5.62% :     0.000007s :     26: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     17: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000005s :     40: predicate.loop_unroll_before_grad
 2.00% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000002s :     12: predicate.merge_addn
 1.25% :     0.000002s :     12: predicate.minmaximum_grad
 1.57% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     17: predicate.partial_eliminate
 1.34% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     12: predicate.reduce_eliminate
 1.54% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     20: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     12: predicate.reshape_eliminate
 1.31% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.63% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.81% :     0.000004s :     20: predicate.switch_defer_inline
 2.45% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.08% :     0.000011s :     65: predicate.switch_simplify
 1.29% :     0.000002s :     12: predicate.tile_eliminate
 1.32% :     0.000002s :     12: predicate.transpose_eliminate
 1.47% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.55% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     12: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000705    15
48.57% :     0.000342s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.05% :     0.000085s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.38% :     0.000278s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015054    72
 0.31% :     0.000046s :      1: add_recomputation
 0.54% :     0.000081s :      1: auto_monad
 0.11% :     0.000016s :      1: auto_monad_reorder
 1.70% :     0.000256s :      1: bootstrap
 0.18% :     0.000027s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
33.62% :     0.005060s :      1: jit_opt_a
 0.90% :     0.000135s :      1: jit_opt_after_cconv
 0.32% :     0.000048s :      1: jit_opt_b
 3.03% :     0.000456s :      1: loop_unroll
 3.39% :     0.000510s :      1: mutable_eliminate
 6.83% :     0.001028s :     26: opt.transform.jit_opt_a
 0.30% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000023s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.97% :     0.000447s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.30% :     0.000046s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.25% :     0.000339s :      1: renormalize.infer
 1.71% :     0.000257s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000034s :      1: rewriter_after_opt_a
 0.57% :     0.000086s :      1: rewriter_before_opt_a
 0.48% :     0.000072s :      1: symbol_engine_optimizer
38.78% :     0.005837s :      1: type_inference


TotalTime = 0.0232899, [33]
[bootstrap]: 0.00050091
[type_inference]: 0.00536322
[event_method]: 1.819e-05
[auto_monad]: 6.881e-05
[graph_reusing]: 6.14999e-06
[pre_auto_parallel]: 2.78e-06
[py_interpret_to_execute]: 3.87e-05
[rewriter_before_opt_a]: 7.478e-05
[expand_dump_flag]: 3.03e-06
[jit_opt_a]: 0.00700231, [2]
    [Cycle 1]: 0.00223469, [27]
        [switch_simplify]: 7.522e-05
        [loop_unroll]: 3.106e-05
        [a_1]: 0.00065529
        [with_stream_mark]: 2.73e-05
        [recompute_prepare]: 8.22e-06
        [updatestate_depend_eliminate]: 4.37e-06
        [updatestate_assign_eliminate]: 2.79999e-06
        [updatestate_loads_eliminate]: 2.58998e-06
        [parameter_eliminate]: 2.46e-06
        [specialize_transform]: 6.28e-06
        [updatestate_useless_node_eliminater]: 5.74e-06
        [accelerated_algorithm]: 6.02001e-06
        [meta_shard_fg_expand]: 2.14999e-06
        [get_grad_eliminate_]: 5.54e-06
        [merge_forward]: 3.94002e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.558e-05
        [j_node_and_user_rematch]: 1.012e-05
        [meta_fg_expand]: 2.53e-06
        [replace_old_param]: 1.019e-05
        [inline_without_move]: 5.59e-06
        [renormalize]: 0.00106174
        [add_forward_monad_depend]: 8.01001e-06
        [auto_monad_grad]: 2.58e-06
        [auto_monad_eliminator]: 1.835e-05
        [cse]: 3.301e-05
        [replace_applicator]: 2.088e-05
    [Cycle 2]: 0.00036046, [27]
        [switch_simplify]: 5.85002e-06
        [loop_unroll]: 5.73002e-06
        [a_1]: 0.00010074
        [with_stream_mark]: 1.402e-05
        [recompute_prepare]: 5.44e-06
        [updatestate_depend_eliminate]: 3.58e-06
        [updatestate_assign_eliminate]: 2.54001e-06
        [updatestate_loads_eliminate]: 2.23002e-06
        [parameter_eliminate]: 1.39998e-06
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 5.59e-06
        [meta_shard_fg_expand]: 1.62999e-06
        [get_grad_eliminate_]: 4.67998e-06
        [merge_forward]: 3.14999e-06
        [cell_reuse_recompute_pass]: 2.68998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.599e-05
        [j_node_and_user_rematch]: 9.00001e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 9.05999e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.64e-06
        [auto_monad_grad]: 8.59989e-07
        [auto_monad_eliminator]: 5.67001e-06
        [cse]: 1.257e-05
        [replace_applicator]: 9.43002e-06
[py_interpret_to_execute_after_opt_a]: 1.457e-05
[rewriter_after_opt_a]: 3.844e-05
[convert_after_rewriter]: 7.56999e-06
[order_py_execute_after_rewriter]: 5.19e-06
[mutable_eliminate]: 0.00085472
[jit_opt_b]: 6.263e-05, [1]
    [Cycle 1]: 5.4e-05, [2]
        [frontend_op_eliminate]: 2.467e-05
        [inline_after_opt_a]: 1.753e-05
[cconv]: 3.46e-05
[loop_unroll]: 0.00048237
[jit_opt_after_cconv]: 0.00016567, [1]
    [Cycle 1]: 0.00015878, [11]
        [c_1]: 2.244e-05
        [parameter_eliminate]: 5.17999e-06
        [updatestate_depend_eliminate]: 9.05999e-06
        [updatestate_assign_eliminate]: 3.18e-06
        [updatestate_loads_eliminate]: 2.56998e-06
        [cse]: 2.766e-05
        [call_graph_tuple_transform]: 2.247e-05
        [tuple_list_get_item_eliminator]: 5.59e-06
        [none_parameter_eliminate]: 1.56002e-06
        [renormalize]: 8.80013e-07
        [switch_simplify]: 5.55001e-06
[remove_dup_value]: 1.608e-05
[partial_unused_args_eliminate]: 2.51e-06
[environ_conv]: 6.04999e-06
[add_recomputation]: 5.533e-05
[cse_after_recomputation]: 2.358e-05, [1]
    [Cycle 1]: 1.702e-05, [1]
        [cse]: 1.056e-05
[auto_monad_reorder]: 1.782e-05
[get_jit_bprop_graph]: 2.17999e-06
[rewriter_after_jit_bprop_graph]: 7.41001e-06
[opt_after_jit_grad]: 0.00051085
[symbol_engine_optimizer]: 7.996e-05, [1]
    [Cycle 1]: 7.208e-05, [6]
        [build]: 4.65001e-06
        [elim_shapecalc]: 8.25999e-06
        [elim_not_effective]: 1.438e-05
        [opt_reshape]: 6.28e-06
        [fold_const_symbol]: 8.90999e-06
        [renormalize]: 6.39993e-07
[validate]: 4.677e-05
[backend_pass]: 9.60019e-07
[task_emit]: 0.00752146
[execute]: 1.091e-05
Sums
    bootstrap                            :     0.000501s :  2.75%
    type_inference                       :     0.005363s : 29.49%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000069s :  0.38%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.21%
    rewriter_before_opt_a                :     0.000075s :  0.41%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000081s :  0.45%
    jit_opt_a.loop_unroll                :     0.000037s :  0.20%
    jit_opt_a.a_1                        :     0.000756s :  4.16%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001062s :  5.84%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.13%
    jit_opt_a.cse                        :     0.000046s :  0.25%
    jit_opt_a.replace_applicator         :     0.000030s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000855s :  4.70%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.10%
    cconv                                :     0.000035s :  0.19%
    loop_unroll                          :     0.000482s :  2.65%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000055s :  0.30%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000018s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000511s :  2.81%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000047s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007521s : 41.36%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000284    23
 0.68% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000002s :      2: substitution.fold_const_symbol
 2.14% :     0.000006s :      3: substitution.graph_param_transform
85.78% :     0.000244s :      4: substitution.inline
 1.37% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.90% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.38% :     0.000007s :      2: substitution.replace_old_param
 5.16% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005286     2
81.83% :     0.004326s :      1: type_inference.infer
18.17% :     0.000960s :      1: type_inference.specialize
------[replace.]   0.000085     6
68.63% :     0.000058s :      4: replace.inline
31.37% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000255     6
94.70% :     0.000241s :      4: match.inline
 5.30% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   691
 1.72% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.23% :     0.000002s :     10: predicate.addn_zero_filter
 2.67% :     0.000004s :     10: predicate.arithmetic_simplify
 1.23% :     0.000002s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.91% :     0.000001s :     10: predicate.compare_switch_simplify
 1.12% :     0.000001s :     10: predicate.depend_value_elim
 1.17% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.24% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.64% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.41% :     0.000001s :      3: predicate.graph_param_transform
 5.53% :     0.000007s :     22: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.28% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.53% :     0.000002s :     15: predicate.load_eliminater
 1.25% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.93% :     0.000001s :     10: predicate.merge_addn
 0.90% :     0.000001s :     10: predicate.minmaximum_grad
 1.70% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.01% :     0.000003s :     15: predicate.partial_eliminate
 1.32% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.70% :     0.000002s :     10: predicate.reshape_eliminate
 1.11% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.02% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.08% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.76% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.12% :     0.000003s :     16: predicate.switch_defer_inline
 1.99% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.83% :     0.000012s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.08% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.02% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.39% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.95% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.44% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.03% :     0.000003s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000768    12
44.76% :     0.000344s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.18% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.06% :     0.000369s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025472    76
 0.23% :     0.000058s :      1: add_recomputation
 0.28% :     0.000072s :      1: auto_monad
 0.08% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 2.03% :     0.000516s :      1: bootstrap
 0.15% :     0.000037s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000026s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000023s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.06% :     0.000015s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
27.51% :     0.007007s :      1: jit_opt_a
 0.66% :     0.000168s :      1: jit_opt_after_cconv
 0.26% :     0.000065s :      1: jit_opt_b
 1.93% :     0.000491s :      1: loop_unroll
 3.40% :     0.000866s :      1: mutable_eliminate
 3.95% :     0.001007s :     26: opt.transform.jit_opt_a
 0.21% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000034s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.04% :     0.000521s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000041s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 2.19% :     0.000559s :      1: renormalize.infer
 1.94% :     0.000493s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000041s :      1: rewriter_after_opt_a
 0.30% :     0.000078s :      1: rewriter_before_opt_a
 0.32% :     0.000082s :      1: symbol_engine_optimizer
29.61% :     0.007543s :      1: task_emit
21.12% :     0.005380s :      1: type_inference
 0.28% :     0.000071s :      1: validate


TotalTime = 0.0137196, [30]
[bootstrap]: 0.00027201
[type_inference]: 0.00620389
[event_method]: 2.032e-05
[auto_monad]: 7.943e-05
[graph_reusing]: 6.07999e-06
[pre_auto_parallel]: 2.54999e-06
[py_interpret_to_execute]: 4.368e-05
[rewriter_before_opt_a]: 8.3e-05
[expand_dump_flag]: 3.18e-06
[jit_opt_a]: 0.0050095, [2]
    [Cycle 1]: 0.00165492, [27]
        [switch_simplify]: 0.00011128
        [loop_unroll]: 3.328e-05
        [a_1]: 0.00064331
        [with_stream_mark]: 1.56e-05
        [recompute_prepare]: 6.95002e-06
        [updatestate_depend_eliminate]: 3.8e-06
        [updatestate_assign_eliminate]: 3.16001e-06
        [updatestate_loads_eliminate]: 2.67001e-06
        [parameter_eliminate]: 2.29001e-06
        [specialize_transform]: 5.62999e-06
        [updatestate_useless_node_eliminater]: 5.14998e-06
        [accelerated_algorithm]: 5.62999e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 5.69999e-06
        [merge_forward]: 3.45e-06
        [cell_reuse_recompute_pass]: 1.14998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.334e-05
        [j_node_and_user_rematch]: 8.85001e-06
        [meta_fg_expand]: 2.44999e-06
        [replace_old_param]: 8.23001e-06
        [inline_without_move]: 5.14e-06
        [renormalize]: 0.00054147
        [add_forward_monad_depend]: 5.15999e-06
        [auto_monad_grad]: 2.14e-06
        [auto_monad_eliminator]: 1.282e-05
        [cse]: 2.655e-05
        [replace_applicator]: 1.123e-05
    [Cycle 2]: 0.00032179, [27]
        [switch_simplify]: 5.42001e-06
        [loop_unroll]: 4.80999e-06
        [a_1]: 0.00010059
        [with_stream_mark]: 8.64e-06
        [recompute_prepare]: 5.07e-06
        [updatestate_depend_eliminate]: 2.39001e-06
        [updatestate_assign_eliminate]: 2.19999e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 4.85999e-06
        [updatestate_useless_node_eliminater]: 4.68001e-06
        [accelerated_algorithm]: 4.74e-06
        [meta_shard_fg_expand]: 1.05001e-06
        [get_grad_eliminate_]: 4.48999e-06
        [merge_forward]: 2.22999e-06
        [cell_reuse_recompute_pass]: 1.19998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.229e-05
        [j_node_and_user_rematch]: 7.16999e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 6.79999e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 7.90023e-07
        [auto_monad_eliminator]: 5.25999e-06
        [cse]: 1.034e-05
        [replace_applicator]: 5.09998e-06
[py_interpret_to_execute_after_opt_a]: 8.32e-06
[rewriter_after_opt_a]: 3.204e-05
[convert_after_rewriter]: 6.85002e-06
[order_py_execute_after_rewriter]: 5.12e-06
[mutable_eliminate]: 0.00046347
[jit_opt_b]: 5.555e-05, [1]
    [Cycle 1]: 4.844e-05, [2]
        [frontend_op_eliminate]: 1.475e-05
        [inline_after_opt_a]: 2.272e-05
[cconv]: 2.391e-05
[loop_unroll]: 0.00041848
[jit_opt_after_cconv]: 0.00014057, [1]
    [Cycle 1]: 0.00013382, [11]
        [c_1]: 2.264e-05
        [parameter_eliminate]: 2.59999e-06
        [updatestate_depend_eliminate]: 5.65001e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.26998e-06
        [cse]: 1.697e-05
        [call_graph_tuple_transform]: 1.817e-05
        [tuple_list_get_item_eliminator]: 5.14998e-06
        [none_parameter_eliminate]: 1.40001e-06
        [renormalize]: 5.3001e-07
        [switch_simplify]: 5.49e-06
[remove_dup_value]: 1.351e-05
[partial_unused_args_eliminate]: 2.06e-06
[environ_conv]: 4.72e-06
[add_recomputation]: 4.676e-05
[cse_after_recomputation]: 2.101e-05, [1]
    [Cycle 1]: 1.52e-05, [1]
        [cse]: 9.56003e-06
[auto_monad_reorder]: 1.456e-05
[get_jit_bprop_graph]: 1.53002e-06
[rewriter_after_jit_bprop_graph]: 5.41998e-06
[opt_after_jit_grad]: 0.00045951
[symbol_engine_optimizer]: 7.251e-05, [1]
    [Cycle 1]: 6.646e-05, [6]
        [build]: 4.05e-06
        [elim_shapecalc]: 7.33e-06
        [elim_not_effective]: 1.273e-05
        [opt_reshape]: 5.59e-06
        [fold_const_symbol]: 8.35001e-06
        [renormalize]: 7.7e-07
[validate]: 3.064e-05
Sums
    bootstrap                            :     0.000272s :  2.69%
    type_inference                       :     0.006204s : 61.31%
    event_method                         :     0.000020s :  0.20%
    auto_monad                           :     0.000079s :  0.78%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000044s :  0.43%
    rewriter_before_opt_a                :     0.000083s :  0.82%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000117s :  1.15%
    jit_opt_a.loop_unroll                :     0.000038s :  0.38%
    jit_opt_a.a_1                        :     0.000744s :  7.35%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000010s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000542s :  5.35%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.18%
    jit_opt_a.cse                        :     0.000037s :  0.36%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.08%
    rewriter_after_opt_a                 :     0.000032s :  0.32%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000463s :  4.58%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000023s :  0.22%
    cconv                                :     0.000024s :  0.24%
    loop_unroll                          :     0.000418s :  4.14%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000047s :  0.46%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000460s :  4.54%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000031s :  0.30%


Time group info:
------[substitution.]   0.000242    26
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.28% :     0.000006s :      3: substitution.graph_param_transform
83.62% :     0.000203s :      6: substitution.inline
 1.13% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.86% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.40% :     0.000003s :      2: substitution.replace_old_param
 3.34% :     0.000008s :      1: substitution.switch_simplify
 5.09% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006143     2
83.97% :     0.005158s :      1: type_inference.infer
16.03% :     0.000985s :      1: type_inference.specialize
------[replace.]   0.000125     9
50.18% :     0.000063s :      6: replace.inline
27.47% :     0.000034s :      1: replace.switch_simplify
22.35% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000218     9
91.50% :     0.000199s :      6: match.inline
 3.38% :     0.000007s :      1: match.switch_simplify
 5.13% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   801
 1.38% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.04% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     12: predicate.addn_check_dump
 1.44% :     0.000002s :     12: predicate.addn_zero_filter
 1.78% :     0.000002s :     12: predicate.arithmetic_simplify
 1.29% :     0.000002s :     12: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     12: predicate.compare_switch_simplify
 1.51% :     0.000002s :     12: predicate.depend_value_elim
 1.19% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     12: predicate.environ_get_eliminate
 1.24% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 6.05% :     0.000008s :     26: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     17: predicate.load_eliminater
 1.30% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.32% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.99% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     12: predicate.merge_addn
 1.15% :     0.000001s :     12: predicate.minmaximum_grad
 1.48% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     17: predicate.partial_eliminate
 1.28% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     12: predicate.reduce_eliminate
 1.55% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     20: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     12: predicate.reshape_eliminate
 1.23% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.71% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.68% :     0.000003s :     20: predicate.switch_defer_inline
 2.43% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.52% :     0.000011s :     65: predicate.switch_simplify
 1.26% :     0.000002s :     12: predicate.tile_eliminate
 1.23% :     0.000002s :     12: predicate.transpose_eliminate
 1.44% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.71% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     12: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000737    15
51.66% :     0.000381s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.44% :     0.000077s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.90% :     0.000279s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015334    72
 0.32% :     0.000049s :      1: add_recomputation
 0.54% :     0.000083s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.89% :     0.000290s :      1: bootstrap
 0.17% :     0.000027s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
32.69% :     0.005012s :      1: jit_opt_a
 0.93% :     0.000143s :      1: jit_opt_after_cconv
 0.38% :     0.000058s :      1: jit_opt_b
 2.78% :     0.000426s :      1: loop_unroll
 3.08% :     0.000472s :      1: mutable_eliminate
 6.57% :     0.001007s :     26: opt.transform.jit_opt_a
 0.31% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000031s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.05% :     0.000468s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.30% :     0.000046s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 1.91% :     0.000294s :      1: renormalize.infer
 1.57% :     0.000241s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000035s :      1: rewriter_after_opt_a
 0.56% :     0.000086s :      1: rewriter_before_opt_a
 0.49% :     0.000075s :      1: symbol_engine_optimizer
40.55% :     0.006218s :      1: type_inference


TotalTime = 0.0184188, [33]
[bootstrap]: 0.00028516
[type_inference]: 0.00446106
[event_method]: 1.785e-05
[auto_monad]: 6.08e-05
[graph_reusing]: 4.68001e-06
[pre_auto_parallel]: 2.07001e-06
[py_interpret_to_execute]: 3.497e-05
[rewriter_before_opt_a]: 6.952e-05
[expand_dump_flag]: 3.22002e-06
[jit_opt_a]: 0.00521449, [2]
    [Cycle 1]: 0.00148372, [27]
        [switch_simplify]: 4.977e-05
        [loop_unroll]: 2.891e-05
        [a_1]: 0.0004951
        [with_stream_mark]: 1.558e-05
        [recompute_prepare]: 7.63001e-06
        [updatestate_depend_eliminate]: 2.97002e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 5.61998e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 5.03002e-06
        [meta_shard_fg_expand]: 1.62001e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 2.89999e-06
        [cell_reuse_recompute_pass]: 8.70001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.074e-05
        [j_node_and_user_rematch]: 8.60001e-06
        [meta_fg_expand]: 2.49999e-06
        [replace_old_param]: 8.32003e-06
        [inline_without_move]: 5.07e-06
        [renormalize]: 0.00059918
        [add_forward_monad_depend]: 4.96002e-06
        [auto_monad_grad]: 1.59e-06
        [auto_monad_eliminator]: 1.17e-05
        [cse]: 1.902e-05
        [replace_applicator]: 1.155e-05
    [Cycle 2]: 0.00031459, [27]
        [switch_simplify]: 5.25001e-06
        [loop_unroll]: 4.63999e-06
        [a_1]: 8.889e-05
        [with_stream_mark]: 9.66e-06
        [recompute_prepare]: 5.14998e-06
        [updatestate_depend_eliminate]: 2.32001e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 4.87998e-06
        [updatestate_useless_node_eliminater]: 4.60999e-06
        [accelerated_algorithm]: 5.13002e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 4.58999e-06
        [merge_forward]: 2.51e-06
        [cell_reuse_recompute_pass]: 1.85001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.204e-05
        [j_node_and_user_rematch]: 7.78999e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 7.13998e-06
        [inline_without_move]: 4.55999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.06002e-06
        [auto_monad_grad]: 1.04e-06
        [auto_monad_eliminator]: 4.80001e-06
        [cse]: 1.014e-05
        [replace_applicator]: 4.71002e-06
[py_interpret_to_execute_after_opt_a]: 9.64e-06
[rewriter_after_opt_a]: 2.952e-05
[convert_after_rewriter]: 6.47001e-06
[order_py_execute_after_rewriter]: 4.82998e-06
[mutable_eliminate]: 0.00054181
[jit_opt_b]: 4.811e-05, [1]
    [Cycle 1]: 4.14e-05, [2]
        [frontend_op_eliminate]: 1.608e-05
        [inline_after_opt_a]: 1.441e-05
[cconv]: 1.974e-05
[loop_unroll]: 0.00042624
[jit_opt_after_cconv]: 0.00013381, [1]
    [Cycle 1]: 0.00012757, [11]
        [c_1]: 1.968e-05
        [parameter_eliminate]: 2.89999e-06
        [updatestate_depend_eliminate]: 5.62001e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 1.798e-05
        [call_graph_tuple_transform]: 1.684e-05
        [tuple_list_get_item_eliminator]: 5.22e-06
        [none_parameter_eliminate]: 9.00007e-07
        [renormalize]: 4.80009e-07
        [switch_simplify]: 5.26002e-06
[remove_dup_value]: 7.38e-06
[partial_unused_args_eliminate]: 1.82999e-06
[environ_conv]: 4.10998e-06
[add_recomputation]: 3.729e-05
[cse_after_recomputation]: 1.931e-05, [1]
    [Cycle 1]: 1.445e-05, [1]
        [cse]: 8.77e-06
[auto_monad_reorder]: 1.103e-05
[get_jit_bprop_graph]: 1.54998e-06
[rewriter_after_jit_bprop_graph]: 5.10999e-06
[opt_after_jit_grad]: 0.00049749
[symbol_engine_optimizer]: 7.387e-05, [1]
    [Cycle 1]: 6.705e-05, [6]
        [build]: 4.57998e-06
        [elim_shapecalc]: 7.45e-06
        [elim_not_effective]: 1.25e-05
        [opt_reshape]: 5.42001e-06
        [fold_const_symbol]: 7.87e-06
        [renormalize]: 2.40019e-07
[validate]: 3.078e-05
[backend_pass]: 8.80013e-07
[task_emit]: 0.00613607
[execute]: 9.87999e-06
Sums
    bootstrap                            :     0.000285s :  1.98%
    type_inference                       :     0.004461s : 30.99%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000061s :  0.42%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000035s :  0.24%
    rewriter_before_opt_a                :     0.000070s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000584s :  4.06%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000599s :  4.16%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.11%
    jit_opt_a.cse                        :     0.000029s :  0.20%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000030s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000542s :  3.76%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000020s :  0.14%
    loop_unroll                          :     0.000426s :  2.96%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000007s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000037s :  0.26%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000011s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000497s :  3.46%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006136s : 42.63%
    execute                              :     0.000010s :  0.07%


Time group info:
------[substitution.]   0.000176    23
 1.12% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000001s :      2: substitution.fold_const_symbol
 2.13% :     0.000004s :      3: substitution.graph_param_transform
83.91% :     0.000147s :      4: substitution.inline
 1.64% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.16% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.36% :     0.000004s :      2: substitution.replace_old_param
 6.16% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004389     2
82.04% :     0.003600s :      1: type_inference.infer
17.96% :     0.000788s :      1: type_inference.specialize
------[replace.]   0.000063     6
67.39% :     0.000043s :      4: replace.inline
32.61% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000154     6
93.66% :     0.000145s :      4: match.inline
 6.34% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.53% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.26% :     0.000003s :     10: predicate.arithmetic_simplify
 1.58% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.18% :     0.000001s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.28% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.45% :     0.000001s :      3: predicate.graph_param_transform
 5.17% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.95% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.15% :     0.000002s :     15: predicate.load_eliminater
 1.58% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.83% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.50% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000002s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.87% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.70% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.40% :     0.000000s :      3: predicate.reset_defer_inline
 1.70% :     0.000002s :     10: predicate.reshape_eliminate
 1.36% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.68% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.32% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.36% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.39% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.04% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000533    12
48.92% :     0.000261s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.97% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.11% :     0.000241s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019898    76
 0.20% :     0.000040s :      1: add_recomputation
 0.32% :     0.000065s :      1: auto_monad
 0.07% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.51% :     0.000301s :      1: bootstrap
 0.11% :     0.000022s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.08% :     0.000015s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
26.22% :     0.005218s :      1: jit_opt_a
 0.69% :     0.000137s :      1: jit_opt_after_cconv
 0.26% :     0.000051s :      1: jit_opt_b
 2.18% :     0.000434s :      1: loop_unroll
 2.77% :     0.000552s :      1: mutable_eliminate
 3.93% :     0.000783s :     26: opt.transform.jit_opt_a
 0.22% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.54% :     0.000505s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000038s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000010s :      1: remove_dup_value
 1.56% :     0.000311s :      1: renormalize.infer
 1.41% :     0.000281s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000032s :      1: rewriter_after_opt_a
 0.36% :     0.000072s :      1: rewriter_before_opt_a
 0.38% :     0.000076s :      1: symbol_engine_optimizer
30.93% :     0.006155s :      1: task_emit
22.51% :     0.004479s :      1: type_inference
 0.25% :     0.000049s :      1: validate


TotalTime = 0.0167552, [33]
[bootstrap]: 0.00027361
[type_inference]: 0.00411555
[event_method]: 1.619e-05
[auto_monad]: 5.394e-05
[graph_reusing]: 4.53001e-06
[pre_auto_parallel]: 2.31e-06
[py_interpret_to_execute]: 2.873e-05
[rewriter_before_opt_a]: 6.693e-05
[expand_dump_flag]: 2.88e-06
[jit_opt_a]: 0.00472695, [2]
    [Cycle 1]: 0.00146706, [27]
        [switch_simplify]: 4.988e-05
        [loop_unroll]: 3.056e-05
        [a_1]: 0.00050593
        [with_stream_mark]: 1.389e-05
        [recompute_prepare]: 7.18998e-06
        [updatestate_depend_eliminate]: 3.48999e-06
        [updatestate_assign_eliminate]: 2.59001e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 1.80001e-06
        [specialize_transform]: 5.96e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 5.72999e-06
        [meta_shard_fg_expand]: 1.76e-06
        [get_grad_eliminate_]: 5.24e-06
        [merge_forward]: 2.92002e-06
        [cell_reuse_recompute_pass]: 8.99978e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.298e-05
        [j_node_and_user_rematch]: 9.02999e-06
        [meta_fg_expand]: 2.12999e-06
        [replace_old_param]: 8.10999e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 0.00051302
        [add_forward_monad_depend]: 4.88001e-06
        [auto_monad_grad]: 1.99999e-06
        [auto_monad_eliminator]: 6.89e-05
        [cse]: 1.868e-05
        [replace_applicator]: 1.089e-05
    [Cycle 2]: 0.00033037, [27]
        [switch_simplify]: 5.89e-06
        [loop_unroll]: 5.20001e-06
        [a_1]: 9.211e-05
        [with_stream_mark]: 8.75001e-06
        [recompute_prepare]: 5.12e-06
        [updatestate_depend_eliminate]: 2.66e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 5.58002e-06
        [updatestate_useless_node_eliminater]: 5.13002e-06
        [accelerated_algorithm]: 4.99998e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 5.35999e-06
        [merge_forward]: 2.61e-06
        [cell_reuse_recompute_pass]: 1.91998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.245e-05
        [j_node_and_user_rematch]: 8.22e-06
        [meta_fg_expand]: 1.49998e-06
        [replace_old_param]: 7.41001e-06
        [inline_without_move]: 4.83001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.09998e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 5.76e-06
        [cse]: 1.007e-05
        [replace_applicator]: 5.42001e-06
[py_interpret_to_execute_after_opt_a]: 8.59e-06
[rewriter_after_opt_a]: 3.059e-05
[convert_after_rewriter]: 5.84e-06
[order_py_execute_after_rewriter]: 5.25001e-06
[mutable_eliminate]: 0.00048057
[jit_opt_b]: 5.024e-05, [1]
    [Cycle 1]: 4.347e-05, [2]
        [frontend_op_eliminate]: 1.67e-05
        [inline_after_opt_a]: 1.492e-05
[cconv]: 2.123e-05
[loop_unroll]: 0.00040957
[jit_opt_after_cconv]: 0.00013749, [1]
    [Cycle 1]: 0.00013146, [11]
        [c_1]: 2.112e-05
        [parameter_eliminate]: 2.73e-06
        [updatestate_depend_eliminate]: 5.24998e-06
        [updatestate_assign_eliminate]: 2.47001e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 1.683e-05
        [call_graph_tuple_transform]: 1.808e-05
        [tuple_list_get_item_eliminator]: 5.77001e-06
        [none_parameter_eliminate]: 9.20001e-07
        [renormalize]: 5.19998e-07
        [switch_simplify]: 5.40001e-06
[remove_dup_value]: 1.128e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 4.50001e-06
[add_recomputation]: 4.267e-05
[cse_after_recomputation]: 2.022e-05, [1]
    [Cycle 1]: 1.481e-05, [1]
        [cse]: 8.89003e-06
[auto_monad_reorder]: 1.361e-05
[get_jit_bprop_graph]: 1.57001e-06
[rewriter_after_jit_bprop_graph]: 4.43001e-06
[opt_after_jit_grad]: 0.00044
[symbol_engine_optimizer]: 7.288e-05, [1]
    [Cycle 1]: 6.645e-05, [6]
        [build]: 3.50998e-06
        [elim_shapecalc]: 7.51999e-06
        [elim_not_effective]: 1.32e-05
        [opt_reshape]: 5.90002e-06
        [fold_const_symbol]: 8.82e-06
        [renormalize]: 6.00005e-07
[validate]: 2.801e-05
[backend_pass]: 8.2e-07
[task_emit]: 0.00549078
[execute]: 6.41998e-06
Sums
    bootstrap                            :     0.000274s :  2.07%
    type_inference                       :     0.004116s : 31.07%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000054s :  0.41%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.22%
    rewriter_before_opt_a                :     0.000067s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.42%
    jit_opt_a.loop_unroll                :     0.000036s :  0.27%
    jit_opt_a.a_1                        :     0.000598s :  4.52%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.08%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000513s :  3.87%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000075s :  0.56%
    jit_opt_a.cse                        :     0.000029s :  0.22%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000481s :  3.63%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000021s :  0.16%
    loop_unroll                          :     0.000410s :  3.09%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000011s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000043s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000440s :  3.32%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000028s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005491s : 41.46%
    execute                              :     0.000006s :  0.05%


Time group info:
------[substitution.]   0.000174    23
 1.14% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000001s :      2: substitution.fold_const_symbol
 2.88% :     0.000005s :      3: substitution.graph_param_transform
83.12% :     0.000145s :      4: substitution.inline
 1.60% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.39% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.94% :     0.000003s :      2: substitution.replace_old_param
 6.18% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004063     2
81.85% :     0.003326s :      1: type_inference.infer
18.15% :     0.000737s :      1: type_inference.specialize
------[replace.]   0.000062     6
66.42% :     0.000041s :      4: replace.inline
33.58% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000152     6
93.69% :     0.000142s :      4: match.inline
 6.31% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.23% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 1.89% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.85% :     0.000007s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.96% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.10% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.83% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.67% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000001s :     10: predicate.same_eliminate
 0.63% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000001s :      6: predicate.special_op_eliminate
 1.15% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.46% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.72% :     0.000009s :     53: predicate.switch_simplify
 1.29% :     0.000002s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.47% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.55% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000499    12
47.88% :     0.000239s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.86% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.27% :     0.000231s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018180    76
 0.25% :     0.000046s :      1: add_recomputation
 0.32% :     0.000057s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.57% :     0.000285s :      1: bootstrap
 0.13% :     0.000024s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.02% :     0.004730s :      1: jit_opt_a
 0.77% :     0.000140s :      1: jit_opt_after_cconv
 0.29% :     0.000053s :      1: jit_opt_b
 2.29% :     0.000417s :      1: loop_unroll
 2.69% :     0.000489s :      1: mutable_eliminate
 4.42% :     0.000803s :     26: opt.transform.jit_opt_a
 0.26% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.46% :     0.000448s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000031s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 1.42% :     0.000259s :      1: renormalize.infer
 1.36% :     0.000248s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000034s :      1: rewriter_after_opt_a
 0.39% :     0.000070s :      1: rewriter_before_opt_a
 0.41% :     0.000075s :      1: symbol_engine_optimizer
30.25% :     0.005500s :      1: task_emit
22.70% :     0.004126s :      1: type_inference
 0.26% :     0.000047s :      1: validate


TotalTime = 0.0138024, [30]
[bootstrap]: 0.00024893
[type_inference]: 0.00583428
[event_method]: 1.928e-05
[auto_monad]: 7.628e-05
[graph_reusing]: 5.92999e-06
[pre_auto_parallel]: 2.62001e-06
[py_interpret_to_execute]: 4.49e-05
[rewriter_before_opt_a]: 8.236e-05
[expand_dump_flag]: 3.23998e-06
[jit_opt_a]: 0.00535823, [2]
    [Cycle 1]: 0.00181305, [27]
        [switch_simplify]: 0.00011206
        [loop_unroll]: 3.377e-05
        [a_1]: 0.0006714
        [with_stream_mark]: 1.773e-05
        [recompute_prepare]: 6.84999e-06
        [updatestate_depend_eliminate]: 3.61001e-06
        [updatestate_assign_eliminate]: 3.35e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 5.85002e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 5.45001e-06
        [meta_shard_fg_expand]: 2.02001e-06
        [get_grad_eliminate_]: 4.99e-06
        [merge_forward]: 3.56001e-06
        [cell_reuse_recompute_pass]: 1.30001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.456e-05
        [j_node_and_user_rematch]: 8.84e-06
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 8.17e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.00065617
        [add_forward_monad_depend]: 5.34e-06
        [auto_monad_grad]: 1.79998e-06
        [auto_monad_eliminator]: 1.47e-05
        [cse]: 2.788e-05
        [replace_applicator]: 1.126e-05
    [Cycle 2]: 0.00032749, [27]
        [switch_simplify]: 5.64998e-06
        [loop_unroll]: 5.49e-06
        [a_1]: 9.372e-05
        [with_stream_mark]: 8.17e-06
        [recompute_prepare]: 5.22999e-06
        [updatestate_depend_eliminate]: 2.71999e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 2.01003e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 5.13002e-06
        [updatestate_useless_node_eliminater]: 5.19998e-06
        [accelerated_algorithm]: 5.40001e-06
        [meta_shard_fg_expand]: 1.32999e-06
        [get_grad_eliminate_]: 5.04e-06
        [merge_forward]: 2.69001e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.337e-05
        [j_node_and_user_rematch]: 8.09002e-06
        [meta_fg_expand]: 1.49998e-06
        [replace_old_param]: 7.62002e-06
        [inline_without_move]: 5.39e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.24998e-06
        [auto_monad_grad]: 8.10018e-07
        [auto_monad_eliminator]: 4.99e-06
        [cse]: 9.71e-06
        [replace_applicator]: 5.22e-06
[py_interpret_to_execute_after_opt_a]: 9.00999e-06
[rewriter_after_opt_a]: 3.285e-05
[convert_after_rewriter]: 8.37e-06
[order_py_execute_after_rewriter]: 5.17999e-06
[mutable_eliminate]: 0.00053783
[jit_opt_b]: 5.08e-05, [1]
    [Cycle 1]: 4.394e-05, [2]
        [frontend_op_eliminate]: 1.635e-05
        [inline_after_opt_a]: 1.561e-05
[cconv]: 2.652e-05
[loop_unroll]: 0.00044213
[jit_opt_after_cconv]: 0.00014626, [1]
    [Cycle 1]: 0.00013916, [11]
        [c_1]: 2.338e-05
        [parameter_eliminate]: 2.49001e-06
        [updatestate_depend_eliminate]: 5.67001e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 1.776e-05
        [call_graph_tuple_transform]: 1.914e-05
        [tuple_list_get_item_eliminator]: 5.58002e-06
        [none_parameter_eliminate]: 1.39998e-06
        [renormalize]: 5.3001e-07
        [switch_simplify]: 5.69e-06
[remove_dup_value]: 1.364e-05
[partial_unused_args_eliminate]: 2.66e-06
[environ_conv]: 4.67e-06
[add_recomputation]: 4.858e-05
[cse_after_recomputation]: 2.177e-05, [1]
    [Cycle 1]: 1.606e-05, [1]
        [cse]: 1.01e-05
[auto_monad_reorder]: 1.571e-05
[get_jit_bprop_graph]: 2.12999e-06
[rewriter_after_jit_bprop_graph]: 6.21998e-06
[opt_after_jit_grad]: 0.00047694
[symbol_engine_optimizer]: 7.468e-05, [1]
    [Cycle 1]: 6.817e-05, [6]
        [build]: 3.88001e-06
        [elim_shapecalc]: 6.98998e-06
        [elim_not_effective]: 1.313e-05
        [opt_reshape]: 5.75001e-06
        [fold_const_symbol]: 8.85001e-06
        [renormalize]: 4.80009e-07
[validate]: 3.414e-05
Sums
    bootstrap                            :     0.000249s :  2.49%
    type_inference                       :     0.005834s : 58.35%
    event_method                         :     0.000019s :  0.19%
    auto_monad                           :     0.000076s :  0.76%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000045s :  0.45%
    rewriter_before_opt_a                :     0.000082s :  0.82%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000118s :  1.18%
    jit_opt_a.loop_unroll                :     0.000039s :  0.39%
    jit_opt_a.a_1                        :     0.000765s :  7.65%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.16%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000656s :  6.56%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.20%
    jit_opt_a.cse                        :     0.000038s :  0.38%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.09%
    rewriter_after_opt_a                 :     0.000033s :  0.33%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000538s :  5.38%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.16%
    cconv                                :     0.000027s :  0.27%
    loop_unroll                          :     0.000442s :  4.42%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000014s :  0.14%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000049s :  0.49%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000477s :  4.77%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.34%


Time group info:
------[substitution.]   0.000258    26
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 1.99% :     0.000005s :      3: substitution.graph_param_transform
84.71% :     0.000219s :      6: substitution.inline
 1.06% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.87% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.33% :     0.000003s :      2: substitution.replace_old_param
 3.12% :     0.000008s :      1: substitution.switch_simplify
 4.67% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005780     2
83.46% :     0.004824s :      1: type_inference.infer
16.54% :     0.000956s :      1: type_inference.specialize
------[replace.]   0.000126     9
50.63% :     0.000064s :      6: replace.inline
25.57% :     0.000032s :      1: replace.switch_simplify
23.79% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000234     9
92.23% :     0.000216s :      6: match.inline
 3.10% :     0.000007s :      1: match.switch_simplify
 4.67% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   801
 1.42% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.16% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000002s :     12: predicate.addn_check_dump
 1.56% :     0.000002s :     12: predicate.addn_zero_filter
 2.10% :     0.000003s :     12: predicate.arithmetic_simplify
 1.50% :     0.000002s :     12: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000002s :     12: predicate.compare_switch_simplify
 1.48% :     0.000002s :     12: predicate.depend_value_elim
 1.16% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     12: predicate.environ_get_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.78% :     0.000001s :      6: predicate.get_grad_eliminate
 0.19% :     0.000000s :      3: predicate.graph_param_transform
 5.71% :     0.000008s :     26: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000002s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.93% :     0.000003s :     17: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.91% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.29% :     0.000002s :     12: predicate.merge_addn
 1.18% :     0.000002s :     12: predicate.minmaximum_grad
 1.79% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     17: predicate.partial_eliminate
 1.19% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     12: predicate.reduce_eliminate
 1.49% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     20: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.53% :     0.000002s :     12: predicate.reshape_eliminate
 1.39% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     12: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.62% :     0.000004s :     20: predicate.switch_defer_inline
 2.27% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.16% :     0.000011s :     65: predicate.switch_simplify
 1.28% :     0.000002s :     12: predicate.tile_eliminate
 1.41% :     0.000002s :     12: predicate.transpose_eliminate
 1.49% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.52% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     12: predicate.value_based_eliminate
 0.42% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000711    15
48.55% :     0.000345s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.15% :     0.000086s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.30% :     0.000279s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015558    72
 0.33% :     0.000052s :      1: add_recomputation
 0.51% :     0.000080s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 1.69% :     0.000263s :      1: bootstrap
 0.19% :     0.000029s :      1: cconv
 0.07% :     0.000011s :      1: convert_after_rewriter
 0.15% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
34.46% :     0.005361s :      1: jit_opt_a
 0.96% :     0.000149s :      1: jit_opt_after_cconv
 0.34% :     0.000053s :      1: jit_opt_b
 2.90% :     0.000451s :      1: loop_unroll
 3.51% :     0.000547s :      1: mutable_eliminate
 6.65% :     0.001035s :     26: opt.transform.jit_opt_a
 0.32% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.12% :     0.000486s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000048s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.47% :     0.000384s :      1: renormalize.infer
 1.71% :     0.000266s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000036s :      1: rewriter_after_opt_a
 0.55% :     0.000085s :      1: rewriter_before_opt_a
 0.50% :     0.000077s :      1: symbol_engine_optimizer
37.58% :     0.005847s :      1: type_inference


TotalTime = 0.022398, [33]
[bootstrap]: 0.00033663
[type_inference]: 0.00523145
[event_method]: 1.727e-05
[auto_monad]: 7.088e-05
[graph_reusing]: 6.86999e-06
[pre_auto_parallel]: 2.61999e-06
[py_interpret_to_execute]: 4.099e-05
[rewriter_before_opt_a]: 7.816e-05
[expand_dump_flag]: 3.45e-06
[jit_opt_a]: 0.00657224, [2]
    [Cycle 1]: 0.00215076, [27]
        [switch_simplify]: 6.671e-05
        [loop_unroll]: 3.129e-05
        [a_1]: 0.00063955
        [with_stream_mark]: 2.355e-05
        [recompute_prepare]: 7.63001e-06
        [updatestate_depend_eliminate]: 4.08001e-06
        [updatestate_assign_eliminate]: 2.92002e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 6.39999e-06
        [updatestate_useless_node_eliminater]: 5.24998e-06
        [accelerated_algorithm]: 6.19001e-06
        [meta_shard_fg_expand]: 2.32999e-06
        [get_grad_eliminate_]: 5.65001e-06
        [merge_forward]: 3.58e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.396e-05
        [j_node_and_user_rematch]: 1.057e-05
        [meta_fg_expand]: 2.31998e-06
        [replace_old_param]: 1.047e-05
        [inline_without_move]: 2.341e-05
        [renormalize]: 0.00101395
        [add_forward_monad_depend]: 6.26e-06
        [auto_monad_grad]: 2.46998e-06
        [auto_monad_eliminator]: 1.641e-05
        [cse]: 3.389e-05
        [replace_applicator]: 1.596e-05
    [Cycle 2]: 0.00034761, [27]
        [switch_simplify]: 6.24999e-06
        [loop_unroll]: 5.51998e-06
        [a_1]: 0.00010501
        [with_stream_mark]: 1.245e-05
        [recompute_prepare]: 4.91002e-06
        [updatestate_depend_eliminate]: 2.99001e-06
        [updatestate_assign_eliminate]: 2.20002e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 5.00999e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.34e-06
        [meta_shard_fg_expand]: 1.60001e-06
        [get_grad_eliminate_]: 4.85001e-06
        [merge_forward]: 2.78998e-06
        [cell_reuse_recompute_pass]: 2.80997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.363e-05
        [j_node_and_user_rematch]: 8.27e-06
        [meta_fg_expand]: 1.78002e-06
        [replace_old_param]: 8.47e-06
        [inline_without_move]: 5.21002e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.45001e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 5.30999e-06
        [cse]: 1.123e-05
        [replace_applicator]: 4.84003e-06
[py_interpret_to_execute_after_opt_a]: 1.339e-05
[rewriter_after_opt_a]: 3.843e-05
[convert_after_rewriter]: 7.70998e-06
[order_py_execute_after_rewriter]: 5.29e-06
[mutable_eliminate]: 0.0008098
[jit_opt_b]: 5.583e-05, [1]
    [Cycle 1]: 4.813e-05, [2]
        [frontend_op_eliminate]: 1.889e-05
        [inline_after_opt_a]: 1.601e-05
[cconv]: 2.886e-05
[loop_unroll]: 0.00047417
[jit_opt_after_cconv]: 0.0001543, [1]
    [Cycle 1]: 0.0001476, [11]
        [c_1]: 2.54e-05
        [parameter_eliminate]: 2.31e-06
        [updatestate_depend_eliminate]: 5.75001e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.11998e-06
        [cse]: 2.351e-05
        [call_graph_tuple_transform]: 2.054e-05
        [tuple_list_get_item_eliminator]: 5.24e-06
        [none_parameter_eliminate]: 1.40999e-06
        [renormalize]: 1.07e-06
        [switch_simplify]: 5.78002e-06
[remove_dup_value]: 1.515e-05
[partial_unused_args_eliminate]: 2.27001e-06
[environ_conv]: 5.59998e-06
[add_recomputation]: 5.215e-05
[cse_after_recomputation]: 2.216e-05, [1]
    [Cycle 1]: 1.562e-05, [1]
        [cse]: 8.95001e-06
[auto_monad_reorder]: 1.697e-05
[get_jit_bprop_graph]: 1.76e-06
[rewriter_after_jit_bprop_graph]: 6.14999e-06
[opt_after_jit_grad]: 0.00049776
[symbol_engine_optimizer]: 8.202e-05, [1]
    [Cycle 1]: 7.413e-05, [6]
        [build]: 4.46002e-06
        [elim_shapecalc]: 8.56002e-06
        [elim_not_effective]: 1.456e-05
        [opt_reshape]: 6.78998e-06
        [fold_const_symbol]: 8.94e-06
        [renormalize]: 8.70001e-07
[validate]: 4.388e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00747462
[execute]: 8.85001e-06
Sums
    bootstrap                            :     0.000337s :  1.91%
    type_inference                       :     0.005231s : 29.61%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000071s :  0.40%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.23%
    rewriter_before_opt_a                :     0.000078s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000073s :  0.41%
    jit_opt_a.loop_unroll                :     0.000037s :  0.21%
    jit_opt_a.a_1                        :     0.000745s :  4.21%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000029s :  0.16%
    jit_opt_a.renormalize                :     0.001014s :  5.74%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.12%
    jit_opt_a.cse                        :     0.000045s :  0.26%
    jit_opt_a.replace_applicator         :     0.000021s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000810s :  4.58%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000029s :  0.16%
    loop_unroll                          :     0.000474s :  2.68%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000052s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000498s :  2.82%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007475s : 42.31%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000274    23
 0.71% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000002s :      2: substitution.fold_const_symbol
 2.19% :     0.000006s :      3: substitution.graph_param_transform
86.46% :     0.000237s :      4: substitution.inline
 1.40% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.55% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.92% :     0.000005s :      2: substitution.replace_old_param
 5.16% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005154     2
81.42% :     0.004196s :      1: type_inference.infer
18.58% :     0.000958s :      1: type_inference.specialize
------[replace.]   0.000088     6
68.18% :     0.000060s :      4: replace.inline
31.82% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000246     6
94.80% :     0.000234s :      4: match.inline
 5.20% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.53% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000002s :     10: predicate.addn_zero_filter
 2.61% :     0.000003s :     10: predicate.arithmetic_simplify
 1.29% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.91% :     0.000001s :     10: predicate.compare_switch_simplify
 1.13% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.25% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000001s :      3: predicate.graph_param_transform
 6.09% :     0.000008s :     22: predicate.inline
 1.06% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.96% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.39% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 1.80% :     0.000002s :      3: predicate.mutable_eliminate
 0.57% :     0.000001s :      3: predicate.opt_reshape
 1.99% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.56% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.51% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.27% :     0.000002s :     10: predicate.reshape_eliminate
 1.10% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.36% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000002s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 1.93% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.15% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000002s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.94% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.26% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000715    12
47.65% :     0.000341s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.73% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.61% :     0.000326s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024518    76
 0.22% :     0.000055s :      1: add_recomputation
 0.30% :     0.000075s :      1: auto_monad
 0.08% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.45% :     0.000355s :      1: bootstrap
 0.13% :     0.000031s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.08% :     0.000021s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.82% :     0.006577s :      1: jit_opt_a
 0.64% :     0.000157s :      1: jit_opt_after_cconv
 0.24% :     0.000059s :      1: jit_opt_b
 1.97% :     0.000482s :      1: loop_unroll
 3.34% :     0.000820s :      1: mutable_eliminate
 4.05% :     0.000994s :     26: opt.transform.jit_opt_a
 0.22% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000028s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.07% :     0.000506s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000044s :      1: py_interpret_to_execute
 0.06% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.25% :     0.000552s :      1: renormalize.infer
 1.85% :     0.000454s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000042s :      1: rewriter_after_opt_a
 0.33% :     0.000081s :      1: rewriter_before_opt_a
 0.35% :     0.000085s :      1: symbol_engine_optimizer
30.55% :     0.007490s :      1: task_emit
21.39% :     0.005244s :      1: type_inference
 0.27% :     0.000067s :      1: validate


TotalTime = 0.0150135, [30]
[bootstrap]: 0.00025547
[type_inference]: 0.00641311
[event_method]: 2.151e-05
[auto_monad]: 8.119e-05
[graph_reusing]: 7.31999e-06
[pre_auto_parallel]: 2.71e-06
[py_interpret_to_execute]: 4.545e-05
[rewriter_before_opt_a]: 8.575e-05
[expand_dump_flag]: 3.42997e-06
[jit_opt_a]: 0.00574347, [2]
    [Cycle 1]: 0.00198454, [27]
        [switch_simplify]: 0.00013465
        [loop_unroll]: 3.449e-05
        [a_1]: 0.00070169
        [with_stream_mark]: 2.146e-05
        [recompute_prepare]: 8.80001e-06
        [updatestate_depend_eliminate]: 3.95e-06
        [updatestate_assign_eliminate]: 3.06001e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 2.31e-06
        [specialize_transform]: 6.00002e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.14e-06
        [meta_shard_fg_expand]: 2.07999e-06
        [get_grad_eliminate_]: 5.07e-06
        [merge_forward]: 3.51999e-06
        [cell_reuse_recompute_pass]: 1.11002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.396e-05
        [j_node_and_user_rematch]: 8.92999e-06
        [meta_fg_expand]: 2.27999e-06
        [replace_old_param]: 8.87e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 0.00074185
        [add_forward_monad_depend]: 6.46e-06
        [auto_monad_grad]: 2.54001e-06
        [auto_monad_eliminator]: 1.467e-05
        [cse]: 3.139e-05
        [replace_applicator]: 1.389e-05
    [Cycle 2]: 0.00033183, [27]
        [switch_simplify]: 5.51e-06
        [loop_unroll]: 4.75001e-06
        [a_1]: 8.954e-05
        [with_stream_mark]: 1.151e-05
        [recompute_prepare]: 5.14998e-06
        [updatestate_depend_eliminate]: 2.61e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 5.59e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 4.85999e-06
        [meta_shard_fg_expand]: 1.55999e-06
        [get_grad_eliminate_]: 4.60001e-06
        [merge_forward]: 2.68e-06
        [cell_reuse_recompute_pass]: 1.92999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.48e-05
        [j_node_and_user_rematch]: 8.55999e-06
        [meta_fg_expand]: 1.64e-06
        [replace_old_param]: 7.61001e-06
        [inline_without_move]: 5.05999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.89999e-06
        [auto_monad_grad]: 9.30013e-07
        [auto_monad_eliminator]: 7.25e-06
        [cse]: 1.177e-05
        [replace_applicator]: 5.97001e-06
[py_interpret_to_execute_after_opt_a]: 1.335e-05
[rewriter_after_opt_a]: 3.804e-05
[convert_after_rewriter]: 9.09998e-06
[order_py_execute_after_rewriter]: 5.17e-06
[mutable_eliminate]: 0.00065791
[jit_opt_b]: 5.127e-05, [1]
    [Cycle 1]: 4.346e-05, [2]
        [frontend_op_eliminate]: 1.55e-05
        [inline_after_opt_a]: 1.526e-05
[cconv]: 3.123e-05
[loop_unroll]: 0.00046892
[jit_opt_after_cconv]: 0.0001492, [1]
    [Cycle 1]: 0.00014145, [11]
        [c_1]: 2.08e-05
        [parameter_eliminate]: 2.93e-06
        [updatestate_depend_eliminate]: 6.86001e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [cse]: 2.266e-05
        [call_graph_tuple_transform]: 2.207e-05
        [tuple_list_get_item_eliminator]: 5.07e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 5.25999e-06
[remove_dup_value]: 1.453e-05
[partial_unused_args_eliminate]: 2.17001e-06
[environ_conv]: 5.05001e-06
[add_recomputation]: 5.289e-05
[cse_after_recomputation]: 2.216e-05, [1]
    [Cycle 1]: 1.618e-05, [1]
        [cse]: 1.028e-05
[auto_monad_reorder]: 1.595e-05
[get_jit_bprop_graph]: 2.19001e-06
[rewriter_after_jit_bprop_graph]: 7.56001e-06
[opt_after_jit_grad]: 0.00047661
[symbol_engine_optimizer]: 7.28e-05, [1]
    [Cycle 1]: 6.667e-05, [6]
        [build]: 4.82e-06
        [elim_shapecalc]: 7.21001e-06
        [elim_not_effective]: 1.292e-05
        [opt_reshape]: 5.67999e-06
        [fold_const_symbol]: 8.22e-06
        [renormalize]: 3.89991e-07
[validate]: 3.852e-05
Sums
    bootstrap                            :     0.000255s :  2.34%
    type_inference                       :     0.006413s : 58.65%
    event_method                         :     0.000022s :  0.20%
    auto_monad                           :     0.000081s :  0.74%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000045s :  0.42%
    rewriter_before_opt_a                :     0.000086s :  0.78%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000140s :  1.28%
    jit_opt_a.loop_unroll                :     0.000039s :  0.36%
    jit_opt_a.a_1                        :     0.000791s :  7.24%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000742s :  6.78%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.20%
    jit_opt_a.cse                        :     0.000043s :  0.39%
    jit_opt_a.replace_applicator         :     0.000020s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.12%
    rewriter_after_opt_a                 :     0.000038s :  0.35%
    convert_after_rewriter               :     0.000009s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000658s :  6.02%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.14%
    cconv                                :     0.000031s :  0.29%
    loop_unroll                          :     0.000469s :  4.29%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000015s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000053s :  0.48%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.07%
    opt_after_jit_grad                   :     0.000477s :  4.36%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000039s :  0.35%


Time group info:
------[substitution.]   0.000298    26
 0.61% :     0.000002s :      2: substitution.elim_not_effective
 0.42% :     0.000001s :      2: substitution.fold_const_symbol
 2.16% :     0.000006s :      3: substitution.graph_param_transform
85.24% :     0.000254s :      6: substitution.inline
 1.14% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.52% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.71% :     0.000005s :      2: substitution.replace_old_param
 2.88% :     0.000009s :      1: substitution.switch_simplify
 4.31% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006345     2
82.96% :     0.005264s :      1: type_inference.infer
17.04% :     0.001081s :      1: type_inference.specialize
------[replace.]   0.000139     9
47.38% :     0.000066s :      6: replace.inline
30.48% :     0.000042s :      1: replace.switch_simplify
22.14% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000271     9
92.76% :     0.000251s :      6: match.inline
 2.89% :     0.000008s :      1: match.switch_simplify
 4.35% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   801
 1.36% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.93% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     12: predicate.addn_check_dump
 1.53% :     0.000002s :     12: predicate.addn_zero_filter
 2.19% :     0.000003s :     12: predicate.arithmetic_simplify
 1.28% :     0.000002s :     12: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000002s :     12: predicate.compare_switch_simplify
 1.30% :     0.000002s :     12: predicate.depend_value_elim
 1.14% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     12: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      3: predicate.elim_not_effective
 0.43% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.23% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     12: predicate.environ_get_eliminate
 1.10% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.79% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000001s :      3: predicate.graph_param_transform
 5.79% :     0.000008s :     26: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.99% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     17: predicate.load_eliminater
 1.90% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.52% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.76% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     12: predicate.merge_addn
 1.11% :     0.000001s :     12: predicate.minmaximum_grad
 2.02% :     0.000003s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     17: predicate.partial_eliminate
 1.16% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     12: predicate.reduce_eliminate
 1.39% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.52% :     0.000002s :     20: predicate.replace_applicator
 0.43% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.29% :     0.000002s :     12: predicate.reshape_eliminate
 1.23% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     12: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.89% :     0.000004s :     20: predicate.switch_defer_inline
 2.25% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.94% :     0.000012s :     65: predicate.switch_simplify
 1.17% :     0.000002s :     12: predicate.tile_eliminate
 1.30% :     0.000002s :     12: predicate.transpose_eliminate
 1.57% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.72% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.37% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.59% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     12: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000810    15
49.37% :     0.000400s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.30% :     0.000108s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.33% :     0.000302s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016903    72
 0.33% :     0.000056s :      1: add_recomputation
 0.50% :     0.000085s :      1: auto_monad
 0.11% :     0.000019s :      1: auto_monad_reorder
 1.61% :     0.000272s :      1: bootstrap
 0.20% :     0.000034s :      1: cconv
 0.07% :     0.000012s :      1: convert_after_rewriter
 0.14% :     0.000024s :      1: cse_after_recomputation
 0.26% :     0.000043s :      1: environ_conv
 0.16% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000010s :      1: graph_reusing
34.00% :     0.005748s :      1: jit_opt_a
 0.90% :     0.000152s :      1: jit_opt_after_cconv
 0.32% :     0.000054s :      1: jit_opt_b
 2.84% :     0.000479s :      1: loop_unroll
 3.97% :     0.000670s :      1: mutable_eliminate
 6.44% :     0.001089s :     26: opt.transform.jit_opt_a
 0.29% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.87% :     0.000485s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000048s :      1: py_interpret_to_execute
 0.09% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.50% :     0.000423s :      1: renormalize.infer
 1.83% :     0.000310s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000042s :      1: rewriter_after_opt_a
 0.52% :     0.000089s :      1: rewriter_before_opt_a
 0.45% :     0.000075s :      1: symbol_engine_optimizer
38.03% :     0.006429s :      1: type_inference


TotalTime = 0.0220024, [33]
[bootstrap]: 0.00040402
[type_inference]: 0.00604982
[event_method]: 1.766e-05
[auto_monad]: 6.775e-05
[graph_reusing]: 6.99001e-06
[pre_auto_parallel]: 2.88e-06
[py_interpret_to_execute]: 4.242e-05
[rewriter_before_opt_a]: 7.834e-05
[expand_dump_flag]: 3.08e-06
[jit_opt_a]: 0.00629027, [2]
    [Cycle 1]: 0.00202511, [27]
        [switch_simplify]: 6.245e-05
        [loop_unroll]: 3.141e-05
        [a_1]: 0.00058602
        [with_stream_mark]: 2.324e-05
        [recompute_prepare]: 8.87e-06
        [updatestate_depend_eliminate]: 3.57997e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.93e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 6.07001e-06
        [updatestate_useless_node_eliminater]: 5.23002e-06
        [accelerated_algorithm]: 6.52001e-06
        [meta_shard_fg_expand]: 2.09999e-06
        [get_grad_eliminate_]: 5.51998e-06
        [merge_forward]: 3.66001e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.44e-05
        [j_node_and_user_rematch]: 1.025e-05
        [meta_fg_expand]: 2.59999e-06
        [replace_old_param]: 9.72999e-06
        [inline_without_move]: 5.84e-06
        [renormalize]: 0.00098125
        [add_forward_monad_depend]: 6.61999e-06
        [auto_monad_grad]: 2.47001e-06
        [auto_monad_eliminator]: 1.5e-05
        [cse]: 3.048e-05
        [replace_applicator]: 1.366e-05
    [Cycle 2]: 0.00033787, [27]
        [switch_simplify]: 6.33e-06
        [loop_unroll]: 4.97999e-06
        [a_1]: 9.394e-05
        [with_stream_mark]: 1.016e-05
        [recompute_prepare]: 5.35001e-06
        [updatestate_depend_eliminate]: 2.68998e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 1.45001e-06
        [specialize_transform]: 4.99998e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.87999e-06
        [meta_shard_fg_expand]: 1.48002e-06
        [get_grad_eliminate_]: 4.55999e-06
        [merge_forward]: 2.63998e-06
        [cell_reuse_recompute_pass]: 2.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.35e-05
        [j_node_and_user_rematch]: 8.71002e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 7.61999e-06
        [inline_without_move]: 4.80001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.40001e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 5.94e-06
        [cse]: 1.19e-05
        [replace_applicator]: 4.84998e-06
[py_interpret_to_execute_after_opt_a]: 1.328e-05
[rewriter_after_opt_a]: 3.773e-05
[convert_after_rewriter]: 7.43999e-06
[order_py_execute_after_rewriter]: 5.42001e-06
[mutable_eliminate]: 0.0006628
[jit_opt_b]: 5.295e-05, [1]
    [Cycle 1]: 4.538e-05, [2]
        [frontend_op_eliminate]: 1.783e-05
        [inline_after_opt_a]: 1.549e-05
[cconv]: 2.714e-05
[loop_unroll]: 0.0004779
[jit_opt_after_cconv]: 0.00014729, [1]
    [Cycle 1]: 0.00014084, [11]
        [c_1]: 2.2e-05
        [parameter_eliminate]: 3.50998e-06
        [updatestate_depend_eliminate]: 5.88998e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.976e-05
        [call_graph_tuple_transform]: 2.1e-05
        [tuple_list_get_item_eliminator]: 5.43002e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 5.52999e-06
[remove_dup_value]: 1.588e-05
[partial_unused_args_eliminate]: 2.59999e-06
[environ_conv]: 5.62001e-06
[add_recomputation]: 5.406e-05
[cse_after_recomputation]: 2.226e-05, [1]
    [Cycle 1]: 1.654e-05, [1]
        [cse]: 1.031e-05
[auto_monad_reorder]: 1.633e-05
[get_jit_bprop_graph]: 2.09e-06
[rewriter_after_jit_bprop_graph]: 6.69999e-06
[opt_after_jit_grad]: 0.00047865
[symbol_engine_optimizer]: 7.434e-05, [1]
    [Cycle 1]: 6.779e-05, [6]
        [build]: 4.17998e-06
        [elim_shapecalc]: 7.85998e-06
        [elim_not_effective]: 1.281e-05
        [opt_reshape]: 5.72001e-06
        [fold_const_symbol]: 8.72e-06
        [renormalize]: 5.40022e-07
[validate]: 4.255e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.0066464
[execute]: 9.45001e-06
Sums
    bootstrap                            :     0.000404s :  2.32%
    type_inference                       :     0.006050s : 34.73%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000068s :  0.39%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000042s :  0.24%
    rewriter_before_opt_a                :     0.000078s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000069s :  0.39%
    jit_opt_a.loop_unroll                :     0.000036s :  0.21%
    jit_opt_a.a_1                        :     0.000680s :  3.90%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000981s :  5.63%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.12%
    jit_opt_a.cse                        :     0.000042s :  0.24%
    jit_opt_a.replace_applicator         :     0.000019s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000663s :  3.80%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000027s :  0.16%
    loop_unroll                          :     0.000478s :  2.74%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000054s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000479s :  2.75%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006646s : 38.15%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000242    23
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.22% :     0.000005s :      3: substitution.graph_param_transform
85.17% :     0.000206s :      4: substitution.inline
 1.83% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.87% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000005s :      2: substitution.replace_old_param
 5.54% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005972     2
83.03% :     0.004959s :      1: type_inference.infer
16.97% :     0.001014s :      1: type_inference.specialize
------[replace.]   0.000075     6
67.80% :     0.000051s :      4: replace.inline
32.20% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000215     6
94.32% :     0.000203s :      4: match.inline
 5.68% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.68% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.20% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 2.17% :     0.000003s :     10: predicate.arithmetic_simplify
 1.26% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.39% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.18% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.53% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.58% :     0.000001s :      3: predicate.graph_param_transform
 5.73% :     0.000007s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.51% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.96% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     10: predicate.reduce_eliminate
 1.57% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.25% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.27% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.77% :     0.000009s :     53: predicate.switch_simplify
 1.40% :     0.000002s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.80% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.81% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000726    12
49.18% :     0.000357s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.70% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.12% :     0.000320s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023986    76
 0.24% :     0.000057s :      1: add_recomputation
 0.30% :     0.000071s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.77% :     0.000425s :      1: bootstrap
 0.12% :     0.000030s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000023s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.24% :     0.006294s :      1: jit_opt_a
 0.63% :     0.000150s :      1: jit_opt_after_cconv
 0.23% :     0.000056s :      1: jit_opt_b
 2.03% :     0.000486s :      1: loop_unroll
 2.81% :     0.000674s :      1: mutable_eliminate
 3.77% :     0.000905s :     26: opt.transform.jit_opt_a
 0.21% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.03% :     0.000487s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000045s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 1.94% :     0.000465s :      1: renormalize.infer
 2.11% :     0.000507s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000041s :      1: rewriter_after_opt_a
 0.34% :     0.000081s :      1: rewriter_before_opt_a
 0.32% :     0.000078s :      1: symbol_engine_optimizer
27.79% :     0.006665s :      1: task_emit
25.30% :     0.006068s :      1: type_inference
 0.27% :     0.000065s :      1: validate


TotalTime = 0.0178728, [33]
[bootstrap]: 0.00027747
[type_inference]: 0.00432276
[event_method]: 1.675e-05
[auto_monad]: 6.408e-05
[graph_reusing]: 6.01e-06
[pre_auto_parallel]: 2.73003e-06
[py_interpret_to_execute]: 3.231e-05
[rewriter_before_opt_a]: 7.461e-05
[expand_dump_flag]: 3.01999e-06
[jit_opt_a]: 0.00491047, [2]
    [Cycle 1]: 0.00148044, [27]
        [switch_simplify]: 5.477e-05
        [loop_unroll]: 3.148e-05
        [a_1]: 0.00051186
        [with_stream_mark]: 1.636e-05
        [recompute_prepare]: 7.63999e-06
        [updatestate_depend_eliminate]: 3.85e-06
        [updatestate_assign_eliminate]: 2.84999e-06
        [updatestate_loads_eliminate]: 2.69999e-06
        [parameter_eliminate]: 1.77999e-06
        [specialize_transform]: 5.92001e-06
        [updatestate_useless_node_eliminater]: 4.96002e-06
        [accelerated_algorithm]: 5.49998e-06
        [meta_shard_fg_expand]: 1.89999e-06
        [get_grad_eliminate_]: 5.25001e-06
        [merge_forward]: 3.86999e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.473e-05
        [j_node_and_user_rematch]: 8.97999e-06
        [meta_fg_expand]: 2.18998e-06
        [replace_old_param]: 8.54e-06
        [inline_without_move]: 5.10999e-06
        [renormalize]: 0.00055359
        [add_forward_monad_depend]: 5.24e-06
        [auto_monad_grad]: 2.11e-06
        [auto_monad_eliminator]: 1.261e-05
        [cse]: 2.744e-05
        [replace_applicator]: 1.075e-05
    [Cycle 2]: 0.00031889, [27]
        [switch_simplify]: 5.67001e-06
        [loop_unroll]: 4.80999e-06
        [a_1]: 8.849e-05
        [with_stream_mark]: 9.00999e-06
        [recompute_prepare]: 5.42999e-06
        [updatestate_depend_eliminate]: 2.43998e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.34e-06
        [meta_shard_fg_expand]: 1.39998e-06
        [get_grad_eliminate_]: 4.49002e-06
        [merge_forward]: 2.40997e-06
        [cell_reuse_recompute_pass]: 1.92001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.161e-05
        [j_node_and_user_rematch]: 7.85e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 6.90998e-06
        [inline_without_move]: 4.53001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 6.99976e-07
        [auto_monad_eliminator]: 5.19e-06
        [cse]: 9.71998e-06
        [replace_applicator]: 5.72001e-06
[py_interpret_to_execute_after_opt_a]: 9.74e-06
[rewriter_after_opt_a]: 3.282e-05
[convert_after_rewriter]: 7.27002e-06
[order_py_execute_after_rewriter]: 4.98001e-06
[mutable_eliminate]: 0.00049456
[jit_opt_b]: 4.956e-05, [1]
    [Cycle 1]: 4.244e-05, [2]
        [frontend_op_eliminate]: 1.66e-05
        [inline_after_opt_a]: 1.439e-05
[cconv]: 2.391e-05
[loop_unroll]: 0.00041745
[jit_opt_after_cconv]: 0.00016455, [1]
    [Cycle 1]: 0.00015757, [11]
        [c_1]: 2.105e-05
        [parameter_eliminate]: 2.85002e-06
        [updatestate_depend_eliminate]: 5.02e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 1.586e-05
        [call_graph_tuple_transform]: 4.354e-05
        [tuple_list_get_item_eliminator]: 5.71e-06
        [none_parameter_eliminate]: 1.84e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 5.77001e-06
[remove_dup_value]: 1.388e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 5.27001e-06
[add_recomputation]: 4.749e-05
[cse_after_recomputation]: 2.053e-05, [1]
    [Cycle 1]: 1.526e-05, [1]
        [cse]: 9.39e-06
[auto_monad_reorder]: 1.542e-05
[get_jit_bprop_graph]: 1.94e-06
[rewriter_after_jit_bprop_graph]: 5.21002e-06
[opt_after_jit_grad]: 0.00048247
[symbol_engine_optimizer]: 7.455e-05, [1]
    [Cycle 1]: 6.758e-05, [6]
        [build]: 3.67002e-06
        [elim_shapecalc]: 7.65e-06
        [elim_not_effective]: 1.363e-05
        [opt_reshape]: 5.90002e-06
        [fold_const_symbol]: 8.48001e-06
        [renormalize]: 5.50004e-07
[validate]: 3.42e-05
[backend_pass]: 1.07e-06
[task_emit]: 0.00605805
[execute]: 7.71999e-06
Sums
    bootstrap                            :     0.000277s :  1.96%
    type_inference                       :     0.004323s : 30.51%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000064s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.23%
    rewriter_before_opt_a                :     0.000075s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000060s :  0.43%
    jit_opt_a.loop_unroll                :     0.000036s :  0.26%
    jit_opt_a.a_1                        :     0.000600s :  4.24%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000554s :  3.91%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000037s :  0.26%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000495s :  3.49%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000417s :  2.95%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000044s :  0.31%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000482s :  3.41%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006058s : 42.76%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000186    23
 1.14% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.73% :     0.000005s :      3: substitution.graph_param_transform
82.90% :     0.000154s :      4: substitution.inline
 1.70% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.05% :     0.000004s :      2: substitution.replace_old_param
 6.53% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004258     2
81.88% :     0.003486s :      1: type_inference.infer
18.12% :     0.000772s :      1: type_inference.specialize
------[replace.]   0.000065     6
67.33% :     0.000044s :      4: replace.inline
32.67% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000163     6
93.24% :     0.000152s :      4: match.inline
 6.76% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.24% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.51% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 1.82% :     0.000002s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.08% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.07% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.59% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.28% :     0.000001s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.87% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.36% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.73% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.09% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.87% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000001s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.66% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.14% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.37% :     0.000002s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.12% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000002s :     10: predicate.value_based_eliminate
 0.45% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000524    12
50.89% :     0.000267s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.97% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.14% :     0.000232s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019373    76
 0.26% :     0.000050s :      1: add_recomputation
 0.35% :     0.000067s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.50% :     0.000291s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.36% :     0.004913s :      1: jit_opt_a
 0.86% :     0.000168s :      1: jit_opt_after_cconv
 0.27% :     0.000052s :      1: jit_opt_b
 2.20% :     0.000425s :      1: loop_unroll
 2.60% :     0.000503s :      1: mutable_eliminate
 4.17% :     0.000809s :     26: opt.transform.jit_opt_a
 0.37% :     0.000072s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.54% :     0.000492s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000035s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.44% :     0.000280s :      1: renormalize.infer
 1.38% :     0.000267s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000038s :      1: rewriter_after_opt_a
 0.40% :     0.000077s :      1: rewriter_before_opt_a
 0.40% :     0.000077s :      1: symbol_engine_optimizer
31.32% :     0.006068s :      1: task_emit
22.39% :     0.004337s :      1: type_inference
 0.27% :     0.000053s :      1: validate


TotalTime = 0.0138134, [30]
[bootstrap]: 0.00025327
[type_inference]: 0.00584354
[event_method]: 2.044e-05
[auto_monad]: 7.641e-05
[graph_reusing]: 5.79e-06
[pre_auto_parallel]: 2.41998e-06
[py_interpret_to_execute]: 4.785e-05
[rewriter_before_opt_a]: 8.103e-05
[expand_dump_flag]: 2.71999e-06
[jit_opt_a]: 0.00539674, [2]
    [Cycle 1]: 0.00179199, [27]
        [switch_simplify]: 0.00011338
        [loop_unroll]: 3.403e-05
        [a_1]: 0.00066978
        [with_stream_mark]: 1.651e-05
        [recompute_prepare]: 7.01001e-06
        [updatestate_depend_eliminate]: 3.16999e-06
        [updatestate_assign_eliminate]: 2.71999e-06
        [updatestate_loads_eliminate]: 2.08002e-06
        [parameter_eliminate]: 1.55999e-06
        [specialize_transform]: 5.59e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.19e-06
        [meta_shard_fg_expand]: 1.96e-06
        [get_grad_eliminate_]: 5.22999e-06
        [merge_forward]: 3.26999e-06
        [cell_reuse_recompute_pass]: 8.99978e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.362e-05
        [j_node_and_user_rematch]: 8.79998e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 8.05999e-06
        [inline_without_move]: 5.22999e-06
        [renormalize]: 0.00066203
        [add_forward_monad_depend]: 5.57999e-06
        [auto_monad_grad]: 2.40002e-06
        [auto_monad_eliminator]: 1.161e-05
        [cse]: 1.699e-05
        [replace_applicator]: 1.13e-05
    [Cycle 2]: 0.00030608, [27]
        [switch_simplify]: 5.77001e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 8.814e-05
        [with_stream_mark]: 9.54e-06
        [recompute_prepare]: 4.80001e-06
        [updatestate_depend_eliminate]: 2.54001e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 9.40025e-07
        [specialize_transform]: 4.82e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 4.55001e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.48999e-06
        [merge_forward]: 2.42001e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.113e-05
        [j_node_and_user_rematch]: 8.03999e-06
        [meta_fg_expand]: 1.53997e-06
        [replace_old_param]: 6.98e-06
        [inline_without_move]: 4.77e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 1.05999e-06
        [auto_monad_eliminator]: 4.63999e-06
        [cse]: 9.51e-06
        [replace_applicator]: 4.84e-06
[py_interpret_to_execute_after_opt_a]: 1.003e-05
[rewriter_after_opt_a]: 3.377e-05
[convert_after_rewriter]: 6.42001e-06
[order_py_execute_after_rewriter]: 4.70001e-06
[mutable_eliminate]: 0.00055012
[jit_opt_b]: 5.646e-05, [1]
    [Cycle 1]: 4.983e-05, [2]
        [frontend_op_eliminate]: 1.527e-05
        [inline_after_opt_a]: 2.278e-05
[cconv]: 2.202e-05
[loop_unroll]: 0.00042218
[jit_opt_after_cconv]: 0.00013593, [1]
    [Cycle 1]: 0.00012948, [11]
        [c_1]: 2.031e-05
        [parameter_eliminate]: 2.36998e-06
        [updatestate_depend_eliminate]: 5.47999e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.702e-05
        [call_graph_tuple_transform]: 1.804e-05
        [tuple_list_get_item_eliminator]: 5.47999e-06
        [none_parameter_eliminate]: 1.15999e-06
        [renormalize]: 5.79981e-07
        [switch_simplify]: 5.47999e-06
[remove_dup_value]: 1.279e-05
[partial_unused_args_eliminate]: 1.98997e-06
[environ_conv]: 5.35999e-06
[add_recomputation]: 4.057e-05
[cse_after_recomputation]: 2.027e-05, [1]
    [Cycle 1]: 1.485e-05, [1]
        [cse]: 9.25999e-06
[auto_monad_reorder]: 1.405e-05
[get_jit_bprop_graph]: 1.82999e-06
[rewriter_after_jit_bprop_graph]: 5.22e-06
[opt_after_jit_grad]: 0.00048163
[symbol_engine_optimizer]: 7.182e-05, [1]
    [Cycle 1]: 6.59e-05, [6]
        [build]: 3.83999e-06
        [elim_shapecalc]: 7.6e-06
        [elim_not_effective]: 1.225e-05
        [opt_reshape]: 5.62001e-06
        [fold_const_symbol]: 8.01001e-06
        [renormalize]: 4.09986e-07
[validate]: 3.019e-05
Sums
    bootstrap                            :     0.000253s :  2.54%
    type_inference                       :     0.005844s : 58.65%
    event_method                         :     0.000020s :  0.21%
    auto_monad                           :     0.000076s :  0.77%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000048s :  0.48%
    rewriter_before_opt_a                :     0.000081s :  0.81%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000119s :  1.20%
    jit_opt_a.loop_unroll                :     0.000039s :  0.39%
    jit_opt_a.a_1                        :     0.000758s :  7.61%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000010s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000662s :  6.65%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.16%
    jit_opt_a.cse                        :     0.000026s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.10%
    rewriter_after_opt_a                 :     0.000034s :  0.34%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000550s :  5.52%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000023s :  0.23%
    cconv                                :     0.000022s :  0.22%
    loop_unroll                          :     0.000422s :  4.24%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000041s :  0.41%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000014s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000482s :  4.83%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000030s :  0.30%


Time group info:
------[substitution.]   0.000267    26
 0.68% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 1.96% :     0.000005s :      3: substitution.graph_param_transform
85.42% :     0.000228s :      6: substitution.inline
 1.24% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.55% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.26% :     0.000003s :      2: substitution.replace_old_param
 3.02% :     0.000008s :      1: substitution.switch_simplify
 4.40% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005789     2
83.08% :     0.004810s :      1: type_inference.infer
16.92% :     0.000980s :      1: type_inference.specialize
------[replace.]   0.000130     9
49.94% :     0.000065s :      6: replace.inline
25.99% :     0.000034s :      1: replace.switch_simplify
24.07% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000243     9
92.60% :     0.000225s :      6: match.inline
 3.02% :     0.000007s :      1: match.switch_simplify
 4.38% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   801
 1.38% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     12: predicate.addn_check_dump
 1.50% :     0.000002s :     12: predicate.addn_zero_filter
 1.95% :     0.000002s :     12: predicate.arithmetic_simplify
 1.36% :     0.000002s :     12: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     12: predicate.compare_switch_simplify
 1.45% :     0.000002s :     12: predicate.depend_value_elim
 1.22% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     12: predicate.environ_get_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.41% :     0.000007s :     26: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.96% :     0.000001s :      6: predicate.less_batch_normalization
 1.80% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     17: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.26% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     12: predicate.merge_addn
 1.15% :     0.000001s :     12: predicate.minmaximum_grad
 1.38% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.41% :     0.000003s :     17: predicate.partial_eliminate
 1.22% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     12: predicate.reduce_eliminate
 1.51% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     20: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     12: predicate.reshape_eliminate
 1.33% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     12: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.80% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.61% :     0.000003s :     20: predicate.switch_defer_inline
 2.40% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.46% :     0.000011s :     65: predicate.switch_simplify
 1.36% :     0.000002s :     12: predicate.tile_eliminate
 1.22% :     0.000002s :     12: predicate.transpose_eliminate
 1.44% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.38% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.64% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     12: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000730    15
45.97% :     0.000336s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.02% :     0.000095s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.01% :     0.000299s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015571    72
 0.28% :     0.000043s :      1: add_recomputation
 0.52% :     0.000080s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.70% :     0.000265s :      1: bootstrap
 0.16% :     0.000025s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
34.68% :     0.005400s :      1: jit_opt_a
 0.89% :     0.000138s :      1: jit_opt_after_cconv
 0.38% :     0.000059s :      1: jit_opt_b
 2.76% :     0.000430s :      1: loop_unroll
 3.59% :     0.000559s :      1: mutable_eliminate
 6.57% :     0.001024s :     26: opt.transform.jit_opt_a
 0.30% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000032s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.14% :     0.000489s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.32% :     0.000050s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000015s :      1: remove_dup_value
 2.39% :     0.000372s :      1: renormalize.infer
 1.82% :     0.000283s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000037s :      1: rewriter_after_opt_a
 0.54% :     0.000084s :      1: rewriter_before_opt_a
 0.48% :     0.000074s :      1: symbol_engine_optimizer
37.61% :     0.005857s :      1: type_inference


TotalTime = 0.0196355, [33]
[bootstrap]: 0.00031557
[type_inference]: 0.00469697
[event_method]: 1.699e-05
[auto_monad]: 6.166e-05
[graph_reusing]: 5.54e-06
[pre_auto_parallel]: 2.72001e-06
[py_interpret_to_execute]: 3.292e-05
[rewriter_before_opt_a]: 7.036e-05
[expand_dump_flag]: 2.99001e-06
[jit_opt_a]: 0.00532516, [2]
    [Cycle 1]: 0.00162877, [27]
        [switch_simplify]: 5.666e-05
        [loop_unroll]: 2.851e-05
        [a_1]: 0.00052744
        [with_stream_mark]: 1.987e-05
        [recompute_prepare]: 7.51999e-06
        [updatestate_depend_eliminate]: 3.93001e-06
        [updatestate_assign_eliminate]: 2.85002e-06
        [updatestate_loads_eliminate]: 2.89999e-06
        [parameter_eliminate]: 2.04999e-06
        [specialize_transform]: 5.86e-06
        [updatestate_useless_node_eliminater]: 4.83001e-06
        [accelerated_algorithm]: 5.32999e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 5.01002e-06
        [merge_forward]: 3.53e-06
        [cell_reuse_recompute_pass]: 1.50999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.332e-05
        [j_node_and_user_rematch]: 8.79e-06
        [meta_fg_expand]: 2.11e-06
        [replace_old_param]: 8.71997e-06
        [inline_without_move]: 5.06997e-06
        [renormalize]: 0.00066734
        [add_forward_monad_depend]: 6.18998e-06
        [auto_monad_grad]: 2.44001e-06
        [auto_monad_eliminator]: 1.358e-05
        [cse]: 2.849e-05
        [replace_applicator]: 1.411e-05
    [Cycle 2]: 0.00033179, [27]
        [switch_simplify]: 5.83002e-06
        [loop_unroll]: 5.44e-06
        [a_1]: 9.132e-05
        [with_stream_mark]: 1.202e-05
        [recompute_prepare]: 5.57001e-06
        [updatestate_depend_eliminate]: 2.47001e-06
        [updatestate_assign_eliminate]: 2.28998e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 4.78001e-06
        [updatestate_useless_node_eliminater]: 4.80999e-06
        [accelerated_algorithm]: 5.15001e-06
        [meta_shard_fg_expand]: 1.49998e-06
        [get_grad_eliminate_]: 4.62998e-06
        [merge_forward]: 2.74001e-06
        [cell_reuse_recompute_pass]: 1.89e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.544e-05
        [j_node_and_user_rematch]: 8.59e-06
        [meta_fg_expand]: 1.77999e-06
        [replace_old_param]: 1.296e-05
        [inline_without_move]: 4.44002e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.21002e-06
        [auto_monad_grad]: 9.10019e-07
        [auto_monad_eliminator]: 5.69e-06
        [cse]: 1.03e-05
        [replace_applicator]: 4.97e-06
[py_interpret_to_execute_after_opt_a]: 1.308e-05
[rewriter_after_opt_a]: 3.649e-05
[convert_after_rewriter]: 6.98e-06
[order_py_execute_after_rewriter]: 5.22e-06
[mutable_eliminate]: 0.00071011
[jit_opt_b]: 5.282e-05, [1]
    [Cycle 1]: 4.419e-05, [2]
        [frontend_op_eliminate]: 1.724e-05
        [inline_after_opt_a]: 1.498e-05
[cconv]: 2.564e-05
[loop_unroll]: 0.00044209
[jit_opt_after_cconv]: 0.0001465, [1]
    [Cycle 1]: 0.00014012, [11]
        [c_1]: 2.115e-05
        [parameter_eliminate]: 2.69999e-06
        [updatestate_depend_eliminate]: 5.52999e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 2.354e-05
        [call_graph_tuple_transform]: 2.024e-05
        [tuple_list_get_item_eliminator]: 5.60001e-06
        [none_parameter_eliminate]: 1.33002e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.42999e-06
[remove_dup_value]: 1.457e-05
[partial_unused_args_eliminate]: 2.01e-06
[environ_conv]: 5.40999e-06
[add_recomputation]: 5.225e-05
[cse_after_recomputation]: 2.08e-05, [1]
    [Cycle 1]: 1.536e-05, [1]
        [cse]: 9.36e-06
[auto_monad_reorder]: 1.479e-05
[get_jit_bprop_graph]: 2.32999e-06
[rewriter_after_jit_bprop_graph]: 7.05002e-06
[opt_after_jit_grad]: 0.00058791
[symbol_engine_optimizer]: 7.915e-05, [1]
    [Cycle 1]: 7.15e-05, [6]
        [build]: 4.79e-06
        [elim_shapecalc]: 8.75999e-06
        [elim_not_effective]: 1.417e-05
        [opt_reshape]: 5.77001e-06
        [fold_const_symbol]: 8.77e-06
        [renormalize]: 5.89993e-07
[validate]: 3.858e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00662347
[execute]: 7.85998e-06
Sums
    bootstrap                            :     0.000316s :  2.02%
    type_inference                       :     0.004697s : 30.02%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000062s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.21%
    rewriter_before_opt_a                :     0.000070s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000619s :  3.95%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000667s :  4.27%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000039s :  0.25%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000036s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000710s :  4.54%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000026s :  0.16%
    loop_unroll                          :     0.000442s :  2.83%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000052s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.05%
    opt_after_jit_grad                   :     0.000588s :  3.76%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006623s : 42.33%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000213    23
 1.17% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000002s :      2: substitution.fold_const_symbol
 2.58% :     0.000005s :      3: substitution.graph_param_transform
81.67% :     0.000174s :      4: substitution.inline
 1.79% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000005s :      4: substitution.remove_not_recompute_node
 4.54% :     0.000010s :      2: substitution.replace_old_param
 5.31% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004629     2
81.41% :     0.003769s :      1: type_inference.infer
18.59% :     0.000861s :      1: type_inference.specialize
------[replace.]   0.000065     6
68.78% :     0.000045s :      4: replace.inline
31.22% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000181     6
94.34% :     0.000171s :      4: match.inline
 5.66% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.23% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.51% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 2.36% :     0.000003s :     10: predicate.arithmetic_simplify
 1.42% :     0.000002s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.41% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.41% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.41% :     0.000000s :      3: predicate.graph_param_transform
 5.40% :     0.000006s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.87% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.13% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.57% :     0.000002s :      3: predicate.mutable_eliminate
 0.57% :     0.000001s :      3: predicate.opt_reshape
 2.04% :     0.000002s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.93% :     0.000002s :     18: predicate.replace_applicator
 0.79% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.14% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.15% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.80% :     0.000009s :     53: predicate.switch_simplify
 1.40% :     0.000002s :     10: predicate.tile_eliminate
 1.28% :     0.000001s :     10: predicate.transpose_eliminate
 1.23% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.14% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.34% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000574    12
49.71% :     0.000285s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.77% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.51% :     0.000255s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021248    76
 0.26% :     0.000055s :      1: add_recomputation
 0.31% :     0.000065s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.58% :     0.000336s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.08% :     0.005329s :      1: jit_opt_a
 0.70% :     0.000149s :      1: jit_opt_after_cconv
 0.26% :     0.000055s :      1: jit_opt_b
 2.12% :     0.000450s :      1: loop_unroll
 3.39% :     0.000720s :      1: mutable_eliminate
 3.93% :     0.000836s :     26: opt.transform.jit_opt_a
 0.23% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.81% :     0.000598s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000036s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.66% :     0.000353s :      1: renormalize.infer
 1.44% :     0.000306s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000040s :      1: rewriter_after_opt_a
 0.34% :     0.000073s :      1: rewriter_before_opt_a
 0.39% :     0.000082s :      1: symbol_engine_optimizer
31.24% :     0.006638s :      1: task_emit
22.17% :     0.004710s :      1: type_inference
 0.28% :     0.000059s :      1: validate


TotalTime = 0.0218323, [33]
[bootstrap]: 0.000335
[type_inference]: 0.00540896
[event_method]: 1.852e-05
[auto_monad]: 6.68e-05
[graph_reusing]: 6.48998e-06
[pre_auto_parallel]: 2.49999e-06
[py_interpret_to_execute]: 3.88e-05
[rewriter_before_opt_a]: 7.515e-05
[expand_dump_flag]: 3.19001e-06
[jit_opt_a]: 0.00601079, [2]
    [Cycle 1]: 0.00188511, [27]
        [switch_simplify]: 6.282e-05
        [loop_unroll]: 3.042e-05
        [a_1]: 0.00058947
        [with_stream_mark]: 2.048e-05
        [recompute_prepare]: 6.78e-06
        [updatestate_depend_eliminate]: 3.5e-06
        [updatestate_assign_eliminate]: 2.84999e-06
        [updatestate_loads_eliminate]: 2.77002e-06
        [parameter_eliminate]: 2.38998e-06
        [specialize_transform]: 6.67002e-06
        [updatestate_useless_node_eliminater]: 5.25999e-06
        [accelerated_algorithm]: 5.80002e-06
        [meta_shard_fg_expand]: 2.04e-06
        [get_grad_eliminate_]: 5.19998e-06
        [merge_forward]: 4.11001e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.519e-05
        [j_node_and_user_rematch]: 9.97001e-06
        [meta_fg_expand]: 2.48e-06
        [replace_old_param]: 9.37999e-06
        [inline_without_move]: 5.57001e-06
        [renormalize]: 0.00084917
        [add_forward_monad_depend]: 5.37999e-06
        [auto_monad_grad]: 2.49001e-06
        [auto_monad_eliminator]: 1.502e-05
        [cse]: 2.89e-05
        [replace_applicator]: 1.135e-05
    [Cycle 2]: 0.00032259, [27]
        [switch_simplify]: 5.61e-06
        [loop_unroll]: 5.08002e-06
        [a_1]: 9.072e-05
        [with_stream_mark]: 9.31e-06
        [recompute_prepare]: 5.09e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 1.19998e-06
        [specialize_transform]: 5.14003e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 4.67998e-06
        [merge_forward]: 2.44001e-06
        [cell_reuse_recompute_pass]: 1.37999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.231e-05
        [j_node_and_user_rematch]: 7.68999e-06
        [meta_fg_expand]: 1.48002e-06
        [replace_old_param]: 7.66001e-06
        [inline_without_move]: 5.22999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.00001e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 5.29e-06
        [cse]: 1.069e-05
        [replace_applicator]: 5.12e-06
[py_interpret_to_execute_after_opt_a]: 1.187e-05
[rewriter_after_opt_a]: 3.687e-05
[convert_after_rewriter]: 8.12998e-06
[order_py_execute_after_rewriter]: 5.72001e-06
[mutable_eliminate]: 0.00054307
[jit_opt_b]: 5.016e-05, [1]
    [Cycle 1]: 4.291e-05, [2]
        [frontend_op_eliminate]: 1.655e-05
        [inline_after_opt_a]: 1.483e-05
[cconv]: 3.213e-05
[loop_unroll]: 0.00047946
[jit_opt_after_cconv]: 0.00014826, [1]
    [Cycle 1]: 0.00014131, [11]
        [c_1]: 2.148e-05
        [parameter_eliminate]: 2.93003e-06
        [updatestate_depend_eliminate]: 6.78e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 2.247e-05
        [call_graph_tuple_transform]: 2.038e-05
        [tuple_list_get_item_eliminator]: 5.35999e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 8.39995e-07
        [switch_simplify]: 5.19e-06
[remove_dup_value]: 1.569e-05
[partial_unused_args_eliminate]: 2.55002e-06
[environ_conv]: 5.92999e-06
[add_recomputation]: 5.098e-05
[cse_after_recomputation]: 2.088e-05, [1]
    [Cycle 1]: 1.518e-05, [1]
        [cse]: 9.41998e-06
[auto_monad_reorder]: 1.597e-05
[get_jit_bprop_graph]: 2.21998e-06
[rewriter_after_jit_bprop_graph]: 5.52999e-06
[opt_after_jit_grad]: 0.00047227
[symbol_engine_optimizer]: 7.364e-05, [1]
    [Cycle 1]: 6.657e-05, [6]
        [build]: 3.73001e-06
        [elim_shapecalc]: 7.45e-06
        [elim_not_effective]: 1.334e-05
        [opt_reshape]: 5.60001e-06
        [fold_const_symbol]: 8.54002e-06
        [renormalize]: 3.89991e-07
[validate]: 3.834e-05
[backend_pass]: 1.04e-06
[task_emit]: 0.00678448
[execute]: 9.54e-06
Sums
    bootstrap                            :     0.000335s :  2.02%
    type_inference                       :     0.005409s : 32.66%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000067s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.23%
    rewriter_before_opt_a                :     0.000075s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.41%
    jit_opt_a.loop_unroll                :     0.000036s :  0.21%
    jit_opt_a.a_1                        :     0.000680s :  4.11%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000849s :  5.13%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000040s :  0.24%
    jit_opt_a.replace_applicator         :     0.000016s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000543s :  3.28%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000032s :  0.19%
    loop_unroll                          :     0.000479s :  2.90%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000051s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000472s :  2.85%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006784s : 40.97%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000244    23
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.36% :     0.000006s :      3: substitution.graph_param_transform
85.38% :     0.000208s :      4: substitution.inline
 1.26% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.86% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.92% :     0.000005s :      2: substitution.replace_old_param
 5.78% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005332     2
81.92% :     0.004368s :      1: type_inference.infer
18.08% :     0.000964s :      1: type_inference.specialize
------[replace.]   0.000074     6
69.16% :     0.000051s :      4: replace.inline
30.84% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000218     6
94.10% :     0.000206s :      4: match.inline
 5.90% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.19% :     0.000001s :     10: predicate.addn_zero_filter
 2.07% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.41% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.43% :     0.000001s :      3: predicate.graph_param_transform
 5.47% :     0.000006s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.80% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.14% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.50% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.56% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.34% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.26% :     0.000001s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.27% :     0.000002s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.76% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.57% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.81% :     0.000009s :     53: predicate.switch_simplify
 1.28% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.73% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.44% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000699    12
49.45% :     0.000345s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.07% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.48% :     0.000304s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022859    76
 0.24% :     0.000054s :      1: add_recomputation
 0.31% :     0.000070s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.56% :     0.000357s :      1: bootstrap
 0.17% :     0.000038s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.31% :     0.006015s :      1: jit_opt_a
 0.66% :     0.000151s :      1: jit_opt_after_cconv
 0.23% :     0.000053s :      1: jit_opt_b
 2.13% :     0.000488s :      1: loop_unroll
 2.42% :     0.000552s :      1: mutable_eliminate
 3.93% :     0.000899s :     26: opt.transform.jit_opt_a
 0.21% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.10% :     0.000480s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000042s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 2.05% :     0.000468s :      1: renormalize.infer
 1.63% :     0.000374s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000040s :      1: rewriter_after_opt_a
 0.34% :     0.000078s :      1: rewriter_before_opt_a
 0.33% :     0.000076s :      1: symbol_engine_optimizer
29.77% :     0.006805s :      1: task_emit
23.74% :     0.005426s :      1: type_inference
 0.26% :     0.000060s :      1: validate


TotalTime = 0.0181688, [33]
[bootstrap]: 0.00028506
[type_inference]: 0.00445756
[event_method]: 1.792e-05
[auto_monad]: 6.418e-05
[graph_reusing]: 5.69e-06
[pre_auto_parallel]: 2.49999e-06
[py_interpret_to_execute]: 3.371e-05
[rewriter_before_opt_a]: 7.343e-05
[expand_dump_flag]: 3.28998e-06
[jit_opt_a]: 0.00503539, [2]
    [Cycle 1]: 0.00151766, [27]
        [switch_simplify]: 5.32e-05
        [loop_unroll]: 3.024e-05
        [a_1]: 0.00054782
        [with_stream_mark]: 1.83e-05
        [recompute_prepare]: 7.80998e-06
        [updatestate_depend_eliminate]: 3.31999e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.78e-06
        [parameter_eliminate]: 1.77001e-06
        [specialize_transform]: 5.87001e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 5.29e-06
        [meta_shard_fg_expand]: 2.34999e-06
        [get_grad_eliminate_]: 5.15001e-06
        [merge_forward]: 3.84002e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.496e-05
        [j_node_and_user_rematch]: 9.24e-06
        [meta_fg_expand]: 2.55002e-06
        [replace_old_param]: 8.87e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.00055059
        [add_forward_monad_depend]: 5.96e-06
        [auto_monad_grad]: 2.39001e-06
        [auto_monad_eliminator]: 1.4e-05
        [cse]: 2.558e-05
        [replace_applicator]: 1.228e-05
    [Cycle 2]: 0.00033704, [27]
        [switch_simplify]: 5.54e-06
        [loop_unroll]: 4.94e-06
        [a_1]: 9.059e-05
        [with_stream_mark]: 9.70002e-06
        [recompute_prepare]: 5.05001e-06
        [updatestate_depend_eliminate]: 2.81999e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.04998e-06
        [specialize_transform]: 5.14e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 1.17999e-06
        [get_grad_eliminate_]: 4.55001e-06
        [merge_forward]: 2.26998e-06
        [cell_reuse_recompute_pass]: 1.74e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.281e-05
        [j_node_and_user_rematch]: 7.49002e-06
        [meta_fg_expand]: 1.95001e-06
        [replace_old_param]: 7.25e-06
        [inline_without_move]: 4.75999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.59998e-06
        [auto_monad_grad]: 9.79984e-07
        [auto_monad_eliminator]: 5.57001e-06
        [cse]: 1.202e-05
        [replace_applicator]: 6.41e-06
[py_interpret_to_execute_after_opt_a]: 1.144e-05
[rewriter_after_opt_a]: 3.557e-05
[convert_after_rewriter]: 7.85998e-06
[order_py_execute_after_rewriter]: 5.99e-06
[mutable_eliminate]: 0.00053774
[jit_opt_b]: 5.036e-05, [1]
    [Cycle 1]: 4.338e-05, [2]
        [frontend_op_eliminate]: 1.642e-05
        [inline_after_opt_a]: 1.531e-05
[cconv]: 2.318e-05
[loop_unroll]: 0.0004288
[jit_opt_after_cconv]: 0.00014825, [1]
    [Cycle 1]: 0.00014183, [11]
        [c_1]: 2.063e-05
        [parameter_eliminate]: 2.43998e-06
        [updatestate_depend_eliminate]: 5.39998e-06
        [updatestate_assign_eliminate]: 2.47001e-06
        [updatestate_loads_eliminate]: 2.18002e-06
        [cse]: 1.737e-05
        [call_graph_tuple_transform]: 2.579e-05
        [tuple_list_get_item_eliminator]: 6.75998e-06
        [none_parameter_eliminate]: 1.76e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 5.59e-06
[remove_dup_value]: 1.361e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 5.07e-06
[add_recomputation]: 4.714e-05
[cse_after_recomputation]: 2.111e-05, [1]
    [Cycle 1]: 1.549e-05, [1]
        [cse]: 9.64e-06
[auto_monad_reorder]: 1.495e-05
[get_jit_bprop_graph]: 1.91998e-06
[rewriter_after_jit_bprop_graph]: 4.90999e-06
[opt_after_jit_grad]: 0.00046902
[symbol_engine_optimizer]: 7.095e-05, [1]
    [Cycle 1]: 6.479e-05, [6]
        [build]: 3.28998e-06
        [elim_shapecalc]: 7.13e-06
        [elim_not_effective]: 1.217e-05
        [opt_reshape]: 5.49e-06
        [fold_const_symbol]: 8.55999e-06
        [renormalize]: 4.49974e-07
[validate]: 3.299e-05
[backend_pass]: 9.49978e-07
[task_emit]: 0.00605301
[execute]: 8.3e-06
Sums
    bootstrap                            :     0.000285s :  1.98%
    type_inference                       :     0.004458s : 31.01%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000064s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.23%
    rewriter_before_opt_a                :     0.000073s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.41%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000638s :  4.44%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000551s :  3.83%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.14%
    jit_opt_a.cse                        :     0.000038s :  0.26%
    jit_opt_a.replace_applicator         :     0.000019s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.08%
    rewriter_after_opt_a                 :     0.000036s :  0.25%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000538s :  3.74%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000023s :  0.16%
    loop_unroll                          :     0.000429s :  2.98%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000469s :  3.26%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006053s : 42.11%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000193    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.86% :     0.000002s :      2: substitution.fold_const_symbol
 6.01% :     0.000012s :      3: substitution.graph_param_transform
80.30% :     0.000155s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.41% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.84% :     0.000004s :      2: substitution.replace_old_param
 6.04% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004397     2
82.01% :     0.003606s :      1: type_inference.infer
17.99% :     0.000791s :      1: type_inference.specialize
------[replace.]   0.000066     6
66.23% :     0.000044s :      4: replace.inline
33.77% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000163     6
93.55% :     0.000152s :      4: match.inline
 6.45% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.54% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 1.89% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000002s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.45% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.30% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 6.13% :     0.000007s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000001s :      6: predicate.less_batch_normalization
 1.81% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.00% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     15: predicate.partial_eliminate
 1.06% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.82% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.97% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.34% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.28% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.25% :     0.000009s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.43% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.43% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000541    12
50.72% :     0.000274s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.27% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.00% :     0.000238s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019679    76
 0.25% :     0.000050s :      1: add_recomputation
 0.34% :     0.000068s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.51% :     0.000297s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.60% :     0.005039s :      1: jit_opt_a
 0.77% :     0.000151s :      1: jit_opt_after_cconv
 0.27% :     0.000053s :      1: jit_opt_b
 2.22% :     0.000437s :      1: loop_unroll
 2.78% :     0.000546s :      1: mutable_eliminate
 4.31% :     0.000847s :     26: opt.transform.jit_opt_a
 0.28% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.42% :     0.000477s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000036s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.38% :     0.000273s :      1: renormalize.infer
 1.38% :     0.000271s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000039s :      1: rewriter_after_opt_a
 0.39% :     0.000076s :      1: rewriter_before_opt_a
 0.37% :     0.000073s :      1: symbol_engine_optimizer
30.81% :     0.006063s :      1: task_emit
22.76% :     0.004479s :      1: type_inference
 0.27% :     0.000053s :      1: validate


TotalTime = 0.0145076, [30]
[bootstrap]: 0.00026401
[type_inference]: 0.00631785
[event_method]: 2.029e-05
[auto_monad]: 7.797e-05
[graph_reusing]: 6.01e-06
[pre_auto_parallel]: 2.66e-06
[py_interpret_to_execute]: 4.623e-05
[rewriter_before_opt_a]: 8.148e-05
[expand_dump_flag]: 3.03e-06
[jit_opt_a]: 0.00560398, [2]
    [Cycle 1]: 0.00193462, [27]
        [switch_simplify]: 0.00011893
        [loop_unroll]: 3.415e-05
        [a_1]: 0.00069406
        [with_stream_mark]: 3.709e-05
        [recompute_prepare]: 7.94002e-06
        [updatestate_depend_eliminate]: 3.75998e-06
        [updatestate_assign_eliminate]: 2.89999e-06
        [updatestate_loads_eliminate]: 2.41998e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 4.83001e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 1.90001e-06
        [get_grad_eliminate_]: 5.14998e-06
        [merge_forward]: 3.52997e-06
        [cell_reuse_recompute_pass]: 9.99979e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.43e-05
        [j_node_and_user_rematch]: 8.97e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 8.72998e-06
        [inline_without_move]: 6.17999e-06
        [renormalize]: 0.00070203
        [add_forward_monad_depend]: 5.67001e-06
        [auto_monad_grad]: 2.12001e-06
        [auto_monad_eliminator]: 1.317e-05
        [cse]: 2.96e-05
        [replace_applicator]: 1.182e-05
    [Cycle 2]: 0.00030383, [27]
        [switch_simplify]: 5.43002e-06
        [loop_unroll]: 4.70001e-06
        [a_1]: 8.706e-05
        [with_stream_mark]: 9.15999e-06
        [recompute_prepare]: 5.02999e-06
        [updatestate_depend_eliminate]: 2.54001e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 4.64998e-06
        [updatestate_useless_node_eliminater]: 4.84003e-06
        [accelerated_algorithm]: 4.73001e-06
        [meta_shard_fg_expand]: 1.03001e-06
        [get_grad_eliminate_]: 4.33001e-06
        [merge_forward]: 2.39999e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.207e-05
        [j_node_and_user_rematch]: 7.73001e-06
        [meta_fg_expand]: 1.45999e-06
        [replace_old_param]: 6.78e-06
        [inline_without_move]: 4.70999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 4.67e-06
        [cse]: 8.57e-06
        [replace_applicator]: 4.72e-06
[py_interpret_to_execute_after_opt_a]: 1.029e-05
[rewriter_after_opt_a]: 3.15e-05
[convert_after_rewriter]: 7.05e-06
[order_py_execute_after_rewriter]: 4.93001e-06
[mutable_eliminate]: 0.00052738
[jit_opt_b]: 4.727e-05, [1]
    [Cycle 1]: 4.017e-05, [2]
        [frontend_op_eliminate]: 1.484e-05
        [inline_after_opt_a]: 1.416e-05
[cconv]: 2.478e-05
[loop_unroll]: 0.00041447
[jit_opt_after_cconv]: 0.0001583, [1]
    [Cycle 1]: 0.00015127, [11]
        [c_1]: 2.18e-05
        [parameter_eliminate]: 2.46998e-06
        [updatestate_depend_eliminate]: 5.42999e-06
        [updatestate_assign_eliminate]: 2.41998e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.558e-05
        [call_graph_tuple_transform]: 3.914e-05
        [tuple_list_get_item_eliminator]: 5.66e-06
        [none_parameter_eliminate]: 1.73002e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 5.31002e-06
[remove_dup_value]: 1.316e-05
[partial_unused_args_eliminate]: 2.68998e-06
[environ_conv]: 5.10001e-06
[add_recomputation]: 4.651e-05
[cse_after_recomputation]: 2.027e-05, [1]
    [Cycle 1]: 1.495e-05, [1]
        [cse]: 9.38002e-06
[auto_monad_reorder]: 1.535e-05
[get_jit_bprop_graph]: 2.24999e-06
[rewriter_after_jit_bprop_graph]: 4.75001e-06
[opt_after_jit_grad]: 0.00047901
[symbol_engine_optimizer]: 7.266e-05, [1]
    [Cycle 1]: 6.667e-05, [6]
        [build]: 3.5e-06
        [elim_shapecalc]: 6.87002e-06
        [elim_not_effective]: 1.344e-05
        [opt_reshape]: 5.86998e-06
        [fold_const_symbol]: 8.47998e-06
        [renormalize]: 3.50003e-07
[validate]: 3.24e-05
Sums
    bootstrap                            :     0.000264s :  2.50%
    type_inference                       :     0.006318s : 59.88%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000078s :  0.74%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000046s :  0.44%
    rewriter_before_opt_a                :     0.000081s :  0.77%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000124s :  1.18%
    jit_opt_a.loop_unroll                :     0.000039s :  0.37%
    jit_opt_a.a_1                        :     0.000781s :  7.40%
    jit_opt_a.with_stream_mark           :     0.000046s :  0.44%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000010s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000702s :  6.65%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.17%
    jit_opt_a.cse                        :     0.000038s :  0.36%
    jit_opt_a.replace_applicator         :     0.000017s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.10%
    rewriter_after_opt_a                 :     0.000032s :  0.30%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000527s :  5.00%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.13%
    cconv                                :     0.000025s :  0.23%
    loop_unroll                          :     0.000414s :  3.93%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000039s :  0.37%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.12%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000047s :  0.44%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000479s :  4.54%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.31%


Time group info:
------[substitution.]   0.000277    26
 0.80% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 1.86% :     0.000005s :      3: substitution.graph_param_transform
85.51% :     0.000237s :      6: substitution.inline
 1.12% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.63% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.18% :     0.000003s :      2: substitution.replace_old_param
 2.88% :     0.000008s :      1: substitution.switch_simplify
 4.47% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006259     2
83.03% :     0.005197s :      1: type_inference.infer
16.97% :     0.001062s :      1: type_inference.specialize
------[replace.]   0.000134     9
48.91% :     0.000066s :      6: replace.inline
28.27% :     0.000038s :      1: replace.switch_simplify
22.83% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000252     9
92.72% :     0.000233s :      6: match.inline
 2.85% :     0.000007s :      1: match.switch_simplify
 4.43% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   801
 1.23% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.23% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     12: predicate.addn_check_dump
 1.59% :     0.000002s :     12: predicate.addn_zero_filter
 2.01% :     0.000003s :     12: predicate.arithmetic_simplify
 1.38% :     0.000002s :     12: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.23% :     0.000002s :     12: predicate.compare_switch_simplify
 1.47% :     0.000002s :     12: predicate.depend_value_elim
 1.17% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     12: predicate.environ_get_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.46% :     0.000007s :     26: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.90% :     0.000002s :     17: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.14% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.95% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     12: predicate.merge_addn
 1.16% :     0.000001s :     12: predicate.minmaximum_grad
 1.47% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     17: predicate.partial_eliminate
 1.23% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     12: predicate.reduce_eliminate
 1.50% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     20: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     12: predicate.reshape_eliminate
 1.25% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     12: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.76% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.75% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.87% :     0.000004s :     20: predicate.switch_defer_inline
 2.35% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.36% :     0.000011s :     65: predicate.switch_simplify
 1.23% :     0.000002s :     12: predicate.tile_eliminate
 1.32% :     0.000002s :     12: predicate.transpose_eliminate
 1.50% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.86% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     12: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000758    15
47.11% :     0.000357s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.15% :     0.000100s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.74% :     0.000301s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016338    72
 0.30% :     0.000049s :      1: add_recomputation
 0.50% :     0.000082s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 1.70% :     0.000278s :      1: bootstrap
 0.17% :     0.000027s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
34.32% :     0.005608s :      1: jit_opt_a
 0.99% :     0.000161s :      1: jit_opt_after_cconv
 0.31% :     0.000050s :      1: jit_opt_b
 2.58% :     0.000422s :      1: loop_unroll
 3.27% :     0.000535s :      1: mutable_eliminate
 6.46% :     0.001055s :     26: opt.transform.jit_opt_a
 0.42% :     0.000068s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000023s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.98% :     0.000487s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000049s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000015s :      1: remove_dup_value
 2.51% :     0.000411s :      1: renormalize.infer
 1.74% :     0.000284s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000034s :      1: rewriter_after_opt_a
 0.52% :     0.000085s :      1: rewriter_before_opt_a
 0.46% :     0.000075s :      1: symbol_engine_optimizer
38.74% :     0.006330s :      1: type_inference


TotalTime = 0.0249007, [33]
[bootstrap]: 0.00046786
[type_inference]: 0.00585198
[event_method]: 2.047e-05
[auto_monad]: 7.332e-05
[graph_reusing]: 6.46e-06
[pre_auto_parallel]: 2.96999e-06
[py_interpret_to_execute]: 4.883e-05
[rewriter_before_opt_a]: 8.503e-05
[expand_dump_flag]: 4.18001e-06
[jit_opt_a]: 0.00806248, [2]
    [Cycle 1]: 0.00192508, [27]
        [switch_simplify]: 6.7e-05
        [loop_unroll]: 3.019e-05
        [a_1]: 0.0005807
        [with_stream_mark]: 2.165e-05
        [recompute_prepare]: 7.20998e-06
        [updatestate_depend_eliminate]: 3.8e-06
        [updatestate_assign_eliminate]: 3.48e-06
        [updatestate_loads_eliminate]: 2.95998e-06
        [parameter_eliminate]: 2.22999e-06
        [specialize_transform]: 5.87999e-06
        [updatestate_useless_node_eliminater]: 5.50001e-06
        [accelerated_algorithm]: 6.11998e-06
        [meta_shard_fg_expand]: 2.12001e-06
        [get_grad_eliminate_]: 5.35999e-06
        [merge_forward]: 3.63999e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.421e-05
        [j_node_and_user_rematch]: 9.47999e-06
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 9.54e-06
        [inline_without_move]: 5.52999e-06
        [renormalize]: 0.00088415
        [add_forward_monad_depend]: 6.16e-06
        [auto_monad_grad]: 2.68e-06
        [auto_monad_eliminator]: 1.518e-05
        [cse]: 3.053e-05
        [replace_applicator]: 1.537e-05
    [Cycle 2]: 0.00033664, [27]
        [switch_simplify]: 5.67001e-06
        [loop_unroll]: 4.89e-06
        [a_1]: 9.665e-05
        [with_stream_mark]: 9.61e-06
        [recompute_prepare]: 5.42001e-06
        [updatestate_depend_eliminate]: 2.98e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 1.18001e-06
        [specialize_transform]: 5.56e-06
        [updatestate_useless_node_eliminater]: 4.99003e-06
        [accelerated_algorithm]: 5.62001e-06
        [meta_shard_fg_expand]: 1.24998e-06
        [get_grad_eliminate_]: 5.21002e-06
        [merge_forward]: 3.33e-06
        [cell_reuse_recompute_pass]: 2.16e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.442e-05
        [j_node_and_user_rematch]: 8.77999e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 7.43999e-06
        [inline_without_move]: 4.83001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.45999e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 4.91002e-06
        [cse]: 1.052e-05
        [replace_applicator]: 5.00999e-06
[py_interpret_to_execute_after_opt_a]: 1.239e-05
[rewriter_after_opt_a]: 3.994e-05
[convert_after_rewriter]: 7.73001e-06
[order_py_execute_after_rewriter]: 4.84e-06
[mutable_eliminate]: 0.00063714
[jit_opt_b]: 5.645e-05, [1]
    [Cycle 1]: 4.915e-05, [2]
        [frontend_op_eliminate]: 2.176e-05
        [inline_after_opt_a]: 1.545e-05
[cconv]: 2.928e-05
[loop_unroll]: 0.00043178
[jit_opt_after_cconv]: 0.0001738, [1]
    [Cycle 1]: 0.00016674, [11]
        [c_1]: 2.4e-05
        [parameter_eliminate]: 3.73001e-06
        [updatestate_depend_eliminate]: 7.85e-06
        [updatestate_assign_eliminate]: 2.48002e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [cse]: 2.024e-05
        [call_graph_tuple_transform]: 3.452e-05
        [tuple_list_get_item_eliminator]: 5.79e-06
        [none_parameter_eliminate]: 2.04999e-06
        [renormalize]: 9.09989e-07
        [switch_simplify]: 6.12001e-06
[remove_dup_value]: 1.498e-05
[partial_unused_args_eliminate]: 2.32999e-06
[environ_conv]: 6.17001e-06
[add_recomputation]: 5.275e-05
[cse_after_recomputation]: 2.264e-05, [1]
    [Cycle 1]: 1.622e-05, [1]
        [cse]: 9.99999e-06
[auto_monad_reorder]: 1.733e-05
[get_jit_bprop_graph]: 2.49001e-06
[rewriter_after_jit_bprop_graph]: 6.69999e-06
[opt_after_jit_grad]: 0.00047206
[symbol_engine_optimizer]: 7.361e-05, [1]
    [Cycle 1]: 6.709e-05, [6]
        [build]: 4.37e-06
        [elim_shapecalc]: 8.28999e-06
        [elim_not_effective]: 1.21e-05
        [opt_reshape]: 5.74999e-06
        [fold_const_symbol]: 8.67e-06
        [renormalize]: 6.10016e-07
[validate]: 4.075e-05
[backend_pass]: 1.19e-06
[task_emit]: 0.00791911
[execute]: 8.90001e-06
Sums
    bootstrap                            :     0.000468s :  2.54%
    type_inference                       :     0.005852s : 31.75%
    event_method                         :     0.000020s :  0.11%
    auto_monad                           :     0.000073s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000049s :  0.26%
    rewriter_before_opt_a                :     0.000085s :  0.46%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000073s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.19%
    jit_opt_a.a_1                        :     0.000677s :  3.68%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000884s :  4.80%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.11%
    jit_opt_a.cse                        :     0.000041s :  0.22%
    jit_opt_a.replace_applicator         :     0.000020s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000040s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000637s :  3.46%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000029s :  0.16%
    loop_unroll                          :     0.000432s :  2.34%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000035s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000053s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000472s :  2.56%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007919s : 42.97%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000231    23
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.62% :     0.000006s :      3: substitution.graph_param_transform
84.82% :     0.000196s :      4: substitution.inline
 1.74% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.92% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.94% :     0.000004s :      2: substitution.replace_old_param
 5.58% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005761     2
81.08% :     0.004671s :      1: type_inference.infer
18.92% :     0.001090s :      1: type_inference.specialize
------[replace.]   0.000081     6
68.37% :     0.000055s :      4: replace.inline
31.63% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000205     6
94.27% :     0.000193s :      4: match.inline
 5.73% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.56% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 2.17% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.18% :     0.000001s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.48% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.34% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.28% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.08% :     0.000001s :      6: predicate.get_grad_eliminate
 0.41% :     0.000001s :      3: predicate.graph_param_transform
 5.96% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.25% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.87% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.34% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.92% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     15: predicate.partial_eliminate
 1.23% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.65% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.05% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.64% :     0.000009s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.07% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.13% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.83% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000716    12
48.23% :     0.000345s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.58% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.19% :     0.000338s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026809    76
 0.21% :     0.000056s :      1: add_recomputation
 0.29% :     0.000078s :      1: auto_monad
 0.08% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.82% :     0.000489s :      1: bootstrap
 0.12% :     0.000032s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.09% :     0.000025s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
30.09% :     0.008066s :      1: jit_opt_a
 0.66% :     0.000177s :      1: jit_opt_after_cconv
 0.22% :     0.000059s :      1: jit_opt_b
 1.64% :     0.000441s :      1: loop_unroll
 2.41% :     0.000647s :      1: mutable_eliminate
 3.38% :     0.000906s :     26: opt.transform.jit_opt_a
 0.25% :     0.000066s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000030s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.79% :     0.000481s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000052s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 1.72% :     0.000462s :      1: renormalize.infer
 1.54% :     0.000413s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000043s :      1: rewriter_after_opt_a
 0.33% :     0.000089s :      1: rewriter_before_opt_a
 0.28% :     0.000076s :      1: symbol_engine_optimizer
29.62% :     0.007941s :      1: task_emit
21.92% :     0.005876s :      1: type_inference
 0.23% :     0.000062s :      1: validate


TotalTime = 0.0181449, [33]
[bootstrap]: 0.00028022
[type_inference]: 0.00446901
[event_method]: 2.407e-05
[auto_monad]: 6.502e-05
[graph_reusing]: 6.76e-06
[pre_auto_parallel]: 2.56e-06
[py_interpret_to_execute]: 3.532e-05
[rewriter_before_opt_a]: 7.184e-05
[expand_dump_flag]: 3.14999e-06
[jit_opt_a]: 0.00520189, [2]
    [Cycle 1]: 0.0015639, [27]
        [switch_simplify]: 5.511e-05
        [loop_unroll]: 2.986e-05
        [a_1]: 0.00053983
        [with_stream_mark]: 1.769e-05
        [recompute_prepare]: 6.89999e-06
        [updatestate_depend_eliminate]: 3.73999e-06
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 2.90998e-06
        [parameter_eliminate]: 2.10002e-06
        [specialize_transform]: 6.06998e-06
        [updatestate_useless_node_eliminater]: 4.97999e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 2.39001e-06
        [get_grad_eliminate_]: 5.03002e-06
        [merge_forward]: 4.08999e-06
        [cell_reuse_recompute_pass]: 1.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.344e-05
        [j_node_and_user_rematch]: 9.02e-06
        [meta_fg_expand]: 2.52001e-06
        [replace_old_param]: 8.35001e-06
        [inline_without_move]: 4.91002e-06
        [renormalize]: 0.00061024
        [add_forward_monad_depend]: 5.06997e-06
        [auto_monad_grad]: 1.96e-06
        [auto_monad_eliminator]: 1.392e-05
        [cse]: 2.564e-05
        [replace_applicator]: 1.113e-05
    [Cycle 2]: 0.00032537, [27]
        [switch_simplify]: 5.50001e-06
        [loop_unroll]: 5.24e-06
        [a_1]: 0.00010147
        [with_stream_mark]: 8.26002e-06
        [recompute_prepare]: 5.54e-06
        [updatestate_depend_eliminate]: 2.79001e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.06002e-06
        [specialize_transform]: 5.05999e-06
        [updatestate_useless_node_eliminater]: 4.78001e-06
        [accelerated_algorithm]: 4.89e-06
        [meta_shard_fg_expand]: 1.17999e-06
        [get_grad_eliminate_]: 4.55001e-06
        [merge_forward]: 2.48002e-06
        [cell_reuse_recompute_pass]: 2.24999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.156e-05
        [j_node_and_user_rematch]: 7.61999e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 6.89001e-06
        [inline_without_move]: 4.65999e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 5.46e-06
        [cse]: 9.73002e-06
        [replace_applicator]: 4.94e-06
[py_interpret_to_execute_after_opt_a]: 9.82999e-06
[rewriter_after_opt_a]: 3.292e-05
[convert_after_rewriter]: 6.71e-06
[order_py_execute_after_rewriter]: 5.33002e-06
[mutable_eliminate]: 0.00051572
[jit_opt_b]: 4.957e-05, [1]
    [Cycle 1]: 4.294e-05, [2]
        [frontend_op_eliminate]: 1.644e-05
        [inline_after_opt_a]: 1.509e-05
[cconv]: 2.269e-05
[loop_unroll]: 0.00041581
[jit_opt_after_cconv]: 0.00013436, [1]
    [Cycle 1]: 0.00012807, [11]
        [c_1]: 2.093e-05
        [parameter_eliminate]: 1.94e-06
        [updatestate_depend_eliminate]: 4.25999e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [cse]: 1.68e-05
        [call_graph_tuple_transform]: 1.667e-05
        [tuple_list_get_item_eliminator]: 5.52001e-06
        [none_parameter_eliminate]: 1.79e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 5.30999e-06
[remove_dup_value]: 1.239e-05
[partial_unused_args_eliminate]: 2.10002e-06
[environ_conv]: 4.82e-06
[add_recomputation]: 4.536e-05
[cse_after_recomputation]: 3.229e-05, [1]
    [Cycle 1]: 1.637e-05, [1]
        [cse]: 1.022e-05
[auto_monad_reorder]: 1.517e-05
[get_jit_bprop_graph]: 1.87001e-06
[rewriter_after_jit_bprop_graph]: 4.87998e-06
[opt_after_jit_grad]: 0.00045544
[symbol_engine_optimizer]: 7.323e-05, [1]
    [Cycle 1]: 6.722e-05, [6]
        [build]: 3.01999e-06
        [elim_shapecalc]: 7.78001e-06
        [elim_not_effective]: 1.284e-05
        [opt_reshape]: 5.70001e-06
        [fold_const_symbol]: 9.29e-06
        [renormalize]: 4.2998e-07
[validate]: 3.184e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00592495
[execute]: 8.69998e-06
Sums
    bootstrap                            :     0.000280s :  1.97%
    type_inference                       :     0.004469s : 31.37%
    event_method                         :     0.000024s :  0.17%
    auto_monad                           :     0.000065s :  0.46%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.25%
    rewriter_before_opt_a                :     0.000072s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.43%
    jit_opt_a.loop_unroll                :     0.000035s :  0.25%
    jit_opt_a.a_1                        :     0.000641s :  4.50%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000610s :  4.28%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.14%
    jit_opt_a.cse                        :     0.000035s :  0.25%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000516s :  3.62%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000023s :  0.16%
    loop_unroll                          :     0.000416s :  2.92%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000045s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000455s :  3.20%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005925s : 41.60%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000208    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.80% :     0.000002s :      2: substitution.fold_const_symbol
 2.39% :     0.000005s :      3: substitution.graph_param_transform
84.83% :     0.000176s :      4: substitution.inline
 1.26% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.98% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.68% :     0.000003s :      2: substitution.replace_old_param
 6.08% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004402     2
80.98% :     0.003564s :      1: type_inference.infer
19.02% :     0.000837s :      1: type_inference.specialize
------[replace.]   0.000068     6
65.85% :     0.000045s :      4: replace.inline
34.15% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000185     6
93.81% :     0.000174s :      4: match.inline
 6.19% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.29% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000001s :     10: predicate.addn_zero_filter
 1.96% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.50% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.38% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.29% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.36% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :      6: predicate.less_batch_normalization
 1.67% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.86% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.10% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.57% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.34% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.56% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.66% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.71% :     0.000003s :     16: predicate.switch_defer_inline
 2.44% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.74% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.26% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.78% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000562    12
47.80% :     0.000269s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.00% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.20% :     0.000254s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019705    76
 0.24% :     0.000048s :      1: add_recomputation
 0.35% :     0.000068s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.48% :     0.000292s :      1: bootstrap
 0.13% :     0.000025s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000035s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.14% :     0.000028s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
26.41% :     0.005205s :      1: jit_opt_a
 0.70% :     0.000137s :      1: jit_opt_after_cconv
 0.27% :     0.000052s :      1: jit_opt_b
 2.14% :     0.000422s :      1: loop_unroll
 2.66% :     0.000524s :      1: mutable_eliminate
 4.29% :     0.000845s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.35% :     0.000464s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000038s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.72% :     0.000339s :      1: renormalize.infer
 1.34% :     0.000264s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000036s :      1: rewriter_after_opt_a
 0.38% :     0.000075s :      1: rewriter_before_opt_a
 0.38% :     0.000076s :      1: symbol_engine_optimizer
30.12% :     0.005935s :      1: task_emit
22.74% :     0.004481s :      1: type_inference
 0.25% :     0.000049s :      1: validate


TotalTime = 0.0137073, [30]
[bootstrap]: 0.00025276
[type_inference]: 0.00591765
[event_method]: 1.861e-05
[auto_monad]: 7.937e-05
[graph_reusing]: 5.40001e-06
[pre_auto_parallel]: 2.59001e-06
[py_interpret_to_execute]: 7.315e-05
[rewriter_before_opt_a]: 8.274e-05
[expand_dump_flag]: 3.03e-06
[jit_opt_a]: 0.00519163, [2]
    [Cycle 1]: 0.00168721, [27]
        [switch_simplify]: 0.00010222
        [loop_unroll]: 3.334e-05
        [a_1]: 0.00064677
        [with_stream_mark]: 1.628e-05
        [recompute_prepare]: 7.13998e-06
        [updatestate_depend_eliminate]: 4.16001e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.50002e-06
        [parameter_eliminate]: 1.69e-06
        [specialize_transform]: 6.24001e-06
        [updatestate_useless_node_eliminater]: 4.89003e-06
        [accelerated_algorithm]: 5.09998e-06
        [meta_shard_fg_expand]: 1.86998e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 3.41001e-06
        [cell_reuse_recompute_pass]: 9.90025e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.397e-05
        [j_node_and_user_rematch]: 8.64003e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 7.78999e-06
        [inline_without_move]: 5.28002e-06
        [renormalize]: 0.0005922
        [add_forward_monad_depend]: 4.99998e-06
        [auto_monad_grad]: 1.70001e-06
        [auto_monad_eliminator]: 1.216e-05
        [cse]: 1.941e-05
        [replace_applicator]: 1.054e-05
    [Cycle 2]: 0.00035382, [27]
        [switch_simplify]: 5.37001e-06
        [loop_unroll]: 4.59998e-06
        [a_1]: 8.813e-05
        [with_stream_mark]: 8.47e-06
        [recompute_prepare]: 5.16002e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 2.15002e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 8.50006e-07
        [specialize_transform]: 4.73001e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 4.72998e-06
        [meta_shard_fg_expand]: 1.33002e-06
        [get_grad_eliminate_]: 4.35e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.227e-05
        [j_node_and_user_rematch]: 7.6e-06
        [meta_fg_expand]: 1.53002e-06
        [replace_old_param]: 4.726e-05
        [inline_without_move]: 5.18002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 5.59e-06
        [cse]: 1.163e-05
        [replace_applicator]: 5.19e-06
[py_interpret_to_execute_after_opt_a]: 9.37001e-06
[rewriter_after_opt_a]: 3.264e-05
[convert_after_rewriter]: 6.48e-06
[order_py_execute_after_rewriter]: 4.87e-06
[mutable_eliminate]: 0.00053556
[jit_opt_b]: 5.357e-05, [1]
    [Cycle 1]: 3.95e-05, [2]
        [frontend_op_eliminate]: 1.467e-05
        [inline_after_opt_a]: 1.392e-05
[cconv]: 2.336e-05
[loop_unroll]: 0.00045622
[jit_opt_after_cconv]: 0.00013387, [1]
    [Cycle 1]: 0.00012768, [11]
        [c_1]: 2.027e-05
        [parameter_eliminate]: 2.46e-06
        [updatestate_depend_eliminate]: 5.31002e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.06003e-06
        [cse]: 1.665e-05
        [call_graph_tuple_transform]: 1.726e-05
        [tuple_list_get_item_eliminator]: 5.04e-06
        [none_parameter_eliminate]: 1.43002e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.29e-06
[remove_dup_value]: 1.217e-05
[partial_unused_args_eliminate]: 1.99999e-06
[environ_conv]: 4.90999e-06
[add_recomputation]: 4.384e-05
[cse_after_recomputation]: 1.974e-05, [1]
    [Cycle 1]: 1.461e-05, [1]
        [cse]: 8.99e-06
[auto_monad_reorder]: 1.313e-05
[get_jit_bprop_graph]: 1.99e-06
[rewriter_after_jit_bprop_graph]: 4.96002e-06
[opt_after_jit_grad]: 0.00046861
[symbol_engine_optimizer]: 7.251e-05, [1]
    [Cycle 1]: 6.623e-05, [6]
        [build]: 3.64002e-06
        [elim_shapecalc]: 7.41001e-06
        [elim_not_effective]: 1.278e-05
        [opt_reshape]: 5.54e-06
        [fold_const_symbol]: 8.24002e-06
        [renormalize]: 4.50003e-07
[validate]: 3.203e-05
Sums
    bootstrap                            :     0.000253s :  2.53%
    type_inference                       :     0.005918s : 59.14%
    event_method                         :     0.000019s :  0.19%
    auto_monad                           :     0.000079s :  0.79%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000073s :  0.73%
    rewriter_before_opt_a                :     0.000083s :  0.83%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000108s :  1.08%
    jit_opt_a.loop_unroll                :     0.000038s :  0.38%
    jit_opt_a.a_1                        :     0.000735s :  7.34%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000055s :  0.55%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000592s :  5.92%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.18%
    jit_opt_a.cse                        :     0.000031s :  0.31%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.09%
    rewriter_after_opt_a                 :     0.000033s :  0.33%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000536s :  5.35%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.14%
    cconv                                :     0.000023s :  0.23%
    loop_unroll                          :     0.000456s :  4.56%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000012s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000044s :  0.44%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000013s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000469s :  4.68%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.32%


Time group info:
------[substitution.]   0.000251    26
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 1.76% :     0.000004s :      3: substitution.graph_param_transform
84.80% :     0.000213s :      6: substitution.inline
 1.10% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.71% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.26% :     0.000003s :      2: substitution.replace_old_param
 3.06% :     0.000008s :      1: substitution.switch_simplify
 5.00% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005865     2
83.97% :     0.004925s :      1: type_inference.infer
16.03% :     0.000940s :      1: type_inference.specialize
------[replace.]   0.000118     9
51.71% :     0.000061s :      6: replace.inline
24.10% :     0.000028s :      1: replace.switch_simplify
24.19% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000228     9
91.95% :     0.000209s :      6: match.inline
 3.04% :     0.000007s :      1: match.switch_simplify
 5.01% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   801
 1.31% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     12: predicate.addn_check_dump
 1.39% :     0.000002s :     12: predicate.addn_zero_filter
 1.85% :     0.000002s :     12: predicate.arithmetic_simplify
 1.28% :     0.000002s :     12: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     12: predicate.compare_switch_simplify
 1.44% :     0.000002s :     12: predicate.depend_value_elim
 1.15% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     12: predicate.environ_get_eliminate
 1.17% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.26% :     0.000007s :     26: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     17: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.22% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     12: predicate.merge_addn
 1.31% :     0.000002s :     12: predicate.minmaximum_grad
 1.40% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     17: predicate.partial_eliminate
 1.24% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     12: predicate.reduce_eliminate
 1.70% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     20: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.60% :     0.000002s :     12: predicate.reshape_eliminate
 1.30% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.62% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     12: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.77% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.72% :     0.000003s :     20: predicate.switch_defer_inline
 2.56% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.34% :     0.000011s :     65: predicate.switch_simplify
 1.38% :     0.000002s :     12: predicate.tile_eliminate
 1.25% :     0.000002s :     12: predicate.transpose_eliminate
 1.40% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.68% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     12: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000692    15
47.83% :     0.000331s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.21% :     0.000084s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.96% :     0.000276s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015388    72
 0.30% :     0.000047s :      1: add_recomputation
 0.54% :     0.000083s :      1: auto_monad
 0.10% :     0.000016s :      1: auto_monad_reorder
 1.70% :     0.000262s :      1: bootstrap
 0.17% :     0.000026s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.14% :     0.000022s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
33.76% :     0.005195s :      1: jit_opt_a
 0.89% :     0.000137s :      1: jit_opt_after_cconv
 0.37% :     0.000056s :      1: jit_opt_b
 3.02% :     0.000464s :      1: loop_unroll
 3.53% :     0.000544s :      1: mutable_eliminate
 6.69% :     0.001030s :     26: opt.transform.jit_opt_a
 0.29% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000023s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.10% :     0.000476s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.50% :     0.000076s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000015s :      1: remove_dup_value
 2.13% :     0.000328s :      1: renormalize.infer
 1.67% :     0.000257s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000036s :      1: rewriter_after_opt_a
 0.56% :     0.000086s :      1: rewriter_before_opt_a
 0.49% :     0.000075s :      1: symbol_engine_optimizer
38.52% :     0.005928s :      1: type_inference


TotalTime = 0.0177957, [33]
[bootstrap]: 0.00026564
[type_inference]: 0.00423442
[event_method]: 1.66e-05
[auto_monad]: 6.19e-05
[graph_reusing]: 5.33002e-06
[pre_auto_parallel]: 2.43998e-06
[py_interpret_to_execute]: 3.37e-05
[rewriter_before_opt_a]: 6.876e-05
[expand_dump_flag]: 2.96001e-06
[jit_opt_a]: 0.00498764, [2]
    [Cycle 1]: 0.00154135, [27]
        [switch_simplify]: 4.917e-05
        [loop_unroll]: 3.008e-05
        [a_1]: 0.0005101
        [with_stream_mark]: 2.577e-05
        [recompute_prepare]: 7.78999e-06
        [updatestate_depend_eliminate]: 3.6e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 2.25002e-06
        [specialize_transform]: 5.88998e-06
        [updatestate_useless_node_eliminater]: 5.12999e-06
        [accelerated_algorithm]: 5.20001e-06
        [meta_shard_fg_expand]: 2.46e-06
        [get_grad_eliminate_]: 5.02e-06
        [merge_forward]: 3.51999e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.362e-05
        [j_node_and_user_rematch]: 8.39998e-06
        [meta_fg_expand]: 2.74001e-06
        [replace_old_param]: 8.92e-06
        [inline_without_move]: 5.09e-06
        [renormalize]: 0.00062042
        [add_forward_monad_depend]: 5.26002e-06
        [auto_monad_grad]: 2.51e-06
        [auto_monad_eliminator]: 1.382e-05
        [cse]: 2.185e-05
        [replace_applicator]: 1.144e-05
    [Cycle 2]: 0.00033803, [27]
        [switch_simplify]: 5.78997e-06
        [loop_unroll]: 4.88001e-06
        [a_1]: 9.461e-05
        [with_stream_mark]: 9.11002e-06
        [recompute_prepare]: 5.84e-06
        [updatestate_depend_eliminate]: 2.76e-06
        [updatestate_assign_eliminate]: 2.20002e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 5.12999e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 1.29998e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.63e-06
        [cell_reuse_recompute_pass]: 2.07999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.18e-05
        [j_node_and_user_rematch]: 7.77e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 1.476e-05
        [inline_without_move]: 4.64998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 5.76e-06
        [cse]: 1.108e-05
        [replace_applicator]: 4.98001e-06
[py_interpret_to_execute_after_opt_a]: 1.234e-05
[rewriter_after_opt_a]: 3.629e-05
[convert_after_rewriter]: 6.91001e-06
[order_py_execute_after_rewriter]: 5.33002e-06
[mutable_eliminate]: 0.00049991
[jit_opt_b]: 4.967e-05, [1]
    [Cycle 1]: 4.211e-05, [2]
        [frontend_op_eliminate]: 1.666e-05
        [inline_after_opt_a]: 1.457e-05
[cconv]: 2.408e-05
[loop_unroll]: 0.00043041
[jit_opt_after_cconv]: 0.00013919, [1]
    [Cycle 1]: 0.0001323, [11]
        [c_1]: 2.096e-05
        [parameter_eliminate]: 2.59001e-06
        [updatestate_depend_eliminate]: 5.42001e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 2.27001e-06
        [cse]: 1.726e-05
        [call_graph_tuple_transform]: 1.738e-05
        [tuple_list_get_item_eliminator]: 5.66998e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 6.40022e-07
        [switch_simplify]: 5.36998e-06
[remove_dup_value]: 1.939e-05
[partial_unused_args_eliminate]: 2.46998e-06
[environ_conv]: 5.40999e-06
[add_recomputation]: 4.802e-05
[cse_after_recomputation]: 2.033e-05, [1]
    [Cycle 1]: 1.495e-05, [1]
        [cse]: 9.47999e-06
[auto_monad_reorder]: 1.44e-05
[get_jit_bprop_graph]: 1.69e-06
[rewriter_after_jit_bprop_graph]: 5.39e-06
[opt_after_jit_grad]: 0.00045451
[symbol_engine_optimizer]: 7.325e-05, [1]
    [Cycle 1]: 6.665e-05, [6]
        [build]: 3.28e-06
        [elim_shapecalc]: 7.92e-06
        [elim_not_effective]: 1.338e-05
        [opt_reshape]: 5.64e-06
        [fold_const_symbol]: 8.27e-06
        [renormalize]: 4.60015e-07
[validate]: 3.338e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00604167
[execute]: 7.72002e-06
Sums
    bootstrap                            :     0.000266s :  1.88%
    type_inference                       :     0.004234s : 30.03%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000062s :  0.44%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.24%
    rewriter_before_opt_a                :     0.000069s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.25%
    jit_opt_a.a_1                        :     0.000605s :  4.29%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000024s :  0.17%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000620s :  4.40%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.14%
    jit_opt_a.cse                        :     0.000033s :  0.23%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.09%
    rewriter_after_opt_a                 :     0.000036s :  0.26%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000500s :  3.55%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000430s :  3.05%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000019s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000048s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000455s :  3.22%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006042s : 42.85%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000183    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.53% :     0.000005s :      3: substitution.graph_param_transform
83.25% :     0.000153s :      4: substitution.inline
 1.47% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.30% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.04% :     0.000004s :      2: substitution.replace_old_param
 6.63% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004176     2
81.18% :     0.003390s :      1: type_inference.infer
18.82% :     0.000786s :      1: type_inference.specialize
------[replace.]   0.000064     6
66.39% :     0.000042s :      4: replace.inline
33.61% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000161     6
93.14% :     0.000150s :      4: match.inline
 6.86% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 1.93% :     0.000002s :     10: predicate.arithmetic_simplify
 1.45% :     0.000002s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.47% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.10% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.45% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.02% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.68% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.24% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.10% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.81% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.07% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.13% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.64% :     0.000009s :     53: predicate.switch_simplify
 1.33% :     0.000002s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.52% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.17% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000523    12
46.95% :     0.000245s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.68% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.37% :     0.000248s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019337    76
 0.26% :     0.000051s :      1: add_recomputation
 0.34% :     0.000065s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.44% :     0.000278s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.81% :     0.004990s :      1: jit_opt_a
 0.73% :     0.000142s :      1: jit_opt_after_cconv
 0.27% :     0.000053s :      1: jit_opt_b
 2.27% :     0.000439s :      1: loop_unroll
 2.63% :     0.000508s :      1: mutable_eliminate
 4.21% :     0.000813s :     26: opt.transform.jit_opt_a
 0.24% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.39% :     0.000463s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000036s :      1: py_interpret_to_execute
 0.08% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000022s :      1: remove_dup_value
 1.70% :     0.000328s :      1: renormalize.infer
 1.47% :     0.000285s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000039s :      1: rewriter_after_opt_a
 0.37% :     0.000071s :      1: rewriter_before_opt_a
 0.39% :     0.000076s :      1: symbol_engine_optimizer
31.31% :     0.006055s :      1: task_emit
21.95% :     0.004245s :      1: type_inference
 0.28% :     0.000055s :      1: validate


TotalTime = 0.0145821, [30]
[bootstrap]: 0.00030806
[type_inference]: 0.00638333
[event_method]: 2.057e-05
[auto_monad]: 8.01e-05
[graph_reusing]: 6.73e-06
[pre_auto_parallel]: 2.81e-06
[py_interpret_to_execute]: 4.611e-05
[rewriter_before_opt_a]: 8.432e-05
[expand_dump_flag]: 3.28998e-06
[jit_opt_a]: 0.00553806, [2]
    [Cycle 1]: 0.00189082, [27]
        [switch_simplify]: 0.00011603
        [loop_unroll]: 5.827e-05
        [a_1]: 0.00068346
        [with_stream_mark]: 1.885e-05
        [recompute_prepare]: 6.66e-06
        [updatestate_depend_eliminate]: 3.33e-06
        [updatestate_assign_eliminate]: 3.09001e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [parameter_eliminate]: 1.99999e-06
        [specialize_transform]: 5.76998e-06
        [updatestate_useless_node_eliminater]: 4.97999e-06
        [accelerated_algorithm]: 5.45001e-06
        [meta_shard_fg_expand]: 2.02001e-06
        [get_grad_eliminate_]: 5.49998e-06
        [merge_forward]: 3.45998e-06
        [cell_reuse_recompute_pass]: 1.31002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.39e-05
        [j_node_and_user_rematch]: 9.80002e-06
        [meta_fg_expand]: 2.12999e-06
        [replace_old_param]: 8.89e-06
        [inline_without_move]: 5.35999e-06
        [renormalize]: 0.00069338
        [add_forward_monad_depend]: 5.44e-06
        [auto_monad_grad]: 2.38998e-06
        [auto_monad_eliminator]: 1.348e-05
        [cse]: 2.809e-05
        [replace_applicator]: 1.12e-05
    [Cycle 2]: 0.00030706, [27]
        [switch_simplify]: 5.38002e-06
        [loop_unroll]: 4.93001e-06
        [a_1]: 8.924e-05
        [with_stream_mark]: 8.94e-06
        [recompute_prepare]: 4.99e-06
        [updatestate_depend_eliminate]: 2.61999e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 9.29984e-07
        [specialize_transform]: 4.92e-06
        [updatestate_useless_node_eliminater]: 4.78001e-06
        [accelerated_algorithm]: 4.74998e-06
        [meta_shard_fg_expand]: 1.21002e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.68e-06
        [cell_reuse_recompute_pass]: 1.20999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.201e-05
        [j_node_and_user_rematch]: 7.48e-06
        [meta_fg_expand]: 1.42999e-06
        [replace_old_param]: 6.56e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.32999e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 4.61002e-06
        [cse]: 9.37999e-06
        [replace_applicator]: 5.12e-06
[py_interpret_to_execute_after_opt_a]: 1.013e-05
[rewriter_after_opt_a]: 3.341e-05
[convert_after_rewriter]: 7.39002e-06
[order_py_execute_after_rewriter]: 5.00999e-06
[mutable_eliminate]: 0.00052629
[jit_opt_b]: 4.875e-05, [1]
    [Cycle 1]: 4.181e-05, [2]
        [frontend_op_eliminate]: 1.582e-05
        [inline_after_opt_a]: 1.42e-05
[cconv]: 2.341e-05
[loop_unroll]: 0.00043243
[jit_opt_after_cconv]: 0.00013411, [1]
    [Cycle 1]: 0.00012783, [11]
        [c_1]: 2.106e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 5.31002e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.608e-05
        [call_graph_tuple_transform]: 1.752e-05
        [tuple_list_get_item_eliminator]: 5.07e-06
        [none_parameter_eliminate]: 1.43002e-06
        [renormalize]: 5.3001e-07
        [switch_simplify]: 5.69e-06
[remove_dup_value]: 2.242e-05
[partial_unused_args_eliminate]: 2.53e-06
[environ_conv]: 4.75001e-06
[add_recomputation]: 6.696e-05
[cse_after_recomputation]: 2.197e-05, [1]
    [Cycle 1]: 1.629e-05, [1]
        [cse]: 9.96e-06
[auto_monad_reorder]: 3.558e-05
[get_jit_bprop_graph]: 1.87001e-06
[rewriter_after_jit_bprop_graph]: 5.72999e-06
[opt_after_jit_grad]: 0.00046367
[symbol_engine_optimizer]: 7.116e-05, [1]
    [Cycle 1]: 6.478e-05, [6]
        [build]: 3.48e-06
        [elim_shapecalc]: 7.06001e-06
        [elim_not_effective]: 1.214e-05
        [opt_reshape]: 5.45001e-06
        [fold_const_symbol]: 8.52e-06
        [renormalize]: 4.19997e-07
[validate]: 3.122e-05
Sums
    bootstrap                            :     0.000308s :  2.88%
    type_inference                       :     0.006383s : 59.77%
    event_method                         :     0.000021s :  0.19%
    auto_monad                           :     0.000080s :  0.75%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000046s :  0.43%
    rewriter_before_opt_a                :     0.000084s :  0.79%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000121s :  1.14%
    jit_opt_a.loop_unroll                :     0.000063s :  0.59%
    jit_opt_a.a_1                        :     0.000773s :  7.23%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000693s :  6.49%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.17%
    jit_opt_a.cse                        :     0.000037s :  0.35%
    jit_opt_a.replace_applicator         :     0.000016s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.09%
    rewriter_after_opt_a                 :     0.000033s :  0.31%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000526s :  4.93%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.13%
    cconv                                :     0.000023s :  0.22%
    loop_unroll                          :     0.000432s :  4.05%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000022s :  0.21%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000067s :  0.63%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000036s :  0.33%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000464s :  4.34%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.29%


Time group info:
------[substitution.]   0.000283    26
 0.67% :     0.000002s :      2: substitution.elim_not_effective
 0.42% :     0.000001s :      2: substitution.fold_const_symbol
 1.89% :     0.000005s :      3: substitution.graph_param_transform
84.98% :     0.000240s :      6: substitution.inline
 1.22% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.58% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.39% :     0.000004s :      2: substitution.replace_old_param
 3.11% :     0.000009s :      1: substitution.switch_simplify
 4.73% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006322     2
83.53% :     0.005280s :      1: type_inference.infer
16.47% :     0.001041s :      1: type_inference.specialize
------[replace.]   0.000130     9
49.24% :     0.000064s :      6: replace.inline
27.80% :     0.000036s :      1: replace.switch_simplify
22.96% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000257     9
92.12% :     0.000237s :      6: match.inline
 3.10% :     0.000008s :      1: match.switch_simplify
 4.78% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   801
 1.36% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.99% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000001s :     12: predicate.addn_check_dump
 1.50% :     0.000002s :     12: predicate.addn_zero_filter
 1.95% :     0.000002s :     12: predicate.arithmetic_simplify
 1.27% :     0.000002s :     12: predicate.cast_eliminate
 0.73% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     12: predicate.compare_switch_simplify
 1.37% :     0.000002s :     12: predicate.depend_value_elim
 1.20% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     12: predicate.environ_get_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.12% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.41% :     0.000007s :     26: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     17: predicate.load_eliminater
 1.60% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000002s :     12: predicate.merge_addn
 1.20% :     0.000002s :     12: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     17: predicate.partial_eliminate
 1.32% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     12: predicate.reduce_eliminate
 1.52% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     20: predicate.replace_applicator
 0.45% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     12: predicate.reshape_eliminate
 1.48% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.54% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     12: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.80% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.63% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.80% :     0.000004s :     20: predicate.switch_defer_inline
 2.41% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.21% :     0.000011s :     65: predicate.switch_simplify
 1.21% :     0.000002s :     12: predicate.tile_eliminate
 1.23% :     0.000002s :     12: predicate.transpose_eliminate
 1.44% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.01% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     12: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000807    15
52.88% :     0.000427s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.29% :     0.000099s :      1: func_graph_cloner_run.FuncGraphClonerNode
34.82% :     0.000281s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016401    72
 0.43% :     0.000070s :      1: add_recomputation
 0.51% :     0.000083s :      1: auto_monad
 0.23% :     0.000038s :      1: auto_monad_reorder
 1.97% :     0.000323s :      1: bootstrap
 0.16% :     0.000026s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
33.79% :     0.005542s :      1: jit_opt_a
 0.83% :     0.000137s :      1: jit_opt_after_cconv
 0.31% :     0.000051s :      1: jit_opt_b
 2.69% :     0.000441s :      1: loop_unroll
 3.26% :     0.000535s :      1: mutable_eliminate
 6.51% :     0.001067s :     26: opt.transform.jit_opt_a
 0.28% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.87% :     0.000472s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000049s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.15% :     0.000025s :      1: remove_dup_value
 2.48% :     0.000406s :      1: renormalize.infer
 1.71% :     0.000280s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000036s :      1: rewriter_after_opt_a
 0.53% :     0.000087s :      1: rewriter_before_opt_a
 0.45% :     0.000074s :      1: symbol_engine_optimizer
38.99% :     0.006396s :      1: type_inference


TotalTime = 0.0215083, [33]
[bootstrap]: 0.00045047
[type_inference]: 0.00605322
[event_method]: 2.033e-05
[auto_monad]: 6.74e-05
[graph_reusing]: 6.28e-06
[pre_auto_parallel]: 3.08e-06
[py_interpret_to_execute]: 3.612e-05
[rewriter_before_opt_a]: 7.414e-05
[expand_dump_flag]: 3.35e-06
[jit_opt_a]: 0.00575848, [2]
    [Cycle 1]: 0.00176466, [27]
        [switch_simplify]: 5.422e-05
        [loop_unroll]: 2.879e-05
        [a_1]: 0.00054824
        [with_stream_mark]: 2.201e-05
        [recompute_prepare]: 6.78e-06
        [updatestate_depend_eliminate]: 4.24002e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 2.66999e-06
        [parameter_eliminate]: 2.02999e-06
        [specialize_transform]: 1.381e-05
        [updatestate_useless_node_eliminater]: 5.19998e-06
        [accelerated_algorithm]: 5.34998e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 5.24e-06
        [merge_forward]: 3.69002e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.485e-05
        [j_node_and_user_rematch]: 8.55999e-06
        [meta_fg_expand]: 2.30002e-06
        [replace_old_param]: 9.87999e-06
        [inline_without_move]: 5.47999e-06
        [renormalize]: 0.00077516
        [add_forward_monad_depend]: 1.072e-05
        [auto_monad_grad]: 2.92002e-06
        [auto_monad_eliminator]: 1.372e-05
        [cse]: 2.865e-05
        [replace_applicator]: 1.229e-05
    [Cycle 2]: 0.00040007, [27]
        [switch_simplify]: 5.46e-06
        [loop_unroll]: 4.80001e-06
        [a_1]: 8.931e-05
        [with_stream_mark]: 9.91998e-06
        [recompute_prepare]: 4.85001e-06
        [updatestate_depend_eliminate]: 2.58e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.17999e-06
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 4.72e-06
        [accelerated_algorithm]: 5.12999e-06
        [meta_shard_fg_expand]: 1.24998e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.43002e-06
        [cell_reuse_recompute_pass]: 1.94e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.196e-05
        [j_node_and_user_rematch]: 1.985e-05
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 7.15998e-06
        [inline_without_move]: 4.73001e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.60999e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 5.61e-06
        [cse]: 1.208e-05
        [replace_applicator]: 5.32001e-06
[py_interpret_to_execute_after_opt_a]: 1.157e-05
[rewriter_after_opt_a]: 3.579e-05
[convert_after_rewriter]: 6.78e-06
[order_py_execute_after_rewriter]: 5.27001e-06
[mutable_eliminate]: 0.00057299
[jit_opt_b]: 4.764e-05, [1]
    [Cycle 1]: 4.01e-05, [2]
        [frontend_op_eliminate]: 1.5e-05
        [inline_after_opt_a]: 1.432e-05
[cconv]: 2.603e-05
[loop_unroll]: 0.0004308
[jit_opt_after_cconv]: 0.00013864, [1]
    [Cycle 1]: 0.0001319, [11]
        [c_1]: 2.04e-05
        [parameter_eliminate]: 3.00002e-06
        [updatestate_depend_eliminate]: 5.92001e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.19999e-06
        [cse]: 1.831e-05
        [call_graph_tuple_transform]: 1.85e-05
        [tuple_list_get_item_eliminator]: 5.16002e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 6.69999e-07
        [switch_simplify]: 5.20001e-06
[remove_dup_value]: 1.443e-05
[partial_unused_args_eliminate]: 2.69999e-06
[environ_conv]: 5.53002e-06
[add_recomputation]: 5.04e-05
[cse_after_recomputation]: 1.969e-05, [1]
    [Cycle 1]: 1.408e-05, [1]
        [cse]: 8.37e-06
[auto_monad_reorder]: 1.482e-05
[get_jit_bprop_graph]: 1.97999e-06
[rewriter_after_jit_bprop_graph]: 5.97999e-06
[opt_after_jit_grad]: 0.0004537
[symbol_engine_optimizer]: 6.979e-05, [1]
    [Cycle 1]: 6.343e-05, [6]
        [build]: 3.48e-06
        [elim_shapecalc]: 7.16001e-06
        [elim_not_effective]: 1.241e-05
        [opt_reshape]: 5.27999e-06
        [fold_const_symbol]: 8.18001e-06
        [renormalize]: 5.19998e-07
[validate]: 3.742e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.0068497
[execute]: 8.59e-06
Sums
    bootstrap                            :     0.000450s :  2.62%
    type_inference                       :     0.006053s : 35.15%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000067s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.21%
    rewriter_before_opt_a                :     0.000074s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000060s :  0.35%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000638s :  3.70%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000019s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000775s :  4.50%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.11%
    jit_opt_a.cse                        :     0.000041s :  0.24%
    jit_opt_a.replace_applicator         :     0.000018s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000573s :  3.33%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.08%
    cconv                                :     0.000026s :  0.15%
    loop_unroll                          :     0.000431s :  2.50%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000050s :  0.29%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000454s :  2.63%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006850s : 39.78%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000217    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.61% :     0.000006s :      3: substitution.graph_param_transform
85.05% :     0.000184s :      4: substitution.inline
 1.35% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.04% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.99% :     0.000004s :      2: substitution.replace_old_param
 5.48% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005971     2
84.32% :     0.005035s :      1: type_inference.infer
15.68% :     0.000936s :      1: type_inference.specialize
------[replace.]   0.000066     6
68.08% :     0.000045s :      4: replace.inline
31.92% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     6
94.42% :     0.000182s :      4: match.inline
 5.58% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.19% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 2.50% :     0.000003s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.55% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.40% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.45% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      6: predicate.less_batch_normalization
 1.67% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.19% :     0.000001s :     10: predicate.minmaximum_grad
 2.18% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.23% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.42% :     0.000002s :     10: predicate.reduce_eliminate
 1.61% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 1.12% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.33% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.25% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000008s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.12% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000658    12
51.11% :     0.000336s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.84% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.05% :     0.000277s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023238    76
 0.23% :     0.000053s :      1: add_recomputation
 0.30% :     0.000071s :      1: auto_monad
 0.07% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 2.04% :     0.000474s :      1: bootstrap
 0.12% :     0.000029s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.11% :     0.000026s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
24.79% :     0.005762s :      1: jit_opt_a
 0.61% :     0.000141s :      1: jit_opt_after_cconv
 0.22% :     0.000050s :      1: jit_opt_b
 1.89% :     0.000439s :      1: loop_unroll
 2.51% :     0.000583s :      1: mutable_eliminate
 3.72% :     0.000863s :     26: opt.transform.jit_opt_a
 0.20% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000023s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000030s :      4: opt.transform.symbol_engine_opt
 1.99% :     0.000462s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000039s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.97% :     0.000457s :      1: renormalize.infer
 1.34% :     0.000311s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000039s :      1: rewriter_after_opt_a
 0.33% :     0.000077s :      1: rewriter_before_opt_a
 0.31% :     0.000072s :      1: symbol_engine_optimizer
29.55% :     0.006866s :      1: task_emit
26.14% :     0.006074s :      1: type_inference
 0.25% :     0.000059s :      1: validate


TotalTime = 0.0182845, [33]
[bootstrap]: 0.00027362
[type_inference]: 0.00422908
[event_method]: 1.637e-05
[auto_monad]: 6.247e-05
[graph_reusing]: 5.61998e-06
[pre_auto_parallel]: 2.26e-06
[py_interpret_to_execute]: 3.164e-05
[rewriter_before_opt_a]: 6.854e-05
[expand_dump_flag]: 2.76999e-06
[jit_opt_a]: 0.00519207, [2]
    [Cycle 1]: 0.00155641, [27]
        [switch_simplify]: 5.24e-05
        [loop_unroll]: 3.071e-05
        [a_1]: 0.0005363
        [with_stream_mark]: 1.702e-05
        [recompute_prepare]: 7.40003e-06
        [updatestate_depend_eliminate]: 3.70998e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.86e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 5.93998e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 5.87999e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 5.97001e-06
        [merge_forward]: 4.08999e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.528e-05
        [j_node_and_user_rematch]: 9.04e-06
        [meta_fg_expand]: 2.56e-06
        [replace_old_param]: 8.60999e-06
        [inline_without_move]: 5.19e-06
        [renormalize]: 0.0006118
        [add_forward_monad_depend]: 5.22e-06
        [auto_monad_grad]: 1.98002e-06
        [auto_monad_eliminator]: 1.382e-05
        [cse]: 1.862e-05
        [replace_applicator]: 1.187e-05
    [Cycle 2]: 0.00035042, [27]
        [switch_simplify]: 5.90002e-06
        [loop_unroll]: 5.35001e-06
        [a_1]: 9.198e-05
        [with_stream_mark]: 1.134e-05
        [recompute_prepare]: 5.46e-06
        [updatestate_depend_eliminate]: 2.89999e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 2.08998e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 5.51998e-06
        [updatestate_useless_node_eliminater]: 5.43002e-06
        [accelerated_algorithm]: 6.01e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 4.94003e-06
        [merge_forward]: 2.94999e-06
        [cell_reuse_recompute_pass]: 1.79e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.399e-05
        [j_node_and_user_rematch]: 8.85999e-06
        [meta_fg_expand]: 1.61998e-06
        [replace_old_param]: 8.12e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 5.74999e-06
        [cse]: 1.149e-05
        [replace_applicator]: 1.406e-05
[py_interpret_to_execute_after_opt_a]: 1.153e-05
[rewriter_after_opt_a]: 3.454e-05
[convert_after_rewriter]: 8e-06
[order_py_execute_after_rewriter]: 5.72999e-06
[mutable_eliminate]: 0.00054455
[jit_opt_b]: 5.497e-05, [1]
    [Cycle 1]: 4.759e-05, [2]
        [frontend_op_eliminate]: 1.905e-05
        [inline_after_opt_a]: 1.615e-05
[cconv]: 2.455e-05
[loop_unroll]: 0.00044915
[jit_opt_after_cconv]: 0.0002075, [1]
    [Cycle 1]: 0.00020042, [11]
        [c_1]: 2.228e-05
        [parameter_eliminate]: 3.23e-06
        [updatestate_depend_eliminate]: 6.17001e-06
        [updatestate_assign_eliminate]: 2.50002e-06
        [updatestate_loads_eliminate]: 2.11998e-06
        [cse]: 1.675e-05
        [call_graph_tuple_transform]: 1.925e-05
        [tuple_list_get_item_eliminator]: 5.49e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 3.50003e-07
        [switch_simplify]: 6.56e-05
[remove_dup_value]: 1.261e-05
[partial_unused_args_eliminate]: 2.71999e-06
[environ_conv]: 4.75001e-06
[add_recomputation]: 4.691e-05
[cse_after_recomputation]: 2.361e-05, [1]
    [Cycle 1]: 1.755e-05, [1]
        [cse]: 1.112e-05
[auto_monad_reorder]: 1.485e-05
[get_jit_bprop_graph]: 2.01998e-06
[rewriter_after_jit_bprop_graph]: 6.21e-06
[opt_after_jit_grad]: 0.0004921
[symbol_engine_optimizer]: 7.469e-05, [1]
    [Cycle 1]: 6.811e-05, [6]
        [build]: 3.76001e-06
        [elim_shapecalc]: 7.46001e-06
        [elim_not_effective]: 1.354e-05
        [opt_reshape]: 5.74999e-06
        [fold_const_symbol]: 8.98002e-06
        [renormalize]: 4.10015e-07
[validate]: 3.596e-05
[backend_pass]: 8.80013e-07
[task_emit]: 0.00613787
[execute]: 8.85999e-06
Sums
    bootstrap                            :     0.000274s :  1.90%
    type_inference                       :     0.004229s : 29.40%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000062s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.22%
    rewriter_before_opt_a                :     0.000069s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.41%
    jit_opt_a.loop_unroll                :     0.000036s :  0.25%
    jit_opt_a.a_1                        :     0.000628s :  4.37%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.08%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000612s :  4.25%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.14%
    jit_opt_a.cse                        :     0.000030s :  0.21%
    jit_opt_a.replace_applicator         :     0.000026s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000035s :  0.24%
    convert_after_rewriter               :     0.000008s :  0.06%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000545s :  3.79%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000449s :  3.12%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000066s :  0.46%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000047s :  0.33%
    cse_after_recomputation.cse          :     0.000011s :  0.08%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000492s :  3.42%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006138s : 42.67%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000202    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.76% :     0.000006s :      3: substitution.graph_param_transform
84.31% :     0.000170s :      4: substitution.inline
 1.44% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.44% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.81% :     0.000004s :      2: substitution.replace_old_param
 5.53% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004170     2
81.60% :     0.003403s :      1: type_inference.infer
18.40% :     0.000767s :      1: type_inference.specialize
------[replace.]   0.000066     6
67.09% :     0.000045s :      4: replace.inline
32.91% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000177     6
94.34% :     0.000167s :      4: match.inline
 5.66% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.22% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.18% :     0.000001s :     10: predicate.addn_zero_filter
 1.87% :     0.000002s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.50% :     0.000002s :     10: predicate.depend_value_elim
 1.27% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.50% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.09% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000001s :      6: predicate.less_batch_normalization
 1.46% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.81% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.87% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.02% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 1.16% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.00% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.51% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.93% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.11% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000520    12
45.82% :     0.000238s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.68% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.50% :     0.000242s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019923    76
 0.25% :     0.000050s :      1: add_recomputation
 0.33% :     0.000066s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.43% :     0.000285s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.13% :     0.000026s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.08% :     0.005195s :      1: jit_opt_a
 1.06% :     0.000211s :      1: jit_opt_after_cconv
 0.29% :     0.000058s :      1: jit_opt_b
 2.30% :     0.000458s :      1: loop_unroll
 2.78% :     0.000554s :      1: mutable_eliminate
 4.28% :     0.000852s :     26: opt.transform.jit_opt_a
 0.54% :     0.000108s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000028s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.51% :     0.000500s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000034s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.64% :     0.000327s :      1: renormalize.infer
 1.39% :     0.000277s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000038s :      1: rewriter_after_opt_a
 0.36% :     0.000071s :      1: rewriter_before_opt_a
 0.39% :     0.000078s :      1: symbol_engine_optimizer
30.89% :     0.006154s :      1: task_emit
21.28% :     0.004240s :      1: type_inference
 0.28% :     0.000056s :      1: validate


TotalTime = 0.0266639, [33]
[bootstrap]: 0.00128788
[type_inference]: 0.0056899
[event_method]: 2.107e-05
[auto_monad]: 7.364e-05
[graph_reusing]: 6.06998e-06
[pre_auto_parallel]: 3.01001e-06
[py_interpret_to_execute]: 0.00096898
[rewriter_before_opt_a]: 0.00010688
[expand_dump_flag]: 5.18002e-06
[jit_opt_a]: 0.00734952, [2]
    [Cycle 1]: 0.0021024, [27]
        [switch_simplify]: 7.531e-05
        [loop_unroll]: 3.084e-05
        [a_1]: 0.00065059
        [with_stream_mark]: 2.253e-05
        [recompute_prepare]: 7.6e-06
        [updatestate_depend_eliminate]: 4.04997e-06
        [updatestate_assign_eliminate]: 3.11001e-06
        [updatestate_loads_eliminate]: 2.99999e-06
        [parameter_eliminate]: 1.83002e-06
        [specialize_transform]: 6.60002e-06
        [updatestate_useless_node_eliminater]: 5.47001e-06
        [accelerated_algorithm]: 6.07001e-06
        [meta_shard_fg_expand]: 2.18002e-06
        [get_grad_eliminate_]: 5.81003e-06
        [merge_forward]: 4.02e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.547e-05
        [j_node_and_user_rematch]: 1.024e-05
        [meta_fg_expand]: 2.89001e-06
        [replace_old_param]: 9.55001e-06
        [inline_without_move]: 5.82001e-06
        [renormalize]: 0.00097561
        [add_forward_monad_depend]: 6.91999e-06
        [auto_monad_grad]: 2.67001e-06
        [auto_monad_eliminator]: 1.505e-05
        [cse]: 2.336e-05
        [replace_applicator]: 1.382e-05
    [Cycle 2]: 0.00035307, [27]
        [switch_simplify]: 7.77998e-06
        [loop_unroll]: 5.86e-06
        [a_1]: 9.842e-05
        [with_stream_mark]: 1.328e-05
        [recompute_prepare]: 5.02999e-06
        [updatestate_depend_eliminate]: 2.89001e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 1.06002e-06
        [specialize_transform]: 4.94e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 5.49998e-06
        [meta_shard_fg_expand]: 1.42999e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 2.41e-06
        [cell_reuse_recompute_pass]: 2.98e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.455e-05
        [j_node_and_user_rematch]: 7.98001e-06
        [meta_fg_expand]: 1.53002e-06
        [replace_old_param]: 1.557e-05
        [inline_without_move]: 5.02e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.76e-06
        [auto_monad_grad]: 1.19e-06
        [auto_monad_eliminator]: 6.45002e-06
        [cse]: 1.19e-05
        [replace_applicator]: 5.07999e-06
[py_interpret_to_execute_after_opt_a]: 1.575e-05
[rewriter_after_opt_a]: 3.883e-05
[convert_after_rewriter]: 8.18001e-06
[order_py_execute_after_rewriter]: 5.38002e-06
[mutable_eliminate]: 0.00078041
[jit_opt_b]: 5.694e-05, [1]
    [Cycle 1]: 4.89e-05, [2]
        [frontend_op_eliminate]: 1.922e-05
        [inline_after_opt_a]: 1.694e-05
[cconv]: 2.873e-05
[loop_unroll]: 0.00047712
[jit_opt_after_cconv]: 0.00015251, [1]
    [Cycle 1]: 0.00014552, [11]
        [c_1]: 2.34e-05
        [parameter_eliminate]: 3.26999e-06
        [updatestate_depend_eliminate]: 5.82001e-06
        [updatestate_assign_eliminate]: 2.69999e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [cse]: 2.124e-05
        [call_graph_tuple_transform]: 2.164e-05
        [tuple_list_get_item_eliminator]: 5.46e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 1.05001e-06
        [switch_simplify]: 5.39e-06
[remove_dup_value]: 1.594e-05
[partial_unused_args_eliminate]: 2.46998e-06
[environ_conv]: 5.93002e-06
[add_recomputation]: 6.066e-05
[cse_after_recomputation]: 2.44e-05, [1]
    [Cycle 1]: 1.735e-05, [1]
        [cse]: 1.121e-05
[auto_monad_reorder]: 1.667e-05
[get_jit_bprop_graph]: 2.43e-06
[rewriter_after_jit_bprop_graph]: 7.4e-06
[opt_after_jit_grad]: 0.0005148
[symbol_engine_optimizer]: 8.612e-05, [1]
    [Cycle 1]: 7.759e-05, [6]
        [build]: 4.57998e-06
        [elim_shapecalc]: 9.02999e-06
        [elim_not_effective]: 1.564e-05
        [opt_reshape]: 6.28002e-06
        [fold_const_symbol]: 9.99001e-06
        [renormalize]: 3.39991e-07
[validate]: 4.778e-05
[backend_pass]: 1.17e-06
[task_emit]: 0.00852488
[execute]: 9.51e-06
Sums
    bootstrap                            :     0.001288s :  6.11%
    type_inference                       :     0.005690s : 27.01%
    event_method                         :     0.000021s :  0.10%
    auto_monad                           :     0.000074s :  0.35%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000969s :  4.60%
    rewriter_before_opt_a                :     0.000107s :  0.51%
    expand_dump_flag                     :     0.000005s :  0.02%
    jit_opt_a.switch_simplify            :     0.000083s :  0.39%
    jit_opt_a.loop_unroll                :     0.000037s :  0.17%
    jit_opt_a.a_1                        :     0.000749s :  3.56%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000025s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.000976s :  4.63%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.10%
    jit_opt_a.cse                        :     0.000035s :  0.17%
    jit_opt_a.replace_applicator         :     0.000019s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.07%
    rewriter_after_opt_a                 :     0.000039s :  0.18%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000780s :  3.71%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.08%
    cconv                                :     0.000029s :  0.14%
    loop_unroll                          :     0.000477s :  2.27%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000061s :  0.29%
    cse_after_recomputation.cse          :     0.000011s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000515s :  2.44%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000048s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008525s : 40.47%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000281    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.25% :     0.000006s :      3: substitution.graph_param_transform
83.20% :     0.000234s :      4: substitution.inline
 1.38% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.90% :     0.000005s :      4: substitution.remove_not_recompute_node
 4.46% :     0.000013s :      2: substitution.replace_old_param
 5.50% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005600     2
81.01% :     0.004537s :      1: type_inference.infer
18.99% :     0.001063s :      1: type_inference.specialize
------[replace.]   0.000079     6
66.68% :     0.000052s :      4: replace.inline
33.32% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000246     6
94.18% :     0.000232s :      4: match.inline
 5.82% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.50% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 2.67% :     0.000004s :     10: predicate.arithmetic_simplify
 1.12% :     0.000001s :     10: predicate.cast_eliminate
 0.59% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.88% :     0.000002s :     10: predicate.depend_value_elim
 1.23% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.58% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.91% :     0.000008s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.55% :     0.000002s :     15: predicate.load_eliminater
 1.23% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.56% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.46% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 1.97% :     0.000003s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.00% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     10: predicate.reduce_eliminate
 1.79% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.44% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.88% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.04% :     0.000003s :     16: predicate.switch_defer_inline
 1.90% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.45% :     0.000010s :     53: predicate.switch_simplify
 1.33% :     0.000002s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.71% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.22% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.48% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.61% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.86% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000754    12
46.88% :     0.000354s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.11% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.00% :     0.000347s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028741    76
 0.23% :     0.000065s :      1: add_recomputation
 0.27% :     0.000078s :      1: auto_monad
 0.07% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 4.56% :     0.001309s :      1: bootstrap
 0.11% :     0.000032s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000027s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.09% :     0.000026s :      1: event_method
 0.05% :     0.000015s :      1: execute
 0.03% :     0.000007s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
25.58% :     0.007353s :      1: jit_opt_a
 0.54% :     0.000155s :      1: jit_opt_after_cconv
 0.21% :     0.000060s :      1: jit_opt_b
 1.69% :     0.000485s :      1: loop_unroll
 2.75% :     0.000789s :      1: mutable_eliminate
 3.48% :     0.000999s :     26: opt.transform.jit_opt_a
 0.18% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000029s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.82% :     0.000524s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 3.42% :     0.000984s :      1: py_interpret_to_execute
 0.06% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000018s :      1: remove_dup_value
 1.89% :     0.000544s :      1: renormalize.infer
 1.47% :     0.000422s :      1: renormalize.specialize
 0.03% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000042s :      1: rewriter_after_opt_a
 0.39% :     0.000113s :      1: rewriter_before_opt_a
 0.31% :     0.000089s :      1: symbol_engine_optimizer
29.72% :     0.008543s :      1: task_emit
19.88% :     0.005713s :      1: type_inference
 0.25% :     0.000071s :      1: validate


TotalTime = 0.0257717, [33]
[bootstrap]: 0.00027183
[type_inference]: 0.00429609
[event_method]: 1.647e-05
[auto_monad]: 6.125e-05
[graph_reusing]: 5.42001e-06
[pre_auto_parallel]: 2.44001e-06
[py_interpret_to_execute]: 3.403e-05
[rewriter_before_opt_a]: 7.249e-05
[expand_dump_flag]: 3.61001e-06
[jit_opt_a]: 0.0124606, [2]
    [Cycle 1]: 0.00875185, [27]
        [switch_simplify]: 5.601e-05
        [loop_unroll]: 3.092e-05
        [a_1]: 0.00767987
        [with_stream_mark]: 1.738e-05
        [recompute_prepare]: 9.47001e-06
        [updatestate_depend_eliminate]: 3.3e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 5.98002e-06
        [updatestate_useless_node_eliminater]: 5.31002e-06
        [accelerated_algorithm]: 5.92999e-06
        [meta_shard_fg_expand]: 2.17001e-06
        [get_grad_eliminate_]: 5.68997e-06
        [merge_forward]: 4.06001e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.38e-05
        [j_node_and_user_rematch]: 8.89003e-06
        [meta_fg_expand]: 2.44001e-06
        [replace_old_param]: 8.99998e-06
        [inline_without_move]: 5.40001e-06
        [renormalize]: 0.00063449
        [add_forward_monad_depend]: 5.44e-06
        [auto_monad_grad]: 2.31e-06
        [auto_monad_eliminator]: 1.334e-05
        [cse]: 2.714e-05
        [replace_applicator]: 1.224e-05
    [Cycle 2]: 0.00032528, [27]
        [switch_simplify]: 5.87999e-06
        [loop_unroll]: 5.22e-06
        [a_1]: 9.281e-05
        [with_stream_mark]: 8.80001e-06
        [recompute_prepare]: 5.08002e-06
        [updatestate_depend_eliminate]: 2.37999e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 5.14e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 5.08002e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.55999e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.72999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.269e-05
        [j_node_and_user_rematch]: 7.90998e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 7.00998e-06
        [inline_without_move]: 4.90999e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 5.34e-06
        [cse]: 1.015e-05
        [replace_applicator]: 5.56e-06
[py_interpret_to_execute_after_opt_a]: 1.072e-05
[rewriter_after_opt_a]: 3.532e-05
[convert_after_rewriter]: 6.63003e-06
[order_py_execute_after_rewriter]: 5.07e-06
[mutable_eliminate]: 0.00082579
[jit_opt_b]: 5.092e-05, [1]
    [Cycle 1]: 4.301e-05, [2]
        [frontend_op_eliminate]: 1.657e-05
        [inline_after_opt_a]: 1.476e-05
[cconv]: 2.636e-05
[loop_unroll]: 0.00043329
[jit_opt_after_cconv]: 0.0001368, [1]
    [Cycle 1]: 0.00012997, [11]
        [c_1]: 2.231e-05
        [parameter_eliminate]: 2.32999e-06
        [updatestate_depend_eliminate]: 4.94e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 1.599e-05
        [call_graph_tuple_transform]: 1.82e-05
        [tuple_list_get_item_eliminator]: 5.36002e-06
        [none_parameter_eliminate]: 1.31002e-06
        [renormalize]: 4.40021e-07
        [switch_simplify]: 5.29e-06
[remove_dup_value]: 1.316e-05
[partial_unused_args_eliminate]: 2.25002e-06
[environ_conv]: 5.25001e-06
[add_recomputation]: 4.371e-05
[cse_after_recomputation]: 1.904e-05, [1]
    [Cycle 1]: 1.394e-05, [1]
        [cse]: 8.50001e-06
[auto_monad_reorder]: 1.491e-05
[get_jit_bprop_graph]: 1.94e-06
[rewriter_after_jit_bprop_graph]: 5.04998e-06
[opt_after_jit_grad]: 0.00045877
[symbol_engine_optimizer]: 7.175e-05, [1]
    [Cycle 1]: 6.534e-05, [6]
        [build]: 2.99999e-06
        [elim_shapecalc]: 7.63001e-06
        [elim_not_effective]: 1.275e-05
        [opt_reshape]: 5.62001e-06
        [fold_const_symbol]: 8.57e-06
        [renormalize]: 3.7998e-07
[validate]: 3.091e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.00615989
[execute]: 7.34002e-06
Sums
    bootstrap                            :     0.000272s :  1.25%
    type_inference                       :     0.004296s : 19.72%
    event_method                         :     0.000016s :  0.08%
    auto_monad                           :     0.000061s :  0.28%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000034s :  0.16%
    rewriter_before_opt_a                :     0.000072s :  0.33%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.28%
    jit_opt_a.loop_unroll                :     0.000036s :  0.17%
    jit_opt_a.a_1                        :     0.007773s : 35.68%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.07%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000635s :  2.91%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.09%
    jit_opt_a.cse                        :     0.000037s :  0.17%
    jit_opt_a.replace_applicator         :     0.000018s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.05%
    rewriter_after_opt_a                 :     0.000035s :  0.16%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000826s :  3.79%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.07%
    cconv                                :     0.000026s :  0.12%
    loop_unroll                          :     0.000433s :  1.99%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000013s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000044s :  0.20%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000015s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000459s :  2.11%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006160s : 28.27%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000205    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.61% :     0.000005s :      3: substitution.graph_param_transform
84.28% :     0.000173s :      4: substitution.inline
 1.40% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.07% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.83% :     0.000004s :      2: substitution.replace_old_param
 6.32% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004237     2
81.19% :     0.003440s :      1: type_inference.infer
18.81% :     0.000797s :      1: type_inference.specialize
------[replace.]   0.000072     6
67.42% :     0.000048s :      4: replace.inline
32.58% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000181     6
93.53% :     0.000169s :      4: match.inline
 6.47% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.20% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     10: predicate.addn_check_dump
 1.64% :     0.000002s :     10: predicate.addn_zero_filter
 2.71% :     0.000003s :     10: predicate.arithmetic_simplify
 1.39% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.38% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.69% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.25% :     0.000002s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.33% :     0.000007s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.01% :     0.000003s :     15: predicate.load_eliminater
 1.30% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.14% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.34% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 1.29% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     10: predicate.reduce_eliminate
 1.71% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.65% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.35% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.27% :     0.000009s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.53% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.35% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000537    12
52.25% :     0.000281s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.29% :     0.000023s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.46% :     0.000233s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.034495    76
 0.14% :     0.000047s :      1: add_recomputation
 0.19% :     0.000064s :      1: auto_monad
 0.05% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.83% :     0.000287s :      1: bootstrap
 0.08% :     0.000029s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.06% :     0.000021s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.06% :     0.000020s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
36.13% :     0.012464s :      1: jit_opt_a
 0.40% :     0.000139s :      1: jit_opt_after_cconv
 0.16% :     0.000054s :      1: jit_opt_b
 1.28% :     0.000441s :      1: loop_unroll
 2.42% :     0.000835s :      1: mutable_eliminate
23.15% :     0.007985s :     26: opt.transform.jit_opt_a
 0.14% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000025s :      4: opt.transform.jit_opt_b
 0.03% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.35% :     0.000467s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.11% :     0.000037s :      1: py_interpret_to_execute
 0.04% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000016s :      1: remove_dup_value
 0.93% :     0.000322s :      1: renormalize.infer
 0.88% :     0.000304s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000038s :      1: rewriter_after_opt_a
 0.22% :     0.000075s :      1: rewriter_before_opt_a
 0.22% :     0.000074s :      1: symbol_engine_optimizer
17.88% :     0.006168s :      1: task_emit
12.49% :     0.004309s :      1: type_inference
 0.14% :     0.000048s :      1: validate


TotalTime = 0.0132849, [30]
[bootstrap]: 0.00026356
[type_inference]: 0.00590369
[event_method]: 4.366e-05
[auto_monad]: 8.006e-05
[graph_reusing]: 5.53002e-06
[pre_auto_parallel]: 2.49001e-06
[py_interpret_to_execute]: 4.139e-05
[rewriter_before_opt_a]: 7.95e-05
[expand_dump_flag]: 3.31001e-06
[jit_opt_a]: 0.00492457, [2]
    [Cycle 1]: 0.00162281, [27]
        [switch_simplify]: 9.732e-05
        [loop_unroll]: 3.314e-05
        [a_1]: 0.00064539
        [with_stream_mark]: 1.339e-05
        [recompute_prepare]: 6.72002e-06
        [updatestate_depend_eliminate]: 2.86e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 6.04999e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 1.44998e-06
        [get_grad_eliminate_]: 5.29998e-06
        [merge_forward]: 2.58e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.222e-05
        [j_node_and_user_rematch]: 8.78001e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 8.41002e-06
        [inline_without_move]: 5.33002e-06
        [renormalize]: 0.00055689
        [add_forward_monad_depend]: 4.35e-06
        [auto_monad_grad]: 2.19001e-06
        [auto_monad_eliminator]: 1.034e-05
        [cse]: 1.481e-05
        [replace_applicator]: 1.104e-05
    [Cycle 2]: 0.000368, [27]
        [switch_simplify]: 5.14e-06
        [loop_unroll]: 4.90999e-06
        [a_1]: 0.0001486
        [with_stream_mark]: 8.67e-06
        [recompute_prepare]: 4.91002e-06
        [updatestate_depend_eliminate]: 2.29999e-06
        [updatestate_assign_eliminate]: 1.92001e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 9.40025e-07
        [specialize_transform]: 4.74002e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 4.70001e-06
        [meta_shard_fg_expand]: 1.14998e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.17999e-06
        [cell_reuse_recompute_pass]: 1.13001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.121e-05
        [j_node_and_user_rematch]: 7.9e-06
        [meta_fg_expand]: 1.90001e-06
        [replace_old_param]: 6.71e-06
        [inline_without_move]: 4.84998e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.02e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 4.97e-06
        [cse]: 9.94001e-06
        [replace_applicator]: 4.77e-06
[py_interpret_to_execute_after_opt_a]: 9.20999e-06
[rewriter_after_opt_a]: 2.871e-05
[convert_after_rewriter]: 6.52001e-06
[order_py_execute_after_rewriter]: 5.15999e-06
[mutable_eliminate]: 0.00049017
[jit_opt_b]: 4.458e-05, [1]
    [Cycle 1]: 3.888e-05, [2]
        [frontend_op_eliminate]: 1.439e-05
        [inline_after_opt_a]: 1.364e-05
[cconv]: 2.022e-05
[loop_unroll]: 0.00040826
[jit_opt_after_cconv]: 0.00012822, [1]
    [Cycle 1]: 0.00012193, [11]
        [c_1]: 1.981e-05
        [parameter_eliminate]: 2.46998e-06
        [updatestate_depend_eliminate]: 4.94998e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.384e-05
        [call_graph_tuple_transform]: 1.669e-05
        [tuple_list_get_item_eliminator]: 5.02e-06
        [none_parameter_eliminate]: 1.29e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 5.17e-06
[remove_dup_value]: 9.72001e-06
[partial_unused_args_eliminate]: 2.28998e-06
[environ_conv]: 4.53999e-06
[add_recomputation]: 3.826e-05
[cse_after_recomputation]: 1.926e-05, [1]
    [Cycle 1]: 1.423e-05, [1]
        [cse]: 8.54e-06
[auto_monad_reorder]: 1.289e-05
[get_jit_bprop_graph]: 1.69e-06
[rewriter_after_jit_bprop_graph]: 4.25e-06
[opt_after_jit_grad]: 0.00043263
[symbol_engine_optimizer]: 6.82e-05, [1]
    [Cycle 1]: 6.28e-05, [6]
        [build]: 2.65002e-06
        [elim_shapecalc]: 7.05e-06
        [elim_not_effective]: 1.219e-05
        [opt_reshape]: 5.56998e-06
        [fold_const_symbol]: 8.15e-06
        [renormalize]: 4.80009e-07
[validate]: 2.936e-05
Sums
    bootstrap                            :     0.000264s :  2.69%
    type_inference                       :     0.005904s : 60.25%
    event_method                         :     0.000044s :  0.45%
    auto_monad                           :     0.000080s :  0.82%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000041s :  0.42%
    rewriter_before_opt_a                :     0.000079s :  0.81%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000102s :  1.05%
    jit_opt_a.loop_unroll                :     0.000038s :  0.39%
    jit_opt_a.a_1                        :     0.000794s :  8.10%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000557s :  5.68%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.16%
    jit_opt_a.cse                        :     0.000025s :  0.25%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.09%
    rewriter_after_opt_a                 :     0.000029s :  0.29%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000490s :  5.00%
    jit_opt_b.frontend_op_eliminate      :     0.000014s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.14%
    cconv                                :     0.000020s :  0.21%
    loop_unroll                          :     0.000408s :  4.17%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000014s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000010s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000038s :  0.39%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000013s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.04%
    opt_after_jit_grad                   :     0.000433s :  4.42%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.30%


Time group info:
------[substitution.]   0.000241    26
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 2.06% :     0.000005s :      3: substitution.graph_param_transform
84.83% :     0.000204s :      6: substitution.inline
 1.10% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.53% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.44% :     0.000003s :      2: substitution.replace_old_param
 3.35% :     0.000008s :      1: substitution.switch_simplify
 4.44% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005853     2
83.96% :     0.004914s :      1: type_inference.infer
16.04% :     0.000939s :      1: type_inference.specialize
------[replace.]   0.000121     9
52.65% :     0.000064s :      6: replace.inline
22.21% :     0.000027s :      1: replace.switch_simplify
25.15% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000218     9
92.24% :     0.000201s :      6: match.inline
 3.38% :     0.000007s :      1: match.switch_simplify
 4.37% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   801
 1.56% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.00% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000001s :     12: predicate.addn_check_dump
 1.57% :     0.000002s :     12: predicate.addn_zero_filter
 2.15% :     0.000003s :     12: predicate.arithmetic_simplify
 1.35% :     0.000002s :     12: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     12: predicate.compare_switch_simplify
 1.41% :     0.000002s :     12: predicate.depend_value_elim
 1.64% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     12: predicate.environ_get_eliminate
 1.17% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.79% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.44% :     0.000007s :     26: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     17: predicate.load_eliminater
 1.04% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.21% :     0.000005s :     40: predicate.loop_unroll_before_grad
 2.17% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     12: predicate.merge_addn
 1.16% :     0.000001s :     12: predicate.minmaximum_grad
 1.22% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     17: predicate.partial_eliminate
 1.42% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     12: predicate.reduce_eliminate
 1.69% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     20: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.62% :     0.000002s :     12: predicate.reshape_eliminate
 1.85% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.63% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.67% :     0.000003s :     20: predicate.switch_defer_inline
 2.37% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.20% :     0.000010s :     65: predicate.switch_simplify
 1.34% :     0.000002s :     12: predicate.tile_eliminate
 1.38% :     0.000002s :     12: predicate.transpose_eliminate
 1.54% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.01% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     12: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.50% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000721    15
51.78% :     0.000374s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.83% :     0.000078s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.39% :     0.000270s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014913    72
 0.27% :     0.000041s :      1: add_recomputation
 0.56% :     0.000083s :      1: auto_monad
 0.10% :     0.000015s :      1: auto_monad_reorder
 1.86% :     0.000277s :      1: bootstrap
 0.15% :     0.000023s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.32% :     0.000047s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
33.04% :     0.004927s :      1: jit_opt_a
 0.88% :     0.000131s :      1: jit_opt_after_cconv
 0.32% :     0.000047s :      1: jit_opt_b
 2.79% :     0.000415s :      1: loop_unroll
 3.33% :     0.000497s :      1: mutable_eliminate
 6.99% :     0.001043s :     26: opt.transform.jit_opt_a
 0.29% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000022s :      4: opt.transform.jit_opt_b
 0.07% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.94% :     0.000439s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.29% :     0.000044s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000012s :      1: remove_dup_value
 2.05% :     0.000306s :      1: renormalize.infer
 1.65% :     0.000245s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000031s :      1: rewriter_after_opt_a
 0.55% :     0.000082s :      1: rewriter_before_opt_a
 0.47% :     0.000071s :      1: symbol_engine_optimizer
39.64% :     0.005912s :      1: type_inference


TotalTime = 0.0220506, [33]
[bootstrap]: 0.00038691
[type_inference]: 0.00523073
[event_method]: 1.923e-05
[auto_monad]: 0.00012153
[graph_reusing]: 5.94e-06
[pre_auto_parallel]: 2.63e-06
[py_interpret_to_execute]: 4.43e-05
[rewriter_before_opt_a]: 7.729e-05
[expand_dump_flag]: 3.54002e-06
[jit_opt_a]: 0.00677986, [2]
    [Cycle 1]: 0.00204415, [27]
        [switch_simplify]: 5.957e-05
        [loop_unroll]: 3.347e-05
        [a_1]: 0.00058311
        [with_stream_mark]: 1.391e-05
        [recompute_prepare]: 7.61999e-06
        [updatestate_depend_eliminate]: 3.13e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 1.98002e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 7.29001e-06
        [updatestate_useless_node_eliminater]: 6.83998e-06
        [accelerated_algorithm]: 6.19001e-06
        [meta_shard_fg_expand]: 1.35001e-06
        [get_grad_eliminate_]: 5.93002e-06
        [merge_forward]: 2.66e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.227e-05
        [j_node_and_user_rematch]: 1.04e-05
        [meta_fg_expand]: 1.99e-06
        [replace_old_param]: 9.85002e-06
        [inline_without_move]: 6.34001e-06
        [renormalize]: 0.00099639
        [add_forward_monad_depend]: 6.70998e-06
        [auto_monad_grad]: 2.61e-06
        [auto_monad_eliminator]: 1.709e-05
        [cse]: 3.533e-05
        [replace_applicator]: 1.685e-05
    [Cycle 2]: 0.00036548, [27]
        [switch_simplify]: 6.54999e-06
        [loop_unroll]: 5.91e-06
        [a_1]: 0.00010903
        [with_stream_mark]: 1.211e-05
        [recompute_prepare]: 5.45001e-06
        [updatestate_depend_eliminate]: 2.86e-06
        [updatestate_assign_eliminate]: 2.40002e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.22999e-06
        [specialize_transform]: 5.59e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.80002e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 4.64998e-06
        [merge_forward]: 2.43002e-06
        [cell_reuse_recompute_pass]: 2.43e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.425e-05
        [j_node_and_user_rematch]: 8.38001e-06
        [meta_fg_expand]: 1.78002e-06
        [replace_old_param]: 7.78999e-06
        [inline_without_move]: 5.15001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.85001e-06
        [auto_monad_grad]: 1.20001e-06
        [auto_monad_eliminator]: 5.62001e-06
        [cse]: 1.19e-05
        [replace_applicator]: 6.10002e-06
[py_interpret_to_execute_after_opt_a]: 1.422e-05
[rewriter_after_opt_a]: 4.066e-05
[convert_after_rewriter]: 7.8e-06
[order_py_execute_after_rewriter]: 5.65001e-06
[mutable_eliminate]: 0.00079683
[jit_opt_b]: 5.361e-05, [1]
    [Cycle 1]: 4.527e-05, [2]
        [frontend_op_eliminate]: 1.768e-05
        [inline_after_opt_a]: 1.579e-05
[cconv]: 2.885e-05
[loop_unroll]: 0.0004966
[jit_opt_after_cconv]: 0.00014894, [1]
    [Cycle 1]: 0.00014259, [11]
        [c_1]: 2.25e-05
        [parameter_eliminate]: 2.64999e-06
        [updatestate_depend_eliminate]: 6.21e-06
        [updatestate_assign_eliminate]: 2.90002e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [cse]: 1.974e-05
        [call_graph_tuple_transform]: 2.102e-05
        [tuple_list_get_item_eliminator]: 5.64998e-06
        [none_parameter_eliminate]: 1.81998e-06
        [renormalize]: 6.29982e-07
        [switch_simplify]: 5.81998e-06
[remove_dup_value]: 1.585e-05
[partial_unused_args_eliminate]: 2.21998e-06
[environ_conv]: 5.40001e-06
[add_recomputation]: 5.499e-05
[cse_after_recomputation]: 2.267e-05, [1]
    [Cycle 1]: 1.64e-05, [1]
        [cse]: 9.59e-06
[auto_monad_reorder]: 1.69e-05
[get_jit_bprop_graph]: 2.02999e-06
[rewriter_after_jit_bprop_graph]: 6.39999e-06
[opt_after_jit_grad]: 0.00048545
[symbol_engine_optimizer]: 7.723e-05, [1]
    [Cycle 1]: 6.975e-05, [6]
        [build]: 4.08999e-06
        [elim_shapecalc]: 8.3e-06
        [elim_not_effective]: 1.321e-05
        [opt_reshape]: 6.11e-06
        [fold_const_symbol]: 8.79e-06
        [renormalize]: 5.60016e-07
[validate]: 3.945e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.00680273
[execute]: 9.72999e-06
Sums
    bootstrap                            :     0.000387s :  2.28%
    type_inference                       :     0.005231s : 30.78%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000122s :  0.72%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000044s :  0.26%
    rewriter_before_opt_a                :     0.000077s :  0.45%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.39%
    jit_opt_a.loop_unroll                :     0.000039s :  0.23%
    jit_opt_a.a_1                        :     0.000692s :  4.07%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000013s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000996s :  5.86%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.13%
    jit_opt_a.cse                        :     0.000047s :  0.28%
    jit_opt_a.replace_applicator         :     0.000023s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000041s :  0.24%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000797s :  4.69%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000029s :  0.17%
    loop_unroll                          :     0.000497s :  2.92%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000055s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000485s :  2.86%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006803s : 40.03%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000207    23
 0.95% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.74% :     0.000006s :      3: substitution.graph_param_transform
85.40% :     0.000177s :      4: substitution.inline
 1.70% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.17% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.04% :     0.000004s :      2: substitution.replace_old_param
 4.32% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005153     2
82.49% :     0.004251s :      1: type_inference.infer
17.51% :     0.000902s :      1: type_inference.specialize
------[replace.]   0.000077     6
64.34% :     0.000050s :      4: replace.inline
35.66% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000182     6
95.66% :     0.000174s :      4: match.inline
 4.34% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.69% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.15% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.43% :     0.000002s :     10: predicate.addn_zero_filter
 2.66% :     0.000003s :     10: predicate.arithmetic_simplify
 1.23% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.81% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.53% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     10: predicate.environ_get_eliminate
 1.22% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.12% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.46% :     0.000007s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.98% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.15% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.23% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.54% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 1.96% :     0.000003s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.84% :     0.000002s :     10: predicate.reduce_eliminate
 1.62% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.82% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 1.19% :     0.000002s :      6: predicate.specialize_transform
 1.51% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.10% :     0.000009s :     53: predicate.switch_simplify
 1.51% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000002s :     10: predicate.transpose_eliminate
 1.56% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.03% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000696    12
48.46% :     0.000337s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.78% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.76% :     0.000326s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024070    76
 0.24% :     0.000058s :      1: add_recomputation
 0.52% :     0.000125s :      1: auto_monad
 0.08% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.70% :     0.000409s :      1: bootstrap
 0.13% :     0.000032s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
28.19% :     0.006785s :      1: jit_opt_a
 0.63% :     0.000152s :      1: jit_opt_after_cconv
 0.23% :     0.000057s :      1: jit_opt_b
 2.10% :     0.000505s :      1: loop_unroll
 3.35% :     0.000807s :      1: mutable_eliminate
 3.85% :     0.000928s :     26: opt.transform.jit_opt_a
 0.21% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.05% :     0.000494s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000047s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 2.23% :     0.000536s :      1: renormalize.infer
 1.88% :     0.000452s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000043s :      1: rewriter_after_opt_a
 0.33% :     0.000080s :      1: rewriter_before_opt_a
 0.33% :     0.000080s :      1: symbol_engine_optimizer
28.35% :     0.006824s :      1: task_emit
21.81% :     0.005249s :      1: type_inference
 0.26% :     0.000062s :      1: validate


TotalTime = 0.0186436, [33]
[bootstrap]: 0.00027846
[type_inference]: 0.00441134
[event_method]: 1.703e-05
[auto_monad]: 6.377e-05
[graph_reusing]: 5.42999e-06
[pre_auto_parallel]: 2.47001e-06
[py_interpret_to_execute]: 3.548e-05
[rewriter_before_opt_a]: 7.035e-05
[expand_dump_flag]: 3.31001e-06
[jit_opt_a]: 0.00534074, [2]
    [Cycle 1]: 0.00153669, [27]
        [switch_simplify]: 5.518e-05
        [loop_unroll]: 3.197e-05
        [a_1]: 0.00051791
        [with_stream_mark]: 1.742e-05
        [recompute_prepare]: 7.13e-06
        [updatestate_depend_eliminate]: 4.23999e-06
        [updatestate_assign_eliminate]: 3.09001e-06
        [updatestate_loads_eliminate]: 2.93e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 5.83997e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 2.21e-06
        [get_grad_eliminate_]: 5.07999e-06
        [merge_forward]: 3.78001e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.315e-05
        [j_node_and_user_rematch]: 8.63001e-06
        [meta_fg_expand]: 2.36e-06
        [replace_old_param]: 8.40999e-06
        [inline_without_move]: 5.32999e-06
        [renormalize]: 0.00059693
        [add_forward_monad_depend]: 5.97001e-06
        [auto_monad_grad]: 2.61999e-06
        [auto_monad_eliminator]: 1.287e-05
        [cse]: 2.866e-05
        [replace_applicator]: 1.25e-05
    [Cycle 2]: 0.00032271, [27]
        [switch_simplify]: 5.87001e-06
        [loop_unroll]: 4.80001e-06
        [a_1]: 8.948e-05
        [with_stream_mark]: 9.26002e-06
        [recompute_prepare]: 5.20001e-06
        [updatestate_depend_eliminate]: 2.54999e-06
        [updatestate_assign_eliminate]: 2.15002e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 5.02e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.07998e-06
        [get_grad_eliminate_]: 4.61002e-06
        [merge_forward]: 2.29999e-06
        [cell_reuse_recompute_pass]: 1.88002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.185e-05
        [j_node_and_user_rematch]: 8.02e-06
        [meta_fg_expand]: 1.77001e-06
        [replace_old_param]: 7.39002e-06
        [inline_without_move]: 4.63999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.19998e-06
        [auto_monad_grad]: 7.99977e-07
        [auto_monad_eliminator]: 5.04e-06
        [cse]: 1.044e-05
        [replace_applicator]: 5.07999e-06
[py_interpret_to_execute_after_opt_a]: 8.97e-06
[rewriter_after_opt_a]: 3.334e-05
[convert_after_rewriter]: 6.56e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00049566
[jit_opt_b]: 4.99e-05, [1]
    [Cycle 1]: 4.301e-05, [2]
        [frontend_op_eliminate]: 1.646e-05
        [inline_after_opt_a]: 1.472e-05
[cconv]: 2.318e-05
[loop_unroll]: 0.00042058
[jit_opt_after_cconv]: 0.00013887, [1]
    [Cycle 1]: 0.00013188, [11]
        [c_1]: 2.135e-05
        [parameter_eliminate]: 2.78e-06
        [updatestate_depend_eliminate]: 5.54e-06
        [updatestate_assign_eliminate]: 2.31998e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [cse]: 1.594e-05
        [call_graph_tuple_transform]: 1.787e-05
        [tuple_list_get_item_eliminator]: 5.34e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 5.66998e-06
[remove_dup_value]: 1.317e-05
[partial_unused_args_eliminate]: 2.43e-06
[environ_conv]: 5.73002e-06
[add_recomputation]: 4.879e-05
[cse_after_recomputation]: 2.096e-05, [1]
    [Cycle 1]: 1.541e-05, [1]
        [cse]: 9.68002e-06
[auto_monad_reorder]: 1.515e-05
[get_jit_bprop_graph]: 1.62999e-06
[rewriter_after_jit_bprop_graph]: 5.05001e-06
[opt_after_jit_grad]: 0.00045796
[symbol_engine_optimizer]: 7.311e-05, [1]
    [Cycle 1]: 6.719e-05, [6]
        [build]: 3.65e-06
        [elim_shapecalc]: 7.58001e-06
        [elim_not_effective]: 1.288e-05
        [opt_reshape]: 5.44998e-06
        [fold_const_symbol]: 8.70001e-06
        [renormalize]: 4.39992e-07
[validate]: 3.236e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.00635814
[execute]: 8.42e-06
Sums
    bootstrap                            :     0.000278s :  1.91%
    type_inference                       :     0.004411s : 30.29%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000064s :  0.44%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.24%
    rewriter_before_opt_a                :     0.000070s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.42%
    jit_opt_a.loop_unroll                :     0.000037s :  0.25%
    jit_opt_a.a_1                        :     0.000607s :  4.17%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000597s :  4.10%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000039s :  0.27%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000496s :  3.40%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000023s :  0.16%
    loop_unroll                          :     0.000421s :  2.89%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000049s :  0.34%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000458s :  3.14%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006358s : 43.66%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000193    23
 1.27% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.66% :     0.000005s :      3: substitution.graph_param_transform
83.33% :     0.000161s :      4: substitution.inline
 1.61% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.10% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000004s :      2: substitution.replace_old_param
 6.36% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004350     2
81.98% :     0.003566s :      1: type_inference.infer
18.02% :     0.000784s :      1: type_inference.specialize
------[replace.]   0.000066     6
66.53% :     0.000044s :      4: replace.inline
33.47% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000170     6
93.42% :     0.000158s :      4: match.inline
 6.58% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 1.86% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000001s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.22% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.39% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.44% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.86% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.82% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.45% :     0.000010s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.45% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.04% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000546    12
49.90% :     0.000272s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.77% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.33% :     0.000242s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020164    76
 0.26% :     0.000051s :      1: add_recomputation
 0.33% :     0.000067s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.44% :     0.000291s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.50% :     0.005344s :      1: jit_opt_a
 0.70% :     0.000142s :      1: jit_opt_after_cconv
 0.26% :     0.000052s :      1: jit_opt_b
 2.12% :     0.000428s :      1: loop_unroll
 2.50% :     0.000505s :      1: mutable_eliminate
 4.04% :     0.000815s :     26: opt.transform.jit_opt_a
 0.23% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.31% :     0.000466s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000038s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.52% :     0.000306s :      1: renormalize.infer
 1.41% :     0.000283s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000036s :      1: rewriter_after_opt_a
 0.36% :     0.000073s :      1: rewriter_before_opt_a
 0.37% :     0.000076s :      1: symbol_engine_optimizer
31.61% :     0.006374s :      1: task_emit
21.94% :     0.004424s :      1: type_inference
 0.26% :     0.000052s :      1: validate


TotalTime = 0.0184474, [33]
[bootstrap]: 0.00030503
[type_inference]: 0.00434652
[event_method]: 1.757e-05
[auto_monad]: 6.255e-05
[graph_reusing]: 5.52001e-06
[pre_auto_parallel]: 2.58e-06
[py_interpret_to_execute]: 3.184e-05
[rewriter_before_opt_a]: 6.983e-05
[expand_dump_flag]: 3.43e-06
[jit_opt_a]: 0.00537519, [2]
    [Cycle 1]: 0.00159563, [27]
        [switch_simplify]: 5.422e-05
        [loop_unroll]: 3.874e-05
        [a_1]: 0.00051322
        [with_stream_mark]: 2.012e-05
        [recompute_prepare]: 7.68001e-06
        [updatestate_depend_eliminate]: 3.86001e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 2.12999e-06
        [specialize_transform]: 5.68002e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.17e-06
        [meta_shard_fg_expand]: 1.86998e-06
        [get_grad_eliminate_]: 5.07999e-06
        [merge_forward]: 3.34001e-06
        [cell_reuse_recompute_pass]: 1.28002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.395e-05
        [j_node_and_user_rematch]: 8.60999e-06
        [meta_fg_expand]: 2.19999e-06
        [replace_old_param]: 8.70001e-06
        [inline_without_move]: 5.00999e-06
        [renormalize]: 0.00064772
        [add_forward_monad_depend]: 5.05999e-06
        [auto_monad_grad]: 2.27999e-06
        [auto_monad_eliminator]: 1.327e-05
        [cse]: 2.965e-05
        [replace_applicator]: 1.203e-05
    [Cycle 2]: 0.00031349, [27]
        [switch_simplify]: 5.42999e-06
        [loop_unroll]: 4.79e-06
        [a_1]: 8.812e-05
        [with_stream_mark]: 9.42999e-06
        [recompute_prepare]: 5.15001e-06
        [updatestate_depend_eliminate]: 2.54999e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 4.77e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 4.71002e-06
        [meta_shard_fg_expand]: 1.08001e-06
        [get_grad_eliminate_]: 4.3e-06
        [merge_forward]: 2.54999e-06
        [cell_reuse_recompute_pass]: 1.89e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.205e-05
        [j_node_and_user_rematch]: 7.63999e-06
        [meta_fg_expand]: 1.39e-06
        [replace_old_param]: 6.87002e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.08001e-06
        [auto_monad_grad]: 1.05001e-06
        [auto_monad_eliminator]: 5.08002e-06
        [cse]: 9.87001e-06
        [replace_applicator]: 4.92e-06
[py_interpret_to_execute_after_opt_a]: 1.027e-05
[rewriter_after_opt_a]: 3.437e-05
[convert_after_rewriter]: 7.28e-06
[order_py_execute_after_rewriter]: 4.87e-06
[mutable_eliminate]: 0.00059839
[jit_opt_b]: 4.913e-05, [1]
    [Cycle 1]: 4.181e-05, [2]
        [frontend_op_eliminate]: 1.584e-05
        [inline_after_opt_a]: 1.46e-05
[cconv]: 2.628e-05
[loop_unroll]: 0.00042009
[jit_opt_after_cconv]: 0.00013401, [1]
    [Cycle 1]: 0.00012748, [11]
        [c_1]: 1.931e-05
        [parameter_eliminate]: 2.64001e-06
        [updatestate_depend_eliminate]: 5.40999e-06
        [updatestate_assign_eliminate]: 2.23002e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.753e-05
        [call_graph_tuple_transform]: 1.77e-05
        [tuple_list_get_item_eliminator]: 5.34e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 6.60017e-07
        [switch_simplify]: 5.22e-06
[remove_dup_value]: 1.421e-05
[partial_unused_args_eliminate]: 2.10002e-06
[environ_conv]: 5.02999e-06
[add_recomputation]: 4.875e-05
[cse_after_recomputation]: 1.868e-05, [1]
    [Cycle 1]: 1.355e-05, [1]
        [cse]: 8.09002e-06
[auto_monad_reorder]: 1.546e-05
[get_jit_bprop_graph]: 1.87999e-06
[rewriter_after_jit_bprop_graph]: 5.34998e-06
[opt_after_jit_grad]: 0.00045406
[symbol_engine_optimizer]: 7.098e-05, [1]
    [Cycle 1]: 6.435e-05, [6]
        [build]: 4.08001e-06
        [elim_shapecalc]: 7.25e-06
        [elim_not_effective]: 1.232e-05
        [opt_reshape]: 5.30999e-06
        [fold_const_symbol]: 8.40001e-06
        [renormalize]: 6.20028e-07
[validate]: 3.686e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00606188
[execute]: 9.02e-06
Sums
    bootstrap                            :     0.000305s :  2.12%
    type_inference                       :     0.004347s : 30.23%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000063s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.22%
    rewriter_before_opt_a                :     0.000070s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000060s :  0.41%
    jit_opt_a.loop_unroll                :     0.000044s :  0.30%
    jit_opt_a.a_1                        :     0.000601s :  4.18%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000648s :  4.51%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000040s :  0.27%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000598s :  4.16%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000026s :  0.18%
    loop_unroll                          :     0.000420s :  2.92%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.34%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000454s :  3.16%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006062s : 42.16%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000199    23
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.61% :     0.000005s :      3: substitution.graph_param_transform
83.55% :     0.000166s :      4: substitution.inline
 1.51% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.38% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.19% :     0.000004s :      2: substitution.replace_old_param
 6.19% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004276     2
81.84% :     0.003500s :      1: type_inference.infer
18.16% :     0.000776s :      1: type_inference.specialize
------[replace.]   0.000065     6
67.93% :     0.000044s :      4: replace.inline
32.07% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000175     6
93.58% :     0.000164s :      4: match.inline
 6.42% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.19% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.10% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000002s :     10: predicate.addn_zero_filter
 1.77% :     0.000002s :     10: predicate.arithmetic_simplify
 1.16% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.15% :     0.000001s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.40% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.19% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.04% :     0.000001s :      6: predicate.less_batch_normalization
 1.79% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
11.28% :     0.000014s :     34: predicate.loop_unroll_before_grad
 1.72% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 1.85% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.32% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.18% :     0.000001s :     10: predicate.reshape_eliminate
 1.43% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.10% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 1.92% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.49% :     0.000009s :     53: predicate.switch_simplify
 1.08% :     0.000001s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.23% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.81% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.32% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.34% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.52% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.40% :     0.000002s :     10: predicate.value_based_eliminate
 0.44% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000523    12
49.28% :     0.000258s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.05% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.66% :     0.000233s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020005    76
 0.26% :     0.000052s :      1: add_recomputation
 0.33% :     0.000066s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.61% :     0.000323s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.88% :     0.005378s :      1: jit_opt_a
 0.68% :     0.000137s :      1: jit_opt_after_cconv
 0.26% :     0.000052s :      1: jit_opt_b
 2.14% :     0.000429s :      1: loop_unroll
 3.04% :     0.000608s :      1: mutable_eliminate
 4.07% :     0.000813s :     26: opt.transform.jit_opt_a
 0.22% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.31% :     0.000462s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000035s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.81% :     0.000363s :      1: renormalize.infer
 1.39% :     0.000278s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000037s :      1: rewriter_after_opt_a
 0.36% :     0.000073s :      1: rewriter_before_opt_a
 0.37% :     0.000073s :      1: symbol_engine_optimizer
30.36% :     0.006074s :      1: task_emit
21.80% :     0.004360s :      1: type_inference
 0.29% :     0.000057s :      1: validate


TotalTime = 0.0198278, [33]
[bootstrap]: 0.00027221
[type_inference]: 0.00435548
[event_method]: 1.688e-05
[auto_monad]: 6.378e-05
[graph_reusing]: 5.84e-06
[pre_auto_parallel]: 2.43e-06
[py_interpret_to_execute]: 3.287e-05
[rewriter_before_opt_a]: 7.261e-05
[expand_dump_flag]: 2.96999e-06
[jit_opt_a]: 0.00515318, [2]
    [Cycle 1]: 0.0016258, [27]
        [switch_simplify]: 5.527e-05
        [loop_unroll]: 3.017e-05
        [a_1]: 0.00053431
        [with_stream_mark]: 1.753e-05
        [recompute_prepare]: 7.23e-06
        [updatestate_depend_eliminate]: 3.51999e-06
        [updatestate_assign_eliminate]: 3.02002e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 2.36998e-06
        [specialize_transform]: 6.28e-06
        [updatestate_useless_node_eliminater]: 5.28002e-06
        [accelerated_algorithm]: 5.51998e-06
        [meta_shard_fg_expand]: 2.14999e-06
        [get_grad_eliminate_]: 5.00001e-06
        [merge_forward]: 3.88999e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.441e-05
        [j_node_and_user_rematch]: 9.64e-06
        [meta_fg_expand]: 2.32999e-06
        [replace_old_param]: 8.87e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.00066163
        [add_forward_monad_depend]: 5.30999e-06
        [auto_monad_grad]: 2.43e-06
        [auto_monad_eliminator]: 1.376e-05
        [cse]: 2.9e-05
        [replace_applicator]: 1.32e-05
    [Cycle 2]: 0.00033159, [27]
        [switch_simplify]: 6.14001e-06
        [loop_unroll]: 4.92999e-06
        [a_1]: 9.193e-05
        [with_stream_mark]: 9.99999e-06
        [recompute_prepare]: 5.05001e-06
        [updatestate_depend_eliminate]: 2.66999e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 4.97999e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 5.27001e-06
        [meta_shard_fg_expand]: 1.05001e-06
        [get_grad_eliminate_]: 4.48999e-06
        [merge_forward]: 2.32999e-06
        [cell_reuse_recompute_pass]: 1.94e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.322e-05
        [j_node_and_user_rematch]: 8e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 7.92e-06
        [inline_without_move]: 4.60999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 8.89995e-07
        [auto_monad_eliminator]: 5.69e-06
        [cse]: 1.048e-05
        [replace_applicator]: 5.69e-06
[py_interpret_to_execute_after_opt_a]: 1.087e-05
[rewriter_after_opt_a]: 3.415e-05
[convert_after_rewriter]: 6.83e-06
[order_py_execute_after_rewriter]: 4.99e-06
[mutable_eliminate]: 0.00070445
[jit_opt_b]: 5.279e-05, [1]
    [Cycle 1]: 4.522e-05, [2]
        [frontend_op_eliminate]: 1.726e-05
        [inline_after_opt_a]: 1.585e-05
[cconv]: 2.894e-05
[loop_unroll]: 0.00047512
[jit_opt_after_cconv]: 0.00016024, [1]
    [Cycle 1]: 0.00015312, [11]
        [c_1]: 2.135e-05
        [parameter_eliminate]: 3.4e-06
        [updatestate_depend_eliminate]: 6.31e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.61999e-06
        [cse]: 2.682e-05
        [call_graph_tuple_transform]: 2.211e-05
        [tuple_list_get_item_eliminator]: 6.27001e-06
        [none_parameter_eliminate]: 1.69e-06
        [renormalize]: 8.30012e-07
        [switch_simplify]: 5.86e-06
[remove_dup_value]: 1.454e-05
[partial_unused_args_eliminate]: 2.68e-06
[environ_conv]: 5.37999e-06
[add_recomputation]: 5.298e-05
[cse_after_recomputation]: 2.281e-05, [1]
    [Cycle 1]: 1.709e-05, [1]
        [cse]: 1.04e-05
[auto_monad_reorder]: 1.46e-05
[get_jit_bprop_graph]: 2.11003e-06
[rewriter_after_jit_bprop_graph]: 8.43001e-06
[opt_after_jit_grad]: 0.00053624
[symbol_engine_optimizer]: 8.052e-05, [1]
    [Cycle 1]: 7.257e-05, [6]
        [build]: 5.77001e-06
        [elim_shapecalc]: 7.89002e-06
        [elim_not_effective]: 1.436e-05
        [opt_reshape]: 5.96998e-06
        [fold_const_symbol]: 9.62001e-06
        [renormalize]: 7.2e-07
[validate]: 4.371e-05
[backend_pass]: 1.02e-06
[task_emit]: 0.00735189
[execute]: 9.19998e-06
Sums
    bootstrap                            :     0.000272s :  1.70%
    type_inference                       :     0.004355s : 27.25%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000064s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.21%
    rewriter_before_opt_a                :     0.000073s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.38%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000626s :  3.92%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000662s :  4.14%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000039s :  0.25%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000704s :  4.41%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000029s :  0.18%
    loop_unroll                          :     0.000475s :  2.97%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000027s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000053s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000536s :  3.36%
    symbol_engine_optimizer.build        :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007352s : 46.00%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000206    23
 0.90% :     0.000002s :      2: substitution.elim_not_effective
 0.78% :     0.000002s :      2: substitution.fold_const_symbol
 2.62% :     0.000005s :      3: substitution.graph_param_transform
83.78% :     0.000173s :      4: substitution.inline
 1.77% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.21% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.16% :     0.000004s :      2: substitution.replace_old_param
 5.78% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004294     2
81.98% :     0.003520s :      1: type_inference.infer
18.02% :     0.000774s :      1: type_inference.specialize
------[replace.]   0.000068     6
67.51% :     0.000046s :      4: replace.inline
32.49% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000181     6
94.06% :     0.000170s :      4: match.inline
 5.94% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.29% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.54% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000002s :     10: predicate.addn_zero_filter
 2.04% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.57% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.51% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.77% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.36% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.95% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 3.30% :     0.000004s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.85% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.22% :     0.000003s :     16: predicate.switch_defer_inline
 2.01% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.88% :     0.000010s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.25% :     0.000002s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.63% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.56% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000560    12
49.39% :     0.000276s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.52% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.09% :     0.000247s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021450    76
 0.26% :     0.000056s :      1: add_recomputation
 0.31% :     0.000067s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.35% :     0.000290s :      1: bootstrap
 0.15% :     0.000032s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
24.04% :     0.005157s :      1: jit_opt_a
 0.76% :     0.000163s :      1: jit_opt_after_cconv
 0.26% :     0.000056s :      1: jit_opt_b
 2.26% :     0.000486s :      1: loop_unroll
 3.33% :     0.000715s :      1: mutable_eliminate
 3.91% :     0.000839s :     26: opt.transform.jit_opt_a
 0.24% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.56% :     0.000549s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.64% :     0.000353s :      1: renormalize.infer
 1.41% :     0.000302s :      1: renormalize.specialize
 0.05% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000038s :      1: rewriter_after_opt_a
 0.35% :     0.000075s :      1: rewriter_before_opt_a
 0.39% :     0.000084s :      1: symbol_engine_optimizer
34.37% :     0.007372s :      1: task_emit
20.36% :     0.004368s :      1: type_inference
 0.32% :     0.000068s :      1: validate


TotalTime = 0.0182529, [33]
[bootstrap]: 0.00029614
[type_inference]: 0.00433288
[event_method]: 1.745e-05
[auto_monad]: 6.145e-05
[graph_reusing]: 5.92999e-06
[pre_auto_parallel]: 2.54999e-06
[py_interpret_to_execute]: 3.218e-05
[rewriter_before_opt_a]: 7.427e-05
[expand_dump_flag]: 3.01999e-06
[jit_opt_a]: 0.00508626, [2]
    [Cycle 1]: 0.0015199, [27]
        [switch_simplify]: 5.466e-05
        [loop_unroll]: 2.881e-05
        [a_1]: 0.00050359
        [with_stream_mark]: 1.841e-05
        [recompute_prepare]: 8e-06
        [updatestate_depend_eliminate]: 3.79002e-06
        [updatestate_assign_eliminate]: 2.91999e-06
        [updatestate_loads_eliminate]: 2.66e-06
        [parameter_eliminate]: 2.19001e-06
        [specialize_transform]: 5.63002e-06
        [updatestate_useless_node_eliminater]: 4.86002e-06
        [accelerated_algorithm]: 5.30999e-06
        [meta_shard_fg_expand]: 2.37001e-06
        [get_grad_eliminate_]: 4.99e-06
        [merge_forward]: 3.87002e-06
        [cell_reuse_recompute_pass]: 1.12999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.308e-05
        [j_node_and_user_rematch]: 8.66997e-06
        [meta_fg_expand]: 2.12999e-06
        [replace_old_param]: 8.3e-06
        [inline_without_move]: 5.37999e-06
        [renormalize]: 0.00059615
        [add_forward_monad_depend]: 5.74e-06
        [auto_monad_grad]: 2.52001e-06
        [auto_monad_eliminator]: 1.362e-05
        [cse]: 2.994e-05
        [replace_applicator]: 1.213e-05
    [Cycle 2]: 0.00033799, [27]
        [switch_simplify]: 5.20999e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 8.832e-05
        [with_stream_mark]: 8.50001e-06
        [recompute_prepare]: 5.20999e-06
        [updatestate_depend_eliminate]: 2.36e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 4.94998e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 5.06002e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.46002e-06
        [merge_forward]: 2.56e-06
        [cell_reuse_recompute_pass]: 1.87999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.204e-05
        [j_node_and_user_rematch]: 2.362e-05
        [meta_fg_expand]: 2.19001e-06
        [replace_old_param]: 7.97e-06
        [inline_without_move]: 4.64998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.57999e-06
        [auto_monad_grad]: 9.29984e-07
        [auto_monad_eliminator]: 5.34e-06
        [cse]: 1.212e-05
        [replace_applicator]: 5.02999e-06
[py_interpret_to_execute_after_opt_a]: 9.15999e-06
[rewriter_after_opt_a]: 3.472e-05
[convert_after_rewriter]: 7.09001e-06
[order_py_execute_after_rewriter]: 4.70999e-06
[mutable_eliminate]: 0.00054686
[jit_opt_b]: 4.877e-05, [1]
    [Cycle 1]: 4.2e-05, [2]
        [frontend_op_eliminate]: 1.602e-05
        [inline_after_opt_a]: 1.448e-05
[cconv]: 2.51e-05
[loop_unroll]: 0.0004262
[jit_opt_after_cconv]: 0.00013772, [1]
    [Cycle 1]: 0.00013088, [11]
        [c_1]: 1.989e-05
        [parameter_eliminate]: 2.64001e-06
        [updatestate_depend_eliminate]: 5.29e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [cse]: 1.871e-05
        [call_graph_tuple_transform]: 1.815e-05
        [tuple_list_get_item_eliminator]: 5.22e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 5.90022e-07
        [switch_simplify]: 5.27999e-06
[remove_dup_value]: 1.3e-05
[partial_unused_args_eliminate]: 2.09e-06
[environ_conv]: 4.80001e-06
[add_recomputation]: 4.935e-05
[cse_after_recomputation]: 2.078e-05, [1]
    [Cycle 1]: 1.494e-05, [1]
        [cse]: 9.46e-06
[auto_monad_reorder]: 1.513e-05
[get_jit_bprop_graph]: 2.24001e-06
[rewriter_after_jit_bprop_graph]: 5.47999e-06
[opt_after_jit_grad]: 0.00045989
[symbol_engine_optimizer]: 0.00013314, [1]
    [Cycle 1]: 0.0001266, [6]
        [build]: 3.53999e-06
        [elim_shapecalc]: 7.03998e-06
        [elim_not_effective]: 7.327e-05
        [opt_reshape]: 5.99e-06
        [fold_const_symbol]: 8.22e-06
        [renormalize]: 4.59986e-07
[validate]: 3.389e-05
[backend_pass]: 1.34e-06
[task_emit]: 0.00610587
[execute]: 7.74002e-06
Sums
    bootstrap                            :     0.000296s :  2.06%
    type_inference                       :     0.004333s : 30.15%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000061s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.22%
    rewriter_before_opt_a                :     0.000074s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000060s :  0.42%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000592s :  4.12%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000032s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000596s :  4.15%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000042s :  0.29%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000035s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000547s :  3.81%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000426s :  2.97%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000460s :  3.20%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000073s :  0.51%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006106s : 42.49%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000199    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.91% :     0.000006s :      3: substitution.graph_param_transform
76.27% :     0.000152s :      4: substitution.inline
 8.86% :     0.000018s :      4: substitution.j_node_and_user_rematch
 2.10% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.21% :     0.000004s :      2: substitution.replace_old_param
 5.97% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004264     2
78.82% :     0.003361s :      1: type_inference.infer
21.18% :     0.000903s :      1: type_inference.specialize
------[replace.]   0.000064     6
67.56% :     0.000043s :      4: replace.inline
32.44% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000160     6
93.26% :     0.000149s :      4: match.inline
 6.74% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.29% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000001s :     10: predicate.addn_zero_filter
 2.29% :     0.000003s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000001s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.68% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.45% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.80% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.95% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.07% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.75% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.34% :     0.000003s :     15: predicate.partial_eliminate
 1.23% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.69% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000001s :     10: predicate.reshape_eliminate
 1.44% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.61% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.44% :     0.000008s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.33% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.93% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000536    12
49.19% :     0.000263s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.10% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.72% :     0.000245s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019822    76
 0.26% :     0.000052s :      1: add_recomputation
 0.33% :     0.000065s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.57% :     0.000311s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.36% :     0.000071s :      1: event_method
 0.07% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.68% :     0.005089s :      1: jit_opt_a
 0.71% :     0.000140s :      1: jit_opt_after_cconv
 0.26% :     0.000051s :      1: jit_opt_b
 2.19% :     0.000434s :      1: loop_unroll
 2.80% :     0.000556s :      1: mutable_eliminate
 4.10% :     0.000812s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.46% :     0.000091s :      4: opt.transform.symbol_engine_opt
 2.36% :     0.000468s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000035s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.56% :     0.000310s :      1: renormalize.infer
 1.41% :     0.000279s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000038s :      1: rewriter_after_opt_a
 0.39% :     0.000077s :      1: rewriter_before_opt_a
 0.69% :     0.000136s :      1: symbol_engine_optimizer
30.88% :     0.006121s :      1: task_emit
21.93% :     0.004347s :      1: type_inference
 0.27% :     0.000053s :      1: validate


TotalTime = 0.0141651, [30]
[bootstrap]: 0.00027118
[type_inference]: 0.00625345
[event_method]: 1.976e-05
[auto_monad]: 8.029e-05
[graph_reusing]: 6.36e-06
[pre_auto_parallel]: 3.2e-06
[py_interpret_to_execute]: 4.183e-05
[rewriter_before_opt_a]: 0.0001201
[expand_dump_flag]: 3.25e-06
[jit_opt_a]: 0.00531622, [2]
    [Cycle 1]: 0.00188326, [27]
        [switch_simplify]: 0.00011163
        [loop_unroll]: 3.319e-05
        [a_1]: 0.00071514
        [with_stream_mark]: 1.824e-05
        [recompute_prepare]: 7.14001e-06
        [updatestate_depend_eliminate]: 2.84999e-06
        [updatestate_assign_eliminate]: 2.71e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.44e-06
        [specialize_transform]: 6.16998e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 5.44e-06
        [meta_shard_fg_expand]: 2.44999e-06
        [get_grad_eliminate_]: 5.30001e-06
        [merge_forward]: 3.40998e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.324e-05
        [j_node_and_user_rematch]: 8.98002e-06
        [meta_fg_expand]: 1.89999e-06
        [replace_old_param]: 8.54e-06
        [inline_without_move]: 5.12999e-06
        [renormalize]: 0.0006867
        [add_forward_monad_depend]: 4.80001e-06
        [auto_monad_grad]: 2.72001e-06
        [auto_monad_eliminator]: 1.201e-05
        [cse]: 2.283e-05
        [replace_applicator]: 1.195e-05
    [Cycle 2]: 0.00030843, [27]
        [switch_simplify]: 5.29998e-06
        [loop_unroll]: 5.02e-06
        [a_1]: 8.711e-05
        [with_stream_mark]: 9.26002e-06
        [recompute_prepare]: 4.94e-06
        [updatestate_depend_eliminate]: 2.45002e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 1.80001e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 5.16002e-06
        [updatestate_useless_node_eliminater]: 4.80999e-06
        [accelerated_algorithm]: 4.80001e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.65999e-06
        [merge_forward]: 2.54999e-06
        [cell_reuse_recompute_pass]: 1.45001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.266e-05
        [j_node_and_user_rematch]: 7.62002e-06
        [meta_fg_expand]: 1.40999e-06
        [replace_old_param]: 6.88998e-06
        [inline_without_move]: 4.89e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 1.17999e-06
        [auto_monad_eliminator]: 4.65999e-06
        [cse]: 9.39e-06
        [replace_applicator]: 5.07e-06
[py_interpret_to_execute_after_opt_a]: 9.85002e-06
[rewriter_after_opt_a]: 3.441e-05
[convert_after_rewriter]: 6.66e-06
[order_py_execute_after_rewriter]: 5.25001e-06
[mutable_eliminate]: 0.00055259
[jit_opt_b]: 4.742e-05, [1]
    [Cycle 1]: 4.012e-05, [2]
        [frontend_op_eliminate]: 1.501e-05
        [inline_after_opt_a]: 1.381e-05
[cconv]: 2.462e-05
[loop_unroll]: 0.00041741
[jit_opt_after_cconv]: 0.00013226, [1]
    [Cycle 1]: 0.00012579, [11]
        [c_1]: 2.042e-05
        [parameter_eliminate]: 2.51e-06
        [updatestate_depend_eliminate]: 4.27e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 1.91998e-06
        [cse]: 1.451e-05
        [call_graph_tuple_transform]: 1.793e-05
        [tuple_list_get_item_eliminator]: 5.04998e-06
        [none_parameter_eliminate]: 1.20999e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.20999e-06
[remove_dup_value]: 1.202e-05
[partial_unused_args_eliminate]: 2.31e-06
[environ_conv]: 4.4e-06
[add_recomputation]: 4.017e-05
[cse_after_recomputation]: 1.965e-05, [1]
    [Cycle 1]: 1.425e-05, [1]
        [cse]: 8.79e-06
[auto_monad_reorder]: 1.5e-05
[get_jit_bprop_graph]: 2.49999e-06
[rewriter_after_jit_bprop_graph]: 5.94999e-06
[opt_after_jit_grad]: 0.00045083
[symbol_engine_optimizer]: 6.879e-05, [1]
    [Cycle 1]: 6.301e-05, [6]
        [build]: 2.81e-06
        [elim_shapecalc]: 6.84999e-06
        [elim_not_effective]: 1.241e-05
        [opt_reshape]: 5.22999e-06
        [fold_const_symbol]: 8.45999e-06
        [renormalize]: 4.89992e-07
[validate]: 3.126e-05
Sums
    bootstrap                            :     0.000271s :  2.59%
    type_inference                       :     0.006253s : 59.76%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000080s :  0.77%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000042s :  0.40%
    rewriter_before_opt_a                :     0.000120s :  1.15%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000117s :  1.12%
    jit_opt_a.loop_unroll                :     0.000038s :  0.37%
    jit_opt_a.a_1                        :     0.000802s :  7.67%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000687s :  6.56%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.16%
    jit_opt_a.cse                        :     0.000032s :  0.31%
    jit_opt_a.replace_applicator         :     0.000017s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.09%
    rewriter_after_opt_a                 :     0.000034s :  0.33%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000553s :  5.28%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.13%
    cconv                                :     0.000025s :  0.24%
    loop_unroll                          :     0.000417s :  3.99%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000012s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000040s :  0.38%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000451s :  4.31%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.30%


Time group info:
------[substitution.]   0.000269    26
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 1.84% :     0.000005s :      3: substitution.graph_param_transform
85.25% :     0.000229s :      6: substitution.inline
 1.08% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.65% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.41% :     0.000004s :      2: substitution.replace_old_param
 3.26% :     0.000009s :      1: substitution.switch_simplify
 4.16% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006195     2
84.15% :     0.005213s :      1: type_inference.infer
15.85% :     0.000982s :      1: type_inference.specialize
------[replace.]   0.000176     9
36.31% :     0.000064s :      6: replace.inline
20.18% :     0.000036s :      1: replace.switch_simplify
43.51% :     0.000077s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000244     9
92.58% :     0.000226s :      6: match.inline
 3.25% :     0.000008s :      1: match.switch_simplify
 4.17% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   801
 1.37% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.99% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.20% :     0.000002s :     12: predicate.addn_check_dump
 1.51% :     0.000002s :     12: predicate.addn_zero_filter
 1.71% :     0.000002s :     12: predicate.arithmetic_simplify
 1.49% :     0.000002s :     12: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.17% :     0.000001s :     12: predicate.compare_switch_simplify
 1.43% :     0.000002s :     12: predicate.depend_value_elim
 1.24% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.66% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.45% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.24% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.18% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.54% :     0.000002s :     12: predicate.environ_get_eliminate
 1.24% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.28% :     0.000002s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.47% :     0.000007s :     26: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.90% :     0.000002s :     17: predicate.load_eliminater
 1.04% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.33% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.93% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000001s :     12: predicate.merge_addn
 1.20% :     0.000002s :     12: predicate.minmaximum_grad
 1.16% :     0.000001s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     17: predicate.partial_eliminate
 1.38% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     12: predicate.reduce_eliminate
 1.44% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.95% :     0.000002s :     20: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     12: predicate.reshape_eliminate
 1.24% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     12: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.60% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.83% :     0.000004s :     20: predicate.switch_defer_inline
 2.36% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.36% :     0.000010s :     65: predicate.switch_simplify
 1.44% :     0.000002s :     12: predicate.tile_eliminate
 1.27% :     0.000002s :     12: predicate.transpose_eliminate
 1.39% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.37% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.34% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000754    15
49.36% :     0.000372s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.42% :     0.000094s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.22% :     0.000288s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015968    72
 0.27% :     0.000043s :      1: add_recomputation
 0.52% :     0.000083s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 1.81% :     0.000289s :      1: bootstrap
 0.17% :     0.000027s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.32% :     0.005320s :      1: jit_opt_a
 0.84% :     0.000135s :      1: jit_opt_after_cconv
 0.31% :     0.000050s :      1: jit_opt_b
 2.66% :     0.000424s :      1: loop_unroll
 3.50% :     0.000560s :      1: mutable_eliminate
 6.70% :     0.001070s :     26: opt.transform.jit_opt_a
 0.28% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000023s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.87% :     0.000458s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000044s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000015s :      1: remove_dup_value
 2.46% :     0.000394s :      1: renormalize.infer
 1.79% :     0.000286s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000037s :      1: rewriter_after_opt_a
 0.77% :     0.000123s :      1: rewriter_before_opt_a
 0.45% :     0.000071s :      1: symbol_engine_optimizer
39.23% :     0.006264s :      1: type_inference


TotalTime = 0.0195527, [33]
[bootstrap]: 0.00035841
[type_inference]: 0.00514704
[event_method]: 1.611e-05
[auto_monad]: 6.641e-05
[graph_reusing]: 6.29999e-06
[pre_auto_parallel]: 2.67001e-06
[py_interpret_to_execute]: 3.533e-05
[rewriter_before_opt_a]: 7.489e-05
[expand_dump_flag]: 3.04999e-06
[jit_opt_a]: 0.00528923, [2]
    [Cycle 1]: 0.0015629, [27]
        [switch_simplify]: 5.716e-05
        [loop_unroll]: 2.975e-05
        [a_1]: 0.00051936
        [with_stream_mark]: 1.574e-05
        [recompute_prepare]: 7.87e-06
        [updatestate_depend_eliminate]: 3.45e-06
        [updatestate_assign_eliminate]: 3.36001e-06
        [updatestate_loads_eliminate]: 2.67001e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 6.09001e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 2.28002e-06
        [get_grad_eliminate_]: 5.44998e-06
        [merge_forward]: 3.9e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.426e-05
        [j_node_and_user_rematch]: 8.90001e-06
        [meta_fg_expand]: 2.38998e-06
        [replace_old_param]: 8.33999e-06
        [inline_without_move]: 5.27001e-06
        [renormalize]: 0.00061115
        [add_forward_monad_depend]: 5.00999e-06
        [auto_monad_grad]: 2.27001e-06
        [auto_monad_eliminator]: 1.407e-05
        [cse]: 3.113e-05
        [replace_applicator]: 1.187e-05
    [Cycle 2]: 0.00032585, [27]
        [switch_simplify]: 5.71003e-06
        [loop_unroll]: 4.82998e-06
        [a_1]: 9.099e-05
        [with_stream_mark]: 9.72001e-06
        [recompute_prepare]: 4.92e-06
        [updatestate_depend_eliminate]: 2.91999e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.06002e-06
        [specialize_transform]: 5.00001e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 1.32999e-06
        [get_grad_eliminate_]: 4.62e-06
        [merge_forward]: 2.50002e-06
        [cell_reuse_recompute_pass]: 2.07999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.264e-05
        [j_node_and_user_rematch]: 7.82e-06
        [meta_fg_expand]: 1.59998e-06
        [replace_old_param]: 1.348e-05
        [inline_without_move]: 4.68999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.57999e-06
        [auto_monad_grad]: 6.40022e-07
        [auto_monad_eliminator]: 5.00999e-06
        [cse]: 1.055e-05
        [replace_applicator]: 4.94e-06
[py_interpret_to_execute_after_opt_a]: 1.087e-05
[rewriter_after_opt_a]: 3.391e-05
[convert_after_rewriter]: 6.84999e-06
[order_py_execute_after_rewriter]: 5.17999e-06
[mutable_eliminate]: 0.00057545
[jit_opt_b]: 5.045e-05, [1]
    [Cycle 1]: 4.231e-05, [2]
        [frontend_op_eliminate]: 1.604e-05
        [inline_after_opt_a]: 1.463e-05
[cconv]: 2.346e-05
[loop_unroll]: 0.00048808
[jit_opt_after_cconv]: 0.00014029, [1]
    [Cycle 1]: 0.00013366, [11]
        [c_1]: 2.108e-05
        [parameter_eliminate]: 2.39001e-06
        [updatestate_depend_eliminate]: 5.72999e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [cse]: 1.936e-05
        [call_graph_tuple_transform]: 1.772e-05
        [tuple_list_get_item_eliminator]: 5.24998e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 9.79984e-07
        [switch_simplify]: 5.29998e-06
[remove_dup_value]: 1.394e-05
[partial_unused_args_eliminate]: 2.19999e-06
[environ_conv]: 5.09e-06
[add_recomputation]: 4.979e-05
[cse_after_recomputation]: 2.096e-05, [1]
    [Cycle 1]: 1.504e-05, [1]
        [cse]: 9.14e-06
[auto_monad_reorder]: 1.502e-05
[get_jit_bprop_graph]: 1.74998e-06
[rewriter_after_jit_bprop_graph]: 4.95999e-06
[opt_after_jit_grad]: 0.00048128
[symbol_engine_optimizer]: 7.719e-05, [1]
    [Cycle 1]: 7.045e-05, [6]
        [build]: 5.09998e-06
        [elim_shapecalc]: 7.9e-06
        [elim_not_effective]: 1.329e-05
        [opt_reshape]: 6.11e-06
        [fold_const_symbol]: 8.80999e-06
        [renormalize]: 4.19997e-07
[validate]: 3.632e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.00631097
[execute]: 7.31001e-06
Sums
    bootstrap                            :     0.000358s :  2.31%
    type_inference                       :     0.005147s : 33.10%
    event_method                         :     0.000016s :  0.10%
    auto_monad                           :     0.000066s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.23%
    rewriter_before_opt_a                :     0.000075s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000610s :  3.93%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000022s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000611s :  3.93%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000042s :  0.27%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000575s :  3.70%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000023s :  0.15%
    loop_unroll                          :     0.000488s :  3.14%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000481s :  3.10%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006311s : 40.59%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000195    23
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.50% :     0.000005s :      3: substitution.graph_param_transform
80.38% :     0.000157s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.28% :     0.000004s :      4: substitution.remove_not_recompute_node
 5.09% :     0.000010s :      2: substitution.replace_old_param
 6.41% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005075     2
83.83% :     0.004255s :      1: type_inference.infer
16.17% :     0.000821s :      1: type_inference.specialize
------[replace.]   0.000067     6
66.63% :     0.000045s :      4: replace.inline
33.37% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000165     6
93.18% :     0.000154s :      4: match.inline
 6.82% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.21% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.68% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.71% :     0.000002s :     10: predicate.addn_zero_filter
 2.24% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.67% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.47% :     0.000002s :      6: predicate.less_batch_normalization
 2.08% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 2.01% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.42% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     10: predicate.reduce_eliminate
 1.35% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.32% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.14% :     0.000010s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.36% :     0.000002s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.97% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.69% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.89% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000579    12
52.52% :     0.000304s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.71% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.77% :     0.000248s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021093    76
 0.25% :     0.000053s :      1: add_recomputation
 0.33% :     0.000070s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.80% :     0.000379s :      1: bootstrap
 0.12% :     0.000026s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.09% :     0.000020s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.09% :     0.005293s :      1: jit_opt_a
 0.68% :     0.000143s :      1: jit_opt_after_cconv
 0.25% :     0.000053s :      1: jit_opt_b
 2.36% :     0.000498s :      1: loop_unroll
 2.76% :     0.000582s :      1: mutable_eliminate
 3.92% :     0.000827s :     26: opt.transform.jit_opt_a
 0.22% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.32% :     0.000490s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000038s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.42% :     0.000299s :      1: renormalize.infer
 1.44% :     0.000305s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000037s :      1: rewriter_after_opt_a
 0.37% :     0.000077s :      1: rewriter_before_opt_a
 0.38% :     0.000080s :      1: symbol_engine_optimizer
29.96% :     0.006320s :      1: task_emit
24.44% :     0.005156s :      1: type_inference
 0.26% :     0.000056s :      1: validate


TotalTime = 0.0176969, [33]
[bootstrap]: 0.00028301
[type_inference]: 0.00435944
[event_method]: 1.597e-05
[auto_monad]: 6.31e-05
[graph_reusing]: 5.15999e-06
[pre_auto_parallel]: 2.56e-06
[py_interpret_to_execute]: 3.052e-05
[rewriter_before_opt_a]: 6.856e-05
[expand_dump_flag]: 3.07002e-06
[jit_opt_a]: 0.00506131, [2]
    [Cycle 1]: 0.00154555, [27]
        [switch_simplify]: 5.152e-05
        [loop_unroll]: 2.853e-05
        [a_1]: 0.00052588
        [with_stream_mark]: 1.841e-05
        [recompute_prepare]: 7.73001e-06
        [updatestate_depend_eliminate]: 3.21001e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.78e-06
        [parameter_eliminate]: 1.90001e-06
        [specialize_transform]: 6.02999e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.23002e-06
        [meta_shard_fg_expand]: 1.98002e-06
        [get_grad_eliminate_]: 5.17e-06
        [merge_forward]: 3.79002e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.394e-05
        [j_node_and_user_rematch]: 9.10001e-06
        [meta_fg_expand]: 2.04999e-06
        [replace_old_param]: 2.389e-05
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.0005948
        [add_forward_monad_depend]: 5.51e-06
        [auto_monad_grad]: 2.51e-06
        [auto_monad_eliminator]: 1.288e-05
        [cse]: 2.827e-05
        [replace_applicator]: 1.161e-05
    [Cycle 2]: 0.00031517, [27]
        [switch_simplify]: 5.47001e-06
        [loop_unroll]: 5.00999e-06
        [a_1]: 8.989e-05
        [with_stream_mark]: 9.07001e-06
        [recompute_prepare]: 5.17e-06
        [updatestate_depend_eliminate]: 2.49999e-06
        [updatestate_assign_eliminate]: 2.13998e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 4.90001e-06
        [updatestate_useless_node_eliminater]: 4.83001e-06
        [accelerated_algorithm]: 4.70001e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.45e-06
        [merge_forward]: 2.42001e-06
        [cell_reuse_recompute_pass]: 2.05002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.152e-05
        [j_node_and_user_rematch]: 7.8e-06
        [meta_fg_expand]: 1.55999e-06
        [replace_old_param]: 7.01999e-06
        [inline_without_move]: 4.59002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 4.77998e-06
        [cse]: 9.47999e-06
        [replace_applicator]: 5.28002e-06
[py_interpret_to_execute_after_opt_a]: 9.95002e-06
[rewriter_after_opt_a]: 3.253e-05
[convert_after_rewriter]: 7.12997e-06
[order_py_execute_after_rewriter]: 5.44e-06
[mutable_eliminate]: 0.00053308
[jit_opt_b]: 4.756e-05, [1]
    [Cycle 1]: 4.077e-05, [2]
        [frontend_op_eliminate]: 1.543e-05
        [inline_after_opt_a]: 1.434e-05
[cconv]: 2.517e-05
[loop_unroll]: 0.00041994
[jit_opt_after_cconv]: 0.00013236, [1]
    [Cycle 1]: 0.00012615, [11]
        [c_1]: 1.965e-05
        [parameter_eliminate]: 2.63e-06
        [updatestate_depend_eliminate]: 5.09e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [cse]: 1.602e-05
        [call_graph_tuple_transform]: 1.751e-05
        [tuple_list_get_item_eliminator]: 5.17999e-06
        [none_parameter_eliminate]: 1.47999e-06
        [renormalize]: 3.00002e-07
        [switch_simplify]: 5.22999e-06
[remove_dup_value]: 1.354e-05
[partial_unused_args_eliminate]: 2.04999e-06
[environ_conv]: 5.00999e-06
[add_recomputation]: 4.317e-05
[cse_after_recomputation]: 1.947e-05, [1]
    [Cycle 1]: 1.421e-05, [1]
        [cse]: 8.47998e-06
[auto_monad_reorder]: 1.534e-05
[get_jit_bprop_graph]: 1.76e-06
[rewriter_after_jit_bprop_graph]: 5.29998e-06
[opt_after_jit_grad]: 0.00045163
[symbol_engine_optimizer]: 7.054e-05, [1]
    [Cycle 1]: 6.415e-05, [6]
        [build]: 3.46001e-06
        [elim_shapecalc]: 7.23e-06
        [elim_not_effective]: 1.225e-05
        [opt_reshape]: 5.74e-06
        [fold_const_symbol]: 8.28999e-06
        [renormalize]: 5.59987e-07
[validate]: 3.265e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00574028
[execute]: 6.49999e-06
Sums
    bootstrap                            :     0.000283s :  2.03%
    type_inference                       :     0.004359s : 31.32%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000063s :  0.45%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.22%
    rewriter_before_opt_a                :     0.000069s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000616s :  4.42%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000031s :  0.22%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000595s :  4.27%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000038s :  0.27%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000533s :  3.83%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.18%
    loop_unroll                          :     0.000420s :  3.02%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000043s :  0.31%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000452s :  3.24%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000033s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005740s : 41.24%
    execute                              :     0.000006s :  0.05%


Time group info:
------[substitution.]   0.000192    23
 1.12% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.81% :     0.000005s :      3: substitution.graph_param_transform
82.66% :     0.000158s :      4: substitution.inline
 1.97% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.24% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.12% :     0.000004s :      2: substitution.replace_old_param
 6.43% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004236     2
82.17% :     0.003481s :      1: type_inference.infer
17.83% :     0.000755s :      1: type_inference.specialize
------[replace.]   0.000066     6
67.56% :     0.000045s :      4: replace.inline
32.44% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000167     6
93.27% :     0.000156s :      4: match.inline
 6.73% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.28% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     10: predicate.addn_check_dump
 1.56% :     0.000002s :     10: predicate.addn_zero_filter
 2.57% :     0.000003s :     10: predicate.arithmetic_simplify
 1.29% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.36% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.63% :     0.000002s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.13% :     0.000006s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      6: predicate.less_batch_normalization
 1.62% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.20% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.79% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.35% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000000s :      3: predicate.opt_reshape
 2.68% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.90% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.67% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.10% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.72% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.79% :     0.000009s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.27% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.56% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.32% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000538    12
49.03% :     0.000264s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.62% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.35% :     0.000238s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019222    76
 0.24% :     0.000046s :      1: add_recomputation
 0.35% :     0.000067s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.54% :     0.000295s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.35% :     0.005064s :      1: jit_opt_a
 0.70% :     0.000135s :      1: jit_opt_after_cconv
 0.26% :     0.000050s :      1: jit_opt_b
 2.22% :     0.000427s :      1: loop_unroll
 2.81% :     0.000541s :      1: mutable_eliminate
 4.33% :     0.000833s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.39% :     0.000459s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.58% :     0.000304s :      1: renormalize.infer
 1.47% :     0.000284s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000035s :      1: rewriter_after_opt_a
 0.37% :     0.000071s :      1: rewriter_before_opt_a
 0.38% :     0.000073s :      1: symbol_engine_optimizer
29.94% :     0.005755s :      1: task_emit
22.74% :     0.004371s :      1: type_inference
 0.27% :     0.000051s :      1: validate


TotalTime = 0.0200171, [33]
[bootstrap]: 0.00029619
[type_inference]: 0.00487701
[event_method]: 1.757e-05
[auto_monad]: 8.782e-05
[graph_reusing]: 7.15998e-06
[pre_auto_parallel]: 3.13e-06
[py_interpret_to_execute]: 3.659e-05
[rewriter_before_opt_a]: 7.453e-05
[expand_dump_flag]: 3.08e-06
[jit_opt_a]: 0.00559761, [2]
    [Cycle 1]: 0.0017419, [27]
        [switch_simplify]: 5.211e-05
        [loop_unroll]: 2.99e-05
        [a_1]: 0.00053386
        [with_stream_mark]: 1.71e-05
        [recompute_prepare]: 8.15e-06
        [updatestate_depend_eliminate]: 3.91001e-06
        [updatestate_assign_eliminate]: 3.23998e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 6.20002e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 6.29999e-06
        [meta_shard_fg_expand]: 2.09e-06
        [get_grad_eliminate_]: 5.66e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 1.69998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.496e-05
        [j_node_and_user_rematch]: 9.91e-06
        [meta_fg_expand]: 2.59999e-06
        [replace_old_param]: 9.00999e-06
        [inline_without_move]: 1.425e-05
        [renormalize]: 0.0007008
        [add_forward_monad_depend]: 5.39e-06
        [auto_monad_grad]: 2.52001e-06
        [auto_monad_eliminator]: 1.496e-05
        [cse]: 3.089e-05
        [replace_applicator]: 1.245e-05
    [Cycle 2]: 0.00035418, [27]
        [switch_simplify]: 5.87999e-06
        [loop_unroll]: 4.75001e-06
        [a_1]: 9.234e-05
        [with_stream_mark]: 9.47999e-06
        [recompute_prepare]: 5.15001e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 1.98997e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 5.05999e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 5.64e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 4.56002e-06
        [merge_forward]: 2.93e-06
        [cell_reuse_recompute_pass]: 2.39001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.208e-05
        [j_node_and_user_rematch]: 8.01001e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 7.4e-06
        [inline_without_move]: 4.82e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 5.81e-06
        [cse]: 1.218e-05
        [replace_applicator]: 5.61e-06
[py_interpret_to_execute_after_opt_a]: 1.095e-05
[rewriter_after_opt_a]: 3.558e-05
[convert_after_rewriter]: 6.62002e-06
[order_py_execute_after_rewriter]: 5.13002e-06
[mutable_eliminate]: 0.00067879
[jit_opt_b]: 5.043e-05, [1]
    [Cycle 1]: 4.34e-05, [2]
        [frontend_op_eliminate]: 1.716e-05
        [inline_after_opt_a]: 1.434e-05
[cconv]: 2.55e-05
[loop_unroll]: 0.00044931
[jit_opt_after_cconv]: 0.00014588, [1]
    [Cycle 1]: 0.00013866, [11]
        [c_1]: 2.638e-05
        [parameter_eliminate]: 2.66e-06
        [updatestate_depend_eliminate]: 5.69e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.01003e-06
        [cse]: 1.783e-05
        [call_graph_tuple_transform]: 1.8e-05
        [tuple_list_get_item_eliminator]: 5.22e-06
        [none_parameter_eliminate]: 1.55001e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 5.22999e-06
[remove_dup_value]: 1.421e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 5.24e-06
[add_recomputation]: 4.94e-05
[cse_after_recomputation]: 2.072e-05, [1]
    [Cycle 1]: 1.547e-05, [1]
        [cse]: 9.66003e-06
[auto_monad_reorder]: 1.493e-05
[get_jit_bprop_graph]: 2.29999e-06
[rewriter_after_jit_bprop_graph]: 5.20999e-06
[opt_after_jit_grad]: 0.00048104
[symbol_engine_optimizer]: 7.465e-05, [1]
    [Cycle 1]: 6.81e-05, [6]
        [build]: 3.86999e-06
        [elim_shapecalc]: 7.67002e-06
        [elim_not_effective]: 1.367e-05
        [opt_reshape]: 5.61e-06
        [fold_const_symbol]: 8.42998e-06
        [renormalize]: 7.00005e-07
[validate]: 3.399e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.00666649
[execute]: 8.40999e-06
Sums
    bootstrap                            :     0.000296s :  1.88%
    type_inference                       :     0.004877s : 30.90%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000088s :  0.56%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.23%
    rewriter_before_opt_a                :     0.000075s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.37%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000626s :  3.97%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000019s :  0.12%
    jit_opt_a.renormalize                :     0.000701s :  4.44%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.13%
    jit_opt_a.cse                        :     0.000043s :  0.27%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000679s :  4.30%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000025s :  0.16%
    loop_unroll                          :     0.000449s :  2.85%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.17%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000481s :  3.05%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006666s : 42.24%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000202    23
 1.26% :     0.000003s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.51% :     0.000005s :      3: substitution.graph_param_transform
83.25% :     0.000168s :      4: substitution.inline
 1.46% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.19% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000004s :      2: substitution.replace_old_param
 6.60% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004808     2
82.73% :     0.003977s :      1: type_inference.infer
17.27% :     0.000831s :      1: type_inference.specialize
------[replace.]   0.000069     6
65.84% :     0.000046s :      4: replace.inline
34.16% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000178     6
93.21% :     0.000166s :      4: match.inline
 6.79% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 1.99% :     0.000002s :     10: predicate.arithmetic_simplify
 1.36% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.47% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.57% :     0.000001s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.23% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.60% :     0.000002s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.81% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.69% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.28% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.80% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.53% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.56% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.12% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.86% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000610    12
51.63% :     0.000315s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.49% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.88% :     0.000268s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021684    76
 0.24% :     0.000052s :      1: add_recomputation
 0.42% :     0.000092s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.57% :     0.000341s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.83% :     0.005600s :      1: jit_opt_a
 0.69% :     0.000149s :      1: jit_opt_after_cconv
 0.24% :     0.000053s :      1: jit_opt_b
 2.11% :     0.000458s :      1: loop_unroll
 3.17% :     0.000688s :      1: mutable_eliminate
 3.90% :     0.000846s :     26: opt.transform.jit_opt_a
 0.24% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.10% :     0.000022s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.26% :     0.000490s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000039s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.78% :     0.000386s :      1: renormalize.infer
 1.42% :     0.000307s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000039s :      1: rewriter_after_opt_a
 0.36% :     0.000077s :      1: rewriter_before_opt_a
 0.36% :     0.000078s :      1: symbol_engine_optimizer
30.80% :     0.006679s :      1: task_emit
22.56% :     0.004892s :      1: type_inference
 0.24% :     0.000053s :      1: validate


TotalTime = 0.0138724, [30]
[bootstrap]: 0.00027707
[type_inference]: 0.00604815
[event_method]: 1.945e-05
[auto_monad]: 8.004e-05
[graph_reusing]: 5.98002e-06
[pre_auto_parallel]: 2.53998e-06
[py_interpret_to_execute]: 4.114e-05
[rewriter_before_opt_a]: 7.826e-05
[expand_dump_flag]: 3.51001e-06
[jit_opt_a]: 0.00530941, [2]
    [Cycle 1]: 0.00181445, [27]
        [switch_simplify]: 0.00010882
        [loop_unroll]: 3.286e-05
        [a_1]: 0.00065421
        [with_stream_mark]: 1.659e-05
        [recompute_prepare]: 7.9e-06
        [updatestate_depend_eliminate]: 3.23e-06
        [updatestate_assign_eliminate]: 2.56998e-06
        [updatestate_loads_eliminate]: 2.45997e-06
        [parameter_eliminate]: 1.37999e-06
        [specialize_transform]: 5.37001e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 2.02999e-06
        [get_grad_eliminate_]: 5.31998e-06
        [merge_forward]: 3.08e-06
        [cell_reuse_recompute_pass]: 1.00001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.284e-05
        [j_node_and_user_rematch]: 9.54999e-06
        [meta_fg_expand]: 2.44001e-06
        [replace_old_param]: 8.97e-06
        [inline_without_move]: 5.55001e-06
        [renormalize]: 0.00070544
        [add_forward_monad_depend]: 4.95001e-06
        [auto_monad_grad]: 2.59001e-06
        [auto_monad_eliminator]: 1.175e-05
        [cse]: 1.661e-05
        [replace_applicator]: 1.118e-05
    [Cycle 2]: 0.00030386, [27]
        [switch_simplify]: 5.20999e-06
        [loop_unroll]: 4.67e-06
        [a_1]: 8.617e-05
        [with_stream_mark]: 8.84e-06
        [recompute_prepare]: 4.97e-06
        [updatestate_depend_eliminate]: 2.48998e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 4.85999e-06
        [updatestate_useless_node_eliminater]: 4.68001e-06
        [accelerated_algorithm]: 4.62e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.37e-06
        [merge_forward]: 2.17999e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.194e-05
        [j_node_and_user_rematch]: 7.33e-06
        [meta_fg_expand]: 1.43002e-06
        [replace_old_param]: 7.01001e-06
        [inline_without_move]: 4.60001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 4.47e-06
        [cse]: 9.59999e-06
        [replace_applicator]: 4.80999e-06
[py_interpret_to_execute_after_opt_a]: 8.37e-06
[rewriter_after_opt_a]: 3.414e-05
[convert_after_rewriter]: 6.54999e-06
[order_py_execute_after_rewriter]: 5.07e-06
[mutable_eliminate]: 0.00053581
[jit_opt_b]: 4.651e-05, [1]
    [Cycle 1]: 4.005e-05, [2]
        [frontend_op_eliminate]: 1.52e-05
        [inline_after_opt_a]: 1.404e-05
[cconv]: 2.354e-05
[loop_unroll]: 0.00040394
[jit_opt_after_cconv]: 0.00014054, [1]
    [Cycle 1]: 0.00013435, [11]
        [c_1]: 1.998e-05
        [parameter_eliminate]: 2.51998e-06
        [updatestate_depend_eliminate]: 5.29e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.627e-05
        [call_graph_tuple_transform]: 1.77e-05
        [tuple_list_get_item_eliminator]: 4.97999e-06
        [none_parameter_eliminate]: 1.91e-06
        [renormalize]: 6.79982e-07
        [switch_simplify]: 5.12e-06
[remove_dup_value]: 1.187e-05
[partial_unused_args_eliminate]: 2.22001e-06
[environ_conv]: 4.49998e-06
[add_recomputation]: 4.121e-05
[cse_after_recomputation]: 1.911e-05, [1]
    [Cycle 1]: 1.363e-05, [1]
        [cse]: 7.92e-06
[auto_monad_reorder]: 1.428e-05
[get_jit_bprop_graph]: 1.94999e-06
[rewriter_after_jit_bprop_graph]: 5.77001e-06
[opt_after_jit_grad]: 0.00044378
[symbol_engine_optimizer]: 7.139e-05, [1]
    [Cycle 1]: 6.544e-05, [6]
        [build]: 3.2e-06
        [elim_shapecalc]: 7.63999e-06
        [elim_not_effective]: 1.264e-05
        [opt_reshape]: 5.47001e-06
        [fold_const_symbol]: 8.02003e-06
        [renormalize]: 5.69999e-07
[validate]: 3.158e-05
Sums
    bootstrap                            :     0.000277s :  2.74%
    type_inference                       :     0.006048s : 59.72%
    event_method                         :     0.000019s :  0.19%
    auto_monad                           :     0.000080s :  0.79%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000041s :  0.41%
    rewriter_before_opt_a                :     0.000078s :  0.77%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000114s :  1.13%
    jit_opt_a.loop_unroll                :     0.000038s :  0.37%
    jit_opt_a.a_1                        :     0.000740s :  7.31%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000705s :  6.97%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.16%
    jit_opt_a.cse                        :     0.000026s :  0.26%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.08%
    rewriter_after_opt_a                 :     0.000034s :  0.34%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000536s :  5.29%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.14%
    cconv                                :     0.000024s :  0.23%
    loop_unroll                          :     0.000404s :  3.99%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000012s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000041s :  0.41%
    cse_after_recomputation.cse          :     0.000008s :  0.08%
    auto_monad_reorder                   :     0.000014s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000444s :  4.38%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000032s :  0.31%


Time group info:
------[substitution.]   0.000257    26
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 1.94% :     0.000005s :      3: substitution.graph_param_transform
84.83% :     0.000218s :      6: substitution.inline
 1.20% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.59% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.45% :     0.000004s :      2: substitution.replace_old_param
 3.17% :     0.000008s :      1: substitution.switch_simplify
 4.60% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005991     2
83.94% :     0.005029s :      1: type_inference.infer
16.06% :     0.000962s :      1: type_inference.specialize
------[replace.]   0.000129     9
49.00% :     0.000063s :      6: replace.inline
28.18% :     0.000036s :      1: replace.switch_simplify
22.82% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000233     9
92.22% :     0.000215s :      6: match.inline
 3.16% :     0.000007s :      1: match.switch_simplify
 4.62% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   801
 1.39% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.10% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.20% :     0.000002s :     12: predicate.addn_check_dump
 1.40% :     0.000002s :     12: predicate.addn_zero_filter
 1.86% :     0.000002s :     12: predicate.arithmetic_simplify
 1.27% :     0.000002s :     12: predicate.cast_eliminate
 0.75% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.20% :     0.000002s :     12: predicate.compare_switch_simplify
 1.59% :     0.000002s :     12: predicate.depend_value_elim
 1.19% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.35% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     12: predicate.environ_get_eliminate
 1.25% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.29% :     0.000007s :     26: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :      6: predicate.less_batch_normalization
 1.62% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     17: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.27% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000002s :     12: predicate.merge_addn
 1.20% :     0.000002s :     12: predicate.minmaximum_grad
 1.62% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     17: predicate.partial_eliminate
 1.26% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     12: predicate.reduce_eliminate
 1.52% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     20: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     12: predicate.reshape_eliminate
 1.27% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.59% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     12: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.72% :     0.000001s :      6: predicate.special_op_eliminate
 0.85% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.68% :     0.000003s :     20: predicate.switch_defer_inline
 2.45% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.24% :     0.000010s :     65: predicate.switch_simplify
 1.27% :     0.000002s :     12: predicate.tile_eliminate
 1.28% :     0.000002s :     12: predicate.transpose_eliminate
 1.43% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.49% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.32% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.78% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     12: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000728    15
48.67% :     0.000354s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.19% :     0.000089s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.13% :     0.000285s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015638    72
 0.28% :     0.000044s :      1: add_recomputation
 0.53% :     0.000084s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.86% :     0.000291s :      1: bootstrap
 0.17% :     0.000026s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.97% :     0.005313s :      1: jit_opt_a
 0.92% :     0.000143s :      1: jit_opt_after_cconv
 0.31% :     0.000049s :      1: jit_opt_b
 2.63% :     0.000412s :      1: loop_unroll
 3.48% :     0.000544s :      1: mutable_eliminate
 6.41% :     0.001003s :     26: opt.transform.jit_opt_a
 0.28% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000023s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.88% :     0.000451s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.28% :     0.000044s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000014s :      1: remove_dup_value
 2.77% :     0.000433s :      1: renormalize.infer
 1.70% :     0.000265s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000037s :      1: rewriter_after_opt_a
 0.52% :     0.000081s :      1: rewriter_before_opt_a
 0.47% :     0.000074s :      1: symbol_engine_optimizer
38.75% :     0.006059s :      1: type_inference

.
TotalTime = 0.0195008, [33]
[bootstrap]: 0.00031098
[type_inference]: 0.00465948
[event_method]: 1.764e-05
[auto_monad]: 6.658e-05
[graph_reusing]: 6.04001e-06
[pre_auto_parallel]: 2.74001e-06
[py_interpret_to_execute]: 3.437e-05
[rewriter_before_opt_a]: 7.261e-05
[expand_dump_flag]: 3.46001e-06
[jit_opt_a]: 0.00554623, [2]
    [Cycle 1]: 0.0018078, [27]
        [switch_simplify]: 9.68e-05
        [loop_unroll]: 5.848e-05
        [a_1]: 0.00060499
        [with_stream_mark]: 2.008e-05
        [recompute_prepare]: 7.45e-06
        [updatestate_depend_eliminate]: 3.65e-06
        [updatestate_assign_eliminate]: 2.96999e-06
        [updatestate_loads_eliminate]: 2.58998e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 6.21e-06
        [updatestate_useless_node_eliminater]: 5.11002e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 2.32999e-06
        [get_grad_eliminate_]: 5.16002e-06
        [merge_forward]: 3.91001e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.451e-05
        [j_node_and_user_rematch]: 9.59e-06
        [meta_fg_expand]: 2.45002e-06
        [replace_old_param]: 9.29e-06
        [inline_without_move]: 5.24e-06
        [renormalize]: 0.00065434
        [add_forward_monad_depend]: 4.99e-06
        [auto_monad_grad]: 2.41e-06
        [auto_monad_eliminator]: 1.445e-05
        [cse]: 5.259e-05
        [replace_applicator]: 1.273e-05
    [Cycle 2]: 0.00039359, [27]
        [switch_simplify]: 6.15997e-06
        [loop_unroll]: 6.00002e-06
        [a_1]: 0.00011152
        [with_stream_mark]: 9.59999e-06
        [recompute_prepare]: 4.99e-06
        [updatestate_depend_eliminate]: 2.78e-06
        [updatestate_assign_eliminate]: 2.11998e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 5.24e-06
        [meta_shard_fg_expand]: 1.18001e-06
        [get_grad_eliminate_]: 4.60999e-06
        [merge_forward]: 2.78e-06
        [cell_reuse_recompute_pass]: 2.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.358e-05
        [j_node_and_user_rematch]: 8.07e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 7.31999e-06
        [inline_without_move]: 4.67e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 6.09001e-06
        [cse]: 1.112e-05
        [replace_applicator]: 5.25999e-06
[py_interpret_to_execute_after_opt_a]: 1.095e-05
[rewriter_after_opt_a]: 3.319e-05
[convert_after_rewriter]: 6.96999e-06
[order_py_execute_after_rewriter]: 8.00999e-06
[mutable_eliminate]: 0.00055017
[jit_opt_b]: 5.011e-05, [1]
    [Cycle 1]: 4.33e-05, [2]
        [frontend_op_eliminate]: 1.649e-05
        [inline_after_opt_a]: 1.527e-05
[cconv]: 2.498e-05
[loop_unroll]: 0.00042354
[jit_opt_after_cconv]: 0.00014043, [1]
    [Cycle 1]: 0.00013422, [11]
        [c_1]: 2.15e-05
        [parameter_eliminate]: 2.39001e-06
        [updatestate_depend_eliminate]: 5.77001e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.08002e-06
        [cse]: 1.702e-05
        [call_graph_tuple_transform]: 1.862e-05
        [tuple_list_get_item_eliminator]: 5.81998e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 7.2e-07
        [switch_simplify]: 5.42999e-06
[remove_dup_value]: 1.244e-05
[partial_unused_args_eliminate]: 2.09e-06
[environ_conv]: 5.16998e-06
[add_recomputation]: 5.029e-05
[cse_after_recomputation]: 2.027e-05, [1]
    [Cycle 1]: 1.502e-05, [1]
        [cse]: 9.24e-06
[auto_monad_reorder]: 1.623e-05
[get_jit_bprop_graph]: 1.77001e-06
[rewriter_after_jit_bprop_graph]: 5.87001e-06
[opt_after_jit_grad]: 0.000576
[symbol_engine_optimizer]: 7.385e-05, [1]
    [Cycle 1]: 6.715e-05, [6]
        [build]: 4.2e-06
        [elim_shapecalc]: 7.88999e-06
        [elim_not_effective]: 1.225e-05
        [opt_reshape]: 5.59e-06
        [fold_const_symbol]: 8.92e-06
        [renormalize]: 4.39992e-07
[validate]: 3.378e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00653687
[execute]: 7.50998e-06
Sums
    bootstrap                            :     0.000311s :  2.01%
    type_inference                       :     0.004659s : 30.09%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000067s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.22%
    rewriter_before_opt_a                :     0.000073s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000103s :  0.66%
    jit_opt_a.loop_unroll                :     0.000064s :  0.42%
    jit_opt_a.a_1                        :     0.000717s :  4.63%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000654s :  4.23%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.13%
    jit_opt_a.cse                        :     0.000064s :  0.41%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000008s :  0.05%
    mutable_eliminate                    :     0.000550s :  3.55%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000025s :  0.16%
    loop_unroll                          :     0.000424s :  2.73%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000576s :  3.72%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006537s : 42.21%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000204    23
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.69% :     0.000006s :      3: substitution.graph_param_transform
83.63% :     0.000171s :      4: substitution.inline
 1.74% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.17% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.39% :     0.000005s :      2: substitution.replace_old_param
 5.76% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004593     2
81.25% :     0.003732s :      1: type_inference.infer
18.75% :     0.000861s :      1: type_inference.specialize
------[replace.]   0.000068     6
67.00% :     0.000046s :      4: replace.inline
33.00% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000179     6
94.06% :     0.000168s :      4: match.inline
 5.94% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 1.96% :     0.000002s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.19% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.76% :     0.000007s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.88% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.74% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.19% :     0.000001s :     10: predicate.minmaximum_grad
 1.77% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.25% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.51% :     0.000003s :     16: predicate.switch_defer_inline
 2.33% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.17% :     0.000010s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.34% :     0.000002s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.87% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.48% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000684    12
47.14% :     0.000322s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.97% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.89% :     0.000328s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021262    76
 0.25% :     0.000053s :      1: add_recomputation
 0.33% :     0.000070s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.53% :     0.000325s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.10% :     0.005549s :      1: jit_opt_a
 0.67% :     0.000143s :      1: jit_opt_after_cconv
 0.25% :     0.000053s :      1: jit_opt_b
 2.03% :     0.000431s :      1: loop_unroll
 2.63% :     0.000559s :      1: mutable_eliminate
 4.70% :     0.000998s :     26: opt.transform.jit_opt_a
 0.23% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.75% :     0.000585s :      1: opt_after_jit_grad
 0.05% :     0.000010s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 1.59% :     0.000338s :      1: renormalize.infer
 1.45% :     0.000308s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000036s :      1: rewriter_after_opt_a
 0.35% :     0.000075s :      1: rewriter_before_opt_a
 0.36% :     0.000076s :      1: symbol_engine_optimizer
30.80% :     0.006549s :      1: task_emit
21.98% :     0.004673s :      1: type_inference
 0.26% :     0.000055s :      1: validate


TotalTime = 0.0184735, [33]
[bootstrap]: 0.00027232
[type_inference]: 0.00428378
[event_method]: 1.648e-05
[auto_monad]: 6.326e-05
[graph_reusing]: 5.76e-06
[pre_auto_parallel]: 2.39999e-06
[py_interpret_to_execute]: 2.83e-05
[rewriter_before_opt_a]: 6.994e-05
[expand_dump_flag]: 3.09001e-06
[jit_opt_a]: 0.0053254, [2]
    [Cycle 1]: 0.00152332, [27]
        [switch_simplify]: 4.925e-05
        [loop_unroll]: 2.888e-05
        [a_1]: 0.00048769
        [with_stream_mark]: 1.619e-05
        [recompute_prepare]: 6.57002e-06
        [updatestate_depend_eliminate]: 3.73001e-06
        [updatestate_assign_eliminate]: 3.18998e-06
        [updatestate_loads_eliminate]: 2.40002e-06
        [parameter_eliminate]: 1.85001e-06
        [specialize_transform]: 5.80002e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.33002e-06
        [meta_shard_fg_expand]: 2.41998e-06
        [get_grad_eliminate_]: 5.33002e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 1.60999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.384e-05
        [j_node_and_user_rematch]: 7.462e-05
        [meta_fg_expand]: 2.43e-06
        [replace_old_param]: 8.87e-06
        [inline_without_move]: 5.42999e-06
        [renormalize]: 0.00056651
        [add_forward_monad_depend]: 5.40999e-06
        [auto_monad_grad]: 2.23002e-06
        [auto_monad_eliminator]: 1.312e-05
        [cse]: 2.685e-05
        [replace_applicator]: 1.043e-05
    [Cycle 2]: 0.00031514, [27]
        [switch_simplify]: 5.14e-06
        [loop_unroll]: 5.00001e-06
        [a_1]: 8.802e-05
        [with_stream_mark]: 8.85999e-06
        [recompute_prepare]: 5.07e-06
        [updatestate_depend_eliminate]: 2.37999e-06
        [updatestate_assign_eliminate]: 2.05002e-06
        [updatestate_loads_eliminate]: 2.06003e-06
        [parameter_eliminate]: 1.32999e-06
        [specialize_transform]: 4.89998e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 4.87998e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.34002e-06
        [merge_forward]: 2.32999e-06
        [cell_reuse_recompute_pass]: 2.34001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.176e-05
        [j_node_and_user_rematch]: 7.34002e-06
        [meta_fg_expand]: 2.47001e-06
        [replace_old_param]: 7.48e-06
        [inline_without_move]: 4.52e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.28002e-06
        [auto_monad_grad]: 7.40023e-07
        [auto_monad_eliminator]: 5.12e-06
        [cse]: 1.15e-05
        [replace_applicator]: 4.67e-06
[py_interpret_to_execute_after_opt_a]: 8.24998e-06
[rewriter_after_opt_a]: 3.494e-05
[convert_after_rewriter]: 6.36e-06
[order_py_execute_after_rewriter]: 4.61002e-06
[mutable_eliminate]: 0.00063104
[jit_opt_b]: 6.579e-05, [1]
    [Cycle 1]: 5.816e-05, [2]
        [frontend_op_eliminate]: 3.063e-05
        [inline_after_opt_a]: 1.477e-05
[cconv]: 2.552e-05
[loop_unroll]: 0.00051336
[jit_opt_after_cconv]: 0.0001412, [1]
    [Cycle 1]: 0.00013497, [11]
        [c_1]: 2.078e-05
        [parameter_eliminate]: 3.10002e-06
        [updatestate_depend_eliminate]: 6.28e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [cse]: 1.894e-05
        [call_graph_tuple_transform]: 1.897e-05
        [tuple_list_get_item_eliminator]: 5.34e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 7.09988e-07
        [switch_simplify]: 5.22e-06
[remove_dup_value]: 1.343e-05
[partial_unused_args_eliminate]: 2.24001e-06
[environ_conv]: 5.54e-06
[add_recomputation]: 4.933e-05
[cse_after_recomputation]: 2.012e-05, [1]
    [Cycle 1]: 1.426e-05, [1]
        [cse]: 8.68001e-06
[auto_monad_reorder]: 1.526e-05
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 5.91e-06
[opt_after_jit_grad]: 0.00045279
[symbol_engine_optimizer]: 7.178e-05, [1]
    [Cycle 1]: 6.562e-05, [6]
        [build]: 4.77998e-06
        [elim_shapecalc]: 7.27002e-06
        [elim_not_effective]: 1.216e-05
        [opt_reshape]: 5.56998e-06
        [fold_const_symbol]: 8.45001e-06
        [renormalize]: 5.10016e-07
[validate]: 3.73e-05
[backend_pass]: 8.90024e-07
[task_emit]: 0.00608249
[execute]: 8.28999e-06
Sums
    bootstrap                            :     0.000272s :  1.89%
    type_inference                       :     0.004284s : 29.78%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000063s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.20%
    rewriter_before_opt_a                :     0.000070s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000576s :  4.00%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000082s :  0.57%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000567s :  3.94%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000038s :  0.27%
    jit_opt_a.replace_applicator         :     0.000015s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000035s :  0.24%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000631s :  4.39%
    jit_opt_b.frontend_op_eliminate      :     0.000031s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000026s :  0.18%
    loop_unroll                          :     0.000513s :  3.57%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000049s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000453s :  3.15%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006082s : 42.29%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000238    23
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.47% :     0.000006s :      3: substitution.graph_param_transform
59.19% :     0.000141s :      4: substitution.inline
28.44% :     0.000068s :      4: substitution.j_node_and_user_rematch
 1.84% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.73% :     0.000004s :      2: substitution.replace_old_param
 4.91% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004223     2
82.19% :     0.003471s :      1: type_inference.infer
17.81% :     0.000752s :      1: type_inference.specialize
------[replace.]   0.000063     6
67.59% :     0.000042s :      4: replace.inline
32.41% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000149     6
92.83% :     0.000139s :      4: match.inline
 7.17% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.25% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.21% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 1.87% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.48% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.22% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.56% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.96% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.89% :     0.000002s :     15: predicate.load_eliminater
 1.81% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 2.38% :     0.000003s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.64% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 1.05% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.58% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.44% :     0.000008s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.90% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.32% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.47% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000527    12
52.76% :     0.000278s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.28% :     0.000023s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.96% :     0.000226s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020013    76
 0.26% :     0.000052s :      1: add_recomputation
 0.33% :     0.000067s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.45% :     0.000290s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.63% :     0.005329s :      1: jit_opt_a
 0.72% :     0.000144s :      1: jit_opt_after_cconv
 0.34% :     0.000069s :      1: jit_opt_b
 2.61% :     0.000523s :      1: loop_unroll
 3.20% :     0.000641s :      1: mutable_eliminate
 4.19% :     0.000839s :     26: opt.transform.jit_opt_a
 0.23% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000039s :      4: opt.transform.jit_opt_b
 0.08% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.16% :     0.000033s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.30% :     0.000461s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000031s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.48% :     0.000296s :      1: renormalize.infer
 1.31% :     0.000263s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000038s :      1: rewriter_after_opt_a
 0.36% :     0.000073s :      1: rewriter_before_opt_a
 0.37% :     0.000074s :      1: symbol_engine_optimizer
30.45% :     0.006094s :      1: task_emit
21.48% :     0.004298s :      1: type_inference
 0.29% :     0.000058s :      1: validate


TotalTime = 0.0141095, [30]
[bootstrap]: 0.00026751
[type_inference]: 0.00626345
[event_method]: 2.136e-05
[auto_monad]: 8.342e-05
[graph_reusing]: 6.48e-06
[pre_auto_parallel]: 2.58e-06
[py_interpret_to_execute]: 4.522e-05
[rewriter_before_opt_a]: 9.207e-05
[expand_dump_flag]: 3.06001e-06
[jit_opt_a]: 0.0052613, [2]
    [Cycle 1]: 0.00170966, [27]
        [switch_simplify]: 0.00010935
        [loop_unroll]: 3.273e-05
        [a_1]: 0.0006602
        [with_stream_mark]: 1.681e-05
        [recompute_prepare]: 7.7e-06
        [updatestate_depend_eliminate]: 3.86001e-06
        [updatestate_assign_eliminate]: 3.45003e-06
        [updatestate_loads_eliminate]: 2.78e-06
        [parameter_eliminate]: 2.18998e-06
        [specialize_transform]: 6.43e-06
        [updatestate_useless_node_eliminater]: 5.24998e-06
        [accelerated_algorithm]: 5.58002e-06
        [meta_shard_fg_expand]: 2.34999e-06
        [get_grad_eliminate_]: 5.07e-06
        [merge_forward]: 3.25998e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.379e-05
        [j_node_and_user_rematch]: 8.77e-06
        [meta_fg_expand]: 2.61999e-06
        [replace_old_param]: 8.55001e-06
        [inline_without_move]: 5.04998e-06
        [renormalize]: 0.00057417
        [add_forward_monad_depend]: 5.25999e-06
        [auto_monad_grad]: 1.98997e-06
        [auto_monad_eliminator]: 1.316e-05
        [cse]: 2.629e-05
        [replace_applicator]: 1.087e-05
    [Cycle 2]: 0.00030983, [27]
        [switch_simplify]: 5.51998e-06
        [loop_unroll]: 4.62e-06
        [a_1]: 8.626e-05
        [with_stream_mark]: 8.53001e-06
        [recompute_prepare]: 4.71002e-06
        [updatestate_depend_eliminate]: 2.37999e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 1.22999e-06
        [specialize_transform]: 4.82e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 4.77e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.33999e-06
        [merge_forward]: 2.29999e-06
        [cell_reuse_recompute_pass]: 1.20999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.238e-05
        [j_node_and_user_rematch]: 7.38999e-06
        [meta_fg_expand]: 1.45001e-06
        [replace_old_param]: 1.301e-05
        [inline_without_move]: 4.68999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 4.92e-06
        [cse]: 1.002e-05
        [replace_applicator]: 5.02999e-06
[py_interpret_to_execute_after_opt_a]: 1.093e-05
[rewriter_after_opt_a]: 3.425e-05
[convert_after_rewriter]: 7.15003e-06
[order_py_execute_after_rewriter]: 5.32001e-06
[mutable_eliminate]: 0.00049222
[jit_opt_b]: 4.71e-05, [1]
    [Cycle 1]: 4.027e-05, [2]
        [frontend_op_eliminate]: 1.544e-05
        [inline_after_opt_a]: 1.364e-05
[cconv]: 2.417e-05
[loop_unroll]: 0.00043441
[jit_opt_after_cconv]: 0.00013566, [1]
    [Cycle 1]: 0.00012937, [11]
        [c_1]: 2.108e-05
        [parameter_eliminate]: 2.71e-06
        [updatestate_depend_eliminate]: 5.10999e-06
        [updatestate_assign_eliminate]: 2.39999e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.612e-05
        [call_graph_tuple_transform]: 1.745e-05
        [tuple_list_get_item_eliminator]: 5.27001e-06
        [none_parameter_eliminate]: 1.42999e-06
        [renormalize]: 6.60017e-07
        [switch_simplify]: 5.20999e-06
[remove_dup_value]: 1.299e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 4.61002e-06
[add_recomputation]: 6.245e-05
[cse_after_recomputation]: 2.135e-05, [1]
    [Cycle 1]: 1.526e-05, [1]
        [cse]: 8.95999e-06
[auto_monad_reorder]: 1.43e-05
[get_jit_bprop_graph]: 1.67001e-06
[rewriter_after_jit_bprop_graph]: 4.75001e-06
[opt_after_jit_grad]: 0.00045748
[symbol_engine_optimizer]: 7.63e-05, [1]
    [Cycle 1]: 6.987e-05, [6]
        [build]: 3.41001e-06
        [elim_shapecalc]: 8.12998e-06
        [elim_not_effective]: 1.31e-05
        [opt_reshape]: 6.05002e-06
        [fold_const_symbol]: 9.28002e-06
        [renormalize]: 8.10018e-07
[validate]: 3.131e-05
Sums
    bootstrap                            :     0.000268s :  2.60%
    type_inference                       :     0.006263s : 60.91%
    event_method                         :     0.000021s :  0.21%
    auto_monad                           :     0.000083s :  0.81%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000045s :  0.44%
    rewriter_before_opt_a                :     0.000092s :  0.90%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000115s :  1.12%
    jit_opt_a.loop_unroll                :     0.000037s :  0.36%
    jit_opt_a.a_1                        :     0.000746s :  7.26%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000022s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000574s :  5.58%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.18%
    jit_opt_a.cse                        :     0.000036s :  0.35%
    jit_opt_a.replace_applicator         :     0.000016s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.11%
    rewriter_after_opt_a                 :     0.000034s :  0.33%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000492s :  4.79%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.13%
    cconv                                :     0.000024s :  0.24%
    loop_unroll                          :     0.000434s :  4.22%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000062s :  0.61%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000014s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000457s :  4.45%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000031s :  0.30%


Time group info:
------[substitution.]   0.000256    26
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 2.00% :     0.000005s :      3: substitution.graph_param_transform
84.49% :     0.000217s :      6: substitution.inline
 1.04% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.80% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.36% :     0.000003s :      2: substitution.replace_old_param
 2.96% :     0.000008s :      1: substitution.switch_simplify
 5.10% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006205     2
83.63% :     0.005190s :      1: type_inference.infer
16.37% :     0.001016s :      1: type_inference.specialize
------[replace.]   0.000126     9
50.90% :     0.000064s :      6: replace.inline
25.52% :     0.000032s :      1: replace.switch_simplify
23.58% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000232     9
91.86% :     0.000213s :      6: match.inline
 2.96% :     0.000007s :      1: match.switch_simplify
 5.18% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   801
 1.48% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.10% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000001s :     12: predicate.addn_check_dump
 1.49% :     0.000002s :     12: predicate.addn_zero_filter
 1.93% :     0.000002s :     12: predicate.arithmetic_simplify
 1.24% :     0.000002s :     12: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.20% :     0.000002s :     12: predicate.compare_switch_simplify
 1.53% :     0.000002s :     12: predicate.depend_value_elim
 1.26% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     12: predicate.environ_get_eliminate
 1.32% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.79% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.62% :     0.000007s :     26: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.96% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     17: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.28% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000002s :     12: predicate.merge_addn
 1.20% :     0.000002s :     12: predicate.minmaximum_grad
 1.77% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     17: predicate.partial_eliminate
 1.26% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     12: predicate.reduce_eliminate
 1.55% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     20: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     12: predicate.reshape_eliminate
 1.30% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.60% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     12: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.84% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.84% :     0.000004s :     20: predicate.switch_defer_inline
 2.42% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.35% :     0.000011s :     65: predicate.switch_simplify
 1.39% :     0.000002s :     12: predicate.tile_eliminate
 1.39% :     0.000002s :     12: predicate.transpose_eliminate
 1.41% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.52% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000748    15
50.27% :     0.000376s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.44% :     0.000086s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.29% :     0.000286s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015759    72
 0.42% :     0.000066s :      1: add_recomputation
 0.55% :     0.000087s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.80% :     0.000284s :      1: bootstrap
 0.17% :     0.000027s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
33.40% :     0.005264s :      1: jit_opt_a
 0.88% :     0.000138s :      1: jit_opt_after_cconv
 0.32% :     0.000050s :      1: jit_opt_b
 2.92% :     0.000461s :      1: loop_unroll
 3.17% :     0.000500s :      1: mutable_eliminate
 6.43% :     0.001014s :     26: opt.transform.jit_opt_a
 0.29% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000023s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.95% :     0.000465s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000048s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000015s :      1: remove_dup_value
 1.87% :     0.000295s :      1: renormalize.infer
 1.73% :     0.000272s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000038s :      1: rewriter_after_opt_a
 0.60% :     0.000095s :      1: rewriter_before_opt_a
 0.50% :     0.000079s :      1: symbol_engine_optimizer
39.83% :     0.006277s :      1: type_inference


TotalTime = 0.0308548, [33]
[bootstrap]: 0.0003742
[type_inference]: 0.00517461
[event_method]: 1.865e-05
[auto_monad]: 6.638e-05
[graph_reusing]: 5.97999e-06
[pre_auto_parallel]: 2.89001e-06
[py_interpret_to_execute]: 4.273e-05
[rewriter_before_opt_a]: 7.711e-05
[expand_dump_flag]: 3.78999e-06
[jit_opt_a]: 0.0147248, [2]
    [Cycle 1]: 0.00194163, [27]
        [switch_simplify]: 6.997e-05
        [loop_unroll]: 3.22e-05
        [a_1]: 0.00060018
        [with_stream_mark]: 2.064e-05
        [recompute_prepare]: 8.38999e-06
        [updatestate_depend_eliminate]: 3.45e-06
        [updatestate_assign_eliminate]: 3.02002e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 6.11998e-06
        [updatestate_useless_node_eliminater]: 5.34e-06
        [accelerated_algorithm]: 5.99e-06
        [meta_shard_fg_expand]: 2.40002e-06
        [get_grad_eliminate_]: 5.52999e-06
        [merge_forward]: 4.12998e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.522e-05
        [j_node_and_user_rematch]: 9.87999e-06
        [meta_fg_expand]: 2.51e-06
        [replace_old_param]: 1.076e-05
        [inline_without_move]: 6.17001e-06
        [renormalize]: 0.00086154
        [add_forward_monad_depend]: 7.28e-06
        [auto_monad_grad]: 3.06001e-06
        [auto_monad_eliminator]: 1.61e-05
        [cse]: 3.291e-05
        [replace_applicator]: 1.473e-05
    [Cycle 2]: 0.00041813, [27]
        [switch_simplify]: 6.36e-06
        [loop_unroll]: 5.10001e-06
        [a_1]: 9.454e-05
        [with_stream_mark]: 6.78e-05
        [recompute_prepare]: 5.72001e-06
        [updatestate_depend_eliminate]: 2.51e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 5.52999e-06
        [updatestate_useless_node_eliminater]: 5.52999e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 1.23002e-06
        [get_grad_eliminate_]: 4.72998e-06
        [merge_forward]: 2.55002e-06
        [cell_reuse_recompute_pass]: 3.06999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.472e-05
        [j_node_and_user_rematch]: 8.84e-06
        [meta_fg_expand]: 1.89e-06
        [replace_old_param]: 7.83999e-06
        [inline_without_move]: 4.73001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.60999e-06
        [auto_monad_grad]: 9.09989e-07
        [auto_monad_eliminator]: 7.31001e-06
        [cse]: 1.256e-05
        [replace_applicator]: 5.78002e-06
[py_interpret_to_execute_after_opt_a]: 1.439e-05
[rewriter_after_opt_a]: 4.036e-05
[convert_after_rewriter]: 8.26002e-06
[order_py_execute_after_rewriter]: 5.67999e-06
[mutable_eliminate]: 0.00090088
[jit_opt_b]: 6.002e-05, [1]
    [Cycle 1]: 5.052e-05, [2]
        [frontend_op_eliminate]: 2.076e-05
        [inline_after_opt_a]: 1.619e-05
[cconv]: 2.977e-05
[loop_unroll]: 0.00055877
[jit_opt_after_cconv]: 0.00016219, [1]
    [Cycle 1]: 0.00015414, [11]
        [c_1]: 2.304e-05
        [parameter_eliminate]: 3.28e-06
        [updatestate_depend_eliminate]: 6.58998e-06
        [updatestate_assign_eliminate]: 2.68998e-06
        [updatestate_loads_eliminate]: 2.86e-06
        [cse]: 2.247e-05
        [call_graph_tuple_transform]: 2.336e-05
        [tuple_list_get_item_eliminator]: 6.33e-06
        [none_parameter_eliminate]: 1.89e-06
        [renormalize]: 8.29983e-07
        [switch_simplify]: 6.34999e-06
[remove_dup_value]: 1.722e-05
[partial_unused_args_eliminate]: 2.53e-06
[environ_conv]: 5.64998e-06
[add_recomputation]: 5.141e-05
[cse_after_recomputation]: 2.317e-05, [1]
    [Cycle 1]: 1.666e-05, [1]
        [cse]: 1.038e-05
[auto_monad_reorder]: 1.664e-05
[get_jit_bprop_graph]: 2.29001e-06
[rewriter_after_jit_bprop_graph]: 8.24002e-06
[opt_after_jit_grad]: 0.00054371
[symbol_engine_optimizer]: 8.079e-05, [1]
    [Cycle 1]: 7.285e-05, [6]
        [build]: 3.68999e-06
        [elim_shapecalc]: 8.65999e-06
        [elim_not_effective]: 1.313e-05
        [opt_reshape]: 6.48998e-06
        [fold_const_symbol]: 1.048e-05
        [renormalize]: 8.2e-07
[validate]: 4.48e-05
[backend_pass]: 9.79984e-07
[task_emit]: 0.00753306
[execute]: 8.99998e-06
Sums
    bootstrap                            :     0.000374s :  2.10%
    type_inference                       :     0.005175s : 29.09%
    event_method                         :     0.000019s :  0.10%
    auto_monad                           :     0.000066s :  0.37%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.24%
    rewriter_before_opt_a                :     0.000077s :  0.43%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.43%
    jit_opt_a.loop_unroll                :     0.000037s :  0.21%
    jit_opt_a.a_1                        :     0.000695s :  3.91%
    jit_opt_a.with_stream_mark           :     0.000088s :  0.50%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000862s :  4.84%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.13%
    jit_opt_a.cse                        :     0.000045s :  0.26%
    jit_opt_a.replace_applicator         :     0.000021s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000040s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000901s :  5.07%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000030s :  0.17%
    loop_unroll                          :     0.000559s :  3.14%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000017s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000051s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000544s :  3.06%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007533s : 42.35%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000244    23
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.65% :     0.000006s :      3: substitution.graph_param_transform
84.49% :     0.000206s :      4: substitution.inline
 1.65% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.03% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.20% :     0.000005s :      2: substitution.replace_old_param
 5.68% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005094     2
81.44% :     0.004149s :      1: type_inference.infer
18.56% :     0.000945s :      1: type_inference.specialize
------[replace.]   0.000081     6
68.08% :     0.000055s :      4: replace.inline
31.92% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000216     6
94.15% :     0.000204s :      4: match.inline
 5.85% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.49% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 2.14% :     0.000003s :     10: predicate.arithmetic_simplify
 1.59% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.09% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.33% :     0.000007s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :      6: predicate.less_batch_normalization
 1.36% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.63% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 2.65% :     0.000003s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.34% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.80% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.09% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.75% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.69% :     0.000010s :     53: predicate.switch_simplify
 1.67% :     0.000002s :     10: predicate.tile_eliminate
 1.06% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.32% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000673    12
49.61% :     0.000334s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.26% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.13% :     0.000304s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.032765    76
 0.17% :     0.000054s :      1: add_recomputation
 0.21% :     0.000070s :      1: auto_monad
 0.06% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.20% :     0.000393s :      1: bootstrap
 0.10% :     0.000033s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.08% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.07% :     0.000023s :      1: event_method
 0.04% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
44.95% :     0.014729s :      1: jit_opt_a
 0.51% :     0.000166s :      1: jit_opt_after_cconv
 0.19% :     0.000063s :      1: jit_opt_b
 1.73% :     0.000568s :      1: loop_unroll
 2.79% :     0.000913s :      1: mutable_eliminate
 2.85% :     0.000934s :     26: opt.transform.jit_opt_a
 0.17% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000030s :      4: opt.transform.jit_opt_b
 0.05% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.69% :     0.000553s :      1: opt_after_jit_grad
 0.03% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000046s :      1: py_interpret_to_execute
 0.05% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000020s :      1: remove_dup_value
 1.39% :     0.000456s :      1: renormalize.infer
 1.21% :     0.000397s :      1: renormalize.specialize
 0.03% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000048s :      1: rewriter_after_opt_a
 0.24% :     0.000080s :      1: rewriter_before_opt_a
 0.25% :     0.000083s :      1: symbol_engine_optimizer
23.05% :     0.007552s :      1: task_emit
15.84% :     0.005191s :      1: type_inference
 0.21% :     0.000068s :      1: validate


TotalTime = 0.0174385, [33]
[bootstrap]: 0.00026233
[type_inference]: 0.00419018
[event_method]: 1.625e-05
[auto_monad]: 5.484e-05
[graph_reusing]: 5.30999e-06
[pre_auto_parallel]: 2.02001e-06
[py_interpret_to_execute]: 3.227e-05
[rewriter_before_opt_a]: 6.785e-05
[expand_dump_flag]: 3.38e-06
[jit_opt_a]: 0.00487973, [2]
    [Cycle 1]: 0.00148293, [27]
        [switch_simplify]: 4.589e-05
        [loop_unroll]: 3.072e-05
        [a_1]: 0.00053438
        [with_stream_mark]: 1.556e-05
        [recompute_prepare]: 8.09002e-06
        [updatestate_depend_eliminate]: 3.66001e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [parameter_eliminate]: 1.44998e-06
        [specialize_transform]: 6.23e-06
        [updatestate_useless_node_eliminater]: 5.25999e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 1.66e-06
        [get_grad_eliminate_]: 5.19998e-06
        [merge_forward]: 3.26999e-06
        [cell_reuse_recompute_pass]: 9.10019e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.245e-05
        [j_node_and_user_rematch]: 8.92e-06
        [meta_fg_expand]: 1.97001e-06
        [replace_old_param]: 8.39998e-06
        [inline_without_move]: 5.87999e-06
        [renormalize]: 0.00056004
        [add_forward_monad_depend]: 4.28999e-06
        [auto_monad_grad]: 2.17999e-06
        [auto_monad_eliminator]: 1.137e-05
        [cse]: 1.942e-05
        [replace_applicator]: 1.174e-05
    [Cycle 2]: 0.00031747, [27]
        [switch_simplify]: 5.56e-06
        [loop_unroll]: 4.78001e-06
        [a_1]: 9.333e-05
        [with_stream_mark]: 8.85999e-06
        [recompute_prepare]: 5.02999e-06
        [updatestate_depend_eliminate]: 2.39999e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 5.02e-06
        [updatestate_useless_node_eliminater]: 5.12999e-06
        [accelerated_algorithm]: 4.72e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.85999e-06
        [merge_forward]: 2.59001e-06
        [cell_reuse_recompute_pass]: 1.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.175e-05
        [j_node_and_user_rematch]: 8.39998e-06
        [meta_fg_expand]: 1.60001e-06
        [replace_old_param]: 6.97002e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.23002e-06
        [auto_monad_grad]: 9.29984e-07
        [auto_monad_eliminator]: 4.83001e-06
        [cse]: 9.84001e-06
        [replace_applicator]: 4.66997e-06
[py_interpret_to_execute_after_opt_a]: 9.91998e-06
[rewriter_after_opt_a]: 2.946e-05
[convert_after_rewriter]: 6.04999e-06
[order_py_execute_after_rewriter]: 5.17999e-06
[mutable_eliminate]: 0.00049319
[jit_opt_b]: 4.871e-05, [1]
    [Cycle 1]: 4.211e-05, [2]
        [frontend_op_eliminate]: 1.599e-05
        [inline_after_opt_a]: 1.515e-05
[cconv]: 2.034e-05
[loop_unroll]: 0.00041648
[jit_opt_after_cconv]: 0.00013482, [1]
    [Cycle 1]: 0.00012859, [11]
        [c_1]: 2.079e-05
        [parameter_eliminate]: 2.47001e-06
        [updatestate_depend_eliminate]: 5.14998e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 1.607e-05
        [call_graph_tuple_transform]: 1.745e-05
        [tuple_list_get_item_eliminator]: 5.39e-06
        [none_parameter_eliminate]: 1.10001e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.49998e-06
[remove_dup_value]: 1.183e-05
[partial_unused_args_eliminate]: 1.89e-06
[environ_conv]: 5.37999e-06
[add_recomputation]: 3.94e-05
[cse_after_recomputation]: 2.099e-05, [1]
    [Cycle 1]: 1.579e-05, [1]
        [cse]: 9.89999e-06
[auto_monad_reorder]: 1.366e-05
[get_jit_bprop_graph]: 1.98002e-06
[rewriter_after_jit_bprop_graph]: 5.14003e-06
[opt_after_jit_grad]: 0.00045622
[symbol_engine_optimizer]: 7.235e-05, [1]
    [Cycle 1]: 6.65e-05, [6]
        [build]: 3.2e-06
        [elim_shapecalc]: 7.42998e-06
        [elim_not_effective]: 1.275e-05
        [opt_reshape]: 5.92001e-06
        [fold_const_symbol]: 8.88002e-06
        [renormalize]: 7.50006e-07
[validate]: 3.124e-05
[backend_pass]: 6.59988e-07
[task_emit]: 0.00590851
[execute]: 7.73999e-06
Sums
    bootstrap                            :     0.000262s :  1.90%
    type_inference                       :     0.004190s : 30.41%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000055s :  0.40%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000032s :  0.23%
    rewriter_before_opt_a                :     0.000068s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000051s :  0.37%
    jit_opt_a.loop_unroll                :     0.000036s :  0.26%
    jit_opt_a.a_1                        :     0.000628s :  4.55%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000560s :  4.06%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.12%
    jit_opt_a.cse                        :     0.000029s :  0.21%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000029s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000493s :  3.58%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000020s :  0.15%
    loop_unroll                          :     0.000416s :  3.02%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000039s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000456s :  3.31%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000031s :  0.23%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005909s : 42.87%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000177    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.61% :     0.000005s :      3: substitution.graph_param_transform
83.33% :     0.000147s :      4: substitution.inline
 1.78% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.26% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000003s :      2: substitution.replace_old_param
 6.30% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004137     2
81.81% :     0.003385s :      1: type_inference.infer
18.19% :     0.000753s :      1: type_inference.specialize
------[replace.]   0.000064     6
64.64% :     0.000041s :      4: replace.inline
35.36% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000155     6
93.58% :     0.000145s :      4: match.inline
 6.42% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   691
 1.22% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.20% :     0.000002s :     10: predicate.addn_zero_filter
 1.68% :     0.000002s :     10: predicate.arithmetic_simplify
 1.21% :     0.000002s :     10: predicate.cast_eliminate
 0.63% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 5.89% :     0.000008s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.72% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 4.27% :     0.000006s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.88% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.14% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.90% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.72% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.54% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 1.88% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.51% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.22% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.78% :     0.000009s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.02% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.09% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 9.28% :     0.000012s :     18: predicate.tuple_list_get_item_eliminator
 1.25% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.33% :     0.000002s :     10: predicate.value_based_eliminate
 0.51% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000513    12
47.58% :     0.000244s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.89% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.53% :     0.000239s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018920    76
 0.22% :     0.000042s :      1: add_recomputation
 0.31% :     0.000058s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.45% :     0.000274s :      1: bootstrap
 0.12% :     0.000023s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.81% :     0.004882s :      1: jit_opt_a
 0.73% :     0.000138s :      1: jit_opt_after_cconv
 0.27% :     0.000051s :      1: jit_opt_b
 2.24% :     0.000424s :      1: loop_unroll
 2.65% :     0.000502s :      1: mutable_eliminate
 4.38% :     0.000828s :     26: opt.transform.jit_opt_a
 0.24% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.45% :     0.000464s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000035s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 1.58% :     0.000299s :      1: renormalize.infer
 1.28% :     0.000243s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000032s :      1: rewriter_after_opt_a
 0.37% :     0.000071s :      1: rewriter_before_opt_a
 0.40% :     0.000075s :      1: symbol_engine_optimizer
31.30% :     0.005922s :      1: task_emit
22.20% :     0.004201s :      1: type_inference
 0.26% :     0.000049s :      1: validate


TotalTime = 0.0138003, [30]
[bootstrap]: 0.00025876
[type_inference]: 0.0060286
[event_method]: 2.028e-05
[auto_monad]: 7.969e-05
[graph_reusing]: 6.16e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 4.374e-05
[rewriter_before_opt_a]: 8.509e-05
[expand_dump_flag]: 3.13e-06
[jit_opt_a]: 0.00528283, [2]
    [Cycle 1]: 0.00176812, [27]
        [switch_simplify]: 0.0001129
        [loop_unroll]: 3.309e-05
        [a_1]: 0.00065106
        [with_stream_mark]: 1.812e-05
        [recompute_prepare]: 7.38999e-06
        [updatestate_depend_eliminate]: 3.46001e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.87002e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 5.60001e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.54998e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 5.05999e-06
        [merge_forward]: 4.01001e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.354e-05
        [j_node_and_user_rematch]: 8.84e-06
        [meta_fg_expand]: 2.19999e-06
        [replace_old_param]: 8.92e-06
        [inline_without_move]: 5.46e-06
        [renormalize]: 0.00063671
        [add_forward_monad_depend]: 4.83001e-06
        [auto_monad_grad]: 2.23998e-06
        [auto_monad_eliminator]: 1.281e-05
        [cse]: 2.891e-05
        [replace_applicator]: 1.121e-05
    [Cycle 2]: 0.00030115, [27]
        [switch_simplify]: 5.48002e-06
        [loop_unroll]: 4.99e-06
        [a_1]: 8.681e-05
        [with_stream_mark]: 8.28001e-06
        [recompute_prepare]: 4.85001e-06
        [updatestate_depend_eliminate]: 2.52001e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 8.89995e-07
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.64998e-06
        [accelerated_algorithm]: 4.65999e-06
        [meta_shard_fg_expand]: 1.06002e-06
        [get_grad_eliminate_]: 4.38001e-06
        [merge_forward]: 2.32001e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.156e-05
        [j_node_and_user_rematch]: 7.38e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 6.51e-06
        [inline_without_move]: 4.82998e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.05999e-06
        [auto_monad_grad]: 7.10017e-07
        [auto_monad_eliminator]: 4.72e-06
        [cse]: 8.97e-06
        [replace_applicator]: 5.19e-06
[py_interpret_to_execute_after_opt_a]: 9.39e-06
[rewriter_after_opt_a]: 3.363e-05
[convert_after_rewriter]: 6.57002e-06
[order_py_execute_after_rewriter]: 4.82998e-06
[mutable_eliminate]: 0.0004906
[jit_opt_b]: 5.413e-05, [1]
    [Cycle 1]: 4.028e-05, [2]
        [frontend_op_eliminate]: 1.458e-05
        [inline_after_opt_a]: 1.4e-05
[cconv]: 2.449e-05
[loop_unroll]: 0.00041289
[jit_opt_after_cconv]: 0.00013456, [1]
    [Cycle 1]: 0.00012833, [11]
        [c_1]: 2.007e-05
        [parameter_eliminate]: 2.56e-06
        [updatestate_depend_eliminate]: 5.14e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.701e-05
        [call_graph_tuple_transform]: 1.741e-05
        [tuple_list_get_item_eliminator]: 5.52001e-06
        [none_parameter_eliminate]: 1.63002e-06
        [renormalize]: 3.30008e-07
        [switch_simplify]: 5.02e-06
[remove_dup_value]: 1.349e-05
[partial_unused_args_eliminate]: 2.04999e-06
[environ_conv]: 5.32001e-06
[add_recomputation]: 4.788e-05
[cse_after_recomputation]: 2.011e-05, [1]
    [Cycle 1]: 1.509e-05, [1]
        [cse]: 9.37001e-06
[auto_monad_reorder]: 1.526e-05
[get_jit_bprop_graph]: 1.54998e-06
[rewriter_after_jit_bprop_graph]: 5.27001e-06
[opt_after_jit_grad]: 0.00044523
[symbol_engine_optimizer]: 7.032e-05, [1]
    [Cycle 1]: 6.403e-05, [6]
        [build]: 3.71999e-06
        [elim_shapecalc]: 7.2e-06
        [elim_not_effective]: 1.201e-05
        [opt_reshape]: 5.34998e-06
        [fold_const_symbol]: 7.78001e-06
        [renormalize]: 3.59985e-07
[validate]: 3.324e-05
Sums
    bootstrap                            :     0.000259s :  2.58%
    type_inference                       :     0.006029s : 60.14%
    event_method                         :     0.000020s :  0.20%
    auto_monad                           :     0.000080s :  0.79%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000044s :  0.44%
    rewriter_before_opt_a                :     0.000085s :  0.85%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000118s :  1.18%
    jit_opt_a.loop_unroll                :     0.000038s :  0.38%
    jit_opt_a.a_1                        :     0.000738s :  7.36%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000010s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000637s :  6.35%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.17%
    jit_opt_a.cse                        :     0.000038s :  0.38%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.09%
    rewriter_after_opt_a                 :     0.000034s :  0.34%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000491s :  4.89%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.14%
    cconv                                :     0.000024s :  0.24%
    loop_unroll                          :     0.000413s :  4.12%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000048s :  0.48%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000445s :  4.44%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.33%


Time group info:
------[substitution.]   0.000254    26
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 1.92% :     0.000005s :      3: substitution.graph_param_transform
85.00% :     0.000216s :      6: substitution.inline
 1.29% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.79% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.43% :     0.000004s :      2: substitution.replace_old_param
 2.98% :     0.000008s :      1: substitution.switch_simplify
 4.37% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005967     2
83.81% :     0.005001s :      1: type_inference.infer
16.19% :     0.000966s :      1: type_inference.specialize
------[replace.]   0.000127     9
50.14% :     0.000063s :      6: replace.inline
27.67% :     0.000035s :      1: replace.switch_simplify
22.19% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000229     9
92.69% :     0.000213s :      6: match.inline
 2.96% :     0.000007s :      1: match.switch_simplify
 4.35% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   801
 1.46% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.08% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.17% :     0.000001s :     12: predicate.addn_check_dump
 1.35% :     0.000002s :     12: predicate.addn_zero_filter
 1.96% :     0.000002s :     12: predicate.arithmetic_simplify
 1.29% :     0.000002s :     12: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.19% :     0.000001s :     12: predicate.compare_switch_simplify
 1.51% :     0.000002s :     12: predicate.depend_value_elim
 1.20% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.18% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.19% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     12: predicate.environ_get_eliminate
 1.20% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.76% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.28% :     0.000007s :     26: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000002s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     17: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.76% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     12: predicate.merge_addn
 1.20% :     0.000002s :     12: predicate.minmaximum_grad
 1.71% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     17: predicate.partial_eliminate
 1.39% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     12: predicate.reduce_eliminate
 1.54% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     20: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     12: predicate.reshape_eliminate
 1.28% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     12: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.84% :     0.000004s :     20: predicate.switch_defer_inline
 2.36% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.45% :     0.000011s :     65: predicate.switch_simplify
 1.28% :     0.000002s :     12: predicate.tile_eliminate
 1.29% :     0.000002s :     12: predicate.transpose_eliminate
 1.55% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.51% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     12: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000721    15
50.78% :     0.000366s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.40% :     0.000082s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.82% :     0.000273s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015492    72
 0.33% :     0.000051s :      1: add_recomputation
 0.54% :     0.000083s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 1.76% :     0.000273s :      1: bootstrap
 0.17% :     0.000027s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
34.12% :     0.005286s :      1: jit_opt_a
 0.89% :     0.000137s :      1: jit_opt_after_cconv
 0.37% :     0.000057s :      1: jit_opt_b
 2.71% :     0.000420s :      1: loop_unroll
 3.22% :     0.000499s :      1: mutable_eliminate
 6.47% :     0.001002s :     26: opt.transform.jit_opt_a
 0.29% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000022s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.92% :     0.000453s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.30% :     0.000046s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.36% :     0.000365s :      1: renormalize.infer
 1.70% :     0.000264s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000036s :      1: rewriter_after_opt_a
 0.57% :     0.000088s :      1: rewriter_before_opt_a
 0.47% :     0.000073s :      1: symbol_engine_optimizer
39.00% :     0.006042s :      1: type_inference


TotalTime = 0.0170395, [33]
[bootstrap]: 0.00027485
[type_inference]: 0.00409935
[event_method]: 1.665e-05
[auto_monad]: 4.263e-05
[graph_reusing]: 3.85e-06
[pre_auto_parallel]: 1.76e-06
[py_interpret_to_execute]: 3.051e-05
[rewriter_before_opt_a]: 7.248e-05
[expand_dump_flag]: 2.39999e-06
[jit_opt_a]: 0.00477541, [2]
    [Cycle 1]: 0.0014049, [27]
        [switch_simplify]: 4.531e-05
        [loop_unroll]: 3.156e-05
        [a_1]: 0.00049387
        [with_stream_mark]: 1.434e-05
        [recompute_prepare]: 7.41999e-06
        [updatestate_depend_eliminate]: 2.89999e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 5.96998e-06
        [updatestate_useless_node_eliminater]: 5.18002e-06
        [accelerated_algorithm]: 5.59998e-06
        [meta_shard_fg_expand]: 1.40001e-06
        [get_grad_eliminate_]: 5.20001e-06
        [merge_forward]: 2.93e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.158e-05
        [j_node_and_user_rematch]: 8.80999e-06
        [meta_fg_expand]: 1.99e-06
        [replace_old_param]: 7.83001e-06
        [inline_without_move]: 5.22e-06
        [renormalize]: 0.00053168
        [add_forward_monad_depend]: 4.21001e-06
        [auto_monad_grad]: 1.73002e-06
        [auto_monad_eliminator]: 1.006e-05
        [cse]: 1.722e-05
        [replace_applicator]: 1.224e-05
    [Cycle 2]: 0.00032373, [27]
        [switch_simplify]: 5.35001e-06
        [loop_unroll]: 5.12e-06
        [a_1]: 9.013e-05
        [with_stream_mark]: 8.56997e-06
        [recompute_prepare]: 4.99998e-06
        [updatestate_depend_eliminate]: 2.66e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 4.70999e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.18002e-06
        [meta_shard_fg_expand]: 1.32999e-06
        [get_grad_eliminate_]: 4.58999e-06
        [merge_forward]: 2.42001e-06
        [cell_reuse_recompute_pass]: 1.51998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.278e-05
        [j_node_and_user_rematch]: 8.17003e-06
        [meta_fg_expand]: 1.54998e-06
        [replace_old_param]: 1.047e-05
        [inline_without_move]: 4.84003e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.27999e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 5.36002e-06
        [cse]: 1.096e-05
        [replace_applicator]: 4.97e-06
[py_interpret_to_execute_after_opt_a]: 8.82999e-06
[rewriter_after_opt_a]: 2.662e-05
[convert_after_rewriter]: 6.29001e-06
[order_py_execute_after_rewriter]: 4.83001e-06
[mutable_eliminate]: 0.00047907
[jit_opt_b]: 5.117e-05, [1]
    [Cycle 1]: 4.477e-05, [2]
        [frontend_op_eliminate]: 1.766e-05
        [inline_after_opt_a]: 1.481e-05
[cconv]: 1.577e-05
[loop_unroll]: 0.00042433
[jit_opt_after_cconv]: 0.0001365, [1]
    [Cycle 1]: 0.00012999, [11]
        [c_1]: 2.013e-05
        [parameter_eliminate]: 2.71e-06
        [updatestate_depend_eliminate]: 5.48002e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.722e-05
        [call_graph_tuple_transform]: 1.701e-05
        [tuple_list_get_item_eliminator]: 5.39998e-06
        [none_parameter_eliminate]: 9.70002e-07
        [renormalize]: 6.29982e-07
        [switch_simplify]: 5.57999e-06
[remove_dup_value]: 6.83e-06
[partial_unused_args_eliminate]: 1.55999e-06
[environ_conv]: 3.3e-06
[add_recomputation]: 3.225e-05
[cse_after_recomputation]: 1.974e-05, [1]
    [Cycle 1]: 1.488e-05, [1]
        [cse]: 9.24e-06
[auto_monad_reorder]: 9.09e-06
[get_jit_bprop_graph]: 1.41002e-06
[rewriter_after_jit_bprop_graph]: 4.48999e-06
[opt_after_jit_grad]: 0.00047098
[symbol_engine_optimizer]: 7.002e-05, [1]
    [Cycle 1]: 6.371e-05, [6]
        [build]: 2.75997e-06
        [elim_shapecalc]: 7.44002e-06
        [elim_not_effective]: 1.199e-05
        [opt_reshape]: 5.51998e-06
        [fold_const_symbol]: 8.32e-06
        [renormalize]: 3.80009e-07
[validate]: 2.687e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00572274
[execute]: 9.25001e-06
Sums
    bootstrap                            :     0.000275s :  2.05%
    type_inference                       :     0.004099s : 30.58%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000043s :  0.32%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000031s :  0.23%
    rewriter_before_opt_a                :     0.000072s :  0.54%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000051s :  0.38%
    jit_opt_a.loop_unroll                :     0.000037s :  0.27%
    jit_opt_a.a_1                        :     0.000584s :  4.36%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000532s :  3.97%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.12%
    jit_opt_a.cse                        :     0.000028s :  0.21%
    jit_opt_a.replace_applicator         :     0.000017s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000027s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000479s :  3.57%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000016s :  0.12%
    loop_unroll                          :     0.000424s :  3.17%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000007s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000003s :  0.02%
    add_recomputation                    :     0.000032s :  0.24%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000009s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000471s :  3.51%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005723s : 42.69%
    execute                              :     0.000009s :  0.07%


Time group info:
------[substitution.]   0.000171    23
 0.84% :     0.000001s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.08% :     0.000004s :      3: substitution.graph_param_transform
84.28% :     0.000144s :      4: substitution.inline
 1.68% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.21% :     0.000004s :      4: substitution.remove_not_recompute_node
 3.85% :     0.000007s :      2: substitution.replace_old_param
 4.51% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004050     2
80.41% :     0.003256s :      1: type_inference.infer
19.59% :     0.000793s :      1: type_inference.specialize
------[replace.]   0.000063     6
65.89% :     0.000042s :      4: replace.inline
34.11% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000148     6
95.65% :     0.000141s :      4: match.inline
 4.35% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.29% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.62% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 1.84% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.46% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.55% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.39% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.20% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.64% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.42% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.84% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000002s :     15: predicate.partial_eliminate
 1.25% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.34% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.97% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.72% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.63% :     0.000002s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.25% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.16% :     0.000008s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.34% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000524    12
45.63% :     0.000239s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.66% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.71% :     0.000245s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018464    76
 0.19% :     0.000035s :      1: add_recomputation
 0.25% :     0.000046s :      1: auto_monad
 0.06% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.55% :     0.000287s :      1: bootstrap
 0.10% :     0.000019s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
25.88% :     0.004778s :      1: jit_opt_a
 0.75% :     0.000139s :      1: jit_opt_after_cconv
 0.29% :     0.000054s :      1: jit_opt_b
 2.34% :     0.000432s :      1: loop_unroll
 2.64% :     0.000488s :      1: mutable_eliminate
 4.26% :     0.000787s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.60% :     0.000480s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000009s :      1: remove_dup_value
 1.45% :     0.000268s :      1: renormalize.infer
 1.39% :     0.000257s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000030s :      1: rewriter_after_opt_a
 0.41% :     0.000076s :      1: rewriter_before_opt_a
 0.39% :     0.000073s :      1: symbol_engine_optimizer
31.07% :     0.005738s :      1: task_emit
22.27% :     0.004111s :      1: type_inference
 0.24% :     0.000044s :      1: validate


TotalTime = 0.0265455, [33]
[bootstrap]: 0.00038198
[type_inference]: 0.00852883
[event_method]: 1.814e-05
[auto_monad]: 6.953e-05
[graph_reusing]: 6.37001e-06
[pre_auto_parallel]: 2.74999e-06
[py_interpret_to_execute]: 4.533e-05
[rewriter_before_opt_a]: 7.978e-05
[expand_dump_flag]: 3.41999e-06
[jit_opt_a]: 0.00783432, [2]
    [Cycle 1]: 0.0018825, [27]
        [switch_simplify]: 6.882e-05
        [loop_unroll]: 3.131e-05
        [a_1]: 0.00058612
        [with_stream_mark]: 2.058e-05
        [recompute_prepare]: 7.85e-06
        [updatestate_depend_eliminate]: 3.82002e-06
        [updatestate_assign_eliminate]: 3.41001e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [parameter_eliminate]: 2.57001e-06
        [specialize_transform]: 6.11e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 5.92001e-06
        [meta_shard_fg_expand]: 2.58e-06
        [get_grad_eliminate_]: 5.69e-06
        [merge_forward]: 3.83001e-06
        [cell_reuse_recompute_pass]: 1.60999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.461e-05
        [j_node_and_user_rematch]: 9.71e-06
        [meta_fg_expand]: 2.38002e-06
        [replace_old_param]: 9.09e-06
        [inline_without_move]: 6.01e-06
        [renormalize]: 0.00083886
        [add_forward_monad_depend]: 6.14001e-06
        [auto_monad_grad]: 2.48e-06
        [auto_monad_eliminator]: 1.506e-05
        [cse]: 3.012e-05
        [replace_applicator]: 1.321e-05
    [Cycle 2]: 0.00033584, [27]
        [switch_simplify]: 6.32001e-06
        [loop_unroll]: 5.55001e-06
        [a_1]: 9.28e-05
        [with_stream_mark]: 9.91e-06
        [recompute_prepare]: 5.42999e-06
        [updatestate_depend_eliminate]: 2.69999e-06
        [updatestate_assign_eliminate]: 2.06998e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 1.19998e-06
        [specialize_transform]: 5.25001e-06
        [updatestate_useless_node_eliminater]: 5.12999e-06
        [accelerated_algorithm]: 5.15999e-06
        [meta_shard_fg_expand]: 1.15999e-06
        [get_grad_eliminate_]: 4.58999e-06
        [merge_forward]: 2.44001e-06
        [cell_reuse_recompute_pass]: 1.86e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.302e-05
        [j_node_and_user_rematch]: 8.1e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 7.69002e-06
        [inline_without_move]: 4.67998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.40001e-06
        [auto_monad_grad]: 1.00999e-06
        [auto_monad_eliminator]: 5.37999e-06
        [cse]: 1.109e-05
        [replace_applicator]: 5.12e-06
[py_interpret_to_execute_after_opt_a]: 1.202e-05
[rewriter_after_opt_a]: 3.827e-05
[convert_after_rewriter]: 8.66002e-06
[order_py_execute_after_rewriter]: 5.72001e-06
[mutable_eliminate]: 0.00077143
[jit_opt_b]: 5.1e-05, [1]
    [Cycle 1]: 4.367e-05, [2]
        [frontend_op_eliminate]: 1.696e-05
        [inline_after_opt_a]: 1.513e-05
[cconv]: 2.724e-05
[loop_unroll]: 0.00044102
[jit_opt_after_cconv]: 0.00014544, [1]
    [Cycle 1]: 0.00013877, [11]
        [c_1]: 2.179e-05
        [parameter_eliminate]: 2.58e-06
        [updatestate_depend_eliminate]: 6.12001e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 1.868e-05
        [call_graph_tuple_transform]: 2.014e-05
        [tuple_list_get_item_eliminator]: 5.64998e-06
        [none_parameter_eliminate]: 1.86e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.56002e-06
[remove_dup_value]: 1.564e-05
[partial_unused_args_eliminate]: 3.01001e-06
[environ_conv]: 4.76997e-06
[add_recomputation]: 4.945e-05
[cse_after_recomputation]: 2.199e-05, [1]
    [Cycle 1]: 1.623e-05, [1]
        [cse]: 9.94999e-06
[auto_monad_reorder]: 1.649e-05
[get_jit_bprop_graph]: 2.14e-06
[rewriter_after_jit_bprop_graph]: 5.35001e-06
[opt_after_jit_grad]: 0.00046708
[symbol_engine_optimizer]: 7.802e-05, [1]
    [Cycle 1]: 7.087e-05, [6]
        [build]: 4e-06
        [elim_shapecalc]: 8.27998e-06
        [elim_not_effective]: 1.387e-05
        [opt_reshape]: 5.74e-06
        [fold_const_symbol]: 9.79999e-06
        [renormalize]: 7.10017e-07
[validate]: 4.106e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.0071234
[execute]: 8.40001e-06
Sums
    bootstrap                            :     0.000382s :  1.88%
    type_inference                       :     0.008529s : 42.08%
    event_method                         :     0.000018s :  0.09%
    auto_monad                           :     0.000070s :  0.34%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000045s :  0.22%
    rewriter_before_opt_a                :     0.000080s :  0.39%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000075s :  0.37%
    jit_opt_a.loop_unroll                :     0.000037s :  0.18%
    jit_opt_a.a_1                        :     0.000679s :  3.35%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.08%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.000839s :  4.14%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.10%
    jit_opt_a.cse                        :     0.000041s :  0.20%
    jit_opt_a.replace_applicator         :     0.000018s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.06%
    rewriter_after_opt_a                 :     0.000038s :  0.19%
    convert_after_rewriter               :     0.000009s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000771s :  3.81%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.07%
    cconv                                :     0.000027s :  0.13%
    loop_unroll                          :     0.000441s :  2.18%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000049s :  0.24%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000467s :  2.30%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.20%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007123s : 35.14%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000229    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000002s :      2: substitution.fold_const_symbol
 2.48% :     0.000006s :      3: substitution.graph_param_transform
84.19% :     0.000193s :      4: substitution.inline
 1.58% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.19% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.86% :     0.000004s :      2: substitution.replace_old_param
 5.95% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008439     2
66.69% :     0.005628s :      1: type_inference.infer
33.31% :     0.002811s :      1: type_inference.specialize
------[replace.]   0.000080     6
67.23% :     0.000054s :      4: replace.inline
32.77% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000203     6
93.85% :     0.000190s :      4: match.inline
 6.15% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.58% :     0.000003s :     10: predicate.arithmetic_simplify
 1.72% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.47% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.48% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.09% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.16% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 5.39% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.76% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 1.21% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.25% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.65% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.01% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.89% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.53% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.30% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.11% :     0.000010s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.60% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.16% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.95% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.67% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.53% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000719    12
48.59% :     0.000350s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.65% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.76% :     0.000336s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028388    76
 0.18% :     0.000052s :      1: add_recomputation
 0.26% :     0.000073s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.42% :     0.000403s :      1: bootstrap
 0.11% :     0.000030s :      1: cconv
 0.04% :     0.000012s :      1: convert_after_rewriter
 0.08% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000023s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
27.61% :     0.007838s :      1: jit_opt_a
 0.52% :     0.000148s :      1: jit_opt_after_cconv
 0.19% :     0.000054s :      1: jit_opt_b
 1.58% :     0.000449s :      1: loop_unroll
 2.75% :     0.000780s :      1: mutable_eliminate
 3.20% :     0.000909s :     26: opt.transform.jit_opt_a
 0.17% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000026s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.67% :     0.000475s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000048s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000018s :      1: remove_dup_value
 1.57% :     0.000446s :      1: renormalize.infer
 1.35% :     0.000384s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000043s :      1: rewriter_after_opt_a
 0.29% :     0.000083s :      1: rewriter_before_opt_a
 0.28% :     0.000081s :      1: symbol_engine_optimizer
25.16% :     0.007144s :      1: task_emit
30.11% :     0.008549s :      1: type_inference
 0.22% :     0.000061s :      1: validate


TotalTime = 0.0190619, [33]
[bootstrap]: 0.00032477
[type_inference]: 0.00457959
[event_method]: 1.682e-05
[auto_monad]: 6.428e-05
[graph_reusing]: 5.62999e-06
[pre_auto_parallel]: 2.44001e-06
[py_interpret_to_execute]: 3.08e-05
[rewriter_before_opt_a]: 7.087e-05
[expand_dump_flag]: 2.88998e-06
[jit_opt_a]: 0.00540819, [2]
    [Cycle 1]: 0.00158947, [27]
        [switch_simplify]: 5.291e-05
        [loop_unroll]: 2.933e-05
        [a_1]: 0.00050883
        [with_stream_mark]: 1.888e-05
        [recompute_prepare]: 7.21001e-06
        [updatestate_depend_eliminate]: 3.95e-06
        [updatestate_assign_eliminate]: 2.71e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 5.84999e-06
        [updatestate_useless_node_eliminater]: 4.67998e-06
        [accelerated_algorithm]: 5.01002e-06
        [meta_shard_fg_expand]: 1.94999e-06
        [get_grad_eliminate_]: 5.23002e-06
        [merge_forward]: 3.87998e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.372e-05
        [j_node_and_user_rematch]: 8.57998e-06
        [meta_fg_expand]: 2.32001e-06
        [replace_old_param]: 9.17001e-06
        [inline_without_move]: 5.72001e-06
        [renormalize]: 0.00065384
        [add_forward_monad_depend]: 5.77001e-06
        [auto_monad_grad]: 2.44999e-06
        [auto_monad_eliminator]: 1.45e-05
        [cse]: 2.733e-05
        [replace_applicator]: 1.254e-05
    [Cycle 2]: 0.00032444, [27]
        [switch_simplify]: 5.75001e-06
        [loop_unroll]: 5.39e-06
        [a_1]: 8.697e-05
        [with_stream_mark]: 9.81998e-06
        [recompute_prepare]: 4.75001e-06
        [updatestate_depend_eliminate]: 2.37999e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 4.90001e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 4.92999e-06
        [meta_shard_fg_expand]: 1.23002e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.29999e-06
        [cell_reuse_recompute_pass]: 1.50001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.353e-05
        [j_node_and_user_rematch]: 7.73999e-06
        [meta_fg_expand]: 1.72999e-06
        [replace_old_param]: 7.28e-06
        [inline_without_move]: 4.79e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.36998e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 5.31998e-06
        [cse]: 1.124e-05
        [replace_applicator]: 4.90001e-06
[py_interpret_to_execute_after_opt_a]: 1.012e-05
[rewriter_after_opt_a]: 3.6e-05
[convert_after_rewriter]: 8.73001e-06
[order_py_execute_after_rewriter]: 4.76002e-06
[mutable_eliminate]: 0.00056215
[jit_opt_b]: 4.716e-05, [1]
    [Cycle 1]: 4.042e-05, [2]
        [frontend_op_eliminate]: 1.474e-05
        [inline_after_opt_a]: 1.467e-05
[cconv]: 2.624e-05
[loop_unroll]: 0.00043891
[jit_opt_after_cconv]: 0.00013844, [1]
    [Cycle 1]: 0.00013191, [11]
        [c_1]: 1.952e-05
        [parameter_eliminate]: 2.63e-06
        [updatestate_depend_eliminate]: 5.88998e-06
        [updatestate_assign_eliminate]: 2.14999e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [cse]: 1.994e-05
        [call_graph_tuple_transform]: 1.879e-05
        [tuple_list_get_item_eliminator]: 4.97e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 7.89994e-07
        [switch_simplify]: 5.09e-06
[remove_dup_value]: 1.283e-05
[partial_unused_args_eliminate]: 2.02999e-06
[environ_conv]: 5.19998e-06
[add_recomputation]: 4.72e-05
[cse_after_recomputation]: 1.998e-05, [1]
    [Cycle 1]: 1.438e-05, [1]
        [cse]: 8.65001e-06
[auto_monad_reorder]: 1.54e-05
[get_jit_bprop_graph]: 1.94999e-06
[rewriter_after_jit_bprop_graph]: 5.66e-06
[opt_after_jit_grad]: 0.00045401
[symbol_engine_optimizer]: 7.115e-05, [1]
    [Cycle 1]: 6.418e-05, [6]
        [build]: 3.65998e-06
        [elim_shapecalc]: 7.41999e-06
        [elim_not_effective]: 1.199e-05
        [opt_reshape]: 5.33002e-06
        [fold_const_symbol]: 8.70999e-06
        [renormalize]: 4.30009e-07
[validate]: 3.595e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00639172
[execute]: 8.42998e-06
Sums
    bootstrap                            :     0.000325s :  2.17%
    type_inference                       :     0.004580s : 30.65%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000064s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.21%
    rewriter_before_opt_a                :     0.000071s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.23%
    jit_opt_a.a_1                        :     0.000596s :  3.99%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000654s :  4.38%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000039s :  0.26%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.24%
    convert_after_rewriter               :     0.000009s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000562s :  3.76%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000026s :  0.18%
    loop_unroll                          :     0.000439s :  2.94%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000047s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000454s :  3.04%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006392s : 42.78%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000192    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 3.10% :     0.000006s :      3: substitution.graph_param_transform
82.56% :     0.000158s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.38% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.38% :     0.000005s :      2: substitution.replace_old_param
 6.39% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004516     2
82.77% :     0.003738s :      1: type_inference.infer
17.23% :     0.000778s :      1: type_inference.specialize
------[replace.]   0.000065     6
68.51% :     0.000045s :      4: replace.inline
31.49% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000167     6
93.31% :     0.000156s :      4: match.inline
 6.69% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.54% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.14% :     0.000002s :     10: predicate.arithmetic_simplify
 1.31% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000001s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.49% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.62% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 1.86% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.04% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.66% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000000s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.27% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.53% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.74% :     0.000009s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.44% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.70% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.34% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000539    12
50.59% :     0.000273s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.49% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.92% :     0.000237s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020618    76
 0.24% :     0.000050s :      1: add_recomputation
 0.33% :     0.000068s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.64% :     0.000338s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.25% :     0.005412s :      1: jit_opt_a
 0.68% :     0.000141s :      1: jit_opt_after_cconv
 0.24% :     0.000050s :      1: jit_opt_b
 2.17% :     0.000448s :      1: loop_unroll
 2.77% :     0.000571s :      1: mutable_eliminate
 3.89% :     0.000801s :     26: opt.transform.jit_opt_a
 0.22% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.24% :     0.000462s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 1.44% :     0.000298s :      1: renormalize.infer
 1.69% :     0.000349s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000039s :      1: rewriter_after_opt_a
 0.36% :     0.000074s :      1: rewriter_before_opt_a
 0.36% :     0.000074s :      1: symbol_engine_optimizer
31.09% :     0.006411s :      1: task_emit
22.34% :     0.004606s :      1: type_inference
 0.27% :     0.000056s :      1: validate


TotalTime = 0.0139049, [30]
[bootstrap]: 0.00027091
[type_inference]: 0.00622041
[event_method]: 2.032e-05
[auto_monad]: 7.905e-05
[graph_reusing]: 6.53e-06
[pre_auto_parallel]: 2.62001e-06
[py_interpret_to_execute]: 4.492e-05
[rewriter_before_opt_a]: 8.594e-05
[expand_dump_flag]: 3.01001e-06
[jit_opt_a]: 0.00517508, [2]
    [Cycle 1]: 0.00173254, [27]
        [switch_simplify]: 0.00010825
        [loop_unroll]: 3.317e-05
        [a_1]: 0.00065678
        [with_stream_mark]: 1.803e-05
        [recompute_prepare]: 6.63998e-06
        [updatestate_depend_eliminate]: 3.56001e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.40002e-06
        [parameter_eliminate]: 1.77001e-06
        [specialize_transform]: 5.81998e-06
        [updatestate_useless_node_eliminater]: 5.39998e-06
        [accelerated_algorithm]: 5.17e-06
        [meta_shard_fg_expand]: 2.75997e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 3.38999e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.33e-05
        [j_node_and_user_rematch]: 8.67998e-06
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 8.09002e-06
        [inline_without_move]: 5.16002e-06
        [renormalize]: 0.0006065
        [add_forward_monad_depend]: 4.89e-06
        [auto_monad_grad]: 2.43e-06
        [auto_monad_eliminator]: 1.352e-05
        [cse]: 2.551e-05
        [replace_applicator]: 1.128e-05
    [Cycle 2]: 0.00031603, [27]
        [switch_simplify]: 5.72001e-06
        [loop_unroll]: 4.77e-06
        [a_1]: 8.756e-05
        [with_stream_mark]: 8.54e-06
        [recompute_prepare]: 4.95001e-06
        [updatestate_depend_eliminate]: 2.51e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 1.14998e-06
        [specialize_transform]: 4.77e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 5.27999e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.23999e-06
        [merge_forward]: 2.29999e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.23e-05
        [j_node_and_user_rematch]: 7.92998e-06
        [meta_fg_expand]: 1.76003e-06
        [replace_old_param]: 7.06001e-06
        [inline_without_move]: 4.80001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 6.60017e-07
        [auto_monad_eliminator]: 5.04e-06
        [cse]: 9.91e-06
        [replace_applicator]: 4.94e-06
[py_interpret_to_execute_after_opt_a]: 8.79e-06
[rewriter_after_opt_a]: 3.191e-05
[convert_after_rewriter]: 7.51999e-06
[order_py_execute_after_rewriter]: 5.08002e-06
[mutable_eliminate]: 0.00049228
[jit_opt_b]: 4.722e-05, [1]
    [Cycle 1]: 4.045e-05, [2]
        [frontend_op_eliminate]: 1.512e-05
        [inline_after_opt_a]: 1.44e-05
[cconv]: 2.449e-05
[loop_unroll]: 0.00040807
[jit_opt_after_cconv]: 0.00013212, [1]
    [Cycle 1]: 0.00012598, [11]
        [c_1]: 1.978e-05
        [parameter_eliminate]: 2.61e-06
        [updatestate_depend_eliminate]: 5.24e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.637e-05
        [call_graph_tuple_transform]: 1.767e-05
        [tuple_list_get_item_eliminator]: 5.10001e-06
        [none_parameter_eliminate]: 1.25999e-06
        [renormalize]: 5.99975e-07
        [switch_simplify]: 5.04e-06
[remove_dup_value]: 1.347e-05
[partial_unused_args_eliminate]: 2.19999e-06
[environ_conv]: 5.64e-06
[add_recomputation]: 4.785e-05
[cse_after_recomputation]: 1.975e-05, [1]
    [Cycle 1]: 1.417e-05, [1]
        [cse]: 8.70001e-06
[auto_monad_reorder]: 1.5e-05
[get_jit_bprop_graph]: 1.62001e-06
[rewriter_after_jit_bprop_graph]: 4.89998e-06
[opt_after_jit_grad]: 0.00046026
[symbol_engine_optimizer]: 7.007e-05, [1]
    [Cycle 1]: 6.408e-05, [6]
        [build]: 3.53e-06
        [elim_shapecalc]: 6.97002e-06
        [elim_not_effective]: 1.232e-05
        [opt_reshape]: 5.35999e-06
        [fold_const_symbol]: 8.35001e-06
        [renormalize]: 4.80009e-07
[validate]: 3.083e-05
Sums
    bootstrap                            :     0.000271s :  2.65%
    type_inference                       :     0.006220s : 60.93%
    event_method                         :     0.000020s :  0.20%
    auto_monad                           :     0.000079s :  0.77%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000045s :  0.44%
    rewriter_before_opt_a                :     0.000086s :  0.84%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000114s :  1.12%
    jit_opt_a.loop_unroll                :     0.000038s :  0.37%
    jit_opt_a.a_1                        :     0.000744s :  7.29%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000607s :  5.94%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.18%
    jit_opt_a.cse                        :     0.000035s :  0.35%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.09%
    rewriter_after_opt_a                 :     0.000032s :  0.31%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000492s :  4.82%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.14%
    cconv                                :     0.000024s :  0.24%
    loop_unroll                          :     0.000408s :  4.00%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000048s :  0.47%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000460s :  4.51%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.30%


Time group info:
------[substitution.]   0.000251    26
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 2.10% :     0.000005s :      3: substitution.graph_param_transform
84.39% :     0.000212s :      6: substitution.inline
 1.05% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.73% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.36% :     0.000003s :      2: substitution.replace_old_param
 3.17% :     0.000008s :      1: substitution.switch_simplify
 4.99% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006158     2
83.74% :     0.005157s :      1: type_inference.infer
16.26% :     0.001001s :      1: type_inference.specialize
------[replace.]   0.000125     9
51.24% :     0.000064s :      6: replace.inline
26.11% :     0.000033s :      1: replace.switch_simplify
22.65% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000227     9
91.81% :     0.000209s :      6: match.inline
 3.17% :     0.000007s :      1: match.switch_simplify
 5.02% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   801
 1.51% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.19% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000001s :     12: predicate.addn_check_dump
 1.44% :     0.000002s :     12: predicate.addn_zero_filter
 2.08% :     0.000003s :     12: predicate.arithmetic_simplify
 1.24% :     0.000002s :     12: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.33% :     0.000002s :     12: predicate.compare_switch_simplify
 1.43% :     0.000002s :     12: predicate.depend_value_elim
 1.18% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.46% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     12: predicate.environ_get_eliminate
 1.16% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.81% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.45% :     0.000007s :     26: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.94% :     0.000001s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     17: predicate.load_eliminater
 1.32% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.26% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000002s :     12: predicate.merge_addn
 1.18% :     0.000001s :     12: predicate.minmaximum_grad
 1.45% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.34% :     0.000003s :     17: predicate.partial_eliminate
 1.28% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     12: predicate.reduce_eliminate
 1.49% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     20: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.60% :     0.000002s :     12: predicate.reshape_eliminate
 1.31% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.79% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.69% :     0.000003s :     20: predicate.switch_defer_inline
 2.34% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.53% :     0.000011s :     65: predicate.switch_simplify
 1.32% :     0.000002s :     12: predicate.tile_eliminate
 1.23% :     0.000002s :     12: predicate.transpose_eliminate
 1.44% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.96% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     12: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.54% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000742    15
50.97% :     0.000378s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.91% :     0.000081s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.11% :     0.000283s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015572    72
 0.33% :     0.000051s :      1: add_recomputation
 0.53% :     0.000083s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.86% :     0.000290s :      1: bootstrap
 0.17% :     0.000027s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
33.25% :     0.005178s :      1: jit_opt_a
 0.87% :     0.000135s :      1: jit_opt_after_cconv
 0.32% :     0.000050s :      1: jit_opt_b
 2.67% :     0.000416s :      1: loop_unroll
 3.21% :     0.000500s :      1: mutable_eliminate
 6.45% :     0.001004s :     26: opt.transform.jit_opt_a
 0.29% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.01% :     0.000469s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000048s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.11% :     0.000328s :      1: renormalize.infer
 1.74% :     0.000271s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000037s :      1: rewriter_after_opt_a
 0.57% :     0.000089s :      1: rewriter_before_opt_a
 0.47% :     0.000073s :      1: symbol_engine_optimizer
40.03% :     0.006233s :      1: type_inference


TotalTime = 0.0176131, [33]
[bootstrap]: 0.00026203
[type_inference]: 0.00413714
[event_method]: 1.626e-05
[auto_monad]: 6.183e-05
[graph_reusing]: 5.17999e-06
[pre_auto_parallel]: 2.51e-06
[py_interpret_to_execute]: 3.192e-05
[rewriter_before_opt_a]: 6.847e-05
[expand_dump_flag]: 2.99999e-06
[jit_opt_a]: 0.00500343, [2]
    [Cycle 1]: 0.0015183, [27]
        [switch_simplify]: 5.123e-05
        [loop_unroll]: 3.146e-05
        [a_1]: 0.00051584
        [with_stream_mark]: 1.781e-05
        [recompute_prepare]: 7.16999e-06
        [updatestate_depend_eliminate]: 3.7e-06
        [updatestate_assign_eliminate]: 3.45998e-06
        [updatestate_loads_eliminate]: 2.83003e-06
        [parameter_eliminate]: 1.84998e-06
        [specialize_transform]: 6.17001e-06
        [updatestate_useless_node_eliminater]: 5.15999e-06
        [accelerated_algorithm]: 5.25001e-06
        [meta_shard_fg_expand]: 2.19999e-06
        [get_grad_eliminate_]: 5.57001e-06
        [merge_forward]: 3.67002e-06
        [cell_reuse_recompute_pass]: 9.20001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.477e-05
        [j_node_and_user_rematch]: 9.00001e-06
        [meta_fg_expand]: 2.48e-06
        [replace_old_param]: 7.98001e-06
        [inline_without_move]: 5.18002e-06
        [renormalize]: 0.00059772
        [add_forward_monad_depend]: 5.03002e-06
        [auto_monad_grad]: 2.25002e-06
        [auto_monad_eliminator]: 1.314e-05
        [cse]: 2.362e-05
        [replace_applicator]: 1.125e-05
    [Cycle 2]: 0.00032504, [27]
        [switch_simplify]: 5.50001e-06
        [loop_unroll]: 4.89003e-06
        [a_1]: 9.179e-05
        [with_stream_mark]: 8.95999e-06
        [recompute_prepare]: 5.03002e-06
        [updatestate_depend_eliminate]: 2.43e-06
        [updatestate_assign_eliminate]: 2.05002e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 4.87e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.17e-06
        [meta_shard_fg_expand]: 1.09998e-06
        [get_grad_eliminate_]: 4.58999e-06
        [merge_forward]: 2.37001e-06
        [cell_reuse_recompute_pass]: 2.01998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.251e-05
        [j_node_and_user_rematch]: 7.75998e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 7.43999e-06
        [inline_without_move]: 1.174e-05
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.30999e-06
        [auto_monad_grad]: 8.89995e-07
        [auto_monad_eliminator]: 5.12999e-06
        [cse]: 1.043e-05
        [replace_applicator]: 4.97e-06
[py_interpret_to_execute_after_opt_a]: 9.18002e-06
[rewriter_after_opt_a]: 3.405e-05
[convert_after_rewriter]: 7.2e-06
[order_py_execute_after_rewriter]: 5.62999e-06
[mutable_eliminate]: 0.00049117
[jit_opt_b]: 4.924e-05, [1]
    [Cycle 1]: 4.263e-05, [2]
        [frontend_op_eliminate]: 1.662e-05
        [inline_after_opt_a]: 1.441e-05
[cconv]: 2.56e-05
[loop_unroll]: 0.00041315
[jit_opt_after_cconv]: 0.00018234, [1]
    [Cycle 1]: 0.00017559, [11]
        [c_1]: 2.156e-05
        [parameter_eliminate]: 2.51e-06
        [updatestate_depend_eliminate]: 5.43002e-06
        [updatestate_assign_eliminate]: 4.129e-05
        [updatestate_loads_eliminate]: 2.34999e-06
        [cse]: 1.652e-05
        [call_graph_tuple_transform]: 1.89e-05
        [tuple_list_get_item_eliminator]: 5.69e-06
        [none_parameter_eliminate]: 1.95001e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.48002e-06
[remove_dup_value]: 1.408e-05
[partial_unused_args_eliminate]: 2.49001e-06
[environ_conv]: 5.25999e-06
[add_recomputation]: 5.422e-05
[cse_after_recomputation]: 2.179e-05, [1]
    [Cycle 1]: 1.609e-05, [1]
        [cse]: 9.99001e-06
[auto_monad_reorder]: 1.558e-05
[get_jit_bprop_graph]: 1.65001e-06
[rewriter_after_jit_bprop_graph]: 4.91002e-06
[opt_after_jit_grad]: 0.00045813
[symbol_engine_optimizer]: 7.408e-05, [1]
    [Cycle 1]: 6.812e-05, [6]
        [build]: 3.68999e-06
        [elim_shapecalc]: 7.41999e-06
        [elim_not_effective]: 1.32e-05
        [opt_reshape]: 5.92999e-06
        [fold_const_symbol]: 9.15001e-06
        [renormalize]: 3.60014e-07
[validate]: 3.237e-05
[backend_pass]: 8.00006e-07
[task_emit]: 0.00593104
[execute]: 7.73999e-06
Sums
    bootstrap                            :     0.000262s :  1.89%
    type_inference                       :     0.004137s : 29.82%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000062s :  0.45%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.23%
    rewriter_before_opt_a                :     0.000068s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.41%
    jit_opt_a.loop_unroll                :     0.000036s :  0.26%
    jit_opt_a.a_1                        :     0.000608s :  4.38%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000017s :  0.12%
    jit_opt_a.renormalize                :     0.000598s :  4.31%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000034s :  0.25%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000491s :  3.54%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000026s :  0.18%
    loop_unroll                          :     0.000413s :  2.98%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000041s :  0.30%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000054s :  0.39%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000016s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000458s :  3.30%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005931s : 42.75%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000192    23
 1.11% :     0.000002s :      2: substitution.elim_not_effective
 0.83% :     0.000002s :      2: substitution.fold_const_symbol
 2.52% :     0.000005s :      3: substitution.graph_param_transform
83.74% :     0.000161s :      4: substitution.inline
 1.68% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.42% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.71% :     0.000003s :      2: substitution.replace_old_param
 6.00% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004081     2
81.53% :     0.003327s :      1: type_inference.infer
18.47% :     0.000754s :      1: type_inference.specialize
------[replace.]   0.000064     6
66.39% :     0.000042s :      4: replace.inline
33.61% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000169     6
93.86% :     0.000158s :      4: match.inline
 6.14% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.26% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.74% :     0.000002s :     10: predicate.addn_zero_filter
 1.86% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000002s :     10: predicate.cast_eliminate
 0.55% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     10: predicate.compare_switch_simplify
 1.24% :     0.000001s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.63% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.34% :     0.000002s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.12% :     0.000001s :      6: predicate.get_grad_eliminate
 0.33% :     0.000000s :      3: predicate.graph_param_transform
 5.36% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.44% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.67% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.55% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000002s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     10: predicate.reduce_eliminate
 1.60% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.07% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.89% :     0.000009s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.63% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.87% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000520    12
48.07% :     0.000250s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.16% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.77% :     0.000238s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019139    76
 0.30% :     0.000057s :      1: add_recomputation
 0.34% :     0.000065s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.44% :     0.000275s :      1: bootstrap
 0.15% :     0.000029s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.13% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.16% :     0.005006s :      1: jit_opt_a
 0.97% :     0.000185s :      1: jit_opt_after_cconv
 0.27% :     0.000052s :      1: jit_opt_b
 2.20% :     0.000421s :      1: loop_unroll
 2.61% :     0.000499s :      1: mutable_eliminate
 4.28% :     0.000820s :     26: opt.transform.jit_opt_a
 0.25% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.43% :     0.000466s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000035s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 1.67% :     0.000321s :      1: renormalize.infer
 1.41% :     0.000270s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000037s :      1: rewriter_after_opt_a
 0.37% :     0.000071s :      1: rewriter_before_opt_a
 0.40% :     0.000077s :      1: symbol_engine_optimizer
31.05% :     0.005942s :      1: task_emit
21.67% :     0.004148s :      1: type_inference
 0.27% :     0.000051s :      1: validate


TotalTime = 0.0268745, [33]
[bootstrap]: 0.00043741
[type_inference]: 0.00619495
[event_method]: 1.768e-05
[auto_monad]: 6.722e-05
[graph_reusing]: 5.42001e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 3.978e-05
[rewriter_before_opt_a]: 7.157e-05
[expand_dump_flag]: 3.18e-06
[jit_opt_a]: 0.00947524, [2]
    [Cycle 1]: 0.00405826, [27]
        [switch_simplify]: 6.787e-05
        [loop_unroll]: 3.151e-05
        [a_1]: 0.00067199
        [with_stream_mark]: 2.221e-05
        [recompute_prepare]: 8.69998e-06
        [updatestate_depend_eliminate]: 3.31001e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 1.92001e-06
        [specialize_transform]: 6.75998e-06
        [updatestate_useless_node_eliminater]: 5.48002e-06
        [accelerated_algorithm]: 5.81e-06
        [meta_shard_fg_expand]: 2.57001e-06
        [get_grad_eliminate_]: 5.89999e-06
        [merge_forward]: 3.47002e-06
        [cell_reuse_recompute_pass]: 1.14998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.614e-05
        [j_node_and_user_rematch]: 9.62999e-06
        [meta_fg_expand]: 2.64001e-06
        [replace_old_param]: 9.64e-06
        [inline_without_move]: 5.39e-06
        [renormalize]: 0.00288529
        [add_forward_monad_depend]: 1.008e-05
        [auto_monad_grad]: 2.68998e-06
        [auto_monad_eliminator]: 1.867e-05
        [cse]: 2.551e-05
        [replace_applicator]: 2.732e-05
    [Cycle 2]: 0.00041005, [27]
        [switch_simplify]: 6.69001e-06
        [loop_unroll]: 5.94999e-06
        [a_1]: 0.00010858
        [with_stream_mark]: 1.646e-05
        [recompute_prepare]: 6.61e-06
        [updatestate_depend_eliminate]: 3.01001e-06
        [updatestate_assign_eliminate]: 2.52001e-06
        [updatestate_loads_eliminate]: 2.69001e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 5.48002e-06
        [updatestate_useless_node_eliminater]: 5.49e-06
        [accelerated_algorithm]: 5.81e-06
        [meta_shard_fg_expand]: 1.71e-06
        [get_grad_eliminate_]: 4.88001e-06
        [merge_forward]: 4.08001e-06
        [cell_reuse_recompute_pass]: 4.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.778e-05
        [j_node_and_user_rematch]: 9.19e-06
        [meta_fg_expand]: 2.19001e-06
        [replace_old_param]: 9.61e-06
        [inline_without_move]: 4.82e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.51002e-06
        [auto_monad_grad]: 9.79984e-07
        [auto_monad_eliminator]: 5.54e-06
        [cse]: 1.19e-05
        [replace_applicator]: 4.98001e-06
[py_interpret_to_execute_after_opt_a]: 2.033e-05
[rewriter_after_opt_a]: 3.709e-05
[convert_after_rewriter]: 7.88001e-06
[order_py_execute_after_rewriter]: 5.27001e-06
[mutable_eliminate]: 0.00085797
[jit_opt_b]: 6.21e-05, [1]
    [Cycle 1]: 5.335e-05, [2]
        [frontend_op_eliminate]: 2.384e-05
        [inline_after_opt_a]: 1.637e-05
[cconv]: 3.066e-05
[loop_unroll]: 0.00046594
[jit_opt_after_cconv]: 0.00015422, [1]
    [Cycle 1]: 0.0001474, [11]
        [c_1]: 2.217e-05
        [parameter_eliminate]: 3.07002e-06
        [updatestate_depend_eliminate]: 6.78e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 2.13002e-06
        [cse]: 2.199e-05
        [call_graph_tuple_transform]: 2.196e-05
        [tuple_list_get_item_eliminator]: 5.81e-06
        [none_parameter_eliminate]: 1.72001e-06
        [renormalize]: 8.50006e-07
        [switch_simplify]: 5.46e-06
[remove_dup_value]: 1.67e-05
[partial_unused_args_eliminate]: 2.76e-06
[environ_conv]: 5.38002e-06
[add_recomputation]: 5.374e-05
[cse_after_recomputation]: 2.236e-05, [1]
    [Cycle 1]: 1.628e-05, [1]
        [cse]: 9.74999e-06
[auto_monad_reorder]: 1.692e-05
[get_jit_bprop_graph]: 2.12999e-06
[rewriter_after_jit_bprop_graph]: 7.41999e-06
[opt_after_jit_grad]: 0.00048697
[symbol_engine_optimizer]: 8.281e-05, [1]
    [Cycle 1]: 7.5e-05, [6]
        [build]: 4.97999e-06
        [elim_shapecalc]: 8.26002e-06
        [elim_not_effective]: 1.468e-05
        [opt_reshape]: 6.58e-06
        [fold_const_symbol]: 1.036e-05
        [renormalize]: 5.10016e-07
[validate]: 4.472e-05
[backend_pass]: 1.02998e-06
[task_emit]: 0.00793554
[execute]: 1.051e-05
Sums
    bootstrap                            :     0.000437s :  2.07%
    type_inference                       :     0.006195s : 29.29%
    event_method                         :     0.000018s :  0.08%
    auto_monad                           :     0.000067s :  0.32%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000040s :  0.19%
    rewriter_before_opt_a                :     0.000072s :  0.34%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000075s :  0.35%
    jit_opt_a.loop_unroll                :     0.000037s :  0.18%
    jit_opt_a.a_1                        :     0.000781s :  3.69%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.002885s : 13.64%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.11%
    jit_opt_a.cse                        :     0.000037s :  0.18%
    jit_opt_a.replace_applicator         :     0.000032s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000020s :  0.10%
    rewriter_after_opt_a                 :     0.000037s :  0.18%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000858s :  4.06%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.000031s :  0.14%
    loop_unroll                          :     0.000466s :  2.20%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000017s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000054s :  0.25%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000487s :  2.30%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.21%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007936s : 37.52%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000267    23
 0.98% :     0.000003s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.38% :     0.000006s :      3: substitution.graph_param_transform
85.13% :     0.000227s :      4: substitution.inline
 1.43% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.17% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.26% :     0.000006s :      2: substitution.replace_old_param
 5.15% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006121     2
84.63% :     0.005180s :      1: type_inference.infer
15.37% :     0.000940s :      1: type_inference.specialize
------[replace.]   0.000081     6
64.85% :     0.000052s :      4: replace.inline
35.15% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000237     6
94.70% :     0.000224s :      4: match.inline
 5.30% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   691
 1.65% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.20% :     0.000002s :     10: predicate.addn_zero_filter
 2.27% :     0.000003s :     10: predicate.arithmetic_simplify
 1.62% :     0.000002s :     10: predicate.cast_eliminate
 0.67% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.16% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     10: predicate.environ_get_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.96% :     0.000008s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      6: predicate.less_batch_normalization
 1.62% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.14% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.15% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.38% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 1.56% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 1.95% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.83% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.97% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     10: predicate.same_eliminate
 0.60% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.64% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.21% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.11% :     0.000010s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.00% :     0.000001s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.11% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.22% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.98% :     0.000003s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000779    12
42.62% :     0.000332s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.66% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.72% :     0.000395s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.030894    76
 0.18% :     0.000057s :      1: add_recomputation
 0.23% :     0.000071s :      1: auto_monad
 0.06% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.47% :     0.000455s :      1: bootstrap
 0.11% :     0.000034s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.07% :     0.000021s :      1: event_method
 0.06% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
30.68% :     0.009480s :      1: jit_opt_a
 0.51% :     0.000157s :      1: jit_opt_after_cconv
 0.21% :     0.000065s :      1: jit_opt_b
 1.54% :     0.000475s :      1: loop_unroll
 2.81% :     0.000867s :      1: mutable_eliminate
 3.34% :     0.001032s :     26: opt.transform.jit_opt_a
 0.17% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000033s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.60% :     0.000495s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000043s :      1: py_interpret_to_execute
 0.07% :     0.000023s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000019s :      1: remove_dup_value
 7.43% :     0.002297s :      1: renormalize.infer
 1.85% :     0.000573s :      1: renormalize.specialize
 0.03% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000040s :      1: rewriter_after_opt_a
 0.24% :     0.000074s :      1: rewriter_before_opt_a
 0.28% :     0.000086s :      1: symbol_engine_optimizer
25.76% :     0.007958s :      1: task_emit
20.10% :     0.006210s :      1: type_inference
 0.22% :     0.000067s :      1: validate


TotalTime = 0.0201981, [33]
[bootstrap]: 0.00035497
[type_inference]: 0.00457855
[event_method]: 1.766e-05
[auto_monad]: 0.00015058
[graph_reusing]: 6.09001e-06
[pre_auto_parallel]: 2.72001e-06
[py_interpret_to_execute]: 3.307e-05
[rewriter_before_opt_a]: 7.211e-05
[expand_dump_flag]: 2.83e-06
[jit_opt_a]: 0.00566066, [2]
    [Cycle 1]: 0.0016014, [27]
        [switch_simplify]: 5.485e-05
        [loop_unroll]: 2.836e-05
        [a_1]: 0.00056407
        [with_stream_mark]: 2e-05
        [recompute_prepare]: 7.51999e-06
        [updatestate_depend_eliminate]: 4.48999e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 3.13e-06
        [parameter_eliminate]: 2.01e-06
        [specialize_transform]: 6.17999e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.42999e-06
        [meta_shard_fg_expand]: 2.02001e-06
        [get_grad_eliminate_]: 5.05001e-06
        [merge_forward]: 3.58e-06
        [cell_reuse_recompute_pass]: 1.51002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.606e-05
        [j_node_and_user_rematch]: 9.43002e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 8.55001e-06
        [inline_without_move]: 5.59e-06
        [renormalize]: 0.00059989
        [add_forward_monad_depend]: 4.95999e-06
        [auto_monad_grad]: 2.60002e-06
        [auto_monad_eliminator]: 1.458e-05
        [cse]: 2.855e-05
        [replace_applicator]: 1.188e-05
    [Cycle 2]: 0.00035372, [27]
        [switch_simplify]: 5.40999e-06
        [loop_unroll]: 4.72e-06
        [a_1]: 0.0001078
        [with_stream_mark]: 1.048e-05
        [recompute_prepare]: 5.56e-06
        [updatestate_depend_eliminate]: 3.23e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 5.29e-06
        [updatestate_useless_node_eliminater]: 5.15999e-06
        [accelerated_algorithm]: 5.62001e-06
        [meta_shard_fg_expand]: 1.44998e-06
        [get_grad_eliminate_]: 4.80999e-06
        [merge_forward]: 2.78e-06
        [cell_reuse_recompute_pass]: 1.76998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.471e-05
        [j_node_and_user_rematch]: 8.43999e-06
        [meta_fg_expand]: 1.66998e-06
        [replace_old_param]: 7.33e-06
        [inline_without_move]: 5.22999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 5.74e-06
        [cse]: 1.157e-05
        [replace_applicator]: 5.32001e-06
[py_interpret_to_execute_after_opt_a]: 1.26e-05
[rewriter_after_opt_a]: 4.97e-05
[convert_after_rewriter]: 7.97998e-06
[order_py_execute_after_rewriter]: 5.47999e-06
[mutable_eliminate]: 0.00060899
[jit_opt_b]: 5.109e-05, [1]
    [Cycle 1]: 4.339e-05, [2]
        [frontend_op_eliminate]: 1.654e-05
        [inline_after_opt_a]: 1.482e-05
[cconv]: 2.7e-05
[loop_unroll]: 0.00044704
[jit_opt_after_cconv]: 0.00015161, [1]
    [Cycle 1]: 0.0001442, [11]
        [c_1]: 2.262e-05
        [parameter_eliminate]: 3.21001e-06
        [updatestate_depend_eliminate]: 5.89999e-06
        [updatestate_assign_eliminate]: 2.54001e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [cse]: 2.178e-05
        [call_graph_tuple_transform]: 2.058e-05
        [tuple_list_get_item_eliminator]: 5.40999e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 5.3001e-07
        [switch_simplify]: 5.72001e-06
[remove_dup_value]: 1.429e-05
[partial_unused_args_eliminate]: 2.34999e-06
[environ_conv]: 7.903e-05
[add_recomputation]: 5.822e-05
[cse_after_recomputation]: 2.321e-05, [1]
    [Cycle 1]: 1.7e-05, [1]
        [cse]: 1.025e-05
[auto_monad_reorder]: 1.503e-05
[get_jit_bprop_graph]: 1.95001e-06
[rewriter_after_jit_bprop_graph]: 5.95002e-06
[opt_after_jit_grad]: 0.00050268
[symbol_engine_optimizer]: 8.038e-05, [1]
    [Cycle 1]: 7.335e-05, [6]
        [build]: 4.2e-06
        [elim_shapecalc]: 7.97e-06
        [elim_not_effective]: 1.46e-05
        [opt_reshape]: 6.69001e-06
        [fold_const_symbol]: 8.85001e-06
        [renormalize]: 6.89994e-07
[validate]: 4.426e-05
[backend_pass]: 1.52001e-06
[task_emit]: 0.00689925
[execute]: 8.38999e-06
Sums
    bootstrap                            :     0.000355s :  2.24%
    type_inference                       :     0.004579s : 28.91%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000151s :  0.95%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.21%
    rewriter_before_opt_a                :     0.000072s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000060s :  0.38%
    jit_opt_a.loop_unroll                :     0.000033s :  0.21%
    jit_opt_a.a_1                        :     0.000672s :  4.24%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000600s :  3.79%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000040s :  0.25%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000050s :  0.31%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000609s :  3.84%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000027s :  0.17%
    loop_unroll                          :     0.000447s :  2.82%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000079s :  0.50%
    add_recomputation                    :     0.000058s :  0.37%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000503s :  3.17%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.28%
    backend_pass                         :     0.000002s :  0.01%
    task_emit                            :     0.006899s : 43.56%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000241    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.50% :     0.000006s :      3: substitution.graph_param_transform
85.48% :     0.000206s :      4: substitution.inline
 1.31% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.68% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.64% :     0.000004s :      2: substitution.replace_old_param
 4.82% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004514     2
82.79% :     0.003737s :      1: type_inference.infer
17.21% :     0.000777s :      1: type_inference.specialize
------[replace.]   0.000065     6
66.48% :     0.000043s :      4: replace.inline
33.52% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000213     6
95.09% :     0.000203s :      4: match.inline
 4.91% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.57% :     0.000002s :     10: predicate.addn_zero_filter
 2.04% :     0.000002s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.59% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000001s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.14% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.84% :     0.000007s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.97% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.83% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.17% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.28% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.95% :     0.000002s :      3: predicate.mutable_eliminate
 0.66% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.67% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.38% :     0.000009s :     53: predicate.switch_simplify
 1.47% :     0.000002s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.80% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.45% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.81% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000561    12
49.25% :     0.000276s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.33% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.42% :     0.000243s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021791    76
 0.28% :     0.000062s :      1: add_recomputation
 0.71% :     0.000155s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 1.72% :     0.000375s :      1: bootstrap
 0.14% :     0.000030s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.12% :     0.000026s :      1: cse_after_recomputation
 0.38% :     0.000083s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.99% :     0.005664s :      1: jit_opt_a
 0.71% :     0.000154s :      1: jit_opt_after_cconv
 0.25% :     0.000054s :      1: jit_opt_b
 2.09% :     0.000456s :      1: loop_unroll
 2.84% :     0.000619s :      1: mutable_eliminate
 4.05% :     0.000883s :     26: opt.transform.jit_opt_a
 0.23% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.35% :     0.000512s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000036s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.37% :     0.000299s :      1: renormalize.infer
 1.35% :     0.000293s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000053s :      1: rewriter_after_opt_a
 0.34% :     0.000075s :      1: rewriter_before_opt_a
 0.38% :     0.000083s :      1: symbol_engine_optimizer
31.75% :     0.006917s :      1: task_emit
21.07% :     0.004591s :      1: type_inference
 0.31% :     0.000067s :      1: validate


TotalTime = 0.0186376, [33]
[bootstrap]: 0.00031673
[type_inference]: 0.00480945
[event_method]: 1.716e-05
[auto_monad]: 6.295e-05
[graph_reusing]: 6.01e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 3.586e-05
[rewriter_before_opt_a]: 7.601e-05
[expand_dump_flag]: 3.14001e-06
[jit_opt_a]: 0.00509436, [2]
    [Cycle 1]: 0.00146666, [27]
        [switch_simplify]: 5.612e-05
        [loop_unroll]: 3.001e-05
        [a_1]: 0.00050894
        [with_stream_mark]: 1.547e-05
        [recompute_prepare]: 7.4e-06
        [updatestate_depend_eliminate]: 3.33e-06
        [updatestate_assign_eliminate]: 2.71999e-06
        [updatestate_loads_eliminate]: 2.39001e-06
        [parameter_eliminate]: 1.94999e-06
        [specialize_transform]: 6.19999e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 5.29998e-06
        [meta_shard_fg_expand]: 2.25002e-06
        [get_grad_eliminate_]: 5.44e-06
        [merge_forward]: 3.58999e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.447e-05
        [j_node_and_user_rematch]: 8.99e-06
        [meta_fg_expand]: 2.81999e-06
        [replace_old_param]: 8.10999e-06
        [inline_without_move]: 5.07999e-06
        [renormalize]: 0.00054445
        [add_forward_monad_depend]: 5.17e-06
        [auto_monad_grad]: 2.14e-06
        [auto_monad_eliminator]: 1.337e-05
        [cse]: 2.84e-05
        [replace_applicator]: 1.148e-05
    [Cycle 2]: 0.00031978, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 5.07999e-06
        [a_1]: 8.94e-05
        [with_stream_mark]: 8.59e-06
        [recompute_prepare]: 5.05001e-06
        [updatestate_depend_eliminate]: 2.32999e-06
        [updatestate_assign_eliminate]: 2.00002e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.24998e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.65001e-06
        [merge_forward]: 2.54001e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.2e-05
        [j_node_and_user_rematch]: 7.66001e-06
        [meta_fg_expand]: 1.65001e-06
        [replace_old_param]: 6.99001e-06
        [inline_without_move]: 4.63001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.53002e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 5.19998e-06
        [cse]: 1.033e-05
        [replace_applicator]: 5.96e-06
[py_interpret_to_execute_after_opt_a]: 1.009e-05
[rewriter_after_opt_a]: 3.359e-05
[convert_after_rewriter]: 7.08998e-06
[order_py_execute_after_rewriter]: 5.52999e-06
[mutable_eliminate]: 0.00051201
[jit_opt_b]: 5.048e-05, [1]
    [Cycle 1]: 4.311e-05, [2]
        [frontend_op_eliminate]: 1.63e-05
        [inline_after_opt_a]: 1.428e-05
[cconv]: 2.473e-05
[loop_unroll]: 0.0004177
[jit_opt_after_cconv]: 0.00013767, [1]
    [Cycle 1]: 0.00013125, [11]
        [c_1]: 2.066e-05
        [parameter_eliminate]: 2.53e-06
        [updatestate_depend_eliminate]: 5.36002e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.643e-05
        [call_graph_tuple_transform]: 1.874e-05
        [tuple_list_get_item_eliminator]: 5.60001e-06
        [none_parameter_eliminate]: 1.79e-06
        [renormalize]: 7.10017e-07
        [switch_simplify]: 5.61e-06
[remove_dup_value]: 1.274e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 4.58999e-06
[add_recomputation]: 4.814e-05
[cse_after_recomputation]: 2e-05, [1]
    [Cycle 1]: 1.454e-05, [1]
        [cse]: 8.69003e-06
[auto_monad_reorder]: 1.486e-05
[get_jit_bprop_graph]: 1.72001e-06
[rewriter_after_jit_bprop_graph]: 4.89e-06
[opt_after_jit_grad]: 0.00047754
[symbol_engine_optimizer]: 7.325e-05, [1]
    [Cycle 1]: 6.623e-05, [6]
        [build]: 3.21999e-06
        [elim_shapecalc]: 7.82e-06
        [elim_not_effective]: 1.244e-05
        [opt_reshape]: 5.35999e-06
        [fold_const_symbol]: 8.93002e-06
        [renormalize]: 5.60016e-07
[validate]: 3.28e-05
[backend_pass]: 8.59989e-07
[task_emit]: 0.00611674
[execute]: 7.09001e-06
Sums
    bootstrap                            :     0.000317s :  2.15%
    type_inference                       :     0.004809s : 32.65%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000063s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.24%
    rewriter_before_opt_a                :     0.000076s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.42%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000598s :  4.06%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000545s :  3.70%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000039s :  0.26%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000512s :  3.48%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000418s :  2.84%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000048s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000478s :  3.24%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000033s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006117s : 41.53%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000184    23
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 3.00% :     0.000006s :      3: substitution.graph_param_transform
82.61% :     0.000152s :      4: substitution.inline
 1.72% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.56% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.02% :     0.000004s :      2: substitution.replace_old_param
 6.31% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004747     2
83.17% :     0.003948s :      1: type_inference.infer
16.83% :     0.000799s :      1: type_inference.specialize
------[replace.]   0.000065     6
66.76% :     0.000043s :      4: replace.inline
33.24% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000160     6
93.48% :     0.000150s :      4: match.inline
 6.52% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.25% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 2.38% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.59% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.08% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.59% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.35% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.76% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.52% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.02% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.60% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.00% :     0.000009s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.95% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.75% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000548    12
51.50% :     0.000282s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.97% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.53% :     0.000238s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020086    76
 0.25% :     0.000051s :      1: add_recomputation
 0.33% :     0.000066s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.66% :     0.000334s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.38% :     0.005097s :      1: jit_opt_a
 0.70% :     0.000140s :      1: jit_opt_after_cconv
 0.26% :     0.000053s :      1: jit_opt_b
 2.12% :     0.000426s :      1: loop_unroll
 2.59% :     0.000521s :      1: mutable_eliminate
 4.01% :     0.000804s :     26: opt.transform.jit_opt_a
 0.23% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.42% :     0.000486s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000039s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.35% :     0.000272s :      1: renormalize.infer
 1.32% :     0.000266s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000037s :      1: rewriter_after_opt_a
 0.39% :     0.000079s :      1: rewriter_before_opt_a
 0.38% :     0.000076s :      1: symbol_engine_optimizer
30.50% :     0.006127s :      1: task_emit
24.02% :     0.004824s :      1: type_inference
 0.25% :     0.000051s :      1: validate


TotalTime = 0.0138489, [30]
[bootstrap]: 0.00025479
[type_inference]: 0.00598049
[event_method]: 2.052e-05
[auto_monad]: 0.00011564
[graph_reusing]: 6.09999e-06
[pre_auto_parallel]: 2.57001e-06
[py_interpret_to_execute]: 4.478e-05
[rewriter_before_opt_a]: 8.061e-05
[expand_dump_flag]: 2.60997e-06
[jit_opt_a]: 0.0053117, [2]
    [Cycle 1]: 0.00192315, [27]
        [switch_simplify]: 0.00010263
        [loop_unroll]: 3.268e-05
        [a_1]: 0.00082853
        [with_stream_mark]: 1.609e-05
        [recompute_prepare]: 7.30998e-06
        [updatestate_depend_eliminate]: 3.38999e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 1.58002e-06
        [specialize_transform]: 5.78997e-06
        [updatestate_useless_node_eliminater]: 4.92999e-06
        [accelerated_algorithm]: 5.42999e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 5.49998e-06
        [merge_forward]: 2.86e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.287e-05
        [j_node_and_user_rematch]: 8.82e-06
        [meta_fg_expand]: 2.27001e-06
        [replace_old_param]: 8.95001e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.00064465
        [add_forward_monad_depend]: 4.85001e-06
        [auto_monad_grad]: 2.37999e-06
        [auto_monad_eliminator]: 1.203e-05
        [cse]: 1.874e-05
        [replace_applicator]: 1.103e-05
    [Cycle 2]: 0.00033012, [27]
        [switch_simplify]: 5.67999e-06
        [loop_unroll]: 4.62e-06
        [a_1]: 8.696e-05
        [with_stream_mark]: 8.73001e-06
        [recompute_prepare]: 4.72998e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 8.80013e-07
        [specialize_transform]: 4.80999e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.30999e-06
        [meta_shard_fg_expand]: 1.00999e-06
        [get_grad_eliminate_]: 4.56002e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.185e-05
        [j_node_and_user_rematch]: 7.66001e-06
        [meta_fg_expand]: 1.47999e-06
        [replace_old_param]: 7.65e-06
        [inline_without_move]: 4.82e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 6.39993e-07
        [auto_monad_eliminator]: 4.81002e-06
        [cse]: 1.002e-05
        [replace_applicator]: 4.84e-06
[py_interpret_to_execute_after_opt_a]: 9.61e-06
[rewriter_after_opt_a]: 3.041e-05
[convert_after_rewriter]: 6.26e-06
[order_py_execute_after_rewriter]: 5.24998e-06
[mutable_eliminate]: 0.00052873
[jit_opt_b]: 5.847e-05, [1]
    [Cycle 1]: 5.128e-05, [2]
        [frontend_op_eliminate]: 2.319e-05
        [inline_after_opt_a]: 1.588e-05
[cconv]: 2.464e-05
[loop_unroll]: 0.00041863
[jit_opt_after_cconv]: 0.0001348, [1]
    [Cycle 1]: 0.0001286, [11]
        [c_1]: 2.083e-05
        [parameter_eliminate]: 2.69001e-06
        [updatestate_depend_eliminate]: 5.66e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 1.626e-05
        [call_graph_tuple_transform]: 1.694e-05
        [tuple_list_get_item_eliminator]: 5.14998e-06
        [none_parameter_eliminate]: 1.36998e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.22999e-06
[remove_dup_value]: 1.107e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 4.27e-06
[add_recomputation]: 4.335e-05
[cse_after_recomputation]: 1.976e-05, [1]
    [Cycle 1]: 1.471e-05, [1]
        [cse]: 9.24998e-06
[auto_monad_reorder]: 1.41e-05
[get_jit_bprop_graph]: 2.15002e-06
[rewriter_after_jit_bprop_graph]: 5.48002e-06
[opt_after_jit_grad]: 0.00045086
[symbol_engine_optimizer]: 7.038e-05, [1]
    [Cycle 1]: 6.453e-05, [6]
        [build]: 3.33e-06
        [elim_shapecalc]: 7.68001e-06
        [elim_not_effective]: 1.213e-05
        [opt_reshape]: 5.72001e-06
        [fold_const_symbol]: 8.42e-06
        [renormalize]: 5.39992e-07
[validate]: 3.299e-05
Sums
    bootstrap                            :     0.000255s :  2.49%
    type_inference                       :     0.005980s : 58.54%
    event_method                         :     0.000021s :  0.20%
    auto_monad                           :     0.000116s :  1.13%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000045s :  0.44%
    rewriter_before_opt_a                :     0.000081s :  0.79%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000108s :  1.06%
    jit_opt_a.loop_unroll                :     0.000037s :  0.37%
    jit_opt_a.a_1                        :     0.000915s :  8.96%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000645s :  6.31%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.16%
    jit_opt_a.cse                        :     0.000029s :  0.28%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.09%
    rewriter_after_opt_a                 :     0.000030s :  0.30%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000529s :  5.18%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.16%
    cconv                                :     0.000025s :  0.24%
    loop_unroll                          :     0.000419s :  4.10%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000011s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000043s :  0.42%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000014s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000451s :  4.41%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000033s :  0.32%


Time group info:
------[substitution.]   0.000290    26
 0.65% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000002s :      2: substitution.fold_const_symbol
 1.59% :     0.000005s :      3: substitution.graph_param_transform
74.57% :     0.000217s :      6: substitution.inline
 1.02% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.44% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.22% :     0.000004s :      2: substitution.replace_old_param
 2.82% :     0.000008s :      1: substitution.switch_simplify
16.12% :     0.000047s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005925     2
83.22% :     0.004931s :      1: type_inference.infer
16.78% :     0.000994s :      1: type_inference.specialize
------[replace.]   0.000167     9
53.90% :     0.000090s :      6: replace.inline
17.94% :     0.000030s :      1: replace.switch_simplify
28.16% :     0.000047s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000250     9
85.39% :     0.000213s :      6: match.inline
 2.96% :     0.000007s :      1: match.switch_simplify
11.65% :     0.000029s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000167   801
 1.12% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.96% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.92% :     0.000002s :     12: predicate.addn_check_dump
 1.06% :     0.000002s :     12: predicate.addn_zero_filter
 1.43% :     0.000002s :     12: predicate.arithmetic_simplify
13.08% :     0.000022s :     12: predicate.cast_eliminate
 0.35% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.91% :     0.000002s :     12: predicate.compare_switch_simplify
 1.11% :     0.000002s :     12: predicate.depend_value_elim
 0.98% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.03% :     0.000002s :     12: predicate.dict_get_item_eliminator
 0.96% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.73% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.22% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.98% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.06% :     0.000002s :     12: predicate.environ_get_add_eliminate
 0.92% :     0.000002s :     12: predicate.environ_get_depend_swap
 0.94% :     0.000002s :     12: predicate.environ_get_eliminate
 0.91% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.17% :     0.000000s :      3: predicate.fold_const_symbol
 0.67% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 3.99% :     0.000007s :     26: predicate.inline
 0.68% :     0.000001s :      6: predicate.inline_without_move
 0.27% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.82% :     0.000001s :      6: predicate.less_batch_normalization
 1.26% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.37% :     0.000002s :     17: predicate.load_eliminater
 0.96% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.18% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.42% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 0.93% :     0.000002s :     12: predicate.merge_addn
 0.92% :     0.000002s :     12: predicate.minmaximum_grad
 1.03% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 1.63% :     0.000003s :     17: predicate.partial_eliminate
 0.97% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.25% :     0.000002s :     12: predicate.reduce_eliminate
 1.13% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.34% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.30% :     0.000002s :     20: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.11% :     0.000002s :     12: predicate.reshape_eliminate
 1.04% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.47% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.95% :     0.000002s :     12: predicate.same_eliminate
 0.36% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.61% :     0.000001s :      6: predicate.special_op_eliminate
 0.74% :     0.000001s :      6: predicate.specialize_transform
 1.07% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.09% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.31% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 9.22% :     0.000015s :     20: predicate.switch_defer_inline
 1.86% :     0.000003s :     20: predicate.switch_layer_defer_inline
 6.33% :     0.000011s :     65: predicate.switch_simplify
 1.13% :     0.000002s :     12: predicate.tile_eliminate
 1.04% :     0.000002s :     12: predicate.transpose_eliminate
 1.12% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 0.97% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 2.46% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.12% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.17% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 5.92% :     0.000010s :     17: predicate.updatestate_pure_node_eliminater
 2.19% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.21% :     0.000002s :     12: predicate.value_based_eliminate
 0.26% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.49% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000703    15
47.24% :     0.000332s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.50% :     0.000088s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.26% :     0.000283s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015731    72
 0.29% :     0.000046s :      1: add_recomputation
 0.76% :     0.000120s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.70% :     0.000268s :      1: bootstrap
 0.17% :     0.000027s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
33.79% :     0.005315s :      1: jit_opt_a
 0.87% :     0.000138s :      1: jit_opt_after_cconv
 0.39% :     0.000061s :      1: jit_opt_b
 2.71% :     0.000426s :      1: loop_unroll
 3.41% :     0.000537s :      1: mutable_eliminate
 7.45% :     0.001172s :     26: opt.transform.jit_opt_a
 0.28% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000032s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.91% :     0.000458s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.30% :     0.000048s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000013s :      1: remove_dup_value
 2.32% :     0.000365s :      1: renormalize.infer
 1.73% :     0.000273s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000034s :      1: rewriter_after_opt_a
 0.53% :     0.000083s :      1: rewriter_before_opt_a
 0.46% :     0.000073s :      1: symbol_engine_optimizer
38.09% :     0.005992s :      1: type_inference


TotalTime = 0.0135446, [30]
[bootstrap]: 0.00024072
[type_inference]: 0.0058277
[event_method]: 1.974e-05
[auto_monad]: 6.937e-05
[graph_reusing]: 4.60999e-06
[pre_auto_parallel]: 2.17999e-06
[py_interpret_to_execute]: 4.201e-05
[rewriter_before_opt_a]: 9.406e-05
[expand_dump_flag]: 2.79001e-06
[jit_opt_a]: 0.00515013, [2]
    [Cycle 1]: 0.00181739, [27]
        [switch_simplify]: 0.00012901
        [loop_unroll]: 3.404e-05
        [a_1]: 0.00064937
        [with_stream_mark]: 1.568e-05
        [recompute_prepare]: 1.932e-05
        [updatestate_depend_eliminate]: 3.90998e-06
        [updatestate_assign_eliminate]: 2.54001e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.35999e-06
        [specialize_transform]: 6.21e-06
        [updatestate_useless_node_eliminater]: 5.56e-06
        [accelerated_algorithm]: 5.50001e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 3.61001e-06
        [cell_reuse_recompute_pass]: 1.09998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.264e-05
        [j_node_and_user_rematch]: 9.75002e-06
        [meta_fg_expand]: 2.02001e-06
        [replace_old_param]: 8.17e-06
        [inline_without_move]: 5.61003e-06
        [renormalize]: 0.0006708
        [add_forward_monad_depend]: 6.34001e-06
        [auto_monad_grad]: 2.32001e-06
        [auto_monad_eliminator]: 1.33e-05
        [cse]: 1.995e-05
        [replace_applicator]: 1.137e-05
    [Cycle 2]: 0.00030387, [27]
        [switch_simplify]: 5.39e-06
        [loop_unroll]: 4.71002e-06
        [a_1]: 8.726e-05
        [with_stream_mark]: 8.32998e-06
        [recompute_prepare]: 4.87e-06
        [updatestate_depend_eliminate]: 2.48e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [parameter_eliminate]: 1.11002e-06
        [specialize_transform]: 4.60001e-06
        [updatestate_useless_node_eliminater]: 4.96997e-06
        [accelerated_algorithm]: 4.58001e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 4.60001e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 1.69998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.213e-05
        [j_node_and_user_rematch]: 7.58999e-06
        [meta_fg_expand]: 1.39998e-06
        [replace_old_param]: 6.59001e-06
        [inline_without_move]: 5.04e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 9.80013e-07
        [auto_monad_grad]: 6.29982e-07
        [auto_monad_eliminator]: 4.89e-06
        [cse]: 9.04e-06
        [replace_applicator]: 4.90999e-06
[py_interpret_to_execute_after_opt_a]: 1.143e-05
[rewriter_after_opt_a]: 3.225e-05
[convert_after_rewriter]: 6.75002e-06
[order_py_execute_after_rewriter]: 4.92e-06
[mutable_eliminate]: 0.00055996
[jit_opt_b]: 5.607e-05, [1]
    [Cycle 1]: 4.877e-05, [2]
        [frontend_op_eliminate]: 2.246e-05
        [inline_after_opt_a]: 1.488e-05
[cconv]: 2.403e-05
[loop_unroll]: 0.00040774
[jit_opt_after_cconv]: 0.0001329, [1]
    [Cycle 1]: 0.00012665, [11]
        [c_1]: 1.905e-05
        [parameter_eliminate]: 2.45002e-06
        [updatestate_depend_eliminate]: 5.27001e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 1.96998e-06
        [cse]: 1.662e-05
        [call_graph_tuple_transform]: 1.723e-05
        [tuple_list_get_item_eliminator]: 5.00001e-06
        [none_parameter_eliminate]: 1.64998e-06
        [renormalize]: 4.90021e-07
        [switch_simplify]: 5.44e-06
[remove_dup_value]: 1.232e-05
[partial_unused_args_eliminate]: 1.94e-06
[environ_conv]: 5.05001e-06
[add_recomputation]: 4.337e-05
[cse_after_recomputation]: 2.012e-05, [1]
    [Cycle 1]: 1.469e-05, [1]
        [cse]: 8.65999e-06
[auto_monad_reorder]: 1.366e-05
[get_jit_bprop_graph]: 2.06e-06
[rewriter_after_jit_bprop_graph]: 6.17999e-06
[opt_after_jit_grad]: 0.00045773
[symbol_engine_optimizer]: 7.63e-05, [1]
    [Cycle 1]: 6.937e-05, [6]
        [build]: 3.36001e-06
        [elim_shapecalc]: 8.51002e-06
        [elim_not_effective]: 1.288e-05
        [opt_reshape]: 5.40001e-06
        [fold_const_symbol]: 8.51002e-06
        [renormalize]: 8.09989e-07
[validate]: 3.442e-05
Sums
    bootstrap                            :     0.000241s :  2.42%
    type_inference                       :     0.005828s : 58.66%
    event_method                         :     0.000020s :  0.20%
    auto_monad                           :     0.000069s :  0.70%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000042s :  0.42%
    rewriter_before_opt_a                :     0.000094s :  0.95%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000134s :  1.35%
    jit_opt_a.loop_unroll                :     0.000039s :  0.39%
    jit_opt_a.a_1                        :     0.000737s :  7.41%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000024s :  0.24%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000671s :  6.75%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.18%
    jit_opt_a.cse                        :     0.000029s :  0.29%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000032s :  0.32%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000560s :  5.64%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.15%
    cconv                                :     0.000024s :  0.24%
    loop_unroll                          :     0.000408s :  4.10%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000012s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000043s :  0.44%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000014s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000458s :  4.61%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000034s :  0.35%


Time group info:
------[substitution.]   0.000254    26
 0.71% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.00% :     0.000005s :      3: substitution.graph_param_transform
84.78% :     0.000215s :      6: substitution.inline
 1.44% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.74% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.14% :     0.000003s :      2: substitution.replace_old_param
 3.05% :     0.000008s :      1: substitution.switch_simplify
 4.67% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005774     2
83.18% :     0.004803s :      1: type_inference.infer
16.82% :     0.000971s :      1: type_inference.specialize
------[replace.]   0.000121     9
50.81% :     0.000061s :      6: replace.inline
25.46% :     0.000031s :      1: replace.switch_simplify
23.73% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000230     9
92.31% :     0.000212s :      6: match.inline
 3.04% :     0.000007s :      1: match.switch_simplify
 4.66% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000153   801
 1.42% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.96% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     12: predicate.addn_check_dump
 1.17% :     0.000002s :     12: predicate.addn_zero_filter
 1.39% :     0.000002s :     12: predicate.arithmetic_simplify
 1.12% :     0.000002s :     12: predicate.cast_eliminate
 0.31% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     12: predicate.compare_switch_simplify
 1.25% :     0.000002s :     12: predicate.depend_value_elim
 0.97% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.15% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.14% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.65% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.51% :     0.000001s :      3: predicate.elim_not_effective
 0.79% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.06% :     0.000002s :     12: predicate.environ_add_const_eliminate
 0.98% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.03% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.06% :     0.000002s :     12: predicate.environ_get_eliminate
 1.00% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.68% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 4.27% :     0.000007s :     26: predicate.inline
 0.78% :     0.000001s :      6: predicate.inline_without_move
 0.29% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.75% :     0.000001s :      6: predicate.less_batch_normalization
 1.28% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.47% :     0.000002s :     17: predicate.load_eliminater
 0.91% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.65% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.47% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000002s :     12: predicate.merge_addn
 1.05% :     0.000002s :     12: predicate.minmaximum_grad
 1.19% :     0.000002s :      3: predicate.mutable_eliminate
 0.36% :     0.000001s :      3: predicate.opt_reshape
 1.76% :     0.000003s :     17: predicate.partial_eliminate
 1.13% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.34% :     0.000002s :     12: predicate.reduce_eliminate
 1.29% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.37% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.39% :     0.000002s :     20: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.12% :     0.000002s :     12: predicate.reshape_eliminate
 1.08% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.58% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000002s :     12: predicate.same_eliminate
 0.38% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.60% :     0.000001s :      6: predicate.special_op_eliminate
 0.80% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.06% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.32% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     20: predicate.switch_defer_inline
 2.02% :     0.000003s :     20: predicate.switch_layer_defer_inline
24.40% :     0.000037s :     65: predicate.switch_simplify
 1.03% :     0.000002s :     12: predicate.tile_eliminate
 1.10% :     0.000002s :     12: predicate.transpose_eliminate
 1.24% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.05% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 2.61% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.19% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.23% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.39% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.44% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.38% :     0.000002s :     12: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000711    15
48.45% :     0.000344s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.50% :     0.000089s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.05% :     0.000278s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015312    72
 0.30% :     0.000046s :      1: add_recomputation
 0.47% :     0.000073s :      1: auto_monad
 0.11% :     0.000016s :      1: auto_monad_reorder
 1.66% :     0.000254s :      1: bootstrap
 0.18% :     0.000027s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
33.66% :     0.005154s :      1: jit_opt_a
 0.89% :     0.000136s :      1: jit_opt_after_cconv
 0.38% :     0.000059s :      1: jit_opt_b
 2.71% :     0.000415s :      1: loop_unroll
 3.71% :     0.000568s :      1: mutable_eliminate
 6.76% :     0.001035s :     26: opt.transform.jit_opt_a
 0.28% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000031s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.21% :     0.000492s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.29% :     0.000044s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000015s :      1: remove_dup_value
 2.56% :     0.000391s :      1: renormalize.infer
 1.77% :     0.000272s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000035s :      1: rewriter_after_opt_a
 0.64% :     0.000098s :      1: rewriter_before_opt_a
 0.51% :     0.000079s :      1: symbol_engine_optimizer
38.13% :     0.005839s :      1: type_inference


TotalTime = 0.0179048, [33]
[bootstrap]: 0.00027235
[type_inference]: 0.00429914
[event_method]: 1.755e-05
[auto_monad]: 6.614e-05
[graph_reusing]: 5.64998e-06
[pre_auto_parallel]: 2.54001e-06
[py_interpret_to_execute]: 3.284e-05
[rewriter_before_opt_a]: 7.349e-05
[expand_dump_flag]: 3.21001e-06
[jit_opt_a]: 0.00493715, [2]
    [Cycle 1]: 0.00145072, [27]
        [switch_simplify]: 5.313e-05
        [loop_unroll]: 3.062e-05
        [a_1]: 0.00050996
        [with_stream_mark]: 1.675e-05
        [recompute_prepare]: 7.04001e-06
        [updatestate_depend_eliminate]: 3.18e-06
        [updatestate_assign_eliminate]: 3.23998e-06
        [updatestate_loads_eliminate]: 3.03e-06
        [parameter_eliminate]: 1.82999e-06
        [specialize_transform]: 5.62999e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 5.73002e-06
        [meta_shard_fg_expand]: 2.09e-06
        [get_grad_eliminate_]: 5.14e-06
        [merge_forward]: 3.75e-06
        [cell_reuse_recompute_pass]: 1.62001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.376e-05
        [j_node_and_user_rematch]: 9.19e-06
        [meta_fg_expand]: 2.41e-06
        [replace_old_param]: 8.17e-06
        [inline_without_move]: 5.09e-06
        [renormalize]: 0.00052913
        [add_forward_monad_depend]: 5.46e-06
        [auto_monad_grad]: 2.19001e-06
        [auto_monad_eliminator]: 1.336e-05
        [cse]: 2.681e-05
        [replace_applicator]: 1.108e-05
    [Cycle 2]: 0.00032155, [27]
        [switch_simplify]: 5.80002e-06
        [loop_unroll]: 4.84e-06
        [a_1]: 9.152e-05
        [with_stream_mark]: 8.43999e-06
        [recompute_prepare]: 5.19e-06
        [updatestate_depend_eliminate]: 2.55002e-06
        [updatestate_assign_eliminate]: 2.13002e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 1.02998e-06
        [specialize_transform]: 4.92999e-06
        [updatestate_useless_node_eliminater]: 5.20001e-06
        [accelerated_algorithm]: 5.02999e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 4.61002e-06
        [merge_forward]: 2.55002e-06
        [cell_reuse_recompute_pass]: 1.52001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.303e-05
        [j_node_and_user_rematch]: 7.99997e-06
        [meta_fg_expand]: 1.56998e-06
        [replace_old_param]: 7.23e-06
        [inline_without_move]: 4.74e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 6.40022e-07
        [auto_monad_eliminator]: 5.60001e-06
        [cse]: 1.011e-05
        [replace_applicator]: 5.15999e-06
[py_interpret_to_execute_after_opt_a]: 2.128e-05
[rewriter_after_opt_a]: 3.338e-05
[convert_after_rewriter]: 7.08998e-06
[order_py_execute_after_rewriter]: 5.24e-06
[mutable_eliminate]: 0.00050166
[jit_opt_b]: 5.414e-05, [1]
    [Cycle 1]: 4.684e-05, [2]
        [frontend_op_eliminate]: 1.759e-05
        [inline_after_opt_a]: 1.71e-05
[cconv]: 2.521e-05
[loop_unroll]: 0.00042304
[jit_opt_after_cconv]: 0.00013849, [1]
    [Cycle 1]: 0.00013187, [11]
        [c_1]: 2.077e-05
        [parameter_eliminate]: 2.41998e-06
        [updatestate_depend_eliminate]: 5.02e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.671e-05
        [call_graph_tuple_transform]: 1.883e-05
        [tuple_list_get_item_eliminator]: 5.51e-06
        [none_parameter_eliminate]: 1.40001e-06
        [renormalize]: 3.50003e-07
        [switch_simplify]: 5.74e-06
[remove_dup_value]: 1.313e-05
[partial_unused_args_eliminate]: 2.31998e-06
[environ_conv]: 4.95999e-06
[add_recomputation]: 4.61e-05
[cse_after_recomputation]: 2.039e-05, [1]
    [Cycle 1]: 1.526e-05, [1]
        [cse]: 8.93002e-06
[auto_monad_reorder]: 1.47e-05
[get_jit_bprop_graph]: 1.55999e-06
[rewriter_after_jit_bprop_graph]: 5.82001e-06
[opt_after_jit_grad]: 0.00045754
[symbol_engine_optimizer]: 7.217e-05, [1]
    [Cycle 1]: 6.539e-05, [6]
        [build]: 3.27002e-06
        [elim_shapecalc]: 7.43e-06
        [elim_not_effective]: 1.245e-05
        [opt_reshape]: 5.37999e-06
        [fold_const_symbol]: 8.70001e-06
        [renormalize]: 4.50003e-07
[validate]: 3.266e-05
[backend_pass]: 7.80012e-07
[task_emit]: 0.00611488
[execute]: 7.88001e-06
Sums
    bootstrap                            :     0.000272s :  1.93%
    type_inference                       :     0.004299s : 30.39%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000066s :  0.47%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.23%
    rewriter_before_opt_a                :     0.000073s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.42%
    jit_opt_a.loop_unroll                :     0.000035s :  0.25%
    jit_opt_a.a_1                        :     0.000601s :  4.25%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000529s :  3.74%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000037s :  0.26%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000021s :  0.15%
    rewriter_after_opt_a                 :     0.000033s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000502s :  3.55%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.12%
    cconv                                :     0.000025s :  0.18%
    loop_unroll                          :     0.000423s :  2.99%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000046s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000458s :  3.23%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006115s : 43.23%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000184    23
 1.10% :     0.000002s :      2: substitution.elim_not_effective
 0.80% :     0.000001s :      2: substitution.fold_const_symbol
 3.06% :     0.000006s :      3: substitution.graph_param_transform
82.97% :     0.000153s :      4: substitution.inline
 1.71% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.26% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.83% :     0.000003s :      2: substitution.replace_old_param
 6.27% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004238     2
81.69% :     0.003462s :      1: type_inference.infer
18.31% :     0.000776s :      1: type_inference.specialize
------[replace.]   0.000065     6
67.07% :     0.000044s :      4: replace.inline
32.93% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000161     6
93.52% :     0.000150s :      4: match.inline
 6.48% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.29% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000002s :     10: predicate.addn_zero_filter
 2.14% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.55% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000001s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.15% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.38% :     0.000002s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.37% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.55% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.28% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.16% :     0.000001s :     10: predicate.minmaximum_grad
 1.51% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000000s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.26% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.93% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.27% :     0.000002s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.57% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.17% :     0.000010s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.51% :     0.000002s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.29% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.84% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000529    12
50.82% :     0.000269s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.17% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.01% :     0.000233s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019347    76
 0.25% :     0.000049s :      1: add_recomputation
 0.36% :     0.000070s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.48% :     0.000286s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.53% :     0.004940s :      1: jit_opt_a
 0.73% :     0.000141s :      1: jit_opt_after_cconv
 0.30% :     0.000057s :      1: jit_opt_b
 2.24% :     0.000432s :      1: loop_unroll
 2.64% :     0.000510s :      1: mutable_eliminate
 4.17% :     0.000807s :     26: opt.transform.jit_opt_a
 0.24% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000027s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.41% :     0.000466s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000036s :      1: py_interpret_to_execute
 0.12% :     0.000024s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.37% :     0.000265s :      1: renormalize.infer
 1.33% :     0.000257s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000036s :      1: rewriter_after_opt_a
 0.39% :     0.000076s :      1: rewriter_before_opt_a
 0.39% :     0.000075s :      1: symbol_engine_optimizer
31.67% :     0.006128s :      1: task_emit
22.28% :     0.004310s :      1: type_inference
 0.27% :     0.000052s :      1: validate


TotalTime = 0.0413997, [33]
[bootstrap]: 0.0004132
[type_inference]: 0.0054572
[event_method]: 1.895e-05
[auto_monad]: 7.078e-05
[graph_reusing]: 6.50997e-06
[pre_auto_parallel]: 2.70002e-06
[py_interpret_to_execute]: 4.553e-05
[rewriter_before_opt_a]: 7.582e-05
[expand_dump_flag]: 3.53999e-06
[jit_opt_a]: 0.0248682, [2]
    [Cycle 1]: 0.00344641, [27]
        [switch_simplify]: 7.698e-05
        [loop_unroll]: 3.304e-05
        [a_1]: 0.00070214
        [with_stream_mark]: 8.072e-05
        [recompute_prepare]: 1.161e-05
        [updatestate_depend_eliminate]: 5.12e-06
        [updatestate_assign_eliminate]: 4.1e-06
        [updatestate_loads_eliminate]: 2.86e-06
        [parameter_eliminate]: 2.51e-06
        [specialize_transform]: 7.34002e-06
        [updatestate_useless_node_eliminater]: 5.31002e-06
        [accelerated_algorithm]: 6.14001e-06
        [meta_shard_fg_expand]: 2.34001e-06
        [get_grad_eliminate_]: 5.94e-06
        [merge_forward]: 3.76001e-06
        [cell_reuse_recompute_pass]: 1.62001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.698e-05
        [j_node_and_user_rematch]: 1.038e-05
        [meta_fg_expand]: 2.88e-06
        [replace_old_param]: 1.03e-05
        [inline_without_move]: 5.50001e-06
        [renormalize]: 0.00212482
        [add_forward_monad_depend]: 9.00001e-06
        [auto_monad_grad]: 2.93e-06
        [auto_monad_eliminator]: 2.058e-05
        [cse]: 3.446e-05
        [replace_applicator]: 2.352e-05
    [Cycle 2]: 0.00042307, [27]
        [switch_simplify]: 6.87002e-06
        [loop_unroll]: 5.35999e-06
        [a_1]: 0.00011045
        [with_stream_mark]: 1.767e-05
        [recompute_prepare]: 6.44999e-06
        [updatestate_depend_eliminate]: 3.08998e-06
        [updatestate_assign_eliminate]: 2.77002e-06
        [updatestate_loads_eliminate]: 2.51998e-06
        [parameter_eliminate]: 1.52999e-06
        [specialize_transform]: 5.70001e-06
        [updatestate_useless_node_eliminater]: 5.54998e-06
        [accelerated_algorithm]: 6.96001e-06
        [meta_shard_fg_expand]: 1.96998e-06
        [get_grad_eliminate_]: 4.79e-06
        [merge_forward]: 3.23e-06
        [cell_reuse_recompute_pass]: 4.23999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.837e-05
        [j_node_and_user_rematch]: 9.53997e-06
        [meta_fg_expand]: 2.32999e-06
        [replace_old_param]: 1.08e-05
        [inline_without_move]: 5.14998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.94999e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 6.61e-06
        [cse]: 1.607e-05
        [replace_applicator]: 7.63001e-06
[py_interpret_to_execute_after_opt_a]: 1.62e-05
[rewriter_after_opt_a]: 5.043e-05
[convert_after_rewriter]: 8.18001e-06
[order_py_execute_after_rewriter]: 5.70001e-06
[mutable_eliminate]: 0.00093021
[jit_opt_b]: 6.486e-05, [1]
    [Cycle 1]: 5.569e-05, [2]
        [frontend_op_eliminate]: 2.309e-05
        [inline_after_opt_a]: 1.85e-05
[cconv]: 3.465e-05
[loop_unroll]: 0.00054087
[jit_opt_after_cconv]: 0.00016851, [1]
    [Cycle 1]: 0.00016093, [11]
        [c_1]: 2.295e-05
        [parameter_eliminate]: 4.85999e-06
        [updatestate_depend_eliminate]: 9.42999e-06
        [updatestate_assign_eliminate]: 3.07002e-06
        [updatestate_loads_eliminate]: 2.56998e-06
        [cse]: 2.94e-05
        [call_graph_tuple_transform]: 2.355e-05
        [tuple_list_get_item_eliminator]: 5.70001e-06
        [none_parameter_eliminate]: 1.81e-06
        [renormalize]: 1.10999e-06
        [switch_simplify]: 5.76e-06
[remove_dup_value]: 1.691e-05
[partial_unused_args_eliminate]: 2.27001e-06
[environ_conv]: 5.98998e-06
[add_recomputation]: 5.774e-05
[cse_after_recomputation]: 2.416e-05, [1]
    [Cycle 1]: 1.688e-05, [1]
        [cse]: 9.72999e-06
[auto_monad_reorder]: 1.688e-05
[get_jit_bprop_graph]: 2.39999e-06
[rewriter_after_jit_bprop_graph]: 9.46998e-06
[opt_after_jit_grad]: 0.00054315
[symbol_engine_optimizer]: 7.943e-05, [1]
    [Cycle 1]: 7.18e-05, [6]
        [build]: 4.68001e-06
        [elim_shapecalc]: 8.33001e-06
        [elim_not_effective]: 1.494e-05
        [opt_reshape]: 5.81e-06
        [fold_const_symbol]: 9.37001e-06
        [renormalize]: 9.5999e-07
[validate]: 4.461e-05
[backend_pass]: 1.32999e-06
[task_emit]: 0.00755316
[execute]: 1.046e-05
Sums
    bootstrap                            :     0.000413s :  2.10%
    type_inference                       :     0.005457s : 27.80%
    event_method                         :     0.000019s :  0.10%
    auto_monad                           :     0.000071s :  0.36%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000046s :  0.23%
    rewriter_before_opt_a                :     0.000076s :  0.39%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000084s :  0.43%
    jit_opt_a.loop_unroll                :     0.000038s :  0.20%
    jit_opt_a.a_1                        :     0.000813s :  4.14%
    jit_opt_a.with_stream_mark           :     0.000098s :  0.50%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000021s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.002125s : 10.82%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.14%
    jit_opt_a.cse                        :     0.000051s :  0.26%
    jit_opt_a.replace_applicator         :     0.000031s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.08%
    rewriter_after_opt_a                 :     0.000050s :  0.26%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000930s :  4.74%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.09%
    cconv                                :     0.000035s :  0.18%
    loop_unroll                          :     0.000541s :  2.76%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000058s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000543s :  2.77%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007553s : 38.47%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000313    23
 0.62% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000002s :      2: substitution.fold_const_symbol
 2.01% :     0.000006s :      3: substitution.graph_param_transform
85.84% :     0.000269s :      4: substitution.inline
 1.49% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.80% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.40% :     0.000007s :      2: substitution.replace_old_param
 5.21% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005373     2
81.77% :     0.004394s :      1: type_inference.infer
18.23% :     0.000980s :      1: type_inference.specialize
------[replace.]   0.000090     6
67.07% :     0.000060s :      4: replace.inline
32.93% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000280     6
94.79% :     0.000265s :      4: match.inline
 5.21% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000142   691
 1.95% :     0.000003s :     10: predicate.accumulaten_eliminater
 1.38% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.86% :     0.000001s :     10: predicate.addn_check_dump
 1.52% :     0.000002s :     10: predicate.addn_zero_filter
 2.99% :     0.000004s :     10: predicate.arithmetic_simplify
 1.15% :     0.000002s :     10: predicate.cast_eliminate
 0.65% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.13% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     10: predicate.environ_get_eliminate
 1.07% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.10% :     0.000002s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.49% :     0.000008s :     22: predicate.inline
 0.72% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.45% :     0.000002s :      6: predicate.less_batch_normalization
 1.29% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.97% :     0.000003s :     15: predicate.load_eliminater
 1.62% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.81% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.32% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.90% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 2.29% :     0.000003s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 1.87% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.64% :     0.000002s :     10: predicate.reshape_eliminate
 0.96% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.60% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.22% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 1.75% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.35% :     0.000010s :     53: predicate.switch_simplify
 1.20% :     0.000002s :     10: predicate.tile_eliminate
 1.12% :     0.000002s :     10: predicate.transpose_eliminate
 1.51% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.14% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 2.94% :     0.000004s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.35% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.35% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.13% :     0.000003s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000766    12
44.03% :     0.000337s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.45% :     0.000065s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.52% :     0.000364s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.044720    76
 0.14% :     0.000062s :      1: add_recomputation
 0.17% :     0.000074s :      1: auto_monad
 0.04% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.96% :     0.000430s :      1: bootstrap
 0.08% :     0.000038s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.06% :     0.000026s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.05% :     0.000023s :      1: event_method
 0.04% :     0.000016s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
55.62% :     0.024874s :      1: jit_opt_a
 0.38% :     0.000172s :      1: jit_opt_after_cconv
 0.15% :     0.000068s :      1: jit_opt_b
 1.23% :     0.000552s :      1: loop_unroll
 2.11% :     0.000945s :      1: mutable_eliminate
 2.41% :     0.001079s :     26: opt.transform.jit_opt_a
 0.12% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000033s :      4: opt.transform.jit_opt_b
 0.03% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.24% :     0.000554s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000048s :      1: py_interpret_to_execute
 0.04% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000020s :      1: remove_dup_value
 3.54% :     0.001585s :      1: renormalize.infer
 1.17% :     0.000525s :      1: renormalize.specialize
 0.03% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000054s :      1: rewriter_after_opt_a
 0.18% :     0.000079s :      1: rewriter_before_opt_a
 0.18% :     0.000082s :      1: symbol_engine_optimizer
16.94% :     0.007575s :      1: task_emit
12.24% :     0.005474s :      1: type_inference
 0.15% :     0.000068s :      1: validate


TotalTime = 0.0264757, [33]
[bootstrap]: 0.00027889
[type_inference]: 0.00706403
[event_method]: 1.729e-05
[auto_monad]: 6.213e-05
[graph_reusing]: 5.69999e-06
[pre_auto_parallel]: 2.34001e-06
[py_interpret_to_execute]: 2.904e-05
[rewriter_before_opt_a]: 7.009e-05
[expand_dump_flag]: 2.98998e-06
[jit_opt_a]: 0.0069635, [2]
    [Cycle 1]: 0.00155892, [27]
        [switch_simplify]: 5.576e-05
        [loop_unroll]: 2.954e-05
        [a_1]: 0.00052736
        [with_stream_mark]: 2.142e-05
        [recompute_prepare]: 7.59002e-06
        [updatestate_depend_eliminate]: 4.3e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.81e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 5.80002e-06
        [updatestate_useless_node_eliminater]: 4.72998e-06
        [accelerated_algorithm]: 5.10001e-06
        [meta_shard_fg_expand]: 1.90001e-06
        [get_grad_eliminate_]: 5.23002e-06
        [merge_forward]: 4.04002e-06
        [cell_reuse_recompute_pass]: 1.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.377e-05
        [j_node_and_user_rematch]: 8.23001e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 8.43999e-06
        [inline_without_move]: 5.31998e-06
        [renormalize]: 0.00060555
        [add_forward_monad_depend]: 5.25001e-06
        [auto_monad_grad]: 2.22001e-06
        [auto_monad_eliminator]: 1.324e-05
        [cse]: 2.891e-05
        [replace_applicator]: 1.189e-05
    [Cycle 2]: 0.00030412, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 4.84e-06
        [a_1]: 8.6e-05
        [with_stream_mark]: 9.64e-06
        [recompute_prepare]: 4.55999e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 2.08998e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 1.21997e-06
        [specialize_transform]: 4.75001e-06
        [updatestate_useless_node_eliminater]: 4.50999e-06
        [accelerated_algorithm]: 4.48001e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.32e-06
        [merge_forward]: 2.35002e-06
        [cell_reuse_recompute_pass]: 1.89e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.215e-05
        [j_node_and_user_rematch]: 7.53e-06
        [meta_fg_expand]: 1.52999e-06
        [replace_old_param]: 6.44001e-06
        [inline_without_move]: 4.76997e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 9.30013e-07
        [auto_monad_eliminator]: 4.79e-06
        [cse]: 9.20001e-06
        [replace_applicator]: 4.75999e-06
[py_interpret_to_execute_after_opt_a]: 1.006e-05
[rewriter_after_opt_a]: 3.509e-05
[convert_after_rewriter]: 6.86001e-06
[order_py_execute_after_rewriter]: 5.23002e-06
[mutable_eliminate]: 0.00127084
[jit_opt_b]: 4.914e-05, [1]
    [Cycle 1]: 4.13e-05, [2]
        [frontend_op_eliminate]: 1.527e-05
        [inline_after_opt_a]: 1.532e-05
[cconv]: 2.675e-05
[loop_unroll]: 0.00046993
[jit_opt_after_cconv]: 0.00013949, [1]
    [Cycle 1]: 0.00013275, [11]
        [c_1]: 1.926e-05
        [parameter_eliminate]: 2.84001e-06
        [updatestate_depend_eliminate]: 5.99e-06
        [updatestate_assign_eliminate]: 2.19999e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [cse]: 2.038e-05
        [call_graph_tuple_transform]: 1.86e-05
        [tuple_list_get_item_eliminator]: 5.11997e-06
        [none_parameter_eliminate]: 1.69998e-06
        [renormalize]: 7.49977e-07
        [switch_simplify]: 5.26998e-06
[remove_dup_value]: 1.294e-05
[partial_unused_args_eliminate]: 2.12001e-06
[environ_conv]: 4.91002e-06
[add_recomputation]: 5.09e-05
[cse_after_recomputation]: 1.948e-05, [1]
    [Cycle 1]: 1.444e-05, [1]
        [cse]: 8.90001e-06
[auto_monad_reorder]: 1.515e-05
[get_jit_bprop_graph]: 2.37001e-06
[rewriter_after_jit_bprop_graph]: 5.86e-06
[opt_after_jit_grad]: 0.00046958
[symbol_engine_optimizer]: 0.00014258, [1]
    [Cycle 1]: 0.00013666, [6]
        [build]: 3.68e-06
        [elim_shapecalc]: 8.74e-06
        [elim_not_effective]: 1.265e-05
        [opt_reshape]: 5.52001e-06
        [fold_const_symbol]: 8.03001e-06
        [renormalize]: 4.50003e-07
[validate]: 3.757e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.00897775
[execute]: 8.74e-06
Sums
    bootstrap                            :     0.000279s :  1.35%
    type_inference                       :     0.007064s : 34.14%
    event_method                         :     0.000017s :  0.08%
    auto_monad                           :     0.000062s :  0.30%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000029s :  0.14%
    rewriter_before_opt_a                :     0.000070s :  0.34%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000061s :  0.30%
    jit_opt_a.loop_unroll                :     0.000034s :  0.17%
    jit_opt_a.a_1                        :     0.000613s :  2.96%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.07%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000606s :  2.93%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.09%
    jit_opt_a.cse                        :     0.000038s :  0.18%
    jit_opt_a.replace_applicator         :     0.000017s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.05%
    rewriter_after_opt_a                 :     0.000035s :  0.17%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.001271s :  6.14%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.07%
    cconv                                :     0.000027s :  0.13%
    loop_unroll                          :     0.000470s :  2.27%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000051s :  0.25%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000015s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000470s :  2.27%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.18%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008978s : 43.39%
    execute                              :     0.000009s :  0.04%


Time group info:
------[substitution.]   0.000207    23
 0.94% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.48% :     0.000005s :      3: substitution.graph_param_transform
79.46% :     0.000165s :      4: substitution.inline
 1.39% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.58% :     0.000003s :      2: substitution.replace_old_param
11.55% :     0.000024s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006995     2
88.99% :     0.006225s :      1: type_inference.infer
11.01% :     0.000770s :      1: type_inference.specialize
------[replace.]   0.000067     6
66.13% :     0.000044s :      4: replace.inline
33.87% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000185     6
87.69% :     0.000162s :      4: match.inline
12.31% :     0.000023s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.26% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 2.19% :     0.000003s :     10: predicate.arithmetic_simplify
 1.15% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000001s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.73% :     0.000007s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.68% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.60% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.80% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 2.63% :     0.000003s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000002s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.73% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.32% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.02% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.53% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.26% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.53% :     0.000002s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.85% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.64% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000547    12
50.61% :     0.000277s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.78% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.60% :     0.000233s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028009    76
 0.19% :     0.000054s :      1: add_recomputation
 0.23% :     0.000066s :      1: auto_monad
 0.06% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.06% :     0.000298s :      1: bootstrap
 0.11% :     0.000029s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.08% :     0.000022s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
24.87% :     0.006967s :      1: jit_opt_a
 0.51% :     0.000142s :      1: jit_opt_after_cconv
 0.19% :     0.000052s :      1: jit_opt_b
 1.71% :     0.000479s :      1: loop_unroll
 4.58% :     0.001282s :      1: mutable_eliminate
 2.91% :     0.000816s :     26: opt.transform.jit_opt_a
 0.16% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.71% :     0.000478s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.11% :     0.000032s :      1: py_interpret_to_execute
 0.04% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000015s :      1: remove_dup_value
 1.16% :     0.000324s :      1: renormalize.infer
 0.98% :     0.000275s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000038s :      1: rewriter_after_opt_a
 0.26% :     0.000073s :      1: rewriter_before_opt_a
 0.52% :     0.000145s :      1: symbol_engine_optimizer
32.12% :     0.008996s :      1: task_emit
25.28% :     0.007080s :      1: type_inference
 0.22% :     0.000062s :      1: validate


TotalTime = 0.0191505, [33]
[bootstrap]: 0.00029863
[type_inference]: 0.00453605
[event_method]: 1.674e-05
[auto_monad]: 6.167e-05
[graph_reusing]: 5.56e-06
[pre_auto_parallel]: 2.99999e-06
[py_interpret_to_execute]: 3.547e-05
[rewriter_before_opt_a]: 7.449e-05
[expand_dump_flag]: 3.38e-06
[jit_opt_a]: 0.00559253, [2]
    [Cycle 1]: 0.00178077, [27]
        [switch_simplify]: 5.562e-05
        [loop_unroll]: 2.911e-05
        [a_1]: 0.00054449
        [with_stream_mark]: 2.138e-05
        [recompute_prepare]: 7.45e-06
        [updatestate_depend_eliminate]: 3.95998e-06
        [updatestate_assign_eliminate]: 3.16999e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [parameter_eliminate]: 2.02999e-06
        [specialize_transform]: 5.62001e-06
        [updatestate_useless_node_eliminater]: 7.71001e-06
        [accelerated_algorithm]: 6.59001e-06
        [meta_shard_fg_expand]: 2.44999e-06
        [get_grad_eliminate_]: 5.67999e-06
        [merge_forward]: 4.47e-06
        [cell_reuse_recompute_pass]: 1.34998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.497e-05
        [j_node_and_user_rematch]: 9.49999e-06
        [meta_fg_expand]: 2.44999e-06
        [replace_old_param]: 9.72999e-06
        [inline_without_move]: 5.54e-06
        [renormalize]: 0.00072783
        [add_forward_monad_depend]: 5.42001e-06
        [auto_monad_grad]: 2.57001e-06
        [auto_monad_eliminator]: 1.341e-05
        [cse]: 2.947e-05
        [replace_applicator]: 1.116e-05
    [Cycle 2]: 0.00032523, [27]
        [switch_simplify]: 5.48002e-06
        [loop_unroll]: 4.91002e-06
        [a_1]: 9.079e-05
        [with_stream_mark]: 9.19e-06
        [recompute_prepare]: 5.40001e-06
        [updatestate_depend_eliminate]: 2.79001e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.33002e-06
        [specialize_transform]: 4.83001e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.14e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 4.60999e-06
        [merge_forward]: 2.51998e-06
        [cell_reuse_recompute_pass]: 1.82001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.268e-05
        [j_node_and_user_rematch]: 7.98999e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 6.95998e-06
        [inline_without_move]: 4.78001e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.31002e-06
        [auto_monad_grad]: 9.60019e-07
        [auto_monad_eliminator]: 5.72001e-06
        [cse]: 1.031e-05
        [replace_applicator]: 5.20999e-06
[py_interpret_to_execute_after_opt_a]: 1.178e-05
[rewriter_after_opt_a]: 3.762e-05
[convert_after_rewriter]: 7.25998e-06
[order_py_execute_after_rewriter]: 5.20001e-06
[mutable_eliminate]: 0.00059189
[jit_opt_b]: 5.007e-05, [1]
    [Cycle 1]: 4.251e-05, [2]
        [frontend_op_eliminate]: 1.646e-05
        [inline_after_opt_a]: 1.475e-05
[cconv]: 2.515e-05
[loop_unroll]: 0.00044273
[jit_opt_after_cconv]: 0.00014377, [1]
    [Cycle 1]: 0.00013749, [11]
        [c_1]: 2.113e-05
        [parameter_eliminate]: 2.98e-06
        [updatestate_depend_eliminate]: 6.17001e-06
        [updatestate_assign_eliminate]: 2.71e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 2e-05
        [call_graph_tuple_transform]: 1.961e-05
        [tuple_list_get_item_eliminator]: 5.32001e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 9.50007e-07
        [switch_simplify]: 5.25999e-06
[remove_dup_value]: 1.362e-05
[partial_unused_args_eliminate]: 2.22001e-06
[environ_conv]: 5.25001e-06
[add_recomputation]: 6.336e-05
[cse_after_recomputation]: 2.153e-05, [1]
    [Cycle 1]: 1.537e-05, [1]
        [cse]: 9.31e-06
[auto_monad_reorder]: 1.554e-05
[get_jit_bprop_graph]: 1.96e-06
[rewriter_after_jit_bprop_graph]: 5.76e-06
[opt_after_jit_grad]: 0.00047382
[symbol_engine_optimizer]: 7.415e-05, [1]
    [Cycle 1]: 6.753e-05, [6]
        [build]: 4.60001e-06
        [elim_shapecalc]: 7.65e-06
        [elim_not_effective]: 1.254e-05
        [opt_reshape]: 5.74e-06
        [fold_const_symbol]: 9.06002e-06
        [renormalize]: 4.80009e-07
[validate]: 3.642e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00626783
[execute]: 9.81998e-06
Sums
    bootstrap                            :     0.000299s :  2.00%
    type_inference                       :     0.004536s : 30.31%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000062s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.24%
    rewriter_before_opt_a                :     0.000074s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000635s :  4.25%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000013s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000728s :  4.86%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000040s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000592s :  3.96%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000443s :  2.96%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000063s :  0.42%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000474s :  3.17%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006268s : 41.89%
    execute                              :     0.000010s :  0.07%


Time group info:
------[substitution.]   0.000222    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.50% :     0.000006s :      3: substitution.graph_param_transform
84.86% :     0.000188s :      4: substitution.inline
 1.59% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.91% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000005s :      2: substitution.replace_old_param
 5.58% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004465     2
82.05% :     0.003664s :      1: type_inference.infer
17.95% :     0.000801s :      1: type_inference.specialize
------[replace.]   0.000068     6
66.75% :     0.000045s :      4: replace.inline
33.25% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000196     6
94.34% :     0.000185s :      4: match.inline
 5.66% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.27% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.28% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 1.94% :     0.000002s :     10: predicate.arithmetic_simplify
 1.45% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.49% :     0.000002s :     10: predicate.depend_value_elim
 1.29% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.62% :     0.000001s :      3: predicate.graph_param_transform
 5.41% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.60% :     0.000002s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.86% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.92% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.13% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.58% :     0.000009s :     53: predicate.switch_simplify
 1.44% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.83% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.35% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.05% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000595    12
50.02% :     0.000297s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.59% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.39% :     0.000252s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020825    76
 0.32% :     0.000067s :      1: add_recomputation
 0.31% :     0.000065s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.52% :     0.000317s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.08% :     0.000016s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.87% :     0.005596s :      1: jit_opt_a
 0.70% :     0.000147s :      1: jit_opt_after_cconv
 0.25% :     0.000053s :      1: jit_opt_b
 2.17% :     0.000452s :      1: loop_unroll
 2.89% :     0.000601s :      1: mutable_eliminate
 4.07% :     0.000848s :     26: opt.transform.jit_opt_a
 0.23% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.32% :     0.000483s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000039s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.93% :     0.000403s :      1: renormalize.infer
 1.52% :     0.000316s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000041s :      1: rewriter_after_opt_a
 0.37% :     0.000077s :      1: rewriter_before_opt_a
 0.37% :     0.000077s :      1: symbol_engine_optimizer
30.18% :     0.006286s :      1: task_emit
21.85% :     0.004550s :      1: type_inference
 0.28% :     0.000058s :      1: validate


TotalTime = 0.0145614, [30]
[bootstrap]: 0.00030283
[type_inference]: 0.00665329
[event_method]: 2.099e-05
[auto_monad]: 8.026e-05
[graph_reusing]: 6.33e-06
[pre_auto_parallel]: 2.59999e-06
[py_interpret_to_execute]: 4.515e-05
[rewriter_before_opt_a]: 8.563e-05
[expand_dump_flag]: 3.06001e-06
[jit_opt_a]: 0.00522195, [2]
    [Cycle 1]: 0.00179219, [27]
        [switch_simplify]: 0.00011058
        [loop_unroll]: 3.356e-05
        [a_1]: 0.00069428
        [with_stream_mark]: 1.927e-05
        [recompute_prepare]: 7.30998e-06
        [updatestate_depend_eliminate]: 3.57002e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 2.27999e-06
        [specialize_transform]: 6.59999e-06
        [updatestate_useless_node_eliminater]: 5.10999e-06
        [accelerated_algorithm]: 6.12999e-06
        [meta_shard_fg_expand]: 2.14e-06
        [get_grad_eliminate_]: 5.67001e-06
        [merge_forward]: 3.52002e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.507e-05
        [j_node_and_user_rematch]: 8.69998e-06
        [meta_fg_expand]: 2.12999e-06
        [replace_old_param]: 8.28001e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.00061211
        [add_forward_monad_depend]: 5.39e-06
        [auto_monad_grad]: 2.76999e-06
        [auto_monad_eliminator]: 1.329e-05
        [cse]: 2.944e-05
        [replace_applicator]: 1.19e-05
    [Cycle 2]: 0.00031614, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 4.76002e-06
        [a_1]: 8.679e-05
        [with_stream_mark]: 9.00999e-06
        [recompute_prepare]: 4.90001e-06
        [updatestate_depend_eliminate]: 2.59001e-06
        [updatestate_assign_eliminate]: 2.13002e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 8.2e-07
        [specialize_transform]: 4.79e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 4.62998e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 4.52998e-06
        [merge_forward]: 2.31998e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.308e-05
        [j_node_and_user_rematch]: 7.77e-06
        [meta_fg_expand]: 1.39e-06
        [replace_old_param]: 6.99001e-06
        [inline_without_move]: 4.81002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 1.36998e-06
        [auto_monad_eliminator]: 5.10001e-06
        [cse]: 1.015e-05
        [replace_applicator]: 5.62999e-06
[py_interpret_to_execute_after_opt_a]: 9.66e-06
[rewriter_after_opt_a]: 3.38e-05
[convert_after_rewriter]: 7.11999e-06
[order_py_execute_after_rewriter]: 5.03002e-06
[mutable_eliminate]: 0.00060198
[jit_opt_b]: 4.939e-05, [1]
    [Cycle 1]: 4.176e-05, [2]
        [frontend_op_eliminate]: 1.594e-05
        [inline_after_opt_a]: 1.464e-05
[cconv]: 2.544e-05
[loop_unroll]: 0.00042876
[jit_opt_after_cconv]: 0.0001383, [1]
    [Cycle 1]: 0.00013196, [11]
        [c_1]: 2.025e-05
        [parameter_eliminate]: 2.56998e-06
        [updatestate_depend_eliminate]: 5.60001e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 1.986e-05
        [call_graph_tuple_transform]: 1.761e-05
        [tuple_list_get_item_eliminator]: 5.08002e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 5.40022e-07
        [switch_simplify]: 4.96997e-06
[remove_dup_value]: 1.349e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 5.01997e-06
[add_recomputation]: 4.874e-05
[cse_after_recomputation]: 2.984e-05, [1]
    [Cycle 1]: 1.6e-05, [1]
        [cse]: 9.41003e-06
[auto_monad_reorder]: 1.477e-05
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 5.82001e-06
[opt_after_jit_grad]: 0.00044049
[symbol_engine_optimizer]: 7.537e-05, [1]
    [Cycle 1]: 6.852e-05, [6]
        [build]: 4.18999e-06
        [elim_shapecalc]: 7.6e-06
        [elim_not_effective]: 1.286e-05
        [opt_reshape]: 5.81e-06
        [fold_const_symbol]: 9.15999e-06
        [renormalize]: 6.79982e-07
[validate]: 3.347e-05
Sums
    bootstrap                            :     0.000303s :  2.79%
    type_inference                       :     0.006653s : 61.26%
    event_method                         :     0.000021s :  0.19%
    auto_monad                           :     0.000080s :  0.74%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000045s :  0.42%
    rewriter_before_opt_a                :     0.000086s :  0.79%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000116s :  1.07%
    jit_opt_a.loop_unroll                :     0.000038s :  0.35%
    jit_opt_a.a_1                        :     0.000781s :  7.19%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000612s :  5.64%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.17%
    jit_opt_a.cse                        :     0.000040s :  0.36%
    jit_opt_a.replace_applicator         :     0.000018s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.09%
    rewriter_after_opt_a                 :     0.000034s :  0.31%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000602s :  5.54%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.13%
    cconv                                :     0.000025s :  0.23%
    loop_unroll                          :     0.000429s :  3.95%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000049s :  0.45%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000440s :  4.06%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000033s :  0.31%


Time group info:
------[substitution.]   0.000252    26
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.00% :     0.000005s :      3: substitution.graph_param_transform
83.82% :     0.000211s :      6: substitution.inline
 1.08% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.98% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.42% :     0.000004s :      2: substitution.replace_old_param
 3.30% :     0.000008s :      1: substitution.switch_simplify
 5.09% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006592     2
84.59% :     0.005576s :      1: type_inference.infer
15.41% :     0.001016s :      1: type_inference.specialize
------[replace.]   0.000125     9
50.53% :     0.000063s :      6: replace.inline
26.36% :     0.000033s :      1: replace.switch_simplify
23.11% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000227     9
91.49% :     0.000208s :      6: match.inline
 3.34% :     0.000008s :      1: match.switch_simplify
 5.17% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   801
 1.61% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.21% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     12: predicate.addn_check_dump
 1.33% :     0.000002s :     12: predicate.addn_zero_filter
 1.81% :     0.000002s :     12: predicate.arithmetic_simplify
 1.48% :     0.000002s :     12: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     12: predicate.compare_switch_simplify
 1.47% :     0.000002s :     12: predicate.depend_value_elim
 1.17% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.73% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.23% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     12: predicate.environ_get_eliminate
 1.15% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.77% :     0.000007s :     26: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.99% :     0.000001s :      6: predicate.less_batch_normalization
 1.65% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     17: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.04% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.80% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     12: predicate.merge_addn
 1.15% :     0.000001s :     12: predicate.minmaximum_grad
 2.16% :     0.000003s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     17: predicate.partial_eliminate
 1.22% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     12: predicate.reduce_eliminate
 1.49% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     20: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     12: predicate.reshape_eliminate
 1.38% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.70% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.67% :     0.000003s :     20: predicate.switch_defer_inline
 2.33% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.21% :     0.000011s :     65: predicate.switch_simplify
 1.27% :     0.000002s :     12: predicate.tile_eliminate
 1.25% :     0.000002s :     12: predicate.transpose_eliminate
 1.39% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.64% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.86% :     0.000002s :     12: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000828    15
55.53% :     0.000460s :      6: func_graph_cloner_run.FuncGraphClonerGraph
 9.87% :     0.000082s :      1: func_graph_cloner_run.FuncGraphClonerNode
34.60% :     0.000287s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016280    72
 0.32% :     0.000051s :      1: add_recomputation
 0.51% :     0.000083s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.97% :     0.000322s :      1: bootstrap
 0.17% :     0.000028s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.20% :     0.000032s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
32.10% :     0.005225s :      1: jit_opt_a
 0.87% :     0.000141s :      1: jit_opt_after_cconv
 0.32% :     0.000052s :      1: jit_opt_b
 2.68% :     0.000436s :      1: loop_unroll
 3.75% :     0.000611s :      1: mutable_eliminate
 6.44% :     0.001049s :     26: opt.transform.jit_opt_a
 0.27% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.75% :     0.000448s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.29% :     0.000048s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 1.99% :     0.000323s :      1: renormalize.infer
 1.73% :     0.000282s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000037s :      1: rewriter_after_opt_a
 0.54% :     0.000088s :      1: rewriter_before_opt_a
 0.48% :     0.000078s :      1: symbol_engine_optimizer
40.95% :     0.006667s :      1: type_inference


TotalTime = 0.0264068, [33]
[bootstrap]: 0.00030621
[type_inference]: 0.00785158
[event_method]: 1.736e-05
[auto_monad]: 6.102e-05
[graph_reusing]: 5.57999e-06
[pre_auto_parallel]: 2.84999e-06
[py_interpret_to_execute]: 3.358e-05
[rewriter_before_opt_a]: 7.285e-05
[expand_dump_flag]: 3.14999e-06
[jit_opt_a]: 0.007622, [2]
    [Cycle 1]: 0.00171852, [27]
        [switch_simplify]: 5.833e-05
        [loop_unroll]: 2.865e-05
        [a_1]: 0.0005535
        [with_stream_mark]: 2.527e-05
        [recompute_prepare]: 7.33e-06
        [updatestate_depend_eliminate]: 3.65e-06
        [updatestate_assign_eliminate]: 2.86e-06
        [updatestate_loads_eliminate]: 2.84001e-06
        [parameter_eliminate]: 2.09e-06
        [specialize_transform]: 5.91003e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.89999e-06
        [get_grad_eliminate_]: 5.27999e-06
        [merge_forward]: 3.78001e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.422e-05
        [j_node_and_user_rematch]: 9.49999e-06
        [meta_fg_expand]: 2.13002e-06
        [replace_old_param]: 9.36998e-06
        [inline_without_move]: 5.23002e-06
        [renormalize]: 0.00072316
        [add_forward_monad_depend]: 5.36002e-06
        [auto_monad_grad]: 2.39001e-06
        [auto_monad_eliminator]: 1.293e-05
        [cse]: 3.125e-05
        [replace_applicator]: 1.212e-05
    [Cycle 2]: 0.0003218, [27]
        [switch_simplify]: 5.59998e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 8.833e-05
        [with_stream_mark]: 1.039e-05
        [recompute_prepare]: 4.89998e-06
        [updatestate_depend_eliminate]: 2.69999e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [parameter_eliminate]: 1.50999e-06
        [specialize_transform]: 4.94e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 1.05001e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 2.74001e-06
        [cell_reuse_recompute_pass]: 2.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.255e-05
        [j_node_and_user_rematch]: 8.07e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 7.36999e-06
        [inline_without_move]: 4.52e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.58002e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 5.99e-06
        [cse]: 1.096e-05
        [replace_applicator]: 5.17999e-06
[py_interpret_to_execute_after_opt_a]: 1.041e-05
[rewriter_after_opt_a]: 3.731e-05
[convert_after_rewriter]: 6.94001e-06
[order_py_execute_after_rewriter]: 5.17e-06
[mutable_eliminate]: 0.00140142
[jit_opt_b]: 5.009e-05, [1]
    [Cycle 1]: 4.283e-05, [2]
        [frontend_op_eliminate]: 1.675e-05
        [inline_after_opt_a]: 1.476e-05
[cconv]: 3.437e-05
[loop_unroll]: 0.00051924
[jit_opt_after_cconv]: 0.00014802, [1]
    [Cycle 1]: 0.00014103, [11]
        [c_1]: 2.042e-05
        [parameter_eliminate]: 3.13998e-06
        [updatestate_depend_eliminate]: 6.88e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.03997e-06
        [cse]: 2.486e-05
        [call_graph_tuple_transform]: 1.968e-05
        [tuple_list_get_item_eliminator]: 5.11002e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 4.99975e-07
        [switch_simplify]: 5.39998e-06
[remove_dup_value]: 1.306e-05
[partial_unused_args_eliminate]: 1.95001e-06
[environ_conv]: 5.45001e-06
[add_recomputation]: 5.523e-05
[cse_after_recomputation]: 1.923e-05, [1]
    [Cycle 1]: 1.378e-05, [1]
        [cse]: 8.21002e-06
[auto_monad_reorder]: 1.468e-05
[get_jit_bprop_graph]: 1.96e-06
[rewriter_after_jit_bprop_graph]: 5.96e-06
[opt_after_jit_grad]: 0.00045871
[symbol_engine_optimizer]: 7.334e-05, [1]
    [Cycle 1]: 6.689e-05, [6]
        [build]: 4.60999e-06
        [elim_shapecalc]: 7.58999e-06
        [elim_not_effective]: 1.257e-05
        [opt_reshape]: 5.56e-06
        [fold_const_symbol]: 8.37e-06
        [renormalize]: 5.19998e-07
[validate]: 4.023e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00730639
[execute]: 9.81e-06
Sums
    bootstrap                            :     0.000306s :  1.52%
    type_inference                       :     0.007852s : 38.87%
    event_method                         :     0.000017s :  0.09%
    auto_monad                           :     0.000061s :  0.30%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000034s :  0.17%
    rewriter_before_opt_a                :     0.000073s :  0.36%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.32%
    jit_opt_a.loop_unroll                :     0.000034s :  0.17%
    jit_opt_a.a_1                        :     0.000642s :  3.18%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000723s :  3.58%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.09%
    jit_opt_a.cse                        :     0.000042s :  0.21%
    jit_opt_a.replace_applicator         :     0.000017s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.05%
    rewriter_after_opt_a                 :     0.000037s :  0.18%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.001401s :  6.94%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.07%
    cconv                                :     0.000034s :  0.17%
    loop_unroll                          :     0.000519s :  2.57%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000055s :  0.27%
    cse_after_recomputation.cse          :     0.000008s :  0.04%
    auto_monad_reorder                   :     0.000015s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000459s :  2.27%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.20%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007306s : 36.17%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000232    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.23% :     0.000005s :      3: substitution.graph_param_transform
85.40% :     0.000198s :      4: substitution.inline
 1.54% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.99% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.98% :     0.000005s :      2: substitution.replace_old_param
 5.49% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007783     2
89.49% :     0.006964s :      1: type_inference.infer
10.51% :     0.000818s :      1: type_inference.specialize
------[replace.]   0.000066     6
68.62% :     0.000045s :      4: replace.inline
31.38% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000207     6
94.39% :     0.000196s :      4: match.inline
 5.61% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.22% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.36% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 2.30% :     0.000003s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.56% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.16% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 6.27% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.68% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.95% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 2.39% :     0.000003s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.34% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 1.18% :     0.000001s :      6: predicate.specialize_transform
 1.19% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.09% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.10% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.42% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000607    12
50.36% :     0.000306s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.97% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.67% :     0.000253s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028096    76
 0.21% :     0.000058s :      1: add_recomputation
 0.23% :     0.000064s :      1: auto_monad
 0.06% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.14% :     0.000321s :      1: bootstrap
 0.13% :     0.000037s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000021s :      1: event_method
 0.05% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
27.14% :     0.007625s :      1: jit_opt_a
 0.54% :     0.000151s :      1: jit_opt_after_cconv
 0.19% :     0.000053s :      1: jit_opt_b
 1.88% :     0.000528s :      1: loop_unroll
 5.03% :     0.001413s :      1: mutable_eliminate
 3.03% :     0.000853s :     26: opt.transform.jit_opt_a
 0.17% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000023s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.66% :     0.000467s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000036s :      1: py_interpret_to_execute
 0.05% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000015s :      1: remove_dup_value
 1.45% :     0.000407s :      1: renormalize.infer
 1.10% :     0.000308s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000042s :      1: rewriter_after_opt_a
 0.27% :     0.000076s :      1: rewriter_before_opt_a
 0.27% :     0.000076s :      1: symbol_engine_optimizer
26.07% :     0.007324s :      1: task_emit
28.01% :     0.007869s :      1: type_inference
 0.22% :     0.000062s :      1: validate


TotalTime = 0.0332687, [33]
[bootstrap]: 0.00027961
[type_inference]: 0.00441865
[event_method]: 1.625e-05
[auto_monad]: 7.193e-05
[graph_reusing]: 5.22999e-06
[pre_auto_parallel]: 2.22001e-06
[py_interpret_to_execute]: 3.352e-05
[rewriter_before_opt_a]: 6.668e-05
[expand_dump_flag]: 2.91999e-06
[jit_opt_a]: 0.0161022, [2]
    [Cycle 1]: 0.0123492, [27]
        [switch_simplify]: 5.251e-05
        [loop_unroll]: 2.965e-05
        [a_1]: 0.00054925
        [with_stream_mark]: 1.989e-05
        [recompute_prepare]: 7.87e-06
        [updatestate_depend_eliminate]: 3.31999e-06
        [updatestate_assign_eliminate]: 2.63003e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [parameter_eliminate]: 1.69998e-06
        [specialize_transform]: 6.10002e-06
        [updatestate_useless_node_eliminater]: 4.79002e-06
        [accelerated_algorithm]: 5.04998e-06
        [meta_shard_fg_expand]: 2.31e-06
        [get_grad_eliminate_]: 5.46e-06
        [merge_forward]: 3.93999e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.313e-05
        [j_node_and_user_rematch]: 9.25999e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 9.26002e-06
        [inline_without_move]: 5.17999e-06
        [renormalize]: 0.0112966
        [add_forward_monad_depend]: 1.238e-05
        [auto_monad_grad]: 3.13e-06
        [auto_monad_eliminator]: 2.299e-05
        [cse]: 3.305e-05
        [replace_applicator]: 2.984e-05
    [Cycle 2]: 0.00043356, [27]
        [switch_simplify]: 7.77e-06
        [loop_unroll]: 8.12e-06
        [a_1]: 0.00012561
        [with_stream_mark]: 1.66e-05
        [recompute_prepare]: 6.73e-06
        [updatestate_depend_eliminate]: 3.8e-06
        [updatestate_assign_eliminate]: 2.79999e-06
        [updatestate_loads_eliminate]: 3.06001e-06
        [parameter_eliminate]: 2.09e-06
        [specialize_transform]: 5.57001e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 8.32998e-06
        [meta_shard_fg_expand]: 2.31e-06
        [get_grad_eliminate_]: 5.46e-06
        [merge_forward]: 4.72998e-06
        [cell_reuse_recompute_pass]: 4.89e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.145e-05
        [j_node_and_user_rematch]: 9.30001e-06
        [meta_fg_expand]: 2.95002e-06
        [replace_old_param]: 9.06002e-06
        [inline_without_move]: 1.382e-05
        [renormalize]: 1.19995e-07
        [add_forward_monad_depend]: 1.59998e-06
        [auto_monad_grad]: 1.42e-06
        [auto_monad_eliminator]: 6.94001e-06
        [cse]: 1.49e-05
        [replace_applicator]: 6.21e-06
[py_interpret_to_execute_after_opt_a]: 2.103e-05
[rewriter_after_opt_a]: 4.524e-05
[convert_after_rewriter]: 7.55e-06
[order_py_execute_after_rewriter]: 6.83998e-06
[mutable_eliminate]: 0.00133707
[jit_opt_b]: 6.308e-05, [1]
    [Cycle 1]: 5.285e-05, [2]
        [frontend_op_eliminate]: 2.182e-05
        [inline_after_opt_a]: 1.826e-05
[cconv]: 3.322e-05
[loop_unroll]: 0.00053146
[jit_opt_after_cconv]: 0.00016717, [1]
    [Cycle 1]: 0.00016009, [11]
        [c_1]: 2.475e-05
        [parameter_eliminate]: 4.4e-06
        [updatestate_depend_eliminate]: 6.92002e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [cse]: 2.55e-05
        [call_graph_tuple_transform]: 2.51e-05
        [tuple_list_get_item_eliminator]: 6.29001e-06
        [none_parameter_eliminate]: 1.92999e-06
        [renormalize]: 6.79982e-07
        [switch_simplify]: 5.62999e-06
[remove_dup_value]: 1.523e-05
[partial_unused_args_eliminate]: 2.69999e-06
[environ_conv]: 6.98998e-06
[add_recomputation]: 5.357e-05
[cse_after_recomputation]: 2.247e-05, [1]
    [Cycle 1]: 1.676e-05, [1]
        [cse]: 9.81e-06
[auto_monad_reorder]: 1.625e-05
[get_jit_bprop_graph]: 2.26e-06
[rewriter_after_jit_bprop_graph]: 7.6e-06
[opt_after_jit_grad]: 0.00057837
[symbol_engine_optimizer]: 8.455e-05, [1]
    [Cycle 1]: 7.665e-05, [6]
        [build]: 4.92999e-06
        [elim_shapecalc]: 8.26002e-06
        [elim_not_effective]: 1.612e-05
        [opt_reshape]: 6.63e-06
        [fold_const_symbol]: 1.011e-05
        [renormalize]: 5.89993e-07
[validate]: 4.174e-05
[backend_pass]: 1.22e-06
[task_emit]: 0.00898458
[execute]: 8.67e-06
Sums
    bootstrap                            :     0.000280s :  0.96%
    type_inference                       :     0.004419s : 15.11%
    event_method                         :     0.000016s :  0.06%
    auto_monad                           :     0.000072s :  0.25%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000034s :  0.11%
    rewriter_before_opt_a                :     0.000067s :  0.23%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000060s :  0.21%
    jit_opt_a.loop_unroll                :     0.000038s :  0.13%
    jit_opt_a.a_1                        :     0.000675s :  2.31%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.06%
    jit_opt_a.inline_without_move        :     0.000019s :  0.06%
    jit_opt_a.renormalize                :     0.011297s : 38.64%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000030s :  0.10%
    jit_opt_a.cse                        :     0.000048s :  0.16%
    jit_opt_a.replace_applicator         :     0.000036s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000021s :  0.07%
    rewriter_after_opt_a                 :     0.000045s :  0.15%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.001337s :  4.57%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000033s :  0.11%
    loop_unroll                          :     0.000531s :  1.82%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000015s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000054s :  0.18%
    cse_after_recomputation.cse          :     0.000010s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.03%
    opt_after_jit_grad                   :     0.000578s :  1.98%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008985s : 30.73%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000228    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 3.33% :     0.000008s :      3: substitution.graph_param_transform
82.14% :     0.000187s :      4: substitution.inline
 1.95% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.58% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.69% :     0.000006s :      2: substitution.replace_old_param
 5.68% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004359     2
82.24% :     0.003585s :      1: type_inference.infer
17.76% :     0.000774s :      1: type_inference.specialize
------[replace.]   0.000069     6
67.53% :     0.000047s :      4: replace.inline
32.47% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000196     6
94.08% :     0.000184s :      4: match.inline
 5.92% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   691
 1.53% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.88% :     0.000001s :     10: predicate.addn_check_dump
 1.58% :     0.000002s :     10: predicate.addn_zero_filter
 2.42% :     0.000003s :     10: predicate.arithmetic_simplify
 1.47% :     0.000002s :     10: predicate.cast_eliminate
 0.61% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.88% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 1.26% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     10: predicate.environ_get_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.21% :     0.000002s :      6: predicate.get_grad_eliminate
 0.33% :     0.000000s :      3: predicate.graph_param_transform
 5.01% :     0.000007s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :      6: predicate.less_batch_normalization
 1.76% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000003s :     15: predicate.load_eliminater
 1.48% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.70% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.60% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000002s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.68% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 1.95% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.11% :     0.000003s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.97% :     0.000003s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.48% :     0.000001s :      3: predicate.reset_defer_inline
 1.72% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.58% :     0.000002s :     10: predicate.same_eliminate
 0.41% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.25% :     0.000002s :      6: predicate.special_op_eliminate
 0.83% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 1.97% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.36% :     0.000009s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000002s :     10: predicate.transpose_eliminate
 1.55% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.89% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.83% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.36% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.32% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.36% :     0.000003s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.010857    12
 2.58% :     0.000280s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 0.40% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
97.02% :     0.010533s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.045618    76
 0.12% :     0.000057s :      1: add_recomputation
 0.17% :     0.000076s :      1: auto_monad
 0.04% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.65% :     0.000295s :      1: bootstrap
 0.08% :     0.000036s :      1: cconv
 0.02% :     0.000011s :      1: convert_after_rewriter
 0.05% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.04% :     0.000020s :      1: event_method
 0.03% :     0.000013s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
35.31% :     0.016107s :      1: jit_opt_a
 0.37% :     0.000170s :      1: jit_opt_after_cconv
 0.14% :     0.000066s :      1: jit_opt_b
 1.19% :     0.000542s :      1: loop_unroll
 2.96% :     0.001352s :      1: mutable_eliminate
 2.03% :     0.000925s :     26: opt.transform.jit_opt_a
 0.13% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000032s :      4: opt.transform.jit_opt_b
 0.04% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.29% :     0.000589s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.08% :     0.000036s :      1: py_interpret_to_execute
 0.05% :     0.000023s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000018s :      1: remove_dup_value
 0.87% :     0.000398s :      1: renormalize.infer
23.86% :     0.010886s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000049s :      1: rewriter_after_opt_a
 0.15% :     0.000069s :      1: rewriter_before_opt_a
 0.19% :     0.000088s :      1: symbol_engine_optimizer
19.73% :     0.009001s :      1: task_emit
 9.71% :     0.004430s :      1: type_inference
 0.14% :     0.000065s :      1: validate


TotalTime = 0.0308603, [33]
[bootstrap]: 0.00048383
[type_inference]: 0.00942558
[event_method]: 1.988e-05
[auto_monad]: 7.438e-05
[graph_reusing]: 5.99e-06
[pre_auto_parallel]: 3.01001e-06
[py_interpret_to_execute]: 5.074e-05
[rewriter_before_opt_a]: 7.973e-05
[expand_dump_flag]: 3.58e-06
[jit_opt_a]: 0.00897636, [2]
    [Cycle 1]: 0.00401812, [27]
        [switch_simplify]: 6.778e-05
        [loop_unroll]: 3.165e-05
        [a_1]: 0.00062548
        [with_stream_mark]: 2.01e-05
        [recompute_prepare]: 7.85998e-06
        [updatestate_depend_eliminate]: 3.71001e-06
        [updatestate_assign_eliminate]: 2.71e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [parameter_eliminate]: 1.64e-06
        [specialize_transform]: 5.87001e-06
        [updatestate_useless_node_eliminater]: 5.14998e-06
        [accelerated_algorithm]: 7.16001e-06
        [meta_shard_fg_expand]: 2.81e-06
        [get_grad_eliminate_]: 6.14001e-06
        [merge_forward]: 3.61999e-06
        [cell_reuse_recompute_pass]: 1.09998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.806e-05
        [j_node_and_user_rematch]: 1.071e-05
        [meta_fg_expand]: 2.49001e-06
        [replace_old_param]: 9.74999e-06
        [inline_without_move]: 5.67999e-06
        [renormalize]: 0.00289174
        [add_forward_monad_depend]: 1.02e-05
        [auto_monad_grad]: 2.66e-06
        [auto_monad_eliminator]: 2.198e-05
        [cse]: 2.997e-05
        [replace_applicator]: 2.271e-05
    [Cycle 2]: 0.00038254, [27]
        [switch_simplify]: 6.84999e-06
        [loop_unroll]: 5.62999e-06
        [a_1]: 0.00011025
        [with_stream_mark]: 1.57e-05
        [recompute_prepare]: 5.21998e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.83e-06
        [parameter_eliminate]: 2.02999e-06
        [specialize_transform]: 4.84998e-06
        [updatestate_useless_node_eliminater]: 4.66002e-06
        [accelerated_algorithm]: 6.23998e-06
        [meta_shard_fg_expand]: 2.36e-06
        [get_grad_eliminate_]: 4.80999e-06
        [merge_forward]: 4.40999e-06
        [cell_reuse_recompute_pass]: 4.69998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.731e-05
        [j_node_and_user_rematch]: 8.73001e-06
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 9.05001e-06
        [inline_without_move]: 4.90001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.69e-06
        [auto_monad_grad]: 1.29998e-06
        [auto_monad_eliminator]: 5.81e-06
        [cse]: 1.177e-05
        [replace_applicator]: 5.56e-06
[py_interpret_to_execute_after_opt_a]: 1.889e-05
[rewriter_after_opt_a]: 4.065e-05
[convert_after_rewriter]: 8.20999e-06
[order_py_execute_after_rewriter]: 4.85001e-06
[mutable_eliminate]: 0.00093412
[jit_opt_b]: 6.059e-05, [1]
    [Cycle 1]: 5.053e-05, [2]
        [frontend_op_eliminate]: 1.949e-05
        [inline_after_opt_a]: 1.706e-05
[cconv]: 3.841e-05
[loop_unroll]: 0.00053329
[jit_opt_after_cconv]: 0.00017865, [1]
    [Cycle 1]: 0.0001704, [11]
        [c_1]: 2.385e-05
        [parameter_eliminate]: 5.72001e-06
        [updatestate_depend_eliminate]: 9.47001e-06
        [updatestate_assign_eliminate]: 2.86e-06
        [updatestate_loads_eliminate]: 2.76e-06
        [cse]: 3.144e-05
        [call_graph_tuple_transform]: 2.668e-05
        [tuple_list_get_item_eliminator]: 6.28e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 9.30013e-07
        [switch_simplify]: 6.26998e-06
[remove_dup_value]: 1.626e-05
[partial_unused_args_eliminate]: 2.44999e-06
[environ_conv]: 7.02002e-06
[add_recomputation]: 6.082e-05
[cse_after_recomputation]: 2.608e-05, [1]
    [Cycle 1]: 1.838e-05, [1]
        [cse]: 1.088e-05
[auto_monad_reorder]: 1.677e-05
[get_jit_bprop_graph]: 2.63003e-06
[rewriter_after_jit_bprop_graph]: 9.27999e-06
[opt_after_jit_grad]: 0.00063015
[symbol_engine_optimizer]: 8.712e-05, [1]
    [Cycle 1]: 7.831e-05, [6]
        [build]: 7.03e-06
        [elim_shapecalc]: 8.77e-06
        [elim_not_effective]: 1.467e-05
        [opt_reshape]: 7.3e-06
        [fold_const_symbol]: 9.77001e-06
        [renormalize]: 9.00007e-07
[validate]: 5.354e-05
[backend_pass]: 1.00999e-06
[task_emit]: 0.00872686
[execute]: 1.179e-05
Sums
    bootstrap                            :     0.000484s :  1.89%
    type_inference                       :     0.009426s : 36.89%
    event_method                         :     0.000020s :  0.08%
    auto_monad                           :     0.000074s :  0.29%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000051s :  0.20%
    rewriter_before_opt_a                :     0.000080s :  0.31%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000075s :  0.29%
    jit_opt_a.loop_unroll                :     0.000037s :  0.15%
    jit_opt_a.a_1                        :     0.000736s :  2.88%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.04%
    jit_opt_a.merge_forward              :     0.000008s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.07%
    jit_opt_a.inline_without_move        :     0.000011s :  0.04%
    jit_opt_a.renormalize                :     0.002892s : 11.32%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.11%
    jit_opt_a.cse                        :     0.000042s :  0.16%
    jit_opt_a.replace_applicator         :     0.000028s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.07%
    rewriter_after_opt_a                 :     0.000041s :  0.16%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000934s :  3.66%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.07%
    cconv                                :     0.000038s :  0.15%
    loop_unroll                          :     0.000533s :  2.09%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000031s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000061s :  0.24%
    cse_after_recomputation.cse          :     0.000011s :  0.04%
    auto_monad_reorder                   :     0.000017s :  0.07%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.04%
    opt_after_jit_grad                   :     0.000630s :  2.47%
    symbol_engine_optimizer.build        :     0.000007s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000054s :  0.21%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008727s : 34.15%
    execute                              :     0.000012s :  0.05%


Time group info:
------[substitution.]   0.000274    23
 0.92% :     0.000003s :      2: substitution.elim_not_effective
 0.71% :     0.000002s :      2: substitution.fold_const_symbol
 2.17% :     0.000006s :      3: substitution.graph_param_transform
85.08% :     0.000233s :      4: substitution.inline
 1.70% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.24% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.14% :     0.000006s :      2: substitution.replace_old_param
 5.04% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009333     2
67.12% :     0.006265s :      1: type_inference.infer
32.88% :     0.003069s :      1: type_inference.specialize
------[replace.]   0.000078     6
66.07% :     0.000052s :      4: replace.inline
33.93% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000243     6
94.82% :     0.000231s :      4: match.inline
 5.18% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   691
 1.47% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 2.70% :     0.000004s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.35% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.90% :     0.000001s :     10: predicate.compare_switch_simplify
 1.14% :     0.000002s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.46% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.55% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.97% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.84% :     0.000008s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.85% :     0.000003s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.62% :     0.000002s :     15: predicate.load_eliminater
 1.89% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.85% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.29% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.94% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 2.35% :     0.000003s :      3: predicate.mutable_eliminate
 0.39% :     0.000001s :      3: predicate.opt_reshape
 1.84% :     0.000002s :     15: predicate.partial_eliminate
 0.96% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.52% :     0.000001s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.08% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.08% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.12% :     0.000003s :     16: predicate.switch_defer_inline
 1.85% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.36% :     0.000010s :     53: predicate.switch_simplify
 1.47% :     0.000002s :     10: predicate.tile_eliminate
 1.20% :     0.000002s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.28% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 2.06% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.31% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.44% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.57% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.002698    12
14.01% :     0.000378s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 2.01% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
83.98% :     0.002266s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.034849    76
 0.18% :     0.000064s :      1: add_recomputation
 0.22% :     0.000078s :      1: auto_monad
 0.06% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.46% :     0.000508s :      1: bootstrap
 0.12% :     0.000041s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000028s :      1: cse_after_recomputation
 0.03% :     0.000011s :      1: environ_conv
 0.07% :     0.000024s :      1: event_method
 0.05% :     0.000017s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
25.77% :     0.008981s :      1: jit_opt_a
 0.52% :     0.000182s :      1: jit_opt_after_cconv
 0.18% :     0.000063s :      1: jit_opt_b
 1.56% :     0.000543s :      1: loop_unroll
 2.72% :     0.000950s :      1: mutable_eliminate
 2.82% :     0.000984s :     26: opt.transform.jit_opt_a
 0.17% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000029s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.84% :     0.000641s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000054s :      1: py_interpret_to_execute
 0.06% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 1.41% :     0.000491s :      1: renormalize.infer
 6.86% :     0.002390s :      1: renormalize.specialize
 0.03% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000044s :      1: rewriter_after_opt_a
 0.24% :     0.000083s :      1: rewriter_before_opt_a
 0.26% :     0.000090s :      1: symbol_engine_optimizer
25.11% :     0.008751s :      1: task_emit
27.09% :     0.009442s :      1: type_inference
 0.23% :     0.000080s :      1: validate


TotalTime = 0.0148308, [30]
[bootstrap]: 0.00029714
[type_inference]: 0.00659066
[event_method]: 2.24e-05
[auto_monad]: 8.228e-05
[graph_reusing]: 7.25e-06
[pre_auto_parallel]: 2.64001e-06
[py_interpret_to_execute]: 4.641e-05
[rewriter_before_opt_a]: 8.545e-05
[expand_dump_flag]: 3.13e-06
[jit_opt_a]: 0.00560059, [2]
    [Cycle 1]: 0.00193191, [27]
        [switch_simplify]: 0.000123
        [loop_unroll]: 3.366e-05
        [a_1]: 0.00068573
        [with_stream_mark]: 2.108e-05
        [recompute_prepare]: 7.7e-06
        [updatestate_depend_eliminate]: 4.08999e-06
        [updatestate_assign_eliminate]: 3.27002e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 5.67001e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 2.02999e-06
        [get_grad_eliminate_]: 5.44998e-06
        [merge_forward]: 3.43999e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.439e-05
        [j_node_and_user_rematch]: 8.40999e-06
        [meta_fg_expand]: 2.46e-06
        [replace_old_param]: 9.39998e-06
        [inline_without_move]: 5.46002e-06
        [renormalize]: 0.00073797
        [add_forward_monad_depend]: 6.26e-06
        [auto_monad_grad]: 2.51e-06
        [auto_monad_eliminator]: 1.421e-05
        [cse]: 3.098e-05
        [replace_applicator]: 1.124e-05
    [Cycle 2]: 0.00033007, [27]
        [switch_simplify]: 5.86998e-06
        [loop_unroll]: 5.03002e-06
        [a_1]: 8.834e-05
        [with_stream_mark]: 9.29e-06
        [recompute_prepare]: 5.22999e-06
        [updatestate_depend_eliminate]: 2.51e-06
        [updatestate_assign_eliminate]: 1.89999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 4.65999e-06
        [updatestate_useless_node_eliminater]: 4.76002e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 1.32999e-06
        [get_grad_eliminate_]: 1.101e-05
        [merge_forward]: 2.69001e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.329e-05
        [j_node_and_user_rematch]: 7.66001e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 7.7e-06
        [inline_without_move]: 5.04e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 1.03001e-06
        [auto_monad_eliminator]: 5.56e-06
        [cse]: 1.215e-05
        [replace_applicator]: 4.77998e-06
[py_interpret_to_execute_after_opt_a]: 1.195e-05
[rewriter_after_opt_a]: 3.669e-05
[convert_after_rewriter]: 6.98998e-06
[order_py_execute_after_rewriter]: 5.56e-06
[mutable_eliminate]: 0.00055216
[jit_opt_b]: 4.819e-05, [1]
    [Cycle 1]: 4.143e-05, [2]
        [frontend_op_eliminate]: 1.567e-05
        [inline_after_opt_a]: 1.445e-05
[cconv]: 2.658e-05
[loop_unroll]: 0.0004158
[jit_opt_after_cconv]: 0.0001419, [1]
    [Cycle 1]: 0.00013567, [11]
        [c_1]: 2.033e-05
        [parameter_eliminate]: 3.5e-06
        [updatestate_depend_eliminate]: 6.02999e-06
        [updatestate_assign_eliminate]: 2.54001e-06
        [updatestate_loads_eliminate]: 2.13002e-06
        [cse]: 2.004e-05
        [call_graph_tuple_transform]: 1.85e-05
        [tuple_list_get_item_eliminator]: 5.21002e-06
        [none_parameter_eliminate]: 1.99e-06
        [renormalize]: 6.60017e-07
        [switch_simplify]: 5.38002e-06
[remove_dup_value]: 1.27e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 5.61003e-06
[add_recomputation]: 5.006e-05
[cse_after_recomputation]: 2.065e-05, [1]
    [Cycle 1]: 1.482e-05, [1]
        [cse]: 8.77e-06
[auto_monad_reorder]: 1.578e-05
[get_jit_bprop_graph]: 2.16e-06
[rewriter_after_jit_bprop_graph]: 6.07001e-06
[opt_after_jit_grad]: 0.00043671
[symbol_engine_optimizer]: 7.211e-05, [1]
    [Cycle 1]: 6.567e-05, [6]
        [build]: 4.77998e-06
        [elim_shapecalc]: 7.18998e-06
        [elim_not_effective]: 1.297e-05
        [opt_reshape]: 5.11002e-06
        [fold_const_symbol]: 8.52e-06
        [renormalize]: 4.60015e-07
[validate]: 3.82e-05
Sums
    bootstrap                            :     0.000297s :  2.73%
    type_inference                       :     0.006591s : 60.50%
    event_method                         :     0.000022s :  0.21%
    auto_monad                           :     0.000082s :  0.76%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000046s :  0.43%
    rewriter_before_opt_a                :     0.000085s :  0.78%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000129s :  1.18%
    jit_opt_a.loop_unroll                :     0.000039s :  0.36%
    jit_opt_a.a_1                        :     0.000774s :  7.11%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000010s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.15%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.16%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000738s :  6.78%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.18%
    jit_opt_a.cse                        :     0.000043s :  0.40%
    jit_opt_a.replace_applicator         :     0.000016s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.11%
    rewriter_after_opt_a                 :     0.000037s :  0.34%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000552s :  5.07%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.13%
    cconv                                :     0.000027s :  0.24%
    loop_unroll                          :     0.000416s :  3.82%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000050s :  0.46%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000437s :  4.01%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.35%


Time group info:
------[substitution.]   0.000281    26
 0.65% :     0.000002s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 1.92% :     0.000005s :      3: substitution.graph_param_transform
85.25% :     0.000239s :      6: substitution.inline
 0.92% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.60% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.86% :     0.000005s :      2: substitution.replace_old_param
 2.82% :     0.000008s :      1: substitution.switch_simplify
 4.54% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006519     2
84.09% :     0.005482s :      1: type_inference.infer
15.91% :     0.001037s :      1: type_inference.specialize
------[replace.]   0.000134     9
49.11% :     0.000066s :      6: replace.inline
29.52% :     0.000040s :      1: replace.switch_simplify
21.37% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000255     9
92.64% :     0.000236s :      6: match.inline
 2.82% :     0.000007s :      1: match.switch_simplify
 4.54% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   801
 1.35% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.96% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000002s :     12: predicate.addn_check_dump
 1.56% :     0.000002s :     12: predicate.addn_zero_filter
 2.30% :     0.000003s :     12: predicate.arithmetic_simplify
 1.24% :     0.000002s :     12: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     12: predicate.compare_switch_simplify
 1.40% :     0.000002s :     12: predicate.depend_value_elim
 1.12% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.57% :     0.000001s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     12: predicate.environ_get_eliminate
 1.13% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.80% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.77% :     0.000008s :     26: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.94% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     17: predicate.load_eliminater
 1.69% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.43% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000002s :     12: predicate.merge_addn
 1.13% :     0.000001s :     12: predicate.minmaximum_grad
 1.60% :     0.000002s :      3: predicate.mutable_eliminate
 0.36% :     0.000000s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     17: predicate.partial_eliminate
 1.22% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.43% :     0.000002s :     12: predicate.reduce_eliminate
 1.63% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     20: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     12: predicate.reshape_eliminate
 1.42% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.66% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.76% :     0.000001s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.79% :     0.000004s :     20: predicate.switch_defer_inline
 2.36% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.76% :     0.000011s :     65: predicate.switch_simplify
 1.27% :     0.000002s :     12: predicate.tile_eliminate
 1.30% :     0.000002s :     12: predicate.transpose_eliminate
 1.38% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.71% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.39% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     12: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000778    15
48.04% :     0.000374s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.03% :     0.000101s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.92% :     0.000303s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016677    72
 0.32% :     0.000053s :      1: add_recomputation
 0.52% :     0.000086s :      1: auto_monad
 0.11% :     0.000019s :      1: auto_monad_reorder
 1.92% :     0.000321s :      1: bootstrap
 0.18% :     0.000029s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.17% :     0.000028s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000010s :      1: graph_reusing
33.60% :     0.005604s :      1: jit_opt_a
 0.87% :     0.000145s :      1: jit_opt_after_cconv
 0.30% :     0.000051s :      1: jit_opt_b
 2.54% :     0.000423s :      1: loop_unroll
 3.37% :     0.000561s :      1: mutable_eliminate
 6.36% :     0.001060s :     26: opt.transform.jit_opt_a
 0.27% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.66% :     0.000444s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000049s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000015s :      1: remove_dup_value
 2.56% :     0.000426s :      1: renormalize.infer
 1.82% :     0.000303s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000040s :      1: rewriter_after_opt_a
 0.53% :     0.000088s :      1: rewriter_before_opt_a
 0.45% :     0.000074s :      1: symbol_engine_optimizer
39.64% :     0.006611s :      1: type_inference


TotalTime = 0.020702, [33]
[bootstrap]: 0.00036576
[type_inference]: 0.00522844
[event_method]: 1.823e-05
[auto_monad]: 6.903e-05
[graph_reusing]: 5.87999e-06
[pre_auto_parallel]: 3.08e-06
[py_interpret_to_execute]: 3.74e-05
[rewriter_before_opt_a]: 7.322e-05
[expand_dump_flag]: 3.31999e-06
[jit_opt_a]: 0.00557297, [2]
    [Cycle 1]: 0.0016565, [27]
        [switch_simplify]: 6.06e-05
        [loop_unroll]: 2.93e-05
        [a_1]: 0.00052429
        [with_stream_mark]: 1.852e-05
        [recompute_prepare]: 7.21999e-06
        [updatestate_depend_eliminate]: 3.96001e-06
        [updatestate_assign_eliminate]: 3.10998e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 2.39001e-06
        [specialize_transform]: 5.59e-06
        [updatestate_useless_node_eliminater]: 4.86002e-06
        [accelerated_algorithm]: 5.10999e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 5.35999e-06
        [merge_forward]: 3.73001e-06
        [cell_reuse_recompute_pass]: 1.24998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.416e-05
        [j_node_and_user_rematch]: 8.54998e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 8.95001e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.00069158
        [add_forward_monad_depend]: 6.13998e-06
        [auto_monad_grad]: 2.56e-06
        [auto_monad_eliminator]: 1.431e-05
        [cse]: 3.067e-05
        [replace_applicator]: 1.149e-05
    [Cycle 2]: 0.00031538, [27]
        [switch_simplify]: 5.81998e-06
        [loop_unroll]: 4.72e-06
        [a_1]: 8.964e-05
        [with_stream_mark]: 8.92999e-06
        [recompute_prepare]: 4.94e-06
        [updatestate_depend_eliminate]: 2.34001e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 1.22999e-06
        [specialize_transform]: 4.99e-06
        [updatestate_useless_node_eliminater]: 4.79998e-06
        [accelerated_algorithm]: 4.70001e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 4.55001e-06
        [merge_forward]: 2.24001e-06
        [cell_reuse_recompute_pass]: 2.01e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.274e-05
        [j_node_and_user_rematch]: 7.82002e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 7.25998e-06
        [inline_without_move]: 4.63999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.49e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 5.14e-06
        [cse]: 1.025e-05
        [replace_applicator]: 4.79e-06
[py_interpret_to_execute_after_opt_a]: 1.095e-05
[rewriter_after_opt_a]: 3.5e-05
[convert_after_rewriter]: 7.53e-06
[order_py_execute_after_rewriter]: 4.80001e-06
[mutable_eliminate]: 0.00058292
[jit_opt_b]: 5.018e-05, [1]
    [Cycle 1]: 4.263e-05, [2]
        [frontend_op_eliminate]: 1.581e-05
        [inline_after_opt_a]: 1.506e-05
[cconv]: 2.715e-05
[loop_unroll]: 0.00043322
[jit_opt_after_cconv]: 0.00014812, [1]
    [Cycle 1]: 0.00014109, [11]
        [c_1]: 2.086e-05
        [parameter_eliminate]: 3.75998e-06
        [updatestate_depend_eliminate]: 6.82002e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [cse]: 2.194e-05
        [call_graph_tuple_transform]: 1.945e-05
        [tuple_list_get_item_eliminator]: 5.05001e-06
        [none_parameter_eliminate]: 1.55001e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 5.27001e-06
[remove_dup_value]: 1.34e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 5.56998e-06
[add_recomputation]: 5.246e-05
[cse_after_recomputation]: 2.032e-05, [1]
    [Cycle 1]: 1.44e-05, [1]
        [cse]: 8.78001e-06
[auto_monad_reorder]: 1.467e-05
[get_jit_bprop_graph]: 2.07001e-06
[rewriter_after_jit_bprop_graph]: 6.71e-06
[opt_after_jit_grad]: 0.00047015
[symbol_engine_optimizer]: 7.431e-05, [1]
    [Cycle 1]: 6.723e-05, [6]
        [build]: 3.81001e-06
        [elim_shapecalc]: 7.27997e-06
        [elim_not_effective]: 1.351e-05
        [opt_reshape]: 5.17999e-06
        [fold_const_symbol]: 8.77999e-06
        [renormalize]: 8.29983e-07
[validate]: 3.762e-05
[backend_pass]: 8.10018e-07
[task_emit]: 0.00708241
[execute]: 1.041e-05
Sums
    bootstrap                            :     0.000366s :  2.22%
    type_inference                       :     0.005228s : 31.78%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000069s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.23%
    rewriter_before_opt_a                :     0.000073s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.21%
    jit_opt_a.a_1                        :     0.000614s :  3.73%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000692s :  4.20%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000041s :  0.25%
    jit_opt_a.replace_applicator         :     0.000016s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000583s :  3.54%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000027s :  0.17%
    loop_unroll                          :     0.000433s :  2.63%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000052s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000470s :  2.86%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000038s :  0.23%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007082s : 43.05%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000200    23
 1.31% :     0.000003s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.97% :     0.000006s :      3: substitution.graph_param_transform
82.93% :     0.000166s :      4: substitution.inline
 1.70% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.12% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.18% :     0.000004s :      2: substitution.replace_old_param
 6.13% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005144     2
83.60% :     0.004300s :      1: type_inference.infer
16.40% :     0.000844s :      1: type_inference.specialize
------[replace.]   0.000068     6
69.57% :     0.000047s :      4: replace.inline
30.43% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000174     6
93.63% :     0.000163s :      4: match.inline
 6.37% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 2.97% :     0.000003s :     10: predicate.arithmetic_simplify
 1.44% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.48% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.16% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.39% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.44% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.48% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.84% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 2.20% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.79% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.22% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.80% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.97% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.00% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.67% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000612    12
53.67% :     0.000329s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.89% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.44% :     0.000254s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022317    76
 0.25% :     0.000055s :      1: add_recomputation
 0.32% :     0.000072s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.75% :     0.000391s :      1: bootstrap
 0.14% :     0.000030s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
24.99% :     0.005577s :      1: jit_opt_a
 0.68% :     0.000151s :      1: jit_opt_after_cconv
 0.24% :     0.000053s :      1: jit_opt_b
 1.98% :     0.000442s :      1: loop_unroll
 2.66% :     0.000593s :      1: mutable_eliminate
 3.70% :     0.000825s :     26: opt.transform.jit_opt_a
 0.21% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.15% :     0.000479s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000040s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.66% :     0.000371s :      1: renormalize.infer
 1.40% :     0.000313s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000038s :      1: rewriter_after_opt_a
 0.34% :     0.000076s :      1: rewriter_before_opt_a
 0.34% :     0.000077s :      1: symbol_engine_optimizer
31.83% :     0.007104s :      1: task_emit
23.52% :     0.005248s :      1: type_inference
 0.26% :     0.000058s :      1: validate


TotalTime = 0.025714, [33]
[bootstrap]: 0.00034998
[type_inference]: 0.00504562
[event_method]: 1.857e-05
[auto_monad]: 6.358e-05
[graph_reusing]: 5.72999e-06
[pre_auto_parallel]: 2.43e-06
[py_interpret_to_execute]: 4.535e-05
[rewriter_before_opt_a]: 7.163e-05
[expand_dump_flag]: 2.96001e-06
[jit_opt_a]: 0.00687985, [2]
    [Cycle 1]: 0.00181109, [27]
        [switch_simplify]: 5.294e-05
        [loop_unroll]: 3.242e-05
        [a_1]: 0.00057632
        [with_stream_mark]: 1.606e-05
        [recompute_prepare]: 6.73e-06
        [updatestate_depend_eliminate]: 3.19001e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 1.95001e-06
        [parameter_eliminate]: 1.24998e-06
        [specialize_transform]: 6.47001e-06
        [updatestate_useless_node_eliminater]: 5.70001e-06
        [accelerated_algorithm]: 1.477e-05
        [meta_shard_fg_expand]: 1.64998e-06
        [get_grad_eliminate_]: 6.23998e-06
        [merge_forward]: 3.55e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.268e-05
        [j_node_and_user_rematch]: 1.026e-05
        [meta_fg_expand]: 2.22999e-06
        [replace_old_param]: 8.60999e-06
        [inline_without_move]: 5.69e-06
        [renormalize]: 0.00081193
        [add_forward_monad_depend]: 6.16e-06
        [auto_monad_grad]: 2.48998e-06
        [auto_monad_eliminator]: 1.286e-05
        [cse]: 2.171e-05
        [replace_applicator]: 1.284e-05
    [Cycle 2]: 0.00033003, [27]
        [switch_simplify]: 6.32001e-06
        [loop_unroll]: 5.04e-06
        [a_1]: 9.413e-05
        [with_stream_mark]: 1.047e-05
        [recompute_prepare]: 5.10001e-06
        [updatestate_depend_eliminate]: 2.63998e-06
        [updatestate_assign_eliminate]: 2.10002e-06
        [updatestate_loads_eliminate]: 1.98002e-06
        [parameter_eliminate]: 1.30999e-06
        [specialize_transform]: 5.54e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.09998e-06
        [meta_shard_fg_expand]: 1.15999e-06
        [get_grad_eliminate_]: 4.68999e-06
        [merge_forward]: 2.43998e-06
        [cell_reuse_recompute_pass]: 1.73002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.26e-05
        [j_node_and_user_rematch]: 8.74003e-06
        [meta_fg_expand]: 1.78002e-06
        [replace_old_param]: 7.8e-06
        [inline_without_move]: 5.28002e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.50999e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 5.77999e-06
        [cse]: 1.134e-05
        [replace_applicator]: 5.62001e-06
[py_interpret_to_execute_after_opt_a]: 1.169e-05
[rewriter_after_opt_a]: 3.584e-05
[convert_after_rewriter]: 7.87003e-06
[order_py_execute_after_rewriter]: 5.17999e-06
[mutable_eliminate]: 0.00073271
[jit_opt_b]: 5.236e-05, [1]
    [Cycle 1]: 4.496e-05, [2]
        [frontend_op_eliminate]: 1.74e-05
        [inline_after_opt_a]: 1.51e-05
[cconv]: 2.593e-05
[loop_unroll]: 0.00045857
[jit_opt_after_cconv]: 0.00015066, [1]
    [Cycle 1]: 0.00014339, [11]
        [c_1]: 2.196e-05
        [parameter_eliminate]: 2.91999e-06
        [updatestate_depend_eliminate]: 6.41e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [cse]: 2.113e-05
        [call_graph_tuple_transform]: 2.176e-05
        [tuple_list_get_item_eliminator]: 5.35001e-06
        [none_parameter_eliminate]: 1.63002e-06
        [renormalize]: 5.50004e-07
        [switch_simplify]: 6.01e-06
[remove_dup_value]: 1.313e-05
[partial_unused_args_eliminate]: 2.51e-06
[environ_conv]: 5.99e-06
[add_recomputation]: 4.792e-05
[cse_after_recomputation]: 2.256e-05, [1]
    [Cycle 1]: 1.682e-05, [1]
        [cse]: 1.096e-05
[auto_monad_reorder]: 1.375e-05
[get_jit_bprop_graph]: 2.18002e-06
[rewriter_after_jit_bprop_graph]: 6.71999e-06
[opt_after_jit_grad]: 0.00050331
[symbol_engine_optimizer]: 7.601e-05, [1]
    [Cycle 1]: 6.998e-05, [6]
        [build]: 4.07e-06
        [elim_shapecalc]: 8.02e-06
        [elim_not_effective]: 1.328e-05
        [opt_reshape]: 5.81998e-06
        [fold_const_symbol]: 9.96998e-06
        [renormalize]: 3.89991e-07
[validate]: 4.395e-05
[backend_pass]: 1.47999e-06
[task_emit]: 0.0107669
[execute]: 9.13002e-06
Sums
    bootstrap                            :     0.000350s :  1.72%
    type_inference                       :     0.005046s : 24.81%
    event_method                         :     0.000019s :  0.09%
    auto_monad                           :     0.000064s :  0.31%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000045s :  0.22%
    rewriter_before_opt_a                :     0.000072s :  0.35%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000059s :  0.29%
    jit_opt_a.loop_unroll                :     0.000037s :  0.18%
    jit_opt_a.a_1                        :     0.000670s :  3.30%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000020s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.08%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.000812s :  3.99%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.09%
    jit_opt_a.cse                        :     0.000033s :  0.16%
    jit_opt_a.replace_applicator         :     0.000018s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.06%
    rewriter_after_opt_a                 :     0.000036s :  0.18%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000733s :  3.60%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.07%
    cconv                                :     0.000026s :  0.13%
    loop_unroll                          :     0.000459s :  2.26%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000013s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000048s :  0.24%
    cse_after_recomputation.cse          :     0.000011s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.03%
    opt_after_jit_grad                   :     0.000503s :  2.48%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000044s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.010767s : 52.95%
    execute                              :     0.000009s :  0.04%


Time group info:
------[substitution.]   0.000222    23
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.96% :     0.000002s :      2: substitution.fold_const_symbol
 2.70% :     0.000006s :      3: substitution.graph_param_transform
84.08% :     0.000186s :      4: substitution.inline
 1.71% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.98% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.69% :     0.000004s :      2: substitution.replace_old_param
 5.86% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004971     2
80.03% :     0.003978s :      1: type_inference.infer
19.97% :     0.000993s :      1: type_inference.specialize
------[replace.]   0.000077     6
66.00% :     0.000051s :      4: replace.inline
34.00% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000196     6
93.97% :     0.000184s :      4: match.inline
 6.03% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.94% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.50% :     0.000002s :     10: predicate.addn_zero_filter
 2.26% :     0.000003s :     10: predicate.arithmetic_simplify
 1.30% :     0.000002s :     10: predicate.cast_eliminate
 0.57% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.24% :     0.000002s :     10: predicate.depend_value_elim
 1.32% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.38% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.13% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.41% :     0.000002s :     10: predicate.environ_get_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 4.73% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.28% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.27% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.27% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 2.09% :     0.000003s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.81% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.57% :     0.000009s :     53: predicate.switch_simplify
 1.28% :     0.000002s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.56% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.37% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000678    12
44.22% :     0.000300s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.76% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.03% :     0.000339s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027525    76
 0.18% :     0.000051s :      1: add_recomputation
 0.24% :     0.000067s :      1: auto_monad
 0.06% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 1.34% :     0.000370s :      1: bootstrap
 0.10% :     0.000029s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000023s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
25.01% :     0.006885s :      1: jit_opt_a
 0.56% :     0.000154s :      1: jit_opt_after_cconv
 0.20% :     0.000055s :      1: jit_opt_b
 1.70% :     0.000468s :      1: loop_unroll
 2.70% :     0.000742s :      1: mutable_eliminate
 3.25% :     0.000895s :     26: opt.transform.jit_opt_a
 0.19% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000026s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.86% :     0.000513s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000048s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000016s :      1: remove_dup_value
 1.55% :     0.000426s :      1: renormalize.infer
 1.37% :     0.000378s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000039s :      1: rewriter_after_opt_a
 0.27% :     0.000074s :      1: rewriter_before_opt_a
 0.29% :     0.000078s :      1: symbol_engine_optimizer
39.19% :     0.010788s :      1: task_emit
18.39% :     0.005063s :      1: type_inference
 0.24% :     0.000066s :      1: validate


TotalTime = 0.0369464, [33]
[bootstrap]: 0.00337775
[type_inference]: 0.00843869
[event_method]: 1.885e-05
[auto_monad]: 6.805e-05
[graph_reusing]: 6.00002e-06
[pre_auto_parallel]: 3.26001e-06
[py_interpret_to_execute]: 4.441e-05
[rewriter_before_opt_a]: 7.635e-05
[expand_dump_flag]: 3.21999e-06
[jit_opt_a]: 0.0062645, [2]
    [Cycle 1]: 0.00194753, [27]
        [switch_simplify]: 5.699e-05
        [loop_unroll]: 3.257e-05
        [a_1]: 0.00061057
        [with_stream_mark]: 1.921e-05
        [recompute_prepare]: 9.51e-06
        [updatestate_depend_eliminate]: 3.78999e-06
        [updatestate_assign_eliminate]: 3.55e-06
        [updatestate_loads_eliminate]: 2.79001e-06
        [parameter_eliminate]: 2.02001e-06
        [specialize_transform]: 7.21999e-06
        [updatestate_useless_node_eliminater]: 5.39e-06
        [accelerated_algorithm]: 5.88002e-06
        [meta_shard_fg_expand]: 2.02999e-06
        [get_grad_eliminate_]: 6.24001e-06
        [merge_forward]: 3.82998e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.756e-05
        [j_node_and_user_rematch]: 1.043e-05
        [meta_fg_expand]: 2.14e-06
        [replace_old_param]: 9.68997e-06
        [inline_without_move]: 5.06997e-06
        [renormalize]: 0.00087316
        [add_forward_monad_depend]: 5.22e-06
        [auto_monad_grad]: 2.04e-06
        [auto_monad_eliminator]: 1.629e-05
        [cse]: 3.171e-05
        [replace_applicator]: 1.52e-05
    [Cycle 2]: 0.00033678, [27]
        [switch_simplify]: 6.16998e-06
        [loop_unroll]: 4.99e-06
        [a_1]: 9.382e-05
        [with_stream_mark]: 9.09998e-06
        [recompute_prepare]: 4.82e-06
        [updatestate_depend_eliminate]: 2.53998e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 1.88997e-06
        [parameter_eliminate]: 1.57001e-06
        [specialize_transform]: 5.19998e-06
        [updatestate_useless_node_eliminater]: 5.39e-06
        [accelerated_algorithm]: 6.09001e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 5.05001e-06
        [merge_forward]: 2.76e-06
        [cell_reuse_recompute_pass]: 1.52999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.428e-05
        [j_node_and_user_rematch]: 7.94002e-06
        [meta_fg_expand]: 1.88002e-06
        [replace_old_param]: 7.56999e-06
        [inline_without_move]: 4.95001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 5.04e-06
        [cse]: 1.143e-05
        [replace_applicator]: 5.51e-06
[py_interpret_to_execute_after_opt_a]: 1.167e-05
[rewriter_after_opt_a]: 3.883e-05
[convert_after_rewriter]: 7.69002e-06
[order_py_execute_after_rewriter]: 5.57999e-06
[mutable_eliminate]: 0.00066231
[jit_opt_b]: 5.302e-05, [1]
    [Cycle 1]: 4.54e-05, [2]
        [frontend_op_eliminate]: 1.819e-05
        [inline_after_opt_a]: 1.537e-05
[cconv]: 2.908e-05
[loop_unroll]: 0.00059098
[jit_opt_after_cconv]: 0.00014575, [1]
    [Cycle 1]: 0.00013881, [11]
        [c_1]: 2.162e-05
        [parameter_eliminate]: 2.69001e-06
        [updatestate_depend_eliminate]: 6.04999e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [cse]: 1.96e-05
        [call_graph_tuple_transform]: 1.982e-05
        [tuple_list_get_item_eliminator]: 5.15001e-06
        [none_parameter_eliminate]: 1.75001e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 5.47001e-06
[remove_dup_value]: 1.397e-05
[partial_unused_args_eliminate]: 2.89001e-06
[environ_conv]: 5.69e-06
[add_recomputation]: 5.739e-05
[cse_after_recomputation]: 2.103e-05, [1]
    [Cycle 1]: 1.532e-05, [1]
        [cse]: 8.74998e-06
[auto_monad_reorder]: 2.112e-05
[get_jit_bprop_graph]: 1.94e-06
[rewriter_after_jit_bprop_graph]: 6.56999e-06
[opt_after_jit_grad]: 0.00048297
[symbol_engine_optimizer]: 7.747e-05, [1]
    [Cycle 1]: 7.035e-05, [6]
        [build]: 4.84e-06
        [elim_shapecalc]: 7.95e-06
        [elim_not_effective]: 1.29e-05
        [opt_reshape]: 6.58e-06
        [fold_const_symbol]: 9.84001e-06
        [renormalize]: 1.8999e-07
[validate]: 4.403e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.0161174
[execute]: 6.02999e-06
Sums
    bootstrap                            :     0.003378s : 10.46%
    type_inference                       :     0.008439s : 26.13%
    event_method                         :     0.000019s :  0.06%
    auto_monad                           :     0.000068s :  0.21%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.14%
    rewriter_before_opt_a                :     0.000076s :  0.24%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000063s :  0.20%
    jit_opt_a.loop_unroll                :     0.000038s :  0.12%
    jit_opt_a.a_1                        :     0.000704s :  2.18%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000017s :  0.05%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.000873s :  2.70%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.07%
    jit_opt_a.cse                        :     0.000043s :  0.13%
    jit_opt_a.replace_applicator         :     0.000021s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000039s :  0.12%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000662s :  2.05%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.05%
    cconv                                :     0.000029s :  0.09%
    loop_unroll                          :     0.000591s :  1.83%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000014s :  0.04%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000057s :  0.18%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000021s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000483s :  1.50%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000044s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.016117s : 49.91%
    execute                              :     0.000006s :  0.02%


Time group info:
------[substitution.]   0.000248    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000002s :      2: substitution.fold_const_symbol
 2.27% :     0.000006s :      3: substitution.graph_param_transform
85.14% :     0.000211s :      4: substitution.inline
 1.42% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.48% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.74% :     0.000004s :      2: substitution.replace_old_param
 5.41% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008355     2
86.34% :     0.007214s :      1: type_inference.infer
13.66% :     0.001141s :      1: type_inference.specialize
------[replace.]   0.000079     6
66.69% :     0.000053s :      4: replace.inline
33.31% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000220     6
94.40% :     0.000208s :      4: match.inline
 5.60% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.64% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.30% :     0.000003s :     10: predicate.arithmetic_simplify
 1.15% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.62% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.07% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.74% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.45% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.65% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.91% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.20% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 2.04% :     0.000003s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     10: predicate.reduce_eliminate
 1.36% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.58% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.52% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 1.92% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.01% :     0.000009s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000002s :     10: predicate.transpose_eliminate
 1.51% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.89% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.00% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000806    12
47.74% :     0.000385s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.83% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.43% :     0.000382s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038851    76
 0.16% :     0.000061s :      1: add_recomputation
 0.18% :     0.000072s :      1: auto_monad
 0.06% :     0.000024s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 8.77% :     0.003407s :      1: bootstrap
 0.08% :     0.000032s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.06% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.06% :     0.000023s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
16.14% :     0.006269s :      1: jit_opt_a
 0.38% :     0.000149s :      1: jit_opt_after_cconv
 0.14% :     0.000056s :      1: jit_opt_b
 1.55% :     0.000600s :      1: loop_unroll
 1.73% :     0.000673s :      1: mutable_eliminate
 2.40% :     0.000932s :     26: opt.transform.jit_opt_a
 0.13% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000027s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.27% :     0.000492s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000047s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000016s :      1: remove_dup_value
 1.27% :     0.000492s :      1: renormalize.infer
 0.96% :     0.000373s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000042s :      1: rewriter_after_opt_a
 0.20% :     0.000079s :      1: rewriter_before_opt_a
 0.21% :     0.000080s :      1: symbol_engine_optimizer
41.52% :     0.016132s :      1: task_emit
21.77% :     0.008457s :      1: type_inference
 0.17% :     0.000067s :      1: validate


TotalTime = 0.0163822, [30]
[bootstrap]: 0.00030942
[type_inference]: 0.00689586
[event_method]: 2.656e-05
[auto_monad]: 8.688e-05
[graph_reusing]: 6.66e-06
[pre_auto_parallel]: 2.99999e-06
[py_interpret_to_execute]: 5.639e-05
[rewriter_before_opt_a]: 8.887e-05
[expand_dump_flag]: 3.01001e-06
[jit_opt_a]: 0.00648476, [2]
    [Cycle 1]: 0.00224177, [27]
        [switch_simplify]: 0.00017
        [loop_unroll]: 3.35e-05
        [a_1]: 0.00081663
        [with_stream_mark]: 2.509e-05
        [recompute_prepare]: 7.93001e-06
        [updatestate_depend_eliminate]: 4.53001e-06
        [updatestate_assign_eliminate]: 3.13e-06
        [updatestate_loads_eliminate]: 2.88e-06
        [parameter_eliminate]: 2.16e-06
        [specialize_transform]: 6.04001e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 5.56002e-06
        [meta_shard_fg_expand]: 2.59001e-06
        [get_grad_eliminate_]: 5.22999e-06
        [merge_forward]: 4.35999e-06
        [cell_reuse_recompute_pass]: 9.50007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.524e-05
        [j_node_and_user_rematch]: 9.08002e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 1.045e-05
        [inline_without_move]: 5.22999e-06
        [renormalize]: 0.00083328
        [add_forward_monad_depend]: 7.16001e-06
        [auto_monad_grad]: 2.44001e-06
        [auto_monad_eliminator]: 1.622e-05
        [cse]: 3.207e-05
        [replace_applicator]: 1.282e-05
    [Cycle 2]: 0.00033088, [27]
        [switch_simplify]: 5.64998e-06
        [loop_unroll]: 4.52e-06
        [a_1]: 8.942e-05
        [with_stream_mark]: 1.849e-05
        [recompute_prepare]: 5.57001e-06
        [updatestate_depend_eliminate]: 3.40998e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 1.77001e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 5.12999e-06
        [updatestate_useless_node_eliminater]: 4.81002e-06
        [accelerated_algorithm]: 5.75001e-06
        [meta_shard_fg_expand]: 1.27999e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 3.11001e-06
        [cell_reuse_recompute_pass]: 1.62001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.447e-05
        [j_node_and_user_rematch]: 8.10999e-06
        [meta_fg_expand]: 1.63002e-06
        [replace_old_param]: 7.71999e-06
        [inline_without_move]: 4.90001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.42e-06
        [auto_monad_grad]: 1.17e-06
        [auto_monad_eliminator]: 5.32001e-06
        [cse]: 1.105e-05
        [replace_applicator]: 5.24998e-06
[py_interpret_to_execute_after_opt_a]: 1.239e-05
[rewriter_after_opt_a]: 3.579e-05
[convert_after_rewriter]: 6.94999e-06
[order_py_execute_after_rewriter]: 4.67e-06
[mutable_eliminate]: 0.00076739
[jit_opt_b]: 4.874e-05, [1]
    [Cycle 1]: 4.065e-05, [2]
        [frontend_op_eliminate]: 1.465e-05
        [inline_after_opt_a]: 1.473e-05
[cconv]: 2.787e-05
[loop_unroll]: 0.00044576
[jit_opt_after_cconv]: 0.00014794, [1]
    [Cycle 1]: 0.00014052, [11]
        [c_1]: 2.153e-05
        [parameter_eliminate]: 3.21001e-06
        [updatestate_depend_eliminate]: 5.94999e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [cse]: 2.102e-05
        [call_graph_tuple_transform]: 1.999e-05
        [tuple_list_get_item_eliminator]: 5.47001e-06
        [none_parameter_eliminate]: 1.86003e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 6.14999e-06
[remove_dup_value]: 1.376e-05
[partial_unused_args_eliminate]: 2.54999e-06
[environ_conv]: 5.88998e-06
[add_recomputation]: 5.042e-05
[cse_after_recomputation]: 2.053e-05, [1]
    [Cycle 1]: 1.439e-05, [1]
        [cse]: 8.99e-06
[auto_monad_reorder]: 1.507e-05
[get_jit_bprop_graph]: 2.09e-06
[rewriter_after_jit_bprop_graph]: 6.09999e-06
[opt_after_jit_grad]: 0.00050399
[symbol_engine_optimizer]: 7.404e-05, [1]
    [Cycle 1]: 6.729e-05, [6]
        [build]: 2.81e-06
        [elim_shapecalc]: 8.00999e-06
        [elim_not_effective]: 1.354e-05
        [opt_reshape]: 5.50001e-06
        [fold_const_symbol]: 8.80001e-06
        [renormalize]: 5.79981e-07
[validate]: 3.644e-05
Sums
    bootstrap                            :     0.000309s :  2.61%
    type_inference                       :     0.006896s : 58.23%
    event_method                         :     0.000027s :  0.22%
    auto_monad                           :     0.000087s :  0.73%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000056s :  0.48%
    rewriter_before_opt_a                :     0.000089s :  0.75%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000176s :  1.48%
    jit_opt_a.loop_unroll                :     0.000038s :  0.32%
    jit_opt_a.a_1                        :     0.000906s :  7.65%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.37%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000833s :  7.04%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.18%
    jit_opt_a.cse                        :     0.000043s :  0.36%
    jit_opt_a.replace_applicator         :     0.000018s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.10%
    rewriter_after_opt_a                 :     0.000036s :  0.30%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000767s :  6.48%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000028s :  0.24%
    loop_unroll                          :     0.000446s :  3.76%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000014s :  0.12%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000050s :  0.43%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000015s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000504s :  4.26%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.31%


Time group info:
------[substitution.]   0.000407    26
 0.58% :     0.000002s :      2: substitution.elim_not_effective
 0.39% :     0.000002s :      2: substitution.fold_const_symbol
 1.43% :     0.000006s :      3: substitution.graph_param_transform
88.80% :     0.000361s :      6: substitution.inline
 0.81% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.23% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.28% :     0.000005s :      2: substitution.replace_old_param
 2.32% :     0.000009s :      1: substitution.switch_simplify
 3.17% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006812     2
83.61% :     0.005696s :      1: type_inference.infer
16.39% :     0.001117s :      1: type_inference.specialize
------[replace.]   0.000149     9
47.24% :     0.000070s :      6: replace.inline
31.38% :     0.000047s :      1: replace.switch_simplify
21.38% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000378     9
94.64% :     0.000358s :      6: match.inline
 2.26% :     0.000009s :      1: match.switch_simplify
 3.10% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000156   801
 1.03% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.12% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     12: predicate.addn_check_dump
 1.40% :     0.000002s :     12: predicate.addn_zero_filter
 1.67% :     0.000003s :     12: predicate.arithmetic_simplify
 1.10% :     0.000002s :     12: predicate.cast_eliminate
 0.37% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     12: predicate.compare_switch_simplify
 1.15% :     0.000002s :     12: predicate.depend_value_elim
 0.97% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.08% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.07% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.58% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.22% :     0.000000s :      3: predicate.elim_not_effective
 0.42% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.01% :     0.000002s :     12: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     12: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     12: predicate.environ_get_eliminate
 0.96% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.17% :     0.000000s :      3: predicate.fold_const_symbol
 0.69% :     0.000001s :      6: predicate.get_grad_eliminate
 0.34% :     0.000001s :      3: predicate.graph_param_transform
 4.95% :     0.000008s :     26: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.27% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.42% :     0.000002s :     17: predicate.load_eliminater
 1.24% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.47% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.51% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000002s :     12: predicate.merge_addn
 1.05% :     0.000002s :     12: predicate.minmaximum_grad
 1.11% :     0.000002s :      3: predicate.mutable_eliminate
 0.28% :     0.000000s :      3: predicate.opt_reshape
 1.90% :     0.000003s :     17: predicate.partial_eliminate
 1.17% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.30% :     0.000002s :     12: predicate.reduce_eliminate
 1.23% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.32% :     0.000002s :     20: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.45% :     0.000001s :      3: predicate.reset_defer_inline
 1.16% :     0.000002s :     12: predicate.reshape_eliminate
 1.31% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.58% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.05% :     0.000002s :     12: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.60% :     0.000001s :      6: predicate.special_op_eliminate
 0.74% :     0.000001s :      6: predicate.specialize_transform
 1.08% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.08% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.31% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000004s :     20: predicate.switch_defer_inline
 2.06% :     0.000003s :     20: predicate.switch_layer_defer_inline
22.34% :     0.000035s :     65: predicate.switch_simplify
 1.04% :     0.000002s :     12: predicate.tile_eliminate
 1.07% :     0.000002s :     12: predicate.transpose_eliminate
 1.10% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.03% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.39% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.34% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.26% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.39% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.36% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.30% :     0.000002s :     12: predicate.value_based_eliminate
 0.26% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000909    15
43.53% :     0.000396s :      6: func_graph_cloner_run.FuncGraphClonerGraph
21.09% :     0.000192s :      1: func_graph_cloner_run.FuncGraphClonerNode
35.38% :     0.000322s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018502    72
 0.29% :     0.000053s :      1: add_recomputation
 0.49% :     0.000091s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 1.80% :     0.000332s :      1: bootstrap
 0.17% :     0.000031s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.17% :     0.000032s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
35.07% :     0.006489s :      1: jit_opt_a
 0.81% :     0.000151s :      1: jit_opt_after_cconv
 0.28% :     0.000052s :      1: jit_opt_b
 2.45% :     0.000454s :      1: loop_unroll
 4.20% :     0.000777s :      1: mutable_eliminate
 6.69% :     0.001238s :     26: opt.transform.jit_opt_a
 0.26% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.77% :     0.000513s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.32% :     0.000060s :      1: py_interpret_to_execute
 0.08% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 2.61% :     0.000483s :      1: renormalize.infer
 1.84% :     0.000341s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000039s :      1: rewriter_after_opt_a
 0.50% :     0.000092s :      1: rewriter_before_opt_a
 0.41% :     0.000077s :      1: symbol_engine_optimizer
37.40% :     0.006920s :      1: type_inference


TotalTime = 0.0187189, [33]
[bootstrap]: 0.00028359
[type_inference]: 0.00455458
[event_method]: 1.666e-05
[auto_monad]: 6.134e-05
[graph_reusing]: 5.95002e-06
[pre_auto_parallel]: 2.49999e-06
[py_interpret_to_execute]: 2.962e-05
[rewriter_before_opt_a]: 7.162e-05
[expand_dump_flag]: 2.89001e-06
[jit_opt_a]: 0.00525904, [2]
    [Cycle 1]: 0.0016393, [27]
        [switch_simplify]: 5.698e-05
        [loop_unroll]: 2.789e-05
        [a_1]: 0.00052446
        [with_stream_mark]: 2.09e-05
        [recompute_prepare]: 7.03e-06
        [updatestate_depend_eliminate]: 3.83001e-06
        [updatestate_assign_eliminate]: 3.11001e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 5.86998e-06
        [updatestate_useless_node_eliminater]: 4.97999e-06
        [accelerated_algorithm]: 5.59e-06
        [meta_shard_fg_expand]: 1.92001e-06
        [get_grad_eliminate_]: 5.37999e-06
        [merge_forward]: 3.35e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.365e-05
        [j_node_and_user_rematch]: 8.69e-06
        [meta_fg_expand]: 2.16998e-06
        [replace_old_param]: 8.97e-06
        [inline_without_move]: 5.42001e-06
        [renormalize]: 0.00067522
        [add_forward_monad_depend]: 7e-06
        [auto_monad_grad]: 2.54001e-06
        [auto_monad_eliminator]: 1.58e-05
        [cse]: 3.195e-05
        [replace_applicator]: 1.338e-05
    [Cycle 2]: 0.000316, [27]
        [switch_simplify]: 5.62999e-06
        [loop_unroll]: 4.63001e-06
        [a_1]: 8.602e-05
        [with_stream_mark]: 1.025e-05
        [recompute_prepare]: 5.39998e-06
        [updatestate_depend_eliminate]: 3.03e-06
        [updatestate_assign_eliminate]: 1.96998e-06
        [updatestate_loads_eliminate]: 1.76e-06
        [parameter_eliminate]: 1.23002e-06
        [specialize_transform]: 5.17e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 4.82e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 4.35999e-06
        [merge_forward]: 3.01001e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.335e-05
        [j_node_and_user_rematch]: 7.64002e-06
        [meta_fg_expand]: 1.75001e-06
        [replace_old_param]: 8e-06
        [inline_without_move]: 4.84998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 4.97e-06
        [cse]: 1.13e-05
        [replace_applicator]: 5.00001e-06
[py_interpret_to_execute_after_opt_a]: 1.347e-05
[rewriter_after_opt_a]: 3.814e-05
[convert_after_rewriter]: 7.39002e-06
[order_py_execute_after_rewriter]: 5.56e-06
[mutable_eliminate]: 0.00062696
[jit_opt_b]: 5.014e-05, [1]
    [Cycle 1]: 4.203e-05, [2]
        [frontend_op_eliminate]: 1.666e-05
        [inline_after_opt_a]: 1.417e-05
[cconv]: 2.668e-05
[loop_unroll]: 0.00042547
[jit_opt_after_cconv]: 0.00013636, [1]
    [Cycle 1]: 0.00013029, [11]
        [c_1]: 1.938e-05
        [parameter_eliminate]: 2.72001e-06
        [updatestate_depend_eliminate]: 5.59e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.897e-05
        [call_graph_tuple_transform]: 1.878e-05
        [tuple_list_get_item_eliminator]: 5.24e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 5.59987e-07
        [switch_simplify]: 5.02e-06
[remove_dup_value]: 1.319e-05
[partial_unused_args_eliminate]: 2.36998e-06
[environ_conv]: 5.24e-06
[add_recomputation]: 5.231e-05
[cse_after_recomputation]: 1.908e-05, [1]
    [Cycle 1]: 1.403e-05, [1]
        [cse]: 8.33999e-06
[auto_monad_reorder]: 1.547e-05
[get_jit_bprop_graph]: 2.14e-06
[rewriter_after_jit_bprop_graph]: 5.46e-06
[opt_after_jit_grad]: 0.00045437
[symbol_engine_optimizer]: 7.037e-05, [1]
    [Cycle 1]: 6.394e-05, [6]
        [build]: 4.34002e-06
        [elim_shapecalc]: 6.77002e-06
        [elim_not_effective]: 1.272e-05
        [opt_reshape]: 5.10001e-06
        [fold_const_symbol]: 7.92998e-06
        [renormalize]: 7.89994e-07
[validate]: 3.713e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.00620304
[execute]: 9.41e-06
Sums
    bootstrap                            :     0.000284s :  1.92%
    type_inference                       :     0.004555s : 30.78%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000061s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.20%
    rewriter_before_opt_a                :     0.000072s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.42%
    jit_opt_a.loop_unroll                :     0.000033s :  0.22%
    jit_opt_a.a_1                        :     0.000610s :  4.13%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000675s :  4.56%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.14%
    jit_opt_a.cse                        :     0.000043s :  0.29%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.09%
    rewriter_after_opt_a                 :     0.000038s :  0.26%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000627s :  4.24%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000027s :  0.18%
    loop_unroll                          :     0.000425s :  2.88%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000052s :  0.35%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000454s :  3.07%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000037s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006203s : 41.92%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000211    23
 1.11% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.54% :     0.000005s :      3: substitution.graph_param_transform
83.60% :     0.000177s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.37% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.52% :     0.000005s :      2: substitution.replace_old_param
 5.71% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004486     2
82.99% :     0.003723s :      1: type_inference.infer
17.01% :     0.000763s :      1: type_inference.specialize
------[replace.]   0.000065     6
68.18% :     0.000044s :      4: replace.inline
31.82% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000185     6
94.05% :     0.000174s :      4: match.inline
 5.95% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.35% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.15% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000001s :     10: predicate.addn_zero_filter
 2.00% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000001s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.63% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.39% :     0.000002s :      6: predicate.less_batch_normalization
 1.70% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.87% :     0.000002s :     15: predicate.load_eliminater
 1.55% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.76% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 2.05% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.65% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000001s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.02% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.62% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.67% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.36% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.28% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.14% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000561    12
51.17% :     0.000287s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.63% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.20% :     0.000237s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020310    76
 0.27% :     0.000056s :      1: add_recomputation
 0.32% :     0.000065s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.50% :     0.000304s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.92% :     0.005263s :      1: jit_opt_a
 0.68% :     0.000139s :      1: jit_opt_after_cconv
 0.26% :     0.000053s :      1: jit_opt_b
 2.13% :     0.000433s :      1: loop_unroll
 3.13% :     0.000636s :      1: mutable_eliminate
 4.04% :     0.000820s :     26: opt.transform.jit_opt_a
 0.22% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.28% :     0.000463s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000032s :      1: py_interpret_to_execute
 0.08% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.58% :     0.000320s :      1: renormalize.infer
 1.71% :     0.000346s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000041s :      1: rewriter_after_opt_a
 0.37% :     0.000074s :      1: rewriter_before_opt_a
 0.36% :     0.000073s :      1: symbol_engine_optimizer
30.63% :     0.006221s :      1: task_emit
22.49% :     0.004568s :      1: type_inference
 0.29% :     0.000060s :      1: validate


TotalTime = 0.0232125, [33]
[bootstrap]: 0.0003253
[type_inference]: 0.00643519
[event_method]: 2.086e-05
[auto_monad]: 7.378e-05
[graph_reusing]: 6.05002e-06
[pre_auto_parallel]: 2.78e-06
[py_interpret_to_execute]: 5.611e-05
[rewriter_before_opt_a]: 8.169e-05
[expand_dump_flag]: 3.4e-06
[jit_opt_a]: 0.00688624, [2]
    [Cycle 1]: 0.0020267, [27]
        [switch_simplify]: 6.689e-05
        [loop_unroll]: 2.927e-05
        [a_1]: 0.00059408
        [with_stream_mark]: 2.409e-05
        [recompute_prepare]: 7.35e-06
        [updatestate_depend_eliminate]: 3.65e-06
        [updatestate_assign_eliminate]: 2.86999e-06
        [updatestate_loads_eliminate]: 2.76e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 6.04999e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.44998e-06
        [meta_shard_fg_expand]: 2.44999e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 4.37e-06
        [cell_reuse_recompute_pass]: 1.47999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.539e-05
        [j_node_and_user_rematch]: 9.55001e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 9.68997e-06
        [inline_without_move]: 5.82001e-06
        [renormalize]: 0.0009649
        [add_forward_monad_depend]: 6.06e-06
        [auto_monad_grad]: 2.59001e-06
        [auto_monad_eliminator]: 1.569e-05
        [cse]: 3.064e-05
        [replace_applicator]: 1.314e-05
    [Cycle 2]: 0.0003608, [27]
        [switch_simplify]: 6.78e-06
        [loop_unroll]: 6.16998e-06
        [a_1]: 9.951e-05
        [with_stream_mark]: 9.67001e-06
        [recompute_prepare]: 5.46e-06
        [updatestate_depend_eliminate]: 2.86e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 5.55001e-06
        [accelerated_algorithm]: 5.34998e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 5.35001e-06
        [merge_forward]: 3.13998e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.463e-05
        [j_node_and_user_rematch]: 9.19998e-06
        [meta_fg_expand]: 1.65001e-06
        [replace_old_param]: 9.02e-06
        [inline_without_move]: 5.15001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 9.5999e-07
        [auto_monad_eliminator]: 5.54e-06
        [cse]: 1.003e-05
        [replace_applicator]: 5.74e-06
[py_interpret_to_execute_after_opt_a]: 1.166e-05
[rewriter_after_opt_a]: 4.319e-05
[convert_after_rewriter]: 7.79002e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.00066896
[jit_opt_b]: 5.743e-05, [1]
    [Cycle 1]: 4.919e-05, [2]
        [frontend_op_eliminate]: 1.885e-05
        [inline_after_opt_a]: 1.647e-05
[cconv]: 3.057e-05
[loop_unroll]: 0.00047097
[jit_opt_after_cconv]: 0.000154, [1]
    [Cycle 1]: 0.00014685, [11]
        [c_1]: 2.222e-05
        [parameter_eliminate]: 3.10998e-06
        [updatestate_depend_eliminate]: 6.24001e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 2.23002e-06
        [cse]: 2.088e-05
        [call_graph_tuple_transform]: 2.215e-05
        [tuple_list_get_item_eliminator]: 6.23998e-06
        [none_parameter_eliminate]: 1.61002e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 6.03998e-06
[remove_dup_value]: 1.548e-05
[partial_unused_args_eliminate]: 3.10998e-06
[environ_conv]: 5.99e-06
[add_recomputation]: 5.308e-05
[cse_after_recomputation]: 2.267e-05, [1]
    [Cycle 1]: 1.631e-05, [1]
        [cse]: 9.76e-06
[auto_monad_reorder]: 1.608e-05
[get_jit_bprop_graph]: 2.48998e-06
[rewriter_after_jit_bprop_graph]: 6.38e-06
[opt_after_jit_grad]: 0.00086038
[symbol_engine_optimizer]: 8.54e-05, [1]
    [Cycle 1]: 7.651e-05, [6]
        [build]: 5.15999e-06
        [elim_shapecalc]: 8.43999e-06
        [elim_not_effective]: 1.557e-05
        [opt_reshape]: 7.23999e-06
        [fold_const_symbol]: 9.17001e-06
        [renormalize]: 9.50007e-07
[validate]: 3.996e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.0064932
[execute]: 9.34e-06
Sums
    bootstrap                            :     0.000325s :  1.81%
    type_inference                       :     0.006435s : 35.75%
    event_method                         :     0.000021s :  0.12%
    auto_monad                           :     0.000074s :  0.41%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000056s :  0.31%
    rewriter_before_opt_a                :     0.000082s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.41%
    jit_opt_a.loop_unroll                :     0.000035s :  0.20%
    jit_opt_a.a_1                        :     0.000694s :  3.85%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000965s :  5.36%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.12%
    jit_opt_a.cse                        :     0.000041s :  0.23%
    jit_opt_a.replace_applicator         :     0.000019s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.06%
    rewriter_after_opt_a                 :     0.000043s :  0.24%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000669s :  3.72%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000031s :  0.17%
    loop_unroll                          :     0.000471s :  2.62%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000053s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000860s :  4.78%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006493s : 36.07%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000259    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.54% :     0.000007s :      3: substitution.graph_param_transform
85.46% :     0.000221s :      4: substitution.inline
 1.55% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.98% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.95% :     0.000005s :      2: substitution.replace_old_param
 5.17% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006336     2
81.26% :     0.005148s :      1: type_inference.infer
18.74% :     0.001188s :      1: type_inference.specialize
------[replace.]   0.000073     6
66.75% :     0.000049s :      4: replace.inline
33.25% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000230     6
94.72% :     0.000218s :      4: match.inline
 5.28% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.66% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.23% :     0.000002s :     10: predicate.addn_zero_filter
 1.86% :     0.000002s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.63% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000002s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.45% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.19% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.36% :     0.000007s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.62% :     0.000002s :     15: predicate.load_eliminater
 1.76% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.81% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.23% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.58% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.97% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.10% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.30% :     0.000002s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.53% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.72% :     0.000010s :     53: predicate.switch_simplify
 1.34% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.52% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.83% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000777    12
43.93% :     0.000341s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.68% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.39% :     0.000384s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025205    76
 0.22% :     0.000056s :      1: add_recomputation
 0.31% :     0.000078s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.37% :     0.000344s :      1: bootstrap
 0.13% :     0.000034s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000025s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
27.34% :     0.006891s :      1: jit_opt_a
 0.62% :     0.000157s :      1: jit_opt_after_cconv
 0.24% :     0.000061s :      1: jit_opt_b
 1.90% :     0.000480s :      1: loop_unroll
 2.69% :     0.000679s :      1: mutable_eliminate
 3.67% :     0.000926s :     26: opt.transform.jit_opt_a
 0.21% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.46% :     0.000873s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000060s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.20% :     0.000554s :      1: renormalize.infer
 1.59% :     0.000401s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000047s :      1: rewriter_after_opt_a
 0.34% :     0.000086s :      1: rewriter_before_opt_a
 0.35% :     0.000088s :      1: symbol_engine_optimizer
25.84% :     0.006512s :      1: task_emit
25.62% :     0.006457s :      1: type_inference
 0.25% :     0.000064s :      1: validate


TotalTime = 0.0167507, [30]
[bootstrap]: 0.00038133
[type_inference]: 0.00750601
[event_method]: 2.432e-05
[auto_monad]: 8.333e-05
[graph_reusing]: 6.68998e-06
[pre_auto_parallel]: 2.78e-06
[py_interpret_to_execute]: 4.995e-05
[rewriter_before_opt_a]: 8.982e-05
[expand_dump_flag]: 3.03e-06
[jit_opt_a]: 0.00622591, [2]
    [Cycle 1]: 0.0021882, [27]
        [switch_simplify]: 0.0001427
        [loop_unroll]: 3.354e-05
        [a_1]: 0.00077491
        [with_stream_mark]: 2.785e-05
        [recompute_prepare]: 9.44e-06
        [updatestate_depend_eliminate]: 4.70001e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 2.48998e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 6.37001e-06
        [updatestate_useless_node_eliminater]: 5.13002e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 5.12999e-06
        [merge_forward]: 3.73001e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.63e-05
        [j_node_and_user_rematch]: 9.64999e-06
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 9.75002e-06
        [inline_without_move]: 5.89e-06
        [renormalize]: 0.00083266
        [add_forward_monad_depend]: 7.69002e-06
        [auto_monad_grad]: 2.44999e-06
        [auto_monad_eliminator]: 1.683e-05
        [cse]: 3.239e-05
        [replace_applicator]: 1.576e-05
    [Cycle 2]: 0.00032337, [27]
        [switch_simplify]: 5.84e-06
        [loop_unroll]: 5.03002e-06
        [a_1]: 8.829e-05
        [with_stream_mark]: 1.254e-05
        [recompute_prepare]: 5.22999e-06
        [updatestate_depend_eliminate]: 2.81e-06
        [updatestate_assign_eliminate]: 2.84001e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.62e-06
        [accelerated_algorithm]: 4.63001e-06
        [meta_shard_fg_expand]: 1.85001e-06
        [get_grad_eliminate_]: 4.37e-06
        [merge_forward]: 3.11001e-06
        [cell_reuse_recompute_pass]: 1.69e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.498e-05
        [j_node_and_user_rematch]: 8.07e-06
        [meta_fg_expand]: 1.37e-06
        [replace_old_param]: 7.45e-06
        [inline_without_move]: 4.87998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.59e-06
        [auto_monad_grad]: 1.24998e-06
        [auto_monad_eliminator]: 5.34e-06
        [cse]: 1.014e-05
        [replace_applicator]: 4.97999e-06
[py_interpret_to_execute_after_opt_a]: 3.451e-05
[rewriter_after_opt_a]: 3.722e-05
[convert_after_rewriter]: 6.49999e-06
[order_py_execute_after_rewriter]: 4.72998e-06
[mutable_eliminate]: 0.00068123
[jit_opt_b]: 5.055e-05, [1]
    [Cycle 1]: 4.297e-05, [2]
        [frontend_op_eliminate]: 1.497e-05
        [inline_after_opt_a]: 1.56e-05
[cconv]: 3.024e-05
[loop_unroll]: 0.00045174
[jit_opt_after_cconv]: 0.00015289, [1]
    [Cycle 1]: 0.00014639, [11]
        [c_1]: 2.12e-05
        [parameter_eliminate]: 4.05e-06
        [updatestate_depend_eliminate]: 8.43001e-06
        [updatestate_assign_eliminate]: 2.66999e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 2.251e-05
        [call_graph_tuple_transform]: 2.211e-05
        [tuple_list_get_item_eliminator]: 6.07999e-06
        [none_parameter_eliminate]: 1.70001e-06
        [renormalize]: 9.5999e-07
        [switch_simplify]: 4.99e-06
[remove_dup_value]: 1.412e-05
[partial_unused_args_eliminate]: 2.15002e-06
[environ_conv]: 5.25999e-06
[add_recomputation]: 4.909e-05
[cse_after_recomputation]: 2.076e-05, [1]
    [Cycle 1]: 1.507e-05, [1]
        [cse]: 8.85001e-06
[auto_monad_reorder]: 1.558e-05
[get_jit_bprop_graph]: 2.23002e-06
[rewriter_after_jit_bprop_graph]: 7.52002e-06
[opt_after_jit_grad]: 0.00049577
[symbol_engine_optimizer]: 7.469e-05, [1]
    [Cycle 1]: 6.848e-05, [6]
        [build]: 3.71999e-06
        [elim_shapecalc]: 7.66001e-06
        [elim_not_effective]: 1.358e-05
        [opt_reshape]: 5.45001e-06
        [fold_const_symbol]: 8.52e-06
        [renormalize]: 5.60016e-07
[validate]: 3.781e-05
Sums
    bootstrap                            :     0.000381s :  3.08%
    type_inference                       :     0.007506s : 60.58%
    event_method                         :     0.000024s :  0.20%
    auto_monad                           :     0.000083s :  0.67%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000050s :  0.40%
    rewriter_before_opt_a                :     0.000090s :  0.72%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000149s :  1.20%
    jit_opt_a.loop_unroll                :     0.000039s :  0.31%
    jit_opt_a.a_1                        :     0.000863s :  6.97%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.08%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.14%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.000833s :  6.72%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.18%
    jit_opt_a.cse                        :     0.000043s :  0.34%
    jit_opt_a.replace_applicator         :     0.000021s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000035s :  0.28%
    rewriter_after_opt_a                 :     0.000037s :  0.30%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000681s :  5.50%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.13%
    cconv                                :     0.000030s :  0.24%
    loop_unroll                          :     0.000452s :  3.65%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.17%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000049s :  0.40%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000016s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.06%
    opt_after_jit_grad                   :     0.000496s :  4.00%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.31%


Time group info:
------[substitution.]   0.000350    26
 0.49% :     0.000002s :      2: substitution.elim_not_effective
 0.38% :     0.000001s :      2: substitution.fold_const_symbol
 1.71% :     0.000006s :      3: substitution.graph_param_transform
86.27% :     0.000302s :      6: substitution.inline
 1.02% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.43% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.68% :     0.000006s :      2: substitution.replace_old_param
 2.87% :     0.000010s :      1: substitution.switch_simplify
 4.15% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007427     2
85.21% :     0.006328s :      1: type_inference.infer
14.79% :     0.001098s :      1: type_inference.specialize
------[replace.]   0.000156     9
49.60% :     0.000077s :      6: replace.inline
29.03% :     0.000045s :      1: replace.switch_simplify
21.37% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000320     9
93.01% :     0.000298s :      6: match.inline
 2.91% :     0.000009s :      1: match.switch_simplify
 4.08% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   801
 1.30% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.05% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     12: predicate.addn_check_dump
 1.85% :     0.000003s :     12: predicate.addn_zero_filter
 2.20% :     0.000003s :     12: predicate.arithmetic_simplify
 1.18% :     0.000002s :     12: predicate.cast_eliminate
 0.38% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     12: predicate.compare_switch_simplify
 1.51% :     0.000002s :     12: predicate.depend_value_elim
 1.12% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.14% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.68% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.36% :     0.000002s :     12: predicate.environ_get_eliminate
 1.10% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.79% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 6.23% :     0.000008s :     26: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.49% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000002s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     17: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.11% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     12: predicate.merge_addn
 1.10% :     0.000002s :     12: predicate.minmaximum_grad
 2.12% :     0.000003s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     17: predicate.partial_eliminate
 1.30% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.41% :     0.000002s :     12: predicate.reduce_eliminate
 1.36% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.50% :     0.000002s :     20: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.60% :     0.000001s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     12: predicate.reshape_eliminate
 1.16% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     12: predicate.same_eliminate
 0.89% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.71% :     0.000001s :      6: predicate.special_op_eliminate
 0.85% :     0.000001s :      6: predicate.specialize_transform
 1.22% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.61% :     0.000004s :     20: predicate.switch_defer_inline
 2.15% :     0.000003s :     20: predicate.switch_layer_defer_inline
 9.00% :     0.000012s :     65: predicate.switch_simplify
 1.25% :     0.000002s :     12: predicate.tile_eliminate
 1.22% :     0.000002s :     12: predicate.transpose_eliminate
 1.30% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.43% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.33% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.59% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     12: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.45% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000900    15
48.90% :     0.000440s :      6: func_graph_cloner_run.FuncGraphClonerGraph
15.30% :     0.000138s :      1: func_graph_cloner_run.FuncGraphClonerNode
35.80% :     0.000322s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018811    72
 0.28% :     0.000052s :      1: add_recomputation
 0.47% :     0.000088s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 2.16% :     0.000406s :      1: bootstrap
 0.18% :     0.000033s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.16% :     0.000029s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.12% :     0.006230s :      1: jit_opt_a
 0.83% :     0.000156s :      1: jit_opt_after_cconv
 0.28% :     0.000053s :      1: jit_opt_b
 2.46% :     0.000462s :      1: loop_unroll
 3.69% :     0.000695s :      1: mutable_eliminate
 6.24% :     0.001174s :     26: opt.transform.jit_opt_a
 0.27% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.68% :     0.000505s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000053s :      1: py_interpret_to_execute
 0.20% :     0.000037s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 2.63% :     0.000494s :      1: renormalize.infer
 1.74% :     0.000328s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000041s :      1: rewriter_after_opt_a
 0.50% :     0.000093s :      1: rewriter_before_opt_a
 0.41% :     0.000077s :      1: symbol_engine_optimizer
40.00% :     0.007525s :      1: type_inference


TotalTime = 0.0259848, [33]
[bootstrap]: 0.00039731
[type_inference]: 0.00622861
[event_method]: 1.902e-05
[auto_monad]: 6.584e-05
[graph_reusing]: 6.62002e-06
[pre_auto_parallel]: 2.52001e-06
[py_interpret_to_execute]: 4.74e-05
[rewriter_before_opt_a]: 7.329e-05
[expand_dump_flag]: 3.49001e-06
[jit_opt_a]: 0.00742297, [2]
    [Cycle 1]: 0.00181767, [27]
        [switch_simplify]: 5.734e-05
        [loop_unroll]: 3.082e-05
        [a_1]: 0.00058049
        [with_stream_mark]: 1.597e-05
        [recompute_prepare]: 7.58999e-06
        [updatestate_depend_eliminate]: 3.21999e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 6.56e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 5.99e-06
        [meta_shard_fg_expand]: 1.71e-06
        [get_grad_eliminate_]: 5.94999e-06
        [merge_forward]: 3.17002e-06
        [cell_reuse_recompute_pass]: 1.14003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.353e-05
        [j_node_and_user_rematch]: 1.048e-05
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 8.98002e-06
        [inline_without_move]: 5.84e-06
        [renormalize]: 0.00082337
        [add_forward_monad_depend]: 5.54998e-06
        [auto_monad_grad]: 1.82999e-06
        [auto_monad_eliminator]: 1.153e-05
        [cse]: 1.946e-05
        [replace_applicator]: 1.341e-05
    [Cycle 2]: 0.00035576, [27]
        [switch_simplify]: 6.06e-06
        [loop_unroll]: 5.47001e-06
        [a_1]: 9.85e-05
        [with_stream_mark]: 8.85999e-06
        [recompute_prepare]: 5.83997e-06
        [updatestate_depend_eliminate]: 2.78e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 9.00007e-07
        [specialize_transform]: 5.92999e-06
        [updatestate_useless_node_eliminater]: 6.16e-06
        [accelerated_algorithm]: 6.61e-06
        [meta_shard_fg_expand]: 1.42999e-06
        [get_grad_eliminate_]: 5.30999e-06
        [merge_forward]: 2.67001e-06
        [cell_reuse_recompute_pass]: 2.48e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.341e-05
        [j_node_and_user_rematch]: 9.02e-06
        [meta_fg_expand]: 1.51002e-06
        [replace_old_param]: 8.28999e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.21002e-06
        [auto_monad_grad]: 8.79983e-07
        [auto_monad_eliminator]: 6.04999e-06
        [cse]: 1.18e-05
        [replace_applicator]: 6.11e-06
[py_interpret_to_execute_after_opt_a]: 1.04e-05
[rewriter_after_opt_a]: 3.07e-05
[convert_after_rewriter]: 6.59001e-06
[order_py_execute_after_rewriter]: 5.11002e-06
[mutable_eliminate]: 0.0007325
[jit_opt_b]: 6.223e-05, [1]
    [Cycle 1]: 5.424e-05, [2]
        [frontend_op_eliminate]: 2.211e-05
        [inline_after_opt_a]: 1.858e-05
[cconv]: 2.445e-05
[loop_unroll]: 0.00059299
[jit_opt_after_cconv]: 0.00172042, [1]
    [Cycle 1]: 0.00170982, [11]
        [c_1]: 2.507e-05
        [parameter_eliminate]: 4e-06
        [updatestate_depend_eliminate]: 8.25999e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 1.567e-05
        [cse]: 2.626e-05
        [call_graph_tuple_transform]: 3.861e-05
        [tuple_list_get_item_eliminator]: 8.72e-06
        [none_parameter_eliminate]: 2.32999e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 8.08001e-06
[remove_dup_value]: 1.122e-05
[partial_unused_args_eliminate]: 2.31e-06
[environ_conv]: 6.53998e-06
[add_recomputation]: 4.366e-05
[cse_after_recomputation]: 2.814e-05, [1]
    [Cycle 1]: 1.865e-05, [1]
        [cse]: 1.19e-05
[auto_monad_reorder]: 1.153e-05
[get_jit_bprop_graph]: 2.14999e-06
[rewriter_after_jit_bprop_graph]: 1.057e-05
[opt_after_jit_grad]: 0.00070735
[symbol_engine_optimizer]: 8.809e-05, [1]
    [Cycle 1]: 7.872e-05, [6]
        [build]: 3.14999e-06
        [elim_shapecalc]: 1.019e-05
        [elim_not_effective]: 1.596e-05
        [opt_reshape]: 8.05e-06
        [fold_const_symbol]: 9.97001e-06
        [renormalize]: 9.89996e-07
[validate]: 3.851e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00733089
[execute]: 5.31998e-06
Sums
    bootstrap                            :     0.000397s :  2.14%
    type_inference                       :     0.006229s : 33.61%
    event_method                         :     0.000019s :  0.10%
    auto_monad                           :     0.000066s :  0.36%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000047s :  0.26%
    rewriter_before_opt_a                :     0.000073s :  0.40%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.34%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000679s :  3.66%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000823s :  4.44%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.09%
    jit_opt_a.cse                        :     0.000031s :  0.17%
    jit_opt_a.replace_applicator         :     0.000020s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.17%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000732s :  3.95%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.10%
    cconv                                :     0.000024s :  0.13%
    loop_unroll                          :     0.000593s :  3.20%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000016s :  0.08%
    jit_opt_after_cconv.cse              :     0.000026s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000039s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000008s :  0.04%
    remove_dup_value                     :     0.000011s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000044s :  0.24%
    cse_after_recomputation.cse          :     0.000012s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.06%
    opt_after_jit_grad                   :     0.000707s :  3.82%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.21%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007331s : 39.56%
    execute                              :     0.000005s :  0.03%


Time group info:
------[substitution.]   0.000206    23
 0.69% :     0.000001s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.79% :     0.000006s :      3: substitution.graph_param_transform
84.54% :     0.000175s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.18% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.77% :     0.000004s :      2: substitution.replace_old_param
 5.82% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006153     2
82.50% :     0.005076s :      1: type_inference.infer
17.50% :     0.001077s :      1: type_inference.specialize
------[replace.]   0.000079     6
65.26% :     0.000051s :      4: replace.inline
34.74% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000183     6
94.09% :     0.000172s :      4: match.inline
 5.91% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.65% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 2.29% :     0.000003s :     10: predicate.arithmetic_simplify
 1.28% :     0.000002s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.59% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.32% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.44% :     0.000002s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 1.16% :     0.000002s :      6: predicate.get_grad_eliminate
 0.56% :     0.000001s :      3: predicate.graph_param_transform
 5.35% :     0.000007s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 1.76% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.58% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 2.03% :     0.000003s :      3: predicate.mutable_eliminate
 0.67% :     0.000001s :      3: predicate.opt_reshape
 1.98% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.32% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.99% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.23% :     0.000002s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.54% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.12% :     0.000003s :     16: predicate.switch_defer_inline
 1.91% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.64% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000002s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.60% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.33% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000712    12
44.10% :     0.000314s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.25% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.65% :     0.000368s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027855    76
 0.17% :     0.000047s :      1: add_recomputation
 0.25% :     0.000069s :      1: auto_monad
 0.05% :     0.000015s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.50% :     0.000417s :      1: bootstrap
 0.10% :     0.000027s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000031s :      1: cse_after_recomputation
 0.04% :     0.000012s :      1: environ_conv
 0.09% :     0.000024s :      1: event_method
 0.03% :     0.000009s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
26.66% :     0.007427s :      1: jit_opt_a
 6.19% :     0.001726s :      1: jit_opt_after_cconv
 0.23% :     0.000065s :      1: jit_opt_b
 2.17% :     0.000606s :      1: loop_unroll
 2.67% :     0.000745s :      1: mutable_eliminate
 3.25% :     0.000906s :     26: opt.transform.jit_opt_a
 0.26% :     0.000074s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000032s :      4: opt.transform.jit_opt_b
 0.06% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000040s :      4: opt.transform.symbol_engine_opt
 2.57% :     0.000716s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000050s :      1: py_interpret_to_execute
 0.05% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000014s :      1: remove_dup_value
 1.55% :     0.000431s :      1: renormalize.infer
 1.38% :     0.000384s :      1: renormalize.specialize
 0.05% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000034s :      1: rewriter_after_opt_a
 0.27% :     0.000076s :      1: rewriter_before_opt_a
 0.33% :     0.000091s :      1: symbol_engine_optimizer
26.37% :     0.007345s :      1: task_emit
22.43% :     0.006247s :      1: type_inference
 0.21% :     0.000059s :      1: validate


TotalTime = 0.0241329, [33]
[bootstrap]: 0.00031838
[type_inference]: 0.00526231
[event_method]: 1.906e-05
[auto_monad]: 6.573e-05
[graph_reusing]: 6.07001e-06
[pre_auto_parallel]: 2.85998e-06
[py_interpret_to_execute]: 4.512e-05
[rewriter_before_opt_a]: 7.569e-05
[expand_dump_flag]: 3.35e-06
[jit_opt_a]: 0.00630391, [2]
    [Cycle 1]: 0.00200295, [27]
        [switch_simplify]: 6.537e-05
        [loop_unroll]: 3.039e-05
        [a_1]: 0.00064453
        [with_stream_mark]: 2.22e-05
        [recompute_prepare]: 8.08001e-06
        [updatestate_depend_eliminate]: 4.05e-06
        [updatestate_assign_eliminate]: 3.04001e-06
        [updatestate_loads_eliminate]: 2.99001e-06
        [parameter_eliminate]: 2.42001e-06
        [specialize_transform]: 6.31998e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 6.49001e-06
        [meta_shard_fg_expand]: 2.47001e-06
        [get_grad_eliminate_]: 6.04001e-06
        [merge_forward]: 3.9e-06
        [cell_reuse_recompute_pass]: 1.72999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.81e-05
        [j_node_and_user_rematch]: 1.105e-05
        [meta_fg_expand]: 2.81e-06
        [replace_old_param]: 1.173e-05
        [inline_without_move]: 6.07001e-06
        [renormalize]: 0.0008738
        [add_forward_monad_depend]: 7.13998e-06
        [auto_monad_grad]: 2.56e-06
        [auto_monad_eliminator]: 1.578e-05
        [cse]: 3.208e-05
        [replace_applicator]: 1.341e-05
    [Cycle 2]: 0.00033619, [27]
        [switch_simplify]: 5.80002e-06
        [loop_unroll]: 5.15001e-06
        [a_1]: 9.7e-05
        [with_stream_mark]: 9.47001e-06
        [recompute_prepare]: 5.34998e-06
        [updatestate_depend_eliminate]: 2.65997e-06
        [updatestate_assign_eliminate]: 2.00002e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.33002e-06
        [specialize_transform]: 4.71002e-06
        [updatestate_useless_node_eliminater]: 4.70999e-06
        [accelerated_algorithm]: 5.96e-06
        [meta_shard_fg_expand]: 1.44998e-06
        [get_grad_eliminate_]: 4.57998e-06
        [merge_forward]: 2.29001e-06
        [cell_reuse_recompute_pass]: 2.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.315e-05
        [j_node_and_user_rematch]: 7.8e-06
        [meta_fg_expand]: 1.56002e-06
        [replace_old_param]: 1.544e-05
        [inline_without_move]: 4.61002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.07e-06
        [cse]: 1.182e-05
        [replace_applicator]: 4.95999e-06
[py_interpret_to_execute_after_opt_a]: 1.362e-05
[rewriter_after_opt_a]: 4.153e-05
[convert_after_rewriter]: 7.56999e-06
[order_py_execute_after_rewriter]: 5.34998e-06
[mutable_eliminate]: 0.00074921
[jit_opt_b]: 5.359e-05, [1]
    [Cycle 1]: 4.512e-05, [2]
        [frontend_op_eliminate]: 1.876e-05
        [inline_after_opt_a]: 1.495e-05
[cconv]: 2.827e-05
[loop_unroll]: 0.00272993
[jit_opt_after_cconv]: 0.00020325, [1]
    [Cycle 1]: 0.00019396, [11]
        [c_1]: 2.875e-05
        [parameter_eliminate]: 5.10999e-06
        [updatestate_depend_eliminate]: 8.57e-06
        [updatestate_assign_eliminate]: 3.18998e-06
        [updatestate_loads_eliminate]: 2.98998e-06
        [cse]: 3.921e-05
        [call_graph_tuple_transform]: 2.845e-05
        [tuple_list_get_item_eliminator]: 7.79002e-06
        [none_parameter_eliminate]: 2.05002e-06
        [renormalize]: 1.47001e-06
        [switch_simplify]: 6.46e-06
[remove_dup_value]: 1.684e-05
[partial_unused_args_eliminate]: 2.48e-06
[environ_conv]: 7.55e-06
[add_recomputation]: 6.158e-05
[cse_after_recomputation]: 2.6e-05, [1]
    [Cycle 1]: 2.006e-05, [1]
        [cse]: 1.155e-05
[auto_monad_reorder]: 1.807e-05
[get_jit_bprop_graph]: 2.54001e-06
[rewriter_after_jit_bprop_graph]: 1.168e-05
[opt_after_jit_grad]: 0.00064346
[symbol_engine_optimizer]: 8.953e-05, [1]
    [Cycle 1]: 8.014e-05, [6]
        [build]: 5.20001e-06
        [elim_shapecalc]: 8.42998e-06
        [elim_not_effective]: 1.601e-05
        [opt_reshape]: 7.26001e-06
        [fold_const_symbol]: 1.078e-05
        [renormalize]: 8.39995e-07
[validate]: 5.049e-05
[backend_pass]: 8.99978e-07
[task_emit]: 0.00699607
[execute]: 8.45001e-06
Sums
    bootstrap                            :     0.000318s :  1.64%
    type_inference                       :     0.005262s : 27.05%
    event_method                         :     0.000019s :  0.10%
    auto_monad                           :     0.000066s :  0.34%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000045s :  0.23%
    rewriter_before_opt_a                :     0.000076s :  0.39%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000071s :  0.37%
    jit_opt_a.loop_unroll                :     0.000036s :  0.18%
    jit_opt_a.a_1                        :     0.000742s :  3.81%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000027s :  0.14%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.000874s :  4.49%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.11%
    jit_opt_a.cse                        :     0.000044s :  0.23%
    jit_opt_a.replace_applicator         :     0.000018s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.07%
    rewriter_after_opt_a                 :     0.000042s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000749s :  3.85%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000028s :  0.15%
    loop_unroll                          :     0.002730s : 14.03%
    jit_opt_after_cconv.c_1              :     0.000029s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000039s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000008s :  0.04%
    add_recomputation                    :     0.000062s :  0.32%
    cse_after_recomputation.cse          :     0.000012s :  0.06%
    auto_monad_reorder                   :     0.000018s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000012s :  0.06%
    opt_after_jit_grad                   :     0.000643s :  3.31%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000050s :  0.26%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006996s : 35.96%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000246    23
 0.80% :     0.000002s :      2: substitution.elim_not_effective
 0.77% :     0.000002s :      2: substitution.fold_const_symbol
 3.15% :     0.000008s :      3: substitution.graph_param_transform
80.34% :     0.000198s :      4: substitution.inline
 1.58% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.41% :     0.000006s :      4: substitution.remove_not_recompute_node
 5.34% :     0.000013s :      2: substitution.replace_old_param
 5.61% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005181     2
79.90% :     0.004139s :      1: type_inference.infer
20.10% :     0.001041s :      1: type_inference.specialize
------[replace.]   0.000081     6
63.37% :     0.000052s :      4: replace.inline
36.63% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000208     6
93.92% :     0.000195s :      4: match.inline
 6.08% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   691
 1.81% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.70% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.56% :     0.000002s :     10: predicate.addn_zero_filter
 2.15% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000002s :     10: predicate.depend_value_elim
 0.93% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.75% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000002s :     10: predicate.environ_get_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.25% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.60% :     0.000002s :      6: predicate.less_batch_normalization
 1.70% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 3.52% :     0.000005s :      3: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.24% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 1.45% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 1.88% :     0.000002s :     15: predicate.partial_eliminate
 1.38% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.03% :     0.000003s :     10: predicate.reduce_eliminate
 1.21% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.95% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.10% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.02% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.36% :     0.000002s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 1.93% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.05% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000002s :     10: predicate.transpose_eliminate
 1.60% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.49% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.77% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.44% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.46% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000675    12
44.75% :     0.000302s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.19% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.06% :     0.000338s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026121    76
 0.25% :     0.000065s :      1: add_recomputation
 0.26% :     0.000069s :      1: auto_monad
 0.08% :     0.000021s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.28% :     0.000334s :      1: bootstrap
 0.12% :     0.000031s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000028s :      1: cse_after_recomputation
 0.05% :     0.000012s :      1: environ_conv
 0.09% :     0.000023s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000006s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
24.15% :     0.006309s :      1: jit_opt_a
 0.79% :     0.000207s :      1: jit_opt_after_cconv
 0.22% :     0.000057s :      1: jit_opt_b
10.53% :     0.002750s :      1: loop_unroll
 2.90% :     0.000759s :      1: mutable_eliminate
 3.75% :     0.000981s :     26: opt.transform.jit_opt_a
 0.26% :     0.000067s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000027s :      4: opt.transform.jit_opt_b
 0.10% :     0.000026s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000039s :      4: opt.transform.symbol_engine_opt
 2.50% :     0.000654s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000048s :      1: py_interpret_to_execute
 0.06% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000020s :      1: remove_dup_value
 1.77% :     0.000462s :      1: renormalize.infer
 1.54% :     0.000402s :      1: renormalize.specialize
 0.05% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000044s :      1: rewriter_after_opt_a
 0.30% :     0.000079s :      1: rewriter_before_opt_a
 0.35% :     0.000092s :      1: symbol_engine_optimizer
26.87% :     0.007018s :      1: task_emit
20.23% :     0.005285s :      1: type_inference
 0.28% :     0.000073s :      1: validate


TotalTime = 0.016104, [30]
[bootstrap]: 0.00031128
[type_inference]: 0.00766339
[event_method]: 2.35e-05
[auto_monad]: 8.617e-05
[graph_reusing]: 6.50002e-06
[pre_auto_parallel]: 2.91999e-06
[py_interpret_to_execute]: 4.978e-05
[rewriter_before_opt_a]: 9.445e-05
[expand_dump_flag]: 3.24001e-06
[jit_opt_a]: 0.0057095, [2]
    [Cycle 1]: 0.00188786, [27]
        [switch_simplify]: 0.00012382
        [loop_unroll]: 3.644e-05
        [a_1]: 0.00070129
        [with_stream_mark]: 2.057e-05
        [recompute_prepare]: 7.09001e-06
        [updatestate_depend_eliminate]: 3.97e-06
        [updatestate_assign_eliminate]: 3.16999e-06
        [updatestate_loads_eliminate]: 2.69999e-06
        [parameter_eliminate]: 1.84998e-06
        [specialize_transform]: 6.57002e-06
        [updatestate_useless_node_eliminater]: 5.40999e-06
        [accelerated_algorithm]: 6.29999e-06
        [meta_shard_fg_expand]: 2.19999e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 4.02e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.421e-05
        [j_node_and_user_rematch]: 9.47001e-06
        [meta_fg_expand]: 2.32001e-06
        [replace_old_param]: 9.51e-06
        [inline_without_move]: 5.69e-06
        [renormalize]: 0.0006723
        [add_forward_monad_depend]: 5.76e-06
        [auto_monad_grad]: 2.43998e-06
        [auto_monad_eliminator]: 1.322e-05
        [cse]: 2.792e-05
        [replace_applicator]: 1.132e-05
    [Cycle 2]: 0.00030594, [27]
        [switch_simplify]: 5.62001e-06
        [loop_unroll]: 4.84998e-06
        [a_1]: 8.645e-05
        [with_stream_mark]: 9.34e-06
        [recompute_prepare]: 5.04e-06
        [updatestate_depend_eliminate]: 2.59999e-06
        [updatestate_assign_eliminate]: 1.96998e-06
        [updatestate_loads_eliminate]: 1.73002e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.72e-06
        [updatestate_useless_node_eliminater]: 4.68999e-06
        [accelerated_algorithm]: 4.50999e-06
        [meta_shard_fg_expand]: 1.29998e-06
        [get_grad_eliminate_]: 4.22e-06
        [merge_forward]: 2.58998e-06
        [cell_reuse_recompute_pass]: 1.39998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.263e-05
        [j_node_and_user_rematch]: 7.5e-06
        [meta_fg_expand]: 1.39e-06
        [replace_old_param]: 6.87002e-06
        [inline_without_move]: 4.75001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 4.68001e-06
        [cse]: 9.57999e-06
        [replace_applicator]: 5.28002e-06
[py_interpret_to_execute_after_opt_a]: 9.46e-06
[rewriter_after_opt_a]: 3.345e-05
[convert_after_rewriter]: 6.65002e-06
[order_py_execute_after_rewriter]: 4.62998e-06
[mutable_eliminate]: 0.00050084
[jit_opt_b]: 5.509e-05, [1]
    [Cycle 1]: 4.832e-05, [2]
        [frontend_op_eliminate]: 2.149e-05
        [inline_after_opt_a]: 1.54e-05
[cconv]: 8.138e-05
[loop_unroll]: 0.00043575
[jit_opt_after_cconv]: 0.00013924, [1]
    [Cycle 1]: 0.00013303, [11]
        [c_1]: 1.966e-05
        [parameter_eliminate]: 2.54001e-06
        [updatestate_depend_eliminate]: 5.44998e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.828e-05
        [call_graph_tuple_transform]: 1.893e-05
        [tuple_list_get_item_eliminator]: 5.68002e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 5.89e-06
[remove_dup_value]: 1.243e-05
[partial_unused_args_eliminate]: 2.69001e-06
[environ_conv]: 5.62001e-06
[add_recomputation]: 4.559e-05
[cse_after_recomputation]: 2.033e-05, [1]
    [Cycle 1]: 1.477e-05, [1]
        [cse]: 8.79e-06
[auto_monad_reorder]: 1.448e-05
[get_jit_bprop_graph]: 2.16e-06
[rewriter_after_jit_bprop_graph]: 5.06997e-06
[opt_after_jit_grad]: 0.00047936
[symbol_engine_optimizer]: 7.649e-05, [1]
    [Cycle 1]: 7.011e-05, [6]
        [build]: 4.26001e-06
        [elim_shapecalc]: 7.16001e-06
        [elim_not_effective]: 1.395e-05
        [opt_reshape]: 6.09999e-06
        [fold_const_symbol]: 9.10001e-06
        [renormalize]: 4.19997e-07
[validate]: 3.665e-05
Sums
    bootstrap                            :     0.000311s :  2.60%
    type_inference                       :     0.007663s : 63.92%
    event_method                         :     0.000024s :  0.20%
    auto_monad                           :     0.000086s :  0.72%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000050s :  0.42%
    rewriter_before_opt_a                :     0.000094s :  0.79%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000129s :  1.08%
    jit_opt_a.loop_unroll                :     0.000041s :  0.34%
    jit_opt_a.a_1                        :     0.000788s :  6.57%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000672s :  5.61%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.15%
    jit_opt_a.cse                        :     0.000037s :  0.31%
    jit_opt_a.replace_applicator         :     0.000017s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.08%
    rewriter_after_opt_a                 :     0.000033s :  0.28%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000501s :  4.18%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.13%
    cconv                                :     0.000081s :  0.68%
    loop_unroll                          :     0.000436s :  3.63%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000012s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000046s :  0.38%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000479s :  4.00%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.31%


Time group info:
------[substitution.]   0.000282    26
 0.69% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 1.86% :     0.000005s :      3: substitution.graph_param_transform
84.87% :     0.000240s :      6: substitution.inline
 1.09% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.80% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.58% :     0.000004s :      2: substitution.replace_old_param
 3.03% :     0.000009s :      1: substitution.switch_simplify
 4.59% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007584     2
84.67% :     0.006421s :      1: type_inference.infer
15.33% :     0.001163s :      1: type_inference.specialize
------[replace.]   0.000129     9
48.57% :     0.000063s :      6: replace.inline
28.18% :     0.000036s :      1: replace.switch_simplify
23.25% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000256     9
92.35% :     0.000236s :      6: match.inline
 3.02% :     0.000008s :      1: match.switch_simplify
 4.63% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   801
 1.51% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.17% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.21% :     0.000002s :     12: predicate.addn_check_dump
 1.65% :     0.000002s :     12: predicate.addn_zero_filter
 2.06% :     0.000003s :     12: predicate.arithmetic_simplify
 1.35% :     0.000002s :     12: predicate.cast_eliminate
 0.35% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.18% :     0.000002s :     12: predicate.compare_switch_simplify
 1.43% :     0.000002s :     12: predicate.depend_value_elim
 1.14% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.41% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.15% :     0.000002s :     12: predicate.environ_get_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.75% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 5.72% :     0.000008s :     26: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     17: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.33% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.26% :     0.000002s :     12: predicate.merge_addn
 1.13% :     0.000002s :     12: predicate.minmaximum_grad
 1.24% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.36% :     0.000003s :     17: predicate.partial_eliminate
 1.29% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     12: predicate.reduce_eliminate
 1.53% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     20: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     12: predicate.reshape_eliminate
 1.52% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.57% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     12: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.74% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.88% :     0.000004s :     20: predicate.switch_defer_inline
 2.46% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.85% :     0.000012s :     65: predicate.switch_simplify
 1.32% :     0.000002s :     12: predicate.tile_eliminate
 1.27% :     0.000002s :     12: predicate.transpose_eliminate
 1.42% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.49% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.39% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.77% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     12: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000905    15
54.20% :     0.000490s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.83% :     0.000098s :      1: func_graph_cloner_run.FuncGraphClonerNode
34.97% :     0.000316s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017903    72
 0.27% :     0.000049s :      1: add_recomputation
 0.50% :     0.000090s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 1.85% :     0.000332s :      1: bootstrap
 0.47% :     0.000085s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.16% :     0.000028s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
31.91% :     0.005712s :      1: jit_opt_a
 0.79% :     0.000142s :      1: jit_opt_after_cconv
 0.32% :     0.000058s :      1: jit_opt_b
 2.47% :     0.000443s :      1: loop_unroll
 2.84% :     0.000509s :      1: mutable_eliminate
 5.99% :     0.001072s :     26: opt.transform.jit_opt_a
 0.26% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000030s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.73% :     0.000489s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000053s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 2.20% :     0.000395s :      1: renormalize.infer
 1.51% :     0.000270s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000036s :      1: rewriter_after_opt_a
 0.55% :     0.000098s :      1: rewriter_before_opt_a
 0.44% :     0.000079s :      1: symbol_engine_optimizer
42.92% :     0.007684s :      1: type_inference


TotalTime = 0.0395708, [33]
[bootstrap]: 0.00050351
[type_inference]: 0.00838024
[event_method]: 2.955e-05
[auto_monad]: 8.285e-05
[graph_reusing]: 7.23e-06
[pre_auto_parallel]: 3.48e-06
[py_interpret_to_execute]: 8.598e-05
[rewriter_before_opt_a]: 9.979e-05
[expand_dump_flag]: 3.83999e-06
[jit_opt_a]: 0.00779355, [2]
    [Cycle 1]: 0.00235499, [27]
        [switch_simplify]: 7.743e-05
        [loop_unroll]: 3.083e-05
        [a_1]: 0.00069789
        [with_stream_mark]: 3.125e-05
        [recompute_prepare]: 9.27999e-06
        [updatestate_depend_eliminate]: 4.62998e-06
        [updatestate_assign_eliminate]: 3.55e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [parameter_eliminate]: 2.44001e-06
        [specialize_transform]: 6.26e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 6.30997e-06
        [meta_shard_fg_expand]: 2.46e-06
        [get_grad_eliminate_]: 5.60001e-06
        [merge_forward]: 4.03999e-06
        [cell_reuse_recompute_pass]: 2.43e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.678e-05
        [j_node_and_user_rematch]: 1.207e-05
        [meta_fg_expand]: 2.63e-06
        [replace_old_param]: 1.174e-05
        [inline_without_move]: 5.39998e-06
        [renormalize]: 0.00109552
        [add_forward_monad_depend]: 8.82e-06
        [auto_monad_grad]: 3.45e-06
        [auto_monad_eliminator]: 2.113e-05
        [cse]: 3.978e-05
        [replace_applicator]: 1.994e-05
    [Cycle 2]: 0.00042776, [27]
        [switch_simplify]: 6.73e-06
        [loop_unroll]: 6.21e-06
        [a_1]: 0.00011188
        [with_stream_mark]: 1.801e-05
        [recompute_prepare]: 7.13998e-06
        [updatestate_depend_eliminate]: 3.36999e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.66e-06
        [specialize_transform]: 5.57001e-06
        [updatestate_useless_node_eliminater]: 5.80002e-06
        [accelerated_algorithm]: 7.16001e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 4.87e-06
        [merge_forward]: 4.02e-06
        [cell_reuse_recompute_pass]: 4.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.867e-05
        [j_node_and_user_rematch]: 1.084e-05
        [meta_fg_expand]: 2.22001e-06
        [replace_old_param]: 1.038e-05
        [inline_without_move]: 5.54e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 2.58e-06
        [auto_monad_grad]: 1.96e-06
        [auto_monad_eliminator]: 8.23001e-06
        [cse]: 1.577e-05
        [replace_applicator]: 2.159e-05
[py_interpret_to_execute_after_opt_a]: 2.004e-05
[rewriter_after_opt_a]: 5.176e-05
[convert_after_rewriter]: 8.93002e-06
[order_py_execute_after_rewriter]: 5.61e-06
[mutable_eliminate]: 0.00125956
[jit_opt_b]: 9.751e-05, [1]
    [Cycle 1]: 6.915e-05, [2]
        [frontend_op_eliminate]: 3.163e-05
        [inline_after_opt_a]: 2.091e-05
[cconv]: 4.893e-05
[loop_unroll]: 0.00069392
[jit_opt_after_cconv]: 0.00023407, [1]
    [Cycle 1]: 0.00022472, [11]
        [c_1]: 2.689e-05
        [parameter_eliminate]: 7.05e-06
        [updatestate_depend_eliminate]: 1.319e-05
        [updatestate_assign_eliminate]: 3.88001e-06
        [updatestate_loads_eliminate]: 2.88e-06
        [cse]: 4.483e-05
        [call_graph_tuple_transform]: 3.159e-05
        [tuple_list_get_item_eliminator]: 7.13e-06
        [none_parameter_eliminate]: 2.29999e-06
        [renormalize]: 8.30012e-07
        [switch_simplify]: 6.56e-06
[remove_dup_value]: 1.868e-05
[partial_unused_args_eliminate]: 3.01999e-06
[environ_conv]: 7.38999e-06
[add_recomputation]: 7.006e-05
[cse_after_recomputation]: 2.879e-05, [1]
    [Cycle 1]: 1.962e-05, [1]
        [cse]: 1.217e-05
[auto_monad_reorder]: 1.884e-05
[get_jit_bprop_graph]: 3.26999e-06
[rewriter_after_jit_bprop_graph]: 1.26e-05
[opt_after_jit_grad]: 0.00072353
[symbol_engine_optimizer]: 0.000109, [1]
    [Cycle 1]: 9.712e-05, [6]
        [build]: 8.07e-06
        [elim_shapecalc]: 1.195e-05
        [elim_not_effective]: 2.103e-05
        [opt_reshape]: 9.02999e-06
        [fold_const_symbol]: 1.171e-05
        [renormalize]: 1.04003e-06
[validate]: 6.552e-05
[backend_pass]: 1.77001e-06
[task_emit]: 0.0187348
[execute]: 1.118e-05
Sums
    bootstrap                            :     0.000504s :  1.50%
    type_inference                       :     0.008380s : 24.90%
    event_method                         :     0.000030s :  0.09%
    auto_monad                           :     0.000083s :  0.25%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000086s :  0.26%
    rewriter_before_opt_a                :     0.000100s :  0.30%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000084s :  0.25%
    jit_opt_a.loop_unroll                :     0.000037s :  0.11%
    jit_opt_a.a_1                        :     0.000810s :  2.41%
    jit_opt_a.with_stream_mark           :     0.000049s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000008s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000023s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000022s :  0.07%
    jit_opt_a.inline_without_move        :     0.000011s :  0.03%
    jit_opt_a.renormalize                :     0.001096s :  3.26%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000029s :  0.09%
    jit_opt_a.cse                        :     0.000056s :  0.17%
    jit_opt_a.replace_applicator         :     0.000042s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000020s :  0.06%
    rewriter_after_opt_a                 :     0.000052s :  0.15%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.001260s :  3.74%
    jit_opt_b.frontend_op_eliminate      :     0.000032s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000021s :  0.06%
    cconv                                :     0.000049s :  0.15%
    loop_unroll                          :     0.000694s :  2.06%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000013s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000045s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000032s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000019s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000070s :  0.21%
    cse_after_recomputation.cse          :     0.000012s :  0.04%
    auto_monad_reorder                   :     0.000019s :  0.06%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000013s :  0.04%
    opt_after_jit_grad                   :     0.000724s :  2.15%
    symbol_engine_optimizer.build        :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000066s :  0.19%
    backend_pass                         :     0.000002s :  0.01%
    task_emit                            :     0.018735s : 55.67%
    execute                              :     0.000011s :  0.03%


Time group info:
------[substitution.]   0.000323    23
 1.14% :     0.000004s :      2: substitution.elim_not_effective
 0.53% :     0.000002s :      2: substitution.fold_const_symbol
 2.11% :     0.000007s :      3: substitution.graph_param_transform
85.39% :     0.000276s :      4: substitution.inline
 1.77% :     0.000006s :      4: substitution.j_node_and_user_rematch
 1.80% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.50% :     0.000008s :      2: substitution.replace_old_param
 4.76% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008245     2
84.27% :     0.006948s :      1: type_inference.infer
15.73% :     0.001297s :      1: type_inference.specialize
------[replace.]   0.000094     6
69.28% :     0.000065s :      4: replace.inline
30.72% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000287     6
95.04% :     0.000272s :      4: match.inline
 4.96% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000150   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.35% :     0.000004s :      3: predicate.ad_related_special_op_eliminate
 0.84% :     0.000001s :     10: predicate.addn_check_dump
 1.16% :     0.000002s :     10: predicate.addn_zero_filter
 2.44% :     0.000004s :     10: predicate.arithmetic_simplify
 1.20% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.03% :     0.000002s :     10: predicate.depend_value_elim
 0.94% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.26% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 1.23% :     0.000002s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.94% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     10: predicate.environ_get_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.17% :     0.000000s :      3: predicate.fold_const_symbol
 0.80% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000001s :      3: predicate.graph_param_transform
 5.82% :     0.000009s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.29% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.62% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.53% :     0.000002s :     15: predicate.load_eliminater
 2.18% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.82% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000002s :     10: predicate.merge_addn
 0.86% :     0.000001s :     10: predicate.minmaximum_grad
 3.64% :     0.000005s :      3: predicate.mutable_eliminate
 0.62% :     0.000001s :      3: predicate.opt_reshape
 1.73% :     0.000003s :     15: predicate.partial_eliminate
 1.00% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.16% :     0.000003s :     10: predicate.reduce_eliminate
 1.21% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000003s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.39% :     0.000001s :      3: predicate.reset_defer_inline
 1.30% :     0.000002s :     10: predicate.reshape_eliminate
 0.87% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.05% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.40% :     0.000002s :      6: predicate.special_op_eliminate
 0.75% :     0.000001s :      6: predicate.specialize_transform
 1.17% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 1.94% :     0.000003s :     16: predicate.switch_defer_inline
 1.69% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.04% :     0.000012s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 0.90% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.06% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.38% :     0.000007s :     18: predicate.tuple_list_get_item_eliminator
 2.31% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.28% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.43% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.27% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.06% :     0.000002s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000976    12
50.48% :     0.000493s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.80% :     0.000066s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.72% :     0.000417s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.041922    76
 0.18% :     0.000075s :      1: add_recomputation
 0.21% :     0.000090s :      1: auto_monad
 0.05% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 1.28% :     0.000535s :      1: bootstrap
 0.13% :     0.000053s :      1: cconv
 0.03% :     0.000012s :      1: convert_after_rewriter
 0.08% :     0.000032s :      1: cse_after_recomputation
 0.03% :     0.000012s :      1: environ_conv
 0.09% :     0.000036s :      1: event_method
 0.04% :     0.000018s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000006s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
18.60% :     0.007799s :      1: jit_opt_a
 0.57% :     0.000238s :      1: jit_opt_after_cconv
 0.24% :     0.000102s :      1: jit_opt_b
 1.69% :     0.000709s :      1: loop_unroll
 3.06% :     0.001284s :      1: mutable_eliminate
 2.60% :     0.001091s :     26: opt.transform.jit_opt_a
 0.16% :     0.000068s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000044s :      4: opt.transform.jit_opt_b
 0.05% :     0.000021s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000033s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000034s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000048s :      4: opt.transform.symbol_engine_opt
 1.76% :     0.000740s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000089s :      1: py_interpret_to_execute
 0.06% :     0.000024s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000021s :      1: remove_dup_value
 1.41% :     0.000591s :      1: renormalize.infer
 1.17% :     0.000491s :      1: renormalize.specialize
 0.04% :     0.000016s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000056s :      1: rewriter_after_opt_a
 0.25% :     0.000106s :      1: rewriter_before_opt_a
 0.27% :     0.000113s :      1: symbol_engine_optimizer
44.75% :     0.018762s :      1: task_emit
20.08% :     0.008417s :      1: type_inference
 0.23% :     0.000098s :      1: validate


TotalTime = 0.0140896, [30]
[bootstrap]: 0.00026634
[type_inference]: 0.00609154
[event_method]: 2.001e-05
[auto_monad]: 8.018e-05
[graph_reusing]: 6.12999e-06
[pre_auto_parallel]: 2.64001e-06
[py_interpret_to_execute]: 4.316e-05
[rewriter_before_opt_a]: 8.54e-05
[expand_dump_flag]: 3.08998e-06
[jit_opt_a]: 0.00540253, [2]
    [Cycle 1]: 0.00185782, [27]
        [switch_simplify]: 0.00011685
        [loop_unroll]: 3.33e-05
        [a_1]: 0.000692
        [with_stream_mark]: 1.714e-05
        [recompute_prepare]: 7.53e-06
        [updatestate_depend_eliminate]: 4.12998e-06
        [updatestate_assign_eliminate]: 2.68998e-06
        [updatestate_loads_eliminate]: 2.56998e-06
        [parameter_eliminate]: 2.07999e-06
        [specialize_transform]: 5.81e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 5.09e-06
        [meta_shard_fg_expand]: 2.12001e-06
        [get_grad_eliminate_]: 5.14003e-06
        [merge_forward]: 4.15e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.379e-05
        [j_node_and_user_rematch]: 8.75999e-06
        [meta_fg_expand]: 2.32001e-06
        [replace_old_param]: 8.53001e-06
        [inline_without_move]: 5.13002e-06
        [renormalize]: 0.00066305
        [add_forward_monad_depend]: 5.74999e-06
        [auto_monad_grad]: 2.71e-06
        [auto_monad_eliminator]: 1.374e-05
        [cse]: 3.042e-05
        [replace_applicator]: 1.239e-05
    [Cycle 2]: 0.00031656, [27]
        [switch_simplify]: 5.59998e-06
        [loop_unroll]: 5.69e-06
        [a_1]: 8.828e-05
        [with_stream_mark]: 9.42999e-06
        [recompute_prepare]: 4.79e-06
        [updatestate_depend_eliminate]: 2.59999e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 4.97999e-06
        [updatestate_useless_node_eliminater]: 5.17999e-06
        [accelerated_algorithm]: 4.67e-06
        [meta_shard_fg_expand]: 1.07998e-06
        [get_grad_eliminate_]: 4.3e-06
        [merge_forward]: 2.68e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.26e-05
        [j_node_and_user_rematch]: 8.45999e-06
        [meta_fg_expand]: 1.66002e-06
        [replace_old_param]: 7.18998e-06
        [inline_without_move]: 5.02e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 1.12999e-06
        [auto_monad_eliminator]: 4.72e-06
        [cse]: 1.011e-05
        [replace_applicator]: 5.52001e-06
[py_interpret_to_execute_after_opt_a]: 1.094e-05
[rewriter_after_opt_a]: 3.705e-05
[convert_after_rewriter]: 7.31001e-06
[order_py_execute_after_rewriter]: 5.47001e-06
[mutable_eliminate]: 0.00057287
[jit_opt_b]: 5.801e-05, [1]
    [Cycle 1]: 5.095e-05, [2]
        [frontend_op_eliminate]: 1.555e-05
        [inline_after_opt_a]: 1.571e-05
[cconv]: 2.382e-05
[loop_unroll]: 0.0004126
[jit_opt_after_cconv]: 0.00013378, [1]
    [Cycle 1]: 0.00012765, [11]
        [c_1]: 1.983e-05
        [parameter_eliminate]: 2.52001e-06
        [updatestate_depend_eliminate]: 5.12e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [cse]: 1.751e-05
        [call_graph_tuple_transform]: 1.744e-05
        [tuple_list_get_item_eliminator]: 5.24e-06
        [none_parameter_eliminate]: 1.37e-06
        [renormalize]: 8.30012e-07
        [switch_simplify]: 5.09e-06
[remove_dup_value]: 1.325e-05
[partial_unused_args_eliminate]: 2.25002e-06
[environ_conv]: 4.94e-06
[add_recomputation]: 4.899e-05
[cse_after_recomputation]: 2.042e-05, [1]
    [Cycle 1]: 1.509e-05, [1]
        [cse]: 9.19e-06
[auto_monad_reorder]: 1.458e-05
[get_jit_bprop_graph]: 2.12001e-06
[rewriter_after_jit_bprop_graph]: 5.59e-06
[opt_after_jit_grad]: 0.00044211
[symbol_engine_optimizer]: 7.085e-05, [1]
    [Cycle 1]: 6.418e-05, [6]
        [build]: 3.96001e-06
        [elim_shapecalc]: 7.08e-06
        [elim_not_effective]: 1.21e-05
        [opt_reshape]: 5.53002e-06
        [fold_const_symbol]: 8.57e-06
        [renormalize]: 4.00003e-07
[validate]: 3.336e-05
Sums
    bootstrap                            :     0.000266s :  2.59%
    type_inference                       :     0.006092s : 59.32%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000080s :  0.78%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000043s :  0.42%
    rewriter_before_opt_a                :     0.000085s :  0.83%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000122s :  1.19%
    jit_opt_a.loop_unroll                :     0.000039s :  0.38%
    jit_opt_a.a_1                        :     0.000780s :  7.60%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000663s :  6.46%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.18%
    jit_opt_a.cse                        :     0.000041s :  0.39%
    jit_opt_a.replace_applicator         :     0.000018s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.11%
    rewriter_after_opt_a                 :     0.000037s :  0.36%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000573s :  5.58%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.15%
    cconv                                :     0.000024s :  0.23%
    loop_unroll                          :     0.000413s :  4.02%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000049s :  0.48%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000442s :  4.30%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.32%


Time group info:
------[substitution.]   0.000288    26
 0.69% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000002s :      2: substitution.fold_const_symbol
 1.71% :     0.000005s :      3: substitution.graph_param_transform
85.60% :     0.000247s :      6: substitution.inline
 1.26% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.47% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.31% :     0.000004s :      2: substitution.replace_old_param
 2.78% :     0.000008s :      1: substitution.switch_simplify
 4.62% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006030     2
84.03% :     0.005067s :      1: type_inference.infer
15.97% :     0.000963s :      1: type_inference.specialize
------[replace.]   0.000131     9
48.73% :     0.000064s :      6: replace.inline
27.91% :     0.000037s :      1: replace.switch_simplify
23.36% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000263     9
92.63% :     0.000244s :      6: match.inline
 2.75% :     0.000007s :      1: match.switch_simplify
 4.62% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   801
 1.33% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     12: predicate.addn_check_dump
 1.43% :     0.000002s :     12: predicate.addn_zero_filter
 1.88% :     0.000002s :     12: predicate.arithmetic_simplify
 1.27% :     0.000002s :     12: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     12: predicate.compare_switch_simplify
 1.40% :     0.000002s :     12: predicate.depend_value_elim
 1.19% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.71% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     12: predicate.environ_get_eliminate
 1.12% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.53% :     0.000007s :     26: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.98% :     0.000001s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     17: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.30% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000002s :     12: predicate.merge_addn
 1.12% :     0.000001s :     12: predicate.minmaximum_grad
 1.62% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     17: predicate.partial_eliminate
 1.34% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     12: predicate.reduce_eliminate
 1.48% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     20: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     12: predicate.reshape_eliminate
 1.25% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     12: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.65% :     0.000003s :     20: predicate.switch_defer_inline
 2.36% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.88% :     0.000011s :     65: predicate.switch_simplify
 1.32% :     0.000002s :     12: predicate.tile_eliminate
 1.22% :     0.000002s :     12: predicate.transpose_eliminate
 1.46% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.12% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.90% :     0.000002s :     12: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000743    15
49.36% :     0.000367s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.16% :     0.000098s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.48% :     0.000279s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015857    72
 0.33% :     0.000052s :      1: add_recomputation
 0.53% :     0.000084s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.80% :     0.000285s :      1: bootstrap
 0.17% :     0.000027s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
34.09% :     0.005406s :      1: jit_opt_a
 0.86% :     0.000136s :      1: jit_opt_after_cconv
 0.38% :     0.000060s :      1: jit_opt_b
 2.65% :     0.000420s :      1: loop_unroll
 3.66% :     0.000581s :      1: mutable_eliminate
 6.65% :     0.001054s :     26: opt.transform.jit_opt_a
 0.28% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.83% :     0.000450s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000046s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.30% :     0.000365s :      1: renormalize.infer
 1.83% :     0.000291s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000040s :      1: rewriter_after_opt_a
 0.56% :     0.000088s :      1: rewriter_before_opt_a
 0.46% :     0.000073s :      1: symbol_engine_optimizer
38.50% :     0.006105s :      1: type_inference


TotalTime = 0.0280927, [33]
[bootstrap]: 0.00032941
[type_inference]: 0.00901674
[event_method]: 1.824e-05
[auto_monad]: 7.009e-05
[graph_reusing]: 5.55001e-06
[pre_auto_parallel]: 2.19001e-06
[py_interpret_to_execute]: 5.86e-05
[rewriter_before_opt_a]: 7.766e-05
[expand_dump_flag]: 4.05e-06
[jit_opt_a]: 0.00621372, [2]
    [Cycle 1]: 0.00190045, [27]
        [switch_simplify]: 6.763e-05
        [loop_unroll]: 3.442e-05
        [a_1]: 0.00059518
        [with_stream_mark]: 1.82e-05
        [recompute_prepare]: 7.02002e-06
        [updatestate_depend_eliminate]: 3.11999e-06
        [updatestate_assign_eliminate]: 3.17002e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 5.83002e-06
        [updatestate_useless_node_eliminater]: 5.25999e-06
        [accelerated_algorithm]: 5.81003e-06
        [meta_shard_fg_expand]: 2.23998e-06
        [get_grad_eliminate_]: 5.54998e-06
        [merge_forward]: 3.51001e-06
        [cell_reuse_recompute_pass]: 1.51002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.439e-05
        [j_node_and_user_rematch]: 9.96e-06
        [meta_fg_expand]: 2.81e-06
        [replace_old_param]: 1.002e-05
        [inline_without_move]: 5.80002e-06
        [renormalize]: 0.00085002
        [add_forward_monad_depend]: 6.05002e-06
        [auto_monad_grad]: 2.34001e-06
        [auto_monad_eliminator]: 1.518e-05
        [cse]: 2.226e-05
        [replace_applicator]: 1.41e-05
    [Cycle 2]: 0.0003478, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 5.29998e-06
        [a_1]: 9.535e-05
        [with_stream_mark]: 1.069e-05
        [recompute_prepare]: 5.20999e-06
        [updatestate_depend_eliminate]: 2.69001e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 2.01998e-06
        [parameter_eliminate]: 1.29998e-06
        [specialize_transform]: 5.22e-06
        [updatestate_useless_node_eliminater]: 4.91002e-06
        [accelerated_algorithm]: 5.72001e-06
        [meta_shard_fg_expand]: 1.45999e-06
        [get_grad_eliminate_]: 5.28002e-06
        [merge_forward]: 2.69999e-06
        [cell_reuse_recompute_pass]: 2.34001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.451e-05
        [j_node_and_user_rematch]: 8.35999e-06
        [meta_fg_expand]: 2.12999e-06
        [replace_old_param]: 8.12e-06
        [inline_without_move]: 4.83001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.46002e-06
        [auto_monad_grad]: 1.00001e-06
        [auto_monad_eliminator]: 6.43e-06
        [cse]: 1.121e-05
        [replace_applicator]: 5.07e-06
[py_interpret_to_execute_after_opt_a]: 1.188e-05
[rewriter_after_opt_a]: 3.554e-05
[convert_after_rewriter]: 7.9e-06
[order_py_execute_after_rewriter]: 5.32001e-06
[mutable_eliminate]: 0.00068474
[jit_opt_b]: 5.606e-05, [1]
    [Cycle 1]: 4.832e-05, [2]
        [frontend_op_eliminate]: 1.92e-05
        [inline_after_opt_a]: 1.638e-05
[cconv]: 2.953e-05
[loop_unroll]: 0.00046397
[jit_opt_after_cconv]: 0.00015224, [1]
    [Cycle 1]: 0.00014553, [11]
        [c_1]: 2.273e-05
        [parameter_eliminate]: 3.35e-06
        [updatestate_depend_eliminate]: 5.48002e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [cse]: 2.299e-05
        [call_graph_tuple_transform]: 2.145e-05
        [tuple_list_get_item_eliminator]: 5.89e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 5.39998e-06
[remove_dup_value]: 1.476e-05
[partial_unused_args_eliminate]: 2.45002e-06
[environ_conv]: 6.31e-06
[add_recomputation]: 5.099e-05
[cse_after_recomputation]: 2.207e-05, [1]
    [Cycle 1]: 1.591e-05, [1]
        [cse]: 9.67001e-06
[auto_monad_reorder]: 1.75e-05
[get_jit_bprop_graph]: 2.18002e-06
[rewriter_after_jit_bprop_graph]: 9.75002e-06
[opt_after_jit_grad]: 0.00054532
[symbol_engine_optimizer]: 8.844e-05, [1]
    [Cycle 1]: 8.007e-05, [6]
        [build]: 4.03999e-06
        [elim_shapecalc]: 9.09e-06
        [elim_not_effective]: 1.709e-05
        [opt_reshape]: 6.93e-06
        [fold_const_symbol]: 9.64e-06
        [renormalize]: 1.50999e-06
[validate]: 4.134e-05
[backend_pass]: 1.22999e-06
[task_emit]: 0.00961402
[execute]: 9.23002e-06
Sums
    bootstrap                            :     0.000329s :  1.42%
    type_inference                       :     0.009017s : 38.76%
    event_method                         :     0.000018s :  0.08%
    auto_monad                           :     0.000070s :  0.30%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000059s :  0.25%
    rewriter_before_opt_a                :     0.000078s :  0.33%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000073s :  0.31%
    jit_opt_a.loop_unroll                :     0.000040s :  0.17%
    jit_opt_a.a_1                        :     0.000691s :  2.97%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.08%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.000850s :  3.65%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.09%
    jit_opt_a.cse                        :     0.000033s :  0.14%
    jit_opt_a.replace_applicator         :     0.000019s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.05%
    rewriter_after_opt_a                 :     0.000036s :  0.15%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000685s :  2.94%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.07%
    cconv                                :     0.000030s :  0.13%
    loop_unroll                          :     0.000464s :  1.99%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000015s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000051s :  0.22%
    cse_after_recomputation.cse          :     0.000010s :  0.04%
    auto_monad_reorder                   :     0.000018s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.04%
    opt_after_jit_grad                   :     0.000545s :  2.34%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000002s :  0.01%
    validate                             :     0.000041s :  0.18%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.009614s : 41.33%
    execute                              :     0.000009s :  0.04%


Time group info:
------[substitution.]   0.000228    23
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000002s :      2: substitution.fold_const_symbol
 2.68% :     0.000006s :      3: substitution.graph_param_transform
84.03% :     0.000192s :      4: substitution.inline
 1.67% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.94% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000005s :      2: substitution.replace_old_param
 5.98% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008942     2
88.28% :     0.007894s :      1: type_inference.infer
11.72% :     0.001048s :      1: type_inference.specialize
------[replace.]   0.000079     6
68.13% :     0.000054s :      4: replace.inline
31.87% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000202     6
93.84% :     0.000189s :      4: match.inline
 6.16% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.65% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 2.18% :     0.000003s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 1.21% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.08% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.11% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.83% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.60% :     0.000001s :      3: predicate.graph_param_transform
 5.28% :     0.000007s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 1.22% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.33% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.51% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 1.50% :     0.000002s :      3: predicate.mutable_eliminate
 0.74% :     0.000001s :      3: predicate.opt_reshape
 1.95% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.87% :     0.000002s :     10: predicate.reduce_eliminate
 1.30% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.53% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.61% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.36% :     0.000002s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.57% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 1.91% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.77% :     0.000010s :     53: predicate.switch_simplify
 1.20% :     0.000002s :     10: predicate.tile_eliminate
 1.22% :     0.000002s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.62% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.94% :     0.000003s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000731    12
49.20% :     0.000360s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.44% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.36% :     0.000339s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.029980    76
 0.18% :     0.000054s :      1: add_recomputation
 0.24% :     0.000073s :      1: auto_monad
 0.07% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.16% :     0.000347s :      1: bootstrap
 0.11% :     0.000032s :      1: cconv
 0.04% :     0.000013s :      1: convert_after_rewriter
 0.08% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.07% :     0.000022s :      1: event_method
 0.04% :     0.000013s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.65% :     0.000194s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
20.74% :     0.006218s :      1: jit_opt_a
 0.52% :     0.000155s :      1: jit_opt_after_cconv
 0.20% :     0.000059s :      1: jit_opt_b
 1.57% :     0.000472s :      1: loop_unroll
 2.32% :     0.000695s :      1: mutable_eliminate
 3.09% :     0.000925s :     26: opt.transform.jit_opt_a
 0.17% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000029s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.86% :     0.000556s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.21% :     0.000062s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 1.49% :     0.000446s :      1: renormalize.infer
 1.32% :     0.000396s :      1: renormalize.specialize
 0.05% :     0.000016s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000039s :      1: rewriter_after_opt_a
 0.27% :     0.000081s :      1: rewriter_before_opt_a
 0.31% :     0.000092s :      1: symbol_engine_optimizer
32.11% :     0.009628s :      1: task_emit
30.13% :     0.009032s :      1: type_inference
 0.21% :     0.000064s :      1: validate


TotalTime = 0.0279415, [33]
[bootstrap]: 0.00035496
[type_inference]: 0.00692995
[event_method]: 1.769e-05
[auto_monad]: 6.789e-05
[graph_reusing]: 5.92999e-06
[pre_auto_parallel]: 2.37001e-06
[py_interpret_to_execute]: 4.513e-05
[rewriter_before_opt_a]: 7.363e-05
[expand_dump_flag]: 3.45e-06
[jit_opt_a]: 0.00964775, [2]
    [Cycle 1]: 0.00209743, [27]
        [switch_simplify]: 7.403e-05
        [loop_unroll]: 3.231e-05
        [a_1]: 0.00065105
        [with_stream_mark]: 2.336e-05
        [recompute_prepare]: 8.00999e-06
        [updatestate_depend_eliminate]: 3.60998e-06
        [updatestate_assign_eliminate]: 3.61001e-06
        [updatestate_loads_eliminate]: 2.61999e-06
        [parameter_eliminate]: 2.18002e-06
        [specialize_transform]: 7.31001e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.81998e-06
        [meta_shard_fg_expand]: 2.32999e-06
        [get_grad_eliminate_]: 5.43002e-06
        [merge_forward]: 3.65998e-06
        [cell_reuse_recompute_pass]: 1.50999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.487e-05
        [j_node_and_user_rematch]: 9.49e-06
        [meta_fg_expand]: 2.59999e-06
        [replace_old_param]: 9.74e-06
        [inline_without_move]: 5.40001e-06
        [renormalize]: 0.00096591
        [add_forward_monad_depend]: 5.77001e-06
        [auto_monad_grad]: 2.94999e-06
        [auto_monad_eliminator]: 1.477e-05
        [cse]: 3.101e-05
        [replace_applicator]: 1.355e-05
    [Cycle 2]: 0.0003336, [27]
        [switch_simplify]: 5.74e-06
        [loop_unroll]: 5.61e-06
        [a_1]: 9.396e-05
        [with_stream_mark]: 1.01e-05
        [recompute_prepare]: 4.97e-06
        [updatestate_depend_eliminate]: 2.54999e-06
        [updatestate_assign_eliminate]: 2.11998e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 4.90999e-06
        [updatestate_useless_node_eliminater]: 5.13002e-06
        [accelerated_algorithm]: 5.35001e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.81e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.226e-05
        [j_node_and_user_rematch]: 8.25999e-06
        [meta_fg_expand]: 1.64e-06
        [replace_old_param]: 7.55e-06
        [inline_without_move]: 5.14998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 8.10018e-07
        [auto_monad_eliminator]: 5.57999e-06
        [cse]: 1.225e-05
        [replace_applicator]: 5.17999e-06
[py_interpret_to_execute_after_opt_a]: 1.265e-05
[rewriter_after_opt_a]: 3.744e-05
[convert_after_rewriter]: 7.36001e-06
[order_py_execute_after_rewriter]: 5.25001e-06
[mutable_eliminate]: 0.00075964
[jit_opt_b]: 5.486e-05, [1]
    [Cycle 1]: 4.663e-05, [2]
        [frontend_op_eliminate]: 1.962e-05
        [inline_after_opt_a]: 1.545e-05
[cconv]: 2.851e-05
[loop_unroll]: 0.00044306
[jit_opt_after_cconv]: 0.00014143, [1]
    [Cycle 1]: 0.00013452, [11]
        [c_1]: 2.15e-05
        [parameter_eliminate]: 2.59999e-06
        [updatestate_depend_eliminate]: 6.19001e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.786e-05
        [call_graph_tuple_transform]: 1.866e-05
        [tuple_list_get_item_eliminator]: 5.63997e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 5.43002e-06
[remove_dup_value]: 1.569e-05
[partial_unused_args_eliminate]: 2.84999e-06
[environ_conv]: 5.42001e-06
[add_recomputation]: 5.042e-05
[cse_after_recomputation]: 2.02e-05, [1]
    [Cycle 1]: 1.494e-05, [1]
        [cse]: 9.15001e-06
[auto_monad_reorder]: 1.61e-05
[get_jit_bprop_graph]: 2.04999e-06
[rewriter_after_jit_bprop_graph]: 5.97999e-06
[opt_after_jit_grad]: 0.00047569
[symbol_engine_optimizer]: 7.709e-05, [1]
    [Cycle 1]: 6.992e-05, [6]
        [build]: 4.90999e-06
        [elim_shapecalc]: 8.48001e-06
        [elim_not_effective]: 1.32e-05
        [opt_reshape]: 5.59998e-06
        [fold_const_symbol]: 9.44e-06
        [renormalize]: 7.00005e-07
[validate]: 3.931e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.008356
[execute]: 1.066e-05
Sums
    bootstrap                            :     0.000355s :  1.77%
    type_inference                       :     0.006930s : 34.53%
    event_method                         :     0.000018s :  0.09%
    auto_monad                           :     0.000068s :  0.34%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000045s :  0.22%
    rewriter_before_opt_a                :     0.000074s :  0.37%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000080s :  0.40%
    jit_opt_a.loop_unroll                :     0.000038s :  0.19%
    jit_opt_a.a_1                        :     0.000745s :  3.71%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.000966s :  4.81%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.10%
    jit_opt_a.cse                        :     0.000043s :  0.22%
    jit_opt_a.replace_applicator         :     0.000019s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.06%
    rewriter_after_opt_a                 :     0.000037s :  0.19%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000760s :  3.79%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000029s :  0.14%
    loop_unroll                          :     0.000443s :  2.21%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.25%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000476s :  2.37%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.20%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008356s : 41.64%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000272    23
 0.94% :     0.000003s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 2.09% :     0.000006s :      3: substitution.graph_param_transform
87.11% :     0.000237s :      4: substitution.inline
 1.16% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.68% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.66% :     0.000005s :      2: substitution.replace_old_param
 4.91% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006853     2
71.65% :     0.004910s :      1: type_inference.infer
28.35% :     0.001943s :      1: type_inference.specialize
------[replace.]   0.000086     6
68.43% :     0.000059s :      4: replace.inline
31.57% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000246     6
95.05% :     0.000234s :      4: match.inline
 4.95% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.49% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.30% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 2.12% :     0.000003s :     10: predicate.arithmetic_simplify
 1.12% :     0.000001s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.26% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.42% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.54% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.81% :     0.000007s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000002s :      6: predicate.less_batch_normalization
 1.32% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.33% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 2.25% :     0.000003s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.83% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.54% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.41% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.03% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 1.13% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.53% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.84% :     0.000010s :     53: predicate.switch_simplify
 1.28% :     0.000002s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.61% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000733    12
42.40% :     0.000311s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.22% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.39% :     0.000369s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.029988    76
 0.18% :     0.000053s :      1: add_recomputation
 0.24% :     0.000071s :      1: auto_monad
 0.06% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.25% :     0.000375s :      1: bootstrap
 0.10% :     0.000031s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.07% :     0.000021s :      1: event_method
 0.05% :     0.000016s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
32.19% :     0.009652s :      1: jit_opt_a
 0.48% :     0.000144s :      1: jit_opt_after_cconv
 0.19% :     0.000058s :      1: jit_opt_b
 1.50% :     0.000451s :      1: loop_unroll
 2.56% :     0.000769s :      1: mutable_eliminate
 3.27% :     0.000981s :     26: opt.transform.jit_opt_a
 0.16% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000028s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.61% :     0.000484s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000048s :      1: py_interpret_to_execute
 0.05% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000018s :      1: remove_dup_value
 1.79% :     0.000537s :      1: renormalize.infer
 1.40% :     0.000420s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000040s :      1: rewriter_after_opt_a
 0.25% :     0.000076s :      1: rewriter_before_opt_a
 0.27% :     0.000080s :      1: symbol_engine_optimizer
27.94% :     0.008378s :      1: task_emit
23.15% :     0.006943s :      1: type_inference
 0.20% :     0.000060s :      1: validate


TotalTime = 0.0134314, [30]
[bootstrap]: 0.00024447
[type_inference]: 0.00589889
[event_method]: 1.996e-05
[auto_monad]: 8.191e-05
[graph_reusing]: 6.23e-06
[pre_auto_parallel]: 2.81999e-06
[py_interpret_to_execute]: 4.431e-05
[rewriter_before_opt_a]: 8.194e-05
[expand_dump_flag]: 3.04001e-06
[jit_opt_a]: 0.00506635, [2]
    [Cycle 1]: 0.0016969, [27]
        [switch_simplify]: 0.00010979
        [loop_unroll]: 3.318e-05
        [a_1]: 0.00065663
        [with_stream_mark]: 1.597e-05
        [recompute_prepare]: 7.45998e-06
        [updatestate_depend_eliminate]: 3.46001e-06
        [updatestate_assign_eliminate]: 2.86999e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [parameter_eliminate]: 1.85001e-06
        [specialize_transform]: 6.20002e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 5.40001e-06
        [meta_shard_fg_expand]: 1.99999e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 3.94002e-06
        [cell_reuse_recompute_pass]: 9.69972e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.431e-05
        [j_node_and_user_rematch]: 8.93002e-06
        [meta_fg_expand]: 2.36998e-06
        [replace_old_param]: 8.70001e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.00056821
        [add_forward_monad_depend]: 5.37999e-06
        [auto_monad_grad]: 2.59999e-06
        [auto_monad_eliminator]: 1.275e-05
        [cse]: 2.695e-05
        [replace_applicator]: 1.219e-05
    [Cycle 2]: 0.00031555, [27]
        [switch_simplify]: 6.34001e-06
        [loop_unroll]: 5.27999e-06
        [a_1]: 9.307e-05
        [with_stream_mark]: 8.94998e-06
        [recompute_prepare]: 5.00001e-06
        [updatestate_depend_eliminate]: 2.71e-06
        [updatestate_assign_eliminate]: 2.00002e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.72998e-06
        [updatestate_useless_node_eliminater]: 4.79998e-06
        [accelerated_algorithm]: 4.97e-06
        [meta_shard_fg_expand]: 1.36002e-06
        [get_grad_eliminate_]: 4.67e-06
        [merge_forward]: 2.79001e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.203e-05
        [j_node_and_user_rematch]: 7.61999e-06
        [meta_fg_expand]: 1.42999e-06
        [replace_old_param]: 6.61e-06
        [inline_without_move]: 4.78001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 5.39e-06
        [cse]: 9.37001e-06
        [replace_applicator]: 5.09e-06
[py_interpret_to_execute_after_opt_a]: 8.86002e-06
[rewriter_after_opt_a]: 3.014e-05
[convert_after_rewriter]: 6.49001e-06
[order_py_execute_after_rewriter]: 5.09998e-06
[mutable_eliminate]: 0.00049014
[jit_opt_b]: 4.73e-05, [1]
    [Cycle 1]: 4.059e-05, [2]
        [frontend_op_eliminate]: 1.525e-05
        [inline_after_opt_a]: 1.395e-05
[cconv]: 2.461e-05
[loop_unroll]: 0.00042024
[jit_opt_after_cconv]: 0.00013451, [1]
    [Cycle 1]: 0.00012746, [11]
        [c_1]: 2.108e-05
        [parameter_eliminate]: 2.44001e-06
        [updatestate_depend_eliminate]: 5.34e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.561e-05
        [call_graph_tuple_transform]: 1.725e-05
        [tuple_list_get_item_eliminator]: 5.09e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 4.60015e-07
        [switch_simplify]: 5.17999e-06
[remove_dup_value]: 1.31e-05
[partial_unused_args_eliminate]: 2.41998e-06
[environ_conv]: 4.90001e-06
[add_recomputation]: 4.672e-05
[cse_after_recomputation]: 1.984e-05, [1]
    [Cycle 1]: 1.479e-05, [1]
        [cse]: 9.00001e-06
[auto_monad_reorder]: 1.471e-05
[get_jit_bprop_graph]: 1.70001e-06
[rewriter_after_jit_bprop_graph]: 5.27999e-06
[opt_after_jit_grad]: 0.00044181
[symbol_engine_optimizer]: 7.11e-05, [1]
    [Cycle 1]: 6.512e-05, [6]
        [build]: 3.40998e-06
        [elim_shapecalc]: 7.14001e-06
        [elim_not_effective]: 1.236e-05
        [opt_reshape]: 5.68002e-06
        [fold_const_symbol]: 8.25e-06
        [renormalize]: 4.30009e-07
[validate]: 3.132e-05
Sums
    bootstrap                            :     0.000244s :  2.49%
    type_inference                       :     0.005899s : 60.07%
    event_method                         :     0.000020s :  0.20%
    auto_monad                           :     0.000082s :  0.83%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000044s :  0.45%
    rewriter_before_opt_a                :     0.000082s :  0.83%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000116s :  1.18%
    jit_opt_a.loop_unroll                :     0.000038s :  0.39%
    jit_opt_a.a_1                        :     0.000750s :  7.63%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000568s :  5.79%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.18%
    jit_opt_a.cse                        :     0.000036s :  0.37%
    jit_opt_a.replace_applicator         :     0.000017s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.09%
    rewriter_after_opt_a                 :     0.000030s :  0.31%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000490s :  4.99%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.14%
    cconv                                :     0.000025s :  0.25%
    loop_unroll                          :     0.000420s :  4.28%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000047s :  0.48%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000442s :  4.50%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.32%


Time group info:
------[substitution.]   0.000245    26
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.04% :     0.000005s :      3: substitution.graph_param_transform
83.94% :     0.000205s :      6: substitution.inline
 1.13% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.80% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.45% :     0.000004s :      2: substitution.replace_old_param
 3.18% :     0.000008s :      1: substitution.switch_simplify
 5.19% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005843     2
83.78% :     0.004895s :      1: type_inference.infer
16.22% :     0.000948s :      1: type_inference.specialize
------[replace.]   0.000126     9
51.43% :     0.000065s :      6: replace.inline
25.76% :     0.000033s :      1: replace.switch_simplify
22.81% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000221     9
91.52% :     0.000202s :      6: match.inline
 3.20% :     0.000007s :      1: match.switch_simplify
 5.28% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   801
 1.41% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.11% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     12: predicate.addn_check_dump
 1.58% :     0.000002s :     12: predicate.addn_zero_filter
 1.88% :     0.000002s :     12: predicate.arithmetic_simplify
 1.33% :     0.000002s :     12: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000002s :     12: predicate.compare_switch_simplify
 1.62% :     0.000002s :     12: predicate.depend_value_elim
 1.22% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.21% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     12: predicate.environ_get_eliminate
 1.14% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.60% :     0.000007s :     26: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     17: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.94% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000002s :     12: predicate.merge_addn
 1.28% :     0.000002s :     12: predicate.minmaximum_grad
 1.89% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     17: predicate.partial_eliminate
 1.38% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     12: predicate.reduce_eliminate
 1.52% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     20: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     12: predicate.reshape_eliminate
 1.34% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     12: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.72% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.94% :     0.000004s :     20: predicate.switch_defer_inline
 2.43% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.32% :     0.000011s :     65: predicate.switch_simplify
 1.17% :     0.000002s :     12: predicate.tile_eliminate
 1.45% :     0.000002s :     12: predicate.transpose_eliminate
 1.50% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.40% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.38% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     12: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000709    15
50.81% :     0.000360s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.92% :     0.000077s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.28% :     0.000271s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015077    72
 0.33% :     0.000050s :      1: add_recomputation
 0.57% :     0.000086s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.73% :     0.000261s :      1: bootstrap
 0.18% :     0.000028s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
33.62% :     0.005069s :      1: jit_opt_a
 0.91% :     0.000137s :      1: jit_opt_after_cconv
 0.33% :     0.000050s :      1: jit_opt_b
 2.84% :     0.000428s :      1: loop_unroll
 3.31% :     0.000499s :      1: mutable_eliminate
 6.73% :     0.001014s :     26: opt.transform.jit_opt_a
 0.30% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000023s :      4: opt.transform.jit_opt_b
 0.09% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.98% :     0.000449s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000047s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000015s :      1: remove_dup_value
 2.06% :     0.000311s :      1: renormalize.infer
 1.66% :     0.000251s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000033s :      1: rewriter_after_opt_a
 0.56% :     0.000085s :      1: rewriter_before_opt_a
 0.49% :     0.000074s :      1: symbol_engine_optimizer
39.21% :     0.005912s :      1: type_inference


TotalTime = 0.0244028, [33]
[bootstrap]: 0.00032414
[type_inference]: 0.00536787
[event_method]: 2.267e-05
[auto_monad]: 7.395e-05
[graph_reusing]: 6.41998e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 6.353e-05
[rewriter_before_opt_a]: 8.274e-05
[expand_dump_flag]: 3.22002e-06
[jit_opt_a]: 0.00791527, [2]
    [Cycle 1]: 0.00228206, [27]
        [switch_simplify]: 7.407e-05
        [loop_unroll]: 2.998e-05
        [a_1]: 0.00067092
        [with_stream_mark]: 2.907e-05
        [recompute_prepare]: 9.81e-06
        [updatestate_depend_eliminate]: 5.47001e-06
        [updatestate_assign_eliminate]: 3.11999e-06
        [updatestate_loads_eliminate]: 3.11001e-06
        [parameter_eliminate]: 2.36e-06
        [specialize_transform]: 6.88e-06
        [updatestate_useless_node_eliminater]: 5.38002e-06
        [accelerated_algorithm]: 6.06e-06
        [meta_shard_fg_expand]: 2.83e-06
        [get_grad_eliminate_]: 5.73002e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 1.99999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.828e-05
        [j_node_and_user_rematch]: 1.19e-05
        [meta_fg_expand]: 2.19001e-06
        [replace_old_param]: 1.066e-05
        [inline_without_move]: 5.54e-06
        [renormalize]: 0.00106252
        [add_forward_monad_depend]: 7.57002e-06
        [auto_monad_grad]: 2.93e-06
        [auto_monad_eliminator]: 1.719e-05
        [cse]: 3.737e-05
        [replace_applicator]: 1.788e-05
    [Cycle 2]: 0.00037212, [27]
        [switch_simplify]: 6.34001e-06
        [loop_unroll]: 5.91e-06
        [a_1]: 0.00010185
        [with_stream_mark]: 1.435e-05
        [recompute_prepare]: 5.47999e-06
        [updatestate_depend_eliminate]: 2.81e-06
        [updatestate_assign_eliminate]: 2.74999e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 1.75001e-06
        [specialize_transform]: 5.37001e-06
        [updatestate_useless_node_eliminater]: 5.32999e-06
        [accelerated_algorithm]: 6.24001e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 5.64e-06
        [merge_forward]: 3.61999e-06
        [cell_reuse_recompute_pass]: 2.70997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.625e-05
        [j_node_and_user_rematch]: 9.42001e-06
        [meta_fg_expand]: 1.99999e-06
        [replace_old_param]: 8.67e-06
        [inline_without_move]: 4.60001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.59998e-06
        [auto_monad_grad]: 9.60019e-07
        [auto_monad_eliminator]: 5.27999e-06
        [cse]: 1.133e-05
        [replace_applicator]: 5.09e-06
[py_interpret_to_execute_after_opt_a]: 1.534e-05
[rewriter_after_opt_a]: 4.283e-05
[convert_after_rewriter]: 8.3e-06
[order_py_execute_after_rewriter]: 5.44e-06
[mutable_eliminate]: 0.00084442
[jit_opt_b]: 6.182e-05, [1]
    [Cycle 1]: 4.948e-05, [2]
        [frontend_op_eliminate]: 2.004e-05
        [inline_after_opt_a]: 1.668e-05
[cconv]: 3.364e-05
[loop_unroll]: 0.00052582
[jit_opt_after_cconv]: 0.00017292, [1]
    [Cycle 1]: 0.00016524, [11]
        [c_1]: 2.63e-05
        [parameter_eliminate]: 3.95e-06
        [updatestate_depend_eliminate]: 8.43999e-06
        [updatestate_assign_eliminate]: 3.13e-06
        [updatestate_loads_eliminate]: 2.80997e-06
        [cse]: 2.694e-05
        [call_graph_tuple_transform]: 2.437e-05
        [tuple_list_get_item_eliminator]: 5.86e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 8.59989e-07
        [switch_simplify]: 5.56e-06
[remove_dup_value]: 1.581e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 5.79e-06
[add_recomputation]: 5.891e-05
[cse_after_recomputation]: 2.419e-05, [1]
    [Cycle 1]: 1.756e-05, [1]
        [cse]: 1.004e-05
[auto_monad_reorder]: 1.611e-05
[get_jit_bprop_graph]: 2.76e-06
[rewriter_after_jit_bprop_graph]: 9.30001e-06
[opt_after_jit_grad]: 0.00057619
[symbol_engine_optimizer]: 8.198e-05, [1]
    [Cycle 1]: 7.458e-05, [6]
        [build]: 5.57001e-06
        [elim_shapecalc]: 8.35999e-06
        [elim_not_effective]: 1.422e-05
        [opt_reshape]: 6.85002e-06
        [fold_const_symbol]: 9.41e-06
        [renormalize]: 4.80009e-07
[validate]: 5.663e-05
[backend_pass]: 1.49998e-06
[task_emit]: 0.00769158
[execute]: 1.075e-05
Sums
    bootstrap                            :     0.000324s :  1.76%
    type_inference                       :     0.005368s : 29.23%
    event_method                         :     0.000023s :  0.12%
    auto_monad                           :     0.000074s :  0.40%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000064s :  0.35%
    rewriter_before_opt_a                :     0.000083s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000080s :  0.44%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000773s :  4.21%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001063s :  5.79%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.12%
    jit_opt_a.cse                        :     0.000049s :  0.27%
    jit_opt_a.replace_applicator         :     0.000023s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.08%
    rewriter_after_opt_a                 :     0.000043s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000844s :  4.60%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000034s :  0.18%
    loop_unroll                          :     0.000526s :  2.86%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000027s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000059s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000576s :  3.14%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000057s :  0.31%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007692s : 41.88%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000300    23
 0.64% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000002s :      2: substitution.fold_const_symbol
 2.17% :     0.000007s :      3: substitution.graph_param_transform
85.86% :     0.000258s :      4: substitution.inline
 1.64% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.97% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.12% :     0.000006s :      2: substitution.replace_old_param
 5.07% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005271     2
79.92% :     0.004213s :      1: type_inference.infer
20.08% :     0.001059s :      1: type_inference.specialize
------[replace.]   0.000088     6
66.98% :     0.000059s :      4: replace.inline
33.02% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000269     6
94.76% :     0.000255s :      4: match.inline
 5.24% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.58% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.88% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 2.17% :     0.000003s :     10: predicate.arithmetic_simplify
 1.30% :     0.000002s :     10: predicate.cast_eliminate
 0.64% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.90% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 0.96% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.97% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.44% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.81% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.07% :     0.000001s :     10: predicate.environ_get_eliminate
 1.64% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.08% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.33% :     0.000007s :     22: predicate.inline
 0.73% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.53% :     0.000002s :     15: predicate.load_eliminater
 2.10% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.18% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.93% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 3.13% :     0.000004s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 1.92% :     0.000003s :     15: predicate.partial_eliminate
 1.26% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.79% :     0.000002s :     10: predicate.reduce_eliminate
 1.27% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.49% :     0.000001s :      3: predicate.reset_defer_inline
 1.59% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.93% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000002s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.16% :     0.000002s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.23% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 2.42% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.38% :     0.000010s :     53: predicate.switch_simplify
 1.12% :     0.000002s :     10: predicate.tile_eliminate
 0.95% :     0.000001s :     10: predicate.transpose_eliminate
 1.20% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.02% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.86% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.90% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.44% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000765    12
45.19% :     0.000346s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.76% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.05% :     0.000360s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026608    76
 0.24% :     0.000063s :      1: add_recomputation
 0.29% :     0.000078s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.30% :     0.000347s :      1: bootstrap
 0.14% :     0.000036s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000027s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.11% :     0.000029s :      1: event_method
 0.06% :     0.000017s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
29.77% :     0.007921s :      1: jit_opt_a
 0.66% :     0.000176s :      1: jit_opt_after_cconv
 0.24% :     0.000065s :      1: jit_opt_b
 2.03% :     0.000539s :      1: loop_unroll
 3.22% :     0.000858s :      1: mutable_eliminate
 3.85% :     0.001025s :     26: opt.transform.jit_opt_a
 0.22% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000030s :      4: opt.transform.jit_opt_b
 0.06% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000023s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000589s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000067s :      1: py_interpret_to_execute
 0.07% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.16% :     0.000576s :      1: renormalize.infer
 1.78% :     0.000475s :      1: renormalize.specialize
 0.04% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000047s :      1: rewriter_after_opt_a
 0.32% :     0.000086s :      1: rewriter_before_opt_a
 0.32% :     0.000085s :      1: symbol_engine_optimizer
28.99% :     0.007714s :      1: task_emit
20.26% :     0.005390s :      1: type_inference
 0.32% :     0.000084s :      1: validate


TotalTime = 0.0230565, [33]
[bootstrap]: 0.00041837
[type_inference]: 0.00636716
[event_method]: 2.024e-05
[auto_monad]: 7.419e-05
[graph_reusing]: 6.63e-06
[pre_auto_parallel]: 3.10002e-06
[py_interpret_to_execute]: 5.124e-05
[rewriter_before_opt_a]: 7.929e-05
[expand_dump_flag]: 3.57002e-06
[jit_opt_a]: 0.00683172, [2]
    [Cycle 1]: 0.00203167, [27]
        [switch_simplify]: 7.411e-05
        [loop_unroll]: 3.718e-05
        [a_1]: 0.00061267
        [with_stream_mark]: 2.278e-05
        [recompute_prepare]: 7.25998e-06
        [updatestate_depend_eliminate]: 3.45e-06
        [updatestate_assign_eliminate]: 2.98998e-06
        [updatestate_loads_eliminate]: 2.84001e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 6.47001e-06
        [updatestate_useless_node_eliminater]: 5.37001e-06
        [accelerated_algorithm]: 6.01998e-06
        [meta_shard_fg_expand]: 2.44001e-06
        [get_grad_eliminate_]: 6.24001e-06
        [merge_forward]: 3.86001e-06
        [cell_reuse_recompute_pass]: 1.62001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.618e-05
        [j_node_and_user_rematch]: 9.39998e-06
        [meta_fg_expand]: 2.61e-06
        [replace_old_param]: 9.57999e-06
        [inline_without_move]: 5.57001e-06
        [renormalize]: 0.00092664
        [add_forward_monad_depend]: 5.80002e-06
        [auto_monad_grad]: 2.65997e-06
        [auto_monad_eliminator]: 1.515e-05
        [cse]: 4.061e-05
        [replace_applicator]: 1.396e-05
    [Cycle 2]: 0.00034075, [27]
        [switch_simplify]: 6.28002e-06
        [loop_unroll]: 5.44e-06
        [a_1]: 9.488e-05
        [with_stream_mark]: 9.47001e-06
        [recompute_prepare]: 5.51e-06
        [updatestate_depend_eliminate]: 2.63998e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [parameter_eliminate]: 1.35001e-06
        [specialize_transform]: 4.95001e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.54e-06
        [get_grad_eliminate_]: 4.81002e-06
        [merge_forward]: 2.99999e-06
        [cell_reuse_recompute_pass]: 2.35002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.392e-05
        [j_node_and_user_rematch]: 7.99997e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 8.42e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 9.30013e-07
        [auto_monad_eliminator]: 5.12999e-06
        [cse]: 1.095e-05
        [replace_applicator]: 5.60001e-06
[py_interpret_to_execute_after_opt_a]: 1.275e-05
[rewriter_after_opt_a]: 3.765e-05
[convert_after_rewriter]: 7.33999e-06
[order_py_execute_after_rewriter]: 5.49e-06
[mutable_eliminate]: 0.00068874
[jit_opt_b]: 5.326e-05, [1]
    [Cycle 1]: 4.503e-05, [2]
        [frontend_op_eliminate]: 1.805e-05
        [inline_after_opt_a]: 1.569e-05
[cconv]: 2.755e-05
[loop_unroll]: 0.00060236
[jit_opt_after_cconv]: 0.00015752, [1]
    [Cycle 1]: 0.00015054, [11]
        [c_1]: 2.64e-05
        [parameter_eliminate]: 2.69999e-06
        [updatestate_depend_eliminate]: 5.35999e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [cse]: 1.853e-05
        [call_graph_tuple_transform]: 1.945e-05
        [tuple_list_get_item_eliminator]: 5.25999e-06
        [none_parameter_eliminate]: 2.01e-06
        [renormalize]: 8.70001e-07
        [switch_simplify]: 5.84999e-06
[remove_dup_value]: 1.454e-05
[partial_unused_args_eliminate]: 2.43002e-06
[environ_conv]: 5.44e-06
[add_recomputation]: 5.056e-05
[cse_after_recomputation]: 1.98e-05, [1]
    [Cycle 1]: 1.416e-05, [1]
        [cse]: 8.84e-06
[auto_monad_reorder]: 1.643e-05
[get_jit_bprop_graph]: 1.99999e-06
[rewriter_after_jit_bprop_graph]: 6.28e-06
[opt_after_jit_grad]: 0.00048867
[symbol_engine_optimizer]: 7.552e-05, [1]
    [Cycle 1]: 6.885e-05, [6]
        [build]: 3.56001e-06
        [elim_shapecalc]: 7.83001e-06
        [elim_not_effective]: 1.34e-05
        [opt_reshape]: 5.92999e-06
        [fold_const_symbol]: 9.50001e-06
        [renormalize]: 9.50007e-07
[validate]: 3.66e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00665421
[execute]: 8.53001e-06
Sums
    bootstrap                            :     0.000418s :  2.33%
    type_inference                       :     0.006367s : 35.51%
    event_method                         :     0.000020s :  0.11%
    auto_monad                           :     0.000074s :  0.41%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000051s :  0.29%
    rewriter_before_opt_a                :     0.000079s :  0.44%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000080s :  0.45%
    jit_opt_a.loop_unroll                :     0.000043s :  0.24%
    jit_opt_a.a_1                        :     0.000708s :  3.95%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000927s :  5.17%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.11%
    jit_opt_a.cse                        :     0.000052s :  0.29%
    jit_opt_a.replace_applicator         :     0.000020s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000689s :  3.84%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000028s :  0.15%
    loop_unroll                          :     0.000602s :  3.36%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000051s :  0.28%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000489s :  2.73%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000037s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006654s : 37.11%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000248    23
 0.90% :     0.000002s :      2: substitution.elim_not_effective
 0.82% :     0.000002s :      2: substitution.fold_const_symbol
 2.43% :     0.000006s :      3: substitution.graph_param_transform
85.01% :     0.000211s :      4: substitution.inline
 1.31% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.87% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.10% :     0.000005s :      2: substitution.replace_old_param
 5.55% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006273     2
82.78% :     0.005193s :      1: type_inference.infer
17.22% :     0.001080s :      1: type_inference.specialize
------[replace.]   0.000083     6
67.58% :     0.000056s :      4: replace.inline
32.42% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000220     6
94.41% :     0.000208s :      4: match.inline
 5.59% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.43% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000002s :     10: predicate.addn_zero_filter
 2.31% :     0.000003s :     10: predicate.arithmetic_simplify
 1.14% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.73% :     0.000002s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.09% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.46% :     0.000002s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.39% :     0.000002s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.66% :     0.000007s :     22: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.99% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.60% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.33% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.49% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.06% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.99% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.30% :     0.000002s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.67% :     0.000003s :     16: predicate.switch_defer_inline
 1.92% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.01% :     0.000010s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.30% :     0.000002s :     10: predicate.transpose_eliminate
 1.80% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.05% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.73% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.08% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000780    12
51.51% :     0.000402s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.49% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.00% :     0.000343s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025032    76
 0.21% :     0.000053s :      1: add_recomputation
 0.31% :     0.000078s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.76% :     0.000441s :      1: bootstrap
 0.12% :     0.000030s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000025s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
27.31% :     0.006836s :      1: jit_opt_a
 0.64% :     0.000160s :      1: jit_opt_after_cconv
 0.22% :     0.000056s :      1: jit_opt_b
 2.45% :     0.000612s :      1: loop_unroll
 2.79% :     0.000698s :      1: mutable_eliminate
 3.80% :     0.000951s :     26: opt.transform.jit_opt_a
 0.21% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.99% :     0.000497s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000054s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.03% :     0.000507s :      1: renormalize.infer
 1.64% :     0.000410s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000041s :      1: rewriter_after_opt_a
 0.33% :     0.000082s :      1: rewriter_before_opt_a
 0.31% :     0.000078s :      1: symbol_engine_optimizer
26.64% :     0.006668s :      1: task_emit
25.51% :     0.006385s :      1: type_inference
 0.22% :     0.000055s :      1: validate


TotalTime = 0.0243139, [33]
[bootstrap]: 0.00040265
[type_inference]: 0.00649417
[event_method]: 1.796e-05
[auto_monad]: 6.757e-05
[graph_reusing]: 5.21002e-06
[pre_auto_parallel]: 2.32999e-06
[py_interpret_to_execute]: 3.873e-05
[rewriter_before_opt_a]: 7.24e-05
[expand_dump_flag]: 3.24001e-06
[jit_opt_a]: 0.00693067, [2]
    [Cycle 1]: 0.00192312, [27]
        [switch_simplify]: 5.871e-05
        [loop_unroll]: 3.145e-05
        [a_1]: 0.00057845
        [with_stream_mark]: 1.901e-05
        [recompute_prepare]: 7.90998e-06
        [updatestate_depend_eliminate]: 3.65998e-06
        [updatestate_assign_eliminate]: 3.45e-06
        [updatestate_loads_eliminate]: 3.04999e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 6.26e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 5.88002e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 5.76e-06
        [merge_forward]: 3.73001e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.483e-05
        [j_node_and_user_rematch]: 9.69e-06
        [meta_fg_expand]: 2.36e-06
        [replace_old_param]: 8.62998e-06
        [inline_without_move]: 5.58002e-06
        [renormalize]: 0.00089712
        [add_forward_monad_depend]: 6.36e-06
        [auto_monad_grad]: 3.06999e-06
        [auto_monad_eliminator]: 1.644e-05
        [cse]: 2.341e-05
        [replace_applicator]: 1.531e-05
    [Cycle 2]: 0.00034047, [27]
        [switch_simplify]: 5.57001e-06
        [loop_unroll]: 5.40001e-06
        [a_1]: 9.727e-05
        [with_stream_mark]: 1.016e-05
        [recompute_prepare]: 4.97e-06
        [updatestate_depend_eliminate]: 2.84999e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 5.20999e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 5.37999e-06
        [meta_shard_fg_expand]: 1.59e-06
        [get_grad_eliminate_]: 4.74998e-06
        [merge_forward]: 2.90002e-06
        [cell_reuse_recompute_pass]: 2.21e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.419e-05
        [j_node_and_user_rematch]: 8.17e-06
        [meta_fg_expand]: 1.74998e-06
        [replace_old_param]: 7.46001e-06
        [inline_without_move]: 5.12e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 9.89996e-07
        [auto_monad_eliminator]: 5.70001e-06
        [cse]: 1.011e-05
        [replace_applicator]: 1.323e-05
[py_interpret_to_execute_after_opt_a]: 1.153e-05
[rewriter_after_opt_a]: 3.661e-05
[convert_after_rewriter]: 6.89001e-06
[order_py_execute_after_rewriter]: 5.35999e-06
[mutable_eliminate]: 0.00061579
[jit_opt_b]: 5.59e-05, [1]
    [Cycle 1]: 4.871e-05, [2]
        [frontend_op_eliminate]: 1.675e-05
        [inline_after_opt_a]: 1.65e-05
[cconv]: 2.824e-05
[loop_unroll]: 0.00044262
[jit_opt_after_cconv]: 0.00159854, [1]
    [Cycle 1]: 0.00159027, [11]
        [c_1]: 4.217e-05
        [parameter_eliminate]: 4.12e-06
        [updatestate_depend_eliminate]: 7.5e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [cse]: 2.493e-05
        [call_graph_tuple_transform]: 2.528e-05
        [tuple_list_get_item_eliminator]: 5.99e-06
        [none_parameter_eliminate]: 1.76998e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 6.02999e-06
[remove_dup_value]: 1.566e-05
[partial_unused_args_eliminate]: 2.59001e-06
[environ_conv]: 6.48e-06
[add_recomputation]: 5.241e-05
[cse_after_recomputation]: 2.326e-05, [1]
    [Cycle 1]: 1.636e-05, [1]
        [cse]: 1.018e-05
[auto_monad_reorder]: 1.65e-05
[get_jit_bprop_graph]: 2.17001e-06
[rewriter_after_jit_bprop_graph]: 7.08998e-06
[opt_after_jit_grad]: 0.00061403
[symbol_engine_optimizer]: 8.108e-05, [1]
    [Cycle 1]: 7.286e-05, [6]
        [build]: 3.64002e-06
        [elim_shapecalc]: 8.3e-06
        [elim_not_effective]: 1.443e-05
        [opt_reshape]: 6.64001e-06
        [fold_const_symbol]: 1.01e-05
        [renormalize]: 5.00004e-07
[validate]: 4.116e-05
[backend_pass]: 1.24e-06
[task_emit]: 0.00637767
[execute]: 9.30001e-06
Sums
    bootstrap                            :     0.000403s :  2.29%
    type_inference                       :     0.006494s : 36.95%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000068s :  0.38%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.22%
    rewriter_before_opt_a                :     0.000072s :  0.41%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.37%
    jit_opt_a.loop_unroll                :     0.000037s :  0.21%
    jit_opt_a.a_1                        :     0.000676s :  3.84%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000897s :  5.11%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.13%
    jit_opt_a.cse                        :     0.000034s :  0.19%
    jit_opt_a.replace_applicator         :     0.000029s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000616s :  3.50%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000028s :  0.16%
    loop_unroll                          :     0.000443s :  2.52%
    jit_opt_after_cconv.c_1              :     0.000042s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000052s :  0.30%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000614s :  3.49%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006378s : 36.29%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000223    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.97% :     0.000002s :      2: substitution.fold_const_symbol
 2.79% :     0.000006s :      3: substitution.graph_param_transform
84.18% :     0.000188s :      4: substitution.inline
 1.60% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.91% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.64% :     0.000004s :      2: substitution.replace_old_param
 5.93% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006422     2
85.09% :     0.005464s :      1: type_inference.infer
14.91% :     0.000958s :      1: type_inference.specialize
------[replace.]   0.000076     6
66.18% :     0.000050s :      4: replace.inline
33.82% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000197     6
93.92% :     0.000185s :      4: match.inline
 6.08% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.53% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.52% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.21% :     0.000002s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.49% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.12% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000001s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000002s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 6.35% :     0.000008s :     22: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.31% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.70% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.55% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.21% :     0.000010s :     53: predicate.switch_simplify
 1.44% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.20% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.26% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000638    12
47.01% :     0.000300s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.03% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.97% :     0.000306s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026238    76
 0.21% :     0.000055s :      1: add_recomputation
 0.27% :     0.000071s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.62% :     0.000425s :      1: bootstrap
 0.12% :     0.000031s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.08% :     0.000022s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
26.43% :     0.006934s :      1: jit_opt_a
 6.11% :     0.001603s :      1: jit_opt_after_cconv
 0.22% :     0.000059s :      1: jit_opt_b
 1.72% :     0.000451s :      1: loop_unroll
 2.38% :     0.000625s :      1: mutable_eliminate
 3.45% :     0.000905s :     26: opt.transform.jit_opt_a
 0.27% :     0.000071s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.37% :     0.000623s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000042s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 1.75% :     0.000459s :      1: renormalize.infer
 1.64% :     0.000430s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000040s :      1: rewriter_after_opt_a
 0.29% :     0.000075s :      1: rewriter_before_opt_a
 0.32% :     0.000084s :      1: symbol_engine_optimizer
24.37% :     0.006395s :      1: task_emit
24.80% :     0.006507s :      1: type_inference
 0.24% :     0.000062s :      1: validate


TotalTime = 0.0132733, [30]
[bootstrap]: 0.00025108
[type_inference]: 0.00576103
[event_method]: 2.084e-05
[auto_monad]: 8.451e-05
[graph_reusing]: 6.24999e-06
[pre_auto_parallel]: 2.71e-06
[py_interpret_to_execute]: 4.605e-05
[rewriter_before_opt_a]: 8.816e-05
[expand_dump_flag]: 3.68999e-06
[jit_opt_a]: 0.00505048, [2]
    [Cycle 1]: 0.00168787, [27]
        [switch_simplify]: 0.000109
        [loop_unroll]: 3.421e-05
        [a_1]: 0.00066852
        [with_stream_mark]: 1.708e-05
        [recompute_prepare]: 7.35998e-06
        [updatestate_depend_eliminate]: 3.61999e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [parameter_eliminate]: 2.22999e-06
        [specialize_transform]: 5.82001e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.32999e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 4.99998e-06
        [merge_forward]: 3.4e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.36e-05
        [j_node_and_user_rematch]: 9.59e-06
        [meta_fg_expand]: 2.54999e-06
        [replace_old_param]: 8.25999e-06
        [inline_without_move]: 5.11002e-06
        [renormalize]: 0.00054719
        [add_forward_monad_depend]: 5.49e-06
        [auto_monad_grad]: 1.75001e-06
        [auto_monad_eliminator]: 1.314e-05
        [cse]: 2.767e-05
        [replace_applicator]: 1.106e-05
    [Cycle 2]: 0.00030591, [27]
        [switch_simplify]: 5.37999e-06
        [loop_unroll]: 5.12e-06
        [a_1]: 8.804e-05
        [with_stream_mark]: 8.38001e-06
        [recompute_prepare]: 4.92e-06
        [updatestate_depend_eliminate]: 2.59001e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 9.30013e-07
        [specialize_transform]: 4.92999e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 4.70999e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 4.48001e-06
        [merge_forward]: 2.38998e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.222e-05
        [j_node_and_user_rematch]: 7.48999e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 6.96001e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 6.40022e-07
        [auto_monad_eliminator]: 4.84e-06
        [cse]: 9.19e-06
        [replace_applicator]: 4.89e-06
[py_interpret_to_execute_after_opt_a]: 9.31e-06
[rewriter_after_opt_a]: 3.118e-05
[convert_after_rewriter]: 6.59001e-06
[order_py_execute_after_rewriter]: 4.61002e-06
[mutable_eliminate]: 0.00047096
[jit_opt_b]: 5.926e-05, [1]
    [Cycle 1]: 5.225e-05, [2]
        [frontend_op_eliminate]: 1.576e-05
        [inline_after_opt_a]: 1.658e-05
[cconv]: 2.326e-05
[loop_unroll]: 0.0004037
[jit_opt_after_cconv]: 0.0001333, [1]
    [Cycle 1]: 0.00012659, [11]
        [c_1]: 2.053e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 5.39e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [cse]: 1.572e-05
        [call_graph_tuple_transform]: 1.732e-05
        [tuple_list_get_item_eliminator]: 5.12e-06
        [none_parameter_eliminate]: 1.49998e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 5.24998e-06
[remove_dup_value]: 1.277e-05
[partial_unused_args_eliminate]: 2.22001e-06
[environ_conv]: 4.73001e-06
[add_recomputation]: 4.725e-05
[cse_after_recomputation]: 1.963e-05, [1]
    [Cycle 1]: 1.415e-05, [1]
        [cse]: 8.69e-06
[auto_monad_reorder]: 1.418e-05
[get_jit_bprop_graph]: 1.37e-06
[rewriter_after_jit_bprop_graph]: 5.40001e-06
[opt_after_jit_grad]: 0.00044675
[symbol_engine_optimizer]: 7.07e-05, [1]
    [Cycle 1]: 6.471e-05, [6]
        [build]: 3.25002e-06
        [elim_shapecalc]: 7.8e-06
        [elim_not_effective]: 1.241e-05
        [opt_reshape]: 5.55001e-06
        [fold_const_symbol]: 8.25e-06
        [renormalize]: 5.09986e-07
[validate]: 3.097e-05
Sums
    bootstrap                            :     0.000251s :  2.60%
    type_inference                       :     0.005761s : 59.68%
    event_method                         :     0.000021s :  0.22%
    auto_monad                           :     0.000085s :  0.88%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000046s :  0.48%
    rewriter_before_opt_a                :     0.000088s :  0.91%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000114s :  1.18%
    jit_opt_a.loop_unroll                :     0.000039s :  0.41%
    jit_opt_a.a_1                        :     0.000757s :  7.84%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000547s :  5.67%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.19%
    jit_opt_a.cse                        :     0.000037s :  0.38%
    jit_opt_a.replace_applicator         :     0.000016s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.10%
    rewriter_after_opt_a                 :     0.000031s :  0.32%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000471s :  4.88%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.17%
    cconv                                :     0.000023s :  0.24%
    loop_unroll                          :     0.000404s :  4.18%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000047s :  0.49%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000014s :  0.15%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000447s :  4.63%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000031s :  0.32%


Time group info:
------[substitution.]   0.000250    26
 0.90% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 1.86% :     0.000005s :      3: substitution.graph_param_transform
84.45% :     0.000211s :      6: substitution.inline
 1.28% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.80% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.34% :     0.000003s :      2: substitution.replace_old_param
 2.83% :     0.000007s :      1: substitution.switch_simplify
 5.01% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005699     2
82.25% :     0.004687s :      1: type_inference.infer
17.75% :     0.001012s :      1: type_inference.specialize
------[replace.]   0.000129     9
50.13% :     0.000065s :      6: replace.inline
26.31% :     0.000034s :      1: replace.switch_simplify
23.56% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000225     9
92.14% :     0.000207s :      6: match.inline
 2.79% :     0.000006s :      1: match.switch_simplify
 5.07% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000140   801
 1.28% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.09% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     12: predicate.addn_check_dump
 1.35% :     0.000002s :     12: predicate.addn_zero_filter
 9.57% :     0.000013s :     12: predicate.arithmetic_simplify
 1.19% :     0.000002s :     12: predicate.cast_eliminate
 0.36% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     12: predicate.compare_switch_simplify
 1.37% :     0.000002s :     12: predicate.depend_value_elim
 1.07% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.13% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.15% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.76% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.17% :     0.000002s :     12: predicate.environ_get_eliminate
 1.11% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.80% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.16% :     0.000007s :     26: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.84% :     0.000001s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     17: predicate.load_eliminater
 1.11% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.92% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     12: predicate.merge_addn
 1.06% :     0.000001s :     12: predicate.minmaximum_grad
 1.31% :     0.000002s :      3: predicate.mutable_eliminate
 0.37% :     0.000001s :      3: predicate.opt_reshape
 2.07% :     0.000003s :     17: predicate.partial_eliminate
 1.16% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     12: predicate.reduce_eliminate
 1.36% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.40% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.44% :     0.000002s :     20: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     12: predicate.reshape_eliminate
 1.21% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.59% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     12: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000004s :     20: predicate.switch_defer_inline
 2.14% :     0.000003s :     20: predicate.switch_layer_defer_inline
 7.78% :     0.000011s :     65: predicate.switch_simplify
 1.14% :     0.000002s :     12: predicate.tile_eliminate
 1.16% :     0.000002s :     12: predicate.transpose_eliminate
 1.36% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.62% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.39% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.54% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     12: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000706    15
49.67% :     0.000351s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.28% :     0.000080s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.04% :     0.000276s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014910    72
 0.34% :     0.000050s :      1: add_recomputation
 0.59% :     0.000088s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.80% :     0.000268s :      1: bootstrap
 0.17% :     0.000026s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000025s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
33.89% :     0.005053s :      1: jit_opt_a
 0.91% :     0.000136s :      1: jit_opt_after_cconv
 0.41% :     0.000062s :      1: jit_opt_b
 2.76% :     0.000411s :      1: loop_unroll
 3.22% :     0.000480s :      1: mutable_eliminate
 6.84% :     0.001019s :     26: opt.transform.jit_opt_a
 0.30% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.19% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.05% :     0.000454s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.32% :     0.000048s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000015s :      1: remove_dup_value
 1.98% :     0.000296s :      1: renormalize.infer
 1.64% :     0.000244s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000034s :      1: rewriter_after_opt_a
 0.61% :     0.000091s :      1: rewriter_before_opt_a
 0.49% :     0.000073s :      1: symbol_engine_optimizer
38.73% :     0.005775s :      1: type_inference


TotalTime = 0.0137072, [30]
[bootstrap]: 0.00026329
[type_inference]: 0.00597894
[event_method]: 2.064e-05
[auto_monad]: 7.99e-05
[graph_reusing]: 6.06998e-06
[pre_auto_parallel]: 2.93e-06
[py_interpret_to_execute]: 4.414e-05
[rewriter_before_opt_a]: 8.487e-05
[expand_dump_flag]: 2.99001e-06
[jit_opt_a]: 0.00518307, [2]
    [Cycle 1]: 0.00173558, [27]
        [switch_simplify]: 0.0001151
        [loop_unroll]: 3.552e-05
        [a_1]: 0.00067927
        [with_stream_mark]: 1.687e-05
        [recompute_prepare]: 7.31999e-06
        [updatestate_depend_eliminate]: 3.56001e-06
        [updatestate_assign_eliminate]: 3.24001e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 1.81998e-06
        [specialize_transform]: 5.97999e-06
        [updatestate_useless_node_eliminater]: 4.92999e-06
        [accelerated_algorithm]: 5.15001e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 5.15001e-06
        [merge_forward]: 3.5e-06
        [cell_reuse_recompute_pass]: 1.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.267e-05
        [j_node_and_user_rematch]: 9.44998e-06
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 8.69e-06
        [inline_without_move]: 5.57001e-06
        [renormalize]: 0.0005894
        [add_forward_monad_depend]: 4.52e-06
        [auto_monad_grad]: 1.32999e-06
        [auto_monad_eliminator]: 9.94999e-06
        [cse]: 1.798e-05
        [replace_applicator]: 1.233e-05
    [Cycle 2]: 0.00030785, [27]
        [switch_simplify]: 5.56e-06
        [loop_unroll]: 4.68999e-06
        [a_1]: 8.876e-05
        [with_stream_mark]: 8.31002e-06
        [recompute_prepare]: 4.72e-06
        [updatestate_depend_eliminate]: 2.64999e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.19998e-06
        [specialize_transform]: 4.82998e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 4.67e-06
        [meta_shard_fg_expand]: 1.26997e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.51e-06
        [cell_reuse_recompute_pass]: 1.24998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.243e-05
        [j_node_and_user_rematch]: 7.66999e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 6.69999e-06
        [inline_without_move]: 4.79e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.04e-06
        [auto_monad_grad]: 9.89996e-07
        [auto_monad_eliminator]: 4.65999e-06
        [cse]: 8.70001e-06
        [replace_applicator]: 4.92e-06
[py_interpret_to_execute_after_opt_a]: 9.28002e-06
[rewriter_after_opt_a]: 3.528e-05
[convert_after_rewriter]: 7.09001e-06
[order_py_execute_after_rewriter]: 4.75001e-06
[mutable_eliminate]: 0.00052903
[jit_opt_b]: 4.76e-05, [1]
    [Cycle 1]: 4.035e-05, [2]
        [frontend_op_eliminate]: 1.527e-05
        [inline_after_opt_a]: 1.405e-05
[cconv]: 2.487e-05
[loop_unroll]: 0.00042757
[jit_opt_after_cconv]: 0.00013408, [1]
    [Cycle 1]: 0.0001278, [11]
        [c_1]: 2.12e-05
        [parameter_eliminate]: 2.32001e-06
        [updatestate_depend_eliminate]: 4.99003e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.621e-05
        [call_graph_tuple_transform]: 1.722e-05
        [tuple_list_get_item_eliminator]: 5.23002e-06
        [none_parameter_eliminate]: 1.76998e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 5.33002e-06
[remove_dup_value]: 1.372e-05
[partial_unused_args_eliminate]: 2.23002e-06
[environ_conv]: 4.70001e-06
[add_recomputation]: 4.451e-05
[cse_after_recomputation]: 2.002e-05, [1]
    [Cycle 1]: 1.448e-05, [1]
        [cse]: 8.83001e-06
[auto_monad_reorder]: 1.429e-05
[get_jit_bprop_graph]: 1.77001e-06
[rewriter_after_jit_bprop_graph]: 4.56002e-06
[opt_after_jit_grad]: 0.00043661
[symbol_engine_optimizer]: 7.152e-05, [1]
    [Cycle 1]: 6.515e-05, [6]
        [build]: 3.43e-06
        [elim_shapecalc]: 7.35998e-06
        [elim_not_effective]: 1.268e-05
        [opt_reshape]: 5.49998e-06
        [fold_const_symbol]: 8.23001e-06
        [renormalize]: 4.80009e-07
[validate]: 3.244e-05
Sums
    bootstrap                            :     0.000263s :  2.63%
    type_inference                       :     0.005979s : 59.81%
    event_method                         :     0.000021s :  0.21%
    auto_monad                           :     0.000080s :  0.80%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000044s :  0.44%
    rewriter_before_opt_a                :     0.000085s :  0.85%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000121s :  1.21%
    jit_opt_a.loop_unroll                :     0.000040s :  0.40%
    jit_opt_a.a_1                        :     0.000768s :  7.68%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000589s :  5.90%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.15%
    jit_opt_a.cse                        :     0.000027s :  0.27%
    jit_opt_a.replace_applicator         :     0.000017s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.09%
    rewriter_after_opt_a                 :     0.000035s :  0.35%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000529s :  5.29%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.14%
    cconv                                :     0.000025s :  0.25%
    loop_unroll                          :     0.000428s :  4.28%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000045s :  0.45%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000014s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000437s :  4.37%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.32%


Time group info:
------[substitution.]   0.000258    26
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 1.81% :     0.000005s :      3: substitution.graph_param_transform
84.46% :     0.000218s :      6: substitution.inline
 1.13% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.90% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.33% :     0.000003s :      2: substitution.replace_old_param
 2.97% :     0.000008s :      1: substitution.switch_simplify
 5.05% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005917     2
83.66% :     0.004950s :      1: type_inference.infer
16.34% :     0.000967s :      1: type_inference.specialize
------[replace.]   0.000130     9
50.32% :     0.000066s :      6: replace.inline
26.86% :     0.000035s :      1: replace.switch_simplify
22.82% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000233     9
91.96% :     0.000214s :      6: match.inline
 2.94% :     0.000007s :      1: match.switch_simplify
 5.10% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   801
 1.32% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.12% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.23% :     0.000002s :     12: predicate.addn_check_dump
 1.37% :     0.000002s :     12: predicate.addn_zero_filter
 1.87% :     0.000002s :     12: predicate.arithmetic_simplify
 1.37% :     0.000002s :     12: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.23% :     0.000002s :     12: predicate.compare_switch_simplify
 1.40% :     0.000002s :     12: predicate.depend_value_elim
 1.18% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.68% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     12: predicate.environ_get_eliminate
 1.13% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.63% :     0.000007s :     26: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.90% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     17: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.47% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.32% :     0.000002s :     12: predicate.merge_addn
 1.10% :     0.000001s :     12: predicate.minmaximum_grad
 1.45% :     0.000002s :      3: predicate.mutable_eliminate
 0.39% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     17: predicate.partial_eliminate
 1.42% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     12: predicate.reduce_eliminate
 1.59% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     20: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     12: predicate.reshape_eliminate
 1.44% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.56% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.77% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.70% :     0.000004s :     20: predicate.switch_defer_inline
 2.49% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.46% :     0.000011s :     65: predicate.switch_simplify
 1.21% :     0.000002s :     12: predicate.tile_eliminate
 1.33% :     0.000002s :     12: predicate.transpose_eliminate
 1.46% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.02% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     12: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000696    15
48.16% :     0.000335s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.96% :     0.000083s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.88% :     0.000277s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015380    72
 0.31% :     0.000048s :      1: add_recomputation
 0.54% :     0.000084s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.83% :     0.000281s :      1: bootstrap
 0.18% :     0.000028s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.16% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
33.72% :     0.005186s :      1: jit_opt_a
 0.89% :     0.000137s :      1: jit_opt_after_cconv
 0.33% :     0.000050s :      1: jit_opt_b
 2.83% :     0.000435s :      1: loop_unroll
 3.49% :     0.000537s :      1: mutable_eliminate
 6.75% :     0.001039s :     26: opt.transform.jit_opt_a
 0.30% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.89% :     0.000444s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000047s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.11% :     0.000325s :      1: renormalize.infer
 1.67% :     0.000258s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000038s :      1: rewriter_after_opt_a
 0.57% :     0.000088s :      1: rewriter_before_opt_a
 0.48% :     0.000074s :      1: symbol_engine_optimizer
38.96% :     0.005993s :      1: type_inference


TotalTime = 0.020305, [33]
[bootstrap]: 0.00032377
[type_inference]: 0.00493672
[event_method]: 1.807e-05
[auto_monad]: 6.722e-05
[graph_reusing]: 6.38e-06
[pre_auto_parallel]: 2.46998e-06
[py_interpret_to_execute]: 3.649e-05
[rewriter_before_opt_a]: 7.493e-05
[expand_dump_flag]: 3.34001e-06
[jit_opt_a]: 0.00560008, [2]
    [Cycle 1]: 0.00174691, [27]
        [switch_simplify]: 5.609e-05
        [loop_unroll]: 3.225e-05
        [a_1]: 0.00054761
        [with_stream_mark]: 1.831e-05
        [recompute_prepare]: 7.03e-06
        [updatestate_depend_eliminate]: 4e-06
        [updatestate_assign_eliminate]: 3.45e-06
        [updatestate_loads_eliminate]: 2.78e-06
        [parameter_eliminate]: 2.24001e-06
        [specialize_transform]: 6.04999e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 6.09999e-06
        [meta_shard_fg_expand]: 2.29001e-06
        [get_grad_eliminate_]: 5.88998e-06
        [merge_forward]: 3.78001e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.528e-05
        [j_node_and_user_rematch]: 1.011e-05
        [meta_fg_expand]: 2.74999e-06
        [replace_old_param]: 9.70002e-06
        [inline_without_move]: 6.07999e-06
        [renormalize]: 0.00076363
        [add_forward_monad_depend]: 5.42001e-06
        [auto_monad_grad]: 2.73e-06
        [auto_monad_eliminator]: 1.45e-05
        [cse]: 2.936e-05
        [replace_applicator]: 1.164e-05
    [Cycle 2]: 0.0003312, [27]
        [switch_simplify]: 5.77999e-06
        [loop_unroll]: 5.22999e-06
        [a_1]: 9.328e-05
        [with_stream_mark]: 9.41998e-06
        [recompute_prepare]: 5.34998e-06
        [updatestate_depend_eliminate]: 2.80997e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 5.14e-06
        [updatestate_useless_node_eliminater]: 5.09998e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.95001e-06
        [merge_forward]: 2.31e-06
        [cell_reuse_recompute_pass]: 1.67999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.24e-05
        [j_node_and_user_rematch]: 8.17e-06
        [meta_fg_expand]: 1.84998e-06
        [replace_old_param]: 8.18001e-06
        [inline_without_move]: 4.67e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 4.99e-06
        [cse]: 1.107e-05
        [replace_applicator]: 5.61003e-06
[py_interpret_to_execute_after_opt_a]: 1.098e-05
[rewriter_after_opt_a]: 3.588e-05
[convert_after_rewriter]: 7.17002e-06
[order_py_execute_after_rewriter]: 5.93002e-06
[mutable_eliminate]: 0.00053577
[jit_opt_b]: 5.121e-05, [1]
    [Cycle 1]: 4.431e-05, [2]
        [frontend_op_eliminate]: 1.767e-05
        [inline_after_opt_a]: 1.504e-05
[cconv]: 2.633e-05
[loop_unroll]: 0.00042952
[jit_opt_after_cconv]: 0.00014602, [1]
    [Cycle 1]: 0.0001393, [11]
        [c_1]: 2.233e-05
        [parameter_eliminate]: 2.62001e-06
        [updatestate_depend_eliminate]: 6.14001e-06
        [updatestate_assign_eliminate]: 2.80002e-06
        [updatestate_loads_eliminate]: 2.21998e-06
        [cse]: 1.938e-05
        [call_graph_tuple_transform]: 1.935e-05
        [tuple_list_get_item_eliminator]: 6.01e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.72001e-06
[remove_dup_value]: 1.448e-05
[partial_unused_args_eliminate]: 2.30002e-06
[environ_conv]: 5.52001e-06
[add_recomputation]: 5.222e-05
[cse_after_recomputation]: 2.158e-05, [1]
    [Cycle 1]: 1.56e-05, [1]
        [cse]: 9.86e-06
[auto_monad_reorder]: 1.618e-05
[get_jit_bprop_graph]: 1.92001e-06
[rewriter_after_jit_bprop_graph]: 5.54e-06
[opt_after_jit_grad]: 0.00047407
[symbol_engine_optimizer]: 7.517e-05, [1]
    [Cycle 1]: 6.795e-05, [6]
        [build]: 3.83999e-06
        [elim_shapecalc]: 7.98001e-06
        [elim_not_effective]: 1.335e-05
        [opt_reshape]: 5.69e-06
        [fold_const_symbol]: 9.22999e-06
        [renormalize]: 4.30009e-07
[validate]: 3.888e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00705021
[execute]: 9.27999e-06
Sums
    bootstrap                            :     0.000324s :  2.00%
    type_inference                       :     0.004937s : 30.56%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000067s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.23%
    rewriter_before_opt_a                :     0.000075s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.38%
    jit_opt_a.loop_unroll                :     0.000037s :  0.23%
    jit_opt_a.a_1                        :     0.000641s :  3.97%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000764s :  4.73%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000040s :  0.25%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000536s :  3.32%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000026s :  0.16%
    loop_unroll                          :     0.000430s :  2.66%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000052s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000474s :  2.94%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000039s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007050s : 43.65%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000209    23
 1.07% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.65% :     0.000006s :      3: substitution.graph_param_transform
83.24% :     0.000174s :      4: substitution.inline
 1.67% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.22% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000004s :      2: substitution.replace_old_param
 6.43% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004868     2
81.54% :     0.003969s :      1: type_inference.infer
18.46% :     0.000898s :      1: type_inference.specialize
------[replace.]   0.000069     6
66.13% :     0.000046s :      4: replace.inline
33.87% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000184     6
93.30% :     0.000172s :      4: match.inline
 6.70% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.32% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 1.93% :     0.000002s :     10: predicate.arithmetic_simplify
 1.32% :     0.000002s :     10: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.54% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.33% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.88% :     0.000007s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.26% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.92% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.40% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.77% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.32% :     0.000003s :     16: predicate.switch_defer_inline
 1.93% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.70% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.28% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.14% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000654    12
49.30% :     0.000322s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.76% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.94% :     0.000294s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022027    76
 0.25% :     0.000055s :      1: add_recomputation
 0.32% :     0.000071s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.56% :     0.000344s :      1: bootstrap
 0.13% :     0.000030s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.44% :     0.005603s :      1: jit_opt_a
 0.68% :     0.000149s :      1: jit_opt_after_cconv
 0.24% :     0.000054s :      1: jit_opt_b
 1.99% :     0.000438s :      1: loop_unroll
 2.47% :     0.000545s :      1: mutable_eliminate
 3.90% :     0.000858s :     26: opt.transform.jit_opt_a
 0.23% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.19% :     0.000482s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000039s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.93% :     0.000424s :      1: renormalize.infer
 1.51% :     0.000332s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000039s :      1: rewriter_after_opt_a
 0.35% :     0.000077s :      1: rewriter_before_opt_a
 0.35% :     0.000078s :      1: symbol_engine_optimizer
32.09% :     0.007068s :      1: task_emit
22.47% :     0.004949s :      1: type_inference
 0.27% :     0.000059s :      1: validate


TotalTime = 0.0257156, [33]
[bootstrap]: 0.00044901
[type_inference]: 0.00630526
[event_method]: 2.357e-05
[auto_monad]: 7.326e-05
[graph_reusing]: 6.15002e-06
[pre_auto_parallel]: 2.83e-06
[py_interpret_to_execute]: 5.144e-05
[rewriter_before_opt_a]: 8.041e-05
[expand_dump_flag]: 3.88001e-06
[jit_opt_a]: 0.00794635, [2]
    [Cycle 1]: 0.00277135, [27]
        [switch_simplify]: 7.507e-05
        [loop_unroll]: 3.078e-05
        [a_1]: 0.00085489
        [with_stream_mark]: 3.045e-05
        [recompute_prepare]: 1.139e-05
        [updatestate_depend_eliminate]: 4.97e-06
        [updatestate_assign_eliminate]: 3.13e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [parameter_eliminate]: 2.21e-06
        [specialize_transform]: 7.15998e-06
        [updatestate_useless_node_eliminater]: 6.23998e-06
        [accelerated_algorithm]: 7.22002e-06
        [meta_shard_fg_expand]: 3.61999e-06
        [get_grad_eliminate_]: 6.63e-06
        [merge_forward]: 4.57e-06
        [cell_reuse_recompute_pass]: 2.76999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.763e-05
        [j_node_and_user_rematch]: 1.192e-05
        [meta_fg_expand]: 2.53998e-06
        [replace_old_param]: 1.186e-05
        [inline_without_move]: 5.78997e-06
        [renormalize]: 0.00133472
        [add_forward_monad_depend]: 9.56998e-06
        [auto_monad_grad]: 2.54999e-06
        [auto_monad_eliminator]: 1.959e-05
        [cse]: 4.023e-05
        [replace_applicator]: 2.153e-05
    [Cycle 2]: 0.00039247, [27]
        [switch_simplify]: 6.89999e-06
        [loop_unroll]: 5.64998e-06
        [a_1]: 0.00011047
        [with_stream_mark]: 1.564e-05
        [recompute_prepare]: 5.76e-06
        [updatestate_depend_eliminate]: 3.95998e-06
        [updatestate_assign_eliminate]: 2.47001e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 5.62001e-06
        [updatestate_useless_node_eliminater]: 5.12999e-06
        [accelerated_algorithm]: 6.14001e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 4.63999e-06
        [merge_forward]: 3.12002e-06
        [cell_reuse_recompute_pass]: 2.89999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.665e-05
        [j_node_and_user_rematch]: 8.68001e-06
        [meta_fg_expand]: 1.64e-06
        [replace_old_param]: 9.37999e-06
        [inline_without_move]: 5.39e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 2.36e-06
        [auto_monad_grad]: 1.60999e-06
        [auto_monad_eliminator]: 7.77002e-06
        [cse]: 1.355e-05
        [replace_applicator]: 7.01999e-06
[py_interpret_to_execute_after_opt_a]: 1.628e-05
[rewriter_after_opt_a]: 4.618e-05
[convert_after_rewriter]: 8.38001e-06
[order_py_execute_after_rewriter]: 6.12001e-06
[mutable_eliminate]: 0.00098786
[jit_opt_b]: 5.942e-05, [1]
    [Cycle 1]: 4.908e-05, [2]
        [frontend_op_eliminate]: 1.998e-05
        [inline_after_opt_a]: 1.625e-05
[cconv]: 3.893e-05
[loop_unroll]: 0.00058828
[jit_opt_after_cconv]: 0.00018827, [1]
    [Cycle 1]: 0.00017953, [11]
        [c_1]: 2.405e-05
        [parameter_eliminate]: 5.54998e-06
        [updatestate_depend_eliminate]: 9.79999e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.35002e-06
        [cse]: 3.434e-05
        [call_graph_tuple_transform]: 2.826e-05
        [tuple_list_get_item_eliminator]: 5.68002e-06
        [none_parameter_eliminate]: 1.82999e-06
        [renormalize]: 8.29983e-07
        [switch_simplify]: 6.44999e-06
[remove_dup_value]: 1.745e-05
[partial_unused_args_eliminate]: 2.79999e-06
[environ_conv]: 6.23002e-06
[add_recomputation]: 6.019e-05
[cse_after_recomputation]: 2.378e-05, [1]
    [Cycle 1]: 1.754e-05, [1]
        [cse]: 1.058e-05
[auto_monad_reorder]: 1.677e-05
[get_jit_bprop_graph]: 2.49001e-06
[rewriter_after_jit_bprop_graph]: 1.105e-05
[opt_after_jit_grad]: 0.00062153
[symbol_engine_optimizer]: 8.865e-05, [1]
    [Cycle 1]: 7.986e-05, [6]
        [build]: 6.30002e-06
        [elim_shapecalc]: 8.87e-06
        [elim_not_effective]: 1.695e-05
        [opt_reshape]: 6.71e-06
        [fold_const_symbol]: 1.004e-05
        [renormalize]: 4.09986e-07
[validate]: 5.398e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.007613
[execute]: 1.197e-05
Sums
    bootstrap                            :     0.000449s :  2.23%
    type_inference                       :     0.006305s : 31.35%
    event_method                         :     0.000024s :  0.12%
    auto_monad                           :     0.000073s :  0.36%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000051s :  0.26%
    rewriter_before_opt_a                :     0.000080s :  0.40%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000082s :  0.41%
    jit_opt_a.loop_unroll                :     0.000036s :  0.18%
    jit_opt_a.a_1                        :     0.000965s :  4.80%
    jit_opt_a.with_stream_mark           :     0.000046s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001335s :  6.64%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.14%
    jit_opt_a.cse                        :     0.000054s :  0.27%
    jit_opt_a.replace_applicator         :     0.000029s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.08%
    rewriter_after_opt_a                 :     0.000046s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000988s :  4.91%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.000039s :  0.19%
    loop_unroll                          :     0.000588s :  2.92%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000034s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000060s :  0.30%
    cse_after_recomputation.cse          :     0.000011s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.05%
    opt_after_jit_grad                   :     0.000622s :  3.09%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000054s :  0.27%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007613s : 37.85%
    execute                              :     0.000012s :  0.06%


Time group info:
------[substitution.]   0.000448    23
 0.51% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000002s :      2: substitution.fold_const_symbol
 1.45% :     0.000006s :      3: substitution.graph_param_transform
89.96% :     0.000403s :      4: substitution.inline
 1.05% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.29% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.48% :     0.000007s :      2: substitution.replace_old_param
 3.81% :     0.000017s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006209     2
82.37% :     0.005114s :      1: type_inference.infer
17.63% :     0.001095s :      1: type_inference.specialize
------[replace.]   0.000103     6
63.71% :     0.000066s :      4: replace.inline
36.29% :     0.000038s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000414     6
96.26% :     0.000399s :      4: match.inline
 3.74% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000152   691
 1.64% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.49% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.15% :     0.000002s :     10: predicate.addn_zero_filter
 2.20% :     0.000003s :     10: predicate.arithmetic_simplify
 1.49% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000002s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 0.89% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.24% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000001s :      3: predicate.elim_not_effective
 0.75% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.04% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.84% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000002s :     10: predicate.environ_get_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.54% :     0.000008s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 2.08% :     0.000003s :      6: predicate.less_batch_normalization
 1.66% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.52% :     0.000002s :     15: predicate.load_eliminater
 2.34% :     0.000004s :      3: predicate.loop_unroll_after_grad
 3.37% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.41% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 0.92% :     0.000001s :     10: predicate.merge_addn
 0.86% :     0.000001s :     10: predicate.minmaximum_grad
 3.10% :     0.000005s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 1.79% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.37% :     0.000002s :     10: predicate.reduce_eliminate
 1.59% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.73% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000003s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.45% :     0.000001s :      3: predicate.reset_defer_inline
 1.19% :     0.000002s :     10: predicate.reshape_eliminate
 1.41% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.79% :     0.000003s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000002s :      6: predicate.special_op_eliminate
 1.11% :     0.000002s :      6: predicate.specialize_transform
 1.18% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.08% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.30% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.51% :     0.000004s :     16: predicate.switch_defer_inline
 1.83% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.83% :     0.000010s :     53: predicate.switch_simplify
 1.25% :     0.000002s :     10: predicate.tile_eliminate
 0.98% :     0.000001s :     10: predicate.transpose_eliminate
 1.17% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.93% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.91% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 2.49% :     0.000004s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.43% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.50% :     0.000005s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000003s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000880    12
45.53% :     0.000401s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.27% :     0.000064s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.21% :     0.000416s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028408    76
 0.23% :     0.000064s :      1: add_recomputation
 0.27% :     0.000078s :      1: auto_monad
 0.07% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.67% :     0.000475s :      1: bootstrap
 0.15% :     0.000042s :      1: cconv
 0.05% :     0.000015s :      1: convert_after_rewriter
 0.09% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.10% :     0.000028s :      1: event_method
 0.06% :     0.000018s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
27.99% :     0.007951s :      1: jit_opt_a
 0.67% :     0.000191s :      1: jit_opt_after_cconv
 0.22% :     0.000062s :      1: jit_opt_b
 2.11% :     0.000599s :      1: loop_unroll
 3.54% :     0.001007s :      1: mutable_eliminate
 4.33% :     0.001230s :     26: opt.transform.jit_opt_a
 0.21% :     0.000061s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000030s :      4: opt.transform.jit_opt_b
 0.06% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000025s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000039s :      4: opt.transform.symbol_engine_opt
 2.24% :     0.000636s :      1: opt_after_jit_grad
 0.03% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000055s :      1: py_interpret_to_execute
 0.07% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000020s :      1: remove_dup_value
 2.76% :     0.000783s :      1: renormalize.infer
 1.89% :     0.000538s :      1: renormalize.specialize
 0.05% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000053s :      1: rewriter_after_opt_a
 0.30% :     0.000084s :      1: rewriter_before_opt_a
 0.32% :     0.000091s :      1: symbol_engine_optimizer
26.90% :     0.007641s :      1: task_emit
22.29% :     0.006331s :      1: type_inference
 0.29% :     0.000082s :      1: validate


TotalTime = 0.0142434, [30]
[bootstrap]: 0.00026978
[type_inference]: 0.00634427
[event_method]: 2.211e-05
[auto_monad]: 8.194e-05
[graph_reusing]: 7.16001e-06
[pre_auto_parallel]: 3.43e-06
[py_interpret_to_execute]: 4.534e-05
[rewriter_before_opt_a]: 8.738e-05
[expand_dump_flag]: 2.98e-06
[jit_opt_a]: 0.0052762, [2]
    [Cycle 1]: 0.00181524, [27]
        [switch_simplify]: 0.00011666
        [loop_unroll]: 3.609e-05
        [a_1]: 0.00067878
        [with_stream_mark]: 1.604e-05
        [recompute_prepare]: 7.58001e-06
        [updatestate_depend_eliminate]: 3.7e-06
        [updatestate_assign_eliminate]: 2.93998e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [parameter_eliminate]: 2.13998e-06
        [specialize_transform]: 5.87999e-06
        [updatestate_useless_node_eliminater]: 5.16998e-06
        [accelerated_algorithm]: 5.61e-06
        [meta_shard_fg_expand]: 2.18998e-06
        [get_grad_eliminate_]: 5.19998e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.427e-05
        [j_node_and_user_rematch]: 9.17001e-06
        [meta_fg_expand]: 2.29999e-06
        [replace_old_param]: 9.04e-06
        [inline_without_move]: 5.42999e-06
        [renormalize]: 0.0006438
        [add_forward_monad_depend]: 4.84e-06
        [auto_monad_grad]: 2.49999e-06
        [auto_monad_eliminator]: 1.303e-05
        [cse]: 3.006e-05
        [replace_applicator]: 1.217e-05
    [Cycle 2]: 0.00031414, [27]
        [switch_simplify]: 5.56998e-06
        [loop_unroll]: 4.84003e-06
        [a_1]: 8.822e-05
        [with_stream_mark]: 9.29998e-06
        [recompute_prepare]: 4.79e-06
        [updatestate_depend_eliminate]: 2.57001e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 4.84998e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.33002e-06
        [get_grad_eliminate_]: 4.58999e-06
        [merge_forward]: 2.79999e-06
        [cell_reuse_recompute_pass]: 1.71002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.215e-05
        [j_node_and_user_rematch]: 7.83999e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 7.01999e-06
        [inline_without_move]: 4.81997e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.03001e-06
        [auto_monad_grad]: 1.19e-06
        [auto_monad_eliminator]: 5.39e-06
        [cse]: 1.065e-05
        [replace_applicator]: 4.97999e-06
[py_interpret_to_execute_after_opt_a]: 1.034e-05
[rewriter_after_opt_a]: 3.288e-05
[convert_after_rewriter]: 7.08998e-06
[order_py_execute_after_rewriter]: 5.22e-06
[mutable_eliminate]: 0.00053842
[jit_opt_b]: 5.859e-05, [1]
    [Cycle 1]: 4.274e-05, [2]
        [frontend_op_eliminate]: 1.636e-05
        [inline_after_opt_a]: 1.468e-05
[cconv]: 2.44e-05
[loop_unroll]: 0.00042336
[jit_opt_after_cconv]: 0.00014006, [1]
    [Cycle 1]: 0.00013335, [11]
        [c_1]: 2.189e-05
        [parameter_eliminate]: 2.84001e-06
        [updatestate_depend_eliminate]: 5.70001e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [cse]: 1.804e-05
        [call_graph_tuple_transform]: 1.786e-05
        [tuple_list_get_item_eliminator]: 5.19e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 7.39994e-07
        [switch_simplify]: 5.46e-06
[remove_dup_value]: 1.439e-05
[partial_unused_args_eliminate]: 2.12001e-06
[environ_conv]: 4.99998e-06
[add_recomputation]: 4.988e-05
[cse_after_recomputation]: 2.162e-05, [1]
    [Cycle 1]: 1.63e-05, [1]
        [cse]: 1.018e-05
[auto_monad_reorder]: 1.488e-05
[get_jit_bprop_graph]: 1.84998e-06
[rewriter_after_jit_bprop_graph]: 5.71998e-06
[opt_after_jit_grad]: 0.00046073
[symbol_engine_optimizer]: 7.487e-05, [1]
    [Cycle 1]: 6.849e-05, [6]
        [build]: 4.07e-06
        [elim_shapecalc]: 8e-06
        [elim_not_effective]: 1.33e-05
        [opt_reshape]: 5.57001e-06
        [fold_const_symbol]: 8.79998e-06
        [renormalize]: 3.50003e-07
[validate]: 3.34e-05
Sums
    bootstrap                            :     0.000270s :  2.57%
    type_inference                       :     0.006344s : 60.40%
    event_method                         :     0.000022s :  0.21%
    auto_monad                           :     0.000082s :  0.78%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000045s :  0.43%
    rewriter_before_opt_a                :     0.000087s :  0.83%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000122s :  1.16%
    jit_opt_a.loop_unroll                :     0.000041s :  0.39%
    jit_opt_a.a_1                        :     0.000767s :  7.30%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000644s :  6.13%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.18%
    jit_opt_a.cse                        :     0.000041s :  0.39%
    jit_opt_a.replace_applicator         :     0.000017s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.10%
    rewriter_after_opt_a                 :     0.000033s :  0.31%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000538s :  5.13%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.14%
    cconv                                :     0.000024s :  0.23%
    loop_unroll                          :     0.000423s :  4.03%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000050s :  0.47%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000461s :  4.39%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.32%


Time group info:
------[substitution.]   0.000261    26
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 1.90% :     0.000005s :      3: substitution.graph_param_transform
84.29% :     0.000220s :      6: substitution.inline
 1.18% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.67% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.49% :     0.000004s :      2: substitution.replace_old_param
 2.95% :     0.000008s :      1: substitution.switch_simplify
 5.14% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006279     2
83.63% :     0.005251s :      1: type_inference.infer
16.37% :     0.001028s :      1: type_inference.specialize
------[replace.]   0.000132     9
50.48% :     0.000067s :      6: replace.inline
25.51% :     0.000034s :      1: replace.switch_simplify
24.01% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000236     9
91.85% :     0.000216s :      6: match.inline
 2.94% :     0.000007s :      1: match.switch_simplify
 5.21% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   801
 1.28% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.00% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     12: predicate.addn_check_dump
 1.59% :     0.000002s :     12: predicate.addn_zero_filter
 1.91% :     0.000003s :     12: predicate.arithmetic_simplify
 1.29% :     0.000002s :     12: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     12: predicate.compare_switch_simplify
 1.43% :     0.000002s :     12: predicate.depend_value_elim
 1.15% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.51% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.38% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     12: predicate.environ_get_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.51% :     0.000007s :     26: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     17: predicate.load_eliminater
 1.30% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.57% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.03% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     12: predicate.merge_addn
 1.20% :     0.000002s :     12: predicate.minmaximum_grad
 1.44% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     17: predicate.partial_eliminate
 1.18% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     12: predicate.reduce_eliminate
 1.56% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     20: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     12: predicate.reshape_eliminate
 1.29% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 1.01% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     12: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.93% :     0.000004s :     20: predicate.switch_defer_inline
 2.34% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.71% :     0.000012s :     65: predicate.switch_simplify
 1.22% :     0.000002s :     12: predicate.tile_eliminate
 1.18% :     0.000002s :     12: predicate.transpose_eliminate
 1.55% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.59% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     12: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000763    15
50.11% :     0.000382s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.04% :     0.000084s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.86% :     0.000296s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015990    72
 0.33% :     0.000053s :      1: add_recomputation
 0.53% :     0.000085s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 1.77% :     0.000283s :      1: bootstrap
 0.17% :     0.000027s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000024s :      1: cse_after_recomputation
 0.12% :     0.000019s :      1: environ_conv
 0.16% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000010s :      1: graph_reusing
33.02% :     0.005279s :      1: jit_opt_a
 0.89% :     0.000143s :      1: jit_opt_after_cconv
 0.38% :     0.000061s :      1: jit_opt_b
 2.70% :     0.000431s :      1: loop_unroll
 3.42% :     0.000547s :      1: mutable_eliminate
 6.52% :     0.001043s :     26: opt.transform.jit_opt_a
 0.29% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.93% :     0.000469s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000048s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.14% :     0.000343s :      1: renormalize.infer
 1.84% :     0.000294s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000036s :      1: rewriter_after_opt_a
 0.56% :     0.000090s :      1: rewriter_before_opt_a
 0.48% :     0.000077s :      1: symbol_engine_optimizer
39.78% :     0.006360s :      1: type_inference


TotalTime = 0.0267749, [33]
[bootstrap]: 0.00026913
[type_inference]: 0.00579659
[event_method]: 2.049e-05
[auto_monad]: 7.497e-05
[graph_reusing]: 6.49999e-06
[pre_auto_parallel]: 2.94001e-06
[py_interpret_to_execute]: 5.237e-05
[rewriter_before_opt_a]: 8.283e-05
[expand_dump_flag]: 3.5e-06
[jit_opt_a]: 0.00931101, [2]
    [Cycle 1]: 0.0019358, [27]
        [switch_simplify]: 6.904e-05
        [loop_unroll]: 3.061e-05
        [a_1]: 0.00059866
        [with_stream_mark]: 2.059e-05
        [recompute_prepare]: 7.23999e-06
        [updatestate_depend_eliminate]: 4.35e-06
        [updatestate_assign_eliminate]: 3.27002e-06
        [updatestate_loads_eliminate]: 2.64001e-06
        [parameter_eliminate]: 2.33002e-06
        [specialize_transform]: 6.05002e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 6.05002e-06
        [meta_shard_fg_expand]: 2.45002e-06
        [get_grad_eliminate_]: 5.46e-06
        [merge_forward]: 3.73999e-06
        [cell_reuse_recompute_pass]: 1.69e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.466e-05
        [j_node_and_user_rematch]: 9.23002e-06
        [meta_fg_expand]: 2.54999e-06
        [replace_old_param]: 9.66998e-06
        [inline_without_move]: 5.54e-06
        [renormalize]: 0.00087541
        [add_forward_monad_depend]: 5.64e-06
        [auto_monad_grad]: 2.26998e-06
        [auto_monad_eliminator]: 1.429e-05
        [cse]: 3.002e-05
        [replace_applicator]: 1.292e-05
    [Cycle 2]: 0.00033566, [27]
        [switch_simplify]: 5.56e-06
        [loop_unroll]: 5.22999e-06
        [a_1]: 9.286e-05
        [with_stream_mark]: 1.102e-05
        [recompute_prepare]: 5.24e-06
        [updatestate_depend_eliminate]: 2.93e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.72999e-06
        [specialize_transform]: 5.00999e-06
        [updatestate_useless_node_eliminater]: 4.78001e-06
        [accelerated_algorithm]: 5.45001e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 2.19001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.35e-05
        [j_node_and_user_rematch]: 8.26002e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 7.47998e-06
        [inline_without_move]: 4.77e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.67999e-06
        [auto_monad_grad]: 1.16997e-06
        [auto_monad_eliminator]: 5.44998e-06
        [cse]: 1.078e-05
        [replace_applicator]: 4.90001e-06
[py_interpret_to_execute_after_opt_a]: 1.336e-05
[rewriter_after_opt_a]: 3.736e-05
[convert_after_rewriter]: 7.98999e-06
[order_py_execute_after_rewriter]: 5.67001e-06
[mutable_eliminate]: 0.00065604
[jit_opt_b]: 5.451e-05, [1]
    [Cycle 1]: 4.68e-05, [2]
        [frontend_op_eliminate]: 1.874e-05
        [inline_after_opt_a]: 1.65e-05
[cconv]: 2.626e-05
[loop_unroll]: 0.00044371
[jit_opt_after_cconv]: 0.00019432, [1]
    [Cycle 1]: 0.00018784, [11]
        [c_1]: 2.166e-05
        [parameter_eliminate]: 3.08e-06
        [updatestate_depend_eliminate]: 6.26e-06
        [updatestate_assign_eliminate]: 2.57001e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 2.022e-05
        [call_graph_tuple_transform]: 6.58e-05
        [tuple_list_get_item_eliminator]: 6.18998e-06
        [none_parameter_eliminate]: 1.72001e-06
        [renormalize]: 9.30013e-07
        [switch_simplify]: 5.61e-06
[remove_dup_value]: 1.618e-05
[partial_unused_args_eliminate]: 2.81999e-06
[environ_conv]: 5.72999e-06
[add_recomputation]: 5.538e-05
[cse_after_recomputation]: 2.137e-05, [1]
    [Cycle 1]: 1.553e-05, [1]
        [cse]: 9.51e-06
[auto_monad_reorder]: 1.594e-05
[get_jit_bprop_graph]: 1.96e-06
[rewriter_after_jit_bprop_graph]: 5.50001e-06
[opt_after_jit_grad]: 0.00047537
[symbol_engine_optimizer]: 7.748e-05, [1]
    [Cycle 1]: 7.031e-05, [6]
        [build]: 4.13001e-06
        [elim_shapecalc]: 8.10999e-06
        [elim_not_effective]: 1.37e-05
        [opt_reshape]: 5.66e-06
        [fold_const_symbol]: 9.42999e-06
        [renormalize]: 7.99977e-07
[validate]: 4.067e-05
[backend_pass]: 1.23002e-06
[task_emit]: 0.00875644
[execute]: 9.05001e-06
Sums
    bootstrap                            :     0.000269s :  1.41%
    type_inference                       :     0.005797s : 30.39%
    event_method                         :     0.000020s :  0.11%
    auto_monad                           :     0.000075s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000052s :  0.27%
    rewriter_before_opt_a                :     0.000083s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000075s :  0.39%
    jit_opt_a.loop_unroll                :     0.000036s :  0.19%
    jit_opt_a.a_1                        :     0.000692s :  3.63%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000875s :  4.59%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.10%
    jit_opt_a.cse                        :     0.000041s :  0.21%
    jit_opt_a.replace_applicator         :     0.000018s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.20%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000656s :  3.44%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000026s :  0.14%
    loop_unroll                          :     0.000444s :  2.33%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000066s :  0.34%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000055s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000475s :  2.49%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008756s : 45.91%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000286    23
 0.77% :     0.000002s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
18.05% :     0.000052s :      3: substitution.graph_param_transform
71.47% :     0.000204s :      4: substitution.inline
 1.17% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.58% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.78% :     0.000005s :      2: substitution.replace_old_param
 4.74% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005702     2
62.23% :     0.003549s :      1: type_inference.infer
37.77% :     0.002153s :      1: type_inference.specialize
------[replace.]   0.000084     6
61.76% :     0.000052s :      4: replace.inline
38.24% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000214     6
94.22% :     0.000202s :      4: match.inline
 5.78% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.00% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.16% :     0.000001s :     10: predicate.addn_zero_filter
 2.13% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.58% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.39% :     0.000002s :     10: predicate.environ_get_eliminate
 1.39% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.40% :     0.000007s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.35% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.49% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.95% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.38% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.25% :     0.000002s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 2.01% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.56% :     0.000009s :     53: predicate.switch_simplify
 1.33% :     0.000002s :     10: predicate.tile_eliminate
 1.07% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.11% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.17% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.83% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000643    12
40.24% :     0.000259s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.80% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
52.96% :     0.000340s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028717    76
 0.20% :     0.000058s :      1: add_recomputation
 0.27% :     0.000078s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.99% :     0.000283s :      1: bootstrap
 0.10% :     0.000029s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.09% :     0.000025s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
32.44% :     0.009315s :      1: jit_opt_a
 0.69% :     0.000197s :      1: jit_opt_after_cconv
 0.20% :     0.000057s :      1: jit_opt_b
 1.57% :     0.000452s :      1: loop_unroll
 2.32% :     0.000665s :      1: mutable_eliminate
 3.20% :     0.000919s :     26: opt.transform.jit_opt_a
 0.33% :     0.000096s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.68% :     0.000483s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000055s :      1: py_interpret_to_execute
 0.06% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 1.62% :     0.000464s :      1: renormalize.infer
 1.40% :     0.000403s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000040s :      1: rewriter_after_opt_a
 0.30% :     0.000086s :      1: rewriter_before_opt_a
 0.28% :     0.000080s :      1: symbol_engine_optimizer
30.57% :     0.008779s :      1: task_emit
20.27% :     0.005820s :      1: type_inference
 0.21% :     0.000060s :      1: validate


TotalTime = 0.0259486, [33]
[bootstrap]: 0.00042778
[type_inference]: 0.00684481
[event_method]: 2.356e-05
[auto_monad]: 7.347e-05
[graph_reusing]: 5.67999e-06
[pre_auto_parallel]: 2.44999e-06
[py_interpret_to_execute]: 5.195e-05
[rewriter_before_opt_a]: 7.962e-05
[expand_dump_flag]: 3.26001e-06
[jit_opt_a]: 0.00783879, [2]
    [Cycle 1]: 0.00268315, [27]
        [switch_simplify]: 8.54e-05
        [loop_unroll]: 3.108e-05
        [a_1]: 0.00075919
        [with_stream_mark]: 3.12e-05
        [recompute_prepare]: 8.80999e-06
        [updatestate_depend_eliminate]: 4.32998e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 2.89999e-06
        [parameter_eliminate]: 2.39001e-06
        [specialize_transform]: 6.61e-06
        [updatestate_useless_node_eliminater]: 5.49e-06
        [accelerated_algorithm]: 6.74999e-06
        [meta_shard_fg_expand]: 2.84999e-06
        [get_grad_eliminate_]: 5.84e-06
        [merge_forward]: 3.85998e-06
        [cell_reuse_recompute_pass]: 2.32001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.778e-05
        [j_node_and_user_rematch]: 1.192e-05
        [meta_fg_expand]: 2.63998e-06
        [replace_old_param]: 1.179e-05
        [inline_without_move]: 5.69999e-06
        [renormalize]: 0.00135341
        [add_forward_monad_depend]: 1.024e-05
        [auto_monad_grad]: 2.61e-06
        [auto_monad_eliminator]: 2.19e-05
        [cse]: 3.811e-05
        [replace_applicator]: 1.983e-05
    [Cycle 2]: 0.00036569, [27]
        [switch_simplify]: 7.28e-06
        [loop_unroll]: 5.82001e-06
        [a_1]: 0.00010469
        [with_stream_mark]: 1.591e-05
        [recompute_prepare]: 5.14e-06
        [updatestate_depend_eliminate]: 2.84999e-06
        [updatestate_assign_eliminate]: 2.54001e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 1.37e-06
        [specialize_transform]: 5.15999e-06
        [updatestate_useless_node_eliminater]: 4.72e-06
        [accelerated_algorithm]: 5.90002e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 4.74e-06
        [merge_forward]: 2.49001e-06
        [cell_reuse_recompute_pass]: 3.80998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.672e-05
        [j_node_and_user_rematch]: 9.31e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 8.74998e-06
        [inline_without_move]: 4.95001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.84e-06
        [auto_monad_grad]: 6.00005e-07
        [auto_monad_eliminator]: 5.99999e-06
        [cse]: 1.18e-05
        [replace_applicator]: 4.97e-06
[py_interpret_to_execute_after_opt_a]: 1.611e-05
[rewriter_after_opt_a]: 4.401e-05
[convert_after_rewriter]: 9.14998e-06
[order_py_execute_after_rewriter]: 5.09e-06
[mutable_eliminate]: 0.00097415
[jit_opt_b]: 6.216e-05, [1]
    [Cycle 1]: 5.198e-05, [2]
        [frontend_op_eliminate]: 2.094e-05
        [inline_after_opt_a]: 1.674e-05
[cconv]: 3.788e-05
[loop_unroll]: 0.00055358
[jit_opt_after_cconv]: 0.0001964, [1]
    [Cycle 1]: 0.00018715, [11]
        [c_1]: 2.332e-05
        [parameter_eliminate]: 4.96002e-06
        [updatestate_depend_eliminate]: 1.072e-05
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.66e-06
        [cse]: 3.808e-05
        [call_graph_tuple_transform]: 3.017e-05
        [tuple_list_get_item_eliminator]: 7.26999e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 1.39e-06
        [switch_simplify]: 7.65998e-06
[remove_dup_value]: 1.722e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 7.66001e-06
[add_recomputation]: 6.303e-05
[cse_after_recomputation]: 2.571e-05, [1]
    [Cycle 1]: 1.797e-05, [1]
        [cse]: 1.112e-05
[auto_monad_reorder]: 1.631e-05
[get_jit_bprop_graph]: 3.03e-06
[rewriter_after_jit_bprop_graph]: 1.088e-05
[opt_after_jit_grad]: 0.00057544
[symbol_engine_optimizer]: 9.145e-05, [1]
    [Cycle 1]: 8.158e-05, [6]
        [build]: 6.41e-06
        [elim_shapecalc]: 9.58002e-06
        [elim_not_effective]: 1.645e-05
        [opt_reshape]: 7.32002e-06
        [fold_const_symbol]: 1.11e-05
        [renormalize]: 8.70001e-07
[validate]: 5.717e-05
[backend_pass]: 1.22e-06
[task_emit]: 0.00752465
[execute]: 1.123e-05
Sums
    bootstrap                            :     0.000428s :  2.10%
    type_inference                       :     0.006845s : 33.59%
    event_method                         :     0.000024s :  0.12%
    auto_monad                           :     0.000073s :  0.36%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000052s :  0.25%
    rewriter_before_opt_a                :     0.000080s :  0.39%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000093s :  0.45%
    jit_opt_a.loop_unroll                :     0.000037s :  0.18%
    jit_opt_a.a_1                        :     0.000864s :  4.24%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.001353s :  6.64%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.14%
    jit_opt_a.cse                        :     0.000050s :  0.24%
    jit_opt_a.replace_applicator         :     0.000025s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.08%
    rewriter_after_opt_a                 :     0.000044s :  0.22%
    convert_after_rewriter               :     0.000009s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000974s :  4.78%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.08%
    cconv                                :     0.000038s :  0.19%
    loop_unroll                          :     0.000554s :  2.72%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000038s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000030s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000008s :  0.04%
    remove_dup_value                     :     0.000017s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000008s :  0.04%
    add_recomputation                    :     0.000063s :  0.31%
    cse_after_recomputation.cse          :     0.000011s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.05%
    opt_after_jit_grad                   :     0.000575s :  2.82%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000057s :  0.28%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007525s : 36.93%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000345    23
 0.61% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000002s :      2: substitution.fold_const_symbol
 1.82% :     0.000006s :      3: substitution.graph_param_transform
86.78% :     0.000300s :      4: substitution.inline
 1.76% :     0.000006s :      4: substitution.j_node_and_user_rematch
 1.78% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.13% :     0.000007s :      2: substitution.replace_old_param
 4.41% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006746     2
81.82% :     0.005519s :      1: type_inference.infer
18.18% :     0.001226s :      1: type_inference.specialize
------[replace.]   0.000099     6
67.07% :     0.000066s :      4: replace.inline
32.93% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000311     6
95.50% :     0.000297s :      4: match.inline
 4.50% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000163   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.90% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.78% :     0.000001s :     10: predicate.addn_check_dump
 1.02% :     0.000002s :     10: predicate.addn_zero_filter
 2.32% :     0.000004s :     10: predicate.arithmetic_simplify
 0.85% :     0.000001s :     10: predicate.cast_eliminate
 0.28% :     0.000000s :      3: predicate.check_bprop_eliminate
10.11% :     0.000016s :     10: predicate.compare_switch_simplify
 1.17% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.98% :     0.000003s :     10: predicate.dict_get_item_eliminator
 1.88% :     0.000003s :     10: predicate.dict_set_item_eliminator
 1.04% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.80% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.75% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.92% :     0.000002s :     10: predicate.environ_get_eliminate
 0.95% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000002s :      6: predicate.get_grad_eliminate
 0.43% :     0.000001s :      3: predicate.graph_param_transform
 4.52% :     0.000007s :     22: predicate.inline
 0.64% :     0.000001s :      6: predicate.inline_without_move
 0.29% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.36% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.39% :     0.000002s :     15: predicate.load_eliminater
 1.67% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.34% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.93% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.76% :     0.000001s :     10: predicate.merge_addn
 0.77% :     0.000001s :     10: predicate.minmaximum_grad
 3.16% :     0.000005s :      3: predicate.mutable_eliminate
 0.36% :     0.000001s :      3: predicate.opt_reshape
 1.66% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000003s :     10: predicate.reduce_eliminate
 1.28% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.38% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.20% :     0.000000s :      3: predicate.reset_defer_inline
 1.60% :     0.000003s :     10: predicate.reshape_eliminate
 0.93% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000002s :     10: predicate.same_eliminate
 0.39% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.28% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.10% :     0.000003s :     16: predicate.switch_defer_inline
 1.74% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.69% :     0.000011s :     53: predicate.switch_simplify
 1.15% :     0.000002s :     10: predicate.tile_eliminate
 0.82% :     0.000001s :     10: predicate.transpose_eliminate
 0.98% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.69% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.83% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.14% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.41% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000003s :     10: predicate.value_based_eliminate
 0.23% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.96% :     0.000002s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000966    12
45.06% :     0.000435s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.73% :     0.000065s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.22% :     0.000466s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028562    76
 0.23% :     0.000066s :      1: add_recomputation
 0.27% :     0.000077s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.60% :     0.000458s :      1: bootstrap
 0.14% :     0.000040s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000029s :      1: cse_after_recomputation
 0.04% :     0.000012s :      1: environ_conv
 0.10% :     0.000028s :      1: event_method
 0.06% :     0.000018s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
27.46% :     0.007844s :      1: jit_opt_a
 0.70% :     0.000199s :      1: jit_opt_after_cconv
 0.23% :     0.000065s :      1: jit_opt_b
 1.98% :     0.000566s :      1: loop_unroll
 3.47% :     0.000991s :      1: mutable_eliminate
 3.96% :     0.001130s :     26: opt.transform.jit_opt_a
 0.22% :     0.000064s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000030s :      4: opt.transform.jit_opt_b
 0.06% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000025s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000041s :      4: opt.transform.symbol_engine_opt
 2.06% :     0.000587s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000055s :      1: py_interpret_to_execute
 0.07% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000020s :      1: remove_dup_value
 2.36% :     0.000673s :      1: renormalize.infer
 2.34% :     0.000669s :      1: renormalize.specialize
 0.05% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000052s :      1: rewriter_after_opt_a
 0.29% :     0.000083s :      1: rewriter_before_opt_a
 0.33% :     0.000095s :      1: symbol_engine_optimizer
26.43% :     0.007548s :      1: task_emit
24.06% :     0.006871s :      1: type_inference
 0.29% :     0.000084s :      1: validate


TotalTime = 0.014515, [30]
[bootstrap]: 0.00027084
[type_inference]: 0.00640607
[event_method]: 2.292e-05
[auto_monad]: 8.598e-05
[graph_reusing]: 6.58e-06
[pre_auto_parallel]: 2.64999e-06
[py_interpret_to_execute]: 4.93e-05
[rewriter_before_opt_a]: 8.812e-05
[expand_dump_flag]: 3.21001e-06
[jit_opt_a]: 0.00548764, [2]
    [Cycle 1]: 0.00194324, [27]
        [switch_simplify]: 0.00012055
        [loop_unroll]: 3.637e-05
        [a_1]: 0.00073243
        [with_stream_mark]: 1.987e-05
        [recompute_prepare]: 7.43e-06
        [updatestate_depend_eliminate]: 3.73001e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.64999e-06
        [parameter_eliminate]: 1.77999e-06
        [specialize_transform]: 6.24001e-06
        [updatestate_useless_node_eliminater]: 5.44e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 2.14e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 3.76999e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.479e-05
        [j_node_and_user_rematch]: 9.34e-06
        [meta_fg_expand]: 2.49999e-06
        [replace_old_param]: 8.60001e-06
        [inline_without_move]: 5.15999e-06
        [renormalize]: 0.00070875
        [add_forward_monad_depend]: 5.10001e-06
        [auto_monad_grad]: 2.35002e-06
        [auto_monad_eliminator]: 1.35e-05
        [cse]: 2.846e-05
        [replace_applicator]: 1.173e-05
    [Cycle 2]: 0.00031548, [27]
        [switch_simplify]: 5.86e-06
        [loop_unroll]: 5.25999e-06
        [a_1]: 9.054e-05
        [with_stream_mark]: 8.42e-06
        [recompute_prepare]: 5.04998e-06
        [updatestate_depend_eliminate]: 2.53003e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 5.09998e-06
        [updatestate_useless_node_eliminater]: 5.20001e-06
        [accelerated_algorithm]: 4.57998e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 5.10001e-06
        [merge_forward]: 2.71e-06
        [cell_reuse_recompute_pass]: 1.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.295e-05
        [j_node_and_user_rematch]: 7.74002e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 7.33999e-06
        [inline_without_move]: 4.85999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.40999e-06
        [auto_monad_grad]: 9.90025e-07
        [auto_monad_eliminator]: 4.52998e-06
        [cse]: 9.78998e-06
        [replace_applicator]: 5.05001e-06
[py_interpret_to_execute_after_opt_a]: 1.07e-05
[rewriter_after_opt_a]: 3.47e-05
[convert_after_rewriter]: 7.21999e-06
[order_py_execute_after_rewriter]: 5.05001e-06
[mutable_eliminate]: 0.0005375
[jit_opt_b]: 5.981e-05, [1]
    [Cycle 1]: 4.381e-05, [2]
        [frontend_op_eliminate]: 1.634e-05
        [inline_after_opt_a]: 1.523e-05
[cconv]: 2.479e-05
[loop_unroll]: 0.00042816
[jit_opt_after_cconv]: 0.00013903, [1]
    [Cycle 1]: 0.00013207, [11]
        [c_1]: 2.131e-05
        [parameter_eliminate]: 2.61e-06
        [updatestate_depend_eliminate]: 5.35001e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [cse]: 1.767e-05
        [call_graph_tuple_transform]: 1.856e-05
        [tuple_list_get_item_eliminator]: 5.18002e-06
        [none_parameter_eliminate]: 1.46002e-06
        [renormalize]: 7.00005e-07
        [switch_simplify]: 5.48002e-06
[remove_dup_value]: 1.369e-05
[partial_unused_args_eliminate]: 2.19001e-06
[environ_conv]: 4.95001e-06
[add_recomputation]: 5.051e-05
[cse_after_recomputation]: 2.118e-05, [1]
    [Cycle 1]: 1.529e-05, [1]
        [cse]: 9.46e-06
[auto_monad_reorder]: 1.505e-05
[get_jit_bprop_graph]: 2.25002e-06
[rewriter_after_jit_bprop_graph]: 4.68999e-06
[opt_after_jit_grad]: 0.0004461
[symbol_engine_optimizer]: 7.24e-05, [1]
    [Cycle 1]: 6.614e-05, [6]
        [build]: 3.99002e-06
        [elim_shapecalc]: 7.52998e-06
        [elim_not_effective]: 1.259e-05
        [opt_reshape]: 5.54998e-06
        [fold_const_symbol]: 8.45001e-06
        [renormalize]: 5.00004e-07
[validate]: 3.569e-05
Sums
    bootstrap                            :     0.000271s :  2.53%
    type_inference                       :     0.006406s : 59.91%
    event_method                         :     0.000023s :  0.21%
    auto_monad                           :     0.000086s :  0.80%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000049s :  0.46%
    rewriter_before_opt_a                :     0.000088s :  0.82%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000126s :  1.18%
    jit_opt_a.loop_unroll                :     0.000042s :  0.39%
    jit_opt_a.a_1                        :     0.000823s :  7.70%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000709s :  6.63%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.17%
    jit_opt_a.cse                        :     0.000038s :  0.36%
    jit_opt_a.replace_applicator         :     0.000017s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.10%
    rewriter_after_opt_a                 :     0.000035s :  0.32%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000537s :  5.03%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.14%
    cconv                                :     0.000025s :  0.23%
    loop_unroll                          :     0.000428s :  4.00%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000051s :  0.47%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000446s :  4.17%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.33%


Time group info:
------[substitution.]   0.000282    26
 0.65% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 1.93% :     0.000005s :      3: substitution.graph_param_transform
85.33% :     0.000241s :      6: substitution.inline
 0.99% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.63% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.30% :     0.000004s :      2: substitution.replace_old_param
 2.95% :     0.000008s :      1: substitution.switch_simplify
 4.76% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006282     2
83.98% :     0.005275s :      1: type_inference.infer
16.02% :     0.001006s :      1: type_inference.specialize
------[replace.]   0.000138     9
49.99% :     0.000069s :      6: replace.inline
26.84% :     0.000037s :      1: replace.switch_simplify
23.18% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000257     9
92.31% :     0.000237s :      6: match.inline
 2.96% :     0.000008s :      1: match.switch_simplify
 4.74% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   801
 1.40% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.10% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000002s :     12: predicate.addn_check_dump
 1.35% :     0.000002s :     12: predicate.addn_zero_filter
 2.03% :     0.000003s :     12: predicate.arithmetic_simplify
 1.21% :     0.000002s :     12: predicate.cast_eliminate
 0.37% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     12: predicate.compare_switch_simplify
 1.52% :     0.000002s :     12: predicate.depend_value_elim
 1.12% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     12: predicate.environ_get_eliminate
 1.13% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.76% :     0.000008s :     26: predicate.inline
 0.76% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.94% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     17: predicate.load_eliminater
 1.19% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.49% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.24% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     12: predicate.merge_addn
 1.15% :     0.000002s :     12: predicate.minmaximum_grad
 1.46% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     17: predicate.partial_eliminate
 1.24% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     12: predicate.reduce_eliminate
 1.47% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     20: predicate.replace_applicator
 0.44% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     12: predicate.reshape_eliminate
 1.24% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.66% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     12: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.75% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.98% :     0.000004s :     20: predicate.switch_defer_inline
 2.36% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.44% :     0.000011s :     65: predicate.switch_simplify
 1.21% :     0.000002s :     12: predicate.tile_eliminate
 1.23% :     0.000002s :     12: predicate.transpose_eliminate
 1.69% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.37% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.74% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.89% :     0.000003s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000775    15
49.20% :     0.000381s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.54% :     0.000097s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.26% :     0.000296s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016389    72
 0.33% :     0.000053s :      1: add_recomputation
 0.55% :     0.000090s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 1.77% :     0.000289s :      1: bootstrap
 0.17% :     0.000028s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.17% :     0.000028s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.50% :     0.005490s :      1: jit_opt_a
 0.86% :     0.000142s :      1: jit_opt_after_cconv
 0.38% :     0.000063s :      1: jit_opt_b
 2.66% :     0.000436s :      1: loop_unroll
 3.33% :     0.000546s :      1: mutable_eliminate
 6.74% :     0.001104s :     26: opt.transform.jit_opt_a
 0.29% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.16% :     0.000026s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.77% :     0.000454s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.32% :     0.000052s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.52% :     0.000413s :      1: renormalize.infer
 1.75% :     0.000287s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000038s :      1: rewriter_after_opt_a
 0.55% :     0.000091s :      1: rewriter_before_opt_a
 0.46% :     0.000075s :      1: symbol_engine_optimizer
39.18% :     0.006421s :      1: type_inference


TotalTime = 0.0246772, [33]
[bootstrap]: 0.00034704
[type_inference]: 0.00571254
[event_method]: 2.111e-05
[auto_monad]: 6.908e-05
[graph_reusing]: 6.11e-06
[pre_auto_parallel]: 2.67001e-06
[py_interpret_to_execute]: 5.618e-05
[rewriter_before_opt_a]: 9.725e-05
[expand_dump_flag]: 4.15e-06
[jit_opt_a]: 0.00720367, [2]
    [Cycle 1]: 0.00214124, [27]
        [switch_simplify]: 7.502e-05
        [loop_unroll]: 2.987e-05
        [a_1]: 0.00065438
        [with_stream_mark]: 2.581e-05
        [recompute_prepare]: 7.45e-06
        [updatestate_depend_eliminate]: 4.27e-06
        [updatestate_assign_eliminate]: 3.07002e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 6.17999e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 2.44001e-06
        [get_grad_eliminate_]: 5.25001e-06
        [merge_forward]: 3.88999e-06
        [cell_reuse_recompute_pass]: 1.52001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.511e-05
        [j_node_and_user_rematch]: 1.06e-05
        [meta_fg_expand]: 2.61e-06
        [replace_old_param]: 1.035e-05
        [inline_without_move]: 5.39e-06
        [renormalize]: 0.0009889
        [add_forward_monad_depend]: 7.78001e-06
        [auto_monad_grad]: 2.37001e-06
        [auto_monad_eliminator]: 1.825e-05
        [cse]: 3.327e-05
        [replace_applicator]: 1.647e-05
    [Cycle 2]: 0.00033263, [27]
        [switch_simplify]: 5.66998e-06
        [loop_unroll]: 5.15999e-06
        [a_1]: 9.456e-05
        [with_stream_mark]: 1.224e-05
        [recompute_prepare]: 4.97999e-06
        [updatestate_depend_eliminate]: 2.64001e-06
        [updatestate_assign_eliminate]: 1.95001e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 5.27001e-06
        [updatestate_useless_node_eliminater]: 5.08002e-06
        [accelerated_algorithm]: 4.78001e-06
        [meta_shard_fg_expand]: 1.57001e-06
        [get_grad_eliminate_]: 4.37e-06
        [merge_forward]: 2.96999e-06
        [cell_reuse_recompute_pass]: 3.02002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.438e-05
        [j_node_and_user_rematch]: 9.40001e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 7.32002e-06
        [inline_without_move]: 4.68001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.59e-06
        [auto_monad_grad]: 6.39993e-07
        [auto_monad_eliminator]: 6.24999e-06
        [cse]: 1.094e-05
        [replace_applicator]: 5.39e-06
[py_interpret_to_execute_after_opt_a]: 1.421e-05
[rewriter_after_opt_a]: 4.25e-05
[convert_after_rewriter]: 7.42002e-06
[order_py_execute_after_rewriter]: 5.94999e-06
[mutable_eliminate]: 0.00084877
[jit_opt_b]: 6.219e-05, [1]
    [Cycle 1]: 5.231e-05, [2]
        [frontend_op_eliminate]: 2.248e-05
        [inline_after_opt_a]: 1.64e-05
[cconv]: 3.6e-05
[loop_unroll]: 0.00057392
[jit_opt_after_cconv]: 0.00019721, [1]
    [Cycle 1]: 0.00018892, [11]
        [c_1]: 2.724e-05
        [parameter_eliminate]: 4.88001e-06
        [updatestate_depend_eliminate]: 1.116e-05
        [updatestate_assign_eliminate]: 2.67001e-06
        [updatestate_loads_eliminate]: 3.32997e-06
        [cse]: 3.414e-05
        [call_graph_tuple_transform]: 3.145e-05
        [tuple_list_get_item_eliminator]: 6.49001e-06
        [none_parameter_eliminate]: 1.91e-06
        [renormalize]: 8.29983e-07
        [switch_simplify]: 6.32001e-06
[remove_dup_value]: 1.7e-05
[partial_unused_args_eliminate]: 3.86999e-06
[environ_conv]: 7.03e-06
[add_recomputation]: 6.816e-05
[cse_after_recomputation]: 2.913e-05, [1]
    [Cycle 1]: 2.056e-05, [1]
        [cse]: 1.37e-05
[auto_monad_reorder]: 1.833e-05
[get_jit_bprop_graph]: 3.03998e-06
[rewriter_after_jit_bprop_graph]: 1.063e-05
[opt_after_jit_grad]: 0.00058975
[symbol_engine_optimizer]: 8.485e-05, [1]
    [Cycle 1]: 7.714e-05, [6]
        [build]: 5.66003e-06
        [elim_shapecalc]: 9.49999e-06
        [elim_not_effective]: 1.522e-05
        [opt_reshape]: 6.89001e-06
        [fold_const_symbol]: 1.011e-05
        [renormalize]: 7.60017e-07
[validate]: 5.172e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.00817189
[execute]: 1.032e-05
Sums
    bootstrap                            :     0.000347s :  1.81%
    type_inference                       :     0.005713s : 29.76%
    event_method                         :     0.000021s :  0.11%
    auto_monad                           :     0.000069s :  0.36%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000056s :  0.29%
    rewriter_before_opt_a                :     0.000097s :  0.51%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000081s :  0.42%
    jit_opt_a.loop_unroll                :     0.000035s :  0.18%
    jit_opt_a.a_1                        :     0.000749s :  3.90%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000989s :  5.15%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.13%
    jit_opt_a.cse                        :     0.000044s :  0.23%
    jit_opt_a.replace_applicator         :     0.000022s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.07%
    rewriter_after_opt_a                 :     0.000042s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000849s :  4.42%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000036s :  0.19%
    loop_unroll                          :     0.000574s :  2.99%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000034s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000031s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.09%
    partial_unused_args_eliminate        :     0.000004s :  0.02%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000068s :  0.36%
    cse_after_recomputation.cse          :     0.000014s :  0.07%
    auto_monad_reorder                   :     0.000018s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.06%
    opt_after_jit_grad                   :     0.000590s :  3.07%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000052s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008172s : 42.58%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000289    23
 0.64% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000002s :      2: substitution.fold_const_symbol
 2.64% :     0.000008s :      3: substitution.graph_param_transform
86.34% :     0.000250s :      4: substitution.inline
 1.48% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.61% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.74% :     0.000005s :      2: substitution.replace_old_param
 4.99% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005622     2
79.84% :     0.004489s :      1: type_inference.infer
20.16% :     0.001133s :      1: type_inference.specialize
------[replace.]   0.000079     6
69.15% :     0.000055s :      4: replace.inline
30.85% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000260     6
94.88% :     0.000247s :      4: match.inline
 5.12% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   691
 1.56% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.65% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.92% :     0.000001s :     10: predicate.addn_check_dump
 1.08% :     0.000001s :     10: predicate.addn_zero_filter
 2.01% :     0.000003s :     10: predicate.arithmetic_simplify
 1.32% :     0.000002s :     10: predicate.cast_eliminate
 0.74% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.77% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.09% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.04% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.09% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.25% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.03% :     0.000001s :     10: predicate.environ_get_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.81% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000001s :      3: predicate.graph_param_transform
 5.33% :     0.000007s :     22: predicate.inline
 0.75% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000002s :      6: predicate.less_batch_normalization
 1.62% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.53% :     0.000002s :     15: predicate.load_eliminater
 2.00% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.34% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.24% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 3.31% :     0.000004s :      3: predicate.mutable_eliminate
 0.66% :     0.000001s :      3: predicate.opt_reshape
 1.95% :     0.000003s :     15: predicate.partial_eliminate
 1.35% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.65% :     0.000004s :     10: predicate.reduce_eliminate
 1.24% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.44% :     0.000002s :     18: predicate.replace_applicator
 0.39% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.53% :     0.000002s :     10: predicate.reshape_eliminate
 1.02% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.32% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.44% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.68% :     0.000002s :      6: predicate.special_op_eliminate
 0.83% :     0.000001s :      6: predicate.specialize_transform
 1.61% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.14% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.95% :     0.000009s :     53: predicate.switch_simplify
 1.49% :     0.000002s :     10: predicate.tile_eliminate
 1.02% :     0.000001s :     10: predicate.transpose_eliminate
 1.90% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.28% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.41% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.43% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000850    12
44.42% :     0.000378s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.12% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.45% :     0.000412s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026748    76
 0.27% :     0.000072s :      1: add_recomputation
 0.27% :     0.000073s :      1: auto_monad
 0.08% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.38% :     0.000368s :      1: bootstrap
 0.15% :     0.000039s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000033s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.10% :     0.000026s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.95% :     0.007208s :      1: jit_opt_a
 0.75% :     0.000201s :      1: jit_opt_after_cconv
 0.25% :     0.000066s :      1: jit_opt_b
 2.18% :     0.000583s :      1: loop_unroll
 3.23% :     0.000864s :      1: mutable_eliminate
 3.68% :     0.000985s :     26: opt.transform.jit_opt_a
 0.25% :     0.000067s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000032s :      4: opt.transform.jit_opt_b
 0.06% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000038s :      4: opt.transform.symbol_engine_opt
 2.25% :     0.000601s :      1: opt_after_jit_grad
 0.03% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000006s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000059s :      1: py_interpret_to_execute
 0.06% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000020s :      1: remove_dup_value
 2.10% :     0.000562s :      1: renormalize.infer
 1.56% :     0.000417s :      1: renormalize.specialize
 0.05% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000046s :      1: rewriter_after_opt_a
 0.38% :     0.000101s :      1: rewriter_before_opt_a
 0.33% :     0.000088s :      1: symbol_engine_optimizer
30.65% :     0.008198s :      1: task_emit
21.44% :     0.005735s :      1: type_inference
 0.29% :     0.000078s :      1: validate


TotalTime = 0.0143089, [30]
[bootstrap]: 0.00027403
[type_inference]: 0.00643076
[event_method]: 2.206e-05
[auto_monad]: 7.933e-05
[graph_reusing]: 6.38998e-06
[pre_auto_parallel]: 2.81999e-06
[py_interpret_to_execute]: 4.86e-05
[rewriter_before_opt_a]: 8.667e-05
[expand_dump_flag]: 2.88e-06
[jit_opt_a]: 0.00535212, [2]
    [Cycle 1]: 0.00181228, [27]
        [switch_simplify]: 0.00011439
        [loop_unroll]: 3.465e-05
        [a_1]: 0.00068164
        [with_stream_mark]: 1.833e-05
        [recompute_prepare]: 7.3e-06
        [updatestate_depend_eliminate]: 3.58e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.81999e-06
        [parameter_eliminate]: 1.98002e-06
        [specialize_transform]: 6.15002e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 5.59998e-06
        [meta_shard_fg_expand]: 1.99999e-06
        [get_grad_eliminate_]: 5.21002e-06
        [merge_forward]: 3.98999e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.488e-05
        [j_node_and_user_rematch]: 8.55999e-06
        [meta_fg_expand]: 2.33998e-06
        [replace_old_param]: 8.80001e-06
        [inline_without_move]: 5.25999e-06
        [renormalize]: 0.00063962
        [add_forward_monad_depend]: 5.04e-06
        [auto_monad_grad]: 1.99999e-06
        [auto_monad_eliminator]: 1.349e-05
        [cse]: 3.025e-05
        [replace_applicator]: 1.196e-05
    [Cycle 2]: 0.00032236, [27]
        [switch_simplify]: 5.70001e-06
        [loop_unroll]: 5.00001e-06
        [a_1]: 9.01e-05
        [with_stream_mark]: 9.07999e-06
        [recompute_prepare]: 5.09e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 2.06998e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 4.92999e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 4.96002e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 4.38999e-06
        [merge_forward]: 2.61e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.899e-05
        [j_node_and_user_rematch]: 7.48e-06
        [meta_fg_expand]: 1.57999e-06
        [replace_old_param]: 7.46001e-06
        [inline_without_move]: 4.80999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.14998e-06
        [auto_monad_grad]: 1.00001e-06
        [auto_monad_eliminator]: 5.28002e-06
        [cse]: 1.071e-05
        [replace_applicator]: 7.63001e-06
[py_interpret_to_execute_after_opt_a]: 8.67998e-06
[rewriter_after_opt_a]: 3.439e-05
[convert_after_rewriter]: 6.84001e-06
[order_py_execute_after_rewriter]: 5.10999e-06
[mutable_eliminate]: 0.00048195
[jit_opt_b]: 4.898e-05, [1]
    [Cycle 1]: 4.195e-05, [2]
        [frontend_op_eliminate]: 1.605e-05
        [inline_after_opt_a]: 1.451e-05
[cconv]: 2.536e-05
[loop_unroll]: 0.00040529
[jit_opt_after_cconv]: 0.00013526, [1]
    [Cycle 1]: 0.00012912, [11]
        [c_1]: 2.042e-05
        [parameter_eliminate]: 2.37999e-06
        [updatestate_depend_eliminate]: 5.17e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.702e-05
        [call_graph_tuple_transform]: 1.843e-05
        [tuple_list_get_item_eliminator]: 5.17999e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 5.52999e-06
[remove_dup_value]: 1.308e-05
[partial_unused_args_eliminate]: 2.04999e-06
[environ_conv]: 5.01997e-06
[add_recomputation]: 4.619e-05
[cse_after_recomputation]: 1.985e-05, [1]
    [Cycle 1]: 1.456e-05, [1]
        [cse]: 9.04998e-06
[auto_monad_reorder]: 1.467e-05
[get_jit_bprop_graph]: 1.63002e-06
[rewriter_after_jit_bprop_graph]: 4.82998e-06
[opt_after_jit_grad]: 0.00046682
[symbol_engine_optimizer]: 7.118e-05, [1]
    [Cycle 1]: 6.476e-05, [6]
        [build]: 3.80998e-06
        [elim_shapecalc]: 7.39002e-06
        [elim_not_effective]: 1.198e-05
        [opt_reshape]: 5.46998e-06
        [fold_const_symbol]: 8.45999e-06
        [renormalize]: 3.89991e-07
[validate]: 3.278e-05
Sums
    bootstrap                            :     0.000274s :  2.61%
    type_inference                       :     0.006431s : 61.14%
    event_method                         :     0.000022s :  0.21%
    auto_monad                           :     0.000079s :  0.75%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000049s :  0.46%
    rewriter_before_opt_a                :     0.000087s :  0.82%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000120s :  1.14%
    jit_opt_a.loop_unroll                :     0.000040s :  0.38%
    jit_opt_a.a_1                        :     0.000772s :  7.34%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000640s :  6.08%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.18%
    jit_opt_a.cse                        :     0.000041s :  0.39%
    jit_opt_a.replace_applicator         :     0.000020s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.08%
    rewriter_after_opt_a                 :     0.000034s :  0.33%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000482s :  4.58%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.14%
    cconv                                :     0.000025s :  0.24%
    loop_unroll                          :     0.000405s :  3.85%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000013s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000046s :  0.44%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000467s :  4.44%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.31%


Time group info:
------[substitution.]   0.000277    26
 0.63% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 1.82% :     0.000005s :      3: substitution.graph_param_transform
83.15% :     0.000230s :      6: substitution.inline
 0.93% :     0.000003s :      4: substitution.j_node_and_user_rematch
 4.36% :     0.000012s :      4: substitution.remove_not_recompute_node
 1.37% :     0.000004s :      2: substitution.replace_old_param
 2.78% :     0.000008s :      1: substitution.switch_simplify
 4.51% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006363     2
82.94% :     0.005277s :      1: type_inference.infer
17.06% :     0.001085s :      1: type_inference.specialize
------[replace.]   0.000132     9
50.97% :     0.000067s :      6: replace.inline
26.12% :     0.000034s :      1: replace.switch_simplify
22.91% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000244     9
92.56% :     0.000226s :      6: match.inline
 2.79% :     0.000007s :      1: match.switch_simplify
 4.65% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   801
 1.35% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.10% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     12: predicate.addn_check_dump
 1.29% :     0.000002s :     12: predicate.addn_zero_filter
 2.13% :     0.000003s :     12: predicate.arithmetic_simplify
 1.44% :     0.000002s :     12: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     12: predicate.compare_switch_simplify
 1.38% :     0.000002s :     12: predicate.depend_value_elim
 1.21% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.41% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.35% :     0.000002s :     12: predicate.environ_get_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.50% :     0.000001s :      3: predicate.graph_param_transform
 5.67% :     0.000007s :     26: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.99% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     17: predicate.load_eliminater
 1.18% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.68% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.81% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     12: predicate.merge_addn
 1.18% :     0.000002s :     12: predicate.minmaximum_grad
 1.48% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     17: predicate.partial_eliminate
 1.18% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.83% :     0.000002s :     12: predicate.reduce_eliminate
 1.54% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     20: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     12: predicate.reshape_eliminate
 1.30% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     12: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.61% :     0.000003s :     20: predicate.switch_defer_inline
 2.37% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.87% :     0.000012s :     65: predicate.switch_simplify
 1.28% :     0.000002s :     12: predicate.tile_eliminate
 1.18% :     0.000002s :     12: predicate.transpose_eliminate
 1.51% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.70% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.67% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000775    15
49.94% :     0.000387s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.80% :     0.000092s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.25% :     0.000297s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016053    72
 0.31% :     0.000049s :      1: add_recomputation
 0.52% :     0.000083s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.82% :     0.000292s :      1: bootstrap
 0.17% :     0.000028s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
33.36% :     0.005356s :      1: jit_opt_a
 0.86% :     0.000138s :      1: jit_opt_after_cconv
 0.32% :     0.000051s :      1: jit_opt_b
 2.57% :     0.000413s :      1: loop_unroll
 3.06% :     0.000491s :      1: mutable_eliminate
 6.55% :     0.001052s :     26: opt.transform.jit_opt_a
 0.29% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.96% :     0.000475s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.32% :     0.000051s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.17% :     0.000349s :      1: renormalize.infer
 1.77% :     0.000284s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000037s :      1: rewriter_after_opt_a
 0.56% :     0.000090s :      1: rewriter_before_opt_a
 0.46% :     0.000074s :      1: symbol_engine_optimizer
40.15% :     0.006446s :      1: type_inference


TotalTime = 0.0232662, [33]
[bootstrap]: 0.00035941
[type_inference]: 0.00589164
[event_method]: 2.4e-05
[auto_monad]: 7.662e-05
[graph_reusing]: 6.89001e-06
[pre_auto_parallel]: 3.35998e-06
[py_interpret_to_execute]: 5.832e-05
[rewriter_before_opt_a]: 8.621e-05
[expand_dump_flag]: 3.63999e-06
[jit_opt_a]: 0.00697942, [2]
    [Cycle 1]: 0.00217904, [27]
        [switch_simplify]: 7.149e-05
        [loop_unroll]: 2.991e-05
        [a_1]: 0.00082578
        [with_stream_mark]: 2.133e-05
        [recompute_prepare]: 9.23002e-06
        [updatestate_depend_eliminate]: 3.82002e-06
        [updatestate_assign_eliminate]: 3.2e-06
        [updatestate_loads_eliminate]: 2.81999e-06
        [parameter_eliminate]: 2.54999e-06
        [specialize_transform]: 6.62002e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 5.86e-06
        [meta_shard_fg_expand]: 2.66e-06
        [get_grad_eliminate_]: 5.76998e-06
        [merge_forward]: 4.18999e-06
        [cell_reuse_recompute_pass]: 2.00002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.593e-05
        [j_node_and_user_rematch]: 1.022e-05
        [meta_fg_expand]: 2.21e-06
        [replace_old_param]: 1.007e-05
        [inline_without_move]: 5.56e-06
        [renormalize]: 0.00087469
        [add_forward_monad_depend]: 6.36998e-06
        [auto_monad_grad]: 2.93998e-06
        [auto_monad_eliminator]: 1.588e-05
        [cse]: 3.022e-05
        [replace_applicator]: 1.507e-05
    [Cycle 2]: 0.00033993, [27]
        [switch_simplify]: 5.79999e-06
        [loop_unroll]: 4.85999e-06
        [a_1]: 9.443e-05
        [with_stream_mark]: 9.79999e-06
        [recompute_prepare]: 5.17e-06
        [updatestate_depend_eliminate]: 2.94999e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.35999e-06
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 1.45999e-06
        [get_grad_eliminate_]: 4.74002e-06
        [merge_forward]: 2.58e-06
        [cell_reuse_recompute_pass]: 3.46001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.431e-05
        [j_node_and_user_rematch]: 8.15999e-06
        [meta_fg_expand]: 1.72001e-06
        [replace_old_param]: 7.72002e-06
        [inline_without_move]: 4.89003e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.09998e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 6.07999e-06
        [cse]: 1.223e-05
        [replace_applicator]: 1.291e-05
[py_interpret_to_execute_after_opt_a]: 1.426e-05
[rewriter_after_opt_a]: 3.788e-05
[convert_after_rewriter]: 7.85998e-06
[order_py_execute_after_rewriter]: 5.10001e-06
[mutable_eliminate]: 0.00073899
[jit_opt_b]: 5.987e-05, [1]
    [Cycle 1]: 5.216e-05, [2]
        [frontend_op_eliminate]: 2.357e-05
        [inline_after_opt_a]: 1.65e-05
[cconv]: 3.186e-05
[loop_unroll]: 0.00046945
[jit_opt_after_cconv]: 0.00015095, [1]
    [Cycle 1]: 0.00014455, [11]
        [c_1]: 2.278e-05
        [parameter_eliminate]: 3.26001e-06
        [updatestate_depend_eliminate]: 7.68999e-06
        [updatestate_assign_eliminate]: 2.87002e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 2.242e-05
        [call_graph_tuple_transform]: 1.992e-05
        [tuple_list_get_item_eliminator]: 5.09e-06
        [none_parameter_eliminate]: 1.77999e-06
        [renormalize]: 6.29982e-07
        [switch_simplify]: 5.40999e-06
[remove_dup_value]: 1.503e-05
[partial_unused_args_eliminate]: 2.56e-06
[environ_conv]: 4.89e-06
[add_recomputation]: 5.395e-05
[cse_after_recomputation]: 2.252e-05, [1]
    [Cycle 1]: 1.579e-05, [1]
        [cse]: 9.72999e-06
[auto_monad_reorder]: 1.673e-05
[get_jit_bprop_graph]: 2.70002e-06
[rewriter_after_jit_bprop_graph]: 7.33999e-06
[opt_after_jit_grad]: 0.00061048
[symbol_engine_optimizer]: 8.407e-05, [1]
    [Cycle 1]: 7.613e-05, [6]
        [build]: 5.35001e-06
        [elim_shapecalc]: 9.37999e-06
        [elim_not_effective]: 1.513e-05
        [opt_reshape]: 5.67001e-06
        [fold_const_symbol]: 8.80999e-06
        [renormalize]: 8.70001e-07
[validate]: 4.461e-05
[backend_pass]: 9.10019e-07
[task_emit]: 0.00714277
[execute]: 7.45e-06
Sums
    bootstrap                            :     0.000359s :  1.98%
    type_inference                       :     0.005892s : 32.49%
    event_method                         :     0.000024s :  0.13%
    auto_monad                           :     0.000077s :  0.42%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000058s :  0.32%
    rewriter_before_opt_a                :     0.000086s :  0.48%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000077s :  0.43%
    jit_opt_a.loop_unroll                :     0.000035s :  0.19%
    jit_opt_a.a_1                        :     0.000920s :  5.08%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000875s :  4.82%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.12%
    jit_opt_a.cse                        :     0.000042s :  0.23%
    jit_opt_a.replace_applicator         :     0.000028s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000739s :  4.08%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000032s :  0.18%
    loop_unroll                          :     0.000469s :  2.59%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000054s :  0.30%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000610s :  3.37%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007143s : 39.39%
    execute                              :     0.000007s :  0.04%


Time group info:
------[substitution.]   0.000442    23
 0.56% :     0.000002s :      2: substitution.elim_not_effective
 0.28% :     0.000001s :      2: substitution.fold_const_symbol
 1.32% :     0.000006s :      3: substitution.graph_param_transform
91.25% :     0.000404s :      4: substitution.inline
 0.77% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.14% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.21% :     0.000005s :      2: substitution.replace_old_param
 3.47% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005802     2
80.49% :     0.004670s :      1: type_inference.infer
19.51% :     0.001132s :      1: type_inference.specialize
------[replace.]   0.000093     6
65.11% :     0.000061s :      4: replace.inline
34.89% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000414     6
96.62% :     0.000400s :      4: match.inline
 3.38% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.27% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 3.45% :     0.000004s :     10: predicate.arithmetic_simplify
 1.18% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.05% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.16% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 6.00% :     0.000008s :     22: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.07% :     0.000001s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.63% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.61% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.08% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.01% :     0.000001s :     10: predicate.same_eliminate
 0.66% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.35% :     0.000002s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.86% :     0.000004s :     16: predicate.switch_defer_inline
 1.94% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.16% :     0.000009s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.96% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.61% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.89% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000948    12
39.46% :     0.000374s :      5: func_graph_cloner_run.FuncGraphClonerGraph
23.51% :     0.000223s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.02% :     0.000351s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025428    76
 0.23% :     0.000058s :      1: add_recomputation
 0.32% :     0.000081s :      1: auto_monad
 0.08% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.49% :     0.000378s :      1: bootstrap
 0.14% :     0.000035s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.11% :     0.000028s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
27.47% :     0.006985s :      1: jit_opt_a
 0.61% :     0.000154s :      1: jit_opt_after_cconv
 0.25% :     0.000063s :      1: jit_opt_b
 1.89% :     0.000480s :      1: loop_unroll
 2.94% :     0.000749s :      1: mutable_eliminate
 4.58% :     0.001164s :     26: opt.transform.jit_opt_a
 0.20% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000033s :      4: opt.transform.jit_opt_b
 0.05% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000037s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.45% :     0.000622s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000062s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 1.81% :     0.000461s :      1: renormalize.infer
 1.59% :     0.000405s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000041s :      1: rewriter_after_opt_a
 0.36% :     0.000091s :      1: rewriter_before_opt_a
 0.34% :     0.000087s :      1: symbol_engine_optimizer
28.14% :     0.007156s :      1: task_emit
23.28% :     0.005919s :      1: type_inference
 0.26% :     0.000066s :      1: validate


TotalTime = 0.0229167, [33]
[bootstrap]: 0.00034647
[type_inference]: 0.00536319
[event_method]: 1.906e-05
[auto_monad]: 6.096e-05
[graph_reusing]: 6.23002e-06
[pre_auto_parallel]: 2.36e-06
[py_interpret_to_execute]: 4.313e-05
[rewriter_before_opt_a]: 7.425e-05
[expand_dump_flag]: 3.38e-06
[jit_opt_a]: 0.0082292, [2]
    [Cycle 1]: 0.00389928, [27]
        [switch_simplify]: 5.533e-05
        [loop_unroll]: 3.03e-05
        [a_1]: 0.00055923
        [with_stream_mark]: 1.517e-05
        [recompute_prepare]: 7.15e-06
        [updatestate_depend_eliminate]: 3.05002e-06
        [updatestate_assign_eliminate]: 2.41998e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 6.07001e-06
        [updatestate_useless_node_eliminater]: 5.10999e-06
        [accelerated_algorithm]: 5.96998e-06
        [meta_shard_fg_expand]: 1.40001e-06
        [get_grad_eliminate_]: 5.92999e-06
        [merge_forward]: 3.26001e-06
        [cell_reuse_recompute_pass]: 1.20999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.175e-05
        [j_node_and_user_rematch]: 9.96e-06
        [meta_fg_expand]: 1.92001e-06
        [replace_old_param]: 8.64003e-06
        [inline_without_move]: 5.69e-06
        [renormalize]: 0.00289888
        [add_forward_monad_depend]: 9.05999e-06
        [auto_monad_grad]: 2.65002e-06
        [auto_monad_eliminator]: 1.771e-05
        [cse]: 2.615e-05
        [replace_applicator]: 1.944e-05
    [Cycle 2]: 0.00035339, [27]
        [switch_simplify]: 6.73e-06
        [loop_unroll]: 5.97999e-06
        [a_1]: 0.00010453
        [with_stream_mark]: 1.274e-05
        [recompute_prepare]: 5.10999e-06
        [updatestate_depend_eliminate]: 3.02002e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 5.34e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 5.40999e-06
        [meta_shard_fg_expand]: 1.97999e-06
        [get_grad_eliminate_]: 4.75999e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 3.88999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.669e-05
        [j_node_and_user_rematch]: 9.05999e-06
        [meta_fg_expand]: 1.86998e-06
        [replace_old_param]: 9.32001e-06
        [inline_without_move]: 4.63999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.40001e-06
        [auto_monad_grad]: 6.59988e-07
        [auto_monad_eliminator]: 5.34e-06
        [cse]: 1.028e-05
        [replace_applicator]: 4.93001e-06
[py_interpret_to_execute_after_opt_a]: 1.62e-05
[rewriter_after_opt_a]: 3.81e-05
[convert_after_rewriter]: 7.7e-06
[order_py_execute_after_rewriter]: 5.18002e-06
[mutable_eliminate]: 0.00086843
[jit_opt_b]: 6.472e-05, [1]
    [Cycle 1]: 4.57e-05, [2]
        [frontend_op_eliminate]: 1.882e-05
        [inline_after_opt_a]: 1.502e-05
[cconv]: 2.915e-05
[loop_unroll]: 0.00045572
[jit_opt_after_cconv]: 0.00014777, [1]
    [Cycle 1]: 0.00014089, [11]
        [c_1]: 2.182e-05
        [parameter_eliminate]: 2.78e-06
        [updatestate_depend_eliminate]: 5.40999e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [cse]: 2.04e-05
        [call_graph_tuple_transform]: 1.977e-05
        [tuple_list_get_item_eliminator]: 5.82999e-06
        [none_parameter_eliminate]: 1.95001e-06
        [renormalize]: 6.69999e-07
        [switch_simplify]: 5.67001e-06
[remove_dup_value]: 1.344e-05
[partial_unused_args_eliminate]: 2.56e-06
[environ_conv]: 4.71002e-06
[add_recomputation]: 4.882e-05
[cse_after_recomputation]: 2.171e-05, [1]
    [Cycle 1]: 1.607e-05, [1]
        [cse]: 9.99001e-06
[auto_monad_reorder]: 1.62e-05
[get_jit_bprop_graph]: 2.47001e-06
[rewriter_after_jit_bprop_graph]: 6.04999e-06
[opt_after_jit_grad]: 0.00047267
[symbol_engine_optimizer]: 7.946e-05, [1]
    [Cycle 1]: 7.275e-05, [6]
        [build]: 4.31002e-06
        [elim_shapecalc]: 8.43999e-06
        [elim_not_effective]: 1.42e-05
        [opt_reshape]: 6.53e-06
        [fold_const_symbol]: 9.36002e-06
        [renormalize]: 1.00001e-06
[validate]: 3.975e-05
[backend_pass]: 1.26002e-06
[task_emit]: 0.0061961
[execute]: 1.01e-05
Sums
    bootstrap                            :     0.000346s :  1.89%
    type_inference                       :     0.005363s : 29.33%
    event_method                         :     0.000019s :  0.10%
    auto_monad                           :     0.000061s :  0.33%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.24%
    rewriter_before_opt_a                :     0.000074s :  0.41%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.34%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000664s :  3.63%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.002899s : 15.85%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.13%
    jit_opt_a.cse                        :     0.000036s :  0.20%
    jit_opt_a.replace_applicator         :     0.000024s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.09%
    rewriter_after_opt_a                 :     0.000038s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000868s :  4.75%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000029s :  0.16%
    loop_unroll                          :     0.000456s :  2.49%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000013s :  0.07%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.27%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000473s :  2.58%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006196s : 33.89%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000224    23
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000002s :      2: substitution.fold_const_symbol
 2.63% :     0.000006s :      3: substitution.graph_param_transform
83.60% :     0.000187s :      4: substitution.inline
 1.98% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.01% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.49% :     0.000006s :      2: substitution.replace_old_param
 5.63% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005292     2
81.22% :     0.004299s :      1: type_inference.infer
18.78% :     0.000994s :      1: type_inference.specialize
------[replace.]   0.000077     6
66.80% :     0.000051s :      4: replace.inline
33.20% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000196     6
94.15% :     0.000185s :      4: match.inline
 5.85% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.58% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 1.94% :     0.000002s :     10: predicate.arithmetic_simplify
 1.53% :     0.000002s :     10: predicate.cast_eliminate
 0.61% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000002s :     10: predicate.depend_value_elim
 1.22% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.39% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.65% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.37% :     0.000002s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.36% :     0.000002s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.27% :     0.000007s :     22: predicate.inline
 1.02% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.52% :     0.000002s :      6: predicate.less_batch_normalization
 1.46% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.61% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.89% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.08% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000003s :     15: predicate.partial_eliminate
 1.34% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.74% :     0.000002s :     10: predicate.reduce_eliminate
 1.75% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.08% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.51% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.14% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.52% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.14% :     0.000003s :     16: predicate.switch_defer_inline
 1.83% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.98% :     0.000009s :     53: predicate.switch_simplify
 1.39% :     0.000002s :     10: predicate.tile_eliminate
 1.02% :     0.000001s :     10: predicate.transpose_eliminate
 1.57% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.04% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.13% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000687    12
46.01% :     0.000316s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.94% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.05% :     0.000330s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026803    76
 0.20% :     0.000052s :      1: add_recomputation
 0.24% :     0.000065s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.34% :     0.000359s :      1: bootstrap
 0.12% :     0.000032s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.09% :     0.000023s :      1: event_method
 0.05% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
30.72% :     0.008234s :      1: jit_opt_a
 0.56% :     0.000151s :      1: jit_opt_after_cconv
 0.25% :     0.000068s :      1: jit_opt_b
 1.73% :     0.000464s :      1: loop_unroll
 3.27% :     0.000877s :      1: mutable_eliminate
 3.32% :     0.000890s :     26: opt.transform.jit_opt_a
 0.19% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000027s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.79% :     0.000481s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000046s :      1: py_interpret_to_execute
 0.07% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000016s :      1: remove_dup_value
 1.95% :     0.000523s :      1: renormalize.infer
 8.83% :     0.002367s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000041s :      1: rewriter_after_opt_a
 0.29% :     0.000077s :      1: rewriter_before_opt_a
 0.31% :     0.000082s :      1: symbol_engine_optimizer
23.19% :     0.006215s :      1: task_emit
20.07% :     0.005380s :      1: type_inference
 0.23% :     0.000061s :      1: validate


TotalTime = 0.0160383, [30]
[bootstrap]: 0.00026482
[type_inference]: 0.00671743
[event_method]: 2.173e-05
[auto_monad]: 8.151e-05
[graph_reusing]: 6.88e-06
[pre_auto_parallel]: 2.92002e-06
[py_interpret_to_execute]: 4.888e-05
[rewriter_before_opt_a]: 8.742e-05
[expand_dump_flag]: 2.94999e-06
[jit_opt_a]: 0.0062106, [2]
    [Cycle 1]: 0.0022184, [27]
        [switch_simplify]: 0.00015156
        [loop_unroll]: 3.606e-05
        [a_1]: 0.00079026
        [with_stream_mark]: 2.579e-05
        [recompute_prepare]: 1.043e-05
        [updatestate_depend_eliminate]: 4.75999e-06
        [updatestate_assign_eliminate]: 3.36999e-06
        [updatestate_loads_eliminate]: 2.84999e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 7.39002e-06
        [updatestate_useless_node_eliminater]: 5.66998e-06
        [accelerated_algorithm]: 6.59999e-06
        [meta_shard_fg_expand]: 2.14e-06
        [get_grad_eliminate_]: 5.86e-06
        [merge_forward]: 3.48e-06
        [cell_reuse_recompute_pass]: 1.71002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.588e-05
        [j_node_and_user_rematch]: 1.072e-05
        [meta_fg_expand]: 2.32001e-06
        [replace_old_param]: 1.01e-05
        [inline_without_move]: 5.86e-06
        [renormalize]: 0.00082075
        [add_forward_monad_depend]: 7.04001e-06
        [auto_monad_grad]: 2.51998e-06
        [auto_monad_eliminator]: 1.768e-05
        [cse]: 3.27e-05
        [replace_applicator]: 1.251e-05
    [Cycle 2]: 0.00033116, [27]
        [switch_simplify]: 6.88e-06
        [loop_unroll]: 5.57001e-06
        [a_1]: 9.491e-05
        [with_stream_mark]: 1.102e-05
        [recompute_prepare]: 5.34e-06
        [updatestate_depend_eliminate]: 2.69001e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 5.51e-06
        [updatestate_useless_node_eliminater]: 5.32999e-06
        [accelerated_algorithm]: 4.96997e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 5.06002e-06
        [merge_forward]: 2.77002e-06
        [cell_reuse_recompute_pass]: 1.88997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.436e-05
        [j_node_and_user_rematch]: 7.84997e-06
        [meta_fg_expand]: 1.47999e-06
        [replace_old_param]: 7.38e-06
        [inline_without_move]: 4.63001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 6.39993e-07
        [auto_monad_eliminator]: 5.02999e-06
        [cse]: 1.033e-05
        [replace_applicator]: 5.67999e-06
[py_interpret_to_execute_after_opt_a]: 1.23e-05
[rewriter_after_opt_a]: 3.573e-05
[convert_after_rewriter]: 7.93001e-06
[order_py_execute_after_rewriter]: 5.88998e-06
[mutable_eliminate]: 0.00073927
[jit_opt_b]: 5.579e-05, [1]
    [Cycle 1]: 4.723e-05, [2]
        [frontend_op_eliminate]: 1.763e-05
        [inline_after_opt_a]: 1.668e-05
[cconv]: 3.072e-05
[loop_unroll]: 0.0005418
[jit_opt_after_cconv]: 0.00017062, [1]
    [Cycle 1]: 0.00016222, [11]
        [c_1]: 2.189e-05
        [parameter_eliminate]: 4.20999e-06
        [updatestate_depend_eliminate]: 8.00999e-06
        [updatestate_assign_eliminate]: 2.70002e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [cse]: 2.379e-05
        [call_graph_tuple_transform]: 2.051e-05
        [tuple_list_get_item_eliminator]: 5.27999e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 1.13001e-06
        [switch_simplify]: 7.00998e-06
[remove_dup_value]: 1.444e-05
[partial_unused_args_eliminate]: 2.32001e-06
[environ_conv]: 5.07999e-06
[add_recomputation]: 5.422e-05
[cse_after_recomputation]: 2.55e-05, [1]
    [Cycle 1]: 1.904e-05, [1]
        [cse]: 1.103e-05
[auto_monad_reorder]: 1.753e-05
[get_jit_bprop_graph]: 2.08002e-06
[rewriter_after_jit_bprop_graph]: 6.98e-06
[opt_after_jit_grad]: 0.00054899
[symbol_engine_optimizer]: 7.761e-05, [1]
    [Cycle 1]: 7.034e-05, [6]
        [build]: 5.12999e-06
        [elim_shapecalc]: 7.98001e-06
        [elim_not_effective]: 1.322e-05
        [opt_reshape]: 5.49e-06
        [fold_const_symbol]: 8.40001e-06
        [renormalize]: 1.18001e-06
[validate]: 3.941e-05
Sums
    bootstrap                            :     0.000265s :  2.26%
    type_inference                       :     0.006717s : 57.42%
    event_method                         :     0.000022s :  0.19%
    auto_monad                           :     0.000082s :  0.70%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000049s :  0.42%
    rewriter_before_opt_a                :     0.000087s :  0.75%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000158s :  1.35%
    jit_opt_a.loop_unroll                :     0.000042s :  0.36%
    jit_opt_a.a_1                        :     0.000885s :  7.57%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000821s :  7.02%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.19%
    jit_opt_a.cse                        :     0.000043s :  0.37%
    jit_opt_a.replace_applicator         :     0.000018s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.11%
    rewriter_after_opt_a                 :     0.000036s :  0.31%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000739s :  6.32%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.14%
    cconv                                :     0.000031s :  0.26%
    loop_unroll                          :     0.000542s :  4.63%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.06%
    remove_dup_value                     :     0.000014s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000054s :  0.46%
    cse_after_recomputation.cse          :     0.000011s :  0.09%
    auto_monad_reorder                   :     0.000018s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000549s :  4.69%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.34%


Time group info:
------[substitution.]   0.000339    26
 0.52% :     0.000002s :      2: substitution.elim_not_effective
 0.37% :     0.000001s :      2: substitution.fold_const_symbol
 1.59% :     0.000005s :      3: substitution.graph_param_transform
86.48% :     0.000293s :      6: substitution.inline
 1.22% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.34% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.44% :     0.000005s :      2: substitution.replace_old_param
 2.98% :     0.000010s :      1: substitution.switch_simplify
 4.05% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006648     2
84.16% :     0.005594s :      1: type_inference.infer
15.84% :     0.001053s :      1: type_inference.specialize
------[replace.]   0.000162     9
49.76% :     0.000081s :      6: replace.inline
28.63% :     0.000046s :      1: replace.switch_simplify
21.61% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000311     9
93.04% :     0.000289s :      6: match.inline
 2.93% :     0.000009s :      1: match.switch_simplify
 4.02% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000150   801
 1.19% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.72% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000002s :     12: predicate.addn_check_dump
 1.31% :     0.000002s :     12: predicate.addn_zero_filter
 2.23% :     0.000003s :     12: predicate.arithmetic_simplify
 1.13% :     0.000002s :     12: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000002s :     12: predicate.compare_switch_simplify
 1.50% :     0.000002s :     12: predicate.depend_value_elim
 1.07% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.15% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     12: predicate.dict_set_item_eliminator
 1.32% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.23% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.06% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     12: predicate.environ_get_eliminate
 1.06% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.45% :     0.000001s :      3: predicate.graph_param_transform
 5.05% :     0.000008s :     26: predicate.inline
 0.78% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000002s :      6: predicate.less_batch_normalization
 1.82% :     0.000003s :     14: predicate.list_to_tuple_eliminator_
 1.78% :     0.000003s :     17: predicate.load_eliminater
 1.96% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.23% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.96% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000002s :     12: predicate.merge_addn
 1.05% :     0.000002s :     12: predicate.minmaximum_grad
 2.30% :     0.000003s :      3: predicate.mutable_eliminate
 0.36% :     0.000001s :      3: predicate.opt_reshape
 2.00% :     0.000003s :     17: predicate.partial_eliminate
 1.23% :     0.000002s :     12: predicate.print_const_string_wrapper
 2.01% :     0.000003s :     12: predicate.reduce_eliminate
 1.59% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.39% :     0.000002s :     20: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     12: predicate.reshape_eliminate
 1.39% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 1.03% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000002s :     12: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.67% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.25% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.09% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.33% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.34% :     0.000005s :     20: predicate.switch_defer_inline
 2.17% :     0.000003s :     20: predicate.switch_layer_defer_inline
 9.34% :     0.000014s :     65: predicate.switch_simplify
 1.15% :     0.000002s :     12: predicate.tile_eliminate
 1.07% :     0.000002s :     12: predicate.transpose_eliminate
 1.37% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.58% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000003s :     12: predicate.tuple_list_set_item_eliminator
 1.35% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     12: predicate.value_based_eliminate
 0.25% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000843    15
48.57% :     0.000410s :      6: func_graph_cloner_run.FuncGraphClonerGraph
14.43% :     0.000122s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.00% :     0.000312s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018134    72
 0.32% :     0.000059s :      1: add_recomputation
 0.47% :     0.000085s :      1: auto_monad
 0.11% :     0.000020s :      1: auto_monad_reorder
 1.57% :     0.000284s :      1: bootstrap
 0.18% :     0.000033s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.15% :     0.000028s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.14% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
34.27% :     0.006214s :      1: jit_opt_a
 0.96% :     0.000174s :      1: jit_opt_after_cconv
 0.32% :     0.000059s :      1: jit_opt_b
 3.06% :     0.000554s :      1: loop_unroll
 4.16% :     0.000754s :      1: mutable_eliminate
 6.67% :     0.001210s :     26: opt.transform.jit_opt_a
 0.28% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000027s :      4: opt.transform.jit_opt_b
 0.09% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.09% :     0.000560s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000052s :      1: py_interpret_to_execute
 0.08% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000017s :      1: remove_dup_value
 2.59% :     0.000469s :      1: renormalize.infer
 1.88% :     0.000341s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000039s :      1: rewriter_after_opt_a
 0.50% :     0.000090s :      1: rewriter_before_opt_a
 0.44% :     0.000081s :      1: symbol_engine_optimizer
37.13% :     0.006734s :      1: type_inference


TotalTime = 0.0345557, [33]
[bootstrap]: 0.00036817
[type_inference]: 0.00542734
[event_method]: 1.891e-05
[auto_monad]: 7.063e-05
[graph_reusing]: 5.91998e-06
[pre_auto_parallel]: 2.73e-06
[py_interpret_to_execute]: 4.354e-05
[rewriter_before_opt_a]: 8.045e-05
[expand_dump_flag]: 3.44001e-06
[jit_opt_a]: 0.00711897, [2]
    [Cycle 1]: 0.00257355, [27]
        [switch_simplify]: 7.081e-05
        [loop_unroll]: 4.803e-05
        [a_1]: 0.00066718
        [with_stream_mark]: 2.65e-05
        [recompute_prepare]: 7.61999e-06
        [updatestate_depend_eliminate]: 3.8e-06
        [updatestate_assign_eliminate]: 2.93998e-06
        [updatestate_loads_eliminate]: 2.98e-06
        [parameter_eliminate]: 2.99001e-06
        [specialize_transform]: 6.82002e-06
        [updatestate_useless_node_eliminater]: 5.42999e-06
        [accelerated_algorithm]: 5.64e-06
        [meta_shard_fg_expand]: 2.45002e-06
        [get_grad_eliminate_]: 5.27999e-06
        [merge_forward]: 3.58e-06
        [cell_reuse_recompute_pass]: 1.71e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.541e-05
        [j_node_and_user_rematch]: 1.043e-05
        [meta_fg_expand]: 2.39999e-06
        [replace_old_param]: 1.06e-05
        [inline_without_move]: 6.21e-06
        [renormalize]: 0.00099282
        [add_forward_monad_depend]: 8.10999e-06
        [auto_monad_grad]: 2.42001e-06
        [auto_monad_eliminator]: 1.698e-05
        [cse]: 3.086e-05
        [replace_applicator]: 1.747e-05
    [Cycle 2]: 0.00035175, [27]
        [switch_simplify]: 5.71e-06
        [loop_unroll]: 5.52999e-06
        [a_1]: 9.887e-05
        [with_stream_mark]: 1.312e-05
        [recompute_prepare]: 5.37001e-06
        [updatestate_depend_eliminate]: 3.6e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 1.57001e-06
        [specialize_transform]: 5.39e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 6.14001e-06
        [meta_shard_fg_expand]: 1.63002e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 2.96999e-06
        [cell_reuse_recompute_pass]: 3.18e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.423e-05
        [j_node_and_user_rematch]: 8.74e-06
        [meta_fg_expand]: 2.00002e-06
        [replace_old_param]: 1.471e-05
        [inline_without_move]: 4.99e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.64e-06
        [auto_monad_grad]: 1.02e-06
        [auto_monad_eliminator]: 6.02001e-06
        [cse]: 1.06e-05
        [replace_applicator]: 5.38002e-06
[py_interpret_to_execute_after_opt_a]: 1.397e-05
[rewriter_after_opt_a]: 3.924e-05
[convert_after_rewriter]: 7.48e-06
[order_py_execute_after_rewriter]: 5.00001e-06
[mutable_eliminate]: 0.00087115
[jit_opt_b]: 5.902e-05, [1]
    [Cycle 1]: 4.858e-05, [2]
        [frontend_op_eliminate]: 2.002e-05
        [inline_after_opt_a]: 1.63e-05
[cconv]: 3.454e-05
[loop_unroll]: 0.00055215
[jit_opt_after_cconv]: 0.00016063, [1]
    [Cycle 1]: 0.00015353, [11]
        [c_1]: 2.27e-05
        [parameter_eliminate]: 4.65001e-06
        [updatestate_depend_eliminate]: 7.17997e-06
        [updatestate_assign_eliminate]: 3.16001e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [cse]: 2.493e-05
        [call_graph_tuple_transform]: 2.308e-05
        [tuple_list_get_item_eliminator]: 5.61998e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 7.09988e-07
        [switch_simplify]: 5.84999e-06
[remove_dup_value]: 1.628e-05
[partial_unused_args_eliminate]: 2.78e-06
[environ_conv]: 7.27997e-06
[add_recomputation]: 5.103e-05
[cse_after_recomputation]: 2.334e-05, [1]
    [Cycle 1]: 1.699e-05, [1]
        [cse]: 1.006e-05
[auto_monad_reorder]: 1.704e-05
[get_jit_bprop_graph]: 2.53998e-06
[rewriter_after_jit_bprop_graph]: 8.70001e-06
[opt_after_jit_grad]: 0.00049427
[symbol_engine_optimizer]: 7.99e-05, [1]
    [Cycle 1]: 7.267e-05, [6]
        [build]: 4.52e-06
        [elim_shapecalc]: 8.45999e-06
        [elim_not_effective]: 1.418e-05
        [opt_reshape]: 5.91e-06
        [fold_const_symbol]: 1.015e-05
        [renormalize]: 7.2e-07
[validate]: 4.59e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.018673
[execute]: 8.15e-06
Sums
    bootstrap                            :     0.000368s :  1.26%
    type_inference                       :     0.005427s : 18.54%
    event_method                         :     0.000019s :  0.06%
    auto_monad                           :     0.000071s :  0.24%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.15%
    rewriter_before_opt_a                :     0.000080s :  0.27%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000077s :  0.26%
    jit_opt_a.loop_unroll                :     0.000054s :  0.18%
    jit_opt_a.a_1                        :     0.000766s :  2.62%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000025s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.04%
    jit_opt_a.renormalize                :     0.000993s :  3.39%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.08%
    jit_opt_a.cse                        :     0.000041s :  0.14%
    jit_opt_a.replace_applicator         :     0.000023s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.05%
    rewriter_after_opt_a                 :     0.000039s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000871s :  2.98%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.06%
    cconv                                :     0.000035s :  0.12%
    loop_unroll                          :     0.000552s :  1.89%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000051s :  0.17%
    cse_after_recomputation.cse          :     0.000010s :  0.03%
    auto_monad_reorder                   :     0.000017s :  0.06%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.03%
    opt_after_jit_grad                   :     0.000494s :  1.69%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.018673s : 63.78%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000301    23
 0.88% :     0.000003s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 2.28% :     0.000007s :      3: substitution.graph_param_transform
84.00% :     0.000253s :      4: substitution.inline
 1.45% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.72% :     0.000005s :      4: substitution.remove_not_recompute_node
 4.22% :     0.000013s :      2: substitution.replace_old_param
 5.00% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005348     2
81.09% :     0.004337s :      1: type_inference.infer
18.91% :     0.001011s :      1: type_inference.specialize
------[replace.]   0.000082     6
67.96% :     0.000056s :      4: replace.inline
32.04% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000264     6
94.73% :     0.000250s :      4: match.inline
 5.27% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   691
 1.77% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 2.28% :     0.000003s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.57% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.67% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.85% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.25% :     0.000000s :      3: predicate.elim_not_effective
 0.41% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.35% :     0.000002s :     10: predicate.environ_get_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.72% :     0.000008s :     22: predicate.inline
 1.38% :     0.000002s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.50% :     0.000002s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.54% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 5.81% :     0.000008s :     34: predicate.loop_unroll_before_grad
 2.39% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.92% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 1.79% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 1.96% :     0.000003s :     15: predicate.partial_eliminate
 1.01% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     10: predicate.reduce_eliminate
 1.28% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.46% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.42% :     0.000001s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.06% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.41% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000002s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.79% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.25% :     0.000003s :     16: predicate.switch_defer_inline
 1.99% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.18% :     0.000010s :     53: predicate.switch_simplify
 1.25% :     0.000002s :     10: predicate.tile_eliminate
 0.99% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.07% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.69% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.39% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.40% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000745    12
46.80% :     0.000348s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.83% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.37% :     0.000338s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.036679    76
 0.15% :     0.000054s :      1: add_recomputation
 0.20% :     0.000074s :      1: auto_monad
 0.05% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.05% :     0.000385s :      1: bootstrap
 0.10% :     0.000038s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.07% :     0.000024s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
19.42% :     0.007124s :      1: jit_opt_a
 0.45% :     0.000164s :      1: jit_opt_after_cconv
 0.17% :     0.000063s :      1: jit_opt_b
 1.53% :     0.000562s :      1: loop_unroll
 2.41% :     0.000882s :      1: mutable_eliminate
 2.80% :     0.001028s :     26: opt.transform.jit_opt_a
 0.15% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000029s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.37% :     0.000503s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000047s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 1.49% :     0.000548s :      1: renormalize.infer
 1.18% :     0.000434s :      1: renormalize.specialize
 0.03% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000042s :      1: rewriter_after_opt_a
 0.23% :     0.000084s :      1: rewriter_before_opt_a
 0.23% :     0.000083s :      1: symbol_engine_optimizer
50.96% :     0.018691s :      1: task_emit
14.85% :     0.005447s :      1: type_inference
 0.19% :     0.000069s :      1: validate


TotalTime = 0.0167181, [30]
[bootstrap]: 0.00032357
[type_inference]: 0.00747921
[event_method]: 2.174e-05
[auto_monad]: 9.143e-05
[graph_reusing]: 6.26e-06
[pre_auto_parallel]: 2.76e-06
[py_interpret_to_execute]: 5.444e-05
[rewriter_before_opt_a]: 8.76e-05
[expand_dump_flag]: 3.3e-06
[jit_opt_a]: 0.00615468, [2]
    [Cycle 1]: 0.00216159, [27]
        [switch_simplify]: 0.00013012
        [loop_unroll]: 3.521e-05
        [a_1]: 0.00072795
        [with_stream_mark]: 2.532e-05
        [recompute_prepare]: 7.25998e-06
        [updatestate_depend_eliminate]: 4.05e-06
        [updatestate_assign_eliminate]: 3.01999e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [parameter_eliminate]: 2.12001e-06
        [specialize_transform]: 6.34999e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.54998e-06
        [meta_shard_fg_expand]: 2.09e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 3.66999e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.551e-05
        [j_node_and_user_rematch]: 1.017e-05
        [meta_fg_expand]: 3.09999e-06
        [replace_old_param]: 9.34998e-06
        [inline_without_move]: 5.57999e-06
        [renormalize]: 0.0008183
        [add_forward_monad_depend]: 6.94001e-06
        [auto_monad_grad]: 2.71e-06
        [auto_monad_eliminator]: 1.606e-05
        [cse]: 2.822e-05
        [replace_applicator]: 7.406e-05
    [Cycle 2]: 0.0003455, [27]
        [switch_simplify]: 6.68e-06
        [loop_unroll]: 5.09e-06
        [a_1]: 9.251e-05
        [with_stream_mark]: 1.488e-05
        [recompute_prepare]: 5.97001e-06
        [updatestate_depend_eliminate]: 3.63e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 5.02999e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 4.62e-06
        [meta_shard_fg_expand]: 1.45999e-06
        [get_grad_eliminate_]: 4.39998e-06
        [merge_forward]: 3.66001e-06
        [cell_reuse_recompute_pass]: 2.10002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.541e-05
        [j_node_and_user_rematch]: 8.84e-06
        [meta_fg_expand]: 1.76e-06
        [replace_old_param]: 9.05999e-06
        [inline_without_move]: 4.69998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.82999e-06
        [auto_monad_grad]: 1.03001e-06
        [auto_monad_eliminator]: 6.61999e-06
        [cse]: 1.431e-05
        [replace_applicator]: 4.87e-06
[py_interpret_to_execute_after_opt_a]: 1.33e-05
[rewriter_after_opt_a]: 3.746e-05
[convert_after_rewriter]: 6.89999e-06
[order_py_execute_after_rewriter]: 5.38002e-06
[mutable_eliminate]: 0.00079305
[jit_opt_b]: 5.144e-05, [1]
    [Cycle 1]: 4.356e-05, [2]
        [frontend_op_eliminate]: 1.681e-05
        [inline_after_opt_a]: 1.532e-05
[cconv]: 3.071e-05
[loop_unroll]: 0.00047589
[jit_opt_after_cconv]: 0.00015293, [1]
    [Cycle 1]: 0.00014615, [11]
        [c_1]: 2.138e-05
        [parameter_eliminate]: 3.34001e-06
        [updatestate_depend_eliminate]: 7.8e-06
        [updatestate_assign_eliminate]: 2.72001e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 2.228e-05
        [call_graph_tuple_transform]: 2.179e-05
        [tuple_list_get_item_eliminator]: 5.27001e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 9.50007e-07
        [switch_simplify]: 5.57001e-06
[remove_dup_value]: 1.542e-05
[partial_unused_args_eliminate]: 2.46998e-06
[environ_conv]: 5.93998e-06
[add_recomputation]: 5.449e-05
[cse_after_recomputation]: 2.156e-05, [1]
    [Cycle 1]: 1.568e-05, [1]
        [cse]: 9.66e-06
[auto_monad_reorder]: 1.562e-05
[get_jit_bprop_graph]: 2.27001e-06
[rewriter_after_jit_bprop_graph]: 7.03998e-06
[opt_after_jit_grad]: 0.00048137
[symbol_engine_optimizer]: 7.741e-05, [1]
    [Cycle 1]: 7.08e-05, [6]
        [build]: 4.45999e-06
        [elim_shapecalc]: 8.03001e-06
        [elim_not_effective]: 1.367e-05
        [opt_reshape]: 5.97001e-06
        [fold_const_symbol]: 8.62e-06
        [renormalize]: 8.50006e-07
[validate]: 4.137e-05
Sums
    bootstrap                            :     0.000324s :  2.60%
    type_inference                       :     0.007479s : 60.21%
    event_method                         :     0.000022s :  0.18%
    auto_monad                           :     0.000091s :  0.74%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000054s :  0.44%
    rewriter_before_opt_a                :     0.000088s :  0.71%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000137s :  1.10%
    jit_opt_a.loop_unroll                :     0.000040s :  0.32%
    jit_opt_a.a_1                        :     0.000820s :  6.60%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000818s :  6.59%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.18%
    jit_opt_a.cse                        :     0.000043s :  0.34%
    jit_opt_a.replace_applicator         :     0.000079s :  0.64%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.11%
    rewriter_after_opt_a                 :     0.000037s :  0.30%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000793s :  6.38%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000031s :  0.25%
    loop_unroll                          :     0.000476s :  3.83%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.17%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000015s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000054s :  0.44%
    cse_after_recomputation.cse          :     0.000010s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000481s :  3.87%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.33%


Time group info:
------[substitution.]   0.000315    26
 0.63% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000002s :      2: substitution.fold_const_symbol
 2.01% :     0.000006s :      3: substitution.graph_param_transform
85.58% :     0.000270s :      6: substitution.inline
 1.21% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.55% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.69% :     0.000005s :      2: substitution.replace_old_param
 2.77% :     0.000009s :      1: substitution.switch_simplify
 4.02% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007400     2
84.53% :     0.006256s :      1: type_inference.infer
15.47% :     0.001145s :      1: type_inference.specialize
------[replace.]   0.000146     9
49.59% :     0.000072s :      6: replace.inline
28.15% :     0.000041s :      1: replace.switch_simplify
22.26% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000286     9
93.14% :     0.000266s :      6: match.inline
 2.79% :     0.000008s :      1: match.switch_simplify
 4.07% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   801
 1.31% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.61% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     12: predicate.addn_check_dump
 1.56% :     0.000002s :     12: predicate.addn_zero_filter
 1.96% :     0.000003s :     12: predicate.arithmetic_simplify
 1.26% :     0.000002s :     12: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     12: predicate.compare_switch_simplify
 1.32% :     0.000002s :     12: predicate.depend_value_elim
 1.11% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.10% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.13% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     12: predicate.environ_get_eliminate
 1.35% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.74% :     0.000008s :     26: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.07% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     17: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.86% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000002s :     12: predicate.merge_addn
 1.20% :     0.000002s :     12: predicate.minmaximum_grad
 1.64% :     0.000002s :      3: predicate.mutable_eliminate
 0.58% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     17: predicate.partial_eliminate
 1.28% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     12: predicate.reduce_eliminate
 1.41% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     20: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.39% :     0.000001s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     12: predicate.reshape_eliminate
 1.28% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     12: predicate.same_eliminate
 0.63% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.65% :     0.000004s :     20: predicate.switch_defer_inline
 2.19% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.27% :     0.000012s :     65: predicate.switch_simplify
 1.20% :     0.000002s :     12: predicate.tile_eliminate
 1.17% :     0.000002s :     12: predicate.transpose_eliminate
 1.42% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.13% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.61% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     12: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000883    15
48.23% :     0.000426s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.96% :     0.000114s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.81% :     0.000343s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018756    72
 0.31% :     0.000057s :      1: add_recomputation
 0.51% :     0.000095s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 1.83% :     0.000343s :      1: bootstrap
 0.18% :     0.000034s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.14% :     0.000027s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
32.84% :     0.006159s :      1: jit_opt_a
 0.83% :     0.000156s :      1: jit_opt_after_cconv
 0.29% :     0.000054s :      1: jit_opt_b
 2.58% :     0.000485s :      1: loop_unroll
 4.29% :     0.000804s :      1: mutable_eliminate
 6.28% :     0.001179s :     26: opt.transform.jit_opt_a
 0.27% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.62% :     0.000491s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000057s :      1: py_interpret_to_execute
 0.08% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000018s :      1: remove_dup_value
 2.52% :     0.000472s :      1: renormalize.infer
 1.80% :     0.000337s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000040s :      1: rewriter_after_opt_a
 0.48% :     0.000090s :      1: rewriter_before_opt_a
 0.43% :     0.000080s :      1: symbol_engine_optimizer
39.97% :     0.007497s :      1: type_inference


TotalTime = 0.027655, [33]
[bootstrap]: 0.00037246
[type_inference]: 0.00886603
[event_method]: 7.309e-05
[auto_monad]: 0.0001012
[graph_reusing]: 7.1e-06
[pre_auto_parallel]: 2.84999e-06
[py_interpret_to_execute]: 5.257e-05
[rewriter_before_opt_a]: 8.635e-05
[expand_dump_flag]: 3.85e-06
[jit_opt_a]: 0.00715276, [2]
    [Cycle 1]: 0.00219394, [27]
        [switch_simplify]: 7.164e-05
        [loop_unroll]: 3.028e-05
        [a_1]: 0.00065385
        [with_stream_mark]: 2.266e-05
        [recompute_prepare]: 8.53001e-06
        [updatestate_depend_eliminate]: 3.7e-06
        [updatestate_assign_eliminate]: 3.21999e-06
        [updatestate_loads_eliminate]: 2.62001e-06
        [parameter_eliminate]: 2.01e-06
        [specialize_transform]: 6.31e-06
        [updatestate_useless_node_eliminater]: 5.57999e-06
        [accelerated_algorithm]: 5.94e-06
        [meta_shard_fg_expand]: 2.50002e-06
        [get_grad_eliminate_]: 5.89e-06
        [merge_forward]: 3.6e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.543e-05
        [j_node_and_user_rematch]: 1.056e-05
        [meta_fg_expand]: 2.42001e-06
        [replace_old_param]: 1.011e-05
        [inline_without_move]: 6.00002e-06
        [renormalize]: 0.00105497
        [add_forward_monad_depend]: 8.2e-06
        [auto_monad_grad]: 2.82002e-06
        [auto_monad_eliminator]: 1.732e-05
        [cse]: 2.276e-05
        [replace_applicator]: 1.663e-05
    [Cycle 2]: 0.00036459, [27]
        [switch_simplify]: 6.69999e-06
        [loop_unroll]: 5.22e-06
        [a_1]: 0.00010261
        [with_stream_mark]: 1.364e-05
        [recompute_prepare]: 5.76998e-06
        [updatestate_depend_eliminate]: 3.41999e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 1.45001e-06
        [specialize_transform]: 4.82e-06
        [updatestate_useless_node_eliminater]: 4.71997e-06
        [accelerated_algorithm]: 6.06998e-06
        [meta_shard_fg_expand]: 2.19001e-06
        [get_grad_eliminate_]: 4.61002e-06
        [merge_forward]: 3.23998e-06
        [cell_reuse_recompute_pass]: 2.65002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.671e-05
        [j_node_and_user_rematch]: 8.50001e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 1.671e-05
        [inline_without_move]: 4.62998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.49998e-06
        [auto_monad_grad]: 8.59989e-07
        [auto_monad_eliminator]: 6.83998e-06
        [cse]: 1.201e-05
        [replace_applicator]: 5.81003e-06
[py_interpret_to_execute_after_opt_a]: 1.885e-05
[rewriter_after_opt_a]: 3.835e-05
[convert_after_rewriter]: 7.32002e-06
[order_py_execute_after_rewriter]: 5.66e-06
[mutable_eliminate]: 0.00122523
[jit_opt_b]: 5.859e-05, [1]
    [Cycle 1]: 4.93e-05, [2]
        [frontend_op_eliminate]: 1.894e-05
        [inline_after_opt_a]: 1.755e-05
[cconv]: 3.593e-05
[loop_unroll]: 0.00052932
[jit_opt_after_cconv]: 0.00017482, [1]
    [Cycle 1]: 0.00016744, [11]
        [c_1]: 2.466e-05
        [parameter_eliminate]: 4.28999e-06
        [updatestate_depend_eliminate]: 1.008e-05
        [updatestate_assign_eliminate]: 3.13998e-06
        [updatestate_loads_eliminate]: 2.43998e-06
        [cse]: 3.071e-05
        [call_graph_tuple_transform]: 2.46e-05
        [tuple_list_get_item_eliminator]: 5.89e-06
        [none_parameter_eliminate]: 2.20002e-06
        [renormalize]: 8.00006e-07
        [switch_simplify]: 5.64e-06
[remove_dup_value]: 1.558e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 6.58e-06
[add_recomputation]: 5.668e-05
[cse_after_recomputation]: 2.188e-05, [1]
    [Cycle 1]: 1.543e-05, [1]
        [cse]: 9.49e-06
[auto_monad_reorder]: 1.635e-05
[get_jit_bprop_graph]: 1.98002e-06
[rewriter_after_jit_bprop_graph]: 8.23999e-06
[opt_after_jit_grad]: 0.00054242
[symbol_engine_optimizer]: 8.638e-05, [1]
    [Cycle 1]: 7.796e-05, [6]
        [build]: 5.50001e-06
        [elim_shapecalc]: 8.70999e-06
        [elim_not_effective]: 1.676e-05
        [opt_reshape]: 6.59999e-06
        [fold_const_symbol]: 9.51e-06
        [renormalize]: 9.49978e-07
[validate]: 4.41e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.00778382
[execute]: 9.66e-06
Sums
    bootstrap                            :     0.000372s :  1.67%
    type_inference                       :     0.008866s : 39.64%
    event_method                         :     0.000073s :  0.33%
    auto_monad                           :     0.000101s :  0.45%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000053s :  0.24%
    rewriter_before_opt_a                :     0.000086s :  0.39%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000078s :  0.35%
    jit_opt_a.loop_unroll                :     0.000035s :  0.16%
    jit_opt_a.a_1                        :     0.000756s :  3.38%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000027s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.001055s :  4.72%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.11%
    jit_opt_a.cse                        :     0.000035s :  0.16%
    jit_opt_a.replace_applicator         :     0.000022s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.17%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.001225s :  5.48%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.08%
    cconv                                :     0.000036s :  0.16%
    loop_unroll                          :     0.000529s :  2.37%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000031s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000057s :  0.25%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000542s :  2.43%
    symbol_engine_optimizer.build        :     0.000006s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.20%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007784s : 34.80%
    execute                              :     0.000010s :  0.04%


Time group info:
------[substitution.]   0.000287    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000002s :      2: substitution.fold_const_symbol
 2.26% :     0.000006s :      3: substitution.graph_param_transform
86.11% :     0.000247s :      4: substitution.inline
 1.65% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.75% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.72% :     0.000005s :      2: substitution.replace_old_param
 5.19% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008767     2
87.04% :     0.007631s :      1: type_inference.infer
12.96% :     0.001136s :      1: type_inference.specialize
------[replace.]   0.000087     6
68.96% :     0.000060s :      4: replace.inline
31.04% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000258     6
94.68% :     0.000244s :      4: match.inline
 5.32% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.66% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.91% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000002s :     10: predicate.addn_zero_filter
 2.56% :     0.000003s :     10: predicate.arithmetic_simplify
 1.64% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.92% :     0.000001s :     10: predicate.compare_switch_simplify
 1.19% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.43% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     10: predicate.environ_get_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.35% :     0.000002s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 6.46% :     0.000009s :     22: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      6: predicate.less_batch_normalization
 2.06% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.61% :     0.000002s :     15: predicate.load_eliminater
 1.84% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.64% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.22% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 0.92% :     0.000001s :     10: predicate.minmaximum_grad
 2.56% :     0.000003s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 1.97% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.35% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.47% :     0.000001s :      3: predicate.reset_defer_inline
 1.63% :     0.000002s :     10: predicate.reshape_eliminate
 1.01% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.04% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.07% :     0.000001s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.01% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.92% :     0.000009s :     53: predicate.switch_simplify
 1.32% :     0.000002s :     10: predicate.tile_eliminate
 0.96% :     0.000001s :     10: predicate.transpose_eliminate
 1.52% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.89% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.44% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.93% :     0.000003s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000810    12
46.75% :     0.000379s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.88% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.37% :     0.000376s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.029838    76
 0.20% :     0.000059s :      1: add_recomputation
 0.36% :     0.000107s :      1: auto_monad
 0.06% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.30% :     0.000388s :      1: bootstrap
 0.13% :     0.000039s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000011s :      1: environ_conv
 0.27% :     0.000079s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
23.99% :     0.007157s :      1: jit_opt_a
 0.59% :     0.000177s :      1: jit_opt_after_cconv
 0.20% :     0.000061s :      1: jit_opt_b
 1.81% :     0.000539s :      1: loop_unroll
 4.15% :     0.001238s :      1: mutable_eliminate
 3.37% :     0.001007s :     26: opt.transform.jit_opt_a
 0.19% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000029s :      4: opt.transform.jit_opt_b
 0.05% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.86% :     0.000554s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000056s :      1: py_interpret_to_execute
 0.07% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000018s :      1: remove_dup_value
 1.97% :     0.000589s :      1: renormalize.infer
 1.53% :     0.000455s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000042s :      1: rewriter_after_opt_a
 0.30% :     0.000090s :      1: rewriter_before_opt_a
 0.30% :     0.000089s :      1: symbol_engine_optimizer
26.15% :     0.007802s :      1: task_emit
29.78% :     0.008887s :      1: type_inference
 0.22% :     0.000066s :      1: validate


TotalTime = 0.0267406, [33]
[bootstrap]: 0.00047139
[type_inference]: 0.00799971
[event_method]: 2.445e-05
[auto_monad]: 7.665e-05
[graph_reusing]: 6.71e-06
[pre_auto_parallel]: 2.93e-06
[py_interpret_to_execute]: 4.778e-05
[rewriter_before_opt_a]: 7.885e-05
[expand_dump_flag]: 3.7e-06
[jit_opt_a]: 0.00775356, [2]
    [Cycle 1]: 0.00236969, [27]
        [switch_simplify]: 9.091e-05
        [loop_unroll]: 3.106e-05
        [a_1]: 0.00073319
        [with_stream_mark]: 2.761e-05
        [recompute_prepare]: 8.33999e-06
        [updatestate_depend_eliminate]: 3.96001e-06
        [updatestate_assign_eliminate]: 3.45998e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 2.47001e-06
        [specialize_transform]: 6.84999e-06
        [updatestate_useless_node_eliminater]: 5.54e-06
        [accelerated_algorithm]: 6.56e-06
        [meta_shard_fg_expand]: 3.06999e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 4.07998e-06
        [cell_reuse_recompute_pass]: 1.76998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.611e-05
        [j_node_and_user_rematch]: 1.056e-05
        [meta_fg_expand]: 2.46e-06
        [replace_old_param]: 1.077e-05
        [inline_without_move]: 5.23002e-06
        [renormalize]: 0.00110181
        [add_forward_monad_depend]: 7.4e-06
        [auto_monad_grad]: 3.11001e-06
        [auto_monad_eliminator]: 1.655e-05
        [cse]: 3.659e-05
        [replace_applicator]: 1.484e-05
    [Cycle 2]: 0.00034356, [27]
        [switch_simplify]: 5.67999e-06
        [loop_unroll]: 5.46e-06
        [a_1]: 9.34e-05
        [with_stream_mark]: 1.327e-05
        [recompute_prepare]: 5.71e-06
        [updatestate_depend_eliminate]: 2.82002e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 1.81003e-06
        [parameter_eliminate]: 1.06002e-06
        [specialize_transform]: 4.58999e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 5.15999e-06
        [meta_shard_fg_expand]: 1.53002e-06
        [get_grad_eliminate_]: 4.71002e-06
        [merge_forward]: 2.48998e-06
        [cell_reuse_recompute_pass]: 4.68001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.589e-05
        [j_node_and_user_rematch]: 7.82998e-06
        [meta_fg_expand]: 1.76e-06
        [replace_old_param]: 8.32e-06
        [inline_without_move]: 4.79e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.38002e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 5.85002e-06
        [cse]: 1.171e-05
        [replace_applicator]: 5.04e-06
[py_interpret_to_execute_after_opt_a]: 1.651e-05
[rewriter_after_opt_a]: 4.057e-05
[convert_after_rewriter]: 6.91999e-06
[order_py_execute_after_rewriter]: 8.75999e-06
[mutable_eliminate]: 0.00099499
[jit_opt_b]: 5.723e-05, [1]
    [Cycle 1]: 4.743e-05, [2]
        [frontend_op_eliminate]: 1.892e-05
        [inline_after_opt_a]: 1.587e-05
[cconv]: 3.356e-05
[loop_unroll]: 0.00049384
[jit_opt_after_cconv]: 0.00015704, [1]
    [Cycle 1]: 0.00015095, [11]
        [c_1]: 2.125e-05
        [parameter_eliminate]: 3.81999e-06
        [updatestate_depend_eliminate]: 8.33999e-06
        [updatestate_assign_eliminate]: 2.69999e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [cse]: 2.671e-05
        [call_graph_tuple_transform]: 2.138e-05
        [tuple_list_get_item_eliminator]: 5.25999e-06
        [none_parameter_eliminate]: 1.70001e-06
        [renormalize]: 8.30012e-07
        [switch_simplify]: 5.25999e-06
[remove_dup_value]: 1.49e-05
[partial_unused_args_eliminate]: 2.54999e-06
[environ_conv]: 6.45002e-06
[add_recomputation]: 6.071e-05
[cse_after_recomputation]: 2.296e-05, [1]
    [Cycle 1]: 1.598e-05, [1]
        [cse]: 9.46e-06
[auto_monad_reorder]: 1.667e-05
[get_jit_bprop_graph]: 2.27999e-06
[rewriter_after_jit_bprop_graph]: 7.9e-06
[opt_after_jit_grad]: 0.0005233
[symbol_engine_optimizer]: 8.148e-05, [1]
    [Cycle 1]: 7.338e-05, [6]
        [build]: 5.51e-06
        [elim_shapecalc]: 8.07e-06
        [elim_not_effective]: 1.463e-05
        [opt_reshape]: 5.46e-06
        [fold_const_symbol]: 8.79e-06
        [renormalize]: 1.13001e-06
[validate]: 5.02e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00739236
[execute]: 1.032e-05
Sums
    bootstrap                            :     0.000471s :  2.25%
    type_inference                       :     0.008000s : 38.15%
    event_method                         :     0.000024s :  0.12%
    auto_monad                           :     0.000077s :  0.37%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000048s :  0.23%
    rewriter_before_opt_a                :     0.000079s :  0.38%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000097s :  0.46%
    jit_opt_a.loop_unroll                :     0.000037s :  0.17%
    jit_opt_a.a_1                        :     0.000827s :  3.94%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.001102s :  5.25%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.11%
    jit_opt_a.cse                        :     0.000048s :  0.23%
    jit_opt_a.replace_applicator         :     0.000020s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.08%
    rewriter_after_opt_a                 :     0.000041s :  0.19%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000009s :  0.04%
    mutable_eliminate                    :     0.000995s :  4.75%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.000034s :  0.16%
    loop_unroll                          :     0.000494s :  2.36%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.07%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000061s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000523s :  2.50%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000050s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007392s : 35.25%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000349    23
 0.65% :     0.000002s :      2: substitution.elim_not_effective
 0.41% :     0.000001s :      2: substitution.fold_const_symbol
 1.74% :     0.000006s :      3: substitution.graph_param_transform
88.54% :     0.000309s :      4: substitution.inline
 1.10% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.64% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.81% :     0.000006s :      2: substitution.replace_old_param
 4.11% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007894     2
80.28% :     0.006337s :      1: type_inference.infer
19.72% :     0.001557s :      1: type_inference.specialize
------[replace.]   0.000096     6
68.85% :     0.000066s :      4: replace.inline
31.15% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000320     6
95.87% :     0.000306s :      4: match.inline
 4.13% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.32% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.92% :     0.000001s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 2.20% :     0.000003s :     10: predicate.arithmetic_simplify
 0.98% :     0.000001s :     10: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.91% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 0.98% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.47% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.50% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     10: predicate.environ_get_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.77% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000001s :      3: predicate.graph_param_transform
 5.68% :     0.000008s :     22: predicate.inline
 0.75% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.54% :     0.000002s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.54% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.22% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.07% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 0.91% :     0.000001s :     10: predicate.minmaximum_grad
 2.49% :     0.000003s :      3: predicate.mutable_eliminate
 0.39% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 0.99% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.81% :     0.000002s :     10: predicate.reduce_eliminate
 1.24% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.49% :     0.000001s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.01% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.31% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.72% :     0.000002s :     10: predicate.same_eliminate
 0.66% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.34% :     0.000002s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.88% :     0.000010s :     53: predicate.switch_simplify
 1.37% :     0.000002s :     10: predicate.tile_eliminate
 1.00% :     0.000001s :     10: predicate.transpose_eliminate
 2.01% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.60% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.51% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001116    12
45.17% :     0.000504s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.01% :     0.000067s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.82% :     0.000545s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.029025    76
 0.22% :     0.000064s :      1: add_recomputation
 0.28% :     0.000080s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.71% :     0.000498s :      1: bootstrap
 0.13% :     0.000037s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.10% :     0.000030s :      1: event_method
 0.05% :     0.000016s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
26.73% :     0.007759s :      1: jit_opt_a
 0.55% :     0.000160s :      1: jit_opt_after_cconv
 0.21% :     0.000060s :      1: jit_opt_b
 1.73% :     0.000503s :      1: loop_unroll
 3.48% :     0.001011s :      1: mutable_eliminate
 3.74% :     0.001086s :     26: opt.transform.jit_opt_a
 0.17% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000028s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.84% :     0.000533s :      1: opt_after_jit_grad
 0.04% :     0.000011s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000051s :      1: py_interpret_to_execute
 0.07% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 2.19% :     0.000636s :      1: renormalize.infer
 1.56% :     0.000453s :      1: renormalize.specialize
 0.03% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000044s :      1: rewriter_after_opt_a
 0.28% :     0.000082s :      1: rewriter_before_opt_a
 0.29% :     0.000084s :      1: symbol_engine_optimizer
25.54% :     0.007414s :      1: task_emit
27.66% :     0.008030s :      1: type_inference
 0.27% :     0.000078s :      1: validate


TotalTime = 0.0165156, [30]
[bootstrap]: 0.00030829
[type_inference]: 0.00742745
[event_method]: 2.347e-05
[auto_monad]: 8.477e-05
[graph_reusing]: 6.25002e-06
[pre_auto_parallel]: 2.83e-06
[py_interpret_to_execute]: 5.102e-05
[rewriter_before_opt_a]: 8.921e-05
[expand_dump_flag]: 3.54002e-06
[jit_opt_a]: 0.0062659, [2]
    [Cycle 1]: 0.0021799, [27]
        [switch_simplify]: 0.00013825
        [loop_unroll]: 3.609e-05
        [a_1]: 0.00081961
        [with_stream_mark]: 2.498e-05
        [recompute_prepare]: 9.54e-06
        [updatestate_depend_eliminate]: 4.22e-06
        [updatestate_assign_eliminate]: 3.71999e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 2.53998e-06
        [specialize_transform]: 6.51e-06
        [updatestate_useless_node_eliminater]: 4.84998e-06
        [accelerated_algorithm]: 6.01e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 5.19e-06
        [merge_forward]: 4.05998e-06
        [cell_reuse_recompute_pass]: 1.89e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.603e-05
        [j_node_and_user_rematch]: 1.041e-05
        [meta_fg_expand]: 2.21e-06
        [replace_old_param]: 1.008e-05
        [inline_without_move]: 5.39998e-06
        [renormalize]: 0.00076722
        [add_forward_monad_depend]: 5.99999e-06
        [auto_monad_grad]: 2.36e-06
        [auto_monad_eliminator]: 1.776e-05
        [cse]: 3.229e-05
        [replace_applicator]: 1.608e-05
    [Cycle 2]: 0.00036029, [27]
        [switch_simplify]: 6.21998e-06
        [loop_unroll]: 5.35001e-06
        [a_1]: 9.454e-05
        [with_stream_mark]: 1.36e-05
        [recompute_prepare]: 6.44999e-06
        [updatestate_depend_eliminate]: 2.79001e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 4.80001e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.50001e-06
        [meta_shard_fg_expand]: 1.46998e-06
        [get_grad_eliminate_]: 5.99999e-06
        [merge_forward]: 3.47002e-06
        [cell_reuse_recompute_pass]: 1.80001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.595e-05
        [j_node_and_user_rematch]: 9.44e-06
        [meta_fg_expand]: 1.66002e-06
        [replace_old_param]: 8.80001e-06
        [inline_without_move]: 4.58001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.91e-06
        [auto_monad_grad]: 1.17e-06
        [auto_monad_eliminator]: 7.88001e-06
        [cse]: 1.347e-05
        [replace_applicator]: 6.24001e-06
[py_interpret_to_execute_after_opt_a]: 1.229e-05
[rewriter_after_opt_a]: 3.939e-05
[convert_after_rewriter]: 8.22e-06
[order_py_execute_after_rewriter]: 5.32999e-06
[mutable_eliminate]: 0.00065568
[jit_opt_b]: 5.121e-05, [1]
    [Cycle 1]: 4.277e-05, [2]
        [frontend_op_eliminate]: 1.568e-05
        [inline_after_opt_a]: 1.485e-05
[cconv]: 2.772e-05
[loop_unroll]: 0.00042476
[jit_opt_after_cconv]: 0.00014371, [1]
    [Cycle 1]: 0.00013737, [11]
        [c_1]: 2.228e-05
        [parameter_eliminate]: 3.08998e-06
        [updatestate_depend_eliminate]: 6.07001e-06
        [updatestate_assign_eliminate]: 2.54001e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [cse]: 1.853e-05
        [call_graph_tuple_transform]: 2.001e-05
        [tuple_list_get_item_eliminator]: 5.12e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 6.60017e-07
        [switch_simplify]: 5.23002e-06
[remove_dup_value]: 1.37e-05
[partial_unused_args_eliminate]: 2.64999e-06
[environ_conv]: 5.48002e-06
[add_recomputation]: 4.788e-05
[cse_after_recomputation]: 2.095e-05, [1]
    [Cycle 1]: 1.523e-05, [1]
        [cse]: 9.24e-06
[auto_monad_reorder]: 1.604e-05
[get_jit_bprop_graph]: 2.19999e-06
[rewriter_after_jit_bprop_graph]: 5.97001e-06
[opt_after_jit_grad]: 0.00045824
[symbol_engine_optimizer]: 7.743e-05, [1]
    [Cycle 1]: 7.107e-05, [6]
        [build]: 4.58999e-06
        [elim_shapecalc]: 7.46999e-06
        [elim_not_effective]: 1.375e-05
        [opt_reshape]: 5.62001e-06
        [fold_const_symbol]: 9.51e-06
        [renormalize]: 5.50004e-07
[validate]: 3.954e-05
Sums
    bootstrap                            :     0.000308s :  2.54%
    type_inference                       :     0.007427s : 61.28%
    event_method                         :     0.000023s :  0.19%
    auto_monad                           :     0.000085s :  0.70%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000051s :  0.42%
    rewriter_before_opt_a                :     0.000089s :  0.74%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000144s :  1.19%
    jit_opt_a.loop_unroll                :     0.000041s :  0.34%
    jit_opt_a.a_1                        :     0.000914s :  7.54%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.09%
    jit_opt_a.merge_forward              :     0.000008s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000767s :  6.33%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.21%
    jit_opt_a.cse                        :     0.000046s :  0.38%
    jit_opt_a.replace_applicator         :     0.000022s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.10%
    rewriter_after_opt_a                 :     0.000039s :  0.32%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000656s :  5.41%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000028s :  0.23%
    loop_unroll                          :     0.000425s :  3.50%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.11%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000048s :  0.40%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000458s :  3.78%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.33%


Time group info:
------[substitution.]   0.000382    26
 0.55% :     0.000002s :      2: substitution.elim_not_effective
 0.35% :     0.000001s :      2: substitution.fold_const_symbol
 1.76% :     0.000007s :      3: substitution.graph_param_transform
87.40% :     0.000334s :      6: substitution.inline
 1.14% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.26% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.48% :     0.000006s :      2: substitution.replace_old_param
 2.31% :     0.000009s :      1: substitution.switch_simplify
 3.74% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007347     2
84.58% :     0.006214s :      1: type_inference.infer
15.42% :     0.001133s :      1: type_inference.specialize
------[replace.]   0.000153     9
50.46% :     0.000077s :      6: replace.inline
28.58% :     0.000044s :      1: replace.switch_simplify
20.96% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000352     9
93.96% :     0.000330s :      6: match.inline
 2.30% :     0.000008s :      1: match.switch_simplify
 3.74% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000140   801
 1.34% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.90% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000002s :     12: predicate.addn_check_dump
 1.63% :     0.000002s :     12: predicate.addn_zero_filter
 2.25% :     0.000003s :     12: predicate.arithmetic_simplify
 1.25% :     0.000002s :     12: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000002s :     12: predicate.compare_switch_simplify
 1.32% :     0.000002s :     12: predicate.depend_value_elim
 1.13% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.44% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.11% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.12% :     0.000002s :     12: predicate.environ_get_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.04% :     0.000007s :     26: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.76% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.81% :     0.000003s :     17: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.31% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.22% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     12: predicate.merge_addn
 1.10% :     0.000002s :     12: predicate.minmaximum_grad
 1.51% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     17: predicate.partial_eliminate
 1.10% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     12: predicate.reduce_eliminate
 1.72% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.53% :     0.000002s :     20: predicate.replace_applicator
 1.03% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000002s :     12: predicate.reshape_eliminate
 1.20% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.58% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     12: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.75% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.90% :     0.000005s :     20: predicate.switch_defer_inline
 2.20% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.87% :     0.000012s :     65: predicate.switch_simplify
 1.20% :     0.000002s :     12: predicate.tile_eliminate
 1.22% :     0.000002s :     12: predicate.transpose_eliminate
 1.47% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.87% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     12: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000863    15
48.99% :     0.000423s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.13% :     0.000113s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.88% :     0.000327s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018560    72
 0.27% :     0.000051s :      1: add_recomputation
 0.48% :     0.000089s :      1: auto_monad
 0.10% :     0.000019s :      1: auto_monad_reorder
 1.78% :     0.000331s :      1: bootstrap
 0.16% :     0.000030s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.15% :     0.000028s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.78% :     0.006269s :      1: jit_opt_a
 0.79% :     0.000147s :      1: jit_opt_after_cconv
 0.29% :     0.000054s :      1: jit_opt_b
 2.33% :     0.000432s :      1: loop_unroll
 3.59% :     0.000666s :      1: mutable_eliminate
 6.63% :     0.001231s :     26: opt.transform.jit_opt_a
 0.26% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.51% :     0.000466s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000055s :      1: py_interpret_to_execute
 0.09% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 2.36% :     0.000438s :      1: renormalize.infer
 1.72% :     0.000320s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000043s :      1: rewriter_after_opt_a
 0.51% :     0.000094s :      1: rewriter_before_opt_a
 0.43% :     0.000080s :      1: symbol_engine_optimizer
40.13% :     0.007447s :      1: type_inference


TotalTime = 0.0239655, [33]
[bootstrap]: 0.00042633
[type_inference]: 0.00689154
[event_method]: 2.553e-05
[auto_monad]: 8.007e-05
[graph_reusing]: 6.64001e-06
[pre_auto_parallel]: 3.26999e-06
[py_interpret_to_execute]: 5.092e-05
[rewriter_before_opt_a]: 7.916e-05
[expand_dump_flag]: 3.31999e-06
[jit_opt_a]: 0.00626858, [2]
    [Cycle 1]: 0.00179592, [27]
        [switch_simplify]: 5.851e-05
        [loop_unroll]: 2.943e-05
        [a_1]: 0.00056635
        [with_stream_mark]: 1.874e-05
        [recompute_prepare]: 7.08998e-06
        [updatestate_depend_eliminate]: 4.01001e-06
        [updatestate_assign_eliminate]: 3.17002e-06
        [updatestate_loads_eliminate]: 2.59001e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 6.11e-06
        [updatestate_useless_node_eliminater]: 5.14998e-06
        [accelerated_algorithm]: 5.52001e-06
        [meta_shard_fg_expand]: 2.14e-06
        [get_grad_eliminate_]: 5.22999e-06
        [merge_forward]: 3.85e-06
        [cell_reuse_recompute_pass]: 1.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.439e-05
        [j_node_and_user_rematch]: 1.003e-05
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 9.37001e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 0.00078518
        [add_forward_monad_depend]: 6.30002e-06
        [auto_monad_grad]: 1.74e-06
        [auto_monad_eliminator]: 1.504e-05
        [cse]: 3.093e-05
        [replace_applicator]: 1.433e-05
    [Cycle 2]: 0.0003519, [27]
        [switch_simplify]: 6.14999e-06
        [loop_unroll]: 5.89e-06
        [a_1]: 9.615e-05
        [with_stream_mark]: 1.059e-05
        [recompute_prepare]: 5.82001e-06
        [updatestate_depend_eliminate]: 3.72998e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 5.57999e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 6.19999e-06
        [meta_shard_fg_expand]: 1.40001e-06
        [get_grad_eliminate_]: 5.21002e-06
        [merge_forward]: 2.91e-06
        [cell_reuse_recompute_pass]: 1.72001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.43e-05
        [j_node_and_user_rematch]: 9.64e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 1.568e-05
        [inline_without_move]: 5.57001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.72001e-06
        [auto_monad_grad]: 1.15999e-06
        [auto_monad_eliminator]: 5.21998e-06
        [cse]: 1.106e-05
        [replace_applicator]: 5.46e-06
[py_interpret_to_execute_after_opt_a]: 1.131e-05
[rewriter_after_opt_a]: 4.043e-05
[convert_after_rewriter]: 7.62998e-06
[order_py_execute_after_rewriter]: 5.20001e-06
[mutable_eliminate]: 0.00077451
[jit_opt_b]: 5.501e-05, [1]
    [Cycle 1]: 4.691e-05, [2]
        [frontend_op_eliminate]: 1.831e-05
        [inline_after_opt_a]: 1.643e-05
[cconv]: 3.119e-05
[loop_unroll]: 0.00054738
[jit_opt_after_cconv]: 0.00019243, [1]
    [Cycle 1]: 0.00018402, [11]
        [c_1]: 2.398e-05
        [parameter_eliminate]: 5.81e-06
        [updatestate_depend_eliminate]: 9.42999e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 2.30002e-06
        [cse]: 3.088e-05
        [call_graph_tuple_transform]: 2.854e-05
        [tuple_list_get_item_eliminator]: 6.68e-06
        [none_parameter_eliminate]: 2.39001e-06
        [renormalize]: 1.14003e-06
        [switch_simplify]: 6.28e-06
[remove_dup_value]: 1.61e-05
[partial_unused_args_eliminate]: 2.51e-06
[environ_conv]: 7e-06
[add_recomputation]: 6.252e-05
[cse_after_recomputation]: 2.682e-05, [1]
    [Cycle 1]: 1.936e-05, [1]
        [cse]: 1.195e-05
[auto_monad_reorder]: 1.781e-05
[get_jit_bprop_graph]: 2.19001e-06
[rewriter_after_jit_bprop_graph]: 9.20999e-06
[opt_after_jit_grad]: 0.00065697
[symbol_engine_optimizer]: 9.418e-05, [1]
    [Cycle 1]: 8.381e-05, [6]
        [build]: 5.84999e-06
        [elim_shapecalc]: 9.92999e-06
        [elim_not_effective]: 1.651e-05
        [opt_reshape]: 7.03e-06
        [fold_const_symbol]: 1.032e-05
        [renormalize]: 5.8001e-07
[validate]: 4.761e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.0072262
[execute]: 7.61001e-06
Sums
    bootstrap                            :     0.000426s :  2.23%
    type_inference                       :     0.006892s : 36.07%
    event_method                         :     0.000026s :  0.13%
    auto_monad                           :     0.000080s :  0.42%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000051s :  0.27%
    rewriter_before_opt_a                :     0.000079s :  0.41%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.34%
    jit_opt_a.loop_unroll                :     0.000035s :  0.18%
    jit_opt_a.a_1                        :     0.000663s :  3.47%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000025s :  0.13%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000785s :  4.11%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.11%
    jit_opt_a.cse                        :     0.000042s :  0.22%
    jit_opt_a.replace_applicator         :     0.000020s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000040s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000775s :  4.05%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000031s :  0.16%
    loop_unroll                          :     0.000547s :  2.86%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000031s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000029s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000063s :  0.33%
    cse_after_recomputation.cse          :     0.000012s :  0.06%
    auto_monad_reorder                   :     0.000018s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000657s :  3.44%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000048s :  0.25%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007226s : 37.82%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000233    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.81% :     0.000002s :      2: substitution.fold_const_symbol
 2.85% :     0.000007s :      3: substitution.graph_param_transform
80.33% :     0.000188s :      4: substitution.inline
 1.90% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.18% :     0.000005s :      4: substitution.remove_not_recompute_node
 5.22% :     0.000012s :      2: substitution.replace_old_param
 5.70% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005820     2
83.03% :     0.004832s :      1: type_inference.infer
16.97% :     0.000988s :      1: type_inference.specialize
------[replace.]   0.000076     6
68.32% :     0.000052s :      4: replace.inline
31.68% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000197     6
93.80% :     0.000185s :      4: match.inline
 6.20% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.39% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.21% :     0.000002s :     10: predicate.addn_zero_filter
 2.09% :     0.000003s :     10: predicate.arithmetic_simplify
 1.09% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.93% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000002s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.28% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.08% :     0.000001s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.90% :     0.000008s :     22: predicate.inline
 1.13% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.91% :     0.000002s :     15: predicate.load_eliminater
 1.74% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.92% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.35% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 1.16% :     0.000001s :     10: predicate.minmaximum_grad
 1.92% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.32% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.42% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.42% :     0.000002s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.60% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.10% :     0.000003s :     16: predicate.switch_defer_inline
 1.84% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.14% :     0.000009s :     53: predicate.switch_simplify
 1.47% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000002s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.91% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.56% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.96% :     0.000003s :     10: predicate.value_based_eliminate
 0.43% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000725    12
52.67% :     0.000382s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.39% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.94% :     0.000311s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025769    76
 0.26% :     0.000066s :      1: add_recomputation
 0.32% :     0.000083s :      1: auto_monad
 0.08% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.76% :     0.000453s :      1: bootstrap
 0.13% :     0.000034s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000030s :      1: cse_after_recomputation
 0.04% :     0.000011s :      1: environ_conv
 0.12% :     0.000031s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
24.34% :     0.006273s :      1: jit_opt_a
 0.76% :     0.000196s :      1: jit_opt_after_cconv
 0.23% :     0.000059s :      1: jit_opt_b
 2.19% :     0.000563s :      1: loop_unroll
 3.05% :     0.000787s :      1: mutable_eliminate
 3.47% :     0.000893s :     26: opt.transform.jit_opt_a
 0.24% :     0.000062s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000028s :      4: opt.transform.jit_opt_b
 0.07% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000040s :      4: opt.transform.symbol_engine_opt
 2.62% :     0.000675s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000054s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000019s :      1: remove_dup_value
 1.68% :     0.000434s :      1: renormalize.infer
 1.33% :     0.000342s :      1: renormalize.specialize
 0.05% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000044s :      1: rewriter_after_opt_a
 0.32% :     0.000082s :      1: rewriter_before_opt_a
 0.38% :     0.000098s :      1: symbol_engine_optimizer
28.11% :     0.007243s :      1: task_emit
26.84% :     0.006915s :      1: type_inference
 0.29% :     0.000074s :      1: validate


TotalTime = 0.0205934, [33]
[bootstrap]: 0.00034348
[type_inference]: 0.00496461
[event_method]: 1.808e-05
[auto_monad]: 6.381e-05
[graph_reusing]: 5.77999e-06
[pre_auto_parallel]: 2.49999e-06
[py_interpret_to_execute]: 4.181e-05
[rewriter_before_opt_a]: 7.435e-05
[expand_dump_flag]: 3.80998e-06
[jit_opt_a]: 0.0065194, [2]
    [Cycle 1]: 0.00204347, [27]
        [switch_simplify]: 5.92e-05
        [loop_unroll]: 3.323e-05
        [a_1]: 0.00060129
        [with_stream_mark]: 1.987e-05
        [recompute_prepare]: 7.34002e-06
        [updatestate_depend_eliminate]: 3.57997e-06
        [updatestate_assign_eliminate]: 2.95998e-06
        [updatestate_loads_eliminate]: 2.34999e-06
        [parameter_eliminate]: 2.16e-06
        [specialize_transform]: 6.71e-06
        [updatestate_useless_node_eliminater]: 6.21e-06
        [accelerated_algorithm]: 5.96e-06
        [meta_shard_fg_expand]: 2.04e-06
        [get_grad_eliminate_]: 5.84999e-06
        [merge_forward]: 3.68e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.535e-05
        [j_node_and_user_rematch]: 9.79999e-06
        [meta_fg_expand]: 2.34999e-06
        [replace_old_param]: 1.032e-05
        [inline_without_move]: 5.92001e-06
        [renormalize]: 0.00094769
        [add_forward_monad_depend]: 7.47998e-06
        [auto_monad_grad]: 2.71e-06
        [auto_monad_eliminator]: 1.415e-05
        [cse]: 2.818e-05
        [replace_applicator]: 1.801e-05
    [Cycle 2]: 0.00034032, [27]
        [switch_simplify]: 5.87001e-06
        [loop_unroll]: 5.00001e-06
        [a_1]: 9.731e-05
        [with_stream_mark]: 1.229e-05
        [recompute_prepare]: 5.09998e-06
        [updatestate_depend_eliminate]: 2.77002e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.37e-06
        [specialize_transform]: 5.09e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 5.81e-06
        [meta_shard_fg_expand]: 1.55999e-06
        [get_grad_eliminate_]: 4.46002e-06
        [merge_forward]: 2.85998e-06
        [cell_reuse_recompute_pass]: 2.74001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.517e-05
        [j_node_and_user_rematch]: 8.45001e-06
        [meta_fg_expand]: 1.54998e-06
        [replace_old_param]: 7.54002e-06
        [inline_without_move]: 4.51002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.11997e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 5.57001e-06
        [cse]: 1.077e-05
        [replace_applicator]: 5.86998e-06
[py_interpret_to_execute_after_opt_a]: 1.302e-05
[rewriter_after_opt_a]: 3.419e-05
[convert_after_rewriter]: 7.62002e-06
[order_py_execute_after_rewriter]: 4.84e-06
[mutable_eliminate]: 0.00078361
[jit_opt_b]: 5.402e-05, [1]
    [Cycle 1]: 4.606e-05, [2]
        [frontend_op_eliminate]: 1.818e-05
        [inline_after_opt_a]: 1.623e-05
[cconv]: 2.819e-05
[loop_unroll]: 0.00044923
[jit_opt_after_cconv]: 0.00014453, [1]
    [Cycle 1]: 0.00013804, [11]
        [c_1]: 2.134e-05
        [parameter_eliminate]: 3.44001e-06
        [updatestate_depend_eliminate]: 5.97999e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [cse]: 1.988e-05
        [call_graph_tuple_transform]: 2.042e-05
        [tuple_list_get_item_eliminator]: 5.69e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 7.60017e-07
        [switch_simplify]: 5.29e-06
[remove_dup_value]: 1.409e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 6.29999e-06
[add_recomputation]: 4.891e-05
[cse_after_recomputation]: 2.104e-05, [1]
    [Cycle 1]: 1.52e-05, [1]
        [cse]: 9.50001e-06
[auto_monad_reorder]: 1.52e-05
[get_jit_bprop_graph]: 2.37001e-06
[rewriter_after_jit_bprop_graph]: 6.13002e-06
[opt_after_jit_grad]: 0.00047794
[symbol_engine_optimizer]: 7.426e-05, [1]
    [Cycle 1]: 6.705e-05, [6]
        [build]: 3.5e-06
        [elim_shapecalc]: 7.43999e-06
        [elim_not_effective]: 1.312e-05
        [opt_reshape]: 5.59998e-06
        [fold_const_symbol]: 8.52e-06
        [renormalize]: 6.10016e-07
[validate]: 3.965e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.00611022
[execute]: 6.44001e-06
Sums
    bootstrap                            :     0.000343s :  2.18%
    type_inference                       :     0.004965s : 31.44%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000064s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000042s :  0.26%
    rewriter_before_opt_a                :     0.000074s :  0.47%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.41%
    jit_opt_a.loop_unroll                :     0.000038s :  0.24%
    jit_opt_a.a_1                        :     0.000699s :  4.42%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000948s :  6.00%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000039s :  0.25%
    jit_opt_a.replace_applicator         :     0.000024s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000034s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000784s :  4.96%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000028s :  0.18%
    loop_unroll                          :     0.000449s :  2.85%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000049s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000478s :  3.03%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006110s : 38.70%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000247    23
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000001s :      2: substitution.fold_const_symbol
 2.38% :     0.000006s :      3: substitution.graph_param_transform
85.17% :     0.000211s :      4: substitution.inline
 1.62% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.95% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.07% :     0.000005s :      2: substitution.replace_old_param
 5.40% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004892     2
81.31% :     0.003978s :      1: type_inference.infer
18.69% :     0.000914s :      1: type_inference.specialize
------[replace.]   0.000079     6
64.84% :     0.000051s :      4: replace.inline
35.16% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000220     6
94.48% :     0.000208s :      4: match.inline
 5.52% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.79% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.40% :     0.000003s :     10: predicate.arithmetic_simplify
 1.30% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.35% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.39% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.11% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.19% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.23% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.36% :     0.000002s :     10: predicate.minmaximum_grad
 1.75% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 1.95% :     0.000002s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.05% :     0.000003s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.52% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.22% :     0.000002s :      6: predicate.specialize_transform
 1.55% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.32% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.82% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.60% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.67% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.84% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000685    12
46.77% :     0.000321s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.90% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.33% :     0.000318s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022569    76
 0.23% :     0.000052s :      1: add_recomputation
 0.30% :     0.000067s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.58% :     0.000356s :      1: bootstrap
 0.14% :     0.000031s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
28.91% :     0.006524s :      1: jit_opt_a
 0.65% :     0.000147s :      1: jit_opt_after_cconv
 0.25% :     0.000057s :      1: jit_opt_b
 2.02% :     0.000457s :      1: loop_unroll
 3.51% :     0.000793s :      1: mutable_eliminate
 4.12% :     0.000931s :     26: opt.transform.jit_opt_a
 0.22% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.15% :     0.000485s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000045s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.26% :     0.000510s :      1: renormalize.infer
 1.89% :     0.000427s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000037s :      1: rewriter_after_opt_a
 0.34% :     0.000078s :      1: rewriter_before_opt_a
 0.34% :     0.000077s :      1: symbol_engine_optimizer
27.15% :     0.006128s :      1: task_emit
22.05% :     0.004978s :      1: type_inference
 0.27% :     0.000060s :      1: validate


TotalTime = 0.0164259, [30]
[bootstrap]: 0.00028458
[type_inference]: 0.00697445
[event_method]: 2.25e-05
[auto_monad]: 8.604e-05
[graph_reusing]: 6.78e-06
[pre_auto_parallel]: 2.42001e-06
[py_interpret_to_execute]: 5.158e-05
[rewriter_before_opt_a]: 8.921e-05
[expand_dump_flag]: 2.99001e-06
[jit_opt_a]: 0.00649434, [2]
    [Cycle 1]: 0.00212183, [27]
        [switch_simplify]: 0.00013534
        [loop_unroll]: 3.426e-05
        [a_1]: 0.00077905
        [with_stream_mark]: 2.662e-05
        [recompute_prepare]: 9.37999e-06
        [updatestate_depend_eliminate]: 4.19002e-06
        [updatestate_assign_eliminate]: 3.13998e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 7e-06
        [updatestate_useless_node_eliminater]: 5.28002e-06
        [accelerated_algorithm]: 5.82001e-06
        [meta_shard_fg_expand]: 1.96e-06
        [get_grad_eliminate_]: 5.52001e-06
        [merge_forward]: 3.42002e-06
        [cell_reuse_recompute_pass]: 1.55001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.574e-05
        [j_node_and_user_rematch]: 9.64999e-06
        [meta_fg_expand]: 2.85002e-06
        [replace_old_param]: 1.135e-05
        [inline_without_move]: 5.34998e-06
        [renormalize]: 0.00076546
        [add_forward_monad_depend]: 5.97001e-06
        [auto_monad_grad]: 2.32999e-06
        [auto_monad_eliminator]: 1.693e-05
        [cse]: 3.153e-05
        [replace_applicator]: 1.42e-05
    [Cycle 2]: 0.00037376, [27]
        [switch_simplify]: 6.04999e-06
        [loop_unroll]: 4.95001e-06
        [a_1]: 9.043e-05
        [with_stream_mark]: 1.191e-05
        [recompute_prepare]: 5.71003e-06
        [updatestate_depend_eliminate]: 2.93998e-06
        [updatestate_assign_eliminate]: 2.20002e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.25999e-06
        [specialize_transform]: 4.83001e-06
        [updatestate_useless_node_eliminater]: 4.64002e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.43001e-06
        [merge_forward]: 2.71999e-06
        [cell_reuse_recompute_pass]: 2.00002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.583e-05
        [j_node_and_user_rematch]: 8.23999e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 9.84999e-06
        [inline_without_move]: 5.35001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 2.36e-06
        [auto_monad_grad]: 1.32999e-06
        [auto_monad_eliminator]: 8.70001e-06
        [cse]: 1.604e-05
        [replace_applicator]: 6.31e-06
[py_interpret_to_execute_after_opt_a]: 1.472e-05
[rewriter_after_opt_a]: 3.948e-05
[convert_after_rewriter]: 7.31999e-06
[order_py_execute_after_rewriter]: 5.71e-06
[mutable_eliminate]: 0.00073994
[jit_opt_b]: 6.551e-05, [1]
    [Cycle 1]: 5.018e-05, [2]
        [frontend_op_eliminate]: 2.323e-05
        [inline_after_opt_a]: 1.535e-05
[cconv]: 2.725e-05
[loop_unroll]: 0.00050204
[jit_opt_after_cconv]: 0.000145, [1]
    [Cycle 1]: 0.00013877, [11]
        [c_1]: 1.968e-05
        [parameter_eliminate]: 3.61999e-06
        [updatestate_depend_eliminate]: 7.01001e-06
        [updatestate_assign_eliminate]: 2.55002e-06
        [updatestate_loads_eliminate]: 2.27001e-06
        [cse]: 2.101e-05
        [call_graph_tuple_transform]: 1.976e-05
        [tuple_list_get_item_eliminator]: 5.46e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 1.19998e-06
        [switch_simplify]: 5.96998e-06
[remove_dup_value]: 1.37e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 4.65001e-06
[add_recomputation]: 5.035e-05
[cse_after_recomputation]: 2.017e-05, [1]
    [Cycle 1]: 1.471e-05, [1]
        [cse]: 9.02e-06
[auto_monad_reorder]: 1.431e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 6.73998e-06
[opt_after_jit_grad]: 0.00045211
[symbol_engine_optimizer]: 7.183e-05, [1]
    [Cycle 1]: 6.569e-05, [6]
        [build]: 3.93001e-06
        [elim_shapecalc]: 7.2600
TotalTime = 0.0222324, [33]
[bootstrap]: 0.0004239
[type_inference]: 0.00571224
[event_method]: 1.84e-05
[auto_monad]: 7.112e-05
[graph_reusing]: 5.76e-06
[pre_auto_parallel]: 2.76999e-06
[py_interpret_to_execute]: 4.181e-05
[rewriter_before_opt_a]: 7.483e-05
[expand_dump_flag]: 3.10002e-06
[jit_opt_a]: 0.00680061, [2]
    [Cycle 1]: 0.00216364, [27]
        [switch_simplify]: 6.356e-05
        [loop_unroll]: 2.906e-05
        [a_1]: 0.00062674
        [with_stream_mark]: 2.493e-05
        [recompute_prepare]: 7.58999e-06
        [updatestate_depend_eliminate]: 3.82002e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.59001e-06
        [parameter_eliminate]: 2.09e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 4.89998e-06
        [accelerated_algorithm]: 6.65002e-06
        [meta_shard_fg_expand]: 2.59001e-06
        [get_grad_eliminate_]: 6.12999e-06
        [merge_forward]: 3.85998e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.514e-05
        [j_node_and_user_rematch]: 1.044e-05
        [meta_fg_expand]: 2.33002e-06
        [replace_old_param]: 1.231e-05
        [inline_without_move]: 5.79e-06
        [renormalize]: 0.00106392
        [add_forward_monad_depend]: 6.59001e-06
        [auto_monad_grad]: 2.37999e-06
        [auto_monad_eliminator]: 1.407e-05
        [cse]: 3.302e-05
        [replace_applicator]: 1.232e-05
    [Cycle 2]: 0.00035462, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 3.067e-05
        [a_1]: 9.015e-05
        [with_stream_mark]: 1.193e-05
        [recompute_prepare]: 4.74e-06
        [updatestate_depend_eliminate]: 2.81e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 9.30013e-07
        [specialize_transform]: 4.85001e-06
        [updatestate_useless_node_eliminater]: 4.77998e-06
        [accelerated_algorithm]: 4.95999e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 4.55001e-06
        [merge_forward]: 2.19001e-06
        [cell_reuse_recompute_pass]: 2.37999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.347e-05
        [j_node_and_user_rematch]: 8.04002e-06
        [meta_fg_expand]: 1.48002e-06
        [replace_old_param]: 6.54999e-06
        [inline_without_move]: 8.77999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.32999e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 5.59e-06
        [cse]: 1.112e-05
        [replace_applicator]: 4.92e-06
[py_interpret_to_execute_after_opt_a]: 1.1e-05
[rewriter_after_opt_a]: 4.047e-05
[convert_after_rewriter]: 7.26001e-06
[order_py_execute_after_rewriter]: 5.66998e-06
[mutable_eliminate]: 0.00065335
[jit_opt_b]: 5.024e-05, [1]
    [Cycle 1]: 4.254e-05, [2]
        [frontend_op_eliminate]: 1.662e-05
        [inline_after_opt_a]: 1.432e-05
[cconv]: 2.744e-05
[loop_unroll]: 0.00042774
[jit_opt_after_cconv]: 0.00014067, [1]
    [Cycle 1]: 0.00013395, [11]
        [c_1]: 2.084e-05
        [parameter_eliminate]: 2.49999e-06
        [updatestate_depend_eliminate]: 6.43e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 1.811e-05
        [call_graph_tuple_transform]: 1.823e-05
        [tuple_list_get_item_eliminator]: 5.25999e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 3.30008e-07
        [switch_simplify]: 5.12e-06
[remove_dup_value]: 1.503e-05
[partial_unused_args_eliminate]: 2.48e-06
[environ_conv]: 5.87999e-06
[add_recomputation]: 5.611e-05
[cse_after_recomputation]: 2.279e-05, [1]
    [Cycle 1]: 1.688e-05, [1]
        [cse]: 9.64e-06
[auto_monad_reorder]: 1.65e-05
[get_jit_bprop_graph]: 2.74001e-06
[rewriter_after_jit_bprop_graph]: 6.14999e-06
[opt_after_jit_grad]: 0.00046266
[symbol_engine_optimizer]: 7.462e-05, [1]
    [Cycle 1]: 6.714e-05, [6]
        [build]: 3.78001e-06
        [elim_shapecalc]: 7.64002e-061e-06
        [elim_not_effective]: 1.238e-05
        [opt_reshape]: 5.41998e-06
        [fold_const_symbol]: 8.55999e-06
        [renormalize]: 4.00003e-07
[validate]: 3.644e-05
Sums
    bootstrap                            :     0.000285s :  2.42%
    type_inference                       :     0.006974s : 59.35%
    event_method                         :     0.000022s :  0.19%
    auto_monad                           :     0.000086s :  0.73%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000052s :  0.44%
    rewriter_before_opt_a                :     0.000089s :  0.76%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000141s :  1.20%
    jit_opt_a.loop_unroll                :     0.000039s :  0.33%
    jit_opt_a.a_1                        :     0.000869s :  7.40%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000021s :  0.18%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.000766s :  6.51%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.22%
    jit_opt_a.cse                        :     0.000048s :  0.40%
    jit_opt_a.replace_applicator         :     0.000021s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.13%
    rewriter_after_opt_a                 :     0.000039s :  0.34%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000740s :  6.30%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.13%
    cconv                                :     0.000027s :  0.23%
    loop_unroll                          :     0.000502s :  4.27%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.17%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000014s :  0.12%
    partial_unused_args_eliminate        :     0.00
        [elim_not_effective]: 1.249e-05
        [opt_reshape]: 5.79e-06
        [fold_const_symbol]: 9.14e-06
        [renormalize]: 5.50004e-07
[validate]: 4.419e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00674828
[execute]: 9.79999e-06
Sums
    bootstrap                            :     0.000424s :  2.45%
    type_inference                       :     0.005712s : 33.07%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000071s :  0.41%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000042s :  0.24%
    rewriter_before_opt_a                :     0.000075s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000069s :  0.40%
    jit_opt_a.loop_unroll                :     0.000060s :  0.35%
    jit_opt_a.a_1                        :     0.000717s :  4.15%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000015s :  0.08%
    jit_opt_a.renormalize                :     0.001064s :  6.16%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.11%
    jit_opt_a.cse                        :     0.000044s :  0.26%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000040s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000653s :  3.78%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.08%
    cconv                                :     0.000027s :  0.16%
    loop_unroll                          :     0.000428s :  2.48%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.00000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000050s :  0.43%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000014s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000452s :  3.85%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.31%

15s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000056s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000463s :  2.68%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006748s : 39.07%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000341    26
 0.60% :     0.000002s :      2: substitution.elim_not_effective
 0.34% :     0.000001s :      2: substitution.fold_const_symbol
 1.69% :     0.000006s :      3: substitution.graph_param_transform
86.61% :     0.000296s :      6: substitution.inline
 1.09% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.45% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.70% :     0.000006s :      2: substitution.replace_old_param
 2.68% :     0.000009s :      1: substitution.switch_simplify
 3.84% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006895     2
83.27% :     0.005742s :      1: type_inference.infer
16.73% :     0.001153s :      1: type_inference.specialize
------[replace.]   0.000156     9
50.28% :     0.000078s :      6: replace.inline
27.34% :     0.000043s :      1: replace.switch_simplify
22.38% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000312     9
93.55% :     0.000292s :      6: match.inline
 2.66% :     0.000008s :      1: match.switch_simplify
 3.80% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000150   801
 1.38% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.91% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     12: predicate.addn_check_dump
 1.14% :     0.000002s :     12: predicate.addn_zero_filter
 2.15% :     0.000003s :     12: predicate.arithmetic_simplify
 1.34% :     0.000002s :     12: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     12: predicate.compare_switch_simplify
 1.24% :     0.000002s :     12: predicate.depend_value_elim
 1.10% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.42% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.24% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.05% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.06% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.03% :     0.000002s :     12: predicate.environ_get_eliminate
 1.03% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.78% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 4.79% :     0.000007s :     26: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.48% :     0.000002s :     17: predicate.load_eliminater
 1.28% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.86% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.67% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.28% :     0.000002s :     12: predicate.merge_addn
 1.04% :     0.000002s :     12: predicate.minmaximum_grad
 2.32% :     0.000003s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 1.90% :     0.000003s :     17: predicate.partial_eliminate
 1.04% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.36% :     0.000002s :     12: predicate.reduce_eliminate
 1.70% :     0.000003s :     14: predicate.redundant_stop_gradient_eliminater
 0.38% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.46% :     0.000002s :     20: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000002s :     12: predicate.reshape_eliminate
 1.37% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 5.42% :     0.000008s :      3: predicate.row_tensor_eliminate
 1.09% :     0.000002s :     12: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.67% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.35% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 4.21% :     0.000006s :     20: predicate.switch_defer_inline
 2.08% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.29% :     0.000012s :     65: predicate.switch_simplify
 1.12% :     0.000002s :     12: predicate.tile_eliminate
 1.23% :     0.000002s :     12: predicate.transpose_eliminate
 1.50% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.07% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.89% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.30% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.30% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.42% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.32% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     12: predicate.value_based_eliminate
 0.24% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000852    15
48.16% :     0.000410s :      6: func_graph_cloner_run.FuncGraphClonerGraph
14.16% :     0.000121s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.68% :     0.000321s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018426    72
 0.29% :     0.000053s :      1: add_recomputation
 0.49% :     0.000090s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 1.64% :     0.000302s :      1: bootstrap
 0.16% :     0.000030s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000027s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
35.27% :     0.006499s :      1: jit_opt_a
 0.80% :     0.000148s :      1: jit_opt_after_cconv
 0.37% :     0.000069s :      1: jit_opt_b
 2.76% :     0.000509s :      1: loop_unroll
 4.09% :     0.000753s :      1: mutable_eliminate
 6.39% :     0.001177s :     26: opt.transform.jit_opt_a
 0.25% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000032s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.50% :     0.000460s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.30% :     0.000055s :      1: py_interpret_to_execute
 0.10% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 2.40% :     0.000442s :      1: renormalize.infer
 1.71% :     0.000315s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000044s :      1: rewriter_after_opt_a
 0.50% :     0.000093s :      1: rewriter_before_opt_a
 0.40% :     0.000075s :      1: symbol_engine_optimizer
37.96% :     0.006994s :      1: type_inference


Time group info:
------[substitution.]   0.000278    23
 0.71% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000002s :      2: substitution.fold_const_symbol
 2.16% :     0.000006s :      3: substitution.graph_param_transform
86.22% :     0.000240s :      4: substitution.inline
 1.44% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.72% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.69% :     0.000005s :      2: substitution.replace_old_param
 5.52% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005632     2
80.62% :     0.004541s :      1: type_inference.infer
19.38% :     0.001091s :      1: type_inference.specialize
------[replace.]   0.000084     6
68.24% :     0.000057s :      4: replace.inline
31.76% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000252     6
94.35% :     0.000237s :      4: match.inline
 5.65% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.72% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000001s :     10: predicate.addn_zero_filter
 2.10% :     0.000002s :     10: predicate.arithmetic_simplify
 1.19% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.22% :     0.000001s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.60% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.12% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.87% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.85% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.27% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.87% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.43% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     18: predicate.replace_applicator
 0.96% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.20% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.22% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.10% :     0.000010s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.25% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.63% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.80% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000785    12
45.27% :     0.000355s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.19% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.53% :     0.000373s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024325    76
 0.24% :     0.000059s :      1: add_recomputation
 0.31% :     0.000075s :      1: auto_monad
 0.08% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.85% :     0.000449s :      1: bootstrap
 0.12% :     0.000030s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
27.97% :     0.006804s :      1: jit_opt_a
 0.59% :     0.000143s :      1: jit_opt_after_cconv
 0.22% :     0.000053s :      1: jit_opt_b
 1.80% :     0.000437s :      1: loop_unroll
 2.73% :     0.000663s :      1: mutable_eliminate
 3.89% :     0.000945s :     26: opt.transform.jit_opt_a
 0.19% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.93% :     0.000471s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000044s :      1: py_interpret_to_execute
 0.05% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.68% :     0.000652s :      1: renormalize.infer
 1.66% :     0.000403s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000043s :      1: rewriter_after_opt_a
 0.32% :     0.000078s :      1: rewriter_before_opt_a
 0.32% :     0.000077s :      1: symbol_engine_optimizer
27.83% :     0.006769s :      1: task_emit
23.57% :     0.005733s :      1: type_inference
 0.27% :     0.000066s :      1: validate


TotalTime = 0.0219855, [33]
[bootstrap]: 0.00033327
[type_inference]: 0.0051712
[event_method]: 1.814e-05
[auto_monad]: 6.693e-05
[graph_reusing]: 6.13002e-06
[pre_auto_parallel]: 2.51e-06
[py_interpret_to_execute]: 3.893e-05
[rewriter_before_opt_a]: 7.482e-05
[expand_dump_flag]: 3.48999e-06
[jit_opt_a]: 0.00708139, [2]
    [Cycle 1]: 0.00204569, [27]
        [switch_simplify]: 6.477e-05
        [loop_unroll]: 3.191e-05
        [a_1]: 0.0006169
        [with_stream_mark]: 2.333e-05
        [recompute_prepare]: 9.42999e-06
        [updatestate_depend_eliminate]: 4.01001e-06
        [updatestate_assign_eliminate]: 2.97002e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [parameter_eliminate]: 2.24999e-06
        [specialize_transform]: 6.28e-06
        [updatestate_useless_node_eliminater]: 5.25001e-06
        [accelerated_algorithm]: 5.77999e-06
        [meta_shard_fg_expand]: 2.77002e-06
        [get_grad_eliminate_]: 6.39999e-06
        [merge_forward]: 4.32998e-06
        [cell_reuse_recompute_pass]: 1.53002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.275e-05
        [j_node_and_user_rematch]: 1.152e-05
        [meta_fg_expand]: 2.71e-06
        [replace_old_param]: 9.74e-06
        [inline_without_move]: 5.92999e-06
        [renormalize]: 0.00092433
        [add_forward_monad_depend]: 7.32002e-06
        [auto_monad_grad]: 2.36e-06
        [auto_monad_eliminator]: 1.816e-05
        [cse]: 2.883e-05
        [replace_applicator]: 1.462e-05
    [Cycle 2]: 0.00033055, [27]
        [switch_simplify]: 6.09001e-06
        [loop_unroll]: 4.77e-06
        [a_1]: 9.323e-05
        [with_stream_mark]: 9.87001e-06
        [recompute_prepare]: 5.21002e-06
        [updatestate_depend_eliminate]: 2.53e-06
        [updatestate_assign_eliminate]: 1.92001e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 1.37999e-06
        [specialize_transform]: 4.95999e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 5.30999e-06
        [meta_shard_fg_expand]: 1.22999e-06
        [get_grad_eliminate_]: 4.74e-06
        [merge_forward]: 2.59001e-06
        [cell_reuse_recompute_pass]: 2.11e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.366e-05
        [j_node_and_user_rematch]: 8.33999e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 7.21999e-06
        [inline_without_move]: 4.63001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 6.02999e-06
        [cse]: 1.033e-05
        [replace_applicator]: 4.97999e-06
[py_interpret_to_execute_after_opt_a]: 1.211e-05
[rewriter_after_opt_a]: 3.953e-05
[convert_after_rewriter]: 7.48999e-06
[order_py_execute_after_rewriter]: 5.27001e-06
[mutable_eliminate]: 0.00077473
[jit_opt_b]: 5.128e-05, [1]
    [Cycle 1]: 4.374e-05, [2]
        [frontend_op_eliminate]: 1.704e-05
        [inline_after_opt_a]: 1.511e-05
[cconv]: 2.739e-05
[loop_unroll]: 0.00043649
[jit_opt_after_cconv]: 0.0001846, [1]
    [Cycle 1]: 0.00017798, [11]
        [c_1]: 2.189e-05
        [parameter_eliminate]: 2.63e-06
        [updatestate_depend_eliminate]: 5.57999e-06
        [updatestate_assign_eliminate]: 2.41998e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 1.778e-05
        [call_graph_tuple_transform]: 2.068e-05
        [tuple_list_get_item_eliminator]: 5.36998e-06
        [none_parameter_eliminate]: 2.01003e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 5.79e-06
[remove_dup_value]: 1.407e-05
[partial_unused_args_eliminate]: 2.19001e-06
[environ_conv]: 5.96e-06
[add_recomputation]: 5.319e-05
[cse_after_recomputation]: 2.26e-05, [1]
    [Cycle 1]: 1.656e-05, [1]
        [cse]: 1.019e-05
[auto_monad_reorder]: 0.00033449
[get_jit_bprop_graph]: 2.86e-06
[rewriter_after_jit_bprop_graph]: 6.94999e-06
[opt_after_jit_grad]: 0.00049518
[symbol_engine_optimizer]: 7.547e-05, [1]
    [Cycle 1]: 6.838e-05, [6]
        [build]: 3.71001e-06
        [elim_shapecalc]: 8.27003e-06
        [elim_not_effective]: 1.301e-05
        [opt_reshape]: 5.85002e-06
        [fold_const_symbol]: 8.58001e-06
        [renormalize]: 5.50004e-07
[validate]: 4.068e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.0063523
[execute]: 8.60001e-06
Sums
    bootstrap                            :     0.000333s :  2.01%
    type_inference                       :     0.005171s : 31.24%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000067s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.24%
    rewriter_before_opt_a                :     0.000075s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000071s :  0.43%
    jit_opt_a.loop_unroll                :     0.000037s :  0.22%
    jit_opt_a.a_1                        :     0.000710s :  4.29%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000924s :  5.58%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.15%
    jit_opt_a.cse                        :     0.000039s :  0.24%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000040s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000775s :  4.68%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000027s :  0.17%
    loop_unroll                          :     0.000436s :  2.64%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000053s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000334s :  2.02%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000495s :  2.99%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006352s : 38.37%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000261    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.34% :     0.000006s :      3: substitution.graph_param_transform
82.81% :     0.000216s :      4: substitution.inline
 1.83% :     0.000005s :      4: substitution.j_node_and_user_rematch
 4.29% :     0.000011s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000005s :      2: substitution.replace_old_param
 5.59% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005098     2
82.12% :     0.004186s :      1: type_inference.infer
17.88% :     0.000912s :      1: type_inference.specialize
------[replace.]   0.000078     6
66.53% :     0.000052s :      4: replace.inline
33.47% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000226     6
94.09% :     0.000213s :      4: match.inline
 5.91% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.49% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.55% :     0.000002s :     10: predicate.addn_zero_filter
 2.16% :     0.000003s :     10: predicate.arithmetic_simplify
 1.54% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.22% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.09% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.43% :     0.000002s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.17% :     0.000002s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.29% :     0.000007s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :      6: predicate.less_batch_normalization
 1.76% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.06% :     0.000003s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.86% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.03% :     0.000003s :     15: predicate.partial_eliminate
 1.26% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.87% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.73% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.43% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.74% :     0.000002s :     10: predicate.reshape_eliminate
 1.29% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.67% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.61% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.17% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.67% :     0.000010s :     53: predicate.switch_simplify
 1.21% :     0.000002s :     10: predicate.tile_eliminate
 1.21% :     0.000002s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.84% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.01% :     0.000003s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000693    12
49.12% :     0.000341s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.95% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.92% :     0.000311s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023952    76
 0.24% :     0.000056s :      1: add_recomputation
 0.29% :     0.000070s :      1: auto_monad
 1.43% :     0.000343s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.47% :     0.000352s :      1: bootstrap
 0.13% :     0.000030s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000006s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
29.58% :     0.007086s :      1: jit_opt_a
 0.78% :     0.000188s :      1: jit_opt_after_cconv
 0.22% :     0.000054s :      1: jit_opt_b
 1.86% :     0.000445s :      1: loop_unroll
 3.27% :     0.000784s :      1: mutable_eliminate
 3.96% :     0.000948s :     26: opt.transform.jit_opt_a
 0.21% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.10% :     0.000504s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000042s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.12% :     0.000508s :      1: renormalize.infer
 1.70% :     0.000407s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000042s :      1: rewriter_after_opt_a
 0.32% :     0.000077s :      1: rewriter_before_opt_a
 0.33% :     0.000078s :      1: symbol_engine_optimizer
26.61% :     0.006373s :      1: task_emit
21.65% :     0.005186s :      1: type_inference
 0.26% :     0.000063s :      1: validate


TotalTime = 0.0479231, [33]
[bootstrap]: 0.00030683
[type_inference]: 0.0300676
[event_method]: 1.793e-05
[auto_monad]: 6.739e-05
[graph_reusing]: 6.27001e-06
[pre_auto_parallel]: 2.89001e-06
[py_interpret_to_execute]: 0.00120354
[rewriter_before_opt_a]: 8.92e-05
[expand_dump_flag]: 4.18999e-06
[jit_opt_a]: 0.00684666, [2]
    [Cycle 1]: 0.00209555, [27]
        [switch_simplify]: 7.044e-05
        [loop_unroll]: 3.069e-05
        [a_1]: 0.00073191
        [with_stream_mark]: 2.231e-05
        [recompute_prepare]: 8.09002e-06
        [updatestate_depend_eliminate]: 3.33998e-06
        [updatestate_assign_eliminate]: 3.26001e-06
        [updatestate_loads_eliminate]: 3.18e-06
        [parameter_eliminate]: 1.92999e-06
        [specialize_transform]: 6.03998e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 6.58e-06
        [meta_shard_fg_expand]: 2.70002e-06
        [get_grad_eliminate_]: 5.32001e-06
        [merge_forward]: 3.81999e-06
        [cell_reuse_recompute_pass]: 1.07998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.556e-05
        [j_node_and_user_rematch]: 9.79e-06
        [meta_fg_expand]: 2.69001e-06
        [replace_old_param]: 9.87001e-06
        [inline_without_move]: 5.56e-06
        [renormalize]: 0.00089411
        [add_forward_monad_depend]: 6.27001e-06
        [auto_monad_grad]: 2.53e-06
        [auto_monad_eliminator]: 1.471e-05
        [cse]: 3.015e-05
        [replace_applicator]: 1.33e-05
    [Cycle 2]: 0.00034622, [27]
        [switch_simplify]: 5.61e-06
        [loop_unroll]: 5.40999e-06
        [a_1]: 9.1e-05
        [with_stream_mark]: 1.018e-05
        [recompute_prepare]: 5.39e-06
        [updatestate_depend_eliminate]: 2.61e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [parameter_eliminate]: 1.25999e-06
        [specialize_transform]: 5.32001e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 1.729e-05
        [merge_forward]: 2.54999e-06
        [cell_reuse_recompute_pass]: 2.58003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.353e-05
        [j_node_and_user_rematch]: 7.82998e-06
        [meta_fg_expand]: 1.89e-06
        [replace_old_param]: 7.19001e-06
        [inline_without_move]: 4.47e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 5.75001e-06
        [cse]: 1.178e-05
        [replace_applicator]: 5.02999e-06
[py_interpret_to_execute_after_opt_a]: 1.248e-05
[rewriter_after_opt_a]: 3.673e-05
[convert_after_rewriter]: 7.23e-06
[order_py_execute_after_rewriter]: 5.62001e-06
[mutable_eliminate]: 0.0007156
[jit_opt_b]: 5.981e-05, [1]
    [Cycle 1]: 5.194e-05, [2]
        [frontend_op_eliminate]: 2.441e-05
        [inline_after_opt_a]: 1.577e-05
[cconv]: 2.709e-05
[loop_unroll]: 0.00061483
[jit_opt_after_cconv]: 0.00015505, [1]
    [Cycle 1]: 0.00014813, [11]
        [c_1]: 2.676e-05
        [parameter_eliminate]: 2.81e-06
        [updatestate_depend_eliminate]: 6.62002e-06
        [updatestate_assign_eliminate]: 2.33998e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 2.24e-05
        [call_graph_tuple_transform]: 2.099e-05
        [tuple_list_get_item_eliminator]: 5.61e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.39998e-06
[remove_dup_value]: 1.596e-05
[partial_unused_args_eliminate]: 2.68e-06
[environ_conv]: 5.74999e-06
[add_recomputation]: 5.175e-05
[cse_after_recomputation]: 2.162e-05, [1]
    [Cycle 1]: 1.57e-05, [1]
        [cse]: 9.61e-06
[auto_monad_reorder]: 1.706e-05
[get_jit_bprop_graph]: 1.72001e-06
[rewriter_after_jit_bprop_graph]: 7.4e-06
[opt_after_jit_grad]: 0.00050493
[symbol_engine_optimizer]: 7.898e-05, [1]
    [Cycle 1]: 7.121e-05, [6]
        [build]: 4.32e-06
        [elim_shapecalc]: 8.11002e-06
        [elim_not_effective]: 1.436e-05
        [opt_reshape]: 6.06e-06
        [fold_const_symbol]: 9.04998e-06
        [renormalize]: 6.89994e-07
[validate]: 4.329e-05
[backend_pass]: 1.17e-06
[task_emit]: 0.00667816
[execute]: 8.68001e-06
Sums
    bootstrap                            :     0.000307s :  0.72%
    type_inference                       :     0.030068s : 70.17%
    event_method                         :     0.000018s :  0.04%
    auto_monad                           :     0.000067s :  0.16%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.001204s :  2.81%
    rewriter_before_opt_a                :     0.000089s :  0.21%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000076s :  0.18%
    jit_opt_a.loop_unroll                :     0.000036s :  0.08%
    jit_opt_a.a_1                        :     0.000823s :  1.92%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.07%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000017s :  0.04%
    jit_opt_a.inline_without_move        :     0.000010s :  0.02%
    jit_opt_a.renormalize                :     0.000894s :  2.09%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.05%
    jit_opt_a.cse                        :     0.000042s :  0.10%
    jit_opt_a.replace_applicator         :     0.000018s :  0.04%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000037s :  0.09%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.01%
    mutable_eliminate                    :     0.000716s :  1.67%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.04%
    cconv                                :     0.000027s :  0.06%
    loop_unroll                          :     0.000615s :  1.43%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.01%
    remove_dup_value                     :     0.000016s :  0.04%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.01%
    add_recomputation                    :     0.000052s :  0.12%
    cse_after_recomputation.cse          :     0.000010s :  0.02%
    auto_monad_reorder                   :     0.000017s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000505s :  1.18%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.10%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006678s : 15.59%
    execute                              :     0.000009s :  0.02%


TotalTime = 0.0209005, [33]
[bootstrap]: 0.00032763
[type_inference]: 0.00556908
[event_method]: 2.01e-05
[auto_monad]: 6.351e-05
[graph_reusing]: 6.27001e-06
[pre_auto_parallel]: 2.09e-06
[py_interpret_to_execute]: 4.52e-05
[rewriter_before_opt_a]: 7.131e-05
[expand_dump_flag]: 2.81999e-06
[jit_opt_a]: 0.00625329, [2]
    [Cycle 1]: 0.00193061, [27]
        [switch_simplify]: 0.00015304
        [loop_unroll]: 3.169e-05
        [a_1]: 0.00053756
        [with_stream_mark]: 1.481e-05
        [recompute_prepare]: 7.49002e-06
        [updatestate_depend_eliminate]: 2.75002e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.19998e-06
        [specialize_transform]: 6.07001e-06
        [updatestate_useless_node_eliminater]: 5.26998e-06
        [accelerated_algorithm]: 5.48002e-06
        [meta_shard_fg_expand]: 1.47999e-06
        [get_grad_eliminate_]: 4.65001e-06
        [merge_forward]: 2.86999e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.269e-05
        [j_node_and_user_rematch]: 8.85999e-06
        [meta_fg_expand]: 1.95001e-06
        [replace_old_param]: 7.45998e-06
        [inline_without_move]: 5.14e-06
        [renormalize]: 0.00089306
        [add_forward_monad_depend]: 6.28998e-06
        [auto_monad_grad]: 1.60999e-06
        [auto_monad_eliminator]: 1.145e-05
        [cse]: 1.783e-05
        [replace_applicator]: 1.343e-05
    [Cycle 2]: 0.00032184, [27]
        [switch_simplify]: 6.05002e-06
        [loop_unroll]: 5.24e-06
        [a_1]: 9.369e-05
        [with_stream_mark]: 9.07999e-06
        [recompute_prepare]: 4.85001e-06
        [updatestate_depend_eliminate]: 2.87002e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.10999e-06
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.45999e-06
        [merge_forward]: 2.41e-06
        [cell_reuse_recompute_pass]: 2.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.312e-05
        [j_node_and_user_rematch]: 7.70998e-06
        [meta_fg_expand]: 1.58002e-06
        [replace_old_param]: 7.38e-06
        [inline_without_move]: 4.52998e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.40001e-06
        [auto_monad_grad]: 1.01002e-06
        [auto_monad_eliminator]: 5.46002e-06
        [cse]: 1.073e-05
        [replace_applicator]: 5.03002e-06
[py_interpret_to_execute_after_opt_a]: 9.87001e-06
[rewriter_after_opt_a]: 3.075e-05
[convert_after_rewriter]: 5.51e-06
[order_py_execute_after_rewriter]: 4.52e-06
[mutable_eliminate]: 0.00066449
[jit_opt_b]: 5.183e-05, [1]
    [Cycle 1]: 4.481e-05, [2]
        [frontend_op_eliminate]: 1.786e-05
        [inline_after_opt_a]: 1.538e-05
[cconv]: 1.872e-05
[loop_unroll]: 0.00045155
[jit_opt_after_cconv]: 0.00014067, [1]
    [Cycle 1]: 0.00013455, [11]
        [c_1]: 2.216e-05
        [parameter_eliminate]: 2.63998e-06
        [updatestate_depend_eliminate]: 5.66998e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.876e-05
        [call_graph_tuple_transform]: 1.788e-05
        [tuple_list_get_item_eliminator]: 5.51e-06
        [none_parameter_eliminate]: 1.37999e-06
        [renormalize]: 5.10016e-07
        [switch_simplify]: 5.25999e-06
[remove_dup_value]: 7.92998e-06
[partial_unused_args_eliminate]: 1.92001e-06
[environ_conv]: 4.38999e-06
[add_recomputation]: 3.635e-05
[cse_after_recomputation]: 2.094e-05, [1]
    [Cycle 1]: 1.586e-05, [1]
        [cse]: 8.94998e-06
[auto_monad_reorder]: 1.129e-05
[get_jit_bprop_graph]: 1.51998e-06
[rewriter_after_jit_bprop_graph]: 5.94e-06
[opt_after_jit_grad]: 0.00047684
[symbol_engine_optimizer]: 7.349e-05, [1]
    [Cycle 1]: 6.613e-05, [6]
        [build]: 3.08e-06
        [elim_shapecalc]: 7.43e-06
        [elim_not_effective]: 1.282e-05
        [opt_reshape]: 6.07001e-06
        [fold_const_symbol]: 8.82e-06
        [renormalize]: 6.00005e-07
[validate]: 3.279e-05
[backend_pass]: 7.2e-07
[task_emit]: 0.00626883
[execute]: 4.27e-06
Sums
    bootstrap                            :     0.000328s :  2.01%
    type_inference                       :     0.005569s : 34.21%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000064s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000045s :  0.28%
    rewriter_before_opt_a                :     0.000071s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000159s :  0.98%
    jit_opt_a.loop_unroll                :     0.000037s :  0.23%
    jit_opt_a.a_1                        :     0.000631s :  3.88%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000893s :  5.49%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.10%
    jit_opt_a.cse                        :     0.000029s :  0.18%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000664s :  4.08%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000019s :  0.11%
    loop_unroll                          :     0.000452s :  2.77%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000008s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000036s :  0.22%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000011s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000477s :  2.93%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000033s :  0.20%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006269s : 38.50%
    execute                              :     0.000004s :  0.03%


Time group info:
------[substitution.]   0.000279    23
 0.80% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.17% :     0.000006s :      3: substitution.graph_param_transform
86.86% :     0.000242s :      4: substitution.inline
 1.29% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.60% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.82% :     0.000005s :      2: substitution.replace_old_param
 4.99% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.029989     2
96.69% :     0.028996s :      1: type_inference.infer
 3.31% :     0.000993s :      1: type_inference.specialize
------[replace.]   0.000169     6
83.99% :     0.000142s :      4: replace.inline
16.01% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000252     6
94.94% :     0.000239s :      4: match.inline
 5.06% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.67% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.23% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.20% :     0.000001s :     10: predicate.addn_zero_filter
 2.27% :     0.000003s :     10: predicate.arithmetic_simplify
 1.36% :     0.000002s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.13% :     0.000001s :     10: predicate.depend_value_elim
 1.23% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.62% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.45% :     0.000002s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.17% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 6.06% :     0.000007s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000001s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.25% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.69% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 1.47% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.04% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.14% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.06% :     0.000001s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.19% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.76% :     0.000003s :     16: predicate.switch_defer_inline
 1.99% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.55% :     0.000009s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.23% :     0.000002s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.82% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.03% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.84% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.025006    12
98.49% :     0.024629s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 0.22% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
 1.28% :     0.000321s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.050134    76
 0.11% :     0.000055s :      1: add_recomputation
 0.14% :     0.000071s :      1: auto_monad
 0.04% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.64% :     0.000323s :      1: bootstrap
 0.06% :     0.000030s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.05% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.04% :     0.000023s :      1: event_method
 0.03% :     0.000013s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
13.66% :     0.006850s :      1: jit_opt_a
 0.32% :     0.000158s :      1: jit_opt_after_cconv
 0.12% :     0.000063s :      1: jit_opt_b
 1.25% :     0.000625s :      1: loop_unroll
 1.45% :     0.000725s :      1: mutable_eliminate
 2.12% :     0.001065s :     26: opt.transform.jit_opt_a
 0.11% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000034s :      4: opt.transform.jit_opt_b
 0.31% :     0.000156s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.02% :     0.000513s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 2.43% :     0.001219s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000018s :      1: remove_dup_value
 0.95% :     0.000477s :      1: renormalize.infer
 0.81% :     0.000408s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000041s :      1: rewriter_after_opt_a
 0.19% :     0.000094s :      1: rewriter_before_opt_a
 0.16% :     0.000082s :      1: symbol_engine_optimizer
13.35% :     0.006693s :      1: task_emit
60.01% :     0.030084s :      1: type_inference
 0.13% :     0.000064s :      1: validate


Time group info:
------[substitution.]   0.000194    23
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.35% :     0.000005s :      3: substitution.graph_param_transform
86.59% :     0.000168s :      4: substitution.inline
 1.41% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.32% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.61% :     0.000003s :      2: substitution.replace_old_param
 4.24% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005488     2
81.38% :     0.004467s :      1: type_inference.infer
18.62% :     0.001022s :      1: type_inference.specialize
------[replace.]   0.000074     6
66.99% :     0.000050s :      4: replace.inline
33.01% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000173     6
95.87% :     0.000166s :      4: match.inline
 4.13% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.72% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.20% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.18% :     0.000001s :     10: predicate.addn_zero_filter
 2.31% :     0.000003s :     10: predicate.arithmetic_simplify
 1.46% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 1.23% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.55% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.44% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.83% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 4.91% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      6: predicate.less_batch_normalization
 1.82% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.08% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.20% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.29% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.58% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 1.98% :     0.000002s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.77% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.83% :     0.000002s :     10: predicate.reshape_eliminate
 1.11% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.52% :     0.000002s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.41% :     0.000002s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.57% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.81% :     0.000009s :     53: predicate.switch_simplify
 1.31% :     0.000002s :     10: predicate.tile_eliminate
 1.06% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.15% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.48% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000664    12
43.81% :     0.000291s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.69% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.50% :     0.000342s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022823    76
 0.17% :     0.000039s :      1: add_recomputation
 0.29% :     0.000067s :      1: auto_monad
 0.06% :     0.000014s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.51% :     0.000344s :      1: bootstrap
 0.09% :     0.000021s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.11% :     0.000024s :      1: event_method
 0.03% :     0.000008s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.42% :     0.006258s :      1: jit_opt_a
 0.63% :     0.000143s :      1: jit_opt_after_cconv
 0.24% :     0.000054s :      1: jit_opt_b
 2.01% :     0.000459s :      1: loop_unroll
 2.95% :     0.000673s :      1: mutable_eliminate
 4.12% :     0.000940s :     26: opt.transform.jit_opt_a
 0.21% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.13% :     0.000485s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.21% :     0.000048s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000010s :      1: remove_dup_value
 2.10% :     0.000480s :      1: renormalize.infer
 1.78% :     0.000406s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000034s :      1: rewriter_after_opt_a
 0.32% :     0.000074s :      1: rewriter_before_opt_a
 0.33% :     0.000076s :      1: symbol_engine_optimizer
27.52% :     0.006281s :      1: task_emit
24.48% :     0.005586s :      1: type_inference
 0.23% :     0.000051s :      1: validate


TotalTime = 0.0137331, [30]
[bootstrap]: 0.00025704
[type_inference]: 0.00611334
[event_method]: 2.075e-05
[auto_monad]: 7.917e-05
[graph_reusing]: 6.61999e-06
[pre_auto_parallel]: 2.58e-06
[py_interpret_to_execute]: 4.713e-05
[rewriter_before_opt_a]: 8.548e-05
[expand_dump_flag]: 3.46001e-06
[jit_opt_a]: 0.00515669, [2]
    [Cycle 1]: 0.00174121, [27]
        [switch_simplify]: 0.00010931
        [loop_unroll]: 3.376e-05
        [a_1]: 0.0006576
        [with_stream_mark]: 1.714e-05
        [recompute_prepare]: 7.19001e-06
        [updatestate_depend_eliminate]: 3.45e-06
        [updatestate_assign_eliminate]: 2.83998e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [parameter_eliminate]: 1.71002e-06
        [specialize_transform]: 6.02999e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 5.40999e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 3.85e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.424e-05
        [j_node_and_user_rematch]: 9.19e-06
        [meta_fg_expand]: 2.34999e-06
        [replace_old_param]: 8.55001e-06
        [inline_without_move]: 5.44998e-06
        [renormalize]: 0.00060897
        [add_forward_monad_depend]: 5.63002e-06
        [auto_monad_grad]: 2.48998e-06
        [auto_monad_eliminator]: 1.315e-05
        [cse]: 2.918e-05
        [replace_applicator]: 1.157e-05
    [Cycle 2]: 0.00031929, [27]
        [switch_simplify]: 5.42001e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 8.858e-05
        [with_stream_mark]: 8.18999e-06
        [recompute_prepare]: 5.23002e-06
        [updatestate_depend_eliminate]: 2.43e-06
        [updatestate_assign_eliminate]: 1.97999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 4.82e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 4.74998e-06
        [meta_shard_fg_expand]: 1.02e-06
        [get_grad_eliminate_]: 4.68999e-06
        [merge_forward]: 2.55002e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.011e-05
        [j_node_and_user_rematch]: 7.56001e-06
        [meta_fg_expand]: 1.45999e-06
        [replace_old_param]: 6.98e-06
        [inline_without_move]: 4.74e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 5.30999e-06
        [cse]: 1.024e-05
        [replace_applicator]: 4.87998e-06
[py_interpret_to_execute_after_opt_a]: 1.005e-05
[rewriter_after_opt_a]: 3.243e-05
[convert_after_rewriter]: 6.58e-06
[order_py_execute_after_rewriter]: 5.07e-06
[mutable_eliminate]: 0.00048542
[jit_opt_b]: 4.913e-05, [1]
    [Cycle 1]: 4.202e-05, [2]
        [frontend_op_eliminate]: 1.671e-05
        [inline_after_opt_a]: 1.428e-05
[cconv]: 2.273e-05
[loop_unroll]: 0.00039749
[jit_opt_after_cconv]: 0.00013989, [1]
    [Cycle 1]: 0.00013351, [11]
        [c_1]: 2.017e-05
        [parameter_eliminate]: 2.26e-06
        [updatestate_depend_eliminate]: 5.71e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 2.01998e-06
        [cse]: 1.568e-05
        [call_graph_tuple_transform]: 1.728e-05
        [tuple_list_get_item_eliminator]: 5.10999e-06
        [none_parameter_eliminate]: 1.40001e-06
        [renormalize]: 7.00005e-07
        [switch_simplify]: 5.40999e-06
[remove_dup_value]: 1.444e-05
[partial_unused_args_eliminate]: 2.26998e-06
[environ_conv]: 4.76997e-06
[add_recomputation]: 4.485e-05
[cse_after_recomputation]: 2.032e-05, [1]
    [Cycle 1]: 1.457e-05, [1]
        [cse]: 8.92e-06
[auto_monad_reorder]: 1.366e-05
[get_jit_bprop_graph]: 1.85001e-06
[rewriter_after_jit_bprop_graph]: 4.80001e-06
[opt_after_jit_grad]: 0.00044119
[symbol_engine_optimizer]: 7.12e-05, [1]
    [Cycle 1]: 6.533e-05, [6]
        [build]: 3.66999e-06
        [elim_shapecalc]: 7.61001e-06
        [elim_not_effective]: 1.166e-05
        [opt_reshape]: 5.77001e-06
        [fold_const_symbol]: 8.25999e-06
        [renormalize]: 3.19997e-07
[validate]: 3.344e-05
Sums
    bootstrap                            :     0.000257s :  2.55%
    type_inference                       :     0.006113s : 60.68%
    event_method                         :     0.000021s :  0.21%
    auto_monad                           :     0.000079s :  0.79%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000047s :  0.47%
    rewriter_before_opt_a                :     0.000085s :  0.85%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000115s :  1.14%
    jit_opt_a.loop_unroll                :     0.000039s :  0.38%
    jit_opt_a.a_1                        :     0.000746s :  7.41%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000609s :  6.05%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.18%
    jit_opt_a.cse                        :     0.000039s :  0.39%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.10%
    rewriter_after_opt_a                 :     0.000032s :  0.32%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000485s :  4.82%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.14%
    cconv                                :     0.000023s :  0.23%
    loop_unroll                          :     0.000397s :  3.95%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000045s :  0.45%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000014s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000441s :  4.38%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.33%


Time group info:
------[substitution.]   0.000254    26
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 1.97% :     0.000005s :      3: substitution.graph_param_transform
84.44% :     0.000215s :      6: substitution.inline
 1.13% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.82% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.53% :     0.000004s :      2: substitution.replace_old_param
 2.96% :     0.000008s :      1: substitution.switch_simplify
 4.90% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006045     2
83.49% :     0.005046s :      1: type_inference.infer
16.51% :     0.000998s :      1: type_inference.specialize
------[replace.]   0.000127     9
50.87% :     0.000065s :      6: replace.inline
26.29% :     0.000033s :      1: replace.switch_simplify
22.84% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000230     9
92.15% :     0.000212s :      6: match.inline
 2.93% :     0.000007s :      1: match.switch_simplify
 4.92% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   801
 1.37% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.00% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     12: predicate.addn_check_dump
 1.43% :     0.000002s :     12: predicate.addn_zero_filter
 1.98% :     0.000003s :     12: predicate.arithmetic_simplify
 1.30% :     0.000002s :     12: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     12: predicate.compare_switch_simplify
 1.50% :     0.000002s :     12: predicate.depend_value_elim
 1.18% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.50% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.46% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     12: predicate.environ_get_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.18% :     0.000002s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 5.85% :     0.000008s :     26: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     17: predicate.load_eliminater
 1.21% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.94% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     12: predicate.merge_addn
 1.14% :     0.000001s :     12: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.43% :     0.000003s :     17: predicate.partial_eliminate
 1.21% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     12: predicate.reduce_eliminate
 1.46% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     20: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.56% :     0.000002s :     12: predicate.reshape_eliminate
 1.36% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.57% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.57% :     0.000003s :     20: predicate.switch_defer_inline
 2.46% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.29% :     0.000011s :     65: predicate.switch_simplify
 1.19% :     0.000002s :     12: predicate.tile_eliminate
 1.24% :     0.000002s :     12: predicate.transpose_eliminate
 1.55% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000002s :     12: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000744    15
50.18% :     0.000373s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.44% :     0.000085s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.37% :     0.000285s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015407    72
 0.31% :     0.000047s :      1: add_recomputation
 0.54% :     0.000083s :      1: auto_monad
 0.10% :     0.000016s :      1: auto_monad_reorder
 1.77% :     0.000273s :      1: bootstrap
 0.16% :     0.000025s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
33.49% :     0.005159s :      1: jit_opt_a
 0.93% :     0.000143s :      1: jit_opt_after_cconv
 0.34% :     0.000052s :      1: jit_opt_b
 2.63% :     0.000404s :      1: loop_unroll
 3.20% :     0.000493s :      1: mutable_eliminate
 6.57% :     0.001012s :     26: opt.transform.jit_opt_a
 0.29% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.91% :     0.000448s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.32% :     0.000050s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 2.21% :     0.000341s :      1: renormalize.infer
 1.69% :     0.000260s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000035s :      1: rewriter_after_opt_a
 0.57% :     0.000088s :      1: rewriter_before_opt_a
 0.48% :     0.000074s :      1: symbol_engine_optimizer
39.76% :     0.006126s :      1: type_inference


TotalTime = 0.0302173, [33]
[bootstrap]: 0.00033692
[type_inference]: 0.0080285
[event_method]: 2.208e-05
[auto_monad]: 6.942e-05
[graph_reusing]: 5.81e-06
[pre_auto_parallel]: 2.63e-06
[py_interpret_to_execute]: 4.699e-05
[rewriter_before_opt_a]: 7.498e-05
[expand_dump_flag]: 3.26001e-06
[jit_opt_a]: 0.00879267, [2]
    [Cycle 1]: 0.00195127, [27]
        [switch_simplify]: 6.036e-05
        [loop_unroll]: 3.058e-05
        [a_1]: 0.00058929
        [with_stream_mark]: 2.205e-05
        [recompute_prepare]: 7.78001e-06
        [updatestate_depend_eliminate]: 3.90998e-06
        [updatestate_assign_eliminate]: 3.13e-06
        [updatestate_loads_eliminate]: 2.79999e-06
        [parameter_eliminate]: 2.61e-06
        [specialize_transform]: 6.32001e-06
        [updatestate_useless_node_eliminater]: 5.20001e-06
        [accelerated_algorithm]: 6.07001e-06
        [meta_shard_fg_expand]: 2.73998e-06
        [get_grad_eliminate_]: 5.56e-06
        [merge_forward]: 4.68999e-06
        [cell_reuse_recompute_pass]: 2.19999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.746e-05
        [j_node_and_user_rematch]: 9.60001e-06
        [meta_fg_expand]: 2.81999e-06
        [replace_old_param]: 1.006e-05
        [inline_without_move]: 5.47001e-06
        [renormalize]: 0.0008921
        [add_forward_monad_depend]: 7.11999e-06
        [auto_monad_grad]: 3.20998e-06
        [auto_monad_eliminator]: 1.403e-05
        [cse]: 3.015e-05
        [replace_applicator]: 1.429e-05
    [Cycle 2]: 0.00286957, [27]
        [switch_simplify]: 6.34001e-06
        [loop_unroll]: 5.67999e-06
        [a_1]: 9.467e-05
        [with_stream_mark]: 1.04e-05
        [recompute_prepare]: 5.77001e-06
        [updatestate_depend_eliminate]: 3.56001e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 1.09003e-06
        [specialize_transform]: 5.37001e-06
        [updatestate_useless_node_eliminater]: 5.54e-06
        [accelerated_algorithm]: 0.00242111
        [meta_shard_fg_expand]: 3.25002e-06
        [get_grad_eliminate_]: 1.152e-05
        [merge_forward]: 1.1e-05
        [cell_reuse_recompute_pass]: 4.16001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.419e-05
        [j_node_and_user_rematch]: 9.82001e-06
        [meta_fg_expand]: 2.41e-06
        [replace_old_param]: 2.147e-05
        [inline_without_move]: 5.76e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 6.33e-06
        [auto_monad_grad]: 2.34999e-06
        [auto_monad_eliminator]: 1.192e-05
        [cse]: 2.659e-05
        [replace_applicator]: 6.32001e-06
[py_interpret_to_execute_after_opt_a]: 2.049e-05
[rewriter_after_opt_a]: 4.286e-05
[convert_after_rewriter]: 6.97002e-06
[order_py_execute_after_rewriter]: 1.169e-05
[mutable_eliminate]: 0.00085724
[jit_opt_b]: 5.979e-05, [1]
    [Cycle 1]: 5.064e-05, [2]
        [frontend_op_eliminate]: 2.033e-05
        [inline_after_opt_a]: 1.785e-05
[cconv]: 3.802e-05
[loop_unroll]: 0.00048921
[jit_opt_after_cconv]: 0.00022171, [1]
    [Cycle 1]: 0.00021426, [11]
        [c_1]: 2.349e-05
        [parameter_eliminate]: 4.96002e-06
        [updatestate_depend_eliminate]: 8.00999e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 2.91e-06
        [cse]: 2.581e-05
        [call_graph_tuple_transform]: 7.603e-05
        [tuple_list_get_item_eliminator]: 6.29001e-06
        [none_parameter_eliminate]: 1.87001e-06
        [renormalize]: 1.20999e-06
        [switch_simplify]: 6.53e-06
[remove_dup_value]: 1.609e-05
[partial_unused_args_eliminate]: 2.43998e-06
[environ_conv]: 6.77002e-06
[add_recomputation]: 5.534e-05
[cse_after_recomputation]: 2.322e-05, [1]
    [Cycle 1]: 1.709e-05, [1]
        [cse]: 1.004e-05
[auto_monad_reorder]: 1.681e-05
[get_jit_bprop_graph]: 2.36998e-06
[rewriter_after_jit_bprop_graph]: 7.38e-06
[opt_after_jit_grad]: 0.00051894
[symbol_engine_optimizer]: 8.693e-05, [1]
    [Cycle 1]: 7.838e-05, [6]
        [build]: 5.33002e-06
        [elim_shapecalc]: 8.55999e-06
        [elim_not_effective]: 1.419e-05
        [opt_reshape]: 7.51001e-06
        [fold_const_symbol]: 1.128e-05
        [renormalize]: 8.29983e-07
[validate]: 4.865e-05
[backend_pass]: 8.30012e-07
[task_emit]: 0.0100463
[execute]: 9.60001e-06
Sums
    bootstrap                            :     0.000337s :  1.32%
    type_inference                       :     0.008029s : 31.46%
    event_method                         :     0.000022s :  0.09%
    auto_monad                           :     0.000069s :  0.27%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000047s :  0.18%
    rewriter_before_opt_a                :     0.000075s :  0.29%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000067s :  0.26%
    jit_opt_a.loop_unroll                :     0.000036s :  0.14%
    jit_opt_a.a_1                        :     0.000684s :  2.68%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.002427s :  9.51%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.07%
    jit_opt_a.merge_forward              :     0.000016s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000042s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000032s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.04%
    jit_opt_a.renormalize                :     0.000892s :  3.50%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000006s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.10%
    jit_opt_a.cse                        :     0.000057s :  0.22%
    jit_opt_a.replace_applicator         :     0.000021s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000020s :  0.08%
    rewriter_after_opt_a                 :     0.000043s :  0.17%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000012s :  0.05%
    mutable_eliminate                    :     0.000857s :  3.36%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.07%
    cconv                                :     0.000038s :  0.15%
    loop_unroll                          :     0.000489s :  1.92%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000076s :  0.30%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000055s :  0.22%
    cse_after_recomputation.cse          :     0.000010s :  0.04%
    auto_monad_reorder                   :     0.000017s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.03%
    opt_after_jit_grad                   :     0.000519s :  2.03%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000049s :  0.19%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010046s : 39.37%
    execute                              :     0.000010s :  0.04%


Time group info:
------[substitution.]   0.000305    23
 0.67% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
19.28% :     0.000059s :      3: substitution.graph_param_transform
69.50% :     0.000212s :      4: substitution.inline
 1.24% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.93% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.44% :     0.000007s :      2: substitution.replace_old_param
 4.45% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007938     2
85.33% :     0.006774s :      1: type_inference.infer
14.67% :     0.001164s :      1: type_inference.specialize
------[replace.]   0.000071     6
68.98% :     0.000049s :      4: replace.inline
31.02% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000222     6
94.37% :     0.000209s :      4: match.inline
 5.63% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.71% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 1.81% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.46% :     0.000002s :     10: predicate.depend_value_elim
 1.22% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.06% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.05% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      3: predicate.fold_const_symbol
 1.19% :     0.000001s :      6: predicate.get_grad_eliminate
 0.19% :     0.000000s :      3: predicate.graph_param_transform
 5.57% :     0.000007s :     22: predicate.inline
 1.42% :     0.000002s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 3.33% :     0.000004s :      6: predicate.less_batch_normalization
 1.31% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.73% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.32% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 1.69% :     0.000002s :      3: predicate.mutable_eliminate
 0.70% :     0.000001s :      3: predicate.opt_reshape
 2.01% :     0.000003s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.80% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.32% :     0.000003s :     16: predicate.switch_defer_inline
 1.92% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.19% :     0.000009s :     53: predicate.switch_simplify
 1.08% :     0.000001s :     10: predicate.tile_eliminate
 1.07% :     0.000001s :     10: predicate.transpose_eliminate
 1.60% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.19% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.93% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.37% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.63% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.51% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000708    12
41.83% :     0.000296s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.17% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.00% :     0.000361s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.034625    76
 0.17% :     0.000059s :      1: add_recomputation
 0.21% :     0.000073s :      1: auto_monad
 0.06% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.02% :     0.000352s :      1: bootstrap
 0.12% :     0.000041s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.07% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000011s :      1: environ_conv
 0.08% :     0.000027s :      1: event_method
 0.04% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
25.41% :     0.008798s :      1: jit_opt_a
 0.65% :     0.000224s :      1: jit_opt_after_cconv
 0.18% :     0.000063s :      1: jit_opt_b
 1.44% :     0.000498s :      1: loop_unroll
 2.51% :     0.000869s :      1: mutable_eliminate
 9.68% :     0.003353s :     26: opt.transform.jit_opt_a
 0.31% :     0.000108s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000031s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.52% :     0.000528s :      1: opt_after_jit_grad
 0.04% :     0.000014s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000050s :      1: py_interpret_to_execute
 0.07% :     0.000023s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000019s :      1: remove_dup_value
 1.47% :     0.000507s :      1: renormalize.infer
 1.09% :     0.000376s :      1: renormalize.specialize
 0.03% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000047s :      1: rewriter_after_opt_a
 0.23% :     0.000078s :      1: rewriter_before_opt_a
 0.26% :     0.000090s :      1: symbol_engine_optimizer
29.07% :     0.010066s :      1: task_emit
23.25% :     0.008051s :      1: type_inference
 0.20% :     0.000071s :      1: validate


TotalTime = 0.0204447, [33]
[bootstrap]: 0.00032683
[type_inference]: 0.00494774
[event_method]: 1.711e-05
[auto_monad]: 6.341e-05
[graph_reusing]: 5.62999e-06
[pre_auto_parallel]: 2.42001e-06
[py_interpret_to_execute]: 3.462e-05
[rewriter_before_opt_a]: 6.984e-05
[expand_dump_flag]: 3.15998e-06
[jit_opt_a]: 0.00594751, [2]
    [Cycle 1]: 0.00201006, [27]
        [switch_simplify]: 5.773e-05
        [loop_unroll]: 2.88e-05
        [a_1]: 0.00057118
        [with_stream_mark]: 2.3e-05
        [recompute_prepare]: 6.83e-06
        [updatestate_depend_eliminate]: 3.86999e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [parameter_eliminate]: 1.55999e-06
        [specialize_transform]: 6.01e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.34998e-06
        [meta_shard_fg_expand]: 2.33002e-06
        [get_grad_eliminate_]: 6.41e-06
        [merge_forward]: 3.91999e-06
        [cell_reuse_recompute_pass]: 1.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.425e-05
        [j_node_and_user_rematch]: 9.43002e-06
        [meta_fg_expand]: 2.29999e-06
        [replace_old_param]: 9.74999e-06
        [inline_without_move]: 5.19e-06
        [renormalize]: 0.00097444
        [add_forward_monad_depend]: 6.04001e-06
        [auto_monad_grad]: 2.61e-06
        [auto_monad_eliminator]: 1.403e-05
        [cse]: 2.817e-05
        [replace_applicator]: 1.182e-05
    [Cycle 2]: 0.00031749, [27]
        [switch_simplify]: 5.43002e-06
        [loop_unroll]: 4.77998e-06
        [a_1]: 9.003e-05
        [with_stream_mark]: 9.39998e-06
        [recompute_prepare]: 5.05999e-06
        [updatestate_depend_eliminate]: 2.79001e-06
        [updatestate_assign_eliminate]: 1.99999e-06
        [updatestate_loads_eliminate]: 2.08002e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.72998e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 4.82e-06
        [meta_shard_fg_expand]: 1.34998e-06
        [get_grad_eliminate_]: 4.80999e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 2.02001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.245e-05
        [j_node_and_user_rematch]: 7.39002e-06
        [meta_fg_expand]: 1.42e-06
        [replace_old_param]: 7.47002e-06
        [inline_without_move]: 5.04e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.36002e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 5.32001e-06
        [cse]: 1.026e-05
        [replace_applicator]: 4.90001e-06
[py_interpret_to_execute_after_opt_a]: 1.143e-05
[rewriter_after_opt_a]: 3.751e-05
[convert_after_rewriter]: 7.76001e-06
[order_py_execute_after_rewriter]: 5.30999e-06
[mutable_eliminate]: 0.0006252
[jit_opt_b]: 5.286e-05, [1]
    [Cycle 1]: 4.508e-05, [2]
        [frontend_op_eliminate]: 1.802e-05
        [inline_after_opt_a]: 1.593e-05
[cconv]: 2.599e-05
[loop_unroll]: 0.00043762
[jit_opt_after_cconv]: 0.00013921, [1]
    [Cycle 1]: 0.00013327, [11]
        [c_1]: 2.048e-05
        [parameter_eliminate]: 2.56e-06
        [updatestate_depend_eliminate]: 5.20001e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 1.857e-05
        [call_graph_tuple_transform]: 1.886e-05
        [tuple_list_get_item_eliminator]: 5.02999e-06
        [none_parameter_eliminate]: 2.00002e-06
        [renormalize]: 7.40023e-07
        [switch_simplify]: 5.07999e-06
[remove_dup_value]: 1.324e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 5.02e-06
[add_recomputation]: 5.268e-05
[cse_after_recomputation]: 2.06e-05, [1]
    [Cycle 1]: 1.464e-05, [1]
        [cse]: 9.04e-06
[auto_monad_reorder]: 1.514e-05
[get_jit_bprop_graph]: 2.26998e-06
[rewriter_after_jit_bprop_graph]: 6.15002e-06
[opt_after_jit_grad]: 0.00046026
[symbol_engine_optimizer]: 7.24e-05, [1]
    [Cycle 1]: 6.553e-05, [6]
        [build]: 3.95998e-06
        [elim_shapecalc]: 7.18e-06
        [elim_not_effective]: 1.253e-05
        [opt_reshape]: 5.49998e-06
        [fold_const_symbol]: 8.67e-06
        [renormalize]: 4.60015e-07
[validate]: 4e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00676789
[execute]: 1.118e-05
Sums
    bootstrap                            :     0.000327s :  2.02%
    type_inference                       :     0.004948s : 30.59%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000063s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000035s :  0.21%
    rewriter_before_opt_a                :     0.000070s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.21%
    jit_opt_a.a_1                        :     0.000661s :  4.09%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000975s :  6.03%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000038s :  0.24%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000625s :  3.87%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000026s :  0.16%
    loop_unroll                          :     0.000438s :  2.71%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000053s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000460s :  2.85%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000040s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006768s : 41.84%
    execute                              :     0.000011s :  0.07%


Time group info:
------[substitution.]   0.000238    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 2.23% :     0.000005s :      3: substitution.graph_param_transform
86.00% :     0.000205s :      4: substitution.inline
 1.27% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.93% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.16% :     0.000005s :      2: substitution.replace_old_param
 5.00% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004879     2
82.05% :     0.004003s :      1: type_inference.infer
17.95% :     0.000876s :      1: type_inference.specialize
------[replace.]   0.000070     6
65.36% :     0.000046s :      4: replace.inline
34.64% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000213     6
94.94% :     0.000202s :      4: match.inline
 5.06% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 2.33% :     0.000003s :     10: predicate.arithmetic_simplify
 1.30% :     0.000002s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000001s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.22% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.54% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.41% :     0.000000s :      3: predicate.graph_param_transform
 5.72% :     0.000007s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000001s :      6: predicate.less_batch_normalization
 1.62% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.78% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.04% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.15% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.58% :     0.000002s :      3: predicate.mutable_eliminate
 0.63% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.28% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.53% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.58% :     0.000002s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.59% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.61% :     0.000009s :     53: predicate.switch_simplify
 1.28% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.40% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000667    12
45.84% :     0.000306s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.50% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.66% :     0.000311s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022385    76
 0.25% :     0.000056s :      1: add_recomputation
 0.30% :     0.000067s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.53% :     0.000342s :      1: bootstrap
 0.13% :     0.000029s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.59% :     0.005952s :      1: jit_opt_a
 0.63% :     0.000142s :      1: jit_opt_after_cconv
 0.25% :     0.000055s :      1: jit_opt_b
 1.99% :     0.000445s :      1: loop_unroll
 2.83% :     0.000633s :      1: mutable_eliminate
 3.90% :     0.000873s :     26: opt.transform.jit_opt_a
 0.21% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.09% :     0.000468s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.41% :     0.000539s :      1: renormalize.infer
 1.91% :     0.000427s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000041s :      1: rewriter_after_opt_a
 0.32% :     0.000073s :      1: rewriter_before_opt_a
 0.34% :     0.000075s :      1: symbol_engine_optimizer
30.33% :     0.006788s :      1: task_emit
22.16% :     0.004961s :      1: type_inference
 0.28% :     0.000063s :      1: validate


TotalTime = 0.0291889, [33]
[bootstrap]: 0.00041818
[type_inference]: 0.0101696
[event_method]: 1.763e-05
[auto_monad]: 6.793e-05
[graph_reusing]: 6.19001e-06
[pre_auto_parallel]: 3.73999e-06
[py_interpret_to_execute]: 4.44e-05
[rewriter_before_opt_a]: 7.721e-05
[expand_dump_flag]: 3.43999e-06
[jit_opt_a]: 0.00601036, [2]
    [Cycle 1]: 0.00188677, [27]
        [switch_simplify]: 6.075e-05
        [loop_unroll]: 2.953e-05
        [a_1]: 0.00054149
        [with_stream_mark]: 1.851e-05
        [recompute_prepare]: 7.95e-06
        [updatestate_depend_eliminate]: 4.3e-06
        [updatestate_assign_eliminate]: 3.28998e-06
        [updatestate_loads_eliminate]: 2.81999e-06
        [parameter_eliminate]: 2.22999e-06
        [specialize_transform]: 7.05e-06
        [updatestate_useless_node_eliminater]: 5.10001e-06
        [accelerated_algorithm]: 6.23e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 5.66e-06
        [merge_forward]: 3.73999e-06
        [cell_reuse_recompute_pass]: 1.92999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.639e-05
        [j_node_and_user_rematch]: 1.039e-05
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 9.06002e-06
        [inline_without_move]: 5.57001e-06
        [renormalize]: 0.00089095
        [add_forward_monad_depend]: 6.19001e-06
        [auto_monad_grad]: 1.99e-06
        [auto_monad_eliminator]: 1.533e-05
        [cse]: 2.944e-05
        [replace_applicator]: 1.362e-05
    [Cycle 2]: 0.00032478, [27]
        [switch_simplify]: 5.59e-06
        [loop_unroll]: 4.92999e-06
        [a_1]: 9.179e-05
        [with_stream_mark]: 1.044e-05
        [recompute_prepare]: 5.19e-06
        [updatestate_depend_eliminate]: 2.88e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 1.26997e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 2.53998e-06
        [cell_reuse_recompute_pass]: 1.54e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.355e-05
        [j_node_and_user_rematch]: 8.05e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 6.99001e-06
        [inline_without_move]: 4.60999e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 8.79983e-07
        [auto_monad_eliminator]: 6.12999e-06
        [cse]: 9.60001e-06
        [replace_applicator]: 4.74998e-06
[py_interpret_to_execute_after_opt_a]: 1.01e-05
[rewriter_after_opt_a]: 3.592e-05
[convert_after_rewriter]: 7.21001e-06
[order_py_execute_after_rewriter]: 5.00001e-06
[mutable_eliminate]: 0.00058878
[jit_opt_b]: 5.009e-05, [1]
    [Cycle 1]: 4.278e-05, [2]
        [frontend_op_eliminate]: 1.658e-05
        [inline_after_opt_a]: 1.494e-05
[cconv]: 2.629e-05
[loop_unroll]: 0.00043238
[jit_opt_after_cconv]: 0.00014004, [1]
    [Cycle 1]: 0.00013303, [11]
        [c_1]: 2.238e-05
        [parameter_eliminate]: 2.64999e-06
        [updatestate_depend_eliminate]: 5.74e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.76e-05
        [call_graph_tuple_transform]: 1.804e-05
        [tuple_list_get_item_eliminator]: 5.23002e-06
        [none_parameter_eliminate]: 1.46002e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 5.20999e-06
[remove_dup_value]: 1.362e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 4.89e-06
[add_recomputation]: 0.00175478
[cse_after_recomputation]: 3.135e-05, [1]
    [Cycle 1]: 2.427e-05, [1]
        [cse]: 1.468e-05
[auto_monad_reorder]: 1.693e-05
[get_jit_bprop_graph]: 1.96e-06
[rewriter_after_jit_bprop_graph]: 7.14001e-06
[opt_after_jit_grad]: 0.00055469
[symbol_engine_optimizer]: 7.927e-05, [1]
    [Cycle 1]: 7.187e-05, [6]
        [build]: 3.42997e-06
        [elim_shapecalc]: 9.27001e-06
        [elim_not_effective]: 1.332e-05
        [opt_reshape]: 6.04001e-06
        [fold_const_symbol]: 9.63002e-06
        [renormalize]: 9.20001e-07
[validate]: 3.647e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.00783632
[execute]: 1.024e-05
Sums
    bootstrap                            :     0.000418s :  1.73%
    type_inference                       :     0.010170s : 41.96%
    event_method                         :     0.000018s :  0.07%
    auto_monad                           :     0.000068s :  0.28%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000004s :  0.02%
    py_interpret_to_execute              :     0.000044s :  0.18%
    rewriter_before_opt_a                :     0.000077s :  0.32%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000066s :  0.27%
    jit_opt_a.loop_unroll                :     0.000034s :  0.14%
    jit_opt_a.a_1                        :     0.000633s :  2.61%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.04%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.07%
    jit_opt_a.inline_without_move        :     0.000010s :  0.04%
    jit_opt_a.renormalize                :     0.000891s :  3.68%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.09%
    jit_opt_a.cse                        :     0.000039s :  0.16%
    jit_opt_a.replace_applicator         :     0.000018s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.04%
    rewriter_after_opt_a                 :     0.000036s :  0.15%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000589s :  2.43%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.06%
    cconv                                :     0.000026s :  0.11%
    loop_unroll                          :     0.000432s :  1.78%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000014s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.001755s :  7.24%
    cse_after_recomputation.cse          :     0.000015s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.03%
    opt_after_jit_grad                   :     0.000555s :  2.29%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007836s : 32.33%
    execute                              :     0.000010s :  0.04%


Time group info:
------[substitution.]   0.000205    23
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.91% :     0.000002s :      2: substitution.fold_const_symbol
 2.54% :     0.000005s :      3: substitution.graph_param_transform
83.46% :     0.000171s :      4: substitution.inline
 1.73% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.18% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.92% :     0.000004s :      2: substitution.replace_old_param
 6.37% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010089     2
90.56% :     0.009136s :      1: type_inference.infer
 9.44% :     0.000953s :      1: type_inference.specialize
------[replace.]   0.000071     6
65.25% :     0.000046s :      4: replace.inline
34.75% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000180     6
93.32% :     0.000168s :      4: match.inline
 6.68% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.47% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.77% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.21% :     0.000001s :     10: predicate.addn_zero_filter
 2.45% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.56% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.50% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.70% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.50% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.39% :     0.000002s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 5.71% :     0.000007s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.79% :     0.000002s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.83% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.60% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.07% :     0.000002s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.66% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.54% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.69% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.43% :     0.000009s :     53: predicate.switch_simplify
 1.10% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.27% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.09% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.24% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.01% :     0.000002s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000724    12
53.58% :     0.000388s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.17% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.25% :     0.000306s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.031029    76
 5.68% :     0.001762s :      1: add_recomputation
 0.23% :     0.000071s :      1: auto_monad
 0.06% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.41% :     0.000439s :      1: bootstrap
 0.09% :     0.000029s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000035s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.07% :     0.000021s :      1: event_method
 0.05% :     0.000016s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 1.60% :     0.000498s :      1: graph_reusing
19.38% :     0.006014s :      1: jit_opt_a
 0.46% :     0.000143s :      1: jit_opt_after_cconv
 0.17% :     0.000053s :      1: jit_opt_b
 1.42% :     0.000441s :      1: loop_unroll
 1.93% :     0.000598s :      1: mutable_eliminate
 2.75% :     0.000853s :     26: opt.transform.jit_opt_a
 0.15% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000025s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.81% :     0.000563s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000007s :      1: pre_auto_parallel
 0.15% :     0.000047s :      1: py_interpret_to_execute
 0.04% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000016s :      1: remove_dup_value
 1.70% :     0.000527s :      1: renormalize.infer
 1.15% :     0.000356s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000041s :      1: rewriter_after_opt_a
 0.26% :     0.000080s :      1: rewriter_before_opt_a
 0.26% :     0.000082s :      1: symbol_engine_optimizer
25.32% :     0.007856s :      1: task_emit
32.82% :     0.010182s :      1: type_inference
 0.20% :     0.000061s :      1: validate


TotalTime = 0.0199897, [33]
[bootstrap]: 0.00033207
[type_inference]: 0.00481527
[event_method]: 1.742e-05
[auto_monad]: 6.218e-05
[graph_reusing]: 6.33998e-06
[pre_auto_parallel]: 2.10002e-06
[py_interpret_to_execute]: 3.967e-05
[rewriter_before_opt_a]: 7.005e-05
[expand_dump_flag]: 2.99999e-06
[jit_opt_a]: 0.00573271, [2]
    [Cycle 1]: 0.00166446, [27]
        [switch_simplify]: 4.951e-05
        [loop_unroll]: 2.936e-05
        [a_1]: 0.00052398
        [with_stream_mark]: 1.549e-05
        [recompute_prepare]: 7.11001e-06
        [updatestate_depend_eliminate]: 3.34001e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 1.95001e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 5.96998e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.57001e-06
        [meta_shard_fg_expand]: 1.90001e-06
        [get_grad_eliminate_]: 5.40999e-06
        [merge_forward]: 2.94001e-06
        [cell_reuse_recompute_pass]: 8.10018e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.132e-05
        [j_node_and_user_rematch]: 9.20001e-06
        [meta_fg_expand]: 1.86e-06
        [replace_old_param]: 8.46002e-06
        [inline_without_move]: 5.54e-06
        [renormalize]: 0.0007561
        [add_forward_monad_depend]: 5.51998e-06
        [auto_monad_grad]: 2.27999e-06
        [auto_monad_eliminator]: 1.273e-05
        [cse]: 1.704e-05
        [replace_applicator]: 1.226e-05
    [Cycle 2]: 0.00031397, [27]
        [switch_simplify]: 5.82999e-06
        [loop_unroll]: 4.95999e-06
        [a_1]: 8.887e-05
        [with_stream_mark]: 8.84e-06
        [recompute_prepare]: 5.47001e-06
        [updatestate_depend_eliminate]: 2.43e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 5.00001e-06
        [updatestate_useless_node_eliminater]: 4.55001e-06
        [accelerated_algorithm]: 5.19998e-06
        [meta_shard_fg_expand]: 1.20001e-06
        [get_grad_eliminate_]: 4.37e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 1.60999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.177e-05
        [j_node_and_user_rematch]: 7.4e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 7.19001e-06
        [inline_without_move]: 4.45999e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.00001e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 4.80001e-06
        [cse]: 9.77999e-06
        [replace_applicator]: 4.84e-06
[py_interpret_to_execute_after_opt_a]: 8.76002e-06
[rewriter_after_opt_a]: 3.069e-05
[convert_after_rewriter]: 6.93e-06
[order_py_execute_after_rewriter]: 4.47e-06
[mutable_eliminate]: 0.00114747
[jit_opt_b]: 5.017e-05, [1]
    [Cycle 1]: 4.32e-05, [2]
        [frontend_op_eliminate]: 1.658e-05
        [inline_after_opt_a]: 1.501e-05
[cconv]: 2.436e-05
[loop_unroll]: 0.00042477
[jit_opt_after_cconv]: 0.00013758, [1]
    [Cycle 1]: 0.00013062, [11]
        [c_1]: 2.027e-05
        [parameter_eliminate]: 2.54999e-06
        [updatestate_depend_eliminate]: 5.50001e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.03002e-06
        [cse]: 1.865e-05
        [call_graph_tuple_transform]: 1.763e-05
        [tuple_list_get_item_eliminator]: 5.40999e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 1.275e-05
[partial_unused_args_eliminate]: 2.36998e-06
[environ_conv]: 5.09e-06
[add_recomputation]: 3.646e-05
[cse_after_recomputation]: 1.972e-05, [1]
    [Cycle 1]: 1.454e-05, [1]
        [cse]: 8.94998e-06
[auto_monad_reorder]: 1.069e-05
[get_jit_bprop_graph]: 1.74e-06
[rewriter_after_jit_bprop_graph]: 4.65999e-06
[opt_after_jit_grad]: 0.00044591
[symbol_engine_optimizer]: 7.006e-05, [1]
    [Cycle 1]: 6.368e-05, [6]
        [build]: 2.88e-06
        [elim_shapecalc]: 7.53999e-06
        [elim_not_effective]: 1.231e-05
        [opt_reshape]: 5.60001e-06
        [fold_const_symbol]: 8.12e-06
        [renormalize]: 5.39992e-07
[validate]: 3.161e-05
[backend_pass]: 7.60017e-07
[task_emit]: 0.00622214
[execute]: 5.69e-06
Sums
    bootstrap                            :     0.000332s :  2.12%
    type_inference                       :     0.004815s : 30.79%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000062s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000040s :  0.25%
    rewriter_before_opt_a                :     0.000070s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.35%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000613s :  3.92%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000756s :  4.84%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.11%
    jit_opt_a.cse                        :     0.000027s :  0.17%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.001147s :  7.34%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000024s :  0.16%
    loop_unroll                          :     0.000425s :  2.72%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000036s :  0.23%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000011s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000446s :  2.85%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.20%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006222s : 39.79%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000193    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 2.42% :     0.000005s :      3: substitution.graph_param_transform
84.88% :     0.000164s :      4: substitution.inline
 1.66% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.08% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.89% :     0.000004s :      2: substitution.replace_old_param
 5.66% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004743     2
80.70% :     0.003827s :      1: type_inference.infer
19.30% :     0.000916s :      1: type_inference.specialize
------[replace.]   0.000068     6
65.60% :     0.000045s :      4: replace.inline
34.40% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000171     6
94.26% :     0.000161s :      4: match.inline
 5.74% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.49% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.10% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000001s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000001s :     10: predicate.depend_value_elim
 1.20% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.61% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.24% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.24% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.29% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.28% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.15% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.22% :     0.000001s :     10: predicate.minmaximum_grad
 2.06% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.07% :     0.000002s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.94% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.41% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.48% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.16% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.62% :     0.000009s :     53: predicate.switch_simplify
 1.47% :     0.000002s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.23% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000630    12
46.08% :     0.000290s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.86% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.06% :     0.000303s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021656    76
 0.18% :     0.000039s :      1: add_recomputation
 0.30% :     0.000065s :      1: auto_monad
 0.06% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.63% :     0.000353s :      1: bootstrap
 0.12% :     0.000027s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.04% :     0.000009s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.49% :     0.005737s :      1: jit_opt_a
 0.65% :     0.000140s :      1: jit_opt_after_cconv
 0.24% :     0.000053s :      1: jit_opt_b
 2.00% :     0.000433s :      1: loop_unroll
 5.35% :     0.001158s :      1: mutable_eliminate
 3.76% :     0.000814s :     26: opt.transform.jit_opt_a
 0.21% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.10% :     0.000454s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000042s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 1.88% :     0.000408s :      1: renormalize.infer
 1.57% :     0.000341s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000034s :      1: rewriter_after_opt_a
 0.34% :     0.000073s :      1: rewriter_before_opt_a
 0.34% :     0.000073s :      1: symbol_engine_optimizer
28.80% :     0.006237s :      1: task_emit
22.29% :     0.004828s :      1: type_inference
 0.23% :     0.000049s :      1: validate


TotalTime = 0.0147748, [30]
[bootstrap]: 0.00026285
[type_inference]: 0.00672485
[event_method]: 2.166e-05
[auto_monad]: 8.418e-05
[graph_reusing]: 6.17999e-06
[pre_auto_parallel]: 2.88e-06
[py_interpret_to_execute]: 4.751e-05
[rewriter_before_opt_a]: 8.529e-05
[expand_dump_flag]: 2.98e-06
[jit_opt_a]: 0.00546086, [2]
    [Cycle 1]: 0.00181096, [27]
        [switch_simplify]: 0.00012099
        [loop_unroll]: 3.412e-05
        [a_1]: 0.00068515
        [with_stream_mark]: 2.004e-05
        [recompute_prepare]: 7.45e-06
        [updatestate_depend_eliminate]: 3.53999e-06
        [updatestate_assign_eliminate]: 3.53999e-06
        [updatestate_loads_eliminate]: 2.48002e-06
        [parameter_eliminate]: 2.24001e-06
        [specialize_transform]: 5.84999e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 5.75001e-06
        [meta_shard_fg_expand]: 2.52001e-06
        [get_grad_eliminate_]: 5.24e-06
        [merge_forward]: 3.65e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.487e-05
        [j_node_and_user_rematch]: 9.15001e-06
        [meta_fg_expand]: 2.07001e-06
        [replace_old_param]: 9.21998e-06
        [inline_without_move]: 5.57999e-06
        [renormalize]: 0.00063288
        [add_forward_monad_depend]: 5.05999e-06
        [auto_monad_grad]: 1.97001e-06
        [auto_monad_eliminator]: 1.262e-05
        [cse]: 2.776e-05
        [replace_applicator]: 1.043e-05
    [Cycle 2]: 0.00030288, [27]
        [switch_simplify]: 5.48002e-06
        [loop_unroll]: 4.65999e-06
        [a_1]: 8.712e-05
        [with_stream_mark]: 8.36002e-06
        [recompute_prepare]: 4.75001e-06
        [updatestate_depend_eliminate]: 2.51e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 5.08002e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 4.60001e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.31002e-06
        [merge_forward]: 2.34999e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.193e-05
        [j_node_and_user_rematch]: 7.75998e-06
        [meta_fg_expand]: 1.41002e-06
        [replace_old_param]: 6.71e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 4.22e-06
        [cse]: 9.91e-06
        [replace_applicator]: 4.75001e-06
[py_interpret_to_execute_after_opt_a]: 9.99999e-06
[rewriter_after_opt_a]: 3.423e-05
[convert_after_rewriter]: 6.58e-06
[order_py_execute_after_rewriter]: 4.95001e-06
[mutable_eliminate]: 0.00048862
[jit_opt_b]: 5.24e-05, [1]
    [Cycle 1]: 4.032e-05, [2]
        [frontend_op_eliminate]: 1.525e-05
        [inline_after_opt_a]: 1.426e-05
[cconv]: 8.116e-05
[loop_unroll]: 0.00043988
[jit_opt_after_cconv]: 0.00013645, [1]
    [Cycle 1]: 0.00013011, [11]
        [c_1]: 2.038e-05
        [parameter_eliminate]: 2.46e-06
        [updatestate_depend_eliminate]: 5.42999e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 1.695e-05
        [call_graph_tuple_transform]: 1.797e-05
        [tuple_list_get_item_eliminator]: 4.99e-06
        [none_parameter_eliminate]: 1.56998e-06
        [renormalize]: 7.60017e-07
        [switch_simplify]: 5.22e-06
[remove_dup_value]: 1.395e-05
[partial_unused_args_eliminate]: 2.51998e-06
[environ_conv]: 4.84e-06
[add_recomputation]: 4.91e-05
[cse_after_recomputation]: 2.068e-05, [1]
    [Cycle 1]: 1.505e-05, [1]
        [cse]: 9.00999e-06
[auto_monad_reorder]: 1.447e-05
[get_jit_bprop_graph]: 1.96e-06
[rewriter_after_jit_bprop_graph]: 5.89e-06
[opt_after_jit_grad]: 0.0004347
[symbol_engine_optimizer]: 7.091e-05, [1]
    [Cycle 1]: 6.52e-05, [6]
        [build]: 3.88001e-06
        [elim_shapecalc]: 7.27002e-06
        [elim_not_effective]: 1.199e-05
        [opt_reshape]: 5.53002e-06
        [fold_const_symbol]: 7.88999e-06
        [renormalize]: 6.19999e-07
[validate]: 3.274e-05
Sums
    bootstrap                            :     0.000263s :  2.42%
    type_inference                       :     0.006725s : 61.95%
    event_method                         :     0.000022s :  0.20%
    auto_monad                           :     0.000084s :  0.78%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000048s :  0.44%
    rewriter_before_opt_a                :     0.000085s :  0.79%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000126s :  1.17%
    jit_opt_a.loop_unroll                :     0.000039s :  0.36%
    jit_opt_a.a_1                        :     0.000772s :  7.11%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000633s :  5.83%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.16%
    jit_opt_a.cse                        :     0.000038s :  0.35%
    jit_opt_a.replace_applicator         :     0.000015s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.09%
    rewriter_after_opt_a                 :     0.000034s :  0.32%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000489s :  4.50%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.13%
    cconv                                :     0.000081s :  0.75%
    loop_unroll                          :     0.000440s :  4.05%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.13%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000049s :  0.45%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000014s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000435s :  4.00%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000033s :  0.30%


Time group info:
------[substitution.]   0.000280    26
 0.66% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 1.86% :     0.000005s :      3: substitution.graph_param_transform
85.51% :     0.000239s :      6: substitution.inline
 1.02% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.51% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.42% :     0.000004s :      2: substitution.replace_old_param
 3.00% :     0.000008s :      1: substitution.switch_simplify
 4.57% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006652     2
84.40% :     0.005614s :      1: type_inference.infer
15.60% :     0.001038s :      1: type_inference.specialize
------[replace.]   0.000133     9
49.52% :     0.000066s :      6: replace.inline
28.15% :     0.000038s :      1: replace.switch_simplify
22.33% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000255     9
92.48% :     0.000236s :      6: match.inline
 2.99% :     0.000008s :      1: match.switch_simplify
 4.54% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   801
 1.38% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.96% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000001s :     12: predicate.addn_check_dump
 1.39% :     0.000002s :     12: predicate.addn_zero_filter
 1.86% :     0.000002s :     12: predicate.arithmetic_simplify
 1.40% :     0.000002s :     12: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000001s :     12: predicate.compare_switch_simplify
 1.52% :     0.000002s :     12: predicate.depend_value_elim
 1.20% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.38% :     0.000002s :     12: predicate.environ_get_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.53% :     0.000007s :     26: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.86% :     0.000002s :     17: predicate.load_eliminater
 1.32% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.28% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     12: predicate.merge_addn
 1.16% :     0.000001s :     12: predicate.minmaximum_grad
 1.48% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.47% :     0.000003s :     17: predicate.partial_eliminate
 1.28% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     12: predicate.reduce_eliminate
 1.48% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     20: predicate.replace_applicator
 0.45% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     12: predicate.reshape_eliminate
 1.31% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     12: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.65% :     0.000003s :     20: predicate.switch_defer_inline
 2.34% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.59% :     0.000011s :     65: predicate.switch_simplify
 1.30% :     0.000002s :     12: predicate.tile_eliminate
 1.46% :     0.000002s :     12: predicate.transpose_eliminate
 1.53% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.47% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     12: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000788    15
49.85% :     0.000393s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.69% :     0.000100s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.47% :     0.000295s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016505    72
 0.31% :     0.000052s :      1: add_recomputation
 0.53% :     0.000088s :      1: auto_monad
 0.10% :     0.000017s :      1: auto_monad_reorder
 1.70% :     0.000281s :      1: bootstrap
 0.51% :     0.000084s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.16% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.11% :     0.005464s :      1: jit_opt_a
 0.84% :     0.000139s :      1: jit_opt_after_cconv
 0.33% :     0.000055s :      1: jit_opt_b
 2.71% :     0.000447s :      1: loop_unroll
 3.01% :     0.000496s :      1: mutable_eliminate
 6.34% :     0.001047s :     26: opt.transform.jit_opt_a
 0.28% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000023s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.68% :     0.000442s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000050s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.15% :     0.000355s :      1: renormalize.infer
 1.64% :     0.000271s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000037s :      1: rewriter_after_opt_a
 0.53% :     0.000088s :      1: rewriter_before_opt_a
 0.44% :     0.000073s :      1: symbol_engine_optimizer
40.83% :     0.006740s :      1: type_inference


TotalTime = 0.0212531, [33]
[bootstrap]: 0.00036677
[type_inference]: 0.00532451
[event_method]: 1.973e-05
[auto_monad]: 7.154e-05
[graph_reusing]: 7.26999e-06
[pre_auto_parallel]: 2.56e-06
[py_interpret_to_execute]: 4.425e-05
[rewriter_before_opt_a]: 7.575e-05
[expand_dump_flag]: 3.37002e-06
[jit_opt_a]: 0.00618876, [2]
    [Cycle 1]: 0.00201424, [27]
        [switch_simplify]: 6.827e-05
        [loop_unroll]: 3.947e-05
        [a_1]: 0.00064699
        [with_stream_mark]: 2.485e-05
        [recompute_prepare]: 8.08001e-06
        [updatestate_depend_eliminate]: 4.05e-06
        [updatestate_assign_eliminate]: 2.99001e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 2.46e-06
        [specialize_transform]: 6.19001e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.86e-06
        [meta_shard_fg_expand]: 2.42001e-06
        [get_grad_eliminate_]: 4.95001e-06
        [merge_forward]: 3.54002e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.367e-05
        [j_node_and_user_rematch]: 8.90999e-06
        [meta_fg_expand]: 2.36e-06
        [replace_old_param]: 9.66e-06
        [inline_without_move]: 5.47999e-06
        [renormalize]: 0.00088993
        [add_forward_monad_depend]: 6.09999e-06
        [auto_monad_grad]: 2.41e-06
        [auto_monad_eliminator]: 1.464e-05
        [cse]: 3.348e-05
        [replace_applicator]: 1.186e-05
    [Cycle 2]: 0.00031762, [27]
        [switch_simplify]: 5.20001e-06
        [loop_unroll]: 4.66002e-06
        [a_1]: 8.915e-05
        [with_stream_mark]: 9.12001e-06
        [recompute_prepare]: 5.12e-06
        [updatestate_depend_eliminate]: 2.39999e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.42999e-06
        [specialize_transform]: 4.94e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 4.92e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 4.37e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.87001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.216e-05
        [j_node_and_user_rematch]: 7.95e-06
        [meta_fg_expand]: 1.81e-06
        [replace_old_param]: 7.21001e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.16997e-06
        [auto_monad_grad]: 1.45999e-06
        [auto_monad_eliminator]: 4.92e-06
        [cse]: 9.73998e-06
        [replace_applicator]: 4.92999e-06
[py_interpret_to_execute_after_opt_a]: 1.247e-05
[rewriter_after_opt_a]: 3.635e-05
[convert_after_rewriter]: 7.39002e-06
[order_py_execute_after_rewriter]: 4.76002e-06
[mutable_eliminate]: 0.00063461
[jit_opt_b]: 5.093e-05, [1]
    [Cycle 1]: 4.341e-05, [2]
        [frontend_op_eliminate]: 1.687e-05
        [inline_after_opt_a]: 1.435e-05
[cconv]: 2.899e-05
[loop_unroll]: 0.0004449
[jit_opt_after_cconv]: 0.00014558, [1]
    [Cycle 1]: 0.00013903, [11]
        [c_1]: 2.111e-05
        [parameter_eliminate]: 3.46001e-06
        [updatestate_depend_eliminate]: 6.14999e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [cse]: 2.239e-05
        [call_graph_tuple_transform]: 1.992e-05
        [tuple_list_get_item_eliminator]: 4.97e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 3.50003e-07
        [switch_simplify]: 5.15999e-06
[remove_dup_value]: 1.599e-05
[partial_unused_args_eliminate]: 2.31e-06
[environ_conv]: 5.82001e-06
[add_recomputation]: 5.513e-05
[cse_after_recomputation]: 2.061e-05, [1]
    [Cycle 1]: 1.516e-05, [1]
        [cse]: 9.14e-06
[auto_monad_reorder]: 1.727e-05
[get_jit_bprop_graph]: 3.04001e-06
[rewriter_after_jit_bprop_graph]: 7.13e-06
[opt_after_jit_grad]: 0.00054475
[symbol_engine_optimizer]: 7.708e-05, [1]
    [Cycle 1]: 6.949e-05, [6]
        [build]: 4.04002e-06
        [elim_shapecalc]: 7.43e-06
        [elim_not_effective]: 1.433e-05
        [opt_reshape]: 6.07999e-06
        [fold_const_symbol]: 9.07999e-06
        [renormalize]: 7.09988e-07
[validate]: 4.599e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.0067401
[execute]: 1.039e-05
Sums
    bootstrap                            :     0.000367s :  2.19%
    type_inference                       :     0.005325s : 31.82%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000072s :  0.43%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000044s :  0.26%
    rewriter_before_opt_a                :     0.000076s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000073s :  0.44%
    jit_opt_a.loop_unroll                :     0.000044s :  0.26%
    jit_opt_a.a_1                        :     0.000736s :  4.40%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000890s :  5.32%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000043s :  0.26%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000635s :  3.79%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000029s :  0.17%
    loop_unroll                          :     0.000445s :  2.66%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000055s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000545s :  3.26%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006740s : 40.28%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000301    23
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000002s :      2: substitution.fold_const_symbol
 1.75% :     0.000005s :      3: substitution.graph_param_transform
88.32% :     0.000266s :      4: substitution.inline
 1.03% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.53% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.59% :     0.000005s :      2: substitution.replace_old_param
 4.51% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005238     2
80.64% :     0.004224s :      1: type_inference.infer
19.36% :     0.001014s :      1: type_inference.specialize
------[replace.]   0.000077     6
68.71% :     0.000053s :      4: replace.inline
31.29% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000276     6
95.45% :     0.000264s :      4: match.inline
 4.55% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.29% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.59% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 1.92% :     0.000002s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.16% :     0.000001s :     10: predicate.depend_value_elim
 1.35% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.39% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.83% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.35% :     0.000007s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.57% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.69% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 1.79% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.51% :     0.000001s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.10% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 1.92% :     0.000002s :     16: predicate.switch_layer_defer_inline
 9.53% :     0.000012s :     53: predicate.switch_simplify
 1.52% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.13% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.34% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.64% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000727    12
49.82% :     0.000362s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.04% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.14% :     0.000314s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023257    76
 0.25% :     0.000058s :      1: add_recomputation
 0.33% :     0.000076s :      1: auto_monad
 0.08% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.67% :     0.000389s :      1: bootstrap
 0.14% :     0.000032s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.63% :     0.006192s :      1: jit_opt_a
 0.64% :     0.000149s :      1: jit_opt_after_cconv
 0.23% :     0.000054s :      1: jit_opt_b
 1.95% :     0.000453s :      1: loop_unroll
 2.77% :     0.000645s :      1: mutable_eliminate
 4.16% :     0.000967s :     26: opt.transform.jit_opt_a
 0.20% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.32% :     0.000075s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.39% :     0.000555s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000047s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000019s :      1: remove_dup_value
 2.16% :     0.000503s :      1: renormalize.infer
 1.63% :     0.000378s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000039s :      1: rewriter_after_opt_a
 0.34% :     0.000078s :      1: rewriter_before_opt_a
 0.34% :     0.000080s :      1: symbol_engine_optimizer
29.06% :     0.006758s :      1: task_emit
22.98% :     0.005344s :      1: type_inference
 0.30% :     0.000070s :      1: validate


TotalTime = 0.0222618, [33]
[bootstrap]: 0.0003536
[type_inference]: 0.00505081
[event_method]: 1.683e-05
[auto_monad]: 4.83e-05
[graph_reusing]: 3.46999e-06
[pre_auto_parallel]: 1.97999e-06
[py_interpret_to_execute]: 4.18e-05
[rewriter_before_opt_a]: 6.707e-05
[expand_dump_flag]: 2.69001e-06
[jit_opt_a]: 0.00623333, [2]
    [Cycle 1]: 0.00212071, [27]
        [switch_simplify]: 6.538e-05
        [loop_unroll]: 2.886e-05
        [a_1]: 0.00077231
        [with_stream_mark]: 2.458e-05
        [recompute_prepare]: 7.60998e-06
        [updatestate_depend_eliminate]: 4.1e-06
        [updatestate_assign_eliminate]: 4.28999e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [parameter_eliminate]: 2.01998e-06
        [specialize_transform]: 6.01e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.68002e-06
        [meta_shard_fg_expand]: 2.30002e-06
        [get_grad_eliminate_]: 5.91e-06
        [merge_forward]: 3.66999e-06
        [cell_reuse_recompute_pass]: 1.58002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.552e-05
        [j_node_and_user_rematch]: 1.009e-05
        [meta_fg_expand]: 2.16e-06
        [replace_old_param]: 1.065e-05
        [inline_without_move]: 5.53002e-06
        [renormalize]: 0.00090537
        [add_forward_monad_depend]: 5.19e-06
        [auto_monad_grad]: 1.34e-06
        [auto_monad_eliminator]: 1.27e-05
        [cse]: 1.772e-05
        [replace_applicator]: 1.289e-05
    [Cycle 2]: 0.00032081, [27]
        [switch_simplify]: 5.63002e-06
        [loop_unroll]: 5.15001e-06
        [a_1]: 9.007e-05
        [with_stream_mark]: 8.83001e-06
        [recompute_prepare]: 5.05999e-06
        [updatestate_depend_eliminate]: 2.63998e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 5.05999e-06
        [updatestate_useless_node_eliminater]: 4.80999e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 1.55999e-06
        [get_grad_eliminate_]: 4.48999e-06
        [merge_forward]: 2.30002e-06
        [cell_reuse_recompute_pass]: 2.06e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.252e-05
        [j_node_and_user_rematch]: 8.03001e-06
        [meta_fg_expand]: 1.45999e-06
        [replace_old_param]: 7.77e-06
        [inline_without_move]: 4.63001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.11002e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 5.28002e-06
        [cse]: 1.013e-05
        [replace_applicator]: 4.84e-06
[py_interpret_to_execute_after_opt_a]: 9.86e-06
[rewriter_after_opt_a]: 2.868e-05
[convert_after_rewriter]: 6.68998e-06
[order_py_execute_after_rewriter]: 3.82998e-06
[mutable_eliminate]: 0.00066146
[jit_opt_b]: 5.231e-05, [1]
    [Cycle 1]: 4.511e-05, [2]
        [frontend_op_eliminate]: 1.783e-05
        [inline_after_opt_a]: 1.504e-05
[cconv]: 1.838e-05
[loop_unroll]: 0.0004646
[jit_opt_after_cconv]: 0.00014324, [1]
    [Cycle 1]: 0.00013692, [11]
        [c_1]: 2.207e-05
        [parameter_eliminate]: 3.03e-06
        [updatestate_depend_eliminate]: 6.24001e-06
        [updatestate_assign_eliminate]: 2.63998e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [cse]: 1.988e-05
        [call_graph_tuple_transform]: 1.869e-05
        [tuple_list_get_item_eliminator]: 5.40999e-06
        [none_parameter_eliminate]: 9.49978e-07
        [renormalize]: 6.60017e-07
        [switch_simplify]: 5.12e-06
[remove_dup_value]: 7.46999e-06
[partial_unused_args_eliminate]: 2.29001e-06
[environ_conv]: 4.22998e-06
[add_recomputation]: 3.459e-05
[cse_after_recomputation]: 2.168e-05, [1]
    [Cycle 1]: 1.579e-05, [1]
        [cse]: 9.14e-06
[auto_monad_reorder]: 9.96e-06
[get_jit_bprop_graph]: 1.76e-06
[rewriter_after_jit_bprop_graph]: 6.96999e-06
[opt_after_jit_grad]: 0.00052543
[symbol_engine_optimizer]: 7.811e-05, [1]
    [Cycle 1]: 7.11e-05, [6]
        [build]: 3.54002e-06
        [elim_shapecalc]: 8.37e-06
        [elim_not_effective]: 1.342e-05
        [opt_reshape]: 6.09001e-06
        [fold_const_symbol]: 9.22001e-06
        [renormalize]: 5.00004e-07
[validate]: 3.392e-05
[backend_pass]: 9.49978e-07
[task_emit]: 0.00809375
[execute]: 1.026e-05
Sums
    bootstrap                            :     0.000354s :  1.98%
    type_inference                       :     0.005051s : 28.33%
    event_method                         :     0.000017s :  0.09%
    auto_monad                           :     0.000048s :  0.27%
    graph_reusing                        :     0.000003s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.23%
    rewriter_before_opt_a                :     0.000067s :  0.38%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000071s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.19%
    jit_opt_a.a_1                        :     0.000862s :  4.84%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000905s :  5.08%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.10%
    jit_opt_a.cse                        :     0.000028s :  0.16%
    jit_opt_a.replace_applicator         :     0.000018s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000029s :  0.16%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000661s :  3.71%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000018s :  0.10%
    loop_unroll                          :     0.000465s :  2.61%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000007s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000035s :  0.19%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000010s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000525s :  2.95%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.19%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008094s : 45.40%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000252    23
 1.08% :     0.000003s :      2: substitution.elim_not_effective
 0.35% :     0.000001s :      2: substitution.fold_const_symbol
 1.53% :     0.000004s :      3: substitution.graph_param_transform
86.21% :     0.000217s :      4: substitution.inline
 1.45% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.86% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.99% :     0.000005s :      2: substitution.replace_old_param
 5.53% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004978     2
81.73% :     0.004069s :      1: type_inference.infer
18.27% :     0.000909s :      1: type_inference.specialize
------[replace.]   0.000082     6
67.19% :     0.000055s :      4: replace.inline
32.81% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000227     6
94.38% :     0.000215s :      4: match.inline
 5.62% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000281   691
 0.68% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.55% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.42% :     0.000001s :     10: predicate.addn_check_dump
 0.52% :     0.000001s :     10: predicate.addn_zero_filter
 0.92% :     0.000003s :     10: predicate.arithmetic_simplify
 0.52% :     0.000001s :     10: predicate.cast_eliminate
 0.22% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.44% :     0.000001s :     10: predicate.compare_switch_simplify
 0.61% :     0.000002s :     10: predicate.depend_value_elim
 0.46% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 0.58% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.48% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.40% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.14% :     0.000000s :      3: predicate.elim_not_effective
 0.37% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.53% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.45% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.48% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.54% :     0.000002s :     10: predicate.environ_get_eliminate
 0.44% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      3: predicate.fold_const_symbol
 0.42% :     0.000001s :      6: predicate.get_grad_eliminate
 0.10% :     0.000000s :      3: predicate.graph_param_transform
 2.39% :     0.000007s :     22: predicate.inline
 0.38% :     0.000001s :      6: predicate.inline_without_move
 0.16% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.49% :     0.000001s :      6: predicate.less_batch_normalization
 0.67% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 0.74% :     0.000002s :     15: predicate.load_eliminater
 0.51% :     0.000001s :      3: predicate.loop_unroll_after_grad
 1.60% :     0.000005s :     34: predicate.loop_unroll_before_grad
 0.81% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.45% :     0.000001s :     10: predicate.merge_addn
 0.45% :     0.000001s :     10: predicate.minmaximum_grad
 0.74% :     0.000002s :      3: predicate.mutable_eliminate
 0.18% :     0.000000s :      3: predicate.opt_reshape
 1.04% :     0.000003s :     15: predicate.partial_eliminate
 0.52% :     0.000001s :     10: predicate.print_const_string_wrapper
 0.76% :     0.000002s :     10: predicate.reduce_eliminate
 0.60% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.24% :     0.000001s :      6: predicate.remove_not_recompute_node
 0.73% :     0.000002s :     18: predicate.replace_applicator
 0.37% :     0.000001s :      6: predicate.replace_old_param
 0.14% :     0.000000s :      3: predicate.reset_defer_inline
 0.54% :     0.000002s :     10: predicate.reshape_eliminate
 0.53% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.39% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.54% :     0.000002s :     10: predicate.same_eliminate
 0.22% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.38% :     0.000001s :      6: predicate.special_op_eliminate
 0.41% :     0.000001s :      6: predicate.specialize_transform
 0.58% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 0.58% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.17% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 1.04% :     0.000003s :     16: predicate.switch_defer_inline
 0.86% :     0.000002s :     16: predicate.switch_layer_defer_inline
 3.26% :     0.000009s :     53: predicate.switch_simplify
 0.60% :     0.000002s :     10: predicate.tile_eliminate
 0.53% :     0.000001s :     10: predicate.transpose_eliminate
 0.74% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.48% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 1.64% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 0.79% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 0.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 0.77% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
58.98% :     0.000166s :     21: predicate.updatestate_useless_node_eliminater
 0.63% :     0.000002s :     10: predicate.value_based_eliminate
 0.13% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.30% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000666    12
46.61% :     0.000311s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.56% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.83% :     0.000319s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024351    76
 0.15% :     0.000038s :      1: add_recomputation
 0.22% :     0.000052s :      1: auto_monad
 0.05% :     0.000012s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.53% :     0.000373s :      1: bootstrap
 0.09% :     0.000021s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000021s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000006s :      1: graph_reusing
25.62% :     0.006238s :      1: jit_opt_a
 0.60% :     0.000146s :      1: jit_opt_after_cconv
 0.23% :     0.000055s :      1: jit_opt_b
 1.95% :     0.000474s :      1: loop_unroll
 2.75% :     0.000671s :      1: mutable_eliminate
 4.45% :     0.001084s :     26: opt.transform.jit_opt_a
 0.20% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.20% :     0.000535s :      1: opt_after_jit_grad
 0.02% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000045s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000010s :      1: remove_dup_value
 2.00% :     0.000487s :      1: renormalize.infer
 1.68% :     0.000410s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000032s :      1: rewriter_after_opt_a
 0.29% :     0.000070s :      1: rewriter_before_opt_a
 0.33% :     0.000081s :      1: symbol_engine_optimizer
33.31% :     0.008111s :      1: task_emit
20.81% :     0.005066s :      1: type_inference
 0.22% :     0.000052s :      1: validate


TotalTime = 0.0134449, [30]
[bootstrap]: 0.00024651
[type_inference]: 0.00585865
[event_method]: 2.004e-05
[auto_monad]: 7.956e-05
[graph_reusing]: 6.48998e-06
[pre_auto_parallel]: 2.62001e-06
[py_interpret_to_execute]: 4.533e-05
[rewriter_before_opt_a]: 8.363e-05
[expand_dump_flag]: 3.36999e-06
[jit_opt_a]: 0.00496872, [2]
    [Cycle 1]: 0.00170493, [27]
        [switch_simplify]: 0.00010543
        [loop_unroll]: 3.421e-05
        [a_1]: 0.00066767
        [with_stream_mark]: 1.573e-05
        [recompute_prepare]: 7.5e-06
        [updatestate_depend_eliminate]: 3.39001e-06
        [updatestate_assign_eliminate]: 3.33e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 2.02999e-06
        [specialize_transform]: 5.85002e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 5.55001e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 5.79e-06
        [merge_forward]: 3.43e-06
        [cell_reuse_recompute_pass]: 9.89996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.459e-05
        [j_node_and_user_rematch]: 8.97e-06
        [meta_fg_expand]: 2.26998e-06
        [replace_old_param]: 8.44002e-06
        [inline_without_move]: 5.33002e-06
        [renormalize]: 0.00055177
        [add_forward_monad_depend]: 5.98998e-06
        [auto_monad_grad]: 2.49999e-06
        [auto_monad_eliminator]: 1.369e-05
        [cse]: 4.184e-05
        [replace_applicator]: 1.284e-05
    [Cycle 2]: 0.0003229, [27]
        [switch_simplify]: 5.94999e-06
        [loop_unroll]: 4.96002e-06
        [a_1]: 8.809e-05
        [with_stream_mark]: 1.015e-05
        [recompute_prepare]: 5.39e-06
        [updatestate_depend_eliminate]: 3.33e-06
        [updatestate_assign_eliminate]: 1.94999e-06
        [updatestate_loads_eliminate]: 1.77001e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 4.95999e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 4.70999e-06
        [meta_shard_fg_expand]: 1.54e-06
        [get_grad_eliminate_]: 4.58999e-06
        [merge_forward]: 2.58998e-06
        [cell_reuse_recompute_pass]: 1.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.353e-05
        [j_node_and_user_rematch]: 8e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 6.73e-06
        [inline_without_move]: 4.92e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.95001e-06
        [auto_monad_grad]: 1.44e-06
        [auto_monad_eliminator]: 5.52001e-06
        [cse]: 1.109e-05
        [replace_applicator]: 5.25001e-06
[py_interpret_to_execute_after_opt_a]: 1.097e-05
[rewriter_after_opt_a]: 3.686e-05
[convert_after_rewriter]: 7.18e-06
[order_py_execute_after_rewriter]: 5.35001e-06
[mutable_eliminate]: 0.00054858
[jit_opt_b]: 5.785e-05, [1]
    [Cycle 1]: 4.313e-05, [2]
        [frontend_op_eliminate]: 1.68e-05
        [inline_after_opt_a]: 1.531e-05
[cconv]: 3.02e-05
[loop_unroll]: 0.00045769
[jit_opt_after_cconv]: 0.00014612, [1]
    [Cycle 1]: 0.0001389, [11]
        [c_1]: 2.049e-05
        [parameter_eliminate]: 3.36001e-06
        [updatestate_depend_eliminate]: 7.07002e-06
        [updatestate_assign_eliminate]: 2.50002e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [cse]: 2.207e-05
        [call_graph_tuple_transform]: 1.878e-05
        [tuple_list_get_item_eliminator]: 5.00001e-06
        [none_parameter_eliminate]: 1.31002e-06
        [renormalize]: 1.00999e-06
        [switch_simplify]: 5.52001e-06
[remove_dup_value]: 1.44e-05
[partial_unused_args_eliminate]: 2.06998e-06
[environ_conv]: 5.77001e-06
[add_recomputation]: 4.923e-05
[cse_after_recomputation]: 2.122e-05, [1]
    [Cycle 1]: 1.55e-05, [1]
        [cse]: 9.55001e-06
[auto_monad_reorder]: 1.436e-05
[get_jit_bprop_graph]: 2.13002e-06
[rewriter_after_jit_bprop_graph]: 6.43998e-06
[opt_after_jit_grad]: 0.00044585
[symbol_engine_optimizer]: 7.083e-05, [1]
    [Cycle 1]: 6.47e-05, [6]
        [build]: 3.63e-06
        [elim_shapecalc]: 7.25e-06
        [elim_not_effective]: 1.21e-05
        [opt_reshape]: 5.57001e-06
        [fold_const_symbol]: 8.12e-06
        [renormalize]: 4.69998e-07
[validate]: 3.494e-05
Sums
    bootstrap                            :     0.000247s :  2.48%
    type_inference                       :     0.005859s : 58.98%
    event_method                         :     0.000020s :  0.20%
    auto_monad                           :     0.000080s :  0.80%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000045s :  0.46%
    rewriter_before_opt_a                :     0.000084s :  0.84%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000111s :  1.12%
    jit_opt_a.loop_unroll                :     0.000039s :  0.39%
    jit_opt_a.a_1                        :     0.000756s :  7.61%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000552s :  5.56%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.19%
    jit_opt_a.cse                        :     0.000053s :  0.53%
    jit_opt_a.replace_applicator         :     0.000018s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.11%
    rewriter_after_opt_a                 :     0.000037s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000549s :  5.52%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.15%
    cconv                                :     0.000030s :  0.30%
    loop_unroll                          :     0.000458s :  4.61%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000014s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000049s :  0.50%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000014s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000446s :  4.49%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.35%


Time group info:
------[substitution.]   0.000247    26
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 2.23% :     0.000006s :      3: substitution.graph_param_transform
83.64% :     0.000206s :      6: substitution.inline
 1.36% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.91% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.41% :     0.000003s :      2: substitution.replace_old_param
 2.95% :     0.000007s :      1: substitution.switch_simplify
 5.27% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005803     2
83.35% :     0.004837s :      1: type_inference.infer
16.65% :     0.000966s :      1: type_inference.specialize
------[replace.]   0.000123     9
51.43% :     0.000063s :      6: replace.inline
25.68% :     0.000032s :      1: replace.switch_simplify
22.88% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000221     9
91.69% :     0.000203s :      6: match.inline
 2.96% :     0.000007s :      1: match.switch_simplify
 5.34% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   801
 1.37% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     12: predicate.addn_check_dump
 1.47% :     0.000002s :     12: predicate.addn_zero_filter
 1.82% :     0.000002s :     12: predicate.arithmetic_simplify
 1.34% :     0.000002s :     12: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.26% :     0.000002s :     12: predicate.compare_switch_simplify
 1.57% :     0.000002s :     12: predicate.depend_value_elim
 1.20% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     12: predicate.environ_get_eliminate
 1.15% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.66% :     0.000007s :     26: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.97% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     17: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.15% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.98% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     12: predicate.merge_addn
 1.43% :     0.000002s :     12: predicate.minmaximum_grad
 2.07% :     0.000003s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     17: predicate.partial_eliminate
 1.26% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     12: predicate.reduce_eliminate
 1.45% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     20: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     12: predicate.reshape_eliminate
 1.36% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.66% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     12: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.79% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.64% :     0.000003s :     20: predicate.switch_defer_inline
 2.46% :     0.000003s :     20: predicate.switch_layer_defer_inline
 7.94% :     0.000010s :     65: predicate.switch_simplify
 1.28% :     0.000002s :     12: predicate.tile_eliminate
 1.26% :     0.000002s :     12: predicate.transpose_eliminate
 1.39% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.60% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     12: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000713    15
51.23% :     0.000365s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.78% :     0.000077s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.00% :     0.000271s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015077    72
 0.35% :     0.000052s :      1: add_recomputation
 0.55% :     0.000083s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.73% :     0.000261s :      1: bootstrap
 0.22% :     0.000033s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
32.98% :     0.004972s :      1: jit_opt_a
 0.99% :     0.000149s :      1: jit_opt_after_cconv
 0.40% :     0.000061s :      1: jit_opt_b
 3.09% :     0.000465s :      1: loop_unroll
 3.71% :     0.000559s :      1: mutable_eliminate
 6.77% :     0.001021s :     26: opt.transform.jit_opt_a
 0.31% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.01% :     0.000453s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.32% :     0.000048s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 1.99% :     0.000300s :      1: renormalize.infer
 1.62% :     0.000245s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000041s :      1: rewriter_after_opt_a
 0.57% :     0.000087s :      1: rewriter_before_opt_a
 0.49% :     0.000073s :      1: symbol_engine_optimizer
38.93% :     0.005870s :      1: type_inference


TotalTime = 0.0280263, [33]
[bootstrap]: 0.00034025
[type_inference]: 0.00768113
[event_method]: 2.397e-05
[auto_monad]: 7.529e-05
[graph_reusing]: 6.24999e-06
[pre_auto_parallel]: 2.49001e-06
[py_interpret_to_execute]: 5.294e-05
[rewriter_before_opt_a]: 8.229e-05
[expand_dump_flag]: 3.43e-06
[jit_opt_a]: 0.00838524, [2]
    [Cycle 1]: 0.0020731, [27]
        [switch_simplify]: 6.93e-05
        [loop_unroll]: 2.936e-05
        [a_1]: 0.00062546
        [with_stream_mark]: 2.481e-05
        [recompute_prepare]: 6.94999e-06
        [updatestate_depend_eliminate]: 4.33001e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.79001e-06
        [parameter_eliminate]: 2.09999e-06
        [specialize_transform]: 5.69e-06
        [updatestate_useless_node_eliminater]: 4.72998e-06
        [accelerated_algorithm]: 5.52999e-06
        [meta_shard_fg_expand]: 2.00002e-06
        [get_grad_eliminate_]: 5.75001e-06
        [merge_forward]: 3.85e-06
        [cell_reuse_recompute_pass]: 1.73997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.485e-05
        [j_node_and_user_rematch]: 9.15999e-06
        [meta_fg_expand]: 2.38002e-06
        [replace_old_param]: 1.024e-05
        [inline_without_move]: 5.42999e-06
        [renormalize]: 0.0009809
        [add_forward_monad_depend]: 5.86e-06
        [auto_monad_grad]: 3.07002e-06
        [auto_monad_eliminator]: 1.47e-05
        [cse]: 3.144e-05
        [replace_applicator]: 1.297e-05
    [Cycle 2]: 0.00032534, [27]
        [switch_simplify]: 5.56e-06
        [loop_unroll]: 5.54e-06
        [a_1]: 9.099e-05
        [with_stream_mark]: 9.84999e-06
        [recompute_prepare]: 5.23002e-06
        [updatestate_depend_eliminate]: 2.59001e-06
        [updatestate_assign_eliminate]: 2.53003e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 1.14998e-06
        [specialize_transform]: 4.73001e-06
        [updatestate_useless_node_eliminater]: 4.68001e-06
        [accelerated_algorithm]: 4.95001e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 4.32e-06
        [merge_forward]: 2.98e-06
        [cell_reuse_recompute_pass]: 1.87001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.302e-05
        [j_node_and_user_rematch]: 7.82e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 7.4e-06
        [inline_without_move]: 4.58001e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 1.32e-06
        [auto_monad_eliminator]: 5.37999e-06
        [cse]: 1.07e-05
        [replace_applicator]: 4.89003e-06
[py_interpret_to_execute_after_opt_a]: 1.425e-05
[rewriter_after_opt_a]: 3.784e-05
[convert_after_rewriter]: 7.15e-06
[order_py_execute_after_rewriter]: 5.24e-06
[mutable_eliminate]: 0.00074755
[jit_opt_b]: 9.817e-05, [1]
    [Cycle 1]: 8.841e-05, [2]
        [frontend_op_eliminate]: 1.801e-05
        [inline_after_opt_a]: 5.435e-05
[cconv]: 3.31e-05
[loop_unroll]: 0.00049139
[jit_opt_after_cconv]: 0.00015266, [1]
    [Cycle 1]: 0.00014586, [11]
        [c_1]: 2.121e-05
        [parameter_eliminate]: 4.08001e-06
        [updatestate_depend_eliminate]: 6.96001e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.89999e-06
        [cse]: 2.415e-05
        [call_graph_tuple_transform]: 2.11e-05
        [tuple_list_get_item_eliminator]: 5.13002e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 5.19998e-06
[remove_dup_value]: 1.627e-05
[partial_unused_args_eliminate]: 2.69001e-06
[environ_conv]: 6.39999e-06
[add_recomputation]: 5.645e-05
[cse_after_recomputation]: 2.193e-05, [1]
    [Cycle 1]: 1.616e-05, [1]
        [cse]: 1.022e-05
[auto_monad_reorder]: 1.633e-05
[get_jit_bprop_graph]: 2.63e-06
[rewriter_after_jit_bprop_graph]: 8.62e-06
[opt_after_jit_grad]: 0.00048553
[symbol_engine_optimizer]: 7.565e-05, [1]
    [Cycle 1]: 6.818e-05, [6]
        [build]: 4.87e-06
        [elim_shapecalc]: 7.9e-06
        [elim_not_effective]: 1.293e-05
        [opt_reshape]: 5.67001e-06
        [fold_const_symbol]: 8.2e-06
        [renormalize]: 4.89992e-07
[validate]: 4.707e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.0087759
[execute]: 1.089e-05
Sums
    bootstrap                            :     0.000340s :  1.59%
    type_inference                       :     0.007681s : 35.98%
    event_method                         :     0.000024s :  0.11%
    auto_monad                           :     0.000075s :  0.35%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000053s :  0.25%
    rewriter_before_opt_a                :     0.000082s :  0.39%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000075s :  0.35%
    jit_opt_a.loop_unroll                :     0.000035s :  0.16%
    jit_opt_a.a_1                        :     0.000716s :  3.36%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000981s :  4.60%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.09%
    jit_opt_a.cse                        :     0.000042s :  0.20%
    jit_opt_a.replace_applicator         :     0.000018s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.18%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000748s :  3.50%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000054s :  0.25%
    cconv                                :     0.000033s :  0.16%
    loop_unroll                          :     0.000491s :  2.30%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000056s :  0.26%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.04%
    opt_after_jit_grad                   :     0.000486s :  2.27%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000047s :  0.22%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008776s : 41.11%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000280    23
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 1.95% :     0.000005s :      3: substitution.graph_param_transform
87.09% :     0.000244s :      4: substitution.inline
 1.24% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.77% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.64% :     0.000005s :      2: substitution.replace_old_param
 5.14% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007588     2
85.01% :     0.006450s :      1: type_inference.infer
14.99% :     0.001138s :      1: type_inference.specialize
------[replace.]   0.000079     6
66.98% :     0.000053s :      4: replace.inline
33.02% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000254     6
94.75% :     0.000241s :      4: match.inline
 5.25% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000153   691
 1.08% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.93% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.79% :     0.000001s :     10: predicate.addn_check_dump
 0.94% :     0.000001s :     10: predicate.addn_zero_filter
 1.64% :     0.000003s :     10: predicate.arithmetic_simplify
 1.08% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.80% :     0.000001s :     10: predicate.compare_switch_simplify
 0.96% :     0.000001s :     10: predicate.depend_value_elim
 0.91% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.08% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.89% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.68% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.79% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.80% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.86% :     0.000001s :     10: predicate.environ_get_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.16% :     0.000000s :      3: predicate.fold_const_symbol
 0.81% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
26.63% :     0.000041s :     22: predicate.inline
 0.67% :     0.000001s :      6: predicate.inline_without_move
 0.27% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.83% :     0.000001s :      6: predicate.less_batch_normalization
 1.20% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.30% :     0.000002s :     15: predicate.load_eliminater
 1.15% :     0.000002s :      3: predicate.loop_unroll_after_grad
 2.92% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.75% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.82% :     0.000001s :     10: predicate.merge_addn
 0.80% :     0.000001s :     10: predicate.minmaximum_grad
 1.60% :     0.000002s :      3: predicate.mutable_eliminate
 0.38% :     0.000001s :      3: predicate.opt_reshape
 1.68% :     0.000003s :     15: predicate.partial_eliminate
 0.91% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.38% :     0.000002s :     10: predicate.reduce_eliminate
 1.07% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.37% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.26% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.13% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.12% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.03% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.76% :     0.000001s :      6: predicate.special_op_eliminate
 0.74% :     0.000001s :      6: predicate.specialize_transform
 1.07% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.32% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 1.65% :     0.000003s :     16: predicate.switch_defer_inline
 1.55% :     0.000002s :     16: predicate.switch_layer_defer_inline
 5.98% :     0.000009s :     53: predicate.switch_simplify
 0.98% :     0.000002s :     10: predicate.tile_eliminate
 1.02% :     0.000002s :     10: predicate.transpose_eliminate
 1.04% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.89% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 2.97% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.21% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.23% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.03% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.52% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000814    12
48.77% :     0.000397s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.67% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.56% :     0.000363s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.030074    76
 0.20% :     0.000060s :      1: add_recomputation
 0.26% :     0.000079s :      1: auto_monad
 0.06% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.21% :     0.000364s :      1: bootstrap
 0.12% :     0.000036s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.10% :     0.000030s :      1: event_method
 0.06% :     0.000017s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
27.89% :     0.008389s :      1: jit_opt_a
 0.52% :     0.000155s :      1: jit_opt_after_cconv
 0.34% :     0.000101s :      1: jit_opt_b
 1.66% :     0.000500s :      1: loop_unroll
 2.52% :     0.000758s :      1: mutable_eliminate
 3.12% :     0.000939s :     26: opt.transform.jit_opt_a
 0.17% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000063s :      4: opt.transform.jit_opt_b
 0.05% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.64% :     0.000494s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000056s :      1: py_interpret_to_execute
 0.05% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000019s :      1: remove_dup_value
 1.81% :     0.000543s :      1: renormalize.infer
 1.43% :     0.000429s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000041s :      1: rewriter_after_opt_a
 0.29% :     0.000086s :      1: rewriter_before_opt_a
 0.26% :     0.000078s :      1: symbol_engine_optimizer
29.26% :     0.008799s :      1: task_emit
25.63% :     0.007707s :      1: type_inference
 0.23% :     0.000071s :      1: validate


TotalTime = 0.0146214, [30]
[bootstrap]: 0.00024605
[type_inference]: 0.00643542
[event_method]: 2.196e-05
[auto_monad]: 8.315e-05
[graph_reusing]: 6.32001e-06
[pre_auto_parallel]: 2.81e-06
[py_interpret_to_execute]: 5.153e-05
[rewriter_before_opt_a]: 8.522e-05
[expand_dump_flag]: 3.13998e-06
[jit_opt_a]: 0.00570872, [2]
    [Cycle 1]: 0.00186209, [27]
        [switch_simplify]: 0.00012701
        [loop_unroll]: 3.34e-05
        [a_1]: 0.00069651
        [with_stream_mark]: 1.973e-05
        [recompute_prepare]: 7.28e-06
        [updatestate_depend_eliminate]: 4.38001e-06
        [updatestate_assign_eliminate]: 3.48999e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 5.76998e-06
        [updatestate_useless_node_eliminater]: 5.08002e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.97001e-06
        [get_grad_eliminate_]: 5.22999e-06
        [merge_forward]: 3.71001e-06
        [cell_reuse_recompute_pass]: 1.03001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.373e-05
        [j_node_and_user_rematch]: 9.42999e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 8.63001e-06
        [inline_without_move]: 5.39998e-06
        [renormalize]: 0.00065983
        [add_forward_monad_depend]: 4.67e-06
        [auto_monad_grad]: 2.33002e-06
        [auto_monad_eliminator]: 1.369e-05
        [cse]: 2.861e-05
        [replace_applicator]: 1.124e-05
    [Cycle 2]: 0.00030355, [27]
        [switch_simplify]: 5.73002e-06
        [loop_unroll]: 4.74e-06
        [a_1]: 8.704e-05
        [with_stream_mark]: 8.36002e-06
        [recompute_prepare]: 5.29e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 1.50001e-06
        [specialize_transform]: 4.82998e-06
        [updatestate_useless_node_eliminater]: 4.70999e-06
        [accelerated_algorithm]: 4.58999e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.31002e-06
        [merge_forward]: 2.22999e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.183e-05
        [j_node_and_user_rematch]: 7.63001e-06
        [meta_fg_expand]: 1.56998e-06
        [replace_old_param]: 6.91999e-06
        [inline_without_move]: 4.52e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 6.99976e-07
        [auto_monad_eliminator]: 4.97e-06
        [cse]: 9.14e-06
        [replace_applicator]: 4.76002e-06
[py_interpret_to_execute_after_opt_a]: 9.58997e-06
[rewriter_after_opt_a]: 3.234e-05
[convert_after_rewriter]: 6.96001e-06
[order_py_execute_after_rewriter]: 4.67998e-06
[mutable_eliminate]: 0.00048175
[jit_opt_b]: 4.605e-05, [1]
    [Cycle 1]: 3.973e-05, [2]
        [frontend_op_eliminate]: 1.499e-05
        [inline_after_opt_a]: 1.37e-05
[cconv]: 3.067e-05
[loop_unroll]: 0.00042195
[jit_opt_after_cconv]: 0.0001319, [1]
    [Cycle 1]: 0.00012559, [11]
        [c_1]: 2.069e-05
        [parameter_eliminate]: 2.46e-06
        [updatestate_depend_eliminate]: 5.15999e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.545e-05
        [call_graph_tuple_transform]: 1.699e-05
        [tuple_list_get_item_eliminator]: 5.02999e-06
        [none_parameter_eliminate]: 1.40999e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 5.34998e-06
[remove_dup_value]: 1.272e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 4.35e-06
[add_recomputation]: 4.418e-05
[cse_after_recomputation]: 1.964e-05, [1]
    [Cycle 1]: 1.442e-05, [1]
        [cse]: 8.82e-06
[auto_monad_reorder]: 1.48e-05
[get_jit_bprop_graph]: 1.87999e-06
[rewriter_after_jit_bprop_graph]: 4.57998e-06
[opt_after_jit_grad]: 0.00043576
[symbol_engine_optimizer]: 7.118e-05, [1]
    [Cycle 1]: 6.526e-05, [6]
        [build]: 3.8e-06
        [elim_shapecalc]: 7.53999e-06
        [elim_not_effective]: 1.219e-05
        [opt_reshape]: 5.22999e-06
        [fold_const_symbol]: 8.40001e-06
        [renormalize]: 4.40021e-07
[validate]: 3.217e-05
Sums
    bootstrap                            :     0.000246s :  2.34%
    type_inference                       :     0.006435s : 61.25%
    event_method                         :     0.000022s :  0.21%
    auto_monad                           :     0.000083s :  0.79%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000052s :  0.49%
    rewriter_before_opt_a                :     0.000085s :  0.81%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000133s :  1.26%
    jit_opt_a.loop_unroll                :     0.000038s :  0.36%
    jit_opt_a.a_1                        :     0.000784s :  7.46%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000660s :  6.28%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.18%
    jit_opt_a.cse                        :     0.000038s :  0.36%
    jit_opt_a.replace_applicator         :     0.000016s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.09%
    rewriter_after_opt_a                 :     0.000032s :  0.31%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000482s :  4.58%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.13%
    cconv                                :     0.000031s :  0.29%
    loop_unroll                          :     0.000422s :  4.02%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000044s :  0.42%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000436s :  4.15%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.31%


Time group info:
------[substitution.]   0.000291    26
 0.65% :     0.000002s :      2: substitution.elim_not_effective
 0.41% :     0.000001s :      2: substitution.fold_const_symbol
 1.59% :     0.000005s :      3: substitution.graph_param_transform
86.21% :     0.000251s :      6: substitution.inline
 1.21% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.42% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.35% :     0.000004s :      2: substitution.replace_old_param
 2.89% :     0.000008s :      1: substitution.switch_simplify
 4.27% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006363     2
82.82% :     0.005270s :      1: type_inference.infer
17.18% :     0.001093s :      1: type_inference.specialize
------[replace.]   0.000138     9
48.46% :     0.000067s :      6: replace.inline
29.87% :     0.000041s :      1: replace.switch_simplify
21.68% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000267     9
92.91% :     0.000248s :      6: match.inline
 2.87% :     0.000008s :      1: match.switch_simplify
 4.23% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   801
 1.43% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.18% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     12: predicate.addn_check_dump
 1.43% :     0.000002s :     12: predicate.addn_zero_filter
 1.86% :     0.000002s :     12: predicate.arithmetic_simplify
 1.36% :     0.000002s :     12: predicate.cast_eliminate
 0.55% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     12: predicate.compare_switch_simplify
 1.54% :     0.000002s :     12: predicate.depend_value_elim
 1.20% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     12: predicate.environ_get_eliminate
 1.33% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.74% :     0.000007s :     26: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.99% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     17: predicate.load_eliminater
 1.13% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.42% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000002s :     12: predicate.merge_addn
 1.14% :     0.000001s :     12: predicate.minmaximum_grad
 1.49% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.53% :     0.000003s :     17: predicate.partial_eliminate
 1.22% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     12: predicate.reduce_eliminate
 1.51% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     20: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     12: predicate.reshape_eliminate
 1.26% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.73% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.74% :     0.000004s :     20: predicate.switch_defer_inline
 2.34% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.56% :     0.000011s :     65: predicate.switch_simplify
 1.30% :     0.000002s :     12: predicate.tile_eliminate
 1.32% :     0.000002s :     12: predicate.transpose_eliminate
 1.35% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.56% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     12: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000793    15
46.98% :     0.000373s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.38% :     0.000106s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.64% :     0.000314s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016396    72
 0.29% :     0.000047s :      1: add_recomputation
 0.53% :     0.000087s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.59% :     0.000260s :      1: bootstrap
 0.20% :     0.000033s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.16% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
34.84% :     0.005713s :      1: jit_opt_a
 0.82% :     0.000135s :      1: jit_opt_after_cconv
 0.30% :     0.000048s :      1: jit_opt_b
 2.62% :     0.000429s :      1: loop_unroll
 2.99% :     0.000490s :      1: mutable_eliminate
 6.49% :     0.001063s :     26: opt.transform.jit_opt_a
 0.27% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.71% :     0.000444s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.33% :     0.000054s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000015s :      1: remove_dup_value
 2.32% :     0.000380s :      1: renormalize.infer
 1.66% :     0.000272s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000035s :      1: rewriter_after_opt_a
 0.54% :     0.000088s :      1: rewriter_before_opt_a
 0.45% :     0.000074s :      1: symbol_engine_optimizer
39.35% :     0.006452s :      1: type_inference


TotalTime = 0.0214399, [33]
[bootstrap]: 0.00040757
[type_inference]: 0.00552886
[event_method]: 1.904e-05
[auto_monad]: 6.465e-05
[graph_reusing]: 5.35999e-06
[pre_auto_parallel]: 2.94999e-06
[py_interpret_to_execute]: 4.213e-05
[rewriter_before_opt_a]: 7.383e-05
[expand_dump_flag]: 3.61001e-06
[jit_opt_a]: 0.00625991, [2]
    [Cycle 1]: 0.00189209, [27]
        [switch_simplify]: 5.739e-05
        [loop_unroll]: 2.878e-05
        [a_1]: 0.00055452
        [with_stream_mark]: 1.852e-05
        [recompute_prepare]: 7.68001e-06
        [updatestate_depend_eliminate]: 3.33998e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 5.88002e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.57001e-06
        [meta_shard_fg_expand]: 2.09999e-06
        [get_grad_eliminate_]: 5.16002e-06
        [merge_forward]: 3.58999e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.361e-05
        [j_node_and_user_rematch]: 8.75999e-06
        [meta_fg_expand]: 2.32001e-06
        [replace_old_param]: 8.94e-06
        [inline_without_move]: 5.22999e-06
        [renormalize]: 0.00089225
        [add_forward_monad_depend]: 6.73998e-06
        [auto_monad_grad]: 2.81e-06
        [auto_monad_eliminator]: 1.474e-05
        [cse]: 3.281e-05
        [replace_applicator]: 1.436e-05
    [Cycle 2]: 0.0003374, [27]
        [switch_simplify]: 5.43002e-06
        [loop_unroll]: 5.04e-06
        [a_1]: 9.142e-05
        [with_stream_mark]: 1.045e-05
        [recompute_prepare]: 5.34e-06
        [updatestate_depend_eliminate]: 2.93998e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.72999e-06
        [specialize_transform]: 4.67998e-06
        [updatestate_useless_node_eliminater]: 4.53999e-06
        [accelerated_algorithm]: 5.10001e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 4.33001e-06
        [merge_forward]: 2.54999e-06
        [cell_reuse_recompute_pass]: 2.11998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.527e-05
        [j_node_and_user_rematch]: 7.68001e-06
        [meta_fg_expand]: 1.77999e-06
        [replace_old_param]: 7.99002e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 5.47001e-06
        [cse]: 1.293e-05
        [replace_applicator]: 5.86e-06
[py_interpret_to_execute_after_opt_a]: 1.482e-05
[rewriter_after_opt_a]: 3.858e-05
[convert_after_rewriter]: 7.67998e-06
[order_py_execute_after_rewriter]: 7.05002e-06
[mutable_eliminate]: 0.00063665
[jit_opt_b]: 5.04e-05, [1]
    [Cycle 1]: 4.229e-05, [2]
        [frontend_op_eliminate]: 1.632e-05
        [inline_after_opt_a]: 1.454e-05
[cconv]: 2.891e-05
[loop_unroll]: 0.00041456
[jit_opt_after_cconv]: 0.00013944, [1]
    [Cycle 1]: 0.00013323, [11]
        [c_1]: 2.116e-05
        [parameter_eliminate]: 2.56e-06
        [updatestate_depend_eliminate]: 5.92001e-06
        [updatestate_assign_eliminate]: 2.38002e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 2.003e-05
        [call_graph_tuple_transform]: 1.858e-05
        [tuple_list_get_item_eliminator]: 5.05999e-06
        [none_parameter_eliminate]: 1.32e-06
        [renormalize]: 6.69999e-07
        [switch_simplify]: 5.17e-06
[remove_dup_value]: 1.373e-05
[partial_unused_args_eliminate]: 2.45002e-06
[environ_conv]: 5.51e-06
[add_recomputation]: 5.422e-05
[cse_after_recomputation]: 2.122e-05, [1]
    [Cycle 1]: 1.549e-05, [1]
        [cse]: 9.51e-06
[auto_monad_reorder]: 1.536e-05
[get_jit_bprop_graph]: 2.31e-06
[rewriter_after_jit_bprop_graph]: 5.29e-06
[opt_after_jit_grad]: 0.00045876
[symbol_engine_optimizer]: 7.253e-05, [1]
    [Cycle 1]: 6.548e-05, [6]
        [build]: 4.08999e-06
        [elim_shapecalc]: 7.16999e-06
        [elim_not_effective]: 1.246e-05
        [opt_reshape]: 5.52001e-06
        [fold_const_symbol]: 8.47e-06
        [renormalize]: 7.00005e-07
[validate]: 4.241e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.00673804
[execute]: 1.001e-05
Sums
    bootstrap                            :     0.000408s :  2.44%
    type_inference                       :     0.005529s : 33.04%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000065s :  0.39%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000042s :  0.25%
    rewriter_before_opt_a                :     0.000074s :  0.44%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000646s :  3.86%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000892s :  5.33%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000046s :  0.27%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.09%
    rewriter_after_opt_a                 :     0.000039s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000007s :  0.04%
    mutable_eliminate                    :     0.000637s :  3.80%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000029s :  0.17%
    loop_unroll                          :     0.000415s :  2.48%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000054s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000459s :  2.74%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006738s : 40.27%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000226    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.57% :     0.000006s :      3: substitution.graph_param_transform
85.13% :     0.000192s :      4: substitution.inline
 1.24% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.22% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.82% :     0.000004s :      2: substitution.replace_old_param
 5.62% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005432     2
81.77% :     0.004442s :      1: type_inference.infer
18.23% :     0.000990s :      1: type_inference.specialize
------[replace.]   0.000069     6
68.63% :     0.000047s :      4: replace.inline
31.37% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000201     6
94.28% :     0.000190s :      4: match.inline
 5.72% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.54% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.53% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000001s :     10: predicate.addn_zero_filter
 1.92% :     0.000002s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.17% :     0.000001s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.82% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.26% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.49% :     0.000001s :      3: predicate.graph_param_transform
 5.50% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.18% :     0.000001s :      6: predicate.less_batch_normalization
 1.46% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.89% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.13% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.51% :     0.000002s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.52% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.39% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.72% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.53% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.28% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.31% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.41% :     0.000008s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.28% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.08% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000762    12
51.23% :     0.000390s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.54% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.22% :     0.000329s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023277    76
 0.25% :     0.000057s :      1: add_recomputation
 0.29% :     0.000068s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.84% :     0.000429s :      1: bootstrap
 0.14% :     0.000032s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.04% :     0.000010s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
26.92% :     0.006266s :      1: jit_opt_a
 0.61% :     0.000143s :      1: jit_opt_after_cconv
 0.23% :     0.000053s :      1: jit_opt_b
 1.82% :     0.000423s :      1: loop_unroll
 2.77% :     0.000646s :      1: mutable_eliminate
 3.69% :     0.000859s :     26: opt.transform.jit_opt_a
 0.20% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.01% :     0.000467s :      1: opt_after_jit_grad
 0.04% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000045s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000029s :      1: remove_dup_value
 2.03% :     0.000473s :      1: renormalize.infer
 1.77% :     0.000411s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000042s :      1: rewriter_after_opt_a
 0.33% :     0.000077s :      1: rewriter_before_opt_a
 0.32% :     0.000075s :      1: symbol_engine_optimizer
29.03% :     0.006757s :      1: task_emit
23.83% :     0.005546s :      1: type_inference
 0.28% :     0.000066s :      1: validate


TotalTime = 0.0136474, [30]
[bootstrap]: 0.0002507
[type_inference]: 0.00602542
[event_method]: 2.062e-05
[auto_monad]: 7.841e-05
[graph_reusing]: 5.96998e-06
[pre_auto_parallel]: 3.25e-06
[py_interpret_to_execute]: 4.292e-05
[rewriter_before_opt_a]: 8.343e-05
[expand_dump_flag]: 3.06999e-06
[jit_opt_a]: 0.00498096, [2]
    [Cycle 1]: 0.00166514, [27]
        [switch_simplify]: 0.00010522
        [loop_unroll]: 3.435e-05
        [a_1]: 0.00064232
        [with_stream_mark]: 1.569e-05
        [recompute_prepare]: 6.81999e-06
        [updatestate_depend_eliminate]: 3.55e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.47001e-06
        [parameter_eliminate]: 1.92001e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 2.07999e-06
        [get_grad_eliminate_]: 4.93001e-06
        [merge_forward]: 3.76999e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.395e-05
        [j_node_and_user_rematch]: 8.75001e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 8.67e-06
        [inline_without_move]: 5.35999e-06
        [renormalize]: 0.00055572
        [add_forward_monad_depend]: 5.08002e-06
        [auto_monad_grad]: 2.17999e-06
        [auto_monad_eliminator]: 1.273e-05
        [cse]: 2.865e-05
        [replace_applicator]: 1.176e-05
    [Cycle 2]: 0.00032832, [27]
        [switch_simplify]: 5.35001e-06
        [loop_unroll]: 4.79e-06
        [a_1]: 8.666e-05
        [with_stream_mark]: 9.18002e-06
        [recompute_prepare]: 4.74e-06
        [updatestate_depend_eliminate]: 2.46998e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 4.79e-06
        [updatestate_useless_node_eliminater]: 4.70999e-06
        [accelerated_algorithm]: 5.01997e-06
        [meta_shard_fg_expand]: 1.03001e-06
        [get_grad_eliminate_]: 4.40999e-06
        [merge_forward]: 3.04001e-06
        [cell_reuse_recompute_pass]: 1.52999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.181e-05
        [j_node_and_user_rematch]: 7.8e-06
        [meta_fg_expand]: 1.65001e-06
        [replace_old_param]: 7.11999e-06
        [inline_without_move]: 5.10999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.64e-06
        [auto_monad_grad]: 8.89995e-07
        [auto_monad_eliminator]: 6.08002e-06
        [cse]: 1.136e-05
        [replace_applicator]: 5.02e-06
[py_interpret_to_execute_after_opt_a]: 9.66e-06
[rewriter_after_opt_a]: 3.647e-05
[convert_after_rewriter]: 7.08998e-06
[order_py_execute_after_rewriter]: 5.56998e-06
[mutable_eliminate]: 0.00055017
[jit_opt_b]: 5.152e-05, [1]
    [Cycle 1]: 4.37e-05, [2]
        [frontend_op_eliminate]: 1.616e-05
        [inline_after_opt_a]: 1.502e-05
[cconv]: 3.001e-05
[loop_unroll]: 0.00046169
[jit_opt_after_cconv]: 0.00014587, [1]
    [Cycle 1]: 0.00013902, [11]
        [c_1]: 1.982e-05
        [parameter_eliminate]: 3.48e-06
        [updatestate_depend_eliminate]: 7.21001e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [cse]: 2.133e-05
        [call_graph_tuple_transform]: 1.899e-05
        [tuple_list_get_item_eliminator]: 5.19e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 9.20001e-07
        [switch_simplify]: 5.39e-06
[remove_dup_value]: 1.376e-05
[partial_unused_args_eliminate]: 2.30002e-06
[environ_conv]: 5.55001e-06
[add_recomputation]: 4.88e-05
[cse_after_recomputation]: 2.18e-05, [1]
    [Cycle 1]: 1.61e-05, [1]
        [cse]: 9.87999e-06
[auto_monad_reorder]: 1.479e-05
[get_jit_bprop_graph]: 2.02999e-06
[rewriter_after_jit_bprop_graph]: 6.02999e-06
[opt_after_jit_grad]: 0.0004673
[symbol_engine_optimizer]: 7.183e-05, [1]
    [Cycle 1]: 6.506e-05, [6]
        [build]: 3.68e-06
        [elim_shapecalc]: 7e-06
        [elim_not_effective]: 1.239e-05
        [opt_reshape]: 5.54e-06
        [fold_const_symbol]: 8.51002e-06
        [renormalize]: 7.50006e-07
[validate]: 3.244e-05
Sums
    bootstrap                            :     0.000251s :  2.49%
    type_inference                       :     0.006025s : 59.75%
    event_method                         :     0.000021s :  0.20%
    auto_monad                           :     0.000078s :  0.78%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000043s :  0.43%
    rewriter_before_opt_a                :     0.000083s :  0.83%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000111s :  1.10%
    jit_opt_a.loop_unroll                :     0.000039s :  0.39%
    jit_opt_a.a_1                        :     0.000729s :  7.23%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000556s :  5.51%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.19%
    jit_opt_a.cse                        :     0.000040s :  0.40%
    jit_opt_a.replace_applicator         :     0.000017s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.10%
    rewriter_after_opt_a                 :     0.000036s :  0.36%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000550s :  5.46%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.15%
    cconv                                :     0.000030s :  0.30%
    loop_unroll                          :     0.000462s :  4.58%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000049s :  0.48%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000467s :  4.63%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000032s :  0.32%


Time group info:
------[substitution.]   0.000252    26
 0.73% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.12% :     0.000005s :      3: substitution.graph_param_transform
80.93% :     0.000204s :      6: substitution.inline
 1.09% :     0.000003s :      4: substitution.j_node_and_user_rematch
 4.93% :     0.000012s :      4: substitution.remove_not_recompute_node
 1.49% :     0.000004s :      2: substitution.replace_old_param
 3.20% :     0.000008s :      1: substitution.switch_simplify
 4.96% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005969     2
83.20% :     0.004966s :      1: type_inference.infer
16.80% :     0.001003s :      1: type_inference.specialize
------[replace.]   0.000121     9
51.09% :     0.000062s :      6: replace.inline
25.90% :     0.000031s :      1: replace.switch_simplify
23.01% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000219     9
91.65% :     0.000201s :      6: match.inline
 3.19% :     0.000007s :      1: match.switch_simplify
 5.15% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   801
 1.43% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.17% :     0.000002s :     12: predicate.addn_check_dump
 1.44% :     0.000002s :     12: predicate.addn_zero_filter
 1.88% :     0.000002s :     12: predicate.arithmetic_simplify
 1.32% :     0.000002s :     12: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     12: predicate.compare_switch_simplify
 1.45% :     0.000002s :     12: predicate.depend_value_elim
 1.15% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.42% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.33% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.39% :     0.000002s :     12: predicate.environ_get_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.60% :     0.000007s :     26: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     17: predicate.load_eliminater
 1.22% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.39% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.72% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000002s :     12: predicate.merge_addn
 1.23% :     0.000002s :     12: predicate.minmaximum_grad
 1.77% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     17: predicate.partial_eliminate
 1.25% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     12: predicate.reduce_eliminate
 1.52% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     20: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     12: predicate.reshape_eliminate
 1.26% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.80% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.64% :     0.000003s :     20: predicate.switch_defer_inline
 2.44% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.31% :     0.000011s :     65: predicate.switch_simplify
 1.31% :     0.000002s :     12: predicate.tile_eliminate
 1.24% :     0.000002s :     12: predicate.transpose_eliminate
 1.53% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.22% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     12: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.99% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000746    15
49.12% :     0.000366s :      6: func_graph_cloner_run.FuncGraphClonerGraph
10.25% :     0.000077s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.63% :     0.000303s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015260    72
 0.34% :     0.000052s :      1: add_recomputation
 0.54% :     0.000082s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 1.75% :     0.000268s :      1: bootstrap
 0.22% :     0.000033s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
32.66% :     0.004984s :      1: jit_opt_a
 0.97% :     0.000149s :      1: jit_opt_after_cconv
 0.35% :     0.000054s :      1: jit_opt_b
 3.08% :     0.000470s :      1: loop_unroll
 3.68% :     0.000561s :      1: mutable_eliminate
 6.54% :     0.000997s :     26: opt.transform.jit_opt_a
 0.30% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.12% :     0.000476s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000046s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000016s :      1: remove_dup_value
 1.86% :     0.000284s :      1: renormalize.infer
 1.74% :     0.000265s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000040s :      1: rewriter_after_opt_a
 0.56% :     0.000086s :      1: rewriter_before_opt_a
 0.49% :     0.000074s :      1: symbol_engine_optimizer
39.56% :     0.006037s :      1: type_inference


TotalTime = 0.0249404, [33]
[bootstrap]: 0.00040901
[type_inference]: 0.00644271
[event_method]: 1.676e-05
[auto_monad]: 5.028e-05
[graph_reusing]: 3.81001e-06
[pre_auto_parallel]: 1.89e-06
[py_interpret_to_execute]: 4.422e-05
[rewriter_before_opt_a]: 6.919e-05
[expand_dump_flag]: 2.07001e-06
[jit_opt_a]: 0.00729499, [2]
    [Cycle 1]: 0.00203005, [27]
        [switch_simplify]: 6.812e-05
        [loop_unroll]: 2.956e-05
        [a_1]: 0.00058428
        [with_stream_mark]: 2.107e-05
        [recompute_prepare]: 8.55001e-06
        [updatestate_depend_eliminate]: 4e-06
        [updatestate_assign_eliminate]: 3.21999e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 5.99e-06
        [updatestate_useless_node_eliminater]: 5.40999e-06
        [accelerated_algorithm]: 6.33e-06
        [meta_shard_fg_expand]: 2.29999e-06
        [get_grad_eliminate_]: 6.26e-06
        [merge_forward]: 3.75e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.497e-05
        [j_node_and_user_rematch]: 1.06e-05
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 1.109e-05
        [inline_without_move]: 5.70001e-06
        [renormalize]: 0.00094395
        [add_forward_monad_depend]: 3.911e-05
        [auto_monad_grad]: 2.92002e-06
        [auto_monad_eliminator]: 1.553e-05
        [cse]: 2.911e-05
        [replace_applicator]: 1.257e-05
    [Cycle 2]: 0.00033514, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 5.09e-06
        [a_1]: 9.482e-05
        [with_stream_mark]: 1.118e-05
        [recompute_prepare]: 4.55001e-06
        [updatestate_depend_eliminate]: 3.16999e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 4.95001e-06
        [updatestate_useless_node_eliminater]: 4.58001e-06
        [accelerated_algorithm]: 5.39998e-06
        [meta_shard_fg_expand]: 1.92001e-06
        [get_grad_eliminate_]: 4.59002e-06
        [merge_forward]: 2.74999e-06
        [cell_reuse_recompute_pass]: 2.27001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.31e-05
        [j_node_and_user_rematch]: 7.95e-06
        [meta_fg_expand]: 1.80001e-06
        [replace_old_param]: 7.87e-06
        [inline_without_move]: 4.70999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.26002e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 4.84998e-06
        [cse]: 9.96e-06
        [replace_applicator]: 4.81002e-06
[py_interpret_to_execute_after_opt_a]: 1.23e-05
[rewriter_after_opt_a]: 3.895e-05
[convert_after_rewriter]: 7.13998e-06
[order_py_execute_after_rewriter]: 5.03002e-06
[mutable_eliminate]: 0.00065532
[jit_opt_b]: 5.221e-05, [1]
    [Cycle 1]: 4.451e-05, [2]
        [frontend_op_eliminate]: 1.812e-05
        [inline_after_opt_a]: 1.531e-05
[cconv]: 2.758e-05
[loop_unroll]: 0.00044961
[jit_opt_after_cconv]: 0.00014711, [1]
    [Cycle 1]: 0.00014008, [11]
        [c_1]: 2.219e-05
        [parameter_eliminate]: 2.61e-06
        [updatestate_depend_eliminate]: 5.94e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 2.068e-05
        [call_graph_tuple_transform]: 1.913e-05
        [tuple_list_get_item_eliminator]: 5.42001e-06
        [none_parameter_eliminate]: 1.81e-06
        [renormalize]: 4.49974e-07
        [switch_simplify]: 5.42999e-06
[remove_dup_value]: 1.429e-05
[partial_unused_args_eliminate]: 2.09e-06
[environ_conv]: 5.30001e-06
[add_recomputation]: 5.523e-05
[cse_after_recomputation]: 2.206e-05, [1]
    [Cycle 1]: 1.593e-05, [1]
        [cse]: 9.97001e-06
[auto_monad_reorder]: 1.55e-05
[get_jit_bprop_graph]: 2.40002e-06
[rewriter_after_jit_bprop_graph]: 6.74001e-06
[opt_after_jit_grad]: 0.00051756
[symbol_engine_optimizer]: 7.841e-05, [1]
    [Cycle 1]: 7.045e-05, [6]
        [build]: 4.52998e-06
        [elim_shapecalc]: 7.65e-06
        [elim_not_effective]: 1.416e-05
        [opt_reshape]: 6.16e-06
        [fold_const_symbol]: 9.67001e-06
        [renormalize]: 6.00005e-07
[validate]: 4.097e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00819577
[execute]: 1.024e-05
Sums
    bootstrap                            :     0.000409s :  2.12%
    type_inference                       :     0.006443s : 33.33%
    event_method                         :     0.000017s :  0.09%
    auto_monad                           :     0.000050s :  0.26%
    graph_reusing                        :     0.000004s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.23%
    rewriter_before_opt_a                :     0.000069s :  0.36%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000074s :  0.38%
    jit_opt_a.loop_unroll                :     0.000035s :  0.18%
    jit_opt_a.a_1                        :     0.000679s :  3.51%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000944s :  4.88%
    jit_opt_a.add_forward_monad_depend   :     0.000040s :  0.21%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.11%
    jit_opt_a.cse                        :     0.000039s :  0.20%
    jit_opt_a.replace_applicator         :     0.000017s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.06%
    rewriter_after_opt_a                 :     0.000039s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000655s :  3.39%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000028s :  0.14%
    loop_unroll                          :     0.000450s :  2.33%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000055s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.03%
    opt_after_jit_grad                   :     0.000518s :  2.68%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008196s : 42.39%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000238    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000002s :      2: substitution.fold_const_symbol
 2.21% :     0.000005s :      3: substitution.graph_param_transform
85.02% :     0.000202s :      4: substitution.inline
 1.40% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000005s :      2: substitution.replace_old_param
 5.62% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006362     2
70.05% :     0.004456s :      1: type_inference.infer
29.95% :     0.001905s :      1: type_inference.specialize
------[replace.]   0.000077     6
67.63% :     0.000052s :      4: replace.inline
32.37% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000212     6
94.22% :     0.000200s :      4: match.inline
 5.78% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.53% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 2.30% :     0.000003s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.60% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.33% :     0.000002s :     10: predicate.compare_switch_simplify
 1.82% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.07% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.49% :     0.000002s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.51% :     0.000001s :      3: predicate.graph_param_transform
 5.18% :     0.000006s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.91% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 1.24% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.84% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 2.02% :     0.000002s :      3: predicate.mutable_eliminate
 0.39% :     0.000000s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.13% :     0.000003s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.76% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.88% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.07% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.21% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.64% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 1.92% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.58% :     0.000009s :     53: predicate.switch_simplify
 1.46% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.94% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.64% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000829    12
57.96% :     0.000480s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.06% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.98% :     0.000315s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026890    76
 0.22% :     0.000059s :      1: add_recomputation
 0.20% :     0.000054s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.61% :     0.000432s :      1: bootstrap
 0.11% :     0.000030s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.08% :     0.000020s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.01% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000006s :      1: graph_reusing
27.15% :     0.007299s :      1: jit_opt_a
 0.56% :     0.000150s :      1: jit_opt_after_cconv
 0.21% :     0.000055s :      1: jit_opt_b
 1.71% :     0.000459s :      1: loop_unroll
 2.47% :     0.000664s :      1: mutable_eliminate
 3.38% :     0.000908s :     26: opt.transform.jit_opt_a
 0.18% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.96% :     0.000527s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000047s :      1: py_interpret_to_execute
 0.05% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 1.90% :     0.000512s :      1: renormalize.infer
 1.57% :     0.000423s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000042s :      1: rewriter_after_opt_a
 0.27% :     0.000072s :      1: rewriter_before_opt_a
 0.30% :     0.000082s :      1: symbol_engine_optimizer
30.57% :     0.008219s :      1: task_emit
24.01% :     0.006457s :      1: type_inference
 0.24% :     0.000065s :      1: validate


TotalTime = 0.0258577, [33]
[bootstrap]: 0.00033414
[type_inference]: 0.00494953
[event_method]: 1.876e-05
[auto_monad]: 6.854e-05
[graph_reusing]: 7.10998e-06
[pre_auto_parallel]: 3.25e-06
[py_interpret_to_execute]: 4.44e-05
[rewriter_before_opt_a]: 8.009e-05
[expand_dump_flag]: 3.40003e-06
[jit_opt_a]: 0.00907048, [2]
    [Cycle 1]: 0.00189359, [27]
        [switch_simplify]: 6.101e-05
        [loop_unroll]: 3.009e-05
        [a_1]: 0.00062962
        [with_stream_mark]: 1.467e-05
        [recompute_prepare]: 7.71999e-06
        [updatestate_depend_eliminate]: 2.98e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.91998e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 6.33998e-06
        [updatestate_useless_node_eliminater]: 5.10999e-06
        [accelerated_algorithm]: 5.51002e-06
        [meta_shard_fg_expand]: 1.30001e-06
        [get_grad_eliminate_]: 5.83997e-06
        [merge_forward]: 2.64001e-06
        [cell_reuse_recompute_pass]: 1.05999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.182e-05
        [j_node_and_user_rematch]: 8.95999e-06
        [meta_fg_expand]: 2.01e-06
        [replace_old_param]: 7.93001e-06
        [inline_without_move]: 4.99998e-06
        [renormalize]: 0.00085981
        [add_forward_monad_depend]: 5.54e-06
        [auto_monad_grad]: 1.51998e-06
        [auto_monad_eliminator]: 1.103e-05
        [cse]: 1.984e-05
        [replace_applicator]: 1.341e-05
    [Cycle 2]: 0.00033679, [27]
        [switch_simplify]: 5.69999e-06
        [loop_unroll]: 5.07999e-06
        [a_1]: 9.185e-05
        [with_stream_mark]: 1.174e-05
        [recompute_prepare]: 5.32999e-06
        [updatestate_depend_eliminate]: 2.63e-06
        [updatestate_assign_eliminate]: 1.98002e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 9.69972e-07
        [specialize_transform]: 5.26002e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.14998e-06
        [meta_shard_fg_expand]: 1.23002e-06
        [get_grad_eliminate_]: 4.62998e-06
        [merge_forward]: 2.78e-06
        [cell_reuse_recompute_pass]: 3.2e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.247e-05
        [j_node_and_user_rematch]: 8.28001e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 7.82e-06
        [inline_without_move]: 4.60999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.48002e-06
        [auto_monad_grad]: 1.27e-06
        [auto_monad_eliminator]: 5.87999e-06
        [cse]: 9.94001e-06
        [replace_applicator]: 5.40999e-06
[py_interpret_to_execute_after_opt_a]: 1.223e-05
[rewriter_after_opt_a]: 3.157e-05
[convert_after_rewriter]: 6.39999e-06
[order_py_execute_after_rewriter]: 4.62e-06
[mutable_eliminate]: 0.00216621
[jit_opt_b]: 7.428e-05, [1]
    [Cycle 1]: 6.575e-05, [2]
        [frontend_op_eliminate]: 3.343e-05
        [inline_after_opt_a]: 1.918e-05
[cconv]: 3.82e-05
[loop_unroll]: 0.00124113
[jit_opt_after_cconv]: 0.00019305, [1]
    [Cycle 1]: 0.0001848, [11]
        [c_1]: 3.351e-05
        [parameter_eliminate]: 5.82001e-06
        [updatestate_depend_eliminate]: 1.076e-05
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.99001e-06
        [cse]: 3.5e-05
        [call_graph_tuple_transform]: 2.459e-05
        [tuple_list_get_item_eliminator]: 6.03998e-06
        [none_parameter_eliminate]: 1.47999e-06
        [renormalize]: 1.07e-06
        [switch_simplify]: 5.71e-06
[remove_dup_value]: 1.545e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 5.70001e-06
[add_recomputation]: 5.298e-05
[cse_after_recomputation]: 2.282e-05, [1]
    [Cycle 1]: 1.61e-05, [1]
        [cse]: 9.37999e-06
[auto_monad_reorder]: 1.636e-05
[get_jit_bprop_graph]: 2.53998e-06
[rewriter_after_jit_bprop_graph]: 9.57999e-06
[opt_after_jit_grad]: 0.00052151
[symbol_engine_optimizer]: 7.981e-05, [1]
    [Cycle 1]: 7.17e-05, [6]
        [build]: 4.31002e-06
        [elim_shapecalc]: 7.68999e-06
        [elim_not_effective]: 1.516e-05
        [opt_reshape]: 6.44001e-06
        [fold_const_symbol]: 9.67999e-06
        [renormalize]: 5.50004e-07
[validate]: 4.453e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00648641
[execute]: 7.58001e-06
Sums
    bootstrap                            :     0.000334s :  1.82%
    type_inference                       :     0.004950s : 26.97%
    event_method                         :     0.000019s :  0.10%
    auto_monad                           :     0.000069s :  0.37%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000044s :  0.24%
    rewriter_before_opt_a                :     0.000080s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.36%
    jit_opt_a.loop_unroll                :     0.000035s :  0.19%
    jit_opt_a.a_1                        :     0.000721s :  3.93%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000860s :  4.69%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.09%
    jit_opt_a.cse                        :     0.000030s :  0.16%
    jit_opt_a.replace_applicator         :     0.000019s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.17%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.002166s : 11.80%
    jit_opt_b.frontend_op_eliminate      :     0.000033s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.10%
    cconv                                :     0.000038s :  0.21%
    loop_unroll                          :     0.001241s :  6.76%
    jit_opt_after_cconv.c_1              :     0.000034s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000035s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000053s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.05%
    opt_after_jit_grad                   :     0.000522s :  2.84%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006486s : 35.35%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000215    23
 1.26% :     0.000003s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 3.26% :     0.000007s :      3: substitution.graph_param_transform
84.73% :     0.000182s :      4: substitution.inline
 1.55% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.94% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.75% :     0.000004s :      2: substitution.replace_old_param
 4.88% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004870     2
80.90% :     0.003940s :      1: type_inference.infer
19.10% :     0.000930s :      1: type_inference.specialize
------[replace.]   0.000078     6
65.85% :     0.000051s :      4: replace.inline
34.15% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000189     6
94.98% :     0.000179s :      4: match.inline
 5.02% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.54% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.56% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000002s :     10: predicate.addn_zero_filter
 2.35% :     0.000003s :     10: predicate.arithmetic_simplify
 1.43% :     0.000002s :     10: predicate.cast_eliminate
 0.58% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.58% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.49% :     0.000002s :     10: predicate.environ_get_eliminate
 1.26% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.11% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.16% :     0.000007s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.87% :     0.000002s :     15: predicate.load_eliminater
 1.99% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.76% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.70% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 2.63% :     0.000003s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 1.87% :     0.000002s :     15: predicate.partial_eliminate
 1.23% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.84% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     18: predicate.replace_applicator
 0.45% :     0.000001s :      6: predicate.replace_old_param
 0.55% :     0.000001s :      3: predicate.reset_defer_inline
 1.14% :     0.000001s :     10: predicate.reshape_eliminate
 1.34% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.57% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.19% :     0.000002s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 1.84% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.32% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000002s :     10: predicate.tile_eliminate
 1.07% :     0.000001s :     10: predicate.transpose_eliminate
 1.53% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.01% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.38% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.43% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.95% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000667    12
47.91% :     0.000320s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.10% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.99% :     0.000313s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027808    76
 0.20% :     0.000056s :      1: add_recomputation
 0.26% :     0.000073s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.25% :     0.000348s :      1: bootstrap
 0.15% :     0.000041s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.08% :     0.000023s :      1: event_method
 0.04% :     0.000012s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
32.63% :     0.009075s :      1: jit_opt_a
 0.71% :     0.000196s :      1: jit_opt_after_cconv
 0.28% :     0.000078s :      1: jit_opt_b
 4.51% :     0.001254s :      1: loop_unroll
 7.85% :     0.002183s :      1: mutable_eliminate
 3.38% :     0.000939s :     26: opt.transform.jit_opt_a
 0.24% :     0.000066s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000045s :      4: opt.transform.jit_opt_b
 0.07% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000027s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.91% :     0.000530s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000047s :      1: py_interpret_to_execute
 0.05% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000018s :      1: remove_dup_value
 1.64% :     0.000456s :      1: renormalize.infer
 1.43% :     0.000396s :      1: renormalize.specialize
 0.04% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000035s :      1: rewriter_after_opt_a
 0.30% :     0.000083s :      1: rewriter_before_opt_a
 0.30% :     0.000083s :      1: symbol_engine_optimizer
23.39% :     0.006504s :      1: task_emit
17.87% :     0.004970s :      1: type_inference
 0.23% :     0.000065s :      1: validate

.
TotalTime = 0.0227864, [33]
[bootstrap]: 0.0003537
[type_inference]: 0.00552238
[event_method]: 1.98e-05
[auto_monad]: 7.081e-05
[graph_reusing]: 6.33998e-06
[pre_auto_parallel]: 2.74001e-06
[py_interpret_to_execute]: 4.347e-05
[rewriter_before_opt_a]: 7.702e-05
[expand_dump_flag]: 3.56999e-06
[jit_opt_a]: 0.00664512, [2]
    [Cycle 1]: 0.00213335, [27]
        [switch_simplify]: 7.028e-05
        [loop_unroll]: 2.95e-05
        [a_1]: 0.00068538
        [with_stream_mark]: 2.605e-05
        [recompute_prepare]: 8.67998e-06
        [updatestate_depend_eliminate]: 3.96001e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.70002e-06
        [parameter_eliminate]: 1.98002e-06
        [specialize_transform]: 5.92001e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 5.67001e-06
        [meta_shard_fg_expand]: 2.21e-06
        [get_grad_eliminate_]: 5.37999e-06
        [merge_forward]: 3.49001e-06
        [cell_reuse_recompute_pass]: 1.90001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.411e-05
        [j_node_and_user_rematch]: 1.037e-05
        [meta_fg_expand]: 3.01001e-06
        [replace_old_param]: 9.67999e-06
        [inline_without_move]: 5.47001e-06
        [renormalize]: 0.00095146
        [add_forward_monad_depend]: 6.89001e-06
        [auto_monad_grad]: 2.63e-06
        [auto_monad_eliminator]: 1.646e-05
        [cse]: 3.209e-05
        [replace_applicator]: 1.531e-05
    [Cycle 2]: 0.00036715, [27]
        [switch_simplify]: 5.29e-06
        [loop_unroll]: 5.19e-06
        [a_1]: 9.453e-05
        [with_stream_mark]: 1.215e-05
        [recompute_prepare]: 4.94e-06
        [updatestate_depend_eliminate]: 3.15998e-06
        [updatestate_assign_eliminate]: 1.89999e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 1.45999e-06
        [specialize_transform]: 4.71002e-06
        [updatestate_useless_node_eliminater]: 4.53001e-06
        [accelerated_algorithm]: 5.41002e-06
        [meta_shard_fg_expand]: 1.73002e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.88e-06
        [cell_reuse_recompute_pass]: 3.47002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.959e-05
        [j_node_and_user_rematch]: 8.79e-06
        [meta_fg_expand]: 2.02001e-06
        [replace_old_param]: 7.46001e-06
        [inline_without_move]: 4.42e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.24003e-06
        [auto_monad_grad]: 1.10001e-06
        [auto_monad_eliminator]: 6.49001e-06
        [cse]: 1.237e-05
        [replace_applicator]: 7.32002e-06
[py_interpret_to_execute_after_opt_a]: 1.501e-05
[rewriter_after_opt_a]: 4.047e-05
[convert_after_rewriter]: 7.60998e-06
[order_py_execute_after_rewriter]: 5.74999e-06
[mutable_eliminate]: 0.0007505
[jit_opt_b]: 5.481e-05, [1]
    [Cycle 1]: 4.648e-05, [2]
        [frontend_op_eliminate]: 1.778e-05
        [inline_after_opt_a]: 1.617e-05
[cconv]: 3.158e-05
[loop_unroll]: 0.00047541
[jit_opt_after_cconv]: 0.00015292, [1]
    [Cycle 1]: 0.00014669, [11]
        [c_1]: 2.176e-05
        [parameter_eliminate]: 3.39001e-06
        [updatestate_depend_eliminate]: 7.01001e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [cse]: 2.459e-05
        [call_graph_tuple_transform]: 2.148e-05
        [tuple_list_get_item_eliminator]: 5.56e-06
        [none_parameter_eliminate]: 1.83002e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 5.33002e-06
[remove_dup_value]: 1.6e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 5.95002e-06
[add_recomputation]: 5.908e-05
[cse_after_recomputation]: 2.079e-05, [1]
    [Cycle 1]: 1.478e-05, [1]
        [cse]: 8.32e-06
[auto_monad_reorder]: 1.633e-05
[get_jit_bprop_graph]: 2.12001e-06
[rewriter_after_jit_bprop_graph]: 9.17999e-06
[opt_after_jit_grad]: 0.00051595
[symbol_engine_optimizer]: 8.091e-05, [1]
    [Cycle 1]: 7.313e-05, [6]
        [build]: 5.12999e-06
        [elim_shapecalc]: 8.06001e-06
        [elim_not_effective]: 1.437e-05
        [opt_reshape]: 5.89e-06
        [fold_const_symbol]: 9.14e-06
        [renormalize]: 6.00005e-07
[validate]: 4.81e-05
[backend_pass]: 1.12e-06
[task_emit]: 0.00746964
[execute]: 1.121e-05
Sums
    bootstrap                            :     0.000354s :  1.97%
    type_inference                       :     0.005522s : 30.81%
    event_method                         :     0.000020s :  0.11%
    auto_monad                           :     0.000071s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.24%
    rewriter_before_opt_a                :     0.000077s :  0.43%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.42%
    jit_opt_a.loop_unroll                :     0.000035s :  0.19%
    jit_opt_a.a_1                        :     0.000780s :  4.35%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000952s :  5.31%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.13%
    jit_opt_a.cse                        :     0.000044s :  0.25%
    jit_opt_a.replace_applicator         :     0.000023s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.08%
    rewriter_after_opt_a                 :     0.000040s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000751s :  4.19%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000032s :  0.18%
    loop_unroll                          :     0.000475s :  2.65%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000059s :  0.33%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000516s :  2.88%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000048s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007470s : 41.68%
    execute                              :     0.000011s :  0.06%


TotalTime = 0.0135159, [30]
[bootstrap]: 0.00024926
[type_inference]: 0.00597688
[event_method]: 2.066e-05
[auto_monad]: 7.896e-05
[graph_reusing]: 6.52001e-06
[pre_auto_parallel]: 3.08998e-06
[py_interpret_to_execute]: 4.405e-05
[rewriter_before_opt_a]: 8.347e-05
[expand_dump_flag]: 2.99999e-06
[jit_opt_a]: 0.00509386, [2]
    [Cycle 1]: 0.00169381, [27]
        [switch_simplify]: 0.00010824
        [loop_unroll]: 3.326e-05
        [a_1]: 0.00064664
        [with_stream_mark]: 1.809e-05
        [recompute_prepare]: 7.13e-06
        [updatestate_depend_eliminate]: 3.13e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.35002e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 5.77001e-06
        [updatestate_useless_node_eliminater]: 5.24998e-06
        [accelerated_algorithm]: 5.35001e-06
        [meta_shard_fg_expand]: 2.12999e-06
        [get_grad_eliminate_]: 4.98001e-06
        [merge_forward]: 3.31999e-06
        [cell_reuse_recompute_pass]: 1.31998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.442e-05
        [j_node_and_user_rematch]: 8.89e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 8.02998e-06
        [inline_without_move]: 5.19998e-06
        [renormalize]: 0.00057831
        [add_forward_monad_depend]: 4.75001e-06
        [auto_monad_grad]: 1.94999e-06
        [auto_monad_eliminator]: 1.339e-05
        [cse]: 2.85e-05
        [replace_applicator]: 1.106e-05
    [Cycle 2]: 0.00032937, [27]
        [switch_simplify]: 5.19e-06
        [loop_unroll]: 5.23002e-06
        [a_1]: 9.762e-05
        [with_stream_mark]: 9.59999e-06
        [recompute_prepare]: 5.37999e-06
        [updatestate_depend_eliminate]: 2.74001e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.86998e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.82e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 1.15999e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.58e-06
        [cell_reuse_recompute_pass]: 1.14998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.217e-05
        [j_node_and_user_rematch]: 7.58001e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 6.89999e-06
        [inline_without_move]: 4.48999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.11002e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 5.43002e-06
        [cse]: 1.09e-05
        [replace_applicator]: 4.77998e-06
[py_interpret_to_execute_after_opt_a]: 8.90999e-06
[rewriter_after_opt_a]: 3.255e-05
[convert_after_rewriter]: 9.37001e-06
[order_py_execute_after_rewriter]: 4.71997e-06
[mutable_eliminate]: 0.0004741
[jit_opt_b]: 4.924e-05, [1]
    [Cycle 1]: 4.227e-05, [2]
        [frontend_op_eliminate]: 1.623e-05
        [inline_after_opt_a]: 1.42e-05
[cconv]: 2.289e-05
[loop_unroll]: 0.00040551
[jit_opt_after_cconv]: 0.00013625, [1]
    [Cycle 1]: 0.00012997, [11]
        [c_1]: 2.142e-05
        [parameter_eliminate]: 2.64999e-06
        [updatestate_depend_eliminate]: 5.27001e-06
        [updatestate_assign_eliminate]: 2.45002e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [cse]: 1.647e-05
        [call_graph_tuple_transform]: 1.75e-05
        [tuple_list_get_item_eliminator]: 5.13002e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 2.30008e-07
        [switch_simplify]: 5.14e-06
[remove_dup_value]: 1.245e-05
[partial_unused_args_eliminate]: 2.58e-06
[environ_conv]: 4.97999e-06
[add_recomputation]: 4.69e-05
[cse_after_recomputation]: 1.929e-05, [1]
    [Cycle 1]: 1.397e-05, [1]
        [cse]: 8.31002e-06
[auto_monad_reorder]: 1.423e-05
[get_jit_bprop_graph]: 1.66e-06
[rewriter_after_jit_bprop_graph]: 4.50999e-06
[opt_after_jit_grad]: 0.00043972
[symbol_engine_optimizer]: 7.058e-05, [1]
    [Cycle 1]: 6.44e-05, [6]
        [build]: 3.68e-06
        [elim_shapecalc]: 7.11001e-06
        [elim_not_effective]: 1.199e-05
        [opt_reshape]: 6.05002e-06
        [fold_const_symbol]: 8.2e-06
        [renormalize]: 3.50003e-07
[validate]: 3.137e-05
Sums
    bootstrap                            :     0.000249s :  2.52%
    type_inference                       :     0.005977s : 60.51%
    event_method                         :     0.000021s :  0.21%
    auto_monad                           :     0.000079s :  0.80%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000044s :  0.45%
    rewriter_before_opt_a                :     0.000083s :  0.85%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000113s :  1.15%
    jit_opt_a.loop_unroll                :     0.000038s :  0.39%
    jit_opt_a.a_1                        :     0.000744s :  7.53%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000578s :  5.86%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.19%
    jit_opt_a.cse                        :     0.000039s :  0.40%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.09%
    rewriter_after_opt_a                 :     0.000033s :  0.33%
    convert_after_rewriter               :     0.000009s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000474s :  4.80%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.14%
    cconv                                :     0.000023s :  0.23%
    loop_unroll                          :     0.000406s :  4.11%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000012s :  0.13%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000047s :  0.47%
    cse_after_recomputation.cse          :     0.000008s :  0.08%
    auto_monad_reorder                   :     0.000014s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000440s :  4.45%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.32%


Time group info:
------[substitution.]   0.000308    23
 0.64% :     0.000002s :      2: substitution.elim_not_effective
 0.42% :     0.000001s :      2: substitution.fold_const_symbol
 2.04% :     0.000006s :      3: substitution.graph_param_transform
85.86% :     0.000264s :      4: substitution.inline
 1.48% :     0.000005s :      4: substitution.j_node_and_user_rematch
 3.19% :     0.000010s :      4: substitution.remove_not_recompute_node
 1.65% :     0.000005s :      2: substitution.replace_old_param
 4.72% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005435     2
81.61% :     0.004436s :      1: type_inference.infer
18.39% :     0.001000s :      1: type_inference.specialize
------[replace.]   0.000095     6
65.43% :     0.000062s :      4: replace.inline
34.57% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000275     6
95.27% :     0.000262s :      4: match.inline
 4.73% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.57% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 2.07% :     0.000003s :     10: predicate.addn_zero_filter
 2.53% :     0.000003s :     10: predicate.arithmetic_simplify
 1.50% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.11% :     0.000001s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.69% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.18% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.03% :     0.000001s :     10: predicate.environ_get_eliminate
 1.19% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.55% :     0.000007s :     22: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000002s :      6: predicate.less_batch_normalization
 1.46% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.55% :     0.000002s :     15: predicate.load_eliminater
 1.10% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.44% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.18% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.35% :     0.000002s :     10: predicate.minmaximum_grad
 1.32% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.28% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.71% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.64% :     0.000001s :      3: predicate.reset_defer_inline
 1.12% :     0.000001s :     10: predicate.reshape_eliminate
 1.05% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.04% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.56% :     0.000003s :     16: predicate.switch_defer_inline
 1.83% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.34% :     0.000009s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.07% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.86% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.21% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.47% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.35% :     0.000003s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.90% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000786    12
49.71% :     0.000391s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.55% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.74% :     0.000336s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024855    76
 0.25% :     0.000062s :      1: add_recomputation
 0.30% :     0.000074s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.51% :     0.000376s :      1: bootstrap
 0.14% :     0.000034s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.07% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.75% :     0.006650s :      1: jit_opt_a
 0.63% :     0.000156s :      1: jit_opt_after_cconv
 0.23% :     0.000058s :      1: jit_opt_b
 1.95% :     0.000484s :      1: loop_unroll
 3.07% :     0.000762s :      1: mutable_eliminate
 4.09% :     0.001018s :     26: opt.transform.jit_opt_a
 0.20% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.12% :     0.000526s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000046s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.11% :     0.000525s :      1: renormalize.infer
 1.68% :     0.000417s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000044s :      1: rewriter_after_opt_a
 0.32% :     0.000080s :      1: rewriter_before_opt_a
 0.34% :     0.000083s :      1: symbol_engine_optimizer
30.14% :     0.007492s :      
Time group info:
------[substitution.]   0.000249    26
 0.77% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000001s :      2: substitution.fold_const_symbol
 1.98% :     0.000005s :      3: substitution.graph_param_transform
84.52% :     0.000210s :      6: substitution.inline
 1.07% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.73% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.41% :     0.000004s :      2: substitution.replace_old_param
 3.06% :     0.000008s :      1: substitution.switch_simplify
 4.94% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005916     2
83.71% :     0.004952s :      1: type_inference.infer
16.29% :     0.000964s :      1: type_inference.specialize
------[replace.]   0.000126     9
50.00% :     0.000063s :      6: replace.inline
26.95% :     0.000034s :      1: replace.switch_simplify
23.05% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000225     9
92.09% :     0.000207s :      6: match.inline
 3.01% :     0.000007s :      1: match.switch_simplify
 4.90% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   801
 1.44% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.05% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000001s :     12: predicate.addn_check_dump
 1.47% :     0.000002s :     12: predicate.addn_zero_filter
 1.97% :     0.000003s :     12: predicate.arithmetic_simplify
 1.35% :     0.000002s :     12: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.20% :     0.000002s :     12: predicate.compare_switch_simplify
 1.48% :     0.000002s :     12: predicate.depend_value_elim
 1.27% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.45% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.46% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     12: predicate.environ_add_const_eliminate
 1.27% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     12: predicate.environ_get_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.61% :     0.000007s :     26: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 2.00% :     0.000003s :     17: predicate.load_eliminater
 1.20% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.40% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.11% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     12: predicate.merge_addn
 1.19% :     0.000002s :     12: predicate.minmaximum_grad
 1.37% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.37% :     0.000003s :     17: predicate.partial_eliminate
 1.34% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     12: predicate.reduce_eliminate
 1.54% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     20: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_d1: task_emit
22.30% :     0.005542s :      1: type_inference
 0.30% :     0.000075s :      1: validate

efer_inline
 1.36% :     0.000002s :     12: predicate.reshape_eliminate
 1.33% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     12: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.79% :     0.000004s :     20: predicate.switch_defer_inline
 2.43% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.13% :     0.000010s :     65: predicate.switch_simplify
 1.21% :     0.000002s :     12: predicate.tile_eliminate
 1.34% :     0.000002s :     12: predicate.transpose_eliminate
 1.42% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.63% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.78% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.57% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     12: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000715    15
50.13% :     0.000358s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.30% :     0.000081s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.57% :     0.000276s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015156    72
 0.33% :     0.000050s :      1: add_recomputation
 0.54% :     0.000082s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.76% :     0.000266s :      1: bootstrap
 0.17% :     0.000025s :      1: cconv
 0.08% :     0.000012s :      1: convert_after_rewriter
 0.14% :     0.000021s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
33.63% :     0.005097s :      1: jit_opt_a
 0.92% :     0.000139s :      1: jit_opt_after_cconv
 0.34% :     0.000052s :      1: jit_opt_b
 2.72% :     0.000413s :      1: loop_unroll
 3.19% :     0.000483s :      1: mutable_eliminate
 6.63% :     0.001005s :     26: opt.transform.jit_opt_a
 0.30% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.95% :     0.000448s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000047s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000015s :      1: remove_dup_value
 2.01% :     0.000304s :      1: renormalize.infer
 1.76% :     0.000267s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000035s :      1: rewriter_after_opt_a
 0.57% :     0.000086s :      1: rewriter_before_opt_a
 0.48% :     0.000073s :      1: symbol_engine_optimizer
39.53% :     0.005991s :      1: type_inference


TotalTime = 0.0236772, [33]
[bootstrap]: 0.00033771
[type_inference]: 0.00702589
[event_method]: 1.773e-05
[auto_monad]: 6.795e-05
[graph_reusing]: 6.10002e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 4.445e-05
[rewriter_before_opt_a]: 7.938e-05
[expand_dump_flag]: 3.58999e-06
[jit_opt_a]: 0.0074708, [2]
    [Cycle 1]: 0.00192412, [27]
        [switch_simplify]: 8.899e-05
        [loop_unroll]: 3.06e-05
        [a_1]: 0.00059161
        [with_stream_mark]: 1.988e-05
        [recompute_prepare]: 7.35998e-06
        [updatestate_depend_eliminate]: 4.07998e-06
        [updatestate_assign_eliminate]: 3.83001e-06
        [updatestate_loads_eliminate]: 2.55002e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 6.98e-06
        [updatestate_useless_node_eliminater]: 5.29998e-06
        [accelerated_algorithm]: 5.96e-06
        [meta_shard_fg_expand]: 2.12001e-06
        [get_grad_eliminate_]: 6.49001e-06
        [merge_forward]: 3.70998e-06
        [cell_reuse_recompute_pass]: 1.63002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.709e-05
        [j_node_and_user_rematch]: 1.075e-05
        [meta_fg_expand]: 2.66e-06
        [replace_old_param]: 9.69e-06
        [inline_without_move]: 5.56e-06
        [renormalize]: 0.00084752
        [add_forward_monad_depend]: 6.56999e-06
        [auto_monad_grad]: 2.03002e-06
        [auto_monad_eliminator]: 1.524e-05
        [cse]: 3.093e-05
        [replace_applicator]: 1.321e-05
    [Cycle 2]: 0.00033208, [27]
        [switch_simplify]: 5.44e-06
        [loop_unroll]: 5.08002e-06
        [a_1]: 9.141e-05
        [with_stream_mark]: 9.86998e-06
        [recompute_prepare]: 5.05999e-06
        [updatestate_depend_eliminate]: 2.64001e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 5.45001e-06
        [updatestate_useless_node_eliminater]: 5.34e-06
        [accelerated_algorithm]: 6.01998e-06
        [meta_shard_fg_expand]: 1.50001e-06
        [get_grad_eliminate_]: 4.97999e-06
        [merge_forward]: 2.84001e-06
        [cell_reuse_recompute_pass]: 1.94e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.441e-05
        [j_node_and_user_rematch]: 8.90999e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 7.58999e-06
        [inline_without_move]: 4.57998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.06002e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 5.69999e-06
        [cse]: 1.087e-05
        [replace_applicator]: 4.87998e-06
[py_interpret_to_execute_after_opt_a]: 1.134e-05
[rewriter_after_opt_a]: 3.94e-05
[convert_after_rewriter]: 8.47e-06
[order_py_execute_after_rewriter]: 5.42999e-06
[mutable_eliminate]: 0.00069578
[jit_opt_b]: 5.226e-05, [1]
    [Cycle 1]: 4.437e-05, [2]
        [frontend_op_eliminate]: 1.721e-05
        [inline_after_opt_a]: 1.494e-05
[cconv]: 2.832e-05
[loop_unroll]: 0.00043992
[jit_opt_after_cconv]: 0.00014638, [1]
    [Cycle 1]: 0.0001396, [11]
        [c_1]: 2.087e-05
        [parameter_eliminate]: 3.36999e-06
        [updatestate_depend_eliminate]: 6.66999e-06
        [updatestate_assign_eliminate]: 2.54001e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [cse]: 2.001e-05
        [call_graph_tuple_transform]: 2.008e-05
        [tuple_list_get_item_eliminator]: 5.39e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 5.52001e-06
[remove_dup_value]: 1.582e-05
[partial_unused_args_eliminate]: 2.50997e-06
[environ_conv]: 4.84e-06
[add_recomputation]: 4.98e-05
[cse_after_recomputation]: 2.044e-05, [1]
    [Cycle 1]: 1.518e-05, [1]
        [cse]: 9.64999e-06
[auto_monad_reorder]: 1.581e-05
[get_jit_bprop_graph]: 2.23002e-06
[rewriter_after_jit_bprop_graph]: 6.14001e-06
[opt_after_jit_grad]: 0.0004757
[symbol_engine_optimizer]: 7.285e-05, [1]
    [Cycle 1]: 6.639e-05, [6]
        [build]: 4.12003e-06
        [elim_shapecalc]: 7.68001e-06
        [elim_not_effective]: 1.279e-05
        [opt_reshape]: 6.24001e-06
        [fold_const_symbol]: 8.27e-06
        [renormalize]: 4.30009e-07
[validate]: 4.172e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00626376
[execute]: 8.22e-06
Sums
    bootstrap                            :     0.000338s :  1.89%
    type_inference                       :     0.007026s : 39.41%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000068s :  0.38%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.25%
    rewriter_before_opt_a                :     0.000079s :  0.45%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000094s :  0.53%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000683s :  3.83%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000848s :  4.75%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.12%
    jit_opt_a.cse                        :     0.000042s :  0.23%
    jit_opt_a.replace_applicator         :     0.000018s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000039s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000696s :  3.90%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000028s :  0.16%
    loop_unroll                          :     0.000440s :  2.47%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.28%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000476s :  2.67%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000042s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006264s : 35.13%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000233    23
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.54% :     0.000006s :      3: substitution.graph_param_transform
83.98% :     0.000195s :      4: substitution.inline
 1.37% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.57% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.93% :     0.000004s :      2: substitution.replace_old_param
 6.24% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006943     2
57.21% :     0.003972s :      1: type_inference.infer
42.79% :     0.002971s :      1: type_inference.specialize
------[replace.]   0.000081     6
67.20% :     0.000055s :      4: replace.inline
32.80% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000206     6
93.50% :     0.000192s :      4: match.inline
 6.50% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.48% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000002s :     10: predicate.addn_zero_filter
 2.64% :     0.000003s :     10: predicate.arithmetic_simplify
 1.36% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.42% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.25% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 4.95% :     0.000006s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.44% :     0.000002s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.24% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.14% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.71% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000000s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.29% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.85% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.36% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.96% :     0.000009s :     53: predicate.switch_simplify
 1.36% :     0.000002s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.75% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000679    12
47.95% :     0.000326s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.99% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.06% :     0.000320s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025538    76
 0.21% :     0.000052s :      1: add_recomputation
 0.28% :     0.000072s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.40% :     0.000358s :      1: bootstrap
 0.12% :     0.000031s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.08% :     0.000021s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
29.27% :     0.007475s :      1: jit_opt_a
 0.59% :     0.000150s :      1: jit_opt_after_cconv
 0.21% :     0.000055s :      1: jit_opt_b
 1.76% :     0.000449s :      1: loop_unroll
 2.76% :     0.000705s :      1: mutable_eliminate
 3.58% :     0.000915s :     26: opt.transform.jit_opt_a
 0.19% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.90% :     0.000484s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000047s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 1.75% :     0.000446s :      1: renormalize.infer
 1.54% :     0.000392s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000042s :      1: rewriter_after_opt_a
 0.32% :     0.000082s :      1: rewriter_before_opt_a
 0.29% :     0.000075s :      1: symbol_engine_optimizer
24.58% :     0.006277s :      1: task_emit
27.58% :     0.007042s :      1: type_inference
 0.24% :     0.000062s :      1: validate


TotalTime = 0.0138404, [30]
[bootstrap]: 0.00025807
[type_inference]: 0.00618493
[event_method]: 2.11e-05
[auto_monad]: 7.816e-05
[graph_reusing]: 5.89e-06
[pre_auto_parallel]: 2.86e-06
[py_interpret_to_execute]: 4.652e-05
[rewriter_before_opt_a]: 8.57e-05
[expand_dump_flag]: 2.99001e-06
[jit_opt_a]: 0.00515411, [2]
    [Cycle 1]: 0.00175678, [27]
        [switch_simplify]: 0.00011194
        [loop_unroll]: 3.311e-05
        [a_1]: 0.00069253
        [with_stream_mark]: 1.801e-05
        [recompute_prepare]: 6.97002e-06
        [updatestate_depend_eliminate]: 3.36001e-06
        [updatestate_assign_eliminate]: 2.77002e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [parameter_eliminate]: 2.20002e-06
        [specialize_transform]: 5.54998e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 5.45001e-06
        [meta_shard_fg_expand]: 2.02999e-06
        [get_grad_eliminate_]: 5.02999e-06
        [merge_forward]: 3.38e-06
        [cell_reuse_recompute_pass]: 1.05999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.484e-05
        [j_node_and_user_rematch]: 8.69e-06
        [meta_fg_expand]: 2.17999e-06
        [replace_old_param]: 8.82e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 0.00058913
        [add_forward_monad_depend]: 5.13002e-06
        [auto_monad_grad]: 1.91e-06
        [auto_monad_eliminator]: 1.304e-05
        [cse]: 2.927e-05
        [replace_applicator]: 1.061e-05
    [Cycle 2]: 0.00030384, [27]
        [switch_simplify]: 5.59998e-06
        [loop_unroll]: 4.72e-06
        [a_1]: 8.683e-05
        [with_stream_mark]: 8.33999e-06
        [recompute_prepare]: 4.79998e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 1.92999e-06
        [updatestate_loads_eliminate]: 1.74e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 4.75999e-06
        [updatestate_useless_node_eliminater]: 4.60999e-06
        [accelerated_algorithm]: 4.40999e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 4.28999e-06
        [merge_forward]: 2.41e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.213e-05
        [j_node_and_user_rematch]: 7.71999e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 7.02997e-06
        [inline_without_move]: 4.73001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.05999e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 4.71002e-06
        [cse]: 9.45001e-06
        [replace_applicator]: 4.75001e-06
[py_interpret_to_execute_after_opt_a]: 9.75002e-06
[rewriter_after_opt_a]: 3.218e-05
[convert_after_rewriter]: 6.46e-06
[order_py_execute_after_rewriter]: 5.19e-06
[mutable_eliminate]: 0.00048208
[jit_opt_b]: 5.728e-05, [1]
    [Cycle 1]: 5.027e-05, [2]
        [frontend_op_eliminate]: 1.522e-05
        [inline_after_opt_a]: 1.543e-05
[cconv]: 2.524e-05
[loop_unroll]: 0.00041894
[jit_opt_after_cconv]: 0.00013248, [1]
    [Cycle 1]: 0.00012623, [11]
        [c_1]: 1.994e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 5.07e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [cse]: 1.64e-05
        [call_graph_tuple_transform]: 1.699e-05
        [tuple_list_get_item_eliminator]: 5.19998e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 4.2998e-07
        [switch_simplify]: 5.22e-06
[remove_dup_value]: 1.413e-05
[partial_unused_args_eliminate]: 2.36998e-06
[environ_conv]: 4.68999e-06
[add_recomputation]: 4.618e-05
[cse_after_recomputation]: 2.002e-05, [1]
    [Cycle 1]: 1.5e-05, [1]
        [cse]: 9.22001e-06
[auto_monad_reorder]: 1.491e-05
[get_jit_bprop_graph]: 1.69e-06
[rewriter_after_jit_bprop_graph]: 4.79e-06
[opt_after_jit_grad]: 0.00044817
[symbol_engine_optimizer]: 7.215e-05, [1]
    [Cycle 1]: 6.609e-05, [6]
        [build]: 4.18001e-06
        [elim_shapecalc]: 7.41001e-06
        [elim_not_effective]: 1.293e-05
        [opt_reshape]: 5.39e-06
        [fold_const_symbol]: 8.42e-06
        [renormalize]: 4.30009e-07
[validate]: 3.329e-05
Sums
    bootstrap                            :     0.000258s :  2.54%
    type_inference                       :     0.006185s : 60.77%
    event_method                         :     0.000021s :  0.21%
    auto_monad                           :     0.000078s :  0.77%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000047s :  0.46%
    rewriter_before_opt_a                :     0.000086s :  0.84%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000118s :  1.15%
    jit_opt_a.loop_unroll                :     0.000038s :  0.37%
    jit_opt_a.a_1                        :     0.000779s :  7.66%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000010s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000589s :  5.79%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.17%
    jit_opt_a.cse                        :     0.000039s :  0.38%
    jit_opt_a.replace_applicator         :     0.000015s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.10%
    rewriter_after_opt_a                 :     0.000032s :  0.32%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000482s :  4.74%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.15%
    cconv                                :     0.000025s :  0.25%
    loop_unroll                          :     0.000419s :  4.12%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000046s :  0.45%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000448s :  4.40%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.33%


Time group info:
------[substitution.]   0.000293    26
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 1.73% :     0.000005s :      3: substitution.graph_param_transform
85.99% :     0.000252s :      6: substitution.inline
 1.04% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.63% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.44% :     0.000004s :      2: substitution.replace_old_param
 2.67% :     0.000008s :      1: substitution.switch_simplify
 4.26% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006121     2
83.03% :     0.005082s :      1: type_inference.infer
16.97% :     0.001039s :      1: type_inference.specialize
------[replace.]   0.000129     9
50.88% :     0.000066s :      6: replace.inline
26.80% :     0.000035s :      1: replace.switch_simplify
22.33% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000267     9
93.11% :     0.000249s :      6: match.inline
 2.64% :     0.000007s :      1: match.switch_simplify
 4.25% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   801
 1.31% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.20% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.17% :     0.000001s :     12: predicate.addn_check_dump
 1.57% :     0.000002s :     12: predicate.addn_zero_filter
 1.88% :     0.000002s :     12: predicate.arithmetic_simplify
 1.28% :     0.000002s :     12: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.24% :     0.000002s :     12: predicate.compare_switch_simplify
 1.38% :     0.000002s :     12: predicate.depend_value_elim
 1.20% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     12: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     12: predicate.environ_get_eliminate
 1.16% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.78% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.53% :     0.000007s :     26: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     17: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.24% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000002s :     12: predicate.merge_addn
 1.21% :     0.000002s :     12: predicate.minmaximum_grad
 1.36% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     17: predicate.partial_eliminate
 1.42% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     12: predicate.reduce_eliminate
 1.56% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     20: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     12: predicate.reshape_eliminate
 1.31% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.46% :     0.000002s :     12: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.72% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.92% :     0.000004s :     20: predicate.switch_defer_inline
 2.35% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.43% :     0.000011s :     65: predicate.switch_simplify
 1.28% :     0.000002s :     12: predicate.tile_eliminate
 1.36% :     0.000002s :     12: predicate.transpose_eliminate
 1.48% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.99% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.39% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000782    15
49.48% :     0.000387s :      6: func_graph_cloner_run.FuncGraphClonerGraph
14.90% :     0.000117s :      1: func_graph_cloner_run.FuncGraphClonerNode
35.62% :     0.000279s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015525    72
 0.32% :     0.000049s :      1: add_recomputation
 0.53% :     0.000082s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.77% :     0.000276s :      1: bootstrap
 0.18% :     0.000028s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.16% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
33.22% :     0.005157s :      1: jit_opt_a
 0.87% :     0.000135s :      1: jit_opt_after_cconv
 0.39% :     0.000060s :      1: jit_opt_b
 2.75% :     0.000426s :      1: loop_unroll
 3.16% :     0.000490s :      1: mutable_eliminate
 6.71% :     0.001042s :     26: opt.transform.jit_opt_a
 0.28% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.94% :     0.000457s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.32% :     0.000049s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000016s :      1: remove_dup_value
 2.08% :     0.000323s :      1: renormalize.infer
 1.67% :     0.000259s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000035s :      1: rewriter_after_opt_a
 0.57% :     0.000088s :      1: rewriter_before_opt_a
 0.48% :     0.000075s :      1: symbol_engine_optimizer
39.94% :     0.006200s :      1: type_inference


TotalTime = 0.0216209, [33]
[bootstrap]: 0.00032941
[type_inference]: 0.00520046
[event_method]: 1.752e-05
[auto_monad]: 6.658e-05
[graph_reusing]: 5.89e-06
[pre_auto_parallel]: 2.68e-06
[py_interpret_to_execute]: 4.047e-05
[rewriter_before_opt_a]: 7.351e-05
[expand_dump_flag]: 3.16001e-06
[jit_opt_a]: 0.00632272, [2]
    [Cycle 1]: 0.00200446, [27]
        [switch_simplify]: 6.427e-05
        [loop_unroll]: 2.933e-05
        [a_1]: 0.00061426
        [with_stream_mark]: 2.301e-05
        [recompute_prepare]: 8e-06
        [updatestate_depend_eliminate]: 3.65e-06
        [updatestate_assign_eliminate]: 3.48e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 2.29001e-06
        [specialize_transform]: 5.86e-06
        [updatestate_useless_node_eliminater]: 5.42001e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 2.46998e-06
        [get_grad_eliminate_]: 5.03002e-06
        [merge_forward]: 3.71999e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.548e-05
        [j_node_and_user_rematch]: 9.32001e-06
        [meta_fg_expand]: 2.79999e-06
        [replace_old_param]: 9.47999e-06
        [inline_without_move]: 5.05999e-06
        [renormalize]: 0.00092209
        [add_forward_monad_depend]: 6.45002e-06
        [auto_monad_grad]: 2.86e-06
        [auto_monad_eliminator]: 1.485e-05
        [cse]: 3.307e-05
        [replace_applicator]: 1.237e-05
    [Cycle 2]: 0.00033186, [27]
        [switch_simplify]: 5.98998e-06
        [loop_unroll]: 5.24998e-06
        [a_1]: 9.115e-05
        [with_stream_mark]: 1.102e-05
        [recompute_prepare]: 5.97001e-06
        [updatestate_depend_eliminate]: 2.83e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.91003e-06
        [parameter_eliminate]: 1.37999e-06
        [specialize_transform]: 5.00001e-06
        [updatestate_useless_node_eliminater]: 4.77998e-06
        [accelerated_algorithm]: 5.52001e-06
        [meta_shard_fg_expand]: 1.59e-06
        [get_grad_eliminate_]: 4.58001e-06
        [merge_forward]: 2.66e-06
        [cell_reuse_recompute_pass]: 2.37999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.369e-05
        [j_node_and_user_rematch]: 9.01002e-06
        [meta_fg_expand]: 2.14999e-06
        [replace_old_param]: 8.22e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.02e-06
        [auto_monad_grad]: 6.50005e-07
        [auto_monad_eliminator]: 5.51998e-06
        [cse]: 1.025e-05
        [replace_applicator]: 5.41002e-06
[py_interpret_to_execute_after_opt_a]: 1.229e-05
[rewriter_after_opt_a]: 3.835e-05
[convert_after_rewriter]: 6.92002e-06
[order_py_execute_after_rewriter]: 5.20999e-06
[mutable_eliminate]: 0.00075566
[jit_opt_b]: 5.095e-05, [1]
    [Cycle 1]: 4.376e-05, [2]
        [frontend_op_eliminate]: 1.769e-05
        [inline_after_opt_a]: 1.485e-05
[cconv]: 2.915e-05
[loop_unroll]: 0.00047704
[jit_opt_after_cconv]: 0.00014303, [1]
    [Cycle 1]: 0.00013688, [11]
        [c_1]: 2.087e-05
        [parameter_eliminate]: 3.46001e-06
        [updatestate_depend_eliminate]: 6.31e-06
        [updatestate_assign_eliminate]: 2.41998e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [cse]: 2.103e-05
        [call_graph_tuple_transform]: 1.869e-05
        [tuple_list_get_item_eliminator]: 5.09e-06
        [none_parameter_eliminate]: 1.61998e-06
        [renormalize]: 7.09988e-07
        [switch_simplify]: 5.79999e-06
[remove_dup_value]: 1.424e-05
[partial_unused_args_eliminate]: 2.64999e-06
[environ_conv]: 5.76e-06
[add_recomputation]: 5.323e-05
[cse_after_recomputation]: 2.02e-05, [1]
    [Cycle 1]: 1.467e-05, [1]
        [cse]: 8.67e-06
[auto_monad_reorder]: 1.489e-05
[get_jit_bprop_graph]: 2.05002e-06
[rewriter_after_jit_bprop_graph]: 5.78997e-06
[opt_after_jit_grad]: 0.00047191
[symbol_engine_optimizer]: 7.587e-05, [1]
    [Cycle 1]: 6.931e-05, [6]
        [build]: 4.53001e-06
        [elim_shapecalc]: 7.71001e-06
        [elim_not_effective]: 1.397e-05
        [opt_reshape]: 6.01998e-06
        [fold_const_symbol]: 9.20999e-06
        [renormalize]: 5.50004e-07
[validate]: 4.412e-05
[backend_pass]: 7.79983e-07
[task_emit]: 0.00708943
[execute]: 8.93002e-06
Sums
    bootstrap                            :     0.000329s :  1.94%
    type_inference                       :     0.005200s : 30.63%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000067s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.24%
    rewriter_before_opt_a                :     0.000074s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000070s :  0.41%
    jit_opt_a.loop_unroll                :     0.000035s :  0.20%
    jit_opt_a.a_1                        :     0.000705s :  4.15%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000922s :  5.43%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000043s :  0.26%
    jit_opt_a.replace_applicator         :     0.000018s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000756s :  4.45%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000029s :  0.17%
    loop_unroll                          :     0.000477s :  2.81%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000053s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000472s :  2.78%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.26%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007089s : 41.76%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000262    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000002s :      2: substitution.fold_const_symbol
 2.16% :     0.000006s :      3: substitution.graph_param_transform
85.53% :     0.000224s :      4: substitution.inline
 1.48% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.10% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000005s :      2: substitution.replace_old_param
 5.27% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005127     2
81.23% :     0.004164s :      1: type_inference.infer
18.77% :     0.000962s :      1: type_inference.specialize
------[replace.]   0.000076     6
67.44% :     0.000051s :      4: replace.inline
32.56% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000234     6
94.53% :     0.000221s :      4: match.inline
 5.47% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.81% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.62% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 1.71% :     0.000002s :     10: predicate.arithmetic_simplify
 1.32% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.48% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.74% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.79% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.40% :     0.000007s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.81% :     0.000002s :      6: predicate.less_batch_normalization
 2.13% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 1.58% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.15% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.98% :     0.000002s :      3: predicate.mutable_eliminate
 0.61% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.47% :     0.000001s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.10% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.16% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     10: predicate.same_eliminate
 0.78% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.45% :     0.000002s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.60% :     0.000003s :     16: predicate.switch_defer_inline
 2.00% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.61% :     0.000009s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.68% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.08% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.09% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000735    12
50.18% :     0.000369s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.28% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.53% :     0.000313s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023570    76
 0.24% :     0.000056s :      1: add_recomputation
 0.30% :     0.000070s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.49% :     0.000350s :      1: bootstrap
 0.14% :     0.000032s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.84% :     0.006326s :      1: jit_opt_a
 0.62% :     0.000146s :      1: jit_opt_after_cconv
 0.23% :     0.000054s :      1: jit_opt_b
 2.06% :     0.000485s :      1: loop_unroll
 3.25% :     0.000765s :      1: mutable_eliminate
 3.94% :     0.000929s :     26: opt.transform.jit_opt_a
 0.20% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.04% :     0.000481s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000043s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.18% :     0.000514s :      1: renormalize.infer
 1.69% :     0.000399s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000041s :      1: rewriter_after_opt_a
 0.32% :     0.000076s :      1: rewriter_before_opt_a
 0.33% :     0.000079s :      1: symbol_engine_optimizer
30.17% :     0.007111s :      1: task_emit
22.14% :     0.005218s :      1: type_inference
 0.29% :     0.000070s :      1: validate


TotalTime = 0.0134891, [30]
[bootstrap]: 0.00026908
[type_inference]: 0.00600302
[event_method]: 2.041e-05
[auto_monad]: 7.905e-05
[graph_reusing]: 6.58e-06
[pre_auto_parallel]: 2.76e-06
[py_interpret_to_execute]: 4.515e-05
[rewriter_before_opt_a]: 8.408e-05
[expand_dump_flag]: 3.15998e-06
[jit_opt_a]: 0.00500751, [2]
    [Cycle 1]: 0.00166109, [27]
        [switch_simplify]: 0.00010553
        [loop_unroll]: 3.326e-05
        [a_1]: 0.00064994
        [with_stream_mark]: 1.699e-05
        [recompute_prepare]: 7.18e-06
        [updatestate_depend_eliminate]: 3.10002e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.66e-06
        [parameter_eliminate]: 1.74e-06
        [specialize_transform]: 5.79e-06
        [updatestate_useless_node_eliminater]: 4.89998e-06
        [accelerated_algorithm]: 5.17e-06
        [meta_shard_fg_expand]: 2.04e-06
        [get_grad_eliminate_]: 5.10001e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.437e-05
        [j_node_and_user_rematch]: 8.79998e-06
        [meta_fg_expand]: 2.25002e-06
        [replace_old_param]: 8.90001e-06
        [inline_without_move]: 5.23002e-06
        [renormalize]: 0.0005463
        [add_forward_monad_depend]: 5.14e-06
        [auto_monad_grad]: 2.12001e-06
        [auto_monad_eliminator]: 1.324e-05
        [cse]: 2.604e-05
        [replace_applicator]: 1.106e-05
    [Cycle 2]: 0.00030376, [27]
        [switch_simplify]: 5.58002e-06
        [loop_unroll]: 4.79e-06
        [a_1]: 8.895e-05
        [with_stream_mark]: 7.63001e-06
        [recompute_prepare]: 4.94e-06
        [updatestate_depend_eliminate]: 2.49999e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 9.20001e-07
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 4.76002e-06
        [accelerated_algorithm]: 4.65001e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.36002e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.41998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.166e-05
        [j_node_and_user_rematch]: 7.46999e-06
        [meta_fg_expand]: 1.46998e-06
        [replace_old_param]: 6.78e-06
        [inline_without_move]: 4.65001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 6.50005e-07
        [auto_monad_eliminator]: 4.43001e-06
        [cse]: 9.46998e-06
        [replace_applicator]: 5.02999e-06
[py_interpret_to_execute_after_opt_a]: 8.79003e-06
[rewriter_after_opt_a]: 3.25e-05
[convert_after_rewriter]: 6.69001e-06
[order_py_execute_after_rewriter]: 4.68001e-06
[mutable_eliminate]: 0.00047449
[jit_opt_b]: 5.618e-05, [1]
    [Cycle 1]: 4.974e-05, [2]
        [frontend_op_eliminate]: 2.31e-05
        [inline_after_opt_a]: 1.53e-05
[cconv]: 2.451e-05
[loop_unroll]: 0.00040695
[jit_opt_after_cconv]: 0.00013472, [1]
    [Cycle 1]: 0.00012775, [11]
        [c_1]: 2.06e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 5.32999e-06
        [updatestate_assign_eliminate]: 2.23998e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [cse]: 1.596e-05
        [call_graph_tuple_transform]: 1.735e-05
        [tuple_list_get_item_eliminator]: 5.27001e-06
        [none_parameter_eliminate]: 1.33002e-06
        [renormalize]: 3.70026e-07
        [switch_simplify]: 5.07999e-06
[remove_dup_value]: 1.366e-05
[partial_unused_args_eliminate]: 2.32999e-06
[environ_conv]: 4.89e-06
[add_recomputation]: 4.497e-05
[cse_after_recomputation]: 2e-05, [1]
    [Cycle 1]: 1.469e-05, [1]
        [cse]: 9.02e-06
[auto_monad_reorder]: 1.516e-05
[get_jit_bprop_graph]: 1.84998e-06
[rewriter_after_jit_bprop_graph]: 4.80999e-06
[opt_after_jit_grad]: 0.00044783
[symbol_engine_optimizer]: 7.002e-05, [1]
    [Cycle 1]: 6.385e-05, [6]
        [build]: 3.91999e-06
        [elim_shapecalc]: 7.23e-06
        [elim_not_effective]: 1.163e-05
        [opt_reshape]: 5.66e-06
        [fold_const_symbol]: 7.88001e-06
        [renormalize]: 4.00003e-07
[validate]: 3.151e-05
Sums
    bootstrap                            :     0.000269s :  2.72%
    type_inference                       :     0.006003s : 60.69%
    event_method                         :     0.000020s :  0.21%
    auto_monad                           :     0.000079s :  0.80%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000045s :  0.46%
    rewriter_before_opt_a                :     0.000084s :  0.85%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000111s :  1.12%
    jit_opt_a.loop_unroll                :     0.000038s :  0.38%
    jit_opt_a.a_1                        :     0.000739s :  7.47%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000546s :  5.52%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.18%
    jit_opt_a.cse                        :     0.000036s :  0.36%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.09%
    rewriter_after_opt_a                 :     0.000032s :  0.33%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000474s :  4.80%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.15%
    cconv                                :     0.000025s :  0.25%
    loop_unroll                          :     0.000407s :  4.11%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000045s :  0.45%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000448s :  4.53%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.32%


Time group info:
------[substitution.]   0.000252    26
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.02% :     0.000005s :      3: substitution.graph_param_transform
84.21% :     0.000212s :      6: substitution.inline
 1.16% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.74% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.38% :     0.000003s :      2: substitution.replace_old_param
 3.16% :     0.000008s :      1: substitution.switch_simplify
 5.07% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005943     2
83.35% :     0.004953s :      1: type_inference.infer
16.65% :     0.000990s :      1: type_inference.specialize
------[replace.]   0.000123     9
52.10% :     0.000064s :      6: replace.inline
25.06% :     0.000031s :      1: replace.switch_simplify
22.85% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000228     9
91.74% :     0.000209s :      6: match.inline
 3.17% :     0.000007s :      1: match.switch_simplify
 5.10% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   801
 1.55% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.16% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.17% :     0.000001s :     12: predicate.addn_check_dump
 1.60% :     0.000002s :     12: predicate.addn_zero_filter
 1.89% :     0.000002s :     12: predicate.arithmetic_simplify
 1.33% :     0.000002s :     12: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.18% :     0.000002s :     12: predicate.compare_switch_simplify
 1.48% :     0.000002s :     12: predicate.depend_value_elim
 1.19% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.37% :     0.000002s :     12: predicate.environ_get_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.79% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.60% :     0.000007s :     26: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     17: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.24% :     0.000005s :     40: predicate.loop_unroll_before_grad
 2.00% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000002s :     12: predicate.merge_addn
 1.15% :     0.000001s :     12: predicate.minmaximum_grad
 1.39% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     17: predicate.partial_eliminate
 1.25% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     12: predicate.reduce_eliminate
 1.50% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     20: predicate.replace_applicator
 0.82% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.58% :     0.000002s :     12: predicate.reshape_eliminate
 1.25% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     12: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.81% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.62% :     0.000003s :     20: predicate.switch_defer_inline
 2.51% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.20% :     0.000010s :     65: predicate.switch_simplify
 1.33% :     0.000002s :     12: predicate.tile_eliminate
 1.27% :     0.000002s :     12: predicate.transpose_eliminate
 1.28% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.80% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000002s :     12: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000726    15
49.93% :     0.000363s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.34% :     0.000082s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.73% :     0.000281s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015094    72
 0.32% :     0.000048s :      1: add_recomputation
 0.55% :     0.000083s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 1.91% :     0.000288s :      1: bootstrap
 0.18% :     0.000027s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
33.19% :     0.005010s :      1: jit_opt_a
 0.91% :     0.000137s :      1: jit_opt_after_cconv
 0.39% :     0.000059s :      1: jit_opt_b
 2.74% :     0.000414s :      1: loop_unroll
 3.20% :     0.000482s :      1: mutable_eliminate
 6.60% :     0.000997s :     26: opt.transform.jit_opt_a
 0.30% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000032s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000029s :      4: opt.transform.symbol_engine_opt
 3.02% :     0.000455s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.32% :     0.000048s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000016s :      1: remove_dup_value
 1.97% :     0.000298s :      1: renormalize.infer
 1.60% :     0.000242s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000035s :      1: rewriter_after_opt_a
 0.57% :     0.000087s :      1: rewriter_before_opt_a
 0.48% :     0.000072s :      1: symbol_engine_optimizer
39.85% :     0.006015s :      1: type_inference


TotalTime = 0.0344013, [33]
[bootstrap]: 0.00042716
[type_inference]: 0.00548369
[event_method]: 1.848e-05
[auto_monad]: 6.973e-05
[graph_reusing]: 5.65001e-06
[pre_auto_parallel]: 2.96999e-06
[py_interpret_to_execute]: 4.342e-05
[rewriter_before_opt_a]: 7.741e-05
[expand_dump_flag]: 3.2e-06
[jit_opt_a]: 0.0190632, [2]
    [Cycle 1]: 0.0146395, [27]
        [switch_simplify]: 6.867e-05
        [loop_unroll]: 3.072e-05
        [a_1]: 0.0131331
        [with_stream_mark]: 2.58e-05
        [recompute_prepare]: 1.139e-05
        [updatestate_depend_eliminate]: 3.82002e-06
        [updatestate_assign_eliminate]: 2.78998e-06
        [updatestate_loads_eliminate]: 2.90998e-06
        [parameter_eliminate]: 2.48e-06
        [specialize_transform]: 7.08e-06
        [updatestate_useless_node_eliminater]: 5.92001e-06
        [accelerated_algorithm]: 6.86001e-06
        [meta_shard_fg_expand]: 2.78e-06
        [get_grad_eliminate_]: 6.32001e-06
        [merge_forward]: 3.88001e-06
        [cell_reuse_recompute_pass]: 1.52001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.553e-05
        [j_node_and_user_rematch]: 1.194e-05
        [meta_fg_expand]: 2.74001e-06
        [replace_old_param]: 1.143e-05
        [inline_without_move]: 6.06e-06
        [renormalize]: 0.00098385
        [add_forward_monad_depend]: 6.94999e-06
        [auto_monad_grad]: 2.32001e-06
        [auto_monad_eliminator]: 1.588e-05
        [cse]: 2.947e-05
        [replace_applicator]: 1.335e-05
    [Cycle 2]: 0.00033156, [27]
        [switch_simplify]: 5.84999e-06
        [loop_unroll]: 5.76998e-06
        [a_1]: 9.641e-05
        [with_stream_mark]: 1.115e-05
        [recompute_prepare]: 4.91997e-06
        [updatestate_depend_eliminate]: 2.68e-06
        [updatestate_assign_eliminate]: 2.18002e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [parameter_eliminate]: 1.24998e-06
        [specialize_transform]: 5.07e-06
        [updatestate_useless_node_eliminater]: 4.68999e-06
        [accelerated_algorithm]: 5.24e-06
        [meta_shard_fg_expand]: 1.45999e-06
        [get_grad_eliminate_]: 4.87998e-06
        [merge_forward]: 2.54999e-06
        [cell_reuse_recompute_pass]: 2.48998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.477e-05
        [j_node_and_user_rematch]: 8.71002e-06
        [meta_fg_expand]: 1.65001e-06
        [replace_old_param]: 7.36999e-06
        [inline_without_move]: 5.12999e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 5.35999e-06
        [cse]: 1.073e-05
        [replace_applicator]: 5.42999e-06
[py_interpret_to_execute_after_opt_a]: 1.398e-05
[rewriter_after_opt_a]: 3.981e-05
[convert_after_rewriter]: 7.23999e-06
[order_py_execute_after_rewriter]: 5.19e-06
[mutable_eliminate]: 0.00077821
[jit_opt_b]: 5.152e-05, [1]
    [Cycle 1]: 4.354e-05, [2]
        [frontend_op_eliminate]: 1.775e-05
        [inline_after_opt_a]: 1.523e-05
[cconv]: 2.735e-05
[loop_unroll]: 0.00050247
[jit_opt_after_cconv]: 0.00015787, [1]
    [Cycle 1]: 0.00015048, [11]
        [c_1]: 2.512e-05
        [parameter_eliminate]: 3.23e-06
        [updatestate_depend_eliminate]: 5.77999e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [cse]: 2.563e-05
        [call_graph_tuple_transform]: 2.04e-05
        [tuple_list_get_item_eliminator]: 5.44e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 6.15002e-06
[remove_dup_value]: 1.465e-05
[partial_unused_args_eliminate]: 2.51998e-06
[environ_conv]: 5.43002e-06
[add_recomputation]: 5.395e-05
[cse_after_recomputation]: 2.561e-05, [1]
    [Cycle 1]: 1.855e-05, [1]
        [cse]: 1.243e-05
[auto_monad_reorder]: 1.573e-05
[get_jit_bprop_graph]: 2.12001e-06
[rewriter_after_jit_bprop_graph]: 6.37001e-06
[opt_after_jit_grad]: 0.00050733
[symbol_engine_optimizer]: 7.802e-05, [1]
    [Cycle 1]: 7.109e-05, [6]
        [build]: 4.02998e-06
        [elim_shapecalc]: 8.13999e-06
        [elim_not_effective]: 1.352e-05
        [opt_reshape]: 6.41e-06
        [fold_const_symbol]: 9.43002e-06
        [renormalize]: 4.40021e-07
[validate]: 3.891e-05
[backend_pass]: 8.79983e-07
[task_emit]: 0.0066093
[execute]: 9.64e-06
Sums
    bootstrap                            :     0.000427s :  1.44%
    type_inference                       :     0.005484s : 18.53%
    event_method                         :     0.000018s :  0.06%
    auto_monad                           :     0.000070s :  0.24%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.15%
    rewriter_before_opt_a                :     0.000077s :  0.26%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000075s :  0.25%
    jit_opt_a.loop_unroll                :     0.000036s :  0.12%
    jit_opt_a.a_1                        :     0.013229s : 44.70%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.04%
    jit_opt_a.merge_forward              :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000019s :  0.06%
    jit_opt_a.inline_without_move        :     0.000011s :  0.04%
    jit_opt_a.renormalize                :     0.000984s :  3.32%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.07%
    jit_opt_a.cse                        :     0.000040s :  0.14%
    jit_opt_a.replace_applicator         :     0.000019s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.05%
    rewriter_after_opt_a                 :     0.000040s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000778s :  2.63%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.05%
    cconv                                :     0.000027s :  0.09%
    loop_unroll                          :     0.000502s :  1.70%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000015s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000054s :  0.18%
    cse_after_recomputation.cse          :     0.000012s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000507s :  1.71%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000039s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006609s : 22.33%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.000332    23
 0.59% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000002s :      2: substitution.fold_const_symbol
 1.83% :     0.000006s :      3: substitution.graph_param_transform
87.07% :     0.000289s :      4: substitution.inline
 1.73% :     0.000006s :      4: substitution.j_node_and_user_rematch
 1.43% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.56% :     0.000005s :      2: substitution.replace_old_param
 5.26% :     0.000017s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005404     2
82.36% :     0.004451s :      1: type_inference.infer
17.64% :     0.000953s :      1: type_inference.specialize
------[replace.]   0.000096     6
70.21% :     0.000067s :      4: replace.inline
29.79% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000303     6
94.60% :     0.000286s :      4: match.inline
 5.40% :     0.000016s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.012505   691
 0.01% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.01% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.01% :     0.000001s :     10: predicate.addn_check_dump
 0.02% :     0.000002s :     10: predicate.addn_zero_filter
 0.04% :     0.000005s :     10: predicate.arithmetic_simplify
 0.02% :     0.000003s :     10: predicate.cast_eliminate
 0.00% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.01% :     0.000001s :     10: predicate.compare_switch_simplify
 0.02% :     0.000003s :     10: predicate.depend_value_elim
 0.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 0.02% :     0.000003s :     10: predicate.dict_get_item_eliminator
 0.02% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.00% :     0.000000s :      3: predicate.elim_not_effective
 0.00% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.01% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.01% :     0.000002s :     10: predicate.environ_get_depend_swap
 0.02% :     0.000002s :     10: predicate.environ_get_eliminate
 0.01% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.00% :     0.000000s :      3: predicate.fold_const_symbol
 0.01% :     0.000002s :      6: predicate.get_grad_eliminate
 0.00% :     0.000000s :      3: predicate.graph_param_transform
 0.07% :     0.000009s :     22: predicate.inline
 0.01% :     0.000001s :      6: predicate.inline_without_move
 0.00% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.02% :     0.000002s :      6: predicate.less_batch_normalization
 0.02% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 0.02% :     0.000002s :     15: predicate.load_eliminater
 0.01% :     0.000002s :      3: predicate.loop_unroll_after_grad
 0.04% :     0.000005s :     34: predicate.loop_unroll_before_grad
 0.03% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 0.02% :     0.000002s :     10: predicate.merge_addn
 0.01% :     0.000002s :     10: predicate.minmaximum_grad
 0.02% :     0.000003s :      3: predicate.mutable_eliminate
 0.00% :     0.000001s :      3: predicate.opt_reshape
 0.02% :     0.000003s :     15: predicate.partial_eliminate
 0.01% :     0.000002s :     10: predicate.print_const_string_wrapper
 0.04% :     0.000005s :     10: predicate.reduce_eliminate
 0.02% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.00% :     0.000001s :      6: predicate.remove_not_recompute_node
 0.02% :     0.000002s :     18: predicate.replace_applicator
 0.01% :     0.000001s :      6: predicate.replace_old_param
 0.00% :     0.000000s :      3: predicate.reset_defer_inline
98.81% :     0.012357s :     10: predicate.reshape_eliminate
 0.01% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.01% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.02% :     0.000002s :     10: predicate.same_eliminate
 0.01% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.01% :     0.000001s :      6: predicate.special_op_eliminate
 0.01% :     0.000001s :      6: predicate.specialize_transform
 0.02% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 0.01% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.00% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 0.05% :     0.000006s :     16: predicate.switch_defer_inline
 0.02% :     0.000003s :     16: predicate.switch_layer_defer_inline
 0.08% :     0.000010s :     53: predicate.switch_simplify
 0.02% :     0.000002s :     10: predicate.tile_eliminate
 0.02% :     0.000002s :     10: predicate.transpose_eliminate
 0.02% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.02% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 0.05% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 0.02% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 0.02% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 0.01% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 0.03% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 0.02% :     0.000003s :     10: predicate.value_based_eliminate
 0.00% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.01% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000737    12
48.99% :     0.000361s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.59% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.43% :     0.000320s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.048955    76
 0.12% :     0.000057s :      1: add_recomputation
 0.15% :     0.000073s :      1: auto_monad
 0.04% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.92% :     0.000449s :      1: bootstrap
 0.06% :     0.000030s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.06% :     0.000029s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.05% :     0.000023s :      1: event_method
 0.03% :     0.000016s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
38.95% :     0.019068s :      1: jit_opt_a
 0.33% :     0.000161s :      1: jit_opt_after_cconv
 0.11% :     0.000054s :      1: jit_opt_b
 1.04% :     0.000511s :      1: loop_unroll
 1.61% :     0.000787s :      1: mutable_eliminate
27.51% :     0.013468s :     26: opt.transform.jit_opt_a
 0.11% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000027s :      4: opt.transform.jit_opt_b
 0.03% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.05% :     0.000515s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.10% :     0.000047s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000017s :      1: remove_dup_value
 1.18% :     0.000575s :      1: renormalize.infer
 0.82% :     0.000399s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000043s :      1: rewriter_after_opt_a
 0.16% :     0.000080s :      1: rewriter_before_opt_a
 0.16% :     0.000080s :      1: symbol_engine_optimizer
13.54% :     0.006628s :      1: task_emit
11.28% :     0.005521s :      1: type_inference
 0.12% :     0.000060s :      1: validate


TotalTime = 0.0293579, [33]
[bootstrap]: 0.00038607
[type_inference]: 0.00698947
[event_method]: 1.888e-05
[auto_monad]: 6.657e-05
[graph_reusing]: 5.73002e-06
[pre_auto_parallel]: 2.76999e-06
[py_interpret_to_execute]: 3.943e-05
[rewriter_before_opt_a]: 7.589e-05
[expand_dump_flag]: 2.99001e-06
[jit_opt_a]: 0.00939742, [2]
    [Cycle 1]: 0.00181367, [27]
        [switch_simplify]: 5.634e-05
        [loop_unroll]: 2.939e-05
        [a_1]: 0.00054514
        [with_stream_mark]: 1.744e-05
        [recompute_prepare]: 7.09001e-06
        [updatestate_depend_eliminate]: 2.96999e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.55999e-06
        [specialize_transform]: 5.92999e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 5.52001e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 5.40001e-06
        [merge_forward]: 3.21999e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.152e-05
        [j_node_and_user_rematch]: 9.20999e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 8.22e-06
        [inline_without_move]: 5.54998e-06
        [renormalize]: 0.00087149
        [add_forward_monad_depend]: 5.22999e-06
        [auto_monad_grad]: 2.20002e-06
        [auto_monad_eliminator]: 1.342e-05
        [cse]: 1.692e-05
        [replace_applicator]: 1.312e-05
    [Cycle 2]: 0.00032677, [27]
        [switch_simplify]: 5.54998e-06
        [loop_unroll]: 5.39e-06
        [a_1]: 9.509e-05
        [with_stream_mark]: 1.076e-05
        [recompute_prepare]: 5.07e-06
        [updatestate_depend_eliminate]: 2.49999e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 5.15001e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.65001e-06
        [meta_shard_fg_expand]: 1.33002e-06
        [get_grad_eliminate_]: 4.63999e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 2.34001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.252e-05
        [j_node_and_user_rematch]: 7.95998e-06
        [meta_fg_expand]: 2.01998e-06
        [replace_old_param]: 7.44002e-06
        [inline_without_move]: 4.72998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.05999e-06
        [auto_monad_grad]: 1.06002e-06
        [auto_monad_eliminator]: 5.05999e-06
        [cse]: 9.76e-06
        [replace_applicator]: 4.86002e-06
[py_interpret_to_execute_after_opt_a]: 1.121e-05
[rewriter_after_opt_a]: 3.456e-05
[convert_after_rewriter]: 6.63e-06
[order_py_execute_after_rewriter]: 4.90999e-06
[mutable_eliminate]: 0.00062429
[jit_opt_b]: 5.001e-05, [1]
    [Cycle 1]: 4.331e-05, [2]
        [frontend_op_eliminate]: 1.76e-05
        [inline_after_opt_a]: 1.474e-05
[cconv]: 2.401e-05
[loop_unroll]: 0.00042691
[jit_opt_after_cconv]: 0.00013609, [1]
    [Cycle 1]: 0.00012996, [11]
        [c_1]: 2.114e-05
        [parameter_eliminate]: 2.34999e-06
        [updatestate_depend_eliminate]: 5.28002e-06
        [updatestate_assign_eliminate]: 2.47001e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.736e-05
        [call_graph_tuple_transform]: 1.837e-05
        [tuple_list_get_item_eliminator]: 5.72001e-06
        [none_parameter_eliminate]: 1.07e-06
        [renormalize]: 5.79981e-07
        [switch_simplify]: 5.10999e-06
[remove_dup_value]: 9.57001e-06
[partial_unused_args_eliminate]: 2.05002e-06
[environ_conv]: 3.86001e-06
[add_recomputation]: 3.747e-05
[cse_after_recomputation]: 2.094e-05, [1]
    [Cycle 1]: 1.59e-05, [1]
        [cse]: 9.50001e-06
[auto_monad_reorder]: 1.147e-05
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 5.74e-06
[opt_after_jit_grad]: 0.00046611
[symbol_engine_optimizer]: 7.177e-05, [1]
    [Cycle 1]: 6.517e-05, [6]
        [build]: 2.98e-06
        [elim_shapecalc]: 8.08999e-06
        [elim_not_effective]: 1.279e-05
        [opt_reshape]: 5.32001e-06
        [fold_const_symbol]: 8.55999e-06
        [renormalize]: 4.00003e-07
[validate]: 3.242e-05
[backend_pass]: 6.89994e-07
[task_emit]: 0.0101579
[execute]: 1.041e-05
Sums
    bootstrap                            :     0.000386s :  1.80%
    type_inference                       :     0.006989s : 32.53%
    event_method                         :     0.000019s :  0.09%
    auto_monad                           :     0.000067s :  0.31%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.18%
    rewriter_before_opt_a                :     0.000076s :  0.35%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000062s :  0.29%
    jit_opt_a.loop_unroll                :     0.000035s :  0.16%
    jit_opt_a.a_1                        :     0.000640s :  2.98%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.07%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000872s :  4.06%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.09%
    jit_opt_a.cse                        :     0.000027s :  0.12%
    jit_opt_a.replace_applicator         :     0.000018s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.05%
    rewriter_after_opt_a                 :     0.000035s :  0.16%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000624s :  2.91%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.07%
    cconv                                :     0.000024s :  0.11%
    loop_unroll                          :     0.000427s :  1.99%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000010s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000037s :  0.17%
    cse_after_recomputation.cse          :     0.000010s :  0.04%
    auto_monad_reorder                   :     0.000011s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000466s :  2.17%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010158s : 47.28%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000213    23
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.28% :     0.000005s :      3: substitution.graph_param_transform
85.89% :     0.000183s :      4: substitution.inline
 1.50% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.66% :     0.000004s :      2: substitution.replace_old_param
 5.19% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006908     2
84.72% :     0.005853s :      1: type_inference.infer
15.28% :     0.001055s :      1: type_inference.specialize
------[replace.]   0.000073     6
66.52% :     0.000049s :      4: replace.inline
33.48% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000190     6
94.72% :     0.000180s :      4: match.inline
 5.28% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.55% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.49% :     0.000002s :     10: predicate.addn_zero_filter
 2.37% :     0.000003s :     10: predicate.arithmetic_simplify
 1.19% :     0.000001s :     10: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.23% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 5.54% :     0.000006s :     22: predicate.inline
 1.10% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.60% :     0.000002s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.91% :     0.000002s :     15: predicate.load_eliminater
 1.17% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.29% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.73% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.45% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000000s :      3: predicate.opt_reshape
 2.01% :     0.000002s :     15: predicate.partial_eliminate
 1.08% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.39% :     0.000002s :     10: predicate.reduce_eliminate
 1.59% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000001s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.64% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.26% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.81% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 2.23% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.99% :     0.000008s :     53: predicate.switch_simplify
 1.34% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.35% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.20% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.97% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.002013    12
82.32% :     0.001657s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 1.79% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
15.89% :     0.000320s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.031172    76
 0.13% :     0.000040s :      1: add_recomputation
 0.22% :     0.000070s :      1: auto_monad
 0.04% :     0.000014s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.29% :     0.000402s :      1: bootstrap
 0.09% :     0.000027s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.07% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000006s :      1: environ_conv
 0.08% :     0.000023s :      1: event_method
 0.05% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
30.16% :     0.009402s :      1: jit_opt_a
 0.45% :     0.000139s :      1: jit_opt_after_cconv
 0.17% :     0.000053s :      1: jit_opt_b
 1.40% :     0.000435s :      1: loop_unroll
 2.03% :     0.000633s :      1: mutable_eliminate
 2.73% :     0.000852s :     26: opt.transform.jit_opt_a
 0.15% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000026s :      4: opt.transform.jit_opt_b
 0.04% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.52% :     0.000474s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000042s :      1: py_interpret_to_execute
 0.04% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000012s :      1: remove_dup_value
 1.53% :     0.000477s :      1: renormalize.infer
 1.24% :     0.000385s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000038s :      1: rewriter_after_opt_a
 0.25% :     0.000079s :      1: rewriter_before_opt_a
 0.24% :     0.000074s :      1: symbol_engine_optimizer
32.65% :     0.010178s :      1: task_emit
22.49% :     0.007009s :      1: type_inference
 0.16% :     0.000051s :      1: validate


TotalTime = 0.0222186, [33]
[bootstrap]: 0.00040356
[type_inference]: 0.005734
[event_method]: 1.824e-05
[auto_monad]: 6.694e-05
[graph_reusing]: 6.53998e-06
[pre_auto_parallel]: 2.89999e-06
[py_interpret_to_execute]: 4.048e-05
[rewriter_before_opt_a]: 7.481e-05
[expand_dump_flag]: 3.14001e-06
[jit_opt_a]: 0.00627515, [2]
    [Cycle 1]: 0.0020071, [27]
        [switch_simplify]: 7.433e-05
        [loop_unroll]: 3.051e-05
        [a_1]: 0.00060314
        [with_stream_mark]: 2.192e-05
        [recompute_prepare]: 7.87e-06
        [updatestate_depend_eliminate]: 4.26001e-06
        [updatestate_assign_eliminate]: 3.16001e-06
        [updatestate_loads_eliminate]: 2.71999e-06
        [parameter_eliminate]: 2.09e-06
        [specialize_transform]: 6.05002e-06
        [updatestate_useless_node_eliminater]: 5.02999e-06
        [accelerated_algorithm]: 5.87999e-06
        [meta_shard_fg_expand]: 2.54001e-06
        [get_grad_eliminate_]: 5.70001e-06
        [merge_forward]: 3.63999e-06
        [cell_reuse_recompute_pass]: 1.67001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.541e-05
        [j_node_and_user_rematch]: 9.74e-06
        [meta_fg_expand]: 2.41e-06
        [replace_old_param]: 1.036e-05
        [inline_without_move]: 5.25999e-06
        [renormalize]: 0.00091908
        [add_forward_monad_depend]: 6.12999e-06
        [auto_monad_grad]: 2.34999e-06
        [auto_monad_eliminator]: 1.563e-05
        [cse]: 3.472e-05
        [replace_applicator]: 1.316e-05
    [Cycle 2]: 0.00032089, [27]
        [switch_simplify]: 5.66e-06
        [loop_unroll]: 5.07e-06
        [a_1]: 9.143e-05
        [with_stream_mark]: 1.053e-05
        [recompute_prepare]: 5.34998e-06
        [updatestate_depend_eliminate]: 2.59001e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 8.80013e-07
        [specialize_transform]: 4.80999e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 4.55001e-06
        [meta_shard_fg_expand]: 1.42999e-06
        [get_grad_eliminate_]: 4.53001e-06
        [merge_forward]: 3.26001e-06
        [cell_reuse_recompute_pass]: 3.71999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.495e-05
        [j_node_and_user_rematch]: 7.93999e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 6.95998e-06
        [inline_without_move]: 5.00001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.11002e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 4.77e-06
        [cse]: 1.078e-05
        [replace_applicator]: 5.37999e-06
[py_interpret_to_execute_after_opt_a]: 1.009e-05
[rewriter_after_opt_a]: 3.79e-05
[convert_after_rewriter]: 7.13e-06
[order_py_execute_after_rewriter]: 5.31002e-06
[mutable_eliminate]: 0.0007622
[jit_opt_b]: 5.324e-05, [1]
    [Cycle 1]: 4.537e-05, [2]
        [frontend_op_eliminate]: 1.783e-05
        [inline_after_opt_a]: 1.55e-05
[cconv]: 3.044e-05
[loop_unroll]: 0.00045649
[jit_opt_after_cconv]: 0.00014921, [1]
    [Cycle 1]: 0.00014244, [11]
        [c_1]: 2.256e-05
        [parameter_eliminate]: 3.05002e-06
        [updatestate_depend_eliminate]: 6.66999e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 2.195e-05
        [call_graph_tuple_transform]: 2.032e-05
        [tuple_list_get_item_eliminator]: 5.29e-06
        [none_parameter_eliminate]: 1.49998e-06
        [renormalize]: 8.70001e-07
        [switch_simplify]: 5.31998e-06
[remove_dup_value]: 1.368e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 5.98998e-06
[add_recomputation]: 5.806e-05
[cse_after_recomputation]: 2.206e-05, [1]
    [Cycle 1]: 1.595e-05, [1]
        [cse]: 1.017e-05
[auto_monad_reorder]: 1.672e-05
[get_jit_bprop_graph]: 1.39e-06
[rewriter_after_jit_bprop_graph]: 7e-06
[opt_after_jit_grad]: 0.00047098
[symbol_engine_optimizer]: 7.902e-05, [1]
    [Cycle 1]: 7.17e-05, [6]
        [build]: 4.68001e-06
        [elim_shapecalc]: 8.07e-06
        [elim_not_effective]: 1.408e-05
        [opt_reshape]: 6.19999e-06
        [fold_const_symbol]: 1.031e-05
        [renormalize]: 4.49974e-07
[validate]: 4.853e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00710906
[execute]: 8.75999e-06
Sums
    bootstrap                            :     0.000404s :  2.29%
    type_inference                       :     0.005734s : 32.56%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000067s :  0.38%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.23%
    rewriter_before_opt_a                :     0.000075s :  0.42%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000080s :  0.45%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000695s :  3.94%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000919s :  5.22%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000045s :  0.26%
    jit_opt_a.replace_applicator         :     0.000019s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000038s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000762s :  4.33%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000030s :  0.17%
    loop_unroll                          :     0.000456s :  2.59%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000058s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000471s :  2.67%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000049s :  0.28%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007109s : 40.37%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000249    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.79% :     0.000002s :      2: substitution.fold_const_symbol
 2.47% :     0.000006s :      3: substitution.graph_param_transform
85.12% :     0.000212s :      4: substitution.inline
 1.47% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.11% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000005s :      2: substitution.replace_old_param
 5.29% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005654     2
80.49% :     0.004551s :      1: type_inference.infer
19.51% :     0.001103s :      1: type_inference.specialize
------[replace.]   0.000079     6
66.95% :     0.000053s :      4: replace.inline
33.05% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000222     6
94.50% :     0.000209s :      4: match.inline
 5.50% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.67% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.13% :     0.000001s :     10: predicate.cast_eliminate
 0.71% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000001s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.81% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.16% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.24% :     0.000006s :     22: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.07% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.62% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.96% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.53% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.36% :     0.000002s :     10: predicate.minmaximum_grad
 1.70% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 1.36% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     10: predicate.reduce_eliminate
 1.61% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.55% :     0.000002s :     18: predicate.replace_applicator
 1.01% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.62% :     0.000002s :     10: predicate.reshape_eliminate
 1.09% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.32% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.88% :     0.000002s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.63% :     0.000002s :      6: predicate.special_op_eliminate
 0.86% :     0.000001s :      6: predicate.specialize_transform
 1.18% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.22% :     0.000003s :     16: predicate.switch_defer_inline
 1.87% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.22% :     0.000009s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.86% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.66% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.84% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000766    12
48.12% :     0.000369s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.25% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.63% :     0.000357s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024159    76
 0.25% :     0.000061s :      1: add_recomputation
 0.29% :     0.000070s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.77% :     0.000427s :      1: bootstrap
 0.14% :     0.000033s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.99% :     0.006280s :      1: jit_opt_a
 0.63% :     0.000152s :      1: jit_opt_after_cconv
 0.23% :     0.000056s :      1: jit_opt_b
 1.92% :     0.000464s :      1: loop_unroll
 3.19% :     0.000772s :      1: mutable_eliminate
 3.84% :     0.000928s :     26: opt.transform.jit_opt_a
 0.20% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.98% :     0.000479s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000043s :      1: py_interpret_to_execute
 0.05% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.11% :     0.000511s :      1: renormalize.infer
 1.65% :     0.000399s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000041s :      1: rewriter_after_opt_a
 0.32% :     0.000078s :      1: rewriter_before_opt_a
 0.34% :     0.000082s :      1: symbol_engine_optimizer
29.49% :     0.007125s :      1: task_emit
23.81% :     0.005752s :      1: type_inference
 0.31% :     0.000075s :      1: validate


TotalTime = 0.0149088, [30]
[bootstrap]: 0.00026586
[type_inference]: 0.00614264
[event_method]: 2.144e-05
[auto_monad]: 8.136e-05
[graph_reusing]: 6.41e-06
[pre_auto_parallel]: 3.13e-06
[py_interpret_to_execute]: 4.566e-05
[rewriter_before_opt_a]: 8.265e-05
[expand_dump_flag]: 3.14999e-06
[jit_opt_a]: 0.00579788, [2]
    [Cycle 1]: 0.00212275, [27]
        [switch_simplify]: 0.00014021
        [loop_unroll]: 3.501e-05
        [a_1]: 0.0007484
        [with_stream_mark]: 2.603e-05
        [recompute_prepare]: 9.29998e-06
        [updatestate_depend_eliminate]: 4.3e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 6.11998e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.81e-06
        [meta_shard_fg_expand]: 2.15002e-06
        [get_grad_eliminate_]: 5.15999e-06
        [merge_forward]: 3.9e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.573e-05
        [j_node_and_user_rematch]: 9.74e-06
        [meta_fg_expand]: 2.29001e-06
        [replace_old_param]: 9.97001e-06
        [inline_without_move]: 5.00001e-06
        [renormalize]: 0.00080373
        [add_forward_monad_depend]: 6.57002e-06
        [auto_monad_grad]: 2.91999e-06
        [auto_monad_eliminator]: 1.636e-05
        [cse]: 3.254e-05
        [replace_applicator]: 1.456e-05
    [Cycle 2]: 0.00033623, [27]
        [switch_simplify]: 5.66e-06
        [loop_unroll]: 5.10001e-06
        [a_1]: 8.699e-05
        [with_stream_mark]: 1.182e-05
        [recompute_prepare]: 6.17999e-06
        [updatestate_depend_eliminate]: 2.81e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.63002e-06
        [specialize_transform]: 4.79998e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.00999e-06
        [meta_shard_fg_expand]: 1.58997e-06
        [get_grad_eliminate_]: 4.58001e-06
        [merge_forward]: 2.83e-06
        [cell_reuse_recompute_pass]: 2.34999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.583e-05
        [j_node_and_user_rematch]: 9.49999e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 8.48001e-06
        [inline_without_move]: 4.89e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.35001e-06
        [auto_monad_grad]: 1.45001e-06
        [auto_monad_eliminator]: 6.38e-06
        [cse]: 1.228e-05
        [replace_applicator]: 6.07999e-06
[py_interpret_to_execute_after_opt_a]: 1.321e-05
[rewriter_after_opt_a]: 4.079e-05
[convert_after_rewriter]: 7.1e-06
[order_py_execute_after_rewriter]: 4.95001e-06
[mutable_eliminate]: 0.00070096
[jit_opt_b]: 5.142e-05, [1]
    [Cycle 1]: 4.295e-05, [2]
        [frontend_op_eliminate]: 1.557e-05
        [inline_after_opt_a]: 1.465e-05
[cconv]: 2.937e-05
[loop_unroll]: 0.0004949
[jit_opt_after_cconv]: 0.00015525, [1]
    [Cycle 1]: 0.00014708, [11]
        [c_1]: 2.117e-05
        [parameter_eliminate]: 3.85e-06
        [updatestate_depend_eliminate]: 8.54e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [cse]: 2.342e-05
        [call_graph_tuple_transform]: 2.077e-05
        [tuple_list_get_item_eliminator]: 5.09998e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 1.17999e-06
        [switch_simplify]: 5.18002e-06
[remove_dup_value]: 1.399e-05
[partial_unused_args_eliminate]: 2.20002e-06
[environ_conv]: 6.16e-06
[add_recomputation]: 5.539e-05
[cse_after_recomputation]: 2.358e-05, [1]
    [Cycle 1]: 1.741e-05, [1]
        [cse]: 1.062e-05
[auto_monad_reorder]: 1.541e-05
[get_jit_bprop_graph]: 2.16e-06
[rewriter_after_jit_bprop_graph]: 7.61999e-06
[opt_after_jit_grad]: 0.00052094
[symbol_engine_optimizer]: 7.86e-05, [1]
    [Cycle 1]: 7.117e-05, [6]
        [build]: 4.99003e-06
        [elim_shapecalc]: 7.39002e-06
        [elim_not_effective]: 1.466e-05
        [opt_reshape]: 5.32001e-06
        [fold_const_symbol]: 9.09e-06
        [renormalize]: 1.10001e-06
[validate]: 3.857e-05
Sums
    bootstrap                            :     0.000266s :  2.43%
    type_inference                       :     0.006143s : 56.24%
    event_method                         :     0.000021s :  0.20%
    auto_monad                           :     0.000081s :  0.74%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000046s :  0.42%
    rewriter_before_opt_a                :     0.000083s :  0.76%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000146s :  1.34%
    jit_opt_a.loop_unroll                :     0.000040s :  0.37%
    jit_opt_a.a_1                        :     0.000835s :  7.65%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.17%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000804s :  7.36%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.21%
    jit_opt_a.cse                        :     0.000045s :  0.41%
    jit_opt_a.replace_applicator         :     0.000021s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.12%
    rewriter_after_opt_a                 :     0.000041s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000701s :  6.42%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.13%
    cconv                                :     0.000029s :  0.27%
    loop_unroll                          :     0.000495s :  4.53%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000055s :  0.51%
    cse_after_recomputation.cse          :     0.000011s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.07%
    opt_after_jit_grad                   :     0.000521s :  4.77%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.35%


Time group info:
------[substitution.]   0.000327    26
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.38% :     0.000001s :      2: substitution.fold_const_symbol
 1.71% :     0.000006s :      3: substitution.graph_param_transform
85.65% :     0.000280s :      6: substitution.inline
 1.24% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.63% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.81% :     0.000006s :      2: substitution.replace_old_param
 2.83% :     0.000009s :      1: substitution.switch_simplify
 4.01% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006076     2
83.84% :     0.005094s :      1: type_inference.infer
16.16% :     0.000982s :      1: type_inference.specialize
------[replace.]   0.000155     9
47.61% :     0.000074s :      6: replace.inline
29.77% :     0.000046s :      1: replace.switch_simplify
22.62% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000297     9
93.08% :     0.000276s :      6: match.inline
 2.86% :     0.000008s :      1: match.switch_simplify
 4.06% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   801
 1.35% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.50% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000002s :     12: predicate.addn_check_dump
 1.30% :     0.000002s :     12: predicate.addn_zero_filter
 1.80% :     0.000002s :     12: predicate.arithmetic_simplify
 1.23% :     0.000002s :     12: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     12: predicate.compare_switch_simplify
 1.39% :     0.000002s :     12: predicate.depend_value_elim
 1.15% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.17% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.26% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     12: predicate.environ_get_eliminate
 1.26% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.78% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 6.23% :     0.000008s :     26: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.49% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.18% :     0.000002s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     17: predicate.load_eliminater
 1.58% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.03% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000002s :     12: predicate.merge_addn
 1.09% :     0.000001s :     12: predicate.minmaximum_grad
 1.66% :     0.000002s :      3: predicate.mutable_eliminate
 0.38% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     17: predicate.partial_eliminate
 1.20% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     12: predicate.reduce_eliminate
 1.46% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     20: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 2.02% :     0.000003s :     12: predicate.reshape_eliminate
 1.23% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000002s :     12: predicate.same_eliminate
 0.62% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.73% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     20: predicate.switch_defer_inline
 2.22% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.19% :     0.000011s :     65: predicate.switch_simplify
 1.27% :     0.000002s :     12: predicate.tile_eliminate
 1.26% :     0.000002s :     12: predicate.transpose_eliminate
 1.34% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.62% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 2.22% :     0.000003s :     12: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.40% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.81% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000805    15
47.20% :     0.000380s :      6: func_graph_cloner_run.FuncGraphClonerGraph
14.26% :     0.000115s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.54% :     0.000310s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016915    72
 0.35% :     0.000059s :      1: add_recomputation
 0.50% :     0.000085s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 1.69% :     0.000285s :      1: bootstrap
 0.19% :     0.000033s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
34.30% :     0.005801s :      1: jit_opt_a
 0.94% :     0.000158s :      1: jit_opt_after_cconv
 0.32% :     0.000054s :      1: jit_opt_b
 2.99% :     0.000506s :      1: loop_unroll
 4.21% :     0.000713s :      1: mutable_eliminate
 6.78% :     0.001147s :     26: opt.transform.jit_opt_a
 0.29% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000024s :      4: opt.transform.jit_opt_b
 0.09% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.15% :     0.000533s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000048s :      1: py_interpret_to_execute
 0.10% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.63% :     0.000445s :      1: renormalize.infer
 2.07% :     0.000350s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000046s :      1: rewriter_after_opt_a
 0.51% :     0.000085s :      1: rewriter_before_opt_a
 0.48% :     0.000081s :      1: symbol_engine_optimizer
36.41% :     0.006159s :      1: type_inference


TotalTime = 0.0224237, [33]
[bootstrap]: 0.00043865
[type_inference]: 0.00537992
[event_method]: 2.062e-05
[auto_monad]: 6.984e-05
[graph_reusing]: 6.12001e-06
[pre_auto_parallel]: 3.08e-06
[py_interpret_to_execute]: 5.402e-05
[rewriter_before_opt_a]: 7.635e-05
[expand_dump_flag]: 3.08998e-06
[jit_opt_a]: 0.0070728, [2]
    [Cycle 1]: 0.00249749, [27]
        [switch_simplify]: 6.701e-05
        [loop_unroll]: 3.009e-05
        [a_1]: 0.00059288
        [with_stream_mark]: 2.259e-05
        [recompute_prepare]: 7.16999e-06
        [updatestate_depend_eliminate]: 4.2e-06
        [updatestate_assign_eliminate]: 2.96999e-06
        [updatestate_loads_eliminate]: 2.38998e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 5.96998e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 5.30001e-06
        [meta_shard_fg_expand]: 1.99999e-06
        [get_grad_eliminate_]: 5.25999e-06
        [merge_forward]: 3.88999e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.499e-05
        [j_node_and_user_rematch]: 9.24e-06
        [meta_fg_expand]: 2.47001e-06
        [replace_old_param]: 9.76e-06
        [inline_without_move]: 5.10001e-06
        [renormalize]: 0.00141785
        [add_forward_monad_depend]: 9.31e-06
        [auto_monad_grad]: 2.74999e-06
        [auto_monad_eliminator]: 2.071e-05
        [cse]: 3.406e-05
        [replace_applicator]: 1.747e-05
    [Cycle 2]: 0.00034305, [27]
        [switch_simplify]: 6.33e-06
        [loop_unroll]: 4.83001e-06
        [a_1]: 9.717e-05
        [with_stream_mark]: 1.23e-05
        [recompute_prepare]: 5.20001e-06
        [updatestate_depend_eliminate]: 3.66999e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.35999e-06
        [specialize_transform]: 5.46e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 5.32001e-06
        [meta_shard_fg_expand]: 1.89999e-06
        [get_grad_eliminate_]: 4.45e-06
        [merge_forward]: 2.98998e-06
        [cell_reuse_recompute_pass]: 2.61e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.567e-05
        [j_node_and_user_rematch]: 8.55001e-06
        [meta_fg_expand]: 1.77999e-06
        [replace_old_param]: 7.97e-06
        [inline_without_move]: 4.38999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.49e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 4.86002e-06
        [cse]: 1.075e-05
        [replace_applicator]: 5.87999e-06
[py_interpret_to_execute_after_opt_a]: 1.537e-05
[rewriter_after_opt_a]: 3.873e-05
[convert_after_rewriter]: 7.28999e-06
[order_py_execute_after_rewriter]: 5.03002e-06
[mutable_eliminate]: 0.00070809
[jit_opt_b]: 5.242e-05, [1]
    [Cycle 1]: 4.474e-05, [2]
        [frontend_op_eliminate]: 1.688e-05
        [inline_after_opt_a]: 1.523e-05
[cconv]: 2.81e-05
[loop_unroll]: 0.00042687
[jit_opt_after_cconv]: 0.00014051, [1]
    [Cycle 1]: 0.00013394, [11]
        [c_1]: 2.085e-05
        [parameter_eliminate]: 3.28e-06
        [updatestate_depend_eliminate]: 5.92001e-06
        [updatestate_assign_eliminate]: 2.66999e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [cse]: 1.749e-05
        [call_graph_tuple_transform]: 1.97e-05
        [tuple_list_get_item_eliminator]: 5.12e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.06997e-06
[remove_dup_value]: 1.389e-05
[partial_unused_args_eliminate]: 2.49001e-06
[environ_conv]: 5.61e-06
[add_recomputation]: 5.255e-05
[cse_after_recomputation]: 2.016e-05, [1]
    [Cycle 1]: 1.481e-05, [1]
        [cse]: 8.79998e-06
[auto_monad_reorder]: 1.492e-05
[get_jit_bprop_graph]: 2.11998e-06
[rewriter_after_jit_bprop_graph]: 5.40001e-06
[opt_after_jit_grad]: 0.00045344
[symbol_engine_optimizer]: 7.48e-05, [1]
    [Cycle 1]: 6.846e-05, [6]
        [build]: 4.62e-06
        [elim_shapecalc]: 7.35e-06
        [elim_not_effective]: 1.342e-05
        [opt_reshape]: 5.62999e-06
        [fold_const_symbol]: 9.18002e-06
        [renormalize]: 3.69997e-07
[validate]: 3.976e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00695542
[execute]: 9.94001e-06
Sums
    bootstrap                            :     0.000439s :  2.50%
    type_inference                       :     0.005380s : 30.69%
    event_method                         :     0.000021s :  0.12%
    auto_monad                           :     0.000070s :  0.40%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000054s :  0.31%
    rewriter_before_opt_a                :     0.000076s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000073s :  0.42%
    jit_opt_a.loop_unroll                :     0.000035s :  0.20%
    jit_opt_a.a_1                        :     0.000690s :  3.94%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000009s :  0.05%
    jit_opt_a.renormalize                :     0.001418s :  8.09%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.15%
    jit_opt_a.cse                        :     0.000045s :  0.26%
    jit_opt_a.replace_applicator         :     0.000023s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.09%
    rewriter_after_opt_a                 :     0.000039s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000708s :  4.04%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000028s :  0.16%
    loop_unroll                          :     0.000427s :  2.43%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000053s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000453s :  2.59%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000040s :  0.23%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006955s : 39.67%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000253    23
 0.95% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.19% :     0.000006s :      3: substitution.graph_param_transform
85.67% :     0.000217s :      4: substitution.inline
 1.45% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.77% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.30% :     0.000006s :      2: substitution.replace_old_param
 5.16% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005290     2
80.66% :     0.004267s :      1: type_inference.infer
19.34% :     0.001023s :      1: type_inference.specialize
------[replace.]   0.000072     6
68.71% :     0.000049s :      4: replace.inline
31.29% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000227     6
94.69% :     0.000215s :      4: match.inline
 5.31% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.73% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.17% :     0.000001s :     10: predicate.addn_zero_filter
 2.16% :     0.000003s :     10: predicate.arithmetic_simplify
 1.15% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.53% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.57% :     0.000002s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.90% :     0.000007s :     22: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.16% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.60% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.13% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 1.84% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.34% :     0.000002s :     10: predicate.reduce_eliminate
 1.99% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.94% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 2.00% :     0.000002s :     10: predicate.reshape_eliminate
 1.08% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.73% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 1.21% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.58% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.23% :     0.000003s :     16: predicate.switch_defer_inline
 1.90% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.56% :     0.000009s :     53: predicate.switch_simplify
 1.58% :     0.000002s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.60% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.06% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.60% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.99% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000736    12
48.20% :     0.000355s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.19% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.60% :     0.000328s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024849    76
 0.22% :     0.000056s :      1: add_recomputation
 0.29% :     0.000073s :      1: auto_monad
 0.07% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.84% :     0.000457s :      1: bootstrap
 0.12% :     0.000031s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000026s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
28.48% :     0.007076s :      1: jit_opt_a
 0.58% :     0.000143s :      1: jit_opt_after_cconv
 0.22% :     0.000055s :      1: jit_opt_b
 1.75% :     0.000434s :      1: loop_unroll
 2.89% :     0.000717s :      1: mutable_eliminate
 3.70% :     0.000919s :     26: opt.transform.jit_opt_a
 0.19% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.04% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.85% :     0.000460s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000056s :      1: py_interpret_to_execute
 0.07% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.05% :     0.000508s :      1: renormalize.infer
 3.62% :     0.000899s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000042s :      1: rewriter_after_opt_a
 0.32% :     0.000079s :      1: rewriter_before_opt_a
 0.31% :     0.000078s :      1: symbol_engine_optimizer
28.08% :     0.006977s :      1: task_emit
21.73% :     0.005399s :      1: type_inference
 0.25% :     0.000062s :      1: validate


TotalTime = 0.0141666, [30]
[bootstrap]: 0.00026393
[type_inference]: 0.00610895
[event_method]: 2.033e-05
[auto_monad]: 8.036e-05
[graph_reusing]: 6.13002e-06
[pre_auto_parallel]: 2.88e-06
[py_interpret_to_execute]: 4.673e-05
[rewriter_before_opt_a]: 8.746e-05
[expand_dump_flag]: 2.96999e-06
[jit_opt_a]: 0.00553309, [2]
    [Cycle 1]: 0.00206771, [27]
        [switch_simplify]: 0.00011987
        [loop_unroll]: 3.362e-05
        [a_1]: 0.0009016
        [with_stream_mark]: 2.145e-05
        [recompute_prepare]: 7.98001e-06
        [updatestate_depend_eliminate]: 3.63e-06
        [updatestate_assign_eliminate]: 3.18998e-06
        [updatestate_loads_eliminate]: 2.86999e-06
        [parameter_eliminate]: 1.78002e-06
        [specialize_transform]: 5.94e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 5.59998e-06
        [meta_shard_fg_expand]: 2.11e-06
        [get_grad_eliminate_]: 5.38002e-06
        [merge_forward]: 3.61999e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.418e-05
        [j_node_and_user_rematch]: 9.42001e-06
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 9.72999e-06
        [inline_without_move]: 5.17e-06
        [renormalize]: 0.0006654
        [add_forward_monad_depend]: 4.93001e-06
        [auto_monad_grad]: 2.57001e-06
        [auto_monad_eliminator]: 1.285e-05
        [cse]: 2.984e-05
        [replace_applicator]: 1.291e-05
    [Cycle 2]: 0.0003045, [27]
        [switch_simplify]: 5.27001e-06
        [loop_unroll]: 4.75001e-06
        [a_1]: 8.664e-05
        [with_stream_mark]: 8.55999e-06
        [recompute_prepare]: 4.95999e-06
        [updatestate_depend_eliminate]: 2.46998e-06
        [updatestate_assign_eliminate]: 2.14999e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 4.72e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 4.56002e-06
        [meta_shard_fg_expand]: 1.22999e-06
        [get_grad_eliminate_]: 4.47998e-06
        [merge_forward]: 2.83e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.307e-05
        [j_node_and_user_rematch]: 7.50998e-06
        [meta_fg_expand]: 1.58002e-06
        [replace_old_param]: 6.80002e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.11002e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 4.74e-06
        [cse]: 9.72001e-06
        [replace_applicator]: 5.05999e-06
[py_interpret_to_execute_after_opt_a]: 1.028e-05
[rewriter_after_opt_a]: 3.434e-05
[convert_after_rewriter]: 6.73e-06
[order_py_execute_after_rewriter]: 5.30001e-06
[mutable_eliminate]: 0.0005137
[jit_opt_b]: 5.554e-05, [1]
    [Cycle 1]: 4.81e-05, [2]
        [frontend_op_eliminate]: 1.531e-05
        [inline_after_opt_a]: 2.163e-05
[cconv]: 2.461e-05
[loop_unroll]: 0.00040193
[jit_opt_after_cconv]: 0.00013352, [1]
    [Cycle 1]: 0.00012692, [11]
        [c_1]: 1.996e-05
        [parameter_eliminate]: 2.36e-06
        [updatestate_depend_eliminate]: 4.95001e-06
        [updatestate_assign_eliminate]: 2.36998e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [cse]: 1.66e-05
        [call_graph_tuple_transform]: 1.818e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.54998e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 5.25001e-06
[remove_dup_value]: 1.311e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 5.00001e-06
[add_recomputation]: 4.794e-05
[cse_after_recomputation]: 2.094e-05, [1]
    [Cycle 1]: 1.567e-05, [1]
        [cse]: 9.54e-06
[auto_monad_reorder]: 1.45e-05
[get_jit_bprop_graph]: 1.93002e-06
[rewriter_after_jit_bprop_graph]: 5.15001e-06
[opt_after_jit_grad]: 0.00043981
[symbol_engine_optimizer]: 7.094e-05, [1]
    [Cycle 1]: 6.496e-05, [6]
        [build]: 3.66001e-06
        [elim_shapecalc]: 7.22002e-06
        [elim_not_effective]: 1.218e-05
        [opt_reshape]: 5.56e-06
        [fold_const_symbol]: 8.24998e-06
        [renormalize]: 6.29982e-07
[validate]: 3.52e-05
Sums
    bootstrap                            :     0.000264s :  2.53%
    type_inference                       :     0.006109s : 58.54%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000080s :  0.77%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000047s :  0.45%
    rewriter_before_opt_a                :     0.000087s :  0.84%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000125s :  1.20%
    jit_opt_a.loop_unroll                :     0.000038s :  0.37%
    jit_opt_a.a_1                        :     0.000988s :  9.47%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000665s :  6.38%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.17%
    jit_opt_a.cse                        :     0.000040s :  0.38%
    jit_opt_a.replace_applicator         :     0.000018s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.10%
    rewriter_after_opt_a                 :     0.000034s :  0.33%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000514s :  4.92%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000022s :  0.21%
    cconv                                :     0.000025s :  0.24%
    loop_unroll                          :     0.000402s :  3.85%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000048s :  0.46%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000014s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000440s :  4.21%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000035s :  0.34%


Time group info:
------[substitution.]   0.000295    26
 0.62% :     0.000002s :      2: substitution.elim_not_effective
 0.41% :     0.000001s :      2: substitution.fold_const_symbol
 1.94% :     0.000006s :      3: substitution.graph_param_transform
85.61% :     0.000253s :      6: substitution.inline
 1.05% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.62% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.51% :     0.000004s :      2: substitution.replace_old_param
 2.63% :     0.000008s :      1: substitution.switch_simplify
 4.62% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006045     2
83.53% :     0.005050s :      1: type_inference.infer
16.47% :     0.000996s :      1: type_inference.specialize
------[replace.]   0.000325     9
79.34% :     0.000258s :      6: replace.inline
11.36% :     0.000037s :      1: replace.switch_simplify
 9.30% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000269     9
92.79% :     0.000250s :      6: match.inline
 2.61% :     0.000007s :      1: match.switch_simplify
 4.59% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   801
 1.51% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.94% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     12: predicate.addn_check_dump
 1.35% :     0.000002s :     12: predicate.addn_zero_filter
 2.20% :     0.000003s :     12: predicate.arithmetic_simplify
 1.35% :     0.000002s :     12: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     12: predicate.compare_switch_simplify
 1.43% :     0.000002s :     12: predicate.depend_value_elim
 1.20% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.54% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.24% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     12: predicate.environ_get_eliminate
 1.13% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.83% :     0.000008s :     26: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.87% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     17: predicate.load_eliminater
 1.23% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000002s :     12: predicate.merge_addn
 1.16% :     0.000002s :     12: predicate.minmaximum_grad
 1.27% :     0.000002s :      3: predicate.mutable_eliminate
 0.34% :     0.000000s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     17: predicate.partial_eliminate
 1.32% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     12: predicate.reduce_eliminate
 1.48% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.18% :     0.000003s :     20: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     12: predicate.reshape_eliminate
 1.33% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.79% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.32% :     0.000004s :     20: predicate.switch_defer_inline
 2.40% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.44% :     0.000011s :     65: predicate.switch_simplify
 1.39% :     0.000002s :     12: predicate.tile_eliminate
 1.26% :     0.000002s :     12: predicate.transpose_eliminate
 1.32% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.97% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.84% :     0.000002s :     12: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000758    15
49.61% :     0.000376s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.43% :     0.000102s :      1: func_graph_cloner_run.FuncGraphClonerNode
36.96% :     0.000280s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016151    72
 0.31% :     0.000051s :      1: add_recomputation
 0.52% :     0.000084s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 1.75% :     0.000283s :      1: bootstrap
 0.17% :     0.000027s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
34.28% :     0.005536s :      1: jit_opt_a
 0.85% :     0.000136s :      1: jit_opt_after_cconv
 0.36% :     0.000058s :      1: jit_opt_b
 2.53% :     0.000409s :      1: loop_unroll
 3.23% :     0.000522s :      1: mutable_eliminate
 7.83% :     0.001265s :     26: opt.transform.jit_opt_a
 0.28% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000030s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.77% :     0.000448s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000049s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000015s :      1: remove_dup_value
 2.38% :     0.000384s :      1: renormalize.infer
 1.69% :     0.000274s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000038s :      1: rewriter_after_opt_a
 0.56% :     0.000090s :      1: rewriter_before_opt_a
 0.45% :     0.000073s :      1: symbol_engine_optimizer
37.90% :     0.006122s :      1: type_inference


TotalTime = 0.0521197, [33]
[bootstrap]: 0.00036035
[type_inference]: 0.00489307
[event_method]: 1.903e-05
[auto_monad]: 6.476e-05
[graph_reusing]: 5.91998e-06
[pre_auto_parallel]: 3.13e-06
[py_interpret_to_execute]: 4.106e-05
[rewriter_before_opt_a]: 7.449e-05
[expand_dump_flag]: 3.33998e-06
[jit_opt_a]: 0.0354589, [2]
    [Cycle 1]: 0.00177975, [27]
        [switch_simplify]: 5.396e-05
        [loop_unroll]: 2.903e-05
        [a_1]: 0.00050424
        [with_stream_mark]: 1.253e-05
        [recompute_prepare]: 7.21999e-06
        [updatestate_depend_eliminate]: 2.76999e-06
        [updatestate_assign_eliminate]: 1.99999e-06
        [updatestate_loads_eliminate]: 1.77001e-06
        [parameter_eliminate]: 7.50006e-07
        [specialize_transform]: 5.72001e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.64e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 8.39995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.1e-05
        [j_node_and_user_rematch]: 9.40001e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 8.94e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 0.00086475
        [add_forward_monad_depend]: 6.71e-06
        [auto_monad_grad]: 2.65002e-06
        [auto_monad_eliminator]: 1.481e-05
        [cse]: 3.176e-05
        [replace_applicator]: 1.643e-05
    [Cycle 2]: 0.00032677, [27]
        [switch_simplify]: 6.34001e-06
        [loop_unroll]: 4.93001e-06
        [a_1]: 9.443e-05
        [with_stream_mark]: 1.086e-05
        [recompute_prepare]: 4.93001e-06
        [updatestate_depend_eliminate]: 2.61999e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.83002e-06
        [parameter_eliminate]: 1.37999e-06
        [specialize_transform]: 4.70001e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 4.98001e-06
        [meta_shard_fg_expand]: 1.32999e-06
        [get_grad_eliminate_]: 4.52998e-06
        [merge_forward]: 2.74001e-06
        [cell_reuse_recompute_pass]: 2.05002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.269e-05
        [j_node_and_user_rematch]: 7.91001e-06
        [meta_fg_expand]: 1.96998e-06
        [replace_old_param]: 7.73001e-06
        [inline_without_move]: 4.80001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.21997e-06
        [auto_monad_grad]: 6.79982e-07
        [auto_monad_eliminator]: 4.94e-06
        [cse]: 1.028e-05
        [replace_applicator]: 4.82e-06
[py_interpret_to_execute_after_opt_a]: 1.171e-05
[rewriter_after_opt_a]: 3.816e-05
[convert_after_rewriter]: 7.30998e-06
[order_py_execute_after_rewriter]: 4.91997e-06
[mutable_eliminate]: 0.00082942
[jit_opt_b]: 5.32e-05, [1]
    [Cycle 1]: 4.577e-05, [2]
        [frontend_op_eliminate]: 1.836e-05
        [inline_after_opt_a]: 1.531e-05
[cconv]: 2.673e-05
[loop_unroll]: 0.00053977
[jit_opt_after_cconv]: 0.00014946, [1]
    [Cycle 1]: 0.00014267, [11]
        [c_1]: 2.216e-05
        [parameter_eliminate]: 3.33e-06
        [updatestate_depend_eliminate]: 5.92999e-06
        [updatestate_assign_eliminate]: 2.40997e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 2.125e-05
        [call_graph_tuple_transform]: 2.083e-05
        [tuple_list_get_item_eliminator]: 5.74e-06
        [none_parameter_eliminate]: 1.41002e-06
        [renormalize]: 4.60015e-07
        [switch_simplify]: 5.64998e-06
[remove_dup_value]: 1.573e-05
[partial_unused_args_eliminate]: 2.47001e-06
[environ_conv]: 6.11e-06
[add_recomputation]: 5.337e-05
[cse_after_recomputation]: 2.189e-05, [1]
    [Cycle 1]: 1.622e-05, [1]
        [cse]: 9.44998e-06
[auto_monad_reorder]: 1.563e-05
[get_jit_bprop_graph]: 1.78002e-06
[rewriter_after_jit_bprop_graph]: 7.01001e-06
[opt_after_jit_grad]: 0.0004493
[symbol_engine_optimizer]: 7.692e-05, [1]
    [Cycle 1]: 7.045e-05, [6]
        [build]: 3.66001e-06
        [elim_shapecalc]: 8.28001e-06
        [elim_not_effective]: 1.305e-05
        [opt_reshape]: 6.29001e-06
        [fold_const_symbol]: 9.82001e-06
        [renormalize]: 4.7998e-07
[validate]: 4.222e-05
[backend_pass]: 9.60019e-07
[task_emit]: 0.00861485
[execute]: 6.64999e-06
Sums
    bootstrap                            :     0.000360s :  1.99%
    type_inference                       :     0.004893s : 26.97%
    event_method                         :     0.000019s :  0.10%
    auto_monad                           :     0.000065s :  0.36%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000041s :  0.23%
    rewriter_before_opt_a                :     0.000074s :  0.41%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000060s :  0.33%
    jit_opt_a.loop_unroll                :     0.000034s :  0.19%
    jit_opt_a.a_1                        :     0.000599s :  3.30%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000010s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000865s :  4.77%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.11%
    jit_opt_a.cse                        :     0.000042s :  0.23%
    jit_opt_a.replace_applicator         :     0.000021s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.06%
    rewriter_after_opt_a                 :     0.000038s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000829s :  4.57%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000027s :  0.15%
    loop_unroll                          :     0.000540s :  2.98%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000053s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000449s :  2.48%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000042s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008615s : 47.49%
    execute                              :     0.000007s :  0.04%


TotalTime = 0.0214025, [33]
[bootstrap]: 0.00034166
[type_inference]: 0.00541963
[event_method]: 1.952e-05
[auto_monad]: 6.837e-05
[graph_reusing]: 6.39999e-06
[pre_auto_parallel]: 2.82002e-06
[py_interpret_to_execute]: 4.688e-05
[rewriter_before_opt_a]: 7.387e-05
[expand_dump_flag]: 3.02002e-06
[jit_opt_a]: 0.00638591, [2]
    [Cycle 1]: 0.00200817, [27]
        [switch_simplify]: 6.015e-05
        [loop_unroll]: 2.829e-05
        [a_1]: 0.00055348
        [with_stream_mark]: 2.363e-05
        [recompute_prepare]: 7.06001e-06
        [updatestate_depend_eliminate]: 4.18999e-06
        [updatestate_assign_eliminate]: 3.09001e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 2.44001e-06
        [specialize_transform]: 5.86e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 5.72001e-06
        [meta_shard_fg_expand]: 2.59001e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 3.89002e-06
        [cell_reuse_recompute_pass]: 1.76003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.723e-05
        [j_node_and_user_rematch]: 9.14e-06
        [meta_fg_expand]: 2.56998e-06
        [replace_old_param]: 1.034e-05
        [inline_without_move]: 6.04001e-06
        [renormalize]: 0.00100235
        [add_forward_monad_depend]: 5.56e-06
        [auto_monad_grad]: 2.54999e-06
        [auto_monad_eliminator]: 1.397e-05
        [cse]: 3.085e-05
        [replace_applicator]: 1.215e-05
    [Cycle 2]: 0.00032074, [27]
        [switch_simplify]: 5.47001e-06
        [loop_unroll]: 4.94998e-06
        [a_1]: 9.301e-05
        [with_stream_mark]: 9.04e-06
        [recompute_prepare]: 5.04e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 9.10019e-07
        [specialize_transform]: 4.95999e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 4.97e-06
        [meta_shard_fg_expand]: 1.35001e-06
        [get_grad_eliminate_]: 4.4e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 1.97001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.27e-05
        [j_node_and_user_rematch]: 7.36999e-06
        [meta_fg_expand]: 1.44998e-06
        [replace_old_param]: 7.11999e-06
        [inline_without_move]: 4.45e-06
        [renormalize]: 1.10012e-07
        [add_forward_monad_depend]: 1.17999e-06
        [auto_monad_grad]: 7.79983e-07
        [auto_monad_eliminator]: 5.54998e-06
        [cse]: 1.023e-05
        [replace_applicator]: 4.82e-06
[py_interpret_to_execute_after_opt_a]: 1.184e-05
[rewriter_after_opt_a]: 3.71e-05
[convert_after_rewriter]: 7.12002e-06
[order_py_execute_after_rewriter]: 4.95999e-06
[mutable_eliminate]: 0.00062007
[jit_opt_b]: 5.029e-05, [1]
    [Cycle 1]: 4.32e-05, [2]
        [frontend_op_eliminate]: 1.666e-05
        [inline_after_opt_a]: 1.455e-05
[cconv]: 2.641e-05
[loop_unroll]: 0.00042228
[jit_opt_after_cconv]: 0.00013675, [1]
    [Cycle 1]: 0.00013015, [11]
        [c_1]: 2.028e-05
        [parameter_eliminate]: 2.59001e-06
        [updatestate_depend_eliminate]: 5.55001e-06
        [updatestate_assign_eliminate]: 2.30002e-06
        [updatestate_loads_eliminate]: 1.95001e-06
        [cse]: 1.775e-05
        [call_graph_tuple_transform]: 1.846e-05
        [tuple_list_get_item_eliminator]: 5.03002e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.26998e-06
[remove_dup_value]: 1.394e-05
[partial_unused_args_eliminate]: 2.53e-06
[environ_conv]: 6.49999e-06
[add_recomputation]: 5.251e-05
[cse_after_recomputation]: 2.109e-05, [1]
    [Cycle 1]: 1.531e-05, [1]
        [cse]: 8.22998e-06
[auto_monad_reorder]: 1.587e-05
[get_jit_bprop_graph]: 2.49001e-06
[rewriter_after_jit_bprop_graph]: 6.27001e-06
[opt_after_jit_grad]: 0.00055494
[symbol_engine_optimizer]: 9.01e-05, [1]
    [Cycle 1]: 8.161e-05, [6]
        [build]: 4.31002e-06
        [elim_shapecalc]: 1.786e-05
        [elim_not_effective]: 1.547e-05
        [opt_reshape]: 5.73002e-06
        [fold_const_symbol]: 8.92999e-06
        [renormalize]: 4.69998e-07
[validate]: 4.278e-05
[backend_pass]: 8.09989e-07
[task_emit]: 0.00666713
[execute]: 1.05e-05
Sums
    bootstrap                            :     0.000342s :  2.05%
    type_inference                       :     0.005420s : 32.46%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000068s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000047s :  0.28%
    rewriter_before_opt_a                :     0.000074s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.39%
    jit_opt_a.loop_unroll                :     0.000033s :  0.20%
    jit_opt_a.a_1                        :     0.000646s :  3.87%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001002s :  6.00%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000041s :  0.25%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000620s :  3.71%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000026s :  0.16%
    loop_unroll                          :     0.000422s :  2.53%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000053s :  0.31%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000555s :  3.32%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000018s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000043s :  0.26%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006667s : 39.94%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000175    23
 1.23% :     0.000002s :      2: substitution.elim_not_effective
 0.95% :     0.000002s :      2: substitution.fold_const_symbol
 3.27% :     0.000006s :      3: substitution.graph_param_transform
82.83% :     0.000145s :      4: substitution.inline
 2.05% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.71% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.35% :     0.000004s :      2: substitution.replace_old_param
 4.60% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004816     2
80.52% :     0.003878s :      1: type_inference.infer
19.48% :     0.000938s :      1: type_inference.specialize
------[replace.]   0.000068     6
65.35% :     0.000045s :      4: replace.inline
34.65% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000149     6
95.32% :     0.000142s :      4: match.inline
 4.68% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.60% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.28% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.23% :     0.000001s :     10: predicate.addn_zero_filter
 2.00% :     0.000002s :     10: predicate.arithmetic_simplify
 1.43% :     0.000002s :     10: predicate.cast_eliminate
 0.69% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.54% :     0.000002s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.48% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.31% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.50% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.07% :     0.000001s :      6: predicate.less_batch_normalization
 1.72% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.57% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.92% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.48% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.51% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 1.93% :     0.000002s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.74% :     0.000002s :     10: predicate.reduce_eliminate
 1.76% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.92% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.41% :     0.000002s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.65% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.71% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.07% :     0.000008s :     53: predicate.switch_simplify
 1.28% :     0.000002s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.27% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.77% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.94% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000654    12
48.75% :     0.000319s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.10% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.15% :     0.000308s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.053905    76
 0.10% :     0.000056s :      1: add_recomputation
 0.13% :     0.000068s :      1: auto_monad
 0.03% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.71% :     0.000381s :      1: bootstrap
 0.05% :     0.000030s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.04% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.04% :     0.000023s :      1: event_method
 0.02% :     0.000011s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
65.79% :     0.035463s :      1: jit_opt_a
 0.28% :     0.000152s :      1: jit_opt_after_cconv
 0.10% :     0.000056s :      1: jit_opt_b
 1.02% :     0.000549s :      1: loop_unroll
 1.56% :     0.000839s :      1: mutable_eliminate
 1.51% :     0.000812s :     26: opt.transform.jit_opt_a
 0.09% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000028s :      4: opt.transform.jit_opt_b
 0.03% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000034s :      4: opt.transform.symbol_engine_opt
 0.85% :     0.000456s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000044s :      1: py_interpret_to_execute
 0.03% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000018s :      1: remove_dup_value
 0.92% :     0.000495s :      1: renormalize.infer
 0.67% :     0.000363s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000041s :      1: rewriter_after_opt_a
 0.14% :     0.000077s :      1: rewriter_before_opt_a
 0.15% :     0.000080s :      1: symbol_engine_optimizer
16.01% :     0.008632s :      1: task_emit
 9.11% :     0.004910s :      1: type_inference
 0.12% :     0.000064s :      1: validate


Time group info:
------[substitution.]   0.000218    23
 1.15% :     0.000003s :      2: substitution.elim_not_effective
 0.74% :     0.000002s :      2: substitution.fold_const_symbol
 2.63% :     0.000006s :      3: substitution.graph_param_transform
82.57% :     0.000180s :      4: substitution.inline
 1.50% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.82% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.30% :     0.000005s :      2: substitution.replace_old_param
 6.30% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005334     2
79.55% :     0.004243s :      1: type_inference.infer
20.45% :     0.001091s :      1: type_inference.specialize
------[replace.]   0.000072     6
64.29% :     0.000046s :      4: replace.inline
35.71% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000190     6
93.34% :     0.000178s :      4: match.inline
 6.66% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.70% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 2.06% :     0.000002s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000001s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.58% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.65% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.47% :     0.000001s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.63% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.25% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.59% :     0.000007s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.07% :     0.000001s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.82% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.77% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     15: predicate.partial_eliminate
 1.24% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.72% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.65% :     0.000002s :     10: predicate.reshape_eliminate
 1.11% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.45% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.22% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.58% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.68% :     0.000009s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.53% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.64% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.98% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000802    12
47.02% :     0.000377s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.58% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.40% :     0.000380s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023363    76
 0.24% :     0.000056s :      1: add_recomputation
 0.31% :     0.000072s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.56% :     0.000363s :      1: bootstrap
 0.12% :     0.000029s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
27.35% :     0.006390s :      1: jit_opt_a
 0.60% :     0.000139s :      1: jit_opt_after_cconv
 0.23% :     0.000053s :      1: jit_opt_b
 1.84% :     0.000430s :      1: loop_unroll
 2.69% :     0.000629s :      1: mutable_eliminate
 3.68% :     0.000860s :     26: opt.transform.jit_opt_a
 0.19% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.42% :     0.000564s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000050s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.45% :     0.000573s :      1: renormalize.infer
 1.80% :     0.000421s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000040s :      1: rewriter_after_opt_a
 0.33% :     0.000077s :      1: rewriter_before_opt_a
 0.40% :     0.000093s :      1: symbol_engine_optimizer
28.62% :     0.006685s :      1: task_emit
23.27% :     0.005436s :      1: type_inference
 0.28% :     0.000066s :      1: validate


TotalTime = 0.0148617, [30]
[bootstrap]: 0.00026213
[type_inference]: 0.00641231
[event_method]: 2.183e-05
[auto_monad]: 8.17e-05
[graph_reusing]: 6.19999e-06
[pre_auto_parallel]: 2.71e-06
[py_interpret_to_execute]: 4.5e-05
[rewriter_before_opt_a]: 8.509e-05
[expand_dump_flag]: 3.31999e-06
[jit_opt_a]: 0.00551042, [2]
    [Cycle 1]: 0.00198968, [27]
        [switch_simplify]: 0.00012571
        [loop_unroll]: 3.725e-05
        [a_1]: 0.00073353
        [with_stream_mark]: 2.321e-05
        [recompute_prepare]: 8.48999e-06
        [updatestate_depend_eliminate]: 4.18999e-06
        [updatestate_assign_eliminate]: 3.27002e-06
        [updatestate_loads_eliminate]: 2.69999e-06
        [parameter_eliminate]: 2.26998e-06
        [specialize_transform]: 6.49999e-06
        [updatestate_useless_node_eliminater]: 5.47999e-06
        [accelerated_algorithm]: 6.76999e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 6.29001e-06
        [merge_forward]: 3.64002e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.66e-05
        [j_node_and_user_rematch]: 1.093e-05
        [meta_fg_expand]: 2.29001e-06
        [replace_old_param]: 1.06e-05
        [inline_without_move]: 5.57001e-06
        [renormalize]: 0.00070661
        [add_forward_monad_depend]: 6.13002e-06
        [auto_monad_grad]: 2.93998e-06
        [auto_monad_eliminator]: 1.574e-05
        [cse]: 3.229e-05
        [replace_applicator]: 1.509e-05
    [Cycle 2]: 0.00035115, [27]
        [switch_simplify]: 6.44999e-06
        [loop_unroll]: 5.80002e-06
        [a_1]: 9.997e-05
        [with_stream_mark]: 1.161e-05
        [recompute_prepare]: 6.23e-06
        [updatestate_depend_eliminate]: 3.04999e-06
        [updatestate_assign_eliminate]: 2.41998e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [parameter_eliminate]: 1.25999e-06
        [specialize_transform]: 5.93998e-06
        [updatestate_useless_node_eliminater]: 5.57001e-06
        [accelerated_algorithm]: 5.26998e-06
        [meta_shard_fg_expand]: 1.60999e-06
        [get_grad_eliminate_]: 5.05999e-06
        [merge_forward]: 3.12002e-06
        [cell_reuse_recompute_pass]: 1.96998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.597e-05
        [j_node_and_user_rematch]: 8.93002e-06
        [meta_fg_expand]: 1.40001e-06
        [replace_old_param]: 8.44998e-06
        [inline_without_move]: 5.11002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.38002e-06
        [auto_monad_grad]: 1.10999e-06
        [auto_monad_eliminator]: 4.92e-06
        [cse]: 1.208e-05
        [replace_applicator]: 5.02e-06
[py_interpret_to_execute_after_opt_a]: 1.411e-05
[rewriter_after_opt_a]: 3.802e-05
[convert_after_rewriter]: 7.63999e-06
[order_py_execute_after_rewriter]: 5.24e-06
[mutable_eliminate]: 0.00067562
[jit_opt_b]: 6.274e-05, [1]
    [Cycle 1]: 5.414e-05, [2]
        [frontend_op_eliminate]: 1.639e-05
        [inline_after_opt_a]: 1.816e-05
[cconv]: 3.223e-05
[loop_unroll]: 0.00047387
[jit_opt_after_cconv]: 0.00016351, [1]
    [Cycle 1]: 0.00015675, [11]
        [c_1]: 2.304e-05
        [parameter_eliminate]: 4.05e-06
        [updatestate_depend_eliminate]: 7.33e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.33998e-06
        [cse]: 2.438e-05
        [call_graph_tuple_transform]: 2.248e-05
        [tuple_list_get_item_eliminator]: 6.76999e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 6.42001e-06
[remove_dup_value]: 2.757e-05
[partial_unused_args_eliminate]: 3.46999e-06
[environ_conv]: 6.21e-06
[add_recomputation]: 5.331e-05
[cse_after_recomputation]: 2.372e-05, [1]
    [Cycle 1]: 1.773e-05, [1]
        [cse]: 1.161e-05
[auto_monad_reorder]: 1.582e-05
[get_jit_bprop_graph]: 2.63e-06
[rewriter_after_jit_bprop_graph]: 7.53e-06
[opt_after_jit_grad]: 0.00050329
[symbol_engine_optimizer]: 7.92e-05, [1]
    [Cycle 1]: 7.188e-05, [6]
        [build]: 4.23999e-06
        [elim_shapecalc]: 8.03999e-06
        [elim_not_effective]: 1.416e-05
        [opt_reshape]: 6.24001e-06
        [fold_const_symbol]: 8.72e-06
        [renormalize]: 4.40021e-07
[validate]: 4.096e-05
Sums
    bootstrap                            :     0.000262s :  2.37%
    type_inference                       :     0.006412s : 58.05%
    event_method                         :     0.000022s :  0.20%
    auto_monad                           :     0.000082s :  0.74%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000045s :  0.41%
    rewriter_before_opt_a                :     0.000085s :  0.77%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000132s :  1.20%
    jit_opt_a.loop_unroll                :     0.000043s :  0.39%
    jit_opt_a.a_1                        :     0.000834s :  7.55%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.17%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000707s :  6.40%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.19%
    jit_opt_a.cse                        :     0.000044s :  0.40%
    jit_opt_a.replace_applicator         :     0.000020s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.13%
    rewriter_after_opt_a                 :     0.000038s :  0.34%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000676s :  6.12%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.16%
    cconv                                :     0.000032s :  0.29%
    loop_unroll                          :     0.000474s :  4.29%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000028s :  0.25%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000053s :  0.48%
    cse_after_recomputation.cse          :     0.000012s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.14%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.07%
    opt_after_jit_grad                   :     0.000503s :  4.56%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000041s :  0.37%


Time group info:
------[substitution.]   0.000293    26
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.42% :     0.000001s :      2: substitution.fold_const_symbol
 1.94% :     0.000006s :      3: substitution.graph_param_transform
84.71% :     0.000248s :      6: substitution.inline
 1.30% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.96% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.72% :     0.000005s :      2: substitution.replace_old_param
 2.90% :     0.000008s :      1: substitution.switch_simplify
 4.32% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006347     2
83.23% :     0.005282s :      1: type_inference.infer
16.77% :     0.001065s :      1: type_inference.specialize
------[replace.]   0.000137     9
50.05% :     0.000069s :      6: replace.inline
27.67% :     0.000038s :      1: replace.switch_simplify
22.28% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000263     9
92.77% :     0.000244s :      6: match.inline
 2.91% :     0.000008s :      1: match.switch_simplify
 4.31% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000145   801
 1.46% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.93% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000002s :     12: predicate.addn_check_dump
 1.45% :     0.000002s :     12: predicate.addn_zero_filter
 2.21% :     0.000003s :     12: predicate.arithmetic_simplify
 1.27% :     0.000002s :     12: predicate.cast_eliminate
 0.35% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     12: predicate.compare_switch_simplify
 1.55% :     0.000002s :     12: predicate.depend_value_elim
 1.19% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     12: predicate.dict_set_item_eliminator
 1.12% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000001s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.11% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     12: predicate.environ_get_eliminate
 1.11% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.33% :     0.000000s :      3: predicate.graph_param_transform
 5.46% :     0.000008s :     26: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000002s :      6: predicate.less_batch_normalization
 1.65% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.75% :     0.000003s :     17: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.32% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.76% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000002s :     12: predicate.merge_addn
 1.14% :     0.000002s :     12: predicate.minmaximum_grad
 2.29% :     0.000003s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     17: predicate.partial_eliminate
 1.32% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     12: predicate.reduce_eliminate
 1.51% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     20: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     12: predicate.reshape_eliminate
 1.37% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.54% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     12: predicate.same_eliminate
 0.42% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.79% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.87% :     0.000004s :     20: predicate.switch_defer_inline
 2.42% :     0.000004s :     20: predicate.switch_layer_defer_inline
 8.64% :     0.000013s :     65: predicate.switch_simplify
 1.31% :     0.000002s :     12: predicate.tile_eliminate
 1.15% :     0.000002s :     12: predicate.transpose_eliminate
 1.40% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.70% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.41% :     0.000002s :     12: predicate.value_based_eliminate
 0.38% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.50% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000822    15
46.36% :     0.000381s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.48% :     0.000103s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.17% :     0.000338s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016769    72
 0.34% :     0.000057s :      1: add_recomputation
 0.51% :     0.000085s :      1: auto_monad
 0.11% :     0.000019s :      1: auto_monad_reorder
 1.68% :     0.000282s :      1: bootstrap
 0.21% :     0.000035s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
32.88% :     0.005514s :      1: jit_opt_a
 0.99% :     0.000167s :      1: jit_opt_after_cconv
 0.39% :     0.000066s :      1: jit_opt_b
 2.88% :     0.000483s :      1: loop_unroll
 4.10% :     0.000687s :      1: mutable_eliminate
 6.79% :     0.001139s :     26: opt.transform.jit_opt_a
 0.32% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.06% :     0.000514s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000006s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000048s :      1: py_interpret_to_execute
 0.10% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.18% :     0.000031s :      1: remove_dup_value
 2.37% :     0.000398s :      1: renormalize.infer
 1.79% :     0.000300s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000041s :      1: rewriter_after_opt_a
 0.53% :     0.000088s :      1: rewriter_before_opt_a
 0.49% :     0.000082s :      1: symbol_engine_optimizer
38.33% :     0.006428s :      1: type_inference


TotalTime = 0.0618347, [33]
[bootstrap]: 0.00042955
[type_inference]: 0.00929361
[event_method]: 1.823e-05
[auto_monad]: 6.673e-05
[graph_reusing]: 5.93002e-06
[pre_auto_parallel]: 2.66999e-06
[py_interpret_to_execute]: 4.449e-05
[rewriter_before_opt_a]: 6.955e-05
[expand_dump_flag]: 3.17002e-06
[jit_opt_a]: 0.00673885, [2]
    [Cycle 1]: 0.00199703, [27]
        [switch_simplify]: 5.502e-05
        [loop_unroll]: 2.865e-05
        [a_1]: 0.00059138
        [with_stream_mark]: 2.013e-05
        [recompute_prepare]: 7.68001e-06
        [updatestate_depend_eliminate]: 3.23998e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [parameter_eliminate]: 1.33002e-06
        [specialize_transform]: 5.96e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 5.63997e-06
        [meta_shard_fg_expand]: 2.53e-06
        [get_grad_eliminate_]: 4.90001e-06
        [merge_forward]: 3.78001e-06
        [cell_reuse_recompute_pass]: 1.38002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.311e-05
        [j_node_and_user_rematch]: 9.29e-06
        [meta_fg_expand]: 2.46e-06
        [replace_old_param]: 9.47999e-06
        [inline_without_move]: 5.10001e-06
        [renormalize]: 0.00098196
        [add_forward_monad_depend]: 6.53e-06
        [auto_monad_grad]: 2.56e-06
        [auto_monad_eliminator]: 1.366e-05
        [cse]: 2.769e-05
        [replace_applicator]: 1.348e-05
    [Cycle 2]: 0.00035915, [27]
        [switch_simplify]: 5.78002e-06
        [loop_unroll]: 4.63999e-06
        [a_1]: 9.637e-05
        [with_stream_mark]: 1.108e-05
        [recompute_prepare]: 5.15001e-06
        [updatestate_depend_eliminate]: 2.88998e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 5.01002e-06
        [updatestate_useless_node_eliminater]: 4.55001e-06
        [accelerated_algorithm]: 4.57e-06
        [meta_shard_fg_expand]: 1.29998e-06
        [get_grad_eliminate_]: 4.63001e-06
        [merge_forward]: 2.70002e-06
        [cell_reuse_recompute_pass]: 2.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.317e-05
        [j_node_and_user_rematch]: 7.94997e-06
        [meta_fg_expand]: 1.89e-06
        [replace_old_param]: 2.641e-05
        [inline_without_move]: 5.09e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 1.359e-05
        [cse]: 1.149e-05
        [replace_applicator]: 4.97e-06
[py_interpret_to_execute_after_opt_a]: 1.406e-05
[rewriter_after_opt_a]: 3.961e-05
[convert_after_rewriter]: 7.2e-06
[order_py_execute_after_rewriter]: 4.75001e-06
[mutable_eliminate]: 0.00078889
[jit_opt_b]: 5.395e-05, [1]
    [Cycle 1]: 4.606e-05, [2]
        [frontend_op_eliminate]: 1.8e-05
        [inline_after_opt_a]: 1.574e-05
[cconv]: 3.001e-05
[loop_unroll]: 0.0004879
[jit_opt_after_cconv]: 0.0001508, [1]
    [Cycle 1]: 0.00014346, [11]
        [c_1]: 2.234e-05
        [parameter_eliminate]: 3.25e-06
        [updatestate_depend_eliminate]: 6.43e-06
        [updatestate_assign_eliminate]: 2.64999e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [cse]: 2.045e-05
        [call_graph_tuple_transform]: 2.159e-05
        [tuple_list_get_item_eliminator]: 5.32001e-06
        [none_parameter_eliminate]: 1.74998e-06
        [renormalize]: 7.00005e-07
        [switch_simplify]: 5.25999e-06
[remove_dup_value]: 1.528e-05
[partial_unused_args_eliminate]: 3.18e-06
[environ_conv]: 6.53e-06
[add_recomputation]: 5.5e-05
[cse_after_recomputation]: 2.263e-05, [1]
    [Cycle 1]: 1.648e-05, [1]
        [cse]: 1.023e-05
[auto_monad_reorder]: 1.704e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 8.2e-06
[opt_after_jit_grad]: 0.00051178
[symbol_engine_optimizer]: 7.705e-05, [1]
    [Cycle 1]: 7.028e-05, [6]
        [build]: 3.97998e-06
        [elim_shapecalc]: 7.61001e-06
        [elim_not_effective]: 1.428e-05
        [opt_reshape]: 6.24001e-06
        [fold_const_symbol]: 9.47999e-06
        [renormalize]: 7.40023e-07
[validate]: 4.497e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.0425642
[execute]: 1.026e-05
Sums
    bootstrap                            :     0.000430s :  0.76%
    type_inference                       :     0.009294s : 16.36%
    event_method                         :     0.000018s :  0.03%
    auto_monad                           :     0.000067s :  0.12%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000044s :  0.08%
    rewriter_before_opt_a                :     0.000070s :  0.12%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000061s :  0.11%
    jit_opt_a.loop_unroll                :     0.000033s :  0.06%
    jit_opt_a.a_1                        :     0.000688s :  1.21%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.05%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.00%
    jit_opt_a.specialize_transform       :     0.000011s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.02%
    jit_opt_a.merge_forward              :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.05%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000036s :  0.06%
    jit_opt_a.inline_without_move        :     0.000010s :  0.02%
    jit_opt_a.renormalize                :     0.000982s :  1.73%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.05%
    jit_opt_a.cse                        :     0.000039s :  0.07%
    jit_opt_a.replace_applicator         :     0.000018s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.02%
    rewriter_after_opt_a                 :     0.000040s :  0.07%
    convert_after_rewriter               :     0.000007s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000789s :  1.39%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.03%
    cconv                                :     0.000030s :  0.05%
    loop_unroll                          :     0.000488s :  0.86%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.04%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000020s :  0.04%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.01%
    remove_dup_value                     :     0.000015s :  0.03%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.01%
    add_recomputation                    :     0.000055s :  0.10%
    cse_after_recomputation.cse          :     0.000010s :  0.02%
    auto_monad_reorder                   :     0.000017s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.01%
    opt_after_jit_grad                   :     0.000512s :  0.90%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.08%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.042564s : 74.94%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000269    23
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000002s :      2: substitution.fold_const_symbol
 2.25% :     0.000006s :      3: substitution.graph_param_transform
80.05% :     0.000216s :      4: substitution.inline
 1.30% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.75% :     0.000005s :      4: substitution.remove_not_recompute_node
 8.30% :     0.000022s :      2: substitution.replace_old_param
 4.93% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009213     2
86.70% :     0.007988s :      1: type_inference.infer
13.30% :     0.001225s :      1: type_inference.specialize
------[replace.]   0.000077     6
65.47% :     0.000050s :      4: replace.inline
34.53% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000225     6
94.56% :     0.000213s :      4: match.inline
 5.44% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.57% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.56% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.19% :     0.000001s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 2.12% :     0.000003s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.52% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.64% :     0.000001s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000001s :      3: predicate.graph_param_transform
 5.64% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.04% :     0.000001s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.87% :     0.000002s :     15: predicate.load_eliminater
 1.51% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.92% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 2.18% :     0.000003s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     15: predicate.partial_eliminate
 1.26% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.59% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.89% :     0.000001s :      6: predicate.replace_old_param
 0.44% :     0.000001s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.03% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.86% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.08% :     0.000003s :     16: predicate.switch_defer_inline
 1.89% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.13% :     0.000009s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.21% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.90% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.88% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 1.01% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000785    12
41.13% :     0.000323s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.23% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
52.64% :     0.000413s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.063828    76
 0.09% :     0.000058s :      1: add_recomputation
 0.11% :     0.000071s :      1: auto_monad
 0.03% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.71% :     0.000454s :      1: bootstrap
 0.05% :     0.000033s :      1: cconv
 0.01% :     0.000010s :      1: convert_after_rewriter
 0.04% :     0.000025s :      1: cse_after_recomputation
 0.01% :     0.000009s :      1: environ_conv
 0.03% :     0.000022s :      1: event_method
 0.02% :     0.000015s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000008s :      1: graph_reusing
10.57% :     0.006744s :      1: jit_opt_a
 0.24% :     0.000155s :      1: jit_opt_after_cconv
 0.09% :     0.000057s :      1: jit_opt_b
 0.78% :     0.000497s :      1: loop_unroll
 1.25% :     0.000798s :      1: mutable_eliminate
 1.44% :     0.000917s :     26: opt.transform.jit_opt_a
 0.08% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000027s :      4: opt.transform.jit_opt_b
 0.02% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.05% :     0.000034s :      4: opt.transform.symbol_engine_opt
 0.82% :     0.000521s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000047s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000018s :      1: remove_dup_value
 0.90% :     0.000577s :      1: renormalize.infer
 0.62% :     0.000396s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000042s :      1: rewriter_after_opt_a
 0.11% :     0.000073s :      1: rewriter_before_opt_a
 0.13% :     0.000080s :      1: symbol_engine_optimizer
66.72% :     0.042585s :      1: task_emit
14.58% :     0.009309s :      1: type_inference
 0.10% :     0.000066s :      1: validate


TotalTime = 0.0221447, [33]
[bootstrap]: 0.00034973
[type_inference]: 0.005219
[event_method]: 2.209e-05
[auto_monad]: 7.104e-05
[graph_reusing]: 6.24999e-06
[pre_auto_parallel]: 2.85002e-06
[py_interpret_to_execute]: 6.328e-05
[rewriter_before_opt_a]: 7.48e-05
[expand_dump_flag]: 3.21001e-06
[jit_opt_a]: 0.00669558, [2]
    [Cycle 1]: 0.00211379, [27]
        [switch_simplify]: 6.636e-05
        [loop_unroll]: 2.842e-05
        [a_1]: 0.00062208
        [with_stream_mark]: 2.647e-05
        [recompute_prepare]: 8.60999e-06
        [updatestate_depend_eliminate]: 3.93999e-06
        [updatestate_assign_eliminate]: 2.92002e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 2.00002e-06
        [specialize_transform]: 6.81001e-06
        [updatestate_useless_node_eliminater]: 5.17999e-06
        [accelerated_algorithm]: 7.33e-06
        [meta_shard_fg_expand]: 2.45002e-06
        [get_grad_eliminate_]: 5.64998e-06
        [merge_forward]: 4.02998e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.528e-05
        [j_node_and_user_rematch]: 1.035e-05
        [meta_fg_expand]: 3.03e-06
        [replace_old_param]: 9.64e-06
        [inline_without_move]: 5.47001e-06
        [renormalize]: 0.00100085
        [add_forward_monad_depend]: 7.08e-06
        [auto_monad_grad]: 2.85002e-06
        [auto_monad_eliminator]: 1.651e-05
        [cse]: 3.426e-05
        [replace_applicator]: 1.359e-05
    [Cycle 2]: 0.00034683, [27]
        [switch_simplify]: 5.61e-06
        [loop_unroll]: 5.32999e-06
        [a_1]: 9.246e-05
        [with_stream_mark]: 1.296e-05
        [recompute_prepare]: 5.27001e-06
        [updatestate_depend_eliminate]: 2.68e-06
        [updatestate_assign_eliminate]: 2.47001e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 9.90025e-07
        [specialize_transform]: 6.07001e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 1.44998e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 3.65e-06
        [cell_reuse_recompute_pass]: 3.25e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.366e-05
        [j_node_and_user_rematch]: 8.47e-06
        [meta_fg_expand]: 2.17999e-06
        [replace_old_param]: 7.5e-06
        [inline_without_move]: 4.90999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.58002e-06
        [auto_monad_grad]: 1.23002e-06
        [auto_monad_eliminator]: 5.49998e-06
        [cse]: 1.1e-05
        [replace_applicator]: 4.89e-06
[py_interpret_to_execute_after_opt_a]: 1.308e-05
[rewriter_after_opt_a]: 4.01e-05
[convert_after_rewriter]: 6.69001e-06
[order_py_execute_after_rewriter]: 4.78001e-06
[mutable_eliminate]: 0.00070996
[jit_opt_b]: 5.097e-05, [1]
    [Cycle 1]: 4.31e-05, [2]
        [frontend_op_eliminate]: 1.7e-05
        [inline_after_opt_a]: 1.45e-05
[cconv]: 2.784e-05
[loop_unroll]: 0.00045231
[jit_opt_after_cconv]: 0.00015877, [1]
    [Cycle 1]: 0.00015109, [11]
        [c_1]: 2.254e-05
        [parameter_eliminate]: 3.63e-06
        [updatestate_depend_eliminate]: 6.86999e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 2.284e-05
        [call_graph_tuple_transform]: 2.176e-05
        [tuple_list_get_item_eliminator]: 5.61998e-06
        [none_parameter_eliminate]: 1.84e-06
        [renormalize]: 7.09988e-07
        [switch_simplify]: 5.45001e-06
[remove_dup_value]: 1.679e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 6.39999e-06
[add_recomputation]: 5.856e-05
[cse_after_recomputation]: 2.546e-05, [1]
    [Cycle 1]: 1.811e-05, [1]
        [cse]: 9.42999e-06
[auto_monad_reorder]: 1.746e-05
[get_jit_bprop_graph]: 2.56e-06
[rewriter_after_jit_bprop_graph]: 5.77999e-06
[opt_after_jit_grad]: 0.00058622
[symbol_engine_optimizer]: 7.909e-05, [1]
    [Cycle 1]: 7.056e-05, [6]
        [build]: 4.77e-06
        [elim_shapecalc]: 7.35003e-06
        [elim_not_effective]: 1.513e-05
        [opt_reshape]: 5.92999e-06
        [fold_const_symbol]: 8.67e-06
        [renormalize]: 6.39993e-07
[validate]: 4.631e-05
[backend_pass]: 1.12999e-06
[task_emit]: 0.00707993
[execute]: 8.60001e-06
Sums
    bootstrap                            :     0.000350s :  2.03%
    type_inference                       :     0.005219s : 30.32%
    event_method                         :     0.000022s :  0.13%
    auto_monad                           :     0.000071s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000063s :  0.37%
    rewriter_before_opt_a                :     0.000075s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000072s :  0.42%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000715s :  4.15%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001001s :  5.81%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.13%
    jit_opt_a.cse                        :     0.000045s :  0.26%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000040s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000710s :  4.12%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.08%
    cconv                                :     0.000028s :  0.16%
    loop_unroll                          :     0.000452s :  2.63%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000017s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000059s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000586s :  3.41%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007080s : 41.13%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000267    23
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.17% :     0.000006s :      3: substitution.graph_param_transform
85.87% :     0.000229s :      4: substitution.inline
 1.56% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.94% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.77% :     0.000005s :      2: substitution.replace_old_param
 5.29% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005128     2
78.47% :     0.004024s :      1: type_inference.infer
21.53% :     0.001104s :      1: type_inference.specialize
------[replace.]   0.000080     6
65.38% :     0.000052s :      4: replace.inline
34.62% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000239     6
94.54% :     0.000226s :      4: match.inline
 5.46% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.29% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.49% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.65% :     0.000002s :     10: predicate.addn_zero_filter
 2.26% :     0.000003s :     10: predicate.arithmetic_simplify
 1.41% :     0.000002s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 0.97% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.70% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.35% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.81% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.61% :     0.000001s :      3: predicate.elim_not_effective
 0.42% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.45% :     0.000001s :      3: predicate.graph_param_transform
 5.09% :     0.000006s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.68% :     0.000002s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.70% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.52% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.80% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.49% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.00% :     0.000002s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     10: predicate.reduce_eliminate
 1.80% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.54% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.09% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.53% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.18% :     0.000009s :     53: predicate.switch_simplify
 1.60% :     0.000002s :     10: predicate.tile_eliminate
 1.02% :     0.000001s :     10: predicate.transpose_eliminate
 1.58% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.93% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.06% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.04% :     0.000003s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000717    12
42.09% :     0.000302s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.60% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.31% :     0.000361s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024191    76
 0.26% :     0.000063s :      1: add_recomputation
 0.31% :     0.000075s :      1: auto_monad
 0.09% :     0.000021s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.52% :     0.000368s :      1: bootstrap
 0.12% :     0.000030s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000028s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.11% :     0.000026s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.69% :     0.006700s :      1: jit_opt_a
 0.67% :     0.000162s :      1: jit_opt_after_cconv
 0.22% :     0.000054s :      1: jit_opt_b
 1.91% :     0.000462s :      1: loop_unroll
 2.97% :     0.000717s :      1: mutable_eliminate
 3.90% :     0.000942s :     26: opt.transform.jit_opt_a
 0.21% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.47% :     0.000597s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000068s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000019s :      1: remove_dup_value
 2.28% :     0.000552s :      1: renormalize.infer
 1.82% :     0.000439s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000043s :      1: rewriter_after_opt_a
 0.32% :     0.000078s :      1: rewriter_before_opt_a
 0.34% :     0.000082s :      1: symbol_engine_optimizer
29.33% :     0.007096s :      1: task_emit
21.65% :     0.005239s :      1: type_inference
 0.30% :     0.000072s :      1: validate


TotalTime = 0.0173709, [30]
[bootstrap]: 0.00031879
[type_inference]: 0.00777048
[event_method]: 2.516e-05
[auto_monad]: 8.956e-05
[graph_reusing]: 7.2e-06
[pre_auto_parallel]: 2.69999e-06
[py_interpret_to_execute]: 5.34e-05
[rewriter_before_opt_a]: 8.996e-05
[expand_dump_flag]: 3.49001e-06
[jit_opt_a]: 0.00645967, [2]
    [Cycle 1]: 0.00237, [27]
        [switch_simplify]: 0.00015196
        [loop_unroll]: 3.666e-05
        [a_1]: 0.00079011
        [with_stream_mark]: 3.056e-05
        [recompute_prepare]: 1.092e-05
        [updatestate_depend_eliminate]: 4.4e-06
        [updatestate_assign_eliminate]: 3.4e-06
        [updatestate_loads_eliminate]: 2.91e-06
        [parameter_eliminate]: 2.28002e-06
        [specialize_transform]: 7.25003e-06
        [updatestate_useless_node_eliminater]: 5.57999e-06
        [accelerated_algorithm]: 6.07001e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 5.71e-06
        [merge_forward]: 4.46002e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.85e-05
        [j_node_and_user_rematch]: 1.015e-05
        [meta_fg_expand]: 2.75002e-06
        [replace_old_param]: 1.043e-05
        [inline_without_move]: 5.46e-06
        [renormalize]: 0.0009468
        [add_forward_monad_depend]: 7.9e-06
        [auto_monad_grad]: 2.71999e-06
        [auto_monad_eliminator]: 1.828e-05
        [cse]: 3.566e-05
        [replace_applicator]: 1.844e-05
    [Cycle 2]: 0.00034926, [27]
        [switch_simplify]: 6.20002e-06
        [loop_unroll]: 5.54e-06
        [a_1]: 9.063e-05
        [with_stream_mark]: 1.444e-05
        [recompute_prepare]: 5.52001e-06
        [updatestate_depend_eliminate]: 3.26001e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [parameter_eliminate]: 1.97999e-06
        [specialize_transform]: 5.39998e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 1.74998e-06
        [get_grad_eliminate_]: 4.48999e-06
        [merge_forward]: 3.54002e-06
        [cell_reuse_recompute_pass]: 2.21998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.658e-05
        [j_node_and_user_rematch]: 8.35001e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 8.52e-06
        [inline_without_move]: 4.73001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 1.27999e-06
        [auto_monad_eliminator]: 6.31998e-06
        [cse]: 1.172e-05
        [replace_applicator]: 6.06e-06
[py_interpret_to_execute_after_opt_a]: 1.476e-05
[rewriter_after_opt_a]: 4.112e-05
[convert_after_rewriter]: 6.83998e-06
[order_py_execute_after_rewriter]: 5.17999e-06
[mutable_eliminate]: 0.00076922
[jit_opt_b]: 5.487e-05, [1]
    [Cycle 1]: 4.553e-05, [2]
        [frontend_op_eliminate]: 1.76e-05
        [inline_after_opt_a]: 1.598e-05
[cconv]: 3.371e-05
[loop_unroll]: 0.00047794
[jit_opt_after_cconv]: 0.00016191, [1]
    [Cycle 1]: 0.00015488, [11]
        [c_1]: 2.138e-05
        [parameter_eliminate]: 4.84998e-06
        [updatestate_depend_eliminate]: 9.69999e-06
        [updatestate_assign_eliminate]: 3.11001e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [cse]: 2.624e-05
        [call_graph_tuple_transform]: 2.285e-05
        [tuple_list_get_item_eliminator]: 5.77001e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 5.39e-06
[remove_dup_value]: 1.479e-05
[partial_unused_args_eliminate]: 2.89001e-06
[environ_conv]: 5.80002e-06
[add_recomputation]: 5.335e-05
[cse_after_recomputation]: 2.197e-05, [1]
    [Cycle 1]: 1.605e-05, [1]
        [cse]: 9.44e-06
[auto_monad_reorder]: 1.525e-05
[get_jit_bprop_graph]: 2.74999e-06
[rewriter_after_jit_bprop_graph]: 8.99998e-06
[opt_after_jit_grad]: 0.0005201
[symbol_engine_optimizer]: 8.013e-05, [1]
    [Cycle 1]: 7.216e-05, [6]
        [build]: 4.62e-06
        [elim_shapecalc]: 9.08002e-06
        [elim_not_effective]: 1.401e-05
        [opt_reshape]: 6.01e-06
        [fold_const_symbol]: 8.84e-06
        [renormalize]: 6.09987e-07
[validate]: 4.328e-05
Sums
    bootstrap                            :     0.000319s :  2.46%
    type_inference                       :     0.007770s : 60.07%
    event_method                         :     0.000025s :  0.19%
    auto_monad                           :     0.000090s :  0.69%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000053s :  0.41%
    rewriter_before_opt_a                :     0.000090s :  0.70%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000158s :  1.22%
    jit_opt_a.loop_unroll                :     0.000042s :  0.33%
    jit_opt_a.a_1                        :     0.000881s :  6.81%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000008s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000947s :  7.32%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.19%
    jit_opt_a.cse                        :     0.000047s :  0.37%
    jit_opt_a.replace_applicator         :     0.000025s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.11%
    rewriter_after_opt_a                 :     0.000041s :  0.32%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000769s :  5.95%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.12%
    cconv                                :     0.000034s :  0.26%
    loop_unroll                          :     0.000478s :  3.69%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.17%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000026s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000015s :  0.11%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000053s :  0.41%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.12%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.07%
    opt_after_jit_grad                   :     0.000520s :  4.02%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.33%


Time group info:
------[substitution.]   0.000359    26
 0.52% :     0.000002s :      2: substitution.elim_not_effective
 0.44% :     0.000002s :      2: substitution.fold_const_symbol
 1.56% :     0.000006s :      3: substitution.graph_param_transform
86.36% :     0.000310s :      6: substitution.inline
 1.06% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.55% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.73% :     0.000006s :      2: substitution.replace_old_param
 2.56% :     0.000009s :      1: substitution.switch_simplify
 4.23% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007686     2
84.80% :     0.006518s :      1: type_inference.infer
15.20% :     0.001168s :      1: type_inference.specialize
------[replace.]   0.000164     9
49.04% :     0.000080s :      6: replace.inline
30.48% :     0.000050s :      1: replace.switch_simplify
20.48% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000329     9
93.24% :     0.000307s :      6: match.inline
 2.57% :     0.000008s :      1: match.switch_simplify
 4.18% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000142   801
 1.28% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.11% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     12: predicate.addn_check_dump
 1.44% :     0.000002s :     12: predicate.addn_zero_filter
 1.87% :     0.000003s :     12: predicate.arithmetic_simplify
 1.26% :     0.000002s :     12: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000002s :     12: predicate.compare_switch_simplify
 1.69% :     0.000002s :     12: predicate.depend_value_elim
 1.04% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.15% :     0.000002s :     12: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      3: predicate.elim_not_effective
 0.92% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.08% :     0.000002s :     12: predicate.environ_get_eliminate
 1.07% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.93% :     0.000008s :     26: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.68% :     0.000002s :      6: predicate.less_batch_normalization
 1.33% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     17: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.29% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.65% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000002s :     12: predicate.merge_addn
 1.04% :     0.000001s :     12: predicate.minmaximum_grad
 2.36% :     0.000003s :      3: predicate.mutable_eliminate
 0.38% :     0.000001s :      3: predicate.opt_reshape
 2.03% :     0.000003s :     17: predicate.partial_eliminate
 1.11% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.39% :     0.000002s :     12: predicate.reduce_eliminate
 1.49% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.82% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     20: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.58% :     0.000001s :      3: predicate.reset_defer_inline
 1.77% :     0.000003s :     12: predicate.reshape_eliminate
 1.29% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000002s :     12: predicate.same_eliminate
 0.42% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.79% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.12% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.85% :     0.000004s :     20: predicate.switch_defer_inline
 2.14% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.94% :     0.000013s :     65: predicate.switch_simplify
 1.35% :     0.000002s :     12: predicate.tile_eliminate
 1.13% :     0.000002s :     12: predicate.transpose_eliminate
 1.31% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.01% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.51% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.41% :     0.000002s :     12: predicate.value_based_eliminate
 0.27% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000948    15
48.14% :     0.000456s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.77% :     0.000130s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.09% :     0.000361s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019588    72
 0.29% :     0.000057s :      1: add_recomputation
 0.48% :     0.000093s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 1.75% :     0.000342s :      1: bootstrap
 0.19% :     0.000037s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.15% :     0.000030s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000010s :      1: graph_reusing
33.00% :     0.006463s :      1: jit_opt_a
 0.84% :     0.000165s :      1: jit_opt_after_cconv
 0.29% :     0.000057s :      1: jit_opt_b
 2.49% :     0.000488s :      1: loop_unroll
 4.01% :     0.000785s :      1: mutable_eliminate
 6.20% :     0.001215s :     26: opt.transform.jit_opt_a
 0.26% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000027s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.72% :     0.000533s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000056s :      1: py_interpret_to_execute
 0.09% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000017s :      1: remove_dup_value
 2.81% :     0.000550s :      1: renormalize.infer
 1.97% :     0.000385s :      1: renormalize.specialize
 0.06% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000045s :      1: rewriter_after_opt_a
 0.48% :     0.000093s :      1: rewriter_before_opt_a
 0.42% :     0.000083s :      1: symbol_engine_optimizer
39.80% :     0.007796s :      1: type_inference


TotalTime = 0.0146423, [30]
[bootstrap]: 0.00029919
[type_inference]: 0.00659559
[event_method]: 2.278e-05
[auto_monad]: 8.096e-05
[graph_reusing]: 6.66999e-06
[pre_auto_parallel]: 2.71e-06
[py_interpret_to_execute]: 4.845e-05
[rewriter_before_opt_a]: 8.681e-05
[expand_dump_flag]: 3.58999e-06
[jit_opt_a]: 0.00543188, [2]
    [Cycle 1]: 0.00185549, [27]
        [switch_simplify]: 0.00012344
        [loop_unroll]: 3.538e-05
        [a_1]: 0.00068386
        [with_stream_mark]: 1.989e-05
        [recompute_prepare]: 7.41001e-06
        [updatestate_depend_eliminate]: 3.64002e-06
        [updatestate_assign_eliminate]: 3.47002e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 2.24001e-06
        [specialize_transform]: 5.92999e-06
        [updatestate_useless_node_eliminater]: 5.33002e-06
        [accelerated_algorithm]: 5.65001e-06
        [meta_shard_fg_expand]: 2.33998e-06
        [get_grad_eliminate_]: 5.27001e-06
        [merge_forward]: 3.86999e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.434e-05
        [j_node_and_user_rematch]: 9.10001e-06
        [meta_fg_expand]: 2.45002e-06
        [replace_old_param]: 9.08002e-06
        [inline_without_move]: 5.52999e-06
        [renormalize]: 0.00066659
        [add_forward_monad_depend]: 5.01002e-06
        [auto_monad_grad]: 2.09e-06
        [auto_monad_eliminator]: 1.376e-05
        [cse]: 2.974e-05
        [replace_applicator]: 1.197e-05
    [Cycle 2]: 0.00031254, [27]
        [switch_simplify]: 5.61e-06
        [loop_unroll]: 4.72e-06
        [a_1]: 8.839e-05
        [with_stream_mark]: 1.054e-05
        [recompute_prepare]: 5.02e-06
        [updatestate_depend_eliminate]: 2.59001e-06
        [updatestate_assign_eliminate]: 2.18002e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 4.82998e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 4.85999e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.4e-06
        [merge_forward]: 3.01001e-06
        [cell_reuse_recompute_pass]: 1.30001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.325e-05
        [j_node_and_user_rematch]: 8.01001e-06
        [meta_fg_expand]: 1.45999e-06
        [replace_old_param]: 6.90998e-06
        [inline_without_move]: 4.93001e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 5.05001e-06
        [cse]: 9.34e-06
        [replace_applicator]: 5.07e-06
[py_interpret_to_execute_after_opt_a]: 9.15999e-06
[rewriter_after_opt_a]: 3.392e-05
[convert_after_rewriter]: 7.52998e-06
[order_py_execute_after_rewriter]: 4.79e-06
[mutable_eliminate]: 0.00051216
[jit_opt_b]: 5.855e-05, [1]
    [Cycle 1]: 5.121e-05, [2]
        [frontend_op_eliminate]: 1.623e-05
        [inline_after_opt_a]: 1.586e-05
[cconv]: 2.479e-05
[loop_unroll]: 0.00043221
[jit_opt_after_cconv]: 0.00013923, [1]
    [Cycle 1]: 0.00013259, [11]
        [c_1]: 2.105e-05
        [parameter_eliminate]: 2.44001e-06
        [updatestate_depend_eliminate]: 5.59998e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.891e-05
        [call_graph_tuple_transform]: 1.803e-05
        [tuple_list_get_item_eliminator]: 5.29e-06
        [none_parameter_eliminate]: 1.42999e-06
        [renormalize]: 6.10016e-07
        [switch_simplify]: 5.19998e-06
[remove_dup_value]: 1.29e-05
[partial_unused_args_eliminate]: 2.12001e-06
[environ_conv]: 4.90999e-06
[add_recomputation]: 4.905e-05
[cse_after_recomputation]: 2.15e-05, [1]
    [Cycle 1]: 1.582e-05, [1]
        [cse]: 9.49e-06
[auto_monad_reorder]: 1.421e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 4.93001e-06
[opt_after_jit_grad]: 0.00044251
[symbol_engine_optimizer]: 7.206e-05, [1]
    [Cycle 1]: 6.596e-05, [6]
        [build]: 3.73999e-06
        [elim_shapecalc]: 7.66999e-06
        [elim_not_effective]: 1.266e-05
        [opt_reshape]: 5.63002e-06
        [fold_const_symbol]: 8.03001e-06
        [renormalize]: 6.10016e-07
[validate]: 3.453e-05
Sums
    bootstrap                            :     0.000299s :  2.77%
    type_inference                       :     0.006596s : 61.16%
    event_method                         :     0.000023s :  0.21%
    auto_monad                           :     0.000081s :  0.75%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000048s :  0.45%
    rewriter_before_opt_a                :     0.000087s :  0.80%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000129s :  1.20%
    jit_opt_a.loop_unroll                :     0.000040s :  0.37%
    jit_opt_a.a_1                        :     0.000772s :  7.16%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000667s :  6.18%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.17%
    jit_opt_a.cse                        :     0.000039s :  0.36%
    jit_opt_a.replace_applicator         :     0.000017s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.08%
    rewriter_after_opt_a                 :     0.000034s :  0.31%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000512s :  4.75%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.15%
    cconv                                :     0.000025s :  0.23%
    loop_unroll                          :     0.000432s :  4.01%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000049s :  0.45%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000014s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000443s :  4.10%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000035s :  0.32%


Time group info:
------[substitution.]   0.000278    26
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 1.95% :     0.000005s :      3: substitution.graph_param_transform
85.71% :     0.000238s :      6: substitution.inline
 1.09% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.65% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.21% :     0.000003s :      2: substitution.replace_old_param
 2.79% :     0.000008s :      1: substitution.switch_simplify
 4.35% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006525     2
84.29% :     0.005500s :      1: type_inference.infer
15.71% :     0.001025s :      1: type_inference.specialize
------[replace.]   0.000135     9
48.99% :     0.000066s :      6: replace.inline
28.88% :     0.000039s :      1: replace.switch_simplify
22.14% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000253     9
92.92% :     0.000235s :      6: match.inline
 2.76% :     0.000007s :      1: match.switch_simplify
 4.32% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   801
 1.34% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.10% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     12: predicate.addn_check_dump
 1.37% :     0.000002s :     12: predicate.addn_zero_filter
 1.96% :     0.000003s :     12: predicate.arithmetic_simplify
 1.27% :     0.000002s :     12: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     12: predicate.compare_switch_simplify
 1.33% :     0.000002s :     12: predicate.depend_value_elim
 1.23% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     12: predicate.environ_get_eliminate
 1.12% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.81% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.90% :     0.000008s :     26: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     17: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.45% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.99% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     12: predicate.merge_addn
 1.16% :     0.000002s :     12: predicate.minmaximum_grad
 1.84% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     17: predicate.partial_eliminate
 1.22% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     12: predicate.reduce_eliminate
 1.46% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     20: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     12: predicate.reshape_eliminate
 1.27% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     12: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.71% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.72% :     0.000004s :     20: predicate.switch_defer_inline
 2.43% :     0.000003s :     20: predicate.switch_layer_defer_inline
 9.08% :     0.000012s :     65: predicate.switch_simplify
 1.22% :     0.000002s :     12: predicate.tile_eliminate
 1.13% :     0.000001s :     12: predicate.transpose_eliminate
 1.34% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.48% :     0.000002s :     12: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000790    15
50.58% :     0.000400s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.08% :     0.000095s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.33% :     0.000295s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016417    72
 0.32% :     0.000052s :      1: add_recomputation
 0.52% :     0.000085s :      1: auto_monad
 0.10% :     0.000017s :      1: auto_monad_reorder
 1.95% :     0.000320s :      1: bootstrap
 0.17% :     0.000028s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.17% :     0.000028s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
33.11% :     0.005435s :      1: jit_opt_a
 0.86% :     0.000142s :      1: jit_opt_after_cconv
 0.37% :     0.000061s :      1: jit_opt_b
 2.69% :     0.000441s :      1: loop_unroll
 3.17% :     0.000521s :      1: mutable_eliminate
 6.42% :     0.001055s :     26: opt.transform.jit_opt_a
 0.28% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.74% :     0.000450s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000051s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000015s :      1: remove_dup_value
 2.32% :     0.000380s :      1: renormalize.infer
 1.70% :     0.000279s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000037s :      1: rewriter_after_opt_a
 0.55% :     0.000090s :      1: rewriter_before_opt_a
 0.45% :     0.000074s :      1: symbol_engine_optimizer
40.29% :     0.006615s :      1: type_inference


TotalTime = 0.0825339, [33]
[bootstrap]: 0.00037644
[type_inference]: 0.00746163
[event_method]: 1.813e-05
[auto_monad]: 6.933e-05
[graph_reusing]: 6.36e-06
[pre_auto_parallel]: 2.58e-06
[py_interpret_to_execute]: 3.787e-05
[rewriter_before_opt_a]: 7.498e-05
[expand_dump_flag]: 3.48e-06
[jit_opt_a]: 0.0515872, [2]
    [Cycle 1]: 0.00216215, [27]
        [switch_simplify]: 7.338e-05
        [loop_unroll]: 3.014e-05
        [a_1]: 0.00064286
        [with_stream_mark]: 2.472e-05
        [recompute_prepare]: 1.18e-05
        [updatestate_depend_eliminate]: 3.97e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.46998e-06
        [parameter_eliminate]: 2.56998e-06
        [specialize_transform]: 6.78e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 6.85998e-06
        [meta_shard_fg_expand]: 2.69999e-06
        [get_grad_eliminate_]: 6.19999e-06
        [merge_forward]: 4.02e-06
        [cell_reuse_recompute_pass]: 1.41002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.658e-05
        [j_node_and_user_rematch]: 1.104e-05
        [meta_fg_expand]: 2.74001e-06
        [replace_old_param]: 1.053e-05
        [inline_without_move]: 5.84e-06
        [renormalize]: 0.00098172
        [add_forward_monad_depend]: 9.69999e-06
        [auto_monad_grad]: 2.69001e-06
        [auto_monad_eliminator]: 2.003e-05
        [cse]: 3.406e-05
        [replace_applicator]: 2.166e-05
    [Cycle 2]: 0.00035676, [27]
        [switch_simplify]: 6.24001e-06
        [loop_unroll]: 5.20001e-06
        [a_1]: 0.00010094
        [with_stream_mark]: 1.376e-05
        [recompute_prepare]: 4.92e-06
        [updatestate_depend_eliminate]: 3.31001e-06
        [updatestate_assign_eliminate]: 2.59999e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 5.25001e-06
        [updatestate_useless_node_eliminater]: 4.67998e-06
        [accelerated_algorithm]: 5.73002e-06
        [meta_shard_fg_expand]: 1.63002e-06
        [get_grad_eliminate_]: 4.89998e-06
        [merge_forward]: 3.25e-06
        [cell_reuse_recompute_pass]: 2.48002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.714e-05
        [j_node_and_user_rematch]: 9.04e-06
        [meta_fg_expand]: 1.88997e-06
        [replace_old_param]: 8.72e-06
        [inline_without_move]: 4.65001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.45999e-06
        [auto_monad_grad]: 1.07e-06
        [auto_monad_eliminator]: 5.54e-06
        [cse]: 1.111e-05
        [replace_applicator]: 4.93001e-06
[py_interpret_to_execute_after_opt_a]: 1.56e-05
[rewriter_after_opt_a]: 4.186e-05
[convert_after_rewriter]: 6.68e-06
[order_py_execute_after_rewriter]: 5.19e-06
[mutable_eliminate]: 0.00083676
[jit_opt_b]: 5.561e-05, [1]
    [Cycle 1]: 4.688e-05, [2]
        [frontend_op_eliminate]: 1.966e-05
        [inline_after_opt_a]: 1.55e-05
[cconv]: 4.223e-05
[loop_unroll]: 0.00047121
[jit_opt_after_cconv]: 0.00015453, [1]
    [Cycle 1]: 0.00014745, [11]
        [c_1]: 2.145e-05
        [parameter_eliminate]: 3.76001e-06
        [updatestate_depend_eliminate]: 6.54999e-06
        [updatestate_assign_eliminate]: 2.77002e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [cse]: 2.627e-05
        [call_graph_tuple_transform]: 1.986e-05
        [tuple_list_get_item_eliminator]: 5.32001e-06
        [none_parameter_eliminate]: 1.67999e-06
        [renormalize]: 7.00005e-07
        [switch_simplify]: 5.50001e-06
[remove_dup_value]: 1.491e-05
[partial_unused_args_eliminate]: 2.76999e-06
[environ_conv]: 5.62999e-06
[add_recomputation]: 5.374e-05
[cse_after_recomputation]: 2.176e-05, [1]
    [Cycle 1]: 1.621e-05, [1]
        [cse]: 9.90002e-06
[auto_monad_reorder]: 1.614e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 7.93999e-06
[opt_after_jit_grad]: 0.00048622
[symbol_engine_optimizer]: 7.648e-05, [1]
    [Cycle 1]: 6.923e-05, [6]
        [build]: 4.07e-06
        [elim_shapecalc]: 7.77e-06
        [elim_not_effective]: 1.343e-05
        [opt_reshape]: 6.07999e-06
        [fold_const_symbol]: 9.30001e-06
        [renormalize]: 6.89994e-07
[validate]: 4.089e-05
[backend_pass]: 1.02e-06
[task_emit]: 0.0202905
[execute]: 1.273e-05
Sums
    bootstrap                            :     0.000376s :  1.15%
    type_inference                       :     0.007462s : 22.77%
    event_method                         :     0.000018s :  0.06%
    auto_monad                           :     0.000069s :  0.21%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000038s :  0.12%
    rewriter_before_opt_a                :     0.000075s :  0.23%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000080s :  0.24%
    jit_opt_a.loop_unroll                :     0.000035s :  0.11%
    jit_opt_a.a_1                        :     0.000744s :  2.27%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000019s :  0.06%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.000982s :  3.00%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.08%
    jit_opt_a.cse                        :     0.000045s :  0.14%
    jit_opt_a.replace_applicator         :     0.000027s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.05%
    rewriter_after_opt_a                 :     0.000042s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000837s :  2.55%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.05%
    cconv                                :     0.000042s :  0.13%
    loop_unroll                          :     0.000471s :  1.44%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000015s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000054s :  0.16%
    cse_after_recomputation.cse          :     0.000010s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000486s :  1.48%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.020291s : 61.93%
    execute                              :     0.000013s :  0.04%


Time group info:
------[substitution.]   0.000265    23
 1.06% :     0.000003s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 2.41% :     0.000006s :      3: substitution.graph_param_transform
84.00% :     0.000222s :      4: substitution.inline
 1.74% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.33% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.53% :     0.000007s :      2: substitution.replace_old_param
 5.45% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007385     2
86.65% :     0.006399s :      1: type_inference.infer
13.35% :     0.000986s :      1: type_inference.specialize
------[replace.]   0.000090     6
64.02% :     0.000058s :      4: replace.inline
35.98% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000233     6
94.28% :     0.000220s :      4: match.inline
 5.72% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   691
 1.81% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.91% :     0.000001s :     10: predicate.addn_check_dump
 1.12% :     0.000002s :     10: predicate.addn_zero_filter
 2.56% :     0.000003s :     10: predicate.arithmetic_simplify
 1.01% :     0.000001s :     10: predicate.cast_eliminate
 0.35% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.17% :     0.000002s :     10: predicate.compare_switch_simplify
 1.98% :     0.000003s :     10: predicate.depend_value_elim
 1.49% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.67% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.01% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.98% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 1.31% :     0.000002s :      6: predicate.get_grad_eliminate
 0.36% :     0.000000s :      3: predicate.graph_param_transform
 4.90% :     0.000007s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000002s :      6: predicate.less_batch_normalization
 1.84% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.58% :     0.000002s :     15: predicate.load_eliminater
 1.14% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.46% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.29% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.94% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000002s :     10: predicate.minmaximum_grad
 2.29% :     0.000003s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.01% :     0.000003s :     15: predicate.partial_eliminate
 0.97% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.18% :     0.000003s :     10: predicate.reduce_eliminate
 1.35% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.15% :     0.000003s :     18: predicate.replace_applicator
 0.43% :     0.000001s :      6: predicate.replace_old_param
 0.21% :     0.000000s :      3: predicate.reset_defer_inline
 1.86% :     0.000003s :     10: predicate.reshape_eliminate
 0.95% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000002s :     10: predicate.same_eliminate
 0.69% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      6: predicate.special_op_eliminate
 0.84% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.88% :     0.000003s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.77% :     0.000004s :     16: predicate.switch_defer_inline
 1.68% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.24% :     0.000010s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.98% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.97% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.17% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.39% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.48% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000734    12
48.21% :     0.000354s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.87% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.92% :     0.000344s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.084609    76
 0.07% :     0.000056s :      1: add_recomputation
 0.09% :     0.000073s :      1: auto_monad
 0.02% :     0.000019s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.47% :     0.000395s :      1: bootstrap
 0.05% :     0.000045s :      1: cconv
 0.01% :     0.000009s :      1: convert_after_rewriter
 0.03% :     0.000024s :      1: cse_after_recomputation
 0.01% :     0.000008s :      1: environ_conv
 0.03% :     0.000022s :      1: event_method
 0.02% :     0.000020s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000009s :      1: graph_reusing
60.98% :     0.051592s :      1: jit_opt_a
 0.19% :     0.000157s :      1: jit_opt_after_cconv
 0.07% :     0.000059s :      1: jit_opt_b
 0.57% :     0.000479s :      1: loop_unroll
 1.00% :     0.000847s :      1: mutable_eliminate
 1.18% :     0.000995s :     26: opt.transform.jit_opt_a
 0.06% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.03% :     0.000028s :      4: opt.transform.jit_opt_b
 0.01% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.04% :     0.000034s :      4: opt.transform.symbol_engine_opt
 0.58% :     0.000494s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.05% :     0.000041s :      1: py_interpret_to_execute
 0.02% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000017s :      1: remove_dup_value
 0.61% :     0.000514s :      1: renormalize.infer
 0.54% :     0.000456s :      1: renormalize.specialize
 0.01% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.05% :     0.000045s :      1: rewriter_after_opt_a
 0.09% :     0.000078s :      1: rewriter_before_opt_a
 0.09% :     0.000079s :      1: symbol_engine_optimizer
24.02% :     0.020319s :      1: task_emit
 8.83% :     0.007474s :      1: type_inference
 0.07% :     0.000061s :      1: validate


TotalTime = 0.0241704, [33]
[bootstrap]: 0.00038563
[type_inference]: 0.00570705
[event_method]: 2.161e-05
[auto_monad]: 7.564e-05
[graph_reusing]: 7.11999e-06
[pre_auto_parallel]: 3.09001e-06
[py_interpret_to_execute]: 5.522e-05
[rewriter_before_opt_a]: 8.465e-05
[expand_dump_flag]: 4.07998e-06
[jit_opt_a]: 0.00746654, [2]
    [Cycle 1]: 0.00232128, [27]
        [switch_simplify]: 7.539e-05
        [loop_unroll]: 3.079e-05
        [a_1]: 0.0007352
        [with_stream_mark]: 2.839e-05
        [recompute_prepare]: 8.94998e-06
        [updatestate_depend_eliminate]: 4.12e-06
        [updatestate_assign_eliminate]: 3.46001e-06
        [updatestate_loads_eliminate]: 2.41998e-06
        [parameter_eliminate]: 2.51e-06
        [specialize_transform]: 6.30002e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 6.00002e-06
        [meta_shard_fg_expand]: 2.31998e-06
        [get_grad_eliminate_]: 5.59e-06
        [merge_forward]: 4.06001e-06
        [cell_reuse_recompute_pass]: 1.66002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.67e-05
        [j_node_and_user_rematch]: 1.063e-05
        [meta_fg_expand]: 3.48e-06
        [replace_old_param]: 1.106e-05
        [inline_without_move]: 5.86e-06
        [renormalize]: 0.00105559
        [add_forward_monad_depend]: 8.61002e-06
        [auto_monad_grad]: 3.08e-06
        [auto_monad_eliminator]: 1.586e-05
        [cse]: 3.746e-05
        [replace_applicator]: 1.752e-05
    [Cycle 2]: 0.0003488, [27]
        [switch_simplify]: 7.08e-06
        [loop_unroll]: 5.61e-06
        [a_1]: 9.635e-05
        [with_stream_mark]: 1.279e-05
        [recompute_prepare]: 5.53002e-06
        [updatestate_depend_eliminate]: 2.97002e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [parameter_eliminate]: 1.17999e-06
        [specialize_transform]: 5.10999e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 1.55999e-06
        [get_grad_eliminate_]: 4.97e-06
        [merge_forward]: 2.73e-06
        [cell_reuse_recompute_pass]: 2.79001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.519e-05
        [j_node_and_user_rematch]: 9.14998e-06
        [meta_fg_expand]: 2.44999e-06
        [replace_old_param]: 7.25998e-06
        [inline_without_move]: 8.85999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 1.08001e-06
        [auto_monad_eliminator]: 4.75999e-06
        [cse]: 1.11e-05
        [replace_applicator]: 4.77e-06
[py_interpret_to_execute_after_opt_a]: 1.62e-05
[rewriter_after_opt_a]: 4.049e-05
[convert_after_rewriter]: 7.63001e-06
[order_py_execute_after_rewriter]: 5.57999e-06
[mutable_eliminate]: 0.00090035
[jit_opt_b]: 5.759e-05, [1]
    [Cycle 1]: 4.672e-05, [2]
        [frontend_op_eliminate]: 1.906e-05
        [inline_after_opt_a]: 1.583e-05
[cconv]: 3.373e-05
[loop_unroll]: 0.00049531
[jit_opt_after_cconv]: 0.00018126, [1]
    [Cycle 1]: 0.000174, [11]
        [c_1]: 2.33e-05
        [parameter_eliminate]: 4.49998e-06
        [updatestate_depend_eliminate]: 9.25999e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [cse]: 2.606e-05
        [call_graph_tuple_transform]: 2.3e-05
        [tuple_list_get_item_eliminator]: 5.97999e-06
        [none_parameter_eliminate]: 1.86998e-06
        [renormalize]: 1.02e-06
        [switch_simplify]: 6.31998e-06
[remove_dup_value]: 1.622e-05
[partial_unused_args_eliminate]: 2.73e-06
[environ_conv]: 6.54999e-06
[add_recomputation]: 5.508e-05
[cse_after_recomputation]: 2.321e-05, [1]
    [Cycle 1]: 1.731e-05, [1]
        [cse]: 1.11e-05
[auto_monad_reorder]: 1.606e-05
[get_jit_bprop_graph]: 2.29001e-06
[rewriter_after_jit_bprop_graph]: 9.44e-06
[opt_after_jit_grad]: 0.00051047
[symbol_engine_optimizer]: 8.212e-05, [1]
    [Cycle 1]: 7.41e-05, [6]
        [build]: 5.30001e-06
        [elim_shapecalc]: 8.69e-06
        [elim_not_effective]: 1.417e-05
        [opt_reshape]: 6.68e-06
        [fold_const_symbol]: 9.81e-06
        [renormalize]: 1.02998e-06
[validate]: 4.95e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.00757985
[execute]: 9.48002e-06
Sums
    bootstrap                            :     0.000386s :  2.07%
    type_inference                       :     0.005707s : 30.62%
    event_method                         :     0.000022s :  0.12%
    auto_monad                           :     0.000076s :  0.41%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000055s :  0.30%
    rewriter_before_opt_a                :     0.000085s :  0.45%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000082s :  0.44%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000832s :  4.46%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000015s :  0.08%
    jit_opt_a.renormalize                :     0.001056s :  5.66%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.11%
    jit_opt_a.cse                        :     0.000049s :  0.26%
    jit_opt_a.replace_applicator         :     0.000022s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.09%
    rewriter_after_opt_a                 :     0.000040s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000900s :  4.83%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.000034s :  0.18%
    loop_unroll                          :     0.000495s :  2.66%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000055s :  0.30%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000510s :  2.74%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000049s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007580s : 40.67%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000337    23
 0.56% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000002s :      2: substitution.fold_const_symbol
 1.86% :     0.000006s :      3: substitution.graph_param_transform
87.44% :     0.000294s :      4: substitution.inline
 1.41% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.53% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.67% :     0.000006s :      2: substitution.replace_old_param
 4.90% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005614     2
80.97% :     0.004546s :      1: type_inference.infer
19.03% :     0.001068s :      1: type_inference.specialize
------[replace.]   0.000098     6
68.13% :     0.000066s :      4: replace.inline
31.87% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000306     6
94.96% :     0.000290s :      4: match.inline
 5.04% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   691
 1.54% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.89% :     0.000001s :     10: predicate.addn_check_dump
 2.05% :     0.000003s :     10: predicate.addn_zero_filter
 3.22% :     0.000004s :     10: predicate.arithmetic_simplify
 1.16% :     0.000002s :     10: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.87% :     0.000001s :     10: predicate.compare_switch_simplify
 1.08% :     0.000001s :     10: predicate.depend_value_elim
 0.95% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.46% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000002s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.17% :     0.000002s :      6: predicate.get_grad_eliminate
 0.39% :     0.000001s :      3: predicate.graph_param_transform
 5.51% :     0.000008s :     22: predicate.inline
 0.74% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.49% :     0.000002s :      6: predicate.less_batch_normalization
 1.66% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.45% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.80% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.81% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.87% :     0.000001s :     10: predicate.merge_addn
 2.06% :     0.000003s :     10: predicate.minmaximum_grad
 1.84% :     0.000003s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.41% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.31% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.53% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.43% :     0.000001s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 0.95% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     10: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 0.82% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 2.27% :     0.000003s :     16: predicate.switch_layer_defer_inline
 9.23% :     0.000013s :     53: predicate.switch_simplify
 1.23% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000002s :     10: predicate.transpose_eliminate
 1.23% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.93% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.08% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.35% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.46% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000789    12
47.86% :     0.000378s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.89% :     0.000062s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.24% :     0.000349s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026417    76
 0.22% :     0.000058s :      1: add_recomputation
 0.30% :     0.000080s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.55% :     0.000408s :      1: bootstrap
 0.14% :     0.000037s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.10% :     0.000027s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
28.28% :     0.007471s :      1: jit_opt_a
 0.70% :     0.000185s :      1: jit_opt_after_cconv
 0.23% :     0.000061s :      1: jit_opt_b
 1.91% :     0.000505s :      1: loop_unroll
 3.45% :     0.000912s :      1: mutable_eliminate
 4.10% :     0.001084s :     26: opt.transform.jit_opt_a
 0.21% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.97% :     0.000520s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000058s :      1: py_interpret_to_execute
 0.07% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 2.20% :     0.000581s :      1: renormalize.infer
 1.76% :     0.000464s :      1: renormalize.specialize
 0.04% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000044s :      1: rewriter_after_opt_a
 0.34% :     0.000090s :      1: rewriter_before_opt_a
 0.32% :     0.000085s :      1: symbol_engine_optimizer
28.77% :     0.007600s :      1: task_emit
21.69% :     0.005729s :      1: type_inference
 0.28% :     0.000075s :      1: validate


TotalTime = 0.0359589, [33]
[bootstrap]: 0.00045304
[type_inference]: 0.00607992
[event_method]: 1.771e-05
[auto_monad]: 6.755e-05
[graph_reusing]: 5.69e-06
[pre_auto_parallel]: 2.27001e-06
[py_interpret_to_execute]: 4.032e-05
[rewriter_before_opt_a]: 7.036e-05
[expand_dump_flag]: 3.30003e-06
[jit_opt_a]: 0.00669034, [2]
    [Cycle 1]: 0.00219666, [27]
        [switch_simplify]: 6.814e-05
        [loop_unroll]: 2.951e-05
        [a_1]: 0.00063832
        [with_stream_mark]: 2.657e-05
        [recompute_prepare]: 7.92003e-06
        [updatestate_depend_eliminate]: 4.02e-06
        [updatestate_assign_eliminate]: 3.25e-06
        [updatestate_loads_eliminate]: 2.77002e-06
        [parameter_eliminate]: 2.11e-06
        [specialize_transform]: 4.467e-05
        [updatestate_useless_node_eliminater]: 5.69e-06
        [accelerated_algorithm]: 6.42001e-06
        [meta_shard_fg_expand]: 2.59999e-06
        [get_grad_eliminate_]: 5.82999e-06
        [merge_forward]: 4.61002e-06
        [cell_reuse_recompute_pass]: 1.37999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 3.97e-05
        [j_node_and_user_rematch]: 1.143e-05
        [meta_fg_expand]: 2.43998e-06
        [replace_old_param]: 1.158e-05
        [inline_without_move]: 5.81e-06
        [renormalize]: 0.00100377
        [add_forward_monad_depend]: 6.29999e-06
        [auto_monad_grad]: 2.59001e-06
        [auto_monad_eliminator]: 1.652e-05
        [cse]: 3.005e-05
        [replace_applicator]: 1.382e-05
    [Cycle 2]: 0.00033738, [27]
        [switch_simplify]: 5.47999e-06
        [loop_unroll]: 4.88001e-06
        [a_1]: 9.426e-05
        [with_stream_mark]: 1.141e-05
        [recompute_prepare]: 6.02999e-06
        [updatestate_depend_eliminate]: 3.43e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.95001e-06
        [parameter_eliminate]: 1.09998e-06
        [specialize_transform]: 5.19998e-06
        [updatestate_useless_node_eliminater]: 4.83001e-06
        [accelerated_algorithm]: 5.30999e-06
        [meta_shard_fg_expand]: 1.39998e-06
        [get_grad_eliminate_]: 4.84003e-06
        [merge_forward]: 2.86999e-06
        [cell_reuse_recompute_pass]: 1.87999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.455e-05
        [j_node_and_user_rematch]: 7.95e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 7.65998e-06
        [inline_without_move]: 4.57e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.47999e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 5.77001e-06
        [cse]: 1.194e-05
        [replace_applicator]: 4.92999e-06
[py_interpret_to_execute_after_opt_a]: 1.444e-05
[rewriter_after_opt_a]: 3.974e-05
[convert_after_rewriter]: 7.06001e-06
[order_py_execute_after_rewriter]: 4.90999e-06
[mutable_eliminate]: 0.0007327
[jit_opt_b]: 5.339e-05, [1]
    [Cycle 1]: 4.528e-05, [2]
        [frontend_op_eliminate]: 1.791e-05
        [inline_after_opt_a]: 1.541e-05
[cconv]: 3.141e-05
[loop_unroll]: 0.0004655
[jit_opt_after_cconv]: 0.0001593, [1]
    [Cycle 1]: 0.0001508, [11]
        [c_1]: 2.253e-05
        [parameter_eliminate]: 3.71999e-06
        [updatestate_depend_eliminate]: 8.62e-06
        [updatestate_assign_eliminate]: 2.59999e-06
        [updatestate_loads_eliminate]: 1.98002e-06
        [cse]: 2.399e-05
        [call_graph_tuple_transform]: 2.233e-05
        [tuple_list_get_item_eliminator]: 5.39e-06
        [none_parameter_eliminate]: 1.47999e-06
        [renormalize]: 1.12e-06
        [switch_simplify]: 5.56e-06
[remove_dup_value]: 1.605e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 6.21e-06
[add_recomputation]: 6.105e-05
[cse_after_recomputation]: 2.398e-05, [1]
    [Cycle 1]: 1.772e-05, [1]
        [cse]: 1.162e-05
[auto_monad_reorder]: 1.582e-05
[get_jit_bprop_graph]: 2.66999e-06
[rewriter_after_jit_bprop_graph]: 7.6e-06
[opt_after_jit_grad]: 0.00051353
[symbol_engine_optimizer]: 9.629e-05, [1]
    [Cycle 1]: 8.825e-05, [6]
        [build]: 4.74002e-06
        [elim_shapecalc]: 7.8e-06
        [elim_not_effective]: 1.506e-05
        [opt_reshape]: 6.26e-06
        [fold_const_symbol]: 9.59e-06
        [renormalize]: 7.80012e-07
[validate]: 4.521e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.0199833
[execute]: 1.114e-05
Sums
    bootstrap                            :     0.000453s :  1.46%
    type_inference                       :     0.006080s : 19.55%
    event_method                         :     0.000018s :  0.06%
    auto_monad                           :     0.000068s :  0.22%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000040s :  0.13%
    rewriter_before_opt_a                :     0.000070s :  0.23%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000074s :  0.24%
    jit_opt_a.loop_unroll                :     0.000034s :  0.11%
    jit_opt_a.a_1                        :     0.000733s :  2.36%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000050s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000054s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000019s :  0.06%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.001004s :  3.23%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.07%
    jit_opt_a.cse                        :     0.000042s :  0.13%
    jit_opt_a.replace_applicator         :     0.000019s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.05%
    rewriter_after_opt_a                 :     0.000040s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000733s :  2.36%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.05%
    cconv                                :     0.000031s :  0.10%
    loop_unroll                          :     0.000465s :  1.50%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000061s :  0.20%
    cse_after_recomputation.cse          :     0.000012s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000514s :  1.65%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.019983s : 64.24%
    execute                              :     0.000011s :  0.04%


Time group info:
------[substitution.]   0.000303    23
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 1.91% :     0.000006s :      3: substitution.graph_param_transform
80.17% :     0.000243s :      4: substitution.inline
 1.30% :     0.000004s :      4: substitution.j_node_and_user_rematch
 9.03% :     0.000027s :      4: substitution.remove_not_recompute_node
 1.93% :     0.000006s :      2: substitution.replace_old_param
 4.45% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006006     2
84.24% :     0.005060s :      1: type_inference.infer
15.76% :     0.000947s :      1: type_inference.specialize
------[replace.]   0.000078     6
67.17% :     0.000053s :      4: replace.inline
32.83% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000252     6
95.10% :     0.000240s :      4: match.inline
 4.90% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   691
 1.23% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.43% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.15% :     0.000002s :     10: predicate.addn_zero_filter
 5.93% :     0.000008s :     10: predicate.arithmetic_simplify
 1.41% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.66% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.02% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.56% :     0.000007s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.21% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.54% :     0.000002s :     15: predicate.load_eliminater
 1.61% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.50% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.76% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 1.93% :     0.000003s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     15: predicate.partial_eliminate
 1.31% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.54% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.53% :     0.000001s :      3: predicate.reset_defer_inline
 1.56% :     0.000002s :     10: predicate.reshape_eliminate
 1.53% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.10% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.06% :     0.000003s :     16: predicate.switch_defer_inline
 1.77% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.90% :     0.000010s :     53: predicate.switch_simplify
 1.19% :     0.000002s :     10: predicate.tile_eliminate
 1.04% :     0.000001s :     10: predicate.transpose_eliminate
 1.16% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.02% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.90% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.66% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.80% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.43% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000744    12
48.98% :     0.000364s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.34% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.68% :     0.000325s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038094    76
 0.17% :     0.000065s :      1: add_recomputation
 0.19% :     0.000071s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.23% :     0.000468s :      1: bootstrap
 0.09% :     0.000034s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.07% :     0.000026s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.06% :     0.000021s :      1: event_method
 0.04% :     0.000017s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
17.57% :     0.006694s :      1: jit_opt_a
 0.43% :     0.000162s :      1: jit_opt_after_cconv
 0.15% :     0.000057s :      1: jit_opt_b
 1.25% :     0.000475s :      1: loop_unroll
 1.95% :     0.000744s :      1: mutable_eliminate
 2.70% :     0.001028s :     26: opt.transform.jit_opt_a
 0.14% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000027s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.37% :     0.000524s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000043s :      1: py_interpret_to_execute
 0.04% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 1.52% :     0.000578s :      1: renormalize.infer
 1.10% :     0.000417s :      1: renormalize.specialize
 0.03% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000043s :      1: rewriter_after_opt_a
 0.19% :     0.000073s :      1: rewriter_before_opt_a
 0.26% :     0.000100s :      1: symbol_engine_optimizer
52.52% :     0.020007s :      1: task_emit
15.99% :     0.006093s :      1: type_inference
 0.18% :     0.000069s :      1: validate


TotalTime = 0.0201649, [30]
[bootstrap]: 0.00029355
[type_inference]: 0.0111771
[event_method]: 2.749e-05
[auto_monad]: 9.116e-05
[graph_reusing]: 7.55e-06
[pre_auto_parallel]: 2.74999e-06
[py_interpret_to_execute]: 5.656e-05
[rewriter_before_opt_a]: 9.375e-05
[expand_dump_flag]: 3.11001e-06
[jit_opt_a]: 0.00626754, [2]
    [Cycle 1]: 0.00207747, [27]
        [switch_simplify]: 0.0001482
        [loop_unroll]: 3.604e-05
        [a_1]: 0.00073296
        [with_stream_mark]: 2.175e-05
        [recompute_prepare]: 7.13e-06
        [updatestate_depend_eliminate]: 3.46999e-06
        [updatestate_assign_eliminate]: 2.87002e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 5.90002e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 5.76003e-06
        [meta_shard_fg_expand]: 2.00002e-06
        [get_grad_eliminate_]: 5.27001e-06
        [merge_forward]: 3.46001e-06
        [cell_reuse_recompute_pass]: 1.62999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.454e-05
        [j_node_and_user_rematch]: 9.42999e-06
        [meta_fg_expand]: 2.32001e-06
        [replace_old_param]: 1.033e-05
        [inline_without_move]: 5.24e-06
        [renormalize]: 0.0007952
        [add_forward_monad_depend]: 5.13002e-06
        [auto_monad_grad]: 2.93e-06
        [auto_monad_eliminator]: 1.4e-05
        [cse]: 3.016e-05
        [replace_applicator]: 1.247e-05
    [Cycle 2]: 0.00031399, [27]
        [switch_simplify]: 5.67001e-06
        [loop_unroll]: 4.68999e-06
        [a_1]: 8.812e-05
        [with_stream_mark]: 1.018e-05
        [recompute_prepare]: 4.80999e-06
        [updatestate_depend_eliminate]: 2.46998e-06
        [updatestate_assign_eliminate]: 1.97999e-06
        [updatestate_loads_eliminate]: 1.86998e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 4.88001e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 4.70999e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 4.58001e-06
        [merge_forward]: 3.14001e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.39e-05
        [j_node_and_user_rematch]: 8.15999e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 7.25e-06
        [inline_without_move]: 5.18002e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 1.14e-06
        [auto_monad_eliminator]: 4.99e-06
        [cse]: 9.58002e-06
        [replace_applicator]: 4.83001e-06
[py_interpret_to_execute_after_opt_a]: 1.125e-05
[rewriter_after_opt_a]: 3.58e-05
[convert_after_rewriter]: 6.88998e-06
[order_py_execute_after_rewriter]: 4.95999e-06
[mutable_eliminate]: 0.00058409
[jit_opt_b]: 4.817e-05, [1]
    [Cycle 1]: 4.135e-05, [2]
        [frontend_op_eliminate]: 1.613e-05
        [inline_after_opt_a]: 1.41e-05
[cconv]: 2.683e-05
[loop_unroll]: 0.00041287
[jit_opt_after_cconv]: 0.00013933, [1]
    [Cycle 1]: 0.00013311, [11]
        [c_1]: 2.103e-05
        [parameter_eliminate]: 3.26001e-06
        [updatestate_depend_eliminate]: 5.32001e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [cse]: 1.859e-05
        [call_graph_tuple_transform]: 1.808e-05
        [tuple_list_get_item_eliminator]: 5.12e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 6.29982e-07
        [switch_simplify]: 5.32001e-06
[remove_dup_value]: 1.404e-05
[partial_unused_args_eliminate]: 2.16998e-06
[environ_conv]: 4.92e-06
[add_recomputation]: 4.888e-05
[cse_after_recomputation]: 2.056e-05, [1]
    [Cycle 1]: 1.554e-05, [1]
        [cse]: 9.44e-06
[auto_monad_reorder]: 1.534e-05
[get_jit_bprop_graph]: 2.14e-06
[rewriter_after_jit_bprop_graph]: 5.42001e-06
[opt_after_jit_grad]: 0.00044008
[symbol_engine_optimizer]: 7.183e-05, [1]
    [Cycle 1]: 6.583e-05, [6]
        [build]: 3.80998e-06
        [elim_shapecalc]: 7.15e-06
        [elim_not_effective]: 1.253e-05
        [opt_reshape]: 5.92999e-06
        [fold_const_symbol]: 8.42e-06
        [renormalize]: 4.80009e-07
[validate]: 3.773e-05
Sums
    bootstrap                            :     0.000294s :  1.87%
    type_inference                       :     0.011177s : 71.38%
    event_method                         :     0.000027s :  0.18%
    auto_monad                           :     0.000091s :  0.58%
    graph_reusing                        :     0.000008s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000057s :  0.36%
    rewriter_before_opt_a                :     0.000094s :  0.60%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000154s :  0.98%
    jit_opt_a.loop_unroll                :     0.000041s :  0.26%
    jit_opt_a.a_1                        :     0.000821s :  5.24%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000795s :  5.08%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000040s :  0.25%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000584s :  3.73%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000027s :  0.17%
    loop_unroll                          :     0.000413s :  2.64%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000440s :  2.81%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.24%


Time group info:
------[substitution.]   0.000322    26
 0.62% :     0.000002s :      2: substitution.elim_not_effective
 0.38% :     0.000001s :      2: substitution.fold_const_symbol
 1.69% :     0.000005s :      3: substitution.graph_param_transform
86.20% :     0.000277s :      6: substitution.inline
 1.12% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.54% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.53% :     0.000005s :      2: substitution.replace_old_param
 2.88% :     0.000009s :      1: substitution.switch_simplify
 4.05% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011075     2
87.64% :     0.009706s :      1: type_inference.infer
12.36% :     0.001369s :      1: type_inference.specialize
------[replace.]   0.000147     9
46.64% :     0.000069s :      6: replace.inline
31.86% :     0.000047s :      1: replace.switch_simplify
21.50% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000294     9
93.12% :     0.000274s :      6: match.inline
 2.87% :     0.000008s :      1: match.switch_simplify
 4.01% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   801
 1.34% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.02% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     12: predicate.addn_check_dump
 1.30% :     0.000002s :     12: predicate.addn_zero_filter
 2.12% :     0.000003s :     12: predicate.arithmetic_simplify
 1.20% :     0.000002s :     12: predicate.cast_eliminate
 0.38% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     12: predicate.compare_switch_simplify
 1.44% :     0.000002s :     12: predicate.depend_value_elim
 1.15% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.22% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     12: predicate.environ_get_eliminate
 1.10% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.35% :     0.000000s :      3: predicate.graph_param_transform
 5.68% :     0.000008s :     26: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000002s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     17: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.26% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.86% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     12: predicate.merge_addn
 1.15% :     0.000002s :     12: predicate.minmaximum_grad
 1.87% :     0.000003s :      3: predicate.mutable_eliminate
 0.59% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     17: predicate.partial_eliminate
 1.16% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     12: predicate.reduce_eliminate
 1.42% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     20: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     12: predicate.reshape_eliminate
 1.30% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.75% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.52% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.77% :     0.000004s :     20: predicate.switch_defer_inline
 2.39% :     0.000003s :     20: predicate.switch_layer_defer_inline
 9.08% :     0.000012s :     65: predicate.switch_simplify
 1.20% :     0.000002s :     12: predicate.tile_eliminate
 1.16% :     0.000002s :     12: predicate.transpose_eliminate
 1.58% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.62% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.57% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     12: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000957    15
46.53% :     0.000445s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.82% :     0.000123s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.65% :     0.000389s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022130    72
 0.23% :     0.000052s :      1: add_recomputation
 0.43% :     0.000096s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 1.43% :     0.000318s :      1: bootstrap
 0.13% :     0.000030s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.15% :     0.000033s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000010s :      1: graph_reusing
28.34% :     0.006271s :      1: jit_opt_a
 0.64% :     0.000142s :      1: jit_opt_after_cconv
 0.23% :     0.000051s :      1: jit_opt_b
 1.90% :     0.000421s :      1: loop_unroll
 2.69% :     0.000594s :      1: mutable_eliminate
 5.11% :     0.001131s :     26: opt.transform.jit_opt_a
 0.21% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.02% :     0.000448s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000060s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.13% :     0.000470s :      1: renormalize.infer
 1.43% :     0.000315s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000039s :      1: rewriter_after_opt_a
 0.44% :     0.000097s :      1: rewriter_before_opt_a
 0.34% :     0.000074s :      1: symbol_engine_optimizer
50.65% :     0.011210s :      1: type_inference


TotalTime = 0.0202093, [30]
[bootstrap]: 0.00032893
[type_inference]: 0.00736607
[event_method]: 2.429e-05
[auto_monad]: 8.402e-05
[graph_reusing]: 7.62998e-06
[pre_auto_parallel]: 3.35e-06
[py_interpret_to_execute]: 5.717e-05
[rewriter_before_opt_a]: 8.837e-05
[expand_dump_flag]: 3.71999e-06
[jit_opt_a]: 0.00607816, [2]
    [Cycle 1]: 0.00214299, [27]
        [switch_simplify]: 0.00014822
        [loop_unroll]: 3.822e-05
        [a_1]: 0.00077156
        [with_stream_mark]: 2.546e-05
        [recompute_prepare]: 8.69e-06
        [updatestate_depend_eliminate]: 4.39002e-06
        [updatestate_assign_eliminate]: 3.04001e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 2.68e-06
        [specialize_transform]: 6.17999e-06
        [updatestate_useless_node_eliminater]: 5.34e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 2.27001e-06
        [get_grad_eliminate_]: 5.46e-06
        [merge_forward]: 4.3e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.574e-05
        [j_node_and_user_rematch]: 9.99001e-06
        [meta_fg_expand]: 2.42001e-06
        [replace_old_param]: 1.041e-05
        [inline_without_move]: 5.25999e-06
        [renormalize]: 0.00079294
        [add_forward_monad_depend]: 6.84001e-06
        [auto_monad_grad]: 2.71e-06
        [auto_monad_eliminator]: 1.509e-05
        [cse]: 3.189e-05
        [replace_applicator]: 1.382e-05
    [Cycle 2]: 0.00032099, [27]
        [switch_simplify]: 5.46e-06
        [loop_unroll]: 4.77e-06
        [a_1]: 8.966e-05
        [with_stream_mark]: 1.131e-05
        [recompute_prepare]: 5.20999e-06
        [updatestate_depend_eliminate]: 2.75997e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 4.80999e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 4.90999e-06
        [meta_shard_fg_expand]: 1.35999e-06
        [get_grad_eliminate_]: 4.4e-06
        [merge_forward]: 2.73003e-06
        [cell_reuse_recompute_pass]: 1.85001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.382e-05
        [j_node_and_user_rematch]: 8.80001e-06
        [meta_fg_expand]: 1.55001e-06
        [replace_old_param]: 7.84002e-06
        [inline_without_move]: 4.75999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 1.20001e-06
        [auto_monad_eliminator]: 5.12e-06
        [cse]: 9.46e-06
        [replace_applicator]: 4.99003e-06
[py_interpret_to_execute_after_opt_a]: 1.158e-05
[rewriter_after_opt_a]: 3.593e-05
[convert_after_rewriter]: 7.26001e-06
[order_py_execute_after_rewriter]: 5.23002e-06
[mutable_eliminate]: 0.00063401
[jit_opt_b]: 4.879e-05, [1]
    [Cycle 1]: 4.156e-05, [2]
        [frontend_op_eliminate]: 1.588e-05
        [inline_after_opt_a]: 1.464e-05
[cconv]: 2.742e-05
[loop_unroll]: 0.0004265
[jit_opt_after_cconv]: 0.00374358, [1]
    [Cycle 1]: 0.00373564, [11]
        [c_1]: 2.092e-05
        [parameter_eliminate]: 3.18e-06
        [updatestate_depend_eliminate]: 6.99001e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.06003e-06
        [cse]: 0.00353862
        [call_graph_tuple_transform]: 4.569e-05
        [tuple_list_get_item_eliminator]: 6.73e-06
        [none_parameter_eliminate]: 3.81001e-06
        [renormalize]: 7.09988e-07
        [switch_simplify]: 6.10002e-06
[remove_dup_value]: 1.853e-05
[partial_unused_args_eliminate]: 2.24999e-06
[environ_conv]: 6.58e-06
[add_recomputation]: 6.283e-05
[cse_after_recomputation]: 3.123e-05, [1]
    [Cycle 1]: 2.152e-05, [1]
        [cse]: 1.524e-05
[auto_monad_reorder]: 1.599e-05
[get_jit_bprop_graph]: 2.24999e-06
[rewriter_after_jit_bprop_graph]: 9.69e-06
[opt_after_jit_grad]: 0.00074572
[symbol_engine_optimizer]: 8.108e-05, [1]
    [Cycle 1]: 7.368e-05, [6]
        [build]: 4.99e-06
        [elim_shapecalc]: 7.73999e-06
        [elim_not_effective]: 1.475e-05
        [opt_reshape]: 6.28e-06
        [fold_const_symbol]: 8.92e-06
        [renormalize]: 8.90024e-07
[validate]: 4.576e-05
Sums
    bootstrap                            :     0.000329s :  2.07%
    type_inference                       :     0.007366s : 46.33%
    event_method                         :     0.000024s :  0.15%
    auto_monad                           :     0.000084s :  0.53%
    graph_reusing                        :     0.000008s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000057s :  0.36%
    rewriter_before_opt_a                :     0.000088s :  0.56%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000154s :  0.97%
    jit_opt_a.loop_unroll                :     0.000043s :  0.27%
    jit_opt_a.a_1                        :     0.000861s :  5.42%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000793s :  4.99%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000041s :  0.26%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000634s :  3.99%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000027s :  0.17%
    loop_unroll                          :     0.000427s :  2.68%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.003539s : 22.26%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000046s :  0.29%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000019s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000063s :  0.40%
    cse_after_recomputation.cse          :     0.000015s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.06%
    opt_after_jit_grad                   :     0.000746s :  4.69%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000046s :  0.29%


Time group info:
------[substitution.]   0.000333    26
 0.62% :     0.000002s :      2: substitution.elim_not_effective
 0.37% :     0.000001s :      2: substitution.fold_const_symbol
 2.32% :     0.000008s :      3: substitution.graph_param_transform
84.96% :     0.000283s :      6: substitution.inline
 1.26% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.65% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.59% :     0.000005s :      2: substitution.replace_old_param
 3.00% :     0.000010s :      1: substitution.switch_simplify
 4.24% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007286     2
83.76% :     0.006103s :      1: type_inference.infer
16.24% :     0.001183s :      1: type_inference.specialize
------[replace.]   0.000156     9
48.01% :     0.000075s :      6: replace.inline
30.58% :     0.000048s :      1: replace.switch_simplify
21.42% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000301     9
92.67% :     0.000279s :      6: match.inline
 3.03% :     0.000009s :      1: match.switch_simplify
 4.30% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000140   801
 1.41% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000002s :     12: predicate.addn_check_dump
 1.38% :     0.000002s :     12: predicate.addn_zero_filter
 2.33% :     0.000003s :     12: predicate.arithmetic_simplify
 1.47% :     0.000002s :     12: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000002s :     12: predicate.compare_switch_simplify
 1.65% :     0.000002s :     12: predicate.depend_value_elim
 1.12% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.18% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000001s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     12: predicate.environ_get_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.76% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000001s :      3: predicate.graph_param_transform
 5.43% :     0.000008s :     26: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :      6: predicate.less_batch_normalization
 1.96% :     0.000003s :     14: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     17: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.39% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.89% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000002s :     12: predicate.merge_addn
 1.10% :     0.000002s :     12: predicate.minmaximum_grad
 1.87% :     0.000003s :      3: predicate.mutable_eliminate
 0.57% :     0.000001s :      3: predicate.opt_reshape
 1.92% :     0.000003s :     17: predicate.partial_eliminate
 1.12% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     12: predicate.reduce_eliminate
 1.41% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     20: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     12: predicate.reshape_eliminate
 1.25% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.60% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.47% :     0.000002s :     12: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.75% :     0.000001s :      6: predicate.special_op_eliminate
 0.86% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.78% :     0.000004s :     20: predicate.switch_defer_inline
 2.47% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.69% :     0.000012s :     65: predicate.switch_simplify
 1.27% :     0.000002s :     12: predicate.tile_eliminate
 1.24% :     0.000002s :     12: predicate.transpose_eliminate
 1.32% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.94% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.49% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     12: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000882    15
49.54% :     0.000437s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.30% :     0.000117s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.17% :     0.000328s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022247    72
 0.30% :     0.000066s :      1: add_recomputation
 0.39% :     0.000088s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 1.59% :     0.000353s :      1: bootstrap
 0.14% :     0.000030s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000033s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.13% :     0.000029s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000010s :      1: graph_reusing
27.34% :     0.006082s :      1: jit_opt_a
16.85% :     0.003749s :      1: jit_opt_after_cconv
 0.23% :     0.000051s :      1: jit_opt_b
 1.95% :     0.000435s :      1: loop_unroll
 2.90% :     0.000645s :      1: mutable_eliminate
 5.30% :     0.001179s :     26: opt.transform.jit_opt_a
 0.32% :     0.000071s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.40% :     0.000755s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000060s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000021s :      1: remove_dup_value
 2.03% :     0.000452s :      1: renormalize.infer
 1.49% :     0.000331s :      1: renormalize.specialize
 0.05% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000039s :      1: rewriter_after_opt_a
 0.41% :     0.000092s :      1: rewriter_before_opt_a
 0.38% :     0.000084s :      1: symbol_engine_optimizer
33.21% :     0.007389s :      1: type_inference


TotalTime = 0.0388761, [33]
[bootstrap]: 0.00034301
[type_inference]: 0.00732684
[event_method]: 1.903e-05
[auto_monad]: 7.151e-05
[graph_reusing]: 6.23e-06
[pre_auto_parallel]: 3.19001e-06
[py_interpret_to_execute]: 4.056e-05
[rewriter_before_opt_a]: 7.545e-05
[expand_dump_flag]: 3.55e-06
[jit_opt_a]: 0.00694182, [2]
    [Cycle 1]: 0.00223755, [27]
        [switch_simplify]: 7.026e-05
        [loop_unroll]: 2.973e-05
        [a_1]: 0.00062609
        [with_stream_mark]: 2.667e-05
        [recompute_prepare]: 7.36999e-06
        [updatestate_depend_eliminate]: 4.35e-06
        [updatestate_assign_eliminate]: 2.88e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [parameter_eliminate]: 2.51e-06
        [specialize_transform]: 6.28e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 5.59e-06
        [meta_shard_fg_expand]: 2.11e-06
        [get_grad_eliminate_]: 5.03002e-06
        [merge_forward]: 4.39002e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.596e-05
        [j_node_and_user_rematch]: 9.36e-06
        [meta_fg_expand]: 2.43e-06
        [replace_old_param]: 1.03e-05
        [inline_without_move]: 5.23002e-06
        [renormalize]: 0.00108109
        [add_forward_monad_depend]: 8.18001e-06
        [auto_monad_grad]: 3.14001e-06
        [auto_monad_eliminator]: 1.793e-05
        [cse]: 5.614e-05
        [replace_applicator]: 1.783e-05
    [Cycle 2]: 0.00035554, [27]
        [switch_simplify]: 7.03e-06
        [loop_unroll]: 5.20001e-06
        [a_1]: 9.741e-05
        [with_stream_mark]: 1.412e-05
        [recompute_prepare]: 5.51e-06
        [updatestate_depend_eliminate]: 3.13998e-06
        [updatestate_assign_eliminate]: 1.94999e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 4.95001e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.25001e-06
        [meta_shard_fg_expand]: 1.89999e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 3.53e-06
        [cell_reuse_recompute_pass]: 2.96001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.702e-05
        [j_node_and_user_rematch]: 7.98001e-06
        [meta_fg_expand]: 2.20002e-06
        [replace_old_param]: 8.82e-06
        [inline_without_move]: 4.80999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 6.30997e-06
        [cse]: 1.164e-05
        [replace_applicator]: 4.92e-06
[py_interpret_to_execute_after_opt_a]: 1.774e-05
[rewriter_after_opt_a]: 4.311e-05
[convert_after_rewriter]: 7.19001e-06
[order_py_execute_after_rewriter]: 5.00001e-06
[mutable_eliminate]: 0.00089581
[jit_opt_b]: 6.201e-05, [1]
    [Cycle 1]: 5.203e-05, [2]
        [frontend_op_eliminate]: 1.98e-05
        [inline_after_opt_a]: 1.752e-05
[cconv]: 3.571e-05
[loop_unroll]: 0.00047894
[jit_opt_after_cconv]: 0.00016402, [1]
    [Cycle 1]: 0.00015702, [11]
        [c_1]: 2.225e-05
        [parameter_eliminate]: 4.05998e-06
        [updatestate_depend_eliminate]: 9.50001e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [cse]: 2.745e-05
        [call_graph_tuple_transform]: 2.374e-05
        [tuple_list_get_item_eliminator]: 5.22999e-06
        [none_parameter_eliminate]: 1.60001e-06
        [renormalize]: 7.50006e-07
        [switch_simplify]: 5.47999e-06
[remove_dup_value]: 1.525e-05
[partial_unused_args_eliminate]: 2.62001e-06
[environ_conv]: 6.11e-06
[add_recomputation]: 6.086e-05
[cse_after_recomputation]: 2.211e-05, [1]
    [Cycle 1]: 1.645e-05, [1]
        [cse]: 8.94998e-06
[auto_monad_reorder]: 1.628e-05
[get_jit_bprop_graph]: 2.64999e-06
[rewriter_after_jit_bprop_graph]: 8.77999e-06
[opt_after_jit_grad]: 0.00049967
[symbol_engine_optimizer]: 7.925e-05, [1]
    [Cycle 1]: 7.131e-05, [6]
        [build]: 4.35e-06
        [elim_shapecalc]: 8.76002e-06
        [elim_not_effective]: 1.458e-05
        [opt_reshape]: 6.41e-06
        [fold_const_symbol]: 9.36998e-06
        [renormalize]: 7.2e-07
[validate]: 4.383e-05
[backend_pass]: 9.10019e-07
[task_emit]: 0.0213097
[execute]: 9.82999e-06
Sums
    bootstrap                            :     0.000343s :  1.01%
    type_inference                       :     0.007327s : 21.67%
    event_method                         :     0.000019s :  0.06%
    auto_monad                           :     0.000072s :  0.21%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.12%
    rewriter_before_opt_a                :     0.000075s :  0.22%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000077s :  0.23%
    jit_opt_a.loop_unroll                :     0.000035s :  0.10%
    jit_opt_a.a_1                        :     0.000724s :  2.14%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000008s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000019s :  0.06%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.001081s :  3.20%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.07%
    jit_opt_a.cse                        :     0.000068s :  0.20%
    jit_opt_a.replace_applicator         :     0.000023s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.05%
    rewriter_after_opt_a                 :     0.000043s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000896s :  2.65%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.05%
    cconv                                :     0.000036s :  0.11%
    loop_unroll                          :     0.000479s :  1.42%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000015s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000061s :  0.18%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.03%
    opt_after_jit_grad                   :     0.000500s :  1.48%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.021310s : 63.04%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.000279    23
 0.66% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000002s :      2: substitution.fold_const_symbol
 2.45% :     0.000007s :      3: substitution.graph_param_transform
86.46% :     0.000241s :      4: substitution.inline
 1.12% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.70% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.25% :     0.000006s :      2: substitution.replace_old_param
 4.62% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007248     2
85.46% :     0.006194s :      1: type_inference.infer
14.54% :     0.001054s :      1: type_inference.specialize
------[replace.]   0.000080     6
68.25% :     0.000055s :      4: replace.inline
31.75% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000250     6
95.29% :     0.000238s :      4: match.inline
 4.71% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.79% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.12% :     0.000001s :     10: predicate.addn_zero_filter
 2.06% :     0.000003s :     10: predicate.arithmetic_simplify
 1.28% :     0.000002s :     10: predicate.cast_eliminate
 0.58% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.60% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.78% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.81% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.87% :     0.000007s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.67% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.56% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.80% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 2.37% :     0.000003s :      3: predicate.mutable_eliminate
 0.66% :     0.000001s :      3: predicate.opt_reshape
 2.43% :     0.000003s :     15: predicate.partial_eliminate
 1.01% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.39% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 1.05% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.45% :     0.000001s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.06% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.36% :     0.000002s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.61% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.18% :     0.000003s :     16: predicate.switch_defer_inline
 1.89% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.69% :     0.000010s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.21% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.09% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.04% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.07% :     0.000003s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000803    12
48.40% :     0.000389s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.60% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.00% :     0.000353s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.041022    76
 0.16% :     0.000064s :      1: add_recomputation
 0.18% :     0.000076s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.88% :     0.000362s :      1: bootstrap
 0.09% :     0.000038s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.06% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.06% :     0.000024s :      1: event_method
 0.04% :     0.000017s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
16.93% :     0.006947s :      1: jit_opt_a
 0.41% :     0.000167s :      1: jit_opt_after_cconv
 0.16% :     0.000065s :      1: jit_opt_b
 1.19% :     0.000487s :      1: loop_unroll
 2.21% :     0.000908s :      1: mutable_eliminate
 2.34% :     0.000960s :     26: opt.transform.jit_opt_a
 0.13% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000029s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.24% :     0.000508s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000043s :      1: py_interpret_to_execute
 0.05% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000018s :      1: remove_dup_value
 1.46% :     0.000600s :      1: renormalize.infer
 1.14% :     0.000470s :      1: renormalize.specialize
 0.03% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000047s :      1: rewriter_after_opt_a
 0.19% :     0.000079s :      1: rewriter_before_opt_a
 0.20% :     0.000082s :      1: symbol_engine_optimizer
52.01% :     0.021336s :      1: task_emit
17.91% :     0.007348s :      1: type_inference
 0.17% :     0.000069s :      1: validate


TotalTime = 0.0691407, [33]
[bootstrap]: 0.0004165
[type_inference]: 0.0226828
[event_method]: 1.966e-05
[auto_monad]: 7.407e-05
[graph_reusing]: 7.13998e-06
[pre_auto_parallel]: 2.87002e-06
[py_interpret_to_execute]: 5.002e-05
[rewriter_before_opt_a]: 7.94e-05
[expand_dump_flag]: 3.25e-06
[jit_opt_a]: 0.00882565, [2]
    [Cycle 1]: 0.00239623, [27]
        [switch_simplify]: 8.497e-05
        [loop_unroll]: 3.032e-05
        [a_1]: 0.00071987
        [with_stream_mark]: 2.961e-05
        [recompute_prepare]: 1.172e-05
        [updatestate_depend_eliminate]: 4.62e-06
        [updatestate_assign_eliminate]: 3.14999e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 2.87002e-06
        [specialize_transform]: 7.2e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 6.31e-06
        [meta_shard_fg_expand]: 3.26001e-06
        [get_grad_eliminate_]: 6.22001e-06
        [merge_forward]: 4.88001e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.784e-05
        [j_node_and_user_rematch]: 1.029e-05
        [meta_fg_expand]: 2.53e-06
        [replace_old_param]: 1.151e-05
        [inline_without_move]: 5.81e-06
        [renormalize]: 0.00111258
        [add_forward_monad_depend]: 1.082e-05
        [auto_monad_grad]: 2.88e-06
        [auto_monad_eliminator]: 1.927e-05
        [cse]: 3.199e-05
        [replace_applicator]: 2.247e-05
    [Cycle 2]: 0.00043654, [27]
        [switch_simplify]: 7.66999e-06
        [loop_unroll]: 5.26002e-06
        [a_1]: 0.00014785
        [with_stream_mark]: 1.945e-05
        [recompute_prepare]: 5.71998e-06
        [updatestate_depend_eliminate]: 3.43999e-06
        [updatestate_assign_eliminate]: 2.84999e-06
        [updatestate_loads_eliminate]: 2.36998e-06
        [parameter_eliminate]: 2.35002e-06
        [specialize_transform]: 5.76998e-06
        [updatestate_useless_node_eliminater]: 5.49998e-06
        [accelerated_algorithm]: 5.53002e-06
        [meta_shard_fg_expand]: 1.71e-06
        [get_grad_eliminate_]: 5.20001e-06
        [merge_forward]: 3.35e-06
        [cell_reuse_recompute_pass]: 4.02998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.978e-05
        [j_node_and_user_rematch]: 9.87999e-06
        [meta_fg_expand]: 1.81e-06
        [replace_old_param]: 8.67e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.86999e-06
        [auto_monad_grad]: 1.29e-06
        [auto_monad_eliminator]: 8.62e-06
        [cse]: 1.424e-05
        [replace_applicator]: 8e-06
[py_interpret_to_execute_after_opt_a]: 1.966e-05
[rewriter_after_opt_a]: 4.445e-05
[convert_after_rewriter]: 7.53e-06
[order_py_execute_after_rewriter]: 4.93001e-06
[mutable_eliminate]: 0.0010909
[jit_opt_b]: 6.672e-05, [1]
    [Cycle 1]: 5.662e-05, [2]
        [frontend_op_eliminate]: 2.269e-05
        [inline_after_opt_a]: 1.878e-05
[cconv]: 4.288e-05
[loop_unroll]: 0.00067489
[jit_opt_after_cconv]: 0.00024019, [1]
    [Cycle 1]: 0.00022991, [11]
        [c_1]: 2.719e-05
        [parameter_eliminate]: 6.76999e-06
        [updatestate_depend_eliminate]: 1.024e-05
        [updatestate_assign_eliminate]: 3.66999e-06
        [updatestate_loads_eliminate]: 3.21001e-06
        [cse]: 3.593e-05
        [call_graph_tuple_transform]: 3.121e-05
        [tuple_list_get_item_eliminator]: 6.42001e-06
        [none_parameter_eliminate]: 2.20002e-06
        [renormalize]: 1.03001e-06
        [switch_simplify]: 6.97002e-06
[remove_dup_value]: 1.871e-05
[partial_unused_args_eliminate]: 2.86999e-06
[environ_conv]: 7.65e-06
[add_recomputation]: 5.92e-05
[cse_after_recomputation]: 2.906e-05, [1]
    [Cycle 1]: 2.171e-05, [1]
        [cse]: 1.325e-05
[auto_monad_reorder]: 1.688e-05
[get_jit_bprop_graph]: 2.94001e-06
[rewriter_after_jit_bprop_graph]: 1.326e-05
[opt_after_jit_grad]: 0.00080769
[symbol_engine_optimizer]: 0.00010056, [1]
    [Cycle 1]: 8.971e-05, [6]
        [build]: 5.62001e-06
        [elim_shapecalc]: 9.90002e-06
        [elim_not_effective]: 1.99e-05
        [opt_reshape]: 7.77002e-06
        [fold_const_symbol]: 1.114e-05
        [renormalize]: 1.59e-06
[validate]: 5.525e-05
[backend_pass]: 1.20999e-06
[task_emit]: 0.0333717
[execute]: 1.066e-05
Sums
    bootstrap                            :     0.000416s :  0.67%
    type_inference                       :     0.022683s : 36.40%
    event_method                         :     0.000020s :  0.03%
    auto_monad                           :     0.000074s :  0.12%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000050s :  0.08%
    rewriter_before_opt_a                :     0.000079s :  0.13%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000093s :  0.15%
    jit_opt_a.loop_unroll                :     0.000036s :  0.06%
    jit_opt_a.a_1                        :     0.000868s :  1.39%
    jit_opt_a.with_stream_mark           :     0.000049s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000013s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.02%
    jit_opt_a.merge_forward              :     0.000008s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000020s :  0.03%
    jit_opt_a.inline_without_move        :     0.000011s :  0.02%
    jit_opt_a.renormalize                :     0.001113s :  1.79%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.04%
    jit_opt_a.cse                        :     0.000046s :  0.07%
    jit_opt_a.replace_applicator         :     0.000030s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000020s :  0.03%
    rewriter_after_opt_a                 :     0.000044s :  0.07%
    convert_after_rewriter               :     0.000008s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.001091s :  1.75%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.03%
    cconv                                :     0.000043s :  0.07%
    loop_unroll                          :     0.000675s :  1.08%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.04%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000036s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000031s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.01%
    remove_dup_value                     :     0.000019s :  0.03%
    partial_unused_args_eliminate        :     0.000003s :  0.00%
    environ_conv                         :     0.000008s :  0.01%
    add_recomputation                    :     0.000059s :  0.10%
    cse_after_recomputation.cse          :     0.000013s :  0.02%
    auto_monad_reorder                   :     0.000017s :  0.03%
    get_jit_bprop_graph                  :     0.000003s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000013s :  0.02%
    opt_after_jit_grad                   :     0.000808s :  1.30%
    symbol_engine_optimizer.build        :     0.000006s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000002s :  0.00%
    validate                             :     0.000055s :  0.09%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.033372s : 53.55%
    execute                              :     0.000011s :  0.02%


Time group info:
------[substitution.]   0.000325    23
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000002s :      2: substitution.fold_const_symbol
 2.38% :     0.000008s :      3: substitution.graph_param_transform
86.03% :     0.000280s :      4: substitution.inline
 1.49% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.99% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.13% :     0.000007s :      2: substitution.replace_old_param
 4.64% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.022583     2
94.04% :     0.021238s :      1: type_inference.infer
 5.96% :     0.001345s :      1: type_inference.specialize
------[replace.]   0.000092     6
67.40% :     0.000062s :      4: replace.inline
32.60% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000290     6
95.18% :     0.000276s :      4: match.inline
 4.82% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000157   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.00% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.78% :     0.000001s :     10: predicate.addn_check_dump
 1.66% :     0.000003s :     10: predicate.addn_zero_filter
 2.76% :     0.000004s :     10: predicate.arithmetic_simplify
 0.98% :     0.000002s :     10: predicate.cast_eliminate
 0.36% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.77% :     0.000001s :     10: predicate.compare_switch_simplify
 2.01% :     0.000003s :     10: predicate.depend_value_elim
 1.29% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.57% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.79% :     0.000003s :     10: predicate.dict_set_item_eliminator
 1.37% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.52% :     0.000001s :      3: predicate.elim_not_effective
 0.84% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.48% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.78% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.97% :     0.000002s :     10: predicate.environ_get_eliminate
 0.78% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.34% :     0.000001s :      3: predicate.graph_param_transform
 5.60% :     0.000009s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.27% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.64% :     0.000003s :      6: predicate.less_batch_normalization
 1.68% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.52% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.24% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.14% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.93% :     0.000001s :     10: predicate.merge_addn
 0.91% :     0.000001s :     10: predicate.minmaximum_grad
 2.86% :     0.000004s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 1.83% :     0.000003s :     15: predicate.partial_eliminate
 0.85% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.29% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.54% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.39% :     0.000001s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 0.95% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.13% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000002s :     10: predicate.same_eliminate
 0.40% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.20% :     0.000002s :      6: predicate.special_op_eliminate
 0.75% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.15% :     0.000003s :     16: predicate.switch_defer_inline
 1.76% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.12% :     0.000011s :     53: predicate.switch_simplify
 1.52% :     0.000002s :     10: predicate.tile_eliminate
 1.19% :     0.000002s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.76% :     0.000007s :     18: predicate.tuple_list_get_item_eliminator
 2.00% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.25% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.51% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.16% :     0.000003s :     10: predicate.value_based_eliminate
 0.37% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001330    12
58.31% :     0.000775s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.51% :     0.000060s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.18% :     0.000494s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.071556    76
 0.09% :     0.000063s :      1: add_recomputation
 0.11% :     0.000077s :      1: auto_monad
 0.03% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.62% :     0.000440s :      1: bootstrap
 0.06% :     0.000046s :      1: cconv
 0.01% :     0.000010s :      1: convert_after_rewriter
 0.04% :     0.000032s :      1: cse_after_recomputation
 0.01% :     0.000010s :      1: environ_conv
 0.03% :     0.000023s :      1: event_method
 0.02% :     0.000016s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000006s :      1: get_jit_bprop_graph
 0.01% :     0.000009s :      1: graph_reusing
12.34% :     0.008831s :      1: jit_opt_a
 0.34% :     0.000245s :      1: jit_opt_after_cconv
 0.10% :     0.000071s :      1: jit_opt_b
 0.96% :     0.000688s :      1: loop_unroll
 1.55% :     0.001111s :      1: mutable_eliminate
 1.60% :     0.001143s :     26: opt.transform.jit_opt_a
 0.09% :     0.000067s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000033s :      4: opt.transform.jit_opt_b
 0.03% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000032s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000041s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000045s :      4: opt.transform.symbol_engine_opt
 1.15% :     0.000825s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000053s :      1: py_interpret_to_execute
 0.03% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000022s :      1: remove_dup_value
 0.85% :     0.000607s :      1: renormalize.infer
 0.69% :     0.000493s :      1: renormalize.specialize
 0.02% :     0.000016s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000048s :      1: rewriter_after_opt_a
 0.12% :     0.000082s :      1: rewriter_before_opt_a
 0.15% :     0.000104s :      1: symbol_engine_optimizer
46.67% :     0.033392s :      1: task_emit
31.73% :     0.022705s :      1: type_inference
 0.12% :     0.000082s :      1: validate


TotalTime = 0.0360188, [33]
[bootstrap]: 0.00048678
[type_inference]: 0.00650971
[event_method]: 2.095e-05
[auto_monad]: 7.258e-05
[graph_reusing]: 6.86999e-06
[pre_auto_parallel]: 3.45e-06
[py_interpret_to_execute]: 5.33e-05
[rewriter_before_opt_a]: 7.991e-05
[expand_dump_flag]: 3.48e-06
[jit_opt_a]: 0.00679554, [2]
    [Cycle 1]: 0.00205546, [27]
        [switch_simplify]: 6.256e-05
        [loop_unroll]: 2.892e-05
        [a_1]: 0.00063023
        [with_stream_mark]: 2.526e-05
        [recompute_prepare]: 7.66001e-06
        [updatestate_depend_eliminate]: 4.16001e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [parameter_eliminate]: 2.56e-06
        [specialize_transform]: 5.46e-06
        [updatestate_useless_node_eliminater]: 5.32999e-06
        [accelerated_algorithm]: 5.74e-06
        [meta_shard_fg_expand]: 2.63e-06
        [get_grad_eliminate_]: 5.32001e-06
        [merge_forward]: 3.94002e-06
        [cell_reuse_recompute_pass]: 1.35001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.431e-05
        [j_node_and_user_rematch]: 9.61e-06
        [meta_fg_expand]: 2.48998e-06
        [replace_old_param]: 1.014e-05
        [inline_without_move]: 5.84e-06
        [renormalize]: 0.00096291
        [add_forward_monad_depend]: 6.18002e-06
        [auto_monad_grad]: 2.83e-06
        [auto_monad_eliminator]: 1.514e-05
        [cse]: 2.984e-05
        [replace_applicator]: 1.181e-05
    [Cycle 2]: 0.0003211, [27]
        [switch_simplify]: 5.27001e-06
        [loop_unroll]: 4.92999e-06
        [a_1]: 9.061e-05
        [with_stream_mark]: 9.52999e-06
        [recompute_prepare]: 5.05999e-06
        [updatestate_depend_eliminate]: 2.58998e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 4.92e-06
        [updatestate_useless_node_eliminater]: 4.80999e-06
        [accelerated_algorithm]: 5.02999e-06
        [meta_shard_fg_expand]: 1.13001e-06
        [get_grad_eliminate_]: 4.87998e-06
        [merge_forward]: 2.40002e-06
        [cell_reuse_recompute_pass]: 2.36e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.186e-05
        [j_node_and_user_rematch]: 7.90998e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 7.02002e-06
        [inline_without_move]: 4.92999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.19998e-06
        [auto_monad_grad]: 9.49978e-07
        [auto_monad_eliminator]: 5.52999e-06
        [cse]: 1.048e-05
        [replace_applicator]: 4.97e-06
[py_interpret_to_execute_after_opt_a]: 1.169e-05
[rewriter_after_opt_a]: 3.779e-05
[convert_after_rewriter]: 7.72002e-06
[order_py_execute_after_rewriter]: 5.17e-06
[mutable_eliminate]: 0.00057773
[jit_opt_b]: 5.028e-05, [1]
    [Cycle 1]: 4.281e-05, [2]
        [frontend_op_eliminate]: 1.605e-05
        [inline_after_opt_a]: 1.549e-05
[cconv]: 2.71e-05
[loop_unroll]: 0.00042842
[jit_opt_after_cconv]: 0.00013996, [1]
    [Cycle 1]: 0.00013375, [11]
        [c_1]: 2.051e-05
        [parameter_eliminate]: 2.96001e-06
        [updatestate_depend_eliminate]: 5.91998e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [cse]: 1.795e-05
        [call_graph_tuple_transform]: 1.945e-05
        [tuple_list_get_item_eliminator]: 5.41002e-06
        [none_parameter_eliminate]: 1.58002e-06
        [renormalize]: 4.60015e-07
        [switch_simplify]: 5.23002e-06
[remove_dup_value]: 1.472e-05
[partial_unused_args_eliminate]: 2.34999e-06
[environ_conv]: 5.34e-06
[add_recomputation]: 5.297e-05
[cse_after_recomputation]: 2.005e-05, [1]
    [Cycle 1]: 1.467e-05, [1]
        [cse]: 8.77e-06
[auto_monad_reorder]: 1.686e-05
[get_jit_bprop_graph]: 2.01003e-06
[rewriter_after_jit_bprop_graph]: 5.51e-06
[opt_after_jit_grad]: 0.00046669
[symbol_engine_optimizer]: 0.0123596, [1]
    [Cycle 1]: 0.0123357, [6]
        [build]: 3.93001e-06
        [elim_shapecalc]: 7.20998e-06
        [elim_not_effective]: 1.322e-05
        [opt_reshape]: 5.42001e-06
        [fold_const_symbol]: 8.99e-06
        [renormalize]: 0.0122188
[validate]: 8.917e-05
[backend_pass]: 1.43002e-06
[task_emit]: 0.00736422
[execute]: 9.76e-06
Sums
    bootstrap                            :     0.000487s :  1.58%
    type_inference                       :     0.006510s : 21.12%
    event_method                         :     0.000021s :  0.07%
    auto_monad                           :     0.000073s :  0.24%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000053s :  0.17%
    rewriter_before_opt_a                :     0.000080s :  0.26%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000068s :  0.22%
    jit_opt_a.loop_unroll                :     0.000034s :  0.11%
    jit_opt_a.a_1                        :     0.000721s :  2.34%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000010s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000017s :  0.06%
    jit_opt_a.inline_without_move        :     0.000011s :  0.03%
    jit_opt_a.renormalize                :     0.000963s :  3.12%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.07%
    jit_opt_a.cse                        :     0.000040s :  0.13%
    jit_opt_a.replace_applicator         :     0.000017s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000038s :  0.12%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000578s :  1.87%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.05%
    cconv                                :     0.000027s :  0.09%
    loop_unroll                          :     0.000428s :  1.39%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000015s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000053s :  0.17%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000017s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000467s :  1.51%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.012219s : 39.65%
    validate                             :     0.000089s :  0.29%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007364s : 23.90%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.000279    23
 0.69% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000002s :      2: substitution.fold_const_symbol
 1.99% :     0.000006s :      3: substitution.graph_param_transform
86.98% :     0.000243s :      4: substitution.inline
 1.39% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.46% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.73% :     0.000005s :      2: substitution.replace_old_param
 5.13% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006414     2
82.28% :     0.005277s :      1: type_inference.infer
17.72% :     0.001137s :      1: type_inference.specialize
------[replace.]   0.000079     6
68.26% :     0.000054s :      4: replace.inline
31.74% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000254     6
94.82% :     0.000240s :      4: match.inline
 5.18% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.69% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.44% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000001s :     10: predicate.addn_zero_filter
 1.95% :     0.000002s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.61% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.49% :     0.000001s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.77% :     0.000007s :     22: predicate.inline
 1.02% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.86% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.82% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.47% :     0.000002s :     10: predicate.same_eliminate
 0.60% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.77% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.80% :     0.000009s :     53: predicate.switch_simplify
 1.29% :     0.000002s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.97% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.09% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000815    12
48.86% :     0.000398s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.00% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.14% :     0.000360s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.037995    76
 0.15% :     0.000056s :      1: add_recomputation
 0.20% :     0.000076s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 1.35% :     0.000512s :      1: bootstrap
 0.08% :     0.000030s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.06% :     0.000022s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.07% :     0.000026s :      1: event_method
 0.04% :     0.000015s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
17.90% :     0.006800s :      1: jit_opt_a
 0.38% :     0.000143s :      1: jit_opt_after_cconv
 0.14% :     0.000053s :      1: jit_opt_b
 1.15% :     0.000437s :      1: loop_unroll
 1.54% :     0.000586s :      1: mutable_eliminate
 2.47% :     0.000938s :     26: opt.transform.jit_opt_a
 0.12% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000025s :      4: opt.transform.jit_opt_b
 0.03% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.25% :     0.000476s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.15% :     0.000056s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000017s :      1: remove_dup_value
 1.45% :     0.000551s :      1: renormalize.infer
 1.06% :     0.000403s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000041s :      1: rewriter_after_opt_a
 0.22% :     0.000083s :      1: rewriter_before_opt_a
32.56% :     0.012370s :      1: symbol_engine_optimizer
19.44% :     0.007385s :      1: task_emit
17.21% :     0.006537s :      1: type_inference
 0.35% :     0.000132s :      1: validate


TotalTime = 0.0144028, [30]
[bootstrap]: 0.00030626
[type_inference]: 0.00620951
[event_method]: 2.049e-05
[auto_monad]: 9.288e-05
[graph_reusing]: 7e-06
[pre_auto_parallel]: 2.98e-06
[py_interpret_to_execute]: 4.552e-05
[rewriter_before_opt_a]: 8.573e-05
[expand_dump_flag]: 3.04001e-06
[jit_opt_a]: 0.00543206, [2]
    [Cycle 1]: 0.00182383, [27]
        [switch_simplify]: 0.00011722
        [loop_unroll]: 3.451e-05
        [a_1]: 0.0006768
        [with_stream_mark]: 1.956e-05
        [recompute_prepare]: 7.35998e-06
        [updatestate_depend_eliminate]: 3.2e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.52001e-06
        [parameter_eliminate]: 1.82999e-06
        [specialize_transform]: 5.97001e-06
        [updatestate_useless_node_eliminater]: 5.17999e-06
        [accelerated_algorithm]: 5.16002e-06
        [meta_shard_fg_expand]: 2.03997e-06
        [get_grad_eliminate_]: 5.21998e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.424e-05
        [j_node_and_user_rematch]: 8.98002e-06
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 8.94e-06
        [inline_without_move]: 5.02e-06
        [renormalize]: 0.0006492
        [add_forward_monad_depend]: 5.80002e-06
        [auto_monad_grad]: 2.64001e-06
        [auto_monad_eliminator]: 1.32e-05
        [cse]: 2.934e-05
        [replace_applicator]: 1.181e-05
    [Cycle 2]: 0.00038402, [27]
        [switch_simplify]: 5.72999e-06
        [loop_unroll]: 4.67e-06
        [a_1]: 8.778e-05
        [with_stream_mark]: 8.70999e-06
        [recompute_prepare]: 4.99e-06
        [updatestate_depend_eliminate]: 2.54999e-06
        [updatestate_assign_eliminate]: 2.14999e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 4.88001e-06
        [updatestate_useless_node_eliminater]: 4.58001e-06
        [accelerated_algorithm]: 4.85999e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 4.45999e-06
        [merge_forward]: 2.39001e-06
        [cell_reuse_recompute_pass]: 7.339e-05
        [cell_reuse_handle_not_recompute_node_pass]: 1.253e-05
        [j_node_and_user_rematch]: 7.73999e-06
        [meta_fg_expand]: 1.46998e-06
        [replace_old_param]: 7.4e-06
        [inline_without_move]: 4.75001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.75001e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 4.85999e-06
        [cse]: 1.127e-05
        [replace_applicator]: 4.95001e-06
[py_interpret_to_execute_after_opt_a]: 1.002e-05
[rewriter_after_opt_a]: 3.514e-05
[convert_after_rewriter]: 6.63e-06
[order_py_execute_after_rewriter]: 4.82998e-06
[mutable_eliminate]: 0.00050716
[jit_opt_b]: 4.855e-05, [1]
    [Cycle 1]: 4.149e-05, [2]
        [frontend_op_eliminate]: 1.606e-05
        [inline_after_opt_a]: 1.378e-05
[cconv]: 2.438e-05
[loop_unroll]: 0.00041907
[jit_opt_after_cconv]: 0.00014062, [1]
    [Cycle 1]: 0.00013438, [11]
        [c_1]: 2.054e-05
        [parameter_eliminate]: 2.71999e-06
        [updatestate_depend_eliminate]: 5.97001e-06
        [updatestate_assign_eliminate]: 2.57001e-06
        [updatestate_loads_eliminate]: 2.39001e-06
        [cse]: 1.844e-05
        [call_graph_tuple_transform]: 1.843e-05
        [tuple_list_get_item_eliminator]: 5.55001e-06
        [none_parameter_eliminate]: 1.53002e-06
        [renormalize]: 5.60016e-07
        [switch_simplify]: 5.44998e-06
[remove_dup_value]: 1.333e-05
[partial_unused_args_eliminate]: 1.97999e-06
[environ_conv]: 4.92e-06
[add_recomputation]: 4.768e-05
[cse_after_recomputation]: 2.137e-05, [1]
    [Cycle 1]: 1.611e-05, [1]
        [cse]: 1.032e-05
[auto_monad_reorder]: 1.497e-05
[get_jit_bprop_graph]: 1.92999e-06
[rewriter_after_jit_bprop_graph]: 4.50001e-06
[opt_after_jit_grad]: 0.00060403
[symbol_engine_optimizer]: 7.511e-05, [1]
    [Cycle 1]: 6.817e-05, [6]
        [build]: 4.88001e-06
        [elim_shapecalc]: 7.46001e-06
        [elim_not_effective]: 1.353e-05
        [opt_reshape]: 5.51002e-06
        [fold_const_symbol]: 8.36002e-06
        [renormalize]: 7.50006e-07
[validate]: 3.597e-05
Sums
    bootstrap                            :     0.000306s :  2.89%
    type_inference                       :     0.006210s : 58.62%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000093s :  0.88%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000046s :  0.43%
    rewriter_before_opt_a                :     0.000086s :  0.81%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000123s :  1.16%
    jit_opt_a.loop_unroll                :     0.000039s :  0.37%
    jit_opt_a.a_1                        :     0.000765s :  7.22%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000075s :  0.71%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000649s :  6.13%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.17%
    jit_opt_a.cse                        :     0.000041s :  0.38%
    jit_opt_a.replace_applicator         :     0.000017s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.09%
    rewriter_after_opt_a                 :     0.000035s :  0.33%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000507s :  4.79%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.13%
    cconv                                :     0.000024s :  0.23%
    loop_unroll                          :     0.000419s :  3.96%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000048s :  0.45%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000604s :  5.70%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000036s :  0.34%


Time group info:
------[substitution.]   0.000271    26
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 1.93% :     0.000005s :      3: substitution.graph_param_transform
84.69% :     0.000229s :      6: substitution.inline
 1.21% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.70% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.59% :     0.000004s :      2: substitution.replace_old_param
 2.76% :     0.000007s :      1: substitution.switch_simplify
 4.92% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006144     2
84.15% :     0.005170s :      1: type_inference.infer
15.85% :     0.000974s :      1: type_inference.specialize
------[replace.]   0.000130     9
49.27% :     0.000064s :      6: replace.inline
27.62% :     0.000036s :      1: replace.switch_simplify
23.11% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000245     9
92.26% :     0.000226s :      6: match.inline
 2.77% :     0.000007s :      1: match.switch_simplify
 4.97% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   801
 1.43% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.20% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000001s :     12: predicate.addn_check_dump
 1.39% :     0.000002s :     12: predicate.addn_zero_filter
 1.80% :     0.000002s :     12: predicate.arithmetic_simplify
 1.49% :     0.000002s :     12: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     12: predicate.compare_switch_simplify
 1.58% :     0.000002s :     12: predicate.depend_value_elim
 1.15% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.46% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     12: predicate.environ_add_const_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     12: predicate.environ_get_eliminate
 1.21% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.59% :     0.000007s :     26: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.89% :     0.000001s :      6: predicate.less_batch_normalization
 1.64% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     17: predicate.load_eliminater
 1.71% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.11% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     12: predicate.merge_addn
 1.16% :     0.000002s :     12: predicate.minmaximum_grad
 2.34% :     0.000003s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.39% :     0.000003s :     17: predicate.partial_eliminate
 1.24% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     12: predicate.reduce_eliminate
 1.68% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     20: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     12: predicate.reshape_eliminate
 1.26% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     12: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.84% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.67% :     0.000003s :     20: predicate.switch_defer_inline
 2.37% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.39% :     0.000011s :     65: predicate.switch_simplify
 1.22% :     0.000002s :     12: predicate.tile_eliminate
 1.23% :     0.000002s :     12: predicate.transpose_eliminate
 1.35% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.69% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     12: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000766    15
50.61% :     0.000388s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.27% :     0.000094s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.12% :     0.000284s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016146    72
 0.31% :     0.000051s :      1: add_recomputation
 0.60% :     0.000097s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 2.01% :     0.000325s :      1: bootstrap
 0.17% :     0.000027s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000010s :      1: graph_reusing
33.66% :     0.005435s :      1: jit_opt_a
 0.89% :     0.000143s :      1: jit_opt_after_cconv
 0.32% :     0.000052s :      1: jit_opt_b
 2.64% :     0.000427s :      1: loop_unroll
 3.19% :     0.000515s :      1: mutable_eliminate
 6.43% :     0.001039s :     26: opt.transform.jit_opt_a
 0.29% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.80% :     0.000614s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000048s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.26% :     0.000365s :      1: renormalize.infer
 1.71% :     0.000276s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000038s :      1: rewriter_after_opt_a
 0.55% :     0.000089s :      1: rewriter_before_opt_a
 0.48% :     0.000078s :      1: symbol_engine_optimizer
38.55% :     0.006225s :      1: type_inference


TotalTime = 0.015461, [30]
[bootstrap]: 0.00029042
[type_inference]: 0.00676569
[event_method]: 2.24e-05
[auto_monad]: 8.155e-05
[graph_reusing]: 6.38e-06
[pre_auto_parallel]: 2.53e-06
[py_interpret_to_execute]: 4.967e-05
[rewriter_before_opt_a]: 8.967e-05
[expand_dump_flag]: 3.24001e-06
[jit_opt_a]: 0.00595193, [2]
    [Cycle 1]: 0.00209244, [27]
        [switch_simplify]: 0.00012949
        [loop_unroll]: 3.356e-05
        [a_1]: 0.00072209
        [with_stream_mark]: 2.091e-05
        [recompute_prepare]: 7.36001e-06
        [updatestate_depend_eliminate]: 3.55e-06
        [updatestate_assign_eliminate]: 3.61999e-06
        [updatestate_loads_eliminate]: 2.52001e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 6.29001e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 2.51998e-06
        [get_grad_eliminate_]: 3.151e-05
        [merge_forward]: 3.73999e-06
        [cell_reuse_recompute_pass]: 1.06002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.601e-05
        [j_node_and_user_rematch]: 9.22001e-06
        [meta_fg_expand]: 2.19001e-06
        [replace_old_param]: 9.00001e-06
        [inline_without_move]: 5.60001e-06
        [renormalize]: 0.000758
        [add_forward_monad_depend]: 5.86e-06
        [auto_monad_grad]: 2.48002e-06
        [auto_monad_eliminator]: 1.541e-05
        [cse]: 3.135e-05
        [replace_applicator]: 1.244e-05
    [Cycle 2]: 0.00033809, [27]
        [switch_simplify]: 5.47999e-06
        [loop_unroll]: 5.07e-06
        [a_1]: 8.88e-05
        [with_stream_mark]: 1.072e-05
        [recompute_prepare]: 5.03002e-06
        [updatestate_depend_eliminate]: 2.82002e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 1.98997e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 5.35999e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 4.67e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 5.19e-06
        [merge_forward]: 3.05998e-06
        [cell_reuse_recompute_pass]: 1.20999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.219e-05
        [j_node_and_user_rematch]: 7.51001e-06
        [meta_fg_expand]: 1.73002e-06
        [replace_old_param]: 1.291e-05
        [inline_without_move]: 5.07e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.70001e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 4.98001e-06
        [cse]: 1.071e-05
        [replace_applicator]: 5.56002e-06
[py_interpret_to_execute_after_opt_a]: 1.044e-05
[rewriter_after_opt_a]: 3.547e-05
[convert_after_rewriter]: 6.69999e-06
[order_py_execute_after_rewriter]: 5.24998e-06
[mutable_eliminate]: 0.00057699
[jit_opt_b]: 4.908e-05, [1]
    [Cycle 1]: 4.173e-05, [2]
        [frontend_op_eliminate]: 1.5e-05
        [inline_after_opt_a]: 1.484e-05
[cconv]: 2.411e-05
[loop_unroll]: 0.0004301
[jit_opt_after_cconv]: 0.00015022, [1]
    [Cycle 1]: 0.00014395, [11]
        [c_1]: 2.396e-05
        [parameter_eliminate]: 2.83998e-06
        [updatestate_depend_eliminate]: 6.30002e-06
        [updatestate_assign_eliminate]: 2.59999e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [cse]: 2.032e-05
        [call_graph_tuple_transform]: 1.893e-05
        [tuple_list_get_item_eliminator]: 5.87001e-06
        [none_parameter_eliminate]: 1.77001e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 5.35001e-06
[remove_dup_value]: 1.346e-05
[partial_unused_args_eliminate]: 2.21998e-06
[environ_conv]: 5.56e-06
[add_recomputation]: 5.017e-05
[cse_after_recomputation]: 2.192e-05, [1]
    [Cycle 1]: 1.612e-05, [1]
        [cse]: 9.80002e-06
[auto_monad_reorder]: 1.502e-05
[get_jit_bprop_graph]: 2.07999e-06
[rewriter_after_jit_bprop_graph]: 5.63002e-06
[opt_after_jit_grad]: 0.00049192
[symbol_engine_optimizer]: 7.844e-05, [1]
    [Cycle 1]: 7.192e-05, [6]
        [build]: 4.95999e-06
        [elim_shapecalc]: 7.31001e-06
        [elim_not_effective]: 1.368e-05
        [opt_reshape]: 6.45002e-06
        [fold_const_symbol]: 8.92e-06
        [renormalize]: 6.19999e-07
[validate]: 3.872e-05
Sums
    bootstrap                            :     0.000290s :  2.58%
    type_inference                       :     0.006766s : 60.10%
    event_method                         :     0.000022s :  0.20%
    auto_monad                           :     0.000082s :  0.72%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000050s :  0.44%
    rewriter_before_opt_a                :     0.000090s :  0.80%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000135s :  1.20%
    jit_opt_a.loop_unroll                :     0.000039s :  0.34%
    jit_opt_a.a_1                        :     0.000811s :  7.20%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000037s :  0.33%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000022s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.000758s :  6.73%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.18%
    jit_opt_a.cse                        :     0.000042s :  0.37%
    jit_opt_a.replace_applicator         :     0.000018s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.09%
    rewriter_after_opt_a                 :     0.000035s :  0.32%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000577s :  5.13%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.13%
    cconv                                :     0.000024s :  0.21%
    loop_unroll                          :     0.000430s :  3.82%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000050s :  0.45%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000492s :  4.37%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.34%


Time group info:
------[substitution.]   0.000308    26
 0.60% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000002s :      2: substitution.fold_const_symbol
 1.73% :     0.000005s :      3: substitution.graph_param_transform
86.17% :     0.000266s :      6: substitution.inline
 1.11% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.53% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.44% :     0.000004s :      2: substitution.replace_old_param
 2.69% :     0.000008s :      1: substitution.switch_simplify
 4.22% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006694     2
84.30% :     0.005643s :      1: type_inference.infer
15.70% :     0.001051s :      1: type_inference.specialize
------[replace.]   0.000140     9
47.73% :     0.000067s :      6: replace.inline
30.29% :     0.000043s :      1: replace.switch_simplify
21.97% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000282     9
93.13% :     0.000262s :      6: match.inline
 2.65% :     0.000007s :      1: match.switch_simplify
 4.22% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   801
 1.32% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.75% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     12: predicate.addn_check_dump
 1.55% :     0.000002s :     12: predicate.addn_zero_filter
 1.98% :     0.000003s :     12: predicate.arithmetic_simplify
 1.36% :     0.000002s :     12: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     12: predicate.compare_switch_simplify
 1.37% :     0.000002s :     12: predicate.depend_value_elim
 1.10% :     0.000001s :     12: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     12: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.35% :     0.000002s :     12: predicate.environ_get_eliminate
 1.12% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.81% :     0.000008s :     26: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.07% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     17: predicate.load_eliminater
 1.24% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.30% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000002s :     12: predicate.merge_addn
 1.12% :     0.000001s :     12: predicate.minmaximum_grad
 1.70% :     0.000002s :      3: predicate.mutable_eliminate
 0.62% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     17: predicate.partial_eliminate
 1.45% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     12: predicate.reduce_eliminate
 1.51% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     20: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.32% :     0.000002s :     12: predicate.reshape_eliminate
 1.24% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.58% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.54% :     0.000002s :     12: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.70% :     0.000004s :     20: predicate.switch_defer_inline
 2.29% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.19% :     0.000011s :     65: predicate.switch_simplify
 1.28% :     0.000002s :     12: predicate.tile_eliminate
 1.24% :     0.000002s :     12: predicate.transpose_eliminate
 1.44% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.85% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     12: predicate.value_based_eliminate
 0.40% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000845    15
50.31% :     0.000425s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.45% :     0.000114s :      1: func_graph_cloner_run.FuncGraphClonerNode
36.24% :     0.000306s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017401    72
 0.31% :     0.000053s :      1: add_recomputation
 0.49% :     0.000085s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 1.78% :     0.000309s :      1: bootstrap
 0.15% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.15% :     0.000027s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
34.23% :     0.005956s :      1: jit_opt_a
 0.88% :     0.000153s :      1: jit_opt_after_cconv
 0.30% :     0.000052s :      1: jit_opt_b
 2.52% :     0.000438s :      1: loop_unroll
 3.37% :     0.000586s :      1: mutable_eliminate
 6.50% :     0.001131s :     26: opt.transform.jit_opt_a
 0.29% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.88% :     0.000501s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000052s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 2.50% :     0.000435s :      1: renormalize.infer
 1.80% :     0.000313s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000039s :      1: rewriter_after_opt_a
 0.53% :     0.000092s :      1: rewriter_before_opt_a
 0.47% :     0.000081s :      1: symbol_engine_optimizer
38.98% :     0.006783s :      1: type_inference


TotalTime = 0.0455486, [33]
[bootstrap]: 0.00042262
[type_inference]: 0.00622272
[event_method]: 1.934e-05
[auto_monad]: 8.719e-05
[graph_reusing]: 7.09001e-06
[pre_auto_parallel]: 2.48002e-06
[py_interpret_to_execute]: 4.943e-05
[rewriter_before_opt_a]: 7.63e-05
[expand_dump_flag]: 3.23e-06
[jit_opt_a]: 0.00675912, [2]
    [Cycle 1]: 0.00206483, [27]
        [switch_simplify]: 7.664e-05
        [loop_unroll]: 3.33e-05
        [a_1]: 0.00062966
        [with_stream_mark]: 2.285e-05
        [recompute_prepare]: 7.33999e-06
        [updatestate_depend_eliminate]: 3.39001e-06
        [updatestate_assign_eliminate]: 2.81999e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 6.07001e-06
        [updatestate_useless_node_eliminater]: 5.23002e-06
        [accelerated_algorithm]: 6.16e-06
        [meta_shard_fg_expand]: 2.57001e-06
        [get_grad_eliminate_]: 5.87999e-06
        [merge_forward]: 4.12e-06
        [cell_reuse_recompute_pass]: 1.55001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.554e-05
        [j_node_and_user_rematch]: 9.78002e-06
        [meta_fg_expand]: 2.58003e-06
        [replace_old_param]: 1.011e-05
        [inline_without_move]: 5.19e-06
        [renormalize]: 0.00095528
        [add_forward_monad_depend]: 5.57999e-06
        [auto_monad_grad]: 1.74e-06
        [auto_monad_eliminator]: 1.196e-05
        [cse]: 2.242e-05
        [replace_applicator]: 1.37e-05
    [Cycle 2]: 0.00033086, [27]
        [switch_simplify]: 5.79e-06
        [loop_unroll]: 5.00999e-06
        [a_1]: 9.353e-05
        [with_stream_mark]: 1.092e-05
        [recompute_prepare]: 5.17e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 5.34e-06
        [updatestate_useless_node_eliminater]: 4.83001e-06
        [accelerated_algorithm]: 5.14e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 4.52998e-06
        [merge_forward]: 2.32001e-06
        [cell_reuse_recompute_pass]: 2.68e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.355e-05
        [j_node_and_user_rematch]: 8.08999e-06
        [meta_fg_expand]: 2.26998e-06
        [replace_old_param]: 8.09002e-06
        [inline_without_move]: 4.82e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.72999e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 5.35001e-06
        [cse]: 9.72001e-06
        [replace_applicator]: 5.15999e-06
[py_interpret_to_execute_after_opt_a]: 1.059e-05
[rewriter_after_opt_a]: 3.285e-05
[convert_after_rewriter]: 5.67999e-06
[order_py_execute_after_rewriter]: 4.60999e-06
[mutable_eliminate]: 0.00072279
[jit_opt_b]: 5.528e-05, [1]
    [Cycle 1]: 4.744e-05, [2]
        [frontend_op_eliminate]: 1.94e-05
        [inline_after_opt_a]: 1.544e-05
[cconv]: 2.023e-05
[loop_unroll]: 0.0211077
[jit_opt_after_cconv]: 0.00023303, [1]
    [Cycle 1]: 0.00021883, [11]
        [c_1]: 3.348e-05
        [parameter_eliminate]: 9.72001e-06
        [updatestate_depend_eliminate]: 1.201e-05
        [updatestate_assign_eliminate]: 3.74002e-06
        [updatestate_loads_eliminate]: 3.53e-06
        [cse]: 4.605e-05
        [call_graph_tuple_transform]: 3.401e-05
        [tuple_list_get_item_eliminator]: 6.94999e-06
        [none_parameter_eliminate]: 1.65001e-06
        [renormalize]: 9.5999e-07
        [switch_simplify]: 6.98e-06
[remove_dup_value]: 1.83e-05
[partial_unused_args_eliminate]: 2.94001e-06
[environ_conv]: 8.69e-06
[add_recomputation]: 6.682e-05
[cse_after_recomputation]: 2.764e-05, [1]
    [Cycle 1]: 2.072e-05, [1]
        [cse]: 1.238e-05
[auto_monad_reorder]: 1.786e-05
[get_jit_bprop_graph]: 2.97002e-06
[rewriter_after_jit_bprop_graph]: 1.473e-05
[opt_after_jit_grad]: 0.00085399
[symbol_engine_optimizer]: 0.00010002, [1]
    [Cycle 1]: 8.871e-05, [6]
        [build]: 7.75e-06
        [elim_shapecalc]: 1.034e-05
        [elim_not_effective]: 1.974e-05
        [opt_reshape]: 5.89999e-06
        [fold_const_symbol]: 9.84001e-06
        [renormalize]: 1.49998e-06
[validate]: 5.656e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.00822965
[execute]: 9.77999e-06
Sums
    bootstrap                            :     0.000423s :  1.05%
    type_inference                       :     0.006223s : 15.40%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000087s :  0.22%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000049s :  0.12%
    rewriter_before_opt_a                :     0.000076s :  0.19%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000082s :  0.20%
    jit_opt_a.loop_unroll                :     0.000038s :  0.09%
    jit_opt_a.a_1                        :     0.000723s :  1.79%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.07%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000018s :  0.05%
    jit_opt_a.inline_without_move        :     0.000010s :  0.02%
    jit_opt_a.renormalize                :     0.000955s :  2.36%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.04%
    jit_opt_a.cse                        :     0.000032s :  0.08%
    jit_opt_a.replace_applicator         :     0.000019s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.03%
    rewriter_after_opt_a                 :     0.000033s :  0.08%
    convert_after_rewriter               :     0.000006s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000723s :  1.79%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.04%
    cconv                                :     0.000020s :  0.05%
    loop_unroll                          :     0.021108s : 52.22%
    jit_opt_after_cconv.c_1              :     0.000033s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000010s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000046s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000034s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000018s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000009s :  0.02%
    add_recomputation                    :     0.000067s :  0.17%
    cse_after_recomputation.cse          :     0.000012s :  0.03%
    auto_monad_reorder                   :     0.000018s :  0.04%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000015s :  0.04%
    opt_after_jit_grad                   :     0.000854s :  2.11%
    symbol_engine_optimizer.build        :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000057s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008230s : 20.36%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000266    23
 1.12% :     0.000003s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 2.94% :     0.000008s :      3: substitution.graph_param_transform
84.64% :     0.000225s :      4: substitution.inline
 1.43% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.92% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.04% :     0.000005s :      2: substitution.replace_old_param
 5.46% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006143     2
81.66% :     0.005016s :      1: type_inference.infer
18.34% :     0.001127s :      1: type_inference.specialize
------[replace.]   0.000087     6
67.96% :     0.000059s :      4: replace.inline
32.04% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000235     6
94.46% :     0.000222s :      4: match.inline
 5.54% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   691
 1.22% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.04% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.11% :     0.000001s :     10: predicate.addn_zero_filter
 2.42% :     0.000003s :     10: predicate.arithmetic_simplify
 1.13% :     0.000002s :     10: predicate.cast_eliminate
 0.35% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.92% :     0.000001s :     10: predicate.compare_switch_simplify
 1.73% :     0.000002s :     10: predicate.depend_value_elim
 0.97% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.61% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.66% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.72% :     0.000001s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000001s :     10: predicate.environ_get_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.19% :     0.000002s :      6: predicate.get_grad_eliminate
 0.34% :     0.000000s :      3: predicate.graph_param_transform
 5.94% :     0.000008s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000002s :      6: predicate.less_batch_normalization
 1.29% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 4.53% :     0.000006s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.24% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.94% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 1.45% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 1.90% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     10: predicate.reduce_eliminate
 1.61% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.72% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.50% :     0.000002s :     18: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 0.99% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.29% :     0.000002s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.49% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.25% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.53% :     0.000010s :     53: predicate.switch_simplify
 1.25% :     0.000002s :     10: predicate.tile_eliminate
 0.98% :     0.000001s :     10: predicate.transpose_eliminate
 1.13% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.52% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.42% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.65% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000832    12
48.32% :     0.000402s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.18% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.50% :     0.000387s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.047633    76
 0.15% :     0.000073s :      1: add_recomputation
 0.19% :     0.000091s :      1: auto_monad
 0.04% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.93% :     0.000442s :      1: bootstrap
 0.05% :     0.000023s :      1: cconv
 0.02% :     0.000008s :      1: convert_after_rewriter
 0.06% :     0.000030s :      1: cse_after_recomputation
 0.03% :     0.000014s :      1: environ_conv
 0.05% :     0.000023s :      1: event_method
 0.03% :     0.000016s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
14.20% :     0.006763s :      1: jit_opt_a
 0.50% :     0.000238s :      1: jit_opt_after_cconv
 0.12% :     0.000058s :      1: jit_opt_b
44.38% :     0.021139s :      1: loop_unroll
 1.54% :     0.000733s :      1: mutable_eliminate
 2.02% :     0.000962s :     26: opt.transform.jit_opt_a
 0.16% :     0.000076s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000028s :      4: opt.transform.jit_opt_b
 0.09% :     0.000044s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000035s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000041s :      4: opt.transform.symbol_engine_opt
 1.82% :     0.000868s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000052s :      1: py_interpret_to_execute
 0.03% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000022s :      1: remove_dup_value
 1.09% :     0.000521s :      1: renormalize.infer
 0.89% :     0.000425s :      1: renormalize.specialize
 0.04% :     0.000017s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000036s :      1: rewriter_after_opt_a
 0.17% :     0.000080s :      1: rewriter_before_opt_a
 0.22% :     0.000104s :      1: symbol_engine_optimizer
17.33% :     0.008256s :      1: task_emit
13.10% :     0.006239s :      1: type_inference
 0.18% :     0.000087s :      1: validate


TotalTime = 0.0530181, [33]
[bootstrap]: 0.00034984
[type_inference]: 0.0193126
[event_method]: 2.43e-05
[auto_monad]: 7.657e-05
[graph_reusing]: 6.17999e-06
[pre_auto_parallel]: 3.07002e-06
[py_interpret_to_execute]: 5.614e-05
[rewriter_before_opt_a]: 8.04e-05
[expand_dump_flag]: 3.70003e-06
[jit_opt_a]: 0.02344, [2]
    [Cycle 1]: 0.00218727, [27]
        [switch_simplify]: 7.217e-05
        [loop_unroll]: 3.038e-05
        [a_1]: 0.00069826
        [with_stream_mark]: 2.619e-05
        [recompute_prepare]: 8.26002e-06
        [updatestate_depend_eliminate]: 4.53001e-06
        [updatestate_assign_eliminate]: 3.6e-06
        [updatestate_loads_eliminate]: 2.80997e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 6.73e-06
        [updatestate_useless_node_eliminater]: 5.41002e-06
        [accelerated_algorithm]: 6.51999e-06
        [meta_shard_fg_expand]: 2.58e-06
        [get_grad_eliminate_]: 6.11998e-06
        [merge_forward]: 3.58999e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.577e-05
        [j_node_and_user_rematch]: 1.033e-05
        [meta_fg_expand]: 2.48002e-06
        [replace_old_param]: 1.045e-05
        [inline_without_move]: 6.14001e-06
        [renormalize]: 0.00100136
        [add_forward_monad_depend]: 6.51e-06
        [auto_monad_grad]: 3.49001e-06
        [auto_monad_eliminator]: 1.623e-05
        [cse]: 3.14e-05
        [replace_applicator]: 1.275e-05
    [Cycle 2]: 0.00038491, [27]
        [switch_simplify]: 6.17001e-06
        [loop_unroll]: 4.85999e-06
        [a_1]: 9.271e-05
        [with_stream_mark]: 9.80002e-06
        [recompute_prepare]: 5.26002e-06
        [updatestate_depend_eliminate]: 2.43002e-06
        [updatestate_assign_eliminate]: 2.00002e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 5.16998e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 5.00999e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 4.90999e-06
        [merge_forward]: 2.33002e-06
        [cell_reuse_recompute_pass]: 2.36998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.246e-05
        [j_node_and_user_rematch]: 8.28999e-06
        [meta_fg_expand]: 1.49998e-06
        [replace_old_param]: 8.28999e-06
        [inline_without_move]: 4.94e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.16002e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 6.43e-06
        [cse]: 1.201e-05
        [replace_applicator]: 5.73002e-06
[py_interpret_to_execute_after_opt_a]: 1.36e-05
[rewriter_after_opt_a]: 4.108e-05
[convert_after_rewriter]: 7.82002e-06
[order_py_execute_after_rewriter]: 4.92e-06
[mutable_eliminate]: 0.00073615
[jit_opt_b]: 5.276e-05, [1]
    [Cycle 1]: 4.494e-05, [2]
        [frontend_op_eliminate]: 1.778e-05
        [inline_after_opt_a]: 1.504e-05
[cconv]: 2.861e-05
[loop_unroll]: 0.00046216
[jit_opt_after_cconv]: 0.00015444, [1]
    [Cycle 1]: 0.00014767, [11]
        [c_1]: 2.251e-05
        [parameter_eliminate]: 3.61999e-06
        [updatestate_depend_eliminate]: 8.57998e-06
        [updatestate_assign_eliminate]: 2.64001e-06
        [updatestate_loads_eliminate]: 2.21998e-06
        [cse]: 2.21e-05
        [call_graph_tuple_transform]: 2.127e-05
        [tuple_list_get_item_eliminator]: 5.39e-06
        [none_parameter_eliminate]: 1.79998e-06
        [renormalize]: 7.39994e-07
        [switch_simplify]: 5.62999e-06
[remove_dup_value]: 1.544e-05
[partial_unused_args_eliminate]: 2.94999e-06
[environ_conv]: 5.28002e-06
[add_recomputation]: 5.734e-05
[cse_after_recomputation]: 2.303e-05, [1]
    [Cycle 1]: 1.684e-05, [1]
        [cse]: 1.043e-05
[auto_monad_reorder]: 1.719e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 7.75e-06
[opt_after_jit_grad]: 0.00053783
[symbol_engine_optimizer]: 7.902e-05, [1]
    [Cycle 1]: 7.195e-05, [6]
        [build]: 4.65001e-06
        [elim_shapecalc]: 7.85e-06
        [elim_not_effective]: 1.44e-05
        [opt_reshape]: 6.76999e-06
        [fold_const_symbol]: 9.50001e-06
        [renormalize]: 5.19998e-07
[validate]: 4.538e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.0071072
[execute]: 9.69e-06
Sums
    bootstrap                            :     0.000350s :  1.11%
    type_inference                       :     0.019313s : 61.48%
    event_method                         :     0.000024s :  0.08%
    auto_monad                           :     0.000077s :  0.24%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000056s :  0.18%
    rewriter_before_opt_a                :     0.000080s :  0.26%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000078s :  0.25%
    jit_opt_a.loop_unroll                :     0.000035s :  0.11%
    jit_opt_a.a_1                        :     0.000791s :  2.52%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.04%
    jit_opt_a.merge_forward              :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000019s :  0.06%
    jit_opt_a.inline_without_move        :     0.000011s :  0.04%
    jit_opt_a.renormalize                :     0.001001s :  3.19%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.07%
    jit_opt_a.cse                        :     0.000043s :  0.14%
    jit_opt_a.replace_applicator         :     0.000018s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.04%
    rewriter_after_opt_a                 :     0.000041s :  0.13%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000736s :  2.34%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.05%
    cconv                                :     0.000029s :  0.09%
    loop_unroll                          :     0.000462s :  1.47%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000015s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000057s :  0.18%
    cse_after_recomputation.cse          :     0.000010s :  0.03%
    auto_monad_reorder                   :     0.000017s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000538s :  1.71%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007107s : 22.63%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.000316    23
 0.73% :     0.000002s :      2: substitution.elim_not_effective
 0.42% :     0.000001s :      2: substitution.fold_const_symbol
 1.99% :     0.000006s :      3: substitution.graph_param_transform
87.48% :     0.000276s :      4: substitution.inline
 1.37% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.58% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.78% :     0.000006s :      2: substitution.replace_old_param
 4.64% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.019203     2
92.06% :     0.017677s :      1: type_inference.infer
 7.94% :     0.001525s :      1: type_inference.specialize
------[replace.]   0.000084     6
69.37% :     0.000059s :      4: replace.inline
30.63% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000288     6
95.31% :     0.000274s :      4: match.inline
 4.69% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.29% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.92% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 2.57% :     0.000003s :     10: predicate.arithmetic_simplify
 1.53% :     0.000002s :     10: predicate.cast_eliminate
 0.63% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.58% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.45% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.15% :     0.000007s :     22: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.66% :     0.000002s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.21% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 2.12% :     0.000003s :      3: predicate.mutable_eliminate
 0.61% :     0.000001s :      3: predicate.opt_reshape
 1.99% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.41% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.40% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.53% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.82% :     0.000002s :     10: predicate.reshape_eliminate
 1.05% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.18% :     0.000002s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.52% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.22% :     0.000003s :     16: predicate.switch_defer_inline
 1.80% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.46% :     0.000010s :     53: predicate.switch_simplify
 1.48% :     0.000002s :     10: predicate.tile_eliminate
 1.28% :     0.000002s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.95% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.46% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.09% :     0.000003s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000912    12
39.08% :     0.000356s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.86% :     0.000063s :      1: func_graph_cloner_run.FuncGraphClonerNode
54.06% :     0.000493s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.055147    76
 0.11% :     0.000061s :      1: add_recomputation
 0.15% :     0.000081s :      1: auto_monad
 0.04% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.67% :     0.000372s :      1: bootstrap
 0.06% :     0.000031s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.05% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.05% :     0.000030s :      1: event_method
 0.03% :     0.000015s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
42.51% :     0.023444s :      1: jit_opt_a
 0.29% :     0.000157s :      1: jit_opt_after_cconv
 0.10% :     0.000055s :      1: jit_opt_b
 0.85% :     0.000471s :      1: loop_unroll
 1.35% :     0.000746s :      1: mutable_eliminate
 1.86% :     0.001028s :     26: opt.transform.jit_opt_a
 0.09% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000026s :      4: opt.transform.jit_opt_b
 0.03% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000035s :      4: opt.transform.symbol_engine_opt
 0.99% :     0.000547s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000059s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000018s :      1: remove_dup_value
 1.06% :     0.000586s :      1: renormalize.infer
 0.74% :     0.000407s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000044s :      1: rewriter_after_opt_a
 0.15% :     0.000083s :      1: rewriter_before_opt_a
 0.15% :     0.000082s :      1: symbol_engine_optimizer
12.92% :     0.007128s :      1: task_emit
35.07% :     0.019339s :      1: type_inference
 0.13% :     0.000070s :      1: validate


TotalTime = 0.0175961, [30]
[bootstrap]: 0.00036159
[type_inference]: 0.0074
[event_method]: 2.438e-05
[auto_monad]: 8.951e-05
[graph_reusing]: 6.97002e-06
[pre_auto_parallel]: 3.58e-06
[py_interpret_to_execute]: 7.058e-05
[rewriter_before_opt_a]: 0.00012589
[expand_dump_flag]: 3.77002e-06
[jit_opt_a]: 0.00688458, [2]
    [Cycle 1]: 0.00230145, [27]
        [switch_simplify]: 0.00015247
        [loop_unroll]: 3.388e-05
        [a_1]: 0.0007875
        [with_stream_mark]: 2.583e-05
        [recompute_prepare]: 1.058e-05
        [updatestate_depend_eliminate]: 3.98001e-06
        [updatestate_assign_eliminate]: 3.71001e-06
        [updatestate_loads_eliminate]: 2.96999e-06
        [parameter_eliminate]: 1.95001e-06
        [specialize_transform]: 7.51001e-06
        [updatestate_useless_node_eliminater]: 5.91e-06
        [accelerated_algorithm]: 6.64001e-06
        [meta_shard_fg_expand]: 2.17001e-06
        [get_grad_eliminate_]: 5.64e-06
        [merge_forward]: 4.42998e-06
        [cell_reuse_recompute_pass]: 1.08001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.885e-05
        [j_node_and_user_rematch]: 1.115e-05
        [meta_fg_expand]: 2.88998e-06
        [replace_old_param]: 1.103e-05
        [inline_without_move]: 6.19001e-06
        [renormalize]: 0.00088626
        [add_forward_monad_depend]: 8.64e-06
        [auto_monad_grad]: 2.64999e-06
        [auto_monad_eliminator]: 1.769e-05
        [cse]: 3.363e-05
        [replace_applicator]: 1.723e-05
    [Cycle 2]: 0.0003551, [27]
        [switch_simplify]: 6.09999e-06
        [loop_unroll]: 4.89e-06
        [a_1]: 9.133e-05
        [with_stream_mark]: 1.555e-05
        [recompute_prepare]: 6.57002e-06
        [updatestate_depend_eliminate]: 3.8e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 5.28002e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 5.03002e-06
        [merge_forward]: 3.37002e-06
        [cell_reuse_recompute_pass]: 2.15002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.698e-05
        [j_node_and_user_rematch]: 8.59e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 8.67e-06
        [inline_without_move]: 4.90001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 2.04e-06
        [auto_monad_grad]: 1.11997e-06
        [auto_monad_eliminator]: 6.22001e-06
        [cse]: 1.325e-05
        [replace_applicator]: 6.20002e-06
[py_interpret_to_execute_after_opt_a]: 1.449e-05
[rewriter_after_opt_a]: 4.294e-05
[convert_after_rewriter]: 7.56001e-06
[order_py_execute_after_rewriter]: 5.34e-06
[mutable_eliminate]: 0.00082814
[jit_opt_b]: 5.606e-05, [1]
    [Cycle 1]: 4.63e-05, [2]
        [frontend_op_eliminate]: 1.748e-05
        [inline_after_opt_a]: 1.628e-05
[cconv]: 3.169e-05
[loop_unroll]: 0.00051536
[jit_opt_after_cconv]: 0.0001566, [1]
    [Cycle 1]: 0.00014852, [11]
        [c_1]: 2.259e-05
        [parameter_eliminate]: 4.3e-06
        [updatestate_depend_eliminate]: 7.08998e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [cse]: 2.568e-05
        [call_graph_tuple_transform]: 2.084e-05
        [tuple_list_get_item_eliminator]: 5.10001e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 9.70002e-07
        [switch_simplify]: 5.21998e-06
[remove_dup_value]: 1.386e-05
[partial_unused_args_eliminate]: 2.12001e-06
[environ_conv]: 5.67001e-06
[add_recomputation]: 5.162e-05
[cse_after_recomputation]: 2.196e-05, [1]
    [Cycle 1]: 1.627e-05, [1]
        [cse]: 1.016e-05
[auto_monad_reorder]: 1.496e-05
[get_jit_bprop_graph]: 2.01998e-06
[rewriter_after_jit_bprop_graph]: 7.31999e-06
[opt_after_jit_grad]: 0.00046513
[symbol_engine_optimizer]: 7.323e-05, [1]
    [Cycle 1]: 6.711e-05, [6]
        [build]: 4.10998e-06
        [elim_shapecalc]: 7.08e-06
        [elim_not_effective]: 1.25e-05
        [opt_reshape]: 6.03002e-06
        [fold_const_symbol]: 8.90999e-06
        [renormalize]: 3.20026e-07
[validate]: 4.012e-05
Sums
    bootstrap                            :     0.000362s :  2.87%
    type_inference                       :     0.007400s : 58.64%
    event_method                         :     0.000024s :  0.19%
    auto_monad                           :     0.000090s :  0.71%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000004s :  0.03%
    py_interpret_to_execute              :     0.000071s :  0.56%
    rewriter_before_opt_a                :     0.000126s :  1.00%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000159s :  1.26%
    jit_opt_a.loop_unroll                :     0.000039s :  0.31%
    jit_opt_a.a_1                        :     0.000879s :  6.96%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.08%
    jit_opt_a.merge_forward              :     0.000008s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000020s :  0.16%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.000886s :  7.02%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.19%
    jit_opt_a.cse                        :     0.000047s :  0.37%
    jit_opt_a.replace_applicator         :     0.000023s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.11%
    rewriter_after_opt_a                 :     0.000043s :  0.34%
    convert_after_rewriter               :     0.000008s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000828s :  6.56%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.13%
    cconv                                :     0.000032s :  0.25%
    loop_unroll                          :     0.000515s :  4.08%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000026s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000052s :  0.41%
    cse_after_recomputation.cse          :     0.000010s :  0.08%
    auto_monad_reorder                   :     0.000015s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000465s :  3.69%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000040s :  0.32%


Time group info:
------[substitution.]   0.000344    26
 0.54% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000002s :      2: substitution.fold_const_symbol
 1.62% :     0.000006s :      3: substitution.graph_param_transform
85.91% :     0.000295s :      6: substitution.inline
 1.12% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.69% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.66% :     0.000006s :      2: substitution.replace_old_param
 2.94% :     0.000010s :      1: substitution.switch_simplify
 4.07% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007315     2
83.63% :     0.006117s :      1: type_inference.infer
16.37% :     0.001198s :      1: type_inference.specialize
------[replace.]   0.000165     9
46.80% :     0.000077s :      6: replace.inline
29.59% :     0.000049s :      1: replace.switch_simplify
23.61% :     0.000039s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000314     9
92.92% :     0.000291s :      6: match.inline
 2.98% :     0.000009s :      1: match.switch_simplify
 4.09% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000143   801
 1.47% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.02% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000002s :     12: predicate.addn_check_dump
 1.27% :     0.000002s :     12: predicate.addn_zero_filter
 1.85% :     0.000003s :     12: predicate.arithmetic_simplify
 1.17% :     0.000002s :     12: predicate.cast_eliminate
 0.36% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000002s :     12: predicate.compare_switch_simplify
 2.21% :     0.000003s :     12: predicate.depend_value_elim
 1.10% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.14% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.71% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.09% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.06% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     12: predicate.environ_get_eliminate
 1.08% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.94% :     0.000008s :     26: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.52% :     0.000002s :      6: predicate.less_batch_normalization
 1.67% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     17: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.85% :     0.000005s :     40: predicate.loop_unroll_before_grad
 2.01% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     12: predicate.merge_addn
 1.10% :     0.000002s :     12: predicate.minmaximum_grad
 2.25% :     0.000003s :      3: predicate.mutable_eliminate
 0.39% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     17: predicate.partial_eliminate
 1.15% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.81% :     0.000003s :     12: predicate.reduce_eliminate
 1.36% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     20: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.38% :     0.000001s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     12: predicate.reshape_eliminate
 1.30% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.43% :     0.000002s :     12: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     20: predicate.switch_defer_inline
 2.13% :     0.000003s :     20: predicate.switch_layer_defer_inline
 9.05% :     0.000013s :     65: predicate.switch_simplify
 1.22% :     0.000002s :     12: predicate.tile_eliminate
 1.14% :     0.000002s :     12: predicate.transpose_eliminate
 1.35% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.22% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     12: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001040    15
54.39% :     0.000565s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.82% :     0.000123s :      1: func_graph_cloner_run.FuncGraphClonerNode
33.79% :     0.000351s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019755    72
 0.28% :     0.000054s :      1: add_recomputation
 0.47% :     0.000093s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 1.94% :     0.000383s :      1: bootstrap
 0.18% :     0.000035s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.15% :     0.000030s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
34.88% :     0.006890s :      1: jit_opt_a
 0.81% :     0.000159s :      1: jit_opt_after_cconv
 0.30% :     0.000059s :      1: jit_opt_b
 2.67% :     0.000527s :      1: loop_unroll
 4.27% :     0.000843s :      1: mutable_eliminate
 6.14% :     0.001213s :     26: opt.transform.jit_opt_a
 0.25% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000027s :      4: opt.transform.jit_opt_b
 0.08% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.14% :     0.000028s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.40% :     0.000474s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000006s :      1: pre_auto_parallel
 0.38% :     0.000075s :      1: py_interpret_to_execute
 0.09% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 2.55% :     0.000504s :      1: renormalize.infer
 1.88% :     0.000371s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000048s :      1: rewriter_after_opt_a
 0.66% :     0.000131s :      1: rewriter_before_opt_a
 0.38% :     0.000076s :      1: symbol_engine_optimizer
37.86% :     0.007478s :      1: type_inference


TotalTime = 0.0253855, [33]
[bootstrap]: 0.00050828
[type_inference]: 0.00595094
[event_method]: 6.838e-05
[auto_monad]: 7.326e-05
[graph_reusing]: 7.43999e-06
[pre_auto_parallel]: 2.59001e-06
[py_interpret_to_execute]: 4.546e-05
[rewriter_before_opt_a]: 7.81e-05
[expand_dump_flag]: 3.91001e-06
[jit_opt_a]: 0.00726869, [2]
    [Cycle 1]: 0.00189435, [27]
        [switch_simplify]: 6.964e-05
        [loop_unroll]: 3.105e-05
        [a_1]: 0.00059897
        [with_stream_mark]: 2.003e-05
        [recompute_prepare]: 7.9e-06
        [updatestate_depend_eliminate]: 3.45e-06
        [updatestate_assign_eliminate]: 2.86999e-06
        [updatestate_loads_eliminate]: 3.08e-06
        [parameter_eliminate]: 2.09e-06
        [specialize_transform]: 5.77001e-06
        [updatestate_useless_node_eliminater]: 5.12999e-06
        [accelerated_algorithm]: 6.39001e-06
        [meta_shard_fg_expand]: 2.12999e-06
        [get_grad_eliminate_]: 5.29998e-06
        [merge_forward]: 3.83999e-06
        [cell_reuse_recompute_pass]: 1.10999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.492e-05
        [j_node_and_user_rematch]: 9.29998e-06
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 9.51998e-06
        [inline_without_move]: 6.27001e-06
        [renormalize]: 0.00083732
        [add_forward_monad_depend]: 6.53e-06
        [auto_monad_grad]: 2.37999e-06
        [auto_monad_eliminator]: 1.499e-05
        [cse]: 2.95e-05
        [replace_applicator]: 1.289e-05
    [Cycle 2]: 0.00032544, [27]
        [switch_simplify]: 5.44e-06
        [loop_unroll]: 5.46002e-06
        [a_1]: 9.083e-05
        [with_stream_mark]: 1.055e-05
        [recompute_prepare]: 5.07999e-06
        [updatestate_depend_eliminate]: 2.89001e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [parameter_eliminate]: 9.49978e-07
        [specialize_transform]: 4.88001e-06
        [updatestate_useless_node_eliminater]: 4.83001e-06
        [accelerated_algorithm]: 5.77999e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 4.50999e-06
        [merge_forward]: 2.48998e-06
        [cell_reuse_recompute_pass]: 1.89999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.536e-05
        [j_node_and_user_rematch]: 7.61001e-06
        [meta_fg_expand]: 1.74998e-06
        [replace_old_param]: 7.57002e-06
        [inline_without_move]: 4.50001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 6.49976e-07
        [auto_monad_eliminator]: 4.76002e-06
        [cse]: 1.044e-05
        [replace_applicator]: 5.30001e-06
[py_interpret_to_execute_after_opt_a]: 1.302e-05
[rewriter_after_opt_a]: 3.718e-05
[convert_after_rewriter]: 6.89999e-06
[order_py_execute_after_rewriter]: 5.37001e-06
[mutable_eliminate]: 0.00080645
[jit_opt_b]: 5.416e-05, [1]
    [Cycle 1]: 4.714e-05, [2]
        [frontend_op_eliminate]: 1.826e-05
        [inline_after_opt_a]: 1.641e-05
[cconv]: 2.931e-05
[loop_unroll]: 0.0004484
[jit_opt_after_cconv]: 0.00014771, [1]
    [Cycle 1]: 0.00014092, [11]
        [c_1]: 2.1e-05
        [parameter_eliminate]: 3.64002e-06
        [updatestate_depend_eliminate]: 7.57998e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 2.226e-05
        [call_graph_tuple_transform]: 1.984e-05
        [tuple_list_get_item_eliminator]: 5.00001e-06
        [none_parameter_eliminate]: 1.58002e-06
        [renormalize]: 8.50006e-07
        [switch_simplify]: 5.22e-06
[remove_dup_value]: 1.473e-05
[partial_unused_args_eliminate]: 2.19001e-06
[environ_conv]: 6.12001e-06
[add_recomputation]: 5.09e-05
[cse_after_recomputation]: 2.013e-05, [1]
    [Cycle 1]: 1.491e-05, [1]
        [cse]: 9.04998e-06
[auto_monad_reorder]: 1.662e-05
[get_jit_bprop_graph]: 1.99e-06
[rewriter_after_jit_bprop_graph]: 7.45e-06
[opt_after_jit_grad]: 0.00046961
[symbol_engine_optimizer]: 7.438e-05, [1]
    [Cycle 1]: 6.76e-05, [6]
        [build]: 4.08001e-06
        [elim_shapecalc]: 7.63001e-06
        [elim_not_effective]: 1.375e-05
        [opt_reshape]: 5.72001e-06
        [fold_const_symbol]: 8.94e-06
        [renormalize]: 7.89994e-07
[validate]: 4.284e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.00886899
[execute]: 1.011e-05
Sums
    bootstrap                            :     0.000508s :  2.58%
    type_inference                       :     0.005951s : 30.24%
    event_method                         :     0.000068s :  0.35%
    auto_monad                           :     0.000073s :  0.37%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000045s :  0.23%
    rewriter_before_opt_a                :     0.000078s :  0.40%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000075s :  0.38%
    jit_opt_a.loop_unroll                :     0.000037s :  0.19%
    jit_opt_a.a_1                        :     0.000690s :  3.51%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.000837s :  4.26%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.10%
    jit_opt_a.cse                        :     0.000040s :  0.20%
    jit_opt_a.replace_applicator         :     0.000018s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.19%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000806s :  4.10%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.000029s :  0.15%
    loop_unroll                          :     0.000448s :  2.28%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000051s :  0.26%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000470s :  2.39%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.22%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008869s : 45.07%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000257    23
 0.95% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000002s :      2: substitution.fold_const_symbol
 2.32% :     0.000006s :      3: substitution.graph_param_transform
85.24% :     0.000219s :      4: substitution.inline
 1.29% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.98% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.82% :     0.000005s :      2: substitution.replace_old_param
 5.76% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005859     2
83.37% :     0.004884s :      1: type_inference.infer
16.63% :     0.000975s :      1: type_inference.specialize
------[replace.]   0.000078     6
66.32% :     0.000052s :      4: replace.inline
33.68% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000230     6
94.08% :     0.000216s :      4: match.inline
 5.92% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.62% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 2.18% :     0.000003s :     10: predicate.arithmetic_simplify
 1.40% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.59% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.09% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.58% :     0.000007s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.66% :     0.000002s :      6: predicate.less_batch_normalization
 1.36% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 1.32% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.83% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 2.58% :     0.000003s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.67% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.46% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.56% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.55% :     0.000003s :     16: predicate.switch_defer_inline
 1.97% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.85% :     0.000009s :     53: predicate.switch_simplify
 1.33% :     0.000002s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.28% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.76% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000749    12
52.42% :     0.000392s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.60% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.98% :     0.000307s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027236    76
 0.20% :     0.000054s :      1: add_recomputation
 0.28% :     0.000077s :      1: auto_monad
 0.07% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.97% :     0.000535s :      1: bootstrap
 0.12% :     0.000032s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.27% :     0.000074s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
26.70% :     0.007273s :      1: jit_opt_a
 0.55% :     0.000150s :      1: jit_opt_after_cconv
 0.21% :     0.000057s :      1: jit_opt_b
 1.68% :     0.000457s :      1: loop_unroll
 3.00% :     0.000816s :      1: mutable_eliminate
 3.37% :     0.000918s :     26: opt.transform.jit_opt_a
 0.17% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000027s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.75% :     0.000477s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000049s :      1: py_interpret_to_execute
 0.06% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 1.65% :     0.000450s :      1: renormalize.infer
 1.39% :     0.000379s :      1: renormalize.specialize
 0.03% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000040s :      1: rewriter_after_opt_a
 0.30% :     0.000081s :      1: rewriter_before_opt_a
 0.28% :     0.000077s :      1: symbol_engine_optimizer
32.64% :     0.008890s :      1: task_emit
21.92% :     0.005971s :      1: type_inference
 0.23% :     0.000063s :      1: validate


TotalTime = 0.0226473, [33]
[bootstrap]: 0.00040465
[type_inference]: 0.0053247
[event_method]: 1.873e-05
[auto_monad]: 0.00015588
[graph_reusing]: 6.16998e-06
[pre_auto_parallel]: 2.62001e-06
[py_interpret_to_execute]: 4.259e-05
[rewriter_before_opt_a]: 7.902e-05
[expand_dump_flag]: 3.2e-06
[jit_opt_a]: 0.00678819, [2]
    [Cycle 1]: 0.00222981, [27]
        [switch_simplify]: 7.475e-05
        [loop_unroll]: 3.082e-05
        [a_1]: 0.00066072
        [with_stream_mark]: 2.648e-05
        [recompute_prepare]: 7.94002e-06
        [updatestate_depend_eliminate]: 4.28001e-06
        [updatestate_assign_eliminate]: 3.09001e-06
        [updatestate_loads_eliminate]: 2.53998e-06
        [parameter_eliminate]: 2.29001e-06
        [specialize_transform]: 6.17001e-06
        [updatestate_useless_node_eliminater]: 5.70001e-06
        [accelerated_algorithm]: 6.29001e-06
        [meta_shard_fg_expand]: 2.29999e-06
        [get_grad_eliminate_]: 5.96e-06
        [merge_forward]: 3.98001e-06
        [cell_reuse_recompute_pass]: 1.71998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.654e-05
        [j_node_and_user_rematch]: 9.31e-06
        [meta_fg_expand]: 2.41e-06
        [replace_old_param]: 1.122e-05
        [inline_without_move]: 5.29998e-06
        [renormalize]: 0.00106793
        [add_forward_monad_depend]: 6.86001e-06
        [auto_monad_grad]: 2.66e-06
        [auto_monad_eliminator]: 1.609e-05
        [cse]: 3.475e-05
        [replace_applicator]: 1.413e-05
    [Cycle 2]: 0.00033815, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 4.97999e-06
        [a_1]: 9.449e-05
        [with_stream_mark]: 1.365e-05
        [recompute_prepare]: 4.87998e-06
        [updatestate_depend_eliminate]: 3.56999e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 1.86998e-06
        [parameter_eliminate]: 1.17999e-06
        [specialize_transform]: 4.82e-06
        [updatestate_useless_node_eliminater]: 4.68999e-06
        [accelerated_algorithm]: 5.96e-06
        [meta_shard_fg_expand]: 1.22999e-06
        [get_grad_eliminate_]: 4.73001e-06
        [merge_forward]: 3.06999e-06
        [cell_reuse_recompute_pass]: 2.61e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.322e-05
        [j_node_and_user_rematch]: 8.12998e-06
        [meta_fg_expand]: 2.13002e-06
        [replace_old_param]: 7.71001e-06
        [inline_without_move]: 4.54002e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.45999e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 5.10001e-06
        [cse]: 1.111e-05
        [replace_applicator]: 5.24e-06
[py_interpret_to_execute_after_opt_a]: 1.312e-05
[rewriter_after_opt_a]: 3.838e-05
[convert_after_rewriter]: 6.73e-06
[order_py_execute_after_rewriter]: 4.94e-06
[mutable_eliminate]: 0.00084124
[jit_opt_b]: 6.363e-05, [1]
    [Cycle 1]: 5.478e-05, [2]
        [frontend_op_eliminate]: 2.482e-05
        [inline_after_opt_a]: 1.668e-05
[cconv]: 2.96e-05
[loop_unroll]: 0.00048737
[jit_opt_after_cconv]: 0.00016278, [1]
    [Cycle 1]: 0.00015573, [11]
        [c_1]: 2.248e-05
        [parameter_eliminate]: 3.60998e-06
        [updatestate_depend_eliminate]: 9.09e-06
        [updatestate_assign_eliminate]: 2.59999e-06
        [updatestate_loads_eliminate]: 2.35002e-06
        [cse]: 2.745e-05
        [call_graph_tuple_transform]: 2.185e-05
        [tuple_list_get_item_eliminator]: 5.43002e-06
        [none_parameter_eliminate]: 1.37999e-06
        [renormalize]: 8.70001e-07
        [switch_simplify]: 5.62999e-06
[remove_dup_value]: 1.475e-05
[partial_unused_args_eliminate]: 2.37001e-06
[environ_conv]: 5.71e-06
[add_recomputation]: 5.684e-05
[cse_after_recomputation]: 2.147e-05, [1]
    [Cycle 1]: 1.588e-05, [1]
        [cse]: 9.82999e-06
[auto_monad_reorder]: 1.6e-05
[get_jit_bprop_graph]: 2.02999e-06
[rewriter_after_jit_bprop_graph]: 8.85001e-06
[opt_after_jit_grad]: 0.00051868
[symbol_engine_optimizer]: 8.658e-05, [1]
    [Cycle 1]: 7.882e-05, [6]
        [build]: 6.04001e-06
        [elim_shapecalc]: 8.42998e-06
        [elim_not_effective]: 1.584e-05
        [opt_reshape]: 6.68e-06
        [fold_const_symbol]: 9.91998e-06
        [renormalize]: 1.11002e-06
[validate]: 4.723e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00713553
[execute]: 8.79e-06
Sums
    bootstrap                            :     0.000405s :  2.28%
    type_inference                       :     0.005325s : 30.03%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000156s :  0.88%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.24%
    rewriter_before_opt_a                :     0.000079s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000080s :  0.45%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000755s :  4.26%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001068s :  6.02%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.12%
    jit_opt_a.cse                        :     0.000046s :  0.26%
    jit_opt_a.replace_applicator         :     0.000019s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000841s :  4.74%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000030s :  0.17%
    loop_unroll                          :     0.000487s :  2.75%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000057s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000519s :  2.93%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000047s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007136s : 40.25%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000301    23
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 2.04% :     0.000006s :      3: substitution.graph_param_transform
87.58% :     0.000263s :      4: substitution.inline
 1.05% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.43% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.98% :     0.000006s :      2: substitution.replace_old_param
 4.74% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005239     2
81.30% :     0.004259s :      1: type_inference.infer
18.70% :     0.000980s :      1: type_inference.specialize
------[replace.]   0.000087     6
68.08% :     0.000059s :      4: replace.inline
31.92% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000274     6
95.25% :     0.000261s :      4: match.inline
 4.75% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000002s :     10: predicate.addn_zero_filter
 2.69% :     0.000003s :     10: predicate.arithmetic_simplify
 1.30% :     0.000002s :     10: predicate.cast_eliminate
 0.61% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.45% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 6.05% :     0.000008s :     22: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.76% :     0.000002s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 1.64% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.93% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 2.07% :     0.000003s :      3: predicate.mutable_eliminate
 0.60% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.32% :     0.000002s :     10: predicate.reshape_eliminate
 1.09% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.05% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.25% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.79% :     0.000010s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000002s :     10: predicate.transpose_eliminate
 1.26% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.51% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.88% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.33% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.85% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.96% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000727    12
45.72% :     0.000332s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.27% :     0.000060s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.01% :     0.000334s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024818    76
 0.24% :     0.000060s :      1: add_recomputation
 0.64% :     0.000160s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.73% :     0.000429s :      1: bootstrap
 0.13% :     0.000033s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000023s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
27.37% :     0.006793s :      1: jit_opt_a
 0.67% :     0.000166s :      1: jit_opt_after_cconv
 0.27% :     0.000066s :      1: jit_opt_b
 2.00% :     0.000496s :      1: loop_unroll
 3.43% :     0.000852s :      1: mutable_eliminate
 4.00% :     0.000992s :     26: opt.transform.jit_opt_a
 0.21% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000034s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.13% :     0.000528s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000046s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.41% :     0.000599s :      1: renormalize.infer
 1.85% :     0.000459s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000041s :      1: rewriter_after_opt_a
 0.33% :     0.000082s :      1: rewriter_before_opt_a
 0.36% :     0.000089s :      1: symbol_engine_optimizer
28.83% :     0.007156s :      1: task_emit
21.54% :     0.005345s :      1: type_inference
 0.29% :     0.000073s :      1: validate


TotalTime = 0.0345075, [33]
[bootstrap]: 0.00033787
[type_inference]: 0.00555067
[event_method]: 1.878e-05
[auto_monad]: 6.83e-05
[graph_reusing]: 6.64999e-06
[pre_auto_parallel]: 3.25e-06
[py_interpret_to_execute]: 4.166e-05
[rewriter_before_opt_a]: 7.61e-05
[expand_dump_flag]: 3.18e-06
[jit_opt_a]: 0.00650164, [2]
    [Cycle 1]: 0.00213307, [27]
        [switch_simplify]: 6.717e-05
        [loop_unroll]: 3.132e-05
        [a_1]: 0.00065222
        [with_stream_mark]: 2.478e-05
        [recompute_prepare]: 8.11002e-06
        [updatestate_depend_eliminate]: 4.22e-06
        [updatestate_assign_eliminate]: 3.40998e-06
        [updatestate_loads_eliminate]: 2.59001e-06
        [parameter_eliminate]: 2.63e-06
        [specialize_transform]: 6.33e-06
        [updatestate_useless_node_eliminater]: 5.37001e-06
        [accelerated_algorithm]: 6.34001e-06
        [meta_shard_fg_expand]: 2.98998e-06
        [get_grad_eliminate_]: 5.63002e-06
        [merge_forward]: 3.74002e-06
        [cell_reuse_recompute_pass]: 1.53002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.62e-05
        [j_node_and_user_rematch]: 9.65002e-06
        [meta_fg_expand]: 2.22999e-06
        [replace_old_param]: 1.005e-05
        [inline_without_move]: 5.77999e-06
        [renormalize]: 0.00099537
        [add_forward_monad_depend]: 7.38999e-06
        [auto_monad_grad]: 2.68003e-06
        [auto_monad_eliminator]: 1.654e-05
        [cse]: 3.011e-05
        [replace_applicator]: 1.49e-05
    [Cycle 2]: 0.00034525, [27]
        [switch_simplify]: 5.68997e-06
        [loop_unroll]: 5.41998e-06
        [a_1]: 9.807e-05
        [with_stream_mark]: 1.157e-05
        [recompute_prepare]: 5.52999e-06
        [updatestate_depend_eliminate]: 3.36999e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 1.39e-06
        [specialize_transform]: 4.99e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 6.02999e-06
        [meta_shard_fg_expand]: 1.61998e-06
        [get_grad_eliminate_]: 4.63999e-06
        [merge_forward]: 3.04001e-06
        [cell_reuse_recompute_pass]: 2.78e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.478e-05
        [j_node_and_user_rematch]: 8.28001e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 7.82e-06
        [inline_without_move]: 4.85001e-06
        [renormalize]: 1.19995e-07
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 5.82001e-06
        [cse]: 1.105e-05
        [replace_applicator]: 6.15002e-06
[py_interpret_to_execute_after_opt_a]: 1.305e-05
[rewriter_after_opt_a]: 3.9e-05
[convert_after_rewriter]: 7.9e-06
[order_py_execute_after_rewriter]: 5.486e-05
[mutable_eliminate]: 0.00087375
[jit_opt_b]: 6.757e-05, [1]
    [Cycle 1]: 5.7e-05, [2]
        [frontend_op_eliminate]: 2.343e-05
        [inline_after_opt_a]: 1.969e-05
[cconv]: 3.385e-05
[loop_unroll]: 0.00048903
[jit_opt_after_cconv]: 0.00017813, [1]
    [Cycle 1]: 0.00017105, [11]
        [c_1]: 2.432e-05
        [parameter_eliminate]: 3.55998e-06
        [updatestate_depend_eliminate]: 7.11001e-06
        [updatestate_assign_eliminate]: 2.77002e-06
        [updatestate_loads_eliminate]: 2.26998e-06
        [cse]: 3.834e-05
        [call_graph_tuple_transform]: 2.499e-05
        [tuple_list_get_item_eliminator]: 6.02999e-06
        [none_parameter_eliminate]: 1.67001e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 6.49999e-06
[remove_dup_value]: 1.748e-05
[partial_unused_args_eliminate]: 2.54001e-06
[environ_conv]: 6.37001e-06
[add_recomputation]: 5.818e-05
[cse_after_recomputation]: 2.525e-05, [1]
    [Cycle 1]: 1.878e-05, [1]
        [cse]: 1.197e-05
[auto_monad_reorder]: 1.664e-05
[get_jit_bprop_graph]: 2.86e-06
[rewriter_after_jit_bprop_graph]: 1.003e-05
[opt_after_jit_grad]: 0.0005124
[symbol_engine_optimizer]: 8.182e-05, [1]
    [Cycle 1]: 7.523e-05, [6]
        [build]: 3.84002e-06
        [elim_shapecalc]: 8.22003e-06
        [elim_not_effective]: 1.459e-05
        [opt_reshape]: 7.64002e-06
        [fold_const_symbol]: 9.76998e-06
        [renormalize]: 3.30008e-07
[validate]: 4.263e-05
[backend_pass]: 8.80013e-07
[task_emit]: 0.00698767
[execute]: 9.40001e-06
Sums
    bootstrap                            :     0.000338s :  1.91%
    type_inference                       :     0.005551s : 31.42%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000068s :  0.39%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000042s :  0.24%
    rewriter_before_opt_a                :     0.000076s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000073s :  0.41%
    jit_opt_a.loop_unroll                :     0.000037s :  0.21%
    jit_opt_a.a_1                        :     0.000750s :  4.25%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000995s :  5.63%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.13%
    jit_opt_a.cse                        :     0.000041s :  0.23%
    jit_opt_a.replace_applicator         :     0.000021s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000039s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000055s :  0.31%
    mutable_eliminate                    :     0.000874s :  4.95%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.11%
    cconv                                :     0.000034s :  0.19%
    loop_unroll                          :     0.000489s :  2.77%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000038s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000017s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000058s :  0.33%
    cse_after_recomputation.cse          :     0.000012s :  0.07%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.06%
    opt_after_jit_grad                   :     0.000512s :  2.90%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000043s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006988s : 39.55%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000284    23
 0.91% :     0.000003s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.39% :     0.000007s :      3: substitution.graph_param_transform
86.57% :     0.000246s :      4: substitution.inline
 1.15% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.77% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.83% :     0.000005s :      2: substitution.replace_old_param
 4.90% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005473     2
81.93% :     0.004484s :      1: type_inference.infer
18.07% :     0.000989s :      1: type_inference.specialize
------[replace.]   0.000088     6
66.36% :     0.000058s :      4: replace.inline
33.64% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000256     6
95.06% :     0.000243s :      4: match.inline
 4.94% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 2.78% :     0.000004s :     10: predicate.arithmetic_simplify
 1.49% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.65% :     0.000002s :     10: predicate.depend_value_elim
 0.96% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.50% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.05% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.20% :     0.000002s :      6: predicate.get_grad_eliminate
 0.39% :     0.000001s :      3: predicate.graph_param_transform
 5.36% :     0.000007s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.48% :     0.000002s :      6: predicate.less_batch_normalization
 1.36% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.58% :     0.000002s :     15: predicate.load_eliminater
 1.55% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.73% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.37% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 2.47% :     0.000003s :      3: predicate.mutable_eliminate
 0.74% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000003s :     15: predicate.partial_eliminate
 1.47% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.85% :     0.000002s :     10: predicate.reduce_eliminate
 1.26% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.11% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.08% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.09% :     0.000003s :     16: predicate.switch_defer_inline
 1.84% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.24% :     0.000010s :     53: predicate.switch_simplify
 1.54% :     0.000002s :     10: predicate.tile_eliminate
 1.35% :     0.000002s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.36% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.86% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000763    12
48.40% :     0.000369s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.45% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.15% :     0.000337s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.036596    76
 0.17% :     0.000062s :      1: add_recomputation
 0.20% :     0.000072s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.98% :     0.000357s :      1: bootstrap
 0.10% :     0.000037s :      1: cconv
33.12% :     0.012122s :      1: convert_after_rewriter
 0.07% :     0.000027s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.07% :     0.000024s :      1: event_method
 0.04% :     0.000015s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
17.78% :     0.006506s :      1: jit_opt_a
 0.50% :     0.000181s :      1: jit_opt_after_cconv
 0.19% :     0.000071s :      1: jit_opt_b
 1.36% :     0.000497s :      1: loop_unroll
 2.42% :     0.000884s :      1: mutable_eliminate
 2.69% :     0.000984s :     26: opt.transform.jit_opt_a
 0.16% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000035s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.42% :     0.000521s :      1: opt_after_jit_grad
 0.17% :     0.000064s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000044s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000020s :      1: remove_dup_value
 1.52% :     0.000558s :      1: renormalize.infer
 1.17% :     0.000429s :      1: renormalize.specialize
 0.03% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000042s :      1: rewriter_after_opt_a
 0.22% :     0.000079s :      1: rewriter_before_opt_a
 0.23% :     0.000084s :      1: symbol_engine_optimizer
19.15% :     0.007008s :      1: task_emit
15.22% :     0.005568s :      1: type_inference
 0.17% :     0.000064s :      1: validate


TotalTime = 0.0247577, [33]
[bootstrap]: 0.0003709
[type_inference]: 0.00703299
[event_method]: 1.997e-05
[auto_monad]: 6.708e-05
[graph_reusing]: 6.39001e-06
[pre_auto_parallel]: 2.76e-06
[py_interpret_to_execute]: 4.388e-05
[rewriter_before_opt_a]: 7.58e-05
[expand_dump_flag]: 3.13e-06
[jit_opt_a]: 0.00858779, [2]
    [Cycle 1]: 0.00297645, [27]
        [switch_simplify]: 7.368e-05
        [loop_unroll]: 2.907e-05
        [a_1]: 0.00061923
        [with_stream_mark]: 2.468e-05
        [recompute_prepare]: 7.36999e-06
        [updatestate_depend_eliminate]: 4.82e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.59001e-06
        [parameter_eliminate]: 2.36e-06
        [specialize_transform]: 6.66999e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 5.37999e-06
        [meta_shard_fg_expand]: 2.39001e-06
        [get_grad_eliminate_]: 5.61003e-06
        [merge_forward]: 3.76001e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.547e-05
        [j_node_and_user_rematch]: 9.79e-06
        [meta_fg_expand]: 2.43e-06
        [replace_old_param]: 1.023e-05
        [inline_without_move]: 5.05999e-06
        [renormalize]: 0.00188432
        [add_forward_monad_depend]: 7.05e-06
        [auto_monad_grad]: 2.51e-06
        [auto_monad_eliminator]: 1.662e-05
        [cse]: 1.881e-05
        [replace_applicator]: 1.478e-05
    [Cycle 2]: 0.00034469, [27]
        [switch_simplify]: 5.71e-06
        [loop_unroll]: 5.35999e-06
        [a_1]: 9.849e-05
        [with_stream_mark]: 1.384e-05
        [recompute_prepare]: 6.02999e-06
        [updatestate_depend_eliminate]: 2.73e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 5.14e-06
        [updatestate_useless_node_eliminater]: 4.83001e-06
        [accelerated_algorithm]: 5.35001e-06
        [meta_shard_fg_expand]: 1.63002e-06
        [get_grad_eliminate_]: 4.77e-06
        [merge_forward]: 3.18998e-06
        [cell_reuse_recompute_pass]: 3.08998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.454e-05
        [j_node_and_user_rematch]: 8.03001e-06
        [meta_fg_expand]: 2.16e-06
        [replace_old_param]: 7.13998e-06
        [inline_without_move]: 1.208e-05
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 9.80013e-07
        [auto_monad_eliminator]: 5.21002e-06
        [cse]: 1.034e-05
        [replace_applicator]: 4.67998e-06
[py_interpret_to_execute_after_opt_a]: 1.471e-05
[rewriter_after_opt_a]: 4.017e-05
[convert_after_rewriter]: 8.80999e-06
[order_py_execute_after_rewriter]: 5.15999e-06
[mutable_eliminate]: 0.00074768
[jit_opt_b]: 5.393e-05, [1]
    [Cycle 1]: 4.592e-05, [2]
        [frontend_op_eliminate]: 1.854e-05
        [inline_after_opt_a]: 1.545e-05
[cconv]: 2.629e-05
[loop_unroll]: 0.00046429
[jit_opt_after_cconv]: 0.0001551, [1]
    [Cycle 1]: 0.00014802, [11]
        [c_1]: 2.181e-05
        [parameter_eliminate]: 4.13001e-06
        [updatestate_depend_eliminate]: 6.11998e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.60002e-06
        [cse]: 2.501e-05
        [call_graph_tuple_transform]: 2.155e-05
        [tuple_list_get_item_eliminator]: 5.29998e-06
        [none_parameter_eliminate]: 1.35001e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 5.08002e-06
[remove_dup_value]: 1.389e-05
[partial_unused_args_eliminate]: 2.10002e-06
[environ_conv]: 5.64998e-06
[add_recomputation]: 5.183e-05
[cse_after_recomputation]: 2.082e-05, [1]
    [Cycle 1]: 1.549e-05, [1]
        [cse]: 9.86998e-06
[auto_monad_reorder]: 1.415e-05
[get_jit_bprop_graph]: 2.04999e-06
[rewriter_after_jit_bprop_graph]: 7.63001e-06
[opt_after_jit_grad]: 0.00049884
[symbol_engine_optimizer]: 7.983e-05, [1]
    [Cycle 1]: 7.219e-05, [6]
        [build]: 4.80999e-06
        [elim_shapecalc]: 8.32e-06
        [elim_not_effective]: 1.434e-05
        [opt_reshape]: 5.96e-06
        [fold_const_symbol]: 9.40001e-06
        [renormalize]: 7.90023e-07
[validate]: 4.824e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00606405
[execute]: 6.53003e-06
Sums
    bootstrap                            :     0.000371s :  1.97%
    type_inference                       :     0.007033s : 37.32%
    event_method                         :     0.000020s :  0.11%
    auto_monad                           :     0.000067s :  0.36%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.23%
    rewriter_before_opt_a                :     0.000076s :  0.40%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000079s :  0.42%
    jit_opt_a.loop_unroll                :     0.000034s :  0.18%
    jit_opt_a.a_1                        :     0.000718s :  3.81%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.09%
    jit_opt_a.inline_without_move        :     0.000017s :  0.09%
    jit_opt_a.renormalize                :     0.001884s : 10.00%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.12%
    jit_opt_a.cse                        :     0.000029s :  0.15%
    jit_opt_a.replace_applicator         :     0.000019s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.08%
    rewriter_after_opt_a                 :     0.000040s :  0.21%
    convert_after_rewriter               :     0.000009s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000748s :  3.97%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000026s :  0.14%
    loop_unroll                          :     0.000464s :  2.46%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000052s :  0.28%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000499s :  2.65%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000048s :  0.26%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006064s : 32.18%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000268    23
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.78% :     0.000002s :      2: substitution.fold_const_symbol
 2.26% :     0.000006s :      3: substitution.graph_param_transform
86.12% :     0.000231s :      4: substitution.inline
 1.50% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.92% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.89% :     0.000005s :      2: substitution.replace_old_param
 4.75% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006948     2
83.82% :     0.005824s :      1: type_inference.infer
16.18% :     0.001124s :      1: type_inference.specialize
------[replace.]   0.000081     6
65.49% :     0.000053s :      4: replace.inline
34.51% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000240     6
95.15% :     0.000228s :      4: match.inline
 4.85% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.99% :     0.000003s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.21% :     0.000002s :     10: predicate.addn_zero_filter
 2.85% :     0.000004s :     10: predicate.arithmetic_simplify
 1.11% :     0.000001s :     10: predicate.cast_eliminate
 0.68% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.59% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.51% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.70% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     10: predicate.environ_get_eliminate
 1.25% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.44% :     0.000001s :      3: predicate.graph_param_transform
 5.08% :     0.000006s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :      6: predicate.less_batch_normalization
 1.46% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.57% :     0.000002s :     15: predicate.load_eliminater
 1.49% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.76% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.25% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 1.24% :     0.000002s :     10: predicate.minmaximum_grad
 1.67% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.07% :     0.000003s :     15: predicate.partial_eliminate
 1.09% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.00% :     0.000003s :     10: predicate.reduce_eliminate
 1.71% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.53% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.43% :     0.000001s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.46% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.03% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.06% :     0.000001s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.17% :     0.000003s :     16: predicate.switch_defer_inline
 1.85% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.64% :     0.000010s :     53: predicate.switch_simplify
 1.60% :     0.000002s :     10: predicate.tile_eliminate
 1.06% :     0.000001s :     10: predicate.transpose_eliminate
 1.25% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.58% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.39% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.46% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.57% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000762    12
41.69% :     0.000318s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.47% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.84% :     0.000395s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027706    76
 0.20% :     0.000055s :      1: add_recomputation
 0.25% :     0.000070s :      1: auto_monad
 0.06% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.39% :     0.000384s :      1: bootstrap
 0.10% :     0.000029s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.08% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.09% :     0.000025s :      1: event_method
 0.04% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
31.01% :     0.008591s :      1: jit_opt_a
 0.57% :     0.000158s :      1: jit_opt_after_cconv
 0.20% :     0.000057s :      1: jit_opt_b
 1.70% :     0.000472s :      1: loop_unroll
 2.74% :     0.000758s :      1: mutable_eliminate
 3.46% :     0.000958s :     26: opt.transform.jit_opt_a
 0.18% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.83% :     0.000507s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000046s :      1: py_interpret_to_execute
 0.06% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000016s :      1: remove_dup_value
 5.02% :     0.001391s :      1: renormalize.infer
 1.73% :     0.000480s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000044s :      1: rewriter_after_opt_a
 0.28% :     0.000079s :      1: rewriter_before_opt_a
 0.30% :     0.000082s :      1: symbol_engine_optimizer
21.94% :     0.006080s :      1: task_emit
25.46% :     0.007054s :      1: type_inference
 0.24% :     0.000068s :      1: validate


TotalTime = 0.0376993, [33]
[bootstrap]: 0.00039589
[type_inference]: 0.00786182
[event_method]: 1.799e-05
[auto_monad]: 6.889e-05
[graph_reusing]: 5.79999e-06
[pre_auto_parallel]: 2.59001e-06
[py_interpret_to_execute]: 4.489e-05
[rewriter_before_opt_a]: 0.00010779
[expand_dump_flag]: 3.92998e-06
[jit_opt_a]: 0.0119733, [2]
    [Cycle 1]: 0.00182836, [27]
        [switch_simplify]: 5.694e-05
        [loop_unroll]: 3.028e-05
        [a_1]: 0.0006163
        [with_stream_mark]: 1.54e-05
        [recompute_prepare]: 7.96001e-06
        [updatestate_depend_eliminate]: 3.25002e-06
        [updatestate_assign_eliminate]: 2.11998e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 6.00002e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 5.65001e-06
        [meta_shard_fg_expand]: 1.45001e-06
        [get_grad_eliminate_]: 6.17001e-06
        [merge_forward]: 2.58e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.129e-05
        [j_node_and_user_rematch]: 9.66e-06
        [meta_fg_expand]: 2.18998e-06
        [replace_old_param]: 8.41002e-06
        [inline_without_move]: 5.34998e-06
        [renormalize]: 0.0007799
        [add_forward_monad_depend]: 6.48998e-06
        [auto_monad_grad]: 2.58998e-06
        [auto_monad_eliminator]: 1.629e-05
        [cse]: 3.124e-05
        [replace_applicator]: 1.403e-05
    [Cycle 2]: 0.00032489, [27]
        [switch_simplify]: 5.81e-06
        [loop_unroll]: 5.90002e-06
        [a_1]: 9.258e-05
        [with_stream_mark]: 1.033e-05
        [recompute_prepare]: 5.00999e-06
        [updatestate_depend_eliminate]: 2.51e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 1.24998e-06
        [specialize_transform]: 4.72998e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 5.14998e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 4.52998e-06
        [merge_forward]: 2.44001e-06
        [cell_reuse_recompute_pass]: 2.53003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.352e-05
        [j_node_and_user_rematch]: 7.58999e-06
        [meta_fg_expand]: 2.19001e-06
        [replace_old_param]: 7.35998e-06
        [inline_without_move]: 4.57e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 8.80013e-07
        [auto_monad_eliminator]: 5.39e-06
        [cse]: 9.77001e-06
        [replace_applicator]: 4.68999e-06
[py_interpret_to_execute_after_opt_a]: 1.45e-05
[rewriter_after_opt_a]: 3.791e-05
[convert_after_rewriter]: 7.48e-06
[order_py_execute_after_rewriter]: 5.02e-06
[mutable_eliminate]: 0.00073684
[jit_opt_b]: 5.284e-05, [1]
    [Cycle 1]: 4.449e-05, [2]
        [frontend_op_eliminate]: 1.651e-05
        [inline_after_opt_a]: 1.543e-05
[cconv]: 3.071e-05
[loop_unroll]: 0.00044687
[jit_opt_after_cconv]: 0.00015562, [1]
    [Cycle 1]: 0.00014852, [11]
        [c_1]: 2.192e-05
        [parameter_eliminate]: 4.55999e-06
        [updatestate_depend_eliminate]: 7.66001e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [cse]: 2.334e-05
        [call_graph_tuple_transform]: 2.171e-05
        [tuple_list_get_item_eliminator]: 5.77999e-06
        [none_parameter_eliminate]: 1.64998e-06
        [renormalize]: 7.90023e-07
        [switch_simplify]: 5.07e-06
[remove_dup_value]: 1.484e-05
[partial_unused_args_eliminate]: 2.16998e-06
[environ_conv]: 6.15002e-06
[add_recomputation]: 5.173e-05
[cse_after_recomputation]: 2.063e-05, [1]
    [Cycle 1]: 1.48e-05, [1]
        [cse]: 8.93002e-06
[auto_monad_reorder]: 1.558e-05
[get_jit_bprop_graph]: 1.49e-06
[rewriter_after_jit_bprop_graph]: 7.58999e-06
[opt_after_jit_grad]: 0.00055311
[symbol_engine_optimizer]: 7.833e-05, [1]
    [Cycle 1]: 7.111e-05, [6]
        [build]: 4.35999e-06
        [elim_shapecalc]: 8.2e-06
        [elim_not_effective]: 1.414e-05
        [opt_reshape]: 5.86e-06
        [fold_const_symbol]: 9.08002e-06
        [renormalize]: 5.79981e-07
[validate]: 4.326e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.0146886
[execute]: 9.20001e-06
Sums
    bootstrap                            :     0.000396s :  1.45%
    type_inference                       :     0.007862s : 28.88%
    event_method                         :     0.000018s :  0.07%
    auto_monad                           :     0.000069s :  0.25%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000045s :  0.16%
    rewriter_before_opt_a                :     0.000108s :  0.40%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000063s :  0.23%
    jit_opt_a.loop_unroll                :     0.000036s :  0.13%
    jit_opt_a.a_1                        :     0.000709s :  2.60%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.04%
    jit_opt_a.merge_forward              :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.06%
    jit_opt_a.inline_without_move        :     0.000010s :  0.04%
    jit_opt_a.renormalize                :     0.000780s :  2.86%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.08%
    jit_opt_a.cse                        :     0.000041s :  0.15%
    jit_opt_a.replace_applicator         :     0.000019s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.05%
    rewriter_after_opt_a                 :     0.000038s :  0.14%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000737s :  2.71%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.06%
    cconv                                :     0.000031s :  0.11%
    loop_unroll                          :     0.000447s :  1.64%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000015s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000052s :  0.19%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.03%
    opt_after_jit_grad                   :     0.000553s :  2.03%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.014689s : 53.95%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000216    23
 1.35% :     0.000003s :      2: substitution.elim_not_effective
 0.72% :     0.000002s :      2: substitution.fold_const_symbol
 2.96% :     0.000006s :      3: substitution.graph_param_transform
85.75% :     0.000185s :      4: substitution.inline
 1.52% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.88% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.67% :     0.000004s :      2: substitution.replace_old_param
 4.16% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007774     2
73.48% :     0.005712s :      1: type_inference.infer
26.52% :     0.002062s :      1: type_inference.specialize
------[replace.]   0.000086     6
72.80% :     0.000062s :      4: replace.inline
27.20% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000191     6
95.88% :     0.000183s :      4: match.inline
 4.12% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.83% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.57% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 2.35% :     0.000003s :     10: predicate.arithmetic_simplify
 1.10% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.85% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.46% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.74% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 4.90% :     0.000006s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.73% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.60% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.04% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.07% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.40% :     0.000002s :     10: predicate.merge_addn
 1.17% :     0.000001s :     10: predicate.minmaximum_grad
 2.02% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000002s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.18% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.61% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.66% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.69% :     0.000009s :     53: predicate.switch_simplify
 1.31% :     0.000002s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.10% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.86% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000794    12
50.87% :     0.000404s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.36% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.77% :     0.000355s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.039505    76
 0.14% :     0.000055s :      1: add_recomputation
 0.18% :     0.000073s :      1: auto_monad
 0.05% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.06% :     0.000419s :      1: bootstrap
 0.09% :     0.000034s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.06% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.05% :     0.000021s :      1: event_method
 0.04% :     0.000014s :      1: execute
 0.02% :     0.000007s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
30.32% :     0.011978s :      1: jit_opt_a
 0.40% :     0.000158s :      1: jit_opt_after_cconv
 0.14% :     0.000055s :      1: jit_opt_b
 1.15% :     0.000455s :      1: loop_unroll
 1.89% :     0.000748s :      1: mutable_eliminate
 2.34% :     0.000924s :     26: opt.transform.jit_opt_a
 0.13% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000026s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.42% :     0.000562s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000048s :      1: py_interpret_to_execute
 0.04% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000017s :      1: remove_dup_value
 1.04% :     0.000412s :      1: renormalize.infer
 0.91% :     0.000361s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000041s :      1: rewriter_after_opt_a
 0.28% :     0.000112s :      1: rewriter_before_opt_a
 0.20% :     0.000081s :      1: symbol_engine_optimizer
37.23% :     0.014709s :      1: task_emit
19.94% :     0.007876s :      1: type_inference
 0.17% :     0.000065s :      1: validate


TotalTime = 0.0171836, [30]
[bootstrap]: 0.00029515
[type_inference]: 0.00733402
[event_method]: 2.403e-05
[auto_monad]: 8.88e-05
[graph_reusing]: 6.85002e-06
[pre_auto_parallel]: 3.22002e-06
[py_interpret_to_execute]: 5.574e-05
[rewriter_before_opt_a]: 9.058e-05
[expand_dump_flag]: 3.69002e-06
[jit_opt_a]: 0.00671329, [2]
    [Cycle 1]: 0.00226425, [27]
        [switch_simplify]: 0.00014662
        [loop_unroll]: 3.463e-05
        [a_1]: 0.00079276
        [with_stream_mark]: 2.851e-05
        [recompute_prepare]: 8.59998e-06
        [updatestate_depend_eliminate]: 4.08999e-06
        [updatestate_assign_eliminate]: 3.25998e-06
        [updatestate_loads_eliminate]: 2.87002e-06
        [parameter_eliminate]: 2.20002e-06
        [specialize_transform]: 6.33998e-06
        [updatestate_useless_node_eliminater]: 5.31998e-06
        [accelerated_algorithm]: 5.97001e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 5.14998e-06
        [merge_forward]: 4.37e-06
        [cell_reuse_recompute_pass]: 9.60019e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.652e-05
        [j_node_and_user_rematch]: 9.61998e-06
        [meta_fg_expand]: 2.71999e-06
        [replace_old_param]: 9.51998e-06
        [inline_without_move]: 5.42001e-06
        [renormalize]: 0.00087226
        [add_forward_monad_depend]: 6.93e-06
        [auto_monad_grad]: 2.49999e-06
        [auto_monad_eliminator]: 1.694e-05
        [cse]: 3.437e-05
        [replace_applicator]: 1.672e-05
    [Cycle 2]: 0.0003245, [27]
        [switch_simplify]: 5.27001e-06
        [loop_unroll]: 4.84e-06
        [a_1]: 9.095e-05
        [with_stream_mark]: 1.216e-05
        [recompute_prepare]: 5.19998e-06
        [updatestate_depend_eliminate]: 2.94001e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 5.19e-06
        [updatestate_useless_node_eliminater]: 5.01002e-06
        [accelerated_algorithm]: 5.14e-06
        [meta_shard_fg_expand]: 1.68002e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 2.96001e-06
        [cell_reuse_recompute_pass]: 1.62999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.57e-05
        [j_node_and_user_rematch]: 8.31002e-06
        [meta_fg_expand]: 1.64e-06
        [replace_old_param]: 8.1e-06
        [inline_without_move]: 5.23002e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 1.10001e-06
        [auto_monad_eliminator]: 4.85999e-06
        [cse]: 1.008e-05
        [replace_applicator]: 4.76002e-06
[py_interpret_to_execute_after_opt_a]: 1.461e-05
[rewriter_after_opt_a]: 3.769e-05
[convert_after_rewriter]: 7.15998e-06
[order_py_execute_after_rewriter]: 5.05001e-06
[mutable_eliminate]: 0.00073585
[jit_opt_b]: 6.505e-05, [1]
    [Cycle 1]: 5.597e-05, [2]
        [frontend_op_eliminate]: 1.732e-05
        [inline_after_opt_a]: 1.734e-05
[cconv]: 3.34e-05
[loop_unroll]: 0.00050637
[jit_opt_after_cconv]: 0.00016233, [1]
    [Cycle 1]: 0.00015387, [11]
        [c_1]: 2.108e-05
        [parameter_eliminate]: 4.47003e-06
        [updatestate_depend_eliminate]: 8.54e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.38002e-06
        [cse]: 2.38e-05
        [call_graph_tuple_transform]: 2.285e-05
        [tuple_list_get_item_eliminator]: 5.53002e-06
        [none_parameter_eliminate]: 2.42001e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 6.07999e-06
[remove_dup_value]: 1.577e-05
[partial_unused_args_eliminate]: 2.32999e-06
[environ_conv]: 6.04999e-06
[add_recomputation]: 5.443e-05
[cse_after_recomputation]: 2.484e-05, [1]
    [Cycle 1]: 1.819e-05, [1]
        [cse]: 1.04e-05
[auto_monad_reorder]: 1.505e-05
[get_jit_bprop_graph]: 2.17001e-06
[rewriter_after_jit_bprop_graph]: 7.23999e-06
[opt_after_jit_grad]: 0.00053186
[symbol_engine_optimizer]: 8.138e-05, [1]
    [Cycle 1]: 7.353e-05, [6]
        [build]: 5.80002e-06
        [elim_shapecalc]: 8.13999e-06
        [elim_not_effective]: 1.516e-05
        [opt_reshape]: 5.44998e-06
        [fold_const_symbol]: 8.37e-06
        [renormalize]: 4.50003e-07
[validate]: 4.257e-05
Sums
    bootstrap                            :     0.000295s :  2.39%
    type_inference                       :     0.007334s : 59.29%
    event_method                         :     0.000024s :  0.19%
    auto_monad                           :     0.000089s :  0.72%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000056s :  0.45%
    rewriter_before_opt_a                :     0.000091s :  0.73%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000152s :  1.23%
    jit_opt_a.loop_unroll                :     0.000039s :  0.32%
    jit_opt_a.a_1                        :     0.000884s :  7.14%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.14%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.000872s :  7.05%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.18%
    jit_opt_a.cse                        :     0.000044s :  0.36%
    jit_opt_a.replace_applicator         :     0.000021s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.12%
    rewriter_after_opt_a                 :     0.000038s :  0.30%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000736s :  5.95%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.14%
    cconv                                :     0.000033s :  0.27%
    loop_unroll                          :     0.000506s :  4.09%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.17%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000016s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000054s :  0.44%
    cse_after_recomputation.cse          :     0.000010s :  0.08%
    auto_monad_reorder                   :     0.000015s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000532s :  4.30%
    symbol_engine_optimizer.build        :     0.000006s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000043s :  0.34%


Time group info:
------[substitution.]   0.000344    26
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.41% :     0.000001s :      2: substitution.fold_const_symbol
 1.72% :     0.000006s :      3: substitution.graph_param_transform
85.83% :     0.000295s :      6: substitution.inline
 1.10% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.39% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.59% :     0.000005s :      2: substitution.replace_old_param
 3.01% :     0.000010s :      1: substitution.switch_simplify
 4.23% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007247     2
83.60% :     0.006058s :      1: type_inference.infer
16.40% :     0.001189s :      1: type_inference.specialize
------[replace.]   0.000158     9
48.18% :     0.000076s :      6: replace.inline
28.97% :     0.000046s :      1: replace.switch_simplify
22.85% :     0.000036s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000314     9
92.66% :     0.000291s :      6: match.inline
 3.06% :     0.000010s :      1: match.switch_simplify
 4.28% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000145   801
 1.18% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     12: predicate.addn_check_dump
 1.26% :     0.000002s :     12: predicate.addn_zero_filter
 2.07% :     0.000003s :     12: predicate.arithmetic_simplify
 1.19% :     0.000002s :     12: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     12: predicate.compare_switch_simplify
 1.75% :     0.000003s :     12: predicate.depend_value_elim
 1.04% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.16% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.65% :     0.000001s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.08% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.11% :     0.000002s :     12: predicate.environ_get_eliminate
 1.05% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.17% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.19% :     0.000000s :      3: predicate.graph_param_transform
 6.08% :     0.000009s :     26: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :      6: predicate.less_batch_normalization
 1.78% :     0.000003s :     14: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     17: predicate.load_eliminater
 1.61% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.85% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.97% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000002s :     12: predicate.merge_addn
 1.04% :     0.000001s :     12: predicate.minmaximum_grad
 2.66% :     0.000004s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.00% :     0.000003s :     17: predicate.partial_eliminate
 1.11% :     0.000002s :     12: predicate.print_const_string_wrapper
 2.20% :     0.000003s :     12: predicate.reduce_eliminate
 1.36% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.49% :     0.000002s :     20: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.41% :     0.000001s :      3: predicate.reset_defer_inline
 1.56% :     0.000002s :     12: predicate.reshape_eliminate
 1.25% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 1.02% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.09% :     0.000002s :     12: predicate.same_eliminate
 0.39% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.94% :     0.000004s :     20: predicate.switch_defer_inline
 2.34% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.42% :     0.000012s :     65: predicate.switch_simplify
 1.36% :     0.000002s :     12: predicate.tile_eliminate
 1.38% :     0.000002s :     12: predicate.transpose_eliminate
 1.67% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.81% :     0.000006s :     20: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.53% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.40% :     0.000002s :     12: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000896    15
47.66% :     0.000427s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.69% :     0.000123s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.65% :     0.000346s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019314    72
 0.30% :     0.000058s :      1: add_recomputation
 0.48% :     0.000093s :      1: auto_monad
 0.10% :     0.000019s :      1: auto_monad_reorder
 1.63% :     0.000315s :      1: bootstrap
 0.19% :     0.000036s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.15% :     0.000028s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
34.78% :     0.006717s :      1: jit_opt_a
 0.86% :     0.000165s :      1: jit_opt_after_cconv
 0.35% :     0.000068s :      1: jit_opt_b
 2.68% :     0.000518s :      1: loop_unroll
 3.89% :     0.000752s :      1: mutable_eliminate
 6.21% :     0.001200s :     26: opt.transform.jit_opt_a
 0.27% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000027s :      4: opt.transform.jit_opt_b
 0.09% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.83% :     0.000546s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000059s :      1: py_interpret_to_execute
 0.09% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000018s :      1: remove_dup_value
 2.59% :     0.000499s :      1: renormalize.infer
 1.87% :     0.000361s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000041s :      1: rewriter_after_opt_a
 0.49% :     0.000095s :      1: rewriter_before_opt_a
 0.44% :     0.000084s :      1: symbol_engine_optimizer
38.10% :     0.007358s :      1: type_inference


TotalTime = 0.0506115, [33]
[bootstrap]: 0.00037455
[type_inference]: 0.00553747
[event_method]: 2.035e-05
[auto_monad]: 6.885e-05
[graph_reusing]: 5.40001e-06
[pre_auto_parallel]: 2.79999e-06
[py_interpret_to_execute]: 4.299e-05
[rewriter_before_opt_a]: 7.57e-05
[expand_dump_flag]: 3.46999e-06
[jit_opt_a]: 0.00684988, [2]
    [Cycle 1]: 0.00228833, [27]
        [switch_simplify]: 6.997e-05
        [loop_unroll]: 3.082e-05
        [a_1]: 0.00078765
        [with_stream_mark]: 2.608e-05
        [recompute_prepare]: 7.68001e-06
        [updatestate_depend_eliminate]: 4.16001e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 2.53998e-06
        [parameter_eliminate]: 2.62001e-06
        [specialize_transform]: 5.72001e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.87001e-06
        [meta_shard_fg_expand]: 2.31998e-06
        [get_grad_eliminate_]: 5.71e-06
        [merge_forward]: 4.19997e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.581e-05
        [j_node_and_user_rematch]: 9.84001e-06
        [meta_fg_expand]: 2.74001e-06
        [replace_old_param]: 1.07e-05
        [inline_without_move]: 5.43002e-06
        [renormalize]: 0.00101457
        [add_forward_monad_depend]: 6.22001e-06
        [auto_monad_grad]: 1.29998e-06
        [auto_monad_eliminator]: 1.554e-05
        [cse]: 3.123e-05
        [replace_applicator]: 1.362e-05
    [Cycle 2]: 0.00033971, [27]
        [switch_simplify]: 5.94e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 9.876e-05
        [with_stream_mark]: 9.88998e-06
        [recompute_prepare]: 5.23002e-06
        [updatestate_depend_eliminate]: 2.59999e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 1.66e-06
        [specialize_transform]: 5.13002e-06
        [updatestate_useless_node_eliminater]: 4.72e-06
        [accelerated_algorithm]: 5.72999e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.79002e-06
        [merge_forward]: 2.49001e-06
        [cell_reuse_recompute_pass]: 2.22001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.242e-05
        [j_node_and_user_rematch]: 8.34002e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 1.572e-05
        [inline_without_move]: 4.63001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.26997e-06
        [auto_monad_grad]: 7.10017e-07
        [auto_monad_eliminator]: 4.96997e-06
        [cse]: 9.92001e-06
        [replace_applicator]: 5.40999e-06
[py_interpret_to_execute_after_opt_a]: 1.165e-05
[rewriter_after_opt_a]: 2.816e-05
[convert_after_rewriter]: 5.75001e-06
[order_py_execute_after_rewriter]: 4.11001e-06
[mutable_eliminate]: 0.00073858
[jit_opt_b]: 5.479e-05, [1]
    [Cycle 1]: 4.651e-05, [2]
        [frontend_op_eliminate]: 1.797e-05
        [inline_after_opt_a]: 1.612e-05
[cconv]: 2.802e-05
[loop_unroll]: 0.00045574
[jit_opt_after_cconv]: 0.00014612, [1]
    [Cycle 1]: 0.0001391, [11]
        [c_1]: 2.216e-05
        [parameter_eliminate]: 2.86e-06
        [updatestate_depend_eliminate]: 6.11e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 1.927e-05
        [call_graph_tuple_transform]: 2.003e-05
        [tuple_list_get_item_eliminator]: 5.22e-06
        [none_parameter_eliminate]: 9.80013e-07
        [renormalize]: 5.79981e-07
        [switch_simplify]: 5.22e-06
[remove_dup_value]: 1.362e-05
[partial_unused_args_eliminate]: 1.97001e-06
[environ_conv]: 5.73002e-06
[add_recomputation]: 5.224e-05
[cse_after_recomputation]: 2.207e-05, [1]
    [Cycle 1]: 1.606e-05, [1]
        [cse]: 1.037e-05
[auto_monad_reorder]: 1.428e-05
[get_jit_bprop_graph]: 2.27999e-06
[rewriter_after_jit_bprop_graph]: 7.03998e-06
[opt_after_jit_grad]: 0.0005563
[symbol_engine_optimizer]: 7.742e-05, [1]
    [Cycle 1]: 7.005e-05, [6]
        [build]: 3.44001e-06
        [elim_shapecalc]: 8.21002e-06
        [elim_not_effective]: 1.419e-05
        [opt_reshape]: 6.07001e-06
        [fold_const_symbol]: 8.77e-06
        [renormalize]: 7.2e-07
[validate]: 4.12e-05
[backend_pass]: 1.17e-06
[task_emit]: 0.0351169
[execute]: 1.031e-05
Sums
    bootstrap                            :     0.000375s :  0.82%
    type_inference                       :     0.005537s : 12.11%
    event_method                         :     0.000020s :  0.04%
    auto_monad                           :     0.000069s :  0.15%
    graph_reusing                        :     0.000005s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.09%
    rewriter_before_opt_a                :     0.000076s :  0.17%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000076s :  0.17%
    jit_opt_a.loop_unroll                :     0.000036s :  0.08%
    jit_opt_a.a_1                        :     0.000886s :  1.94%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.02%
    jit_opt_a.merge_forward              :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000026s :  0.06%
    jit_opt_a.inline_without_move        :     0.000010s :  0.02%
    jit_opt_a.renormalize                :     0.001015s :  2.22%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.04%
    jit_opt_a.cse                        :     0.000041s :  0.09%
    jit_opt_a.replace_applicator         :     0.000019s :  0.04%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000028s :  0.06%
    convert_after_rewriter               :     0.000006s :  0.01%
    order_py_execute_after_rewriter      :     0.000004s :  0.01%
    mutable_eliminate                    :     0.000739s :  1.62%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.04%
    cconv                                :     0.000028s :  0.06%
    loop_unroll                          :     0.000456s :  1.00%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.05%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000019s :  0.04%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.01%
    remove_dup_value                     :     0.000014s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000006s :  0.01%
    add_recomputation                    :     0.000052s :  0.11%
    cse_after_recomputation.cse          :     0.000010s :  0.02%
    auto_monad_reorder                   :     0.000014s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000556s :  1.22%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.09%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.035117s : 76.82%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000289    23
 0.50% :     0.000001s :      2: substitution.elim_not_effective
 0.33% :     0.000001s :      2: substitution.fold_const_symbol
 1.59% :     0.000005s :      3: substitution.graph_param_transform
85.43% :     0.000247s :      4: substitution.inline
 1.25% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.61% :     0.000005s :      4: substitution.remove_not_recompute_node
 4.23% :     0.000012s :      2: substitution.replace_old_param
 5.06% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005453     2
81.46% :     0.004442s :      1: type_inference.infer
18.54% :     0.001011s :      1: type_inference.specialize
------[replace.]   0.000092     6
64.70% :     0.000060s :      4: replace.inline
35.30% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000258     6
94.82% :     0.000244s :      4: match.inline
 5.18% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.37% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 2.58% :     0.000003s :     10: predicate.arithmetic_simplify
 1.23% :     0.000002s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.50% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.45% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.44% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.54% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.07% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.48% :     0.000007s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.58% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.71% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.79% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     15: predicate.partial_eliminate
 1.39% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.82% :     0.000002s :     10: predicate.reshape_eliminate
 1.92% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000001s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.26% :     0.000002s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.60% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.73% :     0.000004s :     16: predicate.switch_defer_inline
 1.95% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.43% :     0.000010s :     53: predicate.switch_simplify
 1.51% :     0.000002s :     10: predicate.tile_eliminate
 1.18% :     0.000002s :     10: predicate.transpose_eliminate
 1.70% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.76% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.61% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.88% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000766    12
47.92% :     0.000367s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.61% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.47% :     0.000340s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.052855    76
 0.10% :     0.000055s :      1: add_recomputation
 0.14% :     0.000073s :      1: auto_monad
 0.03% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.75% :     0.000396s :      1: bootstrap
 0.06% :     0.000031s :      1: cconv
 0.02% :     0.000008s :      1: convert_after_rewriter
 0.05% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.05% :     0.000025s :      1: event_method
 0.03% :     0.000016s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000008s :      1: graph_reusing
12.97% :     0.006855s :      1: jit_opt_a
 0.28% :     0.000149s :      1: jit_opt_after_cconv
 0.11% :     0.000058s :      1: jit_opt_b
 0.88% :     0.000465s :      1: loop_unroll
 1.42% :     0.000750s :      1: mutable_eliminate
 2.13% :     0.001126s :     26: opt.transform.jit_opt_a
 0.09% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000027s :      4: opt.transform.jit_opt_b
 0.02% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.07% :     0.000566s :      1: opt_after_jit_grad
 0.01% :     0.000006s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000046s :      1: py_interpret_to_execute
 0.03% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000016s :      1: remove_dup_value
 1.06% :     0.000560s :      1: renormalize.infer
 0.84% :     0.000445s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.06% :     0.000031s :      1: rewriter_after_opt_a
 0.15% :     0.000078s :      1: rewriter_before_opt_a
 0.15% :     0.000080s :      1: symbol_engine_optimizer
66.48% :     0.035136s :      1: task_emit
10.51% :     0.005555s :      1: type_inference
 0.12% :     0.000063s :      1: validate


TotalTime = 0.0393319, [33]
[bootstrap]: 0.00044759
[type_inference]: 0.00566817
[event_method]: 1.991e-05
[auto_monad]: 6.862e-05
[graph_reusing]: 5.87999e-06
[pre_auto_parallel]: 2.91e-06
[py_interpret_to_execute]: 5.186e-05
[rewriter_before_opt_a]: 8.346e-05
[expand_dump_flag]: 3.64002e-06
[jit_opt_a]: 0.00661767, [2]
    [Cycle 1]: 0.00208322, [27]
        [switch_simplify]: 7.011e-05
        [loop_unroll]: 3.075e-05
        [a_1]: 0.00063098
        [with_stream_mark]: 2.323e-05
        [recompute_prepare]: 8.3e-06
        [updatestate_depend_eliminate]: 4.08999e-06
        [updatestate_assign_eliminate]: 3.38999e-06
        [updatestate_loads_eliminate]: 2.99999e-06
        [parameter_eliminate]: 1.85001e-06
        [specialize_transform]: 6.41998e-06
        [updatestate_useless_node_eliminater]: 5.21998e-06
        [accelerated_algorithm]: 6.17001e-06
        [meta_shard_fg_expand]: 2.34001e-06
        [get_grad_eliminate_]: 6.00002e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.682e-05
        [j_node_and_user_rematch]: 1.064e-05
        [meta_fg_expand]: 2.19001e-06
        [replace_old_param]: 1.032e-05
        [inline_without_move]: 5.58002e-06
        [renormalize]: 0.00096255
        [add_forward_monad_depend]: 7.00002e-06
        [auto_monad_grad]: 3.34001e-06
        [auto_monad_eliminator]: 1.507e-05
        [cse]: 3.016e-05
        [replace_applicator]: 1.532e-05
    [Cycle 2]: 0.00034029, [27]
        [switch_simplify]: 6.01e-06
        [loop_unroll]: 5.70001e-06
        [a_1]: 9.567e-05
        [with_stream_mark]: 1.114e-05
        [recompute_prepare]: 5.61e-06
        [updatestate_depend_eliminate]: 2.86999e-06
        [updatestate_assign_eliminate]: 2.10002e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.32e-06
        [specialize_transform]: 5.42001e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.52001e-06
        [meta_shard_fg_expand]: 1.09998e-06
        [get_grad_eliminate_]: 4.66002e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 2.31e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.422e-05
        [j_node_and_user_rematch]: 8.07e-06
        [meta_fg_expand]: 1.73002e-06
        [replace_old_param]: 7.95e-06
        [inline_without_move]: 4.83001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 7.99977e-07
        [auto_monad_eliminator]: 5.06002e-06
        [cse]: 1.168e-05
        [replace_applicator]: 5.78997e-06
[py_interpret_to_execute_after_opt_a]: 1.339e-05
[rewriter_after_opt_a]: 3.827e-05
[convert_after_rewriter]: 6.86001e-06
[order_py_execute_after_rewriter]: 4.94e-06
[mutable_eliminate]: 0.0007685
[jit_opt_b]: 6.249e-05, [1]
    [Cycle 1]: 5.344e-05, [2]
        [frontend_op_eliminate]: 2.473e-05
        [inline_after_opt_a]: 1.625e-05
[cconv]: 2.826e-05
[loop_unroll]: 0.00054203
[jit_opt_after_cconv]: 0.00015317, [1]
    [Cycle 1]: 0.00014648, [11]
        [c_1]: 2.32e-05
        [parameter_eliminate]: 3.06001e-06
        [updatestate_depend_eliminate]: 6.02001e-06
        [updatestate_assign_eliminate]: 2.69999e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [cse]: 2.447e-05
        [call_graph_tuple_transform]: 2.061e-05
        [tuple_list_get_item_eliminator]: 5.59e-06
        [none_parameter_eliminate]: 1.74e-06
        [renormalize]: 8.39995e-07
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 1.396e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 5.37001e-06
[add_recomputation]: 5.703e-05
[cse_after_recomputation]: 2.156e-05, [1]
    [Cycle 1]: 1.566e-05, [1]
        [cse]: 9.05999e-06
[auto_monad_reorder]: 1.602e-05
[get_jit_bprop_graph]: 2.49001e-06
[rewriter_after_jit_bprop_graph]: 7.29001e-06
[opt_after_jit_grad]: 0.00052059
[symbol_engine_optimizer]: 8.234e-05, [1]
    [Cycle 1]: 7.481e-05, [6]
        [build]: 4.57998e-06
        [elim_shapecalc]: 8.38999e-06
        [elim_not_effective]: 1.491e-05
        [opt_reshape]: 7.21001e-06
        [fold_const_symbol]: 9.62999e-06
        [renormalize]: 9.49978e-07
[validate]: 4.415e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.0236251
[execute]: 1.132e-05
Sums
    bootstrap                            :     0.000448s :  1.30%
    type_inference                       :     0.005668s : 16.50%
    event_method                         :     0.000020s :  0.06%
    auto_monad                           :     0.000069s :  0.20%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000052s :  0.15%
    rewriter_before_opt_a                :     0.000083s :  0.24%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000076s :  0.22%
    jit_opt_a.loop_unroll                :     0.000036s :  0.11%
    jit_opt_a.a_1                        :     0.000727s :  2.11%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.03%
    jit_opt_a.merge_forward              :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000018s :  0.05%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.000963s :  2.80%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.06%
    jit_opt_a.cse                        :     0.000042s :  0.12%
    jit_opt_a.replace_applicator         :     0.000021s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.04%
    rewriter_after_opt_a                 :     0.000038s :  0.11%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000769s :  2.24%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.05%
    cconv                                :     0.000028s :  0.08%
    loop_unroll                          :     0.000542s :  1.58%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000014s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000057s :  0.17%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000521s :  1.52%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.023625s : 68.76%
    execute                              :     0.000011s :  0.03%


Time group info:
------[substitution.]   0.000268    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000002s :      2: substitution.fold_const_symbol
 2.15% :     0.000006s :      3: substitution.graph_param_transform
85.75% :     0.000230s :      4: substitution.inline
 1.61% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.02% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.89% :     0.000005s :      2: substitution.replace_old_param
 5.14% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005581     2
81.38% :     0.004542s :      1: type_inference.infer
18.62% :     0.001039s :      1: type_inference.specialize
------[replace.]   0.000087     6
67.44% :     0.000059s :      4: replace.inline
32.56% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000239     6
94.89% :     0.000227s :      4: match.inline
 5.11% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.55% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.56% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000002s :     10: predicate.addn_zero_filter
 2.76% :     0.000004s :     10: predicate.arithmetic_simplify
 1.41% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.47% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.00% :     0.000001s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.88% :     0.000008s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.73% :     0.000002s :      6: predicate.less_batch_normalization
 1.30% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.60% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.79% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.90% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.04% :     0.000003s :     15: predicate.partial_eliminate
 1.32% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 2.08% :     0.000003s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.54% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.58% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.41% :     0.000002s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.21% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.54% :     0.000010s :     53: predicate.switch_simplify
 1.32% :     0.000002s :     10: predicate.tile_eliminate
 1.04% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.75% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.46% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.57% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.94% :     0.000003s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.88% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000731    12
47.89% :     0.000350s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.19% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.92% :     0.000336s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.041371    76
 0.15% :     0.000060s :      1: add_recomputation
 0.18% :     0.000073s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.14% :     0.000471s :      1: bootstrap
 0.07% :     0.000031s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.06% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.06% :     0.000025s :      1: event_method
 0.04% :     0.000018s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
16.01% :     0.006622s :      1: jit_opt_a
 0.38% :     0.000156s :      1: jit_opt_after_cconv
 0.16% :     0.000065s :      1: jit_opt_b
 1.33% :     0.000552s :      1: loop_unroll
 1.88% :     0.000779s :      1: mutable_eliminate
 2.33% :     0.000966s :     26: opt.transform.jit_opt_a
 0.12% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000034s :      4: opt.transform.jit_opt_b
 0.03% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.28% :     0.000531s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.20% :     0.000084s :      1: pre_auto_parallel
 0.13% :     0.000055s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000016s :      1: remove_dup_value
 1.21% :     0.000499s :      1: renormalize.infer
 1.10% :     0.000455s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000042s :      1: rewriter_after_opt_a
 0.21% :     0.000087s :      1: rewriter_before_opt_a
 0.21% :     0.000085s :      1: symbol_engine_optimizer
57.16% :     0.023649s :      1: task_emit
13.75% :     0.005688s :      1: type_inference
 0.16% :     0.000068s :      1: validate


TotalTime = 0.0304169, [33]
[bootstrap]: 0.00035914
[type_inference]: 0.00832555
[event_method]: 1.992e-05
[auto_monad]: 6.958e-05
[graph_reusing]: 6.26e-06
[pre_auto_parallel]: 2.54001e-06
[py_interpret_to_execute]: 4.327e-05
[rewriter_before_opt_a]: 7.56e-05
[expand_dump_flag]: 3.45e-06
[jit_opt_a]: 0.00954022, [2]
    [Cycle 1]: 0.00486305, [27]
        [switch_simplify]: 6.97e-05
        [loop_unroll]: 2.99e-05
        [a_1]: 0.0006457
        [with_stream_mark]: 4.598e-05
        [recompute_prepare]: 3.136e-05
        [updatestate_depend_eliminate]: 5.63002e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [parameter_eliminate]: 2.44999e-06
        [specialize_transform]: 6.96999e-06
        [updatestate_useless_node_eliminater]: 6.76999e-06
        [accelerated_algorithm]: 7.14001e-06
        [meta_shard_fg_expand]: 3.38e-06
        [get_grad_eliminate_]: 5.94e-06
        [merge_forward]: 4.92e-06
        [cell_reuse_recompute_pass]: 3.30003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.593e-05
        [j_node_and_user_rematch]: 1.091e-05
        [meta_fg_expand]: 2.84999e-06
        [replace_old_param]: 1.259e-05
        [inline_without_move]: 5.83997e-06
        [renormalize]: 0.00114073
        [add_forward_monad_depend]: 7.86001e-06
        [auto_monad_grad]: 2.71e-06
        [auto_monad_eliminator]: 1.931e-05
        [cse]: 3.32e-05
        [replace_applicator]: 1.6e-05
    [Cycle 2]: 0.00035982, [27]
        [switch_simplify]: 7.02002e-06
        [loop_unroll]: 5.52001e-06
        [a_1]: 0.00010986
        [with_stream_mark]: 1.161e-05
        [recompute_prepare]: 5.24e-06
        [updatestate_depend_eliminate]: 3.26001e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [parameter_eliminate]: 1.36002e-06
        [specialize_transform]: 5.13002e-06
        [updatestate_useless_node_eliminater]: 5.07999e-06
        [accelerated_algorithm]: 5.94e-06
        [meta_shard_fg_expand]: 1.43002e-06
        [get_grad_eliminate_]: 4.63001e-06
        [merge_forward]: 3.03e-06
        [cell_reuse_recompute_pass]: 2.66999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.365e-05
        [j_node_and_user_rematch]: 8.25e-06
        [meta_fg_expand]: 1.96003e-06
        [replace_old_param]: 7.1e-06
        [inline_without_move]: 4.91002e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 6.93e-06
        [cse]: 1.101e-05
        [replace_applicator]: 5.62999e-06
[py_interpret_to_execute_after_opt_a]: 1.671e-05
[rewriter_after_opt_a]: 4.32e-05
[convert_after_rewriter]: 7.71999e-06
[order_py_execute_after_rewriter]: 5.47999e-06
[mutable_eliminate]: 0.00092922
[jit_opt_b]: 6.195e-05, [1]
    [Cycle 1]: 5.223e-05, [2]
        [frontend_op_eliminate]: 2.05e-05
        [inline_after_opt_a]: 1.901e-05
[cconv]: 3.288e-05
[loop_unroll]: 0.00048407
[jit_opt_after_cconv]: 0.00016002, [1]
    [Cycle 1]: 0.00015257, [11]
        [c_1]: 2.426e-05
        [parameter_eliminate]: 3.65998e-06
        [updatestate_depend_eliminate]: 7.54002e-06
        [updatestate_assign_eliminate]: 2.50002e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 2.51e-05
        [call_graph_tuple_transform]: 2.307e-05
        [tuple_list_get_item_eliminator]: 5.05999e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 5.73002e-06
[remove_dup_value]: 1.672e-05
[partial_unused_args_eliminate]: 2.53e-06
[environ_conv]: 7.31001e-06
[add_recomputation]: 5.353e-05
[cse_after_recomputation]: 2.145e-05, [1]
    [Cycle 1]: 1.516e-05, [1]
        [cse]: 9.07999e-06
[auto_monad_reorder]: 1.583e-05
[get_jit_bprop_graph]: 2.59001e-06
[rewriter_after_jit_bprop_graph]: 6.18998e-06
[opt_after_jit_grad]: 0.0004781
[symbol_engine_optimizer]: 0.00012161, [1]
    [Cycle 1]: 0.00011403, [6]
        [build]: 4.60999e-06
        [elim_shapecalc]: 4.867e-05
        [elim_not_effective]: 1.441e-05
        [opt_reshape]: 6.46999e-06
        [fold_const_symbol]: 8.72e-06
        [renormalize]: 6.50005e-07
[validate]: 4.652e-05
[backend_pass]: 1.19998e-06
[task_emit]: 0.00921639
[execute]: 8.76002e-06
Sums
    bootstrap                            :     0.000359s :  1.57%
    type_inference                       :     0.008326s : 36.35%
    event_method                         :     0.000020s :  0.09%
    auto_monad                           :     0.000070s :  0.30%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.19%
    rewriter_before_opt_a                :     0.000076s :  0.33%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000077s :  0.33%
    jit_opt_a.loop_unroll                :     0.000035s :  0.15%
    jit_opt_a.a_1                        :     0.000756s :  3.30%
    jit_opt_a.with_stream_mark           :     0.000058s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000037s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.001141s :  4.98%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.11%
    jit_opt_a.cse                        :     0.000044s :  0.19%
    jit_opt_a.replace_applicator         :     0.000022s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.07%
    rewriter_after_opt_a                 :     0.000043s :  0.19%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000929s :  4.06%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.08%
    cconv                                :     0.000033s :  0.14%
    loop_unroll                          :     0.000484s :  2.11%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.07%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000054s :  0.23%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.07%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000478s :  2.09%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000049s :  0.21%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000047s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.009216s : 40.24%
    execute                              :     0.000009s :  0.04%


Time group info:
------[substitution.]   0.000287    23
 0.64% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.14% :     0.000006s :      3: substitution.graph_param_transform
85.98% :     0.000247s :      4: substitution.inline
 1.43% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.44% :     0.000007s :      2: substitution.replace_old_param
 4.88% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008237     2
87.30% :     0.007191s :      1: type_inference.infer
12.70% :     0.001046s :      1: type_inference.specialize
------[replace.]   0.000078     6
71.34% :     0.000056s :      4: replace.inline
28.66% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000257     6
94.98% :     0.000244s :      4: match.inline
 5.02% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.15% :     0.000002s :     10: predicate.addn_zero_filter
 2.30% :     0.000003s :     10: predicate.arithmetic_simplify
 1.44% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.93% :     0.000001s :     10: predicate.compare_switch_simplify
 2.10% :     0.000003s :     10: predicate.depend_value_elim
 0.93% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.05% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.70% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.98% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.16% :     0.000002s :      6: predicate.get_grad_eliminate
 0.35% :     0.000000s :      3: predicate.graph_param_transform
 5.62% :     0.000008s :     22: predicate.inline
 0.78% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 2.16% :     0.000003s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.51% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.72% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 0.89% :     0.000001s :     10: predicate.minmaximum_grad
 2.09% :     0.000003s :      3: predicate.mutable_eliminate
 0.57% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     15: predicate.partial_eliminate
 0.97% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.78% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.83% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.57% :     0.000001s :      3: predicate.reset_defer_inline
 1.16% :     0.000002s :     10: predicate.reshape_eliminate
 1.49% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     10: predicate.same_eliminate
 1.35% :     0.000002s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000002s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.39% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.32% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.14% :     0.000003s :     16: predicate.switch_defer_inline
 1.79% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.94% :     0.000010s :     53: predicate.switch_simplify
 1.36% :     0.000002s :     10: predicate.tile_eliminate
 1.28% :     0.000002s :     10: predicate.transpose_eliminate
 1.62% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.90% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.55% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.83% :     0.000003s :     10: predicate.value_based_eliminate
 0.26% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.87% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000803    12
45.41% :     0.000365s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.26% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.33% :     0.000380s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.032738    76
 0.17% :     0.000056s :      1: add_recomputation
 0.22% :     0.000073s :      1: auto_monad
 0.06% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.16% :     0.000381s :      1: bootstrap
 0.11% :     0.000036s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.07% :     0.000024s :      1: event_method
 0.04% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
29.15% :     0.009545s :      1: jit_opt_a
 0.50% :     0.000164s :      1: jit_opt_after_cconv
 0.20% :     0.000065s :      1: jit_opt_b
 1.51% :     0.000493s :      1: loop_unroll
 2.87% :     0.000941s :      1: mutable_eliminate
 3.13% :     0.001025s :     26: opt.transform.jit_opt_a
 0.17% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000032s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000074s :      4: opt.transform.symbol_engine_opt
 1.49% :     0.000487s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.14% :     0.000046s :      1: py_interpret_to_execute
 0.06% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000019s :      1: remove_dup_value
 1.96% :     0.000643s :      1: renormalize.infer
 1.48% :     0.000486s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000047s :      1: rewriter_after_opt_a
 0.24% :     0.000079s :      1: rewriter_before_opt_a
 0.38% :     0.000124s :      1: symbol_engine_optimizer
28.20% :     0.009232s :      1: task_emit
25.50% :     0.008347s :      1: type_inference
 0.21% :     0.000069s :      1: validate


TotalTime = 0.0240949, [33]
[bootstrap]: 0.00049029
[type_inference]: 0.00679808
[event_method]: 2.173e-05
[auto_monad]: 7.51e-05
[graph_reusing]: 5.94e-06
[pre_auto_parallel]: 3.18998e-06
[py_interpret_to_execute]: 4.879e-05
[rewriter_before_opt_a]: 9.219e-05
[expand_dump_flag]: 3.64002e-06
[jit_opt_a]: 0.00701786, [2]
    [Cycle 1]: 0.00211158, [27]
        [switch_simplify]: 6.301e-05
        [loop_unroll]: 3.119e-05
        [a_1]: 0.00063077
        [with_stream_mark]: 1.964e-05
        [recompute_prepare]: 8.25e-06
        [updatestate_depend_eliminate]: 3.50998e-06
        [updatestate_assign_eliminate]: 3.26001e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [parameter_eliminate]: 2.44001e-06
        [specialize_transform]: 5.99999e-06
        [updatestate_useless_node_eliminater]: 5.14998e-06
        [accelerated_algorithm]: 5.90002e-06
        [meta_shard_fg_expand]: 2.32001e-06
        [get_grad_eliminate_]: 5.92001e-06
        [merge_forward]: 3.77998e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.552e-05
        [j_node_and_user_rematch]: 1.002e-05
        [meta_fg_expand]: 2.50002e-06
        [replace_old_param]: 8.95001e-06
        [inline_without_move]: 5.96e-06
        [renormalize]: 0.00100935
        [add_forward_monad_depend]: 1.258e-05
        [auto_monad_grad]: 3.09001e-06
        [auto_monad_eliminator]: 1.635e-05
        [cse]: 3.247e-05
        [replace_applicator]: 1.46e-05
    [Cycle 2]: 0.00034367, [27]
        [switch_simplify]: 5.72999e-06
        [loop_unroll]: 5.00001e-06
        [a_1]: 9.667e-05
        [with_stream_mark]: 1.345e-05
        [recompute_prepare]: 5.75001e-06
        [updatestate_depend_eliminate]: 2.73e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [parameter_eliminate]: 1.45999e-06
        [specialize_transform]: 4.80001e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 4.99e-06
        [meta_shard_fg_expand]: 2.12001e-06
        [get_grad_eliminate_]: 4.64998e-06
        [merge_forward]: 2.84001e-06
        [cell_reuse_recompute_pass]: 2.79999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.484e-05
        [j_node_and_user_rematch]: 8.3e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 8.28999e-06
        [inline_without_move]: 4.67998e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.72001e-06
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 5.81e-06
        [cse]: 1.173e-05
        [replace_applicator]: 5.76e-06
[py_interpret_to_execute_after_opt_a]: 1.366e-05
[rewriter_after_opt_a]: 3.941e-05
[convert_after_rewriter]: 8.32e-06
[order_py_execute_after_rewriter]: 5.10001e-06
[mutable_eliminate]: 0.0008253
[jit_opt_b]: 6.114e-05, [1]
    [Cycle 1]: 5.202e-05, [2]
        [frontend_op_eliminate]: 2.371e-05
        [inline_after_opt_a]: 1.683e-05
[cconv]: 3.425e-05
[loop_unroll]: 0.00050455
[jit_opt_after_cconv]: 0.00016521, [1]
    [Cycle 1]: 0.00015726, [11]
        [c_1]: 2.489e-05
        [parameter_eliminate]: 4.37e-06
        [updatestate_depend_eliminate]: 7.41001e-06
        [updatestate_assign_eliminate]: 3.14999e-06
        [updatestate_loads_eliminate]: 2.84999e-06
        [cse]: 2.326e-05
        [call_graph_tuple_transform]: 2.195e-05
        [tuple_list_get_item_eliminator]: 6.79999e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 6.29001e-06
[remove_dup_value]: 1.582e-05
[partial_unused_args_eliminate]: 2.78e-06
[environ_conv]: 5.89e-06
[add_recomputation]: 5.65e-05
[cse_after_recomputation]: 2.297e-05, [1]
    [Cycle 1]: 1.664e-05, [1]
        [cse]: 9.86998e-06
[auto_monad_reorder]: 1.593e-05
[get_jit_bprop_graph]: 2.26998e-06
[rewriter_after_jit_bprop_graph]: 7.48999e-06
[opt_after_jit_grad]: 0.00050149
[symbol_engine_optimizer]: 7.973e-05, [1]
    [Cycle 1]: 7.179e-05, [6]
        [build]: 3.56001e-06
        [elim_shapecalc]: 7.83999e-06
        [elim_not_effective]: 1.464e-05
        [opt_reshape]: 6.35002e-06
        [fold_const_symbol]: 9.74e-06
        [renormalize]: 8.29983e-07
[validate]: 4.333e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.00686721
[execute]: 8.02e-06
Sums
    bootstrap                            :     0.000490s :  2.60%
    type_inference                       :     0.006798s : 36.07%
    event_method                         :     0.000022s :  0.12%
    auto_monad                           :     0.000075s :  0.40%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000049s :  0.26%
    rewriter_before_opt_a                :     0.000092s :  0.49%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000069s :  0.36%
    jit_opt_a.loop_unroll                :     0.000036s :  0.19%
    jit_opt_a.a_1                        :     0.000727s :  3.86%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001009s :  5.36%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.12%
    jit_opt_a.cse                        :     0.000044s :  0.23%
    jit_opt_a.replace_applicator         :     0.000020s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.07%
    rewriter_after_opt_a                 :     0.000039s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000825s :  4.38%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000034s :  0.18%
    loop_unroll                          :     0.000505s :  2.68%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000056s :  0.30%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000501s :  2.66%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.23%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006867s : 36.44%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000268    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000002s :      2: substitution.fold_const_symbol
 2.48% :     0.000007s :      3: substitution.graph_param_transform
85.99% :     0.000231s :      4: substitution.inline
 1.46% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.62% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.84% :     0.000005s :      2: substitution.replace_old_param
 5.11% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006708     2
83.72% :     0.005616s :      1: type_inference.infer
16.28% :     0.001092s :      1: type_inference.specialize
------[replace.]   0.000082     6
66.91% :     0.000055s :      4: replace.inline
33.09% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000240     6
94.77% :     0.000228s :      4: match.inline
 5.23% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.23% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 2.93% :     0.000004s :     10: predicate.arithmetic_simplify
 1.15% :     0.000001s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.18% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.83% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.54% :     0.000007s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000002s :      6: predicate.less_batch_normalization
 1.36% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.91% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.95% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 2.31% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.54% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     10: predicate.reduce_eliminate
 1.33% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.93% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.66% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.51% :     0.000002s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.57% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.44% :     0.000009s :     53: predicate.switch_simplify
 1.42% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.42% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.80% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.91% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.81% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000769    12
46.79% :     0.000360s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.63% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.58% :     0.000358s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026171    76
 0.23% :     0.000060s :      1: add_recomputation
 0.30% :     0.000079s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.96% :     0.000514s :      1: bootstrap
 0.14% :     0.000037s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.10% :     0.000027s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.03% :     0.000007s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.83% :     0.007022s :      1: jit_opt_a
 0.64% :     0.000169s :      1: jit_opt_after_cconv
 0.24% :     0.000064s :      1: jit_opt_b
 1.96% :     0.000514s :      1: loop_unroll
 3.20% :     0.000836s :      1: mutable_eliminate
 3.65% :     0.000955s :     26: opt.transform.jit_opt_a
 0.21% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000033s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.95% :     0.000509s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000052s :      1: py_interpret_to_execute
 0.06% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.15% :     0.000562s :      1: renormalize.infer
 1.68% :     0.000439s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000043s :      1: rewriter_after_opt_a
 0.37% :     0.000096s :      1: rewriter_before_opt_a
 0.31% :     0.000082s :      1: symbol_engine_optimizer
26.31% :     0.006886s :      1: task_emit
26.07% :     0.006824s :      1: type_inference
 0.25% :     0.000065s :      1: validate


TotalTime = 0.0151409, [30]
[bootstrap]: 0.00026936
[type_inference]: 0.00665645
[event_method]: 2.199e-05
[auto_monad]: 8.333e-05
[graph_reusing]: 5.97999e-06
[pre_auto_parallel]: 2.65002e-06
[py_interpret_to_execute]: 4.891e-05
[rewriter_before_opt_a]: 8.667e-05
[expand_dump_flag]: 2.91e-06
[jit_opt_a]: 0.00573258, [2]
    [Cycle 1]: 0.00197693, [27]
        [switch_simplify]: 0.00012579
        [loop_unroll]: 3.377e-05
        [a_1]: 0.00070175
        [with_stream_mark]: 2.095e-05
        [recompute_prepare]: 7.30998e-06
        [updatestate_depend_eliminate]: 4.13999e-06
        [updatestate_assign_eliminate]: 3.52997e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 5.56998e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.18002e-06
        [meta_shard_fg_expand]: 2.04999e-06
        [get_grad_eliminate_]: 5.71e-06
        [merge_forward]: 3.56001e-06
        [cell_reuse_recompute_pass]: 1.00999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.479e-05
        [j_node_and_user_rematch]: 9.41e-06
        [meta_fg_expand]: 2.49001e-06
        [replace_old_param]: 8.92999e-06
        [inline_without_move]: 5.17e-06
        [renormalize]: 0.00075609
        [add_forward_monad_depend]: 5.79e-06
        [auto_monad_grad]: 2.65002e-06
        [auto_monad_eliminator]: 1.333e-05
        [cse]: 3.338e-05
        [replace_applicator]: 1.306e-05
    [Cycle 2]: 0.00031774, [27]
        [switch_simplify]: 5.66998e-06
        [loop_unroll]: 4.89e-06
        [a_1]: 9.308e-05
        [with_stream_mark]: 1.099e-05
        [recompute_prepare]: 5.57001e-06
        [updatestate_depend_eliminate]: 2.64999e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 5.44998e-06
        [updatestate_useless_node_eliminater]: 5.31002e-06
        [accelerated_algorithm]: 4.43001e-06
        [meta_shard_fg_expand]: 1.22999e-06
        [get_grad_eliminate_]: 4.38999e-06
        [merge_forward]: 2.69001e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.269e-05
        [j_node_and_user_rematch]: 7.66999e-06
        [meta_fg_expand]: 1.85001e-06
        [replace_old_param]: 7.20998e-06
        [inline_without_move]: 4.70001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.09998e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 4.42e-06
        [cse]: 9.70002e-06
        [replace_applicator]: 5.32999e-06
[py_interpret_to_execute_after_opt_a]: 1.028e-05
[rewriter_after_opt_a]: 3.696e-05
[convert_after_rewriter]: 6.42001e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00056092
[jit_opt_b]: 4.795e-05, [1]
    [Cycle 1]: 4.103e-05, [2]
        [frontend_op_eliminate]: 1.523e-05
        [inline_after_opt_a]: 1.44e-05
[cconv]: 2.545e-05
[loop_unroll]: 0.00043464
[jit_opt_after_cconv]: 0.00022465, [1]
    [Cycle 1]: 0.00021784, [11]
        [c_1]: 2.135e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 5.58002e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 1.98002e-06
        [cse]: 1.864e-05
        [call_graph_tuple_transform]: 9.687e-05
        [tuple_list_get_item_eliminator]: 6.24999e-06
        [none_parameter_eliminate]: 2.19001e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 5.47999e-06
[remove_dup_value]: 1.489e-05
[partial_unused_args_eliminate]: 2.64001e-06
[environ_conv]: 5.15001e-06
[add_recomputation]: 5.008e-05
[cse_after_recomputation]: 2.356e-05, [1]
    [Cycle 1]: 1.747e-05, [1]
        [cse]: 1.155e-05
[auto_monad_reorder]: 1.493e-05
[get_jit_bprop_graph]: 2.53e-06
[rewriter_after_jit_bprop_graph]: 5.87999e-06
[opt_after_jit_grad]: 0.00046272
[symbol_engine_optimizer]: 7.507e-05, [1]
    [Cycle 1]: 6.87e-05, [6]
        [build]: 4.50999e-06
        [elim_shapecalc]: 7.83999e-06
        [elim_not_effective]: 1.261e-05
        [opt_reshape]: 5.80002e-06
        [fold_const_symbol]: 9.09e-06
        [renormalize]: 6.89994e-07
[validate]: 3.569e-05
Sums
    bootstrap                            :     0.000269s :  2.43%
    type_inference                       :     0.006656s : 59.97%
    event_method                         :     0.000022s :  0.20%
    auto_monad                           :     0.000083s :  0.75%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000049s :  0.44%
    rewriter_before_opt_a                :     0.000087s :  0.78%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000131s :  1.18%
    jit_opt_a.loop_unroll                :     0.000039s :  0.35%
    jit_opt_a.a_1                        :     0.000795s :  7.16%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000756s :  6.81%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.16%
    jit_opt_a.cse                        :     0.000043s :  0.39%
    jit_opt_a.replace_applicator         :     0.000018s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.09%
    rewriter_after_opt_a                 :     0.000037s :  0.33%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000561s :  5.05%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.13%
    cconv                                :     0.000025s :  0.23%
    loop_unroll                          :     0.000435s :  3.92%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000097s :  0.87%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000015s :  0.13%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000050s :  0.45%
    cse_after_recomputation.cse          :     0.000012s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.13%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000463s :  4.17%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000036s :  0.32%


Time group info:
------[substitution.]   0.000303    26
 0.64% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000002s :      2: substitution.fold_const_symbol
 2.04% :     0.000006s :      3: substitution.graph_param_transform
85.43% :     0.000259s :      6: substitution.inline
 1.06% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.50% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.37% :     0.000004s :      2: substitution.replace_old_param
 3.15% :     0.000010s :      1: substitution.switch_simplify
 4.20% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006581     2
82.33% :     0.005418s :      1: type_inference.infer
17.67% :     0.001163s :      1: type_inference.specialize
------[replace.]   0.000137     9
47.62% :     0.000065s :      6: replace.inline
30.35% :     0.000042s :      1: replace.switch_simplify
22.02% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000276     9
92.63% :     0.000256s :      6: match.inline
 3.15% :     0.000009s :      1: match.switch_simplify
 4.22% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   801
 1.31% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     12: predicate.addn_check_dump
 1.46% :     0.000002s :     12: predicate.addn_zero_filter
 1.95% :     0.000003s :     12: predicate.arithmetic_simplify
 1.35% :     0.000002s :     12: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     12: predicate.compare_switch_simplify
 1.44% :     0.000002s :     12: predicate.depend_value_elim
 1.26% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     12: predicate.environ_get_eliminate
 1.15% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.52% :     0.000007s :     26: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.93% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     17: predicate.load_eliminater
 1.22% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.26% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.96% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000002s :     12: predicate.merge_addn
 1.25% :     0.000002s :     12: predicate.minmaximum_grad
 1.33% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     17: predicate.partial_eliminate
 1.24% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     12: predicate.reduce_eliminate
 1.55% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     20: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.60% :     0.000002s :     12: predicate.reshape_eliminate
 1.23% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     12: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      6: predicate.special_op_eliminate
 1.15% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.64% :     0.000003s :     20: predicate.switch_defer_inline
 2.33% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.21% :     0.000011s :     65: predicate.switch_simplify
 1.29% :     0.000002s :     12: predicate.tile_eliminate
 1.43% :     0.000002s :     12: predicate.transpose_eliminate
 1.42% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.75% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     12: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000821    15
47.28% :     0.000388s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.70% :     0.000112s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.02% :     0.000320s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017099    72
 0.31% :     0.000053s :      1: add_recomputation
 0.51% :     0.000087s :      1: auto_monad
 0.10% :     0.000017s :      1: auto_monad_reorder
 1.68% :     0.000288s :      1: bootstrap
 0.17% :     0.000028s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000026s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
33.55% :     0.005736s :      1: jit_opt_a
 1.33% :     0.000228s :      1: jit_opt_after_cconv
 0.29% :     0.000050s :      1: jit_opt_b
 2.59% :     0.000442s :      1: loop_unroll
 3.33% :     0.000569s :      1: mutable_eliminate
 6.31% :     0.001078s :     26: opt.transform.jit_opt_a
 0.74% :     0.000126s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000023s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.76% :     0.000471s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000052s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.54% :     0.000434s :      1: renormalize.infer
 1.84% :     0.000314s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000040s :      1: rewriter_after_opt_a
 0.52% :     0.000090s :      1: rewriter_before_opt_a
 0.45% :     0.000078s :      1: symbol_engine_optimizer
39.03% :     0.006674s :      1: type_inference


TotalTime = 0.0290535, [33]
[bootstrap]: 0.00038623
[type_inference]: 0.00802029
[event_method]: 1.884e-05
[auto_monad]: 6.69e-05
[graph_reusing]: 5.64e-06
[pre_auto_parallel]: 2.27999e-06
[py_interpret_to_execute]: 4.243e-05
[rewriter_before_opt_a]: 7.384e-05
[expand_dump_flag]: 3.08e-06
[jit_opt_a]: 0.00704289, [2]
    [Cycle 1]: 0.00240746, [27]
        [switch_simplify]: 6.888e-05
        [loop_unroll]: 2.924e-05
        [a_1]: 0.00065738
        [with_stream_mark]: 2.731e-05
        [recompute_prepare]: 9.89999e-06
        [updatestate_depend_eliminate]: 4.37e-06
        [updatestate_assign_eliminate]: 2.91999e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [parameter_eliminate]: 2.12999e-06
        [specialize_transform]: 7.55e-06
        [updatestate_useless_node_eliminater]: 5.42001e-06
        [accelerated_algorithm]: 6.25002e-06
        [meta_shard_fg_expand]: 1.95001e-06
        [get_grad_eliminate_]: 5.40001e-06
        [merge_forward]: 4.27e-06
        [cell_reuse_recompute_pass]: 1.72999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.936e-05
        [j_node_and_user_rematch]: 1.143e-05
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 1.145e-05
        [inline_without_move]: 5.81e-06
        [renormalize]: 0.0012008
        [add_forward_monad_depend]: 1.034e-05
        [auto_monad_grad]: 3.35e-06
        [auto_monad_eliminator]: 2.063e-05
        [cse]: 3.601e-05
        [replace_applicator]: 2.372e-05
    [Cycle 2]: 0.00036968, [27]
        [switch_simplify]: 5.63997e-06
        [loop_unroll]: 5.01002e-06
        [a_1]: 0.00010236
        [with_stream_mark]: 1.798e-05
        [recompute_prepare]: 5.05001e-06
        [updatestate_depend_eliminate]: 3.2e-06
        [updatestate_assign_eliminate]: 2.82002e-06
        [updatestate_loads_eliminate]: 2.32001e-06
        [parameter_eliminate]: 2.26e-06
        [specialize_transform]: 5.18002e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 6.56999e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 4.27e-06
        [merge_forward]: 2.95998e-06
        [cell_reuse_recompute_pass]: 3.95e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.897e-05
        [j_node_and_user_rematch]: 9.10001e-06
        [meta_fg_expand]: 2.19999e-06
        [replace_old_param]: 7.74002e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.55001e-06
        [auto_monad_grad]: 1.44e-06
        [auto_monad_eliminator]: 5.88002e-06
        [cse]: 1.127e-05
        [replace_applicator]: 5.89999e-06
[py_interpret_to_execute_after_opt_a]: 1.567e-05
[rewriter_after_opt_a]: 4.28e-05
[convert_after_rewriter]: 8.35001e-06
[order_py_execute_after_rewriter]: 5.30001e-06
[mutable_eliminate]: 0.0022354
[jit_opt_b]: 8.378e-05, [1]
    [Cycle 1]: 5.792e-05, [2]
        [frontend_op_eliminate]: 2.489e-05
        [inline_after_opt_a]: 1.923e-05
[cconv]: 4.435e-05
[loop_unroll]: 0.00063915
[jit_opt_after_cconv]: 0.00019009, [1]
    [Cycle 1]: 0.00018267, [11]
        [c_1]: 2.434e-05
        [parameter_eliminate]: 6.73e-06
        [updatestate_depend_eliminate]: 1.184e-05
        [updatestate_assign_eliminate]: 3.81001e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [cse]: 3.925e-05
        [call_graph_tuple_transform]: 2.736e-05
        [tuple_list_get_item_eliminator]: 6.24999e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 9.30013e-07
        [switch_simplify]: 5.91998e-06
[remove_dup_value]: 1.689e-05
[partial_unused_args_eliminate]: 2.32001e-06
[environ_conv]: 8.23999e-06
[add_recomputation]: 5.91e-05
[cse_after_recomputation]: 2.327e-05, [1]
    [Cycle 1]: 1.692e-05, [1]
        [cse]: 1.036e-05
[auto_monad_reorder]: 1.62e-05
[get_jit_bprop_graph]: 2.73e-06
[rewriter_after_jit_bprop_graph]: 1.047e-05
[opt_after_jit_grad]: 0.00056314
[symbol_engine_optimizer]: 9.386e-05, [1]
    [Cycle 1]: 8.453e-05, [6]
        [build]: 5.32001e-06
        [elim_shapecalc]: 8.95001e-06
        [elim_not_effective]: 1.502e-05
        [opt_reshape]: 6.94001e-06
        [fold_const_symbol]: 9.82999e-06
        [renormalize]: 6.39993e-07
[validate]: 4.945e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00901802
[execute]: 9.79e-06
Sums
    bootstrap                            :     0.000386s :  1.61%
    type_inference                       :     0.008020s : 33.38%
    event_method                         :     0.000019s :  0.08%
    auto_monad                           :     0.000067s :  0.28%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.18%
    rewriter_before_opt_a                :     0.000074s :  0.31%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000075s :  0.31%
    jit_opt_a.loop_unroll                :     0.000034s :  0.14%
    jit_opt_a.a_1                        :     0.000760s :  3.16%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.04%
    jit_opt_a.renormalize                :     0.001201s :  5.00%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.11%
    jit_opt_a.cse                        :     0.000047s :  0.20%
    jit_opt_a.replace_applicator         :     0.000030s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.07%
    rewriter_after_opt_a                 :     0.000043s :  0.18%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.002235s :  9.30%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.08%
    cconv                                :     0.000044s :  0.18%
    loop_unroll                          :     0.000639s :  2.66%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000039s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000017s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000008s :  0.03%
    add_recomputation                    :     0.000059s :  0.25%
    cse_after_recomputation.cse          :     0.000010s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.07%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.04%
    opt_after_jit_grad                   :     0.000563s :  2.34%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000049s :  0.21%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009018s : 37.53%
    execute                              :     0.000010s :  0.04%


Time group info:
------[substitution.]   0.000301    23
 0.90% :     0.000003s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 2.09% :     0.000006s :      3: substitution.graph_param_transform
86.09% :     0.000259s :      4: substitution.inline
 1.89% :     0.000006s :      4: substitution.j_node_and_user_rematch
 1.84% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.14% :     0.000006s :      2: substitution.replace_old_param
 4.61% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007934     2
86.63% :     0.006873s :      1: type_inference.infer
13.37% :     0.001061s :      1: type_inference.specialize
------[replace.]   0.000089     6
67.24% :     0.000060s :      4: replace.inline
32.76% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000270     6
95.28% :     0.000257s :      4: match.inline
 4.72% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   691
 1.86% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.11% :     0.000001s :     10: predicate.addn_zero_filter
 2.44% :     0.000003s :     10: predicate.arithmetic_simplify
 1.16% :     0.000002s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.93% :     0.000001s :     10: predicate.compare_switch_simplify
 1.56% :     0.000002s :     10: predicate.depend_value_elim
 0.94% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.09% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.06% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.89% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.03% :     0.000001s :     10: predicate.environ_get_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.73% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 6.21% :     0.000008s :     22: predicate.inline
 0.78% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :      6: predicate.less_batch_normalization
 1.30% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 1.66% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.35% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.12% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.94% :     0.000001s :     10: predicate.merge_addn
 0.93% :     0.000001s :     10: predicate.minmaximum_grad
 4.09% :     0.000005s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 1.89% :     0.000003s :     15: predicate.partial_eliminate
 1.03% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.92% :     0.000003s :     12: predicate.redundant_stop_gradient_eliminater
 0.75% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.04% :     0.000003s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.60% :     0.000001s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.35% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.09% :     0.000001s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.25% :     0.000002s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.68% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.11% :     0.000003s :     16: predicate.switch_defer_inline
 1.72% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.30% :     0.000010s :     53: predicate.switch_simplify
 1.23% :     0.000002s :     10: predicate.tile_eliminate
 1.29% :     0.000002s :     10: predicate.transpose_eliminate
 1.27% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.28% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.42% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.47% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.38% :     0.000002s :     10: predicate.value_based_eliminate
 0.45% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000818    12
44.87% :     0.000367s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.36% :     0.000060s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.77% :     0.000391s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.031408    76
 0.20% :     0.000063s :      1: add_recomputation
 0.22% :     0.000070s :      1: auto_monad
 0.06% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.28% :     0.000401s :      1: bootstrap
 0.15% :     0.000047s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.08% :     0.000026s :      1: cse_after_recomputation
 0.04% :     0.000012s :      1: environ_conv
 0.07% :     0.000022s :      1: event_method
 0.05% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
22.44% :     0.007047s :      1: jit_opt_a
 0.61% :     0.000193s :      1: jit_opt_after_cconv
 0.28% :     0.000087s :      1: jit_opt_b
 2.07% :     0.000650s :      1: loop_unroll
 7.19% :     0.002259s :      1: mutable_eliminate
 3.22% :     0.001010s :     26: opt.transform.jit_opt_a
 0.19% :     0.000060s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000036s :      4: opt.transform.jit_opt_b
 0.05% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000036s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.83% :     0.000574s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.14% :     0.000045s :      1: py_interpret_to_execute
 0.06% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000019s :      1: remove_dup_value
 2.07% :     0.000650s :      1: renormalize.infer
 1.71% :     0.000537s :      1: renormalize.specialize
 0.04% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000047s :      1: rewriter_after_opt_a
 0.25% :     0.000077s :      1: rewriter_before_opt_a
 0.31% :     0.000096s :      1: symbol_engine_optimizer
28.78% :     0.009040s :      1: task_emit
25.59% :     0.008036s :      1: type_inference
 0.23% :     0.000074s :      1: validate


TotalTime = 0.0153931, [30]
[bootstrap]: 0.0002712
[type_inference]: 0.00667571
[event_method]: 2.297e-05
[auto_monad]: 7.982e-05
[graph_reusing]: 7.08e-06
[pre_auto_parallel]: 2.56e-06
[py_interpret_to_execute]: 4.974e-05
[rewriter_before_opt_a]: 8.681e-05
[expand_dump_flag]: 2.89999e-06
[jit_opt_a]: 0.00596164, [2]
    [Cycle 1]: 0.0021152, [27]
        [switch_simplify]: 0.00013154
        [loop_unroll]: 3.333e-05
        [a_1]: 0.00073813
        [with_stream_mark]: 2.391e-05
        [recompute_prepare]: 7.41999e-06
        [updatestate_depend_eliminate]: 4.29002e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 2.50002e-06
        [parameter_eliminate]: 2.05002e-06
        [specialize_transform]: 6.14999e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.45001e-06
        [meta_shard_fg_expand]: 2.47001e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 4.75999e-06
        [cell_reuse_recompute_pass]: 1.72001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.461e-05
        [j_node_and_user_rematch]: 1.016e-05
        [meta_fg_expand]: 2.51e-06
        [replace_old_param]: 9.41e-06
        [inline_without_move]: 5.47999e-06
        [renormalize]: 0.00083581
        [add_forward_monad_depend]: 6.38e-06
        [auto_monad_grad]: 2.57001e-06
        [auto_monad_eliminator]: 1.514e-05
        [cse]: 3.125e-05
        [replace_applicator]: 1.389e-05
    [Cycle 2]: 0.00032487, [27]
        [switch_simplify]: 5.41002e-06
        [loop_unroll]: 5.01997e-06
        [a_1]: 8.997e-05
        [with_stream_mark]: 1.023e-05
        [recompute_prepare]: 4.65999e-06
        [updatestate_depend_eliminate]: 2.63e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 1.95001e-06
        [parameter_eliminate]: 1.51998e-06
        [specialize_transform]: 4.90001e-06
        [updatestate_useless_node_eliminater]: 4.96997e-06
        [accelerated_algorithm]: 5.08002e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 4.38001e-06
        [merge_forward]: 3.04999e-06
        [cell_reuse_recompute_pass]: 1.60001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.371e-05
        [j_node_and_user_rematch]: 7.73001e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 7.18998e-06
        [inline_without_move]: 4.90999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.24998e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 5.31002e-06
        [cse]: 1.078e-05
        [replace_applicator]: 4.84998e-06
[py_interpret_to_execute_after_opt_a]: 1.318e-05
[rewriter_after_opt_a]: 3.682e-05
[convert_after_rewriter]: 7.26001e-06
[order_py_execute_after_rewriter]: 5.32999e-06
[mutable_eliminate]: 0.0006493
[jit_opt_b]: 5.025e-05, [1]
    [Cycle 1]: 4.228e-05, [2]
        [frontend_op_eliminate]: 1.603e-05
        [inline_after_opt_a]: 1.485e-05
[cconv]: 2.766e-05
[loop_unroll]: 0.00042724
[jit_opt_after_cconv]: 0.00014122, [1]
    [Cycle 1]: 0.00013508, [11]
        [c_1]: 2.151e-05
        [parameter_eliminate]: 3.02002e-06
        [updatestate_depend_eliminate]: 6.62002e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [cse]: 1.845e-05
        [call_graph_tuple_transform]: 1.887e-05
        [tuple_list_get_item_eliminator]: 5.34e-06
        [none_parameter_eliminate]: 1.39998e-06
        [renormalize]: 7.59988e-07
        [switch_simplify]: 5.52001e-06
[remove_dup_value]: 1.302e-05
[partial_unused_args_eliminate]: 2.21998e-06
[environ_conv]: 5.45001e-06
[add_recomputation]: 5.04e-05
[cse_after_recomputation]: 2.03e-05, [1]
    [Cycle 1]: 1.501e-05, [1]
        [cse]: 9.23002e-06
[auto_monad_reorder]: 1.46e-05
[get_jit_bprop_graph]: 2.07999e-06
[rewriter_after_jit_bprop_graph]: 6.27001e-06
[opt_after_jit_grad]: 0.00046111
[symbol_engine_optimizer]: 7.413e-05, [1]
    [Cycle 1]: 6.749e-05, [6]
        [build]: 3.98001e-06
        [elim_shapecalc]: 7.81001e-06
        [elim_not_effective]: 1.266e-05
        [opt_reshape]: 5.42999e-06
        [fold_const_symbol]: 8.79e-06
        [renormalize]: 7.59988e-07
[validate]: 4.041e-05
Sums
    bootstrap                            :     0.000271s :  2.41%
    type_inference                       :     0.006676s : 59.30%
    event_method                         :     0.000023s :  0.20%
    auto_monad                           :     0.000080s :  0.71%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000050s :  0.44%
    rewriter_before_opt_a                :     0.000087s :  0.77%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000137s :  1.22%
    jit_opt_a.loop_unroll                :     0.000038s :  0.34%
    jit_opt_a.a_1                        :     0.000828s :  7.36%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000008s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000836s :  7.42%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.18%
    jit_opt_a.cse                        :     0.000042s :  0.37%
    jit_opt_a.replace_applicator         :     0.000019s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.12%
    rewriter_after_opt_a                 :     0.000037s :  0.33%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000649s :  5.77%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.13%
    cconv                                :     0.000028s :  0.25%
    loop_unroll                          :     0.000427s :  3.80%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000013s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000050s :  0.45%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000015s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000461s :  4.10%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.36%


Time group info:
------[substitution.]   0.000307    26
 0.63% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 1.67% :     0.000005s :      3: substitution.graph_param_transform
85.79% :     0.000264s :      6: substitution.inline
 1.24% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.43% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.46% :     0.000004s :      2: substitution.replace_old_param
 2.99% :     0.000009s :      1: substitution.switch_simplify
 4.35% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006601     2
83.35% :     0.005502s :      1: type_inference.infer
16.65% :     0.001099s :      1: type_inference.specialize
------[replace.]   0.000144     9
49.14% :     0.000071s :      6: replace.inline
29.19% :     0.000042s :      1: replace.switch_simplify
21.67% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000281     9
92.62% :     0.000260s :      6: match.inline
 2.99% :     0.000008s :      1: match.switch_simplify
 4.38% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   801
 1.38% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.22% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.20% :     0.000002s :     12: predicate.addn_check_dump
 1.38% :     0.000002s :     12: predicate.addn_zero_filter
 2.00% :     0.000003s :     12: predicate.arithmetic_simplify
 1.38% :     0.000002s :     12: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000002s :     12: predicate.compare_switch_simplify
 1.48% :     0.000002s :     12: predicate.depend_value_elim
 1.14% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     12: predicate.environ_get_eliminate
 1.25% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.51% :     0.000007s :     26: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     17: predicate.load_eliminater
 1.56% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.93% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000002s :     12: predicate.merge_addn
 1.20% :     0.000002s :     12: predicate.minmaximum_grad
 1.24% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     17: predicate.partial_eliminate
 1.17% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.79% :     0.000002s :     12: predicate.reduce_eliminate
 1.52% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     20: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.44% :     0.000001s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     12: predicate.reshape_eliminate
 1.23% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.74% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.99% :     0.000004s :     20: predicate.switch_defer_inline
 2.33% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.31% :     0.000011s :     65: predicate.switch_simplify
 1.25% :     0.000002s :     12: predicate.tile_eliminate
 1.28% :     0.000002s :     12: predicate.transpose_eliminate
 1.46% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.88% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     12: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000826    15
48.37% :     0.000400s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.66% :     0.000113s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.97% :     0.000314s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017392    72
 0.31% :     0.000053s :      1: add_recomputation
 0.48% :     0.000084s :      1: auto_monad
 0.10% :     0.000017s :      1: auto_monad_reorder
 1.67% :     0.000290s :      1: bootstrap
 0.17% :     0.000030s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.13% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000028s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000010s :      1: graph_reusing
34.30% :     0.005965s :      1: jit_opt_a
 0.83% :     0.000144s :      1: jit_opt_after_cconv
 0.30% :     0.000053s :      1: jit_opt_b
 2.50% :     0.000435s :      1: loop_unroll
 3.79% :     0.000659s :      1: mutable_eliminate
 6.43% :     0.001119s :     26: opt.transform.jit_opt_a
 0.27% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.70% :     0.000470s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.30% :     0.000052s :      1: py_interpret_to_execute
 0.09% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 2.55% :     0.000444s :      1: renormalize.infer
 2.21% :     0.000384s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000040s :      1: rewriter_after_opt_a
 0.52% :     0.000090s :      1: rewriter_before_opt_a
 0.44% :     0.000077s :      1: symbol_engine_optimizer
38.48% :     0.006693s :      1: type_inference


TotalTime = 0.0791834, [33]
[bootstrap]: 0.00036098
[type_inference]: 0.0589821
[event_method]: 2.364e-05
[auto_monad]: 9.078e-05
[graph_reusing]: 6.51e-06
[pre_auto_parallel]: 3.11001e-06
[py_interpret_to_execute]: 7.514e-05
[rewriter_before_opt_a]: 8.49e-05
[expand_dump_flag]: 3.45e-06
[jit_opt_a]: 0.00745685, [2]
    [Cycle 1]: 0.00230884, [27]
        [switch_simplify]: 7.818e-05
        [loop_unroll]: 3.172e-05
        [a_1]: 0.00067683
        [with_stream_mark]: 3.011e-05
        [recompute_prepare]: 7.97e-06
        [updatestate_depend_eliminate]: 5.08002e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 2.36e-06
        [specialize_transform]: 6.96001e-06
        [updatestate_useless_node_eliminater]: 6.24999e-06
        [accelerated_algorithm]: 6.84999e-06
        [meta_shard_fg_expand]: 2.46e-06
        [get_grad_eliminate_]: 5.52999e-06
        [merge_forward]: 4.37e-06
        [cell_reuse_recompute_pass]: 1.99e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.654e-05
        [j_node_and_user_rematch]: 1.042e-05
        [meta_fg_expand]: 2.73e-06
        [replace_old_param]: 1.057e-05
        [inline_without_move]: 5.53997e-06
        [renormalize]: 0.00109723
        [add_forward_monad_depend]: 8.77e-06
        [auto_monad_grad]: 2.79999e-06
        [auto_monad_eliminator]: 1.93e-05
        [cse]: 3.535e-05
        [replace_applicator]: 2.013e-05
    [Cycle 2]: 0.00035531, [27]
        [switch_simplify]: 6.89999e-06
        [loop_unroll]: 6.01e-06
        [a_1]: 9.787e-05
        [with_stream_mark]: 1.461e-05
        [recompute_prepare]: 5.77999e-06
        [updatestate_depend_eliminate]: 3.26001e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 1.54e-06
        [specialize_transform]: 4.92e-06
        [updatestate_useless_node_eliminater]: 4.66002e-06
        [accelerated_algorithm]: 5.75001e-06
        [meta_shard_fg_expand]: 2.32999e-06
        [get_grad_eliminate_]: 4.63001e-06
        [merge_forward]: 3.15002e-06
        [cell_reuse_recompute_pass]: 3.88001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.606e-05
        [j_node_and_user_rematch]: 7.57998e-06
        [meta_fg_expand]: 2.12001e-06
        [replace_old_param]: 7.85998e-06
        [inline_without_move]: 4.57998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 5.85002e-06
        [cse]: 1.228e-05
        [replace_applicator]: 5.20999e-06
[py_interpret_to_execute_after_opt_a]: 1.608e-05
[rewriter_after_opt_a]: 0.00011214
[convert_after_rewriter]: 1.049e-05
[order_py_execute_after_rewriter]: 5.17999e-06
[mutable_eliminate]: 0.00094959
[jit_opt_b]: 6.37e-05, [1]
    [Cycle 1]: 5.35e-05, [2]
        [frontend_op_eliminate]: 2.08e-05
        [inline_after_opt_a]: 1.762e-05
[cconv]: 3.539e-05
[loop_unroll]: 0.0005789
[jit_opt_after_cconv]: 0.00019304, [1]
    [Cycle 1]: 0.00018591, [11]
        [c_1]: 2.457e-05
        [parameter_eliminate]: 5.84e-06
        [updatestate_depend_eliminate]: 1.131e-05
        [updatestate_assign_eliminate]: 3.25998e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [cse]: 3.759e-05
        [call_graph_tuple_transform]: 2.726e-05
        [tuple_list_get_item_eliminator]: 7.06001e-06
        [none_parameter_eliminate]: 1.79e-06
        [renormalize]: 1.17e-06
        [switch_simplify]: 7.04001e-06
[remove_dup_value]: 1.635e-05
[partial_unused_args_eliminate]: 3.42002e-06
[environ_conv]: 7.36001e-06
[add_recomputation]: 6.624e-05
[cse_after_recomputation]: 2.729e-05, [1]
    [Cycle 1]: 1.956e-05, [1]
        [cse]: 1.085e-05
[auto_monad_reorder]: 1.771e-05
[get_jit_bprop_graph]: 2.84999e-06
[rewriter_after_jit_bprop_graph]: 9.76998e-06
[opt_after_jit_grad]: 0.00066521
[symbol_engine_optimizer]: 9.31e-05, [1]
    [Cycle 1]: 8.363e-05, [6]
        [build]: 6.71e-06
        [elim_shapecalc]: 8.70999e-06
        [elim_not_effective]: 1.819e-05
        [opt_reshape]: 7.35e-06
        [fold_const_symbol]: 9.97999e-06
        [renormalize]: 1.35001e-06
[validate]: 5.265e-05
[backend_pass]: 8.79983e-07
[task_emit]: 0.00886368
[execute]: 1.145e-05
Sums
    bootstrap                            :     0.000361s :  0.49%
    type_inference                       :     0.058982s : 80.11%
    event_method                         :     0.000024s :  0.03%
    auto_monad                           :     0.000091s :  0.12%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000075s :  0.10%
    rewriter_before_opt_a                :     0.000085s :  0.12%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000085s :  0.12%
    jit_opt_a.loop_unroll                :     0.000038s :  0.05%
    jit_opt_a.a_1                        :     0.000775s :  1.05%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.06%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.01%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.01%
    jit_opt_a.merge_forward              :     0.000008s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.04%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.02%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000018s :  0.03%
    jit_opt_a.inline_without_move        :     0.000010s :  0.01%
    jit_opt_a.renormalize                :     0.001097s :  1.49%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.03%
    jit_opt_a.cse                        :     0.000048s :  0.06%
    jit_opt_a.replace_applicator         :     0.000025s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.02%
    rewriter_after_opt_a                 :     0.000112s :  0.15%
    convert_after_rewriter               :     0.000010s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000950s :  1.29%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.02%
    cconv                                :     0.000035s :  0.05%
    loop_unroll                          :     0.000579s :  0.79%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.03%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000038s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.01%
    remove_dup_value                     :     0.000016s :  0.02%
    partial_unused_args_eliminate        :     0.000003s :  0.00%
    environ_conv                         :     0.000007s :  0.01%
    add_recomputation                    :     0.000066s :  0.09%
    cse_after_recomputation.cse          :     0.000011s :  0.01%
    auto_monad_reorder                   :     0.000018s :  0.02%
    get_jit_bprop_graph                  :     0.000003s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.01%
    opt_after_jit_grad                   :     0.000665s :  0.90%
    symbol_engine_optimizer.build        :     0.000007s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.02%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000053s :  0.07%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008864s : 12.04%
    execute                              :     0.000011s :  0.02%


Time group info:
------[substitution.]   0.000307    23
 1.06% :     0.000003s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 2.44% :     0.000007s :      3: substitution.graph_param_transform
86.55% :     0.000265s :      4: substitution.inline
 1.18% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.71% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000006s :      2: substitution.replace_old_param
 4.67% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.058889     2
98.08% :     0.057756s :      1: type_inference.infer
 1.92% :     0.001133s :      1: type_inference.specialize
------[replace.]   0.000089     6
67.77% :     0.000060s :      4: replace.inline
32.23% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000275     6
95.26% :     0.000262s :      4: match.inline
 4.74% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   691
 2.11% :     0.000003s :     10: predicate.accumulaten_eliminater
 1.21% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000002s :     10: predicate.addn_zero_filter
 2.18% :     0.000003s :     10: predicate.arithmetic_simplify
 1.00% :     0.000001s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.92% :     0.000001s :     10: predicate.compare_switch_simplify
 1.13% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.49% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.23% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.53% :     0.000001s :      3: predicate.elim_not_effective
 0.44% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000002s :     10: predicate.environ_get_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.43% :     0.000001s :      3: predicate.graph_param_transform
 5.24% :     0.000007s :     22: predicate.inline
 0.76% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.95% :     0.000003s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.47% :     0.000002s :     15: predicate.load_eliminater
 2.27% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.00% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.12% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.91% :     0.000001s :     10: predicate.merge_addn
 0.89% :     0.000001s :     10: predicate.minmaximum_grad
 2.68% :     0.000004s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 1.92% :     0.000003s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.90% :     0.000003s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.45% :     0.000001s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.09% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.26% :     0.000002s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.68% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.08% :     0.000003s :     16: predicate.switch_defer_inline
 1.77% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.73% :     0.000011s :     53: predicate.switch_simplify
 1.15% :     0.000002s :     10: predicate.tile_eliminate
 0.98% :     0.000001s :     10: predicate.transpose_eliminate
 1.52% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.00% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.19% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.89% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.95% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.42% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.01% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000920    12
51.61% :     0.000475s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.58% :     0.000060s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.81% :     0.000385s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.081437    76
 0.09% :     0.000071s :      1: add_recomputation
 0.12% :     0.000095s :      1: auto_monad
 0.03% :     0.000021s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.47% :     0.000384s :      1: bootstrap
 0.05% :     0.000038s :      1: cconv
 0.02% :     0.000014s :      1: convert_after_rewriter
 0.04% :     0.000030s :      1: cse_after_recomputation
 0.01% :     0.000011s :      1: environ_conv
 0.04% :     0.000029s :      1: event_method
 0.02% :     0.000019s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000009s :      1: graph_reusing
 9.16% :     0.007462s :      1: jit_opt_a
 0.24% :     0.000197s :      1: jit_opt_after_cconv
 0.08% :     0.000067s :      1: jit_opt_b
 0.73% :     0.000592s :      1: loop_unroll
 1.18% :     0.000964s :      1: mutable_eliminate
 1.26% :     0.001027s :     26: opt.transform.jit_opt_a
 0.08% :     0.000061s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000030s :      4: opt.transform.jit_opt_b
 0.02% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.05% :     0.000040s :      4: opt.transform.symbol_engine_opt
 0.83% :     0.000679s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.10% :     0.000079s :      1: py_interpret_to_execute
 0.02% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000020s :      1: remove_dup_value
 0.71% :     0.000580s :      1: renormalize.infer
 0.62% :     0.000506s :      1: renormalize.specialize
 0.01% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000120s :      1: rewriter_after_opt_a
 0.11% :     0.000088s :      1: rewriter_before_opt_a
 0.12% :     0.000096s :      1: symbol_engine_optimizer
10.92% :     0.008889s :      1: task_emit
72.45% :     0.059005s :      1: type_inference
 0.10% :     0.000082s :      1: validate


TotalTime = 0.0236894, [33]
[bootstrap]: 0.00044768
[type_inference]: 0.00606019
[event_method]: 2.108e-05
[auto_monad]: 7.324e-05
[graph_reusing]: 6.11e-06
[pre_auto_parallel]: 2.97002e-06
[py_interpret_to_execute]: 4.949e-05
[rewriter_before_opt_a]: 7.821e-05
[expand_dump_flag]: 3.47997e-06
[jit_opt_a]: 0.0074004, [2]
    [Cycle 1]: 0.00218106, [27]
        [switch_simplify]: 6.887e-05
        [loop_unroll]: 3.016e-05
        [a_1]: 0.00064351
        [with_stream_mark]: 2.437e-05
        [recompute_prepare]: 7.15e-06
        [updatestate_depend_eliminate]: 4.55001e-06
        [updatestate_assign_eliminate]: 3.46999e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [parameter_eliminate]: 2.04999e-06
        [specialize_transform]: 6.39999e-06
        [updatestate_useless_node_eliminater]: 5.31002e-06
        [accelerated_algorithm]: 5.71003e-06
        [meta_shard_fg_expand]: 2.93e-06
        [get_grad_eliminate_]: 5.38002e-06
        [merge_forward]: 4.23999e-06
        [cell_reuse_recompute_pass]: 1.31002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.487e-05
        [j_node_and_user_rematch]: 1.013e-05
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 1.019e-05
        [inline_without_move]: 5.40001e-06
        [renormalize]: 0.00105389
        [add_forward_monad_depend]: 7.13e-06
        [auto_monad_grad]: 2.86e-06
        [auto_monad_eliminator]: 1.639e-05
        [cse]: 3.097e-05
        [replace_applicator]: 1.377e-05
    [Cycle 2]: 0.00034442, [27]
        [switch_simplify]: 6.12999e-06
        [loop_unroll]: 5.25999e-06
        [a_1]: 9.724e-05
        [with_stream_mark]: 1.043e-05
        [recompute_prepare]: 5.09998e-06
        [updatestate_depend_eliminate]: 2.76e-06
        [updatestate_assign_eliminate]: 1.99999e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 1.28002e-06
        [specialize_transform]: 5.47001e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 5.27001e-06
        [meta_shard_fg_expand]: 1.42999e-06
        [get_grad_eliminate_]: 5.46e-06
        [merge_forward]: 2.93e-06
        [cell_reuse_recompute_pass]: 2.11998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.128e-05
        [j_node_and_user_rematch]: 7.91001e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 7.78999e-06
        [inline_without_move]: 4.89998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.47001e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 5.85002e-06
        [cse]: 1.076e-05
        [replace_applicator]: 7.45998e-06
[py_interpret_to_execute_after_opt_a]: 1.357e-05
[rewriter_after_opt_a]: 4.148e-05
[convert_after_rewriter]: 7.5e-06
[order_py_execute_after_rewriter]: 5.51e-06
[mutable_eliminate]: 0.00077753
[jit_opt_b]: 5.606e-05, [1]
    [Cycle 1]: 4.709e-05, [2]
        [frontend_op_eliminate]: 1.886e-05
        [inline_after_opt_a]: 1.622e-05
[cconv]: 2.985e-05
[loop_unroll]: 0.000459
[jit_opt_after_cconv]: 0.00015691, [1]
    [Cycle 1]: 0.00014991, [11]
        [c_1]: 2.214e-05
        [parameter_eliminate]: 4.36002e-06
        [updatestate_depend_eliminate]: 7.06999e-06
        [updatestate_assign_eliminate]: 2.59999e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 2.309e-05
        [call_graph_tuple_transform]: 2.272e-05
        [tuple_list_get_item_eliminator]: 5.92999e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 7.59988e-07
        [switch_simplify]: 5.35999e-06
[remove_dup_value]: 1.459e-05
[partial_unused_args_eliminate]: 2.29001e-06
[environ_conv]: 5.71e-06
[add_recomputation]: 5.48e-05
[cse_after_recomputation]: 2.153e-05, [1]
    [Cycle 1]: 1.569e-05, [1]
        [cse]: 9.86e-06
[auto_monad_reorder]: 1.637e-05
[get_jit_bprop_graph]: 2.35002e-06
[rewriter_after_jit_bprop_graph]: 6.17001e-06
[opt_after_jit_grad]: 0.00048117
[symbol_engine_optimizer]: 7.841e-05, [1]
    [Cycle 1]: 7.086e-05, [6]
        [build]: 4.16001e-06
        [elim_shapecalc]: 8.29998e-06
        [elim_not_effective]: 1.487e-05
        [opt_reshape]: 6.19999e-06
        [fold_const_symbol]: 8.91997e-06
        [renormalize]: 4.2998e-07
[validate]: 4.448e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.00701004
[execute]: 1.11e-05
Sums
    bootstrap                            :     0.000448s :  2.47%
    type_inference                       :     0.006060s : 33.43%
    event_method                         :     0.000021s :  0.12%
    auto_monad                           :     0.000073s :  0.40%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000049s :  0.27%
    rewriter_before_opt_a                :     0.000078s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000075s :  0.41%
    jit_opt_a.loop_unroll                :     0.000035s :  0.20%
    jit_opt_a.a_1                        :     0.000741s :  4.09%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001054s :  5.81%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.12%
    jit_opt_a.cse                        :     0.000042s :  0.23%
    jit_opt_a.replace_applicator         :     0.000021s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.07%
    rewriter_after_opt_a                 :     0.000041s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000778s :  4.29%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000030s :  0.16%
    loop_unroll                          :     0.000459s :  2.53%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000055s :  0.30%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000481s :  2.65%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000044s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007010s : 38.67%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000289    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.44% :     0.000001s :      2: substitution.fold_const_symbol
 2.17% :     0.000006s :      3: substitution.graph_param_transform
84.95% :     0.000246s :      4: substitution.inline
 1.08% :     0.000003s :      4: substitution.j_node_and_user_rematch
 4.03% :     0.000012s :      4: substitution.remove_not_recompute_node
 1.93% :     0.000006s :      2: substitution.replace_old_param
 4.60% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005971     2
82.00% :     0.004896s :      1: type_inference.infer
18.00% :     0.001075s :      1: type_inference.specialize
------[replace.]   0.000084     6
69.43% :     0.000058s :      4: replace.inline
30.57% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000255     6
95.29% :     0.000243s :      4: match.inline
 4.71% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.51% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.41% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000002s :     10: predicate.addn_zero_filter
 2.26% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 0.99% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.45% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.07% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.19% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.36% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.04% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.08% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.76% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 1.77% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.88% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.22% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.95% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.42% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.10% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.20% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.46% :     0.000002s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.15% :     0.000003s :     16: predicate.switch_defer_inline
 1.88% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.64% :     0.000010s :     53: predicate.switch_simplify
 1.37% :     0.000002s :     10: predicate.tile_eliminate
 1.26% :     0.000002s :     10: predicate.transpose_eliminate
 1.68% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.45% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.88% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000804    12
49.16% :     0.000395s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.21% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.63% :     0.000351s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025827    76
 0.22% :     0.000058s :      1: add_recomputation
 0.30% :     0.000078s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.83% :     0.000472s :      1: bootstrap
 0.13% :     0.000033s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000026s :      1: event_method
 0.07% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
28.67% :     0.007404s :      1: jit_opt_a
 0.62% :     0.000160s :      1: jit_opt_after_cconv
 0.23% :     0.000059s :      1: jit_opt_b
 1.81% :     0.000467s :      1: loop_unroll
 3.05% :     0.000788s :      1: mutable_eliminate
 3.79% :     0.000980s :     26: opt.transform.jit_opt_a
 0.20% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.90% :     0.000491s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000053s :      1: py_interpret_to_execute
 0.06% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.39% :     0.000617s :      1: renormalize.infer
 1.66% :     0.000428s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000045s :      1: rewriter_after_opt_a
 0.32% :     0.000082s :      1: rewriter_before_opt_a
 0.31% :     0.000081s :      1: symbol_engine_optimizer
27.23% :     0.007033s :      1: task_emit
23.54% :     0.006080s :      1: type_inference
 0.26% :     0.000067s :      1: validate


TotalTime = 0.0228151, [33]
[bootstrap]: 0.00036509
[type_inference]: 0.00566312
[event_method]: 1.962e-05
[auto_monad]: 7.12e-05
[graph_reusing]: 6.73998e-06
[pre_auto_parallel]: 2.99999e-06
[py_interpret_to_execute]: 4.776e-05
[rewriter_before_opt_a]: 8.121e-05
[expand_dump_flag]: 3.38e-06
[jit_opt_a]: 0.00684345, [2]
    [Cycle 1]: 0.00219096, [27]
        [switch_simplify]: 7.37e-05
        [loop_unroll]: 3.208e-05
        [a_1]: 0.00065919
        [with_stream_mark]: 2.533e-05
        [recompute_prepare]: 8.95999e-06
        [updatestate_depend_eliminate]: 4.07998e-06
        [updatestate_assign_eliminate]: 4.2e-06
        [updatestate_loads_eliminate]: 2.91e-06
        [parameter_eliminate]: 2.16e-06
        [specialize_transform]: 6.09001e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 6.44999e-06
        [meta_shard_fg_expand]: 2.41e-06
        [get_grad_eliminate_]: 5.91998e-06
        [merge_forward]: 3.93001e-06
        [cell_reuse_recompute_pass]: 1.81998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.581e-05
        [j_node_and_user_rematch]: 1.132e-05
        [meta_fg_expand]: 2.68998e-06
        [replace_old_param]: 1.126e-05
        [inline_without_move]: 5.69e-06
        [renormalize]: 0.00102683
        [add_forward_monad_depend]: 7.73001e-06
        [auto_monad_grad]: 2.95998e-06
        [auto_monad_eliminator]: 1.461e-05
        [cse]: 3.217e-05
        [replace_applicator]: 1.481e-05
    [Cycle 2]: 0.00037317, [27]
        [switch_simplify]: 7.02002e-06
        [loop_unroll]: 5.29998e-06
        [a_1]: 0.0001041
        [with_stream_mark]: 1.198e-05
        [recompute_prepare]: 6.60002e-06
        [updatestate_depend_eliminate]: 3.23e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.24003e-06
        [specialize_transform]: 5.54e-06
        [updatestate_useless_node_eliminater]: 5.89999e-06
        [accelerated_algorithm]: 6.70002e-06
        [meta_shard_fg_expand]: 1.63002e-06
        [get_grad_eliminate_]: 5.25999e-06
        [merge_forward]: 3.03e-06
        [cell_reuse_recompute_pass]: 3.42997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.587e-05
        [j_node_and_user_rematch]: 9.80002e-06
        [meta_fg_expand]: 2.06998e-06
        [replace_old_param]: 8.48001e-06
        [inline_without_move]: 5.28002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 9.99979e-07
        [auto_monad_eliminator]: 6.11e-06
        [cse]: 1.162e-05
        [replace_applicator]: 5.39998e-06
[py_interpret_to_execute_after_opt_a]: 1.365e-05
[rewriter_after_opt_a]: 4.024e-05
[convert_after_rewriter]: 7.51999e-06
[order_py_execute_after_rewriter]: 8.57e-06
[mutable_eliminate]: 0.00077761
[jit_opt_b]: 5.259e-05, [1]
    [Cycle 1]: 4.47e-05, [2]
        [frontend_op_eliminate]: 1.798e-05
        [inline_after_opt_a]: 1.514e-05
[cconv]: 3.068e-05
[loop_unroll]: 0.00052662
[jit_opt_after_cconv]: 0.00016642, [1]
    [Cycle 1]: 0.00015905, [11]
        [c_1]: 2.262e-05
        [parameter_eliminate]: 3.53e-06
        [updatestate_depend_eliminate]: 7.55e-06
        [updatestate_assign_eliminate]: 2.59001e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 2.271e-05
        [call_graph_tuple_transform]: 2.259e-05
        [tuple_list_get_item_eliminator]: 6.09001e-06
        [none_parameter_eliminate]: 1.35999e-06
        [renormalize]: 5.19998e-07
        [switch_simplify]: 5.52001e-06
[remove_dup_value]: 1.43e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 6.28e-06
[add_recomputation]: 5.421e-05
[cse_after_recomputation]: 2.331e-05, [1]
    [Cycle 1]: 1.655e-05, [1]
        [cse]: 9.44998e-06
[auto_monad_reorder]: 1.606e-05
[get_jit_bprop_graph]: 2.18998e-06
[rewriter_after_jit_bprop_graph]: 7.76001e-06
[opt_after_jit_grad]: 0.00077133
[symbol_engine_optimizer]: 9.008e-05, [1]
    [Cycle 1]: 8.216e-05, [6]
        [build]: 5.61003e-06
        [elim_shapecalc]: 1.22e-05
        [elim_not_effective]: 1.482e-05
        [opt_reshape]: 5.94999e-06
        [fold_const_symbol]: 1.034e-05
        [renormalize]: 4.89992e-07
[validate]: 4.353e-05
[backend_pass]: 9.79984e-07
[task_emit]: 0.00679792
[execute]: 1.024e-05
Sums
    bootstrap                            :     0.000365s :  2.05%
    type_inference                       :     0.005663s : 31.79%
    event_method                         :     0.000020s :  0.11%
    auto_monad                           :     0.000071s :  0.40%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000048s :  0.27%
    rewriter_before_opt_a                :     0.000081s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000081s :  0.45%
    jit_opt_a.loop_unroll                :     0.000037s :  0.21%
    jit_opt_a.a_1                        :     0.000763s :  4.28%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000020s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001027s :  5.76%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.12%
    jit_opt_a.cse                        :     0.000044s :  0.25%
    jit_opt_a.replace_applicator         :     0.000020s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000040s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000009s :  0.05%
    mutable_eliminate                    :     0.000778s :  4.36%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000031s :  0.17%
    loop_unroll                          :     0.000527s :  2.96%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000054s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000771s :  4.33%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000044s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006798s : 38.16%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000286    23
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 2.20% :     0.000006s :      3: substitution.graph_param_transform
85.99% :     0.000246s :      4: substitution.inline
 1.77% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.88% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.82% :     0.000005s :      2: substitution.replace_old_param
 5.03% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005580     2
81.97% :     0.004574s :      1: type_inference.infer
18.03% :     0.001006s :      1: type_inference.specialize
------[replace.]   0.000085     6
68.15% :     0.000058s :      4: replace.inline
31.85% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000256     6
94.86% :     0.000243s :      4: match.inline
 5.14% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   691
 1.64% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 2.94% :     0.000004s :     10: predicate.arithmetic_simplify
 1.49% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.92% :     0.000001s :     10: predicate.compare_switch_simplify
 1.46% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.58% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.82% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     10: predicate.environ_get_eliminate
 1.16% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000001s :      3: predicate.graph_param_transform
 5.27% :     0.000007s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.45% :     0.000002s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.75% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.51% :     0.000002s :      3: predicate.mutable_eliminate
 0.39% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.85% :     0.000002s :     10: predicate.reduce_eliminate
 1.65% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.91% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.09% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000002s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.64% :     0.000004s :     16: predicate.switch_defer_inline
 2.24% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.82% :     0.000011s :     53: predicate.switch_simplify
 1.17% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000002s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.46% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.05% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.64% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.89% :     0.000003s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000747    12
47.19% :     0.000352s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.66% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.15% :     0.000337s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024958    76
 0.23% :     0.000057s :      1: add_recomputation
 0.30% :     0.000075s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.56% :     0.000390s :      1: bootstrap
 0.14% :     0.000034s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
27.44% :     0.006848s :      1: jit_opt_a
 0.68% :     0.000169s :      1: jit_opt_after_cconv
 0.22% :     0.000055s :      1: jit_opt_b
 2.14% :     0.000535s :      1: loop_unroll
 3.15% :     0.000787s :      1: mutable_eliminate
 4.06% :     0.001013s :     26: opt.transform.jit_opt_a
 0.21% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000039s :      4: opt.transform.symbol_engine_opt
 3.14% :     0.000783s :      1: opt_after_jit_grad
 0.04% :     0.000011s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000051s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.32% :     0.000579s :      1: renormalize.infer
 1.75% :     0.000438s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000043s :      1: rewriter_after_opt_a
 0.34% :     0.000084s :      1: rewriter_before_opt_a
 0.37% :     0.000093s :      1: symbol_engine_optimizer
27.32% :     0.006818s :      1: task_emit
22.77% :     0.005682s :      1: type_inference
 0.27% :     0.000067s :      1: validate


TotalTime = 0.0211376, [33]
[bootstrap]: 0.00039462
[type_inference]: 0.00533386
[event_method]: 1.948e-05
[auto_monad]: 6.89e-05
[graph_reusing]: 5.96998e-06
[pre_auto_parallel]: 3.01999e-06
[py_interpret_to_execute]: 4.309e-05
[rewriter_before_opt_a]: 7.743e-05
[expand_dump_flag]: 3.38e-06
[jit_opt_a]: 0.00642588, [2]
    [Cycle 1]: 0.00209487, [27]
        [switch_simplify]: 6.434e-05
        [loop_unroll]: 3.081e-05
        [a_1]: 0.00060337
        [with_stream_mark]: 2.146e-05
        [recompute_prepare]: 7.84002e-06
        [updatestate_depend_eliminate]: 3.80998e-06
        [updatestate_assign_eliminate]: 3.16999e-06
        [updatestate_loads_eliminate]: 2.99001e-06
        [parameter_eliminate]: 2.598e-05
        [specialize_transform]: 7.26001e-06
        [updatestate_useless_node_eliminater]: 6.61e-06
        [accelerated_algorithm]: 6.63e-06
        [meta_shard_fg_expand]: 2.26998e-06
        [get_grad_eliminate_]: 5.88002e-06
        [merge_forward]: 4.02e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.677e-05
        [j_node_and_user_rematch]: 1.089e-05
        [meta_fg_expand]: 2.42001e-06
        [replace_old_param]: 1.008e-05
        [inline_without_move]: 5.64998e-06
        [renormalize]: 0.00099256
        [add_forward_monad_depend]: 6.81001e-06
        [auto_monad_grad]: 3.06001e-06
        [auto_monad_eliminator]: 1.558e-05
        [cse]: 3.118e-05
        [replace_applicator]: 1.517e-05
    [Cycle 2]: 0.00035355, [27]
        [switch_simplify]: 6.51e-06
        [loop_unroll]: 5.66003e-06
        [a_1]: 9.817e-05
        [with_stream_mark]: 1.158e-05
        [recompute_prepare]: 5.24998e-06
        [updatestate_depend_eliminate]: 3.06001e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 2.83e-06
        [parameter_eliminate]: 1.52999e-06
        [specialize_transform]: 5.25999e-06
        [updatestate_useless_node_eliminater]: 5.17999e-06
        [accelerated_algorithm]: 6.01e-06
        [meta_shard_fg_expand]: 1.79998e-06
        [get_grad_eliminate_]: 4.70001e-06
        [merge_forward]: 3.28e-06
        [cell_reuse_recompute_pass]: 3.31001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.458e-05
        [j_node_and_user_rematch]: 8.65001e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 7.87e-06
        [inline_without_move]: 5.59e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.23002e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 5.39998e-06
        [cse]: 1.078e-05
        [replace_applicator]: 5.09003e-06
[py_interpret_to_execute_after_opt_a]: 1.286e-05
[rewriter_after_opt_a]: 3.791e-05
[convert_after_rewriter]: 6.95002e-06
[order_py_execute_after_rewriter]: 5.37999e-06
[mutable_eliminate]: 0.00075279
[jit_opt_b]: 5.673e-05, [1]
    [Cycle 1]: 4.762e-05, [2]
        [frontend_op_eliminate]: 1.962e-05
        [inline_after_opt_a]: 1.596e-05
[cconv]: 3.104e-05
[loop_unroll]: 0.00046511
[jit_opt_after_cconv]: 0.00015889, [1]
    [Cycle 1]: 0.0001525, [11]
        [c_1]: 2.26e-05
        [parameter_eliminate]: 3.68999e-06
        [updatestate_depend_eliminate]: 7.75e-06
        [updatestate_assign_eliminate]: 2.57001e-06
        [updatestate_loads_eliminate]: 2.89999e-06
        [cse]: 2.399e-05
        [call_graph_tuple_transform]: 2.041e-05
        [tuple_list_get_item_eliminator]: 6.02999e-06
        [none_parameter_eliminate]: 1.79e-06
        [renormalize]: 9.10019e-07
        [switch_simplify]: 6.17999e-06
[remove_dup_value]: 1.412e-05
[partial_unused_args_eliminate]: 2.41998e-06
[environ_conv]: 6.31e-06
[add_recomputation]: 5.836e-05
[cse_after_recomputation]: 2.274e-05, [1]
    [Cycle 1]: 1.704e-05, [1]
        [cse]: 1.016e-05
[auto_monad_reorder]: 1.663e-05
[get_jit_bprop_graph]: 2.35002e-06
[rewriter_after_jit_bprop_graph]: 7.50998e-06
[opt_after_jit_grad]: 0.00050278
[symbol_engine_optimizer]: 8.215e-05, [1]
    [Cycle 1]: 7.394e-05, [6]
        [build]: 4.49998e-06
        [elim_shapecalc]: 8.72e-06
        [elim_not_effective]: 1.361e-05
        [opt_reshape]: 6.39999e-06
        [fold_const_symbol]: 9.83998e-06
        [renormalize]: 6.30011e-07
[validate]: 0.00014103
[backend_pass]: 9.70002e-07
[task_emit]: 0.00613856
[execute]: 7.1e-06
Sums
    bootstrap                            :     0.000395s :  2.39%
    type_inference                       :     0.005334s : 32.36%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000069s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.26%
    rewriter_before_opt_a                :     0.000077s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000071s :  0.43%
    jit_opt_a.loop_unroll                :     0.000036s :  0.22%
    jit_opt_a.a_1                        :     0.000702s :  4.26%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000028s :  0.17%
    jit_opt_a.specialize_transform       :     0.000013s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000993s :  6.02%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.13%
    jit_opt_a.cse                        :     0.000042s :  0.25%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000753s :  4.57%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000031s :  0.19%
    loop_unroll                          :     0.000465s :  2.82%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000058s :  0.35%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000503s :  3.05%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000141s :  0.86%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006139s : 37.24%
    execute                              :     0.000007s :  0.04%


Time group info:
------[substitution.]   0.000242    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.77% :     0.000002s :      2: substitution.fold_const_symbol
 2.47% :     0.000006s :      3: substitution.graph_param_transform
84.79% :     0.000205s :      4: substitution.inline
 1.69% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.96% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.14% :     0.000005s :      2: substitution.replace_old_param
 5.36% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005258     2
81.71% :     0.004297s :      1: type_inference.infer
18.29% :     0.000962s :      1: type_inference.specialize
------[replace.]   0.000079     6
66.92% :     0.000053s :      4: replace.inline
33.08% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000214     6
94.46% :     0.000203s :      4: match.inline
 5.54% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.26% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.55% :     0.000002s :     10: predicate.addn_zero_filter
 1.85% :     0.000002s :     10: predicate.arithmetic_simplify
 1.20% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.08% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 4.91% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.51% :     0.000002s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.42% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.16% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.42% :     0.000003s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     15: predicate.partial_eliminate
 1.36% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.68% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.64% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.05% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.21% :     0.000002s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.31% :     0.000003s :     16: predicate.switch_defer_inline
 1.89% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.92% :     0.000010s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000002s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.77% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.22% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000686    12
45.57% :     0.000313s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.56% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.87% :     0.000328s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023177    76
 0.26% :     0.000061s :      1: add_recomputation
 0.31% :     0.000072s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.80% :     0.000416s :      1: bootstrap
 0.15% :     0.000034s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.74% :     0.006430s :      1: jit_opt_a
 0.70% :     0.000162s :      1: jit_opt_after_cconv
 0.26% :     0.000060s :      1: jit_opt_b
 2.04% :     0.000473s :      1: loop_unroll
 3.30% :     0.000765s :      1: mutable_eliminate
 4.05% :     0.000938s :     26: opt.transform.jit_opt_a
 0.22% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000029s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000512s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000046s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.27% :     0.000526s :      1: renormalize.infer
 1.98% :     0.000458s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000041s :      1: rewriter_after_opt_a
 0.35% :     0.000081s :      1: rewriter_before_opt_a
 0.37% :     0.000085s :      1: symbol_engine_optimizer
26.55% :     0.006153s :      1: task_emit
23.09% :     0.005351s :      1: type_inference
 0.72% :     0.000168s :      1: validate


TotalTime = 0.0160151, [30]
[bootstrap]: 0.00032256
[type_inference]: 0.00674258
[event_method]: 2.068e-05
[auto_monad]: 8.013e-05
[graph_reusing]: 6.81001e-06
[pre_auto_parallel]: 3.09001e-06
[py_interpret_to_execute]: 4.426e-05
[rewriter_before_opt_a]: 8.97e-05
[expand_dump_flag]: 3.13e-06
[jit_opt_a]: 0.00658295, [2]
    [Cycle 1]: 0.00195034, [27]
        [switch_simplify]: 0.00013243
        [loop_unroll]: 3.266e-05
        [a_1]: 0.000718
        [with_stream_mark]: 2.139e-05
        [recompute_prepare]: 7.46999e-06
        [updatestate_depend_eliminate]: 3.72002e-06
        [updatestate_assign_eliminate]: 4.00998e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [parameter_eliminate]: 1.74e-06
        [specialize_transform]: 5.71e-06
        [updatestate_useless_node_eliminater]: 4.67998e-06
        [accelerated_algorithm]: 5.22999e-06
        [meta_shard_fg_expand]: 2.03002e-06
        [get_grad_eliminate_]: 4.89003e-06
        [merge_forward]: 3.76001e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.417e-05
        [j_node_and_user_rematch]: 9.04998e-06
        [meta_fg_expand]: 2.59999e-06
        [replace_old_param]: 9.14e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.00071236
        [add_forward_monad_depend]: 5.79e-06
        [auto_monad_grad]: 2.37001e-06
        [auto_monad_eliminator]: 1.266e-05
        [cse]: 3.144e-05
        [replace_applicator]: 1.161e-05
    [Cycle 2]: 0.00030479, [27]
        [switch_simplify]: 5.38997e-06
        [loop_unroll]: 4.74e-06
        [a_1]: 8.746e-05
        [with_stream_mark]: 8.88002e-06
        [recompute_prepare]: 5.35001e-06
        [updatestate_depend_eliminate]: 2.43e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.78002e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 4.72e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 4.37e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.41002e-06
        [merge_forward]: 2.51e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.285e-05
        [j_node_and_user_rematch]: 7.71999e-06
        [meta_fg_expand]: 1.39998e-06
        [replace_old_param]: 6.93e-06
        [inline_without_move]: 4.69998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.07998e-06
        [auto_monad_grad]: 6.79982e-07
        [auto_monad_eliminator]: 4.38999e-06
        [cse]: 9.70002e-06
        [replace_applicator]: 4.75001e-06
[py_interpret_to_execute_after_opt_a]: 9.52999e-06
[rewriter_after_opt_a]: 3.377e-05
[convert_after_rewriter]: 6.65002e-06
[order_py_execute_after_rewriter]: 5.08002e-06
[mutable_eliminate]: 0.00058994
[jit_opt_b]: 5.969e-05, [1]
    [Cycle 1]: 5.253e-05, [2]
        [frontend_op_eliminate]: 1.548e-05
        [inline_after_opt_a]: 1.733e-05
[cconv]: 2.466e-05
[loop_unroll]: 0.00042188
[jit_opt_after_cconv]: 0.00013552, [1]
    [Cycle 1]: 0.00012889, [11]
        [c_1]: 2.022e-05
        [parameter_eliminate]: 2.90998e-06
        [updatestate_depend_eliminate]: 5.51e-06
        [updatestate_assign_eliminate]: 2.52001e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.685e-05
        [call_graph_tuple_transform]: 1.773e-05
        [tuple_list_get_item_eliminator]: 4.99e-06
        [none_parameter_eliminate]: 1.42999e-06
        [renormalize]: 6.10016e-07
        [switch_simplify]: 5.44e-06
[remove_dup_value]: 1.286e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 5.17e-06
[add_recomputation]: 4.897e-05
[cse_after_recomputation]: 2.122e-05, [1]
    [Cycle 1]: 1.576e-05, [1]
        [cse]: 9.78002e-06
[auto_monad_reorder]: 1.456e-05
[get_jit_bprop_graph]: 2.07999e-06
[rewriter_after_jit_bprop_graph]: 5.47001e-06
[opt_after_jit_grad]: 0.00043879
[symbol_engine_optimizer]: 7.134e-05, [1]
    [Cycle 1]: 6.484e-05, [6]
        [build]: 4.07e-06
        [elim_shapecalc]: 7.38e-06
        [elim_not_effective]: 1.198e-05
        [opt_reshape]: 5.51e-06
        [fold_const_symbol]: 8.11002e-06
        [renormalize]: 5.09986e-07
[validate]: 3.438e-05
Sums
    bootstrap                            :     0.000323s :  2.91%
    type_inference                       :     0.006743s : 60.78%
    event_method                         :     0.000021s :  0.19%
    auto_monad                           :     0.000080s :  0.72%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000044s :  0.40%
    rewriter_before_opt_a                :     0.000090s :  0.81%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000138s :  1.24%
    jit_opt_a.loop_unroll                :     0.000037s :  0.34%
    jit_opt_a.a_1                        :     0.000805s :  7.26%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000010s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.08%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000712s :  6.42%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.15%
    jit_opt_a.cse                        :     0.000041s :  0.37%
    jit_opt_a.replace_applicator         :     0.000016s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.09%
    rewriter_after_opt_a                 :     0.000034s :  0.30%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000590s :  5.32%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.16%
    cconv                                :     0.000025s :  0.22%
    loop_unroll                          :     0.000422s :  3.80%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000049s :  0.44%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000439s :  3.96%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.31%


Time group info:
------[substitution.]   0.000286    26
 0.73% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 1.73% :     0.000005s :      3: substitution.graph_param_transform
85.12% :     0.000244s :      6: substitution.inline
 1.17% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.57% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.50% :     0.000004s :      2: substitution.replace_old_param
 3.02% :     0.000009s :      1: substitution.switch_simplify
 4.72% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006670     2
83.82% :     0.005591s :      1: type_inference.infer
16.18% :     0.001079s :      1: type_inference.specialize
------[replace.]   0.000137     9
47.74% :     0.000065s :      6: replace.inline
28.88% :     0.000040s :      1: replace.switch_simplify
23.39% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000260     9
92.24% :     0.000240s :      6: match.inline
 3.04% :     0.000008s :      1: match.switch_simplify
 4.72% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   801
 1.41% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.99% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000002s :     12: predicate.addn_check_dump
 1.61% :     0.000002s :     12: predicate.addn_zero_filter
 1.78% :     0.000002s :     12: predicate.arithmetic_simplify
 1.40% :     0.000002s :     12: predicate.cast_eliminate
 0.38% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     12: predicate.compare_switch_simplify
 1.50% :     0.000002s :     12: predicate.depend_value_elim
 1.18% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.42% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     12: predicate.environ_get_eliminate
 1.18% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 5.69% :     0.000007s :     26: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.99% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     17: predicate.load_eliminater
 1.23% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.93% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.26% :     0.000002s :     12: predicate.merge_addn
 1.14% :     0.000002s :     12: predicate.minmaximum_grad
 1.40% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     17: predicate.partial_eliminate
 1.28% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     12: predicate.reduce_eliminate
 1.48% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     20: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.39% :     0.000001s :      3: predicate.reset_defer_inline
 1.68% :     0.000002s :     12: predicate.reshape_eliminate
 1.27% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     12: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.77% :     0.000001s :      6: predicate.special_op_eliminate
 0.85% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.79% :     0.000004s :     20: predicate.switch_defer_inline
 2.35% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.50% :     0.000011s :     65: predicate.switch_simplify
 1.50% :     0.000002s :     12: predicate.tile_eliminate
 1.48% :     0.000002s :     12: predicate.transpose_eliminate
 1.39% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.58% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.69% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000002s :     12: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000804    15
50.72% :     0.000408s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.81% :     0.000103s :      1: func_graph_cloner_run.FuncGraphClonerNode
36.47% :     0.000293s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017863    72
 0.29% :     0.000052s :      1: add_recomputation
 0.47% :     0.000084s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 1.92% :     0.000343s :      1: bootstrap
 0.15% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.14% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
36.87% :     0.006586s :      1: jit_opt_a
 0.77% :     0.000138s :      1: jit_opt_after_cconv
 0.35% :     0.000063s :      1: jit_opt_b
 2.41% :     0.000430s :      1: loop_unroll
 3.35% :     0.000598s :      1: mutable_eliminate
 6.08% :     0.001086s :     26: opt.transform.jit_opt_a
 0.25% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.50% :     0.000446s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000047s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000015s :      1: remove_dup_value
 2.31% :     0.000412s :      1: renormalize.infer
 1.64% :     0.000293s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000037s :      1: rewriter_after_opt_a
 0.52% :     0.000093s :      1: rewriter_before_opt_a
 0.41% :     0.000074s :      1: symbol_engine_optimizer
37.83% :     0.006758s :      1: type_inference


TotalTime = 0.0390685, [33]
[bootstrap]: 0.00040861
[type_inference]: 0.008982
[event_method]: 2.324e-05
[auto_monad]: 7.282e-05
[graph_reusing]: 6.11e-06
[pre_auto_parallel]: 2.78e-06
[py_interpret_to_execute]: 4.876e-05
[rewriter_before_opt_a]: 8.407e-05
[expand_dump_flag]: 3.9e-06
[jit_opt_a]: 0.0106876, [2]
    [Cycle 1]: 0.00208156, [27]
        [switch_simplify]: 6.786e-05
        [loop_unroll]: 3.06e-05
        [a_1]: 0.00067813
        [with_stream_mark]: 2.266e-05
        [recompute_prepare]: 9.73002e-06
        [updatestate_depend_eliminate]: 4.07e-06
        [updatestate_assign_eliminate]: 3.13e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [parameter_eliminate]: 2.34999e-06
        [specialize_transform]: 6.36e-06
        [updatestate_useless_node_eliminater]: 5.04998e-06
        [accelerated_algorithm]: 5.81e-06
        [meta_shard_fg_expand]: 2.69999e-06
        [get_grad_eliminate_]: 5.91e-06
        [merge_forward]: 3.93999e-06
        [cell_reuse_recompute_pass]: 1.33002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.498e-05
        [j_node_and_user_rematch]: 9.20999e-06
        [meta_fg_expand]: 2.89999e-06
        [replace_old_param]: 9.32999e-06
        [inline_without_move]: 5.42999e-06
        [renormalize]: 0.00091176
        [add_forward_monad_depend]: 7.31001e-06
        [auto_monad_grad]: 2.29999e-06
        [auto_monad_eliminator]: 1.565e-05
        [cse]: 3.246e-05
        [replace_applicator]: 1.871e-05
    [Cycle 2]: 0.00036255, [27]
        [switch_simplify]: 5.28002e-06
        [loop_unroll]: 5.09e-06
        [a_1]: 9.553e-05
        [with_stream_mark]: 1.368e-05
        [recompute_prepare]: 5.25001e-06
        [updatestate_depend_eliminate]: 2.91e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 2.59001e-06
        [parameter_eliminate]: 1.49e-06
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 5.15999e-06
        [accelerated_algorithm]: 5.79999e-06
        [meta_shard_fg_expand]: 1.81998e-06
        [get_grad_eliminate_]: 4.75999e-06
        [merge_forward]: 2.88998e-06
        [cell_reuse_recompute_pass]: 3.63999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.721e-05
        [j_node_and_user_rematch]: 8.55999e-06
        [meta_fg_expand]: 1.76998e-06
        [replace_old_param]: 8.62e-06
        [inline_without_move]: 5.07e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.22001e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 5.17e-06
        [cse]: 1.352e-05
        [replace_applicator]: 6.04001e-06
[py_interpret_to_execute_after_opt_a]: 1.622e-05
[rewriter_after_opt_a]: 4.51e-05
[convert_after_rewriter]: 8.62e-06
[order_py_execute_after_rewriter]: 5.04e-06
[mutable_eliminate]: 0.00084637
[jit_opt_b]: 5.726e-05, [1]
    [Cycle 1]: 4.89e-05, [2]
        [frontend_op_eliminate]: 1.896e-05
        [inline_after_opt_a]: 1.778e-05
[cconv]: 3.22e-05
[loop_unroll]: 0.00055062
[jit_opt_after_cconv]: 0.00017848, [1]
    [Cycle 1]: 0.0001695, [11]
        [c_1]: 2.395e-05
        [parameter_eliminate]: 4.32998e-06
        [updatestate_depend_eliminate]: 8.03001e-06
        [updatestate_assign_eliminate]: 2.59999e-06
        [updatestate_loads_eliminate]: 2.96999e-06
        [cse]: 3.273e-05
        [call_graph_tuple_transform]: 2.289e-05
        [tuple_list_get_item_eliminator]: 6.66e-06
        [none_parameter_eliminate]: 1.81003e-06
        [renormalize]: 1.05001e-06
        [switch_simplify]: 5.52999e-06
[remove_dup_value]: 1.636e-05
[partial_unused_args_eliminate]: 3.01999e-06
[environ_conv]: 7.25e-06
[add_recomputation]: 5.814e-05
[cse_after_recomputation]: 2.437e-05, [1]
    [Cycle 1]: 1.786e-05, [1]
        [cse]: 1.048e-05
[auto_monad_reorder]: 1.777e-05
[get_jit_bprop_graph]: 2.07999e-06
[rewriter_after_jit_bprop_graph]: 8.08999e-06
[opt_after_jit_grad]: 0.00488125
[symbol_engine_optimizer]: 0.00010458, [1]
    [Cycle 1]: 9.169e-05, [6]
        [build]: 8.08999e-06
        [elim_shapecalc]: 1.111e-05
        [elim_not_effective]: 2.007e-05
        [opt_reshape]: 7.02002e-06
        [fold_const_symbol]: 1.033e-05
        [renormalize]: 1.17999e-06
[validate]: 5.254e-05
[backend_pass]: 1.25001e-06
[task_emit]: 0.0115374
[execute]: 9.00999e-06
Sums
    bootstrap                            :     0.000409s :  1.36%
    type_inference                       :     0.008982s : 29.88%
    event_method                         :     0.000023s :  0.08%
    auto_monad                           :     0.000073s :  0.24%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000049s :  0.16%
    rewriter_before_opt_a                :     0.000084s :  0.28%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000073s :  0.24%
    jit_opt_a.loop_unroll                :     0.000036s :  0.12%
    jit_opt_a.a_1                        :     0.000774s :  2.57%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.06%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.000912s :  3.03%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.07%
    jit_opt_a.cse                        :     0.000046s :  0.15%
    jit_opt_a.replace_applicator         :     0.000025s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.05%
    rewriter_after_opt_a                 :     0.000045s :  0.15%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000846s :  2.82%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000032s :  0.11%
    loop_unroll                          :     0.000551s :  1.83%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000033s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000058s :  0.19%
    cse_after_recomputation.cse          :     0.000010s :  0.03%
    auto_monad_reorder                   :     0.000018s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.03%
    opt_after_jit_grad                   :     0.004881s : 16.24%
    symbol_engine_optimizer.build        :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000053s :  0.17%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011537s : 38.38%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000256    23
 1.04% :     0.000003s :      2: substitution.elim_not_effective
 0.74% :     0.000002s :      2: substitution.fold_const_symbol
 2.43% :     0.000006s :      3: substitution.graph_param_transform
84.57% :     0.000217s :      4: substitution.inline
 1.44% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.95% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.04% :     0.000005s :      2: substitution.replace_old_param
 5.79% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008882     2
87.32% :     0.007756s :      1: type_inference.infer
12.68% :     0.001126s :      1: type_inference.specialize
------[replace.]   0.000148     6
79.99% :     0.000118s :      4: replace.inline
20.01% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000228     6
93.97% :     0.000214s :      4: match.inline
 6.03% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.59% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.13% :     0.000001s :     10: predicate.addn_zero_filter
 2.53% :     0.000003s :     10: predicate.arithmetic_simplify
 1.06% :     0.000001s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.18% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.03% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.68% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.61% :     0.000001s :      3: predicate.elim_not_effective
 1.22% :     0.000002s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.02% :     0.000001s :     10: predicate.environ_get_eliminate
 1.41% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 4.64% :     0.000006s :     22: predicate.inline
 0.78% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000002s :      6: predicate.less_batch_normalization
 1.35% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.59% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.69% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 0.93% :     0.000001s :     10: predicate.minmaximum_grad
 2.27% :     0.000003s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 1.94% :     0.000003s :     15: predicate.partial_eliminate
 0.97% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.80% :     0.000002s :     10: predicate.reduce_eliminate
 1.29% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.96% :     0.000003s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     10: predicate.reshape_eliminate
 1.34% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.27% :     0.000002s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.66% :     0.000003s :     16: predicate.switch_defer_inline
 1.77% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.34% :     0.000010s :     53: predicate.switch_simplify
 1.99% :     0.000003s :     10: predicate.tile_eliminate
 0.97% :     0.000001s :     10: predicate.transpose_eliminate
 1.20% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.42% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.54% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.90% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.46% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.48% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.34% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000825    12
51.09% :     0.000422s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.88% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.02% :     0.000363s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.041133    76
 0.15% :     0.000062s :      1: add_recomputation
 0.19% :     0.000077s :      1: auto_monad
 0.05% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 1.05% :     0.000430s :      1: bootstrap
 0.08% :     0.000035s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.06% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000011s :      1: environ_conv
 0.07% :     0.000030s :      1: event_method
 0.03% :     0.000014s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
25.99% :     0.010692s :      1: jit_opt_a
 0.44% :     0.000182s :      1: jit_opt_after_cconv
 0.15% :     0.000060s :      1: jit_opt_b
 1.37% :     0.000562s :      1: loop_unroll
 2.09% :     0.000859s :      1: mutable_eliminate
 2.46% :     0.001011s :     26: opt.transform.jit_opt_a
 0.13% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000029s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000047s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000045s :      4: opt.transform.symbol_engine_opt
11.93% :     0.004906s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000052s :      1: py_interpret_to_execute
 0.05% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 1.16% :     0.000477s :      1: renormalize.infer
 1.03% :     0.000425s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000049s :      1: rewriter_after_opt_a
 0.21% :     0.000088s :      1: rewriter_before_opt_a
 0.26% :     0.000108s :      1: symbol_engine_optimizer
28.10% :     0.011558s :      1: task_emit
21.90% :     0.009007s :      1: type_inference
 0.19% :     0.000079s :      1: validate


TotalTime = 0.0250522, [33]
[bootstrap]: 0.00034319
[type_inference]: 0.0049691
[event_method]: 1.895e-05
[auto_monad]: 6.425e-05
[graph_reusing]: 5.32001e-06
[pre_auto_parallel]: 2.47001e-06
[py_interpret_to_execute]: 3.795e-05
[rewriter_before_opt_a]: 7.976e-05
[expand_dump_flag]: 3.25e-06
[jit_opt_a]: 0.0087381, [2]
    [Cycle 1]: 0.00204759, [27]
        [switch_simplify]: 7.13e-05
        [loop_unroll]: 3.003e-05
        [a_1]: 0.00061991
        [with_stream_mark]: 2.432e-05
        [recompute_prepare]: 8.10999e-06
        [updatestate_depend_eliminate]: 3.64002e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.83998e-06
        [parameter_eliminate]: 1.98997e-06
        [specialize_transform]: 6.50997e-06
        [updatestate_useless_node_eliminater]: 5.41002e-06
        [accelerated_algorithm]: 5.69e-06
        [meta_shard_fg_expand]: 2.60002e-06
        [get_grad_eliminate_]: 5.17e-06
        [merge_forward]: 3.71999e-06
        [cell_reuse_recompute_pass]: 1.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.431e-05
        [j_node_and_user_rematch]: 9.56e-06
        [meta_fg_expand]: 3.00998e-06
        [replace_old_param]: 1.004e-05
        [inline_without_move]: 5.31002e-06
        [renormalize]: 0.00094417
        [add_forward_monad_depend]: 5.30999e-06
        [auto_monad_grad]: 2.69999e-06
        [auto_monad_eliminator]: 1.485e-05
        [cse]: 3.252e-05
        [replace_applicator]: 1.229e-05
    [Cycle 2]: 0.00032787, [27]
        [switch_simplify]: 5.70001e-06
        [loop_unroll]: 4.80001e-06
        [a_1]: 9.14e-05
        [with_stream_mark]: 1.134e-05
        [recompute_prepare]: 5.46e-06
        [updatestate_depend_eliminate]: 2.89001e-06
        [updatestate_assign_eliminate]: 2.05002e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 4.61002e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.49998e-06
        [meta_shard_fg_expand]: 1.40999e-06
        [get_grad_eliminate_]: 4.38001e-06
        [merge_forward]: 2.49001e-06
        [cell_reuse_recompute_pass]: 1.38002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.287e-05
        [j_node_and_user_rematch]: 7.93001e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 7.5e-06
        [inline_without_move]: 4.66002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 5.05999e-06
        [cse]: 1.048e-05
        [replace_applicator]: 4.79e-06
[py_interpret_to_execute_after_opt_a]: 1.227e-05
[rewriter_after_opt_a]: 3.665e-05
[convert_after_rewriter]: 7.71001e-06
[order_py_execute_after_rewriter]: 4.82998e-06
[mutable_eliminate]: 0.00061039
[jit_opt_b]: 5.525e-05, [1]
    [Cycle 1]: 4.776e-05, [2]
        [frontend_op_eliminate]: 2.088e-05
        [inline_after_opt_a]: 1.541e-05
[cconv]: 2.754e-05
[loop_unroll]: 0.00047291
[jit_opt_after_cconv]: 0.00014146, [1]
    [Cycle 1]: 0.00013433, [11]
        [c_1]: 2.06e-05
        [parameter_eliminate]: 2.84999e-06
        [updatestate_depend_eliminate]: 6.09999e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [cse]: 1.837e-05
        [call_graph_tuple_transform]: 1.956e-05
        [tuple_list_get_item_eliminator]: 5.00999e-06
        [none_parameter_eliminate]: 1.99e-06
        [renormalize]: 4.40021e-07
        [switch_simplify]: 5.05999e-06
[remove_dup_value]: 1.294e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 5.22999e-06
[add_recomputation]: 5.255e-05
[cse_after_recomputation]: 2.039e-05, [1]
    [Cycle 1]: 1.493e-05, [1]
        [cse]: 8.77999e-06
[auto_monad_reorder]: 1.507e-05
[get_jit_bprop_graph]: 2.03002e-06
[rewriter_after_jit_bprop_graph]: 5.52001e-06
[opt_after_jit_grad]: 0.00048079
[symbol_engine_optimizer]: 7.374e-05, [1]
    [Cycle 1]: 6.71e-05, [6]
        [build]: 4.60001e-06
        [elim_shapecalc]: 7.74002e-06
        [elim_not_effective]: 1.258e-05
        [opt_reshape]: 5.56e-06
        [fold_const_symbol]: 8.79e-06
        [renormalize]: 4.19997e-07
[validate]: 3.808e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.00847906
[execute]: 1.032e-05
Sums
    bootstrap                            :     0.000343s :  1.90%
    type_inference                       :     0.004969s : 27.56%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000064s :  0.36%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000038s :  0.21%
    rewriter_before_opt_a                :     0.000080s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000077s :  0.43%
    jit_opt_a.loop_unroll                :     0.000035s :  0.19%
    jit_opt_a.a_1                        :     0.000711s :  3.95%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000944s :  5.24%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.11%
    jit_opt_a.cse                        :     0.000043s :  0.24%
    jit_opt_a.replace_applicator         :     0.000017s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.20%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000610s :  3.39%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000028s :  0.15%
    loop_unroll                          :     0.000473s :  2.62%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000053s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000481s :  2.67%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008479s : 47.03%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000267    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000002s :      2: substitution.fold_const_symbol
 2.10% :     0.000006s :      3: substitution.graph_param_transform
86.02% :     0.000230s :      4: substitution.inline
 1.37% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.84% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.07% :     0.000006s :      2: substitution.replace_old_param
 5.24% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004897     2
81.25% :     0.003979s :      1: type_inference.infer
18.75% :     0.000918s :      1: type_inference.specialize
------[replace.]   0.000081     6
68.63% :     0.000055s :      4: replace.inline
31.37% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000240     6
94.70% :     0.000227s :      4: match.inline
 5.30% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.20% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.17% :     0.000003s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.45% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 6.05% :     0.000007s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000001s :      6: predicate.less_batch_normalization
 1.71% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.57% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 2.52% :     0.000003s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.32% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.85% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.21% :     0.000001s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.96% :     0.000009s :     53: predicate.switch_simplify
 1.53% :     0.000002s :     10: predicate.tile_eliminate
 1.33% :     0.000002s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.22% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.81% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000690    12
45.89% :     0.000317s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.99% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.12% :     0.000318s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027035    76
 0.21% :     0.000056s :      1: add_recomputation
 0.25% :     0.000068s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.34% :     0.000363s :      1: bootstrap
 0.11% :     0.000031s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000023s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
32.34% :     0.008742s :      1: jit_opt_a
 0.53% :     0.000144s :      1: jit_opt_after_cconv
 0.21% :     0.000058s :      1: jit_opt_b
 1.78% :     0.000482s :      1: loop_unroll
 2.29% :     0.000619s :      1: mutable_eliminate
 3.46% :     0.000936s :     26: opt.transform.jit_opt_a
 0.17% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000030s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.81% :     0.000489s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.15% :     0.000041s :      1: py_interpret_to_execute
 0.05% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000015s :      1: remove_dup_value
 1.91% :     0.000516s :      1: renormalize.infer
 1.55% :     0.000419s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000040s :      1: rewriter_after_opt_a
 0.31% :     0.000083s :      1: rewriter_before_opt_a
 0.28% :     0.000076s :      1: symbol_engine_optimizer
31.44% :     0.008500s :      1: task_emit
18.43% :     0.004984s :      1: type_inference
 0.23% :     0.000061s :      1: validate


TotalTime = 0.0214506, [33]
[bootstrap]: 0.00029951
[type_inference]: 0.00496692
[event_method]: 1.827e-05
[auto_monad]: 6.686e-05
[graph_reusing]: 5.97001e-06
[pre_auto_parallel]: 2.38998e-06
[py_interpret_to_execute]: 4.143e-05
[rewriter_before_opt_a]: 7.499e-05
[expand_dump_flag]: 3.45e-06
[jit_opt_a]: 0.00643186, [2]
    [Cycle 1]: 0.00188718, [27]
        [switch_simplify]: 5.374e-05
        [loop_unroll]: 3.101e-05
        [a_1]: 0.00051603
        [with_stream_mark]: 2.32e-05
        [recompute_prepare]: 7.44002e-06
        [updatestate_depend_eliminate]: 4.13001e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [parameter_eliminate]: 1.93002e-06
        [specialize_transform]: 6.81001e-06
        [updatestate_useless_node_eliminater]: 5.10999e-06
        [accelerated_algorithm]: 6.21998e-06
        [meta_shard_fg_expand]: 2.05002e-06
        [get_grad_eliminate_]: 5.46002e-06
        [merge_forward]: 4.03999e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.529e-05
        [j_node_and_user_rematch]: 9.76998e-06
        [meta_fg_expand]: 2.39999e-06
        [replace_old_param]: 9.62999e-06
        [inline_without_move]: 5.72001e-06
        [renormalize]: 0.00096794
        [add_forward_monad_depend]: 3.36999e-06
        [auto_monad_grad]: 1.09e-06
        [auto_monad_eliminator]: 8.44998e-06
        [cse]: 1.629e-05
        [replace_applicator]: 1.418e-05
    [Cycle 2]: 0.00038264, [27]
        [switch_simplify]: 5.52001e-06
        [loop_unroll]: 4.93001e-06
        [a_1]: 9.022e-05
        [with_stream_mark]: 1.06e-05
        [recompute_prepare]: 4.95999e-06
        [updatestate_depend_eliminate]: 2.27999e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 1.09998e-06
        [specialize_transform]: 4.64002e-06
        [updatestate_useless_node_eliminater]: 4.80999e-06
        [accelerated_algorithm]: 5.00999e-06
        [meta_shard_fg_expand]: 1.05001e-06
        [get_grad_eliminate_]: 4.83001e-06
        [merge_forward]: 3.04001e-06
        [cell_reuse_recompute_pass]: 2.19001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 6.466e-05
        [j_node_and_user_rematch]: 8.57e-06
        [meta_fg_expand]: 1.78002e-06
        [replace_old_param]: 7.92e-06
        [inline_without_move]: 4.63999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.82999e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 5.69999e-06
        [cse]: 1.123e-05
        [replace_applicator]: 6.14999e-06
[py_interpret_to_execute_after_opt_a]: 1.325e-05
[rewriter_after_opt_a]: 3.695e-05
[convert_after_rewriter]: 7.4e-06
[order_py_execute_after_rewriter]: 4.88001e-06
[mutable_eliminate]: 0.00076118
[jit_opt_b]: 5.324e-05, [1]
    [Cycle 1]: 4.504e-05, [2]
        [frontend_op_eliminate]: 1.705e-05
        [inline_after_opt_a]: 1.479e-05
[cconv]: 3.072e-05
[loop_unroll]: 0.0004505
[jit_opt_after_cconv]: 0.00014679, [1]
    [Cycle 1]: 0.00014021, [11]
        [c_1]: 2.065e-05
        [parameter_eliminate]: 3.26999e-06
        [updatestate_depend_eliminate]: 5.52001e-06
        [updatestate_assign_eliminate]: 2.41998e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 2.289e-05
        [call_graph_tuple_transform]: 2.017e-05
        [tuple_list_get_item_eliminator]: 5.59e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 6.89994e-07
        [switch_simplify]: 4.90001e-06
[remove_dup_value]: 1.563e-05
[partial_unused_args_eliminate]: 2.22001e-06
[environ_conv]: 6.19001e-06
[add_recomputation]: 5.345e-05
[cse_after_recomputation]: 2.066e-05, [1]
    [Cycle 1]: 1.534e-05, [1]
        [cse]: 9.87001e-06
[auto_monad_reorder]: 1.687e-05
[get_jit_bprop_graph]: 1.99e-06
[rewriter_after_jit_bprop_graph]: 5.89e-06
[opt_after_jit_grad]: 0.00047537
[symbol_engine_optimizer]: 7.498e-05, [1]
    [Cycle 1]: 6.795e-05, [6]
        [build]: 3.95e-06
        [elim_shapecalc]: 7.57998e-06
        [elim_not_effective]: 1.319e-05
        [opt_reshape]: 5.90002e-06
        [fold_const_symbol]: 8.35999e-06
        [renormalize]: 1.22e-06
[validate]: 4.16e-05
[backend_pass]: 7.89994e-07
[task_emit]: 0.00709546
[execute]: 1.028e-05
Sums
    bootstrap                            :     0.000300s :  1.80%
    type_inference                       :     0.004967s : 29.79%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000067s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.25%
    rewriter_before_opt_a                :     0.000075s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.36%
    jit_opt_a.loop_unroll                :     0.000036s :  0.22%
    jit_opt_a.a_1                        :     0.000606s :  3.64%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000080s :  0.48%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000968s :  5.81%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.08%
    jit_opt_a.cse                        :     0.000028s :  0.17%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000037s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000761s :  4.57%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000031s :  0.18%
    loop_unroll                          :     0.000450s :  2.70%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000053s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000475s :  2.85%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000042s :  0.25%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007095s : 42.56%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000241    23
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.64% :     0.000006s :      3: substitution.graph_param_transform
63.34% :     0.000153s :      4: substitution.inline
 1.52% :     0.000004s :      4: substitution.j_node_and_user_rematch
23.69% :     0.000057s :      4: substitution.remove_not_recompute_node
 1.82% :     0.000004s :      2: substitution.replace_old_param
 5.69% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004894     2
80.91% :     0.003959s :      1: type_inference.infer
19.09% :     0.000934s :      1: type_inference.specialize
------[replace.]   0.000069     6
63.51% :     0.000044s :      4: replace.inline
36.49% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000163     6
92.22% :     0.000150s :      4: match.inline
 7.78% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.83% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.42% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 1.85% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.64% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 4.71% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.28% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.29% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 2.12% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000000s :      3: predicate.opt_reshape
 2.00% :     0.000002s :     15: predicate.partial_eliminate
 1.35% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.73% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.50% :     0.000001s :      3: predicate.reset_defer_inline
 1.29% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.93% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.62% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.30% :     0.000009s :     53: predicate.switch_simplify
 1.37% :     0.000002s :     10: predicate.tile_eliminate
 1.33% :     0.000002s :     10: predicate.transpose_eliminate
 1.71% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.43% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.66% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000748    12
55.18% :     0.000413s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.04% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.78% :     0.000298s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023401    76
 0.24% :     0.000056s :      1: add_recomputation
 0.30% :     0.000070s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.32% :     0.000310s :      1: bootstrap
 0.14% :     0.000033s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
27.51% :     0.006437s :      1: jit_opt_a
 0.64% :     0.000149s :      1: jit_opt_after_cconv
 0.24% :     0.000056s :      1: jit_opt_b
 1.96% :     0.000458s :      1: loop_unroll
 3.30% :     0.000772s :      1: mutable_eliminate
 3.75% :     0.000877s :     26: opt.transform.jit_opt_a
 0.20% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000026s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.07% :     0.000484s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000044s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 2.66% :     0.000622s :      1: renormalize.infer
 1.44% :     0.000337s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000040s :      1: rewriter_after_opt_a
 0.33% :     0.000078s :      1: rewriter_before_opt_a
 0.33% :     0.000077s :      1: symbol_engine_optimizer
30.40% :     0.007115s :      1: task_emit
21.29% :     0.004981s :      1: type_inference
 0.27% :     0.000062s :      1: validate


TotalTime = 0.0154259, [30]
[bootstrap]: 0.00031545
[type_inference]: 0.00685891
[event_method]: 2.112e-05
[auto_monad]: 8.114e-05
[graph_reusing]: 6.39999e-06
[pre_auto_parallel]: 3.31999e-06
[py_interpret_to_execute]: 4.528e-05
[rewriter_before_opt_a]: 8.926e-05
[expand_dump_flag]: 2.93998e-06
[jit_opt_a]: 0.00592553, [2]
    [Cycle 1]: 0.00208141, [27]
        [switch_simplify]: 0.00013105
        [loop_unroll]: 3.383e-05
        [a_1]: 0.00077411
        [with_stream_mark]: 2.379e-05
        [recompute_prepare]: 7.55e-06
        [updatestate_depend_eliminate]: 3.61999e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.66e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 6.39001e-06
        [updatestate_useless_node_eliminater]: 5.29998e-06
        [accelerated_algorithm]: 6.32001e-06
        [meta_shard_fg_expand]: 2.16e-06
        [get_grad_eliminate_]: 5.80002e-06
        [merge_forward]: 3.95e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.505e-05
        [j_node_and_user_rematch]: 1.083e-05
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 1.019e-05
        [inline_without_move]: 6.04001e-06
        [renormalize]: 0.00076339
        [add_forward_monad_depend]: 6.26e-06
        [auto_monad_grad]: 2.21998e-06
        [auto_monad_eliminator]: 1.388e-05
        [cse]: 3.169e-05
        [replace_applicator]: 1.133e-05
    [Cycle 2]: 0.00032328, [27]
        [switch_simplify]: 5.61e-06
        [loop_unroll]: 5.04e-06
        [a_1]: 8.786e-05
        [with_stream_mark]: 1.018e-05
        [recompute_prepare]: 4.79e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 2.18002e-06
        [updatestate_loads_eliminate]: 1.81003e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 4.82e-06
        [updatestate_useless_node_eliminater]: 4.94998e-06
        [accelerated_algorithm]: 4.99e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 1.24998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.99e-05
        [j_node_and_user_rematch]: 8.3e-06
        [meta_fg_expand]: 1.51998e-06
        [replace_old_param]: 7.63999e-06
        [inline_without_move]: 4.94998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.27999e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 5.15999e-06
        [cse]: 1.065e-05
        [replace_applicator]: 4.90999e-06
[py_interpret_to_execute_after_opt_a]: 1.193e-05
[rewriter_after_opt_a]: 3.862e-05
[convert_after_rewriter]: 7.25003e-06
[order_py_execute_after_rewriter]: 5.25001e-06
[mutable_eliminate]: 0.00055981
[jit_opt_b]: 4.888e-05, [1]
    [Cycle 1]: 4.147e-05, [2]
        [frontend_op_eliminate]: 1.569e-05
        [inline_after_opt_a]: 1.474e-05
[cconv]: 2.503e-05
[loop_unroll]: 0.00040898
[jit_opt_after_cconv]: 0.00013728, [1]
    [Cycle 1]: 0.00013103, [11]
        [c_1]: 2.004e-05
        [parameter_eliminate]: 2.61999e-06
        [updatestate_depend_eliminate]: 5.61998e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.843e-05
        [call_graph_tuple_transform]: 1.73e-05
        [tuple_list_get_item_eliminator]: 5.37999e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 4.2998e-07
        [switch_simplify]: 5.40001e-06
[remove_dup_value]: 1.363e-05
[partial_unused_args_eliminate]: 2.45002e-06
[environ_conv]: 4.63999e-06
[add_recomputation]: 4.561e-05
[cse_after_recomputation]: 1.949e-05, [1]
    [Cycle 1]: 1.445e-05, [1]
        [cse]: 8.99998e-06
[auto_monad_reorder]: 1.546e-05
[get_jit_bprop_graph]: 1.92999e-06
[rewriter_after_jit_bprop_graph]: 5.09998e-06
[opt_after_jit_grad]: 0.00044304
[symbol_engine_optimizer]: 7.15e-05, [1]
    [Cycle 1]: 6.526e-05, [6]
        [build]: 3.84002e-06
        [elim_shapecalc]: 7.08998e-06
        [elim_not_effective]: 1.227e-05
        [opt_reshape]: 5.62999e-06
        [fold_const_symbol]: 8.77e-06
        [renormalize]: 3.39991e-07
[validate]: 3.339e-05
Sums
    bootstrap                            :     0.000315s :  2.79%
    type_inference                       :     0.006859s : 60.69%
    event_method                         :     0.000021s :  0.19%
    auto_monad                           :     0.000081s :  0.72%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000045s :  0.40%
    rewriter_before_opt_a                :     0.000089s :  0.79%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000137s :  1.21%
    jit_opt_a.loop_unroll                :     0.000039s :  0.34%
    jit_opt_a.a_1                        :     0.000862s :  7.63%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.16%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000763s :  6.76%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.17%
    jit_opt_a.cse                        :     0.000042s :  0.37%
    jit_opt_a.replace_applicator         :     0.000016s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.11%
    rewriter_after_opt_a                 :     0.000039s :  0.34%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000560s :  4.95%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.13%
    cconv                                :     0.000025s :  0.22%
    loop_unroll                          :     0.000409s :  3.62%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000046s :  0.40%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000443s :  3.92%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.30%


Time group info:
------[substitution.]   0.000309    26
 0.65% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 1.52% :     0.000005s :      3: substitution.graph_param_transform
83.83% :     0.000259s :      6: substitution.inline
 1.12% :     0.000003s :      4: substitution.j_node_and_user_rematch
 3.56% :     0.000011s :      4: substitution.remove_not_recompute_node
 1.60% :     0.000005s :      2: substitution.replace_old_param
 2.97% :     0.000009s :      1: substitution.switch_simplify
 4.27% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006787     2
84.77% :     0.005753s :      1: type_inference.infer
15.23% :     0.001034s :      1: type_inference.specialize
------[replace.]   0.000146     9
46.42% :     0.000068s :      6: replace.inline
28.79% :     0.000042s :      1: replace.switch_simplify
24.79% :     0.000036s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000275     9
92.65% :     0.000255s :      6: match.inline
 3.00% :     0.000008s :      1: match.switch_simplify
 4.35% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   801
 1.53% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.11% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     12: predicate.addn_check_dump
 1.41% :     0.000002s :     12: predicate.addn_zero_filter
 1.88% :     0.000002s :     12: predicate.arithmetic_simplify
 1.28% :     0.000002s :     12: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000002s :     12: predicate.compare_switch_simplify
 1.37% :     0.000002s :     12: predicate.depend_value_elim
 1.17% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.50% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      3: predicate.elim_not_effective
 0.44% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     12: predicate.environ_get_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.81% :     0.000008s :     26: predicate.inline
 1.05% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.74% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     17: predicate.load_eliminater
 1.48% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.29% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.97% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     12: predicate.merge_addn
 1.24% :     0.000002s :     12: predicate.minmaximum_grad
 1.44% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     17: predicate.partial_eliminate
 1.24% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     12: predicate.reduce_eliminate
 1.44% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     20: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     12: predicate.reshape_eliminate
 1.23% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.62% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     12: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.74% :     0.000004s :     20: predicate.switch_defer_inline
 2.41% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.26% :     0.000011s :     65: predicate.switch_simplify
 1.28% :     0.000002s :     12: predicate.tile_eliminate
 1.24% :     0.000002s :     12: predicate.transpose_eliminate
 1.50% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.52% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.61% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.37% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.61% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     12: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000808    15
49.31% :     0.000398s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.60% :     0.000110s :      1: func_graph_cloner_run.FuncGraphClonerNode
37.09% :     0.000300s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017401    72
 0.28% :     0.000048s :      1: add_recomputation
 0.49% :     0.000085s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 1.94% :     0.000337s :      1: bootstrap
 0.16% :     0.000028s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.14% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
34.07% :     0.005929s :      1: jit_opt_a
 0.81% :     0.000140s :      1: jit_opt_after_cconv
 0.30% :     0.000051s :      1: jit_opt_b
 2.39% :     0.000416s :      1: loop_unroll
 3.26% :     0.000568s :      1: mutable_eliminate
 6.68% :     0.001162s :     26: opt.transform.jit_opt_a
 0.26% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.59% :     0.000451s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000048s :      1: py_interpret_to_execute
 0.08% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 2.51% :     0.000437s :      1: renormalize.infer
 1.83% :     0.000318s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000042s :      1: rewriter_after_opt_a
 0.53% :     0.000092s :      1: rewriter_before_opt_a
 0.43% :     0.000074s :      1: symbol_engine_optimizer
39.51% :     0.006875s :      1: type_inference


TotalTime = 0.0514201, [33]
[bootstrap]: 0.00052562
[type_inference]: 0.00496624
[event_method]: 1.837e-05
[auto_monad]: 6.669e-05
[graph_reusing]: 5.39e-06
[pre_auto_parallel]: 2.58e-06
[py_interpret_to_execute]: 4.113e-05
[rewriter_before_opt_a]: 7.525e-05
[expand_dump_flag]: 2.91999e-06
[jit_opt_a]: 0.026768, [2]
    [Cycle 1]: 0.00203203, [27]
        [switch_simplify]: 7.497e-05
        [loop_unroll]: 3.505e-05
        [a_1]: 0.00062304
        [with_stream_mark]: 1.624e-05
        [recompute_prepare]: 7.50998e-06
        [updatestate_depend_eliminate]: 3.24001e-06
        [updatestate_assign_eliminate]: 2.94001e-06
        [updatestate_loads_eliminate]: 2.68003e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 7.18998e-06
        [updatestate_useless_node_eliminater]: 6.47001e-06
        [accelerated_algorithm]: 6.53e-06
        [meta_shard_fg_expand]: 2.27001e-06
        [get_grad_eliminate_]: 5.47999e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 2.43998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.289e-05
        [j_node_and_user_rematch]: 1.041e-05
        [meta_fg_expand]: 1.84e-06
        [replace_old_param]: 9.02e-06
        [inline_without_move]: 6.17999e-06
        [renormalize]: 0.00092927
        [add_forward_monad_depend]: 7.49002e-06
        [auto_monad_grad]: 2.58e-06
        [auto_monad_eliminator]: 1.562e-05
        [cse]: 2.974e-05
        [replace_applicator]: 1.46e-05
    [Cycle 2]: 0.020717, [27]
        [switch_simplify]: 6.06e-06
        [loop_unroll]: 5.00999e-06
        [a_1]: 9.536e-05
        [with_stream_mark]: 7.10998e-06
        [recompute_prepare]: 5.25001e-06
        [updatestate_depend_eliminate]: 2.31e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 6.50005e-07
        [specialize_transform]: 5.32001e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 8.79983e-07
        [get_grad_eliminate_]: 4.78001e-06
        [merge_forward]: 2.19999e-06
        [cell_reuse_recompute_pass]: 2.11e-06
        [cell_reuse_handle_not_recompute_node_pass]: 0.0202168
        [j_node_and_user_rematch]: 2.857e-05
        [meta_fg_expand]: 7.41001e-06
        [replace_old_param]: 1.611e-05
        [inline_without_move]: 6.89999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.342e-05
        [auto_monad_grad]: 3.22002e-06
        [auto_monad_eliminator]: 2.931e-05
        [cse]: 4.3e-05
        [replace_applicator]: 6.55002e-06
[py_interpret_to_execute_after_opt_a]: 2.745e-05
[rewriter_after_opt_a]: 6.089e-05
[convert_after_rewriter]: 8.75001e-06
[order_py_execute_after_rewriter]: 5.45001e-06
[mutable_eliminate]: 0.00109842
[jit_opt_b]: 7.098e-05, [1]
    [Cycle 1]: 5.894e-05, [2]
        [frontend_op_eliminate]: 2.743e-05
        [inline_after_opt_a]: 1.898e-05
[cconv]: 3.95e-05
[loop_unroll]: 0.0005086
[jit_opt_after_cconv]: 0.00018262, [1]
    [Cycle 1]: 0.00017398, [11]
        [c_1]: 2.731e-05
        [parameter_eliminate]: 6.23998e-06
        [updatestate_depend_eliminate]: 8.42e-06
        [updatestate_assign_eliminate]: 3.64002e-06
        [updatestate_loads_eliminate]: 3.25002e-06
        [cse]: 3.102e-05
        [call_graph_tuple_transform]: 2.45e-05
        [tuple_list_get_item_eliminator]: 6.21e-06
        [none_parameter_eliminate]: 2.14e-06
        [renormalize]: 9.89996e-07
        [switch_simplify]: 6.61e-06
[remove_dup_value]: 1.61e-05
[partial_unused_args_eliminate]: 2.56e-06
[environ_conv]: 7.55e-06
[add_recomputation]: 6.598e-05
[cse_after_recomputation]: 2.498e-05, [1]
    [Cycle 1]: 1.84e-05, [1]
        [cse]: 1.183e-05
[auto_monad_reorder]: 1.774e-05
[get_jit_bprop_graph]: 2.58e-06
[rewriter_after_jit_bprop_graph]: 9.07999e-06
[opt_after_jit_grad]: 0.00053391
[symbol_engine_optimizer]: 8.34e-05, [1]
    [Cycle 1]: 7.639e-05, [6]
        [build]: 4.92e-06
        [elim_shapecalc]: 9.09998e-06
        [elim_not_effective]: 1.484e-05
        [opt_reshape]: 5.92001e-06
        [fold_const_symbol]: 1.023e-05
        [renormalize]: 7.89994e-07
[validate]: 5.596e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.0158516
[execute]: 9.46003e-06
Sums
    bootstrap                            :     0.000526s :  1.13%
    type_inference                       :     0.004966s : 10.65%
    event_method                         :     0.000018s :  0.04%
    auto_monad                           :     0.000067s :  0.14%
    graph_reusing                        :     0.000005s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.09%
    rewriter_before_opt_a                :     0.000075s :  0.16%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000081s :  0.17%
    jit_opt_a.loop_unroll                :     0.000040s :  0.09%
    jit_opt_a.a_1                        :     0.000718s :  1.54%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.05%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000013s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.02%
    jit_opt_a.merge_forward              :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.020230s : 43.40%
    jit_opt_a.j_node_and_user_rematch    :     0.000039s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000009s :  0.02%
    jit_opt_a.replace_old_param          :     0.000025s :  0.05%
    jit_opt_a.inline_without_move        :     0.000013s :  0.03%
    jit_opt_a.renormalize                :     0.000929s :  1.99%
    jit_opt_a.add_forward_monad_depend   :     0.000021s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000006s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000045s :  0.10%
    jit_opt_a.cse                        :     0.000073s :  0.16%
    jit_opt_a.replace_applicator         :     0.000021s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000027s :  0.06%
    rewriter_after_opt_a                 :     0.000061s :  0.13%
    convert_after_rewriter               :     0.000009s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.001098s :  2.36%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.04%
    cconv                                :     0.000040s :  0.08%
    loop_unroll                          :     0.000509s :  1.09%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000031s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.01%
    remove_dup_value                     :     0.000016s :  0.03%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000008s :  0.02%
    add_recomputation                    :     0.000066s :  0.14%
    cse_after_recomputation.cse          :     0.000012s :  0.03%
    auto_monad_reorder                   :     0.000018s :  0.04%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.02%
    opt_after_jit_grad                   :     0.000534s :  1.15%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000056s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.015852s : 34.01%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000302    23
 0.87% :     0.000003s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 2.21% :     0.000007s :      3: substitution.graph_param_transform
74.19% :     0.000224s :      4: substitution.inline
 2.24% :     0.000007s :      4: substitution.j_node_and_user_rematch
13.63% :     0.000041s :      4: substitution.remove_not_recompute_node
 3.15% :     0.000010s :      2: substitution.replace_old_param
 3.27% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004894     2
80.90% :     0.003959s :      1: type_inference.infer
19.10% :     0.000935s :      1: type_inference.specialize
------[replace.]   0.000079     6
66.90% :     0.000053s :      4: replace.inline
33.10% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000231     6
96.18% :     0.000222s :      4: match.inline
 3.82% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   691
 1.27% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.46% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.89% :     0.000001s :     10: predicate.addn_check_dump
 1.16% :     0.000002s :     10: predicate.addn_zero_filter
 1.85% :     0.000003s :     10: predicate.arithmetic_simplify
 1.07% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.07% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.18% :     0.000002s :      6: predicate.get_grad_eliminate
 0.34% :     0.000000s :      3: predicate.graph_param_transform
 5.26% :     0.000007s :     22: predicate.inline
 1.45% :     0.000002s :      6: predicate.inline_without_move
 0.42% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.59% :     0.000002s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.23% :     0.000003s :     15: predicate.load_eliminater
 1.67% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000002s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.13% :     0.000003s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 1.92% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.27% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 1.24% :     0.000002s :      6: predicate.remove_not_recompute_node
 1.48% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.39% :     0.000001s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.10% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.62% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.19% :     0.000002s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.14% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.90% :     0.000011s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.06% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.58% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.79% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.57% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.45% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.93% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000683    12
46.78% :     0.000319s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.53% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.69% :     0.000312s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.073646    76
 0.09% :     0.000070s :      1: add_recomputation
 0.10% :     0.000070s :      1: auto_monad
 0.03% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.74% :     0.000547s :      1: bootstrap
 0.06% :     0.000042s :      1: cconv
 0.02% :     0.000011s :      1: convert_after_rewriter
 0.04% :     0.000027s :      1: cse_after_recomputation
 0.01% :     0.000010s :      1: environ_conv
 0.03% :     0.000023s :      1: event_method
 0.02% :     0.000015s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000008s :      1: graph_reusing
36.36% :     0.026775s :      1: jit_opt_a
 0.25% :     0.000186s :      1: jit_opt_after_cconv
 0.10% :     0.000074s :      1: jit_opt_b
 0.70% :     0.000518s :      1: loop_unroll
 1.51% :     0.001112s :      1: mutable_eliminate
28.76% :     0.021178s :     26: opt.transform.jit_opt_a
 0.08% :     0.000060s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000037s :      4: opt.transform.jit_opt_b
 0.02% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.05% :     0.000037s :      4: opt.transform.symbol_engine_opt
 0.74% :     0.000544s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.06% :     0.000044s :      1: py_interpret_to_execute
 0.04% :     0.000030s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000019s :      1: remove_dup_value
 0.70% :     0.000519s :      1: renormalize.infer
 0.55% :     0.000402s :      1: renormalize.specialize
 0.02% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000065s :      1: rewriter_after_opt_a
 0.11% :     0.000078s :      1: rewriter_before_opt_a
 0.12% :     0.000086s :      1: symbol_engine_optimizer
21.56% :     0.015875s :      1: task_emit
 6.76% :     0.004982s :      1: type_inference
 0.11% :     0.000082s :      1: validate


TotalTime = 0.0149487, [30]
[bootstrap]: 0.00028493
[type_inference]: 0.00656203
[event_method]: 2.236e-05
[auto_monad]: 7.977e-05
[graph_reusing]: 6.29001e-06
[pre_auto_parallel]: 3.04999e-06
[py_interpret_to_execute]: 5.188e-05
[rewriter_before_opt_a]: 8.579e-05
[expand_dump_flag]: 2.93003e-06
[jit_opt_a]: 0.00574835, [2]
    [Cycle 1]: 0.00196014, [27]
        [switch_simplify]: 0.00012656
        [loop_unroll]: 3.35e-05
        [a_1]: 0.0007261
        [with_stream_mark]: 2.125e-05
        [recompute_prepare]: 7.07002e-06
        [updatestate_depend_eliminate]: 3.78001e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 2.53998e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 6.13998e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 5.31002e-06
        [meta_shard_fg_expand]: 1.93002e-06
        [get_grad_eliminate_]: 5.20001e-06
        [merge_forward]: 3.81001e-06
        [cell_reuse_recompute_pass]: 1.02998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.504e-05
        [j_node_and_user_rematch]: 9.14e-06
        [meta_fg_expand]: 2.07001e-06
        [replace_old_param]: 9.24e-06
        [inline_without_move]: 5.22e-06
        [renormalize]: 0.00072081
        [add_forward_monad_depend]: 5.28002e-06
        [auto_monad_grad]: 2.60002e-06
        [auto_monad_eliminator]: 1.296e-05
        [cse]: 3.075e-05
        [replace_applicator]: 1.209e-05
    [Cycle 2]: 0.00030893, [27]
        [switch_simplify]: 5.79999e-06
        [loop_unroll]: 5.17e-06
        [a_1]: 9.007e-05
        [with_stream_mark]: 8.48999e-06
        [recompute_prepare]: 4.84998e-06
        [updatestate_depend_eliminate]: 2.49001e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 5.02999e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 4.65001e-06
        [meta_shard_fg_expand]: 1.20001e-06
        [get_grad_eliminate_]: 4.63999e-06
        [merge_forward]: 2.27999e-06
        [cell_reuse_recompute_pass]: 1.07998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.207e-05
        [j_node_and_user_rematch]: 8.12e-06
        [meta_fg_expand]: 1.48002e-06
        [replace_old_param]: 7e-06
        [inline_without_move]: 4.80001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 6.19999e-07
        [auto_monad_eliminator]: 4.65001e-06
        [cse]: 8.88002e-06
        [replace_applicator]: 5.02e-06
[py_interpret_to_execute_after_opt_a]: 1.031e-05
[rewriter_after_opt_a]: 3.428e-05
[convert_after_rewriter]: 6.20002e-06
[order_py_execute_after_rewriter]: 4.82e-06
[mutable_eliminate]: 0.00055426
[jit_opt_b]: 4.831e-05, [1]
    [Cycle 1]: 4.157e-05, [2]
        [frontend_op_eliminate]: 1.584e-05
        [inline_after_opt_a]: 1.442e-05
[cconv]: 2.464e-05
[loop_unroll]: 0.00043058
[jit_opt_after_cconv]: 0.00013975, [1]
    [Cycle 1]: 0.00013356, [11]
        [c_1]: 2.162e-05
        [parameter_eliminate]: 2.94999e-06
        [updatestate_depend_eliminate]: 5.50001e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [cse]: 1.859e-05
        [call_graph_tuple_transform]: 1.833e-05
        [tuple_list_get_item_eliminator]: 5.12999e-06
        [none_parameter_eliminate]: 1.57999e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 5.07999e-06
[remove_dup_value]: 1.276e-05
[partial_unused_args_eliminate]: 2.07999e-06
[environ_conv]: 4.92e-06
[add_recomputation]: 4.695e-05
[cse_after_recomputation]: 2.035e-05, [1]
    [Cycle 1]: 1.502e-05, [1]
        [cse]: 9.25999e-06
[auto_monad_reorder]: 1.453e-05
[get_jit_bprop_graph]: 1.95001e-06
[rewriter_after_jit_bprop_graph]: 4.90999e-06
[opt_after_jit_grad]: 0.00045862
[symbol_engine_optimizer]: 7.103e-05, [1]
    [Cycle 1]: 6.516e-05, [6]
        [build]: 4.05e-06
        [elim_shapecalc]: 7.12002e-06
        [elim_not_effective]: 1.238e-05
        [opt_reshape]: 5.40001e-06
        [fold_const_symbol]: 8.17e-06
        [renormalize]: 4.69998e-07
[validate]: 3.292e-05
Sums
    bootstrap                            :     0.000285s :  2.62%
    type_inference                       :     0.006562s : 60.29%
    event_method                         :     0.000022s :  0.21%
    auto_monad                           :     0.000080s :  0.73%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000052s :  0.48%
    rewriter_before_opt_a                :     0.000086s :  0.79%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000132s :  1.22%
    jit_opt_a.loop_unroll                :     0.000039s :  0.36%
    jit_opt_a.a_1                        :     0.000816s :  7.50%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.000721s :  6.62%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.16%
    jit_opt_a.cse                        :     0.000040s :  0.36%
    jit_opt_a.replace_applicator         :     0.000017s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.09%
    rewriter_after_opt_a                 :     0.000034s :  0.31%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000554s :  5.09%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.13%
    cconv                                :     0.000025s :  0.23%
    loop_unroll                          :     0.000431s :  3.96%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000013s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000047s :  0.43%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000459s :  4.21%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.30%


Time group info:
------[substitution.]   0.000321    26
 0.56% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000002s :      2: substitution.fold_const_symbol
 1.82% :     0.000006s :      3: substitution.graph_param_transform
86.86% :     0.000278s :      6: substitution.inline
 0.92% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.49% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.34% :     0.000004s :      2: substitution.replace_old_param
 2.65% :     0.000008s :      1: substitution.switch_simplify
 3.88% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006489     2
83.57% :     0.005423s :      1: type_inference.infer
16.43% :     0.001066s :      1: type_inference.specialize
------[replace.]   0.000139     9
48.51% :     0.000067s :      6: replace.inline
28.69% :     0.000040s :      1: replace.switch_simplify
22.79% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000294     9
93.54% :     0.000275s :      6: match.inline
 2.61% :     0.000008s :      1: match.switch_simplify
 3.85% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   801
 1.39% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     12: predicate.addn_check_dump
 1.42% :     0.000002s :     12: predicate.addn_zero_filter
 2.05% :     0.000003s :     12: predicate.arithmetic_simplify
 1.21% :     0.000002s :     12: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     12: predicate.compare_switch_simplify
 1.51% :     0.000002s :     12: predicate.depend_value_elim
 1.17% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.51% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.39% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.19% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     12: predicate.environ_get_eliminate
 1.17% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.45% :     0.000007s :     26: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.95% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     17: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.50% :     0.000006s :     40: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     15: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000002s :     12: predicate.merge_addn
 1.16% :     0.000002s :     12: predicate.minmaximum_grad
 1.76% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     17: predicate.partial_eliminate
 1.26% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     12: predicate.reduce_eliminate
 1.45% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     20: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     12: predicate.reshape_eliminate
 1.32% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     12: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.76% :     0.000001s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.99% :     0.000004s :     20: predicate.switch_defer_inline
 2.34% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.70% :     0.000011s :     65: predicate.switch_simplify
 1.29% :     0.000002s :     12: predicate.tile_eliminate
 1.40% :     0.000002s :     12: predicate.transpose_eliminate
 1.43% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.46% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     12: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000833    15
47.15% :     0.000393s :      6: func_graph_cloner_run.FuncGraphClonerGraph
15.87% :     0.000132s :      1: func_graph_cloner_run.FuncGraphClonerNode
36.98% :     0.000308s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016819    72
 0.30% :     0.000050s :      1: add_recomputation
 0.49% :     0.000083s :      1: auto_monad
 0.10% :     0.000017s :      1: auto_monad_reorder
 1.82% :     0.000306s :      1: bootstrap
 0.16% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.16% :     0.000027s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
34.20% :     0.005752s :      1: jit_opt_a
 0.85% :     0.000142s :      1: jit_opt_after_cconv
 0.30% :     0.000051s :      1: jit_opt_b
 2.61% :     0.000438s :      1: loop_unroll
 3.35% :     0.000563s :      1: mutable_eliminate
 6.53% :     0.001099s :     26: opt.transform.jit_opt_a
 0.28% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.77% :     0.000466s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.32% :     0.000055s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000015s :      1: remove_dup_value
 2.45% :     0.000412s :      1: renormalize.infer
 1.79% :     0.000302s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000037s :      1: rewriter_after_opt_a
 0.53% :     0.000089s :      1: rewriter_before_opt_a
 0.44% :     0.000073s :      1: symbol_engine_optimizer
39.13% :     0.006581s :      1: type_inference


TotalTime = 0.0442086, [33]
[bootstrap]: 0.00035435
[type_inference]: 0.00492161
[event_method]: 1.819e-05
[auto_monad]: 6.751e-05
[graph_reusing]: 6.16e-06
[pre_auto_parallel]: 2.53e-06
[py_interpret_to_execute]: 3.56e-05
[rewriter_before_opt_a]: 7.351e-05
[expand_dump_flag]: 3.33e-06
[jit_opt_a]: 0.00880504, [2]
    [Cycle 1]: 0.00438889, [27]
        [switch_simplify]: 0.00018982
        [loop_unroll]: 0.00012989
        [a_1]: 0.00180311
        [with_stream_mark]: 5.165e-05
        [recompute_prepare]: 2.831e-05
        [updatestate_depend_eliminate]: 1.307e-05
        [updatestate_assign_eliminate]: 1.245e-05
        [updatestate_loads_eliminate]: 8.05e-06
        [parameter_eliminate]: 6.74001e-06
        [specialize_transform]: 2.754e-05
        [updatestate_useless_node_eliminater]: 2.129e-05
        [accelerated_algorithm]: 2.195e-05
        [meta_shard_fg_expand]: 7.20003e-06
        [get_grad_eliminate_]: 2.656e-05
        [merge_forward]: 1.404e-05
        [cell_reuse_recompute_pass]: 2.04999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 5.458e-05
        [j_node_and_user_rematch]: 1.341e-05
        [meta_fg_expand]: 2.91e-06
        [replace_old_param]: 1.097e-05
        [inline_without_move]: 6.14999e-06
        [renormalize]: 0.00110473
        [add_forward_monad_depend]: 1.062e-05
        [auto_monad_grad]: 6.98e-06
        [auto_monad_eliminator]: 4.111e-05
        [cse]: 7.434e-05
        [replace_applicator]: 3.774e-05
    [Cycle 2]: 0.00106021, [27]
        [switch_simplify]: 2.178e-05
        [loop_unroll]: 2.124e-05
        [a_1]: 0.00030686
        [with_stream_mark]: 3.469e-05
        [recompute_prepare]: 2.141e-05
        [updatestate_depend_eliminate]: 1.181e-05
        [updatestate_assign_eliminate]: 7.36999e-06
        [updatestate_loads_eliminate]: 7.40998e-06
        [parameter_eliminate]: 1.72999e-06
        [specialize_transform]: 2.255e-05
        [updatestate_useless_node_eliminater]: 2.217e-05
        [accelerated_algorithm]: 2.741e-05
        [meta_shard_fg_expand]: 5.46e-06
        [get_grad_eliminate_]: 1.564e-05
        [merge_forward]: 8.72e-06
        [cell_reuse_recompute_pass]: 6.71999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 4.751e-05
        [j_node_and_user_rematch]: 9.86e-06
        [meta_fg_expand]: 2.70002e-06
        [replace_old_param]: 9.02999e-06
        [inline_without_move]: 1.159e-05
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.35002e-06
        [auto_monad_grad]: 1.81998e-06
        [auto_monad_eliminator]: 8.17e-06
        [cse]: 1.737e-05
        [replace_applicator]: 6.34999e-06
[py_interpret_to_execute_after_opt_a]: 1.151e-05
[rewriter_after_opt_a]: 3.639e-05
[convert_after_rewriter]: 6.96999e-06
[order_py_execute_after_rewriter]: 4.96002e-06
[mutable_eliminate]: 0.00086476
[jit_opt_b]: 5.428e-05, [1]
    [Cycle 1]: 4.597e-05, [2]
        [frontend_op_eliminate]: 1.804e-05
        [inline_after_opt_a]: 1.598e-05
[cconv]: 2.975e-05
[loop_unroll]: 0.00045401
[jit_opt_after_cconv]: 0.0001474, [1]
    [Cycle 1]: 0.00014102, [11]
        [c_1]: 2.208e-05
        [parameter_eliminate]: 3.2e-06
        [updatestate_depend_eliminate]: 6.41998e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [cse]: 1.949e-05
        [call_graph_tuple_transform]: 1.938e-05
        [tuple_list_get_item_eliminator]: 6.02001e-06
        [none_parameter_eliminate]: 1.76e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 5.52999e-06
[remove_dup_value]: 1.324e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 5.76e-06
[add_recomputation]: 8.746e-05
[cse_after_recomputation]: 2.297e-05, [1]
    [Cycle 1]: 1.699e-05, [1]
        [cse]: 1.082e-05
[auto_monad_reorder]: 1.613e-05
[get_jit_bprop_graph]: 2.28002e-06
[rewriter_after_jit_bprop_graph]: 5.92999e-06
[opt_after_jit_grad]: 0.0207878
[symbol_engine_optimizer]: 0.00011063, [1]
    [Cycle 1]: 9.691e-05, [6]
        [build]: 8.70001e-06
        [elim_shapecalc]: 1.006e-05
        [elim_not_effective]: 2.625e-05
        [opt_reshape]: 7.70998e-06
        [fold_const_symbol]: 1.063e-05
        [renormalize]: 1.10001e-06
[validate]: 5.903e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.00692846
[execute]: 1.101e-05
Sums
    bootstrap                            :     0.000354s :  0.90%
    type_inference                       :     0.004922s : 12.49%
    event_method                         :     0.000018s :  0.05%
    auto_monad                           :     0.000068s :  0.17%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000036s :  0.09%
    rewriter_before_opt_a                :     0.000074s :  0.19%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000212s :  0.54%
    jit_opt_a.loop_unroll                :     0.000151s :  0.38%
    jit_opt_a.a_1                        :     0.002110s :  5.36%
    jit_opt_a.with_stream_mark           :     0.000086s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000050s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000025s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000020s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000015s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000008s :  0.02%
    jit_opt_a.specialize_transform       :     0.000050s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000043s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000049s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000013s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000042s :  0.11%
    jit_opt_a.merge_forward              :     0.000023s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000009s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000102s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000023s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.01%
    jit_opt_a.replace_old_param          :     0.000020s :  0.05%
    jit_opt_a.inline_without_move        :     0.000018s :  0.05%
    jit_opt_a.renormalize                :     0.001105s :  2.80%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000049s :  0.13%
    jit_opt_a.cse                        :     0.000092s :  0.23%
    jit_opt_a.replace_applicator         :     0.000044s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000036s :  0.09%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000865s :  2.19%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.04%
    cconv                                :     0.000030s :  0.08%
    loop_unroll                          :     0.000454s :  1.15%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.01%
    remove_dup_value                     :     0.000013s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.01%
    add_recomputation                    :     0.000087s :  0.22%
    cse_after_recomputation.cse          :     0.000011s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.020788s : 52.76%
    symbol_engine_optimizer.build        :     0.000009s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000026s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000059s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006928s : 17.59%
    execute                              :     0.000011s :  0.03%


Time group info:
------[substitution.]   0.000571    23
 0.52% :     0.000003s :      2: substitution.elim_not_effective
 0.31% :     0.000002s :      2: substitution.fold_const_symbol
 0.93% :     0.000005s :      3: substitution.graph_param_transform
86.71% :     0.000495s :      4: substitution.inline
 0.73% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.75% :     0.000016s :      4: substitution.remove_not_recompute_node
 1.06% :     0.000006s :      2: substitution.replace_old_param
 7.01% :     0.000040s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004848     2
81.37% :     0.003945s :      1: type_inference.infer
18.63% :     0.000903s :      1: type_inference.specialize
------[replace.]   0.000249     6
65.96% :     0.000164s :      4: replace.inline
34.04% :     0.000085s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000522     6
93.44% :     0.000488s :      4: match.inline
 6.56% :     0.000034s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000395   691
 2.49% :     0.000010s :     10: predicate.accumulaten_eliminater
 1.58% :     0.000006s :      3: predicate.ad_related_special_op_eliminate
 2.44% :     0.000010s :     10: predicate.addn_check_dump
 0.59% :     0.000002s :     10: predicate.addn_zero_filter
 1.92% :     0.000008s :     10: predicate.arithmetic_simplify
 0.58% :     0.000002s :     10: predicate.cast_eliminate
 0.14% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.54% :     0.000006s :     10: predicate.compare_switch_simplify
 1.81% :     0.000007s :     10: predicate.depend_value_elim
 0.50% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 2.67% :     0.000011s :     10: predicate.dict_get_item_eliminator
 0.45% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000004s :      3: predicate.dumpgradient_eliminate
 0.15% :     0.000001s :      3: predicate.elim_not_effective
 0.22% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 2.16% :     0.000009s :     10: predicate.environ_add_const_eliminate
 1.42% :     0.000006s :     10: predicate.environ_get_add_eliminate
 0.40% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.72% :     0.000007s :     10: predicate.environ_get_eliminate
 1.38% :     0.000005s :     10: predicate.environ_get_set_eliminate
 0.07% :     0.000000s :      3: predicate.fold_const_symbol
 1.44% :     0.000006s :      6: predicate.get_grad_eliminate
 0.10% :     0.000000s :      3: predicate.graph_param_transform
 4.64% :     0.000018s :     22: predicate.inline
 0.32% :     0.000001s :      6: predicate.inline_without_move
 0.16% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.56% :     0.000006s :      6: predicate.less_batch_normalization
 1.81% :     0.000007s :     12: predicate.list_to_tuple_eliminator_
 1.87% :     0.000007s :     15: predicate.load_eliminater
 0.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.87% :     0.000019s :     34: predicate.loop_unroll_before_grad
 2.69% :     0.000011s :     13: predicate.make_slice_get_slice_eliminator
 1.60% :     0.000006s :     10: predicate.merge_addn
 0.43% :     0.000002s :     10: predicate.minmaximum_grad
 0.67% :     0.000003s :      3: predicate.mutable_eliminate
 0.21% :     0.000001s :      3: predicate.opt_reshape
 0.89% :     0.000004s :     15: predicate.partial_eliminate
 0.49% :     0.000002s :     10: predicate.print_const_string_wrapper
 0.90% :     0.000004s :     10: predicate.reduce_eliminate
 1.70% :     0.000007s :     12: predicate.redundant_stop_gradient_eliminater
 0.21% :     0.000001s :      6: predicate.remove_not_recompute_node
 0.69% :     0.000003s :     18: predicate.replace_applicator
 0.26% :     0.000001s :      6: predicate.replace_old_param
 0.12% :     0.000000s :      3: predicate.reset_defer_inline
 0.59% :     0.000002s :     10: predicate.reshape_eliminate
 0.53% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.27% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.61% :     0.000006s :     10: predicate.same_eliminate
 1.17% :     0.000005s :      6: predicate.set_cell_output_no_recompute
 0.32% :     0.000001s :      6: predicate.special_op_eliminate
 1.78% :     0.000007s :      6: predicate.specialize_transform
 1.63% :     0.000006s :     10: predicate.split_environ_get_set_with_tuple_value
 3.58% :     0.000014s :     10: predicate.stack_unstack_eliminate
 0.16% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.73% :     0.000015s :     16: predicate.switch_defer_inline
 0.94% :     0.000004s :     16: predicate.switch_layer_defer_inline
 9.36% :     0.000037s :     53: predicate.switch_simplify
 3.88% :     0.000015s :     10: predicate.tile_eliminate
 1.52% :     0.000006s :     10: predicate.transpose_eliminate
 0.51% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.53% :     0.000006s :     10: predicate.tuple_list_get_item_depend_reorder
 3.78% :     0.000015s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000006s :     10: predicate.tuple_list_set_item_eliminator
 1.80% :     0.000007s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000007s :     15: predicate.updatestate_pure_node_eliminater
 2.56% :     0.000010s :     21: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000007s :     10: predicate.value_based_eliminate
 0.11% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.26% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000754    12
43.53% :     0.000328s :      5: func_graph_cloner_run.FuncGraphClonerGraph
15.31% :     0.000115s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.16% :     0.000310s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.048286    76
 0.19% :     0.000091s :      1: add_recomputation
 0.15% :     0.000071s :      1: auto_monad
 0.04% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.77% :     0.000373s :      1: bootstrap
 0.07% :     0.000032s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.05% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.05% :     0.000023s :      1: event_method
 0.03% :     0.000016s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
18.24% :     0.008809s :      1: jit_opt_a
 0.31% :     0.000150s :      1: jit_opt_after_cconv
 0.12% :     0.000057s :      1: jit_opt_b
 0.96% :     0.000463s :      1: loop_unroll
 1.81% :     0.000876s :      1: mutable_eliminate
 5.86% :     0.002830s :     26: opt.transform.jit_opt_a
 0.10% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000027s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000054s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000050s :      4: opt.transform.symbol_engine_opt
43.11% :     0.020815s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.08% :     0.000038s :      1: py_interpret_to_execute
 0.03% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000016s :      1: remove_dup_value
 1.01% :     0.000490s :      1: renormalize.infer
 1.23% :     0.000595s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000040s :      1: rewriter_after_opt_a
 0.16% :     0.000077s :      1: rewriter_before_opt_a
 0.24% :     0.000114s :      1: symbol_engine_optimizer
14.39% :     0.006950s :      1: task_emit
10.22% :     0.004937s :      1: type_inference
 0.19% :     0.000091s :      1: validate


TotalTime = 0.034033, [33]
[bootstrap]: 0.00036733
[type_inference]: 0.00491997
[event_method]: 1.844e-05
[auto_monad]: 6.674e-05
[graph_reusing]: 6.12999e-06
[pre_auto_parallel]: 2.43e-06
[py_interpret_to_execute]: 4.161e-05
[rewriter_before_opt_a]: 7.707e-05
[expand_dump_flag]: 3.56999e-06
[jit_opt_a]: 0.0191294, [2]
    [Cycle 1]: 0.00186476, [27]
        [switch_simplify]: 6.358e-05
        [loop_unroll]: 2.912e-05
        [a_1]: 0.00056835
        [with_stream_mark]: 2.144e-05
        [recompute_prepare]: 7.48999e-06
        [updatestate_depend_eliminate]: 4.17e-06
        [updatestate_assign_eliminate]: 2.81999e-06
        [updatestate_loads_eliminate]: 2.94999e-06
        [parameter_eliminate]: 2.05002e-06
        [specialize_transform]: 6.07999e-06
        [updatestate_useless_node_eliminater]: 5.08002e-06
        [accelerated_algorithm]: 5.64998e-06
        [meta_shard_fg_expand]: 2.04e-06
        [get_grad_eliminate_]: 5.17999e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.49e-05
        [j_node_and_user_rematch]: 9.53002e-06
        [meta_fg_expand]: 2.09999e-06
        [replace_old_param]: 9.51998e-06
        [inline_without_move]: 5.20999e-06
        [renormalize]: 0.00084453
        [add_forward_monad_depend]: 6.09999e-06
        [auto_monad_grad]: 2.32999e-06
        [auto_monad_eliminator]: 1.63e-05
        [cse]: 3.167e-05
        [replace_applicator]: 1.506e-05
    [Cycle 2]: 0.0132361, [27]
        [switch_simplify]: 5.54998e-06
        [loop_unroll]: 4.93001e-06
        [a_1]: 9.185e-05
        [with_stream_mark]: 1.152e-05
        [recompute_prepare]: 5.00001e-06
        [updatestate_depend_eliminate]: 2.84999e-06
        [updatestate_assign_eliminate]: 2.64001e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 4.70001e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 0.0127413
        [meta_shard_fg_expand]: 4.51002e-06
        [get_grad_eliminate_]: 9.48002e-06
        [merge_forward]: 1.09e-05
        [cell_reuse_recompute_pass]: 4.58999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.496e-05
        [j_node_and_user_rematch]: 1.021e-05
        [meta_fg_expand]: 3.61999e-06
        [replace_old_param]: 1.339e-05
        [inline_without_move]: 5.25999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 7.73001e-06
        [auto_monad_grad]: 2.76999e-06
        [auto_monad_eliminator]: 1.707e-05
        [cse]: 3.431e-05
        [replace_applicator]: 6.48e-06
[py_interpret_to_execute_after_opt_a]: 1.84e-05
[rewriter_after_opt_a]: 4.452e-05
[convert_after_rewriter]: 6.61e-06
[order_py_execute_after_rewriter]: 4.63999e-06
[mutable_eliminate]: 0.00081842
[jit_opt_b]: 7.395e-05, [1]
    [Cycle 1]: 6.498e-05, [2]
        [frontend_op_eliminate]: 3.568e-05
        [inline_after_opt_a]: 1.708e-05
[cconv]: 3.133e-05
[loop_unroll]: 0.00045692
[jit_opt_after_cconv]: 0.00015608, [1]
    [Cycle 1]: 0.0001483, [11]
        [c_1]: 2.358e-05
        [parameter_eliminate]: 3.41999e-06
        [updatestate_depend_eliminate]: 6.79001e-06
        [updatestate_assign_eliminate]: 2.19999e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [cse]: 2.253e-05
        [call_graph_tuple_transform]: 2.31e-05
        [tuple_list_get_item_eliminator]: 5.70001e-06
        [none_parameter_eliminate]: 1.72001e-06
        [renormalize]: 6.80011e-07
        [switch_simplify]: 5.76e-06
[remove_dup_value]: 1.615e-05
[partial_unused_args_eliminate]: 3.5e-06
[environ_conv]: 5.69999e-06
[add_recomputation]: 5.45e-05
[cse_after_recomputation]: 2.382e-05, [1]
    [Cycle 1]: 1.743e-05, [1]
        [cse]: 1.112e-05
[auto_monad_reorder]: 1.79e-05
[get_jit_bprop_graph]: 2.36998e-06
[rewriter_after_jit_bprop_graph]: 7.51001e-06
[opt_after_jit_grad]: 0.00054829
[symbol_engine_optimizer]: 7.907e-05, [1]
    [Cycle 1]: 7.117e-05, [6]
        [build]: 4.50999e-06
        [elim_shapecalc]: 8.44002e-06
        [elim_not_effective]: 1.407e-05
        [opt_reshape]: 5.97999e-06
        [fold_const_symbol]: 9.82999e-06
        [renormalize]: 8.70001e-07
[validate]: 4.337e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00674332
[execute]: 7.68001e-06
Sums
    bootstrap                            :     0.000367s :  1.26%
    type_inference                       :     0.004920s : 16.81%
    event_method                         :     0.000018s :  0.06%
    auto_monad                           :     0.000067s :  0.23%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.14%
    rewriter_before_opt_a                :     0.000077s :  0.26%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000069s :  0.24%
    jit_opt_a.loop_unroll                :     0.000034s :  0.12%
    jit_opt_a.a_1                        :     0.000660s :  2.26%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.012747s : 43.56%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.05%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000023s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.04%
    jit_opt_a.renormalize                :     0.000845s :  2.89%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000033s :  0.11%
    jit_opt_a.cse                        :     0.000066s :  0.23%
    jit_opt_a.replace_applicator         :     0.000022s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.06%
    rewriter_after_opt_a                 :     0.000045s :  0.15%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000818s :  2.80%
    jit_opt_b.frontend_op_eliminate      :     0.000036s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.06%
    cconv                                :     0.000031s :  0.11%
    loop_unroll                          :     0.000457s :  1.56%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000055s :  0.19%
    cse_after_recomputation.cse          :     0.000011s :  0.04%
    auto_monad_reorder                   :     0.000018s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.03%
    opt_after_jit_grad                   :     0.000548s :  1.87%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006743s : 23.04%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000238    23
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.77% :     0.000002s :      2: substitution.fold_const_symbol
 2.94% :     0.000007s :      3: substitution.graph_param_transform
81.17% :     0.000193s :      4: substitution.inline
 1.80% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.76% :     0.000007s :      4: substitution.remove_not_recompute_node
 3.91% :     0.000009s :      2: substitution.replace_old_param
 5.89% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004849     2
81.81% :     0.003967s :      1: type_inference.infer
18.19% :     0.000882s :      1: type_inference.specialize
------[replace.]   0.000077     6
67.40% :     0.000052s :      4: replace.inline
32.60% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000203     6
93.64% :     0.000190s :      4: match.inline
 6.36% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.23% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.23% :     0.000002s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.11% :     0.000001s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.16% :     0.000001s :     10: predicate.depend_value_elim
 0.95% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.15% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.85% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.30% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.03% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.40% :     0.000002s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.23% :     0.000007s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 5.76% :     0.000007s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.24% :     0.000003s :     15: predicate.load_eliminater
 1.48% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.59% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.38% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 1.54% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.03% :     0.000003s :     15: predicate.partial_eliminate
 0.99% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.30% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 1.14% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.24% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.14% :     0.000003s :     16: predicate.switch_defer_inline
 1.85% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.86% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000002s :     10: predicate.tile_eliminate
 1.02% :     0.000001s :     10: predicate.transpose_eliminate
 1.18% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.07% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.66% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.61% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.41% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000655    12
47.90% :     0.000314s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.08% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.01% :     0.000308s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.035945    76
 0.16% :     0.000058s :      1: add_recomputation
 0.20% :     0.000070s :      1: auto_monad
 0.06% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.08% :     0.000389s :      1: bootstrap
 0.09% :     0.000034s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.07% :     0.000026s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.07% :     0.000024s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
53.23% :     0.019134s :      1: jit_opt_a
 0.44% :     0.000159s :      1: jit_opt_after_cconv
 0.21% :     0.000077s :      1: jit_opt_b
 1.30% :     0.000466s :      1: loop_unroll
 2.30% :     0.000827s :      1: mutable_eliminate
 2.61% :     0.000938s :     26: opt.transform.jit_opt_a
 0.15% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000045s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.56% :     0.000560s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000044s :      1: py_interpret_to_execute
 0.06% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 1.20% :     0.000431s :      1: renormalize.infer
 1.12% :     0.000404s :      1: renormalize.specialize
 0.03% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000048s :      1: rewriter_after_opt_a
 0.22% :     0.000080s :      1: rewriter_before_opt_a
 0.23% :     0.000082s :      1: symbol_engine_optimizer
18.81% :     0.006762s :      1: task_emit
13.74% :     0.004939s :      1: type_inference
 0.18% :     0.000064s :      1: validate


TotalTime = 0.0469689, [33]
[bootstrap]: 0.00033973
[type_inference]: 0.00562792
[event_method]: 1.714e-05
[auto_monad]: 6.869e-05
[graph_reusing]: 6.25002e-06
[pre_auto_parallel]: 2.31998e-06
[py_interpret_to_execute]: 4.205e-05
[rewriter_before_opt_a]: 7.924e-05
[expand_dump_flag]: 3.78001e-06
[jit_opt_a]: 0.0312283, [2]
    [Cycle 1]: 0.0265237, [27]
        [switch_simplify]: 6.025e-05
        [loop_unroll]: 3.015e-05
        [a_1]: 0.0247656
        [with_stream_mark]: 4.214e-05
        [recompute_prepare]: 1.464e-05
        [updatestate_depend_eliminate]: 5.56e-06
        [updatestate_assign_eliminate]: 3.16001e-06
        [updatestate_loads_eliminate]: 2.96001e-06
        [parameter_eliminate]: 3.58e-06
        [specialize_transform]: 7.21001e-06
        [updatestate_useless_node_eliminater]: 6.45997e-06
        [accelerated_algorithm]: 7.05e-06
        [meta_shard_fg_expand]: 4.04002e-06
        [get_grad_eliminate_]: 6.76e-06
        [merge_forward]: 4.54002e-06
        [cell_reuse_recompute_pass]: 4.82e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.918e-05
        [j_node_and_user_rematch]: 1.188e-05
        [meta_fg_expand]: 3.06001e-06
        [replace_old_param]: 1.327e-05
        [inline_without_move]: 6.29001e-06
        [renormalize]: 0.0011689
        [add_forward_monad_depend]: 9.45001e-06
        [auto_monad_grad]: 3.08e-06
        [auto_monad_eliminator]: 2.193e-05
        [cse]: 3.556e-05
        [replace_applicator]: 1.598e-05
    [Cycle 2]: 0.00039059, [27]
        [switch_simplify]: 7.39002e-06
        [loop_unroll]: 5.37999e-06
        [a_1]: 0.00011142
        [with_stream_mark]: 1.508e-05
        [recompute_prepare]: 6.29001e-06
        [updatestate_depend_eliminate]: 3.42002e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.99999e-06
        [updatestate_useless_node_eliminater]: 5.97999e-06
        [accelerated_algorithm]: 6.79999e-06
        [meta_shard_fg_expand]: 1.77001e-06
        [get_grad_eliminate_]: 5.77999e-06
        [merge_forward]: 2.54001e-06
        [cell_reuse_recompute_pass]: 3.25998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.703e-05
        [j_node_and_user_rematch]: 9.76998e-06
        [meta_fg_expand]: 2.48002e-06
        [replace_old_param]: 9.72001e-06
        [inline_without_move]: 5.12e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.56002e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 6.63998e-06
        [cse]: 1.22e-05
        [replace_applicator]: 6.44001e-06
[py_interpret_to_execute_after_opt_a]: 1.68e-05
[rewriter_after_opt_a]: 4.461e-05
[convert_after_rewriter]: 7.88001e-06
[order_py_execute_after_rewriter]: 5.64998e-06
[mutable_eliminate]: 0.00095292
[jit_opt_b]: 5.703e-05, [1]
    [Cycle 1]: 4.752e-05, [2]
        [frontend_op_eliminate]: 1.956e-05
        [inline_after_opt_a]: 1.595e-05
[cconv]: 3.23e-05
[loop_unroll]: 0.00050568
[jit_opt_after_cconv]: 0.00015305, [1]
    [Cycle 1]: 0.00014615, [11]
        [c_1]: 2.214e-05
        [parameter_eliminate]: 3.33e-06
        [updatestate_depend_eliminate]: 8.19002e-06
        [updatestate_assign_eliminate]: 2.56998e-06
        [updatestate_loads_eliminate]: 2.08002e-06
        [cse]: 2.159e-05
        [call_graph_tuple_transform]: 2.12e-05
        [tuple_list_get_item_eliminator]: 5.62001e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 6.60017e-07
        [switch_simplify]: 5.86e-06
[remove_dup_value]: 1.439e-05
[partial_unused_args_eliminate]: 2.29001e-06
[environ_conv]: 5.39e-06
[add_recomputation]: 5.79e-05
[cse_after_recomputation]: 2.314e-05, [1]
    [Cycle 1]: 1.688e-05, [1]
        [cse]: 9.93002e-06
[auto_monad_reorder]: 1.679e-05
[get_jit_bprop_graph]: 2.36e-06
[rewriter_after_jit_bprop_graph]: 5.57999e-06
[opt_after_jit_grad]: 0.00047632
[symbol_engine_optimizer]: 7.785e-05, [1]
    [Cycle 1]: 7.023e-05, [6]
        [build]: 4.58999e-06
        [elim_shapecalc]: 7.71999e-06
        [elim_not_effective]: 1.359e-05
        [opt_reshape]: 5.84e-06
        [fold_const_symbol]: 8.95999e-06
        [renormalize]: 7.00005e-07
[validate]: 4.861e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00681114
[execute]: 9.98002e-06
Sums
    bootstrap                            :     0.000340s :  0.81%
    type_inference                       :     0.005628s : 13.42%
    event_method                         :     0.000017s :  0.04%
    auto_monad                           :     0.000069s :  0.16%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.10%
    rewriter_before_opt_a                :     0.000079s :  0.19%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000068s :  0.16%
    jit_opt_a.loop_unroll                :     0.000036s :  0.08%
    jit_opt_a.a_1                        :     0.024877s : 59.34%
    jit_opt_a.with_stream_mark           :     0.000057s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000013s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000014s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000013s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000008s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000022s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.01%
    jit_opt_a.replace_old_param          :     0.000023s :  0.05%
    jit_opt_a.inline_without_move        :     0.000011s :  0.03%
    jit_opt_a.renormalize                :     0.001169s :  2.79%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000029s :  0.07%
    jit_opt_a.cse                        :     0.000048s :  0.11%
    jit_opt_a.replace_applicator         :     0.000022s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.04%
    rewriter_after_opt_a                 :     0.000045s :  0.11%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.01%
    mutable_eliminate                    :     0.000953s :  2.27%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.04%
    cconv                                :     0.000032s :  0.08%
    loop_unroll                          :     0.000506s :  1.21%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.05%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000022s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.01%
    remove_dup_value                     :     0.000014s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.01%
    add_recomputation                    :     0.000058s :  0.14%
    cse_after_recomputation.cse          :     0.000010s :  0.02%
    auto_monad_reorder                   :     0.000017s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000476s :  1.14%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000049s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006811s : 16.25%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000297    23
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 2.18% :     0.000006s :      3: substitution.graph_param_transform
80.18% :     0.000238s :      4: substitution.inline
 2.05% :     0.000006s :      4: substitution.j_node_and_user_rematch
 2.03% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.54% :     0.000008s :      2: substitution.replace_old_param
 9.85% :     0.000029s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005552     2
81.35% :     0.004516s :      1: type_inference.infer
18.65% :     0.001036s :      1: type_inference.specialize
------[replace.]   0.024064     6
 0.22% :     0.000054s :      4: replace.inline
99.78% :     0.024011s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000262     6
89.90% :     0.000236s :      4: match.inline
10.10% :     0.000026s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000172   691
 1.72% :     0.000003s :     10: predicate.accumulaten_eliminater
 1.12% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.87% :     0.000002s :     10: predicate.addn_check_dump
 1.78% :     0.000003s :     10: predicate.addn_zero_filter
 3.28% :     0.000006s :     10: predicate.arithmetic_simplify
 1.39% :     0.000002s :     10: predicate.cast_eliminate
 0.31% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.85% :     0.000001s :     10: predicate.compare_switch_simplify
 1.68% :     0.000003s :     10: predicate.depend_value_elim
 1.79% :     0.000003s :     10: predicate.dict_get_item_const_eliminator
 1.78% :     0.000003s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.66% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.22% :     0.000000s :      3: predicate.elim_not_effective
 0.42% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.39% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.91% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.82% :     0.000003s :     10: predicate.environ_get_eliminate
 0.93% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.17% :     0.000000s :      3: predicate.fold_const_symbol
 1.26% :     0.000002s :      6: predicate.get_grad_eliminate
 0.19% :     0.000000s :      3: predicate.graph_param_transform
 5.82% :     0.000010s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.27% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.65% :     0.000003s :      6: predicate.less_batch_normalization
 1.90% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.89% :     0.000003s :     15: predicate.load_eliminater
 1.17% :     0.000002s :      3: predicate.loop_unroll_after_grad
 2.98% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.74% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000002s :     10: predicate.merge_addn
 0.93% :     0.000002s :     10: predicate.minmaximum_grad
 1.65% :     0.000003s :      3: predicate.mutable_eliminate
 0.37% :     0.000001s :      3: predicate.opt_reshape
 1.93% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.99% :     0.000003s :     10: predicate.reduce_eliminate
 1.73% :     0.000003s :     12: predicate.redundant_stop_gradient_eliminater
 0.34% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.43% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.56% :     0.000003s :     10: predicate.reshape_eliminate
 1.29% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.41% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 0.79% :     0.000001s :      6: predicate.specialize_transform
 1.75% :     0.000003s :     10: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.30% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.39% :     0.000006s :     16: predicate.switch_defer_inline
 2.06% :     0.000004s :     16: predicate.switch_layer_defer_inline
 5.55% :     0.000010s :     53: predicate.switch_simplify
 1.21% :     0.000002s :     10: predicate.tile_eliminate
 1.14% :     0.000002s :     10: predicate.transpose_eliminate
 2.16% :     0.000004s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.03% :     0.000007s :     18: predicate.tuple_list_get_item_eliminator
 2.02% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.20% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.69% :     0.000005s :     21: predicate.updatestate_useless_node_eliminater
 2.05% :     0.000004s :     10: predicate.value_based_eliminate
 0.23% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000813    12
45.17% :     0.000367s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.80% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.03% :     0.000390s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.073367    76
 0.08% :     0.000061s :      1: add_recomputation
 0.10% :     0.000072s :      1: auto_monad
 0.03% :     0.000020s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.48% :     0.000349s :      1: bootstrap
 0.05% :     0.000035s :      1: cconv
 0.01% :     0.000011s :      1: convert_after_rewriter
 0.03% :     0.000025s :      1: cse_after_recomputation
 0.01% :     0.000008s :      1: environ_conv
 0.03% :     0.000021s :      1: event_method
 0.02% :     0.000015s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000009s :      1: graph_reusing
42.57% :     0.031233s :      1: jit_opt_a
 0.21% :     0.000156s :      1: jit_opt_after_cconv
 0.08% :     0.000060s :      1: jit_opt_b
 0.70% :     0.000514s :      1: loop_unroll
 1.31% :     0.000963s :      1: mutable_eliminate
34.25% :     0.025130s :     26: opt.transform.jit_opt_a
 0.07% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000028s :      4: opt.transform.jit_opt_b
 0.02% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.04% :     0.000033s :      4: opt.transform.symbol_engine_opt
 0.66% :     0.000485s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.06% :     0.000046s :      1: py_interpret_to_execute
 0.03% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000017s :      1: remove_dup_value
 0.86% :     0.000628s :      1: renormalize.infer
 0.72% :     0.000529s :      1: renormalize.specialize
 0.01% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000048s :      1: rewriter_after_opt_a
 0.11% :     0.000082s :      1: rewriter_before_opt_a
 0.11% :     0.000081s :      1: symbol_engine_optimizer
 9.31% :     0.006832s :      1: task_emit
 7.69% :     0.005643s :      1: type_inference
 0.10% :     0.000074s :      1: validate


TotalTime = 0.0218324, [30]
[bootstrap]: 0.00030291
[type_inference]: 0.0075163
[event_method]: 2.313e-05
[auto_monad]: 8.364e-05
[graph_reusing]: 6.00002e-06
[pre_auto_parallel]: 2.53e-06
[py_interpret_to_execute]: 5.491e-05
[rewriter_before_opt_a]: 8.802e-05
[expand_dump_flag]: 3.12002e-06
[jit_opt_a]: 0.0111216, [2]
    [Cycle 1]: 0.00699194, [27]
        [switch_simplify]: 0.00014698
        [loop_unroll]: 3.532e-05
        [a_1]: 0.00394722
        [with_stream_mark]: 3.381e-05
        [recompute_prepare]: 1.224e-05
        [updatestate_depend_eliminate]: 4.39998e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.92002e-06
        [parameter_eliminate]: 2.37999e-06
        [specialize_transform]: 6.59001e-06
        [updatestate_useless_node_eliminater]: 5.17999e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 2.61999e-06
        [get_grad_eliminate_]: 5.05999e-06
        [merge_forward]: 4.00998e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.533e-05
        [j_node_and_user_rematch]: 1.017e-05
        [meta_fg_expand]: 2.40002e-06
        [replace_old_param]: 1.071e-05
        [inline_without_move]: 6.07001e-06
        [renormalize]: 0.00236802
        [add_forward_monad_depend]: 1.091e-05
        [auto_monad_grad]: 3.23e-06
        [auto_monad_eliminator]: 2.208e-05
        [cse]: 3.604e-05
        [replace_applicator]: 2.17e-05
    [Cycle 2]: 0.00035792, [27]
        [switch_simplify]: 6.61999e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 9.603e-05
        [with_stream_mark]: 1.849e-05
        [recompute_prepare]: 5.41002e-06
        [updatestate_depend_eliminate]: 3.71001e-06
        [updatestate_assign_eliminate]: 2.73998e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [parameter_eliminate]: 1.61002e-06
        [specialize_transform]: 5.63002e-06
        [updatestate_useless_node_eliminater]: 4.60001e-06
        [accelerated_algorithm]: 5.40999e-06
        [meta_shard_fg_expand]: 2.02001e-06
        [get_grad_eliminate_]: 4.76002e-06
        [merge_forward]: 4.22998e-06
        [cell_reuse_recompute_pass]: 2.41e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.807e-05
        [j_node_and_user_rematch]: 8.73001e-06
        [meta_fg_expand]: 2.58003e-06
        [replace_old_param]: 9.42999e-06
        [inline_without_move]: 4.78001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 1.16002e-06
        [auto_monad_eliminator]: 5.76e-06
        [cse]: 1.278e-05
        [replace_applicator]: 5.09998e-06
[py_interpret_to_execute_after_opt_a]: 2.069e-05
[rewriter_after_opt_a]: 3.901e-05
[convert_after_rewriter]: 6.16e-06
[order_py_execute_after_rewriter]: 5.17999e-06
[mutable_eliminate]: 0.0008339
[jit_opt_b]: 5.789e-05, [1]
    [Cycle 1]: 4.679e-05, [2]
        [frontend_op_eliminate]: 1.731e-05
        [inline_after_opt_a]: 1.623e-05
[cconv]: 3.442e-05
[loop_unroll]: 0.00047941
[jit_opt_after_cconv]: 0.00016735, [1]
    [Cycle 1]: 0.00016057, [11]
        [c_1]: 2.269e-05
        [parameter_eliminate]: 5.39998e-06
        [updatestate_depend_eliminate]: 9.27001e-06
        [updatestate_assign_eliminate]: 2.84001e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [cse]: 3.043e-05
        [call_graph_tuple_transform]: 2.218e-05
        [tuple_list_get_item_eliminator]: 5.75001e-06
        [none_parameter_eliminate]: 1.40999e-06
        [renormalize]: 8.50006e-07
        [switch_simplify]: 5.71e-06
[remove_dup_value]: 1.712e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 6.00002e-06
[add_recomputation]: 5.561e-05
[cse_after_recomputation]: 2.28e-05, [1]
    [Cycle 1]: 1.692e-05, [1]
        [cse]: 1.05e-05
[auto_monad_reorder]: 1.669e-05
[get_jit_bprop_graph]: 2.56e-06
[rewriter_after_jit_bprop_graph]: 8.32e-06
[opt_after_jit_grad]: 0.0005229
[symbol_engine_optimizer]: 8.029e-05, [1]
    [Cycle 1]: 7.26e-05, [6]
        [build]: 5.56e-06
        [elim_shapecalc]: 7.29001e-06
        [elim_not_effective]: 1.363e-05
        [opt_reshape]: 6.12999e-06
        [fold_const_symbol]: 9.50001e-06
        [renormalize]: 8.49977e-07
[validate]: 4.464e-05
Sums
    bootstrap                            :     0.000303s :  1.75%
    type_inference                       :     0.007516s : 43.35%
    event_method                         :     0.000023s :  0.13%
    auto_monad                           :     0.000084s :  0.48%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000055s :  0.32%
    rewriter_before_opt_a                :     0.000088s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000154s :  0.89%
    jit_opt_a.loop_unroll                :     0.000040s :  0.23%
    jit_opt_a.a_1                        :     0.004043s : 23.32%
    jit_opt_a.with_stream_mark           :     0.000052s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000020s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.002368s : 13.66%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.16%
    jit_opt_a.cse                        :     0.000049s :  0.28%
    jit_opt_a.replace_applicator         :     0.000027s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000021s :  0.12%
    rewriter_after_opt_a                 :     0.000039s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000834s :  4.81%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000034s :  0.20%
    loop_unroll                          :     0.000479s :  2.77%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000030s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000056s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000523s :  3.02%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.26%


Time group info:
------[substitution.]   0.003424    26
 0.07% :     0.000002s :      2: substitution.elim_not_effective
 0.04% :     0.000001s :      2: substitution.fold_const_symbol
 0.17% :     0.000006s :      3: substitution.graph_param_transform
98.42% :     0.003370s :      6: substitution.inline
 0.12% :     0.000004s :      4: substitution.j_node_and_user_rematch
 0.18% :     0.000006s :      4: substitution.remove_not_recompute_node
 0.22% :     0.000008s :      2: substitution.replace_old_param
 0.27% :     0.000009s :      1: substitution.switch_simplify
 0.51% :     0.000017s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007433     2
84.49% :     0.006281s :      1: type_inference.infer
15.51% :     0.001153s :      1: type_inference.specialize
------[replace.]   0.000218     9
58.03% :     0.000127s :      6: replace.inline
22.45% :     0.000049s :      1: replace.switch_simplify
19.52% :     0.000043s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.003388     9
99.29% :     0.003364s :      6: match.inline
 0.24% :     0.000008s :      1: match.switch_simplify
 0.47% :     0.000016s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000150   801
 1.39% :     0.000002s :     12: predicate.accumulaten_eliminater
 0.91% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     12: predicate.addn_check_dump
 1.21% :     0.000002s :     12: predicate.addn_zero_filter
 2.15% :     0.000003s :     12: predicate.arithmetic_simplify
 1.19% :     0.000002s :     12: predicate.cast_eliminate
 0.33% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     12: predicate.compare_switch_simplify
 1.57% :     0.000002s :     12: predicate.depend_value_elim
 1.06% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.32% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      3: predicate.elim_not_effective
 0.43% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000002s :     12: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.00% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     12: predicate.environ_get_eliminate
 1.07% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.51% :     0.000001s :      3: predicate.graph_param_transform
 6.40% :     0.000010s :     26: predicate.inline
 0.74% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.63% :     0.000002s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.84% :     0.000003s :     17: predicate.load_eliminater
 1.71% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.60% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.85% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000002s :     12: predicate.merge_addn
 1.00% :     0.000001s :     12: predicate.minmaximum_grad
 2.37% :     0.000004s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.42% :     0.000004s :     17: predicate.partial_eliminate
 1.25% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     12: predicate.reduce_eliminate
 1.71% :     0.000003s :     14: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.39% :     0.000002s :     20: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000001s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     12: predicate.reshape_eliminate
 1.25% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 1.01% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000002s :     12: predicate.same_eliminate
 0.60% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.62% :     0.000001s :      6: predicate.special_op_eliminate
 0.76% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.31% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.51% :     0.000005s :     20: predicate.switch_defer_inline
 2.02% :     0.000003s :     20: predicate.switch_layer_defer_inline
 7.95% :     0.000012s :     65: predicate.switch_simplify
 1.26% :     0.000002s :     12: predicate.tile_eliminate
 1.11% :     0.000002s :     12: predicate.transpose_eliminate
 1.26% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 4.50% :     0.000007s :     20: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000003s :     12: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.44% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.57% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000003s :     12: predicate.value_based_eliminate
 0.25% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.54% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.003992    15
11.06% :     0.000441s :      6: func_graph_cloner_run.FuncGraphClonerGraph
78.94% :     0.003151s :      1: func_graph_cloner_run.FuncGraphClonerNode
10.00% :     0.000399s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028622    72
 0.21% :     0.000059s :      1: add_recomputation
 0.31% :     0.000088s :      1: auto_monad
 0.07% :     0.000020s :      1: auto_monad_reorder
 1.13% :     0.000324s :      1: bootstrap
 0.13% :     0.000037s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.10% :     0.000028s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
38.87% :     0.011126s :      1: jit_opt_a
 0.59% :     0.000170s :      1: jit_opt_after_cconv
 0.21% :     0.000061s :      1: jit_opt_b
 1.72% :     0.000491s :      1: loop_unroll
 2.96% :     0.000847s :      1: mutable_eliminate
15.28% :     0.004372s :     26: opt.transform.jit_opt_a
 0.18% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.86% :     0.000534s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000058s :      1: py_interpret_to_execute
 0.08% :     0.000024s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 6.63% :     0.001897s :      1: renormalize.infer
 1.59% :     0.000456s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000042s :      1: rewriter_after_opt_a
 0.32% :     0.000091s :      1: rewriter_before_opt_a
 0.29% :     0.000083s :      1: symbol_engine_optimizer
26.33% :     0.007537s :      1: type_inference


TotalTime = 0.0438763, [33]
[bootstrap]: 0.0003138
[type_inference]: 0.0240199
[event_method]: 1.895e-05
[auto_monad]: 6.502e-05
[graph_reusing]: 5.85002e-06
[pre_auto_parallel]: 2.36e-06
[py_interpret_to_execute]: 4.287e-05
[rewriter_before_opt_a]: 7.445e-05
[expand_dump_flag]: 3.17002e-06
[jit_opt_a]: 0.00863329, [2]
    [Cycle 1]: 0.00284385, [27]
        [switch_simplify]: 7.096e-05
        [loop_unroll]: 3.105e-05
        [a_1]: 0.0006182
        [with_stream_mark]: 2.624e-05
        [recompute_prepare]: 7.53e-06
        [updatestate_depend_eliminate]: 4.05e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 3.16999e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 6.49001e-06
        [updatestate_useless_node_eliminater]: 5.35001e-06
        [accelerated_algorithm]: 5.76998e-06
        [meta_shard_fg_expand]: 2.19999e-06
        [get_grad_eliminate_]: 5.47001e-06
        [merge_forward]: 3.60998e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.555e-05
        [j_node_and_user_rematch]: 1.007e-05
        [meta_fg_expand]: 2.53e-06
        [replace_old_param]: 9.86998e-06
        [inline_without_move]: 5.39e-06
        [renormalize]: 0.00173656
        [add_forward_monad_depend]: 6.63e-06
        [auto_monad_grad]: 2.94001e-06
        [auto_monad_eliminator]: 1.519e-05
        [cse]: 3.104e-05
        [replace_applicator]: 1.587e-05
    [Cycle 2]: 0.00040944, [27]
        [switch_simplify]: 6.23e-06
        [loop_unroll]: 6.09999e-06
        [a_1]: 9.85e-05
        [with_stream_mark]: 1.221e-05
        [recompute_prepare]: 5.51e-06
        [updatestate_depend_eliminate]: 3.31001e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 1.09003e-06
        [specialize_transform]: 6.23e-06
        [updatestate_useless_node_eliminater]: 5.62999e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.50001e-06
        [get_grad_eliminate_]: 4.50999e-06
        [merge_forward]: 3.78999e-06
        [cell_reuse_recompute_pass]: 2.59999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.531e-05
        [j_node_and_user_rematch]: 8.73001e-06
        [meta_fg_expand]: 1.91003e-06
        [replace_old_param]: 7.87e-06
        [inline_without_move]: 4.90999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.45001e-06
        [auto_monad_grad]: 1.20001e-06
        [auto_monad_eliminator]: 5.79e-06
        [cse]: 1.446e-05
        [replace_applicator]: 5.67999e-06
[py_interpret_to_execute_after_opt_a]: 1.385e-05
[rewriter_after_opt_a]: 3.709e-05
[convert_after_rewriter]: 7.4e-06
[order_py_execute_after_rewriter]: 5.52999e-06
[mutable_eliminate]: 0.00069064
[jit_opt_b]: 5.546e-05, [1]
    [Cycle 1]: 4.704e-05, [2]
        [frontend_op_eliminate]: 1.878e-05
        [inline_after_opt_a]: 1.634e-05
[cconv]: 2.839e-05
[loop_unroll]: 0.00047303
[jit_opt_after_cconv]: 0.00016264, [1]
    [Cycle 1]: 0.00015452, [11]
        [c_1]: 2.337e-05
        [parameter_eliminate]: 3.00002e-06
        [updatestate_depend_eliminate]: 7.93001e-06
        [updatestate_assign_eliminate]: 2.74001e-06
        [updatestate_loads_eliminate]: 2.27001e-06
        [cse]: 2.501e-05
        [call_graph_tuple_transform]: 2.038e-05
        [tuple_list_get_item_eliminator]: 5.77999e-06
        [none_parameter_eliminate]: 1.67999e-06
        [renormalize]: 8.00006e-07
        [switch_simplify]: 6.39001e-06
[remove_dup_value]: 1.481e-05
[partial_unused_args_eliminate]: 2.18998e-06
[environ_conv]: 5.01002e-06
[add_recomputation]: 6.108e-05
[cse_after_recomputation]: 2.247e-05, [1]
    [Cycle 1]: 1.607e-05, [1]
        [cse]: 9.97001e-06
[auto_monad_reorder]: 1.572e-05
[get_jit_bprop_graph]: 2.32001e-06
[rewriter_after_jit_bprop_graph]: 7.56999e-06
[opt_after_jit_grad]: 0.00049562
[symbol_engine_optimizer]: 8.277e-05, [1]
    [Cycle 1]: 7.474e-05, [6]
        [build]: 4.27998e-06
        [elim_shapecalc]: 9.55001e-06
        [elim_not_effective]: 1.404e-05
        [opt_reshape]: 5.87999e-06
        [fold_const_symbol]: 1.011e-05
        [renormalize]: 4.50003e-07
[validate]: 4.304e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.00822119
[execute]: 1.043e-05
Sums
    bootstrap                            :     0.000314s :  0.83%
    type_inference                       :     0.024020s : 63.62%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000065s :  0.17%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.11%
    rewriter_before_opt_a                :     0.000074s :  0.20%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000077s :  0.20%
    jit_opt_a.loop_unroll                :     0.000037s :  0.10%
    jit_opt_a.a_1                        :     0.000717s :  1.90%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000013s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000018s :  0.05%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.001737s :  4.60%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.06%
    jit_opt_a.cse                        :     0.000045s :  0.12%
    jit_opt_a.replace_applicator         :     0.000022s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.04%
    rewriter_after_opt_a                 :     0.000037s :  0.10%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.01%
    mutable_eliminate                    :     0.000691s :  1.83%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.04%
    cconv                                :     0.000028s :  0.08%
    loop_unroll                          :     0.000473s :  1.25%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000015s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.01%
    add_recomputation                    :     0.000061s :  0.16%
    cse_after_recomputation.cse          :     0.000010s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000496s :  1.31%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000043s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008221s : 21.78%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.000270    23
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000002s :      2: substitution.fold_const_symbol
 2.18% :     0.000006s :      3: substitution.graph_param_transform
86.26% :     0.000233s :      4: substitution.inline
 1.42% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.30% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000006s :      2: substitution.replace_old_param
 4.46% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.023939     2
95.77% :     0.022926s :      1: type_inference.infer
 4.23% :     0.001013s :      1: type_inference.specialize
------[replace.]   0.000079     6
69.02% :     0.000055s :      4: replace.inline
30.98% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000242     6
95.44% :     0.000231s :      4: match.inline
 4.56% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.39% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000002s :     10: predicate.addn_zero_filter
 2.39% :     0.000003s :     10: predicate.arithmetic_simplify
 1.19% :     0.000001s :     10: predicate.cast_eliminate
 0.72% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.80% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.36% :     0.000000s :      3: predicate.graph_param_transform
 5.66% :     0.000007s :     22: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 1.48% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 1.18% :     0.000001s :     10: predicate.minmaximum_grad
 1.52% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.04% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.25% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     10: predicate.same_eliminate
 0.67% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.34% :     0.000002s :      6: predicate.special_op_eliminate
 1.28% :     0.000002s :      6: predicate.specialize_transform
 1.54% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.73% :     0.000003s :     16: predicate.switch_defer_inline
 2.20% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.19% :     0.000010s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000002s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.25% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.47% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.016607    12
97.57% :     0.016203s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 0.33% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
 2.10% :     0.000348s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.046679    76
 0.14% :     0.000065s :      1: add_recomputation
 0.15% :     0.000069s :      1: auto_monad
 0.04% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.71% :     0.000334s :      1: bootstrap
 0.07% :     0.000031s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.05% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.05% :     0.000024s :      1: event_method
 0.03% :     0.000016s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
18.50% :     0.008638s :      1: jit_opt_a
 0.36% :     0.000166s :      1: jit_opt_after_cconv
 0.13% :     0.000058s :      1: jit_opt_b
 1.03% :     0.000482s :      1: loop_unroll
 1.50% :     0.000701s :      1: mutable_eliminate
 2.05% :     0.000958s :     26: opt.transform.jit_opt_a
 0.11% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000028s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.08% :     0.000504s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.10% :     0.000046s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000017s :      1: remove_dup_value
 2.69% :     0.001255s :      1: renormalize.infer
 1.01% :     0.000472s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000040s :      1: rewriter_after_opt_a
 0.17% :     0.000077s :      1: rewriter_before_opt_a
 0.18% :     0.000085s :      1: symbol_engine_optimizer
17.66% :     0.008242s :      1: task_emit
51.50% :     0.024039s :      1: type_inference
 0.14% :     0.000067s :      1: validate


TotalTime = 0.0258906, [33]
[bootstrap]: 0.00037734
[type_inference]: 0.00804943
[event_method]: 1.92e-05
[auto_monad]: 6.628e-05
[graph_reusing]: 6.11998e-06
[pre_auto_parallel]: 2.58e-06
[py_interpret_to_execute]: 3.916e-05
[rewriter_before_opt_a]: 7.355e-05
[expand_dump_flag]: 3.09001e-06
[jit_opt_a]: 0.00662585, [2]
    [Cycle 1]: 0.00274305, [27]
        [switch_simplify]: 5.928e-05
        [loop_unroll]: 3.066e-05
        [a_1]: 0.00056483
        [with_stream_mark]: 2.117e-05
        [recompute_prepare]: 7.41999e-06
        [updatestate_depend_eliminate]: 3.48999e-06
        [updatestate_assign_eliminate]: 3.50003e-06
        [updatestate_loads_eliminate]: 2.91e-06
        [parameter_eliminate]: 2.69001e-06
        [specialize_transform]: 6.43e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 5.99e-06
        [meta_shard_fg_expand]: 2.09999e-06
        [get_grad_eliminate_]: 5.38002e-06
        [merge_forward]: 3.71001e-06
        [cell_reuse_recompute_pass]: 1.08001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.532e-05
        [j_node_and_user_rematch]: 1.011e-05
        [meta_fg_expand]: 2.43e-06
        [replace_old_param]: 9.04e-06
        [inline_without_move]: 5.59e-06
        [renormalize]: 0.00172109
        [add_forward_monad_depend]: 6.24999e-06
        [auto_monad_grad]: 2.76e-06
        [auto_monad_eliminator]: 1.659e-05
        [cse]: 3.03e-05
        [replace_applicator]: 1.429e-05
    [Cycle 2]: 0.00036023, [27]
        [switch_simplify]: 6.27001e-06
        [loop_unroll]: 5.02e-06
        [a_1]: 9.544e-05
        [with_stream_mark]: 1.109e-05
        [recompute_prepare]: 5.47001e-06
        [updatestate_depend_eliminate]: 2.74001e-06
        [updatestate_assign_eliminate]: 2.59001e-06
        [updatestate_loads_eliminate]: 2.03002e-06
        [parameter_eliminate]: 1.39998e-06
        [specialize_transform]: 4.87e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 4.72e-06
        [meta_shard_fg_expand]: 1.46998e-06
        [get_grad_eliminate_]: 5.00999e-06
        [merge_forward]: 2.71e-06
        [cell_reuse_recompute_pass]: 2.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.534e-05
        [j_node_and_user_rematch]: 7.83001e-06
        [meta_fg_expand]: 1.84998e-06
        [replace_old_param]: 7.56001e-06
        [inline_without_move]: 6.12999e-06
        [renormalize]: 4.99713e-08
        [add_forward_monad_depend]: 1.66002e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 5.40999e-06
        [cse]: 1.254e-05
        [replace_applicator]: 5.05001e-06
[py_interpret_to_execute_after_opt_a]: 1.12e-05
[rewriter_after_opt_a]: 3.714e-05
[convert_after_rewriter]: 6.86999e-06
[order_py_execute_after_rewriter]: 4.72e-06
[mutable_eliminate]: 0.00063172
[jit_opt_b]: 5.155e-05, [1]
    [Cycle 1]: 4.377e-05, [2]
        [frontend_op_eliminate]: 1.717e-05
        [inline_after_opt_a]: 1.53e-05
[cconv]: 2.782e-05
[loop_unroll]: 0.00043209
[jit_opt_after_cconv]: 0.00014566, [1]
    [Cycle 1]: 0.00013898, [11]
        [c_1]: 2.097e-05
        [parameter_eliminate]: 3.09001e-06
        [updatestate_depend_eliminate]: 6.64001e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 2.056e-05
        [call_graph_tuple_transform]: 1.971e-05
        [tuple_list_get_item_eliminator]: 5.36002e-06
        [none_parameter_eliminate]: 1.60001e-06
        [renormalize]: 5.39992e-07
        [switch_simplify]: 5.61e-06
[remove_dup_value]: 1.37e-05
[partial_unused_args_eliminate]: 2.34001e-06
[environ_conv]: 5.17999e-06
[add_recomputation]: 5.398e-05
[cse_after_recomputation]: 2.133e-05, [1]
    [Cycle 1]: 1.562e-05, [1]
        [cse]: 9.71e-06
[auto_monad_reorder]: 1.496e-05
[get_jit_bprop_graph]: 2.07001e-06
[rewriter_after_jit_bprop_graph]: 6.60997e-06
[opt_after_jit_grad]: 0.00047388
[symbol_engine_optimizer]: 8.338e-05, [1]
    [Cycle 1]: 7.625e-05, [6]
        [build]: 4.94003e-06
        [elim_shapecalc]: 8.52e-06
        [elim_not_effective]: 1.399e-05
        [opt_reshape]: 6.60002e-06
        [fold_const_symbol]: 1.027e-05
        [renormalize]: 4.89992e-07
[validate]: 4.481e-05
[backend_pass]: 1.15999e-06
[task_emit]: 0.00831054
[execute]: 9.84999e-06
Sums
    bootstrap                            :     0.000377s :  1.74%
    type_inference                       :     0.008049s : 37.12%
    event_method                         :     0.000019s :  0.09%
    auto_monad                           :     0.000066s :  0.31%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.18%
    rewriter_before_opt_a                :     0.000074s :  0.34%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000066s :  0.30%
    jit_opt_a.loop_unroll                :     0.000036s :  0.16%
    jit_opt_a.a_1                        :     0.000660s :  3.04%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.08%
    jit_opt_a.inline_without_move        :     0.000012s :  0.05%
    jit_opt_a.renormalize                :     0.001721s :  7.94%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.10%
    jit_opt_a.cse                        :     0.000043s :  0.20%
    jit_opt_a.replace_applicator         :     0.000019s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.05%
    rewriter_after_opt_a                 :     0.000037s :  0.17%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000632s :  2.91%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.07%
    cconv                                :     0.000028s :  0.13%
    loop_unroll                          :     0.000432s :  1.99%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000054s :  0.25%
    cse_after_recomputation.cse          :     0.000010s :  0.04%
    auto_monad_reorder                   :     0.000015s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.03%
    opt_after_jit_grad                   :     0.000474s :  2.19%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000045s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008311s : 38.32%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000226    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.53% :     0.000006s :      3: substitution.graph_param_transform
84.39% :     0.000191s :      4: substitution.inline
 1.63% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.41% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000005s :      2: substitution.replace_old_param
 5.53% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007971     2
87.66% :     0.006987s :      1: type_inference.infer
12.34% :     0.000983s :      1: type_inference.specialize
------[replace.]   0.000073     6
69.09% :     0.000050s :      4: replace.inline
30.91% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     6
94.30% :     0.000188s :      4: match.inline
 5.70% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.15% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 2.86% :     0.000003s :     10: predicate.arithmetic_simplify
 1.40% :     0.000002s :     10: predicate.cast_eliminate
 0.58% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.18% :     0.000001s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.15% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.73% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.06% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.32% :     0.000002s :      6: predicate.get_grad_eliminate
 0.35% :     0.000000s :      3: predicate.graph_param_transform
 5.55% :     0.000007s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.59% :     0.000002s :      6: predicate.less_batch_normalization
 1.78% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.55% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.89% :     0.000002s :      3: predicate.mutable_eliminate
 0.65% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.05% :     0.000003s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.87% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.33% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.19% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.74% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000002s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.47% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000757    12
51.47% :     0.000390s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.59% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.94% :     0.000325s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028593    76
 0.20% :     0.000058s :      1: add_recomputation
 0.24% :     0.000070s :      1: auto_monad
 0.06% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.40% :     0.000400s :      1: bootstrap
 0.11% :     0.000031s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000024s :      1: event_method
 0.05% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
23.19% :     0.006630s :      1: jit_opt_a
 0.52% :     0.000148s :      1: jit_opt_after_cconv
 0.19% :     0.000054s :      1: jit_opt_b
 1.55% :     0.000442s :      1: loop_unroll
 2.24% :     0.000642s :      1: mutable_eliminate
 3.09% :     0.000883s :     26: opt.transform.jit_opt_a
 0.17% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000026s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.69% :     0.000484s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000042s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000016s :      1: remove_dup_value
 1.62% :     0.000463s :      1: renormalize.infer
 4.36% :     0.001248s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000040s :      1: rewriter_after_opt_a
 0.27% :     0.000076s :      1: rewriter_before_opt_a
 0.30% :     0.000086s :      1: symbol_engine_optimizer
29.13% :     0.008330s :      1: task_emit
28.21% :     0.008067s :      1: type_inference
 0.24% :     0.000069s :      1: validate


TotalTime = 0.0413904, [33]
[bootstrap]: 0.0003955
[type_inference]: 0.00716288
[event_method]: 2.027e-05
[auto_monad]: 7.939e-05
[graph_reusing]: 5.94999e-06
[pre_auto_parallel]: 3.23e-06
[py_interpret_to_execute]: 5.632e-05
[rewriter_before_opt_a]: 8.133e-05
[expand_dump_flag]: 3.55e-06
[jit_opt_a]: 0.00902602, [2]
    [Cycle 1]: 0.00386704, [27]
        [switch_simplify]: 7.493e-05
        [loop_unroll]: 3.343e-05
        [a_1]: 0.00067561
        [with_stream_mark]: 2.892e-05
        [recompute_prepare]: 1.062e-05
        [updatestate_depend_eliminate]: 4.28999e-06
        [updatestate_assign_eliminate]: 3.05002e-06
        [updatestate_loads_eliminate]: 3.01999e-06
        [parameter_eliminate]: 2.49999e-06
        [specialize_transform]: 6.90998e-06
        [updatestate_useless_node_eliminater]: 5.99e-06
        [accelerated_algorithm]: 6.15002e-06
        [meta_shard_fg_expand]: 2.72001e-06
        [get_grad_eliminate_]: 6.56999e-06
        [merge_forward]: 4.42998e-06
        [cell_reuse_recompute_pass]: 1.95001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.795e-05
        [j_node_and_user_rematch]: 1.098e-05
        [meta_fg_expand]: 2.71e-06
        [replace_old_param]: 1.823e-05
        [inline_without_move]: 6.17001e-06
        [renormalize]: 0.0026273
        [add_forward_monad_depend]: 1.154e-05
        [auto_monad_grad]: 2.89001e-06
        [auto_monad_eliminator]: 2.077e-05
        [cse]: 3.146e-05
        [replace_applicator]: 2.228e-05
    [Cycle 2]: 0.0003906, [27]
        [switch_simplify]: 7.16001e-06
        [loop_unroll]: 6.32001e-06
        [a_1]: 0.00011603
        [with_stream_mark]: 1.774e-05
        [recompute_prepare]: 5.69e-06
        [updatestate_depend_eliminate]: 3.91999e-06
        [updatestate_assign_eliminate]: 3.36999e-06
        [updatestate_loads_eliminate]: 2.85998e-06
        [parameter_eliminate]: 1.48002e-06
        [specialize_transform]: 5.59998e-06
        [updatestate_useless_node_eliminater]: 5.06997e-06
        [accelerated_algorithm]: 5.62999e-06
        [meta_shard_fg_expand]: 2.56998e-06
        [get_grad_eliminate_]: 5.47001e-06
        [merge_forward]: 4.18999e-06
        [cell_reuse_recompute_pass]: 3.74002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.748e-05
        [j_node_and_user_rematch]: 9.88998e-06
        [meta_fg_expand]: 2.53998e-06
        [replace_old_param]: 9.10001e-06
        [inline_without_move]: 4.90999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.50001e-06
        [auto_monad_grad]: 9.89996e-07
        [auto_monad_eliminator]: 6.17001e-06
        [cse]: 1.071e-05
        [replace_applicator]: 5.06002e-06
[py_interpret_to_execute_after_opt_a]: 1.741e-05
[rewriter_after_opt_a]: 4.042e-05
[convert_after_rewriter]: 7.88999e-06
[order_py_execute_after_rewriter]: 5.12e-06
[mutable_eliminate]: 0.00089473
[jit_opt_b]: 6.125e-05, [1]
    [Cycle 1]: 5.104e-05, [2]
        [frontend_op_eliminate]: 1.99e-05
        [inline_after_opt_a]: 1.742e-05
[cconv]: 3.032e-05
[loop_unroll]: 0.00051645
[jit_opt_after_cconv]: 0.00016726, [1]
    [Cycle 1]: 0.00015938, [11]
        [c_1]: 2.23e-05
        [parameter_eliminate]: 3.91999e-06
        [updatestate_depend_eliminate]: 8.23001e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 2.709e-05
        [call_graph_tuple_transform]: 2.433e-05
        [tuple_list_get_item_eliminator]: 6.23998e-06
        [none_parameter_eliminate]: 1.72001e-06
        [renormalize]: 8.59989e-07
        [switch_simplify]: 6.63003e-06
[remove_dup_value]: 1.593e-05
[partial_unused_args_eliminate]: 2.29999e-06
[environ_conv]: 6.88e-06
[add_recomputation]: 5.583e-05
[cse_after_recomputation]: 2.307e-05, [1]
    [Cycle 1]: 1.674e-05, [1]
        [cse]: 9.42999e-06
[auto_monad_reorder]: 1.703e-05
[get_jit_bprop_graph]: 2.79999e-06
[rewriter_after_jit_bprop_graph]: 8.60001e-06
[opt_after_jit_grad]: 0.00056631
[symbol_engine_optimizer]: 8.598e-05, [1]
    [Cycle 1]: 7.774e-05, [6]
        [build]: 4.94e-06
        [elim_shapecalc]: 9.12001e-06
        [elim_not_effective]: 1.607e-05
        [opt_reshape]: 6.34999e-06
        [fold_const_symbol]: 9.25999e-06
        [renormalize]: 4.19997e-07
[validate]: 4.949e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.0217125
[execute]: 1.142e-05
Sums
    bootstrap                            :     0.000395s :  1.10%
    type_inference                       :     0.007163s : 19.96%
    event_method                         :     0.000020s :  0.06%
    auto_monad                           :     0.000079s :  0.22%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000056s :  0.16%
    rewriter_before_opt_a                :     0.000081s :  0.23%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000082s :  0.23%
    jit_opt_a.loop_unroll                :     0.000040s :  0.11%
    jit_opt_a.a_1                        :     0.000792s :  2.21%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000013s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.03%
    jit_opt_a.merge_forward              :     0.000009s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000027s :  0.08%
    jit_opt_a.inline_without_move        :     0.000011s :  0.03%
    jit_opt_a.renormalize                :     0.002627s :  7.32%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.08%
    jit_opt_a.cse                        :     0.000042s :  0.12%
    jit_opt_a.replace_applicator         :     0.000027s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.05%
    rewriter_after_opt_a                 :     0.000040s :  0.11%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000895s :  2.49%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.05%
    cconv                                :     0.000030s :  0.08%
    loop_unroll                          :     0.000516s :  1.44%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000016s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000056s :  0.16%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000017s :  0.05%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.02%
    opt_after_jit_grad                   :     0.000566s :  1.58%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000049s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.021713s : 60.52%
    execute                              :     0.000011s :  0.03%


Time group info:
------[substitution.]   0.000291    23
 1.07% :     0.000003s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.21% :     0.000006s :      3: substitution.graph_param_transform
84.93% :     0.000247s :      4: substitution.inline
 1.65% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.09% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.29% :     0.000007s :      2: substitution.replace_old_param
 5.24% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007063     2
75.70% :     0.005346s :      1: type_inference.infer
24.30% :     0.001716s :      1: type_inference.specialize
------[replace.]   0.000086     6
66.15% :     0.000057s :      4: replace.inline
33.85% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000259     6
94.63% :     0.000245s :      4: match.inline
 5.37% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000147   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000002s :     10: predicate.addn_check_dump
 1.98% :     0.000003s :     10: predicate.addn_zero_filter
 2.50% :     0.000004s :     10: predicate.arithmetic_simplify
 1.96% :     0.000003s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.86% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.34% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.09% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.35% :     0.000002s :     10: predicate.environ_get_eliminate
 1.44% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.13% :     0.000002s :      6: predicate.get_grad_eliminate
 0.34% :     0.000001s :      3: predicate.graph_param_transform
 5.67% :     0.000008s :     22: predicate.inline
 0.74% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.77% :     0.000003s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.57% :     0.000002s :     15: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.90% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.88% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000002s :     10: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.38% :     0.000001s :      3: predicate.opt_reshape
 1.99% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.83% :     0.000003s :     10: predicate.reduce_eliminate
 1.67% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000002s :     10: predicate.reshape_eliminate
 1.56% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.07% :     0.000002s :     10: predicate.same_eliminate
 0.62% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 0.85% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 1.91% :     0.000003s :     16: predicate.switch_defer_inline
 1.67% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.49% :     0.000011s :     53: predicate.switch_simplify
 1.49% :     0.000002s :     10: predicate.tile_eliminate
 1.07% :     0.000002s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.48% :     0.000007s :     18: predicate.tuple_list_get_item_eliminator
 2.03% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.34% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.36% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.18% :     0.000003s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000832    12
42.62% :     0.000355s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.23% :     0.000060s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.15% :     0.000417s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.045190    76
 0.13% :     0.000059s :      1: add_recomputation
 0.18% :     0.000083s :      1: auto_monad
 0.04% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.92% :     0.000415s :      1: bootstrap
 0.07% :     0.000033s :      1: cconv
 0.02% :     0.000011s :      1: convert_after_rewriter
 0.06% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.06% :     0.000025s :      1: event_method
 0.04% :     0.000018s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
19.98% :     0.009031s :      1: jit_opt_a
 0.38% :     0.000170s :      1: jit_opt_after_cconv
 0.14% :     0.000064s :      1: jit_opt_b
 1.17% :     0.000527s :      1: loop_unroll
 2.01% :     0.000907s :      1: mutable_eliminate
 2.35% :     0.001063s :     26: opt.transform.jit_opt_a
 0.12% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000030s :      4: opt.transform.jit_opt_b
 0.03% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.28% :     0.000578s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000059s :      1: py_interpret_to_execute
 0.04% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000018s :      1: remove_dup_value
 4.57% :     0.002064s :      1: renormalize.infer
 1.21% :     0.000549s :      1: renormalize.specialize
 0.02% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000044s :      1: rewriter_after_opt_a
 0.19% :     0.000084s :      1: rewriter_before_opt_a
 0.20% :     0.000089s :      1: symbol_engine_optimizer
48.10% :     0.021735s :      1: task_emit
15.90% :     0.007186s :      1: type_inference
 0.16% :     0.000073s :      1: validate


TotalTime = 0.0408692, [33]
[bootstrap]: 0.00039422
[type_inference]: 0.00597817
[event_method]: 1.93e-05
[auto_monad]: 6.8e-05
[graph_reusing]: 5.69999e-06
[pre_auto_parallel]: 2.68e-06
[py_interpret_to_execute]: 4.378e-05
[rewriter_before_opt_a]: 7.792e-05
[expand_dump_flag]: 3.08e-06
[jit_opt_a]: 0.0234155, [2]
    [Cycle 1]: 0.00222858, [27]
        [switch_simplify]: 7.047e-05
        [loop_unroll]: 3.192e-05
        [a_1]: 0.0006408
        [with_stream_mark]: 2.816e-05
        [recompute_prepare]: 8e-06
        [updatestate_depend_eliminate]: 4.58001e-06
        [updatestate_assign_eliminate]: 3.12002e-06
        [updatestate_loads_eliminate]: 3.02002e-06
        [parameter_eliminate]: 2.33998e-06
        [specialize_transform]: 6.98e-06
        [updatestate_useless_node_eliminater]: 5.40999e-06
        [accelerated_algorithm]: 6.04001e-06
        [meta_shard_fg_expand]: 2.25002e-06
        [get_grad_eliminate_]: 5.95002e-06
        [merge_forward]: 3.4e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.696e-05
        [j_node_and_user_rematch]: 1.118e-05
        [meta_fg_expand]: 2.67001e-06
        [replace_old_param]: 1.066e-05
        [inline_without_move]: 5.27001e-06
        [renormalize]: 0.00108742
        [add_forward_monad_depend]: 8.1e-06
        [auto_monad_grad]: 2.57001e-06
        [auto_monad_eliminator]: 1.725e-05
        [cse]: 3.301e-05
        [replace_applicator]: 1.609e-05
    [Cycle 2]: 0.00033833, [27]
        [switch_simplify]: 5.42001e-06
        [loop_unroll]: 4.83001e-06
        [a_1]: 9.309e-05
        [with_stream_mark]: 1.231e-05
        [recompute_prepare]: 5.49e-06
        [updatestate_depend_eliminate]: 3.05002e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.91998e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 5.10001e-06
        [updatestate_useless_node_eliminater]: 4.76002e-06
        [accelerated_algorithm]: 5.93998e-06
        [meta_shard_fg_expand]: 1.81003e-06
        [get_grad_eliminate_]: 4.60001e-06
        [merge_forward]: 2.91999e-06
        [cell_reuse_recompute_pass]: 2.52001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.484e-05
        [j_node_and_user_rematch]: 7.68001e-06
        [meta_fg_expand]: 2.29001e-06
        [replace_old_param]: 7.5e-06
        [inline_without_move]: 4.87998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.50001e-06
        [auto_monad_grad]: 1.15999e-06
        [auto_monad_eliminator]: 6.21998e-06
        [cse]: 1.246e-05
        [replace_applicator]: 4.94998e-06
[py_interpret_to_execute_after_opt_a]: 1.565e-05
[rewriter_after_opt_a]: 4.259e-05
[convert_after_rewriter]: 7.81001e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.00079211
[jit_opt_b]: 5.385e-05, [1]
    [Cycle 1]: 4.574e-05, [2]
        [frontend_op_eliminate]: 1.785e-05
        [inline_after_opt_a]: 1.57e-05
[cconv]: 3.066e-05
[loop_unroll]: 0.00049694
[jit_opt_after_cconv]: 0.00015641, [1]
    [Cycle 1]: 0.00014861, [11]
        [c_1]: 2.217e-05
        [parameter_eliminate]: 4.92e-06
        [updatestate_depend_eliminate]: 7.61999e-06
        [updatestate_assign_eliminate]: 2.75997e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [cse]: 2.374e-05
        [call_graph_tuple_transform]: 2.183e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.58002e-06
        [renormalize]: 8.70001e-07
        [switch_simplify]: 6.26e-06
[remove_dup_value]: 1.55e-05
[partial_unused_args_eliminate]: 2.84001e-06
[environ_conv]: 6.68998e-06
[add_recomputation]: 5.755e-05
[cse_after_recomputation]: 2.163e-05, [1]
    [Cycle 1]: 1.563e-05, [1]
        [cse]: 9.31e-06
[auto_monad_reorder]: 1.558e-05
[get_jit_bprop_graph]: 2.40002e-06
[rewriter_after_jit_bprop_graph]: 7.06001e-06
[opt_after_jit_grad]: 0.00054787
[symbol_engine_optimizer]: 7.891e-05, [1]
    [Cycle 1]: 7.12e-05, [6]
        [build]: 4.35999e-06
        [elim_shapecalc]: 7.85e-06
        [elim_not_effective]: 1.417e-05
        [opt_reshape]: 5.86998e-06
        [fold_const_symbol]: 9.56e-06
        [renormalize]: 8.89995e-07
[validate]: 4.572e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00820579
[execute]: 8.90999e-06
Sums
    bootstrap                            :     0.000394s :  2.04%
    type_inference                       :     0.005978s : 30.91%
    event_method                         :     0.000019s :  0.10%
    auto_monad                           :     0.000068s :  0.35%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.23%
    rewriter_before_opt_a                :     0.000078s :  0.40%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.39%
    jit_opt_a.loop_unroll                :     0.000037s :  0.19%
    jit_opt_a.a_1                        :     0.000734s :  3.79%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.001088s :  5.62%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.12%
    jit_opt_a.cse                        :     0.000045s :  0.24%
    jit_opt_a.replace_applicator         :     0.000021s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.08%
    rewriter_after_opt_a                 :     0.000043s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000792s :  4.10%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.000031s :  0.16%
    loop_unroll                          :     0.000497s :  2.57%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000058s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000548s :  2.83%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008206s : 42.43%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000280    23
 0.99% :     0.000003s :      2: substitution.elim_not_effective
 0.56% :     0.000002s :      2: substitution.fold_const_symbol
 2.13% :     0.000006s :      3: substitution.graph_param_transform
86.08% :     0.000241s :      4: substitution.inline
 1.43% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.91% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.79% :     0.000005s :      2: substitution.replace_old_param
 5.11% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005893     2
82.71% :     0.004874s :      1: type_inference.infer
17.29% :     0.001019s :      1: type_inference.specialize
------[replace.]   0.000078     6
68.07% :     0.000053s :      4: replace.inline
31.93% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000252     6
94.85% :     0.000239s :      4: match.inline
 5.15% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.24% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.75% :     0.000003s :     10: predicate.arithmetic_simplify
 1.15% :     0.000001s :     10: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.30% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.47% :     0.000001s :      3: predicate.elim_not_effective
 0.43% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.40% :     0.000001s :      3: predicate.graph_param_transform
 5.75% :     0.000007s :     22: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 2.11% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.85% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.31% :     0.000003s :      3: predicate.mutable_eliminate
 0.39% :     0.000000s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     15: predicate.partial_eliminate
 1.01% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     10: predicate.reduce_eliminate
 1.70% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.55% :     0.000001s :      3: predicate.reset_defer_inline
 1.32% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.63% :     0.000010s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.53% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.43% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000797    12
45.99% :     0.000366s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.21% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.80% :     0.000373s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.043039    76
 0.14% :     0.000061s :      1: add_recomputation
 0.17% :     0.000071s :      1: auto_monad
 0.04% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.97% :     0.000418s :      1: bootstrap
 0.08% :     0.000033s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.06% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.06% :     0.000025s :      1: event_method
 0.03% :     0.000014s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
54.42% :     0.023420s :      1: jit_opt_a
 0.37% :     0.000159s :      1: jit_opt_after_cconv
 0.13% :     0.000057s :      1: jit_opt_b
 1.18% :     0.000507s :      1: loop_unroll
 1.86% :     0.000802s :      1: mutable_eliminate
 2.26% :     0.000974s :     26: opt.transform.jit_opt_a
 0.12% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000027s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.29% :     0.000557s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.11% :     0.000046s :      1: py_interpret_to_execute
 0.04% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000018s :      1: remove_dup_value
 1.40% :     0.000601s :      1: renormalize.infer
 1.11% :     0.000477s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000046s :      1: rewriter_after_opt_a
 0.19% :     0.000081s :      1: rewriter_before_opt_a
 0.19% :     0.000082s :      1: symbol_engine_optimizer
19.12% :     0.008228s :      1: task_emit
13.94% :     0.005998s :      1: type_inference
 0.16% :     0.000069s :      1: validate


TotalTime = 0.0236049, [33]
[bootstrap]: 0.00037355
[type_inference]: 0.00541199
[event_method]: 1.855e-05
[auto_monad]: 4.993e-05
[graph_reusing]: 3.98001e-06
[pre_auto_parallel]: 2.32001e-06
[py_interpret_to_execute]: 4.784e-05
[rewriter_before_opt_a]: 8.595e-05
[expand_dump_flag]: 3.50998e-06
[jit_opt_a]: 0.0072062, [2]
    [Cycle 1]: 0.00220105, [27]
        [switch_simplify]: 7.128e-05
        [loop_unroll]: 3.224e-05
        [a_1]: 0.000655
        [with_stream_mark]: 2.727e-05
        [recompute_prepare]: 8.53001e-06
        [updatestate_depend_eliminate]: 4.05e-06
        [updatestate_assign_eliminate]: 3.38999e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 2.46998e-06
        [specialize_transform]: 6.38003e-06
        [updatestate_useless_node_eliminater]: 5.32999e-06
        [accelerated_algorithm]: 6.36e-06
        [meta_shard_fg_expand]: 2.37999e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 4.03999e-06
        [cell_reuse_recompute_pass]: 1.28002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.421e-05
        [j_node_and_user_rematch]: 1.02e-05
        [meta_fg_expand]: 3.07002e-06
        [replace_old_param]: 1.074e-05
        [inline_without_move]: 5.49998e-06
        [renormalize]: 0.00104783
        [add_forward_monad_depend]: 7.1e-06
        [auto_monad_grad]: 2.77002e-06
        [auto_monad_eliminator]: 1.8e-05
        [cse]: 3.273e-05
        [replace_applicator]: 1.447e-05
    [Cycle 2]: 0.00034587, [27]
        [switch_simplify]: 6.08998e-06
        [loop_unroll]: 5.92001e-06
        [a_1]: 9.678e-05
        [with_stream_mark]: 1.135e-05
        [recompute_prepare]: 6.01e-06
        [updatestate_depend_eliminate]: 3.25e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 2.08002e-06
        [parameter_eliminate]: 1.14003e-06
        [specialize_transform]: 5.41002e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 5.62999e-06
        [meta_shard_fg_expand]: 1.13001e-06
        [get_grad_eliminate_]: 5.67999e-06
        [merge_forward]: 2.89999e-06
        [cell_reuse_recompute_pass]: 2.02999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.369e-05
        [j_node_and_user_rematch]: 8.62e-06
        [meta_fg_expand]: 1.87999e-06
        [replace_old_param]: 7.26999e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.42999e-06
        [auto_monad_grad]: 8.10018e-07
        [auto_monad_eliminator]: 5.13002e-06
        [cse]: 1.129e-05
        [replace_applicator]: 5.51998e-06
[py_interpret_to_execute_after_opt_a]: 1.399e-05
[rewriter_after_opt_a]: 4.213e-05
[convert_after_rewriter]: 7.77e-06
[order_py_execute_after_rewriter]: 4.98001e-06
[mutable_eliminate]: 0.00094661
[jit_opt_b]: 5.908e-05, [1]
    [Cycle 1]: 4.829e-05, [2]
        [frontend_op_eliminate]: 1.835e-05
        [inline_after_opt_a]: 1.678e-05
[cconv]: 3.585e-05
[loop_unroll]: 0.0005006
[jit_opt_after_cconv]: 0.00016584, [1]
    [Cycle 1]: 0.00015782, [11]
        [c_1]: 2.215e-05
        [parameter_eliminate]: 3.79002e-06
        [updatestate_depend_eliminate]: 7.18998e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [cse]: 2.981e-05
        [call_graph_tuple_transform]: 2.378e-05
        [tuple_list_get_item_eliminator]: 5.61003e-06
        [none_parameter_eliminate]: 1.61998e-06
        [renormalize]: 8.39995e-07
        [switch_simplify]: 5.47001e-06
[remove_dup_value]: 1.622e-05
[partial_unused_args_eliminate]: 3.33e-06
[environ_conv]: 6.57002e-06
[add_recomputation]: 5.803e-05
[cse_after_recomputation]: 2.327e-05, [1]
    [Cycle 1]: 1.663e-05, [1]
        [cse]: 1.02e-05
[auto_monad_reorder]: 1.724e-05
[get_jit_bprop_graph]: 2.26998e-06
[rewriter_after_jit_bprop_graph]: 9.00999e-06
[opt_after_jit_grad]: 0.00055771
[symbol_engine_optimizer]: 8.506e-05, [1]
    [Cycle 1]: 7.56e-05, [6]
        [build]: 5.19e-06
        [elim_shapecalc]: 8.04002e-06
        [elim_not_effective]: 1.594e-05
        [opt_reshape]: 5.89999e-06
        [fold_const_symbol]: 9.72001e-06
        [renormalize]: 7.79983e-07
[validate]: 5.073e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.00752763
[execute]: 1.095e-05
Sums
    bootstrap                            :     0.000374s :  2.05%
    type_inference                       :     0.005412s : 29.68%
    event_method                         :     0.000019s :  0.10%
    auto_monad                           :     0.000050s :  0.27%
    graph_reusing                        :     0.000004s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000048s :  0.26%
    rewriter_before_opt_a                :     0.000086s :  0.47%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000077s :  0.42%
    jit_opt_a.loop_unroll                :     0.000038s :  0.21%
    jit_opt_a.a_1                        :     0.000752s :  4.12%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001048s :  5.75%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.13%
    jit_opt_a.cse                        :     0.000044s :  0.24%
    jit_opt_a.replace_applicator         :     0.000020s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000042s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000947s :  5.19%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000036s :  0.20%
    loop_unroll                          :     0.000501s :  2.75%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000030s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000058s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000558s :  3.06%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000051s :  0.28%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007528s : 41.29%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000290    23
 0.88% :     0.000003s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 2.14% :     0.000006s :      3: substitution.graph_param_transform
86.70% :     0.000252s :      4: substitution.inline
 1.16% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.68% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.11% :     0.000006s :      2: substitution.replace_old_param
 4.90% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005335     2
78.78% :     0.004202s :      1: type_inference.infer
21.22% :     0.001132s :      1: type_inference.specialize
------[replace.]   0.000087     6
67.47% :     0.000059s :      4: replace.inline
32.53% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000262     6
95.00% :     0.000249s :      4: match.inline
 5.00% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.23% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.92% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000002s :     10: predicate.addn_zero_filter
 2.53% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000002s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.14% :     0.000001s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.54% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.08% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.74% :     0.000007s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.49% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.39% :     0.000002s :      6: predicate.less_batch_normalization
 1.81% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.84% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.22% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.27% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 2.20% :     0.000003s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 1.95% :     0.000002s :     15: predicate.partial_eliminate
 1.29% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.63% :     0.000001s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.99% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.67% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 1.85% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.44% :     0.000009s :     53: predicate.switch_simplify
 1.47% :     0.000002s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.53% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.59% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.92% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.53% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.98% :     0.000003s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000719    12
44.37% :     0.000319s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.47% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.16% :     0.000339s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025751    76
 0.24% :     0.000062s :      1: add_recomputation
 0.21% :     0.000053s :      1: auto_monad
 0.08% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.53% :     0.000395s :      1: bootstrap
 0.15% :     0.000038s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.06% :     0.000017s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
28.00% :     0.007211s :      1: jit_opt_a
 0.65% :     0.000169s :      1: jit_opt_after_cconv
 0.24% :     0.000062s :      1: jit_opt_b
 1.98% :     0.000510s :      1: loop_unroll
 3.72% :     0.000958s :      1: mutable_eliminate
 3.85% :     0.000992s :     26: opt.transform.jit_opt_a
 0.21% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000028s :      4: opt.transform.jit_opt_b
 0.06% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000570s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000051s :      1: py_interpret_to_execute
 0.06% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 2.27% :     0.000585s :      1: renormalize.infer
 1.76% :     0.000453s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000046s :      1: rewriter_after_opt_a
 0.35% :     0.000089s :      1: rewriter_before_opt_a
 0.34% :     0.000088s :      1: symbol_engine_optimizer
29.32% :     0.007551s :      1: task_emit
21.08% :     0.005428s :      1: type_inference
 0.29% :     0.000075s :      1: validate


TotalTime = 0.0262872, [33]
[bootstrap]: 0.00043347
[type_inference]: 0.00544566
[event_method]: 1.873e-05
[auto_monad]: 6.889e-05
[graph_reusing]: 6.09001e-06
[pre_auto_parallel]: 2.68e-06
[py_interpret_to_execute]: 4.05e-05
[rewriter_before_opt_a]: 7.602e-05
[expand_dump_flag]: 3.07002e-06
[jit_opt_a]: 0.0113482, [2]
    [Cycle 1]: 0.00699987, [27]
        [switch_simplify]: 6.336e-05
        [loop_unroll]: 2.875e-05
        [a_1]: 0.0055525
        [with_stream_mark]: 2.55e-05
        [recompute_prepare]: 1.084e-05
        [updatestate_depend_eliminate]: 4.60999e-06
        [updatestate_assign_eliminate]: 2.95998e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 2.39999e-06
        [specialize_transform]: 6.29001e-06
        [updatestate_useless_node_eliminater]: 5.37001e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 2.75997e-06
        [get_grad_eliminate_]: 5.58002e-06
        [merge_forward]: 3.98001e-06
        [cell_reuse_recompute_pass]: 1.80001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.558e-05
        [j_node_and_user_rematch]: 9.64e-06
        [meta_fg_expand]: 3.60003e-06
        [replace_old_param]: 1.004e-05
        [inline_without_move]: 6.01e-06
        [renormalize]: 0.00095488
        [add_forward_monad_depend]: 5.66003e-06
        [auto_monad_grad]: 2.49001e-06
        [auto_monad_eliminator]: 1.495e-05
        [cse]: 3.198e-05
        [replace_applicator]: 1.209e-05
    [Cycle 2]: 0.00031706, [27]
        [switch_simplify]: 5.51e-06
        [loop_unroll]: 5.06997e-06
        [a_1]: 9.075e-05
        [with_stream_mark]: 8.48001e-06
        [recompute_prepare]: 4.74998e-06
        [updatestate_depend_eliminate]: 2.60002e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 1.02998e-06
        [specialize_transform]: 4.90999e-06
        [updatestate_useless_node_eliminater]: 4.62e-06
        [accelerated_algorithm]: 4.97999e-06
        [meta_shard_fg_expand]: 1.09998e-06
        [get_grad_eliminate_]: 4.40999e-06
        [merge_forward]: 2.29001e-06
        [cell_reuse_recompute_pass]: 1.44998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.279e-05
        [j_node_and_user_rematch]: 7.5e-06
        [meta_fg_expand]: 1.48002e-06
        [replace_old_param]: 7.01001e-06
        [inline_without_move]: 4.53001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.35001e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 5.17e-06
        [cse]: 1.051e-05
        [replace_applicator]: 4.69002e-06
[py_interpret_to_execute_after_opt_a]: 1.234e-05
[rewriter_after_opt_a]: 3.976e-05
[convert_after_rewriter]: 8.18001e-06
[order_py_execute_after_rewriter]: 5.59e-06
[mutable_eliminate]: 0.00086055
[jit_opt_b]: 5.267e-05, [1]
    [Cycle 1]: 4.485e-05, [2]
        [frontend_op_eliminate]: 1.814e-05
        [inline_after_opt_a]: 1.453e-05
[cconv]: 2.837e-05
[loop_unroll]: 0.00043652
[jit_opt_after_cconv]: 0.00013775, [1]
    [Cycle 1]: 0.00013123, [11]
        [c_1]: 2.048e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 5.16002e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 1.827e-05
        [call_graph_tuple_transform]: 1.94e-05
        [tuple_list_get_item_eliminator]: 5.42001e-06
        [none_parameter_eliminate]: 1.57999e-06
        [renormalize]: 5.79981e-07
        [switch_simplify]: 4.90999e-06
[remove_dup_value]: 1.497e-05
[partial_unused_args_eliminate]: 2.22001e-06
[environ_conv]: 5.20001e-06
[add_recomputation]: 5.198e-05
[cse_after_recomputation]: 2.105e-05, [1]
    [Cycle 1]: 1.484e-05, [1]
        [cse]: 9.02e-06
[auto_monad_reorder]: 1.476e-05
[get_jit_bprop_graph]: 1.96e-06
[rewriter_after_jit_bprop_graph]: 4.79e-06
[opt_after_jit_grad]: 0.00045287
[symbol_engine_optimizer]: 7.534e-05, [1]
    [Cycle 1]: 6.899e-05, [6]
        [build]: 4.46002e-06
        [elim_shapecalc]: 7.64002e-06
        [elim_not_effective]: 1.306e-05
        [opt_reshape]: 5.56e-06
        [fold_const_symbol]: 9.25999e-06
        [renormalize]: 4.89992e-07
[validate]: 3.865e-05
[backend_pass]: 8.2e-07
[task_emit]: 0.00634795
[execute]: 9.32001e-06
Sums
    bootstrap                            :     0.000433s :  2.01%
    type_inference                       :     0.005446s : 25.22%
    event_method                         :     0.000019s :  0.09%
    auto_monad                           :     0.000069s :  0.32%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000040s :  0.19%
    rewriter_before_opt_a                :     0.000076s :  0.35%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000069s :  0.32%
    jit_opt_a.loop_unroll                :     0.000034s :  0.16%
    jit_opt_a.a_1                        :     0.005643s : 26.14%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.08%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.000955s :  4.42%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.09%
    jit_opt_a.cse                        :     0.000042s :  0.20%
    jit_opt_a.replace_applicator         :     0.000017s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.06%
    rewriter_after_opt_a                 :     0.000040s :  0.18%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000861s :  3.99%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.07%
    cconv                                :     0.000028s :  0.13%
    loop_unroll                          :     0.000437s :  2.02%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000015s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000052s :  0.24%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000015s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000453s :  2.10%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000039s :  0.18%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006348s : 29.40%
    execute                              :     0.000009s :  0.04%


Time group info:
------[substitution.]   0.005139    23
 0.04% :     0.000002s :      2: substitution.elim_not_effective
 0.03% :     0.000001s :      2: substitution.fold_const_symbol
 0.10% :     0.000005s :      3: substitution.graph_param_transform
99.30% :     0.005103s :      4: substitution.inline
 0.06% :     0.000003s :      4: substitution.j_node_and_user_rematch
 0.09% :     0.000005s :      4: substitution.remove_not_recompute_node
 0.09% :     0.000005s :      2: substitution.replace_old_param
 0.29% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005368     2
82.36% :     0.004421s :      1: type_inference.infer
17.64% :     0.000947s :      1: type_inference.specialize
------[replace.]   0.000096     6
72.04% :     0.000069s :      4: replace.inline
27.96% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.005110     6
99.74% :     0.005096s :      4: match.inline
 0.26% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   691
 1.51% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.37% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.46% :     0.000002s :     10: predicate.addn_zero_filter
 3.17% :     0.000004s :     10: predicate.arithmetic_simplify
 1.22% :     0.000002s :     10: predicate.cast_eliminate
 0.56% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.73% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 6.48% :     0.000008s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000002s :      6: predicate.less_batch_normalization
 1.91% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.72% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 1.16% :     0.000001s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.34% :     0.000003s :     15: predicate.partial_eliminate
 1.33% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.83% :     0.000002s :     10: predicate.reduce_eliminate
 1.56% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.49% :     0.000002s :     18: predicate.replace_applicator
 0.44% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.32% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.16% :     0.000004s :     16: predicate.switch_defer_inline
 1.89% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.95% :     0.000009s :     53: predicate.switch_simplify
 1.43% :     0.000002s :     10: predicate.tile_eliminate
 1.19% :     0.000002s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.87% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.44% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.98% :     0.000003s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000717    12
48.50% :     0.000348s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.50% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.99% :     0.000308s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.033199    76
 0.17% :     0.000055s :      1: add_recomputation
 0.22% :     0.000072s :      1: auto_monad
 0.05% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.37% :     0.000455s :      1: bootstrap
 0.09% :     0.000031s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.07% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.07% :     0.000024s :      1: event_method
 0.04% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
34.20% :     0.011353s :      1: jit_opt_a
 0.42% :     0.000140s :      1: jit_opt_after_cconv
 0.17% :     0.000055s :      1: jit_opt_b
 1.34% :     0.000445s :      1: loop_unroll
 2.62% :     0.000869s :      1: mutable_eliminate
17.66% :     0.005863s :     26: opt.transform.jit_opt_a
 0.14% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000026s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.39% :     0.000461s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000043s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000017s :      1: remove_dup_value
 1.63% :     0.000543s :      1: renormalize.infer
 1.22% :     0.000404s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000043s :      1: rewriter_after_opt_a
 0.24% :     0.000079s :      1: rewriter_before_opt_a
 0.23% :     0.000078s :      1: symbol_engine_optimizer
19.17% :     0.006364s :      1: task_emit
16.46% :     0.005463s :      1: type_inference
 0.18% :     0.000061s :      1: validate


TotalTime = 0.0229301, [30]
[bootstrap]: 0.00034743
[type_inference]: 0.00857107
[event_method]: 2.428e-05
[auto_monad]: 8.916e-05
[graph_reusing]: 6.89999e-06
[pre_auto_parallel]: 3.09999e-06
[py_interpret_to_execute]: 5.833e-05
[rewriter_before_opt_a]: 8.814e-05
[expand_dump_flag]: 2.91999e-06
[jit_opt_a]: 0.00904335, [2]
    [Cycle 1]: 0.00341917, [27]
        [switch_simplify]: 0.00014851
        [loop_unroll]: 3.352e-05
        [a_1]: 0.00077506
        [with_stream_mark]: 2.533e-05
        [recompute_prepare]: 7.4e-06
        [updatestate_depend_eliminate]: 4.01001e-06
        [updatestate_assign_eliminate]: 3.16001e-06
        [updatestate_loads_eliminate]: 2.69001e-06
        [parameter_eliminate]: 1.91998e-06
        [specialize_transform]: 6.27001e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 5.40999e-06
        [meta_shard_fg_expand]: 2.29999e-06
        [get_grad_eliminate_]: 5.22e-06
        [merge_forward]: 3.92998e-06
        [cell_reuse_recompute_pass]: 1.10999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.513e-05
        [j_node_and_user_rematch]: 9.56e-06
        [meta_fg_expand]: 2.44001e-06
        [replace_old_param]: 1.053e-05
        [inline_without_move]: 5.40001e-06
        [renormalize]: 0.00208329
        [add_forward_monad_depend]: 6.51999e-06
        [auto_monad_grad]: 2.43998e-06
        [auto_monad_eliminator]: 1.443e-05
        [cse]: 3.199e-05
        [replace_applicator]: 1.558e-05
    [Cycle 2]: 0.00033861, [27]
        [switch_simplify]: 5.85002e-06
        [loop_unroll]: 5.12e-06
        [a_1]: 9.618e-05
        [with_stream_mark]: 1.241e-05
        [recompute_prepare]: 5.01002e-06
        [updatestate_depend_eliminate]: 2.86e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [parameter_eliminate]: 1.50001e-06
        [specialize_transform]: 5.17e-06
        [updatestate_useless_node_eliminater]: 4.64998e-06
        [accelerated_algorithm]: 5.00999e-06
        [meta_shard_fg_expand]: 1.29998e-06
        [get_grad_eliminate_]: 4.73001e-06
        [merge_forward]: 2.70002e-06
        [cell_reuse_recompute_pass]: 2.56e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.378e-05
        [j_node_and_user_rematch]: 7.83001e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 1.644e-05
        [inline_without_move]: 4.77e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.19998e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 4.75001e-06
        [cse]: 1.031e-05
        [replace_applicator]: 4.90001e-06
[py_interpret_to_execute_after_opt_a]: 1.194e-05
[rewriter_after_opt_a]: 3.759e-05
[convert_after_rewriter]: 7.30998e-06
[order_py_execute_after_rewriter]: 5.22999e-06
[mutable_eliminate]: 0.0019242
[jit_opt_b]: 6.104e-05, [1]
    [Cycle 1]: 5.122e-05, [2]
        [frontend_op_eliminate]: 2.019e-05
        [inline_after_opt_a]: 1.751e-05
[cconv]: 3.481e-05
[loop_unroll]: 0.00051008
[jit_opt_after_cconv]: 0.00016284, [1]
    [Cycle 1]: 0.00015531, [11]
        [c_1]: 2.284e-05
        [parameter_eliminate]: 4.33001e-06
        [updatestate_depend_eliminate]: 7.2e-06
        [updatestate_assign_eliminate]: 3.31999e-06
        [updatestate_loads_eliminate]: 2.34999e-06
        [cse]: 2.86e-05
        [call_graph_tuple_transform]: 2.239e-05
        [tuple_list_get_item_eliminator]: 5.34e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 7.10017e-07
        [switch_simplify]: 5.79999e-06
[remove_dup_value]: 1.432e-05
[partial_unused_args_eliminate]: 2.37001e-06
[environ_conv]: 4.92e-06
[add_recomputation]: 5.597e-05
[cse_after_recomputation]: 2.331e-05, [1]
    [Cycle 1]: 1.698e-05, [1]
        [cse]: 9.62001e-06
[auto_monad_reorder]: 1.672e-05
[get_jit_bprop_graph]: 2.44001e-06
[rewriter_after_jit_bprop_graph]: 8.53001e-06
[opt_after_jit_grad]: 0.00147455
[symbol_engine_optimizer]: 8.026e-05, [1]
    [Cycle 1]: 7.232e-05, [6]
        [build]: 3.80998e-06
        [elim_shapecalc]: 7.79997e-06
        [elim_not_effective]: 1.505e-05
        [opt_reshape]: 6.75002e-06
        [fold_const_symbol]: 9.59e-06
        [renormalize]: 6.50005e-07
[validate]: 4.103e-05
Sums
    bootstrap                            :     0.000347s :  2.04%
    type_inference                       :     0.008571s : 50.44%
    event_method                         :     0.000024s :  0.14%
    auto_monad                           :     0.000089s :  0.52%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000058s :  0.34%
    rewriter_before_opt_a                :     0.000088s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000154s :  0.91%
    jit_opt_a.loop_unroll                :     0.000039s :  0.23%
    jit_opt_a.a_1                        :     0.000871s :  5.13%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000027s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.002083s : 12.26%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.11%
    jit_opt_a.cse                        :     0.000042s :  0.25%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.001924s : 11.32%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.10%
    cconv                                :     0.000035s :  0.20%
    loop_unroll                          :     0.000510s :  3.00%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000056s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.001475s :  8.68%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.24%


Time group info:
------[substitution.]   0.000362    26
 0.65% :     0.000002s :      2: substitution.elim_not_effective
 0.36% :     0.000001s :      2: substitution.fold_const_symbol
 1.39% :     0.000005s :      3: substitution.graph_param_transform
84.87% :     0.000307s :      6: substitution.inline
 0.90% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.48% :     0.000005s :      4: substitution.remove_not_recompute_node
 3.64% :     0.000013s :      2: substitution.replace_old_param
 3.04% :     0.000011s :      1: substitution.switch_simplify
 3.68% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008486     2
83.88% :     0.007118s :      1: type_inference.infer
16.12% :     0.001368s :      1: type_inference.specialize
------[replace.]   0.000159     9
47.90% :     0.000076s :      6: replace.inline
29.96% :     0.000048s :      1: replace.switch_simplify
22.14% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000326     9
93.24% :     0.000304s :      6: match.inline
 3.06% :     0.000010s :      1: match.switch_simplify
 3.70% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000143   801
 1.50% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.41% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000002s :     12: predicate.addn_check_dump
 1.20% :     0.000002s :     12: predicate.addn_zero_filter
 2.35% :     0.000003s :     12: predicate.arithmetic_simplify
 1.25% :     0.000002s :     12: predicate.cast_eliminate
 0.33% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     12: predicate.compare_switch_simplify
 1.23% :     0.000002s :     12: predicate.depend_value_elim
 1.06% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     12: predicate.dict_set_item_eliminator
 1.08% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.66% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.09% :     0.000002s :     12: predicate.environ_get_eliminate
 1.10% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.83% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.32% :     0.000008s :     26: predicate.inline
 0.73% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.97% :     0.000001s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     17: predicate.load_eliminater
 1.64% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.84% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.91% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000002s :     12: predicate.merge_addn
 1.07% :     0.000002s :     12: predicate.minmaximum_grad
 2.75% :     0.000004s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 1.98% :     0.000003s :     17: predicate.partial_eliminate
 1.29% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     12: predicate.reduce_eliminate
 1.51% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.52% :     0.000002s :     20: predicate.replace_applicator
 0.95% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     12: predicate.reshape_eliminate
 1.11% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.97% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.43% :     0.000002s :     12: predicate.same_eliminate
 0.40% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.80% :     0.000001s :      6: predicate.special_op_eliminate
 0.78% :     0.000001s :      6: predicate.specialize_transform
 1.67% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.31% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.63% :     0.000004s :     20: predicate.switch_defer_inline
 2.22% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.29% :     0.000012s :     65: predicate.switch_simplify
 1.20% :     0.000002s :     12: predicate.tile_eliminate
 1.27% :     0.000002s :     12: predicate.transpose_eliminate
 1.49% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.46% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.84% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     12: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.84% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001043    15
43.95% :     0.000458s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.92% :     0.000135s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.13% :     0.000450s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026262    72
 0.23% :     0.000059s :      1: add_recomputation
 0.36% :     0.000094s :      1: auto_monad
 0.07% :     0.000020s :      1: auto_monad_reorder
 1.40% :     0.000367s :      1: bootstrap
 0.14% :     0.000038s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.11% :     0.000029s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
34.45% :     0.009048s :      1: jit_opt_a
 0.63% :     0.000166s :      1: jit_opt_after_cconv
 0.24% :     0.000064s :      1: jit_opt_b
 1.98% :     0.000519s :      1: loop_unroll
 7.39% :     0.001940s :      1: mutable_eliminate
 4.53% :     0.001191s :     26: opt.transform.jit_opt_a
 0.20% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000030s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000036s :      4: opt.transform.symbol_engine_opt
 5.66% :     0.001487s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000062s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 6.15% :     0.001614s :      1: renormalize.infer
 1.74% :     0.000458s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000041s :      1: rewriter_after_opt_a
 0.35% :     0.000091s :      1: rewriter_before_opt_a
 0.32% :     0.000083s :      1: symbol_engine_optimizer
32.71% :     0.008591s :      1: type_inference


TotalTime = 0.0223975, [33]
[bootstrap]: 0.00030384
[type_inference]: 0.0063252
[event_method]: 1.951e-05
[auto_monad]: 6.872e-05
[graph_reusing]: 6.29999e-06
[pre_auto_parallel]: 2.51998e-06
[py_interpret_to_execute]: 3.999e-05
[rewriter_before_opt_a]: 7.628e-05
[expand_dump_flag]: 3.41001e-06
[jit_opt_a]: 0.00631722, [2]
    [Cycle 1]: 0.00212008, [27]
        [switch_simplify]: 6.648e-05
        [loop_unroll]: 3.131e-05
        [a_1]: 0.00063834
        [with_stream_mark]: 2.392e-05
        [recompute_prepare]: 8.78001e-06
        [updatestate_depend_eliminate]: 4.47998e-06
        [updatestate_assign_eliminate]: 3.13e-06
        [updatestate_loads_eliminate]: 2.74001e-06
        [parameter_eliminate]: 2.20002e-06
        [specialize_transform]: 6.34001e-06
        [updatestate_useless_node_eliminater]: 5.51e-06
        [accelerated_algorithm]: 5.94e-06
        [meta_shard_fg_expand]: 2.58998e-06
        [get_grad_eliminate_]: 5.77001e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 1.53002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.634e-05
        [j_node_and_user_rematch]: 1.105e-05
        [meta_fg_expand]: 2.73e-06
        [replace_old_param]: 9.59999e-06
        [inline_without_move]: 5.46e-06
        [renormalize]: 0.00100545
        [add_forward_monad_depend]: 6.58998e-06
        [auto_monad_grad]: 2.54999e-06
        [auto_monad_eliminator]: 1.549e-05
        [cse]: 3.266e-05
        [replace_applicator]: 1.326e-05
    [Cycle 2]: 0.00033902, [27]
        [switch_simplify]: 6.09999e-06
        [loop_unroll]: 5.34e-06
        [a_1]: 9.64e-05
        [with_stream_mark]: 1.064e-05
        [recompute_prepare]: 5.54e-06
        [updatestate_depend_eliminate]: 2.81e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 4.88001e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.44e-06
        [meta_shard_fg_expand]: 1.50001e-06
        [get_grad_eliminate_]: 4.84e-06
        [merge_forward]: 2.59001e-06
        [cell_reuse_recompute_pass]: 2.00002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.361e-05
        [j_node_and_user_rematch]: 8.24998e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 8e-06
        [inline_without_move]: 4.77e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 5.66998e-06
        [cse]: 1.177e-05
        [replace_applicator]: 5.15999e-06
[py_interpret_to_execute_after_opt_a]: 1.288e-05
[rewriter_after_opt_a]: 3.777e-05
[convert_after_rewriter]: 1.021e-05
[order_py_execute_after_rewriter]: 4.92e-06
[mutable_eliminate]: 0.00082936
[jit_opt_b]: 5.754e-05, [1]
    [Cycle 1]: 4.936e-05, [2]
        [frontend_op_eliminate]: 2.006e-05
        [inline_after_opt_a]: 1.725e-05
[cconv]: 3.084e-05
[loop_unroll]: 0.00047363
[jit_opt_after_cconv]: 0.00016523, [1]
    [Cycle 1]: 0.00015769, [11]
        [c_1]: 2.306e-05
        [parameter_eliminate]: 4.28999e-06
        [updatestate_depend_eliminate]: 7.92e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 3.08e-06
        [cse]: 2.529e-05
        [call_graph_tuple_transform]: 2.316e-05
        [tuple_list_get_item_eliminator]: 6.11998e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 9.80013e-07
        [switch_simplify]: 5.77001e-06
[remove_dup_value]: 1.565e-05
[partial_unused_args_eliminate]: 2.34001e-06
[environ_conv]: 6.06e-06
[add_recomputation]: 5.477e-05
[cse_after_recomputation]: 2.141e-05, [1]
    [Cycle 1]: 1.51e-05, [1]
        [cse]: 9.10001e-06
[auto_monad_reorder]: 1.662e-05
[get_jit_bprop_graph]: 2.37001e-06
[rewriter_after_jit_bprop_graph]: 7.98999e-06
[opt_after_jit_grad]: 0.00053315
[symbol_engine_optimizer]: 8.074e-05, [1]
    [Cycle 1]: 7.288e-05, [6]
        [build]: 3.45998e-06
        [elim_shapecalc]: 8.67e-06
        [elim_not_effective]: 1.54e-05
        [opt_reshape]: 6.59999e-06
        [fold_const_symbol]: 9.45001e-06
        [renormalize]: 1.04e-06
[validate]: 4.241e-05
[backend_pass]: 8.79983e-07
[task_emit]: 0.00657916
[execute]: 1.108e-05
Sums
    bootstrap                            :     0.000304s :  1.70%
    type_inference                       :     0.006325s : 35.40%
    event_method                         :     0.000020s :  0.11%
    auto_monad                           :     0.000069s :  0.38%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000040s :  0.22%
    rewriter_before_opt_a                :     0.000076s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000073s :  0.41%
    jit_opt_a.loop_unroll                :     0.000037s :  0.21%
    jit_opt_a.a_1                        :     0.000735s :  4.11%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001006s :  5.63%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.12%
    jit_opt_a.cse                        :     0.000044s :  0.25%
    jit_opt_a.replace_applicator         :     0.000018s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.21%
    convert_after_rewriter               :     0.000010s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000829s :  4.64%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.10%
    cconv                                :     0.000031s :  0.17%
    loop_unroll                          :     0.000474s :  2.65%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000025s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000055s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000533s :  2.98%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000042s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006579s : 36.83%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000275    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000002s :      2: substitution.fold_const_symbol
 2.34% :     0.000006s :      3: substitution.graph_param_transform
85.80% :     0.000236s :      4: substitution.inline
 1.39% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.73% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.92% :     0.000005s :      2: substitution.replace_old_param
 5.31% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006251     2
83.93% :     0.005246s :      1: type_inference.infer
16.07% :     0.001004s :      1: type_inference.specialize
------[replace.]   0.000080     6
65.81% :     0.000053s :      4: replace.inline
34.19% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000246     6
94.58% :     0.000233s :      4: match.inline
 5.42% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   691
 1.27% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.74% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.18% :     0.000002s :     10: predicate.addn_zero_filter
 2.15% :     0.000003s :     10: predicate.arithmetic_simplify
 1.09% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.39% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.47% :     0.000001s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.03% :     0.000001s :     10: predicate.environ_get_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.80% :     0.000008s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.58% :     0.000002s :      6: predicate.less_batch_normalization
 1.79% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.55% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.83% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.27% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 0.95% :     0.000001s :     10: predicate.minmaximum_grad
 2.45% :     0.000003s :      3: predicate.mutable_eliminate
 0.65% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     15: predicate.partial_eliminate
 0.99% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.82% :     0.000002s :     10: predicate.reduce_eliminate
 1.61% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.29% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.10% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.77% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.39% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.46% :     0.000003s :     16: predicate.switch_defer_inline
 1.76% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.46% :     0.000010s :     53: predicate.switch_simplify
 1.85% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.91% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.52% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000729    12
45.67% :     0.000333s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.30% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.03% :     0.000343s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024483    76
 0.24% :     0.000058s :      1: add_recomputation
 0.30% :     0.000073s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.31% :     0.000321s :      1: bootstrap
 0.14% :     0.000033s :      1: cconv
 0.05% :     0.000013s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.07% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
25.82% :     0.006322s :      1: jit_opt_a
 0.69% :     0.000168s :      1: jit_opt_after_cconv
 0.25% :     0.000061s :      1: jit_opt_b
 1.97% :     0.000483s :      1: loop_unroll
 3.43% :     0.000840s :      1: mutable_eliminate
 3.95% :     0.000967s :     26: opt.transform.jit_opt_a
 0.22% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000029s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000541s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000043s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.31% :     0.000565s :      1: renormalize.infer
 1.77% :     0.000432s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000041s :      1: rewriter_after_opt_a
 0.32% :     0.000079s :      1: rewriter_before_opt_a
 0.34% :     0.000083s :      1: symbol_engine_optimizer
26.97% :     0.006604s :      1: task_emit
25.91% :     0.006343s :      1: type_inference
 0.26% :     0.000064s :      1: validate


TotalTime = 0.0285111, [33]
[bootstrap]: 0.00046117
[type_inference]: 0.00690524
[event_method]: 2.24e-05
[auto_monad]: 7.251e-05
[graph_reusing]: 5.91e-06
[pre_auto_parallel]: 2.53998e-06
[py_interpret_to_execute]: 5.048e-05
[rewriter_before_opt_a]: 7.731e-05
[expand_dump_flag]: 3.73999e-06
[jit_opt_a]: 0.0116219, [2]
    [Cycle 1]: 0.00214171, [27]
        [switch_simplify]: 6.02e-05
        [loop_unroll]: 2.875e-05
        [a_1]: 0.00057922
        [with_stream_mark]: 1.793e-05
        [recompute_prepare]: 7.28999e-06
        [updatestate_depend_eliminate]: 2.88e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [parameter_eliminate]: 1.39e-06
        [specialize_transform]: 6.42001e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.42999e-06
        [meta_shard_fg_expand]: 1.81e-06
        [get_grad_eliminate_]: 5.46002e-06
        [merge_forward]: 3.61001e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.151e-05
        [j_node_and_user_rematch]: 8.40999e-06
        [meta_fg_expand]: 1.75001e-06
        [replace_old_param]: 8.80001e-06
        [inline_without_move]: 5.35001e-06
        [renormalize]: 0.00112466
        [add_forward_monad_depend]: 7.58001e-06
        [auto_monad_grad]: 2.39001e-06
        [auto_monad_eliminator]: 1.729e-05
        [cse]: 2.658e-05
        [replace_applicator]: 1.614e-05
    [Cycle 2]: 0.00038903, [27]
        [switch_simplify]: 6.73998e-06
        [loop_unroll]: 5.40001e-06
        [a_1]: 9.892e-05
        [with_stream_mark]: 1.506e-05
        [recompute_prepare]: 6.26998e-06
        [updatestate_depend_eliminate]: 3.81999e-06
        [updatestate_assign_eliminate]: 2.47001e-06
        [updatestate_loads_eliminate]: 1.74998e-06
        [parameter_eliminate]: 1.75001e-06
        [specialize_transform]: 6.19001e-06
        [updatestate_useless_node_eliminater]: 4.83001e-06
        [accelerated_algorithm]: 6.31e-06
        [meta_shard_fg_expand]: 2.56e-06
        [get_grad_eliminate_]: 4.79998e-06
        [merge_forward]: 4.80001e-06
        [cell_reuse_recompute_pass]: 3.21999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.715e-05
        [j_node_and_user_rematch]: 9.29e-06
        [meta_fg_expand]: 1.80001e-06
        [replace_old_param]: 8.33999e-06
        [inline_without_move]: 4.65999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 2.55002e-06
        [auto_monad_grad]: 2.49001e-06
        [auto_monad_eliminator]: 7.58999e-06
        [cse]: 1.522e-05
        [replace_applicator]: 7.85e-06
[py_interpret_to_execute_after_opt_a]: 1.337e-05
[rewriter_after_opt_a]: 4.261e-05
[convert_after_rewriter]: 9.20999e-06
[order_py_execute_after_rewriter]: 5.25001e-06
[mutable_eliminate]: 0.00072293
[jit_opt_b]: 6.538e-05, [1]
    [Cycle 1]: 5.728e-05, [2]
        [frontend_op_eliminate]: 2.664e-05
        [inline_after_opt_a]: 1.68e-05
[cconv]: 3.396e-05
[loop_unroll]: 0.00053083
[jit_opt_after_cconv]: 0.0001679, [1]
    [Cycle 1]: 0.00016047, [11]
        [c_1]: 2.307e-05
        [parameter_eliminate]: 4.74e-06
        [updatestate_depend_eliminate]: 8.05e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [cse]: 3.021e-05
        [call_graph_tuple_transform]: 2.221e-05
        [tuple_list_get_item_eliminator]: 5.84e-06
        [none_parameter_eliminate]: 1.22e-06
        [renormalize]: 8.70001e-07
        [switch_simplify]: 5.83002e-06
[remove_dup_value]: 1.491e-05
[partial_unused_args_eliminate]: 2.71999e-06
[environ_conv]: 7.01001e-06
[add_recomputation]: 5.31e-05
[cse_after_recomputation]: 2.425e-05, [1]
    [Cycle 1]: 1.775e-05, [1]
        [cse]: 1.112e-05
[auto_monad_reorder]: 1.517e-05
[get_jit_bprop_graph]: 2.42001e-06
[rewriter_after_jit_bprop_graph]: 8.42e-06
[opt_after_jit_grad]: 0.00064325
[symbol_engine_optimizer]: 8.474e-05, [1]
    [Cycle 1]: 7.571e-05, [6]
        [build]: 5.15999e-06
        [elim_shapecalc]: 8.99998e-06
        [elim_not_effective]: 1.565e-05
        [opt_reshape]: 6.14001e-06
        [fold_const_symbol]: 9.37999e-06
        [renormalize]: 7.09988e-07
[validate]: 0.00012412
[backend_pass]: 9.39996e-07
[task_emit]: 0.00644812
[execute]: 8.46002e-06
Sums
    bootstrap                            :     0.000461s :  2.46%
    type_inference                       :     0.006905s : 36.91%
    event_method                         :     0.000022s :  0.12%
    auto_monad                           :     0.000073s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000050s :  0.27%
    rewriter_before_opt_a                :     0.000077s :  0.41%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.36%
    jit_opt_a.loop_unroll                :     0.000034s :  0.18%
    jit_opt_a.a_1                        :     0.000678s :  3.62%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.001125s :  6.01%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.13%
    jit_opt_a.cse                        :     0.000042s :  0.22%
    jit_opt_a.replace_applicator         :     0.000024s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000043s :  0.23%
    convert_after_rewriter               :     0.000009s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000723s :  3.86%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000034s :  0.18%
    loop_unroll                          :     0.000531s :  2.84%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000030s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000053s :  0.28%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000643s :  3.44%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000124s :  0.66%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006448s : 34.47%
    execute                              :     0.000008s :  0.05%


TotalTime = 0.0281288, [33]
[bootstrap]: 0.0004508
[type_inference]: 0.00979221
[event_method]: 1.935e-05
[auto_monad]: 6.827e-05
[graph_reusing]: 5.84999e-06
[pre_auto_parallel]: 2.39999e-06
[py_interpret_to_execute]: 4.061e-05
[rewriter_before_opt_a]: 7.029e-05
[expand_dump_flag]: 4.16001e-06
[jit_opt_a]: 0.00765605, [2]
    [Cycle 1]: 0.00263792, [27]
        [switch_simplify]: 6.198e-05
        [loop_unroll]: 3.158e-05
        [a_1]: 0.00067518
        [with_stream_mark]: 2.169e-05
        [recompute_prepare]: 8.72e-06
        [updatestate_depend_eliminate]: 3.28998e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [parameter_eliminate]: 1.60999e-06
        [specialize_transform]: 6.13002e-06
        [updatestate_useless_node_eliminater]: 5.79999e-06
        [accelerated_algorithm]: 6.76e-06
        [meta_shard_fg_expand]: 2.54001e-06
        [get_grad_eliminate_]: 5.84e-06
        [merge_forward]: 3.18998e-06
        [cell_reuse_recompute_pass]: 1.62001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.492e-05
        [j_node_and_user_rematch]: 9.54e-06
        [meta_fg_expand]: 2.64999e-06
        [replace_old_param]: 8.91002e-06
        [inline_without_move]: 6.16e-06
        [renormalize]: 0.00149179
        [add_forward_monad_depend]: 8.79998e-06
        [auto_monad_grad]: 3.26001e-06
        [auto_monad_eliminator]: 1.559e-05
        [cse]: 2.403e-05
        [replace_applicator]: 1.707e-05
    [Cycle 2]: 0.00038016, [27]
        [switch_simplify]: 6.21e-06
        [loop_unroll]: 6.34999e-06
        [a_1]: 0.00011353
        [with_stream_mark]: 1.484e-05
        [recompute_prepare]: 6.23998e-06
        [updatestate_depend_eliminate]: 2.94999e-06
        [updatestate_assign_eliminate]: 3.15002e-06
        [updatestate_loads_eliminate]: 2.66e-06
        [parameter_eliminate]: 1.55999e-06
        [specialize_transform]: 5.59e-06
        [updatestate_useless_node_eliminater]: 5.14998e-06
        [accelerated_algorithm]: 5.42999e-06
        [meta_shard_fg_expand]: 9.17001e-06
        [get_grad_eliminate_]: 5.68002e-06
        [merge_forward]: 4.2e-06
        [cell_reuse_recompute_pass]: 3.88001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.76e-05
        [j_node_and_user_rematch]: 9.14998e-06
        [meta_fg_expand]: 2.41e-06
        [replace_old_param]: 8.97e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.39998e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 5.47001e-06
        [cse]: 1.153e-05
        [replace_applicator]: 5.06997e-06
[py_interpret_to_execute_after_opt_a]: 1.636e-05
[rewriter_after_opt_a]: 3.994e-05
[convert_after_rewriter]: 8.02998e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00088471
[jit_opt_b]: 6.654e-05, [1]
    [Cycle 1]: 4.638e-05, [2]
        [frontend_op_eliminate]: 1.952e-05
        [inline_after_opt_a]: 1.531e-05
[cconv]: 2.981e-05
[loop_unroll]: 0.00046601
[jit_opt_after_cconv]: 0.00015109, [1]
    [Cycle 1]: 0.00014381, [11]
        [c_1]: 2.324e-05
        [parameter_eliminate]: 3.74002e-06
        [updatestate_depend_eliminate]: 5.56e-06
        [updatestate_assign_eliminate]: 2.77002e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 2.082e-05
        [call_graph_tuple_transform]: 2.143e-05
        [tuple_list_get_item_eliminator]: 5.37999e-06
        [none_parameter_eliminate]: 1.87999e-06
        [renormalize]: 7.2e-07
        [switch_simplify]: 5.82001e-06
[remove_dup_value]: 1.477e-05
[partial_unused_args_eliminate]: 2.54001e-06
[environ_conv]: 5.57001e-06
[add_recomputation]: 5.166e-05
[cse_after_recomputation]: 2.346e-05, [1]
    [Cycle 1]: 1.748e-05, [1]
        [cse]: 1.103e-05
[auto_monad_reorder]: 1.615e-05
[get_jit_bprop_graph]: 2.69001e-06
[rewriter_after_jit_bprop_graph]: 7.51001e-06
[opt_after_jit_grad]: 0.00050894
[symbol_engine_optimizer]: 8.267e-05, [1]
    [Cycle 1]: 7.57e-05, [6]
        [build]: 4.19997e-06
        [elim_shapecalc]: 8.58001e-06
        [elim_not_effective]: 1.584e-05
        [opt_reshape]: 6.92002e-06
        [fold_const_symbol]: 9.71e-06
        [renormalize]: 6.80011e-07
[validate]: 4.349e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00734268
[execute]: 1.098e-05
Sums
    bootstrap                            :     0.000451s :  1.98%
    type_inference                       :     0.009792s : 42.94%
    event_method                         :     0.000019s :  0.08%
    auto_monad                           :     0.000068s :  0.30%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.18%
    rewriter_before_opt_a                :     0.000070s :  0.31%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.30%
    jit_opt_a.loop_unroll                :     0.000038s :  0.17%
    jit_opt_a.a_1                        :     0.000789s :  3.46%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000012s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.08%
    jit_opt_a.inline_without_move        :     0.000012s :  0.05%
    jit_opt_a.renormalize                :     0.001492s :  6.54%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.09%
    jit_opt_a.cse                        :     0.000036s :  0.16%
    jit_opt_a.replace_applicator         :     0.000022s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.07%
    rewriter_after_opt_a                 :     0.000040s :  0.18%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000885s :  3.88%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.07%
    cconv                                :     0.000030s :  0.13%
    loop_unroll                          :     0.000466s :  2.04%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000052s :  0.23%
    cse_after_recomputation.cse          :     0.000011s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.07%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.03%
    opt_after_jit_grad                   :     0.000509s :  2.23%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.19%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007343s : 32.20%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000280    23
 1.08% :     0.000003s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 2.12% :     0.000006s :      3: substitution.graph_param_transform
83.75% :     0.000235s :      4: substitution.inline
 1.46% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.04% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.86% :     0.000005s :      2: substitution.replace_old_param
 7.20% :     0.000020s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009702     2
87.76% :     0.008515s :      1: type_inference.infer
12.24% :     0.001188s :      1: type_inference.specialize
------[replace.]   0.000092     6
64.48% :     0.000060s :      4: replace.inline
35.52% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000251     6
92.48% :     0.000232s :      4: match.inline
 7.52% :     0.000019s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000147   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.85% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 2.58% :     0.000004s :     10: predicate.arithmetic_simplify
 1.30% :     0.000002s :     10: predicate.cast_eliminate
 0.57% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.83% :     0.000001s :     10: predicate.compare_switch_simplify
 1.44% :     0.000002s :     10: predicate.depend_value_elim
 0.88% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.79% :     0.000003s :     10: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.24% :     0.000000s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.89% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000002s :     10: predicate.environ_get_eliminate
 1.09% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 5.10% :     0.000007s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :      6: predicate.less_batch_normalization
 1.34% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.44% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.82% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.02% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.86% :     0.000001s :     10: predicate.merge_addn
 0.91% :     0.000001s :     10: predicate.minmaximum_grad
 2.15% :     0.000003s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 1.96% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.64% :     0.000004s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     18: predicate.replace_applicator
 0.42% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 2.00% :     0.000003s :     10: predicate.reshape_eliminate
 1.35% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.32% :     0.000002s :      6: predicate.special_op_eliminate
 0.82% :     0.000001s :      6: predicate.specialize_transform
 1.79% :     0.000003s :     10: predicate.split_environ_get_set_with_tuple_value
 1.89% :     0.000003s :     10: predicate.stack_unstack_eliminate
 0.31% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 1.89% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.90% :     0.000010s :     53: predicate.switch_simplify
 1.53% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000002s :     10: predicate.transpose_eliminate
 1.94% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.61% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.00% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 2.18% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.31% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.45% :     0.000004s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000843    12
38.69% :     0.000326s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.11% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
55.20% :     0.000465s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.030752    76
 0.18% :     0.000055s :      1: add_recomputation
 0.23% :     0.000072s :      1: auto_monad
 0.06% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.53% :     0.000472s :      1: bootstrap
 0.11% :     0.000033s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.07% :     0.000023s :      1: event_method
 0.06% :     0.000017s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
24.91% :     0.007660s :      1: jit_opt_a
 0.50% :     0.000154s :      1: jit_opt_after_cconv
 0.23% :     0.000070s :      1: jit_opt_b
 1.54% :     0.000474s :      1: loop_unroll
 2.91% :     0.000895s :      1: mutable_eliminate
 3.34% :     0.001027s :     26: opt.transform.jit_opt_a
 0.17% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000028s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.68% :     0.000517s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.14% :     0.000043s :      1: py_interpret_to_execute
 0.06% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000018s :      1: remove_dup_value
 3.08% :     0.000946s :      1: renormalize.infer
 1.73% :     0.000533s :      1: renormalize.specialize
 0.03% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000043s :      1: rewriter_after_opt_a
 0.24% :     0.000073s :      1: rewriter_before_opt_a
 0.28% :     0.000086s :      1: symbol_engine_optimizer
23.95% :     0.007365s :      1: task_emit
31.90% :     0.009811s :      1: type_inference
 0.21% :     0.000064s :      1: validate


Time group info:
------[substitution.]   0.000245    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000002s :      2: substitution.fold_const_symbol
 2.54% :     0.000006s :      3: substitution.graph_param_transform
85.23% :     0.000209s :      4: substitution.inline
 1.43% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.11% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000005s :      2: substitution.replace_old_param
 5.20% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006810     2
82.88% :     0.005643s :      1: type_inference.infer
17.12% :     0.001166s :      1: type_inference.specialize
------[replace.]   0.000075     6
67.95% :     0.000051s :      4: replace.inline
32.05% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000218     6
94.64% :     0.000206s :      4: match.inline
 5.36% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.71% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.06% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 2.31% :     0.000003s :     10: predicate.arithmetic_simplify
 1.43% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.91% :     0.000002s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.66% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.28% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      3: predicate.elim_not_effective
 0.74% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.00% :     0.000001s :     10: predicate.environ_get_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.43% :     0.000001s :      3: predicate.graph_param_transform
 5.02% :     0.000006s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.00% :     0.000003s :     15: predicate.load_eliminater
 1.57% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.72% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.79% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 2.69% :     0.000003s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.02% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.42% :     0.000001s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.02% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.03% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.52% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.12% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.72% :     0.000010s :     53: predicate.switch_simplify
 1.47% :     0.000002s :     10: predicate.tile_eliminate
 1.36% :     0.000002s :     10: predicate.transpose_eliminate
 1.24% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.41% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.11% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000859    12
50.09% :     0.000430s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.51% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.40% :     0.000381s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.030656    76
 0.18% :     0.000056s :      1: add_recomputation
 0.25% :     0.000076s :      1: auto_monad
 0.06% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.58% :     0.000484s :      1: bootstrap
 0.12% :     0.000037s :      1: cconv
 0.04% :     0.000012s :      1: convert_after_rewriter
 0.09% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.09% :     0.000027s :      1: event_method
 0.04% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
37.93% :     0.011628s :      1: jit_opt_a
 0.56% :     0.000171s :      1: jit_opt_after_cconv
 0.22% :     0.000068s :      1: jit_opt_b
 1.76% :     0.000540s :      1: loop_unroll
 2.40% :     0.000735s :      1: mutable_eliminate
 2.95% :     0.000905s :     26: opt.transform.jit_opt_a
 0.17% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000036s :      4: opt.transform.jit_opt_b
 0.05% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.13% :     0.000654s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000053s :      1: py_interpret_to_execute
 0.05% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 2.17% :     0.000664s :      1: renormalize.infer
 1.47% :     0.000451s :      1: renormalize.specialize
 0.03% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000047s :      1: rewriter_after_opt_a
 0.26% :     0.000080s :      1: rewriter_before_opt_a
 0.28% :     0.000087s :      1: symbol_engine_optimizer
21.09% :     0.006466s :      1: task_emit
22.59% :     0.006926s :      1: type_inference
 0.48% :     0.000149s :      1: validate


TotalTime = 0.051747, [33]
[bootstrap]: 0.0003929
[type_inference]: 0.0053745
[event_method]: 1.893e-05
[auto_monad]: 6.622e-05
[graph_reusing]: 5.55001e-06
[pre_auto_parallel]: 2.57001e-06
[py_interpret_to_execute]: 4.05e-05
[rewriter_before_opt_a]: 7.44e-05
[expand_dump_flag]: 3.08e-06
[jit_opt_a]: 0.0347191, [2]
    [Cycle 1]: 0.00237439, [27]
        [switch_simplify]: 8.112e-05
        [loop_unroll]: 3.159e-05
        [a_1]: 0.00070828
        [with_stream_mark]: 2.843e-05
        [recompute_prepare]: 9.10999e-06
        [updatestate_depend_eliminate]: 4.55001e-06
        [updatestate_assign_eliminate]: 3.18998e-06
        [updatestate_loads_eliminate]: 2.76999e-06
        [parameter_eliminate]: 2.84001e-06
        [specialize_transform]: 6.29999e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 6.04001e-06
        [meta_shard_fg_expand]: 2.79999e-06
        [get_grad_eliminate_]: 6.884e-05
        [merge_forward]: 4.50001e-06
        [cell_reuse_recompute_pass]: 1.66002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.678e-05
        [j_node_and_user_rematch]: 9.47001e-06
        [meta_fg_expand]: 2.74999e-06
        [replace_old_param]: 1.01e-05
        [inline_without_move]: 5.29998e-06
        [renormalize]: 0.00106131
        [add_forward_monad_depend]: 8.73001e-06
        [auto_monad_grad]: 3.58e-06
        [auto_monad_eliminator]: 1.828e-05
        [cse]: 3.54e-05
        [replace_applicator]: 1.896e-05
    [Cycle 2]: 0.00036343, [27]
        [switch_simplify]: 6.66e-06
        [loop_unroll]: 5.39e-06
        [a_1]: 0.00010256
        [with_stream_mark]: 1.478e-05
        [recompute_prepare]: 6.40002e-06
        [updatestate_depend_eliminate]: 3.04999e-06
        [updatestate_assign_eliminate]: 2.58998e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [parameter_eliminate]: 1.59e-06
        [specialize_transform]: 4.92e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.84e-06
        [meta_shard_fg_expand]: 1.74998e-06
        [get_grad_eliminate_]: 5.05999e-06
        [merge_forward]: 2.94001e-06
        [cell_reuse_recompute_pass]: 2.81e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.598e-05
        [j_node_and_user_rematch]: 9.17001e-06
        [meta_fg_expand]: 1.75001e-06
        [replace_old_param]: 7.95998e-06
        [inline_without_move]: 4.74e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 2.09e-06
        [auto_monad_grad]: 1.08001e-06
        [auto_monad_eliminator]: 6.21e-06
        [cse]: 1.142e-05
        [replace_applicator]: 4.99998e-06
[py_interpret_to_execute_after_opt_a]: 1.485e-05
[rewriter_after_opt_a]: 4.041e-05
[convert_after_rewriter]: 7.55e-06
[order_py_execute_after_rewriter]: 6.02999e-06
[mutable_eliminate]: 0.0009287
[jit_opt_b]: 5.849e-05, [1]
    [Cycle 1]: 4.923e-05, [2]
        [frontend_op_eliminate]: 1.904e-05
        [inline_after_opt_a]: 1.67e-05
[cconv]: 3.577e-05
[loop_unroll]: 0.00052266
[jit_opt_after_cconv]: 0.00018306, [1]
    [Cycle 1]: 0.00017458, [11]
        [c_1]: 2.34e-05
        [parameter_eliminate]: 5.02999e-06
        [updatestate_depend_eliminate]: 9.91e-06
        [updatestate_assign_eliminate]: 4.32e-06
        [updatestate_loads_eliminate]: 2.76e-06
        [cse]: 3.334e-05
        [call_graph_tuple_transform]: 2.768e-05
        [tuple_list_get_item_eliminator]: 6.01e-06
        [none_parameter_eliminate]: 1.47999e-06
        [renormalize]: 1.05001e-06
        [switch_simplify]: 6.34001e-06
[remove_dup_value]: 1.568e-05
[partial_unused_args_eliminate]: 2.20002e-06
[environ_conv]: 7.32002e-06
[add_recomputation]: 5.976e-05
[cse_after_recomputation]: 2.248e-05, [1]
    [Cycle 1]: 1.586e-05, [1]
        [cse]: 9.24e-06
[auto_monad_reorder]: 1.592e-05
[get_jit_bprop_graph]: 2.66e-06
[rewriter_after_jit_bprop_graph]: 1.105e-05
[opt_after_jit_grad]: 0.00058308
[symbol_engine_optimizer]: 9.224e-05, [1]
    [Cycle 1]: 8.27e-05, [6]
        [build]: 6.08002e-06
        [elim_shapecalc]: 8.89003e-06
        [elim_not_effective]: 1.81e-05
        [opt_reshape]: 7.00998e-06
        [fold_const_symbol]: 1.095e-05
        [renormalize]: 8.99978e-07
[validate]: 4.875e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00807194
[execute]: 9.76e-06
Sums
    bootstrap                            :     0.000393s :  2.07%
    type_inference                       :     0.005375s : 28.32%
    event_method                         :     0.000019s :  0.10%
    auto_monad                           :     0.000066s :  0.35%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000040s :  0.21%
    rewriter_before_opt_a                :     0.000074s :  0.39%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000088s :  0.46%
    jit_opt_a.loop_unroll                :     0.000037s :  0.19%
    jit_opt_a.a_1                        :     0.000811s :  4.27%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000074s :  0.39%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.001061s :  5.59%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.13%
    jit_opt_a.cse                        :     0.000047s :  0.25%
    jit_opt_a.replace_applicator         :     0.000024s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.08%
    rewriter_after_opt_a                 :     0.000040s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000929s :  4.89%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000036s :  0.19%
    loop_unroll                          :     0.000523s :  2.75%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000033s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000060s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.06%
    opt_after_jit_grad                   :     0.000583s :  3.07%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000049s :  0.26%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008072s : 42.53%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000323    23
 0.98% :     0.000003s :      2: substitution.elim_not_effective
 0.37% :     0.000001s :      2: substitution.fold_const_symbol
 1.82% :     0.000006s :      3: substitution.graph_param_transform
87.52% :     0.000283s :      4: substitution.inline
 1.34% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.77% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.72% :     0.000006s :      2: substitution.replace_old_param
 4.49% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005290     2
81.16% :     0.004293s :      1: type_inference.infer
18.84% :     0.000997s :      1: type_inference.specialize
------[replace.]   0.000088     6
68.58% :     0.000060s :      4: replace.inline
31.42% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000294     6
95.47% :     0.000280s :      4: match.inline
 4.53% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000142   691
 1.19% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.36% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.84% :     0.000001s :     10: predicate.addn_check_dump
 1.55% :     0.000002s :     10: predicate.addn_zero_filter
 3.02% :     0.000004s :     10: predicate.arithmetic_simplify
 1.43% :     0.000002s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.83% :     0.000001s :     10: predicate.compare_switch_simplify
 1.24% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.39% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.12% :     0.000002s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.67% :     0.000008s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      6: predicate.less_batch_normalization
 1.90% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.55% :     0.000002s :     15: predicate.load_eliminater
 1.83% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.85% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.87% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000002s :     10: predicate.minmaximum_grad
 2.82% :     0.000004s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.93% :     0.000003s :     10: predicate.reduce_eliminate
 1.67% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.40% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.91% :     0.000003s :     18: predicate.replace_applicator
 0.45% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.25% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.07% :     0.000002s :     10: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 0.83% :     0.000001s :      6: predicate.specialize_transform
 1.20% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.18% :     0.000003s :     16: predicate.switch_defer_inline
 1.71% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.64% :     0.000011s :     53: predicate.switch_simplify
 1.31% :     0.000002s :     10: predicate.tile_eliminate
 1.29% :     0.000002s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.13% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 2.01% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.19% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.36% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.39% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.06% :     0.000003s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000767    12
47.10% :     0.000361s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.65% :     0.000066s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.25% :     0.000339s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.054025    76
 0.12% :     0.000064s :      1: add_recomputation
 0.13% :     0.000070s :      1: auto_monad
 0.03% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.77% :     0.000417s :      1: bootstrap
 0.07% :     0.000039s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.05% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000011s :      1: environ_conv
 0.04% :     0.000023s :      1: event_method
 0.03% :     0.000016s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000008s :      1: graph_reusing
64.27% :     0.034724s :      1: jit_opt_a
 0.35% :     0.000187s :      1: jit_opt_after_cconv
 0.11% :     0.000061s :      1: jit_opt_b
 0.99% :     0.000534s :      1: loop_unroll
 1.75% :     0.000945s :      1: mutable_eliminate
 2.09% :     0.001127s :     26: opt.transform.jit_opt_a
 0.11% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000028s :      4: opt.transform.jit_opt_b
 0.03% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000041s :      4: opt.transform.symbol_engine_opt
 1.10% :     0.000596s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.08% :     0.000044s :      1: py_interpret_to_execute
 0.03% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000019s :      1: remove_dup_value
 1.08% :     0.000585s :      1: renormalize.infer
 0.86% :     0.000465s :      1: renormalize.specialize
 0.02% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000048s :      1: rewriter_after_opt_a
 0.14% :     0.000077s :      1: rewriter_before_opt_a
 0.18% :     0.000095s :      1: symbol_engine_optimizer
14.98% :     0.008094s :      1: task_emit
 9.98% :     0.005394s :      1: type_inference
 0.14% :     0.000074s :      1: validate


TotalTime = 0.0371695, [30]
[bootstrap]: 0.00038537
[type_inference]: 0.00877308
[event_method]: 2.462e-05
[auto_monad]: 9.268e-05
[graph_reusing]: 6.96001e-06
[pre_auto_parallel]: 2.64999e-06
[py_interpret_to_execute]: 6.621e-05
[rewriter_before_opt_a]: 9.125e-05
[expand_dump_flag]: 3.6e-06
[jit_opt_a]: 0.0236208, [2]
    [Cycle 1]: 0.0163495, [27]
        [switch_simplify]: 0.0124573
        [loop_unroll]: 6.151e-05
        [a_1]: 0.00091612
        [with_stream_mark]: 3.114e-05
        [recompute_prepare]: 9.34998e-06
        [updatestate_depend_eliminate]: 4.33999e-06
        [updatestate_assign_eliminate]: 3.38999e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 2.54001e-06
        [specialize_transform]: 7.05e-06
        [updatestate_useless_node_eliminater]: 5.66003e-06
        [accelerated_algorithm]: 6.06e-06
        [meta_shard_fg_expand]: 2.99999e-06
        [get_grad_eliminate_]: 6.32001e-06
        [merge_forward]: 3.86999e-06
        [cell_reuse_recompute_pass]: 1.34998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.61e-05
        [j_node_and_user_rematch]: 9.92999e-06
        [meta_fg_expand]: 2.53003e-06
        [replace_old_param]: 1.165e-05
        [inline_without_move]: 5.71003e-06
        [renormalize]: 0.0024571
        [add_forward_monad_depend]: 1.04e-05
        [auto_monad_grad]: 2.81999e-06
        [auto_monad_eliminator]: 1.961e-05
        [cse]: 3.276e-05
        [replace_applicator]: 2.206e-05
    [Cycle 2]: 0.00036188, [27]
        [switch_simplify]: 7.03e-06
        [loop_unroll]: 5.82001e-06
        [a_1]: 0.00010494
        [with_stream_mark]: 1.394e-05
        [recompute_prepare]: 5.17e-06
        [updatestate_depend_eliminate]: 3.45e-06
        [updatestate_assign_eliminate]: 2.68998e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 6.72002e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.51998e-06
        [meta_shard_fg_expand]: 1.95001e-06
        [get_grad_eliminate_]: 5.89e-06
        [merge_forward]: 3.22002e-06
        [cell_reuse_recompute_pass]: 3.53999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.93e-05
        [j_node_and_user_rematch]: 9.05999e-06
        [meta_fg_expand]: 1.84998e-06
        [replace_old_param]: 9.29998e-06
        [inline_without_move]: 4.89e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.64998e-06
        [auto_monad_grad]: 6.30011e-07
        [auto_monad_eliminator]: 5.02e-06
        [cse]: 1.084e-05
        [replace_applicator]: 5.30001e-06
[py_interpret_to_execute_after_opt_a]: 1.722e-05
[rewriter_after_opt_a]: 4.072e-05
[convert_after_rewriter]: 7.36001e-06
[order_py_execute_after_rewriter]: 4.94e-06
[mutable_eliminate]: 0.0009008
[jit_opt_b]: 5.759e-05, [1]
    [Cycle 1]: 4.869e-05, [2]
        [frontend_op_eliminate]: 1.88e-05
        [inline_after_opt_a]: 1.695e-05
[cconv]: 0.00044067
[loop_unroll]: 0.00133179
[jit_opt_after_cconv]: 0.00020283, [1]
    [Cycle 1]: 0.00019167, [11]
        [c_1]: 2.945e-05
        [parameter_eliminate]: 7.43e-06
        [updatestate_depend_eliminate]: 1.314e-05
        [updatestate_assign_eliminate]: 3.4e-06
        [updatestate_loads_eliminate]: 3.26999e-06
        [cse]: 3.438e-05
        [call_graph_tuple_transform]: 2.876e-05
        [tuple_list_get_item_eliminator]: 6.88e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 8.60018e-07
        [switch_simplify]: 7.45e-06
[remove_dup_value]: 1.641e-05
[partial_unused_args_eliminate]: 2.49999e-06
[environ_conv]: 6.54999e-06
[add_recomputation]: 5.948e-05
[cse_after_recomputation]: 2.357e-05, [1]
    [Cycle 1]: 1.723e-05, [1]
        [cse]: 1.158e-05
[auto_monad_reorder]: 1.55e-05
[get_jit_bprop_graph]: 3.01999e-06
[rewriter_after_jit_bprop_graph]: 1.263e-05
[opt_after_jit_grad]: 0.00051267
[symbol_engine_optimizer]: 8.271e-05, [1]
    [Cycle 1]: 7.563e-05, [6]
        [build]: 5.35001e-06
        [elim_shapecalc]: 8.64e-06
        [elim_not_effective]: 1.376e-05
        [opt_reshape]: 6.64001e-06
        [fold_const_symbol]: 9.29e-06
        [renormalize]: 6.30011e-07
[validate]: 4.813e-05
Sums
    bootstrap                            :     0.000385s :  1.31%
    type_inference                       :     0.008773s : 29.79%
    event_method                         :     0.000025s :  0.08%
    auto_monad                           :     0.000093s :  0.31%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000066s :  0.22%
    rewriter_before_opt_a                :     0.000091s :  0.31%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.012464s : 42.32%
    jit_opt_a.loop_unroll                :     0.000067s :  0.23%
    jit_opt_a.a_1                        :     0.001021s :  3.47%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000014s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000021s :  0.07%
    jit_opt_a.inline_without_move        :     0.000011s :  0.04%
    jit_opt_a.renormalize                :     0.002457s :  8.34%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.08%
    jit_opt_a.cse                        :     0.000044s :  0.15%
    jit_opt_a.replace_applicator         :     0.000027s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.06%
    rewriter_after_opt_a                 :     0.000041s :  0.14%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000901s :  3.06%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.06%
    cconv                                :     0.000441s :  1.50%
    loop_unroll                          :     0.001332s :  4.52%
    jit_opt_after_cconv.c_1              :     0.000029s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000013s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000034s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000029s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000059s :  0.20%
    cse_after_recomputation.cse          :     0.000012s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000013s :  0.04%
    opt_after_jit_grad                   :     0.000513s :  1.74%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000048s :  0.16%


Time group info:
------[substitution.]   0.000432    26
 0.41% :     0.000002s :      2: substitution.elim_not_effective
 0.42% :     0.000002s :      2: substitution.fold_const_symbol
 1.50% :     0.000006s :      3: substitution.graph_param_transform
87.95% :     0.000380s :      6: substitution.inline
 1.02% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.37% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.74% :     0.000008s :      2: substitution.replace_old_param
 2.06% :     0.000009s :      1: substitution.switch_simplify
 3.53% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008040     2
82.11% :     0.006602s :      1: type_inference.infer
17.89% :     0.001438s :      1: type_inference.specialize
------[replace.]   0.012435     9
 0.71% :     0.000088s :      6: replace.inline
98.98% :     0.012308s :      1: replace.switch_simplify
 0.31% :     0.000039s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000399     9
94.42% :     0.000377s :      6: match.inline
 2.07% :     0.000008s :      1: match.switch_simplify
 3.51% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000185   801
 1.41% :     0.000003s :     12: predicate.accumulaten_eliminater
 0.92% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.81% :     0.000002s :     12: predicate.addn_check_dump
 1.26% :     0.000002s :     12: predicate.addn_zero_filter
 2.46% :     0.000005s :     12: predicate.arithmetic_simplify
 1.81% :     0.000003s :     12: predicate.cast_eliminate
 0.33% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.80% :     0.000001s :     12: predicate.compare_switch_simplify
 1.26% :     0.000002s :     12: predicate.depend_value_elim
 1.00% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.14% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.60% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.21% :     0.000000s :      3: predicate.elim_not_effective
 0.45% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.03% :     0.000002s :     12: predicate.environ_add_const_eliminate
 0.89% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.03% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     12: predicate.environ_get_eliminate
 0.94% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.14% :     0.000000s :      3: predicate.fold_const_symbol
 1.12% :     0.000002s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.53% :     0.000010s :     26: predicate.inline
 0.58% :     0.000001s :      6: predicate.inline_without_move
 0.24% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.04% :     0.000002s :      6: predicate.less_batch_normalization
 1.63% :     0.000003s :     14: predicate.list_to_tuple_eliminator_
 1.54% :     0.000003s :     17: predicate.load_eliminater
 2.21% :     0.000004s :      3: predicate.loop_unroll_after_grad
 7.78% :     0.000014s :     40: predicate.loop_unroll_before_grad
 2.02% :     0.000004s :     15: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000002s :     12: predicate.merge_addn
 0.88% :     0.000002s :     12: predicate.minmaximum_grad
 1.93% :     0.000004s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 1.62% :     0.000003s :     17: predicate.partial_eliminate
 1.37% :     0.000003s :     12: predicate.print_const_string_wrapper
 1.57% :     0.000003s :     12: predicate.reduce_eliminate
 1.63% :     0.000003s :     14: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.37% :     0.000003s :     20: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000001s :      3: predicate.reset_defer_inline
 1.47% :     0.000003s :     12: predicate.reshape_eliminate
 1.09% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.62% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000002s :     12: predicate.same_eliminate
 0.33% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.70% :     0.000001s :      6: predicate.special_op_eliminate
 0.77% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000003s :     12: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.34% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 1.98% :     0.000004s :     20: predicate.switch_defer_inline
 1.85% :     0.000003s :     20: predicate.switch_layer_defer_inline
10.97% :     0.000020s :     65: predicate.switch_simplify
 1.31% :     0.000002s :     12: predicate.tile_eliminate
 1.18% :     0.000002s :     12: predicate.transpose_eliminate
 1.49% :     0.000003s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000007s :     20: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000003s :     12: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000003s :     14: predicate.tuple_to_list_eliminator_
 1.26% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.04% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.92% :     0.000004s :     12: predicate.value_based_eliminate
 0.21% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001090    15
44.27% :     0.000483s :      6: func_graph_cloner_run.FuncGraphClonerGraph
14.48% :     0.000158s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.25% :     0.000450s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.053381    72
 0.12% :     0.000063s :      1: add_recomputation
 0.18% :     0.000097s :      1: auto_monad
 0.03% :     0.000018s :      1: auto_monad_reorder
 0.76% :     0.000408s :      1: bootstrap
 1.00% :     0.000533s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.05% :     0.000026s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.06% :     0.000030s :      1: event_method
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
44.26% :     0.023626s :      1: jit_opt_a
 0.39% :     0.000207s :      1: jit_opt_after_cconv
 0.11% :     0.000060s :      1: jit_opt_b
 2.52% :     0.001348s :      1: loop_unroll
 1.71% :     0.000913s :      1: mutable_eliminate
25.64% :     0.013689s :     26: opt.transform.jit_opt_a
 0.13% :     0.000068s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000025s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000035s :      4: opt.transform.symbol_engine_opt
 0.98% :     0.000521s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000069s :      1: py_interpret_to_execute
 0.04% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000019s :      1: remove_dup_value
 1.23% :     0.000656s :      1: renormalize.infer
 3.35% :     0.001789s :      1: renormalize.specialize
 0.03% :     0.000015s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000044s :      1: rewriter_after_opt_a
 0.18% :     0.000095s :      1: rewriter_before_opt_a
 0.16% :     0.000085s :      1: symbol_engine_optimizer
16.48% :     0.008797s :      1: type_inference


TotalTime = 0.0331915, [33]
[bootstrap]: 0.00044492
[type_inference]: 0.00776031
[event_method]: 2.311e-05
[auto_monad]: 7.571e-05
[graph_reusing]: 7.06999e-06
[pre_auto_parallel]: 4.22e-06
[py_interpret_to_execute]: 5.508e-05
[rewriter_before_opt_a]: 8.053e-05
[expand_dump_flag]: 3.58999e-06
[jit_opt_a]: 0.0104082, [2]
    [Cycle 1]: 0.00531413, [27]
        [switch_simplify]: 6.872e-05
        [loop_unroll]: 3.037e-05
        [a_1]: 0.00350127
        [with_stream_mark]: 3.754e-05
        [recompute_prepare]: 1.544e-05
        [updatestate_depend_eliminate]: 5.59e-06
        [updatestate_assign_eliminate]: 3.11001e-06
        [updatestate_loads_eliminate]: 2.83e-06
        [parameter_eliminate]: 3.09999e-06
        [specialize_transform]: 7.03e-06
        [updatestate_useless_node_eliminater]: 6.48003e-06
        [accelerated_algorithm]: 7.20998e-06
        [meta_shard_fg_expand]: 3.32997e-06
        [get_grad_eliminate_]: 6.13998e-06
        [merge_forward]: 4.15e-06
        [cell_reuse_recompute_pass]: 4.57e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.603e-05
        [j_node_and_user_rematch]: 1.199e-05
        [meta_fg_expand]: 3.13998e-06
        [replace_old_param]: 1.365e-05
        [inline_without_move]: 5.64e-06
        [renormalize]: 0.00120762
        [add_forward_monad_depend]: 1.084e-05
        [auto_monad_grad]: 2.89001e-06
        [auto_monad_eliminator]: 2.264e-05
        [cse]: 3.61e-05
        [replace_applicator]: 2.258e-05
    [Cycle 2]: 0.00040175, [27]
        [switch_simplify]: 6.76999e-06
        [loop_unroll]: 6.48998e-06
        [a_1]: 0.00010888
        [with_stream_mark]: 1.775e-05
        [recompute_prepare]: 7.07002e-06
        [updatestate_depend_eliminate]: 4.23001e-06
        [updatestate_assign_eliminate]: 3.36001e-06
        [updatestate_loads_eliminate]: 2.84001e-06
        [parameter_eliminate]: 1.83002e-06
        [specialize_transform]: 5.89e-06
        [updatestate_useless_node_eliminater]: 5.47001e-06
        [accelerated_algorithm]: 6.74999e-06
        [meta_shard_fg_expand]: 2.14999e-06
        [get_grad_eliminate_]: 4.68999e-06
        [merge_forward]: 3.93001e-06
        [cell_reuse_recompute_pass]: 3.7e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.062e-05
        [j_node_and_user_rematch]: 9.62999e-06
        [meta_fg_expand]: 1.99e-06
        [replace_old_param]: 9.58997e-06
        [inline_without_move]: 5.39e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 2.01998e-06
        [auto_monad_grad]: 1.44e-06
        [auto_monad_eliminator]: 7.03e-06
        [cse]: 1.443e-05
        [replace_applicator]: 6.26e-06
[py_interpret_to_execute_after_opt_a]: 1.838e-05
[rewriter_after_opt_a]: 4.11e-05
[convert_after_rewriter]: 7.42002e-06
[order_py_execute_after_rewriter]: 6.16998e-06
[mutable_eliminate]: 0.00090784
[jit_opt_b]: 7.189e-05, [1]
    [Cycle 1]: 6.115e-05, [2]
        [frontend_op_eliminate]: 2.512e-05
        [inline_after_opt_a]: 2.237e-05
[cconv]: 2.934e-05
[loop_unroll]: 0.00057387
[jit_opt_after_cconv]: 0.00018376, [1]
    [Cycle 1]: 0.0001751, [11]
        [c_1]: 2.522e-05
        [parameter_eliminate]: 4.02e-06
        [updatestate_depend_eliminate]: 8.45001e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [cse]: 3.559e-05
        [call_graph_tuple_transform]: 2.55e-05
        [tuple_list_get_item_eliminator]: 6.29999e-06
        [none_parameter_eliminate]: 1.60001e-06
        [renormalize]: 1.04e-06
        [switch_simplify]: 7.08998e-06
[remove_dup_value]: 1.758e-05
[partial_unused_args_eliminate]: 2.24001e-06
[environ_conv]: 6.64001e-06
[add_recomputation]: 5.779e-05
[cse_after_recomputation]: 2.527e-05, [1]
    [Cycle 1]: 1.891e-05, [1]
        [cse]: 1.212e-05
[auto_monad_reorder]: 1.854e-05
[get_jit_bprop_graph]: 2.54001e-06
[rewriter_after_jit_bprop_graph]: 1.068e-05
[opt_after_jit_grad]: 0.00064933
[symbol_engine_optimizer]: 8.412e-05, [1]
    [Cycle 1]: 7.617e-05, [6]
        [build]: 5.12e-06
        [elim_shapecalc]: 9.04998e-06
        [elim_
TotalTime = 0.0294084, [33]
[bootstrap]: 0.00040015
[type_inference]: 0.00839759
[event_method]: 2.454e-05
[auto_monad]: 7.57e-05
[graph_reusing]: 6.29999e-06
[pre_auto_parallel]: 2.63e-06
[py_interpret_to_execute]: 5.674e-05
[rewriter_before_opt_a]: 8.128e-05
[expand_dump_flag]: 3.43e-06
[jit_opt_a]: 0.00907767, [2]
    [Cycle 1]: 0.00242686, [27]
        [switch_simplify]: 6.406e-05
        [loop_unroll]: 3.259e-05
        [a_1]: 0.00067434
        [with_stream_mark]: 2.278e-05
        [recompute_prepare]: 1.035e-05
        [updatestate_depend_eliminate]: 4.19002e-06
        [updatestate_assign_eliminate]: 3.06999e-06
        [updatestate_loads_eliminate]: 2.66999e-06
        [parameter_eliminate]: 2.06e-06
        [specialize_transform]: 7.45998e-06
        [updatestate_useless_node_eliminater]: 6.01e-06
        [accelerated_algorithm]: 7.86001e-06
        [meta_shard_fg_expand]: 2.88998e-06
        [get_grad_eliminate_]: 7.12002e-06
        [merge_forward]: 3.90998e-06
        [cell_reuse_recompute_pass]: 3.41001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.449e-05
        [j_node_and_user_rematch]: 1.073e-05
        [meta_fg_expand]: 2.24999e-06
        [replace_old_param]: 1.028e-05
        [inline_without_move]: 6.11998e-06
        [renormalize]: 0.00124601
        [add_forward_monad_depend]: 9.96e-06
        [auto_monad_grad]: 2.54001e-06
        [auto_monad_eliminator]: 1.834e-05
        [cse]: 2.179e-05
        [replace_applicator]: 2.107e-05
    [Cycle 2]: 0.00048095, [27]
        [switch_simplify]: 6.44001e-06
        [loop_unroll]: 5.80002e-06
        [a_1]: 0.00011675
        [with_stream_mark]: 1.648e-05
        [recompute_prepare]: 5.46e-06
        [updatestate_depend_eliminate]: 2.48e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 1.53002e-06
        [specialize_transform]: 5.61998e-06
        [updatestate_useless_node_eliminater]: 5.67999e-06
        [accelerated_algorithm]: 7.91001e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 6.14999e-06
        [merge_forward]: 3.77998e-06
        [cell_reuse_recompute_pass]: 4.05e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.78e-05
        [j_node_and_user_rematch]: 9.86e-06
        [meta_fg_expand]: 2.00002e-06
        [replace_old_param]: 9.46e-06
        [inline_without_move]: 6.34001e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 3.5e-06
        [auto_monad_grad]: 1.80001e-06
        [auto_monad_eliminator]: 8.16002e-06
        [cse]: 1.39e-05
        [replace_applicator]: 5.66998e-06
[py_interpret_to_execute_after_opt_a]: 1.772e-05
[rewriter_after_opt_a]: 4.229e-05
[convert_after_rewriter]: 1.34e-05
[order_py_execute_after_rewriter]: 6.56999e-06
[mutable_eliminate]: 0.00088262
[jit_opt_b]: 6.668e-05, [1]
    [Cycle 1]: 5.728e-05, [2]
        [frontend_op_eliminate]: 2.318e-05
        [inline_after_opt_a]: 2.008e-05
[cconv]: 3.201e-05
[loop_unroll]: 0.00057328
[jit_opt_after_cconv]: 0.00018783, [1]
    [Cycle 1]: 0.00017888, [11]
        [c_1]: 2.504e-05
        [parameter_eliminate]: 5.09e-06
        [updatestate_depend_eliminate]: 7.97e-06
        [updatestate_assign_eliminate]: 3.26999e-06
        [updatestate_loads_eliminate]: 3.11001e-06
        [cse]: 3.584e-05
        [call_graph_tuple_transform]: 2.674e-05
        [tuple_list_get_item_eliminator]: 5.99e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 1.17e-06
        [switch_simplify]: 6.19999e-06
[remove_dup_value]: 1.442e-05
[partial_unused_args_eliminate]: 2.37001e-06
[environ_conv]: 5.77001e-06
[add_recomputation]: 5.628e-05
[cse_after_recomputation]: 2.427e-05, [1]
    [Cycle 1]: 1.705e-05, [1]
        [cse]: 1.002e-05
[auto_monad_reorder]: 1.777e-05
[get_jit_bprop_graph]: 2.76999e-06
[rewriter_after_jit_bprop_graph]: 1.073e-05
[opt_after_jit_grad]: 0.00057998
[symbol_engine_optimizer]: 8.856e-05, [1]
    [Cycle 1]: 7.934e-05, [6]
        [build]: 4.82e-06
        [elim_shapecalc]: 8.85999e-06
        [elimnot_effective]: 1.504e-05
        [opt_reshape]: 7.13998e-06
        [fold_const_symbol]: 9.27001e-06
        [renormalize]: 1.07e-06
[validate]: 4.468e-05
[backend_pass]: 1.10001e-06
[task_emit]: 0.0112457
[execute]: 1.035e-05
Sums
    bootstrap                            :     0.000445s :  1.61%
    type_inference                       :     0.007760s : 28.05%
    event_method                         :     0.000023s :  0.08%
    auto_monad                           :     0.000076s :  0.27%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000004s :  0.02%
    py_interpret_to_execute              :     0.000055s :  0.20%
    rewriter_before_opt_a                :     0.000081s :  0.29%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000075s :  0.27%
    jit_opt_a.loop_unroll                :     0.000037s :  0.13%
    jit_opt_a.a_1                        :     0.003610s : 13.05%
    jit_opt_a.with_stream_mark           :     0.000055s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000023s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000014s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.04%
    jit_opt_a.merge_forward              :     0.000008s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000008s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000022s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000023s :  0.08%
    jit_opt_a.inline_without_move        :     0.000011s :  0.04%
    jit_opt_a.renormalize                :     0.001208s :  4.37%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000030s :  0.11%
    jit_opt_a.cse                        :     0.000051s :  0.18%
    jit_opt_a.replace_applicator         :     0.000029s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.07%
    rewriter_after_opt_a                 :     0.000041s :  0.15%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000908s :  3.28%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000022s :  0.08%
    cconv                                :     0.000029s :  0.11%
    loop_unroll                          :     0.000574s :  2.07%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000036s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000018s :  0.06%
  _not_effective]: 1.612e-05
        [opt_reshape]: 6.31e-06
        [fold_const_symbol]: 1.119e-05
        [renormalize]: 6.79982e-07
[validate]: 4.615e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00833545
[execute]: 1e-05
Sums
    bootstrap                            :     0.000400s :  1.79%
    type_inference                       :     0.008398s : 37.47%
    event_method                         :     0.000025s :  0.11%
    auto_monad                           :     0.000076s :  0.34%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000057s :  0.25%
    rewriter_before_opt_a                :     0.000081s :  0.36%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000070s :  0.31%
    jit_opt_a.loop_unroll                :     0.000038s :  0.17%
    jit_opt_a.a_1                        :     0.000791s :  3.53%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000013s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.09%
    jit_opt_a.inline_without_move        :     0.000012s :  0.06%
    jit_opt_a.renormalize                :     0.001246s :  5.56%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.12%
    jit_opt_a.cse                        :     0.000036s :  0.16%
    jit_opt_a.replace_applicator         :     0.000027s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.08%
    rewriter_after_opt_a                 :     0.000042s :  0.19%
    convert_after_rewriter               :     0.000013s :  0.06%
    order_py_execute_after_rewriter      :     0.000007s :  0.03%
    mutable_eliminate                    :     0.000883s :  3.94%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.09%
    cconv                                :     0.000032s :  0.14%
    loop_unroll                          :     0.000573s :  2.56%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000036s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.06%
    pa  partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000058s :  0.21%
    cse_after_recomputation.cse          :     0.000012s :  0.04%
    auto_monad_reorder                   :     0.000019s :  0.07%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.04%
    opt_after_jit_grad                   :     0.000649s :  2.35%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011246s : 40.65%
    execute                              :     0.000010s :  0.04%

rtial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000056s :  0.25%
    cse_after_recomputation.cse          :     0.000010s :  0.04%
    auto_monad_reorder                   :     0.000018s :  0.08%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.05%
    opt_after_jit_grad                   :     0.000580s :  2.59%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.21%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008335s : 37.19%
    execute                              :     0.000010s :  0.04%


Time group info:
------[substitution.]   0.000278    23
 0.96% :     0.000003s :      2: substitution.elim_not_effective
 0.58% :     0.000002s :      2: substitution.fold_const_symbol
 2.49% :     0.000007s :      3: substitution.graph_param_transform
84.65% :     0.000235s :      4: substitution.inline
 1.32% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.12% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.25% :     0.000006s :      2: substitution.replace_old_param
 5.64% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008297     2
85.35% :     0.007081s :      1: type_inference.infer
14.65% :     0.001216s :      1: type_inference.specialize
------[replace.]   0.000091     6
66.49% :     0.000060s :      4: replace.inline
33.51% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000247     6
94.19% :     0.000232s :      4: match.inline
 5.81% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000151   691
 1.54% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.83% :     0.000001s :     10: predicate.addn_check_dump
 1.77% :     0.000003s :     10: predicate.addn_zero_filter
 2.95% :     0.000004s :     10: predicate.arithmetic_simplify
 1.67% :     0.000003s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.29% :     0.000002s :     10: predicate.compare_switch_simplify
 1.83% :     0.000003s :     10: predicate.depend_value_elim
 1.13% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.74% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.89% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.62% :     0.000002s :     10: predicate.environ_get_eliminate
 0.85% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.71% :     0.000003s :      6: predicate.get_grad_eliminate
 0.19% :     0.000000s :      3: predicate.graph_param_transform
 4.76% :     0.000007s :     22: predicate.inline
 0.75% :     0.000001s :      6: predicate.inline_without_move
 0.29% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.56% :     0.000002s :      6: predicate.less_batch_normalization
 1.22% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.41% :     0.000002s :     15: predicate.load_eliminater
 1.85% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.72% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.09% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.85% :     0.000001s :     10: predicate.merge_addn
 0.89% :     0.000001s :     10: predicate.minmaximum_grad
 2.27% :     0.000003s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 1.76% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000003s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000003s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.16% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.48% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.18% :     0.000002s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.61% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.52% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.91% :     0.000004s :     16: predicate.switch_defer_inline
 1.85% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.68% :     0.000010s :     53: predicate.switch_simplify
 1.18% :     0.000002s :     10: predicate.tile_eliminate
 1.32% :     0.000002s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.45% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.31% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.26% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000856    12
45.96% :     0.000393s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.12% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.92% :     0.000419s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.031826    76
 0.19% :     0.000060s :      1: add_recomputation
 0.25% :     0.000079s :      1: auto_monad
 0.06% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.32% :     0.000420s :      1: bootstrap
 0.11% :     0.000035s :      1: cconv
 0.05% :     0.000016s :      1: convert_after_rewriter
 0.08% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.09% :     0.000029s :      1: event_method
 0.05% :     0.000016s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
28.54% :     0.009083s :      1: jit_opt_a
 0.60% :     0.000191s :      1: jit_opt_after_cconv
 0.22% :     0.000070s :      1: jit_opt_b
 1.84% :     0.000584s :      1: loop_unroll
 2.81% :     0.000896s :      1: mutable_eliminate
 3.29% :     0.001048s :     26: opt.transform.jit_opt_a
 0.19% :     0.000060s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000034s :      4: opt.transform.jit_opt_b
 0.05% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.86% :     0.000591s :      1: opt_after_jit_grad
 0.03% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000060s :      1: py_interpret_to_execute
 0.06% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000017s :      1: remove_dup_value
 2.24% :     0.000713s :      1: renormalize.infer
 1.64% :     0.000522s :      1: renormalize.specialize
 0.04% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000045s :      1: rewriter_after_opt_a
 0.27% :     0.000086s :      1: rewriter_before_opt_a
 0.29% :     0.000091s :      1: symbol_engine_optimizer
26.26% :     0.008357s :      1: task_emit
26.45% :     0.008420s :      1: type_inference
 0.22% :     0.000069s :      1: validate


Time group info:
------[substitution.]   0.000296    23
 0.71% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000002s :      2: substitution.fold_const_symbol
 2.26% :     0.000007s :      3: substitution.graph_param_transform
80.05% :     0.000237s :      4: substitution.inline
 1.92% :     0.000006s :      4: substitution.j_node_and_user_rematch
 2.22% :     0.000007s :      4: substitution.remove_not_recompute_node
 3.13% :     0.000009s :      2: substitution.replace_old_param
 9.12% :     0.000027s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007658     2
83.91% :     0.006426s :      1: type_inference.infer
16.09% :     0.001232s :      1: type_inference.specialize
------[replace.]   0.000170     6
34.35% :     0.000058s :      4: replace.inline
65.65% :     0.000112s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000258     6
90.94% :     0.000234s :      4: match.inline
 9.06% :     0.000023s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000168   691
 1.55% :     0.000003s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.84% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 3.15% :     0.000005s :     10: predicate.arithmetic_simplify
 1.45% :     0.000002s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.86% :     0.000001s :     10: predicate.compare_switch_simplify
 1.99% :     0.000003s :     10: predicate.depend_value_elim
 0.99% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.66% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.24% :     0.000000s :      3: predicate.elim_not_effective
 0.46% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.73% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     10: predicate.environ_get_eliminate
 0.85% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.15% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000002s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 6.00% :     0.000010s :     22: predicate.inline
 0.63% :     0.000001s :      6: predicate.inline_without_move
 0.29% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.41% :     0.000002s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.42% :     0.000002s :     15: predicate.load_eliminater
 1.18% :     0.000002s :      3: predicate.loop_unroll_after_grad
 2.96% :     0.000005s :     34: predicate.loop_unroll_before_grad
 4.13% :     0.000007s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000002s :     10: predicate.merge_addn
 1.54% :     0.000003s :     10: predicate.minmaximum_grad
 1.61% :     0.000003s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.00% :     0.000003s :     15: predicate.partial_eliminate
 0.85% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.73% :     0.000003s :     10: predicate.reduce_eliminate
 1.12% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.38% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.42% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.42% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.80% :     0.000001s :     10: predicate.same_eliminate
 0.39% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000002s :      6: predicate.special_op_eliminate
 0.71% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.41% :     0.000006s :     16: predicate.switch_defer_inline
 1.44% :     0.000002s :     16: predicate.switch_layer_defer_inline
 5.42% :     0.000009s :     53: predicate.switch_simplify
 1.88% :     0.000003s :     10: predicate.tile_eliminate
 1.51% :     0.000003s :     10: predicate.transpose_eliminate
 1.80% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.26% :     0.000007s :     18: predicate.tuple_list_get_item_eliminator
 4.99% :     0.000008s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.17% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.15% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.16% :     0.000004s :     10: predicate.value_based_eliminate
 0.22% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000969    12
46.15% :     0.000447s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.64% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.21% :     0.000467s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038401    76
 0.16% :     0.000062s :      1: add_recomputation
 0.21% :     0.000080s :      1: auto_monad
 0.06% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 1.22% :     0.000467s :      1: bootstrap
 0.08% :     0.000032s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000028s :      1: cse_after_recomputation
 0.03% :     0.000011s :      1: environ_conv
 0.07% :     0.000029s :      1: event_method
 0.04% :     0.000016s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
27.12% :     0.010413s :      1: jit_opt_a
 0.49% :     0.000187s :      1: jit_opt_after_cconv
 0.19% :     0.000075s :      1: jit_opt_b
 1.52% :     0.000584s :      1: loop_unroll
 2.39% :     0.000919s :      1: mutable_eliminate
10.10% :     0.003877s :     26: opt.transform.jit_opt_a
 0.15% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000037s :      4: opt.transform.jit_opt_b
 0.05% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.71% :     0.000659s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000006s :      1: pre_auto_parallel
 0.15% :     0.000058s :      1: py_interpret_to_execute
 0.06% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000021s :      1: remove_dup_value
 1.62% :     0.000621s :      1: renormalize.infer
 1.49% :     0.000573s :      1: renormalize.specialize
 0.03% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000045s :      1: rewriter_after_opt_a
 0.22% :     0.000084s :      1: rewriter_before_opt_a
 0.23% :     0.000087s :      1: symbol_engine_optimizer
29.34% :     0.011267s :      1: task_emit
20.27% :     0.007785s :      1: type_inference
 0.30% :     0.000114s :      1: validate


TotalTime = 0.0200918, [30]
[bootstrap]: 0.00031997
[type_inference]: 0.00825257
[event_method]: 2.337e-05
[auto_monad]: 8.426e-05
[graph_reusing]: 6.26e-06
[pre_auto_parallel]: 2.76999e-06
[py_interpret_to_execute]: 5.641e-05
[rewriter_before_opt_a]: 8.932e-05
[expand_dump_flag]: 3.02002e-06
[jit_opt_a]: 0.00739867, [2]
    [Cycle 1]: 0.00322626, [27]
        [switch_simplify]: 0.00014249
        [loop_unroll]: 3.386e-05
        [a_1]: 0.00166794
        [with_stream_mark]: 2.776e-05
        [recompute_prepare]: 1.014e-05
        [updatestate_depend_eliminate]: 3.9e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [parameter_eliminate]: 2.26e-06
        [specialize_transform]: 6.74999e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 6.26998e-06
        [meta_shard_fg_expand]: 2.61999e-06
        [get_grad_eliminate_]: 5.93998e-06
        [merge_forward]: 3.93001e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.547e-05
        [j_node_and_user_rematch]: 9.93002e-06
        [meta_fg_expand]: 2.75002e-06
        [replace_old_param]: 9.64999e-06
        [inline_without_move]: 6.17001e-06
        [renormalize]: 0.0010081
        [add_forward_monad_depend]: 5.96e-06
        [auto_monad_grad]: 1.66e-06
        [auto_monad_eliminator]: 1.021e-05
        [cse]: 1.738e-05
        [replace_applicator]: 1.257e-05
    [Cycle 2]: 0.00031515, [27]
        [switch_simplify]: 6.19001e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 9.579e-05
        [with_stream_mark]: 1.01e-05
        [recompute_prepare]: 5.05999e-06
        [updatestate_depend_eliminate]: 2.34001e-06
        [updatestate_assign_eliminate]: 1.93002e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 4.95001e-06
        [updatestate_useless_node_eliminater]: 4.48999e-06
        [accelerated_algorithm]: 4.95001e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 4.47998e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 1.65001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.191e-05
        [j_node_and_user_rematch]: 7.75e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 6.71999e-06
        [inline_without_move]: 4.90001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.18001e-06
        [auto_monad_grad]: 6.59988e-07
        [auto_monad_eliminator]: 4.35e-06
        [cse]: 8.95001e-06
        [replace_applicator]: 5.07e-06
[py_interpret_to_execute_after_opt_a]: 9.36e-06
[rewriter_after_opt_a]: 2.697e-05
[convert_after_rewriter]: 5.98002e-06
[order_py_execute_after_rewriter]: 3.83999e-06
[mutable_eliminate]: 0.00070502
[jit_opt_b]: 5.193e-05, [1]
    [Cycle 1]: 4.521e-05, [2]
        [frontend_op_eliminate]: 1.766e-05
        [inline_after_opt_a]: 1.492e-05
[cconv]: 1.869e-05
[loop_unroll]: 0.00190798
[jit_opt_after_cconv]: 0.00016145, [1]
    [Cycle 1]: 0.00015361, [11]
        [c_1]: 2.537e-05
        [parameter_eliminate]: 3.6e-06
        [updatestate_depend_eliminate]: 6.94999e-06
        [updatestate_assign_eliminate]: 2.64001e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [cse]: 2.347e-05
        [call_graph_tuple_transform]: 2.171e-05
        [tuple_list_get_item_eliminator]: 6.03002e-06
        [none_parameter_eliminate]: 9.40025e-07
        [renormalize]: 4.39992e-07
        [switch_simplify]: 6.17001e-06
[remove_dup_value]: 1.177e-05
[partial_unused_args_eliminate]: 1.94e-06
[environ_conv]: 5.12999e-06
[add_recomputation]: 4.117e-05
[cse_after_recomputation]: 2.386e-05, [1]
    [Cycle 1]: 1.771e-05, [1]
        [cse]: 1.09e-05
[auto_monad_reorder]: 1.171e-05
[get_jit_bprop_graph]: 1.99e-06
[rewriter_after_jit_bprop_graph]: 8.54e-06
[opt_after_jit_grad]: 0.00054067
[symbol_engine_optimizer]: 7.828e-05, [1]
    [Cycle 1]: 7.181e-05, [6]
        [build]: 3.71001e-06
        [elim_shapecalc]: 8.64e-06
        [elim_not_effective]: 1.383e-05
        [opt_reshape]: 6.06e-06
        [fold_const_symbol]: 8.64e-06
        [renormalize]: 5.09986e-07
[validate]: 3.72e-05
Sums
    bootstrap                            :     0.000320s :  2.05%
    type_inference                       :     0.008253s : 52.93%
    event_method                         :     0.000023s :  0.15%
    auto_monad                           :     0.000084s :  0.54%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000056s :  0.36%
    rewriter_before_opt_a                :     0.000089s :  0.57%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000149s :  0.95%
    jit_opt_a.loop_unroll                :     0.000039s :  0.25%
    jit_opt_a.a_1                        :     0.001764s : 11.31%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.001008s :  6.47%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.09%
    jit_opt_a.cse                        :     0.000026s :  0.17%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000027s :  0.17%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000705s :  4.52%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000019s :  0.12%
    loop_unroll                          :     0.001908s : 12.24%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000012s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000041s :  0.26%
    cse_after_recomputation.cse          :     0.000011s :  0.07%
    auto_monad_reorder                   :     0.000012s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000541s :  3.47%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.24%


Time group info:
------[substitution.]   0.000396    26
 0.54% :     0.000002s :      2: substitution.elim_not_effective
 0.26% :     0.000001s :      2: substitution.fold_const_symbol
 1.39% :     0.000006s :      3: substitution.graph_param_transform
85.06% :     0.000337s :      6: substitution.inline
 0.87% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.08% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.15% :     0.000005s :      2: substitution.replace_old_param
 2.48% :     0.000010s :      1: substitution.switch_simplify
 7.18% :     0.000028s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008172     2
84.25% :     0.006885s :      1: type_inference.infer
15.75% :     0.001287s :      1: type_inference.specialize
------[replace.]   0.000182     9
52.57% :     0.000096s :      6: replace.inline
24.35% :     0.000044s :      1: replace.switch_simplify
23.08% :     0.000042s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000369     9
90.23% :     0.000333s :      6: match.inline
 2.41% :     0.000009s :      1: match.switch_simplify
 7.35% :     0.000027s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000164   801
 1.27% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.08% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.91% :     0.000001s :     12: predicate.addn_check_dump
 1.30% :     0.000002s :     12: predicate.addn_zero_filter
 2.60% :     0.000004s :     12: predicate.arithmetic_simplify
 1.02% :     0.000002s :     12: predicate.cast_eliminate
 0.34% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.92% :     0.000002s :     12: predicate.compare_switch_simplify
 1.16% :     0.000002s :     12: predicate.depend_value_elim
 1.05% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.06% :     0.000002s :     12: predicate.dict_get_item_eliminator
 0.98% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.64% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.24% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     12: predicate.environ_add_const_eliminate
 0.93% :     0.000002s :     12: predicate.environ_get_add_eliminate
 0.89% :     0.000001s :     12: predicate.environ_get_depend_swap
 1.08% :     0.000002s :     12: predicate.environ_get_eliminate
 0.91% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.15% :     0.000000s :      3: predicate.fold_const_symbol
 0.78% :     0.000001s :      6: predicate.get_grad_eliminate
 0.18% :     0.000000s :      3: predicate.graph_param_transform
11.86% :     0.000019s :     26: predicate.inline
 0.75% :     0.000001s :      6: predicate.inline_without_move
 0.27% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000002s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.39% :     0.000002s :     17: predicate.load_eliminater
 1.84% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.41% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.01% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 0.87% :     0.000001s :     12: predicate.merge_addn
 0.94% :     0.000002s :     12: predicate.minmaximum_grad
 1.39% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 1.86% :     0.000003s :     17: predicate.partial_eliminate
 1.05% :     0.000002s :     12: predicate.print_const_string_wrapper
 5.71% :     0.000009s :     12: predicate.reduce_eliminate
 1.23% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.36% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.24% :     0.000002s :     20: predicate.replace_applicator
 0.36% :     0.000001s :      6: predicate.replace_old_param
 0.21% :     0.000000s :      3: predicate.reset_defer_inline
 2.68% :     0.000004s :     12: predicate.reshape_eliminate
 1.00% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.09% :     0.000002s :     12: predicate.same_eliminate
 0.37% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000002s :      6: predicate.special_op_eliminate
 0.73% :     0.000001s :      6: predicate.specialize_transform
 1.62% :     0.000003s :     12: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.29% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.75% :     0.000004s :     20: predicate.switch_defer_inline
 2.01% :     0.000003s :     20: predicate.switch_layer_defer_inline
 7.39% :     0.000012s :     65: predicate.switch_simplify
 1.22% :     0.000002s :     12: predicate.tile_eliminate
 1.00% :     0.000002s :     12: predicate.transpose_eliminate
 1.13% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.06% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.13% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.25% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.35% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.31% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.09% :     0.000003s :     23: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000003s :     12: predicate.value_based_eliminate
 0.33% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000985    15
45.95% :     0.000452s :      6: func_graph_cloner_run.FuncGraphClonerGraph
13.98% :     0.000138s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.08% :     0.000395s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023233    72
 0.19% :     0.000044s :      1: add_recomputation
 0.38% :     0.000088s :      1: auto_monad
 0.06% :     0.000014s :      1: auto_monad_reorder
 1.47% :     0.000341s :      1: bootstrap
 0.09% :     0.000021s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.11% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.12% :     0.000029s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
31.87% :     0.007403s :      1: jit_opt_a
 0.71% :     0.000165s :      1: jit_opt_after_cconv
 0.24% :     0.000055s :      1: jit_opt_b
 8.28% :     0.001923s :      1: loop_unroll
 3.07% :     0.000714s :      1: mutable_eliminate
 8.91% :     0.002069s :     26: opt.transform.jit_opt_a
 0.24% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.37% :     0.000550s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000059s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000014s :      1: remove_dup_value
 2.70% :     0.000628s :      1: renormalize.infer
 1.59% :     0.000371s :      1: renormalize.specialize
 0.05% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000030s :      1: rewriter_after_opt_a
 0.40% :     0.000092s :      1: rewriter_before_opt_a
 0.35% :     0.000081s :      1: symbol_engine_optimizer
35.60% :     0.008271s :      1: type_inference


TotalTime = 0.0410631, [33]
[bootstrap]: 0.00036694
[type_inference]: 0.00510456
[event_method]: 2.053e-05
[auto_monad]: 6.603e-05
[graph_reusing]: 6.49999e-06
[pre_auto_parallel]: 2.84999e-06
[py_interpret_to_execute]: 4.363e-05
[rewriter_before_opt_a]: 8e-05
[expand_dump_flag]: 3.36001e-06
[jit_opt_a]: 0.0262227, [2]
    [Cycle 1]: 0.0222088, [27]
        [switch_simplify]: 6.113e-05
        [loop_unroll]: 2.975e-05
        [a_1]: 0.0005808
        [with_stream_mark]: 2.343e-05
        [recompute_prepare]: 7.28999e-06
        [updatestate_depend_eliminate]: 3.91001e-06
        [updatestate_assign_eliminate]: 3.03e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [parameter_eliminate]: 2.17001e-06
        [specialize_transform]: 5.98998e-06
        [updatestate_useless_node_eliminater]: 4.99998e-06
        [accelerated_algorithm]: 5.57001e-06
        [meta_shard_fg_expand]: 2.11e-06
        [get_grad_eliminate_]: 5.51e-06
        [merge_forward]: 3.76999e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.39e-05
        [j_node_and_user_rematch]: 8.82999e-06
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 9.00999e-06
        [inline_without_move]: 5.11997e-06
        [renormalize]: 0.0211387
        [add_forward_monad_depend]: 8.89e-06
        [auto_monad_grad]: 2.83e-06
        [auto_monad_eliminator]: 1.855e-05
        [cse]: 3.451e-05
        [replace_applicator]: 2.058e-05
    [Cycle 2]: 0.00038166, [27]
        [switch_simplify]: 6.63e-06
        [loop_unroll]: 6.79999e-06
        [a_1]: 0.00011531
        [with_stream_mark]: 1.225e-05
        [recompute_prepare]: 5.39e-06
        [updatestate_depend_eliminate]: 2.81999e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.25999e-06
        [specialize_transform]: 4.94e-06
        [updatestate_useless_node_eliminater]: 5.35999e-06
        [accelerated_algorithm]: 8.37e-06
        [meta_shard_fg_expand]: 1.84e-06
        [get_grad_eliminate_]: 5.49e-06
        [merge_forward]: 3.23998e-06
        [cell_reuse_recompute_pass]: 3.93999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.429e-05
        [j_node_and_user_rematch]: 9.32999e-06
        [meta_fg_expand]: 2.21998e-06
        [replace_old_param]: 8.13999e-06
        [inline_without_move]: 1.212e-05
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.38002e-06
        [auto_monad_grad]: 8.79983e-07
        [auto_monad_eliminator]: 5.88998e-06
        [cse]: 1.194e-05
        [replace_applicator]: 5.08002e-06
[py_interpret_to_execute_after_opt_a]: 1.956e-05
[rewriter_after_opt_a]: 3.979e-05
[convert_after_rewriter]: 7.3e-06
[order_py_execute_after_rewriter]: 4.73001e-06
[mutable_eliminate]: 0.00078894
[jit_opt_b]: 6.939e-05, [1]
    [Cycle 1]: 6.051e-05, [2]
        [frontend_op_eliminate]: 1.938e-05
        [inline_after_opt_a]: 1.597e-05
[cconv]: 2.981e-05
[loop_unroll]: 0.00043809
[jit_opt_after_cconv]: 0.00015474, [1]
    [Cycle 1]: 0.0001483, [11]
        [c_1]: 2.468e-05
        [parameter_eliminate]: 3.53e-06
        [updatestate_depend_eliminate]: 6.59999e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [cse]: 2.287e-05
        [call_graph_tuple_transform]: 2.139e-05
        [tuple_list_get_item_eliminator]: 5.86e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 1.00001e-06
        [switch_simplify]: 5.47999e-06
[remove_dup_value]: 1.429e-05
[partial_unused_args_eliminate]: 2.10002e-06
[environ_conv]: 4.89e-06
[add_recomputation]: 5.094e-05
[cse_after_recomputation]: 2.092e-05, [1]
    [Cycle 1]: 1.516e-05, [1]
        [cse]: 9.27999e-06
[auto_monad_reorder]: 1.533e-05
[get_jit_bprop_graph]: 2.01e-06
[rewriter_after_jit_bprop_graph]: 8.02e-06
[opt_after_jit_grad]: 0.00046255
[symbol_engine_optimizer]: 7.579e-05, [1]
    [Cycle 1]: 6.861e-05, [6]
        [build]: 5.42999e-06
        [elim_shapecalc]: 7.95e-06
        [elim_not_effective]: 1.286e-05
        [opt_reshape]: 5.61e-06
        [fold_const_symbol]: 9.10001e-06
        [renormalize]: 4.09986e-07
[validate]: 4.342e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00665317
[execute]: 1.045e-05
Sums
    bootstrap                            :     0.000367s :  1.00%
    type_inference                       :     0.005105s : 13.89%
    event_method                         :     0.000021s :  0.06%
    auto_monad                           :     0.000066s :  0.18%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.12%
    rewriter_before_opt_a                :     0.000080s :  0.22%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000068s :  0.18%
    jit_opt_a.loop_unroll                :     0.000037s :  0.10%
    jit_opt_a.a_1                        :     0.000696s :  1.89%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000014s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000017s :  0.05%
    jit_opt_a.inline_without_move        :     0.000017s :  0.05%
    jit_opt_a.renormalize                :     0.021139s : 57.54%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.07%
    jit_opt_a.cse                        :     0.000046s :  0.13%
    jit_opt_a.replace_applicator         :     0.000026s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000020s :  0.05%
    rewriter_after_opt_a                 :     0.000040s :  0.11%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000789s :  2.15%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.04%
    cconv                                :     0.000030s :  0.08%
    loop_unroll                          :     0.000438s :  1.19%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.01%
    remove_dup_value                     :     0.000014s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.01%
    add_recomputation                    :     0.000051s :  0.14%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000015s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000463s :  1.26%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000043s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006653s : 18.11%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.000246    23
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.45% :     0.000006s :      3: substitution.graph_param_transform
85.58% :     0.000211s :      4: substitution.inline
 1.69% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.94% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.80% :     0.000004s :      2: substitution.replace_old_param
 5.17% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005030     2
82.52% :     0.004150s :      1: type_inference.infer
17.48% :     0.000879s :      1: type_inference.specialize
------[replace.]   0.000075     6
66.38% :     0.000050s :      4: replace.inline
33.62% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000220     6
94.77% :     0.000208s :      4: match.inline
 5.23% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.55% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.61% :     0.000002s :     10: predicate.addn_zero_filter
 2.74% :     0.000004s :     10: predicate.arithmetic_simplify
 1.46% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.93% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.33% :     0.000002s :      6: predicate.get_grad_eliminate
 0.36% :     0.000000s :      3: predicate.graph_param_transform
 5.08% :     0.000007s :     22: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.46% :     0.000002s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.17% :     0.000003s :     15: predicate.load_eliminater
 1.29% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.78% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.49% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.18% :     0.000002s :     10: predicate.minmaximum_grad
 2.02% :     0.000003s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 1.90% :     0.000003s :     15: predicate.partial_eliminate
 1.42% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.24% :     0.000003s :     10: predicate.reduce_eliminate
 1.61% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.61% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.05% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000002s :     10: predicate.tile_eliminate
 1.43% :     0.000002s :     10: predicate.transpose_eliminate
 1.26% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.49% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.54% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.46% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.50% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000744    12
54.98% :     0.000409s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.41% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.61% :     0.000287s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.063245    76
 0.09% :     0.000054s :      1: add_recomputation
 0.11% :     0.000070s :      1: auto_monad
 0.03% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.61% :     0.000386s :      1: bootstrap
 0.05% :     0.000033s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.04% :     0.000023s :      1: cse_after_recomputation
 0.01% :     0.000008s :      1: environ_conv
 0.04% :     0.000025s :      1: event_method
 0.02% :     0.000016s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000009s :      1: graph_reusing
41.47% :     0.026227s :      1: jit_opt_a
 0.25% :     0.000157s :      1: jit_opt_after_cconv
 0.11% :     0.000072s :      1: jit_opt_b
 0.71% :     0.000446s :      1: loop_unroll
 1.26% :     0.000799s :      1: mutable_eliminate
 1.48% :     0.000936s :     26: opt.transform.jit_opt_a
 0.09% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000029s :      4: opt.transform.jit_opt_b
 0.02% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.05% :     0.000032s :      4: opt.transform.symbol_engine_opt
 0.75% :     0.000472s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000047s :      1: py_interpret_to_execute
 0.04% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000017s :      1: remove_dup_value
 0.72% :     0.000458s :      1: renormalize.infer
32.68% :     0.020671s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000043s :      1: rewriter_after_opt_a
 0.13% :     0.000083s :      1: rewriter_before_opt_a
 0.12% :     0.000078s :      1: symbol_engine_optimizer
10.55% :     0.006673s :      1: task_emit
 8.10% :     0.005123s :      1: type_inference
 0.11% :     0.000067s :      1: validate


TotalTime = 0.026857, [33]
[bootstrap]: 0.00041854
[type_inference]: 0.00630611
[event_method]: 2.272e-05
[auto_monad]: 7.879e-05
[graph_reusing]: 6.26e-06
[pre_auto_parallel]: 3.43e-06
[py_interpret_to_execute]: 5.411e-05
[rewriter_before_opt_a]: 8.502e-05
[expand_dump_flag]: 3.78001e-06
[jit_opt_a]: 0.00805812, [2]
    [Cycle 1]: 0.00275254, [27]
        [switch_simplify]: 9.211e-05
        [loop_unroll]: 3.274e-05
        [a_1]: 0.00075032
        [with_stream_mark]: 2.821e-05
        [recompute_prepare]: 1.078e-05
        [updatestate_depend_eliminate]: 4.57998e-06
        [updatestate_assign_eliminate]: 3.48999e-06
        [updatestate_loads_eliminate]: 2.74999e-06
        [parameter_eliminate]: 2.43e-06
        [specialize_transform]: 6.15002e-06
        [updatestate_useless_node_eliminater]: 5.69e-06
        [accelerated_algorithm]: 6.61999e-06
        [meta_shard_fg_expand]: 3.03998e-06
        [get_grad_eliminate_]: 5.48002e-06
        [merge_forward]: 4.82998e-06
        [cell_reuse_recompute_pass]: 1.63997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.587e-05
        [j_node_and_user_rematch]: 1.14e-05
        [meta_fg_expand]: 3.07002e-06
        [replace_old_param]: 1.056e-05
        [inline_without_move]: 5.59e-06
        [renormalize]: 0.00104049
        [add_forward_monad_depend]: 9.82001e-06
        [auto_monad_grad]: 2.81e-06
        [auto_monad_eliminator]: 1.913e-05
        [cse]: 3.408e-05
        [replace_applicator]: 2.069e-05
    [Cycle 2]: 0.00038407, [27]
        [switch_simplify]: 7.52998e-06
        [loop_unroll]: 5.54998e-06
        [a_1]: 0.00010284
        [with_stream_mark]: 1.59e-05
        [recompute_prepare]: 6.17999e-06
        [updatestate_depend_eliminate]: 3.48e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [parameter_eliminate]: 1.32e-06
        [specialize_transform]: 5.17999e-06
        [updatestate_useless_node_eliminater]: 5.06997e-06
        [accelerated_algorithm]: 5.89999e-06
        [meta_shard_fg_expand]: 1.67001e-06
        [get_grad_eliminate_]: 4.54998e-06
        [merge_forward]: 4.15999e-06
        [cell_reuse_recompute_pass]: 3.46999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.917e-05
        [j_node_and_user_rematch]: 9.17999e-06
        [meta_fg_expand]: 2.10002e-06
        [replace_old_param]: 9.19e-06
        [inline_without_move]: 4.79002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.59001e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 7e-06
        [cse]: 1.308e-05
        [replace_applicator]: 5.71e-06
[py_interpret_to_execute_after_opt_a]: 1.816e-05
[rewriter_after_opt_a]: 4.359e-05
[convert_after_rewriter]: 8.05999e-06
[order_py_execute_after_rewriter]: 5.24998e-06
[mutable_eliminate]: 0.00095164
[jit_opt_b]: 6.091e-05, [1]
    [Cycle 1]: 5.173e-05, [2]
        [frontend_op_eliminate]: 2.167e-05
        [inline_after_opt_a]: 1.693e-05
[cconv]: 3.418e-05
[loop_unroll]: 0.00059183
[jit_opt_after_cconv]: 0.00017117, [1]
    [Cycle 1]: 0.00016361, [11]
        [c_1]: 2.371e-05
        [parameter_eliminate]: 5.89e-06
        [updatestate_depend_eliminate]: 1.088e-05
        [updatestate_assign_eliminate]: 2.52001e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [cse]: 3.107e-05
        [call_graph_tuple_transform]: 2.398e-05
        [tuple_list_get_item_eliminator]: 5.27001e-06
        [none_parameter_eliminate]: 1.67001e-06
        [renormalize]: 5.50004e-07
        [switch_simplify]: 5.58997e-06
[remove_dup_value]: 1.627e-05
[partial_unused_args_eliminate]: 2.06e-06
[environ_conv]: 5.91e-06
[add_recomputation]: 6.349e-05
[cse_after_recomputation]: 2.392e-05, [1]
    [Cycle 1]: 1.702e-05, [1]
        [cse]: 1.092e-05
[auto_monad_reorder]: 1.61e-05
[get_jit_bprop_graph]: 2.41998e-06
[rewriter_after_jit_bprop_graph]: 9.39998e-06
[opt_after_jit_grad]: 0.00052509
[symbol_engine_optimizer]: 7.995e-05, [1]
    [Cycle 1]: 7.211e-05, [6]
        [build]: 4.58999e-06
        [elim_shapecalc]: 8.12e-06
        [elim_not_effective]: 1.539e-05
        [opt_reshape]: 5.62001e-06
        [fold_const_symbol]: 9.47001e-06
        [renormalize]: 9.39996e-07
[validate]: 4.804e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.00886362
[execute]: 1.114e-05
Sums
    bootstrap                            :     0.000419s :  2.01%
    type_inference                       :     0.006306s : 30.33%
    event_method                         :     0.000023s :  0.11%
    auto_monad                           :     0.000079s :  0.38%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000054s :  0.26%
    rewriter_before_opt_a                :     0.000085s :  0.41%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000100s :  0.48%
    jit_opt_a.loop_unroll                :     0.000038s :  0.18%
    jit_opt_a.a_1                        :     0.000853s :  4.10%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000009s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.001041s :  5.01%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.13%
    jit_opt_a.cse                        :     0.000047s :  0.23%
    jit_opt_a.replace_applicator         :     0.000026s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.09%
    rewriter_after_opt_a                 :     0.000044s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000952s :  4.58%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.08%
    cconv                                :     0.000034s :  0.16%
    loop_unroll                          :     0.000592s :  2.85%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000031s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000063s :  0.31%
    cse_after_recomputation.cse          :     0.000011s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000525s :  2.53%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000048s :  0.23%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008864s : 42.64%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000365    23
 0.82% :     0.000003s :      2: substitution.elim_not_effective
 0.36% :     0.000001s :      2: substitution.fold_const_symbol
 1.61% :     0.000006s :      3: substitution.graph_param_transform
87.86% :     0.000320s :      4: substitution.inline
 1.31% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.47% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.00% :     0.000007s :      2: substitution.replace_old_param
 4.57% :     0.000017s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006206     2
81.58% :     0.005063s :      1: type_inference.infer
18.42% :     0.001143s :      1: type_inference.specialize
------[replace.]   0.000088     6
64.81% :     0.000057s :      4: replace.inline
35.19% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000331     6
95.33% :     0.000316s :      4: match.inline
 4.67% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000141   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.86% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 2.55% :     0.000004s :     10: predicate.arithmetic_simplify
 1.38% :     0.000002s :     10: predicate.cast_eliminate
 0.35% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.83% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 0.88% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.38% :     0.000002s :     10: predicate.environ_get_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.35% :     0.000000s :      3: predicate.graph_param_transform
 5.82% :     0.000008s :     22: predicate.inline
 0.74% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.66% :     0.000002s :      6: predicate.less_batch_normalization
 1.32% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.49% :     0.000002s :     15: predicate.load_eliminater
 2.00% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.82% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.92% :     0.000001s :     10: predicate.merge_addn
 0.88% :     0.000001s :     10: predicate.minmaximum_grad
 2.66% :     0.000004s :      3: predicate.mutable_eliminate
 0.34% :     0.000000s :      3: predicate.opt_reshape
 1.88% :     0.000003s :     15: predicate.partial_eliminate
 1.02% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.42% :     0.000002s :     18: predicate.replace_applicator
 0.41% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.59% :     0.000002s :     10: predicate.reshape_eliminate
 1.02% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.56% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.67% :     0.000001s :      6: predicate.special_op_eliminate
 0.84% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.73% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.10% :     0.000003s :     16: predicate.switch_defer_inline
 1.68% :     0.000002s :     16: predicate.switch_layer_defer_inline
10.76% :     0.000015s :     53: predicate.switch_simplify
 1.32% :     0.000002s :     10: predicate.tile_eliminate
 1.61% :     0.000002s :     10: predicate.transpose_eliminate
 1.78% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.44% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.69% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.32% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.37% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.43% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.37% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000812    12
46.19% :     0.000375s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.30% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.51% :     0.000378s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.029144    76
 0.23% :     0.000067s :      1: add_recomputation
 0.29% :     0.000083s :      1: auto_monad
 0.06% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.51% :     0.000440s :      1: bootstrap
 0.13% :     0.000037s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.10% :     0.000029s :      1: event_method
 0.06% :     0.000017s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
27.67% :     0.008063s :      1: jit_opt_a
 0.60% :     0.000174s :      1: jit_opt_after_cconv
 0.22% :     0.000064s :      1: jit_opt_b
 2.07% :     0.000602s :      1: loop_unroll
 3.32% :     0.000966s :      1: mutable_eliminate
 3.87% :     0.001128s :     26: opt.transform.jit_opt_a
 0.19% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000030s :      4: opt.transform.jit_opt_b
 0.06% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000023s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.83% :     0.000533s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000058s :      1: py_interpret_to_execute
 0.07% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 1.95% :     0.000567s :      1: renormalize.infer
 1.58% :     0.000461s :      1: renormalize.specialize
 0.04% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000048s :      1: rewriter_after_opt_a
 0.30% :     0.000089s :      1: rewriter_before_opt_a
 0.28% :     0.000083s :      1: symbol_engine_optimizer
30.50% :     0.008889s :      1: task_emit
21.73% :     0.006333s :      1: type_inference
 0.24% :     0.000071s :      1: validate


TotalTime = 0.0219634, [33]
[bootstrap]: 0.00039743
[type_inference]: 0.005526
[event_method]: 1.764e-05
[auto_monad]: 6.305e-05
[graph_reusing]: 4.72e-06
[pre_auto_parallel]: 2.00002e-06
[py_interpret_to_execute]: 4.263e-05
[rewriter_before_opt_a]: 7.235e-05
[expand_dump_flag]: 3.07002e-06
[jit_opt_a]: 0.00663251, [2]
    [Cycle 1]: 0.00233949, [27]
        [switch_simplify]: 7.008e-05
        [loop_unroll]: 3.306e-05
        [a_1]: 0.00063009
        [with_stream_mark]: 2.111e-05
        [recompute_prepare]: 7.4e-06
        [updatestate_depend_eliminate]: 3.46999e-06
        [updatestate_assign_eliminate]: 3.16999e-06
        [updatestate_loads_eliminate]: 2.32001e-06
        [parameter_eliminate]: 2.25002e-06
        [specialize_transform]: 5.86e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 5.42999e-06
        [meta_shard_fg_expand]: 2.07001e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 3.9e-06
        [cell_reuse_recompute_pass]: 1.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.491e-05
        [j_node_and_user_rematch]: 9.22999e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 9.05999e-06
        [inline_without_move]: 5.02e-06
        [renormalize]: 0.00123969
        [add_forward_monad_depend]: 8.12e-06
        [auto_monad_grad]: 2.73998e-06
        [auto_monad_eliminator]: 1.736e-05
        [cse]: 2.317e-05
        [replace_applicator]: 1.639e-05
    [Cycle 2]: 0.00036066, [27]
        [switch_simplify]: 6.28998e-06
        [loop_unroll]: 5.29e-06
        [a_1]: 0.00010312
        [with_stream_mark]: 1.295e-05
        [recompute_prepare]: 5.24e-06
        [updatestate_depend_eliminate]: 2.69999e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.45999e-06
        [specialize_transform]: 5.04003e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.94e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 4.90001e-06
        [merge_forward]: 3.29001e-06
        [cell_reuse_recompute_pass]: 2.11003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.52e-05
        [j_node_and_user_rematch]: 8.78001e-06
        [meta_fg_expand]: 1.83002e-06
        [replace_old_param]: 7.97e-06
        [inline_without_move]: 4.87998e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.50999e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 5.86998e-06
        [cse]: 1.073e-05
        [replace_applicator]: 5.70001e-06
[py_interpret_to_execute_after_opt_a]: 1.334e-05
[rewriter_after_opt_a]: 3.911e-05
[convert_after_rewriter]: 7.27002e-06
[order_py_execute_after_rewriter]: 5.66003e-06
[mutable_eliminate]: 0.00068312
[jit_opt_b]: 5.316e-05, [1]
    [Cycle 1]: 4.528e-05, [2]
        [frontend_op_eliminate]: 1.842e-05
        [inline_after_opt_a]: 1.495e-05
[cconv]: 2.818e-05
[loop_unroll]: 0.00046933
[jit_opt_after_cconv]: 0.00015109, [1]
    [Cycle 1]: 0.00014415, [11]
        [c_1]: 2.267e-05
        [parameter_eliminate]: 2.98e-06
        [updatestate_depend_eliminate]: 6.31e-06
        [updatestate_assign_eliminate]: 2.74999e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 2.245e-05
        [call_graph_tuple_transform]: 2.123e-05
        [tuple_list_get_item_eliminator]: 5.52001e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 5.50001e-06
[remove_dup_value]: 1.503e-05
[partial_unused_args_eliminate]: 2.44001e-06
[environ_conv]: 6.04999e-06
[add_recomputation]: 5.233e-05
[cse_after_recomputation]: 2.122e-05, [1]
    [Cycle 1]: 1.584e-05, [1]
        [cse]: 9.46e-06
[auto_monad_reorder]: 1.698e-05
[get_jit_bprop_graph]: 1.96998e-06
[rewriter_after_jit_bprop_graph]: 5.85002e-06
[opt_after_jit_grad]: 0.00048363
[symbol_engine_optimizer]: 7.801e-05, [1]
    [Cycle 1]: 7.083e-05, [6]
        [build]: 4.08999e-06
        [elim_shapecalc]: 8.70999e-06
        [elim_not_effective]: 1.381e-05
        [opt_reshape]: 6.21e-06
        [fold_const_symbol]: 9.37999e-06
        [renormalize]: 5.29981e-07
[validate]: 4.409e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00679673
[execute]: 8.23001e-06
Sums
    bootstrap                            :     0.000397s :  2.29%
    type_inference                       :     0.005526s : 31.81%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000063s :  0.36%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.25%
    rewriter_before_opt_a                :     0.000072s :  0.42%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.44%
    jit_opt_a.loop_unroll                :     0.000038s :  0.22%
    jit_opt_a.a_1                        :     0.000733s :  4.22%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001240s :  7.14%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.13%
    jit_opt_a.cse                        :     0.000034s :  0.20%
    jit_opt_a.replace_applicator         :     0.000022s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000039s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000683s :  3.93%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000028s :  0.16%
    loop_unroll                          :     0.000469s :  2.70%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000052s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000484s :  2.78%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006797s : 39.13%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000272    23
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000002s :      2: substitution.fold_const_symbol
 2.47% :     0.000007s :      3: substitution.graph_param_transform
86.34% :     0.000235s :      4: substitution.inline
 1.41% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.75% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.55% :     0.000004s :      2: substitution.replace_old_param
 5.15% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005455     2
82.55% :     0.004503s :      1: type_inference.infer
17.45% :     0.000952s :      1: type_inference.specialize
------[replace.]   0.000086     6
68.69% :     0.000059s :      4: replace.inline
31.31% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000245     6
94.77% :     0.000232s :      4: match.inline
 5.23% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.53% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 2.05% :     0.000003s :     10: predicate.arithmetic_simplify
 1.26% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000002s :     10: predicate.depend_value_elim
 1.22% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.69% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.07% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.53% :     0.000002s :     10: predicate.environ_get_eliminate
 1.28% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.04% :     0.000007s :     22: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000002s :      6: predicate.less_batch_normalization
 1.83% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.20% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.91% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 1.87% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.31% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.89% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.23% :     0.000002s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.61% :     0.000003s :     16: predicate.switch_defer_inline
 1.81% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.02% :     0.000011s :     53: predicate.switch_simplify
 1.31% :     0.000002s :     10: predicate.tile_eliminate
 1.27% :     0.000002s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.77% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.87% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.48% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.45% :     0.000003s :     10: predicate.value_based_eliminate
 0.47% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000696    12
45.34% :     0.000315s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.35% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.31% :     0.000329s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024273    76
 0.23% :     0.000056s :      1: add_recomputation
 0.28% :     0.000067s :      1: auto_monad
 0.08% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.70% :     0.000413s :      1: bootstrap
 0.13% :     0.000031s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
27.35% :     0.006637s :      1: jit_opt_a
 0.64% :     0.000154s :      1: jit_opt_after_cconv
 0.23% :     0.000056s :      1: jit_opt_b
 1.97% :     0.000478s :      1: loop_unroll
 2.86% :     0.000693s :      1: mutable_eliminate
 3.99% :     0.000969s :     26: opt.transform.jit_opt_a
 0.21% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.03% :     0.000492s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000046s :      1: py_interpret_to_execute
 0.06% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.49% :     0.000604s :      1: renormalize.infer
 2.58% :     0.000626s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000042s :      1: rewriter_after_opt_a
 0.31% :     0.000075s :      1: rewriter_before_opt_a
 0.33% :     0.000081s :      1: symbol_engine_optimizer
28.07% :     0.006812s :      1: task_emit
22.82% :     0.005540s :      1: type_inference
 0.27% :     0.000066s :      1: validate


TotalTime = 0.0187723, [30]
[bootstrap]: 0.00039995
[type_inference]: 0.00870496
[event_method]: 2.511e-05
[auto_monad]: 8.305e-05
[graph_reusing]: 6.50002e-06
[pre_auto_parallel]: 2.58e-06
[py_interpret_to_execute]: 5.717e-05
[rewriter_before_opt_a]: 0.00010277
[expand_dump_flag]: 3.01001e-06
[jit_opt_a]: 0.0069109, [2]
    [Cycle 1]: 0.00248974, [27]
        [switch_simplify]: 0.00014584
        [loop_unroll]: 3.453e-05
        [a_1]: 0.0007921
        [with_stream_mark]: 2.465e-05
        [recompute_prepare]: 7.46999e-06
        [updatestate_depend_eliminate]: 4.37e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [parameter_eliminate]: 2.28998e-06
        [specialize_transform]: 5.92001e-06
        [updatestate_useless_node_eliminater]: 5.28002e-06
        [accelerated_algorithm]: 6.07001e-06
        [meta_shard_fg_expand]: 2.14999e-06
        [get_grad_eliminate_]: 2.643e-05
        [merge_forward]: 3.75998e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.554e-05
        [j_node_and_user_rematch]: 1.012e-05
        [meta_fg_expand]: 2.41998e-06
        [replace_old_param]: 1.08e-05
        [inline_without_move]: 5.54e-06
        [renormalize]: 0.00111756
        [add_forward_monad_depend]: 6.19001e-06
        [auto_monad_grad]: 3.22002e-06
        [auto_monad_eliminator]: 1.428e-05
        [cse]: 2.784e-05
        [replace_applicator]: 1.553e-05
    [Cycle 2]: 0.00032752, [27]
        [switch_simplify]: 5.60001e-06
        [loop_unroll]: 4.90001e-06
        [a_1]: 9.455e-05
        [with_stream_mark]: 1.118e-05
        [recompute_prepare]: 5.20001e-06
        [updatestate_depend_eliminate]: 2.84999e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.17999e-06
        [specialize_transform]: 5.10999e-06
        [updatestate_useless_node_eliminater]: 4.70999e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 1.97999e-06
        [get_grad_eliminate_]: 5.01002e-06
        [merge_forward]: 3.48e-06
        [cell_reuse_recompute_pass]: 2.44999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.445e-05
        [j_node_and_user_rematch]: 8.59002e-06
        [meta_fg_expand]: 1.58002e-06
        [replace_old_param]: 7.58001e-06
        [inline_without_move]: 4.47998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 1.14e-06
        [auto_monad_eliminator]: 4.61002e-06
        [cse]: 9.42001e-06
        [replace_applicator]: 4.91997e-06
[py_interpret_to_execute_after_opt_a]: 1.186e-05
[rewriter_after_opt_a]: 3.819e-05
[convert_after_rewriter]: 7.23e-06
[order_py_execute_after_rewriter]: 5.79999e-06
[mutable_eliminate]: 0.00077216
[jit_opt_b]: 5.245e-05, [1]
    [Cycle 1]: 4.424e-05, [2]
        [frontend_op_eliminate]: 1.742e-05
        [inline_after_opt_a]: 1.503e-05
[cconv]: 3.395e-05
[loop_unroll]: 0.00047124
[jit_opt_after_cconv]: 0.00015304, [1]
    [Cycle 1]: 0.00014569, [11]
        [c_1]: 2.173e-05
        [parameter_eliminate]: 3.08e-06
        [updatestate_depend_eliminate]: 7.45e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 2.249e-05
        [call_graph_tuple_transform]: 1.935e-05
        [tuple_list_get_item_eliminator]: 5.22e-06
        [none_parameter_eliminate]: 1.82001e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 6.17999e-06
[remove_dup_value]: 1.449e-05
[partial_unused_args_eliminate]: 2.83e-06
[environ_conv]: 6.17999e-06
[add_recomputation]: 5.086e-05
[cse_after_recomputation]: 2.217e-05, [1]
    [Cycle 1]: 1.598e-05, [1]
        [cse]: 9.64e-06
[auto_monad_reorder]: 1.592e-05
[get_jit_bprop_graph]: 2.27999e-06
[rewriter_after_jit_bprop_graph]: 8.12e-06
[opt_after_jit_grad]: 0.00047479
[symbol_engine_optimizer]: 7.577e-05, [1]
    [Cycle 1]: 6.955e-05, [6]
        [build]: 4.57998e-06
        [elim_shapecalc]: 7.68999e-06
        [elim_not_effective]: 1.386e-05
        [opt_reshape]: 5.81e-06
        [fold_const_symbol]: 8.61002e-06
        [renormalize]: 4.39992e-07
[validate]: 5.852e-05
Sums
    bootstrap                            :     0.000400s :  2.85%
    type_inference                       :     0.008705s : 61.97%
    event_method                         :     0.000025s :  0.18%
    auto_monad                           :     0.000083s :  0.59%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000057s :  0.41%
    rewriter_before_opt_a                :     0.000103s :  0.73%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000151s :  1.08%
    jit_opt_a.loop_unroll                :     0.000039s :  0.28%
    jit_opt_a.a_1                        :     0.000887s :  6.31%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000031s :  0.22%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.13%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.001118s :  7.96%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000037s :  0.27%
    jit_opt_a.replace_applicator         :     0.000020s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.27%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000772s :  5.50%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000034s :  0.24%
    loop_unroll                          :     0.000471s :  3.35%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000051s :  0.36%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000016s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.06%
    opt_after_jit_grad                   :     0.000475s :  3.38%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000059s :  0.42%


Time group info:
------[substitution.]   0.000361    26
 0.65% :     0.000002s :      2: substitution.elim_not_effective
 0.42% :     0.000002s :      2: substitution.fold_const_symbol
 1.46% :     0.000005s :      3: substitution.graph_param_transform
87.06% :     0.000314s :      6: substitution.inline
 1.17% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.34% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.42% :     0.000005s :      2: substitution.replace_old_param
 2.86% :     0.000010s :      1: substitution.switch_simplify
 3.62% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008623     2
84.23% :     0.007263s :      1: type_inference.infer
15.77% :     0.001360s :      1: type_inference.specialize
------[replace.]   0.000160     9
47.52% :     0.000076s :      6: replace.inline
29.59% :     0.000047s :      1: replace.switch_simplify
22.89% :     0.000037s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000333     9
93.56% :     0.000311s :      6: match.inline
 2.85% :     0.000009s :      1: match.switch_simplify
 3.59% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   801
 1.41% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.17% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     12: predicate.addn_check_dump
 1.31% :     0.000002s :     12: predicate.addn_zero_filter
 2.16% :     0.000003s :     12: predicate.arithmetic_simplify
 1.31% :     0.000002s :     12: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     12: predicate.compare_switch_simplify
 1.36% :     0.000002s :     12: predicate.depend_value_elim
 1.11% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000001s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.39% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.24% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.25% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     12: predicate.environ_get_eliminate
 1.23% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.34% :     0.000000s :      3: predicate.graph_param_transform
 6.00% :     0.000008s :     26: predicate.inline
 0.77% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.55% :     0.000002s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     17: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.90% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.99% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     12: predicate.merge_addn
 1.09% :     0.000001s :     12: predicate.minmaximum_grad
 2.01% :     0.000003s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.03% :     0.000003s :     17: predicate.partial_eliminate
 1.14% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     12: predicate.reduce_eliminate
 1.55% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.53% :     0.000002s :     20: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.38% :     0.000001s :      3: predicate.reset_defer_inline
 1.68% :     0.000002s :     12: predicate.reshape_eliminate
 1.18% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     12: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.73% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.31% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.73% :     0.000004s :     20: predicate.switch_defer_inline
 2.28% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.69% :     0.000012s :     65: predicate.switch_simplify
 1.29% :     0.000002s :     12: predicate.tile_eliminate
 1.22% :     0.000002s :     12: predicate.transpose_eliminate
 1.42% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.64% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.57% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     12: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001202    15
44.48% :     0.000535s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.51% :     0.000138s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.01% :     0.000529s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021137    72
 0.25% :     0.000054s :      1: add_recomputation
 0.41% :     0.000087s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 1.99% :     0.000420s :      1: bootstrap
 0.17% :     0.000037s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.14% :     0.000030s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
32.72% :     0.006916s :      1: jit_opt_a
 0.74% :     0.000156s :      1: jit_opt_after_cconv
 0.26% :     0.000055s :      1: jit_opt_b
 2.27% :     0.000479s :      1: loop_unroll
 3.71% :     0.000784s :      1: mutable_eliminate
 5.78% :     0.001222s :     26: opt.transform.jit_opt_a
 0.23% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.28% :     0.000483s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000060s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 2.76% :     0.000584s :      1: renormalize.infer
 2.48% :     0.000524s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000041s :      1: rewriter_after_opt_a
 0.50% :     0.000107s :      1: rewriter_before_opt_a
 0.37% :     0.000078s :      1: symbol_engine_optimizer
41.28% :     0.008725s :      1: type_inference


TotalTime = 0.0390554, [33]
[bootstrap]: 0.00046158
[type_inference]: 0.00905409
[event_method]: 1.816e-05
[auto_monad]: 7.045e-05
[graph_reusing]: 6.26e-06
[pre_auto_parallel]: 2.51998e-06
[py_interpret_to_execute]: 4.238e-05
[rewriter_before_opt_a]: 7.357e-05
[expand_dump_flag]: 3.53999e-06
[jit_opt_a]: 0.0178426, [2]
    [Cycle 1]: 0.00783393, [27]
        [switch_simplify]: 5.664e-05
        [loop_unroll]: 3.045e-05
        [a_1]: 0.00054418
        [with_stream_mark]: 1.554e-05
        [recompute_prepare]: 6.98e-06
        [updatestate_depend_eliminate]: 2.99999e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.08998e-06
        [parameter_eliminate]: 8.90024e-07
        [specialize_transform]: 5.99e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 5.42999e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 6.00002e-06
        [merge_forward]: 1.941e-05
        [cell_reuse_recompute_pass]: 9.77001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 5.622e-05
        [j_node_and_user_rematch]: 1.539e-05
        [meta_fg_expand]: 6.88e-06
        [replace_old_param]: 1.549e-05
        [inline_without_move]: 7.66001e-06
        [renormalize]: 0.00116579
        [add_forward_monad_depend]: 9.89001e-06
        [auto_monad_grad]: 2.59999e-06
        [auto_monad_eliminator]: 1.595e-05
        [cse]: 2.128e-05
        [replace_applicator]: 1.657e-05
    [Cycle 2]: 0.0003711, [27]
        [switch_simplify]: 6.34001e-06
        [loop_unroll]: 6.59001e-06
        [a_1]: 0.00011394
        [with_stream_mark]: 1.665e-05
        [recompute_prepare]: 5.84999e-06
        [updatestate_depend_eliminate]: 3.6e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 5.69e-06
        [updatestate_useless_node_eliminater]: 5.18002e-06
        [accelerated_algorithm]: 5.74e-06
        [meta_shard_fg_expand]: 2.16e-06
        [get_grad_eliminate_]: 5.57001e-06
        [merge_forward]: 3.07002e-06
        [cell_reuse_recompute_pass]: 2.62001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.45e-05
        [j_node_and_user_rematch]: 8.27e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 1.347e-05
        [inline_without_move]: 4.67998e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.35001e-06
        [auto_monad_grad]: 9.29984e-07
        [auto_monad_eliminator]: 6.11e-06
        [cse]: 1.044e-05
        [replace_applicator]: 4.77e-06
[py_interpret_to_execute_after_opt_a]: 1.618e-05
[rewriter_after_opt_a]: 3.915e-05
[convert_after_rewriter]: 7.06999e-06
[order_py_execute_after_rewriter]: 5.44e-06
[mutable_eliminate]: 0.0009163
[jit_opt_b]: 5.671e-05, [1]
    [Cycle 1]: 4.878e-05, [2]
        [frontend_op_eliminate]: 1.988e-05
        [inline_after_opt_a]: 1.689e-05
[cconv]: 3.011e-05
[loop_unroll]: 0.00045425
[jit_opt_after_cconv]: 0.00016014, [1]
    [Cycle 1]: 0.00015349, [11]
        [c_1]: 2.339e-05
        [parameter_eliminate]: 3.26001e-06
        [updatestate_depend_eliminate]: 7.62002e-06
        [updatestate_assign_eliminate]: 2.86999e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 2.234e-05
        [call_graph_tuple_transform]: 2.303e-05
        [tuple_list_get_item_eliminator]: 6.27001e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 6.84999e-06
[remove_dup_value]: 1.626e-05
[partial_unused_args_eliminate]: 2.93e-06
[environ_conv]: 6.34999e-06
[add_recomputation]: 5.007e-05
[cse_after_recomputation]: 2.229e-05, [1]
    [Cycle 1]: 1.618e-05, [1]
        [cse]: 9.21998e-06
[auto_monad_reorder]: 1.569e-05
[get_jit_bprop_graph]: 2.09e-06
[rewriter_after_jit_bprop_graph]: 7.90998e-06
[opt_after_jit_grad]: 0.00047365
[symbol_engine_optimizer]: 9.985e-05, [1]
    [Cycle 1]: 9.212e-05, [6]
        [build]: 3.32002e-06
        [elim_shapecalc]: 8.28999e-06
        [elim_not_effective]: 3.489e-05
        [opt_reshape]: 6.93e-06
        [fold_const_symbol]: 8.90999e-06
        [renormalize]: 3.80009e-07
[validate]: 4.736e-05
[backend_pass]: 1.35999e-06
[task_emit]: 0.0088066
[execute]: 1.056e-05
Sums
    bootstrap                            :     0.000462s :  1.99%
    type_inference                       :     0.009054s : 39.10%
    event_method                         :     0.000018s :  0.08%
    auto_monad                           :     0.000070s :  0.30%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.18%
    rewriter_before_opt_a                :     0.000074s :  0.32%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.27%
    jit_opt_a.loop_unroll                :     0.000037s :  0.16%
    jit_opt_a.a_1                        :     0.000658s :  2.84%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.05%
    jit_opt_a.merge_forward              :     0.000022s :  0.10%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000012s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000071s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000024s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000009s :  0.04%
    jit_opt_a.replace_old_param          :     0.000029s :  0.13%
    jit_opt_a.inline_without_move        :     0.000012s :  0.05%
    jit_opt_a.renormalize                :     0.001166s :  5.03%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.10%
    jit_opt_a.cse                        :     0.000032s :  0.14%
    jit_opt_a.replace_applicator         :     0.000021s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.07%
    rewriter_after_opt_a                 :     0.000039s :  0.17%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000916s :  3.96%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.07%
    cconv                                :     0.000030s :  0.13%
    loop_unroll                          :     0.000454s :  1.96%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000016s :  0.07%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000050s :  0.22%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.03%
    opt_after_jit_grad                   :     0.000474s :  2.05%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000035s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000047s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008807s : 38.03%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000217    23
 1.11% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.93% :     0.000006s :      3: substitution.graph_param_transform
78.62% :     0.000170s :      4: substitution.inline
 2.42% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.75% :     0.000006s :      4: substitution.remove_not_recompute_node
 6.62% :     0.000014s :      2: substitution.replace_old_param
 4.91% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008974     2
88.16% :     0.007912s :      1: type_inference.infer
11.84% :     0.001062s :      1: type_inference.specialize
------[replace.]   0.000075     6
67.23% :     0.000051s :      4: replace.inline
32.77% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000177     6
94.60% :     0.000168s :      4: match.inline
 5.40% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.63% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 2.53% :     0.000003s :     10: predicate.arithmetic_simplify
 1.49% :     0.000002s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.91% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.53% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.35% :     0.000002s :     10: predicate.environ_get_eliminate
 1.16% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 4.61% :     0.000006s :     22: predicate.inline
 1.42% :     0.000002s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :      6: predicate.less_batch_normalization
 2.22% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.58% :     0.000002s :     15: predicate.load_eliminater
 1.15% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.12% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 1.31% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.03% :     0.000003s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.87% :     0.000002s :     10: predicate.reduce_eliminate
 1.74% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 1.23% :     0.000002s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.17% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.59% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.61% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.06% :     0.000003s :     16: predicate.switch_defer_inline
 1.79% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.15% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000002s :     10: predicate.tile_eliminate
 1.20% :     0.000002s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.74% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.32% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.41% :     0.000003s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000759    12
44.91% :     0.000341s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.25% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.83% :     0.000386s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.041271    76
 0.13% :     0.000053s :      1: add_recomputation
 0.18% :     0.000074s :      1: auto_monad
 0.04% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 1.18% :     0.000485s :      1: bootstrap
 0.08% :     0.000033s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.06% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.05% :     0.000022s :      1: event_method
 0.04% :     0.000016s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
43.24% :     0.017846s :      1: jit_opt_a
 0.40% :     0.000163s :      1: jit_opt_after_cconv
 0.14% :     0.000059s :      1: jit_opt_b
 1.12% :     0.000462s :      1: loop_unroll
 2.24% :     0.000923s :      1: mutable_eliminate
 2.25% :     0.000928s :     26: opt.transform.jit_opt_a
 0.13% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000028s :      4: opt.transform.jit_opt_b
 0.03% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000055s :      4: opt.transform.symbol_engine_opt
 1.17% :     0.000483s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000046s :      1: py_interpret_to_execute
 0.05% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 1.53% :     0.000630s :      1: renormalize.infer
 1.27% :     0.000524s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000042s :      1: rewriter_after_opt_a
 0.19% :     0.000077s :      1: rewriter_before_opt_a
 0.25% :     0.000102s :      1: symbol_engine_optimizer
21.39% :     0.008829s :      1: task_emit
21.97% :     0.009068s :      1: type_inference
 0.17% :     0.000070s :      1: validate


TotalTime = 0.0295809, [33]
[bootstrap]: 0.00045506
[type_inference]: 0.00594506
[event_method]: 2.002e-05
[auto_monad]: 6.687e-05
[graph_reusing]: 6.74001e-06
[pre_auto_parallel]: 2.91e-06
[py_interpret_to_execute]: 4.464e-05
[rewriter_before_opt_a]: 7.863e-05
[expand_dump_flag]: 3.51999e-06
[jit_opt_a]: 0.0126892, [2]
    [Cycle 1]: 0.00470271, [27]
        [switch_simplify]: 6.812e-05
        [loop_unroll]: 3.159e-05
        [a_1]: 0.00312581
        [with_stream_mark]: 2.85e-05
        [recompute_prepare]: 1.18e-05
        [updatestate_depend_eliminate]: 4.29997e-06
        [updatestate_assign_eliminate]: 3.40998e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [parameter_eliminate]: 2.43e-06
        [specialize_transform]: 6.76e-06
        [updatestate_useless_node_eliminater]: 6.07001e-06
        [accelerated_algorithm]: 6.24001e-06
        [meta_shard_fg_expand]: 2.98e-06
        [get_grad_eliminate_]: 6.03002e-06
        [merge_forward]: 4.03001e-06
        [cell_reuse_recompute_pass]: 1.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.608e-05
        [j_node_and_user_rematch]: 1.052e-05
        [meta_fg_expand]: 2.56e-06
        [replace_old_param]: 1.082e-05
        [inline_without_move]: 5.50001e-06
        [renormalize]: 0.00104412
        [add_forward_monad_depend]: 7.37002e-06
        [auto_monad_grad]: 2.56e-06
        [auto_monad_eliminator]: 1.738e-05
        [cse]: 3.233e-05
        [replace_applicator]: 1.502e-05
    [Cycle 2]: 0.00035905, [27]
        [switch_simplify]: 6.02999e-06
        [loop_unroll]: 5.46998e-06
        [a_1]: 0.00010419
        [with_stream_mark]: 1.254e-05
        [recompute_prepare]: 5.64e-06
        [updatestate_depend_eliminate]: 3.08998e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.32e-06
        [specialize_transform]: 5.37001e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 5.27001e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 5.17e-06
        [merge_forward]: 2.68998e-06
        [cell_reuse_recompute_pass]: 2.21e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.566e-05
        [j_node_and_user_rematch]: 9.14e-06
        [meta_fg_expand]: 1.49998e-06
        [replace_old_param]: 8.76997e-06
        [inline_without_move]: 4.72998e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.49998e-06
        [auto_monad_grad]: 1.07e-06
        [auto_monad_eliminator]: 5.52999e-06
        [cse]: 1.095e-05
        [replace_applicator]: 4.95001e-06
[py_interpret_to_execute_after_opt_a]: 1.542e-05
[rewriter_after_opt_a]: 4.451e-05
[convert_after_rewriter]: 8.58001e-06
[order_py_execute_after_rewriter]: 5.89999e-06
[mutable_eliminate]: 0.00092369
[jit_opt_b]: 5.639e-05, [1]
    [Cycle 1]: 4.771e-05, [2]
        [frontend_op_eliminate]: 1.948e-05
        [inline_after_opt_a]: 1.596e-05
[cconv]: 0.00022043
[loop_unroll]: 0.00057269
[jit_opt_after_cconv]: 0.00017658, [1]
    [Cycle 1]: 0.00016903, [11]
        [c_1]: 2.374e-05
        [parameter_eliminate]: 5.34e-06
        [updatestate_depend_eliminate]: 1.011e-05
        [updatestate_assign_eliminate]: 3.86001e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [cse]: 2.995e-05
        [call_graph_tuple_transform]: 2.456e-05
        [tuple_list_get_item_eliminator]: 6.64999e-06
        [none_parameter_eliminate]: 1.70001e-06
        [renormalize]: 9.70002e-07
        [switch_simplify]: 6.36998e-06
[remove_dup_value]: 1.721e-05
[partial_unused_args_eliminate]: 2.19999e-06
[environ_conv]: 6.47001e-06
[add_recomputation]: 5.451e-05
[cse_after_recomputation]: 2.483e-05, [1]
    [Cycle 1]: 1.843e-05, [1]
        [cse]: 1.235e-05
[auto_monad_reorder]: 1.687e-05
[get_jit_bprop_graph]: 2.01998e-06
[rewriter_after_jit_bprop_graph]: 9.48002e-06
[opt_after_jit_grad]: 0.00053488
[symbol_engine_optimizer]: 8.124e-05, [1]
    [Cycle 1]: 7.297e-05, [6]
        [build]: 3.41001e-06
        [elim_shapecalc]: 8.23001e-06
        [elim_not_effective]: 1.506e-05
        [opt_reshape]: 7.00998e-06
        [fold_const_symbol]: 8.99e-06
        [renormalize]: 4.59986e-07
[validate]: 4.568e-05
[backend_pass]: 8.79983e-07
[task_emit]: 0.00716253
[execute]: 1.054e-05
Sums
    bootstrap                            :     0.000455s :  2.15%
    type_inference                       :     0.005945s : 28.05%
    event_method                         :     0.000020s :  0.09%
    auto_monad                           :     0.000067s :  0.32%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000045s :  0.21%
    rewriter_before_opt_a                :     0.000079s :  0.37%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.35%
    jit_opt_a.loop_unroll                :     0.000037s :  0.17%
    jit_opt_a.a_1                        :     0.003230s : 15.24%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.001044s :  4.93%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.11%
    jit_opt_a.cse                        :     0.000043s :  0.20%
    jit_opt_a.replace_applicator         :     0.000020s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.07%
    rewriter_after_opt_a                 :     0.000045s :  0.21%
    convert_after_rewriter               :     0.000009s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000924s :  4.36%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.000220s :  1.04%
    loop_unroll                          :     0.000573s :  2.70%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000030s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000055s :  0.26%
    cse_after_recomputation.cse          :     0.000012s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.04%
    opt_after_jit_grad                   :     0.000535s :  2.52%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000046s :  0.22%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007163s : 33.79%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000378    23
 0.49% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000002s :      2: substitution.fold_const_symbol
 1.82% :     0.000007s :      3: substitution.graph_param_transform
88.90% :     0.000336s :      4: substitution.inline
 1.09% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.40% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.47% :     0.000006s :      2: substitution.replace_old_param
 4.37% :     0.000017s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005859     2
83.76% :     0.004907s :      1: type_inference.infer
16.24% :     0.000951s :      1: type_inference.specialize
------[replace.]   0.002395     6
98.83% :     0.002368s :      4: replace.inline
 1.17% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000347     6
95.60% :     0.000332s :      4: match.inline
 4.40% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000158   691
 1.76% :     0.000003s :     10: predicate.accumulaten_eliminater
 1.46% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000002s :     10: predicate.addn_check_dump
 1.62% :     0.000003s :     10: predicate.addn_zero_filter
 2.07% :     0.000003s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.84% :     0.000001s :     10: predicate.compare_switch_simplify
 1.14% :     0.000002s :     10: predicate.depend_value_elim
 1.33% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.73% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000001s :      3: predicate.elim_not_effective
 0.43% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.86% :     0.000003s :     10: predicate.environ_add_const_eliminate
 0.78% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     10: predicate.environ_get_eliminate
 0.85% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 1.11% :     0.000002s :      6: predicate.get_grad_eliminate
 0.19% :     0.000000s :      3: predicate.graph_param_transform
 6.23% :     0.000010s :     22: predicate.inline
 0.68% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000002s :      6: predicate.less_batch_normalization
 2.39% :     0.000004s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000003s :     15: predicate.load_eliminater
 1.28% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.25% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.83% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.88% :     0.000001s :     10: predicate.merge_addn
 1.17% :     0.000002s :     10: predicate.minmaximum_grad
 2.14% :     0.000003s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 1.86% :     0.000003s :     15: predicate.partial_eliminate
 1.23% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.13% :     0.000003s :     10: predicate.reduce_eliminate
 1.58% :     0.000003s :     12: predicate.redundant_stop_gradient_eliminater
 0.39% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.21% :     0.000002s :     18: predicate.replace_applicator
 0.42% :     0.000001s :      6: predicate.replace_old_param
 0.18% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.46% :     0.000002s :     10: predicate.same_eliminate
 0.37% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.16% :     0.000002s :      6: predicate.special_op_eliminate
 0.85% :     0.000001s :      6: predicate.specialize_transform
 1.72% :     0.000003s :     10: predicate.split_environ_get_set_with_tuple_value
 1.09% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.31% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 5.99% :     0.000009s :     16: predicate.switch_defer_inline
 1.58% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.56% :     0.000010s :     53: predicate.switch_simplify
 1.44% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000002s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.02% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.59% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 2.17% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.24% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.25% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000003s :     10: predicate.value_based_eliminate
 0.27% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.48% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000957    12
60.00% :     0.000574s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.57% :     0.000063s :      1: func_graph_cloner_run.FuncGraphClonerNode
33.43% :     0.000320s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.034205    76
 0.17% :     0.000058s :      1: add_recomputation
 0.20% :     0.000070s :      1: auto_monad
 0.06% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.41% :     0.000481s :      1: bootstrap
 0.67% :     0.000231s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.08% :     0.000027s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.07% :     0.000024s :      1: event_method
 0.05% :     0.000017s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
37.11% :     0.012694s :      1: jit_opt_a
 0.52% :     0.000179s :      1: jit_opt_after_cconv
 0.17% :     0.000059s :      1: jit_opt_b
 1.71% :     0.000585s :      1: loop_unroll
 2.75% :     0.000939s :      1: mutable_eliminate
10.15% :     0.003473s :     26: opt.transform.jit_opt_a
 0.17% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.59% :     0.000544s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000048s :      1: py_interpret_to_execute
 0.05% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000020s :      1: remove_dup_value
 1.75% :     0.000600s :      1: renormalize.infer
 1.27% :     0.000434s :      1: renormalize.specialize
 0.03% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000048s :      1: rewriter_after_opt_a
 0.24% :     0.000082s :      1: rewriter_before_opt_a
 0.25% :     0.000084s :      1: symbol_engine_optimizer
21.01% :     0.007187s :      1: task_emit
17.44% :     0.005964s :      1: type_inference
 0.20% :     0.000069s :      1: validate


TotalTime = 0.0347924, [33]
[bootstrap]: 0.00037735
[type_inference]: 0.0129943
[event_method]: 2.058e-05
[auto_monad]: 7.527e-05
[graph_reusing]: 6.26e-06
[pre_auto_parallel]: 2.74999e-06
[py_interpret_to_execute]: 4.445e-05
[rewriter_before_opt_a]: 7.822e-05
[expand_dump_flag]: 4.15e-06
[jit_opt_a]: 0.007347, [2]
    [Cycle 1]: 0.00249483, [27]
        [switch_simplify]: 7.806e-05
        [loop_unroll]: 2.991e-05
        [a_1]: 0.0008927
        [with_stream_mark]: 2.799e-05
        [recompute_prepare]: 9.22999e-06
        [updatestate_depend_eliminate]: 5.39e-06
        [updatestate_assign_eliminate]: 3.38999e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [parameter_eliminate]: 2.54001e-06
        [specialize_transform]: 7.3e-06
        [updatestate_useless_node_eliminater]: 5.66e-06
        [accelerated_algorithm]: 6.09999e-06
        [meta_shard_fg_expand]: 2.81999e-06
        [get_grad_eliminate_]: 5.60001e-06
        [merge_forward]: 3.91001e-06
        [cell_reuse_recompute_pass]: 1.72999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.646e-05
        [j_node_and_user_rematch]: 1.074e-05
        [meta_fg_expand]: 2.98e-06
        [replace_old_param]: 1.124e-05
        [inline_without_move]: 5.30999e-06
        [renormalize]: 0.00105523
        [add_forward_monad_depend]: 9.35001e-06
        [auto_monad_grad]: 3.04999e-06
        [auto_monad_eliminator]: 1.941e-05
        [cse]: 3.462e-05
        [replace_applicator]: 2.212e-05
    [Cycle 2]: 0.00037461, [27]
        [switch_simplify]: 6.26998e-06
        [loop_unroll]: 5.80002e-06
        [a_1]: 0.00010177
        [with_stream_mark]: 1.678e-05
        [recompute_prepare]: 5.35001e-06
        [updatestate_depend_eliminate]: 4.45e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 4.73001e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 5.84e-06
        [meta_shard_fg_expand]: 2.07001e-06
        [get_grad_eliminate_]: 4.75999e-06
        [merge_forward]: 3.60998e-06
        [cell_reuse_recompute_pass]: 3.19001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.733e-05
        [j_node_and_user_rematch]: 9.12999e-06
        [meta_fg_expand]: 2.06998e-06
        [replace_old_param]: 8.45001e-06
        [inline_without_move]: 4.84003e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 1.09e-06
        [auto_monad_eliminator]: 6.36e-06
        [cse]: 1.338e-05
        [replace_applicator]: 6.66e-06
[py_interpret_to_execute_after_opt_a]: 1.602e-05
[rewriter_after_opt_a]: 4.108e-05
[convert_after_rewriter]: 7.74002e-06
[order_py_execute_after_rewriter]: 5.56e-06
[mutable_eliminate]: 0.00087621
[jit_opt_b]: 5.502e-05, [1]
    [Cycle 1]: 4.589e-05, [2]
        [frontend_op_eliminate]: 1.814e-05
        [inline_after_opt_a]: 1.577e-05
[cconv]: 3.264e-05
[loop_unroll]: 0.0004765
[jit_opt_after_cconv]: 0.00360369, [1]
    [Cycle 1]: 0.00359395, [11]
        [c_1]: 2.197e-05
        [parameter_eliminate]: 4.55001e-06
        [updatestate_depend_eliminate]: 2.222e-05
        [updatestate_assign_eliminate]: 3.36999e-06
        [updatestate_loads_eliminate]: 3.18e-06
        [cse]: 4.364e-05
        [call_graph_tuple_transform]: 5.165e-05
        [tuple_list_get_item_eliminator]: 6.76e-06
        [none_parameter_eliminate]: 5.13002e-06
        [renormalize]: 7.49977e-07
        [switch_simplify]: 6.96001e-06
[remove_dup_value]: 1.829e-05
[partial_unused_args_eliminate]: 2.48e-06
[environ_conv]: 8.80999e-06
[add_recomputation]: 6.264e-05
[cse_after_recomputation]: 2.733e-05, [1]
    [Cycle 1]: 1.775e-05, [1]
        [cse]: 1.095e-05
[auto_monad_reorder]: 1.874e-05
[get_jit_bprop_graph]: 2.32999e-06
[rewriter_after_jit_bprop_graph]: 1.488e-05
[opt_after_jit_grad]: 0.00086076
[symbol_engine_optimizer]: 8.338e-05, [1]
    [Cycle 1]: 7.461e-05, [6]
        [build]: 5.34e-06
        [elim_shapecalc]: 8.48999e-06
        [elim_not_effective]: 1.574e-05
        [opt_reshape]: 6.73998e-06
        [fold_const_symbol]: 9.02999e-06
        [renormalize]: 4.09986e-07
[validate]: 5.133e-05
[backend_pass]: 8.80013e-07
[task_emit]: 0.00731312
[execute]: 1.046e-05
Sums
    bootstrap                            :     0.000377s :  1.44%
    type_inference                       :     0.012994s : 49.58%
    event_method                         :     0.000021s :  0.08%
    auto_monad                           :     0.000075s :  0.29%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.17%
    rewriter_before_opt_a                :     0.000078s :  0.30%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000084s :  0.32%
    jit_opt_a.loop_unroll                :     0.000036s :  0.14%
    jit_opt_a.a_1                        :     0.000994s :  3.79%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000008s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.04%
    jit_opt_a.renormalize                :     0.001055s :  4.03%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.10%
    jit_opt_a.cse                        :     0.000048s :  0.18%
    jit_opt_a.replace_applicator         :     0.000029s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.06%
    rewriter_after_opt_a                 :     0.000041s :  0.16%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000876s :  3.34%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.06%
    cconv                                :     0.000033s :  0.12%
    loop_unroll                          :     0.000477s :  1.82%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000022s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000044s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000052s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000018s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000009s :  0.03%
    add_recomputation                    :     0.000063s :  0.24%
    cse_after_recomputation.cse          :     0.000011s :  0.04%
    auto_monad_reorder                   :     0.000019s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000015s :  0.06%
    opt_after_jit_grad                   :     0.000861s :  3.28%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000051s :  0.20%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007313s : 27.91%
    execute                              :     0.000010s :  0.04%


Time group info:
------[substitution.]   0.000507    23
 0.46% :     0.000002s :      2: substitution.elim_not_effective
 0.34% :     0.000002s :      2: substitution.fold_const_symbol
 1.73% :     0.000009s :      3: substitution.graph_param_transform
91.24% :     0.000462s :      4: substitution.inline
 0.87% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.31% :     0.000007s :      4: substitution.remove_not_recompute_node
 1.27% :     0.000006s :      2: substitution.replace_old_param
 2.78% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012898     2
91.13% :     0.011755s :      1: type_inference.infer
 8.87% :     0.001143s :      1: type_inference.specialize
------[replace.]   0.000092     6
68.30% :     0.000063s :      4: replace.inline
31.70% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000471     6
97.27% :     0.000458s :      4: match.inline
 2.73% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   691
 1.59% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.51% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.89% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000002s :     10: predicate.addn_zero_filter
 2.63% :     0.000004s :     10: predicate.arithmetic_simplify
 1.50% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.89% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 1.23% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.00% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.88% :     0.000003s :     10: predicate.environ_get_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.61% :     0.000001s :      3: predicate.graph_param_transform
 5.95% :     0.000008s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.39% :     0.000002s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.55% :     0.000002s :     15: predicate.load_eliminater
 1.33% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.68% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.10% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.94% :     0.000001s :     10: predicate.merge_addn
 1.42% :     0.000002s :     10: predicate.minmaximum_grad
 2.29% :     0.000003s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     15: predicate.partial_eliminate
 0.92% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.83% :     0.000002s :     10: predicate.reduce_eliminate
 1.26% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.72% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.41% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.02% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.79% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.69% :     0.000004s :     16: predicate.switch_defer_inline
 1.76% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.94% :     0.000011s :     53: predicate.switch_simplify
 1.62% :     0.000002s :     10: predicate.tile_eliminate
 1.14% :     0.000002s :     10: predicate.transpose_eliminate
 1.23% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.09% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.04% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.82% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.42% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.41% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000003s :     10: predicate.value_based_eliminate
 0.44% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000888    12
48.48% :     0.000431s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.85% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.67% :     0.000397s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.037233    76
 0.18% :     0.000066s :      1: add_recomputation
 0.21% :     0.000079s :      1: auto_monad
 0.06% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.06% :     0.000393s :      1: bootstrap
 0.10% :     0.000036s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000030s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.07% :     0.000025s :      1: event_method
 0.04% :     0.000016s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
19.74% :     0.007351s :      1: jit_opt_a
 9.69% :     0.003609s :      1: jit_opt_after_cconv
 0.15% :     0.000058s :      1: jit_opt_b
 1.30% :     0.000484s :      1: loop_unroll
 2.39% :     0.000888s :      1: mutable_eliminate
 3.36% :     0.001250s :     26: opt.transform.jit_opt_a
 0.22% :     0.000081s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000027s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.34% :     0.000870s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000047s :      1: py_interpret_to_execute
 0.05% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000021s :      1: remove_dup_value
 1.53% :     0.000570s :      1: renormalize.infer
 1.27% :     0.000473s :      1: renormalize.specialize
 0.05% :     0.000017s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000045s :      1: rewriter_after_opt_a
 0.22% :     0.000082s :      1: rewriter_before_opt_a
 0.23% :     0.000086s :      1: symbol_engine_optimizer
19.69% :     0.007333s :      1: task_emit
34.97% :     0.013019s :      1: type_inference
 0.20% :     0.000076s :      1: validate


TotalTime = 0.0411285, [33]
[bootstrap]: 0.00036484
[type_inference]: 0.0051215
[event_method]: 1.888e-05
[auto_monad]: 6.798e-05
[graph_reusing]: 5.98998e-06
[pre_auto_parallel]: 2.63e-06
[py_interpret_to_execute]: 4.263e-05
[rewriter_before_opt_a]: 7.472e-05
[expand_dump_flag]: 2.89001e-06
[jit_opt_a]: 0.00615284, [2]
    [Cycle 1]: 0.00206418, [27]
        [switch_simplify]: 6.444e-05
        [loop_unroll]: 3.062e-05
        [a_1]: 0.00064566
        [with_stream_mark]: 2.657e-05
        [recompute_prepare]: 8.22998e-06
        [updatestate_depend_eliminate]: 4.15e-06
        [updatestate_assign_eliminate]: 3.15998e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [parameter_eliminate]: 2.32999e-06
        [specialize_transform]: 6.28e-06
        [updatestate_useless_node_eliminater]: 5.59998e-06
        [accelerated_algorithm]: 5.59998e-06
        [meta_shard_fg_expand]: 2.14e-06
        [get_grad_eliminate_]: 5.18002e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.489e-05
        [j_node_and_user_rematch]: 9.47001e-06
        [meta_fg_expand]: 2.61999e-06
        [replace_old_param]: 9.15001e-06
        [inline_without_move]: 5.16002e-06
        [renormalize]: 0.00093487
        [add_forward_monad_depend]: 7.63999e-06
        [auto_monad_grad]: 2.62001e-06
        [auto_monad_eliminator]: 1.638e-05
        [cse]: 3.074e-05
        [replace_applicator]: 1.469e-05
    [Cycle 2]: 0.00034636, [27]
        [switch_simplify]: 6.68e-06
        [loop_unroll]: 5.15001e-06
        [a_1]: 9.435e-05
        [with_stream_mark]: 1.324e-05
        [recompute_prepare]: 5.77001e-06
        [updatestate_depend_eliminate]: 3.4e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 5.39e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 5.64e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 4.55999e-06
        [merge_forward]: 3.31999e-06
        [cell_reuse_recompute_pass]: 2.78998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.436e-05
        [j_node_and_user_rematch]: 8.69e-06
        [meta_fg_expand]: 1.72001e-06
        [replace_old_param]: 8.38001e-06
        [inline_without_move]: 4.60001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.50999e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 5.26998e-06
        [cse]: 1.169e-05
        [replace_applicator]: 5.50001e-06
[py_interpret_to_execute_after_opt_a]: 1.493e-05
[rewriter_after_opt_a]: 3.873e-05
[convert_after_rewriter]: 6.90002e-06
[order_py_execute_after_rewriter]: 4.87e-06
[mutable_eliminate]: 0.00075574
[jit_opt_b]: 5.748e-05, [1]
    [Cycle 1]: 4.813e-05, [2]
        [frontend_op_eliminate]: 1.931e-05
        [inline_after_opt_a]: 1.68e-05
[cconv]: 3.218e-05
[loop_unroll]: 0.00046422
[jit_opt_after_cconv]: 0.00016186, [1]
    [Cycle 1]: 0.00015476, [11]
        [c_1]: 2.352e-05
        [parameter_eliminate]: 4.50999e-06
        [updatestate_depend_eliminate]: 7.21001e-06
        [updatestate_assign_eliminate]: 2.39999e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [cse]: 2.67e-05
        [call_graph_tuple_transform]: 2.326e-05
        [tuple_list_get_item_eliminator]: 5.49e-06
        [none_parameter_eliminate]: 1.77001e-06
        [renormalize]: 6.69999e-07
        [switch_simplify]: 5.21998e-06
[remove_dup_value]: 1.445e-05
[partial_unused_args_eliminate]: 2.29001e-06
[environ_conv]: 5.57001e-06
[add_recomputation]: 5.854e-05
[cse_after_recomputation]: 2.063e-05, [1]
    [Cycle 1]: 1.465e-05, [1]
        [cse]: 9.49e-06
[auto_monad_reorder]: 1.514e-05
[get_jit_bprop_graph]: 2.24999e-06
[rewriter_after_jit_bprop_graph]: 8.03999e-06
[opt_after_jit_grad]: 0.00050076
[symbol_engine_optimizer]: 0.0202682, [1]
    [Cycle 1]: 0.0202539, [6]
        [build]: 5.48002e-06
        [elim_shapecalc]: 8.13999e-06
        [elim_not_effective]: 0.0201234
        [opt_reshape]: 1.701e-05
        [fold_const_symbol]: 1.266e-05
        [renormalize]: 7.89994e-07
[validate]: 7.137e-05
[backend_pass]: 1.13001e-06
[task_emit]: 0.00650053
[execute]: 1.024e-05
Sums
    bootstrap                            :     0.000365s :  1.00%
    type_inference                       :     0.005121s : 13.99%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000068s :  0.19%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.12%
    rewriter_before_opt_a                :     0.000075s :  0.20%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000071s :  0.19%
    jit_opt_a.loop_unroll                :     0.000036s :  0.10%
    jit_opt_a.a_1                        :     0.000740s :  2.02%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000018s :  0.05%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.000935s :  2.55%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.06%
    jit_opt_a.cse                        :     0.000042s :  0.12%
    jit_opt_a.replace_applicator         :     0.000020s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.04%
    rewriter_after_opt_a                 :     0.000039s :  0.11%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000756s :  2.06%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.05%
    cconv                                :     0.000032s :  0.09%
    loop_unroll                          :     0.000464s :  1.27%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.01%
    remove_dup_value                     :     0.000014s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000059s :  0.16%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000015s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000501s :  1.37%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.020123s : 54.96%
    symbol_engine_optimizer.opt_reshape  :     0.000017s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000071s :  0.19%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006501s : 17.75%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.000266    23
 1.58% :     0.000004s :      2: substitution.elim_not_effective
 0.79% :     0.000002s :      2: substitution.fold_const_symbol
 2.40% :     0.000006s :      3: substitution.graph_param_transform
85.19% :     0.000226s :      4: substitution.inline
 1.50% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.87% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.02% :     0.000005s :      2: substitution.replace_old_param
 4.65% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005040     2
82.51% :     0.004159s :      1: type_inference.infer
17.49% :     0.000882s :      1: type_inference.specialize
------[replace.]   0.000078     6
66.89% :     0.000052s :      4: replace.inline
33.11% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000235     6
95.22% :     0.000224s :      4: match.inline
 4.78% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.32% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.21% :     0.000002s :     10: predicate.addn_zero_filter
 2.12% :     0.000003s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.62% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.06% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.05% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 1.47% :     0.000002s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.40% :     0.000002s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.81% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 6.46% :     0.000008s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.97% :     0.000001s :      6: predicate.less_batch_normalization
 1.31% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.52% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.57% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 2.81% :     0.000004s :      3: predicate.mutable_eliminate
 3.18% :     0.000004s :      3: predicate.opt_reshape
 2.05% :     0.000003s :     15: predicate.partial_eliminate
 1.04% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.38% :     0.000002s :     10: predicate.reduce_eliminate
 1.85% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.49% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.16% :     0.000001s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 1.90% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.79% :     0.000010s :     53: predicate.switch_simplify
 1.47% :     0.000002s :     10: predicate.tile_eliminate
 1.26% :     0.000002s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.43% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.43% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.48% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000702    12
49.84% :     0.000350s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.77% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.39% :     0.000297s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.043217    76
 0.14% :     0.000062s :      1: add_recomputation
 0.17% :     0.000072s :      1: auto_monad
 0.04% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.89% :     0.000386s :      1: bootstrap
 0.08% :     0.000035s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.05% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.05% :     0.000023s :      1: event_method
 0.04% :     0.000016s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
14.25% :     0.006157s :      1: jit_opt_a
 0.38% :     0.000165s :      1: jit_opt_after_cconv
 0.14% :     0.000060s :      1: jit_opt_b
 1.10% :     0.000474s :      1: loop_unroll
 1.78% :     0.000768s :      1: mutable_eliminate
 2.24% :     0.000966s :     26: opt.transform.jit_opt_a
 0.12% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000028s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.27% :     0.000115s :      4: opt.transform.symbol_engine_opt
 1.18% :     0.000510s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.11% :     0.000046s :      1: py_interpret_to_execute
 0.04% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000017s :      1: remove_dup_value
 1.20% :     0.000517s :      1: renormalize.infer
 0.94% :     0.000408s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000042s :      1: rewriter_after_opt_a
 0.18% :     0.000078s :      1: rewriter_before_opt_a
46.91% :     0.020275s :      1: symbol_engine_optimizer
15.09% :     0.006519s :      1: task_emit
11.90% :     0.005142s :      1: type_inference
 0.24% :     0.000105s :      1: validate


TotalTime = 0.0222787, [33]
[bootstrap]: 0.00043429
[type_inference]: 0.00601183
[event_method]: 2.205e-05
[auto_monad]: 7.482e-05
[graph_reusing]: 6.88e-06
[pre_auto_parallel]: 2.98e-06
[py_interpret_to_execute]: 5.782e-05
[rewriter_before_opt_a]: 8.481e-05
[expand_dump_flag]: 3.63999e-06
[jit_opt_a]: 0.00650364, [2]
    [Cycle 1]: 0.00192967, [27]
        [switch_simplify]: 6.403e-05
        [loop_unroll]: 3.076e-05
        [a_1]: 0.00059984
        [with_stream_mark]: 2.489e-05
        [recompute_prepare]: 8.64e-06
        [updatestate_depend_eliminate]: 3.62002e-06
        [updatestate_assign_eliminate]: 3.06001e-06
        [updatestate_loads_eliminate]: 2.74001e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 6.12999e-06
        [updatestate_useless_node_eliminater]: 5.49998e-06
        [accelerated_algorithm]: 5.94e-06
        [meta_shard_fg_expand]: 2.34001e-06
        [get_grad_eliminate_]: 5.42001e-06
        [merge_forward]: 3.83001e-06
        [cell_reuse_recompute_pass]: 1.99e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.552e-05
        [j_node_and_user_rematch]: 9.12001e-06
        [meta_fg_expand]: 2.14999e-06
        [replace_old_param]: 9.39e-06
        [inline_without_move]: 4.98001e-06
        [renormalize]: 0.00086802
        [add_forward_monad_depend]: 6.38e-06
        [auto_monad_grad]: 2.90002e-06
        [auto_monad_eliminator]: 1.415e-05
        [cse]: 3.072e-05
        [replace_applicator]: 1.318e-05
    [Cycle 2]: 0.00033997, [27]
        [switch_simplify]: 6.23e-06
        [loop_unroll]: 5.02999e-06
        [a_1]: 9.538e-05
        [with_stream_mark]: 1.021e-05
        [recompute_prepare]: 5.27999e-06
        [updatestate_depend_eliminate]: 2.88998e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [parameter_eliminate]: 1.33002e-06
        [specialize_transform]: 5.92999e-06
        [updatestate_useless_node_eliminater]: 5.18002e-06
        [accelerated_algorithm]: 5.52001e-06
        [meta_shard_fg_expand]: 1.31998e-06
        [get_grad_eliminate_]: 4.77e-06
        [merge_forward]: 2.52001e-06
        [cell_reuse_recompute_pass]: 2.26e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.29e-05
        [j_node_and_user_rematch]: 8.77e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 7.1e-06
        [inline_without_move]: 4.82998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.01e-06
        [auto_monad_grad]: 1.24e-06
        [auto_monad_eliminator]: 5.34998e-06
        [cse]: 1.036e-05
        [replace_applicator]: 5.07e-06
[py_interpret_to_execute_after_opt_a]: 1.12e-05
[rewriter_after_opt_a]: 3.9e-05
[convert_after_rewriter]: 7.18998e-06
[order_py_execute_after_rewriter]: 5.52001e-06
[mutable_eliminate]: 0.00071023
[jit_opt_b]: 5.331e-05, [1]
    [Cycle 1]: 4.518e-05, [2]
        [frontend_op_eliminate]: 1.697e-05
        [inline_after_opt_a]: 1.608e-05
[cconv]: 2.753e-05
[loop_unroll]: 0.00044598
[jit_opt_after_cconv]: 0.00014377, [1]
    [Cycle 1]: 0.00013736, [11]
        [c_1]: 2.062e-05
        [parameter_eliminate]: 3.13998e-06
        [updatestate_depend_eliminate]: 6.30002e-06
        [updatestate_assign_eliminate]: 2.83003e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [cse]: 1.952e-05
        [call_graph_tuple_transform]: 1.999e-05
        [tuple_list_get_item_eliminator]: 5.48002e-06
        [none_parameter_eliminate]: 1.94999e-06
        [renormalize]: 8.00006e-07
        [switch_simplify]: 5.19e-06
[remove_dup_value]: 1.308e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 5.59e-06
[add_recomputation]: 5.162e-05
[cse_after_recomputation]: 1.991e-05, [1]
    [Cycle 1]: 1.454e-05, [1]
        [cse]: 8.87e-06
[auto_monad_reorder]: 1.624e-05
[get_jit_bprop_graph]: 2.08002e-06
[rewriter_after_jit_bprop_graph]: 6.31998e-06
[opt_after_jit_grad]: 0.00046864
[symbol_engine_optimizer]: 7.477e-05, [1]
    [Cycle 1]: 6.809e-05, [6]
        [build]: 4.55001e-06
        [elim_shapecalc]: 7.85e-06
        [elim_not_effective]: 1.268e-05
        [opt_reshape]: 5.84999e-06
        [fold_const_symbol]: 8.85999e-06
        [renormalize]: 4.19997e-07
[validate]: 4.186e-05
[backend_pass]: 8.59989e-07
[task_emit]: 0.00667356
[execute]: 9.76998e-06
Sums
    bootstrap                            :     0.000434s :  2.50%
    type_inference                       :     0.006012s : 34.60%
    event_method                         :     0.000022s :  0.13%
    auto_monad                           :     0.000075s :  0.43%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000058s :  0.33%
    rewriter_before_opt_a                :     0.000085s :  0.49%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000070s :  0.40%
    jit_opt_a.loop_unroll                :     0.000036s :  0.21%
    jit_opt_a.a_1                        :     0.000695s :  4.00%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000868s :  5.00%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.11%
    jit_opt_a.cse                        :     0.000041s :  0.24%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000039s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000710s :  4.09%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000028s :  0.16%
    loop_unroll                          :     0.000446s :  2.57%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000052s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000469s :  2.70%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000042s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006674s : 38.41%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000257    23
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.25% :     0.000006s :      3: substitution.graph_param_transform
86.13% :     0.000221s :      4: substitution.inline
 1.65% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.76% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.75% :     0.000004s :      2: substitution.replace_old_param
 5.20% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005920     2
82.29% :     0.004872s :      1: type_inference.infer
17.71% :     0.001049s :      1: type_inference.specialize
------[replace.]   0.000072     6
67.75% :     0.000049s :      4: replace.inline
32.25% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000231     6
94.72% :     0.000219s :      4: match.inline
 5.28% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.43% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.59% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 1.96% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000002s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.49% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 6.25% :     0.000008s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.54% :     0.000002s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 1.28% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.65% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.15% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 1.66% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.65% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     10: predicate.same_eliminate
 0.68% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.23% :     0.000002s :      6: predicate.special_op_eliminate
 1.12% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.36% :     0.000011s :     53: predicate.switch_simplify
 1.48% :     0.000002s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.55% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.94% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000790    12
47.27% :     0.000374s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.78% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.95% :     0.000363s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024162    76
 0.23% :     0.000055s :      1: add_recomputation
 0.33% :     0.000080s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.89% :     0.000457s :      1: bootstrap
 0.13% :     0.000031s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.11% :     0.000027s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.94% :     0.006508s :      1: jit_opt_a
 0.61% :     0.000147s :      1: jit_opt_after_cconv
 0.23% :     0.000056s :      1: jit_opt_b
 1.88% :     0.000455s :      1: loop_unroll
 2.98% :     0.000720s :      1: mutable_eliminate
 3.81% :     0.000920s :     26: opt.transform.jit_opt_a
 0.20% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.97% :     0.000476s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000006s :      1: pre_auto_parallel
 0.25% :     0.000061s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000015s :      1: remove_dup_value
 2.10% :     0.000508s :      1: renormalize.infer
 1.46% :     0.000352s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000042s :      1: rewriter_after_opt_a
 0.37% :     0.000089s :      1: rewriter_before_opt_a
 0.32% :     0.000078s :      1: symbol_engine_optimizer
27.71% :     0.006694s :      1: task_emit
24.98% :     0.006035s :      1: type_inference
 0.26% :     0.000063s :      1: validate


TotalTime = 0.0237458, [33]
[bootstrap]: 0.00040937
[type_inference]: 0.00539859
[event_method]: 1.8e-05
[auto_monad]: 6.745e-05
[graph_reusing]: 6.54999e-06
[pre_auto_parallel]: 2.56e-06
[py_interpret_to_execute]: 3.975e-05
[rewriter_before_opt_a]: 7.53e-05
[expand_dump_flag]: 3.68e-06
[jit_opt_a]: 0.0082813, [2]
    [Cycle 1]: 0.00394971, [27]
        [switch_simplify]: 6.655e-05
        [loop_unroll]: 3.165e-05
        [a_1]: 0.0023652
        [with_stream_mark]: 3.118e-05
        [recompute_prepare]: 1.11e-05
        [updatestate_depend_eliminate]: 4.34002e-06
        [updatestate_assign_eliminate]: 2.98998e-06
        [updatestate_loads_eliminate]: 2.84001e-06
        [parameter_eliminate]: 2.64001e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.91998e-06
        [meta_shard_fg_expand]: 2.56e-06
        [get_grad_eliminate_]: 5.53002e-06
        [merge_forward]: 4.57e-06
        [cell_reuse_recompute_pass]: 4.87e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.529e-05
        [j_node_and_user_rematch]: 1.06e-05
        [meta_fg_expand]: 2.69999e-06
        [replace_old_param]: 1.075e-05
        [inline_without_move]: 5.35999e-06
        [renormalize]: 0.00105809
        [add_forward_monad_depend]: 6.51e-06
        [auto_monad_grad]: 2.98e-06
        [auto_monad_eliminator]: 1.918e-05
        [cse]: 3.252e-05
        [replace_applicator]: 1.549e-05
    [Cycle 2]: 0.00033876, [27]
        [switch_simplify]: 6.26998e-06
        [loop_unroll]: 5.54e-06
        [a_1]: 9.509e-05
        [with_stream_mark]: 1.273e-05
        [recompute_prepare]: 4.94998e-06
        [updatestate_depend_eliminate]: 3.18e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [parameter_eliminate]: 1.41002e-06
        [specialize_transform]: 5.02999e-06
        [updatestate_useless_node_eliminater]: 4.73001e-06
        [accelerated_algorithm]: 6.05002e-06
        [meta_shard_fg_expand]: 1.52001e-06
        [get_grad_eliminate_]: 4.72998e-06
        [merge_forward]: 2.69999e-06
        [cell_reuse_recompute_pass]: 2.49999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.432e-05
        [j_node_and_user_rematch]: 8.03999e-06
        [meta_fg_expand]: 2.03002e-06
        [replace_old_param]: 7.48e-06
        [inline_without_move]: 1.115e-05
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.09998e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 5.35001e-06
        [cse]: 1.122e-05
        [replace_applicator]: 4.99003e-06
[py_interpret_to_execute_after_opt_a]: 1.491e-05
[rewriter_after_opt_a]: 4.137e-05
[convert_after_rewriter]: 7.20003e-06
[order_py_execute_after_rewriter]: 6.06e-06
[mutable_eliminate]: 0.00089805
[jit_opt_b]: 5.401e-05, [1]
    [Cycle 1]: 4.601e-05, [2]
        [frontend_op_eliminate]: 1.845e-05
        [inline_after_opt_a]: 1.561e-05
[cconv]: 3.113e-05
[loop_unroll]: 0.00046943
[jit_opt_after_cconv]: 0.00015853, [1]
    [Cycle 1]: 0.00015079, [11]
        [c_1]: 2.239e-05
        [parameter_eliminate]: 4.25999e-06
        [updatestate_depend_eliminate]: 7.34002e-06
        [updatestate_assign_eliminate]: 2.63e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [cse]: 2.724e-05
        [call_graph_tuple_transform]: 2.049e-05
        [tuple_list_get_item_eliminator]: 5.30999e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 1.25999e-06
        [switch_simplify]: 6.16998e-06
[remove_dup_value]: 1.611e-05
[partial_unused_args_eliminate]: 2.28998e-06
[environ_conv]: 6.07999e-06
[add_recomputation]: 5.145e-05
[cse_after_recomputation]: 2.193e-05, [1]
    [Cycle 1]: 1.581e-05, [1]
        [cse]: 1.01e-05
[auto_monad_reorder]: 1.627e-05
[get_jit_bprop_graph]: 2.02999e-06
[rewriter_after_jit_bprop_graph]: 6.36998e-06
[opt_after_jit_grad]: 0.00048803
[symbol_engine_optimizer]: 7.864e-05, [1]
    [Cycle 1]: 7.151e-05, [6]
        [build]: 5.37999e-06
        [elim_shapecalc]: 8.62998e-06
        [elim_not_effective]: 1.35e-05
        [opt_reshape]: 6.34999e-06
        [fold_const_symbol]: 9.04e-06
        [renormalize]: 8.99978e-07
[validate]: 4.373e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00678976
[execute]: 9.80002e-06
Sums
    bootstrap                            :     0.000409s :  2.15%
    type_inference                       :     0.005399s : 28.31%
    event_method                         :     0.000018s :  0.09%
    auto_monad                           :     0.000067s :  0.35%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000040s :  0.21%
    rewriter_before_opt_a                :     0.000075s :  0.39%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000073s :  0.38%
    jit_opt_a.loop_unroll                :     0.000037s :  0.20%
    jit_opt_a.a_1                        :     0.002460s : 12.90%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000017s :  0.09%
    jit_opt_a.renormalize                :     0.001058s :  5.55%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.13%
    jit_opt_a.cse                        :     0.000044s :  0.23%
    jit_opt_a.replace_applicator         :     0.000020s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.08%
    rewriter_after_opt_a                 :     0.000041s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000898s :  4.71%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.000031s :  0.16%
    loop_unroll                          :     0.000469s :  2.46%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000051s :  0.27%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000488s :  2.56%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.23%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006790s : 35.60%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.001951    23
 0.10% :     0.000002s :      2: substitution.elim_not_effective
 0.09% :     0.000002s :      2: substitution.fold_const_symbol
 0.33% :     0.000006s :      3: substitution.graph_param_transform
11.37% :     0.000222s :      4: substitution.inline
 0.18% :     0.000004s :      4: substitution.j_node_and_user_rematch
 0.27% :     0.000005s :      4: substitution.remove_not_recompute_node
 0.31% :     0.000006s :      2: substitution.replace_old_param
87.35% :     0.001704s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005318     2
81.69% :     0.004345s :      1: type_inference.infer
18.31% :     0.000974s :      1: type_inference.specialize
------[replace.]   0.000107     6
46.55% :     0.000050s :      4: replace.inline
53.45% :     0.000057s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.001919     6
11.41% :     0.000219s :      4: match.inline
88.59% :     0.001700s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   691
 1.49% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 2.41% :     0.000003s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.80% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.74% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.93% :     0.000008s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.49% :     0.000002s :      6: predicate.less_batch_normalization
 1.70% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.58% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.82% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.26% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 0.91% :     0.000001s :     10: predicate.minmaximum_grad
 1.68% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.79% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.75% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.45% :     0.000001s :      3: predicate.reset_defer_inline
 1.18% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.15% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.94% :     0.000009s :     53: predicate.switch_simplify
 1.54% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000002s :     10: predicate.transpose_eliminate
 1.79% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.37% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.35% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.86% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.80% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.47% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.64% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.48% :     0.000002s :     10: predicate.value_based_eliminate
 0.57% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000705    12
43.11% :     0.000304s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.44% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.44% :     0.000349s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027610    76
 0.20% :     0.000054s :      1: add_recomputation
 0.26% :     0.000071s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.55% :     0.000428s :      1: bootstrap
 0.12% :     0.000034s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.08% :     0.000022s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
30.01% :     0.008286s :      1: jit_opt_a
 0.58% :     0.000161s :      1: jit_opt_after_cconv
 0.21% :     0.000057s :      1: jit_opt_b
 1.73% :     0.000479s :      1: loop_unroll
 3.30% :     0.000910s :      1: mutable_eliminate
 9.78% :     0.002700s :     26: opt.transform.jit_opt_a
 0.18% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.80% :     0.000496s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000043s :      1: py_interpret_to_execute
 0.06% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 2.10% :     0.000579s :      1: renormalize.infer
 1.70% :     0.000468s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000045s :      1: rewriter_after_opt_a
 0.28% :     0.000078s :      1: rewriter_before_opt_a
 0.29% :     0.000081s :      1: symbol_engine_optimizer
24.66% :     0.006810s :      1: task_emit
19.62% :     0.005417s :      1: type_inference
 0.23% :     0.000065s :      1: validate


TotalTime = 0.0393022, [30]
[bootstrap]: 0.00034944
[type_inference]: 0.0287934
[event_method]: 2.584e-05
[auto_monad]: 8.843e-05
[graph_reusing]: 6.69999e-06
[pre_auto_parallel]: 2.51e-06
[py_interpret_to_execute]: 5.94e-05
[rewriter_before_opt_a]: 8.876e-05
[expand_dump_flag]: 2.98e-06
[jit_opt_a]: 0.0074619, [2]
    [Cycle 1]: 0.00245931, [27]
        [switch_simplify]: 0.00014581
        [loop_unroll]: 9.703e-05
        [a_1]: 0.0008116
        [with_stream_mark]: 2.75e-05
        [recompute_prepare]: 8.12998e-06
        [updatestate_depend_eliminate]: 4.28999e-06
        [updatestate_assign_eliminate]: 3.11001e-06
        [updatestate_loads_eliminate]: 3.13e-06
        [parameter_eliminate]: 3.04999e-06
        [specialize_transform]: 6.74999e-06
        [updatestate_useless_node_eliminater]: 5.59e-06
        [accelerated_algorithm]: 6.31998e-06
        [meta_shard_fg_expand]: 2.09e-06
        [get_grad_eliminate_]: 5.76e-06
        [merge_forward]: 3.79002e-06
        [cell_reuse_recompute_pass]: 1.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.557e-05
        [j_node_and_user_rematch]: 9.67001e-06
        [meta_fg_expand]: 2.44001e-06
        [replace_old_param]: 9.97001e-06
        [inline_without_move]: 5.56e-06
        [renormalize]: 0.00098366
        [add_forward_monad_depend]: 6.84999e-06
        [auto_monad_grad]: 2.58998e-06
        [auto_monad_eliminator]: 1.576e-05
        [cse]: 3.019e-05
        [replace_applicator]: 1.494e-05
    [Cycle 2]: 0.00034769, [27]
        [switch_simplify]: 6.46e-06
        [loop_unroll]: 4.83001e-06
        [a_1]: 9.174e-05
        [with_stream_mark]: 1.017e-05
        [recompute_prepare]: 5.05001e-06
        [updatestate_depend_eliminate]: 2.69001e-06
        [updatestate_assign_eliminate]: 1.91e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 4.70999e-06
        [updatestate_useless_node_eliminater]: 4.68999e-06
        [accelerated_algorithm]: 4.74e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 4.71002e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.94e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.269e-05
        [j_node_and_user_rematch]: 8.12e-06
        [meta_fg_expand]: 1.37999e-06
        [replace_old_param]: 7.23e-06
        [inline_without_move]: 4.77e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 1.72001e-06
        [auto_monad_eliminator]: 5.15999e-06
        [cse]: 1.045e-05
        [replace_applicator]: 6.02001e-06
[py_interpret_to_execute_after_opt_a]: 1.166e-05
[rewriter_after_opt_a]: 5.631e-05
[convert_after_rewriter]: 7.11999e-06
[order_py_execute_after_rewriter]: 5.10999e-06
[mutable_eliminate]: 0.00069319
[jit_opt_b]: 6.197e-05, [1]
    [Cycle 1]: 4.465e-05, [2]
        [frontend_op_eliminate]: 1.785e-05
        [inline_after_opt_a]: 1.547e-05
[cconv]: 2.865e-05
[loop_unroll]: 0.00043565
[jit_opt_after_cconv]: 0.00017326, [1]
    [Cycle 1]: 0.00016599, [11]
        [c_1]: 2.221e-05
        [parameter_eliminate]: 3.2e-06
        [updatestate_depend_eliminate]: 6.46e-06
        [updatestate_assign_eliminate]: 2.71999e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [cse]: 1.993e-05
        [call_graph_tuple_transform]: 1.894e-05
        [tuple_list_get_item_eliminator]: 5.57001e-06
        [none_parameter_eliminate]: 1.99e-06
        [renormalize]: 5.79981e-07
        [switch_simplify]: 5.25999e-06
[remove_dup_value]: 1.33e-05
[partial_unused_args_eliminate]: 2.18002e-06
[environ_conv]: 5.86e-06
[add_recomputation]: 5.545e-05
[cse_after_recomputation]: 2.161e-05, [1]
    [Cycle 1]: 1.521e-05, [1]
        [cse]: 9.15999e-06
[auto_monad_reorder]: 1.604e-05
[get_jit_bprop_graph]: 2.21998e-06
[rewriter_after_jit_bprop_graph]: 5.86e-06
[opt_after_jit_grad]: 0.00051516
[symbol_engine_optimizer]: 7.873e-05, [1]
    [Cycle 1]: 7.211e-05, [6]
        [build]: 4.12e-06
        [elim_shapecalc]: 8.25e-06
        [elim_not_effective]: 1.397e-05
        [opt_reshape]: 5.44e-06
        [fold_const_symbol]: 9.12999e-06
        [renormalize]: 5.39992e-07
[validate]: 3.799e-05
Sums
    bootstrap                            :     0.000349s :  1.03%
    type_inference                       :     0.028793s : 84.88%
    event_method                         :     0.000026s :  0.08%
    auto_monad                           :     0.000088s :  0.26%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000059s :  0.18%
    rewriter_before_opt_a                :     0.000089s :  0.26%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000152s :  0.45%
    jit_opt_a.loop_unroll                :     0.000102s :  0.30%
    jit_opt_a.a_1                        :     0.000903s :  2.66%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000017s :  0.05%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.000984s :  2.90%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.06%
    jit_opt_a.cse                        :     0.000041s :  0.12%
    jit_opt_a.replace_applicator         :     0.000021s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000056s :  0.17%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000693s :  2.04%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.05%
    cconv                                :     0.000029s :  0.08%
    loop_unroll                          :     0.000436s :  1.28%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000013s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000055s :  0.16%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000515s :  1.52%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.11%


Time group info:
------[substitution.]   0.000367    26
 0.55% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000002s :      2: substitution.fold_const_symbol
 1.37% :     0.000005s :      3: substitution.graph_param_transform
87.17% :     0.000320s :      6: substitution.inline
 0.84% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.22% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.42% :     0.000005s :      2: substitution.replace_old_param
 2.75% :     0.000010s :      1: substitution.switch_simplify
 4.09% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.028697     2
94.81% :     0.027209s :      1: type_inference.infer
 5.19% :     0.001488s :      1: type_inference.specialize
------[replace.]   0.000164     9
48.19% :     0.000079s :      6: replace.inline
28.78% :     0.000047s :      1: replace.switch_simplify
23.03% :     0.000038s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000339     9
93.21% :     0.000316s :      6: match.inline
 2.70% :     0.000009s :      1: match.switch_simplify
 4.08% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000142   801
 1.42% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000002s :     12: predicate.addn_check_dump
 1.43% :     0.000002s :     12: predicate.addn_zero_filter
 4.12% :     0.000006s :     12: predicate.arithmetic_simplify
 1.27% :     0.000002s :     12: predicate.cast_eliminate
 0.30% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     12: predicate.compare_switch_simplify
 1.45% :     0.000002s :     12: predicate.depend_value_elim
 1.11% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.38% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     12: predicate.environ_get_add_eliminate
 1.06% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.12% :     0.000002s :     12: predicate.environ_get_eliminate
 1.08% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.58% :     0.000008s :     26: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.61% :     0.000002s :     17: predicate.load_eliminater
 1.25% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.88% :     0.000005s :     40: predicate.loop_unroll_before_grad
 1.88% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000002s :     12: predicate.merge_addn
 1.10% :     0.000002s :     12: predicate.minmaximum_grad
 2.12% :     0.000003s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000003s :     17: predicate.partial_eliminate
 1.39% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     12: predicate.reduce_eliminate
 1.36% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.52% :     0.000002s :     20: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000001s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     12: predicate.reshape_eliminate
 1.16% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 1.03% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     12: predicate.same_eliminate
 0.42% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 0.82% :     0.000001s :      6: predicate.specialize_transform
 1.57% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.72% :     0.000004s :     20: predicate.switch_defer_inline
 2.36% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.18% :     0.000012s :     65: predicate.switch_simplify
 1.40% :     0.000002s :     12: predicate.tile_eliminate
 1.36% :     0.000002s :     12: predicate.transpose_eliminate
 1.21% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.50% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     12: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001085    15
45.17% :     0.000490s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.55% :     0.000136s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.28% :     0.000459s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.041613    72
 0.14% :     0.000058s :      1: add_recomputation
 0.22% :     0.000092s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.89% :     0.000369s :      1: bootstrap
 0.08% :     0.000032s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.06% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.07% :     0.000030s :      1: event_method
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
17.94% :     0.007467s :      1: jit_opt_a
 0.42% :     0.000176s :      1: jit_opt_after_cconv
 0.16% :     0.000065s :      1: jit_opt_b
 1.06% :     0.000443s :      1: loop_unroll
 1.69% :     0.000702s :      1: mutable_eliminate
 3.07% :     0.001277s :     26: opt.transform.jit_opt_a
 0.12% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000027s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.26% :     0.000525s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.15% :     0.000063s :      1: py_interpret_to_execute
 0.03% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000016s :      1: remove_dup_value
 1.41% :     0.000585s :      1: renormalize.infer
 0.94% :     0.000390s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000060s :      1: rewriter_after_opt_a
 0.22% :     0.000092s :      1: rewriter_before_opt_a
 0.20% :     0.000081s :      1: symbol_engine_optimizer
69.25% :     0.028816s :      1: type_inference


TotalTime = 0.027685, [33]
[bootstrap]: 0.00038442
[type_inference]: 0.00528951
[event_method]: 1.98e-05
[auto_monad]: 5.858e-05
[graph_reusing]: 4.85001e-06
[pre_auto_parallel]: 2.27001e-06
[py_interpret_to_execute]: 4.694e-05
[rewriter_before_opt_a]: 7.659e-05
[expand_dump_flag]: 2.86e-06
[jit_opt_a]: 0.00626128, [2]
    [Cycle 1]: 0.00191407, [27]
        [switch_simplify]: 5.074e-05
        [loop_unroll]: 2.958e-05
        [a_1]: 0.00054907
        [with_stream_mark]: 2.146e-05
        [recompute_prepare]: 8.67e-06
        [updatestate_depend_eliminate]: 3.83999e-06
        [updatestate_assign_eliminate]: 2.31998e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.52001e-06
        [specialize_transform]: 6.21e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.55001e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 5.87001e-06
        [merge_forward]: 3.76999e-06
        [cell_reuse_recompute_pass]: 1.54998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.284e-05
        [j_node_and_user_rematch]: 8.90001e-06
        [meta_fg_expand]: 2.09999e-06
        [replace_old_param]: 8.1e-06
        [inline_without_move]: 5.56e-06
        [renormalize]: 0.00091042
        [add_forward_monad_depend]: 8.80001e-06
        [auto_monad_grad]: 2.32999e-06
        [auto_monad_eliminator]: 1.784e-05
        [cse]: 3.069e-05
        [replace_applicator]: 1.661e-05
    [Cycle 2]: 0.00034388, [27]
        [switch_simplify]: 6.46e-06
        [loop_unroll]: 5.17999e-06
        [a_1]: 9.541e-05
        [with_stream_mark]: 1.191e-05
        [recompute_prepare]: 5.04998e-06
        [updatestate_depend_eliminate]: 2.79001e-06
        [updatestate_assign_eliminate]: 2.66999e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [parameter_eliminate]: 1.36998e-06
        [specialize_transform]: 5.29e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.37999e-06
        [meta_shard_fg_expand]: 1.34998e-06
        [get_grad_eliminate_]: 4.93001e-06
        [merge_forward]: 2.68998e-06
        [cell_reuse_recompute_pass]: 2.93998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.721e-05
        [j_node_and_user_rematch]: 8.15e-06
        [meta_fg_expand]: 1.72999e-06
        [replace_old_param]: 7.11999e-06
        [inline_without_move]: 4.63999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.36002e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 5.47999e-06
        [cse]: 1.024e-05
        [replace_applicator]: 5.05001e-06
[py_interpret_to_execute_after_opt_a]: 1.652e-05
[rewriter_after_opt_a]: 4.211e-05
[convert_after_rewriter]: 7.51999e-06
[order_py_execute_after_rewriter]: 5.32001e-06
[mutable_eliminate]: 0.00089113
[jit_opt_b]: 5.721e-05, [1]
    [Cycle 1]: 4.809e-05, [2]
        [frontend_op_eliminate]: 1.926e-05
        [inline_after_opt_a]: 1.691e-05
[cconv]: 3.024e-05
[loop_unroll]: 0.0004663
[jit_opt_after_cconv]: 0.00015889, [1]
    [Cycle 1]: 0.00015211, [11]
        [c_1]: 2.103e-05
        [parameter_eliminate]: 5.02999e-06
        [updatestate_depend_eliminate]: 8.15e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 2.643e-05
        [call_graph_tuple_transform]: 2.248e-05
        [tuple_list_get_item_eliminator]: 5.42001e-06
        [none_parameter_eliminate]: 1.41998e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 5.74e-06
[remove_dup_value]: 1.436e-05
[partial_unused_args_eliminate]: 2.98e-06
[environ_conv]: 6.25002e-06
[add_recomputation]: 5.132e-05
[cse_after_recomputation]: 2.19e-05, [1]
    [Cycle 1]: 1.578e-05, [1]
        [cse]: 8.86002e-06
[auto_monad_reorder]: 1.538e-05
[get_jit_bprop_graph]: 2.26e-06
[rewriter_after_jit_bprop_graph]: 7.3e-06
[opt_after_jit_grad]: 0.00049005
[symbol_engine_optimizer]: 7.822e-05, [1]
    [Cycle 1]: 6.999e-05, [6]
        [build]: 3.81999e-06
        [elim_shapecalc]: 7.4e-06
        [elim_not_effective]: 1.359e-05
        [opt_reshape]: 6.17999e-06
        [fold_const_symbol]: 1.042e-05
        [renormalize]: 5.59987e-07
[validate]: 3.95e-05
[backend_pass]: 1.15999e-06
[task_emit]: 0.0128864
[execute]: 1.063e-05
Sums
    bootstrap                            :     0.000384s :  1.67%
    type_inference                       :     0.005290s : 22.99%
    event_method                         :     0.000020s :  0.09%
    auto_monad                           :     0.000059s :  0.25%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000047s :  0.20%
    rewriter_before_opt_a                :     0.000077s :  0.33%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000057s :  0.25%
    jit_opt_a.loop_unroll                :     0.000035s :  0.15%
    jit_opt_a.a_1                        :     0.000644s :  2.80%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.07%
    jit_opt_a.inline_without_move        :     0.000010s :  0.04%
    jit_opt_a.renormalize                :     0.000911s :  3.96%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.10%
    jit_opt_a.cse                        :     0.000041s :  0.18%
    jit_opt_a.replace_applicator         :     0.000022s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.07%
    rewriter_after_opt_a                 :     0.000042s :  0.18%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000891s :  3.87%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.07%
    cconv                                :     0.000030s :  0.13%
    loop_unroll                          :     0.000466s :  2.03%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000014s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000051s :  0.22%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000015s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.03%
    opt_after_jit_grad                   :     0.000490s :  2.13%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.17%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.012886s : 56.01%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000210    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 3.08% :     0.000006s :      3: substitution.graph_param_transform
85.26% :     0.000179s :      4: substitution.inline
 1.43% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.45% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.68% :     0.000004s :      2: substitution.replace_old_param
 4.55% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005213     2
82.16% :     0.004283s :      1: type_inference.infer
17.84% :     0.000930s :      1: type_inference.specialize
------[replace.]   0.000071     6
65.67% :     0.000047s :      4: replace.inline
34.33% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000185     6
95.42% :     0.000176s :      4: match.inline
 4.58% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.48% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.16% :     0.000001s :     10: predicate.addn_zero_filter
 2.19% :     0.000003s :     10: predicate.arithmetic_simplify
 1.70% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.66% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.45% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.25% :     0.000002s :      6: predicate.get_grad_eliminate
 0.60% :     0.000001s :      3: predicate.graph_param_transform
 4.99% :     0.000006s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.80% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.57% :     0.000002s :     15: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.68% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.22% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 2.32% :     0.000003s :      3: predicate.mutable_eliminate
 0.39% :     0.000000s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.43% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.53% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.85% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     10: predicate.same_eliminate
 0.64% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.29% :     0.000002s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.89% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.22% :     0.000003s :     16: predicate.switch_defer_inline
 1.93% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.12% :     0.000009s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.53% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.88% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000638    12
46.72% :     0.000298s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.30% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.98% :     0.000300s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.029562    76
 0.18% :     0.000054s :      1: add_recomputation
 0.21% :     0.000062s :      1: auto_monad
 0.06% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.38% :     0.000407s :      1: bootstrap
 0.11% :     0.000033s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.08% :     0.000024s :      1: event_method
 0.06% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
21.19% :     0.006265s :      1: jit_opt_a
 0.55% :     0.000161s :      1: jit_opt_after_cconv
 0.20% :     0.000060s :      1: jit_opt_b
 1.60% :     0.000474s :      1: loop_unroll
 3.05% :     0.000902s :      1: mutable_eliminate
 2.90% :     0.000858s :     26: opt.transform.jit_opt_a
 0.17% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000028s :      4: opt.transform.jit_opt_b
 0.03% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.69% :     0.000499s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000050s :      1: py_interpret_to_execute
 0.06% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 1.30% :     0.000385s :      1: renormalize.infer
 1.75% :     0.000517s :      1: renormalize.specialize
 0.03% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000047s :      1: rewriter_after_opt_a
 0.27% :     0.000080s :      1: rewriter_before_opt_a
 0.27% :     0.000081s :      1: symbol_engine_optimizer
43.66% :     0.012908s :      1: task_emit
17.95% :     0.005306s :      1: type_inference
 0.21% :     0.000063s :      1: validate


TotalTime = 0.0321226, [33]
[bootstrap]: 0.00032276
[type_inference]: 0.00483563
[event_method]: 1.768e-05
[auto_monad]: 6.382e-05
[graph_reusing]: 5.24998e-06
[pre_auto_parallel]: 2.24001e-06
[py_interpret_to_execute]: 3.693e-05
[rewriter_before_opt_a]: 7.264e-05
[expand_dump_flag]: 3.41001e-06
[jit_opt_a]: 0.00563271, [2]
    [Cycle 1]: 0.00169232, [27]
        [switch_simplify]: 5.676e-05
        [loop_unroll]: 2.901e-05
        [a_1]: 0.00054146
        [with_stream_mark]: 1.823e-05
        [recompute_prepare]: 7.05e-06
        [updatestate_depend_eliminate]: 3.78001e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.65002e-06
        [parameter_eliminate]: 2.01998e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 5.17999e-06
        [accelerated_algorithm]: 5.02e-06
        [meta_shard_fg_expand]: 1.87999e-06
        [get_grad_eliminate_]: 5.02e-06
        [merge_forward]: 3.70998e-06
        [cell_reuse_recompute_pass]: 1.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.46e-05
        [j_node_and_user_rematch]: 8.17998e-06
        [meta_fg_expand]: 2.10002e-06
        [replace_old_param]: 8.45001e-06
        [inline_without_move]: 4.94e-06
        [renormalize]: 0.00072523
        [add_forward_monad_depend]: 6.12001e-06
        [auto_monad_grad]: 2.37001e-06
        [auto_monad_eliminator]: 1.392e-05
        [cse]: 2.855e-05
        [replace_applicator]: 1.235e-05
    [Cycle 2]: 0.00034925, [27]
        [switch_simplify]: 5.56e-06
        [loop_unroll]: 4.85001e-06
        [a_1]: 9.015e-05
        [with_stream_mark]: 1.039e-05
        [recompute_prepare]: 5.39e-06
        [updatestate_depend_eliminate]: 3.04999e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 4.85999e-06
        [updatestate_useless_node_eliminater]: 6.19999e-06
        [accelerated_algorithm]: 5.25001e-06
        [meta_shard_fg_expand]: 1.28002e-06
        [get_grad_eliminate_]: 4.77998e-06
        [merge_forward]: 3.09999e-06
        [cell_reuse_recompute_pass]: 1.98002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.486e-05
        [j_node_and_user_rematch]: 7.83001e-06
        [meta_fg_expand]: 1.64e-06
        [replace_old_param]: 7.97e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 9.5999e-07
        [auto_monad_eliminator]: 6.12999e-06
        [cse]: 1.272e-05
        [replace_applicator]: 5.54e-06
[py_interpret_to_execute_after_opt_a]: 1.134e-05
[rewriter_after_opt_a]: 3.664e-05
[convert_after_rewriter]: 7.21001e-06
[order_py_execute_after_rewriter]: 5.62001e-06
[mutable_eliminate]: 0.00064041
[jit_opt_b]: 5.143e-05, [1]
    [Cycle 1]: 4.376e-05, [2]
        [frontend_op_eliminate]: 1.716e-05
        [inline_after_opt_a]: 1.506e-05
[cconv]: 2.72e-05
[loop_unroll]: 0.00043952
[jit_opt_after_cconv]: 0.00014285, [1]
    [Cycle 1]: 0.00013644, [11]
        [c_1]: 2.033e-05
        [parameter_eliminate]: 2.91999e-06
        [updatestate_depend_eliminate]: 6.66e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [cse]: 2.127e-05
        [call_graph_tuple_transform]: 1.872e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.70001e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 5.37999e-06
[remove_dup_value]: 1.442e-05
[partial_unused_args_eliminate]: 2.48e-06
[environ_conv]: 5.53002e-06
[add_recomputation]: 4.949e-05
[cse_after_recomputation]: 2.035e-05, [1]
    [Cycle 1]: 1.441e-05, [1]
        [cse]: 8.70999e-06
[auto_monad_reorder]: 1.541e-05
[get_jit_bprop_graph]: 1.92999e-06
[rewriter_after_jit_bprop_graph]: 5.98002e-06
[opt_after_jit_grad]: 0.0004664
[symbol_engine_optimizer]: 8.08e-05, [1]
    [Cycle 1]: 7.352e-05, [6]
        [build]: 5.12999e-06
        [elim_shapecalc]: 8.67e-06
        [elim_not_effective]: 1.466e-05
        [opt_reshape]: 5.82001e-06
        [fold_const_symbol]: 9.44e-06
        [renormalize]: 5.09986e-07
[validate]: 4.278e-05
[backend_pass]: 1.47001e-06
[task_emit]: 0.0188243
[execute]: 9.05001e-06
Sums
    bootstrap                            :     0.000323s :  1.16%
    type_inference                       :     0.004836s : 17.35%
    event_method                         :     0.000018s :  0.06%
    auto_monad                           :     0.000064s :  0.23%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000037s :  0.13%
    rewriter_before_opt_a                :     0.000073s :  0.26%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000062s :  0.22%
    jit_opt_a.loop_unroll                :     0.000034s :  0.12%
    jit_opt_a.a_1                        :     0.000632s :  2.27%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000016s :  0.06%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.000725s :  2.60%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.07%
    jit_opt_a.cse                        :     0.000041s :  0.15%
    jit_opt_a.replace_applicator         :     0.000018s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.04%
    rewriter_after_opt_a                 :     0.000037s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000640s :  2.30%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.05%
    cconv                                :     0.000027s :  0.10%
    loop_unroll                          :     0.000440s :  1.58%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000014s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000049s :  0.18%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000015s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000466s :  1.67%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.15%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.018824s : 67.53%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000210    23
 1.14% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.52% :     0.000005s :      3: substitution.graph_param_transform
83.41% :     0.000175s :      4: substitution.inline
 1.20% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.51% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.35% :     0.000005s :      2: substitution.replace_old_param
 6.27% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004762     2
82.09% :     0.003909s :      1: type_inference.infer
17.91% :     0.000853s :      1: type_inference.specialize
------[replace.]   0.000071     6
66.18% :     0.000047s :      4: replace.inline
33.82% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000185     6
93.52% :     0.000173s :      4: match.inline
 6.48% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.93% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.38% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 2.51% :     0.000003s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.99% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.22% :     0.000001s :     10: predicate.depend_value_elim
 1.36% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.27% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.50% :     0.000006s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.82% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.78% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.42% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.52% :     0.000002s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.27% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.55% :     0.000009s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.56% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.81% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.91% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000648    12
51.99% :     0.000337s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.22% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.78% :     0.000277s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.033793    76
 0.16% :     0.000053s :      1: add_recomputation
 0.20% :     0.000067s :      1: auto_monad
 0.05% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.01% :     0.000341s :      1: bootstrap
 0.09% :     0.000030s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.07% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.07% :     0.000022s :      1: event_method
 0.04% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
16.68% :     0.005637s :      1: jit_opt_a
 0.43% :     0.000146s :      1: jit_opt_after_cconv
 0.16% :     0.000054s :      1: jit_opt_b
 1.33% :     0.000448s :      1: loop_unroll
 1.93% :     0.000651s :      1: mutable_eliminate
 2.49% :     0.000843s :     26: opt.transform.jit_opt_a
 0.14% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000026s :      4: opt.transform.jit_opt_b
 0.03% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.41% :     0.000475s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000040s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000017s :      1: remove_dup_value
 1.16% :     0.000392s :      1: renormalize.infer
 0.96% :     0.000326s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000040s :      1: rewriter_after_opt_a
 0.22% :     0.000075s :      1: rewriter_before_opt_a
 0.25% :     0.000083s :      1: symbol_engine_optimizer
55.76% :     0.018844s :      1: task_emit
14.36% :     0.004852s :      1: type_inference
 0.20% :     0.000068s :      1: validate


TotalTime = 0.0173171, [30]
[bootstrap]: 0.00030847
[type_inference]: 0.00783331
[event_method]: 2.496e-05
[auto_monad]: 8.17e-05
[graph_reusing]: 6.62002e-06
[pre_auto_parallel]: 2.92002e-06
[py_interpret_to_execute]: 5.765e-05
[rewriter_before_opt_a]: 8.508e-05
[expand_dump_flag]: 2.74001e-06
[jit_opt_a]: 0.00661417, [2]
    [Cycle 1]: 0.00251311, [27]
        [switch_simplify]: 0.00012646
        [loop_unroll]: 3.419e-05
        [a_1]: 0.00078639
        [with_stream_mark]: 1.864e-05
        [recompute_prepare]: 7.88001e-06
        [updatestate_depend_eliminate]: 3.11001e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 1.47001e-06
        [specialize_transform]: 6.74001e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 6.54001e-06
        [meta_shard_fg_expand]: 2.71e-06
        [get_grad_eliminate_]: 5.92001e-06
        [merge_forward]: 3.41001e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.445e-05
        [j_node_and_user_rematch]: 9.29e-06
        [meta_fg_expand]: 2.11e-06
        [replace_old_param]: 9.00001e-06
        [inline_without_move]: 5.60001e-06
        [renormalize]: 0.00121867
        [add_forward_monad_depend]: 5.96e-06
        [auto_monad_grad]: 2.41e-06
        [auto_monad_eliminator]: 1.389e-05
        [cse]: 2.575e-05
        [replace_applicator]: 1.389e-05
    [Cycle 2]: 0.00033373, [27]
        [switch_simplify]: 6.11e-06
        [loop_unroll]: 5.47001e-06
        [a_1]: 9.451e-05
        [with_stream_mark]: 1.142e-05
        [recompute_prepare]: 4.98001e-06
        [updatestate_depend_eliminate]: 3.06999e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.02998e-06
        [specialize_transform]: 5.04003e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 5.21002e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 4.32e-06
        [merge_forward]: 2.95998e-06
        [cell_reuse_recompute_pass]: 2.58998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.275e-05
        [j_node_and_user_rematch]: 7.88001e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 7.81001e-06
        [inline_without_move]: 4.67e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 9.89996e-07
        [auto_monad_eliminator]: 4.91002e-06
        [cse]: 1.105e-05
        [replace_applicator]: 4.94e-06
[py_interpret_to_execute_after_opt_a]: 1.217e-05
[rewriter_after_opt_a]: 3.65e-05
[convert_after_rewriter]: 7.38e-06
[order_py_execute_after_rewriter]: 5.15001e-06
[mutable_eliminate]: 0.00073161
[jit_opt_b]: 5.162e-05, [1]
    [Cycle 1]: 4.45e-05, [2]
        [frontend_op_eliminate]: 1.698e-05
        [inline_after_opt_a]: 1.536e-05
[cconv]: 2.727e-05
[loop_unroll]: 0.00044011
[jit_opt_after_cconv]: 0.00014311, [1]
    [Cycle 1]: 0.00013655, [11]
        [c_1]: 2.158e-05
        [parameter_eliminate]: 3.31001e-06
        [updatestate_depend_eliminate]: 6.17001e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [cse]: 2.028e-05
        [call_graph_tuple_transform]: 1.831e-05
        [tuple_list_get_item_eliminator]: 5.36998e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 5.42999e-06
[remove_dup_value]: 1.338e-05
[partial_unused_args_eliminate]: 1.93002e-06
[environ_conv]: 5.59e-06
[add_recomputation]: 4.719e-05
[cse_after_recomputation]: 2.096e-05, [1]
    [Cycle 1]: 1.558e-05, [1]
        [cse]: 1.002e-05
[auto_monad_reorder]: 1.423e-05
[get_jit_bprop_graph]: 1.86e-06
[rewriter_after_jit_bprop_graph]: 5.99e-06
[opt_after_jit_grad]: 0.00045666
[symbol_engine_optimizer]: 7.431e-05, [1]
    [Cycle 1]: 6.825e-05, [6]
        [build]: 3.43e-06
        [elim_shapecalc]: 7.78999e-06
        [elim_not_effective]: 1.217e-05
        [opt_reshape]: 6.01e-06
        [fold_const_symbol]: 8.88002e-06
        [renormalize]: 4.89992e-07
[validate]: 3.503e-05
Sums
    bootstrap                            :     0.000308s :  2.38%
    type_inference                       :     0.007833s : 60.42%
    event_method                         :     0.000025s :  0.19%
    auto_monad                           :     0.000082s :  0.63%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000058s :  0.44%
    rewriter_before_opt_a                :     0.000085s :  0.66%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000133s :  1.02%
    jit_opt_a.loop_unroll                :     0.000040s :  0.31%
    jit_opt_a.a_1                        :     0.000881s :  6.79%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.13%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.001219s :  9.40%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.15%
    jit_opt_a.cse                        :     0.000037s :  0.28%
    jit_opt_a.replace_applicator         :     0.000019s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.09%
    rewriter_after_opt_a                 :     0.000036s :  0.28%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000732s :  5.64%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000027s :  0.21%
    loop_unroll                          :     0.000440s :  3.39%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.17%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000047s :  0.36%
    cse_after_recomputation.cse          :     0.000010s :  0.08%
    auto_monad_reorder                   :     0.000014s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000457s :  3.52%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.27%


Time group info:
------[substitution.]   0.000326    26
 0.59% :     0.000002s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 1.42% :     0.000005s :      3: substitution.graph_param_transform
87.44% :     0.000285s :      6: substitution.inline
 1.01% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.47% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.07% :     0.000003s :      2: substitution.replace_old_param
 2.51% :     0.000008s :      1: substitution.switch_simplify
 4.06% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007757     2
82.45% :     0.006396s :      1: type_inference.infer
17.55% :     0.001361s :      1: type_inference.specialize
------[replace.]   0.000159     9
48.82% :     0.000078s :      6: replace.inline
26.31% :     0.000042s :      1: replace.switch_simplify
24.88% :     0.000040s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000301     9
93.61% :     0.000282s :      6: match.inline
 2.44% :     0.000007s :      1: match.switch_simplify
 3.95% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000141   801
 1.62% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.13% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     12: predicate.addn_check_dump
 1.82% :     0.000003s :     12: predicate.addn_zero_filter
 2.46% :     0.000003s :     12: predicate.arithmetic_simplify
 1.30% :     0.000002s :     12: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000002s :     12: predicate.compare_switch_simplify
 1.44% :     0.000002s :     12: predicate.depend_value_elim
 1.24% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.42% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.46% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.16% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.12% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.34% :     0.000002s :     12: predicate.environ_get_eliminate
 1.35% :     0.000002s :     12: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 4.97% :     0.000007s :     26: predicate.inline
 0.75% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000002s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     17: predicate.load_eliminater
 1.20% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.90% :     0.000005s :     40: predicate.loop_unroll_before_grad
 2.31% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     12: predicate.merge_addn
 1.16% :     0.000002s :     12: predicate.minmaximum_grad
 2.26% :     0.000003s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000003s :     17: predicate.partial_eliminate
 1.38% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     12: predicate.reduce_eliminate
 1.43% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.53% :     0.000002s :     20: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     12: predicate.reshape_eliminate
 1.19% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.61% :     0.000002s :     12: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 0.86% :     0.000001s :      6: predicate.specialize_transform
 1.60% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.31% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.92% :     0.000004s :     20: predicate.switch_defer_inline
 2.33% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.33% :     0.000012s :     65: predicate.switch_simplify
 1.39% :     0.000002s :     12: predicate.tile_eliminate
 1.29% :     0.000002s :     12: predicate.transpose_eliminate
 1.37% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.18% :     0.000004s :     20: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     12: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001014    15
42.79% :     0.000434s :      6: func_graph_cloner_run.FuncGraphClonerGraph
11.82% :     0.000120s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.38% :     0.000460s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019765    72
 0.25% :     0.000050s :      1: add_recomputation
 0.43% :     0.000086s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 1.63% :     0.000322s :      1: bootstrap
 0.15% :     0.000030s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.15% :     0.000029s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.48% :     0.006618s :      1: jit_opt_a
 0.74% :     0.000146s :      1: jit_opt_after_cconv
 0.27% :     0.000054s :      1: jit_opt_b
 2.27% :     0.000448s :      1: loop_unroll
 3.74% :     0.000740s :      1: mutable_eliminate
 5.93% :     0.001173s :     26: opt.transform.jit_opt_a
 0.24% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.35% :     0.000465s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000060s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 3.89% :     0.000768s :      1: renormalize.infer
 2.23% :     0.000442s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000040s :      1: rewriter_after_opt_a
 0.45% :     0.000088s :      1: rewriter_before_opt_a
 0.39% :     0.000077s :      1: symbol_engine_optimizer
39.71% :     0.007849s :      1: type_inference


TotalTime = 0.0383526, [33]
[bootstrap]: 0.00053285
[type_inference]: 0.005821
[event_method]: 2.025e-05
[auto_monad]: 7.051e-05
[graph_reusing]: 7.24001e-06
[pre_auto_parallel]: 3.49001e-06
[py_interpret_to_execute]: 4.421e-05
[rewriter_before_opt_a]: 8.439e-05
[expand_dump_flag]: 3.64002e-06
[jit_opt_a]: 0.0219638, [2]
    [Cycle 1]: 0.0172895, [27]
        [switch_simplify]: 7.456e-05
        [loop_unroll]: 3.033e-05
        [a_1]: 0.0155565
        [with_stream_mark]: 3.351e-05
        [recompute_prepare]: 1.38e-05
        [updatestate_depend_eliminate]: 4.79998e-06
        [updatestate_assign_eliminate]: 3.31001e-06
        [updatestate_loads_eliminate]: 3.28e-06
        [parameter_eliminate]: 3.08e-06
        [specialize_transform]: 6.91001e-06
        [updatestate_useless_node_eliminater]: 6.47001e-06
        [accelerated_algorithm]: 6.73e-06
        [meta_shard_fg_expand]: 3.66999e-06
        [get_grad_eliminate_]: 6.16998e-06
        [merge_forward]: 4.15999e-06
        [cell_reuse_recompute_pass]: 2.26e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.833e-05
        [j_node_and_user_rematch]: 1.169e-05
        [meta_fg_expand]: 3.43e-06
        [replace_old_param]: 1.168e-05
        [inline_without_move]: 6.94999e-06
        [renormalize]: 0.00112958
        [add_forward_monad_depend]: 8.08001e-06
        [auto_monad_grad]: 2.83e-06
        [auto_monad_eliminator]: 1.932e-05
        [cse]: 3.988e-05
        [replace_applicator]: 1.709e-05
    [Cycle 2]: 0.00035357, [27]
        [switch_simplify]: 6.83e-06
        [loop_unroll]: 5.25999e-06
        [a_1]: 0.00010117
        [with_stream_mark]: 1.345e-05
        [recompute_prepare]: 5.40999e-06
        [updatestate_depend_eliminate]: 2.91999e-06
        [updatestate_assign_eliminate]: 2.06998e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 1.49e-06
        [specialize_transform]: 4.93001e-06
        [updatestate_useless_node_eliminater]: 4.97999e-06
        [accelerated_algorithm]: 5.82999e-06
        [meta_shard_fg_expand]: 2.34001e-06
        [get_grad_eliminate_]: 4.78001e-06
        [merge_forward]: 3.36999e-06
        [cell_reuse_recompute_pass]: 1.99e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.412e-05
        [j_node_and_user_rematch]: 9.46998e-06
        [meta_fg_expand]: 2.46e-06
        [replace_old_param]: 7.68999e-06
        [inline_without_move]: 4.92999e-06
        [renormalize]: 1.10012e-07
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 1.15999e-06
        [auto_monad_eliminator]: 5.72001e-06
        [cse]: 1.258e-05
        [replace_applicator]: 5.36998e-06
[py_interpret_to_execute_after_opt_a]: 1.579e-05
[rewriter_after_opt_a]: 4.257e-05
[convert_after_rewriter]: 1.022e-05
[order_py_execute_after_rewriter]: 5.04e-06
[mutable_eliminate]: 0.00093915
[jit_opt_b]: 5.45e-05, [1]
    [Cycle 1]: 4.576e-05, [2]
        [frontend_op_eliminate]: 1.803e-05
        [inline_after_opt_a]: 1.518e-05
[cconv]: 3.487e-05
[loop_unroll]: 0.00047207
[jit_opt_after_cconv]: 0.0001649, [1]
    [Cycle 1]: 0.00015742, [11]
        [c_1]: 2.355e-05
        [parameter_eliminate]: 3.79002e-06
        [updatestate_depend_eliminate]: 7.51999e-06
        [updatestate_assign_eliminate]: 3.41999e-06
        [updatestate_loads_eliminate]: 2.34999e-06
        [cse]: 2.812e-05
        [call_graph_tuple_transform]: 2.283e-05
        [tuple_list_get_item_eliminator]: 5.26002e-06
        [none_parameter_eliminate]: 1.50999e-06
        [renormalize]: 9.10019e-07
        [switch_simplify]: 5.43002e-06
[remove_dup_value]: 1.575e-05
[partial_unused_args_eliminate]: 2.64001e-06
[environ_conv]: 5.97001e-06
[add_recomputation]: 5.792e-05
[cse_after_recomputation]: 2.352e-05, [1]
    [Cycle 1]: 1.803e-05, [1]
        [cse]: 1.072e-05
[auto_monad_reorder]: 1.609e-05
[get_jit_bprop_graph]: 2.36998e-06
[rewriter_after_jit_bprop_graph]: 9.52999e-06
[opt_after_jit_grad]: 0.00051949
[symbol_engine_optimizer]: 8.475e-05, [1]
    [Cycle 1]: 7.72e-05, [6]
        [build]: 5.19e-06
        [elim_shapecalc]: 8.27998e-06
        [elim_not_effective]: 1.63e-05
        [opt_reshape]: 6.26e-06
        [fold_const_symbol]: 1.008e-05
        [renormalize]: 6.80011e-07
[validate]: 4.718e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00702134
[execute]: 9.29998e-06
Sums
    bootstrap                            :     0.000533s :  1.60%
    type_inference                       :     0.005821s : 17.49%
    event_method                         :     0.000020s :  0.06%
    auto_monad                           :     0.000071s :  0.21%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.13%
    rewriter_before_opt_a                :     0.000084s :  0.25%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000081s :  0.24%
    jit_opt_a.loop_unroll                :     0.000036s :  0.11%
    jit_opt_a.a_1                        :     0.015658s : 47.06%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.03%
    jit_opt_a.merge_forward              :     0.000008s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.06%
    jit_opt_a.inline_without_move        :     0.000012s :  0.04%
    jit_opt_a.renormalize                :     0.001130s :  3.40%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.08%
    jit_opt_a.cse                        :     0.000052s :  0.16%
    jit_opt_a.replace_applicator         :     0.000022s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.05%
    rewriter_after_opt_a                 :     0.000043s :  0.13%
    convert_after_rewriter               :     0.000010s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000939s :  2.82%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.05%
    cconv                                :     0.000035s :  0.10%
    loop_unroll                          :     0.000472s :  1.42%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000016s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000058s :  0.17%
    cse_after_recomputation.cse          :     0.000011s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.03%
    opt_after_jit_grad                   :     0.000519s :  1.56%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000047s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007021s : 21.10%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000391    23
 0.86% :     0.000003s :      2: substitution.elim_not_effective
 0.33% :     0.000001s :      2: substitution.fold_const_symbol
 1.78% :     0.000007s :      3: substitution.graph_param_transform
87.50% :     0.000342s :      4: substitution.inline
 1.35% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.46% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.56% :     0.000006s :      2: substitution.replace_old_param
 5.16% :     0.000020s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005736     2
82.73% :     0.004745s :      1: type_inference.infer
17.27% :     0.000991s :      1: type_inference.specialize
------[replace.]   0.000160     6
65.45% :     0.000104s :      4: replace.inline
34.55% :     0.000055s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000358     6
94.71% :     0.000339s :      4: match.inline
 5.29% :     0.000019s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000168   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000002s :     10: predicate.addn_check_dump
 1.15% :     0.000002s :     10: predicate.addn_zero_filter
 2.54% :     0.000004s :     10: predicate.arithmetic_simplify
 1.38% :     0.000002s :     10: predicate.cast_eliminate
 0.30% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000002s :     10: predicate.compare_switch_simplify
 1.51% :     0.000003s :     10: predicate.depend_value_elim
 1.37% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.53% :     0.000003s :     10: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.25% :     0.000000s :      3: predicate.elim_not_effective
 0.46% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.76% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.93% :     0.000002s :     10: predicate.environ_get_depend_swap
 0.99% :     0.000002s :     10: predicate.environ_get_eliminate
 0.91% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.17% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 7.10% :     0.000012s :     22: predicate.inline
 0.94% :     0.000002s :      6: predicate.inline_without_move
 0.27% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.37% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.40% :     0.000002s :     15: predicate.load_eliminater
 1.04% :     0.000002s :      3: predicate.loop_unroll_after_grad
 2.90% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 5.26% :     0.000009s :     10: predicate.merge_addn
 0.89% :     0.000002s :     10: predicate.minmaximum_grad
 1.76% :     0.000003s :      3: predicate.mutable_eliminate
 0.34% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     15: predicate.partial_eliminate
 1.00% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.16% :     0.000004s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.35% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.36% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.19% :     0.000000s :      3: predicate.reset_defer_inline
 1.64% :     0.000003s :     10: predicate.reshape_eliminate
 1.28% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.67% :     0.000001s :      6: predicate.special_op_eliminate
 0.78% :     0.000001s :      6: predicate.specialize_transform
 2.52% :     0.000004s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.31% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.56% :     0.000006s :     16: predicate.switch_defer_inline
 1.55% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.67% :     0.000011s :     53: predicate.switch_simplify
 1.46% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000002s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.31% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.17% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.26% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.16% :     0.000004s :     10: predicate.value_based_eliminate
 0.58% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000845    12
46.22% :     0.000391s :      5: func_graph_cloner_run.FuncGraphClonerGraph
11.36% :     0.000096s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.41% :     0.000359s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.055507    76
 0.11% :     0.000061s :      1: add_recomputation
 0.13% :     0.000074s :      1: auto_monad
 0.03% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.01% :     0.000559s :      1: bootstrap
 0.07% :     0.000038s :      1: cconv
 0.02% :     0.000013s :      1: convert_after_rewriter
 0.05% :     0.000026s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.05% :     0.000025s :      1: event_method
 0.03% :     0.000014s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
39.58% :     0.021968s :      1: jit_opt_a
 0.30% :     0.000168s :      1: jit_opt_after_cconv
 0.10% :     0.000057s :      1: jit_opt_b
 0.87% :     0.000481s :      1: loop_unroll
 1.71% :     0.000950s :      1: mutable_eliminate
28.67% :     0.015911s :     26: opt.transform.jit_opt_a
 0.10% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000027s :      4: opt.transform.jit_opt_b
 0.02% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000037s :      4: opt.transform.symbol_engine_opt
 0.95% :     0.000529s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.09% :     0.000048s :      1: py_interpret_to_execute
 0.03% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000019s :      1: remove_dup_value
 1.12% :     0.000620s :      1: renormalize.infer
 0.90% :     0.000499s :      1: renormalize.specialize
 0.02% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000045s :      1: rewriter_after_opt_a
 0.16% :     0.000087s :      1: rewriter_before_opt_a
 0.16% :     0.000087s :      1: symbol_engine_optimizer
12.69% :     0.007043s :      1: task_emit
10.52% :     0.005841s :      1: type_inference
 0.13% :     0.000070s :      1: validate


TotalTime = 0.0223041, [33]
[bootstrap]: 0.00041994
[type_inference]: 0.00658525
[event_method]: 2.131e-05
[auto_monad]: 6.811e-05
[graph_reusing]: 5.74e-06
[pre_auto_parallel]: 2.84999e-06
[py_interpret_to_execute]: 3.966e-05
[rewriter_before_opt_a]: 7.577e-05
[expand_dump_flag]: 3.20002e-06
[jit_opt_a]: 0.00650868, [2]
    [Cycle 1]: 0.00181592, [27]
        [switch_simplify]: 6.208e-05
        [loop_unroll]: 3.025e-05
        [a_1]: 0.0005833
        [with_stream_mark]: 2.206e-05
        [recompute_prepare]: 7.08998e-06
        [updatestate_depend_eliminate]: 3.80998e-06
        [updatestate_assign_eliminate]: 3.07002e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 5.79999e-06
        [updatestate_useless_node_eliminater]: 5.21002e-06
        [accelerated_algorithm]: 5.67001e-06
        [meta_shard_fg_expand]: 2.48e-06
        [get_grad_eliminate_]: 5.23002e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.503e-05
        [j_node_and_user_rematch]: 9.44e-06
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 9.20001e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.0007779
        [add_forward_monad_depend]: 6.21e-06
        [auto_monad_grad]: 2.66e-06
        [auto_monad_eliminator]: 1.583e-05
        [cse]: 2.932e-05
        [replace_applicator]: 1.241e-05
    [Cycle 2]: 0.00032966, [27]
        [switch_simplify]: 5.93998e-06
        [loop_unroll]: 4.93001e-06
        [a_1]: 9.198e-05
        [with_stream_mark]: 1.08e-05
        [recompute_prepare]: 4.70999e-06
        [updatestate_depend_eliminate]: 3.12002e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 5.70001e-06
        [meta_shard_fg_expand]: 1.33002e-06
        [get_grad_eliminate_]: 4.72e-06
        [merge_forward]: 2.62001e-06
        [cell_reuse_recompute_pass]: 1.73002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.363e-05
        [j_node_and_user_rematch]: 7.62998e-06
        [meta_fg_expand]: 1.47999e-06
        [replace_old_param]: 1.29e-05
        [inline_without_move]: 4.65999e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 6.99976e-07
        [auto_monad_eliminator]: 5.22999e-06
        [cse]: 1.141e-05
        [replace_applicator]: 5.37001e-06
[py_interpret_to_execute_after_opt_a]: 1.322e-05
[rewriter_after_opt_a]: 3.779e-05
[convert_after_rewriter]: 7.63001e-06
[order_py_execute_after_rewriter]: 7.1e-06
[mutable_eliminate]: 0.00062405
[jit_opt_b]: 5.017e-05, [1]
    [Cycle 1]: 4.251e-05, [2]
        [frontend_op_eliminate]: 1.609e-05
        [inline_after_opt_a]: 1.509e-05
[cconv]: 2.875e-05
[loop_unroll]: 0.00044148
[jit_opt_after_cconv]: 0.0001458, [1]
    [Cycle 1]: 0.00013913, [11]
        [c_1]: 2.108e-05
        [parameter_eliminate]: 2.71e-06
        [updatestate_depend_eliminate]: 7.22997e-06
        [updatestate_assign_eliminate]: 2.65002e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [cse]: 2.071e-05
        [call_graph_tuple_transform]: 1.895e-05
        [tuple_list_get_item_eliminator]: 5.68997e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 8.50006e-07
        [switch_simplify]: 5.57001e-06
[remove_dup_value]: 1.442e-05
[partial_unused_args_eliminate]: 2.66e-06
[environ_conv]: 5.95002e-06
[add_recomputation]: 5.423e-05
[cse_after_recomputation]: 1.986e-05, [1]
    [Cycle 1]: 1.461e-05, [1]
        [cse]: 8.82e-06
[auto_monad_reorder]: 1.534e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 5.62001e-06
[opt_after_jit_grad]: 0.00045984
[symbol_engine_optimizer]: 7.263e-05, [1]
    [Cycle 1]: 6.607e-05, [6]
        [build]: 4.18001e-06
        [elim_shapecalc]: 7.95e-06
        [elim_not_effective]: 1.255e-05
        [opt_reshape]: 5.35001e-06
        [fold_const_symbol]: 8.2e-06
        [renormalize]: 6.30011e-07
[validate]: 4.223e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00628298
[execute]: 8.07e-06
Sums
    bootstrap                            :     0.000420s :  2.43%
    type_inference                       :     0.006585s : 38.08%
    event_method                         :     0.000021s :  0.12%
    auto_monad                           :     0.000068s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.23%
    rewriter_before_opt_a                :     0.000076s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.20%
    jit_opt_a.a_1                        :     0.000675s :  3.91%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.13%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000778s :  4.50%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.12%
    jit_opt_a.cse                        :     0.000041s :  0.24%
    jit_opt_a.replace_applicator         :     0.000018s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000007s :  0.04%
    mutable_eliminate                    :     0.000624s :  3.61%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000029s :  0.17%
    loop_unroll                          :     0.000441s :  2.55%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000054s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000460s :  2.66%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006283s : 36.33%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000244    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000001s :      2: substitution.fold_const_symbol
 2.50% :     0.000006s :      3: substitution.graph_param_transform
85.62% :     0.000209s :      4: substitution.inline
 1.50% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.84% :     0.000005s :      2: substitution.replace_old_param
 5.18% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006504     2
85.79% :     0.005580s :      1: type_inference.infer
14.21% :     0.000924s :      1: type_inference.specialize
------[replace.]   0.000070     6
68.39% :     0.000048s :      4: replace.inline
31.61% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000218     6
94.69% :     0.000207s :      4: match.inline
 5.31% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.30% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.05% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000002s :     10: predicate.addn_zero_filter
 2.02% :     0.000003s :     10: predicate.arithmetic_simplify
 1.13% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.63% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.61% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.03% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.16% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 6.01% :     0.000007s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.28% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.25% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.00% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.41% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.85% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.36% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.34% :     0.000002s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.69% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.45% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.14% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.30% :     0.000003s :     16: predicate.switch_defer_inline
 2.01% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.08% :     0.000010s :     53: predicate.switch_simplify
 1.25% :     0.000002s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.00% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.12% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.79% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.26% :     0.000003s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000694    12
52.71% :     0.000366s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.46% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.83% :     0.000276s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024073    76
 0.24% :     0.000057s :      1: add_recomputation
 0.30% :     0.000071s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.83% :     0.000441s :      1: bootstrap
 0.13% :     0.000031s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.11% :     0.000026s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
27.05% :     0.006513s :      1: jit_opt_a
 0.62% :     0.000149s :      1: jit_opt_after_cconv
 0.22% :     0.000053s :      1: jit_opt_b
 1.87% :     0.000450s :      1: loop_unroll
 2.63% :     0.000633s :      1: mutable_eliminate
 3.73% :     0.000897s :     26: opt.transform.jit_opt_a
 0.20% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.95% :     0.000468s :      1: opt_after_jit_grad
 0.04% :     0.000010s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000043s :      1: py_interpret_to_execute
 0.06% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.87% :     0.000450s :      1: renormalize.infer
 1.33% :     0.000320s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000041s :      1: rewriter_after_opt_a
 0.33% :     0.000079s :      1: rewriter_before_opt_a
 0.31% :     0.000075s :      1: symbol_engine_optimizer
26.17% :     0.006300s :      1: task_emit
27.45% :     0.006607s :      1: type_inference
 0.27% :     0.000065s :      1: validate


TotalTime = 0.0315525, [33]
[bootstrap]: 0.00033813
[type_inference]: 0.00501546
[event_method]: 1.834e-05
[auto_monad]: 6.405e-05
[graph_reusing]: 5.61e-06
[pre_auto_parallel]: 2.48002e-06
[py_interpret_to_execute]: 3.765e-05
[rewriter_before_opt_a]: 7.491e-05
[expand_dump_flag]: 2.81999e-06
[jit_opt_a]: 0.00614134, [2]
    [Cycle 1]: 0.00190192, [27]
        [switch_simplify]: 6.049e-05
        [loop_unroll]: 2.974e-05
        [a_1]: 0.00062107
        [with_stream_mark]: 2.579e-05
        [recompute_prepare]: 7.51001e-06
        [updatestate_depend_eliminate]: 4e-06
        [updatestate_assign_eliminate]: 2.91999e-06
        [updatestate_loads_eliminate]: 2.52001e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 6.26e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 5.69999e-06
        [meta_shard_fg_expand]: 2.43e-06
        [get_grad_eliminate_]: 5.30999e-06
        [merge_forward]: 3.43999e-06
        [cell_reuse_recompute_pass]: 1.41998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.473e-05
        [j_node_and_user_rematch]: 9.20001e-06
        [meta_fg_expand]: 2.11003e-06
        [replace_old_param]: 9.25001e-06
        [inline_without_move]: 5.14e-06
        [renormalize]: 0.00081869
        [add_forward_monad_depend]: 7e-06
        [auto_monad_grad]: 2.99001e-06
        [auto_monad_eliminator]: 1.601e-05
        [cse]: 3.177e-05
        [replace_applicator]: 1.437e-05
    [Cycle 2]: 0.00034163, [27]
        [switch_simplify]: 5.85002e-06
        [loop_unroll]: 6.21e-06
        [a_1]: 9.405e-05
        [with_stream_mark]: 1.138e-05
        [recompute_prepare]: 5.20999e-06
        [updatestate_depend_eliminate]: 3.03998e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 4.93001e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 6.01998e-06
        [meta_shard_fg_expand]: 1.37999e-06
        [get_grad_eliminate_]: 5.07e-06
        [merge_forward]: 3.11999e-06
        [cell_reuse_recompute_pass]: 2.62001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.445e-05
        [j_node_and_user_rematch]: 7.8e-06
        [meta_fg_expand]: 1.59998e-06
        [replace_old_param]: 8.2e-06
        [inline_without_move]: 4.43999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.52999e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 5.34998e-06
        [cse]: 1.213e-05
        [replace_applicator]: 4.95001e-06
[py_interpret_to_execute_after_opt_a]: 1.493e-05
[rewriter_after_opt_a]: 3.898e-05
[convert_after_rewriter]: 7.38e-06
[order_py_execute_after_rewriter]: 5.84e-06
[mutable_eliminate]: 0.00068486
[jit_opt_b]: 5.107e-05, [1]
    [Cycle 1]: 4.391e-05, [2]
        [frontend_op_eliminate]: 1.753e-05
        [inline_after_opt_a]: 1.498e-05
[cconv]: 2.76e-05
[loop_unroll]: 0.00041924
[jit_opt_after_cconv]: 0.00013871, [1]
    [Cycle 1]: 0.00013248, [11]
        [c_1]: 2.021e-05
        [parameter_eliminate]: 2.96001e-06
        [updatestate_depend_eliminate]: 5.61998e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.01998e-06
        [cse]: 1.949e-05
        [call_graph_tuple_transform]: 1.905e-05
        [tuple_list_get_item_eliminator]: 5.17e-06
        [none_parameter_eliminate]: 1.50999e-06
        [renormalize]: 6.10016e-07
        [switch_simplify]: 5.06997e-06
[remove_dup_value]: 1.487e-05
[partial_unused_args_eliminate]: 2.31e-06
[environ_conv]: 4.62e-06
[add_recomputation]: 5.134e-05
[cse_after_recomputation]: 1.994e-05, [1]
    [Cycle 1]: 1.439e-05, [1]
        [cse]: 9.04e-06
[auto_monad_reorder]: 1.496e-05
[get_jit_bprop_graph]: 2.44001e-06
[rewriter_after_jit_bprop_graph]: 5.74e-06
[opt_after_jit_grad]: 0.00045015
[symbol_engine_optimizer]: 7.299e-05, [1]
    [Cycle 1]: 6.592e-05, [6]
        [build]: 3.77998e-06
        [elim_shapecalc]: 7.21999e-06
        [elim_not_effective]: 1.305e-05
        [opt_reshape]: 5.97999e-06
        [fold_const_symbol]: 8.36002e-06
        [renormalize]: 5.69999e-07
[validate]: 3.836e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.0175473
[execute]: 1.048e-05
Sums
    bootstrap                            :     0.000338s :  1.25%
    type_inference                       :     0.005015s : 18.57%
    event_method                         :     0.000018s :  0.07%
    auto_monad                           :     0.000064s :  0.24%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000038s :  0.14%
    rewriter_before_opt_a                :     0.000075s :  0.28%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000066s :  0.25%
    jit_opt_a.loop_unroll                :     0.000036s :  0.13%
    jit_opt_a.a_1                        :     0.000715s :  2.65%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000017s :  0.06%
    jit_opt_a.inline_without_move        :     0.000010s :  0.04%
    jit_opt_a.renormalize                :     0.000819s :  3.03%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.08%
    jit_opt_a.cse                        :     0.000044s :  0.16%
    jit_opt_a.replace_applicator         :     0.000019s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.06%
    rewriter_after_opt_a                 :     0.000039s :  0.14%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000685s :  2.54%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.06%
    cconv                                :     0.000028s :  0.10%
    loop_unroll                          :     0.000419s :  1.55%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000015s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000051s :  0.19%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000015s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000450s :  1.67%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.017547s : 64.98%
    execute                              :     0.000010s :  0.04%


Time group info:
------[substitution.]   0.000236    23
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.31% :     0.000005s :      3: substitution.graph_param_transform
85.38% :     0.000201s :      4: substitution.inline
 1.57% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.14% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.25% :     0.000005s :      2: substitution.replace_old_param
 4.88% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004941     2
83.12% :     0.004107s :      1: type_inference.infer
16.88% :     0.000834s :      1: type_inference.specialize
------[replace.]   0.000069     6
68.25% :     0.000047s :      4: replace.inline
31.75% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000209     6
94.99% :     0.000199s :      4: match.inline
 5.01% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 2.26% :     0.000003s :     10: predicate.arithmetic_simplify
 1.47% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.63% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.19% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.34% :     0.000002s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.69% :     0.000007s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.40% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.48% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000000s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.92% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.44% :     0.000002s :      6: predicate.special_op_eliminate
 1.15% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.30% :     0.000003s :     16: predicate.switch_defer_inline
 1.96% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.57% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.58% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.00% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.65% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000644    12
50.89% :     0.000328s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.22% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.89% :     0.000270s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.033403    76
 0.16% :     0.000054s :      1: add_recomputation
 0.20% :     0.000068s :      1: auto_monad
 0.05% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.08% :     0.000360s :      1: bootstrap
 0.09% :     0.000030s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000022s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.07% :     0.000022s :      1: event_method
 0.05% :     0.000016s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
18.40% :     0.006145s :      1: jit_opt_a
 0.42% :     0.000141s :      1: jit_opt_after_cconv
 0.16% :     0.000054s :      1: jit_opt_b
 1.28% :     0.000427s :      1: loop_unroll
 2.08% :     0.000694s :      1: mutable_eliminate
 2.80% :     0.000936s :     26: opt.transform.jit_opt_a
 0.14% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000026s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.37% :     0.000458s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000041s :      1: py_interpret_to_execute
 0.05% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000017s :      1: remove_dup_value
 1.37% :     0.000456s :      1: renormalize.infer
 1.06% :     0.000354s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000042s :      1: rewriter_after_opt_a
 0.23% :     0.000078s :      1: rewriter_before_opt_a
 0.23% :     0.000076s :      1: symbol_engine_optimizer
52.60% :     0.017569s :      1: task_emit
15.06% :     0.005032s :      1: type_inference
 0.18% :     0.000060s :      1: validate


TotalTime = 0.0285622, [33]
[bootstrap]: 0.0003377
[type_inference]: 0.00735711
[event_method]: 2.213e-05
[auto_monad]: 7.331e-05
[graph_reusing]: 5.88002e-06
[pre_auto_parallel]: 2.49001e-06
[py_interpret_to_execute]: 4.891e-05
[rewriter_before_opt_a]: 8.096e-05
[expand_dump_flag]: 3.47002e-06
[jit_opt_a]: 0.00862189, [2]
    [Cycle 1]: 0.00377912, [27]
        [switch_simplify]: 7.442e-05
        [loop_unroll]: 3.18e-05
        [a_1]: 0.00066398
        [with_stream_mark]: 2.666e-05
        [recompute_prepare]: 8.45001e-06
        [updatestate_depend_eliminate]: 4.27e-06
        [updatestate_assign_eliminate]: 3.22002e-06
        [updatestate_loads_eliminate]: 2.56998e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 7.07002e-06
        [updatestate_useless_node_eliminater]: 5.47001e-06
        [accelerated_algorithm]: 5.69e-06
        [meta_shard_fg_expand]: 2.51e-06
        [get_grad_eliminate_]: 5.58002e-06
        [merge_forward]: 3.59002e-06
        [cell_reuse_recompute_pass]: 1.81998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.494e-05
        [j_node_and_user_rematch]: 1.059e-05
        [meta_fg_expand]: 2.44999e-06
        [replace_old_param]: 1.071e-05
        [inline_without_move]: 5.73002e-06
        [renormalize]: 0.00257935
        [add_forward_monad_depend]: 1.073e-05
        [auto_monad_grad]: 2.37999e-06
        [auto_monad_eliminator]: 2.208e-05
        [cse]: 3.588e-05
        [replace_applicator]: 2.169e-05
    [Cycle 2]: 0.00036313, [27]
        [switch_simplify]: 6.56e-06
        [loop_unroll]: 5.42001e-06
        [a_1]: 0.00010324
        [with_stream_mark]: 1.542e-05
        [recompute_prepare]: 5.12e-06
        [updatestate_depend_eliminate]: 2.97002e-06
        [updatestate_assign_eliminate]: 2.67001e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 5.04998e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 6.38e-06
        [meta_shard_fg_expand]: 1.25001e-06
        [get_grad_eliminate_]: 4.87998e-06
        [merge_forward]: 3.95e-06
        [cell_reuse_recompute_pass]: 3.76999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.708e-05
        [j_node_and_user_rematch]: 8.90999e-06
        [meta_fg_expand]: 2.04999e-06
        [replace_old_param]: 8.97e-06
        [inline_without_move]: 4.94e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 9.90025e-07
        [auto_monad_eliminator]: 5.30001e-06
        [cse]: 1.121e-05
        [replace_applicator]: 1.314e-05
[py_interpret_to_execute_after_opt_a]: 1.585e-05
[rewriter_after_opt_a]: 4.09e-05
[convert_after_rewriter]: 7.36001e-06
[order_py_execute_after_rewriter]: 4.93001e-06
[mutable_eliminate]: 0.00084778
[jit_opt_b]: 0.000127, [1]
    [Cycle 1]: 0.00011806, [2]
        [frontend_op_eliminate]: 8.732e-05
        [inline_after_opt_a]: 1.687e-05
[cconv]: 3.238e-05
[loop_unroll]: 0.00046652
[jit_opt_after_cconv]: 0.00015358, [1]
    [Cycle 1]: 0.00014664, [11]
        [c_1]: 2.3e-05
        [parameter_eliminate]: 3.34001e-06
        [updatestate_depend_eliminate]: 6.78e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [cse]: 2.234e-05
        [call_graph_tuple_transform]: 2.116e-05
        [tuple_list_get_item_eliminator]: 5.60001e-06
        [none_parameter_eliminate]: 1.86e-06
        [renormalize]: 7.2e-07
        [switch_simplify]: 5.51e-06
[remove_dup_value]: 1.669e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 7.03998e-06
[add_recomputation]: 5.406e-05
[cse_after_recomputation]: 2.048e-05, [1]
    [Cycle 1]: 1.479e-05, [1]
        [cse]: 8.90999e-06
[auto_monad_reorder]: 1.599e-05
[get_jit_bprop_graph]: 2.58e-06
[rewriter_after_jit_bprop_graph]: 6.53e-06
[opt_after_jit_grad]: 0.00047189
[symbol_engine_optimizer]: 7.939e-05, [1]
    [Cycle 1]: 7.197e-05, [6]
        [build]: 4.65001e-06
        [elim_shapecalc]: 8.55999e-06
        [elim_not_effective]: 1.432e-05
        [opt_reshape]: 5.67999e-06
        [fold_const_symbol]: 9.56e-06
        [renormalize]: 6.30011e-07
[validate]: 4.234e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.00934587
[execute]: 1.17e-05
Sums
    bootstrap                            :     0.000338s :  1.44%
    type_inference                       :     0.007357s : 31.45%
    event_method                         :     0.000022s :  0.09%
    auto_monad                           :     0.000073s :  0.31%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000049s :  0.21%
    rewriter_before_opt_a                :     0.000081s :  0.35%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000081s :  0.35%
    jit_opt_a.loop_unroll                :     0.000037s :  0.16%
    jit_opt_a.a_1                        :     0.000767s :  3.28%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000008s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.08%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.002579s : 11.03%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.12%
    jit_opt_a.cse                        :     0.000047s :  0.20%
    jit_opt_a.replace_applicator         :     0.000035s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.07%
    rewriter_after_opt_a                 :     0.000041s :  0.17%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000848s :  3.62%
    jit_opt_b.frontend_op_eliminate      :     0.000087s :  0.37%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.07%
    cconv                                :     0.000032s :  0.14%
    loop_unroll                          :     0.000467s :  1.99%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000017s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000054s :  0.23%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.07%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.03%
    opt_after_jit_grad                   :     0.000472s :  2.02%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.18%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009346s : 39.95%
    execute                              :     0.000012s :  0.05%


Time group info:
------[substitution.]   0.000295    23
 0.77% :     0.000002s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 2.15% :     0.000006s :      3: substitution.graph_param_transform
86.32% :     0.000255s :      4: substitution.inline
 1.60% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.79% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000006s :      2: substitution.replace_old_param
 4.93% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007269     2
71.64% :     0.005207s :      1: type_inference.infer
28.36% :     0.002062s :      1: type_inference.specialize
------[replace.]   0.000087     6
69.17% :     0.000060s :      4: replace.inline
30.83% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000266     6
94.94% :     0.000253s :      4: match.inline
 5.06% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   691
 1.63% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.91% :     0.000001s :     10: predicate.addn_check_dump
 1.17% :     0.000002s :     10: predicate.addn_zero_filter
 2.71% :     0.000004s :     10: predicate.arithmetic_simplify
 1.29% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.92% :     0.000001s :     10: predicate.compare_switch_simplify
 1.12% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.14% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.43% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.36% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.99% :     0.000008s :     22: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.02% :     0.000003s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.07% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 1.61% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.07% :     0.000003s :     15: predicate.partial_eliminate
 1.29% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.85% :     0.000002s :     10: predicate.reduce_eliminate
 1.34% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.70% :     0.000002s :     10: predicate.reshape_eliminate
 1.34% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.99% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.07% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 1.44% :     0.000002s :      6: predicate.specialize_transform
 1.55% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 1.79% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.45% :     0.000010s :     53: predicate.switch_simplify
 1.28% :     0.000002s :     10: predicate.tile_eliminate
 1.21% :     0.000002s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.98% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000002s :     10: predicate.value_based_eliminate
 0.48% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.84% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000782    12
43.89% :     0.000343s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.16% :     0.000064s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.94% :     0.000375s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.032339    76
 0.18% :     0.000057s :      1: add_recomputation
 0.24% :     0.000076s :      1: auto_monad
 0.06% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.09% :     0.000353s :      1: bootstrap
 0.11% :     0.000035s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.08% :     0.000027s :      1: event_method
 0.05% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.68% :     0.008627s :      1: jit_opt_a
 0.48% :     0.000157s :      1: jit_opt_after_cconv
 0.40% :     0.000130s :      1: jit_opt_b
 1.47% :     0.000475s :      1: loop_unroll
 2.65% :     0.000858s :      1: mutable_eliminate
 3.18% :     0.001028s :     26: opt.transform.jit_opt_a
 0.16% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.30% :     0.000096s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.49% :     0.000481s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000052s :      1: py_interpret_to_execute
 0.06% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000019s :      1: remove_dup_value
 6.37% :     0.002060s :      1: renormalize.infer
 1.57% :     0.000506s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000044s :      1: rewriter_after_opt_a
 0.26% :     0.000084s :      1: rewriter_before_opt_a
 0.25% :     0.000082s :      1: symbol_engine_optimizer
28.97% :     0.009368s :      1: task_emit
22.83% :     0.007381s :      1: type_inference
 0.19% :     0.000062s :      1: validate


TotalTime = 0.0211637, [33]
[bootstrap]: 0.00041526
[type_inference]: 0.00516513
[event_method]: 1.938e-05
[auto_monad]: 6.746e-05
[graph_reusing]: 6.43003e-06
[pre_auto_parallel]: 2.86999e-06
[py_interpret_to_execute]: 4.027e-05
[rewriter_before_opt_a]: 7.444e-05
[expand_dump_flag]: 3.51999e-06
[jit_opt_a]: 0.00616335, [2]
    [Cycle 1]: 0.00188628, [27]
        [switch_simplify]: 5.682e-05
        [loop_unroll]: 2.929e-05
        [a_1]: 0.00058423
        [with_stream_mark]: 2.501e-05
        [recompute_prepare]: 7.45e-06
        [updatestate_depend_eliminate]: 4.19997e-06
        [updatestate_assign_eliminate]: 3.02002e-06
        [updatestate_loads_eliminate]: 2.84999e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 6.02999e-06
        [updatestate_useless_node_eliminater]: 5.14998e-06
        [accelerated_algorithm]: 5.74999e-06
        [meta_shard_fg_expand]: 2.41e-06
        [get_grad_eliminate_]: 5.66e-06
        [merge_forward]: 4.58001e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.526e-05
        [j_node_and_user_rematch]: 9.02e-06
        [meta_fg_expand]: 2.27001e-06
        [replace_old_param]: 9.64e-06
        [inline_without_move]: 4.80001e-06
        [renormalize]: 0.00082066
        [add_forward_monad_depend]: 7.65e-06
        [auto_monad_grad]: 2.54001e-06
        [auto_monad_eliminator]: 1.576e-05
        [cse]: 3.189e-05
        [replace_applicator]: 1.561e-05
    [Cycle 2]: 0.00036464, [27]
        [switch_simplify]: 5.61e-06
        [loop_unroll]: 5.28002e-06
        [a_1]: 9.358e-05
        [with_stream_mark]: 1.281e-05
        [recompute_prepare]: 5.59e-06
        [updatestate_depend_eliminate]: 2.69999e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 5.52999e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 5.78002e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 1.783e-05
        [merge_forward]: 3.08e-06
        [cell_reuse_recompute_pass]: 2.32001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.581e-05
        [j_node_and_user_rematch]: 8.22e-06
        [meta_fg_expand]: 1.85001e-06
        [replace_old_param]: 7.88001e-06
        [inline_without_move]: 7.53e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.51998e-06
        [auto_monad_grad]: 1.28002e-06
        [auto_monad_eliminator]: 6.64001e-06
        [cse]: 1.39e-05
        [replace_applicator]: 6.00002e-06
[py_interpret_to_execute_after_opt_a]: 1.418e-05
[rewriter_after_opt_a]: 4.246e-05
[convert_after_rewriter]: 7.36001e-06
[order_py_execute_after_rewriter]: 5.00001e-06
[mutable_eliminate]: 0.00075062
[jit_opt_b]: 5.332e-05, [1]
    [Cycle 1]: 4.48e-05, [2]
        [frontend_op_eliminate]: 1.78e-05
        [inline_after_opt_a]: 1.521e-05
[cconv]: 3.275e-05
[loop_unroll]: 0.00045174
[jit_opt_after_cconv]: 0.0001573, [1]
    [Cycle 1]: 0.00015043, [11]
        [c_1]: 2.127e-05
        [parameter_eliminate]: 3.81999e-06
        [updatestate_depend_eliminate]: 9.20999e-06
        [updatestate_assign_eliminate]: 2.30002e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 2.549e-05
        [call_graph_tuple_transform]: 2.075e-05
        [tuple_list_get_item_eliminator]: 5.97999e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 5.54e-06
[remove_dup_value]: 1.608e-05
[partial_unused_args_eliminate]: 2.25002e-06
[environ_conv]: 6.67002e-06
[add_recomputation]: 5.765e-05
[cse_after_recomputation]: 2.04e-05, [1]
    [Cycle 1]: 1.479e-05, [1]
        [cse]: 8.85001e-06
[auto_monad_reorder]: 1.596e-05
[get_jit_bprop_graph]: 2.19999e-06
[rewriter_after_jit_bprop_graph]: 8.48001e-06
[opt_after_jit_grad]: 0.000499
[symbol_engine_optimizer]: 7.797e-05, [1]
    [Cycle 1]: 7.02e-05, [6]
        [build]: 5.17999e-06
        [elim_shapecalc]: 8.18999e-06
        [elim_not_effective]: 1.351e-05
        [opt_reshape]: 6.02999e-06
        [fold_const_symbol]: 8.84e-06
        [renormalize]: 8.40024e-07
[validate]: 4.759e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.00668943
[execute]: 1.052e-05
Sums
    bootstrap                            :     0.000415s :  2.51%
    type_inference                       :     0.005165s : 31.18%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000067s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.24%
    rewriter_before_opt_a                :     0.000074s :  0.45%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.38%
    jit_opt_a.loop_unroll                :     0.000035s :  0.21%
    jit_opt_a.a_1                        :     0.000678s :  4.09%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.14%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000012s :  0.07%
    jit_opt_a.renormalize                :     0.000821s :  4.96%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.14%
    jit_opt_a.cse                        :     0.000046s :  0.28%
    jit_opt_a.replace_applicator         :     0.000022s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.09%
    rewriter_after_opt_a                 :     0.000042s :  0.26%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000751s :  4.53%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000033s :  0.20%
    loop_unroll                          :     0.000452s :  2.73%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000058s :  0.35%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000499s :  3.01%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000048s :  0.29%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006689s : 40.39%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000240    23
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.49% :     0.000006s :      3: substitution.graph_param_transform
84.72% :     0.000203s :      4: substitution.inline
 1.53% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.42% :     0.000006s :      2: substitution.replace_old_param
 5.29% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005087     2
83.70% :     0.004257s :      1: type_inference.infer
16.30% :     0.000829s :      1: type_inference.specialize
------[replace.]   0.000073     6
66.03% :     0.000048s :      4: replace.inline
33.97% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000212     6
94.52% :     0.000200s :      4: match.inline
 5.48% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   691
 1.17% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.98% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.88% :     0.000001s :     10: predicate.addn_check_dump
 1.10% :     0.000002s :     10: predicate.addn_zero_filter
 2.14% :     0.000003s :     10: predicate.arithmetic_simplify
 1.26% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 0.99% :     0.000001s :     10: predicate.depend_value_elim
 0.98% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.12% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.92% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.44% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
13.12% :     0.000018s :     10: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.95% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.10% :     0.000002s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.75% :     0.000008s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000002s :      6: predicate.less_batch_normalization
 1.30% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.43% :     0.000002s :     15: predicate.load_eliminater
 1.13% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.32% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.81% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.92% :     0.000001s :     10: predicate.merge_addn
 0.89% :     0.000001s :     10: predicate.minmaximum_grad
 2.04% :     0.000003s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 1.94% :     0.000003s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.28% :     0.000002s :     10: predicate.reduce_eliminate
 1.33% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.43% :     0.000002s :     18: predicate.replace_applicator
 0.44% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.00% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.06% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.56% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 1.92% :     0.000003s :     16: predicate.switch_defer_inline
 1.75% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.60% :     0.000009s :     53: predicate.switch_simplify
 1.04% :     0.000001s :     10: predicate.tile_eliminate
 0.98% :     0.000001s :     10: predicate.transpose_eliminate
 1.06% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.84% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.24% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.39% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.50% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000627    12
51.08% :     0.000320s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.28% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.65% :     0.000255s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023001    76
 0.27% :     0.000061s :      1: add_recomputation
 0.31% :     0.000070s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.90% :     0.000436s :      1: bootstrap
 0.15% :     0.000035s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.81% :     0.006167s :      1: jit_opt_a
 0.70% :     0.000160s :      1: jit_opt_after_cconv
 0.24% :     0.000056s :      1: jit_opt_b
 2.01% :     0.000462s :      1: loop_unroll
 3.32% :     0.000763s :      1: mutable_eliminate
 3.97% :     0.000914s :     26: opt.transform.jit_opt_a
 0.22% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000509s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000043s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000019s :      1: remove_dup_value
 2.04% :     0.000469s :      1: renormalize.infer
 1.49% :     0.000343s :      1: renormalize.specialize
 0.05% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000045s :      1: rewriter_after_opt_a
 0.34% :     0.000077s :      1: rewriter_before_opt_a
 0.35% :     0.000081s :      1: symbol_engine_optimizer
29.17% :     0.006709s :      1: task_emit
22.54% :     0.005184s :      1: type_inference
 0.31% :     0.000070s :      1: validate


TotalTime = 0.0175511, [30]
[bootstrap]: 0.00036113
[type_inference]: 0.00782024
[event_method]: 2.467e-05
[auto_monad]: 8.552e-05
[graph_reusing]: 6.31e-06
[pre_auto_parallel]: 2.66e-06
[py_interpret_to_execute]: 6.078e-05
[rewriter_before_opt_a]: 8.909e-05
[expand_dump_flag]: 3.18998e-06
[jit_opt_a]: 0.00671378, [2]
    [Cycle 1]: 0.00234197, [27]
        [switch_simplify]: 0.00014618
        [loop_unroll]: 3.47e-05
        [a_1]: 0.00078215
        [with_stream_mark]: 2.356e-05
        [recompute_prepare]: 7.65e-06
        [updatestate_depend_eliminate]: 3.76001e-06
        [updatestate_assign_eliminate]: 2.97002e-06
        [updatestate_loads_eliminate]: 2.64001e-06
        [parameter_eliminate]: 1.93002e-06
        [specialize_transform]: 5.87001e-06
        [updatestate_useless_node_eliminater]: 4.78001e-06
        [accelerated_algorithm]: 5.84e-06
        [meta_shard_fg_expand]: 2.88998e-06
        [get_grad_eliminate_]: 5.17e-06
        [merge_forward]: 3.41999e-06
        [cell_reuse_recompute_pass]: 1.54998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.506e-05
        [j_node_and_user_rematch]: 1.008e-05
        [meta_fg_expand]: 2.33002e-06
        [replace_old_param]: 9.91e-06
        [inline_without_move]: 5.35999e-06
        [renormalize]: 0.00098756
        [add_forward_monad_depend]: 6.41998e-06
        [auto_monad_grad]: 2.91e-06
        [auto_monad_eliminator]: 1.461e-05
        [cse]: 3.328e-05
        [replace_applicator]: 1.559e-05
    [Cycle 2]: 0.00033845, [27]
        [switch_simplify]: 6.18002e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 9.338e-05
        [with_stream_mark]: 1.192e-05
        [recompute_prepare]: 5.43002e-06
        [updatestate_depend_eliminate]: 2.80002e-06
        [updatestate_assign_eliminate]: 2.01998e-06
        [updatestate_loads_eliminate]: 1.83997e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 5.47001e-06
        [updatestate_useless_node_eliminater]: 4.69002e-06
        [accelerated_algorithm]: 4.73001e-06
        [meta_shard_fg_expand]: 1.03001e-06
        [get_grad_eliminate_]: 4.82e-06
        [merge_forward]: 2.42001e-06
        [cell_reuse_recompute_pass]: 2.11e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.666e-05
        [j_node_and_user_rematch]: 8.68001e-06
        [meta_fg_expand]: 1.81e-06
        [replace_old_param]: 7.94002e-06
        [inline_without_move]: 4.62998e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 9.29984e-07
        [auto_monad_eliminator]: 5.50001e-06
        [cse]: 1.201e-05
        [replace_applicator]: 5.62999e-06
[py_interpret_to_execute_after_opt_a]: 1.182e-05
[rewriter_after_opt_a]: 3.878e-05
[convert_after_rewriter]: 7.45998e-06
[order_py_execute_after_rewriter]: 4.93001e-06
[mutable_eliminate]: 0.0007383
[jit_opt_b]: 5.329e-05, [1]
    [Cycle 1]: 4.49e-05, [2]
        [frontend_op_eliminate]: 1.774e-05
        [inline_after_opt_a]: 1.582e-05
[cconv]: 2.989e-05
[loop_unroll]: 0.00044578
[jit_opt_after_cconv]: 0.00014753, [1]
    [Cycle 1]: 0.00014021, [11]
        [c_1]: 2.198e-05
        [parameter_eliminate]: 3.13e-06
        [updatestate_depend_eliminate]: 6.72002e-06
        [updatestate_assign_eliminate]: 2.28002e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [cse]: 2.145e-05
        [call_graph_tuple_transform]: 1.887e-05
        [tuple_list_get_item_eliminator]: 5.27999e-06
        [none_parameter_eliminate]: 1.63997e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 5.54e-06
[remove_dup_value]: 1.414e-05
[partial_unused_args_eliminate]: 2.11998e-06
[environ_conv]: 5.59998e-06
[add_recomputation]: 5.235e-05
[cse_after_recomputation]: 2.042e-05, [1]
    [Cycle 1]: 1.48e-05, [1]
        [cse]: 8.65001e-06
[auto_monad_reorder]: 1.521e-05
[get_jit_bprop_graph]: 1.92001e-06
[rewriter_after_jit_bprop_graph]: 6.46e-06
[opt_after_jit_grad]: 0.00047902
[symbol_engine_optimizer]: 7.696e-05, [1]
    [Cycle 1]: 6.943e-05, [6]
        [build]: 4.18999e-06
        [elim_shapecalc]: 8.15999e-06
        [elim_not_effective]: 1.282e-05
        [opt_reshape]: 5.86e-06
        [fold_const_symbol]: 9.33002e-06
        [renormalize]: 4.60015e-07
[validate]: 4.198e-05
Sums
    bootstrap                            :     0.000361s :  2.80%
    type_inference                       :     0.007820s : 60.71%
    event_method                         :     0.000025s :  0.19%
    auto_monad                           :     0.000086s :  0.66%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000061s :  0.47%
    rewriter_before_opt_a                :     0.000089s :  0.69%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000152s :  1.18%
    jit_opt_a.loop_unroll                :     0.000040s :  0.31%
    jit_opt_a.a_1                        :     0.000876s :  6.80%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000988s :  7.67%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.16%
    jit_opt_a.cse                        :     0.000045s :  0.35%
    jit_opt_a.replace_applicator         :     0.000021s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.09%
    rewriter_after_opt_a                 :     0.000039s :  0.30%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000738s :  5.73%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.12%
    cconv                                :     0.000030s :  0.23%
    loop_unroll                          :     0.000446s :  3.46%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.17%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000052s :  0.41%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000479s :  3.72%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000042s :  0.33%


Time group info:
------[substitution.]   0.000357    26
 0.56% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000002s :      2: substitution.fold_const_symbol
 1.59% :     0.000006s :      3: substitution.graph_param_transform
85.79% :     0.000306s :      6: substitution.inline
 1.04% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.14% :     0.000008s :      4: substitution.remove_not_recompute_node
 1.62% :     0.000006s :      2: substitution.replace_old_param
 2.93% :     0.000010s :      1: substitution.switch_simplify
 3.86% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007738     2
81.31% :     0.006291s :      1: type_inference.infer
18.69% :     0.001446s :      1: type_inference.specialize
------[replace.]   0.000160     9
47.86% :     0.000076s :      6: replace.inline
28.68% :     0.000046s :      1: replace.switch_simplify
23.46% :     0.000037s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000325     9
93.12% :     0.000303s :      6: match.inline
 2.98% :     0.000010s :      1: match.switch_simplify
 3.90% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   801
 1.37% :     0.000002s :     12: predicate.accumulaten_eliminater
 1.25% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     12: predicate.addn_check_dump
 1.54% :     0.000002s :     12: predicate.addn_zero_filter
 2.39% :     0.000003s :     12: predicate.arithmetic_simplify
 1.15% :     0.000002s :     12: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     12: predicate.compare_switch_simplify
 1.43% :     0.000002s :     12: predicate.depend_value_elim
 1.12% :     0.000002s :     12: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     12: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     12: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.55% :     0.000002s :     12: predicate.environ_add_const_eliminate
 1.11% :     0.000002s :     12: predicate.environ_get_add_eliminate
 1.28% :     0.000002s :     12: predicate.environ_get_depend_swap
 1.13% :     0.000002s :     12: predicate.environ_get_eliminate
 1.09% :     0.000001s :     12: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 5.47% :     0.000008s :     26: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     14: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     17: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.51% :     0.000006s :     40: predicate.loop_unroll_before_grad
 2.03% :     0.000003s :     15: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     12: predicate.merge_addn
 1.10% :     0.000002s :     12: predicate.minmaximum_grad
 1.54% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 1.99% :     0.000003s :     17: predicate.partial_eliminate
 1.66% :     0.000002s :     12: predicate.print_const_string_wrapper
 1.85% :     0.000003s :     12: predicate.reduce_eliminate
 1.41% :     0.000002s :     14: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     20: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     12: predicate.reshape_eliminate
 1.17% :     0.000002s :     12: predicate.row_tensor_add_zeros_like
 1.02% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.50% :     0.000002s :     12: predicate.same_eliminate
 0.42% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 0.85% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     12: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     12: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     20: predicate.switch_defer_inline
 2.30% :     0.000003s :     20: predicate.switch_layer_defer_inline
 8.00% :     0.000011s :     65: predicate.switch_simplify
 1.31% :     0.000002s :     12: predicate.tile_eliminate
 1.19% :     0.000002s :     12: predicate.transpose_eliminate
 1.69% :     0.000002s :     12: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000002s :     12: predicate.tuple_list_get_item_depend_reorder
 3.74% :     0.000005s :     20: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     14: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     17: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000004s :     23: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     12: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001053    15
44.58% :     0.000469s :      6: func_graph_cloner_run.FuncGraphClonerGraph
12.89% :     0.000136s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.53% :     0.000448s :      8: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019776    72
 0.28% :     0.000055s :      1: add_recomputation
 0.45% :     0.000089s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 1.92% :     0.000380s :      1: bootstrap
 0.16% :     0.000033s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.15% :     0.000029s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
33.97% :     0.006718s :      1: jit_opt_a
 0.76% :     0.000150s :      1: jit_opt_after_cconv
 0.28% :     0.000056s :      1: jit_opt_b
 2.29% :     0.000453s :      1: loop_unroll
 3.78% :     0.000747s :      1: mutable_eliminate
 6.02% :     0.001190s :     26: opt.transform.jit_opt_a
 0.25% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000027s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.46% :     0.000487s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.32% :     0.000064s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 2.84% :     0.000562s :      1: renormalize.infer
 2.11% :     0.000417s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000042s :      1: rewriter_after_opt_a
 0.47% :     0.000092s :      1: rewriter_before_opt_a
 0.40% :     0.000080s :      1: symbol_engine_optimizer
39.64% :     0.007839s :      1: type_inference


TotalTime = 0.0234554, [33]
[bootstrap]: 0.00036882
[type_inference]: 0.00657895
[event_method]: 2.062e-05
[auto_monad]: 7.266e-05
[graph_reusing]: 6.39001e-06
[pre_auto_parallel]: 2.51998e-06
[py_interpret_to_execute]: 4.699e-05
[rewriter_before_opt_a]: 8.038e-05
[expand_dump_flag]: 3.55003e-06
[jit_opt_a]: 0.00673698, [2]
    [Cycle 1]: 0.00215742, [27]
        [switch_simplify]: 7.035e-05
        [loop_unroll]: 3.065e-05
        [a_1]: 0.00064433
        [with_stream_mark]: 2.635e-05
        [recompute_prepare]: 8.06001e-06
        [updatestate_depend_eliminate]: 4.2e-06
        [updatestate_assign_eliminate]: 2.99001e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [parameter_eliminate]: 2.14999e-06
        [specialize_transform]: 6.73e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 5.71998e-06
        [meta_shard_fg_expand]: 2.05002e-06
        [get_grad_eliminate_]: 5.80002e-06
        [merge_forward]: 3.81999e-06
        [cell_reuse_recompute_pass]: 1.20999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.566e-05
        [j_node_and_user_rematch]: 9.67001e-06
        [meta_fg_expand]: 2.59001e-06
        [replace_old_param]: 9.96e-06
        [inline_without_move]: 5.32001e-06
        [renormalize]: 0.00102747
        [add_forward_monad_depend]: 7.11001e-06
        [auto_monad_grad]: 2.73e-06
        [auto_monad_eliminator]: 1.579e-05
        [cse]: 3.374e-05
        [replace_applicator]: 1.502e-05
    [Cycle 2]: 0.00034057, [27]
        [switch_simplify]: 6.04999e-06
        [loop_unroll]: 5.00999e-06
        [a_1]: 9.66e-05
        [with_stream_mark]: 1.214e-05
        [recompute_prepare]: 5.34e-06
        [updatestate_depend_eliminate]: 2.62001e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 5.04e-06
        [updatestate_useless_node_eliminater]: 5.40001e-06
        [accelerated_algorithm]: 5.52999e-06
        [meta_shard_fg_expand]: 1.50999e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 2.64001e-06
        [cell_reuse_recompute_pass]: 1.98002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.27e-05
        [j_node_and_user_rematch]: 8e-06
        [meta_fg_expand]: 1.77999e-06
        [replace_old_param]: 7.63001e-06
        [inline_without_move]: 4.89e-06
        [renormalize]: 4.00178e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 9.09989e-07
        [auto_monad_eliminator]: 6.58e-06
        [cse]: 1.111e-05
        [replace_applicator]: 9.29998e-06
[py_interpret_to_execute_after_opt_a]: 1.328e-05
[rewriter_after_opt_a]: 4.091e-05
[convert_after_rewriter]: 7.68001e-06
[order_py_execute_after_rewriter]: 5.19e-06
[mutable_eliminate]: 0.00071682
[jit_opt_b]: 5.37e-05, [1]
    [Cycle 1]: 4.606e-05, [2]
        [frontend_op_eliminate]: 1.806e-05
        [inline_after_opt_a]: 1.543e-05
[cconv]: 2.886e-05
[loop_unroll]: 0.00044515
[jit_opt_after_cconv]: 0.00015256, [1]
    [Cycle 1]: 0.0001458, [11]
        [c_1]: 2.144e-05
        [parameter_eliminate]: 2.75002e-06
        [updatestate_depend_eliminate]: 7.36999e-06
        [updatestate_assign_eliminate]: 2.69999e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 2.336e-05
        [call_graph_tuple_transform]: 2.087e-05
        [tuple_list_get_item_eliminator]: 6.17999e-06
        [none_parameter_eliminate]: 1.84998e-06
        [renormalize]: 6.40022e-07
        [switch_simplify]: 5.51002e-06
[remove_dup_value]: 1.582e-05
[partial_unused_args_eliminate]: 2.91999e-06
[environ_conv]: 6.51999e-06
[add_recomputation]: 5.403e-05
[cse_after_recomputation]: 2.188e-05, [1]
    [Cycle 1]: 1.592e-05, [1]
        [cse]: 9.42999e-06
[auto_monad_reorder]: 1.647e-05
[get_jit_bprop_graph]: 2.31998e-06
[rewriter_after_jit_bprop_graph]: 6.28e-06
[opt_after_jit_grad]: 0.0004931
[symbol_engine_optimizer]: 7.792e-05, [1]
    [Cycle 1]: 7.05e-05, [6]
        [build]: 4.81002e-06
        [elim_shapecalc]: 8.75999e-06
        [elim_not_effective]: 1.379e-05
        [opt_reshape]: 5.66e-06
        [fold_const_symbol]: 8.99998e-06
        [renormalize]: 4.69998e-07
[validate]: 4.453e-05
[backend_pass]: 1.14e-06
[task_emit]: 0.00706176
[execute]: 1.053e-05
Sums
    bootstrap                            :     0.000369s :  1.99%
    type_inference                       :     0.006579s : 35.51%
    event_method                         :     0.000021s :  0.11%
    auto_monad                           :     0.000073s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000047s :  0.25%
    rewriter_before_opt_a                :     0.000080s :  0.43%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.41%
    jit_opt_a.loop_unroll                :     0.000036s :  0.19%
    jit_opt_a.a_1                        :     0.000741s :  4.00%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001028s :  5.55%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.12%
    jit_opt_a.cse                        :     0.000045s :  0.24%
    jit_opt_a.replace_applicator         :     0.000024s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000041s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000717s :  3.87%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000029s :  0.16%
    loop_unroll                          :     0.000445s :  2.40%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000054s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000493s :  2.66%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000045s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007062s : 38.12%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000286    23
 0.88% :     0.000003s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 2.05% :     0.000006s :      3: substitution.graph_param_transform
87.40% :     0.000250s :      4: substitution.inline
 1.18% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.64% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.84% :     0.000005s :      2: substitution.replace_old_param
 4.56% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006488     2
83.90% :     0.005443s :      1: type_inference.infer
16.10% :     0.001045s :      1: type_inference.specialize
------[replace.]   0.000079     6
69.36% :     0.000055s :      4: replace.inline
30.64% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000259     6
95.46% :     0.000248s :      4: match.inline
 4.54% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.73% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.57% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.18% :     0.000001s :     10: predicate.addn_zero_filter
 2.41% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.57% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.74% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.46% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 5.84% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 1.55% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.15% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.37% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 2.15% :     0.000003s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.34% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.09% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.61% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.23% :     0.000003s :     16: predicate.switch_defer_inline
 1.89% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.65% :     0.000010s :     53: predicate.switch_simplify
 1.45% :     0.000002s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.89% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000784    12
47.96% :     0.000376s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.38% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.66% :     0.000350s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025544    76
 0.22% :     0.000057s :      1: add_recomputation
 0.30% :     0.000076s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.53% :     0.000390s :      1: bootstrap
 0.13% :     0.000032s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000026s :      1: event_method
 0.06% :     0.000017s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
26.39% :     0.006741s :      1: jit_opt_a
 0.61% :     0.000156s :      1: jit_opt_after_cconv
 0.22% :     0.000056s :      1: jit_opt_b
 1.78% :     0.000454s :      1: loop_unroll
 2.85% :     0.000727s :      1: mutable_eliminate
 3.82% :     0.000977s :     26: opt.transform.jit_opt_a
 0.20% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.96% :     0.000501s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000050s :      1: py_interpret_to_execute
 0.06% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.32% :     0.000592s :      1: renormalize.infer
 1.67% :     0.000426s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000044s :      1: rewriter_after_opt_a
 0.33% :     0.000084s :      1: rewriter_before_opt_a
 0.32% :     0.000081s :      1: symbol_engine_optimizer
27.73% :     0.007084s :      1: task_emit
25.83% :     0.006598s :      1: type_inference
 0.27% :     0.000069s :      1: validate


TotalTime = 0.0216925, [33]
[bootstrap]: 0.000466
[type_inference]: 0.00535521
[event_method]: 1.968e-05
[auto_monad]: 6.581e-05
[graph_reusing]: 6.17001e-06
[pre_auto_parallel]: 2.76e-06
[py_interpret_to_execute]: 4.003e-05
[rewriter_before_opt_a]: 7.321e-05
[expand_dump_flag]: 3.46001e-06
[jit_opt_a]: 0.00636783, [2]
    [Cycle 1]: 0.00183668, [27]
        [switch_simplify]: 6.002e-05
        [loop_unroll]: 2.915e-05
        [a_1]: 0.00058296
        [with_stream_mark]: 2.34e-05
        [recompute_prepare]: 7.92e-06
        [updatestate_depend_eliminate]: 3.81999e-06
        [updatestate_assign_eliminate]: 3.43999e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [parameter_eliminate]: 2.06e-06
        [specialize_transform]: 5.91998e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 6.24001e-06
        [meta_shard_fg_expand]: 2.39001e-06
        [get_grad_eliminate_]: 5.35999e-06
        [merge_forward]: 3.64002e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.469e-05
        [j_node_and_user_rematch]: 1.024e-05
        [meta_fg_expand]: 2.34999e-06
        [replace_old_param]: 1.084e-05
        [inline_without_move]: 5.44e-06
        [renormalize]: 0.0007688
        [add_forward_monad_depend]: 8.99003e-06
        [auto_monad_grad]: 2.58003e-06
        [auto_monad_eliminator]: 1.742e-05
        [cse]: 3.069e-05
        [replace_applicator]: 1.666e-05
    [Cycle 2]: 0.00048286, [27]
        [switch_simplify]: 6.23998e-06
        [loop_unroll]: 5.02e-06
        [a_1]: 0.00020663
        [with_stream_mark]: 1.532e-05
        [recompute_prepare]: 7.03e-06
        [updatestate_depend_eliminate]: 2.94999e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [parameter_eliminate]: 1.37999e-06
        [specialize_transform]: 6.12999e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 6.22001e-06
        [meta_shard_fg_expand]: 1.72001e-06
        [get_grad_eliminate_]: 5.05001e-06
        [merge_forward]: 3.13998e-06
        [cell_reuse_recompute_pass]: 2.66999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.534e-05
        [j_node_and_user_rematch]: 7.73999e-06
        [meta_fg_expand]: 2.54001e-06
        [replace_old_param]: 8.60999e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.01003e-06
        [auto_monad_grad]: 1.17e-06
        [auto_monad_eliminator]: 6.21e-06
        [cse]: 1.561e-05
        [replace_applicator]: 5.57001e-06
[py_interpret_to_execute_after_opt_a]: 1.488e-05
[rewriter_after_opt_a]: 4.036e-05
[convert_after_rewriter]: 8.36002e-06
[order_py_execute_after_rewriter]: 6.03002e-06
[mutable_eliminate]: 0.00080801
[jit_opt_b]: 5.517e-05, [1]
    [Cycle 1]: 4.607e-05, [2]
        [frontend_op_eliminate]: 1.763e-05
        [inline_after_opt_a]: 1.591e-05
[cconv]: 3.102e-05
[loop_unroll]: 0.00047969
[jit_opt_after_cconv]: 0.00015655, [1]
    [Cycle 1]: 0.00014992, [11]
        [c_1]: 2.265e-05
        [parameter_eliminate]: 3.55e-06
        [updatestate_depend_eliminate]: 7.39002e-06
        [updatestate_assign_eliminate]: 2.74001e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [cse]: 2.28e-05
        [call_graph_tuple_transform]: 2.277e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 7.00005e-07
        [switch_simplify]: 6.39001e-06
[remove_dup_value]: 1.581e-05
[partial_unused_args_eliminate]: 2.11998e-06
[environ_conv]: 6.07999e-06
[add_recomputation]: 5.669e-05
[cse_after_recomputation]: 2.018e-05, [1]
    [Cycle 1]: 1.452e-05, [1]
        [cse]: 8.76002e-06
[auto_monad_reorder]: 1.604e-05
[get_jit_bprop_graph]: 2.02999e-06
[rewriter_after_jit_bprop_graph]: 7.50998e-06
[opt_after_jit_grad]: 0.00048265
[symbol_engine_optimizer]: 7.376e-05, [1]
    [Cycle 1]: 6.694e-05, [6]
        [build]: 4.28001e-06
        [elim_shapecalc]: 7.93999e-06
        [elim_not_effective]: 1.348e-05
        [opt_reshape]: 5.46e-06
        [fold_const_symbol]: 8.22e-06
        [renormalize]: 6.90023e-07
[validate]: 4.458e-05
[backend_pass]: 8.90024e-07
[task_emit]: 0.00671502
[execute]: 1.03e-05
Sums
    bootstrap                            :     0.000466s :  2.75%
    type_inference                       :     0.005355s : 31.61%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000066s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.24%
    rewriter_before_opt_a                :     0.000073s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000790s :  4.66%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000769s :  4.54%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.14%
    jit_opt_a.cse                        :     0.000046s :  0.27%
    jit_opt_a.replace_applicator         :     0.000022s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.09%
    rewriter_after_opt_a                 :     0.000040s :  0.24%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000808s :  4.77%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000031s :  0.18%
    loop_unroll                          :     0.000480s :  2.83%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000057s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000483s :  2.85%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006715s : 39.63%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000249    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.66% :     0.000007s :      3: substitution.graph_param_transform
84.58% :     0.000211s :      4: substitution.inline
 1.46% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.87% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.66% :     0.000007s :      2: substitution.replace_old_param
 5.30% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005272     2
83.94% :     0.004426s :      1: type_inference.infer
16.06% :     0.000847s :      1: type_inference.specialize
------[replace.]   0.000073     6
67.04% :     0.000049s :      4: replace.inline
32.96% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000220     6
94.61% :     0.000208s :      4: match.inline
 5.39% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.60% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.74% :     0.000002s :     10: predicate.addn_zero_filter
 2.06% :     0.000003s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.50% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.51% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.04% :     0.000001s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.61% :     0.000007s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.70% :     0.000002s :      6: predicate.less_batch_normalization
 1.83% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.58% :     0.000002s :     15: predicate.load_eliminater
 1.56% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.66% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.81% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.31% :     0.000003s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.84% :     0.000002s :     10: predicate.reduce_eliminate
 1.66% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.79% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.60% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.13% :     0.000003s :     16: predicate.switch_defer_inline
 1.87% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.25% :     0.000010s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.03% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.99% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.20% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.40% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000646    12
51.15% :     0.000331s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.18% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.67% :     0.000263s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023586    76
 0.25% :     0.000060s :      1: add_recomputation
 0.30% :     0.000070s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 2.08% :     0.000490s :      1: bootstrap
 0.14% :     0.000033s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000025s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
27.02% :     0.006372s :      1: jit_opt_a
 0.68% :     0.000160s :      1: jit_opt_after_cconv
 0.25% :     0.000058s :      1: jit_opt_b
 2.07% :     0.000489s :      1: loop_unroll
 3.47% :     0.000819s :      1: mutable_eliminate
 4.32% :     0.001019s :     26: opt.transform.jit_opt_a
 0.23% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.08% :     0.000491s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000043s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 1.81% :     0.000427s :      1: renormalize.infer
 1.41% :     0.000333s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000043s :      1: rewriter_after_opt_a
 0.32% :     0.000076s :      1: rewriter_before_opt_a
 0.32% :     0.000077s :      1: symbol_engine_optimizer
28.55% :     0.006734s :      1: task_emit
22.79% :     0.005375s :      1: type_inference
 0.28% :     0.000066s :      1: validate


TotalTime = 0.025954, [33]
[bootstrap]: 0.00037489
[type_inference]: 0.00685754
[event_method]: 2.113e-05
[auto_monad]: 7.486e-05
[graph_reusing]: 6.49001e-06
[pre_auto_parallel]: 2.97002e-06
[py_interpret_to_execute]: 5.02e-05
[rewriter_before_opt_a]: 8.102e-05
[expand_dump_flag]: 3.19001e-06
[jit_opt_a]: 0.00684948, [2]
    [Cycle 1]: 0.00211256, [27]
        [switch_simplify]: 7.314e-05
        [loop_unroll]: 3.254e-05
        [a_1]: 0.00063676
        [with_stream_mark]: 2.525e-05
        [recompute_prepare]: 7.41999e-06
        [updatestate_depend_eliminate]: 4.60999e-06
        [updatestate_assign_eliminate]: 3.04001e-06
        [updatestate_loads_eliminate]: 2.69999e-06
        [parameter_eliminate]: 2.73e-06
        [specialize_transform]: 6.16e-06
        [updatestate_useless_node_eliminater]: 5.36002e-06
        [accelerated_algorithm]: 6.28e-06
        [meta_shard_fg_expand]: 2.89001e-06
        [get_grad_eliminate_]: 5.56e-06
        [merge_forward]: 3.58e-06
        [cell_reuse_recompute_pass]: 1.26002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.474e-05
        [j_node_and_user_rematch]: 9.91998e-06
        [meta_fg_expand]: 2.55002e-06
        [replace_old_param]: 1.265e-05
        [inline_without_move]: 5.52001e-06
        [renormalize]: 0.00097825
        [add_forward_monad_depend]: 6.98e-06
        [auto_monad_grad]: 2.46e-06
        [auto_monad_eliminator]: 1.698e-05
        [cse]: 3.168e-05
        [replace_applicator]: 1.383e-05
    [Cycle 2]: 0.00033767, [27]
        [switch_simplify]: 5.88998e-06
        [loop_unroll]: 5.40999e-06
        [a_1]: 9.37e-05
        [with_stream_mark]: 1.056e-05
        [recompute_prepare]: 5.19e-06
        [updatestate_depend_eliminate]: 2.87002e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.23002e-06
        [specialize_transform]: 5.25001e-06
        [updatestate_useless_node_eliminater]: 5.41998e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 1.31002e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 3.16999e-06
        [cell_reuse_recompute_pass]: 2.09999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.329e-05
        [j_node_and_user_rematch]: 8.1e-06
        [meta_fg_expand]: 1.72999e-06
        [replace_old_param]: 8e-06
        [inline_without_move]: 4.78001e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 5.96998e-06
        [cse]: 1.114e-05
        [replace_applicator]: 5.37001e-06
[py_interpret_to_execute_after_opt_a]: 1.383e-05
[rewriter_after_opt_a]: 4.014e-05
[convert_after_rewriter]: 7.23999e-06
[order_py_execute_after_rewriter]: 8.35001e-06
[mutable_eliminate]: 0.001761
[jit_opt_b]: 5.898e-05, [1]
    [Cycle 1]: 4.94e-05, [2]
        [frontend_op_eliminate]: 1.991e-05
        [inline_after_opt_a]: 1.656e-05
[cconv]: 3.284e-05
[loop_unroll]: 0.00052731
[jit_opt_after_cconv]: 0.00016204, [1]
    [Cycle 1]: 0.0001554, [11]
        [c_1]: 2.325e-05
        [parameter_eliminate]: 3.36999e-06
        [updatestate_depend_eliminate]: 6.81001e-06
        [updatestate_assign_eliminate]: 2.56998e-06
        [updatestate_loads_eliminate]: 2.64001e-06
        [cse]: 2.86e-05
        [call_graph_tuple_transform]: 2.181e-05
        [tuple_list_get_item_eliminator]: 5.75001e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 8.29983e-07
        [switch_simplify]: 5.58002e-06
[remove_dup_value]: 1.582e-05
[partial_unused_args_eliminate]: 2.61e-06
[environ_conv]: 5.92999e-06
[add_recomputation]: 5.29e-05
[cse_after_recomputation]: 2.139e-05, [1]
    [Cycle 1]: 1.528e-05, [1]
        [cse]: 9.57001e-06
[auto_monad_reorder]: 1.541e-05
[get_jit_bprop_graph]: 2.26998e-06
[rewriter_after_jit_bprop_graph]: 7.88001e-06
[opt_after_jit_grad]: 0.00048234
[symbol_engine_optimizer]: 7.978e-05, [1]
    [Cycle 1]: 7.269e-05, [6]
        [build]: 5.32001e-06
        [elim_shapecalc]: 9.15999e-06
        [elim_not_effective]: 1.379e-05
        [opt_reshape]: 6.14001e-06
        [fold_const_symbol]: 9.34998e-06
        [renormalize]: 7.39994e-07
[validate]: 4.51e-05
[backend_pass]: 1.20999e-06
[task_emit]: 0.00801097
[execute]: 1.079e-05
Sums
    bootstrap                            :     0.000375s :  1.80%
    type_inference                       :     0.006858s : 32.90%
    event_method                         :     0.000021s :  0.10%
    auto_monad                           :     0.000075s :  0.36%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000050s :  0.24%
    rewriter_before_opt_a                :     0.000081s :  0.39%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000079s :  0.38%
    jit_opt_a.loop_unroll                :     0.000038s :  0.18%
    jit_opt_a.a_1                        :     0.000730s :  3.50%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000978s :  4.69%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.11%
    jit_opt_a.cse                        :     0.000043s :  0.21%
    jit_opt_a.replace_applicator         :     0.000019s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.07%
    rewriter_after_opt_a                 :     0.000040s :  0.19%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000008s :  0.04%
    mutable_eliminate                    :     0.001761s :  8.45%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.08%
    cconv                                :     0.000033s :  0.16%
    loop_unroll                          :     0.000527s :  2.53%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000053s :  0.25%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000482s :  2.31%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008011s : 38.44%
    execute                              :     0.000011s :  0.05%


TotalTime = 0.0213904, [33]
[bootstrap]: 0.0004012
[type_inference]: 0.00576274
[event_method]: 2.066e-05
[auto_monad]: 6.48e-05
[graph_reusing]: 5.91e-06
[pre_auto_parallel]: 2.71e-06
[py_interpret_to_execute]: 4.085e-05
[rewriter_before_opt_a]: 7.225e-05
[expand_dump_flag]: 3.16999e-06
[jit_opt_a]: 0.00591581, [2]
    [Cycle 1]: 0.00186766, [27]
        [switch_simplify]: 6.153e-05
        [loop_unroll]: 2.93e-05
        [a_1]: 0.00056483
        [with_stream_mark]: 2.377e-05
        [recompute_prepare]: 7.53e-06
        [updatestate_depend_eliminate]: 3.74002e-06
        [updatestate_assign_eliminate]: 2.84999e-06
        [updatestate_loads_eliminate]: 2.59001e-06
        [parameter_eliminate]: 1.76e-06
        [specialize_transform]: 5.61e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 5.27999e-06
        [meta_shard_fg_expand]: 1.95001e-06
        [get_grad_eliminate_]: 5.21998e-06
        [merge_forward]: 6.99e-05
        [cell_reuse_recompute_pass]: 2.01e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.539e-05
        [j_node_and_user_rematch]: 9.59e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 9.29998e-06
        [inline_without_move]: 5.20001e-06
        [renormalize]: 0.00077769
        [add_forward_monad_depend]: 6.09999e-06
        [auto_monad_grad]: 2.73e-06
        [auto_monad_eliminator]: 1.416e-05
        [cse]: 2.7e-05
        [replace_applicator]: 1.252e-05
    [Cycle 2]: 0.00032268, [27]
        [switch_simplify]: 5.38997e-06
        [loop_unroll]: 4.76002e-06
        [a_1]: 8.883e-05
        [with_stream_mark]: 1.038e-05
        [recompute_prepare]: 5.10001e-06
        [updatestate_depend_eliminate]: 2.77002e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 4.78001e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 4.44002e-06
        [merge_forward]: 2.71e-06
        [cell_reuse_recompute_pass]: 1.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.314e-05
        [j_node_and_user_rematch]: 7.96001e-06
        [meta_fg_expand]: 1.72999e-06
        [replace_old_param]: 7.55e-06
        [inline_without_move]: 4.69998e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 5.18002e-06
        [cse]: 1.109e-05
        [replace_applicator]: 4.97e-06
[py_interpret_to_execute_after_opt_a]: 1.171e-05
[rewriter_after_opt_a]: 3.648e-05
[convert_after_rewriter]: 6.98e-06
[order_py_execute_after_rewriter]: 4.87e-06
[mutable_eliminate]: 0.00066518
[jit_opt_b]: 5.253e-05, [1]
    [Cycle 1]: 4.489e-05, [2]
        [frontend_op_eliminate]: 1.861e-05
        [inline_after_opt_a]: 1.526e-05
[cconv]: 2.951e-05
[loop_unroll]: 0.00044769
[jit_opt_after_cconv]: 0.00014607, [1]
    [Cycle 1]: 0.00013838, [11]
        [c_1]: 1.985e-05
        [parameter_eliminate]: 3.55e-06
        [updatestate_depend_eliminate]: 7.06999e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 2.125e-05
        [call_graph_tuple_transform]: 2.021e-05
        [tuple_list_get_item_eliminator]: 5.02999e-06
        [none_parameter_eliminate]: 1.76e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 5.33002e-06
[remove_dup_value]: 1.402e-05
[partial_unused_args_eliminate]: 2.21998e-06
[environ_conv]: 5.26002e-06
[add_recomputation]: 5.341e-05
[cse_after_recomputation]: 2.143e-05, [1]
    [Cycle 1]: 1.592e-05, [1]
        [cse]: 1.009e-05
[auto_monad_reorder]: 1.41e-05
[get_jit_bprop_graph]: 2.18002e-06
[rewriter_after_jit_bprop_graph]: 6.44001e-06
[opt_after_jit_grad]: 0.00049297
[symbol_engine_optimizer]: 8.031e-05, [1]
    [Cycle 1]: 7.239e-05, [6]
        [build]: 4.53999e-06
        [elim_shapecalc]: 7.85e-06
        [elim_not_effective]: 1.419e-05
        [opt_reshape]: 6.36e-06
        [fold_const_symbol]: 9.64999e-06
        [renormalize]: 6.09987e-07
[validate]: 4.078e-05
[backend_pass]: 1.52001e-06
[task_emit]: 0.00671115
[execute]: 9.97999e-06
Sums
    bootstrap                            :     0.000401s :  2.36%
    type_inference                       :     0.005763s : 33.92%
    event_method                         :     0.000021s :  0.12%
    auto_monad                           :     0.000065s :  0.38%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000041s :  0.24%
    rewriter_before_opt_a                :     0.000072s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000654s :  3.85%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000073s :  0.43%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000778s :  4.58%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.11%
    jit_opt_a.cse                        :     0.000038s :  0.22%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000665s :  3.92%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000030s :  0.17%
    loop_unroll                          :     0.000448s :  2.64%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000053s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000493s :  2.90%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.24%
    backend_pass                         :     0.000002s :  0.01%
    task_emit                            :     0.006711s : 39.51%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000275    23
 0.71% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.23% :     0.000006s :      3: substitution.graph_param_transform
86.30% :     0.000237s :      4: substitution.inline
 1.53% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.69% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000006s :      2: substitution.replace_old_param
 4.98% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006768     2
84.09% :     0.005692s :      1: type_inference.infer
15.91% :     0.001076s :      1: type_inference.specialize
------[replace.]   0.000080     6
68.87% :     0.000055s :      4: replace.inline
31.13% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000247     6
94.91% :     0.000234s :      4: match.inline
 5.09% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.32% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.21% :     0.000002s :     10: predicate.addn_zero_filter
 2.54% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.80% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.37% :     0.000007s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.71% :     0.000002s :      6: predicate.less_batch_normalization
 1.37% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.90% :     0.000002s :     15: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.26% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.48% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 0.95% :     0.000001s :     10: predicate.minmaximum_grad
 2.28% :     0.000003s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     15: predicate.partial_eliminate
 1.06% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.99% :     0.000003s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.49% :     0.000001s :      3: predicate.reset_defer_inline
 1.29% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.10% :     0.000001s :      6: predicate.special_op_eliminate
 1.13% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.14% :     0.000003s :     16: predicate.switch_defer_inline
 2.00% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.86% :     0.000010s :     53: predicate.switch_simplify
 1.31% :     0.000002s :     10: predicate.tile_eliminate
 1.04% :     0.000001s :     10: predicate.transpose_eliminate
 1.55% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.44% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.87% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.65% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000807    12
48.88% :     0.000394s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.81% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.30% :     0.000357s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028014    76
 0.20% :     0.000056s :      1: add_recomputation
 0.28% :     0.000079s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.42% :     0.000399s :      1: bootstrap
 0.13% :     0.000036s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.10% :     0.000027s :      1: event_method
 0.06% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
24.46% :     0.006853s :      1: jit_opt_a
 0.59% :     0.000165s :      1: jit_opt_after_cconv
 0.22% :     0.000062s :      1: jit_opt_b
 1.91% :     0.000536s :      1: loop_unroll
 6.34% :     0.001776s :      1: mutable_eliminate
 3.46% :     0.000970s :     26: opt.transform.jit_opt_a
 0.19% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000029s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.75% :     0.000490s :      1: opt_after_jit_grad
 0.04% :     0.000011s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000053s :      1: py_interpret_to_execute
 0.06% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 1.94% :     0.000543s :      1: renormalize.infer
 1.52% :     0.000426s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000043s :      1: rewriter_after_opt_a
 0.30% :     0.000085s :      1: rewriter_before_opt_a
 0.30% :     0.000083s :      1: symbol_engine_optimizer
28.72% :     0.008047s :      1: task_emit
24.56% :     0.006881s :      1: type_inference
 0.24% :     0.000066s :      1: validate


Time group info:
------[substitution.]   0.000239    23
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.86% :     0.000002s :      2: substitution.fold_const_symbol
 2.51% :     0.000006s :      3: substitution.graph_param_transform
85.44% :     0.000204s :      4: substitution.inline
 1.39% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.04% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.98% :     0.000005s :      2: substitution.replace_old_param
 5.01% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005678     2
84.34% :     0.004789s :      1: type_inference.infer
15.66% :     0.000889s :      1: type_inference.specialize
------[replace.]   0.000068     6
69.03% :     0.000047s :      4: replace.inline
30.97% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000213     6
94.91% :     0.000202s :      4: match.inline
 5.09% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000001s :     10: predicate.addn_zero_filter
 2.07% :     0.000002s :     10: predicate.arithmetic_simplify
 1.48% :     0.000002s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000001s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.74% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.50% :     0.000001s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.22% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.36% :     0.000002s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.92% :     0.000007s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.98% :     0.000001s :      6: predicate.less_batch_normalization
 1.76% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.80% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.96% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.98% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000002s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 1.20% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000001s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.60% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.54% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.24% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.82% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.93% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.69% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000712    12
53.86% :     0.000383s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.93% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.22% :     0.000279s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023136    76
 0.24% :     0.000056s :      1: add_recomputation
 0.29% :     0.000068s :      1: auto_monad
 0.07% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.85% :     0.000427s :      1: bootstrap
 0.14% :     0.000032s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.11% :     0.000025s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.59% :     0.005920s :      1: jit_opt_a
 0.64% :     0.000149s :      1: jit_opt_after_cconv
 0.24% :     0.000055s :      1: jit_opt_b
 1.97% :     0.000456s :      1: loop_unroll
 2.93% :     0.000677s :      1: mutable_eliminate
 3.76% :     0.000869s :     26: opt.transform.jit_opt_a
 0.20% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.17% :     0.000502s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000043s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.88% :     0.000434s :      1: renormalize.infer
 1.45% :     0.000335s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000040s :      1: rewriter_after_opt_a
 0.33% :     0.000075s :      1: rewriter_before_opt_a
 0.36% :     0.000083s :      1: symbol_engine_optimizer
29.10% :     0.006732s :      1: task_emit
25.00% :     0.005783s :      1: type_inference
 0.28% :     0.000064s :      1: validate


TotalTime = 0.0262914, [33]
[bootstrap]: 0.00038678
[type_inference]: 0.00534033
[event_method]: 1.794e-05
[auto_monad]: 6.207e-05
[graph_reusing]: 5.49e-06
[pre_auto_parallel]: 2.31e-06
[py_interpret_to_execute]: 3.509e-05
[rewriter_before_opt_a]: 6.929e-05
[expand_dump_flag]: 3.32997e-06
[jit_opt_a]: 0.0120753, [2]
    [Cycle 1]: 0.00174172, [27]
        [switch_simplify]: 5.496e-05
        [loop_unroll]: 2.992e-05
        [a_1]: 0.00053216
        [with_stream_mark]: 1.621e-05
        [recompute_prepare]: 8.48001e-06
        [updatestate_depend_eliminate]: 3.25002e-06
        [updatestate_assign_eliminate]: 2.15002e-06
        [updatestate_loads_eliminate]: 1.79998e-06
        [parameter_eliminate]: 1.49e-06
        [specialize_transform]: 6.23e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 6.31998e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 5.45001e-06
        [merge_forward]: 2.89001e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.109e-05
        [j_node_and_user_rematch]: 8.36002e-06
        [meta_fg_expand]: 1.86e-06
        [replace_old_param]: 8.90999e-06
        [inline_without_move]: 5.42001e-06
        [renormalize]: 0.00080157
        [add_forward_monad_depend]: 6.22001e-06
        [auto_monad_grad]: 1.79e-06
        [auto_monad_eliminator]: 1.291e-05
        [cse]: 1.746e-05
        [replace_applicator]: 1.314e-05
    [Cycle 2]: 0.00034845, [27]
        [switch_simplify]: 5.87001e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 9.398e-05
        [with_stream_mark]: 1.17e-05
        [recompute_prepare]: 5.47001e-06
        [updatestate_depend_eliminate]: 3.03e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [parameter_eliminate]: 1.26002e-06
        [specialize_transform]: 5.26998e-06
        [updatestate_useless_node_eliminater]: 5.49998e-06
        [accelerated_algorithm]: 5.87999e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 5.04e-06
        [merge_forward]: 2.58e-06
        [cell_reuse_recompute_pass]: 2.89001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.388e-05
        [j_node_and_user_rematch]: 9.74e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 8.53001e-06
        [inline_without_move]: 5.09e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.84e-06
        [auto_monad_grad]: 1.04e-06
        [auto_monad_eliminator]: 5.42001e-06
        [cse]: 1.098e-05
        [replace_applicator]: 5.30999e-06
[py_interpret_to_execute_after_opt_a]: 1.227e-05
[rewriter_after_opt_a]: 3.652e-05
[convert_after_rewriter]: 7.46999e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.00073346
[jit_opt_b]: 5.203e-05, [1]
    [Cycle 1]: 4.44e-05, [2]
        [frontend_op_eliminate]: 1.656e-05
        [inline_after_opt_a]: 1.599e-05
[cconv]: 2.805e-05
[loop_unroll]: 0.00049203
[jit_opt_after_cconv]: 0.00016098, [1]
    [Cycle 1]: 0.00015348, [11]
        [c_1]: 2.257e-05
        [parameter_eliminate]: 3.03998e-06
        [updatestate_depend_eliminate]: 7.13e-06
        [updatestate_assign_eliminate]: 3.11999e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [cse]: 2.637e-05
        [call_graph_tuple_transform]: 2.131e-05
        [tuple_list_get_item_eliminator]: 6.02001e-06
        [none_parameter_eliminate]: 1.64998e-06
        [renormalize]: 6.60017e-07
        [switch_simplify]: 5.46e-06
[remove_dup_value]: 1.155e-05
[partial_unused_args_eliminate]: 2.67001e-06
[environ_conv]: 5.92001e-06
[add_recomputation]: 4.527e-05
[cse_after_recomputation]: 2.498e-05, [1]
    [Cycle 1]: 1.862e-05, [1]
        [cse]: 1.072e-05
[auto_monad_reorder]: 1.32e-05
[get_jit_bprop_graph]: 1.94e-06
[rewriter_after_jit_bprop_graph]: 6.91999e-06
[opt_after_jit_grad]: 0.00053557
[symbol_engine_optimizer]: 8.102e-05, [1]
    [Cycle 1]: 7.387e-05, [6]
        [build]: 3.01999e-06
        [elim_shapecalc]: 8.52998e-06
        [elim_not_effective]: 1.567e-05
        [opt_reshape]: 6.19001e-06
        [fold_const_symbol]: 9.09e-06
        [renormalize]: 5.39992e-07
[validate]: 3.494e-05
[backend_pass]: 6.50005e-07
[task_emit]: 0.00576459
[execute]: 1.029e-05
Sums
    bootstrap                            :     0.000387s :  2.47%
    type_inference                       :     0.005340s : 34.13%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000062s :  0.40%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000035s :  0.22%
    rewriter_before_opt_a                :     0.000069s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000626s :  4.00%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000802s :  5.12%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000028s :  0.18%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000037s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000733s :  4.69%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000028s :  0.18%
    loop_unroll                          :     0.000492s :  3.14%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000026s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000012s :  0.07%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000045s :  0.29%
    cse_after_recomputation.cse          :     0.000011s :  0.07%
    auto_monad_reorder                   :     0.000013s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000536s :  3.42%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.22%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005765s : 36.84%
    execute                              :     0.000010s :  0.07%


Time group info:
------[substitution.]   0.000203    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.62% :     0.000005s :      3: substitution.graph_param_transform
84.60% :     0.000172s :      4: substitution.inline
 1.65% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.95% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.21% :     0.000004s :      2: substitution.replace_old_param
 5.48% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005271     2
83.40% :     0.004396s :      1: type_inference.infer
16.60% :     0.000875s :      1: type_inference.specialize
------[replace.]   0.000068     6
66.33% :     0.000045s :      4: replace.inline
33.67% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000179     6
94.42% :     0.000169s :      4: match.inline
 5.58% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.48% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000002s :     10: predicate.addn_zero_filter
 1.84% :     0.000002s :     10: predicate.arithmetic_simplify
 1.49% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.59% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.52% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.32% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.25% :     0.000001s :      6: predicate.get_grad_eliminate
 0.34% :     0.000000s :      3: predicate.graph_param_transform
 5.05% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.85% :     0.000002s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 1.81% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.13% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.63% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000002s :     15: predicate.partial_eliminate
 1.34% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.77% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.89% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.51% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.38% :     0.000009s :     53: predicate.switch_simplify
 1.42% :     0.000002s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.95% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000599    12
48.11% :     0.000288s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.81% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.08% :     0.000276s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028044    76
 0.17% :     0.000049s :      1: add_recomputation
 0.23% :     0.000065s :      1: auto_monad
 0.06% :     0.000016s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.46% :     0.000409s :      1: bootstrap
 0.11% :     0.000031s :      1: cconv
 0.04% :     0.000012s :      1: convert_after_rewriter
 0.10% :     0.000027s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.08% :     0.000021s :      1: event_method
 0.05% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
43.08% :     0.012080s :      1: jit_opt_a
 0.59% :     0.000164s :      1: jit_opt_after_cconv
 0.20% :     0.000055s :      1: jit_opt_b
 1.79% :     0.000503s :      1: loop_unroll
 2.65% :     0.000743s :      1: mutable_eliminate
 3.00% :     0.000841s :     26: opt.transform.jit_opt_a
 0.18% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.94% :     0.000545s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.14% :     0.000038s :      1: py_interpret_to_execute
 0.05% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000014s :      1: remove_dup_value
 1.48% :     0.000414s :      1: renormalize.infer
 1.35% :     0.000380s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000039s :      1: rewriter_after_opt_a
 0.26% :     0.000072s :      1: rewriter_before_opt_a
 0.30% :     0.000083s :      1: symbol_engine_optimizer
20.64% :     0.005787s :      1: task_emit
19.09% :     0.005353s :      1: type_inference
 0.19% :     0.000054s :      1: validate


TotalTime = 0.0403281, [33]
[bootstrap]: 0.00038657
[type_inference]: 0.00511904
[event_method]: 1.954e-05
[auto_monad]: 6.16e-05
[graph_reusing]: 4.75999e-06
[pre_auto_parallel]: 2.31e-06
[py_interpret_to_execute]: 4.192e-05
[rewriter_before_opt_a]: 7.287e-05
[expand_dump_flag]: 3.01001e-06
[jit_opt_a]: 0.00679046, [2]
    [Cycle 1]: 0.0020504, [27]
        [switch_simplify]: 5.7e-05
        [loop_unroll]: 3.097e-05
        [a_1]: 0.000599
        [with_stream_mark]: 2.005e-05
        [recompute_prepare]: 7.40998e-06
        [updatestate_depend_eliminate]: 3.55998e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.72001e-06
        [parameter_eliminate]: 1.42e-06
        [specialize_transform]: 7.65e-06
        [updatestate_useless_node_eliminater]: 6.38003e-06
        [accelerated_algorithm]: 6.67002e-06
        [meta_shard_fg_expand]: 2.86999e-06
        [get_grad_eliminate_]: 5.82001e-06
        [merge_forward]: 4.05e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.549e-05
        [j_node_and_user_rematch]: 1.102e-05
        [meta_fg_expand]: 2.73e-06
        [replace_old_param]: 9.50001e-06
        [inline_without_move]: 6.44999e-06
        [renormalize]: 0.00096217
        [add_forward_monad_depend]: 5.79e-06
        [auto_monad_grad]: 2.66e-06
        [auto_monad_eliminator]: 1.509e-05
        [cse]: 2.883e-05
        [replace_applicator]: 1.544e-05
    [Cycle 2]: 0.00033813, [27]
        [switch_simplify]: 5.56998e-06
        [loop_unroll]: 4.82e-06
        [a_1]: 9.425e-05
        [with_stream_mark]: 1.161e-05
        [recompute_prepare]: 5.29e-06
        [updatestate_depend_eliminate]: 2.94001e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 5.14e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.67999e-06
        [meta_shard_fg_expand]: 1.15999e-06
        [get_grad_eliminate_]: 4.51002e-06
        [merge_forward]: 3.01999e-06
        [cell_reuse_recompute_pass]: 3.38e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.333e-05
        [j_node_and_user_rematch]: 8.15e-06
        [meta_fg_expand]: 2.12999e-06
        [replace_old_param]: 1.452e-05
        [inline_without_move]: 4.59998e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.50001e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 6.19999e-06
        [cse]: 1.127e-05
        [replace_applicator]: 5.25001e-06
[py_interpret_to_execute_after_opt_a]: 1.307e-05
[rewriter_after_opt_a]: 3.795e-05
[convert_after_rewriter]: 7.63001e-06
[order_py_execute_after_rewriter]: 5.87001e-06
[mutable_eliminate]: 0.00067323
[jit_opt_b]: 5.468e-05, [1]
    [Cycle 1]: 4.643e-05, [2]
        [frontend_op_eliminate]: 1.814e-05
        [inline_after_opt_a]: 1.611e-05
[cconv]: 2.687e-05
[loop_unroll]: 0.00046399
[jit_opt_after_cconv]: 0.00015298, [1]
    [Cycle 1]: 0.00014539, [11]
        [c_1]: 2.173e-05
        [parameter_eliminate]: 2.58e-06
        [updatestate_depend_eliminate]: 5.79e-06
        [updatestate_assign_eliminate]: 2.55002e-06
        [updatestate_loads_eliminate]: 2.40002e-06
        [cse]: 2.333e-05
        [call_graph_tuple_transform]: 2.037e-05
        [tuple_list_get_item_eliminator]: 5.91e-06
        [none_parameter_eliminate]: 1.89999e-06
        [renormalize]: 6.89994e-07
        [switch_simplify]: 5.43002e-06
[remove_dup_value]: 1.547e-05
[partial_unused_args_eliminate]: 2.90002e-06
[environ_conv]: 5.79e-06
[add_recomputation]: 5.058e-05
[cse_after_recomputation]: 2.219e-05, [1]
    [Cycle 1]: 1.612e-05, [1]
        [cse]: 1.008e-05
[auto_monad_reorder]: 1.744e-05
[get_jit_bprop_graph]: 2.19999e-06
[rewriter_after_jit_bprop_graph]: 6.83e-06
[opt_after_jit_grad]: 0.00144683
[symbol_engine_optimizer]: 8.739e-05, [1]
    [Cycle 1]: 7.785e-05, [6]
        [build]: 4.23999e-06
        [elim_shapecalc]: 8.82999e-06
        [elim_not_effective]: 1.51e-05
        [opt_reshape]: 7.23999e-06
        [fold_const_symbol]: 1.217e-05
        [renormalize]: 9.20001e-07
[validate]: 4.658e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.0244112
[execute]: 1.576e-05
Sums
    bootstrap                            :     0.000387s :  1.10%
    type_inference                       :     0.005119s : 14.54%
    event_method                         :     0.000020s :  0.06%
    auto_monad                           :     0.000062s :  0.17%
    graph_reusing                        :     0.000005s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.12%
    rewriter_before_opt_a                :     0.000073s :  0.21%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000063s :  0.18%
    jit_opt_a.loop_unroll                :     0.000036s :  0.10%
    jit_opt_a.a_1                        :     0.000693s :  1.97%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000013s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000024s :  0.07%
    jit_opt_a.inline_without_move        :     0.000011s :  0.03%
    jit_opt_a.renormalize                :     0.000962s :  2.73%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.06%
    jit_opt_a.cse                        :     0.000040s :  0.11%
    jit_opt_a.replace_applicator         :     0.000021s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.04%
    rewriter_after_opt_a                 :     0.000038s :  0.11%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000673s :  1.91%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.05%
    cconv                                :     0.000027s :  0.08%
    loop_unroll                          :     0.000464s :  1.32%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000015s :  0.04%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000051s :  0.14%
    cse_after_recomputation.cse          :     0.000010s :  0.03%
    auto_monad_reorder                   :     0.000017s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.001447s :  4.11%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000047s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.024411s : 69.33%
    execute                              :     0.000016s :  0.04%


Time group info:
------[substitution.]   0.000256    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.32% :     0.000006s :      3: substitution.graph_param_transform
83.41% :     0.000214s :      4: substitution.inline
 1.64% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.12% :     0.000005s :      4: substitution.remove_not_recompute_node
 4.18% :     0.000011s :      2: substitution.replace_old_param
 4.92% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005045     2
81.16% :     0.004095s :      1: type_inference.infer
18.84% :     0.000950s :      1: type_inference.specialize
------[replace.]   0.000082     6
68.22% :     0.000056s :      4: replace.inline
31.78% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000223     6
94.86% :     0.000211s :      4: match.inline
 5.14% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.87% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.16% :     0.000001s :     10: predicate.addn_zero_filter
 2.38% :     0.000003s :     10: predicate.arithmetic_simplify
 1.27% :     0.000002s :     10: predicate.cast_eliminate
 0.76% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.08% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.12% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.79% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.39% :     0.000007s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.46% :     0.000002s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.69% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.79% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.08% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 1.67% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.93% :     0.000002s :     10: predicate.reduce_eliminate
 1.34% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.24% :     0.000002s :     10: predicate.reshape_eliminate
 1.53% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.80% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.54% :     0.000002s :      6: predicate.special_op_eliminate
 1.16% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 2.01% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.61% :     0.000008s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.19% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 2.16% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.63% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000002s :     10: predicate.value_based_eliminate
 0.88% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000677    12
45.13% :     0.000306s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.97% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.90% :     0.000324s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042326    76
 0.13% :     0.000054s :      1: add_recomputation
 0.15% :     0.000065s :      1: auto_monad
 0.05% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.96% :     0.000405s :      1: bootstrap
 0.07% :     0.000029s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.06% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.06% :     0.000024s :      1: event_method
 0.05% :     0.000022s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
16.05% :     0.006795s :      1: jit_opt_a
 0.37% :     0.000156s :      1: jit_opt_after_cconv
 0.14% :     0.000057s :      1: jit_opt_b
 1.12% :     0.000472s :      1: loop_unroll
 1.61% :     0.000682s :      1: mutable_eliminate
 2.19% :     0.000927s :     26: opt.transform.jit_opt_a
 0.12% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000028s :      4: opt.transform.jit_opt_b
 0.03% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000039s :      4: opt.transform.symbol_engine_opt
 3.44% :     0.001458s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.10% :     0.000044s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000018s :      1: remove_dup_value
 1.28% :     0.000540s :      1: renormalize.infer
 0.98% :     0.000413s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000041s :      1: rewriter_after_opt_a
 0.18% :     0.000076s :      1: rewriter_before_opt_a
 0.22% :     0.000091s :      1: symbol_engine_optimizer
57.76% :     0.024448s :      1: task_emit
12.13% :     0.005136s :      1: type_inference
 0.17% :     0.000072s :      1: validate


TotalTime = 0.0203126, [33]
[bootstrap]: 0.0005282
[type_inference]: 0.00522386
[event_method]: 1.904e-05
[auto_monad]: 6.701e-05
[graph_reusing]: 5.56e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 3.851e-05
[rewriter_before_opt_a]: 7.857e-05
[expand_dump_flag]: 3.11001e-06
[jit_opt_a]: 0.00549706, [2]
    [Cycle 1]: 0.0017311, [27]
        [switch_simplify]: 7.432e-05
        [loop_unroll]: 3.045e-05
        [a_1]: 0.00056426
        [with_stream_mark]: 2.42e-05
        [recompute_prepare]: 7.19001e-06
        [updatestate_depend_eliminate]: 3.93001e-06
        [updatestate_assign_eliminate]: 2.81999e-06
        [updatestate_loads_eliminate]: 2.80002e-06
        [parameter_eliminate]: 1.76e-06
        [specialize_transform]: 5.77001e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 5.00001e-06
        [merge_forward]: 3.95998e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.375e-05
        [j_node_and_user_rematch]: 8.60001e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 9.76e-06
        [inline_without_move]: 5.27001e-06
        [renormalize]: 0.0006992
        [add_forward_monad_depend]: 5.30999e-06
        [auto_monad_grad]: 2.71999e-06
        [auto_monad_eliminator]: 1.401e-05
        [cse]: 3.09e-05
        [replace_applicator]: 1.291e-05
    [Cycle 2]: 0.00031977, [27]
        [switch_simplify]: 5.57001e-06
        [loop_unroll]: 4.77e-06
        [a_1]: 8.712e-05
        [with_stream_mark]: 9.91998e-06
        [recompute_prepare]: 5.46998e-06
        [updatestate_depend_eliminate]: 3.56999e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.11002e-06
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 4.99e-06
        [meta_shard_fg_expand]: 1.03001e-06
        [get_grad_eliminate_]: 4.44998e-06
        [merge_forward]: 2.79999e-06
        [cell_reuse_recompute_pass]: 2.00002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.38e-05
        [j_node_and_user_rematch]: 8.38999e-06
        [meta_fg_expand]: 1.49998e-06
        [replace_old_param]: 7.85998e-06
        [inline_without_move]: 4.46002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 9.5999e-07
        [auto_monad_eliminator]: 5.47001e-06
        [cse]: 9.77999e-06
        [replace_applicator]: 4.85001e-06
[py_interpret_to_execute_after_opt_a]: 1.13e-05
[rewriter_after_opt_a]: 3.638e-05
[convert_after_rewriter]: 7.08e-06
[order_py_execute_after_rewriter]: 5.12999e-06
[mutable_eliminate]: 0.00060511
[jit_opt_b]: 4.972e-05, [1]
    [Cycle 1]: 4.261e-05, [2]
        [frontend_op_eliminate]: 1.599e-05
        [inline_after_opt_a]: 1.485e-05
[cconv]: 2.566e-05
[loop_unroll]: 0.00041771
[jit_opt_after_cconv]: 0.00013988, [1]
    [Cycle 1]: 0.00013366, [11]
        [c_1]: 2.065e-05
        [parameter_eliminate]: 3.33e-06
        [updatestate_depend_eliminate]: 6.21e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.946e-05
        [call_graph_tuple_transform]: 1.929e-05
        [tuple_list_get_item_eliminator]: 5.09998e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 5.59987e-07
        [switch_simplify]: 5.34998e-06
[remove_dup_value]: 1.287e-05
[partial_unused_args_eliminate]: 2.32001e-06
[environ_conv]: 4.97e-06
[add_recomputation]: 5.131e-05
[cse_after_recomputation]: 1.934e-05, [1]
    [Cycle 1]: 1.431e-05, [1]
        [cse]: 8.02998e-06
[auto_monad_reorder]: 1.572e-05
[get_jit_bprop_graph]: 2.31e-06
[rewriter_after_jit_bprop_graph]: 5.42001e-06
[opt_after_jit_grad]: 0.00045132
[symbol_engine_optimizer]: 0.00019093, [1]
    [Cycle 1]: 0.00018314, [6]
        [build]: 4.37e-06
        [elim_shapecalc]: 1.162e-05
        [elim_not_effective]: 1.352e-05
        [opt_reshape]: 5.87001e-06
        [fold_const_symbol]: 9.54e-06
        [renormalize]: 5.69999e-07
[validate]: 4.378e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.00650567
[execute]: 8.27e-06
Sums
    bootstrap                            :     0.000528s :  3.28%
    type_inference                       :     0.005224s : 32.45%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000067s :  0.42%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.24%
    rewriter_before_opt_a                :     0.000079s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000080s :  0.50%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000651s :  4.05%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000699s :  4.34%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000041s :  0.25%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000605s :  3.76%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000026s :  0.16%
    loop_unroll                          :     0.000418s :  2.59%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000051s :  0.32%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000451s :  2.80%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006506s : 40.41%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000234    23
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000002s :      2: substitution.fold_const_symbol
 2.48% :     0.000006s :      3: substitution.graph_param_transform
84.67% :     0.000198s :      4: substitution.inline
 1.54% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.06% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.35% :     0.000006s :      2: substitution.replace_old_param
 5.20% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005145     2
83.24% :     0.004283s :      1: type_inference.infer
16.76% :     0.000862s :      1: type_inference.specialize
------[replace.]   0.000069     6
66.60% :     0.000046s :      4: replace.inline
33.40% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000207     6
94.66% :     0.000196s :      4: match.inline
 5.34% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.30% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.10% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.51% :     0.000002s :     10: predicate.addn_zero_filter
 1.89% :     0.000002s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 0.97% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.00% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.71% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.74% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.02% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.96% :     0.000001s :     10: predicate.environ_get_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.60% :     0.000007s :     22: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.95% :     0.000001s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.29% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.67% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.64% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.94% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.60% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 1.99% :     0.000003s :     15: predicate.partial_eliminate
 1.03% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.35% :     0.000002s :     10: predicate.reduce_eliminate
 1.29% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.52% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.14% :     0.000001s :     10: predicate.reshape_eliminate
 1.21% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.06% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.77% :     0.000001s :      6: predicate.special_op_eliminate
 0.82% :     0.000001s :      6: predicate.specialize_transform
 1.13% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.08% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.32% :     0.000003s :     16: predicate.switch_defer_inline
 1.96% :     0.000003s :     16: predicate.switch_layer_defer_inline
17.25% :     0.000022s :     53: predicate.switch_simplify
 1.10% :     0.000001s :     10: predicate.tile_eliminate
 1.02% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.04% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.56% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.32% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.37% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.48% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000616    12
51.12% :     0.000315s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.85% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.03% :     0.000253s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021986    76
 0.25% :     0.000054s :      1: add_recomputation
 0.32% :     0.000071s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 2.53% :     0.000557s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.02% :     0.005500s :      1: jit_opt_a
 0.65% :     0.000143s :      1: jit_opt_after_cconv
 0.24% :     0.000052s :      1: jit_opt_b
 1.94% :     0.000426s :      1: loop_unroll
 2.80% :     0.000615s :      1: mutable_eliminate
 4.00% :     0.000880s :     26: opt.transform.jit_opt_a
 0.21% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.09% :     0.000459s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000041s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 1.77% :     0.000389s :      1: renormalize.infer
 1.37% :     0.000301s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000041s :      1: rewriter_after_opt_a
 0.37% :     0.000081s :      1: rewriter_before_opt_a
 0.88% :     0.000194s :      1: symbol_engine_optimizer
29.67% :     0.006524s :      1: task_emit
23.83% :     0.005240s :      1: type_inference
 0.30% :     0.000067s :      1: validate


TotalTime = 0.0326554, [33]
[bootstrap]: 0.00046557
[type_inference]: 0.00598725
[event_method]: 2.393e-05
[auto_monad]: 7.771e-05
[graph_reusing]: 6.14999e-06
[pre_auto_parallel]: 2.61999e-06
[py_interpret_to_execute]: 4.601e-05
[rewriter_before_opt_a]: 8.229e-05
[expand_dump_flag]: 3.58e-06
[jit_opt_a]: 0.0166354, [2]
    [Cycle 1]: 0.0016975, [27]
        [switch_simplify]: 6.079e-05
        [loop_unroll]: 2.937e-05
        [a_1]: 0.0005305
        [with_stream_mark]: 1.679e-05
        [recompute_prepare]: 7.36001e-06
        [updatestate_depend_eliminate]: 3.51001e-06
        [updatestate_assign_eliminate]: 3.41001e-06
        [updatestate_loads_eliminate]: 2.89999e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 5.77001e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 2.67001e-06
        [get_grad_eliminate_]: 5.00001e-06
        [merge_forward]: 3.50003e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.513e-05
        [j_node_and_user_rematch]: 8.88002e-06
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 8.58001e-06
        [inline_without_move]: 5.07e-06
        [renormalize]: 0.00072748
        [add_forward_monad_depend]: 5.49998e-06
        [auto_monad_grad]: 2.43e-06
        [auto_monad_eliminator]: 1.469e-05
        [cse]: 2.989e-05
        [replace_applicator]: 1.367e-05
    [Cycle 2]: 0.00032007, [27]
        [switch_simplify]: 5.56e-06
        [loop_unroll]: 4.77e-06
        [a_1]: 8.836e-05
        [with_stream_mark]: 8.07998e-06
        [recompute_prepare]: 4.92999e-06
        [updatestate_depend_eliminate]: 2.64999e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 5.09e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 1.16002e-06
        [get_grad_eliminate_]: 4.60001e-06
        [merge_forward]: 2.89001e-06
        [cell_reuse_recompute_pass]: 2.02999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.225e-05
        [j_node_and_user_rematch]: 7.46999e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 7.05e-06
        [inline_without_move]: 4.48999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.55999e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 5.44e-06
        [cse]: 1.121e-05
        [replace_applicator]: 5.10999e-06
[py_interpret_to_execute_after_opt_a]: 1.268e-05
[rewriter_after_opt_a]: 3.808e-05
[convert_after_rewriter]: 7.48999e-06
[order_py_execute_after_rewriter]: 5.17999e-06
[mutable_eliminate]: 0.00062369
[jit_opt_b]: 5.056e-05, [1]
    [Cycle 1]: 4.355e-05, [2]
        [frontend_op_eliminate]: 1.645e-05
        [inline_after_opt_a]: 1.542e-05
[cconv]: 2.809e-05
[loop_unroll]: 0.00043377
[jit_opt_after_cconv]: 0.00014345, [1]
    [Cycle 1]: 0.00013668, [11]
        [c_1]: 2.143e-05
        [parameter_eliminate]: 2.83e-06
        [updatestate_depend_eliminate]: 5.79e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 2.19e-05
        [call_graph_tuple_transform]: 1.914e-05
        [tuple_list_get_item_eliminator]: 5.29e-06
        [none_parameter_eliminate]: 1.46998e-06
        [renormalize]: 7.2e-07
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 1.34e-05
[partial_unused_args_eliminate]: 2.48002e-06
[environ_conv]: 6.36e-06
[add_recomputation]: 5.238e-05
[cse_after_recomputation]: 2.051e-05, [1]
    [Cycle 1]: 1.481e-05, [1]
        [cse]: 8.74998e-06
[auto_monad_reorder]: 1.617e-05
[get_jit_bprop_graph]: 1.54998e-06
[rewriter_after_jit_bprop_graph]: 5.77999e-06
[opt_after_jit_grad]: 0.00045927
[symbol_engine_optimizer]: 7.289e-05, [1]
    [Cycle 1]: 6.665e-05, [6]
        [build]: 4.05998e-06
        [elim_shapecalc]: 7.83001e-06
        [elim_not_effective]: 1.234e-05
        [opt_reshape]: 5.94e-06
        [fold_const_symbol]: 8.56002e-06
        [renormalize]: 6.09987e-07
[validate]: 3.707e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.007053
[execute]: 8.68001e-06
Sums
    bootstrap                            :     0.000466s :  2.68%
    type_inference                       :     0.005987s : 34.43%
    event_method                         :     0.000024s :  0.14%
    auto_monad                           :     0.000078s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000046s :  0.26%
    rewriter_before_opt_a                :     0.000082s :  0.47%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000619s :  3.56%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000728s :  4.18%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000041s :  0.24%
    jit_opt_a.replace_applicator         :     0.000019s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000624s :  3.59%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000028s :  0.16%
    loop_unroll                          :     0.000434s :  2.49%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000052s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000459s :  2.64%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007053s : 40.56%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000201    23
 1.10% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.94% :     0.000006s :      3: substitution.graph_param_transform
82.59% :     0.000166s :      4: substitution.inline
 1.66% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.39% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.85% :     0.000004s :      2: substitution.replace_old_param
 6.79% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005897     2
83.25% :     0.004909s :      1: type_inference.infer
16.75% :     0.000988s :      1: type_inference.specialize
------[replace.]   0.000075     6
66.62% :     0.000050s :      4: replace.inline
33.38% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000176     6
92.84% :     0.000164s :      4: match.inline
 7.16% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 1.95% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.06% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.09% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.57% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 6.33% :     0.000007s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.39% :     0.000002s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.82% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.02% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.85% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.29% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.79% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.53% :     0.000001s :      3: predicate.reset_defer_inline
 1.68% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.46% :     0.000002s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.56% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.23% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.82% :     0.000009s :     53: predicate.switch_simplify
 1.43% :     0.000002s :     10: predicate.tile_eliminate
 1.07% :     0.000001s :     10: predicate.transpose_eliminate
 1.26% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.01% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.59% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000662    12
54.26% :     0.000359s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 3.83% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.91% :     0.000278s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.034315    76
 0.16% :     0.000055s :      1: add_recomputation
 0.24% :     0.000081s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.44% :     0.000493s :      1: bootstrap
 0.09% :     0.000031s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.08% :     0.000029s :      1: event_method
 0.04% :     0.000013s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
48.49% :     0.016640s :      1: jit_opt_a
 0.43% :     0.000146s :      1: jit_opt_after_cconv
 0.15% :     0.000053s :      1: jit_opt_b
 1.29% :     0.000442s :      1: loop_unroll
 1.84% :     0.000632s :      1: mutable_eliminate
 2.42% :     0.000832s :     26: opt.transform.jit_opt_a
 0.14% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000025s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.36% :     0.000467s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000050s :      1: py_interpret_to_execute
 0.05% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000016s :      1: remove_dup_value
 1.21% :     0.000417s :      1: renormalize.infer
 0.88% :     0.000302s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000043s :      1: rewriter_after_opt_a
 0.25% :     0.000086s :      1: rewriter_before_opt_a
 0.22% :     0.000075s :      1: symbol_engine_optimizer
20.60% :     0.007068s :      1: task_emit
17.52% :     0.006011s :      1: type_inference
 0.17% :     0.000058s :      1: validate


TotalTime = 0.0235868, [33]
[bootstrap]: 0.00043673
[type_inference]: 0.00542436
[event_method]: 1.659e-05
[auto_monad]: 6.839e-05
[graph_reusing]: 5.64998e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 3.661e-05
[rewriter_before_opt_a]: 8.212e-05
[expand_dump_flag]: 3.33998e-06
[jit_opt_a]: 0.00631455, [2]
    [Cycle 1]: 0.00227109, [27]
        [switch_simplify]: 0.00017688
        [loop_unroll]: 3.367e-05
        [a_1]: 0.00063685
        [with_stream_mark]: 2.453e-05
        [recompute_prepare]: 1.055e-05
        [updatestate_depend_eliminate]: 3.55998e-06
        [updatestate_assign_eliminate]: 2.52001e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 7.93001e-06
        [updatestate_useless_node_eliminater]: 5.97999e-06
        [accelerated_algorithm]: 6.51e-06
        [meta_shard_fg_expand]: 3.33e-06
        [get_grad_eliminate_]: 5.54e-06
        [merge_forward]: 3.83999e-06
        [cell_reuse_recompute_pass]: 2.12001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.829e-05
        [j_node_and_user_rematch]: 1.159e-05
        [meta_fg_expand]: 4.1e-06
        [replace_old_param]: 1.005e-05
        [inline_without_move]: 5.96e-06
        [renormalize]: 0.00102141
        [add_forward_monad_depend]: 9.23002e-06
        [auto_monad_grad]: 2.01998e-06
        [auto_monad_eliminator]: 1.606e-05
        [cse]: 2.654e-05
        [replace_applicator]: 2.046e-05
    [Cycle 2]: 0.00040746, [27]
        [switch_simplify]: 6.15002e-06
        [loop_unroll]: 6.14999e-06
        [a_1]: 0.00010815
        [with_stream_mark]: 1.645e-05
        [recompute_prepare]: 6.86001e-06
        [updatestate_depend_eliminate]: 3.23e-06
        [updatestate_assign_eliminate]: 3.01999e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 5.79e-06
        [updatestate_useless_node_eliminater]: 5.92999e-06
        [accelerated_algorithm]: 7.01001e-06
        [meta_shard_fg_expand]: 2.22999e-06
        [get_grad_eliminate_]: 4.85001e-06
        [merge_forward]: 4.27e-06
        [cell_reuse_recompute_pass]: 4.14997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.816e-05
        [j_node_and_user_rematch]: 8.43001e-06
        [meta_fg_expand]: 2.01003e-06
        [replace_old_param]: 8.29002e-06
        [inline_without_move]: 4.85999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 1.45999e-06
        [auto_monad_eliminator]: 8.22e-06
        [cse]: 1.309e-05
        [replace_applicator]: 6.81001e-06
[py_interpret_to_execute_after_opt_a]: 1.741e-05
[rewriter_after_opt_a]: 3.932e-05
[convert_after_rewriter]: 8.80999e-06
[order_py_execute_after_rewriter]: 5.94999e-06
[mutable_eliminate]: 0.00082414
[jit_opt_b]: 6.422e-05, [1]
    [Cycle 1]: 5.557e-05, [2]
        [frontend_op_eliminate]: 2.196e-05
        [inline_after_opt_a]: 1.755e-05
[cconv]: 3.293e-05
[loop_unroll]: 0.00070153
[jit_opt_after_cconv]: 0.00018148, [1]
    [Cycle 1]: 0.00017286, [11]
        [c_1]: 2.463e-05
        [parameter_eliminate]: 4.68001e-06
        [updatestate_depend_eliminate]: 7.78001e-06
        [updatestate_assign_eliminate]: 3.98999e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [cse]: 2.755e-05
        [call_graph_tuple_transform]: 2.821e-05
        [tuple_list_get_item_eliminator]: 6.28998e-06
        [none_parameter_eliminate]: 1.96e-06
        [renormalize]: 3.19997e-07
        [switch_simplify]: 6.41998e-06
[remove_dup_value]: 1.681e-05
[partial_unused_args_eliminate]: 2.43e-06
[environ_conv]: 7.45998e-06
[add_recomputation]: 5.716e-05
[cse_after_recomputation]: 2.374e-05, [1]
    [Cycle 1]: 1.74e-05, [1]
        [cse]: 1.045e-05
[auto_monad_reorder]: 1.786e-05
[get_jit_bprop_graph]: 2.32999e-06
[rewriter_after_jit_bprop_graph]: 9.01002e-06
[opt_after_jit_grad]: 0.00065282
[symbol_engine_optimizer]: 9.628e-05, [1]
    [Cycle 1]: 8.683e-05, [6]
        [build]: 6.55997e-06
        [elim_shapecalc]: 9.81e-06
        [elim_not_effective]: 1.907e-05
        [opt_reshape]: 6.12999e-06
        [fold_const_symbol]: 1.071e-05
        [renormalize]: 8.00006e-07
[validate]: 4.888e-05
[backend_pass]: 1.15999e-06
[task_emit]: 0.0081057
[execute]: 1.041e-05
Sums
    bootstrap                            :     0.000437s :  2.28%
    type_inference                       :     0.005424s : 28.27%
    event_method                         :     0.000017s :  0.09%
    auto_monad                           :     0.000068s :  0.36%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000037s :  0.19%
    rewriter_before_opt_a                :     0.000082s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000183s :  0.95%
    jit_opt_a.loop_unroll                :     0.000040s :  0.21%
    jit_opt_a.a_1                        :     0.000745s :  3.88%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000014s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000014s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001021s :  5.32%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.13%
    jit_opt_a.cse                        :     0.000040s :  0.21%
    jit_opt_a.replace_applicator         :     0.000027s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.09%
    rewriter_after_opt_a                 :     0.000039s :  0.20%
    convert_after_rewriter               :     0.000009s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000824s :  4.29%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.09%
    cconv                                :     0.000033s :  0.17%
    loop_unroll                          :     0.000702s :  3.66%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000057s :  0.30%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000018s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000653s :  3.40%
    symbol_engine_optimizer.build        :     0.000007s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000049s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008106s : 42.24%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000258    23
 1.28% :     0.000003s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 2.44% :     0.000006s :      3: substitution.graph_param_transform
83.63% :     0.000215s :      4: substitution.inline
 1.66% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.50% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000005s :      2: substitution.replace_old_param
 5.90% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005302     2
82.64% :     0.004382s :      1: type_inference.infer
17.36% :     0.000920s :      1: type_inference.specialize
------[replace.]   0.000090     6
63.33% :     0.000057s :      4: replace.inline
36.67% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000227     6
93.81% :     0.000213s :      4: match.inline
 6.19% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   691
 1.52% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.10% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.89% :     0.000001s :     10: predicate.addn_check_dump
 1.13% :     0.000002s :     10: predicate.addn_zero_filter
 3.13% :     0.000004s :     10: predicate.arithmetic_simplify
 1.09% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000002s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.49% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     10: predicate.environ_get_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 6.60% :     0.000009s :     22: predicate.inline
 1.15% :     0.000002s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.33% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.61% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.77% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.42% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 0.88% :     0.000001s :     10: predicate.minmaximum_grad
 1.30% :     0.000002s :      3: predicate.mutable_eliminate
 0.39% :     0.000001s :      3: predicate.opt_reshape
 1.77% :     0.000002s :     15: predicate.partial_eliminate
 0.94% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.40% :     0.000001s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.03% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.30% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.03% :     0.000001s :     10: predicate.same_eliminate
 0.67% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000002s :      6: predicate.specialize_transform
 1.57% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.57% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.74% :     0.000004s :     16: predicate.switch_defer_inline
 1.82% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.46% :     0.000010s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.35% :     0.000002s :     10: predicate.transpose_eliminate
 1.18% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.33% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.42% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.37% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.94% :     0.000003s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000750    12
48.40% :     0.000363s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.33% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.28% :     0.000340s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025855    76
 0.24% :     0.000062s :      1: add_recomputation
 0.28% :     0.000072s :      1: auto_monad
 0.08% :     0.000021s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.78% :     0.000461s :      1: bootstrap
 0.14% :     0.000036s :      1: cconv
 0.05% :     0.000012s :      1: convert_after_rewriter
 0.10% :     0.000026s :      1: cse_after_recomputation
 0.04% :     0.000011s :      1: environ_conv
 0.08% :     0.000020s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
24.44% :     0.006319s :      1: jit_opt_a
 0.72% :     0.000186s :      1: jit_opt_after_cconv
 0.26% :     0.000068s :      1: jit_opt_b
 2.75% :     0.000712s :      1: loop_unroll
 3.24% :     0.000838s :      1: mutable_eliminate
 4.31% :     0.001114s :     26: opt.transform.jit_opt_a
 0.24% :     0.000061s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000032s :      4: opt.transform.jit_opt_b
 0.07% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000042s :      4: opt.transform.symbol_engine_opt
 2.58% :     0.000666s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000040s :      1: py_interpret_to_execute
 0.08% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000020s :      1: remove_dup_value
 2.08% :     0.000538s :      1: renormalize.infer
 1.82% :     0.000471s :      1: renormalize.specialize
 0.04% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000048s :      1: rewriter_after_opt_a
 0.33% :     0.000085s :      1: rewriter_before_opt_a
 0.39% :     0.000100s :      1: symbol_engine_optimizer
31.44% :     0.008128s :      1: task_emit
21.02% :     0.005434s :      1: type_inference
 0.30% :     0.000076s :      1: validate


TotalTime = 0.0300008, [33]
[bootstrap]: 0.00041965
[type_inference]: 0.00492668
[event_method]: 1.789e-05
[auto_monad]: 6.417e-05
[graph_reusing]: 6.17999e-06
[pre_auto_parallel]: 2.48998e-06
[py_interpret_to_execute]: 3.36e-05
[rewriter_before_opt_a]: 7.337e-05
[expand_dump_flag]: 3.31001e-06
[jit_opt_a]: 0.0154028, [2]
    [Cycle 1]: 0.0116628, [27]
        [switch_simplify]: 5.458e-05
        [loop_unroll]: 2.871e-05
        [a_1]: 0.00052716
        [with_stream_mark]: 2.139e-05
        [recompute_prepare]: 7.13e-06
        [updatestate_depend_eliminate]: 3.45998e-06
        [updatestate_assign_eliminate]: 2.95998e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 1.64998e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 4.83001e-06
        [accelerated_algorithm]: 5.07999e-06
        [meta_shard_fg_expand]: 1.85001e-06
        [get_grad_eliminate_]: 4.94998e-06
        [merge_forward]: 3.61999e-06
        [cell_reuse_recompute_pass]: 9.89996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.352e-05
        [j_node_and_user_rematch]: 8.50001e-06
        [meta_fg_expand]: 2.01e-06
        [replace_old_param]: 0.00968446
        [inline_without_move]: 3.017e-05
        [renormalize]: 0.00092388
        [add_forward_monad_depend]: 8.92e-06
        [auto_monad_grad]: 2.89999e-06
        [auto_monad_eliminator]: 1.965e-05
        [cse]: 3.439e-05
        [replace_applicator]: 1.265e-05
    [Cycle 2]: 0.00035407, [27]
        [switch_simplify]: 6.34001e-06
        [loop_unroll]: 5.74e-06
        [a_1]: 0.00010085
        [with_stream_mark]: 1.634e-05
        [recompute_prepare]: 5.34998e-06
        [updatestate_depend_eliminate]: 3.46001e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.69999e-06
        [parameter_eliminate]: 1.91998e-06
        [specialize_transform]: 5.23002e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 5.62001e-06
        [meta_shard_fg_expand]: 1.87001e-06
        [get_grad_eliminate_]: 4.65001e-06
        [merge_forward]: 3.45998e-06
        [cell_reuse_recompute_pass]: 3.70998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.689e-05
        [j_node_and_user_rematch]: 8.84e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 8.59e-06
        [inline_without_move]: 4.81002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.48002e-06
        [auto_monad_grad]: 9.10019e-07
        [auto_monad_eliminator]: 5.81998e-06
        [cse]: 1.086e-05
        [replace_applicator]: 5.24e-06
[py_interpret_to_execute_after_opt_a]: 1.47e-05
[rewriter_after_opt_a]: 3.793e-05
[convert_after_rewriter]: 6.69001e-06
[order_py_execute_after_rewriter]: 6.70002e-06
[mutable_eliminate]: 0.0008378
[jit_opt_b]: 5.596e-05, [1]
    [Cycle 1]: 4.674e-05, [2]
        [frontend_op_eliminate]: 1.643e-05
        [inline_after_opt_a]: 1.707e-05
[cconv]: 3.239e-05
[loop_unroll]: 0.00048445
[jit_opt_after_cconv]: 0.00015678, [1]
    [Cycle 1]: 0.00014998, [11]
        [c_1]: 2.35e-05
        [parameter_eliminate]: 3.86001e-06
        [updatestate_depend_eliminate]: 7.92e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.21998e-06
        [cse]: 2.447e-05
        [call_graph_tuple_transform]: 2.192e-05
        [tuple_list_get_item_eliminator]: 4.99e-06
        [none_parameter_eliminate]: 1.39998e-06
        [renormalize]: 1.10001e-06
        [switch_simplify]: 5.02e-06
[remove_dup_value]: 1.585e-05
[partial_unused_args_eliminate]: 2.04999e-06
[environ_conv]: 6.06e-06
[add_recomputation]: 5.66e-05
[cse_after_recomputation]: 2.062e-05, [1]
    [Cycle 1]: 1.452e-05, [1]
        [cse]: 8.82999e-06
[auto_monad_reorder]: 1.525e-05
[get_jit_bprop_graph]: 2.21e-06
[rewriter_after_jit_bprop_graph]: 8.62998e-06
[opt_after_jit_grad]: 0.00049345
[symbol_engine_optimizer]: 7.806e-05, [1]
    [Cycle 1]: 7.14e-05, [6]
        [build]: 5.46998e-06
        [elim_shapecalc]: 7.6e-06
        [elim_not_effective]: 1.438e-05
        [opt_reshape]: 5.86003e-06
        [fold_const_symbol]: 8.36002e-06
        [renormalize]: 6.19999e-07
[validate]: 4.252e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00641954
[execute]: 2.236e-05
Sums
    bootstrap                            :     0.000420s :  1.62%
    type_inference                       :     0.004927s : 19.03%
    event_method                         :     0.000018s :  0.07%
    auto_monad                           :     0.000064s :  0.25%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000034s :  0.13%
    rewriter_before_opt_a                :     0.000073s :  0.28%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000061s :  0.24%
    jit_opt_a.loop_unroll                :     0.000034s :  0.13%
    jit_opt_a.a_1                        :     0.000628s :  2.43%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.009693s : 37.43%
    jit_opt_a.inline_without_move        :     0.000035s :  0.14%
    jit_opt_a.renormalize                :     0.000924s :  3.57%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.10%
    jit_opt_a.cse                        :     0.000045s :  0.17%
    jit_opt_a.replace_applicator         :     0.000018s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.06%
    rewriter_after_opt_a                 :     0.000038s :  0.15%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.03%
    mutable_eliminate                    :     0.000838s :  3.24%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.07%
    cconv                                :     0.000032s :  0.13%
    loop_unroll                          :     0.000484s :  1.87%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000057s :  0.22%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000015s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.03%
    opt_after_jit_grad                   :     0.000493s :  1.91%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006420s : 24.79%
    execute                              :     0.000022s :  0.09%


Time group info:
------[substitution.]   0.009851    23
 0.02% :     0.000002s :      2: substitution.elim_not_effective
 0.01% :     0.000001s :      2: substitution.fold_const_symbol
 0.05% :     0.000005s :      3: substitution.graph_param_transform
 1.78% :     0.000175s :      4: substitution.inline
 0.04% :     0.000004s :      4: substitution.j_node_and_user_rematch
 0.05% :     0.000005s :      4: substitution.remove_not_recompute_node
97.92% :     0.009647s :      2: substitution.replace_old_param
 0.11% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004852     2
83.79% :     0.004065s :      1: type_inference.infer
16.21% :     0.000786s :      1: type_inference.specialize
------[replace.]   0.000066     6
68.15% :     0.000045s :      4: replace.inline
31.85% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000183     6
94.48% :     0.000173s :      4: match.inline
 5.52% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.16% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.03% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 2.08% :     0.000003s :     10: predicate.arithmetic_simplify
 1.57% :     0.000002s :     10: predicate.cast_eliminate
 0.69% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.91% :     0.000001s :     10: predicate.compare_switch_simplify
 1.14% :     0.000001s :     10: predicate.depend_value_elim
 0.98% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.14% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.66% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.17% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.71% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.03% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.20% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.90% :     0.000008s :     22: predicate.inline
 5.41% :     0.000007s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :      6: predicate.less_batch_normalization
 1.70% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.60% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.13% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 0.93% :     0.000001s :     10: predicate.minmaximum_grad
 1.69% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000003s :     15: predicate.partial_eliminate
 1.08% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.45% :     0.000002s :     18: predicate.replace_applicator
 0.83% :     0.000001s :      6: predicate.replace_old_param
 0.43% :     0.000001s :      3: predicate.reset_defer_inline
 1.13% :     0.000001s :     10: predicate.reshape_eliminate
 1.19% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.58% :     0.000002s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.84% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.14% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.07% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.08% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.36% :     0.000009s :     53: predicate.switch_simplify
 1.00% :     0.000001s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.28% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.24% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.51% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000616    12
48.05% :     0.000296s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.85% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.10% :     0.000278s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.041562    76
 0.14% :     0.000059s :      1: add_recomputation
 0.16% :     0.000067s :      1: auto_monad
 0.04% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.07% :     0.000446s :      1: bootstrap
 0.08% :     0.000035s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.06% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.05% :     0.000022s :      1: event_method
 0.07% :     0.000027s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
37.07% :     0.015407s :      1: jit_opt_a
 0.38% :     0.000160s :      1: jit_opt_after_cconv
 0.14% :     0.000059s :      1: jit_opt_b
 1.19% :     0.000494s :      1: loop_unroll
 2.05% :     0.000851s :      1: mutable_eliminate
25.34% :     0.010531s :     26: opt.transform.jit_opt_a
 0.12% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000027s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.21% :     0.000503s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.09% :     0.000036s :      1: py_interpret_to_execute
 0.04% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000019s :      1: remove_dup_value
 1.23% :     0.000509s :      1: renormalize.infer
 0.97% :     0.000402s :      1: renormalize.specialize
 0.03% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000041s :      1: rewriter_after_opt_a
 0.18% :     0.000076s :      1: rewriter_before_opt_a
 0.19% :     0.000081s :      1: symbol_engine_optimizer
15.49% :     0.006438s :      1: task_emit
11.89% :     0.004942s :      1: type_inference
 0.17% :     0.000069s :      1: validate


TotalTime = 0.0278242, [33]
[bootstrap]: 0.000297
[type_inference]: 0.00472834
[event_method]: 1.775e-05
[auto_monad]: 6.163e-05
[graph_reusing]: 5.36002e-06
[pre_auto_parallel]: 2.38998e-06
[py_interpret_to_execute]: 3.387e-05
[rewriter_before_opt_a]: 7.181e-05
[expand_dump_flag]: 2.94001e-06
[jit_opt_a]: 0.00570554, [2]
    [Cycle 1]: 0.00195266, [27]
        [switch_simplify]: 5.325e-05
        [loop_unroll]: 2.761e-05
        [a_1]: 0.00052095
        [with_stream_mark]: 2.111e-05
        [recompute_prepare]: 7.35e-06
        [updatestate_depend_eliminate]: 3.33e-06
        [updatestate_assign_eliminate]: 2.62001e-06
        [updatestate_loads_eliminate]: 2.92002e-06
        [parameter_eliminate]: 1.74e-06
        [specialize_transform]: 6.02999e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 5.48002e-06
        [meta_shard_fg_expand]: 2.22001e-06
        [get_grad_eliminate_]: 5.18002e-06
        [merge_forward]: 2.354e-05
        [cell_reuse_recompute_pass]: 1.54e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.537e-05
        [j_node_and_user_rematch]: 9.55001e-06
        [meta_fg_expand]: 2.24999e-06
        [replace_old_param]: 9.42001e-06
        [inline_without_move]: 5.45001e-06
        [renormalize]: 0.00097131
        [add_forward_monad_depend]: 5.46e-06
        [auto_monad_grad]: 2.48e-06
        [auto_monad_eliminator]: 1.406e-05
        [cse]: 3.174e-05
        [replace_applicator]: 1.518e-05
    [Cycle 2]: 0.00034493, [27]
        [switch_simplify]: 6.17001e-06
        [loop_unroll]: 5.69999e-06
        [a_1]: 9.867e-05
        [with_stream_mark]: 1.153e-05
        [recompute_prepare]: 5.20001e-06
        [updatestate_depend_eliminate]: 2.55002e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 5.35999e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 5.57001e-06
        [meta_shard_fg_expand]: 1.76e-06
        [get_grad_eliminate_]: 4.84998e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 2.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.362e-05
        [j_node_and_user_rematch]: 8.32e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 7.20998e-06
        [inline_without_move]: 4.91002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.42e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 5.20001e-06
        [cse]: 1.119e-05
        [replace_applicator]: 6.07999e-06
[py_interpret_to_execute_after_opt_a]: 1.091e-05
[rewriter_after_opt_a]: 3.347e-05
[convert_after_rewriter]: 7.55998e-06
[order_py_execute_after_rewriter]: 5.47001e-06
[mutable_eliminate]: 0.00056318
[jit_opt_b]: 4.877e-05, [1]
    [Cycle 1]: 4.145e-05, [2]
        [frontend_op_eliminate]: 1.564e-05
        [inline_after_opt_a]: 1.44e-05
[cconv]: 2.507e-05
[loop_unroll]: 0.00138821
[jit_opt_after_cconv]: 0.00014, [1]
    [Cycle 1]: 0.00013356, [11]
        [c_1]: 2.047e-05
        [parameter_eliminate]: 3.20002e-06
        [updatestate_depend_eliminate]: 6.69001e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 1.986e-05
        [call_graph_tuple_transform]: 1.854e-05
        [tuple_list_get_item_eliminator]: 5.32999e-06
        [none_parameter_eliminate]: 1.53002e-06
        [renormalize]: 6.79982e-07
        [switch_simplify]: 5.05999e-06
[remove_dup_value]: 1.311e-05
[partial_unused_args_eliminate]: 2.35002e-06
[environ_conv]: 5.33002e-06
[add_recomputation]: 4.88e-05
[cse_after_recomputation]: 1.812e-05, [1]
    [Cycle 1]: 1.295e-05, [1]
        [cse]: 7.73999e-06
[auto_monad_reorder]: 1.442e-05
[get_jit_bprop_graph]: 2.24999e-06
[rewriter_after_jit_bprop_graph]: 6.43e-06
[opt_after_jit_grad]: 0.00055513
[symbol_engine_optimizer]: 7.102e-05, [1]
    [Cycle 1]: 6.463e-05, [6]
        [build]: 3.86999e-06
        [elim_shapecalc]: 7.81001e-06
        [elim_not_effective]: 1.204e-05
        [opt_reshape]: 5.79999e-06
        [fold_const_symbol]: 7.85998e-06
        [renormalize]: 4.69998e-07
[validate]: 3.682e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.0136681
[execute]: 9.72001e-06
Sums
    bootstrap                            :     0.000297s :  1.25%
    type_inference                       :     0.004728s : 19.89%
    event_method                         :     0.000018s :  0.07%
    auto_monad                           :     0.000062s :  0.26%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000034s :  0.14%
    rewriter_before_opt_a                :     0.000072s :  0.30%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000059s :  0.25%
    jit_opt_a.loop_unroll                :     0.000033s :  0.14%
    jit_opt_a.a_1                        :     0.000620s :  2.61%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000026s :  0.11%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.07%
    jit_opt_a.inline_without_move        :     0.000010s :  0.04%
    jit_opt_a.renormalize                :     0.000971s :  4.09%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.08%
    jit_opt_a.cse                        :     0.000043s :  0.18%
    jit_opt_a.replace_applicator         :     0.000021s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.05%
    rewriter_after_opt_a                 :     0.000033s :  0.14%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000563s :  2.37%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.06%
    cconv                                :     0.000025s :  0.11%
    loop_unroll                          :     0.001388s :  5.84%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000013s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000049s :  0.21%
    cse_after_recomputation.cse          :     0.000008s :  0.03%
    auto_monad_reorder                   :     0.000014s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000555s :  2.33%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.013668s : 57.49%
    execute                              :     0.000010s :  0.04%


Time group info:
------[substitution.]   0.000206    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.48% :     0.000005s :      3: substitution.graph_param_transform
83.77% :     0.000173s :      4: substitution.inline
 1.80% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.21% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.11% :     0.000004s :      2: substitution.replace_old_param
 6.19% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004660     2
83.36% :     0.003884s :      1: type_inference.infer
16.64% :     0.000776s :      1: type_inference.specialize
------[replace.]   0.000065     6
68.03% :     0.000044s :      4: replace.inline
31.97% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000182     6
93.62% :     0.000170s :      4: match.inline
 6.38% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.28% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 2.28% :     0.000003s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.22% :     0.000001s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.12% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.29% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000000s :      3: predicate.graph_param_transform
 5.97% :     0.000007s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.71% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.84% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.73% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.35% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.83% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.25% :     0.000001s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.30% :     0.000010s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.51% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.85% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000575    12
50.72% :     0.000292s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.82% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.46% :     0.000244s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.029723    76
 0.17% :     0.000052s :      1: add_recomputation
 0.22% :     0.000065s :      1: auto_monad
 0.06% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.08% :     0.000320s :      1: bootstrap
 0.09% :     0.000028s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000020s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.07% :     0.000022s :      1: event_method
 0.05% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
19.21% :     0.005709s :      1: jit_opt_a
 0.48% :     0.000143s :      1: jit_opt_after_cconv
 0.17% :     0.000051s :      1: jit_opt_b
 4.70% :     0.001398s :      1: loop_unroll
 1.93% :     0.000572s :      1: mutable_eliminate
 2.81% :     0.000834s :     26: opt.transform.jit_opt_a
 0.16% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000030s :      4: opt.transform.symbol_engine_opt
 1.90% :     0.000564s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000036s :      1: py_interpret_to_execute
 0.04% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000016s :      1: remove_dup_value
 2.24% :     0.000664s :      1: renormalize.infer
 1.00% :     0.000298s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000036s :      1: rewriter_after_opt_a
 0.25% :     0.000075s :      1: rewriter_before_opt_a
 0.25% :     0.000074s :      1: symbol_engine_optimizer
46.05% :     0.013687s :      1: task_emit
15.96% :     0.004744s :      1: type_inference
 0.20% :     0.000059s :      1: validate


TotalTime = 0.0223543, [33]
[bootstrap]: 0.00047708
[type_inference]: 0.0062858
[event_method]: 1.916e-05
[auto_monad]: 6.68e-05
[graph_reusing]: 6.01998e-06
[pre_auto_parallel]: 3.23e-06
[py_interpret_to_execute]: 3.953e-05
[rewriter_before_opt_a]: 7.609e-05
[expand_dump_flag]: 3.61001e-06
[jit_opt_a]: 0.00659066, [2]
    [Cycle 1]: 0.00190365, [27]
        [switch_simplify]: 5.993e-05
        [loop_unroll]: 2.918e-05
        [a_1]: 0.00063784
        [with_stream_mark]: 2.494e-05
        [recompute_prepare]: 7.95e-06
        [updatestate_depend_eliminate]: 4.15999e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.59001e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 5.82001e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.51002e-06
        [meta_shard_fg_expand]: 2.88998e-06
        [get_grad_eliminate_]: 5.07999e-06
        [merge_forward]: 4.53001e-06
        [cell_reuse_recompute_pass]: 1.39998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.509e-05
        [j_node_and_user_rematch]: 9.50001e-06
        [meta_fg_expand]: 2.22999e-06
        [replace_old_param]: 1.002e-05
        [inline_without_move]: 5.20999e-06
        [renormalize]: 0.00076949
        [add_forward_monad_depend]: 7.63001e-06
        [auto_monad_grad]: 2.93e-06
        [auto_monad_eliminator]: 1.541e-05
        [cse]: 3.013e-05
        [replace_applicator]: 1.333e-05
    [Cycle 2]: 0.0003249, [27]
        [switch_simplify]: 5.51e-06
        [loop_unroll]: 4.77e-06
        [a_1]: 8.889e-05
        [with_stream_mark]: 1.223e-05
        [recompute_prepare]: 4.99e-06
        [updatestate_depend_eliminate]: 2.78e-06
        [updatestate_assign_eliminate]: 1.92999e-06
        [updatestate_loads_eliminate]: 1.76e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.95999e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 5.07e-06
        [meta_shard_fg_expand]: 1.40999e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 3.41001e-06
        [cell_reuse_recompute_pass]: 2.41e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.347e-05
        [j_node_and_user_rematch]: 8.27e-06
        [meta_fg_expand]: 1.30999e-06
        [replace_old_param]: 7.80998e-06
        [inline_without_move]: 4.72998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.66e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 5.53002e-06
        [cse]: 1.081e-05
        [replace_applicator]: 5.22999e-06
[py_interpret_to_execute_after_opt_a]: 1.386e-05
[rewriter_after_opt_a]: 3.783e-05
[convert_after_rewriter]: 6.93998e-06
[order_py_execute_after_rewriter]: 5.04e-06
[mutable_eliminate]: 0.00073918
[jit_opt_b]: 5.184e-05, [1]
    [Cycle 1]: 4.382e-05, [2]
        [frontend_op_eliminate]: 1.674e-05
        [inline_after_opt_a]: 1.52e-05
[cconv]: 2.977e-05
[loop_unroll]: 0.00043822
[jit_opt_after_cconv]: 0.00013999, [1]
    [Cycle 1]: 0.0001339, [11]
        [c_1]: 1.996e-05
        [parameter_eliminate]: 3.03e-06
        [updatestate_depend_eliminate]: 6.70998e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 1.943e-05
        [call_graph_tuple_transform]: 1.928e-05
        [tuple_list_get_item_eliminator]: 5.17999e-06
        [none_parameter_eliminate]: 1.46998e-06
        [renormalize]: 6.29982e-07
        [switch_simplify]: 4.89e-06
[remove_dup_value]: 1.304e-05
[partial_unused_args_eliminate]: 2.24999e-06
[environ_conv]: 5.60001e-06
[add_recomputation]: 5.361e-05
[cse_after_recomputation]: 2.056e-05, [1]
    [Cycle 1]: 1.47e-05, [1]
        [cse]: 9.05001e-06
[auto_monad_reorder]: 1.537e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 5.29e-06
[opt_after_jit_grad]: 0.0004623
[symbol_engine_optimizer]: 7.247e-05, [1]
    [Cycle 1]: 6.618e-05, [6]
        [build]: 3.69002e-06
        [elim_shapecalc]: 7.77e-06
        [elim_not_effective]: 1.275e-05
        [opt_reshape]: 5.36998e-06
        [fold_const_symbol]: 8.40999e-06
        [renormalize]: 6.29982e-07
[validate]: 3.866e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.00638628
[execute]: 8.67e-06
Sums
    bootstrap                            :     0.000477s :  2.76%
    type_inference                       :     0.006286s : 36.33%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000067s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.23%
    rewriter_before_opt_a                :     0.000076s :  0.44%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000727s :  4.20%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000770s :  4.45%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.12%
    jit_opt_a.cse                        :     0.000041s :  0.24%
    jit_opt_a.replace_applicator         :     0.000019s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000739s :  4.27%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000030s :  0.17%
    loop_unroll                          :     0.000438s :  2.53%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000054s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000462s :  2.67%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006386s : 36.91%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000234    23
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.69% :     0.000006s :      3: substitution.graph_param_transform
83.54% :     0.000195s :      4: substitution.inline
 1.68% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.96% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.41% :     0.000006s :      2: substitution.replace_old_param
 6.23% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006203     2
85.52% :     0.005304s :      1: type_inference.infer
14.48% :     0.000898s :      1: type_inference.specialize
------[replace.]   0.000077     6
62.02% :     0.000048s :      4: replace.inline
37.98% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000206     6
93.48% :     0.000193s :      4: match.inline
 6.52% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.24% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.21% :     0.000001s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.27% :     0.000002s :     10: predicate.cast_eliminate
 0.35% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.07% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.81% :     0.000007s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 1.56% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.68% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.04% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.56% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.71% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.41% :     0.000000s :      3: predicate.reset_defer_inline
 1.66% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.68% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.53% :     0.000003s :     16: predicate.switch_defer_inline
 1.91% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.29% :     0.000009s :     53: predicate.switch_simplify
 1.55% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.91% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.95% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.21% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000670    12
52.29% :     0.000350s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.35% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.36% :     0.000277s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024164    76
 0.23% :     0.000056s :      1: add_recomputation
 0.29% :     0.000071s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 2.10% :     0.000508s :      1: bootstrap
 0.13% :     0.000032s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
27.29% :     0.006595s :      1: jit_opt_a
 0.59% :     0.000143s :      1: jit_opt_after_cconv
 0.23% :     0.000055s :      1: jit_opt_b
 1.85% :     0.000446s :      1: loop_unroll
 3.10% :     0.000749s :      1: mutable_eliminate
 3.90% :     0.000943s :     26: opt.transform.jit_opt_a
 0.19% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.95% :     0.000471s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000042s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000015s :      1: remove_dup_value
 1.85% :     0.000447s :      1: renormalize.infer
 1.29% :     0.000313s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000041s :      1: rewriter_after_opt_a
 0.33% :     0.000079s :      1: rewriter_before_opt_a
 0.31% :     0.000075s :      1: symbol_engine_optimizer
26.50% :     0.006404s :      1: task_emit
26.10% :     0.006306s :      1: type_inference
 0.25% :     0.000060s :      1: validate


TotalTime = 0.0245171, [30]
[bootstrap]: 0.00046507
[type_inference]: 0.0129449
[event_method]: 1.911e-05
[auto_monad]: 7.086e-05
[graph_reusing]: 6.77002e-06
[pre_auto_parallel]: 2.52001e-06
[py_interpret_to_execute]: 4.251e-05
[rewriter_before_opt_a]: 7.709e-05
[expand_dump_flag]: 3.14999e-06
[jit_opt_a]: 0.00815245, [2]
    [Cycle 1]: 0.00218879, [27]
        [switch_simplify]: 7.012e-05
        [loop_unroll]: 3.042e-05
        [a_1]: 0.00065665
        [with_stream_mark]: 2.484e-05
        [recompute_prepare]: 8.20999e-06
        [updatestate_depend_eliminate]: 4.43001e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [parameter_eliminate]: 2.57001e-06
        [specialize_transform]: 6.48e-06
        [updatestate_useless_node_eliminater]: 5.40001e-06
        [accelerated_algorithm]: 5.92001e-06
        [meta_shard_fg_expand]: 2.74001e-06
        [get_grad_eliminate_]: 6.65998e-06
        [merge_forward]: 3.78999e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.529e-05
        [j_node_and_user_rematch]: 1.015e-05
        [meta_fg_expand]: 2.30002e-06
        [replace_old_param]: 1.093e-05
        [inline_without_move]: 5.93002e-06
        [renormalize]: 0.00102491
        [add_forward_monad_depend]: 1.321e-05
        [auto_monad_grad]: 3.47997e-06
        [auto_monad_eliminator]: 1.874e-05
        [cse]: 3.04e-05
        [replace_applicator]: 1.664e-05
    [Cycle 2]: 0.00033955, [27]
        [switch_simplify]: 6.01e-06
        [loop_unroll]: 5.74999e-06
        [a_1]: 9.98e-05
        [with_stream_mark]: 1.308e-05
        [recompute_prepare]: 4.83001e-06
        [updatestate_depend_eliminate]: 3.23e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 1.71e-06
        [specialize_transform]: 4.75999e-06
        [updatestate_useless_node_eliminater]: 5.45001e-06
        [accelerated_algorithm]: 5.34998e-06
        [meta_shard_fg_expand]: 1.48002e-06
        [get_grad_eliminate_]: 4.58001e-06
        [merge_forward]: 3.28998e-06
        [cell_reuse_recompute_pass]: 2.24999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.47e-05
        [j_node_and_user_rematch]: 8.26002e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 8.03999e-06
        [inline_without_move]: 5.05999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.42e-06
        [auto_monad_grad]: 7.80012e-07
        [auto_monad_eliminator]: 5.74e-06
        [cse]: 1.135e-05
        [replace_applicator]: 5.00001e-06
[py_interpret_to_execute_after_opt_a]: 1.366e-05
[rewriter_after_opt_a]: 3.896e-05
[convert_after_rewriter]: 8.15e-06
[order_py_execute_after_rewriter]: 5.39e-06
[mutable_eliminate]: 0.00083738
[jit_opt_b]: 5.8e-05, [1]
    [Cycle 1]: 4.771e-05, [2]
        [frontend_op_eliminate]: 1.958e-05
        [inline_after_opt_a]: 1.613e-05
[cconv]: 3.532e-05
[loop_unroll]: 0.00050485
[jit_opt_after_cconv]: 0.00016542, [1]
    [Cycle 1]: 0.00015738, [11]
        [c_1]: 2.271e-05
        [parameter_eliminate]: 3.83999e-06
        [updatestate_depend_eliminate]: 9.67001e-06
        [updatestate_assign_eliminate]: 2.50997e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 2.894e-05
        [call_graph_tuple_transform]: 2.51e-05
        [tuple_list_get_item_eliminator]: 5.12999e-06
        [none_parameter_eliminate]: 1.56998e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 5.42001e-06
[remove_dup_value]: 1.614e-05
[partial_unused_args_eliminate]: 2.56998e-06
[environ_conv]: 1.308e-05
[add_recomputation]: 6.536e-05
[cse_after_recomputation]: 2.488e-05, [1]
    [Cycle 1]: 1.796e-05, [1]
        [cse]: 1.12e-05
[auto_monad_reorder]: 1.671e-05
[get_jit_bprop_graph]: 2.31e-06
[rewriter_after_jit_bprop_graph]: 1.066e-05
[opt_after_jit_grad]: 0.0005435
[symbol_engine_optimizer]: 8.893e-05, [1]
    [Cycle 1]: 8.167e-05, [6]
        [build]: 5.76e-06
        [elim_shapecalc]: 9.40001e-06
        [elim_not_effective]: 1.865e-05
        [opt_reshape]: 6.71e-06
        [fold_const_symbol]: 9.47999e-06
        [renormalize]: 6.49976e-07
[validate]: 4.651e-05
Sums
    bootstrap                            :     0.000465s :  2.55%
    type_inference                       :     0.012945s : 71.07%
    event_method                         :     0.000019s :  0.10%
    auto_monad                           :     0.000071s :  0.39%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.23%
    rewriter_before_opt_a                :     0.000077s :  0.42%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.42%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000756s :  4.15%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001025s :  5.63%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.13%
    jit_opt_a.cse                        :     0.000042s :  0.23%
    jit_opt_a.replace_applicator         :     0.000022s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000039s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000837s :  4.60%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000035s :  0.19%
    loop_unroll                          :     0.000505s :  2.77%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000013s :  0.07%
    add_recomputation                    :     0.000065s :  0.36%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.06%
    opt_after_jit_grad                   :     0.000544s :  2.98%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000047s :  0.26%


Time group info:
------[substitution.]   0.000291    23
 0.93% :     0.000003s :      2: substitution.elim_not_effective
 0.64% :     0.000002s :      2: substitution.fold_const_symbol
 2.12% :     0.000006s :      3: substitution.graph_param_transform
86.40% :     0.000251s :      4: substitution.inline
 1.57% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.66% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.00% :     0.000006s :      2: substitution.replace_old_param
 4.68% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012855     2
91.41% :     0.011750s :      1: type_inference.infer
 8.59% :     0.001104s :      1: type_inference.specialize
------[replace.]   0.000088     6
67.14% :     0.000059s :      4: replace.inline
32.86% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000261     6
95.22% :     0.000249s :      4: match.inline
 4.78% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.57% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.48% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 2.77% :     0.000004s :     10: predicate.arithmetic_simplify
 1.19% :     0.000002s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.15% :     0.000002s :     10: predicate.depend_value_elim
 1.22% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.07% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.74% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     10: predicate.environ_get_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000001s :      3: predicate.graph_param_transform
 5.72% :     0.000008s :     22: predicate.inline
 1.03% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :      6: predicate.less_batch_normalization
 1.88% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.72% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.00% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 2.15% :     0.000003s :      3: predicate.mutable_eliminate
 0.64% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000003s :     15: predicate.partial_eliminate
 1.41% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.79% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.50% :     0.000002s :     18: predicate.replace_applicator
 0.43% :     0.000001s :      6: predicate.replace_old_param
 0.42% :     0.000001s :      3: predicate.reset_defer_inline
 1.62% :     0.000002s :     10: predicate.reshape_eliminate
 1.07% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.14% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.62% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.14% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000002s :     10: predicate.tile_eliminate
 1.31% :     0.000002s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.62% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.47% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.92% :     0.000003s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.88% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000888    12
50.90% :     0.000452s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.37% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.73% :     0.000379s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026570    72
 0.26% :     0.000070s :      1: add_recomputation
 0.28% :     0.000074s :      1: auto_monad
 0.07% :     0.000020s :      1: auto_monad_reorder
 1.83% :     0.000485s :      1: bootstrap
 0.14% :     0.000038s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000027s :      1: cse_after_recomputation
 0.06% :     0.000016s :      1: environ_conv
 0.09% :     0.000024s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
30.70% :     0.008157s :      1: jit_opt_a
 0.63% :     0.000169s :      1: jit_opt_after_cconv
 0.23% :     0.000061s :      1: jit_opt_b
 1.94% :     0.000516s :      1: loop_unroll
 3.19% :     0.000849s :      1: mutable_eliminate
 3.74% :     0.000995s :     26: opt.transform.jit_opt_a
 0.21% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000029s :      4: opt.transform.jit_opt_b
 0.06% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000040s :      4: opt.transform.symbol_engine_opt
 2.09% :     0.000555s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000046s :      1: py_interpret_to_execute
 0.06% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.14% :     0.000570s :      1: renormalize.infer
 1.67% :     0.000445s :      1: renormalize.specialize
 0.05% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000043s :      1: rewriter_after_opt_a
 0.30% :     0.000080s :      1: rewriter_before_opt_a
 0.35% :     0.000092s :      1: symbol_engine_optimizer
48.80% :     0.012967s :      1: type_inference


TotalTime = 0.0406129, [33]
[bootstrap]: 0.00033598
[type_inference]: 0.0078294
[event_method]: 1.966e-05
[auto_monad]: 6.394e-05
[graph_reusing]: 5.59e-06
[pre_auto_parallel]: 2.56e-06
[py_interpret_to_execute]: 3.858e-05
[rewriter_before_opt_a]: 7.127e-05
[expand_dump_flag]: 2.86e-06
[jit_opt_a]: 0.0137448, [2]
    [Cycle 1]: 0.00694754, [27]
        [switch_simplify]: 5.883e-05
        [loop_unroll]: 2.864e-05
        [a_1]: 0.00064816
        [with_stream_mark]: 2.607e-05
        [recompute_prepare]: 8.13999e-06
        [updatestate_depend_eliminate]: 4.35e-06
        [updatestate_assign_eliminate]: 3.13998e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 2.29999e-06
        [specialize_transform]: 6.938e-05
        [updatestate_useless_node_eliminater]: 8.72e-06
        [accelerated_algorithm]: 7.71001e-06
        [meta_shard_fg_expand]: 3.26001e-06
        [get_grad_eliminate_]: 6.17001e-06
        [merge_forward]: 1.395e-05
        [cell_reuse_recompute_pass]: 3.67998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.583e-05
        [j_node_and_user_rematch]: 1.078e-05
        [meta_fg_expand]: 2.67001e-06
        [replace_old_param]: 1.149e-05
        [inline_without_move]: 5.04998e-06
        [renormalize]: 0.00163793
        [add_forward_monad_depend]: 1.005e-05
        [auto_monad_grad]: 2.78e-06
        [auto_monad_eliminator]: 1.753e-05
        [cse]: 3.477e-05
        [replace_applicator]: 1.497e-05
    [Cycle 2]: 0.00104485, [27]
        [switch_simplify]: 5.59e-06
        [loop_unroll]: 4.68001e-06
        [a_1]: 0.00077615
        [with_stream_mark]: 2.059e-05
        [recompute_prepare]: 6.96001e-06
        [updatestate_depend_eliminate]: 3.38999e-06
        [updatestate_assign_eliminate]: 2.63e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [parameter_eliminate]: 1.85001e-06
        [specialize_transform]: 5.19998e-06
        [updatestate_useless_node_eliminater]: 4.80999e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 1.34998e-06
        [get_grad_eliminate_]: 4.87e-06
        [merge_forward]: 2.78e-06
        [cell_reuse_recompute_pass]: 2.11e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.559e-05
        [j_node_and_user_rematch]: 8.61002e-06
        [meta_fg_expand]: 1.57999e-06
        [replace_old_param]: 8.48001e-06
        [inline_without_move]: 4.65999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.37999e-06
        [auto_monad_grad]: 1.32e-06
        [auto_monad_eliminator]: 6.22001e-06
        [cse]: 1.611e-05
        [replace_applicator]: 5.49e-06
[py_interpret_to_execute_after_opt_a]: 1.517e-05
[rewriter_after_opt_a]: 3.918e-05
[convert_after_rewriter]: 6.39999e-06
[order_py_execute_after_rewriter]: 4.70001e-06
[mutable_eliminate]: 0.00078545
[jit_opt_b]: 5.221e-05, [1]
    [Cycle 1]: 4.231e-05, [2]
        [frontend_op_eliminate]: 1.603e-05
        [inline_after_opt_a]: 1.465e-05
[cconv]: 3.338e-05
[loop_unroll]: 0.00259067
[jit_opt_after_cconv]: 0.00016785, [1]
    [Cycle 1]: 0.00015957, [11]
        [c_1]: 2.117e-05
        [parameter_eliminate]: 6.11e-06
        [updatestate_depend_eliminate]: 1.002e-05
        [updatestate_assign_eliminate]: 3.14999e-06
        [updatestate_loads_eliminate]: 2.43998e-06
        [cse]: 3.212e-05
        [call_graph_tuple_transform]: 2.258e-05
        [tuple_list_get_item_eliminator]: 5.66e-06
        [none_parameter_eliminate]: 1.30999e-06
        [renormalize]: 7.99977e-07
        [switch_simplify]: 5.40999e-06
[remove_dup_value]: 1.518e-05
[partial_unused_args_eliminate]: 2.44999e-06
[environ_conv]: 5.81e-06
[add_recomputation]: 5.377e-05
[cse_after_recomputation]: 2.069e-05, [1]
    [Cycle 1]: 1.463e-05, [1]
        [cse]: 9.19e-06
[auto_monad_reorder]: 1.553e-05
[get_jit_bprop_graph]: 1.99999e-06
[rewriter_after_jit_bprop_graph]: 9.97999e-06
[opt_after_jit_grad]: 0.00059186
[symbol_engine_optimizer]: 7.713e-05, [1]
    [Cycle 1]: 6.833e-05, [6]
        [build]: 4.53999e-06
        [elim_shapecalc]: 7.40003e-06
        [elim_not_effective]: 1.336e-05
        [opt_reshape]: 6.48e-06
        [fold_const_symbol]: 9.13002e-06
        [renormalize]: 5.50004e-07
[validate]: 4.174e-05
[backend_pass]: 9.10019e-07
[task_emit]: 0.0137224
[execute]: 9.02999e-06
Sums
    bootstrap                            :     0.000336s :  1.12%
    type_inference                       :     0.007829s : 26.01%
    event_method                         :     0.000020s :  0.07%
    auto_monad                           :     0.000064s :  0.21%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.13%
    rewriter_before_opt_a                :     0.000071s :  0.24%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000064s :  0.21%
    jit_opt_a.loop_unroll                :     0.000033s :  0.11%
    jit_opt_a.a_1                        :     0.001424s :  4.73%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000075s :  0.25%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000014s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.04%
    jit_opt_a.merge_forward              :     0.000017s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000041s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000020s :  0.07%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.001638s :  5.44%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.08%
    jit_opt_a.cse                        :     0.000051s :  0.17%
    jit_opt_a.replace_applicator         :     0.000020s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.05%
    rewriter_after_opt_a                 :     0.000039s :  0.13%
    convert_after_rewriter               :     0.000006s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000785s :  2.61%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.05%
    cconv                                :     0.000033s :  0.11%
    loop_unroll                          :     0.002591s :  8.61%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000032s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000015s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000054s :  0.18%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.03%
    opt_after_jit_grad                   :     0.000592s :  1.97%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.013722s : 45.59%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000316    23
 0.68% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000002s :      2: substitution.fold_const_symbol
 1.76% :     0.000006s :      3: substitution.graph_param_transform
86.81% :     0.000275s :      4: substitution.inline
 1.74% :     0.000006s :      4: substitution.j_node_and_user_rematch
 1.78% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.62% :     0.000008s :      2: substitution.replace_old_param
 4.12% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007747     2
83.67% :     0.006482s :      1: type_inference.infer
16.33% :     0.001265s :      1: type_inference.specialize
------[replace.]   0.000075     6
67.76% :     0.000051s :      4: replace.inline
32.24% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000284     6
95.79% :     0.000272s :      4: match.inline
 4.21% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.11% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.48% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.16% :     0.000002s :     10: predicate.addn_zero_filter
 1.87% :     0.000002s :     10: predicate.arithmetic_simplify
 1.13% :     0.000001s :     10: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.93% :     0.000001s :     10: predicate.compare_switch_simplify
 1.57% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 0.97% :     0.000001s :     10: predicate.dict_get_item_eliminator
 0.99% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.04% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.45% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.97% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.48% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.41% :     0.000002s :      6: predicate.get_grad_eliminate
 0.33% :     0.000000s :      3: predicate.graph_param_transform
 5.48% :     0.000007s :     22: predicate.inline
 0.76% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 2.41% :     0.000003s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.49% :     0.000002s :     15: predicate.load_eliminater
 2.44% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.37% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 0.90% :     0.000001s :     10: predicate.minmaximum_grad
 2.29% :     0.000003s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 1.85% :     0.000002s :     15: predicate.partial_eliminate
 1.04% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.35% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.68% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.47% :     0.000001s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.11% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.99% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 6.16% :     0.000008s :      6: predicate.specialize_transform
 1.26% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.03% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.49% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.03% :     0.000003s :     16: predicate.switch_defer_inline
 1.85% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.52% :     0.000009s :     53: predicate.switch_simplify
 0.99% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000002s :     10: predicate.transpose_eliminate
 1.11% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.95% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.52% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.28% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.32% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.42% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 4.00% :     0.000005s :     21: predicate.updatestate_useless_node_eliminater
 1.28% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001037    12
31.04% :     0.000322s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.01% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
63.96% :     0.000663s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.044070    76
 0.13% :     0.000057s :      1: add_recomputation
 0.15% :     0.000067s :      1: auto_monad
 0.04% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.81% :     0.000357s :      1: bootstrap
 0.08% :     0.000037s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.05% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.06% :     0.000024s :      1: event_method
 0.03% :     0.000014s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
31.20% :     0.013749s :      1: jit_opt_a
 0.39% :     0.000171s :      1: jit_opt_after_cconv
 0.12% :     0.000055s :      1: jit_opt_b
 5.90% :     0.002602s :      1: loop_unroll
 1.81% :     0.000797s :      1: mutable_eliminate
 3.88% :     0.001710s :     26: opt.transform.jit_opt_a
 0.12% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000024s :      4: opt.transform.jit_opt_b
 0.04% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.37% :     0.000602s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000041s :      1: py_interpret_to_execute
 0.04% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000018s :      1: remove_dup_value
 2.79% :     0.001228s :      1: renormalize.infer
 0.90% :     0.000396s :      1: renormalize.specialize
 0.03% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000042s :      1: rewriter_after_opt_a
 0.17% :     0.000074s :      1: rewriter_before_opt_a
 0.18% :     0.000080s :      1: symbol_engine_optimizer
31.18% :     0.013742s :      1: task_emit
17.81% :     0.007849s :      1: type_inference
 0.15% :     0.000066s :      1: validate


TotalTime = 0.0592582, [33]
[bootstrap]: 0.00034026
[type_inference]: 0.0108611
[event_method]: 2.644e-05
[auto_monad]: 7.956e-05
[graph_reusing]: 6.66999e-06
[pre_auto_parallel]: 3.66999e-06
[py_interpret_to_execute]: 5.738e-05
[rewriter_before_opt_a]: 8.787e-05
[expand_dump_flag]: 3.19001e-06
[jit_opt_a]: 0.00871914, [2]
    [Cycle 1]: 0.00225696, [27]
        [switch_simplify]: 7.421e-05
        [loop_unroll]: 3.148e-05
        [a_1]: 0.00069592
        [with_stream_mark]: 2.945e-05
        [recompute_prepare]: 8.1e-06
        [updatestate_depend_eliminate]: 4.20999e-06
        [updatestate_assign_eliminate]: 3.48999e-06
        [updatestate_loads_eliminate]: 2.81e-06
        [parameter_eliminate]: 2.38002e-06
        [specialize_transform]: 7.48e-06
        [updatestate_useless_node_eliminater]: 6.33e-06
        [accelerated_algorithm]: 5.88998e-06
        [meta_shard_fg_expand]: 2.64001e-06
        [get_grad_eliminate_]: 5.96e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 2.39999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.861e-05
        [j_node_and_user_rematch]: 1.049e-05
        [meta_fg_expand]: 2.74001e-06
        [replace_old_param]: 1.04e-05
        [inline_without_move]: 6.12001e-06
        [renormalize]: 0.00101866
        [add_forward_monad_depend]: 9.70002e-06
        [auto_monad_grad]: 2.92002e-06
        [auto_monad_eliminator]: 1.977e-05
        [cse]: 3.934e-05
        [replace_applicator]: 1.693e-05
    [Cycle 2]: 0.00037135, [27]
        [switch_simplify]: 6.17001e-06
        [loop_unroll]: 5.40001e-06
        [a_1]: 0.00010258
        [with_stream_mark]: 1.391e-05
        [recompute_prepare]: 5.22e-06
        [updatestate_depend_eliminate]: 3.2e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.40002e-06
        [parameter_eliminate]: 1.37e-06
        [specialize_transform]: 5.08002e-06
        [updatestate_useless_node_eliminater]: 5.13002e-06
        [accelerated_algorithm]: 5.61e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 4.60001e-06
        [merge_forward]: 2.74999e-06
        [cell_reuse_recompute_pass]: 4.67e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.798e-05
        [j_node_and_user_rematch]: 9.14e-06
        [meta_fg_expand]: 2.06998e-06
        [replace_old_param]: 9.14e-06
        [inline_without_move]: 5.33002e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.86e-06
        [auto_monad_grad]: 1.00001e-06
        [auto_monad_eliminator]: 6.23002e-06
        [cse]: 1.267e-05
        [replace_applicator]: 5.25999e-06
[py_interpret_to_execute_after_opt_a]: 1.48e-05
[rewriter_after_opt_a]: 4.106e-05
[convert_after_rewriter]: 7.43e-06
[order_py_execute_after_rewriter]: 5.59e-06
[mutable_eliminate]: 0.00085375
[jit_opt_b]: 5.589e-05, [1]
    [Cycle 1]: 4.75e-05, [2]
        [frontend_op_eliminate]: 1.678e-05
        [inline_after_opt_a]: 1.871e-05
[cconv]: 3.318e-05
[loop_unroll]: 0.00049167
[jit_opt_after_cconv]: 0.00017073, [1]
    [Cycle 1]: 0.00016339, [11]
        [c_1]: 2.432e-05
        [parameter_eliminate]: 5.67999e-06
        [updatestate_depend_eliminate]: 8.67998e-06
        [updatestate_assign_eliminate]: 2.88998e-06
        [updatestate_loads_eliminate]: 2.80997e-06
        [cse]: 2.658e-05
        [call_graph_tuple_transform]: 2.33e-05
        [tuple_list_get_item_eliminator]: 6.17999e-06
        [none_parameter_eliminate]: 1.70001e-06
        [renormalize]: 1.02e-06
        [switch_simplify]: 6.01e-06
[remove_dup_value]: 1.615e-05
[partial_unused_args_eliminate]: 2.79999e-06
[environ_conv]: 6.66999e-06
[add_recomputation]: 5.674e-05
[cse_after_recomputation]: 2.227e-05, [1]
    [Cycle 1]: 1.611e-05, [1]
        [cse]: 9.82001e-06
[auto_monad_reorder]: 1.708e-05
[get_jit_bprop_graph]: 2.36e-06
[rewriter_after_jit_bprop_graph]: 8.69e-06
[opt_after_jit_grad]: 0.0005141
[symbol_engine_optimizer]: 8.586e-05, [1]
    [Cycle 1]: 7.818e-05, [6]
        [build]: 4.62e-06
        [elim_shapecalc]: 9.74e-06
        [elim_not_effective]: 1.504e-05
        [opt_reshape]: 6.98e-06
        [fold_const_symbol]: 1.049e-05
        [renormalize]: 6.60017e-07
[validate]: 4.956e-05
[backend_pass]: 1.12e-06
[task_emit]: 0.0363404
[execute]: 1.025e-05
Sums
    bootstrap                            :     0.000340s :  0.65%
    type_inference                       :     0.010861s : 20.72%
    event_method                         :     0.000026s :  0.05%
    auto_monad                           :     0.000080s :  0.15%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000004s :  0.01%
    py_interpret_to_execute              :     0.000057s :  0.11%
    rewriter_before_opt_a                :     0.000088s :  0.17%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000080s :  0.15%
    jit_opt_a.loop_unroll                :     0.000037s :  0.07%
    jit_opt_a.a_1                        :     0.000798s :  1.52%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000013s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.02%
    jit_opt_a.merge_forward              :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.07%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000020s :  0.04%
    jit_opt_a.inline_without_move        :     0.000011s :  0.02%
    jit_opt_a.renormalize                :     0.001019s :  1.94%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.05%
    jit_opt_a.cse                        :     0.000052s :  0.10%
    jit_opt_a.replace_applicator         :     0.000022s :  0.04%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.03%
    rewriter_after_opt_a                 :     0.000041s :  0.08%
    convert_after_rewriter               :     0.000007s :  0.01%
    order_py_execute_after_rewriter      :     0.000006s :  0.01%
    mutable_eliminate                    :     0.000854s :  1.63%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.04%
    cconv                                :     0.000033s :  0.06%
    loop_unroll                          :     0.000492s :  0.94%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.05%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.01%
    remove_dup_value                     :     0.000016s :  0.03%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.01%
    add_recomputation                    :     0.000057s :  0.11%
    cse_after_recomputation.cse          :     0.000010s :  0.02%
    auto_monad_reorder                   :     0.000017s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.02%
    opt_after_jit_grad                   :     0.000514s :  0.98%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000050s :  0.09%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.036340s : 69.32%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000310    23
 0.65% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000002s :      2: substitution.fold_const_symbol
 2.33% :     0.000007s :      3: substitution.graph_param_transform
85.90% :     0.000266s :      4: substitution.inline
 1.34% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.05% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000006s :      2: substitution.replace_old_param
 5.17% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010753     2
39.65% :     0.004264s :      1: type_inference.infer
60.35% :     0.006490s :      1: type_inference.specialize
------[replace.]   0.000089     6
66.73% :     0.000059s :      4: replace.inline
33.27% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000278     6
94.68% :     0.000263s :      4: match.inline
 5.32% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.89% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000002s :     10: predicate.addn_zero_filter
 3.30% :     0.000005s :     10: predicate.arithmetic_simplify
 1.22% :     0.000002s :     10: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.91% :     0.000001s :     10: predicate.compare_switch_simplify
 1.73% :     0.000002s :     10: predicate.depend_value_elim
 0.96% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000002s :     10: predicate.environ_get_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.71% :     0.000008s :     22: predicate.inline
 1.32% :     0.000002s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.46% :     0.000002s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.61% :     0.000002s :     15: predicate.load_eliminater
 1.59% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.80% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.51% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 2.80% :     0.000004s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     15: predicate.partial_eliminate
 1.01% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.25% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.50% :     0.000002s :     18: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.62% :     0.000002s :     10: predicate.reshape_eliminate
 1.05% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.75% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.39% :     0.000010s :     53: predicate.switch_simplify
 1.25% :     0.000002s :     10: predicate.tile_eliminate
 1.01% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.76% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.18% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.43% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.15% :     0.000003s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.84% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.005849    12
 5.92% :     0.000346s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 1.05% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
93.04% :     0.005442s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.061442    76
 0.10% :     0.000060s :      1: add_recomputation
 0.14% :     0.000084s :      1: auto_monad
 0.03% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.59% :     0.000362s :      1: bootstrap
 0.06% :     0.000036s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.04% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.05% :     0.000032s :      1: event_method
 0.03% :     0.000016s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000009s :      1: graph_reusing
14.20% :     0.008724s :      1: jit_opt_a
 0.28% :     0.000173s :      1: jit_opt_after_cconv
 0.10% :     0.000059s :      1: jit_opt_b
 0.81% :     0.000500s :      1: loop_unroll
 1.41% :     0.000865s :      1: mutable_eliminate
 1.71% :     0.001050s :     26: opt.transform.jit_opt_a
 0.09% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000028s :      4: opt.transform.jit_opt_b
 0.02% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000038s :      4: opt.transform.symbol_engine_opt
 0.85% :     0.000524s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.10% :     0.000060s :      1: py_interpret_to_execute
 0.03% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000018s :      1: remove_dup_value
 0.90% :     0.000552s :      1: renormalize.infer
 0.74% :     0.000457s :      1: renormalize.specialize
 0.02% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000049s :      1: rewriter_after_opt_a
 0.15% :     0.000091s :      1: rewriter_before_opt_a
 0.14% :     0.000089s :      1: symbol_engine_optimizer
59.18% :     0.036363s :      1: task_emit
17.72% :     0.010889s :      1: type_inference
 0.12% :     0.000074s :      1: validate


TotalTime = 0.0352554, [33]
[bootstrap]: 0.00042432
[type_inference]: 0.00575608
[event_method]: 2.026e-05
[auto_monad]: 6.695e-05
[graph_reusing]: 6.06998e-06
[pre_auto_parallel]: 2.67001e-06
[py_interpret_to_execute]: 4.003e-05
[rewriter_before_opt_a]: 7.43e-05
[expand_dump_flag]: 3.41999e-06
[jit_opt_a]: 0.00763122, [2]
    [Cycle 1]: 0.00203861, [27]
        [switch_simplify]: 6.785e-05
        [loop_unroll]: 2.924e-05
        [a_1]: 0.00063056
        [with_stream_mark]: 2.744e-05
        [recompute_prepare]: 8.59e-06
        [updatestate_depend_eliminate]: 4.72998e-06
        [updatestate_assign_eliminate]: 3.16999e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [parameter_eliminate]: 2.02001e-06
        [specialize_transform]: 5.92001e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 6.06e-06
        [meta_shard_fg_expand]: 2.27999e-06
        [get_grad_eliminate_]: 5.17e-06
        [merge_forward]: 4.66002e-06
        [cell_reuse_recompute_pass]: 2.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.547e-05
        [j_node_and_user_rematch]: 9.17001e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 1.092e-05
        [inline_without_move]: 5.44e-06
        [renormalize]: 0.00089934
        [add_forward_monad_depend]: 9.07001e-06
        [auto_monad_grad]: 2.44001e-06
        [auto_monad_eliminator]: 1.765e-05
        [cse]: 3.317e-05
        [replace_applicator]: 1.71e-05
    [Cycle 2]: 0.00033315, [27]
        [switch_simplify]: 5.63002e-06
        [loop_unroll]: 4.94e-06
        [a_1]: 9.167e-05
        [with_stream_mark]: 1.384e-05
        [recompute_prepare]: 4.90001e-06
        [updatestate_depend_eliminate]: 2.41e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.32001e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 4.87e-06
        [updatestate_useless_node_eliminater]: 4.67998e-06
        [accelerated_algorithm]: 5.09998e-06
        [meta_shard_fg_expand]: 1.45999e-06
        [get_grad_eliminate_]: 4.45999e-06
        [merge_forward]: 3.14999e-06
        [cell_reuse_recompute_pass]: 1.98997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.488e-05
        [j_node_and_user_rematch]: 7.81001e-06
        [meta_fg_expand]: 1.42999e-06
        [replace_old_param]: 8.15e-06
        [inline_without_move]: 5.06002e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 1.42e-06
        [auto_monad_eliminator]: 5.17e-06
        [cse]: 1.183e-05
        [replace_applicator]: 5.14e-06
[py_interpret_to_execute_after_opt_a]: 1.547e-05
[rewriter_after_opt_a]: 4.187e-05
[convert_after_rewriter]: 6.83998e-06
[order_py_execute_after_rewriter]: 6.76999e-06
[mutable_eliminate]: 0.00087888
[jit_opt_b]: 5.339e-05, [1]
    [Cycle 1]: 4.513e-05, [2]
        [frontend_op_eliminate]: 1.686e-05
        [inline_after_opt_a]: 1.536e-05
[cconv]: 3.39e-05
[loop_unroll]: 0.00052106
[jit_opt_after_cconv]: 0.00016338, [1]
    [Cycle 1]: 0.00015583, [11]
        [c_1]: 2.015e-05
        [parameter_eliminate]: 4.52e-06
        [updatestate_depend_eliminate]: 1.03e-05
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 2.32001e-06
        [cse]: 2.896e-05
        [call_graph_tuple_transform]: 2.263e-05
        [tuple_list_get_item_eliminator]: 5.32001e-06
        [none_parameter_eliminate]: 1.48002e-06
        [renormalize]: 1.27e-06
        [switch_simplify]: 5.22999e-06
[remove_dup_value]: 1.506e-05
[partial_unused_args_eliminate]: 2.57001e-06
[environ_conv]: 5.97001e-06
[add_recomputation]: 5.564e-05
[cse_after_recomputation]: 2.134e-05, [1]
    [Cycle 1]: 1.563e-05, [1]
        [cse]: 1.004e-05
[auto_monad_reorder]: 1.624e-05
[get_jit_bprop_graph]: 2.53998e-06
[rewriter_after_jit_bprop_graph]: 9.06998e-06
[opt_after_jit_grad]: 0.00051844
[symbol_engine_optimizer]: 7.846e-05, [1]
    [Cycle 1]: 7.145e-05, [6]
        [build]: 5.32999e-06
        [elim_shapecalc]: 8.89e-06
        [elim_not_effective]: 1.499e-05
        [opt_reshape]: 5.24e-06
        [fold_const_symbol]: 8.98002e-06
        [renormalize]: 5.90022e-07
[validate]: 4.326e-05
[backend_pass]: 1.15999e-06
[task_emit]: 0.0184896
[execute]: 7.09001e-06
Sums
    bootstrap                            :     0.000424s :  1.45%
    type_inference                       :     0.005756s : 19.64%
    event_method                         :     0.000020s :  0.07%
    auto_monad                           :     0.000067s :  0.23%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000040s :  0.14%
    rewriter_before_opt_a                :     0.000074s :  0.25%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000073s :  0.25%
    jit_opt_a.loop_unroll                :     0.000034s :  0.12%
    jit_opt_a.a_1                        :     0.000722s :  2.46%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000008s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.01%
    jit_opt_a.replace_old_param          :     0.000019s :  0.07%
    jit_opt_a.inline_without_move        :     0.000011s :  0.04%
    jit_opt_a.renormalize                :     0.000899s :  3.07%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.08%
    jit_opt_a.cse                        :     0.000045s :  0.15%
    jit_opt_a.replace_applicator         :     0.000022s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.05%
    rewriter_after_opt_a                 :     0.000042s :  0.14%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000879s :  3.00%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.05%
    cconv                                :     0.000034s :  0.12%
    loop_unroll                          :     0.000521s :  1.78%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000015s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000056s :  0.19%
    cse_after_recomputation.cse          :     0.000010s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.06%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.03%
    opt_after_jit_grad                   :     0.000518s :  1.77%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.018490s : 63.10%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000287    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000002s :      2: substitution.fold_const_symbol
 2.05% :     0.000006s :      3: substitution.graph_param_transform
86.35% :     0.000248s :      4: substitution.inline
 1.32% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.45% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.52% :     0.000007s :      2: substitution.replace_old_param
 4.90% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005674     2
84.35% :     0.004786s :      1: type_inference.infer
15.65% :     0.000888s :      1: type_inference.specialize
------[replace.]   0.000074     6
68.08% :     0.000050s :      4: replace.inline
31.92% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000258     6
94.97% :     0.000245s :      4: match.inline
 5.03% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.72% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.53% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.44% :     0.000002s :     10: predicate.addn_zero_filter
 1.99% :     0.000002s :     10: predicate.arithmetic_simplify
 1.26% :     0.000002s :     10: predicate.cast_eliminate
 0.32% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000002s :     10: predicate.depend_value_elim
 0.99% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.59% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.03% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.22% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.03% :     0.000001s :     10: predicate.environ_get_eliminate
 1.47% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.87% :     0.000007s :     22: predicate.inline
 0.77% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.47% :     0.000002s :      6: predicate.less_batch_normalization
 1.35% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.59% :     0.000002s :     15: predicate.load_eliminater
 2.03% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.51% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 2.44% :     0.000003s :      3: predicate.mutable_eliminate
 0.37% :     0.000000s :      3: predicate.opt_reshape
 2.08% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.37% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.70% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.70% :     0.000002s :     10: predicate.reshape_eliminate
 1.32% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.09% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 1.85% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.99% :     0.000010s :     53: predicate.switch_simplify
 1.44% :     0.000002s :     10: predicate.tile_eliminate
 1.26% :     0.000002s :     10: predicate.transpose_eliminate
 1.14% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.26% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.84% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.56% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.99% :     0.000003s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000692    12
50.26% :     0.000348s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.54% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.19% :     0.000285s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.037216    76
 0.16% :     0.000059s :      1: add_recomputation
 0.19% :     0.000070s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.21% :     0.000449s :      1: bootstrap
 0.10% :     0.000037s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.06% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.07% :     0.000026s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
20.52% :     0.007636s :      1: jit_opt_a
 0.45% :     0.000166s :      1: jit_opt_after_cconv
 0.15% :     0.000056s :      1: jit_opt_b
 1.43% :     0.000531s :      1: loop_unroll
 2.39% :     0.000891s :      1: mutable_eliminate
 2.56% :     0.000952s :     26: opt.transform.jit_opt_a
 0.13% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000026s :      4: opt.transform.jit_opt_b
 0.04% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.42% :     0.000529s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000043s :      1: py_interpret_to_execute
 0.05% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 1.44% :     0.000536s :      1: renormalize.infer
 0.95% :     0.000354s :      1: renormalize.specialize
 0.03% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000045s :      1: rewriter_after_opt_a
 0.21% :     0.000078s :      1: rewriter_before_opt_a
 0.22% :     0.000081s :      1: symbol_engine_optimizer
49.73% :     0.018508s :      1: task_emit
15.53% :     0.005779s :      1: type_inference
 0.18% :     0.000066s :      1: validate


TotalTime = 0.0390636, [33]
[bootstrap]: 0.00032755
[type_inference]: 0.00877355
[event_method]: 2.114e-05
[auto_monad]: 7.051e-05
[graph_reusing]: 5.67001e-06
[pre_auto_parallel]: 2.55002e-06
[py_interpret_to_execute]: 4.214e-05
[rewriter_before_opt_a]: 7.365e-05
[expand_dump_flag]: 3.36999e-06
[jit_opt_a]: 0.0120737, [2]
    [Cycle 1]: 0.00389776, [27]
        [switch_simplify]: 6.316e-05
        [loop_unroll]: 2.813e-05
        [a_1]: 0.00060332
        [with_stream_mark]: 2.624e-05
        [recompute_prepare]: 6.98e-06
        [updatestate_depend_eliminate]: 3.68e-06
        [updatestate_assign_eliminate]: 2.88998e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 1.81998e-06
        [specialize_transform]: 5.66e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.21002e-06
        [meta_shard_fg_expand]: 2.11e-06
        [get_grad_eliminate_]: 5.15001e-06
        [merge_forward]: 3.11999e-06
        [cell_reuse_recompute_pass]: 9.79984e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.442e-05
        [j_node_and_user_rematch]: 9.49999e-06
        [meta_fg_expand]: 2.31998e-06
        [replace_old_param]: 9.22999e-06
        [inline_without_move]: 5.27999e-06
        [renormalize]: 0.00282305
        [add_forward_monad_depend]: 6.65002e-06
        [auto_monad_grad]: 2.49999e-06
        [auto_monad_eliminator]: 1.835e-05
        [cse]: 3.215e-05
        [replace_applicator]: 1.686e-05
    [Cycle 2]: 0.00032631, [27]
        [switch_simplify]: 5.99e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 9.202e-05
        [with_stream_mark]: 1.175e-05
        [recompute_prepare]: 4.79e-06
        [updatestate_depend_eliminate]: 2.71e-06
        [updatestate_assign_eliminate]: 2.94001e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [parameter_eliminate]: 1.19998e-06
        [specialize_transform]: 5.11997e-06
        [updatestate_useless_node_eliminater]: 4.60001e-06
        [accelerated_algorithm]: 4.79998e-06
        [meta_shard_fg_expand]: 1.45999e-06
        [get_grad_eliminate_]: 4.35999e-06
        [merge_forward]: 2.88e-06
        [cell_reuse_recompute_pass]: 2.12001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.327e-05
        [j_node_and_user_rematch]: 8.38999e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 7.78999e-06
        [inline_without_move]: 4.50999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 7.79983e-07
        [auto_monad_eliminator]: 5.32001e-06
        [cse]: 1.149e-05
        [replace_applicator]: 4.73001e-06
[py_interpret_to_execute_after_opt_a]: 1.507e-05
[rewriter_after_opt_a]: 3.909e-05
[convert_after_rewriter]: 6.86001e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.00066835
[jit_opt_b]: 5.007e-05, [1]
    [Cycle 1]: 4.211e-05, [2]
        [frontend_op_eliminate]: 1.605e-05
        [inline_after_opt_a]: 1.512e-05
[cconv]: 2.763e-05
[loop_unroll]: 0.00042106
[jit_opt_after_cconv]: 0.0001598, [1]
    [Cycle 1]: 0.00015292, [11]
        [c_1]: 2.096e-05
        [parameter_eliminate]: 2.79999e-06
        [updatestate_depend_eliminate]: 5.97999e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 3.583e-05
        [call_graph_tuple_transform]: 1.96e-05
        [tuple_list_get_item_eliminator]: 5.16002e-06
        [none_parameter_eliminate]: 1.72001e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 5.22999e-06
[remove_dup_value]: 1.402e-05
[partial_unused_args_eliminate]: 2.10002e-06
[environ_conv]: 5.59998e-06
[add_recomputation]: 5.46e-05
[cse_after_recomputation]: 1.982e-05, [1]
    [Cycle 1]: 1.401e-05, [1]
        [cse]: 8.61002e-06
[auto_monad_reorder]: 1.51e-05
[get_jit_bprop_graph]: 1.91e-06
[rewriter_after_jit_bprop_graph]: 5.68997e-06
[opt_after_jit_grad]: 0.00045997
[symbol_engine_optimizer]: 7.52e-05, [1]
    [Cycle 1]: 6.805e-05, [6]
        [build]: 4.84e-06
        [elim_shapecalc]: 7.53e-06
        [elim_not_effective]: 1.387e-05
        [opt_reshape]: 5.62001e-06
        [fold_const_symbol]: 8.36002e-06
        [renormalize]: 6.00005e-07
[validate]: 4.017e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.0153463
[execute]: 1.102e-05
Sums
    bootstrap                            :     0.000328s :  1.07%
    type_inference                       :     0.008774s : 28.71%
    event_method                         :     0.000021s :  0.07%
    auto_monad                           :     0.000071s :  0.23%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.14%
    rewriter_before_opt_a                :     0.000074s :  0.24%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000069s :  0.23%
    jit_opt_a.loop_unroll                :     0.000033s :  0.11%
    jit_opt_a.a_1                        :     0.000695s :  2.28%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000017s :  0.06%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.002823s :  9.24%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.08%
    jit_opt_a.cse                        :     0.000044s :  0.14%
    jit_opt_a.replace_applicator         :     0.000022s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.05%
    rewriter_after_opt_a                 :     0.000039s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000668s :  2.19%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.05%
    cconv                                :     0.000028s :  0.09%
    loop_unroll                          :     0.000421s :  1.38%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000036s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000014s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000055s :  0.18%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000015s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000460s :  1.51%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.015346s : 50.21%
    execute                              :     0.000011s :  0.04%


Time group info:
------[substitution.]   0.000264    23
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.05% :     0.000005s :      3: substitution.graph_param_transform
87.21% :     0.000230s :      4: substitution.inline
 1.42% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.51% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000005s :      2: substitution.replace_old_param
 4.61% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008689     2
78.58% :     0.006828s :      1: type_inference.infer
21.42% :     0.001861s :      1: type_inference.specialize
------[replace.]   0.000073     6
67.00% :     0.000049s :      4: replace.inline
33.00% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000239     6
95.35% :     0.000228s :      4: match.inline
 4.65% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.21% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.61% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 3.07% :     0.000004s :     10: predicate.arithmetic_simplify
 1.43% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.41% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.07% :     0.000001s :     10: predicate.environ_get_eliminate
 1.53% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.73% :     0.000007s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.98% :     0.000001s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.95% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.84% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.40% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.07% :     0.000002s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.56% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.84% :     0.000001s :      6: predicate.replace_old_param
 0.47% :     0.000001s :      3: predicate.reset_defer_inline
 1.60% :     0.000002s :     10: predicate.reshape_eliminate
 1.38% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.01% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.25% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.60% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.67% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.17% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.39% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.88% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000680    12
47.37% :     0.000322s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.24% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.39% :     0.000302s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042891    76
 0.13% :     0.000057s :      1: add_recomputation
 0.17% :     0.000074s :      1: auto_monad
 0.04% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.81% :     0.000346s :      1: bootstrap
 0.07% :     0.000031s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.05% :     0.000022s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.06% :     0.000026s :      1: event_method
 0.04% :     0.000016s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
28.16% :     0.012077s :      1: jit_opt_a
 0.38% :     0.000163s :      1: jit_opt_after_cconv
 0.12% :     0.000053s :      1: jit_opt_b
 1.00% :     0.000430s :      1: loop_unroll
 1.58% :     0.000678s :      1: mutable_eliminate
 2.13% :     0.000913s :     26: opt.transform.jit_opt_a
 0.11% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000024s :      4: opt.transform.jit_opt_b
 0.03% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.09% :     0.000468s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.10% :     0.000045s :      1: py_interpret_to_execute
 0.04% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000016s :      1: remove_dup_value
 5.66% :     0.002427s :      1: renormalize.infer
 0.89% :     0.000383s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000042s :      1: rewriter_after_opt_a
 0.18% :     0.000076s :      1: rewriter_before_opt_a
 0.18% :     0.000078s :      1: symbol_engine_optimizer
35.83% :     0.015367s :      1: task_emit
20.50% :     0.008792s :      1: type_inference
 0.15% :     0.000064s :      1: validate

.
TotalTime = 0.0154829, [30]
[bootstrap]: 0.00034011
[type_inference]: 0.00573804
[event_method]: 2.022e-05
[auto_monad]: 7.034e-05
[graph_reusing]: 6.78998e-06
[pre_auto_parallel]: 2.93998e-06
[py_interpret_to_execute]: 4.299e-05
[rewriter_before_opt_a]: 7.887e-05
[expand_dump_flag]: 3.58e-06
[jit_opt_a]: 0.00650947, [2]
    [Cycle 1]: 0.00210522, [27]
        [switch_simplify]: 6.687e-05
        [loop_unroll]: 3.014e-05
        [a_1]: 0.00060833
        [with_stream_mark]: 2.354e-05
        [recompute_prepare]: 8.38999e-06
        [updatestate_depend_eliminate]: 4.38001e-06
        [updatestate_assign_eliminate]: 2.92002e-06
        [updatestate_loads_eliminate]: 2.66e-06
        [parameter_eliminate]: 2.04999e-06
        [specialize_transform]: 6.93e-06
        [updatestate_useless_node_eliminater]: 5.77999e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 2.44001e-06
        [get_grad_eliminate_]: 5.95002e-06
        [merge_forward]: 4.07e-06
        [cell_reuse_recompute_pass]: 1.50001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.749e-05
        [j_node_and_user_rematch]: 1.092e-05
        [meta_fg_expand]: 2.78e-06
        [replace_old_param]: 1.082e-05
        [inline_without_move]: 5.89999e-06
        [renormalize]: 0.00099733
        [add_forward_monad_depend]: 7.56999e-06
        [auto_monad_grad]: 3.06001e-06
        [auto_monad_eliminator]: 1.706e-05
        [cse]: 3.503e-05
        [replace_applicator]: 1.701e-05
    [Cycle 2]: 0.00035152, [27]
        [switch_simplify]: 5.71998e-06
        [loop_unroll]: 5.39e-06
        [a_1]: 9.799e-05
        [with_stream_mark]: 1.458e-05
        [recompute_prepare]: 5.72999e-06
        [updatestate_depend_eliminate]: 3.36999e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [parameter_eliminate]: 2.02999e-06
        [specialize_transform]: 5.29998e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.55001e-06
        [meta_shard_fg_expand]: 1.79998e-06
        [get_grad_eliminate_]: 4.85001e-06
        [merge_forward]: 3.04001e-06
        [cell_reuse_recompute_pass]: 2.76999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.576e-05
        [j_node_and_user_rematch]: 9.10999e-06
        [meta_fg_expand]: 1.45999e-06
        [replace_old_param]: 8.62e-06
        [inline_without_move]: 5.10999e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.51002e-06
        [auto_monad_grad]: 1.01002e-06
        [auto_monad_eliminator]: 5.35999e-06
        [cse]: 1.195e-05
        [replace_applicator]: 5.25999e-06
[py_interpret_to_execute_after_opt_a]: 1.366e-05
[rewriter_after_opt_a]: 4.182e-05
[convert_after_rewriter]: 7.41001e-06
[order_py_execute_after_rewriter]: 5.19e-06
[mutable_eliminate]: 0.0008446
[jit_opt_b]: 0.00010392, [1]
    [Cycle 1]: 9.442e-05, [2]
        [frontend_op_eliminate]: 2.492e-05
        [inline_after_opt_a]: 5.404e-05
[cconv]: 3.438e-05
[loop_unroll]: 0.00048493
[jit_opt_after_cconv]: 0.00016676, [1]
    [Cycle 1]: 0.00015913, [11]
        [c_1]: 2.398e-05
        [parameter_eliminate]: 4.17e-06
        [updatestate_depend_eliminate]: 8.99e-06
        [updatestate_assign_eliminate]: 2.92002e-06
        [updatestate_loads_eliminate]: 2.94001e-06
        [cse]: 2.72e-05
        [call_graph_tuple_transform]: 2.245e-05
        [tuple_list_get_item_eliminator]: 5.62999e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 9.39996e-07
        [switch_simplify]: 5.75001e-06
[remove_dup_value]: 1.584e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 6.41e-06
[add_recomputation]: 5.546e-05
[cse_after_recomputation]: 2.352e-05, [1]
    [Cycle 1]: 1.706e-05, [1]
        [cse]: 1.044e-05
[auto_monad_reorder]: 1.644e-05
[get_jit_bprop_graph]: 2.51e-06
[rewriter_after_jit_bprop_graph]: 8.27e-06
[opt_after_jit_grad]: 0.00049879
[symbol_engine_optimizer]: 8.444e-05, [1]
    [Cycle 1]: 7.594e-05, [6]
        [build]: 5.03002e-06
        [elim_shapecalc]: 8.71997e-06
        [elim_not_effective]: 1.454e-05
        [opt_reshape]: 6.79999e-06
        [fold_const_symbol]: 1.015e-05
        [renormalize]: 7.7e-07
[validate]: 4.787e-05
Sums
    bootstrap                            :     0.000340s :  3.16%
    type_inference                       :     0.005738s : 53.27%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000070s :  0.65%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000043s :  0.40%
    rewriter_before_opt_a                :     0.000079s :  0.73%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000073s :  0.67%
    jit_opt_a.loop_unroll                :     0.000036s :  0.33%
    jit_opt_a.a_1                        :     0.000706s :  6.56%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.18%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000997s :  9.26%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.21%
    jit_opt_a.cse                        :     0.000047s :  0.44%
    jit_opt_a.replace_applicator         :     0.000022s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.13%
    rewriter_after_opt_a                 :     0.000042s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000845s :  7.84%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000054s :  0.50%
    cconv                                :     0.000034s :  0.32%
    loop_unroll                          :     0.000485s :  4.50%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000027s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000016s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000055s :  0.51%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.15%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.08%
    opt_after_jit_grad                   :     0.000499s :  4.63%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000048s :  0.44%


Time group info:
------[substitution.]   0.000252    23
 0.80% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000002s :      2: substitution.fold_const_symbol
 2.44% :     0.000006s :      3: substitution.graph_param_transform
84.72% :     0.000214s :      4: substitution.inline
 1.36% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.19% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.27% :     0.000006s :      2: substitution.replace_old_param
 5.53% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005657     2
81.11% :     0.004588s :      1: type_inference.infer
18.89% :     0.001069s :      1: type_inference.specialize
------[replace.]   0.000082     6
67.89% :     0.000056s :      4: replace.inline
32.11% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000224     6
94.29% :     0.000211s :      4: match.inline
 5.71% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.48% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 2.27% :     0.000003s :     10: predicate.arithmetic_simplify
 1.30% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.32% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.60% :     0.000007s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000002s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 1.65% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.80% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.24% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 2.10% :     0.000003s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     15: predicate.partial_eliminate
 1.28% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.68% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.64% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 1.13% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.26% :     0.000010s :     53: predicate.switch_simplify
 1.34% :     0.000002s :     10: predicate.tile_eliminate
 1.15% :     0.000002s :     10: predicate.transpose_eliminate
 1.24% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.02% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.71% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.93% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.81% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000704    12
42.76% :     0.000301s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.77% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.48% :     0.000362s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017516    72
 0.33% :     0.000058s :      1: add_recomputation
 0.42% :     0.000074s :      1: auto_monad
 0.11% :     0.000020s :      1: auto_monad_reorder
 2.03% :     0.000356s :      1: bootstrap
 0.21% :     0.000038s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000010s :      1: environ_conv
 0.14% :     0.000025s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.06% :     0.000010s :      1: graph_reusing
37.19% :     0.006515s :      1: jit_opt_a
 0.97% :     0.000170s :      1: jit_opt_after_cconv
 0.61% :     0.000107s :      1: jit_opt_b
 2.83% :     0.000496s :      1: loop_unroll
 4.91% :     0.000859s :      1: mutable_eliminate
 5.39% :     0.000945s :     26: opt.transform.jit_opt_a
 0.31% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.40% :     0.000070s :      4: opt.transform.jit_opt_b
 0.08% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.90% :     0.000508s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000045s :      1: py_interpret_to_execute
 0.09% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.96% :     0.000519s :      1: renormalize.infer
 2.68% :     0.000469s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000045s :      1: rewriter_after_opt_a
 0.47% :     0.000082s :      1: rewriter_before_opt_a
 0.50% :     0.000087s :      1: symbol_engine_optimizer
32.85% :     0.005755s :      1: type_inference


TotalTime = 0.0428896, [33]
[bootstrap]: 0.00030036
[type_inference]: 0.0130517
[event_method]: 1.979e-05
[auto_monad]: 6.544e-05
[graph_reusing]: 5.89999e-06
[pre_auto_parallel]: 2.86e-06
[py_interpret_to_execute]: 3.926e-05
[rewriter_before_opt_a]: 7.178e-05
[expand_dump_flag]: 3.27002e-06
[jit_opt_a]: 0.0143482, [2]
    [Cycle 1]: 0.00355738, [27]
        [switch_simplify]: 6.644e-05
        [loop_unroll]: 2.838e-05
        [a_1]: 0.00058833
        [with_stream_mark]: 2.612e-05
        [recompute_prepare]: 7.20003e-06
        [updatestate_depend_eliminate]: 3.76001e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.60002e-06
        [parameter_eliminate]: 1.93002e-06
        [specialize_transform]: 5.57999e-06
        [updatestate_useless_node_eliminater]: 4.81002e-06
        [accelerated_algorithm]: 5.67999e-06
        [meta_shard_fg_expand]: 2.41e-06
        [get_grad_eliminate_]: 4.95001e-06
        [merge_forward]: 3.75998e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.371e-05
        [j_node_and_user_rematch]: 8.97e-06
        [meta_fg_expand]: 1.97999e-06
        [replace_old_param]: 9.31998e-06
        [inline_without_move]: 5.05001e-06
        [renormalize]: 0.00247338
        [add_forward_monad_depend]: 9.81e-06
        [auto_monad_grad]: 2.49999e-06
        [auto_monad_eliminator]: 2.056e-05
        [cse]: 3.508e-05
        [replace_applicator]: 1.817e-05
    [Cycle 2]: 0.00123183, [27]
        [switch_simplify]: 5.92001e-06
        [loop_unroll]: 5.04e-06
        [a_1]: 0.00092936
        [with_stream_mark]: 2.378e-05
        [recompute_prepare]: 7.73999e-06
        [updatestate_depend_eliminate]: 3.51999e-06
        [updatestate_assign_eliminate]: 2.88998e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [parameter_eliminate]: 1.74e-06
        [specialize_transform]: 5.06997e-06
        [updatestate_useless_node_eliminater]: 4.92999e-06
        [accelerated_algorithm]: 6.03002e-06
        [meta_shard_fg_expand]: 2.25002e-06
        [get_grad_eliminate_]: 4.53001e-06
        [merge_forward]: 3.43e-06
        [cell_reuse_recompute_pass]: 3.46999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.937e-05
        [j_node_and_user_rematch]: 8.57e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 9.09e-06
        [inline_without_move]: 1.077e-05
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 2.44001e-06
        [auto_monad_grad]: 1.49e-06
        [auto_monad_eliminator]: 8.33999e-06
        [cse]: 1.965e-05
        [replace_applicator]: 5.81003e-06
[py_interpret_to_execute_after_opt_a]: 1.822e-05
[rewriter_after_opt_a]: 3.968e-05
[convert_after_rewriter]: 6.88998e-06
[order_py_execute_after_rewriter]: 4.77e-06
[mutable_eliminate]: 0.00078146
[jit_opt_b]: 5.231e-05, [1]
    [Cycle 1]: 4.356e-05, [2]
        [frontend_op_eliminate]: 1.668e-05
        [inline_after_opt_a]: 1.477e-05
[cconv]: 3.176e-05
[loop_unroll]: 0.0004977
[jit_opt_after_cconv]: 0.00014631, [1]
    [Cycle 1]: 0.00013971, [11]
        [c_1]: 2.1e-05
        [parameter_eliminate]: 3.03e-06
        [updatestate_depend_eliminate]: 8.04002e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 2.261e-05
        [call_graph_tuple_transform]: 1.859e-05
        [tuple_list_get_item_eliminator]: 5.25999e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 5.99975e-07
        [switch_simplify]: 5.39e-06
[remove_dup_value]: 1.422e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 5.83002e-06
[add_recomputation]: 5.484e-05
[cse_after_recomputation]: 2.038e-05, [1]
    [Cycle 1]: 1.495e-05, [1]
        [cse]: 9.45001e-06
[auto_monad_reorder]: 0.00218523
[get_jit_bprop_graph]: 2.91e-06
[rewriter_after_jit_bprop_graph]: 1.114e-05
[opt_after_jit_grad]: 0.00065003
[symbol_engine_optimizer]: 8.202e-05, [1]
    [Cycle 1]: 7.33e-05, [6]
        [build]: 6.12999e-06
        [elim_shapecalc]: 8.05999e-06
        [elim_not_effective]: 1.706e-05
        [opt_reshape]: 5.42001e-06
        [fold_const_symbol]: 8.37e-06
        [renormalize]: 1.00999e-06
[validate]: 4.552e-05
[backend_pass]: 1.14e-06
[task_emit]: 0.0100583
[execute]: 9.31e-06
Sums
    bootstrap                            :     0.000300s :  0.92%
    type_inference                       :     0.013052s : 40.02%
    event_method                         :     0.000020s :  0.06%
    auto_monad                           :     0.000065s :  0.20%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.12%
    rewriter_before_opt_a                :     0.000072s :  0.22%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000072s :  0.22%
    jit_opt_a.loop_unroll                :     0.000033s :  0.10%
    jit_opt_a.a_1                        :     0.001518s :  4.65%
    jit_opt_a.with_stream_mark           :     0.000050s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000018s :  0.06%
    jit_opt_a.inline_without_move        :     0.000016s :  0.05%
    jit_opt_a.renormalize                :     0.002473s :  7.58%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000029s :  0.09%
    jit_opt_a.cse                        :     0.000055s :  0.17%
    jit_opt_a.replace_applicator         :     0.000024s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.06%
    rewriter_after_opt_a                 :     0.000040s :  0.12%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000781s :  2.40%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.05%
    cconv                                :     0.000032s :  0.10%
    loop_unroll                          :     0.000498s :  1.53%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000014s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000055s :  0.17%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.002185s :  6.70%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.03%
    opt_after_jit_grad                   :     0.000650s :  1.99%
    symbol_engine_optimizer.build        :     0.000006s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010058s : 30.84%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000254    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000002s :      2: substitution.fold_const_symbol
 2.09% :     0.000005s :      3: substitution.graph_param_transform
85.79% :     0.000218s :      4: substitution.inline
 1.53% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.83% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.43% :     0.000006s :      2: substitution.replace_old_param
 4.87% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012970     2
92.84% :     0.012042s :      1: type_inference.infer
 7.16% :     0.000928s :      1: type_inference.specialize
------[replace.]   0.000072     6
69.05% :     0.000050s :      4: replace.inline
30.95% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000227     6
95.00% :     0.000216s :      4: match.inline
 5.00% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.13% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.74% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.64% :     0.000002s :     10: predicate.addn_zero_filter
 3.14% :     0.000004s :     10: predicate.arithmetic_simplify
 1.79% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.52% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.05% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.54% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.02% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.55% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000001s :      3: predicate.graph_param_transform
 7.15% :     0.000009s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.51% :     0.000002s :      6: predicate.less_batch_normalization
 2.01% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.58% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.53% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.94% :     0.000001s :     10: predicate.merge_addn
 0.93% :     0.000001s :     10: predicate.minmaximum_grad
 2.00% :     0.000003s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.01% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.22% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.81% :     0.000001s :      6: predicate.replace_old_param
 0.42% :     0.000001s :      3: predicate.reset_defer_inline
 1.22% :     0.000002s :     10: predicate.reshape_eliminate
 1.08% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.97% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.03% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 1.88% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.22% :     0.000009s :     53: predicate.switch_simplify
 1.06% :     0.000001s :     10: predicate.tile_eliminate
 0.99% :     0.000001s :     10: predicate.transpose_eliminate
 1.14% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.66% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.77% :     0.000005s :     21: predicate.updatestate_useless_node_eliminater
 1.48% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000712    12
50.74% :     0.000361s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.58% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.68% :     0.000297s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.047221    76
 0.12% :     0.000058s :      1: add_recomputation
 0.15% :     0.000069s :      1: auto_monad
 4.67% :     0.002206s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.68% :     0.000321s :      1: bootstrap
 0.07% :     0.000034s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.05% :     0.000022s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.05% :     0.000024s :      1: event_method
 0.03% :     0.000016s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000007s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
30.39% :     0.014352s :      1: jit_opt_a
 0.32% :     0.000149s :      1: jit_opt_after_cconv
 0.12% :     0.000055s :      1: jit_opt_b
 1.07% :     0.000506s :      1: loop_unroll
 1.68% :     0.000792s :      1: mutable_eliminate
 3.72% :     0.001756s :     26: opt.transform.jit_opt_a
 0.10% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000024s :      4: opt.transform.jit_opt_b
 0.03% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.40% :     0.000660s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000042s :      1: py_interpret_to_execute
 0.04% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000017s :      1: remove_dup_value
 4.38% :     0.002068s :      1: renormalize.infer
 0.83% :     0.000394s :      1: renormalize.specialize
 0.03% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000043s :      1: rewriter_after_opt_a
 0.16% :     0.000075s :      1: rewriter_before_opt_a
 0.18% :     0.000085s :      1: symbol_engine_optimizer
21.34% :     0.010078s :      1: task_emit
27.68% :     0.013073s :      1: type_inference
 0.15% :     0.000070s :      1: validate


TotalTime = 0.0201657, [33]
[bootstrap]: 0.00054572
[type_inference]: 0.00516243
[event_method]: 1.892e-05
[auto_monad]: 6.681e-05
[graph_reusing]: 6.14999e-06
[pre_auto_parallel]: 2.61999e-06
[py_interpret_to_execute]: 3.539e-05
[rewriter_before_opt_a]: 7.349e-05
[expand_dump_flag]: 3.16999e-06
[jit_opt_a]: 0.00593908, [2]
    [Cycle 1]: 0.00180112, [27]
        [switch_simplify]: 5.722e-05
        [loop_unroll]: 2.817e-05
        [a_1]: 0.00056848
        [with_stream_mark]: 2.425e-05
        [recompute_prepare]: 7.53e-06
        [updatestate_depend_eliminate]: 4.43999e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 2.80002e-06
        [parameter_eliminate]: 2.35002e-06
        [specialize_transform]: 6.26998e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 6.34999e-06
        [meta_shard_fg_expand]: 2.02999e-06
        [get_grad_eliminate_]: 5.07999e-06
        [merge_forward]: 3.5e-06
        [cell_reuse_recompute_pass]: 1.21002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.566e-05
        [j_node_and_user_rematch]: 9.05001e-06
        [meta_fg_expand]: 2.25002e-06
        [replace_old_param]: 8.62998e-06
        [inline_without_move]: 5.44998e-06
        [renormalize]: 0.00077837
        [add_forward_monad_depend]: 6.21e-06
        [auto_monad_grad]: 2.42001e-06
        [auto_monad_eliminator]: 1.506e-05
        [cse]: 3.209e-05
        [replace_applicator]: 1.189e-05
    [Cycle 2]: 0.00031368, [27]
        [switch_simplify]: 5.76e-06
        [loop_unroll]: 4.74e-06
        [a_1]: 8.73e-05
        [with_stream_mark]: 8.67998e-06
        [recompute_prepare]: 4.85999e-06
        [updatestate_depend_eliminate]: 2.81e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 4.87e-06
        [updatestate_useless_node_eliminater]: 4.60001e-06
        [accelerated_algorithm]: 4.89998e-06
        [meta_shard_fg_expand]: 1.11002e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.27001e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.308e-05
        [j_node_and_user_rematch]: 7.43e-06
        [meta_fg_expand]: 1.45001e-06
        [replace_old_param]: 7.5e-06
        [inline_without_move]: 4.48001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 7.79983e-07
        [auto_monad_eliminator]: 4.67998e-06
        [cse]: 1.14e-05
        [replace_applicator]: 4.65999e-06
[py_interpret_to_execute_after_opt_a]: 1.036e-05
[rewriter_after_opt_a]: 3.759e-05
[convert_after_rewriter]: 6.78e-06
[order_py_execute_after_rewriter]: 4.92e-06
[mutable_eliminate]: 0.00058089
[jit_opt_b]: 5.356e-05, [1]
    [Cycle 1]: 4.529e-05, [2]
        [frontend_op_eliminate]: 1.687e-05
        [inline_after_opt_a]: 1.665e-05
[cconv]: 2.843e-05
[loop_unroll]: 0.00042984
[jit_opt_after_cconv]: 0.00014033, [1]
    [Cycle 1]: 0.00013375, [11]
        [c_1]: 2.043e-05
        [parameter_eliminate]: 2.91e-06
        [updatestate_depend_eliminate]: 6.02001e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.875e-05
        [call_graph_tuple_transform]: 1.865e-05
        [tuple_list_get_item_eliminator]: 5.07e-06
        [none_parameter_eliminate]: 1.90001e-06
        [renormalize]: 8.79983e-07
        [switch_simplify]: 5.30999e-06
[remove_dup_value]: 1.424e-05
[partial_unused_args_eliminate]: 2.39001e-06
[environ_conv]: 4.73001e-06
[add_recomputation]: 5.102e-05
[cse_after_recomputation]: 1.972e-05, [1]
    [Cycle 1]: 1.426e-05, [1]
        [cse]: 8.78001e-06
[auto_monad_reorder]: 1.612e-05
[get_jit_bprop_graph]: 1.94999e-06
[rewriter_after_jit_bprop_graph]: 5.02e-06
[opt_after_jit_grad]: 0.00044801
[symbol_engine_optimizer]: 7.089e-05, [1]
    [Cycle 1]: 6.437e-05, [6]
        [build]: 3.59002e-06
        [elim_shapecalc]: 7.06999e-06
        [elim_not_effective]: 1.238e-05
        [opt_reshape]: 5.15999e-06
        [fold_const_symbol]: 8.65001e-06
        [renormalize]: 6.29982e-07
[validate]: 3.926e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00611865
[execute]: 9.48997e-06
Sums
    bootstrap                            :     0.000546s :  3.48%
    type_inference                       :     0.005162s : 32.87%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000067s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.23%
    rewriter_before_opt_a                :     0.000073s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.21%
    jit_opt_a.a_1                        :     0.000656s :  4.18%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000778s :  4.96%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000043s :  0.28%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000581s :  3.70%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.11%
    cconv                                :     0.000028s :  0.18%
    loop_unroll                          :     0.000430s :  2.74%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000051s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000448s :  2.85%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006119s : 38.96%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000229    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000002s :      2: substitution.fold_const_symbol
 2.63% :     0.000006s :      3: substitution.graph_param_transform
84.71% :     0.000194s :      4: substitution.inline
 1.35% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.29% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000004s :      2: substitution.replace_old_param
 5.55% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005091     2
84.08% :     0.004280s :      1: type_inference.infer
15.92% :     0.000811s :      1: type_inference.specialize
------[replace.]   0.000066     6
67.07% :     0.000045s :      4: replace.inline
32.93% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000203     6
94.30% :     0.000191s :      4: match.inline
 5.70% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.64% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.03% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 2.01% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.71% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.78% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000000s :      3: predicate.elim_not_effective
 0.45% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.85% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000001s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.86% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.25% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.50% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.35% :     0.000003s :     15: predicate.partial_eliminate
 1.27% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.43% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 1.99% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.86% :     0.000009s :     53: predicate.switch_simplify
 1.33% :     0.000002s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.88% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.84% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000604    12
52.19% :     0.000315s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.28% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.53% :     0.000239s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021905    76
 0.25% :     0.000054s :      1: add_recomputation
 0.32% :     0.000070s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 2.59% :     0.000567s :      1: bootstrap
 0.14% :     0.000031s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.11% :     0.000023s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.13% :     0.005943s :      1: jit_opt_a
 0.65% :     0.000143s :      1: jit_opt_after_cconv
 0.26% :     0.000056s :      1: jit_opt_b
 2.00% :     0.000437s :      1: loop_unroll
 2.70% :     0.000591s :      1: mutable_eliminate
 3.95% :     0.000865s :     26: opt.transform.jit_opt_a
 0.21% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.08% :     0.000456s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000038s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.94% :     0.000425s :      1: renormalize.infer
 1.58% :     0.000345s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000040s :      1: rewriter_after_opt_a
 0.35% :     0.000076s :      1: rewriter_before_opt_a
 0.34% :     0.000073s :      1: symbol_engine_optimizer
28.02% :     0.006137s :      1: task_emit
23.64% :     0.005179s :      1: type_inference
 0.27% :     0.000059s :      1: validate


TotalTime = 0.0175786, [30]
[bootstrap]: 0.0004189
[type_inference]: 0.00630301
[event_method]: 2.142e-05
[auto_monad]: 7.06e-05
[graph_reusing]: 5.96998e-06
[pre_auto_parallel]: 2.64999e-06
[py_interpret_to_execute]: 4.367e-05
[rewriter_before_opt_a]: 7.596e-05
[expand_dump_flag]: 3.21999e-06
[jit_opt_a]: 0.00651254, [2]
    [Cycle 1]: 0.00203744, [27]
        [switch_simplify]: 6.727e-05
        [loop_unroll]: 3.262e-05
        [a_1]: 0.00061049
        [with_stream_mark]: 2.549e-05
        [recompute_prepare]: 7.67998e-06
        [updatestate_depend_eliminate]: 4.03999e-06
        [updatestate_assign_eliminate]: 3.09001e-06
        [updatestate_loads_eliminate]: 2.83e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 6.41998e-06
        [updatestate_useless_node_eliminater]: 5.61e-06
        [accelerated_algorithm]: 6.16e-06
        [meta_shard_fg_expand]: 2.37001e-06
        [get_grad_eliminate_]: 5.62999e-06
        [merge_forward]: 3.76999e-06
        [cell_reuse_recompute_pass]: 1.34998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.543e-05
        [j_node_and_user_rematch]: 1.069e-05
        [meta_fg_expand]: 2.62001e-06
        [replace_old_param]: 1.083e-05
        [inline_without_move]: 6.22001e-06
        [renormalize]: 0.00094579
        [add_forward_monad_depend]: 5.44e-06
        [auto_monad_grad]: 2.48e-06
        [auto_monad_eliminator]: 1.512e-05
        [cse]: 2.999e-05
        [replace_applicator]: 1.417e-05
    [Cycle 2]: 0.00032621, [27]
        [switch_simplify]: 5.71e-06
        [loop_unroll]: 4.89e-06
        [a_1]: 9.287e-05
        [with_stream_mark]: 9.10001e-06
        [recompute_prepare]: 5.25999e-06
        [updatestate_depend_eliminate]: 2.59999e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 5.10999e-06
        [updatestate_useless_node_eliminater]: 5.06002e-06
        [accelerated_algorithm]: 5.63002e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 4.99e-06
        [merge_forward]: 2.76e-06
        [cell_reuse_recompute_pass]: 1.99e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.397e-05
        [j_node_and_user_rematch]: 8.42e-06
        [meta_fg_expand]: 2.29001e-06
        [replace_old_param]: 8.71002e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 9.89996e-07
        [auto_monad_grad]: 1.12e-06
        [auto_monad_eliminator]: 5.47001e-06
        [cse]: 1.046e-05
        [replace_applicator]: 4.67e-06
[py_interpret_to_execute_after_opt_a]: 1.803e-05
[rewriter_after_opt_a]: 3.975e-05
[convert_after_rewriter]: 7.35e-06
[order_py_execute_after_rewriter]: 6.31e-06
[mutable_eliminate]: 0.0007184
[jit_opt_b]: 5.977e-05, [1]
    [Cycle 1]: 5.125e-05, [2]
        [frontend_op_eliminate]: 1.767e-05
        [inline_after_opt_a]: 2.126e-05
[cconv]: 2.822e-05
[loop_unroll]: 0.00215172
[jit_opt_after_cconv]: 0.00015924, [1]
    [Cycle 1]: 0.00015065, [11]
        [c_1]: 2.268e-05
        [parameter_eliminate]: 3.53999e-06
        [updatestate_depend_eliminate]: 6.93e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 2.327e-05
        [call_graph_tuple_transform]: 2.205e-05
        [tuple_list_get_item_eliminator]: 6.32001e-06
        [none_parameter_eliminate]: 1.65001e-06
        [renormalize]: 9.19972e-07
        [switch_simplify]: 5.56e-06
[remove_dup_value]: 1.706e-05
[partial_unused_args_eliminate]: 2.39999e-06
[environ_conv]: 5.54e-06
[add_recomputation]: 5.27e-05
[cse_after_recomputation]: 2.293e-05, [1]
    [Cycle 1]: 1.626e-05, [1]
        [cse]: 1.043e-05
[auto_monad_reorder]: 1.697e-05
[get_jit_bprop_graph]: 2.24999e-06
[rewriter_after_jit_bprop_graph]: 7.11001e-06
[opt_after_jit_grad]: 0.00048073
[symbol_engine_optimizer]: 8.064e-05, [1]
    [Cycle 1]: 7.27e-05, [6]
        [build]: 4.41002e-06
        [elim_shapecalc]: 8.23999e-06
        [elim_not_effective]: 1.386e-05
        [opt_reshape]: 6.57002e-06
        [fold_const_symbol]: 9.82001e-06
        [renormalize]: 6.10016e-07
[validate]: 4.258e-05
Sums
    bootstrap                            :     0.000419s :  3.27%
    type_inference                       :     0.006303s : 49.27%
    event_method                         :     0.000021s :  0.17%
    auto_monad                           :     0.000071s :  0.55%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000044s :  0.34%
    rewriter_before_opt_a                :     0.000076s :  0.59%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000073s :  0.57%
    jit_opt_a.loop_unroll                :     0.000038s :  0.29%
    jit_opt_a.a_1                        :     0.000703s :  5.50%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.08%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000020s :  0.15%
    jit_opt_a.inline_without_move        :     0.000012s :  0.09%
    jit_opt_a.renormalize                :     0.000946s :  7.39%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.16%
    jit_opt_a.cse                        :     0.000040s :  0.32%
    jit_opt_a.replace_applicator         :     0.000019s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.14%
    rewriter_after_opt_a                 :     0.000040s :  0.31%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000718s :  5.62%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000021s :  0.17%
    cconv                                :     0.000028s :  0.22%
    loop_unroll                          :     0.002152s : 16.82%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000017s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000053s :  0.41%
    cse_after_recomputation.cse          :     0.000010s :  0.08%
    auto_monad_reorder                   :     0.000017s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000481s :  3.76%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.33%


Time group info:
------[substitution.]   0.000260    23
 0.71% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000002s :      2: substitution.fold_const_symbol
 2.55% :     0.000007s :      3: substitution.graph_param_transform
85.79% :     0.000223s :      4: substitution.inline
 1.30% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.98% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000005s :      2: substitution.replace_old_param
 5.00% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006222     2
82.68% :     0.005145s :      1: type_inference.infer
17.32% :     0.001077s :      1: type_inference.specialize
------[replace.]   0.000076     6
67.77% :     0.000052s :      4: replace.inline
32.23% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000232     6
94.88% :     0.000220s :      4: match.inline
 5.12% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.30% :     0.000003s :     10: predicate.arithmetic_simplify
 1.07% :     0.000001s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.22% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000001s :     10: predicate.environ_get_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.45% :     0.000007s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.90% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.99% :     0.000003s :     15: predicate.load_eliminater
 2.13% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.02% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 0.95% :     0.000001s :     10: predicate.minmaximum_grad
 2.47% :     0.000003s :      3: predicate.mutable_eliminate
 0.62% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.82% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.32% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.26% :     0.000011s :     53: predicate.switch_simplify
 1.18% :     0.000002s :     10: predicate.tile_eliminate
 1.04% :     0.000001s :     10: predicate.transpose_eliminate
 1.21% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.24% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.89% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000759    12
48.49% :     0.000368s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.74% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.78% :     0.000340s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019522    72
 0.29% :     0.000057s :      1: add_recomputation
 0.38% :     0.000074s :      1: auto_monad
 0.10% :     0.000020s :      1: auto_monad_reorder
 2.25% :     0.000440s :      1: bootstrap
 0.16% :     0.000031s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.13% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000010s :      1: environ_conv
 0.13% :     0.000026s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
33.38% :     0.006517s :      1: jit_opt_a
 0.83% :     0.000162s :      1: jit_opt_after_cconv
 0.32% :     0.000063s :      1: jit_opt_b
11.07% :     0.002162s :      1: loop_unroll
 3.73% :     0.000728s :      1: mutable_eliminate
 4.81% :     0.000938s :     26: opt.transform.jit_opt_a
 0.27% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000032s :      4: opt.transform.jit_opt_b
 0.08% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.50% :     0.000489s :      1: opt_after_jit_grad
 0.04% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000047s :      1: py_interpret_to_execute
 0.11% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000020s :      1: remove_dup_value
 2.67% :     0.000520s :      1: renormalize.infer
 2.13% :     0.000417s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000043s :      1: rewriter_after_opt_a
 0.40% :     0.000079s :      1: rewriter_before_opt_a
 0.43% :     0.000083s :      1: symbol_engine_optimizer
32.40% :     0.006324s :      1: type_inference


TotalTime = 0.0249614, [33]
[bootstrap]: 0.00035164
[type_inference]: 0.00528102
[event_method]: 2.047e-05
[auto_monad]: 7.174e-05
[graph_reusing]: 5.91e-06
[pre_auto_parallel]: 2.75997e-06
[py_interpret_to_execute]: 4.211e-05
[rewriter_before_opt_a]: 7.532e-05
[expand_dump_flag]: 3.03e-06
[jit_opt_a]: 0.0070193, [2]
    [Cycle 1]: 0.00235674, [27]
        [switch_simplify]: 7.161e-05
        [loop_unroll]: 3.424e-05
        [a_1]: 0.00077515
        [with_stream_mark]: 3.077e-05
        [recompute_prepare]: 1.025e-05
        [updatestate_depend_eliminate]: 4.77e-06
        [updatestate_assign_eliminate]: 3.36999e-06
        [updatestate_loads_eliminate]: 3.08e-06
        [parameter_eliminate]: 2.98998e-06
        [specialize_transform]: 6.86999e-06
        [updatestate_useless_node_eliminater]: 6.44999e-06
        [accelerated_algorithm]: 6.36e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 6.07999e-06
        [merge_forward]: 3.63999e-06
        [cell_reuse_recompute_pass]: 1.62999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.867e-05
        [j_node_and_user_rematch]: 1.162e-05
        [meta_fg_expand]: 2.56998e-06
        [replace_old_param]: 1.143e-05
        [inline_without_move]: 6.94999e-06
        [renormalize]: 0.00103621
        [add_forward_monad_depend]: 8.14002e-06
        [auto_monad_grad]: 2.87002e-06
        [auto_monad_eliminator]: 2.009e-05
        [cse]: 3.085e-05
        [replace_applicator]: 2.117e-05
    [Cycle 2]: 0.0004124, [27]
        [switch_simplify]: 6.30002e-06
        [loop_unroll]: 5.99e-06
        [a_1]: 0.00010836
        [with_stream_mark]: 1.679e-05
        [recompute_prepare]: 6.21998e-06
        [updatestate_depend_eliminate]: 3.41999e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 5.93002e-06
        [updatestate_useless_node_eliminater]: 4.91997e-06
        [accelerated_algorithm]: 6.38003e-06
        [meta_shard_fg_expand]: 2.09e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 3.58999e-06
        [cell_reuse_recompute_pass]: 4.2e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.96e-05
        [j_node_and_user_rematch]: 1.079e-05
        [meta_fg_expand]: 2.64999e-06
        [replace_old_param]: 2.401e-05
        [inline_without_move]: 5.46e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.93997e-06
        [auto_monad_grad]: 1.17e-06
        [auto_monad_eliminator]: 8.77999e-06
        [cse]: 1.348e-05
        [replace_applicator]: 7.31001e-06
[py_interpret_to_execute_after_opt_a]: 2.031e-05
[rewriter_after_opt_a]: 4.228e-05
[convert_after_rewriter]: 9.19e-06
[order_py_execute_after_rewriter]: 6.14001e-06
[mutable_eliminate]: 0.00086533
[jit_opt_b]: 5.95e-05, [1]
    [Cycle 1]: 4.96e-05, [2]
        [frontend_op_eliminate]: 1.957e-05
        [inline_after_opt_a]: 1.738e-05
[cconv]: 3.297e-05
[loop_unroll]: 0.00057424
[jit_opt_after_cconv]: 0.00017962, [1]
    [Cycle 1]: 0.0001713, [11]
        [c_1]: 2.551e-05
        [parameter_eliminate]: 4.35e-06
        [updatestate_depend_eliminate]: 7.33e-06
        [updatestate_assign_eliminate]: 3.33e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [cse]: 3.183e-05
        [call_graph_tuple_transform]: 2.536e-05
        [tuple_list_get_item_eliminator]: 6.06e-06
        [none_parameter_eliminate]: 1.30001e-06
        [renormalize]: 9.20001e-07
        [switch_simplify]: 5.71e-06
[remove_dup_value]: 1.628e-05
[partial_unused_args_eliminate]: 3.11001e-06
[environ_conv]: 6.28002e-06
[add_recomputation]: 6.9e-05
[cse_after_recomputation]: 3.18e-05, [1]
    [Cycle 1]: 2.28e-05, [1]
        [cse]: 1.218e-05
[auto_monad_reorder]: 1.767e-05
[get_jit_bprop_graph]: 2.74001e-06
[rewriter_after_jit_bprop_graph]: 1.122e-05
[opt_after_jit_grad]: 0.00066832
[symbol_engine_optimizer]: 9.324e-05, [1]
    [Cycle 1]: 8.35e-05, [6]
        [build]: 4.87e-06
        [elim_shapecalc]: 9.69999e-06
        [elim_not_effective]: 1.675e-05
        [opt_reshape]: 7.46999e-06
        [fold_const_symbol]: 1.189e-05
        [renormalize]: 6.49976e-07
[validate]: 4.916e-05
[backend_pass]: 1.82999e-06
[task_emit]: 0.00905298
[execute]: 9.25999e-06
Sums
    bootstrap                            :     0.000352s :  1.76%
    type_inference                       :     0.005281s : 26.47%
    event_method                         :     0.000020s :  0.10%
    auto_monad                           :     0.000072s :  0.36%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.21%
    rewriter_before_opt_a                :     0.000075s :  0.38%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000078s :  0.39%
    jit_opt_a.loop_unroll                :     0.000040s :  0.20%
    jit_opt_a.a_1                        :     0.000884s :  4.43%
    jit_opt_a.with_stream_mark           :     0.000048s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000022s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000035s :  0.18%
    jit_opt_a.inline_without_move        :     0.000012s :  0.06%
    jit_opt_a.renormalize                :     0.001036s :  5.19%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000029s :  0.14%
    jit_opt_a.cse                        :     0.000044s :  0.22%
    jit_opt_a.replace_applicator         :     0.000028s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000020s :  0.10%
    rewriter_after_opt_a                 :     0.000042s :  0.21%
    convert_after_rewriter               :     0.000009s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000865s :  4.34%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000033s :  0.17%
    loop_unroll                          :     0.000574s :  2.88%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000032s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000069s :  0.35%
    cse_after_recomputation.cse          :     0.000012s :  0.06%
    auto_monad_reorder                   :     0.000018s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.06%
    opt_after_jit_grad                   :     0.000668s :  3.35%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000049s :  0.25%
    backend_pass                         :     0.000002s :  0.01%
    task_emit                            :     0.009053s : 45.38%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000346    23
 0.80% :     0.000003s :      2: substitution.elim_not_effective
 0.82% :     0.000003s :      2: substitution.fold_const_symbol
 1.95% :     0.000007s :      3: substitution.graph_param_transform
69.94% :     0.000242s :      4: substitution.inline
 1.67% :     0.000006s :      4: substitution.j_node_and_user_rematch
 1.49% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.50% :     0.000005s :      2: substitution.replace_old_param
21.84% :     0.000076s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005200     2
80.90% :     0.004207s :      1: type_inference.infer
19.10% :     0.000994s :      1: type_inference.specialize
------[replace.]   0.000109     6
50.97% :     0.000056s :      4: replace.inline
49.03% :     0.000053s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000312     6
76.47% :     0.000239s :      4: match.inline
23.53% :     0.000073s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000155   691
 1.51% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.53% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.84% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 2.38% :     0.000004s :     10: predicate.arithmetic_simplify
 1.23% :     0.000002s :     10: predicate.cast_eliminate
 0.29% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.79% :     0.000001s :     10: predicate.compare_switch_simplify
 1.16% :     0.000002s :     10: predicate.depend_value_elim
 0.90% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.01% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.06% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000002s :     10: predicate.environ_add_const_eliminate
 8.24% :     0.000013s :     10: predicate.environ_get_add_eliminate
 0.79% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000002s :     10: predicate.environ_get_eliminate
 1.07% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000001s :      3: predicate.graph_param_transform
 5.88% :     0.000009s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.45% :     0.000002s :      6: predicate.less_batch_normalization
 1.16% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.43% :     0.000002s :     15: predicate.load_eliminater
 1.71% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.50% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.64% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 0.88% :     0.000001s :     10: predicate.merge_addn
 0.86% :     0.000001s :     10: predicate.minmaximum_grad
 1.49% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 1.90% :     0.000003s :     15: predicate.partial_eliminate
 0.89% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.80% :     0.000003s :     10: predicate.reduce_eliminate
 1.12% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.38% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.20% :     0.000000s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 0.92% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000002s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 1.20% :     0.000002s :      6: predicate.specialize_transform
 1.86% :     0.000003s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.34% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.07% :     0.000003s :     16: predicate.switch_defer_inline
 1.68% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.29% :     0.000010s :     53: predicate.switch_simplify
 1.53% :     0.000002s :     10: predicate.tile_eliminate
 0.92% :     0.000001s :     10: predicate.transpose_eliminate
 1.20% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.94% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.89% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.28% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000005s :     21: predicate.updatestate_useless_node_eliminater
 1.41% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000783    12
50.08% :     0.000392s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.12% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.80% :     0.000335s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027293    76
 0.28% :     0.000076s :      1: add_recomputation
 0.27% :     0.000075s :      1: auto_monad
 0.08% :     0.000022s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 1.37% :     0.000375s :      1: bootstrap
 0.13% :     0.000036s :      1: cconv
 0.04% :     0.000012s :      1: convert_after_rewriter
 0.13% :     0.000035s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.09% :     0.000025s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
25.74% :     0.007024s :      1: jit_opt_a
 0.67% :     0.000183s :      1: jit_opt_after_cconv
 0.23% :     0.000063s :      1: jit_opt_b
 2.14% :     0.000585s :      1: loop_unroll
 3.22% :     0.000880s :      1: mutable_eliminate
 4.27% :     0.001165s :     26: opt.transform.jit_opt_a
 0.21% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000029s :      4: opt.transform.jit_opt_b
 0.07% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000042s :      4: opt.transform.symbol_engine_opt
 2.50% :     0.000681s :      1: opt_after_jit_grad
 0.03% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000045s :      1: py_interpret_to_execute
 0.09% :     0.000024s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 2.00% :     0.000547s :      1: renormalize.infer
 1.75% :     0.000478s :      1: renormalize.specialize
 0.05% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000047s :      1: rewriter_after_opt_a
 0.29% :     0.000078s :      1: rewriter_before_opt_a
 0.35% :     0.000097s :      1: symbol_engine_optimizer
33.23% :     0.009070s :      1: task_emit
19.41% :     0.005298s :      1: type_inference
 0.27% :     0.000074s :      1: validate


TotalTime = 0.0283611, [33]
[bootstrap]: 0.00039422
[type_inference]: 0.00673522
[event_method]: 2.025e-05
[auto_monad]: 7.195e-05
[graph_reusing]: 5.89999e-06
[pre_auto_parallel]: 3.48999e-06
[py_interpret_to_execute]: 4.454e-05
[rewriter_before_opt_a]: 7.75e-05
[expand_dump_flag]: 3.26001e-06
[jit_opt_a]: 0.00733244, [2]
    [Cycle 1]: 0.00203255, [27]
        [switch_simplify]: 6.868e-05
        [loop_unroll]: 2.942e-05
        [a_1]: 0.00063764
        [with_stream_mark]: 2.52e-05
        [recompute_prepare]: 1.106e-05
        [updatestate_depend_eliminate]: 4.45999e-06
        [updatestate_assign_eliminate]: 3.16001e-06
        [updatestate_loads_eliminate]: 2.89001e-06
        [parameter_eliminate]: 2.44999e-06
        [specialize_transform]: 6.69999e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.99e-06
        [meta_shard_fg_expand]: 2.22999e-06
        [get_grad_eliminate_]: 5.80002e-06
        [merge_forward]: 3.55e-06
        [cell_reuse_recompute_pass]: 1.92001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.738e-05
        [j_node_and_user_rematch]: 1.098e-05
        [meta_fg_expand]: 3.28e-06
        [replace_old_param]: 1.073e-05
        [inline_without_move]: 5.53997e-06
        [renormalize]: 0.00088868
        [add_forward_monad_depend]: 7.65e-06
        [auto_monad_grad]: 3.28e-06
        [auto_monad_eliminator]: 1.875e-05
        [cse]: 2.998e-05
        [replace_applicator]: 1.746e-05
    [Cycle 2]: 0.00034273, [27]
        [switch_simplify]: 6.30002e-06
        [loop_unroll]: 5.61998e-06
        [a_1]: 9.422e-05
        [with_stream_mark]: 1.446e-05
        [recompute_prepare]: 5.20999e-06
        [updatestate_depend_eliminate]: 3.06001e-06
        [updatestate_assign_eliminate]: 2.64001e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [parameter_eliminate]: 1.18001e-06
        [specialize_transform]: 5.22999e-06
        [updatestate_useless_node_eliminater]: 4.63999e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 1.54e-06
        [get_grad_eliminate_]: 4.85999e-06
        [merge_forward]: 3.15998e-06
        [cell_reuse_recompute_pass]: 2.66999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.544e-05
        [j_node_and_user_rematch]: 7.87e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 9.81e-06
        [inline_without_move]: 4.85001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.61002e-06
        [auto_monad_grad]: 9.10019e-07
        [auto_monad_eliminator]: 5.97999e-06
        [cse]: 1.075e-05
        [replace_applicator]: 1.09e-05
[py_interpret_to_execute_after_opt_a]: 1.618e-05
[rewriter_after_opt_a]: 3.975e-05
[convert_after_rewriter]: 7.71999e-06
[order_py_execute_after_rewriter]: 5.12999e-06
[mutable_eliminate]: 0.00408726
[jit_opt_b]: 7.893e-05, [1]
    [Cycle 1]: 6.258e-05, [2]
        [frontend_op_eliminate]: 2.374e-05
        [inline_after_opt_a]: 2.388e-05
[cconv]: 4.83e-05
[loop_unroll]: 0.0010221
[jit_opt_after_cconv]: 0.00019445, [1]
    [Cycle 1]: 0.00018571, [11]
        [c_1]: 2.401e-05
        [parameter_eliminate]: 7.97e-06
        [updatestate_depend_eliminate]: 1.295e-05
        [updatestate_assign_eliminate]: 3.14001e-06
        [updatestate_loads_eliminate]: 3.26001e-06
        [cse]: 4e-05
        [call_graph_tuple_transform]: 2.773e-05
        [tuple_list_get_item_eliminator]: 6.56e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 1.02998e-06
        [switch_simplify]: 5.27001e-06
[remove_dup_value]: 1.729e-05
[partial_unused_args_eliminate]: 2.22999e-06
[environ_conv]: 6.81001e-06
[add_recomputation]: 6.638e-05
[cse_after_recomputation]: 2.435e-05, [1]
    [Cycle 1]: 1.824e-05, [1]
        [cse]: 1.177e-05
[auto_monad_reorder]: 1.722e-05
[get_jit_bprop_graph]: 2.52001e-06
[rewriter_after_jit_bprop_graph]: 1.137e-05
[opt_after_jit_grad]: 0.00061245
[symbol_engine_optimizer]: 8.235e-05, [1]
    [Cycle 1]: 7.437e-05, [6]
        [build]: 3.95e-06
        [elim_shapecalc]: 8.15e-06
        [elim_not_effective]: 1.599e-05
        [opt_reshape]: 5.65001e-06
        [fold_const_symbol]: 8.70001e-06
        [renormalize]: 6.10016e-07
[validate]: 5.062e-05
[backend_pass]: 1.04003e-06
[task_emit]: 0.006994
[execute]: 1.02e-05
Sums
    bootstrap                            :     0.000394s :  1.74%
    type_inference                       :     0.006735s : 29.71%
    event_method                         :     0.000020s :  0.09%
    auto_monad                           :     0.000072s :  0.32%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000045s :  0.20%
    rewriter_before_opt_a                :     0.000078s :  0.34%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000075s :  0.33%
    jit_opt_a.loop_unroll                :     0.000035s :  0.15%
    jit_opt_a.a_1                        :     0.000732s :  3.23%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000889s :  3.92%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.11%
    jit_opt_a.cse                        :     0.000041s :  0.18%
    jit_opt_a.replace_applicator         :     0.000028s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.07%
    rewriter_after_opt_a                 :     0.000040s :  0.18%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.004087s : 18.03%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.11%
    cconv                                :     0.000048s :  0.21%
    loop_unroll                          :     0.001022s :  4.51%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000013s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000040s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000017s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000066s :  0.29%
    cse_after_recomputation.cse          :     0.000012s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.08%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.05%
    opt_after_jit_grad                   :     0.000612s :  2.70%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000051s :  0.22%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006994s : 30.85%
    execute                              :     0.000010s :  0.04%


Time group info:
------[substitution.]   0.000283    23
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.48% :     0.000007s :      3: substitution.graph_param_transform
84.77% :     0.000240s :      4: substitution.inline
 1.61% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.92% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.58% :     0.000007s :      2: substitution.replace_old_param
 5.42% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006647     2
85.06% :     0.005654s :      1: type_inference.infer
14.94% :     0.000993s :      1: type_inference.specialize
------[replace.]   0.000079     6
68.53% :     0.000054s :      4: replace.inline
31.47% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000251     6
94.45% :     0.000237s :      4: match.inline
 5.55% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000141   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.39% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.85% :     0.000001s :     10: predicate.addn_check_dump
 1.52% :     0.000002s :     10: predicate.addn_zero_filter
 2.11% :     0.000003s :     10: predicate.arithmetic_simplify
 1.21% :     0.000002s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.85% :     0.000001s :     10: predicate.compare_switch_simplify
 1.47% :     0.000002s :     10: predicate.depend_value_elim
 0.92% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.15% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.39% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.24% :     0.000000s :      3: predicate.elim_not_effective
 0.41% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.86% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.02% :     0.000001s :     10: predicate.environ_get_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.16% :     0.000002s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 6.43% :     0.000009s :     22: predicate.inline
 0.72% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.43% :     0.000002s :     15: predicate.load_eliminater
 3.05% :     0.000004s :      3: predicate.loop_unroll_after_grad
 3.34% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.33% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.89% :     0.000001s :     10: predicate.merge_addn
 0.88% :     0.000001s :     10: predicate.minmaximum_grad
 5.06% :     0.000007s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 1.84% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.39% :     0.000002s :     10: predicate.reduce_eliminate
 1.33% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.89% :     0.000001s :      3: predicate.reset_defer_inline
 1.13% :     0.000002s :     10: predicate.reshape_eliminate
 1.04% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.47% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.62% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 0.84% :     0.000001s :      6: predicate.specialize_transform
 1.14% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.03% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.02% :     0.000010s :     53: predicate.switch_simplify
 1.08% :     0.000002s :     10: predicate.tile_eliminate
 1.12% :     0.000002s :     10: predicate.transpose_eliminate
 1.15% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.05% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.33% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.35% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.34% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.52% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.45% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.96% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000719    12
48.41% :     0.000348s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.63% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.96% :     0.000316s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.030391    76
 0.23% :     0.000070s :      1: add_recomputation
 0.25% :     0.000076s :      1: auto_monad
 0.06% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 1.37% :     0.000415s :      1: bootstrap
 0.17% :     0.000051s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.09% :     0.000026s :      1: event_method
 0.05% :     0.000017s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
24.14% :     0.007336s :      1: jit_opt_a
 0.65% :     0.000197s :      1: jit_opt_after_cconv
 0.28% :     0.000084s :      1: jit_opt_b
 3.41% :     0.001036s :      1: loop_unroll
13.55% :     0.004118s :      1: mutable_eliminate
 3.22% :     0.000978s :     26: opt.transform.jit_opt_a
 0.20% :     0.000060s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000036s :      4: opt.transform.jit_opt_b
 0.07% :     0.000020s :      1: opt.transform.loop_unroll_optimizer
 0.14% :     0.000043s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.06% :     0.000625s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000047s :      1: py_interpret_to_execute
 0.06% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000020s :      1: remove_dup_value
 1.66% :     0.000505s :      1: renormalize.infer
 1.23% :     0.000375s :      1: renormalize.specialize
 0.04% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000043s :      1: rewriter_after_opt_a
 0.27% :     0.000081s :      1: rewriter_before_opt_a
 0.28% :     0.000085s :      1: symbol_engine_optimizer
23.09% :     0.007016s :      1: task_emit
22.23% :     0.006757s :      1: type_inference
 0.25% :     0.000076s :      1: validate


TotalTime = 0.0786328, [33]
[bootstrap]: 0.00034526
[type_inference]: 0.0289864
[event_method]: 2.525e-05
[auto_monad]: 7.27e-05
[graph_reusing]: 6.59001e-06
[pre_auto_parallel]: 3.36999e-06
[py_interpret_to_execute]: 4.369e-05
[rewriter_before_opt_a]: 7.658e-05
[expand_dump_flag]: 3.27997e-06
[jit_opt_a]: 0.0198156, [2]
    [Cycle 1]: 0.0077534, [27]
        [switch_simplify]: 7.005e-05
        [loop_unroll]: 2.981e-05
        [a_1]: 0.00061893
        [with_stream_mark]: 7.941e-05
        [recompute_prepare]: 1.113e-05
        [updatestate_depend_eliminate]: 4.90999e-06
        [updatestate_assign_eliminate]: 3.21001e-06
        [updatestate_loads_eliminate]: 2.58998e-06
        [parameter_eliminate]: 2.25002e-06
        [specialize_transform]: 6.39999e-06
        [updatestate_useless_node_eliminater]: 5.10001e-06
        [accelerated_algorithm]: 5.30999e-06
        [meta_shard_fg_expand]: 2.14999e-06
        [get_grad_eliminate_]: 4.82e-06
        [merge_forward]: 3.93999e-06
        [cell_reuse_recompute_pass]: 1.71e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.602e-05
        [j_node_and_user_rematch]: 9.58002e-06
        [meta_fg_expand]: 2.22999e-06
        [replace_old_param]: 1.022e-05
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.00650019
        [add_forward_monad_depend]: 1.43e-05
        [auto_monad_grad]: 2.22001e-06
        [auto_monad_eliminator]: 2.6e-05
        [cse]: 3.507e-05
        [replace_applicator]: 2.877e-05
    [Cycle 2]: 0.00128418, [27]
        [switch_simplify]: 6.63998e-06
        [loop_unroll]: 1.497e-05
        [a_1]: 0.00010647
        [with_stream_mark]: 2.685e-05
        [recompute_prepare]: 5.74e-06
        [updatestate_depend_eliminate]: 3.63e-06
        [updatestate_assign_eliminate]: 2.84999e-06
        [updatestate_loads_eliminate]: 2.74999e-06
        [parameter_eliminate]: 2.21e-06
        [specialize_transform]: 5.37999e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.28002e-06
        [meta_shard_fg_expand]: 2.32999e-06
        [get_grad_eliminate_]: 5.07e-06
        [merge_forward]: 4.05e-06
        [cell_reuse_recompute_pass]: 3.68999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.149e-05
        [j_node_and_user_rematch]: 9.44998e-06
        [meta_fg_expand]: 2.41e-06
        [replace_old_param]: 1.018e-05
        [inline_without_move]: 4.91002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.43e-06
        [auto_monad_grad]: 1.49e-06
        [auto_monad_eliminator]: 9.42999e-06
        [cse]: 2.04e-05
        [replace_applicator]: 5.71e-06
[py_interpret_to_execute_after_opt_a]: 2.264e-05
[rewriter_after_opt_a]: 4.79e-05
[convert_after_rewriter]: 7.63001e-06
[order_py_execute_after_rewriter]: 4.97e-06
[mutable_eliminate]: 0.000888
[jit_opt_b]: 5.686e-05, [1]
    [Cycle 1]: 4.621e-05, [2]
        [frontend_op_eliminate]: 1.717e-05
        [inline_after_opt_a]: 1.605e-05
[cconv]: 3.761e-05
[loop_unroll]: 0.00046542
[jit_opt_after_cconv]: 0.00015662, [1]
    [Cycle 1]: 0.00014935, [11]
        [c_1]: 2.244e-05
        [parameter_eliminate]: 3.21001e-06
        [updatestate_depend_eliminate]: 8.23999e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [cse]: 2.492e-05
        [call_graph_tuple_transform]: 2.17e-05
        [tuple_list_get_item_eliminator]: 5.53002e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 5.90022e-07
        [switch_simplify]: 5.44e-06
[remove_dup_value]: 1.495e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 5.86e-06
[add_recomputation]: 6.072e-05
[cse_after_recomputation]: 2.14e-05, [1]
    [Cycle 1]: 1.577e-05, [1]
        [cse]: 9.87999e-06
[auto_monad_reorder]: 1.541e-05
[get_jit_bprop_graph]: 2.24999e-06
[rewriter_after_jit_bprop_graph]: 9.10001e-06
[opt_after_jit_grad]: 0.00049066
[symbol_engine_optimizer]: 7.681e-05, [1]
    [Cycle 1]: 7.049e-05, [6]
        [build]: 5.64e-06
        [elim_shapecalc]: 7.42002e-06
        [elim_not_effective]: 1.371e-05
        [opt_reshape]: 5.92001e-06
        [fold_const_symbol]: 8.58001e-06
        [renormalize]: 5.69999e-07
[validate]: 4.377e-05
[backend_pass]: 8.49977e-07
[task_emit]: 0.026559
[execute]: 1.042e-05
Sums
    bootstrap                            :     0.000345s :  0.52%
    type_inference                       :     0.028986s : 43.77%
    event_method                         :     0.000025s :  0.04%
    auto_monad                           :     0.000073s :  0.11%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.07%
    rewriter_before_opt_a                :     0.000077s :  0.12%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000077s :  0.12%
    jit_opt_a.loop_unroll                :     0.000045s :  0.07%
    jit_opt_a.a_1                        :     0.000725s :  1.10%
    jit_opt_a.with_stream_mark           :     0.000106s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.01%
    jit_opt_a.merge_forward              :     0.000008s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000020s :  0.03%
    jit_opt_a.inline_without_move        :     0.000010s :  0.02%
    jit_opt_a.renormalize                :     0.006500s :  9.81%
    jit_opt_a.add_forward_monad_depend   :     0.000017s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000035s :  0.05%
    jit_opt_a.cse                        :     0.000055s :  0.08%
    jit_opt_a.replace_applicator         :     0.000034s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000023s :  0.03%
    rewriter_after_opt_a                 :     0.000048s :  0.07%
    convert_after_rewriter               :     0.000008s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000888s :  1.34%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.02%
    cconv                                :     0.000038s :  0.06%
    loop_unroll                          :     0.000465s :  0.70%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.03%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000025s :  0.04%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.01%
    remove_dup_value                     :     0.000015s :  0.02%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000006s :  0.01%
    add_recomputation                    :     0.000061s :  0.09%
    cse_after_recomputation.cse          :     0.000010s :  0.01%
    auto_monad_reorder                   :     0.000015s :  0.02%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.01%
    opt_after_jit_grad                   :     0.000491s :  0.74%
    symbol_engine_optimizer.build        :     0.000006s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.02%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.07%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.026559s : 40.10%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000280    23
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 2.08% :     0.000006s :      3: substitution.graph_param_transform
85.76% :     0.000240s :      4: substitution.inline
 1.70% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.92% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.79% :     0.000008s :      2: substitution.replace_old_param
 4.60% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.028826     2
92.77% :     0.026742s :      1: type_inference.infer
 7.23% :     0.002084s :      1: type_inference.specialize
------[replace.]   0.000079     6
69.28% :     0.000055s :      4: replace.inline
30.72% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000249     6
95.28% :     0.000237s :      4: match.inline
 4.72% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.25% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.92% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.87% :     0.000004s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.33% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.56% :     0.000002s :     10: predicate.depend_value_elim
 0.96% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.20% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.45% :     0.000001s :      3: predicate.elim_not_effective
 0.43% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.46% :     0.000002s :     10: predicate.environ_get_eliminate
 1.24% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 7.50% :     0.000010s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.62% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.73% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.63% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 0.91% :     0.000001s :     10: predicate.minmaximum_grad
 2.24% :     0.000003s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.84% :     0.000002s :     18: predicate.replace_applicator
 0.91% :     0.000001s :      6: predicate.replace_old_param
 0.40% :     0.000001s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.16% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.27% :     0.000003s :     16: predicate.switch_defer_inline
 1.99% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.22% :     0.000010s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.38% :     0.000002s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.20% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.43% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.43% :     0.000002s :     10: predicate.value_based_eliminate
 0.27% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.52% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000784    12
41.31% :     0.000324s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.54% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.15% :     0.000401s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.086219    76
 0.07% :     0.000064s :      1: add_recomputation
 0.09% :     0.000077s :      1: auto_monad
 0.02% :     0.000018s :      1: auto_monad_reorder
 0.00% :     0.000003s :      1: backend_pass
 0.43% :     0.000367s :      1: bootstrap
 0.05% :     0.000041s :      1: cconv
 0.01% :     0.000010s :      1: convert_after_rewriter
 0.03% :     0.000024s :      1: cse_after_recomputation
 0.01% :     0.000008s :      1: environ_conv
 0.04% :     0.000030s :      1: event_method
 0.02% :     0.000016s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000009s :      1: graph_reusing
22.99% :     0.019819s :      1: jit_opt_a
 0.18% :     0.000159s :      1: jit_opt_after_cconv
 0.07% :     0.000060s :      1: jit_opt_b
 0.55% :     0.000476s :      1: loop_unroll
 1.05% :     0.000905s :      1: mutable_eliminate
 1.15% :     0.000988s :     26: opt.transform.jit_opt_a
 0.06% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.03% :     0.000026s :      4: opt.transform.jit_opt_b
 0.01% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.04% :     0.000033s :      4: opt.transform.symbol_engine_opt
 0.58% :     0.000500s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.05% :     0.000046s :      1: py_interpret_to_execute
 0.03% :     0.000026s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000017s :      1: remove_dup_value
 6.94% :     0.005982s :      1: renormalize.infer
 0.58% :     0.000498s :      1: renormalize.specialize
 0.01% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.06% :     0.000052s :      1: rewriter_after_opt_a
 0.09% :     0.000080s :      1: rewriter_before_opt_a
 0.09% :     0.000079s :      1: symbol_engine_optimizer
30.83% :     0.026579s :      1: task_emit
33.65% :     0.029014s :      1: type_inference
 0.08% :     0.000069s :      1: validate


TotalTime = 0.0278414, [33]
[bootstrap]: 0.00046151
[type_inference]: 0.00722473
[event_method]: 1.928e-05
[auto_monad]: 6.765e-05
[graph_reusing]: 6.19999e-06
[pre_auto_parallel]: 3.06001e-06
[py_interpret_to_execute]: 4.031e-05
[rewriter_before_opt_a]: 7.779e-05
[expand_dump_flag]: 3.17002e-06
[jit_opt_a]: 0.00960009, [2]
    [Cycle 1]: 0.00192136, [27]
        [switch_simplify]: 6.494e-05
        [loop_unroll]: 3.038e-05
        [a_1]: 0.00064738
        [with_stream_mark]: 2.653e-05
        [recompute_prepare]: 8.32998e-06
        [updatestate_depend_eliminate]: 3.55e-06
        [updatestate_assign_eliminate]: 3.14999e-06
        [updatestate_loads_eliminate]: 2.84001e-06
        [parameter_eliminate]: 1.99999e-06
        [specialize_transform]: 5.80002e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 5.10999e-06
        [meta_shard_fg_expand]: 2.14999e-06
        [get_grad_eliminate_]: 5.20001e-06
        [merge_forward]: 3.93999e-06
        [cell_reuse_recompute_pass]: 1.94e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.618e-05
        [j_node_and_user_rematch]: 9.76998e-06
        [meta_fg_expand]: 2.23002e-06
        [replace_old_param]: 9.19e-06
        [inline_without_move]: 5.01002e-06
        [renormalize]: 0.00079383
        [add_forward_monad_depend]: 6.57002e-06
        [auto_monad_grad]: 2.89001e-06
        [auto_monad_eliminator]: 1.657e-05
        [cse]: 3.136e-05
        [replace_applicator]: 1.35e-05
    [Cycle 2]: 0.00033287, [27]
        [switch_simplify]: 5.92001e-06
        [loop_unroll]: 4.68999e-06
        [a_1]: 9.401e-05
        [with_stream_mark]: 1.082e-05
        [recompute_prepare]: 5.00999e-06
        [updatestate_depend_eliminate]: 2.34001e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 1.28002e-06
        [specialize_transform]: 5.64e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 4.70001e-06
        [meta_shard_fg_expand]: 1.35999e-06
        [get_grad_eliminate_]: 4.53001e-06
        [merge_forward]: 2.79999e-06
        [cell_reuse_recompute_pass]: 2.90998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.315e-05
        [j_node_and_user_rematch]: 8.1e-06
        [meta_fg_expand]: 1.51998e-06
        [replace_old_param]: 8.13001e-06
        [inline_without_move]: 5.32999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.47999e-06
        [auto_monad_grad]: 1.02998e-06
        [auto_monad_eliminator]: 5.33002e-06
        [cse]: 1.151e-05
        [replace_applicator]: 5.44e-06
[py_interpret_to_execute_after_opt_a]: 1.427e-05
[rewriter_after_opt_a]: 4.135e-05
[convert_after_rewriter]: 8.09997e-06
[order_py_execute_after_rewriter]: 5.79999e-06
[mutable_eliminate]: 0.00083204
[jit_opt_b]: 6.198e-05, [1]
    [Cycle 1]: 5.213e-05, [2]
        [frontend_op_eliminate]: 1.863e-05
        [inline_after_opt_a]: 2.1e-05
[cconv]: 3.2e-05
[loop_unroll]: 0.0006352
[jit_opt_after_cconv]: 0.00017149, [1]
    [Cycle 1]: 0.00016325, [11]
        [c_1]: 2.285e-05
        [parameter_eliminate]: 5.00001e-06
        [updatestate_depend_eliminate]: 8.54e-06
        [updatestate_assign_eliminate]: 3.18998e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [cse]: 2.687e-05
        [call_graph_tuple_transform]: 2.325e-05
        [tuple_list_get_item_eliminator]: 6.54999e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 8.70001e-07
        [switch_simplify]: 5.59e-06
[remove_dup_value]: 1.582e-05
[partial_unused_args_eliminate]: 2.86999e-06
[environ_conv]: 6.56e-06
[add_recomputation]: 6.209e-05
[cse_after_recomputation]: 2.403e-05, [1]
    [Cycle 1]: 1.756e-05, [1]
        [cse]: 1.168e-05
[auto_monad_reorder]: 1.715e-05
[get_jit_bprop_graph]: 2.37001e-06
[rewriter_after_jit_bprop_graph]: 9.77999e-06
[opt_after_jit_grad]: 0.0006285
[symbol_engine_optimizer]: 7.885e-05, [1]
    [Cycle 1]: 7.095e-05, [6]
        [build]: 5.46e-06
        [elim_shapecalc]: 7
TotalTime = 0.0592813, [33]
[bootstrap]: 0.00034494
[type_inference]: 0.0080919
[event_method]: 2.356e-05
[auto_monad]: 7.972e-05
[graph_reusing]: 6.22001e-06
[pre_auto_parallel]: 3.01001e-06
[py_interpret_to_execute]: 5.691e-05
[rewriter_before_opt_a]: 8.734e-05
[expand_dump_flag]: 4.87e-06
[jit_opt_a]: 0.00750908, [2]
    [Cycle 1]: 0.00201521, [27]
        [switch_simplify]: 7.013e-05
        [loop_unroll]: 2.965e-05
        [a_1]: 0.00065391
        [with_stream_mark]: 2.521e-05
        [recompute_prepare]: 7.45e-06
        [updatestate_depend_eliminate]: 4.08999e-06
        [updatestate_assign_eliminate]: 2.98998e-06
        [updatestate_loads_eliminate]: 2.67001e-06
        [parameter_eliminate]: 2.35002e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 5.25999e-06
        [accelerated_algorithm]: 5.56e-06
        [meta_shard_fg_expand]: 2.67001e-06
        [get_grad_eliminate_]: 5.29998e-06
        [merge_forward]: 3.46999e-06
        [cell_reuse_recompute_pass]: 1.26002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.527e-05
        [j_node_and_user_rematch]: 8.40999e-06
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 9.46e-06
        [inline_without_move]: 5.39e-06
        [renormalize]: 0.00087305
        [add_forward_monad_depend]: 7.11999e-06
        [auto_monad_grad]: 2.83e-06
        [auto_monad_eliminator]: 1.759e-05
        [cse]: 3.218e-05
        [replace_applicator]: 1.617e-05
    [Cycle 2]: 0.00033032, [27]
        [switch_simplify]: 6.54001e-06
        [loop_unroll]: 4.84003e-06
        [a_1]: 8.974e-05
        [with_stream_mark]: 1.343e-05
        [recompute_prepare]: 4.81002e-06
        [updatestate_depend_eliminate]: 3.30998e-06
        [updatestate_assign_eliminate]: 2.23002e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [parameter_eliminate]: 1.45001e-06
        [specialize_transform]: 4.63001e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.10001e-06
        [meta_shard_fg_expand]: 1.61998e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.63e-06
        [cell_reuse_recompute_pass]: 1.96998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.549e-05
        [j_node_and_user_rematch]: 8.27e-06
        [meta_fg_expand]: 1.87999e-06
        [replace_old_param]: 7.4e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 1.42e-06
        [auto_monad_eliminator]: 6.04001e-06
        [cse]: 1.102e-05
        [replace_applicator]: 4.83001e-06
[py_interpret_to_execute_after_opt_a]: 1.356e-05
[rewriter_after_opt_a]: 3.99e-05
[convert_after_rewriter]: 6.79999e-06
[order_py_execute_after_rewriter]: 4.81002e-06
[mutable_eliminate]: 0.00075189
[jit_opt_b]: 5.068e-05, [1]
    [Cycle 1]: 4.206e-05, [2]
        [frontend_op_eliminate]: 1.638e-05
        [inline_after_opt_a]: 1.408e-05
[cconv]: 3.118e-05
[loop_unroll]: 0.00105401
[jit_opt_after_cconv]: 0.00021114, [1]
    [Cycle 1]: 0.00020163, [11]
        [c_1]: 2.322e-05
        [parameter_eliminate]: 5.32001e-06
        [updatestate_depend_eliminate]: 9.86e-06
        [updatestate_assign_eliminate]: 3.14999e-06
        [updatestate_loads_eliminate]: 2.03002e-06
        [cse]: 3.619e-05
        [call_graph_tuple_transform]: 2.694e-05
        [tuple_list_get_item_eliminator]: 6.49999e-06
        [none_parameter_eliminate]: 1.77999e-06
        [renormalize]: 8.59989e-07
        [switch_simplify]: 6.46e-06
[remove_dup_value]: 1.63e-05
[partial_unused_args_eliminate]: 2.66999e-06
[environ_conv]: 6.85002e-06
[add_recomputation]: 5.964e-05
[cse_after_recomputation]: 2.194e-05, [1]
    [Cycle 1]: 1.553e-05, [1]
        [cse]: 8.79e-06
[auto_monad_reorder]: 1.628e-05
[get_jit_bprop_graph]: 2.71e-06
[rewriter_after_jit_bprop_graph]: 1.084e-05
[opt_after_jit_grad]: 0.00064917
[symbol_engine_optimizer]: 8.12e-05, [1]
    [Cycle 1]: 7.361e-05, [6]
        [build]: 5.19998e-06
        [elim_shapecalc]: 8.50001.28999e-06
        [elim_not_effective]: 1.523e-05
        [opt_reshape]: 5.59e-06
        [fold_const_symbol]: 8.67e-06
        [renormalize]: 1.58002e-06
[validate]: 4.315e-05
[backend_pass]: 9.40025e-07
[task_emit]: 0.00738021
[execute]: 1.019e-05
Sums
    bootstrap                            :     0.000462s :  2.33%
    type_inference                       :     0.007225s : 36.50%
    event_method                         :     0.000019s :  0.10%
    auto_monad                           :     0.000068s :  0.34%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.20%
    rewriter_before_opt_a                :     0.000078s :  0.39%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000071s :  0.36%
    jit_opt_a.loop_unroll                :     0.000035s :  0.18%
    jit_opt_a.a_1                        :     0.000741s :  3.75%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000794s :  4.01%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.11%
    jit_opt_a.cse                        :     0.000043s :  0.22%
    jit_opt_a.replace_applicator         :     0.000019s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.07%
    rewriter_after_opt_a                 :     0.000041s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000832s :  4.20%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000021s :  0.11%
    cconv                                :     0.000032s :  0.16%
    loop_unroll                          :     0.000635s :  3.21%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :   e-06
        [elim_not_effective]: 1.565e-05
        [opt_reshape]: 6.01e-06
        [fold_const_symbol]: 9.04003e-06
        [renormalize]: 1.02998e-06
[validate]: 4.993e-05
[backend_pass]: 1.06002e-06
[task_emit]: 0.0397136
[execute]: 1.014e-05
Sums
    bootstrap                            :     0.000345s :  0.65%
    type_inference                       :     0.008092s : 15.16%
    event_method                         :     0.000024s :  0.04%
    auto_monad                           :     0.000080s :  0.15%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000057s :  0.11%
    rewriter_before_opt_a                :     0.000087s :  0.16%
    expand_dump_flag                     :     0.000005s :  0.01%
    jit_opt_a.switch_simplify            :     0.000077s :  0.14%
    jit_opt_a.loop_unroll                :     0.000034s :  0.06%
    jit_opt_a.a_1                        :     0.000744s :  1.39%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.02%
    jit_opt_a.merge_forward              :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000017s :  0.03%
    jit_opt_a.inline_without_move        :     0.000010s :  0.02%
    jit_opt_a.renormalize                :     0.000873s :  1.64%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.04%
    jit_opt_a.cse                        :     0.000043s :  0.08%
    jit_opt_a.replace_applicator         :     0.000021s :  0.04%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.03%
    rewriter_after_opt_a                 :     0.000040s :  0.07%
    convert_after_rewriter               :     0.000007s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000752s :  1.41%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.03%
    cconv                                :     0.000031s :  0.06%
    loop_unroll                          :     0.001054s :  1.97%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.04%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000036s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.01%
    remove_dup_value                     :     0.  0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000062s :  0.31%
    cse_after_recomputation.cse          :     0.000012s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.05%
    opt_after_jit_grad                   :     0.000628s :  3.17%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000002s :  0.01%
    validate                             :     0.000043s :  0.22%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007380s : 37.28%
    execute                              :     0.000010s :  0.05%

000016s :  0.03%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.01%
    add_recomputation                    :     0.000060s :  0.11%
    cse_after_recomputation.cse          :     0.000009s :  0.02%
    auto_monad_reorder                   :     0.000016s :  0.03%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.02%
    opt_after_jit_grad                   :     0.000649s :  1.22%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000050s :  0.09%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.039714s : 74.40%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000275    23
 0.69% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000001s :      2: substitution.fold_const_symbol
 2.26% :     0.000006s :      3: substitution.graph_param_transform
86.50% :     0.000238s :      4: substitution.inline
 1.28% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.74% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.78% :     0.000005s :      2: substitution.replace_old_param
 5.23% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007133     2
85.70% :     0.006113s :      1: type_inference.infer
14.30% :     0.001020s :      1: type_inference.specialize
------[replace.]   0.000083     6
66.08% :     0.000055s :      4: replace.inline
33.92% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000249     6
94.78% :     0.000236s :      4: match.inline
 5.22% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.96% :     0.000003s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 1.95% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 2.05% :     0.000003s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.51% :     0.000001s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.51% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.24% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     10: predicate.environ_get_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 6.45% :     0.000008s :     22: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.92% :     0.000001s :      6: predicate.less_batch_normalization
 1.73% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.56% :     0.000002s :     15: predicate.load_eliminater
 1.97% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.47% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.93% :     0.000001s :     10: predicate.merge_addn
 0.92% :     0.000001s :     10: predicate.minmaximum_grad
 2.08% :     0.000003s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.07% :     0.000003s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.63% :     0.000002s :     10: predicate.reshape_eliminate
 1.09% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3:
Time group info:
------[substitution.]   0.000298    23
 1.01% :     0.000003s :      2: substitution.elim_not_effective
 0.44% :     0.000001s :      2: substitution.fold_const_symbol
 2.59% :     0.000008s :      3: substitution.graph_param_transform
87.16% :     0.000260s :      4: substitution.inline
 1.28% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.83% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.64% :     0.000005s :      2: substitution.replace_old_param
 4.05% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007982     2
84.19% :     0.006720s :      1: type_inference.infer
15.81% :     0.001262s :      1: type_inference.specialize
------[replace.]   0.000078     6
68.37% :     0.000053s :      4: replace.inline
31.63% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000268     6
95.91% :     0.000257s :      4: match.inline
 4.09% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.39% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000002s :     10: predicate.addn_zero_filter
 2.39% :     0.000003s :     10: predicate.arithmetic_simplify
 1.38% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.57% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.13% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.08% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.09% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.70% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     10: predicate.environ_get_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 6.47% :     0.000008s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.48% :     0.000002s :      6: predicate.less_batch_normalization
 1.71% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 2.52% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.68% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.14% :     0.000003s :      3: predicate.mutable_eliminate
 0.61% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000002s :     10: predicate.same_eliminate
 0.69% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.83% :     0.000002s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.24% :     0.000003s :     16: predicate.switch_defer_inline
 1.89% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.16% :     0.000009s :     53: predicate.switch_simplify
 1.33% :     0.000002s :     10: predicate.tile_eliminate
 1.50% :     0.000002s :     10: predicate.transpose_eliminate
 1.53% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.58% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.47% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.48% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.92% :     0.000003s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000712    12
50.70% :     0.000361s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.06% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.23% :     0.000293s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.029723    76
 0.22% :     0.000066s :      1: add_recomputation
 0.24% :     0.000071s :      1: auto_monad
 0.07% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.64% :     0.000489s :      1: bootstrap
 0.12% :     0.000035s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000027s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.08% :     0.000024s :      1: event_method
 0.05% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
32.32% :     0.009605s :      1: jit_opt_a
 0.59% :     0.000175s :      1: jit_opt_after_cconv
 0.22% :     0.000065s :      1: jit_opt_b
 2.18% :     0.000648s :      1: loop_unroll
 2.84% :     0.000844s :      1: mutable_eliminate
 3.24% :     0.000964s :     26: opt.transform.jit_opt_a
 0.18% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000033s :      4: opt.transform.jit_opt_b
 0.06% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.15% :     0.000640s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.15% :     0.000043s :      1: py_interpret_to_execute
 0.06% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000019s :      1: remove_dup_value
 1.60% :     0.000474s :      1: renormalize.infer
 1.05% :     0.000311s :      1: renormalize.specialize
 0.04% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000045s :      1: rewriter_after_opt_a
 0.27% :     0.000080s :      1: rewriter_before_opt_a
 0.27% :     0.000081s :      1: symbol_engine_optimizer
24.89% :     0.007398s :       predicate.row_tensor_eliminate
 1.10% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.14% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.19% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.87% :     0.000010s :     53: predicate.switch_simplify
 1.25% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.19% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.01% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.05% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.51% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.95% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000865    12
45.36% :     0.000392s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 9.05% :     0.000078s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.59% :     0.000394s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.061244    76
 0.10% :     0.000063s :      1: add_recomputation
 0.14% :     0.000084s :      1: auto_monad
 0.03% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.60% :     0.000370s :      1: bootstrap
 0.06% :     0.000034s :      1: cconv
 0.02% :     0.000011s :      1: convert_after_rewriter
 0.04% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.05% :     0.000029s :      1: event_method
 0.02% :     0.000015s :      1: execute
 0.01% :     0.000007s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000009s :      1: graph_reusing
12.27% :     0.007513s :      1: jit_opt_a
 0.35% :     0.000215s :      1: jit_opt_after_cconv
 0.09% :     0.000053s :      1: jit_opt_b
 1.75% :     0.001069s :      1: loop_unroll
 1.25% :     0.000764s :      1: mutable_eliminate
 1.59% :     0.000973s :     26: opt.transform.jit_opt_a
 0.10% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000025s :      4: opt.transform.jit_opt_b
 0.03% :     0.000020s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.08% :     0.000659s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.10% :     0.000061s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000019s :      1: remove_dup_value
 0.80% :     0.000490s :      1: renormalize.infer
 0.61% :     0.000373s :      1: renormalize.specialize
 0.02% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000043s :      1: rewriter_after_opt_a
 0.15% :     0.000091s :      1: rewriter_before_opt_a
 0.14% :     0.000084s :      1: symbol_engine_optimizer
64.87% :     0.039731s :      1: task_emit
24.38% :     0.007247s :      1: type_inference
 0.23% :     0.000067s :      1: validate

1: task_emit
13.26% :     0.008121s :      1: type_inference
 0.13% :     0.000077s :      1: validate


TotalTime = 0.01733, [30]
[bootstrap]: 0.00041724
[type_inference]: 0.00605411
[event_method]: 2.21e-05
[auto_monad]: 7.663e-05
[graph_reusing]: 6.33998e-06
[pre_auto_parallel]: 2.74999e-06
[py_interpret_to_execute]: 4.927e-05
[rewriter_before_opt_a]: 8.525e-05
[expand_dump_flag]: 3.86999e-06
[jit_opt_a]: 0.00789402, [2]
    [Cycle 1]: 0.00224054, [27]
        [switch_simplify]: 7.19e-05
        [loop_unroll]: 3.131e-05
        [a_1]: 0.00066295
        [with_stream_mark]: 2.597e-05
        [recompute_prepare]: 1.061e-05
        [updatestate_depend_eliminate]: 4.55001e-06
        [updatestate_assign_eliminate]: 3.35e-06
        [updatestate_loads_eliminate]: 3.01001e-06
        [parameter_eliminate]: 1.77999e-06
        [specialize_transform]: 6.79999e-06
        [updatestate_useless_node_eliminater]: 5.51e-06
        [accelerated_algorithm]: 6.46e-06
        [meta_shard_fg_expand]: 3.21001e-06
        [get_grad_eliminate_]: 6.09001e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.71e-05
        [j_node_and_user_rematch]: 1.076e-05
        [meta_fg_expand]: 2.51e-06
        [replace_old_param]: 1.022e-05
        [inline_without_move]: 5.68002e-06
        [renormalize]: 0.00105185
        [add_forward_monad_depend]: 1.43e-05
        [auto_monad_grad]: 2.86e-06
        [auto_monad_eliminator]: 1.729e-05
        [cse]: 3.166e-05
        [replace_applicator]: 2.07e-05
    [Cycle 2]: 0.00040753, [27]
        [switch_simplify]: 7.09001e-06
        [loop_unroll]: 5.61e-06
        [a_1]: 0.00013229
        [with_stream_mark]: 1.439e-05
        [recompute_prepare]: 7.26001e-06
        [updatestate_depend_eliminate]: 3.46999e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 5.47999e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 6.04001e-06
        [meta_shard_fg_expand]: 1.99999e-06
        [get_grad_eliminate_]: 5.20999e-06
        [merge_forward]: 3.05998e-06
        [cell_reuse_recompute_pass]: 3.45e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.583e-05
        [j_node_and_user_rematch]: 9.96e-06
        [meta_fg_expand]: 2.22999e-06
        [replace_old_param]: 8.77999e-06
        [inline_without_move]: 4.94e-06
        [renormalize]: 1.10012e-07
        [add_forward_monad_depend]: 1.38002e-06
        [auto_monad_grad]: 1.11997e-06
        [auto_monad_eliminator]: 5.91e-06
        [cse]: 1.441e-05
        [replace_applicator]: 5.74e-06
[py_interpret_to_execute_after_opt_a]: 1.674e-05
[rewriter_after_opt_a]: 4.204e-05
[convert_after_rewriter]: 8.12998e-06
[order_py_execute_after_rewriter]: 5.89e-06
[mutable_eliminate]: 0.00085835
[jit_opt_b]: 6.968e-05, [1]
    [Cycle 1]: 5.945e-05, [2]
        [frontend_op_eliminate]: 2.752e-05
        [inline_after_opt_a]: 1.848e-05
[cconv]: 3.606e-05
[loop_unroll]: 0.00052008
[jit_opt_after_cconv]: 0.00017358, [1]
    [Cycle 1]: 0.0001659, [11]
        [c_1]: 2.304e-05
        [parameter_eliminate]: 4.70999e-06
        [updatestate_depend_eliminate]: 9.22001e-06
        [updatestate_assign_eliminate]: 3.41999e-06
        [updatestate_loads_eliminate]: 2.76e-06
        [cse]: 3.007e-05
        [call_graph_tuple_transform]: 2.348e-05
        [tuple_list_get_item_eliminator]: 5.59e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 1.02e-06
        [switch_simplify]: 6.06e-06
[remove_dup_value]: 1.676e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 5.86e-06
[add_recomputation]: 5.633e-05
[cse_after_recomputation]: 2.526e-05, [1]
    [Cycle 1]: 1.947e-05, [1]
        [cse]: 1.201e-05
[auto_monad_reorder]: 1.667e-05
[get_jit_bprop_graph]: 2.40002e-06
[rewriter_after_jit_bprop_graph]: 6.48003e-06
[opt_after_jit_grad]: 0.00052346
[symbol_engine_optimizer]: 8.378e-05, [1]
    [Cycle 1]: 7.532e-05, [6]
        [build]: 4.55001e-06
        [elim_shapecalc]: 8.40999e-06
        [elim_not_effective]: 1.517e-05
        [opt_reshape]: 6.74001e-06
        [fold_const_symbol]: 9.42999e-06
        [renormalize]: 9.30013e-07
[validate]: 4.588e-05
Sums
    bootstrap                            :     0.000417s :  3.66%
    type_inference                       :     0.006054s : 53.08%
    event_method                         :     0.000022s :  0.19%
    auto_monad                           :     0.000077s :  0.67%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000049s :  0.43%
    rewriter_before_opt_a                :     0.000085s :  0.75%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000079s :  0.69%
    jit_opt_a.loop_unroll                :     0.000037s :  0.32%
    jit_opt_a.a_1                        :     0.000795s :  6.97%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.17%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.001052s :  9.22%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.14%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.20%
    jit_opt_a.cse                        :     0.000046s :  0.40%
    jit_opt_a.replace_applicator         :     0.000026s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.15%
    rewriter_after_opt_a                 :     0.000042s :  0.37%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000858s :  7.53%
    jit_opt_b.frontend_op_eliminate      :     0.000028s :  0.24%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.16%
    cconv                                :     0.000036s :  0.32%
    loop_unroll                          :     0.000520s :  4.56%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000030s :  0.26%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000017s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000056s :  0.49%
    cse_after_recomputation.cse          :     0.000012s :  0.11%
    auto_monad_reorder                   :     0.000017s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000523s :  4.59%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000046s :  0.40%


Time group info:
------[substitution.]   0.000289    23
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000002s :      2: substitution.fold_const_symbol
 2.23% :     0.000006s :      3: substitution.graph_param_transform
85.63% :     0.000247s :      4: substitution.inline
 1.62% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.88% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000006s :      2: substitution.replace_old_param
 5.22% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005916     2
81.67% :     0.004831s :      1: type_inference.infer
18.33% :     0.001084s :      1: type_inference.specialize
------[replace.]   0.000086     6
62.53% :     0.000054s :      4: replace.inline
37.47% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000259     6
94.63% :     0.000245s :      4: match.inline
 5.37% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.91% :     0.000001s :     10: predicate.addn_check_dump
 1.13% :     0.000002s :     10: predicate.addn_zero_filter
 2.32% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.91% :     0.000001s :     10: predicate.compare_switch_simplify
 1.53% :     0.000002s :     10: predicate.depend_value_elim
 0.96% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.51% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     10: predicate.environ_get_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.90% :     0.000008s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.77% :     0.000002s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.48% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.58% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.74% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.93% :     0.000001s :     10: predicate.merge_addn
 0.88% :     0.000001s :     10: predicate.minmaximum_grad
 2.88% :     0.000004s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 1.99% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.69% :     0.000004s :     10: predicate.reduce_eliminate
 1.32% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.20% :     0.000000s :      3: predicate.reset_defer_inline
 1.72% :     0.000002s :     10: predicate.reshape_eliminate
 1.11% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.20% :     0.000002s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.57% :     0.000004s :     16: predicate.switch_defer_inline
 1.77% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.71% :     0.000009s :     53: predicate.switch_simplify
 1.28% :     0.000002s :     10: predicate.tile_eliminate
 1.15% :     0.000002s :     10: predicate.transpose_eliminate
 1.74% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.01% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.27% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.85% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.42% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.11% :     0.000003s :     10: predicate.value_based_eliminate
 0.37% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.91% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000806    12
46.15% :     0.000372s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.43% :     0.000060s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.42% :     0.000374s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019497    72
 0.30% :     0.000059s :      1: add_recomputation
 0.41% :     0.000080s :      1: auto_monad
 0.10% :     0.000019s :      1: auto_monad_reorder
 2.22% :     0.000433s :      1: bootstrap
 0.20% :     0.000039s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.14% :     0.000028s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.14% :     0.000027s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
40.51% :     0.007898s :      1: jit_opt_a
 0.91% :     0.000176s :      1: jit_opt_after_cconv
 0.37% :     0.000073s :      1: jit_opt_b
 2.72% :     0.000530s :      1: loop_unroll
 4.48% :     0.000874s :      1: mutable_eliminate
 5.38% :     0.001049s :     26: opt.transform.jit_opt_a
 0.28% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000037s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.73% :     0.000532s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000053s :      1: py_interpret_to_execute
 0.10% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000019s :      1: remove_dup_value
 2.83% :     0.000551s :      1: renormalize.infer
 2.51% :     0.000490s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000045s :      1: rewriter_after_opt_a
 0.45% :     0.000088s :      1: rewriter_before_opt_a
 0.44% :     0.000086s :      1: symbol_engine_optimizer
31.16% :     0.006075s :      1: type_inference


TotalTime = 0.0435279, [33]
[bootstrap]: 0.00047163
[type_inference]: 0.00644508
[event_method]: 2.137e-05
[auto_monad]: 6.098e-05
[graph_reusing]: 4.48999e-06
[pre_auto_parallel]: 2.71999e-06
[py_interpret_to_execute]: 5.213e-05
[rewriter_before_opt_a]: 7.782e-05
[expand_dump_flag]: 2.94999e-06
[jit_opt_a]: 0.0237355, [2]
    [Cycle 1]: 0.00194838, [27]
        [switch_simplify]: 6.229e-05
        [loop_unroll]: 3.257e-05
        [a_1]: 0.00059423
        [with_stream_mark]: 1.739e-05
        [recompute_prepare]: 8.65999e-06
        [updatestate_depend_eliminate]: 3.14001e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [parameter_eliminate]: 1.58002e-06
        [specialize_transform]: 7.01999e-06
        [updatestate_useless_node_eliminater]: 6.17999e-06
        [accelerated_algorithm]: 6.15002e-06
        [meta_shard_fg_expand]: 2.32001e-06
        [get_grad_eliminate_]: 6.22001e-06
        [merge_forward]: 3.36001e-06
        [cell_reuse_recompute_pass]: 1.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.358e-05
        [j_node_and_user_rematch]: 1.027e-05
        [meta_fg_expand]: 2.54001e-06
        [replace_old_param]: 9.39998e-06
        [inline_without_move]: 6.77002e-06
        [renormalize]: 0.00088998
        [add_forward_monad_depend]: 6.29999e-06
        [auto_monad_grad]: 1.60999e-06
        [auto_monad_eliminator]: 1.38e-05
        [cse]: 2.195e-05
        [replace_applicator]: 1.76e-05
    [Cycle 2]: 0.0148309, [27]
        [switch_simplify]: 6.56999e-06
        [loop_unroll]: 6.14001e-06
        [a_1]: 0.00011353
        [with_stream_mark]: 1.405e-05
        [recompute_prepare]: 6.54001e-06
        [updatestate_depend_eliminate]: 3.31001e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.11003e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 5.90002e-06
        [updatestate_useless_node_eliminater]: 6.79001e-06
        [accelerated_algorithm]: 7.82998e-06
        [meta_shard_fg_expand]: 2.17001e-06
        [get_grad_eliminate_]: 5.24e-06
        [merge_forward]: 3.01999e-06
        [cell_reuse_recompute_pass]: 3.35e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.712e-05
        [j_node_and_user_rematch]: 9.76e-06
        [meta_fg_expand]: 1.76e-06
        [replace_old_param]: 8.06001e-06
        [inline_without_move]: 5.43002e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.58002e-06
        [auto_monad_grad]: 1.04e-06
        [auto_monad_eliminator]: 7.26999e-06
        [cse]: 1.412e-05
        [replace_applicator]: 1.074e-05
[py_interpret_to_execute_after_opt_a]: 3.213e-05
[rewriter_after_opt_a]: 6.465e-05
[convert_after_rewriter]: 9.27001e-06
[order_py_execute_after_rewriter]: 6.09001e-06
[mutable_eliminate]: 0.00080243
[jit_opt_b]: 6.721e-05, [1]
    [Cycle 1]: 5.929e-05, [2]
        [frontend_op_eliminate]: 2.371e-05
        [inline_after_opt_a]: 2.154e-05
[cconv]: 3.007e-05
[loop_unroll]: 0.00047321
[jit_opt_after_cconv]: 0.00017264, [1]
    [Cycle 1]: 0.00016547, [11]
        [c_1]: 2.489e-05
        [parameter_eliminate]: 3.96001e-06
        [updatestate_depend_eliminate]: 7.77002e-06
        [updatestate_assign_eliminate]: 2.83998e-06
        [updatestate_loads_eliminate]: 2.66999e-06
        [cse]: 2.961e-05
        [call_graph_tuple_transform]: 2.39e-05
        [tuple_list_get_item_eliminator]: 6.01e-06
        [none_parameter_eliminate]: 1.24e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 6.43e-06
[remove_dup_value]: 1.145e-05
[partial_unused_args_eliminate]: 2.80002e-06
[environ_conv]: 6.18998e-06
[add_recomputation]: 4.35e-05
[cse_after_recomputation]: 2.493e-05, [1]
    [Cycle 1]: 1.782e-05, [1]
        [cse]: 1.078e-05
[auto_monad_reorder]: 1.356e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 6.79999e-06
[opt_after_jit_grad]: 0.00063442
[symbol_engine_optimizer]: 8.744e-05, [1]
    [Cycle 1]: 7.8e-05, [6]
        [build]: 4.4e-06
        [elim_shapecalc]: 9.68002e-06
        [elim_not_effective]: 1.516e-05
        [opt_reshape]: 6.73e-06
        [fold_const_symbol]: 1.058e-05
        [renormalize]: 8.39995e-07
[validate]: 4.253e-05
[backend_pass]: 8.00006e-07
[task_emit]: 0.00983082
[execute]: 1.008e-05
Sums
    bootstrap                            :     0.000472s :  2.20%
    type_inference                       :     0.006445s : 30.11%
    event_method                         :     0.000021s :  0.10%
    auto_monad                           :     0.000061s :  0.28%
    graph_reusing                        :     0.000004s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000052s :  0.24%
    rewriter_before_opt_a                :     0.000078s :  0.36%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000069s :  0.32%
    jit_opt_a.loop_unroll                :     0.000039s :  0.18%
    jit_opt_a.a_1                        :     0.000708s :  3.31%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000013s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000014s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000041s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.08%
    jit_opt_a.inline_without_move        :     0.000012s :  0.06%
    jit_opt_a.renormalize                :     0.000890s :  4.16%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.10%
    jit_opt_a.cse                        :     0.000036s :  0.17%
    jit_opt_a.replace_applicator         :     0.000028s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000032s :  0.15%
    rewriter_after_opt_a                 :     0.000065s :  0.30%
    convert_after_rewriter               :     0.000009s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000802s :  3.75%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000022s :  0.10%
    cconv                                :     0.000030s :  0.14%
    loop_unroll                          :     0.000473s :  2.21%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000030s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000011s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000043s :  0.20%
    cse_after_recomputation.cse          :     0.000011s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.03%
    opt_after_jit_grad                   :     0.000634s :  2.96%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.20%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009831s : 45.93%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000234    23
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.71% :     0.000006s :      3: substitution.graph_param_transform
80.92% :     0.000190s :      4: substitution.inline
 1.78% :     0.000004s :      4: substitution.j_node_and_user_rematch
 6.66% :     0.000016s :      4: substitution.remove_not_recompute_node
 1.60% :     0.000004s :      2: substitution.replace_old_param
 4.83% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006362     2
82.44% :     0.005245s :      1: type_inference.infer
17.56% :     0.001117s :      1: type_inference.specialize
------[replace.]   0.000083     6
66.39% :     0.000055s :      4: replace.inline
33.61% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000197     6
94.88% :     0.000187s :      4: match.inline
 5.12% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   691
 1.78% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.92% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 2.73% :     0.000004s :     10: predicate.arithmetic_simplify
 1.56% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.48% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.80% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.62% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.42% :     0.000002s :     10: predicate.environ_get_eliminate
 1.33% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.15% :     0.000002s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 4.71% :     0.000006s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.69% :     0.000002s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.60% :     0.000002s :     15: predicate.load_eliminater
 1.16% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.35% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.25% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 1.80% :     0.000002s :      3: predicate.mutable_eliminate
 0.63% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.03% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.27% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.56% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.03% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.69% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.25% :     0.000003s :     16: predicate.switch_defer_inline
 1.91% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.02% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000002s :     10: predicate.tile_eliminate
 1.27% :     0.000002s :     10: predicate.transpose_eliminate
 1.55% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.95% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000800    12
47.61% :     0.000381s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.96% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.43% :     0.000379s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.045501    76
 0.10% :     0.000047s :      1: add_recomputation
 0.14% :     0.000064s :      1: auto_monad
 0.05% :     0.000023s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.09% :     0.000495s :      1: bootstrap
 0.07% :     0.000033s :      1: cconv
 0.03% :     0.000012s :      1: convert_after_rewriter
 0.06% :     0.000027s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.06% :     0.000026s :      1: event_method
 0.04% :     0.000017s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
52.18% :     0.023744s :      1: jit_opt_a
 0.39% :     0.000176s :      1: jit_opt_after_cconv
 0.15% :     0.000070s :      1: jit_opt_b
 1.06% :     0.000482s :      1: loop_unroll
 1.79% :     0.000812s :      1: mutable_eliminate
 2.13% :     0.000969s :     26: opt.transform.jit_opt_a
 0.12% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000036s :      4: opt.transform.jit_opt_b
 0.03% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.42% :     0.000646s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000056s :      1: py_interpret_to_execute
 0.08% :     0.000036s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000015s :      1: remove_dup_value
 1.01% :     0.000458s :      1: renormalize.infer
 0.93% :     0.000422s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000069s :      1: rewriter_after_opt_a
 0.18% :     0.000081s :      1: rewriter_before_opt_a
 0.20% :     0.000090s :      1: symbol_engine_optimizer
21.65% :     0.009853s :      1: task_emit
14.21% :     0.006466s :      1: type_inference
 0.14% :     0.000063s :      1: validate


TotalTime = 0.0219438, [33]
[bootstrap]: 0.0003913
[type_inference]: 0.00528843
[event_method]: 2.02e-05
[auto_monad]: 6.488e-05
[graph_reusing]: 5.58002e-06
[pre_auto_parallel]: 2.39999e-06
[py_interpret_to_execute]: 3.529e-05
[rewriter_before_opt_a]: 7.552e-05
[expand_dump_flag]: 3.45e-06
[jit_opt_a]: 0.00608951, [2]
    [Cycle 1]: 0.00203659, [27]
        [switch_simplify]: 6.214e-05
        [loop_unroll]: 2.874e-05
        [a_1]: 0.0007079
        [with_stream_mark]: 2.851e-05
        [recompute_prepare]: 9.42001e-06
        [updatestate_depend_eliminate]: 4.26001e-06
        [updatestate_assign_eliminate]: 3.33e-06
        [updatestate_loads_eliminate]: 3.4e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 5.96998e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.94e-06
        [meta_shard_fg_expand]: 1.99999e-06
        [get_grad_eliminate_]: 5.47001e-06
        [merge_forward]: 3.98001e-06
        [cell_reuse_recompute_pass]: 1.64998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.434e-05
        [j_node_and_user_rematch]: 1.006e-05
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 1.044e-05
        [inline_without_move]: 5.87999e-06
        [renormalize]: 0.00082546
        [add_forward_monad_depend]: 7.26999e-06
        [auto_monad_grad]: 3.28e-06
        [auto_monad_eliminator]: 1.759e-05
        [cse]: 3.513e-05
        [replace_applicator]: 1.374e-05
    [Cycle 2]: 0.00034356, [27]
        [switch_simplify]: 6.12999e-06
        [loop_unroll]: 5.08002e-06
        [a_1]: 9.537e-05
        [with_stream_mark]: 1.301e-05
        [recompute_prepare]: 4.66002e-06
        [updatestate_depend_eliminate]: 2.98e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.32999e-06
        [specialize_transform]: 4.72e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 6.02001e-06
        [meta_shard_fg_expand]: 1.05001e-06
        [get_grad_eliminate_]: 5.13002e-06
        [merge_forward]: 2.98e-06
        [cell_reuse_recompute_pass]: 2.32999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.655e-05
        [j_node_and_user_rematch]: 9.31e-06
        [meta_fg_expand]: 2.27999e-06
        [replace_old_param]: 7.97e-06
        [inline_without_move]: 5.19e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.26002e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 5.64998e-06
        [cse]: 1.161e-05
        [replace_applicator]: 5.51e-06
[py_interpret_to_execute_after_opt_a]: 1.529e-05
[rewriter_after_opt_a]: 4.054e-05
[convert_after_rewriter]: 6.91001e-06
[order_py_execute_after_rewriter]: 5.17999e-06
[mutable_eliminate]: 0.00080553
[jit_opt_b]: 5.566e-05, [1]
    [Cycle 1]: 4.641e-05, [2]
        [frontend_op_eliminate]: 1.784e-05
        [inline_after_opt_a]: 1.69e-05
[cconv]: 3.319e-05
[loop_unroll]: 0.00048502
[jit_opt_after_cconv]: 0.00016204, [1]
    [Cycle 1]: 0.00015454, [11]
        [c_1]: 2.195e-05
        [parameter_eliminate]: 3.98999e-06
        [updatestate_depend_eliminate]: 9.61e-06
        [updatestate_assign_eliminate]: 2.49999e-06
        [updatestate_loads_eliminate]: 2.32001e-06
        [cse]: 2.868e-05
        [call_graph_tuple_transform]: 2.215e-05
        [tuple_list_get_item_eliminator]: 5.19e-06
        [none_parameter_eliminate]: 1.72001e-06
        [renormalize]: 1.04e-06
        [switch_simplify]: 5.12999e-06
[remove_dup_value]: 1.616e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 6.02999e-06
[add_recomputation]: 5.996e-05
[cse_after_recomputation]: 2.293e-05, [1]
    [Cycle 1]: 1.699e-05, [1]
        [cse]: 1.028e-05
[auto_monad_reorder]: 1.591e-05
[get_jit_bprop_graph]: 2.79999e-06
[rewriter_after_jit_bprop_graph]: 9.17999e-06
[opt_after_jit_grad]: 0.00061384
[symbol_engine_optimizer]: 8.028e-05, [1]
    [Cycle 1]: 7.163e-05, [6]
        [build]: 4.72e-06
        [elim_shapecalc]: 7.33999e-06
        [elim_not_effective]: 1.549e-05
        [opt_reshape]: 6.66999e-06
        [fold_const_symbol]: 9.32001e-06
        [renormalize]: 8.30012e-07
[validate]: 4.348e-05
[backend_pass]: 1.19998e-06
[task_emit]: 0.00720884
[execute]: 1.113e-05
Sums
    bootstrap                            :     0.000391s :  2.23%
    type_inference                       :     0.005288s : 30.19%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000065s :  0.37%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000035s :  0.20%
    rewriter_before_opt_a                :     0.000076s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.19%
    jit_opt_a.a_1                        :     0.000803s :  4.59%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000826s :  4.71%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.13%
    jit_opt_a.cse                        :     0.000047s :  0.27%
    jit_opt_a.replace_applicator         :     0.000019s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.09%
    rewriter_after_opt_a                 :     0.000041s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000806s :  4.60%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.10%
    cconv                                :     0.000033s :  0.19%
    loop_unroll                          :     0.000485s :  2.77%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000060s :  0.34%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000614s :  3.50%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007209s : 41.16%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000266    23
 0.73% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000002s :      2: substitution.fold_const_symbol
 2.29% :     0.000006s :      3: substitution.graph_param_transform
85.28% :     0.000227s :      4: substitution.inline
 1.80% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.10% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.11% :     0.000006s :      2: substitution.replace_old_param
 5.09% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005205     2
83.77% :     0.004360s :      1: type_inference.infer
16.23% :     0.000845s :      1: type_inference.specialize
------[replace.]   0.000078     6
67.50% :     0.000053s :      4: replace.inline
32.50% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000237     6
94.76% :     0.000224s :      4: match.inline
 5.24% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.36% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.19% :     0.000002s :     10: predicate.addn_zero_filter
 2.24% :     0.000003s :     10: predicate.arithmetic_simplify
 1.10% :     0.000001s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.10% :     0.000001s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.06% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.09% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.43% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.41% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     10: predicate.environ_get_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.41% :     0.000002s :      6: predicate.get_grad_eliminate
 0.36% :     0.000000s :      3: predicate.graph_param_transform
 6.34% :     0.000008s :     22: predicate.inline
 1.17% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.76% :     0.000002s :      6: predicate.less_batch_normalization
 2.55% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.89% :     0.000002s :     15: predicate.load_eliminater
 1.56% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.62% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.63% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 1.85% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 1.92% :     0.000002s :     15: predicate.partial_eliminate
 1.02% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.32% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.86% :     0.000001s :      6: predicate.replace_old_param
 0.70% :     0.000001s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.33% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.02% :     0.000001s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.18% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.05% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 1.96% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.35% :     0.000009s :     53: predicate.switch_simplify
 1.09% :     0.000001s :     10: predicate.tile_eliminate
 1.98% :     0.000002s :     10: predicate.transpose_eliminate
 1.17% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.17% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.36% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000651    12
50.06% :     0.000326s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 9.13% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.81% :     0.000266s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023918    76
 0.27% :     0.000065s :      1: add_recomputation
 0.29% :     0.000069s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.76% :     0.000420s :      1: bootstrap
 0.15% :     0.000036s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.07% :     0.000018s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
25.48% :     0.006093s :      1: jit_opt_a
 0.69% :     0.000165s :      1: jit_opt_after_cconv
 0.24% :     0.000058s :      1: jit_opt_b
 2.07% :     0.000495s :      1: loop_unroll
 3.43% :     0.000820s :      1: mutable_eliminate
 4.31% :     0.001032s :     26: opt.transform.jit_opt_a
 0.21% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.62% :     0.000627s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000038s :      1: py_interpret_to_execute
 0.08% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000019s :      1: remove_dup_value
 1.91% :     0.000458s :      1: renormalize.infer
 1.50% :     0.000359s :      1: renormalize.specialize
 0.05% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000044s :      1: rewriter_after_opt_a
 0.33% :     0.000079s :      1: rewriter_before_opt_a
 0.35% :     0.000083s :      1: symbol_engine_optimizer
30.24% :     0.007232s :      1: task_emit
22.20% :     0.005310s :      1: type_inference
 0.29% :     0.000070s :      1: validate


TotalTime = 0.0226001, [33]
[bootstrap]: 0.00045044
[type_inference]: 0.00556489
[event_method]: 1.826e-05
[auto_monad]: 6.671e-05
[graph_reusing]: 5.46e-06
[pre_auto_parallel]: 2.73998e-06
[py_interpret_to_execute]: 4.451e-05
[rewriter_before_opt_a]: 7.451e-05
[expand_dump_flag]: 3.43e-06
[jit_opt_a]: 0.00600244, [2]
    [Cycle 1]: 0.00182668, [27]
        [switch_simplify]: 5.819e-05
        [loop_unroll]: 3.128e-05
        [a_1]: 0.00057798
        [with_stream_mark]: 1.986e-05
        [recompute_prepare]: 7.37997e-06
        [updatestate_depend_eliminate]: 3.30998e-06
        [updatestate_assign_eliminate]: 2.86999e-06
        [updatestate_loads_eliminate]: 2.98998e-06
        [parameter_eliminate]: 1.91003e-06
        [specialize_transform]: 5.89999e-06
        [updatestate_useless_node_eliminater]: 5.49998e-06
        [accelerated_algorithm]: 6.05002e-06
        [meta_shard_fg_expand]: 2.12001e-06
        [get_grad_eliminate_]: 5.61e-06
        [merge_forward]: 3.57002e-06
        [cell_reuse_recompute_pass]: 1.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.421e-05
        [j_node_and_user_rematch]: 9.36e-06
        [meta_fg_expand]: 2.46e-06
        [replace_old_param]: 9.14e-06
        [inline_without_move]: 5.80002e-06
        [renormalize]: 0.00080794
        [add_forward_monad_depend]: 5.90002e-06
        [auto_monad_grad]: 2.89999e-06
        [auto_monad_eliminator]: 1.502e-05
        [cse]: 3.083e-05
        [replace_applicator]: 1.246e-05
    [Cycle 2]: 0.00033495, [27]
        [switch_simplify]: 5.72999e-06
        [loop_unroll]: 5.14e-06
        [a_1]: 9.299e-05
        [with_stream_mark]: 9.24e-06
        [recompute_prepare]: 5.64998e-06
        [updatestate_depend_eliminate]: 2.69999e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 8.59989e-07
        [specialize_transform]: 5.10001e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 5.52001e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.60001e-06
        [merge_forward]: 2.51e-06
        [cell_reuse_recompute_pass]: 2.68e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.101e-05
        [j_node_and_user_rematch]: 8.05e-06
        [meta_fg_expand]: 2.20002e-06
        [replace_old_param]: 7.82e-06
        [inline_without_move]: 5.05999e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 5.24e-06
        [cse]: 1.086e-05
        [replace_applicator]: 7.25e-06
[py_interpret_to_execute_after_opt_a]: 1.213e-05
[rewriter_after_opt_a]: 3.788e-05
[convert_after_rewriter]: 8.27e-06
[order_py_execute_after_rewriter]: 4.88001e-06
[mutable_eliminate]: 0.00157624
[jit_opt_b]: 5.752e-05, [1]
    [Cycle 1]: 4.914e-05, [2]
        [frontend_op_eliminate]: 1.883e-05
        [inline_after_opt_a]: 1.742e-05
[cconv]: 3.321e-05
[loop_unroll]: 0.00048198
[jit_opt_after_cconv]: 0.00016249, [1]
    [Cycle 1]: 0.00015538, [11]
        [c_1]: 2.196e-05
        [parameter_eliminate]: 4.42003e-06
        [updatestate_depend_eliminate]: 8.54002e-06
        [updatestate_assign_eliminate]: 3.16001e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [cse]: 2.562e-05
        [call_graph_tuple_transform]: 2.219e-05
        [tuple_list_get_item_eliminator]: 5.92999e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 6.06998e-06
[remove_dup_value]: 1.614e-05
[partial_unused_args_eliminate]: 2.54999e-06
[environ_conv]: 6.78998e-06
[add_recomputation]: 5.725e-05
[cse_after_recomputation]: 2.097e-05, [1]
    [Cycle 1]: 1.559e-05, [1]
        [cse]: 9.61e-06
[auto_monad_reorder]: 1.571e-05
[get_jit_bprop_graph]: 2.23998e-06
[rewriter_after_jit_bprop_graph]: 8.47e-06
[opt_after_jit_grad]: 0.00050207
[symbol_engine_optimizer]: 8.06e-05, [1]
    [Cycle 1]: 7.32e-05, [6]
        [build]: 5.50001e-06
        [elim_shapecalc]: 8.77999e-06
        [elim_not_effective]: 1.4e-05
        [opt_reshape]: 6.31e-06
        [fold_const_symbol]: 9.97999e-06
        [renormalize]: 7.10017e-07
[validate]: 4.732e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.00699628
[execute]: 8.55001e-06
Sums
    bootstrap                            :     0.000450s :  2.49%
    type_inference                       :     0.005565s : 30.72%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000067s :  0.37%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000045s :  0.25%
    rewriter_before_opt_a                :     0.000075s :  0.41%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.35%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000671s :  3.70%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000808s :  4.46%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.11%
    jit_opt_a.cse                        :     0.000042s :  0.23%
    jit_opt_a.replace_applicator         :     0.000020s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.001576s :  8.70%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.10%
    cconv                                :     0.000033s :  0.18%
    loop_unroll                          :     0.000482s :  2.66%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000057s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000502s :  2.77%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000047s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006996s : 38.62%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000241    23
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000002s :      2: substitution.fold_const_symbol
 2.76% :     0.000007s :      3: substitution.graph_param_transform
82.42% :     0.000199s :      4: substitution.inline
 1.22% :     0.000003s :      4: substitution.j_node_and_user_rematch
 4.94% :     0.000012s :      4: substitution.remove_not_recompute_node
 1.72% :     0.000004s :      2: substitution.replace_old_param
 5.45% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005490     2
81.96% :     0.004500s :      1: type_inference.infer
18.04% :     0.000990s :      1: type_inference.specialize
------[replace.]   0.000074     6
67.07% :     0.000049s :      4: replace.inline
32.93% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000208     6
94.24% :     0.000196s :      4: match.inline
 5.76% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.39% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000002s :     10: predicate.addn_zero_filter
 2.03% :     0.000003s :     10: predicate.arithmetic_simplify
 1.48% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.57% :     0.000007s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 1.49% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 2.66% :     0.000003s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.69% :     0.000001s :      3: predicate.reset_defer_inline
 1.27% :     0.000002s :     10: predicate.reshape_eliminate
 1.10% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.13% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.28% :     0.000002s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.70% :     0.000010s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.31% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000728    12
51.17% :     0.000372s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.55% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.28% :     0.000315s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024424    76
 0.25% :     0.000061s :      1: add_recomputation
 0.29% :     0.000070s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.93% :     0.000471s :      1: bootstrap
 0.15% :     0.000036s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.09% :     0.000023s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
24.59% :     0.006006s :      1: jit_opt_a
 0.68% :     0.000165s :      1: jit_opt_after_cconv
 0.25% :     0.000060s :      1: jit_opt_b
 2.01% :     0.000491s :      1: loop_unroll
 6.51% :     0.001589s :      1: mutable_eliminate
 3.68% :     0.000898s :     26: opt.transform.jit_opt_a
 0.21% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000029s :      4: opt.transform.jit_opt_b
 0.06% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.09% :     0.000510s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000047s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000019s :      1: remove_dup_value
 1.78% :     0.000434s :      1: renormalize.infer
 1.49% :     0.000365s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000041s :      1: rewriter_after_opt_a
 0.32% :     0.000077s :      1: rewriter_before_opt_a
 0.34% :     0.000083s :      1: symbol_engine_optimizer
28.72% :     0.007014s :      1: task_emit
22.84% :     0.005579s :      1: type_inference
 0.28% :     0.000069s :      1: validate


TotalTime = 0.0300295, [30]
[bootstrap]: 0.00035807
[type_inference]: 0.00513055
[event_method]: 2.048e-05
[auto_monad]: 6.917e-05
[graph_reusing]: 6.09001e-06
[pre_auto_parallel]: 2.58998e-06
[py_interpret_to_execute]: 4.201e-05
[rewriter_before_opt_a]: 7.752e-05
[expand_dump_flag]: 3.38999e-06
[jit_opt_a]: 0.0219653, [2]
    [Cycle 1]: 0.00191216, [27]
        [switch_simplify]: 6.917e-05
        [loop_unroll]: 3.055e-05
        [a_1]: 0.00064677
        [with_stream_mark]: 2.356e-05
        [recompute_prepare]: 7.99002e-06
        [updatestate_depend_eliminate]: 3.58999e-06
        [updatestate_assign_eliminate]: 2.72001e-06
        [updatestate_loads_eliminate]: 2.56998e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 6.46e-06
        [updatestate_useless_node_eliminater]: 5.25001e-06
        [accelerated_algorithm]: 6.64001e-06
        [meta_shard_fg_expand]: 2.30002e-06
        [get_grad_eliminate_]: 5.95002e-06
        [merge_forward]: 4.21001e-06
        [cell_reuse_recompute_pass]: 1.07998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.534e-05
        [j_node_and_user_rematch]: 9.99999e-06
        [meta_fg_expand]: 2.36e-06
        [replace_old_param]: 9.47001e-06
        [inline_without_move]: 5.58002e-06
        [renormalize]: 0.00082994
        [add_forward_monad_depend]: 3.73999e-06
        [auto_monad_grad]: 1.04e-06
        [auto_monad_eliminator]: 8.40999e-06
        [cse]: 1.368e-05
        [replace_applicator]: 1.266e-05
    [Cycle 2]: 0.00032117, [27]
        [switch_simplify]: 5.47999e-06
        [loop_unroll]: 4.84003e-06
        [a_1]: 9.412e-05
        [with_stream_mark]: 7.34002e-06
        [recompute_prepare]: 5.00001e-06
        [updatestate_depend_eliminate]: 2.43e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 6.79982e-07
        [specialize_transform]: 5.13002e-06
        [updatestate_useless_node_eliminater]: 4.84998e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 9.50007e-07
        [get_grad_eliminate_]: 5.34998e-06
        [merge_forward]: 2.26e-06
        [cell_reuse_recompute_pass]: 2.31e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.174e-05
        [j_node_and_user_rematch]: 8e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 7.64002e-06
        [inline_without_move]: 4.94e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.04e-06
        [auto_monad_grad]: 5.69999e-07
        [auto_monad_eliminator]: 4.45e-06
        [cse]: 9.89999e-06
        [replace_applicator]: 4.97e-06
[py_interpret_to_execute_after_opt_a]: 9.92001e-06
[rewriter_after_opt_a]: 3.812e-05
[convert_after_rewriter]: 7.31999e-06
[order_py_execute_after_rewriter]: 4.94e-06
[mutable_eliminate]: 0.00069527
[jit_opt_b]: 5.159e-05, [1]
    [Cycle 1]: 4.453e-05, [2]
        [frontend_op_eliminate]: 1.767e-05
        [inline_after_opt_a]: 1.563e-05
[cconv]: 2.526e-05
[loop_unroll]: 0.00043302
[jit_opt_after_cconv]: 0.00014577, [1]
    [Cycle 1]: 0.00013864, [11]
        [c_1]: 2.144e-05
        [parameter_eliminate]: 1.57999e-06
        [updatestate_depend_eliminate]: 6.84001e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 2.265e-05
        [call_graph_tuple_transform]: 1.991e-05
        [tuple_list_get_item_eliminator]: 5.47999e-06
        [none_parameter_eliminate]: 6.69999e-07
        [renormalize]: 4.00003e-07
        [switch_simplify]: 5.45001e-06
[remove_dup_value]: 1.313e-05
[partial_unused_args_eliminate]: 1.62001e-06
[environ_conv]: 5.34e-06
[add_recomputation]: 6.604e-05
[cse_after_recomputation]: 2.49e-05, [1]
    [Cycle 1]: 1.792e-05, [1]
        [cse]: 1.078e-05
[auto_monad_reorder]: 1.705e-05
[get_jit_bprop_graph]: 2.75002e-06
[rewriter_after_jit_bprop_graph]: 8.64e-06
[opt_after_jit_grad]: 0.00049361
[symbol_engine_optimizer]: 7.68e-05, [1]
    [Cycle 1]: 6.962e-05, [6]
        [build]: 4.35e-06
        [elim_shapecalc]: 7.39002e-06
        [elim_not_effective]: 1.394e-05
        [opt_reshape]: 6.06e-06
        [fold_const_symbol]: 9.09e-06
        [renormalize]: 2.70025e-07
[validate]: 4.415e-05
Sums
    bootstrap                            :     0.000358s :  3.70%
    type_inference                       :     0.005131s : 52.96%
    event_method                         :     0.000020s :  0.21%
    auto_monad                           :     0.000069s :  0.71%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000042s :  0.43%
    rewriter_before_opt_a                :     0.000078s :  0.80%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000075s :  0.77%
    jit_opt_a.loop_unroll                :     0.000035s :  0.37%
    jit_opt_a.a_1                        :     0.000741s :  7.65%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.18%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000830s :  8.57%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.13%
    jit_opt_a.cse                        :     0.000024s :  0.24%
    jit_opt_a.replace_applicator         :     0.000018s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.10%
    rewriter_after_opt_a                 :     0.000038s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000695s :  7.18%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.16%
    cconv                                :     0.000025s :  0.26%
    loop_unroll                          :     0.000433s :  4.47%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000013s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000066s :  0.68%
    cse_after_recomputation.cse          :     0.000011s :  0.11%
    auto_monad_reorder                   :     0.000017s :  0.18%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.09%
    opt_after_jit_grad                   :     0.000494s :  5.10%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000044s :  0.46%


Time group info:
------[substitution.]   0.000268    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.16% :     0.000006s :      3: substitution.graph_param_transform
87.10% :     0.000234s :      4: substitution.inline
 1.35% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.70% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.48% :     0.000004s :      2: substitution.replace_old_param
 4.87% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005054     2
80.18% :     0.004052s :      1: type_inference.infer
19.82% :     0.001001s :      1: type_inference.specialize
------[replace.]   0.000084     6
68.84% :     0.000058s :      4: replace.inline
31.16% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000243     6
95.11% :     0.000231s :      4: match.inline
 4.89% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.24% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.52% :     0.000002s :     10: predicate.addn_zero_filter
 3.05% :     0.000004s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.50% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.04% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.31% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.41% :     0.000002s :     10: predicate.environ_get_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.15% :     0.000007s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000002s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 1.69% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.78% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.30% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 1.86% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 1.95% :     0.000002s :     15: predicate.partial_eliminate
 1.32% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.36% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.54% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.58% :     0.000002s :     10: predicate.reshape_eliminate
 1.06% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     16: predicate.switch_defer_inline
 1.88% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.54% :     0.000010s :     53: predicate.switch_simplify
 1.42% :     0.000002s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.64% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.77% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.84% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.65% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.11% :     0.000003s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.86% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000704    12
48.47% :     0.000341s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.85% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.69% :     0.000308s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.031870    72
 0.22% :     0.000070s :      1: add_recomputation
 0.23% :     0.000072s :      1: auto_monad
 0.06% :     0.000021s :      1: auto_monad_reorder
 1.17% :     0.000373s :      1: bootstrap
 0.09% :     0.000028s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000028s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000026s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
68.93% :     0.021969s :      1: jit_opt_a
 0.47% :     0.000148s :      1: jit_opt_after_cconv
 0.17% :     0.000054s :      1: jit_opt_b
 1.38% :     0.000440s :      1: loop_unroll
 2.21% :     0.000704s :      1: mutable_eliminate
 3.05% :     0.000971s :     26: opt.transform.jit_opt_a
 0.15% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000027s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.57% :     0.000500s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000045s :      1: py_interpret_to_execute
 0.04% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000016s :      1: remove_dup_value
 1.41% :     0.000449s :      1: renormalize.infer
 1.17% :     0.000373s :      1: renormalize.specialize
 0.03% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000041s :      1: rewriter_after_opt_a
 0.25% :     0.000081s :      1: rewriter_before_opt_a
 0.25% :     0.000079s :      1: symbol_engine_optimizer
16.15% :     0.005148s :      1: type_inference


TotalTime = 0.0271532, [33]
[bootstrap]: 0.0004373
[type_inference]: 0.00658007
[event_method]: 2.372e-05
[auto_monad]: 7.085e-05
[graph_reusing]: 6.28e-06
[pre_auto_parallel]: 2.44999e-06
[py_interpret_to_execute]: 4.332e-05
[rewriter_before_opt_a]: 7.47e-05
[expand_dump_flag]: 3.13e-06
[jit_opt_a]: 0.0059175, [2]
    [Cycle 1]: 0.00160833, [27]
        [switch_simplify]: 4.474e-05
        [loop_unroll]: 2.942e-05
        [a_1]: 0.00046351
        [with_stream_mark]: 1.085e-05
        [recompute_prepare]: 6.61e-06
        [updatestate_depend_eliminate]: 2.64999e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 8.29983e-07
        [specialize_transform]: 5.67999e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.57001e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 2.36998e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.066e-05
        [j_node_and_user_rematch]: 8.74e-06
        [meta_fg_expand]: 1.82999e-06
        [replace_old_param]: 7.73001e-06
        [inline_without_move]: 5.57001e-06
        [renormalize]: 0.00075295
        [add_forward_monad_depend]: 7.31001e-06
        [auto_monad_grad]: 3.25e-06
        [auto_monad_eliminator]: 1.554e-05
        [cse]: 3.084e-05
        [replace_applicator]: 1.401e-05
    [Cycle 2]: 0.00032159, [27]
        [switch_simplify]: 5.75001e-06
        [loop_unroll]: 5.02999e-06
        [a_1]: 9.083e-05
        [with_stream_mark]: 9.82999e-06
        [recompute_prepare]: 4.96002e-06
        [updatestate_depend_eliminate]: 2.75002e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.77999e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 4.99e-06
        [updatestate_useless_node_eliminater]: 4.71002e-06
        [accelerated_algorithm]: 4.86997e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 4.47998e-06
        [merge_forward]: 3.13e-06
        [cell_reuse_recompute_pass]: 2.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.255e-05
        [j_node_and_user_rematch]: 7.88999e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 7.41999e-06
        [inline_without_move]: 5.35999e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.36998e-06
        [auto_monad_grad]: 9.40025e-07
        [auto_monad_eliminator]: 4.70999e-06
        [cse]: 1.024e-05
        [replace_applicator]: 4.91002e-06
[py_interpret_to_execute_after_opt_a]: 1.191e-05
[rewriter_after_opt_a]: 3.86e-05
[convert_after_rewriter]: 6.18002e-06
[order_py_execute_after_rewriter]: 5.15999e-06
[mutable_eliminate]: 0.00069495
[jit_opt_b]: 4.881e-05, [1]
    [Cycle 1]: 4.186e-05, [2]
        [frontend_op_eliminate]: 1.624e-05
        [inline_after_opt_a]: 1.454e-05
[cconv]: 2.805e-05
[loop_unroll]: 0.00043303
[jit_opt_after_cconv]: 0.00014523, [1]
    [Cycle 1]: 0.00013852, [11]
        [c_1]: 1.992e-05
        [parameter_eliminate]: 2.56e-06
        [updatestate_depend_eliminate]: 8.43999e-06
        [updatestate_assign_eliminate]: 2.39999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 2.258e-05
        [call_graph_tuple_transform]: 2.037e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.43002e-06
        [renormalize]: 4.40021e-07
        [switch_simplify]: 5.30999e-06
[remove_dup_value]: 1.353e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 5.60001e-06
[add_recomputation]: 0.00052571
[cse_after_recomputation]: 4.015e-05, [1]
    [Cycle 1]: 3.302e-05, [1]
        [cse]: 2.056e-05
[auto_monad_reorder]: 1.278e-05
[get_jit_bprop_graph]: 1.67999e-06
[rewriter_after_jit_bprop_graph]: 6.91001e-06
[opt_after_jit_grad]: 0.00049293
[symbol_engine_optimizer]: 7.044e-05, [1]
    [Cycle 1]: 6.365e-05, [6]
        [build]: 2.86e-06
        [elim_shapecalc]: 7.17002e-06
        [elim_not_effective]: 1.184e-05
        [opt_reshape]: 5.37999e-06
        [fold_const_symbol]: 8.08999e-06
        [renormalize]: 6.89994e-07
[validate]: 2.869e-05
[backend_pass]: 8.2e-07
[task_emit]: 0.0111218
[execute]: 8.78001e-06
Sums
    bootstrap                            :     0.000437s :  1.94%
    type_inference                       :     0.006580s : 29.23%
    event_method                         :     0.000024s :  0.11%
    auto_monad                           :     0.000071s :  0.31%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.19%
    rewriter_before_opt_a                :     0.000075s :  0.33%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000050s :  0.22%
    jit_opt_a.loop_unroll                :     0.000034s :  0.15%
    jit_opt_a.a_1                        :     0.000554s :  2.46%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.07%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.000753s :  3.34%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.09%
    jit_opt_a.cse                        :     0.000041s :  0.18%
    jit_opt_a.replace_applicator         :     0.000019s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.05%
    rewriter_after_opt_a                 :     0.000039s :  0.17%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000695s :  3.09%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.06%
    cconv                                :     0.000028s :  0.12%
    loop_unroll                          :     0.000433s :  1.92%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000014s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000526s :  2.34%
    cse_after_recomputation.cse          :     0.000021s :  0.09%
    auto_monad_reorder                   :     0.000013s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.03%
    opt_after_jit_grad                   :     0.000493s :  2.19%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000029s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011122s : 49.40%
    execute                              :     0.000009s :  0.04%


Time group info:
------[substitution.]   0.000146    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 3.98% :     0.000006s :      3: substitution.graph_param_transform
82.52% :     0.000121s :      4: substitution.inline
 1.96% :     0.000003s :      4: substitution.j_node_and_user_rematch
 3.05% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.27% :     0.000003s :      2: substitution.replace_old_param
 4.60% :     0.000007s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006491     2
82.94% :     0.005384s :      1: type_inference.infer
17.06% :     0.001108s :      1: type_inference.specialize
------[replace.]   0.000058     6
64.18% :     0.000037s :      4: replace.inline
35.82% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000124     6
95.49% :     0.000119s :      4: match.inline
 4.51% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.70% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.24% :     0.000003s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.79% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 4.84% :     0.000006s :     22: predicate.inline
 1.43% :     0.000002s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.51% :     0.000002s :      6: predicate.less_batch_normalization
 1.79% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.91% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 2.05% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.04% :     0.000002s :     15: predicate.partial_eliminate
 1.31% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.53% :     0.000001s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.63% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.28% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.55% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.25% :     0.000008s :     53: predicate.switch_simplify
 1.43% :     0.000002s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.25% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 2.92% :     0.000003s :     18: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000709    12
51.00% :     0.000362s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 3.30% :     0.000023s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.70% :     0.000324s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028751    76
 1.86% :     0.000535s :      1: add_recomputation
 0.26% :     0.000074s :      1: auto_monad
 0.05% :     0.000016s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.62% :     0.000466s :      1: bootstrap
 0.11% :     0.000031s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000043s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000027s :      1: event_method
 0.05% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
20.60% :     0.005922s :      1: jit_opt_a
 0.51% :     0.000148s :      1: jit_opt_after_cconv
 0.18% :     0.000051s :      1: jit_opt_b
 1.53% :     0.000441s :      1: loop_unroll
 2.45% :     0.000704s :      1: mutable_eliminate
 2.62% :     0.000752s :     26: opt.transform.jit_opt_a
 0.17% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000029s :      4: opt.transform.symbol_engine_opt
 1.74% :     0.000500s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000046s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000016s :      1: remove_dup_value
 1.40% :     0.000402s :      1: renormalize.infer
 1.20% :     0.000345s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000043s :      1: rewriter_after_opt_a
 0.27% :     0.000078s :      1: rewriter_before_opt_a
 0.25% :     0.000073s :      1: symbol_engine_optimizer
38.76% :     0.011143s :      1: task_emit
22.97% :     0.006603s :      1: type_inference
 0.17% :     0.000048s :      1: validate


TotalTime = 0.0369345, [33]
[bootstrap]: 0.00037879
[type_inference]: 0.00519399
[event_method]: 2.076e-05
[auto_monad]: 6.797e-05
[graph_reusing]: 6.27001e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 3.543e-05
[rewriter_before_opt_a]: 7.51e-05
[expand_dump_flag]: 3.24001e-06
[jit_opt_a]: 0.00566679, [2]
    [Cycle 1]: 0.00174241, [27]
        [switch_simplify]: 6.27e-05
        [loop_unroll]: 2.978e-05
        [a_1]: 0.00057508
        [with_stream_mark]: 2.247e-05
        [recompute_prepare]: 7.26001e-06
        [updatestate_depend_eliminate]: 3.68e-06
        [updatestate_assign_eliminate]: 3.75e-06
        [updatestate_loads_eliminate]: 2.53003e-06
        [parameter_eliminate]: 2.25002e-06
        [specialize_transform]: 6.05002e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 5.67001e-06
        [merge_forward]: 3.9e-06
        [cell_reuse_recompute_pass]: 1.72001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.483e-05
        [j_node_and_user_rematch]: 9.93002e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 1.048e-05
        [inline_without_move]: 5.57999e-06
        [renormalize]: 0.00070368
        [add_forward_monad_depend]: 6.49999e-06
        [auto_monad_grad]: 2.29001e-06
        [auto_monad_eliminator]: 1.416e-05
        [cse]: 3.255e-05
        [replace_applicator]: 1.452e-05
    [Cycle 2]: 0.00032861, [27]
        [switch_simplify]: 5.77001e-06
        [loop_unroll]: 4.79998e-06
        [a_1]: 9.034e-05
        [with_stream_mark]: 1.038e-05
        [recompute_prepare]: 5.40001e-06
        [updatestate_depend_eliminate]: 3.15998e-06
        [updatestate_assign_eliminate]: 2.11998e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 4.93001e-06
        [updatestate_useless_node_eliminater]: 4.73001e-06
        [accelerated_algorithm]: 5.23002e-06
        [meta_shard_fg_expand]: 1.17999e-06
        [get_grad_eliminate_]: 4.41002e-06
        [merge_forward]: 2.59001e-06
        [cell_reuse_recompute_pass]: 2.34999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.365e-05
        [j_node_and_user_rematch]: 8.40001e-06
        [meta_fg_expand]: 1.41002e-06
        [replace_old_param]: 7.63001e-06
        [inline_without_move]: 4.87998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 1.22e-06
        [auto_monad_eliminator]: 6.51999e-06
        [cse]: 1.174e-05
        [replace_applicator]: 5.56e-06
[py_interpret_to_execute_after_opt_a]: 1.205e-05
[rewriter_after_opt_a]: 3.884e-05
[convert_after_rewriter]: 7.37002e-06
[order_py_execute_after_rewriter]: 4.80001e-06
[mutable_eliminate]: 0.00063248
[jit_opt_b]: 5.185e-05, [1]
    [Cycle 1]: 4.299e-05, [2]
        [frontend_op_eliminate]: 1.664e-05
        [inline_after_opt_a]: 1.536e-05
[cconv]: 2.8e-05
[loop_unroll]: 0.00042005
[jit_opt_after_cconv]: 0.00014494, [1]
    [Cycle 1]: 0.0001383, [11]
        [c_1]: 2.033e-05
        [parameter_eliminate]: 3.26999e-06
        [updatestate_depend_eliminate]: 6.47001e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 2.35002e-06
        [cse]: 2.09e-05
        [call_graph_tuple_transform]: 2.037e-05
        [tuple_list_get_item_eliminator]: 5.34e-06
        [none_parameter_eliminate]: 1.80001e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 5.14e-06
[remove_dup_value]: 1.488e-05
[partial_unused_args_eliminate]: 2.15002e-06
[environ_conv]: 5.07999e-06
[add_recomputation]: 5.409e-05
[cse_after_recomputation]: 2.144e-05, [1]
    [Cycle 1]: 1.553e-05, [1]
        [cse]: 9.82999e-06
[auto_monad_reorder]: 1.452e-05
[get_jit_bprop_graph]: 2.07999e-06
[rewriter_after_jit_bprop_graph]: 6.38998e-06
[opt_after_jit_grad]: 0.00044452
[symbol_engine_optimizer]: 7.36e-05, [1]
    [Cycle 1]: 6.685e-05, [6]
        [build]: 4.68001e-06
        [elim_shapecalc]: 7.71999e-06
        [elim_not_effective]: 1.266e-05
        [opt_reshape]: 5.55001e-06
        [fold_const_symbol]: 8.52998e-06
        [renormalize]: 6.80011e-07
[validate]: 3.956e-05
[backend_pass]: 9.49978e-07
[task_emit]: 0.023223
[execute]: 1.103e-05
Sums
    bootstrap                            :     0.000379s :  1.16%
    type_inference                       :     0.005194s : 15.89%
    event_method                         :     0.000021s :  0.06%
    auto_monad                           :     0.000068s :  0.21%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000035s :  0.11%
    rewriter_before_opt_a                :     0.000075s :  0.23%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000068s :  0.21%
    jit_opt_a.loop_unroll                :     0.000035s :  0.11%
    jit_opt_a.a_1                        :     0.000665s :  2.04%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000018s :  0.06%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.000704s :  2.15%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.06%
    jit_opt_a.cse                        :     0.000044s :  0.14%
    jit_opt_a.replace_applicator         :     0.000020s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000039s :  0.12%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000632s :  1.94%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.05%
    cconv                                :     0.000028s :  0.09%
    loop_unroll                          :     0.000420s :  1.29%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000015s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000054s :  0.17%
    cse_after_recomputation.cse          :     0.000010s :  0.03%
    auto_monad_reorder                   :     0.000015s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000445s :  1.36%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.023223s : 71.05%
    execute                              :     0.000011s :  0.03%


Time group info:
------[substitution.]   0.000241    23
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 2.60% :     0.000006s :      3: substitution.graph_param_transform
84.35% :     0.000203s :      4: substitution.inline
 1.56% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.89% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.15% :     0.000005s :      2: substitution.replace_old_param
 5.97% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005117     2
82.80% :     0.004237s :      1: type_inference.infer
17.20% :     0.000880s :      1: type_inference.specialize
------[replace.]   0.000069     6
67.30% :     0.000047s :      4: replace.inline
32.70% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000214     6
93.90% :     0.000201s :      4: match.inline
 6.10% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.24% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 2.68% :     0.000003s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.60% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.16% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.41% :     0.000000s :      3: predicate.graph_param_transform
 5.69% :     0.000007s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.66% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.05% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 2.92% :     0.000003s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.24% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.36% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.85% :     0.000002s :     18: predicate.replace_applicator
 0.98% :     0.000001s :      6: predicate.replace_old_param
 0.49% :     0.000001s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.13% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.57% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.69% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.25% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.21% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.47% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000635    12
49.70% :     0.000315s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.21% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.09% :     0.000267s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038630    76
 0.15% :     0.000057s :      1: add_recomputation
 0.18% :     0.000071s :      1: auto_monad
 0.04% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.05% :     0.000406s :      1: bootstrap
 0.08% :     0.000031s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.06% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.06% :     0.000025s :      1: event_method
 0.05% :     0.000018s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
14.68% :     0.005670s :      1: jit_opt_a
 0.38% :     0.000148s :      1: jit_opt_after_cconv
 0.14% :     0.000055s :      1: jit_opt_b
 1.11% :     0.000428s :      1: loop_unroll
 1.67% :     0.000643s :      1: mutable_eliminate
 2.30% :     0.000890s :     26: opt.transform.jit_opt_a
 0.12% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000025s :      4: opt.transform.jit_opt_b
 0.03% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.17% :     0.000453s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.10% :     0.000038s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000017s :      1: remove_dup_value
 0.97% :     0.000373s :      1: renormalize.infer
 0.83% :     0.000322s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000043s :      1: rewriter_after_opt_a
 0.20% :     0.000078s :      1: rewriter_before_opt_a
 0.20% :     0.000076s :      1: symbol_engine_optimizer
60.17% :     0.023242s :      1: task_emit
13.49% :     0.005211s :      1: type_inference
 0.16% :     0.000061s :      1: validate


TotalTime = 0.0280196, [33]
[bootstrap]: 0.00046194
[type_inference]: 0.00741598
[event_method]: 2.05e-05
[auto_monad]: 7.533e-05
[graph_reusing]: 6.44001e-06
[pre_auto_parallel]: 2.53003e-06
[py_interpret_to_execute]: 4.933e-05
[rewriter_before_opt_a]: 7.946e-05
[expand_dump_flag]: 3.47002e-06
[jit_opt_a]: 0.00834885, [2]
    [Cycle 1]: 0.00278976, [27]
        [switch_simplify]: 7.425e-05
        [loop_unroll]: 3.151e-05
        [a_1]: 0.00069139
        [with_stream_mark]: 2.772e-05
        [recompute_prepare]: 1.047e-05
        [updatestate_depend_eliminate]: 5.44e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.78e-06
        [parameter_eliminate]: 2.27001e-06
        [specialize_transform]: 7.64002e-06
        [updatestate_useless_node_eliminater]: 5.70001e-06
        [accelerated_algorithm]: 6.26e-06
        [meta_shard_fg_expand]: 2.81e-06
        [get_grad_eliminate_]: 6.07001e-06
        [merge_forward]: 4.60001e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.729e-05
        [j_node_and_user_rematch]: 1.113e-05
        [meta_fg_expand]: 2.91e-06
        [replace_old_param]: 1.073e-05
        [inline_without_move]: 5.43002e-06
        [renormalize]: 0.00154639
        [add_forward_monad_depend]: 1.015e-05
        [auto_monad_grad]: 2.67001e-06
        [auto_monad_eliminator]: 2.129e-05
        [cse]: 3.371e-05
        [replace_applicator]: 2.179e-05
    [Cycle 2]: 0.00037664, [27]
        [switch_simplify]: 6.33e-06
        [loop_unroll]: 5.32999e-06
        [a_1]: 0.00010866
        [with_stream_mark]: 1.661e-05
        [recompute_prepare]: 5.42999e-06
        [updatestate_depend_eliminate]: 3.83001e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [parameter_eliminate]: 1.47999e-06
        [specialize_transform]: 5.45001e-06
        [updatestate_useless_node_eliminater]: 5.45001e-06
        [accelerated_algorithm]: 5.28002e-06
        [meta_shard_fg_expand]: 1.81e-06
        [get_grad_eliminate_]: 4.80999e-06
        [merge_forward]: 3.28e-06
        [cell_reuse_recompute_pass]: 3.85e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.833e-05
        [j_node_and_user_rematch]: 8.85999e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 9.62001e-06
        [inline_without_move]: 5.27999e-06
        [renormalize]: 1.10012e-07
        [add_forward_monad_depend]: 1.52001e-06
        [auto_monad_grad]: 1.09998e-06
        [auto_monad_eliminator]: 7.26001e-06
        [cse]: 1.161e-05
        [replace_applicator]: 5.18002e-06
[py_interpret_to_execute_after_opt_a]: 7.932e-05
[rewriter_after_opt_a]: 4.214e-05
[convert_after_rewriter]: 7.63999e-06
[order_py_execute_after_rewriter]: 5.25999e-06
[mutable_eliminate]: 0.00085373
[jit_opt_b]: 6.183e-05, [1]
    [Cycle 1]: 5.254e-05, [2]
        [frontend_op_eliminate]: 1.962e-05
        [inline_after_opt_a]: 1.787e-05
[cconv]: 3.807e-05
[loop_unroll]: 0.00057275
[jit_opt_after_cconv]: 0.00019261, [1]
    [Cycle 1]: 0.00018391, [11]
        [c_1]: 2.509e-05
        [parameter_eliminate]: 5.74e-06
        [updatestate_depend_eliminate]: 1.162e-05
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [cse]: 3.625e-05
        [call_graph_tuple_transform]: 2.767e-05
        [tuple_list_get_item_eliminator]: 5.93002e-06
        [none_parameter_eliminate]: 1.69e-06
        [renormalize]: 1.37e-06
        [switch_simplify]: 6.21e-06
[remove_dup_value]: 1.685e-05
[partial_unused_args_eliminate]: 2.79001e-06
[environ_conv]: 7.21001e-06
[add_recomputation]: 5.695e-05
[cse_after_recomputation]: 2.773e-05, [1]
    [Cycle 1]: 2.089e-05, [1]
        [cse]: 1.278e-05
[auto_monad_reorder]: 1.778e-05
[get_jit_bprop_graph]: 2.59001e-06
[rewriter_after_jit_bprop_graph]: 9.84999e-06
[opt_after_jit_grad]: 0.00061298
[symbol_engine_optimizer]: 9.229e-05, [1]
    [Cycle 1]: 8.374e-05, [6]
        [build]: 7.97e-06
        [elim_shapecalc]: 8.97e-06
        [elim_not_effective]: 1.733e-05
        [opt_reshape]: 7.85998e-06
        [fold_const_symbol]: 1.055e-05
        [renormalize]: 1.09e-06
[validate]: 5.358e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.00851116
[execute]: 8.80001e-06
Sums
    bootstrap                            :     0.000462s :  2.09%
    type_inference                       :     0.007416s : 33.61%
    event_method                         :     0.000021s :  0.09%
    auto_monad                           :     0.000075s :  0.34%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000049s :  0.22%
    rewriter_before_opt_a                :     0.000079s :  0.36%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000081s :  0.37%
    jit_opt_a.loop_unroll                :     0.000037s :  0.17%
    jit_opt_a.a_1                        :     0.000800s :  3.63%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.001546s :  7.01%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000029s :  0.13%
    jit_opt_a.cse                        :     0.000045s :  0.21%
    jit_opt_a.replace_applicator         :     0.000027s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000079s :  0.36%
    rewriter_after_opt_a                 :     0.000042s :  0.19%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000854s :  3.87%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.08%
    cconv                                :     0.000038s :  0.17%
    loop_unroll                          :     0.000573s :  2.60%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000036s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000057s :  0.26%
    cse_after_recomputation.cse          :     0.000013s :  0.06%
    auto_monad_reorder                   :     0.000018s :  0.08%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.04%
    opt_after_jit_grad                   :     0.000613s :  2.78%
    symbol_engine_optimizer.build        :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000054s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008511s : 38.57%
    execute                              :     0.000009s :  0.04%


Time group info:
------[substitution.]   0.000299    23
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.87% :     0.000003s :      2: substitution.fold_const_symbol
 2.53% :     0.000008s :      3: substitution.graph_param_transform
84.76% :     0.000254s :      4: substitution.inline
 1.74% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.76% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.12% :     0.000006s :      2: substitution.replace_old_param
 5.52% :     0.000017s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007325     2
84.31% :     0.006176s :      1: type_inference.infer
15.69% :     0.001149s :      1: type_inference.specialize
------[replace.]   0.000095     6
66.32% :     0.000063s :      4: replace.inline
33.68% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000265     6
94.29% :     0.000250s :      4: match.inline
 5.71% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000147   691
 1.91% :     0.000003s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.83% :     0.000001s :     10: predicate.addn_check_dump
 1.15% :     0.000002s :     10: predicate.addn_zero_filter
 2.92% :     0.000004s :     10: predicate.arithmetic_simplify
 1.46% :     0.000002s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.85% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 0.95% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.56% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.66% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.23% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.84% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     10: predicate.environ_get_eliminate
 1.08% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 5.56% :     0.000008s :     22: predicate.inline
 0.72% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 2.03% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.52% :     0.000002s :     15: predicate.load_eliminater
 1.87% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.26% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.85% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 0.84% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 2.64% :     0.000004s :      3: predicate.mutable_eliminate
 0.35% :     0.000001s :      3: predicate.opt_reshape
 2.04% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.27% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.46% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000001s :      3: predicate.reset_defer_inline
 1.14% :     0.000002s :     10: predicate.reshape_eliminate
 1.09% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.99% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.10% :     0.000002s :      6: predicate.special_op_eliminate
 1.10% :     0.000002s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.31% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000004s :     16: predicate.switch_defer_inline
 2.22% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.80% :     0.000011s :     53: predicate.switch_simplify
 1.49% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000002s :     10: predicate.transpose_eliminate
 1.28% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.06% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.26% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.54% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000896    12
45.31% :     0.000406s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.48% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.21% :     0.000432s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.030754    76
 0.20% :     0.000061s :      1: add_recomputation
 0.26% :     0.000081s :      1: auto_monad
 0.07% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.58% :     0.000485s :      1: bootstrap
 0.13% :     0.000041s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000031s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.08% :     0.000026s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
27.16% :     0.008354s :      1: jit_opt_a
 0.64% :     0.000196s :      1: jit_opt_after_cconv
 0.21% :     0.000066s :      1: jit_opt_b
 1.90% :     0.000585s :      1: loop_unroll
 2.82% :     0.000869s :      1: mutable_eliminate
 3.44% :     0.001057s :     26: opt.transform.jit_opt_a
 0.20% :     0.000061s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000029s :      4: opt.transform.jit_opt_b
 0.06% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000024s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000041s :      4: opt.transform.symbol_engine_opt
 2.04% :     0.000626s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000054s :      1: py_interpret_to_execute
 0.27% :     0.000084s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000020s :      1: remove_dup_value
 3.36% :     0.001034s :      1: renormalize.infer
 1.62% :     0.000498s :      1: renormalize.specialize
 0.04% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000047s :      1: rewriter_after_opt_a
 0.27% :     0.000083s :      1: rewriter_before_opt_a
 0.31% :     0.000096s :      1: symbol_engine_optimizer
27.74% :     0.008530s :      1: task_emit
24.19% :     0.007440s :      1: type_inference
 0.26% :     0.000080s :      1: validate


TotalTime = 0.0328018, [30]
[bootstrap]: 0.000421
[type_inference]: 0.00574746
[event_method]: 2.017e-05
[auto_monad]: 7.018e-05
[graph_reusing]: 5.86e-06
[pre_auto_parallel]: 2.68003e-06
[py_interpret_to_execute]: 4.411e-05
[rewriter_before_opt_a]: 8.422e-05
[expand_dump_flag]: 3.63999e-06
[jit_opt_a]: 0.0238158, [2]
    [Cycle 1]: 0.0018239, [27]
        [switch_simplify]: 6.491e-05
        [loop_unroll]: 2.988e-05
        [a_1]: 0.00058864
        [with_stream_mark]: 1.603e-05
        [recompute_prepare]: 7.2e-06
        [updatestate_depend_eliminate]: 3.55e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 1.27999e-06
        [specialize_transform]: 6.56999e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 6.73e-06
        [meta_shard_fg_expand]: 1.56998e-06
        [get_grad_eliminate_]: 6.14999e-06
        [merge_forward]: 2.56e-06
        [cell_reuse_recompute_pass]: 1.63002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.165e-05
        [j_node_and_user_rematch]: 1.022e-05
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 8.59e-06
        [inline_without_move]: 5.59e-06
        [renormalize]: 0.00081718
        [add_forward_monad_depend]: 5.04998e-06
        [auto_monad_grad]: 1.44e-06
        [auto_monad_eliminator]: 1.089e-05
        [cse]: 1.905e-05
        [replace_applicator]: 1.296e-05
    [Cycle 2]: 0.00034564, [27]
        [switch_simplify]: 6.21998e-06
        [loop_unroll]: 5.00001e-06
        [a_1]: 9.836e-05
        [with_stream_mark]: 9.51e-06
        [recompute_prepare]: 5.17e-06
        [updatestate_depend_eliminate]: 2.48e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 5.26002e-06
        [updatestate_useless_node_eliminater]: 5.25001e-06
        [accelerated_algorithm]: 5.94e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.45e-06
        [merge_forward]: 2.32001e-06
        [cell_reuse_recompute_pass]: 2.06e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.272e-05
        [j_node_and_user_rematch]: 8.86997e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 8.24002e-06
        [inline_without_move]: 5.38002e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.31998e-06
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 6.90998e-06
        [cse]: 1.211e-05
        [replace_applicator]: 5.76e-06
[py_interpret_to_execute_after_opt_a]: 1.172e-05
[rewriter_after_opt_a]: 3.045e-05
[convert_after_rewriter]: 6.89999e-06
[order_py_execute_after_rewriter]: 4.42e-06
[mutable_eliminate]: 0.00091028
[jit_opt_b]: 6.092e-05, [1]
    [Cycle 1]: 5.247e-05, [2]
        [frontend_op_eliminate]: 2.316e-05
        [inline_after_opt_a]: 1.672e-05
[cconv]: 2.019e-05
[loop_unroll]: 0.00049117
[jit_opt_after_cconv]: 0.00015459, [1]
    [Cycle 1]: 0.00014771, [11]
        [c_1]: 2.342e-05
        [parameter_eliminate]: 3.11999e-06
        [updatestate_depend_eliminate]: 6.48998e-06
        [updatestate_assign_eliminate]: 2.50002e-06
        [updatestate_loads_eliminate]: 2.50002e-06
        [cse]: 2.128e-05
        [call_graph_tuple_transform]: 2.083e-05
        [tuple_list_get_item_eliminator]: 6.34001e-06
        [none_parameter_eliminate]: 1.02998e-06
        [renormalize]: 5.50004e-07
        [switch_simplify]: 6.34999e-06
[remove_dup_value]: 9.09e-06
[partial_unused_args_eliminate]: 1.67999e-06
[environ_conv]: 4.45e-06
[add_recomputation]: 3.71e-05
[cse_after_recomputation]: 2.399e-05, [1]
    [Cycle 1]: 1.797e-05, [1]
        [cse]: 1.093e-05
[auto_monad_reorder]: 1.21e-05
[get_jit_bprop_graph]: 1.87001e-06
[rewriter_after_jit_bprop_graph]: 9.74999e-06
[opt_after_jit_grad]: 0.00048718
[symbol_engine_optimizer]: 7.784e-05, [1]
    [Cycle 1]: 7.068e-05, [6]
        [build]: 3.82998e-06
        [elim_shapecalc]: 8.37998e-06
        [elim_not_effective]: 1.41e-05
        [opt_reshape]: 6.26998e-06
        [fold_const_symbol]: 8.57e-06
        [renormalize]: 4.30009e-07
[validate]: 3.402e-05
Sums
    bootstrap                            :     0.000421s :  4.00%
    type_inference                       :     0.005747s : 54.57%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000070s :  0.67%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000044s :  0.42%
    rewriter_before_opt_a                :     0.000084s :  0.80%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000071s :  0.68%
    jit_opt_a.loop_unroll                :     0.000035s :  0.33%
    jit_opt_a.a_1                        :     0.000687s :  6.52%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.16%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000817s :  7.76%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.17%
    jit_opt_a.cse                        :     0.000031s :  0.30%
    jit_opt_a.replace_applicator         :     0.000019s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.11%
    rewriter_after_opt_a                 :     0.000030s :  0.29%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000910s :  8.64%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.16%
    cconv                                :     0.000020s :  0.19%
    loop_unroll                          :     0.000491s :  4.66%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000009s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000037s :  0.35%
    cse_after_recomputation.cse          :     0.000011s :  0.10%
    auto_monad_reorder                   :     0.000012s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.09%
    opt_after_jit_grad                   :     0.000487s :  4.63%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.32%


Time group info:
------[substitution.]   0.000211    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.34% :     0.000005s :      3: substitution.graph_param_transform
86.02% :     0.000182s :      4: substitution.inline
 1.74% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.19% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.87% :     0.000004s :      2: substitution.replace_old_param
 4.36% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005664     2
81.21% :     0.004600s :      1: type_inference.infer
18.79% :     0.001064s :      1: type_inference.specialize
------[replace.]   0.000078     6
66.28% :     0.000052s :      4: replace.inline
33.72% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000187     6
95.75% :     0.000179s :      4: match.inline
 4.25% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000142   691
 2.00% :     0.000003s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.87% :     0.000001s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 3.06% :     0.000004s :     10: predicate.arithmetic_simplify
 1.50% :     0.000002s :     10: predicate.cast_eliminate
 0.35% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.83% :     0.000001s :     10: predicate.compare_switch_simplify
 1.63% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.51% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.34% :     0.000002s :     10: predicate.environ_get_eliminate
 1.13% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.23% :     0.000002s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 4.41% :     0.000006s :     22: predicate.inline
 0.77% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.62% :     0.000002s :      6: predicate.less_batch_normalization
 2.17% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000003s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.62% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.74% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.91% :     0.000001s :     10: predicate.merge_addn
 0.86% :     0.000001s :     10: predicate.minmaximum_grad
 1.96% :     0.000003s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 1.73% :     0.000002s :     15: predicate.partial_eliminate
 1.48% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.48% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000002s :      6: predicate.special_op_eliminate
 0.84% :     0.000001s :      6: predicate.specialize_transform
 1.74% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.08% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 1.99% :     0.000003s :     16: predicate.switch_defer_inline
 1.81% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.29% :     0.000012s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.08% :     0.000002s :     10: predicate.transpose_eliminate
 1.62% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.46% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.93% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.82% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.41% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.08% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.99% :     0.000003s :     10: predicate.value_based_eliminate
 0.37% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000732    12
48.62% :     0.000356s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.60% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.78% :     0.000342s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.034607    72
 0.12% :     0.000040s :      1: add_recomputation
 0.21% :     0.000074s :      1: auto_monad
 0.04% :     0.000015s :      1: auto_monad_reorder
 1.28% :     0.000441s :      1: bootstrap
 0.07% :     0.000023s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000027s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.07% :     0.000025s :      1: event_method
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
68.83% :     0.023821s :      1: jit_opt_a
 0.46% :     0.000158s :      1: jit_opt_after_cconv
 0.18% :     0.000064s :      1: jit_opt_b
 1.44% :     0.000500s :      1: loop_unroll
 2.66% :     0.000921s :      1: mutable_eliminate
 2.64% :     0.000915s :     26: opt.transform.jit_opt_a
 0.15% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000033s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.44% :     0.000497s :      1: opt_after_jit_grad
 0.02% :     0.000006s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000048s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000012s :      1: remove_dup_value
 1.26% :     0.000435s :      1: renormalize.infer
 1.08% :     0.000375s :      1: renormalize.specialize
 0.04% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000033s :      1: rewriter_after_opt_a
 0.25% :     0.000088s :      1: rewriter_before_opt_a
 0.23% :     0.000081s :      1: symbol_engine_optimizer
16.67% :     0.005767s :      1: type_inference


TotalTime = 0.0223551, [33]
[bootstrap]: 0.0004115
[type_inference]: 0.00561377
[event_method]: 1.866e-05
[auto_monad]: 6.605e-05
[graph_reusing]: 6.54999e-06
[pre_auto_parallel]: 2.58e-06
[py_interpret_to_execute]: 3.592e-05
[rewriter_before_opt_a]: 7.615e-05
[expand_dump_flag]: 3.26001e-06
[jit_opt_a]: 0.00625129, [2]
    [Cycle 1]: 0.00180624, [27]
        [switch_simplify]: 5.953e-05
        [loop_unroll]: 2.958e-05
        [a_1]: 0.00058477
        [with_stream_mark]: 2.383e-05
        [recompute_prepare]: 7.44002e-06
        [updatestate_depend_eliminate]: 4.3e-06
        [updatestate_assign_eliminate]: 3.23998e-06
        [updatestate_loads_eliminate]: 2.67001e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 6.24999e-06
        [updatestate_useless_node_eliminater]: 5.10001e-06
        [accelerated_algorithm]: 5.94e-06
        [meta_shard_fg_expand]: 2.32001e-06
        [get_grad_eliminate_]: 5.52999e-06
        [merge_forward]: 4.27e-06
        [cell_reuse_recompute_pass]: 1.55999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.453e-05
        [j_node_and_user_rematch]: 8.45999e-06
        [meta_fg_expand]: 2.43998e-06
        [replace_old_param]: 9.16002e-06
        [inline_without_move]: 5.28002e-06
        [renormalize]: 0.00077657
        [add_forward_monad_depend]: 6.38003e-06
        [auto_monad_grad]: 2.46e-06
        [auto_monad_eliminator]: 1.646e-05
        [cse]: 2.945e-05
        [replace_applicator]: 1.243e-05
    [Cycle 2]: 0.00032545, [27]
        [switch_simplify]: 5.38002e-06
        [loop_unroll]: 4.62e-06
        [a_1]: 9.521e-05
        [with_stream_mark]: 9.52999e-06
        [recompute_prepare]: 4.72998e-06
        [updatestate_depend_eliminate]: 2.58e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 4.92999e-06
        [updatestate_useless_node_eliminater]: 5.20001e-06
        [accelerated_algorithm]: 5.65001e-06
        [meta_shard_fg_expand]: 1.13001e-06
        [get_grad_eliminate_]: 4.58001e-06
        [merge_forward]: 2.32001e-06
        [cell_reuse_recompute_pass]: 1.92999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.284e-05
        [j_node_and_user_rematch]: 7.35e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 7.19001e-06
        [inline_without_move]: 4.51002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 7.90023e-07
        [auto_monad_eliminator]: 5.10001e-06
        [cse]: 1.13e-05
        [replace_applicator]: 4.94e-06
[py_interpret_to_execute_after_opt_a]: 1.259e-05
[rewriter_after_opt_a]: 3.951e-05
[convert_after_rewriter]: 7.53999e-06
[order_py_execute_after_rewriter]: 5.00001e-06
[mutable_eliminate]: 0.00070115
[jit_opt_b]: 5.078e-05, [1]
    [Cycle 1]: 4.301e-05, [2]
        [frontend_op_eliminate]: 1.696e-05
        [inline_after_opt_a]: 1.489e-05
[cconv]: 2.606e-05
[loop_unroll]: 0.00046619
[jit_opt_after_cconv]: 0.00017937, [1]
    [Cycle 1]: 0.00017158, [11]
        [c_1]: 2.247e-05
        [parameter_eliminate]: 2.94999e-06
        [updatestate_depend_eliminate]: 5.96e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 2.494e-05
        [call_graph_tuple_transform]: 1.919e-05
        [tuple_list_get_item_eliminator]: 5.03002e-06
        [none_parameter_eliminate]: 1.99e-06
        [renormalize]: 7.50006e-07
        [switch_simplify]: 1.771e-05
[remove_dup_value]: 1.395e-05
[partial_unused_args_eliminate]: 2.36998e-06
[environ_conv]: 5.92001e-06
[add_recomputation]: 5.39e-05
[cse_after_recomputation]: 2.14e-05, [1]
    [Cycle 1]: 1.507e-05, [1]
        [cse]: 9.25999e-06
[auto_monad_reorder]: 1.658e-05
[get_jit_bprop_graph]: 2.14999e-06
[rewriter_after_jit_bprop_graph]: 6.12999e-06
[opt_after_jit_grad]: 0.00053392
[symbol_engine_optimizer]: 7.743e-05, [1]
    [Cycle 1]: 7.001e-05, [6]
        [build]: 3.88001e-06
        [elim_shapecalc]: 7.97e-06
        [elim_not_effective]: 1.4e-05
        [opt_reshape]: 5.39e-06
        [fold_const_symbol]: 9.25001e-06
        [renormalize]: 6.09987e-07
[validate]: 4.233e-05
[backend_pass]: 1.97999e-06
[task_emit]: 0.00734915
[execute]: 9.31e-06
Sums
    bootstrap                            :     0.000411s :  2.34%
    type_inference                       :     0.005614s : 31.97%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000066s :  0.38%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000036s :  0.20%
    rewriter_before_opt_a                :     0.000076s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.37%
    jit_opt_a.loop_unroll                :     0.000034s :  0.19%
    jit_opt_a.a_1                        :     0.000680s :  3.87%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000777s :  4.42%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.12%
    jit_opt_a.cse                        :     0.000041s :  0.23%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000040s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000701s :  3.99%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000026s :  0.15%
    loop_unroll                          :     0.000466s :  2.65%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000018s :  0.10%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000054s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000534s :  3.04%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.24%
    backend_pass                         :     0.000002s :  0.01%
    task_emit                            :     0.007349s : 41.85%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000241    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.47% :     0.000006s :      3: substitution.graph_param_transform
85.40% :     0.000206s :      4: substitution.inline
 1.11% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.72% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000005s :      2: substitution.replace_old_param
 5.84% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005545     2
83.12% :     0.004609s :      1: type_inference.infer
16.88% :     0.000936s :      1: type_inference.specialize
------[replace.]   0.000071     6
68.57% :     0.000049s :      4: replace.inline
31.43% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000216     6
94.03% :     0.000203s :      4: match.inline
 5.97% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.51% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.20% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 2.13% :     0.000003s :     10: predicate.arithmetic_simplify
 1.28% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000001s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.12% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.53% :     0.000007s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.52% :     0.000002s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.06% :     0.000003s :     15: predicate.load_eliminater
 1.60% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.30% :     0.000002s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.36% :     0.000002s :      3: predicate.mutable_eliminate
 0.39% :     0.000000s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.26% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.82% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.46% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.55% :     0.000003s :     16: predicate.switch_defer_inline
 2.33% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.58% :     0.000009s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.65% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.45% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.52% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000670    12
50.07% :     0.000335s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.58% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.35% :     0.000284s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024121    76
 0.24% :     0.000057s :      1: add_recomputation
 0.29% :     0.000069s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 1.81% :     0.000436s :      1: bootstrap
 0.12% :     0.000029s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000023s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.93% :     0.006256s :      1: jit_opt_a
 0.76% :     0.000182s :      1: jit_opt_after_cconv
 0.22% :     0.000053s :      1: jit_opt_b
 1.97% :     0.000475s :      1: loop_unroll
 2.94% :     0.000710s :      1: mutable_eliminate
 3.70% :     0.000892s :     26: opt.transform.jit_opt_a
 0.25% :     0.000061s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.25% :     0.000543s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000039s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.81% :     0.000436s :      1: renormalize.infer
 1.38% :     0.000332s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000042s :      1: rewriter_after_opt_a
 0.33% :     0.000080s :      1: rewriter_before_opt_a
 0.33% :     0.000080s :      1: symbol_engine_optimizer
30.55% :     0.007369s :      1: task_emit
23.33% :     0.005627s :      1: type_inference
 0.28% :     0.000067s :      1: validate


TotalTime = 0.0331731, [33]
[bootstrap]: 0.00032733
[type_inference]: 0.0052327
[event_method]: 2.061e-05
[auto_monad]: 6.543e-05
[graph_reusing]: 6.72002e-06
[pre_auto_parallel]: 2.54999e-06
[py_interpret_to_execute]: 3.536e-05
[rewriter_before_opt_a]: 7.289e-05
[expand_dump_flag]: 3.02002e-06
[jit_opt_a]: 0.00575108, [2]
    [Cycle 1]: 0.00187058, [27]
        [switch_simplify]: 6.632e-05
        [loop_unroll]: 3.018e-05
        [a_1]: 0.00059192
        [with_stream_mark]: 2.412e-05
        [recompute_prepare]: 7.22002e-06
        [updatestate_depend_eliminate]: 4.82e-06
        [updatestate_assign_eliminate]: 2.80002e-06
        [updatestate_loads_eliminate]: 3.04999e-06
        [parameter_eliminate]: 2.54001e-06
        [specialize_transform]: 6.28e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 6.19999e-06
        [meta_shard_fg_expand]: 2.21e-06
        [get_grad_eliminate_]: 5.47999e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.564e-05
        [j_node_and_user_rematch]: 1.055e-05
        [meta_fg_expand]: 2.11998e-06
        [replace_old_param]: 8.99e-06
        [inline_without_move]: 5.87999e-06
        [renormalize]: 0.00075101
        [add_forward_monad_depend]: 6.29001e-06
        [auto_monad_grad]: 2.76e-06
        [auto_monad_eliminator]: 1.421e-05
        [cse]: 2.839e-05
        [replace_applicator]: 1.341e-05
    [Cycle 2]: 0.00033305, [27]
        [switch_simplify]: 5.86e-06
        [loop_unroll]: 5.05999e-06
        [a_1]: 9.687e-05
        [with_stream_mark]: 1.016e-05
        [recompute_prepare]: 5.87001e-06
        [updatestate_depend_eliminate]: 3.01999e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 2.27001e-06
        [parameter_eliminate]: 1.04003e-06
        [specialize_transform]: 5.10999e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.17e-06
        [meta_shard_fg_expand]: 1.50001e-06
        [get_grad_eliminate_]: 4.59002e-06
        [merge_forward]: 2.75002e-06
        [cell_reuse_recompute_pass]: 2.21998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.308e-05
        [j_node_and_user_rematch]: 8.50999e-06
        [meta_fg_expand]: 1.77001e-06
        [replace_old_param]: 7.61999e-06
        [inline_without_move]: 5.04e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 9.50007e-07
        [auto_monad_grad]: 1.54e-06
        [auto_monad_eliminator]: 4.68999e-06
        [cse]: 1.036e-05
        [replace_applicator]: 5.71998e-06
[py_interpret_to_execute_after_opt_a]: 1.176e-05
[rewriter_after_opt_a]: 3.518e-05
[convert_after_rewriter]: 7.11001e-06
[order_py_execute_after_rewriter]: 4.55999e-06
[mutable_eliminate]: 0.00063076
[jit_opt_b]: 5.151e-05, [1]
    [Cycle 1]: 4.337e-05, [2]
        [frontend_op_eliminate]: 1.59e-05
        [inline_after_opt_a]: 1.556e-05
[cconv]: 2.811e-05
[loop_unroll]: 0.00043051
[jit_opt_after_cconv]: 0.00014382, [1]
    [Cycle 1]: 0.0001372, [11]
        [c_1]: 2.11e-05
        [parameter_eliminate]: 2.91999e-06
        [updatestate_depend_eliminate]: 6.88998e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 2.114e-05
        [call_graph_tuple_transform]: 1.966e-05
        [tuple_list_get_item_eliminator]: 5.34998e-06
        [none_parameter_eliminate]: 1.36002e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 5.05999e-06
[remove_dup_value]: 1.417e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 5.27001e-06
[add_recomputation]: 4.968e-05
[cse_after_recomputation]: 2.072e-05, [1]
    [Cycle 1]: 1.525e-05, [1]
        [cse]: 9.48002e-06
[auto_monad_reorder]: 1.604e-05
[get_jit_bprop_graph]: 1.96e-06
[rewriter_after_jit_bprop_graph]: 5.24e-06
[opt_after_jit_grad]: 0.00045741
[symbol_engine_optimizer]: 7.574e-05, [1]
    [Cycle 1]: 6.957e-05, [6]
        [build]: 4.78001e-06
        [elim_shapecalc]: 7.69002e-06
        [elim_not_effective]: 1.29e-05
        [opt_reshape]: 6.41e-06
        [fold_const_symbol]: 9.27001e-06
        [renormalize]: 5.89993e-07
[validate]: 3.775e-05
[backend_pass]: 1.12e-06
[task_emit]: 0.0193886
[execute]: 9.97999e-06
Sums
    bootstrap                            :     0.000327s :  1.13%
    type_inference                       :     0.005233s : 18.10%
    event_method                         :     0.000021s :  0.07%
    auto_monad                           :     0.000065s :  0.23%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000035s :  0.12%
    rewriter_before_opt_a                :     0.000073s :  0.25%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000072s :  0.25%
    jit_opt_a.loop_unroll                :     0.000035s :  0.12%
    jit_opt_a.a_1                        :     0.000689s :  2.38%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000017s :  0.06%
    jit_opt_a.inline_without_move        :     0.000011s :  0.04%
    jit_opt_a.renormalize                :     0.000751s :  2.60%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.07%
    jit_opt_a.cse                        :     0.000039s :  0.13%
    jit_opt_a.replace_applicator         :     0.000019s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000035s :  0.12%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000631s :  2.18%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.05%
    cconv                                :     0.000028s :  0.10%
    loop_unroll                          :     0.000431s :  1.49%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000014s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000050s :  0.17%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000457s :  1.58%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.019389s : 67.05%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.000243    23
 0.77% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.13% :     0.000005s :      3: substitution.graph_param_transform
86.40% :     0.000210s :      4: substitution.inline
 1.49% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.07% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.44% :     0.000004s :      2: substitution.replace_old_param
 5.15% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005157     2
82.05% :     0.004232s :      1: type_inference.infer
17.95% :     0.000926s :      1: type_inference.specialize
------[replace.]   0.000073     6
67.82% :     0.000050s :      4: replace.inline
32.18% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000219     6
94.81% :     0.000208s :      4: match.inline
 5.19% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.45% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000001s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.30% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 6.14% :     0.000007s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.40% :     0.000002s :      6: predicate.less_batch_normalization
 1.67% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 2.39% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.12% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.78% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.40% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.43% :     0.000001s :      3: predicate.reset_defer_inline
 1.23% :     0.000001s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.54% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.22% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.14% :     0.000010s :     53: predicate.switch_simplify
 1.08% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.24% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.90% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.32% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000658    12
47.22% :     0.000311s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.00% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.78% :     0.000295s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.034945    76
 0.15% :     0.000053s :      1: add_recomputation
 0.20% :     0.000069s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.00% :     0.000350s :      1: bootstrap
 0.09% :     0.000031s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.07% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.07% :     0.000025s :      1: event_method
 0.05% :     0.000017s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
16.47% :     0.005754s :      1: jit_opt_a
 0.42% :     0.000147s :      1: jit_opt_after_cconv
 0.15% :     0.000054s :      1: jit_opt_b
 1.25% :     0.000438s :      1: loop_unroll
 1.83% :     0.000640s :      1: mutable_eliminate
 2.63% :     0.000918s :     26: opt.transform.jit_opt_a
 0.14% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000025s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.33% :     0.000466s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.11% :     0.000038s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000016s :      1: remove_dup_value
 1.14% :     0.000399s :      1: renormalize.infer
 0.98% :     0.000343s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000038s :      1: rewriter_after_opt_a
 0.22% :     0.000076s :      1: rewriter_before_opt_a
 0.22% :     0.000078s :      1: symbol_engine_optimizer
55.54% :     0.019410s :      1: task_emit
15.04% :     0.005254s :      1: type_inference
 0.17% :     0.000058s :      1: validate


TotalTime = 0.0177186, [30]
[bootstrap]: 0.00045915
[type_inference]: 0.00555858
[event_method]: 1.79e-05
[auto_monad]: 7.061e-05
[graph_reusing]: 6.01e-06
[pre_auto_parallel]: 2.67001e-06
[py_interpret_to_execute]: 3.792e-05
[rewriter_before_opt_a]: 7.723e-05
[expand_dump_flag]: 3.68e-06
[jit_opt_a]: 0.00782563, [2]
    [Cycle 1]: 0.00247392, [27]
        [switch_simplify]: 7.83e-05
        [loop_unroll]: 3.246e-05
        [a_1]: 0.00069283
        [with_stream_mark]: 2.761e-05
        [recompute_prepare]: 1.245e-05
        [updatestate_depend_eliminate]: 4.52e-06
        [updatestate_assign_eliminate]: 3.19001e-06
        [updatestate_loads_eliminate]: 2.75997e-06
        [parameter_eliminate]: 3.19001e-06
        [specialize_transform]: 8.22e-06
        [updatestate_useless_node_eliminater]: 6.06998e-06
        [accelerated_algorithm]: 1.189e-05
        [meta_shard_fg_expand]: 3.24001e-06
        [get_grad_eliminate_]: 6.43e-06
        [merge_forward]: 5.19e-06
        [cell_reuse_recompute_pass]: 2.99999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.119e-05
        [j_node_and_user_rematch]: 1.108e-05
        [meta_fg_expand]: 2.98998e-06
        [replace_old_param]: 1.285e-05
        [inline_without_move]: 7.45e-06
        [renormalize]: 0.00115867
        [add_forward_monad_depend]: 1.204e-05
        [auto_monad_grad]: 3.43e-06
        [auto_monad_eliminator]: 2.568e-05
        [cse]: 3.872e-05
        [replace_applicator]: 2.469e-05
    [Cycle 2]: 0.00042498, [27]
        [switch_simplify]: 7.03998e-06
        [loop_unroll]: 6.14001e-06
        [a_1]: 0.00011522
        [with_stream_mark]: 2.122e-05
        [recompute_prepare]: 6.34001e-06
        [updatestate_depend_eliminate]: 4.23001e-06
        [updatestate_assign_eliminate]: 3.07002e-06
        [updatestate_loads_eliminate]: 2.55002e-06
        [parameter_eliminate]: 2.43e-06
        [specialize_transform]: 6.05002e-06
        [updatestate_useless_node_eliminater]: 6.11e-06
        [accelerated_algorithm]: 8.09002e-06
        [meta_shard_fg_expand]: 2.99001e-06
        [get_grad_eliminate_]: 6.31998e-06
        [merge_forward]: 3.93001e-06
        [cell_reuse_recompute_pass]: 4.48999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.916e-05
        [j_node_and_user_rematch]: 1.144e-05
        [meta_fg_expand]: 2.68998e-06
        [replace_old_param]: 1.158e-05
        [inline_without_move]: 5.52999e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 2.50002e-06
        [auto_monad_grad]: 1.32e-06
        [auto_monad_eliminator]: 8.08999e-06
        [cse]: 1.7e-05
        [replace_applicator]: 7.05002e-06
[py_interpret_to_execute_after_opt_a]: 1.916e-05
[rewriter_after_opt_a]: 4.9e-05
[convert_after_rewriter]: 9.16998e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.00109662
[jit_opt_b]: 8.712e-05, [1]
    [Cycle 1]: 7.452e-05, [2]
        [frontend_op_eliminate]: 3.587e-05
        [inline_after_opt_a]: 2.287e-05
[cconv]: 3.718e-05
[loop_unroll]: 0.0007115
[jit_opt_after_cconv]: 0.00021244, [1]
    [Cycle 1]: 0.00020178, [11]
        [c_1]: 3.108e-05
        [parameter_eliminate]: 6.48e-06
        [updatestate_depend_eliminate]: 1.068e-05
        [updatestate_assign_eliminate]: 3.75e-06
        [updatestate_loads_eliminate]: 3.11001e-06
        [cse]: 4.002e-05
        [call_graph_tuple_transform]: 2.911e-05
        [tuple_list_get_item_eliminator]: 7.1e-06
        [none_parameter_eliminate]: 1.71e-06
        [renormalize]: 8.09989e-07
        [switch_simplify]: 7.55e-06
[remove_dup_value]: 1.878e-05
[partial_unused_args_eliminate]: 3.28998e-06
[environ_conv]: 6.09999e-06
[add_recomputation]: 6.484e-05
[cse_after_recomputation]: 2.865e-05, [1]
    [Cycle 1]: 2.107e-05, [1]
        [cse]: 1.204e-05
[auto_monad_reorder]: 1.826e-05
[get_jit_bprop_graph]: 2.65002e-06
[rewriter_after_jit_bprop_graph]: 1.19e-05
[opt_after_jit_grad]: 0.00084738
[symbol_engine_optimizer]: 0.0001013, [1]
    [Cycle 1]: 8.823e-05, [6]
        [build]: 6.76e-06
        [elim_shapecalc]: 8.92e-06
        [elim_not_effective]: 1.882e-05
        [opt_reshape]: 7.31999e-06
        [fold_const_symbol]: 1.139e-05
        [renormalize]: 3.39991e-07
[validate]: 5.227e-05
Sums
    bootstrap                            :     0.000459s :  3.84%
    type_inference                       :     0.005559s : 46.45%
    event_method                         :     0.000018s :  0.15%
    auto_monad                           :     0.000071s :  0.59%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.32%
    rewriter_before_opt_a                :     0.000077s :  0.65%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000085s :  0.71%
    jit_opt_a.loop_unroll                :     0.000039s :  0.32%
    jit_opt_a.a_1                        :     0.000808s :  6.75%
    jit_opt_a.with_stream_mark           :     0.000049s :  0.41%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.05%
    jit_opt_a.specialize_transform       :     0.000014s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000020s :  0.17%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000013s :  0.11%
    jit_opt_a.merge_forward              :     0.000009s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000023s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.05%
    jit_opt_a.replace_old_param          :     0.000024s :  0.20%
    jit_opt_a.inline_without_move        :     0.000013s :  0.11%
    jit_opt_a.renormalize                :     0.001159s :  9.68%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.12%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000034s :  0.28%
    jit_opt_a.cse                        :     0.000056s :  0.47%
    jit_opt_a.replace_applicator         :     0.000032s :  0.27%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.16%
    rewriter_after_opt_a                 :     0.000049s :  0.41%
    convert_after_rewriter               :     0.000009s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.001097s :  9.16%
    jit_opt_b.frontend_op_eliminate      :     0.000036s :  0.30%
    jit_opt_b.inline_after_opt_a         :     0.000023s :  0.19%
    cconv                                :     0.000037s :  0.31%
    loop_unroll                          :     0.000711s :  5.95%
    jit_opt_after_cconv.c_1              :     0.000031s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.09%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000040s :  0.33%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000029s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000008s :  0.06%
    remove_dup_value                     :     0.000019s :  0.16%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000065s :  0.54%
    cse_after_recomputation.cse          :     0.000012s :  0.10%
    auto_monad_reorder                   :     0.000018s :  0.15%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000012s :  0.10%
    opt_after_jit_grad                   :     0.000847s :  7.08%
    symbol_engine_optimizer.build        :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000052s :  0.44%


Time group info:
------[substitution.]   0.000291    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000002s :      2: substitution.fold_const_symbol
 2.37% :     0.000007s :      3: substitution.graph_param_transform
83.10% :     0.000242s :      4: substitution.inline
 1.60% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.29% :     0.000007s :      4: substitution.remove_not_recompute_node
 3.17% :     0.000009s :      2: substitution.replace_old_param
 5.85% :     0.000017s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005476     2
82.58% :     0.004522s :      1: type_inference.infer
17.42% :     0.000954s :      1: type_inference.specialize
------[replace.]   0.000093     6
65.70% :     0.000061s :      4: replace.inline
34.30% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000255     6
93.92% :     0.000239s :      4: match.inline
 6.08% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000164   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.92% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000002s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.24% :     0.000004s :     10: predicate.arithmetic_simplify
 1.20% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000002s :     10: predicate.compare_switch_simplify
 1.49% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.65% :     0.000003s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.49% :     0.000001s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.03% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.76% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.77% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     10: predicate.environ_get_eliminate
 0.91% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000002s :      6: predicate.get_grad_eliminate
 0.31% :     0.000001s :      3: predicate.graph_param_transform
 4.92% :     0.000008s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.29% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.65% :     0.000003s :      6: predicate.less_batch_normalization
 1.54% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.36% :     0.000002s :     15: predicate.load_eliminater
 2.24% :     0.000004s :      3: predicate.loop_unroll_after_grad
 3.15% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.76% :     0.000001s :     10: predicate.merge_addn
 0.90% :     0.000001s :     10: predicate.minmaximum_grad
 2.57% :     0.000004s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 1.62% :     0.000003s :     15: predicate.partial_eliminate
 1.51% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.85% :     0.000003s :     10: predicate.reduce_eliminate
 2.18% :     0.000004s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000003s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.72% :     0.000003s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     10: predicate.same_eliminate
 0.71% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000002s :      6: predicate.special_op_eliminate
 0.82% :     0.000001s :      6: predicate.specialize_transform
 1.15% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.12% :     0.000005s :     16: predicate.switch_defer_inline
 1.51% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.41% :     0.000012s :     53: predicate.switch_simplify
 1.43% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000002s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.67% :     0.000008s :     18: predicate.tuple_list_get_item_eliminator
 1.98% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.28% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.46% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.18% :     0.000004s :     10: predicate.value_based_eliminate
 0.24% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.81% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000790    12
48.59% :     0.000384s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.64% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.77% :     0.000369s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020064    72
 0.35% :     0.000070s :      1: add_recomputation
 0.37% :     0.000074s :      1: auto_monad
 0.11% :     0.000022s :      1: auto_monad_reorder
 2.41% :     0.000484s :      1: bootstrap
 0.20% :     0.000041s :      1: cconv
 0.06% :     0.000012s :      1: convert_after_rewriter
 0.16% :     0.000031s :      1: cse_after_recomputation
 0.05% :     0.000010s :      1: environ_conv
 0.11% :     0.000023s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
39.03% :     0.007831s :      1: jit_opt_a
 1.08% :     0.000216s :      1: jit_opt_after_cconv
 0.45% :     0.000091s :      1: jit_opt_b
 3.61% :     0.000725s :      1: loop_unroll
 5.56% :     0.001115s :      1: mutable_eliminate
 5.48% :     0.001100s :     26: opt.transform.jit_opt_a
 0.35% :     0.000069s :      4: opt.transform.jit_opt_after_cconv
 0.24% :     0.000049s :      4: opt.transform.jit_opt_b
 0.11% :     0.000023s :      1: opt.transform.loop_unroll_optimizer
 0.14% :     0.000028s :      1: opt.transform.mutable_eliminate
 0.19% :     0.000038s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000042s :      4: opt.transform.symbol_engine_opt
 4.33% :     0.000869s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000040s :      1: py_interpret_to_execute
 0.12% :     0.000024s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000022s :      1: remove_dup_value
 2.93% :     0.000589s :      1: renormalize.infer
 2.77% :     0.000557s :      1: renormalize.specialize
 0.07% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000054s :      1: rewriter_after_opt_a
 0.40% :     0.000080s :      1: rewriter_before_opt_a
 0.52% :     0.000105s :      1: symbol_engine_optimizer
27.81% :     0.005579s :      1: type_inference


TotalTime = 0.029706, [33]
[bootstrap]: 0.00045661
[type_inference]: 0.00656956
[event_method]: 1.96e-05
[auto_monad]: 7.292e-05
[graph_reusing]: 6.51999e-06
[pre_auto_parallel]: 3.04001e-06
[py_interpret_to_execute]: 4.136e-05
[rewriter_before_opt_a]: 7.868e-05
[expand_dump_flag]: 3.49001e-06
[jit_opt_a]: 0.0109246, [2]
    [Cycle 1]: 0.00376931, [27]
        [switch_simplify]: 7.547e-05
        [loop_unroll]: 3.022e-05
        [a_1]: 0.00070419
        [with_stream_mark]: 2.684e-05
        [recompute_prepare]: 1.246e-05
        [updatestate_depend_eliminate]: 4.58001e-06
        [updatestate_assign_eliminate]: 3.4e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [parameter_eliminate]: 2.57001e-06
        [specialize_transform]: 7.78999e-06
        [updatestate_useless_node_eliminater]: 5.57999e-06
        [accelerated_algorithm]: 7.00002e-06
        [meta_shard_fg_expand]: 2.72001e-06
        [get_grad_eliminate_]: 6.04999e-06
        [merge_forward]: 4.37e-06
        [cell_reuse_recompute_pass]: 2.43002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.211e-05
        [j_node_and_user_rematch]: 1.162e-05
        [meta_fg_expand]: 2.42001e-06
        [replace_old_param]: 1.093e-05
        [inline_without_move]: 6.64999e-06
        [renormalize]: 0.00249021
        [add_forward_monad_depend]: 1.246e-05
        [auto_monad_grad]: 2.51e-06
        [auto_monad_eliminator]: 2.545e-05
        [cse]: 3.267e-05
        [replace_applicator]: 2.434e-05
    [Cycle 2]: 0.00040763, [27]
        [switch_simplify]: 6.39001e-06
        [loop_unroll]: 5.34e-06
        [a_1]: 0.0001137
        [with_stream_mark]: 1.981e-05
        [recompute_prepare]: 6.14001e-06
        [updatestate_depend_eliminate]: 3.95e-06
        [updatestate_assign_eliminate]: 3.06999e-06
        [updatestate_loads_eliminate]: 2.81999e-06
        [parameter_eliminate]: 2.27999e-06
        [specialize_transform]: 5.56002e-06
        [updatestate_useless_node_eliminater]: 5.49e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 2.61e-06
        [get_grad_eliminate_]: 4.94e-06
        [merge_forward]: 4.1e-06
        [cell_reuse_recompute_pass]: 4.56002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.14e-05
        [j_node_and_user_rematch]: 9.46e-06
        [meta_fg_expand]: 2.78e-06
        [replace_old_param]: 9.61e-06
        [inline_without_move]: 5.38002e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 2.37001e-06
        [auto_monad_grad]: 9.09989e-07
        [auto_monad_eliminator]: 7.89002e-06
        [cse]: 1.351e-05
        [replace_applicator]: 5.37999e-06
[py_interpret_to_execute_after_opt_a]: 2.14e-05
[rewriter_after_opt_a]: 4.417e-05
[convert_after_rewriter]: 9.67999e-06
[order_py_execute_after_rewriter]: 5.86e-06
[mutable_eliminate]: 0.00088883
[jit_opt_b]: 5.695e-05, [1]
    [Cycle 1]: 4.707e-05, [2]
        [frontend_op_eliminate]: 1.839e-05
        [inline_after_opt_a]: 1.55e-05
[cconv]: 3.564e-05
[loop_unroll]: 0.00057873
[jit_opt_after_cconv]: 0.00016991, [1]
    [Cycle 1]: 0.00016154, [11]
        [c_1]: 2.397e-05
        [parameter_eliminate]: 4.75001e-06
        [updatestate_depend_eliminate]: 1.03e-05
        [updatestate_assign_eliminate]: 3.04999e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [cse]: 2.824e-05
        [call_graph_tuple_transform]: 2.115e-05
        [tuple_list_get_item_eliminator]: 5.26998e-06
        [none_parameter_eliminate]: 1.71e-06
        [renormalize]: 6.39993e-07
        [switch_simplify]: 5.56002e-06
[remove_dup_value]: 1.594e-05
[partial_unused_args_eliminate]: 2.70997e-06
[environ_conv]: 6.74999e-06
[add_recomputation]: 5.924e-05
[cse_after_recomputation]: 2.333e-05, [1]
    [Cycle 1]: 1.694e-05, [1]
        [cse]: 1.019e-05
[auto_monad_reorder]: 1.623e-05
[get_jit_bprop_graph]: 2.64001e-06
[rewriter_after_jit_bprop_graph]: 8.22998e-06
[opt_after_jit_grad]: 0.0005928
[symbol_engine_optimizer]: 8.244e-05, [1]
    [Cycle 1]: 7.421e-05, [6]
        [build]: 5.40999e-06
        [elim_shapecalc]: 8.69e-06
        [elim_not_effective]: 1.412e-05
        [opt_reshape]: 6.94999e-06
        [fold_const_symbol]: 9.57999e-06
        [renormalize]: 8.59989e-07
[validate]: 4.604e-05
[backend_pass]: 9.60019e-07
[task_emit]: 0.00859677
[execute]: 7.82998e-06
Sums
    bootstrap                            :     0.000457s :  2.06%
    type_inference                       :     0.006570s : 29.59%
    event_method                         :     0.000020s :  0.09%
    auto_monad                           :     0.000073s :  0.33%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.19%
    rewriter_before_opt_a                :     0.000079s :  0.35%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000082s :  0.37%
    jit_opt_a.loop_unroll                :     0.000036s :  0.16%
    jit_opt_a.a_1                        :     0.000818s :  3.68%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000044s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.09%
    jit_opt_a.inline_without_move        :     0.000012s :  0.05%
    jit_opt_a.renormalize                :     0.002490s : 11.22%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000033s :  0.15%
    jit_opt_a.cse                        :     0.000046s :  0.21%
    jit_opt_a.replace_applicator         :     0.000030s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000021s :  0.10%
    rewriter_after_opt_a                 :     0.000044s :  0.20%
    convert_after_rewriter               :     0.000010s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000889s :  4.00%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.07%
    cconv                                :     0.000036s :  0.16%
    loop_unroll                          :     0.000579s :  2.61%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.07%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000059s :  0.27%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.07%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000593s :  2.67%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.21%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008597s : 38.72%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000313    23
 0.57% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000002s :      2: substitution.fold_const_symbol
 2.01% :     0.000006s :      3: substitution.graph_param_transform
85.89% :     0.000269s :      4: substitution.inline
 1.49% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.17% :     0.000007s :      4: substitution.remove_not_recompute_node
 2.22% :     0.000007s :      2: substitution.replace_old_param
 5.09% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006477     2
81.90% :     0.005304s :      1: type_inference.infer
18.10% :     0.001172s :      1: type_inference.specialize
------[replace.]   0.000096     6
66.76% :     0.000064s :      4: replace.inline
33.24% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000280     6
94.75% :     0.000266s :      4: match.inline
 5.25% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000152   691
 1.43% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.69% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.81% :     0.000001s :     10: predicate.addn_check_dump
 1.59% :     0.000002s :     10: predicate.addn_zero_filter
 3.44% :     0.000005s :     10: predicate.arithmetic_simplify
 1.15% :     0.000002s :     10: predicate.cast_eliminate
 0.56% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.81% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000002s :     10: predicate.depend_value_elim
 0.87% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.81% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.61% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.00% :     0.000002s :     10: predicate.environ_get_eliminate
 1.01% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000001s :      3: predicate.graph_param_transform
 5.28% :     0.000008s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.84% :     0.000003s :      6: predicate.less_batch_normalization
 1.86% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.53% :     0.000002s :     15: predicate.load_eliminater
 1.63% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.77% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.09% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.25% :     0.000002s :     10: predicate.merge_addn
 0.81% :     0.000001s :     10: predicate.minmaximum_grad
 2.42% :     0.000004s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.93% :     0.000003s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.20% :     0.000000s :      3: predicate.reset_defer_inline
 1.63% :     0.000002s :     10: predicate.reshape_eliminate
 1.29% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.95% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.81% :     0.000001s :      6: predicate.special_op_eliminate
 0.72% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 1.93% :     0.000003s :     16: predicate.switch_defer_inline
 1.75% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.24% :     0.000011s :     53: predicate.switch_simplify
 1.70% :     0.000003s :     10: predicate.tile_eliminate
 1.56% :     0.000002s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.90% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.19% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 2.05% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.24% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.49% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.83% :     0.000003s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.52% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000979    12
47.56% :     0.000465s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.97% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.46% :     0.000455s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.033396    76
 0.19% :     0.000062s :      1: add_recomputation
 0.23% :     0.000077s :      1: auto_monad
 0.06% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.42% :     0.000474s :      1: bootstrap
 0.12% :     0.000039s :      1: cconv
 0.04% :     0.000013s :      1: convert_after_rewriter
 0.08% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.07% :     0.000024s :      1: event_method
 0.04% :     0.000012s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
32.72% :     0.010929s :      1: jit_opt_a
 0.52% :     0.000173s :      1: jit_opt_after_cconv
 0.18% :     0.000060s :      1: jit_opt_b
 1.77% :     0.000591s :      1: loop_unroll
 2.72% :     0.000907s :      1: mutable_eliminate
 3.27% :     0.001091s :     26: opt.transform.jit_opt_a
 0.16% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.81% :     0.000606s :      1: opt_after_jit_grad
 0.03% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000045s :      1: py_interpret_to_execute
 0.08% :     0.000026s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000019s :      1: remove_dup_value
 5.71% :     0.001908s :      1: renormalize.infer
 1.69% :     0.000566s :      1: renormalize.specialize
 0.03% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000048s :      1: rewriter_after_opt_a
 0.25% :     0.000082s :      1: rewriter_before_opt_a
 0.26% :     0.000086s :      1: symbol_engine_optimizer
25.78% :     0.008611s :      1: task_emit
19.72% :     0.006586s :      1: type_inference
 0.22% :     0.000073s :      1: validate


TotalTime = 0.0227638, [33]
[bootstrap]: 0.00043649
[type_inference]: 0.00545145
[event_method]: 1.914e-05
[auto_monad]: 6.828e-05
[graph_reusing]: 6.57002e-06
[pre_auto_parallel]: 2.86e-06
[py_interpret_to_execute]: 4.701e-05
[rewriter_before_opt_a]: 7.461e-05
[expand_dump_flag]: 3.51001e-06
[jit_opt_a]: 0.00581203, [2]
    [Cycle 1]: 0.00163956, [27]
        [switch_simplify]: 5.44e-05
        [loop_unroll]: 3.094e-05
        [a_1]: 0.00052497
        [with_stream_mark]: 1.37e-05
        [recompute_prepare]: 7.56001e-06
        [updatestate_depend_eliminate]: 2.78e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 5.97001e-06
        [updatestate_useless_node_eliminater]: 4.96002e-06
        [accelerated_algorithm]: 6.53e-06
        [meta_shard_fg_expand]: 1.21002e-06
        [get_grad_eliminate_]: 5.81e-06
        [merge_forward]: 2.38998e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.126e-05
        [j_node_and_user_rematch]: 8.85999e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 8.58001e-06
        [inline_without_move]: 5.51998e-06
        [renormalize]: 0.0007215
        [add_forward_monad_depend]: 5.03002e-06
        [auto_monad_grad]: 1.14e-06
        [auto_monad_eliminator]: 1.07e-05
        [cse]: 1.775e-05
        [replace_applicator]: 1.249e-05
    [Cycle 2]: 0.00032222, [27]
        [switch_simplify]: 5.52001e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 9.147e-05
        [with_stream_mark]: 8.58001e-06
        [recompute_prepare]: 4.80999e-06
        [updatestate_depend_eliminate]: 2.43e-06
        [updatestate_assign_eliminate]: 2.03002e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 9.49978e-07
        [specialize_transform]: 4.82e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.19e-06
        [meta_shard_fg_expand]: 1.02e-06
        [get_grad_eliminate_]: 4.33001e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.89e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.854e-05
        [j_node_and_user_rematch]: 7.63001e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 6.81001e-06
        [inline_without_move]: 4.45999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 5.32001e-06
        [cse]: 1.05e-05
        [replace_applicator]: 4.83001e-06
[py_interpret_to_execute_after_opt_a]: 8.79e-06
[rewriter_after_opt_a]: 2.91e-05
[convert_after_rewriter]: 6.14001e-06
[order_py_execute_after_rewriter]: 4.05e-06
[mutable_eliminate]: 0.00057361
[jit_opt_b]: 4.887e-05, [1]
    [Cycle 1]: 4.232e-05, [2]
        [frontend_op_eliminate]: 1.634e-05
        [inline_after_opt_a]: 1.476e-05
[cconv]: 1.708e-05
[loop_unroll]: 0.00114628
[jit_opt_after_cconv]: 0.00017571, [1]
    [Cycle 1]: 0.00016767, [11]
        [c_1]: 2.334e-05
        [parameter_eliminate]: 5.63002e-06
        [updatestate_depend_eliminate]: 1.045e-05
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 3.14999e-06
        [cse]: 2.768e-05
        [call_graph_tuple_transform]: 2.431e-05
        [tuple_list_get_item_eliminator]: 5.86998e-06
        [none_parameter_eliminate]: 1.69998e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 6.22001e-06
[remove_dup_value]: 1.513e-05
[partial_unused_args_eliminate]: 2.62001e-06
[environ_conv]: 6.82002e-06
[add_recomputation]: 5.669e-05
[cse_after_recomputation]: 2.247e-05, [1]
    [Cycle 1]: 1.652e-05, [1]
        [cse]: 9.28002e-06
[auto_monad_reorder]: 1.727e-05
[get_jit_bprop_graph]: 2.64001e-06
[rewriter_after_jit_bprop_graph]: 8.90001e-06
[opt_after_jit_grad]: 0.00057809
[symbol_engine_optimizer]: 8.667e-05, [1]
    [Cycle 1]: 7.707e-05, [6]
        [build]: 6.58998e-06
        [elim_shapecalc]: 9.46e-06
        [elim_not_effective]: 1.449e-05
        [opt_reshape]: 5.74e-06
        [fold_const_symbol]: 9.49e-06
        [renormalize]: 1.09e-06
[validate]: 4.705e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00771313
[execute]: 9.92999e-06
Sums
    bootstrap                            :     0.000436s :  2.39%
    type_inference                       :     0.005451s : 29.90%
    event_method                         :     0.000019s :  0.10%
    auto_monad                           :     0.000068s :  0.37%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000047s :  0.26%
    rewriter_before_opt_a                :     0.000075s :  0.41%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000060s :  0.33%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000616s :  3.38%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000722s :  3.96%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.09%
    jit_opt_a.cse                        :     0.000028s :  0.15%
    jit_opt_a.replace_applicator         :     0.000017s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.05%
    rewriter_after_opt_a                 :     0.000029s :  0.16%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000574s :  3.15%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000017s :  0.09%
    loop_unroll                          :     0.001146s :  6.29%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000028s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000057s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000578s :  3.17%
    symbol_engine_optimizer.build        :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000047s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007713s : 42.30%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000192    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 3.56% :     0.000007s :      3: substitution.graph_param_transform
81.50% :     0.000156s :      4: substitution.inline
 1.54% :     0.000003s :      4: substitution.j_node_and_user_rematch
 5.54% :     0.000011s :      4: substitution.remove_not_recompute_node
 1.68% :     0.000003s :      2: substitution.replace_old_param
 4.42% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005370     2
81.30% :     0.004366s :      1: type_inference.infer
18.70% :     0.001004s :      1: type_inference.specialize
------[replace.]   0.000071     6
67.22% :     0.000048s :      4: replace.inline
32.78% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000161     6
95.47% :     0.000153s :      4: match.inline
 4.53% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.71% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.90% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 2.21% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000002s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.48% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.14% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.22% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.63% :     0.000007s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000001s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 2.48% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.44% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.36% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000000s :      3: predicate.opt_reshape
 1.93% :     0.000002s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.57% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.02% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.32% :     0.000002s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.57% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.22% :     0.000003s :     16: predicate.switch_defer_inline
 2.01% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.11% :     0.000009s :     53: predicate.switch_simplify
 1.29% :     0.000002s :     10: predicate.tile_eliminate
 1.25% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.69% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.86% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000669    12
48.11% :     0.000322s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.28% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.61% :     0.000318s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024433    76
 0.25% :     0.000060s :      1: add_recomputation
 0.29% :     0.000072s :      1: auto_monad
 0.08% :     0.000021s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.87% :     0.000457s :      1: bootstrap
 0.08% :     0.000020s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000011s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
23.80% :     0.005816s :      1: jit_opt_a
 0.74% :     0.000180s :      1: jit_opt_after_cconv
 0.21% :     0.000051s :      1: jit_opt_b
 4.75% :     0.001162s :      1: loop_unroll
 2.39% :     0.000583s :      1: mutable_eliminate
 3.40% :     0.000831s :     26: opt.transform.jit_opt_a
 0.23% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000020s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.45% :     0.000598s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000050s :      1: py_interpret_to_execute
 0.04% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 1.56% :     0.000381s :      1: renormalize.infer
 1.37% :     0.000334s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000033s :      1: rewriter_after_opt_a
 0.32% :     0.000078s :      1: rewriter_before_opt_a
 0.37% :     0.000090s :      1: symbol_engine_optimizer
31.66% :     0.007735s :      1: task_emit
22.38% :     0.005469s :      1: type_inference
 0.30% :     0.000073s :      1: validate


TotalTime = 0.0140505, [30]
[bootstrap]: 0.00036628
[type_inference]: 0.00485399
[event_method]: 1.865e-05
[auto_monad]: 6.895e-05
[graph_reusing]: 7.02002e-06
[pre_auto_parallel]: 3.11001e-06
[py_interpret_to_execute]: 3.746e-05
[rewriter_before_opt_a]: 8.007e-05
[expand_dump_flag]: 3.44001e-06
[jit_opt_a]: 0.00619414, [2]
    [Cycle 1]: 0.0019502, [27]
        [switch_simplify]: 6.073e-05
        [loop_unroll]: 3.109e-05
        [a_1]: 0.00060815
        [with_stream_mark]: 2.388e-05
        [recompute_prepare]: 7.19001e-06
        [updatestate_depend_eliminate]: 4.27e-06
        [updatestate_assign_eliminate]: 3.3e-06
        [updatestate_loads_eliminate]: 2.83003e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 5.82001e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.96e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 6.10002e-06
        [merge_forward]: 3.90998e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.564e-05
        [j_node_and_user_rematch]: 1.008e-05
        [meta_fg_expand]: 2.53e-06
        [replace_old_param]: 9.66e-06
        [inline_without_move]: 5.54e-06
        [renormalize]: 0.00088288
        [add_forward_monad_depend]: 6.19999e-06
        [auto_monad_grad]: 2.85002e-06
        [auto_monad_eliminator]: 1.529e-05
        [cse]: 3.243e-05
        [replace_applicator]: 1.35e-05
    [Cycle 2]: 0.00048074, [27]
        [switch_simplify]: 5.77999e-06
        [loop_unroll]: 5.29e-06
        [a_1]: 0.00010359
        [with_stream_mark]: 1.041e-05
        [recompute_prepare]: 5.39998e-06
        [updatestate_depend_eliminate]: 2.62001e-06
        [updatestate_assign_eliminate]: 2.39999e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [parameter_eliminate]: 1.29998e-06
        [specialize_transform]: 4.98001e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 1.52001e-06
        [get_grad_eliminate_]: 4.97999e-06
        [merge_forward]: 2.99999e-06
        [cell_reuse_recompute_pass]: 2.01e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.468e-05
        [j_node_and_user_rematch]: 8.81002e-06
        [meta_fg_expand]: 3.21999e-06
        [replace_old_param]: 1.29e-05
        [inline_without_move]: 5.77999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 2.49999e-06
        [auto_monad_grad]: 1.24998e-06
        [auto_monad_eliminator]: 7.71001e-06
        [cse]: 1.656e-05
        [replace_applicator]: 5.92001e-06
[py_interpret_to_execute_after_opt_a]: 2.48e-05
[rewriter_after_opt_a]: 4.026e-05
[convert_after_rewriter]: 7.4e-06
[order_py_execute_after_rewriter]: 5.92001e-06
[mutable_eliminate]: 0.0007387
[jit_opt_b]: 5.687e-05, [1]
    [Cycle 1]: 4.798e-05, [2]
        [frontend_op_eliminate]: 1.92e-05
        [inline_after_opt_a]: 1.695e-05
[cconv]: 2.965e-05
[loop_unroll]: 0.00045101
[jit_opt_after_cconv]: 0.00015335, [1]
    [Cycle 1]: 0.00014635, [11]
        [c_1]: 2.283e-05
        [parameter_eliminate]: 3.41999e-06
        [updatestate_depend_eliminate]: 6.60002e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 2.358e-05
        [call_graph_tuple_transform]: 2.015e-05
        [tuple_list_get_item_eliminator]: 5.44e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 7.79983e-07
        [switch_simplify]: 5.56e-06
[remove_dup_value]: 1.694e-05
[partial_unused_args_eliminate]: 2.24001e-06
[environ_conv]: 6.22001e-06
[add_recomputation]: 5.422e-05
[cse_after_recomputation]: 2.34e-05, [1]
    [Cycle 1]: 1.651e-05, [1]
        [cse]: 1.038e-05
[auto_monad_reorder]: 1.618e-05
[get_jit_bprop_graph]: 2.36e-06
[rewriter_after_jit_bprop_graph]: 5.35001e-06
[opt_after_jit_grad]: 0.00047216
[symbol_engine_optimizer]: 7.663e-05, [1]
    [Cycle 1]: 7.005e-05, [6]
        [build]: 4.04002e-06
        [elim_shapecalc]: 7.73001e-06
        [elim_not_effective]: 1.388e-05
        [opt_reshape]: 5.72999e-06
        [fold_const_symbol]: 9.49e-06
        [renormalize]: 7.40023e-07
[validate]: 4.215e-05
Sums
    bootstrap                            :     0.000366s :  3.83%
    type_inference                       :     0.004854s : 50.81%
    event_method                         :     0.000019s :  0.20%
    auto_monad                           :     0.000069s :  0.72%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000037s :  0.39%
    rewriter_before_opt_a                :     0.000080s :  0.84%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000067s :  0.70%
    jit_opt_a.loop_unroll                :     0.000036s :  0.38%
    jit_opt_a.a_1                        :     0.000712s :  7.45%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.36%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.12%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.06%
    jit_opt_a.replace_old_param          :     0.000023s :  0.24%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000883s :  9.24%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.24%
    jit_opt_a.cse                        :     0.000049s :  0.51%
    jit_opt_a.replace_applicator         :     0.000019s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000025s :  0.26%
    rewriter_after_opt_a                 :     0.000040s :  0.42%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000739s :  7.73%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.18%
    cconv                                :     0.000030s :  0.31%
    loop_unroll                          :     0.000451s :  4.72%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000017s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000054s :  0.57%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000472s :  4.94%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000042s :  0.44%


Time group info:
------[substitution.]   0.000260    23
 1.01% :     0.000003s :      2: substitution.elim_not_effective
 0.60% :     0.000002s :      2: substitution.fold_const_symbol
 2.34% :     0.000006s :      3: substitution.graph_param_transform
85.39% :     0.000222s :      4: substitution.inline
 1.56% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.93% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.25% :     0.000006s :      2: substitution.replace_old_param
 4.92% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004786     2
80.84% :     0.003869s :      1: type_inference.infer
19.16% :     0.000917s :      1: type_inference.specialize
------[replace.]   0.000081     6
68.20% :     0.000055s :      4: replace.inline
31.80% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000231     6
94.92% :     0.000220s :      4: match.inline
 5.08% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 2.14% :     0.000003s :     10: predicate.arithmetic_simplify
 1.15% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.08% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.64% :     0.000001s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.79% :     0.000007s :     22: predicate.inline
 1.19% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.56% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.36% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 2.13% :     0.000003s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.54% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.23% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.32% :     0.000002s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.56% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.27% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.65% :     0.000009s :     53: predicate.switch_simplify
 1.29% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.78% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.90% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000696    12
49.20% :     0.000343s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.19% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.61% :     0.000304s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015926    72
 0.36% :     0.000057s :      1: add_recomputation
 0.46% :     0.000073s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 2.44% :     0.000388s :      1: bootstrap
 0.20% :     0.000032s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
38.92% :     0.006199s :      1: jit_opt_a
 0.98% :     0.000156s :      1: jit_opt_after_cconv
 0.37% :     0.000060s :      1: jit_opt_b
 2.88% :     0.000459s :      1: loop_unroll
 4.71% :     0.000750s :      1: mutable_eliminate
 5.91% :     0.000942s :     26: opt.transform.jit_opt_a
 0.32% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000029s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.02% :     0.000481s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000041s :      1: py_interpret_to_execute
 0.17% :     0.000027s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000019s :      1: remove_dup_value
 3.07% :     0.000489s :      1: renormalize.infer
 2.41% :     0.000384s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000043s :      1: rewriter_after_opt_a
 0.52% :     0.000083s :      1: rewriter_before_opt_a
 0.50% :     0.000079s :      1: symbol_engine_optimizer
30.57% :     0.004869s :      1: type_inference


TotalTime = 0.0230503, [33]
[bootstrap]: 0.00030615
[type_inference]: 0.00507541
[event_method]: 1.828e-05
[auto_monad]: 6.747e-05
[graph_reusing]: 5.62001e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 3.636e-05
[rewriter_before_opt_a]: 7.068e-05
[expand_dump_flag]: 3.4e-06
[jit_opt_a]: 0.00643441, [2]
    [Cycle 1]: 0.00203484, [27]
        [switch_simplify]: 8.232e-05
        [loop_unroll]: 3.037e-05
        [a_1]: 0.00057626
        [with_stream_mark]: 2.502e-05
        [recompute_prepare]: 7.28e-06
        [updatestate_depend_eliminate]: 4.25e-06
        [updatestate_assign_eliminate]: 3.03e-06
        [updatestate_loads_eliminate]: 2.48002e-06
        [parameter_eliminate]: 2.07001e-06
        [specialize_transform]: 6.03998e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.97001e-06
        [meta_shard_fg_expand]: 2.14999e-06
        [get_grad_eliminate_]: 5.42001e-06
        [merge_forward]: 3.84002e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.557e-05
        [j_node_and_user_rematch]: 9.12999e-06
        [meta_fg_expand]: 2.58e-06
        [replace_old_param]: 9.27001e-06
        [inline_without_move]: 5.77001e-06
        [renormalize]: 0.00097085
        [add_forward_monad_depend]: 7.46001e-06
        [auto_monad_grad]: 2.84999e-06
        [auto_monad_eliminator]: 1.918e-05
        [cse]: 3.144e-05
        [replace_applicator]: 1.785e-05
    [Cycle 2]: 0.00035219, [27]
        [switch_simplify]: 5.42999e-06
        [loop_unroll]: 5.37999e-06
        [a_1]: 9.409e-05
        [with_stream_mark]: 1.419e-05
        [recompute_prepare]: 6.43e-06
        [updatestate_depend_eliminate]: 2.73998e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [parameter_eliminate]: 1.29998e-06
        [specialize_transform]: 5.09998e-06
        [updatestate_useless_node_eliminater]: 5.10999e-06
        [accelerated_algorithm]: 5.76003e-06
        [meta_shard_fg_expand]: 1.96998e-06
        [get_grad_eliminate_]: 5.19e-06
        [merge_forward]: 3.11001e-06
        [cell_reuse_recompute_pass]: 1.99e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.572e-05
        [j_node_and_user_rematch]: 9.96998e-06
        [meta_fg_expand]: 1.86e-06
        [replace_old_param]: 8.38001e-06
        [inline_without_move]: 4.39002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.82999e-06
        [auto_monad_grad]: 1.38002e-06
        [auto_monad_eliminator]: 6.59999e-06
        [cse]: 1.205e-05
        [replace_applicator]: 5.59e-06
[py_interpret_to_execute_after_opt_a]: 1.514e-05
[rewriter_after_opt_a]: 4.092e-05
[convert_after_rewriter]: 8.13999e-06
[order_py_execute_after_rewriter]: 4.96997e-06
[mutable_eliminate]: 0.00077699
[jit_opt_b]: 5.55e-05, [1]
    [Cycle 1]: 4.632e-05, [2]
        [frontend_op_eliminate]: 1.771e-05
        [inline_after_opt_a]: 1.591e-05
[cconv]: 3.403e-05
[loop_unroll]: 0.00051414
[jit_opt_after_cconv]: 0.00022824, [1]
    [Cycle 1]: 0.00022035, [11]
        [c_1]: 2.213e-05
        [parameter_eliminate]: 3.65e-06
        [updatestate_depend_eliminate]: 8.61002e-06
        [updatestate_assign_eliminate]: 2.33002e-06
        [updatestate_loads_eliminate]: 2.56998e-06
        [cse]: 2.519e-05
        [call_graph_tuple_transform]: 2.173e-05
        [tuple_list_get_item_eliminator]: 5.67999e-06
        [none_parameter_eliminate]: 1.99e-06
        [renormalize]: 7.00005e-07
        [switch_simplify]: 6.71999e-06
[remove_dup_value]: 1.628e-05
[partial_unused_args_eliminate]: 2.52001e-06
[environ_conv]: 6.47001e-06
[add_recomputation]: 5.894e-05
[cse_after_recomputation]: 2.416e-05, [1]
    [Cycle 1]: 1.771e-05, [1]
        [cse]: 1.142e-05
[auto_monad_reorder]: 1.589e-05
[get_jit_bprop_graph]: 2.02999e-06
[rewriter_after_jit_bprop_graph]: 8.89e-06
[opt_after_jit_grad]: 0.00057333
[symbol_engine_optimizer]: 7.967e-05, [1]
    [Cycle 1]: 7.219e-05, [6]
        [build]: 5.56e-06
        [elim_shapecalc]: 8.13999e-06
        [elim_not_effective]: 1.349e-05
        [opt_reshape]: 6.11e-06
        [fold_const_symbol]: 9.69999e-06
        [renormalize]: 8.70001e-07
[validate]: 4.306e-05
[backend_pass]: 8.49977e-07
[task_emit]: 0.00827854
[execute]: 9.17999e-06
Sums
    bootstrap                            :     0.000306s :  1.68%
    type_inference                       :     0.005075s : 27.78%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000067s :  0.37%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000036s :  0.20%
    rewriter_before_opt_a                :     0.000071s :  0.39%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000088s :  0.48%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000670s :  3.67%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000971s :  5.31%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.14%
    jit_opt_a.cse                        :     0.000043s :  0.24%
    jit_opt_a.replace_applicator         :     0.000023s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.08%
    rewriter_after_opt_a                 :     0.000041s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000777s :  4.25%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000034s :  0.19%
    loop_unroll                          :     0.000514s :  2.81%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.04%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000059s :  0.32%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000573s :  3.14%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008279s : 45.31%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000245    23
 0.77% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.42% :     0.000006s :      3: substitution.graph_param_transform
84.41% :     0.000207s :      4: substitution.inline
 1.97% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.06% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.94% :     0.000005s :      2: substitution.replace_old_param
 5.89% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005006     2
81.53% :     0.004081s :      1: type_inference.infer
18.47% :     0.000925s :      1: type_inference.specialize
------[replace.]   0.000069     6
68.97% :     0.000048s :      4: replace.inline
31.03% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000218     6
93.85% :     0.000204s :      4: match.inline
 6.15% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.65% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.48% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.52% :     0.000002s :     10: predicate.addn_zero_filter
 2.78% :     0.000004s :     10: predicate.arithmetic_simplify
 1.12% :     0.000001s :     10: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000002s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.65% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.26% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.42% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.00% :     0.000001s :     10: predicate.environ_get_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.57% :     0.000007s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 2.30% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 2.86% :     0.000004s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.01% :     0.000003s :     15: predicate.partial_eliminate
 1.34% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.63% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.87% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.10% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     10: predicate.same_eliminate
 0.68% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 1.83% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.60% :     0.000010s :     53: predicate.switch_simplify
 1.37% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.23% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.03% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.43% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.87% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.51% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000671    12
44.50% :     0.000299s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.56% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.94% :     0.000322s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025062    76
 0.25% :     0.000063s :      1: add_recomputation
 0.28% :     0.000071s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.27% :     0.000318s :      1: bootstrap
 0.15% :     0.000037s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000026s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
25.69% :     0.006439s :      1: jit_opt_a
 0.92% :     0.000231s :      1: jit_opt_after_cconv
 0.23% :     0.000058s :      1: jit_opt_b
 2.10% :     0.000526s :      1: loop_unroll
 3.15% :     0.000790s :      1: mutable_eliminate
 3.68% :     0.000921s :     26: opt.transform.jit_opt_a
 0.21% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.07% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.34% :     0.000587s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000039s :      1: py_interpret_to_execute
 0.07% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000019s :      1: remove_dup_value
 2.15% :     0.000539s :      1: renormalize.infer
 1.68% :     0.000422s :      1: renormalize.specialize
 0.05% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000044s :      1: rewriter_after_opt_a
 0.29% :     0.000073s :      1: rewriter_before_opt_a
 0.33% :     0.000082s :      1: symbol_engine_optimizer
33.12% :     0.008300s :      1: task_emit
20.32% :     0.005092s :      1: type_inference
 0.27% :     0.000067s :      1: validate


TotalTime = 0.0323015, [33]
[bootstrap]: 0.00031339
[type_inference]: 0.00473876
[event_method]: 1.87e-05
[auto_monad]: 6.516e-05
[graph_reusing]: 6.39001e-06
[pre_auto_parallel]: 2.72001e-06
[py_interpret_to_execute]: 3.623e-05
[rewriter_before_opt_a]: 7.421e-05
[expand_dump_flag]: 3.47002e-06
[jit_opt_a]: 0.00560842, [2]
    [Cycle 1]: 0.00179002, [27]
        [switch_simplify]: 5.903e-05
        [loop_unroll]: 3.006e-05
        [a_1]: 0.00055886
        [with_stream_mark]: 1.985e-05
        [recompute_prepare]: 7.8e-06
        [updatestate_depend_eliminate]: 4.12e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.53998e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 1.186e-05
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 5.75001e-06
        [meta_shard_fg_expand]: 2.11e-06
        [get_grad_eliminate_]: 5.52999e-06
        [merge_forward]: 3.59002e-06
        [cell_reuse_recompute_pass]: 1.35001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.436e-05
        [j_node_and_user_rematch]: 1.01e-05
        [meta_fg_expand]: 2.61999e-06
        [replace_old_param]: 9.09e-06
        [inline_without_move]: 5.51e-06
        [renormalize]: 0.00071988
        [add_forward_monad_depend]: 5.52999e-06
        [auto_monad_grad]: 2.31e-06
        [auto_monad_eliminator]: 1.493e-05
        [cse]: 3.169e-05
        [replace_applicator]: 1.256e-05
    [Cycle 2]: 0.00034342, [27]
        [switch_simplify]: 5.75001e-06
        [loop_unroll]: 4.76002e-06
        [a_1]: 9.034e-05
        [with_stream_mark]: 1.104e-05
        [recompute_prepare]: 6.11e-06
        [updatestate_depend_eliminate]: 3.35e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 4.77998e-06
        [meta_shard_fg_expand]: 1.35999e-06
        [get_grad_eliminate_]: 4.62e-06
        [merge_forward]: 2.45002e-06
        [cell_reuse_recompute_pass]: 1.59998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.412e-05
        [j_node_and_user_rematch]: 7.99997e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 2.619e-05
        [inline_without_move]: 4.95999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 7.80012e-07
        [auto_monad_eliminator]: 5.23002e-06
        [cse]: 1.068e-05
        [replace_applicator]: 4.81002e-06
[py_interpret_to_execute_after_opt_a]: 1.083e-05
[rewriter_after_opt_a]: 3.48e-05
[convert_after_rewriter]: 7.12002e-06
[order_py_execute_after_rewriter]: 5.02e-06
[mutable_eliminate]: 0.00053646
[jit_opt_b]: 5.013e-05, [1]
    [Cycle 1]: 4.279e-05, [2]
        [frontend_op_eliminate]: 1.649e-05
        [inline_after_opt_a]: 1.497e-05
[cconv]: 2.48e-05
[loop_unroll]: 0.00041844
[jit_opt_after_cconv]: 0.00014013, [1]
    [Cycle 1]: 0.0001336, [11]
        [c_1]: 2.116e-05
        [parameter_eliminate]: 2.99999e-06
        [updatestate_depend_eliminate]: 5.77999e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.846e-05
        [call_graph_tuple_transform]: 1.842e-05
        [tuple_list_get_item_eliminator]: 5.34e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 7.00005e-07
        [switch_simplify]: 5.27001e-06
[remove_dup_value]: 1.381e-05
[partial_unused_args_eliminate]: 2.13002e-06
[environ_conv]: 5.89999e-06
[add_recomputation]: 5.36e-05
[cse_after_recomputation]: 2.008e-05, [1]
    [Cycle 1]: 1.488e-05, [1]
        [cse]: 9.12001e-06
[auto_monad_reorder]: 1.563e-05
[get_jit_bprop_graph]: 1.92999e-06
[rewriter_after_jit_bprop_graph]: 5.57001e-06
[opt_after_jit_grad]: 0.00044983
[symbol_engine_optimizer]: 7.21e-05, [1]
    [Cycle 1]: 6.631e-05, [6]
        [build]: 3.73001e-06
        [elim_shapecalc]: 7.34002e-06
        [elim_not_effective]: 1.257e-05
        [opt_reshape]: 5.66e-06
        [fold_const_symbol]: 8.59e-06
        [renormalize]: 5.59987e-07
[validate]: 3.733e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.0192952
[execute]: 1.007e-05
Sums
    bootstrap                            :     0.000313s :  1.11%
    type_inference                       :     0.004739s : 16.85%
    event_method                         :     0.000019s :  0.07%
    auto_monad                           :     0.000065s :  0.23%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000036s :  0.13%
    rewriter_before_opt_a                :     0.000074s :  0.26%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000065s :  0.23%
    jit_opt_a.loop_unroll                :     0.000035s :  0.12%
    jit_opt_a.a_1                        :     0.000649s :  2.31%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000035s :  0.13%
    jit_opt_a.inline_without_move        :     0.000010s :  0.04%
    jit_opt_a.renormalize                :     0.000720s :  2.56%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.07%
    jit_opt_a.cse                        :     0.000042s :  0.15%
    jit_opt_a.replace_applicator         :     0.000017s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.04%
    rewriter_after_opt_a                 :     0.000035s :  0.12%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000536s :  1.91%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.05%
    cconv                                :     0.000025s :  0.09%
    loop_unroll                          :     0.000418s :  1.49%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000014s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000054s :  0.19%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000450s :  1.60%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.019295s : 68.59%
    execute                              :     0.000010s :  0.04%


Time group info:
------[substitution.]   0.000243    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.21% :     0.000005s :      3: substitution.graph_param_transform
78.68% :     0.000191s :      4: substitution.inline
 1.43% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.82% :     0.000004s :      4: substitution.remove_not_recompute_node
 9.37% :     0.000023s :      2: substitution.replace_old_param
 5.10% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004670     2
81.01% :     0.003783s :      1: type_inference.infer
18.99% :     0.000887s :      1: type_inference.specialize
------[replace.]   0.000069     6
68.45% :     0.000047s :      4: replace.inline
31.55% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     6
94.40% :     0.000189s :      4: match.inline
 5.60% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.17% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.44% :     0.000002s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.48% :     0.000002s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.17% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.82% :     0.000007s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.71% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.66% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.80% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.44% :     0.000002s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.83% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.02% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.00% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000639    12
48.43% :     0.000309s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.73% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.84% :     0.000286s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.034006    76
 0.17% :     0.000057s :      1: add_recomputation
 0.20% :     0.000069s :      1: auto_monad
 0.05% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.99% :     0.000335s :      1: bootstrap
 0.08% :     0.000027s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.06% :     0.000022s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.07% :     0.000023s :      1: event_method
 0.04% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
16.50% :     0.005612s :      1: jit_opt_a
 0.42% :     0.000143s :      1: jit_opt_after_cconv
 0.16% :     0.000053s :      1: jit_opt_b
 1.25% :     0.000427s :      1: loop_unroll
 1.60% :     0.000545s :      1: mutable_eliminate
 2.62% :     0.000890s :     26: opt.transform.jit_opt_a
 0.14% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000025s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.35% :     0.000458s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000039s :      1: py_interpret_to_execute
 0.04% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000016s :      1: remove_dup_value
 1.13% :     0.000384s :      1: renormalize.infer
 0.96% :     0.000328s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000038s :      1: rewriter_after_opt_a
 0.23% :     0.000077s :      1: rewriter_before_opt_a
 0.22% :     0.000075s :      1: symbol_engine_optimizer
56.80% :     0.019315s :      1: task_emit
13.98% :     0.004755s :      1: type_inference
 0.18% :     0.000060s :      1: validate


TotalTime = 0.0289214, [33]
[bootstrap]: 0.00046545
[type_inference]: 0.0086566
[event_method]: 2.331e-05
[auto_monad]: 9.061e-05
[graph_reusing]: 6.12999e-06
[pre_auto_parallel]: 2.76999e-06
[py_interpret_to_execute]: 4.491e-05
[rewriter_before_opt_a]: 0.00010007
[expand_dump_flag]: 3.39001e-06
[jit_opt_a]: 0.00785117, [2]
    [Cycle 1]: 0.00231251, [27]
        [switch_simplify]: 6.454e-05
        [loop_unroll]: 3.02e-05
        [a_1]: 0.00070893
        [with_stream_mark]: 5.702e-05
        [recompute_prepare]: 1.336e-05
        [updatestate_depend_eliminate]: 6.40002e-06
        [updatestate_assign_eliminate]: 3.21001e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 2.62001e-06
        [specialize_transform]: 6.83e-06
        [updatestate_useless_node_eliminater]: 5.61e-06
        [accelerated_algorithm]: 6.17999e-06
        [meta_shard_fg_expand]: 2.22001e-06
        [get_grad_eliminate_]: 5.94999e-06
        [merge_forward]: 4.03999e-06
        [cell_reuse_recompute_pass]: 1.54998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.644e-05
        [j_node_and_user_rematch]: 1.054e-05
        [meta_fg_expand]: 3.03e-06
        [replace_old_param]: 1.006e-05
        [inline_without_move]: 5.32001e-06
        [renormalize]: 0.00104022
        [add_forward_monad_depend]: 8.38999e-06
        [auto_monad_grad]: 2.61e-06
        [auto_monad_eliminator]: 1.962e-05
        [cse]: 3.188e-05
        [replace_applicator]: 1.874e-05
    [Cycle 2]: 0.00036614, [27]
        [switch_simplify]: 5.56998e-06
        [loop_unroll]: 4.89998e-06
        [a_1]: 9.739e-05
        [with_stream_mark]: 1.394e-05
        [recompute_prepare]: 5.31002e-06
        [updatestate_depend_eliminate]: 3.42002e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.35999e-06
        [specialize_transform]: 5.52999e-06
        [updatestate_useless_node_eliminater]: 4.78001e-06
        [accelerated_algorithm]: 6.63e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 4.65999e-06
        [merge_forward]: 2.91e-06
        [cell_reuse_recompute_pass]: 3.08998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.648e-05
        [j_node_and_user_rematch]: 8.38999e-06
        [meta_fg_expand]: 2.12001e-06
        [replace_old_param]: 1.893e-05
        [inline_without_move]: 5.97001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.69e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 7.64002e-06
        [cse]: 1.12e-05
        [replace_applicator]: 5.86998e-06
[py_interpret_to_execute_after_opt_a]: 1.514e-05
[rewriter_after_opt_a]: 4.116e-05
[convert_after_rewriter]: 8.08999e-06
[order_py_execute_after_rewriter]: 5.40999e-06
[mutable_eliminate]: 0.00164007
[jit_opt_b]: 5.912e-05, [1]
    [Cycle 1]: 5.045e-05, [2]
        [frontend_op_eliminate]: 1.792e-05
        [inline_after_opt_a]: 1.808e-05
[cconv]: 4.098e-05
[loop_unroll]: 0.00059873
[jit_opt_after_cconv]: 0.00016318, [1]
    [Cycle 1]: 0.00015586, [11]
        [c_1]: 2.251e-05
        [parameter_eliminate]: 1.60999e-06
        [updatestate_depend_eliminate]: 5.52999e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.20002e-06
        [cse]: 3.387e-05
        [call_graph_tuple_transform]: 2.044e-05
        [tuple_list_get_item_eliminator]: 5.70001e-06
        [none_parameter_eliminate]: 2.03997e-06
        [renormalize]: 8.29983e-07
        [switch_simplify]: 5.97999e-06
[remove_dup_value]: 1.562e-05
[partial_unused_args_eliminate]: 2.63e-06
[environ_conv]: 6.58e-06
[add_recomputation]: 9.044e-05
[cse_after_recomputation]: 2.609e-05, [1]
    [Cycle 1]: 1.905e-05, [1]
        [cse]: 1.135e-05
[auto_monad_reorder]: 1.586e-05
[get_jit_bprop_graph]: 2.48e-06
[rewriter_after_jit_bprop_graph]: 7.83001e-06
[opt_after_jit_grad]: 0.00058139
[symbol_engine_optimizer]: 8.498e-05, [1]
    [Cycle 1]: 7.577e-05, [6]
        [build]: 4.75999e-06
        [elim_shapecalc]: 7.97e-06
        [elim_not_effective]: 1.461e-05
        [opt_reshape]: 6.27001e-06
        [fold_const_symbol]: 1.034e-05
        [renormalize]: 9.20001e-07
[validate]: 4.052e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.0079568
[execute]: 9.93998e-06
Sums
    bootstrap                            :     0.000465s :  2.02%
    type_inference                       :     0.008657s : 37.63%
    event_method                         :     0.000023s :  0.10%
    auto_monad                           :     0.000091s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000045s :  0.20%
    rewriter_before_opt_a                :     0.000100s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000070s :  0.30%
    jit_opt_a.loop_unroll                :     0.000035s :  0.15%
    jit_opt_a.a_1                        :     0.000806s :  3.51%
    jit_opt_a.with_stream_mark           :     0.000071s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000029s :  0.13%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.001040s :  4.52%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.12%
    jit_opt_a.cse                        :     0.000043s :  0.19%
    jit_opt_a.replace_applicator         :     0.000025s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.07%
    rewriter_after_opt_a                 :     0.000041s :  0.18%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.001640s :  7.13%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.08%
    cconv                                :     0.000041s :  0.18%
    loop_unroll                          :     0.000599s :  2.60%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000034s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.07%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000090s :  0.39%
    cse_after_recomputation.cse          :     0.000011s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.03%
    opt_after_jit_grad                   :     0.000581s :  2.53%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.18%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007957s : 34.59%
    execute                              :     0.000010s :  0.04%


Time group info:
------[substitution.]   0.000316    23
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 2.01% :     0.000006s :      3: substitution.graph_param_transform
84.20% :     0.000266s :      4: substitution.inline
 1.29% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.80% :     0.000006s :      4: substitution.remove_not_recompute_node
 4.91% :     0.000016s :      2: substitution.replace_old_param
 4.54% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008522     2
86.74% :     0.007392s :      1: type_inference.infer
13.26% :     0.001130s :      1: type_inference.specialize
------[replace.]   0.000091     6
65.43% :     0.000060s :      4: replace.inline
34.57% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000277     6
95.23% :     0.000264s :      4: match.inline
 4.77% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000155   691
 1.09% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.07% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.79% :     0.000001s :     10: predicate.addn_check_dump
 1.14% :     0.000002s :     10: predicate.addn_zero_filter
18.46% :     0.000029s :     10: predicate.arithmetic_simplify
 1.18% :     0.000002s :     10: predicate.cast_eliminate
 0.30% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.78% :     0.000001s :     10: predicate.compare_switch_simplify
 1.15% :     0.000002s :     10: predicate.depend_value_elim
 1.48% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.72% :     0.000003s :     10: predicate.dict_get_item_eliminator
 0.94% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.25% :     0.000000s :      3: predicate.elim_not_effective
 0.70% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.87% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.79% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.82% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.88% :     0.000001s :     10: predicate.environ_get_eliminate
 0.99% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.17% :     0.000000s :      3: predicate.fold_const_symbol
 0.76% :     0.000001s :      6: predicate.get_grad_eliminate
 0.18% :     0.000000s :      3: predicate.graph_param_transform
 5.12% :     0.000008s :     22: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :      6: predicate.less_batch_normalization
 1.08% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.51% :     0.000002s :     15: predicate.load_eliminater
 1.10% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.20% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.54% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000002s :     10: predicate.merge_addn
 0.92% :     0.000001s :     10: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.37% :     0.000001s :      3: predicate.opt_reshape
 1.94% :     0.000003s :     15: predicate.partial_eliminate
 1.53% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.27% :     0.000002s :     10: predicate.reduce_eliminate
 1.29% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.77% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.40% :     0.000002s :     18: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.60% :     0.000002s :     10: predicate.reshape_eliminate
 0.94% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.52% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.74% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000002s :      6: predicate.specialize_transform
 1.08% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.07% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.30% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.11% :     0.000003s :     16: predicate.switch_defer_inline
 1.55% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.49% :     0.000010s :     53: predicate.switch_simplify
 0.91% :     0.000001s :     10: predicate.tile_eliminate
 0.88% :     0.000001s :     10: predicate.transpose_eliminate
 0.97% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.94% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 2.80% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.35% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.09% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.21% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.23% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.42% :     0.000002s :     10: predicate.value_based_eliminate
 0.27% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.43% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000832    12
48.68% :     0.000405s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.62% :     0.000063s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.70% :     0.000364s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.031120    76
 0.30% :     0.000095s :      1: add_recomputation
 0.31% :     0.000095s :      1: auto_monad
 0.06% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.58% :     0.000492s :      1: bootstrap
 0.14% :     0.000044s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000029s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.09% :     0.000029s :      1: event_method
 0.05% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
25.25% :     0.007857s :      1: jit_opt_a
 0.53% :     0.000166s :      1: jit_opt_after_cconv
 0.20% :     0.000062s :      1: jit_opt_b
 1.95% :     0.000606s :      1: loop_unroll
 5.30% :     0.001651s :      1: mutable_eliminate
 3.40% :     0.001058s :     26: opt.transform.jit_opt_a
 0.16% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000028s :      4: opt.transform.jit_opt_b
 0.04% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.91% :     0.000593s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000049s :      1: py_interpret_to_execute
 0.06% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000018s :      1: remove_dup_value
 1.84% :     0.000573s :      1: renormalize.infer
 1.46% :     0.000454s :      1: renormalize.specialize
 0.03% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000047s :      1: rewriter_after_opt_a
 0.34% :     0.000105s :      1: rewriter_before_opt_a
 0.28% :     0.000088s :      1: symbol_engine_optimizer
25.63% :     0.007976s :      1: task_emit
27.89% :     0.008680s :      1: type_inference
 0.21% :     0.000065s :      1: validate


TotalTime = 0.0313213, [33]
[bootstrap]: 0.00041092
[type_inference]: 0.0057361
[event_method]: 2.027e-05
[auto_monad]: 5.413e-05
[graph_reusing]: 4.35e-06
[pre_auto_parallel]: 2.65002e-06
[py_interpret_to_execute]: 0.0039648
[rewriter_before_opt_a]: 0.00019083
[expand_dump_flag]: 7.12002e-06
[jit_opt_a]: 0.00720739, [2]
    [Cycle 1]: 0.00211939, [27]
        [switch_simplify]: 6.814e-05
        [loop_unroll]: 2.999e-05
        [a_1]: 0.00065297
        [with_stream_mark]: 2.366e-05
        [recompute_prepare]: 9.25001e-06
        [updatestate_depend_eliminate]: 3.78001e-06
        [updatestate_assign_eliminate]: 3.08998e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [parameter_eliminate]: 1.95001e-06
        [specialize_transform]: 6.36998e-06
        [updatestate_useless_node_eliminater]: 5.94e-06
        [accelerated_algorithm]: 5.91e-06
        [meta_shard_fg_expand]: 1.72001e-06
        [get_grad_eliminate_]: 5.77999e-06
        [merge_forward]: 3.85998e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.458e-05
        [j_node_and_user_rematch]: 9.64e-06
        [meta_fg_expand]: 2.01e-06
        [replace_old_param]: 1.021e-05
        [inline_without_move]: 5.88998e-06
        [renormalize]: 0.00100168
        [add_forward_monad_depend]: 8.06001e-06
        [auto_monad_grad]: 2.53e-06
        [auto_monad_eliminator]: 1.451e-05
        [cse]: 2.316e-05
        [replace_applicator]: 1.413e-05
    [Cycle 2]: 0.00036526, [27]
        [switch_simplify]: 6.12999e-06
        [loop_unroll]: 4.95001e-06
        [a_1]: 0.00010029
        [with_stream_mark]: 1.458e-05
        [recompute_prepare]: 5.47999e-06
        [updatestate_depend_eliminate]: 2.83e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [parameter_eliminate]: 1.47001e-06
        [specialize_transform]: 5.81e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 5.15001e-06
        [meta_shard_fg_expand]: 1.96e-06
        [get_grad_eliminate_]: 5.02e-06
        [merge_forward]: 3.22002e-06
        [cell_reuse_recompute_pass]: 2.98e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.658e-05
        [j_node_and_user_rematch]: 9.31e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 8.30999e-06
        [inline_without_move]: 5.41002e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.62999e-06
        [auto_monad_grad]: 1.12e-06
        [auto_monad_eliminator]: 5.94e-06
        [cse]: 1.104e-05
        [replace_applicator]: 5.45001e-06
[py_interpret_to_execute_after_opt_a]: 1.413e-05
[rewriter_after_opt_a]: 3.954e-05
[convert_after_rewriter]: 8.16002e-06
[order_py_execute_after_rewriter]: 6.26e-06
[mutable_eliminate]: 0.00214374
[jit_opt_b]: 6.483e-05, [1]
    [Cycle 1]: 5.507e-05, [2]
        [frontend_op_eliminate]: 2.234e-05
        [inline_after_opt_a]: 1.884e-05
[cconv]: 3.64e-05
[loop_unroll]: 0.00058951
[jit_opt_after_cconv]: 0.0001768, [1]
    [Cycle 1]: 0.00016963, [11]
        [c_1]: 2.355e-05
        [parameter_eliminate]: 6.09999e-06
        [updatestate_depend_eliminate]: 8.65999e-06
        [updatestate_assign_eliminate]: 3.41001e-06
        [updatestate_loads_eliminate]: 2.93e-06
        [cse]: 3.134e-05
        [call_graph_tuple_transform]: 2.428e-05
        [tuple_list_get_item_eliminator]: 5.66e-06
        [none_parameter_eliminate]: 1.27999e-06
        [renormalize]: 7.29982e-07
        [switch_simplify]: 5.77999e-06
[remove_dup_value]: 1.667e-05
[partial_unused_args_eliminate]: 3.24001e-06
[environ_conv]: 6.61999e-06
[add_recomputation]: 5.37e-05
[cse_after_recomputation]: 2.469e-05, [1]
    [Cycle 1]: 1.747e-05, [1]
        [cse]: 9.91e-06
[auto_monad_reorder]: 1.593e-05
[get_jit_bprop_graph]: 2.92002e-06
[rewriter_after_jit_bprop_graph]: 9.19998e-06
[opt_after_jit_grad]: 0.00055575
[symbol_engine_optimizer]: 8.502e-05, [1]
    [Cycle 1]: 7.493e-05, [6]
        [build]: 4.36002e-06
        [elim_shapecalc]: 9.15999e-06
        [elim_not_effective]: 1.559e-05
        [opt_reshape]: 6.84999e-06
        [fold_const_symbol]: 9.56e-06
        [renormalize]: 3.9002e-07
[validate]: 4.843e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.00951735
[execute]: 8.32e-06
Sums
    bootstrap                            :     0.000411s :  1.59%
    type_inference                       :     0.005736s : 22.19%
    event_method                         :     0.000020s :  0.08%
    auto_monad                           :     0.000054s :  0.21%
    graph_reusing                        :     0.000004s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.003965s : 15.34%
    rewriter_before_opt_a                :     0.000191s :  0.74%
    expand_dump_flag                     :     0.000007s :  0.03%
    jit_opt_a.switch_simplify            :     0.000074s :  0.29%
    jit_opt_a.loop_unroll                :     0.000035s :  0.14%
    jit_opt_a.a_1                        :     0.000753s :  2.91%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.07%
    jit_opt_a.inline_without_move        :     0.000011s :  0.04%
    jit_opt_a.renormalize                :     0.001002s :  3.88%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.08%
    jit_opt_a.cse                        :     0.000034s :  0.13%
    jit_opt_a.replace_applicator         :     0.000020s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.05%
    rewriter_after_opt_a                 :     0.000040s :  0.15%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.002144s :  8.29%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.07%
    cconv                                :     0.000036s :  0.14%
    loop_unroll                          :     0.000590s :  2.28%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000031s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000017s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000054s :  0.21%
    cse_after_recomputation.cse          :     0.000010s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.06%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.04%
    opt_after_jit_grad                   :     0.000556s :  2.15%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000048s :  0.19%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009517s : 36.82%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000282    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000002s :      2: substitution.fold_const_symbol
 2.26% :     0.000006s :      3: substitution.graph_param_transform
86.56% :     0.000244s :      4: substitution.inline
 1.35% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.56% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.10% :     0.000006s :      2: substitution.replace_old_param
 4.78% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005659     2
81.04% :     0.004586s :      1: type_inference.infer
18.96% :     0.001073s :      1: type_inference.specialize
------[replace.]   0.000083     6
68.51% :     0.000057s :      4: replace.inline
31.49% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000254     6
95.09% :     0.000241s :      4: match.inline
 4.91% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   691
 1.56% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.49% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.89% :     0.000001s :     10: predicate.addn_check_dump
 1.06% :     0.000001s :     10: predicate.addn_zero_filter
 2.58% :     0.000003s :     10: predicate.arithmetic_simplify
 1.41% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.89% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 0.98% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.72% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.51% :     0.000001s :      3: predicate.elim_not_effective
 0.73% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     10: predicate.environ_get_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 6.08% :     0.000008s :     22: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.97% :     0.000001s :      6: predicate.less_batch_normalization
 1.35% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.50% :     0.000002s :     15: predicate.load_eliminater
 2.04% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.47% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.71% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 0.91% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 2.47% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.94% :     0.000003s :     10: predicate.reduce_eliminate
 1.16% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     18: predicate.replace_applicator
 0.40% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.11% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.17% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     10: predicate.same_eliminate
 0.62% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.14% :     0.000002s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.79% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 1.72% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.88% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000002s :     10: predicate.tile_eliminate
 1.14% :     0.000002s :     10: predicate.transpose_eliminate
 1.65% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.44% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.07% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.83% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.98% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.41% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.58% :     0.000005s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000762    12
44.06% :     0.000336s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.10% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.84% :     0.000372s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.033428    76
 0.17% :     0.000057s :      1: add_recomputation
 0.17% :     0.000058s :      1: auto_monad
 0.06% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.29% :     0.000430s :      1: bootstrap
 0.12% :     0.000040s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.08% :     0.000027s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.07% :     0.000025s :      1: event_method
 0.04% :     0.000013s :      1: execute
 0.03% :     0.000010s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
21.57% :     0.007212s :      1: jit_opt_a
 0.54% :     0.000180s :      1: jit_opt_after_cconv
 0.20% :     0.000068s :      1: jit_opt_b
 1.79% :     0.000600s :      1: loop_unroll
 6.46% :     0.002161s :      1: mutable_eliminate
 2.96% :     0.000991s :     26: opt.transform.jit_opt_a
 0.17% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000032s :      4: opt.transform.jit_opt_b
 0.05% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000028s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.70% :     0.000569s :      1: opt_after_jit_grad
 0.03% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
11.94% :     0.003991s :      1: py_interpret_to_execute
 0.05% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000019s :      1: remove_dup_value
 1.60% :     0.000536s :      1: renormalize.infer
 1.36% :     0.000454s :      1: renormalize.specialize
 0.03% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000043s :      1: rewriter_after_opt_a
 0.60% :     0.000200s :      1: rewriter_before_opt_a
 0.26% :     0.000088s :      1: symbol_engine_optimizer
28.53% :     0.009536s :      1: task_emit
17.20% :     0.005750s :      1: type_inference
 0.21% :     0.000069s :      1: validate


TotalTime = 0.0184541, [30]
[bootstrap]: 0.00044724
[type_inference]: 0.00570751
[event_method]: 2.104e-05
[auto_monad]: 7.307e-05
[graph_reusing]: 5.97999e-06
[pre_auto_parallel]: 2.63998e-06
[py_interpret_to_execute]: 4.456e-05
[rewriter_before_opt_a]: 8.059e-05
[expand_dump_flag]: 3.83999e-06
[jit_opt_a]: 0.00919873, [2]
    [Cycle 1]: 0.00229939, [27]
        [switch_simplify]: 9.118e-05
        [loop_unroll]: 3.216e-05
        [a_1]: 0.00069758
        [with_stream_mark]: 2.568e-05
        [recompute_prepare]: 8.35999e-06
        [updatestate_depend_eliminate]: 4.18999e-06
        [updatestate_assign_eliminate]: 3.03e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 2.34999e-06
        [specialize_transform]: 6.33998e-06
        [updatestate_useless_node_eliminater]: 5.42001e-06
        [accelerated_algorithm]: 5.94e-06
        [meta_shard_fg_expand]: 3.3e-06
        [get_grad_eliminate_]: 6.14001e-06
        [merge_forward]: 4e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.72e-05
        [j_node_and_user_rematch]: 1.064e-05
        [meta_fg_expand]: 2.99999e-06
        [replace_old_param]: 1.053e-05
        [inline_without_move]: 6.10002e-06
        [renormalize]: 0.00105869
        [add_forward_monad_depend]: 7.74002e-06
        [auto_monad_grad]: 3.15998e-06
        [auto_monad_eliminator]: 1.908e-05
        [cse]: 3.245e-05
        [replace_applicator]: 1.749e-05
    [Cycle 2]: 0.00037026, [27]
        [switch_simplify]: 6.38e-06
        [loop_unroll]: 5.51998e-06
        [a_1]: 0.00010467
        [with_stream_mark]: 1.286e-05
        [recompute_prepare]: 5.69e-06
        [updatestate_depend_eliminate]: 3.26001e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 5.25001e-06
        [updatestate_useless_node_eliminater]: 5.62001e-06
        [accelerated_algorithm]: 5.48002e-06
        [meta_shard_fg_expand]: 2.11e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 1.67001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.726e-05
        [j_node_and_user_rematch]: 9.44e-06
        [meta_fg_expand]: 2.22999e-06
        [replace_old_param]: 8.02e-06
        [inline_without_move]: 5.22999e-06
        [renormalize]: 1.19995e-07
        [add_forward_monad_depend]: 1.24998e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 7.15e-06
        [cse]: 1.259e-05
        [replace_applicator]: 6.11e-06
[py_interpret_to_execute_after_opt_a]: 1.564e-05
[rewriter_after_opt_a]: 4.351e-05
[convert_after_rewriter]: 7.16001e-06
[order_py_execute_after_rewriter]: 5.42001e-06
[mutable_eliminate]: 0.00092294
[jit_opt_b]: 5.994e-05, [1]
    [Cycle 1]: 4.914e-05, [2]
        [frontend_op_eliminate]: 1.925e-05
        [inline_after_opt_a]: 1.701e-05
[cconv]: 3.28e-05
[loop_unroll]: 0.00051587
[jit_opt_after_cconv]: 0.00017367, [1]
    [Cycle 1]: 0.00016525, [11]
        [c_1]: 2.25e-05
        [parameter_eliminate]: 4.51002e-06
        [updatestate_depend_eliminate]: 9.37999e-06
        [updatestate_assign_eliminate]: 2.53998e-06
        [updatestate_loads_eliminate]: 2.47001e-06
        [cse]: 3.189e-05
        [call_graph_tuple_transform]: 2.45e-05
        [tuple_list_get_item_eliminator]: 5.51e-06
        [none_parameter_eliminate]: 1.82001e-06
        [renormalize]: 8.2e-07
        [switch_simplify]: 5.40001e-06
[remove_dup_value]: 1.722e-05
[partial_unused_args_eliminate]: 2.49001e-06
[environ_conv]: 6.98e-06
[add_recomputation]: 6.05e-05
[cse_after_recomputation]: 2.395e-05, [1]
    [Cycle 1]: 1.728e-05, [1]
        [cse]: 1.121e-05
[auto_monad_reorder]: 1.723e-05
[get_jit_bprop_graph]: 2.43e-06
[rewriter_after_jit_bprop_graph]: 8.63001e-06
[opt_after_jit_grad]: 0.0005829
[symbol_engine_optimizer]: 9.031e-05, [1]
    [Cycle 1]: 8.246e-05, [6]
        [build]: 5.91e-06
        [elim_shapecalc]: 1.001e-05
        [elim_not_effective]: 1.639e-05
        [opt_reshape]: 7.75e-06
        [fold_const_symbol]: 9.62001e-06
        [renormalize]: 1.00999e-06
[validate]: 4.942e-05
Sums
    bootstrap                            :     0.000447s :  3.99%
    type_inference                       :     0.005708s : 50.89%
    event_method                         :     0.000021s :  0.19%
    auto_monad                           :     0.000073s :  0.65%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000045s :  0.40%
    rewriter_before_opt_a                :     0.000081s :  0.72%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000098s :  0.87%
    jit_opt_a.loop_unroll                :     0.000038s :  0.34%
    jit_opt_a.a_1                        :     0.000802s :  7.15%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.34%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.10%
    jit_opt_a.merge_forward              :     0.000008s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000019s :  0.17%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.001059s :  9.44%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.23%
    jit_opt_a.cse                        :     0.000045s :  0.40%
    jit_opt_a.replace_applicator         :     0.000024s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.14%
    rewriter_after_opt_a                 :     0.000044s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000923s :  8.23%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.15%
    cconv                                :     0.000033s :  0.29%
    loop_unroll                          :     0.000516s :  4.60%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000032s :  0.28%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000017s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000007s :  0.06%
    add_recomputation                    :     0.000060s :  0.54%
    cse_after_recomputation.cse          :     0.000011s :  0.10%
    auto_monad_reorder                   :     0.000017s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.08%
    opt_after_jit_grad                   :     0.000583s :  5.20%
    symbol_engine_optimizer.build        :     0.000006s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000049s :  0.44%


Time group info:
------[substitution.]   0.000311    23
 0.83% :     0.000003s :      2: substitution.elim_not_effective
 0.51% :     0.000002s :      2: substitution.fold_const_symbol
 2.09% :     0.000006s :      3: substitution.graph_param_transform
86.83% :     0.000270s :      4: substitution.inline
 1.21% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.12% :     0.000007s :      4: substitution.remove_not_recompute_node
 2.02% :     0.000006s :      2: substitution.replace_old_param
 4.39% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005621     2
81.80% :     0.004598s :      1: type_inference.infer
18.20% :     0.001023s :      1: type_inference.specialize
------[replace.]   0.000087     6
68.67% :     0.000060s :      4: replace.inline
31.33% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000280     6
95.50% :     0.000267s :      4: match.inline
 4.50% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000151   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.49% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.85% :     0.000001s :     10: predicate.addn_check_dump
 1.65% :     0.000002s :     10: predicate.addn_zero_filter
 2.31% :     0.000003s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.86% :     0.000001s :     10: predicate.compare_switch_simplify
 2.13% :     0.000003s :     10: predicate.depend_value_elim
 0.96% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.55% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.06% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.04% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.33% :     0.000002s :     10: predicate.environ_get_eliminate
 0.84% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.19% :     0.000000s :      3: predicate.graph_param_transform
 5.10% :     0.000008s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000002s :      6: predicate.less_batch_normalization
 1.25% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000003s :     15: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.44% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.89% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000002s :     10: predicate.minmaximum_grad
 2.14% :     0.000003s :      3: predicate.mutable_eliminate
 0.38% :     0.000001s :      3: predicate.opt_reshape
 1.88% :     0.000003s :     15: predicate.partial_eliminate
 1.36% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.49% :     0.000002s :     18: predicate.replace_applicator
 0.38% :     0.000001s :      6: predicate.replace_old_param
 0.39% :     0.000001s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.45% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.93% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.32% :     0.000002s :      6: predicate.special_op_eliminate
 0.85% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.00% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 1.92% :     0.000003s :     16: predicate.switch_defer_inline
 1.61% :     0.000002s :     16: predicate.switch_layer_defer_inline
11.98% :     0.000018s :     53: predicate.switch_simplify
 1.25% :     0.000002s :     10: predicate.tile_eliminate
 1.32% :     0.000002s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.64% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.93% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.33% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.51% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000003s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000785    12
47.93% :     0.000376s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.75% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.32% :     0.000348s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020637    72
 0.31% :     0.000064s :      1: add_recomputation
 0.37% :     0.000077s :      1: auto_monad
 0.10% :     0.000020s :      1: auto_monad_reorder
 2.30% :     0.000474s :      1: bootstrap
 0.17% :     0.000036s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.13% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000011s :      1: environ_conv
 0.12% :     0.000026s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
44.60% :     0.009204s :      1: jit_opt_a
 0.86% :     0.000177s :      1: jit_opt_after_cconv
 0.30% :     0.000062s :      1: jit_opt_b
 2.55% :     0.000527s :      1: loop_unroll
 4.55% :     0.000938s :      1: mutable_eliminate
 5.18% :     0.001070s :     26: opt.transform.jit_opt_a
 0.26% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000029s :      4: opt.transform.jit_opt_b
 0.07% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000040s :      4: opt.transform.symbol_engine_opt
 2.88% :     0.000595s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000047s :      1: py_interpret_to_execute
 0.09% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000020s :      1: remove_dup_value
 2.72% :     0.000561s :      1: renormalize.infer
 2.37% :     0.000488s :      1: renormalize.specialize
 0.05% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000047s :      1: rewriter_after_opt_a
 0.41% :     0.000084s :      1: rewriter_before_opt_a
 0.45% :     0.000093s :      1: symbol_engine_optimizer
27.76% :     0.005728s :      1: type_inference


TotalTime = 0.021117, [33]
[bootstrap]: 0.00034524
[type_inference]: 0.00508868
[event_method]: 1.779e-05
[auto_monad]: 6.848e-05
[graph_reusing]: 5.69999e-06
[pre_auto_parallel]: 2.58e-06
[py_interpret_to_execute]: 3.675e-05
[rewriter_before_opt_a]: 7.041e-05
[expand_dump_flag]: 3.09999e-06
[jit_opt_a]: 0.00613477, [2]
    [Cycle 1]: 0.002042, [27]
        [switch_simplify]: 5.845e-05
        [loop_unroll]: 3.239e-05
        [a_1]: 0.00061052
        [with_stream_mark]: 1.955e-05
        [recompute_prepare]: 9.05999e-06
        [updatestate_depend_eliminate]: 4.23001e-06
        [updatestate_assign_eliminate]: 2.79999e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [parameter_eliminate]: 2.07001e-06
        [specialize_transform]: 6.45002e-06
        [updatestate_useless_node_eliminater]: 5.54e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 2.22001e-06
        [get_grad_eliminate_]: 5.81998e-06
        [merge_forward]: 4.15999e-06
        [cell_reuse_recompute_pass]: 1.30001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.688e-05
        [j_node_and_user_rematch]: 1.002e-05
        [meta_fg_expand]: 2.50002e-06
        [replace_old_param]: 1.104e-05
        [inline_without_move]: 5.94e-06
        [renormalize]: 0.00096623
        [add_forward_monad_depend]: 7.61001e-06
        [auto_monad_grad]: 2.88e-06
        [auto_monad_eliminator]: 1.672e-05
        [cse]: 2.62e-05
        [replace_applicator]: 1.59e-05
    [Cycle 2]: 0.00033738, [27]
        [switch_simplify]: 6.12001e-06
        [loop_unroll]: 5.09998e-06
        [a_1]: 9.275e-05
        [with_stream_mark]: 1.156e-05
        [recompute_prepare]: 5.22e-06
        [updatestate_depend_eliminate]: 3.03e-06
        [updatestate_assign_eliminate]: 2.03997e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 5.25001e-06
        [updatestate_useless_node_eliminater]: 4.72e-06
        [accelerated_algorithm]: 5.17e-06
        [meta_shard_fg_expand]: 1.25001e-06
        [get_grad_eliminate_]: 4.92999e-06
        [merge_forward]: 3.15002e-06
        [cell_reuse_recompute_pass]: 1.97999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.519e-05
        [j_node_and_user_rematch]: 9.20999e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 1.353e-05
        [inline_without_move]: 4.76002e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.87001e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 5.26002e-06
        [cse]: 1.099e-05
        [replace_applicator]: 5.12e-06
[py_interpret_to_execute_after_opt_a]: 1.339e-05
[rewriter_after_opt_a]: 3.9e-05
[convert_after_rewriter]: 6.89001e-06
[order_py_execute_after_rewriter]: 7.75e-06
[mutable_eliminate]: 0.00075713
[jit_opt_b]: 5.453e-05, [1]
    [Cycle 1]: 4.616e-05, [2]
        [frontend_op_eliminate]: 1.707e-05
        [inline_after_opt_a]: 1.659e-05
[cconv]: 2.953e-05
[loop_unroll]: 0.00048496
[jit_opt_after_cconv]: 0.00015459, [1]
    [Cycle 1]: 0.00014765, [11]
        [c_1]: 2.24e-05
        [parameter_eliminate]: 2.99001e-06
        [updatestate_depend_eliminate]: 6.33e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [cse]: 2.562e-05
        [call_graph_tuple_transform]: 2.112e-05
        [tuple_list_get_item_eliminator]: 5.74999e-06
        [none_parameter_eliminate]: 1.84e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 5.74999e-06
[remove_dup_value]: 1.421e-05
[partial_unused_args_eliminate]: 2.71e-06
[environ_conv]: 6.03002e-06
[add_recomputation]: 5.272e-05
[cse_after_recomputation]: 2.184e-05, [1]
    [Cycle 1]: 1.655e-05, [1]
        [cse]: 1.016e-05
[auto_monad_reorder]: 1.529e-05
[get_jit_bprop_graph]: 2.14e-06
[rewriter_after_jit_bprop_graph]: 6.99001e-06
[opt_after_jit_grad]: 0.00051003
[symbol_engine_optimizer]: 7.909e-05, [1]
    [Cycle 1]: 7.15e-05, [6]
        [build]: 4.25999e-06
        [elim_shapecalc]: 7.75998e-06
        [elim_not_effective]: 1.42e-05
        [opt_reshape]: 5.49e-06
        [fold_const_symbol]: 9.79e-06
        [renormalize]: 4.69998e-07
[validate]: 4.26e-05
[backend_pass]: 1.00999e-06
[task_emit]: 0.0068048
[execute]: 9.51e-06
Sums
    bootstrap                            :     0.000345s :  2.07%
    type_inference                       :     0.005089s : 30.46%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000068s :  0.41%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.22%
    rewriter_before_opt_a                :     0.000070s :  0.42%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.39%
    jit_opt_a.loop_unroll                :     0.000037s :  0.22%
    jit_opt_a.a_1                        :     0.000703s :  4.21%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000025s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000966s :  5.78%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.13%
    jit_opt_a.cse                        :     0.000037s :  0.22%
    jit_opt_a.replace_applicator         :     0.000021s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000039s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000008s :  0.05%
    mutable_eliminate                    :     0.000757s :  4.53%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.10%
    cconv                                :     0.000030s :  0.18%
    loop_unroll                          :     0.000485s :  2.90%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000053s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000510s :  3.05%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000043s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006805s : 40.73%
    execute                              :     0.000010s :  0.06%


TotalTime = 0.0304073, [33]
[bootstrap]: 0.00041083
[type_inference]: 0.00596746
[event_method]: 1.941e-05
[auto_monad]: 7.013e-05
[graph_reusing]: 6.01e-06
[pre_auto_parallel]: 2.80997e-06
[py_interpret_to_execute]: 5.302e-05
[rewriter_before_opt_a]: 0.00014605
[expand_dump_flag]: 3.36001e-06
[jit_opt_a]: 0.0139069, [2]
    [Cycle 1]: 0.00197738, [27]
        [switch_simplify]: 7.231e-05
        [loop_unroll]: 2.944e-05
        [a_1]: 0.00063686
        [with_stream_mark]: 2.002e-05
        [recompute_prepare]: 8.45001e-06
        [updatestate_depend_eliminate]: 3.88001e-06
        [updatestate_assign_eliminate]: 3.16999e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 1.94999e-06
        [specialize_transform]: 6.41e-06
        [updatestate_useless_node_eliminater]: 4.97999e-06
        [accelerated_algorithm]: 6.56e-06
        [meta_shard_fg_expand]: 2.51e-06
        [get_grad_eliminate_]: 5.57001e-06
        [merge_forward]: 4e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.629e-05
        [j_node_and_user_rematch]: 9.72001e-06
        [meta_fg_expand]: 2.46e-06
        [replace_old_param]: 9.56e-06
        [inline_without_move]: 5.50001e-06
        [renormalize]: 0.00086501
        [add_forward_monad_depend]: 7.6e-06
        [auto_monad_grad]: 1.96998e-06
        [auto_monad_eliminator]: 1.605e-05
        [cse]: 3.171e-05
        [replace_applicator]: 1.517e-05
    [Cycle 2]: 0.00034712, [27]
        [switch_simplify]: 5.77999e-06
        [loop_unroll]: 5.64e-06
        [a_1]: 9.906e-05
        [with_stream_mark]: 1.061e-05
        [recompute_prepare]: 5.75001e-06
        [updatestate_depend_eliminate]: 2.94001e-06
        [updatestate_assign_eliminate]: 2.43002e-06
        [updatestate_loads_eliminate]: 2.13998e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 4.80999e-06
        [updatestate_useless_node_eliminater]: 4.63001e-06
        [accelerated_algorithm]: 5.82001e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 4.59998e-06
        [merge_forward]: 3.18e-06
        [cell_reuse_recompute_pass]: 3.54002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.436e-05
        [j_node_and_user_rematch]: 8.24998e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 8.38001e-06
        [inline_without_move]: 1.156e-05
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.70001e-06
        [auto_monad_grad]: 1.15999e-06
        [auto_monad_eliminator]: 6.15002e-06
        [cse]: 1.163e-05
        [replace_applicator]: 5.05999e-06
[py_interpret_to_execute_after_opt_a]: 1.182e-05
[rewriter_after_opt_a]: 3.966e-05
[convert_after_rewriter]: 7.36999e-06
[order_py_execute_after_rewriter]: 5.28002e-06
[mutable_eliminate]: 0.00086347
[jit_opt_b]: 5.777e-05, [1]
    [Cycle 1]: 4.875e-05, [2]
        [frontend_op_eliminate]: 1.892e-05
        [inline_after_opt_a]: 1.61e-05
[cconv]: 2.941e-05
[loop_unroll]: 0.00048059
[jit_opt_after_cconv]: 0.00015586, [1]
    [Cycle 1]: 0.00014889, [11]
        [c_1]: 2.231e-05
        [parameter_eliminate]: 2.86999e-06
        [updatestate_depend_eliminate]: 8.63001e-06
        [updatestate_assign_eliminate]: 2.61999e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [cse]: 2.532e-05
        [call_graph_tuple_transform]: 2.118e-05
        [tuple_list_get_item_eliminator]: 5.49e-06
        [none_parameter_eliminate]: 1.79998e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.39e-06
[remove_dup_value]: 1.595e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 5.57999e-06
[add_recomputation]: 5.374e-05
[cse_after_recomputation]: 2.18e-05, [1]
    [Cycle 1]: 1.636e-05, [1]
        [cse]: 9.92999e-06
[auto_monad_reorder]: 1.613e-05
[get_jit_bprop_graph]: 2.11998e-06
[rewriter_after_jit_bprop_graph]: 7.68001e-06
[opt_after_jit_grad]: 0.00057428
[symbol_engine_optimizer]: 8.525e-05, [1]
    [Cycle 1]: 7.691e-05, [6]
        [build]: 4.2e-06
        [elim_shapecalc]: 1.01e-05
        [elim_not_effective]: 1.551e-05
        [opt_reshape]: 8e-06
        [fold_const_symbol]: 9.14998e-06
        [renormalize]: 7.30011e-07
[validate]: 4.768e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.00706856
[execute]: 9.08002e-06
Sums
    bootstrap                            :     0.000411s :  2.27%
    type_inference                       :     0.005967s : 32.90%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000070s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000053s :  0.29%
    rewriter_before_opt_a                :     0.000146s :  0.81%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000078s :  0.43%
    jit_opt_a.loop_unroll                :     0.000035s :  0.19%
    jit_opt_a.a_1                        :     0.000736s :  4.06%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000017s :  0.09%
    jit_opt_a.renormalize                :     0.000865s :  4.77%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.12%
    jit_opt_a.cse                        :     0.000043s :  0.24%
    jit_opt_a.replace_applicator         :     0.000020s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000040s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000863s :  4.76%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000029s :  0.16%
    loop_unroll                          :     0.000481s :  2.65%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000054s :  0.30%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000574s :  3.17%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000048s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007069s : 38.97%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000256    23
 1.02% :     0.000003s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.27% :     0.000006s :      3: substitution.graph_param_transform
85.00% :     0.000217s :      4: substitution.inline
 1.63% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.14% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000005s :      2: substitution.replace_old_param
 5.35% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005021     2
81.67% :     0.004101s :      1: type_inference.infer
18.33% :     0.000920s :      1: type_inference.specialize
------[replace.]   0.000076     6
68.24% :     0.000052s :      4: replace.inline
31.76% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000227     6
94.47% :     0.000215s :      4: match.inline
 5.53% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.14% :     0.000003s :     10: predicate.arithmetic_simplify
 1.46% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.48% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000001s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 6.37% :     0.000008s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.39% :     0.000002s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.25% :     0.000002s :     10: predicate.minmaximum_grad
 2.19% :     0.000003s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000002s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.95% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.05% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.78% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.21% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.71% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.24% :     0.000002s :     10: predicate.transpose_eliminate
 1.25% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.36% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.87% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000716    12
48.78% :     0.000349s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.80% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.43% :     0.000318s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023135    76
 0.24% :     0.000056s :      1: add_recomputation
 0.31% :     0.000072s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.59% :     0.000367s :      1: bootstrap
 0.14% :     0.000032s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.54% :     0.006140s :      1: jit_opt_a
 0.68% :     0.000158s :      1: jit_opt_after_cconv
 0.25% :     0.000057s :      1: jit_opt_b
 2.13% :     0.000494s :      1: loop_unroll
 3.33% :     0.000770s :      1: mutable_eliminate
 4.07% :     0.000941s :     26: opt.transform.jit_opt_a
 0.22% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.25% :     0.000521s :      1: opt_after_jit_grad
 0.04% :     0.000010s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000040s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.36% :     0.000545s :      1: renormalize.infer
 1.78% :     0.000411s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000042s :      1: rewriter_after_opt_a
 0.32% :     0.000073s :      1: rewriter_before_opt_a
 0.36% :     0.000082s :      1: symbol_engine_optimizer
29.50% :     0.006824s :      1: task_emit
22.06% :     0.005104s :      1: type_inference
 0.27% :     0.000063s :      1: validate


Time group info:
------[substitution.]   0.000259    23
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.42% :     0.000006s :      3: substitution.graph_param_transform
85.15% :     0.000221s :      4: substitution.inline
 1.47% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.15% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.23% :     0.000006s :      2: substitution.replace_old_param
 5.30% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005878     2
82.27% :     0.004836s :      1: type_inference.infer
17.73% :     0.001042s :      1: type_inference.specialize
------[replace.]   0.000082     6
67.72% :     0.000056s :      4: replace.inline
32.28% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000231     6
94.52% :     0.000218s :      4: match.inline
 5.48% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   691
 2.04% :     0.000003s :     10: predicate.accumulaten_eliminater
 1.97% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.87% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 2.83% :     0.000004s :     10: predicate.arithmetic_simplify
 1.90% :     0.000003s :     10: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.87% :     0.000001s :     10: predicate.compare_switch_simplify
 1.58% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.16% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.19% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.86% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.56% :     0.000002s :     10: predicate.environ_get_eliminate
 1.11% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000001s :      3: predicate.graph_param_transform
 5.31% :     0.000007s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.58% :     0.000002s :      6: predicate.less_batch_normalization
 1.92% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.56% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.50% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.40% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.94% :     0.000001s :     10: predicate.merge_addn
 0.90% :     0.000001s :     10: predicate.minmaximum_grad
 2.09% :     0.000003s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 1.86% :     0.000003s :     15: predicate.partial_eliminate
 0.97% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.85% :     0.000003s :     10: predicate.reduce_eliminate
 1.34% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.76% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.39% :     0.000002s :     18: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.48% :     0.000001s :      3: predicate.reset_defer_inline
 1.66% :     0.000002s :     10: predicate.reshape_eliminate
 1.08% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.01% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.50% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.67% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.12% :     0.000003s :     16: predicate.switch_defer_inline
 1.79% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.81% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.59% :     0.000002s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.56% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.47% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.32% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.35% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.36% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.49% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000757    12
51.88% :     0.000393s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.57% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.55% :     0.000330s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.032366    76
 0.18% :     0.000057s :      1: add_recomputation
 0.23% :     0.000074s :      1: auto_monad
 0.06% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.34% :     0.000434s :      1: bootstrap
 0.10% :     0.000032s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.08% :     0.000024s :      1: event_method
 0.05% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
42.98% :     0.013911s :      1: jit_opt_a
 0.49% :     0.000159s :      1: jit_opt_after_cconv
 0.19% :     0.000061s :      1: jit_opt_b
 1.51% :     0.000490s :      1: loop_unroll
 2.71% :     0.000877s :      1: mutable_eliminate
 3.02% :     0.000979s :     26: opt.transform.jit_opt_a
 0.16% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000028s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.81% :     0.000585s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000057s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000018s :      1: remove_dup_value
 1.45% :     0.000471s :      1: renormalize.infer
 1.19% :     0.000385s :      1: renormalize.specialize
 0.03% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000044s :      1: rewriter_after_opt_a
 0.47% :     0.000151s :      1: rewriter_before_opt_a
 0.27% :     0.000089s :      1: symbol_engine_optimizer
21.90% :     0.007087s :      1: task_emit
18.51% :     0.005991s :      1: type_inference
 0.22% :     0.000071s :      1: validate

.
TotalTime = 0.0527151, [33]
[bootstrap]: 0.00028306
[type_inference]: 0.0047322
[event_method]: 1.885e-05
[auto_monad]: 6.332e-05
[graph_reusing]: 5.84999e-06
[pre_auto_parallel]: 2.59001e-06
[py_interpret_to_execute]: 3.647e-05
[rewriter_before_opt_a]: 7.52e-05
[expand_dump_flag]: 3.08e-06
[jit_opt_a]: 0.0380989, [2]
    [Cycle 1]: 0.0340353, [27]
        [switch_simplify]: 6.522e-05
        [loop_unroll]: 3.042e-05
        [a_1]: 0.00059331
        [with_stream_mark]: 2.279e-05
        [recompute_prepare]: 6.84999e-06
        [updatestate_depend_eliminate]: 3.91999e-06
        [updatestate_assign_eliminate]: 3.12002e-06
        [updatestate_loads_eliminate]: 2.88e-06
        [parameter_eliminate]: 1.85001e-06
        [specialize_transform]: 6.06e-06
        [updatestate_useless_node_eliminater]: 5.40001e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 2.58003e-06
        [get_grad_eliminate_]: 5.44e-06
        [merge_forward]: 3.48999e-06
        [cell_reuse_recompute_pass]: 1.20999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.609e-05
        [j_node_and_user_rematch]: 9.13002e-06
        [meta_fg_expand]: 2.22999e-06
        [replace_old_param]: 1.019e-05
        [inline_without_move]: 5.42999e-06
        [renormalize]: 0.00084476
        [add_forward_monad_depend]: 6.41e-06
        [auto_monad_grad]: 2.69001e-06
        [auto_monad_eliminator]: 1.423e-05
        [cse]: 3.395e-05
        [replace_applicator]: 0.0320895
    [Cycle 2]: 0.00058845, [27]
        [switch_simplify]: 2.075e-05
        [loop_unroll]: 7.1e-06
        [a_1]: 0.00012229
        [with_stream_mark]: 4.089e-05
        [recompute_prepare]: 6.39001e-06
        [updatestate_depend_eliminate]: 4.60001e-06
        [updatestate_assign_eliminate]: 3.4e-06
        [updatestate_loads_eliminate]: 2.94999e-06
        [parameter_eliminate]: 3.11001e-06
        [specialize_transform]: 5.86e-06
        [updatestate_useless_node_eliminater]: 5.97001e-06
        [accelerated_algorithm]: 8.71002e-06
        [meta_shard_fg_expand]: 3.27002e-06
        [get_grad_eliminate_]: 5.86e-06
        [merge_forward]: 4.68999e-06
        [cell_reuse_recompute_pass]: 4.94e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.826e-05
        [j_node_and_user_rematch]: 1.016e-05
        [meta_fg_expand]: 3.84002e-06
        [replace_old_param]: 1.248e-05
        [inline_without_move]: 4.95999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 5.10999e-06
        [auto_monad_grad]: 3.35998e-06
        [auto_monad_eliminator]: 1.87e-05
        [cse]: 4.469e-05
        [replace_applicator]: 8.48999e-06
[py_interpret_to_execute_after_opt_a]: 2.777e-05
[rewriter_after_opt_a]: 4.759e-05
[convert_after_rewriter]: 7.31001e-06
[order_py_execute_after_rewriter]: 4.90001e-06
[mutable_eliminate]: 0.00085504
[jit_opt_b]: 5.674e-05, [1]
    [Cycle 1]: 4.731e-05, [2]
        [frontend_op_eliminate]: 1.809e-05
        [inline_after_opt_a]: 1.674e-05
[cconv]: 3.951e-05
[loop_unroll]: 0.00044568
[jit_opt_after_cconv]: 0.00014944, [1]
    [Cycle 1]: 0.00014217, [11]
        [c_1]: 2.196e-05
        [parameter_eliminate]: 3.06001e-06
        [updatestate_depend_eliminate]: 6.38003e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [cse]: 2.121e-05
        [call_graph_tuple_transform]: 1.924e-05
        [tuple_list_get_item_eliminator]: 5.62001e-06
        [none_parameter_eliminate]: 1.69e-06
        [renormalize]: 7.10017e-07
        [switch_simplify]: 5.71e-06
[remove_dup_value]: 1.513e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 5.30001e-06
[add_recomputation]: 5.608e-05
[cse_after_recomputation]: 2.165e-05, [1]
    [Cycle 1]: 1.558e-05, [1]
        [cse]: 9.82001e-06
[auto_monad_reorder]: 1.501e-05
[get_jit_bprop_graph]: 2.07001e-06
[rewriter_after_jit_bprop_graph]: 8.43001e-06
[opt_after_jit_grad]: 0.00045568
[symbol_engine_optimizer]: 7.917e-05, [1]
    [Cycle 1]: 7.246e-05, [6]
        [build]: 5.51002e-06
        [elim_shapecalc]: 8.34998e-06
        [elim_not_effective]: 1.45e-05
        [opt_reshape]: 6.39001e-06
        [fold_const_symbol]: 8.79e-06
        [renormalize]: 3.50003e-07
[validate]: 4.398e-05
[backend_pass]: 1.00001e-06
[task_emit]: 0.00680461
[execute]: 1.042e-05
Sums
    bootstrap                            :     0.000283s :  0.58%
    type_inference                       :     0.004732s :  9.77%
    event_method                         :     0.000019s :  0.04%
    auto_monad                           :     0.000063s :  0.13%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000036s :  0.08%
    rewriter_before_opt_a                :     0.000075s :  0.16%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000086s :  0.18%
    jit_opt_a.loop_unroll                :     0.000038s :  0.08%
    jit_opt_a.a_1                        :     0.000716s :  1.48%
    jit_opt_a.with_stream_mark           :     0.000064s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000014s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.02%
    jit_opt_a.merge_forward              :     0.000008s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000044s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.01%
    jit_opt_a.replace_old_param          :     0.000023s :  0.05%
    jit_opt_a.inline_without_move        :     0.000010s :  0.02%
    jit_opt_a.renormalize                :     0.000845s :  1.74%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000006s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000033s :  0.07%
    jit_opt_a.cse                        :     0.000079s :  0.16%
    jit_opt_a.replace_applicator         :     0.032098s : 66.27%
    py_interpret_to_execute_after_opt_a  :     0.000028s :  0.06%
    rewriter_after_opt_a                 :     0.000048s :  0.10%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000855s :  1.77%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.03%
    cconv                                :     0.000040s :  0.08%
    loop_unroll                          :     0.000446s :  0.92%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.05%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000021s :  0.04%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.01%
    remove_dup_value                     :     0.000015s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000005s :  0.01%
    add_recomputation                    :     0.000056s :  0.12%
    cse_after_recomputation.cse          :     0.000010s :  0.02%
    auto_monad_reorder                   :     0.000015s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000456s :  0.94%
    symbol_engine_optimizer.build        :     0.000006s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000044s :  0.09%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006805s : 14.05%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000253    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 2.21% :     0.000006s :      3: substitution.graph_param_transform
83.67% :     0.000212s :      4: substitution.inline
 1.62% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.19% :     0.000006s :      4: substitution.remove_not_recompute_node
 3.81% :     0.000010s :      2: substitution.replace_old_param
 5.16% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004660     2
81.09% :     0.003779s :      1: type_inference.infer
18.91% :     0.000881s :      1: type_inference.specialize
------[replace.]   0.000076     6
67.26% :     0.000051s :      4: replace.inline
32.74% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000221     6
94.60% :     0.000209s :      4: match.inline
 5.40% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.91% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.88% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 2.67% :     0.000004s :     10: predicate.arithmetic_simplify
 1.53% :     0.000002s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.88% :     0.000001s :     10: predicate.compare_switch_simplify
 1.42% :     0.000002s :     10: predicate.depend_value_elim
 1.44% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.06% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.07% :     0.000001s :     10: predicate.environ_get_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.33% :     0.000000s :      3: predicate.graph_param_transform
 6.52% :     0.000009s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.75% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000002s :     10: predicate.minmaximum_grad
 1.64% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.99% :     0.000004s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.99% :     0.000001s :     10: predicate.same_eliminate
 0.64% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.80% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.57% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.22% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000003s :     16: predicate.switch_layer_defer_inline
 9.19% :     0.000013s :     53: predicate.switch_simplify
 1.12% :     0.000002s :     10: predicate.tile_eliminate
 1.00% :     0.000001s :     10: predicate.transpose_eliminate
 1.19% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.66% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.43% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000667    12
47.55% :     0.000317s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.52% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.93% :     0.000300s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.086709    76
 0.07% :     0.000059s :      1: add_recomputation
 0.08% :     0.000067s :      1: auto_monad
 0.02% :     0.000018s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.35% :     0.000302s :      1: bootstrap
 0.05% :     0.000042s :      1: cconv
 0.01% :     0.000010s :      1: convert_after_rewriter
 0.03% :     0.000024s :      1: cse_after_recomputation
 0.01% :     0.000008s :      1: environ_conv
 0.03% :     0.000023s :      1: event_method
 0.02% :     0.000017s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000008s :      1: graph_reusing
43.94% :     0.038104s :      1: jit_opt_a
 0.18% :     0.000152s :      1: jit_opt_after_cconv
 0.07% :     0.000059s :      1: jit_opt_b
 0.52% :     0.000455s :      1: loop_unroll
 1.00% :     0.000866s :      1: mutable_eliminate
38.12% :     0.033055s :     26: opt.transform.jit_opt_a
 0.06% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.03% :     0.000028s :      4: opt.transform.jit_opt_b
 0.02% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.02% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.04% :     0.000034s :      4: opt.transform.symbol_engine_opt
 0.53% :     0.000464s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.05% :     0.000040s :      1: py_interpret_to_execute
 0.04% :     0.000031s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000018s :      1: remove_dup_value
 0.51% :     0.000446s :      1: renormalize.infer
 0.45% :     0.000389s :      1: renormalize.specialize
 0.01% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.06% :     0.000051s :      1: rewriter_after_opt_a
 0.09% :     0.000078s :      1: rewriter_before_opt_a
 0.09% :     0.000082s :      1: symbol_engine_optimizer
 7.87% :     0.006823s :      1: task_emit
 5.48% :     0.004748s :      1: type_inference
 0.08% :     0.000068s :      1: validate


TotalTime = 0.0282044, [30]
[bootstrap]: 0.00040578
[type_inference]: 0.00505621
[event_method]: 1.775e-05
[auto_monad]: 6.721e-05
[graph_reusing]: 6.29001e-06
[pre_auto_parallel]: 2.74999e-06
[py_interpret_to_execute]: 3.857e-05
[rewriter_before_opt_a]: 7.508e-05
[expand_dump_flag]: 3.41001e-06
[jit_opt_a]: 0.0197985, [2]
    [Cycle 1]: 0.00218323, [27]
        [switch_simplify]: 7.324e-05
        [loop_unroll]: 3.207e-05
        [a_1]: 0.00065651
        [with_stream_mark]: 2.709e-05
        [recompute_prepare]: 8.72998e-06
        [updatestate_depend_eliminate]: 4.57e-06
        [updatestate_assign_eliminate]: 2.96999e-06
        [updatestate_loads_eliminate]: 2.98998e-06
        [parameter_eliminate]: 2.03002e-06
        [specialize_transform]: 6.59999e-06
        [updatestate_useless_node_eliminater]: 5.69999e-06
        [accelerated_algorithm]: 6.39999e-06
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 5.58002e-06
        [merge_forward]: 3.84002e-06
        [cell_reuse_recompute_pass]: 1.71e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.662e-05
        [j_node_and_user_rematch]: 1.072e-05
        [meta_fg_expand]: 2.86e-06
        [replace_old_param]: 1.05e-05
        [inline_without_move]: 5.51998e-06
        [renormalize]: 0.00101369
        [add_forward_monad_depend]: 8.03001e-06
        [auto_monad_grad]: 2.96999e-06
        [auto_monad_eliminator]: 1.978e-05
        [cse]: 3.637e-05
        [replace_applicator]: 1.863e-05
    [Cycle 2]: 0.00036584, [27]
        [switch_simplify]: 6.16e-06
        [loop_unroll]: 5.75001e-06
        [a_1]: 0.00010301
        [with_stream_mark]: 1.42e-05
        [recompute_prepare]: 5.33002e-06
        [updatestate_depend_eliminate]: 2.89001e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.37e-06
        [specialize_transform]: 5.37001e-06
        [updatestate_useless_node_eliminater]: 5.24998e-06
        [accelerated_algorithm]: 5.85002e-06
        [meta_shard_fg_expand]: 1.63002e-06
        [get_grad_eliminate_]: 4.68999e-06
        [merge_forward]: 2.87002e-06
        [cell_reuse_recompute_pass]: 1.96e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.641e-05
        [j_node_and_user_rematch]: 8.33001e-06
        [meta_fg_expand]: 1.97999e-06
        [replace_old_param]: 7.74002e-06
        [inline_without_move]: 4.87998e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.57001e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 6.02999e-06
        [cse]: 1.32e-05
        [replace_applicator]: 6.07999e-06
[py_interpret_to_execute_after_opt_a]: 1.381e-05
[rewriter_after_opt_a]: 4.092e-05
[convert_after_rewriter]: 7.96001e-06
[order_py_execute_after_rewriter]: 5.20999e-06
[mutable_eliminate]: 0.00082736
[jit_opt_b]: 6.303e-05, [1]
    [Cycle 1]: 5.274e-05, [2]
        [frontend_op_eliminate]: 2.15e-05
        [inline_after_opt_a]: 1.714e-05
[cconv]: 3.399e-05
[loop_unroll]: 0.00054241
[jit_opt_after_cconv]: 0.00017421, [1]
    [Cycle 1]: 0.00016626, [11]
        [c_1]: 2.366e-05
        [parameter_eliminate]: 4.35999e-06
        [updatestate_depend_eliminate]: 9.20999e-06
        [updatestate_assign_eliminate]: 2.74999e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [cse]: 3.02e-05
        [call_graph_tuple_transform]: 2.304e-05
        [tuple_list_get_item_eliminator]: 5.94999e-06
        [none_parameter_eliminate]: 1.48002e-06
        [renormalize]: 7.00005e-07
        [switch_simplify]: 5.67999e-06
[remove_dup_value]: 1.686e-05
[partial_unused_args_eliminate]: 3.43e-06
[environ_conv]: 7.61999e-06
[add_recomputation]: 5.698e-05
[cse_after_recomputation]: 2.371e-05, [1]
    [Cycle 1]: 1.733e-05, [1]
        [cse]: 1.002e-05
[auto_monad_reorder]: 1.715e-05
[get_jit_bprop_graph]: 2.11003e-06
[rewriter_after_jit_bprop_graph]: 9.41998e-06
[opt_after_jit_grad]: 0.00054228
[symbol_engine_optimizer]: 8.765e-05, [1]
    [Cycle 1]: 7.894e-05, [6]
        [build]: 5.59998e-06
        [elim_shapecalc]: 8.85001e-06
        [elim_not_effective]: 1.69e-05
        [opt_reshape]: 6.59999e-06
        [fold_const_symbol]: 1.053e-05
        [renormalize]: 5.49975e-07
[validate]: 4.78e-05
Sums
    bootstrap                            :     0.000406s :  3.95%
    type_inference                       :     0.005056s : 49.18%
    event_method                         :     0.000018s :  0.17%
    auto_monad                           :     0.000067s :  0.65%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000039s :  0.38%
    rewriter_before_opt_a                :     0.000075s :  0.73%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000079s :  0.77%
    jit_opt_a.loop_unroll                :     0.000038s :  0.37%
    jit_opt_a.a_1                        :     0.000760s :  7.39%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.40%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000018s :  0.18%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.001014s :  9.86%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.25%
    jit_opt_a.cse                        :     0.000050s :  0.48%
    jit_opt_a.replace_applicator         :     0.000025s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.13%
    rewriter_after_opt_a                 :     0.000041s :  0.40%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000827s :  8.05%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.17%
    cconv                                :     0.000034s :  0.33%
    loop_unroll                          :     0.000542s :  5.28%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.09%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000030s :  0.29%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000017s :  0.16%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000008s :  0.07%
    add_recomputation                    :     0.000057s :  0.55%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000017s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.09%
    opt_after_jit_grad                   :     0.000542s :  5.27%
    symbol_engine_optimizer.build        :     0.000006s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000048s :  0.46%


Time group info:
------[substitution.]   0.000290    23
 1.02% :     0.000003s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 2.27% :     0.000007s :      3: substitution.graph_param_transform
85.52% :     0.000248s :      4: substitution.inline
 1.32% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.20% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000006s :      2: substitution.replace_old_param
 5.26% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004984     2
81.39% :     0.004057s :      1: type_inference.infer
18.61% :     0.000928s :      1: type_inference.specialize
------[replace.]   0.000084     6
68.24% :     0.000058s :      4: replace.inline
31.76% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000260     6
94.55% :     0.000246s :      4: match.inline
 5.45% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   691
 2.00% :     0.000003s :     10: predicate.accumulaten_eliminater
 1.72% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 1.80% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000002s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.90% :     0.000001s :     10: predicate.compare_switch_simplify
 2.03% :     0.000003s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.04% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.14% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.95% :     0.000001s :     10: predicate.environ_get_eliminate
 1.25% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.83% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 6.27% :     0.000009s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.91% :     0.000003s :     15: predicate.load_eliminater
 1.82% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.24% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000002s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 2.52% :     0.000003s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 1.99% :     0.000003s :     15: predicate.partial_eliminate
 1.27% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     10: predicate.reduce_eliminate
 1.57% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.50% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.27% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.09% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000002s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.19% :     0.000003s :     16: predicate.switch_defer_inline
 1.75% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.35% :     0.000010s :     53: predicate.switch_simplify
 1.45% :     0.000002s :     10: predicate.tile_eliminate
 1.07% :     0.000001s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.62% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.91% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.46% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.55% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.89% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000733    12
47.65% :     0.000349s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.11% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.24% :     0.000324s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.030290    72
 0.20% :     0.000061s :      1: add_recomputation
 0.23% :     0.000071s :      1: auto_monad
 0.07% :     0.000020s :      1: auto_monad_reorder
 1.40% :     0.000425s :      1: bootstrap
 0.12% :     0.000037s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000026s :      1: cse_after_recomputation
 0.04% :     0.000011s :      1: environ_conv
 0.07% :     0.000022s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
65.38% :     0.019803s :      1: jit_opt_a
 0.59% :     0.000177s :      1: jit_opt_after_cconv
 0.22% :     0.000067s :      1: jit_opt_b
 1.83% :     0.000554s :      1: loop_unroll
 2.78% :     0.000842s :      1: mutable_eliminate
 3.33% :     0.001007s :     26: opt.transform.jit_opt_a
 0.18% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000031s :      4: opt.transform.jit_opt_b
 0.06% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.83% :     0.000554s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000006s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000041s :      1: py_interpret_to_execute
 0.05% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000019s :      1: remove_dup_value
 1.83% :     0.000555s :      1: renormalize.infer
 1.48% :     0.000449s :      1: renormalize.specialize
 0.04% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000044s :      1: rewriter_after_opt_a
 0.26% :     0.000078s :      1: rewriter_before_opt_a
 0.30% :     0.000090s :      1: symbol_engine_optimizer
16.74% :     0.005071s :      1: type_inference


TotalTime = 0.0247687, [33]
[bootstrap]: 0.00035742
[type_inference]: 0.00555138
[event_method]: 2.078e-05
[auto_monad]: 0.00097423
[graph_reusing]: 6.84999e-06
[pre_auto_parallel]: 2.84001e-06
[py_interpret_to_execute]: 4.892e-05
[rewriter_before_opt_a]: 8.017e-05
[expand_dump_flag]: 3.45998e-06
[jit_opt_a]: 0.00598292, [2]
    [Cycle 1]: 0.00183009, [27]
        [switch_simplify]: 6.409e-05
        [loop_unroll]: 2.915e-05
        [a_1]: 0.00056422
        [with_stream_mark]: 2.06e-05
        [recompute_prepare]: 6.89001e-06
        [updatestate_depend_eliminate]: 4.03999e-06
        [updatestate_assign_eliminate]: 3.16999e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 2.54001e-06
        [specialize_transform]: 6.04999e-06
        [updatestate_useless_node_eliminater]: 5.08002e-06
        [accelerated_algorithm]: 5.49998e-06
        [meta_shard_fg_expand]: 2.59001e-06
        [get_grad_eliminate_]: 5.46e-06
        [merge_forward]: 3.93001e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.454e-05
        [j_node_and_user_rematch]: 9.05001e-06
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 9.02999e-06
        [inline_without_move]: 5.19e-06
        [renormalize]: 0.00081946
        [add_forward_monad_depend]: 5.91998e-06
        [auto_monad_grad]: 2.40002e-06
        [auto_monad_eliminator]: 1.438e-05
        [cse]: 3.022e-05
        [replace_applicator]: 1.229e-05
    [Cycle 2]: 0.00031659, [27]
        [switch_simplify]: 5.61e-06
        [loop_unroll]: 4.80999e-06
        [a_1]: 8.928e-05
        [with_stream_mark]: 9.99999e-06
        [recompute_prepare]: 4.85999e-06
        [updatestate_depend_eliminate]: 2.76e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 1.42e-06
        [specialize_transform]: 4.86002e-06
        [updatestate_useless_node_eliminater]: 4.78001e-06
        [accelerated_algorithm]: 4.80001e-06
        [meta_shard_fg_expand]: 1.07998e-06
        [get_grad_eliminate_]: 4.57998e-06
        [merge_forward]: 2.63003e-06
        [cell_reuse_recompute_pass]: 1.74998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.226e-05
        [j_node_and_user_rematch]: 7.5e-06
        [meta_fg_expand]: 1.59998e-06
        [replace_old_param]: 7e-06
        [inline_without_move]: 4.62998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 4.75001e-06
        [cse]: 1.035e-05
        [replace_applicator]: 5.23002e-06
[py_interpret_to_execute_after_opt_a]: 1.208e-05
[rewriter_after_opt_a]: 3.745e-05
[convert_after_rewriter]: 7.04001e-06
[order_py_execute_after_rewriter]: 5.19e-06
[mutable_eliminate]: 0.00062689
[jit_opt_b]: 4.989e-05, [1]
    [Cycle 1]: 4.231e-05, [2]
        [frontend_op_eliminate]: 1.631e-05
        [inline_after_opt_a]: 1.482e-05
[cconv]: 3.911e-05
[loop_unroll]: 0.00127772
[jit_opt_after_cconv]: 0.0001477, [1]
    [Cycle 1]: 0.00014002, [11]
        [c_1]: 2.074e-05
        [parameter_eliminate]: 3.01001e-06
        [updatestate_depend_eliminate]: 6.53998e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.13002e-06
        [cse]: 2.26e-05
        [call_graph_tuple_transform]: 1.843e-05
        [tuple_list_get_item_eliminator]: 5.47001e-06
        [none_parameter_eliminate]: 1.42999e-06
        [renormalize]: 8.50006e-07
        [switch_simplify]: 5.39998e-06
[remove_dup_value]: 1.51e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 5.69999e-06
[add_recomputation]: 5.184e-05
[cse_after_recomputation]: 2.031e-05, [1]
    [Cycle 1]: 1.423e-05, [1]
        [cse]: 8.70001e-06
[auto_monad_reorder]: 1.64e-05
[get_jit_bprop_graph]: 2.26998e-06
[rewriter_after_jit_bprop_graph]: 6.39999e-06
[opt_after_jit_grad]: 0.00046353
[symbol_engine_optimizer]: 7.697e-05, [1]
    [Cycle 1]: 6.993e-05, [6]
        [build]: 5.46002e-06
        [elim_shapecalc]: 7.41999e-06
        [elim_not_effective]: 1.403e-05
        [opt_reshape]: 5.30001e-06
        [fold_const_symbol]: 9.41e-06
        [renormalize]: 3.60014e-07
[validate]: 4.039e-05
[backend_pass]: 9.79984e-07
[task_emit]: 0.00858179
[execute]: 8.84e-06
Sums
    bootstrap                            :     0.000357s :  1.76%
    type_inference                       :     0.005551s : 27.38%
    event_method                         :     0.000021s :  0.10%
    auto_monad                           :     0.000974s :  4.81%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000049s :  0.24%
    rewriter_before_opt_a                :     0.000080s :  0.40%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000070s :  0.34%
    jit_opt_a.loop_unroll                :     0.000034s :  0.17%
    jit_opt_a.a_1                        :     0.000653s :  3.22%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000820s :  4.04%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.09%
    jit_opt_a.cse                        :     0.000041s :  0.20%
    jit_opt_a.replace_applicator         :     0.000018s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.06%
    rewriter_after_opt_a                 :     0.000037s :  0.18%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000627s :  3.09%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.07%
    cconv                                :     0.000039s :  0.19%
    loop_unroll                          :     0.001278s :  6.30%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000052s :  0.26%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000464s :  2.29%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000040s :  0.20%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008582s : 42.33%
    execute                              :     0.000009s :  0.04%


Time group info:
------[substitution.]   0.000224    23
 1.16% :     0.000003s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.44% :     0.000005s :      3: substitution.graph_param_transform
84.60% :     0.000189s :      4: substitution.inline
 1.25% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.84% :     0.000004s :      2: substitution.replace_old_param
 5.88% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005454     2
82.27% :     0.004486s :      1: type_inference.infer
17.73% :     0.000967s :      1: type_inference.specialize
------[replace.]   0.000073     6
68.05% :     0.000050s :      4: replace.inline
31.95% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000199     6
93.95% :     0.000187s :      4: match.inline
 6.05% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.25% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 1.75% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.65% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.41% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      3: predicate.elim_not_effective
 0.76% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.22% :     0.000006s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000001s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 2.01% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.10% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 2.07% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000000s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.79% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.66% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.66% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.87% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.29% :     0.000001s :     10: predicate.transpose_eliminate
 1.58% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.42% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.82% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000695    12
50.33% :     0.000350s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.47% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.20% :     0.000314s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026554    76
 0.21% :     0.000054s :      1: add_recomputation
 3.71% :     0.000986s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.44% :     0.000383s :      1: bootstrap
 0.16% :     0.000042s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.10% :     0.000026s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
22.54% :     0.005986s :      1: jit_opt_a
 0.57% :     0.000150s :      1: jit_opt_after_cconv
 0.20% :     0.000052s :      1: jit_opt_b
 4.84% :     0.001286s :      1: loop_unroll
 2.39% :     0.000636s :      1: mutable_eliminate
 3.27% :     0.000869s :     26: opt.transform.jit_opt_a
 0.18% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.78% :     0.000472s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000052s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.69% :     0.000449s :      1: renormalize.infer
 1.36% :     0.000362s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000040s :      1: rewriter_after_opt_a
 0.31% :     0.000083s :      1: rewriter_before_opt_a
 0.30% :     0.000080s :      1: symbol_engine_optimizer
32.39% :     0.008600s :      1: task_emit
20.99% :     0.005573s :      1: type_inference
 0.23% :     0.000061s :      1: validate


TotalTime = 0.0199044, [33]
[bootstrap]: 0.00029012
[type_inference]: 0.00488736
[event_method]: 1.922e-05
[auto_monad]: 6.705e-05
[graph_reusing]: 6.11e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 3.787e-05
[rewriter_before_opt_a]: 7.331e-05
[expand_dump_flag]: 3.43999e-06
[jit_opt_a]: 0.00575709, [2]
    [Cycle 1]: 0.00187441, [27]
        [switch_simplify]: 6.195e-05
        [loop_unroll]: 2.906e-05
        [a_1]: 0.00057918
        [with_stream_mark]: 2.262e-05
        [recompute_prepare]: 7.59002e-06
        [updatestate_depend_eliminate]: 4.12e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 2.63003e-06
        [parameter_eliminate]: 2.04999e-06
        [specialize_transform]: 5.65001e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 5.30001e-06
        [meta_shard_fg_expand]: 1.87999e-06
        [get_grad_eliminate_]: 5.36002e-06
        [merge_forward]: 3.41001e-06
        [cell_reuse_recompute_pass]: 1.12999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.476e-05
        [j_node_and_user_rematch]: 9.37999e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 9.37001e-06
        [inline_without_move]: 5.32999e-06
        [renormalize]: 0.0008438
        [add_forward_monad_depend]: 6.79001e-06
        [auto_monad_grad]: 2.49001e-06
        [auto_monad_eliminator]: 1.5e-05
        [cse]: 3.101e-05
        [replace_applicator]: 1.302e-05
    [Cycle 2]: 0.00031825, [27]
        [switch_simplify]: 5.74999e-06
        [loop_unroll]: 5.12e-06
        [a_1]: 9.161e-05
        [with_stream_mark]: 1.018e-05
        [recompute_prepare]: 5.00999e-06
        [updatestate_depend_eliminate]: 3.28e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 4.94998e-06
        [updatestate_useless_node_eliminater]: 4.66002e-06
        [accelerated_algorithm]: 5.27999e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 4.3e-06
        [merge_forward]: 2.56e-06
        [cell_reuse_recompute_pass]: 1.42e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.26e-05
        [j_node_and_user_rematch]: 8.2e-06
        [meta_fg_expand]: 1.68002e-06
        [replace_old_param]: 7.39002e-06
        [inline_without_move]: 4.67e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 6.59988e-07
        [auto_monad_eliminator]: 5.02999e-06
        [cse]: 1.036e-05
        [replace_applicator]: 4.78001e-06
[py_interpret_to_execute_after_opt_a]: 1.142e-05
[rewriter_after_opt_a]: 3.651e-05
[convert_after_rewriter]: 7.15e-06
[order_py_execute_after_rewriter]: 4.79e-06
[mutable_eliminate]: 0.00060933
[jit_opt_b]: 5.331e-05, [1]
    [Cycle 1]: 4.567e-05, [2]
        [frontend_op_eliminate]: 1.739e-05
        [inline_after_opt_a]: 1.648e-05
[cconv]: 2.652e-05
[loop_unroll]: 0.00042304
[jit_opt_after_cconv]: 0.00014756, [1]
    [Cycle 1]: 0.00014055, [11]
        [c_1]: 2.109e-05
        [parameter_eliminate]: 3.06999e-06
        [updatestate_depend_eliminate]: 6.12999e-06
        [updatestate_assign_eliminate]: 2.58998e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 2.129e-05
        [call_graph_tuple_transform]: 1.999e-05
        [tuple_list_get_item_eliminator]: 5.39e-06
        [none_parameter_eliminate]: 1.40999e-06
        [renormalize]: 9.70002e-07
        [switch_simplify]: 5.51e-06
[remove_dup_value]: 1.427e-05
[partial_unused_args_eliminate]: 2.07001e-06
[environ_conv]: 5.52999e-06
[add_recomputation]: 5.179e-05
[cse_after_recomputation]: 2.085e-05, [1]
    [Cycle 1]: 1.511e-05, [1]
        [cse]: 9.12999e-06
[auto_monad_reorder]: 1.526e-05
[get_jit_bprop_graph]: 1.87001e-06
[rewriter_after_jit_bprop_graph]: 7.25e-06
[opt_after_jit_grad]: 0.00046244
[symbol_engine_optimizer]: 7.675e-05, [1]
    [Cycle 1]: 7.043e-05, [6]
        [build]: 4.45e-06
        [elim_shapecalc]: 8.41002e-06
        [elim_not_effective]: 1.321e-05
        [opt_reshape]: 5.99e-06
        [fold_const_symbol]: 9.37001e-06
        [renormalize]: 9.29984e-07
[validate]: 3.9e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.0065058
[execute]: 9.46998e-06
Sums
    bootstrap                            :     0.000290s :  1.85%
    type_inference                       :     0.004887s : 31.13%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000067s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.24%
    rewriter_before_opt_a                :     0.000073s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.43%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000671s :  4.27%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000844s :  5.38%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000041s :  0.26%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000609s :  3.88%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000027s :  0.17%
    loop_unroll                          :     0.000423s :  2.69%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000052s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.05%
    opt_after_jit_grad                   :     0.000462s :  2.95%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006506s : 41.44%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000234    23
 0.80% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.28% :     0.000005s :      3: substitution.graph_param_transform
86.15% :     0.000202s :      4: substitution.inline
 1.41% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.86% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.87% :     0.000004s :      2: substitution.replace_old_param
 5.05% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004813     2
81.18% :     0.003908s :      1: type_inference.infer
18.82% :     0.000906s :      1: type_inference.specialize
------[replace.]   0.000073     6
66.82% :     0.000049s :      4: replace.inline
33.18% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000210     6
94.94% :     0.000199s :      4: match.inline
 5.06% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.18% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.23% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.13% :     0.000001s :     10: predicate.addn_zero_filter
 2.34% :     0.000003s :     10: predicate.arithmetic_simplify
 1.38% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.18% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 1.03% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.64% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.46% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.49% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.15% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.20% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 2.47% :     0.000003s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.21% :     0.000001s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.21% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.51% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.74% :     0.000009s :     53: predicate.switch_simplify
 1.45% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.61% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.01% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.08% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.12% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.43% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000668    12
48.35% :     0.000323s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.38% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.27% :     0.000296s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021732    76
 0.25% :     0.000054s :      1: add_recomputation
 0.32% :     0.000070s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.42% :     0.000309s :      1: bootstrap
 0.13% :     0.000029s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000024s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.51% :     0.005760s :      1: jit_opt_a
 0.69% :     0.000150s :      1: jit_opt_after_cconv
 0.26% :     0.000056s :      1: jit_opt_b
 1.98% :     0.000431s :      1: loop_unroll
 2.85% :     0.000620s :      1: mutable_eliminate
 4.08% :     0.000886s :     26: opt.transform.jit_opt_a
 0.22% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.17% :     0.000471s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000041s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 2.10% :     0.000457s :      1: renormalize.infer
 1.74% :     0.000378s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000039s :      1: rewriter_after_opt_a
 0.35% :     0.000076s :      1: rewriter_before_opt_a
 0.36% :     0.000079s :      1: symbol_engine_optimizer
30.03% :     0.006526s :      1: task_emit
22.57% :     0.004905s :      1: type_inference
 0.28% :     0.000060s :      1: validate


TotalTime = 0.0243407, [33]
[bootstrap]: 0.00045201
[type_inference]: 0.00568706
[event_method]: 2.12e-05
[auto_monad]: 7.365e-05
[graph_reusing]: 6.19999e-06
[pre_auto_parallel]: 2.73e-06
[py_interpret_to_execute]: 4.7e-05
[rewriter_before_opt_a]: 7.783e-05
[expand_dump_flag]: 3.60998e-06
[jit_opt_a]: 0.00846491, [2]
    [Cycle 1]: 0.00186888, [27]
        [switch_simplify]: 6.637e-05
        [loop_unroll]: 2.929e-05
        [a_1]: 0.00060152
        [with_stream_mark]: 1.949e-05
        [recompute_prepare]: 8.09997e-06
        [updatestate_depend_eliminate]: 3.32002e-06
        [updatestate_assign_eliminate]: 3.33998e-06
        [updatestate_loads_eliminate]: 3.03998e-06
        [parameter_eliminate]: 2.47001e-06
        [specialize_transform]: 6.44001e-06
        [updatestate_useless_node_eliminater]: 5.36998e-06
        [accelerated_algorithm]: 6.49999e-06
        [meta_shard_fg_expand]: 2.21e-06
        [get_grad_eliminate_]: 5.82999e-06
        [merge_forward]: 4.55999e-06
        [cell_reuse_recompute_pass]: 1.12999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.487e-05
        [j_node_and_user_rematch]: 9.76e-06
        [meta_fg_expand]: 2.24999e-06
        [replace_old_param]: 1.016e-05
        [inline_without_move]: 6.25002e-06
        [renormalize]: 0.0007834
        [add_forward_monad_depend]: 6.44999e-06
        [auto_monad_grad]: 2.79999e-06
        [auto_monad_eliminator]: 1.525e-05
        [cse]: 3.153e-05
        [replace_applicator]: 1.569e-05
    [Cycle 2]: 0.0003404, [27]
        [switch_simplify]: 5.93002e-06
        [loop_unroll]: 5.04e-06
        [a_1]: 9.531e-05
        [with_stream_mark]: 1.113e-05
        [recompute_prepare]: 5.51e-06
        [updatestate_depend_eliminate]: 2.71e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 4.81997e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 5.17e-06
        [meta_shard_fg_expand]: 1.02e-06
        [get_grad_eliminate_]: 4.45e-06
        [merge_forward]: 2.41e-06
        [cell_reuse_recompute_pass]: 2.24001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.475e-05
        [j_node_and_user_rematch]: 7.8e-06
        [meta_fg_expand]: 2.11998e-06
        [replace_old_param]: 1.082e-05
        [inline_without_move]: 5.11002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 6.02001e-06
        [cse]: 1.158e-05
        [replace_applicator]: 4.70001e-06
[py_interpret_to_execute_after_opt_a]: 1.144e-05
[rewriter_after_opt_a]: 3.846e-05
[convert_after_rewriter]: 7.78999e-06
[order_py_execute_after_rewriter]: 5.30001e-06
[mutable_eliminate]: 0.00084448
[jit_opt_b]: 4.96e-05, [1]
    [Cycle 1]: 4.228e-05, [2]
        [frontend_op_eliminate]: 1.633e-05
        [inline_after_opt_a]: 1.467e-05
[cconv]: 2.655e-05
[loop_unroll]: 0.00043176
[jit_opt_after_cconv]: 0.00014448, [1]
    [Cycle 1]: 0.00013729, [11]
        [c_1]: 2.081e-05
        [parameter_eliminate]: 3.04001e-06
        [updatestate_depend_eliminate]: 5.99999e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [cse]: 2.027e-05
        [call_graph_tuple_transform]: 1.87e-05
        [tuple_list_get_item_eliminator]: 5.37999e-06
        [none_parameter_eliminate]: 2.01998e-06
        [renormalize]: 6.80011e-07
        [switch_simplify]: 5.57999e-06
[remove_dup_value]: 1.547e-05
[partial_unused_args_eliminate]: 2.66e-06
[environ_conv]: 5.35001e-06
[add_recomputation]: 5.115e-05
[cse_after_recomputation]: 2.009e-05, [1]
    [Cycle 1]: 1.49e-05, [1]
        [cse]: 8.88002e-06
[auto_monad_reorder]: 1.518e-05
[get_jit_bprop_graph]: 1.42999e-06
[rewriter_after_jit_bprop_graph]: 6.02001e-06
[opt_after_jit_grad]: 0.00046369
[symbol_engine_optimizer]: 7.495e-05, [1]
    [Cycle 1]: 6.757e-05, [6]
        [build]: 5.01997e-06
        [elim_shapecalc]: 7.4e-06
        [elim_not_effective]: 1.212e-05
        [opt_reshape]: 6.02001e-06
        [fold_const_symbol]: 8.67e-06
        [renormalize]: 5.00004e-07
[validate]: 4.177e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00700486
[execute]: 9.04998e-06
Sums
    bootstrap                            :     0.000452s :  2.60%
    type_inference                       :     0.005687s : 32.67%
    event_method                         :     0.000021s :  0.12%
    auto_monad                           :     0.000074s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000047s :  0.27%
    rewriter_before_opt_a                :     0.000078s :  0.45%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000072s :  0.42%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000697s :  4.00%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000021s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000783s :  4.50%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.12%
    jit_opt_a.cse                        :     0.000043s :  0.25%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000844s :  4.85%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000027s :  0.15%
    loop_unroll                          :     0.000432s :  2.48%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000051s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000464s :  2.66%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007005s : 40.24%
    execute                              :     0.000009s :  0.05%


TotalTime = 0.0157318, [30]
[bootstrap]: 0.00036186
[type_inference]: 0.00565766
[event_method]: 1.807e-05
[auto_monad]: 6.59e-05
[graph_reusing]: 5.54e-06
[pre_auto_parallel]: 2.56998e-06
[py_interpret_to_execute]: 3.546e-05
[rewriter_before_opt_a]: 7.051e-05
[expand_dump_flag]: 3.4e-06
[jit_opt_a]: 0.00672165, [2]
    [Cycle 1]: 0.00259043, [27]
        [switch_simplify]: 5.562e-05
        [loop_unroll]: 3.032e-05
        [a_1]: 0.0006003
        [with_stream_mark]: 2.2e-05
        [recompute_prepare]: 8.53001e-06
        [updatestate_depend_eliminate]: 3.78999e-06
        [updatestate_assign_eliminate]: 2.76999e-06
        [updatestate_loads_eliminate]: 2.38998e-06
        [parameter_eliminate]: 1.92001e-06
        [specialize_transform]: 6.41998e-06
        [updatestate_useless_node_eliminater]: 5.59998e-06
        [accelerated_algorithm]: 2.717e-05
        [meta_shard_fg_expand]: 4.18001e-06
        [get_grad_eliminate_]: 7.26001e-06
        [merge_forward]: 8.35001e-06
        [cell_reuse_recompute_pass]: 2.22001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.894e-05
        [j_node_and_user_rematch]: 1.026e-05
        [meta_fg_expand]: 2.73998e-06
        [replace_old_param]: 1.04e-05
        [inline_without_move]: 5.61998e-06
        [renormalize]: 0.00103656
        [add_forward_monad_depend]: 7.63001e-06
        [auto_monad_grad]: 2.79999e-06
        [auto_monad_eliminator]: 1.608e-05
        [cse]: 3.315e-05
        [replace_applicator]: 1.368e-05
    [Cycle 2]: 0.0003518, [27]
        [switch_simplify]: 6.01998e-06
        [loop_unroll]: 5.10001e-06
        [a_1]: 9.938e-05
        [with_stream_mark]: 1.296e-05
        [recompute_prepare]: 5.86998e-06
        [updatestate_depend_eliminate]: 2.53003e-06
        [updatestate_assign_eliminate]: 2.60002e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.22999e-06
        [specialize_transform]: 5.04e-06
        [updatestate_useless_node_eliminater]: 5.19998e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 4.82e-06
        [merge_forward]: 2.92002e-06
        [cell_reuse_recompute_pass]: 2.25002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.376e-05
        [j_node_and_user_rematch]: 7.9e-06
        [meta_fg_expand]: 2.09999e-06
        [replace_old_param]: 8.42e-06
        [inline_without_move]: 4.84e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 5.00999e-06
        [cse]: 1.199e-05
        [replace_applicator]: 4.99003e-06
[py_interpret_to_execute_after_opt_a]: 1.631e-05
[rewriter_after_opt_a]: 3.81e-05
[convert_after_rewriter]: 7.45e-06
[order_py_execute_after_rewriter]: 4.77998e-06
[mutable_eliminate]: 0.00092702
[jit_opt_b]: 6.035e-05, [1]
    [Cycle 1]: 5.109e-05, [2]
        [frontend_op_eliminate]: 2.005e-05
        [inline_after_opt_a]: 1.801e-05
[cconv]: 3.492e-05
[loop_unroll]: 0.00052955
[jit_opt_after_cconv]: 0.00017167, [1]
    [Cycle 1]: 0.00016295, [11]
        [c_1]: 2.602e-05
        [parameter_eliminate]: 3.91999e-06
        [updatestate_depend_eliminate]: 6.61999e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.96999e-06
        [cse]: 2.628e-05
        [call_graph_tuple_transform]: 2.322e-05
        [tuple_list_get_item_eliminator]: 6.14999e-06
        [none_parameter_eliminate]: 1.84998e-06
        [renormalize]: 8.60018e-07
        [switch_simplify]: 6.71999e-06
[remove_dup_value]: 1.676e-05
[partial_unused_args_eliminate]: 2.74999e-06
[environ_conv]: 7.18998e-06
[add_recomputation]: 5.221e-05
[cse_after_recomputation]: 2.416e-05, [1]
    [Cycle 1]: 1.798e-05, [1]
        [cse]: 1.132e-05
[auto_monad_reorder]: 1.643e-05
[get_jit_bprop_graph]: 2.64999e-06
[rewriter_after_jit_bprop_graph]: 7.26001e-06
[opt_after_jit_grad]: 0.00054417
[symbol_engine_optimizer]: 8.391e-05, [1]
    [Cycle 1]: 7.698e-05, [6]
        [build]: 4.86002e-06
        [elim_shapecalc]: 7.56001e-06
        [elim_not_effective]: 1.562e-05
        [opt_reshape]: 6.81001e-06
        [fold_const_symbol]: 1.041e-05
        [renormalize]: 5.50004e-07
[validate]: 4.805e-05
Sums
    bootstrap                            :     0.000362s :  3.33%
    type_inference                       :     0.005658s : 52.14%
    event_method                         :     0.000018s :  0.17%
    auto_monad                           :     0.000066s :  0.61%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.33%
    rewriter_before_opt_a                :     0.000071s :  0.65%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000062s :  0.57%
    jit_opt_a.loop_unroll                :     0.000035s :  0.33%
    jit_opt_a.a_1                        :     0.000700s :  6.45%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000033s :  0.30%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.11%
    jit_opt_a.merge_forward              :     0.000011s :  0.10%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.17%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.001037s :  9.55%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.19%
    jit_opt_a.cse                        :     0.000045s :  0.42%
    jit_opt_a.replace_applicator         :     0.000019s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.15%
    rewriter_after_opt_a                 :     0.000038s :  0.35%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000927s :  8.54%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.17%
    cconv                                :     0.000035s :  0.32%
    loop_unroll                          :     0.000530s :  4.88%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000026s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.06%
    remove_dup_value                     :     0.000017s :  0.15%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000007s :  0.07%
    add_recomputation                    :     0.000052s :  0.48%
    cse_after_recomputation.cse          :     0.000011s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.15%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000544s :  5.01%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000048s :  0.44%


Time group info:
------[substitution.]   0.000235    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.46% :     0.000006s :      3: substitution.graph_param_transform
84.02% :     0.000197s :      4: substitution.inline
 1.34% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.09% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.82% :     0.000007s :      2: substitution.replace_old_param
 5.82% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005594     2
81.03% :     0.004532s :      1: type_inference.infer
18.97% :     0.001061s :      1: type_inference.specialize
------[replace.]   0.000077     6
69.03% :     0.000053s :      4: replace.inline
30.97% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000207     6
93.98% :     0.000195s :      4: match.inline
 6.02% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.49% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 2.54% :     0.000003s :     10: predicate.arithmetic_simplify
 1.29% :     0.000002s :     10: predicate.cast_eliminate
 0.61% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.08% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.24% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000002s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.55% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.61% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.74% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.86% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.57% :     0.000002s :     10: predicate.reshape_eliminate
 1.44% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.79% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.61% :     0.000010s :     53: predicate.switch_simplify
 1.72% :     0.000002s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.49% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.32% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.59% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000002s :     10: predicate.value_based_eliminate
 0.46% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000710    12
49.98% :     0.000355s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.81% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.20% :     0.000321s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026143    76
 0.21% :     0.000054s :      1: add_recomputation
 0.30% :     0.000078s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.82% :     0.000475s :      1: bootstrap
 0.11% :     0.000029s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000026s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
32.40% :     0.008469s :      1: jit_opt_a
 0.56% :     0.000148s :      1: jit_opt_after_cconv
 0.20% :     0.000052s :      1: jit_opt_b
 1.68% :     0.000440s :      1: loop_unroll
 3.27% :     0.000854s :      1: mutable_eliminate
 3.56% :     0.000930s :     26: opt.transform.jit_opt_a
 0.18% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000025s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.80% :     0.000472s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000050s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 1.60% :     0.000419s :      1: renormalize.infer
 1.36% :     0.000355s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000041s :      1: rewriter_after_opt_a
 0.31% :     0.000081s :      1: rewriter_before_opt_a
 0.30% :     0.000077s :      1: symbol_engine_optimizer
26.86% :     0.007021s :      1: task_emit
21.83% :     0.005708s :      1: type_inference
 0.24% :     0.000062s :      1: validate


Time group info:
------[substitution.]   0.000246    23
 1.31% :     0.000003s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.78% :     0.000007s :      3: substitution.graph_param_transform
84.49% :     0.000208s :      4: substitution.inline
 1.27% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.93% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.44% :     0.000006s :      2: substitution.replace_old_param
 5.22% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005587     2
82.34% :     0.004600s :      1: type_inference.infer
17.66% :     0.000987s :      1: type_inference.specialize
------[replace.]   0.000082     6
67.37% :     0.000055s :      4: replace.inline
32.63% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000218     6
94.64% :     0.000206s :      4: match.inline
 5.36% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.46% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 2.06% :     0.000003s :     10: predicate.arithmetic_simplify
 1.06% :     0.000001s :     10: predicate.cast_eliminate
 0.55% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.92% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.06% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.18% :     0.000002s :      6: predicate.get_grad_eliminate
 0.34% :     0.000000s :      3: predicate.graph_param_transform
 5.48% :     0.000007s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 2.66% :     0.000004s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.62% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.68% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.33% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.92% :     0.000001s :     10: predicate.merge_addn
 0.93% :     0.000001s :     10: predicate.minmaximum_grad
 1.89% :     0.000003s :      3: predicate.mutable_eliminate
 0.88% :     0.000001s :      3: predicate.opt_reshape
 1.92% :     0.000003s :     15: predicate.partial_eliminate
 1.60% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 1.82% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.52% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.98% :     0.000003s :     10: predicate.reshape_eliminate
 1.25% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.43% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.43% :     0.000002s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.55% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.62% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.21% :     0.000003s :     16: predicate.switch_defer_inline
 1.88% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.23% :     0.000010s :     53: predicate.switch_simplify
 1.43% :     0.000002s :     10: predicate.tile_eliminate
 1.07% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.29% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000749    12
46.94% :     0.000352s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.41% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.65% :     0.000349s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017773    72
 0.31% :     0.000056s :      1: add_recomputation
 0.39% :     0.000069s :      1: auto_monad
 0.11% :     0.000019s :      1: auto_monad_reorder
 2.14% :     0.000381s :      1: bootstrap
 0.21% :     0.000038s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.12% :     0.000022s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
37.84% :     0.006726s :      1: jit_opt_a
 0.99% :     0.000176s :      1: jit_opt_after_cconv
 0.36% :     0.000063s :      1: jit_opt_b
 3.04% :     0.000540s :      1: loop_unroll
 5.28% :     0.000939s :      1: mutable_eliminate
 5.31% :     0.000944s :     26: opt.transform.jit_opt_a
 0.33% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000030s :      4: opt.transform.jit_opt_b
 0.08% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.12% :     0.000554s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000038s :      1: py_interpret_to_execute
 0.11% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000019s :      1: remove_dup_value
 2.95% :     0.000524s :      1: renormalize.infer
 2.83% :     0.000503s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000041s :      1: rewriter_after_opt_a
 0.41% :     0.000073s :      1: rewriter_before_opt_a
 0.49% :     0.000087s :      1: symbol_engine_optimizer
31.90% :     0.005670s :      1: type_inference


TotalTime = 0.0203584, [33]
[bootstrap]: 0.00028353
[type_inference]: 0.00477232
[event_method]: 1.825e-05
[auto_monad]: 6.491e-05
[graph_reusing]: 5.62001e-06
[pre_auto_parallel]: 2.84999e-06
[py_interpret_to_execute]: 3.441e-05
[rewriter_before_opt_a]: 7.403e-05
[expand_dump_flag]: 2.92002e-06
[jit_opt_a]: 0.00562493, [2]
    [Cycle 1]: 0.00188429, [27]
        [switch_simplify]: 5.734e-05
        [loop_unroll]: 2.925e-05
        [a_1]: 0.00057123
        [with_stream_mark]: 2.248e-05
        [recompute_prepare]: 7.91001e-06
        [updatestate_depend_eliminate]: 4.43999e-06
        [updatestate_assign_eliminate]: 3.25e-06
        [updatestate_loads_eliminate]: 2.53998e-06
        [parameter_eliminate]: 1.97999e-06
        [specialize_transform]: 5.90002e-06
        [updatestate_useless_node_eliminater]: 5.02999e-06
        [accelerated_algorithm]: 6.02999e-06
        [meta_shard_fg_expand]: 2.19001e-06
        [get_grad_eliminate_]: 6.34999e-06
        [merge_forward]: 3.91999e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.498e-05
        [j_node_and_user_rematch]: 9.36002e-06
        [meta_fg_expand]: 2.49999e-06
        [replace_old_param]: 9.46e-06
        [inline_without_move]: 5.62001e-06
        [renormalize]: 0.00085543
        [add_forward_monad_depend]: 5.51e-06
        [auto_monad_grad]: 2.76999e-06
        [auto_monad_eliminator]: 1.497e-05
        [cse]: 3.056e-05
        [replace_applicator]: 1.284e-05
    [Cycle 2]: 0.00032676, [27]
        [switch_simplify]: 6.15002e-06
        [loop_unroll]: 4.85001e-06
        [a_1]: 8.992e-05
        [with_stream_mark]: 1.091e-05
        [recompute_prepare]: 5.22e-06
        [updatestate_depend_eliminate]: 2.60002e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 2.11998e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 4.84998e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 5.00001e-06
        [merge_forward]: 2.30002e-06
        [cell_reuse_recompute_pass]: 2.23998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.278e-05
        [j_node_and_user_rematch]: 8.03001e-06
        [meta_fg_expand]: 1.75001e-06
        [replace_old_param]: 8.48999e-06
        [inline_without_move]: 4.68001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 5.27999e-06
        [cse]: 1.072e-05
        [replace_applicator]: 4.99998e-06
[py_interpret_to_execute_after_opt_a]: 1.122e-05
[rewriter_after_opt_a]: 3.626e-05
[convert_after_rewriter]: 6.93e-06
[order_py_execute_after_rewriter]: 7.93999e-06
[mutable_eliminate]: 0.00057639
[jit_opt_b]: 5.091e-05, [1]
    [Cycle 1]: 4.386e-05, [2]
        [frontend_op_eliminate]: 1.68e-05
        [inline_after_opt_a]: 1.513e-05
[cconv]: 2.868e-05
[loop_unroll]: 0.00043432
[jit_opt_after_cconv]: 0.00015225, [1]
    [Cycle 1]: 0.00014541, [11]
        [c_1]: 2.153e-05
        [parameter_eliminate]: 3.43e-06
        [updatestate_depend_eliminate]: 6.70002e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 2.368e-05
        [call_graph_tuple_transform]: 2.125e-05
        [tuple_list_get_item_eliminator]: 5.51002e-06
        [none_parameter_eliminate]: 1.29e-06
        [renormalize]: 7.39994e-07
        [switch_simplify]: 5.44e-06
[remove_dup_value]: 1.359e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 5.01002e-06
[add_recomputation]: 5.15e-05
[cse_after_recomputation]: 2.214e-05, [1]
    [Cycle 1]: 1.617e-05, [1]
        [cse]: 9.79999e-06
[auto_monad_reorder]: 1.525e-05
[get_jit_bprop_graph]: 2.16998e-06
[rewriter_after_jit_bprop_graph]: 6.21e-06
[opt_after_jit_grad]: 0.00048727
[symbol_engine_optimizer]: 8.005e-05, [1]
    [Cycle 1]: 7.273e-05, [6]
        [build]: 4.95001e-06
        [elim_shapecalc]: 7.73999e-06
        [elim_not_effective]: 1.464e-05
        [opt_reshape]: 5.97001e-06
        [fold_const_symbol]: 8.84e-06
        [renormalize]: 4.19997e-07
[validate]: 4.151e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00720413
[execute]: 1.098e-05
Sums
    bootstrap                            :     0.000284s :  1.74%
    type_inference                       :     0.004772s : 29.31%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000065s :  0.40%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.21%
    rewriter_before_opt_a                :     0.000074s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.21%
    jit_opt_a.a_1                        :     0.000661s :  4.06%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000856s :  5.25%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000041s :  0.25%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000008s :  0.05%
    mutable_eliminate                    :     0.000576s :  3.54%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000029s :  0.18%
    loop_unroll                          :     0.000434s :  2.67%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000051s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000487s :  2.99%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000042s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007204s : 44.24%
    execute                              :     0.000011s :  0.07%


Time group info:
------[substitution.]   0.000235    23
 0.95% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.59% :     0.000006s :      3: substitution.graph_param_transform
84.02% :     0.000197s :      4: substitution.inline
 1.44% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.02% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.31% :     0.000005s :      2: substitution.replace_old_param
 6.07% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004705     2
81.04% :     0.003813s :      1: type_inference.infer
18.96% :     0.000892s :      1: type_inference.specialize
------[replace.]   0.000070     6
67.09% :     0.000047s :      4: replace.inline
32.91% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000208     6
93.81% :     0.000195s :      4: match.inline
 6.19% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.67% :     0.000002s :     10: predicate.addn_zero_filter
 2.18% :     0.000003s :     10: predicate.arithmetic_simplify
 1.11% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.24% :     0.000001s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.08% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.20% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.40% :     0.000002s :      6: predicate.get_grad_eliminate
 0.41% :     0.000000s :      3: predicate.graph_param_transform
 5.60% :     0.000007s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.57% :     0.000002s :      6: predicate.less_batch_normalization
 1.37% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.56% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 2.43% :     0.000003s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.76% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.32% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.64% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.67% :     0.000009s :     53: predicate.switch_simplify
 1.09% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.03% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000671    12
48.24% :     0.000324s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.07% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.69% :     0.000300s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022195    76
 0.25% :     0.000055s :      1: add_recomputation
 0.31% :     0.000069s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.38% :     0.000305s :      1: bootstrap
 0.14% :     0.000032s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.36% :     0.005628s :      1: jit_opt_a
 0.70% :     0.000155s :      1: jit_opt_after_cconv
 0.24% :     0.000054s :      1: jit_opt_b
 1.99% :     0.000443s :      1: loop_unroll
 2.64% :     0.000587s :      1: mutable_eliminate
 3.95% :     0.000877s :     26: opt.transform.jit_opt_a
 0.22% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.24% :     0.000497s :      1: opt_after_jit_grad
 0.05% :     0.000010s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.13% :     0.000472s :      1: renormalize.infer
 1.68% :     0.000373s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000039s :      1: rewriter_after_opt_a
 0.35% :     0.000077s :      1: rewriter_before_opt_a
 0.37% :     0.000083s :      1: symbol_engine_optimizer
32.55% :     0.007225s :      1: task_emit
21.57% :     0.004788s :      1: type_inference
 0.29% :     0.000064s :      1: validate


TotalTime = 0.0340273, [30]
[bootstrap]: 0.00038628
[type_inference]: 0.0143611
[event_method]: 2.263e-05
[auto_monad]: 7.338e-05
[graph_reusing]: 6.63003e-06
[pre_auto_parallel]: 3.03998e-06
[py_interpret_to_execute]: 4.782e-05
[rewriter_before_opt_a]: 7.811e-05
[expand_dump_flag]: 3.41001e-06
[jit_opt_a]: 0.0163563, [2]
    [Cycle 1]: 0.0110063, [27]
        [switch_simplify]: 6.646e-05
        [loop_unroll]: 3.21e-05
        [a_1]: 0.00064284
        [with_stream_mark]: 2.303e-05
        [recompute_prepare]: 8.97999e-06
        [updatestate_depend_eliminate]: 4.28999e-06
        [updatestate_assign_eliminate]: 3.36001e-06
        [updatestate_loads_eliminate]: 2.60002e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 6.24999e-06
        [updatestate_useless_node_eliminater]: 5.58002e-06
        [accelerated_algorithm]: 6.71e-06
        [meta_shard_fg_expand]: 2.12001e-06
        [get_grad_eliminate_]: 5.63002e-06
        [merge_forward]: 3.65e-06
        [cell_reuse_recompute_pass]: 1.55999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.707e-05
        [j_node_and_user_rematch]: 9.77999e-06
        [meta_fg_expand]: 2.44001e-06
        [replace_old_param]: 1.025e-05
        [inline_without_move]: 5.67999e-06
        [renormalize]: 0.00979709
        [add_forward_monad_depend]: 1.555e-05
        [auto_monad_grad]: 2.96999e-06
        [auto_monad_eliminator]: 2.814e-05
        [cse]: 3.6e-05
        [replace_applicator]: 3.233e-05
    [Cycle 2]: 0.0004195, [27]
        [switch_simplify]: 7.08998e-06
        [loop_unroll]: 6.49999e-06
        [a_1]: 0.00012018
        [with_stream_mark]: 2e-05
        [recompute_prepare]: 5.97001e-06
        [updatestate_depend_eliminate]: 4.02e-06
        [updatestate_assign_eliminate]: 2.93998e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 2.22999e-06
        [specialize_transform]: 6.01e-06
        [updatestate_useless_node_eliminater]: 5.25999e-06
        [accelerated_algorithm]: 6.67002e-06
        [meta_shard_fg_expand]: 2.53e-06
        [get_grad_eliminate_]: 5.67999e-06
        [merge_forward]: 4.03999e-06
        [cell_reuse_recompute_pass]: 4.04002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.21e-05
        [j_node_and_user_rematch]: 1.038e-05
        [meta_fg_expand]: 2.81999e-06
        [replace_old_param]: 1.069e-05
        [inline_without_move]: 5.25999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.66e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 6.44001e-06
        [cse]: 1.405e-05
        [replace_applicator]: 5.64e-06
[py_interpret_to_execute_after_opt_a]: 2.383e-05
[rewriter_after_opt_a]: 4.221e-05
[convert_after_rewriter]: 7.06001e-06
[order_py_execute_after_rewriter]: 5.29e-06
[mutable_eliminate]: 0.00093094
[jit_opt_b]: 5.929e-05, [1]
    [Cycle 1]: 5.026e-05, [2]
        [frontend_op_eliminate]: 2.037e-05
        [inline_after_opt_a]: 1.695e-05
[cconv]: 3.486e-05
[loop_unroll]: 0.00044535
[jit_opt_after_cconv]: 0.00015667, [1]
    [Cycle 1]: 0.00014864, [11]
        [c_1]: 2.308e-05
        [parameter_eliminate]: 3.37002e-06
        [updatestate_depend_eliminate]: 7.24001e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 2.247e-05
        [call_graph_tuple_transform]: 2.236e-05
        [tuple_list_get_item_eliminator]: 5.72001e-06
        [none_parameter_eliminate]: 1.80001e-06
        [renormalize]: 6.29982e-07
        [switch_simplify]: 6.12999e-06
[remove_dup_value]: 1.717e-05
[partial_unused_args_eliminate]: 2.43002e-06
[environ_conv]: 6.18002e-06
[add_recomputation]: 5.683e-05
[cse_after_recomputation]: 2.282e-05, [1]
    [Cycle 1]: 1.697e-05, [1]
        [cse]: 1.008e-05
[auto_monad_reorder]: 1.559e-05
[get_jit_bprop_graph]: 2.55002e-06
[rewriter_after_jit_bprop_graph]: 7.68001e-06
[opt_after_jit_grad]: 0.00046932
[symbol_engine_optimizer]: 7.987e-05, [1]
    [Cycle 1]: 7.192e-05, [6]
        [build]: 4.79998e-06
        [elim_shapecalc]: 7.96001e-06
        [elim_not_effective]: 1.347e-05
        [opt_reshape]: 5.91998e-06
        [fold_const_symbol]: 9.41e-06
        [renormalize]: 6.39993e-07
[validate]: 9.032e-05
Sums
    bootstrap                            :     0.000386s :  1.36%
    type_inference                       :     0.014361s : 50.59%
    event_method                         :     0.000023s :  0.08%
    auto_monad                           :     0.000073s :  0.26%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000048s :  0.17%
    rewriter_before_opt_a                :     0.000078s :  0.28%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000074s :  0.26%
    jit_opt_a.loop_unroll                :     0.000039s :  0.14%
    jit_opt_a.a_1                        :     0.000763s :  2.69%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.04%
    jit_opt_a.merge_forward              :     0.000008s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.07%
    jit_opt_a.inline_without_move        :     0.000011s :  0.04%
    jit_opt_a.renormalize                :     0.009797s : 34.51%
    jit_opt_a.add_forward_monad_depend   :     0.000017s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000035s :  0.12%
    jit_opt_a.cse                        :     0.000050s :  0.18%
    jit_opt_a.replace_applicator         :     0.000038s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000024s :  0.08%
    rewriter_after_opt_a                 :     0.000042s :  0.15%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000931s :  3.28%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.06%
    cconv                                :     0.000035s :  0.12%
    loop_unroll                          :     0.000445s :  1.57%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000017s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000057s :  0.20%
    cse_after_recomputation.cse          :     0.000010s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.03%
    opt_after_jit_grad                   :     0.000469s :  1.65%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000090s :  0.32%


Time group info:
------[substitution.]   0.000276    23
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.04% :     0.000006s :      3: substitution.graph_param_transform
84.85% :     0.000234s :      4: substitution.inline
 1.75% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.41% :     0.000007s :      4: substitution.remove_not_recompute_node
 2.77% :     0.000008s :      2: substitution.replace_old_param
 4.94% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.014273     2
91.63% :     0.013078s :      1: type_inference.infer
 8.37% :     0.001195s :      1: type_inference.specialize
------[replace.]   0.000089     6
66.28% :     0.000059s :      4: replace.inline
33.72% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000244     6
94.90% :     0.000232s :      4: match.inline
 5.10% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000146   691
 1.49% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.86% :     0.000001s :     10: predicate.addn_check_dump
 1.44% :     0.000002s :     10: predicate.addn_zero_filter
 2.68% :     0.000004s :     10: predicate.arithmetic_simplify
 1.16% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.86% :     0.000001s :     10: predicate.compare_switch_simplify
 1.92% :     0.000003s :     10: predicate.depend_value_elim
 1.10% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.25% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.19% :     0.000002s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.46% :     0.000008s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.94% :     0.000003s :      6: predicate.less_batch_normalization
 1.83% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 1.49% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.85% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.42% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 0.86% :     0.000001s :     10: predicate.merge_addn
 0.88% :     0.000001s :     10: predicate.minmaximum_grad
 1.98% :     0.000003s :      3: predicate.mutable_eliminate
 0.39% :     0.000001s :      3: predicate.opt_reshape
 1.99% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.67% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.38% :     0.000001s :      3: predicate.reset_defer_inline
 1.75% :     0.000003s :     10: predicate.reshape_eliminate
 1.16% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.24% :     0.000002s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 1.79% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.14% :     0.000010s :     53: predicate.switch_simplify
 1.28% :     0.000002s :     10: predicate.tile_eliminate
 1.31% :     0.000002s :     10: predicate.transpose_eliminate
 1.51% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.40% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.36% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.51% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.94% :     0.000003s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000978    12
41.25% :     0.000404s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.32% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
53.43% :     0.000523s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.044844    72
 0.13% :     0.000060s :      1: add_recomputation
 0.17% :     0.000077s :      1: auto_monad
 0.04% :     0.000019s :      1: auto_monad_reorder
 0.91% :     0.000410s :      1: bootstrap
 0.09% :     0.000038s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.06% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.06% :     0.000027s :      1: event_method
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
36.49% :     0.016362s :      1: jit_opt_a
 0.36% :     0.000159s :      1: jit_opt_after_cconv
 0.14% :     0.000062s :      1: jit_opt_b
 1.01% :     0.000454s :      1: loop_unroll
 2.10% :     0.000942s :      1: mutable_eliminate
 2.29% :     0.001028s :     26: opt.transform.jit_opt_a
 0.12% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000030s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.07% :     0.000478s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000051s :      1: py_interpret_to_execute
 0.06% :     0.000028s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000020s :      1: remove_dup_value
20.19% :     0.009053s :      1: renormalize.infer
 1.61% :     0.000724s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000046s :      1: rewriter_after_opt_a
 0.18% :     0.000082s :      1: rewriter_before_opt_a
 0.18% :     0.000082s :      1: symbol_engine_optimizer
32.07% :     0.014382s :      1: type_inference


TotalTime = 0.0208844, [33]
[bootstrap]: 0.00034444
[type_inference]: 0.00520929
[event_method]: 1.837e-05
[auto_monad]: 6.621e-05
[graph_reusing]: 5.75001e-06
[pre_auto_parallel]: 2.79999e-06
[py_interpret_to_execute]: 3.509e-05
[rewriter_before_opt_a]: 7.459e-05
[expand_dump_flag]: 3.11999e-06
[jit_opt_a]: 0.00609459, [2]
    [Cycle 1]: 0.00199806, [27]
        [switch_simplify]: 6.248e-05
        [loop_unroll]: 2.966e-05
        [a_1]: 0.00057297
        [with_stream_mark]: 2.049e-05
        [recompute_prepare]: 7.70998e-06
        [updatestate_depend_eliminate]: 3.64002e-06
        [updatestate_assign_eliminate]: 3.48999e-06
        [updatestate_loads_eliminate]: 4.369e-05
        [parameter_eliminate]: 2.67001e-06
        [specialize_transform]: 8.84e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.14e-06
        [meta_shard_fg_expand]: 2.29001e-06
        [get_grad_eliminate_]: 5.52001e-06
        [merge_forward]: 4.62998e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.66e-05
        [j_node_and_user_rematch]: 9.33997e-06
        [meta_fg_expand]: 2.14999e-06
        [replace_old_param]: 9.22001e-06
        [inline_without_move]: 5.17999e-06
        [renormalize]: 0.00092733
        [add_forward_monad_depend]: 7.02002e-06
        [auto_monad_grad]: 2.93e-06
        [auto_monad_eliminator]: 1.547e-05
        [cse]: 2.72e-05
        [replace_applicator]: 1.209e-05
    [Cycle 2]: 0.00032825, [27]
        [switch_simplify]: 5.89999e-06
        [loop_unroll]: 5.07e-06
        [a_1]: 8.983e-05
        [with_stream_mark]: 1.213e-05
        [recompute_prepare]: 5.27999e-06
        [updatestate_depend_eliminate]: 3.21001e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.96998e-06
        [parameter_eliminate]: 1.35999e-06
        [specialize_transform]: 4.85999e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.42999e-06
        [meta_shard_fg_expand]: 1.24e-06
        [get_grad_eliminate_]: 4.38001e-06
        [merge_forward]: 3.44001e-06
        [cell_reuse_recompute_pass]: 1.74e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.365e-05
        [j_node_and_user_rematch]: 7.8e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 1.069e-05
        [inline_without_move]: 4.60001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.48002e-06
        [auto_monad_grad]: 9.89996e-07
        [auto_monad_eliminator]: 5.51e-06
        [cse]: 1.028e-05
        [replace_applicator]: 4.82e-06
[py_interpret_to_execute_after_opt_a]: 1.296e-05
[rewriter_after_opt_a]: 3.961e-05
[convert_after_rewriter]: 7.1e-06
[order_py_execute_after_rewriter]: 5.00001e-06
[mutable_eliminate]: 0.00063636
[jit_opt_b]: 5.163e-05, [1]
    [Cycle 1]: 4.382e-05, [2]
        [frontend_op_eliminate]: 1.703e-05
        [inline_after_opt_a]: 1.535e-05
[cconv]: 3.201e-05
[loop_unroll]: 0.00042539
[jit_opt_after_cconv]: 0.00015222, [1]
    [Cycle 1]: 0.00014402, [11]
        [c_1]: 2.205e-05
        [parameter_eliminate]: 3.03998e-06
        [updatestate_depend_eliminate]: 7.66001e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 2.354e-05
        [call_graph_tuple_transform]: 2.006e-05
        [tuple_list_get_item_eliminator]: 5.20999e-06
        [none_parameter_eliminate]: 1.81e-06
        [renormalize]: 6.10016e-07
        [switch_simplify]: 5.27999e-06
[remove_dup_value]: 1.413e-05
[partial_unused_args_eliminate]: 2.78998e-06
[environ_conv]: 5.57999e-06
[add_recomputation]: 5.35e-05
[cse_after_recomputation]: 2.277e-05, [1]
    [Cycle 1]: 1.654e-05, [1]
        [cse]: 9.18002e-06
[auto_monad_reorder]: 1.66e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 5.87999e-06
[opt_after_jit_grad]: 0.00048611
[symbol_engine_optimizer]: 7.744e-05, [1]
    [Cycle 1]: 7.01e-05, [6]
        [build]: 5.10999e-06
        [elim_shapecalc]: 7.65e-06
        [elim_not_effective]: 1.248e-05
        [opt_reshape]: 5.76998e-06
        [fold_const_symbol]: 9.89999e-06
        [renormalize]: 8.70001e-07
[validate]: 4.328e-05
[backend_pass]: 1.07998e-06
[task_emit]: 0.00670682
[execute]: 1.151e-05
Sums
    bootstrap                            :     0.000344s :  2.09%
    type_inference                       :     0.005209s : 31.62%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000066s :  0.40%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.21%
    rewriter_before_opt_a                :     0.000075s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.42%
    jit_opt_a.loop_unroll                :     0.000035s :  0.21%
    jit_opt_a.a_1                        :     0.000663s :  4.02%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000046s :  0.28%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000014s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000927s :  5.63%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.13%
    jit_opt_a.cse                        :     0.000037s :  0.23%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000040s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000636s :  3.86%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000032s :  0.19%
    loop_unroll                          :     0.000425s :  2.58%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000053s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000486s :  2.95%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006707s : 40.71%
    execute                              :     0.000012s :  0.07%


Time group info:
------[substitution.]   0.000243    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000002s :      2: substitution.fold_const_symbol
 2.61% :     0.000006s :      3: substitution.graph_param_transform
84.38% :     0.000205s :      4: substitution.inline
 1.39% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.94% :     0.000005s :      4: substitution.remove_not_recompute_node
 3.20% :     0.000008s :      2: substitution.replace_old_param
 4.99% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005145     2
82.02% :     0.004220s :      1: type_inference.infer
17.98% :     0.000925s :      1: type_inference.specialize
------[replace.]   0.000071     6
66.83% :     0.000047s :      4: replace.inline
33.17% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000214     6
94.84% :     0.000203s :      4: match.inline
 5.16% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.21% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.08% :     0.000002s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.61% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.63% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.39% :     0.000002s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 6.31% :     0.000007s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.96% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.78% :     0.000002s :      3: predicate.mutable_eliminate
 0.58% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.39% :     0.000002s :     10: predicate.reduce_eliminate
 1.68% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.07% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 1.18% :     0.000001s :      6: predicate.specialize_transform
 1.23% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.00% :     0.000009s :     53: predicate.switch_simplify
 1.40% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.98% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.54% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.45% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000684    12
47.30% :     0.000323s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.19% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.51% :     0.000311s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022799    76
 0.25% :     0.000056s :      1: add_recomputation
 0.31% :     0.000070s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.57% :     0.000357s :      1: bootstrap
 0.15% :     0.000035s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.07% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.75% :     0.006099s :      1: jit_opt_a
 0.68% :     0.000155s :      1: jit_opt_after_cconv
 0.24% :     0.000054s :      1: jit_opt_b
 1.90% :     0.000433s :      1: loop_unroll
 2.84% :     0.000648s :      1: mutable_eliminate
 3.89% :     0.000887s :     26: opt.transform.jit_opt_a
 0.22% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.17% :     0.000495s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000038s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.33% :     0.000532s :      1: renormalize.infer
 1.69% :     0.000386s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000042s :      1: rewriter_after_opt_a
 0.34% :     0.000077s :      1: rewriter_before_opt_a
 0.35% :     0.000080s :      1: symbol_engine_optimizer
29.51% :     0.006727s :      1: task_emit
22.91% :     0.005223s :      1: type_inference
 0.29% :     0.000067s :      1: validate


TotalTime = 0.0498111, [33]
[bootstrap]: 0.00041208
[type_inference]: 0.00583545
[event_method]: 2.096e-05
[auto_monad]: 6.835e-05
[graph_reusing]: 5.66e-06
[pre_auto_parallel]: 3.12002e-06
[py_interpret_to_execute]: 4.534e-05
[rewriter_before_opt_a]: 7.836e-05
[expand_dump_flag]: 3.09999e-06
[jit_opt_a]: 0.031758, [2]
    [Cycle 1]: 0.00191519, [27]
        [switch_simplify]: 6.099e-05
        [loop_unroll]: 2.9e-05
        [a_1]: 0.00061698
        [with_stream_mark]: 1.9e-05
        [recompute_prepare]: 8.43999e-06
        [updatestate_depend_eliminate]: 3.93999e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 7.71999e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 6.04999e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 5.74999e-06
        [merge_forward]: 3.04999e-06
        [cell_reuse_recompute_pass]: 1.18001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.314e-05
        [j_node_and_user_rematch]: 9.84999e-06
        [meta_fg_expand]: 2.20002e-06
        [replace_old_param]: 8.95999e-06
        [inline_without_move]: 5.91e-06
        [renormalize]: 0.00087942
        [add_forward_monad_depend]: 5.44e-06
        [auto_monad_grad]: 1.99999e-06
        [auto_monad_eliminator]: 1.112e-05
        [cse]: 1.685e-05
        [replace_applicator]: 1.284e-05
    [Cycle 2]: 0.00036128, [27]
        [switch_simplify]: 5.81e-06
        [loop_unroll]: 5.02999e-06
        [a_1]: 9.252e-05
        [with_stream_mark]: 1.017e-05
        [recompute_prepare]: 5.00001e-06
        [updatestate_depend_eliminate]: 2.67001e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.90001e-06
        [updatestate_useless_node_eliminater]: 5.20001e-06
        [accelerated_algorithm]: 5.22999e-06
        [meta_shard_fg_expand]: 1.25001e-06
        [get_grad_eliminate_]: 4.80001e-06
        [merge_forward]: 2.64001e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.218e-05
        [j_node_and_user_rematch]: 4.886e-05
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 7e-06
        [inline_without_move]: 4.99e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.35999e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 5.97001e-06
        [cse]: 1.108e-05
        [replace_applicator]: 4.97e-06
[py_interpret_to_execute_after_opt_a]: 1.363e-05
[rewriter_after_opt_a]: 3.369e-05
[convert_after_rewriter]: 6.29999e-06
[order_py_execute_after_rewriter]: 4.95999e-06
[mutable_eliminate]: 0.00070508
[jit_opt_b]: 5.246e-05, [1]
    [Cycle 1]: 4.558e-05, [2]
        [frontend_op_eliminate]: 1.829e-05
        [inline_after_opt_a]: 1.567e-05
[cconv]: 2.322e-05
[loop_unroll]: 0.00052415
[jit_opt_after_cconv]: 0.00014879, [1]
    [Cycle 1]: 0.00014175, [11]
        [c_1]: 2.32e-05
        [parameter_eliminate]: 3.16999e-06
        [updatestate_depend_eliminate]: 6.41e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [cse]: 2.145e-05
        [call_graph_tuple_transform]: 1.973e-05
        [tuple_list_get_item_eliminator]: 4.92e-06
        [none_parameter_eliminate]: 1.27e-06
        [renormalize]: 8.00006e-07
        [switch_simplify]: 5.42999e-06
[remove_dup_value]: 1.359e-05
[partial_unused_args_eliminate]: 2.81e-06
[environ_conv]: 5.30999e-06
[add_recomputation]: 4.569e-05
[cse_after_recomputation]: 2.279e-05, [1]
    [Cycle 1]: 1.652e-05, [1]
        [cse]: 1.005e-05
[auto_monad_reorder]: 1.406e-05
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 7.66999e-06
[opt_after_jit_grad]: 0.0004779
[symbol_engine_optimizer]: 7.524e-05, [1]
    [Cycle 1]: 6.91e-05, [6]
        [build]: 4.50999e-06
        [elim_shapecalc]: 8.43001e-06
        [elim_not_effective]: 1.305e-05
        [opt_reshape]: 5.75001e-06
        [fold_const_symbol]: 9.05001e-06
        [renormalize]: 7.49977e-07
[validate]: 4.136e-05
[backend_pass]: 6.59988e-07
[task_emit]: 0.00909754
[execute]: 1.205e-05
Sums
    bootstrap                            :     0.000412s :  2.09%
    type_inference                       :     0.005835s : 29.66%
    event_method                         :     0.000021s :  0.11%
    auto_monad                           :     0.000068s :  0.35%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000045s :  0.23%
    rewriter_before_opt_a                :     0.000078s :  0.40%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.34%
    jit_opt_a.loop_unroll                :     0.000034s :  0.17%
    jit_opt_a.a_1                        :     0.000710s :  3.61%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000013s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000059s :  0.30%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.08%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000879s :  4.47%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.09%
    jit_opt_a.cse                        :     0.000028s :  0.14%
    jit_opt_a.replace_applicator         :     0.000018s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.17%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000705s :  3.58%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.000023s :  0.12%
    loop_unroll                          :     0.000524s :  2.66%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.07%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000046s :  0.23%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000478s :  2.43%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.21%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009098s : 46.24%
    execute                              :     0.000012s :  0.06%


Time group info:
------[substitution.]   0.000249    23
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000001s :      2: substitution.fold_const_symbol
 2.26% :     0.000006s :      3: substitution.graph_param_transform
86.66% :     0.000216s :      4: substitution.inline
 1.30% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.18% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.34% :     0.000003s :      2: substitution.replace_old_param
 4.99% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005710     2
81.65% :     0.004662s :      1: type_inference.infer
18.35% :     0.001048s :      1: type_inference.specialize
------[replace.]   0.000078     6
66.59% :     0.000052s :      4: replace.inline
33.41% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000224     6
94.97% :     0.000213s :      4: match.inline
 5.03% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 2.65% :     0.000003s :     10: predicate.arithmetic_simplify
 1.63% :     0.000002s :     10: predicate.cast_eliminate
 0.33% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.66% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     10: predicate.environ_get_eliminate
 1.23% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.21% :     0.000002s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 4.72% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000002s :      6: predicate.less_batch_normalization
 1.67% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.71% :     0.000005s :     34: predicate.loop_unroll_before_grad
 3.10% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.29% :     0.000003s :     10: predicate.reduce_eliminate
 1.35% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.68% :     0.000002s :     10: predicate.reshape_eliminate
 1.36% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.15% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 1.12% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.19% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.12% :     0.000009s :     53: predicate.switch_simplify
 1.39% :     0.000002s :     10: predicate.tile_eliminate
 1.31% :     0.000002s :     10: predicate.transpose_eliminate
 2.01% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.60% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000726    12
46.10% :     0.000335s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.40% :     0.000046s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.50% :     0.000345s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.051752    76
 0.09% :     0.000048s :      1: add_recomputation
 0.14% :     0.000072s :      1: auto_monad
 0.03% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.84% :     0.000432s :      1: bootstrap
 0.05% :     0.000026s :      1: cconv
 0.02% :     0.000008s :      1: convert_after_rewriter
 0.05% :     0.000025s :      1: cse_after_recomputation
 0.01% :     0.000007s :      1: environ_conv
 0.05% :     0.000027s :      1: event_method
 0.04% :     0.000019s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
61.37% :     0.031762s :      1: jit_opt_a
 0.29% :     0.000151s :      1: jit_opt_after_cconv
 0.11% :     0.000055s :      1: jit_opt_b
 1.03% :     0.000534s :      1: loop_unroll
 1.38% :     0.000715s :      1: mutable_eliminate
 1.88% :     0.000972s :     26: opt.transform.jit_opt_a
 0.10% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000027s :      4: opt.transform.jit_opt_b
 0.03% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000033s :      4: opt.transform.symbol_engine_opt
 0.94% :     0.000487s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000048s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000016s :      1: remove_dup_value
 0.96% :     0.000495s :      1: renormalize.infer
 0.72% :     0.000375s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000037s :      1: rewriter_after_opt_a
 0.16% :     0.000081s :      1: rewriter_before_opt_a
 0.15% :     0.000078s :      1: symbol_engine_optimizer
17.63% :     0.009122s :      1: task_emit
11.32% :     0.005857s :      1: type_inference
 0.12% :     0.000061s :      1: validate


TotalTime = 0.0202321, [33]
[bootstrap]: 0.00028049
[type_inference]: 0.00481769
[event_method]: 1.829e-05
[auto_monad]: 6.652e-05
[graph_reusing]: 5.96e-06
[pre_auto_parallel]: 2.93e-06
[py_interpret_to_execute]: 3.596e-05
[rewriter_before_opt_a]: 7.582e-05
[expand_dump_flag]: 3.37002e-06
[jit_opt_a]: 0.00601186, [2]
    [Cycle 1]: 0.00204709, [27]
        [switch_simplify]: 6.403e-05
        [loop_unroll]: 3.213e-05
        [a_1]: 0.00064972
        [with_stream_mark]: 2.494e-05
        [recompute_prepare]: 7.38e-06
        [updatestate_depend_eliminate]: 4.24997e-06
        [updatestate_assign_eliminate]: 3.29001e-06
        [updatestate_loads_eliminate]: 3.08e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 6.07999e-06
        [updatestate_useless_node_eliminater]: 5.30999e-06
        [accelerated_algorithm]: 6.31e-06
        [meta_shard_fg_expand]: 2.04e-06
        [get_grad_eliminate_]: 5.81e-06
        [merge_forward]: 3.86001e-06
        [cell_reuse_recompute_pass]: 9.99979e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.493e-05
        [j_node_and_user_rematch]: 9.39e-06
        [meta_fg_expand]: 2.46e-06
        [replace_old_param]: 1.023e-05
        [inline_without_move]: 5.57001e-06
        [renormalize]: 0.00092375
        [add_forward_monad_depend]: 6.91001e-06
        [auto_monad_grad]: 2.37999e-06
        [auto_monad_eliminator]: 1.491e-05
        [cse]: 3.215e-05
        [replace_applicator]: 1.309e-05
    [Cycle 2]: 0.00033031, [27]
        [switch_simplify]: 6.06e-06
        [loop_unroll]: 5.27001e-06
        [a_1]: 9.402e-05
        [with_stream_mark]: 1.06e-05
        [recompute_prepare]: 5.52999e-06
        [updatestate_depend_eliminate]: 2.66e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 1.22999e-06
        [specialize_transform]: 5.49e-06
        [updatestate_useless_node_eliminater]: 5.54e-06
        [accelerated_algorithm]: 5.99999e-06
        [meta_shard_fg_expand]: 1.39998e-06
        [get_grad_eliminate_]: 4.37e-06
        [merge_forward]: 2.22999e-06
        [cell_reuse_recompute_pass]: 1.78002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.37e-05
        [j_node_and_user_rematch]: 7.92e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 8.31002e-06
        [inline_without_move]: 5.07e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 5.24e-06
        [cse]: 1.104e-05
        [replace_applicator]: 5.25001e-06
[py_interpret_to_execute_after_opt_a]: 1.184e-05
[rewriter_after_opt_a]: 3.803e-05
[convert_after_rewriter]: 7.80998e-06
[order_py_execute_after_rewriter]: 5.00999e-06
[mutable_eliminate]: 0.0005981
[jit_opt_b]: 5.538e-05, [1]
    [Cycle 1]: 4.792e-05, [2]
        [frontend_op_eliminate]: 2.094e-05
        [inline_after_opt_a]: 1.517e-05
[cconv]: 2.698e-05
[loop_unroll]: 0.00042032
[jit_opt_after_cconv]: 0.00014258, [1]
    [Cycle 1]: 0.00013592, [11]
        [c_1]: 2.126e-05
        [parameter_eliminate]: 2.63e-06
        [updatestate_depend_eliminate]: 6.04001e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 1.992e-05
        [call_graph_tuple_transform]: 1.924e-05
        [tuple_list_get_item_eliminator]: 5.47999e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 7.00005e-07
        [switch_simplify]: 5.74e-06
[remove_dup_value]: 1.413e-05
[partial_unused_args_eliminate]: 2.11003e-06
[environ_conv]: 5.17e-06
[add_recomputation]: 5.113e-05
[cse_after_recomputation]: 2.091e-05, [1]
    [Cycle 1]: 1.534e-05, [1]
        [cse]: 9.32999e-06
[auto_monad_reorder]: 1.559e-05
[get_jit_bprop_graph]: 1.97999e-06
[rewriter_after_jit_bprop_graph]: 5.66e-06
[opt_after_jit_grad]: 0.00046626
[symbol_engine_optimizer]: 7.206e-05, [1]
    [Cycle 1]: 6.541e-05, [6]
        [build]: 3.45e-06
        [elim_shapecalc]: 7.88001e-06
        [elim_not_effective]: 1.269e-05
        [opt_reshape]: 5.49e-06
        [fold_const_symbol]: 8.69e-06
        [renormalize]: 4.80009e-07
[validate]: 3.699e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.00668823
[execute]: 9.02e-06
Sums
    bootstrap                            :     0.000280s :  1.76%
    type_inference                       :     0.004818s : 30.19%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000067s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.23%
    rewriter_before_opt_a                :     0.000076s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000070s :  0.44%
    jit_opt_a.loop_unroll                :     0.000037s :  0.23%
    jit_opt_a.a_1                        :     0.000744s :  4.66%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000924s :  5.79%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000043s :  0.27%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.24%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000598s :  3.75%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000027s :  0.17%
    loop_unroll                          :     0.000420s :  2.63%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000051s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000466s :  2.92%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006688s : 41.91%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000257    23
 0.73% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000002s :      2: substitution.fold_const_symbol
 2.11% :     0.000005s :      3: substitution.graph_param_transform
86.07% :     0.000221s :      4: substitution.inline
 1.28% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.96% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.13% :     0.000005s :      2: substitution.replace_old_param
 5.10% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004743     2
80.85% :     0.003835s :      1: type_inference.infer
19.15% :     0.000908s :      1: type_inference.specialize
------[replace.]   0.000113     6
77.31% :     0.000087s :      4: replace.inline
22.69% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000230     6
94.86% :     0.000218s :      4: match.inline
 5.14% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.22% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.20% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 2.26% :     0.000003s :     10: predicate.arithmetic_simplify
 1.26% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.15% :     0.000001s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.73% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.30% :     0.000002s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.81% :     0.000007s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.45% :     0.000002s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 1.59% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 2.52% :     0.000003s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     15: predicate.partial_eliminate
 1.06% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     10: predicate.reduce_eliminate
 1.73% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.26% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.08% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.56% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.33% :     0.000010s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.90% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000684    12
47.45% :     0.000325s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.66% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.89% :     0.000307s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022230    76
 0.24% :     0.000054s :      1: add_recomputation
 0.32% :     0.000070s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.35% :     0.000300s :      1: bootstrap
 0.13% :     0.000030s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.06% :     0.006015s :      1: jit_opt_a
 0.65% :     0.000145s :      1: jit_opt_after_cconv
 0.26% :     0.000059s :      1: jit_opt_b
 1.93% :     0.000429s :      1: loop_unroll
 2.73% :     0.000607s :      1: mutable_eliminate
 4.37% :     0.000972s :     26: opt.transform.jit_opt_a
 0.22% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000030s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.14% :     0.000475s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000039s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.34% :     0.000521s :      1: renormalize.infer
 1.77% :     0.000393s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000041s :      1: rewriter_after_opt_a
 0.35% :     0.000079s :      1: rewriter_before_opt_a
 0.34% :     0.000075s :      1: symbol_engine_optimizer
30.17% :     0.006707s :      1: task_emit
21.74% :     0.004833s :      1: type_inference
 0.26% :     0.000058s :      1: validate


TotalTime = 0.0293396, [30]
[bootstrap]: 0.00045241
[type_inference]: 0.0129673
[event_method]: 2.233e-05
[auto_monad]: 7.918e-05
[graph_reusing]: 6.56999e-06
[pre_auto_parallel]: 3.08e-06
[py_interpret_to_execute]: 5.588e-05
[rewriter_before_opt_a]: 8.644e-05
[expand_dump_flag]: 3.51001e-06
[jit_opt_a]: 0.0132684, [2]
    [Cycle 1]: 0.00210677, [27]
        [switch_simplify]: 6.815e-05
        [loop_unroll]: 3.229e-05
        [a_1]: 0.00059267
        [with_stream_mark]: 1.964e-05
        [recompute_prepare]: 7.64002e-06
        [updatestate_depend_eliminate]: 3.76999e-06
        [updatestate_assign_eliminate]: 3.37997e-06
        [updatestate_loads_eliminate]: 3.3e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 6.59999e-06
        [updatestate_useless_node_eliminater]: 5.63997e-06
        [accelerated_algorithm]: 6.01998e-06
        [meta_shard_fg_expand]: 2.31998e-06
        [get_grad_eliminate_]: 5.76e-06
        [merge_forward]: 3.73001e-06
        [cell_reuse_recompute_pass]: 1.44998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.544e-05
        [j_node_and_user_rematch]: 1.006e-05
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 9.97001e-06
        [inline_without_move]: 5.18002e-06
        [renormalize]: 0.00104412
        [add_forward_monad_depend]: 6.35002e-06
        [auto_monad_grad]: 2.81e-06
        [auto_monad_eliminator]: 1.491e-05
        [cse]: 3.134e-05
        [replace_applicator]: 1.387e-05
    [Cycle 2]: 0.0003341, [27]
        [switch_simplify]: 6.22001e-06
        [loop_unroll]: 5.14e-06
        [a_1]: 9.363e-05
        [with_stream_mark]: 9.89999e-06
        [recompute_prepare]: 5.50001e-06
        [updatestate_depend_eliminate]: 2.84001e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.26002e-06
        [specialize_transform]: 5.17e-06
        [updatestate_useless_node_eliminater]: 5.39998e-06
        [accelerated_algorithm]: 5.56e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.72e-06
        [merge_forward]: 3.14001e-06
        [cell_reuse_recompute_pass]: 1.42e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.244e-05
        [j_node_and_user_rematch]: 7.95e-06
        [meta_fg_expand]: 1.84998e-06
        [replace_old_param]: 7.89002e-06
        [inline_without_move]: 4.83001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 4.95001e-06
        [cse]: 1.095e-05
        [replace_applicator]: 5.56e-06
[py_interpret_to_execute_after_opt_a]: 1.273e-05
[rewriter_after_opt_a]: 3.808e-05
[convert_after_rewriter]: 7.36001e-06
[order_py_execute_after_rewriter]: 4.96997e-06
[mutable_eliminate]: 0.0006506
[jit_opt_b]: 5.553e-05, [1]
    [Cycle 1]: 4.777e-05, [2]
        [frontend_op_eliminate]: 1.975e-05
        [inline_after_opt_a]: 1.56e-05
[cconv]: 2.642e-05
[loop_unroll]: 0.00046344
[jit_opt_after_cconv]: 0.00017906, [1]
    [Cycle 1]: 0.00017204, [11]
        [c_1]: 2.343e-05
        [parameter_eliminate]: 3.06001e-06
        [updatestate_depend_eliminate]: 6.37001e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [cse]: 2.601e-05
        [call_graph_tuple_transform]: 2.196e-05
        [tuple_list_get_item_eliminator]: 6.72002e-06
        [none_parameter_eliminate]: 1.59998e-06
        [renormalize]: 6.70028e-07
        [switch_simplify]: 6.56e-06
[remove_dup_value]: 1.653e-05
[partial_unused_args_eliminate]: 2.55002e-06
[environ_conv]: 5.79e-06
[add_recomputation]: 5.142e-05
[cse_after_recomputation]: 2.279e-05, [1]
    [Cycle 1]: 1.679e-05, [1]
        [cse]: 1.055e-05
[auto_monad_reorder]: 1.666e-05
[get_jit_bprop_graph]: 1.52999e-06
[rewriter_after_jit_bprop_graph]: 6.20002e-06
[opt_after_jit_grad]: 0.00049279
[symbol_engine_optimizer]: 8.039e-05, [1]
    [Cycle 1]: 7.379e-05, [6]
        [build]: 4.09002e-06
        [elim_shapecalc]: 8.79e-06
        [elim_not_effective]: 1.38e-05
        [opt_reshape]: 6.58e-06
        [fold_const_symbol]: 9.59e-06
        [renormalize]: 5.40022e-07
[validate]: 3.909e-05
Sums
    bootstrap                            :     0.000452s :  2.54%
    type_inference                       :     0.012967s : 72.70%
    event_method                         :     0.000022s :  0.13%
    auto_monad                           :     0.000079s :  0.44%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000056s :  0.31%
    rewriter_before_opt_a                :     0.000086s :  0.48%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.42%
    jit_opt_a.loop_unroll                :     0.000037s :  0.21%
    jit_opt_a.a_1                        :     0.000686s :  3.85%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001044s :  5.85%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.11%
    jit_opt_a.cse                        :     0.000042s :  0.24%
    jit_opt_a.replace_applicator         :     0.000019s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000651s :  3.65%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000026s :  0.15%
    loop_unroll                          :     0.000463s :  2.60%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.04%
    remove_dup_value                     :     0.000017s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000051s :  0.29%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000493s :  2.76%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.22%


Time group info:
------[substitution.]   0.000236    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000002s :      2: substitution.fold_const_symbol
 2.95% :     0.000007s :      3: substitution.graph_param_transform
84.25% :     0.000199s :      4: substitution.inline
 1.25% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.02% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.23% :     0.000005s :      2: substitution.replace_old_param
 5.81% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012866     2
90.01% :     0.011581s :      1: type_inference.infer
 9.99% :     0.001286s :      1: type_inference.specialize
------[replace.]   0.000080     6
66.64% :     0.000053s :      4: replace.inline
33.36% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000209     6
94.00% :     0.000197s :      4: match.inline
 6.00% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.44% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.30% :     0.000003s :     10: predicate.arithmetic_simplify
 1.12% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.49% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.64% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.37% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 0.95% :     0.000001s :     10: predicate.minmaximum_grad
 1.63% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.51% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.73% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.18% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.65% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.77% :     0.000010s :     53: predicate.switch_simplify
 1.40% :     0.000002s :     10: predicate.tile_eliminate
 1.03% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.74% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.85% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000946    12
53.83% :     0.000509s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 3.49% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.68% :     0.000404s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.031360    72
 0.17% :     0.000055s :      1: add_recomputation
 0.27% :     0.000084s :      1: auto_monad
 0.06% :     0.000019s :      1: auto_monad_reorder
 1.52% :     0.000476s :      1: bootstrap
 0.09% :     0.000029s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.09% :     0.000027s :      1: event_method
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
42.33% :     0.013273s :      1: jit_opt_a
 0.58% :     0.000182s :      1: jit_opt_after_cconv
 0.19% :     0.000058s :      1: jit_opt_b
 1.55% :     0.000487s :      1: loop_unroll
 2.10% :     0.000660s :      1: mutable_eliminate
 2.93% :     0.000920s :     26: opt.transform.jit_opt_a
 0.17% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.60% :     0.000501s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000060s :      1: py_interpret_to_execute
 0.05% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000019s :      1: remove_dup_value
 2.00% :     0.000626s :      1: renormalize.infer
 1.30% :     0.000409s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000041s :      1: rewriter_after_opt_a
 0.29% :     0.000090s :      1: rewriter_before_opt_a
 0.26% :     0.000083s :      1: symbol_engine_optimizer
41.42% :     0.012988s :      1: type_inference


TotalTime = 0.0206203, [33]
[bootstrap]: 0.00040414
[type_inference]: 0.00511982
[event_method]: 1.67e-05
[auto_monad]: 6.675e-05
[graph_reusing]: 5.84999e-06
[pre_auto_parallel]: 2.32999e-06
[py_interpret_to_execute]: 3.784e-05
[rewriter_before_opt_a]: 7.116e-05
[expand_dump_flag]: 3.16999e-06
[jit_opt_a]: 0.00593736, [2]
    [Cycle 1]: 0.00191204, [27]
        [switch_simplify]: 5.079e-05
        [loop_unroll]: 2.925e-05
        [a_1]: 0.00056126
        [with_stream_mark]: 1.877e-05
        [recompute_prepare]: 7.28e-06
        [updatestate_depend_eliminate]: 3.21001e-06
        [updatestate_assign_eliminate]: 2.63e-06
        [updatestate_loads_eliminate]: 2.01003e-06
        [parameter_eliminate]: 1.69e-06
        [specialize_transform]: 6.31e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 5.12999e-06
        [merge_forward]: 3.46999e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.459e-05
        [j_node_and_user_rematch]: 1.771e-05
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 9.97999e-06
        [inline_without_move]: 5.69e-06
        [renormalize]: 0.00091864
        [add_forward_monad_depend]: 6.61999e-06
        [auto_monad_grad]: 2.60997e-06
        [auto_monad_eliminator]: 1.588e-05
        [cse]: 2.846e-05
        [replace_applicator]: 1.462e-05
    [Cycle 2]: 0.00033301, [27]
        [switch_simplify]: 5.83997e-06
        [loop_unroll]: 4.62e-06
        [a_1]: 9.412e-05
        [with_stream_mark]: 1.209e-05
        [recompute_prepare]: 5.06002e-06
        [updatestate_depend_eliminate]: 2.46998e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 4.82998e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 4.82e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.32001e-06
        [cell_reuse_recompute_pass]: 1.97999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.343e-05
        [j_node_and_user_rematch]: 7.65e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 8.33999e-06
        [inline_without_move]: 4.66002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.11002e-06
        [auto_monad_grad]: 9.30013e-07
        [auto_monad_eliminator]: 5.32999e-06
        [cse]: 1.053e-05
        [replace_applicator]: 4.96002e-06
[py_interpret_to_execute_after_opt_a]: 1.378e-05
[rewriter_after_opt_a]: 3.784e-05
[convert_after_rewriter]: 6.86999e-06
[order_py_execute_after_rewriter]: 4.91002e-06
[mutable_eliminate]: 0.00072167
[jit_opt_b]: 5.436e-05, [1]
    [Cycle 1]: 4.601e-05, [2]
        [frontend_op_eliminate]: 1.858e-05
        [inline_after_opt_a]: 1.511e-05
[cconv]: 2.832e-05
[loop_unroll]: 0.00043686
[jit_opt_after_cconv]: 0.00014479, [1]
    [Cycle 1]: 0.00013754, [11]
        [c_1]: 2.172e-05
        [parameter_eliminate]: 2.74001e-06
        [updatestate_depend_eliminate]: 5.39998e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 1.95001e-06
        [cse]: 1.942e-05
        [call_graph_tuple_transform]: 2.018e-05
        [tuple_list_get_item_eliminator]: 5.24e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 5.00001e-06
[remove_dup_value]: 1.54e-05
[partial_unused_args_eliminate]: 2.46e-06
[environ_conv]: 5.67001e-06
[add_recomputation]: 0.00014183
[cse_after_recomputation]: 2.607e-05, [1]
    [Cycle 1]: 1.955e-05, [1]
        [cse]: 1.254e-05
[auto_monad_reorder]: 1.624e-05
[get_jit_bprop_graph]: 2.02001e-06
[rewriter_after_jit_bprop_graph]: 6.56999e-06
[opt_after_jit_grad]: 0.00048526
[symbol_engine_optimizer]: 8.941e-05, [1]
    [Cycle 1]: 8.22e-05, [6]
        [build]: 3.89002e-06
        [elim_shapecalc]: 7.96001e-06
        [elim_not_effective]: 1.308e-05
        [opt_reshape]: 5.67999e-06
        [fold_const_symbol]: 2.161e-05
        [renormalize]: 8.99978e-07
[validate]: 3.958e-05
[backend_pass]: 8.90024e-07
[task_emit]: 0.00643909
[execute]: 8.90001e-06
Sums
    bootstrap                            :     0.000404s :  2.48%
    type_inference                       :     0.005120s : 31.44%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000067s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000038s :  0.23%
    rewriter_before_opt_a                :     0.000071s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.35%
    jit_opt_a.loop_unroll                :     0.000034s :  0.21%
    jit_opt_a.a_1                        :     0.000655s :  4.02%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000025s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000919s :  5.64%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.13%
    jit_opt_a.cse                        :     0.000039s :  0.24%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000722s :  4.43%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000028s :  0.17%
    loop_unroll                          :     0.000437s :  2.68%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000142s :  0.87%
    cse_after_recomputation.cse          :     0.000013s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000485s :  2.98%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000022s :  0.13%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006439s : 39.55%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000231    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.77% :     0.000002s :      2: substitution.fold_const_symbol
 2.65% :     0.000006s :      3: substitution.graph_param_transform
84.28% :     0.000194s :      4: substitution.inline
 1.76% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.12% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.35% :     0.000005s :      2: substitution.replace_old_param
 5.22% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005049     2
82.48% :     0.004165s :      1: type_inference.infer
17.52% :     0.000884s :      1: type_inference.specialize
------[replace.]   0.000072     6
67.62% :     0.000049s :      4: replace.inline
32.38% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000203     6
94.59% :     0.000192s :      4: match.inline
 5.41% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.38% :     0.000003s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.08% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.53% :     0.000007s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.04% :     0.000001s :      6: predicate.less_batch_normalization
 1.80% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.28% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.24% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.89% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000000s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.26% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.80% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.50% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.04% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.46% :     0.000002s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.96% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.21% :     0.000003s :     16: predicate.switch_defer_inline
 1.97% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.40% :     0.000009s :     53: predicate.switch_simplify
 1.40% :     0.000002s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.28% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.34% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.69% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000678    12
48.69% :     0.000330s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.71% :     0.000046s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.60% :     0.000303s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022508    76
 0.65% :     0.000147s :      1: add_recomputation
 0.31% :     0.000070s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.88% :     0.000424s :      1: bootstrap
 0.14% :     0.000031s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000029s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.09% :     0.000020s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.40% :     0.005942s :      1: jit_opt_a
 0.66% :     0.000148s :      1: jit_opt_after_cconv
 0.25% :     0.000057s :      1: jit_opt_b
 1.98% :     0.000445s :      1: loop_unroll
 3.24% :     0.000730s :      1: mutable_eliminate
 3.88% :     0.000874s :     26: opt.transform.jit_opt_a
 0.21% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.20% :     0.000495s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000041s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 2.26% :     0.000509s :      1: renormalize.infer
 1.78% :     0.000400s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000041s :      1: rewriter_after_opt_a
 0.33% :     0.000074s :      1: rewriter_before_opt_a
 0.41% :     0.000092s :      1: symbol_engine_optimizer
28.70% :     0.006460s :      1: task_emit
22.80% :     0.005133s :      1: type_inference
 0.26% :     0.000059s :      1: validate


TotalTime = 0.0212973, [33]
[bootstrap]: 0.00029177
[type_inference]: 0.00488937
[event_method]: 1.861e-05
[auto_monad]: 6.747e-05
[graph_reusing]: 6.25002e-06
[pre_auto_parallel]: 2.68e-06
[py_interpret_to_execute]: 3.643e-05
[rewriter_before_opt_a]: 7.395e-05
[expand_dump_flag]: 3.52002e-06
[jit_opt_a]: 0.00584085, [2]
    [Cycle 1]: 0.00196265, [27]
        [switch_simplify]: 6.14e-05
        [loop_unroll]: 3.044e-05
        [a_1]: 0.00059856
        [with_stream_mark]: 2.273e-05
        [recompute_prepare]: 8.07e-06
        [updatestate_depend_eliminate]: 4.45e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 2.65002e-06
        [parameter_eliminate]: 2.34001e-06
        [specialize_transform]: 6.34999e-06
        [updatestate_useless_node_eliminater]: 5.48002e-06
        [accelerated_algorithm]: 5.39998e-06
        [meta_shard_fg_expand]: 2.81e-06
        [get_grad_eliminate_]: 5.20999e-06
        [merge_forward]: 3.83999e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.545e-05
        [j_node_and_user_rematch]: 9.54e-06
        [meta_fg_expand]: 2.59001e-06
        [replace_old_param]: 9.71e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.00090223
        [add_forward_monad_depend]: 6.36998e-06
        [auto_monad_grad]: 2.69001e-06
        [auto_monad_eliminator]: 1.48e-05
        [cse]: 3.324e-05
        [replace_applicator]: 1.348e-05
    [Cycle 2]: 0.00034476, [27]
        [switch_simplify]: 6.21e-06
        [loop_unroll]: 5.20001e-06
        [a_1]: 9.53e-05
        [with_stream_mark]: 1.116e-05
        [recompute_prepare]: 5.84999e-06
        [updatestate_depend_eliminate]: 2.75002e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 5.37999e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.46002e-06
        [meta_shard_fg_expand]: 1.90001e-06
        [get_grad_eliminate_]: 4.67e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 2.43e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.438e-05
        [j_node_and_user_rematch]: 8.47998e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 7.66001e-06
        [inline_without_move]: 4.92e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 6.36998e-06
        [cse]: 1.216e-05
        [replace_applicator]: 5.37001e-06
[py_interpret_to_execute_after_opt_a]: 1.293e-05
[rewriter_after_opt_a]: 3.812e-05
[convert_after_rewriter]: 7.23999e-06
[order_py_execute_after_rewriter]: 5.07999e-06
[mutable_eliminate]: 0.00073637
[jit_opt_b]: 5.56e-05, [1]
    [Cycle 1]: 4.748e-05, [2]
        [frontend_op_eliminate]: 1.95e-05
        [inline_after_opt_a]: 1.586e-05
[cconv]: 3.142e-05
[loop_unroll]: 0.00050768
[jit_opt_after_cconv]: 0.00016396, [1]
    [Cycle 1]: 0.00015658, [11]
        [c_1]: 2.451e-05
        [parameter_eliminate]: 3.68999e-06
        [updatestate_depend_eliminate]: 7.66999e-06
        [updatestate_assign_eliminate]: 2.85002e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [cse]: 2.662e-05
        [call_graph_tuple_transform]: 2.32e-05
        [tuple_list_get_item_eliminator]: 5.86e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 5.61998e-06
[remove_dup_value]: 1.581e-05
[partial_unused_args_eliminate]: 2.43e-06
[environ_conv]: 5.89e-06
[add_recomputation]: 5.389e-05
[cse_after_recomputation]: 2.168e-05, [1]
    [Cycle 1]: 1.546e-05, [1]
        [cse]: 9.78002e-06
[auto_monad_reorder]: 1.608e-05
[get_jit_bprop_graph]: 2.43e-06
[rewriter_after_jit_bprop_graph]: 6.93e-06
[opt_after_jit_grad]: 0.00050074
[symbol_engine_optimizer]: 8.173e-05, [1]
    [Cycle 1]: 7.386e-05, [6]
        [build]: 5.26998e-06
        [elim_shapecalc]: 8.37e-06
        [elim_not_effective]: 1.495e-05
        [opt_reshape]: 6.58e-06
        [fold_const_symbol]: 8.92e-06
        [renormalize]: 5.99975e-07
[validate]: 4.445e-05
[backend_pass]: 8.59989e-07
[task_emit]: 0.00749893
[execute]: 9.81e-06
Sums
    bootstrap                            :     0.000292s :  1.71%
    type_inference                       :     0.004889s : 28.62%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000067s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.21%
    rewriter_before_opt_a                :     0.000074s :  0.43%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.40%
    jit_opt_a.loop_unroll                :     0.000036s :  0.21%
    jit_opt_a.a_1                        :     0.000694s :  4.06%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000902s :  5.28%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.12%
    jit_opt_a.cse                        :     0.000045s :  0.27%
    jit_opt_a.replace_applicator         :     0.000019s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000736s :  4.31%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000031s :  0.18%
    loop_unroll                          :     0.000508s :  2.97%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000054s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000501s :  2.93%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007499s : 43.90%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000250    23
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.50% :     0.000006s :      3: substitution.graph_param_transform
85.17% :     0.000213s :      4: substitution.inline
 1.44% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.92% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.04% :     0.000005s :      2: substitution.replace_old_param
 5.47% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004813     2
80.92% :     0.003895s :      1: type_inference.infer
19.08% :     0.000918s :      1: type_inference.specialize
------[replace.]   0.000075     6
66.87% :     0.000050s :      4: replace.inline
33.13% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000222     6
94.37% :     0.000210s :      4: match.inline
 5.63% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.21% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 2.00% :     0.000003s :     10: predicate.arithmetic_simplify
 1.32% :     0.000002s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.22% :     0.000002s :     10: predicate.depend_value_elim
 0.99% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.16% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.71% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.41% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 6.36% :     0.000008s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 2.15% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.43% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 2.41% :     0.000003s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000003s :     15: predicate.partial_eliminate
 1.09% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.00% :     0.000003s :     10: predicate.reduce_eliminate
 1.36% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.39% :     0.000001s :      3: predicate.reset_defer_inline
 1.26% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.14% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.22% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.13% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     16: predicate.switch_defer_inline
 1.91% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.71% :     0.000010s :     53: predicate.switch_simplify
 1.25% :     0.000002s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.12% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.05% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000694    12
47.42% :     0.000329s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.44% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.14% :     0.000313s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023229    76
 0.24% :     0.000057s :      1: add_recomputation
 0.31% :     0.000071s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.34% :     0.000311s :      1: bootstrap
 0.15% :     0.000034s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.16% :     0.005844s :      1: jit_opt_a
 0.72% :     0.000167s :      1: jit_opt_after_cconv
 0.25% :     0.000059s :      1: jit_opt_b
 2.24% :     0.000520s :      1: loop_unroll
 3.23% :     0.000749s :      1: mutable_eliminate
 3.95% :     0.000918s :     26: opt.transform.jit_opt_a
 0.24% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000029s :      4: opt.transform.jit_opt_b
 0.06% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.20% :     0.000512s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000039s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 2.11% :     0.000491s :      1: renormalize.infer
 1.73% :     0.000403s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000041s :      1: rewriter_after_opt_a
 0.33% :     0.000078s :      1: rewriter_before_opt_a
 0.36% :     0.000084s :      1: symbol_engine_optimizer
32.37% :     0.007520s :      1: task_emit
21.12% :     0.004905s :      1: type_inference
 0.30% :     0.000070s :      1: validate


TotalTime = 0.0517182, [30]
[bootstrap]: 0.00044455
[type_inference]: 0.00571848
[event_method]: 2.101e-05
[auto_monad]: 6.987e-05
[graph_reusing]: 6.79999e-06
[pre_auto_parallel]: 2.59999e-06
[py_interpret_to_execute]: 4.361e-05
[rewriter_before_opt_a]: 7.897e-05
[expand_dump_flag]: 3.43e-06
[jit_opt_a]: 0.00676329, [2]
    [Cycle 1]: 0.00201386, [27]
        [switch_simplify]: 6.496e-05
        [loop_unroll]: 3.04e-05
        [a_1]: 0.0006487
        [with_stream_mark]: 2.38e-05
        [recompute_prepare]: 8.14002e-06
        [updatestate_depend_eliminate]: 4.52e-06
        [updatestate_assign_eliminate]: 3.10998e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 2.28002e-06
        [specialize_transform]: 5.72999e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 5.34e-06
        [meta_shard_fg_expand]: 2.39001e-06
        [get_grad_eliminate_]: 5.30999e-06
        [merge_forward]: 3.78999e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.606e-05
        [j_node_and_user_rematch]: 1.007e-05
        [meta_fg_expand]: 2.93998e-06
        [replace_old_param]: 1.023e-05
        [inline_without_move]: 5.70001e-06
        [renormalize]: 0.00086549
        [add_forward_monad_depend]: 5.99999e-06
        [auto_monad_grad]: 2.24999e-06
        [auto_monad_eliminator]: 1.605e-05
        [cse]: 3.2e-05
        [replace_applicator]: 1.461e-05
    [Cycle 2]: 0.00034251, [27]
        [switch_simplify]: 5.89999e-06
        [loop_unroll]: 4.82e-06
        [a_1]: 9.587e-05
        [with_stream_mark]: 1.078e-05
        [recompute_prepare]: 5.02e-06
        [updatestate_depend_eliminate]: 2.98e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 5.42001e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 4.80001e-06
        [merge_forward]: 2.73998e-06
        [cell_reuse_recompute_pass]: 1.54e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.327e-05
        [j_node_and_user_rematch]: 8.18999e-06
        [meta_fg_expand]: 1.80001e-06
        [replace_old_param]: 7.31999e-06
        [inline_without_move]: 4.80999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.62999e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 5.30999e-06
        [cse]: 1.171e-05
        [replace_applicator]: 5.47999e-06
[py_interpret_to_execute_after_opt_a]: 9.52999e-06
[rewriter_after_opt_a]: 3.602e-05
[convert_after_rewriter]: 7.15e-06
[order_py_execute_after_rewriter]: 5.47999e-06
[mutable_eliminate]: 0.00068846
[jit_opt_b]: 5.746e-05, [1]
    [Cycle 1]: 4.951e-05, [2]
        [frontend_op_eliminate]: 2.12e-05
        [inline_after_opt_a]: 1.605e-05
[cconv]: 2.907e-05
[loop_unroll]: 0.00044399
[jit_opt_after_cconv]: 0.00015256, [1]
    [Cycle 1]: 0.00014541, [11]
        [c_1]: 2.238e-05
        [parameter_eliminate]: 3.71999e-06
        [updatestate_depend_eliminate]: 6.76e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 2.286e-05
        [call_graph_tuple_transform]: 2.063e-05
        [tuple_list_get_item_eliminator]: 6.01e-06
        [none_parameter_eliminate]: 1.56002e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 5.20999e-06
[remove_dup_value]: 1.435e-05
[partial_unused_args_eliminate]: 2.58e-06
[environ_conv]: 5.61998e-06
[add_recomputation]: 5.446e-05
[cse_after_recomputation]: 2.143e-05, [1]
    [Cycle 1]: 1.551e-05, [1]
        [cse]: 9.34998e-06
[auto_monad_reorder]: 1.673e-05
[get_jit_bprop_graph]: 2.32999e-06
[rewriter_after_jit_bprop_graph]: 5.74e-06
[opt_after_jit_grad]: 0.036611
[symbol_engine_optimizer]: 0.00011593, [1]
    [Cycle 1]: 0.00010132, [6]
        [build]: 9.72999e-06
        [elim_shapecalc]: 1.007e-05
        [elim_not_effective]: 2.598e-05
        [opt_reshape]: 9.41e-06
        [fold_const_symbol]: 1.025e-05
        [renormalize]: 1.17999e-06
[validate]: 5.622e-05
Sums
    bootstrap                            :     0.000445s :  0.95%
    type_inference                       :     0.005718s : 12.27%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000070s :  0.15%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.09%
    rewriter_before_opt_a                :     0.000079s :  0.17%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000071s :  0.15%
    jit_opt_a.loop_unroll                :     0.000035s :  0.08%
    jit_opt_a.a_1                        :     0.000745s :  1.60%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.02%
    jit_opt_a.merge_forward              :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000018s :  0.04%
    jit_opt_a.inline_without_move        :     0.000011s :  0.02%
    jit_opt_a.renormalize                :     0.000866s :  1.86%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.05%
    jit_opt_a.cse                        :     0.000044s :  0.09%
    jit_opt_a.replace_applicator         :     0.000020s :  0.04%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.02%
    rewriter_after_opt_a                 :     0.000036s :  0.08%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000688s :  1.48%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.03%
    cconv                                :     0.000029s :  0.06%
    loop_unroll                          :     0.000444s :  0.95%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.05%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000023s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.01%
    remove_dup_value                     :     0.000014s :  0.03%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.01%
    add_recomputation                    :     0.000054s :  0.12%
    cse_after_recomputation.cse          :     0.000009s :  0.02%
    auto_monad_reorder                   :     0.000017s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.036611s : 78.56%
    symbol_engine_optimizer.build        :     0.000010s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000026s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000056s :  0.12%


Time group info:
------[substitution.]   0.000295    23
 1.26% :     0.000004s :      2: substitution.elim_not_effective
 0.53% :     0.000002s :      2: substitution.fold_const_symbol
 1.91% :     0.000006s :      3: substitution.graph_param_transform
87.54% :     0.000258s :      4: substitution.inline
 1.28% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.55% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.55% :     0.000005s :      2: substitution.replace_old_param
 4.38% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005628     2
81.87% :     0.004608s :      1: type_inference.infer
18.13% :     0.001020s :      1: type_inference.specialize
------[replace.]   0.000080     6
69.75% :     0.000056s :      4: replace.inline
30.25% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000268     6
95.56% :     0.000256s :      4: match.inline
 4.44% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 3.60% :     0.000005s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 2.07% :     0.000003s :     10: predicate.arithmetic_simplify
 1.14% :     0.000001s :     10: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.15% :     0.000002s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.16% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.09% :     0.000001s :     10: predicate.dict_set_item_eliminator
 2.52% :     0.000003s :      3: predicate.dumpgradient_eliminate
 0.64% :     0.000001s :      3: predicate.elim_not_effective
 0.70% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.04% :     0.000001s :     10: predicate.environ_get_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.58% :     0.000007s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000002s :      6: predicate.less_batch_normalization
 1.35% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.55% :     0.000002s :     15: predicate.load_eliminater
 1.33% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.75% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.76% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 1.62% :     0.000002s :      3: predicate.mutable_eliminate
 0.87% :     0.000001s :      3: predicate.opt_reshape
 2.34% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.41% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.07% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.83% :     0.000002s :      6: predicate.special_op_eliminate
 1.13% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 1.91% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.88% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000002s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.68% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000743    12
50.66% :     0.000376s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.48% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.85% :     0.000333s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.053635    72
 0.11% :     0.000058s :      1: add_recomputation
 0.14% :     0.000073s :      1: auto_monad
 0.04% :     0.000019s :      1: auto_monad_reorder
 0.87% :     0.000467s :      1: bootstrap
 0.06% :     0.000032s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.04% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.05% :     0.000026s :      1: event_method
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
12.62% :     0.006767s :      1: jit_opt_a
 0.29% :     0.000156s :      1: jit_opt_after_cconv
 0.11% :     0.000060s :      1: jit_opt_b
 0.84% :     0.000452s :      1: loop_unroll
 1.30% :     0.000697s :      1: mutable_eliminate
 1.81% :     0.000970s :     26: opt.transform.jit_opt_a
 0.09% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000030s :      4: opt.transform.jit_opt_b
 0.02% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000059s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000051s :      4: opt.transform.symbol_engine_opt
68.31% :     0.036640s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000047s :      1: py_interpret_to_execute
 0.02% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000017s :      1: remove_dup_value
 0.88% :     0.000473s :      1: renormalize.infer
 0.71% :     0.000383s :      1: renormalize.specialize
 0.01% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000039s :      1: rewriter_after_opt_a
 0.15% :     0.000082s :      1: rewriter_before_opt_a
 0.22% :     0.000121s :      1: symbol_engine_optimizer
10.70% :     0.005737s :      1: type_inference


TotalTime = 0.0420442, [33]
[bootstrap]: 0.00034872
[type_inference]: 0.00519677
[event_method]: 1.879e-05
[auto_monad]: 6.51e-05
[graph_reusing]: 5.95002e-06
[pre_auto_parallel]: 2.73e-06
[py_interpret_to_execute]: 3.78e-05
[rewriter_before_opt_a]: 7.713e-05
[expand_dump_flag]: 3.46999e-06
[jit_opt_a]: 0.0266363, [2]
    [Cycle 1]: 0.0224234, [27]
        [switch_simplify]: 6.043e-05
        [loop_unroll]: 2.871e-05
        [a_1]: 0.020793
        [with_stream_mark]: 3.297e-05
        [recompute_prepare]: 1.593e-05
        [updatestate_depend_eliminate]: 4.46002e-06
        [updatestate_assign_eliminate]: 3.2e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 3.54002e-06
        [specialize_transform]: 6.51999e-06
        [updatestate_useless_node_eliminater]: 5.62001e-06
        [accelerated_algorithm]: 6.41e-06
        [meta_shard_fg_expand]: 3.77998e-06
        [get_grad_eliminate_]: 6.24999e-06
        [merge_forward]: 4.37e-06
        [cell_reuse_recompute_pass]: 4.58999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.183e-05
        [j_node_and_user_rematch]: 1.269e-05
        [meta_fg_expand]: 3.01999e-06
        [replace_old_param]: 1.423e-05
        [inline_without_move]: 6.13998e-06
        [renormalize]: 0.00109037
        [add_forward_monad_depend]: 8.17e-06
        [auto_monad_grad]: 3.28998e-06
        [auto_monad_eliminator]: 1.72e-05
        [cse]: 3.157e-05
        [replace_applicator]: 1.49e-05
    [Cycle 2]: 0.00034366, [27]
        [switch_simplify]: 5.71998e-06
        [loop_unroll]: 5.92999e-06
        [a_1]: 9.658e-05
        [with_stream_mark]: 1.263e-05
        [recompute_prepare]: 5.41998e-06
        [updatestate_depend_eliminate]: 2.74999e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 1.50999e-06
        [specialize_transform]: 4.79e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 5.48002e-06
        [meta_shard_fg_expand]: 1.62999e-06
        [get_grad_eliminate_]: 5.30999e-06
        [merge_forward]: 2.88e-06
        [cell_reuse_recompute_pass]: 3.07002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.506e-05
        [j_node_and_user_rematch]: 8.25e-06
        [meta_fg_expand]: 1.39998e-06
        [replace_old_param]: 7.65e-06
        [inline_without_move]: 4.87998e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.05999e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 5.40999e-06
        [cse]: 1.128e-05
        [replace_applicator]: 4.63001e-06
[py_interpret_to_execute_after_opt_a]: 1.392e-05
[rewriter_after_opt_a]: 3.792e-05
[convert_after_rewriter]: 7.29001e-06
[order_py_execute_after_rewriter]: 5.14003e-06
[mutable_eliminate]: 0.00082516
[jit_opt_b]: 5.268e-05, [1]
    [Cycle 1]: 4.42e-05, [2]
        [frontend_op_eliminate]: 1.72e-05
        [inline_after_opt_a]: 1.509e-05
[cconv]: 2.627e-05
[loop_unroll]: 0.00046377
[jit_opt_after_cconv]: 0.00014731, [1]
    [Cycle 1]: 0.00014095, [11]
        [c_1]: 2.164e-05
        [parameter_eliminate]: 3.06001e-06
        [updatestate_depend_eliminate]: 6.04999e-06
        [updatestate_assign_eliminate]: 2.61999e-06
        [updatestate_loads_eliminate]: 2.20002e-06
        [cse]: 1.847e-05
        [call_graph_tuple_transform]: 2.184e-05
        [tuple_list_get_item_eliminator]: 5.99e-06
        [none_parameter_eliminate]: 1.56998e-06
        [renormalize]: 7.50006e-07
        [switch_simplify]: 6.04001e-06
[remove_dup_value]: 1.574e-05
[partial_unused_args_eliminate]: 2.96001e-06
[environ_conv]: 5.41998e-06
[add_recomputation]: 5.183e-05
[cse_after_recomputation]: 1.948e-05, [1]
    [Cycle 1]: 1.449e-05, [1]
        [cse]: 8.92e-06
[auto_monad_reorder]: 1.599e-05
[get_jit_bprop_graph]: 2.13002e-06
[rewriter_after_jit_bprop_graph]: 7.60998e-06
[opt_after_jit_grad]: 0.00057231
[symbol_engine_optimizer]: 8.312e-05, [1]
    [Cycle 1]: 7.568e-05, [6]
        [build]: 3.55e-06
        [elim_shapecalc]: 8.53001e-06
        [elim_not_effective]: 1.644e-05
        [opt_reshape]: 7.53999e-06
        [fold_const_symbol]: 9.42999e-06
        [renormalize]: 6.19999e-07
[validate]: 4.044e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.00701685
[execute]: 9.64e-06
Sums
    bootstrap                            :     0.000349s :  0.93%
    type_inference                       :     0.005197s : 13.86%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000065s :  0.17%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000038s :  0.10%
    rewriter_before_opt_a                :     0.000077s :  0.21%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000066s :  0.18%
    jit_opt_a.loop_unroll                :     0.000035s :  0.09%
    jit_opt_a.a_1                        :     0.020890s : 55.73%
    jit_opt_a.with_stream_mark           :     0.000046s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000008s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000022s :  0.06%
    jit_opt_a.inline_without_move        :     0.000011s :  0.03%
    jit_opt_a.renormalize                :     0.001090s :  2.91%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.06%
    jit_opt_a.cse                        :     0.000043s :  0.11%
    jit_opt_a.replace_applicator         :     0.000020s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.04%
    rewriter_after_opt_a                 :     0.000038s :  0.10%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000825s :  2.20%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.04%
    cconv                                :     0.000026s :  0.07%
    loop_unroll                          :     0.000464s :  1.24%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.04%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000005s :  0.01%
    add_recomputation                    :     0.000052s :  0.14%
    cse_after_recomputation.cse          :     0.000009s :  0.02%
    auto_monad_reorder                   :     0.000016s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000572s :  1.53%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007017s : 18.72%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.000249    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000002s :      2: substitution.fold_const_symbol
 2.88% :     0.000007s :      3: substitution.graph_param_transform
82.56% :     0.000205s :      4: substitution.inline
 2.17% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.62% :     0.000007s :      4: substitution.remove_not_recompute_node
 3.20% :     0.000008s :      2: substitution.replace_old_param
 5.12% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005123     2
81.41% :     0.004171s :      1: type_inference.infer
18.59% :     0.000953s :      1: type_inference.specialize
------[replace.]   0.020210     6
 0.22% :     0.000045s :      4: replace.inline
99.78% :     0.020165s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000215     6
94.53% :     0.000203s :      4: match.inline
 5.47% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000141   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.00% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.86% :     0.000001s :     10: predicate.addn_check_dump
 1.54% :     0.000002s :     10: predicate.addn_zero_filter
 2.72% :     0.000004s :     10: predicate.arithmetic_simplify
 1.84% :     0.000003s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.91% :     0.000001s :     10: predicate.compare_switch_simplify
 1.91% :     0.000003s :     10: predicate.depend_value_elim
 1.70% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.45% :     0.000001s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.78% :     0.000003s :     10: predicate.environ_get_eliminate
 1.23% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 5.53% :     0.000008s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :      6: predicate.less_batch_normalization
 1.82% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.99% :     0.000003s :     15: predicate.load_eliminater
 1.12% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.44% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.80% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.93% :     0.000001s :     10: predicate.merge_addn
 0.88% :     0.000001s :     10: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.58% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.42% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.88% :     0.000003s :     10: predicate.reduce_eliminate
 2.05% :     0.000003s :     12: predicate.redundant_stop_gradient_eliminater
 0.71% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     10: predicate.reshape_eliminate
 1.08% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000002s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.69% :     0.000004s :     16: predicate.switch_defer_inline
 2.13% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.78% :     0.000010s :     53: predicate.switch_simplify
 1.25% :     0.000002s :     10: predicate.tile_eliminate
 1.07% :     0.000002s :     10: predicate.transpose_eliminate
 1.92% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.98% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.36% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.40% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000721    12
45.43% :     0.000327s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.94% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.64% :     0.000343s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.064374    76
 0.09% :     0.000055s :      1: add_recomputation
 0.11% :     0.000069s :      1: auto_monad
 0.03% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.56% :     0.000363s :      1: bootstrap
 0.05% :     0.000029s :      1: cconv
 0.01% :     0.000009s :      1: convert_after_rewriter
 0.03% :     0.000021s :      1: cse_after_recomputation
 0.01% :     0.000008s :      1: environ_conv
 0.04% :     0.000024s :      1: event_method
 0.02% :     0.000016s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000008s :      1: graph_reusing
41.38% :     0.026641s :      1: jit_opt_a
 0.23% :     0.000150s :      1: jit_opt_after_cconv
 0.09% :     0.000055s :      1: jit_opt_b
 0.73% :     0.000472s :      1: loop_unroll
 1.30% :     0.000834s :      1: mutable_eliminate
32.82% :     0.021130s :     26: opt.transform.jit_opt_a
 0.08% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000026s :      4: opt.transform.jit_opt_b
 0.02% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000037s :      4: opt.transform.symbol_engine_opt
 0.90% :     0.000581s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.06% :     0.000040s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000018s :      1: remove_dup_value
 0.94% :     0.000602s :      1: renormalize.infer
 0.74% :     0.000477s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.06% :     0.000041s :      1: rewriter_after_opt_a
 0.12% :     0.000080s :      1: rewriter_before_opt_a
 0.13% :     0.000086s :      1: symbol_engine_optimizer
10.93% :     0.007039s :      1: task_emit
 8.10% :     0.005216s :      1: type_inference
 0.10% :     0.000063s :      1: validate


TotalTime = 0.020531, [33]
[bootstrap]: 0.00033689
[type_inference]: 0.00491021
[event_method]: 1.799e-05
[auto_monad]: 5.152e-05
[graph_reusing]: 4.1e-06
[pre_auto_parallel]: 1.71e-06
[py_interpret_to_execute]: 3.858e-05
[rewriter_before_opt_a]: 6.743e-05
[expand_dump_flag]: 2.31e-06
[jit_opt_a]: 0.00578225, [2]
    [Cycle 1]: 0.0019981, [27]
        [switch_simplify]: 6.419e-05
        [loop_unroll]: 3.15e-05
        [a_1]: 0.00060379
        [with_stream_mark]: 2.601e-05
        [recompute_prepare]: 7.51999e-06
        [updatestate_depend_eliminate]: 3.68e-06
        [updatestate_assign_eliminate]: 3.13e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 1.95001e-06
        [specialize_transform]: 5.81e-06
        [updatestate_useless_node_eliminater]: 5.44998e-06
        [accelerated_algorithm]: 6.07999e-06
        [meta_shard_fg_expand]: 2.66999e-06
        [get_grad_eliminate_]: 5.44e-06
        [merge_forward]: 3.56001e-06
        [cell_reuse_recompute_pass]: 1.60999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.584e-05
        [j_node_and_user_rematch]: 9.96e-06
        [meta_fg_expand]: 2.27999e-06
        [replace_old_param]: 9.17001e-06
        [inline_without_move]: 5.45001e-06
        [renormalize]: 0.0009258
        [add_forward_monad_depend]: 6.56e-06
        [auto_monad_grad]: 2.70997e-06
        [auto_monad_eliminator]: 1.464e-05
        [cse]: 3.219e-05
        [replace_applicator]: 1.194e-05
    [Cycle 2]: 0.00033378, [27]
        [switch_simplify]: 6.45997e-06
        [loop_unroll]: 5.71998e-06
        [a_1]: 9.489e-05
        [with_stream_mark]: 1.017e-05
        [recompute_prepare]: 5.24e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 5.33002e-06
        [updatestate_useless_node_eliminater]: 4.80999e-06
        [accelerated_algorithm]: 5.13002e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 2.62001e-06
        [cell_reuse_recompute_pass]: 2.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.208e-05
        [j_node_and_user_rematch]: 7.78001e-06
        [meta_fg_expand]: 1.92999e-06
        [replace_old_param]: 7.6e-06
        [inline_without_move]: 1.291e-05
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.26002e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 5.07999e-06
        [cse]: 1.101e-05
        [replace_applicator]: 5.05999e-06
[py_interpret_to_execute_after_opt_a]: 1.304e-05
[rewriter_after_opt_a]: 4.125e-05
[convert_after_rewriter]: 6.77002e-06
[order_py_execute_after_rewriter]: 5.69999e-06
[mutable_eliminate]: 0.00067009
[jit_opt_b]: 5.251e-05, [1]
    [Cycle 1]: 4.408e-05, [2]
        [frontend_op_eliminate]: 1.742e-05
        [inline_after_opt_a]: 1.524e-05
[cconv]: 2.722e-05
[loop_unroll]: 0.00043947
[jit_opt_after_cconv]: 0.00014736, [1]
    [Cycle 1]: 0.00014015, [11]
        [c_1]: 2.246e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 5.36998e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [cse]: 2.105e-05
        [call_graph_tuple_transform]: 2.205e-05
        [tuple_list_get_item_eliminator]: 5.27001e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.35001e-06
[remove_dup_value]: 1.464e-05
[partial_unused_args_eliminate]: 2.39999e-06
[environ_conv]: 4.84e-06
[add_recomputation]: 5.442e-05
[cse_after_recomputation]: 2.116e-05, [1]
    [Cycle 1]: 1.532e-05, [1]
        [cse]: 9.00999e-06
[auto_monad_reorder]: 1.564e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 6.79999e-06
[opt_after_jit_grad]: 0.00046768
[symbol_engine_optimizer]: 7.585e-05, [1]
    [Cycle 1]: 6.865e-05, [6]
        [build]: 3.98999e-06
        [elim_shapecalc]: 7.55998e-06
        [elim_not_effective]: 1.346e-05
        [opt_reshape]: 5.64e-06
        [fold_const_symbol]: 8.83001e-06
        [renormalize]: 4.30009e-07
[validate]: 3.829e-05
[backend_pass]: 1.11997e-06
[task_emit]: 0.00697417
[execute]: 1.04e-05
Sums
    bootstrap                            :     0.000337s :  2.05%
    type_inference                       :     0.004910s : 29.88%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000052s :  0.31%
    graph_reusing                        :     0.000004s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.23%
    rewriter_before_opt_a                :     0.000067s :  0.41%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000071s :  0.43%
    jit_opt_a.loop_unroll                :     0.000037s :  0.23%
    jit_opt_a.a_1                        :     0.000699s :  4.25%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000018s :  0.11%
    jit_opt_a.renormalize                :     0.000926s :  5.63%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000043s :  0.26%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000041s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000670s :  4.08%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000027s :  0.17%
    loop_unroll                          :     0.000439s :  2.67%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000054s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000468s :  2.85%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006974s : 42.44%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000254    23
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.38% :     0.000006s :      3: substitution.graph_param_transform
86.27% :     0.000219s :      4: substitution.inline
 1.39% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.02% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.86% :     0.000005s :      2: substitution.replace_old_param
 4.74% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004844     2
81.51% :     0.003948s :      1: type_inference.infer
18.49% :     0.000895s :      1: type_inference.specialize
------[replace.]   0.000078     6
68.49% :     0.000053s :      4: replace.inline
31.51% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000228     6
95.17% :     0.000217s :      4: match.inline
 4.83% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.21% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.24% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 2.28% :     0.000003s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.45% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.32% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.16% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.50% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.98% :     0.000002s :     15: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.56% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 2.32% :     0.000003s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.48% :     0.000003s :     15: predicate.partial_eliminate
 1.09% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.77% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.30% :     0.000002s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.68% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.95% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000695    12
45.88% :     0.000319s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.48% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.63% :     0.000324s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022489    76
 0.26% :     0.000058s :      1: add_recomputation
 0.24% :     0.000055s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.59% :     0.000357s :      1: bootstrap
 0.13% :     0.000030s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
25.73% :     0.005787s :      1: jit_opt_a
 0.67% :     0.000150s :      1: jit_opt_after_cconv
 0.24% :     0.000055s :      1: jit_opt_b
 1.99% :     0.000449s :      1: loop_unroll
 3.03% :     0.000681s :      1: mutable_eliminate
 4.14% :     0.000931s :     26: opt.transform.jit_opt_a
 0.23% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.12% :     0.000476s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000041s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 2.29% :     0.000516s :      1: renormalize.infer
 1.78% :     0.000401s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000044s :      1: rewriter_after_opt_a
 0.31% :     0.000070s :      1: rewriter_before_opt_a
 0.35% :     0.000078s :      1: symbol_engine_optimizer
31.10% :     0.006994s :      1: task_emit
21.89% :     0.004924s :      1: type_inference
 0.26% :     0.000059s :      1: validate


TotalTime = 0.0667139, [33]
[bootstrap]: 0.00038909
[type_inference]: 0.0217461
[event_method]: 2.106e-05
[auto_monad]: 6.763e-05
[graph_reusing]: 6.29999e-06
[pre_auto_parallel]: 2.96999e-06
[py_interpret_to_execute]: 4.425e-05
[rewriter_before_opt_a]: 7.57e-05
[expand_dump_flag]: 3.22002e-06
[jit_opt_a]: 0.00683056, [2]
    [Cycle 1]: 0.00206802, [27]
        [switch_simplify]: 7.723e-05
        [loop_unroll]: 2.997e-05
        [a_1]: 0.00063776
        [with_stream_mark]: 2.516e-05
        [recompute_prepare]: 7.51001e-06
        [updatestate_depend_eliminate]: 3.75e-06
        [updatestate_assign_eliminate]: 2.89001e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.81998e-06
        [specialize_transform]: 5.93998e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 5.92001e-06
        [meta_shard_fg_expand]: 2.34999e-06
        [get_grad_eliminate_]: 5.20999e-06
        [merge_forward]: 4.03001e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.514e-05
        [j_node_and_user_rematch]: 9.59999e-06
        [meta_fg_expand]: 2.21e-06
        [replace_old_param]: 1.005e-05
        [inline_without_move]: 5.30999e-06
        [renormalize]: 0.00095077
        [add_forward_monad_depend]: 6.87002e-06
        [auto_monad_grad]: 2.74001e-06
        [auto_monad_eliminator]: 1.629e-05
        [cse]: 3.171e-05
        [replace_applicator]: 1.364e-05
    [Cycle 2]: 0.00032599, [27]
        [switch_simplify]: 5.92999e-06
        [loop_unroll]: 4.82e-06
        [a_1]: 9.147e-05
        [with_stream_mark]: 9.89999e-06
        [recompute_prepare]: 5.04998e-06
        [updatestate_depend_eliminate]: 2.84999e-06
        [updatestate_assign_eliminate]: 2.00002e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.39e-06
        [specialize_transform]: 5.09e-06
        [updatestate_useless_node_eliminater]: 4.73001e-06
        [accelerated_algorithm]: 4.85001e-06
        [meta_shard_fg_expand]: 1.24998e-06
        [get_grad_eliminate_]: 4.28999e-06
        [merge_forward]: 3.04001e-06
        [cell_reuse_recompute_pass]: 2.21e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.345e-05
        [j_node_and_user_rematch]: 7.73001e-06
        [meta_fg_expand]: 1.55999e-06
        [replace_old_param]: 7.8e-06
        [inline_without_move]: 4.49002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 5.07999e-06
        [cse]: 1.123e-05
        [replace_applicator]: 5.02e-06
[py_interpret_to_execute_after_opt_a]: 1.406e-05
[rewriter_after_opt_a]: 4.104e-05
[convert_after_rewriter]: 7.00002e-06
[order_py_execute_after_rewriter]: 5.34998e-06
[mutable_eliminate]: 0.00069747
[jit_opt_b]: 4.901e-05, [1]
    [Cycle 1]: 4.14e-05, [2]
        [frontend_op_eliminate]: 1.617e-05
        [inline_after_opt_a]: 1.427e-05
[cconv]: 2.935e-05
[loop_unroll]: 0.00044644
[jit_opt_after_cconv]: 0.00015511, [1]
    [Cycle 1]: 0.00014803, [11]
        [c_1]: 2.118e-05
        [parameter_eliminate]: 3.92002e-06
        [updatestate_depend_eliminate]: 8.97999e-06
        [updatestate_assign_eliminate]: 2.82002e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [cse]: 2.42e-05
        [call_graph_tuple_transform]: 2.197e-05
        [tuple_list_get_item_eliminator]: 5.32001e-06
        [none_parameter_eliminate]: 1.96998e-06
        [renormalize]: 9.60019e-07
        [switch_simplify]: 5.37999e-06
[remove_dup_value]: 1.549e-05
[partial_unused_args_eliminate]: 2.37001e-06
[environ_conv]: 5.69e-06
[add_recomputation]: 5.639e-05
[cse_after_recomputation]: 2.114e-05, [1]
    [Cycle 1]: 1.516e-05, [1]
        [cse]: 9.10999e-06
[auto_monad_reorder]: 1.672e-05
[get_jit_bprop_graph]: 2.60002e-06
[rewriter_after_jit_bprop_graph]: 6.76999e-06
[opt_after_jit_grad]: 0.00049765
[symbol_engine_optimizer]: 7.496e-05, [1]
    [Cycle 1]: 6.804e-05, [6]
        [build]: 4.2e-06
        [elim_shapecalc]: 8.10999e-06
        [elim_not_effective]: 1.351e-05
        [opt_reshape]: 5.49998e-06
        [fold_const_symbol]: 8.60999e-06
        [renormalize]: 3.60014e-07
[validate]: 4.457e-05
[backend_pass]: 8.60018e-07
[task_emit]: 0.0350834
[execute]: 1.012e-05
Sums
    bootstrap                            :     0.000389s :  0.63%
    type_inference                       :     0.021746s : 35.30%
    event_method                         :     0.000021s :  0.03%
    auto_monad                           :     0.000068s :  0.11%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000044s :  0.07%
    rewriter_before_opt_a                :     0.000076s :  0.12%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000083s :  0.13%
    jit_opt_a.loop_unroll                :     0.000035s :  0.06%
    jit_opt_a.a_1                        :     0.000729s :  1.18%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.06%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.02%
    jit_opt_a.merge_forward              :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.05%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000018s :  0.03%
    jit_opt_a.inline_without_move        :     0.000010s :  0.02%
    jit_opt_a.renormalize                :     0.000951s :  1.54%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.03%
    jit_opt_a.cse                        :     0.000043s :  0.07%
    jit_opt_a.replace_applicator         :     0.000019s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.02%
    rewriter_after_opt_a                 :     0.000041s :  0.07%
    convert_after_rewriter               :     0.000007s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000697s :  1.13%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.02%
    cconv                                :     0.000029s :  0.05%
    loop_unroll                          :     0.000446s :  0.72%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.03%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000024s :  0.04%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.01%
    remove_dup_value                     :     0.000015s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000006s :  0.01%
    add_recomputation                    :     0.000056s :  0.09%
    cse_after_recomputation.cse          :     0.000009s :  0.01%
    auto_monad_reorder                   :     0.000017s :  0.03%
    get_jit_bprop_graph                  :     0.000003s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.01%
    opt_after_jit_grad                   :     0.000498s :  0.81%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.02%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000045s :  0.07%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.035083s : 56.94%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000282    23
 0.96% :     0.000003s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.23% :     0.000006s :      3: substitution.graph_param_transform
86.83% :     0.000245s :      4: substitution.inline
 1.08% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.58% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.90% :     0.000005s :      2: substitution.replace_old_param
 4.96% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.021658     2
94.90% :     0.020553s :      1: type_inference.infer
 5.10% :     0.001105s :      1: type_inference.specialize
------[replace.]   0.000077     6
66.55% :     0.000051s :      4: replace.inline
33.45% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000255     6
94.97% :     0.000242s :      4: match.inline
 5.03% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.65% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.69% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.46% :     0.000002s :     10: predicate.addn_zero_filter
 2.78% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.55% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.47% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.11% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.81% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.80% :     0.000007s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.34% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.74% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.74% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 2.03% :     0.000003s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.04% :     0.000003s :     15: predicate.partial_eliminate
 1.32% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.27% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.27% :     0.000002s :     10: predicate.reshape_eliminate
 1.42% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.11% :     0.000001s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.16% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.46% :     0.000003s :     16: predicate.switch_defer_inline
 1.96% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.89% :     0.000010s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.53% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.83% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.48% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000834    12
43.70% :     0.000365s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.53% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.76% :     0.000415s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.068729    76
 0.09% :     0.000060s :      1: add_recomputation
 0.10% :     0.000071s :      1: auto_monad
 0.03% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.59% :     0.000405s :      1: bootstrap
 0.05% :     0.000032s :      1: cconv
 0.01% :     0.000009s :      1: convert_after_rewriter
 0.03% :     0.000023s :      1: cse_after_recomputation
 0.01% :     0.000009s :      1: environ_conv
 0.04% :     0.000026s :      1: event_method
 0.02% :     0.000015s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000009s :      1: graph_reusing
 9.94% :     0.006835s :      1: jit_opt_a
 0.23% :     0.000159s :      1: jit_opt_after_cconv
 0.08% :     0.000052s :      1: jit_opt_b
 0.66% :     0.000456s :      1: loop_unroll
 1.03% :     0.000709s :      1: mutable_eliminate
 1.40% :     0.000963s :     26: opt.transform.jit_opt_a
 0.07% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000025s :      4: opt.transform.jit_opt_b
 0.02% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.05% :     0.000033s :      4: opt.transform.symbol_engine_opt
 0.74% :     0.000508s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000047s :      1: py_interpret_to_execute
 0.02% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000018s :      1: remove_dup_value
 0.78% :     0.000535s :      1: renormalize.infer
 0.59% :     0.000406s :      1: renormalize.specialize
 0.01% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.06% :     0.000044s :      1: rewriter_after_opt_a
 0.11% :     0.000079s :      1: rewriter_before_opt_a
 0.11% :     0.000077s :      1: symbol_engine_optimizer
51.08% :     0.035108s :      1: task_emit
31.67% :     0.021767s :      1: type_inference
 0.10% :     0.000069s :      1: validate


TotalTime = 0.0331565, [30]
[bootstrap]: 0.0003704
[type_inference]: 0.00552594
[event_method]: 2.159e-05
[auto_monad]: 7.342e-05
[graph_reusing]: 6.86999e-06
[pre_auto_parallel]: 2.39001e-06
[py_interpret_to_execute]: 4.739e-05
[rewriter_before_opt_a]: 7.929e-05
[expand_dump_flag]: 4.05998e-06
[jit_opt_a]: 0.0233977, [2]
    [Cycle 1]: 0.00384421, [27]
        [switch_simplify]: 6.885e-05
        [loop_unroll]: 3.033e-05
        [a_1]: 0.00060918
        [with_stream_mark]: 2.381e-05
        [recompute_prepare]: 7.50998e-06
        [updatestate_depend_eliminate]: 3.93999e-06
        [updatestate_assign_eliminate]: 3.18998e-06
        [updatestate_loads_eliminate]: 3.06001e-06
        [parameter_eliminate]: 2.09e-06
        [specialize_transform]: 6.53e-06
        [updatestate_useless_node_eliminater]: 5.39998e-06
        [accelerated_algorithm]: 6.17999e-06
        [meta_shard_fg_expand]: 2.57001e-06
        [get_grad_eliminate_]: 5.69e-06
        [merge_forward]: 4.1e-06
        [cell_reuse_recompute_pass]: 1.11002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.671e-05
        [j_node_and_user_rematch]: 1.052e-05
        [meta_fg_expand]: 3.23e-06
        [replace_old_param]: 1.059e-05
        [inline_without_move]: 5.91e-06
        [renormalize]: 0.00271473
        [add_forward_monad_depend]: 9.64999e-06
        [auto_monad_grad]: 3.16999e-06
        [auto_monad_eliminator]: 1.83e-05
        [cse]: 3.428e-05
        [replace_applicator]: 2.376e-05
    [Cycle 2]: 0.00037824, [27]
        [switch_simplify]: 6.67002e-06
        [loop_unroll]: 5.59e-06
        [a_1]: 0.00010937
        [with_stream_mark]: 1.442e-05
        [recompute_prepare]: 5.35999e-06
        [updatestate_depend_eliminate]: 3.61001e-06
        [updatestate_assign_eliminate]: 2.62001e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 1.64e-06
        [specialize_transform]: 5.47999e-06
        [updatestate_useless_node_eliminater]: 5.47999e-06
        [accelerated_algorithm]: 6.11998e-06
        [meta_shard_fg_expand]: 2.02999e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 3.75e-06
        [cell_reuse_recompute_pass]: 3.97e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.89e-05
        [j_node_and_user_rematch]: 9.62001e-06
        [meta_fg_expand]: 2.17001e-06
        [replace_old_param]: 8.97e-06
        [inline_without_move]: 4.65001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.09999e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 5.73997e-06
        [cse]: 1.132e-05
        [replace_applicator]: 5.32999e-06
[py_interpret_to_execute_after_opt_a]: 1.361e-05
[rewriter_after_opt_a]: 3.869e-05
[convert_after_rewriter]: 7.99002e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00084864
[jit_opt_b]: 5.737e-05, [1]
    [Cycle 1]: 4.859e-05, [2]
        [frontend_op_eliminate]: 1.91e-05
        [inline_after_opt_a]: 1.681e-05
[cconv]: 2.919e-05
[loop_unroll]: 0.00054355
[jit_opt_after_cconv]: 0.00016554, [1]
    [Cycle 1]: 0.00015868, [11]
        [c_1]: 2.38e-05
        [parameter_eliminate]: 3.83001e-06
        [updatestate_depend_eliminate]: 9.96e-06
        [updatestate_assign_eliminate]: 2.45002e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [cse]: 2.377e-05
        [call_graph_tuple_transform]: 2.299e-05
        [tuple_list_get_item_eliminator]: 5.93002e-06
        [none_parameter_eliminate]: 2.02999e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 6.24001e-06
[remove_dup_value]: 1.582e-05
[partial_unused_args_eliminate]: 2.75002e-06
[environ_conv]: 7.11999e-06
[add_recomputation]: 5.524e-05
[cse_after_recomputation]: 2.331e-05, [1]
    [Cycle 1]: 1.702e-05, [1]
        [cse]: 1.033e-05
[auto_monad_reorder]: 0.0007914
[get_jit_bprop_graph]: 2.69999e-06
[rewriter_after_jit_bprop_graph]: 9.65002e-06
[opt_after_jit_grad]: 0.00064676
[symbol_engine_optimizer]: 9.202e-05, [1]
    [Cycle 1]: 8.29e-05, [6]
        [build]: 6.86001e-06
        [elim_shapecalc]: 8.28001e-06
        [elim_not_effective]: 1.724e-05
        [opt_reshape]: 7.60998e-06
        [fold_const_symbol]: 1.07e-05
        [renormalize]: 9.79984e-07
[validate]: 4.882e-05
Sums
    bootstrap                            :     0.000370s :  2.79%
    type_inference                       :     0.005526s : 41.59%
    event_method                         :     0.000022s :  0.16%
    auto_monad                           :     0.000073s :  0.55%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000047s :  0.36%
    rewriter_before_opt_a                :     0.000079s :  0.60%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000076s :  0.57%
    jit_opt_a.loop_unroll                :     0.000036s :  0.27%
    jit_opt_a.a_1                        :     0.000719s :  5.41%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.08%
    jit_opt_a.merge_forward              :     0.000008s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000020s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.002715s : 20.43%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.18%
    jit_opt_a.cse                        :     0.000046s :  0.34%
    jit_opt_a.replace_applicator         :     0.000029s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.10%
    rewriter_after_opt_a                 :     0.000039s :  0.29%
    convert_after_rewriter               :     0.000008s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000849s :  6.39%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.13%
    cconv                                :     0.000029s :  0.22%
    loop_unroll                          :     0.000544s :  4.09%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000016s :  0.12%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000007s :  0.05%
    add_recomputation                    :     0.000055s :  0.42%
    cse_after_recomputation.cse          :     0.000010s :  0.08%
    auto_monad_reorder                   :     0.000791s :  5.96%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.07%
    opt_after_jit_grad                   :     0.000647s :  4.87%
    symbol_engine_optimizer.build        :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000049s :  0.37%


Time group info:
------[substitution.]   0.000251    23
 1.02% :     0.000003s :      2: substitution.elim_not_effective
 0.69% :     0.000002s :      2: substitution.fold_const_symbol
 2.58% :     0.000006s :      3: substitution.graph_param_transform
83.74% :     0.000210s :      4: substitution.inline
 1.74% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.44% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.54% :     0.000006s :      2: substitution.replace_old_param
 5.26% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005442     2
81.23% :     0.004421s :      1: type_inference.infer
18.77% :     0.001021s :      1: type_inference.specialize
------[replace.]   0.000080     6
67.96% :     0.000055s :      4: replace.inline
32.04% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000219     6
94.49% :     0.000207s :      4: match.inline
 5.51% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   691
 1.24% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.02% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.87% :     0.000001s :     10: predicate.addn_check_dump
 1.74% :     0.000002s :     10: predicate.addn_zero_filter
 2.47% :     0.000003s :     10: predicate.arithmetic_simplify
 1.25% :     0.000002s :     10: predicate.cast_eliminate
 0.74% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.91% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.26% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.45% :     0.000001s :      3: predicate.elim_not_effective
 0.43% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.39% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.44% :     0.000002s :     10: predicate.environ_get_eliminate
 1.13% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000001s :      3: predicate.graph_param_transform
 5.71% :     0.000008s :     22: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.68% :     0.000002s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.52% :     0.000002s :     15: predicate.load_eliminater
 1.49% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.55% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.32% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.92% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 1.80% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 1.95% :     0.000003s :     15: predicate.partial_eliminate
 1.33% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000003s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.39% :     0.000001s :      3: predicate.reset_defer_inline
 1.28% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.21% :     0.000002s :      6: predicate.special_op_eliminate
 1.16% :     0.000002s :      6: predicate.specialize_transform
 1.57% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.21% :     0.000003s :     16: predicate.switch_defer_inline
 1.72% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.11% :     0.000010s :     53: predicate.switch_simplify
 1.20% :     0.000002s :     10: predicate.tile_eliminate
 1.20% :     0.000002s :     10: predicate.transpose_eliminate
 1.52% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.29% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.44% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000003s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000784    12
48.65% :     0.000381s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.84% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.51% :     0.000365s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.036890    72
 0.16% :     0.000058s :      1: add_recomputation
 0.21% :     0.000077s :      1: auto_monad
 2.17% :     0.000801s :      1: auto_monad_reorder
 1.06% :     0.000392s :      1: bootstrap
 0.09% :     0.000032s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.07% :     0.000025s :      1: event_method
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000006s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
63.44% :     0.023402s :      1: jit_opt_a
 0.46% :     0.000169s :      1: jit_opt_after_cconv
 0.16% :     0.000060s :      1: jit_opt_b
 1.50% :     0.000553s :      1: loop_unroll
 2.33% :     0.000859s :      1: mutable_eliminate
 2.62% :     0.000968s :     26: opt.transform.jit_opt_a
 0.15% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000028s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.78% :     0.000658s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000050s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 5.98% :     0.002207s :      1: renormalize.infer
 1.34% :     0.000494s :      1: renormalize.specialize
 0.03% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000042s :      1: rewriter_after_opt_a
 0.22% :     0.000082s :      1: rewriter_before_opt_a
 0.26% :     0.000095s :      1: symbol_engine_optimizer
15.04% :     0.005549s :      1: type_inference


TotalTime = 0.0342822, [33]
[bootstrap]: 0.00042971
[type_inference]: 0.00585606
[event_method]: 2.11e-05
[auto_monad]: 6.682e-05
[graph_reusing]: 6.23e-06
[pre_auto_parallel]: 2.43e-06
[py_interpret_to_execute]: 5.499e-05
[rewriter_before_opt_a]: 8.365e-05
[expand_dump_flag]: 4.21001e-06
[jit_opt_a]: 0.0069451, [2]
    [Cycle 1]: 0.0022363, [27]
        [switch_simplify]: 6.468e-05
        [loop_unroll]: 2.908e-05
        [a_1]: 0.0006337
        [with_stream_mark]: 2.612e-05
        [recompute_prepare]: 8.77e-06
        [updatestate_depend_eliminate]: 4.87e-06
        [updatestate_assign_eliminate]: 3.38e-06
        [updatestate_loads_eliminate]: 2.91e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 7.75998e-06
        [updatestate_useless_node_eliminater]: 5.84e-06
        [accelerated_algorithm]: 6.48003e-06
        [meta_shard_fg_expand]: 2.68e-06
        [get_grad_eliminate_]: 6.23998e-06
        [merge_forward]: 4.1e-06
        [cell_reuse_recompute_pass]: 2.59001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.655e-05
        [j_node_and_user_rematch]: 1.055e-05
        [meta_fg_expand]: 2.51e-06
        [replace_old_param]: 1.071e-05
        [inline_without_move]: 6.60997e-06
        [renormalize]: 0.00110128
        [add_forward_monad_depend]: 7.41999e-06
        [auto_monad_grad]: 2.69001e-06
        [auto_monad_eliminator]: 2.008e-05
        [cse]: 3.546e-05
        [replace_applicator]: 1.605e-05
    [Cycle 2]: 0.00040072, [27]
        [switch_simplify]: 6.61999e-06
        [loop_unroll]: 5.88002e-06
        [a_1]: 0.0001043
        [with_stream_mark]: 1.018e-05
        [recompute_prepare]: 5.71e-06
        [updatestate_depend_eliminate]: 2.78998e-06
        [updatestate_assign_eliminate]: 2.52001e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 6.05002e-06
        [updatestate_useless_node_eliminater]: 5.40001e-06
        [accelerated_algorithm]: 6.53e-06
        [meta_shard_fg_expand]: 2.26998e-06
        [get_grad_eliminate_]: 4.79e-06
        [merge_forward]: 3.35e-06
        [cell_reuse_recompute_pass]: 2.87002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.652e-05
        [j_node_and_user_rematch]: 8.47998e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 8.45999e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.74998e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 6.33e-06
        [cse]: 1.337e-05
        [replace_applicator]: 5.55001e-06
[py_interpret_to_execute_after_opt_a]: 1.566e-05
[rewriter_after_opt_a]: 4.25e-05
[convert_after_rewriter]: 7.82e-06
[order_py_execute_after_rewriter]: 5.15999e-06
[mutable_eliminate]: 0.00086226
[jit_opt_b]: 5.562e-05, [1]
    [Cycle 1]: 4.703e-05, [2]
        [frontend_op_eliminate]: 1.848e-05
        [inline_after_opt_a]: 1.599e-05
[cconv]: 3.013e-05
[loop_unroll]: 0.00045923
[jit_opt_after_cconv]: 0.00015125, [1]
    [Cycle 1]: 0.00014382, [11]
        [c_1]: 2.233e-05
        [parameter_eliminate]: 2.78998e-06
        [updatestate_depend_eliminate]: 5.79e-06
        [updatestate_assign_eliminate]: 3.01999e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 2.541e-05
        [call_graph_tuple_transform]: 1.948e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 5.15001e-06
[remove_dup_value]: 1.681e-05
[partial_unused_args_eliminate]: 3.06001e-06
[environ_conv]: 6.00002e-06
[add_recomputation]: 5.497e-05
[cse_after_recomputation]: 2.187e-05, [1]
    [Cycle 1]: 1.55e-05, [1]
        [cse]: 9.56998e-06
[auto_monad_reorder]: 1.642e-05
[get_jit_bprop_graph]: 2.41e-06
[rewriter_after_jit_bprop_graph]: 7.17002e-06
[opt_after_jit_grad]: 0.00048964
[symbol_engine_optimizer]: 7.914e-05, [1]
    [Cycle 1]: 7.135e-05, [6]
        [build]: 4.83001e-06
        [elim_shapecalc]: 8.17e-06
        [elim_not_effective]: 1.301e-05
        [opt_reshape]: 6.51e-06
        [fold_const_symbol]: 9.82001e-06
        [renormalize]: 4.30009e-07
[validate]: 4.53e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.0181656
[execute]: 1.116e-05
Sums
    bootstrap                            :     0.000430s :  1.47%
    type_inference                       :     0.005856s : 20.04%
    event_method                         :     0.000021s :  0.07%
    auto_monad                           :     0.000067s :  0.23%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000055s :  0.19%
    rewriter_before_opt_a                :     0.000084s :  0.29%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000071s :  0.24%
    jit_opt_a.loop_unroll                :     0.000035s :  0.12%
    jit_opt_a.a_1                        :     0.000738s :  2.52%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000014s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000019s :  0.07%
    jit_opt_a.inline_without_move        :     0.000011s :  0.04%
    jit_opt_a.renormalize                :     0.001101s :  3.77%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.09%
    jit_opt_a.cse                        :     0.000049s :  0.17%
    jit_opt_a.replace_applicator         :     0.000022s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.05%
    rewriter_after_opt_a                 :     0.000043s :  0.15%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000862s :  2.95%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.05%
    cconv                                :     0.000030s :  0.10%
    loop_unroll                          :     0.000459s :  1.57%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000017s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000055s :  0.19%
    cse_after_recomputation.cse          :     0.000010s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000490s :  1.68%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000045s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.018166s : 62.15%
    execute                              :     0.000011s :  0.04%


Time group info:
------[substitution.]   0.000276    23
 0.65% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000002s :      2: substitution.fold_const_symbol
 2.05% :     0.000006s :      3: substitution.graph_param_transform
84.84% :     0.000234s :      4: substitution.inline
 1.65% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.24% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.30% :     0.000006s :      2: substitution.replace_old_param
 5.54% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005762     2
81.20% :     0.004679s :      1: type_inference.infer
18.80% :     0.001084s :      1: type_inference.specialize
------[replace.]   0.000082     6
65.90% :     0.000054s :      4: replace.inline
34.10% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000246     6
94.25% :     0.000232s :      4: match.inline
 5.75% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.10% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.21% :     0.000002s :     10: predicate.addn_zero_filter
 2.61% :     0.000003s :     10: predicate.arithmetic_simplify
 1.09% :     0.000001s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.90% :     0.000001s :     10: predicate.compare_switch_simplify
 1.19% :     0.000002s :     10: predicate.depend_value_elim
 1.28% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.57% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.01% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.74% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.88% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.75% :     0.000007s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.45% :     0.000002s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.57% :     0.000002s :     15: predicate.load_eliminater
 1.14% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.89% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.28% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 2.62% :     0.000003s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     15: predicate.partial_eliminate
 1.02% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.45% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.97% :     0.000001s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 1.16% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.67% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.93% :     0.000004s :     16: predicate.switch_defer_inline
 1.86% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.95% :     0.000009s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.31% :     0.000002s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.08% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.43% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 2.78% :     0.000004s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.57% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000761    12
43.69% :     0.000333s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.73% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.58% :     0.000370s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.036465    76
 0.16% :     0.000059s :      1: add_recomputation
 0.19% :     0.000071s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.26% :     0.000458s :      1: bootstrap
 0.09% :     0.000033s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.07% :     0.000026s :      1: event_method
 0.05% :     0.000017s :      1: execute
 0.02% :     0.000007s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
19.06% :     0.006950s :      1: jit_opt_a
 0.42% :     0.000154s :      1: jit_opt_after_cconv
 0.16% :     0.000059s :      1: jit_opt_b
 1.29% :     0.000469s :      1: loop_unroll
 2.40% :     0.000874s :      1: mutable_eliminate
 2.69% :     0.000982s :     26: opt.transform.jit_opt_a
 0.13% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000028s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.37% :     0.000498s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000059s :      1: py_interpret_to_execute
 0.05% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 1.64% :     0.000599s :      1: renormalize.infer
 1.34% :     0.000490s :      1: renormalize.specialize
 0.03% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000046s :      1: rewriter_after_opt_a
 0.24% :     0.000088s :      1: rewriter_before_opt_a
 0.22% :     0.000082s :      1: symbol_engine_optimizer
49.88% :     0.018187s :      1: task_emit
16.12% :     0.005880s :      1: type_inference
 0.18% :     0.000067s :      1: validate


TotalTime = 0.0149978, [30]
[bootstrap]: 0.00056558
[type_inference]: 0.00533623
[event_method]: 1.762e-05
[auto_monad]: 6.702e-05
[graph_reusing]: 5.74e-06
[pre_auto_parallel]: 2.40002e-06
[py_interpret_to_execute]: 3.557e-05
[rewriter_before_opt_a]: 7.122e-05
[expand_dump_flag]: 3.31001e-06
[jit_opt_a]: 0.00632285, [2]
    [Cycle 1]: 0.00211365, [27]
        [switch_simplify]: 5.615e-05
        [loop_unroll]: 3.251e-05
        [a_1]: 0.00071333
        [with_stream_mark]: 2.215e-05
        [recompute_prepare]: 8.35999e-06
        [updatestate_depend_eliminate]: 3.71001e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 1.87001e-06
        [specialize_transform]: 6.21998e-06
        [updatestate_useless_node_eliminater]: 5.60001e-06
        [accelerated_algorithm]: 6.29999e-06
        [meta_shard_fg_expand]: 2.55997e-06
        [get_grad_eliminate_]: 5.78002e-06
        [merge_forward]: 3.76001e-06
        [cell_reuse_recompute_pass]: 1.47999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.455e-05
        [j_node_and_user_rematch]: 1.01e-05
        [meta_fg_expand]: 2.35002e-06
        [replace_old_param]: 9.59999e-06
        [inline_without_move]: 6.44999e-06
        [renormalize]: 0.00094459
        [add_forward_monad_depend]: 7.33999e-06
        [auto_monad_grad]: 3.18e-06
        [auto_monad_eliminator]: 1.609e-05
        [cse]: 2.985e-05
        [replace_applicator]: 1.685e-05
    [Cycle 2]: 0.00035231, [27]
        [switch_simplify]: 5.56998e-06
        [loop_unroll]: 5.21998e-06
        [a_1]: 9.871e-05
        [with_stream_mark]: 1.454e-05
        [recompute_prepare]: 6.06998e-06
        [updatestate_depend_eliminate]: 2.88e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [parameter_eliminate]: 1.97999e-06
        [specialize_transform]: 5.60001e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 5.40001e-06
        [meta_shard_fg_expand]: 1.15999e-06
        [get_grad_eliminate_]: 4.95001e-06
        [merge_forward]: 3.2e-06
        [cell_reuse_recompute_pass]: 3.41001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.549e-05
        [j_node_and_user_rematch]: 9.22001e-06
        [meta_fg_expand]: 1.81e-06
        [replace_old_param]: 7.77e-06
        [inline_without_move]: 4.75999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.97999e-06
        [auto_monad_grad]: 9.40025e-07
        [auto_monad_eliminator]: 5.74999e-06
        [cse]: 1.178e-05
        [replace_applicator]: 5.06002e-06
[py_interpret_to_execute_after_opt_a]: 1.448e-05
[rewriter_after_opt_a]: 3.953e-05
[convert_after_rewriter]: 7.42998e-06
[order_py_execute_after_rewriter]: 5.29e-06
[mutable_eliminate]: 0.00081243
[jit_opt_b]: 5.951e-05, [1]
    [Cycle 1]: 5.094e-05, [2]
        [frontend_op_eliminate]: 1.943e-05
        [inline_after_opt_a]: 1.918e-05
[cconv]: 3.254e-05
[loop_unroll]: 0.0004782
[jit_opt_after_cconv]: 0.00015877, [1]
    [Cycle 1]: 0.00015167, [11]
        [c_1]: 2.224e-05
        [parameter_eliminate]: 3.68e-06
        [updatestate_depend_eliminate]: 6.76999e-06
        [updatestate_assign_eliminate]: 3.02002e-06
        [updatestate_loads_eliminate]: 2.64999e-06
        [cse]: 2.265e-05
        [call_graph_tuple_transform]: 2.295e-05
        [tuple_list_get_item_eliminator]: 5.98998e-06
        [none_parameter_eliminate]: 1.79998e-06
        [renormalize]: 8.2e-07
        [switch_simplify]: 5.56e-06
[remove_dup_value]: 1.592e-05
[partial_unused_args_eliminate]: 2.44001e-06
[environ_conv]: 5.86e-06
[add_recomputation]: 5.317e-05
[cse_after_recomputation]: 2.174e-05, [1]
    [Cycle 1]: 1.58e-05, [1]
        [cse]: 1.016e-05
[auto_monad_reorder]: 1.591e-05
[get_jit_bprop_graph]: 2.36e-06
[rewriter_after_jit_bprop_graph]: 7.48999e-06
[opt_after_jit_grad]: 0.00052068
[symbol_engine_optimizer]: 8.03e-05, [1]
    [Cycle 1]: 7.306e-05, [6]
        [build]: 4.47e-06
        [elim_shapecalc]: 8.40001e-06
        [elim_not_effective]: 1.453e-05
        [opt_reshape]: 7.03e-06
        [fold_const_symbol]: 9.51e-06
        [renormalize]: 4.7998e-07
[validate]: 4.414e-05
Sums
    bootstrap                            :     0.000566s :  5.38%
    type_inference                       :     0.005336s : 50.71%
    event_method                         :     0.000018s :  0.17%
    auto_monad                           :     0.000067s :  0.64%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.34%
    rewriter_before_opt_a                :     0.000071s :  0.68%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000062s :  0.59%
    jit_opt_a.loop_unroll                :     0.000038s :  0.36%
    jit_opt_a.a_1                        :     0.000812s :  7.72%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.17%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000945s :  8.98%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.21%
    jit_opt_a.cse                        :     0.000042s :  0.40%
    jit_opt_a.replace_applicator         :     0.000022s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.14%
    rewriter_after_opt_a                 :     0.000040s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000812s :  7.72%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.18%
    cconv                                :     0.000033s :  0.31%
    loop_unroll                          :     0.000478s :  4.54%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000023s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000016s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000053s :  0.51%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000521s :  4.95%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000044s :  0.42%


Time group info:
------[substitution.]   0.000245    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000002s :      2: substitution.fold_const_symbol
 2.60% :     0.000006s :      3: substitution.graph_param_transform
84.22% :     0.000206s :      4: substitution.inline
 1.75% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.98% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.83% :     0.000004s :      2: substitution.replace_old_param
 5.88% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005266     2
82.58% :     0.004349s :      1: type_inference.infer
17.42% :     0.000917s :      1: type_inference.specialize
------[replace.]   0.000081     6
63.02% :     0.000051s :      4: replace.inline
36.98% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000217     6
93.95% :     0.000204s :      4: match.inline
 6.05% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.29% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.51% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 2.12% :     0.000003s :     10: predicate.addn_zero_filter
 2.50% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.58% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.32% :     0.000007s :     22: predicate.inline
 1.37% :     0.000002s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.58% :     0.000002s :     15: predicate.load_eliminater
 1.54% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 2.00% :     0.000003s :      3: predicate.mutable_eliminate
 0.71% :     0.000001s :      3: predicate.opt_reshape
 1.97% :     0.000003s :     15: predicate.partial_eliminate
 1.57% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.77% :     0.000002s :     10: predicate.reduce_eliminate
 1.36% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.64% :     0.000002s :     10: predicate.reshape_eliminate
 1.07% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.20% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.22% :     0.000002s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.59% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.90% :     0.000009s :     53: predicate.switch_simplify
 1.37% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.67% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.06% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.64% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000711    12
50.24% :     0.000357s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.66% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.11% :     0.000307s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017042    72
 0.33% :     0.000057s :      1: add_recomputation
 0.42% :     0.000071s :      1: auto_monad
 0.11% :     0.000019s :      1: auto_monad_reorder
 3.44% :     0.000587s :      1: bootstrap
 0.21% :     0.000036s :      1: cconv
 0.07% :     0.000012s :      1: convert_after_rewriter
 0.14% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.13% :     0.000021s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
37.13% :     0.006327s :      1: jit_opt_a
 0.95% :     0.000162s :      1: jit_opt_after_cconv
 0.37% :     0.000063s :      1: jit_opt_b
 2.86% :     0.000487s :      1: loop_unroll
 4.83% :     0.000822s :      1: mutable_eliminate
 6.11% :     0.001041s :     26: opt.transform.jit_opt_a
 0.31% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000031s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.11% :     0.000530s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000038s :      1: py_interpret_to_execute
 0.10% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.99% :     0.000510s :      1: renormalize.infer
 2.50% :     0.000426s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000043s :      1: rewriter_after_opt_a
 0.44% :     0.000074s :      1: rewriter_before_opt_a
 0.49% :     0.000083s :      1: symbol_engine_optimizer
31.38% :     0.005349s :      1: type_inference


TotalTime = 0.0271104, [33]
[bootstrap]: 0.00031855
[type_inference]: 0.00798881
[event_method]: 1.828e-05
[auto_monad]: 6.733e-05
[graph_reusing]: 5.97999e-06
[pre_auto_parallel]: 2.58e-06
[py_interpret_to_execute]: 4.393e-05
[rewriter_before_opt_a]: 7.647e-05
[expand_dump_flag]: 3.8e-06
[jit_opt_a]: 0.00624697, [2]
    [Cycle 1]: 0.00188133, [27]
        [switch_simplify]: 6.42e-05
        [loop_unroll]: 3.021e-05
        [a_1]: 0.00058179
        [with_stream_mark]: 1.88e-05
        [recompute_prepare]: 7.78001e-06
        [updatestate_depend_eliminate]: 3.64002e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.59001e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 5.95002e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 5.46998e-06
        [meta_shard_fg_expand]: 2.31e-06
        [get_grad_eliminate_]: 5.69e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 1.44998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.56e-05
        [j_node_and_user_rematch]: 8.59e-06
        [meta_fg_expand]: 2.29999e-06
        [replace_old_param]: 1.006e-05
        [inline_without_move]: 5.32999e-06
        [renormalize]: 0.00084316
        [add_forward_monad_depend]: 7.07002e-06
        [auto_monad_grad]: 2.84999e-06
        [auto_monad_eliminator]: 1.515e-05
        [cse]: 3.046e-05
        [replace_applicator]: 1.339e-05
    [Cycle 2]: 0.00031857, [27]
        [switch_simplify]: 5.48002e-06
        [loop_unroll]: 4.97999e-06
        [a_1]: 8.938e-05
        [with_stream_mark]: 1.025e-05
        [recompute_prepare]: 4.85999e-06
        [updatestate_depend_eliminate]: 2.69001e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [parameter_eliminate]: 1.11002e-06
        [specialize_transform]: 4.70001e-06
        [updatestate_useless_node_eliminater]: 4.81002e-06
        [accelerated_algorithm]: 4.77e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.45e-06
        [merge_forward]: 2.61999e-06
        [cell_reuse_recompute_pass]: 1.71998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.218e-05
        [j_node_and_user_rematch]: 7.55e-06
        [meta_fg_expand]: 1.57999e-06
        [replace_old_param]: 7.4e-06
        [inline_without_move]: 4.87998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 5.39e-06
        [cse]: 1.023e-05
        [replace_applicator]: 5.10999e-06
[py_interpret_to_execute_after_opt_a]: 1.217e-05
[rewriter_after_opt_a]: 3.82e-05
[convert_after_rewriter]: 7.23999e-06
[order_py_execute_after_rewriter]: 5.59998e-06
[mutable_eliminate]: 0.00205949
[jit_opt_b]: 5.925e-05, [1]
    [Cycle 1]: 5.017e-05, [2]
        [frontend_op_eliminate]: 1.968e-05
        [inline_after_opt_a]: 1.782e-05
[cconv]: 3.485e-05
[loop_unroll]: 0.000473
[jit_opt_after_cconv]: 0.00016472, [1]
    [Cycle 1]: 0.00015768, [11]
        [c_1]: 2.164e-05
        [parameter_eliminate]: 5.12999e-06
        [updatestate_depend_eliminate]: 9.40001e-06
        [updatestate_assign_eliminate]: 3.51999e-06
        [updatestate_loads_eliminate]: 2.88e-06
        [cse]: 2.85e-05
        [call_graph_tuple_transform]: 2.221e-05
        [tuple_list_get_item_eliminator]: 5.57001e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 8.59989e-07
        [switch_simplify]: 5.24e-06
[remove_dup_value]: 1.565e-05
[partial_unused_args_eliminate]: 2.24001e-06
[environ_conv]: 5.72001e-06
[add_recomputation]: 5.455e-05
[cse_after_recomputation]: 2.165e-05, [1]
    [Cycle 1]: 1.522e-05, [1]
        [cse]: 9.20999e-06
[auto_monad_reorder]: 1.67e-05
[get_jit_bprop_graph]: 1.79e-06
[rewriter_after_jit_bprop_graph]: 7.33999e-06
[opt_after_jit_grad]: 0.00046669
[symbol_engine_optimizer]: 7.935e-05, [1]
    [Cycle 1]: 7.21e-05, [6]
        [build]: 4.89e-06
        [elim_shapecalc]: 8.25e-06
        [elim_not_effective]: 1.365e-05
        [opt_reshape]: 6.51999e-06
        [fold_const_symbol]: 9.82999e-06
        [renormalize]: 6.09987e-07
[validate]: 4.776e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00851888
[execute]: 9.94001e-06
Sums
    bootstrap                            :     0.000319s :  1.42%
    type_inference                       :     0.007989s : 35.66%
    event_method                         :     0.000018s :  0.08%
    auto_monad                           :     0.000067s :  0.30%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.20%
    rewriter_before_opt_a                :     0.000076s :  0.34%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000070s :  0.31%
    jit_opt_a.loop_unroll                :     0.000035s :  0.16%
    jit_opt_a.a_1                        :     0.000671s :  3.00%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000843s :  3.76%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.09%
    jit_opt_a.cse                        :     0.000041s :  0.18%
    jit_opt_a.replace_applicator         :     0.000018s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.05%
    rewriter_after_opt_a                 :     0.000038s :  0.17%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.002059s :  9.19%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.08%
    cconv                                :     0.000035s :  0.16%
    loop_unroll                          :     0.000473s :  2.11%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000016s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000055s :  0.24%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000017s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.03%
    opt_after_jit_grad                   :     0.000467s :  2.08%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000048s :  0.21%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008519s : 38.03%
    execute                              :     0.000010s :  0.04%


Time group info:
------[substitution.]   0.000226    23
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.77% :     0.000006s :      3: substitution.graph_param_transform
83.97% :     0.000190s :      4: substitution.inline
 1.13% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.13% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.33% :     0.000005s :      2: substitution.replace_old_param
 6.15% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007917     2
59.05% :     0.004675s :      1: type_inference.infer
40.95% :     0.003242s :      1: type_inference.specialize
------[replace.]   0.000077     6
67.64% :     0.000052s :      4: replace.inline
32.36% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     6
93.62% :     0.000187s :      4: match.inline
 6.38% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.65% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.06% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 2.03% :     0.000002s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.47% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.92% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.25% :     0.000006s :     22: predicate.inline
 1.04% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.02% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.98% :     0.000002s :     15: predicate.load_eliminater
 1.29% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 3.21% :     0.000004s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000002s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.28% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 2.44% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.58% :     0.000009s :     53: predicate.switch_simplify
 1.33% :     0.000002s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.47% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.44% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.81% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001326    12
70.49% :     0.000935s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 2.42% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
27.09% :     0.000359s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028958    76
 0.20% :     0.000057s :      1: add_recomputation
 0.24% :     0.000070s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.17% :     0.000339s :      1: bootstrap
 0.13% :     0.000037s :      1: cconv
 0.04% :     0.000012s :      1: convert_after_rewriter
 0.08% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000022s :      1: event_method
 0.05% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
21.59% :     0.006251s :      1: jit_opt_a
 0.58% :     0.000167s :      1: jit_opt_after_cconv
 0.21% :     0.000062s :      1: jit_opt_b
 1.66% :     0.000481s :      1: loop_unroll
 7.17% :     0.002076s :      1: mutable_eliminate
 3.08% :     0.000891s :     26: opt.transform.jit_opt_a
 0.18% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000029s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000023s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.64% :     0.000476s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000047s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000018s :      1: remove_dup_value
 1.61% :     0.000465s :      1: renormalize.infer
 1.28% :     0.000369s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000041s :      1: rewriter_after_opt_a
 0.28% :     0.000080s :      1: rewriter_before_opt_a
 0.28% :     0.000082s :      1: symbol_engine_optimizer
29.49% :     0.008539s :      1: task_emit
27.64% :     0.008005s :      1: type_inference
 0.24% :     0.000071s :      1: validate


TotalTime = 0.0358859, [33]
[bootstrap]: 0.00036733
[type_inference]: 0.00659507
[event_method]: 1.877e-05
[auto_monad]: 6.615e-05
[graph_reusing]: 5.89e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 3.897e-05
[rewriter_before_opt_a]: 7.653e-05
[expand_dump_flag]: 3.38999e-06
[jit_opt_a]: 0.00714333, [2]
    [Cycle 1]: 0.00296908, [27]
        [switch_simplify]: 6.18e-05
        [loop_unroll]: 2.863e-05
        [a_1]: 0.00057276
        [with_stream_mark]: 2.247e-05
        [recompute_prepare]: 6.80998e-06
        [updatestate_depend_eliminate]: 3.74002e-06
        [updatestate_assign_eliminate]: 3.15002e-06
        [updatestate_loads_eliminate]: 2.91999e-06
        [parameter_eliminate]: 1.92999e-06
        [specialize_transform]: 6.24001e-06
        [updatestate_useless_node_eliminater]: 5.47999e-06
        [accelerated_algorithm]: 5.18002e-06
        [meta_shard_fg_expand]: 1.95001e-06
        [get_grad_eliminate_]: 5.32001e-06
        [merge_forward]: 3.83001e-06
        [cell_reuse_recompute_pass]: 1.03001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.426e-05
        [j_node_and_user_rematch]: 9.29998e-06
        [meta_fg_expand]: 2.20002e-06
        [replace_old_param]: 9.82001e-06
        [inline_without_move]: 5.35001e-06
        [renormalize]: 0.00195101
        [add_forward_monad_depend]: 6.24001e-06
        [auto_monad_grad]: 2.51e-06
        [auto_monad_eliminator]: 1.555e-05
        [cse]: 2.968e-05
        [replace_applicator]: 1.45e-05
    [Cycle 2]: 0.0003187, [27]
        [switch_simplify]: 6.09001e-06
        [loop_unroll]: 4.73001e-06
        [a_1]: 8.973e-05
        [with_stream_mark]: 1.03e-05
        [recompute_prepare]: 5.24e-06
        [updatestate_depend_eliminate]: 2.67001e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.71e-06
        [parameter_eliminate]: 1.47001e-06
        [specialize_transform]: 5.25001e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.14998e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 4.32e-06
        [merge_forward]: 3.39001e-06
        [cell_reuse_recompute_pass]: 2.29001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.247e-05
        [j_node_and_user_rematch]: 7.46001e-06
        [meta_fg_expand]: 1.69998e-06
        [replace_old_param]: 7.24001e-06
        [inline_without_move]: 4.58999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.05001e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 4.45999e-06
        [cse]: 9.43997e-06
        [replace_applicator]: 5.10999e-06
[py_interpret_to_execute_after_opt_a]: 1.269e-05
[rewriter_after_opt_a]: 3.901e-05
[convert_after_rewriter]: 7.61999e-06
[order_py_execute_after_rewriter]: 4.94e-06
[mutable_eliminate]: 0.0006044
[jit_opt_b]: 4.99e-05, [1]
    [Cycle 1]: 4.229e-05, [2]
        [frontend_op_eliminate]: 1.641e-05
        [inline_after_opt_a]: 1.398e-05
[cconv]: 2.572e-05
[loop_unroll]: 0.00043259
[jit_opt_after_cconv]: 0.00013752, [1]
    [Cycle 1]: 0.00013075, [11]
        [c_1]: 2.055e-05
        [parameter_eliminate]: 2.61e-06
        [updatestate_depend_eliminate]: 5.63002e-06
        [updatestate_assign_eliminate]: 2.33002e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.825e-05
        [call_graph_tuple_transform]: 1.833e-05
        [tuple_list_get_item_eliminator]: 5.40999e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 9.70002e-07
        [switch_simplify]: 5.13002e-06
[remove_dup_value]: 1.532e-05
[partial_unused_args_eliminate]: 2.76999e-06
[environ_conv]: 4.92999e-06
[add_recomputation]: 5.364e-05
[cse_after_recomputation]: 1.98e-05, [1]
    [Cycle 1]: 1.424e-05, [1]
        [cse]: 8.67998e-06
[auto_monad_reorder]: 1.521e-05
[get_jit_bprop_graph]: 2.56998e-06
[rewriter_after_jit_bprop_graph]: 6.17001e-06
[opt_after_jit_grad]: 0.00045441
[symbol_engine_optimizer]: 7.501e-05, [1]
    [Cycle 1]: 6.821e-05, [6]
        [build]: 5.14998e-06
        [elim_shapecalc]: 7.3e-06
        [elim_not_effective]: 1.325e-05
        [opt_reshape]: 5.94e-06
        [fold_const_symbol]: 8.48999e-06
        [renormalize]: 4.40021e-07
[validate]: 3.974e-05
[backend_pass]: 1.00001e-06
[task_emit]: 0.019324
[execute]: 1.12e-05
Sums
    bootstrap                            :     0.000367s :  1.17%
    type_inference                       :     0.006595s : 21.01%
    event_method                         :     0.000019s :  0.06%
    auto_monad                           :     0.000066s :  0.21%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.12%
    rewriter_before_opt_a                :     0.000077s :  0.24%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000068s :  0.22%
    jit_opt_a.loop_unroll                :     0.000033s :  0.11%
    jit_opt_a.a_1                        :     0.000662s :  2.11%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000017s :  0.05%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.001951s :  6.21%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.06%
    jit_opt_a.cse                        :     0.000039s :  0.12%
    jit_opt_a.replace_applicator         :     0.000020s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.04%
    rewriter_after_opt_a                 :     0.000039s :  0.12%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000604s :  1.93%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.04%
    cconv                                :     0.000026s :  0.08%
    loop_unroll                          :     0.000433s :  1.38%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000015s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000054s :  0.17%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000015s :  0.05%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000454s :  1.45%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000040s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.019324s : 61.55%
    execute                              :     0.000011s :  0.04%


Time group info:
------[substitution.]   0.000239    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.10% :     0.000005s :      3: substitution.graph_param_transform
85.86% :     0.000205s :      4: substitution.inline
 1.17% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.05% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.98% :     0.000005s :      2: substitution.replace_old_param
 5.44% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006516     2
85.52% :     0.005573s :      1: type_inference.infer
14.48% :     0.000943s :      1: type_inference.specialize
------[replace.]   0.000073     6
68.47% :     0.000050s :      4: replace.inline
31.53% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000214     6
94.53% :     0.000203s :      4: match.inline
 5.47% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.44% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.18% :     0.000001s :     10: predicate.addn_zero_filter
 2.04% :     0.000002s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000001s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.47% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.65% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.33% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000002s :     15: predicate.partial_eliminate
 1.42% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000001s :     10: predicate.reshape_eliminate
 1.50% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.47% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.30% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.35% :     0.000008s :     53: predicate.switch_simplify
 1.40% :     0.000002s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.25% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000731    12
49.42% :     0.000361s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.75% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.83% :     0.000320s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038801    76
 0.15% :     0.000056s :      1: add_recomputation
 0.18% :     0.000069s :      1: auto_monad
 0.05% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.00% :     0.000388s :      1: bootstrap
 0.07% :     0.000028s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.06% :     0.000022s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.06% :     0.000023s :      1: event_method
 0.05% :     0.000018s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
18.42% :     0.007147s :      1: jit_opt_a
 0.36% :     0.000140s :      1: jit_opt_after_cconv
 0.14% :     0.000052s :      1: jit_opt_b
 1.14% :     0.000441s :      1: loop_unroll
 1.58% :     0.000614s :      1: mutable_eliminate
 2.26% :     0.000877s :     26: opt.transform.jit_opt_a
 0.12% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000024s :      4: opt.transform.jit_opt_b
 0.03% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.19% :     0.000462s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.11% :     0.000042s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 3.97% :     0.001540s :      1: renormalize.infer
 1.03% :     0.000401s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000042s :      1: rewriter_after_opt_a
 0.20% :     0.000079s :      1: rewriter_before_opt_a
 0.20% :     0.000078s :      1: symbol_engine_optimizer
49.86% :     0.019347s :      1: task_emit
17.04% :     0.006612s :      1: type_inference
 0.15% :     0.000060s :      1: validate


TotalTime = 0.0414066, [33]
[bootstrap]: 0.00036434
[type_inference]: 0.0199915
[event_method]: 2.108e-05
[auto_monad]: 7.045e-05
[graph_reusing]: 6.21998e-06
[pre_auto_parallel]: 2.76e-06
[py_interpret_to_execute]: 4.746e-05
[rewriter_before_opt_a]: 7.798e-05
[expand_dump_flag]: 3.61001e-06
[jit_opt_a]: 0.00720881, [2]
    [Cycle 1]: 0.00216066, [27]
        [switch_simplify]: 7.402e-05
        [loop_unroll]: 2.959e-05
        [a_1]: 0.00064166
        [with_stream_mark]: 2.552e-05
        [recompute_prepare]: 8.23001e-06
        [updatestate_depend_eliminate]: 4.52e-06
        [updatestate_assign_eliminate]: 2.98998e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 6.33e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 6.06e-06
        [meta_shard_fg_expand]: 2.15002e-06
        [get_grad_eliminate_]: 6.56999e-06
        [merge_forward]: 3.81999e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.518e-05
        [j_node_and_user_rematch]: 9.59e-06
        [meta_fg_expand]: 2.55002e-06
        [replace_old_param]: 1.115e-05
        [inline_without_move]: 5.83002e-06
        [renormalize]: 0.00101649
        [add_forward_monad_depend]: 8e-06
        [auto_monad_grad]: 2.95998e-06
        [auto_monad_eliminator]: 1.714e-05
        [cse]: 3.288e-05
        [replace_applicator]: 1.75e-05
    [Cycle 2]: 0.00034134, [27]
        [switch_simplify]: 6.30002e-06
        [loop_unroll]: 5.04e-06
        [a_1]: 9.529e-05
        [with_stream_mark]: 1.19e-05
        [recompute_prepare]: 5.10999e-06
        [updatestate_depend_eliminate]: 2.69999e-06
        [updatestate_assign_eliminate]: 1.93002e-06
        [updatestate_loads_eliminate]: 1.74e-06
        [parameter_eliminate]: 1.39e-06
        [specialize_transform]: 5.02999e-06
        [updatestate_useless_node_eliminater]: 4.58001e-06
        [accelerated_algorithm]: 5.74e-06
        [meta_shard_fg_expand]: 1.78997e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.81999e-06
        [cell_reuse_recompute_pass]: 3.20002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.577e-05
        [j_node_and_user_rematch]: 8.01001e-06
        [meta_fg_expand]: 2.01e-06
        [replace_old_param]: 8.05e-06
        [inline_without_move]: 4.75001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.41998e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 7.67998e-06
        [cse]: 1.189e-05
        [replace_applicator]: 5.09e-06
[py_interpret_to_execute_after_opt_a]: 1.662e-05
[rewriter_after_opt_a]: 4.194e-05
[convert_after_rewriter]: 7.44002e-06
[order_py_execute_after_rewriter]: 7.65998e-06
[mutable_eliminate]: 0.0014261
[jit_opt_b]: 6.536e-05, [1]
    [Cycle 1]: 5.471e-05, [2]
        [frontend_op_eliminate]: 2.234e-05
        [inline_after_opt_a]: 1.985e-05
[cconv]: 4.221e-05
[loop_unroll]: 0.00056785
[jit_opt_after_cconv]: 0.00019284, [1]
    [Cycle 1]: 0.00018472, [11]
        [c_1]: 2.458e-05
        [parameter_eliminate]: 5.67999e-06
        [updatestate_depend_eliminate]: 1.104e-05
        [updatestate_assign_eliminate]: 3.36999e-06
        [updatestate_loads_eliminate]: 2.97002e-06
        [cse]: 3.755e-05
        [call_graph_tuple_transform]: 2.8e-05
        [tuple_list_get_item_eliminator]: 6.24999e-06
        [none_parameter_eliminate]: 2.24001e-06
        [renormalize]: 1.29e-06
        [switch_simplify]: 6.23998e-06
[remove_dup_value]: 1.765e-05
[partial_unused_args_eliminate]: 2.47001e-06
[environ_conv]: 6.70002e-06
[add_recomputation]: 6.116e-05
[cse_after_recomputation]: 2.373e-05, [1]
    [Cycle 1]: 1.707e-05, [1]
        [cse]: 1.063e-05
[auto_monad_reorder]: 1.896e-05
[get_jit_bprop_graph]: 2.01003e-06
[rewriter_after_jit_bprop_graph]: 1.095e-05
[opt_after_jit_grad]: 0.00053309
[symbol_engine_optimizer]: 8.506e-05, [1]
    [Cycle 1]: 7.665e-05, [6]
        [build]: 5.76e-06
        [elim_shapecalc]: 8.67e-06
        [elim_not_effective]: 1.556e-05
        [opt_reshape]: 6.89001e-06
        [fold_const_symbol]: 9.42001e-06
        [renormalize]: 1.02998e-06
[validate]: 4.846e-05
[backend_pass]: 1.10001e-06
[task_emit]: 0.0101632
[execute]: 7.91001e-06
Sums
    bootstrap                            :     0.000364s :  1.01%
    type_inference                       :     0.019992s : 55.56%
    event_method                         :     0.000021s :  0.06%
    auto_monad                           :     0.000070s :  0.20%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000047s :  0.13%
    rewriter_before_opt_a                :     0.000078s :  0.22%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000080s :  0.22%
    jit_opt_a.loop_unroll                :     0.000035s :  0.10%
    jit_opt_a.a_1                        :     0.000737s :  2.05%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000019s :  0.05%
    jit_opt_a.inline_without_move        :     0.000011s :  0.03%
    jit_opt_a.renormalize                :     0.001017s :  2.83%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.07%
    jit_opt_a.cse                        :     0.000045s :  0.12%
    jit_opt_a.replace_applicator         :     0.000023s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.05%
    rewriter_after_opt_a                 :     0.000042s :  0.12%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000008s :  0.02%
    mutable_eliminate                    :     0.001426s :  3.96%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.06%
    cconv                                :     0.000042s :  0.12%
    loop_unroll                          :     0.000568s :  1.58%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000038s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000018s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000061s :  0.17%
    cse_after_recomputation.cse          :     0.000011s :  0.03%
    auto_monad_reorder                   :     0.000019s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.03%
    opt_after_jit_grad                   :     0.000533s :  1.48%
    symbol_engine_optimizer.build        :     0.000006s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000048s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010163s : 28.24%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000293    23
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000002s :      2: substitution.fold_const_symbol
 2.51% :     0.000007s :      3: substitution.graph_param_transform
86.38% :     0.000253s :      4: substitution.inline
 1.29% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.59% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000006s :      2: substitution.replace_old_param
 4.73% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.019891     2
93.71% :     0.018640s :      1: type_inference.infer
 6.29% :     0.001251s :      1: type_inference.specialize
------[replace.]   0.000084     6
66.67% :     0.000056s :      4: replace.inline
33.33% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000262     6
95.16% :     0.000250s :      4: match.inline
 4.84% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.67% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000002s :     10: predicate.addn_zero_filter
 2.31% :     0.000003s :     10: predicate.arithmetic_simplify
 1.22% :     0.000002s :     10: predicate.cast_eliminate
 0.35% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.13% :     0.000001s :     10: predicate.depend_value_elim
 0.96% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.38% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000001s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 5.73% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.47% :     0.000002s :      6: predicate.less_batch_normalization
 1.31% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.90% :     0.000002s :     15: predicate.load_eliminater
 1.71% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.57% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.22% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 3.75% :     0.000005s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.00% :     0.000003s :     15: predicate.partial_eliminate
 1.24% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.27% :     0.000002s :     10: predicate.reduce_eliminate
 1.57% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.94% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.17% :     0.000002s :     10: predicate.reshape_eliminate
 1.09% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.69% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.29% :     0.000002s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.07% :     0.000003s :     16: predicate.switch_defer_inline
 1.79% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.63% :     0.000010s :     53: predicate.switch_simplify
 1.19% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000002s :     10: predicate.transpose_eliminate
 1.28% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.66% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.12% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000886    12
44.29% :     0.000392s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.46% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.26% :     0.000436s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.043534    76
 0.15% :     0.000064s :      1: add_recomputation
 0.17% :     0.000074s :      1: auto_monad
 0.05% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.88% :     0.000385s :      1: bootstrap
 0.10% :     0.000045s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.06% :     0.000026s :      1: cse_after_recomputation
 0.02% :     0.000011s :      1: environ_conv
 0.06% :     0.000026s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
16.57% :     0.007213s :      1: jit_opt_a
 0.45% :     0.000196s :      1: jit_opt_after_cconv
 0.16% :     0.000069s :      1: jit_opt_b
 1.33% :     0.000579s :      1: loop_unroll
 3.33% :     0.001448s :      1: mutable_eliminate
 2.25% :     0.000979s :     26: opt.transform.jit_opt_a
 0.14% :     0.000061s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000034s :      4: opt.transform.jit_opt_b
 0.03% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000030s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.25% :     0.000543s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000050s :      1: py_interpret_to_execute
 0.05% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000020s :      1: remove_dup_value
 1.29% :     0.000561s :      1: renormalize.infer
 1.02% :     0.000445s :      1: renormalize.specialize
 0.03% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000045s :      1: rewriter_after_opt_a
 0.19% :     0.000081s :      1: rewriter_before_opt_a
 0.20% :     0.000088s :      1: symbol_engine_optimizer
23.38% :     0.010179s :      1: task_emit
45.98% :     0.020017s :      1: type_inference
 0.16% :     0.000071s :      1: validate


TotalTime = 0.0214479, [33]
[bootstrap]: 0.00029942
[type_inference]: 0.00622813
[event_method]: 1.877e-05
[auto_monad]: 6.615e-05
[graph_reusing]: 5.92001e-06
[pre_auto_parallel]: 2.43998e-06
[py_interpret_to_execute]: 3.744e-05
[rewriter_before_opt_a]: 7.274e-05
[expand_dump_flag]: 2.92002e-06
[jit_opt_a]: 0.00579583, [2]
    [Cycle 1]: 0.00179266, [27]
        [switch_simplify]: 5.97e-05
        [loop_unroll]: 2.814e-05
        [a_1]: 0.0005499
        [with_stream_mark]: 1.8e-05
        [recompute_prepare]: 7.26001e-06
        [updatestate_depend_eliminate]: 4.68001e-06
        [updatestate_assign_eliminate]: 3.11999e-06
        [updatestate_loads_eliminate]: 2.53003e-06
        [parameter_eliminate]: 2.09e-06
        [specialize_transform]: 5.89999e-06
        [updatestate_useless_node_eliminater]: 5.46e-06
        [accelerated_algorithm]: 5.62999e-06
        [meta_shard_fg_expand]: 2.74999e-06
        [get_grad_eliminate_]: 4.87998e-06
        [merge_forward]: 3.71999e-06
        [cell_reuse_recompute_pass]: 1.43002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.378e-05
        [j_node_and_user_rematch]: 8.80001e-06
        [meta_fg_expand]: 2.32999e-06
        [replace_old_param]: 8.79e-06
        [inline_without_move]: 5.18002e-06
        [renormalize]: 0.00080779
        [add_forward_monad_depend]: 6.07999e-06
        [auto_monad_grad]: 2.67001e-06
        [auto_monad_eliminator]: 1.539e-05
        [cse]: 2.9e-05
        [replace_applicator]: 1.243e-05
    [Cycle 2]: 0.00031956, [27]
        [switch_simplify]: 5.46e-06
        [loop_unroll]: 4.43001e-06
        [a_1]: 8.842e-05
        [with_stream_mark]: 8.85001e-06
        [recompute_prepare]: 4.60999e-06
        [updatestate_depend_eliminate]: 2.86e-06
        [updatestate_assign_eliminate]: 1.92999e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 4.90999e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 4.95001e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 2.69001e-06
        [cell_reuse_recompute_pass]: 2.15002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.244e-05
        [j_node_and_user_rematch]: 7.95e-06
        [meta_fg_expand]: 1.64998e-06
        [replace_old_param]: 7.59002e-06
        [inline_without_move]: 4.73001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 4.79e-06
        [cse]: 1.088e-05
        [replace_applicator]: 4.87e-06
[py_interpret_to_execute_after_opt_a]: 1.049e-05
[rewriter_after_opt_a]: 3.792e-05
[convert_after_rewriter]: 7e-06
[order_py_execute_after_rewriter]: 5.72999e-06
[mutable_eliminate]: 0.00058047
[jit_opt_b]: 5.48e-05, [1]
    [Cycle 1]: 4.701e-05, [2]
        [frontend_op_eliminate]: 2.046e-05
        [inline_after_opt_a]: 1.521e-05
[cconv]: 2.544e-05
[loop_unroll]: 0.00041935
[jit_opt_after_cconv]: 0.0001406, [1]
    [Cycle 1]: 0.00013403, [11]
        [c_1]: 2.11e-05
        [parameter_eliminate]: 2.39999e-06
        [updatestate_depend_eliminate]: 5.24e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 1.918e-05
        [call_graph_tuple_transform]: 1.934e-05
        [tuple_list_get_item_eliminator]: 5.32001e-06
        [none_parameter_eliminate]: 1.45001e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 5.15001e-06
[remove_dup_value]: 1.322e-05
[partial_unused_args_eliminate]: 2.72001e-06
[environ_conv]: 4.99e-06
[add_recomputation]: 5.118e-05
[cse_after_recomputation]: 2.016e-05, [1]
    [Cycle 1]: 1.447e-05, [1]
        [cse]: 8.94e-06
[auto_monad_reorder]: 1.566e-05
[get_jit_bprop_graph]: 1.99999e-06
[rewriter_after_jit_bprop_graph]: 4.82e-06
[opt_after_jit_grad]: 0.00066285
[symbol_engine_optimizer]: 7.337e-05, [1]
    [Cycle 1]: 6.561e-05, [6]
        [build]: 3.63e-06
        [elim_shapecalc]: 7.70998e-06
        [elim_not_effective]: 1.262e-05
        [opt_reshape]: 5.39e-06
        [fold_const_symbol]: 8.33999e-06
        [renormalize]: 7.30011e-07
[validate]: 3.589e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.00653695
[execute]: 8.3e-06
Sums
    bootstrap                            :     0.000299s :  1.75%
    type_inference                       :     0.006228s : 36.32%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000066s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000037s :  0.22%
    rewriter_before_opt_a                :     0.000073s :  0.42%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.38%
    jit_opt_a.loop_unroll                :     0.000033s :  0.19%
    jit_opt_a.a_1                        :     0.000638s :  3.72%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000808s :  4.71%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000040s :  0.23%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000038s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000580s :  3.39%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000025s :  0.15%
    loop_unroll                          :     0.000419s :  2.45%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000051s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000663s :  3.87%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006537s : 38.12%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000216    23
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.85% :     0.000006s :      3: substitution.graph_param_transform
84.51% :     0.000183s :      4: substitution.inline
 1.58% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.09% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.78% :     0.000004s :      2: substitution.replace_old_param
 5.77% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006160     2
84.57% :     0.005209s :      1: type_inference.infer
15.43% :     0.000950s :      1: type_inference.specialize
------[replace.]   0.000074     6
68.64% :     0.000051s :      4: replace.inline
31.36% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     6
94.07% :     0.000180s :      4: match.inline
 5.93% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.42% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.10% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 5.61% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.86% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.86% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.37% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.28% :     0.000001s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.97% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.31% :     0.000001s :      6: predicate.special_op_eliminate
 1.24% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.95% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.72% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.88% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000661    12
50.59% :     0.000335s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.66% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.75% :     0.000296s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023205    76
 0.23% :     0.000054s :      1: add_recomputation
 0.30% :     0.000070s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.37% :     0.000317s :      1: bootstrap
 0.12% :     0.000028s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
24.99% :     0.005799s :      1: jit_opt_a
 0.62% :     0.000144s :      1: jit_opt_after_cconv
 0.25% :     0.000057s :      1: jit_opt_b
 1.84% :     0.000428s :      1: loop_unroll
 2.54% :     0.000589s :      1: mutable_eliminate
 3.65% :     0.000847s :     26: opt.transform.jit_opt_a
 0.20% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000029s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.90% :     0.000672s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000040s :      1: py_interpret_to_execute
 0.05% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.90% :     0.000441s :      1: renormalize.infer
 1.54% :     0.000357s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000041s :      1: rewriter_after_opt_a
 0.33% :     0.000075s :      1: rewriter_before_opt_a
 0.33% :     0.000076s :      1: symbol_engine_optimizer
28.22% :     0.006549s :      1: task_emit
26.91% :     0.006244s :      1: type_inference
 0.24% :     0.000056s :      1: validate


TotalTime = 0.0138478, [30]
[bootstrap]: 0.00035142
[type_inference]: 0.00458541
[event_method]: 1.909e-05
[auto_monad]: 6.756e-05
[graph_reusing]: 6.46e-06
[pre_auto_parallel]: 3.41999e-06
[py_interpret_to_execute]: 3.63e-05
[rewriter_before_opt_a]: 7.262e-05
[expand_dump_flag]: 3.51001e-06
[jit_opt_a]: 0.00640452, [2]
    [Cycle 1]: 0.00182642, [27]
        [switch_simplify]: 6.066e-05
        [loop_unroll]: 3.168e-05
        [a_1]: 0.00058008
        [with_stream_mark]: 1.999e-05
        [recompute_prepare]: 7.31001e-06
        [updatestate_depend_eliminate]: 3.43e-06
        [updatestate_assign_eliminate]: 3.53999e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 6.26e-06
        [updatestate_useless_node_eliminater]: 4.94998e-06
        [accelerated_algorithm]: 6.21e-06
        [meta_shard_fg_expand]: 2.16e-06
        [get_grad_eliminate_]: 5.20999e-06
        [merge_forward]: 3.66999e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.52e-05
        [j_node_and_user_rematch]: 9.86e-06
        [meta_fg_expand]: 2.50002e-06
        [replace_old_param]: 9.17001e-06
        [inline_without_move]: 5.74e-06
        [renormalize]: 0.00079948
        [add_forward_monad_depend]: 7.11001e-06
        [auto_monad_grad]: 3.4e-06
        [auto_monad_eliminator]: 1.534e-05
        [cse]: 2.578e-05
        [replace_applicator]: 1.371e-05
    [Cycle 2]: 0.00033658, [27]
        [switch_simplify]: 5.87999e-06
        [loop_unroll]: 5.24e-06
        [a_1]: 9.368e-05
        [with_stream_mark]: 9.49e-06
        [recompute_prepare]: 5.00999e-06
        [updatestate_depend_eliminate]: 2.71999e-06
        [updatestate_assign_eliminate]: 2.18998e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 4.95001e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.82999e-06
        [meta_shard_fg_expand]: 1.33002e-06
        [get_grad_eliminate_]: 5.25999e-06
        [merge_forward]: 2.88e-06
        [cell_reuse_recompute_pass]: 1.84e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.358e-05
        [j_node_and_user_rematch]: 7.60998e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 7.63999e-06
        [inline_without_move]: 4.93001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.59e-06
        [auto_monad_grad]: 7.80012e-07
        [auto_monad_eliminator]: 5.08002e-06
        [cse]: 1.107e-05
        [replace_applicator]: 5.05001e-06
[py_interpret_to_execute_after_opt_a]: 1.318e-05
[rewriter_after_opt_a]: 3.823e-05
[convert_after_rewriter]: 7.99997e-06
[order_py_execute_after_rewriter]: 5.34998e-06
[mutable_eliminate]: 0.00066833
[jit_opt_b]: 5.925e-05, [1]
    [Cycle 1]: 5.143e-05, [2]
        [frontend_op_eliminate]: 2.351e-05
        [inline_after_opt_a]: 1.617e-05
[cconv]: 2.967e-05
[loop_unroll]: 0.00044297
[jit_opt_after_cconv]: 0.00014517, [1]
    [Cycle 1]: 0.00013841, [11]
        [c_1]: 2.168e-05
        [parameter_eliminate]: 2.83998e-06
        [updatestate_depend_eliminate]: 6.42001e-06
        [updatestate_assign_eliminate]: 2.43002e-06
        [updatestate_loads_eliminate]: 2.23002e-06
        [cse]: 1.994e-05
        [call_graph_tuple_transform]: 1.954e-05
        [tuple_list_get_item_eliminator]: 5.29e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 5.59987e-07
        [switch_simplify]: 5.65001e-06
[remove_dup_value]: 1.595e-05
[partial_unused_args_eliminate]: 2.66e-06
[environ_conv]: 6.02001e-06
[add_recomputation]: 5.128e-05
[cse_after_recomputation]: 2.087e-05, [1]
    [Cycle 1]: 1.526e-05, [1]
        [cse]: 9.62001e-06
[auto_monad_reorder]: 1.603e-05
[get_jit_bprop_graph]: 2.45002e-06
[rewriter_after_jit_bprop_graph]: 6.29999e-06
[opt_after_jit_grad]: 0.00047369
[symbol_engine_optimizer]: 7.712e-05, [1]
    [Cycle 1]: 7.028e-05, [6]
        [build]: 3.55e-06
        [elim_shapecalc]: 8.35001e-06
        [elim_not_effective]: 1.314e-05
        [opt_reshape]: 6.66e-06
        [fold_const_symbol]: 9.12001e-06
        [renormalize]: 6.80011e-07
[validate]: 3.944e-05
Sums
    bootstrap                            :     0.000351s :  3.90%
    type_inference                       :     0.004585s : 50.91%
    event_method                         :     0.000019s :  0.21%
    auto_monad                           :     0.000068s :  0.75%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.04%
    py_interpret_to_execute              :     0.000036s :  0.40%
    rewriter_before_opt_a                :     0.000073s :  0.81%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000067s :  0.74%
    jit_opt_a.loop_unroll                :     0.000037s :  0.41%
    jit_opt_a.a_1                        :     0.000674s :  7.48%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000800s :  8.88%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.23%
    jit_opt_a.cse                        :     0.000037s :  0.41%
    jit_opt_a.replace_applicator         :     0.000019s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.15%
    rewriter_after_opt_a                 :     0.000038s :  0.42%
    convert_after_rewriter               :     0.000008s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000668s :  7.42%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.26%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.18%
    cconv                                :     0.000030s :  0.33%
    loop_unroll                          :     0.000443s :  4.92%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000016s :  0.18%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000051s :  0.57%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000474s :  5.26%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.44%


Time group info:
------[substitution.]   0.000226    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.53% :     0.000006s :      3: substitution.graph_param_transform
84.44% :     0.000190s :      4: substitution.inline
 1.38% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.12% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.80% :     0.000004s :      2: substitution.replace_old_param
 6.14% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004516     2
79.68% :     0.003598s :      1: type_inference.infer
20.32% :     0.000917s :      1: type_inference.specialize
------[replace.]   0.000075     6
65.97% :     0.000050s :      4: replace.inline
34.03% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     6
93.67% :     0.000188s :      4: match.inline
 6.33% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.53% :     0.000002s :     10: predicate.addn_zero_filter
 1.85% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.52% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.42% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.41% :     0.000001s :      3: predicate.graph_param_transform
 5.42% :     0.000007s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.68% :     0.000002s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.27% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.14% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.82% :     0.000002s :      3: predicate.mutable_eliminate
 0.66% :     0.000001s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.33% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.08% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.21% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.62% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.81% :     0.000003s :     16: predicate.switch_defer_inline
 2.00% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.18% :     0.000009s :     53: predicate.switch_simplify
 1.38% :     0.000002s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.76% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000727    12
51.78% :     0.000376s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.77% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.45% :     0.000316s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015600    72
 0.35% :     0.000055s :      1: add_recomputation
 0.45% :     0.000071s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 2.33% :     0.000363s :      1: bootstrap
 0.21% :     0.000033s :      1: cconv
 0.07% :     0.000011s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
41.08% :     0.006408s :      1: jit_opt_a
 0.95% :     0.000148s :      1: jit_opt_after_cconv
 0.40% :     0.000062s :      1: jit_opt_b
 2.90% :     0.000452s :      1: loop_unroll
 4.35% :     0.000678s :      1: mutable_eliminate
 5.74% :     0.000895s :     26: opt.transform.jit_opt_a
 0.31% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000032s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.09% :     0.000482s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000039s :      1: py_interpret_to_execute
 0.10% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000018s :      1: remove_dup_value
 2.47% :     0.000386s :      1: renormalize.infer
 2.59% :     0.000405s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000041s :      1: rewriter_after_opt_a
 0.49% :     0.000076s :      1: rewriter_before_opt_a
 0.51% :     0.000080s :      1: symbol_engine_optimizer
29.48% :     0.004599s :      1: type_inference


TotalTime = 0.0379505, [33]
[bootstrap]: 0.00034639
[type_inference]: 0.005422
[event_method]: 2.1e-05
[auto_monad]: 7.095e-05
[graph_reusing]: 6.16998e-06
[pre_auto_parallel]: 2.96999e-06
[py_interpret_to_execute]: 4.234e-05
[rewriter_before_opt_a]: 7.798e-05
[expand_dump_flag]: 3.16999e-06
[jit_opt_a]: 0.0227158, [2]
    [Cycle 1]: 0.0185829, [27]
        [switch_simplify]: 6.976e-05
        [loop_unroll]: 3.423e-05
        [a_1]: 0.0170322
        [with_stream_mark]: 3.132e-05
        [recompute_prepare]: 1.369e-05
        [updatestate_depend_eliminate]: 4.49998e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.75002e-06
        [parameter_eliminate]: 2.69999e-06
        [specialize_transform]: 7.28e-06
        [updatestate_useless_node_eliminater]: 5.57999e-06
        [accelerated_algorithm]: 6.91001e-06
        [meta_shard_fg_expand]: 2.88e-06
        [get_grad_eliminate_]: 6.09999e-06
        [merge_forward]: 3.88001e-06
        [cell_reuse_recompute_pass]: 1.56002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.471e-05
        [j_node_and_user_rematch]: 1.045e-05
        [meta_fg_expand]: 3.14001e-06
        [replace_old_param]: 1.104e-05
        [inline_without_move]: 5.51e-06
        [renormalize]: 0.00100125
        [add_forward_monad_depend]: 7.28e-06
        [auto_monad_grad]: 2.72001e-06
        [auto_monad_eliminator]: 1.73e-05
        [cse]: 3.568e-05
        [replace_applicator]: 1.385e-05
    [Cycle 2]: 0.00034591, [27]
        [switch_simplify]: 6.24999e-06
        [loop_unroll]: 5.48002e-06
        [a_1]: 9.696e-05
        [with_stream_mark]: 1.214e-05
        [recompute_prepare]: 5.20001e-06
        [updatestate_depend_eliminate]: 3.07002e-06
        [updatestate_assign_eliminate]: 2.15002e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 4.94e-06
        [updatestate_useless_node_eliminater]: 4.64998e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 1.38002e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 3.00002e-06
        [cell_reuse_recompute_pass]: 2.22001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.372e-05
        [j_node_and_user_rematch]: 8.03001e-06
        [meta_fg_expand]: 2.24999e-06
        [replace_old_param]: 7.93999e-06
        [inline_without_move]: 5.11002e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.09998e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 6.12999e-06
        [cse]: 1.237e-05
        [replace_applicator]: 5.44e-06
[py_interpret_to_execute_after_opt_a]: 1.503e-05
[rewriter_after_opt_a]: 3.945e-05
[convert_after_rewriter]: 7.11001e-06
[order_py_execute_after_rewriter]: 9.43002e-06
[mutable_eliminate]: 0.00081267
[jit_opt_b]: 5.469e-05, [1]
    [Cycle 1]: 4.551e-05, [2]
        [frontend_op_eliminate]: 1.852e-05
        [inline_after_opt_a]: 1.498e-05
[cconv]: 3.078e-05
[loop_unroll]: 0.00048366
[jit_opt_after_cconv]: 0.00016956, [1]
    [Cycle 1]: 0.00016168, [11]
        [c_1]: 3.323e-05
        [parameter_eliminate]: 3.31001e-06
        [updatestate_depend_eliminate]: 6.68998e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [cse]: 2.523e-05
        [call_graph_tuple_transform]: 2.12e-05
        [tuple_list_get_item_eliminator]: 5.70001e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 5.92999e-06
[remove_dup_value]: 1.434e-05
[partial_unused_args_eliminate]: 2.37001e-06
[environ_conv]: 5.44998e-06
[add_recomputation]: 5.664e-05
[cse_after_recomputation]: 2.138e-05, [1]
    [Cycle 1]: 1.492e-05, [1]
        [cse]: 8.76002e-06
[auto_monad_reorder]: 1.596e-05
[get_jit_bprop_graph]: 2.61999e-06
[rewriter_after_jit_bprop_graph]: 6.78998e-06
[opt_after_jit_grad]: 0.00049132
[symbol_engine_optimizer]: 7.882e-05, [1]
    [Cycle 1]: 7.143e-05, [6]
        [build]: 5.00001e-06
        [elim_shapecalc]: 8.23999e-06
        [elim_not_effective]: 1.462e-05
        [opt_reshape]: 5.64e-06
        [fold_const_symbol]: 9.24e-06
        [renormalize]: 3.7998e-07
[validate]: 4.44e-05
[backend_pass]: 8.2e-07
[task_emit]: 0.00662486
[execute]: 9.92999e-06
Sums
    bootstrap                            :     0.000346s :  1.04%
    type_inference                       :     0.005422s : 16.22%
    event_method                         :     0.000021s :  0.06%
    auto_monad                           :     0.000071s :  0.21%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.13%
    rewriter_before_opt_a                :     0.000078s :  0.23%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000076s :  0.23%
    jit_opt_a.loop_unroll                :     0.000040s :  0.12%
    jit_opt_a.a_1                        :     0.017129s : 51.23%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.06%
    jit_opt_a.inline_without_move        :     0.000011s :  0.03%
    jit_opt_a.renormalize                :     0.001001s :  2.99%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.07%
    jit_opt_a.cse                        :     0.000048s :  0.14%
    jit_opt_a.replace_applicator         :     0.000019s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.04%
    rewriter_after_opt_a                 :     0.000039s :  0.12%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000009s :  0.03%
    mutable_eliminate                    :     0.000813s :  2.43%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.04%
    cconv                                :     0.000031s :  0.09%
    loop_unroll                          :     0.000484s :  1.45%
    jit_opt_after_cconv.c_1              :     0.000033s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000014s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000057s :  0.17%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000491s :  1.47%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000044s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006625s : 19.81%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.016497    23
 0.01% :     0.000002s :      2: substitution.elim_not_effective
 0.01% :     0.000001s :      2: substitution.fold_const_symbol
 0.04% :     0.000006s :      3: substitution.graph_param_transform
99.75% :     0.016455s :      4: substitution.inline
 0.02% :     0.000004s :      4: substitution.j_node_and_user_rematch
 0.03% :     0.000005s :      4: substitution.remove_not_recompute_node
 0.04% :     0.000006s :      2: substitution.replace_old_param
 0.11% :     0.000018s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005339     2
81.27% :     0.004339s :      1: type_inference.infer
18.73% :     0.001000s :      1: type_inference.specialize
------[replace.]   0.000157     6
79.63% :     0.000125s :      4: replace.inline
20.37% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.016465     6
99.90% :     0.016448s :      4: match.inline
 0.10% :     0.000017s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000152   691
 1.24% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.84% :     0.000001s :     10: predicate.addn_check_dump
 1.59% :     0.000002s :     10: predicate.addn_zero_filter
 3.27% :     0.000005s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.30% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.85% :     0.000001s :     10: predicate.compare_switch_simplify
 1.42% :     0.000002s :     10: predicate.depend_value_elim
 1.28% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.80% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.89% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000002s :     10: predicate.environ_get_eliminate
 1.15% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 6.32% :     0.000010s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.64% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.52% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.73% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.16% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000002s :     10: predicate.merge_addn
 1.14% :     0.000002s :     10: predicate.minmaximum_grad
 2.07% :     0.000003s :      3: predicate.mutable_eliminate
 0.37% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     15: predicate.partial_eliminate
 1.65% :     0.000003s :     10: predicate.print_const_string_wrapper
 1.76% :     0.000003s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.51% :     0.000002s :     18: predicate.replace_applicator
 0.38% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000001s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 0.99% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.01% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.04% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 0.82% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.68% :     0.000006s :     16: predicate.switch_defer_inline
 2.09% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.07% :     0.000011s :     53: predicate.switch_simplify
 1.03% :     0.000002s :     10: predicate.tile_eliminate
 1.14% :     0.000002s :     10: predicate.transpose_eliminate
 1.58% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.57% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.18% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 2.05% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.29% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.47% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.92% :     0.000003s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000754    12
47.28% :     0.000357s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 9.44% :     0.000071s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.28% :     0.000326s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.056484    76
 0.11% :     0.000060s :      1: add_recomputation
 0.13% :     0.000074s :      1: auto_monad
 0.03% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.65% :     0.000370s :      1: bootstrap
 0.06% :     0.000034s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.04% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.05% :     0.000027s :      1: event_method
 0.03% :     0.000014s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
40.22% :     0.022720s :      1: jit_opt_a
 0.31% :     0.000172s :      1: jit_opt_after_cconv
 0.10% :     0.000057s :      1: jit_opt_b
 0.87% :     0.000493s :      1: loop_unroll
 1.46% :     0.000824s :      1: mutable_eliminate
30.75% :     0.017372s :     26: opt.transform.jit_opt_a
 0.11% :     0.000062s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000027s :      4: opt.transform.jit_opt_b
 0.10% :     0.000057s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000034s :      4: opt.transform.symbol_engine_opt
 0.89% :     0.000500s :      1: opt_after_jit_grad
 0.02% :     0.000012s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000045s :      1: py_interpret_to_execute
 0.03% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000017s :      1: remove_dup_value
 0.98% :     0.000553s :      1: renormalize.infer
 0.78% :     0.000438s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000043s :      1: rewriter_after_opt_a
 0.14% :     0.000081s :      1: rewriter_before_opt_a
 0.14% :     0.000081s :      1: symbol_engine_optimizer
11.76% :     0.006644s :      1: task_emit
 9.64% :     0.005444s :      1: type_inference
 0.12% :     0.000069s :      1: validate


TotalTime = 0.0233929, [33]
[bootstrap]: 0.00038087
[type_inference]: 0.00575991
[event_method]: 1.945e-05
[auto_monad]: 0.00011512
[graph_reusing]: 6.62002e-06
[pre_auto_parallel]: 2.66999e-06
[py_interpret_to_execute]: 4.346e-05
[rewriter_before_opt_a]: 7.643e-05
[expand_dump_flag]: 3.75e-06
[jit_opt_a]: 0.00690361, [2]
    [Cycle 1]: 0.0023342, [27]
        [switch_simplify]: 6.959e-05
        [loop_unroll]: 0.0001868
        [a_1]: 0.00070898
        [with_stream_mark]: 2.491e-05
        [recompute_prepare]: 7.92998e-06
        [updatestate_depend_eliminate]: 4.50001e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 2.84999e-06
        [parameter_eliminate]: 2.25002e-06
        [specialize_transform]: 6.55002e-06
        [updatestate_useless_node_eliminater]: 6.02001e-06
        [accelerated_algorithm]: 6.12999e-06
        [meta_shard_fg_expand]: 2.39001e-06
        [get_grad_eliminate_]: 5.71e-06
        [merge_forward]: 3.91999e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.475e-05
        [j_node_and_user_rematch]: 9.72999e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 9.54999e-06
        [inline_without_move]: 5.86e-06
        [renormalize]: 0.00096162
        [add_forward_monad_depend]: 7.38e-06
        [auto_monad_grad]: 3.21001e-06
        [auto_monad_eliminator]: 1.918e-05
        [cse]: 3.477e-05
        [replace_applicator]: 1.623e-05
    [Cycle 2]: 0.00033569, [27]
        [switch_simplify]: 6.23002e-06
        [loop_unroll]: 4.87998e-06
        [a_1]: 9.211e-05
        [with_stream_mark]: 1.275e-05
        [recompute_prepare]: 5.45001e-06
        [updatestate_depend_eliminate]: 3.04999e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.73002e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 4.79e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 5.92999e-06
        [meta_shard_fg_expand]: 1.57001e-06
        [get_grad_eliminate_]: 4.37e-06
        [merge_forward]: 3.01001e-06
        [cell_reuse_recompute_pass]: 2.04999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.486e-05
        [j_node_and_user_rematch]: 8.49002e-06
        [meta_fg_expand]: 1.77999e-06
        [replace_old_param]: 8.15999e-06
        [inline_without_move]: 7.61001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.57001e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 6.32001e-06
        [cse]: 1.056e-05
        [replace_applicator]: 4.75999e-06
[py_interpret_to_execute_after_opt_a]: 1.349e-05
[rewriter_after_opt_a]: 4.215e-05
[convert_after_rewriter]: 7.53e-06
[order_py_execute_after_rewriter]: 5.58002e-06
[mutable_eliminate]: 0.0008715
[jit_opt_b]: 5.286e-05, [1]
    [Cycle 1]: 4.393e-05, [2]
        [frontend_op_eliminate]: 1.776e-05
        [inline_after_opt_a]: 1.446e-05
[cconv]: 3.127e-05
[loop_unroll]: 0.00047387
[jit_opt_after_cconv]: 0.00016008, [1]
    [Cycle 1]: 0.00015228, [11]
        [c_1]: 2.263e-05
        [parameter_eliminate]: 3.95998e-06
        [updatestate_depend_eliminate]: 7.75e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [cse]: 2.624e-05
        [call_graph_tuple_transform]: 2.238e-05
        [tuple_list_get_item_eliminator]: 5.24998e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 8.2e-07
        [switch_simplify]: 5.54e-06
[remove_dup_value]: 0.00043135
[partial_unused_args_eliminate]: 4.46002e-06
[environ_conv]: 9.16998e-06
[add_recomputation]: 6.39e-05
[cse_after_recomputation]: 3.27e-05, [1]
    [Cycle 1]: 2.494e-05, [1]
        [cse]: 1.477e-05
[auto_monad_reorder]: 1.737e-05
[get_jit_bprop_graph]: 2.37001e-06
[rewriter_after_jit_bprop_graph]: 8.58001e-06
[opt_after_jit_grad]: 0.00054608
[symbol_engine_optimizer]: 8.014e-05, [1]
    [Cycle 1]: 7.184e-05, [6]
        [build]: 4.40999e-06
        [elim_shapecalc]: 8.28001e-06
        [elim_not_effective]: 1.477e-05
        [opt_reshape]: 6.24001e-06
        [fold_const_symbol]: 9.47001e-06
        [renormalize]: 7.2e-07
[validate]: 4.674e-05
[backend_pass]: 1.30999e-06
[task_emit]: 0.00690398
[execute]: 1.084e-05
Sums
    bootstrap                            :     0.000381s :  2.07%
    type_inference                       :     0.005760s : 31.24%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000115s :  0.62%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.24%
    rewriter_before_opt_a                :     0.000076s :  0.41%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.41%
    jit_opt_a.loop_unroll                :     0.000192s :  1.04%
    jit_opt_a.a_1                        :     0.000801s :  4.34%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000013s :  0.07%
    jit_opt_a.renormalize                :     0.000962s :  5.22%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.14%
    jit_opt_a.cse                        :     0.000045s :  0.25%
    jit_opt_a.replace_applicator         :     0.000021s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000042s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000872s :  4.73%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.08%
    cconv                                :     0.000031s :  0.17%
    loop_unroll                          :     0.000474s :  2.57%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000431s :  2.34%
    partial_unused_args_eliminate        :     0.000004s :  0.02%
    environ_conv                         :     0.000009s :  0.05%
    add_recomputation                    :     0.000064s :  0.35%
    cse_after_recomputation.cse          :     0.000015s :  0.08%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000546s :  2.96%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000047s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006904s : 37.44%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000314    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000002s :      2: substitution.fold_const_symbol
 2.17% :     0.000007s :      3: substitution.graph_param_transform
87.45% :     0.000275s :      4: substitution.inline
 1.13% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.63% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.81% :     0.000006s :      2: substitution.replace_old_param
 4.59% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005673     2
82.84% :     0.004699s :      1: type_inference.infer
17.16% :     0.000974s :      1: type_inference.specialize
------[replace.]   0.000095     6
73.87% :     0.000070s :      4: replace.inline
26.13% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000285     6
95.36% :     0.000272s :      4: match.inline
 4.64% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000002s :     10: predicate.addn_zero_filter
 2.27% :     0.000003s :     10: predicate.arithmetic_simplify
 1.22% :     0.000002s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.26% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     10: predicate.environ_get_eliminate
 1.21% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.95% :     0.000007s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.37% :     0.000002s :      6: predicate.less_batch_normalization
 1.34% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.88% :     0.000002s :     15: predicate.load_eliminater
 1.83% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.59% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.04% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.13% :     0.000003s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.39% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.24% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.37% :     0.000009s :     53: predicate.switch_simplify
 1.71% :     0.000002s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.28% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.92% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000753    12
48.45% :     0.000365s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.75% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.80% :     0.000330s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025644    76
 0.26% :     0.000067s :      1: add_recomputation
 0.47% :     0.000120s :      1: auto_monad
 0.08% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.58% :     0.000405s :      1: bootstrap
 0.13% :     0.000034s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000035s :      1: cse_after_recomputation
 0.05% :     0.000013s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
26.94% :     0.006908s :      1: jit_opt_a
 0.63% :     0.000163s :      1: jit_opt_after_cconv
 0.22% :     0.000055s :      1: jit_opt_b
 1.88% :     0.000482s :      1: loop_unroll
 3.44% :     0.000883s :      1: mutable_eliminate
 4.66% :     0.001195s :     26: opt.transform.jit_opt_a
 0.20% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.16% :     0.000554s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000007s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000046s :      1: py_interpret_to_execute
 0.06% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 1.72% :     0.000441s :      1: remove_dup_value
 2.04% :     0.000522s :      1: renormalize.infer
 1.68% :     0.000430s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000045s :      1: rewriter_after_opt_a
 0.31% :     0.000079s :      1: rewriter_before_opt_a
 0.32% :     0.000083s :      1: symbol_engine_optimizer
27.00% :     0.006924s :      1: task_emit
22.54% :     0.005781s :      1: type_inference
 0.27% :     0.000070s :      1: validate


TotalTime = 0.015111, [30]
[bootstrap]: 0.0004229
[type_inference]: 0.00543767
[event_method]: 2.019e-05
[auto_monad]: 7.11e-05
[graph_reusing]: 6.00002e-06
[pre_auto_parallel]: 2.62001e-06
[py_interpret_to_execute]: 4.351e-05
[rewriter_before_opt_a]: 7.692e-05
[expand_dump_flag]: 3.89002e-06
[jit_opt_a]: 0.00678556, [2]
    [Cycle 1]: 0.00203208, [27]
        [switch_simplify]: 7.437e-05
        [loop_unroll]: 3.163e-05
        [a_1]: 0.00065427
        [with_stream_mark]: 2.566e-05
        [recompute_prepare]: 1.006e-05
        [updatestate_depend_eliminate]: 4.98001e-06
        [updatestate_assign_eliminate]: 3.2e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 2.72001e-06
        [specialize_transform]: 6.90998e-06
        [updatestate_useless_node_eliminater]: 5.67001e-06
        [accelerated_algorithm]: 6.41e-06
        [meta_shard_fg_expand]: 3.22002e-06
        [get_grad_eliminate_]: 5.53002e-06
        [merge_forward]: 4.18999e-06
        [cell_reuse_recompute_pass]: 2.32999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.705e-05
        [j_node_and_user_rematch]: 1.003e-05
        [meta_fg_expand]: 2.64999e-06
        [replace_old_param]: 1.015e-05
        [inline_without_move]: 5.62001e-06
        [renormalize]: 0.00087299
        [add_forward_monad_depend]: 7.13e-06
        [auto_monad_grad]: 2.96001e-06
        [auto_monad_eliminator]: 1.646e-05
        [cse]: 2.797e-05
        [replace_applicator]: 1.597e-05
    [Cycle 2]: 0.0003421, [27]
        [switch_simplify]: 6.14999e-06
        [loop_unroll]: 5.19e-06
        [a_1]: 9.706e-05
        [with_stream_mark]: 1.118e-05
        [recompute_prepare]: 5.22e-06
        [updatestate_depend_eliminate]: 2.60002e-06
        [updatestate_assign_eliminate]: 2.49999e-06
        [updatestate_loads_eliminate]: 1.93997e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.99e-06
        [updatestate_useless_node_eliminater]: 4.89003e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 1.32999e-06
        [get_grad_eliminate_]: 4.75001e-06
        [merge_forward]: 3.49001e-06
        [cell_reuse_recompute_pass]: 1.77001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.426e-05
        [j_node_and_user_rematch]: 7.97998e-06
        [meta_fg_expand]: 1.76e-06
        [replace_old_param]: 7.70998e-06
        [inline_without_move]: 5.09e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 1.03001e-06
        [auto_monad_eliminator]: 5.54e-06
        [cse]: 1.123e-05
        [replace_applicator]: 5.22e-06
[py_interpret_to_execute_after_opt_a]: 1.332e-05
[rewriter_after_opt_a]: 3.797e-05
[convert_after_rewriter]: 7.75e-06
[order_py_execute_after_rewriter]: 5.25999e-06
[mutable_eliminate]: 0.00063481
[jit_opt_b]: 5.507e-05, [1]
    [Cycle 1]: 4.698e-05, [2]
        [frontend_op_eliminate]: 1.889e-05
        [inline_after_opt_a]: 1.518e-05
[cconv]: 2.969e-05
[loop_unroll]: 0.00042907
[jit_opt_after_cconv]: 0.00014436, [1]
    [Cycle 1]: 0.00013744, [11]
        [c_1]: 2.17e-05
        [parameter_eliminate]: 2.76999e-06
        [updatestate_depend_eliminate]: 6.86999e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.963e-05
        [call_graph_tuple_transform]: 1.907e-05
        [tuple_list_get_item_eliminator]: 5.74e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.37001e-06
[remove_dup_value]: 1.583e-05
[partial_unused_args_eliminate]: 2.61999e-06
[environ_conv]: 5.89999e-06
[add_recomputation]: 5.158e-05
[cse_after_recomputation]: 2.041e-05, [1]
    [Cycle 1]: 1.479e-05, [1]
        [cse]: 9.20001e-06
[auto_monad_reorder]: 1.535e-05
[get_jit_bprop_graph]: 2.20002e-06
[rewriter_after_jit_bprop_graph]: 6.00002e-06
[opt_after_jit_grad]: 0.00045588
[symbol_engine_optimizer]: 7.311e-05, [1]
    [Cycle 1]: 6.665e-05, [6]
        [build]: 4.48001e-06
        [elim_shapecalc]: 7.39002e-06
        [elim_not_effective]: 1.271e-05
        [opt_reshape]: 5.92001e-06
        [fold_const_symbol]: 8.67e-06
        [renormalize]: 5.29981e-07
[validate]: 4.21e-05
Sums
    bootstrap                            :     0.000423s :  4.20%
    type_inference                       :     0.005438s : 54.03%
    event_method                         :     0.000020s :  0.20%
    auto_monad                           :     0.000071s :  0.71%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000044s :  0.43%
    rewriter_before_opt_a                :     0.000077s :  0.76%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000081s :  0.80%
    jit_opt_a.loop_unroll                :     0.000037s :  0.37%
    jit_opt_a.a_1                        :     0.000751s :  7.47%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.37%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.18%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000873s :  8.68%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.22%
    jit_opt_a.cse                        :     0.000039s :  0.39%
    jit_opt_a.replace_applicator         :     0.000021s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.13%
    rewriter_after_opt_a                 :     0.000038s :  0.38%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000635s :  6.31%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.15%
    cconv                                :     0.000030s :  0.30%
    loop_unroll                          :     0.000429s :  4.26%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000016s :  0.16%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000052s :  0.51%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000456s :  4.53%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000042s :  0.42%


Time group info:
------[substitution.]   0.000270    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.05% :     0.000006s :      3: substitution.graph_param_transform
85.65% :     0.000232s :      4: substitution.inline
 1.33% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.97% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.78% :     0.000005s :      2: substitution.replace_old_param
 5.88% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005359     2
80.83% :     0.004332s :      1: type_inference.infer
19.17% :     0.001027s :      1: type_inference.specialize
------[replace.]   0.000089     6
64.43% :     0.000057s :      4: replace.inline
35.57% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000244     6
93.96% :     0.000229s :      4: match.inline
 6.04% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.23% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 2.36% :     0.000003s :     10: predicate.arithmetic_simplify
 1.64% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 0.99% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.59% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.02% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.05% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000002s :     10: predicate.environ_get_eliminate
 1.21% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 5.58% :     0.000008s :     22: predicate.inline
 0.77% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.59% :     0.000002s :      6: predicate.less_batch_normalization
 1.32% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.54% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.00% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.07% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.94% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.71% :     0.000002s :      3: predicate.mutable_eliminate
 0.61% :     0.000001s :      3: predicate.opt_reshape
 1.94% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.29% :     0.000003s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.41% :     0.000002s :     18: predicate.replace_applicator
 0.94% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 2.13% :     0.000003s :     10: predicate.reshape_eliminate
 1.10% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000002s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.15% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.55% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000003s :     16: predicate.switch_layer_defer_inline
 9.69% :     0.000013s :     53: predicate.switch_simplify
 1.12% :     0.000002s :     10: predicate.tile_eliminate
 1.67% :     0.000002s :     10: predicate.transpose_eliminate
 1.51% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.56% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.43% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.46% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.48% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000757    12
48.79% :     0.000369s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.22% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.99% :     0.000340s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017021    72
 0.32% :     0.000054s :      1: add_recomputation
 0.44% :     0.000074s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 2.60% :     0.000442s :      1: bootstrap
 0.19% :     0.000032s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.13% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.89% :     0.006790s :      1: jit_opt_a
 0.86% :     0.000147s :      1: jit_opt_after_cconv
 0.34% :     0.000058s :      1: jit_opt_b
 2.57% :     0.000437s :      1: loop_unroll
 3.79% :     0.000645s :      1: mutable_eliminate
 5.85% :     0.000995s :     26: opt.transform.jit_opt_a
 0.28% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000028s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.73% :     0.000464s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000047s :      1: py_interpret_to_execute
 0.09% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000019s :      1: remove_dup_value
 2.66% :     0.000453s :      1: renormalize.infer
 2.42% :     0.000411s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000041s :      1: rewriter_after_opt_a
 0.47% :     0.000080s :      1: rewriter_before_opt_a
 0.44% :     0.000076s :      1: symbol_engine_optimizer
32.05% :     0.005455s :      1: type_inference


TotalTime = 0.0239667, [33]
[bootstrap]: 0.00032464
[type_inference]: 0.00531715
[event_method]: 2.148e-05
[auto_monad]: 6.545e-05
[graph_reusing]: 6.78998e-06
[pre_auto_parallel]: 2.49001e-06
[py_interpret_to_execute]: 4.539e-05
[rewriter_before_opt_a]: 8.009e-05
[expand_dump_flag]: 3.14999e-06
[jit_opt_a]: 0.00684128, [2]
    [Cycle 1]: 0.00182733, [27]
        [switch_simplify]: 5.308e-05
        [loop_unroll]: 2.914e-05
        [a_1]: 0.00054927
        [with_stream_mark]: 1.507e-05
        [recompute_prepare]: 6.91999e-06
        [updatestate_depend_eliminate]: 2.99001e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.35999e-06
        [specialize_transform]: 5.96e-06
        [updatestate_useless_node_eliminater]: 4.76002e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 1.74e-06
        [get_grad_eliminate_]: 5.39998e-06
        [merge_forward]: 2.84999e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.158e-05
        [j_node_and_user_rematch]: 9.09e-06
        [meta_fg_expand]: 2.07999e-06
        [replace_old_param]: 7.93999e-06
        [inline_without_move]: 5.19e-06
        [renormalize]: 0.00087677
        [add_forward_monad_depend]: 5.80002e-06
        [auto_monad_grad]: 2.39001e-06
        [auto_monad_eliminator]: 1.316e-05
        [cse]: 2.09e-05
        [replace_applicator]: 1.355e-05
    [Cycle 2]: 0.00036408, [27]
        [switch_simplify]: 5.52999e-06
        [loop_unroll]: 5.12e-06
        [a_1]: 9.177e-05
        [with_stream_mark]: 1.135e-05
        [recompute_prepare]: 5.57001e-06
        [updatestate_depend_eliminate]: 2.75002e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.50999e-06
        [specialize_transform]: 4.79e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 1.64e-06
        [get_grad_eliminate_]: 4.58001e-06
        [merge_forward]: 2.84999e-06
        [cell_reuse_recompute_pass]: 3.30998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.441e-05
        [j_node_and_user_rematch]: 8.76002e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 8.06001e-06
        [inline_without_move]: 4.49002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.59e-06
        [auto_monad_grad]: 1.43002e-06
        [auto_monad_eliminator]: 7.63999e-06
        [cse]: 1.163e-05
        [replace_applicator]: 5.51e-06
[py_interpret_to_execute_after_opt_a]: 1.225e-05
[rewriter_after_opt_a]: 3.896e-05
[convert_after_rewriter]: 1.195e-05
[order_py_execute_after_rewriter]: 5.15999e-06
[mutable_eliminate]: 0.00069105
[jit_opt_b]: 5.507e-05, [1]
    [Cycle 1]: 4.686e-05, [2]
        [frontend_op_eliminate]: 1.886e-05
        [inline_after_opt_a]: 1.588e-05
[cconv]: 2.758e-05
[loop_unroll]: 0.00044594
[jit_opt_after_cconv]: 0.00015101, [1]
    [Cycle 1]: 0.00014377, [11]
        [c_1]: 2.218e-05
        [parameter_eliminate]: 3.30998e-06
        [updatestate_depend_eliminate]: 6.22001e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 2.213e-05
        [call_graph_tuple_transform]: 2.204e-05
        [tuple_list_get_item_eliminator]: 5.07999e-06
        [none_parameter_eliminate]: 1.72999e-06
        [renormalize]: 1.00999e-06
        [switch_simplify]: 5.29e-06
[remove_dup_value]: 1.427e-05
[partial_unused_args_eliminate]: 2.91999e-06
[environ_conv]: 5.65001e-06
[add_recomputation]: 5.056e-05
[cse_after_recomputation]: 2.018e-05, [1]
    [Cycle 1]: 1.505e-05, [1]
        [cse]: 8.60001e-06
[auto_monad_reorder]: 1.409e-05
[get_jit_bprop_graph]: 2.19001e-06
[rewriter_after_jit_bprop_graph]: 5.67999e-06
[opt_after_jit_grad]: 0.0004847
[symbol_engine_optimizer]: 8.332e-05, [1]
    [Cycle 1]: 7.61e-05, [6]
        [build]: 5.03002e-06
        [elim_shapecalc]: 8.62998e-06
        [elim_not_effective]: 1.597e-05
        [opt_reshape]: 6.39001e-06
        [fold_const_symbol]: 9.51998e-06
        [renormalize]: 6.50005e-07
[validate]: 4.227e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.00883181
[execute]: 1.148e-05
Sums
    bootstrap                            :     0.000325s :  1.74%
    type_inference                       :     0.005317s : 28.54%
    event_method                         :     0.000021s :  0.12%
    auto_monad                           :     0.000065s :  0.35%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000045s :  0.24%
    rewriter_before_opt_a                :     0.000080s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.31%
    jit_opt_a.loop_unroll                :     0.000034s :  0.18%
    jit_opt_a.a_1                        :     0.000641s :  3.44%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000877s :  4.71%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.11%
    jit_opt_a.cse                        :     0.000033s :  0.17%
    jit_opt_a.replace_applicator         :     0.000019s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000039s :  0.21%
    convert_after_rewriter               :     0.000012s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000691s :  3.71%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000028s :  0.15%
    loop_unroll                          :     0.000446s :  2.39%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000051s :  0.27%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000485s :  2.60%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.23%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008832s : 47.41%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000219    23
 1.17% :     0.000003s :      2: substitution.elim_not_effective
 0.76% :     0.000002s :      2: substitution.fold_const_symbol
 2.66% :     0.000006s :      3: substitution.graph_param_transform
84.43% :     0.000185s :      4: substitution.inline
 1.76% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.85% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.66% :     0.000004s :      2: substitution.replace_old_param
 5.70% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005234     2
79.56% :     0.004165s :      1: type_inference.infer
20.44% :     0.001070s :      1: type_inference.specialize
------[replace.]   0.000072     6
66.22% :     0.000047s :      4: replace.inline
33.78% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000194     6
94.14% :     0.000182s :      4: match.inline
 5.86% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.25% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.48% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.59% :     0.000002s :     10: predicate.addn_zero_filter
 2.00% :     0.000002s :     10: predicate.arithmetic_simplify
 1.69% :     0.000002s :     10: predicate.cast_eliminate
 0.62% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.13% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.78% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.27% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.16% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.46% :     0.000002s :      6: predicate.less_batch_normalization
 1.70% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.00% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.88% :     0.000002s :      3: predicate.mutable_eliminate
 0.61% :     0.000001s :      3: predicate.opt_reshape
 2.02% :     0.000002s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.80% :     0.000001s :      6: predicate.replace_old_param
 0.50% :     0.000001s :      3: predicate.reset_defer_inline
 1.32% :     0.000001s :     10: predicate.reshape_eliminate
 1.51% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.40% :     0.000002s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.59% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.57% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     16: predicate.switch_defer_inline
 2.31% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.51% :     0.000008s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.56% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.44% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000699    12
43.24% :     0.000302s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.53% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.24% :     0.000358s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025800    76
 0.21% :     0.000053s :      1: add_recomputation
 0.27% :     0.000070s :      1: auto_monad
 0.06% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.33% :     0.000344s :      1: bootstrap
 0.12% :     0.000030s :      1: cconv
 0.06% :     0.000015s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.10% :     0.000027s :      1: event_method
 0.07% :     0.000018s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.53% :     0.006845s :      1: jit_opt_a
 0.60% :     0.000154s :      1: jit_opt_after_cconv
 0.22% :     0.000058s :      1: jit_opt_b
 1.76% :     0.000455s :      1: loop_unroll
 2.72% :     0.000701s :      1: mutable_eliminate
 3.30% :     0.000851s :     26: opt.transform.jit_opt_a
 0.20% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.92% :     0.000495s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000049s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.85% :     0.000476s :      1: renormalize.infer
 1.52% :     0.000392s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000043s :      1: rewriter_after_opt_a
 0.32% :     0.000083s :      1: rewriter_before_opt_a
 0.33% :     0.000086s :      1: symbol_engine_optimizer
34.32% :     0.008854s :      1: task_emit
20.70% :     0.005341s :      1: type_inference
 0.24% :     0.000062s :      1: validate


TotalTime = 0.0456744, [33]
[bootstrap]: 0.00031258
[type_inference]: 0.00491489
[event_method]: 1.748e-05
[auto_monad]: 6.57e-05
[graph_reusing]: 5.68002e-06
[pre_auto_parallel]: 2.36998e-06
[py_interpret_to_execute]: 3.724e-05
[rewriter_before_opt_a]: 6.926e-05
[expand_dump_flag]: 3.08e-06
[jit_opt_a]: 0.0305219, [2]
    [Cycle 1]: 0.00214339, [27]
        [switch_simplify]: 6.867e-05
        [loop_unroll]: 3.104e-05
        [a_1]: 0.00068972
        [with_stream_mark]: 2.398e-05
        [recompute_prepare]: 8.42e-06
        [updatestate_depend_eliminate]: 3.7e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 2.64999e-06
        [parameter_eliminate]: 1.95001e-06
        [specialize_transform]: 5.96e-06
        [updatestate_useless_node_eliminater]: 5.07999e-06
        [accelerated_algorithm]: 5.69e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 6.33e-06
        [merge_forward]: 3.54002e-06
        [cell_reuse_recompute_pass]: 1.85001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.528e-05
        [j_node_and_user_rematch]: 9.76998e-06
        [meta_fg_expand]: 2.83998e-06
        [replace_old_param]: 1.009e-05
        [inline_without_move]: 5.92999e-06
        [renormalize]: 0.00097271
        [add_forward_monad_depend]: 6.92002e-06
        [auto_monad_grad]: 2.83e-06
        [auto_monad_eliminator]: 1.682e-05
        [cse]: 3.058e-05
        [replace_applicator]: 1.494e-05
    [Cycle 2]: 0.00034974, [27]
        [switch_simplify]: 6.57002e-06
        [loop_unroll]: 5.69e-06
        [a_1]: 9.53e-05
        [with_stream_mark]: 1.371e-05
        [recompute_prepare]: 5.89999e-06
        [updatestate_depend_eliminate]: 3.49001e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 5.05999e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 5.64998e-06
        [meta_shard_fg_expand]: 1.42999e-06
        [get_grad_eliminate_]: 4.69002e-06
        [merge_forward]: 2.71999e-06
        [cell_reuse_recompute_pass]: 2.44001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.517e-05
        [j_node_and_user_rematch]: 8.25e-06
        [meta_fg_expand]: 1.68002e-06
        [replace_old_param]: 7.37002e-06
        [inline_without_move]: 4.65001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.04003e-06
        [auto_monad_grad]: 8.80013e-07
        [auto_monad_eliminator]: 5.77999e-06
        [cse]: 1.159e-05
        [replace_applicator]: 5.82001e-06
[py_interpret_to_execute_after_opt_a]: 1.293e-05
[rewriter_after_opt_a]: 3.912e-05
[convert_after_rewriter]: 7.66001e-06
[order_py_execute_after_rewriter]: 4.87e-06
[mutable_eliminate]: 0.00068921
[jit_opt_b]: 5.442e-05, [1]
    [Cycle 1]: 4.621e-05, [2]
        [frontend_op_eliminate]: 1.827e-05
        [inline_after_opt_a]: 1.568e-05
[cconv]: 3.138e-05
[loop_unroll]: 0.00057825
[jit_opt_after_cconv]: 0.00016419, [1]
    [Cycle 1]: 0.00015654, [11]
        [c_1]: 2.327e-05
        [parameter_eliminate]: 4.43001e-06
        [updatestate_depend_eliminate]: 7.58001e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [cse]: 2.711e-05
        [call_graph_tuple_transform]: 2.366e-05
        [tuple_list_get_item_eliminator]: 5.64998e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 7.2e-07
        [switch_simplify]: 5.46e-06
[remove_dup_value]: 1.388e-05
[partial_unused_args_eliminate]: 2.56e-06
[environ_conv]: 6.32001e-06
[add_recomputation]: 5.558e-05
[cse_after_recomputation]: 2.228e-05, [1]
    [Cycle 1]: 1.603e-05, [1]
        [cse]: 9.45001e-06
[auto_monad_reorder]: 1.573e-05
[get_jit_bprop_graph]: 2.25002e-06
[rewriter_after_jit_bprop_graph]: 4.781e-05
[opt_after_jit_grad]: 0.00051974
[symbol_engine_optimizer]: 8.16e-05, [1]
    [Cycle 1]: 7.389e-05, [6]
        [build]: 5.60001e-06
        [elim_shapecalc]: 8.56002e-06
        [elim_not_effective]: 1.471e-05
        [opt_reshape]: 6.14999e-06
        [fold_const_symbol]: 9.85002e-06
        [renormalize]: 3.69997e-07
[validate]: 4.595e-05
[backend_pass]: 1.05999e-06
[task_emit]: 0.00707674
[execute]: 9.91e-06
Sums
    bootstrap                            :     0.000313s :  1.84%
    type_inference                       :     0.004915s : 28.98%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000066s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000037s :  0.22%
    rewriter_before_opt_a                :     0.000069s :  0.41%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000075s :  0.44%
    jit_opt_a.loop_unroll                :     0.000037s :  0.22%
    jit_opt_a.a_1                        :     0.000785s :  4.63%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000973s :  5.74%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.13%
    jit_opt_a.cse                        :     0.000042s :  0.25%
    jit_opt_a.replace_applicator         :     0.000021s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000039s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000689s :  4.06%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000031s :  0.19%
    loop_unroll                          :     0.000578s :  3.41%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000056s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000048s :  0.28%
    opt_after_jit_grad                   :     0.000520s :  3.06%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000046s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007077s : 41.72%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000334    23
 0.69% :     0.000002s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 1.75% :     0.000006s :      3: substitution.graph_param_transform
89.27% :     0.000298s :      4: substitution.inline
 1.28% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.29% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.32% :     0.000004s :      2: substitution.replace_old_param
 3.96% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004846     2
81.10% :     0.003930s :      1: type_inference.infer
18.90% :     0.000916s :      1: type_inference.specialize
------[replace.]   0.000078     6
67.64% :     0.000053s :      4: replace.inline
32.36% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000308     6
96.12% :     0.000296s :      4: match.inline
 3.88% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.13% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.63% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 2.75% :     0.000004s :     10: predicate.arithmetic_simplify
 1.14% :     0.000001s :     10: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.93% :     0.000001s :     10: predicate.compare_switch_simplify
 1.52% :     0.000002s :     10: predicate.depend_value_elim
 0.97% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.38% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.91% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.05% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000001s :      3: predicate.graph_param_transform
 5.75% :     0.000007s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.35% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.77% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.77% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 2.32% :     0.000003s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     15: predicate.partial_eliminate
 1.01% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.00% :     0.000003s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.20% :     0.000002s :     10: predicate.reshape_eliminate
 1.08% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.02% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.72% :     0.000004s :     16: predicate.switch_defer_inline
 1.84% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.98% :     0.000010s :     53: predicate.switch_simplify
 1.28% :     0.000002s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.99% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.81% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000680    12
46.33% :     0.000315s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.50% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.17% :     0.000314s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.047785    76
 0.12% :     0.000059s :      1: add_recomputation
 0.14% :     0.000069s :      1: auto_monad
 0.04% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.70% :     0.000334s :      1: bootstrap
 0.07% :     0.000034s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.05% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.04% :     0.000021s :      1: event_method
 0.03% :     0.000015s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
63.88% :     0.030525s :      1: jit_opt_a
 0.35% :     0.000168s :      1: jit_opt_after_cconv
 0.12% :     0.000058s :      1: jit_opt_b
 1.24% :     0.000590s :      1: loop_unroll
 1.47% :     0.000702s :      1: mutable_eliminate
 2.14% :     0.001021s :     26: opt.transform.jit_opt_a
 0.11% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000027s :      4: opt.transform.jit_opt_b
 0.03% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.11% :     0.000531s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000040s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000016s :      1: remove_dup_value
 1.12% :     0.000536s :      1: renormalize.infer
 0.89% :     0.000427s :      1: renormalize.specialize
 0.11% :     0.000051s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000045s :      1: rewriter_after_opt_a
 0.15% :     0.000072s :      1: rewriter_before_opt_a
 0.18% :     0.000085s :      1: symbol_engine_optimizer
14.85% :     0.007096s :      1: task_emit
10.31% :     0.004927s :      1: type_inference
 0.15% :     0.000071s :      1: validate


TotalTime = 0.024478, [33]
[bootstrap]: 0.00036895
[type_inference]: 0.00595678
[event_method]: 2.253e-05
[auto_monad]: 7.184e-05
[graph_reusing]: 6.01e-06
[pre_auto_parallel]: 2.98998e-06
[py_interpret_to_execute]: 4.585e-05
[rewriter_before_opt_a]: 7.634e-05
[expand_dump_flag]: 3.51001e-06
[jit_opt_a]: 0.00691424, [2]
    [Cycle 1]: 0.00201383, [27]
        [switch_simplify]: 6.398e-05
        [loop_unroll]: 2.887e-05
        [a_1]: 0.0005914
        [with_stream_mark]: 2.255e-05
        [recompute_prepare]: 7.70998e-06
        [updatestate_depend_eliminate]: 3.87998e-06
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [parameter_eliminate]: 2.24001e-06
        [specialize_transform]: 6.58e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.52999e-06
        [meta_shard_fg_expand]: 1.95001e-06
        [get_grad_eliminate_]: 5.42999e-06
        [merge_forward]: 3.6e-06
        [cell_reuse_recompute_pass]: 1.06002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.463e-05
        [j_node_and_user_rematch]: 9.80002e-06
        [meta_fg_expand]: 2.02999e-06
        [replace_old_param]: 9.17001e-06
        [inline_without_move]: 5.47999e-06
        [renormalize]: 0.00096096
        [add_forward_monad_depend]: 6.59999e-06
        [auto_monad_grad]: 2.53e-06
        [auto_monad_eliminator]: 1.708e-05
        [cse]: 3.315e-05
        [replace_applicator]: 1.54e-05
    [Cycle 2]: 0.00042272, [27]
        [switch_simplify]: 6.14999e-06
        [loop_unroll]: 4.77998e-06
        [a_1]: 9.199e-05
        [with_stream_mark]: 1.157e-05
        [recompute_prepare]: 5.39e-06
        [updatestate_depend_eliminate]: 2.87002e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [parameter_eliminate]: 1.43002e-06
        [specialize_transform]: 5.55001e-06
        [updatestate_useless_node_eliminater]: 4.78001e-06
        [accelerated_algorithm]: 6.31998e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 5.07e-06
        [merge_forward]: 2.99999e-06
        [cell_reuse_recompute_pass]: 2.68e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.536e-05
        [j_node_and_user_rematch]: 7.76001e-06
        [meta_fg_expand]: 1.81e-06
        [replace_old_param]: 8.25e-06
        [inline_without_move]: 5.12e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 1.99e-06
        [auto_monad_eliminator]: 8.63001e-06
        [cse]: 1.495e-05
        [replace_applicator]: 6.94999e-06
[py_interpret_to_execute_after_opt_a]: 1.747e-05
[rewriter_after_opt_a]: 4.16e-05
[convert_after_rewriter]: 8.41002e-06
[order_py_execute_after_rewriter]: 6.13998e-06
[mutable_eliminate]: 0.00081709
[jit_opt_b]: 5.952e-05, [1]
    [Cycle 1]: 5.051e-05, [2]
        [frontend_op_eliminate]: 2.046e-05
        [inline_after_opt_a]: 1.69e-05
[cconv]: 3.672e-05
[loop_unroll]: 0.00054023
[jit_opt_after_cconv]: 0.00017214, [1]
    [Cycle 1]: 0.00016485, [11]
        [c_1]: 2.381e-05
        [parameter_eliminate]: 5.30001e-06
        [updatestate_depend_eliminate]: 9.69999e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [cse]: 2.893e-05
        [call_graph_tuple_transform]: 2.605e-05
        [tuple_list_get_item_eliminator]: 6.07999e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 7.80012e-07
        [switch_simplify]: 5.54e-06
[remove_dup_value]: 1.723e-05
[partial_unused_args_eliminate]: 2.19999e-06
[environ_conv]: 6.77002e-06
[add_recomputation]: 5.913e-05
[cse_after_recomputation]: 2.216e-05, [1]
    [Cycle 1]: 1.557e-05, [1]
        [cse]: 9.28002e-06
[auto_monad_reorder]: 1.664e-05
[get_jit_bprop_graph]: 2.28002e-06
[rewriter_after_jit_bprop_graph]: 8.61997e-06
[opt_after_jit_grad]: 0.00057662
[symbol_engine_optimizer]: 8.587e-05, [1]
    [Cycle 1]: 7.694e-05, [6]
        [build]: 6.12001e-06
        [elim_shapecalc]: 8.82999e-06
        [elim_not_effective]: 1.654e-05
        [opt_reshape]: 6.23998e-06
        [fold_const_symbol]: 9.82999e-06
        [renormalize]: 5.69999e-07
[validate]: 5.432e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.00817478
[execute]: 1.045e-05
Sums
    bootstrap                            :     0.000369s :  1.92%
    type_inference                       :     0.005957s : 30.99%
    event_method                         :     0.000023s :  0.12%
    auto_monad                           :     0.000072s :  0.37%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000046s :  0.24%
    rewriter_before_opt_a                :     0.000076s :  0.40%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000070s :  0.36%
    jit_opt_a.loop_unroll                :     0.000034s :  0.18%
    jit_opt_a.a_1                        :     0.000683s :  3.56%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000961s :  5.00%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.13%
    jit_opt_a.cse                        :     0.000048s :  0.25%
    jit_opt_a.replace_applicator         :     0.000022s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.09%
    rewriter_after_opt_a                 :     0.000042s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000817s :  4.25%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000037s :  0.19%
    loop_unroll                          :     0.000540s :  2.81%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000059s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.04%
    opt_after_jit_grad                   :     0.000577s :  3.00%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000054s :  0.28%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008175s : 42.53%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000255    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000002s :      2: substitution.fold_const_symbol
 2.65% :     0.000007s :      3: substitution.graph_param_transform
84.72% :     0.000216s :      4: substitution.inline
 1.35% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.03% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.24% :     0.000006s :      2: substitution.replace_old_param
 5.46% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005863     2
79.79% :     0.004679s :      1: type_inference.infer
20.21% :     0.001185s :      1: type_inference.specialize
------[replace.]   0.000076     6
68.29% :     0.000052s :      4: replace.inline
31.71% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000226     6
94.27% :     0.000213s :      4: match.inline
 5.73% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.68% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.09% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.18% :     0.000001s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.14% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.29% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.49% :     0.000001s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     10: predicate.environ_get_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 5.34% :     0.000007s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 2.29% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.78% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.08% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 3.11% :     0.000004s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.28% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.70% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.22% :     0.000001s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.32% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.21% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.57% :     0.000009s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.02% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.97% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.66% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000828    12
42.81% :     0.000355s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.24% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.94% :     0.000422s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026476    76
 0.24% :     0.000063s :      1: add_recomputation
 0.28% :     0.000075s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.46% :     0.000387s :      1: bootstrap
 0.15% :     0.000040s :      1: cconv
 0.04% :     0.000012s :      1: convert_after_rewriter
 0.09% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000026s :      1: event_method
 0.06% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.13% :     0.006919s :      1: jit_opt_a
 0.66% :     0.000176s :      1: jit_opt_after_cconv
 0.23% :     0.000062s :      1: jit_opt_b
 2.09% :     0.000553s :      1: loop_unroll
 3.15% :     0.000833s :      1: mutable_eliminate
 3.45% :     0.000912s :     26: opt.transform.jit_opt_a
 0.22% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000030s :      4: opt.transform.jit_opt_b
 0.06% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000038s :      4: opt.transform.symbol_engine_opt
 2.23% :     0.000592s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000049s :      1: py_interpret_to_execute
 0.08% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000020s :      1: remove_dup_value
 1.98% :     0.000525s :      1: renormalize.infer
 1.61% :     0.000426s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000046s :      1: rewriter_after_opt_a
 0.30% :     0.000079s :      1: rewriter_before_opt_a
 0.33% :     0.000089s :      1: symbol_engine_optimizer
30.96% :     0.008197s :      1: task_emit
22.59% :     0.005981s :      1: type_inference
 0.31% :     0.000082s :      1: validate


TotalTime = 0.0304683, [30]
[bootstrap]: 0.00045131
[type_inference]: 0.00903203
[event_method]: 2.886e-05
[auto_monad]: 8.671e-05
[graph_reusing]: 7.1e-06
[pre_auto_parallel]: 3.62002e-06
[py_interpret_to_execute]: 6.403e-05
[rewriter_before_opt_a]: 9.343e-05
[expand_dump_flag]: 3.41999e-06
[jit_opt_a]: 0.00733225, [2]
    [Cycle 1]: 0.00219579, [27]
        [switch_simplify]: 7.162e-05
        [loop_unroll]: 3.091e-05
        [a_1]: 0.00067237
        [with_stream_mark]: 2.731e-05
        [recompute_prepare]: 8.17998e-06
        [updatestate_depend_eliminate]: 4.52998e-06
        [updatestate_assign_eliminate]: 3.10002e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [parameter_eliminate]: 2.76e-06
        [specialize_transform]: 6.48998e-06
        [updatestate_useless_node_eliminater]: 5.42001e-06
        [accelerated_algorithm]: 6.21e-06
        [meta_shard_fg_expand]: 2.71e-06
        [get_grad_eliminate_]: 6.36e-06
        [merge_forward]: 4.2e-06
        [cell_reuse_recompute_pass]: 1.38002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.617e-05
        [j_node_and_user_rematch]: 1.045e-05
        [meta_fg_expand]: 2.23998e-06
        [replace_old_param]: 1.198e-05
        [inline_without_move]: 6.09001e-06
        [renormalize]: 0.00100834
        [add_forward_monad_depend]: 8.84e-06
        [auto_monad_grad]: 3.03e-06
        [auto_monad_eliminator]: 1.835e-05
        [cse]: 3.338e-05
        [replace_applicator]: 1.705e-05
    [Cycle 2]: 0.00036282, [27]
        [switch_simplify]: 6.59001e-06
        [loop_unroll]: 5.15999e-06
        [a_1]: 9.77e-05
        [with_stream_mark]: 1.457e-05
        [recompute_prepare]: 6.29999e-06
        [updatestate_depend_eliminate]: 3.53999e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [parameter_eliminate]: 1.59e-06
        [specialize_transform]: 5.39e-06
        [updatestate_useless_node_eliminater]: 5.65001e-06
        [accelerated_algorithm]: 6.02999e-06
        [meta_shard_fg_expand]: 2.19999e-06
        [get_grad_eliminate_]: 4.69998e-06
        [merge_forward]: 3.28998e-06
        [cell_reuse_recompute_pass]: 2.63998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.656e-05
        [j_node_and_user_rematch]: 8.29998e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 9.24e-06
        [inline_without_move]: 4.77e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.44998e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 6.34001e-06
        [cse]: 1.248e-05
        [replace_applicator]: 5.63002e-06
[py_interpret_to_execute_after_opt_a]: 1.471e-05
[rewriter_after_opt_a]: 4.206e-05
[convert_after_rewriter]: 7.87998e-06
[order_py_execute_after_rewriter]: 5.46002e-06
[mutable_eliminate]: 0.00086381
[jit_opt_b]: 6.591e-05, [1]
    [Cycle 1]: 5.635e-05, [2]
        [frontend_op_eliminate]: 2.222e-05
        [inline_after_opt_a]: 1.934e-05
[cconv]: 3.647e-05
[loop_unroll]: 0.00060778
[jit_opt_after_cconv]: 0.00019368, [1]
    [Cycle 1]: 0.00018563, [11]
        [c_1]: 2.553e-05
        [parameter_eliminate]: 5.99e-06
        [updatestate_depend_eliminate]: 1.115e-05
        [updatestate_assign_eliminate]: 3.13e-06
        [updatestate_loads_eliminate]: 2.39001e-06
        [cse]: 3.434e-05
        [call_graph_tuple_transform]: 2.907e-05
        [tuple_list_get_item_eliminator]: 6.24001e-06
        [none_parameter_eliminate]: 2.16e-06
        [renormalize]: 8.70001e-07
        [switch_simplify]: 6.03998e-06
[remove_dup_value]: 1.818e-05
[partial_unused_args_eliminate]: 3.58e-06
[environ_conv]: 7.33999e-06
[add_recomputation]: 0.00987394
[cse_after_recomputation]: 8.145e-05, [1]
    [Cycle 1]: 6.56e-05, [1]
        [cse]: 4.26e-05
[auto_monad_reorder]: 3.022e-05
[get_jit_bprop_graph]: 2.40002e-06
[rewriter_after_jit_bprop_graph]: 1.425e-05
[opt_after_jit_grad]: 0.00099616
[symbol_engine_optimizer]: 0.00010204, [1]
    [Cycle 1]: 9.1e-05, [6]
        [build]: 7.10002e-06
        [elim_shapecalc]: 9.12001e-06
        [elim_not_effective]: 2.309e-05
        [opt_reshape]: 8.86002e-06
        [fold_const_symbol]: 1.086e-05
        [renormalize]: 1.07e-06
[validate]: 5.457e-05
Sums
    bootstrap                            :     0.000451s :  1.82%
    type_inference                       :     0.009032s : 36.35%
    event_method                         :     0.000029s :  0.12%
    auto_monad                           :     0.000087s :  0.35%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000004s :  0.01%
    py_interpret_to_execute              :     0.000064s :  0.26%
    rewriter_before_opt_a                :     0.000093s :  0.38%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000078s :  0.31%
    jit_opt_a.loop_unroll                :     0.000036s :  0.15%
    jit_opt_a.a_1                        :     0.000770s :  3.10%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.04%
    jit_opt_a.renormalize                :     0.001008s :  4.06%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.10%
    jit_opt_a.cse                        :     0.000046s :  0.18%
    jit_opt_a.replace_applicator         :     0.000023s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.06%
    rewriter_after_opt_a                 :     0.000042s :  0.17%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000864s :  3.48%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.08%
    cconv                                :     0.000036s :  0.15%
    loop_unroll                          :     0.000608s :  2.45%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000034s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000029s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000018s :  0.07%
    partial_unused_args_eliminate        :     0.000004s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.009874s : 39.73%
    cse_after_recomputation.cse          :     0.000043s :  0.17%
    auto_monad_reorder                   :     0.000030s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000014s :  0.06%
    opt_after_jit_grad                   :     0.000996s :  4.01%
    symbol_engine_optimizer.build        :     0.000007s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000023s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000055s :  0.22%


Time group info:
------[substitution.]   0.000309    23
 0.88% :     0.000003s :      2: substitution.elim_not_effective
 0.67% :     0.000002s :      2: substitution.fold_const_symbol
 2.35% :     0.000007s :      3: substitution.graph_param_transform
86.12% :     0.000266s :      4: substitution.inline
 1.29% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.61% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.20% :     0.000007s :      2: substitution.replace_old_param
 4.88% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008913     2
82.92% :     0.007391s :      1: type_inference.infer
17.08% :     0.001522s :      1: type_inference.specialize
------[replace.]   0.000087     6
66.50% :     0.000058s :      4: replace.inline
33.50% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000278     6
94.97% :     0.000264s :      4: match.inline
 5.03% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   691
 1.26% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.24% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.10% :     0.000002s :     10: predicate.addn_zero_filter
 2.02% :     0.000003s :     10: predicate.arithmetic_simplify
 1.19% :     0.000002s :     10: predicate.cast_eliminate
 0.55% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.90% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000002s :     10: predicate.depend_value_elim
 0.93% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.15% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.24% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.80% :     0.000001s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.04% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.89% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 6.60% :     0.000009s :     22: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.78% :     0.000002s :      6: predicate.less_batch_normalization
 1.23% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 2.13% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.46% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.90% :     0.000001s :     10: predicate.merge_addn
 0.90% :     0.000001s :     10: predicate.minmaximum_grad
 2.45% :     0.000003s :      3: predicate.mutable_eliminate
 1.01% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 1.23% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.81% :     0.000002s :     10: predicate.reduce_eliminate
 1.30% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.47% :     0.000001s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 0.97% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.34% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.04% :     0.000001s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.34% :     0.000002s :      6: predicate.special_op_eliminate
 1.11% :     0.000002s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.20% :     0.000003s :     16: predicate.switch_defer_inline
 1.76% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.97% :     0.000010s :     53: predicate.switch_simplify
 1.11% :     0.000002s :     10: predicate.tile_eliminate
 0.95% :     0.000001s :     10: predicate.transpose_eliminate
 1.52% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.32% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.43% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.89% :     0.000003s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.11% :     0.000002s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000972    12
41.67% :     0.000405s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.63% :     0.000074s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.71% :     0.000493s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.032538    72
30.47% :     0.009915s :      1: add_recomputation
 0.28% :     0.000091s :      1: auto_monad
 0.10% :     0.000034s :      1: auto_monad_reorder
 1.47% :     0.000479s :      1: bootstrap
 0.12% :     0.000040s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.27% :     0.000088s :      1: cse_after_recomputation
 0.03% :     0.000011s :      1: environ_conv
 0.11% :     0.000037s :      1: event_method
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
22.55% :     0.007337s :      1: jit_opt_a
 0.61% :     0.000197s :      1: jit_opt_after_cconv
 0.21% :     0.000069s :      1: jit_opt_b
 1.91% :     0.000622s :      1: loop_unroll
 2.70% :     0.000878s :      1: mutable_eliminate
 3.13% :     0.001019s :     26: opt.transform.jit_opt_a
 0.19% :     0.000063s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000034s :      4: opt.transform.jit_opt_b
 0.06% :     0.000021s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000040s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000048s :      4: opt.transform.symbol_engine_opt
 3.10% :     0.001007s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000006s :      1: partial_unused_args_eliminate
 0.02% :     0.000006s :      1: pre_auto_parallel
 0.21% :     0.000068s :      1: py_interpret_to_execute
 0.05% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000022s :      1: remove_dup_value
 1.67% :     0.000544s :      1: renormalize.infer
 1.40% :     0.000455s :      1: renormalize.specialize
 0.05% :     0.000016s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000046s :      1: rewriter_after_opt_a
 0.30% :     0.000098s :      1: rewriter_before_opt_a
 0.32% :     0.000105s :      1: symbol_engine_optimizer
27.87% :     0.009067s :      1: type_inference


TotalTime = 0.0220088, [33]
[bootstrap]: 0.00033423
[type_inference]: 0.00539701
[event_method]: 1.831e-05
[auto_monad]: 6.88e-05
[graph_reusing]: 5.59998e-06
[pre_auto_parallel]: 2.45002e-06
[py_interpret_to_execute]: 4.293e-05
[rewriter_before_opt_a]: 7.732e-05
[expand_dump_flag]: 2.98e-06
[jit_opt_a]: 0.00636989, [2]
    [Cycle 1]: 0.00219559, [27]
        [switch_simplify]: 7.019e-05
        [loop_unroll]: 2.949e-05
        [a_1]: 0.00063238
        [with_stream_mark]: 2.476e-05
        [recompute_prepare]: 8.38999e-06
        [updatestate_depend_eliminate]: 4.17e-06
        [updatestate_assign_eliminate]: 3.3e-06
        [updatestate_loads_eliminate]: 2.91e-06
        [parameter_eliminate]: 1.84998e-06
        [specialize_transform]: 6.73e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 5.94e-06
        [meta_shard_fg_expand]: 2.63e-06
        [get_grad_eliminate_]: 5.52001e-06
        [merge_forward]: 3.48e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.43e-05
        [j_node_and_user_rematch]: 1.01e-05
        [meta_fg_expand]: 2.51e-06
        [replace_old_param]: 1.065e-05
        [inline_without_move]: 6.77002e-06
        [renormalize]: 0.00105444
        [add_forward_monad_depend]: 7.68999e-06
        [auto_monad_grad]: 2.96001e-06
        [auto_monad_eliminator]: 1.79e-05
        [cse]: 3.085e-05
        [replace_applicator]: 1.822e-05
    [Cycle 2]: 0.00034334, [27]
        [switch_simplify]: 6.02999e-06
        [loop_unroll]: 4.94e-06
        [a_1]: 9.751e-05
        [with_stream_mark]: 1.26e-05
        [recompute_prepare]: 6.08002e-06
        [updatestate_depend_eliminate]: 3.96001e-06
        [updatestate_assign_eliminate]: 2.49999e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 4.81002e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.07e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 4.89e-06
        [merge_forward]: 3.26001e-06
        [cell_reuse_recompute_pass]: 3.15998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.41e-05
        [j_node_and_user_rematch]: 8.40999e-06
        [meta_fg_expand]: 1.99e-06
        [replace_old_param]: 8.03999e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 6.53003e-06
        [cse]: 1.227e-05
        [replace_applicator]: 5.35001e-06
[py_interpret_to_execute_after_opt_a]: 1.481e-05
[rewriter_after_opt_a]: 3.843e-05
[convert_after_rewriter]: 6.93998e-06
[order_py_execute_after_rewriter]: 4.88001e-06
[mutable_eliminate]: 0.00076668
[jit_opt_b]: 5.686e-05, [1]
    [Cycle 1]: 4.798e-05, [2]
        [frontend_op_eliminate]: 1.945e-05
        [inline_after_opt_a]: 1.699e-05
[cconv]: 3.388e-05
[loop_unroll]: 0.00048244
[jit_opt_after_cconv]: 0.00016617, [1]
    [Cycle 1]: 0.00015857, [11]
        [c_1]: 2.221e-05
        [parameter_eliminate]: 3.78001e-06
        [updatestate_depend_eliminate]: 7.98999e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.64999e-06
        [cse]: 2.786e-05
        [call_graph_tuple_transform]: 2.323e-05
        [tuple_list_get_item_eliminator]: 6.10002e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 7.00005e-07
        [switch_simplify]: 5.72999e-06
[remove_dup_value]: 1.661e-05
[partial_unused_args_eliminate]: 2.59001e-06
[environ_conv]: 6.29001e-06
[add_recomputation]: 5.662e-05
[cse_after_recomputation]: 2.292e-05, [1]
    [Cycle 1]: 1.672e-05, [1]
        [cse]: 1.068e-05
[auto_monad_reorder]: 1.621e-05
[get_jit_bprop_graph]: 2.10002e-06
[rewriter_after_jit_bprop_graph]: 8.71997e-06
[opt_after_jit_grad]: 0.0005264
[symbol_engine_optimizer]: 8.512e-05, [1]
    [Cycle 1]: 7.823e-05, [6]
        [build]: 5.47999e-06
        [elim_shapecalc]: 8.67e-06
        [elim_not_effective]: 1.562e-05
        [opt_reshape]: 7.06999e-06
        [fold_const_symbol]: 1.014e-05
        [renormalize]: 9.50007e-07
[validate]: 4.834e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00705795
[execute]: 9.89001e-06
Sums
    bootstrap                            :     0.000334s :  1.91%
    type_inference                       :     0.005397s : 30.91%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000069s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.25%
    rewriter_before_opt_a                :     0.000077s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.44%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000730s :  4.18%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.001055s :  6.04%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.14%
    jit_opt_a.cse                        :     0.000043s :  0.25%
    jit_opt_a.replace_applicator         :     0.000024s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000767s :  4.39%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.10%
    cconv                                :     0.000034s :  0.19%
    loop_unroll                          :     0.000482s :  2.76%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000028s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000057s :  0.32%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000526s :  3.01%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000048s :  0.28%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007058s : 40.42%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000281    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.72% :     0.000008s :      3: substitution.graph_param_transform
86.47% :     0.000243s :      4: substitution.inline
 1.38% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.63% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000006s :      2: substitution.replace_old_param
 4.59% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005316     2
81.56% :     0.004336s :      1: type_inference.infer
18.44% :     0.000980s :      1: type_inference.specialize
------[replace.]   0.000080     6
68.79% :     0.000055s :      4: replace.inline
31.21% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000252     6
95.28% :     0.000240s :      4: match.inline
 4.72% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.56% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.64% :     0.000003s :     10: predicate.arithmetic_simplify
 1.12% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.18% :     0.000001s :     10: predicate.depend_value_elim
 1.28% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.17% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.47% :     0.000001s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.35% :     0.000000s :      3: predicate.graph_param_transform
 5.53% :     0.000007s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :      6: predicate.less_batch_normalization
 1.62% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.88% :     0.000002s :     15: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.64% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.31% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 2.43% :     0.000003s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.81% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.22% :     0.000002s :     10: predicate.reshape_eliminate
 1.08% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.07% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.11% :     0.000001s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.20% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.69% :     0.000010s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.26% :     0.000002s :     10: predicate.transpose_eliminate
 1.99% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.02% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.21% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.48% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.61% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.22% :     0.000002s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000772    12
47.46% :     0.000367s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.68% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.86% :     0.000347s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024144    76
 0.25% :     0.000060s :      1: add_recomputation
 0.30% :     0.000072s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.48% :     0.000358s :      1: bootstrap
 0.15% :     0.000037s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.40% :     0.006375s :      1: jit_opt_a
 0.70% :     0.000169s :      1: jit_opt_after_cconv
 0.25% :     0.000060s :      1: jit_opt_b
 2.04% :     0.000492s :      1: loop_unroll
 3.23% :     0.000780s :      1: mutable_eliminate
 4.00% :     0.000966s :     26: opt.transform.jit_opt_a
 0.22% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000030s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000038s :      4: opt.transform.symbol_engine_opt
 2.23% :     0.000537s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000046s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000019s :      1: remove_dup_value
 2.48% :     0.000598s :      1: renormalize.infer
 1.85% :     0.000447s :      1: renormalize.specialize
 0.05% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000042s :      1: rewriter_after_opt_a
 0.33% :     0.000081s :      1: rewriter_before_opt_a
 0.37% :     0.000088s :      1: symbol_engine_optimizer
29.32% :     0.007079s :      1: task_emit
22.43% :     0.005416s :      1: type_inference
 0.31% :     0.000075s :      1: validate


TotalTime = 0.0225358, [33]
[bootstrap]: 0.00037211
[type_inference]: 0.00516034
[event_method]: 1.799e-05
[auto_monad]: 4.81e-05
[graph_reusing]: 4.03999e-06
[pre_auto_parallel]: 1.74e-06
[py_interpret_to_execute]: 4.031e-05
[rewriter_before_opt_a]: 6.718e-05
[expand_dump_flag]: 1.91e-06
[jit_opt_a]: 0.00661969, [2]
    [Cycle 1]: 0.00219991, [27]
        [switch_simplify]: 6.861e-05
        [loop_unroll]: 3.002e-05
        [a_1]: 0.00069598
        [with_stream_mark]: 2.922e-05
        [recompute_prepare]: 8.88002e-06
        [updatestate_depend_eliminate]: 3.66999e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 2.76e-06
        [parameter_eliminate]: 2.46e-06
        [specialize_transform]: 6.31e-06
        [updatestate_useless_node_eliminater]: 5.67999e-06
        [accelerated_algorithm]: 7.51999e-06
        [meta_shard_fg_expand]: 1.72999e-06
        [get_grad_eliminate_]: 5.44e-06
        [merge_forward]: 3.55998e-06
        [cell_reuse_recompute_pass]: 2.68e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.756e-05
        [j_node_and_user_rematch]: 1.035e-05
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 1.105e-05
        [inline_without_move]: 5.39e-06
        [renormalize]: 0.00099193
        [add_forward_monad_depend]: 8.20999e-06
        [auto_monad_grad]: 2.64999e-06
        [auto_monad_eliminator]: 1.867e-05
        [cse]: 2.962e-05
        [replace_applicator]: 1.875e-05
    [Cycle 2]: 0.00034653, [27]
        [switch_simplify]: 6.61e-06
        [loop_unroll]: 6.11e-06
        [a_1]: 0.00010472
        [with_stream_mark]: 9.86e-06
        [recompute_prepare]: 5.27999e-06
        [updatestate_depend_eliminate]: 2.43998e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 1.55999e-06
        [specialize_transform]: 4.65999e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 5.05001e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 4.55999e-06
        [merge_forward]: 2.63e-06
        [cell_reuse_recompute_pass]: 3.04999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.465e-05
        [j_node_and_user_rematch]: 8.43001e-06
        [meta_fg_expand]: 1.42e-06
        [replace_old_param]: 8.48999e-06
        [inline_without_move]: 5.17999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.35001e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 5.47999e-06
        [cse]: 1.181e-05
        [replace_applicator]: 5.24998e-06
[py_interpret_to_execute_after_opt_a]: 1.782e-05
[rewriter_after_opt_a]: 4.176e-05
[convert_after_rewriter]: 8.59002e-06
[order_py_execute_after_rewriter]: 5.60001e-06
[mutable_eliminate]: 0.00080583
[jit_opt_b]: 5.912e-05, [1]
    [Cycle 1]: 4.965e-05, [2]
        [frontend_op_eliminate]: 1.98e-05
        [inline_after_opt_a]: 1.805e-05
[cconv]: 3.133e-05
[loop_unroll]: 0.00057329
[jit_opt_after_cconv]: 0.0001775, [1]
    [Cycle 1]: 0.00016831, [11]
        [c_1]: 2.425e-05
        [parameter_eliminate]: 6.02999e-06
        [updatestate_depend_eliminate]: 1.191e-05
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [cse]: 3.006e-05
        [call_graph_tuple_transform]: 2.56e-05
        [tuple_list_get_item_eliminator]: 4.85001e-06
        [none_parameter_eliminate]: 8.30012e-07
        [renormalize]: 8.39995e-07
        [switch_simplify]: 5.50001e-06
[remove_dup_value]: 1.526e-05
[partial_unused_args_eliminate]: 3.25e-06
[environ_conv]: 6.73e-06
[add_recomputation]: 5.964e-05
[cse_after_recomputation]: 0.00040211, [1]
    [Cycle 1]: 0.00038372, [1]
        [cse]: 1.136e-05
[auto_monad_reorder]: 2.276e-05
[get_jit_bprop_graph]: 2.09e-06
[rewriter_after_jit_bprop_graph]: 9.89999e-06
[opt_after_jit_grad]: 0.00064832
[symbol_engine_optimizer]: 8.492e-05, [1]
    [Cycle 1]: 7.671e-05, [6]
        [build]: 5.79e-06
        [elim_shapecalc]: 8.99e-06
        [elim_not_ef
TotalTime = 0.0232443, [33]
[bootstrap]: 0.00038153
[type_inference]: 0.0055715
[event_method]: 1.967e-05
[auto_monad]: 6.567e-05
[graph_reusing]: 6.33e-06
[pre_auto_parallel]: 2.66999e-06
[py_interpret_to_execute]: 4.315e-05
[rewriter_before_opt_a]: 7.57e-05
[expand_dump_flag]: 3.33998e-06
[jit_opt_a]: 0.0068526, [2]
    [Cycle 1]: 0.0022003, [27]
        [switch_simplify]: 6.685e-05
        [loop_unroll]: 3.068e-05
        [a_1]: 0.00070534
        [with_stream_mark]: 2.73e-05
        [recompute_prepare]: 9.00999e-06
        [updatestate_depend_eliminate]: 3.15002e-06
        [updatestate_assign_eliminate]: 2.86e-06
        [updatestate_loads_eliminate]: 2.76999e-06
        [parameter_eliminate]: 2.54999e-06
        [specialize_transform]: 6.38e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 6.19001e-06
        [meta_shard_fg_expand]: 2.74001e-06
        [get_grad_eliminate_]: 5.96e-06
        [merge_forward]: 3.36001e-06
        [cell_reuse_recompute_pass]: 1.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.739e-05
        [j_node_and_user_rematch]: 9.66e-06
        [meta_fg_expand]: 2.27001e-06
        [replace_old_param]: 1.081e-05
        [inline_without_move]: 5.49998e-06
        [renormalize]: 0.00099123
        [add_forward_monad_depend]: 7.94002e-06
        [auto_monad_grad]: 2.74999e-06
        [auto_monad_eliminator]: 1.823e-05
        [cse]: 2.937e-05
        [replace_applicator]: 1.84e-05
    [Cycle 2]: 0.0003447, [27]
        [switch_simplify]: 6.83e-06
        [loop_unroll]: 6.09999e-06
        [a_1]: 9.917e-05
        [with_stream_mark]: 1.295e-05
        [recompute_prepare]: 5.40001e-06
        [updatestate_depend_eliminate]: 3.10002e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 1.33002e-06
        [specialize_transform]: 5.04e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 4.95999e-06
        [meta_shard_fg_expand]: 1.74e-06
        [get_grad_eliminate_]: 4.31002e-06
        [merge_forward]: 2.96001e-06
        [cell_reuse_recompute_pass]: 2.69999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.605e-05
        [j_node_and_user_rematch]: 9.42001e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 7.21999e-06
        [inline_without_move]: 4.83001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.45999e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 5.89e-06
        [cse]: 9.89999e-06
        [replace_applicator]: 5.15001e-06
[py_interpret_to_execute_after_opt_a]: 1.756e-05
[rewriter_after_opt_a]: 4.157e-05
[convert_after_rewriter]: 6.53e-06
[order_py_execute_after_rewriter]: 5.04e-06
[mutable_eliminate]: 0.00080501
[jit_opt_b]: 5.59e-05, [1]
    [Cycle 1]: 4.66e-05, [2]
        [frontend_op_eliminate]: 1.786e-05
        [inline_after_opt_a]: 1.568e-05
[cconv]: 3.475e-05
[loop_unroll]: 0.00055167
[jit_opt_after_cconv]: 0.0001779, [1]
    [Cycle 1]: 0.00016921, [11]
        [c_1]: 2.397e-05
        [parameter_eliminate]: 5.96e-06
        [updatestate_depend_eliminate]: 1.11e-05
        [updatestate_assign_eliminate]: 2.84001e-06
        [updatestate_loads_eliminate]: 2.47001e-06
        [cse]: 3.106e-05
        [call_graph_tuple_transform]: 2.501e-05
        [tuple_list_get_item_eliminator]: 5.56e-06
        [none_parameter_eliminate]: 1.67001e-06
        [renormalize]: 8.50006e-07
        [switch_simplify]: 6.44001e-06
[remove_dup_value]: 1.496e-05
[partial_unused_args_eliminate]: 3.22002e-06
[environ_conv]: 6.24001e-06
[add_recomputation]: 5.949e-05
[cse_after_recomputation]: 8.722e-05, [1]
    [Cycle 1]: 7.975e-05, [1]
        [cse]: 1.089e-05
[auto_monad_reorder]: 1.804e-05
[get_jit_bprop_graph]: 2.22001e-06
[rewriter_after_jit_bprop_graph]: 1.218e-05
[opt_after_jit_grad]: 0.00059973
[symbol_engine_optimizer]: 8.484e-05, [1]
    [Cycle 1]: 7.698e-05, [6]
        [build]: 6.40002e-06
        [elim_shapecalc]: 8.40001e-fective]: 1.599e-05
        [opt_reshape]: 6.46999e-06
        [fold_const_symbol]: 9.04e-06
        [renormalize]: 6.80011e-07
[validate]: 3.748e-05
[backend_pass]: 7.39994e-07
[task_emit]: 0.00693894
[execute]: 1.007e-05
Sums
    bootstrap                            :     0.000372s :  2.14%
    type_inference                       :     0.005160s : 29.68%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000048s :  0.28%
    graph_reusing                        :     0.000004s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000040s :  0.23%
    rewriter_before_opt_a                :     0.000067s :  0.39%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000075s :  0.43%
    jit_opt_a.loop_unroll                :     0.000036s :  0.21%
    jit_opt_a.a_1                        :     0.000801s :  4.60%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000992s :  5.71%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.14%
    jit_opt_a.cse                        :     0.000041s :  0.24%
    jit_opt_a.replace_applicator         :     0.000024s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.10%
    rewriter_after_opt_a                 :     0.000042s :  0.24%
    convert_after_rewriter               :     0.000009s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000806s :  4.63%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.10%
    cconv                                :     0.000031s :  0.18%
    loop_unroll                          :     0.000573s :  3.30%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000030s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    par06
        [elim_not_effective]: 1.653e-05
        [opt_reshape]: 6.38003e-06
        [fold_const_symbol]: 9.27001e-06
        [renormalize]: 7.39994e-07
[validate]: 4.613e-05
[backend_pass]: 9.60019e-07
[task_emit]: 0.0072933
[execute]: 1.027e-05
Sums
    bootstrap                            :     0.000382s :  2.10%
    type_inference                       :     0.005572s : 30.73%
    event_method                         :     0.000020s :  0.11%
    auto_monad                           :     0.000066s :  0.36%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.24%
    rewriter_before_opt_a                :     0.000076s :  0.42%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.41%
    jit_opt_a.loop_unroll                :     0.000037s :  0.20%
    jit_opt_a.a_1                        :     0.000805s :  4.44%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000991s :  5.47%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.13%
    jit_opt_a.cse                        :     0.000039s :  0.22%
    jit_opt_a.replace_applicator         :     0.000024s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.10%
    rewriter_after_opt_a                 :     0.000042s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000805s :  4.44%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000035s :  0.19%
    loop_unroll                          :     0.000552s :  3.04%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000031s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0tial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000060s :  0.34%
    cse_after_recomputation.cse          :     0.000011s :  0.07%
    auto_monad_reorder                   :     0.000023s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.06%
    opt_after_jit_grad                   :     0.000648s :  3.73%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.22%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006939s : 39.91%
    execute                              :     0.000010s :  0.06%

.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000059s :  0.33%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000018s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000012s :  0.07%
    opt_after_jit_grad                   :     0.000600s :  3.31%
    symbol_engine_optimizer.build        :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007293s : 40.23%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000332    23
 0.62% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000002s :      2: substitution.fold_const_symbol
 1.98% :     0.000007s :      3: substitution.graph_param_transform
89.95% :     0.000299s :      4: substitution.inline
 1.00% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.68% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.53% :     0.000005s :      2: substitution.replace_old_param
 2.76% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005491     2
81.99% :     0.004502s :      1: type_inference.infer
18.01% :     0.000989s :      1: type_inference.specialize
------[replace.]   0.000090     6
67.97% :     0.000061s :      4: replace.inline
32.03% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000303     6
97.37% :     0.000295s :      4: match.inline
 2.63% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.56% :     0.000002s :     10: predicate.addn_zero_filter
 2.07% :     0.000003s :     10: predicate.arithmetic_simplify
 1.70% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.93% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000002s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.04% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.57% :     0.000001s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.00% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.67% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.29% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 6.66% :     0.000009s :     22: predicate.inline
 0.76% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.18% :     0.000002s :      6: predicate.less_batch_normalization
 1.68% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 2.66% :     0.000004s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.65% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 1.28% :     0.000002s :     10: predicate.merge_addn
 0.95% :     0.000001s :     10: predicate.minmaximum_grad
 2.96% :     0.000004s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 1.92% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.29% :     0.000002s :     10: predicate.reduce_eliminate
 1.36% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.03% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3:
Time group info:
------[substitution.]   0.000326    23
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.28% :     0.000001s :      2: substitution.fold_const_symbol
 2.19% :     0.000007s :      3: substitution.graph_param_transform
87.19% :     0.000284s :      4: substitution.inline
 1.08% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.02% :     0.000007s :      4: substitution.remove_not_recompute_node
 2.02% :     0.000007s :      2: substitution.replace_old_param
 4.46% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005093     2
81.29% :     0.004140s :      1: type_inference.infer
18.71% :     0.000953s :      1: type_inference.specialize
------[replace.]   0.000087     6
66.57% :     0.000058s :      4: replace.inline
33.43% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000294     6
95.49% :     0.000281s :      4: match.inline
 4.51% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.22% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 1.23% :     0.000002s :     10: predicate.addn_check_dump
 1.23% :     0.000002s :     10: predicate.addn_zero_filter
 2.21% :     0.000003s :     10: predicate.arithmetic_simplify
 1.15% :     0.000002s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.51% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.70% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.43% :     0.000002s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.83% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.67% :     0.000007s :     22: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.81% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.78% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.93% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 2.26% :     0.000003s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 1.84% :     0.000002s :     15: predicate.partial_eliminate
 1.06% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.39% :     0.000002s :     10: predicate.reduce_eliminate
 1.22% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.72% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.65% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.03% :     0.000001s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.18% :     0.000002s :      6: predicate.special_op_eliminate
 0.84% :     0.000001s :      6: predicate.specialize_transform
 1.95% :     0.000003s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.32% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.18% :     0.000003s :     16: predicate.switch_defer_inline
 1.83% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.25% :     0.000010s :     53: predicate.switch_simplify
 1.03% :     0.000001s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.22% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.28% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.47% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.47% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000737    12
49.43% :     0.000364s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.53% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.03% :     0.000339s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025381    76
 0.25% :     0.000063s :      1: add_recomputation
 0.27% :     0.000069s :      1: auto_monad
 0.09% :     0.000022s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 1.59% :     0.000404s :      1: bootstrap
 0.15% :     0.000038s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.35% :     0.000090s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000025s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
27.02% :     0.006857s :      1: jit_opt_a
 0.71% :     0.000181s :      1: jit_opt_after_cconv
 0.23% :     0.000058s :      1: jit_opt_b
 2.22% :     0.000564s :      1: loop_unroll
 3.23% :     0.000821s :      1: mutable_eliminate
 4.11% :     0.001044s :     26: opt.transform.jit_opt_a
 0.23% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000027s :      4: opt.transform.jit_opt_b
 0.07% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000025s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.40% :     0.000609s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000046s :      1: py_interpret_to_execute
 0.08% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.20% :     0.000557s :      1: renormalize.infer
 1.67% :     0.000425s :      1: renormalize.specialize
 0.06% :     0.000015s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000046s :      1: rewriter_after_opt_a
 0.31% :     0.000079s :      1: rewriter_before_opt_a
 0.35% :     0.000088s :      1: symbol_engine_optimizer
28.82% :     0.007315s :       predicate.row_tensor_eliminate
 1.03% :     0.000001s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.21% :     0.000002s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.06% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 1.97% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.70% :     0.000010s :     53: predicate.switch_simplify
 1.52% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.76% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.51% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.51% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.55% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.43% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000724    12
43.09% :     0.000312s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.48% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.43% :     0.000351s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024697    76
 0.25% :     0.000063s :      1: add_recomputation
 0.21% :     0.000052s :      1: auto_monad
 0.11% :     0.000027s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.57% :     0.000389s :      1: bootstrap
 0.14% :     0.000034s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 1.65% :     0.000407s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
26.82% :     0.006624s :      1: jit_opt_a
 0.73% :     0.000180s :      1: jit_opt_after_cconv
 0.25% :     0.000062s :      1: jit_opt_b
 2.37% :     0.000586s :      1: loop_unroll
 3.32% :     0.000821s :      1: mutable_eliminate
 4.22% :     0.001043s :     26: opt.transform.jit_opt_a
 0.23% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000031s :      4: opt.transform.jit_opt_b
 0.07% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.67% :     0.000659s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000044s :      1: py_interpret_to_execute
 0.08% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.23% :     0.000550s :      1: renormalize.infer
 1.75% :     0.000433s :      1: renormalize.specialize
 0.05% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000045s :      1: rewriter_after_opt_a
 0.28% :     0.000070s :      1: rewriter_before_opt_a
 0.36% :     0.000088s :      1: symbol_engine_optimizer
28.18% :     0.006959s :      1: task_emit
22.03% :     0.005592s :      1: type_inference
 0.29% :     0.000073s :      1: validate

1: task_emit
20.95% :     0.005174s :      1: type_inference
 0.23% :     0.000056s :      1: validate


TotalTime = 0.0146127, [30]
[bootstrap]: 0.00037771
[type_inference]: 0.00512521
[event_method]: 1.931e-05
[auto_monad]: 6.931e-05
[graph_reusing]: 5.72999e-06
[pre_auto_parallel]: 2.64999e-06
[py_interpret_to_execute]: 4.29e-05
[rewriter_before_opt_a]: 7.629e-05
[expand_dump_flag]: 3.65e-06
[jit_opt_a]: 0.00650381, [2]
    [Cycle 1]: 0.00207853, [27]
        [switch_simplify]: 6.378e-05
        [loop_unroll]: 3.055e-05
        [a_1]: 0.00060709
        [with_stream_mark]: 2.057e-05
        [recompute_prepare]: 8.32998e-06
        [updatestate_depend_eliminate]: 3.81999e-06
        [updatestate_assign_eliminate]: 3.21999e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [parameter_eliminate]: 1.93002e-06
        [specialize_transform]: 6.70002e-06
        [updatestate_useless_node_eliminater]: 5.37999e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 2.33998e-06
        [get_grad_eliminate_]: 5.74e-06
        [merge_forward]: 3.90998e-06
        [cell_reuse_recompute_pass]: 1.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.544e-05
        [j_node_and_user_rematch]: 1.018e-05
        [meta_fg_expand]: 2.22999e-06
        [replace_old_param]: 1.024e-05
        [inline_without_move]: 5.74999e-06
        [renormalize]: 0.00100717
        [add_forward_monad_depend]: 7.46999e-06
        [auto_monad_grad]: 2.29001e-06
        [auto_monad_eliminator]: 1.677e-05
        [cse]: 3.249e-05
        [replace_applicator]: 1.554e-05
    [Cycle 2]: 0.00033284, [27]
        [switch_simplify]: 6.13002e-06
        [loop_unroll]: 5.46e-06
        [a_1]: 9.394e-05
        [with_stream_mark]: 1.146e-05
        [recompute_prepare]: 4.88001e-06
        [updatestate_depend_eliminate]: 2.73e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 2.13998e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 4.95999e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 5.27001e-06
        [meta_shard_fg_expand]: 1.60001e-06
        [get_grad_eliminate_]: 5.22e-06
        [merge_forward]: 2.75997e-06
        [cell_reuse_recompute_pass]: 1.71002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.538e-05
        [j_node_and_user_rematch]: 8.3e-06
        [meta_fg_expand]: 2.16e-06
        [replace_old_param]: 7.87e-06
        [inline_without_move]: 4.73001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.36002e-06
        [auto_monad_grad]: 1.22999e-06
        [auto_monad_eliminator]: 5.64e-06
        [cse]: 1.067e-05
        [replace_applicator]: 5.09e-06
[py_interpret_to_execute_after_opt_a]: 1.256e-05
[rewriter_after_opt_a]: 3.979e-05
[convert_after_rewriter]: 8.14997e-06
[order_py_execute_after_rewriter]: 5.04e-06
[mutable_eliminate]: 0.00067655
[jit_opt_b]: 6.339e-05, [1]
    [Cycle 1]: 5.482e-05, [2]
        [frontend_op_eliminate]: 2.231e-05
        [inline_after_opt_a]: 1.865e-05
[cconv]: 3.094e-05
[loop_unroll]: 0.00046954
[jit_opt_after_cconv]: 0.0001564, [1]
    [Cycle 1]: 0.00014906, [11]
        [c_1]: 2.345e-05
        [parameter_eliminate]: 3.14001e-06
        [updatestate_depend_eliminate]: 7.21001e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.36998e-06
        [cse]: 2.465e-05
        [call_graph_tuple_transform]: 2.09e-05
        [tuple_list_get_item_eliminator]: 5.62999e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 9.39996e-07
        [switch_simplify]: 5.71e-06
[remove_dup_value]: 1.47e-05
[partial_unused_args_eliminate]: 2.28998e-06
[environ_conv]: 5.66e-06
[add_recomputation]: 5.247e-05
[cse_after_recomputation]: 2.089e-05, [1]
    [Cycle 1]: 1.498e-05, [1]
        [cse]: 9.25999e-06
[auto_monad_reorder]: 1.584e-05
[get_jit_bprop_graph]: 1.84e-06
[rewriter_after_jit_bprop_graph]: 6.02999e-06
[opt_after_jit_grad]: 0.00048582
[symbol_engine_optimizer]: 7.777e-05, [1]
    [Cycle 1]: 7.061e-05, [6]
        [build]: 4.35999e-06
        [elim_shapecalc]: 7.82998e-06
        [elim_not_effective]: 1.412e-05
        [opt_reshape]: 6.30002e-06
        [fold_const_symbol]: 8.74998e-06
        [renormalize]: 4.99975e-07
[validate]: 4.482e-05
Sums
    bootstrap                            :     0.000378s :  3.81%
    type_inference                       :     0.005125s : 51.75%
    event_method                         :     0.000019s :  0.19%
    auto_monad                           :     0.000069s :  0.70%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000043s :  0.43%
    rewriter_before_opt_a                :     0.000076s :  0.77%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000070s :  0.71%
    jit_opt_a.loop_unroll                :     0.000036s :  0.36%
    jit_opt_a.a_1                        :     0.000701s :  7.08%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.18%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.001007s : 10.17%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.23%
    jit_opt_a.cse                        :     0.000043s :  0.44%
    jit_opt_a.replace_applicator         :     0.000021s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.13%
    rewriter_after_opt_a                 :     0.000040s :  0.40%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000677s :  6.83%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.19%
    cconv                                :     0.000031s :  0.31%
    loop_unroll                          :     0.000470s :  4.74%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000025s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000015s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000052s :  0.53%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000486s :  4.91%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000045s :  0.45%


Time group info:
------[substitution.]   0.000251    23
 1.07% :     0.000003s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.62% :     0.000007s :      3: substitution.graph_param_transform
85.06% :     0.000214s :      4: substitution.inline
 1.33% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.82% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.91% :     0.000005s :      2: substitution.replace_old_param
 5.62% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005048     2
79.88% :     0.004033s :      1: type_inference.infer
20.12% :     0.001016s :      1: type_inference.specialize
------[replace.]   0.000084     6
65.57% :     0.000055s :      4: replace.inline
34.43% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000224     6
94.18% :     0.000211s :      4: match.inline
 5.82% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.69% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.96% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.23% :     0.000002s :     10: predicate.addn_zero_filter
 1.88% :     0.000002s :     10: predicate.arithmetic_simplify
 1.09% :     0.000001s :     10: predicate.cast_eliminate
 0.55% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.08% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.51% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.36% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.35% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.65% :     0.000002s :      3: predicate.mutable_eliminate
 0.62% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.41% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.57% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.49% :     0.000002s :      6: predicate.special_op_eliminate
 1.19% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.63% :     0.000003s :     16: predicate.switch_defer_inline
 1.92% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.28% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.53% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.62% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.07% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.63% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.84% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000717    12
49.77% :     0.000357s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.76% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.47% :     0.000319s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016605    72
 0.34% :     0.000056s :      1: add_recomputation
 0.44% :     0.000073s :      1: auto_monad
 0.11% :     0.000019s :      1: auto_monad_reorder
 2.41% :     0.000400s :      1: bootstrap
 0.20% :     0.000034s :      1: cconv
 0.07% :     0.000011s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.20% :     0.006509s :      1: jit_opt_a
 0.96% :     0.000159s :      1: jit_opt_after_cconv
 0.40% :     0.000066s :      1: jit_opt_b
 2.88% :     0.000478s :      1: loop_unroll
 4.14% :     0.000688s :      1: mutable_eliminate
 5.61% :     0.000931s :     26: opt.transform.jit_opt_a
 0.31% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000032s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.98% :     0.000495s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000046s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.92% :     0.000484s :      1: renormalize.infer
 3.09% :     0.000513s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000043s :      1: rewriter_after_opt_a
 0.48% :     0.000079s :      1: rewriter_before_opt_a
 0.49% :     0.000081s :      1: symbol_engine_optimizer
30.96% :     0.005141s :      1: type_inference


TotalTime = 0.0302403, [33]
[bootstrap]: 0.00031532
[type_inference]: 0.00490051
[event_method]: 1.783e-05
[auto_monad]: 6.527e-05
[graph_reusing]: 5.92999e-06
[pre_auto_parallel]: 2.09999e-06
[py_interpret_to_execute]: 3.446e-05
[rewriter_before_opt_a]: 7.149e-05
[expand_dump_flag]: 3.06999e-06
[jit_opt_a]: 0.00594243, [2]
    [Cycle 1]: 0.00196209, [27]
        [switch_simplify]: 5.662e-05
        [loop_unroll]: 2.853e-05
        [a_1]: 0.0005731
        [with_stream_mark]: 2.017e-05
        [recompute_prepare]: 6.88e-06
        [updatestate_depend_eliminate]: 3.29001e-06
        [updatestate_assign_eliminate]: 2.74001e-06
        [updatestate_loads_eliminate]: 2.99999e-06
        [parameter_eliminate]: 1.40001e-06
        [specialize_transform]: 5.82999e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.57001e-06
        [meta_shard_fg_expand]: 2.22001e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.399e-05
        [j_node_and_user_rematch]: 9.19998e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 1.003e-05
        [inline_without_move]: 5.71003e-06
        [renormalize]: 0.00095481
        [add_forward_monad_depend]: 7.1e-06
        [auto_monad_grad]: 2.66e-06
        [auto_monad_eliminator]: 1.863e-05
        [cse]: 2.648e-05
        [replace_applicator]: 1.514e-05
    [Cycle 2]: 0.00033073, [27]
        [switch_simplify]: 6.69999e-06
        [loop_unroll]: 6.29999e-06
        [a_1]: 9.116e-05
        [with_stream_mark]: 1.077e-05
        [recompute_prepare]: 4.82998e-06
        [updatestate_depend_eliminate]: 2.39999e-06
        [updatestate_assign_eliminate]: 2.19999e-06
        [updatestate_loads_eliminate]: 2.19999e-06
        [parameter_eliminate]: 1.49998e-06
        [specialize_transform]: 5.49e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 5.27001e-06
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.55999e-06
        [merge_forward]: 2.86e-06
        [cell_reuse_recompute_pass]: 3.54002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.42e-05
        [j_node_and_user_rematch]: 7.83001e-06
        [meta_fg_expand]: 1.97001e-06
        [replace_old_param]: 7.61001e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 5.55001e-06
        [cse]: 1.001e-05
        [replace_applicator]: 5.29e-06
[py_interpret_to_execute_after_opt_a]: 1.111e-05
[rewriter_after_opt_a]: 3.741e-05
[convert_after_rewriter]: 9.92999e-06
[order_py_execute_after_rewriter]: 4.80001e-06
[mutable_eliminate]: 0.00091164
[jit_opt_b]: 5.384e-05, [1]
    [Cycle 1]: 4.505e-05, [2]
        [frontend_op_eliminate]: 1.74e-05
        [inline_after_opt_a]: 1.571e-05
[cconv]: 3.44e-05
[loop_unroll]: 0.00046622
[jit_opt_after_cconv]: 0.000159, [1]
    [Cycle 1]: 0.00015235, [11]
        [c_1]: 2.145e-05
        [parameter_eliminate]: 4.64998e-06
        [updatestate_depend_eliminate]: 8.90001e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [cse]: 2.7e-05
        [call_graph_tuple_transform]: 2.199e-05
        [tuple_list_get_item_eliminator]: 5.45001e-06
        [none_parameter_eliminate]: 1.49998e-06
        [renormalize]: 1.09e-06
        [switch_simplify]: 5.38002e-06
[remove_dup_value]: 1.444e-05
[partial_unused_args_eliminate]: 2.87002e-06
[environ_conv]: 6.49001e-06
[add_recomputation]: 5.379e-05
[cse_after_recomputation]: 2.055e-05, [1]
    [Cycle 1]: 1.503e-05, [1]
        [cse]: 9.08002e-06
[auto_monad_reorder]: 1.528e-05
[get_jit_bprop_graph]: 2.34999e-06
[rewriter_after_jit_bprop_graph]: 7.63999e-06
[opt_after_jit_grad]: 0.000477
[symbol_engine_optimizer]: 7.639e-05, [1]
    [Cycle 1]: 6.91e-05, [6]
        [build]: 4.72998e-06
        [elim_shapecalc]: 8.32e-06
        [elim_not_effective]: 1.39e-05
        [opt_reshape]: 5.62999e-06
        [fold_const_symbol]: 8.75001e-06
        [renormalize]: 3.9002e-07
[validate]: 4.501e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.016238
[execute]: 1.047e-05
Sums
    bootstrap                            :     0.000315s :  1.21%
    type_inference                       :     0.004901s : 18.88%
    event_method                         :     0.000018s :  0.07%
    auto_monad                           :     0.000065s :  0.25%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000034s :  0.13%
    rewriter_before_opt_a                :     0.000071s :  0.28%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000063s :  0.24%
    jit_opt_a.loop_unroll                :     0.000035s :  0.13%
    jit_opt_a.a_1                        :     0.000664s :  2.56%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.07%
    jit_opt_a.inline_without_move        :     0.000010s :  0.04%
    jit_opt_a.renormalize                :     0.000955s :  3.68%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.09%
    jit_opt_a.cse                        :     0.000036s :  0.14%
    jit_opt_a.replace_applicator         :     0.000020s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.04%
    rewriter_after_opt_a                 :     0.000037s :  0.14%
    convert_after_rewriter               :     0.000010s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000912s :  3.51%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.06%
    cconv                                :     0.000034s :  0.13%
    loop_unroll                          :     0.000466s :  1.80%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000014s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000054s :  0.21%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000015s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.03%
    opt_after_jit_grad                   :     0.000477s :  1.84%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000045s :  0.17%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.016238s : 62.56%
    execute                              :     0.000010s :  0.04%


Time group info:
------[substitution.]   0.000240    23
 0.90% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.53% :     0.000006s :      3: substitution.graph_param_transform
85.31% :     0.000205s :      4: substitution.inline
 1.37% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.86% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.90% :     0.000005s :      2: substitution.replace_old_param
 5.58% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004835     2
81.59% :     0.003945s :      1: type_inference.infer
18.41% :     0.000890s :      1: type_inference.specialize
------[replace.]   0.000075     6
68.53% :     0.000052s :      4: replace.inline
31.47% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000214     6
94.26% :     0.000202s :      4: match.inline
 5.74% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.52% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.39% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.73% :     0.000002s :     10: predicate.addn_zero_filter
 1.69% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.47% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.49% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.67% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.48% :     0.000002s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.34% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 2.12% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000000s :      3: predicate.opt_reshape
 2.10% :     0.000002s :     15: predicate.partial_eliminate
 1.23% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 1.20% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.39% :     0.000008s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.39% :     0.000002s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.27% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000660    12
46.20% :     0.000305s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.52% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.28% :     0.000305s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.032184    76
 0.18% :     0.000057s :      1: add_recomputation
 0.21% :     0.000069s :      1: auto_monad
 0.06% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.02% :     0.000329s :      1: bootstrap
 0.11% :     0.000037s :      1: cconv
 0.04% :     0.000012s :      1: convert_after_rewriter
 0.07% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.07% :     0.000021s :      1: event_method
 0.05% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
18.48% :     0.005946s :      1: jit_opt_a
 0.50% :     0.000162s :      1: jit_opt_after_cconv
 0.18% :     0.000057s :      1: jit_opt_b
 1.48% :     0.000475s :      1: loop_unroll
 2.88% :     0.000928s :      1: mutable_eliminate
 2.73% :     0.000879s :     26: opt.transform.jit_opt_a
 0.16% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000026s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.51% :     0.000485s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000037s :      1: py_interpret_to_execute
 0.04% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000017s :      1: remove_dup_value
 1.55% :     0.000500s :      1: renormalize.infer
 1.39% :     0.000447s :      1: renormalize.specialize
 0.03% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000040s :      1: rewriter_after_opt_a
 0.23% :     0.000075s :      1: rewriter_before_opt_a
 0.25% :     0.000079s :      1: symbol_engine_optimizer
50.52% :     0.016259s :      1: task_emit
15.27% :     0.004914s :      1: type_inference
 0.21% :     0.000069s :      1: validate


TotalTime = 0.068216, [33]
[bootstrap]: 0.00037295
[type_inference]: 0.00539392
[event_method]: 1.892e-05
[auto_monad]: 6.473e-05
[graph_reusing]: 5.61e-06
[pre_auto_parallel]: 2.16998e-06
[py_interpret_to_execute]: 4.261e-05
[rewriter_before_opt_a]: 7.327e-05
[expand_dump_flag]: 3.38999e-06
[jit_opt_a]: 0.00817107, [2]
    [Cycle 1]: 0.00185099, [27]
        [switch_simplify]: 5.932e-05
        [loop_unroll]: 2.881e-05
        [a_1]: 0.00055434
        [with_stream_mark]: 1.804e-05
        [recompute_prepare]: 7.21001e-06
        [updatestate_depend_eliminate]: 3.15998e-06
        [updatestate_assign_eliminate]: 2.14999e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.47001e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 5.07999e-06
        [accelerated_algorithm]: 5.49998e-06
        [meta_shard_fg_expand]: 1.59998e-06
        [get_grad_eliminate_]: 5.42999e-06
        [merge_forward]: 3.65998e-06
        [cell_reuse_recompute_pass]: 1.97999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.158e-05
        [j_node_and_user_rematch]: 9.31998e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 8.13001e-06
        [inline_without_move]: 5.08002e-06
        [renormalize]: 0.00088809
        [add_forward_monad_depend]: 5.72999e-06
        [auto_monad_grad]: 1.97999e-06
        [auto_monad_eliminator]: 1.39e-05
        [cse]: 2.026e-05
        [replace_applicator]: 1.265e-05
    [Cycle 2]: 0.00222438, [27]
        [switch_simplify]: 6.44001e-06
        [loop_unroll]: 4.85001e-06
        [a_1]: 8.949e-05
        [with_stream_mark]: 8.85999e-06
        [recompute_prepare]: 5.19998e-06
        [updatestate_depend_eliminate]: 1.79e-05
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.86998e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 5.67999e-06
        [updatestate_useless_node_eliminater]: 4.69002e-06
        [accelerated_algorithm]: 5.10999e-06
        [meta_shard_fg_expand]: 1.17999e-06
        [get_grad_eliminate_]: 4.48999e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 2.90998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.286e-05
        [j_node_and_user_rematch]: 7.93001e-06
        [meta_fg_expand]: 1.72999e-06
        [replace_old_param]: 0.00184304
        [inline_without_move]: 9.09e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 3.56999e-06
        [auto_monad_grad]: 1.73997e-06
        [auto_monad_eliminator]: 1.256e-05
        [cse]: 1.915e-05
        [replace_applicator]: 1.278e-05
[py_interpret_to_execute_after_opt_a]: 1.49e-05
[rewriter_after_opt_a]: 3.918e-05
[convert_after_rewriter]: 7.68999e-06
[order_py_execute_after_rewriter]: 5.25001e-06
[mutable_eliminate]: 0.00094802
[jit_opt_b]: 7.103e-05, [1]
    [Cycle 1]: 6.162e-05, [2]
        [frontend_op_eliminate]: 3.077e-05
        [inline_after_opt_a]: 1.858e-05
[cconv]: 3.005e-05
[loop_unroll]: 0.00048795
[jit_opt_after_cconv]: 0.00015429, [1]
    [Cycle 1]: 0.00014774, [11]
        [c_1]: 2.326e-05
        [parameter_eliminate]: 4.80999e-06
        [updatestate_depend_eliminate]: 6.71e-06
        [updatestate_assign_eliminate]: 3.48999e-06
        [updatestate_loads_eliminate]: 3.09999e-06
        [cse]: 2.292e-05
        [call_graph_tuple_transform]: 1.895e-05
        [tuple_list_get_item_eliminator]: 5.17e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 6.11e-06
[remove_dup_value]: 1.563e-05
[partial_unused_args_eliminate]: 2.09e-06
[environ_conv]: 6.52001e-06
[add_recomputation]: 4.913e-05
[cse_after_recomputation]: 3.608e-05, [1]
    [Cycle 1]: 3.041e-05, [1]
        [cse]: 2.4e-05
[auto_monad_reorder]: 1.592e-05
[get_jit_bprop_graph]: 1.93002e-06
[rewriter_after_jit_bprop_graph]: 6.91999e-06
[opt_after_jit_grad]: 0.00047608
[symbol_engine_optimizer]: 7.662e-05, [1]
    [Cycle 1]: 6.945e-05, [6]
        [build]: 3.61999e-06
        [elim_shapecalc]: 8.17e-06
        [elim_not_effective]: 1.392e-05
        [opt_reshape]: 6.42001e-06
        [fold_const_symbol]: 9.60001e-06
        [renormalize]: 4.69998e-07
[validate]: 3.814e-05
[backend_pass]: 1.09e-06
[task_emit]: 0.0512649
[execute]: 1.312e-05
Sums
    bootstrap                            :     0.000373s :  0.59%
    type_inference                       :     0.005394s :  8.51%
    event_method                         :     0.000019s :  0.03%
    auto_monad                           :     0.000065s :  0.10%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.00%
    py_interpret_to_execute              :     0.000043s :  0.07%
    rewriter_before_opt_a                :     0.000073s :  0.12%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000066s :  0.10%
    jit_opt_a.loop_unroll                :     0.000034s :  0.05%
    jit_opt_a.a_1                        :     0.000644s :  1.02%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.04%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000021s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.00%
    jit_opt_a.specialize_transform       :     0.000012s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.00%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.02%
    jit_opt_a.merge_forward              :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.04%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.001851s :  2.92%
    jit_opt_a.inline_without_move        :     0.000014s :  0.02%
    jit_opt_a.renormalize                :     0.000888s :  1.40%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.04%
    jit_opt_a.cse                        :     0.000039s :  0.06%
    jit_opt_a.replace_applicator         :     0.000025s :  0.04%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.02%
    rewriter_after_opt_a                 :     0.000039s :  0.06%
    convert_after_rewriter               :     0.000008s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000948s :  1.50%
    jit_opt_b.frontend_op_eliminate      :     0.000031s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.03%
    cconv                                :     0.000030s :  0.05%
    loop_unroll                          :     0.000488s :  0.77%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.04%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.cse              :     0.000023s :  0.04%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.01%
    remove_dup_value                     :     0.000016s :  0.02%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000007s :  0.01%
    add_recomputation                    :     0.000049s :  0.08%
    cse_after_recomputation.cse          :     0.000024s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.01%
    opt_after_jit_grad                   :     0.000476s :  0.75%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.02%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.06%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.051265s : 80.88%
    execute                              :     0.000013s :  0.02%


Time group info:
------[substitution.]   0.000220    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.91% :     0.000002s :      2: substitution.fold_const_symbol
 2.38% :     0.000005s :      3: substitution.graph_param_transform
84.10% :     0.000185s :      4: substitution.inline
 1.42% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.97% :     0.000004s :      4: substitution.remove_not_recompute_node
 3.19% :     0.000007s :      2: substitution.replace_old_param
 5.00% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005312     2
79.12% :     0.004203s :      1: type_inference.infer
20.88% :     0.001109s :      1: type_inference.specialize
------[replace.]   0.000076     6
68.77% :     0.000053s :      4: replace.inline
31.23% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     6
94.87% :     0.000182s :      4: match.inline
 5.13% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.59% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.54% :     0.000002s :     10: predicate.addn_zero_filter
 2.14% :     0.000003s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.64% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.42% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.17% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.25% :     0.000006s :     22: predicate.inline
 1.50% :     0.000002s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.61% :     0.000002s :     15: predicate.load_eliminater
 1.49% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.67% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.23% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.00% :     0.000002s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.35% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.51% :     0.000001s :      3: predicate.reset_defer_inline
 1.32% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.28% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.65% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.55% :     0.000002s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.48% :     0.000009s :     53: predicate.switch_simplify
 1.29% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.16% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000661    12
45.93% :     0.000304s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.98% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.09% :     0.000324s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.071923    76
 0.07% :     0.000052s :      1: add_recomputation
 0.09% :     0.000068s :      1: auto_monad
 0.03% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.55% :     0.000393s :      1: bootstrap
 0.05% :     0.000033s :      1: cconv
 0.01% :     0.000010s :      1: convert_after_rewriter
 0.05% :     0.000038s :      1: cse_after_recomputation
 0.01% :     0.000009s :      1: environ_conv
 0.03% :     0.000023s :      1: event_method
 0.03% :     0.000021s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000008s :      1: graph_reusing
11.37% :     0.008175s :      1: jit_opt_a
 0.22% :     0.000157s :      1: jit_opt_after_cconv
 0.10% :     0.000074s :      1: jit_opt_b
 0.69% :     0.000496s :      1: loop_unroll
 1.33% :     0.000959s :      1: mutable_eliminate
 3.76% :     0.002702s :     26: opt.transform.jit_opt_a
 0.07% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000040s :      4: opt.transform.jit_opt_b
 0.02% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.05% :     0.000035s :      4: opt.transform.symbol_engine_opt
 0.67% :     0.000484s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.06% :     0.000045s :      1: py_interpret_to_execute
 0.02% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000018s :      1: remove_dup_value
 0.68% :     0.000488s :      1: renormalize.infer
 0.54% :     0.000392s :      1: renormalize.specialize
 0.01% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.06% :     0.000042s :      1: rewriter_after_opt_a
 0.11% :     0.000076s :      1: rewriter_before_opt_a
 0.11% :     0.000079s :      1: symbol_engine_optimizer
71.40% :     0.051356s :      1: task_emit
 7.52% :     0.005411s :      1: type_inference
 0.08% :     0.000059s :      1: validate


TotalTime = 0.018614, [30]
[bootstrap]: 0.0003844
[type_inference]: 0.00518159
[event_method]: 1.988e-05
[auto_monad]: 7.002e-05
[graph_reusing]: 5.56e-06
[pre_auto_parallel]: 2.89999e-06
[py_interpret_to_execute]: 4.034e-05
[rewriter_before_opt_a]: 7.722e-05
[expand_dump_flag]: 3.56999e-06
[jit_opt_a]: 0.0102577, [2]
    [Cycle 1]: 0.00240218, [27]
        [switch_simplify]: 7.052e-05
        [loop_unroll]: 3.164e-05
        [a_1]: 0.00064929
        [with_stream_mark]: 2.602e-05
        [recompute_prepare]: 7.68001e-06
        [updatestate_depend_eliminate]: 4.27998e-06
        [updatestate_assign_eliminate]: 3.33e-06
        [updatestate_loads_eliminate]: 2.47001e-06
        [parameter_eliminate]: 2.29999e-06
        [specialize_transform]: 6.24001e-06
        [updatestate_useless_node_eliminater]: 5.25001e-06
        [accelerated_algorithm]: 5.97001e-06
        [meta_shard_fg_expand]: 2.17999e-06
        [get_grad_eliminate_]: 6.52001e-06
        [merge_forward]: 3.56001e-06
        [cell_reuse_recompute_pass]: 1.72001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.568e-05
        [j_node_and_user_rematch]: 1.046e-05
        [meta_fg_expand]: 2.86e-06
        [replace_old_param]: 1.068e-05
        [inline_without_move]: 5.69e-06
        [renormalize]: 0.00125136
        [add_forward_monad_depend]: 8.37998e-06
        [auto_monad_grad]: 3.04999e-06
        [auto_monad_eliminator]: 1.679e-05
        [cse]: 3.31e-05
        [replace_applicator]: 1.69e-05
    [Cycle 2]: 0.00035254, [27]
        [switch_simplify]: 5.97999e-06
        [loop_unroll]: 5.10999e-06
        [a_1]: 9.932e-05
        [with_stream_mark]: 1.386e-05
        [recompute_prepare]: 5.35999e-06
        [updatestate_depend_eliminate]: 2.84999e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [parameter_eliminate]: 1.49e-06
        [specialize_transform]: 5.10001e-06
        [updatestate_useless_node_eliminater]: 5.10999e-06
        [accelerated_algorithm]: 5.25001e-06
        [meta_shard_fg_expand]: 1.84998e-06
        [get_grad_eliminate_]: 4.96997e-06
        [merge_forward]: 2.93e-06
        [cell_reuse_recompute_pass]: 2.51e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.504e-05
        [j_node_and_user_rematch]: 8.72e-06
        [meta_fg_expand]: 1.88002e-06
        [replace_old_param]: 8.52e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.48002e-06
        [auto_monad_grad]: 9.99979e-07
        [auto_monad_eliminator]: 5.29e-06
        [cse]: 1.08e-05
        [replace_applicator]: 5.72999e-06
[py_interpret_to_execute_after_opt_a]: 1.574e-05
[rewriter_after_opt_a]: 4.072e-05
[convert_after_rewriter]: 7.43e-06
[order_py_execute_after_rewriter]: 4.87e-06
[mutable_eliminate]: 0.00078233
[jit_opt_b]: 5.577e-05, [1]
    [Cycle 1]: 4.725e-05, [2]
        [frontend_op_eliminate]: 1.915e-05
        [inline_after_opt_a]: 1.633e-05
[cconv]: 3.142e-05
[loop_unroll]: 0.00054342
[jit_opt_after_cconv]: 0.00015916, [1]
    [Cycle 1]: 0.00015234, [11]
        [c_1]: 2.3e-05
        [parameter_eliminate]: 4e-06
        [updatestate_depend_eliminate]: 7.56001e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [cse]: 2.394e-05
        [call_graph_tuple_transform]: 2.194e-05
        [tuple_list_get_item_eliminator]: 5.44e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 8.59989e-07
        [switch_simplify]: 5.73002e-06
[remove_dup_value]: 1.581e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 6.26998e-06
[add_recomputation]: 5.529e-05
[cse_after_recomputation]: 2.226e-05, [1]
    [Cycle 1]: 1.656e-05, [1]
        [cse]: 9.72001e-06
[auto_monad_reorder]: 1.588e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 7.41999e-06
[opt_after_jit_grad]: 0.00047523
[symbol_engine_optimizer]: 7.829e-05, [1]
    [Cycle 1]: 7.118e-05, [6]
        [build]: 4.66002e-06
        [elim_shapecalc]: 8.1e-06
        [elim_not_effective]: 1.396e-05
        [opt_reshape]: 5.89999e-06
        [fold_const_symbol]: 8.96998e-06
        [renormalize]: 5.10016e-07
[validate]: 4.128e-05
Sums
    bootstrap                            :     0.000384s :  3.68%
    type_inference                       :     0.005182s : 49.57%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000070s :  0.67%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000040s :  0.39%
    rewriter_before_opt_a                :     0.000077s :  0.74%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000077s :  0.73%
    jit_opt_a.loop_unroll                :     0.000037s :  0.35%
    jit_opt_a.a_1                        :     0.000749s :  7.16%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.38%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000019s :  0.18%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.001251s : 11.97%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.21%
    jit_opt_a.cse                        :     0.000044s :  0.42%
    jit_opt_a.replace_applicator         :     0.000023s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.15%
    rewriter_after_opt_a                 :     0.000041s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000782s :  7.48%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.16%
    cconv                                :     0.000031s :  0.30%
    loop_unroll                          :     0.000543s :  5.20%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000016s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000055s :  0.53%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000475s :  4.55%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.39%


Time group info:
------[substitution.]   0.000284    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 2.09% :     0.000006s :      3: substitution.graph_param_transform
86.78% :     0.000247s :      4: substitution.inline
 1.51% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.66% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000006s :      2: substitution.replace_old_param
 4.66% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005106     2
80.86% :     0.004129s :      1: type_inference.infer
19.14% :     0.000977s :      1: type_inference.specialize
------[replace.]   0.000083     6
66.99% :     0.000055s :      4: replace.inline
33.01% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000256     6
95.27% :     0.000244s :      4: match.inline
 4.73% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.20% :     0.000002s :     10: predicate.addn_zero_filter
 2.76% :     0.000004s :     10: predicate.arithmetic_simplify
 1.07% :     0.000001s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.10% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.72% :     0.000008s :     22: predicate.inline
 1.10% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.84% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.80% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 1.64% :     0.000002s :      3: predicate.mutable_eliminate
 0.69% :     0.000001s :      3: predicate.opt_reshape
 2.03% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.18% :     0.000003s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.55% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.58% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.22% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000010s :     53: predicate.switch_simplify
 1.31% :     0.000002s :     10: predicate.tile_eliminate
 0.98% :     0.000001s :     10: predicate.transpose_eliminate
 1.93% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.59% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.98% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.84% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 1.10% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000745    12
48.18% :     0.000359s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.87% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.95% :     0.000328s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020895    72
 0.28% :     0.000058s :      1: add_recomputation
 0.35% :     0.000073s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 1.95% :     0.000408s :      1: bootstrap
 0.17% :     0.000035s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.12% :     0.000025s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
49.11% :     0.010262s :      1: jit_opt_a
 0.78% :     0.000162s :      1: jit_opt_after_cconv
 0.28% :     0.000059s :      1: jit_opt_b
 2.65% :     0.000553s :      1: loop_unroll
 3.80% :     0.000794s :      1: mutable_eliminate
 4.74% :     0.000990s :     26: opt.transform.jit_opt_a
 0.25% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000028s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.31% :     0.000483s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000043s :      1: py_interpret_to_execute
 0.09% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000018s :      1: remove_dup_value
 3.79% :     0.000792s :      1: renormalize.infer
 2.14% :     0.000447s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000044s :      1: rewriter_after_opt_a
 0.38% :     0.000080s :      1: rewriter_before_opt_a
 0.39% :     0.000081s :      1: symbol_engine_optimizer
24.88% :     0.005199s :      1: type_inference


TotalTime = 0.0436765, [33]
[bootstrap]: 0.00042209
[type_inference]: 0.00497216
[event_method]: 1.827e-05
[auto_monad]: 6.437e-05
[graph_reusing]: 4.72e-06
[pre_auto_parallel]: 2.18002e-06
[py_interpret_to_execute]: 3.536e-05
[rewriter_before_opt_a]: 6.992e-05
[expand_dump_flag]: 3.19001e-06
[jit_opt_a]: 0.017276, [2]
    [Cycle 1]: 0.00212182, [27]
        [switch_simplify]: 7.367e-05
        [loop_unroll]: 2.933e-05
        [a_1]: 0.00066133
        [with_stream_mark]: 2.602e-05
        [recompute_prepare]: 7.58999e-06
        [updatestate_depend_eliminate]: 4.12998e-06
        [updatestate_assign_eliminate]: 3.18e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 2.02999e-06
        [specialize_transform]: 6.12999e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.72001e-06
        [meta_shard_fg_expand]: 2.42001e-06
        [get_grad_eliminate_]: 5.87999e-06
        [merge_forward]: 4.52e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.486e-05
        [j_node_and_user_rematch]: 1.008e-05
        [meta_fg_expand]: 2.94001e-06
        [replace_old_param]: 9.97001e-06
        [inline_without_move]: 5.68002e-06
        [renormalize]: 0.00096239
        [add_forward_monad_depend]: 6.07999e-06
        [auto_monad_grad]: 2.61e-06
        [auto_monad_eliminator]: 1.58e-05
        [cse]: 3.352e-05
        [replace_applicator]: 1.461e-05
    [Cycle 2]: 0.0003365, [27]
        [switch_simplify]: 6.41e-06
        [loop_unroll]: 5.39e-06
        [a_1]: 9.364e-05
        [with_stream_mark]: 1.213e-05
        [recompute_prepare]: 5.28002e-06
        [updatestate_depend_eliminate]: 2.96001e-06
        [updatestate_assign_eliminate]: 2.39999e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 1.22999e-06
        [specialize_transform]: 4.90001e-06
        [updatestate_useless_node_eliminater]: 4.72998e-06
        [accelerated_algorithm]: 5.21002e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 4.62e-06
        [merge_forward]: 2.66e-06
        [cell_reuse_recompute_pass]: 1.90001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.439e-05
        [j_node_and_user_rematch]: 8.55001e-06
        [meta_fg_expand]: 1.77999e-06
        [replace_old_param]: 7.56999e-06
        [inline_without_move]: 4.63001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 6.60017e-07
        [auto_monad_eliminator]: 5.13002e-06
        [cse]: 1.084e-05
        [replace_applicator]: 4.99e-06
[py_interpret_to_execute_after_opt_a]: 1.216e-05
[rewriter_after_opt_a]: 3.994e-05
[convert_after_rewriter]: 6.83998e-06
[order_py_execute_after_rewriter]: 4.90001e-06
[mutable_eliminate]: 0.00074571
[jit_opt_b]: 6.18e-05, [1]
    [Cycle 1]: 5.34e-05, [2]
        [frontend_op_eliminate]: 2.314e-05
        [inline_after_opt_a]: 1.766e-05
[cconv]: 3.219e-05
[loop_unroll]: 0.00058431
[jit_opt_after_cconv]: 0.00016337, [1]
    [Cycle 1]: 0.00015647, [11]
        [c_1]: 2.377e-05
        [parameter_eliminate]: 3.91999e-06
        [updatestate_depend_eliminate]: 8.64e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [cse]: 2.855e-05
        [call_graph_tuple_transform]: 2.208e-05
        [tuple_list_get_item_eliminator]: 5.21002e-06
        [none_parameter_eliminate]: 1.49998e-06
        [renormalize]: 7.89994e-07
        [switch_simplify]: 5.57001e-06
[remove_dup_value]: 1.451e-05
[partial_unused_args_eliminate]: 2.32001e-06
[environ_conv]: 5.64e-06
[add_recomputation]: 5.666e-05
[cse_after_recomputation]: 2.172e-05, [1]
    [Cycle 1]: 1.621e-05, [1]
        [cse]: 9.73002e-06
[auto_monad_reorder]: 1.459e-05
[get_jit_bprop_graph]: 2.11e-06
[rewriter_after_jit_bprop_graph]: 6.89001e-06
[opt_after_jit_grad]: 0.00047262
[symbol_engine_optimizer]: 7.726e-05, [1]
    [Cycle 1]: 6.969e-05, [6]
        [build]: 4.36002e-06
        [elim_shapecalc]: 7.86001e-06
        [elim_not_effective]: 1.306e-05
        [opt_reshape]: 6.86999e-06
        [fold_const_symbol]: 9.29998e-06
        [renormalize]: 4.59986e-07
[validate]: 4.196e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.0181937
[execute]: 9.74999e-06
Sums
    bootstrap                            :     0.000422s :  1.50%
    type_inference                       :     0.004972s : 17.65%
    event_method                         :     0.000018s :  0.06%
    auto_monad                           :     0.000064s :  0.23%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000035s :  0.13%
    rewriter_before_opt_a                :     0.000070s :  0.25%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000080s :  0.28%
    jit_opt_a.loop_unroll                :     0.000035s :  0.12%
    jit_opt_a.a_1                        :     0.000755s :  2.68%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.06%
    jit_opt_a.inline_without_move        :     0.000010s :  0.04%
    jit_opt_a.renormalize                :     0.000962s :  3.42%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.07%
    jit_opt_a.cse                        :     0.000044s :  0.16%
    jit_opt_a.replace_applicator         :     0.000020s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000040s :  0.14%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000746s :  2.65%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000032s :  0.11%
    loop_unroll                          :     0.000584s :  2.07%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000015s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000057s :  0.20%
    cse_after_recomputation.cse          :     0.000010s :  0.03%
    auto_monad_reorder                   :     0.000015s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000473s :  1.68%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000042s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.018194s : 64.58%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.000297    23
 0.61% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 1.82% :     0.000005s :      3: substitution.graph_param_transform
87.79% :     0.000261s :      4: substitution.inline
 1.40% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.77% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.79% :     0.000005s :      2: substitution.replace_old_param
 4.35% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004907     2
82.02% :     0.004025s :      1: type_inference.infer
17.98% :     0.000882s :      1: type_inference.specialize
------[replace.]   0.000081     6
68.75% :     0.000055s :      4: replace.inline
31.25% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000270     6
95.62% :     0.000258s :      4: match.inline
 4.38% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.21% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.54% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.35% :     0.000002s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 6.14% :     0.000008s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.57% :     0.000002s :     15: predicate.load_eliminater
 2.33% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.49% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.92% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 3.02% :     0.000004s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.03% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.85% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     10: predicate.reshape_eliminate
 0.99% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.27% :     0.000002s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.10% :     0.000003s :     16: predicate.switch_defer_inline
 1.85% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.24% :     0.000011s :     53: predicate.switch_simplify
 1.08% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000002s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.08% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.32% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.42% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.48% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.94% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000695    12
46.75% :     0.000325s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.76% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.49% :     0.000309s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.045747    76
 0.13% :     0.000060s :      1: add_recomputation
 0.15% :     0.000068s :      1: auto_monad
 0.04% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.97% :     0.000445s :      1: bootstrap
 0.08% :     0.000036s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.05% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.05% :     0.000022s :      1: event_method
 0.03% :     0.000015s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
37.77% :     0.017280s :      1: jit_opt_a
 0.36% :     0.000167s :      1: jit_opt_after_cconv
 0.14% :     0.000065s :      1: jit_opt_b
 1.30% :     0.000595s :      1: loop_unroll
 1.66% :     0.000760s :      1: mutable_eliminate
 2.16% :     0.000988s :     26: opt.transform.jit_opt_a
 0.12% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000033s :      4: opt.transform.jit_opt_b
 0.04% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.05% :     0.000482s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.08% :     0.000038s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000017s :      1: remove_dup_value
 1.17% :     0.000537s :      1: renormalize.infer
 0.91% :     0.000417s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000043s :      1: rewriter_after_opt_a
 0.16% :     0.000073s :      1: rewriter_before_opt_a
 0.17% :     0.000080s :      1: symbol_engine_optimizer
39.82% :     0.018216s :      1: task_emit
10.90% :     0.004986s :      1: type_inference
 0.14% :     0.000064s :      1: validate


TotalTime = 0.0555661, [33]
[bootstrap]: 0.00038237
[type_inference]: 0.0359715
[event_method]: 2.114e-05
[auto_monad]: 7.558e-05
[graph_reusing]: 6.24001e-06
[pre_auto_parallel]: 2.76e-06
[py_interpret_to_execute]: 5.246e-05
[rewriter_before_opt_a]: 0.0001267
[expand_dump_flag]: 3.63e-06
[jit_opt_a]: 0.00664554, [2]
    [Cycle 1]: 0.00204037, [27]
        [switch_simplify]: 9.956e-05
        [loop_unroll]: 3.172e-05
        [a_1]: 0.00061313
        [with_stream_mark]: 2.067e-05
        [recompute_prepare]: 6.84999e-06
        [updatestate_depend_eliminate]: 3.01999e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 1.32e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 5.12999e-06
        [accelerated_algorithm]: 5.76003e-06
        [meta_shard_fg_expand]: 2.02001e-06
        [get_grad_eliminate_]: 5.61e-06
        [merge_forward]: 4.08999e-06
        [cell_reuse_recompute_pass]: 1.38002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.388e-05
        [j_node_and_user_rematch]: 9.29998e-06
        [meta_fg_expand]: 2.17999e-06
        [replace_old_param]: 9.44998e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.00095133
        [add_forward_monad_depend]: 6.37001e-06
        [auto_monad_grad]: 2.53e-06
        [auto_monad_eliminator]: 1.457e-05
        [cse]: 2.396e-05
        [replace_applicator]: 1.461e-05
    [Cycle 2]: 0.00032925, [27]
        [switch_simplify]: 5.46e-06
        [loop_unroll]: 5.39e-06
        [a_1]: 9.443e-05
        [with_stream_mark]: 1.111e-05
        [recompute_prepare]: 5.01997e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 1.96998e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.35001e-06
        [specialize_transform]: 4.91002e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 1.71e-06
        [get_grad_eliminate_]: 4.53999e-06
        [merge_forward]: 2.97002e-06
        [cell_reuse_recompute_pass]: 2.16e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.319e-05
        [j_node_and_user_rematch]: 7.8e-06
        [meta_fg_expand]: 1.75001e-06
        [replace_old_param]: 7.26999e-06
        [inline_without_move]: 4.70999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.29998e-06
        [auto_monad_grad]: 6.60017e-07
        [auto_monad_eliminator]: 5.99e-06
        [cse]: 1.04e-05
        [replace_applicator]: 5.44998e-06
[py_interpret_to_execute_after_opt_a]: 1.291e-05
[rewriter_after_opt_a]: 3.599e-05
[convert_after_rewriter]: 7.23e-06
[order_py_execute_after_rewriter]: 5.24e-06
[mutable_eliminate]: 0.00073604
[jit_opt_b]: 5.43e-05, [1]
    [Cycle 1]: 4.735e-05, [2]
        [frontend_op_eliminate]: 1.953e-05
        [inline_after_opt_a]: 1.549e-05
[cconv]: 2.811e-05
[loop_unroll]: 0.00043983
[jit_opt_after_cconv]: 0.00014399, [1]
    [Cycle 1]: 0.00013758, [11]
        [c_1]: 2.135e-05
        [parameter_eliminate]: 3.04001e-06
        [updatestate_depend_eliminate]: 5.97999e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [cse]: 2.04e-05
        [call_graph_tuple_transform]: 2.028e-05
        [tuple_list_get_item_eliminator]: 5.15999e-06
        [none_parameter_eliminate]: 1.19e-06
        [renormalize]: 6.89994e-07
        [switch_simplify]: 4.97e-06
[remove_dup_value]: 1.468e-05
[partial_unused_args_eliminate]: 2.34001e-06
[environ_conv]: 5.64e-06
[add_recomputation]: 4.704e-05
[cse_after_recomputation]: 1.977e-05, [1]
    [Cycle 1]: 1.461e-05, [1]
        [cse]: 8.45999e-06
[auto_monad_reorder]: 1.496e-05
[get_jit_bprop_graph]: 2.31e-06
[rewriter_after_jit_bprop_graph]: 6.28e-06
[opt_after_jit_grad]: 0.00046278
[symbol_engine_optimizer]: 7.384e-05, [1]
    [Cycle 1]: 6.65e-05, [6]
        [build]: 3.21999e-06
        [elim_shapecalc]: 7.36999e-06
        [elim_not_effective]: 1.295e-05
        [opt_reshape]: 5.86e-06
        [fold_const_symbol]: 8.98002e-06
        [renormalize]: 3.80009e-07
[validate]: 3.828e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00988371
[execute]: 1.121e-05
Sums
    bootstrap                            :     0.000382s :  0.75%
    type_inference                       :     0.035972s : 71.03%
    event_method                         :     0.000021s :  0.04%
    auto_monad                           :     0.000076s :  0.15%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000052s :  0.10%
    rewriter_before_opt_a                :     0.000127s :  0.25%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000105s :  0.21%
    jit_opt_a.loop_unroll                :     0.000037s :  0.07%
    jit_opt_a.a_1                        :     0.000708s :  1.40%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.06%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.02%
    jit_opt_a.merge_forward              :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.05%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000017s :  0.03%
    jit_opt_a.inline_without_move        :     0.000010s :  0.02%
    jit_opt_a.renormalize                :     0.000951s :  1.88%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.04%
    jit_opt_a.cse                        :     0.000034s :  0.07%
    jit_opt_a.replace_applicator         :     0.000020s :  0.04%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.03%
    rewriter_after_opt_a                 :     0.000036s :  0.07%
    convert_after_rewriter               :     0.000007s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000736s :  1.45%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.03%
    cconv                                :     0.000028s :  0.06%
    loop_unroll                          :     0.000440s :  0.87%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.04%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000020s :  0.04%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.01%
    remove_dup_value                     :     0.000015s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000006s :  0.01%
    add_recomputation                    :     0.000047s :  0.09%
    cse_after_recomputation.cse          :     0.000008s :  0.02%
    auto_monad_reorder                   :     0.000015s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000463s :  0.91%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.08%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009884s : 19.52%
    execute                              :     0.000011s :  0.02%


Time group info:
------[substitution.]   0.000261    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 2.32% :     0.000006s :      3: substitution.graph_param_transform
86.49% :     0.000226s :      4: substitution.inline
 1.20% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.01% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.54% :     0.000004s :      2: substitution.replace_old_param
 5.11% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.035861     2
13.67% :     0.004901s :      1: type_inference.infer
86.33% :     0.030960s :      1: type_inference.specialize
------[replace.]   0.000076     6
68.37% :     0.000052s :      4: replace.inline
31.63% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000235     6
94.77% :     0.000223s :      4: match.inline
 5.23% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.58% :     0.000002s :     10: predicate.addn_zero_filter
 1.97% :     0.000002s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.76% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.27% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.48% :     0.000002s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.18% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.49% :     0.000002s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.25% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.35% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.31% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.01% :     0.000002s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.98% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.99% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.77% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.18% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.26% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.37% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.87% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.54% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000780    12
48.21% :     0.000376s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.72% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.07% :     0.000351s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.057540    76
 0.09% :     0.000050s :      1: add_recomputation
 0.14% :     0.000079s :      1: auto_monad
 0.03% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.70% :     0.000404s :      1: bootstrap
 0.05% :     0.000031s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.04% :     0.000022s :      1: cse_after_recomputation
 0.01% :     0.000008s :      1: environ_conv
 0.04% :     0.000025s :      1: event_method
 0.03% :     0.000017s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000009s :      1: graph_reusing
11.56% :     0.006650s :      1: jit_opt_a
 0.26% :     0.000147s :      1: jit_opt_after_cconv
 0.10% :     0.000057s :      1: jit_opt_b
 0.78% :     0.000449s :      1: loop_unroll
 1.29% :     0.000745s :      1: mutable_eliminate
 1.61% :     0.000924s :     26: opt.transform.jit_opt_a
 0.08% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000028s :      4: opt.transform.jit_opt_b
 0.02% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000032s :      4: opt.transform.symbol_engine_opt
 0.82% :     0.000471s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.10% :     0.000055s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000017s :      1: remove_dup_value
 0.91% :     0.000522s :      1: renormalize.infer
 0.73% :     0.000421s :      1: renormalize.specialize
 0.01% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000039s :      1: rewriter_after_opt_a
 0.23% :     0.000132s :      1: rewriter_before_opt_a
 0.13% :     0.000077s :      1: symbol_engine_optimizer
17.22% :     0.009906s :      1: task_emit
62.55% :     0.035989s :      1: type_inference
 0.10% :     0.000058s :      1: validate


TotalTime = 0.0238293, [33]
[bootstrap]: 0.00033173
[type_inference]: 0.00604129
[event_method]: 2.385e-05
[auto_monad]: 7.268e-05
[graph_reusing]: 6.17001e-06
[pre_auto_parallel]: 2.62001e-06
[py_interpret_to_execute]: 9.605e-05
[rewriter_before_opt_a]: 8.905e-05
[expand_dump_flag]: 3.7e-06
[jit_opt_a]: 0.00659852, [2]
    [Cycle 1]: 0.00250787, [27]
        [switch_simplify]: 6.792e-05
        [loop_unroll]: 4.811e-05
        [a_1]: 0.00062908
        [with_stream_mark]: 2.524e-05
        [recompute_prepare]: 8.55999e-06
        [updatestate_depend_eliminate]: 4.02e-06
        [updatestate_assign_eliminate]: 3.06999e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [parameter_eliminate]: 2.12999e-06
        [specialize_transform]: 6.28e-06
        [updatestate_useless_node_eliminater]: 5.57001e-06
        [accelerated_algorithm]: 5.85002e-06
        [meta_shard_fg_expand]: 3.04001e-06
        [get_grad_eliminate_]: 6.16e-06
        [merge_forward]: 3.67998e-06
        [cell_reuse_recompute_pass]: 1.21002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.603e-05
        [j_node_and_user_rematch]: 1.027e-05
        [meta_fg_expand]: 3.21001e-06
        [replace_old_param]: 1.018e-05
        [inline_without_move]: 5.32001e-06
        [renormalize]: 0.00093547
        [add_forward_monad_depend]: 7.38999e-06
        [auto_monad_grad]: 2.86999e-06
        [auto_monad_eliminator]: 1.605e-05
        [cse]: 3.302e-05
        [replace_applicator]: 1.697e-05
    [Cycle 2]: 0.00034781, [27]
        [switch_simplify]: 5.77001e-06
        [loop_unroll]: 5.32001e-06
        [a_1]: 9.909e-05
        [with_stream_mark]: 1.095e-05
        [recompute_prepare]: 4.98001e-06
        [updatestate_depend_eliminate]: 3.39001e-06
        [updatestate_assign_eliminate]: 2.11003e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [parameter_eliminate]: 1.44998e-06
        [specialize_transform]: 4.93001e-06
        [updatestate_useless_node_eliminater]: 5.44998e-06
        [accelerated_algorithm]: 5.55001e-06
        [meta_shard_fg_expand]: 1.76998e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 2.74999e-06
        [cell_reuse_recompute_pass]: 2.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.466e-05
        [j_node_and_user_rematch]: 8.15e-06
        [meta_fg_expand]: 2.15002e-06
        [replace_old_param]: 8.35001e-06
        [inline_without_move]: 4.74e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.59e-06
        [auto_monad_grad]: 6.79982e-07
        [auto_monad_eliminator]: 5.95002e-06
        [cse]: 1.144e-05
        [replace_applicator]: 5.32001e-06
[py_interpret_to_execute_after_opt_a]: 1.571e-05
[rewriter_after_opt_a]: 3.951e-05
[convert_after_rewriter]: 6.83e-06
[order_py_execute_after_rewriter]: 5.40999e-06
[mutable_eliminate]: 0.00073996
[jit_opt_b]: 6.448e-05, [1]
    [Cycle 1]: 5.46e-05, [2]
        [frontend_op_eliminate]: 2.41e-05
        [inline_after_opt_a]: 1.862e-05
[cconv]: 2.957e-05
[loop_unroll]: 0.00049316
[jit_opt_after_cconv]: 0.00017135, [1]
    [Cycle 1]: 0.00016273, [11]
        [c_1]: 2.267e-05
        [parameter_eliminate]: 4.12003e-06
        [updatestate_depend_eliminate]: 8.33999e-06
        [updatestate_assign_eliminate]: 2.98998e-06
        [updatestate_loads_eliminate]: 2.71999e-06
        [cse]: 2.852e-05
        [call_graph_tuple_transform]: 2.398e-05
        [tuple_list_get_item_eliminator]: 5.82001e-06
        [none_parameter_eliminate]: 1.67999e-06
        [renormalize]: 1.34e-06
        [switch_simplify]: 5.61e-06
[remove_dup_value]: 1.284e-05
[partial_unused_args_eliminate]: 2.09e-06
[environ_conv]: 5.79e-06
[add_recomputation]: 5.287e-05
[cse_after_recomputation]: 2.305e-05, [1]
    [Cycle 1]: 1.589e-05, [1]
        [cse]: 9.97999e-06
[auto_monad_reorder]: 1.501e-05
[get_jit_bprop_graph]: 1.93997e-06
[rewriter_after_jit_bprop_graph]: 1.012e-05
[opt_after_jit_grad]: 0.00050463
[symbol_engine_optimizer]: 8.16e-05, [1]
    [Cycle 1]: 7.312e-05, [6]
        [build]: 4.99e-06
        [elim_shapecalc]: 8.72998e-06
        [elim_not_effective]: 1.406e-05
        [opt_reshape]: 6.47001e-06
        [fold_const_symbol]: 9.74e-06
        [renormalize]: 1.10999e-06
[validate]: 4.146e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00797243
[execute]: 1.014e-05
Sums
    bootstrap                            :     0.000332s :  1.75%
    type_inference                       :     0.006041s : 31.90%
    event_method                         :     0.000024s :  0.13%
    auto_monad                           :     0.000073s :  0.38%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000096s :  0.51%
    rewriter_before_opt_a                :     0.000089s :  0.47%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.39%
    jit_opt_a.loop_unroll                :     0.000053s :  0.28%
    jit_opt_a.a_1                        :     0.000728s :  3.84%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000936s :  4.94%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.12%
    jit_opt_a.cse                        :     0.000044s :  0.23%
    jit_opt_a.replace_applicator         :     0.000022s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.08%
    rewriter_after_opt_a                 :     0.000040s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000740s :  3.91%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.10%
    cconv                                :     0.000030s :  0.16%
    loop_unroll                          :     0.000493s :  2.60%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000013s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000053s :  0.28%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.05%
    opt_after_jit_grad                   :     0.000505s :  2.66%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007972s : 42.10%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000265    23
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000002s :      2: substitution.fold_const_symbol
 2.60% :     0.000007s :      3: substitution.graph_param_transform
84.57% :     0.000224s :      4: substitution.inline
 1.74% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.03% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.26% :     0.000006s :      2: substitution.replace_old_param
 5.32% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005944     2
80.73% :     0.004799s :      1: type_inference.infer
19.27% :     0.001145s :      1: type_inference.specialize
------[replace.]   0.000082     6
66.83% :     0.000055s :      4: replace.inline
33.17% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000234     6
94.51% :     0.000221s :      4: match.inline
 5.49% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   691
 1.39% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.62% :     0.000003s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.93% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000002s :     10: predicate.depend_value_elim
 1.33% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.54% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.37% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.44% :     0.000002s :      6: predicate.get_grad_eliminate
 0.36% :     0.000000s :      3: predicate.graph_param_transform
 5.39% :     0.000007s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.35% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 6.33% :     0.000008s :     34: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     15: predicate.partial_eliminate
 1.00% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.67% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.94% :     0.000003s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.54% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.50% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.13% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 1.93% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.87% :     0.000009s :     53: predicate.switch_simplify
 1.10% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.22% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.36% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 2.00% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.40% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000819    12
50.12% :     0.000411s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.23% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.65% :     0.000366s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025856    76
 0.22% :     0.000056s :      1: add_recomputation
 0.30% :     0.000077s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.37% :     0.000355s :      1: bootstrap
 0.12% :     0.000032s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.11% :     0.000029s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
25.53% :     0.006602s :      1: jit_opt_a
 0.67% :     0.000174s :      1: jit_opt_after_cconv
 0.26% :     0.000067s :      1: jit_opt_b
 1.94% :     0.000503s :      1: loop_unroll
 2.90% :     0.000751s :      1: mutable_eliminate
 3.79% :     0.000979s :     26: opt.transform.jit_opt_a
 0.21% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000034s :      4: opt.transform.jit_opt_b
 0.06% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.99% :     0.000515s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.39% :     0.000101s :      1: py_interpret_to_execute
 0.07% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000015s :      1: remove_dup_value
 1.94% :     0.000503s :      1: renormalize.infer
 1.64% :     0.000423s :      1: renormalize.specialize
 0.05% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000042s :      1: rewriter_after_opt_a
 0.36% :     0.000094s :      1: rewriter_before_opt_a
 0.33% :     0.000084s :      1: symbol_engine_optimizer
30.93% :     0.007999s :      1: task_emit
23.45% :     0.006064s :      1: type_inference
 0.26% :     0.000066s :      1: validate


TotalTime = 0.0533589, [33]
[bootstrap]: 0.00052533
[type_inference]: 0.00734899
[event_method]: 2.112e-05
[auto_monad]: 6.929e-05
[graph_reusing]: 5.69e-06
[pre_auto_parallel]: 2.77002e-06
[py_interpret_to_execute]: 4.188e-05
[rewriter_before_opt_a]: 7.616e-05
[expand_dump_flag]: 3.43999e-06
[jit_opt_a]: 0.00976344, [2]
    [Cycle 1]: 0.00219205, [27]
        [switch_simplify]: 6.444e-05
        [loop_unroll]: 3.037e-05
        [a_1]: 0.00064533
        [with_stream_mark]: 2.494e-05
        [recompute_prepare]: 8.22e-06
        [updatestate_depend_eliminate]: 4.55999e-06
        [updatestate_assign_eliminate]: 3.11001e-06
        [updatestate_loads_eliminate]: 2.409e-05
        [parameter_eliminate]: 2.24001e-06
        [specialize_transform]: 6.83998e-06
        [updatestate_useless_node_eliminater]: 5.99999e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 2.07999e-06
        [get_grad_eliminate_]: 6.88998e-06
        [merge_forward]: 4.08999e-06
        [cell_reuse_recompute_pass]: 1.50999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.636e-05
        [j_node_and_user_rematch]: 1.011e-05
        [meta_fg_expand]: 2.66e-06
        [replace_old_param]: 1.065e-05
        [inline_without_move]: 6.02001e-06
        [renormalize]: 0.00100083
        [add_forward_monad_depend]: 3.651e-05
        [auto_monad_grad]: 2.69001e-06
        [auto_monad_eliminator]: 1.715e-05
        [cse]: 3.103e-05
        [replace_applicator]: 1.592e-05
    [Cycle 2]: 0.00303614, [27]
        [switch_simplify]: 5.91003e-06
        [loop_unroll]: 4.94e-06
        [a_1]: 9.196e-05
        [with_stream_mark]: 1.048e-05
        [recompute_prepare]: 4.73001e-06
        [updatestate_depend_eliminate]: 2.81e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 1.36002e-06
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.92999e-06
        [accelerated_algorithm]: 5.65001e-06
        [meta_shard_fg_expand]: 1.35999e-06
        [get_grad_eliminate_]: 4.62998e-06
        [merge_forward]: 2.66e-06
        [cell_reuse_recompute_pass]: 2.84999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.427e-05
        [j_node_and_user_rematch]: 8.56002e-06
        [meta_fg_expand]: 8.59e-06
        [replace_old_param]: 3.599e-05
        [inline_without_move]: 8.38001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 7.85e-06
        [auto_monad_grad]: 2.62001e-06
        [auto_monad_eliminator]: 2.067e-05
        [cse]: 3.228e-05
        [replace_applicator]: 9.10001e-06
[py_interpret_to_execute_after_opt_a]: 1.93e-05
[rewriter_after_opt_a]: 4.295e-05
[convert_after_rewriter]: 8.08001e-06
[order_py_execute_after_rewriter]: 4.70999e-06
[mutable_eliminate]: 0.0008508
[jit_opt_b]: 6.69e-05, [1]
    [Cycle 1]: 5.598e-05, [2]
        [frontend_op_eliminate]: 1.999e-05
        [inline_after_opt_a]: 2.14e-05
[cconv]: 3.769e-05
[loop_unroll]: 0.0004717
[jit_opt_after_cconv]: 0.00016652, [1]
    [Cycle 1]: 0.00015865, [11]
        [c_1]: 2.255e-05
        [parameter_eliminate]: 4.47e-06
        [updatestate_depend_eliminate]: 9.65002e-06
        [updatestate_assign_eliminate]: 3.51001e-06
        [updatestate_loads_eliminate]: 3.04001e-06
        [cse]: 2.537e-05
        [call_graph_tuple_transform]: 2.338e-05
        [tuple_list_get_item_eliminator]: 5.52001e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 9.30013e-07
        [switch_simplify]: 5.67999e-06
[remove_dup_value]: 1.449e-05
[partial_unused_args_eliminate]: 2.56e-06
[environ_conv]: 6.34001e-06
[add_recomputation]: 5.756e-05
[cse_after_recomputation]: 2.229e-05, [1]
    [Cycle 1]: 1.589e-05, [1]
        [cse]: 9.74e-06
[auto_monad_reorder]: 1.571e-05
[get_jit_bprop_graph]: 2.06e-06
[rewriter_after_jit_bprop_graph]: 7.61999e-06
[opt_after_jit_grad]: 0.00048306
[symbol_engine_optimizer]: 8.16e-05, [1]
    [Cycle 1]: 7.312e-05, [6]
        [build]: 5.41998e-06
        [elim_shapecalc]: 8.61997e-06
        [elim_not_effective]: 1.403e-05
        [opt_reshape]: 6.40002e-06
        [fold_const_symbol]: 8.99998e-06
        [renormalize]: 4.60015e-07
[validate]: 4.27e-05
[backend_pass]: 1.57999e-06
[task_emit]: 0.0328391
[execute]: 8.34998e-06
Sums
    bootstrap                            :     0.000525s :  1.15%
    type_inference                       :     0.007349s : 16.15%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000069s :  0.15%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.09%
    rewriter_before_opt_a                :     0.000076s :  0.17%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000070s :  0.15%
    jit_opt_a.loop_unroll                :     0.000035s :  0.08%
    jit_opt_a.a_1                        :     0.000737s :  1.62%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000026s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.07%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000011s :  0.02%
    jit_opt_a.replace_old_param          :     0.000047s :  0.10%
    jit_opt_a.inline_without_move        :     0.000014s :  0.03%
    jit_opt_a.renormalize                :     0.001001s :  2.20%
    jit_opt_a.add_forward_monad_depend   :     0.000044s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000038s :  0.08%
    jit_opt_a.cse                        :     0.000063s :  0.14%
    jit_opt_a.replace_applicator         :     0.000025s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.04%
    rewriter_after_opt_a                 :     0.000043s :  0.09%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000851s :  1.87%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000021s :  0.05%
    cconv                                :     0.000038s :  0.08%
    loop_unroll                          :     0.000472s :  1.04%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.05%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.01%
    remove_dup_value                     :     0.000014s :  0.03%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.01%
    add_recomputation                    :     0.000058s :  0.13%
    cse_after_recomputation.cse          :     0.000010s :  0.02%
    auto_monad_reorder                   :     0.000016s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000483s :  1.06%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000043s :  0.09%
    backend_pass                         :     0.000002s :  0.00%
    task_emit                            :     0.032839s : 72.17%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000310    23
 0.60% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000002s :      2: substitution.fold_const_symbol
 2.13% :     0.000007s :      3: substitution.graph_param_transform
86.39% :     0.000268s :      4: substitution.inline
 1.28% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.66% :     0.000005s :      4: substitution.remove_not_recompute_node
 3.22% :     0.000010s :      2: substitution.replace_old_param
 4.20% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007264     2
85.68% :     0.006224s :      1: type_inference.infer
14.32% :     0.001040s :      1: type_inference.specialize
------[replace.]   0.000075     6
68.60% :     0.000051s :      4: replace.inline
31.40% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000277     6
95.68% :     0.000265s :      4: match.inline
 4.32% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.17% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.17% :     0.000001s :     10: predicate.addn_zero_filter
 1.82% :     0.000002s :     10: predicate.arithmetic_simplify
 1.12% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.55% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.12% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.08% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.01% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.08% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.40% :     0.000001s :      3: predicate.graph_param_transform
 5.88% :     0.000007s :     22: predicate.inline
 2.38% :     0.000003s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.80% :     0.000002s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.88% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.93% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 2.67% :     0.000003s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 1.97% :     0.000002s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 1.17% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.21% :     0.000001s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.35% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.74% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.18% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.39% :     0.000009s :     53: predicate.switch_simplify
 1.08% :     0.000001s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.25% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.06% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.88% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.39% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000780    12
47.79% :     0.000373s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.20% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.00% :     0.000351s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.055473    76
 0.11% :     0.000061s :      1: add_recomputation
 0.13% :     0.000073s :      1: auto_monad
 0.03% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.99% :     0.000551s :      1: bootstrap
 0.07% :     0.000040s :      1: cconv
 0.02% :     0.000011s :      1: convert_after_rewriter
 0.04% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.05% :     0.000026s :      1: event_method
 0.02% :     0.000013s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000008s :      1: graph_reusing
17.61% :     0.009768s :      1: jit_opt_a
 0.31% :     0.000170s :      1: jit_opt_after_cconv
 0.13% :     0.000070s :      1: jit_opt_b
 0.87% :     0.000480s :      1: loop_unroll
 1.56% :     0.000865s :      1: mutable_eliminate
 1.80% :     0.000999s :     26: opt.transform.jit_opt_a
 0.10% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000032s :      4: opt.transform.jit_opt_b
 0.02% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000034s :      4: opt.transform.symbol_engine_opt
 0.89% :     0.000492s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000045s :      1: py_interpret_to_execute
 0.04% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000017s :      1: remove_dup_value
 1.05% :     0.000581s :      1: renormalize.infer
 0.74% :     0.000411s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000047s :      1: rewriter_after_opt_a
 0.14% :     0.000079s :      1: rewriter_before_opt_a
 0.15% :     0.000084s :      1: symbol_engine_optimizer
59.23% :     0.032856s :      1: task_emit
13.29% :     0.007374s :      1: type_inference
 0.12% :     0.000066s :      1: validate


TotalTime = 0.0157493, [30]
[bootstrap]: 0.00041082
[type_inference]: 0.00580981
[event_method]: 1.964e-05
[auto_monad]: 6.799e-05
[graph_reusing]: 6.01e-06
[pre_auto_parallel]: 2.34999e-06
[py_interpret_to_execute]: 4.008e-05
[rewriter_before_opt_a]: 7.659e-05
[expand_dump_flag]: 3.41001e-06
[jit_opt_a]: 0.00682614, [2]
    [Cycle 1]: 0.00214162, [27]
        [switch_simplify]: 6.956e-05
        [loop_unroll]: 3.036e-05
        [a_1]: 0.00062636
        [with_stream_mark]: 2.48e-05
        [recompute_prepare]: 7.38999e-06
        [updatestate_depend_eliminate]: 3.85e-06
        [updatestate_assign_eliminate]: 3.19001e-06
        [updatestate_loads_eliminate]: 2.93e-06
        [parameter_eliminate]: 1.92001e-06
        [specialize_transform]: 6.52001e-06
        [updatestate_useless_node_eliminater]: 5.30001e-06
        [accelerated_algorithm]: 5.72999e-06
        [meta_shard_fg_expand]: 2.34999e-06
        [get_grad_eliminate_]: 5.36002e-06
        [merge_forward]: 4.05998e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.455e-05
        [j_node_and_user_rematch]: 1.04e-05
        [meta_fg_expand]: 2.74999e-06
        [replace_old_param]: 1.003e-05
        [inline_without_move]: 5.49e-06
        [renormalize]: 0.00103409
        [add_forward_monad_depend]: 7.28e-06
        [auto_monad_grad]: 2.97002e-06
        [auto_monad_eliminator]: 1.633e-05
        [cse]: 2.985e-05
        [replace_applicator]: 1.517e-05
    [Cycle 2]: 0.00033829, [27]
        [switch_simplify]: 6.24001e-06
        [loop_unroll]: 5.05001e-06
        [a_1]: 9.451e-05
        [with_stream_mark]: 1.209e-05
        [recompute_prepare]: 5.54e-06
        [updatestate_depend_eliminate]: 2.72001e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 1.95001e-06
        [parameter_eliminate]: 1.26002e-06
        [specialize_transform]: 5.36998e-06
        [updatestate_useless_node_eliminater]: 5.17999e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.95001e-06
        [merge_forward]: 3.30998e-06
        [cell_reuse_recompute_pass]: 2.06998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.385e-05
        [j_node_and_user_rematch]: 9.04998e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 7.76001e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 1.09983e-07
        [add_forward_monad_depend]: 1.89e-06
        [auto_monad_grad]: 8.40024e-07
        [auto_monad_eliminator]: 5.66003e-06
        [cse]: 1.217e-05
        [replace_applicator]: 5.18002e-06
[py_interpret_to_execute_after_opt_a]: 1.445e-05
[rewriter_after_opt_a]: 3.88e-05
[convert_after_rewriter]: 7.58999e-06
[order_py_execute_after_rewriter]: 5.49e-06
[mutable_eliminate]: 0.00076441
[jit_opt_b]: 6.189e-05, [1]
    [Cycle 1]: 5.401e-05, [2]
        [frontend_op_eliminate]: 2.385e-05
        [inline_after_opt_a]: 1.709e-05
[cconv]: 3.36e-05
[loop_unroll]: 0.00046893
[jit_opt_after_cconv]: 0.00015613, [1]
    [Cycle 1]: 0.0001489, [11]
        [c_1]: 2.267e-05
        [parameter_eliminate]: 3.76999e-06
        [updatestate_depend_eliminate]: 8.71002e-06
        [updatestate_assign_eliminate]: 2.63e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 2.273e-05
        [call_graph_tuple_transform]: 2.087e-05
        [tuple_list_get_item_eliminator]: 5.81e-06
        [none_parameter_eliminate]: 1.78002e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 5.52001e-06
[remove_dup_value]: 1.571e-05
[partial_unused_args_eliminate]: 2.83998e-06
[environ_conv]: 6.44999e-06
[add_recomputation]: 5.623e-05
[cse_after_recomputation]: 2.314e-05, [1]
    [Cycle 1]: 1.664e-05, [1]
        [cse]: 1.019e-05
[auto_monad_reorder]: 1.634e-05
[get_jit_bprop_graph]: 2.38002e-06
[rewriter_after_jit_bprop_graph]: 6.79001e-06
[opt_after_jit_grad]: 0.00048825
[symbol_engine_optimizer]: 8.109e-05, [1]
    [Cycle 1]: 7.346e-05, [6]
        [build]: 4.26001e-06
        [elim_shapecalc]: 7.56001e-06
        [elim_not_effective]: 1.474e-05
        [opt_reshape]: 6.79001e-06
        [fold_const_symbol]: 9.76998e-06
        [renormalize]: 4.49974e-07
[validate]: 4.343e-05
Sums
    bootstrap                            :     0.000411s :  3.81%
    type_inference                       :     0.005810s : 53.93%
    event_method                         :     0.000020s :  0.18%
    auto_monad                           :     0.000068s :  0.63%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.37%
    rewriter_before_opt_a                :     0.000077s :  0.71%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000076s :  0.70%
    jit_opt_a.loop_unroll                :     0.000035s :  0.33%
    jit_opt_a.a_1                        :     0.000721s :  6.69%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.34%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.17%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.001034s :  9.60%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.20%
    jit_opt_a.cse                        :     0.000042s :  0.39%
    jit_opt_a.replace_applicator         :     0.000020s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.13%
    rewriter_after_opt_a                 :     0.000039s :  0.36%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000764s :  7.10%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.16%
    cconv                                :     0.000034s :  0.31%
    loop_unroll                          :     0.000469s :  4.35%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000016s :  0.15%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000056s :  0.52%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000488s :  4.53%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000043s :  0.40%


Time group info:
------[substitution.]   0.000275    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000002s :      2: substitution.fold_const_symbol
 2.05% :     0.000006s :      3: substitution.graph_param_transform
86.44% :     0.000238s :      4: substitution.inline
 1.36% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.58% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.90% :     0.000005s :      2: substitution.replace_old_param
 5.13% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005737     2
78.75% :     0.004517s :      1: type_inference.infer
21.25% :     0.001219s :      1: type_inference.specialize
------[replace.]   0.000081     6
66.95% :     0.000054s :      4: replace.inline
33.05% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000248     6
94.74% :     0.000235s :      4: match.inline
 5.26% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.43% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 2.31% :     0.000003s :     10: predicate.arithmetic_simplify
 1.15% :     0.000001s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 6.40% :     0.000008s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.62% :     0.000002s :     15: predicate.load_eliminater
 1.58% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 1.61% :     0.000002s :      3: predicate.mutable_eliminate
 0.82% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.33% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.45% :     0.000001s :      6: predicate.replace_old_param
 0.56% :     0.000001s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.01% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.71% :     0.000002s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.57% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 1.87% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.81% :     0.000010s :     53: predicate.switch_simplify
 1.25% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.15% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.89% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.48% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.63% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000816    12
52.49% :     0.000428s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.07% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.44% :     0.000330s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017793    72
 0.33% :     0.000059s :      1: add_recomputation
 0.40% :     0.000071s :      1: auto_monad
 0.11% :     0.000019s :      1: auto_monad_reorder
 2.42% :     0.000430s :      1: bootstrap
 0.20% :     0.000036s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.13% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.39% :     0.006831s :      1: jit_opt_a
 0.90% :     0.000159s :      1: jit_opt_after_cconv
 0.36% :     0.000065s :      1: jit_opt_b
 2.68% :     0.000478s :      1: loop_unroll
 4.35% :     0.000774s :      1: mutable_eliminate
 5.37% :     0.000955s :     26: opt.transform.jit_opt_a
 0.29% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000033s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.79% :     0.000497s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000043s :      1: py_interpret_to_execute
 0.10% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000019s :      1: remove_dup_value
 3.03% :     0.000539s :      1: renormalize.infer
 2.73% :     0.000485s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000042s :      1: rewriter_after_opt_a
 0.45% :     0.000079s :      1: rewriter_before_opt_a
 0.47% :     0.000084s :      1: symbol_engine_optimizer
32.74% :     0.005825s :      1: type_inference


TotalTime = 0.03049, [33]
[bootstrap]: 0.00046482
[type_inference]: 0.0057536
[event_method]: 1.832e-05
[auto_monad]: 6.987e-05
[graph_reusing]: 6.01e-06
[pre_auto_parallel]: 2.91e-06
[py_interpret_to_execute]: 4.224e-05
[rewriter_before_opt_a]: 7.49e-05
[expand_dump_flag]: 3.49001e-06
[jit_opt_a]: 0.0106258, [2]
    [Cycle 1]: 0.00361262, [27]
        [switch_simplify]: 6.958e-05
        [loop_unroll]: 2.976e-05
        [a_1]: 0.00186887
        [with_stream_mark]: 2.87e-05
        [recompute_prepare]: 1.082e-05
        [updatestate_depend_eliminate]: 5.20999e-06
        [updatestate_assign_eliminate]: 3.16999e-06
        [updatestate_loads_eliminate]: 2.51998e-06
        [parameter_eliminate]: 2.31e-06
        [specialize_transform]: 6.81001e-06
        [updatestate_useless_node_eliminater]: 5.61003e-06
        [accelerated_algorithm]: 6.79999e-06
        [meta_shard_fg_expand]: 2.81e-06
        [get_grad_eliminate_]: 6.57002e-06
        [merge_forward]: 4.45e-06
        [cell_reuse_recompute_pass]: 1.82999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.609e-05
        [j_node_and_user_rematch]: 1.061e-05
        [meta_fg_expand]: 2.48e-06
        [replace_old_param]: 1.056e-05
        [inline_without_move]: 6.17001e-06
        [renormalize]: 0.00114138
        [add_forward_monad_depend]: 8.87e-06
        [auto_monad_grad]: 3.85998e-06
        [auto_monad_eliminator]: 1.666e-05
        [cse]: 3.089e-05
        [replace_applicator]: 1.64e-05
    [Cycle 2]: 0.00035279, [27]
        [switch_simplify]: 5.39e-06
        [loop_unroll]: 5.32999e-06
        [a_1]: 0.00010138
        [with_stream_mark]: 1.493e-05
        [recompute_prepare]: 5.33002e-06
        [updatestate_depend_eliminate]: 3.01001e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.52999e-06
        [specialize_transform]: 4.87e-06
        [updatestate_useless_node_eliminater]: 4.68999e-06
        [accelerated_algorithm]: 5.12999e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 4.62e-06
        [merge_forward]: 3.00998e-06
        [cell_reuse_recompute_pass]: 2.96999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.619e-05
        [j_node_and_user_rematch]: 8.42e-06
        [meta_fg_expand]: 1.83002e-06
        [replace_old_param]: 8.17e-06
        [inline_without_move]: 4.76002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.52001e-06
        [auto_monad_grad]: 7.40023e-07
        [auto_monad_eliminator]: 5.99999e-06
        [cse]: 1.255e-05
        [replace_applicator]: 4.75999e-06
[py_interpret_to_execute_after_opt_a]: 1.49e-05
[rewriter_after_opt_a]: 3.829e-05
[convert_after_rewriter]: 7.45998e-06
[order_py_execute_after_rewriter]: 4.99e-06
[mutable_eliminate]: 0.00097222
[jit_opt_b]: 5.836e-05, [1]
    [Cycle 1]: 4.968e-05, [2]
        [frontend_op_eliminate]: 2.094e-05
        [inline_after_opt_a]: 1.682e-05
[cconv]: 3.388e-05
[loop_unroll]: 0.0005096
[jit_opt_after_cconv]: 0.00017622, [1]
    [Cycle 1]: 0.00016808, [11]
        [c_1]: 2.135e-05
        [parameter_eliminate]: 5.14e-06
        [updatestate_depend_eliminate]: 9.66e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 3.07e-05
        [call_graph_tuple_transform]: 2.504e-05
        [tuple_list_get_item_eliminator]: 5.42999e-06
        [none_parameter_eliminate]: 2.24001e-06
        [renormalize]: 8.60018e-07
        [switch_simplify]: 6.09001e-06
[remove_dup_value]: 1.632e-05
[partial_unused_args_eliminate]: 2.02001e-06
[environ_conv]: 7.20003e-06
[add_recomputation]: 6.007e-05
[cse_after_recomputation]: 2.3e-05, [1]
    [Cycle 1]: 1.57e-05, [1]
        [cse]: 9.09e-06
[auto_monad_reorder]: 1.689e-05
[get_jit_bprop_graph]: 2.22999e-06
[rewriter_after_jit_bprop_graph]: 1.102e-05
[opt_after_jit_grad]: 0.0005877
[symbol_engine_optimizer]: 8.603e-05, [1]
    [Cycle 1]: 7.642e-05, [6]
        [build]: 7e-06
        [elim_shapecalc]: 8.60999e-06
        [elim_not_effective]: 1.473e-05
        [opt_reshape]: 6.09001e-06
        [fold_const_symbol]: 9.97999e-06
        [renormalize]: 5.59987e-07
[validate]: 5.244e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.0104741
[execute]: 9.34998e-06
Sums
    bootstrap                            :     0.000465s :  2.02%
    type_inference                       :     0.005754s : 25.00%
    event_method                         :     0.000018s :  0.08%
    auto_monad                           :     0.000070s :  0.30%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.18%
    rewriter_before_opt_a                :     0.000075s :  0.33%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000075s :  0.33%
    jit_opt_a.loop_unroll                :     0.000035s :  0.15%
    jit_opt_a.a_1                        :     0.001970s :  8.56%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.08%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.001141s :  4.96%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.10%
    jit_opt_a.cse                        :     0.000043s :  0.19%
    jit_opt_a.replace_applicator         :     0.000021s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.06%
    rewriter_after_opt_a                 :     0.000038s :  0.17%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000972s :  4.22%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.07%
    cconv                                :     0.000034s :  0.15%
    loop_unroll                          :     0.000510s :  2.21%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000031s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000060s :  0.26%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000017s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.05%
    opt_after_jit_grad                   :     0.000588s :  2.55%
    symbol_engine_optimizer.build        :     0.000007s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000052s :  0.23%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010474s : 45.51%
    execute                              :     0.000009s :  0.04%


Time group info:
------[substitution.]   0.001480    23
 0.14% :     0.000002s :      2: substitution.elim_not_effective
 0.09% :     0.000001s :      2: substitution.fold_const_symbol
 0.41% :     0.000006s :      3: substitution.graph_param_transform
97.30% :     0.001440s :      4: substitution.inline
 0.28% :     0.000004s :      4: substitution.j_node_and_user_rematch
 0.37% :     0.000005s :      4: substitution.remove_not_recompute_node
 0.41% :     0.000006s :      2: substitution.replace_old_param
 0.99% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005672     2
82.85% :     0.004699s :      1: type_inference.infer
17.15% :     0.000973s :      1: type_inference.specialize
------[replace.]   0.000094     6
69.41% :     0.000066s :      4: replace.inline
30.59% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.001449     6
99.08% :     0.001435s :      4: match.inline
 0.92% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   691
 1.55% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.44% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.92% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 2.64% :     0.000004s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.55% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.91% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000002s :     10: predicate.depend_value_elim
 0.95% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.51% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.24% :     0.000000s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.70% :     0.000002s :     10: predicate.environ_get_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 6.17% :     0.000008s :     22: predicate.inline
 1.12% :     0.000002s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000002s :      6: predicate.less_batch_normalization
 1.74% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 2.17% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.58% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 0.93% :     0.000001s :     10: predicate.minmaximum_grad
 2.07% :     0.000003s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 1.99% :     0.000003s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.41% :     0.000002s :     10: predicate.reduce_eliminate
 1.34% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.40% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.42% :     0.000001s :      3: predicate.reset_defer_inline
 1.57% :     0.000002s :     10: predicate.reshape_eliminate
 1.11% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.35% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     10: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.19% :     0.000002s :      6: predicate.special_op_eliminate
 0.85% :     0.000001s :      6: predicate.specialize_transform
 1.65% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.72% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.00% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.09% :     0.000010s :     53: predicate.switch_simplify
 1.61% :     0.000002s :     10: predicate.tile_eliminate
 0.96% :     0.000001s :     10: predicate.transpose_eliminate
 1.07% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.82% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.39% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.40% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.36% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.07% :     0.000003s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000809    12
47.44% :     0.000384s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.91% :     0.000064s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.64% :     0.000361s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.033954    76
 0.19% :     0.000064s :      1: add_recomputation
 0.22% :     0.000073s :      1: auto_monad
 0.06% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.44% :     0.000490s :      1: bootstrap
 0.11% :     0.000037s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.07% :     0.000023s :      1: event_method
 0.04% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
31.31% :     0.010630s :      1: jit_opt_a
 0.53% :     0.000179s :      1: jit_opt_after_cconv
 0.18% :     0.000061s :      1: jit_opt_b
 1.53% :     0.000519s :      1: loop_unroll
 2.91% :     0.000987s :      1: mutable_eliminate
 6.51% :     0.002209s :     26: opt.transform.jit_opt_a
 0.16% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000030s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.77% :     0.000601s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000045s :      1: py_interpret_to_execute
 0.05% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000019s :      1: remove_dup_value
 1.76% :     0.000597s :      1: renormalize.infer
 1.57% :     0.000534s :      1: renormalize.specialize
 0.04% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000045s :      1: rewriter_after_opt_a
 0.23% :     0.000078s :      1: rewriter_before_opt_a
 0.26% :     0.000089s :      1: symbol_engine_optimizer
30.92% :     0.010497s :      1: task_emit
17.00% :     0.005773s :      1: type_inference
 0.23% :     0.000077s :      1: validate


TotalTime = 0.025082, [33]
[bootstrap]: 0.00040506
[type_inference]: 0.0063962
[event_method]: 2.135e-05
[auto_monad]: 7.273e-05
[graph_reusing]: 6.07999e-06
[pre_auto_parallel]: 2.99999e-06
[py_interpret_to_execute]: 4.905e-05
[rewriter_before_opt_a]: 8.039e-05
[expand_dump_flag]: 3.08e-06
[jit_opt_a]: 0.0070727, [2]
    [Cycle 1]: 0.00235308, [27]
        [switch_simplify]: 7.297e-05
        [loop_unroll]: 3.187e-05
        [a_1]: 0.00068433
        [with_stream_mark]: 2.953e-05
        [recompute_prepare]: 1.122e-05
        [updatestate_depend_eliminate]: 3.81999e-06
        [updatestate_assign_eliminate]: 2.94001e-06
        [updatestate_loads_eliminate]: 3.06999e-06
        [parameter_eliminate]: 2.54999e-06
        [specialize_transform]: 6.47001e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 7.20998e-06
        [meta_shard_fg_expand]: 2.27001e-06
        [get_grad_eliminate_]: 5.54e-06
        [merge_forward]: 4.28001e-06
        [cell_reuse_recompute_pass]: 1.97001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.645e-05
        [j_node_and_user_rematch]: 1.101e-05
        [meta_fg_expand]: 3.06999e-06
        [replace_old_param]: 1.096e-05
        [inline_without_move]: 5.49998e-06
        [renormalize]: 0.00110075
        [add_forward_monad_depend]: 1.026e-05
        [auto_monad_grad]: 2.59999e-06
        [auto_monad_eliminator]: 2.136e-05
        [cse]: 3.486e-05
        [replace_applicator]: 2.219e-05
    [Cycle 2]: 0.00043369, [27]
        [switch_simplify]: 6.18998e-06
        [loop_unroll]: 5.22999e-06
        [a_1]: 0.00010467
        [with_stream_mark]: 1.766e-05
        [recompute_prepare]: 6.17001e-06
        [updatestate_depend_eliminate]: 3.33998e-06
        [updatestate_assign_eliminate]: 3.16999e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [parameter_eliminate]: 2.07001e-06
        [specialize_transform]: 5.63002e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 6.88998e-06
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 4.94998e-06
        [merge_forward]: 3.36999e-06
        [cell_reuse_recompute_pass]: 2.34999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.336e-05
        [j_node_and_user_rematch]: 1.142e-05
        [meta_fg_expand]: 3.12002e-06
        [replace_old_param]: 1.073e-05
        [inline_without_move]: 5.60001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 2.89999e-06
        [auto_monad_grad]: 1.03001e-06
        [auto_monad_eliminator]: 1.052e-05
        [cse]: 1.922e-05
        [replace_applicator]: 6.83e-06
[py_interpret_to_execute_after_opt_a]: 1.865e-05
[rewriter_after_opt_a]: 4.338e-05
[convert_after_rewriter]: 7.51001e-06
[order_py_execute_after_rewriter]: 5.80002e-06
[mutable_eliminate]: 0.00083028
[jit_opt_b]: 6.982e-05, [1]
    [Cycle 1]: 5.87e-05, [2]
        [frontend_op_eliminate]: 2.018e-05
        [inline_after_opt_a]: 2.548e-05
[cconv]: 4.074e-05
[loop_unroll]: 0.00054905
[jit_opt_after_cconv]: 0.00019171, [1]
    [Cycle 1]: 0.00018139, [11]
        [c_1]: 2.483e-05
        [parameter_eliminate]: 6.07001e-06
        [updatestate_depend_eliminate]: 1.059e-05
        [updatestate_assign_eliminate]: 3.67002e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [cse]: 3.478e-05
        [call_graph_tuple_transform]: 2.829e-05
        [tuple_list_get_item_eliminator]: 6.19001e-06
        [none_parameter_eliminate]: 1.86e-06
        [renormalize]: 8.30012e-07
        [switch_simplify]: 6.11e-06
[remove_dup_value]: 1.578e-05
[partial_unused_args_eliminate]: 2.54001e-06
[environ_conv]: 6.05002e-06
[add_recomputation]: 6.128e-05
[cse_after_recomputation]: 2.718e-05, [1]
    [Cycle 1]: 1.997e-05, [1]
        [cse]: 1.119e-05
[auto_monad_reorder]: 1.661e-05
[get_jit_bprop_graph]: 2.48002e-06
[rewriter_after_jit_bprop_graph]: 1.063e-05
[opt_after_jit_grad]: 0.00059849
[symbol_engine_optimizer]: 9.002e-05, [1]
    [Cycle 1]: 8.031e-05, [6]
        [build]: 6.78e-06
        [elim_shapecalc]: 9.02e-06
        [elim_not_effective]: 1.716e-05
        [opt_reshape]: 6.39001e-06
        [fold_const_symbol]: 1.053e-05
        [renormalize]: 7.79983e-07
[validate]: 4.957e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.00802055
[execute]: 1.006e-05
Sums
    bootstrap                            :     0.000405s :  2.03%
    type_inference                       :     0.006396s : 32.06%
    event_method                         :     0.000021s :  0.11%
    auto_monad                           :     0.000073s :  0.36%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000049s :  0.25%
    rewriter_before_opt_a                :     0.000080s :  0.40%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000079s :  0.40%
    jit_opt_a.loop_unroll                :     0.000037s :  0.19%
    jit_opt_a.a_1                        :     0.000789s :  3.95%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000014s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000022s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.03%
    jit_opt_a.replace_old_param          :     0.000022s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001101s :  5.52%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000032s :  0.16%
    jit_opt_a.cse                        :     0.000054s :  0.27%
    jit_opt_a.replace_applicator         :     0.000029s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.09%
    rewriter_after_opt_a                 :     0.000043s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000830s :  4.16%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.13%
    cconv                                :     0.000041s :  0.20%
    loop_unroll                          :     0.000549s :  2.75%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000035s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000061s :  0.31%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.05%
    opt_after_jit_grad                   :     0.000598s :  3.00%
    symbol_engine_optimizer.build        :     0.000007s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000050s :  0.25%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008021s : 40.20%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000302    23
 0.85% :     0.000003s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 2.32% :     0.000007s :      3: substitution.graph_param_transform
84.78% :     0.000256s :      4: substitution.inline
 1.68% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.39% :     0.000007s :      4: substitution.remove_not_recompute_node
 2.75% :     0.000008s :      2: substitution.replace_old_param
 4.80% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006305     2
80.37% :     0.005067s :      1: type_inference.infer
19.63% :     0.001238s :      1: type_inference.specialize
------[replace.]   0.000089     6
66.79% :     0.000060s :      4: replace.inline
33.21% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000266     6
94.99% :     0.000253s :      4: match.inline
 5.01% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000149   691
 1.47% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.17% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.83% :     0.000001s :     10: predicate.addn_check_dump
 1.07% :     0.000002s :     10: predicate.addn_zero_filter
 2.32% :     0.000003s :     10: predicate.arithmetic_simplify
 1.37% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.82% :     0.000001s :     10: predicate.compare_switch_simplify
 1.44% :     0.000002s :     10: predicate.depend_value_elim
 0.83% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.70% :     0.000003s :     10: predicate.dict_get_item_eliminator
 1.03% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.70% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.30% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.52% :     0.000002s :     10: predicate.environ_get_eliminate
 1.24% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.17% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 6.44% :     0.000010s :     22: predicate.inline
 1.05% :     0.000002s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.63% :     0.000002s :      6: predicate.less_batch_normalization
 1.82% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.91% :     0.000003s :     15: predicate.load_eliminater
 2.56% :     0.000004s :      3: predicate.loop_unroll_after_grad
 3.39% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.82% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.82% :     0.000001s :     10: predicate.merge_addn
 0.86% :     0.000001s :     10: predicate.minmaximum_grad
 3.18% :     0.000005s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 1.71% :     0.000003s :     15: predicate.partial_eliminate
 0.93% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.86% :     0.000003s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000003s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.59% :     0.000002s :     10: predicate.reshape_eliminate
 1.07% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.66% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.91% :     0.000001s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000002s :      6: predicate.special_op_eliminate
 0.72% :     0.000001s :      6: predicate.specialize_transform
 1.16% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.10% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.79% :     0.000004s :     16: predicate.switch_defer_inline
 1.63% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.93% :     0.000010s :     53: predicate.switch_simplify
 1.11% :     0.000002s :     10: predicate.tile_eliminate
 0.96% :     0.000001s :     10: predicate.transpose_eliminate
 1.78% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.56% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.12% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.13% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.26% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.38% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000853    12
45.02% :     0.000384s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.03% :     0.000060s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.95% :     0.000409s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027381    76
 0.24% :     0.000065s :      1: add_recomputation
 0.28% :     0.000077s :      1: auto_monad
 0.07% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.57% :     0.000429s :      1: bootstrap
 0.16% :     0.000044s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000030s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.10% :     0.000027s :      1: event_method
 0.06% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
25.85% :     0.007078s :      1: jit_opt_a
 0.71% :     0.000196s :      1: jit_opt_after_cconv
 0.27% :     0.000074s :      1: jit_opt_b
 2.06% :     0.000564s :      1: loop_unroll
 3.10% :     0.000850s :      1: mutable_eliminate
 3.86% :     0.001058s :     26: opt.transform.jit_opt_a
 0.23% :     0.000062s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000038s :      4: opt.transform.jit_opt_b
 0.07% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000027s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000039s :      4: opt.transform.symbol_engine_opt
 2.24% :     0.000614s :      1: opt_after_jit_grad
 0.03% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000052s :      1: py_interpret_to_execute
 0.08% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 2.15% :     0.000587s :      1: renormalize.infer
 1.83% :     0.000502s :      1: renormalize.specialize
 0.05% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000049s :      1: rewriter_after_opt_a
 0.31% :     0.000084s :      1: rewriter_before_opt_a
 0.34% :     0.000094s :      1: symbol_engine_optimizer
29.37% :     0.008043s :      1: task_emit
23.45% :     0.006421s :      1: type_inference
 0.28% :     0.000077s :      1: validate


TotalTime = 0.0344166, [33]
[bootstrap]: 0.00041327
[type_inference]: 0.00532163
[event_method]: 2.258e-05
[auto_monad]: 7.447e-05
[graph_reusing]: 7.4e-06
[pre_auto_parallel]: 2.76e-06
[py_interpret_to_execute]: 4.869e-05
[rewriter_before_opt_a]: 7.976e-05
[expand_dump_flag]: 3.58e-06
[jit_opt_a]: 0.0187155, [2]
    [Cycle 1]: 0.00244963, [27]
        [switch_simplify]: 7.735e-05
        [loop_unroll]: 3.105e-05
        [a_1]: 0.00069808
        [with_stream_mark]: 2.932e-05
        [recompute_prepare]: 9.17999e-06
        [updatestate_depend_eliminate]: 4.43001e-06
        [updatestate_assign_eliminate]: 2.90002e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [parameter_eliminate]: 2.07999e-06
        [specialize_transform]: 6.42001e-06
        [updatestate_useless_node_eliminater]: 5.39998e-06
        [accelerated_algorithm]: 6.21e-06
        [meta_shard_fg_expand]: 2.54999e-06
        [get_grad_eliminate_]: 5.12999e-06
        [merge_forward]: 4.05e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.622e-05
        [j_node_and_user_rematch]: 1e-05
        [meta_fg_expand]: 2.43e-06
        [replace_old_param]: 1.052e-05
        [inline_without_move]: 5.51e-06
        [renormalize]: 0.00120113
        [add_forward_monad_depend]: 1.035e-05
        [auto_monad_grad]: 3.28e-06
        [auto_monad_eliminator]: 2.15e-05
        [cse]: 3.611e-05
        [replace_applicator]: 2.113e-05
    [Cycle 2]: 0.00036097, [27]
        [switch_simplify]: 5.89999e-06
        [loop_unroll]: 6.56999e-06
        [a_1]: 0.0001031
        [with_stream_mark]: 1.33e-05
        [recompute_prepare]: 5.40999e-06
        [updatestate_depend_eliminate]: 3.51999e-06
        [updatestate_assign_eliminate]: 2.39999e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.56998e-06
        [specialize_transform]: 5.54e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 5.87999e-06
        [meta_shard_fg_expand]: 1.49998e-06
        [get_grad_eliminate_]: 5.00001e-06
        [merge_forward]: 3.67998e-06
        [cell_reuse_recompute_pass]: 3.75998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.757e-05
        [j_node_and_user_rematch]: 8.15e-06
        [meta_fg_expand]: 2.29999e-06
        [replace_old_param]: 8.54002e-06
        [inline_without_move]: 5.20001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.49e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 5.72001e-06
        [cse]: 1.118e-05
        [replace_applicator]: 4.80001e-06
[py_interpret_to_execute_after_opt_a]: 1.534e-05
[rewriter_after_opt_a]: 4.119e-05
[convert_after_rewriter]: 8.47998e-06
[order_py_execute_after_rewriter]: 4.87e-06
[mutable_eliminate]: 0.00085312
[jit_opt_b]: 5.655e-05, [1]
    [Cycle 1]: 4.674e-05, [2]
        [frontend_op_eliminate]: 1.785e-05
        [inline_after_opt_a]: 1.637e-05
[cconv]: 3.395e-05
[loop_unroll]: 0.00047058
[jit_opt_after_cconv]: 0.00016185, [1]
    [Cycle 1]: 0.00015487, [11]
        [c_1]: 2.222e-05
        [parameter_eliminate]: 4.07e-06
        [updatestate_depend_eliminate]: 8.79e-06
        [updatestate_assign_eliminate]: 2.61999e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [cse]: 2.453e-05
        [call_graph_tuple_transform]: 2.097e-05
        [tuple_list_get_item_eliminator]: 6.19001e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 8.70001e-07
        [switch_simplify]: 5.57999e-06
[remove_dup_value]: 1.525e-05
[partial_unused_args_eliminate]: 2.43002e-06
[environ_conv]: 6.17999e-06
[add_recomputation]: 5.79e-05
[cse_after_recomputation]: 2.328e-05, [1]
    [Cycle 1]: 1.627e-05, [1]
        [cse]: 9.66e-06
[auto_monad_reorder]: 1.678e-05
[get_jit_bprop_graph]: 2.32999e-06
[rewriter_after_jit_bprop_graph]: 8.17998e-06
[opt_after_jit_grad]: 0.00051677
[symbol_engine_optimizer]: 8.16e-05, [1]
    [Cycle 1]: 7.377e-05, [6]
        [build]: 4.77e-06
        [elim_shapecalc]: 8.74e-06
        [elim_not_effective]: 1.536e-05
        [opt_reshape]: 6.41e-06
        [fold_const_symbol]: 8.81002e-06
        [renormalize]: 5.99975e-07
[validate]: 4.683e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00705084
[execute]: 9.19998e-06
Sums
    bootstrap                            :     0.000413s :  2.32%
    type_inference                       :     0.005322s : 29.91%
    event_method                         :     0.000023s :  0.13%
    auto_monad                           :     0.000074s :  0.42%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000049s :  0.27%
    rewriter_before_opt_a                :     0.000080s :  0.45%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000083s :  0.47%
    jit_opt_a.loop_unroll                :     0.000038s :  0.21%
    jit_opt_a.a_1                        :     0.000801s :  4.50%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001201s :  6.75%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.15%
    jit_opt_a.cse                        :     0.000047s :  0.27%
    jit_opt_a.replace_applicator         :     0.000026s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.09%
    rewriter_after_opt_a                 :     0.000041s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000853s :  4.80%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000034s :  0.19%
    loop_unroll                          :     0.000471s :  2.65%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000058s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000517s :  2.90%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000047s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007051s : 39.63%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000323    23
 0.84% :     0.000003s :      2: substitution.elim_not_effective
 0.40% :     0.000001s :      2: substitution.fold_const_symbol
 1.89% :     0.000006s :      3: substitution.graph_param_transform
87.15% :     0.000281s :      4: substitution.inline
 1.17% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.89% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.98% :     0.000006s :      2: substitution.replace_old_param
 4.69% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005233     2
79.94% :     0.004184s :      1: type_inference.infer
20.06% :     0.001050s :      1: type_inference.specialize
------[replace.]   0.000089     6
69.11% :     0.000062s :      4: replace.inline
30.89% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000293     6
95.22% :     0.000279s :      4: match.inline
 4.78% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.53% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 2.39% :     0.000003s :     10: predicate.arithmetic_simplify
 1.61% :     0.000002s :     10: predicate.cast_eliminate
 0.57% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.45% :     0.000002s :     10: predicate.depend_value_elim
 1.25% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.42% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.50% :     0.000007s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.34% :     0.000002s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 1.78% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.22% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 0.92% :     0.000001s :     10: predicate.minmaximum_grad
 2.56% :     0.000003s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.03% :     0.000003s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.85% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.66% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.58% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     10: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.02% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.13% :     0.000003s :     16: predicate.switch_defer_inline
 1.91% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.94% :     0.000010s :     53: predicate.switch_simplify
 1.25% :     0.000002s :     10: predicate.tile_eliminate
 1.37% :     0.000002s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.53% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.12% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.42% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.32% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000779    12
45.34% :     0.000353s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.29% :     0.000065s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.37% :     0.000361s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.036776    76
 0.17% :     0.000061s :      1: add_recomputation
 0.22% :     0.000079s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.18% :     0.000435s :      1: bootstrap
 0.10% :     0.000037s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.07% :     0.000026s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.07% :     0.000027s :      1: event_method
 0.03% :     0.000013s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
50.90% :     0.018719s :      1: jit_opt_a
 0.45% :     0.000165s :      1: jit_opt_after_cconv
 0.16% :     0.000059s :      1: jit_opt_b
 1.31% :     0.000480s :      1: loop_unroll
 2.35% :     0.000865s :      1: mutable_eliminate
 2.86% :     0.001054s :     26: opt.transform.jit_opt_a
 0.14% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000027s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.43% :     0.000526s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000052s :      1: py_interpret_to_execute
 0.05% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 1.74% :     0.000638s :      1: renormalize.infer
 1.50% :     0.000551s :      1: renormalize.specialize
 0.03% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000047s :      1: rewriter_after_opt_a
 0.23% :     0.000083s :      1: rewriter_before_opt_a
 0.23% :     0.000084s :      1: symbol_engine_optimizer
19.22% :     0.007069s :      1: task_emit
14.53% :     0.005343s :      1: type_inference
 0.19% :     0.000071s :      1: validate


TotalTime = 0.0416008, [33]
[bootstrap]: 0.00040818
[type_inference]: 0.00938737
[event_method]: 1.904e-05
[auto_monad]: 6.91e-05
[graph_reusing]: 5.71e-06
[pre_auto_parallel]: 2.66999e-06
[py_interpret_to_execute]: 8.816e-05
[rewriter_before_opt_a]: 7.875e-05
[expand_dump_flag]: 3.27997e-06
[jit_opt_a]: 0.00907561, [2]
    [Cycle 1]: 0.00319235, [27]
        [switch_simplify]: 0.00117274
        [loop_unroll]: 3.239e-05
        [a_1]: 0.00061842
        [with_stream_mark]: 2.56e-05
        [recompute_prepare]: 7.25e-06
        [updatestate_depend_eliminate]: 4.08001e-06
        [updatestate_assign_eliminate]: 3.38999e-06
        [updatestate_loads_eliminate]: 2.81999e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 6.31e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.25001e-06
        [meta_shard_fg_expand]: 2.44001e-06
        [get_grad_eliminate_]: 5.18002e-06
        [merge_forward]: 3.5e-06
        [cell_reuse_recompute_pass]: 2.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.526e-05
        [j_node_and_user_rematch]: 9.36998e-06
        [meta_fg_expand]: 2.20002e-06
        [replace_old_param]: 9.86e-06
        [inline_without_move]: 5.39e-06
        [renormalize]: 0.00097952
        [add_forward_monad_depend]: 6.01e-06
        [auto_monad_grad]: 2.39999e-06
        [auto_monad_eliminator]: 1.922e-05
        [cse]: 3.23e-05
        [replace_applicator]: 1.548e-05
    [Cycle 2]: 0.00035434, [27]
        [switch_simplify]: 5.76998e-06
        [loop_unroll]: 5.37001e-06
        [a_1]: 9.574e-05
        [with_stream_mark]: 1.283e-05
        [recompute_prepare]: 5.51e-06
        [updatestate_depend_eliminate]: 3.7e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 6.06998e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 4.41002e-06
        [merge_forward]: 3.21999e-06
        [cell_reuse_recompute_pass]: 2.54001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.757e-05
        [j_node_and_user_rematch]: 9.07999e-06
        [meta_fg_expand]: 1.78002e-06
        [replace_old_param]: 8.61002e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.52001e-06
        [auto_monad_grad]: 8.89995e-07
        [auto_monad_eliminator]: 6.49999e-06
        [cse]: 1.139e-05
        [replace_applicator]: 5.20001e-06
[py_interpret_to_execute_after_opt_a]: 1.345e-05
[rewriter_after_opt_a]: 4.226e-05
[convert_after_rewriter]: 7.98999e-06
[order_py_execute_after_rewriter]: 5.06002e-06
[mutable_eliminate]: 0.0008364
[jit_opt_b]: 5.333e-05, [1]
    [Cycle 1]: 4.526e-05, [2]
        [frontend_op_eliminate]: 1.733e-05
        [inline_after_opt_a]: 1.543e-05
[cconv]: 3.341e-05
[loop_unroll]: 0.00045214
[jit_opt_after_cconv]: 0.00015322, [1]
    [Cycle 1]: 0.00014568, [11]
        [c_1]: 2.231e-05
        [parameter_eliminate]: 3.32002e-06
        [updatestate_depend_eliminate]: 7.65e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [cse]: 2.474e-05
        [call_graph_tuple_transform]: 2.005e-05
        [tuple_list_get_item_eliminator]: 5.24e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 8.50006e-07
        [switch_simplify]: 5.56e-06
[remove_dup_value]: 1.608e-05
[partial_unused_args_eliminate]: 2.59999e-06
[environ_conv]: 6.17001e-06
[add_recomputation]: 5.266e-05
[cse_after_recomputation]: 2.273e-05, [1]
    [Cycle 1]: 1.617e-05, [1]
        [cse]: 1.042e-05
[auto_monad_reorder]: 1.584e-05
[get_jit_bprop_graph]: 2.41998e-06
[rewriter_after_jit_bprop_graph]: 6.89999e-06
[opt_after_jit_grad]: 0.00045894
[symbol_engine_optimizer]: 7.58e-05, [1]
    [Cycle 1]: 6.943e-05, [6]
        [build]: 4.89e-06
        [elim_shapecalc]: 7.95e-06
        [elim_not_effective]: 1.318e-05
        [opt_reshape]: 5.71e-06
        [fold_const_symbol]: 9.07999e-06
        [renormalize]: 9.20001e-07
[validate]: 4.899e-05
[backend_pass]: 8.99978e-07
[task_emit]: 0.0199252
[execute]: 8.17e-06
Sums
    bootstrap                            :     0.000408s :  1.15%
    type_inference                       :     0.009387s : 26.51%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000069s :  0.20%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000088s :  0.25%
    rewriter_before_opt_a                :     0.000079s :  0.22%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.001179s :  3.33%
    jit_opt_a.loop_unroll                :     0.000038s :  0.11%
    jit_opt_a.a_1                        :     0.000714s :  2.02%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000018s :  0.05%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.000980s :  2.77%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.07%
    jit_opt_a.cse                        :     0.000044s :  0.12%
    jit_opt_a.replace_applicator         :     0.000021s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.04%
    rewriter_after_opt_a                 :     0.000042s :  0.12%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000836s :  2.36%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.04%
    cconv                                :     0.000033s :  0.09%
    loop_unroll                          :     0.000452s :  1.28%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000053s :  0.15%
    cse_after_recomputation.cse          :     0.000010s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000459s :  1.30%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000049s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.019925s : 56.28%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000273    23
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000002s :      2: substitution.fold_const_symbol
 2.06% :     0.000006s :      3: substitution.graph_param_transform
85.78% :     0.000234s :      4: substitution.inline
 1.35% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.49% :     0.000007s :      4: substitution.remove_not_recompute_node
 2.34% :     0.000006s :      2: substitution.replace_old_param
 4.66% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009305     2
89.64% :     0.008341s :      1: type_inference.infer
10.36% :     0.000964s :      1: type_inference.specialize
------[replace.]   0.000078     6
67.35% :     0.000052s :      4: replace.inline
32.65% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000243     6
95.21% :     0.000232s :      4: match.inline
 4.79% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.001215   691
 0.13% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.13% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.10% :     0.000001s :     10: predicate.addn_check_dump
 0.13% :     0.000002s :     10: predicate.addn_zero_filter
 0.23% :     0.000003s :     10: predicate.arithmetic_simplify
 0.15% :     0.000002s :     10: predicate.cast_eliminate
 0.05% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.10% :     0.000001s :     10: predicate.compare_switch_simplify
 0.18% :     0.000002s :     10: predicate.depend_value_elim
 0.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 0.14% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.03% :     0.000000s :      3: predicate.elim_not_effective
 0.07% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.13% :     0.000002s :     10: predicate.environ_get_eliminate
 0.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.02% :     0.000000s :      3: predicate.fold_const_symbol
 0.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.03% :     0.000000s :      3: predicate.graph_param_transform
 0.54% :     0.000007s :     22: predicate.inline
 0.09% :     0.000001s :      6: predicate.inline_without_move
 0.04% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.14% :     0.000002s :      6: predicate.less_batch_normalization
 0.17% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 0.19% :     0.000002s :     15: predicate.load_eliminater
 0.14% :     0.000002s :      3: predicate.loop_unroll_after_grad
 0.38% :     0.000005s :     34: predicate.loop_unroll_before_grad
 0.21% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.10% :     0.000001s :     10: predicate.merge_addn
 0.10% :     0.000001s :     10: predicate.minmaximum_grad
 0.30% :     0.000004s :      3: predicate.mutable_eliminate
 0.05% :     0.000001s :      3: predicate.opt_reshape
 0.22% :     0.000003s :     15: predicate.partial_eliminate
 0.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 0.14% :     0.000002s :     10: predicate.reduce_eliminate
 0.14% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.06% :     0.000001s :      6: predicate.remove_not_recompute_node
 0.15% :     0.000002s :     18: predicate.replace_applicator
 0.07% :     0.000001s :      6: predicate.replace_old_param
 0.05% :     0.000001s :      3: predicate.reset_defer_inline
 0.12% :     0.000001s :     10: predicate.reshape_eliminate
 0.14% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.11% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.11% :     0.000001s :     10: predicate.same_eliminate
 0.05% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.11% :     0.000001s :      6: predicate.special_op_eliminate
 0.09% :     0.000001s :      6: predicate.specialize_transform
 0.16% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 0.14% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.04% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 0.24% :     0.000003s :     16: predicate.switch_defer_inline
 0.20% :     0.000002s :     16: predicate.switch_layer_defer_inline
90.75% :     0.001103s :     53: predicate.switch_simplify
 0.13% :     0.000002s :     10: predicate.tile_eliminate
 0.11% :     0.000001s :     10: predicate.transpose_eliminate
 0.12% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.12% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 0.41% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 0.15% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 0.17% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 0.15% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 0.26% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 0.16% :     0.000002s :     10: predicate.value_based_eliminate
 0.04% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.06% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000760    12
50.49% :     0.000384s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.02% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.50% :     0.000323s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.044737    76
 0.12% :     0.000055s :      1: add_recomputation
 0.16% :     0.000073s :      1: auto_monad
 0.04% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.95% :     0.000424s :      1: bootstrap
 0.08% :     0.000036s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.06% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.05% :     0.000023s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
20.30% :     0.009081s :      1: jit_opt_a
 0.35% :     0.000156s :      1: jit_opt_after_cconv
 0.13% :     0.000057s :      1: jit_opt_b
 1.03% :     0.000461s :      1: loop_unroll
 1.90% :     0.000848s :      1: mutable_eliminate
 4.59% :     0.002054s :     26: opt.transform.jit_opt_a
 0.11% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000026s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.04% :     0.000466s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000092s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000018s :      1: remove_dup_value
 1.18% :     0.000529s :      1: renormalize.infer
 0.99% :     0.000442s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000046s :      1: rewriter_after_opt_a
 0.18% :     0.000082s :      1: rewriter_before_opt_a
 0.18% :     0.000078s :      1: symbol_engine_optimizer
44.57% :     0.019938s :      1: task_emit
21.02% :     0.009405s :      1: type_inference
 0.16% :     0.000073s :      1: validate


TotalTime = 0.0244158, [33]
[bootstrap]: 0.00035585
[type_inference]: 0.00742244
[event_method]: 2.152e-05
[auto_monad]: 7.599e-05
[graph_reusing]: 5.82001e-06
[pre_auto_parallel]: 2.52001e-06
[py_interpret_to_execute]: 4.871e-05
[rewriter_before_opt_a]: 8.025e-05
[expand_dump_flag]: 3.67002e-06
[jit_opt_a]: 0.00656479, [2]
    [Cycle 1]: 0.00210994, [27]
        [switch_simplify]: 6.942e-05
        [loop_unroll]: 3.009e-05
        [a_1]: 0.00062478
        [with_stream_mark]: 2.558e-05
        [recompute_prepare]: 8.70001e-06
        [updatestate_depend_eliminate]: 3.9e-06
        [updatestate_assign_eliminate]: 2.65002e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 6.11998e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 5.75001e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 5.15999e-06
        [merge_forward]: 3.73001e-06
        [cell_reuse_recompute_pass]: 1.24003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.5e-05
        [j_node_and_user_rematch]: 9.84999e-06
        [meta_fg_expand]: 2.13002e-06
        [replace_old_param]: 9.95002e-06
        [inline_without_move]: 5.46002e-06
        [renormalize]: 0.00100158
        [add_forward_monad_depend]: 6.91001e-06
        [auto_monad_grad]: 3.16999e-06
        [auto_monad_eliminator]: 1.424e-05
        [cse]: 3.211e-05
        [replace_applicator]: 1.518e-05
    [Cycle 2]: 0.00035604, [27]
        [switch_simplify]: 6.24001e-06
        [loop_unroll]: 5.59e-06
        [a_1]: 0.0001031
        [with_stream_mark]: 1.271e-05
        [recompute_prepare]: 5.31998e-06
        [updatestate_depend_eliminate]: 3.18998e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.79998e-06
        [specialize_transform]: 5.27001e-06
        [updatestate_useless_node_eliminater]: 5.66998e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 1.50001e-06
        [get_grad_eliminate_]: 4.93001e-06
        [merge_forward]: 2.80002e-06
        [cell_reuse_recompute_pass]: 1.54e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.437e-05
        [j_node_and_user_rematch]: 7.66001e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 7.76001e-06
        [inline_without_move]: 5.12e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.00002e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 5.58002e-06
        [cse]: 1.187e-05
        [replace_applicator]: 5.63997e-06
[py_interpret_to_execute_after_opt_a]: 1.411e-05
[rewriter_after_opt_a]: 3.822e-05
[convert_after_rewriter]: 7.38e-06
[order_py_execute_after_rewriter]: 5.10999e-06
[mutable_eliminate]: 0.00075574
[jit_opt_b]: 5.549e-05, [1]
    [Cycle 1]: 4.789e-05, [2]
        [frontend_op_eliminate]: 1.927e-05
        [inline_after_opt_a]: 1.571e-05
[cconv]: 3.118e-05
[loop_unroll]: 0.00049413
[jit_opt_after_cconv]: 0.00016141, [1]
    [Cycle 1]: 0.00015411, [11]
        [c_1]: 2.112e-05
        [parameter_eliminate]: 3.83999e-06
        [updatestate_depend_eliminate]: 8.37e-06
        [updatestate_assign_eliminate]: 2.85998e-06
        [updatestate_loads_eliminate]: 2.64999e-06
        [cse]: 2.657e-05
        [call_graph_tuple_transform]: 2.168e-05
        [tuple_list_get_item_eliminator]: 5.99e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 6.04001e-06
[remove_dup_value]: 1.548e-05
[partial_unused_args_eliminate]: 2.49001e-06
[environ_conv]: 6.45002e-06
[add_recomputation]: 5.485e-05
[cse_after_recomputation]: 2.263e-05, [1]
    [Cycle 1]: 1.623e-05, [1]
        [cse]: 9.36e-06
[auto_monad_reorder]: 1.489e-05
[get_jit_bprop_graph]: 1.99999e-06
[rewriter_after_jit_bprop_graph]: 9.14e-06
[opt_after_jit_grad]: 0.0005026
[symbol_engine_optimizer]: 8.122e-05, [1]
    [Cycle 1]: 7.376e-05, [6]
        [build]: 5.02e-06
        [elim_shapecalc]: 8.41002e-06
        [elim_not_effective]: 1.503e-05
        [opt_reshape]: 6.06998e-06
        [fold_const_symbol]: 9.41998e-06
        [renormalize]: 4.00003e-07
[validate]: 4.549e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00723988
[execute]: 1.132e-05
Sums
    bootstrap                            :     0.000356s :  1.82%
    type_inference                       :     0.007422s : 37.86%
    event_method                         :     0.000022s :  0.11%
    auto_monad                           :     0.000076s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000049s :  0.25%
    rewriter_before_opt_a                :     0.000080s :  0.41%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.39%
    jit_opt_a.loop_unroll                :     0.000036s :  0.18%
    jit_opt_a.a_1                        :     0.000728s :  3.71%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.001002s :  5.11%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.10%
    jit_opt_a.cse                        :     0.000044s :  0.22%
    jit_opt_a.replace_applicator         :     0.000021s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.19%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000756s :  3.86%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.000031s :  0.16%
    loop_unroll                          :     0.000494s :  2.52%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000055s :  0.28%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000503s :  2.56%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000045s :  0.23%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007240s : 36.93%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000274    23
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000002s :      2: substitution.fold_const_symbol
 2.03% :     0.000006s :      3: substitution.graph_param_transform
86.72% :     0.000238s :      4: substitution.inline
 1.46% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.84% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.79% :     0.000005s :      2: substitution.replace_old_param
 4.73% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007323     2
83.53% :     0.006117s :      1: type_inference.infer
16.47% :     0.001206s :      1: type_inference.specialize
------[replace.]   0.000081     6
67.78% :     0.000055s :      4: replace.inline
32.22% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000247     6
95.18% :     0.000235s :      4: match.inline
 4.82% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.27% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.55% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 2.97% :     0.000004s :     10: predicate.arithmetic_simplify
 1.50% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.66% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.14% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.04% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     10: predicate.environ_get_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 6.00% :     0.000008s :     22: predicate.inline
 0.77% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.33% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.59% :     0.000002s :     15: predicate.load_eliminater
 1.63% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.84% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.81% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.18% :     0.000001s :     10: predicate.minmaximum_grad
 2.97% :     0.000004s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.45% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     10: predicate.same_eliminate
 0.63% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 1.22% :     0.000002s :      6: predicate.specialize_transform
 1.17% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.60% :     0.000003s :     16: predicate.switch_defer_inline
 1.96% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.95% :     0.000010s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000002s :     10: predicate.transpose_eliminate
 1.64% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.06% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.52% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.92% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000835    12
44.17% :     0.000369s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.44% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.40% :     0.000412s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026482    76
 0.22% :     0.000058s :      1: add_recomputation
 0.30% :     0.000080s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.43% :     0.000380s :      1: bootstrap
 0.13% :     0.000035s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.10% :     0.000026s :      1: event_method
 0.06% :     0.000017s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
24.81% :     0.006570s :      1: jit_opt_a
 0.62% :     0.000164s :      1: jit_opt_after_cconv
 0.22% :     0.000059s :      1: jit_opt_b
 1.90% :     0.000503s :      1: loop_unroll
 2.90% :     0.000769s :      1: mutable_eliminate
 3.63% :     0.000961s :     26: opt.transform.jit_opt_a
 0.19% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.93% :     0.000512s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000051s :      1: py_interpret_to_execute
 0.06% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.14% :     0.000567s :      1: renormalize.infer
 1.60% :     0.000424s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000042s :      1: rewriter_after_opt_a
 0.32% :     0.000083s :      1: rewriter_before_opt_a
 0.32% :     0.000084s :      1: symbol_engine_optimizer
27.41% :     0.007260s :      1: task_emit
28.14% :     0.007452s :      1: type_inference
 0.26% :     0.000070s :      1: validate


TotalTime = 0.0152264, [30]
[bootstrap]: 0.00040418
[type_inference]: 0.00537625
[event_method]: 1.878e-05
[auto_monad]: 7.295e-05
[graph_reusing]: 5.51e-06
[pre_auto_parallel]: 2.56e-06
[py_interpret_to_execute]: 4.32e-05
[rewriter_before_opt_a]: 7.588e-05
[expand_dump_flag]: 3.47997e-06
[jit_opt_a]: 0.00654308, [2]
    [Cycle 1]: 0.00205607, [27]
        [switch_simplify]: 6.523e-05
        [loop_unroll]: 3.063e-05
        [a_1]: 0.0006349
        [with_stream_mark]: 2.454e-05
        [recompute_prepare]: 7.61999e-06
        [updatestate_depend_eliminate]: 3.90998e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 2.18998e-06
        [specialize_transform]: 6.44001e-06
        [updatestate_useless_node_eliminater]: 5.41002e-06
        [accelerated_algorithm]: 6.36e-06
        [meta_shard_fg_expand]: 2.30002e-06
        [get_grad_eliminate_]: 5.54e-06
        [merge_forward]: 3.92002e-06
        [cell_reuse_recompute_pass]: 1.03001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.558e-05
        [j_node_and_user_rematch]: 1.039e-05
        [meta_fg_expand]: 2.69001e-06
        [replace_old_param]: 1.012e-05
        [inline_without_move]: 6.17001e-06
        [renormalize]: 0.00095302
        [add_forward_monad_depend]: 6.74999e-06
        [auto_monad_grad]: 2.74001e-06
        [auto_monad_eliminator]: 1.578e-05
        [cse]: 3.106e-05
        [replace_applicator]: 1.509e-05
    [Cycle 2]: 0.00035033, [27]
        [switch_simplify]: 5.96e-06
        [loop_unroll]: 5.01002e-06
        [a_1]: 9.6e-05
        [with_stream_mark]: 1.251e-05
        [recompute_prepare]: 5.37999e-06
        [updatestate_depend_eliminate]: 2.83998e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 2.76999e-06
        [parameter_eliminate]: 1.11002e-06
        [specialize_transform]: 5.04998e-06
        [updatestate_useless_node_eliminater]: 5.48002e-06
        [accelerated_algorithm]: 5.24e-06
        [meta_shard_fg_expand]: 2.11998e-06
        [get_grad_eliminate_]: 4.85001e-06
        [merge_forward]: 3.16001e-06
        [cell_reuse_recompute_pass]: 2.24001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.544e-05
        [j_node_and_user_rematch]: 8.34998e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 7.78001e-06
        [inline_without_move]: 5.52001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.63002e-06
        [auto_monad_grad]: 1.01002e-06
        [auto_monad_eliminator]: 5.59998e-06
        [cse]: 1.226e-05
        [replace_applicator]: 5.46998e-06
[py_interpret_to_execute_after_opt_a]: 1.326e-05
[rewriter_after_opt_a]: 4.123e-05
[convert_after_rewriter]: 7.43e-06
[order_py_execute_after_rewriter]: 5.21998e-06
[mutable_eliminate]: 0.00087732
[jit_opt_b]: 6.819e-05, [1]
    [Cycle 1]: 5.787e-05, [2]
        [frontend_op_eliminate]: 2.546e-05
        [inline_after_opt_a]: 1.906e-05
[cconv]: 3.28e-05
[loop_unroll]: 0.00048867
[jit_opt_after_cconv]: 0.00017083, [1]
    [Cycle 1]: 0.00016408, [11]
        [c_1]: 2.289e-05
        [parameter_eliminate]: 5.19e-06
        [updatestate_depend_eliminate]: 9.06998e-06
        [updatestate_assign_eliminate]: 2.73998e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [cse]: 3.159e-05
        [call_graph_tuple_transform]: 2.285e-05
        [tuple_list_get_item_eliminator]: 5.60001e-06
        [none_parameter_eliminate]: 1.56002e-06
        [renormalize]: 7.50006e-07
        [switch_simplify]: 6.46999e-06
[remove_dup_value]: 1.564e-05
[partial_unused_args_eliminate]: 2.39999e-06
[environ_conv]: 6.71999e-06
[add_recomputation]: 5.639e-05
[cse_after_recomputation]: 2.121e-05, [1]
    [Cycle 1]: 1.582e-05, [1]
        [cse]: 9.15001e-06
[auto_monad_reorder]: 1.718e-05
[get_jit_bprop_graph]: 2.80002e-06
[rewriter_after_jit_bprop_graph]: 8.83001e-06
[opt_after_jit_grad]: 0.00050839
[symbol_engine_optimizer]: 8.283e-05, [1]
    [Cycle 1]: 7.527e-05, [6]
        [build]: 5.30001e-06
        [elim_shapecalc]: 8.37998e-06
        [elim_not_effective]: 1.516e-05
        [opt_reshape]: 6.29001e-06
        [fold_const_symbol]: 9.64999e-06
        [renormalize]: 2.89991e-07
[validate]: 5.042e-05
Sums
    bootstrap                            :     0.000404s :  3.87%
    type_inference                       :     0.005376s : 51.45%
    event_method                         :     0.000019s :  0.18%
    auto_monad                           :     0.000073s :  0.70%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.41%
    rewriter_before_opt_a                :     0.000076s :  0.73%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000071s :  0.68%
    jit_opt_a.loop_unroll                :     0.000036s :  0.34%
    jit_opt_a.a_1                        :     0.000731s :  6.99%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.17%
    jit_opt_a.inline_without_move        :     0.000012s :  0.11%
    jit_opt_a.renormalize                :     0.000953s :  9.12%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.20%
    jit_opt_a.cse                        :     0.000043s :  0.41%
    jit_opt_a.replace_applicator         :     0.000021s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.13%
    rewriter_after_opt_a                 :     0.000041s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000877s :  8.40%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.24%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.18%
    cconv                                :     0.000033s :  0.31%
    loop_unroll                          :     0.000489s :  4.68%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.09%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000032s :  0.30%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000016s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000007s :  0.06%
    add_recomputation                    :     0.000056s :  0.54%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000017s :  0.16%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.08%
    opt_after_jit_grad                   :     0.000508s :  4.87%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000050s :  0.48%


Time group info:
------[substitution.]   0.000277    23
 1.07% :     0.000003s :      2: substitution.elim_not_effective
 0.44% :     0.000001s :      2: substitution.fold_const_symbol
 2.35% :     0.000007s :      3: substitution.graph_param_transform
86.04% :     0.000238s :      4: substitution.inline
 1.33% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.91% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.66% :     0.000005s :      2: substitution.replace_old_param
 5.21% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005297     2
79.92% :     0.004234s :      1: type_inference.infer
20.08% :     0.001063s :      1: type_inference.specialize
------[replace.]   0.000086     6
67.56% :     0.000058s :      4: replace.inline
32.44% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000249     6
94.67% :     0.000236s :      4: match.inline
 5.33% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.49% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.60% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 2.12% :     0.000003s :     10: predicate.arithmetic_simplify
 1.07% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.08% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.11% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 6.41% :     0.000008s :     22: predicate.inline
 1.23% :     0.000002s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000002s :      6: predicate.less_batch_normalization
 1.30% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.60% :     0.000002s :     15: predicate.load_eliminater
 1.25% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.66% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.25% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 2.82% :     0.000004s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.33% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.90% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.27% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.74% :     0.000002s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 1.87% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000010s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.04% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.63% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000723    12
47.56% :     0.000344s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.56% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.88% :     0.000325s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017203    72
 0.35% :     0.000060s :      1: add_recomputation
 0.44% :     0.000076s :      1: auto_monad
 0.12% :     0.000021s :      1: auto_monad_reorder
 2.44% :     0.000420s :      1: bootstrap
 0.21% :     0.000036s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000010s :      1: environ_conv
 0.13% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.06% :     0.006547s :      1: jit_opt_a
 1.01% :     0.000174s :      1: jit_opt_after_cconv
 0.42% :     0.000072s :      1: jit_opt_b
 2.89% :     0.000498s :      1: loop_unroll
 5.18% :     0.000891s :      1: mutable_eliminate
 5.59% :     0.000962s :     26: opt.transform.jit_opt_a
 0.31% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000036s :      4: opt.transform.jit_opt_b
 0.09% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.01% :     0.000518s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000046s :      1: py_interpret_to_execute
 0.09% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 3.07% :     0.000529s :      1: renormalize.infer
 2.42% :     0.000416s :      1: renormalize.specialize
 0.07% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000044s :      1: rewriter_after_opt_a
 0.46% :     0.000079s :      1: rewriter_before_opt_a
 0.50% :     0.000086s :      1: symbol_engine_optimizer
31.37% :     0.005397s :      1: type_inference


TotalTime = 0.0240627, [33]
[bootstrap]: 0.00029701
[type_inference]: 0.00504383
[event_method]: 1.886e-05
[auto_monad]: 6.778e-05
[graph_reusing]: 5.65001e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 4.336e-05
[rewriter_before_opt_a]: 7.292e-05
[expand_dump_flag]: 3.28e-06
[jit_opt_a]: 0.00871783, [2]
    [Cycle 1]: 0.00205398, [27]
        [switch_simplify]: 7.253e-05
        [loop_unroll]: 3.027e-05
        [a_1]: 0.00065024
        [with_stream_mark]: 2.674e-05
        [recompute_prepare]: 8.15e-06
        [updatestate_depend_eliminate]: 4.22e-06
        [updatestate_assign_eliminate]: 3.56999e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 2.29001e-06
        [specialize_transform]: 6.02001e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 5.57999e-06
        [meta_shard_fg_expand]: 2.59001e-06
        [get_grad_eliminate_]: 5.64e-06
        [merge_forward]: 3.54002e-06
        [cell_reuse_recompute_pass]: 1.30001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.577e-05
        [j_node_and_user_rematch]: 9.72001e-06
        [meta_fg_expand]: 2.14e-06
        [replace_old_param]: 9.54e-06
        [inline_without_move]: 5.44998e-06
        [renormalize]: 0.00091201
        [add_forward_monad_depend]: 6.68e-06
        [auto_monad_grad]: 2.02001e-06
        [auto_monad_eliminator]: 1.585e-05
        [cse]: 3.179e-05
        [replace_applicator]: 1.387e-05
    [Cycle 2]: 0.00033366, [27]
        [switch_simplify]: 5.61e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 9.299e-05
        [with_stream_mark]: 1.344e-05
        [recompute_prepare]: 5.00999e-06
        [updatestate_depend_eliminate]: 2.74001e-06
        [updatestate_assign_eliminate]: 1.98997e-06
        [updatestate_loads_eliminate]: 1.95001e-06
        [parameter_eliminate]: 1.26997e-06
        [specialize_transform]: 5.17e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 5.15999e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 4.44002e-06
        [merge_forward]: 2.74001e-06
        [cell_reuse_recompute_pass]: 1.83997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.33e-05
        [j_node_and_user_rematch]: 8.05e-06
        [meta_fg_expand]: 1.61998e-06
        [replace_old_param]: 7.33e-06
        [inline_without_move]: 4.53001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.57999e-06
        [auto_monad_grad]: 1.02e-06
        [auto_monad_eliminator]: 5.76e-06
        [cse]: 1.197e-05
        [replace_applicator]: 4.89e-06
[py_interpret_to_execute_after_opt_a]: 1.262e-05
[rewriter_after_opt_a]: 3.922e-05
[convert_after_rewriter]: 6.46999e-06
[order_py_execute_after_rewriter]: 4.67998e-06
[mutable_eliminate]: 0.00073247
[jit_opt_b]: 5.487e-05, [1]
    [Cycle 1]: 4.639e-05, [2]
        [frontend_op_eliminate]: 1.798e-05
        [inline_after_opt_a]: 1.65e-05
[cconv]: 3.048e-05
[loop_unroll]: 0.00044996
[jit_opt_after_cconv]: 0.00015164, [1]
    [Cycle 1]: 0.00014492, [11]
        [c_1]: 2.135e-05
        [parameter_eliminate]: 3.21001e-06
        [updatestate_depend_eliminate]: 6.64999e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [cse]: 2.23e-05
        [call_graph_tuple_transform]: 2.181e-05
        [tuple_list_get_item_eliminator]: 5.55001e-06
        [none_parameter_eliminate]: 1.92999e-06
        [renormalize]: 5.09986e-07
        [switch_simplify]: 5.62001e-06
[remove_dup_value]: 1.451e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 5.73002e-06
[add_recomputation]: 5.44e-05
[cse_after_recomputation]: 2.071e-05, [1]
    [Cycle 1]: 1.493e-05, [1]
        [cse]: 9.39e-06
[auto_monad_reorder]: 1.562e-05
[get_jit_bprop_graph]: 2.32999e-06
[rewriter_after_jit_bprop_graph]: 7.59002e-06
[opt_after_jit_grad]: 0.00047422
[symbol_engine_optimizer]: 7.836e-05, [1]
    [Cycle 1]: 7.105e-05, [6]
        [build]: 4.55001e-06
        [elim_shapecalc]: 8.84e-06
        [elim_not_effective]: 1.396e-05
        [opt_reshape]: 6.02001e-06
        [fold_const_symbol]: 8.43999e-06
        [renormalize]: 8.89995e-07
[validate]: 4.327e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.00734299
[execute]: 1.004e-05
Sums
    bootstrap                            :     0.000297s :  1.74%
    type_inference                       :     0.005044s : 29.57%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000068s :  0.40%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.25%
    rewriter_before_opt_a                :     0.000073s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000078s :  0.46%
    jit_opt_a.loop_unroll                :     0.000035s :  0.21%
    jit_opt_a.a_1                        :     0.000743s :  4.36%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000912s :  5.35%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.13%
    jit_opt_a.cse                        :     0.000044s :  0.26%
    jit_opt_a.replace_applicator         :     0.000019s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000039s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000732s :  4.29%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.10%
    cconv                                :     0.000030s :  0.18%
    loop_unroll                          :     0.000450s :  2.64%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000054s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000474s :  2.78%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007343s : 43.05%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000282    23
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000001s :      2: substitution.fold_const_symbol
 2.22% :     0.000006s :      3: substitution.graph_param_transform
87.33% :     0.000246s :      4: substitution.inline
 1.20% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.65% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.68% :     0.000005s :      2: substitution.replace_old_param
 4.70% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004965     2
80.91% :     0.004017s :      1: type_inference.infer
19.09% :     0.000948s :      1: type_inference.specialize
------[replace.]   0.000078     6
69.00% :     0.000054s :      4: replace.inline
31.00% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000256     6
95.29% :     0.000244s :      4: match.inline
 4.71% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.23% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 2.38% :     0.000003s :     10: predicate.arithmetic_simplify
 1.44% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.08% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.73% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.40% :     0.000002s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.12% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 6.78% :     0.000008s :     22: predicate.inline
 1.02% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.25% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.15% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 2.17% :     0.000003s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.50% :     0.000001s :      3: predicate.reset_defer_inline
 1.29% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 1.97% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.45% :     0.000010s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.00% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000702    12
47.54% :     0.000334s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.42% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.04% :     0.000309s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026050    76
 0.22% :     0.000057s :      1: add_recomputation
 0.27% :     0.000071s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.21% :     0.000316s :      1: bootstrap
 0.13% :     0.000034s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.09% :     0.000023s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
33.48% :     0.008721s :      1: jit_opt_a
 0.59% :     0.000154s :      1: jit_opt_after_cconv
 0.22% :     0.000058s :      1: jit_opt_b
 1.77% :     0.000460s :      1: loop_unroll
 2.86% :     0.000744s :      1: mutable_eliminate
 3.73% :     0.000972s :     26: opt.transform.jit_opt_a
 0.19% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.86% :     0.000484s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000046s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.93% :     0.000503s :      1: renormalize.infer
 1.53% :     0.000399s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000042s :      1: rewriter_after_opt_a
 0.29% :     0.000076s :      1: rewriter_before_opt_a
 0.31% :     0.000081s :      1: symbol_engine_optimizer
28.27% :     0.007364s :      1: task_emit
19.44% :     0.005064s :      1: type_inference
 0.25% :     0.000066s :      1: validate


TotalTime = 0.043014, [33]
[bootstrap]: 0.00039138
[type_inference]: 0.0125678
[event_method]: 2.414e-05
[auto_monad]: 7.734e-05
[graph_reusing]: 6.84999e-06
[pre_auto_parallel]: 2.69999e-06
[py_interpret_to_execute]: 5.817e-05
[rewriter_before_opt_a]: 8.556e-05
[expand_dump_flag]: 3.35e-06
[jit_opt_a]: 0.0105567, [2]
    [Cycle 1]: 0.0039899, [27]
        [switch_simplify]: 7.396e-05
        [loop_unroll]: 3e-05
        [a_1]: 0.00061844
        [with_stream_mark]: 2.065e-05
        [recompute_prepare]: 7.13998e-06
        [updatestate_depend_eliminate]: 3.66999e-06
        [updatestate_assign_eliminate]: 2.90002e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 6.24001e-06
        [updatestate_useless_node_eliminater]: 4.97999e-06
        [accelerated_algorithm]: 5.96e-06
        [meta_shard_fg_expand]: 2.40002e-06
        [get_grad_eliminate_]: 5.37999e-06
        [merge_forward]: 3.86999e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.664e-05
        [j_node_and_user_rematch]: 1.02e-05
        [meta_fg_expand]: 2.46998e-06
        [replace_old_param]: 9.66998e-06
        [inline_without_move]: 5.62999e-06
        [renormalize]: 0.00283485
        [add_forward_monad_depend]: 1.297e-05
        [auto_monad_grad]: 2.60002e-06
        [auto_monad_eliminator]: 2.665e-05
        [cse]: 3.389e-05
        [replace_applicator]: 2.898e-05
    [Cycle 2]: 0.00038032, [27]
        [switch_simplify]: 6.52001e-06
        [loop_unroll]: 6.47001e-06
        [a_1]: 0.00011487
        [with_stream_mark]: 1.606e-05
        [recompute_prepare]: 5.54e-06
        [updatestate_depend_eliminate]: 3.41001e-06
        [updatestate_assign_eliminate]: 2.76999e-06
        [updatestate_loads_eliminate]: 3.00998e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 6.13002e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 5.84e-06
        [meta_shard_fg_expand]: 2.83998e-06
        [get_grad_eliminate_]: 5.22e-06
        [merge_forward]: 3.73001e-06
        [cell_reuse_recompute_pass]: 4.79e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.939e-05
        [j_node_and_user_rematch]: 9.59e-06
        [meta_fg_expand]: 2.68e-06
        [replace_old_param]: 9.15001e-06
        [inline_without_move]: 5.38002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 1.15999e-06
        [auto_monad_eliminator]: 5.54e-06
        [cse]: 1.042e-05
        [replace_applicator]: 5.59e-06
[py_interpret_to_execute_after_opt_a]: 1.977e-05
[rewriter_after_opt_a]: 4.081e-05
[convert_after_rewriter]: 7.17002e-06
[order_py_execute_after_rewriter]: 4.85001e-06
[mutable_eliminate]: 0.00091829
[jit_opt_b]: 5.944e-05, [1]
    [Cycle 1]: 4.83e-05, [2]
        [frontend_op_eliminate]: 1.991e-05
        [inline_after_opt_a]: 1.616e-05
[cconv]: 3.332e-05
[loop_unroll]: 0.00051267
[jit_opt_after_cconv]: 0.00203225, [1]
    [Cycle 1]: 0.00025458, [11]
        [c_1]: 5.506e-05
        [parameter_eliminate]: 5.61e-06
        [updatestate_depend_eliminate]: 1.088e-05
        [updatestate_assign_eliminate]: 3.76001e-06
        [updatestate_loads_eliminate]: 3.63e-06
        [cse]: 4.247e-05
        [call_graph_tuple_transform]: 3.104e-05
        [tuple_list_get_item_eliminator]: 7.00002e-06
        [none_parameter_eliminate]: 1.70001e-06
        [renormalize]: 5.19998e-07
        [switch_simplify]: 6.45997e-06
[remove_dup_value]: 1.851e-05
[partial_unused_args_eliminate]: 2.73e-06
[environ_conv]: 7.5e-06
[add_recomputation]: 6.026e-05
[cse_after_recomputation]: 2.797e-05, [1]
    [Cycle 1]: 1.925e-05, [1]
        [cse]: 1.241e-05
[auto_monad_reorder]: 1.784e-05
[get_jit_bprop_graph]: 2.33998e-06
[rewriter_after_jit_bprop_graph]: 1.271e-05
[opt_after_jit_grad]: 0.00099727
[symbol_engine_optimizer]: 0.00010435, [1]
    [Cycle 1]: 9.407e-05, [6]
        [build]: 8.1e-06
        [elim_shapecalc]: 1.042e-05
        [elim_not_effective]: 2.086e-05
        [opt_reshape]: 8.15999e-06
        [fold_const_symbol]: 1.009e-05
        [renormalize]: 1.14e-06
[validate]: 5.786e-05
[backend_pass]: 1.19e-06
[task_emit]: 0.0140116
[execute]: 8.45999e-06
Sums
    bootstrap                            :     0.000391s :  1.14%
    type_inference                       :     0.012568s : 36.67%
    event_method                         :     0.000024s :  0.07%
    auto_monad                           :     0.000077s :  0.23%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000058s :  0.17%
    rewriter_before_opt_a                :     0.000086s :  0.25%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000080s :  0.23%
    jit_opt_a.loop_unroll                :     0.000036s :  0.11%
    jit_opt_a.a_1                        :     0.000733s :  2.14%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.03%
    jit_opt_a.merge_forward              :     0.000008s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.05%
    jit_opt_a.inline_without_move        :     0.000011s :  0.03%
    jit_opt_a.renormalize                :     0.002835s :  8.27%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000032s :  0.09%
    jit_opt_a.cse                        :     0.000044s :  0.13%
    jit_opt_a.replace_applicator         :     0.000035s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000020s :  0.06%
    rewriter_after_opt_a                 :     0.000041s :  0.12%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000918s :  2.68%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.05%
    cconv                                :     0.000033s :  0.10%
    loop_unroll                          :     0.000513s :  1.50%
    jit_opt_after_cconv.c_1              :     0.000055s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000042s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000031s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000019s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000060s :  0.18%
    cse_after_recomputation.cse          :     0.000012s :  0.04%
    auto_monad_reorder                   :     0.000018s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000013s :  0.04%
    opt_after_jit_grad                   :     0.000997s :  2.91%
    symbol_engine_optimizer.build        :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000058s :  0.17%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.014012s : 40.89%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000262    23
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.78% :     0.000002s :      2: substitution.fold_const_symbol
 3.05% :     0.000008s :      3: substitution.graph_param_transform
83.76% :     0.000219s :      4: substitution.inline
 1.71% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.09% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.37% :     0.000006s :      2: substitution.replace_old_param
 5.38% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012465     2
57.75% :     0.007198s :      1: type_inference.infer
42.25% :     0.005267s :      1: type_inference.specialize
------[replace.]   0.000083     6
67.01% :     0.000056s :      4: replace.inline
32.99% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000230     6
94.32% :     0.000217s :      4: match.inline
 5.68% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000149   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.54% :     0.000004s :      3: predicate.ad_related_special_op_eliminate
 0.82% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 2.60% :     0.000004s :     10: predicate.arithmetic_simplify
 1.60% :     0.000002s :     10: predicate.cast_eliminate
 0.29% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.80% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 0.89% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.83% :     0.000003s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.30% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000001s :      3: predicate.elim_not_effective
 0.76% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.05% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.82% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.42% :     0.000002s :     10: predicate.environ_get_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.16% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.46% :     0.000001s :      3: predicate.graph_param_transform
 8.06% :     0.000012s :     22: predicate.inline
 0.68% :     0.000001s :      6: predicate.inline_without_move
 0.28% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :      6: predicate.less_batch_normalization
 1.27% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.55% :     0.000002s :     15: predicate.load_eliminater
 1.79% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.37% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.29% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 1.66% :     0.000002s :      3: predicate.mutable_eliminate
 0.69% :     0.000001s :      3: predicate.opt_reshape
 1.84% :     0.000003s :     15: predicate.partial_eliminate
 0.89% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.27% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.53% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.21% :     0.000000s :      3: predicate.reset_defer_inline
 1.70% :     0.000003s :     10: predicate.reshape_eliminate
 1.01% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.89% :     0.000001s :     10: predicate.same_eliminate
 0.41% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.33% :     0.000002s :      6: predicate.special_op_eliminate
 0.73% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.56% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 1.82% :     0.000003s :     16: predicate.switch_defer_inline
 1.71% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.80% :     0.000010s :     53: predicate.switch_simplify
 1.19% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000002s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.97% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.43% :     0.000007s :     18: predicate.tuple_list_get_item_eliminator
 1.88% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.27% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.41% :     0.000005s :     21: predicate.updatestate_useless_node_eliminater
 2.09% :     0.000003s :     10: predicate.value_based_eliminate
 0.25% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.52% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000907    12
48.57% :     0.000441s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 3.97% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.46% :     0.000431s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.047009    76
 0.14% :     0.000064s :      1: add_recomputation
 0.17% :     0.000081s :      1: auto_monad
 0.04% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.88% :     0.000412s :      1: bootstrap
 0.08% :     0.000036s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.06% :     0.000030s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.06% :     0.000028s :      1: event_method
 0.03% :     0.000014s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
22.47% :     0.010561s :      1: jit_opt_a
 4.33% :     0.002038s :      1: jit_opt_after_cconv
 0.13% :     0.000062s :      1: jit_opt_b
 1.11% :     0.000522s :      1: loop_unroll
 1.98% :     0.000932s :      1: mutable_eliminate
 2.11% :     0.000990s :     26: opt.transform.jit_opt_a
 0.19% :     0.000091s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000029s :      4: opt.transform.jit_opt_b
 0.03% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000044s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000045s :      4: opt.transform.symbol_engine_opt
 2.17% :     0.001021s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000061s :      1: py_interpret_to_execute
 0.05% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000021s :      1: remove_dup_value
 1.04% :     0.000491s :      1: renormalize.infer
 4.96% :     0.002330s :      1: renormalize.specialize
 0.03% :     0.000015s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000044s :      1: rewriter_after_opt_a
 0.19% :     0.000088s :      1: rewriter_before_opt_a
 0.23% :     0.000108s :      1: symbol_engine_optimizer
29.85% :     0.014034s :      1: task_emit
26.77% :     0.012586s :      1: type_inference
 0.19% :     0.000090s :      1: validate


TotalTime = 0.0446936, [33]
[bootstrap]: 0.00039917
[type_inference]: 0.00625894
[event_method]: 2.276e-05
[auto_monad]: 7.052e-05
[graph_reusing]: 6.98998e-06
[pre_auto_parallel]: 2.36e-06
[py_interpret_to_execute]: 5.109e-05
[rewriter_before_opt_a]: 8.049e-05
[expand_dump_flag]: 3.48e-06
[jit_opt_a]: 0.0272596, [2]
    [Cycle 1]: 0.00213039, [27]
        [switch_simplify]: 6.679e-05
        [loop_unroll]: 3.095e-05
        [a_1]: 0.00061654
        [with_stream_mark]: 2.846e-05
        [recompute_prepare]: 1.089e-05
        [updatestate_depend_eliminate]: 4.47e-06
        [updatestate_assign_eliminate]: 3.05998e-06
        [updatestate_loads_eliminate]: 3.21001e-06
        [parameter_eliminate]: 2.51998e-06
        [specialize_transform]: 6.06e-06
        [updatestate_useless_node_eliminater]: 5.33002e-06
        [accelerated_algorithm]: 5.62001e-06
        [meta_shard_fg_expand]: 2.10002e-06
        [get_grad_eliminate_]: 5.60001e-06
        [merge_forward]: 4.24002e-06
        [cell_reuse_recompute_pass]: 1.72999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.592e-05
        [j_node_and_user_rematch]: 1.006e-05
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 1.036e-05
        [inline_without_move]: 5.52999e-06
        [renormalize]: 0.0010012
        [add_forward_monad_depend]: 7.55998e-06
        [auto_monad_grad]: 3.38999e-06
        [auto_monad_eliminator]: 2.071e-05
        [cse]: 3.137e-05
        [replace_applicator]: 1.933e-05
    [Cycle 2]: 0.00035371, [27]
        [switch_simplify]: 6.39999e-06
        [loop_unroll]: 5.56002e-06
        [a_1]: 9.834e-05
        [with_stream_mark]: 1.407e-05
        [recompute_prepare]: 5.57001e-06
        [updatestate_depend_eliminate]: 3.18998e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.08002e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 5.19998e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 2.09999e-06
        [get_grad_eliminate_]: 4.33999e-06
        [merge_forward]: 3.5e-06
        [cell_reuse_recompute_pass]: 1.92001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.564e-05
        [j_node_and_user_rematch]: 9.02e-06
        [meta_fg_expand]: 1.77001e-06
        [replace_old_param]: 7.53999e-06
        [inline_without_move]: 1.309e-05
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 6.58e-06
        [cse]: 1.146e-05
        [replace_applicator]: 5.04e-06
[py_interpret_to_execute_after_opt_a]: 1.77e-05
[rewriter_after_opt_a]: 4.489e-05
[convert_after_rewriter]: 8.08999e-06
[order_py_execute_after_rewriter]: 5.17e-06
[mutable_eliminate]: 0.00084466
[jit_opt_b]: 5.666e-05, [1]
    [Cycle 1]: 4.68e-05, [2]
        [frontend_op_eliminate]: 1.887e-05
        [inline_after_opt_a]: 1.537e-05
[cconv]: 3.621e-05
[loop_unroll]: 0.00117778
[jit_opt_after_cconv]: 0.00018531, [1]
    [Cycle 1]: 0.00017662, [11]
        [c_1]: 2.35e-05
        [parameter_eliminate]: 7.48e-06
        [updatestate_depend_eliminate]: 1.016e-05
        [updatestate_assign_eliminate]: 2.91999e-06
        [updatestate_loads_eliminate]: 3.38999e-06
        [cse]: 3.67e-05
        [call_graph_tuple_transform]: 2.664e-05
        [tuple_list_get_item_eliminator]: 5.55001e-06
        [none_parameter_eliminate]: 1.40999e-06
        [renormalize]: 8.2e-07
        [switch_simplify]: 5.99e-06
[remove_dup_value]: 1.648e-05
[partial_unused_args_eliminate]: 2.12001e-06
[environ_conv]: 5.92001e-06
[add_recomputation]: 5.565e-05
[cse_after_recomputation]: 2.127e-05, [1]
    [Cycle 1]: 1.487e-05, [1]
        [cse]: 9.19e-06
[auto_monad_reorder]: 1.589e-05
[get_jit_bprop_graph]: 1.99999e-06
[rewriter_after_jit_bprop_graph]: 8.10999e-06
[opt_after_jit_grad]: 0.00057386
[symbol_engine_optimizer]: 8.141e-05, [1]
    [Cycle 1]: 7.301e-05, [6]
        [build]: 3.90998e-06
        [elim_shapecalc]: 9.18002e-06
        [elim_not_effective]: 1.513e-05
        [opt_reshape]: 6.43e-06
        [fold_const_symbol]: 9.74e-06
        [renormalize]: 1.12e-06
[validate]: 4.738e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00705264
[execute]: 9.10999e-06
Sums
    bootstrap                            :     0.000399s :  2.08%
    type_inference                       :     0.006259s : 32.60%
    event_method                         :     0.000023s :  0.12%
    auto_monad                           :     0.000071s :  0.37%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000051s :  0.27%
    rewriter_before_opt_a                :     0.000080s :  0.42%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000073s :  0.38%
    jit_opt_a.loop_unroll                :     0.000037s :  0.19%
    jit_opt_a.a_1                        :     0.000715s :  3.72%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.09%
    jit_opt_a.inline_without_move        :     0.000019s :  0.10%
    jit_opt_a.renormalize                :     0.001001s :  5.22%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.14%
    jit_opt_a.cse                        :     0.000043s :  0.22%
    jit_opt_a.replace_applicator         :     0.000024s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.09%
    rewriter_after_opt_a                 :     0.000045s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000845s :  4.40%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000036s :  0.19%
    loop_unroll                          :     0.001178s :  6.13%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000037s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000056s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000574s :  2.99%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000047s :  0.25%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007053s : 36.74%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000261    23
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000002s :      2: substitution.fold_const_symbol
 2.65% :     0.000007s :      3: substitution.graph_param_transform
85.20% :     0.000223s :      4: substitution.inline
 1.60% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.79% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.82% :     0.000005s :      2: substitution.replace_old_param
 5.55% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006165     2
80.96% :     0.004991s :      1: type_inference.infer
19.04% :     0.001174s :      1: type_inference.specialize
------[replace.]   0.000083     6
67.06% :     0.000055s :      4: replace.inline
32.94% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000233     6
94.34% :     0.000220s :      4: match.inline
 5.66% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.30% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.43% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.19% :     0.000002s :     10: predicate.addn_zero_filter
 1.87% :     0.000002s :     10: predicate.arithmetic_simplify
 1.70% :     0.000002s :     10: predicate.cast_eliminate
 0.55% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 2.05% :     0.000003s :     10: predicate.depend_value_elim
 0.96% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.02% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.47% :     0.000001s :      3: predicate.elim_not_effective
 0.86% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     10: predicate.environ_get_eliminate
 1.18% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.41% :     0.000001s :      3: predicate.graph_param_transform
 5.54% :     0.000007s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.64% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 2.19% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.77% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 2.58% :     0.000003s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.31% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.90% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.25% :     0.000002s :     10: predicate.reshape_eliminate
 1.06% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.05% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.14% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.08% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.53% :     0.000011s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.17% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.08% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.35% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.43% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.00% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000773    12
44.75% :     0.000346s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.30% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.95% :     0.000379s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.046778    76
 0.13% :     0.000059s :      1: add_recomputation
 0.16% :     0.000074s :      1: auto_monad
 0.04% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.90% :     0.000421s :      1: bootstrap
 0.08% :     0.000039s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.05% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.06% :     0.000028s :      1: event_method
 0.03% :     0.000015s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
58.28% :     0.027264s :      1: jit_opt_a
 0.40% :     0.000189s :      1: jit_opt_after_cconv
 0.13% :     0.000060s :      1: jit_opt_b
 2.55% :     0.001194s :      1: loop_unroll
 1.84% :     0.000860s :      1: mutable_eliminate
 2.06% :     0.000962s :     26: opt.transform.jit_opt_a
 0.12% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000022s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.25% :     0.000584s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000054s :      1: py_interpret_to_execute
 0.04% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000019s :      1: remove_dup_value
 1.20% :     0.000562s :      1: renormalize.infer
 0.92% :     0.000429s :      1: renormalize.specialize
 0.02% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000048s :      1: rewriter_after_opt_a
 0.18% :     0.000084s :      1: rewriter_before_opt_a
 0.18% :     0.000085s :      1: symbol_engine_optimizer
15.13% :     0.007075s :      1: task_emit
13.44% :     0.006285s :      1: type_inference
 0.15% :     0.000072s :      1: validate


TotalTime = 0.0214719, [33]
[bootstrap]: 0.00034843
[type_inference]: 0.00527503
[event_method]: 1.974e-05
[auto_monad]: 5.989e-05
[graph_reusing]: 4.79998e-06
[pre_auto_parallel]: 2.39999e-06
[py_interpret_to_execute]: 4.371e-05
[rewriter_before_opt_a]: 7.408e-05
[expand_dump_flag]: 3.73001e-06
[jit_opt_a]: 0.00675234, [2]
    [Cycle 1]: 0.00272034, [27]
        [switch_simplify]: 6.139e-05
        [loop_unroll]: 3.028e-05
        [a_1]: 0.0005675
        [with_stream_mark]: 2.008e-05
        [recompute_prepare]: 7.65998e-06
        [updatestate_depend_eliminate]: 3.19001e-06
        [updatestate_assign_eliminate]: 3.48999e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [parameter_eliminate]: 1.75001e-06
        [specialize_transform]: 6.02999e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 6.82002e-06
        [meta_shard_fg_expand]: 1.77999e-06
        [get_grad_eliminate_]: 5.19e-06
        [merge_forward]: 3.38e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.335e-05
        [j_node_and_user_rematch]: 9.05001e-06
        [meta_fg_expand]: 2.01998e-06
        [replace_old_param]: 8.43001e-06
        [inline_without_move]: 5.40001e-06
        [renormalize]: 0.00170199
        [add_forward_monad_depend]: 7.62998e-06
        [auto_monad_grad]: 2.02001e-06
        [auto_monad_eliminator]: 1.626e-05
        [cse]: 2.883e-05
        [replace_applicator]: 1.473e-05
    [Cycle 2]: 0.00036801, [27]
        [switch_simplify]: 6.04999e-06
        [loop_unroll]: 5.77001e-06
        [a_1]: 0.00010104
        [with_stream_mark]: 1.459e-05
        [recompute_prepare]: 5.20999e-06
        [updatestate_depend_eliminate]: 2.99001e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [parameter_eliminate]: 1.49e-06
        [specialize_transform]: 4.95999e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.72001e-06
        [meta_shard_fg_expand]: 2.33002e-06
        [get_grad_eliminate_]: 4.60999e-06
        [merge_forward]: 2.88e-06
        [cell_reuse_recompute_pass]: 2.83998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.488e-05
        [j_node_and_user_rematch]: 8.74998e-06
        [meta_fg_expand]: 1.64e-06
        [replace_old_param]: 9.10001e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.59998e-06
        [auto_monad_grad]: 7.90023e-07
        [auto_monad_eliminator]: 5.77999e-06
        [cse]: 1.243e-05
        [replace_applicator]: 4.99e-06
[py_interpret_to_execute_after_opt_a]: 1.668e-05
[rewriter_after_opt_a]: 3.316e-05
[convert_after_rewriter]: 6.28e-06
[order_py_execute_after_rewriter]: 4.53999e-06
[mutable_eliminate]: 0.00064009
[jit_opt_b]: 5.297e-05, [1]
    [Cycle 1]: 4.576e-05, [2]
        [frontend_op_eliminate]: 1.796e-05
        [inline_after_opt_a]: 1.605e-05
[cconv]: 2.175e-05
[loop_unroll]: 0.00042726
[jit_opt_after_cconv]: 0.00014767, [1]
    [Cycle 1]: 0.00014094, [11]
        [c_1]: 2.231e-05
        [parameter_eliminate]: 2.27999e-06
        [updatestate_depend_eliminate]: 5.91e-06
        [updatestate_assign_eliminate]: 2.90002e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [cse]: 2.195e-05
        [call_graph_tuple_transform]: 1.87e-05
        [tuple_list_get_item_eliminator]: 5.61998e-06
        [none_parameter_eliminate]: 1.12e-06
        [renormalize]: 5.09986e-07
        [switch_simplify]: 5.65001e-06
[remove_dup_value]: 8.99e-06
[partial_unused_args_eliminate]: 1.49998e-06
[environ_conv]: 4.2e-06
[add_recomputation]: 3.773e-05
[cse_after_recomputation]: 2.259e-05, [1]
    [Cycle 1]: 1.661e-05, [1]
        [cse]: 1.038e-05
[auto_monad_reorder]: 1.141e-05
[get_jit_bprop_graph]: 1.71e-06
[rewriter_after_jit_bprop_graph]: 6.89999e-06
[opt_after_jit_grad]: 0.00046361
[symbol_engine_optimizer]: 7.907e-05, [1]
    [Cycle 1]: 7.145e-05, [6]
        [build]: 4.48999e-06
        [elim_shapecalc]: 8.67998e-06
        [elim_not_effective]: 1.343e-05
        [opt_reshape]: 5.94e-06
        [fold_const_symbol]: 9.19e-06
        [renormalize]: 5.39992e-07
[validate]: 3.485e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00661914
[execute]: 9.04e-06
Sums
    bootstrap                            :     0.000348s :  2.03%
    type_inference                       :     0.005275s : 30.78%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000060s :  0.35%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.26%
    rewriter_before_opt_a                :     0.000074s :  0.43%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.39%
    jit_opt_a.loop_unroll                :     0.000036s :  0.21%
    jit_opt_a.a_1                        :     0.000669s :  3.90%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001702s :  9.93%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.13%
    jit_opt_a.cse                        :     0.000041s :  0.24%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.10%
    rewriter_after_opt_a                 :     0.000033s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000640s :  3.73%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000022s :  0.13%
    loop_unroll                          :     0.000427s :  2.49%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000009s :  0.05%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000038s :  0.22%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000011s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000464s :  2.71%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006619s : 38.62%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000220    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.05% :     0.000005s :      3: substitution.graph_param_transform
84.79% :     0.000186s :      4: substitution.inline
 1.67% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.39% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.42% :     0.000005s :      2: substitution.replace_old_param
 5.18% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005192     2
80.59% :     0.004184s :      1: type_inference.infer
19.41% :     0.001008s :      1: type_inference.specialize
------[replace.]   0.000076     6
66.70% :     0.000051s :      4: replace.inline
33.30% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000194     6
94.92% :     0.000184s :      4: match.inline
 5.08% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.20% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.51% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.66% :     0.000003s :     10: predicate.arithmetic_simplify
 1.92% :     0.000002s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.79% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.12% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.09% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.05% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.22% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.22% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.43% :     0.000001s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.78% :     0.000007s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      6: predicate.less_batch_normalization
 2.29% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.68% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.83% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.76% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 1.94% :     0.000002s :     15: predicate.partial_eliminate
 1.23% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000002s :     10: predicate.reshape_eliminate
 1.36% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.44% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.17% :     0.000003s :     16: predicate.switch_defer_inline
 1.96% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.06% :     0.000009s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.24% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.39% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 2.17% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.47% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.59% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001503    12
21.97% :     0.000330s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 2.18% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
75.84% :     0.001140s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024163    76
 0.17% :     0.000041s :      1: add_recomputation
 0.26% :     0.000063s :      1: auto_monad
 0.06% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.54% :     0.000371s :      1: bootstrap
 0.10% :     0.000025s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
27.96% :     0.006757s :      1: jit_opt_a
 0.62% :     0.000151s :      1: jit_opt_after_cconv
 0.23% :     0.000056s :      1: jit_opt_b
 1.80% :     0.000435s :      1: loop_unroll
 2.69% :     0.000649s :      1: mutable_eliminate
 3.69% :     0.000892s :     26: opt.transform.jit_opt_a
 0.20% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.96% :     0.000474s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000047s :      1: py_interpret_to_execute
 0.08% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000011s :      1: remove_dup_value
 1.80% :     0.000435s :      1: renormalize.infer
 5.20% :     0.001258s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000037s :      1: rewriter_after_opt_a
 0.32% :     0.000077s :      1: rewriter_before_opt_a
 0.34% :     0.000082s :      1: symbol_engine_optimizer
27.48% :     0.006640s :      1: task_emit
21.92% :     0.005297s :      1: type_inference
 0.23% :     0.000054s :      1: validate


TotalTime = 0.016865, [30]
[bootstrap]: 0.00036504
[type_inference]: 0.00548327
[event_method]: 2.135e-05
[auto_monad]: 7.531e-05
[graph_reusing]: 6.61e-06
[pre_auto_parallel]: 2.98e-06
[py_interpret_to_execute]: 4.611e-05
[rewriter_before_opt_a]: 7.791e-05
[expand_dump_flag]: 3.45e-06
[jit_opt_a]: 0.00828294, [2]
    [Cycle 1]: 0.00194871, [27]
        [switch_simplify]: 6.816e-05
        [loop_unroll]: 3.165e-05
        [a_1]: 0.000593
        [with_stream_mark]: 2.045e-05
        [recompute_prepare]: 8.19002e-06
        [updatestate_depend_eliminate]: 3.88001e-06
        [updatestate_assign_eliminate]: 3.14999e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 2.28002e-06
        [specialize_transform]: 5.69999e-06
        [updatestate_useless_node_eliminater]: 5.44e-06
        [accelerated_algorithm]: 5.93002e-06
        [meta_shard_fg_expand]: 2.21e-06
        [get_grad_eliminate_]: 5.74e-06
        [merge_forward]: 3.81999e-06
        [cell_reuse_recompute_pass]: 9.89996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.523e-05
        [j_node_and_user_rematch]: 1.016e-05
        [meta_fg_expand]: 2.61e-06
        [replace_old_param]: 9.74e-06
        [inline_without_move]: 5.71e-06
        [renormalize]: 0.00088427
        [add_forward_monad_depend]: 6.21e-06
        [auto_monad_grad]: 2.98e-06
        [auto_monad_eliminator]: 1.512e-05
        [cse]: 3.305e-05
        [replace_applicator]: 1.462e-05
    [Cycle 2]: 0.00033554, [27]
        [switch_simplify]: 5.79999e-06
        [loop_unroll]: 5.12e-06
        [a_1]: 0.00010043
        [with_stream_mark]: 1.031e-05
        [recompute_prepare]: 5.17999e-06
        [updatestate_depend_eliminate]: 2.61e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 5.29e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 5.53002e-06
        [meta_shard_fg_expand]: 1.56002e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.74999e-06
        [cell_reuse_recompute_pass]: 2.56e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.314e-05
        [j_node_and_user_rematch]: 8e-06
        [meta_fg_expand]: 1.55999e-06
        [replace_old_param]: 7.53e-06
        [inline_without_move]: 5.06997e-06
        [renormalize]: 1.19995e-07
        [add_forward_monad_depend]: 1.54e-06
        [auto_monad_grad]: 7.29982e-07
        [auto_monad_eliminator]: 5.16998e-06
        [cse]: 1.088e-05
        [replace_applicator]: 5.32999e-06
[py_interpret_to_execute_after_opt_a]: 1.353e-05
[rewriter_after_opt_a]: 3.893e-05
[convert_after_rewriter]: 7.38e-06
[order_py_execute_after_rewriter]: 5.54e-06
[mutable_eliminate]: 0.00077998
[jit_opt_b]: 6.122e-05, [1]
    [Cycle 1]: 5.241e-05, [2]
        [frontend_op_eliminate]: 2.302e-05
        [inline_after_opt_a]: 1.684e-05
[cconv]: 2.806e-05
[loop_unroll]: 0.00046515
[jit_opt_after_cconv]: 0.00015072, [1]
    [Cycle 1]: 0.00014365, [11]
        [c_1]: 2.215e-05
        [parameter_eliminate]: 2.94001e-06
        [updatestate_depend_eliminate]: 5.73002e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 2.294e-05
        [call_graph_tuple_transform]: 2.009e-05
        [tuple_list_get_item_eliminator]: 5.99999e-06
        [none_parameter_eliminate]: 1.62001e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 5.27999e-06
[remove_dup_value]: 1.601e-05
[partial_unused_args_eliminate]: 2.56e-06
[environ_conv]: 5.84e-06
[add_recomputation]: 5.437e-05
[cse_after_recomputation]: 2.402e-05, [1]
    [Cycle 1]: 1.714e-05, [1]
        [cse]: 1.07e-05
[auto_monad_reorder]: 1.604e-05
[get_jit_bprop_graph]: 2.61e-06
[rewriter_after_jit_bprop_graph]: 6.64999e-06
[opt_after_jit_grad]: 0.0004917
[symbol_engine_optimizer]: 8.255e-05, [1]
    [Cycle 1]: 7.585e-05, [6]
        [build]: 3.95998e-06
        [elim_shapecalc]: 8.30999e-06
        [elim_not_effective]: 1.473e-05
        [opt_reshape]: 6.03002e-06
        [fold_const_symbol]: 9.33002e-06
        [renormalize]: 7.2e-07
[validate]: 4.028e-05
Sums
    bootstrap                            :     0.000365s :  3.57%
    type_inference                       :     0.005483s : 53.62%
    event_method                         :     0.000021s :  0.21%
    auto_monad                           :     0.000075s :  0.74%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000046s :  0.45%
    rewriter_before_opt_a                :     0.000078s :  0.76%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000074s :  0.72%
    jit_opt_a.loop_unroll                :     0.000037s :  0.36%
    jit_opt_a.a_1                        :     0.000693s :  6.78%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.17%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000884s :  8.65%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.20%
    jit_opt_a.cse                        :     0.000044s :  0.43%
    jit_opt_a.replace_applicator         :     0.000020s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.13%
    rewriter_after_opt_a                 :     0.000039s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000780s :  7.63%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.16%
    cconv                                :     0.000028s :  0.27%
    loop_unroll                          :     0.000465s :  4.55%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000016s :  0.16%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000054s :  0.53%
    cse_after_recomputation.cse          :     0.000011s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.16%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000492s :  4.81%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.39%


Time group info:
------[substitution.]   0.000239    23
 1.15% :     0.000003s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.40% :     0.000006s :      3: substitution.graph_param_transform
84.80% :     0.000202s :      4: substitution.inline
 1.42% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.95% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.99% :     0.000005s :      2: substitution.replace_old_param
 5.71% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005396     2
80.06% :     0.004320s :      1: type_inference.infer
19.94% :     0.001076s :      1: type_inference.specialize
------[replace.]   0.000080     6
68.69% :     0.000055s :      4: replace.inline
31.31% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000212     6
94.10% :     0.000200s :      4: match.inline
 5.90% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.43% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.12% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 2.25% :     0.000003s :     10: predicate.arithmetic_simplify
 1.08% :     0.000001s :     10: predicate.cast_eliminate
 0.64% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 2.05% :     0.000003s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.82% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.58% :     0.000001s :      3: predicate.elim_not_effective
 0.70% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.43% :     0.000002s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.64% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.37% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.49% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.86% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 1.54% :     0.000002s :      3: predicate.mutable_eliminate
 0.62% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.45% :     0.000001s :      6: predicate.replace_old_param
 0.51% :     0.000001s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.11% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 2.01% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.21% :     0.000009s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.20% :     0.000002s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.59% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.92% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000779    12
51.96% :     0.000405s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.24% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.80% :     0.000341s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018719    72
 0.31% :     0.000057s :      1: add_recomputation
 0.42% :     0.000079s :      1: auto_monad
 0.10% :     0.000019s :      1: auto_monad_reorder
 2.06% :     0.000386s :      1: bootstrap
 0.17% :     0.000031s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000026s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.14% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
44.27% :     0.008287s :      1: jit_opt_a
 0.82% :     0.000154s :      1: jit_opt_after_cconv
 0.34% :     0.000064s :      1: jit_opt_b
 2.53% :     0.000473s :      1: loop_unroll
 4.22% :     0.000790s :      1: mutable_eliminate
 4.93% :     0.000923s :     26: opt.transform.jit_opt_a
 0.27% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000032s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.67% :     0.000500s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000006s :      1: pre_auto_parallel
 0.26% :     0.000049s :      1: py_interpret_to_execute
 0.09% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000019s :      1: remove_dup_value
 2.47% :     0.000463s :      1: renormalize.infer
 2.20% :     0.000412s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000042s :      1: rewriter_after_opt_a
 0.43% :     0.000081s :      1: rewriter_before_opt_a
 0.45% :     0.000085s :      1: symbol_engine_optimizer
29.40% :     0.005503s :      1: type_inference


TotalTime = 0.0597148, [33]
[bootstrap]: 0.0003629
[type_inference]: 0.00521307
[event_method]: 1.656e-05
[auto_monad]: 6.766e-05
[graph_reusing]: 5.15999e-06
[pre_auto_parallel]: 2.25002e-06
[py_interpret_to_execute]: 4.221e-05
[rewriter_before_opt_a]: 7.153e-05
[expand_dump_flag]: 3.28e-06
[jit_opt_a]: 0.0419223, [2]
    [Cycle 1]: 0.00197184, [27]
        [switch_simplify]: 6.606e-05
        [loop_unroll]: 2.928e-05
        [a_1]: 0.00059111
        [with_stream_mark]: 1.699e-05
        [recompute_prepare]: 7.81001e-06
        [updatestate_depend_eliminate]: 3.75e-06
        [updatestate_assign_eliminate]: 3.18e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 6.11e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 5.11002e-06
        [meta_shard_fg_expand]: 2.09e-06
        [get_grad_eliminate_]: 5.56e-06
        [merge_forward]: 3.86999e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.433e-05
        [j_node_and_user_rematch]: 8.99003e-06
        [meta_fg_expand]: 2.32001e-06
        [replace_old_param]: 9.37999e-06
        [inline_without_move]: 5.68002e-06
        [renormalize]: 0.00092354
        [add_forward_monad_depend]: 7.05e-06
        [auto_monad_grad]: 2.37001e-06
        [auto_monad_eliminator]: 1.618e-05
        [cse]: 3.139e-05
        [replace_applicator]: 1.551e-05
    [Cycle 2]: 0.00033984, [27]
        [switch_simplify]: 6.14001e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 9.444e-05
        [with_stream_mark]: 1.242e-05
        [recompute_prepare]: 5.97001e-06
        [updatestate_depend_eliminate]: 3.16001e-06
        [updatestate_assign_eliminate]: 2.19999e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 1.58002e-06
        [specialize_transform]: 5.18002e-06
        [updatestate_useless_node_eliminater]: 4.60999e-06
        [accelerated_algorithm]: 5.19e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 5.27001e-06
        [merge_forward]: 2.79999e-06
        [cell_reuse_recompute_pass]: 2.53e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.532e-05
        [j_node_and_user_rematch]: 7.61999e-06
        [meta_fg_expand]: 1.75001e-06
        [replace_old_param]: 7.81001e-06
        [inline_without_move]: 1.173e-05
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.17999e-06
        [auto_monad_grad]: 8.59989e-07
        [auto_monad_eliminator]: 5.47001e-06
        [cse]: 1.116e-05
        [replace_applicator]: 4.89003e-06
[py_interpret_to_execute_after_opt_a]: 1.172e-05
[rewriter_after_opt_a]: 3.85e-05
[convert_after_rewriter]: 6.72002e-06
[order_py_execute_after_rewriter]: 5.22e-06
[mutable_eliminate]: 0.00079069
[jit_opt_b]: 5.766e-05, [1]
    [Cycle 1]: 4.779e-05, [2]
        [frontend_op_eliminate]: 1.877e-05
        [inline_after_opt_a]: 1.644e-05
[cconv]: 3.467e-05
[loop_unroll]: 0.00050113
[jit_opt_after_cconv]: 0.00016223, [1]
    [Cycle 1]: 0.00015437, [11]
        [c_1]: 2.241e-05
        [parameter_eliminate]: 3.29001e-06
        [updatestate_depend_eliminate]: 8.11002e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [cse]: 2.87e-05
        [call_graph_tuple_transform]: 2.151e-05
        [tuple_list_get_item_eliminator]: 5.62001e-06
        [none_parameter_eliminate]: 1.99999e-06
        [renormalize]: 7.00005e-07
        [switch_simplify]: 5.30999e-06
[remove_dup_value]: 1.531e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 5.44998e-06
[add_recomputation]: 5.514e-05
[cse_after_recomputation]: 2.186e-05, [1]
    [Cycle 1]: 1.554e-05, [1]
        [cse]: 9.25999e-06
[auto_monad_reorder]: 1.501e-05
[get_jit_bprop_graph]: 1.99999e-06
[rewriter_after_jit_bprop_graph]: 8.95001e-06
[opt_after_jit_grad]: 0.00050757
[symbol_engine_optimizer]: 8.039e-05, [1]
    [Cycle 1]: 7.169e-05, [6]
        [build]: 4.83001e-06
        [elim_shapecalc]: 8.13001e-06
        [elim_not_effective]: 1.457e-05
        [opt_reshape]: 6.26e-06
        [fold_const_symbol]: 9.14e-06
        [renormalize]: 8.39995e-07
[validate]: 4.41e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.00942584
[execute]: 7.38e-06
Sums
    bootstrap                            :     0.000363s :  1.86%
    type_inference                       :     0.005213s : 26.77%
    event_method                         :     0.000017s :  0.09%
    auto_monad                           :     0.000068s :  0.35%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.22%
    rewriter_before_opt_a                :     0.000072s :  0.37%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000072s :  0.37%
    jit_opt_a.loop_unroll                :     0.000034s :  0.18%
    jit_opt_a.a_1                        :     0.000686s :  3.52%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.09%
    jit_opt_a.inline_without_move        :     0.000017s :  0.09%
    jit_opt_a.renormalize                :     0.000924s :  4.74%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.11%
    jit_opt_a.cse                        :     0.000043s :  0.22%
    jit_opt_a.replace_applicator         :     0.000020s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.06%
    rewriter_after_opt_a                 :     0.000039s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000791s :  4.06%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.000035s :  0.18%
    loop_unroll                          :     0.000501s :  2.57%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000055s :  0.28%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000508s :  2.61%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.23%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009426s : 48.41%
    execute                              :     0.000007s :  0.04%


Time group info:
------[substitution.]   0.000248    23
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000002s :      2: substitution.fold_const_symbol
 2.38% :     0.000006s :      3: substitution.graph_param_transform
85.99% :     0.000213s :      4: substitution.inline
 1.07% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.99% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.81% :     0.000004s :      2: substitution.replace_old_param
 5.25% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005137     2
80.87% :     0.004155s :      1: type_inference.infer
19.13% :     0.000983s :      1: type_inference.specialize
------[replace.]   0.000077     6
67.67% :     0.000052s :      4: replace.inline
32.33% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000222     6
94.62% :     0.000210s :      4: match.inline
 5.38% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.39% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.46% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.19% :     0.000001s :     10: predicate.addn_zero_filter
 2.66% :     0.000003s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.55% :     0.000002s :     10: predicate.depend_value_elim
 0.98% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.51% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.66% :     0.000001s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.47% :     0.000007s :     22: predicate.inline
 1.12% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.07% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 2.00% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.74% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.12% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 2.25% :     0.000003s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.02% :     0.000003s :     15: predicate.partial_eliminate
 1.09% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.64% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.51% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.20% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.44% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.15% :     0.000003s :     16: predicate.switch_defer_inline
 2.24% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.43% :     0.000009s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.25% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.14% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000739    12
46.71% :     0.000345s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.98% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.31% :     0.000357s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.061672    76
 0.10% :     0.000059s :      1: add_recomputation
 0.12% :     0.000071s :      1: auto_monad
 0.03% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.61% :     0.000379s :      1: bootstrap
 0.06% :     0.000037s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.04% :     0.000024s :      1: cse_after_recomputation
 0.01% :     0.000009s :      1: environ_conv
 0.03% :     0.000020s :      1: event_method
 0.02% :     0.000011s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000007s :      1: graph_reusing
67.98% :     0.041927s :      1: jit_opt_a
 0.27% :     0.000165s :      1: jit_opt_after_cconv
 0.10% :     0.000060s :      1: jit_opt_b
 0.83% :     0.000510s :      1: loop_unroll
 1.30% :     0.000801s :      1: mutable_eliminate
 1.49% :     0.000919s :     26: opt.transform.jit_opt_a
 0.08% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000028s :      4: opt.transform.jit_opt_b
 0.02% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000035s :      4: opt.transform.symbol_engine_opt
 0.84% :     0.000516s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000045s :      1: py_interpret_to_execute
 0.02% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000018s :      1: remove_dup_value
 0.81% :     0.000497s :      1: renormalize.infer
 0.68% :     0.000418s :      1: renormalize.specialize
 0.02% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000042s :      1: rewriter_after_opt_a
 0.12% :     0.000074s :      1: rewriter_before_opt_a
 0.13% :     0.000083s :      1: symbol_engine_optimizer
15.31% :     0.009440s :      1: task_emit
 8.47% :     0.005226s :      1: type_inference
 0.11% :     0.000066s :      1: validate


TotalTime = 0.039827, [33]
[bootstrap]: 0.00035766
[type_inference]: 0.0143007
[event_method]: 2.142e-05
[auto_monad]: 6.708e-05
[graph_reusing]: 6.15002e-06
[pre_auto_parallel]: 2.43e-06
[py_interpret_to_execute]: 4.43e-05
[rewriter_before_opt_a]: 7.27e-05
[expand_dump_flag]: 3.23e-06
[jit_opt_a]: 0.0101893, [2]
    [Cycle 1]: 0.00219633, [27]
        [switch_simplify]: 6.543e-05
        [loop_unroll]: 3.156e-05
        [a_1]: 0.00064221
        [with_stream_mark]: 2.517e-05
        [recompute_prepare]: 8.77e-06
        [updatestate_depend_eliminate]: 4.3e-06
        [updatestate_assign_eliminate]: 3.53e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [parameter_eliminate]: 2.62001e-06
        [specialize_transform]: 6.09001e-06
        [updatestate_useless_node_eliminater]: 5.59e-06
        [accelerated_algorithm]: 6.65002e-06
        [meta_shard_fg_expand]: 2.33002e-06
        [get_grad_eliminate_]: 5.53002e-06
        [merge_forward]: 3.42997e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.706e-05
        [j_node_and_user_rematch]: 1.417e-05
        [meta_fg_expand]: 2.51e-06
        [replace_old_param]: 1.051e-05
        [inline_without_move]: 6.24001e-06
        [renormalize]: 0.00096342
        [add_forward_monad_depend]: 6.19001e-06
        [auto_monad_grad]: 2.13002e-06
        [auto_monad_eliminator]: 1.537e-05
        [cse]: 3.014e-05
        [replace_applicator]: 1.678e-05
    [Cycle 2]: 0.00035839, [27]
        [switch_simplify]: 6.24999e-06
        [loop_unroll]: 5.19998e-06
        [a_1]: 0.00010168
        [with_stream_mark]: 1.335e-05
        [recompute_prepare]: 6.12999e-06
        [updatestate_depend_eliminate]: 3.15002e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.25002e-06
        [parameter_eliminate]: 1.41002e-06
        [specialize_transform]: 5.34e-06
        [updatestate_useless_node_eliminater]: 5.04998e-06
        [accelerated_algorithm]: 6.17001e-06
        [meta_shard_fg_expand]: 1.81e-06
        [get_grad_eliminate_]: 5.76e-06
        [merge_forward]: 3.31999e-06
        [cell_reuse_recompute_pass]: 2.50002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.605e-05
        [j_node_and_user_rematch]: 8.42e-06
        [meta_fg_expand]: 1.82001e-06
        [replace_old_param]: 8.48999e-06
        [inline_without_move]: 4.74e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.49e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 4.85999e-06
        [cse]: 1.275e-05
        [replace_applicator]: 5.58002e-06
[py_interpret_to_execute_after_opt_a]: 1.06e-05
[rewriter_after_opt_a]: 3.486e-05
[convert_after_rewriter]: 6.45997e-06
[order_py_execute_after_rewriter]: 5.25999e-06
[mutable_eliminate]: 0.00076542
[jit_opt_b]: 5.598e-05, [1]
    [Cycle 1]: 4.75e-05, [2]
        [frontend_op_eliminate]: 1.786e-05
        [inline_after_opt_a]: 1.697e-05
[cconv]: 3.282e-05
[loop_unroll]: 0.00048644
[jit_opt_after_cconv]: 0.00015783, [1]
    [Cycle 1]: 0.00015072, [11]
        [c_1]: 2.295e-05
        [parameter_eliminate]: 4.2e-06
        [updatestate_depend_eliminate]: 7.08e-06
        [updatestate_assign_eliminate]: 2.80002e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [cse]: 2.387e-05
        [call_graph_tuple_transform]: 2.183e-05
        [tuple_list_get_item_eliminator]: 6.34999e-06
        [none_parameter_eliminate]: 1.55001e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 5.50001e-06
[remove_dup_value]: 1.309e-05
[partial_unused_args_eliminate]: 2.31998e-06
[environ_conv]: 5.22999e-06
[add_recomputation]: 5.362e-05
[cse_after_recomputation]: 2.149e-05, [1]
    [Cycle 1]: 1.517e-05, [1]
        [cse]: 9.07001e-06
[auto_monad_reorder]: 1.51e-05
[get_jit_bprop_graph]: 1.82999e-06
[rewriter_after_jit_bprop_graph]: 8.40001e-06
[opt_after_jit_grad]: 0.00049702
[symbol_engine_optimizer]: 7.927e-05, [1]
    [Cycle 1]: 7.086e-05, [6]
        [build]: 3.33e-06
        [elim_shapecalc]: 7.66001e-06
        [elim_not_effective]: 1.416e-05
        [opt_reshape]: 6.14999e-06
        [fold_const_symbol]: 9.27001e-06
        [renormalize]: 1.10999e-06
[validate]: 3.987e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.0121981
[execute]: 9.49999e-06
Sums
    bootstrap                            :     0.000358s :  1.14%
    type_inference                       :     0.014301s : 45.56%
    event_method                         :     0.000021s :  0.07%
    auto_monad                           :     0.000067s :  0.21%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.14%
    rewriter_before_opt_a                :     0.000073s :  0.23%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000072s :  0.23%
    jit_opt_a.loop_unroll                :     0.000037s :  0.12%
    jit_opt_a.a_1                        :     0.000744s :  2.37%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000023s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000019s :  0.06%
    jit_opt_a.inline_without_move        :     0.000011s :  0.03%
    jit_opt_a.renormalize                :     0.000964s :  3.07%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.06%
    jit_opt_a.cse                        :     0.000043s :  0.14%
    jit_opt_a.replace_applicator         :     0.000022s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.03%
    rewriter_after_opt_a                 :     0.000035s :  0.11%
    convert_after_rewriter               :     0.000006s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000765s :  2.44%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.05%
    cconv                                :     0.000033s :  0.10%
    loop_unroll                          :     0.000486s :  1.55%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000013s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000054s :  0.17%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000015s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.03%
    opt_after_jit_grad                   :     0.000497s :  1.58%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.012198s : 38.86%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000271    23
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.78% :     0.000002s :      2: substitution.fold_const_symbol
 2.30% :     0.000006s :      3: substitution.graph_param_transform
85.20% :     0.000231s :      4: substitution.inline
 1.85% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.99% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.15% :     0.000006s :      2: substitution.replace_old_param
 5.03% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.014212     2
80.54% :     0.011447s :      1: type_inference.infer
19.46% :     0.002765s :      1: type_inference.specialize
------[replace.]   0.000086     6
68.78% :     0.000059s :      4: replace.inline
31.22% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000240     6
94.86% :     0.000228s :      4: match.inline
 5.14% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   691
 1.27% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.21% :     0.000002s :     10: predicate.addn_zero_filter
 3.28% :     0.000004s :     10: predicate.arithmetic_simplify
 1.83% :     0.000002s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.91% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.45% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.54% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.35% :     0.000002s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 4.91% :     0.000007s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.55% :     0.000002s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.60% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.19% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.14% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 0.91% :     0.000001s :     10: predicate.minmaximum_grad
 1.90% :     0.000003s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     15: predicate.partial_eliminate
 1.06% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.86% :     0.000001s :      6: predicate.replace_old_param
 0.40% :     0.000001s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.05% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.02% :     0.000001s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.57% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.24% :     0.000003s :     16: predicate.switch_defer_inline
 2.24% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.00% :     0.000009s :     53: predicate.switch_simplify
 1.06% :     0.000001s :     10: predicate.tile_eliminate
 1.74% :     0.000002s :     10: predicate.transpose_eliminate
 1.58% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.09% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.28% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 2.19% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.37% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.41% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.47% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.10% :     0.000003s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001480    12
65.28% :     0.000966s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 3.50% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
31.22% :     0.000462s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.041884    76
 0.14% :     0.000057s :      1: add_recomputation
 0.17% :     0.000071s :      1: auto_monad
 0.04% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.91% :     0.000381s :      1: bootstrap
 0.08% :     0.000035s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.06% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.06% :     0.000025s :      1: event_method
 0.03% :     0.000014s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
24.34% :     0.010193s :      1: jit_opt_a
 0.38% :     0.000161s :      1: jit_opt_after_cconv
 0.14% :     0.000059s :      1: jit_opt_b
 1.18% :     0.000495s :      1: loop_unroll
 1.86% :     0.000778s :      1: mutable_eliminate
 2.36% :     0.000988s :     26: opt.transform.jit_opt_a
 0.13% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000028s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.21% :     0.000507s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.11% :     0.000047s :      1: py_interpret_to_execute
 0.03% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000015s :      1: remove_dup_value
 1.26% :     0.000527s :      1: renormalize.infer
 1.02% :     0.000427s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000038s :      1: rewriter_after_opt_a
 0.18% :     0.000076s :      1: rewriter_before_opt_a
 0.20% :     0.000082s :      1: symbol_engine_optimizer
29.17% :     0.012217s :      1: task_emit
34.18% :     0.014315s :      1: type_inference
 0.20% :     0.000085s :      1: validate


TotalTime = 0.0169872, [30]
[bootstrap]: 0.00037033
[type_inference]: 0.00592905
[event_method]: 2.022e-05
[auto_monad]: 6.715e-05
[graph_reusing]: 6.11998e-06
[pre_auto_parallel]: 2.72001e-06
[py_interpret_to_execute]: 4.105e-05
[rewriter_before_opt_a]: 7.163e-05
[expand_dump_flag]: 3.26999e-06
[jit_opt_a]: 0.00758323, [2]
    [Cycle 1]: 0.00296055, [27]
        [switch_simplify]: 7.234e-05
        [loop_unroll]: 3.22e-05
        [a_1]: 0.00063741
        [with_stream_mark]: 2.348e-05
        [recompute_prepare]: 1.046e-05
        [updatestate_depend_eliminate]: 4.60999e-06
        [updatestate_assign_eliminate]: 3.18998e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 2.17999e-06
        [specialize_transform]: 6.24001e-06
        [updatestate_useless_node_eliminater]: 5.64998e-06
        [accelerated_algorithm]: 6.26998e-06
        [meta_shard_fg_expand]: 2.58e-06
        [get_grad_eliminate_]: 5.94999e-06
        [merge_forward]: 3.97002e-06
        [cell_reuse_recompute_pass]: 1.73002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.702e-05
        [j_node_and_user_rematch]: 1.029e-05
        [meta_fg_expand]: 2.83e-06
        [replace_old_param]: 1.095e-05
        [inline_without_move]: 5.68997e-06
        [renormalize]: 0.00181023
        [add_forward_monad_depend]: 8.22e-06
        [auto_monad_grad]: 2.54999e-06
        [auto_monad_eliminator]: 1.821e-05
        [cse]: 3.378e-05
        [replace_applicator]: 1.861e-05
    [Cycle 2]: 0.00037099, [27]
        [switch_simplify]: 5.81003e-06
        [loop_unroll]: 5.69e-06
        [a_1]: 0.000105
        [with_stream_mark]: 1.461e-05
        [recompute_prepare]: 5.92999e-06
        [updatestate_depend_eliminate]: 2.74001e-06
        [updatestate_assign_eliminate]: 2.97002e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.21002e-06
        [specialize_transform]: 5.47001e-06
        [updatestate_useless_node_eliminater]: 5.47001e-06
        [accelerated_algorithm]: 5.89999e-06
        [meta_shard_fg_expand]: 1.69998e-06
        [get_grad_eliminate_]: 4.63001e-06
        [merge_forward]: 3.4e-06
        [cell_reuse_recompute_pass]: 3.13e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.63e-05
        [j_node_and_user_rematch]: 8.37e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 9.51e-06
        [inline_without_move]: 5.18002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 1.28002e-06
        [auto_monad_eliminator]: 5.47001e-06
        [cse]: 1.158e-05
        [replace_applicator]: 5.89999e-06
[py_interpret_to_execute_after_opt_a]: 1.642e-05
[rewriter_after_opt_a]: 4.159e-05
[convert_after_rewriter]: 7.83001e-06
[order_py_execute_after_rewriter]: 8e-06
[mutable_eliminate]: 0.00094428
[jit_opt_b]: 6.172e-05, [1]
    [Cycle 1]: 5.254e-05, [2]
        [frontend_op_eliminate]: 2.09e-05
        [inline_after_opt_a]: 1.838e-05
[cconv]: 3.195e-05
[loop_unroll]: 0.00053089
[jit_opt_after_cconv]: 0.00016815, [1]
    [Cycle 1]: 0.00016027, [11]
        [c_1]: 2.343e-05
        [parameter_eliminate]: 3.64002e-06
        [updatestate_depend_eliminate]: 7.75998e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [cse]: 2.632e-05
        [call_graph_tuple_transform]: 2.28e-05
        [tuple_list_get_item_eliminator]: 6.58998e-06
        [none_parameter_eliminate]: 1.77001e-06
        [renormalize]: 9.00007e-07
        [switch_simplify]: 6.61e-06
[remove_dup_value]: 1.632e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 5.78002e-06
[add_recomputation]: 5.516e-05
[cse_after_recomputation]: 2.486e-05, [1]
    [Cycle 1]: 1.763e-05, [1]
        [cse]: 1e-05
[auto_monad_reorder]: 1.642e-05
[get_jit_bprop_graph]: 2.60002e-06
[rewriter_after_jit_bprop_graph]: 7.5e-06
[opt_after_jit_grad]: 0.000617
[symbol_engine_optimizer]: 8.464e-05, [1]
    [Cycle 1]: 7.656e-05, [6]
        [build]: 5.40999e-06
        [elim_shapecalc]: 9.24e-06
        [elim_not_effective]: 1.521e-05
        [opt_reshape]: 6.68e-06
        [fold_const_symbol]: 9.92001e-06
        [renormalize]: 5.69999e-07
[validate]: 4.386e-05
Sums
    bootstrap                            :     0.000370s :  3.07%
    type_inference                       :     0.005929s : 49.15%
    event_method                         :     0.000020s :  0.17%
    auto_monad                           :     0.000067s :  0.56%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000041s :  0.34%
    rewriter_before_opt_a                :     0.000072s :  0.59%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000078s :  0.65%
    jit_opt_a.loop_unroll                :     0.000038s :  0.31%
    jit_opt_a.a_1                        :     0.000742s :  6.15%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000020s :  0.17%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.001810s : 15.01%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.20%
    jit_opt_a.cse                        :     0.000045s :  0.38%
    jit_opt_a.replace_applicator         :     0.000025s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.14%
    rewriter_after_opt_a                 :     0.000042s :  0.34%
    convert_after_rewriter               :     0.000008s :  0.06%
    order_py_execute_after_rewriter      :     0.000008s :  0.07%
    mutable_eliminate                    :     0.000944s :  7.83%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.15%
    cconv                                :     0.000032s :  0.26%
    loop_unroll                          :     0.000531s :  4.40%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000026s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.05%
    remove_dup_value                     :     0.000016s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000055s :  0.46%
    cse_after_recomputation.cse          :     0.000010s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.14%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000617s :  5.11%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.36%


Time group info:
------[substitution.]   0.000262    23
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.64% :     0.000007s :      3: substitution.graph_param_transform
84.09% :     0.000220s :      4: substitution.inline
 1.48% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.22% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.57% :     0.000007s :      2: substitution.replace_old_param
 5.58% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005855     2
82.91% :     0.004855s :      1: type_inference.infer
17.09% :     0.001001s :      1: type_inference.specialize
------[replace.]   0.000090     6
65.99% :     0.000060s :      4: replace.inline
34.01% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000231     6
94.15% :     0.000217s :      4: match.inline
 5.85% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   691
 1.21% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.82% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.92% :     0.000001s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 2.59% :     0.000004s :     10: predicate.arithmetic_simplify
 1.25% :     0.000002s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.88% :     0.000001s :     10: predicate.compare_switch_simplify
 1.60% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.65% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.02% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.61% :     0.000008s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.48% :     0.000002s :     15: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.74% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.21% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.90% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 1.71% :     0.000002s :      3: predicate.mutable_eliminate
 0.58% :     0.000001s :      3: predicate.opt_reshape
 1.95% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.98% :     0.000003s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.49% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.73% :     0.000002s :     10: predicate.reshape_eliminate
 0.98% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.39% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.14% :     0.000002s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.77% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.31% :     0.000003s :     16: predicate.switch_defer_inline
 1.91% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.04% :     0.000010s :     53: predicate.switch_simplify
 1.23% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000002s :     10: predicate.transpose_eliminate
 1.62% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.59% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.48% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.90% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.37% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.43% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.90% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001485    12
21.67% :     0.000322s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 2.53% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
75.80% :     0.001126s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019851    72
 0.29% :     0.000058s :      1: add_recomputation
 0.36% :     0.000071s :      1: auto_monad
 0.10% :     0.000019s :      1: auto_monad_reorder
 1.98% :     0.000392s :      1: bootstrap
 0.17% :     0.000035s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000027s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.12% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
38.22% :     0.007588s :      1: jit_opt_a
 0.86% :     0.000171s :      1: jit_opt_after_cconv
 0.33% :     0.000065s :      1: jit_opt_b
 2.72% :     0.000540s :      1: loop_unroll
 4.82% :     0.000957s :      1: mutable_eliminate
 5.00% :     0.000992s :     26: opt.transform.jit_opt_a
 0.28% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000031s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.16% :     0.000628s :      1: opt_after_jit_grad
 0.05% :     0.000010s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000044s :      1: py_interpret_to_execute
 0.10% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000019s :      1: remove_dup_value
 2.67% :     0.000531s :      1: renormalize.infer
 6.39% :     0.001269s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000045s :      1: rewriter_after_opt_a
 0.38% :     0.000075s :      1: rewriter_before_opt_a
 0.44% :     0.000087s :      1: symbol_engine_optimizer
29.93% :     0.005942s :      1: type_inference


TotalTime = 0.0454501, [33]
[bootstrap]: 0.00041342
[type_inference]: 0.0149216
[event_method]: 1.901e-05
[auto_monad]: 6.679e-05
[graph_reusing]: 6.00002e-06
[pre_auto_parallel]: 2.47001e-06
[py_interpret_to_execute]: 3.686e-05
[rewriter_before_opt_a]: 7.488e-05
[expand_dump_flag]: 3.21999e-06
[jit_opt_a]: 0.0135798, [2]
    [Cycle 1]: 0.00185968, [27]
        [switch_simplify]: 7.586e-05
        [loop_unroll]: 3.045e-05
        [a_1]: 0.0005784
        [with_stream_mark]: 1.856e-05
        [recompute_prepare]: 6.88998e-06
        [updatestate_depend_eliminate]: 3.81999e-06
        [updatestate_assign_eliminate]: 2.88998e-06
        [updatestate_loads_eliminate]: 2.76e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 5.97001e-06
        [updatestate_useless_node_eliminater]: 5.21998e-06
        [accelerated_algorithm]: 6.64999e-06
        [meta_shard_fg_expand]: 2.17999e-06
        [get_grad_eliminate_]: 5.07999e-06
        [merge_forward]: 3.97e-06
        [cell_reuse_recompute_pass]: 1.71e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.399e-05
        [j_node_and_user_rematch]: 9.62001e-06
        [meta_fg_expand]: 3.14001e-06
        [replace_old_param]: 9.02e-06
        [inline_without_move]: 5.59e-06
        [renormalize]: 0.00081688
        [add_forward_monad_depend]: 5.83997e-06
        [auto_monad_grad]: 2.34999e-06
        [auto_monad_eliminator]: 1.502e-05
        [cse]: 2.895e-05
        [replace_applicator]: 1.216e-05
    [Cycle 2]: 0.00032784, [27]
        [switch_simplify]: 5.67001e-06
        [loop_unroll]: 5.35999e-06
        [a_1]: 9.37e-05
        [with_stream_mark]: 9.32999e-06
        [recompute_prepare]: 5.15001e-06
        [updatestate_depend_eliminate]: 2.49001e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 9.30013e-07
        [specialize_transform]: 4.90001e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.40999e-06
        [meta_shard_fg_expand]: 1.00001e-06
        [get_grad_eliminate_]: 4.59998e-06
        [merge_forward]: 2.83e-06
        [cell_reuse_recompute_pass]: 2.21e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.32e-05
        [j_node_and_user_rematch]: 8.12003e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 7.23999e-06
        [inline_without_move]: 7.98999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.36002e-06
        [auto_monad_grad]: 1.32999e-06
        [auto_monad_eliminator]: 5.64e-06
        [cse]: 9.86e-06
        [replace_applicator]: 4.84e-06
[py_interpret_to_execute_after_opt_a]: 1.22e-05
[rewriter_after_opt_a]: 3.881e-05
[convert_after_rewriter]: 6.89001e-06
[order_py_execute_after_rewriter]: 4.84e-06
[mutable_eliminate]: 0.00074907
[jit_opt_b]: 5.061e-05, [1]
    [Cycle 1]: 4.314e-05, [2]
        [frontend_op_eliminate]: 1.678e-05
        [inline_after_opt_a]: 1.488e-05
[cconv]: 2.732e-05
[loop_unroll]: 0.00045182
[jit_opt_after_cconv]: 0.00014351, [1]
    [Cycle 1]: 0.00013664, [11]
        [c_1]: 2.199e-05
        [parameter_eliminate]: 2.74001e-06
        [updatestate_depend_eliminate]: 5.73002e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [cse]: 1.905e-05
        [call_graph_tuple_transform]: 1.945e-05
        [tuple_list_get_item_eliminator]: 5.30999e-06
        [none_parameter_eliminate]: 1.50999e-06
        [renormalize]: 5.39992e-07
        [switch_simplify]: 5.47999e-06
[remove_dup_value]: 1.47e-05
[partial_unused_args_eliminate]: 2.25002e-06
[environ_conv]: 5.97001e-06
[add_recomputation]: 5.158e-05
[cse_after_recomputation]: 2.258e-05, [1]
    [Cycle 1]: 1.659e-05, [1]
        [cse]: 1.011e-05
[auto_monad_reorder]: 1.584e-05
[get_jit_bprop_graph]: 1.39e-06
[rewriter_after_jit_bprop_graph]: 5.85002e-06
[opt_after_jit_grad]: 0.0004781
[symbol_engine_optimizer]: 7.946e-05, [1]
    [Cycle 1]: 7.243e-05, [6]
        [build]: 5.17e-06
        [elim_shapecalc]: 7.68999e-06
        [elim_not_effective]: 1.535e-05
        [opt_reshape]: 5.87001e-06
        [fold_const_symbol]: 9.60001e-06
        [renormalize]: 8.00006e-07
[validate]: 4.075e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.0138737
[execute]: 1.032e-05
Sums
    bootstrap                            :     0.000413s :  1.24%
    type_inference                       :     0.014922s : 44.68%
    event_method                         :     0.000019s :  0.06%
    auto_monad                           :     0.000067s :  0.20%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000037s :  0.11%
    rewriter_before_opt_a                :     0.000075s :  0.22%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000082s :  0.24%
    jit_opt_a.loop_unroll                :     0.000036s :  0.11%
    jit_opt_a.a_1                        :     0.000672s :  2.01%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.05%
    jit_opt_a.inline_without_move        :     0.000014s :  0.04%
    jit_opt_a.renormalize                :     0.000817s :  2.45%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.06%
    jit_opt_a.cse                        :     0.000039s :  0.12%
    jit_opt_a.replace_applicator         :     0.000017s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000039s :  0.12%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000749s :  2.24%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.04%
    cconv                                :     0.000027s :  0.08%
    loop_unroll                          :     0.000452s :  1.35%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000015s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000052s :  0.15%
    cse_after_recomputation.cse          :     0.000010s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000478s :  1.43%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.013874s : 41.54%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.000223    23
 1.56% :     0.000003s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.72% :     0.000006s :      3: substitution.graph_param_transform
83.41% :     0.000186s :      4: substitution.inline
 1.58% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.07% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.17% :     0.000005s :      2: substitution.replace_old_param
 5.88% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.014840     2
93.30% :     0.013845s :      1: type_inference.infer
 6.70% :     0.000995s :      1: type_inference.specialize
------[replace.]   0.000075     6
66.98% :     0.000050s :      4: replace.inline
33.02% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000196     6
93.86% :     0.000184s :      4: match.inline
 6.14% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   691
 1.49% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.28% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.55% :     0.000002s :     10: predicate.addn_zero_filter
 2.59% :     0.000003s :     10: predicate.arithmetic_simplify
 1.51% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.60% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000001s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.03% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.83% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.27% :     0.000007s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.62% :     0.000002s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.07% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.46% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 1.95% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.01% :     0.000003s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.42% :     0.000002s :     18: predicate.replace_applicator
 0.45% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.10% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.72% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.13% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.19% :     0.000003s :     16: predicate.switch_defer_inline
 1.97% :     0.000003s :     16: predicate.switch_layer_defer_inline
10.50% :     0.000014s :     53: predicate.switch_simplify
 1.38% :     0.000002s :     10: predicate.tile_eliminate
 1.20% :     0.000002s :     10: predicate.transpose_eliminate
 1.94% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.33% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.95% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.34% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.009717    12
96.34% :     0.009361s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 0.32% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
 3.35% :     0.000325s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.047272    76
 0.12% :     0.000055s :      1: add_recomputation
 0.15% :     0.000070s :      1: auto_monad
 0.04% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.92% :     0.000437s :      1: bootstrap
 0.06% :     0.000030s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.05% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.05% :     0.000024s :      1: event_method
 0.03% :     0.000017s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
28.74% :     0.013584s :      1: jit_opt_a
 0.31% :     0.000146s :      1: jit_opt_after_cconv
 0.11% :     0.000053s :      1: jit_opt_b
 0.97% :     0.000459s :      1: loop_unroll
 1.61% :     0.000759s :      1: mutable_eliminate
 1.92% :     0.000907s :     26: opt.transform.jit_opt_a
 0.10% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000025s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.03% :     0.000486s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.08% :     0.000040s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000017s :      1: remove_dup_value
 0.93% :     0.000437s :      1: renormalize.infer
 0.78% :     0.000371s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000042s :      1: rewriter_after_opt_a
 0.16% :     0.000077s :      1: rewriter_before_opt_a
 0.17% :     0.000082s :      1: symbol_engine_optimizer
29.40% :     0.013896s :      1: task_emit
31.61% :     0.014943s :      1: type_inference
 0.13% :     0.000063s :      1: validate


TotalTime = 0.0248997, [33]
[bootstrap]: 0.00040598
[type_inference]: 0.00576732
[event_method]: 1.757e-05
[auto_monad]: 6.726e-05
[graph_reusing]: 5.34e-06
[pre_auto_parallel]: 2.32001e-06
[py_interpret_to_execute]: 4.415e-05
[rewriter_before_opt_a]: 7.475e-05
[expand_dump_flag]: 3.11001e-06
[jit_opt_a]: 0.00720583, [2]
    [Cycle 1]: 0.00218562, [27]
        [switch_simplify]: 7.207e-05
        [loop_unroll]: 3.005e-05
        [a_1]: 0.0006525
        [with_stream_mark]: 2.412e-05
        [recompute_prepare]: 8.87999e-06
        [updatestate_depend_eliminate]: 3.72002e-06
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 2.32999e-06
        [specialize_transform]: 6.33e-06
        [updatestate_useless_node_eliminater]: 4.96002e-06
        [accelerated_algorithm]: 5.82001e-06
        [meta_shard_fg_expand]: 2.70002e-06
        [get_grad_eliminate_]: 6.38e-06
        [merge_forward]: 3.86999e-06
        [cell_reuse_recompute_pass]: 2.29999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.825e-05
        [j_node_and_user_rematch]: 1.03e-05
        [meta_fg_expand]: 2.40002e-06
        [replace_old_param]: 1.165e-05
        [inline_without_move]: 5.64998e-06
        [renormalize]: 0.0010205
        [add_forward_monad_depend]: 8.63001e-06
        [auto_monad_grad]: 2.78e-06
        [auto_monad_eliminator]: 1.719e-05
        [cse]: 2.906e-05
        [replace_applicator]: 1.728e-05
    [Cycle 2]: 0.00045252, [27]
        [switch_simplify]: 6.86999e-06
        [loop_unroll]: 5.75001e-06
        [a_1]: 0.00010107
        [with_stream_mark]: 1.613e-05
        [recompute_prepare]: 7.615e-05
        [updatestate_depend_eliminate]: 3.61001e-06
        [updatestate_assign_eliminate]: 2.69999e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 1.74998e-06
        [specialize_transform]: 5.76e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 6.38e-06
        [meta_shard_fg_expand]: 2.24999e-06
        [get_grad_eliminate_]: 5.00001e-06
        [merge_forward]: 3.3e-06
        [cell_reuse_recompute_pass]: 4.52e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.941e-05
        [j_node_and_user_rematch]: 9.07001e-06
        [meta_fg_expand]: 1.52999e-06
        [replace_old_param]: 9.29e-06
        [inline_without_move]: 4.63001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.90001e-06
        [auto_monad_grad]: 1.46002e-06
        [auto_monad_eliminator]: 6.06998e-06
        [cse]: 1.447e-05
        [replace_applicator]: 5.67999e-06
[py_interpret_to_execute_after_opt_a]: 1.752e-05
[rewriter_after_opt_a]: 3.714e-05
[convert_after_rewriter]: 1.256e-05
[order_py_execute_after_rewriter]: 4.92999e-06
[mutable_eliminate]: 0.00157716
[jit_opt_b]: 7.099e-05, [1]
    [Cycle 1]: 6.035e-05, [2]
        [frontend_op_eliminate]: 2.379e-05
        [inline_after_opt_a]: 2.192e-05
[cconv]: 3.735e-05
[loop_unroll]: 0.00064931
[jit_opt_after_cconv]: 0.00017442, [1]
    [Cycle 1]: 0.00016732, [11]
        [c_1]: 2.383e-05
        [parameter_eliminate]: 5.87001e-06
        [updatestate_depend_eliminate]: 8.68001e-06
        [updatestate_assign_eliminate]: 2.92002e-06
        [updatestate_loads_eliminate]: 2.85002e-06
        [cse]: 2.801e-05
        [call_graph_tuple_transform]: 2.616e-05
        [tuple_list_get_item_eliminator]: 5.97001e-06
        [none_parameter_eliminate]: 2.12999e-06
        [renormalize]: 8.39995e-07
        [switch_simplify]: 6.11e-06
[remove_dup_value]: 1.63e-05
[partial_unused_args_eliminate]: 2.33002e-06
[environ_conv]: 6.01998e-06
[add_recomputation]: 5.466e-05
[cse_after_recomputation]: 2.202e-05, [1]
    [Cycle 1]: 1.566e-05, [1]
        [cse]: 9.21998e-06
[auto_monad_reorder]: 1.582e-05
[get_jit_bprop_graph]: 2.47001e-06
[rewriter_after_jit_bprop_graph]: 1.13e-05
[opt_after_jit_grad]: 0.00055213
[symbol_engine_optimizer]: 8.507e-05, [1]
    [Cycle 1]: 7.619e-05, [6]
        [build]: 5.42999e-06
        [elim_shapecalc]: 8.89e-06
        [elim_not_effective]: 1.635e-05
        [opt_reshape]: 6.65002e-06
        [fold_const_symbol]: 9.34e-06
        [renormalize]: 9.70002e-07
[validate]: 4.681e-05
[backend_pass]: 1.12e-06
[task_emit]: 0.00765325
[execute]: 8.57998e-06
Sums
    bootstrap                            :     0.000406s :  2.07%
    type_inference                       :     0.005767s : 29.41%
    event_method                         :     0.000018s :  0.09%
    auto_monad                           :     0.000067s :  0.34%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.23%
    rewriter_before_opt_a                :     0.000075s :  0.38%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000079s :  0.40%
    jit_opt_a.loop_unroll                :     0.000036s :  0.18%
    jit_opt_a.a_1                        :     0.000754s :  3.84%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000085s :  0.43%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000021s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.001021s :  5.20%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.12%
    jit_opt_a.cse                        :     0.000044s :  0.22%
    jit_opt_a.replace_applicator         :     0.000023s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.09%
    rewriter_after_opt_a                 :     0.000037s :  0.19%
    convert_after_rewriter               :     0.000013s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.001577s :  8.04%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000022s :  0.11%
    cconv                                :     0.000037s :  0.19%
    loop_unroll                          :     0.000649s :  3.31%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000055s :  0.28%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.06%
    opt_after_jit_grad                   :     0.000552s :  2.82%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000047s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007653s : 39.03%
    execute                              :     0.000009s :  0.04%


Time group info:
------[substitution.]   0.000291    23
 0.73% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000002s :      2: substitution.fold_const_symbol
 2.40% :     0.000007s :      3: substitution.graph_param_transform
85.60% :     0.000249s :      4: substitution.inline
 1.51% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.04% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.52% :     0.000007s :      2: substitution.replace_old_param
 4.62% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005687     2
81.28% :     0.004622s :      1: type_inference.infer
18.72% :     0.001065s :      1: type_inference.specialize
------[replace.]   0.000084     6
65.75% :     0.000055s :      4: replace.inline
34.25% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000258     6
95.30% :     0.000246s :      4: match.inline
 4.70% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000140   691
 1.26% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.91% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.85% :     0.000001s :     10: predicate.addn_check_dump
 1.16% :     0.000002s :     10: predicate.addn_zero_filter
 2.85% :     0.000004s :     10: predicate.arithmetic_simplify
 1.08% :     0.000002s :     10: predicate.cast_eliminate
 0.67% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.84% :     0.000001s :     10: predicate.compare_switch_simplify
 1.55% :     0.000002s :     10: predicate.depend_value_elim
 0.87% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.24% :     0.000000s :      3: predicate.elim_not_effective
 0.70% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.00% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.79% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.45% :     0.000008s :     22: predicate.inline
 0.77% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000002s :      6: predicate.less_batch_normalization
 1.81% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.52% :     0.000002s :     15: predicate.load_eliminater
 2.01% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.83% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.26% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.94% :     0.000001s :     10: predicate.merge_addn
 0.88% :     0.000001s :     10: predicate.minmaximum_grad
 3.98% :     0.000006s :      3: predicate.mutable_eliminate
 0.37% :     0.000001s :      3: predicate.opt_reshape
 1.82% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.32% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.40% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.59% :     0.000001s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.42% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.00% :     0.000001s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.13% :     0.000002s :      6: predicate.specialize_transform
 1.14% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.61% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.31% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 1.67% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.57% :     0.000011s :     53: predicate.switch_simplify
 1.15% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000002s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.95% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.43% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.34% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.58% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.99% :     0.000003s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.84% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000738    12
45.45% :     0.000336s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.20% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.35% :     0.000350s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027148    76
 0.21% :     0.000058s :      1: add_recomputation
 0.26% :     0.000070s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.56% :     0.000424s :      1: bootstrap
 0.15% :     0.000040s :      1: cconv
 0.06% :     0.000015s :      1: convert_after_rewriter
 0.09% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.08% :     0.000021s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000006s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
26.56% :     0.007211s :      1: jit_opt_a
 0.66% :     0.000178s :      1: jit_opt_after_cconv
 0.27% :     0.000074s :      1: jit_opt_b
 2.44% :     0.000661s :      1: loop_unroll
 5.89% :     0.001598s :      1: mutable_eliminate
 3.97% :     0.001077s :     26: opt.transform.jit_opt_a
 0.22% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000036s :      4: opt.transform.jit_opt_b
 0.06% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.13% :     0.000035s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000038s :      4: opt.transform.symbol_engine_opt
 2.07% :     0.000563s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000047s :      1: py_interpret_to_execute
 0.07% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 2.04% :     0.000553s :      1: renormalize.infer
 1.68% :     0.000457s :      1: renormalize.specialize
 0.05% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000040s :      1: rewriter_after_opt_a
 0.29% :     0.000078s :      1: rewriter_before_opt_a
 0.32% :     0.000088s :      1: symbol_engine_optimizer
28.26% :     0.007671s :      1: task_emit
21.31% :     0.005785s :      1: type_inference
 0.25% :     0.000069s :      1: validate


TotalTime = 0.0401032, [33]
[bootstrap]: 0.00037798
[type_inference]: 0.00908385
[event_method]: 1.978e-05
[auto_monad]: 7.168e-05
[graph_reusing]: 6.28e-06
[pre_auto_parallel]: 3.66001e-06
[py_interpret_to_execute]: 4.591e-05
[rewriter_before_opt_a]: 8.362e-05
[expand_dump_flag]: 3.61001e-06
[jit_opt_a]: 0.0195894, [2]
    [Cycle 1]: 0.0151481, [27]
        [switch_simplify]: 6.326e-05
        [loop_unroll]: 2.973e-05
        [a_1]: 0.00059823
        [with_stream_mark]: 2.344e-05
        [recompute_prepare]: 7.94002e-06
        [updatestate_depend_eliminate]: 4.50001e-06
        [updatestate_assign_eliminate]: 3.08998e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [parameter_eliminate]: 1.69998e-06
        [specialize_transform]: 6.34001e-06
        [updatestate_useless_node_eliminater]: 5.74999e-06
        [accelerated_algorithm]: 6.32001e-06
        [meta_shard_fg_expand]: 2.02999e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 3.76999e-06
        [cell_reuse_recompute_pass]: 1.70001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.491e-05
        [j_node_and_user_rematch]: 9.52001e-06
        [meta_fg_expand]: 2.42001e-06
        [replace_old_param]: 1.026e-05
        [inline_without_move]: 5.42999e-06
        [renormalize]: 0.0139961
        [add_forward_monad_depend]: 1.51e-05
        [auto_monad_grad]: 3.51001e-06
        [auto_monad_eliminator]: 2.975e-05
        [cse]: 3.513e-05
        [replace_applicator]: 3.033e-05
    [Cycle 2]: 0.00040814, [27]
        [switch_simplify]: 7.41001e-06
        [loop_unroll]: 6.48e-06
        [a_1]: 0.00011322
        [with_stream_mark]: 1.856e-05
        [recompute_prepare]: 5.69999e-06
        [updatestate_depend_eliminate]: 3.51001e-06
        [updatestate_assign_eliminate]: 3.03e-06
        [updatestate_loads_eliminate]: 2.87002e-06
        [parameter_eliminate]: 1.57999e-06
        [specialize_transform]: 5.54998e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 7.95e-06
        [meta_shard_fg_expand]: 2.21e-06
        [get_grad_eliminate_]: 5.53002e-06
        [merge_forward]: 4.08001e-06
        [cell_reuse_recompute_pass]: 4.50001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.468e-05
        [j_node_and_user_rematch]: 9.88002e-06
        [meta_fg_expand]: 2.14999e-06
        [replace_old_param]: 1.124e-05
        [inline_without_move]: 4.94998e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 1.17e-06
        [auto_monad_eliminator]: 4.97e-06
        [cse]: 1.306e-05
        [replace_applicator]: 5.30001e-06
[py_interpret_to_execute_after_opt_a]: 2.461e-05
[rewriter_after_opt_a]: 4.504e-05
[convert_after_rewriter]: 7.26001e-06
[order_py_execute_after_rewriter]: 5.04003e-06
[mutable_eliminate]: 0.00090154
[jit_opt_b]: 6.088e-05, [1]
    [Cycle 1]: 5.096e-05, [2]
        [frontend_op_eliminate]: 2.063e-05
        [inline_after_opt_a]: 1.773e-05
[cconv]: 3.517e-05
[loop_unroll]: 0.0004584
[jit_opt_after_cconv]: 0.00153903, [1]
    [Cycle 1]: 0.00152919, [11]
        [c_1]: 2.22e-05
        [parameter_eliminate]: 3.65998e-06
        [updatestate_depend_eliminate]: 7.04001e-06
        [updatestate_assign_eliminate]: 2.52001e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [cse]: 0.00011594
        [call_graph_tuple_transform]: 0.00065925
        [tuple_list_get_item_eliminator]: 8.70999e-06
        [none_parameter_eliminate]: 4.12e-06
        [renormalize]: 5.39992e-07
        [switch_simplify]: 6.06998e-06
[remove_dup_value]: 1.832e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 7.51999e-06
[add_recomputation]: 6.109e-05
[cse_after_recomputation]: 2.797e-05, [1]
    [Cycle 1]: 2.074e-05, [1]
        [cse]: 1.46e-05
[auto_monad_reorder]: 1.631e-05
[get_jit_bprop_graph]: 2.15002e-06
[rewriter_after_jit_bprop_graph]: 1.073e-05
[opt_after_jit_grad]: 0.00057248
[symbol_engine_optimizer]: 8.107e-05, [1]
    [Cycle 1]: 7.286e-05, [6]
        [build]: 5.10001e-06
        [elim_shapecalc]: 8.45999e-06
        [elim_not_effective]: 1.444e-05
        [opt_reshape]: 6.93e-06
        [fold_const_symbol]: 9.49999e-06
        [renormalize]: 3.89991e-07
[validate]: 4.753e-05
[backend_pass]: 8.79983e-07
[task_emit]: 0.00662376
[execute]: 9.69e-06
Sums
    bootstrap                            :     0.000378s :  1.09%
    type_inference                       :     0.009084s : 26.20%
    event_method                         :     0.000020s :  0.06%
    auto_monad                           :     0.000072s :  0.21%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000004s :  0.01%
    py_interpret_to_execute              :     0.000046s :  0.13%
    rewriter_before_opt_a                :     0.000084s :  0.24%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000071s :  0.20%
    jit_opt_a.loop_unroll                :     0.000036s :  0.10%
    jit_opt_a.a_1                        :     0.000711s :  2.05%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000014s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.03%
    jit_opt_a.merge_forward              :     0.000008s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000022s :  0.06%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.013996s : 40.37%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000035s :  0.10%
    jit_opt_a.cse                        :     0.000048s :  0.14%
    jit_opt_a.replace_applicator         :     0.000036s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000025s :  0.07%
    rewriter_after_opt_a                 :     0.000045s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000902s :  2.60%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.05%
    cconv                                :     0.000035s :  0.10%
    loop_unroll                          :     0.000458s :  1.32%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000116s :  0.33%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000659s :  1.90%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000018s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000008s :  0.02%
    add_recomputation                    :     0.000061s :  0.18%
    cse_after_recomputation.cse          :     0.000015s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.03%
    opt_after_jit_grad                   :     0.000572s :  1.65%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000048s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006624s : 19.10%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.000266    23
 0.71% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000002s :      2: substitution.fold_const_symbol
 3.63% :     0.000010s :      3: substitution.graph_param_transform
82.50% :     0.000219s :      4: substitution.inline
 1.86% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.46% :     0.000007s :      4: substitution.remove_not_recompute_node
 3.11% :     0.000008s :      2: substitution.replace_old_param
 4.97% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008996     2
89.68% :     0.008068s :      1: type_inference.infer
10.32% :     0.000928s :      1: type_inference.specialize
------[replace.]   0.000073     6
68.92% :     0.000050s :      4: replace.inline
31.08% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000228     6
94.72% :     0.000216s :      4: match.inline
 5.28% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.07% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.60% :     0.000002s :     10: predicate.addn_zero_filter
 2.73% :     0.000004s :     10: predicate.arithmetic_simplify
 1.54% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.99% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.47% :     0.000001s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.72% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     10: predicate.environ_get_eliminate
 1.12% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.62% :     0.000001s :      3: predicate.graph_param_transform
 5.88% :     0.000008s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.20% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.59% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.77% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 2.74% :     0.000004s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     10: predicate.reduce_eliminate
 1.90% :     0.000003s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.91% :     0.000003s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.99% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000002s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 1.93% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.03% :     0.000010s :     53: predicate.switch_simplify
 1.25% :     0.000002s :     10: predicate.tile_eliminate
 1.26% :     0.000002s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.33% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.46% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000732    12
53.36% :     0.000391s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.09% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.54% :     0.000290s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.055814    76
 0.11% :     0.000064s :      1: add_recomputation
 0.14% :     0.000076s :      1: auto_monad
 0.03% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.73% :     0.000408s :      1: bootstrap
 0.07% :     0.000038s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.05% :     0.000030s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.04% :     0.000025s :      1: event_method
 0.03% :     0.000015s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
35.11% :     0.019594s :      1: jit_opt_a
 2.76% :     0.001543s :      1: jit_opt_after_cconv
 0.11% :     0.000063s :      1: jit_opt_b
 0.84% :     0.000467s :      1: loop_unroll
 1.64% :     0.000915s :      1: mutable_eliminate
 1.74% :     0.000970s :     26: opt.transform.jit_opt_a
 1.24% :     0.000690s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000031s :      4: opt.transform.jit_opt_b
 0.03% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.04% :     0.000582s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.09% :     0.000049s :      1: py_interpret_to_execute
 0.05% :     0.000027s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000021s :      1: remove_dup_value
 1.02% :     0.000570s :      1: renormalize.infer
24.03% :     0.013412s :      1: renormalize.specialize
 0.02% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000049s :      1: rewriter_after_opt_a
 0.16% :     0.000087s :      1: rewriter_before_opt_a
 0.15% :     0.000084s :      1: symbol_engine_optimizer
11.90% :     0.006644s :      1: task_emit
16.31% :     0.009106s :      1: type_inference
 0.13% :     0.000071s :      1: validate


TotalTime = 0.0548096, [33]
[bootstrap]: 0.00041798
[type_inference]: 0.00610575
[event_method]: 2.006e-05
[auto_monad]: 7.041e-05
[graph_reusing]: 6.13998e-06
[pre_auto_parallel]: 2.97002e-06
[py_interpret_to_execute]: 4.602e-05
[rewriter_before_opt_a]: 7.708e-05
[expand_dump_flag]: 3.33e-06
[jit_opt_a]: 0.0389262, [2]
    [Cycle 1]: 0.00197991, [27]
        [switch_simplify]: 5.735e-05
        [loop_unroll]: 2.907e-05
        [a_1]: 0.00061978
        [with_stream_mark]: 2.37e-05
        [recompute_prepare]: 7.3e-06
        [updatestate_depend_eliminate]: 3.05002e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 1.37e-06
        [specialize_transform]: 6.91999e-06
        [updatestate_useless_node_eliminater]: 5.64e-06
        [accelerated_algorithm]: 5.63002e-06
        [meta_shard_fg_expand]: 1.59e-06
        [get_grad_eliminate_]: 5.51e-06
        [merge_forward]: 3.54002e-06
        [cell_reuse_recompute_pass]: 1.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.39e-05
        [j_node_and_user_rematch]: 8.80999e-06
        [meta_fg_expand]: 2.16e-06
        [replace_old_param]: 9.33002e-06
        [inline_without_move]: 5.55001e-06
        [renormalize]: 0.00092263
        [add_forward_monad_depend]: 6.36e-06
        [auto_monad_grad]: 2.66999e-06
        [auto_monad_eliminator]: 1.379e-05
        [cse]: 2.409e-05
        [replace_applicator]: 1.317e-05
    [Cycle 2]: 0.00033558, [27]
        [switch_simplify]: 5.86998e-06
        [loop_unroll]: 5.25999e-06
        [a_1]: 9.902e-05
        [with_stream_mark]: 1.114e-05
        [recompute_prepare]: 4.87e-06
        [updatestate_depend_eliminate]: 2.87002e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [parameter_eliminate]: 9.09989e-07
        [specialize_transform]: 4.95001e-06
        [updatestate_useless_node_eliminater]: 4.64002e-06
        [accelerated_algorithm]: 5.86998e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 4.84e-06
        [merge_forward]: 2.78e-06
        [cell_reuse_recompute_pass]: 2.00002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.353e-05
        [j_node_and_user_rematch]: 7.63999e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 7.55e-06
        [inline_without_move]: 4.77e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.38002e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 5.51998e-06
        [cse]: 1.132e-05
        [replace_applicator]: 4.72e-06
[py_interpret_to_execute_after_opt_a]: 1.235e-05
[rewriter_after_opt_a]: 3.458e-05
[convert_after_rewriter]: 6.79999e-06
[order_py_execute_after_rewriter]: 4.28001e-06
[mutable_eliminate]: 0.00074002
[jit_opt_b]: 5.644e-05, [1]
    [Cycle 1]: 4.936e-05, [2]
        [frontend_op_eliminate]: 2.244e-05
        [inline_after_opt_a]: 1.559e-05
[cconv]: 2.831e-05
[loop_unroll]: 0.00045381
[jit_opt_after_cconv]: 0.00014488, [1]
    [Cycle 1]: 0.00013775, [11]
        [c_1]: 2.174e-05
        [parameter_eliminate]: 3.06001e-06
        [updatestate_depend_eliminate]: 6.17999e-06
        [updatestate_assign_eliminate]: 2.86999e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 2.034e-05
        [call_graph_tuple_transform]: 1.906e-05
        [tuple_list_get_item_eliminator]: 4.95001e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.81e-06
[remove_dup_value]: 1.368e-05
[partial_unused_args_eliminate]: 1.93002e-06
[environ_conv]: 5.65001e-06
[add_recomputation]: 4.768e-05
[cse_after_recomputation]: 2.09e-05, [1]
    [Cycle 1]: 1.56e-05, [1]
        [cse]: 9.82999e-06
[auto_monad_reorder]: 1.543e-05
[get_jit_bprop_graph]: 1.97999e-06
[rewriter_after_jit_bprop_graph]: 7.23e-06
[opt_after_jit_grad]: 0.00051467
[symbol_engine_optimizer]: 7.928e-05, [1]
    [Cycle 1]: 7.059e-05, [6]
        [build]: 4.58999e-06
        [elim_shapecalc]: 7.98999e-06
        [elim_not_effective]: 1.406e-05
        [opt_reshape]: 6.40002e-06
        [fold_const_symbol]: 9.47999e-06
        [renormalize]: 3.80009e-07
[validate]: 4.24e-05
[backend_pass]: 9.99979e-07
[task_emit]: 0.00663236
[execute]: 9.39e-06
Sums
    bootstrap                            :     0.000418s :  2.39%
    type_inference                       :     0.006106s : 34.85%
    event_method                         :     0.000020s :  0.11%
    auto_monad                           :     0.000070s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000046s :  0.26%
    rewriter_before_opt_a                :     0.000077s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.36%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000719s :  4.10%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000923s :  5.27%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.11%
    jit_opt_a.cse                        :     0.000035s :  0.20%
    jit_opt_a.replace_applicator         :     0.000018s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000740s :  4.22%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000028s :  0.16%
    loop_unroll                          :     0.000454s :  2.59%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000048s :  0.27%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000515s :  2.94%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000042s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006632s : 37.86%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000256    23
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000002s :      2: substitution.fold_const_symbol
 1.95% :     0.000005s :      3: substitution.graph_param_transform
86.49% :     0.000222s :      4: substitution.inline
 1.13% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.84% :     0.000005s :      2: substitution.replace_old_param
 5.08% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006016     2
82.95% :     0.004990s :      1: type_inference.infer
17.05% :     0.001026s :      1: type_inference.specialize
------[replace.]   0.000079     6
69.04% :     0.000055s :      4: replace.inline
30.96% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000231     6
94.84% :     0.000219s :      4: match.inline
 5.16% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.62% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.51% :     0.000002s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.56% :     0.000002s :     10: predicate.cast_eliminate
 0.55% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000001s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.28% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.15% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.46% :     0.000002s :      6: predicate.less_batch_normalization
 1.75% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.61% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.32% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.74% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.07% :     0.000002s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.53% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.69% :     0.000002s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.68% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.63% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.28% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.18% :     0.000009s :     53: predicate.switch_simplify
 1.36% :     0.000002s :     10: predicate.tile_eliminate
 1.30% :     0.000002s :     10: predicate.transpose_eliminate
 1.24% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.09% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.75% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.92% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000776    12
49.50% :     0.000384s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.58% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.92% :     0.000341s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.056772    76
 0.09% :     0.000050s :      1: add_recomputation
 0.13% :     0.000074s :      1: auto_monad
 0.03% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.78% :     0.000442s :      1: bootstrap
 0.05% :     0.000031s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.04% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.04% :     0.000024s :      1: event_method
 0.03% :     0.000015s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000008s :      1: graph_reusing
68.57% :     0.038931s :      1: jit_opt_a
 0.26% :     0.000148s :      1: jit_opt_after_cconv
 0.10% :     0.000059s :      1: jit_opt_b
 0.81% :     0.000463s :      1: loop_unroll
 1.32% :     0.000751s :      1: mutable_eliminate
 1.65% :     0.000934s :     26: opt.transform.jit_opt_a
 0.09% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000031s :      4: opt.transform.jit_opt_b
 0.02% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000035s :      4: opt.transform.symbol_engine_opt
 0.92% :     0.000524s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000049s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000016s :      1: remove_dup_value
 0.89% :     0.000508s :      1: renormalize.infer
 0.72% :     0.000406s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000037s :      1: rewriter_after_opt_a
 0.14% :     0.000080s :      1: rewriter_before_opt_a
 0.14% :     0.000082s :      1: symbol_engine_optimizer
11.71% :     0.006651s :      1: task_emit
10.79% :     0.006126s :      1: type_inference
 0.15% :     0.000085s :      1: validate


TotalTime = 0.0316189, [30]
[bootstrap]: 0.00039823
[type_inference]: 0.00540156
[event_method]: 1.979e-05
[auto_monad]: 7.24e-05
[graph_reusing]: 6.54999e-06
[pre_auto_parallel]: 2.93e-06
[py_interpret_to_execute]: 4.351e-05
[rewriter_before_opt_a]: 7.996e-05
[expand_dump_flag]: 3.43e-06
[jit_opt_a]: 0.0229767, [2]
    [Cycle 1]: 0.00221993, [27]
        [switch_simplify]: 7.489e-05
        [loop_unroll]: 3.04e-05
        [a_1]: 0.00068851
        [with_stream_mark]: 2.702e-05
        [recompute_prepare]: 7.96001e-06
        [updatestate_depend_eliminate]: 4.74e-06
        [updatestate_assign_eliminate]: 3.51001e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 2.84999e-06
        [specialize_transform]: 5.87999e-06
        [updatestate_useless_node_eliminater]: 5.46e-06
        [accelerated_algorithm]: 6.07001e-06
        [meta_shard_fg_expand]: 2.22001e-06
        [get_grad_eliminate_]: 5.72999e-06
        [merge_forward]: 4.14002e-06
        [cell_reuse_recompute_pass]: 1.47999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.569e-05
        [j_node_and_user_rematch]: 1.022e-05
        [meta_fg_expand]: 2.69001e-06
        [replace_old_param]: 1.096e-05
        [inline_without_move]: 6.09001e-06
        [renormalize]: 0.00102362
        [add_forward_monad_depend]: 7.73001e-06
        [auto_monad_grad]: 2.56e-06
        [auto_monad_eliminator]: 1.74e-05
        [cse]: 3.421e-05
        [replace_applicator]: 1.481e-05
    [Cycle 2]: 0.00034544, [27]
        [switch_simplify]: 6.26e-06
        [loop_unroll]: 5.07e-06
        [a_1]: 9.709e-05
        [with_stream_mark]: 1.374e-05
        [recompute_prepare]: 4.94003e-06
        [updatestate_depend_eliminate]: 3.3e-06
        [updatestate_assign_eliminate]: 2.01998e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 5.05999e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 5.62999e-06
        [meta_shard_fg_expand]: 1.75001e-06
        [get_grad_eliminate_]: 4.57e-06
        [merge_forward]: 2.97002e-06
        [cell_reuse_recompute_pass]: 2.53e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.392e-05
        [j_node_and_user_rematch]: 8.64e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 8e-06
        [inline_without_move]: 5.61003e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.67001e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 5.34998e-06
        [cse]: 1.08e-05
        [replace_applicator]: 5.57001e-06
[py_interpret_to_execute_after_opt_a]: 1.606e-05
[rewriter_after_opt_a]: 4.09e-05
[convert_after_rewriter]: 8.20999e-06
[order_py_execute_after_rewriter]: 5.41998e-06
[mutable_eliminate]: 0.00084779
[jit_opt_b]: 5.621e-05, [1]
    [Cycle 1]: 4.766e-05, [2]
        [frontend_op_eliminate]: 1.932e-05
        [inline_after_opt_a]: 1.631e-05
[cconv]: 3.5e-05
[loop_unroll]: 0.00047781
[jit_opt_after_cconv]: 0.00015856, [1]
    [Cycle 1]: 0.0001509, [11]
        [c_1]: 2.225e-05
        [parameter_eliminate]: 3.68e-06
        [updatestate_depend_eliminate]: 7.7e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [cse]: 2.511e-05
        [call_graph_tuple_transform]: 2.094e-05
        [tuple_list_get_item_eliminator]: 6.26998e-06
        [none_parameter_eliminate]: 1.92001e-06
        [renormalize]: 5.79981e-07
        [switch_simplify]: 5.35999e-06
[remove_dup_value]: 1.57e-05
[partial_unused_args_eliminate]: 2.72001e-06
[environ_conv]: 5.97999e-06
[add_recomputation]: 5.61e-05
[cse_after_recomputation]: 2.121e-05, [1]
    [Cycle 1]: 1.481e-05, [1]
        [cse]: 8.65001e-06
[auto_monad_reorder]: 1.618e-05
[get_jit_bprop_graph]: 2.69001e-06
[rewriter_after_jit_bprop_graph]: 7.25e-06
[opt_after_jit_grad]: 0.00049802
[symbol_engine_optimizer]: 8.398e-05, [1]
    [Cycle 1]: 7.626e-05, [6]
        [build]: 4.55999e-06
        [elim_shapecalc]: 8.27e-06
        [elim_not_effective]: 1.534e-05
        [opt_reshape]: 7.58999e-06
        [fold_const_symbol]: 9.79999e-06
        [renormalize]: 7.7e-07
[validate]: 4.598e-05
Sums
    bootstrap                            :     0.000398s :  3.78%
    type_inference                       :     0.005402s : 51.23%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000072s :  0.69%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000044s :  0.41%
    rewriter_before_opt_a                :     0.000080s :  0.76%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000081s :  0.77%
    jit_opt_a.loop_unroll                :     0.000035s :  0.34%
    jit_opt_a.a_1                        :     0.000786s :  7.45%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.39%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.18%
    jit_opt_a.inline_without_move        :     0.000012s :  0.11%
    jit_opt_a.renormalize                :     0.001024s :  9.71%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.22%
    jit_opt_a.cse                        :     0.000045s :  0.43%
    jit_opt_a.replace_applicator         :     0.000020s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.15%
    rewriter_after_opt_a                 :     0.000041s :  0.39%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000848s :  8.04%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.15%
    cconv                                :     0.000035s :  0.33%
    loop_unroll                          :     0.000478s :  4.53%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000025s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000016s :  0.15%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000056s :  0.53%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.15%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000498s :  4.72%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000046s :  0.44%


Time group info:
------[substitution.]   0.000307    23
 0.83% :     0.000003s :      2: substitution.elim_not_effective
 0.57% :     0.000002s :      2: substitution.fold_const_symbol
 2.03% :     0.000006s :      3: substitution.graph_param_transform
86.65% :     0.000266s :      4: substitution.inline
 1.44% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.67% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.94% :     0.000006s :      2: substitution.replace_old_param
 4.87% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005315     2
80.72% :     0.004290s :      1: type_inference.infer
19.28% :     0.001025s :      1: type_inference.specialize
------[replace.]   0.000087     6
66.87% :     0.000058s :      4: replace.inline
33.13% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000277     6
95.18% :     0.000263s :      4: match.inline
 4.82% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   691
 1.11% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.47% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000002s :     10: predicate.addn_zero_filter
 2.25% :     0.000003s :     10: predicate.arithmetic_simplify
 1.64% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.93% :     0.000001s :     10: predicate.compare_switch_simplify
 1.56% :     0.000002s :     10: predicate.depend_value_elim
 1.24% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.61% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.34% :     0.000002s :     10: predicate.environ_get_eliminate
 1.16% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.68% :     0.000008s :     22: predicate.inline
 1.07% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000002s :      6: predicate.less_batch_normalization
 1.35% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.44% :     0.000002s :     15: predicate.load_eliminater
 1.65% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.60% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.93% :     0.000001s :     10: predicate.merge_addn
 1.24% :     0.000002s :     10: predicate.minmaximum_grad
 3.06% :     0.000004s :      3: predicate.mutable_eliminate
 0.59% :     0.000001s :      3: predicate.opt_reshape
 1.95% :     0.000003s :     15: predicate.partial_eliminate
 1.09% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.25% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.40% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.45% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.49% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.18% :     0.000002s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.21% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.52% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 1.88% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.20% :     0.000010s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.07% :     0.000001s :     10: predicate.transpose_eliminate
 1.75% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.45% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.41% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.02% :     0.000003s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.99% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000764    12
48.01% :     0.000367s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.17% :     0.000062s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.81% :     0.000335s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.033717    72
 0.18% :     0.000059s :      1: add_recomputation
 0.23% :     0.000076s :      1: auto_monad
 0.06% :     0.000019s :      1: auto_monad_reorder
 1.26% :     0.000424s :      1: bootstrap
 0.11% :     0.000038s :      1: cconv
 0.04% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.08% :     0.000025s :      1: event_method
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
68.16% :     0.022981s :      1: jit_opt_a
 0.48% :     0.000161s :      1: jit_opt_after_cconv
 0.18% :     0.000059s :      1: jit_opt_b
 1.44% :     0.000487s :      1: loop_unroll
 2.55% :     0.000860s :      1: mutable_eliminate
 3.04% :     0.001025s :     26: opt.transform.jit_opt_a
 0.15% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000028s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.51% :     0.000508s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000046s :      1: py_interpret_to_execute
 0.05% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 1.69% :     0.000571s :      1: renormalize.infer
 1.31% :     0.000442s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000045s :      1: rewriter_after_opt_a
 0.25% :     0.000083s :      1: rewriter_before_opt_a
 0.26% :     0.000087s :      1: symbol_engine_optimizer
16.08% :     0.005422s :      1: type_inference


TotalTime = 0.020951, [33]
[bootstrap]: 0.00037123
[type_inference]: 0.00512117
[event_method]: 1.963e-05
[auto_monad]: 6.779e-05
[graph_reusing]: 6.39999e-06
[pre_auto_parallel]: 2.57001e-06
[py_interpret_to_execute]: 4.202e-05
[rewriter_before_opt_a]: 7.711e-05
[expand_dump_flag]: 3.78999e-06
[jit_opt_a]: 0.00613525, [2]
    [Cycle 1]: 0.00196214, [27]
        [switch_simplify]: 6.039e-05
        [loop_unroll]: 3.043e-05
        [a_1]: 0.00058535
        [with_stream_mark]: 2.318e-05
        [recompute_prepare]: 8.38999e-06
        [updatestate_depend_eliminate]: 4.23001e-06
        [updatestate_assign_eliminate]: 3.04999e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 2.01e-06
        [specialize_transform]: 6.14001e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.57001e-06
        [meta_shard_fg_expand]: 2.14e-06
        [get_grad_eliminate_]: 5.15001e-06
        [merge_forward]: 4.07e-06
        [cell_reuse_recompute_pass]: 1.77999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.473e-05
        [j_node_and_user_rematch]: 9.24e-06
        [meta_fg_expand]: 2.36998e-06
        [replace_old_param]: 9.20999e-06
        [inline_without_move]: 5.59998e-06
        [renormalize]: 0.00091758
        [add_forward_monad_depend]: 7.25003e-06
        [auto_monad_grad]: 2.64001e-06
        [auto_monad_eliminator]: 1.537e-05
        [cse]: 3.16e-05
        [replace_applicator]: 1.423e-05
    [Cycle 2]: 0.00033642, [27]
        [switch_simplify]: 5.72001e-06
        [loop_unroll]: 5.07e-06
        [a_1]: 9.289e-05
        [with_stream_mark]: 1.07e-05
        [recompute_prepare]: 5.25999e-06
        [updatestate_depend_eliminate]: 2.66e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 9.49978e-07
        [specialize_transform]: 4.99e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 5.36998e-06
        [meta_shard_fg_expand]: 1.37999e-06
        [get_grad_eliminate_]: 4.70999e-06
        [merge_forward]: 2.92002e-06
        [cell_reuse_recompute_pass]: 1.64998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.408e-05
        [j_node_and_user_rematch]: 8.60001e-06
        [meta_fg_expand]: 1.82001e-06
        [replace_old_param]: 7.38999e-06
        [inline_without_move]: 4.80999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.42e-06
        [auto_monad_grad]: 1.02e-06
        [auto_monad_eliminator]: 5.64e-06
        [cse]: 1.198e-05
        [replace_applicator]: 5.77001e-06
[py_interpret_to_execute_after_opt_a]: 1.536e-05
[rewriter_after_opt_a]: 4.288e-05
[convert_after_rewriter]: 1.116e-05
[order_py_execute_after_rewriter]: 5.34e-06
[mutable_eliminate]: 0.00072145
[jit_opt_b]: 5.495e-05, [1]
    [Cycle 1]: 4.688e-05, [2]
        [frontend_op_eliminate]: 1.737e-05
        [inline_after_opt_a]: 1.655e-05
[cconv]: 3.106e-05
[loop_unroll]: 0.00046653
[jit_opt_after_cconv]: 0.00016054, [1]
    [Cycle 1]: 0.00015366, [11]
        [c_1]: 2.148e-05
        [parameter_eliminate]: 3.8e-06
        [updatestate_depend_eliminate]: 9.26002e-06
        [updatestate_assign_eliminate]: 2.43998e-06
        [updatestate_loads_eliminate]: 2.30002e-06
        [cse]: 2.443e-05
        [call_graph_tuple_transform]: 2.225e-05
        [tuple_list_get_item_eliminator]: 6.49001e-06
        [none_parameter_eliminate]: 1.76003e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.84e-06
[remove_dup_value]: 1.519e-05
[partial_unused_args_eliminate]: 2.65997e-06
[environ_conv]: 5.81e-06
[add_recomputation]: 5.244e-05
[cse_after_recomputation]: 2.104e-05, [1]
    [Cycle 1]: 1.483e-05, [1]
        [cse]: 9.15001e-06
[auto_monad_reorder]: 1.593e-05
[get_jit_bprop_graph]: 2.21e-06
[rewriter_after_jit_bprop_graph]: 6.96999e-06
[opt_after_jit_grad]: 0.00046882
[symbol_engine_optimizer]: 7.723e-05, [1]
    [Cycle 1]: 7.05e-05, [6]
        [build]: 4.34002e-06
        [elim_shapecalc]: 7.62002e-06
        [elim_not_effective]: 1.422e-05
        [opt_reshape]: 6.06e-06
        [fold_const_symbol]: 9.42999e-06
        [renormalize]: 5.19998e-07
[validate]: 4.155e-05
[backend_pass]: 8.60018e-07
[task_emit]: 0.00663951
[execute]: 1.14e-05
Sums
    bootstrap                            :     0.000371s :  2.26%
    type_inference                       :     0.005121s : 31.13%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000068s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000042s :  0.26%
    rewriter_before_opt_a                :     0.000077s :  0.47%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.40%
    jit_opt_a.loop_unroll                :     0.000036s :  0.22%
    jit_opt_a.a_1                        :     0.000678s :  4.12%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000918s :  5.58%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.13%
    jit_opt_a.cse                        :     0.000044s :  0.26%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.09%
    rewriter_after_opt_a                 :     0.000043s :  0.26%
    convert_after_rewriter               :     0.000011s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000721s :  4.39%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.10%
    cconv                                :     0.000031s :  0.19%
    loop_unroll                          :     0.000467s :  2.84%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000052s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000469s :  2.85%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006640s : 40.36%
    execute                              :     0.000011s :  0.07%


Time group info:
------[substitution.]   0.000244    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.43% :     0.000006s :      3: substitution.graph_param_transform
85.53% :     0.000209s :      4: substitution.inline
 1.31% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.06% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.79% :     0.000004s :      2: substitution.replace_old_param
 5.35% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005046     2
82.91% :     0.004183s :      1: type_inference.infer
17.09% :     0.000862s :      1: type_inference.specialize
------[replace.]   0.000073     6
68.10% :     0.000049s :      4: replace.inline
31.90% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000218     6
94.57% :     0.000206s :      4: match.inline
 5.43% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.05% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 1.84% :     0.000002s :     10: predicate.arithmetic_simplify
 1.45% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.24% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.06% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.46% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.48% :     0.000001s :      3: predicate.graph_param_transform
 6.73% :     0.000008s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.97% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.61% :     0.000002s :     15: predicate.load_eliminater
 1.73% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.07% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 2.10% :     0.000003s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.46% :     0.000003s :     15: predicate.partial_eliminate
 1.06% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     10: predicate.reduce_eliminate
 1.35% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.49% :     0.000001s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.73% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.12% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.21% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.80% :     0.000009s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.67% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.86% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.96% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000660    12
50.57% :     0.000334s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.68% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.75% :     0.000275s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022876    76
 0.24% :     0.000055s :      1: add_recomputation
 0.31% :     0.000071s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.71% :     0.000390s :      1: bootstrap
 0.15% :     0.000034s :      1: cconv
 0.06% :     0.000014s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000025s :      1: event_method
 0.08% :     0.000017s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.84% :     0.006139s :      1: jit_opt_a
 0.72% :     0.000164s :      1: jit_opt_after_cconv
 0.25% :     0.000058s :      1: jit_opt_b
 2.08% :     0.000476s :      1: loop_unroll
 3.21% :     0.000734s :      1: mutable_eliminate
 3.93% :     0.000898s :     26: opt.transform.jit_opt_a
 0.23% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.09% :     0.000478s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000045s :      1: py_interpret_to_execute
 0.08% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 2.52% :     0.000576s :      1: renormalize.infer
 1.45% :     0.000332s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000047s :      1: rewriter_after_opt_a
 0.35% :     0.000080s :      1: rewriter_before_opt_a
 0.35% :     0.000080s :      1: symbol_engine_optimizer
29.11% :     0.006659s :      1: task_emit
22.47% :     0.005139s :      1: type_inference
 0.28% :     0.000065s :      1: validate


TotalTime = 0.0475616, [33]
[bootstrap]: 0.00042856
[type_inference]: 0.00544456
[event_method]: 1.915e-05
[auto_monad]: 6.505e-05
[graph_reusing]: 6.24001e-06
[pre_auto_parallel]: 2.89999e-06
[py_interpret_to_execute]: 4.422e-05
[rewriter_before_opt_a]: 7.513e-05
[expand_dump_flag]: 3.16999e-06
[jit_opt_a]: 0.00697635, [2]
    [Cycle 1]: 0.00210625, [27]
        [switch_simplify]: 6.745e-05
        [loop_unroll]: 2.984e-05
        [a_1]: 0.00066665
        [with_stream_mark]: 2.718e-05
        [recompute_prepare]: 8.32998e-06
        [updatestate_depend_eliminate]: 3.88001e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 3.10998e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 6.06998e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 6.26998e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 5.61e-06
        [merge_forward]: 3.66999e-06
        [cell_reuse_recompute_pass]: 2.11e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.481e-05
        [j_node_and_user_rematch]: 1.016e-05
        [meta_fg_expand]: 2.91999e-06
        [replace_old_param]: 1.082e-05
        [inline_without_move]: 5.56e-06
        [renormalize]: 0.00094128
        [add_forward_monad_depend]: 7.96001e-06
        [auto_monad_grad]: 2.52001e-06
        [auto_monad_eliminator]: 1.781e-05
        [cse]: 3.376e-05
        [replace_applicator]: 1.781e-05
    [Cycle 2]: 0.00037457, [27]
        [switch_simplify]: 7.01999e-06
        [loop_unroll]: 5.81e-06
        [a_1]: 9.678e-05
        [with_stream_mark]: 1.51e-05
        [recompute_prepare]: 5.69e-06
        [updatestate_depend_eliminate]: 3.20998e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [parameter_eliminate]: 1.60999e-06
        [specialize_transform]: 4.71002e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 5.89999e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 5.17e-06
        [merge_forward]: 3.91999e-06
        [cell_reuse_recompute_pass]: 3.19001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.692e-05
        [j_node_and_user_rematch]: 9.52999e-06
        [meta_fg_expand]: 1.82001e-06
        [replace_old_param]: 8.43999e-06
        [inline_without_move]: 6.66e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.52001e-06
        [auto_monad_grad]: 1.60999e-06
        [auto_monad_eliminator]: 7.13998e-06
        [cse]: 1.341e-05
        [replace_applicator]: 5.22e-06
[py_interpret_to_execute_after_opt_a]: 1.679e-05
[rewriter_after_opt_a]: 3.848e-05
[convert_after_rewriter]: 7.48e-06
[order_py_execute_after_rewriter]: 5.35999e-06
[mutable_eliminate]: 0.00086094
[jit_opt_b]: 6.316e-05, [1]
    [Cycle 1]: 5.415e-05, [2]
        [frontend_op_eliminate]: 2.062e-05
        [inline_after_opt_a]: 1.944e-05
[cconv]: 3.44e-05
[loop_unroll]: 0.00052441
[jit_opt_after_cconv]: 0.00017348, [1]
    [Cycle 1]: 0.00016568, [11]
        [c_1]: 2.338e-05
        [parameter_eliminate]: 4.38999e-06
        [updatestate_depend_eliminate]: 9.96e-06
        [updatestate_assign_eliminate]: 2.92002e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [cse]: 3.067e-05
        [call_graph_tuple_transform]: 2.451e-05
        [tuple_list_get_item_eliminator]: 5.51998e-06
        [none_parameter_eliminate]: 1.77999e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 5.62999e-06
[remove_dup_value]: 8.245e-05
[partial_unused_args_eliminate]: 2.89999e-06
[environ_conv]: 6.65002e-06
[add_recomputation]: 6.065e-05
[cse_after_recomputation]: 2.543e-05, [1]
    [Cycle 1]: 1.827e-05, [1]
        [cse]: 1.077e-05
[auto_monad_reorder]: 1.567e-05
[get_jit_bprop_graph]: 2.15002e-06
[rewriter_after_jit_bprop_graph]: 9.96998e-06
[opt_after_jit_grad]: 0.00055053
[symbol_engine_optimizer]: 8.642e-05, [1]
    [Cycle 1]: 7.769e-05, [6]
        [build]: 5.71e-06
        [elim_shapecalc]: 8.99e-06
        [elim_not_effective]: 1.427e-05
        [opt_reshape]: 6.38e-06
        [fold_const_symbol]: 1.057e-05
        [renormalize]: 1.05001e-06
[validate]: 4.84e-05
[backend_pass]: 9.99979e-07
[task_emit]: 0.0316237
[execute]: 8.72998e-06
Sums
    bootstrap                            :     0.000429s :  1.01%
    type_inference                       :     0.005445s : 12.86%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000065s :  0.15%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.10%
    rewriter_before_opt_a                :     0.000075s :  0.18%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000074s :  0.18%
    jit_opt_a.loop_unroll                :     0.000036s :  0.08%
    jit_opt_a.a_1                        :     0.000763s :  1.80%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.03%
    jit_opt_a.merge_forward              :     0.000008s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.07%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000019s :  0.05%
    jit_opt_a.inline_without_move        :     0.000012s :  0.03%
    jit_opt_a.renormalize                :     0.000941s :  2.22%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.06%
    jit_opt_a.cse                        :     0.000047s :  0.11%
    jit_opt_a.replace_applicator         :     0.000023s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.04%
    rewriter_after_opt_a                 :     0.000038s :  0.09%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000861s :  2.03%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.05%
    cconv                                :     0.000034s :  0.08%
    loop_unroll                          :     0.000524s :  1.24%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000031s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.01%
    remove_dup_value                     :     0.000082s :  0.19%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000061s :  0.14%
    cse_after_recomputation.cse          :     0.000011s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.02%
    opt_after_jit_grad                   :     0.000551s :  1.30%
    symbol_engine_optimizer.build        :     0.000006s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000048s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.031624s : 74.68%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000317    23
 0.62% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 2.28% :     0.000007s :      3: substitution.graph_param_transform
87.58% :     0.000277s :      4: substitution.inline
 1.29% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.66% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.89% :     0.000006s :      2: substitution.replace_old_param
 4.22% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005365     2
81.68% :     0.004382s :      1: type_inference.infer
18.32% :     0.000983s :      1: type_inference.specialize
------[replace.]   0.000080     6
69.98% :     0.000056s :      4: replace.inline
30.02% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000288     6
95.69% :     0.000275s :      4: match.inline
 4.31% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000002s :     10: predicate.addn_zero_filter
 2.20% :     0.000003s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.46% :     0.000002s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.02% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.06% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.44% :     0.000002s :      6: predicate.get_grad_eliminate
 0.36% :     0.000000s :      3: predicate.graph_param_transform
 5.63% :     0.000007s :     22: predicate.inline
 1.34% :     0.000002s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.58% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.61% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000002s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 2.09% :     0.000003s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.09% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.82% :     0.000002s :     10: predicate.reduce_eliminate
 1.92% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.77% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.94% :     0.000001s :      6: predicate.replace_old_param
 0.64% :     0.000001s :      3: predicate.reset_defer_inline
 1.21% :     0.000002s :     10: predicate.reshape_eliminate
 1.06% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.10% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.63% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.10% :     0.000001s :      6: predicate.special_op_eliminate
 0.84% :     0.000001s :      6: predicate.specialize_transform
 1.21% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.61% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.66% :     0.000003s :     16: predicate.switch_defer_inline
 1.85% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.50% :     0.000010s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.33% :     0.000002s :     10: predicate.transpose_eliminate
 1.25% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.95% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000714    12
46.75% :     0.000334s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.28% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.97% :     0.000328s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.049629    76
 0.13% :     0.000064s :      1: add_recomputation
 0.14% :     0.000069s :      1: auto_monad
 0.04% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.90% :     0.000447s :      1: bootstrap
 0.08% :     0.000038s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.06% :     0.000028s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.05% :     0.000023s :      1: event_method
 0.03% :     0.000013s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
14.07% :     0.006982s :      1: jit_opt_a
 0.36% :     0.000177s :      1: jit_opt_after_cconv
 0.13% :     0.000066s :      1: jit_opt_b
 1.07% :     0.000533s :      1: loop_unroll
 1.76% :     0.000876s :      1: mutable_eliminate
 2.03% :     0.001007s :     26: opt.transform.jit_opt_a
 0.11% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000032s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.13% :     0.000560s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000047s :      1: py_interpret_to_execute
 0.04% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.17% :     0.000086s :      1: remove_dup_value
 1.04% :     0.000516s :      1: renormalize.infer
 0.83% :     0.000414s :      1: renormalize.specialize
 0.03% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000042s :      1: rewriter_after_opt_a
 0.16% :     0.000078s :      1: rewriter_before_opt_a
 0.18% :     0.000089s :      1: symbol_engine_optimizer
63.76% :     0.031642s :      1: task_emit
11.01% :     0.005464s :      1: type_inference
 0.15% :     0.000073s :      1: validate


TotalTime = 0.034385, [33]
[bootstrap]: 0.0003795
[type_inference]: 0.00548863
[event_method]: 1.848e-05
[auto_monad]: 6.491e-05
[graph_reusing]: 6.11e-06
[pre_auto_parallel]: 2.54999e-06
[py_interpret_to_execute]: 4.298e-05
[rewriter_before_opt_a]: 7.485e-05
[expand_dump_flag]: 3.06999e-06
[jit_opt_a]: 0.0186113, [2]
    [Cycle 1]: 0.0145544, [27]
        [switch_simplify]: 6.404e-05
        [loop_unroll]: 3.246e-05
        [a_1]: 0.0129294
        [with_stream_mark]: 3.619e-05
        [recompute_prepare]: 1.408e-05
        [updatestate_depend_eliminate]: 5.24e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.67001e-06
        [parameter_eliminate]: 2.29999e-06
        [specialize_transform]: 7.28999e-06
        [updatestate_useless_node_eliminater]: 6.21e-06
        [accelerated_algorithm]: 7.66001e-06
        [meta_shard_fg_expand]: 3.93001e-06
        [get_grad_eliminate_]: 6.51e-06
        [merge_forward]: 4.43999e-06
        [cell_reuse_recompute_pass]: 4.57998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.809e-05
        [j_node_and_user_rematch]: 1.188e-05
        [meta_fg_expand]: 3.33998e-06
        [replace_old_param]: 1.178e-05
        [inline_without_move]: 6.19001e-06
        [renormalize]: 0.00106041
        [add_forward_monad_depend]: 8.43001e-06
        [auto_monad_grad]: 2.68003e-06
        [auto_monad_eliminator]: 1.737e-05
        [cse]: 2.938e-05
        [replace_applicator]: 1.761e-05
    [Cycle 2]: 0.0003895, [27]
        [switch_simplify]: 6.68e-06
        [loop_unroll]: 6.27001e-06
        [a_1]: 0.00011837
        [with_stream_mark]: 1.518e-05
        [recompute_prepare]: 6.46e-06
        [updatestate_depend_eliminate]: 3.66999e-06
        [updatestate_assign_eliminate]: 2.69999e-06
        [updatestate_loads_eliminate]: 2.36998e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 5.78002e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 5.88002e-06
        [meta_shard_fg_expand]: 1.72001e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 2.82002e-06
        [cell_reuse_recompute_pass]: 3.03e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.712e-05
        [j_node_and_user_rematch]: 9.01002e-06
        [meta_fg_expand]: 1.92001e-06
        [replace_old_param]: 1.676e-05
        [inline_without_move]: 5.42001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.78002e-06
        [auto_monad_grad]: 9.29984e-07
        [auto_monad_eliminator]: 5.47999e-06
        [cse]: 1.069e-05
        [replace_applicator]: 5.27001e-06
[py_interpret_to_execute_after_opt_a]: 1.566e-05
[rewriter_after_opt_a]: 3.875e-05
[convert_after_rewriter]: 6.88e-06
[order_py_execute_after_rewriter]: 4.88001e-06
[mutable_eliminate]: 0.00084971
[jit_opt_b]: 6.292e-05, [1]
    [Cycle 1]: 5.395e-05, [2]
        [frontend_op_eliminate]: 1.998e-05
        [inline_after_opt_a]: 2.012e-05
[cconv]: 3.493e-05
[loop_unroll]: 0.00051276
[jit_opt_after_cconv]: 0.00027409, [1]
    [Cycle 1]: 0.00026575, [11]
        [c_1]: 2.526e-05
        [parameter_eliminate]: 5.81e-06
        [updatestate_depend_eliminate]: 9.79e-06
        [updatestate_assign_eliminate]: 3.5e-06
        [updatestate_loads_eliminate]: 3.91999e-06
        [cse]: 3.267e-05
        [call_graph_tuple_transform]: 2.938e-05
        [tuple_list_get_item_eliminator]: 6.55997e-06
        [none_parameter_eliminate]: 2.39999e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 7.08998e-06
[remove_dup_value]: 1.387e-05
[partial_unused_args_eliminate]: 2.84999e-06
[environ_conv]: 6.29999e-06
[add_recomputation]: 5.599e-05
[cse_after_recomputation]: 2.361e-05, [1]
    [Cycle 1]: 1.679e-05, [1]
        [cse]: 1.023e-05
[auto_monad_reorder]: 1.626e-05
[get_jit_bprop_graph]: 2.58e-06
[rewriter_after_jit_bprop_graph]: 1.014e-05
[opt_after_jit_grad]: 0.00052767
[symbol_engine_optimizer]: 8.434e-05, [1]
    [Cycle 1]: 7.645e-05, [6]
        [build]: 5.12e-06
        [elim_shapecalc]: 9.61e-06
        [elim_not_effective]: 1.516e-05
        [opt_reshape]: 6.66e-06
        [fold_const_symbol]: 9.86998e-06
        [renormalize]: 6.50005e-07
[validate]: 4.056e-05
[backend_pass]: 8.59989e-07
[task_emit]: 0.00685981
[execute]: 9.14e-06
Sums
    bootstrap                            :     0.000380s :  1.27%
    type_inference                       :     0.005489s : 18.36%
    event_method                         :     0.000018s :  0.06%
    auto_monad                           :     0.000065s :  0.22%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.14%
    rewriter_before_opt_a                :     0.000075s :  0.25%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000071s :  0.24%
    jit_opt_a.loop_unroll                :     0.000039s :  0.13%
    jit_opt_a.a_1                        :     0.013048s : 43.64%
    jit_opt_a.with_stream_mark           :     0.000051s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000013s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000014s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000008s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000029s :  0.10%
    jit_opt_a.inline_without_move        :     0.000012s :  0.04%
    jit_opt_a.renormalize                :     0.001060s :  3.55%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.08%
    jit_opt_a.cse                        :     0.000040s :  0.13%
    jit_opt_a.replace_applicator         :     0.000023s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.05%
    rewriter_after_opt_a                 :     0.000039s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000850s :  2.84%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.07%
    cconv                                :     0.000035s :  0.12%
    loop_unroll                          :     0.000513s :  1.71%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000033s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000029s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000014s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000056s :  0.19%
    cse_after_recomputation.cse          :     0.000010s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.03%
    opt_after_jit_grad                   :     0.000528s :  1.76%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006860s : 22.94%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000275    23
 0.97% :     0.000003s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.59% :     0.000007s :      3: substitution.graph_param_transform
78.31% :     0.000216s :      4: substitution.inline
 1.62% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.37% :     0.000007s :      4: substitution.remove_not_recompute_node
 5.30% :     0.000015s :      2: substitution.replace_old_param
 8.34% :     0.000023s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005410     2
81.45% :     0.004407s :      1: type_inference.infer
18.55% :     0.001003s :      1: type_inference.specialize
------[replace.]   0.000147     6
35.33% :     0.000052s :      4: replace.inline
64.67% :     0.000095s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000232     6
91.84% :     0.000213s :      4: match.inline
 8.16% :     0.000019s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000166   691
 1.53% :     0.000003s :     10: predicate.accumulaten_eliminater
 1.04% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.43% :     0.000002s :     10: predicate.addn_zero_filter
 2.78% :     0.000005s :     10: predicate.arithmetic_simplify
 1.58% :     0.000003s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000002s :     10: predicate.compare_switch_simplify
 1.22% :     0.000002s :     10: predicate.depend_value_elim
 1.20% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.69% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.24% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.06% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.87% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     10: predicate.environ_get_eliminate
 1.16% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.17% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000002s :      6: predicate.get_grad_eliminate
 0.31% :     0.000001s :      3: predicate.graph_param_transform
 5.59% :     0.000009s :     22: predicate.inline
 0.77% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :      6: predicate.less_batch_normalization
 3.90% :     0.000006s :     12: predicate.list_to_tuple_eliminator_
 1.49% :     0.000002s :     15: predicate.load_eliminater
 1.52% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.50% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.16% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000002s :     10: predicate.merge_addn
 0.90% :     0.000001s :     10: predicate.minmaximum_grad
 2.17% :     0.000004s :      3: predicate.mutable_eliminate
 0.59% :     0.000001s :      3: predicate.opt_reshape
 1.88% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.76% :     0.000003s :     10: predicate.reduce_eliminate
 1.55% :     0.000003s :     12: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.93% :     0.000005s :     18: predicate.replace_applicator
 0.42% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.51% :     0.000003s :     10: predicate.reshape_eliminate
 1.33% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.72% :     0.000001s :      6: predicate.special_op_eliminate
 0.69% :     0.000001s :      6: predicate.specialize_transform
 1.25% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.82% :     0.000005s :     16: predicate.switch_defer_inline
 1.68% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.33% :     0.000011s :     53: predicate.switch_simplify
 1.37% :     0.000002s :     10: predicate.tile_eliminate
 1.25% :     0.000002s :     10: predicate.transpose_eliminate
 1.76% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.01% :     0.000007s :     18: predicate.tuple_list_get_item_eliminator
 1.99% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.34% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.31% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000003s :     10: predicate.value_based_eliminate
 0.27% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.46% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000767    12
45.97% :     0.000353s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.43% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.60% :     0.000365s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.048881    76
 0.12% :     0.000059s :      1: add_recomputation
 0.14% :     0.000068s :      1: auto_monad
 0.04% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.81% :     0.000398s :      1: bootstrap
 0.08% :     0.000038s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.05% :     0.000026s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.05% :     0.000022s :      1: event_method
 0.03% :     0.000014s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
38.08% :     0.018616s :      1: jit_opt_a
 0.57% :     0.000278s :      1: jit_opt_after_cconv
 0.13% :     0.000066s :      1: jit_opt_b
 1.07% :     0.000523s :      1: loop_unroll
 1.76% :     0.000862s :      1: mutable_eliminate
27.23% :     0.013308s :     26: opt.transform.jit_opt_a
 0.13% :     0.000064s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000032s :      4: opt.transform.jit_opt_b
 0.03% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.10% :     0.000536s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.09% :     0.000046s :      1: py_interpret_to_execute
 0.04% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000017s :      1: remove_dup_value
 1.14% :     0.000557s :      1: renormalize.infer
 1.00% :     0.000491s :      1: renormalize.specialize
 0.03% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000042s :      1: rewriter_after_opt_a
 0.16% :     0.000078s :      1: rewriter_before_opt_a
 0.18% :     0.000087s :      1: symbol_engine_optimizer
14.07% :     0.006879s :      1: task_emit
11.27% :     0.005507s :      1: type_inference
 0.13% :     0.000062s :      1: validate


TotalTime = 0.0150896, [30]
[bootstrap]: 0.00035882
[type_inference]: 0.00533723
[event_method]: 1.79e-05
[auto_monad]: 6.866e-05
[graph_reusing]: 6.49001e-06
[pre_auto_parallel]: 3.25998e-06
[py_interpret_to_execute]: 4.486e-05
[rewriter_before_opt_a]: 7.56e-05
[expand_dump_flag]: 3.18e-06
[jit_opt_a]: 0.00655521, [2]
    [Cycle 1]: 0.00205935, [27]
        [switch_simplify]: 5.5e-05
        [loop_unroll]: 3.016e-05
        [a_1]: 0.00058753
        [with_stream_mark]: 2.136e-05
        [recompute_prepare]: 7.36001e-06
        [updatestate_depend_eliminate]: 3.45e-06
        [updatestate_assign_eliminate]: 2.40002e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 6.29001e-06
        [updatestate_useless_node_eliminater]: 5.25001e-06
        [accelerated_algorithm]: 5.69e-06
        [meta_shard_fg_expand]: 2.10002e-06
        [get_grad_eliminate_]: 5.01997e-06
        [merge_forward]: 4.43001e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.332e-05
        [j_node_and_user_rematch]: 9.84001e-06
        [meta_fg_expand]: 2.61999e-06
        [replace_old_param]: 9.31998e-06
        [inline_without_move]: 5.94e-06
        [renormalize]: 0.00103362
        [add_forward_monad_depend]: 7.73001e-06
        [auto_monad_grad]: 2.56e-06
        [auto_monad_eliminator]: 1.622e-05
        [cse]: 2.838e-05
        [replace_applicator]: 1.521e-05
    [Cycle 2]: 0.0003553, [27]
        [switch_simplify]: 5.89e-06
        [loop_unroll]: 5.44e-06
        [a_1]: 9.725e-05
        [with_stream_mark]: 1.422e-05
        [recompute_prepare]: 5.25999e-06
        [updatestate_depend_eliminate]: 2.68e-06
        [updatestate_assign_eliminate]: 2.38998e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [parameter_eliminate]: 1.04003e-06
        [specialize_transform]: 4.95001e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.30999e-06
        [meta_shard_fg_expand]: 2.17001e-06
        [get_grad_eliminate_]: 4.78001e-06
        [merge_forward]: 4.07e-06
        [cell_reuse_recompute_pass]: 2.70002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.539e-05
        [j_node_and_user_rematch]: 8.35001e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 8.78001e-06
        [inline_without_move]: 5.10999e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 1.02e-06
        [auto_monad_eliminator]: 6.26e-06
        [cse]: 1.207e-05
        [replace_applicator]: 5.81998e-06
[py_interpret_to_execute_after_opt_a]: 1.658e-05
[rewriter_after_opt_a]: 3.812e-05
[convert_after_rewriter]: 7.48e-06
[order_py_execute_after_rewriter]: 5.82999e-06
[mutable_eliminate]: 0.0008405
[jit_opt_b]: 6.603e-05, [1]
    [Cycle 1]: 5.686e-05, [2]
        [frontend_op_eliminate]: 2.476e-05
        [inline_after_opt_a]: 1.834e-05
[cconv]: 3.435e-05
[loop_unroll]: 0.00050198
[jit_opt_after_cconv]: 0.00016172, [1]
    [Cycle 1]: 0.00015467, [11]
        [c_1]: 2.358e-05
        [parameter_eliminate]: 4.63001e-06
        [updatestate_depend_eliminate]: 6.94001e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 3.02002e-06
        [cse]: 2.669e-05
        [call_graph_tuple_transform]: 2.205e-05
        [tuple_list_get_item_eliminator]: 5.47999e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 9.39996e-07
        [switch_simplify]: 5.76e-06
[remove_dup_value]: 1.515e-05
[partial_unused_args_eliminate]: 2.66e-06
[environ_conv]: 5.79999e-06
[add_recomputation]: 5.4e-05
[cse_after_recomputation]: 2.133e-05, [1]
    [Cycle 1]: 1.495e-05, [1]
        [cse]: 9.54999e-06
[auto_monad_reorder]: 1.545e-05
[get_jit_bprop_graph]: 2.49999e-06
[rewriter_after_jit_bprop_graph]: 8.75001e-06
[opt_after_jit_grad]: 0.00050581
[symbol_engine_optimizer]: 7.978e-05, [1]
    [Cycle 1]: 7.167e-05, [6]
        [build]: 4.84e-06
        [elim_shapecalc]: 8.45999e-06
        [elim_not_effective]: 1.332e-05
        [opt_reshape]: 6.76999e-06
        [fold_const_symbol]: 8.52e-06
        [renormalize]: 6.60017e-07
[validate]: 4.185e-05
Sums
    bootstrap                            :     0.000359s :  3.47%
    type_inference                       :     0.005337s : 51.68%
    event_method                         :     0.000018s :  0.17%
    auto_monad                           :     0.000069s :  0.66%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000045s :  0.43%
    rewriter_before_opt_a                :     0.000076s :  0.73%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000061s :  0.59%
    jit_opt_a.loop_unroll                :     0.000036s :  0.34%
    jit_opt_a.a_1                        :     0.000685s :  6.63%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.34%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000009s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.18%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.001034s : 10.01%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.22%
    jit_opt_a.cse                        :     0.000040s :  0.39%
    jit_opt_a.replace_applicator         :     0.000021s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.16%
    rewriter_after_opt_a                 :     0.000038s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000840s :  8.14%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.24%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.18%
    cconv                                :     0.000034s :  0.33%
    loop_unroll                          :     0.000502s :  4.86%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000027s :  0.26%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000015s :  0.15%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000054s :  0.52%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.08%
    opt_after_jit_grad                   :     0.000506s :  4.90%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000042s :  0.41%


TotalTime = 0.0228184, [33]
[bootstrap]: 0.00040046
[type_inference]: 0.0057655
[event_method]: 2.068e-05
[auto_monad]: 0.00010019
[graph_reusing]: 5.77999e-06
[pre_auto_parallel]: 2.63e-06
[py_interpret_to_execute]: 3.882e-05
[rewriter_before_opt_a]: 7.667e-05
[expand_dump_flag]: 3.71999e-06
[jit_opt_a]: 0.0063218, [2]
    [Cycle 1]: 0.0019847, [27]
        [switch_simplify]: 5.754e-05
        [loop_unroll]: 3.13e-05
        [a_1]: 0.0005512
        [with_stream_mark]: 1.859e-05
        [recompute_prepare]: 8.23999e-06
        [updatestate_depend_eliminate]: 4.02e-06
        [updatestate_assign_eliminate]: 3.06001e-06
        [updatestate_loads_eliminate]: 2.86999e-06
        [parameter_eliminate]: 2.12001e-06
        [specialize_transform]: 6.14001e-06
        [updatestate_useless_node_eliminater]: 5.48002e-06
        [accelerated_algorithm]: 5.67001e-06
        [meta_shard_fg_expand]: 2.02001e-06
        [get_grad_eliminate_]: 5.54e-06
        [merge_forward]: 3.95e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.641e-05
        [j_node_and_user_rematch]: 9.90002e-06
        [meta_fg_expand]: 2.53e-06
        [replace_old_param]: 1.032e-05
        [inline_without_move]: 5.75001e-06
        [renormalize]: 0.00096819
        [add_forward_monad_depend]: 7.44002e-06
        [auto_monad_grad]: 1.71e-06
        [auto_monad_eliminator]: 1.837e-05
        [cse]: 3.136e-05
        [replace_applicator]: 1.693e-05
    [Cycle 2]: 0.00037527, [27]
        [switch_simplify]: 6.31e-06
        [loop_unroll]: 4.90001e-06
        [a_1]: 9.607e-05
        [with_stream_mark]: 1.453e-05
        [recompute_prepare]: 5.61e-06
        [updatestate_depend_eliminate]: 3.8e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [parameter_eliminate]: 1.60999e-06
        [specialize_transform]: 5.90002e-06
        [updatestate_useless_node_eliminater]: 5.09998e-06
        [accelerated_algorithm]: 5.72001e-06
        [meta_shard_fg_expand]: 1.67001e-06
        [get_grad_eliminate_]: 4.75001e-06
        [merge_forward]: 4.12003e-06
        [cell_reuse_recompute_pass]: 3.46001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.841e-05
        [j_node_and_user_rematch]: 8.76002e-06
        [meta_fg_expand]: 2.06998e-06
        [replace_old_param]: 8.54998e-06
        [inline_without_move]: 4.82e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.22999e-06
        [auto_monad_grad]: 1.20999e-06
        [auto_monad_eliminator]: 7.2e-06
        [cse]: 1.556e-05
        [replace_applicator]: 5.91998e-06
[py_interpret_to_execute_after_opt_a]: 1.644e-05
[rewriter_after_opt_a]: 9.345e-05
[convert_after_rewriter]: 9.51e-06
[order_py_execute_after_rewriter]: 5.91e-06
[mutable_eliminate]: 0.00078948
[jit_opt_b]: 5.674e-05, [1]
    [Cycle 1]: 4.793e-05, [2]
        [frontend_op_eliminate]: 1.775e-05
        [inline_after_opt_a]: 1.663e-05
[cconv]: 3.349e-05
[loop_unroll]: 0.00045433
[jit_opt_after_cconv]: 0.00015833, [1]
    [Cycle 1]: 0.00015056, [11]
        [c_1]: 2.274e-05
        [parameter_eliminate]: 3.63999e-06
        [updatestate_depend_eliminate]: 8.79998e-06
        [updatestate_assign_eliminate]: 2.59999e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [cse]: 2.552e-05
        [call_graph_tuple_transform]: 2.054e-05
        [tuple_list_get_item_eliminator]: 5.26002e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 5.3001e-07
        [switch_simplify]: 5.47001e-06
[remove_dup_value]: 1.413e-05
[partial_unused_args_eliminate]: 2.49999e-06
[environ_conv]: 5.39e-06
[add_recomputation]: 9.877e-05
[cse_after_recomputation]: 2.498e-05, [1]
    [Cycle 1]: 1.846e-05, [1]
        [cse]: 1.111e-05
[auto_monad_reorder]: 1.661e-05
[get_jit_bprop_graph]: 2.28998e-06
[rewriter_after_jit_bprop_graph]: 7.59002e-06
[opt_after_jit_grad]: 0.00048036
[symbol_engine_optimizer]: 7.851e-05, [1]
    [Cycle 1]: 7.098e-05, [6]
        [build]: 4.37e-06
        [elim_shapecalc]: 7.92e-06
        [elim_not_effective]: 1.338e-05
        [opt_reshape]: 6.36e-06
        [fold_const_symbol]: 9.24998e-06
        [renormalize]: 8.00006e-07
[validate]: 7.132e-05
[backend_pass]: 1.45001e-06
[task_emit]: 0.00741542
[execute]: 9.41003e-06
Sums
    bootstrap                            :     0.000400s :  2.20%
    type_inference                       :     0.005765s : 31.73%
    event_method                         :     0.000021s :  0.11%
    auto_monad                           :     0.000100s :  0.55%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.21%
    rewriter_before_opt_a                :     0.000077s :  0.42%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.35%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000647s :  3.56%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000968s :  5.33%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.14%
    jit_opt_a.cse                        :     0.000047s :  0.26%
    jit_opt_a.replace_applicator         :     0.000023s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.09%
    rewriter_after_opt_a                 :     0.000093s :  0.51%
    convert_after_rewriter               :     0.000010s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000789s :  4.34%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000033s :  0.18%
    loop_unroll                          :     0.000454s :  2.50%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000099s :  0.54%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000480s :  2.64%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000071s :  0.39%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007415s : 40.81%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000211    23
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 3.09% :     0.000007s :      3: substitution.graph_param_transform
81.71% :     0.000172s :      4: substitution.inline
 1.86% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.78% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.79% :     0.000006s :      2: substitution.replace_old_param
 6.18% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005670     2
84.36% :     0.004783s :      1: type_inference.infer
15.64% :     0.000887s :      1: type_inference.specialize
------[replace.]   0.000071     6
67.55% :     0.000048s :      4: replace.inline
32.45% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000182     6
93.41% :     0.000170s :      4: match.inline
 6.59% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.28% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.15% :     0.000001s :     10: predicate.addn_zero_filter
 2.18% :     0.000003s :     10: predicate.arithmetic_simplify
 1.46% :     0.000002s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.41% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.15% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.45% :     0.000001s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.35% :     0.000002s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.48% :     0.000007s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 1.82% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.86% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 2.91% :     0.000004s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.44% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.84% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.27% :     0.000002s :     10: predicate.reshape_eliminate
 1.40% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.80% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.79% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.58% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.63% :     0.000009s :     53: predicate.switch_simplify
 1.09% :     0.000001s :     10: predicate.tile_eliminate
 1.40% :     0.000002s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.35% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.03% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000650    12
53.69% :     0.000349s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.41% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.90% :     0.000272s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024773    76
 0.41% :     0.000103s :      1: add_recomputation
 0.42% :     0.000105s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.72% :     0.000427s :      1: bootstrap
 0.15% :     0.000036s :      1: cconv
 0.05% :     0.000012s :      1: convert_after_rewriter
 0.11% :     0.000027s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
25.53% :     0.006326s :      1: jit_opt_a
 0.65% :     0.000162s :      1: jit_opt_after_cconv
 0.24% :     0.000060s :      1: jit_opt_b
 1.87% :     0.000462s :      1: loop_unroll
 3.24% :     0.000803s :      1: mutable_eliminate
 3.55% :     0.000878s :     26: opt.transform.jit_opt_a
 0.20% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.97% :     0.000489s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000042s :      1: py_interpret_to_execute
 0.08% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.57% :     0.000636s :      1: renormalize.infer
 1.30% :     0.000322s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.40% :     0.000100s :      1: rewriter_after_opt_a
 0.32% :     0.000080s :      1: rewriter_before_opt_a
 0.33% :     0.000081s :      1: symbol_engine_optimizer
30.00% :     0.007433s :      1: task_emit
23.32% :     0.005778s :      1: type_inference
 0.39% :     0.000096s :      1: validate


Time group info:
------[substitution.]   0.000247    23
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.43% :     0.000006s :      3: substitution.graph_param_transform
85.30% :     0.000211s :      4: substitution.inline
 1.44% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.22% :     0.000005s :      2: substitution.replace_old_param
 5.34% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005258     2
80.84% :     0.004251s :      1: type_inference.infer
19.16% :     0.001007s :      1: type_inference.specialize
------[replace.]   0.000079     6
65.16% :     0.000051s :      4: replace.inline
34.84% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000220     6
94.48% :     0.000208s :      4: match.inline
 5.52% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.26% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.56% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.77% :     0.000002s :     10: predicate.addn_zero_filter
 2.41% :     0.000003s :     10: predicate.arithmetic_simplify
 1.23% :     0.000002s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000002s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.67% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.09% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.32% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.25% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.45% :     0.000002s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.80% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.69% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.86% :     0.000002s :     15: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.25% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 1.93% :     0.000002s :      3: predicate.mutable_eliminate
 0.63% :     0.000001s :      3: predicate.opt_reshape
 1.90% :     0.000002s :     15: predicate.partial_eliminate
 1.23% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.88% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.66% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.06% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.33% :     0.000002s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.22% :     0.000003s :     16: predicate.switch_defer_inline
 1.99% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.78% :     0.000008s :     53: predicate.switch_simplify
 1.08% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.55% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.32% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.18% :     0.000003s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000740    12
47.48% :     0.000351s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.39% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.14% :     0.000341s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017092    72
 0.34% :     0.000058s :      1: add_recomputation
 0.42% :     0.000072s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 2.22% :     0.000379s :      1: bootstrap
 0.22% :     0.000037s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.13% :     0.000022s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
38.38% :     0.006560s :      1: jit_opt_a
 0.96% :     0.000165s :      1: jit_opt_after_cconv
 0.41% :     0.000069s :      1: jit_opt_b
 2.99% :     0.000511s :      1: loop_unroll
 4.99% :     0.000853s :      1: mutable_eliminate
 5.30% :     0.000906s :     26: opt.transform.jit_opt_a
 0.31% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000035s :      4: opt.transform.jit_opt_b
 0.08% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.01% :     0.000515s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000048s :      1: py_interpret_to_execute
 0.11% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000018s :      1: remove_dup_value
 3.43% :     0.000586s :      1: renormalize.infer
 2.55% :     0.000436s :      1: renormalize.specialize
 0.07% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000041s :      1: rewriter_after_opt_a
 0.46% :     0.000078s :      1: rewriter_before_opt_a
 0.48% :     0.000083s :      1: symbol_engine_optimizer
31.30% :     0.005350s :      1: type_inference


TotalTime = 0.0264405, [33]
[bootstrap]: 0.0003166
[type_inference]: 0.00652714
[event_method]: 2.069e-05
[auto_monad]: 7.021e-05
[graph_reusing]: 6.37001e-06
[pre_auto_parallel]: 2.58998e-06
[py_interpret_to_execute]: 4.159e-05
[rewriter_before_opt_a]: 7.824e-05
[expand_dump_flag]: 3.28e-06
[jit_opt_a]: 0.0062758, [2]
    [Cycle 1]: 0.00201057, [27]
        [switch_simplify]: 6.322e-05
        [loop_unroll]: 2.966e-05
        [a_1]: 0.00065417
        [with_stream_mark]: 2.83e-05
        [recompute_prepare]: 8.42e-06
        [updatestate_depend_eliminate]: 3.96001e-06
        [updatestate_assign_eliminate]: 2.86e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [parameter_eliminate]: 2.18002e-06
        [specialize_transform]: 5.82001e-06
        [updatestate_useless_node_eliminater]: 5.62001e-06
        [accelerated_algorithm]: 6.16998e-06
        [meta_shard_fg_expand]: 2.64999e-06
        [get_grad_eliminate_]: 5.59e-06
        [merge_forward]: 4.30999e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.617e-05
        [j_node_and_user_rematch]: 1.119e-05
        [meta_fg_expand]: 2.46998e-06
        [replace_old_param]: 9.40001e-06
        [inline_without_move]: 5.47001e-06
        [renormalize]: 0.00085895
        [add_forward_monad_depend]: 7.09001e-06
        [auto_monad_grad]: 3.08e-06
        [auto_monad_eliminator]: 1.793e-05
        [cse]: 3.338e-05
        [replace_applicator]: 1.661e-05
    [Cycle 2]: 0.00035605, [27]
        [switch_simplify]: 5.89e-06
        [loop_unroll]: 5.09e-06
        [a_1]: 9.879e-05
        [with_stream_mark]: 1.201e-05
        [recompute_prepare]: 6.16e-06
        [updatestate_depend_eliminate]: 3.50998e-06
        [updatestate_assign_eliminate]: 2.64001e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 5.17999e-06
        [updatestate_useless_node_eliminater]: 5.52001e-06
        [accelerated_algorithm]: 5.92001e-06
        [meta_shard_fg_expand]: 1.57999e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 3.00002e-06
        [cell_reuse_recompute_pass]: 2.91e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.508e-05
        [j_node_and_user_rematch]: 9.24e-06
        [meta_fg_expand]: 1.77001e-06
        [replace_old_param]: 7.92e-06
        [inline_without_move]: 4.67998e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.24998e-06
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 5.82001e-06
        [cse]: 1.291e-05
        [replace_applicator]: 5.70001e-06
[py_interpret_to_execute_after_opt_a]: 1.469e-05
[rewriter_after_opt_a]: 4.033e-05
[convert_after_rewriter]: 7.36001e-06
[order_py_execute_after_rewriter]: 5.19e-06
[mutable_eliminate]: 0.00078834
[jit_opt_b]: 5.68e-05, [1]
    [Cycle 1]: 4.762e-05, [2]
        [frontend_op_eliminate]: 1.766e-05
        [inline_after_opt_a]: 1.778e-05
[cconv]: 3.492e-05
[loop_unroll]: 0.00046765
[jit_opt_after_cconv]: 0.00016984, [1]
    [Cycle 1]: 0.00016332, [11]
        [c_1]: 2.254e-05
        [parameter_eliminate]: 4.83001e-06
        [updatestate_depend_eliminate]: 9.87001e-06
        [updatestate_assign_eliminate]: 2.69001e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [cse]: 3.184e-05
        [call_graph_tuple_transform]: 2.356e-05
        [tuple_list_get_item_eliminator]: 5.96998e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 6.99976e-07
        [switch_simplify]: 6.46e-06
[remove_dup_value]: 1.425e-05
[partial_unused_args_eliminate]: 2.12001e-06
[environ_conv]: 6.02001e-06
[add_recomputation]: 5.755e-05
[cse_after_recomputation]: 2.033e-05, [1]
    [Cycle 1]: 1.486e-05, [1]
        [cse]: 8.77e-06
[auto_monad_reorder]: 1.617e-05
[get_jit_bprop_graph]: 2.01e-06
[rewriter_after_jit_bprop_graph]: 9.39998e-06
[opt_after_jit_grad]: 0.00272134
[symbol_engine_optimizer]: 9.38e-05, [1]
    [Cycle 1]: 8.232e-05, [6]
        [build]: 6.26e-06
        [elim_shapecalc]: 8.90001e-06
        [elim_not_effective]: 1.96e-05
        [opt_reshape]: 6.86999e-06
        [fold_const_symbol]: 8.85999e-06
        [renormalize]: 9.70002e-07
[validate]: 5.146e-05
[backend_pass]: 1.62001e-06
[task_emit]: 0.00824565
[execute]: 9.81998e-06
Sums
    bootstrap                            :     0.000317s :  1.45%
    type_inference                       :     0.006527s : 29.93%
    event_method                         :     0.000021s :  0.09%
    auto_monad                           :     0.000070s :  0.32%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.19%
    rewriter_before_opt_a                :     0.000078s :  0.36%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000069s :  0.32%
    jit_opt_a.loop_unroll                :     0.000035s :  0.16%
    jit_opt_a.a_1                        :     0.000753s :  3.45%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000859s :  3.94%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.11%
    jit_opt_a.cse                        :     0.000046s :  0.21%
    jit_opt_a.replace_applicator         :     0.000022s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.07%
    rewriter_after_opt_a                 :     0.000040s :  0.18%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000788s :  3.61%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.08%
    cconv                                :     0.000035s :  0.16%
    loop_unroll                          :     0.000468s :  2.14%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000032s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000058s :  0.26%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.04%
    opt_after_jit_grad                   :     0.002721s : 12.48%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000051s :  0.24%
    backend_pass                         :     0.000002s :  0.01%
    task_emit                            :     0.008246s : 37.81%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000269    23
 0.94% :     0.000003s :      2: substitution.elim_not_effective
 0.59% :     0.000002s :      2: substitution.fold_const_symbol
 2.21% :     0.000006s :      3: substitution.graph_param_transform
85.47% :     0.000230s :      4: substitution.inline
 1.72% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.71% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.16% :     0.000006s :      2: substitution.replace_old_param
 5.20% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006450     2
84.87% :     0.005474s :      1: type_inference.infer
15.13% :     0.000976s :      1: type_inference.specialize
------[replace.]   0.000109     6
74.75% :     0.000081s :      4: replace.inline
25.25% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000240     6
94.70% :     0.000227s :      4: match.inline
 5.30% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.55% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 2.24% :     0.000003s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.47% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.01% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.95% :     0.000003s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000001s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.02% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000001s :     10: predicate.environ_get_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 6.09% :     0.000008s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.73% :     0.000002s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.25% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.51% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.71% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 2.79% :     0.000004s :      3: predicate.mutable_eliminate
 0.75% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.02% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.51% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.47% :     0.000001s :      3: predicate.reset_defer_inline
 1.20% :     0.000002s :     10: predicate.reshape_eliminate
 1.11% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.07% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     10: predicate.same_eliminate
 0.65% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.25% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.76% :     0.000011s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.20% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.08% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.59% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.25% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.99% :     0.000003s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.46% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000708    12
48.26% :     0.000342s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.63% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.10% :     0.000312s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028419    76
 0.22% :     0.000061s :      1: add_recomputation
 0.26% :     0.000074s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 1.19% :     0.000338s :      1: bootstrap
 0.13% :     0.000038s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.09% :     0.000026s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
22.10% :     0.006279s :      1: jit_opt_a
 0.61% :     0.000173s :      1: jit_opt_after_cconv
 0.21% :     0.000060s :      1: jit_opt_b
 1.68% :     0.000478s :      1: loop_unroll
 2.83% :     0.000804s :      1: mutable_eliminate
 3.47% :     0.000986s :     26: opt.transform.jit_opt_a
 0.19% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000028s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000039s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000041s :      4: opt.transform.symbol_engine_opt
 9.65% :     0.002742s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000045s :      1: py_interpret_to_execute
 0.06% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 1.64% :     0.000465s :      1: renormalize.infer
 1.35% :     0.000385s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000043s :      1: rewriter_after_opt_a
 0.29% :     0.000082s :      1: rewriter_before_opt_a
 0.34% :     0.000098s :      1: symbol_engine_optimizer
29.08% :     0.008265s :      1: task_emit
23.03% :     0.006546s :      1: type_inference
 0.28% :     0.000079s :      1: validate


TotalTime = 0.0202704, [33]
[bootstrap]: 0.00036405
[type_inference]: 0.0049574
[event_method]: 1.62e-05
[auto_monad]: 4.353e-05
[graph_reusing]: 3.7e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 3.753e-05
[rewriter_before_opt_a]: 6.241e-05
[expand_dump_flag]: 2.26e-06
[jit_opt_a]: 0.00619688, [2]
    [Cycle 1]: 0.0020907, [27]
        [switch_simplify]: 5.718e-05
        [loop_unroll]: 2.955e-05
        [a_1]: 0.00056722
        [with_stream_mark]: 1.861e-05
        [recompute_prepare]: 6.74999e-06
        [updatestate_depend_eliminate]: 2.99999e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.14998e-06
        [specialize_transform]: 6.36998e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 3.73999e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.239e-05
        [j_node_and_user_rematch]: 9.57001e-06
        [meta_fg_expand]: 1.94e-06
        [replace_old_param]: 8.74e-06
        [inline_without_move]: 5.40999e-06
        [renormalize]: 0.00087726
        [add_forward_monad_depend]: 5.34e-06
        [auto_monad_grad]: 2.44999e-06
        [auto_monad_eliminator]: 1.859e-05
        [cse]: 2.555e-05
        [replace_applicator]: 1.57e-05
    [Cycle 2]: 0.00034791, [27]
        [switch_simplify]: 6.23998e-06
        [loop_unroll]: 5.30001e-06
        [a_1]: 9.464e-05
        [with_stream_mark]: 1.525e-05
        [recompute_prepare]: 4.95001e-06
        [updatestate_depend_eliminate]: 2.58e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [parameter_eliminate]: 1.53002e-06
        [specialize_transform]: 4.81002e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 1.66e-06
        [get_grad_eliminate_]: 4.43001e-06
        [merge_forward]: 2.94001e-06
        [cell_reuse_recompute_pass]: 3.3e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.546e-05
        [j_node_and_user_rematch]: 7.99997e-06
        [meta_fg_expand]: 2.02999e-06
        [replace_old_param]: 7.84002e-06
        [inline_without_move]: 1.39e-05
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.99999e-06
        [auto_monad_grad]: 1.34e-06
        [auto_monad_eliminator]: 5.35001e-06
        [cse]: 1.169e-05
        [replace_applicator]: 5.02e-06
[py_interpret_to_execute_after_opt_a]: 1.399e-05
[rewriter_after_opt_a]: 3.562e-05
[convert_after_rewriter]: 6.81999e-06
[order_py_execute_after_rewriter]: 4.58999e-06
[mutable_eliminate]: 0.00075765
[jit_opt_b]: 5.249e-05, [1]
    [Cycle 1]: 4.39e-05, [2]
        [frontend_op_eliminate]: 1.772e-05
        [inline_after_opt_a]: 1.513e-05
[cconv]: 2.62e-05
[loop_unroll]: 0.00044569
[jit_opt_after_cconv]: 0.0001505, [1]
    [Cycle 1]: 0.00014327, [11]
        [c_1]: 2.246e-05
        [parameter_eliminate]: 3.26001e-06
        [updatestate_depend_eliminate]: 6.59999e-06
        [updatestate_assign_eliminate]: 2.57001e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [cse]: 2.231e-05
        [call_graph_tuple_transform]: 2.062e-05
        [tuple_list_get_item_eliminator]: 5.15001e-06
        [none_parameter_eliminate]: 1.30999e-06
        [renormalize]: 2.3999e-07
        [switch_simplify]: 5.25999e-06
[remove_dup_value]: 1.294e-05
[partial_unused_args_eliminate]: 1.76e-06
[environ_conv]: 4.90001e-06
[add_recomputation]: 4.742e-05
[cse_after_recomputation]: 2.103e-05, [1]
    [Cycle 1]: 1.533e-05, [1]
        [cse]: 8.98002e-06
[auto_monad_reorder]: 1.374e-05
[get_jit_bprop_graph]: 2.18998e-06
[rewriter_after_jit_bprop_graph]: 5.39e-06
[opt_after_jit_grad]: 0.00047169
[symbol_engine_optimizer]: 7.379e-05, [1]
    [Cycle 1]: 6.678e-05, [6]
        [build]: 4.18999e-06
        [elim_shapecalc]: 7.51999e-06
        [elim_not_effective]: 1.307e-05
        [opt_reshape]: 6.10002e-06
        [fold_const_symbol]: 8.94998e-06
        [renormalize]: 4.69998e-07
[validate]: 3.955e-05
[backend_pass]: 9.79984e-07
[task_emit]: 0.00618309
[execute]: 7.4e-06
Sums
    bootstrap                            :     0.000364s :  2.32%
    type_inference                       :     0.004957s : 31.62%
    event_method                         :     0.000016s :  0.10%
    auto_monad                           :     0.000044s :  0.28%
    graph_reusing                        :     0.000004s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.24%
    rewriter_before_opt_a                :     0.000062s :  0.40%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000063s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000662s :  4.22%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000019s :  0.12%
    jit_opt_a.renormalize                :     0.000877s :  5.60%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.15%
    jit_opt_a.cse                        :     0.000037s :  0.24%
    jit_opt_a.replace_applicator         :     0.000021s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.09%
    rewriter_after_opt_a                 :     0.000036s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000758s :  4.83%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000026s :  0.17%
    loop_unroll                          :     0.000446s :  2.84%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000047s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000472s :  3.01%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000040s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006183s : 39.43%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000232    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.53% :     0.000006s :      3: substitution.graph_param_transform
85.22% :     0.000198s :      4: substitution.inline
 1.33% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.30% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.02% :     0.000005s :      2: substitution.replace_old_param
 5.13% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004899     2
81.99% :     0.004017s :      1: type_inference.infer
18.01% :     0.000882s :      1: type_inference.specialize
------[replace.]   0.000077     6
67.83% :     0.000053s :      4: replace.inline
32.17% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000206     6
94.76% :     0.000196s :      4: match.inline
 5.24% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.39% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 2.50% :     0.000003s :     10: predicate.arithmetic_simplify
 1.13% :     0.000001s :     10: predicate.cast_eliminate
 0.57% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.24% :     0.000001s :     10: predicate.depend_value_elim
 1.23% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.50% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.32% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.53% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.41% :     0.000002s :     10: predicate.environ_get_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.15% :     0.000006s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.43% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 2.15% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000002s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.86% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.80% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.36% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.19% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.56% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.56% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.13% :     0.000008s :     53: predicate.switch_simplify
 1.37% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.41% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.28% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.12% :     0.000003s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000689    12
47.72% :     0.000329s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.44% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.84% :     0.000316s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022139    76
 0.23% :     0.000050s :      1: add_recomputation
 0.21% :     0.000047s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.70% :     0.000376s :      1: bootstrap
 0.13% :     0.000029s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.09% :     0.000020s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
28.01% :     0.006201s :      1: jit_opt_a
 0.69% :     0.000153s :      1: jit_opt_after_cconv
 0.25% :     0.000055s :      1: jit_opt_b
 2.05% :     0.000454s :      1: loop_unroll
 3.47% :     0.000768s :      1: mutable_eliminate
 4.02% :     0.000889s :     26: opt.transform.jit_opt_a
 0.22% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.17% :     0.000480s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000040s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 2.20% :     0.000486s :      1: renormalize.infer
 1.73% :     0.000383s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000039s :      1: rewriter_after_opt_a
 0.29% :     0.000065s :      1: rewriter_before_opt_a
 0.34% :     0.000076s :      1: symbol_engine_optimizer
28.00% :     0.006200s :      1: task_emit
22.45% :     0.004970s :      1: type_inference
 0.26% :     0.000058s :      1: validate


TotalTime = 0.0459517, [33]
[bootstrap]: 0.00040084
[type_inference]: 0.00511189
[event_method]: 1.814e-05
[auto_monad]: 6.073e-05
[graph_reusing]: 5.51e-06
[pre_auto_parallel]: 2.29001e-06
[py_interpret_to_execute]: 4.193e-05
[rewriter_before_opt_a]: 7.254e-05
[expand_dump_flag]: 3.24001e-06
[jit_opt_a]: 0.0185934, [2]
    [Cycle 1]: 0.00195672, [27]
        [switch_simplify]: 6.432e-05
        [loop_unroll]: 3.019e-05
        [a_1]: 0.0006049
        [with_stream_mark]: 2.343e-05
        [recompute_prepare]: 8.41002e-06
        [updatestate_depend_eliminate]: 4.13001e-06
        [updatestate_assign_eliminate]: 3.25e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [parameter_eliminate]: 2.44999e-06
        [specialize_transform]: 6.40002e-06
        [updatestate_useless_node_eliminater]: 5.32999e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 2.34001e-06
        [get_grad_eliminate_]: 5.99e-06
        [merge_forward]: 3.48e-06
        [cell_reuse_recompute_pass]: 1.39998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.415e-05
        [j_node_and_user_rematch]: 9.39e-06
        [meta_fg_expand]: 2.31998e-06
        [replace_old_param]: 9.05999e-06
        [inline_without_move]: 5.20999e-06
        [renormalize]: 0.00089354
        [add_forward_monad_depend]: 6.66e-06
        [auto_monad_grad]: 2.46e-06
        [auto_monad_eliminator]: 1.589e-05
        [cse]: 2.98e-05
        [replace_applicator]: 1.439e-05
    [Cycle 2]: 0.00042837, [27]
        [switch_simplify]: 5.89e-06
        [loop_unroll]: 5.27001e-06
        [a_1]: 0.00018213
        [with_stream_mark]: 1.431e-05
        [recompute_prepare]: 5.86e-06
        [updatestate_depend_eliminate]: 2.99999e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 5.95002e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.73002e-06
        [meta_shard_fg_expand]: 2.02001e-06
        [get_grad_eliminate_]: 4.84e-06
        [merge_forward]: 3.15002e-06
        [cell_reuse_recompute_pass]: 2.41998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.523e-05
        [j_node_and_user_rematch]: 8.31002e-06
        [meta_fg_expand]: 1.68997e-06
        [replace_old_param]: 7.63001e-06
        [inline_without_move]: 4.67e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.28002e-06
        [auto_monad_grad]: 1.14998e-06
        [auto_monad_eliminator]: 5.76e-06
        [cse]: 1.157e-05
        [replace_applicator]: 4.84998e-06
[py_interpret_to_execute_after_opt_a]: 1.46e-05
[rewriter_after_opt_a]: 3.885e-05
[convert_after_rewriter]: 6.93998e-06
[order_py_execute_after_rewriter]: 5.30999e-06
[mutable_eliminate]: 0.00073672
[jit_opt_b]: 5.436e-05, [1]
    [Cycle 1]: 4.615e-05, [2]
        [frontend_op_eliminate]: 1.892e-05
        [inline_after_opt_a]: 1.549e-05
[cconv]: 2.985e-05
[loop_unroll]: 0.00045478
[jit_opt_after_cconv]: 0.00017158, [1]
    [Cycle 1]: 0.00016369, [11]
        [c_1]: 2.217e-05
        [parameter_eliminate]: 4.17e-06
        [updatestate_depend_eliminate]: 7.95e-06
        [updatestate_assign_eliminate]: 3.27002e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 2.321e-05
        [call_graph_tuple_transform]: 2.155e-05
        [tuple_list_get_item_eliminator]: 6.04001e-06
        [none_parameter_eliminate]: 1.79998e-06
        [renormalize]: 7.89994e-07
        [switch_simplify]: 5.59e-06
[remove_dup_value]: 1.453e-05
[partial_unused_args_eliminate]: 2.79001e-06
[environ_conv]: 5.52001e-06
[add_recomputation]: 5.074e-05
[cse_after_recomputation]: 2.303e-05, [1]
    [Cycle 1]: 1.666e-05, [1]
        [cse]: 1.047e-05
[auto_monad_reorder]: 1.498e-05
[get_jit_bprop_graph]: 1.95001e-06
[rewriter_after_jit_bprop_graph]: 8.80999e-06
[opt_after_jit_grad]: 0.00047856
[symbol_engine_optimizer]: 7.631e-05, [1]
    [Cycle 1]: 6.974e-05, [6]
        [build]: 4.29002e-06
        [elim_shapecalc]: 7.7e-06
        [elim_not_effective]: 1.314e-05
        [opt_reshape]: 6.03998e-06
        [fold_const_symbol]: 9.61e-06
        [renormalize]: 4.39992e-07
[validate]: 4.251e-05
[backend_pass]: 9.49978e-07
[task_emit]: 0.0191749
[execute]: 8.36002e-06
Sums
    bootstrap                            :     0.000401s :  1.38%
    type_inference                       :     0.005112s : 17.58%
    event_method                         :     0.000018s :  0.06%
    auto_monad                           :     0.000061s :  0.21%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.14%
    rewriter_before_opt_a                :     0.000073s :  0.25%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000070s :  0.24%
    jit_opt_a.loop_unroll                :     0.000035s :  0.12%
    jit_opt_a.a_1                        :     0.000787s :  2.71%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000017s :  0.06%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.000894s :  3.07%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.07%
    jit_opt_a.cse                        :     0.000041s :  0.14%
    jit_opt_a.replace_applicator         :     0.000019s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.05%
    rewriter_after_opt_a                 :     0.000039s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000737s :  2.53%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.05%
    cconv                                :     0.000030s :  0.10%
    loop_unroll                          :     0.000455s :  1.56%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000015s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000051s :  0.17%
    cse_after_recomputation.cse          :     0.000010s :  0.04%
    auto_monad_reorder                   :     0.000015s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.03%
    opt_after_jit_grad                   :     0.000479s :  1.65%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000043s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.019175s : 65.93%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000246    23
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 2.73% :     0.000007s :      3: substitution.graph_param_transform
85.59% :     0.000211s :      4: substitution.inline
 1.40% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.06% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.76% :     0.000004s :      2: substitution.replace_old_param
 5.13% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005038     2
80.64% :     0.004063s :      1: type_inference.infer
19.36% :     0.000975s :      1: type_inference.specialize
------[replace.]   0.000074     6
67.34% :     0.000050s :      4: replace.inline
32.66% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000219     6
94.77% :     0.000208s :      4: match.inline
 5.23% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 2.45% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000002s :     10: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.50% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.82% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.37% :     0.000002s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.90% :     0.000008s :     22: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.81% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.17% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.62% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.36% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.52% :     0.000003s :     10: predicate.reduce_eliminate
 1.34% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.20% :     0.000002s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.48% :     0.000010s :     53: predicate.switch_simplify
 1.20% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.59% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.13% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.89% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000685    12
45.00% :     0.000308s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.28% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.72% :     0.000327s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.047963    76
 0.11% :     0.000054s :      1: add_recomputation
 0.13% :     0.000064s :      1: auto_monad
 0.04% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.87% :     0.000419s :      1: bootstrap
 0.07% :     0.000033s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.05% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.05% :     0.000022s :      1: event_method
 0.03% :     0.000013s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
38.78% :     0.018598s :      1: jit_opt_a
 0.37% :     0.000175s :      1: jit_opt_after_cconv
 0.12% :     0.000057s :      1: jit_opt_b
 0.97% :     0.000465s :      1: loop_unroll
 1.56% :     0.000747s :      1: mutable_eliminate
 2.11% :     0.001014s :     26: opt.transform.jit_opt_a
 0.11% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000028s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.02% :     0.000487s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.09% :     0.000045s :      1: py_interpret_to_execute
 0.04% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000017s :      1: remove_dup_value
 1.04% :     0.000497s :      1: renormalize.infer
 0.81% :     0.000387s :      1: renormalize.specialize
 0.02% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000042s :      1: rewriter_after_opt_a
 0.16% :     0.000076s :      1: rewriter_before_opt_a
 0.16% :     0.000079s :      1: symbol_engine_optimizer
40.02% :     0.019193s :      1: task_emit
10.69% :     0.005126s :      1: type_inference
 0.14% :     0.000065s :      1: validate


TotalTime = 0.0349298, [30]
[bootstrap]: 0.00049474
[type_inference]: 0.0115177
[event_method]: 2.114e-05
[auto_monad]: 7.345e-05
[graph_reusing]: 6.83e-06
[pre_auto_parallel]: 2.66e-06
[py_interpret_to_execute]: 0.0122394
[rewriter_before_opt_a]: 0.00014723
[expand_dump_flag]: 7.9e-06
[jit_opt_a]: 0.00736532, [2]
    [Cycle 1]: 0.0021344, [27]
        [switch_simplify]: 6.819e-05
        [loop_unroll]: 2.918e-05
        [a_1]: 0.00062282
        [with_stream_mark]: 2.567e-05
        [recompute_prepare]: 8.37e-06
        [updatestate_depend_eliminate]: 4.22e-06
        [updatestate_assign_eliminate]: 3.51001e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 1.97999e-06
        [specialize_transform]: 6.12001e-06
        [updatestate_useless_node_eliminater]: 5.37001e-06
        [accelerated_algorithm]: 5.44e-06
        [meta_shard_fg_expand]: 2.27999e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 3.73001e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.462e-05
        [j_node_and_user_rematch]: 1.012e-05
        [meta_fg_expand]: 3.14001e-06
        [replace_old_param]: 1.026e-05
        [inline_without_move]: 6.07999e-06
        [renormalize]: 0.00101729
        [add_forward_monad_depend]: 9.44e-06
        [auto_monad_grad]: 2.63998e-06
        [auto_monad_eliminator]: 1.918e-05
        [cse]: 3.105e-05
        [replace_applicator]: 1.852e-05
    [Cycle 2]: 0.00037875, [27]
        [switch_simplify]: 6.16e-06
        [loop_unroll]: 5.04e-06
        [a_1]: 0.00010444
        [with_stream_mark]: 1.409e-05
        [recompute_prepare]: 6.06e-06
        [updatestate_depend_eliminate]: 3.01001e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 2.13998e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 5.29998e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 1.66e-06
        [get_grad_eliminate_]: 4.92999e-06
        [merge_forward]: 3.26001e-06
        [cell_reuse_recompute_pass]: 2.64001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.712e-05
        [j_node_and_user_rematch]: 8.38001e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 1.008e-05
        [inline_without_move]: 5.71e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.66e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 7.5e-06
        [cse]: 1.251e-05
        [replace_applicator]: 5.79999e-06
[py_interpret_to_execute_after_opt_a]: 1.662e-05
[rewriter_after_opt_a]: 3.979e-05
[convert_after_rewriter]: 8.87e-06
[order_py_execute_after_rewriter]: 5.37001e-06
[mutable_eliminate]: 0.00087971
[jit_opt_b]: 6.004e-05, [1]
    [Cycle 1]: 4.996e-05, [2]
        [frontend_op_eliminate]: 2.019e-05
        [inline_after_opt_a]: 1.719e-05
[cconv]: 3.308e-05
[loop_unroll]: 0.00059051
[jit_opt_after_cconv]: 0.00017264, [1]
    [Cycle 1]: 0.00016393, [11]
        [c_1]: 2.255e-05
        [parameter_eliminate]: 4.00998e-06
        [updatestate_depend_eliminate]: 1.1e-05
        [updatestate_assign_eliminate]: 2.74001e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [cse]: 2.784e-05
        [call_graph_tuple_transform]: 2.526e-05
        [tuple_list_get_item_eliminator]: 5.89e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 5.63002e-06
[remove_dup_value]: 1.641e-05
[partial_unused_args_eliminate]: 2.66999e-06
[environ_conv]: 6.69001e-06
[add_recomputation]: 5.852e-05
[cse_after_recomputation]: 2.503e-05, [1]
    [Cycle 1]: 1.784e-05, [1]
        [cse]: 9.93998e-06
[auto_monad_reorder]: 1.721e-05
[get_jit_bprop_graph]: 5.603e-05
[rewriter_after_jit_bprop_graph]: 9.30001e-06
[opt_after_jit_grad]: 0.00064145
[symbol_engine_optimizer]: 8.633e-05, [1]
    [Cycle 1]: 7.685e-05, [6]
        [build]: 5.94e-06
        [elim_shapecalc]: 8.38001e-06
        [elim_not_effective]: 1.657e-05
        [opt_reshape]: 6.71e-06
        [fold_const_symbol]: 9.17001e-06
        [renormalize]: 8.2e-07
[validate]: 4.329e-05
Sums
    bootstrap                            :     0.000495s :  1.69%
    type_inference                       :     0.011518s : 39.28%
    event_method                         :     0.000021s :  0.07%
    auto_monad                           :     0.000073s :  0.25%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.012239s : 41.74%
    rewriter_before_opt_a                :     0.000147s :  0.50%
    expand_dump_flag                     :     0.000008s :  0.03%
    jit_opt_a.switch_simplify            :     0.000074s :  0.25%
    jit_opt_a.loop_unroll                :     0.000034s :  0.12%
    jit_opt_a.a_1                        :     0.000727s :  2.48%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.07%
    jit_opt_a.inline_without_move        :     0.000012s :  0.04%
    jit_opt_a.renormalize                :     0.001017s :  3.47%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.09%
    jit_opt_a.cse                        :     0.000044s :  0.15%
    jit_opt_a.replace_applicator         :     0.000024s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.06%
    rewriter_after_opt_a                 :     0.000040s :  0.14%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000880s :  3.00%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.06%
    cconv                                :     0.000033s :  0.11%
    loop_unroll                          :     0.000591s :  2.01%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000059s :  0.20%
    cse_after_recomputation.cse          :     0.000010s :  0.03%
    auto_monad_reorder                   :     0.000017s :  0.06%
    get_jit_bprop_graph                  :     0.000056s :  0.19%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.03%
    opt_after_jit_grad                   :     0.000641s :  2.19%
    symbol_engine_optimizer.build        :     0.000006s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.15%


Time group info:
------[substitution.]   0.000279    23
 1.02% :     0.000003s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 2.51% :     0.000007s :      3: substitution.graph_param_transform
85.91% :     0.000240s :      4: substitution.inline
 1.30% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.84% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.18% :     0.000006s :      2: substitution.replace_old_param
 4.76% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011426     2
90.24% :     0.010311s :      1: type_inference.infer
 9.76% :     0.001115s :      1: type_inference.specialize
------[replace.]   0.000080     6
68.38% :     0.000055s :      4: replace.inline
31.62% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000249     6
95.22% :     0.000237s :      4: match.inline
 4.78% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.65% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.20% :     0.000002s :     10: predicate.addn_zero_filter
 2.15% :     0.000003s :     10: predicate.arithmetic_simplify
 1.19% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.45% :     0.000002s :     10: predicate.depend_value_elim
 1.43% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.04% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.71% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.34% :     0.000002s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.77% :     0.000001s :      6: predicate.get_grad_eliminate
 0.45% :     0.000001s :      3: predicate.graph_param_transform
 5.98% :     0.000008s :     22: predicate.inline
 1.10% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000002s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.89% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.69% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.79% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 2.76% :     0.000004s :      3: predicate.mutable_eliminate
 0.60% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.25% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.47% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.63% :     0.000002s :     10: predicate.reshape_eliminate
 1.05% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.13% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.36% :     0.000002s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.64% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.09% :     0.000003s :     16: predicate.switch_defer_inline
 1.81% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.89% :     0.000009s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.21% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.15% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.51% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.44% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.85% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000872    12
50.17% :     0.000438s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.46% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.37% :     0.000378s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.036968    72
 0.17% :     0.000062s :      1: add_recomputation
 0.21% :     0.000077s :      1: auto_monad
 0.05% :     0.000020s :      1: auto_monad_reorder
 1.40% :     0.000518s :      1: bootstrap
 0.10% :     0.000036s :      1: cconv
 0.04% :     0.000014s :      1: convert_after_rewriter
 0.07% :     0.000028s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.07% :     0.000026s :      1: event_method
 0.03% :     0.000010s :      1: expand_dump_flag
 0.17% :     0.000062s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
19.93% :     0.007369s :      1: jit_opt_a
 0.47% :     0.000175s :      1: jit_opt_after_cconv
 0.17% :     0.000063s :      1: jit_opt_b
 1.62% :     0.000600s :      1: loop_unroll
 2.43% :     0.000897s :      1: mutable_eliminate
 2.62% :     0.000968s :     26: opt.transform.jit_opt_a
 0.15% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000029s :      4: opt.transform.jit_opt_b
 0.05% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.77% :     0.000655s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
33.20% :     0.012272s :      1: py_interpret_to_execute
 0.05% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 1.45% :     0.000538s :      1: renormalize.infer
 1.27% :     0.000468s :      1: renormalize.specialize
 0.03% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000043s :      1: rewriter_after_opt_a
 0.42% :     0.000157s :      1: rewriter_before_opt_a
 0.24% :     0.000089s :      1: symbol_engine_optimizer
31.22% :     0.011541s :      1: type_inference


TotalTime = 0.0245901, [33]
[bootstrap]: 0.00032707
[type_inference]: 0.0055222
[event_method]: 1.864e-05
[auto_monad]: 7.011e-05
[graph_reusing]: 5.30999e-06
[pre_auto_parallel]: 2.68003e-06
[py_interpret_to_execute]: 3.819e-05
[rewriter_before_opt_a]: 7.506e-05
[expand_dump_flag]: 3.36999e-06
[jit_opt_a]: 0.00604078, [2]
    [Cycle 1]: 0.00184646, [27]
        [switch_simplify]: 6e-05
        [loop_unroll]: 2.997e-05
        [a_1]: 0.00058156
        [with_stream_mark]: 2.03e-05
        [recompute_prepare]: 8.13999e-06
        [updatestate_depend_eliminate]: 3.6e-06
        [updatestate_assign_eliminate]: 2.69001e-06
        [updatestate_loads_eliminate]: 2.56998e-06
        [parameter_eliminate]: 2.09e-06
        [specialize_transform]: 6.14999e-06
        [updatestate_useless_node_eliminater]: 5.40001e-06
        [accelerated_algorithm]: 5.64998e-06
        [meta_shard_fg_expand]: 2.25002e-06
        [get_grad_eliminate_]: 5.31002e-06
        [merge_forward]: 3.64002e-06
        [cell_reuse_recompute_pass]: 1.05999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.514e-05
        [j_node_and_user_rematch]: 9.69e-06
        [meta_fg_expand]: 2.78e-06
        [replace_old_param]: 9.58997e-06
        [inline_without_move]: 5.00999e-06
        [renormalize]: 0.00081493
        [add_forward_monad_depend]: 5.86e-06
        [auto_monad_grad]: 2.96999e-06
        [auto_monad_eliminator]: 1.475e-05
        [cse]: 2.639e-05
        [replace_applicator]: 1.551e-05
    [Cycle 2]: 0.00033667, [27]
        [switch_simplify]: 6.28e-06
        [loop_unroll]: 6.07001e-06
        [a_1]: 9.263e-05
        [with_stream_mark]: 1.187e-05
        [recompute_prepare]: 5.40001e-06
        [updatestate_depend_eliminate]: 2.82002e-06
        [updatestate_assign_eliminate]: 1.97999e-06
        [updatestate_loads_eliminate]: 2.13002e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 5.27999e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.12e-06
        [meta_shard_fg_expand]: 1.72999e-06
        [get_grad_eliminate_]: 4.45e-06
        [merge_forward]: 3.07002e-06
        [cell_reuse_recompute_pass]: 2.30002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.515e-05
        [j_node_and_user_rematch]: 8.71002e-06
        [meta_fg_expand]: 1.78997e-06
        [replace_old_param]: 1.073e-05
        [inline_without_move]: 4.81002e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 6.50005e-07
        [auto_monad_eliminator]: 5.18002e-06
        [cse]: 9.94001e-06
        [replace_applicator]: 5.09e-06
[py_interpret_to_execute_after_opt_a]: 1.379e-05
[rewriter_after_opt_a]: 3.802e-05
[convert_after_rewriter]: 7.68999e-06
[order_py_execute_after_rewriter]: 6.66999e-06
[mutable_eliminate]: 0.00070789
[jit_opt_b]: 5.794e-05, [1]
    [Cycle 1]: 4.936e-05, [2]
        [frontend_op_eliminate]: 1.768e-05
        [inline_after_opt_a]: 1.771e-05
[cconv]: 2.961e-05
[loop_unroll]: 0.00258071
[jit_opt_after_cconv]: 0.00019225, [1]
    [Cycle 1]: 0.00018111, [11]
        [c_1]: 2.485e-05
        [parameter_eliminate]: 7.03998e-06
        [updatestate_depend_eliminate]: 1.271e-05
        [updatestate_assign_eliminate]: 3.3e-06
        [updatestate_loads_eliminate]: 2.78e-06
        [cse]: 3.354e-05
        [call_graph_tuple_transform]: 2.747e-05
        [tuple_list_get_item_eliminator]: 6.76e-06
        [none_parameter_eliminate]: 1.73002e-06
        [renormalize]: 7.59988e-07
        [switch_simplify]: 5.76e-06
[remove_dup_value]: 1.687e-05
[partial_unused_args_eliminate]: 2.39999e-06
[environ_conv]: 6.91999e-06
[add_recomputation]: 5.75e-05
[cse_after_recomputation]: 2.394e-05, [1]
    [Cycle 1]: 1.747e-05, [1]
        [cse]: 1.099e-05
[auto_monad_reorder]: 1.686e-05
[get_jit_bprop_graph]: 2.07999e-06
[rewriter_after_jit_bprop_graph]: 8.57e-06
[opt_after_jit_grad]: 0.00057348
[symbol_engine_optimizer]: 8.029e-05, [1]
    [Cycle 1]: 7.377e-05, [6]
        [build]: 5.76e-06
        [elim_shapecalc]: 7.65e-06
        [elim_not_effective]: 1.505e-05
        [opt_reshape]: 6.41998e-06
        [fold_const_symbol]: 9.97001e-06
        [renormalize]: 5.19998e-07
[validate]: 4.289e-05
[backend_pass]: 1.00001e-06
[task_emit]: 0.00776305
[execute]: 1.024e-05
Sums
    bootstrap                            :     0.000327s :  1.63%
    type_inference                       :     0.005522s : 27.54%
    event_method                         :     0.000019s :  0.09%
    auto_monad                           :     0.000070s :  0.35%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000038s :  0.19%
    rewriter_before_opt_a                :     0.000075s :  0.37%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.33%
    jit_opt_a.loop_unroll                :     0.000036s :  0.18%
    jit_opt_a.a_1                        :     0.000674s :  3.36%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000815s :  4.06%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.10%
    jit_opt_a.cse                        :     0.000036s :  0.18%
    jit_opt_a.replace_applicator         :     0.000021s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.19%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000007s :  0.03%
    mutable_eliminate                    :     0.000708s :  3.53%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.09%
    cconv                                :     0.000030s :  0.15%
    loop_unroll                          :     0.002581s : 12.87%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000013s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000034s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000058s :  0.29%
    cse_after_recomputation.cse          :     0.000011s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.04%
    opt_after_jit_grad                   :     0.000573s :  2.86%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.21%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007763s : 38.72%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000239    23
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.83% :     0.000002s :      2: substitution.fold_const_symbol
 2.78% :     0.000007s :      3: substitution.graph_param_transform
84.22% :     0.000201s :      4: substitution.inline
 1.60% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.08% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.15% :     0.000005s :      2: substitution.replace_old_param
 5.51% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005443     2
81.46% :     0.004434s :      1: type_inference.infer
18.54% :     0.001009s :      1: type_inference.specialize
------[replace.]   0.000071     6
64.79% :     0.000046s :      4: replace.inline
35.21% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000210     6
94.28% :     0.000198s :      4: match.inline
 5.72% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.71% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 2.04% :     0.000003s :     10: predicate.arithmetic_simplify
 1.29% :     0.000002s :     10: predicate.cast_eliminate
 0.73% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 0.97% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.05% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      3: predicate.elim_not_effective
 0.46% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.05% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.46% :     0.000001s :      3: predicate.graph_param_transform
 6.80% :     0.000009s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.30% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 3.33% :     0.000004s :      3: predicate.loop_unroll_after_grad
 3.84% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.83% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 2.25% :     0.000003s :      3: predicate.mutable_eliminate
 0.37% :     0.000000s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.36% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.63% :     0.000002s :     10: predicate.reshape_eliminate
 1.09% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.15% :     0.000001s :      6: predicate.specialize_transform
 1.74% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.61% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 1.87% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.85% :     0.000010s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.56% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.06% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.82% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.30% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000688    12
48.15% :     0.000331s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.98% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.87% :     0.000309s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026439    76
 0.23% :     0.000060s :      1: add_recomputation
 0.28% :     0.000074s :      1: auto_monad
 0.07% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.30% :     0.000344s :      1: bootstrap
 0.12% :     0.000033s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000027s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.09% :     0.000023s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
22.86% :     0.006045s :      1: jit_opt_a
 0.74% :     0.000196s :      1: jit_opt_after_cconv
 0.23% :     0.000061s :      1: jit_opt_b
 9.83% :     0.002598s :      1: loop_unroll
 2.72% :     0.000719s :      1: mutable_eliminate
 3.41% :     0.000902s :     26: opt.transform.jit_opt_a
 0.23% :     0.000061s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000028s :      4: opt.transform.jit_opt_b
 0.10% :     0.000026s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000586s :      1: opt_after_jit_grad
 0.03% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.15% :     0.000041s :      1: py_interpret_to_execute
 0.06% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 1.70% :     0.000451s :      1: renormalize.infer
 1.34% :     0.000355s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000041s :      1: rewriter_after_opt_a
 0.29% :     0.000078s :      1: rewriter_before_opt_a
 0.31% :     0.000083s :      1: symbol_engine_optimizer
29.44% :     0.007785s :      1: task_emit
20.95% :     0.005539s :      1: type_inference
 0.26% :     0.000068s :      1: validate


TotalTime = 0.0308795, [33]
[bootstrap]: 0.00035072
[type_inference]: 0.0091287
[event_method]: 2.374e-05
[auto_monad]: 7.813e-05
[graph_reusing]: 6.61e-06
[pre_auto_parallel]: 2.58e-06
[py_interpret_to_execute]: 4.659e-05
[rewriter_before_opt_a]: 8.213e-05
[expand_dump_flag]: 3.52997e-06
[jit_opt_a]: 0.00793652, [2]
    [Cycle 1]: 0.00209094, [27]
        [switch_simplify]: 6.8e-05
        [loop_unroll]: 3.132e-05
        [a_1]: 0.00065529
        [with_stream_mark]: 2.886e-05
        [recompute_prepare]: 9.49e-06
        [updatestate_depend_eliminate]: 4.45999e-06
        [updatestate_assign_eliminate]: 3.18998e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 2.83e-06
        [specialize_transform]: 6.36e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 6.24999e-06
        [meta_shard_fg_expand]: 2.23002e-06
        [get_grad_eliminate_]: 5.57999e-06
        [merge_forward]: 4.65999e-06
        [cell_reuse_recompute_pass]: 1.79e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.62e-05
        [j_node_and_user_rematch]: 1.14e-05
        [meta_fg_expand]: 2.44001e-06
        [replace_old_param]: 1.049e-05
        [inline_without_move]: 5.44998e-06
        [renormalize]: 0.00091068
        [add_forward_monad_depend]: 7.76001e-06
        [auto_monad_grad]: 2.74001e-06
        [auto_monad_eliminator]: 1.965e-05
        [cse]: 3.763e-05
        [replace_applicator]: 1.948e-05
    [Cycle 2]: 0.00034576, [27]
        [switch_simplify]: 6.18998e-06
        [loop_unroll]: 4.98001e-06
        [a_1]: 9.453e-05
        [with_stream_mark]: 1.492e-05
        [recompute_prepare]: 5.45001e-06
        [updatestate_depend_eliminate]: 3.04001e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.20002e-06
        [parameter_eliminate]: 1.24998e-06
        [specialize_transform]: 5.19998e-06
        [updatestate_useless_node_eliminater]: 5.07999e-06
        [accelerated_algorithm]: 5.84e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 4.77998e-06
        [merge_forward]: 3.01001e-06
        [cell_reuse_recompute_pass]: 2.32001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.727e-05
        [j_node_and_user_rematch]: 8.40001e-06
        [meta_fg_expand]: 1.64998e-06
        [replace_old_param]: 8.32998e-06
        [inline_without_move]: 5.04e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.37999e-06
        [auto_monad_grad]: 1.22999e-06
        [auto_monad_eliminator]: 6.81001e-06
        [cse]: 1.25e-05
        [replace_applicator]: 5.19e-06
[py_interpret_to_execute_after_opt_a]: 1.635e-05
[rewriter_after_opt_a]: 4.316e-05
[convert_after_rewriter]: 8.07e-06
[order_py_execute_after_rewriter]: 4.97e-06
[mutable_eliminate]: 0.00076851
[jit_opt_b]: 5.961e-05, [1]
    [Cycle 1]: 4.966e-05, [2]
        [frontend_op_eliminate]: 2.056e-05
        [inline_after_opt_a]: 1.693e-05
[cconv]: 3.31e-05
[loop_unroll]: 0.00050559
[jit_opt_after_cconv]: 0.00016871, [1]
    [Cycle 1]: 0.00016186, [11]
        [c_1]: 2.197e-05
        [parameter_eliminate]: 5.20999e-06
        [updatestate_depend_eliminate]: 1.025e-05
        [updatestate_assign_eliminate]: 3.14001e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [cse]: 2.905e-05
        [call_graph_tuple_transform]: 2.355e-05
        [tuple_list_get_item_eliminator]: 5.51002e-06
        [none_parameter_eliminate]: 1.92999e-06
        [renormalize]: 8.70001e-07
        [switch_simplify]: 5.64998e-06
[remove_dup_value]: 1.519e-05
[partial_unused_args_eliminate]: 2.88e-06
[environ_conv]: 6.04001e-06
[add_recomputation]: 5.51e-05
[cse_after_recomputation]: 2.199e-05, [1]
    [Cycle 1]: 1.612e-05, [1]
        [cse]: 1.041e-05
[auto_monad_reorder]: 1.734e-05
[get_jit_bprop_graph]: 2.41e-06
[rewriter_after_jit_bprop_graph]: 9.62999e-06
[opt_after_jit_grad]: 0.00051006
[symbol_engine_optimizer]: 7.726e-05, [1]
    [Cycle 1]: 6.991e-05, [6]
        [build]: 4.22998e-06
        [elim_shapecalc]: 8.64e-06
        [elim_not_effective]: 1.408e-05
        [opt_reshape]: 5.27001e-06
        [fold_const_symbol]: 9.41998e-06
        [renormalize]: 5.09986e-07
[validate]: 4.598e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.0105238
[execute]: 8.95999e-06
Sums
    bootstrap                            :     0.000351s :  1.42%
    type_inference                       :     0.009129s : 37.09%
    event_method                         :     0.000024s :  0.10%
    auto_monad                           :     0.000078s :  0.32%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000047s :  0.19%
    rewriter_before_opt_a                :     0.000082s :  0.33%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000074s :  0.30%
    jit_opt_a.loop_unroll                :     0.000036s :  0.15%
    jit_opt_a.a_1                        :     0.000750s :  3.05%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000008s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.04%
    jit_opt_a.renormalize                :     0.000911s :  3.70%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.11%
    jit_opt_a.cse                        :     0.000050s :  0.20%
    jit_opt_a.replace_applicator         :     0.000025s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.07%
    rewriter_after_opt_a                 :     0.000043s :  0.18%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000769s :  3.12%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.07%
    cconv                                :     0.000033s :  0.13%
    loop_unroll                          :     0.000506s :  2.05%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000015s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000055s :  0.22%
    cse_after_recomputation.cse          :     0.000010s :  0.04%
    auto_monad_reorder                   :     0.000017s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.04%
    opt_after_jit_grad                   :     0.000510s :  2.07%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.19%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010524s : 42.76%
    execute                              :     0.000009s :  0.04%


Time group info:
------[substitution.]   0.000292    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000002s :      2: substitution.fold_const_symbol
 2.06% :     0.000006s :      3: substitution.graph_param_transform
86.48% :     0.000253s :      4: substitution.inline
 1.39% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.69% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.20% :     0.000006s :      2: substitution.replace_old_param
 4.92% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009037     2
88.31% :     0.007980s :      1: type_inference.infer
11.69% :     0.001057s :      1: type_inference.specialize
------[replace.]   0.000085     6
68.45% :     0.000058s :      4: replace.inline
31.55% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000262     6
94.96% :     0.000249s :      4: match.inline
 5.04% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   691
 1.29% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 1.89% :     0.000003s :     10: predicate.arithmetic_simplify
 1.21% :     0.000002s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.12% :     0.000001s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.45% :     0.000001s :      3: predicate.graph_param_transform
 6.35% :     0.000008s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :      6: predicate.less_batch_normalization
 1.26% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.58% :     0.000002s :     15: predicate.load_eliminater
 2.33% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.89% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.62% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.93% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 2.85% :     0.000004s :      3: predicate.mutable_eliminate
 0.37% :     0.000000s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.02% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.78% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.65% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.10% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.66% :     0.000004s :     16: predicate.switch_defer_inline
 1.83% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.92% :     0.000010s :     53: predicate.switch_simplify
 1.10% :     0.000001s :     10: predicate.tile_eliminate
 1.02% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.46% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.99% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.32% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.43% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.84% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000823    12
52.41% :     0.000432s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.29% :     0.000060s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.30% :     0.000332s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.032899    76
 0.18% :     0.000058s :      1: add_recomputation
 0.25% :     0.000083s :      1: auto_monad
 0.06% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.13% :     0.000372s :      1: bootstrap
 0.29% :     0.000097s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.07% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000029s :      1: event_method
 0.04% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
24.14% :     0.007941s :      1: jit_opt_a
 0.52% :     0.000171s :      1: jit_opt_after_cconv
 0.19% :     0.000063s :      1: jit_opt_b
 1.57% :     0.000516s :      1: loop_unroll
 2.38% :     0.000784s :      1: mutable_eliminate
 3.02% :     0.000993s :     26: opt.transform.jit_opt_a
 0.16% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000030s :      4: opt.transform.jit_opt_b
 0.05% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.58% :     0.000519s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000050s :      1: py_interpret_to_execute
 0.06% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 1.58% :     0.000519s :      1: renormalize.infer
 1.16% :     0.000383s :      1: renormalize.specialize
 0.04% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000046s :      1: rewriter_after_opt_a
 0.26% :     0.000086s :      1: rewriter_before_opt_a
 0.24% :     0.000080s :      1: symbol_engine_optimizer
32.05% :     0.010545s :      1: task_emit
27.83% :     0.009155s :      1: type_inference
 0.21% :     0.000070s :      1: validate


TotalTime = 0.0230343, [33]
[bootstrap]: 0.00042303
[type_inference]: 0.00599758
[event_method]: 1.976e-05
[auto_monad]: 6.785e-05
[graph_reusing]: 6.61999e-06
[pre_auto_parallel]: 3.09999e-06
[py_interpret_to_execute]: 4.473e-05
[rewriter_before_opt_a]: 7.904e-05
[expand_dump_flag]: 3.56999e-06
[jit_opt_a]: 0.00692637, [2]
    [Cycle 1]: 0.00212993, [27]
        [switch_simplify]: 6.627e-05
        [loop_unroll]: 2.875e-05
        [a_1]: 0.0006196
        [with_stream_mark]: 2.669e-05
        [recompute_prepare]: 9.30001e-06
        [updatestate_depend_eliminate]: 4e-06
        [updatestate_assign_eliminate]: 2.84001e-06
        [updatestate_loads_eliminate]: 2.52001e-06
        [parameter_eliminate]: 1.76003e-06
        [specialize_transform]: 6.45002e-06
        [updatestate_useless_node_eliminater]: 5.18002e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 5.29998e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.501e-05
        [j_node_and_user_rematch]: 9.42001e-06
        [meta_fg_expand]: 2.27999e-06
        [replace_old_param]: 9.93002e-06
        [inline_without_move]: 5.73002e-06
        [renormalize]: 0.00102026
        [add_forward_monad_depend]: 6.78998e-06
        [auto_monad_grad]: 2.46998e-06
        [auto_monad_eliminator]: 1.719e-05
        [cse]: 3.181e-05
        [replace_applicator]: 1.694e-05
    [Cycle 2]: 0.00034, [27]
        [switch_simplify]: 6.98e-06
        [loop_unroll]: 5.32001e-06
        [a_1]: 9.626e-05
        [with_stream_mark]: 1.146e-05
        [recompute_prepare]: 6.24001e-06
        [updatestate_depend_eliminate]: 2.79999e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 5.20999e-06
        [updatestate_useless_node_eliminater]: 4.80999e-06
        [accelerated_algorithm]: 5.40001e-06
        [meta_shard_fg_expand]: 2.35002e-06
        [get_grad_eliminate_]: 4.4e-06
        [merge_forward]: 2.81e-06
        [cell_reuse_recompute_pass]: 3.3e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.527e-05
        [j_node_and_user_rematch]: 8.57e-06
        [meta_fg_expand]: 1.95001e-06
        [replace_old_param]: 9.22999e-06
        [inline_without_move]: 4.91997e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.33002e-06
        [auto_monad_grad]: 6.39993e-07
        [auto_monad_eliminator]: 6.58e-06
        [cse]: 1.107e-05
        [replace_applicator]: 5.22e-06
[py_interpret_to_execute_after_opt_a]: 1.405e-05
[rewriter_after_opt_a]: 4.067e-05
[convert_after_rewriter]: 7.28999e-06
[order_py_execute_after_rewriter]: 4.84e-06
[mutable_eliminate]: 0.00073346
[jit_opt_b]: 5.399e-05, [1]
    [Cycle 1]: 4.622e-05, [2]
        [frontend_op_eliminate]: 1.761e-05
        [inline_after_opt_a]: 1.627e-05
[cconv]: 2.871e-05
[loop_unroll]: 0.00044525
[jit_opt_after_cconv]: 0.00014993, [1]
    [Cycle 1]: 0.00014295, [11]
        [c_1]: 2.226e-05
        [parameter_eliminate]: 3.13e-06
        [updatestate_depend_eliminate]: 6.58998e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 2.346e-05
        [call_graph_tuple_transform]: 1.997e-05
        [tuple_list_get_item_eliminator]: 5.07e-06
        [none_parameter_eliminate]: 1.84e-06
        [renormalize]: 2.60014e-07
        [switch_simplify]: 5.24e-06
[remove_dup_value]: 1.544e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 5.46e-06
[add_recomputation]: 5.471e-05
[cse_after_recomputation]: 2.192e-05, [1]
    [Cycle 1]: 1.565e-05, [1]
        [cse]: 9.31002e-06
[auto_monad_reorder]: 1.523e-05
[get_jit_bprop_graph]: 2.19001e-06
[rewriter_after_jit_bprop_graph]: 7.05998e-06
[opt_after_jit_grad]: 0.00049647
[symbol_engine_optimizer]: 7.714e-05, [1]
    [Cycle 1]: 7.036e-05, [6]
        [build]: 5.35001e-06
        [elim_shapecalc]: 7.73001e-06
        [elim_not_effective]: 1.359e-05
        [opt_reshape]: 5.77999e-06
        [fold_const_symbol]: 9.36998e-06
        [renormalize]: 9.39996e-07
[validate]: 4.29e-05
[backend_pass]: 1.04e-06
[task_emit]: 0.00695725
[execute]: 1.016e-05
Sums
    bootstrap                            :     0.000423s :  2.37%
    type_inference                       :     0.005998s : 33.57%
    event_method                         :     0.000020s :  0.11%
    auto_monad                           :     0.000068s :  0.38%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000045s :  0.25%
    rewriter_before_opt_a                :     0.000079s :  0.44%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000073s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.19%
    jit_opt_a.a_1                        :     0.000716s :  4.01%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001020s :  5.71%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.13%
    jit_opt_a.cse                        :     0.000043s :  0.24%
    jit_opt_a.replace_applicator         :     0.000022s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000041s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000733s :  4.11%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000029s :  0.16%
    loop_unroll                          :     0.000445s :  2.49%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000055s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000496s :  2.78%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006957s : 38.94%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000272    23
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 2.03% :     0.000006s :      3: substitution.graph_param_transform
85.90% :     0.000234s :      4: substitution.inline
 1.48% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.82% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.44% :     0.000007s :      2: substitution.replace_old_param
 4.98% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005911     2
82.44% :     0.004873s :      1: type_inference.infer
17.56% :     0.001038s :      1: type_inference.specialize
------[replace.]   0.000079     6
68.20% :     0.000054s :      4: replace.inline
31.80% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000244     6
94.89% :     0.000231s :      4: match.inline
 5.11% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.26% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000002s :     10: predicate.addn_zero_filter
 2.71% :     0.000003s :     10: predicate.arithmetic_simplify
 1.10% :     0.000001s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.06% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.06% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.41% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 6.44% :     0.000008s :     22: predicate.inline
 1.23% :     0.000002s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.46% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.58% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.19% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.82% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 1.06% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.34% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.46% :     0.000002s :     18: predicate.replace_applicator
 0.91% :     0.000001s :      6: predicate.replace_old_param
 0.45% :     0.000001s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.17% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.36% :     0.000002s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.23% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.90% :     0.000010s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000002s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.33% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.90% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.69% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000843    12
51.91% :     0.000437s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.55% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.54% :     0.000350s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025096    76
 0.23% :     0.000058s :      1: add_recomputation
 0.29% :     0.000072s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.89% :     0.000474s :      1: bootstrap
 0.12% :     0.000031s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
27.62% :     0.006931s :      1: jit_opt_a
 0.61% :     0.000153s :      1: jit_opt_after_cconv
 0.23% :     0.000057s :      1: jit_opt_b
 1.81% :     0.000455s :      1: loop_unroll
 2.96% :     0.000743s :      1: mutable_eliminate
 3.79% :     0.000951s :     26: opt.transform.jit_opt_a
 0.20% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.02% :     0.000506s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000048s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.33% :     0.000585s :      1: renormalize.infer
 1.69% :     0.000423s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000044s :      1: rewriter_after_opt_a
 0.33% :     0.000082s :      1: rewriter_before_opt_a
 0.32% :     0.000080s :      1: symbol_engine_optimizer
27.80% :     0.006977s :      1: task_emit
23.98% :     0.006019s :      1: type_inference
 0.26% :     0.000065s :      1: validate


TotalTime = 0.0367785, [30]
[bootstrap]: 0.00044008
[type_inference]: 0.00582003
[event_method]: 2.097e-05
[auto_monad]: 0.00012878
[graph_reusing]: 6.46999e-06
[pre_auto_parallel]: 3.26999e-06
[py_interpret_to_execute]: 4.655e-05
[rewriter_before_opt_a]: 8.066e-05
[expand_dump_flag]: 4.12e-06
[jit_opt_a]: 0.0275581, [2]
    [Cycle 1]: 0.00220051, [27]
        [switch_simplify]: 8.082e-05
        [loop_unroll]: 3.143e-05
        [a_1]: 0.00069601
        [with_stream_mark]: 2.743e-05
        [recompute_prepare]: 7.83001e-06
        [updatestate_depend_eliminate]: 4.55001e-06
        [updatestate_assign_eliminate]: 3.4e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 2.59999e-06
        [specialize_transform]: 6.24999e-06
        [updatestate_useless_node_eliminater]: 6.54999e-06
        [accelerated_algorithm]: 6.01e-06
        [meta_shard_fg_expand]: 2.62001e-06
        [get_grad_eliminate_]: 5.79e-06
        [merge_forward]: 4.30999e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.544e-05
        [j_node_and_user_rematch]: 1.046e-05
        [meta_fg_expand]: 2.54001e-06
        [replace_old_param]: 1.104e-05
        [inline_without_move]: 5.84999e-06
        [renormalize]: 0.00098658
        [add_forward_monad_depend]: 6.84001e-06
        [auto_monad_grad]: 2.49001e-06
        [auto_monad_eliminator]: 1.737e-05
        [cse]: 3.373e-05
        [replace_applicator]: 1.464e-05
    [Cycle 2]: 0.00033915, [27]
        [switch_simplify]: 5.27001e-06
        [loop_unroll]: 5.01002e-06
        [a_1]: 9.742e-05
        [with_stream_mark]: 1.141e-05
        [recompute_prepare]: 5.39e-06
        [updatestate_depend_eliminate]: 3.09001e-06
        [updatestate_assign_eliminate]: 2.06998e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.10001e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 5.89999e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 4.68999e-06
        [merge_forward]: 2.98e-06
        [cell_reuse_recompute_pass]: 2.49001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.481e-05
        [j_node_and_user_rematch]: 7.51999e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 7.71999e-06
        [inline_without_move]: 4.82e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.35999e-06
        [auto_monad_grad]: 1.13001e-06
        [auto_monad_eliminator]: 5.42999e-06
        [cse]: 1.082e-05
        [replace_applicator]: 4.95999e-06
[py_interpret_to_execute_after_opt_a]: 1.368e-05
[rewriter_after_opt_a]: 4.19e-05
[convert_after_rewriter]: 7.25e-06
[order_py_execute_after_rewriter]: 5.32999e-06
[mutable_eliminate]: 0.00088096
[jit_opt_b]: 5.549e-05, [1]
    [Cycle 1]: 4.656e-05, [2]
        [frontend_op_eliminate]: 1.729e-05
        [inline_after_opt_a]: 1.716e-05
[cconv]: 3.025e-05
[loop_unroll]: 0.0004774
[jit_opt_after_cconv]: 0.00016333, [1]
    [Cycle 1]: 0.0001555, [11]
        [c_1]: 2.186e-05
        [parameter_eliminate]: 4.42e-06
        [updatestate_depend_eliminate]: 8.66002e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.41998e-06
        [cse]: 2.633e-05
        [call_graph_tuple_transform]: 2.555e-05
        [tuple_list_get_item_eliminator]: 5.50001e-06
        [none_parameter_eliminate]: 1.73002e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 5.47999e-06
[remove_dup_value]: 1.575e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 5.87999e-06
[add_recomputation]: 5.565e-05
[cse_after_recomputation]: 2.293e-05, [1]
    [Cycle 1]: 1.62e-05, [1]
        [cse]: 9.71e-06
[auto_monad_reorder]: 1.846e-05
[get_jit_bprop_graph]: 2.48998e-06
[rewriter_after_jit_bprop_graph]: 9.35001e-06
[opt_after_jit_grad]: 0.00052066
[symbol_engine_optimizer]: 8.241e-05, [1]
    [Cycle 1]: 7.522e-05, [6]
        [build]: 4.95001e-06
        [elim_shapecalc]: 8.07e-06
        [elim_not_effective]: 1.461e-05
        [opt_reshape]: 6.59999e-06
        [fold_const_symbol]: 9.67001e-06
        [renormalize]: 7.89994e-07
[validate]: 4.655e-05
Sums
    bootstrap                            :     0.000440s :  3.97%
    type_inference                       :     0.005820s : 52.45%
    event_method                         :     0.000021s :  0.19%
    auto_monad                           :     0.000129s :  1.16%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000047s :  0.42%
    rewriter_before_opt_a                :     0.000081s :  0.73%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000086s :  0.78%
    jit_opt_a.loop_unroll                :     0.000036s :  0.33%
    jit_opt_a.a_1                        :     0.000793s :  7.15%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.17%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000987s :  8.89%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.21%
    jit_opt_a.cse                        :     0.000045s :  0.40%
    jit_opt_a.replace_applicator         :     0.000020s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.12%
    rewriter_after_opt_a                 :     0.000042s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000881s :  7.94%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.15%
    cconv                                :     0.000030s :  0.27%
    loop_unroll                          :     0.000477s :  4.30%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000026s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000016s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000056s :  0.50%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000018s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.08%
    opt_after_jit_grad                   :     0.000521s :  4.69%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000047s :  0.42%


Time group info:
------[substitution.]   0.000311    23
 0.67% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000002s :      2: substitution.fold_const_symbol
 2.29% :     0.000007s :      3: substitution.graph_param_transform
87.15% :     0.000271s :      4: substitution.inline
 1.19% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.61% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.77% :     0.000005s :      2: substitution.replace_old_param
 4.78% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005738     2
82.44% :     0.004730s :      1: type_inference.infer
17.56% :     0.001007s :      1: type_inference.specialize
------[replace.]   0.000090     6
67.10% :     0.000060s :      4: replace.inline
32.90% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000282     6
95.25% :     0.000268s :      4: match.inline
 4.75% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   691
 1.57% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.23% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000002s :     10: predicate.addn_zero_filter
 2.72% :     0.000004s :     10: predicate.arithmetic_simplify
 1.24% :     0.000002s :     10: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.90% :     0.000001s :     10: predicate.compare_switch_simplify
 2.19% :     0.000003s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000002s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 1.11% :     0.000002s :      6: predicate.get_grad_eliminate
 0.35% :     0.000000s :      3: predicate.graph_param_transform
 5.98% :     0.000008s :     22: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000002s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.42% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.50% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.34% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.92% :     0.000001s :     10: predicate.merge_addn
 0.90% :     0.000001s :     10: predicate.minmaximum_grad
 2.81% :     0.000004s :      3: predicate.mutable_eliminate
 0.57% :     0.000001s :      3: predicate.opt_reshape
 2.04% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.24% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.42% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.51% :     0.000001s :      3: predicate.reset_defer_inline
 1.79% :     0.000002s :     10: predicate.reshape_eliminate
 1.04% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.63% :     0.000002s :     10: predicate.same_eliminate
 0.63% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.18% :     0.000003s :     16: predicate.switch_defer_inline
 1.78% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.97% :     0.000011s :     53: predicate.switch_simplify
 1.28% :     0.000002s :     10: predicate.tile_eliminate
 0.96% :     0.000001s :     10: predicate.transpose_eliminate
 1.70% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.70% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.60% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.26% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.39% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.54% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000783    12
49.66% :     0.000389s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.78% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.56% :     0.000333s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038850    72
 0.15% :     0.000059s :      1: add_recomputation
 0.34% :     0.000134s :      1: auto_monad
 0.05% :     0.000021s :      1: auto_monad_reorder
 1.19% :     0.000461s :      1: bootstrap
 0.08% :     0.000033s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.06% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.07% :     0.000027s :      1: event_method
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
70.95% :     0.027563s :      1: jit_opt_a
 0.43% :     0.000166s :      1: jit_opt_after_cconv
 0.15% :     0.000058s :      1: jit_opt_b
 1.25% :     0.000485s :      1: loop_unroll
 2.30% :     0.000893s :      1: mutable_eliminate
 2.67% :     0.001038s :     26: opt.transform.jit_opt_a
 0.14% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000027s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.37% :     0.000531s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.13% :     0.000050s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 1.40% :     0.000545s :      1: renormalize.infer
 1.11% :     0.000432s :      1: renormalize.specialize
 0.03% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000046s :      1: rewriter_after_opt_a
 0.22% :     0.000084s :      1: rewriter_before_opt_a
 0.22% :     0.000085s :      1: symbol_engine_optimizer
15.03% :     0.005841s :      1: type_inference


TotalTime = 0.0224604, [33]
[bootstrap]: 0.00035951
[type_inference]: 0.00555604
[event_method]: 2.193e-05
[auto_monad]: 6.893e-05
[graph_reusing]: 5.75001e-06
[pre_auto_parallel]: 2.76e-06
[py_interpret_to_execute]: 4.324e-05
[rewriter_before_opt_a]: 7.853e-05
[expand_dump_flag]: 3.3e-06
[jit_opt_a]: 0.00649895, [2]
    [Cycle 1]: 0.00234861, [27]
        [switch_simplify]: 6.688e-05
        [loop_unroll]: 3.189e-05
        [a_1]: 0.00062229
        [with_stream_mark]: 2.441e-05
        [recompute_prepare]: 8.02e-06
        [updatestate_depend_eliminate]: 4.72998e-06
        [updatestate_assign_eliminate]: 3.46999e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [parameter_eliminate]: 2.31998e-06
        [specialize_transform]: 7.13e-06
        [updatestate_useless_node_eliminater]: 5.35999e-06
        [accelerated_algorithm]: 5.83002e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 5.83002e-06
        [merge_forward]: 3.53999e-06
        [cell_reuse_recompute_pass]: 1.62001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.739e-05
        [j_node_and_user_rematch]: 9.92001e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 9.84999e-06
        [inline_without_move]: 6.49001e-06
        [renormalize]: 0.00121345
        [add_forward_monad_depend]: 9.50001e-06
        [auto_monad_grad]: 2.49999e-06
        [auto_monad_eliminator]: 1.903e-05
        [cse]: 3.405e-05
        [replace_applicator]: 1.704e-05
    [Cycle 2]: 0.00036635, [27]
        [switch_simplify]: 7.05002e-06
        [loop_unroll]: 5.59998e-06
        [a_1]: 0.00010181
        [with_stream_mark]: 1.453e-05
        [recompute_prepare]: 5.84e-06
        [updatestate_depend_eliminate]: 3.51999e-06
        [updatestate_assign_eliminate]: 2.72001e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [parameter_eliminate]: 1.09998e-06
        [specialize_transform]: 5.39e-06
        [updatestate_useless_node_eliminater]: 5.08002e-06
        [accelerated_algorithm]: 6.09001e-06
        [meta_shard_fg_expand]: 1.91998e-06
        [get_grad_eliminate_]: 4.89e-06
        [merge_forward]: 3.81001e-06
        [cell_reuse_recompute_pass]: 2.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.746e-05
        [j_node_and_user_rematch]: 8.72998e-06
        [meta_fg_expand]: 2.02999e-06
        [replace_old_param]: 9.20999e-06
        [inline_without_move]: 4.84e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.70001e-06
        [auto_monad_grad]: 1.29998e-06
        [auto_monad_eliminator]: 5.65001e-06
        [cse]: 1.317e-05
        [replace_applicator]: 5.57001e-06
[py_interpret_to_execute_after_opt_a]: 1.524e-05
[rewriter_after_opt_a]: 3.867e-05
[convert_after_rewriter]: 6.69999e-06
[order_py_execute_after_rewriter]: 4.90999e-06
[mutable_eliminate]: 0.00078157
[jit_opt_b]: 6.412e-05, [1]
    [Cycle 1]: 5.502e-05, [2]
        [frontend_op_eliminate]: 2.464e-05
        [inline_after_opt_a]: 1.703e-05
[cconv]: 3.101e-05
[loop_unroll]: 0.00048358
[jit_opt_after_cconv]: 0.00016043, [1]
    [Cycle 1]: 0.00015253, [11]
        [c_1]: 2.25e-05
        [parameter_eliminate]: 3.33e-06
        [updatestate_depend_eliminate]: 8.38001e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [cse]: 2.608e-05
        [call_graph_tuple_transform]: 2.223e-05
        [tuple_list_get_item_eliminator]: 5.60001e-06
        [none_parameter_eliminate]: 1.62001e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 5.92999e-06
[remove_dup_value]: 1.592e-05
[partial_unused_args_eliminate]: 1.94999e-06
[environ_conv]: 6.21998e-06
[add_recomputation]: 5.58e-05
[cse_after_recomputation]: 2.473e-05, [1]
    [Cycle 1]: 1.886e-05, [1]
        [cse]: 1.145e-05
[auto_monad_reorder]: 1.596e-05
[get_jit_bprop_graph]: 2.56e-06
[rewriter_after_jit_bprop_graph]: 8.42998e-06
[opt_after_jit_grad]: 0.00054331
[symbol_engine_optimizer]: 8.572e-05, [1]
    [Cycle 1]: 7.745e-05, [6]
        [build]: 5.86e-06
        [elim_shapecalc]: 8.27998e-06
        [elim_not_effective]: 1.758e-05
        [opt_reshape]: 6.29001e-06
        [fold_const_symbol]: 8.60001e-06
        [renormalize]: 5.79981e-07
[validate]: 4.677e-05
[backend_pass]: 1.09003e-06
[task_emit]: 0.00716573
[execute]: 9.97001e-06
Sums
    bootstrap                            :     0.000360s :  2.00%
    type_inference                       :     0.005556s : 30.94%
    event_method                         :     0.000022s :  0.12%
    auto_monad                           :     0.000069s :  0.38%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.24%
    rewriter_before_opt_a                :     0.000079s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.41%
    jit_opt_a.loop_unroll                :     0.000037s :  0.21%
    jit_opt_a.a_1                        :     0.000724s :  4.03%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001214s :  6.76%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.14%
    jit_opt_a.cse                        :     0.000047s :  0.26%
    jit_opt_a.replace_applicator         :     0.000023s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.08%
    rewriter_after_opt_a                 :     0.000039s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000782s :  4.35%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000031s :  0.17%
    loop_unroll                          :     0.000484s :  2.69%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000056s :  0.31%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000543s :  3.03%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000047s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007166s : 39.90%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000266    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000002s :      2: substitution.fold_const_symbol
 2.11% :     0.000006s :      3: substitution.graph_param_transform
85.54% :     0.000227s :      4: substitution.inline
 1.52% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.18% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.39% :     0.000006s :      2: substitution.replace_old_param
 4.85% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005475     2
81.81% :     0.004479s :      1: type_inference.infer
18.19% :     0.000996s :      1: type_inference.specialize
------[replace.]   0.000081     6
67.48% :     0.000055s :      4: replace.inline
32.52% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000236     6
95.07% :     0.000225s :      4: match.inline
 4.93% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   691
 1.11% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.68% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 1.28% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.04% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.15% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.47% :     0.000007s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.59% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.77% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 2.62% :     0.000003s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     15: predicate.partial_eliminate
 1.43% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.12% :     0.000003s :     10: predicate.reduce_eliminate
 1.67% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.81% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.45% :     0.000001s :      3: predicate.reset_defer_inline
 1.27% :     0.000002s :     10: predicate.reshape_eliminate
 1.29% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.55% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.37% :     0.000002s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.51% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.22% :     0.000009s :     53: predicate.switch_simplify
 1.23% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000002s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.77% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.35% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.83% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000818    12
52.04% :     0.000426s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.36% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.60% :     0.000340s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024775    76
 0.24% :     0.000059s :      1: add_recomputation
 0.29% :     0.000072s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.52% :     0.000378s :      1: bootstrap
 0.14% :     0.000034s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.11% :     0.000027s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.25% :     0.006504s :      1: jit_opt_a
 0.66% :     0.000163s :      1: jit_opt_after_cconv
 0.27% :     0.000067s :      1: jit_opt_b
 1.99% :     0.000493s :      1: loop_unroll
 3.22% :     0.000797s :      1: mutable_eliminate
 3.91% :     0.000969s :     26: opt.transform.jit_opt_a
 0.21% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000034s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.19% :     0.000046s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.24% :     0.000554s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000046s :      1: py_interpret_to_execute
 0.07% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 2.96% :     0.000734s :      1: renormalize.infer
 1.89% :     0.000467s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000043s :      1: rewriter_after_opt_a
 0.33% :     0.000082s :      1: rewriter_before_opt_a
 0.36% :     0.000088s :      1: symbol_engine_optimizer
29.00% :     0.007185s :      1: task_emit
22.50% :     0.005574s :      1: type_inference
 0.29% :     0.000072s :      1: validate


TotalTime = 0.0566663, [33]
[bootstrap]: 0.00030463
[type_inference]: 0.02805
[event_method]: 2.294e-05
[auto_monad]: 7.366e-05
[graph_reusing]: 7.01001e-06
[pre_auto_parallel]: 3.38e-06
[py_interpret_to_execute]: 4.357e-05
[rewriter_before_opt_a]: 8.116e-05
[expand_dump_flag]: 3.60003e-06
[jit_opt_a]: 0.0111684, [2]
    [Cycle 1]: 0.00318822, [27]
        [switch_simplify]: 6.538e-05
        [loop_unroll]: 3.34e-05
        [a_1]: 0.00071025
        [with_stream_mark]: 2.619e-05
        [recompute_prepare]: 9.38002e-06
        [updatestate_depend_eliminate]: 4.37e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 2.75002e-06
        [parameter_eliminate]: 2.07001e-06
        [specialize_transform]: 6.76999e-06
        [updatestate_useless_node_eliminater]: 5.99e-06
        [accelerated_algorithm]: 7.22002e-06
        [meta_shard_fg_expand]: 2.45002e-06
        [get_grad_eliminate_]: 5.46998e-06
        [merge_forward]: 4.55001e-06
        [cell_reuse_recompute_pass]: 1.68002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.566e-05
        [j_node_and_user_rematch]: 1.079e-05
        [meta_fg_expand]: 2.91999e-06
        [replace_old_param]: 1.065e-05
        [inline_without_move]: 5.42001e-06
        [renormalize]: 0.00193503
        [add_forward_monad_depend]: 1.047e-05
        [auto_monad_grad]: 2.59999e-06
        [auto_monad_eliminator]: 2.152e-05
        [cse]: 3.621e-05
        [replace_applicator]: 2.232e-05
    [Cycle 2]: 0.00035626, [27]
        [switch_simplify]: 6.23e-06
        [loop_unroll]: 5.22999e-06
        [a_1]: 9.727e-05
        [with_stream_mark]: 1.56e-05
        [recompute_prepare]: 5.69999e-06
        [updatestate_depend_eliminate]: 4.1e-06
        [updatestate_assign_eliminate]: 2.75002e-06
        [updatestate_loads_eliminate]: 2.28002e-06
        [parameter_eliminate]: 1.30999e-06
        [specialize_transform]: 4.82e-06
        [updatestate_useless_node_eliminater]: 4.78001e-06
        [accelerated_algorithm]: 5.76998e-06
        [meta_shard_fg_expand]: 2.22001e-06
        [get_grad_eliminate_]: 4.50999e-06
        [merge_forward]: 4.05e-06
        [cell_reuse_recompute_pass]: 2.83e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.633e-05
        [j_node_and_user_rematch]: 9.29e-06
        [meta_fg_expand]: 1.89999e-06
        [replace_old_param]: 8.88002e-06
        [inline_without_move]: 4.84998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.28002e-06
        [auto_monad_grad]: 1.28002e-06
        [auto_monad_eliminator]: 6.58998e-06
        [cse]: 1.235e-05
        [replace_applicator]: 5.04e-06
[py_interpret_to_execute_after_opt_a]: 1.936e-05
[rewriter_after_opt_a]: 4.163e-05
[convert_after_rewriter]: 7.6e-06
[order_py_execute_after_rewriter]: 5.62001e-06
[mutable_eliminate]: 0.00075776
[jit_opt_b]: 5.063e-05, [1]
    [Cycle 1]: 4.26e-05, [2]
        [frontend_op_eliminate]: 1.556e-05
        [inline_after_opt_a]: 1.551e-05
[cconv]: 3.228e-05
[loop_unroll]: 0.00052192
[jit_opt_after_cconv]: 0.00015922, [1]
    [Cycle 1]: 0.00015191, [11]
        [c_1]: 2.058e-05
        [parameter_eliminate]: 3.85e-06
        [updatestate_depend_eliminate]: 9.72999e-06
        [updatestate_assign_eliminate]: 2.56998e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [cse]: 2.74e-05
        [call_graph_tuple_transform]: 2.071e-05
        [tuple_list_get_item_eliminator]: 5.70001e-06
        [none_parameter_eliminate]: 1.89e-06
        [renormalize]: 6.40022e-07
        [switch_simplify]: 5.37001e-06
[remove_dup_value]: 1.407e-05
[partial_unused_args_eliminate]: 2.37001e-06
[environ_conv]: 5.72999e-06
[add_recomputation]: 5.986e-05
[cse_after_recomputation]: 2.2e-05, [1]
    [Cycle 1]: 1.535e-05, [1]
        [cse]: 9.54999e-06
[auto_monad_reorder]: 1.565e-05
[get_jit_bprop_graph]: 2.17999e-06
[rewriter_after_jit_bprop_graph]: 9.41e-06
[opt_after_jit_grad]: 0.00049722
[symbol_engine_optimizer]: 7.54e-05, [1]
    [Cycle 1]: 6.839e-05, [6]
        [build]: 4.82998e-06
        [elim_shapecalc]: 7.41001e-06
        [elim_not_effective]: 1.365e-05
        [opt_reshape]: 5.69e-06
        [fold_const_symbol]: 8.95001e-06
        [renormalize]: 5.50004e-07
[validate]: 9.298e-05
[backend_pass]: 1.35001e-06
[task_emit]: 0.0142425
[execute]: 1.09e-05
Sums
    bootstrap                            :     0.000305s :  0.63%
    type_inference                       :     0.028050s : 58.06%
    event_method                         :     0.000023s :  0.05%
    auto_monad                           :     0.000074s :  0.15%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.09%
    rewriter_before_opt_a                :     0.000081s :  0.17%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000072s :  0.15%
    jit_opt_a.loop_unroll                :     0.000039s :  0.08%
    jit_opt_a.a_1                        :     0.000808s :  1.67%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.02%
    jit_opt_a.merge_forward              :     0.000009s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.07%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000020s :  0.04%
    jit_opt_a.inline_without_move        :     0.000010s :  0.02%
    jit_opt_a.renormalize                :     0.001935s :  4.01%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.06%
    jit_opt_a.cse                        :     0.000049s :  0.10%
    jit_opt_a.replace_applicator         :     0.000027s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.04%
    rewriter_after_opt_a                 :     0.000042s :  0.09%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.01%
    mutable_eliminate                    :     0.000758s :  1.57%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.03%
    cconv                                :     0.000032s :  0.07%
    loop_unroll                          :     0.000522s :  1.08%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.04%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000027s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.01%
    remove_dup_value                     :     0.000014s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000006s :  0.01%
    add_recomputation                    :     0.000060s :  0.12%
    cse_after_recomputation.cse          :     0.000010s :  0.02%
    auto_monad_reorder                   :     0.000016s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.02%
    opt_after_jit_grad                   :     0.000497s :  1.03%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000093s :  0.19%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.014243s : 29.48%
    execute                              :     0.000011s :  0.02%


Time group info:
------[substitution.]   0.000279    23
 0.71% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 1.97% :     0.000006s :      3: substitution.graph_param_transform
85.93% :     0.000240s :      4: substitution.inline
 1.46% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.79% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.48% :     0.000007s :      2: substitution.replace_old_param
 5.16% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.027937     2
91.40% :     0.025533s :      1: type_inference.infer
 8.60% :     0.002404s :      1: type_inference.specialize
------[replace.]   0.000110     6
45.44% :     0.000050s :      4: replace.inline
54.56% :     0.000060s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000250     6
94.74% :     0.000237s :      4: match.inline
 5.26% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.21% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.43% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 2.27% :     0.000003s :     10: predicate.arithmetic_simplify
 1.55% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     10: predicate.environ_get_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.79% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 6.89% :     0.000009s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.79% :     0.000002s :      6: predicate.less_batch_normalization
 1.77% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.83% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.93% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.81% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.24% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.60% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.85% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.50% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.77% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.85% :     0.000010s :     53: predicate.switch_simplify
 1.17% :     0.000002s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.03% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.92% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.61% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.43% :     0.000002s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000940    12
54.55% :     0.000513s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.17% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.28% :     0.000369s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.059754    76
 0.11% :     0.000063s :      1: add_recomputation
 0.13% :     0.000077s :      1: auto_monad
 0.03% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.53% :     0.000316s :      1: bootstrap
 0.06% :     0.000035s :      1: cconv
 0.02% :     0.000012s :      1: convert_after_rewriter
 0.04% :     0.000024s :      1: cse_after_recomputation
 0.01% :     0.000008s :      1: environ_conv
 0.05% :     0.000028s :      1: event_method
 0.03% :     0.000016s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
18.70% :     0.011172s :      1: jit_opt_a
 0.27% :     0.000162s :      1: jit_opt_after_cconv
 0.09% :     0.000053s :      1: jit_opt_b
 0.89% :     0.000533s :      1: loop_unroll
 1.29% :     0.000770s :      1: mutable_eliminate
 1.76% :     0.001055s :     26: opt.transform.jit_opt_a
 0.08% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000025s :      4: opt.transform.jit_opt_b
 0.03% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.05% :     0.000033s :      4: opt.transform.symbol_engine_opt
 0.85% :     0.000506s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.08% :     0.000046s :      1: py_interpret_to_execute
 0.04% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000017s :      1: remove_dup_value
 2.52% :     0.001503s :      1: renormalize.infer
 0.70% :     0.000416s :      1: renormalize.specialize
 0.02% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000045s :      1: rewriter_after_opt_a
 0.14% :     0.000084s :      1: rewriter_before_opt_a
 0.13% :     0.000078s :      1: symbol_engine_optimizer
23.87% :     0.014263s :      1: task_emit
47.02% :     0.028099s :      1: type_inference
 0.20% :     0.000117s :      1: validate


TotalTime = 0.0156693, [30]
[bootstrap]: 0.00043089
[type_inference]: 0.0055342
[event_method]: 2.008e-05
[auto_monad]: 6.925e-05
[graph_reusing]: 5.85002e-06
[pre_auto_parallel]: 2.72001e-06
[py_interpret_to_execute]: 4.523e-05
[rewriter_before_opt_a]: 7.944e-05
[expand_dump_flag]: 3.48999e-06
[jit_opt_a]: 0.00672937, [2]
    [Cycle 1]: 0.00217191, [27]
        [switch_simplify]: 7.183e-05
        [loop_unroll]: 2.94e-05
        [a_1]: 0.0006427
        [with_stream_mark]: 2.5e-05
        [recompute_prepare]: 7.99002e-06
        [updatestate_depend_eliminate]: 4.08999e-06
        [updatestate_assign_eliminate]: 3.38e-06
        [updatestate_loads_eliminate]: 2.76e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 5.92999e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 6.29001e-06
        [meta_shard_fg_expand]: 2.38998e-06
        [get_grad_eliminate_]: 5.07999e-06
        [merge_forward]: 4.00998e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.664e-05
        [j_node_and_user_rematch]: 1.125e-05
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 9.97999e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.00103481
        [add_forward_monad_depend]: 8.40999e-06
        [auto_monad_grad]: 2.52001e-06
        [auto_monad_eliminator]: 1.759e-05
        [cse]: 3.149e-05
        [replace_applicator]: 1.758e-05
    [Cycle 2]: 0.00035175, [27]
        [switch_simplify]: 5.99e-06
        [loop_unroll]: 5.69e-06
        [a_1]: 0.00010009
        [with_stream_mark]: 1.439e-05
        [recompute_prepare]: 5.30999e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 2.03002e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 2.18998e-06
        [specialize_transform]: 4.99e-06
        [updatestate_useless_node_eliminater]: 5.08002e-06
        [accelerated_algorithm]: 5.20001e-06
        [meta_shard_fg_expand]: 1.58002e-06
        [get_grad_eliminate_]: 4.50999e-06
        [merge_forward]: 3.3e-06
        [cell_reuse_recompute_pass]: 2.11998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.531e-05
        [j_node_and_user_rematch]: 8.17e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 8.68001e-06
        [inline_without_move]: 5.05001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.35001e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 5.27999e-06
        [cse]: 1.181e-05
        [replace_applicator]: 4.93001e-06
[py_interpret_to_execute_after_opt_a]: 1.592e-05
[rewriter_after_opt_a]: 4.053e-05
[convert_after_rewriter]: 8.01001e-06
[order_py_execute_after_rewriter]: 5.00001e-06
[mutable_eliminate]: 0.00088095
[jit_opt_b]: 5.857e-05, [1]
    [Cycle 1]: 4.919e-05, [2]
        [frontend_op_eliminate]: 1.967e-05
        [inline_after_opt_a]: 1.769e-05
[cconv]: 3.663e-05
[loop_unroll]: 0.00050959
[jit_opt_after_cconv]: 0.00017597, [1]
    [Cycle 1]: 0.00016892, [11]
        [c_1]: 2.299e-05
        [parameter_eliminate]: 5.27001e-06
        [updatestate_depend_eliminate]: 1.16e-05
        [updatestate_assign_eliminate]: 3.03998e-06
        [updatestate_loads_eliminate]: 2.91e-06
        [cse]: 3.089e-05
        [call_graph_tuple_transform]: 2.505e-05
        [tuple_list_get_item_eliminator]: 5.62999e-06
        [none_parameter_eliminate]: 1.97999e-06
        [renormalize]: 8.59989e-07
        [switch_simplify]: 5.66e-06
[remove_dup_value]: 1.669e-05
[partial_unused_args_eliminate]: 2.29001e-06
[environ_conv]: 6.06998e-06
[add_recomputation]: 6.046e-05
[cse_after_recomputation]: 2.392e-05, [1]
    [Cycle 1]: 1.734e-05, [1]
        [cse]: 1.031e-05
[auto_monad_reorder]: 1.624e-05
[get_jit_bprop_graph]: 2.76999e-06
[rewriter_after_jit_bprop_graph]: 1.012e-05
[opt_after_jit_grad]: 0.00054411
[symbol_engine_optimizer]: 8.583e-05, [1]
    [Cycle 1]: 7.762e-05, [6]
        [build]: 6.66e-06
        [elim_shapecalc]: 8.48999e-06
        [elim_not_effective]: 1.617e-05
        [opt_reshape]: 7.38e-06
        [fold_const_symbol]: 1.003e-05
        [renormalize]: 8.80013e-07
[validate]: 5.019e-05
Sums
    bootstrap                            :     0.000431s :  3.98%
    type_inference                       :     0.005534s : 51.16%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000069s :  0.64%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000045s :  0.42%
    rewriter_before_opt_a                :     0.000079s :  0.73%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000078s :  0.72%
    jit_opt_a.loop_unroll                :     0.000035s :  0.32%
    jit_opt_a.a_1                        :     0.000743s :  6.87%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.36%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.17%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.001035s :  9.57%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.21%
    jit_opt_a.cse                        :     0.000043s :  0.40%
    jit_opt_a.replace_applicator         :     0.000023s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.15%
    rewriter_after_opt_a                 :     0.000041s :  0.37%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000881s :  8.14%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.16%
    cconv                                :     0.000037s :  0.34%
    loop_unroll                          :     0.000510s :  4.71%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.11%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000031s :  0.29%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000017s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000060s :  0.56%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.15%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.09%
    opt_after_jit_grad                   :     0.000544s :  5.03%
    symbol_engine_optimizer.build        :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000050s :  0.46%


Time group info:
------[substitution.]   0.000288    23
 0.87% :     0.000003s :      2: substitution.elim_not_effective
 0.75% :     0.000002s :      2: substitution.fold_const_symbol
 2.20% :     0.000006s :      3: substitution.graph_param_transform
86.49% :     0.000249s :      4: substitution.inline
 1.64% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.65% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.92% :     0.000006s :      2: substitution.replace_old_param
 4.48% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005446     2
81.33% :     0.004429s :      1: type_inference.infer
18.67% :     0.001017s :      1: type_inference.specialize
------[replace.]   0.000083     6
68.43% :     0.000057s :      4: replace.inline
31.57% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000259     6
95.46% :     0.000247s :      4: match.inline
 4.54% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.21% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.67% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.20% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.42% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.17% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000001s :      3: predicate.elim_not_effective
 0.77% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.88% :     0.000008s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.01% :     0.000003s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.92% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.94% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 2.41% :     0.000003s :      3: predicate.mutable_eliminate
 0.80% :     0.000001s :      3: predicate.opt_reshape
 2.02% :     0.000003s :     15: predicate.partial_eliminate
 1.49% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.29% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.75% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.44% :     0.000001s :      3: predicate.reset_defer_inline
 1.92% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.19% :     0.000002s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.51% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.54% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.14% :     0.000003s :     16: predicate.switch_defer_inline
 1.77% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.77% :     0.000010s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000002s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.61% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.43% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.88% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.81% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000792    12
49.15% :     0.000389s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.68% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.17% :     0.000342s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017738    72
 0.36% :     0.000064s :      1: add_recomputation
 0.41% :     0.000073s :      1: auto_monad
 0.11% :     0.000019s :      1: auto_monad_reorder
 2.51% :     0.000445s :      1: bootstrap
 0.23% :     0.000040s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.14% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
37.96% :     0.006734s :      1: jit_opt_a
 1.01% :     0.000179s :      1: jit_opt_after_cconv
 0.35% :     0.000061s :      1: jit_opt_b
 2.92% :     0.000518s :      1: loop_unroll
 5.04% :     0.000894s :      1: mutable_eliminate
 5.52% :     0.000980s :     26: opt.transform.jit_opt_a
 0.31% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000030s :      4: opt.transform.jit_opt_b
 0.08% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000038s :      4: opt.transform.symbol_engine_opt
 3.12% :     0.000553s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000048s :      1: py_interpret_to_execute
 0.10% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000019s :      1: remove_dup_value
 3.11% :     0.000551s :      1: renormalize.infer
 2.67% :     0.000474s :      1: renormalize.specialize
 0.07% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000043s :      1: rewriter_after_opt_a
 0.47% :     0.000083s :      1: rewriter_before_opt_a
 0.50% :     0.000088s :      1: symbol_engine_optimizer
31.30% :     0.005552s :      1: type_inference


TotalTime = 0.0603419, [33]
[bootstrap]: 0.00030133
[type_inference]: 0.00603568
[event_method]: 1.663e-05
[auto_monad]: 6.4e-05
[graph_reusing]: 5.76e-06
[pre_auto_parallel]: 3.03e-06
[py_interpret_to_execute]: 4.244e-05
[rewriter_before_opt_a]: 7.478e-05
[expand_dump_flag]: 3.21001e-06
[jit_opt_a]: 0.00591772, [2]
    [Cycle 1]: 0.00189069, [27]
        [switch_simplify]: 5.985e-05
        [loop_unroll]: 3.047e-05
        [a_1]: 0.00056091
        [with_stream_mark]: 1.981e-05
        [recompute_prepare]: 8.35999e-06
        [updatestate_depend_eliminate]: 4.2e-06
        [updatestate_assign_eliminate]: 3.11999e-06
        [updatestate_loads_eliminate]: 2.58998e-06
        [parameter_eliminate]: 2.07999e-06
        [specialize_transform]: 6.27001e-06
        [updatestate_useless_node_eliminater]: 5.15999e-06
        [accelerated_algorithm]: 6.12999e-06
        [meta_shard_fg_expand]: 2.37001e-06
        [get_grad_eliminate_]: 5.61e-06
        [merge_forward]: 3.61999e-06
        [cell_reuse_recompute_pass]: 1.19998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.507e-05
        [j_node_and_user_rematch]: 9.97001e-06
        [meta_fg_expand]: 2.40002e-06
        [replace_old_param]: 9.22999e-06
        [inline_without_move]: 5.42999e-06
        [renormalize]: 0.00087054
        [add_forward_monad_depend]: 6.14001e-06
        [auto_monad_grad]: 2.12001e-06
        [auto_monad_eliminator]: 1.475e-05
        [cse]: 2.935e-05
        [replace_applicator]: 1.508e-05
    [Cycle 2]: 0.00035353, [27]
        [switch_simplify]: 5.68002e-06
        [loop_unroll]: 5.17e-06
        [a_1]: 9.578e-05
        [with_stream_mark]: 1.412e-05
        [recompute_prepare]: 5.94e-06
        [updatestate_depend_eliminate]: 2.54999e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 1.35001e-06
        [specialize_transform]: 5.34e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.84999e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 4.83001e-06
        [merge_forward]: 2.91e-06
        [cell_reuse_recompute_pass]: 1.91003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.406e-05
        [j_node_and_user_rematch]: 8.38001e-06
        [meta_fg_expand]: 1.80001e-06
        [replace_old_param]: 8.28999e-06
        [inline_without_move]: 4.67998e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.81e-06
        [auto_monad_grad]: 8.89995e-07
        [auto_monad_eliminator]: 5.32001e-06
        [cse]: 1.301e-05
        [replace_applicator]: 5.62001e-06
[py_interpret_to_execute_after_opt_a]: 1.279e-05
[rewriter_after_opt_a]: 3.735e-05
[convert_after_rewriter]: 7.23e-06
[order_py_execute_after_rewriter]: 5.20999e-06
[mutable_eliminate]: 0.00073281
[jit_opt_b]: 6.059e-05, [1]
    [Cycle 1]: 5.208e-05, [2]
        [frontend_op_eliminate]: 2.039e-05
        [inline_after_opt_a]: 1.751e-05
[cconv]: 3.111e-05
[loop_unroll]: 0.0369925
[jit_opt_after_cconv]: 0.00022495, [1]
    [Cycle 1]: 0.00021309, [11]
        [c_1]: 2.961e-05
        [parameter_eliminate]: 6.07001e-06
        [updatestate_depend_eliminate]: 9.54e-06
        [updatestate_assign_eliminate]: 3.65e-06
        [updatestate_loads_eliminate]: 3.55998e-06
        [cse]: 5.698e-05
        [call_graph_tuple_transform]: 2.778e-05
        [tuple_list_get_item_eliminator]: 6.68e-06
        [none_parameter_eliminate]: 1.67999e-06
        [renormalize]: 8.99978e-07
        [switch_simplify]: 6.83e-06
[remove_dup_value]: 1.599e-05
[partial_unused_args_eliminate]: 2.84999e-06
[environ_conv]: 6.61e-06
[add_recomputation]: 5.85e-05
[cse_after_recomputation]: 2.614e-05, [1]
    [Cycle 1]: 1.831e-05, [1]
        [cse]: 1.143e-05
[auto_monad_reorder]: 1.718e-05
[get_jit_bprop_graph]: 2.29999e-06
[rewriter_after_jit_bprop_graph]: 1.153e-05
[opt_after_jit_grad]: 0.00070919
[symbol_engine_optimizer]: 0.00014966, [1]
    [Cycle 1]: 0.00013891, [6]
        [build]: 6.48e-06
        [elim_shapecalc]: 8.94e-06
        [elim_not_effective]: 7.134e-05
        [opt_reshape]: 8.21002e-06
        [fold_const_symbol]: 1.118e-05
        [renormalize]: 9.60019e-07
[validate]: 5.062e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.007779
[execute]: 0.00063881
Sums
    bootstrap                            :     0.000301s :  0.54%
    type_inference                       :     0.006036s : 10.80%
    event_method                         :     0.000017s :  0.03%
    auto_monad                           :     0.000064s :  0.11%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.08%
    rewriter_before_opt_a                :     0.000075s :  0.13%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000066s :  0.12%
    jit_opt_a.loop_unroll                :     0.000036s :  0.06%
    jit_opt_a.a_1                        :     0.000657s :  1.17%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.06%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.02%
    jit_opt_a.merge_forward              :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.05%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000018s :  0.03%
    jit_opt_a.inline_without_move        :     0.000010s :  0.02%
    jit_opt_a.renormalize                :     0.000871s :  1.56%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.04%
    jit_opt_a.cse                        :     0.000042s :  0.08%
    jit_opt_a.replace_applicator         :     0.000021s :  0.04%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.02%
    rewriter_after_opt_a                 :     0.000037s :  0.07%
    convert_after_rewriter               :     0.000007s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000733s :  1.31%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.03%
    cconv                                :     0.000031s :  0.06%
    loop_unroll                          :     0.036992s : 66.18%
    jit_opt_after_cconv.c_1              :     0.000030s :  0.05%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000057s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.01%
    remove_dup_value                     :     0.000016s :  0.03%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.01%
    add_recomputation                    :     0.000059s :  0.10%
    cse_after_recomputation.cse          :     0.000011s :  0.02%
    auto_monad_reorder                   :     0.000017s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000012s :  0.02%
    opt_after_jit_grad                   :     0.000709s :  1.27%
    symbol_engine_optimizer.build        :     0.000006s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000071s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000051s :  0.09%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007779s : 13.92%
    execute                              :     0.000639s :  1.14%


Time group info:
------[substitution.]   0.000214    23
 0.94% :     0.000002s :      2: substitution.elim_not_effective
 1.05% :     0.000002s :      2: substitution.fold_const_symbol
 3.13% :     0.000007s :      3: substitution.graph_param_transform
82.35% :     0.000177s :      4: substitution.inline
 1.99% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.35% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.27% :     0.000005s :      2: substitution.replace_old_param
 5.93% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005958     2
83.25% :     0.004961s :      1: type_inference.infer
16.75% :     0.000998s :      1: type_inference.specialize
------[replace.]   0.000077     6
67.20% :     0.000052s :      4: replace.inline
32.80% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000185     6
93.79% :     0.000174s :      4: match.inline
 6.21% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   691
 1.13% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.53% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.04% :     0.000003s :     10: predicate.arithmetic_simplify
 1.16% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.45% :     0.000002s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.36% :     0.000000s :      3: predicate.graph_param_transform
 5.36% :     0.000007s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.47% :     0.000002s :      6: predicate.less_batch_normalization
 1.82% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.59% :     0.000002s :     15: predicate.load_eliminater
 2.92% :     0.000004s :      3: predicate.loop_unroll_after_grad
 3.81% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.38% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 0.93% :     0.000001s :     10: predicate.minmaximum_grad
 2.45% :     0.000003s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.06% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.12% :     0.000003s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.24% :     0.000002s :     10: predicate.reshape_eliminate
 1.04% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.07% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.26% :     0.000002s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.07% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 1.89% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.25% :     0.000009s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.36% :     0.000002s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.27% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.33% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000783    12
46.23% :     0.000362s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 3.73% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.05% :     0.000392s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.062292    76
 0.10% :     0.000062s :      1: add_recomputation
 0.11% :     0.000067s :      1: auto_monad
 0.03% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.52% :     0.000322s :      1: bootstrap
 0.06% :     0.000035s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.05% :     0.000029s :      1: cse_after_recomputation
 0.02% :     0.000011s :      1: environ_conv
 0.03% :     0.000021s :      1: event_method
 1.08% :     0.000671s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000008s :      1: graph_reusing
 9.51% :     0.005922s :      1: jit_opt_a
 0.37% :     0.000229s :      1: jit_opt_after_cconv
 0.10% :     0.000064s :      1: jit_opt_b
59.43% :     0.037019s :      1: loop_unroll
 1.20% :     0.000746s :      1: mutable_eliminate
 1.42% :     0.000882s :     26: opt.transform.jit_opt_a
 0.11% :     0.000067s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000029s :      4: opt.transform.jit_opt_b
 0.05% :     0.000028s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000095s :      4: opt.transform.symbol_engine_opt
 1.16% :     0.000721s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000045s :      1: py_interpret_to_execute
 0.02% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000019s :      1: remove_dup_value
 0.68% :     0.000425s :      1: renormalize.infer
 0.70% :     0.000437s :      1: renormalize.specialize
 0.02% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000041s :      1: rewriter_after_opt_a
 0.13% :     0.000078s :      1: rewriter_before_opt_a
 0.25% :     0.000154s :      1: symbol_engine_optimizer
12.51% :     0.007795s :      1: task_emit
 9.72% :     0.006053s :      1: type_inference
 0.12% :     0.000077s :      1: validate


TotalTime = 0.0473438, [33]
[bootstrap]: 0.00047644
[type_inference]: 0.00742101
[event_method]: 2.264e-05
[auto_monad]: 7.669e-05
[graph_reusing]: 7.02002e-06
[pre_auto_parallel]: 3.35e-06
[py_interpret_to_execute]: 4.691e-05
[rewriter_before_opt_a]: 7.862e-05
[expand_dump_flag]: 3.23e-06
[jit_opt_a]: 0.0279381, [2]
    [Cycle 1]: 0.00666952, [27]
        [switch_simplify]: 6.638e-05
        [loop_unroll]: 3.03e-05
        [a_1]: 0.00062906
        [with_stream_mark]: 2.798e-05
        [recompute_prepare]: 8.33999e-06
        [updatestate_depend_eliminate]: 4.25e-06
        [updatestate_assign_eliminate]: 3.04999e-06
        [updatestate_loads_eliminate]: 2.83e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 6.54001e-06
        [updatestate_useless_node_eliminater]: 5.20001e-06
        [accelerated_algorithm]: 6.04001e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 5.19e-06
        [merge_forward]: 3.50003e-06
        [cell_reuse_recompute_pass]: 1.76998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.432e-05
        [j_node_and_user_rematch]: 1.643e-05
        [meta_fg_expand]: 3.25998e-06
        [replace_old_param]: 1.071e-05
        [inline_without_move]: 5.77001e-06
        [renormalize]: 0.00547705
        [add_forward_monad_depend]: 1.353e-05
        [auto_monad_grad]: 2.70002e-06
        [auto_monad_eliminator]: 2.762e-05
        [cse]: 3.593e-05
        [replace_applicator]: 2.77e-05
    [Cycle 2]: 0.00039163, [27]
        [switch_simplify]: 6.39999e-06
        [loop_unroll]: 5.64998e-06
        [a_1]: 0.0001049
        [with_stream_mark]: 1.937e-05
        [recompute_prepare]: 5.46e-06
        [updatestate_depend_eliminate]: 4.68001e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 2.17001e-06
        [specialize_transform]: 5.44e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 5.83997e-06
        [meta_shard_fg_expand]: 2.46e-06
        [get_grad_eliminate_]: 5.15999e-06
        [merge_forward]: 4.92999e-06
        [cell_reuse_recompute_pass]: 4.15e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.024e-05
        [j_node_and_user_rematch]: 1.106e-05
        [meta_fg_expand]: 2.02999e-06
        [replace_old_param]: 1.333e-05
        [inline_without_move]: 5.20001e-06
        [renormalize]: 1.19995e-07
        [add_forward_monad_depend]: 2.33002e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 7.36999e-06
        [cse]: 1.307e-05
        [replace_applicator]: 5.97001e-06
[py_interpret_to_execute_after_opt_a]: 2.373e-05
[rewriter_after_opt_a]: 4.511e-05
[convert_after_rewriter]: 8.60999e-06
[order_py_execute_after_rewriter]: 5.50001e-06
[mutable_eliminate]: 0.00089013
[jit_opt_b]: 5.907e-05, [1]
    [Cycle 1]: 4.868e-05, [2]
        [frontend_op_eliminate]: 1.846e-05
        [inline_after_opt_a]: 1.768e-05
[cconv]: 3.76e-05
[loop_unroll]: 0.00049447
[jit_opt_after_cconv]: 0.00017133, [1]
    [Cycle 1]: 0.00016362, [11]
        [c_1]: 2.163e-05
        [parameter_eliminate]: 5.32999e-06
        [updatestate_depend_eliminate]: 9.91003e-06
        [updatestate_assign_eliminate]: 2.66999e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [cse]: 3.195e-05
        [call_graph_tuple_transform]: 2.32e-05
        [tuple_list_get_item_eliminator]: 5.49e-06
        [none_parameter_eliminate]: 1.88002e-06
        [renormalize]: 7.10017e-07
        [switch_simplify]: 5.53002e-06
[remove_dup_value]: 1.571e-05
[partial_unused_args_eliminate]: 2.72001e-06
[environ_conv]: 6.18002e-06
[add_recomputation]: 6.116e-05
[cse_after_recomputation]: 2.398e-05, [1]
    [Cycle 1]: 1.757e-05, [1]
        [cse]: 1.043e-05
[auto_monad_reorder]: 1.501e-05
[get_jit_bprop_graph]: 2.19999e-06
[rewriter_after_jit_bprop_graph]: 9.04e-06
[opt_after_jit_grad]: 0.00049849
[symbol_engine_optimizer]: 9.516e-05, [1]
    [Cycle 1]: 7.412e-05, [6]
        [build]: 5.62999e-06
        [elim_shapecalc]: 9.25999e-06
        [elim_not_effective]: 1.471e-05
        [opt_reshape]: 5.71003e-06
        [fold_const_symbol]: 9.00001e-06
        [renormalize]: 2.3999e-07
[validate]: 4.847e-05
[backend_pass]: 9.90025e-07
[task_emit]: 0.00849706
[execute]: 1.07e-05
Sums
    bootstrap                            :     0.000476s :  1.85%
    type_inference                       :     0.007421s : 28.85%
    event_method                         :     0.000023s :  0.09%
    auto_monad                           :     0.000077s :  0.30%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000047s :  0.18%
    rewriter_before_opt_a                :     0.000079s :  0.31%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000073s :  0.28%
    jit_opt_a.loop_unroll                :     0.000036s :  0.14%
    jit_opt_a.a_1                        :     0.000734s :  2.85%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000008s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000024s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.04%
    jit_opt_a.renormalize                :     0.005477s : 21.30%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000035s :  0.14%
    jit_opt_a.cse                        :     0.000049s :  0.19%
    jit_opt_a.replace_applicator         :     0.000034s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000024s :  0.09%
    rewriter_after_opt_a                 :     0.000045s :  0.18%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000890s :  3.46%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.07%
    cconv                                :     0.000038s :  0.15%
    loop_unroll                          :     0.000494s :  1.92%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000032s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000061s :  0.24%
    cse_after_recomputation.cse          :     0.000010s :  0.04%
    auto_monad_reorder                   :     0.000015s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.04%
    opt_after_jit_grad                   :     0.000498s :  1.94%
    symbol_engine_optimizer.build        :     0.000006s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000048s :  0.19%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008497s : 33.04%
    execute                              :     0.000011s :  0.04%


Time group info:
------[substitution.]   0.000284    23
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 1.97% :     0.000006s :      3: substitution.graph_param_transform
85.80% :     0.000243s :      4: substitution.inline
 1.52% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.99% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.66% :     0.000008s :      2: substitution.replace_old_param
 4.80% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007327     2
85.73% :     0.006281s :      1: type_inference.infer
14.27% :     0.001046s :      1: type_inference.specialize
------[replace.]   0.000077     6
68.78% :     0.000053s :      4: replace.inline
31.22% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000253     6
95.12% :     0.000241s :      4: match.inline
 4.88% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   691
 1.55% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.92% :     0.000001s :     10: predicate.addn_check_dump
 1.48% :     0.000002s :     10: predicate.addn_zero_filter
 2.59% :     0.000003s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.11% :     0.000001s :     10: predicate.depend_value_elim
 0.98% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.06% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.12% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.78% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 6.30% :     0.000008s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.61% :     0.000002s :      6: predicate.less_batch_normalization
 1.79% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.54% :     0.000002s :     15: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.64% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.76% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 2.88% :     0.000004s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.51% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.83% :     0.000001s :      6: predicate.replace_old_param
 0.47% :     0.000001s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.05% :     0.000001s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.29% :     0.000002s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.15% :     0.000003s :     16: predicate.switch_defer_inline
 1.96% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.42% :     0.000010s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.61% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.29% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.42% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.54% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000813    12
43.40% :     0.000353s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.07% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.53% :     0.000402s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.053926    76
 0.12% :     0.000064s :      1: add_recomputation
 0.15% :     0.000080s :      1: auto_monad
 0.03% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.92% :     0.000499s :      1: bootstrap
 0.08% :     0.000041s :      1: cconv
 0.02% :     0.000011s :      1: convert_after_rewriter
 0.05% :     0.000026s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.05% :     0.000028s :      1: event_method
 0.03% :     0.000017s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
51.81% :     0.027942s :      1: jit_opt_a
 0.32% :     0.000175s :      1: jit_opt_after_cconv
 0.12% :     0.000062s :      1: jit_opt_b
 0.93% :     0.000504s :      1: loop_unroll
 1.68% :     0.000905s :      1: mutable_eliminate
 1.85% :     0.000995s :     26: opt.transform.jit_opt_a
 0.10% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000028s :      4: opt.transform.jit_opt_b
 0.03% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000023s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000035s :      4: opt.transform.symbol_engine_opt
 0.94% :     0.000507s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000050s :      1: py_interpret_to_execute
 0.05% :     0.000026s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000018s :      1: remove_dup_value
 9.17% :     0.004943s :      1: renormalize.infer
 0.96% :     0.000516s :      1: renormalize.specialize
 0.02% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000048s :      1: rewriter_after_opt_a
 0.15% :     0.000082s :      1: rewriter_before_opt_a
 0.18% :     0.000098s :      1: symbol_engine_optimizer
15.80% :     0.008518s :      1: task_emit
13.80% :     0.007443s :      1: type_inference
 0.14% :     0.000073s :      1: validate


TotalTime = 0.0234558, [33]
[bootstrap]: 0.0003929
[type_inference]: 0.00580697
[event_method]: 1.799e-05
[auto_monad]: 6.745e-05
[graph_reusing]: 6.73998e-06
[pre_auto_parallel]: 2.59999e-06
[py_interpret_to_execute]: 4.269e-05
[rewriter_before_opt_a]: 7.67e-05
[expand_dump_flag]: 3.71001e-06
[jit_opt_a]: 0.00691338, [2]
    [Cycle 1]: 0.0019313, [27]
        [switch_simplify]: 7.163e-05
        [loop_unroll]: 2.837e-05
        [a_1]: 0.00060913
        [with_stream_mark]: 2.202e-05
        [recompute_prepare]: 8.29002e-06
        [updatestate_depend_eliminate]: 3.63999e-06
        [updatestate_assign_eliminate]: 2.74999e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 5.92999e-06
        [updatestate_useless_node_eliminater]: 5.52001e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 3.29001e-06
        [get_grad_eliminate_]: 5.20999e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 1.56998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.596e-05
        [j_node_and_user_rematch]: 9.01998e-06
        [meta_fg_expand]: 2.17999e-06
        [replace_old_param]: 8.60999e-06
        [inline_without_move]: 5.44998e-06
        [renormalize]: 0.00085322
        [add_forward_monad_depend]: 8.62e-06
        [auto_monad_grad]: 2.79999e-06
        [auto_monad_eliminator]: 1.584e-05
        [cse]: 2.006e-05
        [replace_applicator]: 1.786e-05
    [Cycle 2]: 0.00035937, [27]
        [switch_simplify]: 5.96998e-06
        [loop_unroll]: 5.49998e-06
        [a_1]: 9.948e-05
        [with_stream_mark]: 1.239e-05
        [recompute_prepare]: 5.15001e-06
        [updatestate_depend_eliminate]: 3.00998e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 2.36998e-06
        [parameter_eliminate]: 1.12999e-06
        [specialize_transform]: 5.25001e-06
        [updatestate_useless_node_eliminater]: 5.42001e-06
        [accelerated_algorithm]: 5.64e-06
        [meta_shard_fg_expand]: 1.75001e-06
        [get_grad_eliminate_]: 5.48002e-06
        [merge_forward]: 3.53e-06
        [cell_reuse_recompute_pass]: 4.07003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.642e-05
        [j_node_and_user_rematch]: 8.87e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 8.72e-06
        [inline_without_move]: 5.17999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 5.07999e-06
        [cse]: 1.159e-05
        [replace_applicator]: 5.10999e-06
[py_interpret_to_execute_after_opt_a]: 1.695e-05
[rewriter_after_opt_a]: 4.087e-05
[convert_after_rewriter]: 6.83e-06
[order_py_execute_after_rewriter]: 4.86997e-06
[mutable_eliminate]: 0.00092082
[jit_opt_b]: 5.886e-05, [1]
    [Cycle 1]: 4.92e-05, [2]
        [frontend_op_eliminate]: 1.9e-05
        [inline_after_opt_a]: 1.789e-05
[cconv]: 3.235e-05
[loop_unroll]: 0.00062342
[jit_opt_after_cconv]: 0.00017852, [1]
    [Cycle 1]: 0.00017091, [11]
        [c_1]: 2.458e-05
        [parameter_eliminate]: 5.84999e-06
        [updatestate_depend_eliminate]: 8.99e-06
        [updatestate_assign_eliminate]: 3.8e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [cse]: 3.286e-05
        [call_graph_tuple_transform]: 2.375e-05
        [tuple_list_get_item_eliminator]: 6.51e-06
        [none_parameter_eliminate]: 1.67999e-06
        [renormalize]: 8.59989e-07
        [switch_simplify]: 5.87999e-06
[remove_dup_value]: 1.627e-05
[partial_unused_args_eliminate]: 2.63e-06
[environ_conv]: 6.05002e-06
[add_recomputation]: 5.711e-05
[cse_after_recomputation]: 2.279e-05, [1]
    [Cycle 1]: 1.601e-05, [1]
        [cse]: 9.34e-06
[auto_monad_reorder]: 1.842e-05
[get_jit_bprop_graph]: 2.31e-06
[rewriter_after_jit_bprop_graph]: 1.096e-05
[opt_after_jit_grad]: 0.00061594
[symbol_engine_optimizer]: 8.802e-05, [1]
    [Cycle 1]: 7.814e-05, [6]
        [build]: 5.91998e-06
        [elim_shapecalc]: 9.13002e-06
        [elim_not_effective]: 1.759e-05
        [opt_reshape]: 6.32001e-06
        [fold_const_symbol]: 8.72998e-06
        [renormalize]: 6.80011e-07
[validate]: 5.199e-05
[backend_pass]: 1.20001e-06
[task_emit]: 0.00709963
[execute]: 8.70999e-06
Sums
    bootstrap                            :     0.000393s :  2.17%
    type_inference                       :     0.005807s : 32.01%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000067s :  0.37%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.24%
    rewriter_before_opt_a                :     0.000077s :  0.42%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000078s :  0.43%
    jit_opt_a.loop_unroll                :     0.000034s :  0.19%
    jit_opt_a.a_1                        :     0.000709s :  3.91%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000853s :  4.70%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.12%
    jit_opt_a.cse                        :     0.000032s :  0.17%
    jit_opt_a.replace_applicator         :     0.000023s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.09%
    rewriter_after_opt_a                 :     0.000041s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000921s :  5.08%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.10%
    cconv                                :     0.000032s :  0.18%
    loop_unroll                          :     0.000623s :  3.44%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000033s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000057s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000018s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.06%
    opt_after_jit_grad                   :     0.000616s :  3.40%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000052s :  0.29%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007100s : 39.14%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000261    23
 0.90% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.44% :     0.000006s :      3: substitution.graph_param_transform
85.69% :     0.000224s :      4: substitution.inline
 1.38% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.24% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.90% :     0.000005s :      2: substitution.replace_old_param
 4.93% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005720     2
81.86% :     0.004682s :      1: type_inference.infer
18.14% :     0.001038s :      1: type_inference.specialize
------[replace.]   0.000079     6
64.69% :     0.000051s :      4: replace.inline
35.31% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000233     6
94.91% :     0.000221s :      4: match.inline
 5.09% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   691
 2.37% :     0.000003s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 2.35% :     0.000003s :     10: predicate.arithmetic_simplify
 1.28% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.92% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 0.94% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.70% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.05% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.30% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     10: predicate.environ_get_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.17% :     0.000002s :      6: predicate.get_grad_eliminate
 0.38% :     0.000001s :      3: predicate.graph_param_transform
 5.47% :     0.000007s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      6: predicate.less_batch_normalization
 1.72% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.96% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.54% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 0.93% :     0.000001s :     10: predicate.minmaximum_grad
 2.93% :     0.000004s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 1.89% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.40% :     0.000002s :     18: predicate.replace_applicator
 0.80% :     0.000001s :      6: predicate.replace_old_param
 0.40% :     0.000001s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.06% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.10% :     0.000001s :      6: predicate.special_op_eliminate
 0.86% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.49% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 1.97% :     0.000003s :     16: predicate.switch_defer_inline
 1.76% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.98% :     0.000011s :     53: predicate.switch_simplify
 1.31% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000002s :     10: predicate.transpose_eliminate
 1.20% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.96% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.99% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.03% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.43% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000754    12
49.06% :     0.000370s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.08% :     0.000046s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.86% :     0.000338s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025387    76
 0.24% :     0.000060s :      1: add_recomputation
 0.28% :     0.000071s :      1: auto_monad
 0.09% :     0.000022s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.64% :     0.000416s :      1: bootstrap
 0.14% :     0.000036s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.08% :     0.000021s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
27.25% :     0.006919s :      1: jit_opt_a
 0.72% :     0.000182s :      1: jit_opt_after_cconv
 0.24% :     0.000062s :      1: jit_opt_b
 2.50% :     0.000634s :      1: loop_unroll
 3.67% :     0.000931s :      1: mutable_eliminate
 3.73% :     0.000946s :     26: opt.transform.jit_opt_a
 0.23% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000029s :      4: opt.transform.jit_opt_b
 0.07% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000025s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000038s :      4: opt.transform.symbol_engine_opt
 2.47% :     0.000626s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000045s :      1: py_interpret_to_execute
 0.08% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 1.80% :     0.000458s :      1: renormalize.infer
 1.52% :     0.000387s :      1: renormalize.specialize
 0.05% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000044s :      1: rewriter_after_opt_a
 0.31% :     0.000079s :      1: rewriter_before_opt_a
 0.36% :     0.000092s :      1: symbol_engine_optimizer
28.03% :     0.007115s :      1: task_emit
22.93% :     0.005821s :      1: type_inference
 0.30% :     0.000077s :      1: validate


TotalTime = 0.0324098, [30]
[bootstrap]: 0.00043603
[type_inference]: 0.00561177
[event_method]: 2.097e-05
[auto_monad]: 7.182e-05
[graph_reusing]: 6.04999e-06
[pre_auto_parallel]: 2.52001e-06
[py_interpret_to_execute]: 4.709e-05
[rewriter_before_opt_a]: 8.42e-05
[expand_dump_flag]: 3.68e-06
[jit_opt_a]: 0.0232555, [2]
    [Cycle 1]: 0.00230645, [27]
        [switch_simplify]: 9.026e-05
        [loop_unroll]: 3.079e-05
        [a_1]: 0.00071249
        [with_stream_mark]: 2.976e-05
        [recompute_prepare]: 8.15e-06
        [updatestate_depend_eliminate]: 4.60001e-06
        [updatestate_assign_eliminate]: 3.76999e-06
        [updatestate_loads_eliminate]: 3.00002e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 6.79001e-06
        [updatestate_useless_node_eliminater]: 5.37999e-06
        [accelerated_algorithm]: 6.54001e-06
        [meta_shard_fg_expand]: 2.84001e-06
        [get_grad_eliminate_]: 6.56e-06
        [merge_forward]: 4.32e-06
        [cell_reuse_recompute_pass]: 1.63002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.553e-05
        [j_node_and_user_rematch]: 1.087e-05
        [meta_fg_expand]: 2.69001e-06
        [replace_old_param]: 1.188e-05
        [inline_without_move]: 6.26e-06
        [renormalize]: 0.00105767
        [add_forward_monad_depend]: 8.93002e-06
        [auto_monad_grad]: 2.99999e-06
        [auto_monad_eliminator]: 1.901e-05
        [cse]: 3.165e-05
        [replace_applicator]: 1.846e-05
    [Cycle 2]: 0.00035053, [27]
        [switch_simplify]: 6.84001e-06
        [loop_unroll]: 4.99998e-06
        [a_1]: 0.00010193
        [with_stream_mark]: 1.512e-05
        [recompute_prepare]: 4.63999e-06
        [updatestate_depend_eliminate]: 3.26001e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 2.08002e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 5.22e-06
        [updatestate_useless_node_eliminater]: 4.76002e-06
        [accelerated_algorithm]: 5.82001e-06
        [meta_shard_fg_expand]: 1.80001e-06
        [get_grad_eliminate_]: 4.95999e-06
        [merge_forward]: 3.06001e-06
        [cell_reuse_recompute_pass]: 1.98002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.435e-05
        [j_node_and_user_rematch]: 8.11002e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 8.12e-06
        [inline_without_move]: 5.77001e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 9.89996e-07
        [auto_monad_grad]: 1.27999e-06
        [auto_monad_eliminator]: 5.81e-06
        [cse]: 1.239e-05
        [replace_applicator]: 5.02999e-06
[py_interpret_to_execute_after_opt_a]: 1.676e-05
[rewriter_after_opt_a]: 4.283e-05
[convert_after_rewriter]: 1.015e-05
[order_py_execute_after_rewriter]: 5.75001e-06
[mutable_eliminate]: 0.00098074
[jit_opt_b]: 5.93e-05, [1]
    [Cycle 1]: 5.064e-05, [2]
        [frontend_op_eliminate]: 1.989e-05
        [inline_after_opt_a]: 1.789e-05
[cconv]: 3.339e-05
[loop_unroll]: 0.00050279
[jit_opt_after_cconv]: 0.00016871, [1]
    [Cycle 1]: 0.00016206, [11]
        [c_1]: 2.304e-05
        [parameter_eliminate]: 6.33e-06
        [updatestate_depend_eliminate]: 9.94001e-06
        [updatestate_assign_eliminate]: 2.65002e-06
        [updatestate_loads_eliminate]: 2.69001e-06
        [cse]: 2.669e-05
        [call_graph_tuple_transform]: 2.413e-05
        [tuple_list_get_item_eliminator]: 6.17999e-06
        [none_parameter_eliminate]: 1.97999e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 5.94e-06
[remove_dup_value]: 1.611e-05
[partial_unused_args_eliminate]: 2.59001e-06
[environ_conv]: 6.27001e-06
[add_recomputation]: 5.686e-05
[cse_after_recomputation]: 2.088e-05, [1]
    [Cycle 1]: 1.49e-05, [1]
        [cse]: 9.27001e-06
[auto_monad_reorder]: 1.599e-05
[get_jit_bprop_graph]: 2.51998e-06
[rewriter_after_jit_bprop_graph]: 9.25001e-06
[opt_after_jit_grad]: 0.00052206
[symbol_engine_optimizer]: 8.334e-05, [1]
    [Cycle 1]: 7.534e-05, [6]
        [build]: 5.30999e-06
        [elim_shapecalc]: 8.77e-06
        [elim_not_effective]: 1.477e-05
        [opt_reshape]: 6.11e-06
        [fold_const_symbol]: 1.051e-05
        [renormalize]: 4.89992e-07
[validate]: 4.742e-05
Sums
    bootstrap                            :     0.000436s :  3.93%
    type_inference                       :     0.005612s : 50.57%
    event_method                         :     0.000021s :  0.19%
    auto_monad                           :     0.000072s :  0.65%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000047s :  0.42%
    rewriter_before_opt_a                :     0.000084s :  0.76%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000097s :  0.88%
    jit_opt_a.loop_unroll                :     0.000036s :  0.32%
    jit_opt_a.a_1                        :     0.000814s :  7.34%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.40%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000020s :  0.18%
    jit_opt_a.inline_without_move        :     0.000012s :  0.11%
    jit_opt_a.renormalize                :     0.001058s :  9.53%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.22%
    jit_opt_a.cse                        :     0.000044s :  0.40%
    jit_opt_a.replace_applicator         :     0.000023s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.15%
    rewriter_after_opt_a                 :     0.000043s :  0.39%
    convert_after_rewriter               :     0.000010s :  0.09%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000981s :  8.84%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.16%
    cconv                                :     0.000033s :  0.30%
    loop_unroll                          :     0.000503s :  4.53%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.09%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000027s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000016s :  0.15%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000057s :  0.51%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.14%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.08%
    opt_after_jit_grad                   :     0.000522s :  4.70%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000047s :  0.43%


Time group info:
------[substitution.]   0.000326    23
 0.63% :     0.000002s :      2: substitution.elim_not_effective
 0.42% :     0.000001s :      2: substitution.fold_const_symbol
 2.01% :     0.000007s :      3: substitution.graph_param_transform
87.23% :     0.000284s :      4: substitution.inline
 1.40% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.57% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.29% :     0.000007s :      2: substitution.replace_old_param
 4.44% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005529     2
80.90% :     0.004473s :      1: type_inference.infer
19.10% :     0.001056s :      1: type_inference.specialize
------[replace.]   0.000088     6
67.71% :     0.000059s :      4: replace.inline
32.29% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000295     6
95.46% :     0.000282s :      4: match.inline
 4.54% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000144   691
 1.27% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.85% :     0.000001s :     10: predicate.addn_check_dump
 1.53% :     0.000002s :     10: predicate.addn_zero_filter
 2.99% :     0.000004s :     10: predicate.arithmetic_simplify
 1.54% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.86% :     0.000001s :     10: predicate.compare_switch_simplify
 1.98% :     0.000003s :     10: predicate.depend_value_elim
 1.14% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.23% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     10: predicate.environ_get_eliminate
 0.83% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.45% :     0.000002s :      6: predicate.get_grad_eliminate
 0.36% :     0.000001s :      3: predicate.graph_param_transform
 5.96% :     0.000009s :     22: predicate.inline
 0.71% :     0.000001s :      6: predicate.inline_without_move
 0.29% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.46% :     0.000002s :      6: predicate.less_batch_normalization
 1.90% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.54% :     0.000002s :     15: predicate.load_eliminater
 1.32% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.23% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.17% :     0.000002s :     10: predicate.minmaximum_grad
 2.04% :     0.000003s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 1.93% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.85% :     0.000003s :     10: predicate.reduce_eliminate
 1.15% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.72% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.47% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.61% :     0.000002s :     10: predicate.reshape_eliminate
 1.05% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.01% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.42% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000002s :      6: predicate.special_op_eliminate
 1.10% :     0.000002s :      6: predicate.specialize_transform
 1.55% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.61% :     0.000004s :     16: predicate.switch_defer_inline
 1.81% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.09% :     0.000010s :     53: predicate.switch_simplify
 1.19% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000002s :     10: predicate.transpose_eliminate
 1.76% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.51% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.97% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.31% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.50% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.04% :     0.000003s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000798    12
47.43% :     0.000379s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.02% :     0.000064s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.55% :     0.000356s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.034537    72
 0.17% :     0.000060s :      1: add_recomputation
 0.22% :     0.000075s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 1.33% :     0.000460s :      1: bootstrap
 0.11% :     0.000036s :      1: cconv
 0.04% :     0.000013s :      1: convert_after_rewriter
 0.07% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.07% :     0.000026s :      1: event_method
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
67.35% :     0.023260s :      1: jit_opt_a
 0.50% :     0.000172s :      1: jit_opt_after_cconv
 0.18% :     0.000062s :      1: jit_opt_b
 1.48% :     0.000513s :      1: loop_unroll
 2.88% :     0.000994s :      1: mutable_eliminate
 3.10% :     0.001071s :     26: opt.transform.jit_opt_a
 0.16% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000031s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.54% :     0.000531s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000050s :      1: py_interpret_to_execute
 0.06% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 1.69% :     0.000584s :      1: renormalize.infer
 1.34% :     0.000463s :      1: renormalize.specialize
 0.03% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000046s :      1: rewriter_after_opt_a
 0.25% :     0.000088s :      1: rewriter_before_opt_a
 0.25% :     0.000086s :      1: symbol_engine_optimizer
16.31% :     0.005632s :      1: type_inference


TotalTime = 0.0505013, [33]
[bootstrap]: 0.00041967
[type_inference]: 0.00535528
[event_method]: 2.091e-05
[auto_monad]: 6.69e-05
[graph_reusing]: 6.22001e-06
[pre_auto_parallel]: 3.55998e-06
[py_interpret_to_execute]: 4.208e-05
[rewriter_before_opt_a]: 7.386e-05
[expand_dump_flag]: 3.15998e-06
[jit_opt_a]: 0.0347683, [2]
    [Cycle 1]: 0.00197316, [27]
        [switch_simplify]: 6.74e-05
        [loop_unroll]: 3.186e-05
        [a_1]: 0.00059622
        [with_stream_mark]: 2.553e-05
        [recompute_prepare]: 7.26999e-06
        [updatestate_depend_eliminate]: 3.73001e-06
        [updatestate_assign_eliminate]: 2.78998e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 2.02999e-06
        [specialize_transform]: 5.58002e-06
        [updatestate_useless_node_eliminater]: 5.27999e-06
        [accelerated_algorithm]: 5.17e-06
        [meta_shard_fg_expand]: 2.19001e-06
        [get_grad_eliminate_]: 5.09998e-06
        [merge_forward]: 3.38999e-06
        [cell_reuse_recompute_pass]: 1.08001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.382e-05
        [j_node_and_user_rematch]: 9.24e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 9.68997e-06
        [inline_without_move]: 5.08002e-06
        [renormalize]: 0.00090597
        [add_forward_monad_depend]: 6.88e-06
        [auto_monad_grad]: 2.64999e-06
        [auto_monad_eliminator]: 1.589e-05
        [cse]: 3.087e-05
        [replace_applicator]: 1.312e-05
    [Cycle 2]: 0.0287316, [27]
        [switch_simplify]: 5.79999e-06
        [loop_unroll]: 4.74e-06
        [a_1]: 0.0282669
        [with_stream_mark]: 3.292e-05
        [recompute_prepare]: 1.084e-05
        [updatestate_depend_eliminate]: 4.85999e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 2.82002e-06
        [specialize_transform]: 5.74e-06
        [updatestate_useless_node_eliminater]: 5.69e-06
        [accelerated_algorithm]: 8.09002e-06
        [meta_shard_fg_expand]: 3.30998e-06
        [get_grad_eliminate_]: 5.40001e-06
        [merge_forward]: 4.01001e-06
        [cell_reuse_recompute_pass]: 4.92e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.432e-05
        [j_node_and_user_rematch]: 1.038e-05
        [meta_fg_expand]: 2.97002e-06
        [replace_old_param]: 1.158e-05
        [inline_without_move]: 5.46998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 5.76e-06
        [auto_monad_grad]: 2.68003e-06
        [auto_monad_eliminator]: 1.607e-05
        [cse]: 3.85e-05
        [replace_applicator]: 7.49002e-06
[py_interpret_to_execute_after_opt_a]: 2.22e-05
[rewriter_after_opt_a]: 4.145e-05
[convert_after_rewriter]: 6.77002e-06
[order_py_execute_after_rewriter]: 4.88001e-06
[mutable_eliminate]: 0.00088264
[jit_opt_b]: 6.092e-05, [1]
    [Cycle 1]: 5.07e-05, [2]
        [frontend_op_eliminate]: 2.042e-05
        [inline_after_opt_a]: 1.701e-05
[cconv]: 3.371e-05
[loop_unroll]: 0.00044856
[jit_opt_after_cconv]: 0.00015097, [1]
    [Cycle 1]: 0.00014363, [11]
        [c_1]: 2.133e-05
        [parameter_eliminate]: 3.3e-06
        [updatestate_depend_eliminate]: 6.25002e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.21998e-06
        [cse]: 2.283e-05
        [call_graph_tuple_transform]: 2.116e-05
        [tuple_list_get_item_eliminator]: 5.60001e-06
        [none_parameter_eliminate]: 1.40999e-06
        [renormalize]: 8.09989e-07
        [switch_simplify]: 4.97999e-06
[remove_dup_value]: 1.549e-05
[partial_unused_args_eliminate]: 2.46e-06
[environ_conv]: 6.89999e-06
[add_recomputation]: 5.686e-05
[cse_after_recomputation]: 2.225e-05, [1]
    [Cycle 1]: 1.629e-05, [1]
        [cse]: 9.94999e-06
[auto_monad_reorder]: 1.585e-05
[get_jit_bprop_graph]: 2.34999e-06
[rewriter_after_jit_bprop_graph]: 8.64e-06
[opt_after_jit_grad]: 0.00049708
[symbol_engine_optimizer]: 7.809e-05, [1]
    [Cycle 1]: 7.189e-05, [6]
        [build]: 4.90001e-06
        [elim_shapecalc]: 8.3e-06
        [elim_not_effective]: 1.481e-05
        [opt_reshape]: 5.61998e-06
        [fold_const_symbol]: 8.66002e-06
        [renormalize]: 3.50003e-07
[validate]: 4.223e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.00707651
[execute]: 1.017e-05
Sums
    bootstrap                            :     0.000420s :  0.92%
    type_inference                       :     0.005355s : 11.74%
    event_method                         :     0.000021s :  0.05%
    auto_monad                           :     0.000067s :  0.15%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000004s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.09%
    rewriter_before_opt_a                :     0.000074s :  0.16%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000073s :  0.16%
    jit_opt_a.loop_unroll                :     0.000037s :  0.08%
    jit_opt_a.a_1                        :     0.028863s : 63.26%
    jit_opt_a.with_stream_mark           :     0.000058s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.02%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000021s :  0.05%
    jit_opt_a.inline_without_move        :     0.000011s :  0.02%
    jit_opt_a.renormalize                :     0.000906s :  1.99%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000032s :  0.07%
    jit_opt_a.cse                        :     0.000069s :  0.15%
    jit_opt_a.replace_applicator         :     0.000021s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000022s :  0.05%
    rewriter_after_opt_a                 :     0.000041s :  0.09%
    convert_after_rewriter               :     0.000007s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000883s :  1.93%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.04%
    cconv                                :     0.000034s :  0.07%
    loop_unroll                          :     0.000449s :  0.98%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.05%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000023s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.01%
    remove_dup_value                     :     0.000015s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000057s :  0.12%
    cse_after_recomputation.cse          :     0.000010s :  0.02%
    auto_monad_reorder                   :     0.000016s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.02%
    opt_after_jit_grad                   :     0.000497s :  1.09%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000042s :  0.09%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007077s : 15.51%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000261    23
 0.94% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 2.26% :     0.000006s :      3: substitution.graph_param_transform
84.64% :     0.000221s :      4: substitution.inline
 1.91% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.15% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.98% :     0.000008s :      2: substitution.replace_old_param
 4.66% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005267     2
82.05% :     0.004322s :      1: type_inference.infer
17.95% :     0.000945s :      1: type_inference.specialize
------[replace.]   0.000077     6
67.31% :     0.000052s :      4: replace.inline
32.69% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000230     6
95.18% :     0.000219s :      4: match.inline
 4.82% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000147   691
 1.52% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.03% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.84% :     0.000001s :     10: predicate.addn_check_dump
 1.23% :     0.000002s :     10: predicate.addn_zero_filter
 2.74% :     0.000004s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.81% :     0.000001s :     10: predicate.compare_switch_simplify
 1.24% :     0.000002s :     10: predicate.depend_value_elim
 0.99% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.23% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.33% :     0.000002s :     10: predicate.environ_get_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 6.90% :     0.000010s :     22: predicate.inline
 0.69% :     0.000001s :      6: predicate.inline_without_move
 0.29% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.61% :     0.000002s :     15: predicate.load_eliminater
 1.19% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.22% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.82% :     0.000001s :     10: predicate.merge_addn
 0.82% :     0.000001s :     10: predicate.minmaximum_grad
 1.53% :     0.000002s :      3: predicate.mutable_eliminate
 0.35% :     0.000001s :      3: predicate.opt_reshape
 1.90% :     0.000003s :     15: predicate.partial_eliminate
 1.28% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     10: predicate.reduce_eliminate
 1.74% :     0.000003s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.25% :     0.000002s :     18: predicate.replace_applicator
 0.90% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.08% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 0.75% :     0.000001s :      6: predicate.specialize_transform
 1.50% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.61% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 6.67% :     0.000010s :     16: predicate.switch_defer_inline
 2.09% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.67% :     0.000010s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.43% :     0.000002s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.17% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.28% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.54% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.89% :     0.000003s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000693    12
49.30% :     0.000342s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.28% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.42% :     0.000301s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.052518    76
 0.11% :     0.000060s :      1: add_recomputation
 0.13% :     0.000071s :      1: auto_monad
 0.04% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.84% :     0.000439s :      1: bootstrap
 0.07% :     0.000037s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.05% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.05% :     0.000026s :      1: event_method
 0.03% :     0.000015s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
66.21% :     0.034773s :      1: jit_opt_a
 0.29% :     0.000154s :      1: jit_opt_after_cconv
 0.12% :     0.000064s :      1: jit_opt_b
 0.87% :     0.000457s :      1: loop_unroll
 1.70% :     0.000894s :      1: mutable_eliminate
 1.95% :     0.001026s :     26: opt.transform.jit_opt_a
 0.09% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000030s :      4: opt.transform.jit_opt_b
 0.03% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000034s :      4: opt.transform.symbol_engine_opt
 0.96% :     0.000505s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.09% :     0.000045s :      1: py_interpret_to_execute
 0.05% :     0.000025s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000018s :      1: remove_dup_value
 0.97% :     0.000509s :      1: renormalize.infer
 0.74% :     0.000387s :      1: renormalize.specialize
 0.02% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000045s :      1: rewriter_after_opt_a
 0.15% :     0.000077s :      1: rewriter_before_opt_a
 0.15% :     0.000081s :      1: symbol_engine_optimizer
13.51% :     0.007097s :      1: task_emit
10.23% :     0.005373s :      1: type_inference
 0.12% :     0.000064s :      1: validate


TotalTime = 0.0456806, [33]
[bootstrap]: 0.00035291
[type_inference]: 0.00656198
[event_method]: 1.848e-05
[auto_monad]: 6.87e-05
[graph_reusing]: 6.06e-06
[pre_auto_parallel]: 2.74999e-06
[py_interpret_to_execute]: 4.061e-05
[rewriter_before_opt_a]: 7.877e-05
[expand_dump_flag]: 3.18998e-06
[jit_opt_a]: 0.00648101, [2]
    [Cycle 1]: 0.00204331, [27]
        [switch_simplify]: 7.118e-05
        [loop_unroll]: 3.212e-05
        [a_1]: 0.00065229
        [with_stream_mark]: 2.408e-05
        [recompute_prepare]: 8.15e-06
        [updatestate_depend_eliminate]: 4.43999e-06
        [updatestate_assign_eliminate]: 3.20998e-06
        [updatestate_loads_eliminate]: 2.89999e-06
        [parameter_eliminate]: 2.02999e-06
        [specialize_transform]: 5.84999e-06
        [updatestate_useless_node_eliminater]: 6.59999e-06
        [accelerated_algorithm]: 6.12001e-06
        [meta_shard_fg_expand]: 1.99999e-06
        [get_grad_eliminate_]: 7.23e-06
        [merge_forward]: 4.93001e-06
        [cell_reuse_recompute_pass]: 1.49998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.585e-05
        [j_node_and_user_rematch]: 9.82001e-06
        [meta_fg_expand]: 2.43e-06
        [replace_old_param]: 9.91e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.00082113
        [add_forward_monad_depend]: 6.11e-06
        [auto_monad_grad]: 2.31998e-06
        [auto_monad_eliminator]: 1.776e-05
        [cse]: 3.415e-05
        [replace_applicator]: 1.721e-05
    [Cycle 2]: 0.00035594, [27]
        [switch_simplify]: 5.97999e-06
        [loop_unroll]: 5.12e-06
        [a_1]: 9.351e-05
        [with_stream_mark]: 1.427e-05
        [recompute_prepare]: 5.49998e-06
        [updatestate_depend_eliminate]: 2.68998e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 5.22e-06
        [updatestate_useless_node_eliminater]: 5.35999e-06
        [accelerated_algorithm]: 5.59e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 4.58999e-06
        [merge_forward]: 3.66001e-06
        [cell_reuse_recompute_pass]: 2.99999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.605e-05
        [j_node_and_user_rematch]: 8.3e-06
        [meta_fg_expand]: 1.78002e-06
        [replace_old_param]: 8.35001e-06
        [inline_without_move]: 5.59e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.49e-06
        [auto_monad_grad]: 1.75001e-06
        [auto_monad_eliminator]: 6.79001e-06
        [cse]: 1.246e-05
        [replace_applicator]: 5.76e-06
[py_interpret_to_execute_after_opt_a]: 1.624e-05
[rewriter_after_opt_a]: 4.237e-05
[convert_after_rewriter]: 7.78999e-06
[order_py_execute_after_rewriter]: 5.15001e-06
[mutable_eliminate]: 0.00080038
[jit_opt_b]: 5.651e-05, [1]
    [Cycle 1]: 4.738e-05, [2]
        [frontend_op_eliminate]: 1.799e-05
        [inline_after_opt_a]: 1.61e-05
[cconv]: 3.86e-05
[loop_unroll]: 0.00051557
[jit_opt_after_cconv]: 0.00018018, [1]
    [Cycle 1]: 0.00017237, [11]
        [c_1]: 2.227e-05
        [parameter_eliminate]: 5.12999e-06
        [updatestate_depend_eliminate]: 9.56998e-06
        [updatestate_assign_eliminate]: 2.84999e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [cse]: 3.432e-05
        [call_graph_tuple_transform]: 2.574e-05
        [tuple_list_get_item_eliminator]: 5.71e-06
        [none_parameter_eliminate]: 1.59998e-06
        [renormalize]: 7.50006e-07
        [switch_simplify]: 5.52999e-06
[remove_dup_value]: 1.839e-05
[partial_unused_args_eliminate]: 2.83e-06
[environ_conv]: 6.50002e-06
[add_recomputation]: 6.056e-05
[cse_after_recomputation]: 2.436e-05, [1]
    [Cycle 1]: 1.808e-05, [1]
        [cse]: 1.065e-05
[auto_monad_reorder]: 1.626e-05
[get_jit_bprop_graph]: 2.56998e-06
[rewriter_after_jit_bprop_graph]: 9.24998e-06
[opt_after_jit_grad]: 0.0005697
[symbol_engine_optimizer]: 0.00011825, [1]
    [Cycle 1]: 0.00011006, [6]
        [build]: 4.97e-06
        [elim_shapecalc]: 4.269e-05
        [elim_not_effective]: 1.536e-05
        [opt_reshape]: 6.31998e-06
        [fold_const_symbol]: 8.59998e-06
        [renormalize]: 8.60018e-07
[validate]: 5.042e-05
[backend_pass]: 1.12999e-06
[task_emit]: 0.029241
[execute]: 1.064e-05
Sums
    bootstrap                            :     0.000353s :  0.87%
    type_inference                       :     0.006562s : 16.09%
    event_method                         :     0.000018s :  0.05%
    auto_monad                           :     0.000069s :  0.17%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.10%
    rewriter_before_opt_a                :     0.000079s :  0.19%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000077s :  0.19%
    jit_opt_a.loop_unroll                :     0.000037s :  0.09%
    jit_opt_a.a_1                        :     0.000746s :  1.83%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.03%
    jit_opt_a.merge_forward              :     0.000009s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000018s :  0.04%
    jit_opt_a.inline_without_move        :     0.000011s :  0.03%
    jit_opt_a.renormalize                :     0.000821s :  2.01%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.06%
    jit_opt_a.cse                        :     0.000047s :  0.11%
    jit_opt_a.replace_applicator         :     0.000023s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.04%
    rewriter_after_opt_a                 :     0.000042s :  0.10%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000800s :  1.96%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.04%
    cconv                                :     0.000039s :  0.09%
    loop_unroll                          :     0.000516s :  1.26%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.05%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000034s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.01%
    remove_dup_value                     :     0.000018s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000061s :  0.15%
    cse_after_recomputation.cse          :     0.000011s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.04%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.02%
    opt_after_jit_grad                   :     0.000570s :  1.40%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000043s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000050s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.029241s : 71.68%
    execute                              :     0.000011s :  0.03%


Time group info:
------[substitution.]   0.000272    23
 0.77% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.36% :     0.000006s :      3: substitution.graph_param_transform
85.80% :     0.000234s :      4: substitution.inline
 1.42% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.70% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.16% :     0.000006s :      2: substitution.replace_old_param
 5.29% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006488     2
85.67% :     0.005558s :      1: type_inference.infer
14.33% :     0.000930s :      1: type_inference.specialize
------[replace.]   0.000074     6
69.54% :     0.000052s :      4: replace.inline
30.46% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000244     6
94.60% :     0.000231s :      4: match.inline
 5.40% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   691
 1.51% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.51% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 2.51% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000002s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.41% :     0.000002s :     10: predicate.depend_value_elim
 0.99% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.11% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.02% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000001s :      3: predicate.elim_not_effective
 0.76% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.99% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000002s :     10: predicate.environ_get_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000001s :      3: predicate.graph_param_transform
 6.33% :     0.000008s :     22: predicate.inline
 1.07% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.37% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 2.44% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.02% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.70% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 3.78% :     0.000005s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.00% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.76% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.47% :     0.000001s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.97% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.07% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 0.86% :     0.000001s :      6: predicate.specialize_transform
 1.20% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 1.81% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.06% :     0.000011s :     53: predicate.switch_simplify
 1.16% :     0.000002s :     10: predicate.tile_eliminate
 1.04% :     0.000001s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.44% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.29% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.20% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.48% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.17% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.39% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000678    12
47.25% :     0.000320s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.51% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.24% :     0.000300s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.047619    76
 0.14% :     0.000064s :      1: add_recomputation
 0.15% :     0.000072s :      1: auto_monad
 0.04% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.79% :     0.000376s :      1: bootstrap
 0.09% :     0.000042s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.06% :     0.000027s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.05% :     0.000024s :      1: event_method
 0.04% :     0.000018s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
13.62% :     0.006485s :      1: jit_opt_a
 0.39% :     0.000184s :      1: jit_opt_after_cconv
 0.13% :     0.000060s :      1: jit_opt_b
 1.11% :     0.000527s :      1: loop_unroll
 1.72% :     0.000818s :      1: mutable_eliminate
 2.08% :     0.000989s :     26: opt.transform.jit_opt_a
 0.12% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000027s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000025s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.22% :     0.000583s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000043s :      1: py_interpret_to_execute
 0.04% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000021s :      1: remove_dup_value
 0.96% :     0.000456s :      1: renormalize.infer
 0.75% :     0.000356s :      1: renormalize.specialize
 0.02% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000047s :      1: rewriter_after_opt_a
 0.17% :     0.000082s :      1: rewriter_before_opt_a
 0.26% :     0.000121s :      1: symbol_engine_optimizer
61.46% :     0.029265s :      1: task_emit
13.82% :     0.006582s :      1: type_inference
 0.16% :     0.000078s :      1: validate


TotalTime = 0.0189028, [30]
[bootstrap]: 0.00052647
[type_inference]: 0.00722081
[event_method]: 2.569e-05
[auto_monad]: 8.077e-05
[graph_reusing]: 7.6e-06
[pre_auto_parallel]: 3.65e-06
[py_interpret_to_execute]: 5.651e-05
[rewriter_before_opt_a]: 8.833e-05
[expand_dump_flag]: 3.66001e-06
[jit_opt_a]: 0.00786354, [2]
    [Cycle 1]: 0.00223216, [27]
        [switch_simplify]: 7.407e-05
        [loop_unroll]: 3.303e-05
        [a_1]: 0.00067272
        [with_stream_mark]: 2.496e-05
        [recompute_prepare]: 1.121e-05
        [updatestate_depend_eliminate]: 4.1e-06
        [updatestate_assign_eliminate]: 2.86e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [parameter_eliminate]: 2.91999e-06
        [specialize_transform]: 6.84001e-06
        [updatestate_useless_node_eliminater]: 5.70001e-06
        [accelerated_algorithm]: 6.53e-06
        [meta_shard_fg_expand]: 2.96001e-06
        [get_grad_eliminate_]: 6.06e-06
        [merge_forward]: 3.98001e-06
        [cell_reuse_recompute_pass]: 1.85001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.738e-05
        [j_node_and_user_rematch]: 1.098e-05
        [meta_fg_expand]: 3.76001e-06
        [replace_old_param]: 1.094e-05
        [inline_without_move]: 5.91998e-06
        [renormalize]: 0.00098518
        [add_forward_monad_depend]: 7.99002e-06
        [auto_monad_grad]: 2.35002e-06
        [auto_monad_eliminator]: 1.839e-05
        [cse]: 3.327e-05
        [replace_applicator]: 2.254e-05
    [Cycle 2]: 0.00039564, [27]
        [switch_simplify]: 6.38e-06
        [loop_unroll]: 5.62999e-06
        [a_1]: 0.00011038
        [with_stream_mark]: 1.556e-05
        [recompute_prepare]: 5.99999e-06
        [updatestate_depend_eliminate]: 3.84002e-06
        [updatestate_assign_eliminate]: 2.73998e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 5.46e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 6.44999e-06
        [meta_shard_fg_expand]: 2.20002e-06
        [get_grad_eliminate_]: 5.46e-06
        [merge_forward]: 3.16999e-06
        [cell_reuse_recompute_pass]: 3.19001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.839e-05
        [j_node_and_user_rematch]: 1.009e-05
        [meta_fg_expand]: 3.31001e-06
        [replace_old_param]: 8.65001e-06
        [inline_without_move]: 1.78e-05
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 2.17001e-06
        [auto_monad_grad]: 8.79983e-07
        [auto_monad_eliminator]: 7.21001e-06
        [cse]: 1.472e-05
        [replace_applicator]: 6.51e-06
[py_interpret_to_execute_after_opt_a]: 1.971e-05
[rewriter_after_opt_a]: 4.595e-05
[convert_after_rewriter]: 8.13001e-06
[order_py_execute_after_rewriter]: 5.25999e-06
[mutable_eliminate]: 0.00091037
[jit_opt_b]: 6.857e-05, [1]
    [Cycle 1]: 5.678e-05, [2]
        [frontend_op_eliminate]: 2.276e-05
        [inline_after_opt_a]: 1.896e-05
[cconv]: 3.525e-05
[loop_unroll]: 0.00058641
[jit_opt_after_cconv]: 0.00016426, [1]
    [Cycle 1]: 0.00015704, [11]
        [c_1]: 2.385e-05
        [parameter_eliminate]: 4.72998e-06
        [updatestate_depend_eliminate]: 7.09001e-06
        [updatestate_assign_eliminate]: 3.13e-06
        [updatestate_loads_eliminate]: 2.76e-06
        [cse]: 2.477e-05
        [call_graph_tuple_transform]: 2.329e-05
        [tuple_list_get_item_eliminator]: 6.52001e-06
        [none_parameter_eliminate]: 1.62001e-06
        [renormalize]: 8.39995e-07
        [switch_simplify]: 6.46e-06
[remove_dup_value]: 1.634e-05
[partial_unused_args_eliminate]: 2.24001e-06
[environ_conv]: 6.30002e-06
[add_recomputation]: 5.786e-05
[cse_after_recomputation]: 2.333e-05, [1]
    [Cycle 1]: 1.687e-05, [1]
        [cse]: 1.007e-05
[auto_monad_reorder]: 1.639e-05
[get_jit_bprop_graph]: 1.42e-06
[rewriter_after_jit_bprop_graph]: 9.13002e-06
[opt_after_jit_grad]: 0.00064896
[symbol_engine_optimizer]: 9.394e-05, [1]
    [Cycle 1]: 8.437e-05, [6]
        [build]: 5.32999e-06
        [elim_shapecalc]: 8.82e-06
        [elim_not_effective]: 1.91e-05
        [opt_reshape]: 6.96999e-06
        [fold_const_symbol]: 1.083e-05
        [renormalize]: 6.99976e-07
[validate]: 4.745e-05
Sums
    bootstrap                            :     0.000526s :  4.08%
    type_inference                       :     0.007221s : 55.99%
    event_method                         :     0.000026s :  0.20%
    auto_monad                           :     0.000081s :  0.63%
    graph_reusing                        :     0.000008s :  0.06%
    pre_auto_parallel                    :     0.000004s :  0.03%
    py_interpret_to_execute              :     0.000057s :  0.44%
    rewriter_before_opt_a                :     0.000088s :  0.68%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000080s :  0.62%
    jit_opt_a.loop_unroll                :     0.000039s :  0.30%
    jit_opt_a.a_1                        :     0.000783s :  6.07%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.05%
    jit_opt_a.replace_old_param          :     0.000020s :  0.15%
    jit_opt_a.inline_without_move        :     0.000024s :  0.18%
    jit_opt_a.renormalize                :     0.000985s :  7.64%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.20%
    jit_opt_a.cse                        :     0.000048s :  0.37%
    jit_opt_a.replace_applicator         :     0.000029s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000020s :  0.15%
    rewriter_after_opt_a                 :     0.000046s :  0.36%
    convert_after_rewriter               :     0.000008s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000910s :  7.06%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.15%
    cconv                                :     0.000035s :  0.27%
    loop_unroll                          :     0.000586s :  4.55%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000025s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000016s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000058s :  0.45%
    cse_after_recomputation.cse          :     0.000010s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.13%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.07%
    opt_after_jit_grad                   :     0.000649s :  5.03%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000047s :  0.37%


Time group info:
------[substitution.]   0.000293    23
 0.77% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 2.03% :     0.000006s :      3: substitution.graph_param_transform
85.48% :     0.000250s :      4: substitution.inline
 1.85% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.18% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000006s :      2: substitution.replace_old_param
 5.23% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007105     2
81.82% :     0.005813s :      1: type_inference.infer
18.18% :     0.001291s :      1: type_inference.specialize
------[replace.]   0.000085     6
65.94% :     0.000056s :      4: replace.inline
34.06% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000261     6
94.55% :     0.000247s :      4: match.inline
 5.45% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000148   691
 1.25% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.46% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.84% :     0.000001s :     10: predicate.addn_check_dump
 1.85% :     0.000003s :     10: predicate.addn_zero_filter
 2.80% :     0.000004s :     10: predicate.arithmetic_simplify
 1.38% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.85% :     0.000001s :     10: predicate.compare_switch_simplify
 1.78% :     0.000003s :     10: predicate.depend_value_elim
 1.13% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.66% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.54% :     0.000002s :     10: predicate.environ_get_eliminate
 1.30% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.86% :     0.000009s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.28% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.86% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.95% :     0.000003s :     15: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.59% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.87% :     0.000001s :     10: predicate.merge_addn
 1.19% :     0.000002s :     10: predicate.minmaximum_grad
 1.94% :     0.000003s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 1.90% :     0.000003s :     15: predicate.partial_eliminate
 1.38% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.47% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.20% :     0.000000s :      3: predicate.reset_defer_inline
 1.71% :     0.000003s :     10: predicate.reshape_eliminate
 1.20% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.04% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.39% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000002s :      6: predicate.special_op_eliminate
 0.82% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.07% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000004s :     16: predicate.switch_defer_inline
 1.61% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.22% :     0.000012s :     53: predicate.switch_simplify
 1.40% :     0.000002s :     10: predicate.tile_eliminate
 1.06% :     0.000002s :     10: predicate.transpose_eliminate
 1.56% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.92% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.12% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.27% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.41% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.37% :     0.000002s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000958    12
50.51% :     0.000484s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.43% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.06% :     0.000422s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021019    72
 0.29% :     0.000061s :      1: add_recomputation
 0.41% :     0.000085s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 2.67% :     0.000561s :      1: bootstrap
 0.18% :     0.000038s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000010s :      1: environ_conv
 0.15% :     0.000032s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000010s :      1: graph_reusing
37.44% :     0.007870s :      1: jit_opt_a
 0.80% :     0.000167s :      1: jit_opt_after_cconv
 0.34% :     0.000072s :      1: jit_opt_b
 2.84% :     0.000596s :      1: loop_unroll
 4.41% :     0.000927s :      1: mutable_eliminate
 5.04% :     0.001060s :     26: opt.transform.jit_opt_a
 0.27% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000033s :      4: opt.transform.jit_opt_b
 0.08% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000042s :      4: opt.transform.symbol_engine_opt
 3.16% :     0.000664s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000006s :      1: pre_auto_parallel
 0.28% :     0.000060s :      1: py_interpret_to_execute
 0.11% :     0.000023s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000019s :      1: remove_dup_value
 2.44% :     0.000513s :      1: renormalize.infer
 2.20% :     0.000463s :      1: renormalize.specialize
 0.05% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000049s :      1: rewriter_after_opt_a
 0.44% :     0.000093s :      1: rewriter_before_opt_a
 0.47% :     0.000098s :      1: symbol_engine_optimizer
34.51% :     0.007253s :      1: type_inference


TotalTime = 0.0331281, [33]
[bootstrap]: 0.00038661
[type_inference]: 0.00493497
[event_method]: 1.972e-05
[auto_monad]: 6.142e-05
[graph_reusing]: 5.65001e-06
[pre_auto_parallel]: 2.37999e-06
[py_interpret_to_execute]: 3.846e-05
[rewriter_before_opt_a]: 7.22e-05
[expand_dump_flag]: 3.06999e-06
[jit_opt_a]: 0.0183805, [2]
    [Cycle 1]: 0.00186919, [27]
        [switch_simplify]: 6.527e-05
        [loop_unroll]: 2.89e-05
        [a_1]: 0.00056875
        [with_stream_mark]: 2.155e-05
        [recompute_prepare]: 7.12002e-06
        [updatestate_depend_eliminate]: 3.94002e-06
        [updatestate_assign_eliminate]: 2.81999e-06
        [updatestate_loads_eliminate]: 2.76e-06
        [parameter_eliminate]: 1.77999e-06
        [specialize_transform]: 6.01e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 5.31998e-06
        [meta_shard_fg_expand]: 2.01998e-06
        [get_grad_eliminate_]: 5.33002e-06
        [merge_forward]: 3.66999e-06
        [cell_reuse_recompute_pass]: 1.34998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.501e-05
        [j_node_and_user_rematch]: 8.64998e-06
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 9.41998e-06
        [inline_without_move]: 5.07e-06
        [renormalize]: 0.00083395
        [add_forward_monad_depend]: 6.19999e-06
        [auto_monad_grad]: 2.48e-06
        [auto_monad_eliminator]: 1.658e-05
        [cse]: 3.076e-05
        [replace_applicator]: 1.528e-05
    [Cycle 2]: 0.00033464, [27]
        [switch_simplify]: 5.90002e-06
        [loop_unroll]: 5.42001e-06
        [a_1]: 9.289e-05
        [with_stream_mark]: 1.2e-05
        [recompute_prepare]: 5.56998e-06
        [updatestate_depend_eliminate]: 2.91999e-06
        [updatestate_assign_eliminate]: 2.10002e-06
        [updatestate_loads_eliminate]: 2.03002e-06
        [parameter_eliminate]: 1.40001e-06
        [specialize_transform]: 5.04998e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.30999e-06
        [meta_shard_fg_expand]: 2.03002e-06
        [get_grad_eliminate_]: 4.54002e-06
        [merge_forward]: 3.46999e-06
        [cell_reuse_recompute_pass]: 2.61e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.478e-05
        [j_node_and_user_rematch]: 7.49002e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 7.72002e-06
        [inline_without_move]: 4.50999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 8.99978e-07
        [auto_monad_eliminator]: 6.59999e-06
        [cse]: 1.068e-05
        [replace_applicator]: 5.10001e-06
[py_interpret_to_execute_after_opt_a]: 1.61e-05
[rewriter_after_opt_a]: 3.804e-05
[convert_after_rewriter]: 6.93998e-06
[order_py_execute_after_rewriter]: 4.72e-06
[mutable_eliminate]: 0.00081311
[jit_opt_b]: 5.633e-05, [1]
    [Cycle 1]: 4.778e-05, [2]
        [frontend_op_eliminate]: 1.946e-05
        [inline_after_opt_a]: 1.594e-05
[cconv]: 3.191e-05
[loop_unroll]: 0.0004986
[jit_opt_after_cconv]: 0.00016071, [1]
    [Cycle 1]: 0.0001538, [11]
        [c_1]: 2.134e-05
        [parameter_eliminate]: 5.42999e-06
        [updatestate_depend_eliminate]: 7.83999e-06
        [updatestate_assign_eliminate]: 3.01999e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [cse]: 2.891e-05
        [call_graph_tuple_transform]: 2.066e-05
        [tuple_list_get_item_eliminator]: 5.20999e-06
        [none_parameter_eliminate]: 1.69e-06
        [renormalize]: 4.60015e-07
        [switch_simplify]: 5.27999e-06
[remove_dup_value]: 1.599e-05
[partial_unused_args_eliminate]: 1.90001e-06
[environ_conv]: 5.40001e-06
[add_recomputation]: 5.32e-05
[cse_after_recomputation]: 2.178e-05, [1]
    [Cycle 1]: 1.565e-05, [1]
        [cse]: 9.49999e-06
[auto_monad_reorder]: 1.536e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 7.26999e-06
[opt_after_jit_grad]: 0.00051934
[symbol_engine_optimizer]: 7.781e-05, [1]
    [Cycle 1]: 6.971e-05, [6]
        [build]: 4.22e-06
        [elim_shapecalc]: 7.57998e-06
        [elim_not_effective]: 1.351e-05
        [opt_reshape]: 5.49e-06
        [fold_const_symbol]: 9.59e-06
        [renormalize]: 8.89995e-07
[validate]: 4.341e-05
[backend_pass]: 1.13001e-06
[task_emit]: 0.00660414
[execute]: 7.87998e-06
Sums
    bootstrap                            :     0.000387s :  2.37%
    type_inference                       :     0.004935s : 30.29%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000061s :  0.38%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000038s :  0.24%
    rewriter_before_opt_a                :     0.000072s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000071s :  0.44%
    jit_opt_a.loop_unroll                :     0.000034s :  0.21%
    jit_opt_a.a_1                        :     0.000662s :  4.06%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000834s :  5.12%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.14%
    jit_opt_a.cse                        :     0.000041s :  0.25%
    jit_opt_a.replace_applicator         :     0.000020s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.10%
    rewriter_after_opt_a                 :     0.000038s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000813s :  4.99%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000032s :  0.20%
    loop_unroll                          :     0.000499s :  3.06%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000029s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000053s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000519s :  3.19%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006604s : 40.53%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000235    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.82% :     0.000002s :      2: substitution.fold_const_symbol
 2.47% :     0.000006s :      3: substitution.graph_param_transform
84.90% :     0.000200s :      4: substitution.inline
 1.36% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.90% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000005s :      2: substitution.replace_old_param
 5.71% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004858     2
82.44% :     0.004005s :      1: type_inference.infer
17.56% :     0.000853s :      1: type_inference.specialize
------[replace.]   0.000072     6
66.43% :     0.000048s :      4: replace.inline
33.57% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000210     6
94.07% :     0.000197s :      4: match.inline
 5.93% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.63% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 1.99% :     0.000002s :     10: predicate.arithmetic_simplify
 1.09% :     0.000001s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.81% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.14% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.06% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 5.61% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :      6: predicate.less_batch_normalization
 1.37% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 1.55% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.15% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 2.14% :     0.000003s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.38% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.47% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.65% :     0.000001s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.05% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.61% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.20% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.09% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.71% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.98% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.06% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.84% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000622    12
48.98% :     0.000304s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.19% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.83% :     0.000272s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.034954    76
 0.16% :     0.000057s :      1: add_recomputation
 0.19% :     0.000065s :      1: auto_monad
 0.05% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.15% :     0.000402s :      1: bootstrap
 0.10% :     0.000035s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.07% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.07% :     0.000025s :      1: event_method
 0.04% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
52.60% :     0.018385s :      1: jit_opt_a
 0.47% :     0.000163s :      1: jit_opt_after_cconv
 0.17% :     0.000059s :      1: jit_opt_b
 1.45% :     0.000508s :      1: loop_unroll
 2.36% :     0.000825s :      1: mutable_eliminate
 2.52% :     0.000882s :     26: opt.transform.jit_opt_a
 0.14% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000029s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.51% :     0.000529s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000041s :      1: py_interpret_to_execute
 0.05% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 1.32% :     0.000460s :      1: renormalize.infer
 1.05% :     0.000366s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000041s :      1: rewriter_after_opt_a
 0.22% :     0.000075s :      1: rewriter_before_opt_a
 0.23% :     0.000080s :      1: symbol_engine_optimizer
18.94% :     0.006619s :      1: task_emit
14.16% :     0.004951s :      1: type_inference
 0.18% :     0.000065s :      1: validate


TotalTime = 0.0228036, [33]
[bootstrap]: 0.00036361
[type_inference]: 0.00551345
[event_method]: 1.763e-05
[auto_monad]: 9.067e-05
[graph_reusing]: 6.42001e-06
[pre_auto_parallel]: 2.56e-06
[py_interpret_to_execute]: 4.808e-05
[rewriter_before_opt_a]: 7.569e-05
[expand_dump_flag]: 3.34001e-06
[jit_opt_a]: 0.00635258, [2]
    [Cycle 1]: 0.00210051, [27]
        [switch_simplify]: 6.032e-05
        [loop_unroll]: 3.125e-05
        [a_1]: 0.00058733
        [with_stream_mark]: 2.59e-05
        [recompute_prepare]: 7.95998e-06
        [updatestate_depend_eliminate]: 4.01001e-06
        [updatestate_assign_eliminate]: 3.25e-06
        [updatestate_loads_eliminate]: 2.78e-06
        [parameter_eliminate]: 2.44999e-06
        [specialize_transform]: 6.23e-06
        [updatestate_useless_node_eliminater]: 5.37999e-06
        [accelerated_algorithm]: 5.63997e-06
        [meta_shard_fg_expand]: 2.75002e-06
        [get_grad_eliminate_]: 5.82999e-06
        [merge_forward]: 3.81999e-06
        [cell_reuse_recompute_pass]: 2.15002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.605e-05
        [j_node_and_user_rematch]: 1.842e-05
        [meta_fg_expand]: 3.2e-06
        [replace_old_param]: 9.84001e-06
        [inline_without_move]: 5.74999e-06
        [renormalize]: 0.00080898
        [add_forward_monad_depend]: 7.76001e-06
        [auto_monad_grad]: 2.73998e-06
        [auto_monad_eliminator]: 1.882e-05
        [cse]: 3.343e-05
        [replace_applicator]: 1.791e-05
    [Cycle 2]: 0.00035019, [27]
        [switch_simplify]: 6.26e-06
        [loop_unroll]: 4.96002e-06
        [a_1]: 9.435e-05
        [with_stream_mark]: 1.482e-05
        [recompute_prepare]: 5.81e-06
        [updatestate_depend_eliminate]: 3.83001e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.21998e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 5.54e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 5.08002e-06
        [meta_shard_fg_expand]: 1.81e-06
        [get_grad_eliminate_]: 5.02e-06
        [merge_forward]: 3.73999e-06
        [cell_reuse_recompute_pass]: 3.23998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.434e-05
        [j_node_and_user_rematch]: 8.92999e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 8.13999e-06
        [inline_without_move]: 4.97999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.84e-06
        [auto_monad_grad]: 1.06002e-06
        [auto_monad_eliminator]: 5.68002e-06
        [cse]: 1.306e-05
        [replace_applicator]: 5.22999e-06
[py_interpret_to_execute_after_opt_a]: 1.598e-05
[rewriter_after_opt_a]: 4.049e-05
[convert_after_rewriter]: 7.88999e-06
[order_py_execute_after_rewriter]: 5.66e-06
[mutable_eliminate]: 0.00080827
[jit_opt_b]: 5.56e-05, [1]
    [Cycle 1]: 4.717e-05, [2]
        [frontend_op_eliminate]: 1.772e-05
        [inline_after_opt_a]: 1.669e-05
[cconv]: 3.815e-05
[loop_unroll]: 0.0005445
[jit_opt_after_cconv]: 0.00017878, [1]
    [Cycle 1]: 0.00017077, [11]
        [c_1]: 2.193e-05
        [parameter_eliminate]: 5.52001e-06
        [updatestate_depend_eliminate]: 1.072e-05
        [updatestate_assign_eliminate]: 3.03e-06
        [updatestate_loads_eliminate]: 2.63998e-06
        [cse]: 3.659e-05
        [call_graph_tuple_transform]: 2.377e-05
        [tuple_list_get_item_eliminator]: 5.32001e-06
        [none_parameter_eliminate]: 2.04e-06
        [renormalize]: 9.39996e-07
        [switch_simplify]: 5.39e-06
[remove_dup_value]: 1.634e-05
[partial_unused_args_eliminate]: 2.49001e-06
[environ_conv]: 6.42001e-06
[add_recomputation]: 6.032e-05
[cse_after_recomputation]: 2.187e-05, [1]
    [Cycle 1]: 1.535e-05, [1]
        [cse]: 9.01002e-06
[auto_monad_reorder]: 1.629e-05
[get_jit_bprop_graph]: 2.39001e-06
[rewriter_after_jit_bprop_graph]: 8.90001e-06
[opt_after_jit_grad]: 0.00060183
[symbol_engine_optimizer]: 8.391e-05, [1]
    [Cycle 1]: 7.63e-05, [6]
        [build]: 6.39999e-06
        [elim_shapecalc]: 8.02e-06
        [elim_not_effective]: 1.655e-05
        [opt_reshape]: 5.77999e-06
        [fold_const_symbol]: 9.89999e-06
        [renormalize]: 3.19997e-07
[validate]: 4.831e-05
[backend_pass]: 1.02e-06
[task_emit]: 0.00749224
[execute]: 1.029e-05
Sums
    bootstrap                            :     0.000364s :  2.02%
    type_inference                       :     0.005513s : 30.63%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000091s :  0.50%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000048s :  0.27%
    rewriter_before_opt_a                :     0.000076s :  0.42%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.37%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000682s :  3.79%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000809s :  4.50%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.14%
    jit_opt_a.cse                        :     0.000046s :  0.26%
    jit_opt_a.replace_applicator         :     0.000023s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.09%
    rewriter_after_opt_a                 :     0.000040s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000808s :  4.49%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000038s :  0.21%
    loop_unroll                          :     0.000545s :  3.03%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000037s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000060s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000602s :  3.34%
    symbol_engine_optimizer.build        :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000048s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007492s : 41.63%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000251    23
 0.95% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000002s :      2: substitution.fold_const_symbol
 2.23% :     0.000006s :      3: substitution.graph_param_transform
84.35% :     0.000212s :      4: substitution.inline
 2.06% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.52% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.00% :     0.000005s :      2: substitution.replace_old_param
 5.17% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005440     2
83.63% :     0.004549s :      1: type_inference.infer
16.37% :     0.000890s :      1: type_inference.specialize
------[replace.]   0.000070     6
67.68% :     0.000047s :      4: replace.inline
32.32% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000221     6
94.69% :     0.000209s :      4: match.inline
 5.31% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.38% :     0.000003s :     10: predicate.arithmetic_simplify
 1.53% :     0.000002s :     10: predicate.cast_eliminate
 0.61% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.53% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.14% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.22% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     10: predicate.environ_get_eliminate
 1.32% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.19% :     0.000002s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.82% :     0.000007s :     22: predicate.inline
 1.22% :     0.000002s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.58% :     0.000002s :     15: predicate.load_eliminater
 1.76% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.63% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.78% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 3.02% :     0.000004s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 1.02% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.54% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.22% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.05% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.16% :     0.000001s :      6: predicate.specialize_transform
 1.58% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.65% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.62% :     0.000010s :     53: predicate.switch_simplify
 1.10% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.92% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.33% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000650    12
48.63% :     0.000316s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.64% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.73% :     0.000284s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024661    76
 0.26% :     0.000064s :      1: add_recomputation
 0.38% :     0.000095s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 1.59% :     0.000391s :      1: bootstrap
 0.17% :     0.000041s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.78% :     0.006357s :      1: jit_opt_a
 0.74% :     0.000182s :      1: jit_opt_after_cconv
 0.24% :     0.000059s :      1: jit_opt_b
 2.26% :     0.000556s :      1: loop_unroll
 3.35% :     0.000826s :      1: mutable_eliminate
 3.77% :     0.000929s :     26: opt.transform.jit_opt_a
 0.21% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.07% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.49% :     0.000614s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000052s :      1: py_interpret_to_execute
 0.07% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000019s :      1: remove_dup_value
 1.82% :     0.000449s :      1: renormalize.infer
 1.42% :     0.000351s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000044s :      1: rewriter_after_opt_a
 0.32% :     0.000079s :      1: rewriter_before_opt_a
 0.35% :     0.000087s :      1: symbol_engine_optimizer
30.47% :     0.007514s :      1: task_emit
22.41% :     0.005526s :      1: type_inference
 0.31% :     0.000075s :      1: validate


TotalTime = 0.0529777, [33]
[bootstrap]: 0.00029039
[type_inference]: 0.00563607
[event_method]: 1.888e-05
[auto_monad]: 6.847e-05
[graph_reusing]: 5.87999e-06
[pre_auto_parallel]: 2.66e-06
[py_interpret_to_execute]: 4.744e-05
[rewriter_before_opt_a]: 7.554e-05
[expand_dump_flag]: 3.36001e-06
[jit_opt_a]: 0.00578936, [2]
    [Cycle 1]: 0.00195598, [27]
        [switch_simplify]: 6.277e-05
        [loop_unroll]: 3.511e-05
        [a_1]: 0.00059931
        [with_stream_mark]: 1.905e-05
        [recompute_prepare]: 8.34998e-06
        [updatestate_depend_eliminate]: 4.62e-06
        [updatestate_assign_eliminate]: 3.38e-06
        [updatestate_loads_eliminate]: 3.16999e-06
        [parameter_eliminate]: 2.21e-06
        [specialize_transform]: 6.36998e-06
        [updatestate_useless_node_eliminater]: 5.63002e-06
        [accelerated_algorithm]: 5.94e-06
        [meta_shard_fg_expand]: 2.20002e-06
        [get_grad_eliminate_]: 5.25001e-06
        [merge_forward]: 3.33998e-06
        [cell_reuse_recompute_pass]: 1.14998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.542e-05
        [j_node_and_user_rematch]: 9.78002e-06
        [meta_fg_expand]: 2.31998e-06
        [replace_old_param]: 9.61e-06
        [inline_without_move]: 5.52001e-06
        [renormalize]: 0.00087508
        [add_forward_monad_depend]: 6.94999e-06
        [auto_monad_grad]: 2.59999e-06
        [auto_monad_eliminator]: 1.537e-05
        [cse]: 3.143e-05
        [replace_applicator]: 1.337e-05
    [Cycle 2]: 0.00034666, [27]
        [switch_simplify]: 5.71e-06
        [loop_unroll]: 5.19e-06
        [a_1]: 0.00010242
        [with_stream_mark]: 1.127e-05
        [recompute_prepare]: 5.39e-06
        [updatestate_depend_eliminate]: 2.66999e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 4.90001e-06
        [updatestate_useless_node_eliminater]: 4.91997e-06
        [accelerated_algorithm]: 5.36002e-06
        [meta_shard_fg_expand]: 1.02998e-06
        [get_grad_eliminate_]: 4.63999e-06
        [merge_forward]: 2.61e-06
        [cell_reuse_recompute_pass]: 2.06e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.352e-05
        [j_node_and_user_rematch]: 8.17e-06
        [meta_fg_expand]: 1.60001e-06
        [replace_old_param]: 8.10999e-06
        [inline_without_move]: 4.63001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.55001e-06
        [auto_monad_grad]: 1.03001e-06
        [auto_monad_eliminator]: 5.18002e-06
        [cse]: 1.106e-05
        [replace_applicator]: 5.91998e-06
[py_interpret_to_execute_after_opt_a]: 1.176e-05
[rewriter_after_opt_a]: 3.819e-05
[convert_after_rewriter]: 6.91999e-06
[order_py_execute_after_rewriter]: 5.22999e-06
[mutable_eliminate]: 0.00061725
[jit_opt_b]: 5.558e-05, [1]
    [Cycle 1]: 4.707e-05, [2]
        [frontend_op_eliminate]: 1.956e-05
        [inline_after_opt_a]: 1.492e-05
[cconv]: 2.839e-05
[loop_unroll]: 0.00045421
[jit_opt_after_cconv]: 0.00015003, [1]
    [Cycle 1]: 0.00014282, [11]
        [c_1]: 2.255e-05
        [parameter_eliminate]: 2.94999e-06
        [updatestate_depend_eliminate]: 6.74001e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [cse]: 2.109e-05
        [call_graph_tuple_transform]: 1.935e-05
        [tuple_list_get_item_eliminator]: 5.54e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 5.43002e-06
[remove_dup_value]: 1.557e-05
[partial_unused_args_eliminate]: 2.14999e-06
[environ_conv]: 5.77999e-06
[add_recomputation]: 5.035e-05
[cse_after_recomputation]: 2.157e-05, [1]
    [Cycle 1]: 1.624e-05, [1]
        [cse]: 9.90002e-06
[auto_monad_reorder]: 1.56e-05
[get_jit_bprop_graph]: 2.10002e-06
[rewriter_after_jit_bprop_graph]: 6.14001e-06
[opt_after_jit_grad]: 0.00052646
[symbol_engine_optimizer]: 8.238e-05, [1]
    [Cycle 1]: 7.52e-05, [6]
        [build]: 4.13001e-06
        [elim_shapecalc]: 9.00999e-06
        [elim_not_effective]: 1.42e-05
        [opt_reshape]: 6.12999e-06
        [fold_const_symbol]: 1.069e-05
        [renormalize]: 4.59986e-07
[validate]: 4.342e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.038669
[execute]: 7.5e-06
Sums
    bootstrap                            :     0.000290s :  0.59%
    type_inference                       :     0.005636s : 11.55%
    event_method                         :     0.000019s :  0.04%
    auto_monad                           :     0.000068s :  0.14%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000047s :  0.10%
    rewriter_before_opt_a                :     0.000076s :  0.15%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000068s :  0.14%
    jit_opt_a.loop_unroll                :     0.000040s :  0.08%
    jit_opt_a.a_1                        :     0.000702s :  1.44%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.06%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.02%
    jit_opt_a.merge_forward              :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000018s :  0.04%
    jit_opt_a.inline_without_move        :     0.000010s :  0.02%
    jit_opt_a.renormalize                :     0.000875s :  1.79%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.04%
    jit_opt_a.cse                        :     0.000042s :  0.09%
    jit_opt_a.replace_applicator         :     0.000019s :  0.04%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.02%
    rewriter_after_opt_a                 :     0.000038s :  0.08%
    convert_after_rewriter               :     0.000007s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000617s :  1.26%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.03%
    cconv                                :     0.000028s :  0.06%
    loop_unroll                          :     0.000454s :  0.93%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.05%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000021s :  0.04%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.01%
    remove_dup_value                     :     0.000016s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000006s :  0.01%
    add_recomputation                    :     0.000050s :  0.10%
    cse_after_recomputation.cse          :     0.000010s :  0.02%
    auto_monad_reorder                   :     0.000016s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000526s :  1.08%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000043s :  0.09%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.038669s : 79.22%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000233    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.37% :     0.000006s :      3: substitution.graph_param_transform
84.60% :     0.000197s :      4: substitution.inline
 1.52% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.19% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.19% :     0.000005s :      2: substitution.replace_old_param
 5.58% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005554     2
80.37% :     0.004463s :      1: type_inference.infer
19.63% :     0.001091s :      1: type_inference.specialize
------[replace.]   0.000081     6
67.04% :     0.000055s :      4: replace.inline
32.96% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000206     6
94.26% :     0.000194s :      4: match.inline
 5.74% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   691
 1.27% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 2.13% :     0.000003s :     10: predicate.arithmetic_simplify
 1.26% :     0.000002s :     10: predicate.cast_eliminate
 0.62% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000002s :     10: predicate.compare_switch_simplify
 1.70% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.47% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.34% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.71% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.88% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.83% :     0.000001s :      6: predicate.get_grad_eliminate
 0.35% :     0.000000s :      3: predicate.graph_param_transform
 5.42% :     0.000007s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.04% :     0.000001s :      6: predicate.less_batch_normalization
 1.73% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.36% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 2.14% :     0.000003s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.04% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.31% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.08% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.27% :     0.000002s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.02% :     0.000004s :     16: predicate.switch_defer_inline
 1.94% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.41% :     0.000010s :     53: predicate.switch_simplify
 1.22% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.74% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.13% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.75% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.84% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000699    12
46.71% :     0.000327s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.27% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.02% :     0.000329s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.054889    76
 0.10% :     0.000054s :      1: add_recomputation
 0.13% :     0.000072s :      1: auto_monad
 0.03% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.56% :     0.000308s :      1: bootstrap
 0.06% :     0.000031s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.04% :     0.000024s :      1: cse_after_recomputation
 0.01% :     0.000008s :      1: environ_conv
 0.04% :     0.000024s :      1: event_method
 0.02% :     0.000011s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000008s :      1: graph_reusing
10.55% :     0.005793s :      1: jit_opt_a
 0.28% :     0.000153s :      1: jit_opt_after_cconv
 0.11% :     0.000058s :      1: jit_opt_b
 0.84% :     0.000462s :      1: loop_unroll
 1.14% :     0.000628s :      1: mutable_eliminate
 1.70% :     0.000931s :     26: opt.transform.jit_opt_a
 0.09% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000029s :      4: opt.transform.jit_opt_b
 0.02% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000036s :      4: opt.transform.symbol_engine_opt
 0.98% :     0.000536s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000051s :      1: py_interpret_to_execute
 0.03% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000018s :      1: remove_dup_value
 0.86% :     0.000472s :      1: renormalize.infer
 0.72% :     0.000393s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000041s :      1: rewriter_after_opt_a
 0.14% :     0.000078s :      1: rewriter_before_opt_a
 0.15% :     0.000085s :      1: symbol_engine_optimizer
70.48% :     0.038684s :      1: task_emit
10.30% :     0.005653s :      1: type_inference
 0.12% :     0.000066s :      1: validate


TotalTime = 0.0211456, [33]
[bootstrap]: 0.00034025
[type_inference]: 0.00524914
[event_method]: 1.985e-05
[auto_monad]: 6.369e-05
[graph_reusing]: 6.39001e-06
[pre_auto_parallel]: 2.58e-06
[py_interpret_to_execute]: 4.033e-05
[rewriter_before_opt_a]: 7.296e-05
[expand_dump_flag]: 3.4e-06
[jit_opt_a]: 0.00634371, [2]
    [Cycle 1]: 0.00174778, [27]
        [switch_simplify]: 6.068e-05
        [loop_unroll]: 2.914e-05
        [a_1]: 0.00055207
        [with_stream_mark]: 2.104e-05
        [recompute_prepare]: 8.52998e-06
        [updatestate_depend_eliminate]: 3.76999e-06
        [updatestate_assign_eliminate]: 2.81999e-06
        [updatestate_loads_eliminate]: 2.45997e-06
        [parameter_eliminate]: 1.76e-06
        [specialize_transform]: 6.09999e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.91e-06
        [meta_shard_fg_expand]: 2.69999e-06
        [get_grad_eliminate_]: 5.15999e-06
        [merge_forward]: 4e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.499e-05
        [j_node_and_user_rematch]: 8.97999e-06
        [meta_fg_expand]: 2.37001e-06
        [replace_old_param]: 8.94998e-06
        [inline_without_move]: 5.09e-06
        [renormalize]: 0.00074942
        [add_forward_monad_depend]: 5.86e-06
        [auto_monad_grad]: 2.62001e-06
        [auto_monad_eliminator]: 1.463e-05
        [cse]: 2.747e-05
        [replace_applicator]: 1.376e-05
    [Cycle 2]: 0.00035523, [27]
        [switch_simplify]: 5.49998e-06
        [loop_unroll]: 5.25999e-06
        [a_1]: 9.373e-05
        [with_stream_mark]: 1.307e-05
        [recompute_prepare]: 4.76002e-06
        [updatestate_depend_eliminate]: 2.76999e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 4.96002e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 1.60999e-06
        [get_grad_eliminate_]: 4.99998e-06
        [merge_forward]: 2.73003e-06
        [cell_reuse_recompute_pass]: 2.46e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.491e-05
        [j_node_and_user_rematch]: 7.43999e-06
        [meta_fg_expand]: 1.71998e-06
        [replace_old_param]: 1.298e-05
        [inline_without_move]: 4.47e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 7.80012e-07
        [auto_monad_eliminator]: 6.54001e-06
        [cse]: 1.273e-05
        [replace_applicator]: 5.52999e-06
[py_interpret_to_execute_after_opt_a]: 1.336e-05
[rewriter_after_opt_a]: 3.733e-05
[convert_after_rewriter]: 6.81999e-06
[order_py_execute_after_rewriter]: 6.93e-06
[mutable_eliminate]: 0.00072258
[jit_opt_b]: 5.174e-05, [1]
    [Cycle 1]: 4.401e-05, [2]
        [frontend_op_eliminate]: 1.663e-05
        [inline_after_opt_a]: 1.603e-05
[cconv]: 2.655e-05
[loop_unroll]: 0.00043838
[jit_opt_after_cconv]: 0.00013799, [1]
    [Cycle 1]: 0.00013157, [11]
        [c_1]: 2.001e-05
        [parameter_eliminate]: 2.81999e-06
        [updatestate_depend_eliminate]: 5.20999e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [cse]: 1.92e-05
        [call_graph_tuple_transform]: 1.907e-05
        [tuple_list_get_item_eliminator]: 4.99e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 8.00006e-07
        [switch_simplify]: 5.18002e-06
[remove_dup_value]: 1.519e-05
[partial_unused_args_eliminate]: 2.33998e-06
[environ_conv]: 5.24e-06
[add_recomputation]: 4.772e-05
[cse_after_recomputation]: 1.961e-05, [1]
    [Cycle 1]: 1.404e-05, [1]
        [cse]: 8.25999e-06
[auto_monad_reorder]: 1.502e-05
[get_jit_bprop_graph]: 2.27999e-06
[rewriter_after_jit_bprop_graph]: 5.68002e-06
[opt_after_jit_grad]: 0.00047206
[symbol_engine_optimizer]: 7.234e-05, [1]
    [Cycle 1]: 6.541e-05, [6]
        [build]: 3.34001e-06
        [elim_shapecalc]: 7.3e-06
        [elim_not_effective]: 1.285e-05
        [opt_reshape]: 5.22999e-06
        [fold_const_symbol]: 8.55001e-06
        [renormalize]: 2.89991e-07
[validate]: 3.737e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00663478
[execute]: 8.30999e-06
Sums
    bootstrap                            :     0.000340s :  2.09%
    type_inference                       :     0.005249s : 32.30%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000064s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.25%
    rewriter_before_opt_a                :     0.000073s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.21%
    jit_opt_a.a_1                        :     0.000646s :  3.97%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000022s :  0.13%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000749s :  4.61%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.13%
    jit_opt_a.cse                        :     0.000040s :  0.25%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000037s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000007s :  0.04%
    mutable_eliminate                    :     0.000723s :  4.45%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000027s :  0.16%
    loop_unroll                          :     0.000438s :  2.70%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000048s :  0.29%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000472s :  2.90%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006635s : 40.83%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000218    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.42% :     0.000005s :      3: substitution.graph_param_transform
84.36% :     0.000184s :      4: substitution.inline
 1.44% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.28% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.30% :     0.000005s :      2: substitution.replace_old_param
 5.59% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005167     2
82.49% :     0.004262s :      1: type_inference.infer
17.51% :     0.000905s :      1: type_inference.specialize
------[replace.]   0.000073     6
68.27% :     0.000050s :      4: replace.inline
31.73% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     6
94.19% :     0.000181s :      4: match.inline
 5.81% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.56% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.19% :     0.000001s :     10: predicate.addn_zero_filter
 2.34% :     0.000003s :     10: predicate.arithmetic_simplify
 1.39% :     0.000002s :     10: predicate.cast_eliminate
 0.57% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 2.14% :     0.000003s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.52% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.43% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.48% :     0.000001s :      3: predicate.graph_param_transform
 5.27% :     0.000006s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.64% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.39% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 2.10% :     0.000003s :      3: predicate.mutable_eliminate
 0.37% :     0.000000s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.47% :     0.000001s :      3: predicate.reset_defer_inline
 1.19% :     0.000001s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.60% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.31% :     0.000002s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.18% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.85% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.51% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000705    12
55.68% :     0.000393s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.39% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.93% :     0.000282s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022862    76
 0.22% :     0.000051s :      1: add_recomputation
 0.29% :     0.000067s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.58% :     0.000361s :      1: bootstrap
 0.13% :     0.000030s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.11% :     0.000025s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
27.76% :     0.006348s :      1: jit_opt_a
 0.62% :     0.000141s :      1: jit_opt_after_cconv
 0.24% :     0.000054s :      1: jit_opt_b
 1.95% :     0.000446s :      1: loop_unroll
 3.20% :     0.000732s :      1: mutable_eliminate
 3.81% :     0.000870s :     26: opt.transform.jit_opt_a
 0.20% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.10% :     0.000480s :      1: opt_after_jit_grad
 0.04% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000044s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 1.72% :     0.000393s :      1: renormalize.infer
 1.52% :     0.000348s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000040s :      1: rewriter_after_opt_a
 0.33% :     0.000076s :      1: rewriter_before_opt_a
 0.33% :     0.000075s :      1: symbol_engine_optimizer
29.10% :     0.006653s :      1: task_emit
23.05% :     0.005270s :      1: type_inference
 0.25% :     0.000056s :      1: validate


TotalTime = 0.0182095, [30]
[bootstrap]: 0.00047519
[type_inference]: 0.00628927
[event_method]: 2.293e-05
[auto_monad]: 7.554e-05
[graph_reusing]: 7.08e-06
[pre_auto_parallel]: 2.89999e-06
[py_interpret_to_execute]: 5.028e-05
[rewriter_before_opt_a]: 8.464e-05
[expand_dump_flag]: 3.49001e-06
[jit_opt_a]: 0.00785085, [2]
    [Cycle 1]: 0.00257858, [27]
        [switch_simplify]: 8.466e-05
        [loop_unroll]: 3.436e-05
        [a_1]: 0.00076376
        [with_stream_mark]: 3.16e-05
        [recompute_prepare]: 1.358e-05
        [updatestate_depend_eliminate]: 4.70001e-06
        [updatestate_assign_eliminate]: 3.16999e-06
        [updatestate_loads_eliminate]: 3.08e-06
        [parameter_eliminate]: 2.88998e-06
        [specialize_transform]: 7.97e-06
        [updatestate_useless_node_eliminater]: 6.46e-06
        [accelerated_algorithm]: 8.3e-06
        [meta_shard_fg_expand]: 2.84001e-06
        [get_grad_eliminate_]: 5.99999e-06
        [merge_forward]: 4.64002e-06
        [cell_reuse_recompute_pass]: 2.26e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.124e-05
        [j_node_and_user_rematch]: 1.221e-05
        [meta_fg_expand]: 3.03e-06
        [replace_old_param]: 1.377e-05
        [inline_without_move]: 6.17001e-06
        [renormalize]: 0.00121066
        [add_forward_monad_depend]: 1.086e-05
        [auto_monad_grad]: 3.26001e-06
        [auto_monad_eliminator]: 1.879e-05
        [cse]: 3.674e-05
        [replace_applicator]: 1.457e-05
    [Cycle 2]: 0.00044497, [27]
        [switch_simplify]: 6.99001e-06
        [loop_unroll]: 6.61999e-06
        [a_1]: 0.00010561
        [with_stream_mark]: 4.84e-05
        [recompute_prepare]: 6.81999e-06
        [updatestate_depend_eliminate]: 4.03999e-06
        [updatestate_assign_eliminate]: 2.20002e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [parameter_eliminate]: 2.17999e-06
        [specialize_transform]: 5.04e-06
        [updatestate_useless_node_eliminater]: 5.34e-06
        [accelerated_algorithm]: 6.26998e-06
        [meta_shard_fg_expand]: 1.87001e-06
        [get_grad_eliminate_]: 4.80999e-06
        [merge_forward]: 4.29002e-06
        [cell_reuse_recompute_pass]: 2.89999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.846e-05
        [j_node_and_user_rematch]: 1.082e-05
        [meta_fg_expand]: 2.88e-06
        [replace_old_param]: 1.018e-05
        [inline_without_move]: 5.10001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 2.51998e-06
        [auto_monad_grad]: 1.14998e-06
        [auto_monad_eliminator]: 8.80001e-06
        [cse]: 1.691e-05
        [replace_applicator]: 7.68001e-06
[py_interpret_to_execute_after_opt_a]: 2.005e-05
[rewriter_after_opt_a]: 4.523e-05
[convert_after_rewriter]: 8.59e-06
[order_py_execute_after_rewriter]: 5.80002e-06
[mutable_eliminate]: 0.00099577
[jit_opt_b]: 7.434e-05, [1]
    [Cycle 1]: 6.117e-05, [2]
        [frontend_op_eliminate]: 2.458e-05
        [inline_after_opt_a]: 2.005e-05
[cconv]: 4.316e-05
[loop_unroll]: 0.00065669
[jit_opt_after_cconv]: 0.00018896, [1]
    [Cycle 1]: 0.00017881, [11]
        [c_1]: 2.533e-05
        [parameter_eliminate]: 2.88e-06
        [updatestate_depend_eliminate]: 7.26001e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [cse]: 4.237e-05
        [call_graph_tuple_transform]: 2.439e-05
        [tuple_list_get_item_eliminator]: 6.55002e-06
        [none_parameter_eliminate]: 1.87999e-06
        [renormalize]: 6.80011e-07
        [switch_simplify]: 6.11e-06
[remove_dup_value]: 1.793e-05
[partial_unused_args_eliminate]: 2.31e-06
[environ_conv]: 8.83001e-06
[add_recomputation]: 6.147e-05
[cse_after_recomputation]: 2.916e-05, [1]
    [Cycle 1]: 2.033e-05, [1]
        [cse]: 1.248e-05
[auto_monad_reorder]: 1.805e-05
[get_jit_bprop_graph]: 2.76999e-06
[rewriter_after_jit_bprop_graph]: 1.176e-05
[opt_after_jit_grad]: 0.00076386
[symbol_engine_optimizer]: 9.289e-05, [1]
    [Cycle 1]: 8.092e-05, [6]
        [build]: 5.69e-06
        [elim_shapecalc]: 9.17999e-06
        [elim_not_effective]: 1.766e-05
        [opt_reshape]: 7.14001e-06
        [fold_const_symbol]: 1.008e-05
        [renormalize]: 8.2e-07
[validate]: 4.947e-05
Sums
    bootstrap                            :     0.000475s :  3.78%
    type_inference                       :     0.006289s : 49.97%
    event_method                         :     0.000023s :  0.18%
    auto_monad                           :     0.000076s :  0.60%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000050s :  0.40%
    rewriter_before_opt_a                :     0.000085s :  0.67%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000092s :  0.73%
    jit_opt_a.loop_unroll                :     0.000041s :  0.33%
    jit_opt_a.a_1                        :     0.000869s :  6.91%
    jit_opt_a.with_stream_mark           :     0.000080s :  0.64%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.04%
    jit_opt_a.specialize_transform       :     0.000013s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000015s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.09%
    jit_opt_a.merge_forward              :     0.000009s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000023s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.05%
    jit_opt_a.replace_old_param          :     0.000024s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.001211s :  9.62%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.11%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.22%
    jit_opt_a.cse                        :     0.000054s :  0.43%
    jit_opt_a.replace_applicator         :     0.000022s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000020s :  0.16%
    rewriter_after_opt_a                 :     0.000045s :  0.36%
    convert_after_rewriter               :     0.000009s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000996s :  7.91%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.16%
    cconv                                :     0.000043s :  0.34%
    loop_unroll                          :     0.000657s :  5.22%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000042s :  0.34%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000018s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000009s :  0.07%
    add_recomputation                    :     0.000061s :  0.49%
    cse_after_recomputation.cse          :     0.000012s :  0.10%
    auto_monad_reorder                   :     0.000018s :  0.14%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000012s :  0.09%
    opt_after_jit_grad                   :     0.000764s :  6.07%
    symbol_engine_optimizer.build        :     0.000006s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000049s :  0.39%


Time group info:
------[substitution.]   0.000341    23
 0.82% :     0.000003s :      2: substitution.elim_not_effective
 0.52% :     0.000002s :      2: substitution.fold_const_symbol
 2.23% :     0.000008s :      3: substitution.graph_param_transform
85.57% :     0.000292s :      4: substitution.inline
 1.80% :     0.000006s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000007s :      4: substitution.remove_not_recompute_node
 2.45% :     0.000008s :      2: substitution.replace_old_param
 4.60% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006192     2
81.00% :     0.005016s :      1: type_inference.infer
19.00% :     0.001176s :      1: type_inference.specialize
------[replace.]   0.000110     6
66.46% :     0.000073s :      4: replace.inline
33.54% :     0.000037s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000303     6
95.28% :     0.000289s :      4: match.inline
 4.72% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000160   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.94% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.77% :     0.000001s :     10: predicate.addn_check_dump
 1.49% :     0.000002s :     10: predicate.addn_zero_filter
 2.20% :     0.000004s :     10: predicate.arithmetic_simplify
 1.97% :     0.000003s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.76% :     0.000001s :     10: predicate.compare_switch_simplify
 1.72% :     0.000003s :     10: predicate.depend_value_elim
 1.30% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.55% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.21% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.01% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.78% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.03% :     0.000002s :     10: predicate.environ_get_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.81% :     0.000001s :      6: predicate.get_grad_eliminate
 0.34% :     0.000001s :      3: predicate.graph_param_transform
 5.87% :     0.000009s :     22: predicate.inline
 0.70% :     0.000001s :      6: predicate.inline_without_move
 0.26% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.64% :     0.000003s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.23% :     0.000002s :     15: predicate.load_eliminater
 2.00% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.27% :     0.000007s :     34: predicate.loop_unroll_before_grad
 2.09% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000002s :     10: predicate.merge_addn
 0.91% :     0.000001s :     10: predicate.minmaximum_grad
 2.96% :     0.000005s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 1.64% :     0.000003s :     15: predicate.partial_eliminate
 1.08% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.31% :     0.000004s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.36% :     0.000002s :     18: predicate.replace_applicator
 0.44% :     0.000001s :      6: predicate.replace_old_param
 0.38% :     0.000001s :      3: predicate.reset_defer_inline
 1.81% :     0.000003s :     10: predicate.reshape_eliminate
 0.98% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.37% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.74% :     0.000003s :     10: predicate.same_eliminate
 0.75% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000002s :      6: predicate.special_op_eliminate
 0.75% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.08% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.29% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.20% :     0.000004s :     16: predicate.switch_defer_inline
 1.72% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.73% :     0.000012s :     53: predicate.switch_simplify
 1.20% :     0.000002s :     10: predicate.tile_eliminate
 1.25% :     0.000002s :     10: predicate.transpose_eliminate
 1.07% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.93% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.73% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 2.39% :     0.000004s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.21% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.29% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000003s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.85% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000887    12
48.04% :     0.000426s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.91% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.06% :     0.000399s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020646    72
 0.32% :     0.000065s :      1: add_recomputation
 0.38% :     0.000079s :      1: auto_monad
 0.11% :     0.000022s :      1: auto_monad_reorder
 2.44% :     0.000504s :      1: bootstrap
 0.23% :     0.000047s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.16% :     0.000032s :      1: cse_after_recomputation
 0.06% :     0.000013s :      1: environ_conv
 0.13% :     0.000028s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
38.05% :     0.007856s :      1: jit_opt_a
 0.93% :     0.000192s :      1: jit_opt_after_cconv
 0.38% :     0.000078s :      1: jit_opt_b
 3.23% :     0.000667s :      1: loop_unroll
 4.93% :     0.001018s :      1: mutable_eliminate
 5.58% :     0.001153s :     26: opt.transform.jit_opt_a
 0.28% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000035s :      4: opt.transform.jit_opt_b
 0.08% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.15% :     0.000031s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000034s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000040s :      4: opt.transform.symbol_engine_opt
 3.76% :     0.000775s :      1: opt_after_jit_grad
 0.04% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000053s :      1: py_interpret_to_execute
 0.11% :     0.000024s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000021s :      1: remove_dup_value
 3.19% :     0.000658s :      1: renormalize.infer
 2.61% :     0.000540s :      1: renormalize.specialize
 0.07% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000051s :      1: rewriter_after_opt_a
 0.42% :     0.000087s :      1: rewriter_before_opt_a
 0.46% :     0.000096s :      1: symbol_engine_optimizer
30.55% :     0.006308s :      1: type_inference


TotalTime = 0.0260801, [33]
[bootstrap]: 0.00034722
[type_inference]: 0.00542222
[event_method]: 2.184e-05
[auto_monad]: 7.186e-05
[graph_reusing]: 6.19001e-06
[pre_auto_parallel]: 2.70997e-06
[py_interpret_to_execute]: 3.857e-05
[rewriter_before_opt_a]: 7.759e-05
[expand_dump_flag]: 3.37002e-06
[jit_opt_a]: 0.00835232, [2]
    [Cycle 1]: 0.00372235, [27]
        [switch_simplify]: 5.795e-05
        [loop_unroll]: 3.113e-05
        [a_1]: 0.00226999
        [with_stream_mark]: 2.912e-05
        [recompute_prepare]: 1.122e-05
        [updatestate_depend_eliminate]: 4.68999e-06
        [updatestate_assign_eliminate]: 2.84999e-06
        [updatestate_loads_eliminate]: 2.94001e-06
        [parameter_eliminate]: 2.81999e-06
        [specialize_transform]: 6.98e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 6.23e-06
        [meta_shard_fg_expand]: 2.39001e-06
        [get_grad_eliminate_]: 5.84999e-06
        [merge_forward]: 4.08001e-06
        [cell_reuse_recompute_pass]: 2.90002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.679e-05
        [j_node_and_user_rematch]: 1.073e-05
        [meta_fg_expand]: 2.51998e-06
        [replace_old_param]: 1.671e-05
        [inline_without_move]: 6.13998e-06
        [renormalize]: 0.00093214
        [add_forward_monad_depend]: 9.99999e-06
        [auto_monad_grad]: 3.09001e-06
        [auto_monad_eliminator]: 1.766e-05
        [cse]: 2.826e-05
        [replace_applicator]: 1.438e-05
    [Cycle 2]: 0.0003705, [27]
        [switch_simplify]: 5.76998e-06
        [loop_unroll]: 5.75001e-06
        [a_1]: 9.645e-05
        [with_stream_mark]: 1.796e-05
        [recompute_prepare]: 5.23002e-06
        [updatestate_depend_eliminate]: 3.04999e-06
        [updatestate_assign_eliminate]: 2.86e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 5.76998e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 5.33002e-06
        [meta_shard_fg_expand]: 1.87999e-06
        [get_grad_eliminate_]: 4.60999e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 3.38e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.576e-05
        [j_node_and_user_rematch]: 9.65002e-06
        [meta_fg_expand]: 2.09999e-06
        [replace_old_param]: 8.97e-06
        [inline_without_move]: 4.65001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 2.42001e-06
        [auto_monad_grad]: 1.08001e-06
        [auto_monad_eliminator]: 7.6e-06
        [cse]: 1.375e-05
        [replace_applicator]: 6.34001e-06
[py_interpret_to_execute_after_opt_a]: 1.875e-05
[rewriter_after_opt_a]: 4.656e-05
[convert_after_rewriter]: 8.42e-06
[order_py_execute_after_rewriter]: 5.56e-06
[mutable_eliminate]: 0.00097978
[jit_opt_b]: 6.73e-05, [1]
    [Cycle 1]: 5.66e-05, [2]
        [frontend_op_eliminate]: 2.122e-05
        [inline_after_opt_a]: 2.094e-05
[cconv]: 3.787e-05
[loop_unroll]: 0.00060466
[jit_opt_after_cconv]: 0.00017755, [1]
    [Cycle 1]: 0.00016934, [11]
        [c_1]: 2.356e-05
        [parameter_eliminate]: 5.27001e-06
        [updatestate_depend_eliminate]: 9.36e-06
        [updatestate_assign_eliminate]: 2.88e-06
        [updatestate_loads_eliminate]: 2.78e-06
        [cse]: 3.184e-05
        [call_graph_tuple_transform]: 2.767e-05
        [tuple_list_get_item_eliminator]: 5.40999e-06
        [none_parameter_eliminate]: 1.74e-06
        [renormalize]: 9.79984e-07
        [switch_simplify]: 5.30001e-06
[remove_dup_value]: 1.68e-05
[partial_unused_args_eliminate]: 2.43998e-06
[environ_conv]: 6.99001e-06
[add_recomputation]: 5.615e-05
[cse_after_recomputation]: 2.362e-05, [1]
    [Cycle 1]: 1.71e-05, [1]
        [cse]: 1.092e-05
[auto_monad_reorder]: 1.814e-05
[get_jit_bprop_graph]: 2.43998e-06
[rewriter_after_jit_bprop_graph]: 1.098e-05
[opt_after_jit_grad]: 0.00064057
[symbol_engine_optimizer]: 8.578e-05, [1]
    [Cycle 1]: 7.659e-05, [6]
        [build]: 5.30001e-06
        [elim_shapecalc]: 8.26002e-06
        [elim_not_effective]: 1.615e-05
        [opt_reshape]: 6.61999e-06
        [fold_const_symbol]: 9.83998e-06
        [renormalize]: 7.80012e-07
[validate]: 4.989e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.00859441
[execute]: 1.331e-05
Sums
    bootstrap                            :     0.000347s :  1.65%
    type_inference                       :     0.005422s : 25.73%
    event_method                         :     0.000022s :  0.10%
    auto_monad                           :     0.000072s :  0.34%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.18%
    rewriter_before_opt_a                :     0.000078s :  0.37%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.30%
    jit_opt_a.loop_unroll                :     0.000037s :  0.18%
    jit_opt_a.a_1                        :     0.002366s : 11.23%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000026s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.000932s :  4.42%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.12%
    jit_opt_a.cse                        :     0.000042s :  0.20%
    jit_opt_a.replace_applicator         :     0.000021s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.09%
    rewriter_after_opt_a                 :     0.000047s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000980s :  4.65%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000021s :  0.10%
    cconv                                :     0.000038s :  0.18%
    loop_unroll                          :     0.000605s :  2.87%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000032s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000017s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000056s :  0.27%
    cse_after_recomputation.cse          :     0.000011s :  0.05%
    auto_monad_reorder                   :     0.000018s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.05%
    opt_after_jit_grad                   :     0.000641s :  3.04%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000050s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008594s : 40.79%
    execute                              :     0.000013s :  0.06%


Time group info:
------[substitution.]   0.001858    23
 0.11% :     0.000002s :      2: substitution.elim_not_effective
 0.11% :     0.000002s :      2: substitution.fold_const_symbol
 0.40% :     0.000007s :      3: substitution.graph_param_transform
97.30% :     0.001808s :      4: substitution.inline
 0.24% :     0.000005s :      4: substitution.j_node_and_user_rematch
 0.29% :     0.000005s :      4: substitution.remove_not_recompute_node
 0.68% :     0.000013s :      2: substitution.replace_old_param
 0.87% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005343     2
82.36% :     0.004400s :      1: type_inference.infer
17.64% :     0.000943s :      1: type_inference.specialize
------[replace.]   0.000113     6
74.56% :     0.000084s :      4: replace.inline
25.44% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.001818     6
99.17% :     0.001803s :      4: match.inline
 0.83% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000142   691
 1.68% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.47% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.86% :     0.000001s :     10: predicate.addn_check_dump
 1.70% :     0.000002s :     10: predicate.addn_zero_filter
 2.33% :     0.000003s :     10: predicate.arithmetic_simplify
 1.60% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.89% :     0.000001s :     10: predicate.compare_switch_simplify
 1.47% :     0.000002s :     10: predicate.depend_value_elim
 1.23% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      3: predicate.elim_not_effective
 0.46% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.02% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000002s :     10: predicate.environ_get_eliminate
 1.26% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.54% :     0.000001s :      3: predicate.graph_param_transform
 5.73% :     0.000008s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :      6: predicate.less_batch_normalization
 2.02% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 2.22% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.56% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.48% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 0.86% :     0.000001s :     10: predicate.minmaximum_grad
 2.39% :     0.000003s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 1.84% :     0.000003s :     15: predicate.partial_eliminate
 0.99% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.90% :     0.000003s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.70% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.11% :     0.000002s :     10: predicate.same_eliminate
 0.75% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000002s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.56% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.72% :     0.000004s :     16: predicate.switch_defer_inline
 1.79% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.09% :     0.000010s :     53: predicate.switch_simplify
 1.20% :     0.000002s :     10: predicate.tile_eliminate
 1.20% :     0.000002s :     10: predicate.transpose_eliminate
 1.28% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.06% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.58% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.86% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.59% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.59% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000670    12
44.80% :     0.000300s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 9.17% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.02% :     0.000308s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.029761    76
 0.20% :     0.000060s :      1: add_recomputation
 0.25% :     0.000075s :      1: auto_monad
 0.07% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.23% :     0.000367s :      1: bootstrap
 0.14% :     0.000041s :      1: cconv
 0.05% :     0.000014s :      1: convert_after_rewriter
 0.09% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.09% :     0.000026s :      1: event_method
 0.07% :     0.000019s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
28.08% :     0.008357s :      1: jit_opt_a
 0.61% :     0.000181s :      1: jit_opt_after_cconv
 0.24% :     0.000071s :      1: jit_opt_b
 2.07% :     0.000616s :      1: loop_unroll
 3.36% :     0.000999s :      1: mutable_eliminate
 8.75% :     0.002605s :     26: opt.transform.jit_opt_a
 0.20% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000032s :      4: opt.transform.jit_opt_b
 0.06% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000027s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.20% :     0.000655s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000043s :      1: py_interpret_to_execute
 0.07% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000019s :      1: remove_dup_value
 1.70% :     0.000505s :      1: renormalize.infer
 1.40% :     0.000418s :      1: renormalize.specialize
 0.04% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000050s :      1: rewriter_after_opt_a
 0.27% :     0.000081s :      1: rewriter_before_opt_a
 0.30% :     0.000089s :      1: symbol_engine_optimizer
28.96% :     0.008620s :      1: task_emit
18.29% :     0.005442s :      1: type_inference
 0.24% :     0.000073s :      1: validate


TotalTime = 0.0418958, [33]
[bootstrap]: 0.0003468
[type_inference]: 0.00530899
[event_method]: 2.103e-05
[auto_monad]: 6.762e-05
[graph_reusing]: 6.39001e-06
[pre_auto_parallel]: 2.48002e-06
[py_interpret_to_execute]: 4.409e-05
[rewriter_before_opt_a]: 7.706e-05
[expand_dump_flag]: 3.42002e-06
[jit_opt_a]: 0.0269402, [2]
    [Cycle 1]: 0.00198883, [27]
        [switch_simplify]: 6.634e-05
        [loop_unroll]: 3.094e-05
        [a_1]: 0.00064168
        [with_stream_mark]: 2.553e-05
        [recompute_prepare]: 7.68999e-06
        [updatestate_depend_eliminate]: 4.08999e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [parameter_eliminate]: 1.87001e-06
        [specialize_transform]: 6.02999e-06
        [updatestate_useless_node_eliminater]: 5.11002e-06
        [accelerated_algorithm]: 5.49998e-06
        [meta_shard_fg_expand]: 2.15002e-06
        [get_grad_eliminate_]: 5.51e-06
        [merge_forward]: 3.30003e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.451e-05
        [j_node_and_user_rematch]: 9.91998e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 9.15001e-06
        [inline_without_move]: 5.25001e-06
        [renormalize]: 0.00086537
        [add_forward_monad_depend]: 6.31e-06
        [auto_monad_grad]: 2.73e-06
        [auto_monad_eliminator]: 1.708e-05
        [cse]: 2.981e-05
        [replace_applicator]: 1.505e-05
    [Cycle 2]: 0.00033701, [27]
        [switch_simplify]: 5.69999e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 9.312e-05
        [with_stream_mark]: 1.156e-05
        [recompute_prepare]: 5.05001e-06
        [updatestate_depend_eliminate]: 2.99999e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [parameter_eliminate]: 1.71e-06
        [specialize_transform]: 5.01997e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 5.23002e-06
        [meta_shard_fg_expand]: 1.61002e-06
        [get_grad_eliminate_]: 4.61002e-06
        [merge_forward]: 3.18998e-06
        [cell_reuse_recompute_pass]: 2.69001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.474e-05
        [j_node_and_user_rematch]: 9.08002e-06
        [meta_fg_expand]: 1.81e-06
        [replace_old_param]: 7.38999e-06
        [inline_without_move]: 4.84e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.50999e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 5.70001e-06
        [cse]: 1.18e-05
        [replace_applicator]: 7.95e-06
[py_interpret_to_execute_after_opt_a]: 1.267e-05
[rewriter_after_opt_a]: 3.762e-05
[convert_after_rewriter]: 6.96001e-06
[order_py_execute_after_rewriter]: 4.77e-06
[mutable_eliminate]: 0.00076847
[jit_opt_b]: 5.796e-05, [1]
    [Cycle 1]: 4.989e-05, [2]
        [frontend_op_eliminate]: 2.151e-05
        [inline_after_opt_a]: 1.672e-05
[cconv]: 3.041e-05
[loop_unroll]: 0.00043816
[jit_opt_after_cconv]: 0.00014363, [1]
    [Cycle 1]: 0.00013675, [11]
        [c_1]: 2.093e-05
        [parameter_eliminate]: 3.48e-06
        [updatestate_depend_eliminate]: 6.51e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.51998e-06
        [cse]: 1.958e-05
        [call_graph_tuple_transform]: 1.902e-05
        [tuple_list_get_item_eliminator]: 5.86e-06
        [none_parameter_eliminate]: 1.34998e-06
        [renormalize]: 5.29981e-07
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 1.503e-05
[partial_unused_args_eliminate]: 1.94e-06
[environ_conv]: 4.80999e-06
[add_recomputation]: 5.18e-05
[cse_after_recomputation]: 2.062e-05, [1]
    [Cycle 1]: 1.503e-05, [1]
        [cse]: 9.22001e-06
[auto_monad_reorder]: 1.452e-05
[get_jit_bprop_graph]: 2.23002e-06
[rewriter_after_jit_bprop_graph]: 7.16001e-06
[opt_after_jit_grad]: 0.00046364
[symbol_engine_optimizer]: 9.097e-05, [1]
    [Cycle 1]: 8.379e-05, [6]
        [build]: 4.15e-06
        [elim_shapecalc]: 7.10998e-06
        [elim_not_effective]: 1.431e-05
        [opt_reshape]: 5.71e-06
        [fold_const_symbol]: 8.78001e-06
        [renormalize]: 4.80009e-07
[validate]: 4.283e-05
[backend_pass]: 1.00999e-06
[task_emit]: 0.00662788
[execute]: 9.29998e-06
Sums
    bootstrap                            :     0.000347s :  2.09%
    type_inference                       :     0.005309s : 31.97%
    event_method                         :     0.000021s :  0.13%
    auto_monad                           :     0.000068s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.27%
    rewriter_before_opt_a                :     0.000077s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000072s :  0.43%
    jit_opt_a.loop_unroll                :     0.000036s :  0.22%
    jit_opt_a.a_1                        :     0.000735s :  4.42%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000865s :  5.21%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.14%
    jit_opt_a.cse                        :     0.000042s :  0.25%
    jit_opt_a.replace_applicator         :     0.000023s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000768s :  4.63%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.10%
    cconv                                :     0.000030s :  0.18%
    loop_unroll                          :     0.000438s :  2.64%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000052s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000464s :  2.79%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000043s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006628s : 39.91%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000289    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 1.88% :     0.000005s :      3: substitution.graph_param_transform
87.55% :     0.000253s :      4: substitution.inline
 1.60% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.67% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.64% :     0.000005s :      2: substitution.replace_old_param
 4.50% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005230     2
81.67% :     0.004271s :      1: type_inference.infer
18.33% :     0.000959s :      1: type_inference.specialize
------[replace.]   0.000077     6
69.07% :     0.000053s :      4: replace.inline
30.93% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000262     6
95.46% :     0.000251s :      4: match.inline
 4.54% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.47% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.21% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 2.47% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.09% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.47% :     0.000001s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.10% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.84% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.34% :     0.000002s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.17% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.88% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 2.88% :     0.000004s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.35% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.50% :     0.000001s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.10% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.61% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.50% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.97% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 2.00% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.94% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000719    12
46.33% :     0.000333s :      5: func_graph_cloner_run.FuncGraphClonerGraph
12.21% :     0.000088s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.46% :     0.000298s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.043830    76
 0.12% :     0.000054s :      1: add_recomputation
 0.16% :     0.000071s :      1: auto_monad
 0.04% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.83% :     0.000364s :      1: bootstrap
 0.08% :     0.000033s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.05% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.06% :     0.000026s :      1: event_method
 0.03% :     0.000015s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
61.47% :     0.026944s :      1: jit_opt_a
 0.33% :     0.000147s :      1: jit_opt_after_cconv
 0.14% :     0.000061s :      1: jit_opt_b
 1.02% :     0.000446s :      1: loop_unroll
 1.78% :     0.000778s :      1: mutable_eliminate
 2.20% :     0.000966s :     26: opt.transform.jit_opt_a
 0.11% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000031s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.08% :     0.000472s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000047s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000017s :      1: remove_dup_value
 1.10% :     0.000482s :      1: renormalize.infer
 0.85% :     0.000374s :      1: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000041s :      1: rewriter_after_opt_a
 0.18% :     0.000081s :      1: rewriter_before_opt_a
 0.21% :     0.000094s :      1: symbol_engine_optimizer
15.17% :     0.006647s :      1: task_emit
12.16% :     0.005329s :      1: type_inference
 0.15% :     0.000064s :      1: validate


TotalTime = 0.0206254, [33]
[bootstrap]: 0.00035087
[type_inference]: 0.00503497
[event_method]: 1.84e-05
[auto_monad]: 6.445e-05
[graph_reusing]: 5.92999e-06
[pre_auto_parallel]: 2.64999e-06
[py_interpret_to_execute]: 3.601e-05
[rewriter_before_opt_a]: 7.22e-05
[expand_dump_flag]: 3.07002e-06
[jit_opt_a]: 0.00612939, [2]
    [Cycle 1]: 0.00188442, [27]
        [switch_simplify]: 6.021e-05
        [loop_unroll]: 2.985e-05
        [a_1]: 0.00057795
        [with_stream_mark]: 2.337e-05
        [recompute_prepare]: 8.08999e-06
        [updatestate_depend_eliminate]: 4.14002e-06
        [updatestate_assign_eliminate]: 3.21999e-06
        [updatestate_loads_eliminate]: 2.51998e-06
        [parameter_eliminate]: 1.97999e-06
        [specialize_transform]: 6.28e-06
        [updatestate_useless_node_eliminater]: 4.96002e-06
        [accelerated_algorithm]: 5.37999e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 5.35001e-06
        [merge_forward]: 4.28999e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.448e-05
        [j_node_and_user_rematch]: 1.016e-05
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 9.61998e-06
        [inline_without_move]: 5.05001e-06
        [renormalize]: 0.00084975
        [add_forward_monad_depend]: 5.94999e-06
        [auto_monad_grad]: 2.97002e-06
        [auto_monad_eliminator]: 1.434e-05
        [cse]: 3.036e-05
        [replace_applicator]: 1.328e-05
    [Cycle 2]: 0.0003248, [27]
        [switch_simplify]: 5.76998e-06
        [loop_unroll]: 5.07e-06
        [a_1]: 9.045e-05
        [with_stream_mark]: 1.059e-05
        [recompute_prepare]: 4.98001e-06
        [updatestate_depend_eliminate]: 3.03e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 5.47001e-06
        [updatestate_useless_node_eliminater]: 5.15999e-06
        [accelerated_algorithm]: 4.74e-06
        [meta_shard_fg_expand]: 1.25001e-06
        [get_grad_eliminate_]: 4.33001e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 2.44999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.298e-05
        [j_node_and_user_rematch]: 8.02e-06
        [meta_fg_expand]: 1.64e-06
        [replace_old_param]: 6.94001e-06
        [inline_without_move]: 4.85001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.52001e-06
        [auto_monad_grad]: 8.59989e-07
        [auto_monad_eliminator]: 4.91002e-06
        [cse]: 1.088e-05
        [replace_applicator]: 5.14e-06
[py_interpret_to_execute_after_opt_a]: 1.253e-05
[rewriter_after_opt_a]: 3.662e-05
[convert_after_rewriter]: 7.28e-06
[order_py_execute_after_rewriter]: 5.18002e-06
[mutable_eliminate]: 0.00075756
[jit_opt_b]: 5.078e-05, [1]
    [Cycle 1]: 4.306e-05, [2]
        [frontend_op_eliminate]: 1.696e-05
        [inline_after_opt_a]: 1.455e-05
[cconv]: 2.654e-05
[loop_unroll]: 0.00044651
[jit_opt_after_cconv]: 0.00014396, [1]
    [Cycle 1]: 0.00013695, [11]
        [c_1]: 2.105e-05
        [parameter_eliminate]: 3.31001e-06
        [updatestate_depend_eliminate]: 6.40997e-06
        [updatestate_assign_eliminate]: 2.64999e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [cse]: 2.018e-05
        [call_graph_tuple_transform]: 1.83e-05
        [tuple_list_get_item_eliminator]: 5.03002e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 5.79981e-07
        [switch_simplify]: 5.28002e-06
[remove_dup_value]: 1.398e-05
[partial_unused_args_eliminate]: 1.97999e-06
[environ_conv]: 4.96997e-06
[add_recomputation]: 5.114e-05
[cse_after_recomputation]: 2.052e-05, [1]
    [Cycle 1]: 1.517e-05, [1]
        [cse]: 9.04e-06
[auto_monad_reorder]: 1.549e-05
[get_jit_bprop_graph]: 2.10002e-06
[rewriter_after_jit_bprop_graph]: 5.24e-06
[opt_after_jit_grad]: 0.0004613
[symbol_engine_optimizer]: 7.399e-05, [1]
    [Cycle 1]: 6.674e-05, [6]
        [build]: 4.82e-06
        [elim_shapecalc]: 7.6e-06
        [elim_not_effective]: 1.252e-05
        [opt_reshape]: 5.40001e-06
        [fold_const_symbol]: 8.64998e-06
        [renormalize]: 5.39992e-07
[validate]: 4.093e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00650492
[execute]: 8.37e-06
Sums
    bootstrap                            :     0.000351s :  2.18%
    type_inference                       :     0.005035s : 31.34%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000064s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.22%
    rewriter_before_opt_a                :     0.000072s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.41%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000668s :  4.16%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000850s :  5.29%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000041s :  0.26%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000037s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000758s :  4.72%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000027s :  0.17%
    loop_unroll                          :     0.000447s :  2.78%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000051s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000461s :  2.87%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006505s : 40.49%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000235    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.26% :     0.000005s :      3: substitution.graph_param_transform
85.87% :     0.000201s :      4: substitution.inline
 1.28% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.86% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.98% :     0.000005s :      2: substitution.replace_old_param
 5.30% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004959     2
80.00% :     0.003967s :      1: type_inference.infer
20.00% :     0.000992s :      1: type_inference.specialize
------[replace.]   0.000076     6
67.02% :     0.000051s :      4: replace.inline
32.98% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000210     6
94.57% :     0.000199s :      4: match.inline
 5.43% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.27% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.24% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 1.97% :     0.000002s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.12% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.21% :     0.000006s :     22: predicate.inline
 1.09% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.96% :     0.000001s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.94% :     0.000002s :     15: predicate.load_eliminater
 1.19% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 2.26% :     0.000003s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.50% :     0.000003s :     15: predicate.partial_eliminate
 1.29% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.27% :     0.000001s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.16% :     0.000001s :      6: predicate.special_op_eliminate
 1.40% :     0.000002s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.27% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.60% :     0.000009s :     53: predicate.switch_simplify
 1.45% :     0.000002s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.51% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.07% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.98% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000714    12
50.51% :     0.000361s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.63% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.86% :     0.000306s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022458    76
 0.24% :     0.000054s :      1: add_recomputation
 0.30% :     0.000068s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.65% :     0.000371s :      1: bootstrap
 0.13% :     0.000029s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.31% :     0.006134s :      1: jit_opt_a
 0.65% :     0.000147s :      1: jit_opt_after_cconv
 0.24% :     0.000053s :      1: jit_opt_b
 2.02% :     0.000455s :      1: loop_unroll
 3.42% :     0.000768s :      1: mutable_eliminate
 3.94% :     0.000886s :     26: opt.transform.jit_opt_a
 0.21% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.09% :     0.000469s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000039s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.06% :     0.000463s :      1: renormalize.infer
 1.68% :     0.000377s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000042s :      1: rewriter_after_opt_a
 0.33% :     0.000075s :      1: rewriter_before_opt_a
 0.34% :     0.000077s :      1: symbol_engine_optimizer
29.04% :     0.006521s :      1: task_emit
22.50% :     0.005052s :      1: type_inference
 0.27% :     0.000060s :      1: validate


TotalTime = 0.060426, [30]
[bootstrap]: 0.00043512
[type_inference]: 0.0242784
[event_method]: 2.906e-05
[auto_monad]: 9.164e-05
[graph_reusing]: 7.75e-06
[pre_auto_parallel]: 2.78e-06
[py_interpret_to_execute]: 6.129e-05
[rewriter_before_opt_a]: 9.25e-05
[expand_dump_flag]: 4.22e-06
[jit_opt_a]: 0.0323335, [2]
    [Cycle 1]: 0.002394, [27]
        [switch_simplify]: 6.759e-05
        [loop_unroll]: 3.129e-05
        [a_1]: 0.00066363
        [with_stream_mark]: 1.776e-05
        [recompute_prepare]: 9.45001e-06
        [updatestate_depend_eliminate]: 3.59002e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.69998e-06
        [specialize_transform]: 7.56001e-06
        [updatestate_useless_node_eliminater]: 5.57001e-06
        [accelerated_algorithm]: 6.94999e-06
        [meta_shard_fg_expand]: 2.69999e-06
        [get_grad_eliminate_]: 6.12999e-06
        [merge_forward]: 3.33e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.381e-05
        [j_node_and_user_rematch]: 1.084e-05
        [meta_fg_expand]: 2.74999e-06
        [replace_old_param]: 9.84001e-06
        [inline_without_move]: 7.1e-06
        [renormalize]: 0.00120556
        [add_forward_monad_depend]: 1.168e-05
        [auto_monad_grad]: 2.46e-06
        [auto_monad_eliminator]: 2.383e-05
        [cse]: 3.482e-05
        [replace_applicator]: 1.954e-05
    [Cycle 2]: 0.00044421, [27]
        [switch_simplify]: 7.30003e-06
        [loop_unroll]: 5.99e-06
        [a_1]: 0.00012737
        [with_stream_mark]: 2.423e-05
        [recompute_prepare]: 8.30999e-06
        [updatestate_depend_eliminate]: 5.40999e-06
        [updatestate_assign_eliminate]: 3.06999e-06
        [updatestate_loads_eliminate]: 2.67001e-06
        [parameter_eliminate]: 2.61e-06
        [specialize_transform]: 5.42001e-06
        [updatestate_useless_node_eliminater]: 5.82999e-06
        [accelerated_algorithm]: 6.83998e-06
        [meta_shard_fg_expand]: 2.98e-06
        [get_grad_eliminate_]: 5.27999e-06
        [merge_forward]: 4.07e-06
        [cell_reuse_recompute_pass]: 3.55e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.007e-05
        [j_node_and_user_rematch]: 9.19e-06
        [meta_fg_expand]: 2.41998e-06
        [replace_old_param]: 9.55001e-06
        [inline_without_move]: 4.92999e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 2.64001e-06
        [auto_monad_grad]: 8.99978e-07
        [auto_monad_eliminator]: 8.03001e-06
        [cse]: 1.205e-05
        [replace_applicator]: 5.84999e-06
[py_interpret_to_execute_after_opt_a]: 2.028e-05
[rewriter_after_opt_a]: 4.244e-05
[convert_after_rewriter]: 7.48999e-06
[order_py_execute_after_rewriter]: 5.44998e-06
[mutable_eliminate]: 0.00077479
[jit_opt_b]: 6.509e-05, [1]
    [Cycle 1]: 5.656e-05, [2]
        [frontend_op_eliminate]: 2.285e-05
        [inline_after_opt_a]: 2.026e-05
[cconv]: 3.751e-05
[loop_unroll]: 0.00062722
[jit_opt_after_cconv]: 0.00018907, [1]
    [Cycle 1]: 0.00017872, [11]
        [c_1]: 2.432e-05
        [parameter_eliminate]: 4.67998e-06
        [updatestate_depend_eliminate]: 9.49e-06
        [updatestate_assign_eliminate]: 2.89999e-06
        [updatestate_loads_eliminate]: 3.01001e-06
        [cse]: 3.199e-05
        [call_graph_tuple_transform]: 2.718e-05
        [tuple_list_get_item_eliminator]: 5.79999e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 1.00001e-06
        [switch_simplify]: 6.41e-06
[remove_dup_value]: 1.612e-05
[partial_unused_args_eliminate]: 2.44999e-06
[environ_conv]: 6.11998e-06
[add_recomputation]: 5.406e-05
[cse_after_recomputation]: 2.658e-05, [1]
    [Cycle 1]: 1.996e-05, [1]
        [cse]: 1.072e-05
[auto_monad_reorder]: 1.461e-05
[get_jit_bprop_graph]: 2.26998e-06
[rewriter_after_jit_bprop_graph]: 9.67001e-06
[opt_after_jit_grad]: 0.00076717
[symbol_engine_optimizer]: 0.00010103, [1]
    [Cycle 1]: 9.097e-05, [6]
        [build]: 6.29999e-06
        [elim_shapecalc]: 9.47001e-06
        [elim_not_effective]: 1.926e-05
        [opt_reshape]: 1.004e-05
        [fold_const_symbol]: 1.135e-05
        [renormalize]: 7.39994e-07
[validate]: 5.529e-05
Sums
    bootstrap                            :     0.000435s :  1.44%
    type_inference                       :     0.024278s : 80.53%
    event_method                         :     0.000029s :  0.10%
    auto_monad                           :     0.000092s :  0.30%
    graph_reusing                        :     0.000008s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000061s :  0.20%
    rewriter_before_opt_a                :     0.000092s :  0.31%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000075s :  0.25%
    jit_opt_a.loop_unroll                :     0.000037s :  0.12%
    jit_opt_a.a_1                        :     0.000791s :  2.62%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000013s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000014s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.06%
    jit_opt_a.inline_without_move        :     0.000012s :  0.04%
    jit_opt_a.renormalize                :     0.001206s :  4.00%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000032s :  0.11%
    jit_opt_a.cse                        :     0.000047s :  0.16%
    jit_opt_a.replace_applicator         :     0.000025s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000020s :  0.07%
    rewriter_after_opt_a                 :     0.000042s :  0.14%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000775s :  2.57%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.07%
    cconv                                :     0.000038s :  0.12%
    loop_unroll                          :     0.000627s :  2.08%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000032s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000054s :  0.18%
    cse_after_recomputation.cse          :     0.000011s :  0.04%
    auto_monad_reorder                   :     0.000015s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.03%
    opt_after_jit_grad                   :     0.000767s :  2.54%
    symbol_engine_optimizer.build        :     0.000006s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000055s :  0.18%


Time group info:
------[substitution.]   0.000273    23
 1.08% :     0.000003s :      2: substitution.elim_not_effective
 0.81% :     0.000002s :      2: substitution.fold_const_symbol
 2.85% :     0.000008s :      3: substitution.graph_param_transform
84.61% :     0.000231s :      4: substitution.inline
 1.37% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.09% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.81% :     0.000005s :      2: substitution.replace_old_param
 5.38% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.024133     2
93.24% :     0.022502s :      1: type_inference.infer
 6.76% :     0.001631s :      1: type_inference.specialize
------[replace.]   0.000093     6
63.93% :     0.000059s :      4: replace.inline
36.07% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000242     6
94.37% :     0.000228s :      4: match.inline
 5.63% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000148   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.81% :     0.000001s :     10: predicate.addn_check_dump
 1.46% :     0.000002s :     10: predicate.addn_zero_filter
 2.67% :     0.000004s :     10: predicate.arithmetic_simplify
 1.59% :     0.000002s :     10: predicate.cast_eliminate
 0.32% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.87% :     0.000001s :     10: predicate.compare_switch_simplify
 2.12% :     0.000003s :     10: predicate.depend_value_elim
 1.05% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.87% :     0.000003s :     10: predicate.dict_get_item_eliminator
 1.03% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000001s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.98% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.83% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.83% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.17% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 5.32% :     0.000008s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.27% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000002s :      6: predicate.less_batch_normalization
 1.82% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.97% :     0.000003s :     15: predicate.load_eliminater
 1.26% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.73% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.02% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.88% :     0.000001s :     10: predicate.merge_addn
 1.18% :     0.000002s :     10: predicate.minmaximum_grad
 1.64% :     0.000002s :      3: predicate.mutable_eliminate
 1.27% :     0.000002s :      3: predicate.opt_reshape
 1.77% :     0.000003s :     15: predicate.partial_eliminate
 1.51% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.09% :     0.000003s :     10: predicate.reduce_eliminate
 1.67% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000003s :     18: predicate.replace_applicator
 0.39% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000001s :      3: predicate.reset_defer_inline
 2.13% :     0.000003s :     10: predicate.reshape_eliminate
 1.07% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.14% :     0.000002s :      6: predicate.special_op_eliminate
 0.82% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.21% :     0.000003s :     16: predicate.switch_defer_inline
 1.99% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.70% :     0.000010s :     53: predicate.switch_simplify
 1.08% :     0.000002s :     10: predicate.tile_eliminate
 1.18% :     0.000002s :     10: predicate.transpose_eliminate
 1.65% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.05% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.15% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.95% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.37% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.69% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.35% :     0.000003s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001176    12
51.97% :     0.000611s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.03% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.00% :     0.000518s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.062743    72
 0.09% :     0.000059s :      1: add_recomputation
 0.15% :     0.000096s :      1: auto_monad
 0.03% :     0.000018s :      1: auto_monad_reorder
 0.74% :     0.000461s :      1: bootstrap
 0.06% :     0.000041s :      1: cconv
 0.02% :     0.000015s :      1: convert_after_rewriter
 0.05% :     0.000029s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.06% :     0.000035s :      1: event_method
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
51.54% :     0.032338s :      1: jit_opt_a
 0.31% :     0.000193s :      1: jit_opt_after_cconv
 0.11% :     0.000068s :      1: jit_opt_b
 1.02% :     0.000643s :      1: loop_unroll
 1.25% :     0.000786s :      1: mutable_eliminate
 1.67% :     0.001048s :     26: opt.transform.jit_opt_a
 0.10% :     0.000060s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000035s :      4: opt.transform.jit_opt_b
 0.03% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000036s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000046s :      4: opt.transform.symbol_engine_opt
 1.25% :     0.000782s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000066s :      1: py_interpret_to_execute
 0.04% :     0.000023s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000020s :      1: remove_dup_value
 0.95% :     0.000594s :      1: renormalize.infer
 0.95% :     0.000598s :      1: renormalize.specialize
 0.02% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000046s :      1: rewriter_after_opt_a
 0.16% :     0.000097s :      1: rewriter_before_opt_a
 0.17% :     0.000104s :      1: symbol_engine_optimizer
38.75% :     0.024311s :      1: type_inference


TotalTime = 0.0403962, [33]
[bootstrap]: 0.00046172
[type_inference]: 0.00793134
[event_method]: 2.629e-05
[auto_monad]: 7.999e-05
[graph_reusing]: 6.61999e-06
[pre_auto_parallel]: 3.63e-06
[py_interpret_to_execute]: 4.492e-05
[rewriter_before_opt_a]: 8.68e-05
[expand_dump_flag]: 3.45998e-06
[jit_opt_a]: 0.00984604, [2]
    [Cycle 1]: 0.00222531, [27]
        [switch_simplify]: 7.237e-05
        [loop_unroll]: 3.486e-05
        [a_1]: 0.00067507
        [with_stream_mark]: 3.205e-05
        [recompute_prepare]: 1.117e-05
        [updatestate_depend_eliminate]: 4.85001e-06
        [updatestate_assign_eliminate]: 3.73001e-06
        [updatestate_loads_eliminate]: 2.81999e-06
        [parameter_eliminate]: 2.63003e-06
        [specialize_transform]: 7.53e-06
        [updatestate_useless_node_eliminater]: 5.48002e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 2.74001e-06
        [get_grad_eliminate_]: 5.46998e-06
        [merge_forward]: 4.54002e-06
        [cell_reuse_recompute_pass]: 1.77999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.001e-05
        [j_node_and_user_rematch]: 1.033e-05
        [meta_fg_expand]: 2.74999e-06
        [replace_old_param]: 1.042e-05
        [inline_without_move]: 5.96998e-06
        [renormalize]: 0.00098796
        [add_forward_monad_depend]: 9.89999e-06
        [auto_monad_grad]: 3.03e-06
        [auto_monad_eliminator]: 2.086e-05
        [cse]: 3.538e-05
        [replace_applicator]: 1.546e-05
    [Cycle 2]: 0.0003656, [27]
        [switch_simplify]: 5.95002e-06
        [loop_unroll]: 5.09998e-06
        [a_1]: 9.523e-05
        [with_stream_mark]: 1.891e-05
        [recompute_prepare]: 5.97001e-06
        [updatestate_depend_eliminate]: 3.98999e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [parameter_eliminate]: 1.60999e-06
        [specialize_transform]: 5.56998e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 4.61002e-06
        [merge_forward]: 4.18001e-06
        [cell_reuse_recompute_pass]: 2.93998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.731e-05
        [j_node_and_user_rematch]: 8.89998e-06
        [meta_fg_expand]: 2.07999e-06
        [replace_old_param]: 8.13001e-06
        [inline_without_move]: 5.10999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.76e-06
        [auto_monad_grad]: 8.60018e-07
        [auto_monad_eliminator]: 8.51002e-06
        [cse]: 1.437e-05
        [replace_applicator]: 5.99999e-06
[py_interpret_to_execute_after_opt_a]: 1.669e-05
[rewriter_after_opt_a]: 4.499e-05
[convert_after_rewriter]: 8.61002e-06
[order_py_execute_after_rewriter]: 5.67999e-06
[mutable_eliminate]: 0.0008756
[jit_opt_b]: 5.977e-05, [1]
    [Cycle 1]: 4.898e-05, [2]
        [frontend_op_eliminate]: 1.771e-05
        [inline_after_opt_a]: 1.782e-05
[cconv]: 3.168e-05
[loop_unroll]: 0.00072532
[jit_opt_after_cconv]: 0.0001596, [1]
    [Cycle 1]: 0.00015113, [11]
        [c_1]: 2.291e-05
        [parameter_eliminate]: 3.04999e-06
        [updatestate_depend_eliminate]: 6.43e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.89001e-06
        [cse]: 2.519e-05
        [call_graph_tuple_transform]: 2.18e-05
        [tuple_list_get_item_eliminator]: 6.21e-06
        [none_parameter_eliminate]: 1.69998e-06
        [renormalize]: 6.69999e-07
        [switch_simplify]: 5.74e-06
[remove_dup_value]: 1.663e-05
[partial_unused_args_eliminate]: 2.34999e-06
[environ_conv]: 6.63e-06
[add_recomputation]: 6.23e-05
[cse_after_recomputation]: 2.378e-05, [1]
    [Cycle 1]: 1.792e-05, [1]
        [cse]: 1.064e-05
[auto_monad_reorder]: 1.63e-05
[get_jit_bprop_graph]: 2.28998e-06
[rewriter_after_jit_bprop_graph]: 9.48997e-06
[opt_after_jit_grad]: 0.00065012
[symbol_engine_optimizer]: 8.589e-05, [1]
    [Cycle 1]: 7.602e-05, [6]
        [build]: 5.97999e-06
        [elim_shapecalc]: 8.12e-06
        [elim_not_effective]: 1.556e-05
        [opt_reshape]: 6.12001e-06
        [fold_const_symbol]: 9.32001e-06
        [renormalize]: 1.05999e-06
[validate]: 4.683e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.0187427
[execute]: 1.241e-05
Sums
    bootstrap                            :     0.000462s :  1.43%
    type_inference                       :     0.007931s : 24.52%
    event_method                         :     0.000026s :  0.08%
    auto_monad                           :     0.000080s :  0.25%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000004s :  0.01%
    py_interpret_to_execute              :     0.000045s :  0.14%
    rewriter_before_opt_a                :     0.000087s :  0.27%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000078s :  0.24%
    jit_opt_a.loop_unroll                :     0.000040s :  0.12%
    jit_opt_a.a_1                        :     0.000770s :  2.38%
    jit_opt_a.with_stream_mark           :     0.000051s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000013s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000009s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000019s :  0.06%
    jit_opt_a.inline_without_move        :     0.000011s :  0.03%
    jit_opt_a.renormalize                :     0.000988s :  3.05%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000029s :  0.09%
    jit_opt_a.cse                        :     0.000050s :  0.15%
    jit_opt_a.replace_applicator         :     0.000021s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.05%
    rewriter_after_opt_a                 :     0.000045s :  0.14%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000876s :  2.71%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000032s :  0.10%
    loop_unroll                          :     0.000725s :  2.24%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000017s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000062s :  0.19%
    cse_after_recomputation.cse          :     0.000011s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.03%
    opt_after_jit_grad                   :     0.000650s :  2.01%
    symbol_engine_optimizer.build        :     0.000006s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000047s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.018743s : 57.93%
    execute                              :     0.000012s :  0.04%


Time group info:
------[substitution.]   0.000301    23
 0.65% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000002s :      2: substitution.fold_const_symbol
 2.11% :     0.000006s :      3: substitution.graph_param_transform
86.25% :     0.000260s :      4: substitution.inline
 1.51% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.23% :     0.000007s :      2: substitution.replace_old_param
 4.74% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007821     2
83.94% :     0.006565s :      1: type_inference.infer
16.06% :     0.001256s :      1: type_inference.specialize
------[replace.]   0.000086     6
68.31% :     0.000059s :      4: replace.inline
31.69% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000269     6
95.11% :     0.000256s :      4: match.inline
 4.89% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   691
 1.54% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.00% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.88% :     0.000001s :     10: predicate.addn_check_dump
 1.51% :     0.000002s :     10: predicate.addn_zero_filter
 1.89% :     0.000003s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.90% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 0.90% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.59% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.93% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.03% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.98% :     0.000001s :     10: predicate.environ_get_eliminate
 1.28% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.92% :     0.000008s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.50% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.40% :     0.000002s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000003s :     15: predicate.load_eliminater
 3.22% :     0.000004s :      3: predicate.loop_unroll_after_grad
 3.85% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.15% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.88% :     0.000001s :     10: predicate.merge_addn
 0.91% :     0.000001s :     10: predicate.minmaximum_grad
 2.70% :     0.000004s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.03% :     0.000003s :     15: predicate.partial_eliminate
 1.02% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.88% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.62% :     0.000001s :      3: predicate.reset_defer_inline
 1.89% :     0.000003s :     10: predicate.reshape_eliminate
 1.24% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.06% :     0.000001s :     10: predicate.same_eliminate
 0.64% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.16% :     0.000002s :      6: predicate.special_op_eliminate
 0.79% :     0.000001s :      6: predicate.specialize_transform
 1.11% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.04% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.23% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000010s :     53: predicate.switch_simplify
 1.02% :     0.000001s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.26% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.03% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.38% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.84% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.48% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.49% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000856    12
47.78% :     0.000409s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.40% :     0.000063s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.82% :     0.000384s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042526    76
 0.15% :     0.000066s :      1: add_recomputation
 0.20% :     0.000085s :      1: auto_monad
 0.04% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.15% :     0.000489s :      1: bootstrap
 0.08% :     0.000035s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.06% :     0.000026s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.07% :     0.000032s :      1: event_method
 0.04% :     0.000019s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
23.16% :     0.009850s :      1: jit_opt_a
 0.38% :     0.000163s :      1: jit_opt_after_cconv
 0.15% :     0.000064s :      1: jit_opt_b
 1.73% :     0.000737s :      1: loop_unroll
 2.10% :     0.000893s :      1: mutable_eliminate
 2.40% :     0.001022s :     26: opt.transform.jit_opt_a
 0.12% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000022s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.57% :     0.000666s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.11% :     0.000048s :      1: py_interpret_to_execute
 0.05% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 1.29% :     0.000550s :      1: renormalize.infer
 1.00% :     0.000425s :      1: renormalize.specialize
 0.03% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000049s :      1: rewriter_after_opt_a
 0.21% :     0.000091s :      1: rewriter_before_opt_a
 0.21% :     0.000090s :      1: symbol_engine_optimizer
44.14% :     0.018770s :      1: task_emit
18.73% :     0.007964s :      1: type_inference
 0.17% :     0.000074s :      1: validate


TotalTime = 0.0208195, [33]
[bootstrap]: 0.00030983
[type_inference]: 0.00493137
[event_method]: 1.83e-05
[auto_monad]: 6.462e-05
[graph_reusing]: 5.68002e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 4.071e-05
[rewriter_before_opt_a]: 7.493e-05
[expand_dump_flag]: 3.63999e-06
[jit_opt_a]: 0.00616785, [2]
    [Cycle 1]: 0.00218529, [27]
        [switch_simplify]: 6.247e-05
        [loop_unroll]: 3.201e-05
        [a_1]: 0.00057936
        [with_stream_mark]: 1.788e-05
        [recompute_prepare]: 8.54998e-06
        [updatestate_depend_eliminate]: 3.98999e-06
        [updatestate_assign_eliminate]: 2.94001e-06
        [updatestate_loads_eliminate]: 2.91e-06
        [parameter_eliminate]: 1.80001e-06
        [specialize_transform]: 6.17001e-06
        [updatestate_useless_node_eliminater]: 1.289e-05
        [accelerated_algorithm]: 7.36999e-06
        [meta_shard_fg_expand]: 2.88e-06
        [get_grad_eliminate_]: 6.42001e-06
        [merge_forward]: 6.73e-06
        [cell_reuse_recompute_pass]: 1.76998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.797e-05
        [j_node_and_user_rematch]: 1.044e-05
        [meta_fg_expand]: 3.32002e-06
        [replace_old_param]: 1.085e-05
        [inline_without_move]: 5.97001e-06
        [renormalize]: 0.00081896
        [add_forward_monad_depend]: 7.61999e-06
        [auto_monad_grad]: 2.49999e-06
        [auto_monad_eliminator]: 1.456e-05
        [cse]: 2.915e-05
        [replace_applicator]: 1.569e-05
    [Cycle 2]: 0.00035577, [27]
        [switch_simplify]: 5.69e-06
        [loop_unroll]: 5.27999e-06
        [a_1]: 0.00010107
        [with_stream_mark]: 1.148e-05
        [recompute_prepare]: 5.24e-06
        [updatestate_depend_eliminate]: 2.74001e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [parameter_eliminate]: 1.32e-06
        [specialize_transform]: 5.14998e-06
        [updatestate_useless_node_eliminater]: 5.25001e-06
        [accelerated_algorithm]: 5.94e-06
        [meta_shard_fg_expand]: 1.45001e-06
        [get_grad_eliminate_]: 4.70999e-06
        [merge_forward]: 2.80997e-06
        [cell_reuse_recompute_pass]: 2.31998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.437e-05
        [j_node_and_user_rematch]: 8.47e-06
        [meta_fg_expand]: 2.36e-06
        [replace_old_param]: 8.48001e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 1.00999e-06
        [auto_monad_eliminator]: 5.71e-06
        [cse]: 1.232e-05
        [replace_applicator]: 6.16e-06
[py_interpret_to_execute_after_opt_a]: 9.47001e-06
[rewriter_after_opt_a]: 3.483e-05
[convert_after_rewriter]: 7.33e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00068195
[jit_opt_b]: 5.777e-05, [1]
    [Cycle 1]: 5.02e-05, [2]
        [frontend_op_eliminate]: 2.075e-05
        [inline_after_opt_a]: 1.715e-05
[cconv]: 2.971e-05
[loop_unroll]: 0.00045494
[jit_opt_after_cconv]: 0.00015789, [1]
    [Cycle 1]: 0.00015097, [11]
        [c_1]: 2.331e-05
        [parameter_eliminate]: 3.23e-06
        [updatestate_depend_eliminate]: 7.85e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.46998e-06
        [cse]: 2.355e-05
        [call_graph_tuple_transform]: 2.146e-05
        [tuple_list_get_item_eliminator]: 6.02001e-06
        [none_parameter_eliminate]: 1.47999e-06
        [renormalize]: 5.39992e-07
        [switch_simplify]: 5.66e-06
[remove_dup_value]: 1.397e-05
[partial_unused_args_eliminate]: 2.31e-06
[environ_conv]: 5.60001e-06
[add_recomputation]: 5.043e-05
[cse_after_recomputation]: 2.275e-05, [1]
    [Cycle 1]: 1.607e-05, [1]
        [cse]: 9.44e-06
[auto_monad_reorder]: 1.617e-05
[get_jit_bprop_graph]: 1.94999e-06
[rewriter_after_jit_bprop_graph]: 7.62998e-06
[opt_after_jit_grad]: 0.0005014
[symbol_engine_optimizer]: 8.018e-05, [1]
    [Cycle 1]: 7.249e-05, [6]
        [build]: 3.9e-06
        [elim_shapecalc]: 8.07e-06
        [elim_not_effective]: 1.397e-05
        [opt_reshape]: 7e-06
        [fold_const_symbol]: 9.84001e-06
        [renormalize]: 6.39993e-07
[validate]: 9.958e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00671187
[execute]: 9.15999e-06
Sums
    bootstrap                            :     0.000310s :  1.91%
    type_inference                       :     0.004931s : 30.43%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000065s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000041s :  0.25%
    rewriter_before_opt_a                :     0.000075s :  0.46%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.42%
    jit_opt_a.loop_unroll                :     0.000037s :  0.23%
    jit_opt_a.a_1                        :     0.000680s :  4.20%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000018s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000010s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000819s :  5.05%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000041s :  0.26%
    jit_opt_a.replace_applicator         :     0.000022s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000035s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000682s :  4.21%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.11%
    cconv                                :     0.000030s :  0.18%
    loop_unroll                          :     0.000455s :  2.81%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000050s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000501s :  3.09%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000100s :  0.61%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006712s : 41.41%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000223    23
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.74% :     0.000006s :      3: substitution.graph_param_transform
83.62% :     0.000187s :      4: substitution.inline
 1.84% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.05% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000005s :      2: substitution.replace_old_param
 6.20% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004856     2
80.60% :     0.003914s :      1: type_inference.infer
19.40% :     0.000942s :      1: type_inference.specialize
------[replace.]   0.000079     6
68.33% :     0.000054s :      4: replace.inline
31.67% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000196     6
93.58% :     0.000184s :      4: match.inline
 6.42% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 2.07% :     0.000003s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.49% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000001s :      3: predicate.elim_not_effective
 0.71% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.26% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.03% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.17% :     0.000002s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.10% :     0.000007s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.62% :     0.000002s :      6: predicate.less_batch_normalization
 1.83% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.90% :     0.000003s :     15: predicate.load_eliminater
 1.15% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.73% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.15% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000002s :     10: predicate.minmaximum_grad
 2.14% :     0.000003s :      3: predicate.mutable_eliminate
 0.71% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     15: predicate.partial_eliminate
 1.49% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.61% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.07% :     0.000003s :     18: predicate.replace_applicator
 0.81% :     0.000001s :      6: predicate.replace_old_param
 0.39% :     0.000001s :      3: predicate.reset_defer_inline
 1.26% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.32% :     0.000002s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.19% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.43% :     0.000010s :     53: predicate.switch_simplify
 1.21% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000002s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.07% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.53% :     0.000005s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000682    12
49.93% :     0.000341s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.76% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.30% :     0.000309s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022673    76
 0.24% :     0.000054s :      1: add_recomputation
 0.30% :     0.000068s :      1: auto_monad
 0.09% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.46% :     0.000331s :      1: bootstrap
 0.14% :     0.000033s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
27.22% :     0.006172s :      1: jit_opt_a
 0.71% :     0.000161s :      1: jit_opt_after_cconv
 0.27% :     0.000061s :      1: jit_opt_b
 2.05% :     0.000464s :      1: loop_unroll
 3.05% :     0.000692s :      1: mutable_eliminate
 4.08% :     0.000925s :     26: opt.transform.jit_opt_a
 0.23% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000031s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.25% :     0.000510s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000044s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.86% :     0.000421s :      1: renormalize.infer
 1.72% :     0.000389s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000038s :      1: rewriter_after_opt_a
 0.34% :     0.000078s :      1: rewriter_before_opt_a
 0.36% :     0.000083s :      1: symbol_engine_optimizer
29.68% :     0.006730s :      1: task_emit
21.82% :     0.004948s :      1: type_inference
 0.55% :     0.000124s :      1: validate


TotalTime = 0.042228, [33]
[bootstrap]: 0.00032032
[type_inference]: 0.00495011
[event_method]: 1.724e-05
[auto_monad]: 6.2e-05
[graph_reusing]: 5.54e-06
[pre_auto_parallel]: 2.73e-06
[py_interpret_to_execute]: 4.028e-05
[rewriter_before_opt_a]: 7.163e-05
[expand_dump_flag]: 3.32997e-06
[jit_opt_a]: 0.0269215, [2]
    [Cycle 1]: 0.00207143, [27]
        [switch_simplify]: 6.887e-05
        [loop_unroll]: 3.135e-05
        [a_1]: 0.00062111
        [with_stream_mark]: 2.606e-05
        [recompute_prepare]: 9.24e-06
        [updatestate_depend_eliminate]: 4.54998e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [parameter_eliminate]: 2.22001e-06
        [specialize_transform]: 6.48998e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.94e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 5.42001e-06
        [merge_forward]: 3.48e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.46e-05
        [j_node_and_user_rematch]: 9.87001e-06
        [meta_fg_expand]: 2.33998e-06
        [replace_old_param]: 9.48997e-06
        [inline_without_move]: 5.51e-06
        [renormalize]: 0.00095467
        [add_forward_monad_depend]: 8.48001e-06
        [auto_monad_grad]: 3.13e-06
        [auto_monad_eliminator]: 1.869e-05
        [cse]: 3.12e-05
        [replace_applicator]: 1.703e-05
    [Cycle 2]: 0.0003571, [27]
        [switch_simplify]: 6.39001e-06
        [loop_unroll]: 5.39e-06
        [a_1]: 0.00010112
        [with_stream_mark]: 1.284e-05
        [recompute_prepare]: 5.00999e-06
        [updatestate_depend_eliminate]: 3.16001e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 2.18002e-06
        [parameter_eliminate]: 1.45001e-06
        [specialize_transform]: 5.05001e-06
        [updatestate_useless_node_eliminater]: 5.87999e-06
        [accelerated_algorithm]: 5.20001e-06
        [meta_shard_fg_expand]: 1.81998e-06
        [get_grad_eliminate_]: 5.31998e-06
        [merge_forward]: 2.84001e-06
        [cell_reuse_recompute_pass]: 2.81e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.512e-05
        [j_node_and_user_rematch]: 8.72e-06
        [meta_fg_expand]: 1.60001e-06
        [replace_old_param]: 8.13999e-06
        [inline_without_move]: 5.05001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.86e-06
        [auto_monad_grad]: 1.35001e-06
        [auto_monad_eliminator]: 5.47001e-06
        [cse]: 1.365e-05
        [replace_applicator]: 5.15999e-06
[py_interpret_to_execute_after_opt_a]: 1.381e-05
[rewriter_after_opt_a]: 3.906e-05
[convert_after_rewriter]: 7.8e-06
[order_py_execute_after_rewriter]: 5.94e-06
[mutable_eliminate]: 0.00082469
[jit_opt_b]: 5.444e-05, [1]
    [Cycle 1]: 4.71e-05, [2]
        [frontend_op_eliminate]: 1.886e-05
        [inline_after_opt_a]: 1.623e-05
[cconv]: 3.234e-05
[loop_unroll]: 0.00051154
[jit_opt_after_cconv]: 0.00018367, [1]
    [Cycle 1]: 0.00017586, [11]
        [c_1]: 2.298e-05
        [parameter_eliminate]: 4.87e-06
        [updatestate_depend_eliminate]: 1.079e-05
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [cse]: 3.184e-05
        [call_graph_tuple_transform]: 2.621e-05
        [tuple_list_get_item_eliminator]: 5.61998e-06
        [none_parameter_eliminate]: 1.87999e-06
        [renormalize]: 8.50006e-07
        [switch_simplify]: 6.65002e-06
[remove_dup_value]: 1.549e-05
[partial_unused_args_eliminate]: 3.09001e-06
[environ_conv]: 6.43e-06
[add_recomputation]: 5.624e-05
[cse_after_recomputation]: 2.247e-05, [1]
    [Cycle 1]: 1.623e-05, [1]
        [cse]: 9.34998e-06
[auto_monad_reorder]: 1.675e-05
[get_jit_bprop_graph]: 3.21001e-06
[rewriter_after_jit_bprop_graph]: 1.036e-05
[opt_after_jit_grad]: 0.00059161
[symbol_engine_optimizer]: 8.443e-05, [1]
    [Cycle 1]: 7.57e-05, [6]
        [build]: 5.02999e-06
        [elim_shapecalc]: 8.25999e-06
        [elim_not_effective]: 1.522e-05
        [opt_reshape]: 6.28e-06
        [fold_const_symbol]: 9.47001e-06
        [renormalize]: 8.39995e-07
[validate]: 5.275e-05
[backend_pass]: 1.10999e-06
[task_emit]: 0.00705504
[execute]: 9.35001e-06
Sums
    bootstrap                            :     0.000320s :  1.88%
    type_inference                       :     0.004950s : 29.04%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000062s :  0.36%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.24%
    rewriter_before_opt_a                :     0.000072s :  0.42%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000075s :  0.44%
    jit_opt_a.loop_unroll                :     0.000037s :  0.22%
    jit_opt_a.a_1                        :     0.000722s :  4.24%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000955s :  5.60%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.14%
    jit_opt_a.cse                        :     0.000045s :  0.26%
    jit_opt_a.replace_applicator         :     0.000022s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000039s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000825s :  4.84%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000032s :  0.19%
    loop_unroll                          :     0.000512s :  3.00%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000032s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.04%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000056s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.06%
    opt_after_jit_grad                   :     0.000592s :  3.47%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000053s :  0.31%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007055s : 41.39%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000266    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000002s :      2: substitution.fold_const_symbol
 2.44% :     0.000006s :      3: substitution.graph_param_transform
86.13% :     0.000229s :      4: substitution.inline
 1.51% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.83% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.91% :     0.000005s :      2: substitution.replace_old_param
 4.73% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004880     2
81.30% :     0.003967s :      1: type_inference.infer
18.70% :     0.000912s :      1: type_inference.specialize
------[replace.]   0.000077     6
68.03% :     0.000053s :      4: replace.inline
31.97% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000238     6
95.17% :     0.000227s :      4: match.inline
 4.83% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.51% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.61% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.93% :     0.000002s :     10: predicate.addn_zero_filter
 2.42% :     0.000003s :     10: predicate.arithmetic_simplify
 1.54% :     0.000002s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.49% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.42% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     10: predicate.environ_get_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.25% :     0.000002s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.59% :     0.000007s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.99% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.82% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.41% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     15: predicate.partial_eliminate
 1.09% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.94% :     0.000003s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.43% :     0.000001s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.15% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.57% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.17% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.45% :     0.000010s :     53: predicate.switch_simplify
 1.16% :     0.000002s :     10: predicate.tile_eliminate
 1.19% :     0.000002s :     10: predicate.transpose_eliminate
 1.26% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.07% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.70% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.30% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000669    12
45.20% :     0.000302s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.44% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.36% :     0.000310s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.044267    76
 0.14% :     0.000060s :      1: add_recomputation
 0.15% :     0.000065s :      1: auto_monad
 0.05% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.76% :     0.000335s :      1: bootstrap
 0.08% :     0.000035s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.06% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.05% :     0.000021s :      1: event_method
 0.03% :     0.000014s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000006s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
60.83% :     0.026926s :      1: jit_opt_a
 0.42% :     0.000187s :      1: jit_opt_after_cconv
 0.13% :     0.000057s :      1: jit_opt_b
 1.18% :     0.000520s :      1: loop_unroll
 1.89% :     0.000835s :      1: mutable_eliminate
 2.17% :     0.000960s :     26: opt.transform.jit_opt_a
 0.13% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000028s :      4: opt.transform.jit_opt_b
 0.03% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.37% :     0.000606s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.10% :     0.000043s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000018s :      1: remove_dup_value
 1.17% :     0.000517s :      1: renormalize.infer
 0.97% :     0.000430s :      1: renormalize.specialize
 0.03% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000042s :      1: rewriter_after_opt_a
 0.17% :     0.000074s :      1: rewriter_before_opt_a
 0.20% :     0.000088s :      1: symbol_engine_optimizer
15.98% :     0.007075s :      1: task_emit
11.21% :     0.004964s :      1: type_inference
 0.18% :     0.000080s :      1: validate


TotalTime = 0.0519722, [33]
[bootstrap]: 0.0003916
[type_inference]: 0.00635415
[event_method]: 2.492e-05
[auto_monad]: 7.591e-05
[graph_reusing]: 6.86001e-06
[pre_auto_parallel]: 2.97002e-06
[py_interpret_to_execute]: 5.022e-05
[rewriter_before_opt_a]: 9.013e-05
[expand_dump_flag]: 3.49001e-06
[jit_opt_a]: 0.0300682, [2]
    [Cycle 1]: 0.0251027, [27]
        [switch_simplify]: 7.24e-05
        [loop_unroll]: 3.186e-05
        [a_1]: 0.00069253
        [with_stream_mark]: 3.158e-05
        [recompute_prepare]: 1.332e-05
        [updatestate_depend_eliminate]: 5.27001e-06
        [updatestate_assign_eliminate]: 3.4e-06
        [updatestate_loads_eliminate]: 2.94999e-06
        [parameter_eliminate]: 2.79001e-06
        [specialize_transform]: 8.27e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 6.58e-06
        [meta_shard_fg_expand]: 2.39001e-06
        [get_grad_eliminate_]: 5.86e-06
        [merge_forward]: 4.17e-06
        [cell_reuse_recompute_pass]: 2.43e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.978e-05
        [j_node_and_user_rematch]: 1.311e-05
        [meta_fg_expand]: 2.56e-06
        [replace_old_param]: 1.234e-05
        [inline_without_move]: 7.56001e-06
        [renormalize]: 0.0237543
        [add_forward_monad_depend]: 1.492e-05
        [auto_monad_grad]: 2.76e-06
        [auto_monad_eliminator]: 3.239e-05
        [cse]: 3.869e-05
        [replace_applicator]: 3.045e-05
    [Cycle 2]: 0.00044243, [27]
        [switch_simplify]: 6.61999e-06
        [loop_unroll]: 6.11998e-06
        [a_1]: 0.00010775
        [with_stream_mark]: 2.115e-05
        [recompute_prepare]: 7.4e-06
        [updatestate_depend_eliminate]: 4.07e-06
        [updatestate_assign_eliminate]: 3.71999e-06
        [updatestate_loads_eliminate]: 2.81999e-06
        [parameter_eliminate]: 2.21998e-06
        [specialize_transform]: 6.68998e-06
        [updatestate_useless_node_eliminater]: 5.13002e-06
        [accelerated_algorithm]: 6.60002e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 4.94e-06
        [merge_forward]: 5.22e-06
        [cell_reuse_recompute_pass]: 5.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.158e-05
        [j_node_and_user_rematch]: 1.103e-05
        [meta_fg_expand]: 2.83e-06
        [replace_old_param]: 1.145e-05
        [inline_without_move]: 5.09998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 3.47997e-06
        [auto_monad_grad]: 1.25001e-06
        [auto_monad_eliminator]: 1.302e-05
        [cse]: 1.764e-05
        [replace_applicator]: 8.47e-06
[py_interpret_to_execute_after_opt_a]: 2.666e-05
[rewriter_after_opt_a]: 7.554e-05
[convert_after_rewriter]: 9.59e-06
[order_py_execute_after_rewriter]: 9.44e-06
[mutable_eliminate]: 0.00098185
[jit_opt_b]: 6.678e-05, [1]
    [Cycle 1]: 5.372e-05, [2]
        [frontend_op_eliminate]: 1.984e-05
        [inline_after_opt_a]: 1.957e-05
[cconv]: 4.486e-05
[loop_unroll]: 0.00064985
[jit_opt_after_cconv]: 0.00021012, [1]
    [Cycle 1]: 0.00020061, [11]
        [c_1]: 2.81e-05
        [parameter_eliminate]: 7.44002e-06
        [updatestate_depend_eliminate]: 1.395e-05
        [updatestate_assign_eliminate]: 3.35003e-06
        [updatestate_loads_eliminate]: 3.18998e-06
        [cse]: 3.775e-05
        [call_graph_tuple_transform]: 2.932e-05
        [tuple_list_get_item_eliminator]: 6.36e-06
        [none_parameter_eliminate]: 2.29001e-06
        [renormalize]: 5.50004e-07
        [switch_simplify]: 5.71e-06
[remove_dup_value]: 1.71e-05
[partial_unused_args_eliminate]: 2.56998e-06
[environ_conv]: 6.46999e-06
[add_recomputation]: 6.814e-05
[cse_after_recomputation]: 3.217e-05, [1]
    [Cycle 1]: 2.434e-05, [1]
        [cse]: 1.347e-05
[auto_monad_reorder]: 1.87e-05
[get_jit_bprop_graph]: 2.33998e-06
[rewriter_after_jit_bprop_graph]: 1.193e-05
[opt_after_jit_grad]: 0.00067653
[symbol_engine_optimizer]: 9.475e-05, [1]
    [Cycle 1]: 8.48e-05, [6]
        [build]: 7.03e-06
        [elim_shapecalc]: 1.065e-05
        [elim_not_effective]: 1.882e-05
        [opt_reshape]: 6.38998e-06
        [fold_const_symbol]: 8.59002e-06
        [renormalize]: 7.39994e-07
[validate]: 5.5e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.0115096
[execute]: 1.133e-05
Sums
    bootstrap                            :     0.000392s :  0.84%
    type_inference                       :     0.006354s : 13.65%
    event_method                         :     0.000025s :  0.05%
    auto_monad                           :     0.000076s :  0.16%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000050s :  0.11%
    rewriter_before_opt_a                :     0.000090s :  0.19%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000079s :  0.17%
    jit_opt_a.loop_unroll                :     0.000038s :  0.08%
    jit_opt_a.a_1                        :     0.000800s :  1.72%
    jit_opt_a.with_stream_mark           :     0.000053s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000015s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.02%
    jit_opt_a.merge_forward              :     0.000009s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000008s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000041s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000024s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000024s :  0.05%
    jit_opt_a.inline_without_move        :     0.000013s :  0.03%
    jit_opt_a.renormalize                :     0.023754s : 51.05%
    jit_opt_a.add_forward_monad_depend   :     0.000018s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000045s :  0.10%
    jit_opt_a.cse                        :     0.000056s :  0.12%
    jit_opt_a.replace_applicator         :     0.000039s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000027s :  0.06%
    rewriter_after_opt_a                 :     0.000076s :  0.16%
    convert_after_rewriter               :     0.000010s :  0.02%
    order_py_execute_after_rewriter      :     0.000009s :  0.02%
    mutable_eliminate                    :     0.000982s :  2.11%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.04%
    cconv                                :     0.000045s :  0.10%
    loop_unroll                          :     0.000650s :  1.40%
    jit_opt_after_cconv.c_1              :     0.000028s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000014s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000038s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000029s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.01%
    remove_dup_value                     :     0.000017s :  0.04%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.01%
    add_recomputation                    :     0.000068s :  0.15%
    cse_after_recomputation.cse          :     0.000013s :  0.03%
    auto_monad_reorder                   :     0.000019s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000012s :  0.03%
    opt_after_jit_grad                   :     0.000677s :  1.45%
    symbol_engine_optimizer.build        :     0.000007s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000055s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011510s : 24.73%
    execute                              :     0.000011s :  0.02%


Time group info:
------[substitution.]   0.000319    23
 0.64% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000002s :      2: substitution.fold_const_symbol
 2.15% :     0.000007s :      3: substitution.graph_param_transform
85.25% :     0.000272s :      4: substitution.inline
 1.85% :     0.000006s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.81% :     0.000009s :      2: substitution.replace_old_param
 4.82% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006250     2
82.70% :     0.005168s :      1: type_inference.infer
17.30% :     0.001081s :      1: type_inference.specialize
------[replace.]   0.000086     6
68.90% :     0.000060s :      4: replace.inline
31.10% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000283     6
94.99% :     0.000269s :      4: match.inline
 5.01% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000154   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.35% :     0.000004s :      3: predicate.ad_related_special_op_eliminate
 0.79% :     0.000001s :     10: predicate.addn_check_dump
 1.52% :     0.000002s :     10: predicate.addn_zero_filter
 2.74% :     0.000004s :     10: predicate.arithmetic_simplify
 1.57% :     0.000002s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.79% :     0.000001s :     10: predicate.compare_switch_simplify
 1.07% :     0.000002s :     10: predicate.depend_value_elim
 0.89% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.94% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.52% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000001s :      3: predicate.elim_not_effective
 0.72% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.97% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.81% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     10: predicate.environ_get_eliminate
 0.84% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 7.51% :     0.000012s :     22: predicate.inline
 1.39% :     0.000002s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.59% :     0.000002s :      6: predicate.less_batch_normalization
 1.64% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.60% :     0.000002s :     15: predicate.load_eliminater
 2.01% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.58% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.81% :     0.000001s :     10: predicate.merge_addn
 0.79% :     0.000001s :     10: predicate.minmaximum_grad
 2.50% :     0.000004s :      3: predicate.mutable_eliminate
 0.72% :     0.000001s :      3: predicate.opt_reshape
 1.96% :     0.000003s :     15: predicate.partial_eliminate
 0.94% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.36% :     0.000002s :     10: predicate.reduce_eliminate
 1.31% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.04% :     0.000003s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.43% :     0.000001s :      3: predicate.reset_defer_inline
 1.57% :     0.000002s :     10: predicate.reshape_eliminate
 0.94% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.19% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.05% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 1.33% :     0.000002s :      6: predicate.specialize_transform
 1.14% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.00% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.49% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.15% :     0.000003s :     16: predicate.switch_defer_inline
 1.58% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.54% :     0.000012s :     53: predicate.switch_simplify
 1.07% :     0.000002s :     10: predicate.tile_eliminate
 1.23% :     0.000002s :     10: predicate.transpose_eliminate
 1.18% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.85% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.54% :     0.000007s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.34% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.29% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.45% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.39% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.88% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000896    12
43.84% :     0.000393s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.20% :     0.000065s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.96% :     0.000439s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.076956    76
 0.10% :     0.000073s :      1: add_recomputation
 0.11% :     0.000081s :      1: auto_monad
 0.03% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.54% :     0.000414s :      1: bootstrap
 0.06% :     0.000048s :      1: cconv
 0.02% :     0.000014s :      1: convert_after_rewriter
 0.04% :     0.000035s :      1: cse_after_recomputation
 0.01% :     0.000010s :      1: environ_conv
 0.04% :     0.000031s :      1: event_method
 0.02% :     0.000017s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.01% :     0.000010s :      1: graph_reusing
39.08% :     0.030073s :      1: jit_opt_a
 0.28% :     0.000215s :      1: jit_opt_after_cconv
 0.09% :     0.000070s :      1: jit_opt_b
 0.87% :     0.000666s :      1: loop_unroll
 1.30% :     0.001004s :      1: mutable_eliminate
 1.41% :     0.001085s :     26: opt.transform.jit_opt_a
 0.08% :     0.000065s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000030s :      4: opt.transform.jit_opt_b
 0.03% :     0.000023s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000028s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000037s :      1: opt.transform.opt_after_jit_grad
 0.05% :     0.000040s :      4: opt.transform.symbol_engine_opt
 0.90% :     0.000696s :      1: opt_after_jit_grad
 0.02% :     0.000013s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000054s :      1: py_interpret_to_execute
 0.04% :     0.000031s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000021s :      1: remove_dup_value
30.08% :     0.023147s :      1: renormalize.infer
 0.76% :     0.000584s :      1: renormalize.specialize
 0.02% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000084s :      1: rewriter_after_opt_a
 0.12% :     0.000095s :      1: rewriter_before_opt_a
 0.13% :     0.000098s :      1: symbol_engine_optimizer
14.99% :     0.011533s :      1: task_emit
 8.29% :     0.006381s :      1: type_inference
 0.11% :     0.000085s :      1: validate


TotalTime = 0.0148799, [30]
[bootstrap]: 0.00037534
[type_inference]: 0.00554307
[event_method]: 1.84e-05
[auto_monad]: 8.189e-05
[graph_reusing]: 5.71998e-06
[pre_auto_parallel]: 3.21001e-06
[py_interpret_to_execute]: 5.02e-05
[rewriter_before_opt_a]: 0.00010938
[expand_dump_flag]: 2.58998e-06
[jit_opt_a]: 0.00619547, [2]
    [Cycle 1]: 0.0019047, [27]
        [switch_simplify]: 6.267e-05
        [loop_unroll]: 3.059e-05
        [a_1]: 0.00056567
        [with_stream_mark]: 1.504e-05
        [recompute_prepare]: 7.56999e-06
        [updatestate_depend_eliminate]: 3.23e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [parameter_eliminate]: 9.60019e-07
        [specialize_transform]: 6.27001e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 6.46999e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 5.37001e-06
        [merge_forward]: 2.51998e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.314e-05
        [j_node_and_user_rematch]: 9.94001e-06
        [meta_fg_expand]: 1.68002e-06
        [replace_old_param]: 9.16002e-06
        [inline_without_move]: 5.84e-06
        [renormalize]: 0.00090148
        [add_forward_monad_depend]: 6.34999e-06
        [auto_monad_grad]: 1.94e-06
        [auto_monad_eliminator]: 1.493e-05
        [cse]: 2.532e-05
        [replace_applicator]: 1.492e-05
    [Cycle 2]: 0.00035669, [27]
        [switch_simplify]: 6.03998e-06
        [loop_unroll]: 5.19e-06
        [a_1]: 9.991e-05
        [with_stream_mark]: 1.302e-05
        [recompute_prepare]: 5.52999e-06
        [updatestate_depend_eliminate]: 3.56001e-06
        [updatestate_assign_eliminate]: 2.36998e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.69998e-06
        [specialize_transform]: 5.39998e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 6.57002e-06
        [meta_shard_fg_expand]: 2.29999e-06
        [get_grad_eliminate_]: 4.46002e-06
        [merge_forward]: 3.00002e-06
        [cell_reuse_recompute_pass]: 3.83999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.531e-05
        [j_node_and_user_rematch]: 8.42e-06
        [meta_fg_expand]: 1.37e-06
        [replace_old_param]: 8.57e-06
        [inline_without_move]: 5.09e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 1.32999e-06
        [auto_monad_eliminator]: 4.89003e-06
        [cse]: 1.126e-05
        [replace_applicator]: 1.013e-05
[py_interpret_to_execute_after_opt_a]: 1.349e-05
[rewriter_after_opt_a]: 3.395e-05
[convert_after_rewriter]: 6.43e-06
[order_py_execute_after_rewriter]: 4.99e-06
[mutable_eliminate]: 0.00073286
[jit_opt_b]: 5.925e-05, [1]
    [Cycle 1]: 5.228e-05, [2]
        [frontend_op_eliminate]: 2.426e-05
        [inline_after_opt_a]: 1.538e-05
[cconv]: 2.33e-05
[loop_unroll]: 0.00047935
[jit_opt_after_cconv]: 0.00014729, [1]
    [Cycle 1]: 0.00014055, [11]
        [c_1]: 2.19e-05
        [parameter_eliminate]: 3.43e-06
        [updatestate_depend_eliminate]: 5.88002e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [cse]: 2.065e-05
        [call_graph_tuple_transform]: 1.932e-05
        [tuple_list_get_item_eliminator]: 5.15999e-06
        [none_parameter_eliminate]: 1.07998e-06
        [renormalize]: 5.49975e-07
        [switch_simplify]: 5.66e-06
[remove_dup_value]: 9.16998e-06
[partial_unused_args_eliminate]: 5.758e-05
[environ_conv]: 5.67001e-06
[add_recomputation]: 4.262e-05
[cse_after_recomputation]: 2.295e-05, [1]
    [Cycle 1]: 1.731e-05, [1]
        [cse]: 9.96e-06
[auto_monad_reorder]: 1.213e-05
[get_jit_bprop_graph]: 1.87001e-06
[rewriter_after_jit_bprop_graph]: 7.71001e-06
[opt_after_jit_grad]: 0.00052499
[symbol_engine_optimizer]: 8.011e-05, [1]
    [Cycle 1]: 7.17e-05, [6]
        [build]: 3.97998e-06
        [elim_shapecalc]: 9.34998e-06
        [elim_not_effective]: 1.375e-05
        [opt_reshape]: 5.61e-06
        [fold_const_symbol]: 9.59e-06
        [renormalize]: 7.80012e-07
[validate]: 3.886e-05
Sums
    bootstrap                            :     0.000375s :  3.63%
    type_inference                       :     0.005543s : 53.68%
    event_method                         :     0.000018s :  0.18%
    auto_monad                           :     0.000082s :  0.79%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000050s :  0.49%
    rewriter_before_opt_a                :     0.000109s :  1.06%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000069s :  0.67%
    jit_opt_a.loop_unroll                :     0.000036s :  0.35%
    jit_opt_a.a_1                        :     0.000666s :  6.45%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.17%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000902s :  8.73%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.19%
    jit_opt_a.cse                        :     0.000037s :  0.35%
    jit_opt_a.replace_applicator         :     0.000025s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.13%
    rewriter_after_opt_a                 :     0.000034s :  0.33%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000733s :  7.10%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.15%
    cconv                                :     0.000023s :  0.23%
    loop_unroll                          :     0.000479s :  4.64%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000009s :  0.09%
    partial_unused_args_eliminate        :     0.000058s :  0.56%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000043s :  0.41%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000012s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.07%
    opt_after_jit_grad                   :     0.000525s :  5.08%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.38%


Time group info:
------[substitution.]   0.000217    23
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.24% :     0.000005s :      3: substitution.graph_param_transform
84.47% :     0.000184s :      4: substitution.inline
 1.65% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.48% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.17% :     0.000005s :      2: substitution.replace_old_param
 5.51% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005459     2
77.90% :     0.004253s :      1: type_inference.infer
22.10% :     0.001206s :      1: type_inference.specialize
------[replace.]   0.000077     6
65.87% :     0.000051s :      4: replace.inline
34.13% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     6
94.37% :     0.000181s :      4: match.inline
 5.63% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.61% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.37% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 2.25% :     0.000003s :     10: predicate.arithmetic_simplify
 1.23% :     0.000002s :     10: predicate.cast_eliminate
 0.75% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.20% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.12% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 5.51% :     0.000007s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 2.15% :     0.000003s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.58% :     0.000002s :     15: predicate.load_eliminater
 1.12% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.30% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.20% :     0.000002s :     10: predicate.minmaximum_grad
 1.84% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 1.90% :     0.000002s :     15: predicate.partial_eliminate
 1.53% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.33% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.67% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.49% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.35% :     0.000002s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.65% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.00% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.28% :     0.000009s :     53: predicate.switch_simplify
 1.37% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.56% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.94% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.84% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000707    12
48.18% :     0.000341s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.76% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.06% :     0.000333s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016744    72
 0.27% :     0.000046s :      1: add_recomputation
 0.51% :     0.000086s :      1: auto_monad
 0.09% :     0.000015s :      1: auto_monad_reorder
 2.36% :     0.000396s :      1: bootstrap
 0.16% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.13% :     0.000022s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
37.02% :     0.006199s :      1: jit_opt_a
 0.90% :     0.000150s :      1: jit_opt_after_cconv
 0.37% :     0.000062s :      1: jit_opt_b
 2.91% :     0.000487s :      1: loop_unroll
 4.43% :     0.000742s :      1: mutable_eliminate
 5.38% :     0.000900s :     26: opt.transform.jit_opt_a
 0.29% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000033s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.19% :     0.000534s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.36% :     0.000060s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.32% :     0.000053s :      1: py_interpret_to_execute
 0.09% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000012s :      1: remove_dup_value
 2.84% :     0.000476s :      1: renormalize.infer
 2.49% :     0.000417s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000037s :      1: rewriter_after_opt_a
 0.67% :     0.000113s :      1: rewriter_before_opt_a
 0.49% :     0.000083s :      1: symbol_engine_optimizer
33.22% :     0.005562s :      1: type_inference


TotalTime = 0.0342247, [33]
[bootstrap]: 0.00036599
[type_inference]: 0.00529867
[event_method]: 2.004e-05
[auto_monad]: 6.43e-05
[graph_reusing]: 5.79e-06
[pre_auto_parallel]: 3.31999e-06
[py_interpret_to_execute]: 4.084e-05
[rewriter_before_opt_a]: 7.655e-05
[expand_dump_flag]: 2.91999e-06
[jit_opt_a]: 0.00617949, [2]
    [Cycle 1]: 0.00188851, [27]
        [switch_simplify]: 5.965e-05
        [loop_unroll]: 3.121e-05
        [a_1]: 0.00059727
        [with_stream_mark]: 2.225e-05
        [recompute_prepare]: 8.23999e-06
        [updatestate_depend_eliminate]: 3.18e-06
        [updatestate_assign_eliminate]: 2.47001e-06
        [updatestate_loads_eliminate]: 3.28998e-06
        [parameter_eliminate]: 1.55999e-06
        [specialize_transform]: 7e-06
        [updatestate_useless_node_eliminater]: 5.33002e-06
        [accelerated_algorithm]: 6.49001e-06
        [meta_shard_fg_expand]: 2.14e-06
        [get_grad_eliminate_]: 5.71e-06
        [merge_forward]: 3.6e-06
        [cell_reuse_recompute_pass]: 1.50001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.348e-05
        [j_node_and_user_rematch]: 8.82e-06
        [meta_fg_expand]: 2.69001e-06
        [replace_old_param]: 9.54e-06
        [inline_without_move]: 5.86e-06
        [renormalize]: 0.00084168
        [add_forward_monad_depend]: 5.76998e-06
        [auto_monad_grad]: 2.19001e-06
        [auto_monad_eliminator]: 1.428e-05
        [cse]: 2.544e-05
        [replace_applicator]: 1.686e-05
    [Cycle 2]: 0.00034734, [27]
        [switch_simplify]: 6.29001e-06
        [loop_unroll]: 5.66e-06
        [a_1]: 9.566e-05
        [with_stream_mark]: 1.24e-05
        [recompute_prepare]: 5.12e-06
        [updatestate_depend_eliminate]: 3.12002e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [parameter_eliminate]: 1.67999e-06
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.81002e-06
        [accelerated_algorithm]: 5.81e-06
        [meta_shard_fg_expand]: 1.77001e-06
        [get_grad_eliminate_]: 4.83001e-06
        [merge_forward]: 2.83e-06
        [cell_reuse_recompute_pass]: 2.83e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.532e-05
        [j_node_and_user_rematch]: 7.93001e-06
        [meta_fg_expand]: 1.61998e-06
        [replace_old_param]: 7.31001e-06
        [inline_without_move]: 4.45999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.98002e-06
        [auto_monad_grad]: 1.40001e-06
        [auto_monad_eliminator]: 6.51999e-06
        [cse]: 1.124e-05
        [replace_applicator]: 5.24e-06
[py_interpret_to_execute_after_opt_a]: 1.394e-05
[rewriter_after_opt_a]: 3.634e-05
[convert_after_rewriter]: 7.46001e-06
[order_py_execute_after_rewriter]: 5.04e-06
[mutable_eliminate]: 0.00080667
[jit_opt_b]: 5.936e-05, [1]
    [Cycle 1]: 5.076e-05, [2]
        [frontend_op_eliminate]: 2.053e-05
        [inline_after_opt_a]: 1.84e-05
[cconv]: 2.857e-05
[loop_unroll]: 0.0127855
[jit_opt_after_cconv]: 0.00021799, [1]
    [Cycle 1]: 0.00020484, [11]
        [c_1]: 3.052e-05
        [parameter_eliminate]: 7.75e-06
        [updatestate_depend_eliminate]: 1.407e-05
        [updatestate_assign_eliminate]: 4.12003e-06
        [updatestate_loads_eliminate]: 2.81e-06
        [cse]: 4.272e-05
        [call_graph_tuple_transform]: 3.016e-05
        [tuple_list_get_item_eliminator]: 7.41999e-06
        [none_parameter_eliminate]: 1.86e-06
        [renormalize]: 1.20001e-06
        [switch_simplify]: 6.22001e-06
[remove_dup_value]: 1.892e-05
[partial_unused_args_eliminate]: 3.37002e-06
[environ_conv]: 6.99001e-06
[add_recomputation]: 6.449e-05
[cse_after_recomputation]: 2.32e-05, [1]
    [Cycle 1]: 1.702e-05, [1]
        [cse]: 1.078e-05
[auto_monad_reorder]: 1.595e-05
[get_jit_bprop_graph]: 2.49999e-06
[rewriter_after_jit_bprop_graph]: 1.36e-05
[opt_after_jit_grad]: 0.00086451
[symbol_engine_optimizer]: 8.739e-05, [1]
    [Cycle 1]: 7.783e-05, [6]
        [build]: 5.68002e-06
        [elim_shapecalc]: 9.14e-06
        [elim_not_effective]: 1.575e-05
        [opt_reshape]: 7e-06
        [fold_const_symbol]: 1.047e-05
        [renormalize]: 6.59988e-07
[validate]: 4.932e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00678098
[execute]: 9.68002e-06
Sums
    bootstrap                            :     0.000366s :  1.24%
    type_inference                       :     0.005299s : 17.92%
    event_method                         :     0.000020s :  0.07%
    auto_monad                           :     0.000064s :  0.22%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.14%
    rewriter_before_opt_a                :     0.000077s :  0.26%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000066s :  0.22%
    jit_opt_a.loop_unroll                :     0.000037s :  0.12%
    jit_opt_a.a_1                        :     0.000693s :  2.34%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.04%
    jit_opt_a.merge_forward              :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000017s :  0.06%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.000842s :  2.85%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.07%
    jit_opt_a.cse                        :     0.000037s :  0.12%
    jit_opt_a.replace_applicator         :     0.000022s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.05%
    rewriter_after_opt_a                 :     0.000036s :  0.12%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000807s :  2.73%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000029s :  0.10%
    loop_unroll                          :     0.012785s : 43.23%
    jit_opt_after_cconv.c_1              :     0.000031s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000014s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000043s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000030s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000019s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000064s :  0.22%
    cse_after_recomputation.cse          :     0.000011s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000014s :  0.05%
    opt_after_jit_grad                   :     0.000865s :  2.92%
    symbol_engine_optimizer.build        :     0.000006s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000049s :  0.17%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006781s : 22.93%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.000239    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.77% :     0.000002s :      2: substitution.fold_const_symbol
 2.86% :     0.000007s :      3: substitution.graph_param_transform
84.89% :     0.000203s :      4: substitution.inline
 1.20% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.59% :     0.000004s :      2: substitution.replace_old_param
 5.64% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005220     2
80.92% :     0.004224s :      1: type_inference.infer
19.08% :     0.000996s :      1: type_inference.specialize
------[replace.]   0.000080     6
65.06% :     0.000052s :      4: replace.inline
34.94% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000213     6
94.18% :     0.000201s :      4: match.inline
 5.82% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   691
 1.48% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.54% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.54% :     0.000002s :     10: predicate.addn_zero_filter
 2.43% :     0.000003s :     10: predicate.arithmetic_simplify
 1.06% :     0.000001s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.91% :     0.000001s :     10: predicate.compare_switch_simplify
 1.44% :     0.000002s :     10: predicate.depend_value_elim
 0.96% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.59% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.02% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.44% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.39% :     0.000002s :     10: predicate.environ_get_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.75% :     0.000001s :      6: predicate.get_grad_eliminate
 0.34% :     0.000000s :      3: predicate.graph_param_transform
 5.69% :     0.000008s :     22: predicate.inline
 0.78% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.51% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 4.04% :     0.000006s :      3: predicate.loop_unroll_after_grad
 3.60% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.24% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.88% :     0.000001s :     10: predicate.merge_addn
 0.91% :     0.000001s :     10: predicate.minmaximum_grad
 2.24% :     0.000003s :      3: predicate.mutable_eliminate
 0.62% :     0.000001s :      3: predicate.opt_reshape
 1.74% :     0.000002s :     15: predicate.partial_eliminate
 1.02% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.69% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.54% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.14% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.02% :     0.000001s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.25% :     0.000002s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.16% :     0.000003s :     16: predicate.switch_defer_inline
 1.81% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.14% :     0.000010s :     53: predicate.switch_simplify
 1.18% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000002s :     10: predicate.transpose_eliminate
 1.12% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.02% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.52% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.98% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.32% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.40% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.98% :     0.000003s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000655    12
47.01% :     0.000308s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.29% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.70% :     0.000306s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.036154    76
 0.19% :     0.000067s :      1: add_recomputation
 0.19% :     0.000068s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.06% :     0.000384s :      1: bootstrap
 0.09% :     0.000031s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.07% :     0.000024s :      1: event_method
 0.04% :     0.000015s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
17.10% :     0.006184s :      1: jit_opt_a
 0.61% :     0.000222s :      1: jit_opt_after_cconv
 0.17% :     0.000062s :      1: jit_opt_b
35.44% :     0.012814s :      1: loop_unroll
 2.26% :     0.000818s :      1: mutable_eliminate
 2.55% :     0.000921s :     26: opt.transform.jit_opt_a
 0.20% :     0.000071s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000030s :      4: opt.transform.jit_opt_b
 0.11% :     0.000041s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000039s :      4: opt.transform.symbol_engine_opt
 2.42% :     0.000877s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000044s :      1: py_interpret_to_execute
 0.05% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000022s :      1: remove_dup_value
 1.20% :     0.000434s :      1: renormalize.infer
 1.10% :     0.000398s :      1: renormalize.specialize
 0.04% :     0.000016s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000040s :      1: rewriter_after_opt_a
 0.22% :     0.000079s :      1: rewriter_before_opt_a
 0.25% :     0.000090s :      1: symbol_engine_optimizer
18.81% :     0.006801s :      1: task_emit
14.70% :     0.005315s :      1: type_inference
 0.21% :     0.000075s :      1: validate


TotalTime = 0.0454173, [33]
[bootstrap]: 0.00036686
[type_inference]: 0.00686325
[event_method]: 2.293e-05
[auto_monad]: 7.789e-05
[graph_reusing]: 6.14001e-06
[pre_auto_parallel]: 3.44001e-06
[py_interpret_to_execute]: 4.114e-05
[rewriter_before_opt_a]: 8.109e-05
[expand_dump_flag]: 3.55e-06
[jit_opt_a]: 0.00655295, [2]
    [Cycle 1]: 0.00203187, [27]
        [switch_simplify]: 6.647e-05
        [loop_unroll]: 3.014e-05
        [a_1]: 0.0006245
        [with_stream_mark]: 2.873e-05
        [recompute_prepare]: 1.093e-05
        [updatestate_depend_eliminate]: 4.42e-06
        [updatestate_assign_eliminate]: 2.88e-06
        [updatestate_loads_eliminate]: 2.93e-06
        [parameter_eliminate]: 2.34001e-06
        [specialize_transform]: 6.81001e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.92001e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 5.74999e-06
        [merge_forward]: 4.62e-06
        [cell_reuse_recompute_pass]: 1.31002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.779e-05
        [j_node_and_user_rematch]: 1.005e-05
        [meta_fg_expand]: 2.60997e-06
        [replace_old_param]: 1.102e-05
        [inline_without_move]: 7.01999e-06
        [renormalize]: 0.00085106
        [add_forward_monad_depend]: 1.012e-05
        [auto_monad_grad]: 3.02002e-06
        [auto_monad_eliminator]: 2.102e-05
        [cse]: 3.486e-05
        [replace_applicator]: 2.024e-05
    [Cycle 2]: 0.00047385, [27]
        [switch_simplify]: 6.13002e-06
        [loop_unroll]: 5.10999e-06
        [a_1]: 9.854e-05
        [with_stream_mark]: 1.575e-05
        [recompute_prepare]: 5.46e-06
        [updatestate_depend_eliminate]: 3.10002e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [parameter_eliminate]: 2.49999e-06
        [specialize_transform]: 6.69001e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.94e-06
        [meta_shard_fg_expand]: 3.08e-06
        [get_grad_eliminate_]: 5.84999e-06
        [merge_forward]: 4.73001e-06
        [cell_reuse_recompute_pass]: 2.82002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.976e-05
        [j_node_and_user_rematch]: 1.108e-05
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 8.84998e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 1.09983e-07
        [add_forward_monad_depend]: 3.32002e-06
        [auto_monad_grad]: 1.30001e-06
        [auto_monad_eliminator]: 6.672e-05
        [cse]: 1.975e-05
        [replace_applicator]: 8.70999e-06
[py_interpret_to_execute_after_opt_a]: 1.917e-05
[rewriter_after_opt_a]: 4.531e-05
[convert_after_rewriter]: 7.98001e-06
[order_py_execute_after_rewriter]: 5.73002e-06
[mutable_eliminate]: 0.00077374
[jit_opt_b]: 6.024e-05, [1]
    [Cycle 1]: 4.943e-05, [2]
        [frontend_op_eliminate]: 1.705e-05
        [inline_after_opt_a]: 1.827e-05
[cconv]: 3.957e-05
[loop_unroll]: 0.00054833
[jit_opt_after_cconv]: 0.00018331, [1]
    [Cycle 1]: 0.00017476, [11]
        [c_1]: 2.243e-05
        [parameter_eliminate]: 5.59e-06
        [updatestate_depend_eliminate]: 1.142e-05
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.76e-06
        [cse]: 3.374e-05
        [call_graph_tuple_transform]: 2.452e-05
        [tuple_list_get_item_eliminator]: 5.82999e-06
        [none_parameter_eliminate]: 1.85001e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 5.84999e-06
[remove_dup_value]: 1.593e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 6.02999e-06
[add_recomputation]: 6.071e-05
[cse_after_recomputation]: 2.258e-05, [1]
    [Cycle 1]: 1.671e-05, [1]
        [cse]: 9.44998e-06
[auto_monad_reorder]: 1.869e-05
[get_jit_bprop_graph]: 2.31998e-06
[rewriter_after_jit_bprop_graph]: 9.77999e-06
[opt_after_jit_grad]: 0.00061198
[symbol_engine_optimizer]: 8.431e-05, [1]
    [Cycle 1]: 7.57e-05, [6]
        [build]: 5.82999e-06
        [elim_shapecalc]: 8.13999e-06
        [elim_not_effective]: 1.447e-05
        [opt_reshape]: 5.84e-06
        [fold_const_symbol]: 9.41e-06
        [renormalize]: 8.2e-07
[validate]: 4.909e-05
[backend_pass]: 1.05001e-06
[task_emit]: 0.0285253
[execute]: 1.184e-05
Sums
    bootstrap                            :     0.000367s :  0.90%
    type_inference                       :     0.006863s : 16.93%
    event_method                         :     0.000023s :  0.06%
    auto_monad                           :     0.000078s :  0.19%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.10%
    rewriter_before_opt_a                :     0.000081s :  0.20%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000073s :  0.18%
    jit_opt_a.loop_unroll                :     0.000035s :  0.09%
    jit_opt_a.a_1                        :     0.000723s :  1.78%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000014s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.03%
    jit_opt_a.merge_forward              :     0.000009s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000020s :  0.05%
    jit_opt_a.inline_without_move        :     0.000012s :  0.03%
    jit_opt_a.renormalize                :     0.000851s :  2.10%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000088s :  0.22%
    jit_opt_a.cse                        :     0.000055s :  0.13%
    jit_opt_a.replace_applicator         :     0.000029s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.05%
    rewriter_after_opt_a                 :     0.000045s :  0.11%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.01%
    mutable_eliminate                    :     0.000774s :  1.91%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.05%
    cconv                                :     0.000040s :  0.10%
    loop_unroll                          :     0.000548s :  1.35%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000034s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.01%
    remove_dup_value                     :     0.000016s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.01%
    add_recomputation                    :     0.000061s :  0.15%
    cse_after_recomputation.cse          :     0.000009s :  0.02%
    auto_monad_reorder                   :     0.000019s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.02%
    opt_after_jit_grad                   :     0.000612s :  1.51%
    symbol_engine_optimizer.build        :     0.000006s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000049s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.028525s : 70.36%
    execute                              :     0.000012s :  0.03%


Time group info:
------[substitution.]   0.000279    23
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000002s :      2: substitution.fold_const_symbol
 2.12% :     0.000006s :      3: substitution.graph_param_transform
84.99% :     0.000237s :      4: substitution.inline
 1.93% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.19% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.35% :     0.000007s :      2: substitution.replace_old_param
 5.10% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006764     2
85.12% :     0.005758s :      1: type_inference.infer
14.88% :     0.001007s :      1: type_inference.specialize
------[replace.]   0.000076     6
66.12% :     0.000050s :      4: replace.inline
33.88% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000246     6
94.88% :     0.000234s :      4: match.inline
 5.12% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   691
 1.47% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.85% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 1.83% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000002s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 0.94% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.04% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.00% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.13% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000001s :      3: predicate.elim_not_effective
 0.72% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.98% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.04% :     0.000001s :     10: predicate.environ_get_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.13% :     0.000002s :      6: predicate.get_grad_eliminate
 0.53% :     0.000001s :      3: predicate.graph_param_transform
 6.70% :     0.000009s :     22: predicate.inline
 1.14% :     0.000002s :      6: predicate.inline_without_move
 0.68% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.85% :     0.000002s :      6: predicate.less_batch_normalization
 1.46% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.97% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.48% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.93% :     0.000001s :     10: predicate.merge_addn
 0.89% :     0.000001s :     10: predicate.minmaximum_grad
 2.74% :     0.000004s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     15: predicate.partial_eliminate
 0.97% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 1.25% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.79% :     0.000001s :      6: predicate.replace_old_param
 0.48% :     0.000001s :      3: predicate.reset_defer_inline
 1.72% :     0.000002s :     10: predicate.reshape_eliminate
 1.06% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.08% :     0.000001s :     10: predicate.same_eliminate
 0.42% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.22% :     0.000002s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.09% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.69% :     0.000004s :     16: predicate.switch_defer_inline
 1.88% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.75% :     0.000012s :     53: predicate.switch_simplify
 1.04% :     0.000001s :     10: predicate.tile_eliminate
 1.01% :     0.000001s :     10: predicate.transpose_eliminate
 1.17% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.29% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.28% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.44% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.58% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.37% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000756    12
50.72% :     0.000383s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.70% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.57% :     0.000314s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.047373    76
 0.14% :     0.000065s :      1: add_recomputation
 0.17% :     0.000083s :      1: auto_monad
 0.05% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.83% :     0.000394s :      1: bootstrap
 0.09% :     0.000043s :      1: cconv
 0.02% :     0.000011s :      1: convert_after_rewriter
 0.05% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.06% :     0.000029s :      1: event_method
 0.04% :     0.000019s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
13.84% :     0.006556s :      1: jit_opt_a
 0.39% :     0.000187s :      1: jit_opt_after_cconv
 0.13% :     0.000064s :      1: jit_opt_b
 1.19% :     0.000562s :      1: loop_unroll
 1.67% :     0.000789s :      1: mutable_eliminate
 2.06% :     0.000977s :     26: opt.transform.jit_opt_a
 0.12% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000028s :      4: opt.transform.jit_opt_b
 0.04% :     0.000020s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000024s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.32% :     0.000627s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.09% :     0.000044s :      1: py_interpret_to_execute
 0.05% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000019s :      1: remove_dup_value
 1.01% :     0.000479s :      1: renormalize.infer
 0.76% :     0.000359s :      1: renormalize.specialize
 0.03% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000051s :      1: rewriter_after_opt_a
 0.18% :     0.000085s :      1: rewriter_before_opt_a
 0.18% :     0.000088s :      1: symbol_engine_optimizer
60.27% :     0.028549s :      1: task_emit
14.55% :     0.006891s :      1: type_inference
 0.16% :     0.000078s :      1: validate


TotalTime = 0.0145239, [30]
[bootstrap]: 0.00040586
[type_inference]: 0.00468681
[event_method]: 1.704e-05
[auto_monad]: 5.845e-05
[graph_reusing]: 5.74e-06
[pre_auto_parallel]: 2.54999e-06
[py_interpret_to_execute]: 3.679e-05
[rewriter_before_opt_a]: 6.896e-05
[expand_dump_flag]: 2.73e-06
[jit_opt_a]: 0.00581305, [2]
    [Cycle 1]: 0.00178002, [27]
        [switch_simplify]: 5.11e-05
        [loop_unroll]: 2.802e-05
        [a_1]: 0.00054215
        [with_stream_mark]: 1.916e-05
        [recompute_prepare]: 7.28e-06
        [updatestate_depend_eliminate]: 2.69001e-06
        [updatestate_assign_eliminate]: 2.57001e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [parameter_eliminate]: 1.35001e-06
        [specialize_transform]: 6.43e-06
        [updatestate_useless_node_eliminater]: 5.39998e-06
        [accelerated_algorithm]: 5.57999e-06
        [meta_shard_fg_expand]: 1.56002e-06
        [get_grad_eliminate_]: 5.30999e-06
        [merge_forward]: 3.78001e-06
        [cell_reuse_recompute_pass]: 1.16002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.29e-05
        [j_node_and_user_rematch]: 9.50001e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 8.23999e-06
        [inline_without_move]: 5.40999e-06
        [renormalize]: 0.00082112
        [add_forward_monad_depend]: 5.34e-06
        [auto_monad_grad]: 2.02001e-06
        [auto_monad_eliminator]: 1.329e-05
        [cse]: 2.39e-05
        [replace_applicator]: 1.315e-05
    [Cycle 2]: 0.00032432, [27]
        [switch_simplify]: 5.40001e-06
        [loop_unroll]: 4.79e-06
        [a_1]: 9.044e-05
        [with_stream_mark]: 1.02e-05
        [recompute_prepare]: 4.85001e-06
        [updatestate_depend_eliminate]: 2.89001e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 4.90001e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 5.01997e-06
        [meta_shard_fg_expand]: 1.67999e-06
        [get_grad_eliminate_]: 4.48001e-06
        [merge_forward]: 3.08998e-06
        [cell_reuse_recompute_pass]: 1.67999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.344e-05
        [j_node_and_user_rematch]: 8.07998e-06
        [meta_fg_expand]: 1.70001e-06
        [replace_old_param]: 7.54002e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.47999e-06
        [auto_monad_grad]: 1.29998e-06
        [auto_monad_eliminator]: 5.16002e-06
        [cse]: 1.019e-05
        [replace_applicator]: 5.73002e-06
[py_interpret_to_execute_after_opt_a]: 1.123e-05
[rewriter_after_opt_a]: 3.4e-05
[convert_after_rewriter]: 7.51001e-06
[order_py_execute_after_rewriter]: 5.11002e-06
[mutable_eliminate]: 0.00067514
[jit_opt_b]: 5.238e-05, [1]
    [Cycle 1]: 4.501e-05, [2]
        [frontend_op_eliminate]: 1.741e-05
        [inline_after_opt_a]: 1.514e-05
[cconv]: 2.606e-05
[loop_unroll]: 0.00138888
[jit_opt_after_cconv]: 0.00016605, [1]
    [Cycle 1]: 0.00015801, [11]
        [c_1]: 2.339e-05
        [parameter_eliminate]: 5.23002e-06
        [updatestate_depend_eliminate]: 8.12003e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 2.79001e-06
        [cse]: 2.397e-05
        [call_graph_tuple_transform]: 2.359e-05
        [tuple_list_get_item_eliminator]: 6.04001e-06
        [none_parameter_eliminate]: 1.48002e-06
        [renormalize]: 7.30011e-07
        [switch_simplify]: 5.76e-06
[remove_dup_value]: 1.426e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 6.09001e-06
[add_recomputation]: 4.949e-05
[cse_after_recomputation]: 2.157e-05, [1]
    [Cycle 1]: 1.493e-05, [1]
        [cse]: 8.79e-06
[auto_monad_reorder]: 1.465e-05
[get_jit_bprop_graph]: 2.07999e-06
[rewriter_after_jit_bprop_graph]: 9.89001e-06
[opt_after_jit_grad]: 0.00062899
[symbol_engine_optimizer]: 8.306e-05, [1]
    [Cycle 1]: 7.459e-05, [6]
        [build]: 3.97e-06
        [elim_shapecalc]: 8.15999e-06
        [elim_not_effective]: 1.48e-05
        [opt_reshape]: 6.29999e-06
        [fold_const_symbol]: 1.039e-05
        [renormalize]: 7.2e-07
[validate]: 4.097e-05
Sums
    bootstrap                            :     0.000406s :  3.98%
    type_inference                       :     0.004687s : 45.94%
    event_method                         :     0.000017s :  0.17%
    auto_monad                           :     0.000058s :  0.57%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.36%
    rewriter_before_opt_a                :     0.000069s :  0.68%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000056s :  0.55%
    jit_opt_a.loop_unroll                :     0.000033s :  0.32%
    jit_opt_a.a_1                        :     0.000633s :  6.20%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000821s :  8.05%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.18%
    jit_opt_a.cse                        :     0.000034s :  0.33%
    jit_opt_a.replace_applicator         :     0.000019s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.11%
    rewriter_after_opt_a                 :     0.000034s :  0.33%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000675s :  6.62%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.15%
    cconv                                :     0.000026s :  0.26%
    loop_unroll                          :     0.001389s : 13.61%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000024s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000014s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000049s :  0.49%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.10%
    opt_after_jit_grad                   :     0.000629s :  6.16%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.40%


Time group info:
------[substitution.]   0.000213    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.65% :     0.000006s :      3: substitution.graph_param_transform
84.90% :     0.000181s :      4: substitution.inline
 1.46% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.10% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000004s :      2: substitution.replace_old_param
 5.33% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004621     2
80.73% :     0.003731s :      1: type_inference.infer
19.27% :     0.000890s :      1: type_inference.specialize
------[replace.]   0.000074     6
67.36% :     0.000050s :      4: replace.inline
32.64% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000189     6
94.54% :     0.000178s :      4: match.inline
 5.46% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.54% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.39% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000001s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.33% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.46% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.71% :     0.000002s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 6.17% :     0.000007s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.18% :     0.000001s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 2.17% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.84% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.43% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.78% :     0.000002s :      3: predicate.mutable_eliminate
 0.68% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000002s :     15: predicate.partial_eliminate
 1.52% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.32% :     0.000002s :     10: predicate.reduce_eliminate
 1.56% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.48% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.10% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.04% :     0.000008s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.23% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.11% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.84% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000672    12
49.27% :     0.000331s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.44% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.29% :     0.000297s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016251    72
 0.33% :     0.000053s :      1: add_recomputation
 0.38% :     0.000062s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 2.60% :     0.000422s :      1: bootstrap
 0.18% :     0.000029s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000024s :      1: cse_after_recomputation
 0.06% :     0.000010s :      1: environ_conv
 0.13% :     0.000021s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
35.80% :     0.005817s :      1: jit_opt_a
 1.05% :     0.000170s :      1: jit_opt_after_cconv
 0.34% :     0.000055s :      1: jit_opt_b
 8.63% :     0.001402s :      1: loop_unroll
 4.21% :     0.000685s :      1: mutable_eliminate
 5.16% :     0.000838s :     26: opt.transform.jit_opt_a
 0.34% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000026s :      4: opt.transform.jit_opt_b
 0.12% :     0.000020s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.94% :     0.000640s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000039s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.70% :     0.000439s :      1: renormalize.infer
 2.30% :     0.000374s :      1: renormalize.specialize
 0.07% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000037s :      1: rewriter_after_opt_a
 0.44% :     0.000072s :      1: rewriter_before_opt_a
 0.53% :     0.000086s :      1: symbol_engine_optimizer
28.91% :     0.004698s :      1: type_inference


TotalTime = 0.0202785, [33]
[bootstrap]: 0.00032734
[type_inference]: 0.00520045
[event_method]: 1.851e-05
[auto_monad]: 6.896e-05
[graph_reusing]: 5.51002e-06
[pre_auto_parallel]: 3.2e-06
[py_interpret_to_execute]: 8.045e-05
[rewriter_before_opt_a]: 7.54e-05
[expand_dump_flag]: 3.66001e-06
[jit_opt_a]: 0.00570696, [2]
    [Cycle 1]: 0.00167055, [27]
        [switch_simplify]: 5.718e-05
        [loop_unroll]: 2.924e-05
        [a_1]: 0.00055021
        [with_stream_mark]: 1.643e-05
        [recompute_prepare]: 7.38e-06
        [updatestate_depend_eliminate]: 3.91001e-06
        [updatestate_assign_eliminate]: 2.74999e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 5.82999e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 2.49999e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 3.55998e-06
        [cell_reuse_recompute_pass]: 1.10999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.458e-05
        [j_node_and_user_rematch]: 9.52999e-06
        [meta_fg_expand]: 2.01003e-06
        [replace_old_param]: 9.56998e-06
        [inline_without_move]: 6.02999e-06
        [renormalize]: 0.00069187
        [add_forward_monad_depend]: 6.40002e-06
        [auto_monad_grad]: 1.87001e-06
        [auto_monad_eliminator]: 1.43e-05
        [cse]: 2.938e-05
        [replace_applicator]: 1.261e-05
    [Cycle 2]: 0.00033487, [27]
        [switch_simplify]: 5.71e-06
        [loop_unroll]: 5.29e-06
        [a_1]: 8.969e-05
        [with_stream_mark]: 1.093e-05
        [recompute_prepare]: 5.14998e-06
        [updatestate_depend_eliminate]: 2.58e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.91998e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 5.27999e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.00001e-06
        [meta_shard_fg_expand]: 1.30001e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 2.54999e-06
        [cell_reuse_recompute_pass]: 3.66001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.436e-05
        [j_node_and_user_rematch]: 9.43002e-06
        [meta_fg_expand]: 1.47999e-06
        [replace_old_param]: 8.04997e-06
        [inline_without_move]: 5.19998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 6.79982e-07
        [auto_monad_eliminator]: 5.19e-06
        [cse]: 1.196e-05
        [replace_applicator]: 5.25001e-06
[py_interpret_to_execute_after_opt_a]: 1.023e-05
[rewriter_after_opt_a]: 3.547e-05
[convert_after_rewriter]: 7.93001e-06
[order_py_execute_after_rewriter]: 5.51998e-06
[mutable_eliminate]: 0.00062178
[jit_opt_b]: 5.148e-05, [1]
    [Cycle 1]: 4.413e-05, [2]
        [frontend_op_eliminate]: 1.796e-05
        [inline_after_opt_a]: 1.472e-05
[cconv]: 2.714e-05
[loop_unroll]: 0.00043534
[jit_opt_after_cconv]: 0.00014761, [1]
    [Cycle 1]: 0.00014108, [11]
        [c_1]: 2.147e-05
        [parameter_eliminate]: 3.46999e-06
        [updatestate_depend_eliminate]: 7.11999e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [cse]: 2.174e-05
        [call_graph_tuple_transform]: 1.928e-05
        [tuple_list_get_item_eliminator]: 5.15999e-06
        [none_parameter_eliminate]: 1.43002e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 5.27999e-06
[remove_dup_value]: 1.388e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 4.99003e-06
[add_recomputation]: 5.082e-05
[cse_after_recomputation]: 1.961e-05, [1]
    [Cycle 1]: 1.427e-05, [1]
        [cse]: 9.09e-06
[auto_monad_reorder]: 1.476e-05
[get_jit_bprop_graph]: 2.06e-06
[rewriter_after_jit_bprop_graph]: 4.78001e-06
[opt_after_jit_grad]: 0.00045333
[symbol_engine_optimizer]: 7.325e-05, [1]
    [Cycle 1]: 6.688e-05, [6]
        [build]: 4.18999e-06
        [elim_shapecalc]: 7.53e-06
        [elim_not_effective]: 1.247e-05
        [opt_reshape]: 5.40999e-06
        [fold_const_symbol]: 9.02e-06
        [renormalize]: 6.19999e-07
[validate]: 3.785e-05
[backend_pass]: 4.872e-05
[task_emit]: 0.00650164
[execute]: 7.53e-06
Sums
    bootstrap                            :     0.000327s :  2.05%
    type_inference                       :     0.005200s : 32.60%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000069s :  0.43%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000080s :  0.50%
    rewriter_before_opt_a                :     0.000075s :  0.47%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000640s :  4.01%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000692s :  4.34%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000041s :  0.26%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000035s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000622s :  3.90%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000027s :  0.17%
    loop_unroll                          :     0.000435s :  2.73%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000051s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000453s :  2.84%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.24%
    backend_pass                         :     0.000049s :  0.31%
    task_emit                            :     0.006502s : 40.75%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000205    23
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.86% :     0.000006s :      3: substitution.graph_param_transform
83.00% :     0.000170s :      4: substitution.inline
 1.94% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.44% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.04% :     0.000004s :      2: substitution.replace_old_param
 6.12% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005123     2
78.75% :     0.004035s :      1: type_inference.infer
21.25% :     0.001089s :      1: type_inference.specialize
------[replace.]   0.000069     6
66.58% :     0.000046s :      4: replace.inline
33.42% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000180     6
93.62% :     0.000168s :      4: match.inline
 6.38% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.07% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 2.05% :     0.000002s :     10: predicate.addn_zero_filter
 2.27% :     0.000003s :     10: predicate.arithmetic_simplify
 1.36% :     0.000002s :     10: predicate.cast_eliminate
 0.58% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.24% :     0.000001s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.57% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.36% :     0.000002s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.85% :     0.000007s :     22: predicate.inline
 1.07% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.87% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.70% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.82% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.25% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.31% :     0.000002s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 2.25% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.41% :     0.000009s :     53: predicate.switch_simplify
 1.49% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.92% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.94% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000656    12
50.89% :     0.000334s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.36% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.75% :     0.000293s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021932    76
 0.25% :     0.000054s :      1: add_recomputation
 0.33% :     0.000072s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.24% :     0.000052s :      1: backend_pass
 1.58% :     0.000346s :      1: bootstrap
 0.14% :     0.000030s :      1: cconv
 0.06% :     0.000013s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.04% :     0.005711s :      1: jit_opt_a
 0.68% :     0.000150s :      1: jit_opt_after_cconv
 0.25% :     0.000054s :      1: jit_opt_b
 2.02% :     0.000443s :      1: loop_unroll
 2.88% :     0.000632s :      1: mutable_eliminate
 3.90% :     0.000856s :     26: opt.transform.jit_opt_a
 0.22% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.10% :     0.000461s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.38% :     0.000083s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.68% :     0.000367s :      1: renormalize.infer
 1.44% :     0.000317s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000039s :      1: rewriter_after_opt_a
 0.36% :     0.000078s :      1: rewriter_before_opt_a
 0.35% :     0.000076s :      1: symbol_engine_optimizer
29.70% :     0.006514s :      1: task_emit
23.79% :     0.005217s :      1: type_inference
 0.26% :     0.000058s :      1: validate


TotalTime = 0.0412805, [33]
[bootstrap]: 0.00033546
[type_inference]: 0.00524236
[event_method]: 1.927e-05
[auto_monad]: 6.969e-05
[graph_reusing]: 6.06e-06
[pre_auto_parallel]: 2.33998e-06
[py_interpret_to_execute]: 4.088e-05
[rewriter_before_opt_a]: 7.902e-05
[expand_dump_flag]: 2.85002e-06
[jit_opt_a]: 0.00589935, [2]
    [Cycle 1]: 0.00201027, [27]
        [switch_simplify]: 6.423e-05
        [loop_unroll]: 3.063e-05
        [a_1]: 0.00059126
        [with_stream_mark]: 2.222e-05
        [recompute_prepare]: 8.38999e-06
        [updatestate_depend_eliminate]: 4.28999e-06
        [updatestate_assign_eliminate]: 3.18e-06
        [updatestate_loads_eliminate]: 2.78998e-06
        [parameter_eliminate]: 1.80001e-06
        [specialize_transform]: 6.14001e-06
        [updatestate_useless_node_eliminater]: 5.35999e-06
        [accelerated_algorithm]: 5.64e-06
        [meta_shard_fg_expand]: 2.31998e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 3.82002e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.541e-05
        [j_node_and_user_rematch]: 9.82999e-06
        [meta_fg_expand]: 2.49999e-06
        [replace_old_param]: 1e-05
        [inline_without_move]: 5.68002e-06
        [renormalize]: 0.00096305
        [add_forward_monad_depend]: 5.87001e-06
        [auto_monad_grad]: 2.16e-06
        [auto_monad_eliminator]: 1.45e-05
        [cse]: 2.948e-05
        [replace_applicator]: 1.258e-05
    [Cycle 2]: 0.00034189, [27]
        [switch_simplify]: 5.80002e-06
        [loop_unroll]: 5.35999e-06
        [a_1]: 9.468e-05
        [with_stream_mark]: 9.17001e-06
        [recompute_prepare]: 5.56e-06
        [updatestate_depend_eliminate]: 3.43e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [parameter_eliminate]: 1.07998e-06
        [specialize_transform]: 5.51002e-06
        [updatestate_useless_node_eliminater]: 6.22001e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 1.31002e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 2.81999e-06
        [cell_reuse_recompute_pass]: 1.45001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.342e-05
        [j_node_and_user_rematch]: 8.18999e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 8.08999e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.28002e-06
        [auto_monad_grad]: 1.22e-06
        [auto_monad_eliminator]: 5.58002e-06
        [cse]: 1.171e-05
        [replace_applicator]: 6.11e-06
[py_interpret_to_execute_after_opt_a]: 1.212e-05
[rewriter_after_opt_a]: 3.692e-05
[convert_after_rewriter]: 7.03e-06
[order_py_execute_after_rewriter]: 4.85999e-06
[mutable_eliminate]: 0.00069786
[jit_opt_b]: 5.437e-05, [1]
    [Cycle 1]: 4.649e-05, [2]
        [frontend_op_eliminate]: 1.81e-05
        [inline_after_opt_a]: 1.615e-05
[cconv]: 3.018e-05
[loop_unroll]: 0.00045725
[jit_opt_after_cconv]: 0.00017509, [1]
    [Cycle 1]: 0.00016767, [11]
        [c_1]: 2.543e-05
        [parameter_eliminate]: 4.86002e-06
        [updatestate_depend_eliminate]: 8.47e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.86e-06
        [cse]: 2.544e-05
        [call_graph_tuple_transform]: 2.495e-05
        [tuple_list_get_item_eliminator]: 6.30002e-06
        [none_parameter_eliminate]: 2.12001e-06
        [renormalize]: 7.49977e-07
        [switch_simplify]: 6.73e-06
[remove_dup_value]: 1.435e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 5.45001e-06
[add_recomputation]: 5.426e-05
[cse_after_recomputation]: 2.1e-05, [1]
    [Cycle 1]: 1.537e-05, [1]
        [cse]: 9.74e-06
[auto_monad_reorder]: 1.46e-05
[get_jit_bprop_graph]: 1.97999e-06
[rewriter_after_jit_bprop_graph]: 6.54999e-06
[opt_after_jit_grad]: 0.00050363
[symbol_engine_optimizer]: 8.309e-05, [1]
    [Cycle 1]: 7.596e-05, [6]
        [build]: 5.52001e-06
        [elim_shapecalc]: 8.45001e-06
        [elim_not_effective]: 1.631e-05
        [opt_reshape]: 6.14999e-06
        [fold_const_symbol]: 9.41003e-06
        [renormalize]: 5.39992e-07
[validate]: 4.217e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.0271179
[execute]: 9.12999e-06
Sums
    bootstrap                            :     0.000335s :  0.90%
    type_inference                       :     0.005242s : 14.14%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000070s :  0.19%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.11%
    rewriter_before_opt_a                :     0.000079s :  0.21%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000070s :  0.19%
    jit_opt_a.loop_unroll                :     0.000036s :  0.10%
    jit_opt_a.a_1                        :     0.000686s :  1.85%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000018s :  0.05%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.000963s :  2.60%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.05%
    jit_opt_a.cse                        :     0.000041s :  0.11%
    jit_opt_a.replace_applicator         :     0.000019s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000037s :  0.10%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000698s :  1.88%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.04%
    cconv                                :     0.000030s :  0.08%
    loop_unroll                          :     0.000457s :  1.23%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000014s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.01%
    add_recomputation                    :     0.000054s :  0.15%
    cse_after_recomputation.cse          :     0.000010s :  0.03%
    auto_monad_reorder                   :     0.000015s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000504s :  1.36%
    symbol_engine_optimizer.build        :     0.000006s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.027118s : 73.16%
    execute                              :     0.000009s :  0.02%


Time group info:
------[substitution.]   0.000245    23
 1.32% :     0.000003s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.61% :     0.000006s :      3: substitution.graph_param_transform
85.14% :     0.000209s :      4: substitution.inline
 1.42% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.88% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.04% :     0.000005s :      2: substitution.replace_old_param
 5.09% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005174     2
81.45% :     0.004214s :      1: type_inference.infer
18.55% :     0.000960s :      1: type_inference.specialize
------[replace.]   0.000077     6
66.96% :     0.000052s :      4: replace.inline
33.04% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000218     6
94.77% :     0.000206s :      4: match.inline
 5.23% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.12% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.53% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 2.10% :     0.000003s :     10: predicate.arithmetic_simplify
 1.46% :     0.000002s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.42% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      3: predicate.elim_not_effective
 0.98% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.55% :     0.000002s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.17% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 6.04% :     0.000008s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.70% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.65% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 2.94% :     0.000004s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.35% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.11% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.71% :     0.000003s :     16: predicate.switch_defer_inline
 1.99% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.14% :     0.000010s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.06% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.49% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000767    12
53.08% :     0.000407s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.15% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.77% :     0.000313s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.043278    76
 0.13% :     0.000058s :      1: add_recomputation
 0.17% :     0.000073s :      1: auto_monad
 0.04% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.82% :     0.000353s :      1: bootstrap
 0.08% :     0.000033s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.05% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.05% :     0.000023s :      1: event_method
 0.03% :     0.000014s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
13.64% :     0.005903s :      1: jit_opt_a
 0.41% :     0.000179s :      1: jit_opt_after_cconv
 0.13% :     0.000058s :      1: jit_opt_b
 1.08% :     0.000468s :      1: loop_unroll
 1.63% :     0.000707s :      1: mutable_eliminate
 2.12% :     0.000915s :     26: opt.transform.jit_opt_a
 0.14% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000027s :      4: opt.transform.jit_opt_b
 0.03% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.18% :     0.000513s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.10% :     0.000044s :      1: py_interpret_to_execute
 0.03% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000017s :      1: remove_dup_value
 1.12% :     0.000483s :      1: renormalize.infer
 1.09% :     0.000471s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000040s :      1: rewriter_after_opt_a
 0.19% :     0.000082s :      1: rewriter_before_opt_a
 0.20% :     0.000086s :      1: symbol_engine_optimizer
62.71% :     0.027139s :      1: task_emit
12.14% :     0.005256s :      1: type_inference
 0.15% :     0.000066s :      1: validate


TotalTime = 0.044691, [33]
[bootstrap]: 0.00034336
[type_inference]: 0.00569997
[event_method]: 2.306e-05
[auto_monad]: 7.181e-05
[graph_reusing]: 6.78e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 3.855e-05
[rewriter_before_opt_a]: 8.193e-05
[expand_dump_flag]: 3.76001e-06
[jit_opt_a]: 0.0270515, [2]
    [Cycle 1]: 0.022509, [27]
        [switch_simplify]: 6.839e-05
        [loop_unroll]: 3.079e-05
        [a_1]: 0.0006349
        [with_stream_mark]: 7.222e-05
        [recompute_prepare]: 4.346e-05
        [updatestate_depend_eliminate]: 8.15e-06
        [updatestate_assign_eliminate]: 3.21999e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 3.34001e-06
        [specialize_transform]: 8.67e-06
        [updatestate_useless_node_eliminater]: 7.13e-06
        [accelerated_algorithm]: 8.17e-06
        [meta_shard_fg_expand]: 3.89002e-06
        [get_grad_eliminate_]: 6.79999e-06
        [merge_forward]: 7.1e-06
        [cell_reuse_recompute_pass]: 4.58999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.944e-05
        [j_node_and_user_rematch]: 1.136e-05
        [meta_fg_expand]: 2.96001e-06
        [replace_old_param]: 1.323e-05
        [inline_without_move]: 5.44e-06
        [renormalize]: 0.00105125
        [add_forward_monad_depend]: 9.79e-06
        [auto_monad_grad]: 3.33e-06
        [auto_monad_eliminator]: 2.695e-05
        [cse]: 4.077e-05
        [replace_applicator]: 2.592e-05
    [Cycle 2]: 0.00043663, [27]
        [switch_simplify]: 7.48e-06
        [loop_unroll]: 6.15002e-06
        [a_1]: 0.00010647
        [with_stream_mark]: 1.96e-05
        [recompute_prepare]: 6.98998e-06
        [updatestate_depend_eliminate]: 4.46002e-06
        [updatestate_assign_eliminate]: 2.58998e-06
        [updatestate_loads_eliminate]: 3.31999e-06
        [parameter_eliminate]: 2.16998e-06
        [specialize_transform]: 6.41e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 5.87001e-06
        [meta_shard_fg_expand]: 1.77001e-06
        [get_grad_eliminate_]: 4.77998e-06
        [merge_forward]: 5.05999e-06
        [cell_reuse_recompute_pass]: 3.65e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.222e-05
        [j_node_and_user_rematch]: 9.41998e-06
        [meta_fg_expand]: 2.68e-06
        [replace_old_param]: 1.077e-05
        [inline_without_move]: 5.67001e-06
        [renormalize]: 1.10012e-07
        [add_forward_monad_depend]: 2.74999e-06
        [auto_monad_grad]: 1.29e-06
        [auto_monad_eliminator]: 1.089e-05
        [cse]: 1.808e-05
        [replace_applicator]: 7.68001e-06
[py_interpret_to_execute_after_opt_a]: 1.826e-05
[rewriter_after_opt_a]: 5.268e-05
[convert_after_rewriter]: 8.79e-06
[order_py_execute_after_rewriter]: 6.14001e-06
[mutable_eliminate]: 0.00094009
[jit_opt_b]: 6.154e-05, [1]
    [Cycle 1]: 4.962e-05, [2]
        [frontend_op_eliminate]: 1.862e-05
        [inline_after_opt_a]: 1.714e-05
[cconv]: 3.811e-05
[loop_unroll]: 0.000579
[jit_opt_after_cconv]: 0.00024101, [1]
    [Cycle 1]: 0.00023072, [11]
        [c_1]: 2.164e-05
        [parameter_eliminate]: 5.62999e-06
        [updatestate_depend_eliminate]: 1.226e-05
        [updatestate_assign_eliminate]: 4.943e-05
        [updatestate_loads_eliminate]: 2.91999e-06
        [cse]: 3.573e-05
        [call_graph_tuple_transform]: 2.615e-05
        [tuple_list_get_item_eliminator]: 5.40999e-06
        [none_parameter_eliminate]: 2.21e-06
        [renormalize]: 9.89996e-07
        [switch_simplify]: 5.68002e-06
[remove_dup_value]: 1.765e-05
[partial_unused_args_eliminate]: 2.46e-06
[environ_conv]: 6.64001e-06
[add_recomputation]: 6.506e-05
[cse_after_recomputation]: 3.03e-05, [1]
    [Cycle 1]: 2.145e-05, [1]
        [cse]: 1.034e-05
[auto_monad_reorder]: 1.93e-05
[get_jit_bprop_graph]: 2.59999e-06
[rewriter_after_jit_bprop_graph]: 9.61998e-06
[opt_after_jit_grad]: 0.00065542
[symbol_engine_optimizer]: 9.214e-05, [1]
    [Cycle 1]: 8.291e-05, [6]
        [build]: 6.61e-06
        [elim_shapecalc]: 9.51e-06
        [elim_not_effective]: 1.744e-05
        [opt_reshape]: 6.49001e-06
        [fold_const_symbol]: 9.24998e-06
        [renormalize]: 1.52999e-06
[validate]: 4.886e-05
[backend_pass]: 1.20999e-06
[task_emit]: 0.00815436
[execute]: 1.089e-05
Sums
    bootstrap                            :     0.000343s :  1.75%
    type_inference                       :     0.005700s : 29.09%
    event_method                         :     0.000023s :  0.12%
    auto_monad                           :     0.000072s :  0.37%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.20%
    rewriter_before_opt_a                :     0.000082s :  0.42%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.39%
    jit_opt_a.loop_unroll                :     0.000037s :  0.19%
    jit_opt_a.a_1                        :     0.000741s :  3.78%
    jit_opt_a.with_stream_mark           :     0.000092s :  0.47%
    jit_opt_a.recompute_prepare          :     0.000050s :  0.26%
    jit_opt_a.updatestate_depend_eliminate :     0.000013s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.03%
    jit_opt_a.specialize_transform       :     0.000015s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000014s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.06%
    jit_opt_a.merge_forward              :     0.000012s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000052s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.03%
    jit_opt_a.replace_old_param          :     0.000024s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001051s :  5.37%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000038s :  0.19%
    jit_opt_a.cse                        :     0.000059s :  0.30%
    jit_opt_a.replace_applicator         :     0.000034s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.09%
    rewriter_after_opt_a                 :     0.000053s :  0.27%
    convert_after_rewriter               :     0.000009s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000940s :  4.80%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000038s :  0.19%
    loop_unroll                          :     0.000579s :  2.96%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000049s :  0.25%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000036s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000018s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000065s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000019s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.05%
    opt_after_jit_grad                   :     0.000655s :  3.35%
    symbol_engine_optimizer.build        :     0.000007s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000002s :  0.01%
    validate                             :     0.000049s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008154s : 41.62%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000290    23
 0.67% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000002s :      2: substitution.fold_const_symbol
 2.37% :     0.000007s :      3: substitution.graph_param_transform
83.13% :     0.000241s :      4: substitution.inline
 2.00% :     0.000006s :      4: substitution.j_node_and_user_rematch
 2.54% :     0.000007s :      4: substitution.remove_not_recompute_node
 3.60% :     0.000010s :      2: substitution.replace_old_param
 5.10% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005608     2
83.32% :     0.004673s :      1: type_inference.infer
16.68% :     0.000935s :      1: type_inference.specialize
------[replace.]   0.000080     6
66.53% :     0.000053s :      4: replace.inline
33.47% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000252     6
94.72% :     0.000239s :      4: match.inline
 5.28% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000143   691
 1.32% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.09% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.86% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 2.06% :     0.000003s :     10: predicate.arithmetic_simplify
 1.39% :     0.000002s :     10: predicate.cast_eliminate
 0.33% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.87% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.12% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.98% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.87% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000002s :     10: predicate.environ_get_eliminate
 1.06% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 5.71% :     0.000008s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 2.57% :     0.000004s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.39% :     0.000002s :     15: predicate.load_eliminater
 1.76% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.68% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.83% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.88% :     0.000001s :     10: predicate.merge_addn
 0.84% :     0.000001s :     10: predicate.minmaximum_grad
 2.47% :     0.000004s :      3: predicate.mutable_eliminate
 0.36% :     0.000001s :      3: predicate.opt_reshape
 1.94% :     0.000003s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.33% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.93% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.41% :     0.000003s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.80% :     0.000001s :      3: predicate.reset_defer_inline
 2.10% :     0.000003s :     10: predicate.reshape_eliminate
 1.02% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.19% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     10: predicate.same_eliminate
 1.24% :     0.000002s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 1.95% :     0.000003s :      6: predicate.specialize_transform
 1.19% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.21% :     0.000003s :     16: predicate.switch_defer_inline
 1.67% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.98% :     0.000011s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.09% :     0.000002s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.07% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.08% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.19% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.41% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.47% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.42% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000836    12
52.96% :     0.000443s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.02% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.02% :     0.000335s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.046927    76
 0.15% :     0.000070s :      1: add_recomputation
 0.16% :     0.000077s :      1: auto_monad
 0.05% :     0.000023s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.79% :     0.000370s :      1: bootstrap
 0.09% :     0.000042s :      1: cconv
 0.03% :     0.000013s :      1: convert_after_rewriter
 0.07% :     0.000033s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.06% :     0.000028s :      1: event_method
 0.04% :     0.000019s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
57.66% :     0.027056s :      1: jit_opt_a
 0.52% :     0.000245s :      1: jit_opt_after_cconv
 0.14% :     0.000065s :      1: jit_opt_b
 1.27% :     0.000595s :      1: loop_unroll
 2.05% :     0.000961s :      1: mutable_eliminate
 2.24% :     0.001051s :     26: opt.transform.jit_opt_a
 0.12% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000028s :      4: opt.transform.jit_opt_b
 0.04% :     0.000020s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000026s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.43% :     0.000671s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000042s :      1: py_interpret_to_execute
 0.05% :     0.000023s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000020s :      1: remove_dup_value
 1.22% :     0.000574s :      1: renormalize.infer
 0.98% :     0.000461s :      1: renormalize.specialize
 0.03% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000058s :      1: rewriter_after_opt_a
 0.18% :     0.000086s :      1: rewriter_before_opt_a
 0.21% :     0.000096s :      1: symbol_engine_optimizer
17.43% :     0.008180s :      1: task_emit
12.20% :     0.005724s :      1: type_inference
 0.16% :     0.000077s :      1: validate


TotalTime = 0.0657611, [33]
[bootstrap]: 0.0003701
[type_inference]: 0.00579346
[event_method]: 1.736e-05
[auto_monad]: 6.652e-05
[graph_reusing]: 2.052e-05
[pre_auto_parallel]: 2.48002e-06
[py_interpret_to_execute]: 3.754e-05
[rewriter_before_opt_a]: 0.00098067
[expand_dump_flag]: 4.37e-06
[jit_opt_a]: 0.0253324, [2]
    [Cycle 1]: 0.0107046, [27]
        [switch_simplify]: 0.00019283
        [loop_unroll]: 0.00010689
        [a_1]: 0.00320632
        [with_stream_mark]: 0.00042204
        [recompute_prepare]: 0.00060401
        [updatestate_depend_eliminate]: 1.203e-05
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 2.79999e-06
        [parameter_eliminate]: 4.97e-06
        [specialize_transform]: 0.00017108
        [updatestate_useless_node_eliminater]: 2.921e-05
        [accelerated_algorithm]: 2.995e-05
        [meta_shard_fg_expand]: 7.31999e-06
        [get_grad_eliminate_]: 2.658e-05
        [merge_forward]: 1.56e-05
        [cell_reuse_recompute_pass]: 1.28e-05
        [cell_reuse_handle_not_recompute_node_pass]: 6.448e-05
        [j_node_and_user_rematch]: 4.212e-05
        [meta_fg_expand]: 7.77e-06
        [replace_old_param]: 4.309e-05
        [inline_without_move]: 2.844e-05
        [renormalize]: 0.00482939
        [add_forward_monad_depend]: 1.468e-05
        [auto_monad_grad]: 3.22002e-06
        [auto_monad_eliminator]: 2.742e-05
        [cse]: 3.52e-05
        [replace_applicator]: 3.057e-05
    [Cycle 2]: 0.0003929, [27]
        [switch_simplify]: 6.64001e-06
        [loop_unroll]: 6.01e-06
        [a_1]: 0.00011495
        [with_stream_mark]: 1.737e-05
        [recompute_prepare]: 5.04e-06
        [updatestate_depend_eliminate]: 4.04002e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 3.08e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 5.49e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 6.81001e-06
        [meta_shard_fg_expand]: 1.89999e-06
        [get_grad_eliminate_]: 4.99003e-06
        [merge_forward]: 4.08001e-06
        [cell_reuse_recompute_pass]: 4.44002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.199e-05
        [j_node_and_user_rematch]: 9.54e-06
        [meta_fg_expand]: 2.43e-06
        [replace_old_param]: 1.031e-05
        [inline_without_move]: 5.42001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.58002e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 5.25999e-06
        [cse]: 1.11e-05
        [replace_applicator]: 5.06997e-06
[py_interpret_to_execute_after_opt_a]: 1.941e-05
[rewriter_after_opt_a]: 3.923e-05
[convert_after_rewriter]: 6.74001e-06
[order_py_execute_after_rewriter]: 5.04003e-06
[mutable_eliminate]: 0.00090137
[jit_opt_b]: 0.0191489, [1]
    [Cycle 1]: 5.236e-05, [2]
        [frontend_op_eliminate]: 2.312e-05
        [inline_after_opt_a]: 1.738e-05
[cconv]: 5.924e-05
[loop_unroll]: 0.00094072
[jit_opt_after_cconv]: 0.00020726, [1]
    [Cycle 1]: 0.00019577, [11]
        [c_1]: 3.039e-05
        [parameter_eliminate]: 7.41001e-06
        [updatestate_depend_eliminate]: 1.266e-05
        [updatestate_assign_eliminate]: 2.95002e-06
        [updatestate_loads_eliminate]: 3.03e-06
        [cse]: 4.04e-05
        [call_graph_tuple_transform]: 2.878e-05
        [tuple_list_get_item_eliminator]: 6.23e-06
        [none_parameter_eliminate]: 2.04e-06
        [renormalize]: 8.00006e-07
        [switch_simplify]: 6.71e-06
[remove_dup_value]: 1.643e-05
[partial_unused_args_eliminate]: 2.69001e-06
[environ_conv]: 7.26999e-06
[add_recomputation]: 5.897e-05
[cse_after_recomputation]: 2.329e-05, [1]
    [Cycle 1]: 1.612e-05, [1]
        [cse]: 9.76e-06
[auto_monad_reorder]: 1.586e-05
[get_jit_bprop_graph]: 2.63e-06
[rewriter_after_jit_bprop_graph]: 1.269e-05
[opt_after_jit_grad]: 0.00061503
[symbol_engine_optimizer]: 8.398e-05, [1]
    [Cycle 1]: 7.567e-05, [6]
        [build]: 5.91e-06
        [elim_shapecalc]: 7.92e-06
        [elim_not_effective]: 1.417e-05
        [opt_reshape]: 7.19001e-06
        [fold_const_symbol]: 9.86998e-06
        [renormalize]: 7.2e-07
[validate]: 4.69e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.0106082
[execute]: 9.49999e-06
Sums
    bootstrap                            :     0.000370s :  1.19%
    type_inference                       :     0.005793s : 18.60%
    event_method                         :     0.000017s :  0.06%
    auto_monad                           :     0.000067s :  0.21%
    graph_reusing                        :     0.000021s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000038s :  0.12%
    rewriter_before_opt_a                :     0.000981s :  3.15%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000199s :  0.64%
    jit_opt_a.loop_unroll                :     0.000113s :  0.36%
    jit_opt_a.a_1                        :     0.003321s : 10.66%
    jit_opt_a.with_stream_mark           :     0.000439s :  1.41%
    jit_opt_a.recompute_prepare          :     0.000609s :  1.96%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000007s :  0.02%
    jit_opt_a.specialize_transform       :     0.000177s :  0.57%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000034s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000037s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000009s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000032s :  0.10%
    jit_opt_a.merge_forward              :     0.000020s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000017s :  0.06%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000086s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000052s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000010s :  0.03%
    jit_opt_a.replace_old_param          :     0.000053s :  0.17%
    jit_opt_a.inline_without_move        :     0.000034s :  0.11%
    jit_opt_a.renormalize                :     0.004829s : 15.51%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000033s :  0.10%
    jit_opt_a.cse                        :     0.000046s :  0.15%
    jit_opt_a.replace_applicator         :     0.000036s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.06%
    rewriter_after_opt_a                 :     0.000039s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000901s :  2.89%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.06%
    cconv                                :     0.000059s :  0.19%
    loop_unroll                          :     0.000941s :  3.02%
    jit_opt_after_cconv.c_1              :     0.000030s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000013s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000040s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000029s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000016s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000059s :  0.19%
    cse_after_recomputation.cse          :     0.000010s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000013s :  0.04%
    opt_after_jit_grad                   :     0.000615s :  1.97%
    symbol_engine_optimizer.build        :     0.000006s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000047s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010608s : 34.06%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.001356    23
 0.18% :     0.000002s :      2: substitution.elim_not_effective
 0.10% :     0.000001s :      2: substitution.fold_const_symbol
 0.49% :     0.000007s :      3: substitution.graph_param_transform
92.20% :     0.001250s :      4: substitution.inline
 0.73% :     0.000010s :      4: substitution.j_node_and_user_rematch
 1.14% :     0.000015s :      4: substitution.remove_not_recompute_node
 1.70% :     0.000023s :      2: substitution.replace_old_param
 3.44% :     0.000047s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005719     2
84.16% :     0.004813s :      1: type_inference.infer
15.84% :     0.000906s :      1: type_inference.specialize
------[replace.]   0.000759     6
87.20% :     0.000662s :      4: replace.inline
12.80% :     0.000097s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.001279     6
96.78% :     0.001238s :      4: match.inline
 3.22% :     0.000041s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000959   691
 0.70% :     0.000007s :     10: predicate.accumulaten_eliminater
 0.16% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.16% :     0.000002s :     10: predicate.addn_check_dump
 0.79% :     0.000008s :     10: predicate.addn_zero_filter
 0.65% :     0.000006s :     10: predicate.arithmetic_simplify
 1.22% :     0.000012s :     10: predicate.cast_eliminate
 0.05% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.43% :     0.000014s :     10: predicate.compare_switch_simplify
 0.60% :     0.000006s :     10: predicate.depend_value_elim
 0.58% :     0.000006s :     10: predicate.dict_get_item_const_eliminator
 0.78% :     0.000007s :     10: predicate.dict_get_item_eliminator
 0.21% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.14% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.04% :     0.000000s :      3: predicate.elim_not_effective
 0.09% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.20% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.61% :     0.000006s :     10: predicate.environ_get_add_eliminate
 0.24% :     0.000002s :     10: predicate.environ_get_depend_swap
 0.74% :     0.000007s :     10: predicate.environ_get_eliminate
 0.19% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.03% :     0.000000s :      3: predicate.fold_const_symbol
 0.62% :     0.000006s :      6: predicate.get_grad_eliminate
 0.03% :     0.000000s :      3: predicate.graph_param_transform
 1.58% :     0.000015s :     22: predicate.inline
 0.66% :     0.000006s :      6: predicate.inline_without_move
 0.07% :     0.000001s :      6: predicate.j_node_and_user_rematch
 0.31% :     0.000003s :      6: predicate.less_batch_normalization
 0.43% :     0.000004s :     12: predicate.list_to_tuple_eliminator_
 0.31% :     0.000003s :     15: predicate.load_eliminater
 0.32% :     0.000003s :      3: predicate.loop_unroll_after_grad
 1.16% :     0.000011s :     34: predicate.loop_unroll_before_grad
 0.88% :     0.000008s :     13: predicate.make_slice_get_slice_eliminator
 0.16% :     0.000002s :     10: predicate.merge_addn
 0.30% :     0.000003s :     10: predicate.minmaximum_grad
 0.24% :     0.000002s :      3: predicate.mutable_eliminate
 0.08% :     0.000001s :      3: predicate.opt_reshape
 1.18% :     0.000011s :     15: predicate.partial_eliminate
 0.33% :     0.000003s :     10: predicate.print_const_string_wrapper
 0.81% :     0.000008s :     10: predicate.reduce_eliminate
 0.69% :     0.000007s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000005s :      6: predicate.remove_not_recompute_node
 0.27% :     0.000003s :     18: predicate.replace_applicator
 0.16% :     0.000002s :      6: predicate.replace_old_param
 0.07% :     0.000001s :      3: predicate.reset_defer_inline
 1.18% :     0.000011s :     10: predicate.reshape_eliminate
 1.20% :     0.000012s :     10: predicate.row_tensor_add_zeros_like
 0.15% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.26% :     0.000003s :     10: predicate.same_eliminate
59.77% :     0.000573s :      6: predicate.set_cell_output_no_recompute
 0.16% :     0.000002s :      6: predicate.special_op_eliminate
 0.96% :     0.000009s :      6: predicate.specialize_transform
 0.66% :     0.000006s :     10: predicate.split_environ_get_set_with_tuple_value
 0.67% :     0.000006s :     10: predicate.stack_unstack_eliminate
 0.06% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 0.66% :     0.000006s :     16: predicate.switch_defer_inline
 1.20% :     0.000012s :     16: predicate.switch_layer_defer_inline
 3.33% :     0.000032s :     53: predicate.switch_simplify
 0.71% :     0.000007s :     10: predicate.tile_eliminate
 0.85% :     0.000008s :     10: predicate.transpose_eliminate
 1.26% :     0.000012s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.20% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 1.67% :     0.000016s :     18: predicate.tuple_list_get_item_eliminator
 0.39% :     0.000004s :     10: predicate.tuple_list_set_item_eliminator
 0.39% :     0.000004s :     12: predicate.tuple_to_list_eliminator_
 0.66% :     0.000006s :     15: predicate.updatestate_pure_node_eliminater
 1.42% :     0.000014s :     21: predicate.updatestate_useless_node_eliminater
 1.25% :     0.000012s :     10: predicate.value_based_eliminate
 0.05% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.09% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001437    12
51.28% :     0.000737s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 9.01% :     0.000129s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.72% :     0.000571s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.075419    76
 0.08% :     0.000062s :      1: add_recomputation
 0.09% :     0.000070s :      1: auto_monad
 0.02% :     0.000019s :      1: auto_monad_reorder
 0.00% :     0.000003s :      1: backend_pass
 0.52% :     0.000395s :      1: bootstrap
 0.09% :     0.000066s :      1: cconv
 0.01% :     0.000009s :      1: convert_after_rewriter
 0.03% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000012s :      1: environ_conv
 0.03% :     0.000022s :      1: event_method
 0.02% :     0.000015s :      1: execute
 0.01% :     0.000007s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000024s :      1: graph_reusing
33.59% :     0.025336s :      1: jit_opt_a
 0.28% :     0.000211s :      1: jit_opt_after_cconv
25.42% :     0.019173s :      1: jit_opt_b
 1.26% :     0.000952s :      1: loop_unroll
 1.21% :     0.000915s :      1: mutable_eliminate
 6.25% :     0.004714s :     26: opt.transform.jit_opt_a
 0.09% :     0.000068s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000033s :      4: opt.transform.jit_opt_b
 0.03% :     0.000024s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.05% :     0.000036s :      4: opt.transform.symbol_engine_opt
 0.83% :     0.000626s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.05% :     0.000040s :      1: py_interpret_to_execute
 0.03% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000019s :      1: remove_dup_value
 3.50% :     0.002637s :      1: renormalize.infer
 2.87% :     0.002165s :      1: renormalize.specialize
 0.02% :     0.000015s :      1: rewriter_after_jit_bprop_graph
 0.06% :     0.000043s :      1: rewriter_after_opt_a
 1.32% :     0.000993s :      1: rewriter_before_opt_a
 0.12% :     0.000087s :      1: symbol_engine_optimizer
14.09% :     0.010628s :      1: task_emit
 7.71% :     0.005814s :      1: type_inference
 0.09% :     0.000072s :      1: validate


TotalTime = 0.0222228, [33]
[bootstrap]: 0.00032828
[type_inference]: 0.00550522
[event_method]: 2.208e-05
[auto_monad]: 7.241e-05
[graph_reusing]: 6.55997e-06
[pre_auto_parallel]: 2.74999e-06
[py_interpret_to_execute]: 3.831e-05
[rewriter_before_opt_a]: 7.793e-05
[expand_dump_flag]: 3.08e-06
[jit_opt_a]: 0.00627571, [2]
    [Cycle 1]: 0.00192838, [27]
        [switch_simplify]: 6.977e-05
        [loop_unroll]: 3.11e-05
        [a_1]: 0.00062043
        [with_stream_mark]: 2.549e-05
        [recompute_prepare]: 9.35001e-06
        [updatestate_depend_eliminate]: 4.59998e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 2.67001e-06
        [parameter_eliminate]: 1.99999e-06
        [specialize_transform]: 6.20002e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 5.67999e-06
        [meta_shard_fg_expand]: 1.91998e-06
        [get_grad_eliminate_]: 4.90999e-06
        [merge_forward]: 4.33001e-06
        [cell_reuse_recompute_pass]: 1.46998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.623e-05
        [j_node_and_user_rematch]: 9.57999e-06
        [meta_fg_expand]: 2.11e-06
        [replace_old_param]: 9.30001e-06
        [inline_without_move]: 5.00001e-06
        [renormalize]: 0.00080541
        [add_forward_monad_depend]: 7.40998e-06
        [auto_monad_grad]: 2.54999e-06
        [auto_monad_eliminator]: 1.713e-05
        [cse]: 3.312e-05
        [replace_applicator]: 1.617e-05
    [Cycle 2]: 0.00034136, [27]
        [switch_simplify]: 5.22999e-06
        [loop_unroll]: 4.75999e-06
        [a_1]: 8.914e-05
        [with_stream_mark]: 1.246e-05
        [recompute_prepare]: 5.20999e-06
        [updatestate_depend_eliminate]: 3.06999e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.50999e-06
        [specialize_transform]: 5.16002e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 5.25001e-06
        [meta_shard_fg_expand]: 1.35999e-06
        [get_grad_eliminate_]: 4.33999e-06
        [merge_forward]: 3.61001e-06
        [cell_reuse_recompute_pass]: 2.41998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.446e-05
        [j_node_and_user_rematch]: 8.43001e-06
        [meta_fg_expand]: 1.81e-06
        [replace_old_param]: 7.85998e-06
        [inline_without_move]: 4.57998e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.64e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 7.02002e-06
        [cse]: 1.273e-05
        [replace_applicator]: 5.81e-06
[py_interpret_to_execute_after_opt_a]: 1.426e-05
[rewriter_after_opt_a]: 4.1e-05
[convert_after_rewriter]: 7.65998e-06
[order_py_execute_after_rewriter]: 5.00001e-06
[mutable_eliminate]: 0.00069746
[jit_opt_b]: 5.272e-05, [1]
    [Cycle 1]: 4.416e-05, [2]
        [frontend_op_eliminate]: 1.561e-05
        [inline_after_opt_a]: 1.607e-05
[cconv]: 3.234e-05
[loop_unroll]: 0.00050928
[jit_opt_after_cconv]: 0.00016444, [1]
    [Cycle 1]: 0.00015686, [11]
        [c_1]: 2.104e-05
        [parameter_eliminate]: 4.82e-06
        [updatestate_depend_eliminate]: 1.095e-05
        [updatestate_assign_eliminate]: 2.67001e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [cse]: 2.944e-05
        [call_graph_tuple_transform]: 2.177e-05
        [tuple_list_get_item_eliminator]: 4.92999e-06
        [none_parameter_eliminate]: 1.71998e-06
        [renormalize]: 1.22e-06
        [switch_simplify]: 5.17e-06
[remove_dup_value]: 1.395e-05
[partial_unused_args_eliminate]: 2.32999e-06
[environ_conv]: 5.87001e-06
[add_recomputation]: 5.728e-05
[cse_after_recomputation]: 2.046e-05, [1]
    [Cycle 1]: 1.529e-05, [1]
        [cse]: 9.37999e-06
[auto_monad_reorder]: 1.766e-05
[get_jit_bprop_graph]: 2.11e-06
[rewriter_after_jit_bprop_graph]: 7.76001e-06
[opt_after_jit_grad]: 0.00053674
[symbol_engine_optimizer]: 8.12e-05, [1]
    [Cycle 1]: 7.299e-05, [6]
        [build]: 5.86e-06
        [elim_shapecalc]: 8.18001e-06
        [elim_not_effective]: 1.407e-05
        [opt_reshape]: 5.40001e-06
        [fold_const_symbol]: 9.12001e-06
        [renormalize]: 9.50007e-07
[validate]: 4.412e-05
[backend_pass]: 1.49e-06
[task_emit]: 0.00730836
[execute]: 1.116e-05
Sums
    bootstrap                            :     0.000328s :  1.88%
    type_inference                       :     0.005505s : 31.45%
    event_method                         :     0.000022s :  0.13%
    auto_monad                           :     0.000072s :  0.41%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.22%
    rewriter_before_opt_a                :     0.000078s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000075s :  0.43%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000710s :  4.05%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000805s :  4.60%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.14%
    jit_opt_a.cse                        :     0.000046s :  0.26%
    jit_opt_a.replace_applicator         :     0.000022s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000041s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000697s :  3.98%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000032s :  0.18%
    loop_unroll                          :     0.000509s :  2.91%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000057s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000018s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000537s :  3.07%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007308s : 41.76%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000271    23
 0.69% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.28% :     0.000006s :      3: substitution.graph_param_transform
85.50% :     0.000231s :      4: substitution.inline
 1.75% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.90% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.04% :     0.000006s :      2: substitution.replace_old_param
 5.30% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005420     2
82.70% :     0.004482s :      1: type_inference.infer
17.30% :     0.000938s :      1: type_inference.specialize
------[replace.]   0.000078     6
68.71% :     0.000054s :      4: replace.inline
31.29% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000242     6
94.51% :     0.000228s :      4: match.inline
 5.49% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.43% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.74% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000002s :     10: predicate.addn_zero_filter
 1.68% :     0.000002s :     10: predicate.arithmetic_simplify
 1.09% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.17% :     0.000001s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.13% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.09% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.24% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.80% :     0.000001s :      6: predicate.get_grad_eliminate
 0.45% :     0.000001s :      3: predicate.graph_param_transform
 6.78% :     0.000008s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.54% :     0.000002s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.64% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 2.42% :     0.000003s :      3: predicate.mutable_eliminate
 0.40% :     0.000000s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     15: predicate.partial_eliminate
 1.09% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.33% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.65% :     0.000001s :      3: predicate.reset_defer_inline
 1.82% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.67% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.76% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000003s :     16: predicate.switch_layer_defer_inline
 9.05% :     0.000011s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.97% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.31% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000671    12
50.61% :     0.000340s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.41% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.97% :     0.000275s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024085    76
 0.25% :     0.000060s :      1: add_recomputation
 0.32% :     0.000077s :      1: auto_monad
 0.09% :     0.000021s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 1.45% :     0.000350s :      1: bootstrap
 0.15% :     0.000035s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.11% :     0.000027s :      1: event_method
 0.07% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.07% :     0.006279s :      1: jit_opt_a
 0.70% :     0.000168s :      1: jit_opt_after_cconv
 0.23% :     0.000055s :      1: jit_opt_b
 2.16% :     0.000520s :      1: loop_unroll
 2.95% :     0.000709s :      1: mutable_eliminate
 3.91% :     0.000941s :     26: opt.transform.jit_opt_a
 0.21% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.28% :     0.000549s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000041s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.90% :     0.000457s :      1: renormalize.infer
 1.41% :     0.000339s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000045s :      1: rewriter_after_opt_a
 0.34% :     0.000082s :      1: rewriter_before_opt_a
 0.35% :     0.000084s :      1: symbol_engine_optimizer
30.44% :     0.007332s :      1: task_emit
22.96% :     0.005531s :      1: type_inference
 0.29% :     0.000069s :      1: validate


TotalTime = 0.0187585, [30]
[bootstrap]: 0.00036148
[type_inference]: 0.00534055
[event_method]: 2.016e-05
[auto_monad]: 6.996e-05
[graph_reusing]: 5.80002e-06
[pre_auto_parallel]: 2.62001e-06
[py_interpret_to_execute]: 4.278e-05
[rewriter_before_opt_a]: 7.951e-05
[expand_dump_flag]: 3.37002e-06
[jit_opt_a]: 0.009009, [2]
    [Cycle 1]: 0.00350661, [27]
        [switch_simplify]: 0.00012418
        [loop_unroll]: 5.819e-05
        [a_1]: 0.00064665
        [with_stream_mark]: 2.496e-05
        [recompute_prepare]: 7.23e-06
        [updatestate_depend_eliminate]: 3.88999e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 1.90001e-06
        [specialize_transform]: 6.04001e-06
        [updatestate_useless_node_eliminater]: 5.42999e-06
        [accelerated_algorithm]: 6.00002e-06
        [meta_shard_fg_expand]: 2.71e-06
        [get_grad_eliminate_]: 5.69999e-06
        [merge_forward]: 3.98001e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.494e-05
        [j_node_and_user_rematch]: 9.84999e-06
        [meta_fg_expand]: 2.84001e-06
        [replace_old_param]: 1.109e-05
        [inline_without_move]: 5.52999e-06
        [renormalize]: 0.00103248
        [add_forward_monad_depend]: 6.96001e-06
        [auto_monad_grad]: 2.58e-06
        [auto_monad_eliminator]: 1.847e-05
        [cse]: 3.502e-05
        [replace_applicator]: 1.597e-05
    [Cycle 2]: 0.00035012, [27]
        [switch_simplify]: 6.82002e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 9.588e-05
        [with_stream_mark]: 1.323e-05
        [recompute_prepare]: 5.64998e-06
        [updatestate_depend_eliminate]: 3.2e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 5.53002e-06
        [updatestate_useless_node_eliminater]: 5.39e-06
        [accelerated_algorithm]: 5.59e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 4.57998e-06
        [merge_forward]: 2.54999e-06
        [cell_reuse_recompute_pass]: 2.74001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.552e-05
        [j_node_and_user_rematch]: 8.52e-06
        [meta_fg_expand]: 1.41002e-06
        [replace_old_param]: 8.42e-06
        [inline_without_move]: 4.77e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 9.79984e-07
        [auto_monad_eliminator]: 6.17001e-06
        [cse]: 1.215e-05
        [replace_applicator]: 6.42001e-06
[py_interpret_to_execute_after_opt_a]: 1.433e-05
[rewriter_after_opt_a]: 0.00063476
[convert_after_rewriter]: 1.54e-05
[order_py_execute_after_rewriter]: 5.49e-06
[mutable_eliminate]: 0.00092936
[jit_opt_b]: 6.011e-05, [1]
    [Cycle 1]: 5.042e-05, [2]
        [frontend_op_eliminate]: 1.951e-05
        [inline_after_opt_a]: 1.778e-05
[cconv]: 3.606e-05
[loop_unroll]: 0.00047415
[jit_opt_after_cconv]: 0.00016966, [1]
    [Cycle 1]: 0.00016277, [11]
        [c_1]: 2.103e-05
        [parameter_eliminate]: 4.81002e-06
        [updatestate_depend_eliminate]: 7.68999e-06
        [updatestate_assign_eliminate]: 3.04999e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [cse]: 3.168e-05
        [call_graph_tuple_transform]: 2.306e-05
        [tuple_list_get_item_eliminator]: 5.32999e-06
        [none_parameter_eliminate]: 1.73002e-06
        [renormalize]: 1.28002e-06
        [switch_simplify]: 6.63e-06
[remove_dup_value]: 1.59e-05
[partial_unused_args_eliminate]: 2.51e-06
[environ_conv]: 6.17999e-06
[add_recomputation]: 5.36e-05
[cse_after_recomputation]: 2.363e-05, [1]
    [Cycle 1]: 1.732e-05, [1]
        [cse]: 1.081e-05
[auto_monad_reorder]: 1.653e-05
[get_jit_bprop_graph]: 2.16998e-06
[rewriter_after_jit_bprop_graph]: 9.32001e-06
[opt_after_jit_grad]: 0.00051144
[symbol_engine_optimizer]: 7.867e-05, [1]
    [Cycle 1]: 7.064e-05, [6]
        [build]: 4.74998e-06
        [elim_shapecalc]: 7.78999e-06
        [elim_not_effective]: 1.373e-05
        [opt_reshape]: 6.34999e-06
        [fold_const_symbol]: 8.67e-06
        [renormalize]: 8.50006e-07
[validate]: 4.518e-05
Sums
    bootstrap                            :     0.000361s :  3.23%
    type_inference                       :     0.005341s : 47.75%
    event_method                         :     0.000020s :  0.18%
    auto_monad                           :     0.000070s :  0.63%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.38%
    rewriter_before_opt_a                :     0.000080s :  0.71%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000131s :  1.17%
    jit_opt_a.loop_unroll                :     0.000063s :  0.56%
    jit_opt_a.a_1                        :     0.000743s :  6.64%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.34%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000020s :  0.17%
    jit_opt_a.inline_without_move        :     0.000010s :  0.09%
    jit_opt_a.renormalize                :     0.001033s :  9.23%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.22%
    jit_opt_a.cse                        :     0.000047s :  0.42%
    jit_opt_a.replace_applicator         :     0.000022s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.13%
    rewriter_after_opt_a                 :     0.000635s :  5.67%
    convert_after_rewriter               :     0.000015s :  0.14%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000929s :  8.31%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.16%
    cconv                                :     0.000036s :  0.32%
    loop_unroll                          :     0.000474s :  4.24%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000032s :  0.28%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.06%
    remove_dup_value                     :     0.000016s :  0.14%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000054s :  0.48%
    cse_after_recomputation.cse          :     0.000011s :  0.10%
    auto_monad_reorder                   :     0.000017s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.08%
    opt_after_jit_grad                   :     0.000511s :  4.57%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000045s :  0.40%


Time group info:
------[substitution.]   0.000289    23
 0.71% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 2.14% :     0.000006s :      3: substitution.graph_param_transform
86.75% :     0.000250s :      4: substitution.inline
 1.31% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.78% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000006s :      2: substitution.replace_old_param
 4.74% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005253     2
81.02% :     0.004255s :      1: type_inference.infer
18.98% :     0.000997s :      1: type_inference.specialize
------[replace.]   0.000080     6
68.84% :     0.000055s :      4: replace.inline
31.16% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000261     6
95.16% :     0.000248s :      4: match.inline
 4.84% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 2.23% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.90% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.31% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.98% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 5.83% :     0.000008s :     22: predicate.inline
 0.77% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.52% :     0.000002s :     15: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 7.75% :     0.000011s :     34: predicate.loop_unroll_before_grad
 1.77% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 0.90% :     0.000001s :     10: predicate.minmaximum_grad
 2.16% :     0.000003s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 1.83% :     0.000002s :     15: predicate.partial_eliminate
 1.15% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.91% :     0.000003s :     10: predicate.reduce_eliminate
 1.56% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.40% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.54% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.40% :     0.000001s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.04% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.20% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 1.87% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.78% :     0.000011s :     53: predicate.switch_simplify
 1.33% :     0.000002s :     10: predicate.tile_eliminate
 1.04% :     0.000001s :     10: predicate.transpose_eliminate
 1.22% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.04% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.94% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.41% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000748    12
49.27% :     0.000369s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.56% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.17% :     0.000323s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020840    72
 0.28% :     0.000058s :      1: add_recomputation
 0.35% :     0.000073s :      1: auto_monad
 0.09% :     0.000020s :      1: auto_monad_reorder
 1.83% :     0.000381s :      1: bootstrap
 0.19% :     0.000039s :      1: cconv
 0.09% :     0.000019s :      1: convert_after_rewriter
 0.12% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000010s :      1: environ_conv
 0.12% :     0.000025s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
43.25% :     0.009014s :      1: jit_opt_a
 0.83% :     0.000173s :      1: jit_opt_after_cconv
 0.30% :     0.000063s :      1: jit_opt_b
 2.32% :     0.000484s :      1: loop_unroll
 4.52% :     0.000942s :      1: mutable_eliminate
 4.84% :     0.001008s :     26: opt.transform.jit_opt_a
 0.25% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000030s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.50% :     0.000522s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000046s :      1: py_interpret_to_execute
 2.42% :     0.000504s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000018s :      1: remove_dup_value
 2.81% :     0.000585s :      1: renormalize.infer
 2.10% :     0.000438s :      1: renormalize.specialize
 0.05% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 3.11% :     0.000648s :      1: rewriter_after_opt_a
 0.40% :     0.000083s :      1: rewriter_before_opt_a
 0.39% :     0.000081s :      1: symbol_engine_optimizer
25.74% :     0.005364s :      1: type_inference


TotalTime = 0.0215164, [33]
[bootstrap]: 0.00032569
[type_inference]: 0.00475147
[event_method]: 1.872e-05
[auto_monad]: 6.48e-05
[graph_reusing]: 6.37001e-06
[pre_auto_parallel]: 2.40002e-06
[py_interpret_to_execute]: 3.861e-05
[rewriter_before_opt_a]: 7.616e-05
[expand_dump_flag]: 2.98e-06
[jit_opt_a]: 0.00654865, [2]
    [Cycle 1]: 0.00180788, [27]
        [switch_simplify]: 6.449e-05
        [loop_unroll]: 2.979e-05
        [a_1]: 0.00057077
        [with_stream_mark]: 1.899e-05
        [recompute_prepare]: 7.56001e-06
        [updatestate_depend_eliminate]: 4.1e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 2.24001e-06
        [specialize_transform]: 5.56e-06
        [updatestate_useless_node_eliminater]: 4.96002e-06
        [accelerated_algorithm]: 5.42999e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 5.74999e-06
        [merge_forward]: 3.63999e-06
        [cell_reuse_recompute_pass]: 2.24999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.406e-05
        [j_node_and_user_rematch]: 9.46e-06
        [meta_fg_expand]: 2.09999e-06
        [replace_old_param]: 9.81998e-06
        [inline_without_move]: 5.81e-06
        [renormalize]: 0.00079336
        [add_forward_monad_depend]: 6.23e-06
        [auto_monad_grad]: 2.58e-06
        [auto_monad_eliminator]: 1.435e-05
        [cse]: 2.829e-05
        [replace_applicator]: 1.413e-05
    [Cycle 2]: 0.00032273, [27]
        [switch_simplify]: 5.36998e-06
        [loop_unroll]: 4.89003e-06
        [a_1]: 9.222e-05
        [with_stream_mark]: 1.025e-05
        [recompute_prepare]: 5.27999e-06
        [updatestate_depend_eliminate]: 2.75997e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.36002e-06
        [specialize_transform]: 4.63999e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 5.12e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 4.26001e-06
        [merge_forward]: 2.76999e-06
        [cell_reuse_recompute_pass]: 2.18002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.293e-05
        [j_node_and_user_rematch]: 7.75998e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 7.25e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.11002e-06
        [auto_monad_grad]: 8.79983e-07
        [auto_monad_eliminator]: 5.82001e-06
        [cse]: 9.75002e-06
        [replace_applicator]: 4.85001e-06
[py_interpret_to_execute_after_opt_a]: 1.152e-05
[rewriter_after_opt_a]: 3.487e-05
[convert_after_rewriter]: 6.73998e-06
[order_py_execute_after_rewriter]: 4.99e-06
[mutable_eliminate]: 0.00069599
[jit_opt_b]: 5.025e-05, [1]
    [Cycle 1]: 4.339e-05, [2]
        [frontend_op_eliminate]: 1.729e-05
        [inline_after_opt_a]: 1.454e-05
[cconv]: 2.49e-05
[loop_unroll]: 0.00045007
[jit_opt_after_cconv]: 0.00015101, [1]
    [Cycle 1]: 0.00014467, [11]
        [c_1]: 2.19e-05
        [parameter_eliminate]: 3.68999e-06
        [updatestate_depend_eliminate]: 6.62002e-06
        [updatestate_assign_eliminate]: 2.51998e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [cse]: 2.194e-05
        [call_graph_tuple_transform]: 2.044e-05
        [tuple_list_get_item_eliminator]: 5.59e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 8.09989e-07
        [switch_simplify]: 5.71998e-06
[remove_dup_value]: 1.608e-05
[partial_unused_args_eliminate]: 2.85998e-06
[environ_conv]: 0.00025369
[add_recomputation]: 5.48e-05
[cse_after_recomputation]: 3.009e-05, [1]
    [Cycle 1]: 2.325e-05, [1]
        [cse]: 1.343e-05
[auto_monad_reorder]: 1.677e-05
[get_jit_bprop_graph]: 2.41998e-06
[rewriter_after_jit_bprop_graph]: 6.14001e-06
[opt_after_jit_grad]: 0.00054677
[symbol_engine_optimizer]: 7.919e-05, [1]
    [Cycle 1]: 7.235e-05, [6]
        [build]: 5.09998e-06
        [elim_shapecalc]: 8.95001e-06
        [elim_not_effective]: 1.41e-05
        [opt_reshape]: 5.97999e-06
        [fold_const_symbol]: 8.55999e-06
        [renormalize]: 1.02e-06
[validate]: 4.376e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.00692107
[execute]: 9.79999e-06
Sums
    bootstrap                            :     0.000326s :  1.98%
    type_inference                       :     0.004751s : 28.95%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000065s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.24%
    rewriter_before_opt_a                :     0.000076s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000070s :  0.43%
    jit_opt_a.loop_unroll                :     0.000035s :  0.21%
    jit_opt_a.a_1                        :     0.000663s :  4.04%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000793s :  4.83%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000038s :  0.23%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000696s :  4.24%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000025s :  0.15%
    loop_unroll                          :     0.000450s :  2.74%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000254s :  1.55%
    add_recomputation                    :     0.000055s :  0.33%
    cse_after_recomputation.cse          :     0.000013s :  0.08%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000547s :  3.33%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006921s : 42.17%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000224    23
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.82% :     0.000006s :      3: substitution.graph_param_transform
84.30% :     0.000189s :      4: substitution.inline
 1.59% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.31% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.05% :     0.000005s :      2: substitution.replace_old_param
 5.50% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004679     2
80.57% :     0.003770s :      1: type_inference.infer
19.43% :     0.000909s :      1: type_inference.specialize
------[replace.]   0.000074     6
66.00% :     0.000049s :      4: replace.inline
34.00% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000197     6
94.29% :     0.000186s :      4: match.inline
 5.71% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.39% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.54% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.67% :     0.000002s :     10: predicate.addn_zero_filter
 2.49% :     0.000003s :     10: predicate.arithmetic_simplify
 1.46% :     0.000002s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.24% :     0.000001s :     10: predicate.depend_value_elim
 1.24% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.24% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.48% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.07% :     0.000002s :     15: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.85% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.16% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 2.25% :     0.000003s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.09% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.41% :     0.000002s :     10: predicate.reduce_eliminate
 1.56% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.53% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.18% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.08% :     0.000004s :     16: predicate.switch_defer_inline
 2.01% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.26% :     0.000009s :     53: predicate.switch_simplify
 1.32% :     0.000002s :     10: predicate.tile_eliminate
 1.25% :     0.000002s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.72% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.58% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000598    12
45.83% :     0.000274s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.38% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.79% :     0.000292s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023298    76
 0.25% :     0.000058s :      1: add_recomputation
 0.29% :     0.000068s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.46% :     0.000340s :      1: bootstrap
 0.12% :     0.000027s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000033s :      1: cse_after_recomputation
 1.12% :     0.000262s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
28.12% :     0.006552s :      1: jit_opt_a
 0.66% :     0.000154s :      1: jit_opt_after_cconv
 0.23% :     0.000053s :      1: jit_opt_b
 1.97% :     0.000458s :      1: loop_unroll
 3.03% :     0.000706s :      1: mutable_eliminate
 3.79% :     0.000883s :     26: opt.transform.jit_opt_a
 0.21% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.39% :     0.000557s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000041s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.28% :     0.000065s :      1: remove_dup_value
 1.79% :     0.000418s :      1: renormalize.infer
 1.58% :     0.000368s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000038s :      1: rewriter_after_opt_a
 0.34% :     0.000079s :      1: rewriter_before_opt_a
 0.35% :     0.000082s :      1: symbol_engine_optimizer
29.78% :     0.006937s :      1: task_emit
20.46% :     0.004768s :      1: type_inference
 0.28% :     0.000064s :      1: validate


TotalTime = 0.0458458, [33]
[bootstrap]: 0.00036586
[type_inference]: 0.00533103
[event_method]: 1.919e-05
[auto_monad]: 6.655e-05
[graph_reusing]: 6.40997e-06
[pre_auto_parallel]: 3.26001e-06
[py_interpret_to_execute]: 3.608e-05
[rewriter_before_opt_a]: 7.523e-05
[expand_dump_flag]: 3.46001e-06
[jit_opt_a]: 0.00605121, [2]
    [Cycle 1]: 0.00181672, [27]
        [switch_simplify]: 6.355e-05
        [loop_unroll]: 2.85e-05
        [a_1]: 0.00058578
        [with_stream_mark]: 2.36e-05
        [recompute_prepare]: 8.11002e-06
        [updatestate_depend_eliminate]: 4.42998e-06
        [updatestate_assign_eliminate]: 3.31999e-06
        [updatestate_loads_eliminate]: 2.59001e-06
        [parameter_eliminate]: 2.27999e-06
        [specialize_transform]: 5.87001e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 5.83002e-06
        [meta_shard_fg_expand]: 2.14999e-06
        [get_grad_eliminate_]: 4.97e-06
        [merge_forward]: 3.81999e-06
        [cell_reuse_recompute_pass]: 1.52999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.542e-05
        [j_node_and_user_rematch]: 9.47999e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 8.97e-06
        [inline_without_move]: 5.52001e-06
        [renormalize]: 0.00076913
        [add_forward_monad_depend]: 6.57002e-06
        [auto_monad_grad]: 2.51e-06
        [auto_monad_eliminator]: 1.427e-05
        [cse]: 3.203e-05
        [replace_applicator]: 1.432e-05
    [Cycle 2]: 0.00033234, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 8.786e-05
        [with_stream_mark]: 1.272e-05
        [recompute_prepare]: 4.82998e-06
        [updatestate_depend_eliminate]: 2.74001e-06
        [updatestate_assign_eliminate]: 2.59001e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 4.57998e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 4.48001e-06
        [merge_forward]: 2.54999e-06
        [cell_reuse_recompute_pass]: 2.64001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.661e-05
        [j_node_and_user_rematch]: 8.1e-06
        [meta_fg_expand]: 1.58002e-06
        [replace_old_param]: 7.93001e-06
        [inline_without_move]: 4.59002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.85001e-06
        [auto_monad_grad]: 1.44e-06
        [auto_monad_eliminator]: 5.54e-06
        [cse]: 1.09e-05
        [replace_applicator]: 4.88001e-06
[py_interpret_to_execute_after_opt_a]: 1.379e-05
[rewriter_after_opt_a]: 4.108e-05
[convert_after_rewriter]: 7.63999e-06
[order_py_execute_after_rewriter]: 4.62e-06
[mutable_eliminate]: 0.0251044
[jit_opt_b]: 6.95e-05, [1]
    [Cycle 1]: 5.592e-05, [2]
        [frontend_op_eliminate]: 2.002e-05
        [inline_after_opt_a]: 2.153e-05
[cconv]: 4.588e-05
[loop_unroll]: 0.0004956
[jit_opt_after_cconv]: 0.0001943, [1]
    [Cycle 1]: 0.00018683, [11]
        [c_1]: 2.283e-05
        [parameter_eliminate]: 8.13999e-06
        [updatestate_depend_eliminate]: 1.307e-05
        [updatestate_assign_eliminate]: 3.56001e-06
        [updatestate_loads_eliminate]: 3.04999e-06
        [cse]: 4.442e-05
        [call_graph_tuple_transform]: 2.588e-05
        [tuple_list_get_item_eliminator]: 5.72999e-06
        [none_parameter_eliminate]: 1.74e-06
        [renormalize]: 8.00006e-07
        [switch_simplify]: 5.76998e-06
[remove_dup_value]: 1.68e-05
[partial_unused_args_eliminate]: 2.49999e-06
[environ_conv]: 6.74001e-06
[add_recomputation]: 6.249e-05
[cse_after_recomputation]: 2.081e-05, [1]
    [Cycle 1]: 1.523e-05, [1]
        [cse]: 9.10999e-06
[auto_monad_reorder]: 1.703e-05
[get_jit_bprop_graph]: 2.68e-06
[rewriter_after_jit_bprop_graph]: 1.282e-05
[opt_after_jit_grad]: 0.00051474
[symbol_engine_optimizer]: 7.976e-05, [1]
    [Cycle 1]: 7.258e-05, [6]
        [build]: 6.51999e-06
        [elim_shapecalc]: 8.54e-06
        [elim_not_effective]: 1.404e-05
        [opt_reshape]: 5.99999e-06
        [fold_const_symbol]: 8.77e-06
        [renormalize]: 7.80012e-07
[validate]: 4.839e-05
[backend_pass]: 1.00001e-06
[task_emit]: 0.00686126
[execute]: 9.81e-06
Sums
    bootstrap                            :     0.000366s :  0.89%
    type_inference                       :     0.005331s : 12.92%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000067s :  0.16%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000036s :  0.09%
    rewriter_before_opt_a                :     0.000075s :  0.18%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000069s :  0.17%
    jit_opt_a.loop_unroll                :     0.000033s :  0.08%
    jit_opt_a.a_1                        :     0.000674s :  1.63%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.09%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.02%
    jit_opt_a.merge_forward              :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000017s :  0.04%
    jit_opt_a.inline_without_move        :     0.000010s :  0.02%
    jit_opt_a.renormalize                :     0.000769s :  1.86%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.05%
    jit_opt_a.cse                        :     0.000043s :  0.10%
    jit_opt_a.replace_applicator         :     0.000019s :  0.05%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.03%
    rewriter_after_opt_a                 :     0.000041s :  0.10%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.025104s : 60.86%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000022s :  0.05%
    cconv                                :     0.000046s :  0.11%
    loop_unroll                          :     0.000496s :  1.20%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000008s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000013s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000044s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.01%
    remove_dup_value                     :     0.000017s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000062s :  0.15%
    cse_after_recomputation.cse          :     0.000009s :  0.02%
    auto_monad_reorder                   :     0.000017s :  0.04%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000013s :  0.03%
    opt_after_jit_grad                   :     0.000515s :  1.25%
    symbol_engine_optimizer.build        :     0.000007s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000048s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006861s : 16.63%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000257    23
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.61% :     0.000007s :      3: substitution.graph_param_transform
85.39% :     0.000219s :      4: substitution.inline
 1.62% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.79% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.11% :     0.000005s :      2: substitution.replace_old_param
 5.12% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005255     2
83.78% :     0.004402s :      1: type_inference.infer
16.22% :     0.000853s :      1: type_inference.specialize
------[replace.]   0.000071     6
67.99% :     0.000048s :      4: replace.inline
32.01% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000229     6
94.74% :     0.000217s :      4: match.inline
 5.26% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.41% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 1.76% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000002s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.15% :     0.000001s :     10: predicate.depend_value_elim
 0.97% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.09% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.45% :     0.000001s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.01% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     10: predicate.environ_get_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000001s :      3: predicate.graph_param_transform
 5.69% :     0.000007s :     22: predicate.inline
 1.18% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.62% :     0.000002s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.61% :     0.000002s :     15: predicate.load_eliminater
 2.13% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.54% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.08% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 5.11% :     0.000006s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.28% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.78% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.45% :     0.000001s :      6: predicate.replace_old_param
 1.03% :     0.000001s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.30% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.07% :     0.000001s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.03% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.15% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.33% :     0.000009s :     53: predicate.switch_simplify
 1.07% :     0.000001s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.25% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.07% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.21% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.29% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.30% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.66% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.83% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000638    12
50.82% :     0.000324s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.54% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.63% :     0.000259s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.047653    76
 0.14% :     0.000066s :      1: add_recomputation
 0.15% :     0.000070s :      1: auto_monad
 0.04% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.82% :     0.000389s :      1: bootstrap
 0.10% :     0.000049s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.05% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.05% :     0.000024s :      1: event_method
 0.03% :     0.000016s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000006s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
12.71% :     0.006055s :      1: jit_opt_a
 0.41% :     0.000197s :      1: jit_opt_after_cconv
 0.15% :     0.000073s :      1: jit_opt_b
 1.06% :     0.000505s :      1: loop_unroll
52.73% :     0.025128s :      1: mutable_eliminate
 1.88% :     0.000896s :     26: opt.transform.jit_opt_a
 0.12% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000033s :      4: opt.transform.jit_opt_b
 0.03% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000035s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.10% :     0.000524s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000039s :      1: py_interpret_to_execute
 0.03% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000019s :      1: remove_dup_value
 0.93% :     0.000441s :      1: renormalize.infer
 0.67% :     0.000320s :      1: renormalize.specialize
 0.03% :     0.000015s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000045s :      1: rewriter_after_opt_a
 0.16% :     0.000078s :      1: rewriter_before_opt_a
 0.17% :     0.000082s :      1: symbol_engine_optimizer
14.44% :     0.006880s :      1: task_emit
11.23% :     0.005351s :      1: type_inference
 0.15% :     0.000073s :      1: validate


TotalTime = 0.0215367, [33]
[bootstrap]: 0.0003556
[type_inference]: 0.00487577
[event_method]: 1.967e-05
[auto_monad]: 6.427e-05
[graph_reusing]: 5.71e-06
[pre_auto_parallel]: 2.56998e-06
[py_interpret_to_execute]: 4.283e-05
[rewriter_before_opt_a]: 7.542e-05
[expand_dump_flag]: 3.27002e-06
[jit_opt_a]: 0.00606899, [2]
    [Cycle 1]: 0.00192351, [27]
        [switch_simplify]: 7e-05
        [loop_unroll]: 3.108e-05
        [a_1]: 0.00059148
        [with_stream_mark]: 1.826e-05
        [recompute_prepare]: 7.35998e-06
        [updatestate_depend_eliminate]: 3.97e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [parameter_eliminate]: 2.02001e-06
        [specialize_transform]: 6.06998e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 6.50997e-06
        [meta_shard_fg_expand]: 2.35002e-06
        [get_grad_eliminate_]: 5.51e-06
        [merge_forward]: 3.51999e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.436e-05
        [j_node_and_user_rematch]: 9.39998e-06
        [meta_fg_expand]: 2.51e-06
        [replace_old_param]: 9.57001e-06
        [inline_without_move]: 5.66e-06
        [renormalize]: 0.00083175
        [add_forward_monad_depend]: 7.4e-06
        [auto_monad_grad]: 2.30002e-06
        [auto_monad_eliminator]: 1.603e-05
        [cse]: 3.019e-05
        [replace_applicator]: 1.526e-05
    [Cycle 2]: 0.00035336, [27]
        [switch_simplify]: 6.26e-06
        [loop_unroll]: 5.35999e-06
        [a_1]: 9.535e-05
        [with_stream_mark]: 1.077e-05
        [recompute_prepare]: 5.71e-06
        [updatestate_depend_eliminate]: 2.95002e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 5.07999e-06
        [updatestate_useless_node_eliminater]: 4.99998e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 1.66998e-06
        [get_grad_eliminate_]: 4.82e-06
        [merge_forward]: 2.55002e-06
        [cell_reuse_recompute_pass]: 1.90001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.495e-05
        [j_node_and_user_rematch]: 8.63001e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 8.32e-06
        [inline_without_move]: 4.63999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.46998e-06
        [auto_monad_grad]: 1.12e-06
        [auto_monad_eliminator]: 5.04e-06
        [cse]: 1.211e-05
        [replace_applicator]: 5.67999e-06
[py_interpret_to_execute_after_opt_a]: 1.262e-05
[rewriter_after_opt_a]: 3.665e-05
[convert_after_rewriter]: 7.70998e-06
[order_py_execute_after_rewriter]: 6.06998e-06
[mutable_eliminate]: 0.00073421
[jit_opt_b]: 5.583e-05, [1]
    [Cycle 1]: 4.788e-05, [2]
        [frontend_op_eliminate]: 1.838e-05
        [inline_after_opt_a]: 1.708e-05
[cconv]: 3.038e-05
[loop_unroll]: 0.00048299
[jit_opt_after_cconv]: 0.00016757, [1]
    [Cycle 1]: 0.0001609, [11]
        [c_1]: 2.454e-05
        [parameter_eliminate]: 3.83001e-06
        [updatestate_depend_eliminate]: 6.83e-06
        [updatestate_assign_eliminate]: 2.84001e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 2.894e-05
        [call_graph_tuple_transform]: 2.368e-05
        [tuple_list_get_item_eliminator]: 5.91998e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 8.40024e-07
        [switch_simplify]: 6.33e-06
[remove_dup_value]: 1.479e-05
[partial_unused_args_eliminate]: 2.04e-06
[environ_conv]: 4.92e-06
[add_recomputation]: 5.381e-05
[cse_after_recomputation]: 2.295e-05, [1]
    [Cycle 1]: 1.693e-05, [1]
        [cse]: 1.069e-05
[auto_monad_reorder]: 1.545e-05
[get_jit_bprop_graph]: 1.59e-06
[rewriter_after_jit_bprop_graph]: 8.12e-06
[opt_after_jit_grad]: 0.00052718
[symbol_engine_optimizer]: 8.536e-05, [1]
    [Cycle 1]: 7.704e-05, [6]
        [build]: 5.82001e-06
        [elim_shapecalc]: 8.62998e-06
        [elim_not_effective]: 1.501e-05
        [opt_reshape]: 6.83e-06
        [fold_const_symbol]: 1.044e-05
        [renormalize]: 9.20001e-07
[validate]: 4.481e-05
[backend_pass]: 1.20999e-06
[task_emit]: 0.00745851
[execute]: 7.88999e-06
Sums
    bootstrap                            :     0.000356s :  2.09%
    type_inference                       :     0.004876s : 28.64%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000064s :  0.38%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.25%
    rewriter_before_opt_a                :     0.000075s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.45%
    jit_opt_a.loop_unroll                :     0.000036s :  0.21%
    jit_opt_a.a_1                        :     0.000687s :  4.03%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000832s :  4.89%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.12%
    jit_opt_a.cse                        :     0.000042s :  0.25%
    jit_opt_a.replace_applicator         :     0.000021s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000734s :  4.31%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.10%
    cconv                                :     0.000030s :  0.18%
    loop_unroll                          :     0.000483s :  2.84%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000054s :  0.32%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000527s :  3.10%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000045s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007459s : 43.81%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000232    23
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.99% :     0.000002s :      2: substitution.fold_const_symbol
 2.90% :     0.000007s :      3: substitution.graph_param_transform
83.99% :     0.000195s :      4: substitution.inline
 1.50% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000005s :      2: substitution.replace_old_param
 5.76% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004802     2
80.58% :     0.003869s :      1: type_inference.infer
19.42% :     0.000932s :      1: type_inference.specialize
------[replace.]   0.000086     6
66.71% :     0.000057s :      4: replace.inline
33.29% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000205     6
93.99% :     0.000192s :      4: match.inline
 6.01% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.07% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.83% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000002s :     10: predicate.addn_zero_filter
 2.31% :     0.000003s :     10: predicate.arithmetic_simplify
 1.27% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000002s :     10: predicate.depend_value_elim
 1.24% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.34% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.76% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     10: predicate.environ_get_eliminate
 1.18% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.08% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.36% :     0.000007s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.52% :     0.000002s :      6: predicate.less_batch_normalization
 1.90% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.87% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.15% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.25% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.06% :     0.000003s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.03% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.57% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.27% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.09% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 1.97% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.95% :     0.000010s :     53: predicate.switch_simplify
 1.10% :     0.000001s :     10: predicate.tile_eliminate
 1.07% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.54% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.25% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.36% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000657    12
48.98% :     0.000322s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.98% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.04% :     0.000302s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023407    76
 0.24% :     0.000057s :      1: add_recomputation
 0.29% :     0.000068s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.61% :     0.000376s :      1: bootstrap
 0.14% :     0.000033s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
25.95% :     0.006074s :      1: jit_opt_a
 0.73% :     0.000171s :      1: jit_opt_after_cconv
 0.25% :     0.000059s :      1: jit_opt_b
 2.10% :     0.000492s :      1: loop_unroll
 3.19% :     0.000747s :      1: mutable_eliminate
 3.94% :     0.000922s :     26: opt.transform.jit_opt_a
 0.24% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000029s :      4: opt.transform.jit_opt_b
 0.06% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.30% :     0.000538s :      1: opt_after_jit_grad
 0.04% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000046s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.86% :     0.000434s :      1: renormalize.infer
 1.66% :     0.000388s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000043s :      1: rewriter_after_opt_a
 0.34% :     0.000078s :      1: rewriter_before_opt_a
 0.38% :     0.000088s :      1: symbol_engine_optimizer
31.93% :     0.007474s :      1: task_emit
20.90% :     0.004893s :      1: type_inference
 0.29% :     0.000069s :      1: validate


TotalTime = 0.0245649, [33]
[bootstrap]: 0.00032772
[type_inference]: 0.00518857
[event_method]: 1.754e-05
[auto_monad]: 6.259e-05
[graph_reusing]: 5.84e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 3.714e-05
[rewriter_before_opt_a]: 7.19e-05
[expand_dump_flag]: 2.68e-06
[jit_opt_a]: 0.00700978, [2]
    [Cycle 1]: 0.00205077, [27]
        [switch_simplify]: 6.859e-05
        [loop_unroll]: 2.915e-05
        [a_1]: 0.00064656
        [with_stream_mark]: 2.314e-05
        [recompute_prepare]: 8.10999e-06
        [updatestate_depend_eliminate]: 3.43999e-06
        [updatestate_assign_eliminate]: 3.35e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [parameter_eliminate]: 2.15002e-06
        [specialize_transform]: 6.28e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.32999e-06
        [meta_shard_fg_expand]: 2.08998e-06
        [get_grad_eliminate_]: 5.20001e-06
        [merge_forward]: 3.48999e-06
        [cell_reuse_recompute_pass]: 1.79998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.54e-05
        [j_node_and_user_rematch]: 9.55001e-06
        [meta_fg_expand]: 2.48e-06
        [replace_old_param]: 9.47001e-06
        [inline_without_move]: 5.40001e-06
        [renormalize]: 0.00093965
        [add_forward_monad_depend]: 6.91001e-06
        [auto_monad_grad]: 2.59999e-06
        [auto_monad_eliminator]: 1.41e-05
        [cse]: 2.872e-05
        [replace_applicator]: 1.606e-05
    [Cycle 2]: 0.0003294, [27]
        [switch_simplify]: 5.84e-06
        [loop_unroll]: 4.74e-06
        [a_1]: 9.37e-05
        [with_stream_mark]: 1.15e-05
        [recompute_prepare]: 5.52999e-06
        [updatestate_depend_eliminate]: 2.61e-06
        [updatestate_assign_eliminate]: 2.10002e-06
        [updatestate_loads_eliminate]: 2.11998e-06
        [parameter_eliminate]: 1.30999e-06
        [specialize_transform]: 4.79e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 4.90999e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.34997e-06
        [merge_forward]: 2.81e-06
        [cell_reuse_recompute_pass]: 1.84e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.352e-05
        [j_node_and_user_rematch]: 8.05e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 7.75998e-06
        [inline_without_move]: 4.54002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 1.32e-06
        [auto_monad_eliminator]: 5.37001e-06
        [cse]: 1.086e-05
        [replace_applicator]: 4.95001e-06
[py_interpret_to_execute_after_opt_a]: 1.177e-05
[rewriter_after_opt_a]: 3.79e-05
[convert_after_rewriter]: 6.86001e-06
[order_py_execute_after_rewriter]: 8.43999e-06
[mutable_eliminate]: 0.00077068
[jit_opt_b]: 5.705e-05, [1]
    [Cycle 1]: 4.777e-05, [2]
        [frontend_op_eliminate]: 1.886e-05
        [inline_after_opt_a]: 1.629e-05
[cconv]: 3.135e-05
[loop_unroll]: 0.00048139
[jit_opt_after_cconv]: 0.00015694, [1]
    [Cycle 1]: 0.00015001, [11]
        [c_1]: 2.175e-05
        [parameter_eliminate]: 3.68e-06
        [updatestate_depend_eliminate]: 7.3e-06
        [updatestate_assign_eliminate]: 2.71999e-06
        [updatestate_loads_eliminate]: 3.03e-06
        [cse]: 2.379e-05
        [call_graph_tuple_transform]: 2.161e-05
        [tuple_list_get_item_eliminator]: 5.05001e-06
        [none_parameter_eliminate]: 1.71e-06
        [renormalize]: 1.13001e-06
        [switch_simplify]: 5.63002e-06
[remove_dup_value]: 1.453e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 5.56e-06
[add_recomputation]: 5.034e-05
[cse_after_recomputation]: 2.021e-05, [1]
    [Cycle 1]: 1.456e-05, [1]
        [cse]: 8.75001e-06
[auto_monad_reorder]: 1.457e-05
[get_jit_bprop_graph]: 2.36e-06
[rewriter_after_jit_bprop_graph]: 7.73999e-06
[opt_after_jit_grad]: 0.00047963
[symbol_engine_optimizer]: 7.31e-05, [1]
    [Cycle 1]: 6.676e-05, [6]
        [build]: 4.38999e-06
        [elim_shapecalc]: 7.45e-06
        [elim_not_effective]: 1.316e-05
        [opt_reshape]: 5.34998e-06
        [fold_const_symbol]: 8.32e-06
        [renormalize]: 7.89994e-07
[validate]: 4.554e-05
[backend_pass]: 1.15001e-06
[task_emit]: 0.0092903
[execute]: 1.067e-05
Sums
    bootstrap                            :     0.000328s :  1.70%
    type_inference                       :     0.005189s : 26.95%
    event_method                         :     0.000018s :  0.09%
    auto_monad                           :     0.000063s :  0.33%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000037s :  0.19%
    rewriter_before_opt_a                :     0.000072s :  0.37%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000074s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.18%
    jit_opt_a.a_1                        :     0.000740s :  3.85%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000940s :  4.88%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.10%
    jit_opt_a.cse                        :     0.000040s :  0.21%
    jit_opt_a.replace_applicator         :     0.000021s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.06%
    rewriter_after_opt_a                 :     0.000038s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000008s :  0.04%
    mutable_eliminate                    :     0.000771s :  4.00%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.000031s :  0.16%
    loop_unroll                          :     0.000481s :  2.50%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000050s :  0.26%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000480s :  2.49%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.009290s : 48.26%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000262    23
 0.68% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 2.20% :     0.000006s :      3: substitution.graph_param_transform
86.39% :     0.000226s :      4: substitution.inline
 1.52% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.76% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.73% :     0.000005s :      2: substitution.replace_old_param
 5.24% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005118     2
82.35% :     0.004215s :      1: type_inference.infer
17.65% :     0.000903s :      1: type_inference.specialize
------[replace.]   0.000077     6
67.27% :     0.000052s :      4: replace.inline
32.73% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000236     6
94.64% :     0.000224s :      4: match.inline
 5.36% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.54% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.49% :     0.000002s :     10: predicate.addn_zero_filter
 2.75% :     0.000003s :     10: predicate.arithmetic_simplify
 1.30% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000001s :     10: predicate.depend_value_elim
 1.23% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.66% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.06% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.04% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.99% :     0.000007s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.34% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.64% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.72% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 2.33% :     0.000003s :      3: predicate.mutable_eliminate
 0.40% :     0.000000s :      3: predicate.opt_reshape
 2.02% :     0.000002s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.89% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.82% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.84% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.51% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.69% :     0.000009s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.50% :     0.000002s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.04% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.89% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000702    12
48.33% :     0.000339s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.58% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.09% :     0.000310s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026537    76
 0.20% :     0.000054s :      1: add_recomputation
 0.25% :     0.000066s :      1: auto_monad
 0.06% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.28% :     0.000340s :      1: bootstrap
 0.13% :     0.000034s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.08% :     0.000022s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.43% :     0.007014s :      1: jit_opt_a
 0.60% :     0.000160s :      1: jit_opt_after_cconv
 0.23% :     0.000060s :      1: jit_opt_b
 1.85% :     0.000491s :      1: loop_unroll
 2.95% :     0.000782s :      1: mutable_eliminate
 3.65% :     0.000967s :     26: opt.transform.jit_opt_a
 0.19% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.84% :     0.000489s :      1: opt_after_jit_grad
 0.04% :     0.000011s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000040s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 1.93% :     0.000512s :      1: renormalize.infer
 1.57% :     0.000417s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000041s :      1: rewriter_after_opt_a
 0.28% :     0.000075s :      1: rewriter_before_opt_a
 0.29% :     0.000076s :      1: symbol_engine_optimizer
35.08% :     0.009310s :      1: task_emit
19.62% :     0.005206s :      1: type_inference
 0.25% :     0.000067s :      1: validate


TotalTime = 0.0173363, [30]
[bootstrap]: 0.00043179
[type_inference]: 0.00622979
[event_method]: 2.095e-05
[auto_monad]: 6.643e-05
[graph_reusing]: 6.17999e-06
[pre_auto_parallel]: 2.61999e-06
[py_interpret_to_execute]: 4.391e-05
[rewriter_before_opt_a]: 7.963e-05
[expand_dump_flag]: 2.91999e-06
[jit_opt_a]: 0.00727698, [2]
    [Cycle 1]: 0.00267823, [27]
        [switch_simplify]: 7.207e-05
        [loop_unroll]: 2.949e-05
        [a_1]: 0.00067763
        [with_stream_mark]: 2.467e-05
        [recompute_prepare]: 8.45999e-06
        [updatestate_depend_eliminate]: 3.35e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.72001e-06
        [parameter_eliminate]: 2.12999e-06
        [specialize_transform]: 6.55002e-06
        [updatestate_useless_node_eliminater]: 5.16002e-06
        [accelerated_algorithm]: 6.54001e-06
        [meta_shard_fg_expand]: 2.14e-06
        [get_grad_eliminate_]: 6.36e-06
        [merge_forward]: 3.71999e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.609e-05
        [j_node_and_user_rematch]: 1.077e-05
        [meta_fg_expand]: 2.17001e-06
        [replace_old_param]: 1.109e-05
        [inline_without_move]: 5.74999e-06
        [renormalize]: 0.00148839
        [add_forward_monad_depend]: 8.95999e-06
        [auto_monad_grad]: 3.81999e-06
        [auto_monad_eliminator]: 1.689e-05
        [cse]: 2.821e-05
        [replace_applicator]: 1.986e-05
    [Cycle 2]: 0.00034774, [27]
        [switch_simplify]: 5.77001e-06
        [loop_unroll]: 5.27001e-06
        [a_1]: 0.00010065
        [with_stream_mark]: 1.356e-05
        [recompute_prepare]: 6.23e-06
        [updatestate_depend_eliminate]: 2.88e-06
        [updatestate_assign_eliminate]: 2.62001e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [parameter_eliminate]: 1.69e-06
        [specialize_transform]: 5.30999e-06
        [updatestate_useless_node_eliminater]: 4.63001e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 1.50999e-06
        [get_grad_eliminate_]: 5.20001e-06
        [merge_forward]: 3.48999e-06
        [cell_reuse_recompute_pass]: 3.16999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.653e-05
        [j_node_and_user_rematch]: 8.97e-06
        [meta_fg_expand]: 1.92001e-06
        [replace_old_param]: 8.31002e-06
        [inline_without_move]: 4.86002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.32999e-06
        [auto_monad_grad]: 1.15999e-06
        [auto_monad_eliminator]: 5.46e-06
        [cse]: 1.182e-05
        [replace_applicator]: 4.97999e-06
[py_interpret_to_execute_after_opt_a]: 1.488e-05
[rewriter_after_opt_a]: 3.933e-05
[convert_after_rewriter]: 7.06999e-06
[order_py_execute_after_rewriter]: 5.12999e-06
[mutable_eliminate]: 0.00082471
[jit_opt_b]: 6.193e-05, [1]
    [Cycle 1]: 5.176e-05, [2]
        [frontend_op_eliminate]: 1.928e-05
        [inline_after_opt_a]: 1.774e-05
[cconv]: 3.614e-05
[loop_unroll]: 0.00060268
[jit_opt_after_cconv]: 0.00020577, [1]
    [Cycle 1]: 0.00019761, [11]
        [c_1]: 2.611e-05
        [parameter_eliminate]: 5.74e-06
        [updatestate_depend_eliminate]: 1.133e-05
        [updatestate_assign_eliminate]: 3.06001e-06
        [updatestate_loads_eliminate]: 2.98e-06
        [cse]: 3.891e-05
        [call_graph_tuple_transform]: 3.209e-05
        [tuple_list_get_item_eliminator]: 6.79001e-06
        [none_parameter_eliminate]: 1.76e-06
        [renormalize]: 9.49978e-07
        [switch_simplify]: 7.35998e-06
[remove_dup_value]: 1.704e-05
[partial_unused_args_eliminate]: 2.73e-06
[environ_conv]: 6.93e-06
[add_recomputation]: 0.00012936
[cse_after_recomputation]: 3.275e-05, [1]
    [Cycle 1]: 2.358e-05, [1]
        [cse]: 1.442e-05
[auto_monad_reorder]: 1.751e-05
[get_jit_bprop_graph]: 2.53e-06
[rewriter_after_jit_bprop_graph]: 1.404e-05
[opt_after_jit_grad]: 0.00070793
[symbol_engine_optimizer]: 8.701e-05, [1]
    [Cycle 1]: 7.916e-05, [6]
        [build]: 6.57002e-06
        [elim_shapecalc]: 8.67998e-06
        [elim_not_effective]: 1.847e-05
        [opt_reshape]: 6.12999e-06
        [fold_const_symbol]: 9.04e-06
        [renormalize]: 8.39995e-07
[validate]: 5.052e-05
Sums
    bootstrap                            :     0.000432s :  3.51%
    type_inference                       :     0.006230s : 50.64%
    event_method                         :     0.000021s :  0.17%
    auto_monad                           :     0.000066s :  0.54%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000044s :  0.36%
    rewriter_before_opt_a                :     0.000080s :  0.65%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000078s :  0.63%
    jit_opt_a.loop_unroll                :     0.000035s :  0.28%
    jit_opt_a.a_1                        :     0.000778s :  6.33%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.16%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.001488s : 12.10%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.18%
    jit_opt_a.cse                        :     0.000040s :  0.33%
    jit_opt_a.replace_applicator         :     0.000025s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.12%
    rewriter_after_opt_a                 :     0.000039s :  0.32%
    convert_after_rewriter               :     0.000007s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000825s :  6.70%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.14%
    cconv                                :     0.000036s :  0.29%
    loop_unroll                          :     0.000603s :  4.90%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.09%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000039s :  0.32%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000032s :  0.26%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.06%
    remove_dup_value                     :     0.000017s :  0.14%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000007s :  0.06%
    add_recomputation                    :     0.000129s :  1.05%
    cse_after_recomputation.cse          :     0.000014s :  0.12%
    auto_monad_reorder                   :     0.000018s :  0.14%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000014s :  0.11%
    opt_after_jit_grad                   :     0.000708s :  5.75%
    symbol_engine_optimizer.build        :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000051s :  0.41%


Time group info:
------[substitution.]   0.000283    23
 1.08% :     0.000003s :      2: substitution.elim_not_effective
 0.53% :     0.000002s :      2: substitution.fold_const_symbol
 2.54% :     0.000007s :      3: substitution.graph_param_transform
85.25% :     0.000241s :      4: substitution.inline
 1.48% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.95% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.07% :     0.000006s :      2: substitution.replace_old_param
 5.10% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006138     2
82.80% :     0.005082s :      1: type_inference.infer
17.20% :     0.001056s :      1: type_inference.specialize
------[replace.]   0.000083     6
68.95% :     0.000057s :      4: replace.inline
31.05% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000252     6
94.69% :     0.000238s :      4: match.inline
 5.31% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000170   691
 0.88% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.77% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.73% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
25.74% :     0.000044s :     10: predicate.arithmetic_simplify
 0.85% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.73% :     0.000001s :     10: predicate.compare_switch_simplify
 0.89% :     0.000002s :     10: predicate.depend_value_elim
 0.83% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 0.95% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.79% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.22% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.22% :     0.000000s :      3: predicate.elim_not_effective
 0.44% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.78% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.70% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.99% :     0.000002s :     10: predicate.environ_get_eliminate
 0.85% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.15% :     0.000000s :      3: predicate.fold_const_symbol
 0.67% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.04% :     0.000009s :     22: predicate.inline
 0.65% :     0.000001s :      6: predicate.inline_without_move
 0.26% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000002s :      6: predicate.less_batch_normalization
 1.01% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.36% :     0.000002s :     15: predicate.load_eliminater
 1.81% :     0.000003s :      3: predicate.loop_unroll_after_grad
 2.68% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.49% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.76% :     0.000001s :     10: predicate.merge_addn
 0.73% :     0.000001s :     10: predicate.minmaximum_grad
 1.79% :     0.000003s :      3: predicate.mutable_eliminate
 0.36% :     0.000001s :      3: predicate.opt_reshape
 1.48% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000002s :     10: predicate.print_const_string_wrapper
 0.98% :     0.000002s :     10: predicate.reduce_eliminate
 1.09% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.26% :     0.000002s :     18: predicate.replace_applicator
 0.38% :     0.000001s :      6: predicate.replace_old_param
 0.38% :     0.000001s :      3: predicate.reset_defer_inline
 0.97% :     0.000002s :     10: predicate.reshape_eliminate
 0.85% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.66% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.82% :     0.000001s :     10: predicate.same_eliminate
 0.39% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.65% :     0.000001s :      6: predicate.special_op_eliminate
 0.65% :     0.000001s :      6: predicate.specialize_transform
 1.04% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.06% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.26% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 1.89% :     0.000003s :     16: predicate.switch_defer_inline
 1.46% :     0.000002s :     16: predicate.switch_layer_defer_inline
 5.79% :     0.000010s :     53: predicate.switch_simplify
 0.88% :     0.000001s :     10: predicate.tile_eliminate
 0.86% :     0.000001s :     10: predicate.transpose_eliminate
 1.03% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.85% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.15% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.12% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.19% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.13% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 1.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.29% :     0.000002s :     10: predicate.value_based_eliminate
 0.26% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.51% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000802    12
46.21% :     0.000371s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.15% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.64% :     0.000374s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019913    72
 0.67% :     0.000134s :      1: add_recomputation
 0.35% :     0.000070s :      1: auto_monad
 0.10% :     0.000021s :      1: auto_monad_reorder
 2.29% :     0.000456s :      1: bootstrap
 0.20% :     0.000040s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000035s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.13% :     0.000026s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.39% :     0.000078s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
36.56% :     0.007281s :      1: jit_opt_a
 1.05% :     0.000210s :      1: jit_opt_after_cconv
 0.33% :     0.000065s :      1: jit_opt_b
 3.09% :     0.000616s :      1: loop_unroll
 4.21% :     0.000837s :      1: mutable_eliminate
 5.15% :     0.001026s :     26: opt.transform.jit_opt_a
 0.34% :     0.000069s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000029s :      4: opt.transform.jit_opt_b
 0.09% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.18% :     0.000037s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000039s :      4: opt.transform.symbol_engine_opt
 3.62% :     0.000721s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000047s :      1: py_interpret_to_execute
 0.09% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000020s :      1: remove_dup_value
 5.01% :     0.000998s :      1: renormalize.infer
 2.33% :     0.000464s :      1: renormalize.specialize
 0.10% :     0.000019s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000043s :      1: rewriter_after_opt_a
 0.42% :     0.000084s :      1: rewriter_before_opt_a
 0.46% :     0.000091s :      1: symbol_engine_optimizer
31.38% :     0.006250s :      1: type_inference


TotalTime = 0.0397592, [33]
[bootstrap]: 0.00037151
[type_inference]: 0.00571173
[event_method]: 2.245e-05
[auto_monad]: 6.813e-05
[graph_reusing]: 6.55002e-06
[pre_auto_parallel]: 2.49001e-06
[py_interpret_to_execute]: 4.681e-05
[rewriter_before_opt_a]: 7.946e-05
[expand_dump_flag]: 3.23e-06
[jit_opt_a]: 0.00694087, [2]
    [Cycle 1]: 0.00242271, [27]
        [switch_simplify]: 7.379e-05
        [loop_unroll]: 3.433e-05
        [a_1]: 0.00076591
        [with_stream_mark]: 3.064e-05
        [recompute_prepare]: 1.186e-05
        [updatestate_depend_eliminate]: 4.95999e-06
        [updatestate_assign_eliminate]: 2.93998e-06
        [updatestate_loads_eliminate]: 2.91999e-06
        [parameter_eliminate]: 2.73e-06
        [specialize_transform]: 7.44002e-06
        [updatestate_useless_node_eliminater]: 6.06e-06
        [accelerated_algorithm]: 7.15e-06
        [meta_shard_fg_expand]: 2.73e-06
        [get_grad_eliminate_]: 6.38e-06
        [merge_forward]: 4.35e-06
        [cell_reuse_recompute_pass]: 1.92999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.862e-05
        [j_node_and_user_rematch]: 1.143e-05
        [meta_fg_expand]: 2.66e-06
        [replace_old_param]: 1.182e-05
        [inline_without_move]: 5.99e-06
        [renormalize]: 0.00109752
        [add_forward_monad_depend]: 9.32001e-06
        [auto_monad_grad]: 2.68e-06
        [auto_monad_eliminator]: 2.109e-05
        [cse]: 3.073e-05
        [replace_applicator]: 1.979e-05
    [Cycle 2]: 0.00039414, [27]
        [switch_simplify]: 6.68e-06
        [loop_unroll]: 5.56e-06
        [a_1]: 0.00010493
        [with_stream_mark]: 1.404e-05
        [recompute_prepare]: 6.14999e-06
        [updatestate_depend_eliminate]: 3.41999e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.80001e-06
        [specialize_transform]: 5.38002e-06
        [updatestate_useless_node_eliminater]: 5.27999e-06
        [accelerated_algorithm]: 5.70001e-06
        [meta_shard_fg_expand]: 1.64998e-06
        [get_grad_eliminate_]: 5.11002e-06
        [merge_forward]: 3.77998e-06
        [cell_reuse_recompute_pass]: 2.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.797e-05
        [j_node_and_user_rematch]: 9.24e-06
        [meta_fg_expand]: 2.12001e-06
        [replace_old_param]: 1.038e-05
        [inline_without_move]: 5.35999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 2.10002e-06
        [auto_monad_grad]: 1.87001e-06
        [auto_monad_eliminator]: 6.69999e-06
        [cse]: 1.265e-05
        [replace_applicator]: 7.10998e-06
[py_interpret_to_execute_after_opt_a]: 1.567e-05
[rewriter_after_opt_a]: 4.058e-05
[convert_after_rewriter]: 7.66001e-06
[order_py_execute_after_rewriter]: 5.59e-06
[mutable_eliminate]: 0.00082239
[jit_opt_b]: 7.552e-05, [1]
    [Cycle 1]: 6.629e-05, [2]
        [frontend_op_eliminate]: 3.161e-05
        [inline_after_opt_a]: 1.979e-05
[cconv]: 3.723e-05
[loop_unroll]: 0.00054563
[jit_opt_after_cconv]: 0.00018755, [1]
    [Cycle 1]: 0.00017901, [11]
        [c_1]: 2.57e-05
        [parameter_eliminate]: 5.47999e-06
        [updatestate_depend_eliminate]: 1.113e-05
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [cse]: 3.623e-05
        [call_graph_tuple_transform]: 2.345e-05
        [tuple_list_get_item_eliminator]: 5.59e-06
        [none_parameter_eliminate]: 1.35999e-06
        [renormalize]: 7.00005e-07
        [switch_simplify]: 6.06e-06
[remove_dup_value]: 1.622e-05
[partial_unused_args_eliminate]: 2.20002e-06
[environ_conv]: 6.53e-06
[add_recomputation]: 5.45e-05
[cse_after_recomputation]: 2.43e-05, [1]
    [Cycle 1]: 1.732e-05, [1]
        [cse]: 1.028e-05
[auto_monad_reorder]: 1.556e-05
[get_jit_bprop_graph]: 2.02001e-06
[rewriter_after_jit_bprop_graph]: 1.04e-05
[opt_after_jit_grad]: 0.00056182
[symbol_engine_optimizer]: 8.408e-05, [1]
    [Cycle 1]: 7.611e-05, [6]
        [build]: 5.20001e-06
        [elim_shapecalc]: 8.85999e-06
        [elim_not_effective]: 1.513e-05
        [opt_reshape]: 6.84001e-06
        [fold_const_symbol]: 9.74e-06
        [renormalize]: 5.89993e-07
[validate]: 4.748e-05
[backend_pass]: 1.02e-06
[task_emit]: 0.0236689
[execute]: 9.57001e-06
Sums
    bootstrap                            :     0.000372s :  1.07%
    type_inference                       :     0.005712s : 16.38%
    event_method                         :     0.000022s :  0.06%
    auto_monad                           :     0.000068s :  0.20%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000047s :  0.13%
    rewriter_before_opt_a                :     0.000079s :  0.23%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000080s :  0.23%
    jit_opt_a.loop_unroll                :     0.000040s :  0.11%
    jit_opt_a.a_1                        :     0.000871s :  2.50%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000013s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.03%
    jit_opt_a.merge_forward              :     0.000008s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.01%
    jit_opt_a.replace_old_param          :     0.000022s :  0.06%
    jit_opt_a.inline_without_move        :     0.000011s :  0.03%
    jit_opt_a.renormalize                :     0.001098s :  3.15%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.08%
    jit_opt_a.cse                        :     0.000043s :  0.12%
    jit_opt_a.replace_applicator         :     0.000027s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.04%
    rewriter_after_opt_a                 :     0.000041s :  0.12%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000822s :  2.36%
    jit_opt_b.frontend_op_eliminate      :     0.000032s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.06%
    cconv                                :     0.000037s :  0.11%
    loop_unroll                          :     0.000546s :  1.57%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000036s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000055s :  0.16%
    cse_after_recomputation.cse          :     0.000010s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.03%
    opt_after_jit_grad                   :     0.000562s :  1.61%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000047s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.023669s : 67.89%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.000353    23
 0.52% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000002s :      2: substitution.fold_const_symbol
 1.73% :     0.000006s :      3: substitution.graph_param_transform
87.53% :     0.000309s :      4: substitution.inline
 1.46% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.73% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.25% :     0.000008s :      2: substitution.replace_old_param
 4.26% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005627     2
81.03% :     0.004559s :      1: type_inference.infer
18.97% :     0.001068s :      1: type_inference.specialize
------[replace.]   0.000107     6
64.56% :     0.000069s :      4: replace.inline
35.44% :     0.000038s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000319     6
95.71% :     0.000305s :      4: match.inline
 4.29% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000149   691
 1.47% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.10% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.10% :     0.000002s :     10: predicate.addn_zero_filter
 2.61% :     0.000004s :     10: predicate.arithmetic_simplify
 1.56% :     0.000002s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.90% :     0.000001s :     10: predicate.compare_switch_simplify
 1.67% :     0.000002s :     10: predicate.depend_value_elim
 0.96% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.16% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.04% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.00% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.86% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000002s :     10: predicate.environ_get_eliminate
 1.06% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.08% :     0.000002s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 6.04% :     0.000009s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.74% :     0.000003s :      6: predicate.less_batch_normalization
 1.34% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.91% :     0.000003s :     15: predicate.load_eliminater
 1.57% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.20% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.07% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.92% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 2.62% :     0.000004s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 1.92% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.28% :     0.000002s :     10: predicate.reduce_eliminate
 1.59% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.39% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.55% :     0.000001s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.01% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.03% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.98% :     0.000004s :     16: predicate.switch_defer_inline
 1.89% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.59% :     0.000011s :     53: predicate.switch_simplify
 1.14% :     0.000002s :     10: predicate.tile_eliminate
 1.51% :     0.000002s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.79% :     0.000007s :     18: predicate.tuple_list_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.19% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.28% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000785    12
44.80% :     0.000352s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.61% :     0.000060s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.59% :     0.000374s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042129    76
 0.14% :     0.000058s :      1: add_recomputation
 0.17% :     0.000072s :      1: auto_monad
 0.04% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.93% :     0.000392s :      1: bootstrap
 0.10% :     0.000040s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.06% :     0.000027s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.06% :     0.000027s :      1: event_method
 0.04% :     0.000015s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
16.49% :     0.006945s :      1: jit_opt_a
 0.45% :     0.000191s :      1: jit_opt_after_cconv
 0.19% :     0.000080s :      1: jit_opt_b
 1.32% :     0.000557s :      1: loop_unroll
 1.99% :     0.000840s :      1: mutable_eliminate
 2.70% :     0.001140s :     26: opt.transform.jit_opt_a
 0.13% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000023s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.36% :     0.000572s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000050s :      1: py_interpret_to_execute
 0.04% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000019s :      1: remove_dup_value
 1.40% :     0.000592s :      1: renormalize.infer
 1.17% :     0.000494s :      1: renormalize.specialize
 0.03% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000045s :      1: rewriter_after_opt_a
 0.20% :     0.000083s :      1: rewriter_before_opt_a
 0.21% :     0.000087s :      1: symbol_engine_optimizer
56.23% :     0.023690s :      1: task_emit
13.60% :     0.005731s :      1: type_inference
 0.17% :     0.000072s :      1: validate


TotalTime = 0.0276336, [33]
[bootstrap]: 0.00038299
[type_inference]: 0.00605539
[event_method]: 2.252e-05
[auto_monad]: 7.402e-05
[graph_reusing]: 5.61998e-06
[pre_auto_parallel]: 3.26999e-06
[py_interpret_to_execute]: 5.343e-05
[rewriter_before_opt_a]: 8.511e-05
[expand_dump_flag]: 2.84001e-06
[jit_opt_a]: 0.0102372, [2]
    [Cycle 1]: 0.00253024, [27]
        [switch_simplify]: 8.398e-05
        [loop_unroll]: 3.092e-05
        [a_1]: 0.00081482
        [with_stream_mark]: 4.242e-05
        [recompute_prepare]: 1.297e-05
        [updatestate_depend_eliminate]: 4.89e-06
        [updatestate_assign_eliminate]: 3.9e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 2.64999e-06
        [specialize_transform]: 8.37e-06
        [updatestate_useless_node_eliminater]: 1.358e-05
        [accelerated_algorithm]: 7.28999e-06
        [meta_shard_fg_expand]: 2.61999e-06
        [get_grad_eliminate_]: 6.19999e-06
        [merge_forward]: 3.87998e-06
        [cell_reuse_recompute_pass]: 3.64002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.906e-05
        [j_node_and_user_rematch]: 1.03e-05
        [meta_fg_expand]: 2.91999e-06
        [replace_old_param]: 1.143e-05
        [inline_without_move]: 5.85002e-06
        [renormalize]: 0.00108544
        [add_forward_monad_depend]: 9.63002e-06
        [auto_monad_grad]: 3.50998e-06
        [auto_monad_eliminator]: 1.974e-05
        [cse]: 3.504e-05
        [replace_applicator]: 2.256e-05
    [Cycle 2]: 0.00041439, [27]
        [switch_simplify]: 6.29999e-06
        [loop_unroll]: 5.10999e-06
        [a_1]: 0.00010642
        [with_stream_mark]: 1.48e-05
        [recompute_prepare]: 6.87002e-06
        [updatestate_depend_eliminate]: 3.8e-06
        [updatestate_assign_eliminate]: 2.41998e-06
        [updatestate_loads_eliminate]: 2.34999e-06
        [parameter_eliminate]: 1.74e-06
        [specialize_transform]: 5.69e-06
        [updatestate_useless_node_eliminater]: 5.35999e-06
        [accelerated_algorithm]: 5.92999e-06
        [meta_shard_fg_expand]: 2.18998e-06
        [get_grad_eliminate_]: 5.81e-06
        [merge_forward]: 4.31002e-06
        [cell_reuse_recompute_pass]: 3.37002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.232e-05
        [j_node_and_user_rematch]: 9.77999e-06
        [meta_fg_expand]: 2.51e-06
        [replace_old_param]: 1.036e-05
        [inline_without_move]: 5.39e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.04e-06
        [auto_monad_grad]: 1.54e-06
        [auto_monad_eliminator]: 9.46e-06
        [cse]: 1.645e-05
        [replace_applicator]: 7.08e-06
[py_interpret_to_execute_after_opt_a]: 1.888e-05
[rewriter_after_opt_a]: 4.201e-05
[convert_after_rewriter]: 7.84997e-06
[order_py_execute_after_rewriter]: 5.56998e-06
[mutable_eliminate]: 0.0009325
[jit_opt_b]: 6.793e-05, [1]
    [Cycle 1]: 5.688e-05, [2]
        [frontend_op_eliminate]: 2.345e-05
        [inline_after_opt_a]: 1.9e-05
[cconv]: 4.237e-05
[loop_unroll]: 0.00063201
[jit_opt_after_cconv]: 0.00019423, [1]
    [Cycle 1]: 0.00018513, [11]
        [c_1]: 2.602e-05
        [parameter_eliminate]: 6.40002e-06
        [updatestate_depend_eliminate]: 1.069e-05
        [updatestate_assign_eliminate]: 3.16001e-06
        [updatestate_loads_eliminate]: 2.93998e-06
        [cse]: 3.81e-05
        [call_graph_tuple_transform]: 2.889e-05
        [tuple_list_get_item_eliminator]: 6.15002e-06
        [none_parameter_eliminate]: 1.60001e-06
        [renormalize]: 8.49977e-07
        [switch_simplify]: 6.04001e-06
[remove_dup_value]: 1.802e-05
[partial_unused_args_eliminate]: 2.51e-06
[environ_conv]: 8.03999e-06
[add_recomputation]: 6.07e-05
[cse_after_recomputation]: 2.75e-05, [1]
    [Cycle 1]: 1.994e-05, [1]
        [cse]: 1.122e-05
[auto_monad_reorder]: 1.773e-05
[get_jit_bprop_graph]: 2.32001e-06
[rewriter_after_jit_bprop_graph]: 1.164e-05
[opt_after_jit_grad]: 0.00063848
[symbol_engine_optimizer]: 8.672e-05, [1]
    [Cycle 1]: 7.81e-05, [6]
        [build]: 6.04999e-06
        [elim_shapecalc]: 8.2e-06
        [elim_not_effective]: 1.772e-05
        [opt_reshape]: 6.51e-06
        [fold_const_symbol]: 9.44e-06
        [renormalize]: 1.19e-06
[validate]: 5.032e-05
[backend_pass]: 1.11002e-06
[task_emit]: 0.00754245
[execute]: 8.50001e-06
Sums
    bootstrap                            :     0.000383s :  1.96%
    type_inference                       :     0.006055s : 31.05%
    event_method                         :     0.000023s :  0.12%
    auto_monad                           :     0.000074s :  0.38%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000053s :  0.27%
    rewriter_before_opt_a                :     0.000085s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000090s :  0.46%
    jit_opt_a.loop_unroll                :     0.000036s :  0.18%
    jit_opt_a.a_1                        :     0.000921s :  4.72%
    jit_opt_a.with_stream_mark           :     0.000057s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000020s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000014s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000019s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000041s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000022s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001086s :  5.57%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000029s :  0.15%
    jit_opt_a.cse                        :     0.000051s :  0.26%
    jit_opt_a.replace_applicator         :     0.000030s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.10%
    rewriter_after_opt_a                 :     0.000042s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000932s :  4.78%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.10%
    cconv                                :     0.000042s :  0.22%
    loop_unroll                          :     0.000632s :  3.24%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000038s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000029s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000018s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000008s :  0.04%
    add_recomputation                    :     0.000061s :  0.31%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000018s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000012s :  0.06%
    opt_after_jit_grad                   :     0.000638s :  3.27%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000050s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007542s : 38.68%
    execute                              :     0.000009s :  0.04%


Time group info:
------[substitution.]   0.000378    23
 0.75% :     0.000003s :      2: substitution.elim_not_effective
 0.34% :     0.000001s :      2: substitution.fold_const_symbol
 1.74% :     0.000007s :      3: substitution.graph_param_transform
74.39% :     0.000281s :      4: substitution.inline
 1.25% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.61% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.14% :     0.000008s :      2: substitution.replace_old_param
17.78% :     0.000067s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005962     2
82.09% :     0.004894s :      1: type_inference.infer
17.91% :     0.001068s :      1: type_inference.specialize
------[replace.]   0.000105     6
64.55% :     0.000068s :      4: replace.inline
35.45% :     0.000037s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000343     6
80.97% :     0.000278s :      4: match.inline
19.03% :     0.000065s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000168   691
 1.57% :     0.000003s :     10: predicate.accumulaten_eliminater
 1.06% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.73% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 2.43% :     0.000004s :     10: predicate.arithmetic_simplify
 0.97% :     0.000002s :     10: predicate.cast_eliminate
 0.29% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.82% :     0.000001s :     10: predicate.compare_switch_simplify
 1.01% :     0.000002s :     10: predicate.depend_value_elim
 0.72% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.55% :     0.000003s :     10: predicate.dict_get_item_eliminator
 0.94% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.21% :     0.000000s :      3: predicate.elim_not_effective
 0.45% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.72% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.72% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.79% :     0.000001s :     10: predicate.environ_get_eliminate
 1.01% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.16% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000002s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 6.10% :     0.000010s :     22: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.26% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :      6: predicate.less_batch_normalization
 1.28% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.03% :     0.000003s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 2.98% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.23% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 0.73% :     0.000001s :     10: predicate.merge_addn
 0.84% :     0.000001s :     10: predicate.minmaximum_grad
 2.51% :     0.000004s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 1.50% :     0.000003s :     15: predicate.partial_eliminate
 0.93% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000003s :     10: predicate.reduce_eliminate
 1.25% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.36% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.53% :     0.000003s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.43% :     0.000001s :      3: predicate.reset_defer_inline
 1.09% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.99% :     0.000002s :      3: predicate.row_tensor_eliminate
 0.80% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000002s :      6: predicate.special_op_eliminate
 0.91% :     0.000002s :      6: predicate.specialize_transform
10.05% :     0.000017s :     10: predicate.split_environ_get_set_with_tuple_value
 0.88% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.26% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.74% :     0.000005s :     16: predicate.switch_defer_inline
 1.41% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.06% :     0.000012s :     53: predicate.switch_simplify
 1.11% :     0.000002s :     10: predicate.tile_eliminate
 1.34% :     0.000002s :     10: predicate.transpose_eliminate
 1.57% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.06% :     0.000007s :     18: predicate.tuple_list_get_item_eliminator
 2.32% :     0.000004s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.10% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.22% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000003s :     10: predicate.value_based_eliminate
 0.25% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.84% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000806    12
46.84% :     0.000377s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.66% :     0.000062s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.50% :     0.000367s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.030075    76
 0.22% :     0.000066s :      1: add_recomputation
 0.26% :     0.000079s :      1: auto_monad
 0.07% :     0.000021s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 1.32% :     0.000398s :      1: bootstrap
 0.16% :     0.000047s :      1: cconv
 0.05% :     0.000015s :      1: convert_after_rewriter
 0.10% :     0.000030s :      1: cse_after_recomputation
 0.04% :     0.000011s :      1: environ_conv
 0.09% :     0.000028s :      1: event_method
 0.04% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
34.05% :     0.010242s :      1: jit_opt_a
 0.66% :     0.000197s :      1: jit_opt_after_cconv
 0.24% :     0.000072s :      1: jit_opt_b
 2.21% :     0.000665s :      1: loop_unroll
 3.17% :     0.000953s :      1: mutable_eliminate
 4.02% :     0.001209s :     26: opt.transform.jit_opt_a
 0.21% :     0.000063s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000034s :      4: opt.transform.jit_opt_b
 0.07% :     0.000021s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000029s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000038s :      4: opt.transform.symbol_engine_opt
 2.16% :     0.000650s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000006s :      1: pre_auto_parallel
 0.19% :     0.000057s :      1: py_interpret_to_execute
 0.07% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000021s :      1: remove_dup_value
 1.92% :     0.000577s :      1: renormalize.infer
 1.64% :     0.000495s :      1: renormalize.specialize
 0.05% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000047s :      1: rewriter_after_opt_a
 0.29% :     0.000089s :      1: rewriter_before_opt_a
 0.30% :     0.000090s :      1: symbol_engine_optimizer
25.14% :     0.007560s :      1: task_emit
20.21% :     0.006077s :      1: type_inference
 0.25% :     0.000075s :      1: validate


TotalTime = 0.0354465, [33]
[bootstrap]: 0.0004088
[type_inference]: 0.00577888
[event_method]: 2.154e-05
[auto_monad]: 7.171e-05
[graph_reusing]: 6.14001e-06
[pre_auto_parallel]: 2.76999e-06
[py_interpret_to_execute]: 4.103e-05
[rewriter_before_opt_a]: 7.863e-05
[expand_dump_flag]: 3.98001e-06
[jit_opt_a]: 0.018821, [2]
    [Cycle 1]: 0.0144017, [27]
        [switch_simplify]: 6.55e-05
        [loop_unroll]: 2.874e-05
        [a_1]: 0.0129134
        [with_stream_mark]: 3.267e-05
        [recompute_prepare]: 1.399e-05
        [updatestate_depend_eliminate]: 4.53999e-06
        [updatestate_assign_eliminate]: 2.84999e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [parameter_eliminate]: 2.57001e-06
        [specialize_transform]: 6.36e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 3.33e-06
        [get_grad_eliminate_]: 5.17e-06
        [merge_forward]: 4.59002e-06
        [cell_reuse_recompute_pass]: 1.39998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.667e-05
        [j_node_and_user_rematch]: 1.048e-05
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 1.197e-05
        [inline_without_move]: 5.47999e-06
        [renormalize]: 0.0009127
        [add_forward_monad_depend]: 8.75999e-06
        [auto_monad_grad]: 3.45e-06
        [auto_monad_eliminator]: 1.89e-05
        [cse]: 3.579e-05
        [replace_applicator]: 1.759e-05
    [Cycle 2]: 0.00034617, [27]
        [switch_simplify]: 6.07001e-06
        [loop_unroll]: 5.51e-06
        [a_1]: 9.516e-05
        [with_stream_mark]: 1.355e-05
        [recompute_prepare]: 5.00999e-06
        [updatestate_depend_eliminate]: 3.59002e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 4.73001e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 5.27999e-06
        [meta_shard_fg_expand]: 1.61998e-06
        [get_grad_eliminate_]: 4.64002e-06
        [merge_forward]: 3.47002e-06
        [cell_reuse_recompute_pass]: 3.45998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.797e-05
        [j_node_and_user_rematch]: 8.23999e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 8.2e-06
        [inline_without_move]: 4.67e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 5.68002e-06
        [cse]: 1.16e-05
        [replace_applicator]: 5.09e-06
[py_interpret_to_execute_after_opt_a]: 1.584e-05
[rewriter_after_opt_a]: 4.135e-05
[convert_after_rewriter]: 7.28999e-06
[order_py_execute_after_rewriter]: 5.30001e-06
[mutable_eliminate]: 0.00087008
[jit_opt_b]: 5.344e-05, [1]
    [Cycle 1]: 4.491e-05, [2]
        [frontend_op_eliminate]: 1.626e-05
        [inline_after_opt_a]: 1.643e-05
[cconv]: 3.561e-05
[loop_unroll]: 0.00050752
[jit_opt_after_cconv]: 0.00017566, [1]
    [Cycle 1]: 0.000168, [11]
        [c_1]: 2.321e-05
        [parameter_eliminate]: 6.64999e-06
        [updatestate_depend_eliminate]: 1.095e-05
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.72001e-06
        [cse]: 3.245e-05
        [call_graph_tuple_transform]: 2.449e-05
        [tuple_list_get_item_eliminator]: 5.10001e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 5.60016e-07
        [switch_simplify]: 5.25001e-06
[remove_dup_value]: 1.574e-05
[partial_unused_args_eliminate]: 2.23998e-06
[environ_conv]: 6.01e-06
[add_recomputation]: 5.989e-05
[cse_after_recomputation]: 2.25e-05, [1]
    [Cycle 1]: 1.634e-05, [1]
        [cse]: 9.25001e-06
[auto_monad_reorder]: 1.757e-05
[get_jit_bprop_graph]: 2.30002e-06
[rewriter_after_jit_bprop_graph]: 9.37001e-06
[opt_after_jit_grad]: 0.00052195
[symbol_engine_optimizer]: 7.65e-05, [1]
    [Cycle 1]: 6.902e-05, [6]
        [build]: 4.57e-06
        [elim_shapecalc]: 8.12003e-06
        [elim_not_effective]: 1.382e-05
        [opt_reshape]: 5.62001e-06
        [fold_const_symbol]: 8.56002e-06
        [renormalize]: 1.17e-06
[validate]: 4.382e-05
[backend_pass]: 8.80013e-07
[task_emit]: 0.00745307
[execute]: 1.122e-05
Sums
    bootstrap                            :     0.000409s :  1.34%
    type_inference                       :     0.005779s : 18.88%
    event_method                         :     0.000022s :  0.07%
    auto_monad                           :     0.000072s :  0.23%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.13%
    rewriter_before_opt_a                :     0.000079s :  0.26%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000072s :  0.23%
    jit_opt_a.loop_unroll                :     0.000034s :  0.11%
    jit_opt_a.a_1                        :     0.013009s : 42.50%
    jit_opt_a.with_stream_mark           :     0.000046s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000008s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.11%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000020s :  0.07%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.000913s :  2.98%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.08%
    jit_opt_a.cse                        :     0.000047s :  0.15%
    jit_opt_a.replace_applicator         :     0.000023s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.05%
    rewriter_after_opt_a                 :     0.000041s :  0.14%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000870s :  2.84%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.05%
    cconv                                :     0.000036s :  0.12%
    loop_unroll                          :     0.000508s :  1.66%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000032s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000016s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000060s :  0.20%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000018s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.03%
    opt_after_jit_grad                   :     0.000522s :  1.71%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007453s : 24.35%
    execute                              :     0.000011s :  0.04%


Time group info:
------[substitution.]   0.000385    23
 0.57% :     0.000002s :      2: substitution.elim_not_effective
 0.34% :     0.000001s :      2: substitution.fold_const_symbol
 1.62% :     0.000006s :      3: substitution.graph_param_transform
88.64% :     0.000342s :      4: substitution.inline
 1.23% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.27% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.92% :     0.000007s :      2: substitution.replace_old_param
 4.42% :     0.000017s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005695     2
83.82% :     0.004774s :      1: type_inference.infer
16.18% :     0.000921s :      1: type_inference.specialize
------[replace.]   0.012187     6
99.79% :     0.012162s :      4: replace.inline
 0.21% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000353     6
95.65% :     0.000338s :      4: match.inline
 4.35% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000144   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.02% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.85% :     0.000001s :     10: predicate.addn_check_dump
 1.65% :     0.000002s :     10: predicate.addn_zero_filter
 2.73% :     0.000004s :     10: predicate.arithmetic_simplify
 1.76% :     0.000003s :     10: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.86% :     0.000001s :     10: predicate.compare_switch_simplify
 1.85% :     0.000003s :     10: predicate.depend_value_elim
 1.16% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.03% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.07% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.25% :     0.000000s :      3: predicate.elim_not_effective
 0.42% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.96% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000002s :     10: predicate.environ_get_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.17% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 8.94% :     0.000013s :     22: predicate.inline
 0.70% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000002s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.99% :     0.000003s :     15: predicate.load_eliminater
 1.67% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.10% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.64% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.90% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 2.77% :     0.000004s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     15: predicate.partial_eliminate
 1.01% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     10: predicate.reduce_eliminate
 1.80% :     0.000003s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.37% :     0.000002s :     18: predicate.replace_applicator
 0.41% :     0.000001s :      6: predicate.replace_old_param
 0.48% :     0.000001s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.09% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 0.81% :     0.000001s :      6: predicate.specialize_transform
 1.12% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.94% :     0.000006s :     16: predicate.switch_defer_inline
 1.98% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.78% :     0.000010s :     53: predicate.switch_simplify
 1.36% :     0.000002s :     10: predicate.tile_eliminate
 1.09% :     0.000002s :     10: predicate.transpose_eliminate
 1.57% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.05% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.29% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.97% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.37% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.30% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.26% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.40% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000695    12
50.05% :     0.000348s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.71% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.24% :     0.000287s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.049719    76
 0.13% :     0.000064s :      1: add_recomputation
 0.15% :     0.000075s :      1: auto_monad
 0.04% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.88% :     0.000439s :      1: bootstrap
 0.08% :     0.000039s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.05% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.05% :     0.000027s :      1: event_method
 0.04% :     0.000017s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
37.86% :     0.018825s :      1: jit_opt_a
 0.36% :     0.000178s :      1: jit_opt_after_cconv
 0.11% :     0.000056s :      1: jit_opt_b
 1.04% :     0.000516s :      1: loop_unroll
 1.78% :     0.000883s :      1: mutable_eliminate
26.64% :     0.013246s :     26: opt.transform.jit_opt_a
 0.11% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000026s :      4: opt.transform.jit_opt_b
 0.03% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.07% :     0.000532s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000044s :      1: py_interpret_to_execute
 0.04% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000018s :      1: remove_dup_value
 1.05% :     0.000522s :      1: renormalize.infer
 0.77% :     0.000381s :      1: renormalize.specialize
 0.02% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000044s :      1: rewriter_after_opt_a
 0.17% :     0.000082s :      1: rewriter_before_opt_a
 0.16% :     0.000079s :      1: symbol_engine_optimizer
15.04% :     0.007476s :      1: task_emit
11.67% :     0.005800s :      1: type_inference
 0.14% :     0.000069s :      1: validate


TotalTime = 0.0264568, [33]
[bootstrap]: 0.00052396
[type_inference]: 0.00644107
[event_method]: 2.4e-05
[auto_monad]: 7.6e-05
[graph_reusing]: 6.59001e-06
[pre_auto_parallel]: 3.08e-06
[py_interpret_to_execute]: 5.244e-05
[rewriter_before_opt_a]: 8.509e-05
[expand_dump_flag]: 2.98e-06
[jit_opt_a]: 0.00757153, [2]
    [Cycle 1]: 0.00224787, [27]
        [switch_simplify]: 7.87e-05
        [loop_unroll]: 3.125e-05
        [a_1]: 0.00067863
        [with_stream_mark]: 2.701e-05
        [recompute_prepare]: 9.78002e-06
        [updatestate_depend_eliminate]: 4.75001e-06
        [updatestate_assign_eliminate]: 3.14001e-06
        [updatestate_loads_eliminate]: 2.56998e-06
        [parameter_eliminate]: 2.43e-06
        [specialize_transform]: 7.05998e-06
        [updatestate_useless_node_eliminater]: 5.66e-06
        [accelerated_algorithm]: 5.72999e-06
        [meta_shard_fg_expand]: 2.35002e-06
        [get_grad_eliminate_]: 5.19e-06
        [merge_forward]: 3.70998e-06
        [cell_reuse_recompute_pass]: 2.94999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.795e-05
        [j_node_and_user_rematch]: 1.004e-05
        [meta_fg_expand]: 2.44001e-06
        [replace_old_param]: 1.132e-05
        [inline_without_move]: 5.35999e-06
        [renormalize]: 0.00102797
        [add_forward_monad_depend]: 9.44e-06
        [auto_monad_grad]: 2.41e-06
        [auto_monad_eliminator]: 2.004e-05
        [cse]: 3.094e-05
        [replace_applicator]: 2.215e-05
    [Cycle 2]: 0.00038341, [27]
        [switch_simplify]: 6.19999e-06
        [loop_unroll]: 5.71e-06
        [a_1]: 0.00010161
        [with_stream_mark]: 1.672e-05
        [recompute_prepare]: 5.49998e-06
        [updatestate_depend_eliminate]: 3.33e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.40002e-06
        [parameter_eliminate]: 1.45999e-06
        [specialize_transform]: 5.09e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.51002e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 5.04e-06
        [merge_forward]: 3.06001e-06
        [cell_reuse_recompute_pass]: 3.78001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.841e-05
        [j_node_and_user_rematch]: 9.54e-06
        [meta_fg_expand]: 1.63002e-06
        [replace_old_param]: 8.17998e-06
        [inline_without_move]: 4.94e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.31e-06
        [auto_monad_grad]: 1.39e-06
        [auto_monad_eliminator]: 6.74999e-06
        [cse]: 1.304e-05
        [replace_applicator]: 4.78001e-06
[py_interpret_to_execute_after_opt_a]: 1.654e-05
[rewriter_after_opt_a]: 3.838e-05
[convert_after_rewriter]: 7.65998e-06
[order_py_execute_after_rewriter]: 5.77999e-06
[mutable_eliminate]: 0.00092967
[jit_opt_b]: 6.369e-05, [1]
    [Cycle 1]: 5.386e-05, [2]
        [frontend_op_eliminate]: 2.12e-05
        [inline_after_opt_a]: 1.671e-05
[cconv]: 3.856e-05
[loop_unroll]: 0.00056958
[jit_opt_after_cconv]: 0.00018254, [1]
    [Cycle 1]: 0.00017441, [11]
        [c_1]: 2.364e-05
        [parameter_eliminate]: 4.94e-06
        [updatestate_depend_eliminate]: 1.091e-05
        [updatestate_assign_eliminate]: 3.31999e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [cse]: 3.329e-05
        [call_graph_tuple_transform]: 2.62e-05
        [tuple_list_get_item_eliminator]: 5.64e-06
        [none_parameter_eliminate]: 1.87001e-06
        [renormalize]: 1.37999e-06
        [switch_simplify]: 5.79e-06
[remove_dup_value]: 1.577e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 7.92e-06
[add_recomputation]: 5.716e-05
[cse_after_recomputation]: 2.536e-05, [1]
    [Cycle 1]: 1.864e-05, [1]
        [cse]: 1.108e-05
[auto_monad_reorder]: 1.615e-05
[get_jit_bprop_graph]: 2.49001e-06
[rewriter_after_jit_bprop_graph]: 9.32999e-06
[opt_after_jit_grad]: 0.00056748
[symbol_engine_optimizer]: 8.52e-05, [1]
    [Cycle 1]: 7.733e-05, [6]
        [build]: 5.30001e-06
        [elim_shapecalc]: 9.12999e-06
        [elim_not_effective]: 1.582e-05
        [opt_reshape]: 6.38e-06
        [fold_const_symbol]: 9.55001e-06
        [renormalize]: 7.7e-07
[validate]: 4.646e-05
[backend_pass]: 1.07e-06
[task_emit]: 0.00869717
[execute]: 1.093e-05
Sums
    bootstrap                            :     0.000524s :  2.53%
    type_inference                       :     0.006441s : 31.04%
    event_method                         :     0.000024s :  0.12%
    auto_monad                           :     0.000076s :  0.37%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000052s :  0.25%
    rewriter_before_opt_a                :     0.000085s :  0.41%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000085s :  0.41%
    jit_opt_a.loop_unroll                :     0.000037s :  0.18%
    jit_opt_a.a_1                        :     0.000780s :  3.76%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.001028s :  4.95%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.13%
    jit_opt_a.cse                        :     0.000044s :  0.21%
    jit_opt_a.replace_applicator         :     0.000027s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.18%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000930s :  4.48%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.08%
    cconv                                :     0.000039s :  0.19%
    loop_unroll                          :     0.000570s :  2.75%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000033s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000008s :  0.04%
    add_recomputation                    :     0.000057s :  0.28%
    cse_after_recomputation.cse          :     0.000011s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.04%
    opt_after_jit_grad                   :     0.000567s :  2.74%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008697s : 41.92%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000305    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.42% :     0.000001s :      2: substitution.fold_const_symbol
 2.30% :     0.000007s :      3: substitution.graph_param_transform
85.45% :     0.000261s :      4: substitution.inline
 1.52% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.10% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.50% :     0.000008s :      2: substitution.replace_old_param
 4.90% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006334     2
74.57% :     0.004723s :      1: type_inference.infer
25.43% :     0.001611s :      1: type_inference.specialize
------[replace.]   0.000091     6
64.32% :     0.000058s :      4: replace.inline
35.68% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000272     6
94.97% :     0.000258s :      4: match.inline
 5.03% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   691
 1.19% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.74% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.91% :     0.000001s :     10: predicate.addn_check_dump
 1.16% :     0.000002s :     10: predicate.addn_zero_filter
 3.52% :     0.000005s :     10: predicate.arithmetic_simplify
 1.26% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.88% :     0.000001s :     10: predicate.compare_switch_simplify
 1.13% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.52% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.52% :     0.000001s :      3: predicate.elim_not_effective
 0.80% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.00% :     0.000001s :     10: predicate.environ_get_eliminate
 1.38% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.14% :     0.000002s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 6.13% :     0.000008s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :      6: predicate.less_batch_normalization
 1.29% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.60% :     0.000002s :     15: predicate.load_eliminater
 1.58% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.42% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.79% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 2.37% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 1.72% :     0.000002s :     15: predicate.partial_eliminate
 0.93% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.35% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.43% :     0.000001s :      3: predicate.reset_defer_inline
 1.27% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.14% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.11% :     0.000001s :     10: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 0.86% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 1.72% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.01% :     0.000011s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.06% :     0.000001s :     10: predicate.transpose_eliminate
 1.11% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.80% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.91% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.39% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.58% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.02% :     0.000003s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.96% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000818    12
45.72% :     0.000374s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.64% :     0.000062s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.64% :     0.000382s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028641    76
 0.21% :     0.000061s :      1: add_recomputation
 0.28% :     0.000081s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.91% :     0.000547s :      1: bootstrap
 0.15% :     0.000042s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000011s :      1: environ_conv
 0.10% :     0.000030s :      1: event_method
 0.06% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
26.45% :     0.007576s :      1: jit_opt_a
 0.65% :     0.000187s :      1: jit_opt_after_cconv
 0.23% :     0.000067s :      1: jit_opt_b
 2.03% :     0.000580s :      1: loop_unroll
 3.31% :     0.000947s :      1: mutable_eliminate
 3.63% :     0.001039s :     26: opt.transform.jit_opt_a
 0.20% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000030s :      4: opt.transform.jit_opt_b
 0.05% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.01% :     0.000577s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000056s :      1: py_interpret_to_execute
 0.07% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 1.94% :     0.000556s :      1: renormalize.infer
 1.61% :     0.000461s :      1: renormalize.specialize
 0.04% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000042s :      1: rewriter_after_opt_a
 0.31% :     0.000089s :      1: rewriter_before_opt_a
 0.31% :     0.000088s :      1: symbol_engine_optimizer
30.44% :     0.008720s :      1: task_emit
22.57% :     0.006464s :      1: type_inference
 0.25% :     0.000071s :      1: validate


TotalTime = 0.0260269, [33]
[bootstrap]: 0.00042247
[type_inference]: 0.00671628
[event_method]: 2.711e-05
[auto_monad]: 7.646e-05
[graph_reusing]: 6.73e-06
[pre_auto_parallel]: 2.88998e-06
[py_interpret_to_execute]: 5.5e-05
[rewriter_before_opt_a]: 8.717e-05
[expand_dump_flag]: 3.31001e-06
[jit_opt_a]: 0.00792246, [2]
    [Cycle 1]: 0.00228046, [27]
        [switch_simplify]: 7.237e-05
        [loop_unroll]: 2.944e-05
        [a_1]: 0.00066249
        [with_stream_mark]: 2.723e-05
        [recompute_prepare]: 8.82e-06
        [updatestate_depend_eliminate]: 4.04002e-06
        [updatestate_assign_eliminate]: 2.71e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 5.76e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 5.82001e-06
        [meta_shard_fg_expand]: 2.20002e-06
        [get_grad_eliminate_]: 5.91e-06
        [merge_forward]: 3.81999e-06
        [cell_reuse_recompute_pass]: 1.50999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.509e-05
        [j_node_and_user_rematch]: 9.27001e-06
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 1.039e-05
        [inline_without_move]: 5.21002e-06
        [renormalize]: 0.001099
        [add_forward_monad_depend]: 1.031e-05
        [auto_monad_grad]: 3.29001e-06
        [auto_monad_eliminator]: 2.02e-05
        [cse]: 3.386e-05
        [replace_applicator]: 2.105e-05
    [Cycle 2]: 0.00037048, [27]
        [switch_simplify]: 6.61e-06
        [loop_unroll]: 5.29e-06
        [a_1]: 0.00010185
        [with_stream_mark]: 1.39e-05
        [recompute_prepare]: 5.79e-06
        [updatestate_depend_eliminate]: 2.83e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.88002e-06
        [specialize_transform]: 5.29998e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 5.29e-06
        [meta_shard_fg_expand]: 2.65002e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 3.28e-06
        [cell_reuse_recompute_pass]: 3.93001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.637e-05
        [j_node_and_user_rematch]: 8.34002e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 9.19e-06
        [inline_without_move]: 5.07e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 6.29001e-06
        [cse]: 1.317e-05
        [replace_applicator]: 5.33002e-06
[py_interpret_to_execute_after_opt_a]: 1.492e-05
[rewriter_after_opt_a]: 3.924e-05
[convert_after_rewriter]: 7.56001e-06
[order_py_execute_after_rewriter]: 4.82e-06
[mutable_eliminate]: 0.00087876
[jit_opt_b]: 5.86e-05, [1]
    [Cycle 1]: 4.96e-05, [2]
        [frontend_op_eliminate]: 2.102e-05
        [inline_after_opt_a]: 1.611e-05
[cconv]: 2.84e-05
[loop_unroll]: 0.00050254
[jit_opt_after_cconv]: 0.00016118, [1]
    [Cycle 1]: 0.00015351, [11]
        [c_1]: 2.177e-05
        [parameter_eliminate]: 3.61001e-06
        [updatestate_depend_eliminate]: 7.21001e-06
        [updatestate_assign_eliminate]: 3.14001e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [cse]: 2.834e-05
        [call_graph_tuple_transform]: 2.23e-05
        [tuple_list_get_item_eliminator]: 5.07e-06
        [none_parameter_eliminate]: 1.34e-06
        [renormalize]: 8.59989e-07
        [switch_simplify]: 5.49998e-06
[remove_dup_value]: 1.534e-05
[partial_unused_args_eliminate]: 2.53e-06
[environ_conv]: 5.27001e-06
[add_recomputation]: 5.322e-05
[cse_after_recomputation]: 2.224e-05, [1]
    [Cycle 1]: 1.656e-05, [1]
        [cse]: 1.03e-05
[auto_monad_reorder]: 1.623e-05
[get_jit_bprop_graph]: 2.22001e-06
[rewriter_after_jit_bprop_graph]: 7.78999e-06
[opt_after_jit_grad]: 0.00052193
[symbol_engine_optimizer]: 8.207e-05, [1]
    [Cycle 1]: 7.243e-05, [6]
        [build]: 5.32999e-06
        [elim_shapecalc]: 7.83001e-06
        [elim_not_effective]: 1.357e-05
        [opt_reshape]: 6.27001e-06
        [fold_const_symbol]: 9.39e-06
        [renormalize]: 5.89993e-07
[validate]: 3.972e-05
[backend_pass]: 1.15001e-06
[task_emit]: 0.00798139
[execute]: 7.92e-06
Sums
    bootstrap                            :     0.000422s :  2.11%
    type_inference                       :     0.006716s : 33.53%
    event_method                         :     0.000027s :  0.14%
    auto_monad                           :     0.000076s :  0.38%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000055s :  0.27%
    rewriter_before_opt_a                :     0.000087s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000079s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.17%
    jit_opt_a.a_1                        :     0.000764s :  3.82%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.001099s :  5.49%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.13%
    jit_opt_a.cse                        :     0.000047s :  0.23%
    jit_opt_a.replace_applicator         :     0.000026s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.07%
    rewriter_after_opt_a                 :     0.000039s :  0.20%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000879s :  4.39%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.000028s :  0.14%
    loop_unroll                          :     0.000503s :  2.51%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000053s :  0.27%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000522s :  2.61%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007981s : 39.84%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000303    23
 0.59% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000002s :      2: substitution.fold_const_symbol
 2.11% :     0.000006s :      3: substitution.graph_param_transform
86.92% :     0.000264s :      4: substitution.inline
 1.35% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.68% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.13% :     0.000006s :      2: substitution.replace_old_param
 4.60% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006608     2
79.84% :     0.005275s :      1: type_inference.infer
20.16% :     0.001332s :      1: type_inference.specialize
------[replace.]   0.000087     6
70.18% :     0.000061s :      4: replace.inline
29.82% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000274     6
95.33% :     0.000261s :      4: match.inline
 4.67% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.58% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.15% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 2.68% :     0.000003s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.36% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.14% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 6.23% :     0.000008s :     22: predicate.inline
 1.17% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000002s :      6: predicate.less_batch_normalization
 1.89% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.82% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.31% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 2.39% :     0.000003s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 1.93% :     0.000002s :     15: predicate.partial_eliminate
 1.03% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.82% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.08% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.09% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.37% :     0.000009s :     53: predicate.switch_simplify
 1.34% :     0.000002s :     10: predicate.tile_eliminate
 1.34% :     0.000002s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.20% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.67% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000932    12
43.20% :     0.000403s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.60% :     0.000062s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.20% :     0.000468s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028240    76
 0.20% :     0.000057s :      1: add_recomputation
 0.29% :     0.000082s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.58% :     0.000447s :      1: bootstrap
 0.11% :     0.000031s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.12% :     0.000033s :      1: event_method
 0.04% :     0.000011s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
28.07% :     0.007927s :      1: jit_opt_a
 0.58% :     0.000164s :      1: jit_opt_after_cconv
 0.22% :     0.000062s :      1: jit_opt_b
 1.81% :     0.000511s :      1: loop_unroll
 3.16% :     0.000891s :      1: mutable_eliminate
 3.57% :     0.001007s :     26: opt.transform.jit_opt_a
 0.18% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000030s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.88% :     0.000532s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000059s :      1: py_interpret_to_execute
 0.06% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000018s :      1: remove_dup_value
 1.95% :     0.000551s :      1: renormalize.infer
 1.90% :     0.000537s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000043s :      1: rewriter_after_opt_a
 0.32% :     0.000091s :      1: rewriter_before_opt_a
 0.30% :     0.000085s :      1: symbol_engine_optimizer
28.31% :     0.007995s :      1: task_emit
23.88% :     0.006744s :      1: type_inference
 0.22% :     0.000062s :      1: validate


TotalTime = 0.0396251, [33]
[bootstrap]: 0.00035222
[type_inference]: 0.0220519
[event_method]: 1.897e-05
[auto_monad]: 6.776e-05
[graph_reusing]: 5.85002e-06
[pre_auto_parallel]: 2.53e-06
[py_interpret_to_execute]: 4.768e-05
[rewriter_before_opt_a]: 7.723e-05
[expand_dump_flag]: 2.96999e-06
[jit_opt_a]: 0.00657257, [2]
    [Cycle 1]: 0.00205721, [27]
        [switch_simplify]: 7.008e-05
        [loop_unroll]: 3.191e-05
        [a_1]: 0.00064627
        [with_stream_mark]: 2.342e-05
        [recompute_prepare]: 9.32999e-06
        [updatestate_depend_eliminate]: 4.03001e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 6.00002e-06
        [updatestate_useless_node_eliminater]: 5.59e-06
        [accelerated_algorithm]: 5.77001e-06
        [meta_shard_fg_expand]: 2.26998e-06
        [get_grad_eliminate_]: 5.49e-06
        [merge_forward]: 3.45e-06
        [cell_reuse_recompute_pass]: 1.45001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.498e-05
        [j_node_and_user_rematch]: 1.047e-05
        [meta_fg_expand]: 2.28998e-06
        [replace_old_param]: 9.14e-06
        [inline_without_move]: 5.72999e-06
        [renormalize]: 0.00093732
        [add_forward_monad_depend]: 6.71e-06
        [auto_monad_grad]: 2.25002e-06
        [auto_monad_eliminator]: 1.42e-05
        [cse]: 2.252e-05
        [replace_applicator]: 1.684e-05
    [Cycle 2]: 0.00034201, [27]
        [switch_simplify]: 6.11e-06
        [loop_unroll]: 5.78002e-06
        [a_1]: 0.00010244
        [with_stream_mark]: 1.179e-05
        [recompute_prepare]: 5.27001e-06
        [updatestate_depend_eliminate]: 2.96001e-06
        [updatestate_assign_eliminate]: 2.14999e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 4.68999e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 6.01e-06
        [meta_shard_fg_expand]: 1.49998e-06
        [get_grad_eliminate_]: 4.76002e-06
        [merge_forward]: 3.18e-06
        [cell_reuse_recompute_pass]: 1.71e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.539e-05
        [j_node_and_user_rematch]: 8.69e-06
        [meta_fg_expand]: 1.55999e-06
        [replace_old_param]: 7.76001e-06
        [inline_without_move]: 4.89e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.66998e-06
        [auto_monad_grad]: 1.19998e-06
        [auto_monad_eliminator]: 5.69999e-06
        [cse]: 1.04e-05
        [replace_applicator]: 5.79999e-06
[py_interpret_to_execute_after_opt_a]: 1.327e-05
[rewriter_after_opt_a]: 3.816e-05
[convert_after_rewriter]: 6.54999e-06
[order_py_execute_after_rewriter]: 4.98001e-06
[mutable_eliminate]: 0.00087825
[jit_opt_b]: 5.711e-05, [1]
    [Cycle 1]: 4.835e-05, [2]
        [frontend_op_eliminate]: 1.917e-05
        [inline_after_opt_a]: 1.672e-05
[cconv]: 3.209e-05
[loop_unroll]: 0.00052747
[jit_opt_after_cconv]: 0.00017256, [1]
    [Cycle 1]: 0.00016436, [11]
        [c_1]: 2.378e-05
        [parameter_eliminate]: 3.5e-06
        [updatestate_depend_eliminate]: 9.27001e-06
        [updatestate_assign_eliminate]: 4.1e-06
        [updatestate_loads_eliminate]: 2.83998e-06
        [cse]: 2.882e-05
        [call_graph_tuple_transform]: 2.421e-05
        [tuple_list_get_item_eliminator]: 5.67999e-06
        [none_parameter_eliminate]: 1.74998e-06
        [renormalize]: 6.10016e-07
        [switch_simplify]: 5.68002e-06
[remove_dup_value]: 1.499e-05
[partial_unused_args_eliminate]: 2.91999e-06
[environ_conv]: 5.74e-06
[add_recomputation]: 5.251e-05
[cse_after_recomputation]: 2.382e-05, [1]
    [Cycle 1]: 1.711e-05, [1]
        [cse]: 1.122e-05
[auto_monad_reorder]: 1.569e-05
[get_jit_bprop_graph]: 2.70002e-06
[rewriter_after_jit_bprop_graph]: 9.72999e-06
[opt_after_jit_grad]: 0.00055829
[symbol_engine_optimizer]: 8.462e-05, [1]
    [Cycle 1]: 7.708e-05, [6]
        [build]: 5.25001e-06
        [elim_shapecalc]: 8.99e-06
        [elim_not_effective]: 1.575e-05
        [opt_reshape]: 6.93998e-06
        [fold_const_symbol]: 9.51e-06
        [renormalize]: 2.80008e-07
[validate]: 4.51e-05
[backend_pass]: 1.17999e-06
[task_emit]: 0.00762837
[execute]: 9.54999e-06
Sums
    bootstrap                            :     0.000352s :  1.01%
    type_inference                       :     0.022052s : 63.42%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000068s :  0.19%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000048s :  0.14%
    rewriter_before_opt_a                :     0.000077s :  0.22%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000076s :  0.22%
    jit_opt_a.loop_unroll                :     0.000038s :  0.11%
    jit_opt_a.a_1                        :     0.000749s :  2.15%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.06%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000017s :  0.05%
    jit_opt_a.inline_without_move        :     0.000011s :  0.03%
    jit_opt_a.renormalize                :     0.000937s :  2.70%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.06%
    jit_opt_a.cse                        :     0.000033s :  0.09%
    jit_opt_a.replace_applicator         :     0.000023s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.04%
    rewriter_after_opt_a                 :     0.000038s :  0.11%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000878s :  2.53%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.05%
    cconv                                :     0.000032s :  0.09%
    loop_unroll                          :     0.000527s :  1.52%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000015s :  0.04%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000053s :  0.15%
    cse_after_recomputation.cse          :     0.000011s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.03%
    opt_after_jit_grad                   :     0.000558s :  1.61%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000045s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007628s : 21.94%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.000281    23
 1.03% :     0.000003s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.64% :     0.000007s :      3: substitution.graph_param_transform
85.85% :     0.000241s :      4: substitution.inline
 1.32% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.03% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.56% :     0.000004s :      2: substitution.replace_old_param
 5.10% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.021972     2
95.07% :     0.020889s :      1: type_inference.infer
 4.93% :     0.001083s :      1: type_inference.specialize
------[replace.]   0.000086     6
65.48% :     0.000056s :      4: replace.inline
34.52% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000252     6
94.76% :     0.000239s :      4: match.inline
 5.24% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   691
 1.73% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.50% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.92% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 2.25% :     0.000003s :     10: predicate.arithmetic_simplify
 1.49% :     0.000002s :     10: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.75% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.31% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000002s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 6.23% :     0.000008s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 2.06% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.78% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.17% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.24% :     0.000002s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 1.94% :     0.000003s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.01% :     0.000003s :     15: predicate.partial_eliminate
 1.04% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.31% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.82% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.65% :     0.000002s :     10: predicate.reshape_eliminate
 1.03% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.27% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.69% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.07% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 1.84% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.39% :     0.000010s :     53: predicate.switch_simplify
 1.34% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.95% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.12% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.64% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000757    12
48.39% :     0.000366s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.04% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.57% :     0.000337s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.041666    76
 0.13% :     0.000056s :      1: add_recomputation
 0.17% :     0.000071s :      1: auto_monad
 0.04% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.90% :     0.000374s :      1: bootstrap
 0.08% :     0.000035s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.06% :     0.000026s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.05% :     0.000022s :      1: event_method
 0.04% :     0.000015s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
15.79% :     0.006578s :      1: jit_opt_a
 0.42% :     0.000176s :      1: jit_opt_after_cconv
 0.14% :     0.000060s :      1: jit_opt_b
 1.29% :     0.000537s :      1: loop_unroll
 2.13% :     0.000889s :      1: mutable_eliminate
 2.37% :     0.000989s :     26: opt.transform.jit_opt_a
 0.13% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000029s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.36% :     0.000567s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000050s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000018s :      1: remove_dup_value
 1.25% :     0.000522s :      1: renormalize.infer
 0.97% :     0.000406s :      1: renormalize.specialize
 0.03% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000041s :      1: rewriter_after_opt_a
 0.19% :     0.000080s :      1: rewriter_before_opt_a
 0.21% :     0.000087s :      1: symbol_engine_optimizer
18.36% :     0.007651s :      1: task_emit
52.96% :     0.022065s :      1: type_inference
 0.17% :     0.000069s :      1: validate


TotalTime = 0.0261076, [33]
[bootstrap]: 0.00053205
[type_inference]: 0.0059391
[event_method]: 1.84e-05
[auto_monad]: 4.504e-05
[graph_reusing]: 4.06001e-06
[pre_auto_parallel]: 2.07999e-06
[py_interpret_to_execute]: 4.316e-05
[rewriter_before_opt_a]: 6.977e-05
[expand_dump_flag]: 2.31998e-06
[jit_opt_a]: 0.00713956, [2]
    [Cycle 1]: 0.00235263, [27]
        [switch_simplify]: 7.548e-05
        [loop_unroll]: 3.141e-05
        [a_1]: 0.00067945
        [with_stream_mark]: 2.735e-05
        [recompute_prepare]: 1.278e-05
        [updatestate_depend_eliminate]: 4.92e-06
        [updatestate_assign_eliminate]: 4.08999e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 2.36e-06
        [specialize_transform]: 8e-06
        [updatestate_useless_node_eliminater]: 6.88e-06
        [accelerated_algorithm]: 6.67002e-06
        [meta_shard_fg_expand]: 3.66001e-06
        [get_grad_eliminate_]: 5.82999e-06
        [merge_forward]: 4.65999e-06
        [cell_reuse_recompute_pass]: 2.44999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.883e-05
        [j_node_and_user_rematch]: 1.185e-05
        [meta_fg_expand]: 2.36998e-06
        [replace_old_param]: 1.169e-05
        [inline_without_move]: 6.64999e-06
        [renormalize]: 0.00110272
        [add_forward_monad_depend]: 1.045e-05
        [auto_monad_grad]: 3.08998e-06
        [auto_monad_eliminator]: 2.021e-05
        [cse]: 3.397e-05
        [replace_applicator]: 2.31e-05
    [Cycle 2]: 0.00038821, [27]
        [switch_simplify]: 6.67002e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 0.00010151
        [with_stream_mark]: 1.796e-05
        [recompute_prepare]: 6.01e-06
        [updatestate_depend_eliminate]: 3.19001e-06
        [updatestate_assign_eliminate]: 3.58e-06
        [updatestate_loads_eliminate]: 2.01003e-06
        [parameter_eliminate]: 1.67001e-06
        [specialize_transform]: 5.66e-06
        [updatestate_useless_node_eliminater]: 5.34998e-06
        [accelerated_algorithm]: 5.82001e-06
        [meta_shard_fg_expand]: 1.96e-06
        [get_grad_eliminate_]: 5.20001e-06
        [merge_forward]: 3.34001e-06
        [cell_reuse_recompute_pass]: 3.74002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.665e-05
        [j_node_and_user_rematch]: 1.033e-05
        [meta_fg_expand]: 1.89999e-06
        [replace_old_param]: 1.034e-05
        [inline_without_move]: 1.085e-05
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 2.46e-06
        [auto_monad_grad]: 1.62999e-06
        [auto_monad_eliminator]: 7.75e-06
        [cse]: 1.369e-05
        [replace_applicator]: 6.53e-06
[py_interpret_to_execute_after_opt_a]: 1.596e-05
[rewriter_after_opt_a]: 4.349e-05
[convert_after_rewriter]: 6.97002e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.00174995
[jit_opt_b]: 7.321e-05, [1]
    [Cycle 1]: 6.163e-05, [2]
        [frontend_op_eliminate]: 2.424e-05
        [inline_after_opt_a]: 2.213e-05
[cconv]: 4.278e-05
[loop_unroll]: 0.00060372
[jit_opt_after_cconv]: 0.00020056, [1]
    [Cycle 1]: 0.00019195, [11]
        [c_1]: 2.555e-05
        [parameter_eliminate]: 6.59001e-06
        [updatestate_depend_eliminate]: 1.233e-05
        [updatestate_assign_eliminate]: 2.90002e-06
        [updatestate_loads_eliminate]: 2.91e-06
        [cse]: 3.995e-05
        [call_graph_tuple_transform]: 2.967e-05
        [tuple_list_get_item_eliminator]: 5.85002e-06
        [none_parameter_eliminate]: 1.88002e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 6.50997e-06
[remove_dup_value]: 1.819e-05
[partial_unused_args_eliminate]: 2.99999e-06
[environ_conv]: 7.58001e-06
[add_recomputation]: 7.665e-05
[cse_after_recomputation]: 3.216e-05, [1]
    [Cycle 1]: 2.388e-05, [1]
        [cse]: 1.52e-05
[auto_monad_reorder]: 1.684e-05
[get_jit_bprop_graph]: 2.66e-06
[rewriter_after_jit_bprop_graph]: 1.34e-05
[opt_after_jit_grad]: 0.00077201
[symbol_engine_optimizer]: 0.00010648, [1]
    [Cycle 1]: 9.452e-05, [6]
        [build]: 9.19e-06
        [elim_shapecalc]: 1.048e-05
        [elim_not_effective]: 2.155e-05
        [opt_reshape]: 8.00999e-06
        [fold_const_symbol]: 1.152e-05
        [renormalize]: 1.40999e-06
[validate]: 5.749e-05
[backend_pass]: 1.67001e-06
[task_emit]: 0.00813002
[execute]: 8.57998e-06
Sums
    bootstrap                            :     0.000532s :  2.55%
    type_inference                       :     0.005939s : 28.45%
    event_method                         :     0.000018s :  0.09%
    auto_monad                           :     0.000045s :  0.22%
    graph_reusing                        :     0.000004s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.21%
    rewriter_before_opt_a                :     0.000070s :  0.33%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000082s :  0.39%
    jit_opt_a.loop_unroll                :     0.000036s :  0.17%
    jit_opt_a.a_1                        :     0.000781s :  3.74%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000014s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000022s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.11%
    jit_opt_a.inline_without_move        :     0.000017s :  0.08%
    jit_opt_a.renormalize                :     0.001103s :  5.28%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.13%
    jit_opt_a.cse                        :     0.000048s :  0.23%
    jit_opt_a.replace_applicator         :     0.000030s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.08%
    rewriter_after_opt_a                 :     0.000043s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.001750s :  8.38%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000022s :  0.11%
    cconv                                :     0.000043s :  0.20%
    loop_unroll                          :     0.000604s :  2.89%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000040s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000030s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000018s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000008s :  0.04%
    add_recomputation                    :     0.000077s :  0.37%
    cse_after_recomputation.cse          :     0.000015s :  0.07%
    auto_monad_reorder                   :     0.000017s :  0.08%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000013s :  0.06%
    opt_after_jit_grad                   :     0.000772s :  3.70%
    symbol_engine_optimizer.build        :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000022s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000057s :  0.28%
    backend_pass                         :     0.000002s :  0.01%
    task_emit                            :     0.008130s : 38.95%
    execute                              :     0.000009s :  0.04%


Time group info:
------[substitution.]   0.000295    23
 0.73% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000002s :      2: substitution.fold_const_symbol
 2.37% :     0.000007s :      3: substitution.graph_param_transform
84.84% :     0.000250s :      4: substitution.inline
 1.63% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.15% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.47% :     0.000007s :      2: substitution.replace_old_param
 5.26% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005865     2
82.40% :     0.004833s :      1: type_inference.infer
17.60% :     0.001032s :      1: type_inference.specialize
------[replace.]   0.000092     6
63.30% :     0.000059s :      4: replace.inline
36.70% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000261     6
94.46% :     0.000247s :      4: match.inline
 5.54% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000148   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.37% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.85% :     0.000001s :     10: predicate.addn_check_dump
 1.17% :     0.000002s :     10: predicate.addn_zero_filter
 2.49% :     0.000004s :     10: predicate.arithmetic_simplify
 1.10% :     0.000002s :     10: predicate.cast_eliminate
 0.56% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.86% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000002s :     10: predicate.depend_value_elim
 0.90% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.53% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000001s :      3: predicate.elim_not_effective
 0.81% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.87% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.86% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.92% :     0.000001s :     10: predicate.environ_get_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.33% :     0.000000s :      3: predicate.graph_param_transform
 6.20% :     0.000009s :     22: predicate.inline
 0.77% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.50% :     0.000002s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000003s :     15: predicate.load_eliminater
 2.18% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.90% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.41% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 0.90% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000002s :     10: predicate.minmaximum_grad
 3.25% :     0.000005s :      3: predicate.mutable_eliminate
 0.34% :     0.000001s :      3: predicate.opt_reshape
 1.69% :     0.000002s :     15: predicate.partial_eliminate
 1.08% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.21% :     0.000003s :     10: predicate.reduce_eliminate
 1.36% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.52% :     0.000001s :      3: predicate.reset_defer_inline
 1.23% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.15% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000002s :     10: predicate.same_eliminate
 0.60% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.16% :     0.000002s :      6: predicate.special_op_eliminate
 1.14% :     0.000002s :      6: predicate.specialize_transform
 1.08% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.30% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.71% :     0.000004s :     16: predicate.switch_defer_inline
 1.82% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.63% :     0.000011s :     53: predicate.switch_simplify
 1.01% :     0.000001s :     10: predicate.tile_eliminate
 1.04% :     0.000002s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.87% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.18% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.97% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.32% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.28% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.26% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000831    12
42.22% :     0.000351s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.69% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.09% :     0.000424s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028437    76
 0.28% :     0.000081s :      1: add_recomputation
 0.17% :     0.000048s :      1: auto_monad
 0.07% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 1.93% :     0.000549s :      1: bootstrap
 0.16% :     0.000046s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000035s :      1: cse_after_recomputation
 0.04% :     0.000012s :      1: environ_conv
 0.08% :     0.000022s :      1: event_method
 0.04% :     0.000013s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000006s :      1: graph_reusing
25.12% :     0.007144s :      1: jit_opt_a
 0.72% :     0.000203s :      1: jit_opt_after_cconv
 0.27% :     0.000077s :      1: jit_opt_b
 2.18% :     0.000620s :      1: loop_unroll
 6.23% :     0.001773s :      1: mutable_eliminate
 3.71% :     0.001054s :     26: opt.transform.jit_opt_a
 0.22% :     0.000064s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000037s :      4: opt.transform.jit_opt_b
 0.07% :     0.000020s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000032s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000039s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000047s :      4: opt.transform.symbol_engine_opt
 2.79% :     0.000795s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000006s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000046s :      1: py_interpret_to_execute
 0.06% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.24% :     0.000068s :      1: remove_dup_value
 1.99% :     0.000565s :      1: renormalize.infer
 1.85% :     0.000525s :      1: renormalize.specialize
 0.05% :     0.000016s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000048s :      1: rewriter_after_opt_a
 0.26% :     0.000073s :      1: rewriter_before_opt_a
 0.39% :     0.000110s :      1: symbol_engine_optimizer
28.65% :     0.008146s :      1: task_emit
20.93% :     0.005953s :      1: type_inference
 0.32% :     0.000090s :      1: validate


TotalTime = 0.0368746, [30]
[bootstrap]: 0.00035374
[type_inference]: 0.00567547
[event_method]: 2.076e-05
[auto_monad]: 6.84e-05
[graph_reusing]: 6.46999e-06
[pre_auto_parallel]: 2.60002e-06
[py_interpret_to_execute]: 5.05e-05
[rewriter_before_opt_a]: 7.927e-05
[expand_dump_flag]: 3.31001e-06
[jit_opt_a]: 0.0276254, [2]
    [Cycle 1]: 0.00235869, [27]
        [switch_simplify]: 7.567e-05
        [loop_unroll]: 3.075e-05
        [a_1]: 0.00066744
        [with_stream_mark]: 2.743e-05
        [recompute_prepare]: 9.90002e-06
        [updatestate_depend_eliminate]: 4.65999e-06
        [updatestate_assign_eliminate]: 3.13e-06
        [updatestate_loads_eliminate]: 2.96001e-06
        [parameter_eliminate]: 2.23998e-06
        [specialize_transform]: 7.28e-06
        [updatestate_useless_node_eliminater]: 5.52001e-06
        [accelerated_algorithm]: 5.91e-06
        [meta_shard_fg_expand]: 2.83e-06
        [get_grad_eliminate_]: 6.06998e-06
        [merge_forward]: 3.48e-06
        [cell_reuse_recompute_pass]: 1.74998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.724e-05
        [j_node_and_user_rematch]: 1.068e-05
        [meta_fg_expand]: 2.47001e-06
        [replace_old_param]: 9.66998e-06
        [inline_without_move]: 5.71e-06
        [renormalize]: 0.0011357
        [add_forward_monad_depend]: 1.054e-05
        [auto_monad_grad]: 3.51999e-06
        [auto_monad_eliminator]: 2.188e-05
        [cse]: 3.352e-05
        [replace_applicator]: 2.33e-05
    [Cycle 2]: 0.00039156, [27]
        [switch_simplify]: 5.94999e-06
        [loop_unroll]: 5.44998e-06
        [a_1]: 0.00010209
        [with_stream_mark]: 1.721e-05
        [recompute_prepare]: 6.82002e-06
        [updatestate_depend_eliminate]: 3.85e-06
        [updatestate_assign_eliminate]: 2.59001e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [parameter_eliminate]: 1.94999e-06
        [specialize_transform]: 6.28998e-06
        [updatestate_useless_node_eliminater]: 5.13002e-06
        [accelerated_algorithm]: 5.69999e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 4.77998e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 3.48999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.886e-05
        [j_node_and_user_rematch]: 9.17999e-06
        [meta_fg_expand]: 1.66002e-06
        [replace_old_param]: 1.134e-05
        [inline_without_move]: 5.40999e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.68002e-06
        [auto_monad_grad]: 1.14e-06
        [auto_monad_eliminator]: 6.61999e-06
        [cse]: 1.424e-05
        [replace_applicator]: 7.77e-06
[py_interpret_to_execute_after_opt_a]: 1.945e-05
[rewriter_after_opt_a]: 4.31e-05
[convert_after_rewriter]: 8.56002e-06
[order_py_execute_after_rewriter]: 5.05999e-06
[mutable_eliminate]: 0.0008943
[jit_opt_b]: 6.669e-05, [1]
    [Cycle 1]: 5.566e-05, [2]
        [frontend_op_eliminate]: 2.167e-05
        [inline_after_opt_a]: 1.88e-05
[cconv]: 3.916e-05
[loop_unroll]: 0.00055388
[jit_opt_after_cconv]: 0.00018767, [1]
    [Cycle 1]: 0.00017859, [11]
        [c_1]: 2.398e-05
        [parameter_eliminate]: 5.67001e-06
        [updatestate_depend_eliminate]: 1.078e-05
        [updatestate_assign_eliminate]: 3.41999e-06
        [updatestate_loads_eliminate]: 3.24001e-06
        [cse]: 3.304e-05
        [call_graph_tuple_transform]: 2.663e-05
        [tuple_list_get_item_eliminator]: 5.85002e-06
        [none_parameter_eliminate]: 2.30002e-06
        [renormalize]: 9.49978e-07
        [switch_simplify]: 6.14999e-06
[remove_dup_value]: 1.731e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 6.79999e-06
[add_recomputation]: 5.866e-05
[cse_after_recomputation]: 2.407e-05, [1]
    [Cycle 1]: 1.692e-05, [1]
        [cse]: 9.72999e-06
[auto_monad_reorder]: 1.656e-05
[get_jit_bprop_graph]: 2.34001e-06
[rewriter_after_jit_bprop_graph]: 1.169e-05
[opt_after_jit_grad]: 0.00059587
[symbol_engine_optimizer]: 0.00013556, [1]
    [Cycle 1]: 0.00012586, [6]
        [build]: 5.72999e-06
        [elim_shapecalc]: 9.91998e-06
        [elim_not_effective]: 1.7e-05
        [opt_reshape]: 6.35997e-06
        [fold_const_symbol]: 5.542e-05
        [renormalize]: 5.39992e-07
[validate]: 5.376e-05
Sums
    bootstrap                            :     0.000354s :  3.15%
    type_inference                       :     0.005675s : 50.47%
    event_method                         :     0.000021s :  0.18%
    auto_monad                           :     0.000068s :  0.61%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000050s :  0.45%
    rewriter_before_opt_a                :     0.000079s :  0.70%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000082s :  0.73%
    jit_opt_a.loop_unroll                :     0.000036s :  0.32%
    jit_opt_a.a_1                        :     0.000770s :  6.84%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.40%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000014s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000021s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.001136s : 10.10%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.11%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.25%
    jit_opt_a.cse                        :     0.000048s :  0.42%
    jit_opt_a.replace_applicator         :     0.000031s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.17%
    rewriter_after_opt_a                 :     0.000043s :  0.38%
    convert_after_rewriter               :     0.000009s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000894s :  7.95%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.17%
    cconv                                :     0.000039s :  0.35%
    loop_unroll                          :     0.000554s :  4.93%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.10%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000033s :  0.29%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000017s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000007s :  0.06%
    add_recomputation                    :     0.000059s :  0.52%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000017s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000012s :  0.10%
    opt_after_jit_grad                   :     0.000596s :  5.30%
    symbol_engine_optimizer.build        :     0.000006s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000055s :  0.49%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000054s :  0.48%


Time group info:
------[substitution.]   0.000299    23
 0.69% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000002s :      2: substitution.fold_const_symbol
 2.08% :     0.000006s :      3: substitution.graph_param_transform
85.72% :     0.000256s :      4: substitution.inline
 1.81% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.99% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.55% :     0.000008s :      2: substitution.replace_old_param
 4.47% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005586     2
80.25% :     0.004482s :      1: type_inference.infer
19.75% :     0.001103s :      1: type_inference.specialize
------[replace.]   0.000086     6
67.93% :     0.000058s :      4: replace.inline
32.07% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000266     6
95.41% :     0.000254s :      4: match.inline
 4.59% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000140   691
 1.08% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.63% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.89% :     0.000001s :     10: predicate.addn_check_dump
 1.19% :     0.000002s :     10: predicate.addn_zero_filter
 2.61% :     0.000004s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.90% :     0.000001s :     10: predicate.compare_switch_simplify
 1.91% :     0.000003s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 0.99% :     0.000001s :     10: predicate.dict_get_item_eliminator
 0.95% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.41% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.89% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     10: predicate.environ_get_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.34% :     0.000000s :      3: predicate.graph_param_transform
 6.56% :     0.000009s :     22: predicate.inline
 0.77% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :      6: predicate.less_batch_normalization
 1.80% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.58% :     0.000002s :     15: predicate.load_eliminater
 2.19% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.69% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 0.92% :     0.000001s :     10: predicate.minmaximum_grad
 2.79% :     0.000004s :      3: predicate.mutable_eliminate
 0.39% :     0.000001s :      3: predicate.opt_reshape
 1.87% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.47% :     0.000001s :      3: predicate.reset_defer_inline
 1.09% :     0.000002s :     10: predicate.reshape_eliminate
 1.03% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000002s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.16% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.08% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.23% :     0.000012s :     53: predicate.switch_simplify
 0.97% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000002s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.29% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.46% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.93% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000834    12
46.44% :     0.000387s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.43% :     0.000062s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.13% :     0.000385s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.039149    72
 0.16% :     0.000062s :      1: add_recomputation
 0.19% :     0.000073s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.96% :     0.000374s :      1: bootstrap
 0.11% :     0.000042s :      1: cconv
 0.03% :     0.000012s :      1: convert_after_rewriter
 0.07% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.07% :     0.000026s :      1: event_method
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
70.58% :     0.027631s :      1: jit_opt_a
 0.49% :     0.000191s :      1: jit_opt_after_cconv
 0.18% :     0.000070s :      1: jit_opt_b
 1.45% :     0.000566s :      1: loop_unroll
 2.33% :     0.000914s :      1: mutable_eliminate
 2.64% :     0.001034s :     26: opt.transform.jit_opt_a
 0.15% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000032s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000027s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000084s :      4: opt.transform.symbol_engine_opt
 1.56% :     0.000611s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.14% :     0.000054s :      1: py_interpret_to_execute
 0.06% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000020s :      1: remove_dup_value
 1.62% :     0.000634s :      1: renormalize.infer
 1.25% :     0.000490s :      1: renormalize.specialize
 0.04% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000047s :      1: rewriter_after_opt_a
 0.21% :     0.000084s :      1: rewriter_before_opt_a
 0.36% :     0.000139s :      1: symbol_engine_optimizer
14.56% :     0.005699s :      1: type_inference


TotalTime = 0.0241219, [33]
[bootstrap]: 0.00042236
[type_inference]: 0.00674753
[event_method]: 1.999e-05
[auto_monad]: 6.73e-05
[graph_reusing]: 6.69001e-06
[pre_auto_parallel]: 2.71999e-06
[py_interpret_to_execute]: 4.977e-05
[rewriter_before_opt_a]: 7.898e-05
[expand_dump_flag]: 3.00002e-06
[jit_opt_a]: 0.00675532, [2]
    [Cycle 1]: 0.00181128, [27]
        [switch_simplify]: 6.174e-05
        [loop_unroll]: 2.948e-05
        [a_1]: 0.00057886
        [with_stream_mark]: 1.986e-05
        [recompute_prepare]: 7.75e-06
        [updatestate_depend_eliminate]: 4.08999e-06
        [updatestate_assign_eliminate]: 2.88e-06
        [updatestate_loads_eliminate]: 2.82002e-06
        [parameter_eliminate]: 1.82999e-06
        [specialize_transform]: 6.14001e-06
        [updatestate_useless_node_eliminater]: 4.79998e-06
        [accelerated_algorithm]: 6.54001e-06
        [meta_shard_fg_expand]: 2.09999e-06
        [get_grad_eliminate_]: 5.60001e-06
        [merge_forward]: 3.56001e-06
        [cell_reuse_recompute_pass]: 1.29998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.457e-05
        [j_node_and_user_rematch]: 9.39998e-06
        [meta_fg_expand]: 2.58e-06
        [replace_old_param]: 9.42001e-06
        [inline_without_move]: 5.52999e-06
        [renormalize]: 0.0007873
        [add_forward_monad_depend]: 6.15002e-06
        [auto_monad_grad]: 2.63998e-06
        [auto_monad_eliminator]: 1.503e-05
        [cse]: 2.828e-05
        [replace_applicator]: 1.715e-05
    [Cycle 2]: 0.00069332, [27]
        [switch_simplify]: 6.38e-06
        [loop_unroll]: 5.00001e-06
        [a_1]: 0.00035428
        [with_stream_mark]: 2.933e-05
        [recompute_prepare]: 7.18e-06
        [updatestate_depend_eliminate]: 3.48e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [parameter_eliminate]: 1.336e-05
        [specialize_transform]: 5.59e-06
        [updatestate_useless_node_eliminater]: 5.42001e-06
        [accelerated_algorithm]: 6.41e-06
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 4.73001e-06
        [merge_forward]: 3.22002e-06
        [cell_reuse_recompute_pass]: 3.23e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.631e-05
        [j_node_and_user_rematch]: 1.62e-05
        [meta_fg_expand]: 1.91998e-06
        [replace_old_param]: 9.62001e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 2.06998e-06
        [auto_monad_grad]: 1.59e-06
        [auto_monad_eliminator]: 8.79e-06
        [cse]: 1.809e-05
        [replace_applicator]: 6.09001e-06
[py_interpret_to_execute_after_opt_a]: 1.471e-05
[rewriter_after_opt_a]: 3.807e-05
[convert_after_rewriter]: 7.35e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00081108
[jit_opt_b]: 5.609e-05, [1]
    [Cycle 1]: 4.823e-05, [2]
        [frontend_op_eliminate]: 1.778e-05
        [inline_after_opt_a]: 1.739e-05
[cconv]: 2.692e-05
[loop_unroll]: 0.00048452
[jit_opt_after_cconv]: 0.00016198, [1]
    [Cycle 1]: 0.0001548, [11]
        [c_1]: 2.256e-05
        [parameter_eliminate]: 3.92998e-06
        [updatestate_depend_eliminate]: 6.08998e-06
        [updatestate_assign_eliminate]: 3.09001e-06
        [updatestate_loads_eliminate]: 3.06999e-06
        [cse]: 2.668e-05
        [call_graph_tuple_transform]: 2.245e-05
        [tuple_list_get_item_eliminator]: 5.66998e-06
        [none_parameter_eliminate]: 1.35001e-06
        [renormalize]: 8.80013e-07
        [switch_simplify]: 5.72999e-06
[remove_dup_value]: 1.459e-05
[partial_unused_args_eliminate]: 2.39001e-06
[environ_conv]: 6.09999e-06
[add_recomputation]: 5.022e-05
[cse_after_recomputation]: 2.347e-05, [1]
    [Cycle 1]: 1.75e-05, [1]
        [cse]: 1.073e-05
[auto_monad_reorder]: 1.452e-05
[get_jit_bprop_graph]: 2.33002e-06
[rewriter_after_jit_bprop_graph]: 7.97e-06
[opt_after_jit_grad]: 0.00055973
[symbol_engine_optimizer]: 8.04e-05, [1]
    [Cycle 1]: 7.353e-05, [6]
        [build]: 4.23001e-06
        [elim_shapecalc]: 7.97e-06
        [elim_not_effective]: 1.592e-05
        [opt_reshape]: 5.94999e-06
        [fold_const_symbol]: 8.95999e-06
        [renormalize]: 6.69999e-07
[validate]: 4.221e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00730538
[execute]: 1.08e-05
Sums
    bootstrap                            :     0.000422s :  2.20%
    type_inference                       :     0.006748s : 35.20%
    event_method                         :     0.000020s :  0.10%
    auto_monad                           :     0.000067s :  0.35%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000050s :  0.26%
    rewriter_before_opt_a                :     0.000079s :  0.41%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.36%
    jit_opt_a.loop_unroll                :     0.000034s :  0.18%
    jit_opt_a.a_1                        :     0.000933s :  4.87%
    jit_opt_a.with_stream_mark           :     0.000049s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000015s :  0.08%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000026s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000787s :  4.11%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.12%
    jit_opt_a.cse                        :     0.000046s :  0.24%
    jit_opt_a.replace_applicator         :     0.000023s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000811s :  4.23%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000027s :  0.14%
    loop_unroll                          :     0.000485s :  2.53%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000027s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000050s :  0.26%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000560s :  2.92%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.22%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007305s : 38.11%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000244    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.52% :     0.000006s :      3: substitution.graph_param_transform
84.42% :     0.000206s :      4: substitution.inline
 1.64% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.20% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.39% :     0.000006s :      2: substitution.replace_old_param
 5.30% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006662     2
84.22% :     0.005610s :      1: type_inference.infer
15.78% :     0.001051s :      1: type_inference.specialize
------[replace.]   0.000074     6
66.68% :     0.000049s :      4: replace.inline
33.32% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000215     6
94.46% :     0.000203s :      4: match.inline
 5.54% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.91% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.76% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.50% :     0.000002s :     10: predicate.addn_zero_filter
 2.96% :     0.000004s :     10: predicate.arithmetic_simplify
 1.06% :     0.000001s :     10: predicate.cast_eliminate
 0.35% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.54% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.35% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     10: predicate.environ_get_eliminate
 1.27% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.35% :     0.000000s :      3: predicate.graph_param_transform
 5.67% :     0.000007s :     22: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.66% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.68% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.25% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.22% :     0.000002s :     10: predicate.minmaximum_grad
 1.78% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 1.91% :     0.000002s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.38% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.10% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.97% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.27% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.86% :     0.000010s :     53: predicate.switch_simplify
 2.34% :     0.000003s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.40% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.53% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000761    12
50.69% :     0.000386s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.17% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.14% :     0.000328s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026197    76
 0.20% :     0.000053s :      1: add_recomputation
 0.27% :     0.000070s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.70% :     0.000446s :      1: bootstrap
 0.11% :     0.000029s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000024s :      1: event_method
 0.06% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
25.80% :     0.006760s :      1: jit_opt_a
 0.63% :     0.000165s :      1: jit_opt_after_cconv
 0.23% :     0.000059s :      1: jit_opt_b
 1.88% :     0.000492s :      1: loop_unroll
 3.13% :     0.000821s :      1: mutable_eliminate
 4.48% :     0.001172s :     26: opt.transform.jit_opt_a
 0.20% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.18% :     0.000572s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000053s :      1: py_interpret_to_execute
 0.06% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.57% :     0.000412s :      1: renormalize.infer
 1.40% :     0.000367s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000041s :      1: rewriter_after_opt_a
 0.31% :     0.000082s :      1: rewriter_before_opt_a
 0.32% :     0.000084s :      1: symbol_engine_optimizer
27.97% :     0.007328s :      1: task_emit
25.83% :     0.006766s :      1: type_inference
 0.24% :     0.000064s :      1: validate


TotalTime = 0.0254137, [33]
[bootstrap]: 0.00038787
[type_inference]: 0.00565798
[event_method]: 2.031e-05
[auto_monad]: 7.031e-05
[graph_reusing]: 6.43e-06
[pre_auto_parallel]: 2.58e-06
[py_interpret_to_execute]: 5.295e-05
[rewriter_before_opt_a]: 8.064e-05
[expand_dump_flag]: 3.91999e-06
[jit_opt_a]: 0.00801375, [2]
    [Cycle 1]: 0.00350518, [27]
        [switch_simplify]: 7.006e-05
        [loop_unroll]: 2.94e-05
        [a_1]: 0.00062672
        [with_stream_mark]: 2.938e-05
        [recompute_prepare]: 1.033e-05
        [updatestate_depend_eliminate]: 4.25999e-06
        [updatestate_assign_eliminate]: 3.49001e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 2.39999e-06
        [specialize_transform]: 6.17001e-06
        [updatestate_useless_node_eliminater]: 5.13002e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 2.00002e-06
        [get_grad_eliminate_]: 5.15999e-06
        [merge_forward]: 3.98001e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.601e-05
        [j_node_and_user_rematch]: 9.10999e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 1.056e-05
        [inline_without_move]: 5.33002e-06
        [renormalize]: 0.00228941
        [add_forward_monad_depend]: 1.239e-05
        [auto_monad_grad]: 2.91e-06
        [auto_monad_eliminator]: 2.73e-05
        [cse]: 3.955e-05
        [replace_applicator]: 2.675e-05
    [Cycle 2]: 0.00038932, [27]
        [switch_simplify]: 5.81e-06
        [loop_unroll]: 4.93001e-06
        [a_1]: 0.00010322
        [with_stream_mark]: 2.107e-05
        [recompute_prepare]: 5.50001e-06
        [updatestate_depend_eliminate]: 4.02e-06
        [updatestate_assign_eliminate]: 2.97002e-06
        [updatestate_loads_eliminate]: 2.50002e-06
        [parameter_eliminate]: 2.02001e-06
        [specialize_transform]: 5.15001e-06
        [updatestate_useless_node_eliminater]: 4.81002e-06
        [accelerated_algorithm]: 5.45001e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 4.65001e-06
        [merge_forward]: 4.27e-06
        [cell_reuse_recompute_pass]: 3.42002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 3.1e-05
        [j_node_and_user_rematch]: 1.021e-05
        [meta_fg_expand]: 2.02001e-06
        [replace_old_param]: 9.81998e-06
        [inline_without_move]: 5.25001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.95001e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 6.89001e-06
        [cse]: 1.262e-05
        [replace_applicator]: 5.27001e-06
[py_interpret_to_execute_after_opt_a]: 2.129e-05
[rewriter_after_opt_a]: 4.335e-05
[convert_after_rewriter]: 8.78001e-06
[order_py_execute_after_rewriter]: 5.42001e-06
[mutable_eliminate]: 0.00083649
[jit_opt_b]: 5.82e-05, [1]
    [Cycle 1]: 4.917e-05, [2]
        [frontend_op_eliminate]: 2.001e-05
        [inline_after_opt_a]: 1.642e-05
[cconv]: 3.762e-05
[loop_unroll]: 0.00048475
[jit_opt_after_cconv]: 0.00016758, [1]
    [Cycle 1]: 0.00016028, [11]
        [c_1]: 2.187e-05
        [parameter_eliminate]: 5.07e-06
        [updatestate_depend_eliminate]: 1.01e-05
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [cse]: 2.994e-05
        [call_graph_tuple_transform]: 2.343e-05
        [tuple_list_get_item_eliminator]: 5.06997e-06
        [none_parameter_eliminate]: 1.54998e-06
        [renormalize]: 6.40022e-07
        [switch_simplify]: 5.56e-06
[remove_dup_value]: 1.5e-05
[partial_unused_args_eliminate]: 2.83e-06
[environ_conv]: 5.60001e-06
[add_recomputation]: 5.98e-05
[cse_after_recomputation]: 2.273e-05, [1]
    [Cycle 1]: 1.658e-05, [1]
        [cse]: 1.027e-05
[auto_monad_reorder]: 1.624e-05
[get_jit_bprop_graph]: 2.22999e-06
[rewriter_after_jit_bprop_graph]: 9.84999e-06
[opt_after_jit_grad]: 0.00053124
[symbol_engine_optimizer]: 8.258e-05, [1]
    [Cycle 1]: 7.527e-05, [6]
        [build]: 5.99e-06
        [elim_shapecalc]: 8.40999e-06
        [elim_not_effective]: 1.54e-05
        [opt_reshape]: 6.51999e-06
        [fold_const_symbol]: 9.93002e-06
        [renormalize]: 1.02e-06
[validate]: 4.923e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00838289
[execute]: 1.087e-05
Sums
    bootstrap                            :     0.000388s :  1.89%
    type_inference                       :     0.005658s : 27.56%
    event_method                         :     0.000020s :  0.10%
    auto_monad                           :     0.000070s :  0.34%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000053s :  0.26%
    rewriter_before_opt_a                :     0.000081s :  0.39%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.37%
    jit_opt_a.loop_unroll                :     0.000034s :  0.17%
    jit_opt_a.a_1                        :     0.000730s :  3.56%
    jit_opt_a.with_stream_mark           :     0.000050s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000047s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.002289s : 11.15%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000034s :  0.17%
    jit_opt_a.cse                        :     0.000052s :  0.25%
    jit_opt_a.replace_applicator         :     0.000032s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000021s :  0.10%
    rewriter_after_opt_a                 :     0.000043s :  0.21%
    convert_after_rewriter               :     0.000009s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000836s :  4.08%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.000038s :  0.18%
    loop_unroll                          :     0.000485s :  2.36%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000030s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.07%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000060s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.05%
    opt_after_jit_grad                   :     0.000531s :  2.59%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000049s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008383s : 40.84%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000298    23
 0.63% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 1.92% :     0.000006s :      3: substitution.graph_param_transform
82.24% :     0.000245s :      4: substitution.inline
 1.72% :     0.000005s :      4: substitution.j_node_and_user_rematch
 5.86% :     0.000017s :      4: substitution.remove_not_recompute_node
 2.78% :     0.000008s :      2: substitution.replace_old_param
 4.38% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005574     2
84.34% :     0.004701s :      1: type_inference.infer
15.66% :     0.000873s :      1: type_inference.specialize
------[replace.]   0.000075     6
68.16% :     0.000051s :      4: replace.inline
31.84% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000255     6
95.29% :     0.000243s :      4: match.inline
 4.71% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.59% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.85% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.88% :     0.000002s :     10: predicate.addn_zero_filter
 2.24% :     0.000003s :     10: predicate.arithmetic_simplify
 1.47% :     0.000002s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.92% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.07% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.45% :     0.000001s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.44% :     0.000002s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 6.67% :     0.000009s :     22: predicate.inline
 1.24% :     0.000002s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.91% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.75% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.50% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 2.44% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     15: predicate.partial_eliminate
 1.01% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     18: predicate.replace_applicator
 0.83% :     0.000001s :      6: predicate.replace_old_param
 0.44% :     0.000001s :      3: predicate.reset_defer_inline
 1.75% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.09% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.11% :     0.000001s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.10% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.05% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.24% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.53% :     0.000010s :     53: predicate.switch_simplify
 1.17% :     0.000002s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.27% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.46% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.35% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.32% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.46% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.69% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.27% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.51% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000743    12
47.41% :     0.000352s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.01% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.58% :     0.000331s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028807    76
 0.22% :     0.000063s :      1: add_recomputation
 0.26% :     0.000074s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.44% :     0.000415s :      1: bootstrap
 0.14% :     0.000040s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000025s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
27.83% :     0.008018s :      1: jit_opt_a
 0.59% :     0.000171s :      1: jit_opt_after_cconv
 0.21% :     0.000062s :      1: jit_opt_b
 1.72% :     0.000495s :      1: loop_unroll
 2.96% :     0.000852s :      1: mutable_eliminate
 3.44% :     0.000991s :     26: opt.transform.jit_opt_a
 0.18% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000030s :      4: opt.transform.jit_opt_b
 0.05% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.89% :     0.000543s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000057s :      1: py_interpret_to_execute
 0.08% :     0.000024s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000018s :      1: remove_dup_value
 6.24% :     0.001797s :      1: renormalize.infer
 1.65% :     0.000476s :      1: renormalize.specialize
 0.04% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000047s :      1: rewriter_after_opt_a
 0.29% :     0.000084s :      1: rewriter_before_opt_a
 0.30% :     0.000086s :      1: symbol_engine_optimizer
29.18% :     0.008405s :      1: task_emit
19.72% :     0.005682s :      1: type_inference
 0.26% :     0.000076s :      1: validate


TotalTime = 0.0379938, [33]
[bootstrap]: 0.00033206
[type_inference]: 0.00549892
[event_method]: 1.798e-05
[auto_monad]: 6.897e-05
[graph_reusing]: 6.66e-06
[pre_auto_parallel]: 2.59999e-06
[py_interpret_to_execute]: 4.674e-05
[rewriter_before_opt_a]: 7.824e-05
[expand_dump_flag]: 3.41001e-06
[jit_opt_a]: 0.0226294, [2]
    [Cycle 1]: 0.00192504, [27]
        [switch_simplify]: 7.152e-05
        [loop_unroll]: 2.939e-05
        [a_1]: 0.00059701
        [with_stream_mark]: 1.959e-05
        [recompute_prepare]: 9.05001e-06
        [updatestate_depend_eliminate]: 4.32998e-06
        [updatestate_assign_eliminate]: 2.66999e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.97001e-06
        [specialize_transform]: 6.17001e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 2.49999e-06
        [get_grad_eliminate_]: 5.96e-06
        [merge_forward]: 3.65e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.717e-05
        [j_node_and_user_rematch]: 1.039e-05
        [meta_fg_expand]: 2.21e-06
        [replace_old_param]: 1.019e-05
        [inline_without_move]: 5.91e-06
        [renormalize]: 0.0008541
        [add_forward_monad_depend]: 6.56e-06
        [auto_monad_grad]: 2.94999e-06
        [auto_monad_eliminator]: 1.476e-05
        [cse]: 2.873e-05
        [replace_applicator]: 1.59e-05
    [Cycle 2]: 0.00034022, [27]
        [switch_simplify]: 5.76e-06
        [loop_unroll]: 5.61e-06
        [a_1]: 9.416e-05
        [with_stream_mark]: 1.047e-05
        [recompute_prepare]: 5.53002e-06
        [updatestate_depend_eliminate]: 2.86e-06
        [updatestate_assign_eliminate]: 2.31998e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 1.64e-06
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.68002e-06
        [meta_shard_fg_expand]: 1.66e-06
        [get_grad_eliminate_]: 4.63999e-06
        [merge_forward]: 3.21999e-06
        [cell_reuse_recompute_pass]: 2.02999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.53e-05
        [j_node_and_user_rematch]: 7.90998e-06
        [meta_fg_expand]: 1.89e-06
        [replace_old_param]: 8.21002e-06
        [inline_without_move]: 5.09e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.29998e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 5.95002e-06
        [cse]: 1.001e-05
        [replace_applicator]: 8.84e-06
[py_interpret_to_execute_after_opt_a]: 1.186e-05
[rewriter_after_opt_a]: 3.608e-05
[convert_after_rewriter]: 6.69001e-06
[order_py_execute_after_rewriter]: 4.50999e-06
[mutable_eliminate]: 0.00075426
[jit_opt_b]: 6.562e-05, [1]
    [Cycle 1]: 5.74e-05, [2]
        [frontend_op_eliminate]: 2.611e-05
        [inline_after_opt_a]: 1.785e-05
[cconv]: 2.615e-05
[loop_unroll]: 0.00050444
[jit_opt_after_cconv]: 0.00016557, [1]
    [Cycle 1]: 0.00015876, [11]
        [c_1]: 2.575e-05
        [parameter_eliminate]: 3.16001e-06
        [updatestate_depend_eliminate]: 7.2e-06
        [updatestate_assign_eliminate]: 2.64999e-06
        [updatestate_loads_eliminate]: 3.11001e-06
        [cse]: 2.622e-05
        [call_graph_tuple_transform]: 2.186e-05
        [tuple_list_get_item_eliminator]: 5.69e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 8.00006e-07
        [switch_simplify]: 5.57001e-06
[remove_dup_value]: 1.495e-05
[partial_unused_args_eliminate]: 2.00002e-06
[environ_conv]: 5.58002e-06
[add_recomputation]: 5.278e-05
[cse_after_recomputation]: 2.282e-05, [1]
    [Cycle 1]: 1.643e-05, [1]
        [cse]: 9.84999e-06
[auto_monad_reorder]: 1.693e-05
[get_jit_bprop_graph]: 2.24001e-06
[rewriter_after_jit_bprop_graph]: 8.82e-06
[opt_after_jit_grad]: 0.00054117
[symbol_engine_optimizer]: 8.659e-05, [1]
    [Cycle 1]: 7.886e-05, [6]
        [build]: 5.54998e-06
        [elim_shapecalc]: 8.74e-06
        [elim_not_effective]: 1.477e-05
        [opt_reshape]: 6.68e-06
        [fold_const_symbol]: 1.024e-05
        [renormalize]: 6.70028e-07
[validate]: 4.254e-05
[backend_pass]: 1.13001e-06
[task_emit]: 0.00670968
[execute]: 7.65e-06
Sums
    bootstrap                            :     0.000332s :  1.96%
    type_inference                       :     0.005499s : 32.41%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000069s :  0.41%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000047s :  0.28%
    rewriter_before_opt_a                :     0.000078s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000077s :  0.46%
    jit_opt_a.loop_unroll                :     0.000035s :  0.21%
    jit_opt_a.a_1                        :     0.000691s :  4.07%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000854s :  5.03%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.12%
    jit_opt_a.cse                        :     0.000039s :  0.23%
    jit_opt_a.replace_applicator         :     0.000025s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000754s :  4.44%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.11%
    cconv                                :     0.000026s :  0.15%
    loop_unroll                          :     0.000504s :  2.97%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000026s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000053s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000541s :  3.19%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006710s : 39.54%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000236    23
 1.15% :     0.000003s :      2: substitution.elim_not_effective
 0.64% :     0.000002s :      2: substitution.fold_const_symbol
 2.52% :     0.000006s :      3: substitution.graph_param_transform
82.92% :     0.000196s :      4: substitution.inline
 1.76% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.54% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.18% :     0.000005s :      2: substitution.replace_old_param
 6.29% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005422     2
79.73% :     0.004323s :      1: type_inference.infer
20.27% :     0.001099s :      1: type_inference.specialize
------[replace.]   0.000080     6
68.69% :     0.000055s :      4: replace.inline
31.31% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000207     6
93.33% :     0.000193s :      4: match.inline
 6.67% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.61% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.48% :     0.000002s :     10: predicate.addn_zero_filter
 2.64% :     0.000004s :     10: predicate.arithmetic_simplify
 1.32% :     0.000002s :     10: predicate.cast_eliminate
 0.58% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.91% :     0.000003s :     10: predicate.depend_value_elim
 1.24% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.51% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.71% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     10: predicate.environ_get_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.27% :     0.000007s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.28% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.72% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 0.93% :     0.000001s :     10: predicate.minmaximum_grad
 2.52% :     0.000003s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000003s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.22% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.46% :     0.000002s :     18: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.05% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.93% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     10: predicate.same_eliminate
 0.65% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.18% :     0.000002s :      6: predicate.special_op_eliminate
 0.84% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.09% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.29% :     0.000004s :     16: predicate.switch_defer_inline
 1.78% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.70% :     0.000010s :     53: predicate.switch_simplify
 1.02% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000002s :     10: predicate.transpose_eliminate
 2.09% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.94% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.52% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.85% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000720    12
47.10% :     0.000339s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.39% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.51% :     0.000349s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.039912    76
 0.14% :     0.000056s :      1: add_recomputation
 0.18% :     0.000072s :      1: auto_monad
 0.05% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.87% :     0.000345s :      1: bootstrap
 0.07% :     0.000029s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.06% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.06% :     0.000022s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
56.71% :     0.022634s :      1: jit_opt_a
 0.42% :     0.000168s :      1: jit_opt_after_cconv
 0.17% :     0.000069s :      1: jit_opt_b
 1.29% :     0.000514s :      1: loop_unroll
 1.92% :     0.000765s :      1: mutable_eliminate
 2.34% :     0.000936s :     26: opt.transform.jit_opt_a
 0.14% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000036s :      4: opt.transform.jit_opt_b
 0.03% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.38% :     0.000551s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000050s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000018s :      1: remove_dup_value
 1.12% :     0.000448s :      1: renormalize.infer
 1.00% :     0.000398s :      1: renormalize.specialize
 0.03% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000039s :      1: rewriter_after_opt_a
 0.20% :     0.000081s :      1: rewriter_before_opt_a
 0.22% :     0.000090s :      1: symbol_engine_optimizer
16.84% :     0.006723s :      1: task_emit
13.82% :     0.005516s :      1: type_inference
 0.16% :     0.000065s :      1: validate


TotalTime = 0.0146026, [30]
[bootstrap]: 0.00035353
[type_inference]: 0.00497282
[event_method]: 1.793e-05
[auto_monad]: 6.429e-05
[graph_reusing]: 6.14001e-06
[pre_auto_parallel]: 2.33002e-06
[py_interpret_to_execute]: 3.438e-05
[rewriter_before_opt_a]: 7.055e-05
[expand_dump_flag]: 2.99001e-06
[jit_opt_a]: 0.00665377, [2]
    [Cycle 1]: 0.00246086, [27]
        [switch_simplify]: 5.682e-05
        [loop_unroll]: 2.926e-05
        [a_1]: 0.00112535
        [with_stream_mark]: 2.216e-05
        [recompute_prepare]: 8.57e-06
        [updatestate_depend_eliminate]: 3.31999e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [parameter_eliminate]: 2.04999e-06
        [specialize_transform]: 6.01003e-06
        [updatestate_useless_node_eliminater]: 5.37001e-06
        [accelerated_algorithm]: 5.59998e-06
        [meta_shard_fg_expand]: 2.56998e-06
        [get_grad_eliminate_]: 5.19e-06
        [merge_forward]: 3.51001e-06
        [cell_reuse_recompute_pass]: 1.31002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.483e-05
        [j_node_and_user_rematch]: 1.007e-05
        [meta_fg_expand]: 2.46998e-06
        [replace_old_param]: 9.45001e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 0.00088074
        [add_forward_monad_depend]: 5.99e-06
        [auto_monad_grad]: 2.99999e-06
        [auto_monad_eliminator]: 1.431e-05
        [cse]: 2.706e-05
        [replace_applicator]: 1.238e-05
    [Cycle 2]: 0.0003294, [27]
        [switch_simplify]: 5.99e-06
        [loop_unroll]: 5.07e-06
        [a_1]: 9.204e-05
        [with_stream_mark]: 1.032e-05
        [recompute_prepare]: 5.09e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 2.01998e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 5.31002e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.50001e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 4.71002e-06
        [merge_forward]: 2.41e-06
        [cell_reuse_recompute_pass]: 1.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.277e-05
        [j_node_and_user_rematch]: 8.68001e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 7.55998e-06
        [inline_without_move]: 4.90001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.53002e-06
        [auto_monad_grad]: 1.14e-06
        [auto_monad_eliminator]: 5.19e-06
        [cse]: 1.106e-05
        [replace_applicator]: 5.00999e-06
[py_interpret_to_execute_after_opt_a]: 1.273e-05
[rewriter_after_opt_a]: 3.377e-05
[convert_after_rewriter]: 9.92999e-06
[order_py_execute_after_rewriter]: 4.79998e-06
[mutable_eliminate]: 0.00072158
[jit_opt_b]: 5.331e-05, [1]
    [Cycle 1]: 4.514e-05, [2]
        [frontend_op_eliminate]: 1.76e-05
        [inline_after_opt_a]: 1.553e-05
[cconv]: 2.738e-05
[loop_unroll]: 0.00045973
[jit_opt_after_cconv]: 0.00020648, [1]
    [Cycle 1]: 0.00019924, [11]
        [c_1]: 2.218e-05
        [parameter_eliminate]: 2.87002e-06
        [updatestate_depend_eliminate]: 6.48e-06
        [updatestate_assign_eliminate]: 2.31998e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.835e-05
        [call_graph_tuple_transform]: 2.152e-05
        [tuple_list_get_item_eliminator]: 5.66e-06
        [none_parameter_eliminate]: 1.89e-06
        [renormalize]: 5.79981e-07
        [switch_simplify]: 5.61003e-06
[remove_dup_value]: 1.499e-05
[partial_unused_args_eliminate]: 2.14999e-06
[environ_conv]: 5.49998e-06
[add_recomputation]: 4.688e-05
[cse_after_recomputation]: 2.198e-05, [1]
    [Cycle 1]: 1.574e-05, [1]
        [cse]: 9.66998e-06
[auto_monad_reorder]: 1.574e-05
[get_jit_bprop_graph]: 2.11998e-06
[rewriter_after_jit_bprop_graph]: 6.66e-06
[opt_after_jit_grad]: 0.00047837
[symbol_engine_optimizer]: 7.664e-05, [1]
    [Cycle 1]: 6.943e-05, [6]
        [build]: 3.43e-06
        [elim_shapecalc]: 7.76001e-06
        [elim_not_effective]: 1.399e-05
        [opt_reshape]: 6.84999e-06
        [fold_const_symbol]: 9.10999e-06
        [renormalize]: 3.7998e-07
[validate]: 3.865e-05
Sums
    bootstrap                            :     0.000354s :  3.51%
    type_inference                       :     0.004973s : 49.43%
    event_method                         :     0.000018s :  0.18%
    auto_monad                           :     0.000064s :  0.64%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.34%
    rewriter_before_opt_a                :     0.000071s :  0.70%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000063s :  0.62%
    jit_opt_a.loop_unroll                :     0.000034s :  0.34%
    jit_opt_a.a_1                        :     0.001217s : 12.10%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.17%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000881s :  8.76%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.19%
    jit_opt_a.cse                        :     0.000038s :  0.38%
    jit_opt_a.replace_applicator         :     0.000017s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.13%
    rewriter_after_opt_a                 :     0.000034s :  0.34%
    convert_after_rewriter               :     0.000010s :  0.10%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000722s :  7.17%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.15%
    cconv                                :     0.000027s :  0.27%
    loop_unroll                          :     0.000460s :  4.57%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000015s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000047s :  0.47%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000478s :  4.76%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000039s :  0.38%


Time group info:
------[substitution.]   0.000754    23
 0.29% :     0.000002s :      2: substitution.elim_not_effective
 0.16% :     0.000001s :      2: substitution.fold_const_symbol
 0.83% :     0.000006s :      3: substitution.graph_param_transform
95.43% :     0.000719s :      4: substitution.inline
 0.48% :     0.000004s :      4: substitution.j_node_and_user_rematch
 0.60% :     0.000005s :      4: substitution.remove_not_recompute_node
 0.57% :     0.000004s :      2: substitution.replace_old_param
 1.64% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004907     2
82.53% :     0.004050s :      1: type_inference.infer
17.47% :     0.000857s :      1: type_inference.specialize
------[replace.]   0.000090     6
69.95% :     0.000063s :      4: replace.inline
30.05% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000727     6
98.46% :     0.000715s :      4: match.inline
 1.54% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.43% :     0.000002s :     10: predicate.addn_zero_filter
 2.53% :     0.000003s :     10: predicate.arithmetic_simplify
 1.15% :     0.000001s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.51% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 6.10% :     0.000007s :     22: predicate.inline
 1.04% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.81% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.86% :     0.000002s :     15: predicate.load_eliminater
 1.67% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.80% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 1.64% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.75% :     0.000003s :     16: predicate.switch_defer_inline
 2.23% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.20% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.47% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.15% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.87% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000672    12
51.48% :     0.000346s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.70% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.82% :     0.000281s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016968    72
 0.29% :     0.000050s :      1: add_recomputation
 0.40% :     0.000068s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 2.22% :     0.000376s :      1: bootstrap
 0.18% :     0.000030s :      1: cconv
 0.07% :     0.000012s :      1: convert_after_rewriter
 0.14% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.13% :     0.000022s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.24% :     0.006658s :      1: jit_opt_a
 1.23% :     0.000209s :      1: jit_opt_after_cconv
 0.33% :     0.000056s :      1: jit_opt_b
 2.76% :     0.000468s :      1: loop_unroll
 4.31% :     0.000731s :      1: mutable_eliminate
 8.44% :     0.001433s :     26: opt.transform.jit_opt_a
 0.30% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000027s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.86% :     0.000486s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.22% :     0.000037s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000018s :      1: remove_dup_value
 2.90% :     0.000492s :      1: renormalize.infer
 2.24% :     0.000381s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000037s :      1: rewriter_after_opt_a
 0.43% :     0.000074s :      1: rewriter_before_opt_a
 0.47% :     0.000079s :      1: symbol_engine_optimizer
29.38% :     0.004985s :      1: type_inference


TotalTime = 0.0199729, [33]
[bootstrap]: 0.00033607
[type_inference]: 0.005097
[event_method]: 1.803e-05
[auto_monad]: 5.259e-05
[graph_reusing]: 5.70001e-06
[pre_auto_parallel]: 2.02001e-06
[py_interpret_to_execute]: 3.771e-05
[rewriter_before_opt_a]: 7.01e-05
[expand_dump_flag]: 2.47001e-06
[jit_opt_a]: 0.00573287, [2]
    [Cycle 1]: 0.00167411, [27]
        [switch_simplify]: 5.363e-05
        [loop_unroll]: 2.939e-05
        [a_1]: 0.00054024
        [with_stream_mark]: 1.475e-05
        [recompute_prepare]: 7.33e-06
        [updatestate_depend_eliminate]: 3.18998e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 1.88997e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 6.02001e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 1.08001e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 2.73e-06
        [cell_reuse_recompute_pass]: 9.5999e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.153e-05
        [j_node_and_user_rematch]: 8.70001e-06
        [meta_fg_expand]: 2.02001e-06
        [replace_old_param]: 8.02e-06
        [inline_without_move]: 5.45001e-06
        [renormalize]: 0.00073863
        [add_forward_monad_depend]: 5.41998e-06
        [auto_monad_grad]: 1.55999e-06
        [auto_monad_eliminator]: 1.132e-05
        [cse]: 1.723e-05
        [replace_applicator]: 1.176e-05
    [Cycle 2]: 0.00031343, [27]
        [switch_simplify]: 5.62001e-06
        [loop_unroll]: 5.14998e-06
        [a_1]: 9.101e-05
        [with_stream_mark]: 9.37999e-06
        [recompute_prepare]: 4.72e-06
        [updatestate_depend_eliminate]: 2.86e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.01998e-06
        [parameter_eliminate]: 9.29984e-07
        [specialize_transform]: 4.74e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.10001e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.80999e-06
        [merge_forward]: 2.24001e-06
        [cell_reuse_recompute_pass]: 2.37001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.251e-05
        [j_node_and_user_rematch]: 7.78001e-06
        [meta_fg_expand]: 1.60001e-06
        [replace_old_param]: 7.12002e-06
        [inline_without_move]: 4.58999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 9.49978e-07
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 4.93001e-06
        [cse]: 9.48002e-06
        [replace_applicator]: 4.67e-06
[py_interpret_to_execute_after_opt_a]: 9.69e-06
[rewriter_after_opt_a]: 2.693e-05
[convert_after_rewriter]: 5.97001e-06
[order_py_execute_after_rewriter]: 4.51002e-06
[mutable_eliminate]: 0.00062387
[jit_opt_b]: 5.388e-05, [1]
    [Cycle 1]: 4.72e-05, [2]
        [frontend_op_eliminate]: 2.088e-05
        [inline_after_opt_a]: 1.506e-05
[cconv]: 1.788e-05
[loop_unroll]: 0.0005005
[jit_opt_after_cconv]: 0.00013869, [1]
    [Cycle 1]: 0.00013219, [11]
        [c_1]: 2.164e-05
        [parameter_eliminate]: 3.3e-06
        [updatestate_depend_eliminate]: 5.77999e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [cse]: 1.75e-05
        [call_graph_tuple_transform]: 1.768e-05
        [tuple_list_get_item_eliminator]: 5.15001e-06
        [none_parameter_eliminate]: 1.15001e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.25999e-06
[remove_dup_value]: 7.94997e-06
[partial_unused_args_eliminate]: 1.44e-06
[environ_conv]: 3.56001e-06
[add_recomputation]: 3.534e-05
[cse_after_recomputation]: 1.981e-05, [1]
    [Cycle 1]: 1.437e-05, [1]
        [cse]: 8.55001e-06
[auto_monad_reorder]: 9.30001e-06
[get_jit_bprop_graph]: 1.44e-06
[rewriter_after_jit_bprop_graph]: 5.52001e-06
[opt_after_jit_grad]: 0.00047594
[symbol_engine_optimizer]: 7.29e-05, [1]
    [Cycle 1]: 6.601e-05, [6]
        [build]: 2.89999e-06
        [elim_shapecalc]: 8.07e-06
        [elim_not_effective]: 1.195e-05
        [opt_reshape]: 5.97999e-06
        [fold_const_symbol]: 8.52e-06
        [renormalize]: 7.2e-07
[validate]: 3.217e-05
[backend_pass]: 8.80013e-07
[task_emit]: 0.00635402
[execute]: 9.20999e-06
Sums
    bootstrap                            :     0.000336s :  2.15%
    type_inference                       :     0.005097s : 32.63%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000053s :  0.34%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000038s :  0.24%
    rewriter_before_opt_a                :     0.000070s :  0.45%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.38%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000631s :  4.04%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000739s :  4.73%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.10%
    jit_opt_a.cse                        :     0.000027s :  0.17%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000027s :  0.17%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000624s :  3.99%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000018s :  0.11%
    loop_unroll                          :     0.000500s :  3.20%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000008s :  0.05%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000035s :  0.23%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000009s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000476s :  3.05%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006354s : 40.68%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000198    23
 0.70% :     0.000001s :      2: substitution.elim_not_effective
 0.81% :     0.000002s :      2: substitution.fold_const_symbol
 2.26% :     0.000004s :      3: substitution.graph_param_transform
86.85% :     0.000172s :      4: substitution.inline
 1.36% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.93% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.70% :     0.000003s :      2: substitution.replace_old_param
 4.40% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005031     2
81.46% :     0.004098s :      1: type_inference.infer
18.54% :     0.000933s :      1: type_inference.specialize
------[replace.]   0.000076     6
67.91% :     0.000051s :      4: replace.inline
32.09% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000177     6
95.72% :     0.000170s :      4: match.inline
 4.28% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.48% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 2.11% :     0.000002s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.39% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.25% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 6.32% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.87% :     0.000002s :     15: predicate.load_eliminater
 1.51% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.79% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.35% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 2.27% :     0.000003s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.54% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.14% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.22% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.09% :     0.000008s :     53: predicate.switch_simplify
 1.53% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.68% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.88% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000634    12
46.24% :     0.000293s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.22% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.54% :     0.000308s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021651    76
 0.18% :     0.000038s :      1: add_recomputation
 0.26% :     0.000056s :      1: auto_monad
 0.05% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.64% :     0.000354s :      1: bootstrap
 0.09% :     0.000020s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.50% :     0.005737s :      1: jit_opt_a
 0.65% :     0.000141s :      1: jit_opt_after_cconv
 0.26% :     0.000056s :      1: jit_opt_b
 2.35% :     0.000509s :      1: loop_unroll
 2.92% :     0.000632s :      1: mutable_eliminate
 3.86% :     0.000836s :     26: opt.transform.jit_opt_a
 0.21% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000029s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.24% :     0.000485s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000040s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000010s :      1: remove_dup_value
 1.79% :     0.000387s :      1: renormalize.infer
 1.59% :     0.000343s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000030s :      1: rewriter_after_opt_a
 0.34% :     0.000073s :      1: rewriter_before_opt_a
 0.35% :     0.000075s :      1: symbol_engine_optimizer
29.44% :     0.006374s :      1: task_emit
23.60% :     0.005110s :      1: type_inference
 0.23% :     0.000050s :      1: validate


TotalTime = 0.0227758, [33]
[bootstrap]: 0.00035606
[type_inference]: 0.00558278
[event_method]: 2.054e-05
[auto_monad]: 6.963e-05
[graph_reusing]: 5.99999e-06
[pre_auto_parallel]: 2.66e-06
[py_interpret_to_execute]: 4.17e-05
[rewriter_before_opt_a]: 7.554e-05
[expand_dump_flag]: 3.39001e-06
[jit_opt_a]: 0.00697134, [2]
    [Cycle 1]: 0.00225944, [27]
        [switch_simplify]: 7.434e-05
        [loop_unroll]: 2.925e-05
        [a_1]: 0.00069587
        [with_stream_mark]: 2.946e-05
        [recompute_prepare]: 9.73998e-06
        [updatestate_depend_eliminate]: 4.05e-06
        [updatestate_assign_eliminate]: 3.68e-06
        [updatestate_loads_eliminate]: 2.69001e-06
        [parameter_eliminate]: 2.74001e-06
        [specialize_transform]: 6.39001e-06
        [updatestate_useless_node_eliminater]: 5.39e-06
        [accelerated_algorithm]: 6.30002e-06
        [meta_shard_fg_expand]: 2.19001e-06
        [get_grad_eliminate_]: 5.91e-06
        [merge_forward]: 4.05e-06
        [cell_reuse_recompute_pass]: 2.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.659e-05
        [j_node_and_user_rematch]: 1.117e-05
        [meta_fg_expand]: 2.94001e-06
        [replace_old_param]: 1.094e-05
        [inline_without_move]: 6.27001e-06
        [renormalize]: 0.0010468
        [add_forward_monad_depend]: 7.55e-06
        [auto_monad_grad]: 2.59999e-06
        [auto_monad_eliminator]: 1.611e-05
        [cse]: 3.146e-05
        [replace_applicator]: 1.789e-05
    [Cycle 2]: 0.00035367, [27]
        [switch_simplify]: 6.16e-06
        [loop_unroll]: 4.91002e-06
        [a_1]: 9.828e-05
        [with_stream_mark]: 1.241e-05
        [recompute_prepare]: 6.34001e-06
        [updatestate_depend_eliminate]: 2.98e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 5.13002e-06
        [updatestate_useless_node_eliminater]: 5.46e-06
        [accelerated_algorithm]: 5.29998e-06
        [meta_shard_fg_expand]: 1.54998e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 3.08e-06
        [cell_reuse_recompute_pass]: 2.49001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.621e-05
        [j_node_and_user_rematch]: 8.99e-06
        [meta_fg_expand]: 1.51002e-06
        [replace_old_param]: 8.47e-06
        [inline_without_move]: 1.257e-05
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.32999e-06
        [auto_monad_grad]: 1.07e-06
        [auto_monad_eliminator]: 5.72999e-06
        [cse]: 1.153e-05
        [replace_applicator]: 5.55001e-06
[py_interpret_to_execute_after_opt_a]: 1.494e-05
[rewriter_after_opt_a]: 4.289e-05
[convert_after_rewriter]: 7.14001e-06
[order_py_execute_after_rewriter]: 5.56998e-06
[mutable_eliminate]: 0.00088084
[jit_opt_b]: 5.843e-05, [1]
    [Cycle 1]: 4.86e-05, [2]
        [frontend_op_eliminate]: 1.867e-05
        [inline_after_opt_a]: 1.746e-05
[cconv]: 3.274e-05
[loop_unroll]: 0.00048663
[jit_opt_after_cconv]: 0.00016385, [1]
    [Cycle 1]: 0.00015697, [11]
        [c_1]: 2.303e-05
        [parameter_eliminate]: 5.15999e-06
        [updatestate_depend_eliminate]: 8.82999e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [cse]: 2.796e-05
        [call_graph_tuple_transform]: 2.143e-05
        [tuple_list_get_item_eliminator]: 5.40999e-06
        [none_parameter_eliminate]: 1.67999e-06
        [renormalize]: 7.2e-07
        [switch_simplify]: 5.69e-06
[remove_dup_value]: 1.507e-05
[partial_unused_args_eliminate]: 2.11998e-06
[environ_conv]: 5.00999e-06
[add_recomputation]: 5.419e-05
[cse_after_recomputation]: 2.139e-05, [1]
    [Cycle 1]: 1.557e-05, [1]
        [cse]: 9.68002e-06
[auto_monad_reorder]: 1.562e-05
[get_jit_bprop_graph]: 2.24999e-06
[rewriter_after_jit_bprop_graph]: 9.00001e-06
[opt_after_jit_grad]: 0.00048188
[symbol_engine_optimizer]: 7.759e-05, [1]
    [Cycle 1]: 7.043e-05, [6]
        [build]: 5.20999e-06
        [elim_shapecalc]: 8.18999e-06
        [elim_not_effective]: 1.356e-05
        [opt_reshape]: 5.89e-06
        [fold_const_symbol]: 8.40999e-06
        [renormalize]: 8.60018e-07
[validate]: 4.575e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00697299
[execute]: 8.88002e-06
Sums
    bootstrap                            :     0.000356s :  2.01%
    type_inference                       :     0.005583s : 31.49%
    event_method                         :     0.000021s :  0.12%
    auto_monad                           :     0.000070s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000042s :  0.24%
    rewriter_before_opt_a                :     0.000076s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000081s :  0.45%
    jit_opt_a.loop_unroll                :     0.000034s :  0.19%
    jit_opt_a.a_1                        :     0.000794s :  4.48%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000019s :  0.11%
    jit_opt_a.renormalize                :     0.001047s :  5.91%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.12%
    jit_opt_a.cse                        :     0.000043s :  0.24%
    jit_opt_a.replace_applicator         :     0.000023s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.08%
    rewriter_after_opt_a                 :     0.000043s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000881s :  4.97%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.10%
    cconv                                :     0.000033s :  0.18%
    loop_unroll                          :     0.000487s :  2.75%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000054s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000482s :  2.72%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.26%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006973s : 39.33%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000312    23
 0.64% :     0.000002s :      2: substitution.elim_not_effective
 0.37% :     0.000001s :      2: substitution.fold_const_symbol
 1.94% :     0.000006s :      3: substitution.graph_param_transform
87.43% :     0.000273s :      4: substitution.inline
 1.39% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.89% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.91% :     0.000006s :      2: substitution.replace_old_param
 4.43% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005496     2
80.90% :     0.004446s :      1: type_inference.infer
19.10% :     0.001050s :      1: type_inference.specialize
------[replace.]   0.000098     6
64.62% :     0.000063s :      4: replace.inline
35.38% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000283     6
95.46% :     0.000270s :      4: match.inline
 4.54% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 2.87% :     0.000004s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 2.32% :     0.000003s :     10: predicate.depend_value_elim
 0.94% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000001s :     10: predicate.environ_get_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.72% :     0.000008s :     22: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.39% :     0.000002s :      6: predicate.less_batch_normalization
 1.29% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.62% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.47% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.38% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 2.54% :     0.000003s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.62% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.20% :     0.000002s :     10: predicate.reshape_eliminate
 1.05% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000001s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 0.82% :     0.000001s :      6: predicate.specialize_transform
 1.23% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.52% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.67% :     0.000004s :     16: predicate.switch_defer_inline
 2.19% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.81% :     0.000010s :     53: predicate.switch_simplify
 1.16% :     0.000002s :     10: predicate.tile_eliminate
 2.06% :     0.000003s :     10: predicate.transpose_eliminate
 1.64% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.78% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.32% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.58% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000750    12
46.96% :     0.000352s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.59% :     0.000064s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.45% :     0.000333s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024978    76
 0.23% :     0.000057s :      1: add_recomputation
 0.30% :     0.000074s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.52% :     0.000379s :      1: bootstrap
 0.14% :     0.000035s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.11% :     0.000027s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
27.92% :     0.006975s :      1: jit_opt_a
 0.67% :     0.000166s :      1: jit_opt_after_cconv
 0.25% :     0.000061s :      1: jit_opt_b
 1.99% :     0.000496s :      1: loop_unroll
 3.58% :     0.000894s :      1: mutable_eliminate
 4.20% :     0.001050s :     26: opt.transform.jit_opt_a
 0.21% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.96% :     0.000490s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000044s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.33% :     0.000581s :      1: renormalize.infer
 1.82% :     0.000456s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000046s :      1: rewriter_after_opt_a
 0.31% :     0.000078s :      1: rewriter_before_opt_a
 0.32% :     0.000080s :      1: symbol_engine_optimizer
27.98% :     0.006990s :      1: task_emit
22.44% :     0.005604s :      1: type_inference
 0.28% :     0.000069s :      1: validate


TotalTime = 0.0538331, [30]
[bootstrap]: 0.00035268
[type_inference]: 0.0445416
[event_method]: 1.983e-05
[auto_monad]: 6.793e-05
[graph_reusing]: 6.16e-06
[pre_auto_parallel]: 2.77002e-06
[py_interpret_to_execute]: 4.533e-05
[rewriter_before_opt_a]: 7.619e-05
[expand_dump_flag]: 3.51999e-06
[jit_opt_a]: 0.00635209, [2]
    [Cycle 1]: 0.00194024, [27]
        [switch_simplify]: 6.28e-05
        [loop_unroll]: 2.987e-05
        [a_1]: 0.00062929
        [with_stream_mark]: 2.366e-05
        [recompute_prepare]: 7.92e-06
        [updatestate_depend_eliminate]: 3.95e-06
        [updatestate_assign_eliminate]: 3.08998e-06
        [updatestate_loads_eliminate]: 2.74999e-06
        [parameter_eliminate]: 1.93002e-06
        [specialize_transform]: 6.48e-06
        [updatestate_useless_node_eliminater]: 5.27999e-06
        [accelerated_algorithm]: 5.32001e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 5.30999e-06
        [merge_forward]: 3.61001e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.507e-05
        [j_node_and_user_rematch]: 1.003e-05
        [meta_fg_expand]: 2.22001e-06
        [replace_old_param]: 9.89001e-06
        [inline_without_move]: 5.90002e-06
        [renormalize]: 0.00084345
        [add_forward_monad_depend]: 8.07998e-06
        [auto_monad_grad]: 2.87002e-06
        [auto_monad_eliminator]: 1.539e-05
        [cse]: 2.939e-05
        [replace_applicator]: 1.402e-05
    [Cycle 2]: 0.00033468, [27]
        [switch_simplify]: 5.81998e-06
        [loop_unroll]: 4.89e-06
        [a_1]: 9.285e-05
        [with_stream_mark]: 1.14e-05
        [recompute_prepare]: 5.22e-06
        [updatestate_depend_eliminate]: 2.69999e-06
        [updatestate_assign_eliminate]: 2.11998e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [parameter_eliminate]: 1.02998e-06
        [specialize_transform]: 4.95001e-06
        [updatestate_useless_node_eliminater]: 4.96002e-06
        [accelerated_algorithm]: 5.76998e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 4.89998e-06
        [merge_forward]: 3.03e-06
        [cell_reuse_recompute_pass]: 2.51e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.471e-05
        [j_node_and_user_rematch]: 8.16002e-06
        [meta_fg_expand]: 1.60001e-06
        [replace_old_param]: 7.53999e-06
        [inline_without_move]: 4.55999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.54998e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 5.52999e-06
        [cse]: 1.184e-05
        [replace_applicator]: 4.84e-06
[py_interpret_to_execute_after_opt_a]: 1.391e-05
[rewriter_after_opt_a]: 3.833e-05
[convert_after_rewriter]: 9.52001e-06
[order_py_execute_after_rewriter]: 5.61e-06
[mutable_eliminate]: 0.00071432
[jit_opt_b]: 5.28e-05, [1]
    [Cycle 1]: 4.485e-05, [2]
        [frontend_op_eliminate]: 1.741e-05
        [inline_after_opt_a]: 1.595e-05
[cconv]: 2.987e-05
[loop_unroll]: 0.00044233
[jit_opt_after_cconv]: 0.00014587, [1]
    [Cycle 1]: 0.00013883, [11]
        [c_1]: 2.132e-05
        [parameter_eliminate]: 3.40998e-06
        [updatestate_depend_eliminate]: 6.94999e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [cse]: 2.159e-05
        [call_graph_tuple_transform]: 1.954e-05
        [tuple_list_get_item_eliminator]: 5.45001e-06
        [none_parameter_eliminate]: 1.84e-06
        [renormalize]: 3.30008e-07
        [switch_simplify]: 5.11997e-06
[remove_dup_value]: 1.43e-05
[partial_unused_args_eliminate]: 2.24001e-06
[environ_conv]: 6.88e-06
[add_recomputation]: 5.198e-05
[cse_after_recomputation]: 2.137e-05, [1]
    [Cycle 1]: 1.58e-05, [1]
        [cse]: 9.07999e-06
[auto_monad_reorder]: 1.542e-05
[get_jit_bprop_graph]: 2.42001e-06
[rewriter_after_jit_bprop_graph]: 5.94999e-06
[opt_after_jit_grad]: 0.00047218
[symbol_engine_optimizer]: 7.73e-05, [1]
    [Cycle 1]: 6.976e-05, [6]
        [build]: 4.80999e-06
        [elim_shapecalc]: 8.03001e-06
        [elim_not_effective]: 1.363e-05
        [opt_reshape]: 5.81e-06
        [fold_const_symbol]: 8.77e-06
        [renormalize]: 6.19999e-07
[validate]: 4.311e-05
Sums
    bootstrap                            :     0.000353s :  0.72%
    type_inference                       :     0.044542s : 90.67%
    event_method                         :     0.000020s :  0.04%
    auto_monad                           :     0.000068s :  0.14%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000045s :  0.09%
    rewriter_before_opt_a                :     0.000076s :  0.16%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000069s :  0.14%
    jit_opt_a.loop_unroll                :     0.000035s :  0.07%
    jit_opt_a.a_1                        :     0.000722s :  1.47%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.02%
    jit_opt_a.merge_forward              :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000017s :  0.04%
    jit_opt_a.inline_without_move        :     0.000010s :  0.02%
    jit_opt_a.renormalize                :     0.000844s :  1.72%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.04%
    jit_opt_a.cse                        :     0.000041s :  0.08%
    jit_opt_a.replace_applicator         :     0.000019s :  0.04%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.03%
    rewriter_after_opt_a                 :     0.000038s :  0.08%
    convert_after_rewriter               :     0.000010s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.01%
    mutable_eliminate                    :     0.000714s :  1.45%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.03%
    cconv                                :     0.000030s :  0.06%
    loop_unroll                          :     0.000442s :  0.90%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.04%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000022s :  0.04%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.01%
    remove_dup_value                     :     0.000014s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000007s :  0.01%
    add_recomputation                    :     0.000052s :  0.11%
    cse_after_recomputation.cse          :     0.000009s :  0.02%
    auto_monad_reorder                   :     0.000015s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000472s :  0.96%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.09%


Time group info:
------[substitution.]   0.000263    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.19% :     0.000006s :      3: substitution.graph_param_transform
87.01% :     0.000229s :      4: substitution.inline
 1.42% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.69% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.59% :     0.000004s :      2: substitution.replace_old_param
 4.79% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.044452     2
97.35% :     0.043273s :      1: type_inference.infer
 2.65% :     0.001179s :      1: type_inference.specialize
------[replace.]   0.000082     6
65.21% :     0.000054s :      4: replace.inline
34.79% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000238     6
95.17% :     0.000226s :      4: match.inline
 4.83% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.17% :     0.000001s :     10: predicate.addn_zero_filter
 2.18% :     0.000003s :     10: predicate.arithmetic_simplify
 1.27% :     0.000002s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.24% :     0.000001s :     10: predicate.depend_value_elim
 1.50% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.13% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.09% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     10: predicate.environ_get_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.93% :     0.000007s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :      6: predicate.less_batch_normalization
 1.46% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 2.14% :     0.000003s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.80% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.71% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.47% :     0.000001s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.09% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.93% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.55% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.63% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.97% :     0.000010s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.56% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.03% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.33% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.59% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000792    12
45.34% :     0.000359s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.86% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.79% :     0.000379s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.055663    72
 0.10% :     0.000055s :      1: add_recomputation
 0.13% :     0.000072s :      1: auto_monad
 0.03% :     0.000018s :      1: auto_monad_reorder
 0.67% :     0.000375s :      1: bootstrap
 0.06% :     0.000032s :      1: cconv
 0.02% :     0.000012s :      1: convert_after_rewriter
 0.04% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.04% :     0.000025s :      1: event_method
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
11.42% :     0.006356s :      1: jit_opt_a
 0.27% :     0.000149s :      1: jit_opt_after_cconv
 0.10% :     0.000055s :      1: jit_opt_b
 0.81% :     0.000451s :      1: loop_unroll
 1.30% :     0.000726s :      1: mutable_eliminate
 1.70% :     0.000945s :     26: opt.transform.jit_opt_a
 0.09% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000027s :      4: opt.transform.jit_opt_b
 0.02% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000032s :      4: opt.transform.symbol_engine_opt
 0.86% :     0.000480s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000048s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000017s :      1: remove_dup_value
 0.83% :     0.000461s :      1: renormalize.infer
 0.67% :     0.000374s :      1: renormalize.specialize
 0.01% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000042s :      1: rewriter_after_opt_a
 0.14% :     0.000079s :      1: rewriter_before_opt_a
 0.14% :     0.000080s :      1: symbol_engine_optimizer
80.06% :     0.044561s :      1: type_inference


TotalTime = 0.0216244, [33]
[bootstrap]: 0.00035765
[type_inference]: 0.00510884
[event_method]: 1.799e-05
[auto_monad]: 6.385e-05
[graph_reusing]: 5.76003e-06
[pre_auto_parallel]: 2.49001e-06
[py_interpret_to_execute]: 3.936e-05
[rewriter_before_opt_a]: 7.42e-05
[expand_dump_flag]: 3.21999e-06
[jit_opt_a]: 0.00630128, [2]
    [Cycle 1]: 0.00226789, [27]
        [switch_simplify]: 6.896e-05
        [loop_unroll]: 3.028e-05
        [a_1]: 0.00063818
        [with_stream_mark]: 2.448e-05
        [recompute_prepare]: 8.59e-06
        [updatestate_depend_eliminate]: 4.32998e-06
        [updatestate_assign_eliminate]: 3.02002e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 6.33e-06
        [updatestate_useless_node_eliminater]: 5.47999e-06
        [accelerated_algorithm]: 5.68002e-06
        [meta_shard_fg_expand]: 3.09001e-06
        [get_grad_eliminate_]: 5.40999e-06
        [merge_forward]: 4.50001e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.592e-05
        [j_node_and_user_rematch]: 9.51e-06
        [meta_fg_expand]: 2.32001e-06
        [replace_old_param]: 1.123e-05
        [inline_without_move]: 5.40999e-06
        [renormalize]: 0.00112037
        [add_forward_monad_depend]: 9.45001e-06
        [auto_monad_grad]: 2.51e-06
        [auto_monad_eliminator]: 2.123e-05
        [cse]: 3.13e-05
        [replace_applicator]: 2.043e-05
    [Cycle 2]: 0.00037172, [27]
        [switch_simplify]: 6.69001e-06
        [loop_unroll]: 5.41002e-06
        [a_1]: 0.0001017
        [with_stream_mark]: 1.439e-05
        [recompute_prepare]: 5.37001e-06
        [updatestate_depend_eliminate]: 3.33998e-06
        [updatestate_assign_eliminate]: 2.52001e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [parameter_eliminate]: 1.49998e-06
        [specialize_transform]: 5.54e-06
        [updatestate_useless_node_eliminater]: 5.11997e-06
        [accelerated_algorithm]: 6.16e-06
        [meta_shard_fg_expand]: 2.07001e-06
        [get_grad_eliminate_]: 4.93001e-06
        [merge_forward]: 3.46999e-06
        [cell_reuse_recompute_pass]: 2.61e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.811e-05
        [j_node_and_user_rematch]: 8.45001e-06
        [meta_fg_expand]: 1.77001e-06
        [replace_old_param]: 9.20999e-06
        [inline_without_move]: 4.92e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 1.07e-06
        [auto_monad_eliminator]: 6.36998e-06
        [cse]: 1.278e-05
        [replace_applicator]: 5.51e-06
[py_interpret_to_execute_after_opt_a]: 1.429e-05
[rewriter_after_opt_a]: 4.112e-05
[convert_after_rewriter]: 1.232e-05
[order_py_execute_after_rewriter]: 5.01997e-06
[mutable_eliminate]: 0.00084668
[jit_opt_b]: 5.883e-05, [1]
    [Cycle 1]: 4.984e-05, [2]
        [frontend_op_eliminate]: 2.051e-05
        [inline_after_opt_a]: 1.631e-05
[cconv]: 3.565e-05
[loop_unroll]: 0.00049535
[jit_opt_after_cconv]: 0.00017898, [1]
    [Cycle 1]: 0.00016977, [11]
        [c_1]: 2.468e-05
        [parameter_eliminate]: 5.87999e-06
        [updatestate_depend_eliminate]: 1.105e-05
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [cse]: 3.032e-05
        [call_graph_tuple_transform]: 2.532e-05
        [tuple_list_get_item_eliminator]: 5.37999e-06
        [none_parameter_eliminate]: 1.90001e-06
        [renormalize]: 6.79982e-07
        [switch_simplify]: 5.37001e-06
[remove_dup_value]: 1.5e-05
[partial_unused_args_eliminate]: 2.30002e-06
[environ_conv]: 6.02999e-06
[add_recomputation]: 5.418e-05
[cse_after_recomputation]: 2.244e-05, [1]
    [Cycle 1]: 1.61e-05, [1]
        [cse]: 9.64e-06
[auto_monad_reorder]: 1.643e-05
[get_jit_bprop_graph]: 2.39999e-06
[rewriter_after_jit_bprop_graph]: 9.97999e-06
[opt_after_jit_grad]: 0.00051712
[symbol_engine_optimizer]: 8.358e-05, [1]
    [Cycle 1]: 7.453e-05, [6]
        [build]: 5.23002e-06
        [elim_shapecalc]: 8.35001e-06
        [elim_not_effective]: 1.488e-05
        [opt_reshape]: 6.64999e-06
        [fold_const_symbol]: 9.20999e-06
        [renormalize]: 3.50003e-07
[validate]: 5.03e-05
[backend_pass]: 1.15999e-06
[task_emit]: 0.00691966
[execute]: 9.04e-06
Sums
    bootstrap                            :     0.000358s :  2.07%
    type_inference                       :     0.005109s : 29.63%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000064s :  0.37%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.23%
    rewriter_before_opt_a                :     0.000074s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.44%
    jit_opt_a.loop_unroll                :     0.000036s :  0.21%
    jit_opt_a.a_1                        :     0.000740s :  4.29%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001120s :  6.50%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.16%
    jit_opt_a.cse                        :     0.000044s :  0.26%
    jit_opt_a.replace_applicator         :     0.000026s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000041s :  0.24%
    convert_after_rewriter               :     0.000012s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000847s :  4.91%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000036s :  0.21%
    loop_unroll                          :     0.000495s :  2.87%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000030s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000054s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.06%
    opt_after_jit_grad                   :     0.000517s :  3.00%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000050s :  0.29%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006920s : 40.13%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000284    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.48% :     0.000007s :      3: substitution.graph_param_transform
84.25% :     0.000239s :      4: substitution.inline
 1.46% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.19% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.70% :     0.000008s :      2: substitution.replace_old_param
 5.58% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005033     2
81.23% :     0.004088s :      1: type_inference.infer
18.77% :     0.000945s :      1: type_inference.specialize
------[replace.]   0.000089     6
66.96% :     0.000059s :      4: replace.inline
33.04% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000251     6
94.34% :     0.000236s :      4: match.inline
 5.66% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.52% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 2.09% :     0.000003s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.13% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.88% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.08% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 6.11% :     0.000008s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.40% :     0.000002s :      6: predicate.less_batch_normalization
 1.69% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 1.58% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.68% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 2.40% :     0.000003s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     10: predicate.reduce_eliminate
 1.33% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.82% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000002s :     10: predicate.reshape_eliminate
 1.02% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.05% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     10: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 1.89% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.68% :     0.000010s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.63% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.66% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.95% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.53% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000829    12
48.10% :     0.000399s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.42% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.48% :     0.000377s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023847    76
 0.24% :     0.000057s :      1: add_recomputation
 0.28% :     0.000067s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.60% :     0.000381s :      1: bootstrap
 0.16% :     0.000039s :      1: cconv
 0.06% :     0.000015s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.44% :     0.006306s :      1: jit_opt_a
 0.76% :     0.000182s :      1: jit_opt_after_cconv
 0.26% :     0.000062s :      1: jit_opt_b
 2.12% :     0.000506s :      1: loop_unroll
 3.62% :     0.000863s :      1: mutable_eliminate
 4.14% :     0.000986s :     26: opt.transform.jit_opt_a
 0.24% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000030s :      4: opt.transform.jit_opt_b
 0.07% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000527s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000042s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.32% :     0.000554s :      1: renormalize.infer
 2.33% :     0.000555s :      1: renormalize.specialize
 0.05% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000045s :      1: rewriter_after_opt_a
 0.32% :     0.000077s :      1: rewriter_before_opt_a
 0.36% :     0.000086s :      1: symbol_engine_optimizer
29.10% :     0.006939s :      1: task_emit
21.50% :     0.005127s :      1: type_inference
 0.32% :     0.000076s :      1: validate


TotalTime = 0.0228503, [33]
[bootstrap]: 0.00035453
[type_inference]: 0.00559861
[event_method]: 2.119e-05
[auto_monad]: 7.145e-05
[graph_reusing]: 5.99e-06
[pre_auto_parallel]: 2.93998e-06
[py_interpret_to_execute]: 3.985e-05
[rewriter_before_opt_a]: 7.576e-05
[expand_dump_flag]: 3.78999e-06
[jit_opt_a]: 0.00651689, [2]
    [Cycle 1]: 0.00206905, [27]
        [switch_simplify]: 6.856e-05
        [loop_unroll]: 2.964e-05
        [a_1]: 0.00062209
        [with_stream_mark]: 2.781e-05
        [recompute_prepare]: 8.27e-06
        [updatestate_depend_eliminate]: 4.82e-06
        [updatestate_assign_eliminate]: 3.63999e-06
        [updatestate_loads_eliminate]: 2.91999e-06
        [parameter_eliminate]: 2.19999e-06
        [specialize_transform]: 6.67002e-06
        [updatestate_useless_node_eliminater]: 4.96002e-06
        [accelerated_algorithm]: 5.59998e-06
        [meta_shard_fg_expand]: 2.46998e-06
        [get_grad_eliminate_]: 5.44e-06
        [merge_forward]: 4.17e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.571e-05
        [j_node_and_user_rematch]: 1.022e-05
        [meta_fg_expand]: 2.70002e-06
        [replace_old_param]: 1.005e-05
        [inline_without_move]: 5.30999e-06
        [renormalize]: 0.00092551
        [add_forward_monad_depend]: 8.83001e-06
        [auto_monad_grad]: 2.89999e-06
        [auto_monad_eliminator]: 2.037e-05
        [cse]: 3.592e-05
        [replace_applicator]: 1.911e-05
    [Cycle 2]: 0.00035363, [27]
        [switch_simplify]: 5.46e-06
        [loop_unroll]: 5.24e-06
        [a_1]: 9.225e-05
        [with_stream_mark]: 1.616e-05
        [recompute_prepare]: 5.07999e-06
        [updatestate_depend_eliminate]: 3.38e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 2.79001e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 4.90999e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 4.97999e-06
        [meta_shard_fg_expand]: 1.59e-06
        [get_grad_eliminate_]: 4.45999e-06
        [merge_forward]: 3.60998e-06
        [cell_reuse_recompute_pass]: 2.21e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.856e-05
        [j_node_and_user_rematch]: 8.67e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 8.80001e-06
        [inline_without_move]: 4.92e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.96998e-06
        [auto_monad_grad]: 1.74e-06
        [auto_monad_eliminator]: 7.15e-06
        [cse]: 1.33e-05
        [replace_applicator]: 5.55001e-06
[py_interpret_to_execute_after_opt_a]: 1.643e-05
[rewriter_after_opt_a]: 4.339e-05
[convert_after_rewriter]: 7.6e-06
[order_py_execute_after_rewriter]: 5.84e-06
[mutable_eliminate]: 0.00089111
[jit_opt_b]: 6.464e-05, [1]
    [Cycle 1]: 5.192e-05, [2]
        [frontend_op_eliminate]: 1.892e-05
        [inline_after_opt_a]: 1.838e-05
[cconv]: 4.195e-05
[loop_unroll]: 0.00056331
[jit_opt_after_cconv]: 0.00018338, [1]
    [Cycle 1]: 0.00017456, [11]
        [c_1]: 2.113e-05
        [parameter_eliminate]: 6.58998e-06
        [updatestate_depend_eliminate]: 1.166e-05
        [updatestate_assign_eliminate]: 2.67001e-06
        [updatestate_loads_eliminate]: 2.60997e-06
        [cse]: 3.894e-05
        [call_graph_tuple_transform]: 2.476e-05
        [tuple_list_get_item_eliminator]: 5.91e-06
        [none_parameter_eliminate]: 1.54998e-06
        [renormalize]: 1.31998e-06
        [switch_simplify]: 5.76e-06
[remove_dup_value]: 1.556e-05
[partial_unused_args_eliminate]: 2.59001e-06
[environ_conv]: 6.61e-06
[add_recomputation]: 5.796e-05
[cse_after_recomputation]: 2.346e-05, [1]
    [Cycle 1]: 1.769e-05, [1]
        [cse]: 1.037e-05
[auto_monad_reorder]: 1.594e-05
[get_jit_bprop_graph]: 1.94e-06
[rewriter_after_jit_bprop_graph]: 9.74e-06
[opt_after_jit_grad]: 0.00072501
[symbol_engine_optimizer]: 8.587e-05, [1]
    [Cycle 1]: 7.703e-05, [6]
        [build]: 5.96e-06
        [elim_shapecalc]: 7.95e-06
        [elim_not_effective]: 1.664e-05
        [opt_reshape]: 5.67001e-06
        [fold_const_symbol]: 8.72e-06
        [renormalize]: 8.40024e-07
[validate]: 4.836e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.00705886
[execute]: 9.95002e-06
Sums
    bootstrap                            :     0.000355s :  1.97%
    type_inference                       :     0.005599s : 31.10%
    event_method                         :     0.000021s :  0.12%
    auto_monad                           :     0.000071s :  0.40%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.22%
    rewriter_before_opt_a                :     0.000076s :  0.42%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.41%
    jit_opt_a.loop_unroll                :     0.000035s :  0.19%
    jit_opt_a.a_1                        :     0.000714s :  3.97%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000926s :  5.14%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.15%
    jit_opt_a.cse                        :     0.000049s :  0.27%
    jit_opt_a.replace_applicator         :     0.000025s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.09%
    rewriter_after_opt_a                 :     0.000043s :  0.24%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000891s :  4.95%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.10%
    cconv                                :     0.000042s :  0.23%
    loop_unroll                          :     0.000563s :  3.13%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000039s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000058s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.05%
    opt_after_jit_grad                   :     0.000725s :  4.03%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000048s :  0.27%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007059s : 39.21%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000282    23
 0.90% :     0.000003s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.25% :     0.000006s :      3: substitution.graph_param_transform
85.59% :     0.000242s :      4: substitution.inline
 1.40% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.03% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.25% :     0.000006s :      2: substitution.replace_old_param
 5.07% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005516     2
83.69% :     0.004616s :      1: type_inference.infer
16.31% :     0.000900s :      1: type_inference.specialize
------[replace.]   0.000073     6
69.45% :     0.000051s :      4: replace.inline
30.55% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000252     6
94.79% :     0.000239s :      4: match.inline
 5.21% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.76% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.20% :     0.000002s :     10: predicate.addn_zero_filter
 1.75% :     0.000002s :     10: predicate.arithmetic_simplify
 1.40% :     0.000002s :     10: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.92% :     0.000001s :     10: predicate.compare_switch_simplify
 1.56% :     0.000002s :     10: predicate.depend_value_elim
 0.94% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.53% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.49% :     0.000001s :      3: predicate.elim_not_effective
 0.75% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 6.24% :     0.000008s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000002s :      6: predicate.less_batch_normalization
 1.28% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.53% :     0.000002s :     15: predicate.load_eliminater
 2.30% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.69% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.76% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 3.76% :     0.000005s :      3: predicate.mutable_eliminate
 0.38% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     15: predicate.partial_eliminate
 1.03% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.70% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.46% :     0.000002s :     18: predicate.replace_applicator
 0.80% :     0.000001s :      6: predicate.replace_old_param
 0.41% :     0.000001s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.29% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.09% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.59% :     0.000002s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.25% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.55% :     0.000003s :     16: predicate.switch_defer_inline
 1.89% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.85% :     0.000010s :     53: predicate.switch_simplify
 1.21% :     0.000002s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.93% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.01% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.31% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.47% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000715    12
52.54% :     0.000376s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.21% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.25% :     0.000281s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024863    76
 0.25% :     0.000063s :      1: add_recomputation
 0.30% :     0.000075s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.52% :     0.000377s :      1: bootstrap
 0.18% :     0.000045s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000026s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.11% :     0.000026s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.23% :     0.006521s :      1: jit_opt_a
 0.75% :     0.000187s :      1: jit_opt_after_cconv
 0.28% :     0.000069s :      1: jit_opt_b
 2.34% :     0.000582s :      1: loop_unroll
 3.68% :     0.000914s :      1: mutable_eliminate
 3.83% :     0.000953s :     26: opt.transform.jit_opt_a
 0.22% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000029s :      4: opt.transform.jit_opt_b
 0.07% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000027s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.98% :     0.000740s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000043s :      1: py_interpret_to_execute
 0.08% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.08% :     0.000517s :      1: renormalize.infer
 1.60% :     0.000398s :      1: renormalize.specialize
 0.05% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000047s :      1: rewriter_after_opt_a
 0.32% :     0.000079s :      1: rewriter_before_opt_a
 0.36% :     0.000089s :      1: symbol_engine_optimizer
28.48% :     0.007081s :      1: task_emit
22.61% :     0.005621s :      1: type_inference
 0.30% :     0.000075s :      1: validate


TotalTime = 0.0275252, [30]
[bootstrap]: 0.00032809
[type_inference]: 0.00520148
[event_method]: 1.851e-05
[auto_monad]: 6.81e-05
[graph_reusing]: 5.81e-06
[pre_auto_parallel]: 2.75002e-06
[py_interpret_to_execute]: 4.118e-05
[rewriter_before_opt_a]: 7.447e-05
[expand_dump_flag]: 3.6e-06
[jit_opt_a]: 0.0065379, [2]
    [Cycle 1]: 0.00200008, [27]
        [switch_simplify]: 6.919e-05
        [loop_unroll]: 2.965e-05
        [a_1]: 0.00063549
        [with_stream_mark]: 2.272e-05
        [recompute_prepare]: 7.75998e-06
        [updatestate_depend_eliminate]: 3.93999e-06
        [updatestate_assign_eliminate]: 2.79999e-06
        [updatestate_loads_eliminate]: 2.48998e-06
        [parameter_eliminate]: 1.76998e-06
        [specialize_transform]: 5.92001e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.91e-06
        [meta_shard_fg_expand]: 1.98002e-06
        [get_grad_eliminate_]: 5.27001e-06
        [merge_forward]: 3.81001e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.485e-05
        [j_node_and_user_rematch]: 9.72001e-06
        [meta_fg_expand]: 1.97001e-06
        [replace_old_param]: 9.65002e-06
        [inline_without_move]: 5.86e-06
        [renormalize]: 0.00088711
        [add_forward_monad_depend]: 7.77e-06
        [auto_monad_grad]: 2.66999e-06
        [auto_monad_eliminator]: 1.568e-05
        [cse]: 3.148e-05
        [replace_applicator]: 1.523e-05
    [Cycle 2]: 0.0003363, [27]
        [switch_simplify]: 5.93002e-06
        [loop_unroll]: 5.18002e-06
        [a_1]: 9.117e-05
        [with_stream_mark]: 1.332e-05
        [recompute_prepare]: 5.30001e-06
        [updatestate_depend_eliminate]: 2.88e-06
        [updatestate_assign_eliminate]: 2.10002e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 1.30999e-06
        [specialize_transform]: 5.25999e-06
        [updatestate_useless_node_eliminater]: 5.42001e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.54002e-06
        [merge_forward]: 2.99999e-06
        [cell_reuse_recompute_pass]: 1.74e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.469e-05
        [j_node_and_user_rematch]: 8.39002e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 8.18999e-06
        [inline_without_move]: 4.73001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.70001e-06
        [auto_monad_grad]: 6.59988e-07
        [auto_monad_eliminator]: 5.06002e-06
        [cse]: 1.141e-05
        [replace_applicator]: 4.85999e-06
[py_interpret_to_execute_after_opt_a]: 1.446e-05
[rewriter_after_opt_a]: 3.919e-05
[convert_after_rewriter]: 7.87998e-06
[order_py_execute_after_rewriter]: 5.31002e-06
[mutable_eliminate]: 0.00077627
[jit_opt_b]: 5.47e-05, [1]
    [Cycle 1]: 4.603e-05, [2]
        [frontend_op_eliminate]: 1.875e-05
        [inline_after_opt_a]: 1.482e-05
[cconv]: 3.138e-05
[loop_unroll]: 0.00045255
[jit_opt_after_cconv]: 0.0123389, [1]
    [Cycle 1]: 0.0123168, [11]
        [c_1]: 2.211e-05
        [parameter_eliminate]: 4.23001e-06
        [updatestate_depend_eliminate]: 8.80001e-06
        [updatestate_assign_eliminate]: 2.64999e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 2.443e-05
        [call_graph_tuple_transform]: 2.279e-05
        [tuple_list_get_item_eliminator]: 5.42999e-06
        [none_parameter_eliminate]: 1.79998e-06
        [renormalize]: 7.89994e-07
        [switch_simplify]: 0.012062
[remove_dup_value]: 3.046e-05
[partial_unused_args_eliminate]: 7.02002e-06
[environ_conv]: 1.17e-05
[add_recomputation]: 8.023e-05
[cse_after_recomputation]: 5.916e-05, [1]
    [Cycle 1]: 4.716e-05, [1]
        [cse]: 3.22e-05
[auto_monad_reorder]: 1.899e-05
[get_jit_bprop_graph]: 2.48998e-06
[rewriter_after_jit_bprop_graph]: 1.249e-05
[opt_after_jit_grad]: 0.00093913
[symbol_engine_optimizer]: 9.163e-05, [1]
    [Cycle 1]: 8.224e-05, [6]
        [build]: 5.17e-06
        [elim_shapecalc]: 8.84998e-06
        [elim_not_effective]: 1.883e-05
        [opt_reshape]: 7.08e-06
        [fold_const_symbol]: 1.067e-05
        [renormalize]: 1.12999e-06
[validate]: 4.858e-05
Sums
    bootstrap                            :     0.000328s :  1.46%
    type_inference                       :     0.005201s : 23.10%
    event_method                         :     0.000019s :  0.08%
    auto_monad                           :     0.000068s :  0.30%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.18%
    rewriter_before_opt_a                :     0.000074s :  0.33%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000075s :  0.33%
    jit_opt_a.loop_unroll                :     0.000035s :  0.15%
    jit_opt_a.a_1                        :     0.000727s :  3.23%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.08%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.000887s :  3.94%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.09%
    jit_opt_a.cse                        :     0.000043s :  0.19%
    jit_opt_a.replace_applicator         :     0.000020s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.06%
    rewriter_after_opt_a                 :     0.000039s :  0.17%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000776s :  3.45%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.07%
    cconv                                :     0.000031s :  0.14%
    loop_unroll                          :     0.000453s :  2.01%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.012062s : 53.56%
    remove_dup_value                     :     0.000030s :  0.14%
    partial_unused_args_eliminate        :     0.000007s :  0.03%
    environ_conv                         :     0.000012s :  0.05%
    add_recomputation                    :     0.000080s :  0.36%
    cse_after_recomputation.cse          :     0.000032s :  0.14%
    auto_monad_reorder                   :     0.000019s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000012s :  0.06%
    opt_after_jit_grad                   :     0.000939s :  4.17%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000049s :  0.22%


Time group info:
------[substitution.]   0.000281    23
 1.05% :     0.000003s :      2: substitution.elim_not_effective
 0.65% :     0.000002s :      2: substitution.fold_const_symbol
 2.22% :     0.000006s :      3: substitution.graph_param_transform
86.60% :     0.000244s :      4: substitution.inline
 1.30% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.57% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000006s :      2: substitution.replace_old_param
 4.54% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005129     2
81.94% :     0.004203s :      1: type_inference.infer
18.06% :     0.000926s :      1: type_inference.specialize
------[replace.]   0.000079     6
67.69% :     0.000054s :      4: replace.inline
32.31% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000253     6
95.39% :     0.000241s :      4: match.inline
 4.61% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.42% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000002s :     10: predicate.addn_zero_filter
 2.10% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.24% :     0.000002s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.51% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.58% :     0.000001s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.12% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.71% :     0.000007s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.49% :     0.000002s :      6: predicate.less_batch_normalization
 1.37% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.55% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.90% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.80% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 2.04% :     0.000003s :      3: predicate.mutable_eliminate
 0.65% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.36% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.20% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.27% :     0.000010s :     53: predicate.switch_simplify
 1.09% :     0.000001s :     10: predicate.tile_eliminate
 1.27% :     0.000002s :     10: predicate.transpose_eliminate
 1.61% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.11% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.34% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.05% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000713    12
50.82% :     0.000362s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.07% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.11% :     0.000293s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.041476    72
 0.20% :     0.000084s :      1: add_recomputation
 0.17% :     0.000072s :      1: auto_monad
 0.05% :     0.000023s :      1: auto_monad_reorder
 0.84% :     0.000347s :      1: bootstrap
 0.08% :     0.000034s :      1: cconv
 0.03% :     0.000012s :      1: convert_after_rewriter
 0.15% :     0.000062s :      1: cse_after_recomputation
 0.04% :     0.000017s :      1: environ_conv
 0.05% :     0.000022s :      1: event_method
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
15.77% :     0.006542s :      1: jit_opt_a
29.77% :     0.012347s :      1: jit_opt_after_cconv
 0.14% :     0.000057s :      1: jit_opt_b
 1.11% :     0.000461s :      1: loop_unroll
 1.90% :     0.000788s :      1: mutable_eliminate
 2.31% :     0.000958s :     26: opt.transform.jit_opt_a
29.18% :     0.012101s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000027s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000036s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000041s :      4: opt.transform.symbol_engine_opt
 2.30% :     0.000953s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000010s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000044s :      1: py_interpret_to_execute
 0.04% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000036s :      1: remove_dup_value
 1.20% :     0.000499s :      1: renormalize.infer
 0.91% :     0.000379s :      1: renormalize.specialize
 0.04% :     0.000015s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000042s :      1: rewriter_after_opt_a
 0.19% :     0.000077s :      1: rewriter_before_opt_a
 0.23% :     0.000094s :      1: symbol_engine_optimizer
12.58% :     0.005218s :      1: type_inference


TotalTime = 0.0175019, [33]
[bootstrap]: 0.00026384
[type_inference]: 0.00431101
[event_method]: 1.75e-05
[auto_monad]: 6.292e-05
[graph_reusing]: 6.38998e-06
[pre_auto_parallel]: 2.72001e-06
[py_interpret_to_execute]: 2.988e-05
[rewriter_before_opt_a]: 7.263e-05
[expand_dump_flag]: 2.99001e-06
[jit_opt_a]: 0.0048568, [2]
    [Cycle 1]: 0.00140279, [27]
        [switch_simplify]: 4.885e-05
        [loop_unroll]: 2.878e-05
        [a_1]: 0.0005189
        [with_stream_mark]: 1.511e-05
        [recompute_prepare]: 7.3e-06
        [updatestate_depend_eliminate]: 3.41001e-06
        [updatestate_assign_eliminate]: 3.14001e-06
        [updatestate_loads_eliminate]: 2.94001e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 5.57001e-06
        [updatestate_useless_node_eliminater]: 4.96002e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 2.21998e-06
        [get_grad_eliminate_]: 5.04e-06
        [merge_forward]: 3.43e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.335e-05
        [j_node_and_user_rematch]: 8.40001e-06
        [meta_fg_expand]: 2.05002e-06
        [replace_old_param]: 8.67998e-06
        [inline_without_move]: 5.36002e-06
        [renormalize]: 0.00048461
        [add_forward_monad_depend]: 4.85001e-06
        [auto_monad_grad]: 1.86e-06
        [auto_monad_eliminator]: 1.249e-05
        [cse]: 2.931e-05
        [replace_applicator]: 1.099e-05
    [Cycle 2]: 0.00030879, [27]
        [switch_simplify]: 5.37999e-06
        [loop_unroll]: 4.80999e-06
        [a_1]: 8.729e-05
        [with_stream_mark]: 8.17e-06
        [recompute_prepare]: 4.69002e-06
        [updatestate_depend_eliminate]: 2.53e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 4.87998e-06
        [updatestate_useless_node_eliminater]: 4.71002e-06
        [accelerated_algorithm]: 4.87998e-06
        [meta_shard_fg_expand]: 1.16002e-06
        [get_grad_eliminate_]: 4.38001e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 1.55999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.253e-05
        [j_node_and_user_rematch]: 7.71999e-06
        [meta_fg_expand]: 1.50999e-06
        [replace_old_param]: 1.081e-05
        [inline_without_move]: 4.68999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 7.10017e-07
        [auto_monad_eliminator]: 4.87998e-06
        [cse]: 9.77001e-06
        [replace_applicator]: 4.87e-06
[py_interpret_to_execute_after_opt_a]: 8.84998e-06
[rewriter_after_opt_a]: 3.266e-05
[convert_after_rewriter]: 6.38e-06
[order_py_execute_after_rewriter]: 4.52998e-06
[mutable_eliminate]: 0.00046277
[jit_opt_b]: 4.746e-05, [1]
    [Cycle 1]: 4.106e-05, [2]
        [frontend_op_eliminate]: 1.523e-05
        [inline_after_opt_a]: 1.449e-05
[cconv]: 2.396e-05
[loop_unroll]: 0.00039355
[jit_opt_after_cconv]: 0.00013353, [1]
    [Cycle 1]: 0.00012701, [11]
        [c_1]: 2.012e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 5.40001e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 1.692e-05
        [call_graph_tuple_transform]: 1.8e-05
        [tuple_list_get_item_eliminator]: 5.00001e-06
        [none_parameter_eliminate]: 1.34003e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.14998e-06
[remove_dup_value]: 1.36e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 5.57001e-06
[add_recomputation]: 4.697e-05
[cse_after_recomputation]: 2.004e-05, [1]
    [Cycle 1]: 1.484e-05, [1]
        [cse]: 9.12001e-06
[auto_monad_reorder]: 1.46e-05
[get_jit_bprop_graph]: 1.52999e-06
[rewriter_after_jit_bprop_graph]: 4.95001e-06
[opt_after_jit_grad]: 0.00052849
[symbol_engine_optimizer]: 7.196e-05, [1]
    [Cycle 1]: 6.525e-05, [6]
        [build]: 3.63e-06
        [elim_shapecalc]: 7.83999e-06
        [elim_not_effective]: 1.247e-05
        [opt_reshape]: 5.28002e-06
        [fold_const_symbol]: 8.11002e-06
        [renormalize]: 3.89991e-07
[validate]: 3.192e-05
[backend_pass]: 1.02e-06
[task_emit]: 0.00582421
[execute]: 7.61001e-06
Sums
    bootstrap                            :     0.000264s :  1.91%
    type_inference                       :     0.004311s : 31.28%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000063s :  0.46%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.22%
    rewriter_before_opt_a                :     0.000073s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000606s :  4.40%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000485s :  3.52%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000039s :  0.28%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.24%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000463s :  3.36%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000394s :  2.86%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000047s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000528s :  3.83%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005824s : 42.26%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000192    23
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.79% :     0.000005s :      3: substitution.graph_param_transform
82.01% :     0.000158s :      4: substitution.inline
 1.51% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.12% :     0.000004s :      4: substitution.remove_not_recompute_node
 4.04% :     0.000008s :      2: substitution.replace_old_param
 5.90% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004250     2
82.45% :     0.003504s :      1: type_inference.infer
17.55% :     0.000746s :      1: type_inference.specialize
------[replace.]   0.000063     6
67.98% :     0.000043s :      4: replace.inline
32.02% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000165     6
93.83% :     0.000155s :      4: match.inline
 6.17% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.20% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.82% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000002s :     10: predicate.addn_zero_filter
 1.71% :     0.000002s :     10: predicate.arithmetic_simplify
 1.13% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
10.99% :     0.000014s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.14% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.03% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.06% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 4.98% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.99% :     0.000001s :      6: predicate.less_batch_normalization
 1.32% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.56% :     0.000002s :     15: predicate.load_eliminater
 1.54% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.62% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.71% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 1.49% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.32% :     0.000002s :     10: predicate.reduce_eliminate
 1.29% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.06% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.17% :     0.000003s :     16: predicate.switch_defer_inline
 1.96% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.98% :     0.000009s :     53: predicate.switch_simplify
 1.01% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000002s :     10: predicate.transpose_eliminate
 1.17% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.07% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.31% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.35% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.38% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000539    12
54.15% :     0.000292s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.31% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.54% :     0.000219s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018895    76
 0.26% :     0.000050s :      1: add_recomputation
 0.35% :     0.000067s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.50% :     0.000283s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
25.72% :     0.004860s :      1: jit_opt_a
 0.72% :     0.000136s :      1: jit_opt_after_cconv
 0.26% :     0.000050s :      1: jit_opt_b
 2.12% :     0.000401s :      1: loop_unroll
 2.49% :     0.000471s :      1: mutable_eliminate
 4.26% :     0.000805s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.85% :     0.000538s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.32% :     0.000249s :      1: renormalize.infer
 1.21% :     0.000229s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000036s :      1: rewriter_after_opt_a
 0.40% :     0.000075s :      1: rewriter_before_opt_a
 0.39% :     0.000074s :      1: symbol_engine_optimizer
30.88% :     0.005834s :      1: task_emit
22.88% :     0.004323s :      1: type_inference
 0.26% :     0.000049s :      1: validate


TotalTime = 0.0156418, [30]
[bootstrap]: 0.00032415
[type_inference]: 0.00606168
[event_method]: 2.044e-05
[auto_monad]: 6.546e-05
[graph_reusing]: 6.84999e-06
[pre_auto_parallel]: 2.85002e-06
[py_interpret_to_execute]: 3.674e-05
[rewriter_before_opt_a]: 7.744e-05
[expand_dump_flag]: 3.6e-06
[jit_opt_a]: 0.00702478, [2]
    [Cycle 1]: 0.00177826, [27]
        [switch_simplify]: 5.223e-05
        [loop_unroll]: 2.925e-05
        [a_1]: 0.00054351
        [with_stream_mark]: 1.84e-05
        [recompute_prepare]: 7.46999e-06
        [updatestate_depend_eliminate]: 3.24001e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 6.33e-06
        [updatestate_useless_node_eliminater]: 0.00010398
        [accelerated_algorithm]: 8.05e-06
        [meta_shard_fg_expand]: 2.49001e-06
        [get_grad_eliminate_]: 5.91e-06
        [merge_forward]: 5.02e-06
        [cell_reuse_recompute_pass]: 1.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.696e-05
        [j_node_and_user_rematch]: 9.69e-06
        [meta_fg_expand]: 2.45002e-06
        [replace_old_param]: 1.013e-05
        [inline_without_move]: 6.11e-06
        [renormalize]: 0.00070234
        [add_forward_monad_depend]: 5.41998e-06
        [auto_monad_grad]: 2.44001e-06
        [auto_monad_eliminator]: 1.345e-05
        [cse]: 2.806e-05
        [replace_applicator]: 1.134e-05
    [Cycle 2]: 0.00034158, [27]
        [switch_simplify]: 5.50001e-06
        [loop_unroll]: 5.14e-06
        [a_1]: 0.00010606
        [with_stream_mark]: 9.62999e-06
        [recompute_prepare]: 5.27001e-06
        [updatestate_depend_eliminate]: 2.82002e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.22999e-06
        [specialize_transform]: 4.99e-06
        [updatestate_useless_node_eliminater]: 4.81002e-06
        [accelerated_algorithm]: 5.15001e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 4.76002e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 1.54e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.313e-05
        [j_node_and_user_rematch]: 7.91001e-06
        [meta_fg_expand]: 1.60001e-06
        [replace_old_param]: 7.73999e-06
        [inline_without_move]: 4.74998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 5.18002e-06
        [cse]: 1.081e-05
        [replace_applicator]: 4.73001e-06
[py_interpret_to_execute_after_opt_a]: 9.11998e-06
[rewriter_after_opt_a]: 3.252e-05
[convert_after_rewriter]: 6.66e-06
[order_py_execute_after_rewriter]: 4.99e-06
[mutable_eliminate]: 0.0005143
[jit_opt_b]: 5.014e-05, [1]
    [Cycle 1]: 4.351e-05, [2]
        [frontend_op_eliminate]: 1.654e-05
        [inline_after_opt_a]: 1.508e-05
[cconv]: 2.453e-05
[loop_unroll]: 0.00039952
[jit_opt_after_cconv]: 0.00013628, [1]
    [Cycle 1]: 0.00012984, [11]
        [c_1]: 2.055e-05
        [parameter_eliminate]: 2.16e-06
        [updatestate_depend_eliminate]: 5.91e-06
        [updatestate_assign_eliminate]: 2.45002e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 1.608e-05
        [call_graph_tuple_transform]: 1.797e-05
        [tuple_list_get_item_eliminator]: 5.36998e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 5.64e-06
[remove_dup_value]: 1.415e-05
[partial_unused_args_eliminate]: 2.69999e-06
[environ_conv]: 5.22999e-06
[add_recomputation]: 5.104e-05
[cse_after_recomputation]: 2.046e-05, [1]
    [Cycle 1]: 1.52e-05, [1]
        [cse]: 9.29e-06
[auto_monad_reorder]: 1.517e-05
[get_jit_bprop_graph]: 1.40001e-06
[rewriter_after_jit_bprop_graph]: 4.67e-06
[opt_after_jit_grad]: 0.00043866
[symbol_engine_optimizer]: 7.235e-05, [1]
    [Cycle 1]: 6.66e-05, [6]
        [build]: 3.58e-06
        [elim_shapecalc]: 7.66001e-06
        [elim_not_effective]: 1.219e-05
        [opt_reshape]: 6.24999e-06
        [fold_const_symbol]: 9.12001e-06
        [renormalize]: 4.80009e-07
[validate]: 3.496e-05
Sums
    bootstrap                            :     0.000324s :  3.20%
    type_inference                       :     0.006062s : 59.75%
    event_method                         :     0.000020s :  0.20%
    auto_monad                           :     0.000065s :  0.65%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000037s :  0.36%
    rewriter_before_opt_a                :     0.000077s :  0.76%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000058s :  0.57%
    jit_opt_a.loop_unroll                :     0.000034s :  0.34%
    jit_opt_a.a_1                        :     0.000650s :  6.40%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000109s :  1.07%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.18%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000702s :  6.92%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.18%
    jit_opt_a.cse                        :     0.000039s :  0.38%
    jit_opt_a.replace_applicator         :     0.000016s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.09%
    rewriter_after_opt_a                 :     0.000033s :  0.32%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000514s :  5.07%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.15%
    cconv                                :     0.000025s :  0.24%
    loop_unroll                          :     0.000400s :  3.94%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000014s :  0.14%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000051s :  0.50%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.15%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000439s :  4.32%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.34%


Time group info:
------[substitution.]   0.000212    23
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.78% :     0.000002s :      2: substitution.fold_const_symbol
 2.29% :     0.000005s :      3: substitution.graph_param_transform
84.19% :     0.000179s :      4: substitution.inline
 1.44% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.35% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000004s :      2: substitution.replace_old_param
 6.01% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005983     2
81.54% :     0.004879s :      1: type_inference.infer
18.46% :     0.001105s :      1: type_inference.specialize
------[replace.]   0.000069     6
68.47% :     0.000047s :      4: replace.inline
31.53% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000188     6
93.78% :     0.000176s :      4: match.inline
 6.22% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.47% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.52% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.23% :     0.000001s :     10: predicate.addn_zero_filter
 2.32% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.48% :     0.000006s :     22: predicate.inline
 1.06% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.34% :     0.000002s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.96% :     0.000002s :     15: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.96% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.48% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.03% :     0.000002s :     15: predicate.partial_eliminate
 1.29% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.34% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.11% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.56% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.31% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.47% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.28% :     0.000002s :     10: predicate.transpose_eliminate
 1.55% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.80% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.81% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000806    12
49.86% :     0.000402s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.54% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.61% :     0.000367s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017347    72
 0.31% :     0.000054s :      1: add_recomputation
 0.40% :     0.000069s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 1.97% :     0.000342s :      1: bootstrap
 0.16% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000003s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
40.51% :     0.007027s :      1: jit_opt_a
 0.80% :     0.000139s :      1: jit_opt_after_cconv
 0.30% :     0.000053s :      1: jit_opt_b
 2.34% :     0.000407s :      1: loop_unroll
 3.01% :     0.000522s :      1: mutable_eliminate
 5.54% :     0.000961s :     26: opt.transform.jit_opt_a
 0.26% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.57% :     0.000445s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000039s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.09% :     0.000363s :      1: renormalize.infer
 1.91% :     0.000332s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000035s :      1: rewriter_after_opt_a
 0.47% :     0.000081s :      1: rewriter_before_opt_a
 0.43% :     0.000075s :      1: symbol_engine_optimizer
35.04% :     0.006078s :      1: type_inference


TotalTime = 0.0179731, [33]
[bootstrap]: 0.0002788
[type_inference]: 0.00445412
[event_method]: 1.583e-05
[auto_monad]: 6.288e-05
[graph_reusing]: 6.60002e-06
[pre_auto_parallel]: 2.33002e-06
[py_interpret_to_execute]: 3.364e-05
[rewriter_before_opt_a]: 7.327e-05
[expand_dump_flag]: 3.14001e-06
[jit_opt_a]: 0.00499294, [2]
    [Cycle 1]: 0.00143323, [27]
        [switch_simplify]: 5.279e-05
        [loop_unroll]: 3.091e-05
        [a_1]: 0.00051975
        [with_stream_mark]: 1.511e-05
        [recompute_prepare]: 8.2e-06
        [updatestate_depend_eliminate]: 3.75998e-06
        [updatestate_assign_eliminate]: 3.5e-06
        [updatestate_loads_eliminate]: 2.74999e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 5.99e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.26002e-06
        [meta_shard_fg_expand]: 2.21e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 3.39001e-06
        [cell_reuse_recompute_pass]: 1.04003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.361e-05
        [j_node_and_user_rematch]: 8.74e-06
        [meta_fg_expand]: 2.30002e-06
        [replace_old_param]: 8.60999e-06
        [inline_without_move]: 4.88001e-06
        [renormalize]: 0.0004898
        [add_forward_monad_depend]: 4.68001e-06
        [auto_monad_grad]: 1.88002e-06
        [auto_monad_eliminator]: 1.322e-05
        [cse]: 2.848e-05
        [replace_applicator]: 1.064e-05
    [Cycle 2]: 0.00032976, [27]
        [switch_simplify]: 5.52999e-06
        [loop_unroll]: 4.81997e-06
        [a_1]: 9.502e-05
        [with_stream_mark]: 8.50999e-06
        [recompute_prepare]: 5.04e-06
        [updatestate_depend_eliminate]: 2.70002e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 5.49e-06
        [updatestate_useless_node_eliminater]: 5.28002e-06
        [accelerated_algorithm]: 5.56e-06
        [meta_shard_fg_expand]: 1.21002e-06
        [get_grad_eliminate_]: 4.82e-06
        [merge_forward]: 2.75002e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.209e-05
        [j_node_and_user_rematch]: 7.82e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 7.76001e-06
        [inline_without_move]: 4.56002e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 5.33002e-06
        [cse]: 1.047e-05
        [replace_applicator]: 5.42001e-06
[py_interpret_to_execute_after_opt_a]: 8.83001e-06
[rewriter_after_opt_a]: 2.946e-05
[convert_after_rewriter]: 9.80002e-06
[order_py_execute_after_rewriter]: 5.46e-06
[mutable_eliminate]: 0.00049029
[jit_opt_b]: 4.953e-05, [1]
    [Cycle 1]: 4.266e-05, [2]
        [frontend_op_eliminate]: 1.642e-05
        [inline_after_opt_a]: 1.413e-05
[cconv]: 2.403e-05
[loop_unroll]: 0.00042075
[jit_opt_after_cconv]: 0.00013667, [1]
    [Cycle 1]: 0.00013046, [11]
        [c_1]: 2.059e-05
        [parameter_eliminate]: 2.32001e-06
        [updatestate_depend_eliminate]: 5.11002e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.691e-05
        [call_graph_tuple_transform]: 1.771e-05
        [tuple_list_get_item_eliminator]: 5.11002e-06
        [none_parameter_eliminate]: 1.38002e-06
        [renormalize]: 4.40021e-07
        [switch_simplify]: 5.17e-06
[remove_dup_value]: 1.358e-05
[partial_unused_args_eliminate]: 2.13002e-06
[environ_conv]: 5.05001e-06
[add_recomputation]: 4.704e-05
[cse_after_recomputation]: 1.98e-05, [1]
    [Cycle 1]: 1.463e-05, [1]
        [cse]: 8.67e-06
[auto_monad_reorder]: 1.442e-05
[get_jit_bprop_graph]: 1.67001e-06
[rewriter_after_jit_bprop_graph]: 4.80999e-06
[opt_after_jit_grad]: 0.00045145
[symbol_engine_optimizer]: 7.151e-05, [1]
    [Cycle 1]: 6.479e-05, [6]
        [build]: 3.41999e-06
        [elim_shapecalc]: 7.66999e-06
        [elim_not_effective]: 1.239e-05
        [opt_reshape]: 5.61e-06
        [fold_const_symbol]: 8.2e-06
        [renormalize]: 3.30008e-07
[validate]: 3.055e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00601855
[execute]: 8.25999e-06
Sums
    bootstrap                            :     0.000279s :  1.97%
    type_inference                       :     0.004454s : 31.51%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000063s :  0.44%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.24%
    rewriter_before_opt_a                :     0.000073s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.41%
    jit_opt_a.loop_unroll                :     0.000036s :  0.25%
    jit_opt_a.a_1                        :     0.000615s :  4.35%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000490s :  3.47%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000039s :  0.28%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000029s :  0.21%
    convert_after_rewriter               :     0.000010s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000490s :  3.47%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000421s :  2.98%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000451s :  3.19%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006019s : 42.57%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000184    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.77% :     0.000005s :      3: substitution.graph_param_transform
82.87% :     0.000152s :      4: substitution.inline
 1.66% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.21% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.32% :     0.000004s :      2: substitution.replace_old_param
 6.50% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004395     2
82.57% :     0.003629s :      1: type_inference.infer
17.43% :     0.000766s :      1: type_inference.specialize
------[replace.]   0.000068     6
67.95% :     0.000046s :      4: replace.inline
32.05% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000161     6
93.22% :     0.000150s :      4: match.inline
 6.78% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.43% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.05% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 1.99% :     0.000002s :     10: predicate.arithmetic_simplify
 1.36% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.60% :     0.000002s :     10: predicate.depend_value_elim
 1.24% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 5.30% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.32% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.45% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     15: predicate.partial_eliminate
 1.31% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.57% :     0.000002s :     10: predicate.reshape_eliminate
 1.33% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.45% :     0.000002s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.60% :     0.000003s :     16: predicate.switch_defer_inline
 2.31% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.81% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.58% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000530    12
52.04% :     0.000276s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.97% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.99% :     0.000228s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019386    76
 0.26% :     0.000050s :      1: add_recomputation
 0.34% :     0.000066s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.53% :     0.000296s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.06% :     0.000012s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
25.77% :     0.004996s :      1: jit_opt_a
 0.72% :     0.000139s :      1: jit_opt_after_cconv
 0.27% :     0.000052s :      1: jit_opt_b
 2.21% :     0.000428s :      1: loop_unroll
 2.57% :     0.000499s :      1: mutable_eliminate
 4.24% :     0.000821s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.37% :     0.000460s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.23% :     0.000238s :      1: renormalize.infer
 1.26% :     0.000245s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000032s :      1: rewriter_after_opt_a
 0.39% :     0.000076s :      1: rewriter_before_opt_a
 0.38% :     0.000074s :      1: symbol_engine_optimizer
31.11% :     0.006031s :      1: task_emit
23.04% :     0.004466s :      1: type_inference
 0.24% :     0.000047s :      1: validate


TotalTime = 0.0177883, [33]
[bootstrap]: 0.00025781
[type_inference]: 0.00427006
[event_method]: 1.675e-05
[auto_monad]: 0.00015994
[graph_reusing]: 5.66998e-06
[pre_auto_parallel]: 2.81999e-06
[py_interpret_to_execute]: 3.392e-05
[rewriter_before_opt_a]: 7.529e-05
[expand_dump_flag]: 3.03e-06
[jit_opt_a]: 0.0051068, [2]
    [Cycle 1]: 0.00147152, [27]
        [switch_simplify]: 5.086e-05
        [loop_unroll]: 3.085e-05
        [a_1]: 0.00051707
        [with_stream_mark]: 1.479e-05
        [recompute_prepare]: 7.53e-06
        [updatestate_depend_eliminate]: 3.55998e-06
        [updatestate_assign_eliminate]: 3.21001e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [parameter_eliminate]: 1.72001e-06
        [specialize_transform]: 5.98002e-06
        [updatestate_useless_node_eliminater]: 5.25999e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 2.29999e-06
        [get_grad_eliminate_]: 5.40001e-06
        [merge_forward]: 3.62002e-06
        [cell_reuse_recompute_pass]: 1.09003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.403e-05
        [j_node_and_user_rematch]: 9.14998e-06
        [meta_fg_expand]: 2.38998e-06
        [replace_old_param]: 8.90999e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 0.00054079
        [add_forward_monad_depend]: 5.58002e-06
        [auto_monad_grad]: 2.11e-06
        [auto_monad_eliminator]: 1.282e-05
        [cse]: 2.699e-05
        [replace_applicator]: 1.131e-05
    [Cycle 2]: 0.00032937, [27]
        [switch_simplify]: 5.61998e-06
        [loop_unroll]: 5.17999e-06
        [a_1]: 9.214e-05
        [with_stream_mark]: 8.90001e-06
        [recompute_prepare]: 4.99998e-06
        [updatestate_depend_eliminate]: 2.73e-06
        [updatestate_assign_eliminate]: 1.99999e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 1.09998e-06
        [specialize_transform]: 5.27999e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 5.30999e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.72998e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 1.79998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.291e-05
        [j_node_and_user_rematch]: 8.00999e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 7.91001e-06
        [inline_without_move]: 4.97e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.32999e-06
        [cse]: 1.035e-05
        [replace_applicator]: 5.22e-06
[py_interpret_to_execute_after_opt_a]: 1.014e-05
[rewriter_after_opt_a]: 3.01e-05
[convert_after_rewriter]: 7.02002e-06
[order_py_execute_after_rewriter]: 7.18998e-06
[mutable_eliminate]: 0.00047543
[jit_opt_b]: 4.969e-05, [1]
    [Cycle 1]: 4.293e-05, [2]
        [frontend_op_eliminate]: 1.643e-05
        [inline_after_opt_a]: 1.513e-05
[cconv]: 2.348e-05
[loop_unroll]: 0.00040728
[jit_opt_after_cconv]: 0.0001363, [1]
    [Cycle 1]: 0.00012968, [11]
        [c_1]: 2.069e-05
        [parameter_eliminate]: 2.47001e-06
        [updatestate_depend_eliminate]: 5.27999e-06
        [updatestate_assign_eliminate]: 2.33998e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.669e-05
        [call_graph_tuple_transform]: 1.85e-05
        [tuple_list_get_item_eliminator]: 5.61003e-06
        [none_parameter_eliminate]: 1.27e-06
        [renormalize]: 4.40021e-07
        [switch_simplify]: 5.27001e-06
[remove_dup_value]: 1.286e-05
[partial_unused_args_eliminate]: 2.79999e-06
[environ_conv]: 5.66003e-06
[add_recomputation]: 4.869e-05
[cse_after_recomputation]: 2.071e-05, [1]
    [Cycle 1]: 1.483e-05, [1]
        [cse]: 9.13002e-06
[auto_monad_reorder]: 1.463e-05
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 4.65999e-06
[opt_after_jit_grad]: 0.00044779
[symbol_engine_optimizer]: 7.376e-05, [1]
    [Cycle 1]: 6.739e-05, [6]
        [build]: 3.73001e-06
        [elim_shapecalc]: 7.6e-06
        [elim_not_effective]: 1.302e-05
        [opt_reshape]: 5.64e-06
        [fold_const_symbol]: 9.01002e-06
        [renormalize]: 4.7998e-07
[validate]: 3.039e-05
[backend_pass]: 8.00006e-07
[task_emit]: 0.00586358
[execute]: 6.33998e-06
Sums
    bootstrap                            :     0.000258s :  1.86%
    type_inference                       :     0.004270s : 30.73%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000160s :  1.15%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.24%
    rewriter_before_opt_a                :     0.000075s :  0.54%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.41%
    jit_opt_a.loop_unroll                :     0.000036s :  0.26%
    jit_opt_a.a_1                        :     0.000609s :  4.38%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000541s :  3.89%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000037s :  0.27%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000030s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000007s :  0.05%
    mutable_eliminate                    :     0.000475s :  3.42%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000407s :  2.93%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000049s :  0.35%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000448s :  3.22%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000030s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005864s : 42.20%
    execute                              :     0.000006s :  0.05%


TotalTime = 0.0114361, [30]
[bootstrap]: 0.00027017
[type_inference]: 0.00419174
[event_method]: 1.565e-05
[auto_monad]: 5.123e-05
[graph_reusing]: 5.35999e-06
[pre_auto_parallel]: 2.44999e-06
[py_interpret_to_execute]: 2.797e-05
[rewriter_before_opt_a]: 6.41e-05
[expand_dump_flag]: 2.67001e-06
[jit_opt_a]: 0.00488067, [2]
    [Cycle 1]: 0.00139617, [27]
        [switch_simplify]: 4.607e-05
        [loop_unroll]: 2.873e-05
        [a_1]: 0.00048713
        [with_stream_mark]: 1.431e-05
        [recompute_prepare]: 6.93e-06
        [updatestate_depend_eliminate]: 3.2e-06
        [updatestate_assign_eliminate]: 2.57001e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.07998e-06
        [specialize_transform]: 5.86998e-06
        [updatestate_useless_node_eliminater]: 5.15999e-06
        [accelerated_algorithm]: 5.97999e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 5.14e-06
        [merge_forward]: 2.63e-06
        [cell_reuse_recompute_pass]: 9.89996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.174e-05
        [j_node_and_user_rematch]: 8.62998e-06
        [meta_fg_expand]: 2.14999e-06
        [replace_old_param]: 7.65e-06
        [inline_without_move]: 5.31998e-06
        [renormalize]: 0.00053467
        [add_forward_monad_depend]: 4.41002e-06
        [auto_monad_grad]: 1.64e-06
        [auto_monad_eliminator]: 1.041e-05
        [cse]: 1.709e-05
        [replace_applicator]: 1.033e-05
    [Cycle 2]: 0.0003148, [27]
        [switch_simplify]: 5.78002e-06
        [loop_unroll]: 4.77998e-06
        [a_1]: 8.939e-05
        [with_stream_mark]: 8.15e-06
        [recompute_prepare]: 5.32999e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 1.92001e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.72e-06
        [accelerated_algorithm]: 4.90001e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 4.35999e-06
        [merge_forward]: 2.27001e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.142e-05
        [j_node_and_user_rematch]: 7.72002e-06
        [meta_fg_expand]: 1.47999e-06
        [replace_old_param]: 7.36999e-06
        [inline_without_move]: 4.63001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.06002e-06
        [auto_monad_grad]: 6.29982e-07
        [auto_monad_eliminator]: 4.99e-06
        [cse]: 1.063e-05
        [replace_applicator]: 5.04003e-06
[py_interpret_to_execute_after_opt_a]: 8.15999e-06
[rewriter_after_opt_a]: 2.662e-05
[convert_after_rewriter]: 5.30001e-06
[order_py_execute_after_rewriter]: 4.39002e-06
[mutable_eliminate]: 0.0004608
[jit_opt_b]: 7.097e-05, [1]
    [Cycle 1]: 6.431e-05, [2]
        [frontend_op_eliminate]: 3.667e-05
        [inline_after_opt_a]: 1.515e-05
[cconv]: 1.86e-05
[loop_unroll]: 0.00041649
[jit_opt_after_cconv]: 0.00013131, [1]
    [Cycle 1]: 0.00012503, [11]
        [c_1]: 1.998e-05
        [parameter_eliminate]: 2.27001e-06
        [updatestate_depend_eliminate]: 5.15999e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.576e-05
        [call_graph_tuple_transform]: 1.7e-05
        [tuple_list_get_item_eliminator]: 5.15999e-06
        [none_parameter_eliminate]: 1.07e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.15999e-06
[remove_dup_value]: 1.001e-05
[partial_unused_args_eliminate]: 1.69e-06
[environ_conv]: 3.76999e-06
[add_recomputation]: 3.345e-05
[cse_after_recomputation]: 1.95e-05, [1]
    [Cycle 1]: 1.429e-05, [1]
        [cse]: 8.73001e-06
[auto_monad_reorder]: 1.112e-05
[get_jit_bprop_graph]: 1.39e-06
[rewriter_after_jit_bprop_graph]: 4.70999e-06
[opt_after_jit_grad]: 0.0004471
[symbol_engine_optimizer]: 7.141e-05, [1]
    [Cycle 1]: 6.453e-05, [6]
        [build]: 3.06999e-06
        [elim_shapecalc]: 7.33999e-06
        [elim_not_effective]: 1.239e-05
        [opt_reshape]: 5.81e-06
        [fold_const_symbol]: 8.54e-06
        [renormalize]: 4.19997e-07
[validate]: 2.699e-05
Sums
    bootstrap                            :     0.000270s :  3.50%
    type_inference                       :     0.004192s : 54.31%
    event_method                         :     0.000016s :  0.20%
    auto_monad                           :     0.000051s :  0.66%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000028s :  0.36%
    rewriter_before_opt_a                :     0.000064s :  0.83%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000052s :  0.67%
    jit_opt_a.loop_unroll                :     0.000034s :  0.43%
    jit_opt_a.a_1                        :     0.000577s :  7.47%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.12%
    jit_opt_a.merge_forward              :     0.000005s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000535s :  6.93%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.20%
    jit_opt_a.cse                        :     0.000028s :  0.36%
    jit_opt_a.replace_applicator         :     0.000015s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000027s :  0.34%
    convert_after_rewriter               :     0.000005s :  0.07%
    order_py_execute_after_rewriter      :     0.000004s :  0.06%
    mutable_eliminate                    :     0.000461s :  5.97%
    jit_opt_b.frontend_op_eliminate      :     0.000037s :  0.48%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.20%
    cconv                                :     0.000019s :  0.24%
    loop_unroll                          :     0.000416s :  5.40%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000010s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000033s :  0.43%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000011s :  0.14%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000447s :  5.79%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000027s :  0.35%


Time group info:
------[substitution.]   0.000191    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.71% :     0.000005s :      3: substitution.graph_param_transform
83.72% :     0.000160s :      4: substitution.inline
 1.58% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.27% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.05% :     0.000004s :      2: substitution.replace_old_param
 6.00% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004215     2
81.89% :     0.003452s :      1: type_inference.infer
18.11% :     0.000763s :      1: type_inference.specialize
------[replace.]   0.000063     6
66.08% :     0.000041s :      4: replace.inline
33.92% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000167     6
93.92% :     0.000157s :      4: match.inline
 6.08% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.29% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 1.91% :     0.000002s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.70% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.65% :     0.000007s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.58% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.23% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.75% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     15: predicate.partial_eliminate
 1.36% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.94% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.41% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.62% :     0.000002s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 1.13% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.61% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.95% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000001s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.74% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000537    12
48.86% :     0.000262s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.79% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.35% :     0.000233s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019255    76
 0.27% :     0.000052s :      1: add_recomputation
 0.85% :     0.000164s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.40% :     0.000270s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.54% :     0.005109s :      1: jit_opt_a
 0.72% :     0.000139s :      1: jit_opt_after_cconv
 0.27% :     0.000052s :      1: jit_opt_b
 2.15% :     0.000415s :      1: loop_unroll
 2.51% :     0.000484s :      1: mutable_eliminate
 4.24% :     0.000817s :     26: opt.transform.jit_opt_a
 0.24% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.37% :     0.000456s :      1: opt_after_jit_grad
 0.05% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.43% :     0.000275s :      1: renormalize.infer
 1.35% :     0.000259s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000033s :      1: rewriter_after_opt_a
 0.41% :     0.000078s :      1: rewriter_before_opt_a
 0.40% :     0.000076s :      1: symbol_engine_optimizer
30.50% :     0.005873s :      1: task_emit
22.25% :     0.004283s :      1: type_inference
 0.24% :     0.000047s :      1: validate


Time group info:
------[substitution.]   0.000165    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.49% :     0.000004s :      3: substitution.graph_param_transform
83.66% :     0.000138s :      4: substitution.inline
 1.75% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.46% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.89% :     0.000003s :      2: substitution.replace_old_param
 6.05% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004140     2
78.90% :     0.003266s :      1: type_inference.infer
21.10% :     0.000874s :      1: type_inference.specialize
------[replace.]   0.000061     6
67.15% :     0.000041s :      4: replace.inline
32.85% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000144     6
93.90% :     0.000136s :      4: match.inline
 6.10% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.56% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.20% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.15% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000001s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.46% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.21% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.43% :     0.000000s :      3: predicate.graph_param_transform
 5.02% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000001s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.66% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.41% :     0.000003s :     15: predicate.partial_eliminate
 1.28% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.05% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.51% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.62% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.43% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.66% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000632    12
48.09% :     0.000304s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.69% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.23% :     0.000299s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012829    72
 0.28% :     0.000036s :      1: add_recomputation
 0.42% :     0.000054s :      1: auto_monad
 0.11% :     0.000014s :      1: auto_monad_reorder
 2.21% :     0.000284s :      1: bootstrap
 0.17% :     0.000022s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000006s :      1: environ_conv
 0.15% :     0.000019s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.07% :     0.004883s :      1: jit_opt_a
 1.05% :     0.000134s :      1: jit_opt_after_cconv
 0.58% :     0.000074s :      1: jit_opt_b
 3.31% :     0.000424s :      1: loop_unroll
 3.65% :     0.000469s :      1: mutable_eliminate
 6.01% :     0.000771s :     26: opt.transform.jit_opt_a
 0.34% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.35% :     0.000045s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.55% :     0.000455s :      1: opt_after_jit_grad
 0.05% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000031s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000012s :      1: remove_dup_value
 2.10% :     0.000269s :      1: renormalize.infer
 2.02% :     0.000259s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000029s :      1: rewriter_after_opt_a
 0.52% :     0.000067s :      1: rewriter_before_opt_a
 0.58% :     0.000074s :      1: symbol_engine_optimizer
32.76% :     0.004202s :      1: type_inference


TotalTime = 0.0113708, [30]
[bootstrap]: 0.00030064
[type_inference]: 0.00400099
[event_method]: 1.669e-05
[auto_monad]: 4.473e-05
[graph_reusing]: 3.81999e-06
[pre_auto_parallel]: 1.78002e-06
[py_interpret_to_execute]: 2.832e-05
[rewriter_before_opt_a]: 6.316e-05
[expand_dump_flag]: 2.98e-06
[jit_opt_a]: 0.00480266, [2]
    [Cycle 1]: 0.00141193, [27]
        [switch_simplify]: 4.612e-05
        [loop_unroll]: 2.833e-05
        [a_1]: 0.00051098
        [with_stream_mark]: 1.237e-05
        [recompute_prepare]: 7.35998e-06
        [updatestate_depend_eliminate]: 2.86999e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 6.21e-06
        [updatestate_useless_node_eliminater]: 5.37999e-06
        [accelerated_algorithm]: 5.61003e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 5.31998e-06
        [merge_forward]: 2.28998e-06
        [cell_reuse_recompute_pass]: 1.11002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.167e-05
        [j_node_and_user_rematch]: 8.94998e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 7.63999e-06
        [inline_without_move]: 5.42999e-06
        [renormalize]: 0.00051273
        [add_forward_monad_depend]: 4.30999e-06
        [auto_monad_grad]: 1.47001e-06
        [auto_monad_eliminator]: 1.126e-05
        [cse]: 1.726e-05
        [replace_applicator]: 1.083e-05
    [Cycle 2]: 0.00031014, [27]
        [switch_simplify]: 5.38002e-06
        [loop_unroll]: 4.89e-06
        [a_1]: 8.621e-05
        [with_stream_mark]: 7.98001e-06
        [recompute_prepare]: 5.31998e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 2.01003e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 9.20001e-07
        [specialize_transform]: 4.68999e-06
        [updatestate_useless_node_eliminater]: 4.63999e-06
        [accelerated_algorithm]: 5.07999e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.67e-06
        [merge_forward]: 2.22999e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.157e-05
        [j_node_and_user_rematch]: 7.56999e-06
        [meta_fg_expand]: 1.57999e-06
        [replace_old_param]: 7.25998e-06
        [inline_without_move]: 4.63999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.16002e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 4.95001e-06
        [cse]: 1.008e-05
        [replace_applicator]: 4.97e-06
[py_interpret_to_execute_after_opt_a]: 7.26001e-06
[rewriter_after_opt_a]: 2.637e-05
[convert_after_rewriter]: 7.01999e-06
[order_py_execute_after_rewriter]: 4.72e-06
[mutable_eliminate]: 0.00049265
[jit_opt_b]: 4.969e-05, [1]
    [Cycle 1]: 4.352e-05, [2]
        [frontend_op_eliminate]: 1.687e-05
        [inline_after_opt_a]: 1.483e-05
[cconv]: 1.677e-05
[loop_unroll]: 0.00053218
[jit_opt_after_cconv]: 0.0001598, [1]
    [Cycle 1]: 0.00015363, [11]
        [c_1]: 2.084e-05
        [parameter_eliminate]: 2.54001e-06
        [updatestate_depend_eliminate]: 5.34e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.731e-05
        [call_graph_tuple_transform]: 1.667e-05
        [tuple_list_get_item_eliminator]: 5.10999e-06
        [none_parameter_eliminate]: 1.11002e-06
        [renormalize]: 4.10015e-07
        [switch_simplify]: 5.89e-06
[remove_dup_value]: 1.021e-05
[partial_unused_args_eliminate]: 1.74e-06
[environ_conv]: 3.56001e-06
[add_recomputation]: 3.739e-05
[cse_after_recomputation]: 2.108e-05, [1]
    [Cycle 1]: 1.536e-05, [1]
        [cse]: 9.71e-06
[auto_monad_reorder]: 1.104e-05
[get_jit_bprop_graph]: 1.48002e-06
[rewriter_after_jit_bprop_graph]: 5.10001e-06
[opt_after_jit_grad]: 0.00044276
[symbol_engine_optimizer]: 7.382e-05, [1]
    [Cycle 1]: 6.649e-05, [6]
        [build]: 3.39001e-06
        [elim_shapecalc]: 7.6e-06
        [elim_not_effective]: 1.327e-05
        [opt_reshape]: 6.10002e-06
        [fold_const_symbol]: 8.17e-06
        [renormalize]: 3.59985e-07
[validate]: 2.886e-05
Sums
    bootstrap                            :     0.000301s :  3.91%
    type_inference                       :     0.004001s : 52.09%
    event_method                         :     0.000017s :  0.22%
    auto_monad                           :     0.000045s :  0.58%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.37%
    rewriter_before_opt_a                :     0.000063s :  0.82%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000052s :  0.67%
    jit_opt_a.loop_unroll                :     0.000033s :  0.43%
    jit_opt_a.a_1                        :     0.000597s :  7.77%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000005s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000513s :  6.68%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.21%
    jit_opt_a.cse                        :     0.000027s :  0.36%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.09%
    rewriter_after_opt_a                 :     0.000026s :  0.34%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000493s :  6.41%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000017s :  0.22%
    loop_unroll                          :     0.000532s :  6.93%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.08%
    remove_dup_value                     :     0.000010s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000037s :  0.49%
    cse_after_recomputation.cse          :     0.000010s :  0.13%
    auto_monad_reorder                   :     0.000011s :  0.14%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000443s :  5.76%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.38%


Time group info:
------[substitution.]   0.000166    23
 1.14% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.64% :     0.000004s :      3: substitution.graph_param_transform
83.66% :     0.000139s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.79% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.12% :     0.000004s :      2: substitution.replace_old_param
 5.52% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003950     2
80.38% :     0.003175s :      1: type_inference.infer
19.62% :     0.000775s :      1: type_inference.specialize
------[replace.]   0.000078     6
74.38% :     0.000058s :      4: replace.inline
25.62% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000144     6
94.43% :     0.000136s :      4: match.inline
 5.57% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.53% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.44% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 2.01% :     0.000002s :     10: predicate.addn_zero_filter
 2.05% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.22% :     0.000001s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.54% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.02% :     0.000006s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.63% :     0.000002s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.61% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.87% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.36% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.51% :     0.000002s :      3: predicate.mutable_eliminate
 0.60% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000002s :     15: predicate.partial_eliminate
 1.35% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.61% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.53% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.10% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.52% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.66% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.29% :     0.000008s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.81% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.13% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000571    12
48.22% :     0.000275s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.50% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.28% :     0.000259s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012746    72
 0.32% :     0.000040s :      1: add_recomputation
 0.38% :     0.000048s :      1: auto_monad
 0.11% :     0.000014s :      1: auto_monad_reorder
 2.45% :     0.000313s :      1: bootstrap
 0.15% :     0.000019s :      1: cconv
 0.08% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000006s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000006s :      1: graph_reusing
37.70% :     0.004805s :      1: jit_opt_a
 1.28% :     0.000163s :      1: jit_opt_after_cconv
 0.41% :     0.000052s :      1: jit_opt_b
 4.24% :     0.000541s :      1: loop_unroll
 3.93% :     0.000501s :      1: mutable_eliminate
 6.22% :     0.000793s :     26: opt.transform.jit_opt_a
 0.35% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000025s :      4: opt.transform.jit_opt_b
 0.10% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.13% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.53% :     0.000450s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000031s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000013s :      1: remove_dup_value
 1.96% :     0.000250s :      1: renormalize.infer
 2.01% :     0.000257s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.39% :     0.000050s :      1: rewriter_after_opt_a
 0.52% :     0.000066s :      1: rewriter_before_opt_a
 0.60% :     0.000076s :      1: symbol_engine_optimizer
31.46% :     0.004010s :      1: type_inference


TotalTime = 0.01764, [33]
[bootstrap]: 0.00026761
[type_inference]: 0.00431499
[event_method]: 1.666e-05
[auto_monad]: 6.104e-05
[graph_reusing]: 6.01e-06
[pre_auto_parallel]: 2.39999e-06
[py_interpret_to_execute]: 3.147e-05
[rewriter_before_opt_a]: 7.121e-05
[expand_dump_flag]: 2.92002e-06
[jit_opt_a]: 0.00490552, [2]
    [Cycle 1]: 0.00141829, [27]
        [switch_simplify]: 5.203e-05
        [loop_unroll]: 2.944e-05
        [a_1]: 0.00050861
        [with_stream_mark]: 1.556e-05
        [recompute_prepare]: 7.41001e-06
        [updatestate_depend_eliminate]: 3.75e-06
        [updatestate_assign_eliminate]: 2.91999e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [parameter_eliminate]: 2.36e-06
        [specialize_transform]: 5.87001e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.47999e-06
        [meta_shard_fg_expand]: 2.31e-06
        [get_grad_eliminate_]: 5.40001e-06
        [merge_forward]: 4.02e-06
        [cell_reuse_recompute_pass]: 1.28002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.491e-05
        [j_node_and_user_rematch]: 9.12999e-06
        [meta_fg_expand]: 2.24999e-06
        [replace_old_param]: 8.92e-06
        [inline_without_move]: 5.43002e-06
        [renormalize]: 0.00050199
        [add_forward_monad_depend]: 4.72e-06
        [auto_monad_grad]: 1.77001e-06
        [auto_monad_eliminator]: 1.329e-05
        [cse]: 2.776e-05
        [replace_applicator]: 1.105e-05
    [Cycle 2]: 0.00031756, [27]
        [switch_simplify]: 5.37999e-06
        [loop_unroll]: 4.74e-06
        [a_1]: 8.918e-05
        [with_stream_mark]: 8.77e-06
        [recompute_prepare]: 5.15999e-06
        [updatestate_depend_eliminate]: 2.76999e-06
        [updatestate_assign_eliminate]: 2.08998e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 4.95999e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 4.97999e-06
        [meta_shard_fg_expand]: 1.13001e-06
        [get_grad_eliminate_]: 4.48001e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.61998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.199e-05
        [j_node_and_user_rematch]: 7.5e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 6.86001e-06
        [inline_without_move]: 4.57e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.11002e-06
        [auto_monad_grad]: 7.80012e-07
        [auto_monad_eliminator]: 5.19e-06
        [cse]: 1.001e-05
        [replace_applicator]: 4.76002e-06
[py_interpret_to_execute_after_opt_a]: 8.47e-06
[rewriter_after_opt_a]: 3.092e-05
[convert_after_rewriter]: 6.77002e-06
[order_py_execute_after_rewriter]: 5.01997e-06
[mutable_eliminate]: 0.00047095
[jit_opt_b]: 4.8e-05, [1]
    [Cycle 1]: 4.14e-05, [2]
        [frontend_op_eliminate]: 1.628e-05
        [inline_after_opt_a]: 1.426e-05
[cconv]: 2.343e-05
[loop_unroll]: 0.00041922
[jit_opt_after_cconv]: 0.00013647, [1]
    [Cycle 1]: 0.00012995, [11]
        [c_1]: 2.064e-05
        [parameter_eliminate]: 2.34001e-06
        [updatestate_depend_eliminate]: 5.46e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [cse]: 1.62e-05
        [call_graph_tuple_transform]: 1.824e-05
        [tuple_list_get_item_eliminator]: 5.42999e-06
        [none_parameter_eliminate]: 1.94e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 5.32999e-06
[remove_dup_value]: 1.386e-05
[partial_unused_args_eliminate]: 2.40002e-06
[environ_conv]: 4.87e-06
[add_recomputation]: 4.697e-05
[cse_after_recomputation]: 2.022e-05, [1]
    [Cycle 1]: 1.501e-05, [1]
        [cse]: 9.45001e-06
[auto_monad_reorder]: 1.537e-05
[get_jit_bprop_graph]: 1.39e-06
[rewriter_after_jit_bprop_graph]: 4.51002e-06
[opt_after_jit_grad]: 0.00048528
[symbol_engine_optimizer]: 8.033e-05, [1]
    [Cycle 1]: 6.666e-05, [6]
        [build]: 3.25998e-06
        [elim_shapecalc]: 8.03999e-06
        [elim_not_effective]: 1.262e-05
        [opt_reshape]: 5.86e-06
        [fold_const_symbol]: 8.85999e-06
        [renormalize]: 3.80009e-07
[validate]: 3.13e-05
[backend_pass]: 8.2e-07
[task_emit]: 0.00592073
[execute]: 7.26999e-06
Sums
    bootstrap                            :     0.000268s :  1.93%
    type_inference                       :     0.004315s : 31.07%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000061s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.23%
    rewriter_before_opt_a                :     0.000071s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000598s :  4.30%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000502s :  3.61%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000038s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000471s :  3.39%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000419s :  3.02%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000485s :  3.49%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005921s : 42.63%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000180    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.82% :     0.000001s :      2: substitution.fold_const_symbol
 2.80% :     0.000005s :      3: substitution.graph_param_transform
82.85% :     0.000149s :      4: substitution.inline
 1.59% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.32% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.89% :     0.000003s :      2: substitution.replace_old_param
 6.72% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004253     2
81.54% :     0.003468s :      1: type_inference.infer
18.46% :     0.000785s :      1: type_inference.specialize
------[replace.]   0.000066     6
65.79% :     0.000043s :      4: replace.inline
34.21% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000158     6
93.15% :     0.000147s :      4: match.inline
 6.85% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.26% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.43% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 1.96% :     0.000002s :     10: predicate.arithmetic_simplify
 1.36% :     0.000002s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.44% :     0.000002s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.32% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.44% :     0.000001s :      3: predicate.graph_param_transform
 5.44% :     0.000006s :     22: predicate.inline
 1.02% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.46% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.80% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.38% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.84% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.09% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.67% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.93% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000516    12
49.42% :     0.000255s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.28% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.30% :     0.000234s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019053    76
 0.26% :     0.000050s :      1: add_recomputation
 0.34% :     0.000064s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.47% :     0.000280s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.76% :     0.004908s :      1: jit_opt_a
 0.73% :     0.000139s :      1: jit_opt_after_cconv
 0.26% :     0.000050s :      1: jit_opt_b
 2.24% :     0.000427s :      1: loop_unroll
 2.52% :     0.000479s :      1: mutable_eliminate
 4.21% :     0.000801s :     26: opt.transform.jit_opt_a
 0.24% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.59% :     0.000493s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 1.29% :     0.000245s :      1: renormalize.infer
 1.31% :     0.000250s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.39% :     0.000074s :      1: rewriter_before_opt_a
 0.44% :     0.000083s :      1: symbol_engine_optimizer
31.13% :     0.005931s :      1: task_emit
22.72% :     0.004328s :      1: type_inference
 0.25% :     0.000048s :      1: validate


TotalTime = 0.0116267, [30]
[bootstrap]: 0.00025591
[type_inference]: 0.00411773
[event_method]: 1.529e-05
[auto_monad]: 4.626e-05
[graph_reusing]: 4.50999e-06
[pre_auto_parallel]: 1.71e-06
[py_interpret_to_execute]: 3.127e-05
[rewriter_before_opt_a]: 6.349e-05
[expand_dump_flag]: 2.04e-06
[jit_opt_a]: 0.00506973, [2]
    [Cycle 1]: 0.00149444, [27]
        [switch_simplify]: 4.561e-05
        [loop_unroll]: 2.956e-05
        [a_1]: 0.00051089
        [with_stream_mark]: 1.287e-05
        [recompute_prepare]: 7.36999e-06
        [updatestate_depend_eliminate]: 3.19001e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 6.28e-06
        [updatestate_useless_node_eliminater]: 5.62999e-06
        [accelerated_algorithm]: 5.92999e-06
        [meta_shard_fg_expand]: 1.35001e-06
        [get_grad_eliminate_]: 5.43002e-06
        [merge_forward]: 2.66999e-06
        [cell_reuse_recompute_pass]: 8.40024e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.104e-05
        [j_node_and_user_rematch]: 1.01e-05
        [meta_fg_expand]: 1.74998e-06
        [replace_old_param]: 8.70001e-06
        [inline_without_move]: 6.33002e-06
        [renormalize]: 0.00060028
        [add_forward_monad_depend]: 4.82998e-06
        [auto_monad_grad]: 1.17e-06
        [auto_monad_eliminator]: 1.15e-05
        [cse]: 1.635e-05
        [replace_applicator]: 1.153e-05
    [Cycle 2]: 0.00031427, [27]
        [switch_simplify]: 5.57001e-06
        [loop_unroll]: 4.65999e-06
        [a_1]: 8.834e-05
        [with_stream_mark]: 9.02999e-06
        [recompute_prepare]: 4.79e-06
        [updatestate_depend_eliminate]: 2.38998e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 5.01002e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 4.97e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.90999e-06
        [merge_forward]: 2.37001e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.186e-05
        [j_node_and_user_rematch]: 7.65e-06
        [meta_fg_expand]: 1.59998e-06
        [replace_old_param]: 7.42002e-06
        [inline_without_move]: 4.65001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.19003e-06
        [auto_monad_grad]: 7.90023e-07
        [auto_monad_eliminator]: 4.88001e-06
        [cse]: 1.002e-05
        [replace_applicator]: 4.77998e-06
[py_interpret_to_execute_after_opt_a]: 8.89e-06
[rewriter_after_opt_a]: 2.6e-05
[convert_after_rewriter]: 5.67001e-06
[order_py_execute_after_rewriter]: 4.77e-06
[mutable_eliminate]: 0.0005767
[jit_opt_b]: 5.02e-05, [1]
    [Cycle 1]: 4.346e-05, [2]
        [frontend_op_eliminate]: 1.677e-05
        [inline_after_opt_a]: 1.423e-05
[cconv]: 1.793e-05
[loop_unroll]: 0.00041039
[jit_opt_after_cconv]: 0.00013276, [1]
    [Cycle 1]: 0.00012646, [11]
        [c_1]: 2.016e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 5.05999e-06
        [updatestate_assign_eliminate]: 2.23998e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [cse]: 1.646e-05
        [call_graph_tuple_transform]: 1.705e-05
        [tuple_list_get_item_eliminator]: 5.61e-06
        [none_parameter_eliminate]: 9.70002e-07
        [renormalize]: 2.00002e-07
        [switch_simplify]: 5.25001e-06
[remove_dup_value]: 9.41e-06
[partial_unused_args_eliminate]: 1.54e-06
[environ_conv]: 3.75998e-06
[add_recomputation]: 3.385e-05
[cse_after_recomputation]: 2.055e-05, [1]
    [Cycle 1]: 1.537e-05, [1]
        [cse]: 9.17999e-06
[auto_monad_reorder]: 1.026e-05
[get_jit_bprop_graph]: 1.66e-06
[rewriter_after_jit_bprop_graph]: 4.45e-06
[opt_after_jit_grad]: 0.00044462
[symbol_engine_optimizer]: 7.066e-05, [1]
    [Cycle 1]: 6.497e-05, [6]
        [build]: 3.06999e-06
        [elim_shapecalc]: 7.3e-06
        [elim_not_effective]: 1.227e-05
        [opt_reshape]: 5.70001e-06
        [fold_const_symbol]: 8.28999e-06
        [renormalize]: 6.00005e-07
[validate]: 2.925e-05
Sums
    bootstrap                            :     0.000256s :  3.28%
    type_inference                       :     0.004118s : 52.74%
    event_method                         :     0.000015s :  0.20%
    auto_monad                           :     0.000046s :  0.59%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.40%
    rewriter_before_opt_a                :     0.000063s :  0.81%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000051s :  0.66%
    jit_opt_a.loop_unroll                :     0.000034s :  0.44%
    jit_opt_a.a_1                        :     0.000599s :  7.67%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000005s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.21%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000600s :  7.69%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.21%
    jit_opt_a.cse                        :     0.000026s :  0.34%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000026s :  0.33%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000577s :  7.39%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.18%
    cconv                                :     0.000018s :  0.23%
    loop_unroll                          :     0.000410s :  5.26%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000009s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000034s :  0.43%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000010s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.06%
    opt_after_jit_grad                   :     0.000445s :  5.69%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000029s :  0.37%


Time group info:
------[substitution.]   0.000162    23
 1.11% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.64% :     0.000004s :      3: substitution.graph_param_transform
83.33% :     0.000135s :      4: substitution.inline
 1.71% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.57% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.98% :     0.000003s :      2: substitution.replace_old_param
 5.98% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004063     2
78.26% :     0.003180s :      1: type_inference.infer
21.74% :     0.000883s :      1: type_inference.specialize
------[replace.]   0.000067     6
67.61% :     0.000045s :      4: replace.inline
32.39% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000141     6
93.97% :     0.000132s :      4: match.inline
 6.03% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.58% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.69% :     0.000002s :     10: predicate.addn_zero_filter
 2.41% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000001s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.06% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.50% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.18% :     0.000006s :     22: predicate.inline
 1.03% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.69% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.55% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.70% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 1.47% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.63% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.88% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.01% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.16% :     0.000009s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.24% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.83% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000616    12
41.72% :     0.000257s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.68% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
53.60% :     0.000330s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013094    72
 0.28% :     0.000037s :      1: add_recomputation
 0.38% :     0.000050s :      1: auto_monad
 0.10% :     0.000013s :      1: auto_monad_reorder
 2.06% :     0.000269s :      1: bootstrap
 0.16% :     0.000021s :      1: cconv
 0.08% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.14% :     0.000019s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
38.74% :     0.005072s :      1: jit_opt_a
 1.04% :     0.000136s :      1: jit_opt_after_cconv
 0.40% :     0.000053s :      1: jit_opt_b
 3.20% :     0.000419s :      1: loop_unroll
 4.47% :     0.000586s :      1: mutable_eliminate
 6.11% :     0.000800s :     26: opt.transform.jit_opt_a
 0.34% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.45% :     0.000452s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000003s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000034s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000012s :      1: remove_dup_value
 2.33% :     0.000305s :      1: renormalize.infer
 2.20% :     0.000288s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000029s :      1: rewriter_after_opt_a
 0.51% :     0.000066s :      1: rewriter_before_opt_a
 0.56% :     0.000073s :      1: symbol_engine_optimizer
31.55% :     0.004130s :      1: type_inference


TotalTime = 0.0179276, [33]
[bootstrap]: 0.00026374
[type_inference]: 0.00436467
[event_method]: 1.699e-05
[auto_monad]: 6.304e-05
[graph_reusing]: 6.11e-06
[pre_auto_parallel]: 2.56998e-06
[py_interpret_to_execute]: 3.271e-05
[rewriter_before_opt_a]: 7.305e-05
[expand_dump_flag]: 3.31999e-06
[jit_opt_a]: 0.0050014, [2]
    [Cycle 1]: 0.00146289, [27]
        [switch_simplify]: 5.071e-05
        [loop_unroll]: 2.965e-05
        [a_1]: 0.00050982
        [with_stream_mark]: 1.58e-05
        [recompute_prepare]: 6.77002e-06
        [updatestate_depend_eliminate]: 3.52002e-06
        [updatestate_assign_eliminate]: 2.77002e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 1.92001e-06
        [specialize_transform]: 6.11e-06
        [updatestate_useless_node_eliminater]: 5.11002e-06
        [accelerated_algorithm]: 5.74e-06
        [meta_shard_fg_expand]: 1.96998e-06
        [get_grad_eliminate_]: 5.22e-06
        [merge_forward]: 3.60998e-06
        [cell_reuse_recompute_pass]: 1.08001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.44e-05
        [j_node_and_user_rematch]: 9.12001e-06
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 8.64e-06
        [inline_without_move]: 5.36998e-06
        [renormalize]: 0.00054192
        [add_forward_monad_depend]: 5.66e-06
        [auto_monad_grad]: 1.77999e-06
        [auto_monad_eliminator]: 1.396e-05
        [cse]: 2.89e-05
        [replace_applicator]: 1.199e-05
    [Cycle 2]: 0.0003313, [27]
        [switch_simplify]: 6.02001e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 9.158e-05
        [with_stream_mark]: 9.12001e-06
        [recompute_prepare]: 5.76e-06
        [updatestate_depend_eliminate]: 2.76e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [parameter_eliminate]: 9.09989e-07
        [specialize_transform]: 5.42999e-06
        [updatestate_useless_node_eliminater]: 5.89999e-06
        [accelerated_algorithm]: 6.06e-06
        [meta_shard_fg_expand]: 1.43002e-06
        [get_grad_eliminate_]: 4.72998e-06
        [merge_forward]: 2.39999e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.256e-05
        [j_node_and_user_rematch]: 8.24998e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 7.26999e-06
        [inline_without_move]: 4.63001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 5.42001e-06
        [cse]: 1.027e-05
        [replace_applicator]: 4.87998e-06
[py_interpret_to_execute_after_opt_a]: 8.80999e-06
[rewriter_after_opt_a]: 3.336e-05
[convert_after_rewriter]: 9.37999e-06
[order_py_execute_after_rewriter]: 5.79999e-06
[mutable_eliminate]: 0.00048437
[jit_opt_b]: 4.949e-05, [1]
    [Cycle 1]: 4.281e-05, [2]
        [frontend_op_eliminate]: 1.613e-05
        [inline_after_opt_a]: 1.491e-05
[cconv]: 2.511e-05
[loop_unroll]: 0.0004212
[jit_opt_after_cconv]: 0.0001373, [1]
    [Cycle 1]: 0.00013075, [11]
        [c_1]: 2.014e-05
        [parameter_eliminate]: 2.58e-06
        [updatestate_depend_eliminate]: 4.94e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [cse]: 1.72e-05
        [call_graph_tuple_transform]: 1.813e-05
        [tuple_list_get_item_eliminator]: 5.59e-06
        [none_parameter_eliminate]: 1.78997e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.25001e-06
[remove_dup_value]: 1.396e-05
[partial_unused_args_eliminate]: 2.09e-06
[environ_conv]: 5.67999e-06
[add_recomputation]: 4.835e-05
[cse_after_recomputation]: 1.962e-05, [1]
    [Cycle 1]: 1.445e-05, [1]
        [cse]: 8.69003e-06
[auto_monad_reorder]: 1.483e-05
[get_jit_bprop_graph]: 1.51998e-06
[rewriter_after_jit_bprop_graph]: 5.13002e-06
[opt_after_jit_grad]: 0.00045604
[symbol_engine_optimizer]: 7.738e-05, [1]
    [Cycle 1]: 7.091e-05, [6]
        [build]: 3.85e-06
        [elim_shapecalc]: 8.77e-06
        [elim_not_effective]: 1.417e-05
        [opt_reshape]: 6.49999e-06
        [fold_const_symbol]: 8.28001e-06
        [renormalize]: 4.59986e-07
[validate]: 3.431e-05
[backend_pass]: 7.60017e-07
[task_emit]: 0.00605456
[execute]: 8.32e-06
Sums
    bootstrap                            :     0.000264s :  1.87%
    type_inference                       :     0.004365s : 30.89%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000063s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.23%
    rewriter_before_opt_a                :     0.000073s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.25%
    jit_opt_a.a_1                        :     0.000601s :  4.26%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000542s :  3.84%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.14%
    jit_opt_a.cse                        :     0.000039s :  0.28%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.24%
    convert_after_rewriter               :     0.000009s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000484s :  3.43%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000025s :  0.18%
    loop_unroll                          :     0.000421s :  2.98%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000048s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000456s :  3.23%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006055s : 42.85%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000182    23
 1.13% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.82% :     0.000005s :      3: substitution.graph_param_transform
82.74% :     0.000151s :      4: substitution.inline
 1.70% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.30% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000004s :      2: substitution.replace_old_param
 6.53% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004305     2
81.99% :     0.003530s :      1: type_inference.infer
18.01% :     0.000775s :      1: type_inference.specialize
------[replace.]   0.000066     6
66.82% :     0.000044s :      4: replace.inline
33.18% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000159     6
93.23% :     0.000148s :      4: match.inline
 6.77% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 1.69% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.24% :     0.000001s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.04% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000000s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.44% :     0.000006s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.24% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.61% :     0.000002s :      3: predicate.mutable_eliminate
 0.66% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.65% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.48% :     0.000002s :     10: predicate.same_eliminate
 0.62% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.36% :     0.000002s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.49% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.92% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.31% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000002s :     10: predicate.value_based_eliminate
 0.43% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000538    12
49.62% :     0.000267s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.12% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.26% :     0.000243s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019390    76
 0.26% :     0.000051s :      1: add_recomputation
 0.34% :     0.000066s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.43% :     0.000278s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.06% :     0.000012s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.07% :     0.000013s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.81% :     0.005004s :      1: jit_opt_a
 0.72% :     0.000140s :      1: jit_opt_after_cconv
 0.27% :     0.000052s :      1: jit_opt_b
 2.21% :     0.000429s :      1: loop_unroll
 2.54% :     0.000492s :      1: mutable_eliminate
 4.16% :     0.000808s :     26: opt.transform.jit_opt_a
 0.24% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.40% :     0.000464s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000035s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.39% :     0.000269s :      1: renormalize.infer
 1.37% :     0.000265s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000036s :      1: rewriter_after_opt_a
 0.39% :     0.000076s :      1: rewriter_before_opt_a
 0.41% :     0.000080s :      1: symbol_engine_optimizer
31.29% :     0.006067s :      1: task_emit
22.57% :     0.004376s :      1: type_inference
 0.28% :     0.000055s :      1: validate


TotalTime = 0.0114041, [30]
[bootstrap]: 0.00025685
[type_inference]: 0.00406944
[event_method]: 1.604e-05
[auto_monad]: 6.221e-05
[graph_reusing]: 5.62001e-06
[pre_auto_parallel]: 2.71999e-06
[py_interpret_to_execute]: 3.05e-05
[rewriter_before_opt_a]: 0.00012795
[expand_dump_flag]: 3.65e-06
[jit_opt_a]: 0.0048661, [2]
    [Cycle 1]: 0.00147436, [27]
        [switch_simplify]: 5.142e-05
        [loop_unroll]: 2.969e-05
        [a_1]: 0.00050781
        [with_stream_mark]: 1.542e-05
        [recompute_prepare]: 7.43e-06
        [updatestate_depend_eliminate]: 3.44001e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 6.05002e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 5.94e-06
        [meta_shard_fg_expand]: 2.17999e-06
        [get_grad_eliminate_]: 5.00001e-06
        [merge_forward]: 3.53999e-06
        [cell_reuse_recompute_pass]: 1.05999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.422e-05
        [j_node_and_user_rematch]: 1.014e-05
        [meta_fg_expand]: 2.05002e-06
        [replace_old_param]: 8.69e-06
        [inline_without_move]: 5.69999e-06
        [renormalize]: 0.00055628
        [add_forward_monad_depend]: 5.12e-06
        [auto_monad_grad]: 2.09e-06
        [auto_monad_eliminator]: 1.343e-05
        [cse]: 2.675e-05
        [replace_applicator]: 1.113e-05
    [Cycle 2]: 0.00032646, [27]
        [switch_simplify]: 5.42001e-06
        [loop_unroll]: 4.78001e-06
        [a_1]: 9e-05
        [with_stream_mark]: 8.84e-06
        [recompute_prepare]: 5.02e-06
        [updatestate_depend_eliminate]: 2.65002e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.83997e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 4.67998e-06
        [accelerated_algorithm]: 5.32001e-06
        [meta_shard_fg_expand]: 9.89996e-07
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.202e-05
        [j_node_and_user_rematch]: 7.74002e-06
        [meta_fg_expand]: 1.52999e-06
        [replace_old_param]: 7.14001e-06
        [inline_without_move]: 4.63999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.45999e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 5.46e-06
        [cse]: 1.069e-05
        [replace_applicator]: 5.41998e-06
[py_interpret_to_execute_after_opt_a]: 8.32e-06
[rewriter_after_opt_a]: 3.173e-05
[convert_after_rewriter]: 6.34999e-06
[order_py_execute_after_rewriter]: 4.68999e-06
[mutable_eliminate]: 0.00046185
[jit_opt_b]: 4.825e-05, [1]
    [Cycle 1]: 4.169e-05, [2]
        [frontend_op_eliminate]: 1.543e-05
        [inline_after_opt_a]: 1.489e-05
[cconv]: 2.285e-05
[loop_unroll]: 0.00041863
[jit_opt_after_cconv]: 0.00014047, [1]
    [Cycle 1]: 0.00013422, [11]
        [c_1]: 2.239e-05
        [parameter_eliminate]: 2.84999e-06
        [updatestate_depend_eliminate]: 5.56e-06
        [updatestate_assign_eliminate]: 2.55002e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [cse]: 1.653e-05
        [call_graph_tuple_transform]: 1.847e-05
        [tuple_list_get_item_eliminator]: 5.05999e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.57001e-06
[remove_dup_value]: 1.329e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 4.95999e-06
[add_recomputation]: 4.705e-05
[cse_after_recomputation]: 1.956e-05, [1]
    [Cycle 1]: 1.442e-05, [1]
        [cse]: 8.64998e-06
[auto_monad_reorder]: 1.459e-05
[get_jit_bprop_graph]: 1.60999e-06
[rewriter_after_jit_bprop_graph]: 4.94998e-06
[opt_after_jit_grad]: 0.0004559
[symbol_engine_optimizer]: 7.203e-05, [1]
    [Cycle 1]: 6.587e-05, [6]
        [build]: 3.26001e-06
        [elim_shapecalc]: 8.01001e-06
        [elim_not_effective]: 1.233e-05
        [opt_reshape]: 5.47001e-06
        [fold_const_symbol]: 8.60001e-06
        [renormalize]: 5.10016e-07
[validate]: 3.013e-05
Sums
    bootstrap                            :     0.000257s :  3.30%
    type_inference                       :     0.004069s : 52.35%
    event_method                         :     0.000016s :  0.21%
    auto_monad                           :     0.000062s :  0.80%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000030s :  0.39%
    rewriter_before_opt_a                :     0.000128s :  1.65%
    expand_dump_flag                     :     0.000004s :  0.05%
    jit_opt_a.switch_simplify            :     0.000057s :  0.73%
    jit_opt_a.loop_unroll                :     0.000034s :  0.44%
    jit_opt_a.a_1                        :     0.000598s :  7.69%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000556s :  7.16%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.24%
    jit_opt_a.cse                        :     0.000037s :  0.48%
    jit_opt_a.replace_applicator         :     0.000017s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000032s :  0.41%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000462s :  5.94%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000023s :  0.29%
    loop_unroll                          :     0.000419s :  5.39%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.29%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000013s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000047s :  0.61%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000456s :  5.86%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000030s :  0.39%


Time group info:
------[substitution.]   0.000186    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.65% :     0.000005s :      3: substitution.graph_param_transform
82.79% :     0.000154s :      4: substitution.inline
 1.91% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.49% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000004s :      2: substitution.replace_old_param
 6.51% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004013     2
79.42% :     0.003187s :      1: type_inference.infer
20.58% :     0.000826s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.93% :     0.000042s :      4: replace.inline
32.07% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     6
93.23% :     0.000151s :      4: match.inline
 6.77% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.73% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.61% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000001s :     10: predicate.addn_zero_filter
 1.95% :     0.000002s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.54% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.54% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.14% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.40% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.51% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.29% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.56% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.75% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.47% :     0.000008s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.71% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.84% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000560    12
49.33% :     0.000276s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.19% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.48% :     0.000254s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012828    72
 0.39% :     0.000050s :      1: add_recomputation
 0.51% :     0.000065s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.11% :     0.000271s :      1: bootstrap
 0.20% :     0.000026s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.15% :     0.000019s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
37.95% :     0.004869s :      1: jit_opt_a
 1.12% :     0.000143s :      1: jit_opt_after_cconv
 0.40% :     0.000051s :      1: jit_opt_b
 3.33% :     0.000427s :      1: loop_unroll
 3.66% :     0.000470s :      1: mutable_eliminate
 6.26% :     0.000803s :     26: opt.transform.jit_opt_a
 0.38% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000024s :      4: opt.transform.jit_opt_b
 0.10% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.61% :     0.000463s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.04% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000033s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.24% :     0.000287s :      1: renormalize.infer
 2.04% :     0.000262s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000035s :      1: rewriter_after_opt_a
 1.03% :     0.000132s :      1: rewriter_before_opt_a
 0.58% :     0.000074s :      1: symbol_engine_optimizer
31.81% :     0.004080s :      1: type_inference


TotalTime = 0.0351184, [33]
[bootstrap]: 0.00032816
[type_inference]: 0.00538243
[event_method]: 2.182e-05
[auto_monad]: 7.08e-05
[graph_reusing]: 6.79001e-06
[pre_auto_parallel]: 3.36999e-06
[py_interpret_to_execute]: 5.051e-05
[rewriter_before_opt_a]: 8.492e-05
[expand_dump_flag]: 3.36999e-06
[jit_opt_a]: 0.0188436, [2]
    [Cycle 1]: 0.00203572, [27]
        [switch_simplify]: 6.702e-05
        [loop_unroll]: 3.131e-05
        [a_1]: 0.00059927
        [with_stream_mark]: 2.413e-05
        [recompute_prepare]: 7.05e-06
        [updatestate_depend_eliminate]: 3.68999e-06
        [updatestate_assign_eliminate]: 3.06001e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 1.82999e-06
        [specialize_transform]: 5.82999e-06
        [updatestate_useless_node_eliminater]: 5.09998e-06
        [accelerated_algorithm]: 6.19001e-06
        [meta_shard_fg_expand]: 2.43998e-06
        [get_grad_eliminate_]: 5.20999e-06
        [merge_forward]: 3.91999e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.431e-05
        [j_node_and_user_rematch]: 9.40001e-06
        [meta_fg_expand]: 2.40002e-06
        [replace_old_param]: 9.64e-06
        [inline_without_move]: 5.45001e-06
        [renormalize]: 0.00095825
        [add_forward_monad_depend]: 6.53e-06
        [auto_monad_grad]: 2.79999e-06
        [auto_monad_eliminator]: 1.737e-05
        [cse]: 3.104e-05
        [replace_applicator]: 1.516e-05
    [Cycle 2]: 0.00034575, [27]
        [switch_simplify]: 5.84999e-06
        [loop_unroll]: 5.11002e-06
        [a_1]: 9.801e-05
        [with_stream_mark]: 1.077e-05
        [recompute_prepare]: 5.27001e-06
        [updatestate_depend_eliminate]: 2.82002e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 5.32999e-06
        [updatestate_useless_node_eliminater]: 4.67998e-06
        [accelerated_algorithm]: 6.29999e-06
        [meta_shard_fg_expand]: 1.64998e-06
        [get_grad_eliminate_]: 4.65001e-06
        [merge_forward]: 2.52001e-06
        [cell_reuse_recompute_pass]: 2.49999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.451e-05
        [j_node_and_user_rematch]: 8.18001e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 8.25e-06
        [inline_without_move]: 4.57e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.42999e-06
        [auto_monad_grad]: 9.60019e-07
        [auto_monad_eliminator]: 5.39e-06
        [cse]: 1.261e-05
        [replace_applicator]: 5.09e-06
[py_interpret_to_execute_after_opt_a]: 1.311e-05
[rewriter_after_opt_a]: 3.723e-05
[convert_after_rewriter]: 7.15e-06
[order_py_execute_after_rewriter]: 4.77e-06
[mutable_eliminate]: 0.00072267
[jit_opt_b]: 5.63e-05, [1]
    [Cycle 1]: 4.748e-05, [2]
        [frontend_op_eliminate]: 2.036e-05
        [inline_after_opt_a]: 1.529e-05
[cconv]: 2.686e-05
[loop_unroll]: 0.00044009
[jit_opt_after_cconv]: 0.00014051, [1]
    [Cycle 1]: 0.00013377, [11]
        [c_1]: 2.043e-05
        [parameter_eliminate]: 3.05002e-06
        [updatestate_depend_eliminate]: 5.96e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [cse]: 1.864e-05
        [call_graph_tuple_transform]: 1.959e-05
        [tuple_list_get_item_eliminator]: 5.47001e-06
        [none_parameter_eliminate]: 1.50999e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 5.50001e-06
[remove_dup_value]: 1.382e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 5.05999e-06
[add_recomputation]: 4.99e-05
[cse_after_recomputation]: 2.166e-05, [1]
    [Cycle 1]: 1.545e-05, [1]
        [cse]: 9.39998e-06
[auto_monad_reorder]: 1.464e-05
[get_jit_bprop_graph]: 1.91003e-06
[rewriter_after_jit_bprop_graph]: 6.51999e-06
[opt_after_jit_grad]: 0.00045691
[symbol_engine_optimizer]: 7.522e-05, [1]
    [Cycle 1]: 6.803e-05, [6]
        [build]: 4.08999e-06
        [elim_shapecalc]: 7.4e-06
        [elim_not_effective]: 1.33e-05
        [opt_reshape]: 5.80002e-06
        [fold_const_symbol]: 9.29e-06
        [renormalize]: 6.39993e-07
[validate]: 4.208e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.00793414
[execute]: 9.56e-06
Sums
    bootstrap                            :     0.000328s :  1.83%
    type_inference                       :     0.005382s : 29.94%
    event_method                         :     0.000022s :  0.12%
    auto_monad                           :     0.000071s :  0.39%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000051s :  0.28%
    rewriter_before_opt_a                :     0.000085s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000073s :  0.41%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000697s :  3.88%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000958s :  5.33%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.13%
    jit_opt_a.cse                        :     0.000044s :  0.24%
    jit_opt_a.replace_applicator         :     0.000020s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000723s :  4.02%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000027s :  0.15%
    loop_unroll                          :     0.000440s :  2.45%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.28%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000457s :  2.54%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.23%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007934s : 44.13%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000255    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.19% :     0.000006s :      3: substitution.graph_param_transform
86.06% :     0.000219s :      4: substitution.inline
 1.44% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.90% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000005s :      2: substitution.replace_old_param
 4.99% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005296     2
80.09% :     0.004242s :      1: type_inference.infer
19.91% :     0.001054s :      1: type_inference.specialize
------[replace.]   0.000079     6
67.68% :     0.000053s :      4: replace.inline
32.32% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000228     6
94.94% :     0.000217s :      4: match.inline
 5.06% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 2.39% :     0.000003s :     10: predicate.arithmetic_simplify
 1.74% :     0.000002s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     10: predicate.environ_get_eliminate
 1.22% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.19% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.52% :     0.000002s :      6: predicate.less_batch_normalization
 1.93% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.72% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.79% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.94% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.96% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.48% :     0.000001s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.13% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.75% :     0.000003s :     16: predicate.switch_defer_inline
 1.90% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.60% :     0.000009s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.33% :     0.000002s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.40% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000749    12
48.08% :     0.000360s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.94% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.98% :     0.000337s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.037098    76
 0.14% :     0.000053s :      1: add_recomputation
 0.20% :     0.000075s :      1: auto_monad
 0.05% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.94% :     0.000347s :      1: bootstrap
 0.08% :     0.000030s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.06% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.07% :     0.000026s :      1: event_method
 0.04% :     0.000015s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
50.81% :     0.018848s :      1: jit_opt_a
 0.39% :     0.000143s :      1: jit_opt_after_cconv
 0.16% :     0.000059s :      1: jit_opt_b
 1.21% :     0.000450s :      1: loop_unroll
 1.97% :     0.000732s :      1: mutable_eliminate
 2.50% :     0.000927s :     26: opt.transform.jit_opt_a
 0.13% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000029s :      4: opt.transform.jit_opt_b
 0.03% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.25% :     0.000465s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.15% :     0.000054s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000016s :      1: remove_dup_value
 1.42% :     0.000528s :      1: renormalize.infer
 1.13% :     0.000420s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000041s :      1: rewriter_after_opt_a
 0.24% :     0.000089s :      1: rewriter_before_opt_a
 0.21% :     0.000078s :      1: symbol_engine_optimizer
21.44% :     0.007954s :      1: task_emit
14.58% :     0.005408s :      1: type_inference
 0.17% :     0.000062s :      1: validate


TotalTime = 0.0161213, [33]
[bootstrap]: 0.00026367
[type_inference]: 0.00396141
[event_method]: 1.594e-05
[auto_monad]: 5.173e-05
[graph_reusing]: 4.1e-06
[pre_auto_parallel]: 2.22001e-06
[py_interpret_to_execute]: 2.923e-05
[rewriter_before_opt_a]: 6.541e-05
[expand_dump_flag]: 2.72001e-06
[jit_opt_a]: 0.00454724, [2]
    [Cycle 1]: 0.00130845, [27]
        [switch_simplify]: 4.365e-05
        [loop_unroll]: 3.019e-05
        [a_1]: 0.00048056
        [with_stream_mark]: 1.294e-05
        [recompute_prepare]: 7.23999e-06
        [updatestate_depend_eliminate]: 3.22002e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 1.95001e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 5.71998e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 5.51998e-06
        [meta_shard_fg_expand]: 1.64e-06
        [get_grad_eliminate_]: 5.12999e-06
        [merge_forward]: 2.61e-06
        [cell_reuse_recompute_pass]: 9.29984e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.144e-05
        [j_node_and_user_rematch]: 8.73001e-06
        [meta_fg_expand]: 1.99999e-06
        [replace_old_param]: 7.61999e-06
        [inline_without_move]: 5.29998e-06
        [renormalize]: 0.00045882
        [add_forward_monad_depend]: 4.94e-06
        [auto_monad_grad]: 1.52999e-06
        [auto_monad_eliminator]: 1.057e-05
        [cse]: 1.631e-05
        [replace_applicator]: 1.117e-05
    [Cycle 2]: 0.00031202, [27]
        [switch_simplify]: 5.52999e-06
        [loop_unroll]: 4.69998e-06
        [a_1]: 8.926e-05
        [with_stream_mark]: 8.27998e-06
        [recompute_prepare]: 4.90999e-06
        [updatestate_depend_eliminate]: 2.46998e-06
        [updatestate_assign_eliminate]: 1.97999e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 9.00007e-07
        [specialize_transform]: 5.01002e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 4.90001e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 4.53001e-06
        [merge_forward]: 2.19999e-06
        [cell_reuse_recompute_pass]: 1.47999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.15e-05
        [j_node_and_user_rematch]: 7.63999e-06
        [meta_fg_expand]: 1.52999e-06
        [replace_old_param]: 6.84999e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 4.77e-06
        [cse]: 9.69999e-06
        [replace_applicator]: 4.90001e-06
[py_interpret_to_execute_after_opt_a]: 8.29998e-06
[rewriter_after_opt_a]: 2.648e-05
[convert_after_rewriter]: 5.76e-06
[order_py_execute_after_rewriter]: 4.55001e-06
[mutable_eliminate]: 0.00046494
[jit_opt_b]: 4.868e-05, [1]
    [Cycle 1]: 4.246e-05, [2]
        [frontend_op_eliminate]: 1.609e-05
        [inline_after_opt_a]: 1.517e-05
[cconv]: 1.906e-05
[loop_unroll]: 0.00039816
[jit_opt_after_cconv]: 0.00013195, [1]
    [Cycle 1]: 0.000126, [11]
        [c_1]: 2.051e-05
        [parameter_eliminate]: 2.79999e-06
        [updatestate_depend_eliminate]: 4.87998e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.548e-05
        [call_graph_tuple_transform]: 1.643e-05
        [tuple_list_get_item_eliminator]: 5.48002e-06
        [none_parameter_eliminate]: 1.35001e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.54e-06
[remove_dup_value]: 1.175e-05
[partial_unused_args_eliminate]: 1.65001e-06
[environ_conv]: 4.44002e-06
[add_recomputation]: 3.62e-05
[cse_after_recomputation]: 1.997e-05, [1]
    [Cycle 1]: 1.499e-05, [1]
        [cse]: 9.42001e-06
[auto_monad_reorder]: 1.156e-05
[get_jit_bprop_graph]: 1.90001e-06
[rewriter_after_jit_bprop_graph]: 4.59998e-06
[opt_after_jit_grad]: 0.00043568
[symbol_engine_optimizer]: 6.986e-05, [1]
    [Cycle 1]: 6.372e-05, [6]
        [build]: 2.75997e-06
        [elim_shapecalc]: 7.28e-06
        [elim_not_effective]: 1.208e-05
        [opt_reshape]: 5.40001e-06
        [fold_const_symbol]: 8.54002e-06
        [renormalize]: 5.09986e-07
[validate]: 2.778e-05
[backend_pass]: 7.09988e-07
[task_emit]: 0.00526686
[execute]: 5.67001e-06
Sums
    bootstrap                            :     0.000264s :  2.09%
    type_inference                       :     0.003961s : 31.36%
    event_method                         :     0.000016s :  0.13%
    auto_monad                           :     0.000052s :  0.41%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.23%
    rewriter_before_opt_a                :     0.000065s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000049s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.28%
    jit_opt_a.a_1                        :     0.000570s :  4.51%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000459s :  3.63%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.12%
    jit_opt_a.cse                        :     0.000026s :  0.21%
    jit_opt_a.replace_applicator         :     0.000016s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.07%
    rewriter_after_opt_a                 :     0.000026s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000465s :  3.68%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000019s :  0.15%
    loop_unroll                          :     0.000398s :  3.15%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000036s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000012s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000436s :  3.45%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000028s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005267s : 41.69%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000159    23
 1.07% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.54% :     0.000004s :      3: substitution.graph_param_transform
84.02% :     0.000133s :      4: substitution.inline
 1.68% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.28% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.97% :     0.000003s :      2: substitution.replace_old_param
 5.72% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003910     2
81.58% :     0.003190s :      1: type_inference.infer
18.42% :     0.000720s :      1: type_inference.specialize
------[replace.]   0.000063     6
67.78% :     0.000042s :      4: replace.inline
32.22% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000139     6
94.34% :     0.000131s :      4: match.inline
 5.66% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.39% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.24% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000001s :     10: predicate.addn_zero_filter
 1.96% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.29% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.41% :     0.000000s :      3: predicate.graph_param_transform
 5.10% :     0.000006s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.44% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.41% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.67% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000002s :     15: predicate.partial_eliminate
 1.33% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000001s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.20% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.50% :     0.000008s :     53: predicate.switch_simplify
 1.27% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000484    12
48.28% :     0.000234s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.80% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.92% :     0.000222s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017447    76
 0.22% :     0.000039s :      1: add_recomputation
 0.32% :     0.000055s :      1: auto_monad
 0.08% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.58% :     0.000275s :      1: bootstrap
 0.12% :     0.000022s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.13% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
26.08% :     0.004550s :      1: jit_opt_a
 0.77% :     0.000135s :      1: jit_opt_after_cconv
 0.29% :     0.000051s :      1: jit_opt_b
 2.33% :     0.000406s :      1: loop_unroll
 2.71% :     0.000474s :      1: mutable_eliminate
 4.37% :     0.000763s :     26: opt.transform.jit_opt_a
 0.26% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.54% :     0.000444s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 1.28% :     0.000223s :      1: renormalize.infer
 1.32% :     0.000230s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000029s :      1: rewriter_after_opt_a
 0.39% :     0.000068s :      1: rewriter_before_opt_a
 0.41% :     0.000072s :      1: symbol_engine_optimizer
30.23% :     0.005275s :      1: task_emit
22.77% :     0.003972s :      1: type_inference
 0.25% :     0.000043s :      1: validate


TotalTime = 0.0235519, [33]
[bootstrap]: 0.00038075
[type_inference]: 0.00600425
[event_method]: 2.211e-05
[auto_monad]: 0.00012047
[graph_reusing]: 6.19999e-06
[pre_auto_parallel]: 2.98998e-06
[py_interpret_to_execute]: 4.253e-05
[rewriter_before_opt_a]: 7.846e-05
[expand_dump_flag]: 3.42002e-06
[jit_opt_a]: 0.00668504, [2]
    [Cycle 1]: 0.0019728, [27]
        [switch_simplify]: 6.818e-05
        [loop_unroll]: 2.835e-05
        [a_1]: 0.00060746
        [with_stream_mark]: 2.722e-05
        [recompute_prepare]: 8.12e-06
        [updatestate_depend_eliminate]: 5.20999e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 1.88002e-06
        [specialize_transform]: 6.19999e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 2.31998e-06
        [get_grad_eliminate_]: 4.94e-06
        [merge_forward]: 3.66001e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.555e-05
        [j_node_and_user_rematch]: 8.69003e-06
        [meta_fg_expand]: 2.46998e-06
        [replace_old_param]: 1.007e-05
        [inline_without_move]: 5.09e-06
        [renormalize]: 0.00086278
        [add_forward_monad_depend]: 9.45001e-06
        [auto_monad_grad]: 2.49999e-06
        [auto_monad_eliminator]: 1.832e-05
        [cse]: 3.552e-05
        [replace_applicator]: 1.737e-05
    [Cycle 2]: 0.00034288, [27]
        [switch_simplify]: 5.56e-06
        [loop_unroll]: 4.83001e-06
        [a_1]: 9.062e-05
        [with_stream_mark]: 1.562e-05
        [recompute_prepare]: 6.01998e-06
        [updatestate_depend_eliminate]: 3.33998e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 1.95001e-06
        [parameter_eliminate]: 1.32999e-06
        [specialize_transform]: 5.46e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.19e-06
        [meta_shard_fg_expand]: 2.39999e-06
        [get_grad_eliminate_]: 4.48999e-06
        [merge_forward]: 2.67001e-06
        [cell_reuse_recompute_pass]: 3.68999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.637e-05
        [j_node_and_user_rematch]: 8.13001e-06
        [meta_fg_expand]: 1.87999e-06
        [replace_old_param]: 8.53001e-06
        [inline_without_move]: 4.99e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.84e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 5.77001e-06
        [cse]: 1.175e-05
        [replace_applicator]: 4.63999e-06
[py_interpret_to_execute_after_opt_a]: 1.596e-05
[rewriter_after_opt_a]: 4.165e-05
[convert_after_rewriter]: 6.81999e-06
[order_py_execute_after_rewriter]: 4.87998e-06
[mutable_eliminate]: 0.00084398
[jit_opt_b]: 5.919e-05, [1]
    [Cycle 1]: 4.927e-05, [2]
        [frontend_op_eliminate]: 2.02e-05
        [inline_after_opt_a]: 1.697e-05
[cconv]: 3.674e-05
[loop_unroll]: 0.00052063
[jit_opt_after_cconv]: 0.00016122, [1]
    [Cycle 1]: 0.00015422, [11]
        [c_1]: 2.029e-05
        [parameter_eliminate]: 4.19002e-06
        [updatestate_depend_eliminate]: 1.005e-05
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [cse]: 2.97e-05
        [call_graph_tuple_transform]: 2.199e-05
        [tuple_list_get_item_eliminator]: 5.47999e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 5.32999e-06
[remove_dup_value]: 1.576e-05
[partial_unused_args_eliminate]: 2.24001e-06
[environ_conv]: 6.29001e-06
[add_recomputation]: 6.137e-05
[cse_after_recomputation]: 2.01e-05, [1]
    [Cycle 1]: 1.479e-05, [1]
        [cse]: 9.22001e-06
[auto_monad_reorder]: 1.592e-05
[get_jit_bprop_graph]: 2.17999e-06
[rewriter_after_jit_bprop_graph]: 8.75001e-06
[opt_after_jit_grad]: 0.00048109
[symbol_engine_optimizer]: 7.634e-05, [1]
    [Cycle 1]: 6.953e-05, [6]
        [build]: 4.70999e-06
        [elim_shapecalc]: 7.65e-06
        [elim_not_effective]: 1.353e-05
        [opt_reshape]: 5.83997e-06
        [fold_const_symbol]: 8.67e-06
        [renormalize]: 6.30011e-07
[validate]: 4.553e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00750839
[execute]: 1.156e-05
Sums
    bootstrap                            :     0.000381s :  2.06%
    type_inference                       :     0.006004s : 32.50%
    event_method                         :     0.000022s :  0.12%
    auto_monad                           :     0.000120s :  0.65%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.23%
    rewriter_before_opt_a                :     0.000078s :  0.42%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.18%
    jit_opt_a.a_1                        :     0.000698s :  3.78%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000863s :  4.67%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.13%
    jit_opt_a.cse                        :     0.000047s :  0.26%
    jit_opt_a.replace_applicator         :     0.000022s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.09%
    rewriter_after_opt_a                 :     0.000042s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000844s :  4.57%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000037s :  0.20%
    loop_unroll                          :     0.000521s :  2.82%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000030s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000061s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000481s :  2.60%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007508s : 40.64%
    execute                              :     0.000012s :  0.06%


Time group info:
------[substitution.]   0.000274    23
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.25% :     0.000006s :      3: substitution.graph_param_transform
85.62% :     0.000235s :      4: substitution.inline
 1.48% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.14% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.27% :     0.000006s :      2: substitution.replace_old_param
 4.92% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005912     2
83.83% :     0.004956s :      1: type_inference.infer
16.17% :     0.000956s :      1: type_inference.specialize
------[replace.]   0.000072     6
69.00% :     0.000050s :      4: replace.inline
31.00% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000245     6
94.93% :     0.000233s :      4: match.inline
 5.07% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.57% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.95% :     0.000002s :     10: predicate.addn_zero_filter
 1.77% :     0.000002s :     10: predicate.arithmetic_simplify
 1.45% :     0.000002s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.09% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.48% :     0.000001s :      3: predicate.graph_param_transform
 6.64% :     0.000008s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.71% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.61% :     0.000002s :     15: predicate.load_eliminater
 1.61% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.63% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 3.05% :     0.000004s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.08% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     10: predicate.reduce_eliminate
 1.29% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.78% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.92% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.48% :     0.000001s :      3: predicate.reset_defer_inline
 1.15% :     0.000001s :     10: predicate.reshape_eliminate
 1.39% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.68% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.18% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 1.93% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.56% :     0.000011s :     53: predicate.switch_simplify
 1.07% :     0.000001s :     10: predicate.tile_eliminate
 1.25% :     0.000002s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.02% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.23% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.42% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000715    12
49.05% :     0.000350s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.20% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.75% :     0.000306s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025459    76
 0.26% :     0.000065s :      1: add_recomputation
 0.49% :     0.000126s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.59% :     0.000406s :      1: bootstrap
 0.16% :     0.000040s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.11% :     0.000028s :      1: event_method
 0.07% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
26.27% :     0.006688s :      1: jit_opt_a
 0.64% :     0.000164s :      1: jit_opt_after_cconv
 0.24% :     0.000062s :      1: jit_opt_b
 2.08% :     0.000530s :      1: loop_unroll
 3.38% :     0.000861s :      1: mutable_eliminate
 3.64% :     0.000928s :     26: opt.transform.jit_opt_a
 0.20% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000030s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.93% :     0.000491s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000045s :      1: py_interpret_to_execute
 0.07% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 1.92% :     0.000490s :      1: renormalize.infer
 1.43% :     0.000364s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000045s :      1: rewriter_after_opt_a
 0.32% :     0.000082s :      1: rewriter_before_opt_a
 0.31% :     0.000079s :      1: symbol_engine_optimizer
29.58% :     0.007530s :      1: task_emit
23.69% :     0.006032s :      1: type_inference
 0.28% :     0.000070s :      1: validate


TotalTime = 0.0113011, [30]
[bootstrap]: 0.00025513
[type_inference]: 0.00399321
[event_method]: 1.599e-05
[auto_monad]: 6.156e-05
[graph_reusing]: 5.91e-06
[pre_auto_parallel]: 2.49001e-06
[py_interpret_to_execute]: 3.245e-05
[rewriter_before_opt_a]: 6.92e-05
[expand_dump_flag]: 2.51e-06
[jit_opt_a]: 0.00491714, [2]
    [Cycle 1]: 0.00149085, [27]
        [switch_simplify]: 7.287e-05
        [loop_unroll]: 3.746e-05
        [a_1]: 0.00050512
        [with_stream_mark]: 1.63e-05
        [recompute_prepare]: 7.34002e-06
        [updatestate_depend_eliminate]: 3.63e-06
        [updatestate_assign_eliminate]: 3.10002e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [parameter_eliminate]: 1.55999e-06
        [specialize_transform]: 5.57999e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 5.85002e-06
        [meta_shard_fg_expand]: 1.89999e-06
        [get_grad_eliminate_]: 5.07999e-06
        [merge_forward]: 3.61999e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.359e-05
        [j_node_and_user_rematch]: 8.72e-06
        [meta_fg_expand]: 2.32999e-06
        [replace_old_param]: 8.3e-06
        [inline_without_move]: 5.41998e-06
        [renormalize]: 0.00055268
        [add_forward_monad_depend]: 5.04e-06
        [auto_monad_grad]: 1.84e-06
        [auto_monad_eliminator]: 1.196e-05
        [cse]: 2.618e-05
        [replace_applicator]: 1.177e-05
    [Cycle 2]: 0.00031074, [27]
        [switch_simplify]: 5.52999e-06
        [loop_unroll]: 4.75999e-06
        [a_1]: 8.792e-05
        [with_stream_mark]: 8.41002e-06
        [recompute_prepare]: 5.16998e-06
        [updatestate_depend_eliminate]: 2.58e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.98001e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.04998e-06
        [get_grad_eliminate_]: 4.67e-06
        [merge_forward]: 2.44999e-06
        [cell_reuse_recompute_pass]: 1.43002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.174e-05
        [j_node_and_user_rematch]: 7.63001e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 7.01999e-06
        [inline_without_move]: 4.70001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 4.62998e-06
        [cse]: 9.76e-06
        [replace_applicator]: 5.03002e-06
[py_interpret_to_execute_after_opt_a]: 8.65999e-06
[rewriter_after_opt_a]: 3.009e-05
[convert_after_rewriter]: 6.88e-06
[order_py_execute_after_rewriter]: 4.65001e-06
[mutable_eliminate]: 0.00047246
[jit_opt_b]: 5.516e-05, [1]
    [Cycle 1]: 4.863e-05, [2]
        [frontend_op_eliminate]: 2.079e-05
        [inline_after_opt_a]: 1.649e-05
[cconv]: 2.439e-05
[loop_unroll]: 0.00041649
[jit_opt_after_cconv]: 0.00013381, [1]
    [Cycle 1]: 0.00012741, [11]
        [c_1]: 2.052e-05
        [parameter_eliminate]: 2.43998e-06
        [updatestate_depend_eliminate]: 5.10999e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.588e-05
        [call_graph_tuple_transform]: 1.795e-05
        [tuple_list_get_item_eliminator]: 5.38002e-06
        [none_parameter_eliminate]: 1.29e-06
        [renormalize]: 4.40021e-07
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 1.343e-05
[partial_unused_args_eliminate]: 2.56e-06
[environ_conv]: 4.65001e-06
[add_recomputation]: 4.393e-05
[cse_after_recomputation]: 2.034e-05, [1]
    [Cycle 1]: 1.461e-05, [1]
        [cse]: 8.77e-06
[auto_monad_reorder]: 1.343e-05
[get_jit_bprop_graph]: 1.59998e-06
[rewriter_after_jit_bprop_graph]: 4.62e-06
[opt_after_jit_grad]: 0.00043951
[symbol_engine_optimizer]: 6.934e-05, [1]
    [Cycle 1]: 6.3e-05, [6]
        [build]: 3.28998e-06
        [elim_shapecalc]: 6.95998e-06
        [elim_not_effective]: 1.17e-05
        [opt_reshape]: 5.30999e-06
        [fold_const_symbol]: 8.33999e-06
        [renormalize]: 3.50003e-07
[validate]: 3.008e-05
Sums
    bootstrap                            :     0.000255s :  3.34%
    type_inference                       :     0.003993s : 52.27%
    event_method                         :     0.000016s :  0.21%
    auto_monad                           :     0.000062s :  0.81%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.42%
    rewriter_before_opt_a                :     0.000069s :  0.91%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000078s :  1.03%
    jit_opt_a.loop_unroll                :     0.000042s :  0.55%
    jit_opt_a.a_1                        :     0.000593s :  7.76%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000553s :  7.23%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.22%
    jit_opt_a.cse                        :     0.000036s :  0.47%
    jit_opt_a.replace_applicator         :     0.000017s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000030s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000472s :  6.18%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.27%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.22%
    cconv                                :     0.000024s :  0.32%
    loop_unroll                          :     0.000416s :  5.45%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000013s :  0.18%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000044s :  0.57%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000013s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000440s :  5.75%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000030s :  0.39%


Time group info:
------[substitution.]   0.000185    23
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.80% :     0.000005s :      3: substitution.graph_param_transform
83.74% :     0.000155s :      4: substitution.inline
 1.61% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.32% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.05% :     0.000004s :      2: substitution.replace_old_param
 5.82% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003938     2
79.31% :     0.003123s :      1: type_inference.infer
20.69% :     0.000815s :      1: type_inference.specialize
------[replace.]   0.000063     6
67.81% :     0.000042s :      4: replace.inline
32.19% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     6
94.04% :     0.000152s :      4: match.inline
 5.96% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.42% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000001s :     10: predicate.addn_zero_filter
 2.28% :     0.000003s :     10: predicate.arithmetic_simplify
 1.14% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000001s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.67% :     0.000007s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000001s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 2.00% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.94% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     15: predicate.partial_eliminate
 1.24% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.86% :     0.000002s :     10: predicate.reduce_eliminate
 1.69% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.34% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.57% :     0.000003s :     16: predicate.switch_defer_inline
 2.26% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.99% :     0.000008s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.64% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.66% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.36% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000545    12
47.64% :     0.000260s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.17% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.18% :     0.000257s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012746    72
 0.37% :     0.000047s :      1: add_recomputation
 0.51% :     0.000065s :      1: auto_monad
 0.12% :     0.000016s :      1: auto_monad_reorder
 2.08% :     0.000266s :      1: bootstrap
 0.21% :     0.000027s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.15% :     0.000019s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.60% :     0.004920s :      1: jit_opt_a
 1.07% :     0.000136s :      1: jit_opt_after_cconv
 0.45% :     0.000058s :      1: jit_opt_b
 3.33% :     0.000424s :      1: loop_unroll
 3.77% :     0.000480s :      1: mutable_eliminate
 6.46% :     0.000824s :     26: opt.transform.jit_opt_a
 0.36% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.23% :     0.000030s :      4: opt.transform.jit_opt_b
 0.10% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000029s :      4: opt.transform.symbol_engine_opt
 3.51% :     0.000447s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000035s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000016s :      1: remove_dup_value
 2.20% :     0.000281s :      1: renormalize.infer
 2.08% :     0.000265s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000033s :      1: rewriter_after_opt_a
 0.56% :     0.000072s :      1: rewriter_before_opt_a
 0.56% :     0.000072s :      1: symbol_engine_optimizer
31.42% :     0.004004s :      1: type_inference


TotalTime = 0.0120385, [30]
[bootstrap]: 0.00027351
[type_inference]: 0.00444526
[event_method]: 1.707e-05
[auto_monad]: 6.146e-05
[graph_reusing]: 5.98998e-06
[pre_auto_parallel]: 2.52001e-06
[py_interpret_to_execute]: 3.075e-05
[rewriter_before_opt_a]: 7.052e-05
[expand_dump_flag]: 3.37997e-06
[jit_opt_a]: 0.00515301, [2]
    [Cycle 1]: 0.00149179, [27]
        [switch_simplify]: 4.966e-05
        [loop_unroll]: 2.886e-05
        [a_1]: 0.00052698
        [with_stream_mark]: 1.595e-05
        [recompute_prepare]: 7.46999e-06
        [updatestate_depend_eliminate]: 3.36001e-06
        [updatestate_assign_eliminate]: 3.42002e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 1.88002e-06
        [specialize_transform]: 6.86001e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 6.38e-06
        [meta_shard_fg_expand]: 2.22999e-06
        [get_grad_eliminate_]: 5.13002e-06
        [merge_forward]: 3.82998e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.53e-05
        [j_node_and_user_rematch]: 9.70002e-06
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 8.77e-06
        [inline_without_move]: 5.46e-06
        [renormalize]: 0.00054683
        [add_forward_monad_depend]: 5.67999e-06
        [auto_monad_grad]: 1.67999e-06
        [auto_monad_eliminator]: 1.255e-05
        [cse]: 2.7e-05
        [replace_applicator]: 1.095e-05
    [Cycle 2]: 0.00031371, [27]
        [switch_simplify]: 5.55001e-06
        [loop_unroll]: 4.73001e-06
        [a_1]: 8.896e-05
        [with_stream_mark]: 8.45999e-06
        [recompute_prepare]: 5.05999e-06
        [updatestate_depend_eliminate]: 2.54001e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.02998e-06
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 4.92e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 4.33001e-06
        [merge_forward]: 2.30002e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.23e-05
        [j_node_and_user_rematch]: 7.9e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 7.11001e-06
        [inline_without_move]: 4.66002e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.30999e-06
        [auto_monad_grad]: 9.29984e-07
        [auto_monad_eliminator]: 4.78001e-06
        [cse]: 1.024e-05
        [replace_applicator]: 4.77e-06
[py_interpret_to_execute_after_opt_a]: 7.58001e-06
[rewriter_after_opt_a]: 3.04e-05
[convert_after_rewriter]: 7.06001e-06
[order_py_execute_after_rewriter]: 5.10001e-06
[mutable_eliminate]: 0.00047405
[jit_opt_b]: 6.025e-05, [1]
    [Cycle 1]: 4.46e-05, [2]
        [frontend_op_eliminate]: 1.695e-05
        [inline_after_opt_a]: 1.499e-05
[cconv]: 2.246e-05
[loop_unroll]: 0.00041475
[jit_opt_after_cconv]: 0.00013318, [1]
    [Cycle 1]: 0.00012703, [11]
        [c_1]: 2e-05
        [parameter_eliminate]: 2.31e-06
        [updatestate_depend_eliminate]: 5.14998e-06
        [updatestate_assign_eliminate]: 2.23002e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.595e-05
        [call_graph_tuple_transform]: 1.822e-05
        [tuple_list_get_item_eliminator]: 5.30001e-06
        [none_parameter_eliminate]: 1.51002e-06
        [renormalize]: 3.00002e-07
        [switch_simplify]: 5.11997e-06
[remove_dup_value]: 1.314e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 4.78001e-06
[add_recomputation]: 4.631e-05
[cse_after_recomputation]: 1.982e-05, [1]
    [Cycle 1]: 1.43e-05, [1]
        [cse]: 8.69e-06
[auto_monad_reorder]: 1.471e-05
[get_jit_bprop_graph]: 1.53002e-06
[rewriter_after_jit_bprop_graph]: 4.25e-06
[opt_after_jit_grad]: 0.00045016
[symbol_engine_optimizer]: 7.102e-05, [1]
    [Cycle 1]: 6.434e-05, [6]
        [build]: 3.31001e-06
        [elim_shapecalc]: 7.06001e-06
        [elim_not_effective]: 1.222e-05
        [opt_reshape]: 5.70001e-06
        [fold_const_symbol]: 8.54e-06
        [renormalize]: 4.00003e-07
[validate]: 3.196e-05
Sums
    bootstrap                            :     0.000274s :  3.37%
    type_inference                       :     0.004445s : 54.79%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000061s :  0.76%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000031s :  0.38%
    rewriter_before_opt_a                :     0.000071s :  0.87%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000055s :  0.68%
    jit_opt_a.loop_unroll                :     0.000034s :  0.41%
    jit_opt_a.a_1                        :     0.000616s :  7.59%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000547s :  6.74%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.21%
    jit_opt_a.cse                        :     0.000037s :  0.46%
    jit_opt_a.replace_applicator         :     0.000016s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.09%
    rewriter_after_opt_a                 :     0.000030s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000474s :  5.84%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000022s :  0.28%
    loop_unroll                          :     0.000415s :  5.11%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000013s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000046s :  0.57%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.05%
    opt_after_jit_grad                   :     0.000450s :  5.55%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.39%


Time group info:
------[substitution.]   0.000185    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.65% :     0.000005s :      3: substitution.graph_param_transform
82.91% :     0.000153s :      4: substitution.inline
 1.52% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.73% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000004s :      2: substitution.replace_old_param
 6.39% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004383     2
80.12% :     0.003512s :      1: type_inference.infer
19.88% :     0.000871s :      1: type_inference.specialize
------[replace.]   0.000065     6
66.77% :     0.000043s :      4: replace.inline
33.23% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     6
93.40% :     0.000151s :      4: match.inline
 6.60% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.20% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.18% :     0.000001s :     10: predicate.addn_zero_filter
 2.39% :     0.000003s :     10: predicate.arithmetic_simplify
 1.15% :     0.000001s :     10: predicate.cast_eliminate
 0.60% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000001s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.13% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.41% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.16% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.65% :     0.000002s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.60% :     0.000002s :     15: predicate.load_eliminater
 1.33% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.75% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 1.44% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 7.44% :     0.000009s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.81% :     0.000002s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.08% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.25% :     0.000003s :     16: predicate.switch_defer_inline
 1.96% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.09% :     0.000009s :     53: predicate.switch_simplify
 1.05% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.46% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000609    12
49.21% :     0.000300s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.33% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.46% :     0.000283s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013461    72
 0.36% :     0.000049s :      1: add_recomputation
 0.48% :     0.000065s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.15% :     0.000290s :      1: bootstrap
 0.19% :     0.000025s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.30% :     0.005156s :      1: jit_opt_a
 1.01% :     0.000136s :      1: jit_opt_after_cconv
 0.47% :     0.000063s :      1: jit_opt_b
 3.13% :     0.000422s :      1: loop_unroll
 3.59% :     0.000483s :      1: mutable_eliminate
 6.09% :     0.000819s :     26: opt.transform.jit_opt_a
 0.34% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.40% :     0.000457s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000033s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000015s :      1: remove_dup_value
 1.96% :     0.000264s :      1: renormalize.infer
 2.05% :     0.000275s :      1: renormalize.specialize
 0.05% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000033s :      1: rewriter_after_opt_a
 0.54% :     0.000073s :      1: rewriter_before_opt_a
 0.55% :     0.000073s :      1: symbol_engine_optimizer
33.11% :     0.004457s :      1: type_inference


TotalTime = 0.0161117, [33]
[bootstrap]: 0.00026061
[type_inference]: 0.00396051
[event_method]: 1.611e-05
[auto_monad]: 4.773e-05
[graph_reusing]: 4.07003e-06
[pre_auto_parallel]: 2.13002e-06
[py_interpret_to_execute]: 2.868e-05
[rewriter_before_opt_a]: 6.453e-05
[expand_dump_flag]: 2.17001e-06
[jit_opt_a]: 0.0045831, [2]
    [Cycle 1]: 0.00134202, [27]
        [switch_simplify]: 4.503e-05
        [loop_unroll]: 3.022e-05
        [a_1]: 0.00048187
        [with_stream_mark]: 1.379e-05
        [recompute_prepare]: 7.2e-06
        [updatestate_depend_eliminate]: 3.02002e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 5.79e-06
        [updatestate_useless_node_eliminater]: 5.15999e-06
        [accelerated_algorithm]: 5.51998e-06
        [meta_shard_fg_expand]: 1.58002e-06
        [get_grad_eliminate_]: 5.52999e-06
        [merge_forward]: 2.78998e-06
        [cell_reuse_recompute_pass]: 1.00999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.137e-05
        [j_node_and_user_rematch]: 8.61002e-06
        [meta_fg_expand]: 2.44999e-06
        [replace_old_param]: 7.73001e-06
        [inline_without_move]: 5.17e-06
        [renormalize]: 0.00048438
        [add_forward_monad_depend]: 4.58999e-06
        [auto_monad_grad]: 1.49e-06
        [auto_monad_eliminator]: 1.033e-05
        [cse]: 1.894e-05
        [replace_applicator]: 1.217e-05
    [Cycle 2]: 0.00032162, [27]
        [switch_simplify]: 5.69e-06
        [loop_unroll]: 4.62e-06
        [a_1]: 8.972e-05
        [with_stream_mark]: 9.19e-06
        [recompute_prepare]: 5.07999e-06
        [updatestate_depend_eliminate]: 2.74999e-06
        [updatestate_assign_eliminate]: 2.01998e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 4.75001e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.03002e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.62e-06
        [merge_forward]: 2.27999e-06
        [cell_reuse_recompute_pass]: 1.38002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.198e-05
        [j_node_and_user_rematch]: 8.17e-06
        [meta_fg_expand]: 1.81e-06
        [replace_old_param]: 7.61999e-06
        [inline_without_move]: 4.98001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 5.07999e-06
        [cse]: 1.062e-05
        [replace_applicator]: 5.35999e-06
[py_interpret_to_execute_after_opt_a]: 9.59e-06
[rewriter_after_opt_a]: 3.006e-05
[convert_after_rewriter]: 5.82001e-06
[order_py_execute_after_rewriter]: 6.37001e-06
[mutable_eliminate]: 0.00053204
[jit_opt_b]: 4.883e-05, [1]
    [Cycle 1]: 4.197e-05, [2]
        [frontend_op_eliminate]: 1.604e-05
        [inline_after_opt_a]: 1.431e-05
[cconv]: 1.957e-05
[loop_unroll]: 0.00041416
[jit_opt_after_cconv]: 0.00013736, [1]
    [Cycle 1]: 0.00013173, [11]
        [c_1]: 2.102e-05
        [parameter_eliminate]: 2.42001e-06
        [updatestate_depend_eliminate]: 5.97999e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 1.833e-05
        [call_graph_tuple_transform]: 1.805e-05
        [tuple_list_get_item_eliminator]: 5.25001e-06
        [none_parameter_eliminate]: 1.09998e-06
        [renormalize]: 5.50004e-07
        [switch_simplify]: 5.04e-06
[remove_dup_value]: 1.01e-05
[partial_unused_args_eliminate]: 1.91998e-06
[environ_conv]: 4.03999e-06
[add_recomputation]: 3.857e-05
[cse_after_recomputation]: 2.025e-05, [1]
    [Cycle 1]: 1.55e-05, [1]
        [cse]: 9.93002e-06
[auto_monad_reorder]: 1.088e-05
[get_jit_bprop_graph]: 1.89e-06
[rewriter_after_jit_bprop_graph]: 4.43999e-06
[opt_after_jit_grad]: 0.00044171
[symbol_engine_optimizer]: 7.069e-05, [1]
    [Cycle 1]: 6.475e-05, [6]
        [build]: 2.58003e-06
        [elim_shapecalc]: 7.44002e-06
        [elim_not_effective]: 1.277e-05
        [opt_reshape]: 6.02999e-06
        [fold_const_symbol]: 8.50001e-06
        [renormalize]: 3.89991e-07
[validate]: 3.095e-05
[backend_pass]: 6.69999e-07
[task_emit]: 0.00512464
[execute]: 5.81998e-06
Sums
    bootstrap                            :     0.000261s :  2.06%
    type_inference                       :     0.003961s : 31.37%
    event_method                         :     0.000016s :  0.13%
    auto_monad                           :     0.000048s :  0.38%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.23%
    rewriter_before_opt_a                :     0.000065s :  0.51%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000051s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.28%
    jit_opt_a.a_1                        :     0.000572s :  4.53%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000484s :  3.84%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.12%
    jit_opt_a.cse                        :     0.000030s :  0.23%
    jit_opt_a.replace_applicator         :     0.000018s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.08%
    rewriter_after_opt_a                 :     0.000030s :  0.24%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000532s :  4.21%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000020s :  0.16%
    loop_unroll                          :     0.000414s :  3.28%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.17%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000010s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000039s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.08%
    auto_monad_reorder                   :     0.000011s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.04%
    opt_after_jit_grad                   :     0.000442s :  3.50%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005125s : 40.59%
    execute                              :     0.000006s :  0.05%


Time group info:
------[substitution.]   0.000162    23
 1.07% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000001s :      2: substitution.fold_const_symbol
 3.00% :     0.000005s :      3: substitution.graph_param_transform
82.80% :     0.000134s :      4: substitution.inline
 1.70% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.48% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.95% :     0.000003s :      2: substitution.replace_old_param
 6.26% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003914     2
82.20% :     0.003217s :      1: type_inference.infer
17.80% :     0.000697s :      1: type_inference.specialize
------[replace.]   0.000061     6
66.42% :     0.000041s :      4: replace.inline
33.58% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000140     6
93.69% :     0.000131s :      4: match.inline
 6.31% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.24% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 1.92% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000001s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000001s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.74% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.56% :     0.000002s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.66% :     0.000006s :     22: predicate.inline
 1.05% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.24% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.58% :     0.000002s :      3: predicate.mutable_eliminate
 0.59% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000002s :     15: predicate.partial_eliminate
 1.30% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.60% :     0.000002s :     10: predicate.reshape_eliminate
 1.33% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.05% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.50% :     0.000002s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.46% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.47% :     0.000008s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.53% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.43% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000491    12
48.31% :     0.000237s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.12% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.57% :     0.000224s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017474    76
 0.24% :     0.000041s :      1: add_recomputation
 0.29% :     0.000051s :      1: auto_monad
 0.08% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.57% :     0.000274s :      1: bootstrap
 0.13% :     0.000022s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.13% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.03% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
26.24% :     0.004586s :      1: jit_opt_a
 0.80% :     0.000140s :      1: jit_opt_after_cconv
 0.30% :     0.000052s :      1: jit_opt_b
 2.41% :     0.000422s :      1: loop_unroll
 3.09% :     0.000540s :      1: mutable_eliminate
 4.40% :     0.000770s :     26: opt.transform.jit_opt_a
 0.26% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.57% :     0.000449s :      1: opt_after_jit_grad
 0.05% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000031s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000012s :      1: remove_dup_value
 1.40% :     0.000245s :      1: renormalize.infer
 1.33% :     0.000232s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000033s :      1: rewriter_after_opt_a
 0.38% :     0.000067s :      1: rewriter_before_opt_a
 0.42% :     0.000073s :      1: symbol_engine_optimizer
29.39% :     0.005135s :      1: task_emit
22.73% :     0.003972s :      1: type_inference
 0.27% :     0.000047s :      1: validate


TotalTime = 0.0122714, [30]
[bootstrap]: 0.00028874
[type_inference]: 0.00460184
[event_method]: 1.608e-05
[auto_monad]: 6.217e-05
[graph_reusing]: 5.97001e-06
[pre_auto_parallel]: 2.53998e-06
[py_interpret_to_execute]: 3.162e-05
[rewriter_before_opt_a]: 7.07e-05
[expand_dump_flag]: 2.99999e-06
[jit_opt_a]: 0.00506813, [2]
    [Cycle 1]: 0.00145502, [27]
        [switch_simplify]: 4.977e-05
        [loop_unroll]: 2.759e-05
        [a_1]: 0.00050708
        [with_stream_mark]: 1.659e-05
        [recompute_prepare]: 7.48999e-06
        [updatestate_depend_eliminate]: 3.51999e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.76e-06
        [parameter_eliminate]: 2.17001e-06
        [specialize_transform]: 6.17999e-06
        [updatestate_useless_node_eliminater]: 4.89998e-06
        [accelerated_algorithm]: 5.69e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 4.90999e-06
        [merge_forward]: 3.47002e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.47e-05
        [j_node_and_user_rematch]: 8.99998e-06
        [meta_fg_expand]: 2.73e-06
        [replace_old_param]: 8.86002e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.00054477
        [add_forward_monad_depend]: 5.25001e-06
        [auto_monad_grad]: 2.01e-06
        [auto_monad_eliminator]: 1.289e-05
        [cse]: 2.812e-05
        [replace_applicator]: 1.155e-05
    [Cycle 2]: 0.00030995, [27]
        [switch_simplify]: 5.61003e-06
        [loop_unroll]: 4.70999e-06
        [a_1]: 8.875e-05
        [with_stream_mark]: 8.33001e-06
        [recompute_prepare]: 4.84e-06
        [updatestate_depend_eliminate]: 2.41e-06
        [updatestate_assign_eliminate]: 1.91998e-06
        [updatestate_loads_eliminate]: 1.83002e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 4.95999e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 4.92e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 4.45999e-06
        [merge_forward]: 2.25002e-06
        [cell_reuse_recompute_pass]: 1.10999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.141e-05
        [j_node_and_user_rematch]: 7.58999e-06
        [meta_fg_expand]: 1.46002e-06
        [replace_old_param]: 6.63e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 7.10017e-07
        [auto_monad_eliminator]: 4.74998e-06
        [cse]: 9.45001e-06
        [replace_applicator]: 5.03002e-06
[py_interpret_to_execute_after_opt_a]: 7.31999e-06
[rewriter_after_opt_a]: 3.178e-05
[convert_after_rewriter]: 6.44001e-06
[order_py_execute_after_rewriter]: 5.22e-06
[mutable_eliminate]: 0.00051178
[jit_opt_b]: 4.845e-05, [1]
    [Cycle 1]: 4.196e-05, [2]
        [frontend_op_eliminate]: 1.605e-05
        [inline_after_opt_a]: 1.439e-05
[cconv]: 2.571e-05
[loop_unroll]: 0.0004542
[jit_opt_after_cconv]: 0.00013653, [1]
    [Cycle 1]: 0.00013, [11]
        [c_1]: 2.056e-05
        [parameter_eliminate]: 2.38998e-06
        [updatestate_depend_eliminate]: 5.51e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 1.675e-05
        [call_graph_tuple_transform]: 1.828e-05
        [tuple_list_get_item_eliminator]: 5.45001e-06
        [none_parameter_eliminate]: 1.76e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 5.24e-06
[remove_dup_value]: 1.527e-05
[partial_unused_args_eliminate]: 2.32999e-06
[environ_conv]: 5.62999e-06
[add_recomputation]: 4.725e-05
[cse_after_recomputation]: 1.933e-05, [1]
    [Cycle 1]: 1.4e-05, [1]
        [cse]: 8.25e-06
[auto_monad_reorder]: 1.474e-05
[get_jit_bprop_graph]: 1.62999e-06
[rewriter_after_jit_bprop_graph]: 5.69e-06
[opt_after_jit_grad]: 0.00050267
[symbol_engine_optimizer]: 7.388e-05, [1]
    [Cycle 1]: 6.736e-05, [6]
        [build]: 3.47002e-06
        [elim_shapecalc]: 7.84002e-06
        [elim_not_effective]: 1.339e-05
        [opt_reshape]: 5.73002e-06
        [fold_const_symbol]: 8.48001e-06
        [renormalize]: 6.00005e-07
[validate]: 3.651e-05
Sums
    bootstrap                            :     0.000289s :  3.44%
    type_inference                       :     0.004602s : 54.75%
    event_method                         :     0.000016s :  0.19%
    auto_monad                           :     0.000062s :  0.74%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.38%
    rewriter_before_opt_a                :     0.000071s :  0.84%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000055s :  0.66%
    jit_opt_a.loop_unroll                :     0.000032s :  0.38%
    jit_opt_a.a_1                        :     0.000596s :  7.09%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.18%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000545s :  6.48%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.21%
    jit_opt_a.cse                        :     0.000038s :  0.45%
    jit_opt_a.replace_applicator         :     0.000017s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.09%
    rewriter_after_opt_a                 :     0.000032s :  0.38%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000512s :  6.09%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.17%
    cconv                                :     0.000026s :  0.31%
    loop_unroll                          :     0.000454s :  5.40%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000047s :  0.56%
    cse_after_recomputation.cse          :     0.000008s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000503s :  5.98%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000037s :  0.43%


Time group info:
------[substitution.]   0.000181    23
 1.07% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 3.13% :     0.000006s :      3: substitution.graph_param_transform
82.43% :     0.000149s :      4: substitution.inline
 1.54% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.48% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.81% :     0.000003s :      2: substitution.replace_old_param
 6.82% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004538     2
81.25% :     0.003687s :      1: type_inference.infer
18.75% :     0.000851s :      1: type_inference.specialize
------[replace.]   0.000062     6
68.53% :     0.000042s :      4: replace.inline
31.47% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000158     6
92.88% :     0.000147s :      4: match.inline
 7.12% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.48% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 2.13% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.58% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000001s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.64% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.08% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.85% :     0.000007s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000001s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.89% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.93% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.58% :     0.000002s :      3: predicate.mutable_eliminate
 0.59% :     0.000001s :      3: predicate.opt_reshape
 2.55% :     0.000003s :     15: predicate.partial_eliminate
 1.37% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.22% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.81% :     0.000009s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000616    12
52.83% :     0.000325s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.32% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.85% :     0.000264s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013661    72
 0.37% :     0.000050s :      1: add_recomputation
 0.48% :     0.000066s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.24% :     0.000305s :      1: bootstrap
 0.21% :     0.000028s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000021s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
37.12% :     0.005071s :      1: jit_opt_a
 1.02% :     0.000139s :      1: jit_opt_after_cconv
 0.37% :     0.000051s :      1: jit_opt_b
 3.38% :     0.000462s :      1: loop_unroll
 3.81% :     0.000521s :      1: mutable_eliminate
 5.81% :     0.000794s :     26: opt.transform.jit_opt_a
 0.34% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000024s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.75% :     0.000512s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000034s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000018s :      1: remove_dup_value
 1.95% :     0.000267s :      1: renormalize.infer
 1.97% :     0.000270s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000035s :      1: rewriter_after_opt_a
 0.54% :     0.000074s :      1: rewriter_before_opt_a
 0.56% :     0.000076s :      1: symbol_engine_optimizer
33.77% :     0.004614s :      1: type_inference


TotalTime = 0.0127335, [30]
[bootstrap]: 0.00028023
[type_inference]: 0.00434719
[event_method]: 1.545e-05
[auto_monad]: 5.995e-05
[graph_reusing]: 5.34998e-06
[pre_auto_parallel]: 2.46998e-06
[py_interpret_to_execute]: 3.049e-05
[rewriter_before_opt_a]: 7.054e-05
[expand_dump_flag]: 2.96999e-06
[jit_opt_a]: 0.00536016, [2]
    [Cycle 1]: 0.00183727, [27]
        [switch_simplify]: 4.994e-05
        [loop_unroll]: 2.855e-05
        [a_1]: 0.00053234
        [with_stream_mark]: 1.92e-05
        [recompute_prepare]: 1.01e-05
        [updatestate_depend_eliminate]: 4.27e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.61999e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 6.49001e-06
        [updatestate_useless_node_eliminater]: 5.65001e-06
        [accelerated_algorithm]: 5.95002e-06
        [meta_shard_fg_expand]: 2.80002e-06
        [get_grad_eliminate_]: 5.45001e-06
        [merge_forward]: 3.93001e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.749e-05
        [j_node_and_user_rematch]: 9.34e-06
        [meta_fg_expand]: 3.18e-06
        [replace_old_param]: 9.87001e-06
        [inline_without_move]: 6.04001e-06
        [renormalize]: 0.00084061
        [add_forward_monad_depend]: 6.69999e-06
        [auto_monad_grad]: 2.46e-06
        [auto_monad_eliminator]: 1.762e-05
        [cse]: 3.291e-05
        [replace_applicator]: 1.48e-05
    [Cycle 2]: 0.00037603, [27]
        [switch_simplify]: 5.45001e-06
        [loop_unroll]: 4.85999e-06
        [a_1]: 9.672e-05
        [with_stream_mark]: 1.231e-05
        [recompute_prepare]: 6.06e-06
        [updatestate_depend_eliminate]: 3.30998e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [parameter_eliminate]: 1.30999e-06
        [specialize_transform]: 5.54e-06
        [updatestate_useless_node_eliminater]: 5.47001e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.71998e-06
        [get_grad_eliminate_]: 4.45999e-06
        [merge_forward]: 3.31001e-06
        [cell_reuse_recompute_pass]: 1.74e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.656e-05
        [j_node_and_user_rematch]: 8.57998e-06
        [meta_fg_expand]: 2.45002e-06
        [replace_old_param]: 8.93002e-06
        [inline_without_move]: 4.94998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.83002e-06
        [auto_monad_grad]: 1.27e-06
        [auto_monad_eliminator]: 8.50001e-06
        [cse]: 1.489e-05
        [replace_applicator]: 7.48e-06
[py_interpret_to_execute_after_opt_a]: 1.434e-05
[rewriter_after_opt_a]: 4.009e-05
[convert_after_rewriter]: 7.75e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00066519
[jit_opt_b]: 5.753e-05, [1]
    [Cycle 1]: 4.823e-05, [2]
        [frontend_op_eliminate]: 1.917e-05
        [inline_after_opt_a]: 1.62e-05
[cconv]: 3.098e-05
[loop_unroll]: 0.00051531
[jit_opt_after_cconv]: 0.0001824, [1]
    [Cycle 1]: 0.00017443, [11]
        [c_1]: 2.287e-05
        [parameter_eliminate]: 4e-06
        [updatestate_depend_eliminate]: 7.06999e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 3.06001e-06
        [cse]: 2.339e-05
        [call_graph_tuple_transform]: 2.245e-05
        [tuple_list_get_item_eliminator]: 5.89e-06
        [none_parameter_eliminate]: 2.24001e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 5.30999e-06
[remove_dup_value]: 1.614e-05
[partial_unused_args_eliminate]: 2.20002e-06
[environ_conv]: 5.79e-06
[add_recomputation]: 5.535e-05
[cse_after_recomputation]: 2.728e-05, [1]
    [Cycle 1]: 1.844e-05, [1]
        [cse]: 1.031e-05
[auto_monad_reorder]: 1.695e-05
[get_jit_bprop_graph]: 1.81e-06
[rewriter_after_jit_bprop_graph]: 7.22002e-06
[opt_after_jit_grad]: 0.00057615
[symbol_engine_optimizer]: 9.031e-05, [1]
    [Cycle 1]: 8.207e-05, [6]
        [build]: 6.17999e-06
        [elim_shapecalc]: 1.368e-05
        [elim_not_effective]: 1.599e-05
        [opt_reshape]: 6.63e-06
        [fold_const_symbol]: 8.99e-06
        [renormalize]: 5.89993e-07
[validate]: 4.234e-05
Sums
    bootstrap                            :     0.000280s :  3.15%
    type_inference                       :     0.004347s : 48.87%
    event_method                         :     0.000015s :  0.17%
    auto_monad                           :     0.000060s :  0.67%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000030s :  0.34%
    rewriter_before_opt_a                :     0.000071s :  0.79%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000055s :  0.62%
    jit_opt_a.loop_unroll                :     0.000033s :  0.38%
    jit_opt_a.a_1                        :     0.000629s :  7.07%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.18%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.09%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.38%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.06%
    jit_opt_a.replace_old_param          :     0.000019s :  0.21%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000841s :  9.45%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.29%
    jit_opt_a.cse                        :     0.000048s :  0.54%
    jit_opt_a.replace_applicator         :     0.000022s :  0.25%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.16%
    rewriter_after_opt_a                 :     0.000040s :  0.45%
    convert_after_rewriter               :     0.000008s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000665s :  7.48%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.18%
    cconv                                :     0.000031s :  0.35%
    loop_unroll                          :     0.000515s :  5.79%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000023s :  0.26%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.25%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000016s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000055s :  0.62%
    cse_after_recomputation.cse          :     0.000010s :  0.12%
    auto_monad_reorder                   :     0.000017s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.08%
    opt_after_jit_grad                   :     0.000576s :  6.48%
    symbol_engine_optimizer.build        :     0.000006s :  0.07%
    symbol_engine_optimizer.elim_shapecalc :     0.000014s :  0.15%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.18%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000042s :  0.48%


Time group info:
------[substitution.]   0.000193    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.82% :     0.000002s :      2: substitution.fold_const_symbol
 2.87% :     0.000006s :      3: substitution.graph_param_transform
79.64% :     0.000154s :      4: substitution.inline
 1.96% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.99% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.77% :     0.000005s :      2: substitution.replace_old_param
 7.90% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004286     2
80.82% :     0.003464s :      1: type_inference.infer
19.18% :     0.000822s :      1: type_inference.specialize
------[replace.]   0.000073     6
59.90% :     0.000044s :      4: replace.inline
40.10% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000165     6
91.80% :     0.000151s :      4: match.inline
 8.20% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.96% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.69% :     0.000002s :     10: predicate.addn_zero_filter
 1.75% :     0.000002s :     10: predicate.arithmetic_simplify
 1.06% :     0.000001s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.19% :     0.000002s :     10: predicate.depend_value_elim
 1.49% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.64% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.64% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.08% :     0.000001s :     10: predicate.environ_get_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.75% :     0.000001s :      6: predicate.get_grad_eliminate
 0.48% :     0.000001s :      3: predicate.graph_param_transform
 5.86% :     0.000007s :     22: predicate.inline
 1.03% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.89% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.56% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.24% :     0.000002s :     10: predicate.minmaximum_grad
 2.04% :     0.000003s :      3: predicate.mutable_eliminate
 0.39% :     0.000001s :      3: predicate.opt_reshape
 1.99% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.31% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.53% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.65% :     0.000002s :     10: predicate.reshape_eliminate
 1.39% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.33% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.63% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.13% :     0.000001s :      6: predicate.special_op_eliminate
 1.21% :     0.000002s :      6: predicate.specialize_transform
 1.22% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 1.86% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.90% :     0.000009s :     53: predicate.switch_simplify
 1.01% :     0.000001s :     10: predicate.tile_eliminate
 1.01% :     0.000001s :     10: predicate.transpose_eliminate
 1.20% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.68% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.69% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.87% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000626    12
50.07% :     0.000313s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.76% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.17% :     0.000283s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014497    72
 0.41% :     0.000059s :      1: add_recomputation
 0.44% :     0.000063s :      1: auto_monad
 0.14% :     0.000021s :      1: auto_monad_reorder
 2.07% :     0.000300s :      1: bootstrap
 0.23% :     0.000034s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.21% :     0.000030s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.13% :     0.000019s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
37.01% :     0.005365s :      1: jit_opt_a
 1.28% :     0.000186s :      1: jit_opt_after_cconv
 0.41% :     0.000060s :      1: jit_opt_b
 3.65% :     0.000528s :      1: loop_unroll
 4.68% :     0.000679s :      1: mutable_eliminate
 5.87% :     0.000851s :     26: opt.transform.jit_opt_a
 0.37% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000029s :      4: opt.transform.jit_opt_b
 0.11% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.13% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.20% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.29% :     0.000042s :      4: opt.transform.symbol_engine_opt
 4.06% :     0.000588s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.23% :     0.000033s :      1: py_interpret_to_execute
 0.12% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000019s :      1: remove_dup_value
 2.33% :     0.000338s :      1: renormalize.infer
 3.38% :     0.000491s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.30% :     0.000044s :      1: rewriter_after_opt_a
 0.50% :     0.000073s :      1: rewriter_before_opt_a
 0.64% :     0.000093s :      1: symbol_engine_optimizer
30.07% :     0.004359s :      1: type_inference


TotalTime = 0.0151486, [30]
[bootstrap]: 0.00032137
[type_inference]: 0.00471785
[event_method]: 1.681e-05
[auto_monad]: 6.64e-05
[graph_reusing]: 6.16e-06
[pre_auto_parallel]: 2.88e-06
[py_interpret_to_execute]: 3.697e-05
[rewriter_before_opt_a]: 7.377e-05
[expand_dump_flag]: 3.42997e-06
[jit_opt_a]: 0.0068388, [2]
    [Cycle 1]: 0.002383, [27]
        [switch_simplify]: 7.387e-05
        [loop_unroll]: 3.076e-05
        [a_1]: 0.00074384
        [with_stream_mark]: 3.056e-05
        [recompute_prepare]: 8.95001e-06
        [updatestate_depend_eliminate]: 4.25999e-06
        [updatestate_assign_eliminate]: 3.05002e-06
        [updatestate_loads_eliminate]: 2.64999e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 6.54999e-06
        [updatestate_useless_node_eliminater]: 6.46e-06
        [accelerated_algorithm]: 6.69999e-06
        [meta_shard_fg_expand]: 2.94001e-06
        [get_grad_eliminate_]: 7.26999e-06
        [merge_forward]: 3.61999e-06
        [cell_reuse_recompute_pass]: 1.56002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.659e-05
        [j_node_and_user_rematch]: 1.153e-05
        [meta_fg_expand]: 2.69001e-06
        [replace_old_param]: 2.207e-05
        [inline_without_move]: 6.94001e-06
        [renormalize]: 0.00106438
        [add_forward_monad_depend]: 1.068e-05
        [auto_monad_grad]: 3.00002e-06
        [auto_monad_eliminator]: 2.252e-05
        [cse]: 3.393e-05
        [replace_applicator]: 2.204e-05
    [Cycle 2]: 0.00039551, [27]
        [switch_simplify]: 6.86999e-06
        [loop_unroll]: 6.54999e-06
        [a_1]: 0.00011506
        [with_stream_mark]: 1.809e-05
        [recompute_prepare]: 5.76e-06
        [updatestate_depend_eliminate]: 3.41999e-06
        [updatestate_assign_eliminate]: 2.87002e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [parameter_eliminate]: 1.90001e-06
        [specialize_transform]: 5.37001e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 6.34999e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 4.90999e-06
        [merge_forward]: 3.14001e-06
        [cell_reuse_recompute_pass]: 3.06999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.83e-05
        [j_node_and_user_rematch]: 9.44e-06
        [meta_fg_expand]: 2.23002e-06
        [replace_old_param]: 1.007e-05
        [inline_without_move]: 4.92e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.88002e-06
        [auto_monad_grad]: 1.14998e-06
        [auto_monad_eliminator]: 6.61999e-06
        [cse]: 1.422e-05
        [replace_applicator]: 9.07999e-06
[py_interpret_to_execute_after_opt_a]: 1.725e-05
[rewriter_after_opt_a]: 4.443e-05
[convert_after_rewriter]: 6.76e-06
[order_py_execute_after_rewriter]: 5.03002e-06
[mutable_eliminate]: 0.0008787
[jit_opt_b]: 6.343e-05, [1]
    [Cycle 1]: 5.213e-05, [2]
        [frontend_op_eliminate]: 1.872e-05
        [inline_after_opt_a]: 1.794e-05
[cconv]: 3.755e-05
[loop_unroll]: 0.00055292
[jit_opt_after_cconv]: 0.00026487, [1]
    [Cycle 1]: 0.00025625, [11]
        [c_1]: 2.588e-05
        [parameter_eliminate]: 6.43e-06
        [updatestate_depend_eliminate]: 1.126e-05
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.32001e-06
        [cse]: 3.705e-05
        [call_graph_tuple_transform]: 3.28e-05
        [tuple_list_get_item_eliminator]: 6.52001e-06
        [none_parameter_eliminate]: 2.61e-06
        [renormalize]: 5.79981e-07
        [switch_simplify]: 6.65998e-06
[remove_dup_value]: 1.764e-05
[partial_unused_args_eliminate]: 2.99001e-06
[environ_conv]: 7.08e-06
[add_recomputation]: 6.114e-05
[cse_after_recomputation]: 3.517e-05, [1]
    [Cycle 1]: 2.086e-05, [1]
        [cse]: 1.22e-05
[auto_monad_reorder]: 1.728e-05
[get_jit_bprop_graph]: 3.69002e-06
[rewriter_after_jit_bprop_graph]: 1.326e-05
[opt_after_jit_grad]: 0.00064769
[symbol_engine_optimizer]: 0.00010542, [1]
    [Cycle 1]: 9.605e-05, [6]
        [build]: 7.03e-06
        [elim_shapecalc]: 1.352e-05
        [elim_not_effective]: 1.939e-05
        [opt_reshape]: 9.44e-06
        [fold_const_symbol]: 1.1e-05
        [renormalize]: 6.30011e-07
[validate]: 5.732e-05
Sums
    bootstrap                            :     0.000321s :  3.13%
    type_inference                       :     0.004718s : 45.88%
    event_method                         :     0.000017s :  0.16%
    auto_monad                           :     0.000066s :  0.65%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000037s :  0.36%
    rewriter_before_opt_a                :     0.000074s :  0.72%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000081s :  0.79%
    jit_opt_a.loop_unroll                :     0.000037s :  0.36%
    jit_opt_a.a_1                        :     0.000859s :  8.35%
    jit_opt_a.with_stream_mark           :     0.000049s :  0.47%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.12%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000032s :  0.31%
    jit_opt_a.inline_without_move        :     0.000012s :  0.12%
    jit_opt_a.renormalize                :     0.001064s : 10.35%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.12%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000029s :  0.28%
    jit_opt_a.cse                        :     0.000048s :  0.47%
    jit_opt_a.replace_applicator         :     0.000031s :  0.30%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.17%
    rewriter_after_opt_a                 :     0.000044s :  0.43%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000879s :  8.54%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.17%
    cconv                                :     0.000038s :  0.37%
    loop_unroll                          :     0.000553s :  5.38%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.11%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000037s :  0.36%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000033s :  0.32%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.06%
    remove_dup_value                     :     0.000018s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000007s :  0.07%
    add_recomputation                    :     0.000061s :  0.59%
    cse_after_recomputation.cse          :     0.000012s :  0.12%
    auto_monad_reorder                   :     0.000017s :  0.17%
    get_jit_bprop_graph                  :     0.000004s :  0.04%
    rewriter_after_jit_bprop_graph       :     0.000013s :  0.13%
    opt_after_jit_grad                   :     0.000648s :  6.30%
    symbol_engine_optimizer.build        :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_shapecalc :     0.000014s :  0.13%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.19%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.09%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000057s :  0.56%


Time group info:
------[substitution.]   0.000376    23
 0.65% :     0.000002s :      2: substitution.elim_not_effective
 0.39% :     0.000001s :      2: substitution.fold_const_symbol
 1.93% :     0.000007s :      3: substitution.graph_param_transform
87.77% :     0.000330s :      4: substitution.inline
 1.48% :     0.000006s :      4: substitution.j_node_and_user_rematch
 1.43% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.26% :     0.000008s :      2: substitution.replace_old_param
 4.09% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004645     2
80.35% :     0.003732s :      1: type_inference.infer
19.65% :     0.000913s :      1: type_inference.specialize
------[replace.]   0.000090     6
66.80% :     0.000060s :      4: replace.inline
33.20% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000341     6
95.85% :     0.000327s :      4: match.inline
 4.15% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000155   691
 1.13% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.79% :     0.000001s :     10: predicate.addn_check_dump
 1.14% :     0.000002s :     10: predicate.addn_zero_filter
 2.58% :     0.000004s :     10: predicate.arithmetic_simplify
 1.32% :     0.000002s :     10: predicate.cast_eliminate
 0.32% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.79% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 0.92% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.97% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.20% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000001s :      3: predicate.elim_not_effective
 0.76% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.94% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.81% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.99% :     0.000002s :     10: predicate.environ_get_eliminate
 0.87% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000002s :      6: predicate.get_grad_eliminate
 0.38% :     0.000001s :      3: predicate.graph_param_transform
 6.18% :     0.000010s :     22: predicate.inline
 0.69% :     0.000001s :      6: predicate.inline_without_move
 0.28% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.48% :     0.000002s :      6: predicate.less_batch_normalization
 1.27% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.42% :     0.000002s :     15: predicate.load_eliminater
 1.69% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.55% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.81% :     0.000001s :     10: predicate.merge_addn
 0.89% :     0.000001s :     10: predicate.minmaximum_grad
 2.65% :     0.000004s :      3: predicate.mutable_eliminate
 0.75% :     0.000001s :      3: predicate.opt_reshape
 1.82% :     0.000003s :     15: predicate.partial_eliminate
 1.26% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.63% :     0.000003s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.52% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 0.99% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.61% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 0.85% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.52% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.30% :     0.000004s :     16: predicate.switch_defer_inline
 1.57% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.42% :     0.000011s :     53: predicate.switch_simplify
 1.00% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000002s :     10: predicate.transpose_eliminate
 1.14% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.06% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.63% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.25% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.45% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 6.95% :     0.000011s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000803    12
43.41% :     0.000349s :      5: func_graph_cloner_run.FuncGraphClonerGraph
15.43% :     0.000124s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.16% :     0.000330s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017435    72
 0.37% :     0.000065s :      1: add_recomputation
 0.40% :     0.000070s :      1: auto_monad
 0.12% :     0.000020s :      1: auto_monad_reorder
 1.97% :     0.000343s :      1: bootstrap
 0.24% :     0.000041s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.22% :     0.000038s :      1: cse_after_recomputation
 0.06% :     0.000010s :      1: environ_conv
 0.12% :     0.000021s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000006s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.25% :     0.006843s :      1: jit_opt_a
 1.54% :     0.000268s :      1: jit_opt_after_cconv
 0.39% :     0.000067s :      1: jit_opt_b
 3.25% :     0.000567s :      1: loop_unroll
 5.14% :     0.000896s :      1: mutable_eliminate
 6.51% :     0.001135s :     26: opt.transform.jit_opt_a
 0.39% :     0.000067s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000029s :      4: opt.transform.jit_opt_b
 0.10% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.14% :     0.000025s :      1: opt.transform.mutable_eliminate
 0.24% :     0.000042s :      1: opt.transform.opt_after_jit_grad
 0.27% :     0.000048s :      4: opt.transform.symbol_engine_opt
 3.80% :     0.000662s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000039s :      1: py_interpret_to_execute
 0.11% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000021s :      1: remove_dup_value
 3.33% :     0.000581s :      1: renormalize.infer
 2.70% :     0.000471s :      1: renormalize.specialize
 0.09% :     0.000016s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000048s :      1: rewriter_after_opt_a
 0.44% :     0.000077s :      1: rewriter_before_opt_a
 0.63% :     0.000109s :      1: symbol_engine_optimizer
27.15% :     0.004733s :      1: type_inference


TotalTime = 0.0212034, [33]
[bootstrap]: 0.00031616
[type_inference]: 0.00490309
[event_method]: 1.845e-05
[auto_monad]: 6.524e-05
[graph_reusing]: 5.77999e-06
[pre_auto_parallel]: 2.47001e-06
[py_interpret_to_execute]: 3.278e-05
[rewriter_before_opt_a]: 7.388e-05
[expand_dump_flag]: 3.06999e-06
[jit_opt_a]: 0.00589519, [2]
    [Cycle 1]: 0.00183779, [27]
        [switch_simplify]: 5.986e-05
        [loop_unroll]: 3.006e-05
        [a_1]: 0.0006012
        [with_stream_mark]: 2.411e-05
        [recompute_prepare]: 9.75002e-06
        [updatestate_depend_eliminate]: 4.51002e-06
        [updatestate_assign_eliminate]: 3.16999e-06
        [updatestate_loads_eliminate]: 2.79001e-06
        [parameter_eliminate]: 2.58e-06
        [specialize_transform]: 6.52001e-06
        [updatestate_useless_node_eliminater]: 4.99003e-06
        [accelerated_algorithm]: 6.46e-06
        [meta_shard_fg_expand]: 2.27001e-06
        [get_grad_eliminate_]: 5.59e-06
        [merge_forward]: 4.90999e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.713e-05
        [j_node_and_user_rematch]: 1.006e-05
        [meta_fg_expand]: 3.35e-06
        [replace_old_param]: 1.016e-05
        [inline_without_move]: 5.22e-06
        [renormalize]: 0.00074558
        [add_forward_monad_depend]: 6.23e-06
        [auto_monad_grad]: 2.22001e-06
        [auto_monad_eliminator]: 1.692e-05
        [cse]: 2.952e-05
        [replace_applicator]: 1.59e-05
    [Cycle 2]: 0.00037037, [27]
        [switch_simplify]: 7.46999e-06
        [loop_unroll]: 5.19e-06
        [a_1]: 9.227e-05
        [with_stream_mark]: 1.264e-05
        [recompute_prepare]: 5.99999e-06
        [updatestate_depend_eliminate]: 3.51001e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.71999e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 5.11002e-06
        [updatestate_useless_node_eliminater]: 5.69999e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 2.14999e-06
        [get_grad_eliminate_]: 4.78001e-06
        [merge_forward]: 3.88001e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.583e-05
        [j_node_and_user_rematch]: 9.02999e-06
        [meta_fg_expand]: 2.23998e-06
        [replace_old_param]: 7.98999e-06
        [inline_without_move]: 5.05001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.62999e-06
        [auto_monad_grad]: 1.40001e-06
        [auto_monad_eliminator]: 7.80998e-06
        [cse]: 1.493e-05
        [replace_applicator]: 7.71999e-06
[py_interpret_to_execute_after_opt_a]: 1.31e-05
[rewriter_after_opt_a]: 3.872e-05
[convert_after_rewriter]: 8.27e-06
[order_py_execute_after_rewriter]: 4.94e-06
[mutable_eliminate]: 0.0006799
[jit_opt_b]: 6.036e-05, [1]
    [Cycle 1]: 5.012e-05, [2]
        [frontend_op_eliminate]: 2.04e-05
        [inline_after_opt_a]: 1.574e-05
[cconv]: 3.199e-05
[loop_unroll]: 0.00053525
[jit_opt_after_cconv]: 0.00016797, [1]
    [Cycle 1]: 0.00015939, [11]
        [c_1]: 2.171e-05
        [parameter_eliminate]: 5.62999e-06
        [updatestate_depend_eliminate]: 9.35001e-06
        [updatestate_assign_eliminate]: 2.51998e-06
        [updatestate_loads_eliminate]: 2.48998e-06
        [cse]: 2.707e-05
        [call_graph_tuple_transform]: 2.237e-05
        [tuple_list_get_item_eliminator]: 5.60001e-06
        [none_parameter_eliminate]: 1.65001e-06
        [renormalize]: 9.00007e-07
        [switch_simplify]: 5.66e-06
[remove_dup_value]: 1.57e-05
[partial_unused_args_eliminate]: 2.31e-06
[environ_conv]: 6.61999e-06
[add_recomputation]: 5.579e-05
[cse_after_recomputation]: 2.619e-05, [1]
    [Cycle 1]: 1.892e-05, [1]
        [cse]: 1.09e-05
[auto_monad_reorder]: 1.603e-05
[get_jit_bprop_graph]: 2.15002e-06
[rewriter_after_jit_bprop_graph]: 9.12001e-06
[opt_after_jit_grad]: 0.00058712
[symbol_engine_optimizer]: 8.063e-05, [1]
    [Cycle 1]: 7.187e-05, [6]
        [build]: 3.97e-06
        [elim_shapecalc]: 8.13001e-06
        [elim_not_effective]: 1.537e-05
        [opt_reshape]: 5.54e-06
        [fold_const_symbol]: 8.38999e-06
        [renormalize]: 9.39996e-07
[validate]: 3.66e-05
[backend_pass]: 9.79984e-07
[task_emit]: 0.00726503
[execute]: 8.02998e-06
Sums
    bootstrap                            :     0.000316s :  1.88%
    type_inference                       :     0.004903s : 29.18%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000065s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000033s :  0.20%
    rewriter_before_opt_a                :     0.000074s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.21%
    jit_opt_a.a_1                        :     0.000693s :  4.13%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000009s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000746s :  4.44%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.15%
    jit_opt_a.cse                        :     0.000044s :  0.26%
    jit_opt_a.replace_applicator         :     0.000024s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000039s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000680s :  4.05%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000032s :  0.19%
    loop_unroll                          :     0.000535s :  3.19%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000056s :  0.33%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000587s :  3.49%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000037s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007265s : 43.24%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000240    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.48% :     0.000006s :      3: substitution.graph_param_transform
84.87% :     0.000204s :      4: substitution.inline
 1.71% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.02% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.20% :     0.000005s :      2: substitution.replace_old_param
 5.33% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004834     2
82.19% :     0.003973s :      1: type_inference.infer
17.81% :     0.000861s :      1: type_inference.specialize
------[replace.]   0.000078     6
64.16% :     0.000050s :      4: replace.inline
35.84% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000212     6
94.56% :     0.000200s :      4: match.inline
 5.44% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.52% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000002s :     10: predicate.addn_zero_filter
 1.92% :     0.000002s :     10: predicate.arithmetic_simplify
 1.47% :     0.000002s :     10: predicate.cast_eliminate
 0.32% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000002s :     10: predicate.depend_value_elim
 0.97% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.09% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.25% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.94% :     0.000008s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.32% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.57% :     0.000002s :     15: predicate.load_eliminater
 2.10% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.79% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.80% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.85% :     0.000004s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.36% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.53% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.76% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.90% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 1.90% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.37% :     0.000011s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000002s :     10: predicate.transpose_eliminate
 1.60% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.51% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.33% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.56% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.14% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000626    12
48.89% :     0.000306s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.03% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.08% :     0.000270s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023003    76
 0.26% :     0.000061s :      1: add_recomputation
 0.30% :     0.000068s :      1: auto_monad
 0.09% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.46% :     0.000335s :      1: bootstrap
 0.16% :     0.000036s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.12% :     0.000028s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
25.64% :     0.005899s :      1: jit_opt_a
 0.74% :     0.000171s :      1: jit_opt_after_cconv
 0.28% :     0.000063s :      1: jit_opt_b
 2.39% :     0.000549s :      1: loop_unroll
 3.03% :     0.000696s :      1: mutable_eliminate
 4.03% :     0.000928s :     26: opt.transform.jit_opt_a
 0.22% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000029s :      4: opt.transform.jit_opt_b
 0.08% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.61% :     0.000600s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000035s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 1.79% :     0.000411s :      1: renormalize.infer
 1.41% :     0.000324s :      1: renormalize.specialize
 0.05% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000043s :      1: rewriter_after_opt_a
 0.33% :     0.000077s :      1: rewriter_before_opt_a
 0.36% :     0.000083s :      1: symbol_engine_optimizer
31.65% :     0.007281s :      1: task_emit
21.38% :     0.004918s :      1: type_inference
 0.25% :     0.000058s :      1: validate


TotalTime = 0.023526, [33]
[bootstrap]: 0.00038472
[type_inference]: 0.00581518
[event_method]: 2.248e-05
[auto_monad]: 7.333e-05
[graph_reusing]: 7.13998e-06
[pre_auto_parallel]: 3.06999e-06
[py_interpret_to_execute]: 4.485e-05
[rewriter_before_opt_a]: 8.385e-05
[expand_dump_flag]: 3.45e-06
[jit_opt_a]: 0.00692815, [2]
    [Cycle 1]: 0.00227918, [27]
        [switch_simplify]: 6.926e-05
        [loop_unroll]: 0.00021571
        [a_1]: 0.00064272
        [with_stream_mark]: 2.753e-05
        [recompute_prepare]: 7.75998e-06
        [updatestate_depend_eliminate]: 4.79002e-06
        [updatestate_assign_eliminate]: 2.88e-06
        [updatestate_loads_eliminate]: 2.61999e-06
        [parameter_eliminate]: 2.79001e-06
        [specialize_transform]: 6.88e-06
        [updatestate_useless_node_eliminater]: 5.32999e-06
        [accelerated_algorithm]: 5.84e-06
        [meta_shard_fg_expand]: 2.11998e-06
        [get_grad_eliminate_]: 5.32999e-06
        [merge_forward]: 4.53001e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.541e-05
        [j_node_and_user_rematch]: 9.42001e-06
        [meta_fg_expand]: 2.43e-06
        [replace_old_param]: 1.165e-05
        [inline_without_move]: 5.00999e-06
        [renormalize]: 0.00090297
        [add_forward_monad_depend]: 9.29998e-06
        [auto_monad_grad]: 2.74999e-06
        [auto_monad_eliminator]: 1.89e-05
        [cse]: 3.72e-05
        [replace_applicator]: 2.016e-05
    [Cycle 2]: 0.00035952, [27]
        [switch_simplify]: 6.36998e-06
        [loop_unroll]: 4.94e-06
        [a_1]: 9.121e-05
        [with_stream_mark]: 1.323e-05
        [recompute_prepare]: 5.47999e-06
        [updatestate_depend_eliminate]: 3.93999e-06
        [updatestate_assign_eliminate]: 2.40002e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [parameter_eliminate]: 1.71e-06
        [specialize_transform]: 4.90999e-06
        [updatestate_useless_node_eliminater]: 4.77998e-06
        [accelerated_algorithm]: 5.44e-06
        [meta_shard_fg_expand]: 1.97999e-06
        [get_grad_eliminate_]: 4.55999e-06
        [merge_forward]: 4.52e-06
        [cell_reuse_recompute_pass]: 3.78001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.885e-05
        [j_node_and_user_rematch]: 1.016e-05
        [meta_fg_expand]: 2.27999e-06
        [replace_old_param]: 1.302e-05
        [inline_without_move]: 4.99e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.24999e-06
        [auto_monad_grad]: 9.79984e-07
        [auto_monad_eliminator]: 7.61999e-06
        [cse]: 1.322e-05
        [replace_applicator]: 5.97999e-06
[py_interpret_to_execute_after_opt_a]: 1.902e-05
[rewriter_after_opt_a]: 4.646e-05
[convert_after_rewriter]: 9.86e-06
[order_py_execute_after_rewriter]: 8.12e-06
[mutable_eliminate]: 0.00088472
[jit_opt_b]: 5.986e-05, [1]
    [Cycle 1]: 4.799e-05, [2]
        [frontend_op_eliminate]: 1.695e-05
        [inline_after_opt_a]: 1.903e-05
[cconv]: 4.104e-05
[loop_unroll]: 0.00063477
[jit_opt_after_cconv]: 0.00018528, [1]
    [Cycle 1]: 0.00017636, [11]
        [c_1]: 2.239e-05
        [parameter_eliminate]: 6.86001e-06
        [updatestate_depend_eliminate]: 8.36002e-06
        [updatestate_assign_eliminate]: 3.63999e-06
        [updatestate_loads_eliminate]: 3.16001e-06
        [cse]: 3.774e-05
        [call_graph_tuple_transform]: 2.623e-05
        [tuple_list_get_item_eliminator]: 5.69e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 9.89996e-07
        [switch_simplify]: 5.54e-06
[remove_dup_value]: 1.647e-05
[partial_unused_args_eliminate]: 2.47001e-06
[environ_conv]: 5.87999e-06
[add_recomputation]: 6.314e-05
[cse_after_recomputation]: 2.342e-05, [1]
    [Cycle 1]: 1.661e-05, [1]
        [cse]: 9.56e-06
[auto_monad_reorder]: 1.62e-05
[get_jit_bprop_graph]: 1.94999e-06
[rewriter_after_jit_bprop_graph]: 1.035e-05
[opt_after_jit_grad]: 0.0006184
[symbol_engine_optimizer]: 8.247e-05, [1]
    [Cycle 1]: 7.469e-05, [6]
        [build]: 6.12999e-06
        [elim_shapecalc]: 8.24002e-06
        [elim_not_effective]: 1.553e-05
        [opt_reshape]: 6.23e-06
        [fold_const_symbol]: 8.3e-06
        [renormalize]: 4.60015e-07
[validate]: 4.621e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00709578
[execute]: 1.054e-05
Sums
    bootstrap                            :     0.000385s :  2.08%
    type_inference                       :     0.005815s : 31.49%
    event_method                         :     0.000022s :  0.12%
    auto_monad                           :     0.000073s :  0.40%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000045s :  0.24%
    rewriter_before_opt_a                :     0.000084s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.41%
    jit_opt_a.loop_unroll                :     0.000221s :  1.19%
    jit_opt_a.a_1                        :     0.000734s :  3.97%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000009s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000025s :  0.13%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000903s :  4.89%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.14%
    jit_opt_a.cse                        :     0.000050s :  0.27%
    jit_opt_a.replace_applicator         :     0.000026s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.10%
    rewriter_after_opt_a                 :     0.000046s :  0.25%
    convert_after_rewriter               :     0.000010s :  0.05%
    order_py_execute_after_rewriter      :     0.000008s :  0.04%
    mutable_eliminate                    :     0.000885s :  4.79%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.10%
    cconv                                :     0.000041s :  0.22%
    loop_unroll                          :     0.000635s :  3.44%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000038s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000063s :  0.34%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.06%
    opt_after_jit_grad                   :     0.000618s :  3.35%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000046s :  0.25%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007096s : 38.42%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000294    23
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 2.12% :     0.000006s :      3: substitution.graph_param_transform
85.73% :     0.000252s :      4: substitution.inline
 1.85% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.86% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.17% :     0.000006s :      2: substitution.replace_old_param
 5.05% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005725     2
83.68% :     0.004790s :      1: type_inference.infer
16.32% :     0.000934s :      1: type_inference.specialize
------[replace.]   0.000079     6
71.87% :     0.000057s :      4: replace.inline
28.13% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000263     6
94.79% :     0.000249s :      4: match.inline
 5.21% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   691
 1.14% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.52% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000002s :     10: predicate.addn_zero_filter
 1.90% :     0.000002s :     10: predicate.arithmetic_simplify
 1.05% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.41% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.47% :     0.000001s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.02% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     10: predicate.environ_get_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.80% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 6.34% :     0.000008s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.02% :     0.000003s :     15: predicate.load_eliminater
 2.25% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.48% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.92% :     0.000001s :     10: predicate.merge_addn
 0.95% :     0.000001s :     10: predicate.minmaximum_grad
 3.35% :     0.000004s :      3: predicate.mutable_eliminate
 0.76% :     0.000001s :      3: predicate.opt_reshape
 1.96% :     0.000003s :     15: predicate.partial_eliminate
 1.02% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.35% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.75% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 1.25% :     0.000002s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.08% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.02% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.12% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.73% :     0.000004s :     16: predicate.switch_defer_inline
 1.92% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.85% :     0.000010s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.18% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.09% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.54% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.32% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.47% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000715    12
49.84% :     0.000357s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.49% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.67% :     0.000298s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025723    76
 0.26% :     0.000067s :      1: add_recomputation
 0.30% :     0.000077s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.59% :     0.000409s :      1: bootstrap
 0.17% :     0.000044s :      1: cconv
 0.05% :     0.000012s :      1: convert_after_rewriter
 0.10% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.11% :     0.000028s :      1: event_method
 0.07% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
26.95% :     0.006932s :      1: jit_opt_a
 0.73% :     0.000189s :      1: jit_opt_after_cconv
 0.25% :     0.000064s :      1: jit_opt_b
 2.53% :     0.000650s :      1: loop_unroll
 3.51% :     0.000902s :      1: mutable_eliminate
 4.53% :     0.001166s :     26: opt.transform.jit_opt_a
 0.22% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000026s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.45% :     0.000629s :      1: opt_after_jit_grad
 0.04% :     0.000011s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000048s :      1: py_interpret_to_execute
 0.09% :     0.000023s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 2.00% :     0.000515s :      1: renormalize.infer
 1.46% :     0.000376s :      1: renormalize.specialize
 0.05% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000051s :      1: rewriter_after_opt_a
 0.34% :     0.000088s :      1: rewriter_before_opt_a
 0.33% :     0.000085s :      1: symbol_engine_optimizer
27.67% :     0.007117s :      1: task_emit
22.73% :     0.005846s :      1: type_inference
 0.27% :     0.000071s :      1: validate


TotalTime = 0.0155191, [30]
[bootstrap]: 0.0003353
[type_inference]: 0.00537544
[event_method]: 2.001e-05
[auto_monad]: 7.06e-05
[graph_reusing]: 5.82999e-06
[pre_auto_parallel]: 3.09999e-06
[py_interpret_to_execute]: 5.75e-05
[rewriter_before_opt_a]: 7.861e-05
[expand_dump_flag]: 3.25998e-06
[jit_opt_a]: 0.00685984, [2]
    [Cycle 1]: 0.00211891, [27]
        [switch_simplify]: 6.486e-05
        [loop_unroll]: 2.962e-05
        [a_1]: 0.00067061
        [with_stream_mark]: 2.436e-05
        [recompute_prepare]: 8.56997e-06
        [updatestate_depend_eliminate]: 4.1e-06
        [updatestate_assign_eliminate]: 3.53999e-06
        [updatestate_loads_eliminate]: 2.89001e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 6.52001e-06
        [updatestate_useless_node_eliminater]: 5.81e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 2.78003e-06
        [get_grad_eliminate_]: 6.12999e-06
        [merge_forward]: 4.67e-06
        [cell_reuse_recompute_pass]: 1.55999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.597e-05
        [j_node_and_user_rematch]: 1.032e-05
        [meta_fg_expand]: 2.81999e-06
        [replace_old_param]: 1.062e-05
        [inline_without_move]: 5.89999e-06
        [renormalize]: 0.00095591
        [add_forward_monad_depend]: 8.05999e-06
        [auto_monad_grad]: 2.99001e-06
        [auto_monad_eliminator]: 1.875e-05
        [cse]: 2.906e-05
        [replace_applicator]: 1.522e-05
    [Cycle 2]: 0.00040371, [27]
        [switch_simplify]: 6.17001e-06
        [loop_unroll]: 5.30999e-06
        [a_1]: 9.722e-05
        [with_stream_mark]: 1.401e-05
        [recompute_prepare]: 5.96e-06
        [updatestate_depend_eliminate]: 3.26001e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 1.14998e-06
        [specialize_transform]: 6.41e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 5.99e-06
        [meta_shard_fg_expand]: 1.87001e-06
        [get_grad_eliminate_]: 5.05001e-06
        [merge_forward]: 3.28e-06
        [cell_reuse_recompute_pass]: 1.86e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.813e-05
        [j_node_and_user_rematch]: 1.067e-05
        [meta_fg_expand]: 3.22997e-06
        [replace_old_param]: 9.04998e-06
        [inline_without_move]: 4.97e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.95001e-06
        [auto_monad_grad]: 1.14e-06
        [auto_monad_eliminator]: 8.03999e-06
        [cse]: 1.426e-05
        [replace_applicator]: 8.08999e-06
[py_interpret_to_execute_after_opt_a]: 1.479e-05
[rewriter_after_opt_a]: 4.113e-05
[convert_after_rewriter]: 9.03002e-06
[order_py_execute_after_rewriter]: 5.47999e-06
[mutable_eliminate]: 0.00076618
[jit_opt_b]: 6.044e-05, [1]
    [Cycle 1]: 5.093e-05, [2]
        [frontend_op_eliminate]: 1.927e-05
        [inline_after_opt_a]: 1.767e-05
[cconv]: 3.161e-05
[loop_unroll]: 0.00053487
[jit_opt_after_cconv]: 0.00015535, [1]
    [Cycle 1]: 0.00014733, [11]
        [c_1]: 2.205e-05
        [parameter_eliminate]: 3.81999e-06
        [updatestate_depend_eliminate]: 7.98001e-06
        [updatestate_assign_eliminate]: 2.38998e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [cse]: 2.259e-05
        [call_graph_tuple_transform]: 2.199e-05
        [tuple_list_get_item_eliminator]: 5.29e-06
        [none_parameter_eliminate]: 1.41002e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 5.71003e-06
[remove_dup_value]: 1.594e-05
[partial_unused_args_eliminate]: 2.84999e-06
[environ_conv]: 5.59998e-06
[add_recomputation]: 5.127e-05
[cse_after_recomputation]: 2.457e-05, [1]
    [Cycle 1]: 1.744e-05, [1]
        [cse]: 9.84999e-06
[auto_monad_reorder]: 1.679e-05
[get_jit_bprop_graph]: 2.01e-06
[rewriter_after_jit_bprop_graph]: 7.34002e-06
[opt_after_jit_grad]: 0.00062674
[symbol_engine_optimizer]: 8.239e-05, [1]
    [Cycle 1]: 7.379e-05, [6]
        [build]: 4.66002e-06
        [elim_shapecalc]: 8.72e-06
        [elim_not_effective]: 1.579e-05
        [opt_reshape]: 6.23e-06
        [fold_const_symbol]: 8.89e-06
        [renormalize]: 5.00004e-07
[validate]: 4.112e-05
Sums
    bootstrap                            :     0.000335s :  3.20%
    type_inference                       :     0.005375s : 51.31%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000071s :  0.67%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000058s :  0.55%
    rewriter_before_opt_a                :     0.000079s :  0.75%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000071s :  0.68%
    jit_opt_a.loop_unroll                :     0.000035s :  0.33%
    jit_opt_a.a_1                        :     0.000768s :  7.33%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.37%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.06%
    jit_opt_a.replace_old_param          :     0.000020s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000956s :  9.13%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.26%
    jit_opt_a.cse                        :     0.000043s :  0.41%
    jit_opt_a.replace_applicator         :     0.000023s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.14%
    rewriter_after_opt_a                 :     0.000041s :  0.39%
    convert_after_rewriter               :     0.000009s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000766s :  7.31%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.17%
    cconv                                :     0.000032s :  0.30%
    loop_unroll                          :     0.000535s :  5.11%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000016s :  0.15%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000051s :  0.49%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000017s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000627s :  5.98%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.39%


Time group info:
------[substitution.]   0.000275    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 2.16% :     0.000006s :      3: substitution.graph_param_transform
85.84% :     0.000236s :      4: substitution.inline
 1.73% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.67% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.27% :     0.000006s :      2: substitution.replace_old_param
 5.08% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005292     2
78.96% :     0.004179s :      1: type_inference.infer
21.04% :     0.001114s :      1: type_inference.specialize
------[replace.]   0.000106     6
67.79% :     0.000072s :      4: replace.inline
32.21% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000246     6
94.81% :     0.000233s :      4: match.inline
 5.19% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000140   691
 1.25% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.72% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.91% :     0.000001s :     10: predicate.addn_check_dump
 1.70% :     0.000002s :     10: predicate.addn_zero_filter
 1.97% :     0.000003s :     10: predicate.arithmetic_simplify
 1.21% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.89% :     0.000001s :     10: predicate.compare_switch_simplify
 1.50% :     0.000002s :     10: predicate.depend_value_elim
 1.23% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.02% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.19% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000001s :      3: predicate.elim_not_effective
 0.77% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.65% :     0.000002s :     10: predicate.environ_get_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 1.14% :     0.000002s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 6.91% :     0.000010s :     22: predicate.inline
 1.08% :     0.000002s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :      6: predicate.less_batch_normalization
 1.34% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.41% :     0.000002s :     15: predicate.load_eliminater
 2.16% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.72% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.19% :     0.000002s :     10: predicate.minmaximum_grad
 2.33% :     0.000003s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 1.93% :     0.000003s :     15: predicate.partial_eliminate
 1.04% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.19% :     0.000003s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.39% :     0.000002s :     18: predicate.replace_applicator
 0.42% :     0.000001s :      6: predicate.replace_old_param
 0.43% :     0.000001s :      3: predicate.reset_defer_inline
 1.12% :     0.000002s :     10: predicate.reshape_eliminate
 0.96% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.17% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000002s :     10: predicate.same_eliminate
 0.39% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.21% :     0.000002s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.11% :     0.000004s :     16: predicate.switch_defer_inline
 1.78% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.51% :     0.000010s :     53: predicate.switch_simplify
 1.40% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000002s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.07% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.10% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.40% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.34% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.43% :     0.000002s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000803    12
46.93% :     0.000377s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.89% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.17% :     0.000371s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017539    72
 0.31% :     0.000055s :      1: add_recomputation
 0.43% :     0.000075s :      1: auto_monad
 0.12% :     0.000020s :      1: auto_monad_reorder
 1.97% :     0.000346s :      1: bootstrap
 0.20% :     0.000035s :      1: cconv
 0.07% :     0.000012s :      1: convert_after_rewriter
 0.15% :     0.000027s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.14% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
39.13% :     0.006864s :      1: jit_opt_a
 0.90% :     0.000158s :      1: jit_opt_after_cconv
 0.36% :     0.000063s :      1: jit_opt_b
 3.12% :     0.000547s :      1: loop_unroll
 4.45% :     0.000780s :      1: mutable_eliminate
 5.76% :     0.001010s :     26: opt.transform.jit_opt_a
 0.29% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000029s :      4: opt.transform.jit_opt_b
 0.10% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.65% :     0.000641s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.35% :     0.000061s :      1: py_interpret_to_execute
 0.11% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000018s :      1: remove_dup_value
 2.90% :     0.000509s :      1: renormalize.infer
 2.48% :     0.000435s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000046s :      1: rewriter_after_opt_a
 0.47% :     0.000083s :      1: rewriter_before_opt_a
 0.49% :     0.000085s :      1: symbol_engine_optimizer
30.74% :     0.005392s :      1: type_inference


TotalTime = 0.021761, [33]
[bootstrap]: 0.00037086
[type_inference]: 0.00552989
[event_method]: 1.978e-05
[auto_monad]: 6.466e-05
[graph_reusing]: 5.98998e-06
[pre_auto_parallel]: 2.53e-06
[py_interpret_to_execute]: 4.275e-05
[rewriter_before_opt_a]: 7.565e-05
[expand_dump_flag]: 3.19001e-06
[jit_opt_a]: 0.00630151, [2]
    [Cycle 1]: 0.00229352, [27]
        [switch_simplify]: 6.47e-05
        [loop_unroll]: 2.899e-05
        [a_1]: 0.00060003
        [with_stream_mark]: 2.304e-05
        [recompute_prepare]: 7.77002e-06
        [updatestate_depend_eliminate]: 3.42002e-06
        [updatestate_assign_eliminate]: 3.01999e-06
        [updatestate_loads_eliminate]: 2.62001e-06
        [parameter_eliminate]: 2.16e-06
        [specialize_transform]: 6.32001e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 2.04e-06
        [get_grad_eliminate_]: 5.07999e-06
        [merge_forward]: 3.38e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.5e-05
        [j_node_and_user_rematch]: 9.56e-06
        [meta_fg_expand]: 2.27999e-06
        [replace_old_param]: 9.68002e-06
        [inline_without_move]: 5.02e-06
        [renormalize]: 0.00122597
        [add_forward_monad_depend]: 7.70998e-06
        [auto_monad_grad]: 2.64001e-06
        [auto_monad_eliminator]: 1.596e-05
        [cse]: 3.077e-05
        [replace_applicator]: 1.508e-05
    [Cycle 2]: 0.00033537, [27]
        [switch_simplify]: 6.13002e-06
        [loop_unroll]: 5.02999e-06
        [a_1]: 9.366e-05
        [with_stream_mark]: 1.138e-05
        [recompute_prepare]: 4.92999e-06
        [updatestate_depend_eliminate]: 2.76e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 1.86998e-06
        [parameter_eliminate]: 1.55999e-06
        [specialize_transform]: 4.97999e-06
        [updatestate_useless_node_eliminater]: 4.68999e-06
        [accelerated_algorithm]: 5.14e-06
        [meta_shard_fg_expand]: 1.57001e-06
        [get_grad_eliminate_]: 4.49998e-06
        [merge_forward]: 3.04001e-06
        [cell_reuse_recompute_pass]: 2.32001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.398e-05
        [j_node_and_user_rematch]: 7.70998e-06
        [meta_fg_expand]: 2.21e-06
        [replace_old_param]: 7.58999e-06
        [inline_without_move]: 4.40999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 9.5999e-07
        [auto_monad_eliminator]: 5.73997e-06
        [cse]: 1.095e-05
        [replace_applicator]: 5.00999e-06
[py_interpret_to_execute_after_opt_a]: 1.33e-05
[rewriter_after_opt_a]: 3.899e-05
[convert_after_rewriter]: 7.05e-06
[order_py_execute_after_rewriter]: 5.03002e-06
[mutable_eliminate]: 0.00073495
[jit_opt_b]: 5.302e-05, [1]
    [Cycle 1]: 4.497e-05, [2]
        [frontend_op_eliminate]: 1.723e-05
        [inline_after_opt_a]: 1.517e-05
[cconv]: 3.051e-05
[loop_unroll]: 0.00045444
[jit_opt_after_cconv]: 0.00015619, [1]
    [Cycle 1]: 0.00014956, [11]
        [c_1]: 2.209e-05
        [parameter_eliminate]: 3.45998e-06
        [updatestate_depend_eliminate]: 7.28e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 2.364e-05
        [call_graph_tuple_transform]: 2.179e-05
        [tuple_list_get_item_eliminator]: 5.59e-06
        [none_parameter_eliminate]: 1.73002e-06
        [renormalize]: 9.80013e-07
        [switch_simplify]: 5.77999e-06
[remove_dup_value]: 1.453e-05
[partial_unused_args_eliminate]: 2.11998e-06
[environ_conv]: 5.47999e-06
[add_recomputation]: 5.424e-05
[cse_after_recomputation]: 2.188e-05, [1]
    [Cycle 1]: 1.62e-05, [1]
        [cse]: 9.84999e-06
[auto_monad_reorder]: 1.622e-05
[get_jit_bprop_graph]: 2.21e-06
[rewriter_after_jit_bprop_graph]: 8.85001e-06
[opt_after_jit_grad]: 0.00050094
[symbol_engine_optimizer]: 8.116e-05, [1]
    [Cycle 1]: 7.22e-05, [6]
        [build]: 5.67999e-06
        [elim_shapecalc]: 8.08999e-06
        [elim_not_effective]: 1.359e-05
        [opt_reshape]: 6.19999e-06
        [fold_const_symbol]: 9.46e-06
        [renormalize]: 4.80009e-07
[validate]: 4.938e-05
[backend_pass]: 8.59989e-07
[task_emit]: 0.00682825
[execute]: 1.036e-05
Sums
    bootstrap                            :     0.000371s :  2.13%
    type_inference                       :     0.005530s : 31.79%
    event_method                         :     0.000020s :  0.11%
    auto_monad                           :     0.000065s :  0.37%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.25%
    rewriter_before_opt_a                :     0.000076s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000071s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000694s :  3.99%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000009s :  0.05%
    jit_opt_a.renormalize                :     0.001226s :  7.05%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.12%
    jit_opt_a.cse                        :     0.000042s :  0.24%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000039s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000735s :  4.22%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000031s :  0.18%
    loop_unroll                          :     0.000454s :  2.61%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000054s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000501s :  2.88%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000049s :  0.28%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006828s : 39.25%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000257    23
 0.80% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 2.21% :     0.000006s :      3: substitution.graph_param_transform
86.34% :     0.000222s :      4: substitution.inline
 1.29% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.06% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.93% :     0.000005s :      2: substitution.replace_old_param
 4.84% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005447     2
82.31% :     0.004483s :      1: type_inference.infer
17.69% :     0.000964s :      1: type_inference.specialize
------[replace.]   0.000078     6
66.21% :     0.000052s :      4: replace.inline
33.79% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000230     6
95.07% :     0.000219s :      4: match.inline
 4.93% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.24% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.42% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.18% :     0.000001s :     10: predicate.addn_zero_filter
 2.73% :     0.000003s :     10: predicate.arithmetic_simplify
 1.40% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.29% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.78% :     0.000007s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.98% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 1.92% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.32% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.19% :     0.000001s :     10: predicate.minmaximum_grad
 2.23% :     0.000003s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 1.99% :     0.000002s :     15: predicate.partial_eliminate
 1.26% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.83% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.90% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.43% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.10% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.12% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.43% :     0.000009s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.96% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.61% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000761    12
50.91% :     0.000387s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.80% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.29% :     0.000322s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024005    76
 0.24% :     0.000057s :      1: add_recomputation
 0.28% :     0.000068s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.64% :     0.000394s :      1: bootstrap
 0.14% :     0.000033s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000025s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.27% :     0.006306s :      1: jit_opt_a
 0.66% :     0.000159s :      1: jit_opt_after_cconv
 0.23% :     0.000056s :      1: jit_opt_b
 1.93% :     0.000464s :      1: loop_unroll
 3.11% :     0.000746s :      1: mutable_eliminate
 3.82% :     0.000917s :     26: opt.transform.jit_opt_a
 0.22% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.13% :     0.000511s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000045s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 3.24% :     0.000778s :      1: renormalize.infer
 1.83% :     0.000438s :      1: renormalize.specialize
 0.05% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000042s :      1: rewriter_after_opt_a
 0.33% :     0.000080s :      1: rewriter_before_opt_a
 0.35% :     0.000083s :      1: symbol_engine_optimizer
28.53% :     0.006848s :      1: task_emit
23.13% :     0.005552s :      1: type_inference
 0.31% :     0.000075s :      1: validate


TotalTime = 0.0126088, [30]
[bootstrap]: 0.00025697
[type_inference]: 0.00410219
[event_method]: 1.689e-05
[auto_monad]: 6.323e-05
[graph_reusing]: 5.74e-06
[pre_auto_parallel]: 2.84001e-06
[py_interpret_to_execute]: 3.277e-05
[rewriter_before_opt_a]: 6.996e-05
[expand_dump_flag]: 2.78998e-06
[jit_opt_a]: 0.00579702, [2]
    [Cycle 1]: 0.00198877, [27]
        [switch_simplify]: 5.997e-05
        [loop_unroll]: 2.941e-05
        [a_1]: 0.00059479
        [with_stream_mark]: 2.281e-05
        [recompute_prepare]: 6.62002e-06
        [updatestate_depend_eliminate]: 3.6e-06
        [updatestate_assign_eliminate]: 2.92002e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 2.00002e-06
        [specialize_transform]: 5.70001e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 6.07001e-06
        [meta_shard_fg_expand]: 2.08002e-06
        [get_grad_eliminate_]: 5.47999e-06
        [merge_forward]: 3.45e-06
        [cell_reuse_recompute_pass]: 9.80013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.425e-05
        [j_node_and_user_rematch]: 9.52001e-06
        [meta_fg_expand]: 2.29001e-06
        [replace_old_param]: 9.61e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 0.00094332
        [add_forward_monad_depend]: 5.67999e-06
        [auto_monad_grad]: 2.40002e-06
        [auto_monad_eliminator]: 1.405e-05
        [cse]: 3.147e-05
        [replace_applicator]: 1.246e-05
    [Cycle 2]: 0.00032194, [27]
        [switch_simplify]: 5.47999e-06
        [loop_unroll]: 4.70999e-06
        [a_1]: 9.044e-05
        [with_stream_mark]: 1.024e-05
        [recompute_prepare]: 5.02999e-06
        [updatestate_depend_eliminate]: 2.84999e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 4.88001e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.52999e-06
        [meta_shard_fg_expand]: 1.16002e-06
        [get_grad_eliminate_]: 4.3e-06
        [merge_forward]: 3.40003e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.42e-05
        [j_node_and_user_rematch]: 8.15e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 7.13e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 6.50005e-07
        [auto_monad_eliminator]: 5.11002e-06
        [cse]: 9.87001e-06
        [replace_applicator]: 4.95999e-06
[py_interpret_to_execute_after_opt_a]: 1.149e-05
[rewriter_after_opt_a]: 3.589e-05
[convert_after_rewriter]: 7.01999e-06
[order_py_execute_after_rewriter]: 5.27001e-06
[mutable_eliminate]: 0.0006657
[jit_opt_b]: 5.301e-05, [1]
    [Cycle 1]: 4.463e-05, [2]
        [frontend_op_eliminate]: 1.743e-05
        [inline_after_opt_a]: 1.573e-05
[cconv]: 2.999e-05
[loop_unroll]: 0.00043625
[jit_opt_after_cconv]: 0.00014506, [1]
    [Cycle 1]: 0.00013817, [11]
        [c_1]: 2.084e-05
        [parameter_eliminate]: 2.69999e-06
        [updatestate_depend_eliminate]: 5.75001e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [cse]: 2.054e-05
        [call_graph_tuple_transform]: 1.942e-05
        [tuple_list_get_item_eliminator]: 5.37999e-06
        [none_parameter_eliminate]: 1.63002e-06
        [renormalize]: 7.30011e-07
        [switch_simplify]: 5.62999e-06
[remove_dup_value]: 1.399e-05
[partial_unused_args_eliminate]: 2.73998e-06
[environ_conv]: 5.17999e-06
[add_recomputation]: 4.884e-05
[cse_after_recomputation]: 2.105e-05, [1]
    [Cycle 1]: 1.55e-05, [1]
        [cse]: 9.36e-06
[auto_monad_reorder]: 1.499e-05
[get_jit_bprop_graph]: 2.32999e-06
[rewriter_after_jit_bprop_graph]: 6.34001e-06
[opt_after_jit_grad]: 0.00047197
[symbol_engine_optimizer]: 7.466e-05, [1]
    [Cycle 1]: 6.802e-05, [6]
        [build]: 4.68999e-06
        [elim_shapecalc]: 7.93001e-06
        [elim_not_effective]: 1.263e-05
        [opt_reshape]: 5.67999e-06
        [fold_const_symbol]: 8.45001e-06
        [renormalize]: 6.10016e-07
[validate]: 3.956e-05
Sums
    bootstrap                            :     0.000257s :  3.01%
    type_inference                       :     0.004102s : 48.08%
    event_method                         :     0.000017s :  0.20%
    auto_monad                           :     0.000063s :  0.74%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.38%
    rewriter_before_opt_a                :     0.000070s :  0.82%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000065s :  0.77%
    jit_opt_a.loop_unroll                :     0.000034s :  0.40%
    jit_opt_a.a_1                        :     0.000685s :  8.03%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.39%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000943s : 11.06%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.22%
    jit_opt_a.cse                        :     0.000041s :  0.48%
    jit_opt_a.replace_applicator         :     0.000017s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.13%
    rewriter_after_opt_a                 :     0.000036s :  0.42%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000666s :  7.80%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.18%
    cconv                                :     0.000030s :  0.35%
    loop_unroll                          :     0.000436s :  5.11%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000021s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000014s :  0.16%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000049s :  0.57%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000472s :  5.53%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.46%


Time group info:
------[substitution.]   0.000253    23
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.16% :     0.000005s :      3: substitution.graph_param_transform
86.47% :     0.000219s :      4: substitution.inline
 1.43% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.89% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.87% :     0.000005s :      2: substitution.replace_old_param
 4.87% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004040     2
80.01% :     0.003232s :      1: type_inference.infer
19.99% :     0.000808s :      1: type_inference.specialize
------[replace.]   0.000074     6
68.40% :     0.000051s :      4: replace.inline
31.60% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000228     6
95.08% :     0.000217s :      4: match.inline
 4.92% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.28% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 2.38% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.16% :     0.000001s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 6.13% :     0.000007s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.33% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.75% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 2.37% :     0.000003s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     15: predicate.partial_eliminate
 1.39% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.41% :     0.000002s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.52% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.58% :     0.000003s :     16: predicate.switch_defer_inline
 2.20% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.45% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.07% :     0.000001s :     10: predicate.transpose_eliminate
 1.58% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.02% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000597    12
46.19% :     0.000276s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 9.08% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.73% :     0.000267s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014507    72
 0.36% :     0.000052s :      1: add_recomputation
 0.46% :     0.000067s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 1.87% :     0.000271s :      1: bootstrap
 0.23% :     0.000033s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000021s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.99% :     0.005802s :      1: jit_opt_a
 1.02% :     0.000148s :      1: jit_opt_after_cconv
 0.38% :     0.000056s :      1: jit_opt_b
 3.06% :     0.000445s :      1: loop_unroll
 4.65% :     0.000675s :      1: mutable_eliminate
 6.21% :     0.000901s :     26: opt.transform.jit_opt_a
 0.33% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.31% :     0.000480s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000035s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000016s :      1: remove_dup_value
 3.66% :     0.000531s :      1: renormalize.infer
 2.78% :     0.000404s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000039s :      1: rewriter_after_opt_a
 0.50% :     0.000073s :      1: rewriter_before_opt_a
 0.53% :     0.000078s :      1: symbol_engine_optimizer
28.36% :     0.004115s :      1: type_inference


TotalTime = 0.0202273, [33]
[bootstrap]: 0.0003042
[type_inference]: 0.00525593
[event_method]: 2.008e-05
[auto_monad]: 7.078e-05
[graph_reusing]: 6.39001e-06
[pre_auto_parallel]: 2.28998e-06
[py_interpret_to_execute]: 3.554e-05
[rewriter_before_opt_a]: 7.448e-05
[expand_dump_flag]: 3.00002e-06
[jit_opt_a]: 0.0055121, [2]
    [Cycle 1]: 0.00161847, [27]
        [switch_simplify]: 5.891e-05
        [loop_unroll]: 2.989e-05
        [a_1]: 0.00055186
        [with_stream_mark]: 2.031e-05
        [recompute_prepare]: 7.30998e-06
        [updatestate_depend_eliminate]: 4.47998e-06
        [updatestate_assign_eliminate]: 2.87002e-06
        [updatestate_loads_eliminate]: 2.58998e-06
        [parameter_eliminate]: 2.16e-06
        [specialize_transform]: 6.43e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 5.34e-06
        [meta_shard_fg_expand]: 2.16e-06
        [get_grad_eliminate_]: 5.51e-06
        [merge_forward]: 3.45998e-06
        [cell_reuse_recompute_pass]: 1.05999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.352e-05
        [j_node_and_user_rematch]: 9.17999e-06
        [meta_fg_expand]: 2.04999e-06
        [replace_old_param]: 8.32998e-06
        [inline_without_move]: 5.50001e-06
        [renormalize]: 0.00063035
        [add_forward_monad_depend]: 5.32001e-06
        [auto_monad_grad]: 2.53e-06
        [auto_monad_eliminator]: 1.522e-05
        [cse]: 2.75e-05
        [replace_applicator]: 1.394e-05
    [Cycle 2]: 0.00031957, [27]
        [switch_simplify]: 5.74999e-06
        [loop_unroll]: 4.74e-06
        [a_1]: 8.987e-05
        [with_stream_mark]: 1.055e-05
        [recompute_prepare]: 4.94e-06
        [updatestate_depend_eliminate]: 3.09001e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 4.95001e-06
        [updatestate_useless_node_eliminater]: 5.45001e-06
        [accelerated_algorithm]: 4.75001e-06
        [meta_shard_fg_expand]: 1.24e-06
        [get_grad_eliminate_]: 4.53999e-06
        [merge_forward]: 2.76e-06
        [cell_reuse_recompute_pass]: 2.14999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.292e-05
        [j_node_and_user_rematch]: 8.13001e-06
        [meta_fg_expand]: 1.56002e-06
        [replace_old_param]: 6.76e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.72999e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 5.30001e-06
        [cse]: 1.016e-05
        [replace_applicator]: 5.09e-06
[py_interpret_to_execute_after_opt_a]: 1.111e-05
[rewriter_after_opt_a]: 3.67e-05
[convert_after_rewriter]: 7.38e-06
[order_py_execute_after_rewriter]: 4.99e-06
[mutable_eliminate]: 0.0005488
[jit_opt_b]: 5.253e-05, [1]
    [Cycle 1]: 4.45e-05, [2]
        [frontend_op_eliminate]: 1.672e-05
        [inline_after_opt_a]: 1.545e-05
[cconv]: 2.721e-05
[loop_unroll]: 0.00043617
[jit_opt_after_cconv]: 0.00014959, [1]
    [Cycle 1]: 0.00014262, [11]
        [c_1]: 2.038e-05
        [parameter_eliminate]: 3.75e-06
        [updatestate_depend_eliminate]: 7.33999e-06
        [updatestate_assign_eliminate]: 2.38998e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [cse]: 2.538e-05
        [call_graph_tuple_transform]: 1.86e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.40001e-06
        [renormalize]: 5.40022e-07
        [switch_simplify]: 5.84999e-06
[remove_dup_value]: 1.357e-05
[partial_unused_args_eliminate]: 2.35002e-06
[environ_conv]: 5.62999e-06
[add_recomputation]: 5.305e-05
[cse_after_recomputation]: 2.224e-05, [1]
    [Cycle 1]: 1.671e-05, [1]
        [cse]: 1.019e-05
[auto_monad_reorder]: 1.538e-05
[get_jit_bprop_graph]: 1.89999e-06
[rewriter_after_jit_bprop_graph]: 6.06e-06
[opt_after_jit_grad]: 0.0005421
[symbol_engine_optimizer]: 7.374e-05, [1]
    [Cycle 1]: 6.69e-05, [6]
        [build]: 4.07998e-06
        [elim_shapecalc]: 7.65998e-06
        [elim_not_effective]: 1.303e-05
        [opt_reshape]: 5.64998e-06
        [fold_const_symbol]: 8.64998e-06
        [renormalize]: 8.60018e-07
[validate]: 4.05e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00667265
[execute]: 8.95999e-06
Sums
    bootstrap                            :     0.000304s :  1.90%
    type_inference                       :     0.005256s : 32.78%
    event_method                         :     0.000020s :  0.13%
    auto_monad                           :     0.000071s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000036s :  0.22%
    rewriter_before_opt_a                :     0.000074s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000642s :  4.00%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000630s :  3.93%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.13%
    jit_opt_a.cse                        :     0.000038s :  0.23%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000549s :  3.42%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000027s :  0.17%
    loop_unroll                          :     0.000436s :  2.72%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000053s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000542s :  3.38%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006673s : 41.62%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000223    23
 0.94% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.52% :     0.000006s :      3: substitution.graph_param_transform
85.95% :     0.000191s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.78% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.47% :     0.000003s :      2: substitution.replace_old_param
 5.17% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005183     2
82.36% :     0.004269s :      1: type_inference.infer
17.64% :     0.000914s :      1: type_inference.specialize
------[replace.]   0.000066     6
66.86% :     0.000044s :      4: replace.inline
33.14% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000199     6
94.79% :     0.000189s :      4: match.inline
 5.21% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.73% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.47% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.23% :     0.000001s :     10: predicate.addn_zero_filter
 1.75% :     0.000002s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.81% :     0.000007s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.18% :     0.000001s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.81% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 2.20% :     0.000003s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.66% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.45% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.44% :     0.000002s :      6: predicate.special_op_eliminate
 1.18% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.30% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.46% :     0.000010s :     53: predicate.switch_simplify
 1.06% :     0.000001s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.78% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.02% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000632    12
51.46% :     0.000325s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.90% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.64% :     0.000257s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021820    76
 0.26% :     0.000056s :      1: add_recomputation
 0.34% :     0.000075s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.48% :     0.000322s :      1: bootstrap
 0.14% :     0.000030s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000024s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.28% :     0.005515s :      1: jit_opt_a
 0.70% :     0.000152s :      1: jit_opt_after_cconv
 0.26% :     0.000056s :      1: jit_opt_b
 2.04% :     0.000445s :      1: loop_unroll
 2.56% :     0.000559s :      1: mutable_eliminate
 3.91% :     0.000854s :     26: opt.transform.jit_opt_a
 0.21% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.53% :     0.000551s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000038s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.63% :     0.000355s :      1: renormalize.infer
 1.22% :     0.000267s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000040s :      1: rewriter_after_opt_a
 0.35% :     0.000077s :      1: rewriter_before_opt_a
 0.35% :     0.000077s :      1: symbol_engine_optimizer
30.64% :     0.006687s :      1: task_emit
24.16% :     0.005273s :      1: type_inference
 0.28% :     0.000061s :      1: validate


TotalTime = 0.0136028, [30]
[bootstrap]: 0.00031492
[type_inference]: 0.00519084
[event_method]: 1.68e-05
[auto_monad]: 6.262e-05
[graph_reusing]: 6.34001e-06
[pre_auto_parallel]: 2.91e-06
[py_interpret_to_execute]: 3.612e-05
[rewriter_before_opt_a]: 7.234e-05
[expand_dump_flag]: 3.5e-06
[jit_opt_a]: 0.00588296, [2]
    [Cycle 1]: 0.00176478, [27]
        [switch_simplify]: 5.833e-05
        [loop_unroll]: 2.968e-05
        [a_1]: 0.00057758
        [with_stream_mark]: 2.178e-05
        [recompute_prepare]: 7.08e-06
        [updatestate_depend_eliminate]: 3.85998e-06
        [updatestate_assign_eliminate]: 2.91999e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 5.79e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 5.62001e-06
        [meta_shard_fg_expand]: 2.02001e-06
        [get_grad_eliminate_]: 5.24e-06
        [merge_forward]: 3.61999e-06
        [cell_reuse_recompute_pass]: 1.01997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.383e-05
        [j_node_and_user_rematch]: 8.76997e-06
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 8.47e-06
        [inline_without_move]: 5.69e-06
        [renormalize]: 0.00076512
        [add_forward_monad_depend]: 4.65001e-06
        [auto_monad_grad]: 2.12999e-06
        [auto_monad_eliminator]: 1.296e-05
        [cse]: 2.624e-05
        [replace_applicator]: 1.131e-05
    [Cycle 2]: 0.00033788, [27]
        [switch_simplify]: 5.27001e-06
        [loop_unroll]: 4.75999e-06
        [a_1]: 8.783e-05
        [with_stream_mark]: 8.44998e-06
        [recompute_prepare]: 5.00999e-06
        [updatestate_depend_eliminate]: 2.55002e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.95001e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 4.76997e-06
        [updatestate_useless_node_eliminater]: 4.69002e-06
        [accelerated_algorithm]: 4.99e-06
        [meta_shard_fg_expand]: 1.00001e-06
        [get_grad_eliminate_]: 4.36002e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 1.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.219e-05
        [j_node_and_user_rematch]: 7.25e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 6.88e-06
        [inline_without_move]: 5.31998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 5.94e-06
        [cse]: 1.166e-05
        [replace_applicator]: 4.81002e-06
[py_interpret_to_execute_after_opt_a]: 1.141e-05
[rewriter_after_opt_a]: 3.289e-05
[convert_after_rewriter]: 7.41001e-06
[order_py_execute_after_rewriter]: 5.30999e-06
[mutable_eliminate]: 0.00051403
[jit_opt_b]: 4.925e-05, [1]
    [Cycle 1]: 4.245e-05, [2]
        [frontend_op_eliminate]: 1.658e-05
        [inline_after_opt_a]: 1.438e-05
[cconv]: 2.542e-05
[loop_unroll]: 0.00040542
[jit_opt_after_cconv]: 0.00013272, [1]
    [Cycle 1]: 0.00012623, [11]
        [c_1]: 1.992e-05
        [parameter_eliminate]: 2.40002e-06
        [updatestate_depend_eliminate]: 5.08002e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.587e-05
        [call_graph_tuple_transform]: 1.731e-05
        [tuple_list_get_item_eliminator]: 5.14003e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 7.00005e-07
        [switch_simplify]: 5.51e-06
[remove_dup_value]: 1.354e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 5.10999e-06
[add_recomputation]: 4.655e-05
[cse_after_recomputation]: 2.017e-05, [1]
    [Cycle 1]: 1.447e-05, [1]
        [cse]: 8.65001e-06
[auto_monad_reorder]: 1.438e-05
[get_jit_bprop_graph]: 2.32999e-06
[rewriter_after_jit_bprop_graph]: 5.05001e-06
[opt_after_jit_grad]: 0.00043548
[symbol_engine_optimizer]: 8.429e-05, [1]
    [Cycle 1]: 7.792e-05, [6]
        [build]: 3.2e-06
        [elim_shapecalc]: 7.45e-06
        [elim_not_effective]: 1.329e-05
        [opt_reshape]: 5.79999e-06
        [fold_const_symbol]: 8.52998e-06
        [renormalize]: 4.60015e-07
[validate]: 3.459e-05
Sums
    bootstrap                            :     0.000315s :  3.42%
    type_inference                       :     0.005191s : 56.30%
    event_method                         :     0.000017s :  0.18%
    auto_monad                           :     0.000063s :  0.68%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.39%
    rewriter_before_opt_a                :     0.000072s :  0.78%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000064s :  0.69%
    jit_opt_a.loop_unroll                :     0.000034s :  0.37%
    jit_opt_a.a_1                        :     0.000665s :  7.22%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.17%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000765s :  8.30%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.20%
    jit_opt_a.cse                        :     0.000038s :  0.41%
    jit_opt_a.replace_applicator         :     0.000016s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000033s :  0.36%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000514s :  5.58%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.16%
    cconv                                :     0.000025s :  0.28%
    loop_unroll                          :     0.000405s :  4.40%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000014s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000047s :  0.50%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000014s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000435s :  4.72%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.38%


Time group info:
------[substitution.]   0.000225    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.06% :     0.000005s :      3: substitution.graph_param_transform
85.98% :     0.000193s :      4: substitution.inline
 1.27% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.07% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.62% :     0.000004s :      2: substitution.replace_old_param
 5.50% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005122     2
81.50% :     0.004174s :      1: type_inference.infer
18.50% :     0.000947s :      1: type_inference.specialize
------[replace.]   0.000072     6
68.66% :     0.000049s :      4: replace.inline
31.34% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000202     6
94.45% :     0.000191s :      4: match.inline
 5.55% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 7.82% :     0.000010s :     10: predicate.accumulaten_eliminater
 1.15% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.16% :     0.000001s :     10: predicate.addn_zero_filter
 1.94% :     0.000002s :     10: predicate.arithmetic_simplify
 1.11% :     0.000001s :     10: predicate.cast_eliminate
 0.35% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.08% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.01% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.32% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.77% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.23% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 1.47% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 1.98% :     0.000002s :     15: predicate.partial_eliminate
 1.27% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.67% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.52% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.69% :     0.000002s :     10: predicate.reshape_eliminate
 1.08% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.16% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.84% :     0.000009s :     53: predicate.switch_simplify
 1.36% :     0.000002s :     10: predicate.tile_eliminate
 1.06% :     0.000001s :     10: predicate.transpose_eliminate
 1.21% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.07% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.62% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.23% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.07% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000705    12
51.07% :     0.000360s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.63% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.30% :     0.000298s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015295    72
 0.32% :     0.000049s :      1: add_recomputation
 0.44% :     0.000067s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 2.15% :     0.000328s :      1: bootstrap
 0.18% :     0.000028s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.48% :     0.005886s :      1: jit_opt_a
 0.89% :     0.000135s :      1: jit_opt_after_cconv
 0.34% :     0.000052s :      1: jit_opt_b
 2.70% :     0.000413s :      1: loop_unroll
 3.41% :     0.000522s :      1: mutable_eliminate
 5.70% :     0.000872s :     26: opt.transform.jit_opt_a
 0.29% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.90% :     0.000443s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000039s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 2.78% :     0.000425s :      1: renormalize.infer
 2.17% :     0.000332s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000036s :      1: rewriter_after_opt_a
 0.49% :     0.000075s :      1: rewriter_before_opt_a
 0.57% :     0.000087s :      1: symbol_engine_optimizer
34.05% :     0.005208s :      1: type_inference


TotalTime = 0.0174742, [33]
[bootstrap]: 0.00025004
[type_inference]: 0.00438748
[event_method]: 1.778e-05
[auto_monad]: 6.551e-05
[graph_reusing]: 4.844e-05
[pre_auto_parallel]: 2.71e-06
[py_interpret_to_execute]: 3.429e-05
[rewriter_before_opt_a]: 7.38e-05
[expand_dump_flag]: 3.86999e-06
[jit_opt_a]: 0.0050648, [2]
    [Cycle 1]: 0.00149526, [27]
        [switch_simplify]: 5.092e-05
        [loop_unroll]: 2.858e-05
        [a_1]: 0.00055323
        [with_stream_mark]: 1.783e-05
        [recompute_prepare]: 6.83e-06
        [updatestate_depend_eliminate]: 3.56999e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [parameter_eliminate]: 1.83002e-06
        [specialize_transform]: 5.67001e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.26002e-06
        [meta_shard_fg_expand]: 1.94999e-06
        [get_grad_eliminate_]: 5.27001e-06
        [merge_forward]: 3.28e-06
        [cell_reuse_recompute_pass]: 9.79984e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.407e-05
        [j_node_and_user_rematch]: 8.62998e-06
        [meta_fg_expand]: 2.58e-06
        [replace_old_param]: 7.87003e-06
        [inline_without_move]: 5.28002e-06
        [renormalize]: 0.00054881
        [add_forward_monad_depend]: 4.90999e-06
        [auto_monad_grad]: 1.99e-06
        [auto_monad_eliminator]: 1.227e-05
        [cse]: 2.116e-05
        [replace_applicator]: 1.246e-05
    [Cycle 2]: 0.00031517, [27]
        [switch_simplify]: 6.07001e-06
        [loop_unroll]: 5.40999e-06
        [a_1]: 8.95e-05
        [with_stream_mark]: 8.40999e-06
        [recompute_prepare]: 4.65999e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 4.73001e-06
        [updatestate_useless_node_eliminater]: 4.72998e-06
        [accelerated_algorithm]: 5.01002e-06
        [meta_shard_fg_expand]: 1.05001e-06
        [get_grad_eliminate_]: 4.68999e-06
        [merge_forward]: 2.42001e-06
        [cell_reuse_recompute_pass]: 1.42e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.174e-05
        [j_node_and_user_rematch]: 7.7e-06
        [meta_fg_expand]: 1.89999e-06
        [replace_old_param]: 7.01999e-06
        [inline_without_move]: 4.58999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 4.95001e-06
        [cse]: 9.71e-06
        [replace_applicator]: 4.69998e-06
[py_interpret_to_execute_after_opt_a]: 8.97e-06
[rewriter_after_opt_a]: 3.214e-05
[convert_after_rewriter]: 6.61e-06
[order_py_execute_after_rewriter]: 4.72998e-06
[mutable_eliminate]: 0.00045888
[jit_opt_b]: 4.836e-05, [1]
    [Cycle 1]: 4.149e-05, [2]
        [frontend_op_eliminate]: 1.559e-05
        [inline_after_opt_a]: 1.413e-05
[cconv]: 2.478e-05
[loop_unroll]: 0.0004016
[jit_opt_after_cconv]: 0.00013146, [1]
    [Cycle 1]: 0.00012504, [11]
        [c_1]: 1.979e-05
        [parameter_eliminate]: 2.44001e-06
        [updatestate_depend_eliminate]: 5.19e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 2.01998e-06
        [cse]: 1.559e-05
        [call_graph_tuple_transform]: 1.691e-05
        [tuple_list_get_item_eliminator]: 5.02999e-06
        [none_parameter_eliminate]: 1.40999e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 4.96997e-06
[remove_dup_value]: 1.391e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 4.62998e-06
[add_recomputation]: 4.646e-05
[cse_after_recomputation]: 1.963e-05, [1]
    [Cycle 1]: 1.424e-05, [1]
        [cse]: 8.82e-06
[auto_monad_reorder]: 1.392e-05
[get_jit_bprop_graph]: 1.92001e-06
[rewriter_after_jit_bprop_graph]: 4.53001e-06
[opt_after_jit_grad]: 0.00043166
[symbol_engine_optimizer]: 7.095e-05, [1]
    [Cycle 1]: 6.463e-05, [6]
        [build]: 3.53e-06
        [elim_shapecalc]: 6.80002e-06
        [elim_not_effective]: 1.199e-05
        [opt_reshape]: 5.46002e-06
        [fold_const_symbol]: 8.93002e-06
        [renormalize]: 6.39993e-07
[validate]: 3.196e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00558301
[execute]: 6.90998e-06
Sums
    bootstrap                            :     0.000250s :  1.83%
    type_inference                       :     0.004387s : 32.14%
    event_method                         :     0.000018s :  0.13%
    auto_monad                           :     0.000066s :  0.48%
    graph_reusing                        :     0.000048s :  0.35%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.25%
    rewriter_before_opt_a                :     0.000074s :  0.54%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000057s :  0.42%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000643s :  4.71%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000549s :  4.02%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000031s :  0.23%
    jit_opt_a.replace_applicator         :     0.000017s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000459s :  3.36%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.18%
    loop_unroll                          :     0.000402s :  2.94%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000046s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000432s :  3.16%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005583s : 40.90%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000186    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.57% :     0.000005s :      3: substitution.graph_param_transform
83.55% :     0.000155s :      4: substitution.inline
 1.54% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.25% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.97% :     0.000004s :      2: substitution.replace_old_param
 6.46% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004320     2
81.16% :     0.003506s :      1: type_inference.infer
18.84% :     0.000814s :      1: type_inference.specialize
------[replace.]   0.000059     6
67.46% :     0.000040s :      4: replace.inline
32.54% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000164     6
93.30% :     0.000153s :      4: match.inline
 6.70% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000001s :     10: predicate.addn_zero_filter
 1.83% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000001s :     10: predicate.cast_eliminate
 0.81% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000001s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.15% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.80% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.32% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.45% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.40% :     0.000003s :     15: predicate.partial_eliminate
 1.36% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.35% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.52% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.63% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.60% :     0.000008s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000525    12
47.74% :     0.000251s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.10% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.16% :     0.000237s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018965    76
 0.26% :     0.000049s :      1: add_recomputation
 0.36% :     0.000069s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.38% :     0.000261s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.12% :     0.000022s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.27% :     0.000051s :      1: graph_reusing
26.72% :     0.005067s :      1: jit_opt_a
 0.71% :     0.000134s :      1: jit_opt_after_cconv
 0.27% :     0.000051s :      1: jit_opt_b
 2.16% :     0.000409s :      1: loop_unroll
 2.46% :     0.000467s :      1: mutable_eliminate
 4.45% :     0.000843s :     26: opt.transform.jit_opt_a
 0.23% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.32% :     0.000439s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 1.52% :     0.000289s :      1: renormalize.infer
 1.33% :     0.000252s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000035s :      1: rewriter_after_opt_a
 0.41% :     0.000077s :      1: rewriter_before_opt_a
 0.39% :     0.000073s :      1: symbol_engine_optimizer
29.49% :     0.005592s :      1: task_emit
23.20% :     0.004401s :      1: type_inference
 0.26% :     0.000050s :      1: validate


TotalTime = 0.011404, [30]
[bootstrap]: 0.00027243
[type_inference]: 0.00397003
[event_method]: 1.695e-05
[auto_monad]: 5.97e-05
[graph_reusing]: 4.95001e-06
[pre_auto_parallel]: 2.36998e-06
[py_interpret_to_execute]: 3.026e-05
[rewriter_before_opt_a]: 6.715e-05
[expand_dump_flag]: 2.84999e-06
[jit_opt_a]: 0.00501556, [2]
    [Cycle 1]: 0.00153949, [27]
        [switch_simplify]: 5.118e-05
        [loop_unroll]: 2.822e-05
        [a_1]: 0.00051769
        [with_stream_mark]: 1.722e-05
        [recompute_prepare]: 7.01001e-06
        [updatestate_depend_eliminate]: 3.39001e-06
        [updatestate_assign_eliminate]: 2.74999e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 6.02001e-06
        [updatestate_useless_node_eliminater]: 4.91997e-06
        [accelerated_algorithm]: 5.84999e-06
        [meta_shard_fg_expand]: 2.16e-06
        [get_grad_eliminate_]: 5.54e-06
        [merge_forward]: 3.21001e-06
        [cell_reuse_recompute_pass]: 1.12999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.382e-05
        [j_node_and_user_rematch]: 9.28997e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 8.40001e-06
        [inline_without_move]: 5.34998e-06
        [renormalize]: 0.00061725
        [add_forward_monad_depend]: 5.69e-06
        [auto_monad_grad]: 2.61e-06
        [auto_monad_eliminator]: 1.267e-05
        [cse]: 2.512e-05
        [replace_applicator]: 1.119e-05
    [Cycle 2]: 0.00031398, [27]
        [switch_simplify]: 5.37999e-06
        [loop_unroll]: 4.70999e-06
        [a_1]: 8.821e-05
        [with_stream_mark]: 8.46002e-06
        [recompute_prepare]: 4.84e-06
        [updatestate_depend_eliminate]: 2.63e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.79e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 4.83001e-06
        [meta_shard_fg_expand]: 1.40999e-06
        [get_grad_eliminate_]: 4.88001e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.189e-05
        [j_node_and_user_rematch]: 7.40998e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 7.3e-06
        [inline_without_move]: 4.60001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 5.01997e-06
        [cse]: 1.054e-05
        [replace_applicator]: 7.71001e-06
[py_interpret_to_execute_after_opt_a]: 8.60001e-06
[rewriter_after_opt_a]: 3.039e-05
[convert_after_rewriter]: 7.15e-06
[order_py_execute_after_rewriter]: 4.67e-06
[mutable_eliminate]: 0.00048396
[jit_opt_b]: 4.9e-05, [1]
    [Cycle 1]: 4.235e-05, [2]
        [frontend_op_eliminate]: 1.652e-05
        [inline_after_opt_a]: 1.47e-05
[cconv]: 3.766e-05
[loop_unroll]: 0.00040991
[jit_opt_after_cconv]: 0.00013202, [1]
    [Cycle 1]: 0.00012588, [11]
        [c_1]: 1.948e-05
        [parameter_eliminate]: 2.38002e-06
        [updatestate_depend_eliminate]: 5.22e-06
        [updatestate_assign_eliminate]: 2.31998e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.673e-05
        [call_graph_tuple_transform]: 1.749e-05
        [tuple_list_get_item_eliminator]: 5.24e-06
        [none_parameter_eliminate]: 1.29e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 5.02e-06
[remove_dup_value]: 1.484e-05
[partial_unused_args_eliminate]: 2.59999e-06
[environ_conv]: 4.50001e-06
[add_recomputation]: 4.756e-05
[cse_after_recomputation]: 1.926e-05, [1]
    [Cycle 1]: 1.44e-05, [1]
        [cse]: 8.70001e-06
[auto_monad_reorder]: 1.433e-05
[get_jit_bprop_graph]: 1.69998e-06
[rewriter_after_jit_bprop_graph]: 4.48999e-06
[opt_after_jit_grad]: 0.00043366
[symbol_engine_optimizer]: 6.854e-05, [1]
    [Cycle 1]: 6.242e-05, [6]
        [build]: 3.23998e-06
        [elim_shapecalc]: 7.11001e-06
        [elim_not_effective]: 1.151e-05
        [opt_reshape]: 5.79e-06
        [fold_const_symbol]: 8.05e-06
        [renormalize]: 4.90021e-07
[validate]: 3.317e-05
Sums
    bootstrap                            :     0.000272s :  3.54%
    type_inference                       :     0.003970s : 51.58%
    event_method                         :     0.000017s :  0.22%
    auto_monad                           :     0.000060s :  0.78%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000030s :  0.39%
    rewriter_before_opt_a                :     0.000067s :  0.87%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000057s :  0.73%
    jit_opt_a.loop_unroll                :     0.000033s :  0.43%
    jit_opt_a.a_1                        :     0.000606s :  7.87%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.14%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000617s :  8.02%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.23%
    jit_opt_a.cse                        :     0.000036s :  0.46%
    jit_opt_a.replace_applicator         :     0.000019s :  0.25%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000030s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000484s :  6.29%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000038s :  0.49%
    loop_unroll                          :     0.000410s :  5.33%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000015s :  0.19%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000048s :  0.62%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000014s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.06%
    opt_after_jit_grad                   :     0.000434s :  5.63%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000033s :  0.43%


Time group info:
------[substitution.]   0.000195    23
 0.90% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.60% :     0.000005s :      3: substitution.graph_param_transform
83.72% :     0.000163s :      4: substitution.inline
 1.54% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.27% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.87% :     0.000004s :      2: substitution.replace_old_param
 6.50% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003914     2
79.21% :     0.003100s :      1: type_inference.infer
20.79% :     0.000814s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.85% :     0.000042s :      4: replace.inline
32.15% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000172     6
93.30% :     0.000161s :      4: match.inline
 6.70% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 2.23% :     0.000003s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000001s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.54% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.23% :     0.000001s :      6: predicate.get_grad_eliminate
 0.43% :     0.000000s :      3: predicate.graph_param_transform
 5.23% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.93% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 2.06% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     15: predicate.partial_eliminate
 1.28% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     10: predicate.reduce_eliminate
 1.68% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.19% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.71% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.78% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000572    12
48.03% :     0.000275s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.92% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.05% :     0.000264s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012886    72
 0.39% :     0.000050s :      1: add_recomputation
 0.49% :     0.000063s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.22% :     0.000285s :      1: bootstrap
 0.32% :     0.000041s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000021s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
38.94% :     0.005019s :      1: jit_opt_a
 1.05% :     0.000135s :      1: jit_opt_after_cconv
 0.40% :     0.000051s :      1: jit_opt_b
 3.24% :     0.000418s :      1: loop_unroll
 3.82% :     0.000493s :      1: mutable_eliminate
 6.28% :     0.000810s :     26: opt.transform.jit_opt_a
 0.34% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000029s :      4: opt.transform.symbol_engine_opt
 3.43% :     0.000442s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.04% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000033s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000017s :      1: remove_dup_value
 2.50% :     0.000322s :      1: renormalize.infer
 2.23% :     0.000288s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000033s :      1: rewriter_after_opt_a
 0.54% :     0.000070s :      1: rewriter_before_opt_a
 0.55% :     0.000071s :      1: symbol_engine_optimizer
30.89% :     0.003981s :      1: type_inference


TotalTime = 0.0172181, [33]
[bootstrap]: 0.00025051
[type_inference]: 0.00401111
[event_method]: 1.695e-05
[auto_monad]: 6.134e-05
[graph_reusing]: 5.34e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 3.103e-05
[rewriter_before_opt_a]: 6.78e-05
[expand_dump_flag]: 2.97002e-06
[jit_opt_a]: 0.00501266, [2]
    [Cycle 1]: 0.00145514, [27]
        [switch_simplify]: 5.113e-05
        [loop_unroll]: 2.832e-05
        [a_1]: 0.0005173
        [with_stream_mark]: 1.685e-05
        [recompute_prepare]: 7.61999e-06
        [updatestate_depend_eliminate]: 3.68999e-06
        [updatestate_assign_eliminate]: 2.98998e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [parameter_eliminate]: 1.92001e-06
        [specialize_transform]: 6.10002e-06
        [updatestate_useless_node_eliminater]: 5.19998e-06
        [accelerated_algorithm]: 5.07e-06
        [meta_shard_fg_expand]: 1.96998e-06
        [get_grad_eliminate_]: 5.51998e-06
        [merge_forward]: 3.31999e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.382e-05
        [j_node_and_user_rematch]: 8.55001e-06
        [meta_fg_expand]: 2.19001e-06
        [replace_old_param]: 8.42998e-06
        [inline_without_move]: 4.97999e-06
        [renormalize]: 0.00053664
        [add_forward_monad_depend]: 4.97e-06
        [auto_monad_grad]: 1.79998e-06
        [auto_monad_eliminator]: 1.43e-05
        [cse]: 2.052e-05
        [replace_applicator]: 1.185e-05
    [Cycle 2]: 0.00031364, [27]
        [switch_simplify]: 5.25999e-06
        [loop_unroll]: 4.60001e-06
        [a_1]: 8.828e-05
        [with_stream_mark]: 9.32001e-06
        [recompute_prepare]: 5.19e-06
        [updatestate_depend_eliminate]: 2.78e-06
        [updatestate_assign_eliminate]: 2.13998e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 4.83001e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 4.90999e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.45e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 1.69e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.252e-05
        [j_node_and_user_rematch]: 7.79002e-06
        [meta_fg_expand]: 1.50999e-06
        [replace_old_param]: 6.71e-06
        [inline_without_move]: 4.37998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.59e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 5.09e-06
        [cse]: 1.034e-05
        [replace_applicator]: 4.59002e-06
[py_interpret_to_execute_after_opt_a]: 1.16e-05
[rewriter_after_opt_a]: 3.231e-05
[convert_after_rewriter]: 7.1e-06
[order_py_execute_after_rewriter]: 5.42001e-06
[mutable_eliminate]: 0.00052376
[jit_opt_b]: 5.001e-05, [1]
    [Cycle 1]: 4.274e-05, [2]
        [frontend_op_eliminate]: 1.611e-05
        [inline_after_opt_a]: 1.5e-05
[cconv]: 2.665e-05
[loop_unroll]: 0.00041749
[jit_opt_after_cconv]: 0.00014294, [1]
    [Cycle 1]: 0.00013654, [11]
        [c_1]: 2.079e-05
        [parameter_eliminate]: 3.23998e-06
        [updatestate_depend_eliminate]: 6.21e-06
        [updatestate_assign_eliminate]: 2.33998e-06
        [updatestate_loads_eliminate]: 2.34999e-06
        [cse]: 1.981e-05
        [call_graph_tuple_transform]: 1.851e-05
        [tuple_list_get_item_eliminator]: 5.49998e-06
        [none_parameter_eliminate]: 1.48002e-06
        [renormalize]: 5.00004e-07
        [switch_simplify]: 5.47999e-06
[remove_dup_value]: 1.299e-05
[partial_unused_args_eliminate]: 2.24001e-06
[environ_conv]: 5.25001e-06
[add_recomputation]: 4.639e-05
[cse_after_recomputation]: 2.027e-05, [1]
    [Cycle 1]: 1.494e-05, [1]
        [cse]: 9.10001e-06
[auto_monad_reorder]: 1.42e-05
[get_jit_bprop_graph]: 1.97999e-06
[rewriter_after_jit_bprop_graph]: 5.40001e-06
[opt_after_jit_grad]: 0.00046062
[symbol_engine_optimizer]: 7.267e-05, [1]
    [Cycle 1]: 6.638e-05, [6]
        [build]: 3.54002e-06
        [elim_shapecalc]: 7.41001e-06
        [elim_not_effective]: 1.218e-05
        [opt_reshape]: 5.89e-06
        [fold_const_symbol]: 8.89e-06
        [renormalize]: 5.19998e-07
[validate]: 3.621e-05
[backend_pass]: 1.03001e-06
[task_emit]: 0.00564112
[execute]: 6.27001e-06
Sums
    bootstrap                            :     0.000251s :  1.87%
    type_inference                       :     0.004011s : 30.02%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000061s :  0.46%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.23%
    rewriter_before_opt_a                :     0.000068s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.42%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000606s :  4.53%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000537s :  4.02%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.15%
    jit_opt_a.cse                        :     0.000031s :  0.23%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.09%
    rewriter_after_opt_a                 :     0.000032s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000524s :  3.92%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000027s :  0.20%
    loop_unroll                          :     0.000417s :  3.12%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000046s :  0.35%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000461s :  3.45%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005641s : 42.22%
    execute                              :     0.000006s :  0.05%


Time group info:
------[substitution.]   0.000195    23
 1.07% :     0.000002s :      2: substitution.elim_not_effective
 0.77% :     0.000002s :      2: substitution.fold_const_symbol
 2.64% :     0.000005s :      3: substitution.graph_param_transform
80.90% :     0.000158s :      4: substitution.inline
 1.35% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.36% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.04% :     0.000004s :      2: substitution.replace_old_param
 8.88% :     0.000017s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003955     2
81.37% :     0.003218s :      1: type_inference.infer
18.63% :     0.000737s :      1: type_inference.specialize
------[replace.]   0.000061     6
65.35% :     0.000040s :      4: replace.inline
34.65% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000172     6
90.60% :     0.000156s :      4: match.inline
 9.40% :     0.000016s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.67% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000001s :     10: predicate.addn_zero_filter
 1.93% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.20% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.17% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.43% :     0.000002s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 6.27% :     0.000007s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.91% :     0.000002s :     15: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.14% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.65% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     15: predicate.partial_eliminate
 1.36% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.56% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.49% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.49% :     0.000002s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 2.24% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.45% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.58% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000508    12
47.82% :     0.000243s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.43% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.75% :     0.000227s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018670    76
 0.26% :     0.000049s :      1: add_recomputation
 0.35% :     0.000065s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.40% :     0.000262s :      1: bootstrap
 0.16% :     0.000029s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.86% :     0.005016s :      1: jit_opt_a
 0.78% :     0.000146s :      1: jit_opt_after_cconv
 0.28% :     0.000053s :      1: jit_opt_b
 2.28% :     0.000425s :      1: loop_unroll
 2.85% :     0.000532s :      1: mutable_eliminate
 4.32% :     0.000806s :     26: opt.transform.jit_opt_a
 0.25% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.51% :     0.000468s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000034s :      1: py_interpret_to_execute
 0.08% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.49% :     0.000278s :      1: renormalize.infer
 1.35% :     0.000251s :      1: renormalize.specialize
 0.20% :     0.000038s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000035s :      1: rewriter_after_opt_a
 0.38% :     0.000071s :      1: rewriter_before_opt_a
 0.40% :     0.000075s :      1: symbol_engine_optimizer
30.28% :     0.005654s :      1: task_emit
21.55% :     0.004023s :      1: type_inference
 0.30% :     0.000057s :      1: validate


TotalTime = 0.0120473, [30]
[bootstrap]: 0.00025779
[type_inference]: 0.00425181
[event_method]: 1.654e-05
[auto_monad]: 6.047e-05
[graph_reusing]: 5.46e-06
[pre_auto_parallel]: 2.53e-06
[py_interpret_to_execute]: 3.274e-05
[rewriter_before_opt_a]: 7.007e-05
[expand_dump_flag]: 3.14999e-06
[jit_opt_a]: 0.0052974, [2]
    [Cycle 1]: 0.0016766, [27]
        [switch_simplify]: 5.294e-05
        [loop_unroll]: 2.848e-05
        [a_1]: 0.00053749
        [with_stream_mark]: 1.835e-05
        [recompute_prepare]: 7.15998e-06
        [updatestate_depend_eliminate]: 3.21999e-06
        [updatestate_assign_eliminate]: 3.04999e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 2.13998e-06
        [specialize_transform]: 5.83997e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.59e-06
        [meta_shard_fg_expand]: 2.54001e-06
        [get_grad_eliminate_]: 5.44998e-06
        [merge_forward]: 3.75e-06
        [cell_reuse_recompute_pass]: 1.11002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.4e-05
        [j_node_and_user_rematch]: 9.77001e-06
        [meta_fg_expand]: 2.62001e-06
        [replace_old_param]: 8.57e-06
        [inline_without_move]: 5.81e-06
        [renormalize]: 0.00072266
        [add_forward_monad_depend]: 5.00001e-06
        [auto_monad_grad]: 2.76999e-06
        [auto_monad_eliminator]: 1.425e-05
        [cse]: 2.537e-05
        [replace_applicator]: 1.222e-05
    [Cycle 2]: 0.00031827, [27]
        [switch_simplify]: 5.51998e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 8.851e-05
        [with_stream_mark]: 9.31998e-06
        [recompute_prepare]: 5.06997e-06
        [updatestate_depend_eliminate]: 2.51e-06
        [updatestate_assign_eliminate]: 1.90001e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 5.25999e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 1.15999e-06
        [get_grad_eliminate_]: 5.23002e-06
        [merge_forward]: 2.24001e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.22e-05
        [j_node_and_user_rematch]: 7.77e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 6.89001e-06
        [inline_without_move]: 4.63001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 6.60017e-07
        [auto_monad_eliminator]: 5.19e-06
        [cse]: 9.74e-06
        [replace_applicator]: 5.47999e-06
[py_interpret_to_execute_after_opt_a]: 9.99001e-06
[rewriter_after_opt_a]: 3.239e-05
[convert_after_rewriter]: 6.31e-06
[order_py_execute_after_rewriter]: 5.24e-06
[mutable_eliminate]: 0.00052813
[jit_opt_b]: 5.106e-05, [1]
    [Cycle 1]: 4.413e-05, [2]
        [frontend_op_eliminate]: 1.706e-05
        [inline_after_opt_a]: 1.544e-05
[cconv]: 2.44e-05
[loop_unroll]: 0.00042126
[jit_opt_after_cconv]: 0.00014298, [1]
    [Cycle 1]: 0.0001358, [11]
        [c_1]: 2.1e-05
        [parameter_eliminate]: 2.76e-06
        [updatestate_depend_eliminate]: 6.43003e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [cse]: 1.856e-05
        [call_graph_tuple_transform]: 1.877e-05
        [tuple_list_get_item_eliminator]: 5.62001e-06
        [none_parameter_eliminate]: 1.64998e-06
        [renormalize]: 6.39993e-07
        [switch_simplify]: 5.09e-06
[remove_dup_value]: 1.335e-05
[partial_unused_args_eliminate]: 2.42001e-06
[environ_conv]: 4.97e-06
[add_recomputation]: 4.472e-05
[cse_after_recomputation]: 1.988e-05, [1]
    [Cycle 1]: 1.445e-05, [1]
        [cse]: 8.28999e-06
[auto_monad_reorder]: 1.425e-05
[get_jit_bprop_graph]: 1.89e-06
[rewriter_after_jit_bprop_graph]: 5.34e-06
[opt_after_jit_grad]: 0.00045201
[symbol_engine_optimizer]: 7.243e-05, [1]
    [Cycle 1]: 6.605e-05, [6]
        [build]: 3.87998e-06
        [elim_shapecalc]: 7.08e-06
        [elim_not_effective]: 1.167e-05
        [opt_reshape]: 5.54998e-06
        [fold_const_symbol]: 9.41e-06
        [renormalize]: 5.19998e-07
[validate]: 3.429e-05
Sums
    bootstrap                            :     0.000258s :  3.15%
    type_inference                       :     0.004252s : 52.01%
    event_method                         :     0.000017s :  0.20%
    auto_monad                           :     0.000060s :  0.74%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.40%
    rewriter_before_opt_a                :     0.000070s :  0.86%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000058s :  0.72%
    jit_opt_a.loop_unroll                :     0.000033s :  0.41%
    jit_opt_a.a_1                        :     0.000626s :  7.66%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.34%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000723s :  8.84%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.24%
    jit_opt_a.cse                        :     0.000035s :  0.43%
    jit_opt_a.replace_applicator         :     0.000018s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000032s :  0.40%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000528s :  6.46%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000024s :  0.30%
    loop_unroll                          :     0.000421s :  5.15%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000019s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000013s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000045s :  0.55%
    cse_after_recomputation.cse          :     0.000008s :  0.10%
    auto_monad_reorder                   :     0.000014s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000452s :  5.53%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000034s :  0.42%


Time group info:
------[substitution.]   0.000209    23
 0.90% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.70% :     0.000006s :      3: substitution.graph_param_transform
84.96% :     0.000177s :      4: substitution.inline
 1.65% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.14% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.57% :     0.000003s :      2: substitution.replace_old_param
 5.50% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004194     2
80.28% :     0.003367s :      1: type_inference.infer
19.72% :     0.000827s :      1: type_inference.specialize
------[replace.]   0.000067     6
65.84% :     0.000044s :      4: replace.inline
34.16% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000185     6
94.39% :     0.000175s :      4: match.inline
 5.61% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.66% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.39% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000001s :     10: predicate.addn_zero_filter
 2.03% :     0.000002s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.54% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.15% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.06% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.28% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.10% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.96% :     0.000007s :     22: predicate.inline
 1.14% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000001s :      6: predicate.less_batch_normalization
 1.74% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.32% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000002s :     15: predicate.partial_eliminate
 1.42% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.62% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.57% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.65% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.26% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.52% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.07% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.99% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000596    12
48.26% :     0.000287s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.66% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.08% :     0.000269s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013658    72
 0.35% :     0.000047s :      1: add_recomputation
 0.47% :     0.000064s :      1: auto_monad
 0.12% :     0.000017s :      1: auto_monad_reorder
 1.97% :     0.000269s :      1: bootstrap
 0.20% :     0.000027s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.81% :     0.005301s :      1: jit_opt_a
 1.07% :     0.000146s :      1: jit_opt_after_cconv
 0.39% :     0.000054s :      1: jit_opt_b
 3.14% :     0.000429s :      1: loop_unroll
 3.93% :     0.000537s :      1: mutable_eliminate
 6.10% :     0.000833s :     26: opt.transform.jit_opt_a
 0.34% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.37% :     0.000460s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000035s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.85% :     0.000389s :      1: renormalize.infer
 2.38% :     0.000326s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000038s :      1: rewriter_after_opt_a
 0.53% :     0.000073s :      1: rewriter_before_opt_a
 0.55% :     0.000075s :      1: symbol_engine_optimizer
31.21% :     0.004263s :      1: type_inference


TotalTime = 0.0313874, [33]
[bootstrap]: 0.00046197
[type_inference]: 0.0100217
[event_method]: 2.285e-05
[auto_monad]: 7.312e-05
[graph_reusing]: 5.97001e-06
[pre_auto_parallel]: 2.94999e-06
[py_interpret_to_execute]: 4.485e-05
[rewriter_before_opt_a]: 7.785e-05
[expand_dump_flag]: 3.84002e-06
[jit_opt_a]: 0.00751762, [2]
    [Cycle 1]: 0.00208446, [27]
        [switch_simplify]: 6.603e-05
        [loop_unroll]: 3.07e-05
        [a_1]: 0.00069652
        [with_stream_mark]: 3.123e-05
        [recompute_prepare]: 8.07998e-06
        [updatestate_depend_eliminate]: 4.08001e-06
        [updatestate_assign_eliminate]: 3.85e-06
        [updatestate_loads_eliminate]: 2.50997e-06
        [parameter_eliminate]: 2.43e-06
        [specialize_transform]: 6.07001e-06
        [updatestate_useless_node_eliminater]: 5.61e-06
        [accelerated_algorithm]: 6.09001e-06
        [meta_shard_fg_expand]: 2.36e-06
        [get_grad_eliminate_]: 6.19999e-06
        [merge_forward]: 4.63999e-06
        [cell_reuse_recompute_pass]: 1.77001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.607e-05
        [j_node_and_user_rematch]: 1.002e-05
        [meta_fg_expand]: 2.21998e-06
        [replace_old_param]: 1.098e-05
        [inline_without_move]: 6.18998e-06
        [renormalize]: 0.00086987
        [add_forward_monad_depend]: 1.017e-05
        [auto_monad_grad]: 2.74001e-06
        [auto_monad_eliminator]: 1.852e-05
        [cse]: 3.405e-05
        [replace_applicator]: 1.776e-05
    [Cycle 2]: 0.00034562, [27]
        [switch_simplify]: 5.94999e-06
        [loop_unroll]: 5.32001e-06
        [a_1]: 9.391e-05
        [with_stream_mark]: 1.35e-05
        [recompute_prepare]: 5.30001e-06
        [updatestate_depend_eliminate]: 3.41999e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 5.30999e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 4.68001e-06
        [meta_shard_fg_expand]: 1.69e-06
        [get_grad_eliminate_]: 4.50999e-06
        [merge_forward]: 3.93999e-06
        [cell_reuse_recompute_pass]: 3.60998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.682e-05
        [j_node_and_user_rematch]: 7.27002e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 8.27998e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.92999e-06
        [auto_monad_grad]: 1.39998e-06
        [auto_monad_eliminator]: 5.74e-06
        [cse]: 1.209e-05
        [replace_applicator]: 4.70999e-06
[py_interpret_to_execute_after_opt_a]: 1.595e-05
[rewriter_after_opt_a]: 4.177e-05
[convert_after_rewriter]: 8.25e-06
[order_py_execute_after_rewriter]: 4.94998e-06
[mutable_eliminate]: 0.00081269
[jit_opt_b]: 6.251e-05, [1]
    [Cycle 1]: 4.329e-05, [2]
        [frontend_op_eliminate]: 1.609e-05
        [inline_after_opt_a]: 1.607e-05
[cconv]: 3.674e-05
[loop_unroll]: 0.00050808
[jit_opt_after_cconv]: 0.0001698, [1]
    [Cycle 1]: 0.00016275, [11]
        [c_1]: 2.083e-05
        [parameter_eliminate]: 6.17999e-06
        [updatestate_depend_eliminate]: 1.013e-05
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [cse]: 3.314e-05
        [call_graph_tuple_transform]: 2.4e-05
        [tuple_list_get_item_eliminator]: 5.10001e-06
        [none_parameter_eliminate]: 1.57999e-06
        [renormalize]: 1.05999e-06
        [switch_simplify]: 5.27999e-06
[remove_dup_value]: 1.608e-05
[partial_unused_args_eliminate]: 2.48002e-06
[environ_conv]: 5.63997e-06
[add_recomputation]: 6.407e-05
[cse_after_recomputation]: 2.181e-05, [1]
    [Cycle 1]: 1.618e-05, [1]
        [cse]: 9.93002e-06
[auto_monad_reorder]: 1.673e-05
[get_jit_bprop_graph]: 2.37999e-06
[rewriter_after_jit_bprop_graph]: 8.93002e-06
[opt_after_jit_grad]: 0.00058543
[symbol_engine_optimizer]: 8.17e-05, [1]
    [Cycle 1]: 7.41e-05, [6]
        [build]: 5.43002e-06
        [elim_shapecalc]: 8.44998e-06
        [elim_not_effective]: 1.534e-05
        [opt_reshape]: 5.74999e-06
        [fold_const_symbol]: 8.69998e-06
        [renormalize]: 1.09e-06
[validate]: 4.987e-05
[backend_pass]: 1.36998e-06
[task_emit]: 0.0103507
[execute]: 1.106e-05
Sums
    bootstrap                            :     0.000462s :  1.81%
    type_inference                       :     0.010022s : 39.21%
    event_method                         :     0.000023s :  0.09%
    auto_monad                           :     0.000073s :  0.29%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000045s :  0.18%
    rewriter_before_opt_a                :     0.000078s :  0.30%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000072s :  0.28%
    jit_opt_a.loop_unroll                :     0.000036s :  0.14%
    jit_opt_a.a_1                        :     0.000790s :  3.09%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.08%
    jit_opt_a.inline_without_move        :     0.000011s :  0.04%
    jit_opt_a.renormalize                :     0.000870s :  3.40%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.09%
    jit_opt_a.cse                        :     0.000046s :  0.18%
    jit_opt_a.replace_applicator         :     0.000022s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.06%
    rewriter_after_opt_a                 :     0.000042s :  0.16%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000813s :  3.18%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.06%
    cconv                                :     0.000037s :  0.14%
    loop_unroll                          :     0.000508s :  1.99%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000033s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000064s :  0.25%
    cse_after_recomputation.cse          :     0.000010s :  0.04%
    auto_monad_reorder                   :     0.000017s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.03%
    opt_after_jit_grad                   :     0.000585s :  2.29%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000050s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.010351s : 40.49%
    execute                              :     0.000011s :  0.04%


Time group info:
------[substitution.]   0.000279    23
 0.80% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 2.28% :     0.000006s :      3: substitution.graph_param_transform
85.68% :     0.000239s :      4: substitution.inline
 1.21% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.14% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.11% :     0.000006s :      2: substitution.replace_old_param
 5.33% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009919     2
88.72% :     0.008800s :      1: type_inference.infer
11.28% :     0.001119s :      1: type_inference.specialize
------[replace.]   0.000126     6
76.15% :     0.000096s :      4: replace.inline
23.85% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000250     6
94.57% :     0.000236s :      4: match.inline
 5.43% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.77% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 1.87% :     0.000002s :     10: predicate.arithmetic_simplify
 1.61% :     0.000002s :     10: predicate.cast_eliminate
 0.33% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.46% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.12% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.25% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.48% :     0.000002s :     10: predicate.environ_get_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.50% :     0.000001s :      3: predicate.graph_param_transform
 7.27% :     0.000009s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.59% :     0.000002s :     15: predicate.load_eliminater
 2.08% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.73% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.66% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 2.78% :     0.000004s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.00% :     0.000003s :     15: predicate.partial_eliminate
 1.31% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.42% :     0.000002s :     10: predicate.reduce_eliminate
 1.78% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.90% :     0.000001s :      6: predicate.replace_old_param
 0.79% :     0.000001s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.54% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.30% :     0.000002s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.19% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.58% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.31% :     0.000003s :     16: predicate.switch_defer_inline
 1.86% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.33% :     0.000009s :     53: predicate.switch_simplify
 1.07% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000002s :     10: predicate.transpose_eliminate
 1.25% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.03% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.59% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.88% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.37% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.37% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000783    12
44.46% :     0.000348s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.63% :     0.000060s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.91% :     0.000375s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.033401    76
 0.20% :     0.000067s :      1: add_recomputation
 0.23% :     0.000077s :      1: auto_monad
 0.06% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.47% :     0.000492s :      1: bootstrap
 0.12% :     0.000040s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.07% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.08% :     0.000027s :      1: event_method
 0.05% :     0.000017s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
22.52% :     0.007521s :      1: jit_opt_a
 0.52% :     0.000173s :      1: jit_opt_after_cconv
 0.20% :     0.000066s :      1: jit_opt_b
 1.55% :     0.000518s :      1: loop_unroll
 2.49% :     0.000831s :      1: mutable_eliminate
 3.07% :     0.001027s :     26: opt.transform.jit_opt_a
 0.16% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.79% :     0.000597s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000048s :      1: py_interpret_to_execute
 0.05% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 1.49% :     0.000497s :      1: renormalize.infer
 1.09% :     0.000363s :      1: renormalize.specialize
 0.03% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000045s :      1: rewriter_after_opt_a
 0.24% :     0.000081s :      1: rewriter_before_opt_a
 0.25% :     0.000084s :      1: symbol_engine_optimizer
31.06% :     0.010373s :      1: task_emit
30.10% :     0.010053s :      1: type_inference
 0.23% :     0.000077s :      1: validate


TotalTime = 0.0168543, [33]
[bootstrap]: 0.00033751
[type_inference]: 0.00419572
[event_method]: 1.68e-05
[auto_monad]: 5.767e-05
[graph_reusing]: 6.44001e-06
[pre_auto_parallel]: 2.36e-06
[py_interpret_to_execute]: 2.726e-05
[rewriter_before_opt_a]: 7.173e-05
[expand_dump_flag]: 2.99999e-06
[jit_opt_a]: 0.00471026, [2]
    [Cycle 1]: 0.00135869, [27]
        [switch_simplify]: 4.805e-05
        [loop_unroll]: 2.877e-05
        [a_1]: 0.00050826
        [with_stream_mark]: 1.541e-05
        [recompute_prepare]: 7.07002e-06
        [updatestate_depend_eliminate]: 3.3e-06
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 2.69001e-06
        [parameter_eliminate]: 1.76e-06
        [specialize_transform]: 5.98998e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 5.31998e-06
        [meta_shard_fg_expand]: 1.80001e-06
        [get_grad_eliminate_]: 5.53997e-06
        [merge_forward]: 3.73001e-06
        [cell_reuse_recompute_pass]: 1.00001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.259e-05
        [j_node_and_user_rematch]: 8.50999e-06
        [meta_fg_expand]: 2.27001e-06
        [replace_old_param]: 8.09002e-06
        [inline_without_move]: 5.14998e-06
        [renormalize]: 0.00046696
        [add_forward_monad_depend]: 4.52e-06
        [auto_monad_grad]: 2.09999e-06
        [auto_monad_eliminator]: 1.235e-05
        [cse]: 1.931e-05
        [replace_applicator]: 1.185e-05
    [Cycle 2]: 0.0003096, [27]
        [switch_simplify]: 5.30999e-06
        [loop_unroll]: 4.85001e-06
        [a_1]: 8.729e-05
        [with_stream_mark]: 8.08001e-06
        [recompute_prepare]: 5.02999e-06
        [updatestate_depend_eliminate]: 2.82002e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 9.60019e-07
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 4.92e-06
        [meta_shard_fg_expand]: 1.26002e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.35002e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.183e-05
        [j_node_and_user_rematch]: 7.53e-06
        [meta_fg_expand]: 1.40001e-06
        [replace_old_param]: 7.07002e-06
        [inline_without_move]: 4.58999e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.18001e-06
        [auto_monad_grad]: 9.89996e-07
        [auto_monad_eliminator]: 5.12e-06
        [cse]: 8.80001e-06
        [replace_applicator]: 5.05001e-06
[py_interpret_to_execute_after_opt_a]: 8.77999e-06
[rewriter_after_opt_a]: 3.157e-05
[convert_after_rewriter]: 6.93998e-06
[order_py_execute_after_rewriter]: 4.84e-06
[mutable_eliminate]: 0.00046649
[jit_opt_b]: 4.824e-05, [1]
    [Cycle 1]: 4.167e-05, [2]
        [frontend_op_eliminate]: 1.583e-05
        [inline_after_opt_a]: 1.452e-05
[cconv]: 2.373e-05
[loop_unroll]: 0.0003956
[jit_opt_after_cconv]: 0.00013, [1]
    [Cycle 1]: 0.00012409, [11]
        [c_1]: 1.943e-05
        [parameter_eliminate]: 2.34001e-06
        [updatestate_depend_eliminate]: 5.29998e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [cse]: 1.547e-05
        [call_graph_tuple_transform]: 1.695e-05
        [tuple_list_get_item_eliminator]: 5.04998e-06
        [none_parameter_eliminate]: 1.39003e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 5.00999e-06
[remove_dup_value]: 1.255e-05
[partial_unused_args_eliminate]: 2.06998e-06
[environ_conv]: 4.48001e-06
[add_recomputation]: 4.212e-05
[cse_after_recomputation]: 1.962e-05, [1]
    [Cycle 1]: 1.453e-05, [1]
        [cse]: 8.93002e-06
[auto_monad_reorder]: 1.346e-05
[get_jit_bprop_graph]: 1.66e-06
[rewriter_after_jit_bprop_graph]: 4.68999e-06
[opt_after_jit_grad]: 0.00042746
[symbol_engine_optimizer]: 6.961e-05, [1]
    [Cycle 1]: 6.321e-05, [6]
        [build]: 2.99999e-06
        [elim_shapecalc]: 7.31999e-06
        [elim_not_effective]: 1.177e-05
        [opt_reshape]: 5.40999e-06
        [fold_const_symbol]: 8.59e-06
        [renormalize]: 4.19997e-07
[validate]: 3.11e-05
[backend_pass]: 8.29983e-07
[task_emit]: 0.00550169
[execute]: 7.10998e-06
Sums
    bootstrap                            :     0.000338s :  2.55%
    type_inference                       :     0.004196s : 31.66%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000058s :  0.44%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000027s :  0.21%
    rewriter_before_opt_a                :     0.000072s :  0.54%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000596s :  4.49%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000467s :  3.52%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000028s :  0.21%
    jit_opt_a.replace_applicator         :     0.000017s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000466s :  3.52%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000396s :  2.98%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000042s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000013s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000427s :  3.23%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005502s : 41.51%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000172    23
 1.20% :     0.000002s :      2: substitution.elim_not_effective
 0.77% :     0.000001s :      2: substitution.fold_const_symbol
 2.83% :     0.000005s :      3: substitution.graph_param_transform
82.35% :     0.000142s :      4: substitution.inline
 1.74% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.30% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.00% :     0.000003s :      2: substitution.replace_old_param
 6.80% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004145     2
81.89% :     0.003394s :      1: type_inference.infer
18.11% :     0.000751s :      1: type_inference.specialize
------[replace.]   0.000059     6
66.87% :     0.000039s :      4: replace.inline
33.13% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000149     6
93.08% :     0.000139s :      4: match.inline
 6.92% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.32% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000001s :     10: predicate.addn_zero_filter
 1.75% :     0.000002s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000001s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.18% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.74% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.33% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.58% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     15: predicate.partial_eliminate
 1.90% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.89% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.49% :     0.000002s :     10: predicate.same_eliminate
 0.63% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.52% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.29% :     0.000008s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.93% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000497    12
49.67% :     0.000247s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.94% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.39% :     0.000220s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018213    76
 0.25% :     0.000045s :      1: add_recomputation
 0.34% :     0.000061s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.92% :     0.000349s :      1: bootstrap
 0.15% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
25.87% :     0.004713s :      1: jit_opt_a
 0.73% :     0.000133s :      1: jit_opt_after_cconv
 0.28% :     0.000051s :      1: jit_opt_b
 2.21% :     0.000403s :      1: loop_unroll
 2.61% :     0.000475s :      1: mutable_eliminate
 4.34% :     0.000791s :     26: opt.transform.jit_opt_a
 0.24% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.38% :     0.000434s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.30% :     0.000236s :      1: renormalize.infer
 1.22% :     0.000223s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000035s :      1: rewriter_after_opt_a
 0.41% :     0.000075s :      1: rewriter_before_opt_a
 0.40% :     0.000072s :      1: symbol_engine_optimizer
30.27% :     0.005512s :      1: task_emit
23.09% :     0.004206s :      1: type_inference
 0.26% :     0.000048s :      1: validate


TotalTime = 0.0112044, [30]
[bootstrap]: 0.00025536
[type_inference]: 0.00400599
[event_method]: 1.568e-05
[auto_monad]: 6.077e-05
[graph_reusing]: 5.38002e-06
[pre_auto_parallel]: 2.38998e-06
[py_interpret_to_execute]: 3.068e-05
[rewriter_before_opt_a]: 7.281e-05
[expand_dump_flag]: 2.96001e-06
[jit_opt_a]: 0.00481802, [2]
    [Cycle 1]: 0.00139997, [27]
        [switch_simplify]: 4.912e-05
        [loop_unroll]: 2.852e-05
        [a_1]: 0.00049199
        [with_stream_mark]: 1.55e-05
        [recompute_prepare]: 6.56999e-06
        [updatestate_depend_eliminate]: 3.31999e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 5.74e-06
        [updatestate_useless_node_eliminater]: 5.09003e-06
        [accelerated_algorithm]: 5.94e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 5.87001e-06
        [merge_forward]: 3.63e-06
        [cell_reuse_recompute_pass]: 8.30012e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.326e-05
        [j_node_and_user_rematch]: 9.09e-06
        [meta_fg_expand]: 2.41998e-06
        [replace_old_param]: 8.47e-06
        [inline_without_move]: 5.13002e-06
        [renormalize]: 0.00052021
        [add_forward_monad_depend]: 4.87998e-06
        [auto_monad_grad]: 1.84998e-06
        [auto_monad_eliminator]: 1.238e-05
        [cse]: 1.895e-05
        [replace_applicator]: 1.056e-05
    [Cycle 2]: 0.00031169, [27]
        [switch_simplify]: 5.38002e-06
        [loop_unroll]: 5.14e-06
        [a_1]: 8.957e-05
        [with_stream_mark]: 8.31002e-06
        [recompute_prepare]: 4.99e-06
        [updatestate_depend_eliminate]: 2.55002e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 9.20001e-07
        [specialize_transform]: 4.83001e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.37999e-06
        [meta_shard_fg_expand]: 1.14003e-06
        [get_grad_eliminate_]: 4.90001e-06
        [merge_forward]: 2.46998e-06
        [cell_reuse_recompute_pass]: 1.31002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.203e-05
        [j_node_and_user_rematch]: 7.58001e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 7.24001e-06
        [inline_without_move]: 4.63999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 6.99976e-07
        [auto_monad_eliminator]: 4.85001e-06
        [cse]: 9.24e-06
        [replace_applicator]: 4.94998e-06
[py_interpret_to_execute_after_opt_a]: 8.17998e-06
[rewriter_after_opt_a]: 2.958e-05
[convert_after_rewriter]: 6.29999e-06
[order_py_execute_after_rewriter]: 4.77e-06
[mutable_eliminate]: 0.00046747
[jit_opt_b]: 5.227e-05, [1]
    [Cycle 1]: 4.594e-05, [2]
        [frontend_op_eliminate]: 1.912e-05
        [inline_after_opt_a]: 1.542e-05
[cconv]: 2.302e-05
[loop_unroll]: 0.0004014
[jit_opt_after_cconv]: 0.00013266, [1]
    [Cycle 1]: 0.0001262, [11]
        [c_1]: 2.055e-05
        [parameter_eliminate]: 2.36e-06
        [updatestate_depend_eliminate]: 4.90001e-06
        [updatestate_assign_eliminate]: 2.30002e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.607e-05
        [call_graph_tuple_transform]: 1.673e-05
        [tuple_list_get_item_eliminator]: 5.12e-06
        [none_parameter_eliminate]: 1.71e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 5.24e-06
[remove_dup_value]: 1.332e-05
[partial_unused_args_eliminate]: 1.82001e-06
[environ_conv]: 4.82998e-06
[add_recomputation]: 4.414e-05
[cse_after_recomputation]: 1.927e-05, [1]
    [Cycle 1]: 1.396e-05, [1]
        [cse]: 8.53001e-06
[auto_monad_reorder]: 1.434e-05
[get_jit_bprop_graph]: 1.55999e-06
[rewriter_after_jit_bprop_graph]: 4.55001e-06
[opt_after_jit_grad]: 0.0004494
[symbol_engine_optimizer]: 7.305e-05, [1]
    [Cycle 1]: 6.646e-05, [6]
        [build]: 3.24001e-06
        [elim_shapecalc]: 8.35999e-06
        [elim_not_effective]: 1.25e-05
        [opt_reshape]: 5.79e-06
        [fold_const_symbol]: 8.52e-06
        [renormalize]: 4.20026e-07
[validate]: 3.055e-05
Sums
    bootstrap                            :     0.000255s :  3.38%
    type_inference                       :     0.004006s : 53.03%
    event_method                         :     0.000016s :  0.21%
    auto_monad                           :     0.000061s :  0.80%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000031s :  0.41%
    rewriter_before_opt_a                :     0.000073s :  0.96%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000055s :  0.72%
    jit_opt_a.loop_unroll                :     0.000034s :  0.45%
    jit_opt_a.a_1                        :     0.000582s :  7.70%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.14%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000520s :  6.89%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.23%
    jit_opt_a.cse                        :     0.000028s :  0.37%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000030s :  0.39%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000467s :  6.19%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.25%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.20%
    cconv                                :     0.000023s :  0.30%
    loop_unroll                          :     0.000401s :  5.31%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000013s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000044s :  0.58%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000014s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000449s :  5.95%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000031s :  0.40%


Time group info:
------[substitution.]   0.000172    23
 1.12% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.68% :     0.000005s :      3: substitution.graph_param_transform
82.48% :     0.000142s :      4: substitution.inline
 1.68% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.76% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000004s :      2: substitution.replace_old_param
 6.49% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003951     2
79.20% :     0.003129s :      1: type_inference.infer
20.80% :     0.000822s :      1: type_inference.specialize
------[replace.]   0.000059     6
66.80% :     0.000039s :      4: replace.inline
33.20% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000149     6
93.27% :     0.000139s :      4: match.inline
 6.73% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.72% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 1.78% :     0.000002s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.23% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.70% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.12% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.42% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.48% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.95% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.58% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000002s :     15: predicate.partial_eliminate
 1.32% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     10: predicate.reduce_eliminate
 1.66% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.83% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.05% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.55% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000009s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.88% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.05% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000548    12
47.36% :     0.000259s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.11% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.53% :     0.000260s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012572    72
 0.37% :     0.000047s :      1: add_recomputation
 0.51% :     0.000064s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.12% :     0.000267s :      1: bootstrap
 0.20% :     0.000026s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000021s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.15% :     0.000019s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.35% :     0.004821s :      1: jit_opt_a
 1.08% :     0.000135s :      1: jit_opt_after_cconv
 0.44% :     0.000055s :      1: jit_opt_b
 3.25% :     0.000409s :      1: loop_unroll
 3.78% :     0.000476s :      1: mutable_eliminate
 6.21% :     0.000781s :     26: opt.transform.jit_opt_a
 0.35% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000028s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.64% :     0.000458s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000033s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.02% :     0.000254s :      1: renormalize.infer
 2.06% :     0.000259s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000035s :      1: rewriter_after_opt_a
 0.60% :     0.000075s :      1: rewriter_before_opt_a
 0.60% :     0.000076s :      1: symbol_engine_optimizer
31.95% :     0.004017s :      1: type_inference


TotalTime = 0.0118439, [30]
[bootstrap]: 0.00025906
[type_inference]: 0.00418523
[event_method]: 1.678e-05
[auto_monad]: 6.583e-05
[graph_reusing]: 5.75001e-06
[pre_auto_parallel]: 2.39001e-06
[py_interpret_to_execute]: 3.358e-05
[rewriter_before_opt_a]: 7.239e-05
[expand_dump_flag]: 3.26001e-06
[jit_opt_a]: 0.00513634, [2]
    [Cycle 1]: 0.00159194, [27]
        [switch_simplify]: 5.395e-05
        [loop_unroll]: 3.06e-05
        [a_1]: 0.00054802
        [with_stream_mark]: 1.897e-05
        [recompute_prepare]: 7.3e-06
        [updatestate_depend_eliminate]: 3.36001e-06
        [updatestate_assign_eliminate]: 2.89001e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [parameter_eliminate]: 1.94999e-06
        [specialize_transform]: 6.93998e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 6.25002e-06
        [meta_shard_fg_expand]: 2.02999e-06
        [get_grad_eliminate_]: 5.81e-06
        [merge_forward]: 3.57002e-06
        [cell_reuse_recompute_pass]: 9.5999e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.521e-05
        [j_node_and_user_rematch]: 1.018e-05
        [meta_fg_expand]: 2.53e-06
        [replace_old_param]: 9.09e-06
        [inline_without_move]: 5.67001e-06
        [renormalize]: 0.00060937
        [add_forward_monad_depend]: 5.79999e-06
        [auto_monad_grad]: 1.93002e-06
        [auto_monad_eliminator]: 1.396e-05
        [cse]: 3.031e-05
        [replace_applicator]: 1.108e-05
    [Cycle 2]: 0.0003646, [27]
        [switch_simplify]: 5.25999e-06
        [loop_unroll]: 5.25999e-06
        [a_1]: 9.405e-05
        [with_stream_mark]: 8.94e-06
        [recompute_prepare]: 5.66998e-06
        [updatestate_depend_eliminate]: 3.04999e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.19998e-06
        [meta_shard_fg_expand]: 1.15999e-06
        [get_grad_eliminate_]: 4.55999e-06
        [merge_forward]: 2.31e-06
        [cell_reuse_recompute_pass]: 1.24003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.285e-05
        [j_node_and_user_rematch]: 8.29998e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 7.62998e-06
        [inline_without_move]: 5.04e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 6.50005e-07
        [auto_monad_eliminator]: 5.39e-06
        [cse]: 1.209e-05
        [replace_applicator]: 5.72001e-06
[py_interpret_to_execute_after_opt_a]: 1.092e-05
[rewriter_after_opt_a]: 3.26e-05
[convert_after_rewriter]: 6.88e-06
[order_py_execute_after_rewriter]: 5.20999e-06
[mutable_eliminate]: 0.0005332
[jit_opt_b]: 5.161e-05, [1]
    [Cycle 1]: 4.432e-05, [2]
        [frontend_op_eliminate]: 1.706e-05
        [inline_after_opt_a]: 1.526e-05
[cconv]: 2.425e-05
[loop_unroll]: 0.00042855
[jit_opt_after_cconv]: 0.00014176, [1]
    [Cycle 1]: 0.00013575, [11]
        [c_1]: 2.274e-05
        [parameter_eliminate]: 2.59999e-06
        [updatestate_depend_eliminate]: 5.74e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.13002e-06
        [cse]: 1.66e-05
        [call_graph_tuple_transform]: 1.957e-05
        [tuple_list_get_item_eliminator]: 6.02999e-06
        [none_parameter_eliminate]: 1.37999e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 5.46e-06
[remove_dup_value]: 1.398e-05
[partial_unused_args_eliminate]: 2.67001e-06
[environ_conv]: 4.75001e-06
[add_recomputation]: 4.607e-05
[cse_after_recomputation]: 2.008e-05, [1]
    [Cycle 1]: 1.47e-05, [1]
        [cse]: 8.85001e-06
[auto_monad_reorder]: 1.443e-05
[get_jit_bprop_graph]: 1.62999e-06
[rewriter_after_jit_bprop_graph]: 5.59e-06
[opt_after_jit_grad]: 0.00044744
[symbol_engine_optimizer]: 7.368e-05, [1]
    [Cycle 1]: 6.684e-05, [6]
        [build]: 3.41001e-06
        [elim_shapecalc]: 7.61001e-06
        [elim_not_effective]: 1.326e-05
        [opt_reshape]: 5.56998e-06
        [fold_const_symbol]: 9.24998e-06
        [renormalize]: 4.40021e-07
[validate]: 3.278e-05
Sums
    bootstrap                            :     0.000259s :  3.22%
    type_inference                       :     0.004185s : 52.00%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000066s :  0.82%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.42%
    rewriter_before_opt_a                :     0.000072s :  0.90%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000059s :  0.74%
    jit_opt_a.loop_unroll                :     0.000036s :  0.45%
    jit_opt_a.a_1                        :     0.000642s :  7.98%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.21%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000609s :  7.57%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.24%
    jit_opt_a.cse                        :     0.000042s :  0.53%
    jit_opt_a.replace_applicator         :     0.000017s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.14%
    rewriter_after_opt_a                 :     0.000033s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000533s :  6.63%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000024s :  0.30%
    loop_unroll                          :     0.000429s :  5.32%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000046s :  0.57%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000014s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000447s :  5.56%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000033s :  0.41%


Time group info:
------[substitution.]   0.000203    23
 1.13% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.71% :     0.000006s :      3: substitution.graph_param_transform
83.72% :     0.000170s :      4: substitution.inline
 1.77% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.28% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.61% :     0.000003s :      2: substitution.replace_old_param
 6.14% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004124     2
80.22% :     0.003308s :      1: type_inference.infer
19.78% :     0.000816s :      1: type_inference.specialize
------[replace.]   0.000068     6
67.48% :     0.000046s :      4: replace.inline
32.52% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000179     6
93.66% :     0.000167s :      4: match.inline
 6.34% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.64% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.50% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000002s :     10: predicate.addn_zero_filter
 2.33% :     0.000003s :     10: predicate.arithmetic_simplify
 1.22% :     0.000002s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000002s :     10: predicate.depend_value_elim
 1.80% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.63% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.23% :     0.000002s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.45% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.00% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     15: predicate.partial_eliminate
 1.04% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.85% :     0.000002s :     10: predicate.reduce_eliminate
 1.68% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.58% :     0.000002s :     10: predicate.reshape_eliminate
 1.08% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.43% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.40% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.33% :     0.000009s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.07% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.85% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000583    12
49.38% :     0.000288s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.91% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.71% :     0.000261s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013377    72
 0.37% :     0.000049s :      1: add_recomputation
 0.52% :     0.000069s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.05% :     0.000275s :      1: bootstrap
 0.20% :     0.000027s :      1: cconv
 0.09% :     0.000012s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.42% :     0.005140s :      1: jit_opt_a
 1.08% :     0.000145s :      1: jit_opt_after_cconv
 0.41% :     0.000054s :      1: jit_opt_b
 3.26% :     0.000436s :      1: loop_unroll
 4.05% :     0.000542s :      1: mutable_eliminate
 6.40% :     0.000856s :     26: opt.transform.jit_opt_a
 0.38% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.41% :     0.000456s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.04% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000036s :      1: py_interpret_to_execute
 0.10% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.35% :     0.000314s :      1: renormalize.infer
 2.15% :     0.000288s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000035s :      1: rewriter_after_opt_a
 0.56% :     0.000075s :      1: rewriter_before_opt_a
 0.57% :     0.000076s :      1: symbol_engine_optimizer
31.38% :     0.004198s :      1: type_inference


TotalTime = 0.0113897, [30]
[bootstrap]: 0.00029275
[type_inference]: 0.00389934
[event_method]: 1.681e-05
[auto_monad]: 6.105e-05
[graph_reusing]: 6.24001e-06
[pre_auto_parallel]: 2.27999e-06
[py_interpret_to_execute]: 2.963e-05
[rewriter_before_opt_a]: 6.897e-05
[expand_dump_flag]: 2.84001e-06
[jit_opt_a]: 0.00504097, [2]
    [Cycle 1]: 0.00156345, [27]
        [switch_simplify]: 5.286e-05
        [loop_unroll]: 3.005e-05
        [a_1]: 0.00055919
        [with_stream_mark]: 1.727e-05
        [recompute_prepare]: 8.75999e-06
        [updatestate_depend_eliminate]: 3.71999e-06
        [updatestate_assign_eliminate]: 3.17002e-06
        [updatestate_loads_eliminate]: 2.94001e-06
        [parameter_eliminate]: 2.04999e-06
        [specialize_transform]: 6.41e-06
        [updatestate_useless_node_eliminater]: 5.25001e-06
        [accelerated_algorithm]: 6.71999e-06
        [meta_shard_fg_expand]: 2.11e-06
        [get_grad_eliminate_]: 5.73997e-06
        [merge_forward]: 3.53e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.469e-05
        [j_node_and_user_rematch]: 9.77999e-06
        [meta_fg_expand]: 2.25002e-06
        [replace_old_param]: 8.97e-06
        [inline_without_move]: 5.67999e-06
        [renormalize]: 0.00058555
        [add_forward_monad_depend]: 5.11002e-06
        [auto_monad_grad]: 2.18002e-06
        [auto_monad_eliminator]: 1.282e-05
        [cse]: 2.374e-05
        [replace_applicator]: 1.094e-05
    [Cycle 2]: 0.0003247, [27]
        [switch_simplify]: 5.68002e-06
        [loop_unroll]: 5.09003e-06
        [a_1]: 9.3e-05
        [with_stream_mark]: 8.50999e-06
        [recompute_prepare]: 5.04e-06
        [updatestate_depend_eliminate]: 2.49001e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 5.19e-06
        [updatestate_useless_node_eliminater]: 5.12999e-06
        [accelerated_algorithm]: 5.52999e-06
        [meta_shard_fg_expand]: 1.15999e-06
        [get_grad_eliminate_]: 4.93001e-06
        [merge_forward]: 2.49001e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.197e-05
        [j_node_and_user_rematch]: 8.30999e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 7.38e-06
        [inline_without_move]: 4.92e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 4.74998e-06
        [cse]: 1.041e-05
        [replace_applicator]: 5.32001e-06
[py_interpret_to_execute_after_opt_a]: 9.46998e-06
[rewriter_after_opt_a]: 3.119e-05
[convert_after_rewriter]: 6.52001e-06
[order_py_execute_after_rewriter]: 5.61e-06
[mutable_eliminate]: 0.00047586
[jit_opt_b]: 5.202e-05, [1]
    [Cycle 1]: 4.51e-05, [2]
        [frontend_op_eliminate]: 1.74e-05
        [inline_after_opt_a]: 1.574e-05
[cconv]: 2.387e-05
[loop_unroll]: 0.00041395
[jit_opt_after_cconv]: 0.00014122, [1]
    [Cycle 1]: 0.00013417, [11]
        [c_1]: 2.125e-05
        [parameter_eliminate]: 2.53003e-06
        [updatestate_depend_eliminate]: 5.85002e-06
        [updatestate_assign_eliminate]: 2.20002e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 1.798e-05
        [call_graph_tuple_transform]: 1.81e-05
        [tuple_list_get_item_eliminator]: 5.61e-06
        [none_parameter_eliminate]: 1.70001e-06
        [renormalize]: 3.10014e-07
        [switch_simplify]: 5.47999e-06
[remove_dup_value]: 1.501e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 4.94e-06
[add_recomputation]: 4.439e-05
[cse_after_recomputation]: 2.012e-05, [1]
    [Cycle 1]: 1.469e-05, [1]
        [cse]: 9.07001e-06
[auto_monad_reorder]: 1.414e-05
[get_jit_bprop_graph]: 1.52001e-06
[rewriter_after_jit_bprop_graph]: 4.99e-06
[opt_after_jit_grad]: 0.00044319
[symbol_engine_optimizer]: 7.256e-05, [1]
    [Cycle 1]: 6.564e-05, [6]
        [build]: 3.24001e-06
        [elim_shapecalc]: 7.45e-06
        [elim_not_effective]: 1.263e-05
        [opt_reshape]: 5.93002e-06
        [fold_const_symbol]: 9.04e-06
        [renormalize]: 3.7998e-07
[validate]: 3.134e-05
Sums
    bootstrap                            :     0.000293s :  3.82%
    type_inference                       :     0.003899s : 50.83%
    event_method                         :     0.000017s :  0.22%
    auto_monad                           :     0.000061s :  0.80%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000030s :  0.39%
    rewriter_before_opt_a                :     0.000069s :  0.90%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000059s :  0.76%
    jit_opt_a.loop_unroll                :     0.000035s :  0.46%
    jit_opt_a.a_1                        :     0.000652s :  8.50%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.34%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.18%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.16%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.14%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.24%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.21%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000586s :  7.63%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.23%
    jit_opt_a.cse                        :     0.000034s :  0.45%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.12%
    rewriter_after_opt_a                 :     0.000031s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.07%
    mutable_eliminate                    :     0.000476s :  6.20%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.21%
    cconv                                :     0.000024s :  0.31%
    loop_unroll                          :     0.000414s :  5.40%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000018s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000015s :  0.20%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000044s :  0.58%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000014s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000443s :  5.78%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.41%


Time group info:
------[substitution.]   0.000194    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.50% :     0.000005s :      3: substitution.graph_param_transform
84.30% :     0.000164s :      4: substitution.inline
 1.64% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.14% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.83% :     0.000004s :      2: substitution.replace_old_param
 5.94% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003835     2
80.19% :     0.003075s :      1: type_inference.infer
19.81% :     0.000760s :      1: type_inference.specialize
------[replace.]   0.000066     6
67.52% :     0.000045s :      4: replace.inline
32.48% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000171     6
93.96% :     0.000161s :      4: match.inline
 6.04% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.22% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000002s :     10: predicate.addn_zero_filter
 2.42% :     0.000003s :     10: predicate.arithmetic_simplify
 1.28% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.20% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      6: predicate.less_batch_normalization
 1.73% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.10% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000002s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.66% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.28% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 2.23% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.46% :     0.000009s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.76% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.83% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000534    12
50.29% :     0.000269s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.97% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.73% :     0.000234s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012903    72
 0.37% :     0.000047s :      1: add_recomputation
 0.50% :     0.000064s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.37% :     0.000306s :      1: bootstrap
 0.21% :     0.000027s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.07% :     0.000008s :      1: graph_reusing
39.09% :     0.005044s :      1: jit_opt_a
 1.12% :     0.000144s :      1: jit_opt_after_cconv
 0.43% :     0.000055s :      1: jit_opt_b
 3.27% :     0.000422s :      1: loop_unroll
 3.75% :     0.000484s :      1: mutable_eliminate
 6.69% :     0.000864s :     26: opt.transform.jit_opt_a
 0.36% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.50% :     0.000452s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000032s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.14% :     0.000018s :      1: remove_dup_value
 2.27% :     0.000293s :      1: renormalize.infer
 2.20% :     0.000284s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000034s :      1: rewriter_after_opt_a
 0.56% :     0.000072s :      1: rewriter_before_opt_a
 0.58% :     0.000075s :      1: symbol_engine_optimizer
30.30% :     0.003910s :      1: type_inference


TotalTime = 0.0165932, [33]
[bootstrap]: 0.00028826
[type_inference]: 0.00396341
[event_method]: 1.573e-05
[auto_monad]: 5.555e-05
[graph_reusing]: 4.72e-06
[pre_auto_parallel]: 2.09999e-06
[py_interpret_to_execute]: 2.778e-05
[rewriter_before_opt_a]: 7.1e-05
[expand_dump_flag]: 2.54999e-06
[jit_opt_a]: 0.00457058, [2]
    [Cycle 1]: 0.00127109, [27]
        [switch_simplify]: 4.204e-05
        [loop_unroll]: 2.88e-05
        [a_1]: 0.0004839
        [with_stream_mark]: 1.35e-05
        [recompute_prepare]: 7.35003e-06
        [updatestate_depend_eliminate]: 3.21999e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 5.77001e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 1.65001e-06
        [get_grad_eliminate_]: 5.28002e-06
        [merge_forward]: 2.88e-06
        [cell_reuse_recompute_pass]: 8.89995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.149e-05
        [j_node_and_user_rematch]: 8.60999e-06
        [meta_fg_expand]: 1.71998e-06
        [replace_old_param]: 7.87e-06
        [inline_without_move]: 5.02999e-06
        [renormalize]: 0.00042579
        [add_forward_monad_depend]: 4.03001e-06
        [auto_monad_grad]: 1.47999e-06
        [auto_monad_eliminator]: 1.026e-05
        [cse]: 1.561e-05
        [replace_applicator]: 1.108e-05
    [Cycle 2]: 0.00030449, [27]
        [switch_simplify]: 5.87999e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 8.966e-05
        [with_stream_mark]: 7.68001e-06
        [recompute_prepare]: 4.80999e-06
        [updatestate_depend_eliminate]: 2.25002e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.72e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 4.60001e-06
        [meta_shard_fg_expand]: 1.19998e-06
        [get_grad_eliminate_]: 4.27e-06
        [merge_forward]: 2.62001e-06
        [cell_reuse_recompute_pass]: 1.26997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.151e-05
        [j_node_and_user_rematch]: 7.6e-06
        [meta_fg_expand]: 1.42999e-06
        [replace_old_param]: 6.66e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.04e-06
        [auto_monad_grad]: 6.19999e-07
        [auto_monad_eliminator]: 4.62e-06
        [cse]: 9.56998e-06
        [replace_applicator]: 4.72998e-06
[py_interpret_to_execute_after_opt_a]: 7.82e-06
[rewriter_after_opt_a]: 2.679e-05
[convert_after_rewriter]: 6.39001e-06
[order_py_execute_after_rewriter]: 4.65001e-06
[mutable_eliminate]: 0.0004483
[jit_opt_b]: 4.526e-05, [1]
    [Cycle 1]: 3.944e-05, [2]
        [frontend_op_eliminate]: 1.461e-05
        [inline_after_opt_a]: 1.448e-05
[cconv]: 1.929e-05
[loop_unroll]: 0.00039113
[jit_opt_after_cconv]: 0.00013073, [1]
    [Cycle 1]: 0.00012445, [11]
        [c_1]: 2.023e-05
        [parameter_eliminate]: 2.47001e-06
        [updatestate_depend_eliminate]: 5.18002e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [cse]: 1.569e-05
        [call_graph_tuple_transform]: 1.679e-05
        [tuple_list_get_item_eliminator]: 4.99e-06
        [none_parameter_eliminate]: 1.05001e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.15001e-06
[remove_dup_value]: 9.59e-06
[partial_unused_args_eliminate]: 1.98002e-06
[environ_conv]: 3.33e-06
[add_recomputation]: 3.451e-05
[cse_after_recomputation]: 1.958e-05, [1]
    [Cycle 1]: 1.471e-05, [1]
        [cse]: 9.44e-06
[auto_monad_reorder]: 1.094e-05
[get_jit_bprop_graph]: 1.57999e-06
[rewriter_after_jit_bprop_graph]: 4.48999e-06
[opt_after_jit_grad]: 0.00042157
[symbol_engine_optimizer]: 6.711e-05, [1]
    [Cycle 1]: 6.146e-05, [6]
        [build]: 2.32001e-06
        [elim_shapecalc]: 6.94999e-06
        [elim_not_effective]: 1.203e-05
        [opt_reshape]: 5.30999e-06
        [fold_const_symbol]: 8.00999e-06
        [renormalize]: 4.2998e-07
[validate]: 2.704e-05
[backend_pass]: 6.79982e-07
[task_emit]: 0.00573836
[execute]: 6.25002e-06
Sums
    bootstrap                            :     0.000288s :  2.21%
    type_inference                       :     0.003963s : 30.36%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000056s :  0.43%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.21%
    rewriter_before_opt_a                :     0.000071s :  0.54%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000048s :  0.37%
    jit_opt_a.loop_unroll                :     0.000034s :  0.26%
    jit_opt_a.a_1                        :     0.000574s :  4.39%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000426s :  3.26%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000025s :  0.19%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000027s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000448s :  3.43%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000019s :  0.15%
    loop_unroll                          :     0.000391s :  3.00%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000010s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000003s :  0.03%
    add_recomputation                    :     0.000035s :  0.26%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000011s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000422s :  3.23%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005738s : 43.95%
    execute                              :     0.000006s :  0.05%


Time group info:
------[substitution.]   0.000149    23
 1.27% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.87% :     0.000004s :      3: substitution.graph_param_transform
82.76% :     0.000123s :      4: substitution.inline
 1.77% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.58% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000003s :      2: substitution.replace_old_param
 6.26% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003914     2
81.02% :     0.003171s :      1: type_inference.infer
18.98% :     0.000743s :      1: type_inference.specialize
------[replace.]   0.000059     6
64.61% :     0.000038s :      4: replace.inline
35.39% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000129     6
93.63% :     0.000121s :      4: match.inline
 6.37% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.54% :     0.000002s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.37% :     0.000002s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.42% :     0.000002s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.32% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.34% :     0.000001s :     10: predicate.environ_get_eliminate
 1.35% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.26% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.23% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.19% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.66% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000002s :     15: predicate.partial_eliminate
 1.33% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.56% :     0.000002s :     10: predicate.reshape_eliminate
 1.42% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.44% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.72% :     0.000003s :     16: predicate.switch_defer_inline
 2.30% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000008s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.25% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.23% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.04% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000475    12
49.10% :     0.000233s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.48% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.42% :     0.000216s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017885    76
 0.21% :     0.000037s :      1: add_recomputation
 0.33% :     0.000059s :      1: auto_monad
 0.08% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.67% :     0.000299s :      1: bootstrap
 0.12% :     0.000022s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
25.57% :     0.004573s :      1: jit_opt_a
 0.75% :     0.000133s :      1: jit_opt_after_cconv
 0.27% :     0.000048s :      1: jit_opt_b
 2.23% :     0.000398s :      1: loop_unroll
 2.55% :     0.000456s :      1: mutable_eliminate
 4.26% :     0.000763s :     26: opt.transform.jit_opt_a
 0.24% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.40% :     0.000429s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000012s :      1: remove_dup_value
 1.20% :     0.000215s :      1: renormalize.infer
 1.14% :     0.000204s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000030s :      1: rewriter_after_opt_a
 0.41% :     0.000074s :      1: rewriter_before_opt_a
 0.39% :     0.000070s :      1: symbol_engine_optimizer
32.14% :     0.005749s :      1: task_emit
22.22% :     0.003974s :      1: type_inference
 0.23% :     0.000042s :      1: validate


TotalTime = 0.0309931, [33]
[bootstrap]: 0.00041001
[type_inference]: 0.00720721
[event_method]: 2.152e-05
[auto_monad]: 7.105e-05
[graph_reusing]: 6.38e-06
[pre_auto_parallel]: 2.93e-06
[py_interpret_to_execute]: 4.071e-05
[rewriter_before_opt_a]: 7.616e-05
[expand_dump_flag]: 3.22002e-06
[jit_opt_a]: 0.00850724, [2]
    [Cycle 1]: 0.00205638, [27]
        [switch_simplify]: 6.955e-05
        [loop_unroll]: 2.974e-05
        [a_1]: 0.00063979
        [with_stream_mark]: 2.845e-05
        [recompute_prepare]: 7.77998e-06
        [updatestate_depend_eliminate]: 5.02e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.84999e-06
        [parameter_eliminate]: 2.13998e-06
        [specialize_transform]: 6.74999e-06
        [updatestate_useless_node_eliminater]: 5.16998e-06
        [accelerated_algorithm]: 5.58997e-06
        [meta_shard_fg_expand]: 2.56998e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 3.74002e-06
        [cell_reuse_recompute_pass]: 1.56002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.48e-05
        [j_node_and_user_rematch]: 9.82001e-06
        [meta_fg_expand]: 2.42001e-06
        [replace_old_param]: 9.69e-06
        [inline_without_move]: 5.26998e-06
        [renormalize]: 0.00089547
        [add_forward_monad_depend]: 9.42001e-06
        [auto_monad_grad]: 3.97e-06
        [auto_monad_eliminator]: 1.933e-05
        [cse]: 3.489e-05
        [replace_applicator]: 2.005e-05
    [Cycle 2]: 0.0003505, [27]
        [switch_simplify]: 6.04999e-06
        [loop_unroll]: 5.10999e-06
        [a_1]: 9.329e-05
        [with_stream_mark]: 1.584e-05
        [recompute_prepare]: 4.85001e-06
        [updatestate_depend_eliminate]: 3.20002e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [parameter_eliminate]: 1.76e-06
        [specialize_transform]: 5.40001e-06
        [updatestate_useless_node_eliminater]: 4.79998e-06
        [accelerated_algorithm]: 5.26998e-06
        [meta_shard_fg_expand]: 1.91998e-06
        [get_grad_eliminate_]: 4.62998e-06
        [merge_forward]: 3.97e-06
        [cell_reuse_recompute_pass]: 3.45e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.701e-05
        [j_node_and_user_rematch]: 8.16002e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 1.394e-05
        [inline_without_move]: 5.12999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.76e-06
        [auto_monad_grad]: 1.09e-06
        [auto_monad_eliminator]: 6.21998e-06
        [cse]: 1.217e-05
        [replace_applicator]: 4.94003e-06
[py_interpret_to_execute_after_opt_a]: 1.702e-05
[rewriter_after_opt_a]: 4.355e-05
[convert_after_rewriter]: 7.95e-06
[order_py_execute_after_rewriter]: 6.20997e-06
[mutable_eliminate]: 0.00088895
[jit_opt_b]: 5.59e-05, [1]
    [Cycle 1]: 4.689e-05, [2]
        [frontend_op_eliminate]: 1.779e-05
        [inline_after_opt_a]: 1.712e-05
[cconv]: 3.897e-05
[loop_unroll]: 0.00047183
[jit_opt_after_cconv]: 0.00017213, [1]
    [Cycle 1]: 0.00016495, [11]
        [c_1]: 2.078e-05
        [parameter_eliminate]: 4.66002e-06
        [updatestate_depend_eliminate]: 1.117e-05
        [updatestate_assign_eliminate]: 2.70002e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [cse]: 3.61e-05
        [call_graph_tuple_transform]: 2.271e-05
        [tuple_list_get_item_eliminator]: 5.78002e-06
        [none_parameter_eliminate]: 1.55001e-06
        [renormalize]: 8.2e-07
        [switch_simplify]: 5.90002e-06
[remove_dup_value]: 1.532e-05
[partial_unused_args_eliminate]: 2.24999e-06
[environ_conv]: 5.69e-06
[add_recomputation]: 6.204e-05
[cse_after_recomputation]: 2.152e-05, [1]
    [Cycle 1]: 1.563e-05, [1]
        [cse]: 9.42001e-06
[auto_monad_reorder]: 1.67e-05
[get_jit_bprop_graph]: 2.48e-06
[rewriter_after_jit_bprop_graph]: 1.032e-05
[opt_after_jit_grad]: 0.00049761
[symbol_engine_optimizer]: 7.669e-05, [1]
    [Cycle 1]: 6.876e-05, [6]
        [build]: 4.40999e-06
        [elim_shapecalc]: 7.42002e-06
        [elim_not_effective]: 1.398e-05
        [opt_reshape]: 5.54e-06
        [fold_const_symbol]: 9.05999e-06
        [renormalize]: 1.05001e-06
[validate]: 4.429e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.0119186
[execute]: 1.006e-05
Sums
    bootstrap                            :     0.000410s :  1.70%
    type_inference                       :     0.007207s : 29.81%
    event_method                         :     0.000022s :  0.09%
    auto_monad                           :     0.000071s :  0.29%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.17%
    rewriter_before_opt_a                :     0.000076s :  0.31%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000076s :  0.31%
    jit_opt_a.loop_unroll                :     0.000035s :  0.14%
    jit_opt_a.a_1                        :     0.000733s :  3.03%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000008s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000024s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.04%
    jit_opt_a.renormalize                :     0.000896s :  3.70%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.11%
    jit_opt_a.cse                        :     0.000047s :  0.19%
    jit_opt_a.replace_applicator         :     0.000025s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.07%
    rewriter_after_opt_a                 :     0.000044s :  0.18%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000889s :  3.68%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.07%
    cconv                                :     0.000039s :  0.16%
    loop_unroll                          :     0.000472s :  1.95%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000036s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000015s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000062s :  0.26%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000017s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.04%
    opt_after_jit_grad                   :     0.000498s :  2.06%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.18%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011919s : 49.29%
    execute                              :     0.000010s :  0.04%


Time group info:
------[substitution.]   0.000293    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 2.10% :     0.000006s :      3: substitution.graph_param_transform
85.91% :     0.000251s :      4: substitution.inline
 1.46% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.91% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.33% :     0.000007s :      2: substitution.replace_old_param
 5.07% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.007120     2
86.92% :     0.006189s :      1: type_inference.infer
13.08% :     0.000931s :      1: type_inference.specialize
------[replace.]   0.000078     6
69.68% :     0.000054s :      4: replace.inline
30.32% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000262     6
94.79% :     0.000249s :      4: match.inline
 5.21% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.30% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.07% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 1.99% :     0.000002s :     10: predicate.arithmetic_simplify
 1.50% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.78% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.06% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.13% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 6.68% :     0.000008s :     22: predicate.inline
 1.02% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 1.79% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.58% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 3.56% :     0.000004s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.03% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.88% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.61% :     0.000001s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.39% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.21% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.20% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.07% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.16% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.04% :     0.000010s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.25% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.42% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.57% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.64% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000767    12
52.76% :     0.000405s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.96% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.28% :     0.000301s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.032973    76
 0.20% :     0.000065s :      1: add_recomputation
 0.23% :     0.000075s :      1: auto_monad
 0.06% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.34% :     0.000443s :      1: bootstrap
 0.13% :     0.000042s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.08% :     0.000026s :      1: event_method
 0.05% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
25.81% :     0.008511s :      1: jit_opt_a
 0.53% :     0.000175s :      1: jit_opt_after_cconv
 0.18% :     0.000059s :      1: jit_opt_b
 1.46% :     0.000481s :      1: loop_unroll
 2.74% :     0.000902s :      1: mutable_eliminate
 2.96% :     0.000975s :     26: opt.transform.jit_opt_a
 0.16% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000028s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.54% :     0.000507s :      1: opt_after_jit_grad
 0.03% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000044s :      1: py_interpret_to_execute
 0.06% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 1.51% :     0.000499s :      1: renormalize.infer
 1.17% :     0.000387s :      1: renormalize.specialize
 0.04% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000047s :      1: rewriter_after_opt_a
 0.24% :     0.000080s :      1: rewriter_before_opt_a
 0.24% :     0.000079s :      1: symbol_engine_optimizer
36.21% :     0.011940s :      1: task_emit
21.93% :     0.007232s :      1: type_inference
 0.20% :     0.000067s :      1: validate


TotalTime = 0.0171958, [33]
[bootstrap]: 0.00025963
[type_inference]: 0.00420238
[event_method]: 1.792e-05
[auto_monad]: 6.066e-05
[graph_reusing]: 6.00002e-06
[pre_auto_parallel]: 2.76999e-06
[py_interpret_to_execute]: 3.237e-05
[rewriter_before_opt_a]: 7.118e-05
[expand_dump_flag]: 3.23e-06
[jit_opt_a]: 0.00492053, [2]
    [Cycle 1]: 0.0014227, [27]
        [switch_simplify]: 5.019e-05
        [loop_unroll]: 2.89e-05
        [a_1]: 0.00049089
        [with_stream_mark]: 1.711e-05
        [recompute_prepare]: 6.81999e-06
        [updatestate_depend_eliminate]: 3.56999e-06
        [updatestate_assign_eliminate]: 3.04001e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [parameter_eliminate]: 2.27001e-06
        [specialize_transform]: 5.66e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 2.14e-06
        [get_grad_eliminate_]: 5.14e-06
        [merge_forward]: 3.53e-06
        [cell_reuse_recompute_pass]: 9.79984e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.309e-05
        [j_node_and_user_rematch]: 8.60999e-06
        [meta_fg_expand]: 2.02999e-06
        [replace_old_param]: 8.67998e-06
        [inline_without_move]: 5.25001e-06
        [renormalize]: 0.00052358
        [add_forward_monad_depend]: 5.10001e-06
        [auto_monad_grad]: 2.19001e-06
        [auto_monad_eliminator]: 1.318e-05
        [cse]: 2.811e-05
        [replace_applicator]: 1.139e-05
    [Cycle 2]: 0.00030528, [27]
        [switch_simplify]: 5.39998e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 8.642e-05
        [with_stream_mark]: 8.32e-06
        [recompute_prepare]: 5.09998e-06
        [updatestate_depend_eliminate]: 2.36e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 4.95001e-06
        [updatestate_useless_node_eliminater]: 4.56002e-06
        [accelerated_algorithm]: 4.90001e-06
        [meta_shard_fg_expand]: 9.89996e-07
        [get_grad_eliminate_]: 4.51002e-06
        [merge_forward]: 2.31e-06
        [cell_reuse_recompute_pass]: 1.51998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.183e-05
        [j_node_and_user_rematch]: 7.69002e-06
        [meta_fg_expand]: 1.43002e-06
        [replace_old_param]: 7.48e-06
        [inline_without_move]: 4.53001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 4.87e-06
        [cse]: 1.015e-05
        [replace_applicator]: 4.80999e-06
[py_interpret_to_execute_after_opt_a]: 1.366e-05
[rewriter_after_opt_a]: 3.106e-05
[convert_after_rewriter]: 7.01001e-06
[order_py_execute_after_rewriter]: 5.18002e-06
[mutable_eliminate]: 0.00045705
[jit_opt_b]: 4.807e-05, [1]
    [Cycle 1]: 4.136e-05, [2]
        [frontend_op_eliminate]: 1.557e-05
        [inline_after_opt_a]: 1.455e-05
[cconv]: 2.625e-05
[loop_unroll]: 0.0004162
[jit_opt_after_cconv]: 0.00013204, [1]
    [Cycle 1]: 0.00012593, [11]
        [c_1]: 1.905e-05
        [parameter_eliminate]: 2.58003e-06
        [updatestate_depend_eliminate]: 5.22999e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 1.636e-05
        [call_graph_tuple_transform]: 1.79e-05
        [tuple_list_get_item_eliminator]: 5.22e-06
        [none_parameter_eliminate]: 1.35999e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 5.19e-06
[remove_dup_value]: 1.336e-05
[partial_unused_args_eliminate]: 2.17001e-06
[environ_conv]: 4.79998e-06
[add_recomputation]: 4.82e-05
[cse_after_recomputation]: 1.926e-05, [1]
    [Cycle 1]: 1.404e-05, [1]
        [cse]: 8.45999e-06
[auto_monad_reorder]: 1.449e-05
[get_jit_bprop_graph]: 1.94e-06
[rewriter_after_jit_bprop_graph]: 4.85999e-06
[opt_after_jit_grad]: 0.00043542
[symbol_engine_optimizer]: 6.977e-05, [1]
    [Cycle 1]: 6.334e-05, [6]
        [build]: 3.16001e-06
        [elim_shapecalc]: 7.18e-06
        [elim_not_effective]: 1.245e-05
        [opt_reshape]: 5.42999e-06
        [fold_const_symbol]: 8.08999e-06
        [renormalize]: 4.89992e-07
[validate]: 3.15e-05
[backend_pass]: 9.99979e-07
[task_emit]: 0.00564936
[execute]: 6.78e-06
Sums
    bootstrap                            :     0.000260s :  1.93%
    type_inference                       :     0.004202s : 31.29%
    event_method                         :     0.000018s :  0.13%
    auto_monad                           :     0.000061s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.24%
    rewriter_before_opt_a                :     0.000071s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000577s :  4.30%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000524s :  3.90%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000038s :  0.28%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.10%
    rewriter_after_opt_a                 :     0.000031s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000457s :  3.40%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000026s :  0.20%
    loop_unroll                          :     0.000416s :  3.10%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000048s :  0.36%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000435s :  3.24%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005649s : 42.07%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000180    23
 1.14% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000001s :      2: substitution.fold_const_symbol
 2.97% :     0.000005s :      3: substitution.graph_param_transform
82.62% :     0.000149s :      4: substitution.inline
 1.60% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.25% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.32% :     0.000004s :      2: substitution.replace_old_param
 6.36% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004143     2
81.53% :     0.003378s :      1: type_inference.infer
18.47% :     0.000765s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.34% :     0.000042s :      4: replace.inline
32.66% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000157     6
93.37% :     0.000146s :      4: match.inline
 6.63% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.29% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000001s :     10: predicate.addn_zero_filter
 1.73% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000001s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.96% :     0.000007s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.96% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.23% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.49% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000002s :     15: predicate.partial_eliminate
 1.38% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     10: predicate.reduce_eliminate
 1.56% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.77% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.49% :     0.000002s :     10: predicate.same_eliminate
 0.63% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.67% :     0.000008s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.39% :     0.000002s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.98% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000523    12
50.99% :     0.000267s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.64% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.37% :     0.000227s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018596    76
 0.27% :     0.000051s :      1: add_recomputation
 0.34% :     0.000064s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.49% :     0.000277s :      1: bootstrap
 0.15% :     0.000029s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.12% :     0.000022s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.47% :     0.004923s :      1: jit_opt_a
 0.73% :     0.000135s :      1: jit_opt_after_cconv
 0.27% :     0.000051s :      1: jit_opt_b
 2.28% :     0.000424s :      1: loop_unroll
 2.50% :     0.000465s :      1: mutable_eliminate
 4.17% :     0.000776s :     26: opt.transform.jit_opt_a
 0.24% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.38% :     0.000443s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000035s :      1: py_interpret_to_execute
 0.09% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.45% :     0.000270s :      1: renormalize.infer
 1.33% :     0.000246s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.40% :     0.000074s :      1: rewriter_before_opt_a
 0.39% :     0.000072s :      1: symbol_engine_optimizer
30.43% :     0.005659s :      1: task_emit
22.66% :     0.004214s :      1: type_inference
 0.27% :     0.000051s :      1: validate


TotalTime = 0.0114033, [30]
[bootstrap]: 0.00027119
[type_inference]: 0.00413481
[event_method]: 1.593e-05
[auto_monad]: 5.637e-05
[graph_reusing]: 5.62999e-06
[pre_auto_parallel]: 2.25002e-06
[py_interpret_to_execute]: 3.649e-05
[rewriter_before_opt_a]: 9.028e-05
[expand_dump_flag]: 2.41998e-06
[jit_opt_a]: 0.00484772, [2]
    [Cycle 1]: 0.00142456, [27]
        [switch_simplify]: 4.844e-05
        [loop_unroll]: 2.999e-05
        [a_1]: 0.00049093
        [with_stream_mark]: 1.364e-05
        [recompute_prepare]: 7.13e-06
        [updatestate_depend_eliminate]: 3.25e-06
        [updatestate_assign_eliminate]: 2.58998e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [parameter_eliminate]: 1.39998e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 5.09003e-06
        [accelerated_algorithm]: 5.67999e-06
        [meta_shard_fg_expand]: 1.60999e-06
        [get_grad_eliminate_]: 5.17e-06
        [merge_forward]: 3.03e-06
        [cell_reuse_recompute_pass]: 9.90025e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.202e-05
        [j_node_and_user_rematch]: 8.84998e-06
        [meta_fg_expand]: 2.76e-06
        [replace_old_param]: 8.22e-06
        [inline_without_move]: 5.28002e-06
        [renormalize]: 0.00054945
        [add_forward_monad_depend]: 4.32e-06
        [auto_monad_grad]: 1.99999e-06
        [auto_monad_eliminator]: 1.233e-05
        [cse]: 1.864e-05
        [replace_applicator]: 1.145e-05
    [Cycle 2]: 0.00032025, [27]
        [switch_simplify]: 5.76e-06
        [loop_unroll]: 4.85999e-06
        [a_1]: 8.917e-05
        [with_stream_mark]: 8.59e-06
        [recompute_prepare]: 5.30999e-06
        [updatestate_depend_eliminate]: 2.41e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 5.12e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.10999e-06
        [meta_shard_fg_expand]: 1.45001e-06
        [get_grad_eliminate_]: 4.72998e-06
        [merge_forward]: 2.61999e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.207e-05
        [j_node_and_user_rematch]: 7.63001e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 7.15e-06
        [inline_without_move]: 4.55999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.08001e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 5.15001e-06
        [cse]: 1.041e-05
        [replace_applicator]: 5.19e-06
[py_interpret_to_execute_after_opt_a]: 8.47e-06
[rewriter_after_opt_a]: 3.076e-05
[convert_after_rewriter]: 6.09001e-06
[order_py_execute_after_rewriter]: 5.09998e-06
[mutable_eliminate]: 0.00047379
[jit_opt_b]: 5.013e-05, [1]
    [Cycle 1]: 4.352e-05, [2]
        [frontend_op_eliminate]: 1.714e-05
        [inline_after_opt_a]: 1.506e-05
[cconv]: 2.294e-05
[loop_unroll]: 0.00040301
[jit_opt_after_cconv]: 0.000138, [1]
    [Cycle 1]: 0.00013165, [11]
        [c_1]: 2.069e-05
        [parameter_eliminate]: 2.51e-06
        [updatestate_depend_eliminate]: 4.85001e-06
        [updatestate_assign_eliminate]: 2.19999e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [cse]: 1.712e-05
        [call_graph_tuple_transform]: 1.786e-05
        [tuple_list_get_item_eliminator]: 5.37999e-06
        [none_parameter_eliminate]: 1.49998e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.40999e-06
[remove_dup_value]: 1.252e-05
[partial_unused_args_eliminate]: 1.82001e-06
[environ_conv]: 4.60001e-06
[add_recomputation]: 4.188e-05
[cse_after_recomputation]: 2.028e-05, [1]
    [Cycle 1]: 1.505e-05, [1]
        [cse]: 9.15999e-06
[auto_monad_reorder]: 1.399e-05
[get_jit_bprop_graph]: 1.35999e-06
[rewriter_after_jit_bprop_graph]: 4.89e-06
[opt_after_jit_grad]: 0.0004448
[symbol_engine_optimizer]: 7.339e-05, [1]
    [Cycle 1]: 6.677e-05, [6]
        [build]: 3.11999e-06
        [elim_shapecalc]: 7.82e-06
        [elim_not_effective]: 1.275e-05
        [opt_reshape]: 5.52001e-06
        [fold_const_symbol]: 9.10999e-06
        [renormalize]: 4.2998e-07
[validate]: 2.974e-05
Sums
    bootstrap                            :     0.000271s :  3.50%
    type_inference                       :     0.004135s : 53.38%
    event_method                         :     0.000016s :  0.21%
    auto_monad                           :     0.000056s :  0.73%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.47%
    rewriter_before_opt_a                :     0.000090s :  1.17%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000054s :  0.70%
    jit_opt_a.loop_unroll                :     0.000035s :  0.45%
    jit_opt_a.a_1                        :     0.000580s :  7.49%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.06%
    jit_opt_a.replace_old_param          :     0.000015s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000550s :  7.09%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.23%
    jit_opt_a.cse                        :     0.000029s :  0.38%
    jit_opt_a.replace_applicator         :     0.000017s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000031s :  0.40%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000474s :  6.12%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000023s :  0.30%
    loop_unroll                          :     0.000403s :  5.20%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000013s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000042s :  0.54%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000014s :  0.18%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000445s :  5.74%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000030s :  0.38%


Time group info:
------[substitution.]   0.000169    23
 1.12% :     0.000002s :      2: substitution.elim_not_effective
 1.00% :     0.000002s :      2: substitution.fold_const_symbol
 2.82% :     0.000005s :      3: substitution.graph_param_transform
82.61% :     0.000140s :      4: substitution.inline
 1.67% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.41% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.91% :     0.000003s :      2: substitution.replace_old_param
 6.46% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004079     2
79.42% :     0.003239s :      1: type_inference.infer
20.58% :     0.000839s :      1: type_inference.specialize
------[replace.]   0.000062     6
65.99% :     0.000041s :      4: replace.inline
34.01% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000147     6
93.35% :     0.000137s :      4: match.inline
 6.65% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.74% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000001s :     10: predicate.addn_zero_filter
 2.04% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000001s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.06% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.73% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.53% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.34% :     0.000002s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.25% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.80% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.81% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     15: predicate.partial_eliminate
 1.31% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.55% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 2.28% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.38% :     0.000009s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.55% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.33% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000590    12
49.24% :     0.000291s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.80% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.95% :     0.000271s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012798    72
 0.35% :     0.000045s :      1: add_recomputation
 0.47% :     0.000060s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.21% :     0.000283s :      1: bootstrap
 0.20% :     0.000025s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.18% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.15% :     0.000019s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
37.90% :     0.004850s :      1: jit_opt_a
 1.10% :     0.000141s :      1: jit_opt_after_cconv
 0.41% :     0.000053s :      1: jit_opt_b
 3.21% :     0.000410s :      1: loop_unroll
 3.76% :     0.000482s :      1: mutable_eliminate
 6.10% :     0.000780s :     26: opt.transform.jit_opt_a
 0.36% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.54% :     0.000453s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.31% :     0.000039s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000015s :      1: remove_dup_value
 2.18% :     0.000279s :      1: renormalize.infer
 2.06% :     0.000264s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000034s :      1: rewriter_after_opt_a
 0.73% :     0.000094s :      1: rewriter_before_opt_a
 0.59% :     0.000076s :      1: symbol_engine_optimizer
32.39% :     0.004145s :      1: type_inference


TotalTime = 0.0356652, [33]
[bootstrap]: 0.00041796
[type_inference]: 0.0189796
[event_method]: 2.034e-05
[auto_monad]: 6.785e-05
[graph_reusing]: 6.06998e-06
[pre_auto_parallel]: 2.91e-06
[py_interpret_to_execute]: 4.454e-05
[rewriter_before_opt_a]: 7.58e-05
[expand_dump_flag]: 2.98998e-06
[jit_opt_a]: 0.00641205, [2]
    [Cycle 1]: 0.00215888, [27]
        [switch_simplify]: 7.052e-05
        [loop_unroll]: 2.959e-05
        [a_1]: 0.00065305
        [with_stream_mark]: 2.444e-05
        [recompute_prepare]: 8.90001e-06
        [updatestate_depend_eliminate]: 3.94002e-06
        [updatestate_assign_eliminate]: 3.33998e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 5.77999e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 5.20001e-06
        [meta_shard_fg_expand]: 2.12999e-06
        [get_grad_eliminate_]: 5.13002e-06
        [merge_forward]: 3.93001e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.672e-05
        [j_node_and_user_rematch]: 9.25999e-06
        [meta_fg_expand]: 2.09999e-06
        [replace_old_param]: 9.71e-06
        [inline_without_move]: 5.29998e-06
        [renormalize]: 0.00102047
        [add_forward_monad_depend]: 6.17999e-06
        [auto_monad_grad]: 2.54999e-06
        [auto_monad_eliminator]: 1.572e-05
        [cse]: 3.163e-05
        [replace_applicator]: 1.526e-05
    [Cycle 2]: 0.00033541, [27]
        [switch_simplify]: 6.05002e-06
        [loop_unroll]: 5.15999e-06
        [a_1]: 9.632e-05
        [with_stream_mark]: 1.151e-05
        [recompute_prepare]: 5.67999e-06
        [updatestate_depend_eliminate]: 3.14999e-06
        [updatestate_assign_eliminate]: 2.11998e-06
        [updatestate_loads_eliminate]: 1.77001e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 5.14998e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.02999e-06
        [meta_shard_fg_expand]: 1.42999e-06
        [get_grad_eliminate_]: 4.35999e-06
        [merge_forward]: 2.98e-06
        [cell_reuse_recompute_pass]: 2.62001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.372e-05
        [j_node_and_user_rematch]: 7.72002e-06
        [meta_fg_expand]: 1.82999e-06
        [replace_old_param]: 7.53e-06
        [inline_without_move]: 4.84e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.50001e-06
        [auto_monad_grad]: 7.79983e-07
        [auto_monad_eliminator]: 5.67001e-06
        [cse]: 1.003e-05
        [replace_applicator]: 5.22999e-06
[py_interpret_to_execute_after_opt_a]: 1.335e-05
[rewriter_after_opt_a]: 3.96e-05
[convert_after_rewriter]: 7.69002e-06
[order_py_execute_after_rewriter]: 5.47001e-06
[mutable_eliminate]: 0.00072597
[jit_opt_b]: 5.798e-05, [1]
    [Cycle 1]: 5.013e-05, [2]
        [frontend_op_eliminate]: 1.763e-05
        [inline_after_opt_a]: 1.61e-05
[cconv]: 2.953e-05
[loop_unroll]: 0.00044815
[jit_opt_after_cconv]: 0.0001488, [1]
    [Cycle 1]: 0.0001425, [11]
        [c_1]: 2.098e-05
        [parameter_eliminate]: 3.18998e-06
        [updatestate_depend_eliminate]: 7.9e-06
        [updatestate_assign_eliminate]: 3.46999e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 2.238e-05
        [call_graph_tuple_transform]: 1.99e-05
        [tuple_list_get_item_eliminator]: 5.13002e-06
        [none_parameter_eliminate]: 1.60001e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 5.52999e-06
[remove_dup_value]: 1.364e-05
[partial_unused_args_eliminate]: 2.43e-06
[environ_conv]: 6.10002e-06
[add_recomputation]: 5.482e-05
[cse_after_recomputation]: 2.05e-05, [1]
    [Cycle 1]: 1.462e-05, [1]
        [cse]: 8.73001e-06
[auto_monad_reorder]: 1.467e-05
[get_jit_bprop_graph]: 2.27999e-06
[rewriter_after_jit_bprop_graph]: 6.86001e-06
[opt_after_jit_grad]: 0.00050162
[symbol_engine_optimizer]: 7.697e-05, [1]
    [Cycle 1]: 6.985e-05, [6]
        [build]: 4.52e-06
        [elim_shapecalc]: 7.17002e-06
        [elim_not_effective]: 1.358e-05
        [opt_reshape]: 6.06e-06
        [fold_const_symbol]: 9.12001e-06
        [renormalize]: 5.69999e-07
[validate]: 4.481e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00715408
[execute]: 9.69e-06
Sums
    bootstrap                            :     0.000418s :  1.35%
    type_inference                       :     0.018980s : 61.11%
    event_method                         :     0.000020s :  0.07%
    auto_monad                           :     0.000068s :  0.22%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000045s :  0.14%
    rewriter_before_opt_a                :     0.000076s :  0.24%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000077s :  0.25%
    jit_opt_a.loop_unroll                :     0.000035s :  0.11%
    jit_opt_a.a_1                        :     0.000749s :  2.41%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000017s :  0.06%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.001021s :  3.29%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.07%
    jit_opt_a.cse                        :     0.000042s :  0.13%
    jit_opt_a.replace_applicator         :     0.000020s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.04%
    rewriter_after_opt_a                 :     0.000040s :  0.13%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000726s :  2.34%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.05%
    cconv                                :     0.000030s :  0.10%
    loop_unroll                          :     0.000448s :  1.44%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000014s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000055s :  0.18%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000015s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000502s :  1.62%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007154s : 23.04%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.000303    23
 0.66% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000002s :      2: substitution.fold_const_symbol
 1.91% :     0.000006s :      3: substitution.graph_param_transform
87.54% :     0.000266s :      4: substitution.inline
 1.18% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.94% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000006s :      2: substitution.replace_old_param
 4.24% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.018897     2
94.64% :     0.017884s :      1: type_inference.infer
 5.36% :     0.001013s :      1: type_inference.specialize
------[replace.]   0.000084     6
68.61% :     0.000058s :      4: replace.inline
31.39% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000275     6
95.75% :     0.000263s :      4: match.inline
 4.25% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.61% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 2.08% :     0.000003s :     10: predicate.arithmetic_simplify
 1.49% :     0.000002s :     10: predicate.cast_eliminate
 0.57% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.64% :     0.000002s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.48% :     0.000007s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.97% :     0.000001s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.96% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.86% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.22% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 2.35% :     0.000003s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 1.99% :     0.000002s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.34% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.45% :     0.000001s :      3: predicate.reset_defer_inline
 1.86% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.75% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.14% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.57% :     0.000009s :     53: predicate.switch_simplify
 1.10% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.04% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.23% :     0.000002s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.013861    12
97.14% :     0.013465s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 0.42% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
 2.44% :     0.000338s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.037767    76
 0.15% :     0.000058s :      1: add_recomputation
 0.19% :     0.000071s :      1: auto_monad
 0.05% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.17% :     0.000441s :      1: bootstrap
 0.09% :     0.000032s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.06% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.07% :     0.000026s :      1: event_method
 0.04% :     0.000015s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
16.99% :     0.006417s :      1: jit_opt_a
 0.40% :     0.000151s :      1: jit_opt_after_cconv
 0.16% :     0.000061s :      1: jit_opt_b
 1.21% :     0.000458s :      1: loop_unroll
 1.95% :     0.000738s :      1: mutable_eliminate
 2.60% :     0.000981s :     26: opt.transform.jit_opt_a
 0.13% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000027s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.36% :     0.000512s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000047s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000016s :      1: remove_dup_value
 1.50% :     0.000565s :      1: renormalize.infer
 1.18% :     0.000445s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000046s :      1: rewriter_after_opt_a
 0.21% :     0.000079s :      1: rewriter_before_opt_a
 0.21% :     0.000080s :      1: symbol_engine_optimizer
18.99% :     0.007173s :      1: task_emit
50.32% :     0.019004s :      1: type_inference
 0.18% :     0.000068s :      1: validate


TotalTime = 0.0171707, [33]
[bootstrap]: 0.00024652
[type_inference]: 0.00414004
[event_method]: 1.584e-05
[auto_monad]: 6.114e-05
[graph_reusing]: 6.32001e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 3.071e-05
[rewriter_before_opt_a]: 7.024e-05
[expand_dump_flag]: 2.81999e-06
[jit_opt_a]: 0.0051077, [2]
    [Cycle 1]: 0.00143993, [27]
        [switch_simplify]: 5.041e-05
        [loop_unroll]: 2.843e-05
        [a_1]: 0.00050167
        [with_stream_mark]: 1.635e-05
        [recompute_prepare]: 7.23e-06
        [updatestate_depend_eliminate]: 3.18e-06
        [updatestate_assign_eliminate]: 2.66999e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 1.50999e-06
        [specialize_transform]: 5.96998e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.82999e-06
        [meta_shard_fg_expand]: 1.77999e-06
        [get_grad_eliminate_]: 4.85001e-06
        [merge_forward]: 3.41999e-06
        [cell_reuse_recompute_pass]: 1.08001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.311e-05
        [j_node_and_user_rematch]: 8.94998e-06
        [meta_fg_expand]: 2.04999e-06
        [replace_old_param]: 8.43001e-06
        [inline_without_move]: 5.04998e-06
        [renormalize]: 0.00054304
        [add_forward_monad_depend]: 4.97e-06
        [auto_monad_grad]: 2.13998e-06
        [auto_monad_eliminator]: 1.288e-05
        [cse]: 2.245e-05
        [replace_applicator]: 1.247e-05
    [Cycle 2]: 0.00032912, [27]
        [switch_simplify]: 5.17999e-06
        [loop_unroll]: 4.74998e-06
        [a_1]: 9.87e-05
        [with_stream_mark]: 9.82001e-06
        [recompute_prepare]: 4.98001e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 4.80999e-06
        [updatestate_useless_node_eliminater]: 4.53999e-06
        [accelerated_algorithm]: 4.62e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 4.35999e-06
        [merge_forward]: 2.29999e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.221e-05
        [j_node_and_user_rematch]: 7.23999e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 7.43e-06
        [inline_without_move]: 4.55999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.40999e-06
        [auto_monad_grad]: 9.60019e-07
        [auto_monad_eliminator]: 5.31002e-06
        [cse]: 1.127e-05
        [replace_applicator]: 5.35999e-06
[py_interpret_to_execute_after_opt_a]: 1.004e-05
[rewriter_after_opt_a]: 3.45e-05
[convert_after_rewriter]: 7.23999e-06
[order_py_execute_after_rewriter]: 5.16002e-06
[mutable_eliminate]: 0.0005214
[jit_opt_b]: 5.127e-05, [1]
    [Cycle 1]: 4.444e-05, [2]
        [frontend_op_eliminate]: 1.858e-05
        [inline_after_opt_a]: 1.45e-05
[cconv]: 2.456e-05
[loop_unroll]: 0.00040456
[jit_opt_after_cconv]: 0.00013173, [1]
    [Cycle 1]: 0.00012511, [11]
        [c_1]: 1.997e-05
        [parameter_eliminate]: 2.21e-06
        [updatestate_depend_eliminate]: 5.09e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [cse]: 1.573e-05
        [call_graph_tuple_transform]: 1.758e-05
        [tuple_list_get_item_eliminator]: 5.20999e-06
        [none_parameter_eliminate]: 1.37e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 5.04e-06
[remove_dup_value]: 1.336e-05
[partial_unused_args_eliminate]: 2.18002e-06
[environ_conv]: 4.77e-06
[add_recomputation]: 4.66e-05
[cse_after_recomputation]: 1.908e-05, [1]
    [Cycle 1]: 1.39e-05, [1]
        [cse]: 8.55999e-06
[auto_monad_reorder]: 1.334e-05
[get_jit_bprop_graph]: 1.68002e-06
[rewriter_after_jit_bprop_graph]: 4.94e-06
[opt_after_jit_grad]: 0.00044349
[symbol_engine_optimizer]: 6.708e-05, [1]
    [Cycle 1]: 6.167e-05, [6]
        [build]: 3.17002e-06
        [elim_shapecalc]: 7.36999e-06
        [elim_not_effective]: 1.131e-05
        [opt_reshape]: 5.25001e-06
        [fold_const_symbol]: 8.05e-06
        [renormalize]: 4.10015e-07
[validate]: 3.28e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00546535
[execute]: 7.58999e-06
Sums
    bootstrap                            :     0.000247s :  1.86%
    type_inference                       :     0.004140s : 31.21%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000061s :  0.46%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.23%
    rewriter_before_opt_a                :     0.000070s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.42%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000600s :  4.53%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000543s :  4.10%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.14%
    jit_opt_a.cse                        :     0.000034s :  0.25%
    jit_opt_a.replace_applicator         :     0.000018s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.08%
    rewriter_after_opt_a                 :     0.000035s :  0.26%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000521s :  3.93%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000025s :  0.19%
    loop_unroll                          :     0.000405s :  3.05%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.35%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000013s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000443s :  3.34%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005465s : 41.21%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000189    23
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.77% :     0.000005s :      3: substitution.graph_param_transform
82.78% :     0.000156s :      4: substitution.inline
 1.53% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.19% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.39% :     0.000004s :      2: substitution.replace_old_param
 6.72% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004082     2
81.28% :     0.003318s :      1: type_inference.infer
18.72% :     0.000764s :      1: type_inference.specialize
------[replace.]   0.000059     6
67.95% :     0.000040s :      4: replace.inline
32.05% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000165     6
92.95% :     0.000154s :      4: match.inline
 7.05% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
10.45% :     0.000013s :     10: predicate.addn_check_dump
 1.12% :     0.000001s :     10: predicate.addn_zero_filter
 1.58% :     0.000002s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.01% :     0.000001s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.61% :     0.000007s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.90% :     0.000001s :      6: predicate.less_batch_normalization
 1.46% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 1.14% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.63% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000003s :     15: predicate.partial_eliminate
 1.23% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.34% :     0.000002s :     10: predicate.reduce_eliminate
 1.31% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.18% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 1.93% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.63% :     0.000008s :     53: predicate.switch_simplify
 1.06% :     0.000001s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.22% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.06% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.37% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.31% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.52% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000512    12
46.06% :     0.000236s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.25% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.69% :     0.000239s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018620    76
 0.26% :     0.000049s :      1: add_recomputation
 0.35% :     0.000064s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.38% :     0.000257s :      1: bootstrap
 0.15% :     0.000027s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
27.45% :     0.005111s :      1: jit_opt_a
 0.72% :     0.000134s :      1: jit_opt_after_cconv
 0.29% :     0.000054s :      1: jit_opt_b
 2.21% :     0.000412s :      1: loop_unroll
 2.84% :     0.000530s :      1: mutable_eliminate
 4.30% :     0.000801s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.42% :     0.000451s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000033s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.55% :     0.000288s :      1: renormalize.infer
 1.33% :     0.000247s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000038s :      1: rewriter_after_opt_a
 0.39% :     0.000073s :      1: rewriter_before_opt_a
 0.37% :     0.000070s :      1: symbol_engine_optimizer
29.41% :     0.005475s :      1: task_emit
22.31% :     0.004153s :      1: type_inference
 0.27% :     0.000051s :      1: validate


TotalTime = 0.0181378, [33]
[bootstrap]: 0.00028389
[type_inference]: 0.00471257
[event_method]: 1.81e-05
[auto_monad]: 6.729e-05
[graph_reusing]: 6.07001e-06
[pre_auto_parallel]: 2.72001e-06
[py_interpret_to_execute]: 3.422e-05
[rewriter_before_opt_a]: 7.31e-05
[expand_dump_flag]: 3.35e-06
[jit_opt_a]: 0.00518467, [2]
    [Cycle 1]: 0.00147158, [27]
        [switch_simplify]: 5.269e-05
        [loop_unroll]: 2.826e-05
        [a_1]: 0.00050571
        [with_stream_mark]: 1.872e-05
        [recompute_prepare]: 6.64999e-06
        [updatestate_depend_eliminate]: 3.58e-06
        [updatestate_assign_eliminate]: 3.06001e-06
        [updatestate_loads_eliminate]: 2.96001e-06
        [parameter_eliminate]: 1.80001e-06
        [specialize_transform]: 5.67001e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 5.10001e-06
        [meta_shard_fg_expand]: 2.98998e-06
        [get_grad_eliminate_]: 4.97999e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 1.00999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.308e-05
        [j_node_and_user_rematch]: 8.94e-06
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 8.55999e-06
        [inline_without_move]: 5.00001e-06
        [renormalize]: 0.00055191
        [add_forward_monad_depend]: 4.63001e-06
        [auto_monad_grad]: 2.02999e-06
        [auto_monad_eliminator]: 1.345e-05
        [cse]: 2.853e-05
        [replace_applicator]: 1.179e-05
    [Cycle 2]: 0.00035906, [27]
        [switch_simplify]: 5.44e-06
        [loop_unroll]: 4.74e-06
        [a_1]: 0.00010574
        [with_stream_mark]: 3.447e-05
        [recompute_prepare]: 5.94e-06
        [updatestate_depend_eliminate]: 2.47001e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.76e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 5.05999e-06
        [updatestate_useless_node_eliminater]: 4.68001e-06
        [accelerated_algorithm]: 4.95001e-06
        [meta_shard_fg_expand]: 1.03001e-06
        [get_grad_eliminate_]: 4.27e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 1.63002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.181e-05
        [j_node_and_user_rematch]: 7.34002e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 7.01001e-06
        [inline_without_move]: 4.57e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.69998e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 5.14e-06
        [cse]: 1.119e-05
        [replace_applicator]: 4.82e-06
[py_interpret_to_execute_after_opt_a]: 9.59e-06
[rewriter_after_opt_a]: 3.236e-05
[convert_after_rewriter]: 7.39002e-06
[order_py_execute_after_rewriter]: 4.97e-06
[mutable_eliminate]: 0.00048822
[jit_opt_b]: 4.74e-05, [1]
    [Cycle 1]: 4.07e-05, [2]
        [frontend_op_eliminate]: 1.538e-05
        [inline_after_opt_a]: 1.441e-05
[cconv]: 2.4e-05
[loop_unroll]: 0.00040033
[jit_opt_after_cconv]: 0.00013076, [1]
    [Cycle 1]: 0.00012483, [11]
        [c_1]: 1.934e-05
        [parameter_eliminate]: 2.36e-06
        [updatestate_depend_eliminate]: 5.35001e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.605e-05
        [call_graph_tuple_transform]: 1.765e-05
        [tuple_list_get_item_eliminator]: 5.09e-06
        [none_parameter_eliminate]: 1.43002e-06
        [renormalize]: 4.40021e-07
        [switch_simplify]: 5.09998e-06
[remove_dup_value]: 1.373e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 4.92e-06
[add_recomputation]: 4.772e-05
[cse_after_recomputation]: 1.988e-05, [1]
    [Cycle 1]: 1.468e-05, [1]
        [cse]: 9.13002e-06
[auto_monad_reorder]: 1.529e-05
[get_jit_bprop_graph]: 1.86998e-06
[rewriter_after_jit_bprop_graph]: 4.68999e-06
[opt_after_jit_grad]: 0.0004294
[symbol_engine_optimizer]: 6.894e-05, [1]
    [Cycle 1]: 6.28e-05, [6]
        [build]: 2.96001e-06
        [elim_shapecalc]: 7.03998e-06
        [elim_not_effective]: 1.204e-05
        [opt_reshape]: 5.72001e-06
        [fold_const_symbol]: 8.36002e-06
        [renormalize]: 5.69999e-07
[validate]: 3.146e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00576708
[execute]: 6.79999e-06
Sums
    bootstrap                            :     0.000284s :  2.00%
    type_inference                       :     0.004713s : 33.20%
    event_method                         :     0.000018s :  0.13%
    auto_monad                           :     0.000067s :  0.47%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.24%
    rewriter_before_opt_a                :     0.000073s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.41%
    jit_opt_a.loop_unroll                :     0.000033s :  0.23%
    jit_opt_a.a_1                        :     0.000611s :  4.31%
    jit_opt_a.with_stream_mark           :     0.000053s :  0.37%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000552s :  3.89%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000040s :  0.28%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000488s :  3.44%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000400s :  2.82%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000048s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000429s :  3.03%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000031s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005767s : 40.63%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000190    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.44% :     0.000005s :      3: substitution.graph_param_transform
83.60% :     0.000158s :      4: substitution.inline
 1.72% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.10% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.05% :     0.000004s :      2: substitution.replace_old_param
 6.47% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004639     2
82.54% :     0.003829s :      1: type_inference.infer
17.46% :     0.000810s :      1: type_inference.specialize
------[replace.]   0.000061     6
67.71% :     0.000041s :      4: replace.inline
32.29% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000167     6
93.28% :     0.000156s :      4: match.inline
 6.72% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.33% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.08% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000001s :     10: predicate.addn_zero_filter
 1.86% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000001s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.73% :     0.000006s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.69% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000001s :     10: predicate.merge_addn
 1.17% :     0.000001s :     10: predicate.minmaximum_grad
 1.57% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.35% :     0.000003s :     15: predicate.partial_eliminate
 1.40% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.90% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.90% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.60% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.43% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 1.17% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.74% :     0.000008s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.25% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.01% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.66% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000560    12
51.79% :     0.000290s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.79% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.43% :     0.000238s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019598    76
 0.26% :     0.000051s :      1: add_recomputation
 0.36% :     0.000071s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.54% :     0.000302s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.47% :     0.005188s :      1: jit_opt_a
 0.68% :     0.000133s :      1: jit_opt_after_cconv
 0.25% :     0.000050s :      1: jit_opt_b
 2.08% :     0.000407s :      1: loop_unroll
 2.53% :     0.000496s :      1: mutable_eliminate
 4.14% :     0.000811s :     26: opt.transform.jit_opt_a
 0.23% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.23% :     0.000437s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.54% :     0.000301s :      1: renormalize.infer
 1.24% :     0.000243s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000037s :      1: rewriter_after_opt_a
 0.39% :     0.000076s :      1: rewriter_before_opt_a
 0.36% :     0.000071s :      1: symbol_engine_optimizer
29.47% :     0.005776s :      1: task_emit
24.14% :     0.004732s :      1: type_inference
 0.26% :     0.000050s :      1: validate


TotalTime = 0.0113934, [30]
[bootstrap]: 0.00027646
[type_inference]: 0.00409045
[event_method]: 1.79e-05
[auto_monad]: 5.716e-05
[graph_reusing]: 5.05999e-06
[pre_auto_parallel]: 2.34999e-06
[py_interpret_to_execute]: 3.598e-05
[rewriter_before_opt_a]: 7.03e-05
[expand_dump_flag]: 3.15002e-06
[jit_opt_a]: 0.00484526, [2]
    [Cycle 1]: 0.00141051, [27]
        [switch_simplify]: 4.63e-05
        [loop_unroll]: 3.395e-05
        [a_1]: 0.00049875
        [with_stream_mark]: 1.373e-05
        [recompute_prepare]: 7.00998e-06
        [updatestate_depend_eliminate]: 3.03998e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 5.99e-06
        [updatestate_useless_node_eliminater]: 5.13002e-06
        [accelerated_algorithm]: 5.44e-06
        [meta_shard_fg_expand]: 1.53002e-06
        [get_grad_eliminate_]: 5.14e-06
        [merge_forward]: 2.96001e-06
        [cell_reuse_recompute_pass]: 1.13001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.229e-05
        [j_node_and_user_rematch]: 9.45001e-06
        [meta_fg_expand]: 2.19999e-06
        [replace_old_param]: 7.68001e-06
        [inline_without_move]: 5.26998e-06
        [renormalize]: 0.00052797
        [add_forward_monad_depend]: 4.88001e-06
        [auto_monad_grad]: 1.77999e-06
        [auto_monad_eliminator]: 1.174e-05
        [cse]: 1.647e-05
        [replace_applicator]: 1.132e-05
    [Cycle 2]: 0.0003211, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 4.99998e-06
        [a_1]: 8.94e-05
        [with_stream_mark]: 7.95e-06
        [recompute_prepare]: 5.14e-06
        [updatestate_depend_eliminate]: 2.67001e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 9.20001e-07
        [specialize_transform]: 5.25001e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.30999e-06
        [meta_shard_fg_expand]: 1.07998e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.223e-05
        [j_node_and_user_rematch]: 8.01001e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 7.41999e-06
        [inline_without_move]: 8.33999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 4.84e-06
        [cse]: 1.086e-05
        [replace_applicator]: 5.21998e-06
[py_interpret_to_execute_after_opt_a]: 8.18999e-06
[rewriter_after_opt_a]: 3.146e-05
[convert_after_rewriter]: 6.26998e-06
[order_py_execute_after_rewriter]: 5.05999e-06
[mutable_eliminate]: 0.0004654
[jit_opt_b]: 4.904e-05, [1]
    [Cycle 1]: 4.238e-05, [2]
        [frontend_op_eliminate]: 1.618e-05
        [inline_after_opt_a]: 1.484e-05
[cconv]: 2.092e-05
[loop_unroll]: 0.00047476
[jit_opt_after_cconv]: 0.00013591, [1]
    [Cycle 1]: 0.00012899, [11]
        [c_1]: 2.05e-05
        [parameter_eliminate]: 2.39001e-06
        [updatestate_depend_eliminate]: 5.32999e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 1.638e-05
        [call_graph_tuple_transform]: 1.73e-05
        [tuple_list_get_item_eliminator]: 5.34e-06
        [none_parameter_eliminate]: 1.25001e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 5.64e-06
[remove_dup_value]: 1.083e-05
[partial_unused_args_eliminate]: 1.51002e-06
[environ_conv]: 4.57e-06
[add_recomputation]: 4.005e-05
[cse_after_recomputation]: 1.996e-05, [1]
    [Cycle 1]: 1.495e-05, [1]
        [cse]: 9.54999e-06
[auto_monad_reorder]: 1.17e-05
[get_jit_bprop_graph]: 1.49e-06
[rewriter_after_jit_bprop_graph]: 4.92e-06
[opt_after_jit_grad]: 0.00044262
[symbol_engine_optimizer]: 7.399e-05, [1]
    [Cycle 1]: 6.765e-05, [6]
        [build]: 3.03e-06
        [elim_shapecalc]: 7.88999e-06
        [elim_not_effective]: 1.313e-05
        [opt_reshape]: 5.87999e-06
        [fold_const_symbol]: 9.02e-06
        [renormalize]: 4.19997e-07
[validate]: 2.969e-05
Sums
    bootstrap                            :     0.000276s :  3.58%
    type_inference                       :     0.004090s : 52.92%
    event_method                         :     0.000018s :  0.23%
    auto_monad                           :     0.000057s :  0.74%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.47%
    rewriter_before_opt_a                :     0.000070s :  0.91%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000052s :  0.67%
    jit_opt_a.loop_unroll                :     0.000039s :  0.50%
    jit_opt_a.a_1                        :     0.000588s :  7.61%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.20%
    jit_opt_a.inline_without_move        :     0.000014s :  0.18%
    jit_opt_a.renormalize                :     0.000528s :  6.83%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.21%
    jit_opt_a.cse                        :     0.000027s :  0.35%
    jit_opt_a.replace_applicator         :     0.000017s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000031s :  0.41%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000465s :  6.02%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000021s :  0.27%
    loop_unroll                          :     0.000475s :  6.14%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000011s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000040s :  0.52%
    cse_after_recomputation.cse          :     0.000010s :  0.12%
    auto_monad_reorder                   :     0.000012s :  0.15%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000443s :  5.73%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000030s :  0.38%


Time group info:
------[substitution.]   0.000169    23
 1.14% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.42% :     0.000004s :      3: substitution.graph_param_transform
83.60% :     0.000141s :      4: substitution.inline
 1.87% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.59% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.72% :     0.000003s :      2: substitution.replace_old_param
 5.93% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004035     2
79.75% :     0.003218s :      1: type_inference.infer
20.25% :     0.000817s :      1: type_inference.specialize
------[replace.]   0.000063     6
65.55% :     0.000041s :      4: replace.inline
34.45% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000147     6
93.98% :     0.000139s :      4: match.inline
 6.02% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.39% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000001s :     10: predicate.addn_zero_filter
 2.07% :     0.000002s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.58% :     0.000007s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.48% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.90% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.19% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.19% :     0.000001s :     10: predicate.minmaximum_grad
 1.42% :     0.000002s :      3: predicate.mutable_eliminate
 0.58% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.26% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     10: predicate.reduce_eliminate
 1.64% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.60% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 2.28% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.43% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.59% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.60% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.86% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000561    12
50.05% :     0.000281s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.84% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.11% :     0.000253s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012785    72
 0.34% :     0.000043s :      1: add_recomputation
 0.47% :     0.000060s :      1: auto_monad
 0.11% :     0.000014s :      1: auto_monad_reorder
 2.26% :     0.000289s :      1: bootstrap
 0.18% :     0.000024s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.17% :     0.000022s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
37.92% :     0.004848s :      1: jit_opt_a
 1.09% :     0.000139s :      1: jit_opt_after_cconv
 0.40% :     0.000051s :      1: jit_opt_b
 3.78% :     0.000484s :      1: loop_unroll
 3.70% :     0.000473s :      1: mutable_eliminate
 6.22% :     0.000796s :     26: opt.transform.jit_opt_a
 0.36% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000025s :      4: opt.transform.jit_opt_b
 0.10% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.52% :     0.000450s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000003s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.30% :     0.000039s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000013s :      1: remove_dup_value
 2.04% :     0.000261s :      1: renormalize.infer
 2.04% :     0.000261s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000035s :      1: rewriter_after_opt_a
 0.57% :     0.000073s :      1: rewriter_before_opt_a
 0.60% :     0.000077s :      1: symbol_engine_optimizer
32.08% :     0.004101s :      1: type_inference


TotalTime = 0.0162605, [33]
[bootstrap]: 0.00024932
[type_inference]: 0.00399793
[event_method]: 1.611e-05
[auto_monad]: 5.166e-05
[graph_reusing]: 4.70001e-06
[pre_auto_parallel]: 2.11998e-06
[py_interpret_to_execute]: 3.009e-05
[rewriter_before_opt_a]: 8.199e-05
[expand_dump_flag]: 2.74001e-06
[jit_opt_a]: 0.00468737, [2]
    [Cycle 1]: 0.00139763, [27]
        [switch_simplify]: 4.48e-05
        [loop_unroll]: 2.882e-05
        [a_1]: 0.00047864
        [with_stream_mark]: 1.313e-05
        [recompute_prepare]: 7.29001e-06
        [updatestate_depend_eliminate]: 2.71e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 1.57999e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 4.99998e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 1.66e-06
        [get_grad_eliminate_]: 5.00999e-06
        [merge_forward]: 3.26999e-06
        [cell_reuse_recompute_pass]: 9.20001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.237e-05
        [j_node_and_user_rematch]: 8.48999e-06
        [meta_fg_expand]: 2.14999e-06
        [replace_old_param]: 7.87e-06
        [inline_without_move]: 5.00999e-06
        [renormalize]: 0.0005469
        [add_forward_monad_depend]: 4.35999e-06
        [auto_monad_grad]: 1.79e-06
        [auto_monad_eliminator]: 1.134e-05
        [cse]: 1.768e-05
        [replace_applicator]: 1.154e-05
    [Cycle 2]: 0.0003116, [27]
        [switch_simplify]: 5.44e-06
        [loop_unroll]: 4.74e-06
        [a_1]: 8.753e-05
        [with_stream_mark]: 8.43001e-06
        [recompute_prepare]: 4.85001e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 4.74002e-06
        [accelerated_algorithm]: 4.80001e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.40999e-06
        [merge_forward]: 2.37001e-06
        [cell_reuse_recompute_pass]: 1.69998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.147e-05
        [j_node_and_user_rematch]: 7.12002e-06
        [meta_fg_expand]: 1.38002e-06
        [replace_old_param]: 1.193e-05
        [inline_without_move]: 4.80999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.21002e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 5.02999e-06
        [cse]: 1.014e-05
        [replace_applicator]: 5.00001e-06
[py_interpret_to_execute_after_opt_a]: 1.103e-05
[rewriter_after_opt_a]: 2.921e-05
[convert_after_rewriter]: 5.84e-06
[order_py_execute_after_rewriter]: 5.44998e-06
[mutable_eliminate]: 0.00046915
[jit_opt_b]: 4.669e-05, [1]
    [Cycle 1]: 4.045e-05, [2]
        [frontend_op_eliminate]: 1.552e-05
        [inline_after_opt_a]: 1.414e-05
[cconv]: 1.969e-05
[loop_unroll]: 0.00040568
[jit_opt_after_cconv]: 0.00013001, [1]
    [Cycle 1]: 0.00012385, [11]
        [c_1]: 1.953e-05
        [parameter_eliminate]: 1.92999e-06
        [updatestate_depend_eliminate]: 4.68001e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.598e-05
        [call_graph_tuple_transform]: 1.721e-05
        [tuple_list_get_item_eliminator]: 5.04e-06
        [none_parameter_eliminate]: 1.15999e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 4.94e-06
[remove_dup_value]: 1.086e-05
[partial_unused_args_eliminate]: 1.57999e-06
[environ_conv]: 4.15999e-06
[add_recomputation]: 4.086e-05
[cse_after_recomputation]: 1.919e-05, [1]
    [Cycle 1]: 1.401e-05, [1]
        [cse]: 8.51002e-06
[auto_monad_reorder]: 1.148e-05
[get_jit_bprop_graph]: 1.42e-06
[rewriter_after_jit_bprop_graph]: 4.82e-06
[opt_after_jit_grad]: 0.00045344
[symbol_engine_optimizer]: 6.922e-05, [1]
    [Cycle 1]: 6.309e-05, [6]
        [build]: 2.94999e-06
        [elim_shapecalc]: 7.29001e-06
        [elim_not_effective]: 1.227e-05
        [opt_reshape]: 5.53002e-06
        [fold_const_symbol]: 8.2e-06
        [renormalize]: 3.99974e-07
[validate]: 2.907e-05
[backend_pass]: 7.50006e-07
[task_emit]: 0.00519911
[execute]: 5.86e-06
Sums
    bootstrap                            :     0.000249s :  1.96%
    type_inference                       :     0.003998s : 31.39%
    event_method                         :     0.000016s :  0.13%
    auto_monad                           :     0.000052s :  0.41%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.24%
    rewriter_before_opt_a                :     0.000082s :  0.64%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000050s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.26%
    jit_opt_a.a_1                        :     0.000566s :  4.45%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000020s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000547s :  4.29%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.13%
    jit_opt_a.cse                        :     0.000028s :  0.22%
    jit_opt_a.replace_applicator         :     0.000017s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.09%
    rewriter_after_opt_a                 :     0.000029s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000469s :  3.68%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000020s :  0.15%
    loop_unroll                          :     0.000406s :  3.19%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000011s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000041s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000011s :  0.09%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000453s :  3.56%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005199s : 40.82%
    execute                              :     0.000006s :  0.05%


Time group info:
------[substitution.]   0.000163    23
 1.25% :     0.000002s :      2: substitution.elim_not_effective
 0.77% :     0.000001s :      2: substitution.fold_const_symbol
 3.10% :     0.000005s :      3: substitution.graph_param_transform
82.79% :     0.000135s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.49% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.99% :     0.000003s :      2: substitution.replace_old_param
 6.05% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003951     2
81.18% :     0.003207s :      1: type_inference.infer
18.82% :     0.000744s :      1: type_inference.specialize
------[replace.]   0.000058     6
66.61% :     0.000039s :      4: replace.inline
33.39% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000141     6
93.75% :     0.000132s :      4: match.inline
 6.25% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.39% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.48% :     0.000002s :     10: predicate.addn_zero_filter
 1.73% :     0.000002s :     10: predicate.arithmetic_simplify
 1.34% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.51% :     0.000002s :     10: predicate.depend_value_elim
 1.22% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000001s :     10: predicate.environ_get_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.21% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.92% :     0.000001s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.74% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.21% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.19% :     0.000001s :     10: predicate.minmaximum_grad
 1.47% :     0.000002s :      3: predicate.mutable_eliminate
 0.57% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000002s :     15: predicate.partial_eliminate
 1.37% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.54% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 2.26% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.27% :     0.000008s :     53: predicate.switch_simplify
 1.30% :     0.000001s :     10: predicate.tile_eliminate
 1.36% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.34% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.08% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000490    12
45.67% :     0.000224s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.52% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.81% :     0.000239s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017674    76
 0.25% :     0.000044s :      1: add_recomputation
 0.31% :     0.000055s :      1: auto_monad
 0.08% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.46% :     0.000258s :      1: bootstrap
 0.13% :     0.000022s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.54% :     0.004690s :      1: jit_opt_a
 0.75% :     0.000133s :      1: jit_opt_after_cconv
 0.28% :     0.000049s :      1: jit_opt_b
 2.33% :     0.000412s :      1: loop_unroll
 2.70% :     0.000478s :      1: mutable_eliminate
 4.32% :     0.000763s :     26: opt.transform.jit_opt_a
 0.25% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.61% :     0.000461s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000033s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000013s :      1: remove_dup_value
 1.67% :     0.000295s :      1: renormalize.infer
 1.39% :     0.000246s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000032s :      1: rewriter_after_opt_a
 0.48% :     0.000085s :      1: rewriter_before_opt_a
 0.41% :     0.000072s :      1: symbol_engine_optimizer
29.47% :     0.005208s :      1: task_emit
22.68% :     0.004008s :      1: type_inference
 0.25% :     0.000044s :      1: validate


TotalTime = 0.0106895, [30]
[bootstrap]: 0.00024421
[type_inference]: 0.00369525
[event_method]: 1.529e-05
[auto_monad]: 4.953e-05
[graph_reusing]: 4.17998e-06
[pre_auto_parallel]: 2.17001e-06
[py_interpret_to_execute]: 2.965e-05
[rewriter_before_opt_a]: 6.542e-05
[expand_dump_flag]: 2.30002e-06
[jit_opt_a]: 0.00469383, [2]
    [Cycle 1]: 0.00135195, [27]
        [switch_simplify]: 4.297e-05
        [loop_unroll]: 2.896e-05
        [a_1]: 0.00047611
        [with_stream_mark]: 1.276e-05
        [recompute_prepare]: 7.13998e-06
        [updatestate_depend_eliminate]: 3.23e-06
        [updatestate_assign_eliminate]: 2.69999e-06
        [updatestate_loads_eliminate]: 2.26998e-06
        [parameter_eliminate]: 1.42999e-06
        [specialize_transform]: 6.12999e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 1.57001e-06
        [get_grad_eliminate_]: 5.27999e-06
        [merge_forward]: 2.74001e-06
        [cell_reuse_recompute_pass]: 8.70001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.19e-05
        [j_node_and_user_rematch]: 9.09e-06
        [meta_fg_expand]: 1.97001e-06
        [replace_old_param]: 7.55e-06
        [inline_without_move]: 5.17e-06
        [renormalize]: 0.0005082
        [add_forward_monad_depend]: 4.53999e-06
        [auto_monad_grad]: 2.01e-06
        [auto_monad_eliminator]: 1.018e-05
        [cse]: 1.567e-05
        [replace_applicator]: 1.09e-05
    [Cycle 2]: 0.0003198, [27]
        [switch_simplify]: 5.86998e-06
        [loop_unroll]: 4.67998e-06
        [a_1]: 9.009e-05
        [with_stream_mark]: 8.97999e-06
        [recompute_prepare]: 5.32001e-06
        [updatestate_depend_eliminate]: 2.46998e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.95001e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 4.87e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.31002e-06
        [meta_shard_fg_expand]: 1.35999e-06
        [get_grad_eliminate_]: 4.74998e-06
        [merge_forward]: 2.58e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.225e-05
        [j_node_and_user_rematch]: 7.56999e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 6.96999e-06
        [inline_without_move]: 4.68001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 4.88001e-06
        [cse]: 9.72001e-06
        [replace_applicator]: 4.95999e-06
[py_interpret_to_execute_after_opt_a]: 8.84e-06
[rewriter_after_opt_a]: 2.586e-05
[convert_after_rewriter]: 8.38001e-06
[order_py_execute_after_rewriter]: 4.55999e-06
[mutable_eliminate]: 0.0004678
[jit_opt_b]: 4.952e-05, [1]
    [Cycle 1]: 4.317e-05, [2]
        [frontend_op_eliminate]: 1.671e-05
        [inline_after_opt_a]: 1.455e-05
[cconv]: 1.917e-05
[loop_unroll]: 0.00040688
[jit_opt_after_cconv]: 0.00013126, [1]
    [Cycle 1]: 0.00012478, [11]
        [c_1]: 1.988e-05
        [parameter_eliminate]: 2.31e-06
        [updatestate_depend_eliminate]: 4.77e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.475e-05
        [call_graph_tuple_transform]: 1.67e-05
        [tuple_list_get_item_eliminator]: 5.28002e-06
        [none_parameter_eliminate]: 1.32e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.10001e-06
[remove_dup_value]: 9.22001e-06
[partial_unused_args_eliminate]: 1.69998e-06
[environ_conv]: 4.1e-06
[add_recomputation]: 3.477e-05
[cse_after_recomputation]: 1.954e-05, [1]
    [Cycle 1]: 1.448e-05, [1]
        [cse]: 8.85999e-06
[auto_monad_reorder]: 1.058e-05
[get_jit_bprop_graph]: 1.71e-06
[rewriter_after_jit_bprop_graph]: 4e-06
[opt_after_jit_grad]: 0.00043973
[symbol_engine_optimizer]: 7.135e-05, [1]
    [Cycle 1]: 6.488e-05, [6]
        [build]: 2.81e-06
        [elim_shapecalc]: 6.94999e-06
        [elim_not_effective]: 1.22e-05
        [opt_reshape]: 5.87999e-06
        [fold_const_symbol]: 8.77e-06
        [renormalize]: 4.50003e-07
[validate]: 2.632e-05
Sums
    bootstrap                            :     0.000244s :  3.43%
    type_inference                       :     0.003695s : 51.84%
    event_method                         :     0.000015s :  0.21%
    auto_monad                           :     0.000050s :  0.69%
    graph_reusing                        :     0.000004s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000030s :  0.42%
    rewriter_before_opt_a                :     0.000065s :  0.92%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000049s :  0.69%
    jit_opt_a.loop_unroll                :     0.000034s :  0.47%
    jit_opt_a.a_1                        :     0.000566s :  7.94%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000508s :  7.13%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.21%
    jit_opt_a.cse                        :     0.000025s :  0.36%
    jit_opt_a.replace_applicator         :     0.000016s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.12%
    rewriter_after_opt_a                 :     0.000026s :  0.36%
    convert_after_rewriter               :     0.000008s :  0.12%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000468s :  6.56%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.20%
    cconv                                :     0.000019s :  0.27%
    loop_unroll                          :     0.000407s :  5.71%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000015s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000009s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000035s :  0.49%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000011s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.06%
    opt_after_jit_grad                   :     0.000440s :  6.17%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000026s :  0.37%


Time group info:
------[substitution.]   0.000157    23
 1.20% :     0.000002s :      2: substitution.elim_not_effective
 0.79% :     0.000001s :      2: substitution.fold_const_symbol
 2.56% :     0.000004s :      3: substitution.graph_param_transform
82.95% :     0.000130s :      4: substitution.inline
 2.03% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.47% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.92% :     0.000003s :      2: substitution.replace_old_param
 6.08% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003644     2
79.99% :     0.002915s :      1: type_inference.infer
20.01% :     0.000729s :      1: type_inference.specialize
------[replace.]   0.000060     6
66.18% :     0.000040s :      4: replace.inline
33.82% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000136     6
93.87% :     0.000128s :      4: match.inline
 6.13% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.53% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.30% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 2.08% :     0.000002s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000002s :     10: predicate.depend_value_elim
 1.33% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.51% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      3: predicate.fold_const_symbol
 1.16% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.30% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.25% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.14% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.60% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.30% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.79% :     0.000002s :     10: predicate.reduce_eliminate
 1.56% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.48% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.27% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.29% :     0.000008s :     53: predicate.switch_simplify
 1.28% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.50% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.04% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.92% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000523    12
50.42% :     0.000264s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.12% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.46% :     0.000233s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012026    72
 0.31% :     0.000038s :      1: add_recomputation
 0.44% :     0.000053s :      1: auto_monad
 0.11% :     0.000013s :      1: auto_monad_reorder
 2.10% :     0.000252s :      1: bootstrap
 0.18% :     0.000022s :      1: cconv
 0.09% :     0.000011s :      1: convert_after_rewriter
 0.18% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000019s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000006s :      1: graph_reusing
39.05% :     0.004697s :      1: jit_opt_a
 1.11% :     0.000134s :      1: jit_opt_after_cconv
 0.44% :     0.000052s :      1: jit_opt_b
 3.44% :     0.000414s :      1: loop_unroll
 3.95% :     0.000475s :      1: mutable_eliminate
 6.32% :     0.000760s :     26: opt.transform.jit_opt_a
 0.36% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.72% :     0.000447s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000032s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000012s :      1: remove_dup_value
 2.05% :     0.000246s :      1: renormalize.infer
 2.13% :     0.000256s :      1: renormalize.specialize
 0.05% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000029s :      1: rewriter_after_opt_a
 0.57% :     0.000068s :      1: rewriter_before_opt_a
 0.62% :     0.000074s :      1: symbol_engine_optimizer
30.81% :     0.003706s :      1: type_inference


TotalTime = 0.0304201, [33]
[bootstrap]: 0.00034453
[type_inference]: 0.00597303
[event_method]: 1.956e-05
[auto_monad]: 6.656e-05
[graph_reusing]: 6.51999e-06
[pre_auto_parallel]: 2.57001e-06
[py_interpret_to_execute]: 3.519e-05
[rewriter_before_opt_a]: 7.534e-05
[expand_dump_flag]: 3.55e-06
[jit_opt_a]: 0.00914066, [2]
    [Cycle 1]: 0.00202973, [27]
        [switch_simplify]: 6.745e-05
        [loop_unroll]: 2.975e-05
        [a_1]: 0.00068961
        [with_stream_mark]: 2.754e-05
        [recompute_prepare]: 8.86997e-06
        [updatestate_depend_eliminate]: 4.25e-06
        [updatestate_assign_eliminate]: 2.98998e-06
        [updatestate_loads_eliminate]: 2.64999e-06
        [parameter_eliminate]: 2.94999e-06
        [specialize_transform]: 6.44999e-06
        [updatestate_useless_node_eliminater]: 4.86997e-06
        [accelerated_algorithm]: 5.44e-06
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 5.15001e-06
        [merge_forward]: 3.95e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.544e-05
        [j_node_and_user_rematch]: 9.61e-06
        [meta_fg_expand]: 2.53e-06
        [replace_old_param]: 1.12e-05
        [inline_without_move]: 5.02999e-06
        [renormalize]: 0.00082451
        [add_forward_monad_depend]: 7.71001e-06
        [auto_monad_grad]: 2.61e-06
        [auto_monad_eliminator]: 1.768e-05
        [cse]: 3.563e-05
        [replace_applicator]: 1.805e-05
    [Cycle 2]: 0.00033891, [27]
        [switch_simplify]: 5.42999e-06
        [loop_unroll]: 5.08002e-06
        [a_1]: 9.463e-05
        [with_stream_mark]: 1.36e-05
        [recompute_prepare]: 5.15999e-06
        [updatestate_depend_eliminate]: 3.03e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 5.24998e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 4.87e-06
        [meta_shard_fg_expand]: 1.62001e-06
        [get_grad_eliminate_]: 4.30999e-06
        [merge_forward]: 3.08998e-06
        [cell_reuse_recompute_pass]: 3.03e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.606e-05
        [j_node_and_user_rematch]: 8.08999e-06
        [meta_fg_expand]: 1.57999e-06
        [replace_old_param]: 7.11999e-06
        [inline_without_move]: 4.86002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.57999e-06
        [auto_monad_grad]: 1.18001e-06
        [auto_monad_eliminator]: 6.06e-06
        [cse]: 1.19e-05
        [replace_applicator]: 5.27001e-06
[py_interpret_to_execute_after_opt_a]: 1.71e-05
[rewriter_after_opt_a]: 4.384e-05
[convert_after_rewriter]: 7.43999e-06
[order_py_execute_after_rewriter]: 5.65001e-06
[mutable_eliminate]: 0.00089999
[jit_opt_b]: 7.772e-05, [1]
    [Cycle 1]: 6.715e-05, [2]
        [frontend_op_eliminate]: 1.902e-05
        [inline_after_opt_a]: 2.049e-05
[cconv]: 4.201e-05
[loop_unroll]: 0.00059966
[jit_opt_after_cconv]: 0.00023642, [1]
    [Cycle 1]: 0.0002265, [11]
        [c_1]: 2.312e-05
        [parameter_eliminate]: 8.13001e-06
        [updatestate_depend_eliminate]: 1.212e-05
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 3.04999e-06
        [cse]: 3.987e-05
        [call_graph_tuple_transform]: 3.024e-05
        [tuple_list_get_item_eliminator]: 6.56e-06
        [none_parameter_eliminate]: 1.99e-06
        [renormalize]: 1.00001e-06
        [switch_simplify]: 5.77999e-06
[remove_dup_value]: 1.685e-05
[partial_unused_args_eliminate]: 2.32001e-06
[environ_conv]: 7.15e-06
[add_recomputation]: 6.574e-05
[cse_after_recomputation]: 2.376e-05, [1]
    [Cycle 1]: 1.721e-05, [1]
        [cse]: 1.01e-05
[auto_monad_reorder]: 1.606e-05
[get_jit_bprop_graph]: 3.18998e-06
[rewriter_after_jit_bprop_graph]: 1.16e-05
[opt_after_jit_grad]: 0.00099723
[symbol_engine_optimizer]: 8.798e-05, [1]
    [Cycle 1]: 7.903e-05, [6]
        [build]: 6.75998e-06
        [elim_shapecalc]: 8.11002e-06
        [elim_not_effective]: 1.787e-05
        [opt_reshape]: 6.35002e-06
        [fold_const_symbol]: 9.27001e-06
        [renormalize]: 1.41002e-06
[validate]: 4.947e-05
[backend_pass]: 8.90024e-07
[task_emit]: 0.0112318
[execute]: 1.113e-05
Sums
    bootstrap                            :     0.000345s :  1.51%
    type_inference                       :     0.005973s : 26.16%
    event_method                         :     0.000020s :  0.09%
    auto_monad                           :     0.000067s :  0.29%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000035s :  0.15%
    rewriter_before_opt_a                :     0.000075s :  0.33%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000073s :  0.32%
    jit_opt_a.loop_unroll                :     0.000035s :  0.15%
    jit_opt_a.a_1                        :     0.000784s :  3.44%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.04%
    jit_opt_a.renormalize                :     0.000825s :  3.61%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.10%
    jit_opt_a.cse                        :     0.000048s :  0.21%
    jit_opt_a.replace_applicator         :     0.000023s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.07%
    rewriter_after_opt_a                 :     0.000044s :  0.19%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000900s :  3.94%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.09%
    cconv                                :     0.000042s :  0.18%
    loop_unroll                          :     0.000600s :  2.63%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000040s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000030s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000066s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.07%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000012s :  0.05%
    opt_after_jit_grad                   :     0.000997s :  4.37%
    symbol_engine_optimizer.build        :     0.000007s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000049s :  0.22%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011232s : 49.20%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000354    23
 0.85% :     0.000003s :      2: substitution.elim_not_effective
 0.47% :     0.000002s :      2: substitution.fold_const_symbol
 1.79% :     0.000006s :      3: substitution.graph_param_transform
88.57% :     0.000314s :      4: substitution.inline
 1.17% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.41% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.86% :     0.000007s :      2: substitution.replace_old_param
 3.89% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005892     2
83.91% :     0.004944s :      1: type_inference.infer
16.09% :     0.000948s :      1: type_inference.specialize
------[replace.]   0.000071     6
70.27% :     0.000050s :      4: replace.inline
29.73% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000323     6
96.07% :     0.000310s :      4: match.inline
 3.93% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.73% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.00% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.53% :     0.000002s :     10: predicate.addn_zero_filter
 2.13% :     0.000003s :     10: predicate.arithmetic_simplify
 1.54% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.09% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.02% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.38% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.03% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.81% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000000s :      3: predicate.graph_param_transform
 6.55% :     0.000008s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000002s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.06% :     0.000003s :     15: predicate.load_eliminater
 1.60% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.56% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.68% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.94% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 2.78% :     0.000004s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.00% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.32% :     0.000002s :     10: predicate.reduce_eliminate
 1.59% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.88% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.47% :     0.000001s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.03% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.07% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.55% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 1.97% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.42% :     0.000009s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 0.99% :     0.000001s :     10: predicate.transpose_eliminate
 1.18% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.03% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.35% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000764    12
47.84% :     0.000365s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.04% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.12% :     0.000337s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.032405    76
 0.22% :     0.000070s :      1: add_recomputation
 0.22% :     0.000070s :      1: auto_monad
 0.06% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.14% :     0.000370s :      1: bootstrap
 0.14% :     0.000045s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.08% :     0.000025s :      1: event_method
 0.05% :     0.000018s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000006s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
28.22% :     0.009145s :      1: jit_opt_a
 0.74% :     0.000240s :      1: jit_opt_after_cconv
 0.25% :     0.000081s :      1: jit_opt_b
 1.89% :     0.000612s :      1: loop_unroll
 2.89% :     0.000936s :      1: mutable_eliminate
 3.14% :     0.001016s :     26: opt.transform.jit_opt_a
 0.19% :     0.000062s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000029s :      4: opt.transform.jit_opt_b
 0.05% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000026s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000038s :      4: opt.transform.symbol_engine_opt
 3.14% :     0.001017s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000038s :      1: py_interpret_to_execute
 0.06% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000019s :      1: remove_dup_value
 1.49% :     0.000484s :      1: renormalize.infer
 1.02% :     0.000332s :      1: renormalize.specialize
 0.04% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000047s :      1: rewriter_after_opt_a
 0.24% :     0.000078s :      1: rewriter_before_opt_a
 0.28% :     0.000091s :      1: symbol_engine_optimizer
34.74% :     0.011256s :      1: task_emit
18.49% :     0.005993s :      1: type_inference
 0.24% :     0.000079s :      1: validate


TotalTime = 0.0116884, [30]
[bootstrap]: 0.00027025
[type_inference]: 0.00406833
[event_method]: 1.713e-05
[auto_monad]: 6.19e-05
[graph_reusing]: 5.47999e-06
[pre_auto_parallel]: 2.51e-06
[py_interpret_to_execute]: 3.325e-05
[rewriter_before_opt_a]: 7.546e-05
[expand_dump_flag]: 3.38e-06
[jit_opt_a]: 0.00510401, [2]
    [Cycle 1]: 0.00152427, [27]
        [switch_simplify]: 5.022e-05
        [loop_unroll]: 2.985e-05
        [a_1]: 0.00052779
        [with_stream_mark]: 1.676e-05
        [recompute_prepare]: 7.88001e-06
        [updatestate_depend_eliminate]: 3.61001e-06
        [updatestate_assign_eliminate]: 3.07002e-06
        [updatestate_loads_eliminate]: 2.79001e-06
        [parameter_eliminate]: 1.90001e-06
        [specialize_transform]: 6.09001e-06
        [updatestate_useless_node_eliminater]: 5.20001e-06
        [accelerated_algorithm]: 5.60001e-06
        [meta_shard_fg_expand]: 2.64001e-06
        [get_grad_eliminate_]: 6.19999e-06
        [merge_forward]: 4.13001e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.62e-05
        [j_node_and_user_rematch]: 1.017e-05
        [meta_fg_expand]: 2.68e-06
        [replace_old_param]: 8.48001e-06
        [inline_without_move]: 5.31998e-06
        [renormalize]: 0.00057795
        [add_forward_monad_depend]: 4.92e-06
        [auto_monad_grad]: 1.99999e-06
        [auto_monad_eliminator]: 1.341e-05
        [cse]: 2.694e-05
        [replace_applicator]: 1.072e-05
    [Cycle 2]: 0.0003316, [27]
        [switch_simplify]: 5.85002e-06
        [loop_unroll]: 5.37999e-06
        [a_1]: 9.544e-05
        [with_stream_mark]: 8.24998e-06
        [recompute_prepare]: 5.30999e-06
        [updatestate_depend_eliminate]: 2.84999e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 9.30013e-07
        [specialize_transform]: 5.49998e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.34998e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.85001e-06
        [merge_forward]: 2.71999e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.309e-05
        [j_node_and_user_rematch]: 7.97003e-06
        [meta_fg_expand]: 1.64998e-06
        [replace_old_param]: 7.55e-06
        [inline_without_move]: 4.79e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 5.51e-06
        [cse]: 1.184e-05
        [replace_applicator]: 5.04e-06
[py_interpret_to_execute_after_opt_a]: 9.82999e-06
[rewriter_after_opt_a]: 3.297e-05
[convert_after_rewriter]: 7.43e-06
[order_py_execute_after_rewriter]: 4.95999e-06
[mutable_eliminate]: 0.00051248
[jit_opt_b]: 5.2e-05, [1]
    [Cycle 1]: 4.538e-05, [2]
        [frontend_op_eliminate]: 1.654e-05
        [inline_after_opt_a]: 1.665e-05
[cconv]: 2.439e-05
[loop_unroll]: 0.00041847
[jit_opt_after_cconv]: 0.00013623, [1]
    [Cycle 1]: 0.00012992, [11]
        [c_1]: 2.054e-05
        [parameter_eliminate]: 2.39001e-06
        [updatestate_depend_eliminate]: 5.21002e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 1.708e-05
        [call_graph_tuple_transform]: 1.845e-05
        [tuple_list_get_item_eliminator]: 5.24e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.45001e-06
[remove_dup_value]: 1.462e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 4.99e-06
[add_recomputation]: 4.742e-05
[cse_after_recomputation]: 2.05e-05, [1]
    [Cycle 1]: 1.495e-05, [1]
        [cse]: 9.22001e-06
[auto_monad_reorder]: 1.515e-05
[get_jit_bprop_graph]: 1.60999e-06
[rewriter_after_jit_bprop_graph]: 4.94998e-06
[opt_after_jit_grad]: 0.00047093
[symbol_engine_optimizer]: 7.454e-05, [1]
    [Cycle 1]: 6.799e-05, [6]
        [build]: 3.66999e-06
        [elim_shapecalc]: 7.92e-06
        [elim_not_effective]: 1.312e-05
        [opt_reshape]: 5.74e-06
        [fold_const_symbol]: 8.98002e-06
        [renormalize]: 4.80009e-07
[validate]: 3.388e-05
Sums
    bootstrap                            :     0.000270s :  3.43%
    type_inference                       :     0.004068s : 51.67%
    event_method                         :     0.000017s :  0.22%
    auto_monad                           :     0.000062s :  0.79%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.42%
    rewriter_before_opt_a                :     0.000075s :  0.96%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000056s :  0.71%
    jit_opt_a.loop_unroll                :     0.000035s :  0.45%
    jit_opt_a.a_1                        :     0.000623s :  7.91%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.14%
    jit_opt_a.merge_forward              :     0.000007s :  0.09%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.37%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000578s :  7.34%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.24%
    jit_opt_a.cse                        :     0.000039s :  0.49%
    jit_opt_a.replace_applicator         :     0.000016s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000033s :  0.42%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000512s :  6.51%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000024s :  0.31%
    loop_unroll                          :     0.000418s :  5.31%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000015s :  0.19%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000047s :  0.60%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000015s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000471s :  5.98%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000034s :  0.43%


Time group info:
------[substitution.]   0.000192    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.87% :     0.000006s :      3: substitution.graph_param_transform
81.90% :     0.000157s :      4: substitution.inline
 1.84% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.80% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.02% :     0.000004s :      2: substitution.replace_old_param
 6.89% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004007     2
80.63% :     0.003231s :      1: type_inference.infer
19.37% :     0.000776s :      1: type_inference.specialize
------[replace.]   0.000069     6
65.98% :     0.000045s :      4: replace.inline
34.02% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000167     6
92.77% :     0.000154s :      4: match.inline
 7.23% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.42% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.08% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000002s :     10: predicate.cast_eliminate
 0.33% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.11% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.26% :     0.000002s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.64% :     0.000007s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.20% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.94% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.37% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.94% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.69% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.09% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     10: predicate.same_eliminate
 0.63% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.32% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.38% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.71% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.13% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000525    12
49.70% :     0.000261s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.06% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.24% :     0.000232s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013163    72
 0.38% :     0.000050s :      1: add_recomputation
 0.50% :     0.000065s :      1: auto_monad
 0.14% :     0.000018s :      1: auto_monad_reorder
 2.14% :     0.000282s :      1: bootstrap
 0.20% :     0.000027s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.80% :     0.005107s :      1: jit_opt_a
 1.06% :     0.000139s :      1: jit_opt_after_cconv
 0.42% :     0.000055s :      1: jit_opt_b
 3.24% :     0.000426s :      1: loop_unroll
 3.96% :     0.000521s :      1: mutable_eliminate
 6.32% :     0.000832s :     26: opt.transform.jit_opt_a
 0.35% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.64% :     0.000479s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000036s :      1: py_interpret_to_execute
 0.10% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000017s :      1: remove_dup_value
 2.28% :     0.000300s :      1: renormalize.infer
 2.06% :     0.000271s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000036s :      1: rewriter_after_opt_a
 0.59% :     0.000078s :      1: rewriter_before_opt_a
 0.59% :     0.000077s :      1: symbol_engine_optimizer
30.99% :     0.004079s :      1: type_inference


TotalTime = 0.0164601, [33]
[bootstrap]: 0.00025297
[type_inference]: 0.00387291
[event_method]: 1.593e-05
[auto_monad]: 4.156e-05
[graph_reusing]: 3.40998e-06
[pre_auto_parallel]: 2.22001e-06
[py_interpret_to_execute]: 2.82e-05
[rewriter_before_opt_a]: 6.252e-05
[expand_dump_flag]: 2.01998e-06
[jit_opt_a]: 0.00453305, [2]
    [Cycle 1]: 0.00125321, [27]
        [switch_simplify]: 4.595e-05
        [loop_unroll]: 2.887e-05
        [a_1]: 0.00045335
        [with_stream_mark]: 1.154e-05
        [recompute_prepare]: 7.05e-06
        [updatestate_depend_eliminate]: 2.73e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [parameter_eliminate]: 8.79983e-07
        [specialize_transform]: 5.82999e-06
        [updatestate_useless_node_eliminater]: 5.27999e-06
        [accelerated_algorithm]: 5.38002e-06
        [meta_shard_fg_expand]: 1.35999e-06
        [get_grad_eliminate_]: 5.57001e-06
        [merge_forward]: 2.93e-06
        [cell_reuse_recompute_pass]: 1.07998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.067e-05
        [j_node_and_user_rematch]: 8.08001e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 7.56001e-06
        [inline_without_move]: 5.27999e-06
        [renormalize]: 0.00043501
        [add_forward_monad_depend]: 3.98999e-06
        [auto_monad_grad]: 1.52001e-06
        [auto_monad_eliminator]: 9.70002e-06
        [cse]: 1.544e-05
        [replace_applicator]: 1.235e-05
    [Cycle 2]: 0.00031753, [27]
        [switch_simplify]: 5.49998e-06
        [loop_unroll]: 4.94998e-06
        [a_1]: 9.009e-05
        [with_stream_mark]: 7.97e-06
        [recompute_prepare]: 4.89998e-06
        [updatestate_depend_eliminate]: 2.35002e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 9.69972e-07
        [specialize_transform]: 5.09e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.07e-06
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.22001e-06
        [cell_reuse_recompute_pass]: 1.29998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.221e-05
        [j_node_and_user_rematch]: 8.17e-06
        [meta_fg_expand]: 1.30001e-06
        [replace_old_param]: 6.68e-06
        [inline_without_move]: 4.79e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.49e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 5.10001e-06
        [cse]: 1.03e-05
        [replace_applicator]: 5.57001e-06
[py_interpret_to_execute_after_opt_a]: 8.45999e-06
[rewriter_after_opt_a]: 2.352e-05
[convert_after_rewriter]: 6.27001e-06
[order_py_execute_after_rewriter]: 4.23999e-06
[mutable_eliminate]: 0.00047175
[jit_opt_b]: 4.71e-05, [1]
    [Cycle 1]: 4.097e-05, [2]
        [frontend_op_eliminate]: 1.56e-05
        [inline_after_opt_a]: 1.412e-05
[cconv]: 1.626e-05
[loop_unroll]: 0.00039856
[jit_opt_after_cconv]: 0.00014711, [1]
    [Cycle 1]: 0.00014106, [11]
        [c_1]: 3.693e-05
        [parameter_eliminate]: 2.61999e-06
        [updatestate_depend_eliminate]: 5.07e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 1.96998e-06
        [cse]: 1.6e-05
        [call_graph_tuple_transform]: 1.555e-05
        [tuple_list_get_item_eliminator]: 4.95001e-06
        [none_parameter_eliminate]: 7.89994e-07
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.39e-06
[remove_dup_value]: 6.93e-06
[partial_unused_args_eliminate]: 1.37999e-06
[environ_conv]: 3.25998e-06
[add_recomputation]: 3.06e-05
[cse_after_recomputation]: 1.89e-05, [1]
    [Cycle 1]: 1.369e-05, [1]
        [cse]: 8.03999e-06
[auto_monad_reorder]: 9.45001e-06
[get_jit_bprop_graph]: 1.50999e-06
[rewriter_after_jit_bprop_graph]: 4.84e-06
[opt_after_jit_grad]: 0.00044039
[symbol_engine_optimizer]: 6.803e-05, [1]
    [Cycle 1]: 6.22e-05, [6]
        [build]: 2.42001e-06
        [elim_shapecalc]: 7.08e-06
        [elim_not_effective]: 1.214e-05
        [opt_reshape]: 5.37001e-06
        [fold_const_symbol]: 7.92e-06
        [renormalize]: 5.20027e-07
[validate]: 2.462e-05
[backend_pass]: 8.49977e-07
[task_emit]: 0.00573108
[execute]: 7.06001e-06
Sums
    bootstrap                            :     0.000253s :  1.96%
    type_inference                       :     0.003873s : 29.95%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000042s :  0.32%
    graph_reusing                        :     0.000003s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.22%
    rewriter_before_opt_a                :     0.000063s :  0.48%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000051s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.26%
    jit_opt_a.a_1                        :     0.000543s :  4.20%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000014s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000435s :  3.36%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000026s :  0.20%
    jit_opt_a.replace_applicator         :     0.000018s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.07%
    rewriter_after_opt_a                 :     0.000024s :  0.18%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000472s :  3.65%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000016s :  0.13%
    loop_unroll                          :     0.000399s :  3.08%
    jit_opt_after_cconv.c_1              :     0.000037s :  0.29%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000007s :  0.05%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000003s :  0.03%
    add_recomputation                    :     0.000031s :  0.24%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000009s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000440s :  3.41%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000025s :  0.19%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005731s : 44.32%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000141    23
 1.14% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.32% :     0.000003s :      3: substitution.graph_param_transform
83.83% :     0.000118s :      4: substitution.inline
 1.88% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.74% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000003s :      2: substitution.replace_old_param
 5.49% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003826     2
81.53% :     0.003119s :      1: type_inference.infer
18.47% :     0.000707s :      1: type_inference.specialize
------[replace.]   0.000056     6
65.60% :     0.000037s :      4: replace.inline
34.40% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000123     6
94.56% :     0.000116s :      4: match.inline
 5.44% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.28% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 2.03% :     0.000002s :     10: predicate.arithmetic_simplify
 1.43% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.17% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.20% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.45% :     0.000002s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.41% :     0.000006s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.65% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000002s :     15: predicate.partial_eliminate
 1.38% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.53% :     0.000002s :     10: predicate.reshape_eliminate
 1.34% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.35% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 1.13% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.63% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.86% :     0.000009s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.32% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.30% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000471    12
47.45% :     0.000223s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.28% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.28% :     0.000223s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017750    76
 0.19% :     0.000033s :      1: add_recomputation
 0.25% :     0.000045s :      1: auto_monad
 0.07% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.49% :     0.000265s :      1: bootstrap
 0.11% :     0.000019s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
25.55% :     0.004536s :      1: jit_opt_a
 0.84% :     0.000150s :      1: jit_opt_after_cconv
 0.28% :     0.000050s :      1: jit_opt_b
 2.28% :     0.000405s :      1: loop_unroll
 2.70% :     0.000480s :      1: mutable_eliminate
 4.17% :     0.000740s :     26: opt.transform.jit_opt_a
 0.34% :     0.000060s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.52% :     0.000448s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000031s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000009s :      1: remove_dup_value
 1.21% :     0.000214s :      1: renormalize.infer
 1.21% :     0.000214s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000026s :      1: rewriter_after_opt_a
 0.37% :     0.000065s :      1: rewriter_before_opt_a
 0.40% :     0.000070s :      1: symbol_engine_optimizer
32.35% :     0.005743s :      1: task_emit
21.89% :     0.003885s :      1: type_inference
 0.22% :     0.000039s :      1: validate


TotalTime = 0.0118145, [30]
[bootstrap]: 0.00027741
[type_inference]: 0.00416849
[event_method]: 1.68e-05
[auto_monad]: 6.582e-05
[graph_reusing]: 5.50001e-06
[pre_auto_parallel]: 2.39999e-06
[py_interpret_to_execute]: 3.231e-05
[rewriter_before_opt_a]: 7.206e-05
[expand_dump_flag]: 3.29001e-06
[jit_opt_a]: 0.0050853, [2]
    [Cycle 1]: 0.0014711, [27]
        [switch_simplify]: 5.129e-05
        [loop_unroll]: 2.961e-05
        [a_1]: 0.00051614
        [with_stream_mark]: 1.776e-05
        [recompute_prepare]: 7.25e-06
        [updatestate_depend_eliminate]: 3.53e-06
        [updatestate_assign_eliminate]: 3.60998e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [parameter_eliminate]: 1.92999e-06
        [specialize_transform]: 5.77001e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 2.38998e-06
        [get_grad_eliminate_]: 5.02e-06
        [merge_forward]: 3.61001e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.42e-05
        [j_node_and_user_rematch]: 8.70999e-06
        [meta_fg_expand]: 2.53e-06
        [replace_old_param]: 9.19e-06
        [inline_without_move]: 5.21002e-06
        [renormalize]: 0.00053848
        [add_forward_monad_depend]: 5.39e-06
        [auto_monad_grad]: 2.34999e-06
        [auto_monad_eliminator]: 1.333e-05
        [cse]: 2.911e-05
        [replace_applicator]: 1.119e-05
    [Cycle 2]: 0.00032476, [27]
        [switch_simplify]: 5.71e-06
        [loop_unroll]: 5.08002e-06
        [a_1]: 9.059e-05
        [with_stream_mark]: 8.75999e-06
        [recompute_prepare]: 5.31002e-06
        [updatestate_depend_eliminate]: 2.63e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.11002e-06
        [specialize_transform]: 5.19e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.81002e-06
        [merge_forward]: 3.06999e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.298e-05
        [j_node_and_user_rematch]: 8.42e-06
        [meta_fg_expand]: 1.50999e-06
        [replace_old_param]: 7.15e-06
        [inline_without_move]: 4.76997e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.07998e-06
        [auto_monad_grad]: 7.29982e-07
        [auto_monad_eliminator]: 5.27001e-06
        [cse]: 9.92999e-06
        [replace_applicator]: 5.11002e-06
[py_interpret_to_execute_after_opt_a]: 1.001e-05
[rewriter_after_opt_a]: 4.039e-05
[convert_after_rewriter]: 7.46999e-06
[order_py_execute_after_rewriter]: 8.15e-06
[mutable_eliminate]: 0.00051531
[jit_opt_b]: 5.208e-05, [1]
    [Cycle 1]: 4.521e-05, [2]
        [frontend_op_eliminate]: 1.725e-05
        [inline_after_opt_a]: 1.61e-05
[cconv]: 2.467e-05
[loop_unroll]: 0.00042309
[jit_opt_after_cconv]: 0.00013772, [1]
    [Cycle 1]: 0.00013066, [11]
        [c_1]: 2.067e-05
        [parameter_eliminate]: 2.49999e-06
        [updatestate_depend_eliminate]: 5.99999e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [cse]: 1.745e-05
        [call_graph_tuple_transform]: 1.801e-05
        [tuple_list_get_item_eliminator]: 5.47001e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 4.10015e-07
        [switch_simplify]: 5.32999e-06
[remove_dup_value]: 1.434e-05
[partial_unused_args_eliminate]: 2.11e-06
[environ_conv]: 5.37001e-06
[add_recomputation]: 5.103e-05
[cse_after_recomputation]: 2.055e-05, [1]
    [Cycle 1]: 1.531e-05, [1]
        [cse]: 9.52999e-06
[auto_monad_reorder]: 1.583e-05
[get_jit_bprop_graph]: 1.49e-06
[rewriter_after_jit_bprop_graph]: 6.16e-06
[opt_after_jit_grad]: 0.00047603
[symbol_engine_optimizer]: 7.35e-05, [1]
    [Cycle 1]: 6.66e-05, [6]
        [build]: 3.82998e-06
        [elim_shapecalc]: 7.60998e-06
        [elim_not_effective]: 1.279e-05
        [opt_reshape]: 5.49e-06
        [fold_const_symbol]: 8.52e-06
        [renormalize]: 4.69998e-07
[validate]: 3.539e-05
Sums
    bootstrap                            :     0.000277s :  3.49%
    type_inference                       :     0.004168s : 52.42%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000066s :  0.83%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.41%
    rewriter_before_opt_a                :     0.000072s :  0.91%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000057s :  0.72%
    jit_opt_a.loop_unroll                :     0.000035s :  0.44%
    jit_opt_a.a_1                        :     0.000607s :  7.63%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000539s :  6.77%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.23%
    jit_opt_a.cse                        :     0.000039s :  0.49%
    jit_opt_a.replace_applicator         :     0.000016s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.13%
    rewriter_after_opt_a                 :     0.000040s :  0.51%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000008s :  0.10%
    mutable_eliminate                    :     0.000515s :  6.48%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.20%
    cconv                                :     0.000025s :  0.31%
    loop_unroll                          :     0.000423s :  5.32%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000051s :  0.64%
    cse_after_recomputation.cse          :     0.000010s :  0.12%
    auto_monad_reorder                   :     0.000016s :  0.20%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.08%
    opt_after_jit_grad                   :     0.000476s :  5.99%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000035s :  0.45%


Time group info:
------[substitution.]   0.000192    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.75% :     0.000005s :      3: substitution.graph_param_transform
83.48% :     0.000160s :      4: substitution.inline
 1.54% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.41% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.26% :     0.000004s :      2: substitution.replace_old_param
 5.81% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004102     2
81.17% :     0.003329s :      1: type_inference.infer
18.83% :     0.000772s :      1: type_inference.specialize
------[replace.]   0.000066     6
67.52% :     0.000044s :      4: replace.inline
32.48% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000168     6
94.08% :     0.000158s :      4: match.inline
 5.92% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.68% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000001s :     10: predicate.addn_zero_filter
 1.81% :     0.000002s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000001s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.13% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.22% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 6.36% :     0.000007s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.25% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.85% :     0.000002s :      3: predicate.mutable_eliminate
 0.58% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.29% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.25% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.39% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.64% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.87% :     0.000002s :     10: predicate.value_based_eliminate
 0.45% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000551    12
52.46% :     0.000289s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.87% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.67% :     0.000229s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013227    72
 0.41% :     0.000054s :      1: add_recomputation
 0.52% :     0.000069s :      1: auto_monad
 0.14% :     0.000018s :      1: auto_monad_reorder
 2.22% :     0.000294s :      1: bootstrap
 0.21% :     0.000027s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.47% :     0.005088s :      1: jit_opt_a
 1.06% :     0.000140s :      1: jit_opt_after_cconv
 0.41% :     0.000055s :      1: jit_opt_b
 3.25% :     0.000430s :      1: loop_unroll
 3.96% :     0.000524s :      1: mutable_eliminate
 6.14% :     0.000812s :     26: opt.transform.jit_opt_a
 0.35% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000027s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.18% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.66% :     0.000485s :      1: opt_after_jit_grad
 0.08% :     0.000011s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000035s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000017s :      1: remove_dup_value
 2.06% :     0.000272s :      1: renormalize.infer
 1.96% :     0.000260s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.33% :     0.000044s :      1: rewriter_after_opt_a
 0.57% :     0.000075s :      1: rewriter_before_opt_a
 0.58% :     0.000076s :      1: symbol_engine_optimizer
31.60% :     0.004180s :      1: type_inference


TotalTime = 0.0118101, [30]
[bootstrap]: 0.00030002
[type_inference]: 0.00418142
[event_method]: 1.7e-05
[auto_monad]: 6.278e-05
[graph_reusing]: 5.55001e-06
[pre_auto_parallel]: 2.69001e-06
[py_interpret_to_execute]: 3.522e-05
[rewriter_before_opt_a]: 7.725e-05
[expand_dump_flag]: 3.04999e-06
[jit_opt_a]: 0.00506684, [2]
    [Cycle 1]: 0.00151733, [27]
        [switch_simplify]: 5.25e-05
        [loop_unroll]: 3.127e-05
        [a_1]: 0.00052868
        [with_stream_mark]: 1.75e-05
        [recompute_prepare]: 7.03e-06
        [updatestate_depend_eliminate]: 3.63e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.78998e-06
        [parameter_eliminate]: 2.01e-06
        [specialize_transform]: 6.12001e-06
        [updatestate_useless_node_eliminater]: 5.93002e-06
        [accelerated_algorithm]: 6.74999e-06
        [meta_shard_fg_expand]: 2.36e-06
        [get_grad_eliminate_]: 6.17999e-06
        [merge_forward]: 3.88001e-06
        [cell_reuse_recompute_pass]: 1.06002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.601e-05
        [j_node_and_user_rematch]: 1.055e-05
        [meta_fg_expand]: 2.51998e-06
        [replace_old_param]: 8.48001e-06
        [inline_without_move]: 5.34998e-06
        [renormalize]: 0.00056222
        [add_forward_monad_depend]: 5.15999e-06
        [auto_monad_grad]: 2.39001e-06
        [auto_monad_eliminator]: 1.356e-05
        [cse]: 2.924e-05
        [replace_applicator]: 1.118e-05
    [Cycle 2]: 0.00033022, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 5.29e-06
        [a_1]: 9.037e-05
        [with_stream_mark]: 8.99e-06
        [recompute_prepare]: 5.67001e-06
        [updatestate_depend_eliminate]: 2.88e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 5.14e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.19e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.71002e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.246e-05
        [j_node_and_user_rematch]: 8.05e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 8.33999e-06
        [inline_without_move]: 5.51e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.45001e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 5.25001e-06
        [cse]: 1.039e-05
        [replace_applicator]: 5.80002e-06
[py_interpret_to_execute_after_opt_a]: 1.001e-05
[rewriter_after_opt_a]: 3.404e-05
[convert_after_rewriter]: 6.81001e-06
[order_py_execute_after_rewriter]: 4.67e-06
[mutable_eliminate]: 0.00050463
[jit_opt_b]: 5.085e-05, [1]
    [Cycle 1]: 4.389e-05, [2]
        [frontend_op_eliminate]: 1.705e-05
        [inline_after_opt_a]: 1.48e-05
[cconv]: 2.517e-05
[loop_unroll]: 0.00042776
[jit_opt_after_cconv]: 0.00014126, [1]
    [Cycle 1]: 0.00013487, [11]
        [c_1]: 2.066e-05
        [parameter_eliminate]: 2.69999e-06
        [updatestate_depend_eliminate]: 5.72001e-06
        [updatestate_assign_eliminate]: 2.49999e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [cse]: 1.61e-05
        [call_graph_tuple_transform]: 1.988e-05
        [tuple_list_get_item_eliminator]: 5.47001e-06
        [none_parameter_eliminate]: 1.69e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 5.17999e-06
[remove_dup_value]: 1.378e-05
[partial_unused_args_eliminate]: 2.51998e-06
[environ_conv]: 5.35001e-06
[add_recomputation]: 4.908e-05
[cse_after_recomputation]: 1.991e-05, [1]
    [Cycle 1]: 1.456e-05, [1]
        [cse]: 8.77999e-06
[auto_monad_reorder]: 1.526e-05
[get_jit_bprop_graph]: 1.72001e-06
[rewriter_after_jit_bprop_graph]: 5.02e-06
[opt_after_jit_grad]: 0.00045707
[symbol_engine_optimizer]: 7.506e-05, [1]
    [Cycle 1]: 6.886e-05, [6]
        [build]: 3.93001e-06
        [elim_shapecalc]: 8.55999e-06
        [elim_not_effective]: 1.363e-05
        [opt_reshape]: 6.21998e-06
        [fold_const_symbol]: 8.33999e-06
        [renormalize]: 4.80009e-07
[validate]: 3.388e-05
Sums
    bootstrap                            :     0.000300s :  3.75%
    type_inference                       :     0.004181s : 52.25%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000063s :  0.78%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.44%
    rewriter_before_opt_a                :     0.000077s :  0.97%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000058s :  0.73%
    jit_opt_a.loop_unroll                :     0.000037s :  0.46%
    jit_opt_a.a_1                        :     0.000619s :  7.74%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.14%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.36%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.21%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000562s :  7.03%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.24%
    jit_opt_a.cse                        :     0.000040s :  0.50%
    jit_opt_a.replace_applicator         :     0.000017s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.13%
    rewriter_after_opt_a                 :     0.000034s :  0.43%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000505s :  6.31%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000025s :  0.31%
    loop_unroll                          :     0.000428s :  5.35%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.25%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000049s :  0.61%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000457s :  5.71%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000034s :  0.42%


Time group info:
------[substitution.]   0.000194    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.85% :     0.000006s :      3: substitution.graph_param_transform
82.69% :     0.000161s :      4: substitution.inline
 2.12% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.32% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.10% :     0.000004s :      2: substitution.replace_old_param
 6.20% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004115     2
80.30% :     0.003304s :      1: type_inference.infer
19.70% :     0.000811s :      1: type_inference.specialize
------[replace.]   0.000069     6
66.99% :     0.000046s :      4: replace.inline
33.01% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000169     6
93.59% :     0.000158s :      4: match.inline
 6.41% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.62% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.78% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.18% :     0.000001s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.17% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.91% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.33% :     0.000007s :     22: predicate.inline
 1.02% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.51% :     0.000002s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.57% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.17% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.21% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 1.77% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 1.41% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.58% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.52% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.23% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.37% :     0.000009s :     53: predicate.switch_simplify
 1.08% :     0.000001s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.90% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.94% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000569    12
50.55% :     0.000288s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.47% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.98% :     0.000250s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013268    72
 0.39% :     0.000052s :      1: add_recomputation
 0.50% :     0.000066s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 2.41% :     0.000319s :      1: bootstrap
 0.21% :     0.000028s :      1: cconv
 0.08% :     0.000011s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.21% :     0.005070s :      1: jit_opt_a
 1.09% :     0.000144s :      1: jit_opt_after_cconv
 0.40% :     0.000054s :      1: jit_opt_b
 3.29% :     0.000436s :      1: loop_unroll
 3.87% :     0.000513s :      1: mutable_eliminate
 6.29% :     0.000834s :     26: opt.transform.jit_opt_a
 0.36% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000026s :      4: opt.transform.jit_opt_b
 0.10% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.50% :     0.000465s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.04% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000038s :      1: py_interpret_to_execute
 0.10% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.18% :     0.000290s :      1: renormalize.infer
 2.00% :     0.000265s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000037s :      1: rewriter_after_opt_a
 0.61% :     0.000080s :      1: rewriter_before_opt_a
 0.59% :     0.000078s :      1: symbol_engine_optimizer
31.61% :     0.004194s :      1: type_inference


TotalTime = 0.011825, [30]
[bootstrap]: 0.00025025
[type_inference]: 0.00392976
[event_method]: 1.72e-05
[auto_monad]: 6.228e-05
[graph_reusing]: 5.35001e-06
[pre_auto_parallel]: 2.44999e-06
[py_interpret_to_execute]: 3.09e-05
[rewriter_before_opt_a]: 7.056e-05
[expand_dump_flag]: 2.78e-06
[jit_opt_a]: 0.00520747, [2]
    [Cycle 1]: 0.00159866, [27]
        [switch_simplify]: 5.082e-05
        [loop_unroll]: 3.021e-05
        [a_1]: 0.00055274
        [with_stream_mark]: 1.984e-05
        [recompute_prepare]: 8.39998e-06
        [updatestate_depend_eliminate]: 3.34001e-06
        [updatestate_assign_eliminate]: 3.18e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [parameter_eliminate]: 2.27001e-06
        [specialize_transform]: 6.54001e-06
        [updatestate_useless_node_eliminater]: 5.68002e-06
        [accelerated_algorithm]: 5.94999e-06
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 5.86998e-06
        [merge_forward]: 3.99002e-06
        [cell_reuse_recompute_pass]: 1.14998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.474e-05
        [j_node_and_user_rematch]: 8.82e-06
        [meta_fg_expand]: 2.27001e-06
        [replace_old_param]: 8.77e-06
        [inline_without_move]: 5.45001e-06
        [renormalize]: 0.00061027
        [add_forward_monad_depend]: 5.57001e-06
        [auto_monad_grad]: 1.95001e-06
        [auto_monad_eliminator]: 1.459e-05
        [cse]: 2.951e-05
        [replace_applicator]: 1.317e-05
    [Cycle 2]: 0.00034119, [27]
        [switch_simplify]: 5.82999e-06
        [loop_unroll]: 5.37001e-06
        [a_1]: 9.738e-05
        [with_stream_mark]: 1.145e-05
        [recompute_prepare]: 4.98001e-06
        [updatestate_depend_eliminate]: 2.81e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.98002e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 4.95999e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 1.40001e-06
        [get_grad_eliminate_]: 5.06997e-06
        [merge_forward]: 2.64999e-06
        [cell_reuse_recompute_pass]: 1.55999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.333e-05
        [j_node_and_user_rematch]: 8.03001e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 7.51999e-06
        [inline_without_move]: 4.78001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 5.77001e-06
        [cse]: 1.073e-05
        [replace_applicator]: 5.19e-06
[py_interpret_to_execute_after_opt_a]: 1.198e-05
[rewriter_after_opt_a]: 3.441e-05
[convert_after_rewriter]: 6.98998e-06
[order_py_execute_after_rewriter]: 4.72e-06
[mutable_eliminate]: 0.00066635
[jit_opt_b]: 5.801e-05, [1]
    [Cycle 1]: 5.059e-05, [2]
        [frontend_op_eliminate]: 2.001e-05
        [inline_after_opt_a]: 1.74e-05
[cconv]: 2.766e-05
[loop_unroll]: 0.00043132
[jit_opt_after_cconv]: 0.00014716, [1]
    [Cycle 1]: 0.0001403, [11]
        [c_1]: 2.068e-05
        [parameter_eliminate]: 2.58003e-06
        [updatestate_depend_eliminate]: 6.58998e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.25002e-06
        [cse]: 2.075e-05
        [call_graph_tuple_transform]: 1.937e-05
        [tuple_list_get_item_eliminator]: 5.60001e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 6.09001e-06
[remove_dup_value]: 1.418e-05
[partial_unused_args_eliminate]: 2.27001e-06
[environ_conv]: 5.45001e-06
[add_recomputation]: 4.866e-05
[cse_after_recomputation]: 2.024e-05, [1]
    [Cycle 1]: 1.497e-05, [1]
        [cse]: 9.25001e-06
[auto_monad_reorder]: 1.455e-05
[get_jit_bprop_graph]: 2.37999e-06
[rewriter_after_jit_bprop_graph]: 5.97999e-06
[opt_after_jit_grad]: 0.00046899
[symbol_engine_optimizer]: 7.217e-05, [1]
    [Cycle 1]: 6.584e-05, [6]
        [build]: 3.68e-06
        [elim_shapecalc]: 7.21001e-06
        [elim_not_effective]: 1.236e-05
        [opt_reshape]: 6.32001e-06
        [fold_const_symbol]: 8.76997e-06
        [renormalize]: 4.09986e-07
[validate]: 3.699e-05
Sums
    bootstrap                            :     0.000250s :  3.14%
    type_inference                       :     0.003930s : 49.33%
    event_method                         :     0.000017s :  0.22%
    auto_monad                           :     0.000062s :  0.78%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000031s :  0.39%
    rewriter_before_opt_a                :     0.000071s :  0.89%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000057s :  0.71%
    jit_opt_a.loop_unroll                :     0.000036s :  0.45%
    jit_opt_a.a_1                        :     0.000650s :  8.16%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.39%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.14%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000610s :  7.66%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.26%
    jit_opt_a.cse                        :     0.000040s :  0.51%
    jit_opt_a.replace_applicator         :     0.000018s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.15%
    rewriter_after_opt_a                 :     0.000034s :  0.43%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000666s :  8.37%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.25%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.22%
    cconv                                :     0.000028s :  0.35%
    loop_unroll                          :     0.000431s :  5.41%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000021s :  0.26%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.08%
    remove_dup_value                     :     0.000014s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000049s :  0.61%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000015s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.08%
    opt_after_jit_grad                   :     0.000469s :  5.89%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000037s :  0.46%


Time group info:
------[substitution.]   0.000208    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.71% :     0.000006s :      3: substitution.graph_param_transform
84.04% :     0.000175s :      4: substitution.inline
 1.35% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.40% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000004s :      2: substitution.replace_old_param
 5.96% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003875     2
80.57% :     0.003122s :      1: type_inference.infer
19.43% :     0.000753s :      1: type_inference.specialize
------[replace.]   0.000069     6
67.80% :     0.000047s :      4: replace.inline
32.20% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000184     6
93.88% :     0.000173s :      4: match.inline
 6.12% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.55% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.20% :     0.000002s :     10: predicate.addn_zero_filter
 2.45% :     0.000003s :     10: predicate.arithmetic_simplify
 1.15% :     0.000001s :     10: predicate.cast_eliminate
 0.69% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.16% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.79% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.46% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.03% :     0.000001s :     10: predicate.environ_get_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.73% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.51% :     0.000002s :      6: predicate.less_batch_normalization
 1.66% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.58% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.71% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.04% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 2.04% :     0.000003s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.06% :     0.000003s :     10: predicate.reduce_eliminate
 1.31% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.51% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.02% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.62% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.37% :     0.000002s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.08% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 1.91% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.28% :     0.000009s :     53: predicate.switch_simplify
 1.10% :     0.000001s :     10: predicate.tile_eliminate
 1.65% :     0.000002s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.04% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.94% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000529    12
49.71% :     0.000263s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.97% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.31% :     0.000229s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013362    72
 0.39% :     0.000051s :      1: add_recomputation
 0.49% :     0.000066s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 1.98% :     0.000265s :      1: bootstrap
 0.23% :     0.000030s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
39.00% :     0.005211s :      1: jit_opt_a
 1.13% :     0.000151s :      1: jit_opt_after_cconv
 0.46% :     0.000061s :      1: jit_opt_b
 3.29% :     0.000439s :      1: loop_unroll
 5.06% :     0.000677s :      1: mutable_eliminate
 6.44% :     0.000861s :     26: opt.transform.jit_opt_a
 0.36% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000030s :      4: opt.transform.jit_opt_b
 0.10% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.57% :     0.000477s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000034s :      1: py_interpret_to_execute
 0.11% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000017s :      1: remove_dup_value
 2.39% :     0.000319s :      1: renormalize.infer
 2.12% :     0.000283s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.29% :     0.000038s :      1: rewriter_after_opt_a
 0.55% :     0.000073s :      1: rewriter_before_opt_a
 0.56% :     0.000075s :      1: symbol_engine_optimizer
29.49% :     0.003940s :      1: type_inference


TotalTime = 0.0748199, [33]
[bootstrap]: 0.00055433
[type_inference]: 0.00594956
[event_method]: 2.004e-05
[auto_monad]: 6.739e-05
[graph_reusing]: 6.53e-06
[pre_auto_parallel]: 2.98998e-06
[py_interpret_to_execute]: 3.662e-05
[rewriter_before_opt_a]: 0.00011447
[expand_dump_flag]: 3.29001e-06
[jit_opt_a]: 0.0539868, [2]
    [Cycle 1]: 0.00219285, [27]
        [switch_simplify]: 0.00012198
        [loop_unroll]: 3.042e-05
        [a_1]: 0.00068778
        [with_stream_mark]: 3.075e-05
        [recompute_prepare]: 1.045e-05
        [updatestate_depend_eliminate]: 5.12e-06
        [updatestate_assign_eliminate]: 2.77002e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 3.03998e-06
        [specialize_transform]: 8.54e-06
        [updatestate_useless_node_eliminater]: 5.69999e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 2.72001e-06
        [get_grad_eliminate_]: 6.19001e-06
        [merge_forward]: 4.22e-06
        [cell_reuse_recompute_pass]: 2.44001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.761e-05
        [j_node_and_user_rematch]: 9.97999e-06
        [meta_fg_expand]: 2.88e-06
        [replace_old_param]: 1.075e-05
        [inline_without_move]: 5.54e-06
        [renormalize]: 0.00089677
        [add_forward_monad_depend]: 9.90002e-06
        [auto_monad_grad]: 2.82002e-06
        [auto_monad_eliminator]: 2.041e-05
        [cse]: 3.387e-05
        [replace_applicator]: 1.88e-05
    [Cycle 2]: 0.00035411, [27]
        [switch_simplify]: 6.53e-06
        [loop_unroll]: 4.62e-06
        [a_1]: 9.362e-05
        [with_stream_mark]: 1.755e-05
        [recompute_prepare]: 5.63002e-06
        [updatestate_depend_eliminate]: 4.06001e-06
        [updatestate_assign_eliminate]: 2.18002e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [parameter_eliminate]: 2.03997e-06
        [specialize_transform]: 4.82e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 2.54999e-06
        [get_grad_eliminate_]: 4.51002e-06
        [merge_forward]: 3.18e-06
        [cell_reuse_recompute_pass]: 2.94999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.824e-05
        [j_node_and_user_rematch]: 8.2e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 9.37001e-06
        [inline_without_move]: 4.66002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.32999e-06
        [auto_monad_grad]: 1.16002e-06
        [auto_monad_eliminator]: 6.86001e-06
        [cse]: 1.305e-05
        [replace_applicator]: 5.40001e-06
[py_interpret_to_execute_after_opt_a]: 1.707e-05
[rewriter_after_opt_a]: 4.23e-05
[convert_after_rewriter]: 8.20999e-06
[order_py_execute_after_rewriter]: 6.04001e-06
[mutable_eliminate]: 0.0009779
[jit_opt_b]: 5.647e-05, [1]
    [Cycle 1]: 4.683e-05, [2]
        [frontend_op_eliminate]: 1.689e-05
        [inline_after_opt_a]: 1.754e-05
[cconv]: 4.361e-05
[loop_unroll]: 0.00061603
[jit_opt_after_cconv]: 0.00021438, [1]
    [Cycle 1]: 0.00020509, [11]
        [c_1]: 3.912e-05
        [parameter_eliminate]: 7.66001e-06
        [updatestate_depend_eliminate]: 1.403e-05
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 3.16001e-06
        [cse]: 4.129e-05
        [call_graph_tuple_transform]: 2.759e-05
        [tuple_list_get_item_eliminator]: 5.57001e-06
        [none_parameter_eliminate]: 1.66002e-06
        [renormalize]: 1.39e-06
        [switch_simplify]: 5.32001e-06
[remove_dup_value]: 1.567e-05
[partial_unused_args_eliminate]: 4.55001e-06
[environ_conv]: 7.95e-06
[add_recomputation]: 6.643e-05
[cse_after_recomputation]: 3.117e-05, [1]
    [Cycle 1]: 2.359e-05, [1]
        [cse]: 1.364e-05
[auto_monad_reorder]: 1.699e-05
[get_jit_bprop_graph]: 2.41998e-06
[rewriter_after_jit_bprop_graph]: 1.125e-05
[opt_after_jit_grad]: 0.00062454
[symbol_engine_optimizer]: 8.336e-05, [1]
    [Cycle 1]: 7.581e-05, [6]
        [build]: 5.35999e-06
        [elim_shapecalc]: 8.85001e-06
        [elim_not_effective]: 1.561e-05
        [opt_reshape]: 5.64e-06
        [fold_const_symbol]: 9.79e-06
        [renormalize]: 4.69998e-07
[validate]: 4.885e-05
[backend_pass]: 1.00999e-06
[task_emit]: 0.0106862
[execute]: 1.072e-05
Sums
    bootstrap                            :     0.000554s :  2.47%
    type_inference                       :     0.005950s : 26.56%
    event_method                         :     0.000020s :  0.09%
    auto_monad                           :     0.000067s :  0.30%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000037s :  0.16%
    rewriter_before_opt_a                :     0.000114s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000129s :  0.57%
    jit_opt_a.loop_unroll                :     0.000035s :  0.16%
    jit_opt_a.a_1                        :     0.000781s :  3.49%
    jit_opt_a.with_stream_mark           :     0.000048s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000897s :  4.00%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.12%
    jit_opt_a.cse                        :     0.000047s :  0.21%
    jit_opt_a.replace_applicator         :     0.000024s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.08%
    rewriter_after_opt_a                 :     0.000042s :  0.19%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000978s :  4.36%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.08%
    cconv                                :     0.000044s :  0.19%
    loop_unroll                          :     0.000616s :  2.75%
    jit_opt_after_cconv.c_1              :     0.000039s :  0.17%
    jit_opt_after_cconv.parameter_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000014s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000041s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000016s :  0.07%
    partial_unused_args_eliminate        :     0.000005s :  0.02%
    environ_conv                         :     0.000008s :  0.04%
    add_recomputation                    :     0.000066s :  0.30%
    cse_after_recomputation.cse          :     0.000014s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.05%
    opt_after_jit_grad                   :     0.000625s :  2.79%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000049s :  0.22%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010686s : 47.70%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000304    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.34% :     0.000007s :      3: substitution.graph_param_transform
85.88% :     0.000262s :      4: substitution.inline
 1.35% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.83% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.32% :     0.000007s :      2: substitution.replace_old_param
 5.06% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005863     2
85.32% :     0.005002s :      1: type_inference.infer
14.68% :     0.000861s :      1: type_inference.specialize
------[replace.]   0.000095     6
75.69% :     0.000072s :      4: replace.inline
24.31% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000272     6
94.95% :     0.000258s :      4: match.inline
 5.05% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000140   691
 1.62% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.74% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.86% :     0.000001s :     10: predicate.addn_check_dump
 1.21% :     0.000002s :     10: predicate.addn_zero_filter
 2.66% :     0.000004s :     10: predicate.arithmetic_simplify
 1.72% :     0.000002s :     10: predicate.cast_eliminate
 0.35% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.87% :     0.000001s :     10: predicate.compare_switch_simplify
 1.60% :     0.000002s :     10: predicate.depend_value_elim
 0.94% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.10% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.07% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.26% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.54% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.87% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000002s :     10: predicate.environ_get_eliminate
 1.22% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 6.29% :     0.000009s :     22: predicate.inline
 0.73% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 2.51% :     0.000004s :      3: predicate.loop_unroll_after_grad
 3.30% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.84% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.87% :     0.000001s :     10: predicate.merge_addn
 0.87% :     0.000001s :     10: predicate.minmaximum_grad
 2.78% :     0.000004s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 1.84% :     0.000003s :     15: predicate.partial_eliminate
 1.08% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.62% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.59% :     0.000001s :      3: predicate.reset_defer_inline
 1.74% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.21% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.07% :     0.000001s :     10: predicate.same_eliminate
 0.96% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.75% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.18% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.04% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 1.82% :     0.000003s :     16: predicate.switch_defer_inline
 1.89% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.03% :     0.000011s :     53: predicate.switch_simplify
 1.25% :     0.000002s :     10: predicate.tile_eliminate
 1.24% :     0.000002s :     10: predicate.transpose_eliminate
 1.14% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.46% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.84% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.66% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.84% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.38% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.34% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.27% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000704    12
52.56% :     0.000370s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.64% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.80% :     0.000273s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.076896    76
 0.09% :     0.000070s :      1: add_recomputation
 0.09% :     0.000071s :      1: auto_monad
 0.03% :     0.000020s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.80% :     0.000612s :      1: bootstrap
 0.06% :     0.000047s :      1: cconv
 0.01% :     0.000011s :      1: convert_after_rewriter
 0.04% :     0.000034s :      1: cse_after_recomputation
 0.01% :     0.000011s :      1: environ_conv
 0.03% :     0.000025s :      1: event_method
 0.02% :     0.000019s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000009s :      1: graph_reusing
70.21% :     0.053991s :      1: jit_opt_a
 0.28% :     0.000218s :      1: jit_opt_after_cconv
 0.08% :     0.000060s :      1: jit_opt_b
 0.82% :     0.000633s :      1: loop_unroll
 1.30% :     0.000997s :      1: mutable_eliminate
 1.35% :     0.001037s :     26: opt.transform.jit_opt_a
 0.10% :     0.000074s :      4: opt.transform.jit_opt_after_cconv
 0.03% :     0.000026s :      4: opt.transform.jit_opt_b
 0.03% :     0.000023s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000026s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.05% :     0.000036s :      4: opt.transform.symbol_engine_opt
 0.83% :     0.000636s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000008s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000052s :      1: py_interpret_to_execute
 0.03% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.19% :     0.000143s :      1: remove_dup_value
 0.66% :     0.000510s :      1: renormalize.infer
 0.49% :     0.000377s :      1: renormalize.specialize
 0.02% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.06% :     0.000046s :      1: rewriter_after_opt_a
 0.16% :     0.000120s :      1: rewriter_before_opt_a
 0.11% :     0.000086s :      1: symbol_engine_optimizer
13.95% :     0.010727s :      1: task_emit
 7.77% :     0.005975s :      1: type_inference
 0.10% :     0.000075s :      1: validate


TotalTime = 0.0111758, [30]
[bootstrap]: 0.00026353
[type_inference]: 0.00403113
[event_method]: 2.792e-05
[auto_monad]: 6.361e-05
[graph_reusing]: 5.56e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 3.377e-05
[rewriter_before_opt_a]: 7.16e-05
[expand_dump_flag]: 3.03e-06
[jit_opt_a]: 0.00475932, [2]
    [Cycle 1]: 0.00143952, [27]
        [switch_simplify]: 5.018e-05
        [loop_unroll]: 3.068e-05
        [a_1]: 0.0005052
        [with_stream_mark]: 1.617e-05
        [recompute_prepare]: 7.03e-06
        [updatestate_depend_eliminate]: 3.15998e-06
        [updatestate_assign_eliminate]: 2.79999e-06
        [updatestate_loads_eliminate]: 2.76999e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 6.03002e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 5.79999e-06
        [meta_shard_fg_expand]: 1.92001e-06
        [get_grad_eliminate_]: 5.33002e-06
        [merge_forward]: 3.65e-06
        [cell_reuse_recompute_pass]: 1.02998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.459e-05
        [j_node_and_user_rematch]: 9.62999e-06
        [meta_fg_expand]: 2.69999e-06
        [replace_old_param]: 1.057e-05
        [inline_without_move]: 5.83997e-06
        [renormalize]: 0.00046948
        [add_forward_monad_depend]: 5.10999e-06
        [auto_monad_grad]: 1.94999e-06
        [auto_monad_eliminator]: 1.286e-05
        [cse]: 2.385e-05
        [replace_applicator]: 1.127e-05
    [Cycle 2]: 0.00031807, [27]
        [switch_simplify]: 5.59e-06
        [loop_unroll]: 5.25999e-06
        [a_1]: 9.017e-05
        [with_stream_mark]: 8.85999e-06
        [recompute_prepare]: 5.17999e-06
        [updatestate_depend_eliminate]: 2.59999e-06
        [updatestate_assign_eliminate]: 2.15002e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 9.99979e-07
        [specialize_transform]: 4.79002e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 5.13002e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.75999e-06
        [merge_forward]: 2.56e-06
        [cell_reuse_recompute_pass]: 1.19998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.215e-05
        [j_node_and_user_rematch]: 7.85e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 7.05e-06
        [inline_without_move]: 4.60999e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.21002e-06
        [auto_monad_grad]: 7.79983e-07
        [auto_monad_eliminator]: 4.61002e-06
        [cse]: 9.47001e-06
        [replace_applicator]: 4.87998e-06
[py_interpret_to_execute_after_opt_a]: 8.77999e-06
[rewriter_after_opt_a]: 3.003e-05
[convert_after_rewriter]: 7.1e-06
[order_py_execute_after_rewriter]: 4.95001e-06
[mutable_eliminate]: 0.00045689
[jit_opt_b]: 4.985e-05, [1]
    [Cycle 1]: 4.31e-05, [2]
        [frontend_op_eliminate]: 1.67e-05
        [inline_after_opt_a]: 1.479e-05
[cconv]: 2.274e-05
[loop_unroll]: 0.00040097
[jit_opt_after_cconv]: 0.00013264, [1]
    [Cycle 1]: 0.00012647, [11]
        [c_1]: 2.053e-05
        [parameter_eliminate]: 2.49001e-06
        [updatestate_depend_eliminate]: 5.02e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 1.547e-05
        [call_graph_tuple_transform]: 1.724e-05
        [tuple_list_get_item_eliminator]: 5.28002e-06
        [none_parameter_eliminate]: 1.67999e-06
        [renormalize]: 3.80009e-07
        [switch_simplify]: 5.44e-06
[remove_dup_value]: 1.334e-05
[partial_unused_args_eliminate]: 2.10002e-06
[environ_conv]: 4.67e-06
[add_recomputation]: 4.398e-05
[cse_after_recomputation]: 2.049e-05, [1]
    [Cycle 1]: 1.505e-05, [1]
        [cse]: 9.10999e-06
[auto_monad_reorder]: 1.353e-05
[get_jit_bprop_graph]: 1.62001e-06
[rewriter_after_jit_bprop_graph]: 4.92999e-06
[opt_after_jit_grad]: 0.00043638
[symbol_engine_optimizer]: 7.28e-05, [1]
    [Cycle 1]: 6.655e-05, [6]
        [build]: 3.70998e-06
        [elim_shapecalc]: 7.76001e-06
        [elim_not_effective]: 1.257e-05
        [opt_reshape]: 5.70001e-06
        [fold_const_symbol]: 8.77e-06
        [renormalize]: 4.80009e-07
[validate]: 2.984e-05
Sums
    bootstrap                            :     0.000264s :  3.49%
    type_inference                       :     0.004031s : 53.32%
    event_method                         :     0.000028s :  0.37%
    auto_monad                           :     0.000064s :  0.84%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.45%
    rewriter_before_opt_a                :     0.000072s :  0.95%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000056s :  0.74%
    jit_opt_a.loop_unroll                :     0.000036s :  0.48%
    jit_opt_a.a_1                        :     0.000595s :  7.88%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.06%
    jit_opt_a.replace_old_param          :     0.000018s :  0.23%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000470s :  6.21%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.23%
    jit_opt_a.cse                        :     0.000033s :  0.44%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.12%
    rewriter_after_opt_a                 :     0.000030s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000457s :  6.04%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.20%
    cconv                                :     0.000023s :  0.30%
    loop_unroll                          :     0.000401s :  5.30%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000015s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000013s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000044s :  0.58%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000014s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000436s :  5.77%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000030s :  0.39%


Time group info:
------[substitution.]   0.000178    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000001s :      2: substitution.fold_const_symbol
 2.53% :     0.000005s :      3: substitution.graph_param_transform
82.53% :     0.000147s :      4: substitution.inline
 1.73% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.64% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.05% :     0.000004s :      2: substitution.replace_old_param
 6.74% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003974     2
81.12% :     0.003224s :      1: type_inference.infer
18.88% :     0.000750s :      1: type_inference.specialize
------[replace.]   0.000065     6
68.14% :     0.000044s :      4: replace.inline
31.86% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000155     6
92.99% :     0.000144s :      4: match.inline
 7.01% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000001s :     10: predicate.addn_zero_filter
 2.05% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000001s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.44% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.54% :     0.000006s :     22: predicate.inline
 1.02% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.14% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.50% :     0.000003s :     15: predicate.partial_eliminate
 1.30% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.43% :     0.000002s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.59% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.72% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000505    12
50.95% :     0.000257s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.22% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.82% :     0.000221s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012511    72
 0.37% :     0.000047s :      1: add_recomputation
 0.53% :     0.000067s :      1: auto_monad
 0.13% :     0.000016s :      1: auto_monad_reorder
 2.24% :     0.000280s :      1: bootstrap
 0.20% :     0.000025s :      1: cconv
 0.08% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.25% :     0.000032s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.06% :     0.004762s :      1: jit_opt_a
 1.08% :     0.000135s :      1: jit_opt_after_cconv
 0.42% :     0.000052s :      1: jit_opt_b
 3.26% :     0.000408s :      1: loop_unroll
 3.72% :     0.000465s :      1: mutable_eliminate
 6.41% :     0.000802s :     26: opt.transform.jit_opt_a
 0.36% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.55% :     0.000444s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.29% :     0.000036s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000016s :      1: remove_dup_value
 1.82% :     0.000228s :      1: renormalize.infer
 1.88% :     0.000235s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000033s :      1: rewriter_after_opt_a
 0.59% :     0.000074s :      1: rewriter_before_opt_a
 0.60% :     0.000075s :      1: symbol_engine_optimizer
32.31% :     0.004042s :      1: type_inference


TotalTime = 0.0107711, [30]
[bootstrap]: 0.00024682
[type_inference]: 0.00380444
[event_method]: 1.557e-05
[auto_monad]: 5.752e-05
[graph_reusing]: 5.27999e-06
[pre_auto_parallel]: 2.64999e-06
[py_interpret_to_execute]: 3.005e-05
[rewriter_before_opt_a]: 6.788e-05
[expand_dump_flag]: 2.74999e-06
[jit_opt_a]: 0.0046055, [2]
    [Cycle 1]: 0.00133251, [27]
        [switch_simplify]: 4.676e-05
        [loop_unroll]: 3.003e-05
        [a_1]: 0.00048245
        [with_stream_mark]: 1.248e-05
        [recompute_prepare]: 6.66e-06
        [updatestate_depend_eliminate]: 2.96999e-06
        [updatestate_assign_eliminate]: 2.63e-06
        [updatestate_loads_eliminate]: 2.34999e-06
        [parameter_eliminate]: 1.78002e-06
        [specialize_transform]: 5.67999e-06
        [updatestate_useless_node_eliminater]: 5.18002e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 1.70001e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 2.93e-06
        [cell_reuse_recompute_pass]: 1.10999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.24e-05
        [j_node_and_user_rematch]: 8.86002e-06
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 7.92e-06
        [inline_without_move]: 5.30001e-06
        [renormalize]: 0.00046549
        [add_forward_monad_depend]: 4.80001e-06
        [auto_monad_grad]: 1.97001e-06
        [auto_monad_eliminator]: 1.235e-05
        [cse]: 2.294e-05
        [replace_applicator]: 1.111e-05
    [Cycle 2]: 0.00032385, [27]
        [switch_simplify]: 5.59e-06
        [loop_unroll]: 4.84e-06
        [a_1]: 8.994e-05
        [with_stream_mark]: 7.93999e-06
        [recompute_prepare]: 4.84e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 5.56e-06
        [updatestate_useless_node_eliminater]: 5.77999e-06
        [accelerated_algorithm]: 6.10002e-06
        [meta_shard_fg_expand]: 1.17999e-06
        [get_grad_eliminate_]: 4.77e-06
        [merge_forward]: 2.39001e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.191e-05
        [j_node_and_user_rematch]: 8.00999e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 7.11001e-06
        [inline_without_move]: 4.80999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.17e-06
        [cse]: 1.003e-05
        [replace_applicator]: 5.50001e-06
[py_interpret_to_execute_after_opt_a]: 8.2e-06
[rewriter_after_opt_a]: 3.022e-05
[convert_after_rewriter]: 6.61e-06
[order_py_execute_after_rewriter]: 4.85999e-06
[mutable_eliminate]: 0.00046127
[jit_opt_b]: 5.117e-05, [1]
    [Cycle 1]: 4.445e-05, [2]
        [frontend_op_eliminate]: 1.79e-05
        [inline_after_opt_a]: 1.518e-05
[cconv]: 2.175e-05
[loop_unroll]: 0.00040483
[jit_opt_after_cconv]: 0.00013752, [1]
    [Cycle 1]: 0.00013091, [11]
        [c_1]: 2.136e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 5.24e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [cse]: 1.62e-05
        [call_graph_tuple_transform]: 1.782e-05
        [tuple_list_get_item_eliminator]: 5.38002e-06
        [none_parameter_eliminate]: 1.20999e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.77001e-06
[remove_dup_value]: 1.262e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 4.92999e-06
[add_recomputation]: 4.236e-05
[cse_after_recomputation]: 2.022e-05, [1]
    [Cycle 1]: 1.449e-05, [1]
        [cse]: 8.85999e-06
[auto_monad_reorder]: 1.357e-05
[get_jit_bprop_graph]: 1.63002e-06
[rewriter_after_jit_bprop_graph]: 4.62e-06
[opt_after_jit_grad]: 0.00044963
[symbol_engine_optimizer]: 7.32e-05, [1]
    [Cycle 1]: 6.652e-05, [6]
        [build]: 3.21001e-06
        [elim_shapecalc]: 7.7e-06
        [elim_not_effective]: 1.29e-05
        [opt_reshape]: 5.91e-06
        [fold_const_symbol]: 8.76997e-06
        [renormalize]: 3.69997e-07
[validate]: 2.927e-05
Sums
    bootstrap                            :     0.000247s :  3.40%
    type_inference                       :     0.003804s : 52.35%
    event_method                         :     0.000016s :  0.21%
    auto_monad                           :     0.000058s :  0.79%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.04%
    py_interpret_to_execute              :     0.000030s :  0.41%
    rewriter_before_opt_a                :     0.000068s :  0.93%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000052s :  0.72%
    jit_opt_a.loop_unroll                :     0.000035s :  0.48%
    jit_opt_a.a_1                        :     0.000572s :  7.88%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.15%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.16%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000466s :  6.41%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.24%
    jit_opt_a.cse                        :     0.000033s :  0.45%
    jit_opt_a.replace_applicator         :     0.000017s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000030s :  0.42%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000461s :  6.35%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.25%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.21%
    cconv                                :     0.000022s :  0.30%
    loop_unroll                          :     0.000405s :  5.57%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.29%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.25%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.08%
    remove_dup_value                     :     0.000013s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000042s :  0.58%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000014s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000450s :  6.19%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.18%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000029s :  0.40%


Time group info:
------[substitution.]   0.000165    23
 1.13% :     0.000002s :      2: substitution.elim_not_effective
 0.82% :     0.000001s :      2: substitution.fold_const_symbol
 2.99% :     0.000005s :      3: substitution.graph_param_transform
82.17% :     0.000135s :      4: substitution.inline
 1.66% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.30% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.93% :     0.000003s :      2: substitution.replace_old_param
 6.99% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003754     2
80.52% :     0.003023s :      1: type_inference.infer
19.48% :     0.000731s :      1: type_inference.specialize
------[replace.]   0.000061     6
65.68% :     0.000040s :      4: replace.inline
34.32% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000143     6
92.77% :     0.000133s :      4: match.inline
 7.23% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000001s :     10: predicate.addn_zero_filter
 1.79% :     0.000002s :     10: predicate.arithmetic_simplify
 1.32% :     0.000002s :     10: predicate.cast_eliminate
 0.87% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000001s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.54% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000001s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.15% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.18% :     0.000001s :     10: predicate.minmaximum_grad
 1.68% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.51% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.73% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.38% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.04% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.56% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.01% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.51% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.70% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.12% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.47% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000493    12
49.34% :     0.000243s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.13% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.53% :     0.000224s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012076    72
 0.38% :     0.000045s :      1: add_recomputation
 0.50% :     0.000061s :      1: auto_monad
 0.13% :     0.000016s :      1: auto_monad_reorder
 2.14% :     0.000258s :      1: bootstrap
 0.20% :     0.000024s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000019s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
38.16% :     0.004608s :      1: jit_opt_a
 1.16% :     0.000140s :      1: jit_opt_after_cconv
 0.45% :     0.000054s :      1: jit_opt_b
 3.42% :     0.000412s :      1: loop_unroll
 3.89% :     0.000469s :      1: mutable_eliminate
 6.39% :     0.000772s :     26: opt.transform.jit_opt_a
 0.38% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.23% :     0.000027s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.79% :     0.000457s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.04% :     0.000004s :      1: partial_unused_args_eliminate
 0.04% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000033s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000015s :      1: remove_dup_value
 1.72% :     0.000207s :      1: renormalize.infer
 2.08% :     0.000251s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000033s :      1: rewriter_after_opt_a
 0.59% :     0.000071s :      1: rewriter_before_opt_a
 0.63% :     0.000076s :      1: symbol_engine_optimizer
31.59% :     0.003814s :      1: type_inference


TotalTime = 0.0114674, [30]
[bootstrap]: 0.00026471
[type_inference]: 0.00415381
[event_method]: 1.645e-05
[auto_monad]: 6.197e-05
[graph_reusing]: 5.92999e-06
[pre_auto_parallel]: 3.43e-06
[py_interpret_to_execute]: 3.161e-05
[rewriter_before_opt_a]: 7.114e-05
[expand_dump_flag]: 3.16999e-06
[jit_opt_a]: 0.00488833, [2]
    [Cycle 1]: 0.00146891, [27]
        [switch_simplify]: 5.232e-05
        [loop_unroll]: 3.062e-05
        [a_1]: 0.00052889
        [with_stream_mark]: 1.496e-05
        [recompute_prepare]: 7.15e-06
        [updatestate_depend_eliminate]: 3.28998e-06
        [updatestate_assign_eliminate]: 3.09001e-06
        [updatestate_loads_eliminate]: 2.46998e-06
        [parameter_eliminate]: 1.87001e-06
        [specialize_transform]: 5.91998e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 5.74e-06
        [meta_shard_fg_expand]: 2.64999e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 4.33001e-06
        [cell_reuse_recompute_pass]: 1.13001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.505e-05
        [j_node_and_user_rematch]: 9.62001e-06
        [meta_fg_expand]: 2.22999e-06
        [replace_old_param]: 9.01998e-06
        [inline_without_move]: 5.39998e-06
        [renormalize]: 0.00052282
        [add_forward_monad_depend]: 4.87e-06
        [auto_monad_grad]: 2.31e-06
        [auto_monad_eliminator]: 1.341e-05
        [cse]: 2.903e-05
        [replace_applicator]: 1.139e-05
    [Cycle 2]: 0.00031894, [27]
        [switch_simplify]: 5.69e-06
        [loop_unroll]: 5.10999e-06
        [a_1]: 9.085e-05
        [with_stream_mark]: 8.65999e-06
        [recompute_prepare]: 5.14e-06
        [updatestate_depend_eliminate]: 2.49001e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 5.17999e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 5.09e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.38999e-06
        [merge_forward]: 2.37001e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.205e-05
        [j_node_and_user_rematch]: 7.48999e-06
        [meta_fg_expand]: 1.53002e-06
        [replace_old_param]: 6.85998e-06
        [inline_without_move]: 4.59998e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.08001e-06
        [auto_monad_grad]: 1.04e-06
        [auto_monad_eliminator]: 5.00001e-06
        [cse]: 1.051e-05
        [replace_applicator]: 5.18002e-06
[py_interpret_to_execute_after_opt_a]: 9.36998e-06
[rewriter_after_opt_a]: 3.259e-05
[convert_after_rewriter]: 6.27001e-06
[order_py_execute_after_rewriter]: 4.80999e-06
[mutable_eliminate]: 0.00048105
[jit_opt_b]: 5.043e-05, [1]
    [Cycle 1]: 4.417e-05, [2]
        [frontend_op_eliminate]: 1.702e-05
        [inline_after_opt_a]: 1.527e-05
[cconv]: 2.357e-05
[loop_unroll]: 0.00040894
[jit_opt_after_cconv]: 0.00013473, [1]
    [Cycle 1]: 0.00012868, [11]
        [c_1]: 2.032e-05
        [parameter_eliminate]: 2.49001e-06
        [updatestate_depend_eliminate]: 5.24e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.03002e-06
        [cse]: 1.67e-05
        [call_graph_tuple_transform]: 1.718e-05
        [tuple_list_get_item_eliminator]: 5.21002e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 3.09985e-07
        [switch_simplify]: 5.37999e-06
[remove_dup_value]: 1.443e-05
[partial_unused_args_eliminate]: 2.37001e-06
[environ_conv]: 5.43002e-06
[add_recomputation]: 4.686e-05
[cse_after_recomputation]: 2.04e-05, [1]
    [Cycle 1]: 1.495e-05, [1]
        [cse]: 9.19e-06
[auto_monad_reorder]: 1.418e-05
[get_jit_bprop_graph]: 1.72001e-06
[rewriter_after_jit_bprop_graph]: 4.97e-06
[opt_after_jit_grad]: 0.00044315
[symbol_engine_optimizer]: 7.06e-05, [1]
    [Cycle 1]: 6.432e-05, [6]
        [build]: 3.66999e-06
        [elim_shapecalc]: 6.94001e-06
        [elim_not_effective]: 1.174e-05
        [opt_reshape]: 5.82001e-06
        [fold_const_symbol]: 8.60999e-06
        [renormalize]: 5.19998e-07
[validate]: 3.298e-05
Sums
    bootstrap                            :     0.000265s :  3.39%
    type_inference                       :     0.004154s : 53.22%
    event_method                         :     0.000016s :  0.21%
    auto_monad                           :     0.000062s :  0.79%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000003s :  0.04%
    py_interpret_to_execute              :     0.000032s :  0.40%
    rewriter_before_opt_a                :     0.000071s :  0.91%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000058s :  0.74%
    jit_opt_a.loop_unroll                :     0.000036s :  0.46%
    jit_opt_a.a_1                        :     0.000620s :  7.94%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.12%
    jit_opt_a.merge_forward              :     0.000007s :  0.09%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000523s :  6.70%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.24%
    jit_opt_a.cse                        :     0.000040s :  0.51%
    jit_opt_a.replace_applicator         :     0.000017s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.12%
    rewriter_after_opt_a                 :     0.000033s :  0.42%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000481s :  6.16%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.20%
    cconv                                :     0.000024s :  0.30%
    loop_unroll                          :     0.000409s :  5.24%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000047s :  0.60%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000014s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000443s :  5.68%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000033s :  0.42%


Time group info:
------[substitution.]   0.000186    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.58% :     0.000005s :      3: substitution.graph_param_transform
83.09% :     0.000155s :      4: substitution.inline
 1.61% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.46% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.15% :     0.000004s :      2: substitution.replace_old_param
 6.40% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004096     2
79.87% :     0.003271s :      1: type_inference.infer
20.13% :     0.000825s :      1: type_inference.specialize
------[replace.]   0.000076     6
56.18% :     0.000043s :      4: replace.inline
43.82% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000163     6
93.43% :     0.000153s :      4: match.inline
 6.57% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.63% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.21% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 1.95% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.80% :     0.000007s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.53% :     0.000002s :      6: predicate.less_batch_normalization
 1.36% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.24% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.23% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.49% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     15: predicate.partial_eliminate
 1.29% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.81% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.68% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.62% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.77% :     0.000009s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.96% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000549    12
49.40% :     0.000271s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.25% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.35% :     0.000249s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012873    72
 0.39% :     0.000050s :      1: add_recomputation
 0.51% :     0.000065s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.18% :     0.000281s :      1: bootstrap
 0.20% :     0.000026s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
37.99% :     0.004891s :      1: jit_opt_a
 1.07% :     0.000138s :      1: jit_opt_after_cconv
 0.41% :     0.000053s :      1: jit_opt_b
 3.23% :     0.000416s :      1: loop_unroll
 3.80% :     0.000489s :      1: mutable_eliminate
 6.42% :     0.000826s :     26: opt.transform.jit_opt_a
 0.34% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.51% :     0.000451s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.04% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000034s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000017s :      1: remove_dup_value
 1.98% :     0.000254s :      1: renormalize.infer
 2.03% :     0.000261s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000035s :      1: rewriter_after_opt_a
 0.57% :     0.000074s :      1: rewriter_before_opt_a
 0.57% :     0.000073s :      1: symbol_engine_optimizer
32.36% :     0.004166s :      1: type_inference


TotalTime = 0.0283598, [33]
[bootstrap]: 0.00034522
[type_inference]: 0.0054546
[event_method]: 1.841e-05
[auto_monad]: 6.474e-05
[graph_reusing]: 5.86e-06
[pre_auto_parallel]: 3.15998e-06
[py_interpret_to_execute]: 4.435e-05
[rewriter_before_opt_a]: 7.408e-05
[expand_dump_flag]: 3.61001e-06
[jit_opt_a]: 0.00759132, [2]
    [Cycle 1]: 0.00212144, [27]
        [switch_simplify]: 7.213e-05
        [loop_unroll]: 3.046e-05
        [a_1]: 0.00064944
        [with_stream_mark]: 2.725e-05
        [recompute_prepare]: 7.93999e-06
        [updatestate_depend_eliminate]: 3.91001e-06
        [updatestate_assign_eliminate]: 2.72001e-06
        [updatestate_loads_eliminate]: 2.88e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 5.89e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.92001e-06
        [meta_shard_fg_expand]: 1.97001e-06
        [get_grad_eliminate_]: 5.82001e-06
        [merge_forward]: 3.5e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.637e-05
        [j_node_and_user_rematch]: 1.093e-05
        [meta_fg_expand]: 2.49001e-06
        [replace_old_param]: 1.026e-05
        [inline_without_move]: 5.54998e-06
        [renormalize]: 0.00097858
        [add_forward_monad_depend]: 6.40002e-06
        [auto_monad_grad]: 2.54001e-06
        [auto_monad_eliminator]: 1.534e-05
        [cse]: 3.15e-05
        [replace_applicator]: 1.624e-05
    [Cycle 2]: 0.0003342, [27]
        [switch_simplify]: 5.62999e-06
        [loop_unroll]: 5.25001e-06
        [a_1]: 9.442e-05
        [with_stream_mark]: 1.229e-05
        [recompute_prepare]: 5.04e-06
        [updatestate_depend_eliminate]: 3.02002e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.76998e-06
        [parameter_eliminate]: 1.28002e-06
        [specialize_transform]: 4.87e-06
        [updatestate_useless_node_eliminater]: 4.77998e-06
        [accelerated_algorithm]: 5.24e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 4.53999e-06
        [merge_forward]: 2.79999e-06
        [cell_reuse_recompute_pass]: 2.16e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.351e-05
        [j_node_and_user_rematch]: 8.13999e-06
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 7.73001e-06
        [inline_without_move]: 4.70999e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.59e-06
        [auto_monad_grad]: 9.29984e-07
        [auto_monad_eliminator]: 4.80001e-06
        [cse]: 1.074e-05
        [replace_applicator]: 4.72e-06
[py_interpret_to_execute_after_opt_a]: 1.34e-05
[rewriter_after_opt_a]: 3.805e-05
[convert_after_rewriter]: 6.53e-06
[order_py_execute_after_rewriter]: 5.10001e-06
[mutable_eliminate]: 0.00170594
[jit_opt_b]: 6.032e-05, [1]
    [Cycle 1]: 4.916e-05, [2]
        [frontend_op_eliminate]: 1.826e-05
        [inline_after_opt_a]: 1.815e-05
[cconv]: 3.831e-05
[loop_unroll]: 0.00047088
[jit_opt_after_cconv]: 0.00017777, [1]
    [Cycle 1]: 0.00016942, [11]
        [c_1]: 2.296e-05
        [parameter_eliminate]: 5.51e-06
        [updatestate_depend_eliminate]: 1.042e-05
        [updatestate_assign_eliminate]: 2.64999e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [cse]: 3.11e-05
        [call_graph_tuple_transform]: 2.567e-05
        [tuple_list_get_item_eliminator]: 5.25001e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 1.10001e-06
        [switch_simplify]: 5.56998e-06
[remove_dup_value]: 1.649e-05
[partial_unused_args_eliminate]: 2.46e-06
[environ_conv]: 5.44e-06
[add_recomputation]: 5.631e-05
[cse_after_recomputation]: 2.229e-05, [1]
    [Cycle 1]: 1.613e-05, [1]
        [cse]: 9.23002e-06
[auto_monad_reorder]: 1.613e-05
[get_jit_bprop_graph]: 2.32001e-06
[rewriter_after_jit_bprop_graph]: 9.25001e-06
[opt_after_jit_grad]: 0.00048299
[symbol_engine_optimizer]: 8.109e-05, [1]
    [Cycle 1]: 7.338e-05, [6]
        [build]: 5.58002e-06
        [elim_shapecalc]: 7.97e-06
        [elim_not_effective]: 1.389e-05
        [opt_reshape]: 6.96999e-06
        [fold_const_symbol]: 9.51e-06
        [renormalize]: 8.10018e-07
[validate]: 4.63e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.0112453
[execute]: 9.12001e-06
Sums
    bootstrap                            :     0.000345s :  1.53%
    type_inference                       :     0.005455s : 24.21%
    event_method                         :     0.000018s :  0.08%
    auto_monad                           :     0.000065s :  0.29%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.20%
    rewriter_before_opt_a                :     0.000074s :  0.33%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000078s :  0.35%
    jit_opt_a.loop_unroll                :     0.000036s :  0.16%
    jit_opt_a.a_1                        :     0.000744s :  3.30%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000979s :  4.34%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.09%
    jit_opt_a.cse                        :     0.000042s :  0.19%
    jit_opt_a.replace_applicator         :     0.000021s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.06%
    rewriter_after_opt_a                 :     0.000038s :  0.17%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.001706s :  7.57%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.08%
    cconv                                :     0.000038s :  0.17%
    loop_unroll                          :     0.000471s :  2.09%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000031s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000056s :  0.25%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.04%
    opt_after_jit_grad                   :     0.000483s :  2.14%
    symbol_engine_optimizer.build        :     0.000006s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.21%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011245s : 49.91%
    execute                              :     0.000009s :  0.04%


Time group info:
------[substitution.]   0.000291    23
 0.90% :     0.000003s :      2: substitution.elim_not_effective
 0.42% :     0.000001s :      2: substitution.fold_const_symbol
 2.44% :     0.000007s :      3: substitution.graph_param_transform
86.25% :     0.000251s :      4: substitution.inline
 1.71% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.71% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.77% :     0.000005s :      2: substitution.replace_old_param
 4.79% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005373     2
81.88% :     0.004400s :      1: type_inference.infer
18.12% :     0.000974s :      1: type_inference.specialize
------[replace.]   0.000084     6
69.61% :     0.000059s :      4: replace.inline
30.39% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000261     6
95.09% :     0.000248s :      4: match.inline
 4.91% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.10% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.19% :     0.000001s :     10: predicate.addn_zero_filter
 2.23% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.44% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.71% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.34% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.46% :     0.000002s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.16% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.87% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.39% :     0.000002s :      6: predicate.less_batch_normalization
 1.88% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.64% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.90% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.12% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 3.15% :     0.000004s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.57% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.81% :     0.000001s :      6: predicate.replace_old_param
 0.52% :     0.000001s :      3: predicate.reset_defer_inline
 1.57% :     0.000002s :     10: predicate.reshape_eliminate
 1.08% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.17% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.06% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.21% :     0.000003s :     16: predicate.switch_defer_inline
 1.91% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.58% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.50% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.14% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000739    12
47.85% :     0.000354s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.04% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.11% :     0.000326s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.030431    76
 0.20% :     0.000060s :      1: add_recomputation
 0.22% :     0.000068s :      1: auto_monad
 0.06% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.19% :     0.000361s :      1: bootstrap
 0.14% :     0.000041s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.08% :     0.000023s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
24.96% :     0.007595s :      1: jit_opt_a
 0.59% :     0.000181s :      1: jit_opt_after_cconv
 0.21% :     0.000064s :      1: jit_opt_b
 1.58% :     0.000481s :      1: loop_unroll
 5.65% :     0.001721s :      1: mutable_eliminate
 3.22% :     0.000981s :     26: opt.transform.jit_opt_a
 0.18% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000028s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.61% :     0.000491s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000047s :      1: py_interpret_to_execute
 0.05% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000020s :      1: remove_dup_value
 1.76% :     0.000537s :      1: renormalize.infer
 1.42% :     0.000432s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000041s :      1: rewriter_after_opt_a
 0.25% :     0.000077s :      1: rewriter_before_opt_a
 0.28% :     0.000084s :      1: symbol_engine_optimizer
37.02% :     0.011266s :      1: task_emit
17.99% :     0.005476s :      1: type_inference
 0.23% :     0.000070s :      1: validate


TotalTime = 0.0116865, [30]
[bootstrap]: 0.00026604
[type_inference]: 0.00407319
[event_method]: 1.664e-05
[auto_monad]: 6.33e-05
[graph_reusing]: 6.10002e-06
[pre_auto_parallel]: 2.27001e-06
[py_interpret_to_execute]: 3.331e-05
[rewriter_before_opt_a]: 7.197e-05
[expand_dump_flag]: 3.08998e-06
[jit_opt_a]: 0.00504579, [2]
    [Cycle 1]: 0.00150938, [27]
        [switch_simplify]: 5.086e-05
        [loop_unroll]: 3.049e-05
        [a_1]: 0.00052319
        [with_stream_mark]: 1.636e-05
        [recompute_prepare]: 6.85002e-06
        [updatestate_depend_eliminate]: 3.93999e-06
        [updatestate_assign_eliminate]: 2.89001e-06
        [updatestate_loads_eliminate]: 2.55002e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 5.97001e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 5.71998e-06
        [meta_shard_fg_expand]: 2.41e-06
        [get_grad_eliminate_]: 5.45001e-06
        [merge_forward]: 3.86001e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.516e-05
        [j_node_and_user_rematch]: 9.42999e-06
        [meta_fg_expand]: 2.90002e-06
        [replace_old_param]: 8.57e-06
        [inline_without_move]: 5.77001e-06
        [renormalize]: 0.0005709
        [add_forward_monad_depend]: 5.64e-06
        [auto_monad_grad]: 2.01e-06
        [auto_monad_eliminator]: 1.304e-05
        [cse]: 2.726e-05
        [replace_applicator]: 1.214e-05
    [Cycle 2]: 0.00032645, [27]
        [switch_simplify]: 5.74999e-06
        [loop_unroll]: 5.30001e-06
        [a_1]: 9.075e-05
        [with_stream_mark]: 9.27001e-06
        [recompute_prepare]: 4.97e-06
        [updatestate_depend_eliminate]: 2.47001e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 5.12999e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.25001e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.67e-06
        [merge_forward]: 2.58e-06
        [cell_reuse_recompute_pass]: 1.45001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.253e-05
        [j_node_and_user_rematch]: 8e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 7.25e-06
        [inline_without_move]: 4.75001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.40999e-06
        [auto_monad_grad]: 9.29984e-07
        [auto_monad_eliminator]: 5.12999e-06
        [cse]: 1.05e-05
        [replace_applicator]: 5.58002e-06
[py_interpret_to_execute_after_opt_a]: 9.51e-06
[rewriter_after_opt_a]: 3.39e-05
[convert_after_rewriter]: 8.11002e-06
[order_py_execute_after_rewriter]: 5.80002e-06
[mutable_eliminate]: 0.0005114
[jit_opt_b]: 5.092e-05, [1]
    [Cycle 1]: 4.434e-05, [2]
        [frontend_op_eliminate]: 1.792e-05
        [inline_after_opt_a]: 1.464e-05
[cconv]: 2.458e-05
[loop_unroll]: 0.00041092
[jit_opt_after_cconv]: 0.00020272, [1]
    [Cycle 1]: 0.0001964, [11]
        [c_1]: 2.124e-05
        [parameter_eliminate]: 2.39999e-06
        [updatestate_depend_eliminate]: 5.35001e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 1.659e-05
        [call_graph_tuple_transform]: 1.881e-05
        [tuple_list_get_item_eliminator]: 5.55001e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 3.19997e-07
        [switch_simplify]: 6.14999e-06
[remove_dup_value]: 1.362e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 6.12001e-06
[add_recomputation]: 4.891e-05
[cse_after_recomputation]: 2.165e-05, [1]
    [Cycle 1]: 1.609e-05, [1]
        [cse]: 1.007e-05
[auto_monad_reorder]: 1.5e-05
[get_jit_bprop_graph]: 1.87001e-06
[rewriter_after_jit_bprop_graph]: 4.90999e-06
[opt_after_jit_grad]: 0.00046063
[symbol_engine_optimizer]: 7.375e-05, [1]
    [Cycle 1]: 6.711e-05, [6]
        [build]: 4.06001e-06
        [elim_shapecalc]: 7.99002e-06
        [elim_not_effective]: 1.21e-05
        [opt_reshape]: 5.91e-06
        [fold_const_symbol]: 8.37e-06
        [renormalize]: 3.89991e-07
[validate]: 3.328e-05
Sums
    bootstrap                            :     0.000266s :  3.39%
    type_inference                       :     0.004073s : 51.96%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000063s :  0.81%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.42%
    rewriter_before_opt_a                :     0.000072s :  0.92%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000057s :  0.72%
    jit_opt_a.loop_unroll                :     0.000036s :  0.46%
    jit_opt_a.a_1                        :     0.000614s :  7.83%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.06%
    jit_opt_a.replace_old_param          :     0.000016s :  0.20%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000571s :  7.28%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.23%
    jit_opt_a.cse                        :     0.000038s :  0.48%
    jit_opt_a.replace_applicator         :     0.000018s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000034s :  0.43%
    convert_after_rewriter               :     0.000008s :  0.10%
    order_py_execute_after_rewriter      :     0.000006s :  0.07%
    mutable_eliminate                    :     0.000511s :  6.52%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000025s :  0.31%
    loop_unroll                          :     0.000411s :  5.24%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.08%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.08%
    add_recomputation                    :     0.000049s :  0.62%
    cse_after_recomputation.cse          :     0.000010s :  0.13%
    auto_monad_reorder                   :     0.000015s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000461s :  5.88%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.42%


Time group info:
------[substitution.]   0.000185    23
 0.94% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.93% :     0.000005s :      3: substitution.graph_param_transform
82.80% :     0.000154s :      4: substitution.inline
 1.87% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.39% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.94% :     0.000004s :      2: substitution.replace_old_param
 6.45% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004016     2
79.76% :     0.003203s :      1: type_inference.infer
20.24% :     0.000813s :      1: type_inference.specialize
------[replace.]   0.000067     6
66.70% :     0.000044s :      4: replace.inline
33.30% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000161     6
93.36% :     0.000151s :      4: match.inline
 6.64% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.53% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.20% :     0.000001s :     10: predicate.addn_zero_filter
 2.11% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.58% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000001s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.11% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.28% :     0.000006s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.14% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.14% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.65% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.39% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.79% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.89% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.58% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.46% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.79% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.69% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.09% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000551    12
48.91% :     0.000269s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.31% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.78% :     0.000252s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013136    72
 0.39% :     0.000052s :      1: add_recomputation
 0.51% :     0.000067s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 2.13% :     0.000280s :      1: bootstrap
 0.21% :     0.000027s :      1: cconv
 0.08% :     0.000011s :      1: convert_after_rewriter
 0.18% :     0.000024s :      1: cse_after_recomputation
 0.07% :     0.000009s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.44% :     0.005049s :      1: jit_opt_a
 1.57% :     0.000206s :      1: jit_opt_after_cconv
 0.41% :     0.000054s :      1: jit_opt_b
 3.19% :     0.000419s :      1: loop_unroll
 3.96% :     0.000520s :      1: mutable_eliminate
 6.25% :     0.000821s :     26: opt.transform.jit_opt_a
 0.36% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.57% :     0.000469s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000036s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.13% :     0.000280s :      1: renormalize.infer
 2.15% :     0.000283s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000037s :      1: rewriter_after_opt_a
 0.57% :     0.000075s :      1: rewriter_before_opt_a
 0.58% :     0.000076s :      1: symbol_engine_optimizer
31.10% :     0.004085s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.483.395 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc41d30> should be wrapped by jit only once.

TotalTime = 0.0313774, [33]
[bootstrap]: 0.00032798
[type_inference]: 0.00676274
[event_method]: 2.252e-05
[auto_monad]: 7.069e-05
[graph_reusing]: 6.42001e-06
[pre_auto_parallel]: 2.56e-06
[py_interpret_to_execute]: 3.656e-05
[rewriter_before_opt_a]: 7.868e-05
[expand_dump_flag]: 3.21999e-06
[jit_opt_a]: 0.0118244, [2]
    [Cycle 1]: 0.00471795, [27]
        [switch_simplify]: 7.141e-05
        [loop_unroll]: 2.961e-05
        [a_1]: 0.00064219
        [with_stream_mark]: 2.897e-05
        [recompute_prepare]: 7.88001e-06
        [updatestate_depend_eliminate]: 4.3e-06
        [updatestate_assign_eliminate]: 3.33998e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 2.09e-06
        [specialize_transform]: 5.91998e-06
        [updatestate_useless_node_eliminater]: 5.54e-06
        [accelerated_algorithm]: 6.36998e-06
        [meta_shard_fg_expand]: 2.59001e-06
        [get_grad_eliminate_]: 5.94999e-06
        [merge_forward]: 3.85e-06
        [cell_reuse_recompute_pass]: 1.20999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.698e-05
        [j_node_and_user_rematch]: 1.062e-05
        [meta_fg_expand]: 2.49999e-06
        [replace_old_param]: 9.81e-06
        [inline_without_move]: 5.05999e-06
        [renormalize]: 0.0008462
        [add_forward_monad_depend]: 6.94999e-06
        [auto_monad_grad]: 3.04999e-06
        [auto_monad_eliminator]: 1.757e-05
        [cse]: 3.254e-05
        [replace_applicator]: 4.618e-05
    [Cycle 2]: 0.00052003, [27]
        [switch_simplify]: 1.052e-05
        [loop_unroll]: 6.48e-06
        [a_1]: 0.00010906
        [with_stream_mark]: 4.001e-05
        [recompute_prepare]: 6.36998e-06
        [updatestate_depend_eliminate]: 4.85999e-06
        [updatestate_assign_eliminate]: 3.02002e-06
        [updatestate_loads_eliminate]: 6.92002e-06
        [parameter_eliminate]: 3.08998e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 6.71e-06
        [meta_shard_fg_expand]: 2.99999e-06
        [get_grad_eliminate_]: 4.79e-06
        [merge_forward]: 5.37001e-06
        [cell_reuse_recompute_pass]: 5.61e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.863e-05
        [j_node_and_user_rematch]: 1.067e-05
        [meta_fg_expand]: 3.06001e-06
        [replace_old_param]: 1.233e-05
        [inline_without_move]: 1.787e-05
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 6.11e-06
        [auto_monad_grad]: 2.91e-06
        [auto_monad_eliminator]: 1.756e-05
        [cse]: 4.249e-05
        [replace_applicator]: 6.17001e-06
[py_interpret_to_execute_after_opt_a]: 2.527e-05
[rewriter_after_opt_a]: 5.715e-05
[convert_after_rewriter]: 8.16002e-06
[order_py_execute_after_rewriter]: 5.61998e-06
[mutable_eliminate]: 0.0009105
[jit_opt_b]: 6.123e-05, [1]
    [Cycle 1]: 5.01e-05, [2]
        [frontend_op_eliminate]: 1.972e-05
        [inline_after_opt_a]: 1.758e-05
[cconv]: 4.521e-05
[loop_unroll]: 0.00052489
[jit_opt_after_cconv]: 0.00018462, [1]
    [Cycle 1]: 0.00017714, [11]
        [c_1]: 2.278e-05
        [parameter_eliminate]: 6.39999e-06
        [updatestate_depend_eliminate]: 1.107e-05
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.86e-06
        [cse]: 3.753e-05
        [call_graph_tuple_transform]: 2.674e-05
        [tuple_list_get_item_eliminator]: 5.42001e-06
        [none_parameter_eliminate]: 1.80001e-06
        [renormalize]: 1.30999e-06
        [switch_simplify]: 5.29e-06
[remove_dup_value]: 1.646e-05
[partial_unused_args_eliminate]: 2.39001e-06
[environ_conv]: 5.82001e-06
[add_recomputation]: 6.458e-05
[cse_after_recomputation]: 2.418e-05, [1]
    [Cycle 1]: 1.823e-05, [1]
        [cse]: 1.101e-05
[auto_monad_reorder]: 1.625e-05
[get_jit_bprop_graph]: 2.46e-06
[rewriter_after_jit_bprop_graph]: 1.108e-05
[opt_after_jit_grad]: 0.00056766
[symbol_engine_optimizer]: 8.455e-05, [1]
    [Cycle 1]: 7.63e-05, [6]
        [build]: 6.43e-06
        [elim_shapecalc]: 9.25999e-06
        [elim_not_effective]: 1.598e-05
        [opt_reshape]: 5.96e-06
        [fold_const_symbol]: 9.24e-06
        [renormalize]: 9.99979e-07
[validate]: 4.876e-05
[backend_pass]: 1.15001e-06
[task_emit]: 0.0092895
[execute]: 1.058e-05
Sums
    bootstrap                            :     0.000328s :  1.54%
    type_inference                       :     0.006763s : 31.69%
    event_method                         :     0.000023s :  0.11%
    auto_monad                           :     0.000071s :  0.33%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000037s :  0.17%
    rewriter_before_opt_a                :     0.000079s :  0.37%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000082s :  0.38%
    jit_opt_a.loop_unroll                :     0.000036s :  0.17%
    jit_opt_a.a_1                        :     0.000751s :  3.52%
    jit_opt_a.with_stream_mark           :     0.000069s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000009s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000046s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.03%
    jit_opt_a.replace_old_param          :     0.000022s :  0.10%
    jit_opt_a.inline_without_move        :     0.000023s :  0.11%
    jit_opt_a.renormalize                :     0.000846s :  3.97%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000006s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000035s :  0.16%
    jit_opt_a.cse                        :     0.000075s :  0.35%
    jit_opt_a.replace_applicator         :     0.000052s :  0.25%
    py_interpret_to_execute_after_opt_a  :     0.000025s :  0.12%
    rewriter_after_opt_a                 :     0.000057s :  0.27%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000910s :  4.27%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.08%
    cconv                                :     0.000045s :  0.21%
    loop_unroll                          :     0.000525s :  2.46%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000038s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000065s :  0.30%
    cse_after_recomputation.cse          :     0.000011s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.05%
    opt_after_jit_grad                   :     0.000568s :  2.66%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000049s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.009290s : 43.53%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000297    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.07% :     0.000006s :      3: substitution.graph_param_transform
84.48% :     0.000251s :      4: substitution.inline
 1.96% :     0.000006s :      4: substitution.j_node_and_user_rematch
 2.35% :     0.000007s :      4: substitution.remove_not_recompute_node
 3.30% :     0.000010s :      2: substitution.replace_old_param
 4.55% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006676     2
85.53% :     0.005710s :      1: type_inference.infer
14.47% :     0.000966s :      1: type_inference.specialize
------[replace.]   0.000077     6
69.10% :     0.000053s :      4: replace.inline
30.90% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000261     6
95.22% :     0.000248s :      4: match.inline
 4.78% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000146   691
 1.43% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.69% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.88% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 2.63% :     0.000004s :     10: predicate.arithmetic_simplify
 1.39% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.86% :     0.000001s :     10: predicate.compare_switch_simplify
 1.80% :     0.000003s :     10: predicate.depend_value_elim
 1.08% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.16% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.06% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.08% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000001s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.05% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.03% :     0.000002s :     10: predicate.environ_get_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.17% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 7.02% :     0.000010s :     22: predicate.inline
 0.68% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.48% :     0.000002s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.97% :     0.000003s :     15: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.51% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.94% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 2.30% :     0.000003s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 1.97% :     0.000003s :     15: predicate.partial_eliminate
 1.08% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.60% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.32% :     0.000003s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.38% :     0.000001s :      3: predicate.reset_defer_inline
 1.53% :     0.000002s :     10: predicate.reshape_eliminate
 1.06% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.16% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     10: predicate.same_eliminate
 0.41% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.12% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.05% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 1.84% :     0.000003s :     16: predicate.switch_defer_inline
 1.79% :     0.000003s :     16: predicate.switch_layer_defer_inline
 9.13% :     0.000013s :     53: predicate.switch_simplify
 1.63% :     0.000002s :     10: predicate.tile_eliminate
 1.28% :     0.000002s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.99% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.99% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.35% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.42% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.99% :     0.000003s :     10: predicate.value_based_eliminate
 0.26% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000706    12
48.41% :     0.000342s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.59% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.00% :     0.000303s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.033402    76
 0.20% :     0.000068s :      1: add_recomputation
 0.22% :     0.000075s :      1: auto_monad
 0.06% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 1.05% :     0.000351s :      1: bootstrap
 0.14% :     0.000048s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.08% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.08% :     0.000028s :      1: event_method
 0.05% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
35.41% :     0.011828s :      1: jit_opt_a
 0.56% :     0.000188s :      1: jit_opt_after_cconv
 0.19% :     0.000064s :      1: jit_opt_b
 1.61% :     0.000538s :      1: loop_unroll
 2.78% :     0.000928s :      1: mutable_eliminate
 3.14% :     0.001050s :     26: opt.transform.jit_opt_a
 0.17% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000029s :      4: opt.transform.jit_opt_b
 0.05% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000026s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.73% :     0.000579s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000039s :      1: py_interpret_to_execute
 0.08% :     0.000028s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000019s :      1: remove_dup_value
 1.45% :     0.000484s :      1: renormalize.infer
 1.05% :     0.000352s :      1: renormalize.specialize
 0.04% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000062s :      1: rewriter_after_opt_a
 0.25% :     0.000082s :      1: rewriter_before_opt_a
 0.26% :     0.000087s :      1: symbol_engine_optimizer
27.88% :     0.009313s :      1: task_emit
20.33% :     0.006792s :      1: type_inference
 0.23% :     0.000077s :      1: validate


TotalTime = 0.0115154, [30]
[bootstrap]: 0.00025712
[type_inference]: 0.00404512
[event_method]: 1.675e-05
[auto_monad]: 6.403e-05
[graph_reusing]: 5.56e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 3.235e-05
[rewriter_before_opt_a]: 9.015e-05
[expand_dump_flag]: 3.46001e-06
[jit_opt_a]: 0.00502907, [2]
    [Cycle 1]: 0.00148487, [27]
        [switch_simplify]: 5.388e-05
        [loop_unroll]: 2.998e-05
        [a_1]: 0.00051615
        [with_stream_mark]: 1.548e-05
        [recompute_prepare]: 6.98e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 2.97002e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 5.97999e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.99e-06
        [meta_shard_fg_expand]: 1.94999e-06
        [get_grad_eliminate_]: 5.74e-06
        [merge_forward]: 1.366e-05
        [cell_reuse_recompute_pass]: 1.10999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.512e-05
        [j_node_and_user_rematch]: 9.55001e-06
        [meta_fg_expand]: 2.50002e-06
        [replace_old_param]: 8.82999e-06
        [inline_without_move]: 5.81e-06
        [renormalize]: 0.00054652
        [add_forward_monad_depend]: 5.17999e-06
        [auto_monad_grad]: 2.37001e-06
        [auto_monad_eliminator]: 1.263e-05
        [cse]: 2.693e-05
        [replace_applicator]: 1.071e-05
    [Cycle 2]: 0.00032239, [27]
        [switch_simplify]: 5.43002e-06
        [loop_unroll]: 4.84998e-06
        [a_1]: 9.056e-05
        [with_stream_mark]: 8.27e-06
        [recompute_prepare]: 4.99e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 5.32001e-06
        [updatestate_useless_node_eliminater]: 4.99998e-06
        [accelerated_algorithm]: 5.54998e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 4.61002e-06
        [merge_forward]: 2.51998e-06
        [cell_reuse_recompute_pass]: 1.13001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.291e-05
        [j_node_and_user_rematch]: 7.88999e-06
        [meta_fg_expand]: 1.55001e-06
        [replace_old_param]: 7.70998e-06
        [inline_without_move]: 4.67e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.17999e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 4.94e-06
        [cse]: 1.043e-05
        [replace_applicator]: 5.17e-06
[py_interpret_to_execute_after_opt_a]: 9.64999e-06
[rewriter_after_opt_a]: 3.093e-05
[convert_after_rewriter]: 6.51e-06
[order_py_execute_after_rewriter]: 5.54998e-06
[mutable_eliminate]: 0.00047542
[jit_opt_b]: 5.092e-05, [1]
    [Cycle 1]: 4.375e-05, [2]
        [frontend_op_eliminate]: 1.706e-05
        [inline_after_opt_a]: 1.507e-05
[cconv]: 2.449e-05
[loop_unroll]: 0.00041531
[jit_opt_after_cconv]: 0.00013515, [1]
    [Cycle 1]: 0.00012817, [11]
        [c_1]: 2.15e-05
        [parameter_eliminate]: 2.32999e-06
        [updatestate_depend_eliminate]: 5.07e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.03002e-06
        [cse]: 1.565e-05
        [call_graph_tuple_transform]: 1.74e-05
        [tuple_list_get_item_eliminator]: 5.57001e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 4.2998e-07
        [switch_simplify]: 5.39e-06
[remove_dup_value]: 1.292e-05
[partial_unused_args_eliminate]: 2.05002e-06
[environ_conv]: 5.06002e-06
[add_recomputation]: 4.673e-05
[cse_after_recomputation]: 1.957e-05, [1]
    [Cycle 1]: 1.425e-05, [1]
        [cse]: 8.43999e-06
[auto_monad_reorder]: 1.424e-05
[get_jit_bprop_graph]: 1.60999e-06
[rewriter_after_jit_bprop_graph]: 4.80001e-06
[opt_after_jit_grad]: 0.00044768
[symbol_engine_optimizer]: 7.329e-05, [1]
    [Cycle 1]: 6.674e-05, [6]
        [build]: 3.65e-06
        [elim_shapecalc]: 7.66999e-06
        [elim_not_effective]: 1.216e-05
        [opt_reshape]: 5.49e-06
        [fold_const_symbol]: 8.74e-06
        [renormalize]: 4.30009e-07
[validate]: 3.097e-05
Sums
    bootstrap                            :     0.000257s :  3.33%
    type_inference                       :     0.004045s : 52.32%
    event_method                         :     0.000017s :  0.22%
    auto_monad                           :     0.000064s :  0.83%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.42%
    rewriter_before_opt_a                :     0.000090s :  1.17%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000059s :  0.77%
    jit_opt_a.loop_unroll                :     0.000035s :  0.45%
    jit_opt_a.a_1                        :     0.000607s :  7.85%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000016s :  0.21%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.36%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000547s :  7.07%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.23%
    jit_opt_a.cse                        :     0.000037s :  0.48%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000031s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.07%
    mutable_eliminate                    :     0.000475s :  6.15%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000024s :  0.32%
    loop_unroll                          :     0.000415s :  5.37%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000013s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000047s :  0.60%
    cse_after_recomputation.cse          :     0.000008s :  0.11%
    auto_monad_reorder                   :     0.000014s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000448s :  5.79%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000031s :  0.40%


Time group info:
------[substitution.]   0.000186    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.90% :     0.000002s :      2: substitution.fold_const_symbol
 2.58% :     0.000005s :      3: substitution.graph_param_transform
82.19% :     0.000153s :      4: substitution.inline
 1.67% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.66% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.26% :     0.000004s :      2: substitution.replace_old_param
 6.71% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003987     2
79.29% :     0.003161s :      1: type_inference.infer
20.71% :     0.000826s :      1: type_inference.specialize
------[replace.]   0.000066     6
67.38% :     0.000044s :      4: replace.inline
32.62% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     6
93.02% :     0.000150s :      4: match.inline
 6.98% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 1.81% :     0.000002s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.78% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.38% :     0.000006s :     22: predicate.inline
 1.10% :     0.000001s :      6: predicate.inline_without_move
 0.48% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :      6: predicate.less_batch_normalization
 1.46% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.70% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.17% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.54% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.37% :     0.000003s :     15: predicate.partial_eliminate
 1.29% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.01% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.56% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.51% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.11% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.00% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000555    12
49.01% :     0.000272s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.79% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.20% :     0.000256s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012943    72
 0.38% :     0.000049s :      1: add_recomputation
 0.52% :     0.000067s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.09% :     0.000270s :      1: bootstrap
 0.21% :     0.000027s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.88% :     0.005032s :      1: jit_opt_a
 1.07% :     0.000138s :      1: jit_opt_after_cconv
 0.41% :     0.000054s :      1: jit_opt_b
 3.27% :     0.000423s :      1: loop_unroll
 3.74% :     0.000484s :      1: mutable_eliminate
 6.30% :     0.000816s :     26: opt.transform.jit_opt_a
 0.36% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000026s :      4: opt.transform.jit_opt_b
 0.10% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.52% :     0.000455s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000035s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000015s :      1: remove_dup_value
 2.04% :     0.000264s :      1: renormalize.infer
 2.13% :     0.000275s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000034s :      1: rewriter_after_opt_a
 0.72% :     0.000094s :      1: rewriter_before_opt_a
 0.59% :     0.000076s :      1: symbol_engine_optimizer
31.34% :     0.004056s :      1: type_inference


TotalTime = 0.0215309, [33]
[bootstrap]: 0.00036208
[type_inference]: 0.00531324
[event_method]: 1.759e-05
[auto_monad]: 6.417e-05
[graph_reusing]: 6.07001e-06
[pre_auto_parallel]: 2.29001e-06
[py_interpret_to_execute]: 3.968e-05
[rewriter_before_opt_a]: 7.471e-05
[expand_dump_flag]: 3.35e-06
[jit_opt_a]: 0.00624798, [2]
    [Cycle 1]: 0.00217356, [27]
        [switch_simplify]: 6.553e-05
        [loop_unroll]: 2.999e-05
        [a_1]: 0.00074654
        [with_stream_mark]: 2.188e-05
        [recompute_prepare]: 7.73001e-06
        [updatestate_depend_eliminate]: 4.1e-06
        [updatestate_assign_eliminate]: 2.70997e-06
        [updatestate_loads_eliminate]: 3.07002e-06
        [parameter_eliminate]: 1.97999e-06
        [specialize_transform]: 6.29999e-06
        [updatestate_useless_node_eliminater]: 4.84998e-06
        [accelerated_algorithm]: 5.84999e-06
        [meta_shard_fg_expand]: 1.99999e-06
        [get_grad_eliminate_]: 5.36998e-06
        [merge_forward]: 3.4e-06
        [cell_reuse_recompute_pass]: 1.19998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.46e-05
        [j_node_and_user_rematch]: 9.64e-06
        [meta_fg_expand]: 2.16998e-06
        [replace_old_param]: 9.89999e-06
        [inline_without_move]: 5.24998e-06
        [renormalize]: 0.00096727
        [add_forward_monad_depend]: 6.83e-06
        [auto_monad_grad]: 2.49001e-06
        [auto_monad_eliminator]: 1.545e-05
        [cse]: 2.276e-05
        [replace_applicator]: 1.622e-05
    [Cycle 2]: 0.00033319, [27]
        [switch_simplify]: 5.87001e-06
        [loop_unroll]: 5.00001e-06
        [a_1]: 9.123e-05
        [with_stream_mark]: 1.125e-05
        [recompute_prepare]: 5.32999e-06
        [updatestate_depend_eliminate]: 3.03998e-06
        [updatestate_assign_eliminate]: 2.41998e-06
        [updatestate_loads_eliminate]: 1.98002e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 4.90999e-06
        [updatestate_useless_node_eliminater]: 5.10001e-06
        [accelerated_algorithm]: 4.92e-06
        [meta_shard_fg_expand]: 1.26002e-06
        [get_grad_eliminate_]: 4.45999e-06
        [merge_forward]: 3.3e-06
        [cell_reuse_recompute_pass]: 1.87999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.313e-05
        [j_node_and_user_rematch]: 8.61002e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 7.41001e-06
        [inline_without_move]: 4.47e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 4.85999e-06
        [cse]: 1.242e-05
        [replace_applicator]: 5.00999e-06
[py_interpret_to_execute_after_opt_a]: 1.198e-05
[rewriter_after_opt_a]: 3.425e-05
[convert_after_rewriter]: 6.53e-06
[order_py_execute_after_rewriter]: 5.87001e-06
[mutable_eliminate]: 0.00080121
[jit_opt_b]: 5.463e-05, [1]
    [Cycle 1]: 4.608e-05, [2]
        [frontend_op_eliminate]: 1.714e-05
        [inline_after_opt_a]: 1.572e-05
[cconv]: 2.927e-05
[loop_unroll]: 0.00045929
[jit_opt_after_cconv]: 0.00015621, [1]
    [Cycle 1]: 0.00014926, [11]
        [c_1]: 2.107e-05
        [parameter_eliminate]: 3.63e-06
        [updatestate_depend_eliminate]: 8.08999e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [cse]: 2.595e-05
        [call_graph_tuple_transform]: 2.258e-05
        [tuple_list_get_item_eliminator]: 5.31002e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 1.14998e-06
        [switch_simplify]: 5.79e-06
[remove_dup_value]: 1.426e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 5.62999e-06
[add_recomputation]: 5.195e-05
[cse_after_recomputation]: 2.123e-05, [1]
    [Cycle 1]: 1.528e-05, [1]
        [cse]: 9.75002e-06
[auto_monad_reorder]: 1.506e-05
[get_jit_bprop_graph]: 2.94999e-06
[rewriter_after_jit_bprop_graph]: 8.33001e-06
[opt_after_jit_grad]: 0.00047891
[symbol_engine_optimizer]: 7.782e-05, [1]
    [Cycle 1]: 7.051e-05, [6]
        [build]: 4.38999e-06
        [elim_shapecalc]: 7.33999e-06
        [elim_not_effective]: 1.394e-05
        [opt_reshape]: 5.90002e-06
        [fold_const_symbol]: 9.56e-06
        [renormalize]: 6.60017e-07
[validate]: 4.31e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00688356
[execute]: 1.015e-05
Sums
    bootstrap                            :     0.000362s :  2.11%
    type_inference                       :     0.005313s : 31.02%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000064s :  0.37%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000040s :  0.23%
    rewriter_before_opt_a                :     0.000075s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000071s :  0.42%
    jit_opt_a.loop_unroll                :     0.000035s :  0.20%
    jit_opt_a.a_1                        :     0.000838s :  4.89%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000967s :  5.65%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000035s :  0.21%
    jit_opt_a.replace_applicator         :     0.000021s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000801s :  4.68%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000029s :  0.17%
    loop_unroll                          :     0.000459s :  2.68%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000052s :  0.30%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000479s :  2.80%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006884s : 40.18%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000261    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.37% :     0.000006s :      3: substitution.graph_param_transform
85.16% :     0.000222s :      4: substitution.inline
 1.86% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.88% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.95% :     0.000005s :      2: substitution.replace_old_param
 5.41% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005238     2
80.83% :     0.004234s :      1: type_inference.infer
19.17% :     0.001004s :      1: type_inference.specialize
------[replace.]   0.000079     6
66.75% :     0.000053s :      4: replace.inline
33.25% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000232     6
94.44% :     0.000219s :      4: match.inline
 5.56% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.22% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 2.30% :     0.000003s :     10: predicate.arithmetic_simplify
 1.28% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.18% :     0.000001s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.39% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.08% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 6.25% :     0.000007s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.68% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.62% :     0.000002s :     15: predicate.load_eliminater
 1.51% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.45% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.12% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 2.50% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 1.99% :     0.000002s :     15: predicate.partial_eliminate
 1.24% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.35% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.51% :     0.000001s :      3: predicate.reset_defer_inline
 1.62% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.56% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.13% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.70% :     0.000003s :     16: predicate.switch_defer_inline
 2.30% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.49% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.45% :     0.000002s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.94% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000699    12
47.41% :     0.000331s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.18% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.41% :     0.000317s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023664    76
 0.23% :     0.000055s :      1: add_recomputation
 0.28% :     0.000067s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.59% :     0.000376s :      1: bootstrap
 0.13% :     0.000032s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.42% :     0.006252s :      1: jit_opt_a
 0.67% :     0.000159s :      1: jit_opt_after_cconv
 0.24% :     0.000057s :      1: jit_opt_b
 1.98% :     0.000468s :      1: loop_unroll
 3.43% :     0.000812s :      1: mutable_eliminate
 4.49% :     0.001063s :     26: opt.transform.jit_opt_a
 0.22% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.06% :     0.000488s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000042s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.23% :     0.000528s :      1: renormalize.infer
 1.82% :     0.000430s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000040s :      1: rewriter_after_opt_a
 0.33% :     0.000078s :      1: rewriter_before_opt_a
 0.34% :     0.000081s :      1: symbol_engine_optimizer
29.18% :     0.006906s :      1: task_emit
22.52% :     0.005330s :      1: type_inference
 0.28% :     0.000065s :      1: validate


TotalTime = 0.0117637, [30]
[bootstrap]: 0.00026955
[type_inference]: 0.0040888
[event_method]: 1.711e-05
[auto_monad]: 6.932e-05
[graph_reusing]: 5.93998e-06
[pre_auto_parallel]: 2.84999e-06
[py_interpret_to_execute]: 3.546e-05
[rewriter_before_opt_a]: 7.068e-05
[expand_dump_flag]: 2.90002e-06
[jit_opt_a]: 0.00518591, [2]
    [Cycle 1]: 0.00155271, [27]
        [switch_simplify]: 5.064e-05
        [loop_unroll]: 3.082e-05
        [a_1]: 0.00052948
        [with_stream_mark]: 1.476e-05
        [recompute_prepare]: 8.15e-06
        [updatestate_depend_eliminate]: 3.04999e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 1.55001e-06
        [specialize_transform]: 6.38e-06
        [updatestate_useless_node_eliminater]: 5.66e-06
        [accelerated_algorithm]: 5.84e-06
        [meta_shard_fg_expand]: 2.09e-06
        [get_grad_eliminate_]: 5.89999e-06
        [merge_forward]: 2.74001e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.257e-05
        [j_node_and_user_rematch]: 8.82e-06
        [meta_fg_expand]: 2.32999e-06
        [replace_old_param]: 8.57e-06
        [inline_without_move]: 5.74e-06
        [renormalize]: 0.00062434
        [add_forward_monad_depend]: 4.92e-06
        [auto_monad_grad]: 2.04999e-06
        [auto_monad_eliminator]: 1.175e-05
        [cse]: 1.672e-05
        [replace_applicator]: 1.145e-05
    [Cycle 2]: 0.00033944, [27]
        [switch_simplify]: 5.59998e-06
        [loop_unroll]: 5.04e-06
        [a_1]: 9.844e-05
        [with_stream_mark]: 1.021e-05
        [recompute_prepare]: 5.25999e-06
        [updatestate_depend_eliminate]: 2.51e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 4.99e-06
        [updatestate_useless_node_eliminater]: 5.09003e-06
        [accelerated_algorithm]: 5.30001e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 4.60999e-06
        [merge_forward]: 2.32001e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.195e-05
        [j_node_and_user_rematch]: 7.85998e-06
        [meta_fg_expand]: 1.70001e-06
        [replace_old_param]: 7.61999e-06
        [inline_without_move]: 4.83001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 8.80013e-07
        [auto_monad_eliminator]: 5.71998e-06
        [cse]: 1.042e-05
        [replace_applicator]: 9.79999e-06
[py_interpret_to_execute_after_opt_a]: 9.89999e-06
[rewriter_after_opt_a]: 3.009e-05
[convert_after_rewriter]: 6.24001e-06
[order_py_execute_after_rewriter]: 8.15e-06
[mutable_eliminate]: 0.00052988
[jit_opt_b]: 5.093e-05, [1]
    [Cycle 1]: 4.432e-05, [2]
        [frontend_op_eliminate]: 1.739e-05
        [inline_after_opt_a]: 1.487e-05
[cconv]: 2.094e-05
[loop_unroll]: 0.00041096
[jit_opt_after_cconv]: 0.000136, [1]
    [Cycle 1]: 0.00012944, [11]
        [c_1]: 2.115e-05
        [parameter_eliminate]: 2.56e-06
        [updatestate_depend_eliminate]: 5.10001e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 1.631e-05
        [call_graph_tuple_transform]: 1.795e-05
        [tuple_list_get_item_eliminator]: 5.51e-06
        [none_parameter_eliminate]: 1.20001e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.43002e-06
[remove_dup_value]: 1.043e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 3.98999e-06
[add_recomputation]: 3.815e-05
[cse_after_recomputation]: 1.978e-05, [1]
    [Cycle 1]: 1.421e-05, [1]
        [cse]: 8.75001e-06
[auto_monad_reorder]: 1.178e-05
[get_jit_bprop_graph]: 1.64e-06
[rewriter_after_jit_bprop_graph]: 1.822e-05
[opt_after_jit_grad]: 0.00045059
[symbol_engine_optimizer]: 7.2e-05, [1]
    [Cycle 1]: 6.586e-05, [6]
        [build]: 2.93e-06
        [elim_shapecalc]: 7.88999e-06
        [
TotalTime = 0.0443236, [33]
[bootstrap]: 0.00037543
[type_inference]: 0.02175
[event_method]: 3.923e-05
[auto_monad]: 0.00011984
[graph_reusing]: 6.16e-06
[pre_auto_parallel]: 2.59001e-06
[py_interpret_to_execute]: 2.477e-05
[rewriter_before_opt_a]: 8.454e-05
[expand_dump_flag]: 3.18e-06
[jit_opt_a]: 0.0122701, [3]
    [Cycle 1]: 0.00631918, [27]
        [switch_simplify]: 5.609e-05
        [loop_unroll]: 3.714e-05
        [a_1]: 0.00079842
        [with_stream_mark]: 2.146e-05
        [recompute_prepare]: 1.726e-05
        [updatestate_depend_eliminate]: 6.91001e-06
        [updatestate_assign_eliminate]: 6.07999e-06
        [updatestate_loads_eliminate]: 5.36998e-06
        [parameter_eliminate]: 2.50002e-06
        [specialize_transform]: 1.275e-05
        [updatestate_useless_node_eliminater]: 1.17e-05
        [accelerated_algorithm]: 1.176e-05
        [meta_shard_fg_expand]: 2.73e-06
        [get_grad_eliminate_]: 1.162e-05
        [merge_forward]: 6.59999e-06
        [cell_reuse_recompute_pass]: 9.60019e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.385e-05
        [j_node_and_user_rematch]: 3.271e-05
        [meta_fg_expand]: 0.00200125
        [replace_old_param]: 4.987e-05
        [inline_without_move]: 5.064e-05
        [renormalize]: 0.00267888
        [add_forward_monad_depend]: 1.35e-05
        [auto_monad_grad]: 5.99e-06
        [auto_monad_eliminator]: 4.799e-05
        [cse]: 0.00013863
        [replace_applicator]: 6.29e-05
    [Cycle 2]: 0.00181215, [27]
        [switch_simplify]: 3.458e-05
        [loop_unroll]: 3.328e-05
        [a_1]: 0.00099506
        [with_stream_mark]: 1.24e-05
        [recompute_prepare]: 7.54002e-06
        [updatestate_depend_eliminate]: 3.51999e-06
        [updatestate_assign_eliminate]: 2.98998e-06
        [updatestate_loads_eliminate]: 3.01999e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 6.43e-06
        [updatestate_useless_node_eliminater]: 5.90002e-06
        [accelerated_algorithm]: 5.91998e-06
        [meta_shard_fg_expand]: 1.75001e-06
        [get_grad_eliminate_]: 6.02999e-06
        [merge_forward]: 3.16001e-06
        [cell_reuse_recompute_pass]: 7.7e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.322e-05
        [j_node_and_user_rematch]: 9.51e-06
        [meta_fg_expand]: 3.393e-05
        [replace_old_param]: 1.06e-05
        [inline_without_move]: 5.96998e-06
        [renormalize]: 0.00042216
        [add_forward_monad_depend]: 3.86001e-06
        [auto_monad_grad]: 1.58002e-06
        [auto_monad_eliminator]: 1.17e-05
        [cse]: 2.093e-05
        [replace_applicator]: 1.203e-05
    [Cycle 3]: 0.00035516, [27]
        [switch_simplify]: 6.34001e-06
        [loop_unroll]: 5.74999e-06
        [a_1]: 0.0001142
        [with_stream_mark]: 9.02999e-06
        [recompute_prepare]: 5.82999e-06
        [updatestate_depend_eliminate]: 3.26001e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.61999e-06
        [parameter_eliminate]: 9.29984e-07
        [specialize_transform]: 6.22001e-06
        [updatestate_useless_node_eliminater]: 5.76003e-06
        [accelerated_algorithm]: 6.27001e-06
        [meta_shard_fg_expand]: 1.21002e-06
        [get_grad_eliminate_]: 5.48002e-06
        [merge_forward]: 3.03e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.384e-05
        [j_node_and_user_rematch]: 9.79e-06
        [meta_fg_expand]: 1.92999e-06
        [replace_old_param]: 7.61999e-06
        [inline_without_move]: 5.45001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 9.50007e-07
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 6.38998e-06
        [cse]: 1.254e-05
        [replace_applicator]: 5.71e-06
[py_interpret_to_execute_after_opt_a]: 1.116e-05
[rewriter_after_opt_a]: 3.818e-05
[convert_after_rewriter]: 8.04997e-06
[order_py_execute_after_rewriter]: 5.49e-06
[mutable_eliminate]: 0.00057452
[jit_opt_b]: 5.291e-05, [1]
    [Cycle 1]: 4.587e-05, [2]
        [frelim_not_effective]: 1.223e-05
        [opt_reshape]: 5.63002e-06
        [fold_const_symbol]: 8.94e-06
        [renormalize]: 4.10015e-07
[validate]: 2.799e-05
Sums
    bootstrap                            :     0.000270s :  3.41%
    type_inference                       :     0.004089s : 51.73%
    event_method                         :     0.000017s :  0.22%
    auto_monad                           :     0.000069s :  0.88%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000003s :  0.04%
    py_interpret_to_execute              :     0.000035s :  0.45%
    rewriter_before_opt_a                :     0.000071s :  0.89%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000056s :  0.71%
    jit_opt_a.loop_unroll                :     0.000036s :  0.45%
    jit_opt_a.a_1                        :     0.000628s :  7.94%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.13%
    jit_opt_a.merge_forward              :     0.000005s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.20%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000624s :  7.90%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.22%
    jit_opt_a.cse                        :     0.000027s :  0.34%
    jit_opt_a.replace_applicator         :     0.000021s :  0.27%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.13%
    rewriter_after_opt_a                 :     0.000030s :  0.38%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000008s :  0.10%
    mutable_eliminate                    :     0.000530s :  6.70%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000021s :  0.26%
    loop_unroll                          :     0.000411s :  5.20%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000010s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
   ontend_op_eliminate]: 1.766e-05
        [inline_after_opt_a]: 1.699e-05
[cconv]: 2.313e-05
[loop_unroll]: 0.00048716
[jit_opt_after_cconv]: 0.00015827, [1]
    [Cycle 1]: 0.00015184, [11]
        [c_1]: 2.479e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 6.16e-06
        [updatestate_assign_eliminate]: 3.03e-06
        [updatestate_loads_eliminate]: 2.85002e-06
        [cse]: 2.241e-05
        [call_graph_tuple_transform]: 2.045e-05
        [tuple_list_get_item_eliminator]: 6.57002e-06
        [none_parameter_eliminate]: 2.00002e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 6.20002e-06
[remove_dup_value]: 1.565e-05
[partial_unused_args_eliminate]: 1.86e-06
[environ_conv]: 5.75001e-06
[add_recomputation]: 5.857e-05
[cse_after_recomputation]: 2.713e-05, [1]
    [Cycle 1]: 2.145e-05, [1]
        [cse]: 1.545e-05
[auto_monad_reorder]: 1.75e-05
[get_jit_bprop_graph]: 1.97001e-06
[rewriter_after_jit_bprop_graph]: 5.08002e-06
[opt_after_jit_grad]: 0.00049305
[symbol_engine_optimizer]: 8.62e-05, [1]
    [Cycle 1]: 7.992e-05, [6]
        [build]: 3.91999e-06
        [elim_shapecalc]: 9.27999e-06
        [elim_not_effective]: 1.65e-05
        [opt_reshape]: 7.38999e-06
        [fold_const_symbol]: 1.146e-05
        [renormalize]: 5.19998e-07
[validate]: 3.537e-05
[backend_pass]: 1.22e-06
[task_emit]: 0.00733364
[execute]: 6.59001e-06
Sums
    bootstrap                            :     0.000375s :  0.94%
    type_inference                       :     0.021750s : 54.71%
    event_method                         :     0.000039s :  0.10%
    auto_monad                           :     0.000120s :  0.30%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000025s :  0.06%
    rewriter_before_opt_a                :     0.000085s :  0.21%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000097s :  0.24%
    jit_opt_a.loop_unroll                :     0.000076s :  0.19%
    jit_opt_a.a_1                        :     0.001908s :  4.80%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000031s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000025s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.06%
    jit_opt_a.merge_forward              :     0.000013s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000051s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000052s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.002037s :  5.12%
    jit_opt_a.replace_old_param          :     0.000068s :  0.17%
    jit_opt_a.inline_without_move        :     0.000062s :  0.16%
    jit_opt_a.renormalize                :     0.003101s :  7.80%
    jit_opt_a.add_forward_monad_depend   :     0.000018s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000066s :  0.17%
    jit_opt_a.cse                        :     0.000172s :  0.43%
    jit_opt_a.replace_applicator         :     0.000081s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.03%
    rewriter_after_opt_a                 :     0.000038s :  0.10%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_elimin environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000038s :  0.48%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000012s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000018s :  0.23%
    opt_after_jit_grad                   :     0.000451s :  5.70%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000028s :  0.35%

ate                    :     0.000575s :  1.45%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.04%
    cconv                                :     0.000023s :  0.06%
    loop_unroll                          :     0.000487s :  1.23%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000006s :  0.01%
    add_recomputation                    :     0.000059s :  0.15%
    cse_after_recomputation.cse          :     0.000015s :  0.04%
    auto_monad_reorder                   :     0.000017s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000493s :  1.24%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.09%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007334s : 18.45%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000180    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.90% :     0.000002s :      2: substitution.fold_const_symbol
 2.54% :     0.000005s :      3: substitution.graph_param_transform
83.94% :     0.000151s :      4: substitution.inline
 1.67% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.33% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.00% :     0.000004s :      2: substitution.replace_old_param
 5.62% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004031     2
79.01% :     0.003185s :      1: type_inference.infer
20.99% :     0.000846s :      1: type_inference.specialize
------[replace.]   0.000067     6
65.45% :     0.000044s :      4: replace.inline
34.55% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000158     6
94.37% :     0.000149s :      4: match.inline
 5.63% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.58% :     0.000002s :     10: predicate.addn_zero_filter
 2.05% :     0.000002s :     10: predicate.arithmetic_simplify
 1.14% :     0.000001s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.13% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.22% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.20% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.81% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.07% :     0.000002s :     15: predicate.partial_eliminate
 1.28% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.99% :     0.000002s :     10: predicate.reduce_eliminate
 1.93% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.53% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.88% :     0.000009s :     53: predicate.switch_simplify
 1.06% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.44% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.09% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.89% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000561    12
45.96% :     0.000258s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.52% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.51% :     0.000272s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013297    72
 0.31% :     0.000041s :      1: add_recomputation
 0.55% :     0.000073s :      1: auto_monad
 0.11% :     0.000014s :      1: auto_monad_reorder
 2.12% :     0.000282s :      1: bootstrap
 0.18% :     0.000024s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
39.02% :     0.005189s :      1: jit_opt_a
 1.04% :     0.000139s :      1: jit_opt_after_cconv
 0.40% :     0.000054s :      1: jit_opt_b
 3.15% :     0.000419s :      1: loop_unroll
 4.05% :     0.000538s :      1: mutable_eliminate
 6.32% :     0.000841s :     26: opt.transform.jit_opt_a
 0.35% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.45% :     0.000458s :      1: opt_after_jit_grad
 0.08% :     0.000010s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.04% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000038s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000013s :      1: remove_dup_value
 2.32% :     0.000308s :      1: renormalize.infer
 2.32% :     0.000309s :      1: renormalize.specialize
 0.16% :     0.000021s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000033s :      1: rewriter_after_opt_a
 0.55% :     0.000073s :      1: rewriter_before_opt_a
 0.56% :     0.000075s :      1: symbol_engine_optimizer
30.83% :     0.004099s :      1: type_inference


Time group info:
------[substitution.]   0.000466   110
 0.53% :     0.000002s :      3: substitution.elim_not_effective
 0.34% :     0.000002s :      3: substitution.fold_const_symbol
 1.14% :     0.000005s :      4: substitution.graph_param_transform
57.75% :     0.000269s :     12: substitution.inline
 3.34% :     0.000016s :      2: substitution.inline_without_move
 3.92% :     0.000018s :     14: substitution.j_node_and_user_rematch
 2.24% :     0.000010s :      7: substitution.minmaximum_grad
 5.80% :     0.000027s :      5: substitution.partial_eliminate
 2.15% :     0.000010s :     14: substitution.remove_not_recompute_node
 5.31% :     0.000025s :     10: substitution.replace_applicator
 1.47% :     0.000007s :      7: substitution.replace_old_param
 0.53% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.29% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.11% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 8.07% :     0.000038s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.021676     2
95.52% :     0.020703s :      1: type_inference.infer
 4.48% :     0.000972s :      1: type_inference.specialize
------[replace.]   0.000131    19
64.14% :     0.000084s :     12: replace.inline
35.86% :     0.000047s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000281    19
93.50% :     0.000263s :     12: match.inline
 6.50% :     0.000018s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000354  2568
 1.62% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.50% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.43% :     0.000005s :     43: predicate.addn_check_dump
 1.70% :     0.000006s :     43: predicate.addn_zero_filter
 2.09% :     0.000007s :     43: predicate.arithmetic_simplify
 1.63% :     0.000006s :     43: predicate.cast_eliminate
 0.17% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.43% :     0.000005s :     43: predicate.compare_switch_simplify
 1.57% :     0.000006s :     43: predicate.depend_value_elim
 1.45% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.56% :     0.000006s :     43: predicate.dict_get_item_eliminator
 1.59% :     0.000006s :     43: predicate.dict_set_item_eliminator
 0.32% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.15% :     0.000001s :      4: predicate.elim_not_effective
 0.20% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.51% :     0.000005s :     43: predicate.environ_add_const_eliminate
 1.43% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.46% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.50% :     0.000005s :     43: predicate.environ_get_eliminate
 1.46% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.12% :     0.000000s :      4: predicate.fold_const_symbol
 0.89% :     0.000003s :     19: predicate.get_grad_eliminate
 0.18% :     0.000001s :      4: predicate.graph_param_transform
 4.52% :     0.000016s :     70: predicate.inline
 1.80% :     0.000006s :     40: predicate.inline_without_move
 0.38% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.85% :     0.000003s :     19: predicate.less_batch_normalization
 1.78% :     0.000006s :     50: predicate.list_to_tuple_eliminator_
 1.95% :     0.000007s :     54: predicate.load_eliminater
 0.49% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.18% :     0.000011s :     85: predicate.loop_unroll_before_grad
 1.78% :     0.000006s :     47: predicate.make_slice_get_slice_eliminator
 1.44% :     0.000005s :     43: predicate.merge_addn
 1.51% :     0.000005s :     43: predicate.minmaximum_grad
 0.53% :     0.000002s :      4: predicate.mutable_eliminate
 0.25% :     0.000001s :      4: predicate.opt_reshape
 2.35% :     0.000008s :     54: predicate.partial_eliminate
 1.58% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.83% :     0.000006s :     43: predicate.reduce_eliminate
 1.81% :     0.000006s :     50: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.53% :     0.000009s :     96: predicate.replace_applicator
 0.94% :     0.000003s :     40: predicate.replace_old_param
 0.11% :     0.000000s :      4: predicate.reset_defer_inline
 1.82% :     0.000006s :     43: predicate.reshape_eliminate
 1.64% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.27% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.61% :     0.000006s :     43: predicate.same_eliminate
 0.47% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.38% :     0.000001s :      8: predicate.special_op_eliminate
 0.87% :     0.000003s :     19: predicate.specialize_transform
 1.66% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000005s :     43: predicate.stack_unstack_eliminate
 0.18% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.67% :     0.000009s :     62: predicate.switch_defer_inline
 2.52% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.31% :     0.000022s :    151: predicate.switch_simplify
 1.49% :     0.000005s :     43: predicate.tile_eliminate
 1.51% :     0.000005s :     43: predicate.transpose_eliminate
 1.86% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.63% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.05% :     0.000011s :     58: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000006s :     43: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000006s :     50: predicate.tuple_to_list_eliminator_
 1.90% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000010s :     73: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000006s :     43: predicate.value_based_eliminate
 0.15% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.21% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001408    29
57.85% :     0.000815s :     13: func_graph_cloner_run.FuncGraphClonerGraph
42.15% :     0.000594s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.050038    91
 0.12% :     0.000062s :      1: add_recomputation
 0.25% :     0.000125s :      1: auto_monad
 0.04% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.79% :     0.000397s :      1: bootstrap
 0.05% :     0.000026s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.06% :     0.000029s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.09% :     0.000043s :      1: event_method
 0.02% :     0.000010s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
24.53% :     0.012273s :      1: jit_opt_a
 0.32% :     0.000161s :      1: jit_opt_after_cconv
 0.11% :     0.000055s :      1: jit_opt_b
 0.99% :     0.000494s :      1: loop_unroll
 1.16% :     0.000582s :      1: mutable_eliminate
 4.96% :     0.002481s :     39: opt.transform.jit_opt_a
 0.11% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000029s :      4: opt.transform.jit_opt_b
 0.03% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000041s :      4: opt.transform.symbol_engine_opt
 1.00% :     0.000501s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.05% :     0.000027s :      1: py_interpret_to_execute
 0.03% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000018s :      1: remove_dup_value
 3.49% :     0.001745s :      2: renormalize.infer
 2.68% :     0.001343s :      2: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000041s :      1: rewriter_after_opt_a
 0.17% :     0.000087s :      1: rewriter_before_opt_a
 0.18% :     0.000089s :      1: symbol_engine_optimizer
14.67% :     0.007343s :      1: task_emit
43.49% :     0.021762s :      1: type_inference
 0.13% :     0.000066s :      1: validate


TotalTime = 0.0128892, [30]
[bootstrap]: 0.00026932
[type_inference]: 0.00447397
[event_method]: 1.78e-05
[auto_monad]: 6.481e-05
[graph_reusing]: 6.43e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 3.537e-05
[rewriter_before_opt_a]: 7.323e-05
[expand_dump_flag]: 3.03e-06
[jit_opt_a]: 0.0057777, [2]
    [Cycle 1]: 0.00180808, [27]
        [switch_simplify]: 6.128e-05
        [loop_unroll]: 3.068e-05
        [a_1]: 0.00058525
        [with_stream_mark]: 2.186e-05
        [recompute_prepare]: 7.66999e-06
        [updatestate_depend_eliminate]: 3.98999e-06
        [updatestate_assign_eliminate]: 3.41001e-06
        [updatestate_loads_eliminate]: 2.84999e-06
        [parameter_eliminate]: 2.07001e-06
        [specialize_transform]: 6.37001e-06
        [updatestate_useless_node_eliminater]: 5.46002e-06
        [accelerated_algorithm]: 5.64e-06
        [meta_shard_fg_expand]: 2.49001e-06
        [get_grad_eliminate_]: 5.36002e-06
        [merge_forward]: 3.60998e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.397e-05
        [j_node_and_user_rematch]: 8.97999e-06
        [meta_fg_expand]: 2.58e-06
        [replace_old_param]: 9.56e-06
        [inline_without_move]: 5.66998e-06
        [renormalize]: 0.00076699
        [add_forward_monad_depend]: 5.64e-06
        [auto_monad_grad]: 2.51998e-06
        [auto_monad_eliminator]: 1.51e-05
        [cse]: 2.947e-05
        [replace_applicator]: 1.241e-05
    [Cycle 2]: 0.00034087, [27]
        [switch_simplify]: 5.59998e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 9.156e-05
        [with_stream_mark]: 1.031e-05
        [recompute_prepare]: 5.02e-06
        [updatestate_depend_eliminate]: 2.89999e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 4.98001e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.58999e-06
        [merge_forward]: 2.71e-06
        [cell_reuse_recompute_pass]: 1.36002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.379e-05
        [j_node_and_user_rematch]: 7.91001e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 7.33999e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.09998e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 5.21002e-06
        [cse]: 1.102e-05
        [replace_applicator]: 1.672e-05
[py_interpret_to_execute_after_opt_a]: 1.228e-05
[rewriter_after_opt_a]: 3.596e-05
[convert_after_rewriter]: 7.2e-06
[order_py_execute_after_rewriter]: 4.97999e-06
[mutable_eliminate]: 0.00058029
[jit_opt_b]: 5.134e-05, [1]
    [Cycle 1]: 4.437e-05, [2]
        [frontend_op_eliminate]: 1.723e-05
        [inline_after_opt_a]: 1.555e-05
[cconv]: 2.653e-05
[loop_unroll]: 0.00043929
[jit_opt_after_cconv]: 0.00014256, [1]
    [Cycle 1]: 0.00013615, [11]
        [c_1]: 2.226e-05
        [parameter_eliminate]: 2.50002e-06
        [updatestate_depend_eliminate]: 6.49001e-06
        [updatestate_assign_eliminate]: 2.35997e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 1.858e-05
        [call_graph_tuple_transform]: 1.813e-05
        [tuple_list_get_item_eliminator]: 5.75001e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 5.19998e-07
        [switch_simplify]: 5.56e-06
[remove_dup_value]: 1.635e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 5.55001e-06
[add_recomputation]: 4.963e-05
[cse_after_recomputation]: 1.983e-05, [1]
    [Cycle 1]: 1.45e-05, [1]
        [cse]: 8.99e-06
[auto_monad_reorder]: 1.444e-05
[get_jit_bprop_graph]: 2.26998e-06
[rewriter_after_jit_bprop_graph]: 5.69e-06
[opt_after_jit_grad]: 0.00046796
[symbol_engine_optimizer]: 7.372e-05, [1]
    [Cycle 1]: 6.702e-05, [6]
        [build]: 3.81001e-06
        [elim_shapecalc]: 7.61999e-06
        [elim_not_effective]: 1.256e-05
        [opt_reshape]: 5.46998e-06
        [fold_const_symbol]: 8.68001e-06
        [renormalize]: 4.40021e-07
[validate]: 3.832e-05
Sums
    bootstrap                            :     0.000269s :  3.11%
    type_inference                       :     0.004474s : 51.63%
    event_method                         :     0.000018s :  0.21%
    auto_monad                           :     0.000065s :  0.75%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.41%
    rewriter_before_opt_a                :     0.000073s :  0.85%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000067s :  0.77%
    jit_opt_a.loop_unroll                :     0.000036s :  0.41%
    jit_opt_a.a_1                        :     0.000677s :  7.81%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.37%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000767s :  8.85%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.23%
    jit_opt_a.cse                        :     0.000040s :  0.47%
    jit_opt_a.replace_applicator         :     0.000029s :  0.34%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.14%
    rewriter_after_opt_a                 :     0.000036s :  0.42%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000580s :  6.70%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.18%
    cconv                                :     0.000027s :  0.31%
    loop_unroll                          :     0.000439s :  5.07%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000016s :  0.19%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000050s :  0.57%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000014s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000468s :  5.40%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000038s :  0.44%


Time group info:
------[substitution.]   0.000232    23
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.11% :     0.000005s :      3: substitution.graph_param_transform
85.30% :     0.000198s :      4: substitution.inline
 1.32% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.06% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.10% :     0.000005s :      2: substitution.replace_old_param
 5.62% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004409     2
79.02% :     0.003484s :      1: type_inference.infer
20.98% :     0.000925s :      1: type_inference.specialize
------[replace.]   0.000073     6
67.91% :     0.000049s :      4: replace.inline
32.09% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000208     6
94.26% :     0.000196s :      4: match.inline
 5.74% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.53% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.55% :     0.000002s :     10: predicate.addn_zero_filter
 2.05% :     0.000003s :     10: predicate.arithmetic_simplify
 1.16% :     0.000001s :     10: predicate.cast_eliminate
 0.58% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.46% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.07% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.63% :     0.000007s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.49% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.83% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.39% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 1.53% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.01% :     0.000002s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.82% :     0.000010s :     53: predicate.switch_simplify
 1.08% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000002s :     10: predicate.transpose_eliminate
 1.56% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.27% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.73% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000667    12
46.40% :     0.000310s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.68% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.92% :     0.000313s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014614    72
 0.36% :     0.000053s :      1: add_recomputation
 0.47% :     0.000068s :      1: auto_monad
 0.12% :     0.000017s :      1: auto_monad_reorder
 1.92% :     0.000281s :      1: bootstrap
 0.20% :     0.000029s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.15% :     0.000021s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
39.56% :     0.005781s :      1: jit_opt_a
 0.99% :     0.000145s :      1: jit_opt_after_cconv
 0.37% :     0.000054s :      1: jit_opt_b
 3.06% :     0.000447s :      1: loop_unroll
 4.03% :     0.000589s :      1: mutable_eliminate
 6.20% :     0.000906s :     26: opt.transform.jit_opt_a
 0.33% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.26% :     0.000477s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000038s :      1: py_interpret_to_execute
 0.10% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000019s :      1: remove_dup_value
 2.80% :     0.000409s :      1: renormalize.infer
 2.39% :     0.000350s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000039s :      1: rewriter_after_opt_a
 0.52% :     0.000076s :      1: rewriter_before_opt_a
 0.52% :     0.000076s :      1: symbol_engine_optimizer
30.72% :     0.004490s :      1: type_inference


TotalTime = 0.031415, [33]
[bootstrap]: 0.00036618
[type_inference]: 0.00568022
[event_method]: 2.046e-05
[auto_monad]: 7.275e-05
[graph_reusing]: 6.38e-06
[pre_auto_parallel]: 3.09001e-06
[py_interpret_to_execute]: 3.704e-05
[rewriter_before_opt_a]: 7.781e-05
[expand_dump_flag]: 3.5e-06
[jit_opt_a]: 0.00934781, [2]
    [Cycle 1]: 0.00217756, [27]
        [switch_simplify]: 7.266e-05
        [loop_unroll]: 2.953e-05
        [a_1]: 0.00066392
        [with_stream_mark]: 3.034e-05
        [recompute_prepare]: 9.84999e-06
        [updatestate_depend_eliminate]: 4.38999e-06
        [updatestate_assign_eliminate]: 2.86999e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [parameter_eliminate]: 2.44999e-06
        [specialize_transform]: 5.74e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.90002e-06
        [meta_shard_fg_expand]: 2.96999e-06
        [get_grad_eliminate_]: 5.81e-06
        [merge_forward]: 3.86001e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.633e-05
        [j_node_and_user_rematch]: 1.07e-05
        [meta_fg_expand]: 2.36e-06
        [replace_old_param]: 1.095e-05
        [inline_without_move]: 6.17999e-06
        [renormalize]: 0.00097674
        [add_forward_monad_depend]: 7.93001e-06
        [auto_monad_grad]: 3.04999e-06
        [auto_monad_eliminator]: 1.979e-05
        [cse]: 3.324e-05
        [replace_applicator]: 1.99e-05
    [Cycle 2]: 0.0003462, [27]
        [switch_simplify]: 6.04001e-06
        [loop_unroll]: 5.34998e-06
        [a_1]: 9.187e-05
        [with_stream_mark]: 1.433e-05
        [recompute_prepare]: 4.98001e-06
        [updatestate_depend_eliminate]: 3.4e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 2.14999e-06
        [specialize_transform]: 4.98001e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 4.95001e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 4.48999e-06
        [merge_forward]: 2.96001e-06
        [cell_reuse_recompute_pass]: 3.91999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.705e-05
        [j_node_and_user_rematch]: 8.94e-06
        [meta_fg_expand]: 2.08002e-06
        [replace_old_param]: 8.59e-06
        [inline_without_move]: 4.59998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.64998e-06
        [auto_monad_grad]: 1.13001e-06
        [auto_monad_eliminator]: 5.44998e-06
        [cse]: 1.275e-05
        [replace_applicator]: 4.90999e-06
[py_interpret_to_execute_after_opt_a]: 1.721e-05
[rewriter_after_opt_a]: 4.413e-05
[convert_after_rewriter]: 8.15999e-06
[order_py_execute_after_rewriter]: 5.59998e-06
[mutable_eliminate]: 0.00086216
[jit_opt_b]: 5.583e-05, [1]
    [Cycle 1]: 4.568e-05, [2]
        [frontend_op_eliminate]: 1.616e-05
        [inline_after_opt_a]: 1.665e-05
[cconv]: 3.581e-05
[loop_unroll]: 0.00055676
[jit_opt_after_cconv]: 0.00018489, [1]
    [Cycle 1]: 0.00017661, [11]
        [c_1]: 2.206e-05
        [parameter_eliminate]: 6.54001e-06
        [updatestate_depend_eliminate]: 1.26e-05
        [updatestate_assign_eliminate]: 2.91999e-06
        [updatestate_loads_eliminate]: 2.92002e-06
        [cse]: 3.782e-05
        [call_graph_tuple_transform]: 2.649e-05
        [tuple_list_get_item_eliminator]: 5.28002e-06
        [none_parameter_eliminate]: 1.91003e-06
        [renormalize]: 1.19e-06
        [switch_simplify]: 5.89999e-06
[remove_dup_value]: 1.691e-05
[partial_unused_args_eliminate]: 2.66e-06
[environ_conv]: 6.44001e-06
[add_recomputation]: 6.236e-05
[cse_after_recomputation]: 2.286e-05, [1]
    [Cycle 1]: 1.63e-05, [1]
        [cse]: 9.76e-06
[auto_monad_reorder]: 1.625e-05
[get_jit_bprop_graph]: 2.69001e-06
[rewriter_after_jit_bprop_graph]: 1.039e-05
[opt_after_jit_grad]: 0.00056603
[symbol_engine_optimizer]: 8.323e-05, [1]
    [Cycle 1]: 7.491e-05, [6]
        [build]: 6.04999e-06
        [elim_shapecalc]: 8.59002e-06
        [elim_not_effective]: 1.572e-05
        [opt_reshape]: 5.40999e-06
        [fold_const_symbol]: 8.84998e-06
        [renormalize]: 1.10999e-06
[validate]: 4.905e-05
[backend_pass]: 1.23002e-06
[task_emit]: 0.0128912
[execute]: 1.069e-05
Sums
    bootstrap                            :     0.000366s :  1.54%
    type_inference                       :     0.005680s : 23.83%
    event_method                         :     0.000020s :  0.09%
    auto_monad                           :     0.000073s :  0.31%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000037s :  0.16%
    rewriter_before_opt_a                :     0.000078s :  0.33%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000079s :  0.33%
    jit_opt_a.loop_unroll                :     0.000035s :  0.15%
    jit_opt_a.a_1                        :     0.000756s :  3.17%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.08%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.000977s :  4.10%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.11%
    jit_opt_a.cse                        :     0.000046s :  0.19%
    jit_opt_a.replace_applicator         :     0.000025s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.07%
    rewriter_after_opt_a                 :     0.000044s :  0.19%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000862s :  3.62%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.07%
    cconv                                :     0.000036s :  0.15%
    loop_unroll                          :     0.000557s :  2.34%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000013s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000038s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000017s :  0.07%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000062s :  0.26%
    cse_after_recomputation.cse          :     0.000010s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.07%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.04%
    opt_after_jit_grad                   :     0.000566s :  2.38%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000049s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.012891s : 54.09%
    execute                              :     0.000011s :  0.04%


Time group info:
------[substitution.]   0.000304    23
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000002s :      2: substitution.fold_const_symbol
 2.16% :     0.000007s :      3: substitution.graph_param_transform
85.77% :     0.000261s :      4: substitution.inline
 1.69% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.65% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.38% :     0.000007s :      2: substitution.replace_old_param
 5.09% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005590     2
83.82% :     0.004686s :      1: type_inference.infer
16.18% :     0.000904s :      1: type_inference.specialize
------[replace.]   0.000077     6
70.64% :     0.000054s :      4: replace.inline
29.36% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000272     6
94.73% :     0.000258s :      4: match.inline
 5.27% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 2.48% :     0.000003s :     10: predicate.arithmetic_simplify
 1.41% :     0.000002s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 2.09% :     0.000003s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.10% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.70% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.43% :     0.000002s :     10: predicate.environ_get_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 7.49% :     0.000010s :     22: predicate.inline
 1.26% :     0.000002s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000002s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 2.24% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.34% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.73% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 2.85% :     0.000004s :      3: predicate.mutable_eliminate
 0.39% :     0.000001s :      3: predicate.opt_reshape
 2.04% :     0.000003s :     15: predicate.partial_eliminate
 1.08% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.28% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.44% :     0.000001s :      3: predicate.reset_defer_inline
 1.22% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.06% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.77% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.02% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.61% :     0.000010s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000002s :     10: predicate.transpose_eliminate
 1.66% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.95% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.86% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.26% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.46% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000716    12
49.26% :     0.000353s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 9.10% :     0.000065s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.65% :     0.000298s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.033504    76
 0.19% :     0.000065s :      1: add_recomputation
 0.23% :     0.000077s :      1: auto_monad
 0.06% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.17% :     0.000394s :      1: bootstrap
 0.12% :     0.000039s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.08% :     0.000026s :      1: event_method
 0.05% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
27.91% :     0.009352s :      1: jit_opt_a
 0.56% :     0.000188s :      1: jit_opt_after_cconv
 0.18% :     0.000059s :      1: jit_opt_b
 1.70% :     0.000569s :      1: loop_unroll
 2.63% :     0.000881s :      1: mutable_eliminate
 2.98% :     0.000998s :     26: opt.transform.jit_opt_a
 0.17% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.73% :     0.000580s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000040s :      1: py_interpret_to_execute
 0.06% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000020s :      1: remove_dup_value
 1.77% :     0.000594s :      1: renormalize.infer
 1.11% :     0.000372s :      1: renormalize.specialize
 0.04% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000048s :      1: rewriter_after_opt_a
 0.24% :     0.000081s :      1: rewriter_before_opt_a
 0.26% :     0.000086s :      1: symbol_engine_optimizer
38.55% :     0.012915s :      1: task_emit
17.05% :     0.005713s :      1: type_inference
 0.24% :     0.000079s :      1: validate


TotalTime = 0.0131024, [30]
[bootstrap]: 0.00026901
[type_inference]: 0.00464872
[event_method]: 1.813e-05
[auto_monad]: 6.735e-05
[graph_reusing]: 5.33002e-06
[pre_auto_parallel]: 2.27999e-06
[py_interpret_to_execute]: 3.668e-05
[rewriter_before_opt_a]: 7.482e-05
[expand_dump_flag]: 3.26001e-06
[jit_opt_a]: 0.00584449, [2]
    [Cycle 1]: 0.0017517, [27]
        [switch_simplify]: 5.531e-05
        [loop_unroll]: 3.201e-05
        [a_1]: 0.0005705
        [with_stream_mark]: 1.92e-05
        [recompute_prepare]: 7.13e-06
        [updatestate_depend_eliminate]: 4.1e-06
        [updatestate_assign_eliminate]: 3.13998e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 6.51e-06
        [updatestate_useless_node_eliminater]: 5.40999e-06
        [accelerated_algorithm]: 6.48998e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 5.47001e-06
        [merge_forward]: 4.03001e-06
        [cell_reuse_recompute_pass]: 1.38002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.413e-05
        [j_node_and_user_rematch]: 9.92999e-06
        [meta_fg_expand]: 2.74001e-06
        [replace_old_param]: 8.97999e-06
        [inline_without_move]: 5.77001e-06
        [renormalize]: 0.00075026
        [add_forward_monad_depend]: 5.20999e-06
        [auto_monad_grad]: 2.41998e-06
        [auto_monad_eliminator]: 1.308e-05
        [cse]: 2.879e-05
        [replace_applicator]: 1.106e-05
    [Cycle 2]: 0.0003307, [27]
        [switch_simplify]: 6.09001e-06
        [loop_unroll]: 5.20999e-06
        [a_1]: 9.199e-05
        [with_stream_mark]: 8.55999e-06
        [recompute_prepare]: 4.97e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 5.24e-06
        [updatestate_useless_node_eliminater]: 4.94998e-06
        [accelerated_algorithm]: 5.59998e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.72e-06
        [merge_forward]: 2.61999e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.311e-05
        [j_node_and_user_rematch]: 8.15999e-06
        [meta_fg_expand]: 1.66998e-06
        [replace_old_param]: 7.75998e-06
        [inline_without_move]: 4.75001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.27999e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 5.32999e-06
        [cse]: 1.05e-05
        [replace_applicator]: 5.39e-06
[py_interpret_to_execute_after_opt_a]: 9.86e-06
[rewriter_after_opt_a]: 3.231e-05
[convert_after_rewriter]: 7.21999e-06
[order_py_execute_after_rewriter]: 1.835e-05
[mutable_eliminate]: 0.00054131
[jit_opt_b]: 6.477e-05, [1]
    [Cycle 1]: 5.754e-05, [2]
        [frontend_op_eliminate]: 2.947e-05
        [inline_after_opt_a]: 1.56e-05
[cconv]: 2.509e-05
[loop_unroll]: 0.00042838
[jit_opt_after_cconv]: 0.00014058, [1]
    [Cycle 1]: 0.00013372, [11]
        [c_1]: 2.156e-05
        [parameter_eliminate]: 2.43998e-06
        [updatestate_depend_eliminate]: 5.72001e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.703e-05
        [call_graph_tuple_transform]: 1.826e-05
        [tuple_list_get_item_eliminator]: 5.71e-06
        [none_parameter_eliminate]: 1.33002e-06
        [renormalize]: 4.10015e-07
        [switch_simplify]: 5.67999e-06
[remove_dup_value]: 1.413e-05
[partial_unused_args_eliminate]: 2.23002e-06
[environ_conv]: 5.15999e-06
[add_recomputation]: 4.847e-05
[cse_after_recomputation]: 2.001e-05, [1]
    [Cycle 1]: 1.465e-05, [1]
        [cse]: 9.13002e-06
[auto_monad_reorder]: 1.572e-05
[get_jit_bprop_graph]: 1.69998e-06
[rewriter_after_jit_bprop_graph]: 5.40001e-06
[opt_after_jit_grad]: 0.00046107
[symbol_engine_optimizer]: 7.4e-05, [1]
    [Cycle 1]: 6.732e-05, [6]
        [build]: 3.6e-06
        [elim_shapecalc]: 7.73999e-06
        [elim_not_effective]: 1.314e-05
        [opt_reshape]: 5.66998e-06
        [fold_const_symbol]: 8.88002e-06
        [renormalize]: 3.29979e-07
[validate]: 3.598e-05
Sums
    bootstrap                            :     0.000269s :  3.08%
    type_inference                       :     0.004649s : 53.17%
    event_method                         :     0.000018s :  0.21%
    auto_monad                           :     0.000067s :  0.77%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000037s :  0.42%
    rewriter_before_opt_a                :     0.000075s :  0.86%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000061s :  0.70%
    jit_opt_a.loop_unroll                :     0.000037s :  0.43%
    jit_opt_a.a_1                        :     0.000662s :  7.58%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000750s :  8.58%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.21%
    jit_opt_a.cse                        :     0.000039s :  0.45%
    jit_opt_a.replace_applicator         :     0.000016s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.11%
    rewriter_after_opt_a                 :     0.000032s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000018s :  0.21%
    mutable_eliminate                    :     0.000541s :  6.19%
    jit_opt_b.frontend_op_eliminate      :     0.000029s :  0.34%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.18%
    cconv                                :     0.000025s :  0.29%
    loop_unroll                          :     0.000428s :  4.90%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000014s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000048s :  0.55%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000461s :  5.27%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.41%


Time group info:
------[substitution.]   0.000221    23
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.29% :     0.000005s :      3: substitution.graph_param_transform
84.85% :     0.000187s :      4: substitution.inline
 1.45% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.05% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.79% :     0.000004s :      2: substitution.replace_old_param
 6.17% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004584     2
79.25% :     0.003632s :      1: type_inference.infer
20.75% :     0.000951s :      1: type_inference.specialize
------[replace.]   0.000072     6
67.77% :     0.000049s :      4: replace.inline
32.23% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000197     6
93.77% :     0.000185s :      4: match.inline
 6.23% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   691
 1.27% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.60% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.91% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000002s :     10: predicate.addn_zero_filter
 1.95% :     0.000003s :     10: predicate.arithmetic_simplify
 1.08% :     0.000001s :     10: predicate.cast_eliminate
 9.17% :     0.000012s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.12% :     0.000001s :     10: predicate.depend_value_elim
 0.99% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000001s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.19% :     0.000007s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000002s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.60% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.68% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 0.93% :     0.000001s :     10: predicate.minmaximum_grad
 1.52% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.00% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.31% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.49% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.20% :     0.000002s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 1.82% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.05% :     0.000009s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.23% :     0.000002s :     10: predicate.transpose_eliminate
 1.24% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.16% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.39% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.53% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000692    12
47.38% :     0.000328s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.32% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.29% :     0.000320s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014796    72
 0.35% :     0.000052s :      1: add_recomputation
 0.48% :     0.000071s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 1.93% :     0.000285s :      1: bootstrap
 0.19% :     0.000028s :      1: cconv
 0.08% :     0.000012s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
39.52% :     0.005847s :      1: jit_opt_a
 0.97% :     0.000143s :      1: jit_opt_after_cconv
 0.46% :     0.000068s :      1: jit_opt_b
 2.95% :     0.000437s :      1: loop_unroll
 3.72% :     0.000550s :      1: mutable_eliminate
 5.93% :     0.000878s :     26: opt.transform.jit_opt_a
 0.32% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.26% :     0.000039s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.17% :     0.000469s :      1: opt_after_jit_grad
 0.14% :     0.000021s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000040s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 2.69% :     0.000397s :      1: renormalize.infer
 2.33% :     0.000345s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000035s :      1: rewriter_after_opt_a
 0.52% :     0.000078s :      1: rewriter_before_opt_a
 0.52% :     0.000077s :      1: symbol_engine_optimizer
31.51% :     0.004663s :      1: type_inference


TotalTime = 0.0217151, [33]
[bootstrap]: 0.00034837
[type_inference]: 0.00554036
[event_method]: 1.885e-05
[auto_monad]: 6.22e-05
[graph_reusing]: 5.92001e-06
[pre_auto_parallel]: 2.32999e-06
[py_interpret_to_execute]: 3.901e-05
[rewriter_before_opt_a]: 7.208e-05
[expand_dump_flag]: 2.82002e-06
[jit_opt_a]: 0.00627012, [2]
    [Cycle 1]: 0.0020904, [27]
        [switch_simplify]: 6.438e-05
        [loop_unroll]: 2.882e-05
        [a_1]: 0.00060693
        [with_stream_mark]: 2.342e-05
        [recompute_prepare]: 8.69e-06
        [updatestate_depend_eliminate]: 4.08001e-06
        [updatestate_assign_eliminate]: 3.20002e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [parameter_eliminate]: 1.85001e-06
        [specialize_transform]: 6.51e-06
        [updatestate_useless_node_eliminater]: 5.15999e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 2.09e-06
        [get_grad_eliminate_]: 5.19998e-06
        [merge_forward]: 3.56001e-06
        [cell_reuse_recompute_pass]: 1.12999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.439e-05
        [j_node_and_user_rematch]: 9.02e-06
        [meta_fg_expand]: 2.54001e-06
        [replace_old_param]: 1.063e-05
        [inline_without_move]: 5.67999e-06
        [renormalize]: 0.00101039
        [add_forward_monad_depend]: 7.11999e-06
        [auto_monad_grad]: 2.44001e-06
        [auto_monad_eliminator]: 1.397e-05
        [cse]: 3.172e-05
        [replace_applicator]: 1.402e-05
    [Cycle 2]: 0.00033771, [27]
        [switch_simplify]: 5.74e-06
        [loop_unroll]: 5.02999e-06
        [a_1]: 9.462e-05
        [with_stream_mark]: 1.134e-05
        [recompute_prepare]: 5.00999e-06
        [updatestate_depend_eliminate]: 2.69001e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 9.00007e-07
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 4.87e-06
        [merge_forward]: 2.86e-06
        [cell_reuse_recompute_pass]: 1.80001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.383e-05
        [j_node_and_user_rematch]: 8.50999e-06
        [meta_fg_expand]: 1.79998e-06
        [replace_old_param]: 1.415e-05
        [inline_without_move]: 4.80001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.57001e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 5.87001e-06
        [cse]: 1.122e-05
        [replace_applicator]: 5.66e-06
[py_interpret_to_execute_after_opt_a]: 1.393e-05
[rewriter_after_opt_a]: 4.032e-05
[convert_after_rewriter]: 7.16999e-06
[order_py_execute_after_rewriter]: 5.20001e-06
[mutable_eliminate]: 0.00077967
[jit_opt_b]: 5.361e-05, [1]
    [Cycle 1]: 4.466e-05, [2]
        [frontend_op_eliminate]: 1.908e-05
        [inline_after_opt_a]: 1.484e-05
[cconv]: 3.056e-05
[loop_unroll]: 0.00045869
[jit_opt_after_cconv]: 0.0001583, [1]
    [Cycle 1]: 0.00015136, [11]
        [c_1]: 2.289e-05
        [parameter_eliminate]: 2.95002e-06
        [updatestate_depend_eliminate]: 6.99001e-06
        [updatestate_assign_eliminate]: 2.59001e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [cse]: 2.731e-05
        [call_graph_tuple_transform]: 2.074e-05
        [tuple_list_get_item_eliminator]: 5.99999e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 8.09989e-07
        [switch_simplify]: 5.56e-06
[remove_dup_value]: 1.374e-05
[partial_unused_args_eliminate]: 2.66e-06
[environ_conv]: 6.79001e-06
[add_recomputation]: 5.026e-05
[cse_after_recomputation]: 2.197e-05, [1]
    [Cycle 1]: 1.616e-05, [1]
        [cse]: 9.52999e-06
[auto_monad_reorder]: 1.522e-05
[get_jit_bprop_graph]: 2.09999e-06
[rewriter_after_jit_bprop_graph]: 8.12e-06
[opt_after_jit_grad]: 0.00047656
[symbol_engine_optimizer]: 7.556e-05, [1]
    [Cycle 1]: 6.807e-05, [6]
        [build]: 4.05e-06
        [elim_shapecalc]: 7.48e-06
        [elim_not_effective]: 1.39e-05
        [opt_reshape]: 5.81003e-06
        [fold_const_symbol]: 8.65001e-06
        [renormalize]: 7.7e-07
[validate]: 4.353e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00685045
[execute]: 8.32998e-06
Sums
    bootstrap                            :     0.000348s :  2.02%
    type_inference                       :     0.005540s : 32.19%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000062s :  0.36%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.23%
    rewriter_before_opt_a                :     0.000072s :  0.42%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000070s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000702s :  4.08%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000025s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001010s :  5.87%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000043s :  0.25%
    jit_opt_a.replace_applicator         :     0.000020s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000040s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000780s :  4.53%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000031s :  0.18%
    loop_unroll                          :     0.000459s :  2.67%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000050s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000477s :  2.77%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006850s : 39.80%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000272    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 2.46% :     0.000007s :      3: substitution.graph_param_transform
84.32% :     0.000229s :      4: substitution.inline
 1.28% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.68% :     0.000005s :      4: substitution.remove_not_recompute_node
 4.49% :     0.000012s :      2: substitution.replace_old_param
 4.51% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005465     2
82.84% :     0.004527s :      1: type_inference.infer
17.16% :     0.000938s :      1: type_inference.specialize
------[replace.]   0.000084     6
68.24% :     0.000057s :      4: replace.inline
31.76% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000238     6
95.30% :     0.000227s :      4: match.inline
 4.70% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.19% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.04% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.69% :     0.000002s :     10: predicate.addn_zero_filter
 2.37% :     0.000003s :     10: predicate.arithmetic_simplify
 1.25% :     0.000002s :     10: predicate.cast_eliminate
 0.59% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.59% :     0.000002s :     10: predicate.depend_value_elim
 1.22% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.08% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.40% :     0.000002s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.06% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.54% :     0.000002s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.88% :     0.000002s :      3: predicate.mutable_eliminate
 0.58% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.55% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.82% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.85% :     0.000002s :     18: predicate.replace_applicator
 0.82% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.34% :     0.000002s :      6: predicate.specialize_transform
 1.22% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.37% :     0.000009s :     53: predicate.switch_simplify
 1.42% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.18% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.36% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000671    12
45.05% :     0.000302s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.96% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.99% :     0.000315s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023758    76
 0.23% :     0.000053s :      1: add_recomputation
 0.28% :     0.000066s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.55% :     0.000368s :      1: bootstrap
 0.14% :     0.000033s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.41% :     0.006274s :      1: jit_opt_a
 0.68% :     0.000161s :      1: jit_opt_after_cconv
 0.24% :     0.000056s :      1: jit_opt_b
 1.97% :     0.000467s :      1: loop_unroll
 3.33% :     0.000791s :      1: mutable_eliminate
 3.93% :     0.000933s :     26: opt.transform.jit_opt_a
 0.22% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000028s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.04% :     0.000485s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000042s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.48% :     0.000588s :      1: renormalize.infer
 1.74% :     0.000413s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000044s :      1: rewriter_after_opt_a
 0.31% :     0.000075s :      1: rewriter_before_opt_a
 0.33% :     0.000079s :      1: symbol_engine_optimizer
28.91% :     0.006870s :      1: task_emit
23.39% :     0.005557s :      1: type_inference
 0.28% :     0.000066s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.592.891 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc41940> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.593.623 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.028209, [33]
[bootstrap]: 0.00039797
[type_inference]: 0.00601877
[event_method]: 2.35e-05
[auto_monad]: 7.762e-05
[graph_reusing]: 5.96e-06
[pre_auto_parallel]: 3.18e-06
[py_interpret_to_execute]: 4.673e-05
[rewriter_before_opt_a]: 8.058e-05
[expand_dump_flag]: 3.27002e-06
[jit_opt_a]: 0.00964094, [2]
    [Cycle 1]: 0.00456471, [27]
        [switch_simplify]: 0.00227579
        [loop_unroll]: 4.167e-05
        [a_1]: 0.00069793
        [with_stream_mark]: 3.099e-05
        [recompute_prepare]: 9.29998e-06
        [updatestate_depend_eliminate]: 4.80999e-06
        [updatestate_assign_eliminate]: 3.00002e-06
        [updatestate_loads_eliminate]: 2.50002e-06
        [parameter_eliminate]: 2.75002e-06
        [specialize_transform]: 6.53e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 6.04999e-06
        [meta_shard_fg_expand]: 2.79001e-06
        [get_grad_eliminate_]: 5.40999e-06
        [merge_forward]: 4.45e-06
        [cell_reuse_recompute_pass]: 2.02999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.691e-05
        [j_node_and_user_rematch]: 1.086e-05
        [meta_fg_expand]: 3.71999e-06
        [replace_old_param]: 1.069e-05
        [inline_without_move]: 5.22999e-06
        [renormalize]: 0.00103452
        [add_forward_monad_depend]: 1.1e-05
        [auto_monad_grad]: 2.88998e-06
        [auto_monad_eliminator]: 2.372e-05
        [cse]: 4.244e-05
        [replace_applicator]: 2.381e-05
    [Cycle 2]: 0.00037511, [27]
        [switch_simplify]: 6.23998e-06
        [loop_unroll]: 5.41998e-06
        [a_1]: 9.962e-05
        [with_stream_mark]: 1.737e-05
        [recompute_prepare]: 5.81998e-06
        [updatestate_depend_eliminate]: 4.13001e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 4.93001e-06
        [updatestate_useless_node_eliminater]: 5.25999e-06
        [accelerated_algorithm]: 5.30999e-06
        [meta_shard_fg_expand]: 2.11003e-06
        [get_grad_eliminate_]: 4.68999e-06
        [merge_forward]: 4.12e-06
        [cell_reuse_recompute_pass]: 3.38e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.12e-05
        [j_node_and_user_rematch]: 8.99e-06
        [meta_fg_expand]: 1.97001e-06
        [replace_old_param]: 1.044e-05
        [inline_without_move]: 5.76998e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.50001e-06
        [auto_monad_grad]: 1.09e-06
        [auto_monad_eliminator]: 6.51e-06
        [cse]: 1.392e-05
        [replace_applicator]: 5.19998e-06
[py_interpret_to_execute_after_opt_a]: 2.065e-05
[rewriter_after_opt_a]: 4.728e-05
[convert_after_rewriter]: 7.5e-06
[order_py_execute_after_rewriter]: 5.87999e-06
[mutable_eliminate]: 0.00090146
[jit_opt_b]: 6.123e-05, [1]
    [Cycle 1]: 5.064e-05, [2]
        [frontend_op_eliminate]: 1.93e-05
        [inline_after_opt_a]: 1.879e-05
[cconv]: 4.22e-05
[loop_unroll]: 0.00060368
[jit_opt_after_cconv]: 0.00019183, [1]
    [Cycle 1]: 0.00018311, [11]
        [c_1]: 2.297e-05
        [parameter_eliminate]: 7.26001e-06
        [updatestate_depend_eliminate]: 1.276e-05
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 3.26999e-06
        [cse]: 3.789e-05
        [call_graph_tuple_transform]: 2.931e-05
        [tuple_list_get_item_eliminator]: 5.77001e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 8.10018e-07
        [switch_simplify]: 5.86998e-06
[remove_dup_value]: 1.696e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 7.4e-06
[add_recomputation]: 6.038e-05
[cse_after_recomputation]: 2.312e-05, [1]
    [Cycle 1]: 1.68e-05, [1]
        [cse]: 9.66e-06
[auto_monad_reorder]: 1.655e-05
[get_jit_bprop_graph]: 2.19999e-06
[rewriter_after_jit_bprop_graph]: 1.204e-05
[opt_after_jit_grad]: 0.00065213
[symbol_engine_optimizer]: 8.606e-05, [1]
    [Cycle 1]: 7.696e-05, [6]
        [build]: 6.41e-06
        [elim_shapecalc]: 8.43999e-06
        [elim_not_effective]: 1.741e-05
        [opt_reshape]: 5.82999e-06
        [fold_const_symbol]: 8.97999e-06
        [renormalize]: 5.29981e-07
[validate]: 5.082e-05
[backend_pass]: 1.39998e-06
[task_emit]: 0.00880645
[execute]: 1.134e-05
Sums
    bootstrap                            :     0.000398s :  1.75%
    type_inference                       :     0.006019s : 26.53%
    event_method                         :     0.000024s :  0.10%
    auto_monad                           :     0.000078s :  0.34%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000047s :  0.21%
    rewriter_before_opt_a                :     0.000081s :  0.36%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.002282s : 10.06%
    jit_opt_a.loop_unroll                :     0.000047s :  0.21%
    jit_opt_a.a_1                        :     0.000798s :  3.51%
    jit_opt_a.with_stream_mark           :     0.000048s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.03%
    jit_opt_a.replace_old_param          :     0.000021s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.001035s :  4.56%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000030s :  0.13%
    jit_opt_a.cse                        :     0.000056s :  0.25%
    jit_opt_a.replace_applicator         :     0.000029s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000021s :  0.09%
    rewriter_after_opt_a                 :     0.000047s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000901s :  3.97%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.08%
    cconv                                :     0.000042s :  0.19%
    loop_unroll                          :     0.000604s :  2.66%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000013s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000038s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000029s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000060s :  0.27%
    cse_after_recomputation.cse          :     0.000010s :  0.04%
    auto_monad_reorder                   :     0.000017s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000012s :  0.05%
    opt_after_jit_grad                   :     0.000652s :  2.87%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000051s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008806s : 38.81%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000325    23
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000002s :      2: substitution.fold_const_symbol
 2.06% :     0.000007s :      3: substitution.graph_param_transform
86.13% :     0.000280s :      4: substitution.inline
 1.45% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.81% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.66% :     0.000009s :      2: substitution.replace_old_param
 4.63% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005923     2
82.96% :     0.004914s :      1: type_inference.infer
17.04% :     0.001009s :      1: type_inference.specialize
------[replace.]   0.000084     6
70.20% :     0.000059s :      4: replace.inline
29.80% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000291     6
95.30% :     0.000277s :      4: match.inline
 4.70% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000158   691
 1.27% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.48% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.77% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000002s :     10: predicate.addn_zero_filter
 2.33% :     0.000004s :     10: predicate.arithmetic_simplify
 1.79% :     0.000003s :     10: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.77% :     0.000001s :     10: predicate.compare_switch_simplify
 1.79% :     0.000003s :     10: predicate.depend_value_elim
 0.79% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.05% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.13% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000001s :      3: predicate.elim_not_effective
 0.42% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.81% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.89% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.94% :     0.000001s :     10: predicate.environ_get_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 6.11% :     0.000010s :     22: predicate.inline
 0.74% :     0.000001s :      6: predicate.inline_without_move
 0.27% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.40% :     0.000002s :      6: predicate.less_batch_normalization
 1.65% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.61% :     0.000003s :     15: predicate.load_eliminater
 2.03% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000007s :     34: predicate.loop_unroll_before_grad
 1.58% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.85% :     0.000001s :     10: predicate.merge_addn
 0.89% :     0.000001s :     10: predicate.minmaximum_grad
 2.51% :     0.000004s :      3: predicate.mutable_eliminate
 0.38% :     0.000001s :      3: predicate.opt_reshape
 1.76% :     0.000003s :     15: predicate.partial_eliminate
 0.97% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000003s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.25% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.42% :     0.000001s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.10% :     0.000002s :      3: predicate.row_tensor_eliminate
 0.94% :     0.000001s :     10: predicate.same_eliminate
 0.37% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 0.73% :     0.000001s :      6: predicate.specialize_transform
 1.25% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.10% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.08% :     0.000003s :     16: predicate.switch_defer_inline
 1.61% :     0.000003s :     16: predicate.switch_layer_defer_inline
13.42% :     0.000021s :     53: predicate.switch_simplify
 1.46% :     0.000002s :     10: predicate.tile_eliminate
 0.95% :     0.000002s :     10: predicate.transpose_eliminate
 1.13% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.94% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.86% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.12% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.19% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.07% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.01% :     0.000003s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.54% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000797    12
48.59% :     0.000388s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.05% :     0.000064s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.35% :     0.000346s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.032635    76
 0.19% :     0.000063s :      1: add_recomputation
 0.25% :     0.000081s :      1: auto_monad
 0.06% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.29% :     0.000420s :      1: bootstrap
 0.14% :     0.000045s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000011s :      1: environ_conv
 0.09% :     0.000029s :      1: event_method
 0.05% :     0.000018s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
29.55% :     0.009645s :      1: jit_opt_a
 0.60% :     0.000195s :      1: jit_opt_after_cconv
 0.20% :     0.000064s :      1: jit_opt_b
 1.88% :     0.000614s :      1: loop_unroll
 2.81% :     0.000918s :      1: mutable_eliminate
 9.99% :     0.003262s :     26: opt.transform.jit_opt_a
 0.18% :     0.000060s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000029s :      4: opt.transform.jit_opt_b
 0.05% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000024s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.04% :     0.000665s :      1: opt_after_jit_grad
 0.03% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.15% :     0.000049s :      1: py_interpret_to_execute
 0.07% :     0.000023s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000020s :      1: remove_dup_value
 1.74% :     0.000568s :      1: renormalize.infer
 1.40% :     0.000456s :      1: renormalize.specialize
 0.04% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000051s :      1: rewriter_after_opt_a
 0.26% :     0.000084s :      1: rewriter_before_opt_a
 0.27% :     0.000089s :      1: symbol_engine_optimizer
27.06% :     0.008831s :      1: task_emit
18.53% :     0.006049s :      1: type_inference
 0.24% :     0.000080s :      1: validate


TotalTime = 0.0142042, [30]
[bootstrap]: 0.00032284
[type_inference]: 0.0045477
[event_method]: 1.949e-05
[auto_monad]: 6.808e-05
[graph_reusing]: 6.59001e-06
[pre_auto_parallel]: 2.70002e-06
[py_interpret_to_execute]: 3.648e-05
[rewriter_before_opt_a]: 7.707e-05
[expand_dump_flag]: 3.68999e-06
[jit_opt_a]: 0.00660589, [2]
    [Cycle 1]: 0.00230463, [27]
        [switch_simplify]: 7.828e-05
        [loop_unroll]: 3.364e-05
        [a_1]: 0.00075122
        [with_stream_mark]: 2.575e-05
        [recompute_prepare]: 1.52e-05
        [updatestate_depend_eliminate]: 4.25e-06
        [updatestate_assign_eliminate]: 3.00002e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [parameter_eliminate]: 2.41998e-06
        [specialize_transform]: 6.09999e-06
        [updatestate_useless_node_eliminater]: 5.58997e-06
        [accelerated_algorithm]: 5.77999e-06
        [meta_shard_fg_expand]: 2.48002e-06
        [get_grad_eliminate_]: 6.36e-06
        [merge_forward]: 4.13001e-06
        [cell_reuse_recompute_pass]: 1.35001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.741e-05
        [j_node_and_user_rematch]: 1.094e-05
        [meta_fg_expand]: 3.48999e-06
        [replace_old_param]: 1.208e-05
        [inline_without_move]: 6.46e-06
        [renormalize]: 0.0010117
        [add_forward_monad_depend]: 6.24001e-06
        [auto_monad_grad]: 2.53e-06
        [auto_monad_eliminator]: 1.744e-05
        [cse]: 3.275e-05
        [replace_applicator]: 1.361e-05
    [Cycle 2]: 0.00037533, [27]
        [switch_simplify]: 6.48e-06
        [loop_unroll]: 5.18002e-06
        [a_1]: 0.00010071
        [with_stream_mark]: 1.035e-05
        [recompute_prepare]: 5.70001e-06
        [updatestate_depend_eliminate]: 3.43999e-06
        [updatestate_assign_eliminate]: 2.14999e-06
        [updatestate_loads_eliminate]: 2.34999e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 5.71998e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 1.71e-06
        [get_grad_eliminate_]: 4.60999e-06
        [merge_forward]: 2.84001e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.457e-05
        [j_node_and_user_rematch]: 9.79e-06
        [meta_fg_expand]: 1.92999e-06
        [replace_old_param]: 8.19002e-06
        [inline_without_move]: 5.80002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.47001e-06
        [auto_monad_grad]: 9.09989e-07
        [auto_monad_eliminator]: 7.31001e-06
        [cse]: 1.242e-05
        [replace_applicator]: 5.89999e-06
[py_interpret_to_execute_after_opt_a]: 1.522e-05
[rewriter_after_opt_a]: 3.617e-05
[convert_after_rewriter]: 7.31001e-06
[order_py_execute_after_rewriter]: 5.05999e-06
[mutable_eliminate]: 0.00062937
[jit_opt_b]: 5.644e-05, [1]
    [Cycle 1]: 4.768e-05, [2]
        [frontend_op_eliminate]: 1.968e-05
        [inline_after_opt_a]: 1.554e-05
[cconv]: 2.714e-05
[loop_unroll]: 0.00051485
[jit_opt_after_cconv]: 0.00015885, [1]
    [Cycle 1]: 0.00015112, [11]
        [c_1]: 2.373e-05
        [parameter_eliminate]: 2.86e-06
        [updatestate_depend_eliminate]: 7.3e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.27001e-06
        [cse]: 2.038e-05
        [call_graph_tuple_transform]: 2.159e-05
        [tuple_list_get_item_eliminator]: 7.15e-06
        [none_parameter_eliminate]: 1.63002e-06
        [renormalize]: 6.89994e-07
        [switch_simplify]: 7.43999e-06
[remove_dup_value]: 1.469e-05
[partial_unused_args_eliminate]: 2.42001e-06
[environ_conv]: 5.81e-06
[add_recomputation]: 6.158e-05
[cse_after_recomputation]: 2.428e-05, [1]
    [Cycle 1]: 1.82e-05, [1]
        [cse]: 9.84001e-06
[auto_monad_reorder]: 1.637e-05
[get_jit_bprop_graph]: 1.99999e-06
[rewriter_after_jit_bprop_graph]: 7.17997e-06
[opt_after_jit_grad]: 0.0005896
[symbol_engine_optimizer]: 8.013e-05, [1]
    [Cycle 1]: 7.162e-05, [6]
        [build]: 3.93001e-06
        [elim_shapecalc]: 7.83001e-06
        [elim_not_effective]: 1.396e-05
        [opt_reshape]: 7.07002e-06
        [fold_const_symbol]: 9.07001e-06
        [renormalize]: 3.9002e-07
[validate]: 4.727e-05
Sums
    bootstrap                            :     0.000323s :  3.37%
    type_inference                       :     0.004548s : 47.54%
    event_method                         :     0.000019s :  0.20%
    auto_monad                           :     0.000068s :  0.71%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.38%
    rewriter_before_opt_a                :     0.000077s :  0.81%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000085s :  0.89%
    jit_opt_a.loop_unroll                :     0.000039s :  0.41%
    jit_opt_a.a_1                        :     0.000852s :  8.90%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.38%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.22%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.06%
    jit_opt_a.replace_old_param          :     0.000020s :  0.21%
    jit_opt_a.inline_without_move        :     0.000012s :  0.13%
    jit_opt_a.renormalize                :     0.001012s : 10.58%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.26%
    jit_opt_a.cse                        :     0.000045s :  0.47%
    jit_opt_a.replace_applicator         :     0.000020s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.16%
    rewriter_after_opt_a                 :     0.000036s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000629s :  6.58%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.16%
    cconv                                :     0.000027s :  0.28%
    loop_unroll                          :     0.000515s :  5.38%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.08%
    remove_dup_value                     :     0.000015s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000062s :  0.64%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.08%
    opt_after_jit_grad                   :     0.000590s :  6.16%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000047s :  0.49%


Time group info:
------[substitution.]   0.000317    23
 0.67% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000002s :      2: substitution.fold_const_symbol
 1.77% :     0.000006s :      3: substitution.graph_param_transform
87.42% :     0.000277s :      4: substitution.inline
 1.58% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.62% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.70% :     0.000005s :      2: substitution.replace_old_param
 4.73% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004474     2
78.87% :     0.003528s :      1: type_inference.infer
21.13% :     0.000945s :      1: type_inference.specialize
------[replace.]   0.000118     6
61.90% :     0.000073s :      4: replace.inline
38.10% :     0.000045s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000287     6
95.33% :     0.000273s :      4: match.inline
 4.67% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000144   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.57% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.91% :     0.000001s :     10: predicate.addn_check_dump
 2.15% :     0.000003s :     10: predicate.addn_zero_filter
 2.64% :     0.000004s :     10: predicate.arithmetic_simplify
 1.58% :     0.000002s :     10: predicate.cast_eliminate
 0.61% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.90% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.02% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.39% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.53% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.89% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.97% :     0.000001s :     10: predicate.environ_get_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.81% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 6.41% :     0.000009s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.46% :     0.000002s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.43% :     0.000002s :     15: predicate.load_eliminater
 1.72% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000006s :     34: predicate.loop_unroll_before_grad
 3.21% :     0.000005s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     10: predicate.merge_addn
 0.91% :     0.000001s :     10: predicate.minmaximum_grad
 1.85% :     0.000003s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 1.82% :     0.000003s :     15: predicate.partial_eliminate
 0.97% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.75% :     0.000003s :     10: predicate.reduce_eliminate
 1.14% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.40% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.51% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.19% :     0.000002s :     10: predicate.reshape_eliminate
 0.99% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     10: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.26% :     0.000002s :      6: predicate.special_op_eliminate
 0.86% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 1.73% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.39% :     0.000012s :     53: predicate.switch_simplify
 1.04% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000002s :     10: predicate.transpose_eliminate
 1.20% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.23% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 2.58% :     0.000004s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.34% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.30% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000002s :     10: predicate.value_based_eliminate
 0.26% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000698    12
44.37% :     0.000310s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.49% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.14% :     0.000329s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016369    72
 0.40% :     0.000066s :      1: add_recomputation
 0.44% :     0.000072s :      1: auto_monad
 0.12% :     0.000020s :      1: auto_monad_reorder
 2.10% :     0.000343s :      1: bootstrap
 0.18% :     0.000030s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
40.39% :     0.006611s :      1: jit_opt_a
 0.99% :     0.000162s :      1: jit_opt_after_cconv
 0.36% :     0.000059s :      1: jit_opt_b
 3.21% :     0.000525s :      1: loop_unroll
 3.91% :     0.000641s :      1: mutable_eliminate
 6.77% :     0.001109s :     26: opt.transform.jit_opt_a
 0.34% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000029s :      4: opt.transform.jit_opt_b
 0.09% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.66% :     0.000599s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000039s :      1: py_interpret_to_execute
 0.11% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 3.32% :     0.000543s :      1: renormalize.infer
 2.79% :     0.000457s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000040s :      1: rewriter_after_opt_a
 0.49% :     0.000080s :      1: rewriter_before_opt_a
 0.50% :     0.000083s :      1: symbol_engine_optimizer
27.90% :     0.004566s :      1: type_inference


TotalTime = 0.0329236, [33]
[bootstrap]: 0.00023634
[type_inference]: 0.0109477
[event_method]: 4.111e-05
[auto_monad]: 8.055e-05
[graph_reusing]: 5.77001e-06
[pre_auto_parallel]: 2.18998e-06
[py_interpret_to_execute]: 2.478e-05
[rewriter_before_opt_a]: 8.575e-05
[expand_dump_flag]: 2.17999e-06
[jit_opt_a]: 0.012519, [3]
    [Cycle 1]: 0.00631633, [27]
        [switch_simplify]: 6.918e-05
        [loop_unroll]: 3.865e-05
        [a_1]: 0.0008498
        [with_stream_mark]: 2.728e-05
        [recompute_prepare]: 1.818e-05
        [updatestate_depend_eliminate]: 8.08001e-06
        [updatestate_assign_eliminate]: 5.61e-06
        [updatestate_loads_eliminate]: 5.44998e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 1.278e-05
        [updatestate_useless_node_eliminater]: 1.19e-05
        [accelerated_algorithm]: 1.236e-05
        [meta_shard_fg_expand]: 3.527e-05
        [get_grad_eliminate_]: 1.291e-05
        [merge_forward]: 8.26002e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.538e-05
        [j_node_and_user_rematch]: 2.313e-05
        [meta_fg_expand]: 0.00174643
        [replace_old_param]: 5.979e-05
        [inline_without_move]: 5.52e-05
        [renormalize]: 0.00277767
        [add_forward_monad_depend]: 9.76003e-06
        [auto_monad_grad]: 5.97999e-06
        [auto_monad_eliminator]: 5.128e-05
        [cse]: 0.00014748
        [replace_applicator]: 7.03e-05
    [Cycle 2]: 0.00207476, [27]
        [switch_simplify]: 3.576e-05
        [loop_unroll]: 3.462e-05
        [a_1]: 0.00107083
        [with_stream_mark]: 1.516e-05
        [recompute_prepare]: 8.71002e-06
        [updatestate_depend_eliminate]: 3.98999e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 3.13e-06
        [parameter_eliminate]: 1.42999e-06
        [specialize_transform]: 6.77002e-06
        [updatestate_useless_node_eliminater]: 6.46e-06
        [accelerated_algorithm]: 6.38998e-06
        [meta_shard_fg_expand]: 2.41998e-06
        [get_grad_eliminate_]: 6.44999e-06
        [merge_forward]: 3.78999e-06
        [cell_reuse_recompute_pass]: 1.42e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.569e-05
        [j_node_and_user_rematch]: 1.001e-05
        [meta_fg_expand]: 4.014e-05
        [replace_old_param]: 1.073e-05
        [inline_without_move]: 6.33998e-06
        [renormalize]: 0.00051921
        [add_forward_monad_depend]: 4.53001e-06
        [auto_monad_grad]: 1.45999e-06
        [auto_monad_eliminator]: 1.173e-05
        [cse]: 2.071e-05
        [replace_applicator]: 1.282e-05
    [Cycle 3]: 0.0003892, [27]
        [switch_simplify]: 6.74001e-06
        [loop_unroll]: 6.22001e-06
        [a_1]: 0.000141
        [with_stream_mark]: 1.041e-05
        [recompute_prepare]: 5.86e-06
        [updatestate_depend_eliminate]: 3.48999e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 6.35997e-06
        [updatestate_useless_node_eliminater]: 5.72001e-06
        [accelerated_algorithm]: 5.74e-06
        [meta_shard_fg_expand]: 1.54998e-06
        [get_grad_eliminate_]: 5.64e-06
        [merge_forward]: 2.91e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.462e-05
        [j_node_and_user_rematch]: 9.65002e-06
        [meta_fg_expand]: 2.02999e-06
        [replace_old_param]: 7.95e-06
        [inline_without_move]: 5.82001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 6.99976e-07
        [auto_monad_eliminator]: 6.96001e-06
        [cse]: 1.426e-05
        [replace_applicator]: 6.10002e-06
[py_interpret_to_execute_after_opt_a]: 1.22e-05
[rewriter_after_opt_a]: 3.685e-05
[convert_after_rewriter]: 7.57002e-06
[order_py_execute_after_rewriter]: 5.17e-06
[mutable_eliminate]: 0.00067956
[jit_opt_b]: 5.698e-05, [1]
    [Cycle 1]: 4.937e-05, [2]
        [frontend_op_eliminate]: 1.997e-05
        [inline_after_opt_a]: 1.772e-05
[cconv]: 2.627e-05
[loop_unroll]: 0.00044009
[jit_opt_after_cconv]: 0.00016279, [1]
    [Cycle 1]: 0.00015552, [11]
        [c_1]: 2.629e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 5.82999e-06
        [updatestate_assign_eliminate]: 3.31001e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [cse]: 2.25e-05
        [call_graph_tuple_transform]: 2.063e-05
        [tuple_list_get_item_eliminator]: 6.52001e-06
        [none_parameter_eliminate]: 1.59998e-06
        [renormalize]: 7.50006e-07
        [switch_simplify]: 6.16e-06
[remove_dup_value]: 1.593e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 5.49e-06
[add_recomputation]: 4.968e-05
[cse_after_recomputation]: 2.682e-05, [1]
    [Cycle 1]: 2.036e-05, [1]
        [cse]: 1.412e-05
[auto_monad_reorder]: 1.755e-05
[get_jit_bprop_graph]: 1.81998e-06
[rewriter_after_jit_bprop_graph]: 6.25002e-06
[opt_after_jit_grad]: 0.00053755
[symbol_engine_optimizer]: 8.477e-05, [1]
    [Cycle 1]: 7.77e-05, [6]
        [build]: 4.77998e-06
        [elim_shapecalc]: 9.04e-06
        [elim_not_effective]: 1.778e-05
        [opt_reshape]: 6.92002e-06
        [fold_const_symbol]: 1.075e-05
        [renormalize]: 5.69999e-07
[validate]: 4.002e-05
[backend_pass]: 1.62001e-06
[task_emit]: 0.00652355
[execute]: 6.29001e-06
Sums
    bootstrap                            :     0.000236s :  0.84%
    type_inference                       :     0.010948s : 38.72%
    event_method                         :     0.000041s :  0.15%
    auto_monad                           :     0.000081s :  0.28%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000025s :  0.09%
    rewriter_before_opt_a                :     0.000086s :  0.30%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000112s :  0.39%
    jit_opt_a.loop_unroll                :     0.000079s :  0.28%
    jit_opt_a.a_1                        :     0.002062s :  7.29%
    jit_opt_a.with_stream_mark           :     0.000053s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000039s :  0.14%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.09%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000056s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000043s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001789s :  6.33%
    jit_opt_a.replace_old_param          :     0.000078s :  0.28%
    jit_opt_a.inline_without_move        :     0.000067s :  0.24%
    jit_opt_a.renormalize                :     0.003297s : 11.66%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000070s :  0.25%
    jit_opt_a.cse                        :     0.000182s :  0.65%
    jit_opt_a.replace_applicator         :     0.000089s :  0.32%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000037s :  0.13%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000680s :  2.40%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000026s :  0.09%
    loop_unroll                          :     0.000440s :  1.56%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000050s :  0.18%
    cse_after_recomputation.cse          :     0.000014s :  0.05%
    auto_monad_reorder                   :     0.000018s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000538s :  1.90%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.14%
    backend_pass                         :     0.000002s :  0.01%
    task_emit                            :     0.006524s : 23.07%
    execute                              :     0.000006s :  0.02%


Time group info:
------[substitution.]   0.000498   110
 0.46% :     0.000002s :      3: substitution.elim_not_effective
 0.32% :     0.000002s :      3: substitution.fold_const_symbol
 1.16% :     0.000006s :      4: substitution.graph_param_transform
63.67% :     0.000317s :     12: substitution.inline
 3.78% :     0.000019s :      2: substitution.inline_without_move
 1.77% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.27% :     0.000011s :      7: substitution.minmaximum_grad
 1.41% :     0.000007s :      5: substitution.partial_eliminate
 2.22% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.77% :     0.000029s :     10: substitution.replace_applicator
 1.66% :     0.000008s :      7: substitution.replace_old_param
 0.51% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.61% :     0.000023s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.01% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 7.38% :     0.000037s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010879     2
90.77% :     0.009874s :      1: type_inference.infer
 9.23% :     0.001005s :      1: type_inference.specialize
------[replace.]   0.000157    19
66.66% :     0.000105s :     12: replace.inline
33.34% :     0.000052s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000327    19
94.86% :     0.000310s :     12: match.inline
 5.14% :     0.000017s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000373  2568
 1.64% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.52% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.41% :     0.000005s :     43: predicate.addn_check_dump
 1.80% :     0.000007s :     43: predicate.addn_zero_filter
 2.01% :     0.000007s :     43: predicate.arithmetic_simplify
 1.60% :     0.000006s :     43: predicate.cast_eliminate
 0.18% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.40% :     0.000005s :     43: predicate.compare_switch_simplify
 1.72% :     0.000006s :     43: predicate.depend_value_elim
 1.45% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.56% :     0.000006s :     43: predicate.dict_get_item_eliminator
 1.50% :     0.000006s :     43: predicate.dict_set_item_eliminator
 0.32% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.21% :     0.000001s :      4: predicate.elim_not_effective
 0.20% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000005s :     43: predicate.environ_add_const_eliminate
 1.44% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.49% :     0.000006s :     43: predicate.environ_get_depend_swap
 1.54% :     0.000006s :     43: predicate.environ_get_eliminate
 1.44% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.90% :     0.000003s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.33% :     0.000016s :     70: predicate.inline
 1.82% :     0.000007s :     40: predicate.inline_without_move
 0.35% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.93% :     0.000003s :     19: predicate.less_batch_normalization
 1.77% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.88% :     0.000007s :     54: predicate.load_eliminater
 0.47% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.19% :     0.000012s :     85: predicate.loop_unroll_before_grad
 1.77% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.40% :     0.000005s :     43: predicate.merge_addn
 1.43% :     0.000005s :     43: predicate.minmaximum_grad
 0.58% :     0.000002s :      4: predicate.mutable_eliminate
 0.20% :     0.000001s :      4: predicate.opt_reshape
 2.31% :     0.000009s :     54: predicate.partial_eliminate
 1.60% :     0.000006s :     43: predicate.print_const_string_wrapper
 2.00% :     0.000007s :     43: predicate.reduce_eliminate
 1.83% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.46% :     0.000009s :     96: predicate.replace_applicator
 0.91% :     0.000003s :     40: predicate.replace_old_param
 0.11% :     0.000000s :      4: predicate.reset_defer_inline
 1.79% :     0.000007s :     43: predicate.reshape_eliminate
 1.55% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.28% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.71% :     0.000006s :     43: predicate.same_eliminate
 0.47% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.50% :     0.000002s :      8: predicate.special_op_eliminate
 0.84% :     0.000003s :     19: predicate.specialize_transform
 1.66% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.55% :     0.000006s :     43: predicate.stack_unstack_eliminate
 0.16% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.83% :     0.000011s :     62: predicate.switch_defer_inline
 2.47% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.30% :     0.000023s :    151: predicate.switch_simplify
 1.50% :     0.000006s :     43: predicate.tile_eliminate
 1.50% :     0.000006s :     43: predicate.transpose_eliminate
 1.85% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.67% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.01% :     0.000011s :     58: predicate.tuple_list_get_item_eliminator
 1.87% :     0.000007s :     43: predicate.tuple_list_set_item_eliminator
 1.81% :     0.000007s :     50: predicate.tuple_to_list_eliminator_
 1.89% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000011s :     73: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000007s :     43: predicate.value_based_eliminate
 0.14% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.27% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001537    29
61.27% :     0.000941s :     13: func_graph_cloner_run.FuncGraphClonerGraph
38.73% :     0.000595s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.039025    91
 0.14% :     0.000053s :      1: add_recomputation
 0.22% :     0.000085s :      1: auto_monad
 0.05% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.65% :     0.000254s :      1: bootstrap
 0.07% :     0.000029s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000029s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.12% :     0.000046s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.01% :     0.000004s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
32.09% :     0.012522s :      1: jit_opt_a
 0.43% :     0.000166s :      1: jit_opt_after_cconv
 0.15% :     0.000059s :      1: jit_opt_b
 1.15% :     0.000448s :      1: loop_unroll
 1.76% :     0.000688s :      1: mutable_eliminate
 6.85% :     0.002673s :     39: opt.transform.jit_opt_a
 0.14% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000031s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000041s :      4: opt.transform.symbol_engine_opt
 1.40% :     0.000547s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000027s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 4.49% :     0.001751s :      2: renormalize.infer
 3.92% :     0.001529s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000040s :      1: rewriter_after_opt_a
 0.23% :     0.000088s :      1: rewriter_before_opt_a
 0.22% :     0.000087s :      1: symbol_engine_optimizer
16.75% :     0.006537s :      1: task_emit
28.10% :     0.010965s :      1: type_inference
 0.22% :     0.000087s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.639.795 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc414c0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.640.485 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0140921, [30]
[bootstrap]: 0.00029757
[type_inference]: 0.00466762
[event_method]: 1.758e-05
[auto_monad]: 7.139e-05
[graph_reusing]: 6.58998e-06
[pre_auto_parallel]: 2.74001e-06
[py_interpret_to_execute]: 4.371e-05
[rewriter_before_opt_a]: 8.106e-05
[expand_dump_flag]: 3.90998e-06
[jit_opt_a]: 0.0061855, [2]
    [Cycle 1]: 0.00211602, [27]
        [switch_simplify]: 6.465e-05
        [loop_unroll]: 3.184e-05
        [a_1]: 0.00061841
        [with_stream_mark]: 2.184e-05
        [recompute_prepare]: 8.14002e-06
        [updatestate_depend_eliminate]: 4.38999e-06
        [updatestate_assign_eliminate]: 3.07002e-06
        [updatestate_loads_eliminate]: 2.55002e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 6.28e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.71998e-06
        [meta_shard_fg_expand]: 2.02001e-06
        [get_grad_eliminate_]: 5.16998e-06
        [merge_forward]: 3.73001e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.442e-05
        [j_node_and_user_rematch]: 9.97999e-06
        [meta_fg_expand]: 2.39999e-06
        [replace_old_param]: 9.39e-06
        [inline_without_move]: 5.22e-06
        [renormalize]: 0.00102022
        [add_forward_monad_depend]: 7.25e-06
        [auto_monad_grad]: 2.46998e-06
        [auto_monad_eliminator]: 1.801e-05
        [cse]: 3.133e-05
        [replace_applicator]: 1.712e-05
    [Cycle 2]: 0.00040255, [27]
        [switch_simplify]: 8.52e-06
        [loop_unroll]: 6.17999e-06
        [a_1]: 0.00010244
        [with_stream_mark]: 1.316e-05
        [recompute_prepare]: 6.59999e-06
        [updatestate_depend_eliminate]: 3.73001e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 1.14003e-06
        [specialize_transform]: 6.24999e-06
        [updatestate_useless_node_eliminater]: 5.88002e-06
        [accelerated_algorithm]: 6.42001e-06
        [meta_shard_fg_expand]: 2.57001e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 3.33e-06
        [cell_reuse_recompute_pass]: 1.52001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.65e-05
        [j_node_and_user_rematch]: 8.72e-06
        [meta_fg_expand]: 2.63e-06
        [replace_old_param]: 8.72e-06
        [inline_without_move]: 5.87999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.94999e-06
        [auto_monad_grad]: 1.14e-06
        [auto_monad_eliminator]: 8.30999e-06
        [cse]: 1.535e-05
        [replace_applicator]: 6.33e-06
[py_interpret_to_execute_after_opt_a]: 1.46e-05
[rewriter_after_opt_a]: 9.578e-05
[convert_after_rewriter]: 8.1e-06
[order_py_execute_after_rewriter]: 5.17e-06
[mutable_eliminate]: 0.000795
[jit_opt_b]: 6.137e-05, [1]
    [Cycle 1]: 5.076e-05, [2]
        [frontend_op_eliminate]: 2.022e-05
        [inline_after_opt_a]: 1.775e-05
[cconv]: 3.019e-05
[loop_unroll]: 0.00054029
[jit_opt_after_cconv]: 0.00017438, [1]
    [Cycle 1]: 0.00016564, [11]
        [c_1]: 2.354e-05
        [parameter_eliminate]: 3.41001e-06
        [updatestate_depend_eliminate]: 7.16001e-06
        [updatestate_assign_eliminate]: 2.55002e-06
        [updatestate_loads_eliminate]: 2.79999e-06
        [cse]: 2.594e-05
        [call_graph_tuple_transform]: 2.744e-05
        [tuple_list_get_item_eliminator]: 6.93e-06
        [none_parameter_eliminate]: 1.65001e-06
        [renormalize]: 7.60017e-07
        [switch_simplify]: 5.72999e-06
[remove_dup_value]: 1.655e-05
[partial_unused_args_eliminate]: 2.92002e-06
[environ_conv]: 5.87001e-06
[add_recomputation]: 5.376e-05
[cse_after_recomputation]: 2.601e-05, [1]
    [Cycle 1]: 1.948e-05, [1]
        [cse]: 1.109e-05
[auto_monad_reorder]: 1.76e-05
[get_jit_bprop_graph]: 1.92001e-06
[rewriter_after_jit_bprop_graph]: 6.98e-06
[opt_after_jit_grad]: 0.0005448
[symbol_engine_optimizer]: 7.871e-05, [1]
    [Cycle 1]: 7.106e-05, [6]
        [build]: 4.32e-06
        [elim_shapecalc]: 9.31e-06
        [elim_not_effective]: 1.413e-05
        [opt_reshape]: 5.79e-06
        [fold_const_symbol]: 8.99e-06
        [renormalize]: 4.90021e-07
[validate]: 3.688e-05
Sums
    bootstrap                            :     0.000298s :  3.05%
    type_inference                       :     0.004668s : 47.90%
    event_method                         :     0.000018s :  0.18%
    auto_monad                           :     0.000071s :  0.73%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000044s :  0.45%
    rewriter_before_opt_a                :     0.000081s :  0.83%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000073s :  0.75%
    jit_opt_a.loop_unroll                :     0.000038s :  0.39%
    jit_opt_a.a_1                        :     0.000721s :  7.40%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.36%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000018s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.001020s : 10.47%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.27%
    jit_opt_a.cse                        :     0.000047s :  0.48%
    jit_opt_a.replace_applicator         :     0.000023s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.15%
    rewriter_after_opt_a                 :     0.000096s :  0.98%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000795s :  8.16%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.18%
    cconv                                :     0.000030s :  0.31%
    loop_unroll                          :     0.000540s :  5.54%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000026s :  0.27%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.28%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000017s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000054s :  0.55%
    cse_after_recomputation.cse          :     0.000011s :  0.11%
    auto_monad_reorder                   :     0.000018s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000545s :  5.59%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000037s :  0.38%


Time group info:
------[substitution.]   0.000270    23
 0.96% :     0.000003s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 3.04% :     0.000008s :      3: substitution.graph_param_transform
85.78% :     0.000231s :      4: substitution.inline
 1.49% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.91% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.98% :     0.000005s :      2: substitution.replace_old_param
 4.36% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004592     2
77.97% :     0.003580s :      1: type_inference.infer
22.03% :     0.001012s :      1: type_inference.specialize
------[replace.]   0.000080     6
67.98% :     0.000055s :      4: replace.inline
32.02% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000239     6
95.53% :     0.000228s :      4: match.inline
 4.47% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   691
 1.56% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.53% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 2.40% :     0.000003s :     10: predicate.arithmetic_simplify
 1.05% :     0.000001s :     10: predicate.cast_eliminate
 0.64% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.93% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 0.97% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.01% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000001s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.61% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000002s :     10: predicate.environ_get_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.81% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.67% :     0.000008s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.65% :     0.000002s :      6: predicate.less_batch_normalization
 1.64% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.45% :     0.000002s :     15: predicate.load_eliminater
 1.80% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.56% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.02% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.89% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 2.33% :     0.000003s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.96% :     0.000003s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.58% :     0.000002s :     10: predicate.reshape_eliminate
 0.97% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.29% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.35% :     0.000002s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.21% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.06% :     0.000003s :     16: predicate.switch_defer_inline
 1.77% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.22% :     0.000011s :     53: predicate.switch_simplify
 1.37% :     0.000002s :     10: predicate.tile_eliminate
 1.04% :     0.000001s :     10: predicate.transpose_eliminate
 1.51% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.57% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.94% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 2.02% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.41% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.64% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000772    12
43.87% :     0.000338s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.85% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.28% :     0.000380s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016150    72
 0.37% :     0.000059s :      1: add_recomputation
 0.46% :     0.000075s :      1: auto_monad
 0.13% :     0.000021s :      1: auto_monad_reorder
 1.92% :     0.000310s :      1: bootstrap
 0.20% :     0.000033s :      1: cconv
 0.07% :     0.000011s :      1: convert_after_rewriter
 0.17% :     0.000028s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.14% :     0.000022s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
38.33% :     0.006191s :      1: jit_opt_a
 1.10% :     0.000178s :      1: jit_opt_after_cconv
 0.40% :     0.000064s :      1: jit_opt_b
 3.42% :     0.000553s :      1: loop_unroll
 5.00% :     0.000807s :      1: mutable_eliminate
 5.95% :     0.000961s :     26: opt.transform.jit_opt_a
 0.37% :     0.000060s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000029s :      4: opt.transform.jit_opt_b
 0.10% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.43% :     0.000554s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000046s :      1: py_interpret_to_execute
 0.11% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000019s :      1: remove_dup_value
 3.13% :     0.000505s :      1: renormalize.infer
 3.13% :     0.000505s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.62% :     0.000100s :      1: rewriter_after_opt_a
 0.53% :     0.000085s :      1: rewriter_before_opt_a
 0.50% :     0.000081s :      1: symbol_engine_optimizer
29.01% :     0.004685s :      1: type_inference


TotalTime = 0.0258285, [33]
[bootstrap]: 0.00031788
[type_inference]: 0.00575763
[event_method]: 2.168e-05
[auto_monad]: 7.064e-05
[graph_reusing]: 6.22001e-06
[pre_auto_parallel]: 2.69999e-06
[py_interpret_to_execute]: 3.748e-05
[rewriter_before_opt_a]: 7.443e-05
[expand_dump_flag]: 2.99001e-06
[jit_opt_a]: 0.00804374, [2]
    [Cycle 1]: 0.00200359, [27]
        [switch_simplify]: 6.473e-05
        [loop_unroll]: 2.861e-05
        [a_1]: 0.00062249
        [with_stream_mark]: 3.116e-05
        [recompute_prepare]: 9.91e-06
        [updatestate_depend_eliminate]: 4.13999e-06
        [updatestate_assign_eliminate]: 3.33998e-06
        [updatestate_loads_eliminate]: 2.93998e-06
        [parameter_eliminate]: 2.66e-06
        [specialize_transform]: 7.03998e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 5.53002e-06
        [meta_shard_fg_expand]: 2.60002e-06
        [get_grad_eliminate_]: 5.04e-06
        [merge_forward]: 4.45999e-06
        [cell_reuse_recompute_pass]: 1.60999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.55e-05
        [j_node_and_user_rematch]: 1.089e-05
        [meta_fg_expand]: 2.62001e-06
        [replace_old_param]: 9.89001e-06
        [inline_without_move]: 6.19999e-06
        [renormalize]: 0.00086591
        [add_forward_monad_depend]: 8.00999e-06
        [auto_monad_grad]: 2.98e-06
        [auto_monad_eliminator]: 1.879e-05
        [cse]: 3.16e-05
        [replace_applicator]: 1.973e-05
    [Cycle 2]: 0.00033682, [27]
        [switch_simplify]: 5.66998e-06
        [loop_unroll]: 4.57e-06
        [a_1]: 9.397e-05
        [with_stream_mark]: 1.345e-05
        [recompute_prepare]: 6.01e-06
        [updatestate_depend_eliminate]: 2.83e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 1.50999e-06
        [specialize_transform]: 4.62998e-06
        [updatestate_useless_node_eliminater]: 4.63001e-06
        [accelerated_algorithm]: 4.75999e-06
        [meta_shard_fg_expand]: 1.24e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 2.78e-06
        [cell_reuse_recompute_pass]: 2.88998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.542e-05
        [j_node_and_user_rematch]: 8.79998e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 8.13001e-06
        [inline_without_move]: 4.60999e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 1.15999e-06
        [auto_monad_eliminator]: 5.46e-06
        [cse]: 1.036e-05
        [replace_applicator]: 5.56e-06
[py_interpret_to_execute_after_opt_a]: 1.531e-05
[rewriter_after_opt_a]: 4.243e-05
[convert_after_rewriter]: 8.04002e-06
[order_py_execute_after_rewriter]: 5.55001e-06
[mutable_eliminate]: 0.00081757
[jit_opt_b]: 5.528e-05, [1]
    [Cycle 1]: 4.657e-05, [2]
        [frontend_op_eliminate]: 1.718e-05
        [inline_after_opt_a]: 1.657e-05
[cconv]: 3.343e-05
[loop_unroll]: 0.00049735
[jit_opt_after_cconv]: 0.00017, [1]
    [Cycle 1]: 0.00016236, [11]
        [c_1]: 2.049e-05
        [parameter_eliminate]: 5.40999e-06
        [updatestate_depend_eliminate]: 1.14e-05
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [cse]: 3.093e-05
        [call_graph_tuple_transform]: 2.408e-05
        [tuple_list_get_item_eliminator]: 5.47001e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 9.10019e-07
        [switch_simplify]: 5.46e-06
[remove_dup_value]: 1.576e-05
[partial_unused_args_eliminate]: 2.59001e-06
[environ_conv]: 5.72001e-06
[add_recomputation]: 5.938e-05
[cse_after_recomputation]: 2.149e-05, [1]
    [Cycle 1]: 1.526e-05, [1]
        [cse]: 9.36e-06
[auto_monad_reorder]: 1.602e-05
[get_jit_bprop_graph]: 2.26998e-06
[rewriter_after_jit_bprop_graph]: 9.24998e-06
[opt_after_jit_grad]: 0.00054837
[symbol_engine_optimizer]: 8.111e-05, [1]
    [Cycle 1]: 7.423e-05, [6]
        [build]: 6.36e-06
        [elim_shapecalc]: 8.12e-06
        [elim_not_effective]: 1.606e-05
        [opt_reshape]: 5.35001e-06
        [fold_const_symbol]: 8.69e-06
        [renormalize]: 5.89993e-07
[validate]: 4.874e-05
[backend_pass]: 1.18001e-06
[task_emit]: 0.00877189
[execute]: 1.03e-05
Sums
    bootstrap                            :     0.000318s :  1.64%
    type_inference                       :     0.005758s : 29.65%
    event_method                         :     0.000022s :  0.11%
    auto_monad                           :     0.000071s :  0.36%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000037s :  0.19%
    rewriter_before_opt_a                :     0.000074s :  0.38%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000070s :  0.36%
    jit_opt_a.loop_unroll                :     0.000033s :  0.17%
    jit_opt_a.a_1                        :     0.000716s :  3.69%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000866s :  4.46%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.12%
    jit_opt_a.cse                        :     0.000042s :  0.22%
    jit_opt_a.replace_applicator         :     0.000025s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.08%
    rewriter_after_opt_a                 :     0.000042s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000818s :  4.21%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000033s :  0.17%
    loop_unroll                          :     0.000497s :  2.56%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000031s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000059s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000548s :  2.82%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000049s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008772s : 45.18%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000281    23
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 2.22% :     0.000006s :      3: substitution.graph_param_transform
85.88% :     0.000242s :      4: substitution.inline
 1.83% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.90% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.30% :     0.000006s :      2: substitution.replace_old_param
 4.64% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005672     2
84.16% :     0.004774s :      1: type_inference.infer
15.84% :     0.000898s :      1: type_inference.specialize
------[replace.]   0.000073     6
67.99% :     0.000050s :      4: replace.inline
32.01% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000250     6
95.33% :     0.000239s :      4: match.inline
 4.67% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.50% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.54% :     0.000002s :     10: predicate.addn_zero_filter
 1.95% :     0.000002s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.09% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.25% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.47% :     0.000001s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 6.37% :     0.000008s :     22: predicate.inline
 1.05% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.35% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 1.80% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.53% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.27% :     0.000002s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 3.29% :     0.000004s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 2.02% :     0.000002s :     10: predicate.reshape_eliminate
 1.09% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.03% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.11% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.78% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.18% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.08% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.10% :     0.000003s :     16: predicate.switch_defer_inline
 2.24% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.40% :     0.000010s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.16% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.11% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.35% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000763    12
56.80% :     0.000434s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.71% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
35.49% :     0.000271s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027764    76
 0.23% :     0.000063s :      1: add_recomputation
 0.27% :     0.000074s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.23% :     0.000342s :      1: bootstrap
 0.13% :     0.000036s :      1: cconv
 0.04% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000027s :      1: event_method
 0.06% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
28.98% :     0.008047s :      1: jit_opt_a
 0.62% :     0.000173s :      1: jit_opt_after_cconv
 0.21% :     0.000058s :      1: jit_opt_b
 1.84% :     0.000509s :      1: loop_unroll
 2.99% :     0.000831s :      1: mutable_eliminate
 3.42% :     0.000949s :     26: opt.transform.jit_opt_a
 0.19% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.02% :     0.000561s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000040s :      1: py_interpret_to_execute
 0.06% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 1.81% :     0.000502s :      1: renormalize.infer
 1.28% :     0.000355s :      1: renormalize.specialize
 0.04% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000046s :      1: rewriter_after_opt_a
 0.28% :     0.000077s :      1: rewriter_before_opt_a
 0.30% :     0.000084s :      1: symbol_engine_optimizer
31.67% :     0.008793s :      1: task_emit
20.81% :     0.005778s :      1: type_inference
 0.28% :     0.000078s :      1: validate


TotalTime = 0.021672, [33]
[bootstrap]: 0.00032611
[type_inference]: 0.00501818
[event_method]: 1.789e-05
[auto_monad]: 6.265e-05
[graph_reusing]: 6.79001e-06
[pre_auto_parallel]: 2.56e-06
[py_interpret_to_execute]: 3.809e-05
[rewriter_before_opt_a]: 7.316e-05
[expand_dump_flag]: 3.26001e-06
[jit_opt_a]: 0.00669401, [2]
    [Cycle 1]: 0.00210214, [27]
        [switch_simplify]: 6.38e-05
        [loop_unroll]: 2.909e-05
        [a_1]: 0.00060718
        [with_stream_mark]: 2.258e-05
        [recompute_prepare]: 7.53e-06
        [updatestate_depend_eliminate]: 3.91999e-06
        [updatestate_assign_eliminate]: 3.54002e-06
        [updatestate_loads_eliminate]: 2.62001e-06
        [parameter_eliminate]: 2.02999e-06
        [specialize_transform]: 5.93998e-06
        [updatestate_useless_node_eliminater]: 5.62001e-06
        [accelerated_algorithm]: 5.79999e-06
        [meta_shard_fg_expand]: 1.99999e-06
        [get_grad_eliminate_]: 5.61e-06
        [merge_forward]: 3.86999e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.401e-05
        [j_node_and_user_rematch]: 8.99998e-06
        [meta_fg_expand]: 2.54999e-06
        [replace_old_param]: 2.661e-05
        [inline_without_move]: 6.49001e-06
        [renormalize]: 0.00100021
        [add_forward_monad_depend]: 6.79001e-06
        [auto_monad_grad]: 2.44001e-06
        [auto_monad_eliminator]: 1.739e-05
        [cse]: 2.755e-05
        [replace_applicator]: 1.702e-05
    [Cycle 2]: 0.00036509, [27]
        [switch_simplify]: 6.01998e-06
        [loop_unroll]: 5.54e-06
        [a_1]: 9.523e-05
        [with_stream_mark]: 1.098e-05
        [recompute_prepare]: 5.71998e-06
        [updatestate_depend_eliminate]: 2.49999e-06
        [updatestate_assign_eliminate]: 2.30002e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 5.15999e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.39998e-06
        [meta_shard_fg_expand]: 1.30001e-06
        [get_grad_eliminate_]: 4.63001e-06
        [merge_forward]: 3.14999e-06
        [cell_reuse_recompute_pass]: 2.46998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.444e-05
        [j_node_and_user_rematch]: 7.83999e-06
        [meta_fg_expand]: 1.85001e-06
        [replace_old_param]: 7.78999e-06
        [inline_without_move]: 4.80999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.94e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 6.43998e-06
        [cse]: 1.143e-05
        [replace_applicator]: 6.17001e-06
[py_interpret_to_execute_after_opt_a]: 1.376e-05
[rewriter_after_opt_a]: 3.66e-05
[convert_after_rewriter]: 7.01999e-06
[order_py_execute_after_rewriter]: 4.68001e-06
[mutable_eliminate]: 0.00073614
[jit_opt_b]: 5.23e-05, [1]
    [Cycle 1]: 4.343e-05, [2]
        [frontend_op_eliminate]: 1.693e-05
        [inline_after_opt_a]: 1.467e-05
[cconv]: 2.911e-05
[loop_unroll]: 0.00044529
[jit_opt_after_cconv]: 0.00017769, [1]
    [Cycle 1]: 0.00017005, [11]
        [c_1]: 2.041e-05
        [parameter_eliminate]: 3.31001e-06
        [updatestate_depend_eliminate]: 5.59998e-06
        [updatestate_assign_eliminate]: 2.619e-05
        [updatestate_loads_eliminate]: 2.27001e-06
        [cse]: 2.686e-05
        [call_graph_tuple_transform]: 1.914e-05
        [tuple_list_get_item_eliminator]: 5.52001e-06
        [none_parameter_eliminate]: 1.89999e-06
        [renormalize]: 5.50004e-07
        [switch_simplify]: 5.51002e-06
[remove_dup_value]: 1.499e-05
[partial_unused_args_eliminate]: 2.39001e-06
[environ_conv]: 5.57001e-06
[add_recomputation]: 5.492e-05
[cse_after_recomputation]: 2.206e-05, [1]
    [Cycle 1]: 1.621e-05, [1]
        [cse]: 1.04e-05
[auto_monad_reorder]: 1.572e-05
[get_jit_bprop_graph]: 2.43e-06
[rewriter_after_jit_bprop_graph]: 6.78e-06
[opt_after_jit_grad]: 0.0004734
[symbol_engine_optimizer]: 7.363e-05, [1]
    [Cycle 1]: 6.702e-05, [6]
        [build]: 4.55001e-06
        [elim_shapecalc]: 7.55998e-06
        [elim_not_effective]: 1.209e-05
        [opt_reshape]: 5.64e-06
        [fold_const_symbol]: 9.31e-06
        [renormalize]: 8.60018e-07
[validate]: 4.05e-05
[backend_pass]: 1.10999e-06
[task_emit]: 0.00696825
[execute]: 9.03002e-06
Sums
    bootstrap                            :     0.000326s :  1.95%
    type_inference                       :     0.005018s : 29.98%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000063s :  0.37%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.23%
    rewriter_before_opt_a                :     0.000073s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000070s :  0.42%
    jit_opt_a.loop_unroll                :     0.000035s :  0.21%
    jit_opt_a.a_1                        :     0.000702s :  4.20%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000034s :  0.21%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.001000s :  5.98%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.14%
    jit_opt_a.cse                        :     0.000039s :  0.23%
    jit_opt_a.replace_applicator         :     0.000023s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000037s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000736s :  4.40%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000029s :  0.17%
    loop_unroll                          :     0.000445s :  2.66%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000026s :  0.16%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000055s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000473s :  2.83%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006968s : 41.63%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000268    23
 0.66% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000002s :      2: substitution.fold_const_symbol
 2.30% :     0.000006s :      3: substitution.graph_param_transform
86.85% :     0.000233s :      4: substitution.inline
 1.33% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.72% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.64% :     0.000004s :      2: substitution.replace_old_param
 4.85% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004937     2
81.07% :     0.004003s :      1: type_inference.infer
18.93% :     0.000935s :      1: type_inference.specialize
------[replace.]   0.000076     6
67.02% :     0.000051s :      4: replace.inline
32.98% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000243     6
95.05% :     0.000231s :      4: match.inline
 4.95% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.59% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.21% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 2.08% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.62% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.35% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.23% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.74% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.21% :     0.000001s :     10: predicate.minmaximum_grad
 1.83% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.24% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.32% :     0.000003s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.02% :     0.000010s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.10% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.32% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.58% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000651    12
48.48% :     0.000315s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.29% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.23% :     0.000301s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023713    76
 0.25% :     0.000058s :      1: add_recomputation
 0.28% :     0.000066s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.45% :     0.000344s :      1: bootstrap
 0.13% :     0.000032s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
28.25% :     0.006699s :      1: jit_opt_a
 0.76% :     0.000181s :      1: jit_opt_after_cconv
 0.23% :     0.000055s :      1: jit_opt_b
 1.92% :     0.000455s :      1: loop_unroll
 3.16% :     0.000749s :      1: mutable_eliminate
 3.99% :     0.000946s :     26: opt.transform.jit_opt_a
 0.20% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.03% :     0.000482s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000041s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.65% :     0.000628s :      1: renormalize.infer
 1.53% :     0.000362s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000039s :      1: rewriter_after_opt_a
 0.32% :     0.000076s :      1: rewriter_before_opt_a
 0.32% :     0.000076s :      1: symbol_engine_optimizer
29.47% :     0.006987s :      1: task_emit
21.26% :     0.005042s :      1: type_inference
 0.26% :     0.000062s :      1: validate


TotalTime = 0.012894, [30]
[bootstrap]: 0.00031679
[type_inference]: 0.00457256
[event_method]: 1.821e-05
[auto_monad]: 6.212e-05
[graph_reusing]: 6.07001e-06
[pre_auto_parallel]: 2.14999e-06
[py_interpret_to_execute]: 3.693e-05
[rewriter_before_opt_a]: 6.976e-05
[expand_dump_flag]: 3.13e-06
[jit_opt_a]: 0.00557476, [2]
    [Cycle 1]: 0.00180592, [27]
        [switch_simplify]: 5.278e-05
        [loop_unroll]: 3.018e-05
        [a_1]: 0.00056435
        [with_stream_mark]: 1.927e-05
        [recompute_prepare]: 7.51001e-06
        [updatestate_depend_eliminate]: 2.84999e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.37e-06
        [specialize_transform]: 6.49001e-06
        [updatestate_useless_node_eliminater]: 5.71e-06
        [accelerated_algorithm]: 6.37001e-06
        [meta_shard_fg_expand]: 2.02999e-06
        [get_grad_eliminate_]: 5.87001e-06
        [merge_forward]: 3.55e-06
        [cell_reuse_recompute_pass]: 1.31998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.222e-05
        [j_node_and_user_rematch]: 1.016e-05
        [meta_fg_expand]: 2.41e-06
        [replace_old_param]: 8.97e-06
        [inline_without_move]: 5.52001e-06
        [renormalize]: 0.00081609
        [add_forward_monad_depend]: 5.10001e-06
        [auto_monad_grad]: 2.28002e-06
        [auto_monad_eliminator]: 1.377e-05
        [cse]: 1.848e-05
        [replace_applicator]: 1.169e-05
    [Cycle 2]: 0.00032584, [27]
        [switch_simplify]: 5.66998e-06
        [loop_unroll]: 4.94e-06
        [a_1]: 9.193e-05
        [with_stream_mark]: 9.84001e-06
        [recompute_prepare]: 5.13002e-06
        [updatestate_depend_eliminate]: 2.63e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 9.60019e-07
        [specialize_transform]: 4.94e-06
        [updatestate_useless_node_eliminater]: 4.76002e-06
        [accelerated_algorithm]: 5.09e-06
        [meta_shard_fg_expand]: 1.32999e-06
        [get_grad_eliminate_]: 4.76002e-06
        [merge_forward]: 2.54001e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.225e-05
        [j_node_and_user_rematch]: 8.05e-06
        [meta_fg_expand]: 1.60001e-06
        [replace_old_param]: 7.50998e-06
        [inline_without_move]: 4.96002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 4.85999e-06
        [cse]: 1.019e-05
        [replace_applicator]: 5.12e-06
[py_interpret_to_execute_after_opt_a]: 1.096e-05
[rewriter_after_opt_a]: 3.202e-05
[convert_after_rewriter]: 9.02999e-06
[order_py_execute_after_rewriter]: 4.68001e-06
[mutable_eliminate]: 0.00058351
[jit_opt_b]: 5.218e-05, [1]
    [Cycle 1]: 4.563e-05, [2]
        [frontend_op_eliminate]: 1.82e-05
        [inline_after_opt_a]: 1.552e-05
[cconv]: 2.421e-05
[loop_unroll]: 0.0004223
[jit_opt_after_cconv]: 0.00014172, [1]
    [Cycle 1]: 0.00013541, [11]
        [c_1]: 2.132e-05
        [parameter_eliminate]: 2.35002e-06
        [updatestate_depend_eliminate]: 5.55001e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 2.23002e-06
        [cse]: 1.738e-05
        [call_graph_tuple_transform]: 1.922e-05
        [tuple_list_get_item_eliminator]: 5.52001e-06
        [none_parameter_eliminate]: 1.19e-06
        [renormalize]: 3.9002e-07
        [switch_simplify]: 5.29e-06
[remove_dup_value]: 1.345e-05
[partial_unused_args_eliminate]: 1.90001e-06
[environ_conv]: 4.79998e-06
[add_recomputation]: 4.231e-05
[cse_after_recomputation]: 2.189e-05, [1]
    [Cycle 1]: 1.574e-05, [1]
        [cse]: 9.62001e-06
[auto_monad_reorder]: 1.299e-05
[get_jit_bprop_graph]: 1.94e-06
[rewriter_after_jit_bprop_graph]: 5.70001e-06
[opt_after_jit_grad]: 0.00046925
[symbol_engine_optimizer]: 0.00017058, [1]
    [Cycle 1]: 0.00016429, [6]
        [build]: 4.33001e-06
        [elim_shapecalc]: 7.41001e-06
        [elim_not_effective]: 1.28e-05
        [opt_reshape]: 8.882e-05
        [fold_const_symbol]: 1.315e-05
        [renormalize]: 4.50003e-07
[validate]: 3.857e-05
Sums
    bootstrap                            :     0.000317s :  3.58%
    type_inference                       :     0.004573s : 51.69%
    event_method                         :     0.000018s :  0.21%
    auto_monad                           :     0.000062s :  0.70%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.42%
    rewriter_before_opt_a                :     0.000070s :  0.79%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000058s :  0.66%
    jit_opt_a.loop_unroll                :     0.000035s :  0.40%
    jit_opt_a.a_1                        :     0.000656s :  7.42%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000816s :  9.23%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.21%
    jit_opt_a.cse                        :     0.000029s :  0.32%
    jit_opt_a.replace_applicator         :     0.000017s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000032s :  0.36%
    convert_after_rewriter               :     0.000009s :  0.10%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000584s :  6.60%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.18%
    cconv                                :     0.000024s :  0.27%
    loop_unroll                          :     0.000422s :  4.77%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000013s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000042s :  0.48%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000013s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000469s :  5.31%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000089s :  1.00%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.15%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000039s :  0.44%


Time group info:
------[substitution.]   0.000221    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.83% :     0.000002s :      2: substitution.fold_const_symbol
 2.10% :     0.000005s :      3: substitution.graph_param_transform
85.57% :     0.000189s :      4: substitution.inline
 1.69% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.85% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.46% :     0.000003s :      2: substitution.replace_old_param
 5.64% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004512     2
81.50% :     0.003677s :      1: type_inference.infer
18.50% :     0.000835s :      1: type_inference.specialize
------[replace.]   0.000073     6
66.13% :     0.000048s :      4: replace.inline
33.87% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000198     6
94.31% :     0.000187s :      4: match.inline
 5.69% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.47% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.43% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.43% :     0.000002s :     10: predicate.addn_zero_filter
 2.07% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.60% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.65% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.08% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.09% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.05% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.04% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.16% :     0.000001s :     10: predicate.minmaximum_grad
 1.56% :     0.000002s :      3: predicate.mutable_eliminate
 0.81% :     0.000001s :      3: predicate.opt_reshape
 2.07% :     0.000002s :     15: predicate.partial_eliminate
 1.26% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.86% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.12% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.31% :     0.000002s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.39% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.81% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.55% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.92% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000642    12
49.45% :     0.000317s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.70% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.85% :     0.000281s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014715    72
 0.31% :     0.000045s :      1: add_recomputation
 0.44% :     0.000065s :      1: auto_monad
 0.10% :     0.000015s :      1: auto_monad_reorder
 2.25% :     0.000331s :      1: bootstrap
 0.18% :     0.000027s :      1: cconv
 0.08% :     0.000011s :      1: convert_after_rewriter
 0.16% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000022s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
37.91% :     0.005578s :      1: jit_opt_a
 0.99% :     0.000145s :      1: jit_opt_after_cconv
 0.37% :     0.000055s :      1: jit_opt_b
 2.92% :     0.000430s :      1: loop_unroll
 4.03% :     0.000593s :      1: mutable_eliminate
 5.90% :     0.000868s :     26: opt.transform.jit_opt_a
 0.33% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000027s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.80% :     0.000118s :      4: opt.transform.symbol_engine_opt
 3.24% :     0.000477s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000040s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000016s :      1: remove_dup_value
 3.09% :     0.000455s :      1: renormalize.infer
 2.39% :     0.000352s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000035s :      1: rewriter_after_opt_a
 0.49% :     0.000073s :      1: rewriter_before_opt_a
 1.18% :     0.000174s :      1: symbol_engine_optimizer
31.15% :     0.004584s :      1: type_inference


TotalTime = 0.0349906, [33]
[bootstrap]: 0.00025063
[type_inference]: 0.0117742
[event_method]: 4.342e-05
[auto_monad]: 9.835e-05
[graph_reusing]: 7.76001e-06
[pre_auto_parallel]: 2.72001e-06
[py_interpret_to_execute]: 2.722e-05
[rewriter_before_opt_a]: 9.35e-05
[expand_dump_flag]: 3.38e-06
[jit_opt_a]: 0.0128578, [3]
    [Cycle 1]: 0.00620023, [27]
        [switch_simplify]: 6.853e-05
        [loop_unroll]: 3.993e-05
        [a_1]: 0.00095376
        [with_stream_mark]: 3.399e-05
        [recompute_prepare]: 1.896e-05
        [updatestate_depend_eliminate]: 8.29002e-06
        [updatestate_assign_eliminate]: 7.47002e-06
        [updatestate_loads_eliminate]: 6.35997e-06
        [parameter_eliminate]: 2.98003e-06
        [specialize_transform]: 1.308e-05
        [updatestate_useless_node_eliminater]: 1.184e-05
        [accelerated_algorithm]: 1.293e-05
        [meta_shard_fg_expand]: 3.56001e-06
        [get_grad_eliminate_]: 1.205e-05
        [merge_forward]: 7.97e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.733e-05
        [j_node_and_user_rematch]: 2.291e-05
        [meta_fg_expand]: 0.00165236
        [replace_old_param]: 6.039e-05
        [inline_without_move]: 5.745e-05
        [renormalize]: 0.00260045
        [add_forward_monad_depend]: 1.191e-05
        [auto_monad_grad]: 5.81e-06
        [auto_monad_eliminator]: 5.389e-05
        [cse]: 0.00016071
        [replace_applicator]: 7.778e-05
    [Cycle 2]: 0.00217136, [27]
        [switch_simplify]: 3.557e-05
        [loop_unroll]: 3.44e-05
        [a_1]: 0.00113195
        [with_stream_mark]: 1.672e-05
        [recompute_prepare]: 8.35001e-06
        [updatestate_depend_eliminate]: 5.22999e-06
        [updatestate_assign_eliminate]: 4.33001e-06
        [updatestate_loads_eliminate]: 3.67998e-06
        [parameter_eliminate]: 2.36e-06
        [specialize_transform]: 6.89001e-06
        [updatestate_useless_node_eliminater]: 6.16998e-06
        [accelerated_algorithm]: 6.56e-06
        [meta_shard_fg_expand]: 2.85002e-06
        [get_grad_eliminate_]: 6.53e-06
        [merge_forward]: 4.38001e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.602e-05
        [j_node_and_user_rematch]: 9.97001e-06
        [meta_fg_expand]: 4.331e-05
        [replace_old_param]: 9.97999e-06
        [inline_without_move]: 6.31e-06
        [renormalize]: 0.00059274
        [add_forward_monad_depend]: 6.33e-06
        [auto_monad_grad]: 1.28002e-06
        [auto_monad_eliminator]: 1.33e-05
        [cse]: 2.578e-05
        [replace_applicator]: 1.404e-05
    [Cycle 3]: 0.00042889, [27]
        [switch_simplify]: 7.4e-06
        [loop_unroll]: 6.16e-06
        [a_1]: 0.00011941
        [with_stream_mark]: 1.031e-05
        [recompute_prepare]: 6.06e-06
        [updatestate_depend_eliminate]: 3.73999e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [parameter_eliminate]: 1.44998e-06
        [specialize_transform]: 6.28998e-06
        [updatestate_useless_node_eliminater]: 5.79999e-06
        [accelerated_algorithm]: 5.97999e-06
        [meta_shard_fg_expand]: 1.47999e-06
        [get_grad_eliminate_]: 4.596e-05
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 2.02001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.78e-05
        [j_node_and_user_rematch]: 1.034e-05
        [meta_fg_expand]: 2.13002e-06
        [replace_old_param]: 9.18002e-06
        [inline_without_move]: 6.31e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.94e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 7.95e-06
        [cse]: 1.816e-05
        [replace_applicator]: 7.35e-06
[py_interpret_to_execute_after_opt_a]: 1.164e-05
[rewriter_after_opt_a]: 3.923e-05
[convert_after_rewriter]: 7.43999e-06
[order_py_execute_after_rewriter]: 5.19e-06
[mutable_eliminate]: 0.00073569
[jit_opt_b]: 5.577e-05, [1]
    [Cycle 1]: 4.772e-05, [2]
        [frontend_op_eliminate]: 1.908e-05
        [inline_after_opt_a]: 1.73e-05
[cconv]: 2.795e-05
[loop_unroll]: 0.00044651
[jit_opt_after_cconv]: 0.00016421, [1]
    [Cycle 1]: 0.00015762, [11]
        [c_1]: 2.561e-05
        [parameter_eliminate]: 2.71999e-06
        [updatestate_depend_eliminate]: 8.06001e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.66e-06
        [cse]: 2.806e-05
        [call_graph_tuple_transform]: 2.248e-05
        [tuple_list_get_item_eliminator]: 6.00002e-06
        [none_parameter_eliminate]: 1.87001e-06
        [renormalize]: 7.00005e-07
        [switch_simplify]: 6.30002e-06
[remove_dup_value]: 1.738e-05
[partial_unused_args_eliminate]: 2.72001e-06
[environ_conv]: 6.54001e-06
[add_recomputation]: 6.074e-05
[cse_after_recomputation]: 2.856e-05, [1]
    [Cycle 1]: 2.26e-05, [1]
        [cse]: 1.595e-05
[auto_monad_reorder]: 1.968e-05
[get_jit_bprop_graph]: 2.06e-06
[rewriter_after_jit_bprop_graph]: 6.86001e-06
[opt_after_jit_grad]: 0.00048401
[symbol_engine_optimizer]: 8.346e-05, [1]
    [Cycle 1]: 7.688e-05, [6]
        [build]: 5.29e-06
        [elim_shapecalc]: 9.02999e-06
        [elim_not_effective]: 1.571e-05
        [opt_reshape]: 7.26001e-06
        [fold_const_symbol]: 1.113e-05
        [renormalize]: 5.00004e-07
[validate]: 4.49e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.00734165
[execute]: 8.46002e-06
Sums
    bootstrap                            :     0.000251s :  0.83%
    type_inference                       :     0.011774s : 39.19%
    event_method                         :     0.000043s :  0.14%
    auto_monad                           :     0.000098s :  0.33%
    graph_reusing                        :     0.000008s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000027s :  0.09%
    rewriter_before_opt_a                :     0.000094s :  0.31%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000112s :  0.37%
    jit_opt_a.loop_unroll                :     0.000080s :  0.27%
    jit_opt_a.a_1                        :     0.002205s :  7.34%
    jit_opt_a.with_stream_mark           :     0.000061s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000014s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000013s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000007s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000025s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000065s :  0.21%
    jit_opt_a.merge_forward              :     0.000016s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000061s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000043s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.001698s :  5.65%
    jit_opt_a.replace_old_param          :     0.000080s :  0.26%
    jit_opt_a.inline_without_move        :     0.000070s :  0.23%
    jit_opt_a.renormalize                :     0.003193s : 10.63%
    jit_opt_a.add_forward_monad_depend   :     0.000020s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000075s :  0.25%
    jit_opt_a.cse                        :     0.000205s :  0.68%
    jit_opt_a.replace_applicator         :     0.000099s :  0.33%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000039s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000736s :  2.45%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.06%
    cconv                                :     0.000028s :  0.09%
    loop_unroll                          :     0.000447s :  1.49%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000017s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000061s :  0.20%
    cse_after_recomputation.cse          :     0.000016s :  0.05%
    auto_monad_reorder                   :     0.000020s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000484s :  1.61%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007342s : 24.44%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000590   110
 0.38% :     0.000002s :      3: substitution.elim_not_effective
 0.28% :     0.000002s :      3: substitution.fold_const_symbol
 1.05% :     0.000006s :      4: substitution.graph_param_transform
62.86% :     0.000371s :     12: substitution.inline
 3.45% :     0.000020s :      2: substitution.inline_without_move
 1.63% :     0.000010s :     14: substitution.j_node_and_user_rematch
 1.99% :     0.000012s :      7: substitution.minmaximum_grad
 1.30% :     0.000008s :      5: substitution.partial_eliminate
 1.91% :     0.000011s :     14: substitution.remove_not_recompute_node
 9.21% :     0.000054s :     10: substitution.replace_applicator
 1.50% :     0.000009s :      7: substitution.replace_old_param
 0.42% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.02% :     0.000024s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.60% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 7.40% :     0.000044s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011697     2
90.53% :     0.010589s :      1: type_inference.infer
 9.47% :     0.001108s :      1: type_inference.specialize
------[replace.]   0.000157    19
62.97% :     0.000099s :     12: replace.inline
37.03% :     0.000058s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000387    19
94.20% :     0.000364s :     12: match.inline
 5.80% :     0.000022s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000383  2568
 1.52% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.45% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.36% :     0.000005s :     43: predicate.addn_check_dump
 1.63% :     0.000006s :     43: predicate.addn_zero_filter
 2.21% :     0.000008s :     43: predicate.arithmetic_simplify
 1.70% :     0.000007s :     43: predicate.cast_eliminate
 0.15% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.35% :     0.000005s :     43: predicate.compare_switch_simplify
 1.69% :     0.000006s :     43: predicate.depend_value_elim
 1.41% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.50% :     0.000006s :     43: predicate.dict_get_item_eliminator
 1.47% :     0.000006s :     43: predicate.dict_set_item_eliminator
 0.33% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.17% :     0.000001s :      4: predicate.elim_not_effective
 0.21% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000006s :     43: predicate.environ_add_const_eliminate
 1.40% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.40% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.55% :     0.000006s :     43: predicate.environ_get_eliminate
 1.43% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.08% :     0.000000s :      4: predicate.fold_const_symbol
 0.84% :     0.000003s :     19: predicate.get_grad_eliminate
 0.11% :     0.000000s :      4: predicate.graph_param_transform
 4.37% :     0.000017s :     70: predicate.inline
 1.80% :     0.000007s :     40: predicate.inline_without_move
 0.35% :     0.000001s :     19: predicate.j_node_and_user_rematch
 1.02% :     0.000004s :     19: predicate.less_batch_normalization
 1.91% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 2.03% :     0.000008s :     54: predicate.load_eliminater
 0.52% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.10% :     0.000012s :     85: predicate.loop_unroll_before_grad
 1.76% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.42% :     0.000005s :     43: predicate.merge_addn
 1.48% :     0.000006s :     43: predicate.minmaximum_grad
 0.66% :     0.000003s :      4: predicate.mutable_eliminate
 0.20% :     0.000001s :      4: predicate.opt_reshape
 2.32% :     0.000009s :     54: predicate.partial_eliminate
 1.62% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.82% :     0.000007s :     43: predicate.reduce_eliminate
 1.75% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.56% :     0.000010s :     96: predicate.replace_applicator
 0.99% :     0.000004s :     40: predicate.replace_old_param
 0.10% :     0.000000s :      4: predicate.reset_defer_inline
 1.85% :     0.000007s :     43: predicate.reshape_eliminate
 1.51% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.25% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.65% :     0.000006s :     43: predicate.same_eliminate
 0.45% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.41% :     0.000002s :      8: predicate.special_op_eliminate
 0.86% :     0.000003s :     19: predicate.specialize_transform
 1.72% :     0.000007s :     43: predicate.split_environ_get_set_with_tuple_value
 1.55% :     0.000006s :     43: predicate.stack_unstack_eliminate
 0.18% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.65% :     0.000010s :     62: predicate.switch_defer_inline
 2.43% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.08% :     0.000023s :    151: predicate.switch_simplify
 1.56% :     0.000006s :     43: predicate.tile_eliminate
 1.47% :     0.000006s :     43: predicate.transpose_eliminate
 1.81% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.65% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.61% :     0.000014s :     58: predicate.tuple_list_get_item_eliminator
 2.09% :     0.000008s :     43: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000007s :     50: predicate.tuple_to_list_eliminator_
 1.86% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000011s :     73: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000007s :     43: predicate.value_based_eliminate
 0.15% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.24% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001538    29
62.20% :     0.000957s :     13: func_graph_cloner_run.FuncGraphClonerGraph
37.80% :     0.000581s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.041174    91
 0.16% :     0.000065s :      1: add_recomputation
 0.25% :     0.000103s :      1: auto_monad
 0.05% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.65% :     0.000268s :      1: bootstrap
 0.08% :     0.000031s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000031s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.12% :     0.000048s :      1: event_method
 0.03% :     0.000013s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
31.24% :     0.012861s :      1: jit_opt_a
 0.41% :     0.000167s :      1: jit_opt_after_cconv
 0.14% :     0.000058s :      1: jit_opt_b
 1.11% :     0.000456s :      1: loop_unroll
 1.81% :     0.000746s :      1: mutable_eliminate
 6.98% :     0.002872s :     39: opt.transform.jit_opt_a
 0.14% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000030s :      4: opt.transform.jit_opt_b
 0.03% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000040s :      4: opt.transform.symbol_engine_opt
 1.20% :     0.000492s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000030s :      1: py_interpret_to_execute
 0.03% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000020s :      1: remove_dup_value
 4.07% :     0.001674s :      2: renormalize.infer
 3.65% :     0.001503s :      2: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000042s :      1: rewriter_after_opt_a
 0.23% :     0.000096s :      1: rewriter_before_opt_a
 0.21% :     0.000087s :      1: symbol_engine_optimizer
17.87% :     0.007357s :      1: task_emit
28.65% :     0.011795s :      1: type_inference
 0.17% :     0.000071s :      1: validate


TotalTime = 0.0247244, [33]
[bootstrap]: 0.00035242
[type_inference]: 0.00676966
[event_method]: 2.29e-05
[auto_monad]: 7.782e-05
[graph_reusing]: 7.06999e-06
[pre_auto_parallel]: 2.58e-06
[py_interpret_to_execute]: 4.144e-05
[rewriter_before_opt_a]: 7.882e-05
[expand_dump_flag]: 2.98e-06
[jit_opt_a]: 0.00649109, [2]
    [Cycle 1]: 0.00200381, [27]
        [switch_simplify]: 6.822e-05
        [loop_unroll]: 2.944e-05
        [a_1]: 0.00062759
        [with_stream_mark]: 2.826e-05
        [recompute_prepare]: 9.96998e-06
        [updatestate_depend_eliminate]: 4.33001e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.83e-06
        [parameter_eliminate]: 2.41e-06
        [specialize_transform]: 5.98002e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 5.99e-06
        [meta_shard_fg_expand]: 2.83e-06
        [get_grad_eliminate_]: 5.44e-06
        [merge_forward]: 3.91999e-06
        [cell_reuse_recompute_pass]: 2.54999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.478e-05
        [j_node_and_user_rematch]: 9.79999e-06
        [meta_fg_expand]: 2.14e-06
        [replace_old_param]: 1.099e-05
        [inline_without_move]: 5.97001e-06
        [renormalize]: 0.00084701
        [add_forward_monad_depend]: 8.91002e-06
        [auto_monad_grad]: 2.88e-06
        [auto_monad_eliminator]: 2.054e-05
        [cse]: 3.455e-05
        [replace_applicator]: 1.823e-05
    [Cycle 2]: 0.00034341, [27]
        [switch_simplify]: 6.30002e-06
        [loop_unroll]: 4.95999e-06
        [a_1]: 9.04e-05
        [with_stream_mark]: 1.564e-05
        [recompute_prepare]: 4.58999e-06
        [updatestate_depend_eliminate]: 3.66999e-06
        [updatestate_assign_eliminate]: 2.63e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [parameter_eliminate]: 1.62001e-06
        [specialize_transform]: 4.67e-06
        [updatestate_useless_node_eliminater]: 4.55999e-06
        [accelerated_algorithm]: 4.57e-06
        [meta_shard_fg_expand]: 1.66e-06
        [get_grad_eliminate_]: 4.60001e-06
        [merge_forward]: 3.91999e-06
        [cell_reuse_recompute_pass]: 3.40998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.756e-05
        [j_node_and_user_rematch]: 8.64e-06
        [meta_fg_expand]: 1.81998e-06
        [replace_old_param]: 7.60998e-06
        [inline_without_move]: 5.09e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.84e-06
        [auto_monad_grad]: 1.39003e-06
        [auto_monad_eliminator]: 5.87001e-06
        [cse]: 1.228e-05
        [replace_applicator]: 5.42001e-06
[py_interpret_to_execute_after_opt_a]: 1.619e-05
[rewriter_after_opt_a]: 4.352e-05
[convert_after_rewriter]: 7.36999e-06
[order_py_execute_after_rewriter]: 5.09e-06
[mutable_eliminate]: 0.00084193
[jit_opt_b]: 5.241e-05, [1]
    [Cycle 1]: 4.441e-05, [2]
        [frontend_op_eliminate]: 1.682e-05
        [inline_after_opt_a]: 1.587e-05
[cconv]: 3.888e-05
[loop_unroll]: 0.00050346
[jit_opt_after_cconv]: 0.00017449, [1]
    [Cycle 1]: 0.0001668, [11]
        [c_1]: 2.065e-05
        [parameter_eliminate]: 6.84001e-06
        [updatestate_depend_eliminate]: 9.66e-06
        [updatestate_assign_eliminate]: 3.33e-06
        [updatestate_loads_eliminate]: 2.50002e-06
        [cse]: 3.417e-05
        [call_graph_tuple_transform]: 2.591e-05
        [tuple_list_get_item_eliminator]: 5.28002e-06
        [none_parameter_eliminate]: 1.43002e-06
        [renormalize]: 9.70002e-07
        [switch_simplify]: 5.43002e-06
[remove_dup_value]: 1.546e-05
[partial_unused_args_eliminate]: 2.63e-06
[environ_conv]: 6.16e-06
[add_recomputation]: 6.271e-05
[cse_after_recomputation]: 2.267e-05, [1]
    [Cycle 1]: 1.691e-05, [1]
        [cse]: 1.036e-05
[auto_monad_reorder]: 1.696e-05
[get_jit_bprop_graph]: 2.83998e-06
[rewriter_after_jit_bprop_graph]: 1.049e-05
[opt_after_jit_grad]: 0.00054048
[symbol_engine_optimizer]: 7.479e-05, [1]
    [Cycle 1]: 6.826e-05, [6]
        [build]: 5.06002e-06
        [elim_shapecalc]: 7.68001e-06
        [elim_not_effective]: 1.371e-05
        [opt_reshape]: 5.22e-06
        [fold_const_symbol]: 8.13001e-06
        [renormalize]: 5.69999e-07
[validate]: 5.095e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.00810288
[execute]: 1.113e-05
Sums
    bootstrap                            :     0.000352s :  1.78%
    type_inference                       :     0.006770s : 34.11%
    event_method                         :     0.000023s :  0.12%
    auto_monad                           :     0.000078s :  0.39%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.21%
    rewriter_before_opt_a                :     0.000079s :  0.40%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000075s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.17%
    jit_opt_a.a_1                        :     0.000718s :  3.62%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000847s :  4.27%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.13%
    jit_opt_a.cse                        :     0.000047s :  0.24%
    jit_opt_a.replace_applicator         :     0.000024s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.08%
    rewriter_after_opt_a                 :     0.000044s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000842s :  4.24%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.000039s :  0.20%
    loop_unroll                          :     0.000503s :  2.54%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000034s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000063s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.05%
    opt_after_jit_grad                   :     0.000540s :  2.72%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000051s :  0.26%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008103s : 40.83%
    execute                              :     0.000011s :  0.06%


Time group info:
------[substitution.]   0.000284    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 2.19% :     0.000006s :      3: substitution.graph_param_transform
85.96% :     0.000244s :      4: substitution.inline
 1.60% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.76% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.11% :     0.000006s :      2: substitution.replace_old_param
 5.10% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006670     2
67.18% :     0.004481s :      1: type_inference.infer
32.82% :     0.002189s :      1: type_inference.specialize
------[replace.]   0.000076     6
70.17% :     0.000054s :      4: replace.inline
29.83% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000254     6
94.73% :     0.000240s :      4: match.inline
 5.27% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.36% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.58% :     0.000002s :     10: predicate.addn_zero_filter
 2.08% :     0.000003s :     10: predicate.arithmetic_simplify
 1.57% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 0.99% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.05% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     10: predicate.environ_get_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.81% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 6.65% :     0.000008s :     22: predicate.inline
 1.29% :     0.000002s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.65% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.58% :     0.000002s :     15: predicate.load_eliminater
 2.11% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.81% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 2.65% :     0.000003s :      3: predicate.mutable_eliminate
 0.39% :     0.000000s :      3: predicate.opt_reshape
 2.01% :     0.000003s :     15: predicate.partial_eliminate
 1.33% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     10: predicate.reduce_eliminate
 1.33% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.52% :     0.000002s :     18: predicate.replace_applicator
 0.87% :     0.000001s :      6: predicate.replace_old_param
 0.52% :     0.000001s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.38% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.74% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.18% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.13% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 1.93% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.07% :     0.000010s :     53: predicate.switch_simplify
 1.02% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.14% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.06% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.04% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.07% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.30% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.42% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000744    12
46.72% :     0.000348s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.28% :     0.000062s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.00% :     0.000335s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026639    76
 0.25% :     0.000066s :      1: add_recomputation
 0.31% :     0.000082s :      1: auto_monad
 0.07% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.42% :     0.000379s :      1: bootstrap
 0.16% :     0.000042s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.11% :     0.000029s :      1: event_method
 0.07% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
24.39% :     0.006496s :      1: jit_opt_a
 0.67% :     0.000178s :      1: jit_opt_after_cconv
 0.21% :     0.000055s :      1: jit_opt_b
 1.93% :     0.000515s :      1: loop_unroll
 3.22% :     0.000857s :      1: mutable_eliminate
 3.58% :     0.000953s :     26: opt.transform.jit_opt_a
 0.20% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.07% :     0.000552s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000044s :      1: py_interpret_to_execute
 0.07% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 1.79% :     0.000476s :      1: renormalize.infer
 1.36% :     0.000361s :      1: renormalize.specialize
 0.05% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000048s :      1: rewriter_after_opt_a
 0.31% :     0.000082s :      1: rewriter_before_opt_a
 0.29% :     0.000078s :      1: symbol_engine_optimizer
30.50% :     0.008126s :      1: task_emit
25.53% :     0.006800s :      1: type_inference
 0.29% :     0.000077s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.688.765 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dd27e50> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.689.443 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0139116, [30]
[bootstrap]: 0.000335
[type_inference]: 0.0048906
[event_method]: 1.928e-05
[auto_monad]: 6.578e-05
[graph_reusing]: 5.77999e-06
[pre_auto_parallel]: 2.53003e-06
[py_interpret_to_execute]: 4.028e-05
[rewriter_before_opt_a]: 8.586e-05
[expand_dump_flag]: 3.67998e-06
[jit_opt_a]: 0.0060103, [2]
    [Cycle 1]: 0.00198092, [27]
        [switch_simplify]: 5.895e-05
        [loop_unroll]: 3.176e-05
        [a_1]: 0.00069725
        [with_stream_mark]: 2.006e-05
        [recompute_prepare]: 8.64e-06
        [updatestate_depend_eliminate]: 3.86001e-06
        [updatestate_assign_eliminate]: 2.91999e-06
        [updatestate_loads_eliminate]: 3.04999e-06
        [parameter_eliminate]: 2.44999e-06
        [specialize_transform]: 6.11998e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.74e-06
        [meta_shard_fg_expand]: 3.35e-06
        [get_grad_eliminate_]: 5.95002e-06
        [merge_forward]: 4.08001e-06
        [cell_reuse_recompute_pass]: 1.79998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.753e-05
        [j_node_and_user_rematch]: 1.127e-05
        [meta_fg_expand]: 3.01001e-06
        [replace_old_param]: 8.84003e-06
        [inline_without_move]: 5.81e-06
        [renormalize]: 0.00079058
        [add_forward_monad_depend]: 6.32001e-06
        [auto_monad_grad]: 2.89999e-06
        [auto_monad_eliminator]: 1.54e-05
        [cse]: 3.157e-05
        [replace_applicator]: 1.55e-05
    [Cycle 2]: 0.00037419, [27]
        [switch_simplify]: 6.71999e-06
        [loop_unroll]: 5.89999e-06
        [a_1]: 0.00010002
        [with_stream_mark]: 1.112e-05
        [recompute_prepare]: 5.89e-06
        [updatestate_depend_eliminate]: 3.19001e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 2.13002e-06
        [parameter_eliminate]: 9.29984e-07
        [specialize_transform]: 5.25999e-06
        [updatestate_useless_node_eliminater]: 5.38002e-06
        [accelerated_algorithm]: 5.82999e-06
        [meta_shard_fg_expand]: 2.76e-06
        [get_grad_eliminate_]: 4.88001e-06
        [merge_forward]: 2.86e-06
        [cell_reuse_recompute_pass]: 1.51998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.532e-05
        [j_node_and_user_rematch]: 8.32998e-06
        [meta_fg_expand]: 2.16e-06
        [replace_old_param]: 8.45001e-06
        [inline_without_move]: 4.94998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.71e-06
        [auto_monad_grad]: 1.00001e-06
        [auto_monad_eliminator]: 8.87e-06
        [cse]: 1.368e-05
        [replace_applicator]: 6.39999e-06
[py_interpret_to_execute_after_opt_a]: 1.19e-05
[rewriter_after_opt_a]: 3.693e-05
[convert_after_rewriter]: 1.183e-05
[order_py_execute_after_rewriter]: 5.76e-06
[mutable_eliminate]: 0.00061254
[jit_opt_b]: 5.93e-05, [1]
    [Cycle 1]: 5.029e-05, [2]
        [frontend_op_eliminate]: 2.008e-05
        [inline_after_opt_a]: 1.736e-05
[cconv]: 2.765e-05
[loop_unroll]: 0.00050457
[jit_opt_after_cconv]: 0.00015485, [1]
    [Cycle 1]: 0.00014643, [11]
        [c_1]: 2.342e-05
        [parameter_eliminate]: 2.81e-06
        [updatestate_depend_eliminate]: 6.56e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 2.31998e-06
        [cse]: 2.053e-05
        [call_graph_tuple_transform]: 2.18e-05
        [tuple_list_get_item_eliminator]: 6.24001e-06
        [none_parameter_eliminate]: 1.81e-06
        [renormalize]: 7.39994e-07
        [switch_simplify]: 5.79e-06
[remove_dup_value]: 1.494e-05
[partial_unused_args_eliminate]: 2.09e-06
[environ_conv]: 4.92e-06
[add_recomputation]: 4.865e-05
[cse_after_recomputation]: 2.35e-05, [1]
    [Cycle 1]: 1.719e-05, [1]
        [cse]: 1.054e-05
[auto_monad_reorder]: 1.587e-05
[get_jit_bprop_graph]: 2.07999e-06
[rewriter_after_jit_bprop_graph]: 6.81999e-06
[opt_after_jit_grad]: 0.00058241
[symbol_engine_optimizer]: 8.517e-05, [1]
    [Cycle 1]: 7.745e-05, [6]
        [build]: 4.37e-06
        [elim_shapecalc]: 7.99002e-06
        [elim_not_effective]: 1.676e-05
        [opt_reshape]: 6.96001e-06
        [fold_const_symbol]: 1.003e-05
        [renormalize]: 5.3001e-07
[validate]: 3.799e-05
Sums
    bootstrap                            :     0.000335s :  3.50%
    type_inference                       :     0.004891s : 51.09%
    event_method                         :     0.000019s :  0.20%
    auto_monad                           :     0.000066s :  0.69%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000040s :  0.42%
    rewriter_before_opt_a                :     0.000086s :  0.90%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000066s :  0.69%
    jit_opt_a.loop_unroll                :     0.000038s :  0.39%
    jit_opt_a.a_1                        :     0.000797s :  8.33%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.06%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.18%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000791s :  8.26%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.25%
    jit_opt_a.cse                        :     0.000045s :  0.47%
    jit_opt_a.replace_applicator         :     0.000022s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.12%
    rewriter_after_opt_a                 :     0.000037s :  0.39%
    convert_after_rewriter               :     0.000012s :  0.12%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000613s :  6.40%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.18%
    cconv                                :     0.000028s :  0.29%
    loop_unroll                          :     0.000505s :  5.27%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000049s :  0.51%
    cse_after_recomputation.cse          :     0.000011s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000582s :  6.08%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.18%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000038s :  0.40%


Time group info:
------[substitution.]   0.000311    23
 0.68% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 1.73% :     0.000005s :      3: substitution.graph_param_transform
88.00% :     0.000273s :      4: substitution.inline
 1.13% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.55% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.64% :     0.000005s :      2: substitution.replace_old_param
 4.80% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004818     2
80.14% :     0.003862s :      1: type_inference.infer
19.86% :     0.000957s :      1: type_inference.specialize
------[replace.]   0.000086     6
64.51% :     0.000056s :      4: replace.inline
35.49% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000283     6
95.21% :     0.000270s :      4: match.inline
 4.79% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000153   691
 1.19% :     0.000002s :     10: predicate.accumulaten_eliminater
10.59% :     0.000016s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.55% :     0.000002s :     10: predicate.addn_zero_filter
 2.00% :     0.000003s :     10: predicate.arithmetic_simplify
 0.96% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.93% :     0.000001s :     10: predicate.compare_switch_simplify
 1.84% :     0.000003s :     10: predicate.depend_value_elim
 1.49% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.15% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.92% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.45% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.28% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.83% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.91% :     0.000001s :     10: predicate.environ_get_eliminate
 0.99% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.81% :     0.000001s :      6: predicate.get_grad_eliminate
 0.44% :     0.000001s :      3: predicate.graph_param_transform
 4.53% :     0.000007s :     22: predicate.inline
 0.77% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000002s :      6: predicate.less_batch_normalization
 1.18% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.39% :     0.000002s :     15: predicate.load_eliminater
 1.65% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.44% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.78% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 0.86% :     0.000001s :     10: predicate.minmaximum_grad
 1.47% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 1.81% :     0.000003s :     15: predicate.partial_eliminate
 1.00% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.21% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000003s :     18: predicate.replace_applicator
 0.44% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.85% :     0.000003s :     10: predicate.reshape_eliminate
 1.03% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.44% :     0.000002s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.31% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000004s :     16: predicate.switch_defer_inline
 1.87% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.25% :     0.000010s :     53: predicate.switch_simplify
 1.05% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000002s :     10: predicate.transpose_eliminate
 1.51% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.06% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.77% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.38% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.69% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000658    12
49.90% :     0.000328s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.44% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.66% :     0.000294s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015813    72
 0.33% :     0.000052s :      1: add_recomputation
 0.44% :     0.000069s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 2.24% :     0.000355s :      1: bootstrap
 0.19% :     0.000031s :      1: cconv
 0.10% :     0.000015s :      1: convert_after_rewriter
 0.16% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.03% :     0.006014s :      1: jit_opt_a
 1.00% :     0.000158s :      1: jit_opt_after_cconv
 0.39% :     0.000062s :      1: jit_opt_b
 3.27% :     0.000517s :      1: loop_unroll
 3.96% :     0.000626s :      1: mutable_eliminate
 6.50% :     0.001028s :     26: opt.transform.jit_opt_a
 0.34% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000030s :      4: opt.transform.jit_opt_b
 0.09% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.26% :     0.000041s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000038s :      4: opt.transform.symbol_engine_opt
 3.75% :     0.000593s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000043s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.53% :     0.000400s :      1: renormalize.infer
 2.40% :     0.000379s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000040s :      1: rewriter_after_opt_a
 0.57% :     0.000090s :      1: rewriter_before_opt_a
 0.56% :     0.000088s :      1: symbol_engine_optimizer
31.03% :     0.004907s :      1: type_inference


TotalTime = 0.025719, [33]
[bootstrap]: 0.00034374
[type_inference]: 0.00656152
[event_method]: 2.336e-05
[auto_monad]: 7.395e-05
[graph_reusing]: 6.94999e-06
[pre_auto_parallel]: 3.40998e-06
[py_interpret_to_execute]: 3.812e-05
[rewriter_before_opt_a]: 7.719e-05
[expand_dump_flag]: 3.68999e-06
[jit_opt_a]: 0.00635822, [2]
    [Cycle 1]: 0.00199939, [27]
        [switch_simplify]: 6.847e-05
        [loop_unroll]: 2.852e-05
        [a_1]: 0.00060646
        [with_stream_mark]: 2.8e-05
        [recompute_prepare]: 8.57e-06
        [updatestate_depend_eliminate]: 5.16998e-06
        [updatestate_assign_eliminate]: 3.66999e-06
        [updatestate_loads_eliminate]: 2.94001e-06
        [parameter_eliminate]: 2.06e-06
        [specialize_transform]: 6.12001e-06
        [updatestate_useless_node_eliminater]: 5.22999e-06
        [accelerated_algorithm]: 5.68002e-06
        [meta_shard_fg_expand]: 2.32999e-06
        [get_grad_eliminate_]: 4.88001e-06
        [merge_forward]: 3.86999e-06
        [cell_reuse_recompute_pass]: 1.45001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.652e-05
        [j_node_and_user_rematch]: 9.16998e-06
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 8.95999e-06
        [inline_without_move]: 5.84999e-06
        [renormalize]: 0.00087724
        [add_forward_monad_depend]: 8.15e-06
        [auto_monad_grad]: 3.23e-06
        [auto_monad_eliminator]: 1.921e-05
        [cse]: 3.545e-05
        [replace_applicator]: 1.909e-05
    [Cycle 2]: 0.00034061, [27]
        [switch_simplify]: 5.81e-06
        [loop_unroll]: 5.15999e-06
        [a_1]: 8.96e-05
        [with_stream_mark]: 1.447e-05
        [recompute_prepare]: 5.00999e-06
        [updatestate_depend_eliminate]: 3.46999e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.43998e-06
        [parameter_eliminate]: 1.70001e-06
        [specialize_transform]: 4.58999e-06
        [updatestate_useless_node_eliminater]: 4.62e-06
        [accelerated_algorithm]: 5.18002e-06
        [meta_shard_fg_expand]: 1.64e-06
        [get_grad_eliminate_]: 4.89998e-06
        [merge_forward]: 3.71999e-06
        [cell_reuse_recompute_pass]: 2.76999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.542e-05
        [j_node_and_user_rematch]: 8.64e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 8.97e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.62001e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 6.26e-06
        [cse]: 1.264e-05
        [replace_applicator]: 5.26002e-06
[py_interpret_to_execute_after_opt_a]: 1.552e-05
[rewriter_after_opt_a]: 4.166e-05
[convert_after_rewriter]: 6.66e-06
[order_py_execute_after_rewriter]: 5.06997e-06
[mutable_eliminate]: 0.00076611
[jit_opt_b]: 5.394e-05, [1]
    [Cycle 1]: 4.379e-05, [2]
        [frontend_op_eliminate]: 1.597e-05
        [inline_after_opt_a]: 1.58e-05
[cconv]: 3.666e-05
[loop_unroll]: 0.00046344
[jit_opt_after_cconv]: 0.00017081, [1]
    [Cycle 1]: 0.00016317, [11]
        [c_1]: 2.048e-05
        [parameter_eliminate]: 4.82e-06
        [updatestate_depend_eliminate]: 1.067e-05
        [updatestate_assign_eliminate]: 2.74999e-06
        [updatestate_loads_eliminate]: 2.67001e-06
        [cse]: 3.378e-05
        [call_graph_tuple_transform]: 2.431e-05
        [tuple_list_get_item_eliminator]: 5.18002e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 6.39993e-07
        [switch_simplify]: 5.37001e-06
[remove_dup_value]: 1.518e-05
[partial_unused_args_eliminate]: 2.39999e-06
[environ_conv]: 5.54e-06
[add_recomputation]: 6.154e-05
[cse_after_recomputation]: 2.304e-05, [1]
    [Cycle 1]: 1.69e-05, [1]
        [cse]: 1.03e-05
[auto_monad_reorder]: 1.608e-05
[get_jit_bprop_graph]: 2.18998e-06
[rewriter_after_jit_bprop_graph]: 9.44e-06
[opt_after_jit_grad]: 0.00054394
[symbol_engine_optimizer]: 7.767e-05, [1]
    [Cycle 1]: 7.071e-05, [6]
        [build]: 5.43002e-06
        [elim_shapecalc]: 7.83001e-06
        [elim_not_effective]: 1.404e-05
        [opt_reshape]: 5.57999e-06
        [fold_const_symbol]: 9.06002e-06
        [renormalize]: 7.10017e-07
[validate]: 4.977e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.0095808
[execute]: 1.092e-05
Sums
    bootstrap                            :     0.000344s :  1.64%
    type_inference                       :     0.006562s : 31.28%
    event_method                         :     0.000023s :  0.11%
    auto_monad                           :     0.000074s :  0.35%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.18%
    rewriter_before_opt_a                :     0.000077s :  0.37%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.35%
    jit_opt_a.loop_unroll                :     0.000034s :  0.16%
    jit_opt_a.a_1                        :     0.000696s :  3.32%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.000877s :  4.18%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.12%
    jit_opt_a.cse                        :     0.000048s :  0.23%
    jit_opt_a.replace_applicator         :     0.000024s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.07%
    rewriter_after_opt_a                 :     0.000042s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000766s :  3.65%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.000037s :  0.17%
    loop_unroll                          :     0.000463s :  2.21%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000034s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000062s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000544s :  2.59%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000050s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009581s : 45.68%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000276    23
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 2.32% :     0.000006s :      3: substitution.graph_param_transform
85.70% :     0.000237s :      4: substitution.inline
 1.60% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.18% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.27% :     0.000006s :      2: substitution.replace_old_param
 4.69% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006472     2
85.51% :     0.005534s :      1: type_inference.infer
14.49% :     0.000938s :      1: type_inference.specialize
------[replace.]   0.000072     6
70.63% :     0.000051s :      4: replace.inline
29.37% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000245     6
95.23% :     0.000234s :      4: match.inline
 4.77% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.61% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.23% :     0.000001s :     10: predicate.addn_zero_filter
 1.63% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 0.99% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.07% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 6.40% :     0.000008s :     22: predicate.inline
 1.39% :     0.000002s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.60% :     0.000002s :      6: predicate.less_batch_normalization
 1.79% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.87% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.95% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 1.95% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.77% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.55% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.49% :     0.000001s :      3: predicate.reset_defer_inline
 1.60% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.51% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.15% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.13% :     0.000010s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.19% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.02% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.24% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.35% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.37% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.89% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000712    12
47.61% :     0.000339s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.46% :     0.000060s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.93% :     0.000313s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027638    76
 0.23% :     0.000065s :      1: add_recomputation
 0.28% :     0.000078s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.35% :     0.000373s :      1: bootstrap
 0.14% :     0.000040s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000028s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
23.02% :     0.006362s :      1: jit_opt_a
 0.63% :     0.000174s :      1: jit_opt_after_cconv
 0.21% :     0.000057s :      1: jit_opt_b
 1.72% :     0.000475s :      1: loop_unroll
 2.83% :     0.000782s :      1: mutable_eliminate
 3.36% :     0.000930s :     26: opt.transform.jit_opt_a
 0.19% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.01% :     0.000555s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.15% :     0.000041s :      1: py_interpret_to_execute
 0.07% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 1.79% :     0.000495s :      1: renormalize.infer
 1.35% :     0.000373s :      1: renormalize.specialize
 0.04% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000046s :      1: rewriter_after_opt_a
 0.29% :     0.000080s :      1: rewriter_before_opt_a
 0.29% :     0.000080s :      1: symbol_engine_optimizer
34.75% :     0.009603s :      1: task_emit
23.85% :     0.006591s :      1: type_inference
 0.28% :     0.000077s :      1: validate


TotalTime = 0.0343215, [33]
[bootstrap]: 0.00022594
[type_inference]: 0.0111378
[event_method]: 4.606e-05
[auto_monad]: 0.00011205
[graph_reusing]: 7.68999e-06
[pre_auto_parallel]: 2.94999e-06
[py_interpret_to_execute]: 2.848e-05
[rewriter_before_opt_a]: 9.906e-05
[expand_dump_flag]: 3.4e-06
[jit_opt_a]: 0.0124387, [3]
    [Cycle 1]: 0.00613886, [27]
        [switch_simplify]: 7.522e-05
        [loop_unroll]: 5.338e-05
        [a_1]: 0.00086783
        [with_stream_mark]: 3.384e-05
        [recompute_prepare]: 2.002e-05
        [updatestate_depend_eliminate]: 8.90999e-06
        [updatestate_assign_eliminate]: 7.36999e-06
        [updatestate_loads_eliminate]: 6.21e-06
        [parameter_eliminate]: 2.56e-06
        [specialize_transform]: 1.352e-05
        [updatestate_useless_node_eliminater]: 1.294e-05
        [accelerated_algorithm]: 1.342e-05
        [meta_shard_fg_expand]: 3.63999e-06
        [get_grad_eliminate_]: 1.282e-05
        [merge_forward]: 8.97e-06
        [cell_reuse_recompute_pass]: 9.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.875e-05
        [j_node_and_user_rematch]: 2.342e-05
        [meta_fg_expand]: 0.00169186
        [replace_old_param]: 6.15e-05
        [inline_without_move]: 5.549e-05
        [renormalize]: 0.00259653
        [add_forward_monad_depend]: 1.199e-05
        [auto_monad_grad]: 6.31e-06
        [auto_monad_eliminator]: 5.126e-05
        [cse]: 0.00015619
        [replace_applicator]: 7.37e-05
    [Cycle 2]: 0.00221192, [27]
        [switch_simplify]: 3.548e-05
        [loop_unroll]: 3.507e-05
        [a_1]: 0.00106421
        [with_stream_mark]: 1.798e-05
        [recompute_prepare]: 8.2e-06
        [updatestate_depend_eliminate]: 4.45e-06
        [updatestate_assign_eliminate]: 3.87998e-06
        [updatestate_loads_eliminate]: 3.46999e-06
        [parameter_eliminate]: 2.01998e-06
        [specialize_transform]: 6.72002e-06
        [updatestate_useless_node_eliminater]: 5.92001e-06
        [accelerated_algorithm]: 5.97001e-06
        [meta_shard_fg_expand]: 2.61e-06
        [get_grad_eliminate_]: 6.46e-06
        [merge_forward]: 4.11001e-06
        [cell_reuse_recompute_pass]: 1.10999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.607e-05
        [j_node_and_user_rematch]: 1.07e-05
        [meta_fg_expand]: 4.761e-05
        [replace_old_param]: 1.076e-05
        [inline_without_move]: 6.56999e-06
        [renormalize]: 0.00069245
        [add_forward_monad_depend]: 5.08002e-06
        [auto_monad_grad]: 2.74001e-06
        [auto_monad_eliminator]: 1.525e-05
        [cse]: 2.739e-05
        [replace_applicator]: 1.412e-05
    [Cycle 3]: 0.00038157, [27]
        [switch_simplify]: 6.84999e-06
        [loop_unroll]: 6.26e-06
        [a_1]: 0.00012078
        [with_stream_mark]: 1.099e-05
        [recompute_prepare]: 6.16998e-06
        [updatestate_depend_eliminate]: 4.22003e-06
        [updatestate_assign_eliminate]: 2.69999e-06
        [updatestate_loads_eliminate]: 2.66e-06
        [parameter_eliminate]: 1.06002e-06
        [specialize_transform]: 6.59999e-06
        [updatestate_useless_node_eliminater]: 5.87999e-06
        [accelerated_algorithm]: 6.49999e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 5.67001e-06
        [merge_forward]: 3.68e-06
        [cell_reuse_recompute_pass]: 2.14999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.506e-05
        [j_node_and_user_rematch]: 1.055e-05
        [meta_fg_expand]: 2.38998e-06
        [replace_old_param]: 8.64998e-06
        [inline_without_move]: 5.94999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 2.07999e-06
        [auto_monad_grad]: 6.59988e-07
        [auto_monad_eliminator]: 7.26001e-06
        [cse]: 1.664e-05
        [replace_applicator]: 5.76e-06
[py_interpret_to_execute_after_opt_a]: 1.454e-05
[rewriter_after_opt_a]: 4.345e-05
[convert_after_rewriter]: 7.9e-06
[order_py_execute_after_rewriter]: 5.74e-06
[mutable_eliminate]: 0.00076471
[jit_opt_b]: 6.008e-05, [1]
    [Cycle 1]: 5.01e-05, [2]
        [frontend_op_eliminate]: 2.146e-05
        [inline_after_opt_a]: 1.735e-05
[cconv]: 2.991e-05
[loop_unroll]: 0.00052112
[jit_opt_after_cconv]: 0.00017453, [1]
    [Cycle 1]: 0.00016591, [11]
        [c_1]: 2.725e-05
        [parameter_eliminate]: 3.85998e-06
        [updatestate_depend_eliminate]: 8.87999e-06
        [updatestate_assign_eliminate]: 3.36999e-06
        [updatestate_loads_eliminate]: 3.01001e-06
        [cse]: 3.027e-05
        [call_graph_tuple_transform]: 2.283e-05
        [tuple_list_get_item_eliminator]: 6.46e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 7.40023e-07
        [switch_simplify]: 6.38e-06
[remove_dup_value]: 3.112e-05
[partial_unused_args_eliminate]: 2.18002e-06
[environ_conv]: 6.46e-06
[add_recomputation]: 5.755e-05
[cse_after_recomputation]: 2.877e-05, [1]
    [Cycle 1]: 2.275e-05, [1]
        [cse]: 1.598e-05
[auto_monad_reorder]: 1.974e-05
[get_jit_bprop_graph]: 1.91e-06
[rewriter_after_jit_bprop_graph]: 7.41999e-06
[opt_after_jit_grad]: 0.00054924
[symbol_engine_optimizer]: 8.441e-05, [1]
    [Cycle 1]: 7.616e-05, [6]
        [build]: 4.07e-06
        [elim_shapecalc]: 9.07001e-06
        [elim_not_effective]: 1.619e-05
        [opt_reshape]: 6.81001e-06
        [fold_const_symbol]: 1.053e-05
        [renormalize]: 6.10016e-07
[validate]: 4.252e-05
[backend_pass]: 1.02998e-06
[task_emit]: 0.0075384
[execute]: 7.49002e-06
Sums
    bootstrap                            :     0.000226s :  0.76%
    type_inference                       :     0.011138s : 37.43%
    event_method                         :     0.000046s :  0.15%
    auto_monad                           :     0.000112s :  0.38%
    graph_reusing                        :     0.000008s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.10%
    rewriter_before_opt_a                :     0.000099s :  0.33%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000118s :  0.40%
    jit_opt_a.loop_unroll                :     0.000095s :  0.32%
    jit_opt_a.a_1                        :     0.002053s :  6.90%
    jit_opt_a.with_stream_mark           :     0.000063s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000034s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000018s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000014s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.02%
    jit_opt_a.specialize_transform       :     0.000027s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000026s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.08%
    jit_opt_a.merge_forward              :     0.000017s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000060s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000045s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001742s :  5.85%
    jit_opt_a.replace_old_param          :     0.000081s :  0.27%
    jit_opt_a.inline_without_move        :     0.000068s :  0.23%
    jit_opt_a.renormalize                :     0.003289s : 11.05%
    jit_opt_a.add_forward_monad_depend   :     0.000019s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000010s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000074s :  0.25%
    jit_opt_a.cse                        :     0.000200s :  0.67%
    jit_opt_a.replace_applicator         :     0.000094s :  0.31%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.05%
    rewriter_after_opt_a                 :     0.000043s :  0.15%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000765s :  2.57%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.06%
    cconv                                :     0.000030s :  0.10%
    loop_unroll                          :     0.000521s :  1.75%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000030s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000031s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000058s :  0.19%
    cse_after_recomputation.cse          :     0.000016s :  0.05%
    auto_monad_reorder                   :     0.000020s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000549s :  1.85%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007538s : 25.33%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000536   110
 0.49% :     0.000003s :      3: substitution.elim_not_effective
 0.31% :     0.000002s :      3: substitution.fold_const_symbol
 1.21% :     0.000006s :      4: substitution.graph_param_transform
64.88% :     0.000347s :     12: substitution.inline
 3.60% :     0.000019s :      2: substitution.inline_without_move
 1.77% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.03% :     0.000011s :      7: substitution.minmaximum_grad
 1.30% :     0.000007s :      5: substitution.partial_eliminate
 2.11% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.31% :     0.000028s :     10: substitution.replace_applicator
 1.67% :     0.000009s :      7: substitution.replace_old_param
 0.52% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.23% :     0.000023s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.79% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 7.78% :     0.000042s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011062     2
89.79% :     0.009932s :      1: type_inference.infer
10.21% :     0.001129s :      1: type_inference.specialize
------[replace.]   0.000155    19
64.17% :     0.000100s :     12: replace.inline
35.83% :     0.000056s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000361    19
94.25% :     0.000340s :     12: match.inline
 5.75% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000381  2568
 1.62% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.56% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.37% :     0.000005s :     43: predicate.addn_check_dump
 1.71% :     0.000007s :     43: predicate.addn_zero_filter
 2.03% :     0.000008s :     43: predicate.arithmetic_simplify
 1.60% :     0.000006s :     43: predicate.cast_eliminate
 0.19% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.40% :     0.000005s :     43: predicate.compare_switch_simplify
 1.53% :     0.000006s :     43: predicate.depend_value_elim
 1.45% :     0.000006s :     43: predicate.dict_get_item_const_eliminator
 1.46% :     0.000006s :     43: predicate.dict_get_item_eliminator
 1.54% :     0.000006s :     43: predicate.dict_set_item_eliminator
 0.29% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.14% :     0.000001s :      4: predicate.elim_not_effective
 0.22% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000006s :     43: predicate.environ_add_const_eliminate
 1.40% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.39% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.54% :     0.000006s :     43: predicate.environ_get_eliminate
 1.43% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.91% :     0.000003s :     19: predicate.get_grad_eliminate
 0.12% :     0.000000s :      4: predicate.graph_param_transform
 4.57% :     0.000017s :     70: predicate.inline
 1.80% :     0.000007s :     40: predicate.inline_without_move
 0.35% :     0.000001s :     19: predicate.j_node_and_user_rematch
 1.10% :     0.000004s :     19: predicate.less_batch_normalization
 1.70% :     0.000006s :     50: predicate.list_to_tuple_eliminator_
 1.88% :     0.000007s :     54: predicate.load_eliminater
 0.69% :     0.000003s :      4: predicate.loop_unroll_after_grad
 3.29% :     0.000013s :     85: predicate.loop_unroll_before_grad
 1.82% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.39% :     0.000005s :     43: predicate.merge_addn
 1.44% :     0.000005s :     43: predicate.minmaximum_grad
 0.78% :     0.000003s :      4: predicate.mutable_eliminate
 0.20% :     0.000001s :      4: predicate.opt_reshape
 2.19% :     0.000008s :     54: predicate.partial_eliminate
 1.58% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.99% :     0.000008s :     43: predicate.reduce_eliminate
 1.74% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.52% :     0.000010s :     96: predicate.replace_applicator
 0.99% :     0.000004s :     40: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.71% :     0.000007s :     43: predicate.reshape_eliminate
 1.51% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.45% :     0.000002s :      4: predicate.row_tensor_eliminate
 1.60% :     0.000006s :     43: predicate.same_eliminate
 0.52% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.42% :     0.000002s :      8: predicate.special_op_eliminate
 0.82% :     0.000003s :     19: predicate.specialize_transform
 1.58% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.59% :     0.000006s :     43: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.66% :     0.000010s :     62: predicate.switch_defer_inline
 2.49% :     0.000010s :     62: predicate.switch_layer_defer_inline
 6.18% :     0.000024s :    151: predicate.switch_simplify
 1.50% :     0.000006s :     43: predicate.tile_eliminate
 1.46% :     0.000006s :     43: predicate.transpose_eliminate
 1.81% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.62% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.18% :     0.000012s :     58: predicate.tuple_list_get_item_eliminator
 1.98% :     0.000008s :     43: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000007s :     50: predicate.tuple_to_list_eliminator_
 1.86% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000011s :     73: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000007s :     43: predicate.value_based_eliminate
 0.14% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.27% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001547    29
59.79% :     0.000925s :     13: func_graph_cloner_run.FuncGraphClonerGraph
40.21% :     0.000622s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.040447    91
 0.15% :     0.000061s :      1: add_recomputation
 0.29% :     0.000117s :      1: auto_monad
 0.06% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.60% :     0.000242s :      1: bootstrap
 0.08% :     0.000033s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000031s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.12% :     0.000050s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000011s :      1: graph_reusing
30.76% :     0.012443s :      1: jit_opt_a
 0.44% :     0.000177s :      1: jit_opt_after_cconv
 0.15% :     0.000063s :      1: jit_opt_b
 1.32% :     0.000532s :      1: loop_unroll
 1.92% :     0.000776s :      1: mutable_eliminate
 6.68% :     0.002700s :     39: opt.transform.jit_opt_a
 0.15% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000032s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.38% :     0.000559s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000031s :      1: py_interpret_to_execute
 0.04% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000034s :      1: remove_dup_value
 4.36% :     0.001762s :      2: renormalize.infer
 3.73% :     0.001509s :      2: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000046s :      1: rewriter_after_opt_a
 0.25% :     0.000102s :      1: rewriter_before_opt_a
 0.21% :     0.000087s :      1: symbol_engine_optimizer
18.68% :     0.007554s :      1: task_emit
27.57% :     0.011152s :      1: type_inference
 0.17% :     0.000068s :      1: validate


TotalTime = 0.0139574, [30]
[bootstrap]: 0.00033469
[type_inference]: 0.00487361
[event_method]: 1.913e-05
[auto_monad]: 6.981e-05
[graph_reusing]: 5.29e-06
[pre_auto_parallel]: 2.54999e-06
[py_interpret_to_execute]: 4.532e-05
[rewriter_before_opt_a]: 7.674e-05
[expand_dump_flag]: 3.01999e-06
[jit_opt_a]: 0.00595544, [2]
    [Cycle 1]: 0.00195784, [27]
        [switch_simplify]: 6.658e-05
        [loop_unroll]: 3.145e-05
        [a_1]: 0.00061494
        [with_stream_mark]: 2.474e-05
        [recompute_prepare]: 8.10999e-06
        [updatestate_depend_eliminate]: 3.91999e-06
        [updatestate_assign_eliminate]: 3.27002e-06
        [updatestate_loads_eliminate]: 2.86999e-06
        [parameter_eliminate]: 1.92999e-06
        [specialize_transform]: 5.58002e-06
        [updatestate_useless_node_eliminater]: 5.91e-06
        [accelerated_algorithm]: 5.77001e-06
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 5.61e-06
        [merge_forward]: 3.76999e-06
        [cell_reuse_recompute_pass]: 1.47999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.497e-05
        [j_node_and_user_rematch]: 9.91e-06
        [meta_fg_expand]: 2.43998e-06
        [replace_old_param]: 9.27999e-06
        [inline_without_move]: 5.87001e-06
        [renormalize]: 0.00087692
        [add_forward_monad_depend]: 5.94999e-06
        [auto_monad_grad]: 2.83e-06
        [auto_monad_eliminator]: 1.485e-05
        [cse]: 2.651e-05
        [replace_applicator]: 1.488e-05
    [Cycle 2]: 0.00035325, [27]
        [switch_simplify]: 5.92999e-06
        [loop_unroll]: 5.16002e-06
        [a_1]: 9.845e-05
        [with_stream_mark]: 1.074e-05
        [recompute_prepare]: 5.07999e-06
        [updatestate_depend_eliminate]: 3.11001e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 5.04998e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.99999e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 4.80001e-06
        [merge_forward]: 2.99999e-06
        [cell_reuse_recompute_pass]: 1.58002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.524e-05
        [j_node_and_user_rematch]: 9.27001e-06
        [meta_fg_expand]: 1.99999e-06
        [replace_old_param]: 8.05999e-06
        [inline_without_move]: 5.09e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.81998e-06
        [auto_monad_grad]: 1.34998e-06
        [auto_monad_eliminator]: 5.84e-06
        [cse]: 1.131e-05
        [replace_applicator]: 6.38e-06
[py_interpret_to_execute_after_opt_a]: 1.303e-05
[rewriter_after_opt_a]: 3.643e-05
[convert_after_rewriter]: 7.15e-06
[order_py_execute_after_rewriter]: 5.34e-06
[mutable_eliminate]: 0.00070238
[jit_opt_b]: 6.045e-05, [1]
    [Cycle 1]: 5.075e-05, [2]
        [frontend_op_eliminate]: 2.042e-05
        [inline_after_opt_a]: 1.647e-05
[cconv]: 2.95e-05
[loop_unroll]: 0.00052616
[jit_opt_after_cconv]: 0.00017828, [1]
    [Cycle 1]: 0.00017001, [11]
        [c_1]: 2.386e-05
        [parameter_eliminate]: 3.68999e-06
        [updatestate_depend_eliminate]: 7.31001e-06
        [updatestate_assign_eliminate]: 2.45002e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [cse]: 2.371e-05
        [call_graph_tuple_transform]: 2.353e-05
        [tuple_list_get_item_eliminator]: 7.31001e-06
        [none_parameter_eliminate]: 1.74e-06
        [renormalize]: 3.60014e-07
        [switch_simplify]: 5.85002e-06
[remove_dup_value]: 1.511e-05
[partial_unused_args_eliminate]: 2.44001e-06
[environ_conv]: 5.57001e-06
[add_recomputation]: 5.223e-05
[cse_after_recomputation]: 2.326e-05, [1]
    [Cycle 1]: 1.624e-05, [1]
        [cse]: 9.71998e-06
[auto_monad_reorder]: 1.601e-05
[get_jit_bprop_graph]: 2.14e-06
[rewriter_after_jit_bprop_graph]: 5.87999e-06
[opt_after_jit_grad]: 0.00056116
[symbol_engine_optimizer]: 8.364e-05, [1]
    [Cycle 1]: 7.482e-05, [6]
        [build]: 4.74e-06
        [elim_shapecalc]: 8.65001e-06
        [elim_not_effective]: 1.518e-05
        [opt_reshape]: 6.86999e-06
        [fold_const_symbol]: 9.34e-06
        [renormalize]: 4.39992e-07
[validate]: 4.466e-05
Sums
    bootstrap                            :     0.000335s :  3.47%
    type_inference                       :     0.004874s : 50.51%
    event_method                         :     0.000019s :  0.20%
    auto_monad                           :     0.000070s :  0.72%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000045s :  0.47%
    rewriter_before_opt_a                :     0.000077s :  0.80%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000073s :  0.75%
    jit_opt_a.loop_unroll                :     0.000037s :  0.38%
    jit_opt_a.a_1                        :     0.000713s :  7.39%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.37%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.18%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000877s :  9.09%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.21%
    jit_opt_a.cse                        :     0.000038s :  0.39%
    jit_opt_a.replace_applicator         :     0.000021s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.14%
    rewriter_after_opt_a                 :     0.000036s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000702s :  7.28%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.17%
    cconv                                :     0.000030s :  0.31%
    loop_unroll                          :     0.000526s :  5.45%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000052s :  0.54%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000561s :  5.82%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000045s :  0.46%


Time group info:
------[substitution.]   0.000254    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000001s :      2: substitution.fold_const_symbol
 2.30% :     0.000006s :      3: substitution.graph_param_transform
85.68% :     0.000218s :      4: substitution.inline
 1.58% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.83% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.75% :     0.000004s :      2: substitution.replace_old_param
 5.48% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004802     2
78.89% :     0.003788s :      1: type_inference.infer
21.11% :     0.001014s :      1: type_inference.specialize
------[replace.]   0.000078     6
67.96% :     0.000053s :      4: replace.inline
32.04% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000228     6
94.52% :     0.000215s :      4: match.inline
 5.48% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   691
 1.23% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.79% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.99% :     0.000003s :     10: predicate.addn_zero_filter
 2.31% :     0.000003s :     10: predicate.arithmetic_simplify
 1.03% :     0.000001s :     10: predicate.cast_eliminate
 0.32% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.91% :     0.000001s :     10: predicate.compare_switch_simplify
 2.31% :     0.000003s :     10: predicate.depend_value_elim
 0.97% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.69% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.98% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000001s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.01% :     0.000001s :     10: predicate.environ_get_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000001s :      3: predicate.graph_param_transform
 5.14% :     0.000007s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.52% :     0.000002s :      6: predicate.less_batch_normalization
 1.30% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.57% :     0.000002s :     15: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.55% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.26% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 0.95% :     0.000001s :     10: predicate.minmaximum_grad
 2.00% :     0.000003s :      3: predicate.mutable_eliminate
 0.62% :     0.000001s :      3: predicate.opt_reshape
 1.94% :     0.000003s :     15: predicate.partial_eliminate
 1.09% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.68% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.41% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000002s :     10: predicate.reshape_eliminate
 1.01% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.08% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.14% :     0.000002s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.53% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.86% :     0.000004s :     16: predicate.switch_defer_inline
 1.89% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.97% :     0.000010s :     53: predicate.switch_simplify
 1.03% :     0.000001s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 5.12% :     0.000007s :     18: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.43% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.86% :     0.000003s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.12% :     0.000002s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000711    12
46.84% :     0.000333s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.49% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.67% :     0.000325s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015840    72
 0.35% :     0.000055s :      1: add_recomputation
 0.46% :     0.000073s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 2.20% :     0.000349s :      1: bootstrap
 0.21% :     0.000033s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
37.62% :     0.005959s :      1: jit_opt_a
 1.14% :     0.000181s :      1: jit_opt_after_cconv
 0.40% :     0.000063s :      1: jit_opt_b
 3.41% :     0.000539s :      1: loop_unroll
 4.52% :     0.000715s :      1: mutable_eliminate
 5.98% :     0.000947s :     26: opt.transform.jit_opt_a
 0.36% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000030s :      4: opt.transform.jit_opt_b
 0.10% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.62% :     0.000574s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000048s :      1: py_interpret_to_execute
 0.10% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.95% :     0.000467s :      1: renormalize.infer
 2.52% :     0.000400s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000039s :      1: rewriter_after_opt_a
 0.50% :     0.000080s :      1: rewriter_before_opt_a
 0.54% :     0.000086s :      1: symbol_engine_optimizer
30.87% :     0.004890s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.741.441 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dcafca0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.742.060 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0230483, [33]
[bootstrap]: 0.00033412
[type_inference]: 0.00635902
[event_method]: 3.574e-05
[auto_monad]: 8.582e-05
[graph_reusing]: 6.14999e-06
[pre_auto_parallel]: 2.84001e-06
[py_interpret_to_execute]: 4.448e-05
[rewriter_before_opt_a]: 8.497e-05
[expand_dump_flag]: 3.51999e-06
[jit_opt_a]: 0.0063, [2]
    [Cycle 1]: 0.00190139, [27]
        [switch_simplify]: 6.385e-05
        [loop_unroll]: 2.924e-05
        [a_1]: 0.00060086
        [with_stream_mark]: 2.623e-05
        [recompute_prepare]: 8e-06
        [updatestate_depend_eliminate]: 4.38001e-06
        [updatestate_assign_eliminate]: 3.08998e-06
        [updatestate_loads_eliminate]: 2.86e-06
        [parameter_eliminate]: 1.95001e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 6.01e-06
        [meta_shard_fg_expand]: 1.90001e-06
        [get_grad_eliminate_]: 4.98001e-06
        [merge_forward]: 4.14002e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.544e-05
        [j_node_and_user_rematch]: 9.01998e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 9.52001e-06
        [inline_without_move]: 5.25999e-06
        [renormalize]: 0.00081579
        [add_forward_monad_depend]: 6.99001e-06
        [auto_monad_grad]: 3.38999e-06
        [auto_monad_eliminator]: 1.756e-05
        [cse]: 3.42e-05
        [replace_applicator]: 1.43e-05
    [Cycle 2]: 0.00040275, [27]
        [switch_simplify]: 5.71003e-06
        [loop_unroll]: 5.05001e-06
        [a_1]: 8.81e-05
        [with_stream_mark]: 1.245e-05
        [recompute_prepare]: 5.13002e-06
        [updatestate_depend_eliminate]: 2.71999e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [parameter_eliminate]: 1.30999e-06
        [specialize_transform]: 5.39998e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 4.80001e-06
        [meta_shard_fg_expand]: 1.00999e-06
        [get_grad_eliminate_]: 4.34002e-06
        [merge_forward]: 2.19001e-06
        [cell_reuse_recompute_pass]: 2.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.397e-05
        [j_node_and_user_rematch]: 8.35001e-06
        [meta_fg_expand]: 1.63002e-06
        [replace_old_param]: 7.97e-06
        [inline_without_move]: 4.61002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.35999e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 5.62001e-06
        [cse]: 7.8e-05
        [replace_applicator]: 1.017e-05
[py_interpret_to_execute_after_opt_a]: 1.452e-05
[rewriter_after_opt_a]: 4.409e-05
[convert_after_rewriter]: 7.08e-06
[order_py_execute_after_rewriter]: 4.99e-06
[mutable_eliminate]: 0.0008176
[jit_opt_b]: 5.213e-05, [1]
    [Cycle 1]: 4.296e-05, [2]
        [frontend_op_eliminate]: 1.576e-05
        [inline_after_opt_a]: 1.565e-05
[cconv]: 3.335e-05
[loop_unroll]: 0.00047491
[jit_opt_after_cconv]: 0.00016551, [1]
    [Cycle 1]: 0.00015828, [11]
        [c_1]: 2.129e-05
        [parameter_eliminate]: 4.80001e-06
        [updatestate_depend_eliminate]: 1.052e-05
        [updatestate_assign_eliminate]: 3.18e-06
        [updatestate_loads_eliminate]: 2.63003e-06
        [cse]: 3.217e-05
        [call_graph_tuple_transform]: 2.158e-05
        [tuple_list_get_item_eliminator]: 5.64e-06
        [none_parameter_eliminate]: 1.43002e-06
        [renormalize]: 1.20001e-06
        [switch_simplify]: 5.09e-06
[remove_dup_value]: 1.465e-05
[partial_unused_args_eliminate]: 2.34001e-06
[environ_conv]: 4.93001e-06
[add_recomputation]: 5.86e-05
[cse_after_recomputation]: 2.104e-05, [1]
    [Cycle 1]: 1.499e-05, [1]
        [cse]: 9.14e-06
[auto_monad_reorder]: 1.539e-05
[get_jit_bprop_graph]: 2.16e-06
[rewriter_after_jit_bprop_graph]: 9.08002e-06
[opt_after_jit_grad]: 0.00052199
[symbol_engine_optimizer]: 7.807e-05, [1]
    [Cycle 1]: 7.114e-05, [6]
        [build]: 5.99e-06
        [elim_shapecalc]: 7.83999e-06
        [elim_not_effective]: 1.385e-05
        [opt_reshape]: 5.79e-06
        [fold_const_symbol]: 8.17e-06
        [renormalize]: 9.5999e-07
[validate]: 4.557e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.00713344
[execute]: 9.66e-06
Sums
    bootstrap                            :     0.000334s :  1.82%
    type_inference                       :     0.006359s : 34.65%
    event_method                         :     0.000036s :  0.19%
    auto_monad                           :     0.000086s :  0.47%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000044s :  0.24%
    rewriter_before_opt_a                :     0.000085s :  0.46%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000070s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.19%
    jit_opt_a.a_1                        :     0.000689s :  3.75%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000816s :  4.45%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.13%
    jit_opt_a.cse                        :     0.000112s :  0.61%
    jit_opt_a.replace_applicator         :     0.000024s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.08%
    rewriter_after_opt_a                 :     0.000044s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000818s :  4.46%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000033s :  0.18%
    loop_unroll                          :     0.000475s :  2.59%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000032s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000059s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000522s :  2.84%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000046s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007133s : 38.87%
    execute                              :     0.000010s :  0.05%


TotalTime = 0.0114873, [30]
[bootstrap]: 0.00023732
[type_inference]: 0.00390358
[event_method]: 1.705e-05
[auto_monad]: 6.493e-05
[graph_reusing]: 5.74e-06
[pre_auto_parallel]: 2.30002e-06
[py_interpret_to_execute]: 3.496e-05
[rewriter_before_opt_a]: 7.154e-05
[expand_dump_flag]: 3.31001e-06
[jit_opt_a]: 0.00513809, [2]
    [Cycle 1]: 0.00153927, [27]
        [switch_simplify]: 5.788e-05
        [loop_unroll]: 3.25e-05
        [a_1]: 0.00055229
        [with_stream_mark]: 1.692e-05
        [recompute_prepare]: 7.14001e-06
        [updatestate_depend_eliminate]: 3.42002e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.58003e-06
        [parameter_eliminate]: 1.88997e-06
        [specialize_transform]: 6.14001e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 6.43e-06
        [meta_shard_fg_expand]: 2.69001e-06
        [get_grad_eliminate_]: 5.86998e-06
        [merge_forward]: 4.03999e-06
        [cell_reuse_recompute_pass]: 9.5999e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.483e-05
        [j_node_and_user_rematch]: 1.073e-05
        [meta_fg_expand]: 2.22001e-06
        [replace_old_param]: 9.54999e-06
        [inline_without_move]: 5.96998e-06
        [renormalize]: 0.00055817
        [add_forward_monad_depend]: 4.92e-06
        [auto_monad_grad]: 1.82001e-06
        [auto_monad_eliminator]: 1.349e-05
        [cse]: 2.652e-05
        [replace_applicator]: 1.109e-05
    [Cycle 2]: 0.00032387, [27]
        [switch_simplify]: 5.73002e-06
        [loop_unroll]: 5.42001e-06
        [a_1]: 9.164e-05
        [with_stream_mark]: 8.08001e-06
        [recompute_prepare]: 5.19003e-06
        [updatestate_depend_eliminate]: 2.58998e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 4.88001e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.53999e-06
        [merge_forward]: 2.52001e-06
        [cell_reuse_recompute_pass]: 1.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.164e-05
        [j_node_and_user_rematch]: 7.66999e-06
        [meta_fg_expand]: 1.54998e-06
        [replace_old_param]: 6.98998e-06
        [inline_without_move]: 4.75001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.27999e-06
        [cse]: 9.84999e-06
        [replace_applicator]: 5.59e-06
[py_interpret_to_execute_after_opt_a]: 8.33999e-06
[rewriter_after_opt_a]: 3.049e-05
[convert_after_rewriter]: 7.26001e-06
[order_py_execute_after_rewriter]: 5.20001e-06
[mutable_eliminate]: 0.00049585
[jit_opt_b]: 4.996e-05, [1]
    [Cycle 1]: 4.28e-05, [2]
        [frontend_op_eliminate]: 1.641e-05
        [inline_after_opt_a]: 1.466e-05
[cconv]: 2.444e-05
[loop_unroll]: 0.00041451
[jit_opt_after_cconv]: 0.00013923, [1]
    [Cycle 1]: 0.00013315, [11]
        [c_1]: 2.062e-05
        [parameter_eliminate]: 2.34999e-06
        [updatestate_depend_eliminate]: 5.37999e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.733e-05
        [call_graph_tuple_transform]: 1.802e-05
        [tuple_list_get_item_eliminator]: 5.60001e-06
        [none_parameter_eliminate]: 1.85001e-06
        [renormalize]: 8.40024e-07
        [switch_simplify]: 5.61e-06
[remove_dup_value]: 1.402e-05
[partial_unused_args_eliminate]: 2.48e-06
[environ_conv]: 5.20001e-06
[add_recomputation]: 4.776e-05
[cse_after_recomputation]: 2.048e-05, [1]
    [Cycle 1]: 1.498e-05, [1]
        [cse]: 8.86002e-06
[auto_monad_reorder]: 1.49e-05
[get_jit_bprop_graph]: 1.34e-06
[rewriter_after_jit_bprop_graph]: 4.95999e-06
[opt_after_jit_grad]: 0.00045647
[symbol_engine_optimizer]: 7.506e-05, [1]
    [Cycle 1]: 6.836e-05, [6]
        [build]: 3.97e-06
        [elim_shapecalc]: 7.75e-06
        [elim_not_effective]: 1.309e-05
        [opt_reshape]: 6.02999e-06
        [fold_const_symbol]: 8.87999e-06
        [renormalize]: 4.20026e-07
[validate]: 3.27e-05
Sums
    bootstrap                            :     0.000237s :  3.11%
    type_inference                       :     0.003904s : 51.09%
    event_method                         :     0.000017s :  0.22%
    auto_monad                           :     0.000065s :  0.85%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.46%
    rewriter_before_opt_a                :     0.000072s :  0.94%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000064s :  0.83%
    jit_opt_a.loop_unroll                :     0.000038s :  0.50%
    jit_opt_a.a_1                        :     0.000644s :  8.43%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.14%
    jit_opt_a.merge_forward              :     0.000007s :  0.09%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.24%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.22%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000558s :  7.31%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.25%
    jit_opt_a.cse                        :     0.000036s :  0.48%
    jit_opt_a.replace_applicator         :     0.000017s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000030s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.10%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000496s :  6.49%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000024s :  0.32%
    loop_unroll                          :     0.000415s :  5.43%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000014s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000048s :  0.63%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000015s :  0.20%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000456s :  5.97%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000033s :  0.43%


Time group info:
------[substitution.]   0.000268    23
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 2.10% :     0.000006s :      3: substitution.graph_param_transform
86.57% :     0.000232s :      4: substitution.inline
 1.38% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.65% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.16% :     0.000006s :      2: substitution.replace_old_param
 4.89% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005209     2
82.97% :     0.004322s :      1: type_inference.infer
17.03% :     0.000887s :      1: type_inference.specialize
------[replace.]   0.000071     6
68.22% :     0.000049s :      4: replace.inline
31.78% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000241     6
95.03% :     0.000229s :      4: match.inline
 4.97% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.68% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 1.94% :     0.000002s :     10: predicate.arithmetic_simplify
 1.37% :     0.000002s :     10: predicate.cast_eliminate
 0.70% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000001s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.45% :     0.000001s :      3: predicate.graph_param_transform
 7.53% :     0.000009s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.02% :     0.000001s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.65% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.05% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.18% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 2.03% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000002s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.56% :     0.000001s :      3: predicate.reset_defer_inline
 1.24% :     0.000001s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.25% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.10% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.07% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.24% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.06% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.12% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.31% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000660    12
50.01% :     0.000330s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.73% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.27% :     0.000272s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024883    76
 0.25% :     0.000061s :      1: add_recomputation
 0.36% :     0.000090s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.44% :     0.000358s :      1: bootstrap
 0.14% :     0.000036s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.17% :     0.000042s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.33% :     0.006304s :      1: jit_opt_a
 0.68% :     0.000169s :      1: jit_opt_after_cconv
 0.22% :     0.000055s :      1: jit_opt_b
 1.95% :     0.000485s :      1: loop_unroll
 3.33% :     0.000829s :      1: mutable_eliminate
 3.67% :     0.000914s :     26: opt.transform.jit_opt_a
 0.20% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.15% :     0.000534s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000048s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.88% :     0.000468s :      1: renormalize.infer
 1.36% :     0.000339s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000047s :      1: rewriter_after_opt_a
 0.36% :     0.000089s :      1: rewriter_before_opt_a
 0.32% :     0.000081s :      1: symbol_engine_optimizer
28.75% :     0.007154s :      1: task_emit
25.67% :     0.006387s :      1: type_inference
 0.29% :     0.000073s :      1: validate


Time group info:
------[substitution.]   0.000192    23
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.59% :     0.000005s :      3: substitution.graph_param_transform
82.52% :     0.000158s :      4: substitution.inline
 2.12% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.55% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.67% :     0.000003s :      2: substitution.replace_old_param
 6.81% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003848     2
79.50% :     0.003059s :      1: type_inference.infer
20.50% :     0.000789s :      1: type_inference.specialize
------[replace.]   0.000070     6
65.55% :     0.000046s :      4: replace.inline
34.45% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000167     6
92.95% :     0.000156s :      4: match.inline
 7.05% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.48% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 2.36% :     0.000003s :     10: predicate.arithmetic_simplify
 1.14% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.47% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.09% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.45% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.54% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.42% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.45% :     0.000007s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.29% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.76% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 1.40% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.39% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.60% :     0.000002s :     10: predicate.same_eliminate
 0.60% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.16% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.57% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.71% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.57% :     0.000009s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.93% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.96% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.87% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.91% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000531    12
48.62% :     0.000258s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.19% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.20% :     0.000245s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012969    72
 0.39% :     0.000051s :      1: add_recomputation
 0.53% :     0.000068s :      1: auto_monad
 0.14% :     0.000018s :      1: auto_monad_reorder
 1.92% :     0.000249s :      1: bootstrap
 0.21% :     0.000027s :      1: cconv
 0.08% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
39.64% :     0.005141s :      1: jit_opt_a
 1.09% :     0.000142s :      1: jit_opt_after_cconv
 0.41% :     0.000053s :      1: jit_opt_b
 3.26% :     0.000423s :      1: loop_unroll
 3.89% :     0.000504s :      1: mutable_eliminate
 6.65% :     0.000862s :     26: opt.transform.jit_opt_a
 0.36% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.58% :     0.000465s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.04% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.29% :     0.000038s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000016s :      1: remove_dup_value
 2.12% :     0.000275s :      1: renormalize.infer
 2.13% :     0.000276s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000033s :      1: rewriter_after_opt_a
 0.57% :     0.000074s :      1: rewriter_before_opt_a
 0.60% :     0.000078s :      1: symbol_engine_optimizer
30.19% :     0.003915s :      1: type_inference


TotalTime = 0.0292123, [33]
[bootstrap]: 0.00020925
[type_inference]: 0.00958138
[event_method]: 4.09e-05
[auto_monad]: 7.575e-05
[graph_reusing]: 5.95002e-06
[pre_auto_parallel]: 2.16e-06
[py_interpret_to_execute]: 2.299e-05
[rewriter_before_opt_a]: 8.47e-05
[expand_dump_flag]: 2.99001e-06
[jit_opt_a]: 0.0103762, [3]
    [Cycle 1]: 0.00513142, [27]
        [switch_simplify]: 5.756e-05
        [loop_unroll]: 3.858e-05
        [a_1]: 0.00078686
        [with_stream_mark]: 2.112e-05
        [recompute_prepare]: 1.781e-05
        [updatestate_depend_eliminate]: 7.58999e-06
        [updatestate_assign_eliminate]: 5.95002e-06
        [updatestate_loads_eliminate]: 5.52999e-06
        [parameter_eliminate]: 1.77001e-06
        [specialize_transform]: 1.284e-05
        [updatestate_useless_node_eliminater]: 1.211e-05
        [accelerated_algorithm]: 1.241e-05
        [meta_shard_fg_expand]: 2.63e-06
        [get_grad_eliminate_]: 1.21e-05
        [merge_forward]: 7.28e-06
        [cell_reuse_recompute_pass]: 7.7e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.286e-05
        [j_node_and_user_rematch]: 2.187e-05
        [meta_fg_expand]: 0.00137978
        [replace_old_param]: 5.2e-05
        [inline_without_move]: 5.181e-05
        [renormalize]: 0.00215151
        [add_forward_monad_depend]: 8.51002e-06
        [auto_monad_grad]: 5.40001e-06
        [auto_monad_eliminator]: 4.562e-05
        [cse]: 0.00013683
        [replace_applicator]: 6.308e-05
    [Cycle 2]: 0.00189511, [27]
        [switch_simplify]: 3.523e-05
        [loop_unroll]: 3.431e-05
        [a_1]: 0.0010502
        [with_stream_mark]: 1.235e-05
        [recompute_prepare]: 7.87e-06
        [updatestate_depend_eliminate]: 3.63999e-06
        [updatestate_assign_eliminate]: 3.03998e-06
        [updatestate_loads_eliminate]: 2.46998e-06
        [parameter_eliminate]: 1.12999e-06
        [specialize_transform]: 6.48e-06
        [updatestate_useless_node_eliminater]: 6.02001e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 1.66e-06
        [get_grad_eliminate_]: 6.06e-06
        [merge_forward]: 3.06999e-06
        [cell_reuse_recompute_pass]: 8.99978e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.292e-05
        [j_node_and_user_rematch]: 9.53002e-06
        [meta_fg_expand]: 3.252e-05
        [replace_old_param]: 9.08002e-06
        [inline_without_move]: 5.79999e-06
        [renormalize]: 0.00044983
        [add_forward_monad_depend]: 4.17e-06
        [auto_monad_grad]: 1.05001e-06
        [auto_monad_eliminator]: 1.23e-05
        [cse]: 1.963e-05
        [replace_applicator]: 1.22e-05
    [Cycle 3]: 0.00036318, [27]
        [switch_simplify]: 6.89999e-06
        [loop_unroll]: 5.79999e-06
        [a_1]: 0.00011627
        [with_stream_mark]: 8.71997e-06
        [recompute_prepare]: 6.00002e-06
        [updatestate_depend_eliminate]: 3.26999e-06
        [updatestate_assign_eliminate]: 2.74999e-06
        [updatestate_loads_eliminate]: 2.53998e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 6.33e-06
        [updatestate_useless_node_eliminater]: 5.75001e-06
        [accelerated_algorithm]: 5.85002e-06
        [meta_shard_fg_expand]: 1.32999e-06
        [get_grad_eliminate_]: 5.81e-06
        [merge_forward]: 3.01999e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.361e-05
        [j_node_and_user_rematch]: 9.62999e-06
        [meta_fg_expand]: 2.12001e-06
        [replace_old_param]: 7.84002e-06
        [inline_without_move]: 5.61e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.47999e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 7.03e-06
        [cse]: 1.534e-05
        [replace_applicator]: 6.66e-06
[py_interpret_to_execute_after_opt_a]: 1.216e-05
[rewriter_after_opt_a]: 3.417e-05
[convert_after_rewriter]: 7.01999e-06
[order_py_execute_after_rewriter]: 5.42001e-06
[mutable_eliminate]: 0.00051795
[jit_opt_b]: 5.372e-05, [1]
    [Cycle 1]: 4.687e-05, [2]
        [frontend_op_eliminate]: 1.869e-05
        [inline_after_opt_a]: 1.749e-05
[cconv]: 2.035e-05
[loop_unroll]: 0.00041668
[jit_opt_after_cconv]: 0.00015701, [1]
    [Cycle 1]: 0.00015086, [11]
        [c_1]: 2.564e-05
        [parameter_eliminate]: 2.22999e-06
        [updatestate_depend_eliminate]: 5.76e-06
        [updatestate_assign_eliminate]: 2.89999e-06
        [updatestate_loads_eliminate]: 2.61999e-06
        [cse]: 2.237e-05
        [call_graph_tuple_transform]: 1.959e-05
        [tuple_list_get_item_eliminator]: 6.49001e-06
        [none_parameter_eliminate]: 1.40999e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 6.51e-06
[remove_dup_value]: 1.358e-05
[partial_unused_args_eliminate]: 1.79998e-06
[environ_conv]: 5.19998e-06
[add_recomputation]: 4.481e-05
[cse_after_recomputation]: 2.652e-05, [1]
    [Cycle 1]: 2.108e-05, [1]
        [cse]: 1.528e-05
[auto_monad_reorder]: 1.574e-05
[get_jit_bprop_graph]: 1.71998e-06
[rewriter_after_jit_bprop_graph]: 4.80001e-06
[opt_after_jit_grad]: 0.00045542
[symbol_engine_optimizer]: 7.923e-05, [1]
    [Cycle 1]: 7.314e-05, [6]
        [build]: 3.99997e-06
        [elim_shapecalc]: 9.29998e-06
        [elim_not_effective]: 1.468e-05
        [opt_reshape]: 7.03998e-06
        [fold_const_symbol]: 1.055e-05
        [renormalize]: 5.69999e-07
[validate]: 3.149e-05
[backend_pass]: 7.80012e-07
[task_emit]: 0.00671523
[execute]: 7.68999e-06
Sums
    bootstrap                            :     0.000209s :  0.82%
    type_inference                       :     0.009581s : 37.61%
    event_method                         :     0.000041s :  0.16%
    auto_monad                           :     0.000076s :  0.30%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000023s :  0.09%
    rewriter_before_opt_a                :     0.000085s :  0.33%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000100s :  0.39%
    jit_opt_a.loop_unroll                :     0.000079s :  0.31%
    jit_opt_a.a_1                        :     0.001953s :  7.67%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.09%
    jit_opt_a.merge_forward              :     0.000013s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000049s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000041s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001414s :  5.55%
    jit_opt_a.replace_old_param          :     0.000069s :  0.27%
    jit_opt_a.inline_without_move        :     0.000063s :  0.25%
    jit_opt_a.renormalize                :     0.002601s : 10.21%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000065s :  0.25%
    jit_opt_a.cse                        :     0.000172s :  0.67%
    jit_opt_a.replace_applicator         :     0.000082s :  0.32%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.05%
    rewriter_after_opt_a                 :     0.000034s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000518s :  2.03%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.07%
    cconv                                :     0.000020s :  0.08%
    loop_unroll                          :     0.000417s :  1.64%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000014s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000045s :  0.18%
    cse_after_recomputation.cse          :     0.000015s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000455s :  1.79%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000031s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006715s : 26.36%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000465   110
 0.45% :     0.000002s :      3: substitution.elim_not_effective
 0.36% :     0.000002s :      3: substitution.fold_const_symbol
 1.16% :     0.000005s :      4: substitution.graph_param_transform
65.20% :     0.000303s :     12: substitution.inline
 3.35% :     0.000016s :      2: substitution.inline_without_move
 1.72% :     0.000008s :     14: substitution.j_node_and_user_rematch
 2.07% :     0.000010s :      7: substitution.minmaximum_grad
 1.32% :     0.000006s :      5: substitution.partial_eliminate
 2.09% :     0.000010s :     14: substitution.remove_not_recompute_node
 5.34% :     0.000025s :     10: substitution.replace_applicator
 1.44% :     0.000007s :      7: substitution.replace_old_param
 0.50% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.36% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.83% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 7.81% :     0.000036s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009520     2
89.81% :     0.008549s :      1: type_inference.infer
10.19% :     0.000970s :      1: type_inference.specialize
------[replace.]   0.000139    19
62.78% :     0.000087s :     12: replace.inline
37.22% :     0.000052s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000313    19
94.39% :     0.000296s :     12: match.inline
 5.61% :     0.000018s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000366  2568
 1.61% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.46% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.43% :     0.000005s :     43: predicate.addn_check_dump
 1.69% :     0.000006s :     43: predicate.addn_zero_filter
 2.13% :     0.000008s :     43: predicate.arithmetic_simplify
 1.55% :     0.000006s :     43: predicate.cast_eliminate
 0.16% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.43% :     0.000005s :     43: predicate.compare_switch_simplify
 1.58% :     0.000006s :     43: predicate.depend_value_elim
 1.45% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.54% :     0.000006s :     43: predicate.dict_get_item_eliminator
 1.49% :     0.000005s :     43: predicate.dict_set_item_eliminator
 0.33% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.14% :     0.000001s :      4: predicate.elim_not_effective
 0.21% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.46% :     0.000005s :     43: predicate.environ_add_const_eliminate
 1.47% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.44% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.53% :     0.000006s :     43: predicate.environ_get_eliminate
 1.43% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.84% :     0.000003s :     19: predicate.get_grad_eliminate
 0.11% :     0.000000s :      4: predicate.graph_param_transform
 4.33% :     0.000016s :     70: predicate.inline
 1.78% :     0.000007s :     40: predicate.inline_without_move
 0.37% :     0.000001s :     19: predicate.j_node_and_user_rematch
 1.07% :     0.000004s :     19: predicate.less_batch_normalization
 1.87% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.94% :     0.000007s :     54: predicate.load_eliminater
 0.48% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.26% :     0.000012s :     85: predicate.loop_unroll_before_grad
 1.85% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.45% :     0.000005s :     43: predicate.merge_addn
 1.45% :     0.000005s :     43: predicate.minmaximum_grad
 0.54% :     0.000002s :      4: predicate.mutable_eliminate
 0.21% :     0.000001s :      4: predicate.opt_reshape
 2.34% :     0.000009s :     54: predicate.partial_eliminate
 1.52% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.93% :     0.000007s :     43: predicate.reduce_eliminate
 1.77% :     0.000006s :     50: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.57% :     0.000009s :     96: predicate.replace_applicator
 0.95% :     0.000003s :     40: predicate.replace_old_param
 0.11% :     0.000000s :      4: predicate.reset_defer_inline
 1.60% :     0.000006s :     43: predicate.reshape_eliminate
 1.59% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.27% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.58% :     0.000006s :     43: predicate.same_eliminate
 0.52% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.35% :     0.000001s :      8: predicate.special_op_eliminate
 0.90% :     0.000003s :     19: predicate.specialize_transform
 1.70% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.54% :     0.000006s :     43: predicate.stack_unstack_eliminate
 0.18% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.75% :     0.000010s :     62: predicate.switch_defer_inline
 2.52% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.14% :     0.000022s :    151: predicate.switch_simplify
 1.51% :     0.000006s :     43: predicate.tile_eliminate
 1.55% :     0.000006s :     43: predicate.transpose_eliminate
 1.84% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.65% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.16% :     0.000012s :     58: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000007s :     43: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000007s :     50: predicate.tuple_to_list_eliminator_
 1.92% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 3.05% :     0.000011s :     73: predicate.updatestate_useless_node_eliminater
 1.86% :     0.000007s :     43: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.27% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001324    29
59.88% :     0.000793s :     13: func_graph_cloner_run.FuncGraphClonerGraph
40.12% :     0.000531s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.034470    91
 0.14% :     0.000048s :      1: add_recomputation
 0.23% :     0.000079s :      1: auto_monad
 0.05% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.65% :     0.000224s :      1: bootstrap
 0.07% :     0.000023s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000029s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.13% :     0.000045s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
30.11% :     0.010379s :      1: jit_opt_a
 0.46% :     0.000160s :      1: jit_opt_after_cconv
 0.16% :     0.000056s :      1: jit_opt_b
 1.23% :     0.000424s :      1: loop_unroll
 1.53% :     0.000526s :      1: mutable_eliminate
 7.33% :     0.002527s :     39: opt.transform.jit_opt_a
 0.16% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000030s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.34% :     0.000463s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000026s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000016s :      1: remove_dup_value
 3.88% :     0.001338s :      2: renormalize.infer
 3.63% :     0.001251s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000037s :      1: rewriter_after_opt_a
 0.25% :     0.000087s :      1: rewriter_before_opt_a
 0.24% :     0.000082s :      1: symbol_engine_optimizer
19.52% :     0.006729s :      1: task_emit
27.83% :     0.009592s :      1: type_inference
 0.15% :     0.000051s :      1: validate


TotalTime = 0.0425202, [33]
[bootstrap]: 0.00032541
[type_inference]: 0.0134796
[event_method]: 1.858e-05
[auto_monad]: 6.32e-05
[graph_reusing]: 6.23e-06
[pre_auto_parallel]: 2.71999e-06
[py_interpret_to_execute]: 3.911e-05
[rewriter_before_opt_a]: 7.135e-05
[expand_dump_flag]: 3.03e-06
[jit_opt_a]: 0.0099878, [2]
    [Cycle 1]: 0.00205557, [27]
        [switch_simplify]: 7.034e-05
        [loop_unroll]: 2.916e-05
        [a_1]: 0.00062042
        [with_stream_mark]: 2.437e-05
        [recompute_prepare]: 7.72002e-06
        [updatestate_depend_eliminate]: 4.51002e-06
        [updatestate_assign_eliminate]: 2.77002e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 5.97999e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 5.66003e-06
        [meta_shard_fg_expand]: 2.07001e-06
        [get_grad_eliminate_]: 5.42999e-06
        [merge_forward]: 3.75e-06
        [cell_reuse_recompute_pass]: 1.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.374e-05
        [j_node_and_user_rematch]: 9.74999e-06
        [meta_fg_expand]: 2.21998e-06
        [replace_old_param]: 9.49e-06
        [inline_without_move]: 5.40001e-06
        [renormalize]: 0.00094965
        [add_forward_monad_depend]: 7.12002e-06
        [auto_monad_grad]: 2.76e-06
        [auto_monad_eliminator]: 1.728e-05
        [cse]: 3.073e-05
        [replace_applicator]: 1.55e-05
    [Cycle 2]: 0.00033358, [27]
        [switch_simplify]: 5.73002e-06
        [loop_unroll]: 4.82998e-06
        [a_1]: 9.283e-05
        [with_stream_mark]: 1.339e-05
        [recompute_prepare]: 5.15999e-06
        [updatestate_depend_eliminate]: 3.21001e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 4.95999e-06
        [updatestate_useless_node_eliminater]: 4.70999e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 4.89003e-06
        [merge_forward]: 2.79001e-06
        [cell_reuse_recompute_pass]: 2.20002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.272e-05
        [j_node_and_user_rematch]: 7.90998e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 8.13001e-06
        [inline_without_move]: 5.09e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 5.40001e-06
        [cse]: 1.126e-05
        [replace_applicator]: 5.10001e-06
[py_interpret_to_execute_after_opt_a]: 1.332e-05
[rewriter_after_opt_a]: 3.75e-05
[convert_after_rewriter]: 7.10002e-06
[order_py_execute_after_rewriter]: 4.55999e-06
[mutable_eliminate]: 0.00075612
[jit_opt_b]: 5.217e-05, [1]
    [Cycle 1]: 4.364e-05, [2]
        [frontend_op_eliminate]: 1.728e-05
        [inline_after_opt_a]: 1.524e-05
[cconv]: 3.285e-05
[loop_unroll]: 0.00043983
[jit_opt_after_cconv]: 0.00014725, [1]
    [Cycle 1]: 0.00014037, [11]
        [c_1]: 2.12e-05
        [parameter_eliminate]: 3.88999e-06
        [updatestate_depend_eliminate]: 7.31999e-06
        [updatestate_assign_eliminate]: 3.03998e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [cse]: 2.106e-05
        [call_graph_tuple_transform]: 2.053e-05
        [tuple_list_get_item_eliminator]: 5.14998e-06
        [none_parameter_eliminate]: 1.35999e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 5.04e-06
[remove_dup_value]: 1.536e-05
[partial_unused_args_eliminate]: 2.03002e-06
[environ_conv]: 5.39e-06
[add_recomputation]: 5.017e-05
[cse_after_recomputation]: 2.15e-05, [1]
    [Cycle 1]: 1.627e-05, [1]
        [cse]: 9.82999e-06
[auto_monad_reorder]: 1.597e-05
[get_jit_bprop_graph]: 2.06e-06
[rewriter_after_jit_bprop_graph]: 5.98998e-06
[opt_after_jit_grad]: 0.00045958
[symbol_engine_optimizer]: 7.319e-05, [1]
    [Cycle 1]: 6.655e-05, [6]
        [build]: 4.25e-06
        [elim_shapecalc]: 7.51999e-06
        [elim_not_effective]: 1.305e-05
        [opt_reshape]: 5.66e-06
        [fold_const_symbol]: 8.77e-06
        [renormalize]: 4.00003e-07
[validate]: 4.128e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.0160874
[execute]: 1.063e-05
Sums
    bootstrap                            :     0.000325s :  0.95%
    type_inference                       :     0.013480s : 39.36%
    event_method                         :     0.000019s :  0.05%
    auto_monad                           :     0.000063s :  0.18%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.11%
    rewriter_before_opt_a                :     0.000071s :  0.21%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000076s :  0.22%
    jit_opt_a.loop_unroll                :     0.000034s :  0.10%
    jit_opt_a.a_1                        :     0.000713s :  2.08%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.03%
    jit_opt_a.merge_forward              :     0.000007s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.08%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000018s :  0.05%
    jit_opt_a.inline_without_move        :     0.000010s :  0.03%
    jit_opt_a.renormalize                :     0.000950s :  2.77%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.07%
    jit_opt_a.cse                        :     0.000042s :  0.12%
    jit_opt_a.replace_applicator         :     0.000021s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.04%
    rewriter_after_opt_a                 :     0.000037s :  0.11%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000756s :  2.21%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.04%
    cconv                                :     0.000033s :  0.10%
    loop_unroll                          :     0.000440s :  1.28%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.01%
    remove_dup_value                     :     0.000015s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000050s :  0.15%
    cse_after_recomputation.cse          :     0.000010s :  0.03%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000460s :  1.34%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000041s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.016087s : 46.98%
    execute                              :     0.000011s :  0.03%


Time group info:
------[substitution.]   0.000270    23
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.07% :     0.000006s :      3: substitution.graph_param_transform
86.48% :     0.000233s :      4: substitution.inline
 1.39% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.70% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.18% :     0.000006s :      2: substitution.replace_old_param
 4.76% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.013395     2
92.94% :     0.012449s :      1: type_inference.infer
 7.06% :     0.000946s :      1: type_inference.specialize
------[replace.]   0.000079     6
69.04% :     0.000054s :      4: replace.inline
30.96% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000243     6
95.19% :     0.000231s :      4: match.inline
 4.81% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.19% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 2.16% :     0.000003s :     10: predicate.arithmetic_simplify
 1.19% :     0.000001s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.22% :     0.000002s :     10: predicate.compare_switch_simplify
 1.84% :     0.000002s :     10: predicate.depend_value_elim
 1.28% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.10% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.14% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 5.35% :     0.000007s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000002s :      6: predicate.less_batch_normalization
 1.78% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.67% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 2.30% :     0.000003s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.25% :     0.000003s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.16% :     0.000001s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.13% :     0.000003s :     16: predicate.switch_defer_inline
 2.30% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.88% :     0.000010s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.30% :     0.000002s :     10: predicate.transpose_eliminate
 1.76% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.67% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.42% :     0.000003s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000774    12
52.66% :     0.000408s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.85% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.49% :     0.000313s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.044504    76
 0.12% :     0.000054s :      1: add_recomputation
 0.15% :     0.000067s :      1: auto_monad
 0.04% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.78% :     0.000348s :      1: bootstrap
 0.08% :     0.000036s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.05% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.05% :     0.000023s :      1: event_method
 0.04% :     0.000016s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
22.45% :     0.009993s :      1: jit_opt_a
 0.34% :     0.000150s :      1: jit_opt_after_cconv
 0.12% :     0.000056s :      1: jit_opt_b
 1.01% :     0.000448s :      1: loop_unroll
 1.72% :     0.000766s :      1: mutable_eliminate
 2.12% :     0.000941s :     26: opt.transform.jit_opt_a
 0.11% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000026s :      4: opt.transform.jit_opt_b
 0.03% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.05% :     0.000467s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000042s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000018s :      1: remove_dup_value
 1.12% :     0.000498s :      1: renormalize.infer
 1.00% :     0.000443s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000041s :      1: rewriter_after_opt_a
 0.17% :     0.000074s :      1: rewriter_before_opt_a
 0.17% :     0.000076s :      1: symbol_engine_optimizer
36.20% :     0.016112s :      1: task_emit
30.33% :     0.013500s :      1: type_inference
 0.14% :     0.000063s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.777.523 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dcaff70> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.778.047 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0121452, [30]
[bootstrap]: 0.0002996
[type_inference]: 0.00439814
[event_method]: 1.575e-05
[auto_monad]: 4.355e-05
[graph_reusing]: 3.85e-06
[pre_auto_parallel]: 1.89e-06
[py_interpret_to_execute]: 3.295e-05
[rewriter_before_opt_a]: 6.232e-05
[expand_dump_flag]: 2.22001e-06
[jit_opt_a]: 0.0052457, [2]
    [Cycle 1]: 0.00164332, [27]
        [switch_simplify]: 5.507e-05
        [loop_unroll]: 3.057e-05
        [a_1]: 0.00053725
        [with_stream_mark]: 1.526e-05
        [recompute_prepare]: 7.53e-06
        [updatestate_depend_eliminate]: 3.06999e-06
        [updatestate_assign_eliminate]: 2.70002e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 1.88997e-06
        [specialize_transform]: 6.17999e-06
        [updatestate_useless_node_eliminater]: 5.62001e-06
        [accelerated_algorithm]: 5.86998e-06
        [meta_shard_fg_expand]: 1.50999e-06
        [get_grad_eliminate_]: 5.31002e-06
        [merge_forward]: 3.93999e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.325e-05
        [j_node_and_user_rematch]: 9.27999e-06
        [meta_fg_expand]: 1.82999e-06
        [replace_old_param]: 7.97003e-06
        [inline_without_move]: 5.80002e-06
        [renormalize]: 0.00069385
        [add_forward_monad_depend]: 5.01997e-06
        [auto_monad_grad]: 2.48e-06
        [auto_monad_eliminator]: 1.425e-05
        [cse]: 2.636e-05
        [replace_applicator]: 1.149e-05
    [Cycle 2]: 0.0003251, [27]
        [switch_simplify]: 5.59e-06
        [loop_unroll]: 4.95999e-06
        [a_1]: 9.288e-05
        [with_stream_mark]: 8.90001e-06
        [recompute_prepare]: 5.05999e-06
        [updatestate_depend_eliminate]: 2.48998e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 9.99979e-07
        [specialize_transform]: 4.74e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.32999e-06
        [meta_shard_fg_expand]: 1.01997e-06
        [get_grad_eliminate_]: 4.42998e-06
        [merge_forward]: 2.40002e-06
        [cell_reuse_recompute_pass]: 1.50999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.26e-05
        [j_node_and_user_rematch]: 8.27998e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 7.27002e-06
        [inline_without_move]: 4.77998e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.08001e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 5.27001e-06
        [cse]: 1.058e-05
        [replace_applicator]: 5.42999e-06
[py_interpret_to_execute_after_opt_a]: 1.065e-05
[rewriter_after_opt_a]: 3.109e-05
[convert_after_rewriter]: 7.01999e-06
[order_py_execute_after_rewriter]: 7.58999e-06
[mutable_eliminate]: 0.00053146
[jit_opt_b]: 5.02e-05, [1]
    [Cycle 1]: 4.369e-05, [2]
        [frontend_op_eliminate]: 1.713e-05
        [inline_after_opt_a]: 1.495e-05
[cconv]: 2.334e-05
[loop_unroll]: 0.00041779
[jit_opt_after_cconv]: 0.00013675, [1]
    [Cycle 1]: 0.00013011, [11]
        [c_1]: 2.245e-05
        [parameter_eliminate]: 2.24001e-06
        [updatestate_depend_eliminate]: 5.42001e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.65e-05
        [call_graph_tuple_transform]: 1.655e-05
        [tuple_list_get_item_eliminator]: 5.61e-06
        [none_parameter_eliminate]: 1.06997e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 5.44e-06
[remove_dup_value]: 7.38e-06
[partial_unused_args_eliminate]: 1.75001e-06
[environ_conv]: 4.17e-06
[add_recomputation]: 4.669e-05
[cse_after_recomputation]: 2.065e-05, [1]
    [Cycle 1]: 1.499e-05, [1]
        [cse]: 9.42001e-06
[auto_monad_reorder]: 1.263e-05
[get_jit_bprop_graph]: 2.04999e-06
[rewriter_after_jit_bprop_graph]: 5.15999e-06
[opt_after_jit_grad]: 0.00045229
[symbol_engine_optimizer]: 7.389e-05, [1]
    [Cycle 1]: 6.695e-05, [6]
        [build]: 3.41001e-06
        [elim_shapecalc]: 8.38999e-06
        [elim_not_effective]: 1.261e-05
        [opt_reshape]: 5.96e-06
        [fold_const_symbol]: 8.76002e-06
        [renormalize]: 5.00004e-07
[validate]: 3.158e-05
Sums
    bootstrap                            :     0.000300s :  3.61%
    type_inference                       :     0.004398s : 53.00%
    event_method                         :     0.000016s :  0.19%
    auto_monad                           :     0.000044s :  0.52%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.40%
    rewriter_before_opt_a                :     0.000062s :  0.75%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000061s :  0.73%
    jit_opt_a.loop_unroll                :     0.000036s :  0.43%
    jit_opt_a.a_1                        :     0.000630s :  7.59%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.18%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000694s :  8.36%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.24%
    jit_opt_a.cse                        :     0.000037s :  0.45%
    jit_opt_a.replace_applicator         :     0.000017s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.13%
    rewriter_after_opt_a                 :     0.000031s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000008s :  0.09%
    mutable_eliminate                    :     0.000531s :  6.40%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000023s :  0.28%
    loop_unroll                          :     0.000418s :  5.03%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000007s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000047s :  0.56%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000013s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000452s :  5.45%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000032s :  0.38%


Time group info:
------[substitution.]   0.000202    23
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 1.67% :     0.000003s :      3: substitution.graph_param_transform
85.57% :     0.000173s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.28% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.59% :     0.000003s :      2: substitution.replace_old_param
 5.77% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004347     2
81.66% :     0.003550s :      1: type_inference.infer
18.34% :     0.000797s :      1: type_inference.specialize
------[replace.]   0.000066     6
66.42% :     0.000044s :      4: replace.inline
33.58% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000180     6
94.19% :     0.000170s :      4: match.inline
 5.81% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.65% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.28% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 2.60% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.19% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.32% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.08% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 1.00% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.38% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.26% :     0.000006s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.40% :     0.000002s :      6: predicate.less_batch_normalization
 1.66% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 1.32% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.82% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.02% :     0.000002s :     15: predicate.partial_eliminate
 1.28% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.53% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.49% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.74% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 1.92% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.56% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.59% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000608    12
48.37% :     0.000294s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.62% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.02% :     0.000274s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013736    72
 0.36% :     0.000049s :      1: add_recomputation
 0.34% :     0.000047s :      1: auto_monad
 0.11% :     0.000015s :      1: auto_monad_reorder
 2.31% :     0.000317s :      1: bootstrap
 0.19% :     0.000026s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.14% :     0.000019s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
38.21% :     0.005249s :      1: jit_opt_a
 1.02% :     0.000140s :      1: jit_opt_after_cconv
 0.38% :     0.000053s :      1: jit_opt_b
 3.10% :     0.000426s :      1: loop_unroll
 3.93% :     0.000539s :      1: mutable_eliminate
 6.11% :     0.000840s :     26: opt.transform.jit_opt_a
 0.34% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.35% :     0.000460s :      1: opt_after_jit_grad
 0.07% :     0.000010s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000036s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000010s :      1: remove_dup_value
 2.71% :     0.000372s :      1: renormalize.infer
 2.29% :     0.000314s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000034s :      1: rewriter_after_opt_a
 0.47% :     0.000065s :      1: rewriter_before_opt_a
 0.56% :     0.000076s :      1: symbol_engine_optimizer
32.10% :     0.004409s :      1: type_inference


TotalTime = 0.0227247, [33]
[bootstrap]: 0.00034454
[type_inference]: 0.00821331
[event_method]: 1.774e-05
[auto_monad]: 6.47e-05
[graph_reusing]: 6.49999e-06
[pre_auto_parallel]: 2.88e-06
[py_interpret_to_execute]: 4.667e-05
[rewriter_before_opt_a]: 7.474e-05
[expand_dump_flag]: 2.84999e-06
[jit_opt_a]: 0.00607711, [2]
    [Cycle 1]: 0.00163066, [27]
        [switch_simplify]: 4.706e-05
        [loop_unroll]: 2.9e-05
        [a_1]: 0.00048814
        [with_stream_mark]: 1.807e-05
        [recompute_prepare]: 7.01999e-06
        [updatestate_depend_eliminate]: 3.11001e-06
        [updatestate_assign_eliminate]: 3.13e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 6.06e-06
        [updatestate_useless_node_eliminater]: 5.11002e-06
        [accelerated_algorithm]: 5.24998e-06
        [meta_shard_fg_expand]: 1.94999e-06
        [get_grad_eliminate_]: 5.84e-06
        [merge_forward]: 4.35e-06
        [cell_reuse_recompute_pass]: 1.09998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.363e-05
        [j_node_and_user_rematch]: 8.57998e-06
        [meta_fg_expand]: 2.12001e-06
        [replace_old_param]: 8.08999e-06
        [inline_without_move]: 5.57999e-06
        [renormalize]: 0.00075871
        [add_forward_monad_depend]: 4.76002e-06
        [auto_monad_grad]: 1.22999e-06
        [auto_monad_eliminator]: 1.001e-05
        [cse]: 1.651e-05
        [replace_applicator]: 1.161e-05
    [Cycle 2]: 0.00031588, [27]
        [switch_simplify]: 5.86e-06
        [loop_unroll]: 4.89e-06
        [a_1]: 9.041e-05
        [with_stream_mark]: 9.09998e-06
        [recompute_prepare]: 4.79998e-06
        [updatestate_depend_eliminate]: 2.79001e-06
        [updatestate_assign_eliminate]: 1.99999e-06
        [updatestate_loads_eliminate]: 1.91998e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 4.96002e-06
        [updatestate_useless_node_eliminater]: 4.58001e-06
        [accelerated_algorithm]: 5.39998e-06
        [meta_shard_fg_expand]: 9.5999e-07
        [get_grad_eliminate_]: 4.58001e-06
        [merge_forward]: 2.09e-06
        [cell_reuse_recompute_pass]: 1.89e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.198e-05
        [j_node_and_user_rematch]: 7.65e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 7.05e-06
        [inline_without_move]: 1.039e-05
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.16002e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 4.83001e-06
        [cse]: 1.016e-05
        [replace_applicator]: 4.90001e-06
[py_interpret_to_execute_after_opt_a]: 8.08999e-06
[rewriter_after_opt_a]: 2.66e-05
[convert_after_rewriter]: 5.37001e-06
[order_py_execute_after_rewriter]: 4.48999e-06
[mutable_eliminate]: 0.00051442
[jit_opt_b]: 4.818e-05, [1]
    [Cycle 1]: 4.173e-05, [2]
        [frontend_op_eliminate]: 1.643e-05
        [inline_after_opt_a]: 1.414e-05
[cconv]: 1.651e-05
[loop_unroll]: 0.00041688
[jit_opt_after_cconv]: 0.00013267, [1]
    [Cycle 1]: 0.0001264, [11]
        [c_1]: 1.966e-05
        [parameter_eliminate]: 2.53998e-06
        [updatestate_depend_eliminate]: 5.40001e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [cse]: 1.592e-05
        [call_graph_tuple_transform]: 1.829e-05
        [tuple_list_get_item_eliminator]: 4.89e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 5.99975e-07
        [switch_simplify]: 5.14998e-06
[remove_dup_value]: 1.277e-05
[partial_unused_args_eliminate]: 2.01998e-06
[environ_conv]: 5.69999e-06
[add_recomputation]: 4.666e-05
[cse_after_recomputation]: 1.961e-05, [1]
    [Cycle 1]: 1.403e-05, [1]
        [cse]: 8.60999e-06
[auto_monad_reorder]: 1.402e-05
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 5.07e-06
[opt_after_jit_grad]: 0.00048854
[symbol_engine_optimizer]: 7.005e-05, [1]
    [Cycle 1]: 6.369e-05, [6]
        [build]: 2.49999e-06
        [elim_shapecalc]: 7.38999e-06
        [elim_not_effective]: 1.281e-05
        [opt_reshape]: 5.54e-06
        [fold_const_symbol]: 8.23999e-06
        [renormalize]: 4.59986e-07
[validate]: 2.705e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00579564
[execute]: 7.85e-06
Sums
    bootstrap                            :     0.000345s :  1.91%
    type_inference                       :     0.008213s : 45.61%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000065s :  0.36%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000047s :  0.26%
    rewriter_before_opt_a                :     0.000075s :  0.42%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.29%
    jit_opt_a.loop_unroll                :     0.000034s :  0.19%
    jit_opt_a.a_1                        :     0.000579s :  3.21%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.08%
    jit_opt_a.inline_without_move        :     0.000016s :  0.09%
    jit_opt_a.renormalize                :     0.000759s :  4.21%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.08%
    jit_opt_a.cse                        :     0.000027s :  0.15%
    jit_opt_a.replace_applicator         :     0.000017s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.04%
    rewriter_after_opt_a                 :     0.000027s :  0.15%
    convert_after_rewriter               :     0.000005s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000514s :  2.86%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.08%
    cconv                                :     0.000017s :  0.09%
    loop_unroll                          :     0.000417s :  2.32%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000047s :  0.26%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.08%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000489s :  2.71%
    symbol_engine_optimizer.build        :     0.000002s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.15%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005796s : 32.19%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000166    23
 0.79% :     0.000001s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 3.32% :     0.000006s :      3: substitution.graph_param_transform
84.27% :     0.000140s :      4: substitution.inline
 1.66% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.77% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000003s :      2: substitution.replace_old_param
 4.59% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008128     2
65.67% :     0.005338s :      1: type_inference.infer
34.33% :     0.002790s :      1: type_inference.specialize
------[replace.]   0.000062     6
65.31% :     0.000041s :      4: replace.inline
34.69% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000144     6
95.48% :     0.000137s :      4: match.inline
 4.52% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.76% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 2.34% :     0.000003s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.64% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.65% :     0.000001s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.33% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.22% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 4.65% :     0.000005s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.15% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.70% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000002s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.82% :     0.000002s :     10: predicate.reduce_eliminate
 1.60% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.89% :     0.000002s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.62% :     0.000009s :     53: predicate.switch_simplify
 1.37% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.54% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.35% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.39% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000735    12
45.57% :     0.000335s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 3.79% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.64% :     0.000372s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024360    76
 0.20% :     0.000049s :      1: add_recomputation
 0.28% :     0.000068s :      1: auto_monad
 0.07% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.52% :     0.000370s :      1: bootstrap
 0.08% :     0.000019s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
24.96% :     0.006081s :      1: jit_opt_a
 0.56% :     0.000135s :      1: jit_opt_after_cconv
 0.21% :     0.000051s :      1: jit_opt_b
 1.74% :     0.000424s :      1: loop_unroll
 2.14% :     0.000522s :      1: mutable_eliminate
 3.22% :     0.000784s :     26: opt.transform.jit_opt_a
 0.18% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.04% :     0.000496s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000049s :      1: py_interpret_to_execute
 0.04% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000015s :      1: remove_dup_value
 1.83% :     0.000445s :      1: renormalize.infer
 1.25% :     0.000306s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000029s :      1: rewriter_after_opt_a
 0.32% :     0.000078s :      1: rewriter_before_opt_a
 0.30% :     0.000073s :      1: symbol_engine_optimizer
23.85% :     0.005809s :      1: task_emit
33.77% :     0.008226s :      1: type_inference
 0.18% :     0.000043s :      1: validate


TotalTime = 0.0112427, [30]
[bootstrap]: 0.00029273
[type_inference]: 0.00407652
[event_method]: 1.607e-05
[auto_monad]: 5.761e-05
[graph_reusing]: 4.98001e-06
[pre_auto_parallel]: 2.34001e-06
[py_interpret_to_execute]: 3.217e-05
[rewriter_before_opt_a]: 6.74e-05
[expand_dump_flag]: 3.08e-06
[jit_opt_a]: 0.00478989, [2]
    [Cycle 1]: 0.00139939, [27]
        [switch_simplify]: 4.56e-05
        [loop_unroll]: 4.664e-05
        [a_1]: 0.00049951
        [with_stream_mark]: 1.231e-05
        [recompute_prepare]: 7.9e-06
        [updatestate_depend_eliminate]: 2.88e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.02998e-06
        [specialize_transform]: 6.09001e-06
        [updatestate_useless_node_eliminater]: 5.24998e-06
        [accelerated_algorithm]: 5.59e-06
        [meta_shard_fg_expand]: 1.40001e-06
        [get_grad_eliminate_]: 5.24e-06
        [merge_forward]: 2.64999e-06
        [cell_reuse_recompute_pass]: 1.00999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.174e-05
        [j_node_and_user_rematch]: 9.10001e-06
        [meta_fg_expand]: 1.88997e-06
        [replace_old_param]: 7.65e-06
        [inline_without_move]: 5.32999e-06
        [renormalize]: 0.00051407
        [add_forward_monad_depend]: 4.22e-06
        [auto_monad_grad]: 1.53002e-06
        [auto_monad_eliminator]: 1.024e-05
        [cse]: 1.415e-05
        [replace_applicator]: 1.073e-05
    [Cycle 2]: 0.0003172, [27]
        [switch_simplify]: 5.59998e-06
        [loop_unroll]: 4.83001e-06
        [a_1]: 8.904e-05
        [with_stream_mark]: 8.45001e-06
        [recompute_prepare]: 4.93001e-06
        [updatestate_depend_eliminate]: 2.48e-06
        [updatestate_assign_eliminate]: 2.00002e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 5.12e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.46002e-06
        [meta_shard_fg_expand]: 1.01002e-06
        [get_grad_eliminate_]: 4.76002e-06
        [merge_forward]: 2.32001e-06
        [cell_reuse_recompute_pass]: 1.19003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.177e-05
        [j_node_and_user_rematch]: 7.65e-06
        [meta_fg_expand]: 1.37999e-06
        [replace_old_param]: 6.80002e-06
        [inline_without_move]: 4.66002e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 4.99e-06
        [cse]: 1.023e-05
        [replace_applicator]: 5.24e-06
[py_interpret_to_execute_after_opt_a]: 7.41999e-06
[rewriter_after_opt_a]: 2.393e-05
[convert_after_rewriter]: 6.12999e-06
[order_py_execute_after_rewriter]: 4.32e-06
[mutable_eliminate]: 0.0004639
[jit_opt_b]: 4.763e-05, [1]
    [Cycle 1]: 4.169e-05, [2]
        [frontend_op_eliminate]: 1.599e-05
        [inline_after_opt_a]: 1.463e-05
[cconv]: 1.809e-05
[loop_unroll]: 0.00041686
[jit_opt_after_cconv]: 0.00013446, [1]
    [Cycle 1]: 0.00012816, [11]
        [c_1]: 2.097e-05
        [parameter_eliminate]: 2.36e-06
        [updatestate_depend_eliminate]: 5.34998e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [cse]: 1.582e-05
        [call_graph_tuple_transform]: 1.687e-05
        [tuple_list_get_item_eliminator]: 5.46e-06
        [none_parameter_eliminate]: 1.22e-06
        [renormalize]: 4.49974e-07
        [switch_simplify]: 5.25999e-06
[remove_dup_value]: 7.56999e-06
[partial_unused_args_eliminate]: 1.71e-06
[environ_conv]: 3.33998e-06
[add_recomputation]: 3.308e-05
[cse_after_recomputation]: 1.923e-05, [1]
    [Cycle 1]: 1.407e-05, [1]
        [cse]: 8.57e-06
[auto_monad_reorder]: 1.007e-05
[get_jit_bprop_graph]: 1.44e-06
[rewriter_after_jit_bprop_graph]: 4.52998e-06
[opt_after_jit_grad]: 0.00044312
[symbol_engine_optimizer]: 7.116e-05, [1]
    [Cycle 1]: 6.479e-05, [6]
        [build]: 2.66e-06
        [elim_shapecalc]: 7.53999e-06
        [elim_not_effective]: 1.203e-05
        [opt_reshape]: 5.87001e-06
        [fold_const_symbol]: 8.22e-06
        [renormalize]: 4.30009e-07
[validate]: 2.62e-05
Sums
    bootstrap                            :     0.000293s :  3.84%
    type_inference                       :     0.004077s : 53.52%
    event_method                         :     0.000016s :  0.21%
    auto_monad                           :     0.000058s :  0.76%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.42%
    rewriter_before_opt_a                :     0.000067s :  0.88%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000051s :  0.67%
    jit_opt_a.loop_unroll                :     0.000051s :  0.68%
    jit_opt_a.a_1                        :     0.000589s :  7.73%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.04%
    jit_opt_a.replace_old_param          :     0.000014s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000514s :  6.75%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.20%
    jit_opt_a.cse                        :     0.000024s :  0.32%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.10%
    rewriter_after_opt_a                 :     0.000024s :  0.31%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000004s :  0.06%
    mutable_eliminate                    :     0.000464s :  6.09%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000018s :  0.24%
    loop_unroll                          :     0.000417s :  5.47%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000008s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000003s :  0.04%
    add_recomputation                    :     0.000033s :  0.43%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000010s :  0.13%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000443s :  5.82%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000026s :  0.34%


Time group info:
------[substitution.]   0.000160    23
 0.91% :     0.000001s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.40% :     0.000004s :      3: substitution.graph_param_transform
84.14% :     0.000135s :      4: substitution.inline
 1.66% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.78% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.78% :     0.000003s :      2: substitution.replace_old_param
 5.62% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004022     2
79.61% :     0.003202s :      1: type_inference.infer
20.39% :     0.000820s :      1: type_inference.specialize
------[replace.]   0.000064     6
65.58% :     0.000042s :      4: replace.inline
34.42% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000140     6
94.36% :     0.000133s :      4: match.inline
 5.64% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.56% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.49% :     0.000003s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.44% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.24% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 5.20% :     0.000006s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.21% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.81% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.55% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.14% :     0.000008s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.41% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.74% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000559    12
48.64% :     0.000272s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.77% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.58% :     0.000261s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012624    72
 0.29% :     0.000036s :      1: add_recomputation
 0.48% :     0.000061s :      1: auto_monad
 0.10% :     0.000013s :      1: auto_monad_reorder
 2.42% :     0.000306s :      1: bootstrap
 0.17% :     0.000021s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.17% :     0.000021s :      1: cse_after_recomputation
 0.05% :     0.000006s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
37.96% :     0.004792s :      1: jit_opt_a
 1.09% :     0.000137s :      1: jit_opt_after_cconv
 0.40% :     0.000050s :      1: jit_opt_b
 3.37% :     0.000425s :      1: loop_unroll
 3.74% :     0.000472s :      1: mutable_eliminate
 6.36% :     0.000803s :     26: opt.transform.jit_opt_a
 0.36% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.58% :     0.000451s :      1: opt_after_jit_grad
 0.05% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.04% :     0.000004s :      1: pre_auto_parallel
 0.28% :     0.000035s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000010s :      1: remove_dup_value
 2.00% :     0.000252s :      1: renormalize.infer
 2.02% :     0.000255s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000027s :      1: rewriter_after_opt_a
 0.56% :     0.000070s :      1: rewriter_before_opt_a
 0.58% :     0.000074s :      1: symbol_engine_optimizer
32.38% :     0.004087s :      1: type_inference


TotalTime = 0.0297775, [33]
[bootstrap]: 0.00018572
[type_inference]: 0.00940184
[event_method]: 4.075e-05
[auto_monad]: 9.114e-05
[graph_reusing]: 6.49999e-06
[pre_auto_parallel]: 2.61999e-06
[py_interpret_to_execute]: 2.606e-05
[rewriter_before_opt_a]: 9.244e-05
[expand_dump_flag]: 3.16001e-06
[jit_opt_a]: 0.0109911, [3]
    [Cycle 1]: 0.00541883, [27]
        [switch_simplify]: 6.495e-05
        [loop_unroll]: 3.913e-05
        [a_1]: 0.000831
        [with_stream_mark]: 2.775e-05
        [recompute_prepare]: 1.815e-05
        [updatestate_depend_eliminate]: 8.02e-06
        [updatestate_assign_eliminate]: 6.83e-06
        [updatestate_loads_eliminate]: 6.16998e-06
        [parameter_eliminate]: 2.56998e-06
        [specialize_transform]: 1.29e-05
        [updatestate_useless_node_eliminater]: 1.206e-05
        [accelerated_algorithm]: 1.233e-05
        [meta_shard_fg_expand]: 3.6e-06
        [get_grad_eliminate_]: 1.207e-05
        [merge_forward]: 8.42e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.578e-05
        [j_node_and_user_rematch]: 2.259e-05
        [meta_fg_expand]: 0.00146458
        [replace_old_param]: 5.613e-05
        [inline_without_move]: 5.303e-05
        [renormalize]: 0.00225219
        [add_forward_monad_depend]: 9.15001e-06
        [auto_monad_grad]: 5.60001e-06
        [auto_monad_eliminator]: 4.919e-05
        [cse]: 0.00014587
        [replace_applicator]: 6.728e-05
    [Cycle 2]: 0.00190834, [27]
        [switch_simplify]: 3.576e-05
        [loop_unroll]: 3.475e-05
        [a_1]: 0.00105317
        [with_stream_mark]: 1.346e-05
        [recompute_prepare]: 7.81001e-06
        [updatestate_depend_eliminate]: 4.2e-06
        [updatestate_assign_eliminate]: 3.41001e-06
        [updatestate_loads_eliminate]: 2.80997e-06
        [parameter_eliminate]: 1.58002e-06
        [specialize_transform]: 6.92002e-06
        [updatestate_useless_node_eliminater]: 6.00002e-06
        [accelerated_algorithm]: 5.92999e-06
        [meta_shard_fg_expand]: 2.28002e-06
        [get_grad_eliminate_]: 5.70001e-06
        [merge_forward]: 3.83999e-06
        [cell_reuse_recompute_pass]: 9.80013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.408e-05
        [j_node_and_user_rematch]: 1.004e-05
        [meta_fg_expand]: 3.533e-05
        [replace_old_param]: 9.45001e-06
        [inline_without_move]: 5.75001e-06
        [renormalize]: 0.00044552
        [add_forward_monad_depend]: 4.04002e-06
        [auto_monad_grad]: 1.08001e-06
        [auto_monad_eliminator]: 1.207e-05
        [cse]: 2.148e-05
        [replace_applicator]: 1.216e-05
    [Cycle 3]: 0.00036415, [27]
        [switch_simplify]: 6.43e-06
        [loop_unroll]: 6.01e-06
        [a_1]: 0.0001186
        [with_stream_mark]: 9.04e-06
        [recompute_prepare]: 5.79e-06
        [updatestate_depend_eliminate]: 3.83999e-06
        [updatestate_assign_eliminate]: 2.71e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 1.07998e-06
        [specialize_transform]: 6.12999e-06
        [updatestate_useless_node_eliminater]: 5.85002e-06
        [accelerated_algorithm]: 5.82999e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 5.59998e-06
        [merge_forward]: 2.89001e-06
        [cell_reuse_recompute_pass]: 1.48002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.464e-05
        [j_node_and_user_rematch]: 9.76e-06
        [meta_fg_expand]: 2.02001e-06
        [replace_old_param]: 7.92e-06
        [inline_without_move]: 5.61e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.23002e-06
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 6.73e-06
        [cse]: 1.35e-05
        [replace_applicator]: 6.34999e-06
[py_interpret_to_execute_after_opt_a]: 8.80001e-06
[rewriter_after_opt_a]: 3.532e-05
[convert_after_rewriter]: 7.57002e-06
[order_py_execute_after_rewriter]: 5.35999e-06
[mutable_eliminate]: 0.00056034
[jit_opt_b]: 5.402e-05, [1]
    [Cycle 1]: 4.715e-05, [2]
        [frontend_op_eliminate]: 1.862e-05
        [inline_after_opt_a]: 1.708e-05
[cconv]: 2.291e-05
[loop_unroll]: 0.00040956
[jit_opt_after_cconv]: 0.00015447, [1]
    [Cycle 1]: 0.00014769, [11]
        [c_1]: 2.585e-05
        [parameter_eliminate]: 2.56e-06
        [updatestate_depend_eliminate]: 7.16001e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [cse]: 2.306e-05
        [call_graph_tuple_transform]: 2.073e-05
        [tuple_list_get_item_eliminator]: 6.21e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 3.60014e-07
        [switch_simplify]: 6.14999e-06
[remove_dup_value]: 1.63e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 5.39e-06
[add_recomputation]: 5.076e-05
[cse_after_recomputation]: 2.723e-05, [1]
    [Cycle 1]: 2.132e-05, [1]
        [cse]: 1.501e-05
[auto_monad_reorder]: 1.852e-05
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 4.45999e-06
[opt_after_jit_grad]: 0.00045286
[symbol_engine_optimizer]: 8.052e-05, [1]
    [Cycle 1]: 7.431e-05, [6]
        [build]: 3.98999e-06
        [elim_shapecalc]: 8.64003e-06
        [elim_not_effective]: 1.511e-05
        [opt_reshape]: 7.01001e-06
        [fold_const_symbol]: 1.092e-05
        [renormalize]: 4.69998e-07
[validate]: 3.501e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.00679154
[execute]: 7.97e-06
Sums
    bootstrap                            :     0.000186s :  0.72%
    type_inference                       :     0.009402s : 36.56%
    event_method                         :     0.000041s :  0.16%
    auto_monad                           :     0.000091s :  0.35%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.10%
    rewriter_before_opt_a                :     0.000092s :  0.36%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000107s :  0.42%
    jit_opt_a.loop_unroll                :     0.000080s :  0.31%
    jit_opt_a.a_1                        :     0.002003s :  7.79%
    jit_opt_a.with_stream_mark           :     0.000050s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.09%
    jit_opt_a.merge_forward              :     0.000015s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000054s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001502s :  5.84%
    jit_opt_a.replace_old_param          :     0.000074s :  0.29%
    jit_opt_a.inline_without_move        :     0.000064s :  0.25%
    jit_opt_a.renormalize                :     0.002698s : 10.49%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000068s :  0.26%
    jit_opt_a.cse                        :     0.000181s :  0.70%
    jit_opt_a.replace_applicator         :     0.000086s :  0.33%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.03%
    rewriter_after_opt_a                 :     0.000035s :  0.14%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000560s :  2.18%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.07%
    cconv                                :     0.000023s :  0.09%
    loop_unroll                          :     0.000410s :  1.59%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000051s :  0.20%
    cse_after_recomputation.cse          :     0.000015s :  0.06%
    auto_monad_reorder                   :     0.000019s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.02%
    opt_after_jit_grad                   :     0.000453s :  1.76%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006792s : 26.41%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000477   110
 0.47% :     0.000002s :      3: substitution.elim_not_effective
 0.33% :     0.000002s :      3: substitution.fold_const_symbol
 1.19% :     0.000006s :      4: substitution.graph_param_transform
62.17% :     0.000297s :     12: substitution.inline
 3.36% :     0.000016s :      2: substitution.inline_without_move
 1.88% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.23% :     0.000011s :      7: substitution.minmaximum_grad
 1.25% :     0.000006s :      5: substitution.partial_eliminate
 2.28% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.57% :     0.000027s :     10: substitution.replace_applicator
 1.41% :     0.000007s :      7: substitution.replace_old_param
 0.48% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.16% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.10% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
10.12% :     0.000048s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009334     2
89.45% :     0.008349s :      1: type_inference.infer
10.55% :     0.000985s :      1: type_inference.specialize
------[replace.]   0.000146    19
64.17% :     0.000094s :     12: replace.inline
35.83% :     0.000052s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000309    19
93.65% :     0.000290s :     12: match.inline
 6.35% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000368  2568
 1.63% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.41% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.44% :     0.000005s :     43: predicate.addn_check_dump
 1.66% :     0.000006s :     43: predicate.addn_zero_filter
 2.07% :     0.000008s :     43: predicate.arithmetic_simplify
 1.52% :     0.000006s :     43: predicate.cast_eliminate
 0.17% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.44% :     0.000005s :     43: predicate.compare_switch_simplify
 1.58% :     0.000006s :     43: predicate.depend_value_elim
 1.42% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.50% :     0.000006s :     43: predicate.dict_get_item_eliminator
 1.51% :     0.000006s :     43: predicate.dict_set_item_eliminator
 0.36% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      4: predicate.elim_not_effective
 0.20% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.48% :     0.000005s :     43: predicate.environ_add_const_eliminate
 1.44% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.46% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.53% :     0.000006s :     43: predicate.environ_get_eliminate
 1.42% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.81% :     0.000003s :     19: predicate.get_grad_eliminate
 0.11% :     0.000000s :      4: predicate.graph_param_transform
 4.45% :     0.000016s :     70: predicate.inline
 1.95% :     0.000007s :     40: predicate.inline_without_move
 0.37% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.85% :     0.000003s :     19: predicate.less_batch_normalization
 1.78% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.92% :     0.000007s :     54: predicate.load_eliminater
 0.45% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.17% :     0.000012s :     85: predicate.loop_unroll_before_grad
 1.70% :     0.000006s :     47: predicate.make_slice_get_slice_eliminator
 1.45% :     0.000005s :     43: predicate.merge_addn
 1.49% :     0.000005s :     43: predicate.minmaximum_grad
 0.50% :     0.000002s :      4: predicate.mutable_eliminate
 0.18% :     0.000001s :      4: predicate.opt_reshape
 2.40% :     0.000009s :     54: predicate.partial_eliminate
 1.57% :     0.000006s :     43: predicate.print_const_string_wrapper
 2.10% :     0.000008s :     43: predicate.reduce_eliminate
 1.79% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.56% :     0.000009s :     96: predicate.replace_applicator
 0.92% :     0.000003s :     40: predicate.replace_old_param
 0.15% :     0.000001s :      4: predicate.reset_defer_inline
 1.72% :     0.000006s :     43: predicate.reshape_eliminate
 1.54% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.34% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.56% :     0.000006s :     43: predicate.same_eliminate
 0.49% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.36% :     0.000001s :      8: predicate.special_op_eliminate
 0.90% :     0.000003s :     19: predicate.specialize_transform
 1.72% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000005s :     43: predicate.stack_unstack_eliminate
 0.18% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.70% :     0.000010s :     62: predicate.switch_defer_inline
 2.52% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.37% :     0.000023s :    151: predicate.switch_simplify
 1.56% :     0.000006s :     43: predicate.tile_eliminate
 1.45% :     0.000005s :     43: predicate.transpose_eliminate
 1.89% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.70% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.28% :     0.000012s :     58: predicate.tuple_list_get_item_eliminator
 1.87% :     0.000007s :     43: predicate.tuple_list_set_item_eliminator
 1.88% :     0.000007s :     50: predicate.tuple_to_list_eliminator_
 1.88% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000010s :     73: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000007s :     43: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.21% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001343    29
60.83% :     0.000817s :     13: func_graph_cloner_run.FuncGraphClonerGraph
39.17% :     0.000526s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.035195    91
 0.15% :     0.000054s :      1: add_recomputation
 0.27% :     0.000095s :      1: auto_monad
 0.06% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.56% :     0.000197s :      1: bootstrap
 0.07% :     0.000025s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000029s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.13% :     0.000045s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
31.24% :     0.010994s :      1: jit_opt_a
 0.45% :     0.000157s :      1: jit_opt_after_cconv
 0.16% :     0.000057s :      1: jit_opt_b
 1.18% :     0.000417s :      1: loop_unroll
 1.61% :     0.000568s :      1: mutable_eliminate
 7.38% :     0.002597s :     39: opt.transform.jit_opt_a
 0.16% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000030s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.31% :     0.000460s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000028s :      1: py_interpret_to_execute
 0.03% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 3.99% :     0.001405s :      2: renormalize.infer
 3.63% :     0.001279s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000038s :      1: rewriter_after_opt_a
 0.27% :     0.000095s :      1: rewriter_before_opt_a
 0.24% :     0.000083s :      1: symbol_engine_optimizer
19.33% :     0.006802s :      1: task_emit
26.75% :     0.009414s :      1: type_inference
 0.16% :     0.000055s :      1: validate


TotalTime = 0.01668, [33]
[bootstrap]: 0.00032728
[type_inference]: 0.00429546
[event_method]: 1.6e-05
[auto_monad]: 4.368e-05
[graph_reusing]: 4.39002e-06
[pre_auto_parallel]: 1.73002e-06
[py_interpret_to_execute]: 3.048e-05
[rewriter_before_opt_a]: 6.319e-05
[expand_dump_flag]: 2.27001e-06
[jit_opt_a]: 0.00486188, [2]
    [Cycle 1]: 0.00132339, [27]
        [switch_simplify]: 4.654e-05
        [loop_unroll]: 2.9e-05
        [a_1]: 0.00045933
        [with_stream_mark]: 1.199e-05
        [recompute_prepare]: 6.93e-06
        [updatestate_depend_eliminate]: 2.72001e-06
        [updatestate_assign_eliminate]: 2.03997e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 6.01e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.62999e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 5.05001e-06
        [merge_forward]: 2.51e-06
        [cell_reuse_recompute_pass]: 1.20999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.119e-05
        [j_node_and_user_rematch]: 8.88002e-06
        [meta_fg_expand]: 1.76998e-06
        [replace_old_param]: 7.33e-06
        [inline_without_move]: 5.18002e-06
        [renormalize]: 0.00049463
        [add_forward_monad_depend]: 4.45e-06
        [auto_monad_grad]: 1.20999e-06
        [auto_monad_eliminator]: 9.44e-06
        [cse]: 1.441e-05
        [replace_applicator]: 1.173e-05
    [Cycle 2]: 0.00032647, [27]
        [switch_simplify]: 5.12e-06
        [loop_unroll]: 4.85001e-06
        [a_1]: 9.006e-05
        [with_stream_mark]: 8.64998e-06
        [recompute_prepare]: 4.96997e-06
        [updatestate_depend_eliminate]: 2.44001e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 4.72998e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 5.11002e-06
        [meta_shard_fg_expand]: 9.30013e-07
        [get_grad_eliminate_]: 4.22e-06
        [merge_forward]: 2.24001e-06
        [cell_reuse_recompute_pass]: 1.56002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.149e-05
        [j_node_and_user_rematch]: 8.13999e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 7.06999e-06
        [inline_without_move]: 4.54998e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.26002e-06
        [auto_monad_grad]: 8.40024e-07
        [auto_monad_eliminator]: 5.30001e-06
        [cse]: 1.179e-05
        [replace_applicator]: 7.16001e-06
[py_interpret_to_execute_after_opt_a]: 8.07e-06
[rewriter_after_opt_a]: 2.534e-05
[convert_after_rewriter]: 5.78002e-06
[order_py_execute_after_rewriter]: 4.23999e-06
[mutable_eliminate]: 0.00046824
[jit_opt_b]: 4.672e-05, [1]
    [Cycle 1]: 4.034e-05, [2]
        [frontend_op_eliminate]: 1.504e-05
        [inline_after_opt_a]: 1.412e-05
[cconv]: 1.564e-05
[loop_unroll]: 0.00042439
[jit_opt_after_cconv]: 0.00013756, [1]
    [Cycle 1]: 0.00013109, [11]
        [c_1]: 2.028e-05
        [parameter_eliminate]: 2.82002e-06
        [updatestate_depend_eliminate]: 5.42001e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.32001e-06
        [cse]: 1.693e-05
        [call_graph_tuple_transform]: 1.787e-05
        [tuple_list_get_item_eliminator]: 5.62999e-06
        [none_parameter_eliminate]: 1.07e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 5.30999e-06
[remove_dup_value]: 7.13e-06
[partial_unused_args_eliminate]: 1.46998e-06
[environ_conv]: 4.02e-06
[add_recomputation]: 3.288e-05
[cse_after_recomputation]: 2.013e-05, [1]
    [Cycle 1]: 1.491e-05, [1]
        [cse]: 9.02e-06
[auto_monad_reorder]: 1.002e-05
[get_jit_bprop_graph]: 1.49e-06
[rewriter_after_jit_bprop_graph]: 5.66e-06
[opt_after_jit_grad]: 0.00048595
[symbol_engine_optimizer]: 7.441e-05, [1]
    [Cycle 1]: 6.795e-05, [6]
        [build]: 3.13998e-06
        [elim_shapecalc]: 8.05e-06
        [elim_not_effective]: 1.293e-05
        [opt_reshape]: 5.74e-06
        [fold_const_symbol]: 8.64e-06
        [renormalize]: 5.09986e-07
[validate]: 2.663e-05
[backend_pass]: 1.31002e-06
[task_emit]: 0.00504819
[execute]: 4.63001e-06
Sums
    bootstrap                            :     0.000327s :  2.54%
    type_inference                       :     0.004295s : 33.31%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000044s :  0.34%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000030s :  0.24%
    rewriter_before_opt_a                :     0.000063s :  0.49%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000052s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.26%
    jit_opt_a.a_1                        :     0.000549s :  4.26%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000495s :  3.84%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000026s :  0.20%
    jit_opt_a.replace_applicator         :     0.000019s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000025s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000468s :  3.63%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000016s :  0.12%
    loop_unroll                          :     0.000424s :  3.29%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000007s :  0.06%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000033s :  0.25%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000010s :  0.08%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000486s :  3.77%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000027s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005048s : 39.15%
    execute                              :     0.000005s :  0.04%


Time group info:
------[substitution.]   0.000147    23
 0.94% :     0.000001s :      2: substitution.elim_not_effective
 0.74% :     0.000001s :      2: substitution.fold_const_symbol
 2.77% :     0.000004s :      3: substitution.graph_param_transform
82.11% :     0.000121s :      4: substitution.inline
 1.69% :     0.000002s :      4: substitution.j_node_and_user_rematch
 4.60% :     0.000007s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000003s :      2: substitution.replace_old_param
 5.11% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004241     2
81.30% :     0.003448s :      1: type_inference.infer
18.70% :     0.000793s :      1: type_inference.specialize
------[replace.]   0.000058     6
66.07% :     0.000039s :      4: replace.inline
33.93% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000125     6
94.82% :     0.000118s :      4: match.inline
 5.18% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.05% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000001s :     10: predicate.addn_zero_filter
 1.90% :     0.000002s :     10: predicate.arithmetic_simplify
 1.31% :     0.000001s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000001s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.39% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.14% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.54% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 4.86% :     0.000005s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.73% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.89% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.09% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.78% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.36% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 2.01% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.62% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.91% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000555    12
46.85% :     0.000260s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.57% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.58% :     0.000269s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018039    76
 0.20% :     0.000036s :      1: add_recomputation
 0.26% :     0.000047s :      1: auto_monad
 0.07% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.88% :     0.000339s :      1: bootstrap
 0.10% :     0.000018s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.04% :     0.000008s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.97% :     0.004864s :      1: jit_opt_a
 0.78% :     0.000140s :      1: jit_opt_after_cconv
 0.27% :     0.000049s :      1: jit_opt_b
 2.40% :     0.000432s :      1: loop_unroll
 2.64% :     0.000477s :      1: mutable_eliminate
 4.19% :     0.000756s :     26: opt.transform.jit_opt_a
 0.25% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.74% :     0.000495s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000009s :      1: remove_dup_value
 1.33% :     0.000240s :      1: renormalize.infer
 1.38% :     0.000248s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000029s :      1: rewriter_after_opt_a
 0.37% :     0.000066s :      1: rewriter_before_opt_a
 0.43% :     0.000077s :      1: symbol_engine_optimizer
28.03% :     0.005057s :      1: task_emit
23.87% :     0.004306s :      1: type_inference
 0.24% :     0.000043s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.817.887 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc39040> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.818.390 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0113945, [30]
[bootstrap]: 0.00027274
[type_inference]: 0.00413675
[event_method]: 1.699e-05
[auto_monad]: 6.143e-05
[graph_reusing]: 5.64e-06
[pre_auto_parallel]: 2.49999e-06
[py_interpret_to_execute]: 3.254e-05
[rewriter_before_opt_a]: 7.169e-05
[expand_dump_flag]: 3.23e-06
[jit_opt_a]: 0.00483818, [2]
    [Cycle 1]: 0.00145671, [27]
        [switch_simplify]: 4.853e-05
        [loop_unroll]: 2.939e-05
        [a_1]: 0.00054829
        [with_stream_mark]: 1.53e-05
        [recompute_prepare]: 7.15e-06
        [updatestate_depend_eliminate]: 3.26001e-06
        [updatestate_assign_eliminate]: 3.04999e-06
        [updatestate_loads_eliminate]: 2.76e-06
        [parameter_eliminate]: 2.12999e-06
        [specialize_transform]: 5.97999e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 5.83002e-06
        [meta_shard_fg_expand]: 2.04999e-06
        [get_grad_eliminate_]: 5.43002e-06
        [merge_forward]: 3.42002e-06
        [cell_reuse_recompute_pass]: 1.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.476e-05
        [j_node_and_user_rematch]: 9.29998e-06
        [meta_fg_expand]: 2.32999e-06
        [replace_old_param]: 9.17999e-06
        [inline_without_move]: 5.76e-06
        [renormalize]: 0.00051003
        [add_forward_monad_depend]: 5.07e-06
        [auto_monad_grad]: 1.85001e-06
        [auto_monad_eliminator]: 1.279e-05
        [cse]: 2.555e-05
        [replace_applicator]: 1.107e-05
    [Cycle 2]: 0.00031943, [27]
        [switch_simplify]: 5.66998e-06
        [loop_unroll]: 5.02e-06
        [a_1]: 9.075e-05
        [with_stream_mark]: 8.80001e-06
        [recompute_prepare]: 5.19e-06
        [updatestate_depend_eliminate]: 2.46998e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 5.02e-06
        [updatestate_useless_node_eliminater]: 4.68999e-06
        [accelerated_algorithm]: 5.12999e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 5.09998e-06
        [merge_forward]: 2.41e-06
        [cell_reuse_recompute_pass]: 1.18001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.123e-05
        [j_node_and_user_rematch]: 7.45998e-06
        [meta_fg_expand]: 1.47999e-06
        [replace_old_param]: 6.68e-06
        [inline_without_move]: 4.70999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 4.86002e-06
        [cse]: 1.015e-05
        [replace_applicator]: 5.07999e-06
[py_interpret_to_execute_after_opt_a]: 9.30001e-06
[rewriter_after_opt_a]: 3.125e-05
[convert_after_rewriter]: 6.93e-06
[order_py_execute_after_rewriter]: 5.51e-06
[mutable_eliminate]: 0.00046844
[jit_opt_b]: 4.936e-05, [1]
    [Cycle 1]: 4.212e-05, [2]
        [frontend_op_eliminate]: 1.631e-05
        [inline_after_opt_a]: 1.469e-05
[cconv]: 2.427e-05
[loop_unroll]: 0.00041004
[jit_opt_after_cconv]: 0.00013362, [1]
    [Cycle 1]: 0.00012707, [11]
        [c_1]: 2.042e-05
        [parameter_eliminate]: 2.27999e-06
        [updatestate_depend_eliminate]: 5.10999e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.595e-05
        [call_graph_tuple_transform]: 1.751e-05
        [tuple_list_get_item_eliminator]: 5.39998e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 5.3001e-07
        [switch_simplify]: 5.05999e-06
[remove_dup_value]: 1.342e-05
[partial_unused_args_eliminate]: 2.29001e-06
[environ_conv]: 5.15001e-06
[add_recomputation]: 4.482e-05
[cse_after_recomputation]: 1.967e-05, [1]
    [Cycle 1]: 1.423e-05, [1]
        [cse]: 8.62e-06
[auto_monad_reorder]: 1.493e-05
[get_jit_bprop_graph]: 1.40999e-06
[rewriter_after_jit_bprop_graph]: 4.86997e-06
[opt_after_jit_grad]: 0.00045082
[symbol_engine_optimizer]: 7.516e-05, [1]
    [Cycle 1]: 6.861e-05, [6]
        [build]: 3.90998e-06
        [elim_shapecalc]: 7.55e-06
        [elim_not_effective]: 1.33e-05
        [opt_reshape]: 5.62001e-06
        [fold_const_symbol]: 9.19e-06
        [renormalize]: 4.49974e-07
[validate]: 2.953e-05
Sums
    bootstrap                            :     0.000273s :  3.50%
    type_inference                       :     0.004137s : 53.16%
    event_method                         :     0.000017s :  0.22%
    auto_monad                           :     0.000061s :  0.79%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.42%
    rewriter_before_opt_a                :     0.000072s :  0.92%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000054s :  0.70%
    jit_opt_a.loop_unroll                :     0.000034s :  0.44%
    jit_opt_a.a_1                        :     0.000639s :  8.21%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.14%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000510s :  6.55%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.23%
    jit_opt_a.cse                        :     0.000036s :  0.46%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.12%
    rewriter_after_opt_a                 :     0.000031s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000006s :  0.07%
    mutable_eliminate                    :     0.000468s :  6.02%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000024s :  0.31%
    loop_unroll                          :     0.000410s :  5.27%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000013s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000045s :  0.58%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.19%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000451s :  5.79%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000030s :  0.38%


Time group info:
------[substitution.]   0.000176    23
 1.31% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000001s :      2: substitution.fold_const_symbol
 2.57% :     0.000005s :      3: substitution.graph_param_transform
81.90% :     0.000144s :      4: substitution.inline
 1.77% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.59% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000004s :      2: substitution.replace_old_param
 7.07% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004081     2
78.98% :     0.003223s :      1: type_inference.infer
21.02% :     0.000858s :      1: type_inference.specialize
------[replace.]   0.000066     6
65.35% :     0.000043s :      4: replace.inline
34.65% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000153     6
92.60% :     0.000141s :      4: match.inline
 7.40% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.42% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000001s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000001s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.36% :     0.000006s :     22: predicate.inline
 1.03% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.70% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.08% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.83% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000002s :     15: predicate.partial_eliminate
 1.32% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.73% :     0.000009s :     53: predicate.switch_simplify
 1.28% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.99% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000577    12
49.65% :     0.000286s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.61% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.74% :     0.000264s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012812    72
 0.37% :     0.000048s :      1: add_recomputation
 0.51% :     0.000065s :      1: auto_monad
 0.14% :     0.000018s :      1: auto_monad_reorder
 2.21% :     0.000283s :      1: bootstrap
 0.21% :     0.000027s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
37.78% :     0.004841s :      1: jit_opt_a
 1.06% :     0.000136s :      1: jit_opt_after_cconv
 0.41% :     0.000052s :      1: jit_opt_b
 3.26% :     0.000418s :      1: loop_unroll
 3.72% :     0.000476s :      1: mutable_eliminate
 6.56% :     0.000841s :     26: opt.transform.jit_opt_a
 0.35% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.58% :     0.000458s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.04% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000035s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 1.87% :     0.000240s :      1: renormalize.infer
 2.05% :     0.000262s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000036s :      1: rewriter_after_opt_a
 0.58% :     0.000074s :      1: rewriter_before_opt_a
 0.61% :     0.000078s :      1: symbol_engine_optimizer
32.37% :     0.004147s :      1: type_inference


TotalTime = 0.0161073, [33]
[bootstrap]: 0.00025637
[type_inference]: 0.00404533
[event_method]: 1.535e-05
[auto_monad]: 4.32e-05
[graph_reusing]: 3.9e-06
[pre_auto_parallel]: 1.46002e-06
[py_interpret_to_execute]: 2.765e-05
[rewriter_before_opt_a]: 6.09e-05
[expand_dump_flag]: 2.25002e-06
[jit_opt_a]: 0.00477152, [2]
    [Cycle 1]: 0.00128649, [27]
        [switch_simplify]: 4.379e-05
        [loop_unroll]: 2.858e-05
        [a_1]: 0.00046952
        [with_stream_mark]: 1.146e-05
        [recompute_prepare]: 6.33e-06
        [updatestate_depend_eliminate]: 2.83998e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 5.74999e-06
        [updatestate_useless_node_eliminater]: 4.79998e-06
        [accelerated_algorithm]: 5.81e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 4.77e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 8.70001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.114e-05
        [j_node_and_user_rematch]: 9.31e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 8e-06
        [inline_without_move]: 5.10999e-06
        [renormalize]: 0.00047159
        [add_forward_monad_depend]: 3.45e-06
        [auto_monad_grad]: 1.09e-06
        [auto_monad_eliminator]: 8.27998e-06
        [cse]: 1.423e-05
        [replace_applicator]: 8.77e-06
    [Cycle 2]: 0.00031085, [27]
        [switch_simplify]: 5.29e-06
        [loop_unroll]: 4.60999e-06
        [a_1]: 8.899e-05
        [with_stream_mark]: 7.6e-06
        [recompute_prepare]: 4.74e-06
        [updatestate_depend_eliminate]: 2.49999e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.80001e-06
        [parameter_eliminate]: 8.90024e-07
        [specialize_transform]: 4.78001e-06
        [updatestate_useless_node_eliminater]: 4.56002e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 9.80013e-07
        [get_grad_eliminate_]: 4.47998e-06
        [merge_forward]: 2.31e-06
        [cell_reuse_recompute_pass]: 1.72999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.253e-05
        [j_node_and_user_rematch]: 7.71999e-06
        [meta_fg_expand]: 1.52999e-06
        [replace_old_param]: 7.21999e-06
        [inline_without_move]: 4.45999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 7.29982e-07
        [auto_monad_eliminator]: 4.78001e-06
        [cse]: 9.72999e-06
        [replace_applicator]: 4.76002e-06
[py_interpret_to_execute_after_opt_a]: 7.53999e-06
[rewriter_after_opt_a]: 2.551e-05
[convert_after_rewriter]: 5.65001e-06
[order_py_execute_after_rewriter]: 4.46002e-06
[mutable_eliminate]: 0.00044392
[jit_opt_b]: 4.749e-05, [1]
    [Cycle 1]: 4.147e-05, [2]
        [frontend_op_eliminate]: 1.617e-05
        [inline_after_opt_a]: 1.407e-05
[cconv]: 1.438e-05
[loop_unroll]: 0.00040531
[jit_opt_after_cconv]: 0.00013917, [1]
    [Cycle 1]: 0.00013295, [11]
        [c_1]: 1.96e-05
        [parameter_eliminate]: 1.89e-06
        [updatestate_depend_eliminate]: 4.48999e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.591e-05
        [call_graph_tuple_transform]: 1.554e-05
        [tuple_list_get_item_eliminator]: 5.04e-06
        [none_parameter_eliminate]: 7.7e-07
        [renormalize]: 3.29979e-07
        [switch_simplify]: 6.15002e-06
[remove_dup_value]: 8.52e-06
[partial_unused_args_eliminate]: 1.70001e-06
[environ_conv]: 3.80003e-06
[add_recomputation]: 3.376e-05
[cse_after_recomputation]: 2.078e-05, [1]
    [Cycle 1]: 1.543e-05, [1]
        [cse]: 9.27001e-06
[auto_monad_reorder]: 1.004e-05
[get_jit_bprop_graph]: 1.39e-06
[rewriter_after_jit_bprop_graph]: 4.80001e-06
[opt_after_jit_grad]: 0.00046081
[symbol_engine_optimizer]: 6.961e-05, [1]
    [Cycle 1]: 6.324e-05, [6]
        [build]: 2.38998e-06
        [elim_shapecalc]: 6.89999e-06
        [elim_not_effective]: 1.209e-05
        [opt_reshape]: 5.42001e-06
        [fold_const_symbol]: 8.92999e-06
        [renormalize]: 3.60014e-07
[validate]: 2.476e-05
[backend_pass]: 7.59988e-07
[task_emit]: 0.0049749
[execute]: 4.92e-06
Sums
    bootstrap                            :     0.000256s :  2.07%
    type_inference                       :     0.004045s : 32.69%
    event_method                         :     0.000015s :  0.12%
    auto_monad                           :     0.000043s :  0.35%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000001s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.22%
    rewriter_before_opt_a                :     0.000061s :  0.49%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000049s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.27%
    jit_opt_a.a_1                        :     0.000559s :  4.51%
    jit_opt_a.with_stream_mark           :     0.000019s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000472s :  3.81%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.11%
    jit_opt_a.cse                        :     0.000024s :  0.19%
    jit_opt_a.replace_applicator         :     0.000014s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000026s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000444s :  3.59%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000014s :  0.12%
    loop_unroll                          :     0.000405s :  3.27%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000009s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000034s :  0.27%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000010s :  0.08%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000461s :  3.72%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000025s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.004975s : 40.20%
    execute                              :     0.000005s :  0.04%


Time group info:
------[substitution.]   0.000151    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 1.09% :     0.000002s :      2: substitution.fold_const_symbol
 2.59% :     0.000004s :      3: substitution.graph_param_transform
82.00% :     0.000124s :      4: substitution.inline
 2.06% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.91% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.40% :     0.000004s :      2: substitution.replace_old_param
 5.85% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003997     2
81.34% :     0.003251s :      1: type_inference.infer
18.66% :     0.000746s :      1: type_inference.specialize
------[replace.]   0.000057     6
66.27% :     0.000038s :      4: replace.inline
33.73% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000129     6
93.95% :     0.000121s :      4: match.inline
 6.05% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.53% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.44% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 2.06% :     0.000002s :     10: predicate.arithmetic_simplify
 1.34% :     0.000001s :     10: predicate.cast_eliminate
 0.60% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000001s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.70% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 4.87% :     0.000005s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.28% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     10: predicate.merge_addn
 1.22% :     0.000001s :     10: predicate.minmaximum_grad
 1.40% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000002s :     15: predicate.partial_eliminate
 1.38% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     10: predicate.reduce_eliminate
 1.63% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.83% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.93% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.49% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.73% :     0.000009s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.10% :     0.000003s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000515    12
46.44% :     0.000239s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.01% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.55% :     0.000250s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017427    76
 0.21% :     0.000037s :      1: add_recomputation
 0.27% :     0.000047s :      1: auto_monad
 0.07% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.53% :     0.000267s :      1: bootstrap
 0.10% :     0.000017s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.13% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
27.40% :     0.004774s :      1: jit_opt_a
 0.81% :     0.000142s :      1: jit_opt_after_cconv
 0.29% :     0.000050s :      1: jit_opt_b
 2.37% :     0.000413s :      1: loop_unroll
 2.59% :     0.000452s :      1: mutable_eliminate
 4.29% :     0.000748s :     26: opt.transform.jit_opt_a
 0.25% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.69% :     0.000468s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000003s :      1: pre_auto_parallel
 0.17% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000011s :      1: remove_dup_value
 1.33% :     0.000231s :      1: renormalize.infer
 1.35% :     0.000235s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000029s :      1: rewriter_after_opt_a
 0.37% :     0.000064s :      1: rewriter_before_opt_a
 0.41% :     0.000072s :      1: symbol_engine_optimizer
28.59% :     0.004983s :      1: task_emit
23.27% :     0.004055s :      1: type_inference
 0.23% :     0.000039s :      1: validate


TotalTime = 0.0276027, [33]
[bootstrap]: 0.00017486
[type_inference]: 0.0088802
[event_method]: 3.93e-05
[auto_monad]: 7.003e-05
[graph_reusing]: 4.75999e-06
[pre_auto_parallel]: 1.84998e-06
[py_interpret_to_execute]: 2.281e-05
[rewriter_before_opt_a]: 8.351e-05
[expand_dump_flag]: 2.02001e-06
[jit_opt_a]: 0.0101338, [3]
    [Cycle 1]: 0.00499397, [27]
        [switch_simplify]: 5.194e-05
        [loop_unroll]: 3.914e-05
        [a_1]: 0.00078259
        [with_stream_mark]: 1.952e-05
        [recompute_prepare]: 1.867e-05
        [updatestate_depend_eliminate]: 7.11001e-06
        [updatestate_assign_eliminate]: 5.89999e-06
        [updatestate_loads_eliminate]: 5.61e-06
        [parameter_eliminate]: 1.65001e-06
        [specialize_transform]: 1.393e-05
        [updatestate_useless_node_eliminater]: 1.219e-05
        [accelerated_algorithm]: 1.237e-05
        [meta_shard_fg_expand]: 2.75002e-06
        [get_grad_eliminate_]: 1.217e-05
        [merge_forward]: 6.84001e-06
        [cell_reuse_recompute_pass]: 8.2e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.415e-05
        [j_node_and_user_rematch]: 2.297e-05
        [meta_fg_expand]: 0.00135707
        [replace_old_param]: 4.877e-05
        [inline_without_move]: 6.183e-05
        [renormalize]: 0.00205855
        [add_forward_monad_depend]: 7.68001e-06
        [auto_monad_grad]: 4.36002e-06
        [auto_monad_eliminator]: 4.257e-05
        [cse]: 0.00013318
        [replace_applicator]: 6.238e-05
    [Cycle 2]: 0.00188092, [27]
        [switch_simplify]: 3.542e-05
        [loop_unroll]: 3.435e-05
        [a_1]: 0.0010184
        [with_stream_mark]: 1.087e-05
        [recompute_prepare]: 8.02e-06
        [updatestate_depend_eliminate]: 3.36999e-06
        [updatestate_assign_eliminate]: 3.01999e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 6.91001e-06
        [updatestate_useless_node_eliminater]: 6.19001e-06
        [accelerated_algorithm]: 5.84999e-06
        [meta_shard_fg_expand]: 1.82999e-06
        [get_grad_eliminate_]: 6.17999e-06
        [merge_forward]: 3.01001e-06
        [cell_reuse_recompute_pass]: 7.50006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.318e-05
        [j_node_and_user_rematch]: 1.023e-05
        [meta_fg_expand]: 3.307e-05
        [replace_old_param]: 9.72999e-06
        [inline_without_move]: 6.97997e-06
        [renormalize]: 0.00046852
        [add_forward_monad_depend]: 4.42e-06
        [auto_monad_grad]: 1.19e-06
        [auto_monad_eliminator]: 1.078e-05
        [cse]: 1.999e-05
        [replace_applicator]: 1.254e-05
    [Cycle 3]: 0.00036852, [27]
        [switch_simplify]: 7.06999e-06
        [loop_unroll]: 6.09001e-06
        [a_1]: 0.00012378
        [with_stream_mark]: 8.71002e-06
        [recompute_prepare]: 6.19999e-06
        [updatestate_depend_eliminate]: 3.48e-06
        [updatestate_assign_eliminate]: 2.79999e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [parameter_eliminate]: 8.39995e-07
        [specialize_transform]: 6.35002e-06
        [updatestate_useless_node_eliminater]: 5.78997e-06
        [accelerated_algorithm]: 5.72999e-06
        [meta_shard_fg_expand]: 1.24998e-06
        [get_grad_eliminate_]: 5.59e-06
        [merge_forward]: 3.09999e-06
        [cell_reuse_recompute_pass]: 1.70001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.405e-05
        [j_node_and_user_rematch]: 9.72001e-06
        [meta_fg_expand]: 2.07999e-06
        [replace_old_param]: 7.97e-06
        [inline_without_move]: 6.03998e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 6.27001e-06
        [cse]: 1.363e-05
        [replace_applicator]: 6.15002e-06
[py_interpret_to_execute_after_opt_a]: 9.67999e-06
[rewriter_after_opt_a]: 2.982e-05
[convert_after_rewriter]: 6.84999e-06
[order_py_execute_after_rewriter]: 5.86998e-06
[mutable_eliminate]: 0.00051258
[jit_opt_b]: 5.627e-05, [1]
    [Cycle 1]: 5.016e-05, [2]
        [frontend_op_eliminate]: 1.935e-05
        [inline_after_opt_a]: 1.96e-05
[cconv]: 1.817e-05
[loop_unroll]: 0.00046521
[jit_opt_after_cconv]: 0.00015475, [1]
    [Cycle 1]: 0.00014863, [11]
        [c_1]: 2.809e-05
        [parameter_eliminate]: 2.61e-06
        [updatestate_depend_eliminate]: 5.96e-06
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 2.46998e-06
        [cse]: 2.205e-05
        [call_graph_tuple_transform]: 1.959e-05
        [tuple_list_get_item_eliminator]: 6.30002e-06
        [none_parameter_eliminate]: 9.79984e-07
        [renormalize]: 4.60015e-07
        [switch_simplify]: 6.93e-06
[remove_dup_value]: 1.21e-05
[partial_unused_args_eliminate]: 1.77999e-06
[environ_conv]: 4.48001e-06
[add_recomputation]: 4.704e-05
[cse_after_recomputation]: 2.624e-05, [1]
    [Cycle 1]: 2.041e-05, [1]
        [cse]: 1.427e-05
[auto_monad_reorder]: 1.563e-05
[get_jit_bprop_graph]: 1.54e-06
[rewriter_after_jit_bprop_graph]: 5.04e-06
[opt_after_jit_grad]: 0.00046685
[symbol_engine_optimizer]: 7.766e-05, [1]
    [Cycle 1]: 7.178e-05, [6]
        [build]: 2.96999e-06
        [elim_shapecalc]: 9.09e-06
        [elim_not_effective]: 1.452e-05
        [opt_reshape]: 6.88e-06
        [fold_const_symbol]: 1.066e-05
        [renormalize]: 3.60014e-07
[validate]: 2.929e-05
[backend_pass]: 7.60017e-07
[task_emit]: 0.00606114
[execute]: 5.92999e-06
Sums
    bootstrap                            :     0.000175s :  0.73%
    type_inference                       :     0.008880s : 37.02%
    event_method                         :     0.000039s :  0.16%
    auto_monad                           :     0.000070s :  0.29%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000023s :  0.10%
    rewriter_before_opt_a                :     0.000084s :  0.35%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000094s :  0.39%
    jit_opt_a.loop_unroll                :     0.000080s :  0.33%
    jit_opt_a.a_1                        :     0.001925s :  8.02%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000027s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.10%
    jit_opt_a.merge_forward              :     0.000013s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000051s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000043s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001392s :  5.80%
    jit_opt_a.replace_old_param          :     0.000066s :  0.28%
    jit_opt_a.inline_without_move        :     0.000075s :  0.31%
    jit_opt_a.renormalize                :     0.002527s : 10.53%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000006s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000060s :  0.25%
    jit_opt_a.cse                        :     0.000167s :  0.70%
    jit_opt_a.replace_applicator         :     0.000081s :  0.34%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.04%
    rewriter_after_opt_a                 :     0.000030s :  0.12%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000513s :  2.14%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.08%
    cconv                                :     0.000018s :  0.08%
    loop_unroll                          :     0.000465s :  1.94%
    jit_opt_after_cconv.c_1              :     0.000028s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000012s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000047s :  0.20%
    cse_after_recomputation.cse          :     0.000014s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000467s :  1.95%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006061s : 25.27%
    execute                              :     0.000006s :  0.02%


Time group info:
------[substitution.]   0.000406   110
 0.49% :     0.000002s :      3: substitution.elim_not_effective
 0.39% :     0.000002s :      3: substitution.fold_const_symbol
 1.02% :     0.000004s :      4: substitution.graph_param_transform
61.25% :     0.000249s :     12: substitution.inline
 4.00% :     0.000016s :      2: substitution.inline_without_move
 1.99% :     0.000008s :     14: substitution.j_node_and_user_rematch
 2.46% :     0.000010s :      7: substitution.minmaximum_grad
 1.54% :     0.000006s :      5: substitution.partial_eliminate
 2.51% :     0.000010s :     14: substitution.remove_not_recompute_node
 5.90% :     0.000024s :     10: substitution.replace_applicator
 1.55% :     0.000006s :      7: substitution.replace_old_param
 0.62% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.92% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.38% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.98% :     0.000032s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008828     2
89.82% :     0.007929s :      1: type_inference.infer
10.18% :     0.000899s :      1: type_inference.specialize
------[replace.]   0.000135    19
62.61% :     0.000085s :     12: replace.inline
37.39% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000256    19
94.36% :     0.000242s :     12: match.inline
 5.64% :     0.000014s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000370  2568
 1.61% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.47% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.43% :     0.000005s :     43: predicate.addn_check_dump
 1.66% :     0.000006s :     43: predicate.addn_zero_filter
 2.02% :     0.000007s :     43: predicate.arithmetic_simplify
 1.50% :     0.000006s :     43: predicate.cast_eliminate
 0.18% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.43% :     0.000005s :     43: predicate.compare_switch_simplify
 1.56% :     0.000006s :     43: predicate.depend_value_elim
 1.43% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.50% :     0.000006s :     43: predicate.dict_get_item_eliminator
 1.46% :     0.000005s :     43: predicate.dict_set_item_eliminator
 0.32% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.14% :     0.000000s :      4: predicate.elim_not_effective
 0.20% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000005s :     43: predicate.environ_add_const_eliminate
 1.44% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.45% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.48% :     0.000005s :     43: predicate.environ_get_eliminate
 1.50% :     0.000006s :     43: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.84% :     0.000003s :     19: predicate.get_grad_eliminate
 0.11% :     0.000000s :      4: predicate.graph_param_transform
 4.27% :     0.000016s :     70: predicate.inline
 1.94% :     0.000007s :     40: predicate.inline_without_move
 0.35% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.83% :     0.000003s :     19: predicate.less_batch_normalization
 1.78% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.92% :     0.000007s :     54: predicate.load_eliminater
 0.45% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.22% :     0.000012s :     85: predicate.loop_unroll_before_grad
 1.84% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.46% :     0.000005s :     43: predicate.merge_addn
 1.52% :     0.000006s :     43: predicate.minmaximum_grad
 0.55% :     0.000002s :      4: predicate.mutable_eliminate
 0.20% :     0.000001s :      4: predicate.opt_reshape
 2.42% :     0.000009s :     54: predicate.partial_eliminate
 1.59% :     0.000006s :     43: predicate.print_const_string_wrapper
 2.02% :     0.000007s :     43: predicate.reduce_eliminate
 1.78% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.48% :     0.000009s :     96: predicate.replace_applicator
 0.90% :     0.000003s :     40: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.69% :     0.000006s :     43: predicate.reshape_eliminate
 1.66% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.28% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.53% :     0.000006s :     43: predicate.same_eliminate
 0.49% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.45% :     0.000002s :      8: predicate.special_op_eliminate
 0.94% :     0.000003s :     19: predicate.specialize_transform
 1.84% :     0.000007s :     43: predicate.split_environ_get_set_with_tuple_value
 1.71% :     0.000006s :     43: predicate.stack_unstack_eliminate
 0.18% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.78% :     0.000010s :     62: predicate.switch_defer_inline
 2.70% :     0.000010s :     62: predicate.switch_layer_defer_inline
 6.17% :     0.000023s :    151: predicate.switch_simplify
 1.55% :     0.000006s :     43: predicate.tile_eliminate
 1.54% :     0.000006s :     43: predicate.transpose_eliminate
 1.85% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.70% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.00% :     0.000011s :     58: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000007s :     43: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000007s :     50: predicate.tuple_to_list_eliminator_
 1.96% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000011s :     73: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000007s :     43: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.20% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001276    29
58.70% :     0.000749s :     13: func_graph_cloner_run.FuncGraphClonerGraph
41.30% :     0.000527s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.032775    91
 0.15% :     0.000050s :      1: add_recomputation
 0.23% :     0.000074s :      1: auto_monad
 0.06% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.57% :     0.000187s :      1: bootstrap
 0.06% :     0.000021s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000029s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.13% :     0.000043s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.01% :     0.000004s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
30.93% :     0.010137s :      1: jit_opt_a
 0.48% :     0.000158s :      1: jit_opt_after_cconv
 0.18% :     0.000059s :      1: jit_opt_b
 1.44% :     0.000471s :      1: loop_unroll
 1.59% :     0.000520s :      1: mutable_eliminate
 7.65% :     0.002509s :     39: opt.transform.jit_opt_a
 0.18% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000032s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.44% :     0.000473s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.08% :     0.000025s :      1: py_interpret_to_execute
 0.04% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000015s :      1: remove_dup_value
 3.93% :     0.001286s :      2: renormalize.infer
 3.75% :     0.001229s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000033s :      1: rewriter_after_opt_a
 0.26% :     0.000086s :      1: rewriter_before_opt_a
 0.24% :     0.000080s :      1: symbol_engine_optimizer
18.53% :     0.006074s :      1: task_emit
27.12% :     0.008889s :      1: type_inference
 0.14% :     0.000047s :      1: validate


TotalTime = 0.0180466, [33]
[bootstrap]: 0.00027249
[type_inference]: 0.00438231
[event_method]: 1.713e-05
[auto_monad]: 6.225e-05
[graph_reusing]: 5.77999e-06
[pre_auto_parallel]: 2.25002e-06
[py_interpret_to_execute]: 3.614e-05
[rewriter_before_opt_a]: 6.706e-05
[expand_dump_flag]: 2.71999e-06
[jit_opt_a]: 0.00521186, [2]
    [Cycle 1]: 0.00143437, [27]
        [switch_simplify]: 4.962e-05
        [loop_unroll]: 2.924e-05
        [a_1]: 0.00049858
        [with_stream_mark]: 1.438e-05
        [recompute_prepare]: 7.39002e-06
        [updatestate_depend_eliminate]: 3.44001e-06
        [updatestate_assign_eliminate]: 2.86e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.66e-06
        [specialize_transform]: 6.16998e-06
        [updatestate_useless_node_eliminater]: 5.37001e-06
        [accelerated_algorithm]: 5.37999e-06
        [meta_shard_fg_expand]: 1.77001e-06
        [get_grad_eliminate_]: 5.27999e-06
        [merge_forward]: 3.18e-06
        [cell_reuse_recompute_pass]: 1.12999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.27e-05
        [j_node_and_user_rematch]: 9.52999e-06
        [meta_fg_expand]: 2.07001e-06
        [replace_old_param]: 8.57e-06
        [inline_without_move]: 5.37999e-06
        [renormalize]: 0.00054132
        [add_forward_monad_depend]: 5.56e-06
        [auto_monad_grad]: 1.78997e-06
        [auto_monad_eliminator]: 1.283e-05
        [cse]: 2.165e-05
        [replace_applicator]: 1.212e-05
    [Cycle 2]: 0.00033009, [27]
        [switch_simplify]: 5.88002e-06
        [loop_unroll]: 5.13002e-06
        [a_1]: 9.161e-05
        [with_stream_mark]: 8.85001e-06
        [recompute_prepare]: 4.90001e-06
        [updatestate_depend_eliminate]: 2.83e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 5.23002e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.53002e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 5.21002e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.234e-05
        [j_node_and_user_rematch]: 8.43999e-06
        [meta_fg_expand]: 1.79998e-06
        [replace_old_param]: 1.241e-05
        [inline_without_move]: 4.50001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 6.49976e-07
        [auto_monad_eliminator]: 5.46e-06
        [cse]: 1.127e-05
        [replace_applicator]: 5.35999e-06
[py_interpret_to_execute_after_opt_a]: 8.35999e-06
[rewriter_after_opt_a]: 3.006e-05
[convert_after_rewriter]: 6.27001e-06
[order_py_execute_after_rewriter]: 4.64002e-06
[mutable_eliminate]: 0.00051551
[jit_opt_b]: 4.849e-05, [1]
    [Cycle 1]: 4.155e-05, [2]
        [frontend_op_eliminate]: 1.545e-05
        [inline_after_opt_a]: 1.435e-05
[cconv]: 2.336e-05
[loop_unroll]: 0.00044595
[jit_opt_after_cconv]: 0.0001412, [1]
    [Cycle 1]: 0.0001342, [11]
        [c_1]: 2.098e-05
        [parameter_eliminate]: 2.94001e-06
        [updatestate_depend_eliminate]: 7.1e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.18998e-06
        [cse]: 1.902e-05
        [call_graph_tuple_transform]: 1.828e-05
        [tuple_list_get_item_eliminator]: 4.99998e-06
        [none_parameter_eliminate]: 2.01998e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 5.18002e-06
[remove_dup_value]: 1.314e-05
[partial_unused_args_eliminate]: 2.53998e-06
[environ_conv]: 5.25999e-06
[add_recomputation]: 4.807e-05
[cse_after_recomputation]: 1.991e-05, [1]
    [Cycle 1]: 1.419e-05, [1]
        [cse]: 8.54e-06
[auto_monad_reorder]: 1.417e-05
[get_jit_bprop_graph]: 1.19998e-06
[rewriter_after_jit_bprop_graph]: 5.40001e-06
[opt_after_jit_grad]: 0.00046497
[symbol_engine_optimizer]: 7.6e-05, [1]
    [Cycle 1]: 6.912e-05, [6]
        [build]: 3.28998e-06
        [elim_shapecalc]: 7.59002e-06
        [elim_not_effective]: 1.394e-05
        [opt_reshape]: 6.09001e-06
        [fold_const_symbol]: 9.46e-06
        [renormalize]: 4.69998e-07
[validate]: 3.76e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00588365
[execute]: 7.16001e-06
Sums
    bootstrap                            :     0.000272s :  1.94%
    type_inference                       :     0.004382s : 31.26%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000062s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.26%
    rewriter_before_opt_a                :     0.000067s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000590s :  4.21%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000021s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000541s :  3.86%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000033s :  0.23%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000030s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000516s :  3.68%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000446s :  3.18%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000048s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000465s :  3.32%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005884s : 41.98%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000186    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.66% :     0.000005s :      3: substitution.graph_param_transform
80.08% :     0.000149s :      4: substitution.inline
 1.79% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.36% :     0.000004s :      4: substitution.remove_not_recompute_node
 4.87% :     0.000009s :      2: substitution.replace_old_param
 6.50% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004323     2
80.78% :     0.003492s :      1: type_inference.infer
19.22% :     0.000831s :      1: type_inference.specialize
------[replace.]   0.000060     6
67.64% :     0.000041s :      4: replace.inline
32.36% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000157     6
93.00% :     0.000146s :      4: match.inline
 7.00% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.20% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 2.19% :     0.000003s :     10: predicate.arithmetic_simplify
 1.60% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000001s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.14% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.25% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.18% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.73% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.11% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000002s :     15: predicate.partial_eliminate
 1.46% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.32% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.25% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.77% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.42% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.51% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000579    12
49.70% :     0.000288s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.86% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.44% :     0.000263s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019495    76
 0.26% :     0.000051s :      1: add_recomputation
 0.34% :     0.000066s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.45% :     0.000282s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000003s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.75% :     0.005215s :      1: jit_opt_a
 0.74% :     0.000144s :      1: jit_opt_after_cconv
 0.26% :     0.000051s :      1: jit_opt_b
 2.34% :     0.000456s :      1: loop_unroll
 2.69% :     0.000524s :      1: mutable_eliminate
 4.11% :     0.000800s :     26: opt.transform.jit_opt_a
 0.24% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.43% :     0.000473s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000039s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.46% :     0.000285s :      1: renormalize.infer
 1.28% :     0.000249s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000033s :      1: rewriter_after_opt_a
 0.36% :     0.000070s :      1: rewriter_before_opt_a
 0.40% :     0.000078s :      1: symbol_engine_optimizer
30.24% :     0.005895s :      1: task_emit
22.54% :     0.004393s :      1: type_inference
 0.29% :     0.000057s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.870.803 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc390d0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.871.360 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0116105, [30]
[bootstrap]: 0.00027203
[type_inference]: 0.00432642
[event_method]: 1.648e-05
[auto_monad]: 6.152e-05
[graph_reusing]: 5.54998e-06
[pre_auto_parallel]: 2.09999e-06
[py_interpret_to_execute]: 3.476e-05
[rewriter_before_opt_a]: 7.016e-05
[expand_dump_flag]: 3.08998e-06
[jit_opt_a]: 0.00487865, [2]
    [Cycle 1]: 0.00139281, [27]
        [switch_simplify]: 4.826e-05
        [loop_unroll]: 2.837e-05
        [a_1]: 0.00050381
        [with_stream_mark]: 1.427e-05
        [recompute_prepare]: 7.56999e-06
        [updatestate_depend_eliminate]: 3.65998e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 1.57999e-06
        [specialize_transform]: 5.99999e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.62999e-06
        [meta_shard_fg_expand]: 2.14999e-06
        [get_grad_eliminate_]: 5.55001e-06
        [merge_forward]: 3.13e-06
        [cell_reuse_recompute_pass]: 1.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.404e-05
        [j_node_and_user_rematch]: 9.05001e-06
        [meta_fg_expand]: 2.16e-06
        [replace_old_param]: 8.49998e-06
        [inline_without_move]: 5.10999e-06
        [renormalize]: 0.00049736
        [add_forward_monad_depend]: 5.16998e-06
        [auto_monad_grad]: 1.91e-06
        [auto_monad_eliminator]: 1.329e-05
        [cse]: 2.313e-05
        [replace_applicator]: 1.056e-05
    [Cycle 2]: 0.00031644, [27]
        [switch_simplify]: 5.51998e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 8.932e-05
        [with_stream_mark]: 8.89e-06
        [recompute_prepare]: 4.90999e-06
        [updatestate_depend_eliminate]: 2.54001e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.91998e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 4.77998e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 4.98001e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.33002e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.232e-05
        [j_node_and_user_rematch]: 7.82998e-06
        [meta_fg_expand]: 1.50999e-06
        [replace_old_param]: 6.98e-06
        [inline_without_move]: 4.53999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.30999e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 4.87e-06
        [cse]: 9.76e-06
        [replace_applicator]: 5.20999e-06
[py_interpret_to_execute_after_opt_a]: 8.27e-06
[rewriter_after_opt_a]: 2.969e-05
[convert_after_rewriter]: 6.42001e-06
[order_py_execute_after_rewriter]: 4.53001e-06
[mutable_eliminate]: 0.00046779
[jit_opt_b]: 4.806e-05, [1]
    [Cycle 1]: 4.176e-05, [2]
        [frontend_op_eliminate]: 1.621e-05
        [inline_after_opt_a]: 1.437e-05
[cconv]: 2.432e-05
[loop_unroll]: 0.00040282
[jit_opt_after_cconv]: 0.00013319, [1]
    [Cycle 1]: 0.00012671, [11]
        [c_1]: 1.979e-05
        [parameter_eliminate]: 2.48998e-06
        [updatestate_depend_eliminate]: 5.29e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.546e-05
        [call_graph_tuple_transform]: 1.725e-05
        [tuple_list_get_item_eliminator]: 5.21998e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 5.36998e-06
[remove_dup_value]: 1.258e-05
[partial_unused_args_eliminate]: 1.97001e-06
[environ_conv]: 4.84998e-06
[add_recomputation]: 4.25e-05
[cse_after_recomputation]: 1.934e-05, [1]
    [Cycle 1]: 1.412e-05, [1]
        [cse]: 8.60001e-06
[auto_monad_reorder]: 1.333e-05
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 4.82998e-06
[opt_after_jit_grad]: 0.00045356
[symbol_engine_optimizer]: 7.301e-05, [1]
    [Cycle 1]: 6.641e-05, [6]
        [build]: 3.68999e-06
        [elim_shapecalc]: 7.00002e-06
        [elim_not_effective]: 1.223e-05
        [opt_reshape]: 6.16e-06
        [fold_const_symbol]: 8.91002e-06
        [renormalize]: 4.00003e-07
[validate]: 3.232e-05
Sums
    bootstrap                            :     0.000272s :  3.45%
    type_inference                       :     0.004326s : 54.84%
    event_method                         :     0.000016s :  0.21%
    auto_monad                           :     0.000062s :  0.78%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.44%
    rewriter_before_opt_a                :     0.000070s :  0.89%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000054s :  0.68%
    jit_opt_a.loop_unroll                :     0.000033s :  0.42%
    jit_opt_a.a_1                        :     0.000593s :  7.52%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000497s :  6.30%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.23%
    jit_opt_a.cse                        :     0.000033s :  0.42%
    jit_opt_a.replace_applicator         :     0.000016s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.10%
    rewriter_after_opt_a                 :     0.000030s :  0.38%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000468s :  5.93%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.18%
    cconv                                :     0.000024s :  0.31%
    loop_unroll                          :     0.000403s :  5.11%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000015s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000013s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000043s :  0.54%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000013s :  0.17%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000454s :  5.75%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000032s :  0.41%


Time group info:
------[substitution.]   0.000173    23
 1.07% :     0.000002s :      2: substitution.elim_not_effective
 0.79% :     0.000001s :      2: substitution.fold_const_symbol
 2.66% :     0.000005s :      3: substitution.graph_param_transform
82.67% :     0.000143s :      4: substitution.inline
 1.95% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.57% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.84% :     0.000003s :      2: substitution.replace_old_param
 6.44% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004269     2
77.65% :     0.003315s :      1: type_inference.infer
22.35% :     0.000954s :      1: type_inference.specialize
------[replace.]   0.000063     6
68.27% :     0.000043s :      4: replace.inline
31.73% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000151     6
93.30% :     0.000141s :      4: match.inline
 6.70% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.68% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.55% :     0.000002s :     10: predicate.addn_zero_filter
 1.80% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.53% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.19% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000001s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.93% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.69% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.30% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.58% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.49% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.62% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.48% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.32% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.37% :     0.000008s :     53: predicate.switch_simplify
 1.42% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.86% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.81% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.98% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000644    12
46.67% :     0.000300s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.06% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.27% :     0.000317s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012959    72
 0.35% :     0.000045s :      1: add_recomputation
 0.50% :     0.000065s :      1: auto_monad
 0.12% :     0.000016s :      1: auto_monad_reorder
 2.19% :     0.000284s :      1: bootstrap
 0.21% :     0.000027s :      1: cconv
 0.08% :     0.000011s :      1: convert_after_rewriter
 0.17% :     0.000021s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
37.67% :     0.004881s :      1: jit_opt_a
 1.05% :     0.000136s :      1: jit_opt_after_cconv
 0.39% :     0.000051s :      1: jit_opt_b
 3.17% :     0.000410s :      1: loop_unroll
 3.67% :     0.000476s :      1: mutable_eliminate
 6.11% :     0.000792s :     26: opt.transform.jit_opt_a
 0.34% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000024s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.56% :     0.000461s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.29% :     0.000037s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000015s :      1: remove_dup_value
 1.80% :     0.000233s :      1: renormalize.infer
 1.98% :     0.000257s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000033s :      1: rewriter_after_opt_a
 0.56% :     0.000073s :      1: rewriter_before_opt_a
 0.58% :     0.000076s :      1: symbol_engine_optimizer
33.48% :     0.004339s :      1: type_inference


TotalTime = 0.015803, [33]
[bootstrap]: 0.00024901
[type_inference]: 0.00393161
[event_method]: 1.537e-05
[auto_monad]: 4.496e-05
[graph_reusing]: 3.74002e-06
[pre_auto_parallel]: 2.05002e-06
[py_interpret_to_execute]: 2.739e-05
[rewriter_before_opt_a]: 6.194e-05
[expand_dump_flag]: 2.68998e-06
[jit_opt_a]: 0.00470725, [2]
    [Cycle 1]: 0.00138168, [27]
        [switch_simplify]: 4.301e-05
        [loop_unroll]: 2.819e-05
        [a_1]: 0.00045651
        [with_stream_mark]: 1.045e-05
        [recompute_prepare]: 6.59999e-06
        [updatestate_depend_eliminate]: 2.74001e-06
        [updatestate_assign_eliminate]: 2.03002e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 5.87999e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 5.77001e-06
        [meta_shard_fg_expand]: 1.23002e-06
        [get_grad_eliminate_]: 5.02999e-06
        [merge_forward]: 2.31e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.031e-05
        [j_node_and_user_rematch]: 8.25e-06
        [meta_fg_expand]: 1.90001e-06
        [replace_old_param]: 7.73999e-06
        [inline_without_move]: 5.39e-06
        [renormalize]: 0.00056849
        [add_forward_monad_depend]: 4.82998e-06
        [auto_monad_grad]: 1.61002e-06
        [auto_monad_eliminator]: 1.033e-05
        [cse]: 1.603e-05
        [replace_applicator]: 1.155e-05
    [Cycle 2]: 0.0003139, [27]
        [switch_simplify]: 5.48002e-06
        [loop_unroll]: 4.79998e-06
        [a_1]: 8.911e-05
        [with_stream_mark]: 8.22e-06
        [recompute_prepare]: 4.75999e-06
        [updatestate_depend_eliminate]: 2.27001e-06
        [updatestate_assign_eliminate]: 2.08998e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 9.09989e-07
        [specialize_transform]: 4.82e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 4.90999e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.44003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.177e-05
        [j_node_and_user_rematch]: 7.93001e-06
        [meta_fg_expand]: 2.14e-06
        [replace_old_param]: 7.68001e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 7.99977e-07
        [auto_monad_eliminator]: 5.00999e-06
        [cse]: 9.95002e-06
        [replace_applicator]: 4.87e-06
[py_interpret_to_execute_after_opt_a]: 7.70998e-06
[rewriter_after_opt_a]: 2.558e-05
[convert_after_rewriter]: 8.03001e-06
[order_py_execute_after_rewriter]: 4.62e-06
[mutable_eliminate]: 0.00046466
[jit_opt_b]: 4.93e-05, [1]
    [Cycle 1]: 4.329e-05, [2]
        [frontend_op_eliminate]: 1.667e-05
        [inline_after_opt_a]: 1.477e-05
[cconv]: 1.67e-05
[loop_unroll]: 0.00040689
[jit_opt_after_cconv]: 0.00013, [1]
    [Cycle 1]: 0.00012372, [11]
        [c_1]: 1.997e-05
        [parameter_eliminate]: 2.39001e-06
        [updatestate_depend_eliminate]: 5.02e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.569e-05
        [call_graph_tuple_transform]: 1.679e-05
        [tuple_list_get_item_eliminator]: 4.99003e-06
        [none_parameter_eliminate]: 8.39995e-07
        [renormalize]: 3.80009e-07
        [switch_simplify]: 5.25999e-06
[remove_dup_value]: 8.31002e-06
[partial_unused_args_eliminate]: 1.62001e-06
[environ_conv]: 3.7e-06
[add_recomputation]: 3.398e-05
[cse_after_recomputation]: 1.977e-05, [1]
    [Cycle 1]: 1.421e-05, [1]
        [cse]: 8.50001e-06
[auto_monad_reorder]: 1.074e-05
[get_jit_bprop_graph]: 1.42999e-06
[rewriter_after_jit_bprop_graph]: 4.72e-06
[opt_after_jit_grad]: 0.00042529
[symbol_engine_optimizer]: 6.619e-05, [1]
    [Cycle 1]: 6.069e-05, [6]
        [build]: 2.78e-06
        [elim_shapecalc]: 6.32001e-06
        [elim_not_effective]: 1.171e-05
        [opt_reshape]: 5.15999e-06
        [fold_const_symbol]: 7.87e-06
        [renormalize]: 4.89992e-07
[validate]: 2.489e-05
[backend_pass]: 7.2e-07
[task_emit]: 0.00487725
[execute]: 4.3e-06
Sums
    bootstrap                            :     0.000249s :  2.03%
    type_inference                       :     0.003932s : 32.11%
    event_method                         :     0.000015s :  0.13%
    auto_monad                           :     0.000045s :  0.37%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000027s :  0.22%
    rewriter_before_opt_a                :     0.000062s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000048s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.27%
    jit_opt_a.a_1                        :     0.000546s :  4.46%
    jit_opt_a.with_stream_mark           :     0.000019s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.13%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000569s :  4.64%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.13%
    jit_opt_a.cse                        :     0.000026s :  0.21%
    jit_opt_a.replace_applicator         :     0.000016s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000026s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000465s :  3.79%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000017s :  0.14%
    loop_unroll                          :     0.000407s :  3.32%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000008s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000034s :  0.28%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000011s :  0.09%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000425s :  3.47%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000025s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.004877s : 39.83%
    execute                              :     0.000004s :  0.04%


Time group info:
------[substitution.]   0.000144    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.87% :     0.000001s :      2: substitution.fold_const_symbol
 2.77% :     0.000004s :      3: substitution.graph_param_transform
81.98% :     0.000118s :      4: substitution.inline
 1.90% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.68% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.65% :     0.000004s :      2: substitution.replace_old_param
 6.05% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003883     2
81.67% :     0.003172s :      1: type_inference.infer
18.33% :     0.000712s :      1: type_inference.specialize
------[replace.]   0.000058     6
66.40% :     0.000038s :      4: replace.inline
33.60% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000124     6
93.76% :     0.000116s :      4: match.inline
 6.24% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.61% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000001s :     10: predicate.addn_zero_filter
 1.88% :     0.000002s :     10: predicate.arithmetic_simplify
 1.35% :     0.000001s :     10: predicate.cast_eliminate
 0.64% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000001s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.44% :     0.000000s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.00% :     0.000005s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.51% :     0.000002s :      6: predicate.less_batch_normalization
 1.46% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.13% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.40% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000002s :     15: predicate.partial_eliminate
 1.62% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.71% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.11% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.14% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.50% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.56% :     0.000008s :     53: predicate.switch_simplify
 1.33% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.34% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.08% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000491    12
47.31% :     0.000232s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.92% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.76% :     0.000235s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017211    76
 0.21% :     0.000037s :      1: add_recomputation
 0.28% :     0.000048s :      1: auto_monad
 0.08% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.51% :     0.000260s :      1: bootstrap
 0.11% :     0.000019s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.13% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.04% :     0.000007s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
27.37% :     0.004710s :      1: jit_opt_a
 0.77% :     0.000133s :      1: jit_opt_after_cconv
 0.30% :     0.000052s :      1: jit_opt_b
 2.41% :     0.000414s :      1: loop_unroll
 2.75% :     0.000473s :      1: mutable_eliminate
 4.28% :     0.000736s :     26: opt.transform.jit_opt_a
 0.25% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000028s :      4: opt.transform.symbol_engine_opt
 2.51% :     0.000432s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000011s :      1: remove_dup_value
 1.37% :     0.000237s :      1: renormalize.infer
 1.89% :     0.000326s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000028s :      1: rewriter_after_opt_a
 0.38% :     0.000065s :      1: rewriter_before_opt_a
 0.40% :     0.000068s :      1: symbol_engine_optimizer
28.38% :     0.004884s :      1: task_emit
22.90% :     0.003941s :      1: type_inference
 0.23% :     0.000040s :      1: validate


TotalTime = 0.0271765, [33]
[bootstrap]: 0.00018489
[type_inference]: 0.00870508
[event_method]: 3.87e-05
[auto_monad]: 6.868e-05
[graph_reusing]: 4.80001e-06
[pre_auto_parallel]: 2.09999e-06
[py_interpret_to_execute]: 2.176e-05
[rewriter_before_opt_a]: 8.04e-05
[expand_dump_flag]: 2.17999e-06
[jit_opt_a]: 0.010017, [3]
    [Cycle 1]: 0.00489659, [27]
        [switch_simplify]: 5.225e-05
        [loop_unroll]: 4.049e-05
        [a_1]: 0.00081373
        [with_stream_mark]: 1.984e-05
        [recompute_prepare]: 1.791e-05
        [updatestate_depend_eliminate]: 7.35e-06
        [updatestate_assign_eliminate]: 5.67999e-06
        [updatestate_loads_eliminate]: 6.01e-06
        [parameter_eliminate]: 1.83002e-06
        [specialize_transform]: 1.379e-05
        [updatestate_useless_node_eliminater]: 1.236e-05
        [accelerated_algorithm]: 1.259e-05
        [meta_shard_fg_expand]: 2.46e-06
        [get_grad_eliminate_]: 1.195e-05
        [merge_forward]: 7.29001e-06
        [cell_reuse_recompute_pass]: 7.40023e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.39e-05
        [j_node_and_user_rematch]: 2.21e-05
        [meta_fg_expand]: 0.00131504
        [replace_old_param]: 5.009e-05
        [inline_without_move]: 5.046e-05
        [renormalize]: 0.00198159
        [add_forward_monad_depend]: 7.93001e-06
        [auto_monad_grad]: 4.60001e-06
        [auto_monad_eliminator]: 4.189e-05
        [cse]: 0.00013064
        [replace_applicator]: 6.264e-05
    [Cycle 2]: 0.00182013, [27]
        [switch_simplify]: 3.488e-05
        [loop_unroll]: 3.343e-05
        [a_1]: 0.00101196
        [with_stream_mark]: 1.11e-05
        [recompute_prepare]: 7.58001e-06
        [updatestate_depend_eliminate]: 3.48e-06
        [updatestate_assign_eliminate]: 2.90998e-06
        [updatestate_loads_eliminate]: 2.62001e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 6.48998e-06
        [updatestate_useless_node_eliminater]: 6.04001e-06
        [accelerated_algorithm]: 6.02999e-06
        [meta_shard_fg_expand]: 1.55001e-06
        [get_grad_eliminate_]: 6.16e-06
        [merge_forward]: 3.07002e-06
        [cell_reuse_recompute_pass]: 9.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.25e-05
        [j_node_and_user_rematch]: 9.71e-06
        [meta_fg_expand]: 2.861e-05
        [replace_old_param]: 9.45001e-06
        [inline_without_move]: 6.19001e-06
        [renormalize]: 0.00042097
        [add_forward_monad_depend]: 4.1e-06
        [auto_monad_grad]: 1.15999e-06
        [auto_monad_eliminator]: 1.111e-05
        [cse]: 1.971e-05
        [replace_applicator]: 1.214e-05
    [Cycle 3]: 0.00036588, [27]
        [switch_simplify]: 6.26e-06
        [loop_unroll]: 5.64e-06
        [a_1]: 0.00011815
        [with_stream_mark]: 9.18002e-06
        [recompute_prepare]: 5.85002e-06
        [updatestate_depend_eliminate]: 3.37002e-06
        [updatestate_assign_eliminate]: 2.61999e-06
        [updatestate_loads_eliminate]: 2.56998e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 6.16e-06
        [updatestate_useless_node_eliminater]: 5.87999e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 6.51999e-06
        [merge_forward]: 3.16001e-06
        [cell_reuse_recompute_pass]: 1.50999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.565e-05
        [j_node_and_user_rematch]: 9.61e-06
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 8.22e-06
        [inline_without_move]: 5.82999e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 9.09989e-07
        [auto_monad_eliminator]: 6.87002e-06
        [cse]: 1.369e-05
        [replace_applicator]: 6.05002e-06
[py_interpret_to_execute_after_opt_a]: 9.67999e-06
[rewriter_after_opt_a]: 3.138e-05
[convert_after_rewriter]: 6.71e-06
[order_py_execute_after_rewriter]: 5.17e-06
[mutable_eliminate]: 0.00047613
[jit_opt_b]: 5.752e-05, [1]
    [Cycle 1]: 5.084e-05, [2]
        [frontend_op_eliminate]: 2.159e-05
        [inline_after_opt_a]: 1.797e-05
[cconv]: 1.706e-05
[loop_unroll]: 0.00040855
[jit_opt_after_cconv]: 0.00014881, [1]
    [Cycle 1]: 0.00014268, [11]
        [c_1]: 2.617e-05
        [parameter_eliminate]: 2.47001e-06
        [updatestate_depend_eliminate]: 5.62999e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.52001e-06
        [cse]: 2.171e-05
        [call_graph_tuple_transform]: 1.91e-05
        [tuple_list_get_item_eliminator]: 6.24999e-06
        [none_parameter_eliminate]: 1.04e-06
        [renormalize]: 3.19997e-07
        [switch_simplify]: 6.06e-06
[remove_dup_value]: 1.182e-05
[partial_unused_args_eliminate]: 1.84e-06
[environ_conv]: 5.21998e-06
[add_recomputation]: 4.052e-05
[cse_after_recomputation]: 2.675e-05, [1]
    [Cycle 1]: 2.068e-05, [1]
        [cse]: 1.458e-05
[auto_monad_reorder]: 1.478e-05
[get_jit_bprop_graph]: 1.18001e-06
[rewriter_after_jit_bprop_graph]: 4.15999e-06
[opt_after_jit_grad]: 0.00043907
[symbol_engine_optimizer]: 7.781e-05, [1]
    [Cycle 1]: 7.179e-05, [6]
        [build]: 3.13e-06
        [elim_shapecalc]: 8.87999e-06
        [elim_not_effective]: 1.478e-05
        [opt_reshape]: 6.82002e-06
        [fold_const_symbol]: 1.06e-05
        [renormalize]: 3.7998e-07
[validate]: 3.014e-05
[backend_pass]: 7.79983e-07
[task_emit]: 0.00605228
[execute]: 4.89998e-06
Sums
    bootstrap                            :     0.000185s :  0.79%
    type_inference                       :     0.008705s : 37.03%
    event_method                         :     0.000039s :  0.16%
    auto_monad                           :     0.000069s :  0.29%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000022s :  0.09%
    rewriter_before_opt_a                :     0.000080s :  0.34%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000093s :  0.40%
    jit_opt_a.loop_unroll                :     0.000080s :  0.34%
    jit_opt_a.a_1                        :     0.001944s :  8.27%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000031s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.10%
    jit_opt_a.merge_forward              :     0.000014s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000052s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000041s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001346s :  5.72%
    jit_opt_a.replace_old_param          :     0.000068s :  0.29%
    jit_opt_a.inline_without_move        :     0.000062s :  0.27%
    jit_opt_a.renormalize                :     0.002403s : 10.22%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000060s :  0.25%
    jit_opt_a.cse                        :     0.000164s :  0.70%
    jit_opt_a.replace_applicator         :     0.000081s :  0.34%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.04%
    rewriter_after_opt_a                 :     0.000031s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000476s :  2.03%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.08%
    cconv                                :     0.000017s :  0.07%
    loop_unroll                          :     0.000409s :  1.74%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000012s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000041s :  0.17%
    cse_after_recomputation.cse          :     0.000015s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.02%
    opt_after_jit_grad                   :     0.000439s :  1.87%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000030s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006052s : 25.74%
    execute                              :     0.000005s :  0.02%


Time group info:
------[substitution.]   0.000435   110
 0.51% :     0.000002s :      3: substitution.elim_not_effective
 0.39% :     0.000002s :      3: substitution.fold_const_symbol
 1.12% :     0.000005s :      4: substitution.graph_param_transform
63.85% :     0.000278s :     12: substitution.inline
 3.47% :     0.000015s :      2: substitution.inline_without_move
 1.84% :     0.000008s :     14: substitution.j_node_and_user_rematch
 2.34% :     0.000010s :      7: substitution.minmaximum_grad
 1.42% :     0.000006s :      5: substitution.partial_eliminate
 2.42% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.50% :     0.000024s :     10: substitution.replace_applicator
 1.45% :     0.000006s :      7: substitution.replace_old_param
 0.52% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.48% :     0.000019s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.08% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 7.62% :     0.000033s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008654     2
89.49% :     0.007745s :      1: type_inference.infer
10.51% :     0.000909s :      1: type_inference.specialize
------[replace.]   0.000133    19
62.80% :     0.000084s :     12: replace.inline
37.20% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000287    19
94.53% :     0.000271s :     12: match.inline
 5.47% :     0.000016s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000379  2568
 1.51% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.48% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.35% :     0.000005s :     43: predicate.addn_check_dump
 1.60% :     0.000006s :     43: predicate.addn_zero_filter
 1.88% :     0.000007s :     43: predicate.arithmetic_simplify
 1.45% :     0.000005s :     43: predicate.cast_eliminate
 0.18% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.36% :     0.000005s :     43: predicate.compare_switch_simplify
 1.58% :     0.000006s :     43: predicate.depend_value_elim
 1.36% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.46% :     0.000006s :     43: predicate.dict_get_item_eliminator
 1.51% :     0.000006s :     43: predicate.dict_set_item_eliminator
 0.32% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      4: predicate.elim_not_effective
 0.22% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.41% :     0.000005s :     43: predicate.environ_add_const_eliminate
 1.39% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.40% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.61% :     0.000006s :     43: predicate.environ_get_eliminate
 1.42% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.85% :     0.000003s :     19: predicate.get_grad_eliminate
 0.11% :     0.000000s :      4: predicate.graph_param_transform
 4.27% :     0.000016s :     70: predicate.inline
 1.75% :     0.000007s :     40: predicate.inline_without_move
 0.36% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.86% :     0.000003s :     19: predicate.less_batch_normalization
 1.84% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.98% :     0.000008s :     54: predicate.load_eliminater
 0.56% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.24% :     0.000012s :     85: predicate.loop_unroll_before_grad
 1.82% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.37% :     0.000005s :     43: predicate.merge_addn
 1.41% :     0.000005s :     43: predicate.minmaximum_grad
 0.61% :     0.000002s :      4: predicate.mutable_eliminate
 0.20% :     0.000001s :      4: predicate.opt_reshape
 2.24% :     0.000008s :     54: predicate.partial_eliminate
 1.67% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.79% :     0.000007s :     43: predicate.reduce_eliminate
 1.70% :     0.000006s :     50: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.35% :     0.000009s :     96: predicate.replace_applicator
 0.88% :     0.000003s :     40: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.61% :     0.000006s :     43: predicate.reshape_eliminate
 1.48% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.51% :     0.000002s :      4: predicate.row_tensor_eliminate
 1.54% :     0.000006s :     43: predicate.same_eliminate
 0.54% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.37% :     0.000001s :      8: predicate.special_op_eliminate
 0.87% :     0.000003s :     19: predicate.specialize_transform
 1.68% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000006s :     43: predicate.stack_unstack_eliminate
 0.18% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.65% :     0.000010s :     62: predicate.switch_defer_inline
 2.48% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.05% :     0.000023s :    151: predicate.switch_simplify
 1.42% :     0.000005s :     43: predicate.tile_eliminate
 1.39% :     0.000005s :     43: predicate.transpose_eliminate
 1.84% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 4.19% :     0.000016s :     43: predicate.tuple_list_get_item_depend_reorder
 2.88% :     0.000011s :     58: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000007s :     43: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000006s :     50: predicate.tuple_to_list_eliminator_
 2.10% :     0.000008s :     54: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000011s :     73: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000007s :     43: predicate.value_based_eliminate
 0.15% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.31% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001212    29
59.75% :     0.000724s :     13: func_graph_cloner_run.FuncGraphClonerGraph
40.25% :     0.000488s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.032229    91
 0.14% :     0.000044s :      1: add_recomputation
 0.22% :     0.000072s :      1: auto_monad
 0.05% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.61% :     0.000198s :      1: bootstrap
 0.06% :     0.000020s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000029s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.13% :     0.000042s :      1: event_method
 0.03% :     0.000008s :      1: execute
 0.01% :     0.000004s :      1: expand_dump_flag
 0.01% :     0.000003s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
31.09% :     0.010020s :      1: jit_opt_a
 0.47% :     0.000151s :      1: jit_opt_after_cconv
 0.19% :     0.000060s :      1: jit_opt_b
 1.29% :     0.000416s :      1: loop_unroll
 1.50% :     0.000485s :      1: mutable_eliminate
 7.80% :     0.002514s :     39: opt.transform.jit_opt_a
 0.17% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000034s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.39% :     0.000447s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.08% :     0.000024s :      1: py_interpret_to_execute
 0.04% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000019s :      1: remove_dup_value
 3.86% :     0.001243s :      2: renormalize.infer
 3.56% :     0.001149s :      2: renormalize.specialize
 0.02% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000034s :      1: rewriter_after_opt_a
 0.26% :     0.000083s :      1: rewriter_before_opt_a
 0.25% :     0.000080s :      1: symbol_engine_optimizer
18.81% :     0.006062s :      1: task_emit
27.04% :     0.008714s :      1: type_inference
 0.15% :     0.000049s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.905.627 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc391f0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.906.167 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0174967, [33]
[bootstrap]: 0.00026211
[type_inference]: 0.00408739
[event_method]: 1.672e-05
[auto_monad]: 5.791e-05
[graph_reusing]: 4.79998e-06
[pre_auto_parallel]: 2.15002e-06
[py_interpret_to_execute]: 2.898e-05
[rewriter_before_opt_a]: 6.657e-05
[expand_dump_flag]: 2.84999e-06
[jit_opt_a]: 0.00500293, [2]
    [Cycle 1]: 0.00143446, [27]
        [switch_simplify]: 4.703e-05
        [loop_unroll]: 2.884e-05
        [a_1]: 0.00048808
        [with_stream_mark]: 1.397e-05
        [recompute_prepare]: 7.9e-06
        [updatestate_depend_eliminate]: 3.03998e-06
        [updatestate_assign_eliminate]: 2.50002e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 6.28002e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 1.66e-06
        [get_grad_eliminate_]: 6.06e-06
        [merge_forward]: 3.28998e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.337e-05
        [j_node_and_user_rematch]: 9.60001e-06
        [meta_fg_expand]: 2.14e-06
        [replace_old_param]: 8.81002e-06
        [inline_without_move]: 5.57999e-06
        [renormalize]: 0.00055416
        [add_forward_monad_depend]: 5.15999e-06
        [auto_monad_grad]: 1.75001e-06
        [auto_monad_eliminator]: 1.275e-05
        [cse]: 1.965e-05
        [replace_applicator]: 1.146e-05
    [Cycle 2]: 0.00033222, [27]
        [switch_simplify]: 5.73997e-06
        [loop_unroll]: 5.12e-06
        [a_1]: 9.19e-05
        [with_stream_mark]: 8.37e-06
        [recompute_prepare]: 5.08002e-06
        [updatestate_depend_eliminate]: 2.49001e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 5.03002e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.99e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 2.79999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.59e-05
        [j_node_and_user_rematch]: 8.37e-06
        [meta_fg_expand]: 1.77001e-06
        [replace_old_param]: 7.73999e-06
        [inline_without_move]: 7.92998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 5.22999e-06
        [cse]: 1.021e-05
        [replace_applicator]: 5.22e-06
[py_interpret_to_execute_after_opt_a]: 8.71002e-06
[rewriter_after_opt_a]: 3.329e-05
[convert_after_rewriter]: 7.55e-06
[order_py_execute_after_rewriter]: 5.26998e-06
[mutable_eliminate]: 0.00048417
[jit_opt_b]: 4.931e-05, [1]
    [Cycle 1]: 4.237e-05, [2]
        [frontend_op_eliminate]: 1.577e-05
        [inline_after_opt_a]: 1.496e-05
[cconv]: 2.275e-05
[loop_unroll]: 0.00043438
[jit_opt_after_cconv]: 0.00013609, [1]
    [Cycle 1]: 0.00012976, [11]
        [c_1]: 2.173e-05
        [parameter_eliminate]: 2.66e-06
        [updatestate_depend_eliminate]: 5.17e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 1.686e-05
        [call_graph_tuple_transform]: 1.747e-05
        [tuple_list_get_item_eliminator]: 5.29e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.03002e-06
[remove_dup_value]: 1.076e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 4.19002e-06
[add_recomputation]: 4.133e-05
[cse_after_recomputation]: 2.025e-05, [1]
    [Cycle 1]: 1.481e-05, [1]
        [cse]: 8.92e-06
[auto_monad_reorder]: 1.285e-05
[get_jit_bprop_graph]: 1.71e-06
[rewriter_after_jit_bprop_graph]: 5.35999e-06
[opt_after_jit_grad]: 0.00046717
[symbol_engine_optimizer]: 7.308e-05, [1]
    [Cycle 1]: 6.659e-05, [6]
        [build]: 3.3e-06
        [elim_shapecalc]: 7.58001e-06
        [elim_not_effective]: 1.266e-05
        [opt_reshape]: 5.89e-06
        [fold_const_symbol]: 8.08999e-06
        [renormalize]: 4.50003e-07
[validate]: 3.011e-05
[backend_pass]: 1.12e-06
[task_emit]: 0.00592011
[execute]: 8.68001e-06
Sums
    bootstrap                            :     0.000262s :  1.92%
    type_inference                       :     0.004087s : 29.89%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000058s :  0.42%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.21%
    rewriter_before_opt_a                :     0.000067s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000580s :  4.24%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.08%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.12%
    jit_opt_a.inline_without_move        :     0.000014s :  0.10%
    jit_opt_a.renormalize                :     0.000554s :  4.05%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000030s :  0.22%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.24%
    convert_after_rewriter               :     0.000008s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000484s :  3.54%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000434s :  3.18%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000011s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000041s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000013s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000467s :  3.42%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000030s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005920s : 43.30%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000173    23
 1.16% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.89% :     0.000005s :      3: substitution.graph_param_transform
82.70% :     0.000143s :      4: substitution.inline
 1.81% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.63% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.93% :     0.000003s :      2: substitution.replace_old_param
 6.19% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004037     2
82.13% :     0.003316s :      1: type_inference.infer
17.87% :     0.000722s :      1: type_inference.specialize
------[replace.]   0.000060     6
66.43% :     0.000040s :      4: replace.inline
33.57% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000150     6
93.63% :     0.000141s :      4: match.inline
 6.37% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.41% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000001s :     10: predicate.addn_zero_filter
 1.84% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.06% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.66% :     0.000006s :     22: predicate.inline
 1.02% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.86% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.96% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.07% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.63% :     0.000002s :      3: predicate.mutable_eliminate
 0.62% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000002s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.73% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000001s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.84% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.42% :     0.000008s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.48% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.14% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000498    12
46.31% :     0.000231s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.67% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.03% :     0.000239s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018944    76
 0.23% :     0.000044s :      1: add_recomputation
 0.32% :     0.000061s :      1: auto_monad
 0.08% :     0.000015s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.44% :     0.000272s :      1: bootstrap
 0.13% :     0.000025s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.42% :     0.005006s :      1: jit_opt_a
 0.73% :     0.000139s :      1: jit_opt_after_cconv
 0.27% :     0.000052s :      1: jit_opt_b
 2.34% :     0.000443s :      1: loop_unroll
 2.60% :     0.000492s :      1: mutable_eliminate
 4.17% :     0.000790s :     26: opt.transform.jit_opt_a
 0.25% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.51% :     0.000476s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000013s :      1: remove_dup_value
 1.44% :     0.000273s :      1: renormalize.infer
 1.45% :     0.000274s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000036s :      1: rewriter_after_opt_a
 0.37% :     0.000069s :      1: rewriter_before_opt_a
 0.40% :     0.000076s :      1: symbol_engine_optimizer
31.34% :     0.005938s :      1: task_emit
21.63% :     0.004098s :      1: type_inference
 0.25% :     0.000047s :      1: validate


TotalTime = 0.0217804, [33]
[bootstrap]: 0.00033913
[type_inference]: 0.00517139
[event_method]: 1.883e-05
[auto_monad]: 6.532e-05
[graph_reusing]: 5.67001e-06
[pre_auto_parallel]: 2.37999e-06
[py_interpret_to_execute]: 3.897e-05
[rewriter_before_opt_a]: 7.349e-05
[expand_dump_flag]: 2.92002e-06
[jit_opt_a]: 0.00715634, [2]
    [Cycle 1]: 0.00193329, [27]
        [switch_simplify]: 6.34e-05
        [loop_unroll]: 2.951e-05
        [a_1]: 0.0005873
        [with_stream_mark]: 2.281e-05
        [recompute_prepare]: 7.71999e-06
        [updatestate_depend_eliminate]: 3.68999e-06
        [updatestate_assign_eliminate]: 2.69999e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [parameter_eliminate]: 2.22001e-06
        [specialize_transform]: 5.56e-06
        [updatestate_useless_node_eliminater]: 5.10001e-06
        [accelerated_algorithm]: 5.16998e-06
        [meta_shard_fg_expand]: 2.16998e-06
        [get_grad_eliminate_]: 5.54e-06
        [merge_forward]: 3.45e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.45e-05
        [j_node_and_user_rematch]: 8.71002e-06
        [meta_fg_expand]: 2.14999e-06
        [replace_old_param]: 9.20001e-06
        [inline_without_move]: 5.07e-06
        [renormalize]: 0.00089061
        [add_forward_monad_depend]: 5.97999e-06
        [auto_monad_grad]: 2.68998e-06
        [auto_monad_eliminator]: 1.558e-05
        [cse]: 3.148e-05
        [replace_applicator]: 1.285e-05
    [Cycle 2]: 0.00036573, [27]
        [switch_simplify]: 6.00002e-06
        [loop_unroll]: 4.95999e-06
        [a_1]: 0.00012505
        [with_stream_mark]: 1.232e-05
        [recompute_prepare]: 5.25999e-06
        [updatestate_depend_eliminate]: 2.39001e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 1.37999e-06
        [specialize_transform]: 4.99e-06
        [updatestate_useless_node_eliminater]: 4.96002e-06
        [accelerated_algorithm]: 5.12e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 4.74e-06
        [merge_forward]: 2.72001e-06
        [cell_reuse_recompute_pass]: 1.80001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.281e-05
        [j_node_and_user_rematch]: 7.46001e-06
        [meta_fg_expand]: 1.72001e-06
        [replace_old_param]: 7.9e-06
        [inline_without_move]: 1.061e-05
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.50999e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 5.15999e-06
        [cse]: 1.207e-05
        [replace_applicator]: 4.95001e-06
[py_interpret_to_execute_after_opt_a]: 1.335e-05
[rewriter_after_opt_a]: 3.821e-05
[convert_after_rewriter]: 6.88e-06
[order_py_execute_after_rewriter]: 4.95001e-06
[mutable_eliminate]: 0.0006522
[jit_opt_b]: 4.888e-05, [1]
    [Cycle 1]: 4.182e-05, [2]
        [frontend_op_eliminate]: 1.575e-05
        [inline_after_opt_a]: 1.43e-05
[cconv]: 2.761e-05
[loop_unroll]: 0.00041852
[jit_opt_after_cconv]: 0.00013872, [1]
    [Cycle 1]: 0.00013213, [11]
        [c_1]: 2.023e-05
        [parameter_eliminate]: 2.71e-06
        [updatestate_depend_eliminate]: 5.89e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.682e-05
        [call_graph_tuple_transform]: 1.945e-05
        [tuple_list_get_item_eliminator]: 5.28002e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 5.29e-06
[remove_dup_value]: 1.41e-05
[partial_unused_args_eliminate]: 2.81e-06
[environ_conv]: 4.90999e-06
[add_recomputation]: 5.028e-05
[cse_after_recomputation]: 2.037e-05, [1]
    [Cycle 1]: 1.444e-05, [1]
        [cse]: 8.62998e-06
[auto_monad_reorder]: 1.473e-05
[get_jit_bprop_graph]: 2.09999e-06
[rewriter_after_jit_bprop_graph]: 6.12001e-06
[opt_after_jit_grad]: 0.00045906
[symbol_engine_optimizer]: 7.379e-05, [1]
    [Cycle 1]: 6.68e-05, [6]
        [build]: 4.17e-06
        [elim_shapecalc]: 7.41001e-06
        [elim_not_effective]: 1.247e-05
        [opt_reshape]: 5.65001e-06
        [fold_const_symbol]: 9.49999e-06
        [renormalize]: 5.10016e-07
[validate]: 4.023e-05
[backend_pass]: 9.40025e-07
[task_emit]: 0.00662813
[execute]: 1.076e-05
Sums
    bootstrap                            :     0.000339s :  2.08%
    type_inference                       :     0.005171s : 31.77%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000065s :  0.40%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.24%
    rewriter_before_opt_a                :     0.000073s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000069s :  0.43%
    jit_opt_a.loop_unroll                :     0.000034s :  0.21%
    jit_opt_a.a_1                        :     0.000712s :  4.38%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000016s :  0.10%
    jit_opt_a.renormalize                :     0.000891s :  5.47%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.13%
    jit_opt_a.cse                        :     0.000044s :  0.27%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000652s :  4.01%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000028s :  0.17%
    loop_unroll                          :     0.000419s :  2.57%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000459s :  2.82%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006628s : 40.72%
    execute                              :     0.000011s :  0.07%


Time group info:
------[substitution.]   0.000251    23
 0.77% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.14% :     0.000005s :      3: substitution.graph_param_transform
86.67% :     0.000217s :      4: substitution.inline
 1.15% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.75% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000005s :      2: substitution.replace_old_param
 4.93% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005102     2
81.81% :     0.004174s :      1: type_inference.infer
18.19% :     0.000928s :      1: type_inference.specialize
------[replace.]   0.000076     6
68.86% :     0.000052s :      4: replace.inline
31.14% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000226     6
95.00% :     0.000215s :      4: match.inline
 5.00% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000149   691
 0.98% :     0.000001s :     10: predicate.accumulaten_eliminater
 0.95% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.83% :     0.000001s :     10: predicate.addn_check_dump
 0.99% :     0.000001s :     10: predicate.addn_zero_filter
 2.06% :     0.000003s :     10: predicate.arithmetic_simplify
 0.99% :     0.000001s :     10: predicate.cast_eliminate
 0.34% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.83% :     0.000001s :     10: predicate.compare_switch_simplify
 1.01% :     0.000002s :     10: predicate.depend_value_elim
 0.87% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.06% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.97% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.71% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.91% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.81% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.83% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.93% :     0.000001s :     10: predicate.environ_get_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 4.69% :     0.000007s :     22: predicate.inline
 0.75% :     0.000001s :      6: predicate.inline_without_move
 0.28% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.85% :     0.000001s :      6: predicate.less_batch_normalization
 1.27% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.41% :     0.000002s :     15: predicate.load_eliminater
 1.20% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.04% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.64% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.84% :     0.000001s :     10: predicate.merge_addn
 0.91% :     0.000001s :     10: predicate.minmaximum_grad
 1.45% :     0.000002s :      3: predicate.mutable_eliminate
 0.38% :     0.000001s :      3: predicate.opt_reshape
 1.79% :     0.000003s :     15: predicate.partial_eliminate
 0.93% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.19% :     0.000002s :     10: predicate.reduce_eliminate
 1.12% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.50% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.40% :     0.000001s :      3: predicate.reset_defer_inline
 1.04% :     0.000002s :     10: predicate.reshape_eliminate
 0.95% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.06% :     0.000002s :     10: predicate.same_eliminate
 0.39% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 0.81% :     0.000001s :      6: predicate.specialize_transform
 1.13% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
22.38% :     0.000033s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 1.86% :     0.000003s :     16: predicate.switch_defer_inline
 1.65% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.21% :     0.000009s :     53: predicate.switch_simplify
 0.93% :     0.000001s :     10: predicate.tile_eliminate
 1.00% :     0.000001s :     10: predicate.transpose_eliminate
 1.03% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.87% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 2.90% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.22% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.12% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.28% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.46% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.18% :     0.000002s :     10: predicate.value_based_eliminate
 0.27% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.50% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000681    12
49.48% :     0.000337s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.12% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.40% :     0.000289s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023697    76
 0.22% :     0.000053s :      1: add_recomputation
 0.29% :     0.000069s :      1: auto_monad
 0.07% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.51% :     0.000359s :      1: bootstrap
 0.13% :     0.000030s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.07% :     0.000017s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
30.22% :     0.007160s :      1: jit_opt_a
 0.60% :     0.000142s :      1: jit_opt_after_cconv
 0.22% :     0.000051s :      1: jit_opt_b
 1.80% :     0.000427s :      1: loop_unroll
 2.79% :     0.000661s :      1: mutable_eliminate
 3.95% :     0.000935s :     26: opt.transform.jit_opt_a
 0.20% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.97% :     0.000467s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000042s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.20% :     0.000520s :      1: renormalize.infer
 1.53% :     0.000362s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000042s :      1: rewriter_after_opt_a
 0.32% :     0.000076s :      1: rewriter_before_opt_a
 0.32% :     0.000076s :      1: symbol_engine_optimizer
28.06% :     0.006650s :      1: task_emit
21.91% :     0.005192s :      1: type_inference
 0.26% :     0.000061s :      1: validate


TotalTime = 0.0301907, [33]
[bootstrap]: 0.00018034
[type_inference]: 0.00940916
[event_method]: 4.057e-05
[auto_monad]: 9.393e-05
[graph_reusing]: 6.83e-06
[pre_auto_parallel]: 2.59001e-06
[py_interpret_to_execute]: 2.702e-05
[rewriter_before_opt_a]: 9.261e-05
[expand_dump_flag]: 3.28e-06
[jit_opt_a]: 0.0114993, [3]
    [Cycle 1]: 0.00573648, [27]
        [switch_simplify]: 7.076e-05
        [loop_unroll]: 4.053e-05
        [a_1]: 0.00086921
        [with_stream_mark]: 3.286e-05
        [recompute_prepare]: 2.008e-05
        [updatestate_depend_eliminate]: 8.52e-06
        [updatestate_assign_eliminate]: 6.94999e-06
        [updatestate_loads_eliminate]: 6.21e-06
        [parameter_eliminate]: 2.56e-06
        [specialize_transform]: 1.346e-05
        [updatestate_useless_node_eliminater]: 1.222e-05
        [accelerated_algorithm]: 1.247e-05
        [meta_shard_fg_expand]: 4.05e-06
        [get_grad_eliminate_]: 1.184e-05
        [merge_forward]: 9.77001e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.777e-05
        [j_node_and_user_rematch]: 2.305e-05
        [meta_fg_expand]: 0.00155397
        [replace_old_param]: 5.488e-05
        [inline_without_move]: 5.256e-05
        [renormalize]: 0.00234591
        [add_forward_monad_depend]: 1.095e-05
        [auto_monad_grad]: 5.71e-06
        [auto_monad_eliminator]: 5.037e-05
        [cse]: 0.0001895
        [replace_applicator]: 6.696e-05
    [Cycle 2]: 0.00188962, [27]
        [switch_simplify]: 3.547e-05
        [loop_unroll]: 3.515e-05
        [a_1]: 0.00106279
        [with_stream_mark]: 1.188e-05
        [recompute_prepare]: 7.81001e-06
        [updatestate_depend_eliminate]: 4.10998e-06
        [updatestate_assign_eliminate]: 3.13998e-06
        [updatestate_loads_eliminate]: 2.69001e-06
        [parameter_eliminate]: 1.32e-06
        [specialize_transform]: 6.49001e-06
        [updatestate_useless_node_eliminater]: 5.92001e-06
        [accelerated_algorithm]: 5.80002e-06
        [meta_shard_fg_expand]: 1.70001e-06
        [get_grad_eliminate_]: 5.74e-06
        [merge_forward]: 3.38e-06
        [cell_reuse_recompute_pass]: 9.29984e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.344e-05
        [j_node_and_user_rematch]: 1.007e-05
        [meta_fg_expand]: 3.516e-05
        [replace_old_param]: 1.006e-05
        [inline_without_move]: 6.01e-06
        [renormalize]: 0.00042699
        [add_forward_monad_depend]: 4.28999e-06
        [auto_monad_grad]: 1.05999e-06
        [auto_monad_eliminator]: 1.164e-05
        [cse]: 1.913e-05
        [replace_applicator]: 1.27e-05
    [Cycle 3]: 0.00035944, [27]
        [switch_simplify]: 6.38e-06
        [loop_unroll]: 5.84e-06
        [a_1]: 0.00011898
        [with_stream_mark]: 8.25999e-06
        [recompute_prepare]: 5.84999e-06
        [updatestate_depend_eliminate]: 3.18e-06
        [updatestate_assign_eliminate]: 2.69001e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 7.59988e-07
        [specialize_transform]: 6.26e-06
        [updatestate_useless_node_eliminater]: 6.04999e-06
        [accelerated_algorithm]: 5.81003e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 5.55001e-06
        [merge_forward]: 2.92002e-06
        [cell_reuse_recompute_pass]: 1.48002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.333e-05
        [j_node_and_user_rematch]: 9.92999e-06
        [meta_fg_expand]: 2.18998e-06
        [replace_old_param]: 7.82e-06
        [inline_without_move]: 5.51e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 6.99976e-07
        [auto_monad_eliminator]: 6.49001e-06
        [cse]: 1.393e-05
        [replace_applicator]: 5.87999e-06
[py_interpret_to_execute_after_opt_a]: 1.202e-05
[rewriter_after_opt_a]: 3.717e-05
[convert_after_rewriter]: 8.64998e-06
[order_py_execute_after_rewriter]: 5.66e-06
[mutable_eliminate]: 0.00056644
[jit_opt_b]: 5.346e-05, [1]
    [Cycle 1]: 4.745e-05, [2]
        [frontend_op_eliminate]: 1.878e-05
        [inline_after_opt_a]: 1.753e-05
[cconv]: 2.211e-05
[loop_unroll]: 0.00043091
[jit_opt_after_cconv]: 0.0001578, [1]
    [Cycle 1]: 0.00015141, [11]
        [c_1]: 2.561e-05
        [parameter_eliminate]: 2.75002e-06
        [updatestate_depend_eliminate]: 5.57001e-06
        [updatestate_assign_eliminate]: 2.87002e-06
        [updatestate_loads_eliminate]: 2.81999e-06
        [cse]: 2.301e-05
        [call_graph_tuple_transform]: 1.983e-05
        [tuple_list_get_item_eliminator]: 6.23e-06
        [none_parameter_eliminate]: 1.37999e-06
        [renormalize]: 3.9002e-07
        [switch_simplify]: 7.12002e-06
[remove_dup_value]: 1.639e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 5.77001e-06
[add_recomputation]: 4.969e-05
[cse_after_recomputation]: 2.518e-05, [1]
    [Cycle 1]: 1.976e-05, [1]
        [cse]: 1.416e-05
[auto_monad_reorder]: 1.867e-05
[get_jit_bprop_graph]: 1.46002e-06
[rewriter_after_jit_bprop_graph]: 5.67001e-06
[opt_after_jit_grad]: 0.00047765
[symbol_engine_optimizer]: 8.025e-05, [1]
    [Cycle 1]: 7.408e-05, [6]
        [build]: 3.83999e-06
        [elim_shapecalc]: 9.60001e-06
        [elim_not_effective]: 1.493e-05
        [opt_reshape]: 6.97002e-06
        [fold_const_symbol]: 1.046e-05
        [renormalize]: 3.7998e-07
[validate]: 3.392e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00663851
[execute]: 7.42002e-06
Sums
    bootstrap                            :     0.000180s :  0.70%
    type_inference                       :     0.009409s : 36.34%
    event_method                         :     0.000041s :  0.16%
    auto_monad                           :     0.000094s :  0.36%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000027s :  0.10%
    rewriter_before_opt_a                :     0.000093s :  0.36%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000113s :  0.43%
    jit_opt_a.loop_unroll                :     0.000082s :  0.31%
    jit_opt_a.a_1                        :     0.002051s :  7.92%
    jit_opt_a.with_stream_mark           :     0.000053s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000034s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.09%
    jit_opt_a.merge_forward              :     0.000016s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000055s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000043s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001591s :  6.15%
    jit_opt_a.replace_old_param          :     0.000073s :  0.28%
    jit_opt_a.inline_without_move        :     0.000064s :  0.25%
    jit_opt_a.renormalize                :     0.002773s : 10.71%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000068s :  0.26%
    jit_opt_a.cse                        :     0.000223s :  0.86%
    jit_opt_a.replace_applicator         :     0.000086s :  0.33%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.05%
    rewriter_after_opt_a                 :     0.000037s :  0.14%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000566s :  2.19%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.07%
    cconv                                :     0.000022s :  0.09%
    loop_unroll                          :     0.000431s :  1.66%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000050s :  0.19%
    cse_after_recomputation.cse          :     0.000014s :  0.05%
    auto_monad_reorder                   :     0.000019s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000478s :  1.84%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006639s : 25.64%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000498   110
 0.47% :     0.000002s :      3: substitution.elim_not_effective
 0.32% :     0.000002s :      3: substitution.fold_const_symbol
 1.15% :     0.000006s :      4: substitution.graph_param_transform
63.50% :     0.000316s :     12: substitution.inline
 3.46% :     0.000017s :      2: substitution.inline_without_move
 1.88% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.22% :     0.000011s :      7: substitution.minmaximum_grad
 1.55% :     0.000008s :      5: substitution.partial_eliminate
 2.15% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.38% :     0.000027s :     10: substitution.replace_applicator
 1.70% :     0.000008s :      7: substitution.replace_old_param
 0.65% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.31% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.10% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 8.16% :     0.000041s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009343     2
89.41% :     0.008354s :      1: type_inference.infer
10.59% :     0.000990s :      1: type_inference.specialize
------[replace.]   0.000145    19
64.74% :     0.000094s :     12: replace.inline
35.26% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000331    19
93.50% :     0.000309s :     12: match.inline
 6.50% :     0.000022s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000380  2568
 1.59% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.46% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.44% :     0.000005s :     43: predicate.addn_check_dump
 1.76% :     0.000007s :     43: predicate.addn_zero_filter
 2.13% :     0.000008s :     43: predicate.arithmetic_simplify
 1.69% :     0.000006s :     43: predicate.cast_eliminate
 0.17% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.42% :     0.000005s :     43: predicate.compare_switch_simplify
 1.62% :     0.000006s :     43: predicate.depend_value_elim
 1.45% :     0.000006s :     43: predicate.dict_get_item_const_eliminator
 1.54% :     0.000006s :     43: predicate.dict_get_item_eliminator
 1.50% :     0.000006s :     43: predicate.dict_set_item_eliminator
 0.33% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      4: predicate.elim_not_effective
 0.20% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.61% :     0.000006s :     43: predicate.environ_add_const_eliminate
 1.41% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.70% :     0.000006s :     43: predicate.environ_get_depend_swap
 1.53% :     0.000006s :     43: predicate.environ_get_eliminate
 1.68% :     0.000006s :     43: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.79% :     0.000003s :     19: predicate.get_grad_eliminate
 0.13% :     0.000001s :      4: predicate.graph_param_transform
 4.55% :     0.000017s :     70: predicate.inline
 1.73% :     0.000007s :     40: predicate.inline_without_move
 0.37% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.86% :     0.000003s :     19: predicate.less_batch_normalization
 1.80% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.88% :     0.000007s :     54: predicate.load_eliminater
 0.49% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.18% :     0.000012s :     85: predicate.loop_unroll_before_grad
 1.80% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.44% :     0.000005s :     43: predicate.merge_addn
 1.50% :     0.000006s :     43: predicate.minmaximum_grad
 0.45% :     0.000002s :      4: predicate.mutable_eliminate
 0.21% :     0.000001s :      4: predicate.opt_reshape
 2.35% :     0.000009s :     54: predicate.partial_eliminate
 1.59% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.88% :     0.000007s :     43: predicate.reduce_eliminate
 1.79% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.40% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.41% :     0.000009s :     96: predicate.replace_applicator
 0.94% :     0.000004s :     40: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.80% :     0.000007s :     43: predicate.reshape_eliminate
 1.54% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.25% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.60% :     0.000006s :     43: predicate.same_eliminate
 0.48% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.36% :     0.000001s :      8: predicate.special_op_eliminate
 0.88% :     0.000003s :     19: predicate.specialize_transform
 1.75% :     0.000007s :     43: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000005s :     43: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.78% :     0.000011s :     62: predicate.switch_defer_inline
 2.52% :     0.000010s :     62: predicate.switch_layer_defer_inline
 6.33% :     0.000024s :    151: predicate.switch_simplify
 1.46% :     0.000006s :     43: predicate.tile_eliminate
 1.45% :     0.000006s :     43: predicate.transpose_eliminate
 1.79% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.69% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.13% :     0.000012s :     58: predicate.tuple_list_get_item_eliminator
 1.83% :     0.000007s :     43: predicate.tuple_list_set_item_eliminator
 1.75% :     0.000007s :     50: predicate.tuple_to_list_eliminator_
 1.86% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000011s :     73: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000007s :     43: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.24% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001289    29
60.98% :     0.000786s :     13: func_graph_cloner_run.FuncGraphClonerGraph
39.02% :     0.000503s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.035752    91
 0.15% :     0.000053s :      1: add_recomputation
 0.27% :     0.000098s :      1: auto_monad
 0.06% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.53% :     0.000189s :      1: bootstrap
 0.07% :     0.000025s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.08% :     0.000027s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.12% :     0.000045s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
32.17% :     0.011502s :      1: jit_opt_a
 0.45% :     0.000161s :      1: jit_opt_after_cconv
 0.16% :     0.000056s :      1: jit_opt_b
 1.23% :     0.000439s :      1: loop_unroll
 1.60% :     0.000573s :      1: mutable_eliminate
 7.41% :     0.002650s :     39: opt.transform.jit_opt_a
 0.15% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000030s :      4: opt.transform.jit_opt_b
 0.07% :     0.000024s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.36% :     0.000486s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000029s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 4.17% :     0.001491s :      2: renormalize.infer
 3.55% :     0.001268s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000040s :      1: rewriter_after_opt_a
 0.27% :     0.000096s :      1: rewriter_before_opt_a
 0.23% :     0.000083s :      1: symbol_engine_optimizer
18.59% :     0.006647s :      1: task_emit
26.36% :     0.009423s :      1: type_inference
 0.15% :     0.000055s :      1: validate


TotalTime = 0.0109672, [30]
[bootstrap]: 0.0002809
[type_inference]: 0.00397154
[event_method]: 1.531e-05
[auto_monad]: 4.145e-05
[graph_reusing]: 3.67002e-06
[pre_auto_parallel]: 1.69e-06
[py_interpret_to_execute]: 7.152e-05
[rewriter_before_opt_a]: 6.219e-05
[expand_dump_flag]: 2.27999e-06
[jit_opt_a]: 0.00462841, [2]
    [Cycle 1]: 0.00131554, [27]
        [switch_simplify]: 4.18e-05
        [loop_unroll]: 2.813e-05
        [a_1]: 0.00047124
        [with_stream_mark]: 1.147e-05
        [recompute_prepare]: 7.08998e-06
        [updatestate_depend_eliminate]: 2.91e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 6.13002e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 5.79999e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 5.27001e-06
        [merge_forward]: 2.44999e-06
        [cell_reuse_recompute_pass]: 8.00006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.137e-05
        [j_node_and_user_rematch]: 8.74003e-06
        [meta_fg_expand]: 1.79998e-06
        [replace_old_param]: 7.53999e-06
        [inline_without_move]: 5.09003e-06
        [renormalize]: 0.00048223
        [add_forward_monad_depend]: 4.51002e-06
        [auto_monad_grad]: 1.30999e-06
        [auto_monad_eliminator]: 9.92999e-06
        [cse]: 1.727e-05
        [replace_applicator]: 1.173e-05
    [Cycle 2]: 0.00031972, [27]
        [switch_simplify]: 5.57001e-06
        [loop_unroll]: 4.90999e-06
        [a_1]: 9.005e-05
        [with_stream_mark]: 8.72e-06
        [recompute_prepare]: 4.72e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 5.17999e-06
        [updatestate_useless_node_eliminater]: 4.96997e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.27998e-06
        [merge_forward]: 2.53998e-06
        [cell_reuse_recompute_pass]: 2.00002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.242e-05
        [j_node_and_user_rematch]: 8.17e-06
        [meta_fg_expand]: 1.65001e-06
        [replace_old_param]: 7.14001e-06
        [inline_without_move]: 4.70001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 5.00999e-06
        [cse]: 9.87001e-06
        [replace_applicator]: 5.32999e-06
[py_interpret_to_execute_after_opt_a]: 8.69e-06
[rewriter_after_opt_a]: 2.531e-05
[convert_after_rewriter]: 8.32e-06
[order_py_execute_after_rewriter]: 4.13001e-06
[mutable_eliminate]: 0.0004644
[jit_opt_b]: 4.807e-05, [1]
    [Cycle 1]: 4.231e-05, [2]
        [frontend_op_eliminate]: 1.628e-05
        [inline_after_opt_a]: 1.448e-05
[cconv]: 1.861e-05
[loop_unroll]: 0.00040483
[jit_opt_after_cconv]: 0.00013188, [1]
    [Cycle 1]: 0.00012575, [11]
        [c_1]: 2.051e-05
        [parameter_eliminate]: 2.48002e-06
        [updatestate_depend_eliminate]: 5.20001e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.563e-05
        [call_graph_tuple_transform]: 1.702e-05
        [tuple_list_get_item_eliminator]: 5.30999e-06
        [none_parameter_eliminate]: 9.89996e-07
        [renormalize]: 4.60015e-07
        [switch_simplify]: 5.24e-06
[remove_dup_value]: 8.42e-06
[partial_unused_args_eliminate]: 1.89e-06
[environ_conv]: 4.18001e-06
[add_recomputation]: 3.272e-05
[cse_after_recomputation]: 1.88e-05, [1]
    [Cycle 1]: 1.395e-05, [1]
        [cse]: 8.62e-06
[auto_monad_reorder]: 9.11998e-06
[get_jit_bprop_graph]: 1.55999e-06
[rewriter_after_jit_bprop_graph]: 4.78001e-06
[opt_after_jit_grad]: 0.00044164
[symbol_engine_optimizer]: 7.143e-05, [1]
    [Cycle 1]: 6.509e-05, [6]
        [build]: 2.72001e-06
        [elim_shapecalc]: 7.35e-06
        [elim_not_effective]: 1.26e-05
        [opt_reshape]: 5.94e-06
        [fold_const_symbol]: 8.40001e-06
        [renormalize]: 5.69999e-07
[validate]: 2.988e-05
Sums
    bootstrap                            :     0.000281s :  3.78%
    type_inference                       :     0.003972s : 53.43%
    event_method                         :     0.000015s :  0.21%
    auto_monad                           :     0.000041s :  0.56%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000072s :  0.96%
    rewriter_before_opt_a                :     0.000062s :  0.84%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000047s :  0.64%
    jit_opt_a.loop_unroll                :     0.000033s :  0.44%
    jit_opt_a.a_1                        :     0.000561s :  7.55%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000482s :  6.49%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.20%
    jit_opt_a.cse                        :     0.000027s :  0.37%
    jit_opt_a.replace_applicator         :     0.000017s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.12%
    rewriter_after_opt_a                 :     0.000025s :  0.34%
    convert_after_rewriter               :     0.000008s :  0.11%
    order_py_execute_after_rewriter      :     0.000004s :  0.06%
    mutable_eliminate                    :     0.000464s :  6.25%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.19%
    cconv                                :     0.000019s :  0.25%
    loop_unroll                          :     0.000405s :  5.45%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000008s :  0.11%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000033s :  0.44%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000009s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000442s :  5.94%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000030s :  0.40%


TotalTime = 0.0170941, [33]
[bootstrap]: 0.00031066
[type_inference]: 0.00474412
[event_method]: 1.63e-05
[auto_monad]: 6.041e-05
[graph_reusing]: 5.10001e-06
[pre_auto_parallel]: 2.11e-06
[py_interpret_to_execute]: 3.013e-05
[rewriter_before_opt_a]: 6.53e-05
[expand_dump_flag]: 3.04999e-06
[jit_opt_a]: 0.00492001, [2]
    [Cycle 1]: 0.00127396, [27]
        [switch_simplify]: 4.381e-05
        [loop_unroll]: 2.862e-05
        [a_1]: 0.00047515
        [with_stream_mark]: 1.039e-05
        [recompute_prepare]: 6.83e-06
        [updatestate_depend_eliminate]: 2.76999e-06
        [updatestate_assign_eliminate]: 2.18998e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.26002e-06
        [specialize_transform]: 6.06e-06
        [updatestate_useless_node_eliminater]: 4.72998e-06
        [accelerated_algorithm]: 6.51e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 5.20001e-06
        [merge_forward]: 2.32999e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.099e-05
        [j_node_and_user_rematch]: 9.22999e-06
        [meta_fg_expand]: 1.92001e-06
        [replace_old_param]: 7.6e-06
        [inline_without_move]: 5.30999e-06
        [renormalize]: 0.00044751
        [add_forward_monad_depend]: 4.37e-06
        [auto_monad_grad]: 1.39e-06
        [auto_monad_eliminator]: 8.42e-06
        [cse]: 1.399e-05
        [replace_applicator]: 1.099e-05
    [Cycle 2]: 0.0003, [27]
        [switch_simplify]: 5.47001e-06
        [loop_unroll]: 5.00001e-06
        [a_1]: 8.832e-05
        [with_stream_mark]: 7.9e-06
        [recompute_prepare]: 4.69002e-06
        [updatestate_depend_eliminate]: 2.43e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 9.60019e-07
        [specialize_transform]: 4.58999e-06
        [updatestate_useless_node_eliminater]: 4.55001e-06
        [accelerated_algorithm]: 4.52e-06
        [meta_shard_fg_expand]: 9.39996e-07
        [get_grad_eliminate_]: 4.25e-06
        [merge_forward]: 2.21e-06
        [cell_reuse_recompute_pass]: 1.18001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.182e-05
        [j_node_and_user_rematch]: 7.67998e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 6.36e-06
        [inline_without_move]: 4.43001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 6.29982e-07
        [auto_monad_eliminator]: 4.32e-06
        [cse]: 8.65001e-06
        [replace_applicator]: 5.15001e-06
[py_interpret_to_execute_after_opt_a]: 7.56999e-06
[rewriter_after_opt_a]: 2.511e-05
[convert_after_rewriter]: 5.61e-06
[order_py_execute_after_rewriter]: 4.19997e-06
[mutable_eliminate]: 0.00052476
[jit_opt_b]: 4.706e-05, [1]
    [Cycle 1]: 4.08e-05, [2]
        [frontend_op_eliminate]: 1.585e-05
        [inline_after_opt_a]: 1.409e-05
[cconv]: 1.488e-05
[loop_unroll]: 0.00041006
[jit_opt_after_cconv]: 0.00012986, [1]
    [Cycle 1]: 0.00012343, [11]
        [c_1]: 2.122e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 4.07e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.542e-05
        [call_graph_tuple_transform]: 1.575e-05
        [tuple_list_get_item_eliminator]: 5.17e-06
        [none_parameter_eliminate]: 9.10019e-07
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.01002e-06
[remove_dup_value]: 7.18e-06
[partial_unused_args_eliminate]: 2.02999e-06
[environ_conv]: 3.24001e-06
[add_recomputation]: 3.618e-05
[cse_after_recomputation]: 1.886e-05, [1]
    [Cycle 1]: 1.37e-05, [1]
        [cse]: 8e-06
[auto_monad_reorder]: 8.83001e-06
[get_jit_bprop_graph]: 1.40999e-06
[rewriter_after_jit_bprop_graph]: 4.33001e-06
[opt_after_jit_grad]: 0.00044726
[symbol_engine_optimizer]: 6.619e-05, [1]
    [Cycle 1]: 6.084e-05, [6]
        [build]: 2.06998e-06
        [elim_shapecalc]: 7.35003e-06
        [elim_not_effective]: 1.152e-05
        [opt_reshape]: 5.67999e-06
        [fold_const_symbol]: 7.56999e-06
        [renormalize]: 4.89992e-07
[validate]: 2.402e-05
[backend_pass]: 7.29982e-07
[task_emit]: 0.00498326
[execute]: 4.75001e-06
Sums
    bootstrap                            :     0.000311s :  2.35%
    type_inference                       :     0.004744s : 35.90%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000060s :  0.46%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.23%
    rewriter_before_opt_a                :     0.000065s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000049s :  0.37%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000563s :  4.26%
    jit_opt_a.with_stream_mark           :     0.000018s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000448s :  3.39%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.10%
    jit_opt_a.cse                        :     0.000023s :  0.17%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000025s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000525s :  3.97%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000015s :  0.11%
    loop_unroll                          :     0.000410s :  3.10%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000007s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000003s :  0.02%
    add_recomputation                    :     0.000036s :  0.27%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000009s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000447s :  3.38%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000024s :  0.18%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.004983s : 37.71%
    execute                              :     0.000005s :  0.04%


Time group info:
------[substitution.]   0.000148    23
 1.18% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.75% :     0.000004s :      3: substitution.graph_param_transform
83.18% :     0.000123s :      4: substitution.inline
 1.64% :     0.000002s :      4: substitution.j_node_and_user_rematch
 3.19% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000003s :      2: substitution.replace_old_param
 5.34% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003925     2
80.46% :     0.003158s :      1: type_inference.infer
19.54% :     0.000767s :      1: type_inference.specialize
------[replace.]   0.000060     6
65.66% :     0.000039s :      4: replace.inline
34.34% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000128     6
94.59% :     0.000121s :      4: match.inline
 5.41% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.81% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.23% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.16% :     0.000001s :     10: predicate.addn_zero_filter
 2.31% :     0.000003s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.58% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.36% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.01% :     0.000006s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.37% :     0.000002s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.14% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.21% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.57% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000002s :     15: predicate.partial_eliminate
 1.40% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.41% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.61% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.49% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.60% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.56% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.19% :     0.000008s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.51% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.00% :     0.000002s :     10: predicate.value_based_eliminate
 0.45% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000579    12
50.37% :     0.000292s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.54% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.09% :     0.000261s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012270    72
 0.29% :     0.000036s :      1: add_recomputation
 0.37% :     0.000045s :      1: auto_monad
 0.09% :     0.000011s :      1: auto_monad_reorder
 2.40% :     0.000295s :      1: bootstrap
 0.17% :     0.000021s :      1: cconv
 0.09% :     0.000011s :      1: convert_after_rewriter
 0.17% :     0.000021s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000019s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000006s :      1: graph_reusing
37.74% :     0.004631s :      1: jit_opt_a
 1.10% :     0.000135s :      1: jit_opt_after_cconv
 0.41% :     0.000051s :      1: jit_opt_b
 3.36% :     0.000412s :      1: loop_unroll
 3.85% :     0.000472s :      1: mutable_eliminate
 6.15% :     0.000755s :     26: opt.transform.jit_opt_a
 0.36% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.66% :     0.000449s :      1: opt_after_jit_grad
 0.05% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.61% :     0.000075s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000011s :      1: remove_dup_value
 1.82% :     0.000224s :      1: renormalize.infer
 2.05% :     0.000252s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000028s :      1: rewriter_after_opt_a
 0.53% :     0.000065s :      1: rewriter_before_opt_a
 0.60% :     0.000074s :      1: symbol_engine_optimizer
32.45% :     0.003982s :      1: type_inference


Time group info:
------[substitution.]   0.000153    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.42% :     0.000004s :      3: substitution.graph_param_transform
85.02% :     0.000130s :      4: substitution.inline
 1.95% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.65% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.58% :     0.000002s :      2: substitution.replace_old_param
 4.81% :     0.000007s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004690     2
81.62% :     0.003828s :      1: type_inference.infer
18.38% :     0.000862s :      1: type_inference.specialize
------[replace.]   0.000058     6
66.66% :     0.000039s :      4: replace.inline
33.34% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000134     6
95.29% :     0.000127s :      4: match.inline
 4.71% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.58% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.48% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.23% :     0.000001s :     10: predicate.addn_zero_filter
 2.00% :     0.000002s :     10: predicate.arithmetic_simplify
 1.50% :     0.000002s :     10: predicate.cast_eliminate
 0.74% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.46% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 4.78% :     0.000005s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.55% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.95% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.28% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000002s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     10: predicate.reduce_eliminate
 1.80% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.62% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.19% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.52% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.29% :     0.000008s :     53: predicate.switch_simplify
 1.27% :     0.000001s :     10: predicate.tile_eliminate
 1.25% :     0.000001s :     10: predicate.transpose_eliminate
 1.27% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.42% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.38% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 2.01% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.44% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000598    12
46.16% :     0.000276s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.27% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.57% :     0.000296s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018397    76
 0.21% :     0.000039s :      1: add_recomputation
 0.35% :     0.000064s :      1: auto_monad
 0.06% :     0.000011s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.74% :     0.000320s :      1: bootstrap
 0.09% :     0.000017s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000005s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.04% :     0.000008s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.76% :     0.004923s :      1: jit_opt_a
 0.72% :     0.000133s :      1: jit_opt_after_cconv
 0.27% :     0.000050s :      1: jit_opt_b
 2.27% :     0.000418s :      1: loop_unroll
 2.89% :     0.000532s :      1: mutable_eliminate
 4.10% :     0.000754s :     26: opt.transform.jit_opt_a
 0.24% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.47% :     0.000454s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000033s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000009s :      1: remove_dup_value
 1.21% :     0.000223s :      1: renormalize.infer
 1.19% :     0.000219s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000028s :      1: rewriter_after_opt_a
 0.37% :     0.000068s :      1: rewriter_before_opt_a
 0.37% :     0.000069s :      1: symbol_engine_optimizer
27.14% :     0.004992s :      1: task_emit
25.83% :     0.004752s :      1: type_inference
 0.21% :     0.000038s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.943.015 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc5c9d0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.943.526 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0124701, [30]
[bootstrap]: 0.00025959
[type_inference]: 0.00419411
[event_method]: 1.653e-05
[auto_monad]: 5.728e-05
[graph_reusing]: 4.72e-06
[pre_auto_parallel]: 2.07999e-06
[py_interpret_to_execute]: 3.44e-05
[rewriter_before_opt_a]: 6.848e-05
[expand_dump_flag]: 2.49999e-06
[jit_opt_a]: 0.00562895, [2]
    [Cycle 1]: 0.00168869, [27]
        [switch_simplify]: 5.932e-05
        [loop_unroll]: 3.266e-05
        [a_1]: 0.00057031
        [with_stream_mark]: 1.583e-05
        [recompute_prepare]: 7.23999e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 5.87999e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 1.88002e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 3.41999e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.27e-05
        [j_node_and_user_rematch]: 9.52999e-06
        [meta_fg_expand]: 2.17999e-06
        [replace_old_param]: 8.32003e-06
        [inline_without_move]: 5.51e-06
        [renormalize]: 0.00071443
        [add_forward_monad_depend]: 5.60001e-06
        [auto_monad_grad]: 1.48002e-06
        [auto_monad_eliminator]: 9.91e-06
        [cse]: 1.673e-05
        [replace_applicator]: 1.147e-05
    [Cycle 2]: 0.00031704, [27]
        [switch_simplify]: 5.35001e-06
        [loop_unroll]: 4.79002e-06
        [a_1]: 8.822e-05
        [with_stream_mark]: 9.57001e-06
        [recompute_prepare]: 4.94e-06
        [updatestate_depend_eliminate]: 2.53998e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 4.82998e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 5.11997e-06
        [meta_shard_fg_expand]: 1.04003e-06
        [get_grad_eliminate_]: 4.37e-06
        [merge_forward]: 2.32999e-06
        [cell_reuse_recompute_pass]: 1.45001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.249e-05
        [j_node_and_user_rematch]: 7.95e-06
        [meta_fg_expand]: 1.53002e-06
        [replace_old_param]: 7.1e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 6.30011e-07
        [auto_monad_eliminator]: 4.77998e-06
        [cse]: 1.016e-05
        [replace_applicator]: 5.08002e-06
[py_interpret_to_execute_after_opt_a]: 1.02e-05
[rewriter_after_opt_a]: 3.3e-05
[convert_after_rewriter]: 6.74001e-06
[order_py_execute_after_rewriter]: 4.94e-06
[mutable_eliminate]: 0.00057071
[jit_opt_b]: 5.172e-05, [1]
    [Cycle 1]: 4.47e-05, [2]
        [frontend_op_eliminate]: 1.792e-05
        [inline_after_opt_a]: 1.483e-05
[cconv]: 2.527e-05
[loop_unroll]: 0.00043272
[jit_opt_after_cconv]: 0.00014032, [1]
    [Cycle 1]: 0.00013365, [11]
        [c_1]: 2.087e-05
        [parameter_eliminate]: 3.21001e-06
        [updatestate_depend_eliminate]: 5.35001e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [cse]: 1.712e-05
        [call_graph_tuple_transform]: 1.925e-05
        [tuple_list_get_item_eliminator]: 5.52001e-06
        [none_parameter_eliminate]: 1.46002e-06
        [renormalize]: 6.10016e-07
        [switch_simplify]: 5.87001e-06
[remove_dup_value]: 1.232e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 4.96002e-06
[add_recomputation]: 4.339e-05
[cse_after_recomputation]: 2.051e-05, [1]
    [Cycle 1]: 1.487e-05, [1]
        [cse]: 8.82e-06
[auto_monad_reorder]: 1.345e-05
[get_jit_bprop_graph]: 1.92001e-06
[rewriter_after_jit_bprop_graph]: 5.12999e-06
[opt_after_jit_grad]: 0.00053152
[symbol_engine_optimizer]: 7.967e-05, [1]
    [Cycle 1]: 7.164e-05, [6]
        [build]: 4.74e-06
        [elim_shapecalc]: 7.25998e-06
        [elim_not_effective]: 1.455e-05
        [opt_reshape]: 6.24001e-06
        [fold_const_symbol]: 9.56e-06
        [renormalize]: 5.39992e-07
[validate]: 3.818e-05
Sums
    bootstrap                            :     0.000260s :  3.14%
    type_inference                       :     0.004194s : 50.75%
    event_method                         :     0.000017s :  0.20%
    auto_monad                           :     0.000057s :  0.69%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.42%
    rewriter_before_opt_a                :     0.000068s :  0.83%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000065s :  0.78%
    jit_opt_a.loop_unroll                :     0.000037s :  0.45%
    jit_opt_a.a_1                        :     0.000659s :  7.97%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000714s :  8.65%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.18%
    jit_opt_a.cse                        :     0.000027s :  0.33%
    jit_opt_a.replace_applicator         :     0.000017s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000033s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000571s :  6.91%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000025s :  0.31%
    loop_unroll                          :     0.000433s :  5.24%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000012s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000043s :  0.52%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000013s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000532s :  6.43%
    symbol_engine_optimizer.build        :     0.000005s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.18%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000038s :  0.46%


Time group info:
------[substitution.]   0.000221    23
 1.31% :     0.000003s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.51% :     0.000006s :      3: substitution.graph_param_transform
85.41% :     0.000189s :      4: substitution.inline
 1.46% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.14% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.45% :     0.000003s :      2: substitution.replace_old_param
 5.18% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004134     2
77.38% :     0.003199s :      1: type_inference.infer
22.62% :     0.000935s :      1: type_inference.specialize
------[replace.]   0.000078     6
64.95% :     0.000051s :      4: replace.inline
35.05% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000197     6
94.71% :     0.000186s :      4: match.inline
 5.29% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.69% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.44% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.72% :     0.000002s :     10: predicate.addn_zero_filter
 2.74% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000001s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.65% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 4.82% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.58% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.36% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 1.86% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     15: predicate.partial_eliminate
 1.45% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 1.35% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.57% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.86% :     0.000002s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.53% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 1.97% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.13% :     0.000008s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.07% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.52% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.98% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.66% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.86% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000639    12
42.91% :     0.000274s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.84% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.26% :     0.000321s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014116    72
 0.33% :     0.000046s :      1: add_recomputation
 0.43% :     0.000061s :      1: auto_monad
 0.11% :     0.000016s :      1: auto_monad_reorder
 1.92% :     0.000271s :      1: bootstrap
 0.20% :     0.000028s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000020s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
39.91% :     0.005634s :      1: jit_opt_a
 1.01% :     0.000143s :      1: jit_opt_after_cconv
 0.39% :     0.000054s :      1: jit_opt_b
 3.13% :     0.000442s :      1: loop_unroll
 4.10% :     0.000579s :      1: mutable_eliminate
 6.20% :     0.000875s :     26: opt.transform.jit_opt_a
 0.34% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000027s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.83% :     0.000541s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000037s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000015s :      1: remove_dup_value
 2.73% :     0.000385s :      1: renormalize.infer
 2.28% :     0.000322s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000036s :      1: rewriter_after_opt_a
 0.51% :     0.000071s :      1: rewriter_before_opt_a
 0.58% :     0.000082s :      1: symbol_engine_optimizer
29.81% :     0.004208s :      1: type_inference


TotalTime = 0.0246715, [33]
[bootstrap]: 0.00039008
[type_inference]: 0.00477502
[event_method]: 1.766e-05
[auto_monad]: 6.416e-05
[graph_reusing]: 5.66e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 3.73e-05
[rewriter_before_opt_a]: 7.174e-05
[expand_dump_flag]: 3.16001e-06
[jit_opt_a]: 0.00702587, [2]
    [Cycle 1]: 0.00166816, [27]
        [switch_simplify]: 6.134e-05
        [loop_unroll]: 2.924e-05
        [a_1]: 0.00053397
        [with_stream_mark]: 1.766e-05
        [recompute_prepare]: 6.85998e-06
        [updatestate_depend_eliminate]: 3.78999e-06
        [updatestate_assign_eliminate]: 3.35e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 1.83002e-06
        [specialize_transform]: 6.34001e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 5.52001e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 5.08002e-06
        [merge_forward]: 3.84002e-06
        [cell_reuse_recompute_pass]: 1.74e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.446e-05
        [j_node_and_user_rematch]: 8.94e-06
        [meta_fg_expand]: 2.05002e-06
        [replace_old_param]: 8.74e-06
        [inline_without_move]: 5.14e-06
        [renormalize]: 0.00069847
        [add_forward_monad_depend]: 6.10002e-06
        [auto_monad_grad]: 2.69999e-06
        [auto_monad_eliminator]: 1.525e-05
        [cse]: 2.794e-05
        [replace_applicator]: 1.324e-05
    [Cycle 2]: 0.00039607, [27]
        [switch_simplify]: 5.69999e-06
        [loop_unroll]: 4.90999e-06
        [a_1]: 0.00015585
        [with_stream_mark]: 1.043e-05
        [recompute_prepare]: 6.59999e-06
        [updatestate_depend_eliminate]: 3.18e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 4.72998e-06
        [updatestate_useless_node_eliminater]: 4.58999e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.38001e-06
        [merge_forward]: 2.86e-06
        [cell_reuse_recompute_pass]: 1.37999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.306e-05
        [j_node_and_user_rematch]: 8.37e-06
        [meta_fg_expand]: 1.45001e-06
        [replace_old_param]: 7.36001e-06
        [inline_without_move]: 4.70001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.24998e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 5.57999e-06
        [cse]: 1.202e-05
        [replace_applicator]: 4.92e-06
[py_interpret_to_execute_after_opt_a]: 1.253e-05
[rewriter_after_opt_a]: 3.59e-05
[convert_after_rewriter]: 7.01999e-06
[order_py_execute_after_rewriter]: 4.60999e-06
[mutable_eliminate]: 0.00072392
[jit_opt_b]: 4.952e-05, [1]
    [Cycle 1]: 4.154e-05, [2]
        [frontend_op_eliminate]: 1.588e-05
        [inline_after_opt_a]: 1.487e-05
[cconv]: 3.233e-05
[loop_unroll]: 0.00044133
[jit_opt_after_cconv]: 0.00015251, [1]
    [Cycle 1]: 0.00014538, [11]
        [c_1]: 2.244e-05
        [parameter_eliminate]: 2.98998e-06
        [updatestate_depend_eliminate]: 6.81001e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [cse]: 2.391e-05
        [call_graph_tuple_transform]: 2.049e-05
        [tuple_list_get_item_eliminator]: 5.46e-06
        [none_parameter_eliminate]: 1.47999e-06
        [renormalize]: 6.39993e-07
        [switch_simplify]: 5.35999e-06
[remove_dup_value]: 1.455e-05
[partial_unused_args_eliminate]: 2.22999e-06
[environ_conv]: 5.32999e-06
[add_recomputation]: 4.953e-05
[cse_after_recomputation]: 2.114e-05, [1]
    [Cycle 1]: 1.543e-05, [1]
        [cse]: 8.55999e-06
[auto_monad_reorder]: 1.579e-05
[get_jit_bprop_graph]: 2.19001e-06
[rewriter_after_jit_bprop_graph]: 7.13e-06
[opt_after_jit_grad]: 0.00051062
[symbol_engine_optimizer]: 7.989e-05, [1]
    [Cycle 1]: 7.271e-05, [6]
        [build]: 4.84e-06
        [elim_shapecalc]: 8.92999e-06
        [elim_not_effective]: 1.414e-05
        [opt_reshape]: 6.29001e-06
        [fold_const_symbol]: 8.59e-06
        [renormalize]: 8.70001e-07
[validate]: 4.272e-05
[backend_pass]: 1.15999e-06
[task_emit]: 0.00982816
[execute]: 8.70001e-06
Sums
    bootstrap                            :     0.000390s :  2.05%
    type_inference                       :     0.004775s : 25.05%
    event_method                         :     0.000018s :  0.09%
    auto_monad                           :     0.000064s :  0.34%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000037s :  0.20%
    rewriter_before_opt_a                :     0.000072s :  0.38%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.35%
    jit_opt_a.loop_unroll                :     0.000034s :  0.18%
    jit_opt_a.a_1                        :     0.000690s :  3.62%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000699s :  3.67%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.11%
    jit_opt_a.cse                        :     0.000040s :  0.21%
    jit_opt_a.replace_applicator         :     0.000018s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.19%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000724s :  3.80%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000032s :  0.17%
    loop_unroll                          :     0.000441s :  2.32%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.26%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000511s :  2.68%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.009828s : 51.57%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000206    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.83% :     0.000006s :      3: substitution.graph_param_transform
83.51% :     0.000172s :      4: substitution.inline
 1.85% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.98% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000004s :      2: substitution.replace_old_param
 6.26% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004709     2
82.46% :     0.003883s :      1: type_inference.infer
17.54% :     0.000826s :      1: type_inference.specialize
------[replace.]   0.000071     6
68.15% :     0.000049s :      4: replace.inline
31.85% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000181     6
93.50% :     0.000169s :      4: match.inline
 6.50% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.59% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000001s :     10: predicate.addn_zero_filter
 2.21% :     0.000003s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.66% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.13% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.53% :     0.000001s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.36% :     0.000000s :      3: predicate.graph_param_transform
 5.68% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.81% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.58% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.92% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.32% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.51% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.24% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.22% :     0.000003s :     16: predicate.switch_defer_inline
 1.99% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.60% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.34% :     0.000002s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.91% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000601    12
52.65% :     0.000317s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.95% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.39% :     0.000255s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026375    76
 0.20% :     0.000052s :      1: add_recomputation
 0.26% :     0.000067s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.56% :     0.000412s :      1: bootstrap
 0.13% :     0.000035s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000023s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.65% :     0.007030s :      1: jit_opt_a
 0.59% :     0.000155s :      1: jit_opt_after_cconv
 0.20% :     0.000052s :      1: jit_opt_b
 1.71% :     0.000450s :      1: loop_unroll
 2.78% :     0.000734s :      1: mutable_eliminate
 3.43% :     0.000904s :     26: opt.transform.jit_opt_a
 0.19% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.98% :     0.000521s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000040s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.36% :     0.000359s :      1: renormalize.infer
 1.26% :     0.000331s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000039s :      1: rewriter_after_opt_a
 0.28% :     0.000075s :      1: rewriter_before_opt_a
 0.31% :     0.000082s :      1: symbol_engine_optimizer
37.33% :     0.009845s :      1: task_emit
18.16% :     0.004791s :      1: type_inference
 0.26% :     0.000068s :      1: validate


TotalTime = 0.0241483, [33]
[bootstrap]: 0.00024906
[type_inference]: 0.00409751
[event_method]: 1.645e-05
[auto_monad]: 5.952e-05
[graph_reusing]: 5.81998e-06
[pre_auto_parallel]: 2.31e-06
[py_interpret_to_execute]: 3.281e-05
[rewriter_before_opt_a]: 7.2e-05
[expand_dump_flag]: 3.03998e-06
[jit_opt_a]: 0.00545801, [2]
    [Cycle 1]: 0.00173036, [27]
        [switch_simplify]: 5.118e-05
        [loop_unroll]: 3.165e-05
        [a_1]: 0.00053445
        [with_stream_mark]: 1.695e-05
        [recompute_prepare]: 7.56001e-06
        [updatestate_depend_eliminate]: 3.63e-06
        [updatestate_assign_eliminate]: 3.45e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.87001e-06
        [specialize_transform]: 6.43e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 6.57002e-06
        [meta_shard_fg_expand]: 1.97001e-06
        [get_grad_eliminate_]: 5.32001e-06
        [merge_forward]: 3.51999e-06
        [cell_reuse_recompute_pass]: 2.22001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.528e-05
        [j_node_and_user_rematch]: 9.79e-06
        [meta_fg_expand]: 3.26999e-06
        [replace_old_param]: 1.032e-05
        [inline_without_move]: 5.92001e-06
        [renormalize]: 0.00077597
        [add_forward_monad_depend]: 5.15999e-06
        [auto_monad_grad]: 1.71998e-06
        [auto_monad_eliminator]: 1.223e-05
        [cse]: 3.117e-05
        [replace_applicator]: 9.49999e-06
    [Cycle 2]: 0.00031588, [27]
        [switch_simplify]: 5.39e-06
        [loop_unroll]: 4.91997e-06
        [a_1]: 8.955e-05
        [with_stream_mark]: 8.40999e-06
        [recompute_prepare]: 4.63001e-06
        [updatestate_depend_eliminate]: 2.74999e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.81998e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 4.87998e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 4.82998e-06
        [meta_shard_fg_expand]: 1.26002e-06
        [get_grad_eliminate_]: 4.55001e-06
        [merge_forward]: 2.46998e-06
        [cell_reuse_recompute_pass]: 2.07001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.183e-05
        [j_node_and_user_rematch]: 7.55e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 7.15e-06
        [inline_without_move]: 4.72998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.19998e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 5.34e-06
        [cse]: 1.052e-05
        [replace_applicator]: 4.95999e-06
[py_interpret_to_execute_after_opt_a]: 8.71002e-06
[rewriter_after_opt_a]: 3.2e-05
[convert_after_rewriter]: 7.06999e-06
[order_py_execute_after_rewriter]: 5.09e-06
[mutable_eliminate]: 0.00066597
[jit_opt_b]: 5.45e-05, [1]
    [Cycle 1]: 4.708e-05, [2]
        [frontend_op_eliminate]: 2.012e-05
        [inline_after_opt_a]: 1.517e-05
[cconv]: 2.52e-05
[loop_unroll]: 0.00045251
[jit_opt_after_cconv]: 0.0001424, [1]
    [Cycle 1]: 0.00013562, [11]
        [c_1]: 1.972e-05
        [parameter_eliminate]: 2.22001e-06
        [updatestate_depend_eliminate]: 5.02e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [cse]: 2.227e-05
        [call_graph_tuple_transform]: 1.864e-05
        [tuple_list_get_item_eliminator]: 5.35999e-06
        [none_parameter_eliminate]: 1.41002e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 5.74999e-06
[remove_dup_value]: 1.34e-05
[partial_unused_args_eliminate]: 2.69999e-06
[environ_conv]: 5.82999e-06
[add_recomputation]: 4.641e-05
[cse_after_recomputation]: 2.092e-05, [1]
    [Cycle 1]: 1.507e-05, [1]
        [cse]: 8.77e-06
[auto_monad_reorder]: 1.488e-05
[get_jit_bprop_graph]: 1.82999e-06
[rewriter_after_jit_bprop_graph]: 6.40997e-06
[opt_after_jit_grad]: 0.00047555
[symbol_engine_optimizer]: 7.587e-05, [1]
    [Cycle 1]: 6.849e-05, [6]
        [build]: 4.27003e-06
        [elim_shapecalc]: 7.2e-06
        [elim_not_effective]: 1.282e-05
        [opt_reshape]: 6.02999e-06
        [fold_const_symbol]: 9.38002e-06
        [renormalize]: 5.10016e-07
[validate]: 3.694e-05
[backend_pass]: 1.22e-06
[task_emit]: 0.0118424
[execute]: 9.99001e-06
Sums
    bootstrap                            :     0.000249s :  1.24%
    type_inference                       :     0.004098s : 20.36%
    event_method                         :     0.000016s :  0.08%
    auto_monad                           :     0.000060s :  0.30%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000033s :  0.16%
    rewriter_before_opt_a                :     0.000072s :  0.36%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.28%
    jit_opt_a.loop_unroll                :     0.000037s :  0.18%
    jit_opt_a.a_1                        :     0.000624s :  3.10%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.000776s :  3.86%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.09%
    jit_opt_a.cse                        :     0.000042s :  0.21%
    jit_opt_a.replace_applicator         :     0.000014s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.04%
    rewriter_after_opt_a                 :     0.000032s :  0.16%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000666s :  3.31%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000025s :  0.13%
    loop_unroll                          :     0.000453s :  2.25%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000013s :  0.07%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000046s :  0.23%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000015s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000476s :  2.36%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.18%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.011842s : 58.84%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000204    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.85% :     0.000006s :      3: substitution.graph_param_transform
83.34% :     0.000170s :      4: substitution.inline
 1.67% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.38% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.00% :     0.000004s :      2: substitution.replace_old_param
 6.25% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004039     2
79.97% :     0.003230s :      1: type_inference.infer
20.03% :     0.000809s :      1: type_inference.specialize
------[replace.]   0.000070     6
67.00% :     0.000047s :      4: replace.inline
33.00% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000179     6
93.49% :     0.000167s :      4: match.inline
 6.51% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.37% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000001s :     10: predicate.addn_zero_filter
 2.20% :     0.000003s :     10: predicate.arithmetic_simplify
 1.37% :     0.000002s :     10: predicate.cast_eliminate
 0.71% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000001s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.44% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.26% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.12% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.50% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.35% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.24% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 2.33% :     0.000003s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000002s :     15: predicate.partial_eliminate
 1.33% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.65% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.28% :     0.000008s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.87% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.38% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.28% :     0.000003s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000551    12
47.45% :     0.000262s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.73% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.83% :     0.000258s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025860    76
 0.19% :     0.000049s :      1: add_recomputation
 0.24% :     0.000063s :      1: auto_monad
 0.07% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.01% :     0.000260s :      1: bootstrap
 0.11% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.08% :     0.000020s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
21.12% :     0.005462s :      1: jit_opt_a
 0.56% :     0.000145s :      1: jit_opt_after_cconv
 0.22% :     0.000057s :      1: jit_opt_b
 1.78% :     0.000461s :      1: loop_unroll
 2.61% :     0.000676s :      1: mutable_eliminate
 3.21% :     0.000831s :     26: opt.transform.jit_opt_a
 0.18% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000029s :      4: opt.transform.jit_opt_b
 0.04% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000032s :      4: opt.transform.symbol_engine_opt
 1.87% :     0.000482s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.14% :     0.000035s :      1: py_interpret_to_execute
 0.04% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000016s :      1: remove_dup_value
 1.77% :     0.000457s :      1: renormalize.infer
 1.21% :     0.000312s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000035s :      1: rewriter_after_opt_a
 0.29% :     0.000075s :      1: rewriter_before_opt_a
 0.30% :     0.000078s :      1: symbol_engine_optimizer
45.87% :     0.011863s :      1: task_emit
15.90% :     0.004111s :      1: type_inference
 0.22% :     0.000058s :      1: validate


TotalTime = 0.0318656, [33]
[bootstrap]: 0.00018313
[type_inference]: 0.00911152
[event_method]: 3.792e-05
[auto_monad]: 7.2e-05
[graph_reusing]: 4.70001e-06
[pre_auto_parallel]: 1.92999e-06
[py_interpret_to_execute]: 2.25e-05
[rewriter_before_opt_a]: 8.334e-05
[expand_dump_flag]: 2.22999e-06
[jit_opt_a]: 0.0106391, [3]
    [Cycle 1]: 0.00540761, [27]
        [switch_simplify]: 5.4e-05
        [loop_unroll]: 3.936e-05
        [a_1]: 0.00078062
        [with_stream_mark]: 1.971e-05
        [recompute_prepare]: 1.763e-05
        [updatestate_depend_eliminate]: 6.66999e-06
        [updatestate_assign_eliminate]: 5.62001e-06
        [updatestate_loads_eliminate]: 5.15001e-06
        [parameter_eliminate]: 1.83002e-06
        [specialize_transform]: 1.272e-05
        [updatestate_useless_node_eliminater]: 1.169e-05
        [accelerated_algorithm]: 1.214e-05
        [meta_shard_fg_expand]: 2.80997e-06
        [get_grad_eliminate_]: 1.19e-05
        [merge_forward]: 7.05e-06
        [cell_reuse_recompute_pass]: 7.60017e-07
        [cell_reuse_handle_not_recompute_node_pass]: 6.925e-05
        [j_node_and_user_rematch]: 2.311e-05
        [meta_fg_expand]: 0.00137702
        [replace_old_param]: 5.116e-05
        [inline_without_move]: 6.246e-05
        [renormalize]: 0.00236385
        [add_forward_monad_depend]: 1.028e-05
        [auto_monad_grad]: 5.63002e-06
        [auto_monad_eliminator]: 4.68e-05
        [cse]: 0.00014048
        [replace_applicator]: 6.598e-05
    [Cycle 2]: 0.00187494, [27]
        [switch_simplify]: 3.598e-05
        [loop_unroll]: 3.637e-05
        [a_1]: 0.00100547
        [with_stream_mark]: 1.238e-05
        [recompute_prepare]: 7.66999e-06
        [updatestate_depend_eliminate]: 3.56999e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 2.69999e-06
        [parameter_eliminate]: 1.17999e-06
        [specialize_transform]: 6.35002e-06
        [updatestate_useless_node_eliminater]: 5.92999e-06
        [accelerated_algorithm]: 6.12999e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 6.64001e-06
        [merge_forward]: 2.99001e-06
        [cell_reuse_recompute_pass]: 9.49978e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.281e-05
        [j_node_and_user_rematch]: 9.73002e-06
        [meta_fg_expand]: 4.031e-05
        [replace_old_param]: 9.57999e-06
        [inline_without_move]: 6.09999e-06
        [renormalize]: 0.00046469
        [add_forward_monad_depend]: 4.08001e-06
        [auto_monad_grad]: 1.40999e-06
        [auto_monad_eliminator]: 1.115e-05
        [cse]: 1.94e-05
        [replace_applicator]: 1.214e-05
    [Cycle 3]: 0.0003611, [27]
        [switch_simplify]: 6.32001e-06
        [loop_unroll]: 6.31e-06
        [a_1]: 0.00011523
        [with_stream_mark]: 8.77e-06
        [recompute_prepare]: 6.56e-06
        [updatestate_depend_eliminate]: 3.54002e-06
        [updatestate_assign_eliminate]: 2.70002e-06
        [updatestate_loads_eliminate]: 2.66e-06
        [parameter_eliminate]: 1.09003e-06
        [specialize_transform]: 6.43e-06
        [updatestate_useless_node_eliminater]: 5.84e-06
        [accelerated_algorithm]: 5.87001e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 5.76e-06
        [merge_forward]: 2.86999e-06
        [cell_reuse_recompute_pass]: 1.33002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.462e-05
        [j_node_and_user_rematch]: 9.37001e-06
        [meta_fg_expand]: 1.92999e-06
        [replace_old_param]: 7.84002e-06
        [inline_without_move]: 5.71e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.47999e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 6.69999e-06
        [cse]: 1.388e-05
        [replace_applicator]: 6.02001e-06
[py_interpret_to_execute_after_opt_a]: 1.129e-05
[rewriter_after_opt_a]: 3.404e-05
[convert_after_rewriter]: 6.73998e-06
[order_py_execute_after_rewriter]: 5.56002e-06
[mutable_eliminate]: 0.0011926
[jit_opt_b]: 8.059e-05, [1]
    [Cycle 1]: 6.757e-05, [2]
        [frontend_op_eliminate]: 2.723e-05
        [inline_after_opt_a]: 2.436e-05
[cconv]: 4.024e-05
[loop_unroll]: 0.00069503
[jit_opt_after_cconv]: 0.00020057, [1]
    [Cycle 1]: 0.00019228, [11]
        [c_1]: 2.961e-05
        [parameter_eliminate]: 7.17002e-06
        [updatestate_depend_eliminate]: 1.102e-05
        [updatestate_assign_eliminate]: 3.91001e-06
        [updatestate_loads_eliminate]: 3.18e-06
        [cse]: 3.655e-05
        [call_graph_tuple_transform]: 3.094e-05
        [tuple_list_get_item_eliminator]: 6.84999e-06
        [none_parameter_eliminate]: 1.77999e-06
        [renormalize]: 1.15999e-06
        [switch_simplify]: 7.1e-06
[remove_dup_value]: 4.833e-05
[partial_unused_args_eliminate]: 3.14999e-06
[environ_conv]: 7.48e-06
[add_recomputation]: 6.729e-05
[cse_after_recomputation]: 3.241e-05, [1]
    [Cycle 1]: 2.51e-05, [1]
        [cse]: 1.823e-05
[auto_monad_reorder]: 1.999e-05
[get_jit_bprop_graph]: 2.78e-06
[rewriter_after_jit_bprop_graph]: 1.058e-05
[opt_after_jit_grad]: 0.00052281
[symbol_engine_optimizer]: 0.00010136, [1]
    [Cycle 1]: 9.423e-05, [6]
        [build]: 5.29e-06
        [elim_shapecalc]: 1.24e-05
        [elim_not_effective]: 1.944e-05
        [opt_reshape]: 8.81997e-06
        [fold_const_symbol]: 1.478e-05
        [renormalize]: 6.10016e-07
[validate]: 5.109e-05
[backend_pass]: 8.80013e-07
[task_emit]: 0.00831338
[execute]: 1.274e-05
Sums
    bootstrap                            :     0.000183s :  0.65%
    type_inference                       :     0.009112s : 32.51%
    event_method                         :     0.000038s :  0.14%
    auto_monad                           :     0.000072s :  0.26%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000023s :  0.08%
    rewriter_before_opt_a                :     0.000083s :  0.30%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000096s :  0.34%
    jit_opt_a.loop_unroll                :     0.000082s :  0.29%
    jit_opt_a.a_1                        :     0.001901s :  6.78%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000025s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.09%
    jit_opt_a.merge_forward              :     0.000013s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000097s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001419s :  5.06%
    jit_opt_a.replace_old_param          :     0.000069s :  0.24%
    jit_opt_a.inline_without_move        :     0.000074s :  0.27%
    jit_opt_a.renormalize                :     0.002829s : 10.09%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000065s :  0.23%
    jit_opt_a.cse                        :     0.000174s :  0.62%
    jit_opt_a.replace_applicator         :     0.000084s :  0.30%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.04%
    rewriter_after_opt_a                 :     0.000034s :  0.12%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.001193s :  4.26%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.09%
    cconv                                :     0.000040s :  0.14%
    loop_unroll                          :     0.000695s :  2.48%
    jit_opt_after_cconv.c_1              :     0.000030s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000037s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000031s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000048s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000067s :  0.24%
    cse_after_recomputation.cse          :     0.000018s :  0.07%
    auto_monad_reorder                   :     0.000020s :  0.07%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.04%
    opt_after_jit_grad                   :     0.000523s :  1.87%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000051s :  0.18%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008313s : 29.67%
    execute                              :     0.000013s :  0.05%


Time group info:
------[substitution.]   0.000441   110
 0.70% :     0.000003s :      3: substitution.elim_not_effective
 0.61% :     0.000003s :      3: substitution.fold_const_symbol
 1.62% :     0.000007s :      4: substitution.graph_param_transform
59.15% :     0.000261s :     12: substitution.inline
 5.90% :     0.000026s :      2: substitution.inline_without_move
 1.75% :     0.000008s :     14: substitution.j_node_and_user_rematch
 2.49% :     0.000011s :      7: substitution.minmaximum_grad
 1.51% :     0.000007s :      5: substitution.partial_eliminate
 2.53% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.88% :     0.000026s :     10: substitution.replace_applicator
 1.44% :     0.000006s :      7: substitution.replace_old_param
 0.54% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.99% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.41% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 7.48% :     0.000033s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009052     2
89.63% :     0.008113s :      1: type_inference.infer
10.37% :     0.000939s :      1: type_inference.specialize
------[replace.]   0.000135    19
63.18% :     0.000085s :     12: replace.inline
36.82% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000270    19
94.24% :     0.000255s :     12: match.inline
 5.76% :     0.000016s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000373  2568
 1.54% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.64% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.36% :     0.000005s :     43: predicate.addn_check_dump
 1.87% :     0.000007s :     43: predicate.addn_zero_filter
 2.04% :     0.000008s :     43: predicate.arithmetic_simplify
 1.54% :     0.000006s :     43: predicate.cast_eliminate
 0.22% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.35% :     0.000005s :     43: predicate.compare_switch_simplify
 1.51% :     0.000006s :     43: predicate.depend_value_elim
 1.44% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.45% :     0.000005s :     43: predicate.dict_get_item_eliminator
 1.42% :     0.000005s :     43: predicate.dict_set_item_eliminator
 0.34% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      4: predicate.elim_not_effective
 0.42% :     0.000002s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.59% :     0.000006s :     43: predicate.environ_add_const_eliminate
 1.42% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.37% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.45% :     0.000005s :     43: predicate.environ_get_eliminate
 1.42% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.80% :     0.000003s :     19: predicate.get_grad_eliminate
 0.11% :     0.000000s :      4: predicate.graph_param_transform
 4.28% :     0.000016s :     70: predicate.inline
 1.87% :     0.000007s :     40: predicate.inline_without_move
 0.34% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.89% :     0.000003s :     19: predicate.less_batch_normalization
 1.70% :     0.000006s :     50: predicate.list_to_tuple_eliminator_
 1.92% :     0.000007s :     54: predicate.load_eliminater
 0.80% :     0.000003s :      4: predicate.loop_unroll_after_grad
 3.31% :     0.000012s :     85: predicate.loop_unroll_before_grad
 1.85% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.36% :     0.000005s :     43: predicate.merge_addn
 1.44% :     0.000005s :     43: predicate.minmaximum_grad
 1.60% :     0.000006s :      4: predicate.mutable_eliminate
 0.25% :     0.000001s :      4: predicate.opt_reshape
 2.22% :     0.000008s :     54: predicate.partial_eliminate
 1.58% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.83% :     0.000007s :     43: predicate.reduce_eliminate
 1.80% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.45% :     0.000009s :     96: predicate.replace_applicator
 0.82% :     0.000003s :     40: predicate.replace_old_param
 0.24% :     0.000001s :      4: predicate.reset_defer_inline
 1.62% :     0.000006s :     43: predicate.reshape_eliminate
 1.47% :     0.000005s :     43: predicate.row_tensor_add_zeros_like
 0.50% :     0.000002s :      4: predicate.row_tensor_eliminate
 1.58% :     0.000006s :     43: predicate.same_eliminate
 0.45% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.45% :     0.000002s :      8: predicate.special_op_eliminate
 0.83% :     0.000003s :     19: predicate.specialize_transform
 1.66% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000005s :     43: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.60% :     0.000010s :     62: predicate.switch_defer_inline
 2.42% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.05% :     0.000023s :    151: predicate.switch_simplify
 1.48% :     0.000006s :     43: predicate.tile_eliminate
 1.50% :     0.000006s :     43: predicate.transpose_eliminate
 1.78% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.67% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.11% :     0.000012s :     58: predicate.tuple_list_get_item_eliminator
 1.97% :     0.000007s :     43: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000007s :     50: predicate.tuple_to_list_eliminator_
 1.89% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000010s :     73: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000006s :     43: predicate.value_based_eliminate
 0.17% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.32% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001321    29
58.26% :     0.000770s :     13: func_graph_cloner_run.FuncGraphClonerGraph
41.74% :     0.000552s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.037417    91
 0.19% :     0.000071s :      1: add_recomputation
 0.20% :     0.000076s :      1: auto_monad
 0.06% :     0.000023s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.52% :     0.000195s :      1: bootstrap
 0.12% :     0.000043s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000035s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.11% :     0.000042s :      1: event_method
 0.05% :     0.000018s :      1: execute
 0.01% :     0.000004s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
28.44% :     0.010642s :      1: jit_opt_a
 0.54% :     0.000204s :      1: jit_opt_after_cconv
 0.23% :     0.000084s :      1: jit_opt_b
 1.88% :     0.000705s :      1: loop_unroll
 3.25% :     0.001215s :      1: mutable_eliminate
 6.78% :     0.002535s :     39: opt.transform.jit_opt_a
 0.19% :     0.000070s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000042s :      4: opt.transform.jit_opt_b
 0.05% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000043s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.42% :     0.000531s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000006s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000025s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.14% :     0.000052s :      1: remove_dup_value
 3.99% :     0.001492s :      2: renormalize.infer
 3.54% :     0.001323s :      2: renormalize.specialize
 0.03% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000038s :      1: rewriter_after_opt_a
 0.23% :     0.000086s :      1: rewriter_before_opt_a
 0.28% :     0.000105s :      1: symbol_engine_optimizer
22.29% :     0.008341s :      1: task_emit
24.38% :     0.009123s :      1: type_inference
 0.20% :     0.000074s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.987.297 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc39f70> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:16.988.015 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0141285, [30]
[bootstrap]: 0.00027764
[type_inference]: 0.00446832
[event_method]: 2.091e-05
[auto_monad]: 6.836e-05
[graph_reusing]: 5.90002e-06
[pre_auto_parallel]: 2.51e-06
[py_interpret_to_execute]: 4.295e-05
[rewriter_before_opt_a]: 7.585e-05
[expand_dump_flag]: 3.35e-06
[jit_opt_a]: 0.00655765, [2]
    [Cycle 1]: 0.00209685, [27]
        [switch_simplify]: 6.724e-05
        [loop_unroll]: 2.912e-05
        [a_1]: 0.00066999
        [with_stream_mark]: 2.257e-05
        [recompute_prepare]: 8.04002e-06
        [updatestate_depend_eliminate]: 3.61001e-06
        [updatestate_assign_eliminate]: 2.85002e-06
        [updatestate_loads_eliminate]: 3.01001e-06
        [parameter_eliminate]: 1.45001e-06
        [specialize_transform]: 6.11e-06
        [updatestate_useless_node_eliminater]: 4.72e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 5.70001e-06
        [merge_forward]: 3.69002e-06
        [cell_reuse_recompute_pass]: 1.39998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.545e-05
        [j_node_and_user_rematch]: 1.047e-05
        [meta_fg_expand]: 1.97001e-06
        [replace_old_param]: 1.012e-05
        [inline_without_move]: 5.97999e-06
        [renormalize]: 0.00094804
        [add_forward_monad_depend]: 7.55e-06
        [auto_monad_grad]: 2.76e-06
        [auto_monad_eliminator]: 1.698e-05
        [cse]: 3.102e-05
        [replace_applicator]: 1.581e-05
    [Cycle 2]: 0.00034013, [27]
        [switch_simplify]: 6.19001e-06
        [loop_unroll]: 4.86002e-06
        [a_1]: 9.865e-05
        [with_stream_mark]: 1.045e-05
        [recompute_prepare]: 5.33002e-06
        [updatestate_depend_eliminate]: 2.43e-06
        [updatestate_assign_eliminate]: 2.53998e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.46002e-06
        [specialize_transform]: 4.69002e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 5.69e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 3.01999e-06
        [cell_reuse_recompute_pass]: 1.90001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.392e-05
        [j_node_and_user_rematch]: 9.09e-06
        [meta_fg_expand]: 1.78002e-06
        [replace_old_param]: 7.32002e-06
        [inline_without_move]: 4.67e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.82001e-06
        [auto_monad_grad]: 7.29982e-07
        [auto_monad_eliminator]: 5.98002e-06
        [cse]: 1.103e-05
        [replace_applicator]: 5.04003e-06
[py_interpret_to_execute_after_opt_a]: 1.2e-05
[rewriter_after_opt_a]: 3.869e-05
[convert_after_rewriter]: 7.09001e-06
[order_py_execute_after_rewriter]: 5.53002e-06
[mutable_eliminate]: 0.00082297
[jit_opt_b]: 5.757e-05, [1]
    [Cycle 1]: 4.838e-05, [2]
        [frontend_op_eliminate]: 1.987e-05
        [inline_after_opt_a]: 1.595e-05
[cconv]: 3.282e-05
[loop_unroll]: 0.00046591
[jit_opt_after_cconv]: 0.00016071, [1]
    [Cycle 1]: 0.00015313, [11]
        [c_1]: 2.204e-05
        [parameter_eliminate]: 3.93001e-06
        [updatestate_depend_eliminate]: 7.23e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [cse]: 2.661e-05
        [call_graph_tuple_transform]: 2.171e-05
        [tuple_list_get_item_eliminator]: 5.52999e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 1.34e-06
        [switch_simplify]: 5.30999e-06
[remove_dup_value]: 1.438e-05
[partial_unused_args_eliminate]: 2.58e-06
[environ_conv]: 5.47001e-06
[add_recomputation]: 5.35e-05
[cse_after_recomputation]: 2.269e-05, [1]
    [Cycle 1]: 1.614e-05, [1]
        [cse]: 9.67999e-06
[auto_monad_reorder]: 1.416e-05
[get_jit_bprop_graph]: 2.52001e-06
[rewriter_after_jit_bprop_graph]: 9.91e-06
[opt_after_jit_grad]: 0.00055066
[symbol_engine_optimizer]: 8.173e-05, [1]
    [Cycle 1]: 7.328e-05, [6]
        [build]: 5.02e-06
        [elim_shapecalc]: 8.18999e-06
        [elim_not_effective]: 1.543e-05
        [opt_reshape]: 6.02999e-06
        [fold_const_symbol]: 9.36998e-06
        [renormalize]: 4.00003e-07
[validate]: 5.188e-05
Sums
    bootstrap                            :     0.000278s :  2.96%
    type_inference                       :     0.004468s : 47.69%
    event_method                         :     0.000021s :  0.22%
    auto_monad                           :     0.000068s :  0.73%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000043s :  0.46%
    rewriter_before_opt_a                :     0.000076s :  0.81%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000073s :  0.78%
    jit_opt_a.loop_unroll                :     0.000034s :  0.36%
    jit_opt_a.a_1                        :     0.000769s :  8.20%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000948s : 10.12%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.25%
    jit_opt_a.cse                        :     0.000042s :  0.45%
    jit_opt_a.replace_applicator         :     0.000021s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.13%
    rewriter_after_opt_a                 :     0.000039s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000823s :  8.78%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.17%
    cconv                                :     0.000033s :  0.35%
    loop_unroll                          :     0.000466s :  4.97%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000027s :  0.28%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.15%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000053s :  0.57%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000014s :  0.15%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.11%
    opt_after_jit_grad                   :     0.000551s :  5.88%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000052s :  0.55%


Time group info:
------[substitution.]   0.000315    23
 0.88% :     0.000003s :      2: substitution.elim_not_effective
 0.29% :     0.000001s :      2: substitution.fold_const_symbol
 1.93% :     0.000006s :      3: substitution.graph_param_transform
89.46% :     0.000282s :      4: substitution.inline
 1.10% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.49% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.45% :     0.000005s :      2: substitution.replace_old_param
 3.40% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004393     2
78.96% :     0.003469s :      1: type_inference.infer
21.04% :     0.000924s :      1: type_inference.specialize
------[replace.]   0.000082     6
69.76% :     0.000057s :      4: replace.inline
30.24% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000289     6
96.68% :     0.000279s :      4: match.inline
 3.32% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.68% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.15% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.60% :     0.000002s :     10: predicate.addn_zero_filter
 2.05% :     0.000003s :     10: predicate.arithmetic_simplify
 1.14% :     0.000001s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.68% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.13% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.51% :     0.000001s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.41% :     0.000002s :     10: predicate.environ_get_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.00% :     0.000006s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000001s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.53% :     0.000002s :     15: predicate.load_eliminater
 1.54% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.38% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 2.72% :     0.000004s :      3: predicate.mutable_eliminate
 0.77% :     0.000001s :      3: predicate.opt_reshape
 1.91% :     0.000002s :     15: predicate.partial_eliminate
 1.28% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.68% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.30% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.68% :     0.000002s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.23% :     0.000002s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.68% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.92% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.25% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.02% :     0.000009s :     53: predicate.switch_simplify
 1.33% :     0.000002s :     10: predicate.tile_eliminate
 1.18% :     0.000002s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.72% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.96% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.46% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.44% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.89% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000686    12
46.52% :     0.000319s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.82% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.66% :     0.000313s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016124    72
 0.35% :     0.000057s :      1: add_recomputation
 0.45% :     0.000073s :      1: auto_monad
 0.10% :     0.000017s :      1: auto_monad_reorder
 1.83% :     0.000296s :      1: bootstrap
 0.22% :     0.000036s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.16% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
40.70% :     0.006562s :      1: jit_opt_a
 1.02% :     0.000164s :      1: jit_opt_after_cconv
 0.37% :     0.000060s :      1: jit_opt_b
 2.95% :     0.000475s :      1: loop_unroll
 5.17% :     0.000834s :      1: mutable_eliminate
 6.19% :     0.000998s :     26: opt.transform.jit_opt_a
 0.32% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000029s :      4: opt.transform.jit_opt_b
 0.08% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.48% :     0.000561s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000046s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 3.21% :     0.000517s :      1: renormalize.infer
 2.61% :     0.000420s :      1: renormalize.specialize
 0.07% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000042s :      1: rewriter_after_opt_a
 0.49% :     0.000079s :      1: rewriter_before_opt_a
 0.53% :     0.000085s :      1: symbol_engine_optimizer
27.82% :     0.004486s :      1: type_inference


TotalTime = 0.0206178, [33]
[bootstrap]: 0.00027615
[type_inference]: 0.00452712
[event_method]: 1.721e-05
[auto_monad]: 4.379e-05
[graph_reusing]: 4.16001e-06
[pre_auto_parallel]: 1.85001e-06
[py_interpret_to_execute]: 3.709e-05
[rewriter_before_opt_a]: 6.346e-05
[expand_dump_flag]: 1.84e-06
[jit_opt_a]: 0.00644516, [2]
    [Cycle 1]: 0.00209723, [27]
        [switch_simplify]: 6.783e-05
        [loop_unroll]: 2.909e-05
        [a_1]: 0.00066817
        [with_stream_mark]: 2.493e-05
        [recompute_prepare]: 7.38999e-06
        [updatestate_depend_eliminate]: 3.71999e-06
        [updatestate_assign_eliminate]: 2.76e-06
        [updatestate_loads_eliminate]: 2.88e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 5.82001e-06
        [updatestate_useless_node_eliminater]: 5.28002e-06
        [accelerated_algorithm]: 6.49001e-06
        [meta_shard_fg_expand]: 1.31998e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 3.04001e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.502e-05
        [j_node_and_user_rematch]: 9.91e-06
        [meta_fg_expand]: 2.29001e-06
        [replace_old_param]: 1.044e-05
        [inline_without_move]: 5.79e-06
        [renormalize]: 0.000946
        [add_forward_monad_depend]: 7.21999e-06
        [auto_monad_grad]: 2.44001e-06
        [auto_monad_eliminator]: 1.704e-05
        [cse]: 3.104e-05
        [replace_applicator]: 1.583e-05
    [Cycle 2]: 0.00033661, [27]
        [switch_simplify]: 5.44e-06
        [loop_unroll]: 5.06002e-06
        [a_1]: 9.668e-05
        [with_stream_mark]: 1.302e-05
        [recompute_prepare]: 4.99003e-06
        [updatestate_depend_eliminate]: 2.84999e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.30999e-06
        [specialize_transform]: 4.87e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 6.06e-06
        [meta_shard_fg_expand]: 1.77001e-06
        [get_grad_eliminate_]: 4.60001e-06
        [merge_forward]: 2.68e-06
        [cell_reuse_recompute_pass]: 2.41998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.312e-05
        [j_node_and_user_rematch]: 8.02e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 7.28e-06
        [inline_without_move]: 4.53001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.30999e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 5.59998e-06
        [cse]: 1.07e-05
        [replace_applicator]: 5.52001e-06
[py_interpret_to_execute_after_opt_a]: 1.502e-05
[rewriter_after_opt_a]: 3.939e-05
[convert_after_rewriter]: 8.21002e-06
[order_py_execute_after_rewriter]: 5.29998e-06
[mutable_eliminate]: 0.00082456
[jit_opt_b]: 5.694e-05, [1]
    [Cycle 1]: 4.757e-05, [2]
        [frontend_op_eliminate]: 1.98e-05
        [inline_after_opt_a]: 1.505e-05
[cconv]: 3.277e-05
[loop_unroll]: 0.00046911
[jit_opt_after_cconv]: 0.00015861, [1]
    [Cycle 1]: 0.00015151, [11]
        [c_1]: 2.128e-05
        [parameter_eliminate]: 4.51002e-06
        [updatestate_depend_eliminate]: 7.8e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [cse]: 2.661e-05
        [call_graph_tuple_transform]: 2.18e-05
        [tuple_list_get_item_eliminator]: 5.59e-06
        [none_parameter_eliminate]: 1.66002e-06
        [renormalize]: 9.39996e-07
        [switch_simplify]: 5.82999e-06
[remove_dup_value]: 1.441e-05
[partial_unused_args_eliminate]: 2.46998e-06
[environ_conv]: 5.46e-06
[add_recomputation]: 5.447e-05
[cse_after_recomputation]: 2.159e-05, [1]
    [Cycle 1]: 1.518e-05, [1]
        [cse]: 8.83001e-06
[auto_monad_reorder]: 1.598e-05
[get_jit_bprop_graph]: 2.48002e-06
[rewriter_after_jit_bprop_graph]: 8.66002e-06
[opt_after_jit_grad]: 0.00055173
[symbol_engine_optimizer]: 8.133e-05, [1]
    [Cycle 1]: 7.355e-05, [6]
        [build]: 5.34e-06
        [elim_shapecalc]: 8.27e-06
        [elim_not_effective]: 1.509e-05
        [opt_reshape]: 6.38003e-06
        [fold_const_symbol]: 9.19e-06
        [renormalize]: 4.80009e-07
[validate]: 4.432e-05
[backend_pass]: 1.10001e-06
[task_emit]: 0.00656239
[execute]: 9.75002e-06
Sums
    bootstrap                            :     0.000276s :  1.73%
    type_inference                       :     0.004527s : 28.38%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000044s :  0.27%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000037s :  0.23%
    rewriter_before_opt_a                :     0.000063s :  0.40%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000073s :  0.46%
    jit_opt_a.loop_unroll                :     0.000034s :  0.21%
    jit_opt_a.a_1                        :     0.000765s :  4.80%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000946s :  5.93%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.14%
    jit_opt_a.cse                        :     0.000042s :  0.26%
    jit_opt_a.replace_applicator         :     0.000021s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.09%
    rewriter_after_opt_a                 :     0.000039s :  0.25%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000825s :  5.17%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000033s :  0.21%
    loop_unroll                          :     0.000469s :  2.94%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000027s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000054s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000552s :  3.46%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000044s :  0.28%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006562s : 41.15%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000323    23
 0.64% :     0.000002s :      2: substitution.elim_not_effective
 0.40% :     0.000001s :      2: substitution.fold_const_symbol
 1.89% :     0.000006s :      3: substitution.graph_param_transform
88.49% :     0.000286s :      4: substitution.inline
 1.51% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.38% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.66% :     0.000005s :      2: substitution.replace_old_param
 4.02% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004464     2
80.42% :     0.003590s :      1: type_inference.infer
19.58% :     0.000874s :      1: type_inference.specialize
------[replace.]   0.000079     6
68.43% :     0.000054s :      4: replace.inline
31.57% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000295     6
96.00% :     0.000283s :      4: match.inline
 4.00% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 1.76% :     0.000002s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.36% :     0.000002s :     10: predicate.compare_switch_simplify
 1.24% :     0.000002s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.59% :     0.000002s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.46% :     0.000007s :     22: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.44% :     0.000002s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.25% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.71% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.64% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.78% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.07% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.79% :     0.000002s :     10: predicate.reshape_eliminate
 1.43% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.39% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.84% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.43% :     0.000002s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.64% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.30% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.69% :     0.000009s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.26% :     0.000002s :     10: predicate.transpose_eliminate
 1.16% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.04% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.33% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 2.00% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.52% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.06% :     0.000003s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.87% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000640    12
44.13% :     0.000282s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.32% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.55% :     0.000311s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022670    76
 0.25% :     0.000057s :      1: add_recomputation
 0.21% :     0.000047s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.27% :     0.000288s :      1: bootstrap
 0.16% :     0.000036s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
28.45% :     0.006450s :      1: jit_opt_a
 0.71% :     0.000161s :      1: jit_opt_after_cconv
 0.26% :     0.000060s :      1: jit_opt_b
 2.11% :     0.000478s :      1: loop_unroll
 3.69% :     0.000836s :      1: mutable_eliminate
 4.38% :     0.000993s :     26: opt.transform.jit_opt_a
 0.23% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.48% :     0.000562s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000040s :      1: py_interpret_to_execute
 0.08% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.28% :     0.000518s :      1: renormalize.infer
 1.84% :     0.000418s :      1: renormalize.specialize
 0.05% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000042s :      1: rewriter_after_opt_a
 0.29% :     0.000066s :      1: rewriter_before_opt_a
 0.37% :     0.000084s :      1: symbol_engine_optimizer
29.03% :     0.006582s :      1: task_emit
20.02% :     0.004539s :      1: type_inference
 0.30% :     0.000067s :      1: validate


TotalTime = 0.0151019, [30]
[bootstrap]: 0.00035508
[type_inference]: 0.00529673
[event_method]: 1.939e-05
[auto_monad]: 6.858e-05
[graph_reusing]: 5.45001e-06
[pre_auto_parallel]: 2.62001e-06
[py_interpret_to_execute]: 4.135e-05
[rewriter_before_opt_a]: 8.325e-05
[expand_dump_flag]: 3.88999e-06
[jit_opt_a]: 0.00658735, [2]
    [Cycle 1]: 0.00217535, [27]
        [switch_simplify]: 6.6e-05
        [loop_unroll]: 3.042e-05
        [a_1]: 0.00064692
        [with_stream_mark]: 2.615e-05
        [recompute_prepare]: 9.15999e-06
        [updatestate_depend_eliminate]: 3.62998e-06
        [updatestate_assign_eliminate]: 3.67002e-06
        [updatestate_loads_eliminate]: 2.50002e-06
        [parameter_eliminate]: 2.32999e-06
        [specialize_transform]: 6.99001e-06
        [updatestate_useless_node_eliminater]: 5.64e-06
        [accelerated_algorithm]: 6.25002e-06
        [meta_shard_fg_expand]: 2.49001e-06
        [get_grad_eliminate_]: 6.17999e-06
        [merge_forward]: 4.05e-06
        [cell_reuse_recompute_pass]: 1.38002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.7e-05
        [j_node_and_user_rematch]: 1.074e-05
        [meta_fg_expand]: 3.08998e-06
        [replace_old_param]: 1.106e-05
        [inline_without_move]: 5.92999e-06
        [renormalize]: 0.00102142
        [add_forward_monad_depend]: 7.33e-06
        [auto_monad_grad]: 2.63e-06
        [auto_monad_eliminator]: 1.671e-05
        [cse]: 3.461e-05
        [replace_applicator]: 1.516e-05
    [Cycle 2]: 0.00036269, [27]
        [switch_simplify]: 6.39001e-06
        [loop_unroll]: 5.61e-06
        [a_1]: 9.969e-05
        [with_stream_mark]: 9.90002e-06
        [recompute_prepare]: 4.88001e-06
        [updatestate_depend_eliminate]: 2.68e-06
        [updatestate_assign_eliminate]: 2.08002e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 9.09989e-07
        [specialize_transform]: 4.82e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.78997e-06
        [meta_shard_fg_expand]: 1.40001e-06
        [get_grad_eliminate_]: 5.05001e-06
        [merge_forward]: 3.23e-06
        [cell_reuse_recompute_pass]: 1.04998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.285e-05
        [j_node_and_user_rematch]: 8.77e-06
        [meta_fg_expand]: 2.11e-06
        [replace_old_param]: 8.16002e-06
        [inline_without_move]: 4.68001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 2.28998e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 9.56e-06
        [cse]: 1.486e-05
        [replace_applicator]: 5.86e-06
[py_interpret_to_execute_after_opt_a]: 1.458e-05
[rewriter_after_opt_a]: 3.793e-05
[convert_after_rewriter]: 8.99e-06
[order_py_execute_after_rewriter]: 6.21998e-06
[mutable_eliminate]: 0.00074485
[jit_opt_b]: 5.808e-05, [1]
    [Cycle 1]: 4.923e-05, [2]
        [frontend_op_eliminate]: 1.909e-05
        [inline_after_opt_a]: 1.69e-05
[cconv]: 2.919e-05
[loop_unroll]: 0.0005477
[jit_opt_after_cconv]: 0.00015986, [1]
    [Cycle 1]: 0.00015263, [11]
        [c_1]: 2.27e-05
        [parameter_eliminate]: 3.46001e-06
        [updatestate_depend_eliminate]: 7.15e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [cse]: 2.422e-05
        [call_graph_tuple_transform]: 2.296e-05
        [tuple_list_get_item_eliminator]: 5.47001e-06
        [none_parameter_eliminate]: 1.64998e-06
        [renormalize]: 5.40022e-07
        [switch_simplify]: 5.17e-06
[remove_dup_value]: 1.568e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 5.57001e-06
[add_recomputation]: 5.362e-05
[cse_after_recomputation]: 2.365e-05, [1]
    [Cycle 1]: 1.712e-05, [1]
        [cse]: 1.016e-05
[auto_monad_reorder]: 1.623e-05
[get_jit_bprop_graph]: 1.86e-06
[rewriter_after_jit_bprop_graph]: 6.68998e-06
[opt_after_jit_grad]: 0.00056055
[symbol_engine_optimizer]: 8.254e-05, [1]
    [Cycle 1]: 7.385e-05, [6]
        [build]: 5.89e-06
        [elim_shapecalc]: 8.68001e-06
        [elim_not_effective]: 1.486e-05
        [opt_reshape]: 6.54001e-06
        [fold_const_symbol]: 8.65999e-06
        [renormalize]: 3.59985e-07
[validate]: 4.363e-05
Sums
    bootstrap                            :     0.000355s :  3.43%
    type_inference                       :     0.005297s : 51.12%
    event_method                         :     0.000019s :  0.19%
    auto_monad                           :     0.000069s :  0.66%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000041s :  0.40%
    rewriter_before_opt_a                :     0.000083s :  0.80%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000072s :  0.70%
    jit_opt_a.loop_unroll                :     0.000036s :  0.35%
    jit_opt_a.a_1                        :     0.000747s :  7.21%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000019s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.001021s :  9.86%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.25%
    jit_opt_a.cse                        :     0.000049s :  0.48%
    jit_opt_a.replace_applicator         :     0.000021s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.14%
    rewriter_after_opt_a                 :     0.000038s :  0.37%
    convert_after_rewriter               :     0.000009s :  0.09%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000745s :  7.19%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.16%
    cconv                                :     0.000029s :  0.28%
    loop_unroll                          :     0.000548s :  5.29%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000016s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000054s :  0.52%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000561s :  5.41%
    symbol_engine_optimizer.build        :     0.000006s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000044s :  0.42%


Time group info:
------[substitution.]   0.000269    23
 0.77% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000002s :      2: substitution.fold_const_symbol
 2.38% :     0.000006s :      3: substitution.graph_param_transform
85.88% :     0.000231s :      4: substitution.inline
 1.64% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.87% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000005s :      2: substitution.replace_old_param
 5.00% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005212     2
80.10% :     0.004174s :      1: type_inference.infer
19.90% :     0.001037s :      1: type_inference.specialize
------[replace.]   0.000092     6
64.66% :     0.000059s :      4: replace.inline
35.34% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000240     6
95.06% :     0.000228s :      4: match.inline
 4.94% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   691
 1.72% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.73% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.89% :     0.000001s :     10: predicate.addn_check_dump
 1.72% :     0.000002s :     10: predicate.addn_zero_filter
 2.85% :     0.000004s :     10: predicate.arithmetic_simplify
 1.20% :     0.000002s :     10: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.92% :     0.000001s :     10: predicate.compare_switch_simplify
 1.96% :     0.000003s :     10: predicate.depend_value_elim
 1.23% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.04% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.06% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.39% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.41% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.87% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.99% :     0.000001s :     10: predicate.environ_get_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 6.60% :     0.000009s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.47% :     0.000002s :      6: predicate.less_batch_normalization
 1.27% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.49% :     0.000002s :     15: predicate.load_eliminater
 1.85% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.80% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.68% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.90% :     0.000001s :     10: predicate.merge_addn
 0.91% :     0.000001s :     10: predicate.minmaximum_grad
 1.98% :     0.000003s :      3: predicate.mutable_eliminate
 0.88% :     0.000001s :      3: predicate.opt_reshape
 1.90% :     0.000003s :     15: predicate.partial_eliminate
 1.06% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.40% :     0.000002s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.44% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.30% :     0.000002s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.21% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.59% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.32% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.31% :     0.000003s :     16: predicate.switch_defer_inline
 1.79% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.67% :     0.000011s :     53: predicate.switch_simplify
 1.09% :     0.000001s :     10: predicate.tile_eliminate
 0.96% :     0.000001s :     10: predicate.transpose_eliminate
 1.81% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.89% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.41% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.40% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000774    12
46.89% :     0.000363s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.42% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.69% :     0.000354s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017160    72
 0.33% :     0.000057s :      1: add_recomputation
 0.42% :     0.000073s :      1: auto_monad
 0.12% :     0.000020s :      1: auto_monad_reorder
 2.19% :     0.000376s :      1: bootstrap
 0.19% :     0.000032s :      1: cconv
 0.07% :     0.000012s :      1: convert_after_rewriter
 0.15% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000024s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.42% :     0.006592s :      1: jit_opt_a
 0.95% :     0.000163s :      1: jit_opt_after_cconv
 0.36% :     0.000061s :      1: jit_opt_b
 3.27% :     0.000560s :      1: loop_unroll
 4.41% :     0.000756s :      1: mutable_eliminate
 5.72% :     0.000982s :     26: opt.transform.jit_opt_a
 0.31% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000029s :      4: opt.transform.jit_opt_b
 0.10% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.34% :     0.000573s :      1: opt_after_jit_grad
 0.05% :     0.000009s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000044s :      1: py_interpret_to_execute
 0.10% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 3.05% :     0.000524s :      1: renormalize.infer
 2.83% :     0.000486s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000042s :      1: rewriter_after_opt_a
 0.51% :     0.000087s :      1: rewriter_before_opt_a
 0.50% :     0.000085s :      1: symbol_engine_optimizer
30.97% :     0.005315s :      1: type_inference


TotalTime = 0.0217084, [33]
[bootstrap]: 0.00037071
[type_inference]: 0.00537754
[event_method]: 2.046e-05
[auto_monad]: 5.549e-05
[graph_reusing]: 4.62e-06
[pre_auto_parallel]: 2.29001e-06
[py_interpret_to_execute]: 4.395e-05
[rewriter_before_opt_a]: 7.469e-05
[expand_dump_flag]: 2.93e-06
[jit_opt_a]: 0.006448, [2]
    [Cycle 1]: 0.00193716, [27]
        [switch_simplify]: 5.908e-05
        [loop_unroll]: 3.099e-05
        [a_1]: 0.00064439
        [with_stream_mark]: 1.814e-05
        [recompute_prepare]: 1.035e-05
        [updatestate_depend_eliminate]: 3.37997e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [parameter_eliminate]: 1.35999e-06
        [specialize_transform]: 6.43e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 6.33998e-06
        [meta_shard_fg_expand]: 2.27001e-06
        [get_grad_eliminate_]: 5.89999e-06
        [merge_forward]: 2.98998e-06
        [cell_reuse_recompute_pass]: 2.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.416e-05
        [j_node_and_user_rematch]: 9.70002e-06
        [meta_fg_expand]: 1.86e-06
        [replace_old_param]: 9.20999e-06
        [inline_without_move]: 5.79999e-06
        [renormalize]: 0.00085168
        [add_forward_monad_depend]: 5.66e-06
        [auto_monad_grad]: 1.64998e-06
        [auto_monad_eliminator]: 1.414e-05
        [cse]: 1.937e-05
        [replace_applicator]: 1.518e-05
    [Cycle 2]: 0.00036603, [27]
        [switch_simplify]: 5.34e-06
        [loop_unroll]: 5.29998e-06
        [a_1]: 9.933e-05
        [with_stream_mark]: 1.193e-05
        [recompute_prepare]: 5.17999e-06
        [updatestate_depend_eliminate]: 2.72001e-06
        [updatestate_assign_eliminate]: 1.96998e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 5.27999e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 6.09999e-06
        [meta_shard_fg_expand]: 1.70001e-06
        [get_grad_eliminate_]: 4.26001e-06
        [merge_forward]: 3.43e-06
        [cell_reuse_recompute_pass]: 2.16e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.484e-05
        [j_node_and_user_rematch]: 8.03999e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 8.60999e-06
        [inline_without_move]: 5.51e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 2.08998e-06
        [auto_monad_grad]: 9.10019e-07
        [auto_monad_eliminator]: 7.88001e-06
        [cse]: 1.38e-05
        [replace_applicator]: 6.29001e-06
[py_interpret_to_execute_after_opt_a]: 1.34e-05
[rewriter_after_opt_a]: 3.303e-05
[convert_after_rewriter]: 6.74001e-06
[order_py_execute_after_rewriter]: 1.037e-05
[mutable_eliminate]: 0.00072419
[jit_opt_b]: 5.374e-05, [1]
    [Cycle 1]: 4.666e-05, [2]
        [frontend_op_eliminate]: 1.899e-05
        [inline_after_opt_a]: 1.554e-05
[cconv]: 2.188e-05
[loop_unroll]: 0.00050629
[jit_opt_after_cconv]: 0.00014825, [1]
    [Cycle 1]: 0.00014147, [11]
        [c_1]: 2.154e-05
        [parameter_eliminate]: 3.56001e-06
        [updatestate_depend_eliminate]: 6.20002e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [cse]: 2.19e-05
        [call_graph_tuple_transform]: 1.904e-05
        [tuple_list_get_item_eliminator]: 6.03002e-06
        [none_parameter_eliminate]: 1.36998e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.61998e-06
[remove_dup_value]: 9.66e-06
[partial_unused_args_eliminate]: 1.81e-06
[environ_conv]: 4.43999e-06
[add_recomputation]: 4.141e-05
[cse_after_recomputation]: 2.378e-05, [1]
    [Cycle 1]: 1.752e-05, [1]
        [cse]: 9.87001e-06
[auto_monad_reorder]: 1.125e-05
[get_jit_bprop_graph]: 1.54e-06
[rewriter_after_jit_bprop_graph]: 6.14001e-06
[opt_after_jit_grad]: 0.00063543
[symbol_engine_optimizer]: 7.845e-05, [1]
    [Cycle 1]: 7.046e-05, [6]
        [build]: 4.37e-06
        [elim_shapecalc]: 8.05999e-06
        [elim_not_effective]: 1.412e-05
        [opt_reshape]: 5.91998e-06
        [fold_const_symbol]: 8.79e-06
        [renormalize]: 4.69998e-07
[validate]: 3.454e-05
[backend_pass]: 6.89994e-07
[task_emit]: 0.00669859
[execute]: 9.61e-06
Sums
    bootstrap                            :     0.000371s :  2.20%
    type_inference                       :     0.005378s : 31.85%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000055s :  0.33%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.26%
    rewriter_before_opt_a                :     0.000075s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.38%
    jit_opt_a.loop_unroll                :     0.000036s :  0.21%
    jit_opt_a.a_1                        :     0.000744s :  4.40%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000852s :  5.04%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.13%
    jit_opt_a.cse                        :     0.000033s :  0.20%
    jit_opt_a.replace_applicator         :     0.000021s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000033s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000010s :  0.06%
    mutable_eliminate                    :     0.000724s :  4.29%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000022s :  0.13%
    loop_unroll                          :     0.000506s :  3.00%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000010s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000041s :  0.25%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000011s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000635s :  3.76%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.20%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006699s : 39.67%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000268    23
 0.80% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 1.56% :     0.000004s :      3: substitution.graph_param_transform
88.22% :     0.000237s :      4: substitution.inline
 1.39% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.51% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.78% :     0.000005s :      2: substitution.replace_old_param
 4.18% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005298     2
80.96% :     0.004289s :      1: type_inference.infer
19.04% :     0.001008s :      1: type_inference.specialize
------[replace.]   0.000087     6
64.70% :     0.000057s :      4: replace.inline
35.30% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000244     6
95.96% :     0.000234s :      4: match.inline
 4.04% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.68% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.62% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 2.61% :     0.000003s :     10: predicate.arithmetic_simplify
 1.08% :     0.000001s :     10: predicate.cast_eliminate
 0.68% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.03% :     0.000001s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.09% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      3: predicate.elim_not_effective
 0.85% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.03% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 4.60% :     0.000006s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.71% :     0.000002s :      6: predicate.less_batch_normalization
 1.75% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.59% :     0.000002s :     15: predicate.load_eliminater
 1.75% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.11% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.67% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.13% :     0.000003s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 1.97% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.93% :     0.000003s :     10: predicate.reduce_eliminate
 1.29% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.79% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.49% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.65% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.46% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.56% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.33% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 1.92% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.23% :     0.000011s :     53: predicate.switch_simplify
 1.06% :     0.000001s :     10: predicate.tile_eliminate
 1.74% :     0.000002s :     10: predicate.transpose_eliminate
 1.20% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.03% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.33% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.84% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.41% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.15% :     0.000003s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.47% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000699    12
44.36% :     0.000310s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.14% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.50% :     0.000353s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023638    76
 0.19% :     0.000045s :      1: add_recomputation
 0.25% :     0.000060s :      1: auto_monad
 0.06% :     0.000014s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.64% :     0.000388s :      1: bootstrap
 0.10% :     0.000025s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000027s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
27.30% :     0.006452s :      1: jit_opt_a
 0.64% :     0.000151s :      1: jit_opt_after_cconv
 0.24% :     0.000057s :      1: jit_opt_b
 2.18% :     0.000515s :      1: loop_unroll
 3.10% :     0.000734s :      1: mutable_eliminate
 4.12% :     0.000973s :     26: opt.transform.jit_opt_a
 0.21% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.73% :     0.000646s :      1: opt_after_jit_grad
 0.06% :     0.000013s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000047s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000012s :      1: remove_dup_value
 1.91% :     0.000451s :      1: renormalize.infer
 1.65% :     0.000391s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000037s :      1: rewriter_after_opt_a
 0.33% :     0.000078s :      1: rewriter_before_opt_a
 0.34% :     0.000081s :      1: symbol_engine_optimizer
28.43% :     0.006720s :      1: task_emit
22.82% :     0.005395s :      1: type_inference
 0.23% :     0.000053s :      1: validate


TotalTime = 0.0613557, [33]
[bootstrap]: 0.00027744
[type_inference]: 0.0171689
[event_method]: 5.556e-05
[auto_monad]: 0.00010909
[graph_reusing]: 7.66999e-06
[pre_auto_parallel]: 3.71999e-06
[py_interpret_to_execute]: 3.668e-05
[rewriter_before_opt_a]: 0.00010194
[expand_dump_flag]: 4.60999e-06
[jit_opt_a]: 0.0255391, [3]
    [Cycle 1]: 0.00755036, [27]
        [switch_simplify]: 9.075e-05
        [loop_unroll]: 4.053e-05
        [a_1]: 0.00096621
        [with_stream_mark]: 3.829e-05
        [recompute_prepare]: 2.005e-05
        [updatestate_depend_eliminate]: 1.022e-05
        [updatestate_assign_eliminate]: 6.61e-06
        [updatestate_loads_eliminate]: 5.94e-06
        [parameter_eliminate]: 2.66e-06
        [specialize_transform]: 1.368e-05
        [updatestate_useless_node_eliminater]: 1.405e-05
        [accelerated_algorithm]: 1.377e-05
        [meta_shard_fg_expand]: 4.90001e-06
        [get_grad_eliminate_]: 1.324e-05
        [merge_forward]: 9.86e-06
        [cell_reuse_recompute_pass]: 1.49998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.798e-05
        [j_node_and_user_rematch]: 0.00014973
        [meta_fg_expand]: 0.00189891
        [replace_old_param]: 6.335e-05
        [inline_without_move]: 5.773e-05
        [renormalize]: 0.00350608
        [add_forward_monad_depend]: 1.35e-05
        [auto_monad_grad]: 6.61e-06
        [auto_monad_eliminator]: 5.377e-05
        [cse]: 0.00017042
        [replace_applicator]: 8.27e-05
    [Cycle 2]: 0.0023032, [27]
        [switch_simplify]: 3.688e-05
        [loop_unroll]: 3.518e-05
        [a_1]: 0.00115347
        [with_stream_mark]: 1.781e-05
        [recompute_prepare]: 8.70001e-06
        [updatestate_depend_eliminate]: 4.45999e-06
        [updatestate_assign_eliminate]: 3.91999e-06
        [updatestate_loads_eliminate]: 3.39001e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 7.41999e-06
        [updatestate_useless_node_eliminater]: 6.94999e-06
        [accelerated_algorithm]: 6.44999e-06
        [meta_shard_fg_expand]: 2.64999e-06
        [get_grad_eliminate_]: 7.36999e-06
        [merge_forward]: 4.50999e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.707e-05
        [j_node_and_user_rematch]: 1.04e-05
        [meta_fg_expand]: 5.731e-05
        [replace_old_param]: 1.44e-05
        [inline_without_move]: 7.75e-06
        [renormalize]: 0.00066878
        [add_forward_monad_depend]: 5.29e-06
        [auto_monad_grad]: 1.81e-06
        [auto_monad_eliminator]: 1.427e-05
        [cse]: 2.519e-05
        [replace_applicator]: 1.547e-05
    [Cycle 3]: 0.00038651, [27]
        [switch_simplify]: 7.58999e-06
        [loop_unroll]: 6.39001e-06
        [a_1]: 0.00012147
        [with_stream_mark]: 1.04e-05
        [recompute_prepare]: 6.09999e-06
        [updatestate_depend_eliminate]: 3.97e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.72001e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 6.86001e-06
        [updatestate_useless_node_eliminater]: 6.21998e-06
        [accelerated_algorithm]: 6.68e-06
        [meta_shard_fg_expand]: 1.60001e-06
        [get_grad_eliminate_]: 5.47999e-06
        [merge_forward]: 3.09999e-06
        [cell_reuse_recompute_pass]: 2.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.576e-05
        [j_node_and_user_rematch]: 1.004e-05
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 9.47001e-06
        [inline_without_move]: 6.43998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 7.56999e-06
        [cse]: 1.476e-05
        [replace_applicator]: 6.07001e-06
[py_interpret_to_execute_after_opt_a]: 1.371e-05
[rewriter_after_opt_a]: 4.261e-05
[convert_after_rewriter]: 8.37e-06
[order_py_execute_after_rewriter]: 5.81998e-06
[mutable_eliminate]: 0.00086238
[jit_opt_b]: 5.927e-05, [1]
    [Cycle 1]: 5.178e-05, [2]
        [frontend_op_eliminate]: 2.103e-05
        [inline_after_opt_a]: 1.909e-05
[cconv]: 2.623e-05
[loop_unroll]: 0.00047743
[jit_opt_after_cconv]: 0.00022824, [1]
    [Cycle 1]: 0.00022167, [11]
        [c_1]: 2.706e-05
        [parameter_eliminate]: 3.83001e-06
        [updatestate_depend_eliminate]: 6.26e-06
        [updatestate_assign_eliminate]: 4.05e-06
        [updatestate_loads_eliminate]: 2.79999e-06
        [cse]: 2.623e-05
        [call_graph_tuple_transform]: 8.287e-05
        [tuple_list_get_item_eliminator]: 7.37997e-06
        [none_parameter_eliminate]: 1.84998e-06
        [renormalize]: 3.09985e-07
        [switch_simplify]: 6.89999e-06
[remove_dup_value]: 1.761e-05
[partial_unused_args_eliminate]: 2.32999e-06
[environ_conv]: 5.57999e-06
[add_recomputation]: 5.961e-05
[cse_after_recomputation]: 3.032e-05, [1]
    [Cycle 1]: 2.34e-05, [1]
        [cse]: 1.703e-05
[auto_monad_reorder]: 2.016e-05
[get_jit_bprop_graph]: 2.54001e-06
[rewriter_after_jit_bprop_graph]: 7.33999e-06
[opt_after_jit_grad]: 0.00052133
[symbol_engine_optimizer]: 8.646e-05, [1]
    [Cycle 1]: 7.899e-05, [6]
        [build]: 4.21001e-06
        [elim_shapecalc]: 9.15999e-06
        [elim_not_effective]: 1.702e-05
        [opt_reshape]: 7.37997e-06
        [fold_const_symbol]: 1.15e-05
        [renormalize]: 7.00005e-07
[validate]: 4.959e-05
[backend_pass]: 1.34e-06
[task_emit]: 0.0152624
[execute]: 6.99001e-06
Sums
    bootstrap                            :     0.000277s :  0.61%
    type_inference                       :     0.017169s : 38.04%
    event_method                         :     0.000056s :  0.12%
    auto_monad                           :     0.000109s :  0.24%
    graph_reusing                        :     0.000008s :  0.02%
    pre_auto_parallel                    :     0.000004s :  0.01%
    py_interpret_to_execute              :     0.000037s :  0.08%
    rewriter_before_opt_a                :     0.000102s :  0.23%
    expand_dump_flag                     :     0.000005s :  0.01%
    jit_opt_a.switch_simplify            :     0.000135s :  0.30%
    jit_opt_a.loop_unroll                :     0.000082s :  0.18%
    jit_opt_a.a_1                        :     0.002241s :  4.97%
    jit_opt_a.with_stream_mark           :     0.000067s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000035s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000019s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.01%
    jit_opt_a.specialize_transform       :     0.000028s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000027s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000027s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000009s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000026s :  0.06%
    jit_opt_a.merge_forward              :     0.000017s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000061s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000170s :  0.38%
    jit_opt_a.meta_fg_expand             :     0.001958s :  4.34%
    jit_opt_a.replace_old_param          :     0.000087s :  0.19%
    jit_opt_a.inline_without_move        :     0.000072s :  0.16%
    jit_opt_a.renormalize                :     0.004175s :  9.25%
    jit_opt_a.add_forward_monad_depend   :     0.000020s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000076s :  0.17%
    jit_opt_a.cse                        :     0.000210s :  0.47%
    jit_opt_a.replace_applicator         :     0.000104s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.03%
    rewriter_after_opt_a                 :     0.000043s :  0.09%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.01%
    mutable_eliminate                    :     0.000862s :  1.91%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.04%
    cconv                                :     0.000026s :  0.06%
    loop_unroll                          :     0.000477s :  1.06%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000083s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000018s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.01%
    add_recomputation                    :     0.000060s :  0.13%
    cse_after_recomputation.cse          :     0.000017s :  0.04%
    auto_monad_reorder                   :     0.000020s :  0.04%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000521s :  1.16%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000050s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.015262s : 33.82%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000650   110
 0.45% :     0.000003s :      3: substitution.elim_not_effective
 0.31% :     0.000002s :      3: substitution.fold_const_symbol
 0.99% :     0.000006s :      4: substitution.graph_param_transform
67.62% :     0.000440s :     12: substitution.inline
 3.26% :     0.000021s :      2: substitution.inline_without_move
 1.74% :     0.000011s :     14: substitution.j_node_and_user_rematch
 1.88% :     0.000012s :      7: substitution.minmaximum_grad
 1.23% :     0.000008s :      5: substitution.partial_eliminate
 1.78% :     0.000012s :     14: substitution.remove_not_recompute_node
 5.37% :     0.000035s :     10: substitution.replace_applicator
 1.47% :     0.000010s :      7: substitution.replace_old_param
 0.45% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.00% :     0.000026s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.43% :     0.000016s :      7: substitution.tuple_list_get_item_depend_reorder
 7.02% :     0.000046s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.017058     2
90.86% :     0.015499s :      1: type_inference.infer
 9.14% :     0.001559s :      1: type_inference.specialize
------[replace.]   0.000171    19
65.52% :     0.000112s :     12: replace.inline
34.48% :     0.000059s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000432    19
94.78% :     0.000409s :     12: match.inline
 5.22% :     0.000023s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000401  2568
 1.63% :     0.000007s :     43: predicate.accumulaten_eliminater
 0.45% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.33% :     0.000005s :     43: predicate.addn_check_dump
 1.74% :     0.000007s :     43: predicate.addn_zero_filter
 2.42% :     0.000010s :     43: predicate.arithmetic_simplify
 1.49% :     0.000006s :     43: predicate.cast_eliminate
 0.16% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.35% :     0.000005s :     43: predicate.compare_switch_simplify
 1.62% :     0.000007s :     43: predicate.depend_value_elim
 1.33% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.60% :     0.000006s :     43: predicate.dict_get_item_eliminator
 1.48% :     0.000006s :     43: predicate.dict_set_item_eliminator
 0.36% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.17% :     0.000001s :      4: predicate.elim_not_effective
 0.21% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.67% :     0.000007s :     43: predicate.environ_add_const_eliminate
 1.38% :     0.000006s :     43: predicate.environ_get_add_eliminate
 1.40% :     0.000006s :     43: predicate.environ_get_depend_swap
 1.58% :     0.000006s :     43: predicate.environ_get_eliminate
 1.41% :     0.000006s :     43: predicate.environ_get_set_eliminate
 0.08% :     0.000000s :      4: predicate.fold_const_symbol
 0.95% :     0.000004s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.30% :     0.000017s :     70: predicate.inline
 1.90% :     0.000008s :     40: predicate.inline_without_move
 0.59% :     0.000002s :     19: predicate.j_node_and_user_rematch
 1.02% :     0.000004s :     19: predicate.less_batch_normalization
 1.89% :     0.000008s :     50: predicate.list_to_tuple_eliminator_
 1.83% :     0.000007s :     54: predicate.load_eliminater
 0.60% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.11% :     0.000012s :     85: predicate.loop_unroll_before_grad
 1.72% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.34% :     0.000005s :     43: predicate.merge_addn
 1.38% :     0.000006s :     43: predicate.minmaximum_grad
 0.73% :     0.000003s :      4: predicate.mutable_eliminate
 0.17% :     0.000001s :      4: predicate.opt_reshape
 2.25% :     0.000009s :     54: predicate.partial_eliminate
 1.46% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.88% :     0.000008s :     43: predicate.reduce_eliminate
 1.80% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.34% :     0.000009s :     96: predicate.replace_applicator
 1.00% :     0.000004s :     40: predicate.replace_old_param
 0.09% :     0.000000s :      4: predicate.reset_defer_inline
 1.92% :     0.000008s :     43: predicate.reshape_eliminate
 1.51% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.36% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.41% :     0.000006s :     43: predicate.same_eliminate
 0.44% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.41% :     0.000002s :      8: predicate.special_op_eliminate
 0.84% :     0.000003s :     19: predicate.specialize_transform
 1.71% :     0.000007s :     43: predicate.split_environ_get_set_with_tuple_value
 1.52% :     0.000006s :     43: predicate.stack_unstack_eliminate
 0.15% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.69% :     0.000011s :     62: predicate.switch_defer_inline
 2.36% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.24% :     0.000025s :    151: predicate.switch_simplify
 1.60% :     0.000006s :     43: predicate.tile_eliminate
 1.63% :     0.000007s :     43: predicate.transpose_eliminate
 1.88% :     0.000008s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.53% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.21% :     0.000013s :     58: predicate.tuple_list_get_item_eliminator
 1.95% :     0.000008s :     43: predicate.tuple_list_set_item_eliminator
 1.75% :     0.000007s :     50: predicate.tuple_to_list_eliminator_
 1.82% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000011s :     73: predicate.updatestate_useless_node_eliminater
 1.90% :     0.000008s :     43: predicate.value_based_eliminate
 0.18% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.32% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001955    29
59.17% :     0.001157s :     13: func_graph_cloner_run.FuncGraphClonerGraph
40.83% :     0.000798s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.068762    91
 0.09% :     0.000063s :      1: add_recomputation
 0.17% :     0.000114s :      1: auto_monad
 0.03% :     0.000023s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.44% :     0.000303s :      1: bootstrap
 0.04% :     0.000029s :      1: cconv
 0.02% :     0.000011s :      1: convert_after_rewriter
 0.05% :     0.000033s :      1: cse_after_recomputation
 0.03% :     0.000018s :      1: environ_conv
 0.09% :     0.000061s :      1: event_method
 0.02% :     0.000010s :      1: execute
 0.01% :     0.000007s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
37.15% :     0.025543s :      1: jit_opt_a
 0.34% :     0.000232s :      1: jit_opt_after_cconv
 0.09% :     0.000062s :      1: jit_opt_b
 0.71% :     0.000486s :      1: loop_unroll
 1.27% :     0.000872s :      1: mutable_eliminate
 4.42% :     0.003041s :     39: opt.transform.jit_opt_a
 0.18% :     0.000120s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000033s :      4: opt.transform.jit_opt_b
 0.02% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.06% :     0.000042s :      4: opt.transform.symbol_engine_opt
 0.77% :     0.000531s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.06% :     0.000040s :      1: py_interpret_to_execute
 0.02% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000020s :      1: remove_dup_value
 2.96% :     0.002032s :      2: renormalize.infer
 3.09% :     0.002124s :      2: renormalize.specialize
 0.01% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000046s :      1: rewriter_after_opt_a
 0.15% :     0.000105s :      1: rewriter_before_opt_a
 0.13% :     0.000089s :      1: symbol_engine_optimizer
22.21% :     0.015275s :      1: task_emit
25.01% :     0.017196s :      1: type_inference
 0.11% :     0.000074s :      1: validate


TotalTime = 0.0117561, [30]
[bootstrap]: 0.00031959
[type_inference]: 0.00425184
[event_method]: 1.641e-05
[auto_monad]: 6.422e-05
[graph_reusing]: 5.44e-06
[pre_auto_parallel]: 2.99001e-06
[py_interpret_to_execute]: 3.476e-05
[rewriter_before_opt_a]: 7.123e-05
[expand_dump_flag]: 2.94999e-06
[jit_opt_a]: 0.00501937, [2]
    [Cycle 1]: 0.00148475, [27]
        [switch_simplify]: 4.828e-05
        [loop_unroll]: 2.971e-05
        [a_1]: 0.00053416
        [with_stream_mark]: 1.483e-05
        [recompute_prepare]: 6.89999e-06
        [updatestate_depend_eliminate]: 3.54002e-06
        [updatestate_assign_eliminate]: 2.91999e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.77001e-06
        [specialize_transform]: 6.11e-06
        [updatestate_useless_node_eliminater]: 5.18002e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 2.39001e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 3.9e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.438e-05
        [j_node_and_user_rematch]: 9.09e-06
        [meta_fg_expand]: 2.37001e-06
        [replace_old_param]: 7.92998e-06
        [inline_without_move]: 5.60001e-06
        [renormalize]: 0.00052104
        [add_forward_monad_depend]: 5.03002e-06
        [auto_monad_grad]: 2.22999e-06
        [auto_monad_eliminator]: 1.363e-05
        [cse]: 1.751e-05
        [replace_applicator]: 1.151e-05
    [Cycle 2]: 0.00032149, [27]
        [switch_simplify]: 5.52999e-06
        [loop_unroll]: 5.05999e-06
        [a_1]: 9.413e-05
        [with_stream_mark]: 7.8e-06
        [recompute_prepare]: 5.05999e-06
        [updatestate_depend_eliminate]: 2.36e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 4.95999e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 4.89e-06
        [meta_shard_fg_expand]: 1.02e-06
        [get_grad_eliminate_]: 4.4e-06
        [merge_forward]: 2.27999e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.228e-05
        [j_node_and_user_rematch]: 8.12998e-06
        [meta_fg_expand]: 1.47999e-06
        [replace_old_param]: 7.46001e-06
        [inline_without_move]: 4.89e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 8.10018e-07
        [auto_monad_eliminator]: 5.19e-06
        [cse]: 9.69999e-06
        [replace_applicator]: 1.047e-05
[py_interpret_to_execute_after_opt_a]: 8.20999e-06
[rewriter_after_opt_a]: 2.924e-05
[convert_after_rewriter]: 6.93998e-06
[order_py_execute_after_rewriter]: 4.74e-06
[mutable_eliminate]: 0.00047424
[jit_opt_b]: 5.21e-05, [1]
    [Cycle 1]: 4.543e-05, [2]
        [frontend_op_eliminate]: 1.964e-05
        [inline_after_opt_a]: 1.457e-05
[cconv]: 2.455e-05
[loop_unroll]: 0.00041499
[jit_opt_after_cconv]: 0.00013583, [1]
    [Cycle 1]: 0.00012906, [11]
        [c_1]: 1.986e-05
        [parameter_eliminate]: 2.53e-06
        [updatestate_depend_eliminate]: 5.72999e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.34999e-06
        [cse]: 1.659e-05
        [call_graph_tuple_transform]: 1.823e-05
        [tuple_list_get_item_eliminator]: 5.15001e-06
        [none_parameter_eliminate]: 1.32999e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.22999e-06
[remove_dup_value]: 1.44e-05
[partial_unused_args_eliminate]: 2.18998e-06
[environ_conv]: 5.20999e-06
[add_recomputation]: 4.501e-05
[cse_after_recomputation]: 1.962e-05, [1]
    [Cycle 1]: 1.408e-05, [1]
        [cse]: 8.55999e-06
[auto_monad_reorder]: 1.488e-05
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 5.17e-06
[opt_after_jit_grad]: 0.00044794
[symbol_engine_optimizer]: 6.964e-05, [1]
    [Cycle 1]: 6.304e-05, [6]
        [build]: 3.12997e-06
        [elim_shapecalc]: 6.86001e-06
        [elim_not_effective]: 1.199e-05
        [opt_reshape]: 5.44e-06
        [fold_const_symbol]: 8.25999e-06
        [renormalize]: 4.50003e-07
[validate]: 3.149e-05
Sums
    bootstrap                            :     0.000320s :  4.02%
    type_inference                       :     0.004252s : 53.46%
    event_method                         :     0.000016s :  0.21%
    auto_monad                           :     0.000064s :  0.81%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.04%
    py_interpret_to_execute              :     0.000035s :  0.44%
    rewriter_before_opt_a                :     0.000071s :  0.90%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000054s :  0.68%
    jit_opt_a.loop_unroll                :     0.000035s :  0.44%
    jit_opt_a.a_1                        :     0.000628s :  7.90%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000521s :  6.55%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.24%
    jit_opt_a.cse                        :     0.000027s :  0.34%
    jit_opt_a.replace_applicator         :     0.000022s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.10%
    rewriter_after_opt_a                 :     0.000029s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000474s :  5.96%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.25%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000025s :  0.31%
    loop_unroll                          :     0.000415s :  5.22%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000045s :  0.57%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.19%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000448s :  5.63%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000031s :  0.40%


Time group info:
------[substitution.]   0.000180    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.85% :     0.000005s :      3: substitution.graph_param_transform
82.58% :     0.000149s :      4: substitution.inline
 1.74% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.60% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000004s :      2: substitution.replace_old_param
 6.58% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004191     2
79.60% :     0.003336s :      1: type_inference.infer
20.40% :     0.000855s :      1: type_inference.specialize
------[replace.]   0.000096     6
77.53% :     0.000075s :      4: replace.inline
22.47% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000157     6
93.17% :     0.000147s :      4: match.inline
 6.83% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.55% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.01% :     0.000002s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.56% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.46% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.69% :     0.000007s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.62% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.52% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.31% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     10: predicate.reduce_eliminate
 1.62% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.59% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.43% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.39% :     0.000009s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.53% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.82% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.15% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000609    12
51.87% :     0.000316s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.58% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.55% :     0.000265s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013173    72
 0.36% :     0.000048s :      1: add_recomputation
 0.51% :     0.000068s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.56% :     0.000337s :      1: bootstrap
 0.20% :     0.000027s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.12% :     0.005022s :      1: jit_opt_a
 1.05% :     0.000139s :      1: jit_opt_after_cconv
 0.41% :     0.000055s :      1: jit_opt_b
 3.21% :     0.000422s :      1: loop_unroll
 3.67% :     0.000483s :      1: mutable_eliminate
 6.33% :     0.000834s :     26: opt.transform.jit_opt_a
 0.34% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000028s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.46% :     0.000456s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.04% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000038s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000017s :      1: remove_dup_value
 1.98% :     0.000261s :      1: renormalize.infer
 1.92% :     0.000254s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000032s :      1: rewriter_after_opt_a
 0.56% :     0.000074s :      1: rewriter_before_opt_a
 0.55% :     0.000072s :      1: symbol_engine_optimizer
32.36% :     0.004262s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.568.89 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc39160> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.575.24 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.016526, [33]
[bootstrap]: 0.00026051
[type_inference]: 0.00409446
[event_method]: 1.555e-05
[auto_monad]: 4.077e-05
[graph_reusing]: 3.75998e-06
[pre_auto_parallel]: 1.71e-06
[py_interpret_to_execute]: 3.165e-05
[rewriter_before_opt_a]: 6.182e-05
[expand_dump_flag]: 1.87999e-06
[jit_opt_a]: 0.00491909, [2]
    [Cycle 1]: 0.00132967, [27]
        [switch_simplify]: 4.293e-05
        [loop_unroll]: 2.815e-05
        [a_1]: 0.00046338
        [with_stream_mark]: 1.096e-05
        [recompute_prepare]: 7.21999e-06
        [updatestate_depend_eliminate]: 2.96999e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.07998e-06
        [specialize_transform]: 6.24999e-06
        [updatestate_useless_node_eliminater]: 5.35999e-06
        [accelerated_algorithm]: 5.24998e-06
        [meta_shard_fg_expand]: 1.28002e-06
        [get_grad_eliminate_]: 5.04998e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.06002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.124e-05
        [j_node_and_user_rematch]: 8.57e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 8.59998e-06
        [inline_without_move]: 5.27999e-06
        [renormalize]: 0.00050474
        [add_forward_monad_depend]: 4.67998e-06
        [auto_monad_grad]: 1.49e-06
        [auto_monad_eliminator]: 9.26002e-06
        [cse]: 1.471e-05
        [replace_applicator]: 1.139e-05
    [Cycle 2]: 0.00030961, [27]
        [switch_simplify]: 5.27999e-06
        [loop_unroll]: 5.00999e-06
        [a_1]: 8.867e-05
        [with_stream_mark]: 8.22e-06
        [recompute_prepare]: 4.57e-06
        [updatestate_depend_eliminate]: 2.28002e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 5.05999e-06
        [updatestate_useless_node_eliminater]: 4.64002e-06
        [accelerated_algorithm]: 4.65001e-06
        [meta_shard_fg_expand]: 1.00001e-06
        [get_grad_eliminate_]: 4.50999e-06
        [merge_forward]: 2.31e-06
        [cell_reuse_recompute_pass]: 1.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.218e-05
        [j_node_and_user_rematch]: 7.74002e-06
        [meta_fg_expand]: 1.37e-06
        [replace_old_param]: 6.79001e-06
        [inline_without_move]: 4.53001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.06002e-06
        [auto_monad_grad]: 7.99977e-07
        [auto_monad_eliminator]: 4.93001e-06
        [cse]: 9.44e-06
        [replace_applicator]: 1.012e-05
[py_interpret_to_execute_after_opt_a]: 7.4e-06
[rewriter_after_opt_a]: 2.373e-05
[convert_after_rewriter]: 5.14e-06
[order_py_execute_after_rewriter]: 3.91999e-06
[mutable_eliminate]: 0.00047617
[jit_opt_b]: 4.728e-05, [1]
    [Cycle 1]: 4.107e-05, [2]
        [frontend_op_eliminate]: 1.594e-05
        [inline_after_opt_a]: 1.414e-05
[cconv]: 1.48e-05
[loop_unroll]: 0.00043254
[jit_opt_after_cconv]: 0.00012877, [1]
    [Cycle 1]: 0.0001227, [11]
        [c_1]: 1.986e-05
        [parameter_eliminate]: 2.51e-06
        [updatestate_depend_eliminate]: 5.03002e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 1.467e-05
        [call_graph_tuple_transform]: 1.594e-05
        [tuple_list_get_item_eliminator]: 5.20001e-06
        [none_parameter_eliminate]: 8.09989e-07
        [renormalize]: 4.10015e-07
        [switch_simplify]: 5.00001e-06
[remove_dup_value]: 7.66001e-06
[partial_unused_args_eliminate]: 1.40999e-06
[environ_conv]: 3.45e-06
[add_recomputation]: 3.101e-05
[cse_after_recomputation]: 1.888e-05, [1]
    [Cycle 1]: 1.395e-05, [1]
        [cse]: 8.40999e-06
[auto_monad_reorder]: 9.09e-06
[get_jit_bprop_graph]: 1.39998e-06
[rewriter_after_jit_bprop_graph]: 4.12e-06
[opt_after_jit_grad]: 0.00044958
[symbol_engine_optimizer]: 6.993e-05, [1]
    [Cycle 1]: 6.379e-05, [6]
        [build]: 2.36e-06
        [elim_shapecalc]: 7.42002e-06
        [elim_not_effective]: 1.18e-05
        [opt_reshape]: 5.61e-06
        [fold_const_symbol]: 8.35001e-06
        [renormalize]: 4.2998e-07
[validate]: 2.488e-05
[backend_pass]: 6.90023e-07
[task_emit]: 0.00515171
[execute]: 8.17e-06
Sums
    bootstrap                            :     0.000261s :  2.05%
    type_inference                       :     0.004094s : 32.27%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000041s :  0.32%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000032s :  0.25%
    rewriter_before_opt_a                :     0.000062s :  0.49%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000048s :  0.38%
    jit_opt_a.loop_unroll                :     0.000033s :  0.26%
    jit_opt_a.a_1                        :     0.000552s :  4.35%
    jit_opt_a.with_stream_mark           :     0.000019s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000505s :  3.98%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.11%
    jit_opt_a.cse                        :     0.000024s :  0.19%
    jit_opt_a.replace_applicator         :     0.000022s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.06%
    rewriter_after_opt_a                 :     0.000024s :  0.19%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000476s :  3.75%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000015s :  0.12%
    loop_unroll                          :     0.000433s :  3.41%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000008s :  0.06%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000003s :  0.03%
    add_recomputation                    :     0.000031s :  0.24%
    cse_after_recomputation.cse          :     0.000008s :  0.07%
    auto_monad_reorder                   :     0.000009s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000450s :  3.54%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000025s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005152s : 40.60%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000146    23
 1.12% :     0.000002s :      2: substitution.elim_not_effective
 0.82% :     0.000001s :      2: substitution.fold_const_symbol
 2.45% :     0.000004s :      3: substitution.graph_param_transform
83.12% :     0.000121s :      4: substitution.inline
 1.87% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.75% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.62% :     0.000004s :      2: substitution.replace_old_param
 5.27% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004044     2
80.96% :     0.003274s :      1: type_inference.infer
19.04% :     0.000770s :      1: type_inference.specialize
------[replace.]   0.000061     6
66.89% :     0.000041s :      4: replace.inline
33.11% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000126     6
94.73% :     0.000119s :      4: match.inline
 5.27% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.42% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 1.88% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.47% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.33% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.29% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.19% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000002s :     15: predicate.partial_eliminate
 1.64% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.66% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.44% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.28% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.30% :     0.000008s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.42% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000541    12
46.45% :     0.000251s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.74% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.82% :     0.000264s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017878    76
 0.19% :     0.000034s :      1: add_recomputation
 0.24% :     0.000044s :      1: auto_monad
 0.06% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.52% :     0.000272s :      1: bootstrap
 0.10% :     0.000017s :      1: cconv
 0.04% :     0.000007s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.07% :     0.000012s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
27.53% :     0.004921s :      1: jit_opt_a
 0.73% :     0.000131s :      1: jit_opt_after_cconv
 0.28% :     0.000050s :      1: jit_opt_b
 2.46% :     0.000440s :      1: loop_unroll
 2.71% :     0.000484s :      1: mutable_eliminate
 4.18% :     0.000748s :     26: opt.transform.jit_opt_a
 0.24% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.56% :     0.000458s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000034s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000010s :      1: remove_dup_value
 1.40% :     0.000251s :      1: renormalize.infer
 1.39% :     0.000248s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000026s :      1: rewriter_after_opt_a
 0.36% :     0.000064s :      1: rewriter_before_opt_a
 0.40% :     0.000072s :      1: symbol_engine_optimizer
28.89% :     0.005165s :      1: task_emit
22.96% :     0.004105s :      1: type_inference
 0.22% :     0.000040s :      1: validate


TotalTime = 0.0130329, [30]
[bootstrap]: 0.00037634
[type_inference]: 0.00477505
[event_method]: 1.601e-05
[auto_monad]: 5.305e-05
[graph_reusing]: 4.25e-06
[pre_auto_parallel]: 1.95001e-06
[py_interpret_to_execute]: 3.259e-05
[rewriter_before_opt_a]: 6.491e-05
[expand_dump_flag]: 2.55002e-06
[jit_opt_a]: 0.00505129, [2]
    [Cycle 1]: 0.00158564, [27]
        [switch_simplify]: 4.407e-05
        [loop_unroll]: 2.837e-05
        [a_1]: 0.00049883
        [with_stream_mark]: 1.292e-05
        [recompute_prepare]: 7.54002e-06
        [updatestate_depend_eliminate]: 3.08998e-06
        [updatestate_assign_eliminate]: 2.13002e-06
        [updatestate_loads_eliminate]: 1.91998e-06
        [parameter_eliminate]: 1.11997e-06
        [specialize_transform]: 5.85002e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 6.51999e-06
        [meta_shard_fg_expand]: 1.45001e-06
        [get_grad_eliminate_]: 5.84999e-06
        [merge_forward]: 2.91999e-06
        [cell_reuse_recompute_pass]: 9.29984e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.52e-05
        [j_node_and_user_rematch]: 1.079e-05
        [meta_fg_expand]: 1.99999e-06
        [replace_old_param]: 1.032e-05
        [inline_without_move]: 6.23e-06
        [renormalize]: 0.00069985
        [add_forward_monad_depend]: 5.02e-06
        [auto_monad_grad]: 1.35999e-06
        [auto_monad_eliminator]: 9.29e-06
        [cse]: 1.825e-05
        [replace_applicator]: 1.151e-05
    [Cycle 2]: 0.00033673, [27]
        [switch_simplify]: 5.47999e-06
        [loop_unroll]: 5.82999e-06
        [a_1]: 9.685e-05
        [with_stream_mark]: 8.64e-06
        [recompute_prepare]: 4.74002e-06
        [updatestate_depend_eliminate]: 2.58e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 5.30999e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.92001e-06
        [meta_shard_fg_expand]: 1.40001e-06
        [get_grad_eliminate_]: 5.10001e-06
        [merge_forward]: 2.54999e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.326e-05
        [j_node_and_user_rematch]: 7.44002e-06
        [meta_fg_expand]: 1.59998e-06
        [replace_old_param]: 7.24001e-06
        [inline_without_move]: 4.99998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.08001e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 5.48997e-06
        [cse]: 1.163e-05
        [replace_applicator]: 5.50001e-06
[py_interpret_to_execute_after_opt_a]: 1.261e-05
[rewriter_after_opt_a]: 3.6e-05
[convert_after_rewriter]: 7.3e-06
[order_py_execute_after_rewriter]: 5.59e-06
[mutable_eliminate]: 0.0007204
[jit_opt_b]: 5.642e-05, [1]
    [Cycle 1]: 4.732e-05, [2]
        [frontend_op_eliminate]: 1.91e-05
        [inline_after_opt_a]: 1.533e-05
[cconv]: 2.694e-05
[loop_unroll]: 0.00053407
[jit_opt_after_cconv]: 0.00014785, [1]
    [Cycle 1]: 0.00013982, [11]
        [c_1]: 2.236e-05
        [parameter_eliminate]: 3.35003e-06
        [updatestate_depend_eliminate]: 5.46998e-06
        [updatestate_assign_eliminate]: 2.52001e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [cse]: 1.988e-05
        [call_graph_tuple_transform]: 1.903e-05
        [tuple_list_get_item_eliminator]: 5.23002e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 7.39994e-07
        [switch_simplify]: 5.29e-06
[remove_dup_value]: 1.664e-05
[partial_unused_args_eliminate]: 2.26998e-06
[environ_conv]: 5.87999e-06
[add_recomputation]: 5.049e-05
[cse_after_recomputation]: 2.177e-05, [1]
    [Cycle 1]: 1.571e-05, [1]
        [cse]: 9.19e-06
[auto_monad_reorder]: 1.559e-05
[get_jit_bprop_graph]: 2.30002e-06
[rewriter_after_jit_bprop_graph]: 6.83998e-06
[opt_after_jit_grad]: 0.00067167
[symbol_engine_optimizer]: 8.32e-05, [1]
    [Cycle 1]: 7.279e-05, [6]
        [build]: 5.28002e-06
        [elim_shapecalc]: 7.90998e-06
        [elim_not_effective]: 1.486e-05
        [opt_reshape]: 7.15e-06
        [fold_const_symbol]: 9.71998e-06
        [renormalize]: 4.89992e-07
[validate]: 4.225e-05
Sums
    bootstrap                            :     0.000376s :  4.05%
    type_inference                       :     0.004775s : 51.38%
    event_method                         :     0.000016s :  0.17%
    auto_monad                           :     0.000053s :  0.57%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.35%
    rewriter_before_opt_a                :     0.000065s :  0.70%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000050s :  0.53%
    jit_opt_a.loop_unroll                :     0.000034s :  0.37%
    jit_opt_a.a_1                        :     0.000596s :  6.41%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.12%
    jit_opt_a.merge_forward              :     0.000005s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000700s :  7.53%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.16%
    jit_opt_a.cse                        :     0.000030s :  0.32%
    jit_opt_a.replace_applicator         :     0.000017s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.14%
    rewriter_after_opt_a                 :     0.000036s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000720s :  7.75%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000027s :  0.29%
    loop_unroll                          :     0.000534s :  5.75%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000020s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000017s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000050s :  0.54%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000672s :  7.23%
    symbol_engine_optimizer.build        :     0.000005s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000042s :  0.45%


Time group info:
------[substitution.]   0.000170    23
 1.65% :     0.000003s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.86% :     0.000005s :      3: substitution.graph_param_transform
81.45% :     0.000139s :      4: substitution.inline
 1.80% :     0.000003s :      4: substitution.j_node_and_user_rematch
 3.21% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.17% :     0.000004s :      2: substitution.replace_old_param
 6.15% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004718     2
81.88% :     0.003863s :      1: type_inference.infer
18.12% :     0.000855s :      1: type_inference.specialize
------[replace.]   0.000063     6
66.14% :     0.000042s :      4: replace.inline
33.86% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000146     6
93.66% :     0.000136s :      4: match.inline
 6.34% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.92% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000002s :     10: predicate.addn_zero_filter
 2.46% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.28% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.70% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.62% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.24% :     0.000002s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 4.76% :     0.000006s :     22: predicate.inline
 1.04% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 2.16% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.71% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.31% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 2.26% :     0.000003s :      3: predicate.mutable_eliminate
 0.70% :     0.000001s :      3: predicate.opt_reshape
 1.91% :     0.000002s :     15: predicate.partial_eliminate
 1.25% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.08% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.45% :     0.000002s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 1.97% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.88% :     0.000009s :     53: predicate.switch_simplify
 1.23% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.44% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.08% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.84% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000676    12
52.42% :     0.000354s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.47% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.12% :     0.000291s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014599    72
 0.37% :     0.000053s :      1: add_recomputation
 0.39% :     0.000056s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 2.68% :     0.000391s :      1: bootstrap
 0.20% :     0.000030s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000024s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.13% :     0.000019s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
34.63% :     0.005055s :      1: jit_opt_a
 1.03% :     0.000150s :      1: jit_opt_after_cconv
 0.40% :     0.000059s :      1: jit_opt_b
 3.72% :     0.000543s :      1: loop_unroll
 5.01% :     0.000732s :      1: mutable_eliminate
 5.50% :     0.000803s :     26: opt.transform.jit_opt_a
 0.33% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000027s :      4: opt.transform.jit_opt_b
 0.11% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.15% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.20% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000036s :      4: opt.transform.symbol_engine_opt
 4.68% :     0.000684s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000035s :      1: py_interpret_to_execute
 0.11% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000019s :      1: remove_dup_value
 2.39% :     0.000349s :      1: renormalize.infer
 2.35% :     0.000344s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000040s :      1: rewriter_after_opt_a
 0.46% :     0.000068s :      1: rewriter_before_opt_a
 0.59% :     0.000086s :      1: symbol_engine_optimizer
32.78% :     0.004786s :      1: type_inference


TotalTime = 0.0321905, [33]
[bootstrap]: 0.00023011
[type_inference]: 0.0113801
[event_method]: 4.243e-05
[auto_monad]: 9.897e-05
[graph_reusing]: 7.51999e-06
[pre_auto_parallel]: 2.88003e-06
[py_interpret_to_execute]: 2.723e-05
[rewriter_before_opt_a]: 9.69e-05
[expand_dump_flag]: 3.95998e-06
[jit_opt_a]: 0.0117788, [3]
    [Cycle 1]: 0.00548369, [27]
        [switch_simplify]: 9.732e-05
        [loop_unroll]: 4.036e-05
        [a_1]: 0.00081824
        [with_stream_mark]: 2.496e-05
        [recompute_prepare]: 1.755e-05
        [updatestate_depend_eliminate]: 8.08001e-06
        [updatestate_assign_eliminate]: 6.88e-06
        [updatestate_loads_eliminate]: 6.51999e-06
        [parameter_eliminate]: 2.74001e-06
        [specialize_transform]: 1.299e-05
        [updatestate_useless_node_eliminater]: 1.171e-05
        [accelerated_algorithm]: 1.198e-05
        [meta_shard_fg_expand]: 3.49001e-06
        [get_grad_eliminate_]: 1.201e-05
        [merge_forward]: 7.86001e-06
        [cell_reuse_recompute_pass]: 9.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.621e-05
        [j_node_and_user_rematch]: 2.242e-05
        [meta_fg_expand]: 0.00142874
        [replace_old_param]: 5.297e-05
        [inline_without_move]: 5.061e-05
        [renormalize]: 0.00232074
        [add_forward_monad_depend]: 1.095e-05
        [auto_monad_grad]: 6.73998e-06
        [auto_monad_eliminator]: 5.183e-05
        [cse]: 0.00015348
        [replace_applicator]: 6.721e-05
    [Cycle 2]: 0.00222471, [27]
        [switch_simplify]: 3.589e-05
        [loop_unroll]: 3.46e-05
        [a_1]: 0.00110773
        [with_stream_mark]: 1.705e-05
        [recompute_prepare]: 1.073e-05
        [updatestate_depend_eliminate]: 4.82e-06
        [updatestate_assign_eliminate]: 3.25998e-06
        [updatestate_loads_eliminate]: 2.81999e-06
        [parameter_eliminate]: 1.52001e-06
        [specialize_transform]: 8.74e-06
        [updatestate_useless_node_eliminater]: 6.81999e-06
        [accelerated_algorithm]: 7.34002e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 6.03998e-06
        [merge_forward]: 4.13999e-06
        [cell_reuse_recompute_pass]: 1.99e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.585e-05
        [j_node_and_user_rematch]: 1.132e-05
        [meta_fg_expand]: 4.535e-05
        [replace_old_param]: 1.359e-05
        [inline_without_move]: 6.63e-06
        [renormalize]: 0.00063866
        [add_forward_monad_depend]: 5.10001e-06
        [auto_monad_grad]: 2.11e-06
        [auto_monad_eliminator]: 1.609e-05
        [cse]: 2.809e-05
        [replace_applicator]: 1.4e-05
    [Cycle 3]: 0.00045636, [27]
        [switch_simplify]: 6.41998e-06
        [loop_unroll]: 5.97999e-06
        [a_1]: 0.00011904
        [with_stream_mark]: 1.098e-05
        [recompute_prepare]: 6.41e-06
        [updatestate_depend_eliminate]: 3.68999e-06
        [updatestate_assign_eliminate]: 2.89999e-06
        [updatestate_loads_eliminate]: 2.53998e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 6.49001e-06
        [updatestate_useless_node_eliminater]: 5.85002e-06
        [accelerated_algorithm]: 6.28002e-06
        [meta_shard_fg_expand]: 1.79998e-06
        [get_grad_eliminate_]: 5.56002e-06
        [merge_forward]: 3.46999e-06
        [cell_reuse_recompute_pass]: 1.81e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.714e-05
        [j_node_and_user_rematch]: 1.015e-05
        [meta_fg_expand]: 2.89999e-06
        [replace_old_param]: 9.02999e-06
        [inline_without_move]: 6.919e-05
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.79998e-06
        [auto_monad_grad]: 1.60999e-06
        [auto_monad_eliminator]: 9.09e-06
        [cse]: 1.649e-05
        [replace_applicator]: 8.05e-06
[py_interpret_to_execute_after_opt_a]: 1.472e-05
[rewriter_after_opt_a]: 3.939e-05
[convert_after_rewriter]: 7.55e-06
[order_py_execute_after_rewriter]: 5.54998e-06
[mutable_eliminate]: 0.00065809
[jit_opt_b]: 5.858e-05, [1]
    [Cycle 1]: 5.074e-05, [2]
        [frontend_op_eliminate]: 2.016e-05
        [inline_after_opt_a]: 1.945e-05
[cconv]: 1.953e-05
[loop_unroll]: 0.00046309
[jit_opt_after_cconv]: 0.00016198, [1]
    [Cycle 1]: 0.00015535, [11]
        [c_1]: 2.626e-05
        [parameter_eliminate]: 3.18998e-06
        [updatestate_depend_eliminate]: 6.75002e-06
        [updatestate_assign_eliminate]: 2.97002e-06
        [updatestate_loads_eliminate]: 2.70997e-06
        [cse]: 2.832e-05
        [call_graph_tuple_transform]: 2.048e-05
        [tuple_list_get_item_eliminator]: 6.53998e-06
        [none_parameter_eliminate]: 1.22999e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 6.68e-06
[remove_dup_value]: 1.316e-05
[partial_unused_args_eliminate]: 2.02999e-06
[environ_conv]: 5.01997e-06
[add_recomputation]: 4.701e-05
[cse_after_recomputation]: 2.785e-05, [1]
    [Cycle 1]: 2.181e-05, [1]
        [cse]: 1.553e-05
[auto_monad_reorder]: 1.44e-05
[get_jit_bprop_graph]: 1.29e-06
[rewriter_after_jit_bprop_graph]: 5.89e-06
[opt_after_jit_grad]: 0.00048141
[symbol_engine_optimizer]: 8.592e-05, [1]
    [Cycle 1]: 7.897e-05, [6]
        [build]: 2.94999e-06
        [elim_shapecalc]: 1.037e-05
        [elim_not_effective]: 1.709e-05
        [opt_reshape]: 8.00999e-06
        [fold_const_symbol]: 1.046e-05
        [renormalize]: 9.50007e-07
[validate]: 3.66e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.00615098
[execute]: 4.68999e-06
Sums
    bootstrap                            :     0.000230s :  0.83%
    type_inference                       :     0.011380s : 41.03%
    event_method                         :     0.000042s :  0.15%
    auto_monad                           :     0.000099s :  0.36%
    graph_reusing                        :     0.000008s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000027s :  0.10%
    rewriter_before_opt_a                :     0.000097s :  0.35%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000140s :  0.50%
    jit_opt_a.loop_unroll                :     0.000081s :  0.29%
    jit_opt_a.a_1                        :     0.002045s :  7.37%
    jit_opt_a.with_stream_mark           :     0.000053s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000035s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000028s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000026s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.09%
    jit_opt_a.merge_forward              :     0.000015s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000059s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000044s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001477s :  5.33%
    jit_opt_a.replace_old_param          :     0.000076s :  0.27%
    jit_opt_a.inline_without_move        :     0.000126s :  0.46%
    jit_opt_a.renormalize                :     0.002959s : 10.67%
    jit_opt_a.add_forward_monad_depend   :     0.000018s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000010s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000077s :  0.28%
    jit_opt_a.cse                        :     0.000198s :  0.71%
    jit_opt_a.replace_applicator         :     0.000089s :  0.32%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.05%
    rewriter_after_opt_a                 :     0.000039s :  0.14%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000658s :  2.37%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.07%
    cconv                                :     0.000020s :  0.07%
    loop_unroll                          :     0.000463s :  1.67%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000013s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000047s :  0.17%
    cse_after_recomputation.cse          :     0.000016s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000481s :  1.74%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006151s : 22.18%
    execute                              :     0.000005s :  0.02%


Time group info:
------[substitution.]   0.000498   110
 0.46% :     0.000002s :      3: substitution.elim_not_effective
 0.31% :     0.000002s :      3: substitution.fold_const_symbol
 0.98% :     0.000005s :      4: substitution.graph_param_transform
64.27% :     0.000320s :     12: substitution.inline
 3.12% :     0.000016s :      2: substitution.inline_without_move
 1.80% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.19% :     0.000011s :      7: substitution.minmaximum_grad
 1.40% :     0.000007s :      5: substitution.partial_eliminate
 2.26% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.20% :     0.000026s :     10: substitution.replace_applicator
 1.83% :     0.000009s :      7: substitution.replace_old_param
 0.53% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.37% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.07% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 8.20% :     0.000041s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011284     2
89.68% :     0.010120s :      1: type_inference.infer
10.32% :     0.001164s :      1: type_inference.specialize
------[replace.]   0.000151    19
63.51% :     0.000096s :     12: replace.inline
36.49% :     0.000055s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000335    19
93.41% :     0.000312s :     12: match.inline
 6.59% :     0.000022s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000412  2568
 1.44% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.40% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.30% :     0.000005s :     43: predicate.addn_check_dump
 1.56% :     0.000006s :     43: predicate.addn_zero_filter
 1.76% :     0.000007s :     43: predicate.arithmetic_simplify
 1.45% :     0.000006s :     43: predicate.cast_eliminate
 0.16% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.32% :     0.000005s :     43: predicate.compare_switch_simplify
 1.52% :     0.000006s :     43: predicate.depend_value_elim
 1.40% :     0.000006s :     43: predicate.dict_get_item_const_eliminator
 1.40% :     0.000006s :     43: predicate.dict_get_item_eliminator
 1.31% :     0.000005s :     43: predicate.dict_set_item_eliminator
 0.28% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      4: predicate.elim_not_effective
 0.21% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000005s :     43: predicate.environ_add_const_eliminate
 1.30% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.30% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.35% :     0.000006s :     43: predicate.environ_get_eliminate
 1.36% :     0.000006s :     43: predicate.environ_get_set_eliminate
 0.08% :     0.000000s :      4: predicate.fold_const_symbol
 0.87% :     0.000004s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.32% :     0.000018s :     70: predicate.inline
 1.69% :     0.000007s :     40: predicate.inline_without_move
 0.33% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.92% :     0.000004s :     19: predicate.less_batch_normalization
 1.61% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.81% :     0.000007s :     54: predicate.load_eliminater
 0.44% :     0.000002s :      4: predicate.loop_unroll_after_grad
 2.96% :     0.000012s :     85: predicate.loop_unroll_before_grad
 1.76% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.31% :     0.000005s :     43: predicate.merge_addn
 1.35% :     0.000006s :     43: predicate.minmaximum_grad
 0.64% :     0.000003s :      4: predicate.mutable_eliminate
 0.25% :     0.000001s :      4: predicate.opt_reshape
 2.23% :     0.000009s :     54: predicate.partial_eliminate
 1.40% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.87% :     0.000008s :     43: predicate.reduce_eliminate
 1.60% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.32% :     0.000010s :     96: predicate.replace_applicator
 0.82% :     0.000003s :     40: predicate.replace_old_param
 0.12% :     0.000001s :      4: predicate.reset_defer_inline
 1.56% :     0.000006s :     43: predicate.reshape_eliminate
 1.44% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.26% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.40% :     0.000006s :     43: predicate.same_eliminate
 0.45% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.40% :     0.000002s :      8: predicate.special_op_eliminate
 0.85% :     0.000003s :     19: predicate.specialize_transform
 1.90% :     0.000008s :     43: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000006s :     43: predicate.stack_unstack_eliminate
 0.16% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.56% :     0.000011s :     62: predicate.switch_defer_inline
 2.29% :     0.000009s :     62: predicate.switch_layer_defer_inline
13.36% :     0.000055s :    151: predicate.switch_simplify
 1.39% :     0.000006s :     43: predicate.tile_eliminate
 1.31% :     0.000005s :     43: predicate.transpose_eliminate
 1.64% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.55% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 2.84% :     0.000012s :     58: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000007s :     43: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000007s :     50: predicate.tuple_to_list_eliminator_
 1.71% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.67% :     0.000011s :     73: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000007s :     43: predicate.value_based_eliminate
 0.13% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.23% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001597    29
58.36% :     0.000932s :     13: func_graph_cloner_run.FuncGraphClonerGraph
41.64% :     0.000665s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038025    91
 0.13% :     0.000051s :      1: add_recomputation
 0.27% :     0.000103s :      1: auto_monad
 0.04% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.64% :     0.000245s :      1: bootstrap
 0.06% :     0.000022s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000030s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.12% :     0.000046s :      1: event_method
 0.02% :     0.000008s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
30.98% :     0.011782s :      1: jit_opt_a
 0.43% :     0.000165s :      1: jit_opt_after_cconv
 0.16% :     0.000061s :      1: jit_opt_b
 1.24% :     0.000472s :      1: loop_unroll
 1.76% :     0.000668s :      1: mutable_eliminate
 7.23% :     0.002750s :     39: opt.transform.jit_opt_a
 0.15% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000033s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000042s :      4: opt.transform.symbol_engine_opt
 1.29% :     0.000490s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000030s :      1: py_interpret_to_execute
 0.05% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000016s :      1: remove_dup_value
 3.78% :     0.001436s :      2: renormalize.infer
 3.97% :     0.001508s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000043s :      1: rewriter_after_opt_a
 0.26% :     0.000100s :      1: rewriter_before_opt_a
 0.23% :     0.000089s :      1: symbol_engine_optimizer
16.20% :     0.006161s :      1: task_emit
29.99% :     0.011402s :      1: type_inference
 0.15% :     0.000058s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.979.40 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc39d30> should be wrapped by jit only once.

TotalTime = 0.0218726, [33]
[bootstrap]: 0.0002946
[type_inference]: 0.0059992
[event_method]: 1.814e-05
[auto_monad]: 6.638e-05
[graph_reusing]: 6.02999e-06
[pre_auto_parallel]: 2.56e-06
[py_interpret_to_execute]: 3.992e-05
[rewriter_before_opt_a]: 7.811e-05
[expand_dump_flag]: 3.62002e-06
[jit_opt_a]: 0.00623322, [2]
    [Cycle 1]: 0.00205921, [27]
        [switch_simplify]: 6.311e-05
        [loop_unroll]: 2.891e-05
        [a_1]: 0.00059585
        [with_stream_mark]: 2.447e-05
        [recompute_prepare]: 1.12e-05
        [updatestate_depend_eliminate]: 4.14002e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 2.07999e-06
        [specialize_transform]: 7.67998e-06
        [updatestate_useless_node_eliminater]: 5.39998e-06
        [accelerated_algorithm]: 6.51999e-06
        [meta_shard_fg_expand]: 2.35002e-06
        [get_grad_eliminate_]: 6.11998e-06
        [merge_forward]: 4.18999e-06
        [cell_reuse_recompute_pass]: 1.71998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.501e-05
        [j_node_and_user_rematch]: 9.29e-06
        [meta_fg_expand]: 2.36e-06
        [replace_old_param]: 1.099e-05
        [inline_without_move]: 6.30002e-06
        [renormalize]: 0.00095777
        [add_forward_monad_depend]: 7.57998e-06
        [auto_monad_grad]: 2.92002e-06
        [auto_monad_eliminator]: 1.51e-05
        [cse]: 3.036e-05
        [replace_applicator]: 1.686e-05
    [Cycle 2]: 0.00033174, [27]
        [switch_simplify]: 6.09999e-06
        [loop_unroll]: 5.47999e-06
        [a_1]: 9.107e-05
        [with_stream_mark]: 1.153e-05
        [recompute_prepare]: 5.80002e-06
        [updatestate_depend_eliminate]: 2.56998e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 1.58002e-06
        [specialize_transform]: 5.06002e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 5.64e-06
        [meta_shard_fg_expand]: 1.55999e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 2.61999e-06
        [cell_reuse_recompute_pass]: 2.49001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.386e-05
        [j_node_and_user_rematch]: 8.24002e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 8.71002e-06
        [inline_without_move]: 4.81997e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.28002e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 5.98998e-06
        [cse]: 1.027e-05
        [replace_applicator]: 4.87e-06
[py_interpret_to_execute_after_opt_a]: 1.286e-05
[rewriter_after_opt_a]: 3.69e-05
[convert_after_rewriter]: 6.56e-06
[order_py_execute_after_rewriter]: 4.59002e-06
[mutable_eliminate]: 0.00083097
[jit_opt_b]: 5.379e-05, [1]
    [Cycle 1]: 4.48e-05, [2]
        [frontend_op_eliminate]: 1.78e-05
        [inline_after_opt_a]: 1.564e-05
[cconv]: 2.889e-05
[loop_unroll]: 0.00046477
[jit_opt_after_cconv]: 0.00015339, [1]
    [Cycle 1]: 0.00014684, [11]
        [c_1]: 2.063e-05
        [parameter_eliminate]: 4.12998e-06
        [updatestate_depend_eliminate]: 8.08001e-06
        [updatestate_assign_eliminate]: 2.76e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [cse]: 2.633e-05
        [call_graph_tuple_transform]: 2.152e-05
        [tuple_list_get_item_eliminator]: 5.17999e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 1.02998e-06
        [switch_simplify]: 5.33002e-06
[remove_dup_value]: 1.461e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 5.59e-06
[add_recomputation]: 5.641e-05
[cse_after_recomputation]: 2.116e-05, [1]
    [Cycle 1]: 1.534e-05, [1]
        [cse]: 9.49e-06
[auto_monad_reorder]: 1.494e-05
[get_jit_bprop_graph]: 2.27001e-06
[rewriter_after_jit_bprop_graph]: 8.28001e-06
[opt_after_jit_grad]: 0.00048061
[symbol_engine_optimizer]: 7.791e-05, [1]
    [Cycle 1]: 7.065e-05, [6]
        [build]: 4.93001e-06
        [elim_shapecalc]: 8.05e-06
        [elim_not_effective]: 1.469e-05
        [opt_reshape]: 5.56e-06
        [fold_const_symbol]: 8.92e-06
        [renormalize]: 6.00005e-07
[validate]: 4.591e-05
[backend_pass]: 1.17999e-06
[task_emit]: 0.00656317
[execute]: 9.35001e-06
Sums
    bootstrap                            :     0.000295s :  1.70%
    type_inference                       :     0.005999s : 34.59%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000066s :  0.38%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000040s :  0.23%
    rewriter_before_opt_a                :     0.000078s :  0.45%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000069s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000687s :  3.96%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000958s :  5.52%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.12%
    jit_opt_a.cse                        :     0.000041s :  0.23%
    jit_opt_a.replace_applicator         :     0.000022s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000831s :  4.79%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000029s :  0.17%
    loop_unroll                          :     0.000465s :  2.68%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000056s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000481s :  2.77%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006563s : 37.84%
    execute                              :     0.000009s :  0.05%

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.985.72 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

Time group info:
------[substitution.]   0.000256    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000002s :      2: substitution.fold_const_symbol
 2.28% :     0.000006s :      3: substitution.graph_param_transform
85.57% :     0.000219s :      4: substitution.inline
 1.52% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.73% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.44% :     0.000006s :      2: substitution.replace_old_param
 4.94% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005924     2
84.09% :     0.004981s :      1: type_inference.infer
15.91% :     0.000942s :      1: type_inference.specialize
------[replace.]   0.000075     6
68.13% :     0.000051s :      4: replace.inline
31.87% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000228     6
94.91% :     0.000217s :      4: match.inline
 5.09% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 1.97% :     0.000002s :     10: predicate.arithmetic_simplify
 1.90% :     0.000002s :     10: predicate.cast_eliminate
 0.79% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 0.97% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.15% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.04% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.04% :     0.000001s :     10: predicate.environ_get_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 6.45% :     0.000008s :     22: predicate.inline
 1.11% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.64% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.02% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.25% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.63% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.72% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.08% :     0.000003s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.24% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.45% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.17% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.68% :     0.000003s :     16: predicate.switch_defer_inline
 1.99% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.29% :     0.000009s :     53: predicate.switch_simplify
 1.45% :     0.000002s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.80% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000637    12
44.47% :     0.000283s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.22% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.31% :     0.000302s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023851    76
 0.25% :     0.000059s :      1: add_recomputation
 0.30% :     0.000070s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.30% :     0.000310s :      1: bootstrap
 0.13% :     0.000032s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.15% :     0.006237s :      1: jit_opt_a
 0.66% :     0.000156s :      1: jit_opt_after_cconv
 0.24% :     0.000057s :      1: jit_opt_b
 1.99% :     0.000474s :      1: loop_unroll
 3.54% :     0.000845s :      1: mutable_eliminate
 3.86% :     0.000921s :     26: opt.transform.jit_opt_a
 0.21% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.05% :     0.000489s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000043s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.47% :     0.000589s :      1: renormalize.infer
 1.49% :     0.000356s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000040s :      1: rewriter_after_opt_a
 0.34% :     0.000082s :      1: rewriter_before_opt_a
 0.34% :     0.000080s :      1: symbol_engine_optimizer
27.60% :     0.006583s :      1: task_emit
25.24% :     0.006019s :      1: type_inference
 0.29% :     0.000070s :      1: validate


TotalTime = 0.0130961, [30]
[bootstrap]: 0.00034172
[type_inference]: 0.00507976
[event_method]: 1.687e-05
[auto_monad]: 6.44e-05
[graph_reusing]: 5.24e-06
[pre_auto_parallel]: 2.37999e-06
[py_interpret_to_execute]: 3.798e-05
[rewriter_before_opt_a]: 7.25e-05
[expand_dump_flag]: 3.86999e-06
[jit_opt_a]: 0.0054667, [2]
    [Cycle 1]: 0.00158058, [27]
        [switch_simplify]: 5.696e-05
        [loop_unroll]: 3.017e-05
        [a_1]: 0.00054036
        [with_stream_mark]: 1.64e-05
        [recompute_prepare]: 6.88998e-06
        [updatestate_depend_eliminate]: 3.64002e-06
        [updatestate_assign_eliminate]: 2.99001e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 1.76998e-06
        [specialize_transform]: 5.49998e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.57001e-06
        [meta_shard_fg_expand]: 2.09e-06
        [get_grad_eliminate_]: 5.34998e-06
        [merge_forward]: 3.77002e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.395e-05
        [j_node_and_user_rematch]: 9.61e-06
        [meta_fg_expand]: 2.14999e-06
        [replace_old_param]: 8.55001e-06
        [inline_without_move]: 5.38002e-06
        [renormalize]: 0.00062155
        [add_forward_monad_depend]: 5.12e-06
        [auto_monad_grad]: 2.37999e-06
        [auto_monad_eliminator]: 1.337e-05
        [cse]: 2.809e-05
        [replace_applicator]: 1.11e-05
    [Cycle 2]: 0.00031497, [27]
        [switch_simplify]: 5.27001e-06
        [loop_unroll]: 4.78001e-06
        [a_1]: 8.998e-05
        [with_stream_mark]: 8.80001e-06
        [recompute_prepare]: 4.79e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.86998e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 4.79e-06
        [updatestate_useless_node_eliminater]: 4.61002e-06
        [accelerated_algorithm]: 4.92999e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 4.62998e-06
        [merge_forward]: 2.27001e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.189e-05
        [j_node_and_user_rematch]: 7.43e-06
        [meta_fg_expand]: 1.45001e-06
        [replace_old_param]: 7.61999e-06
        [inline_without_move]: 4.77e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 1.15001e-06
        [auto_monad_eliminator]: 4.90999e-06
        [cse]: 1.036e-05
        [replace_applicator]: 4.97999e-06
[py_interpret_to_execute_after_opt_a]: 9.20999e-06
[rewriter_after_opt_a]: 3.159e-05
[convert_after_rewriter]: 6.38e-06
[order_py_execute_after_rewriter]: 4.66002e-06
[mutable_eliminate]: 0.00049144
[jit_opt_b]: 5.12e-05, [1]
    [Cycle 1]: 4.497e-05, [2]
        [frontend_op_eliminate]: 1.916e-05
        [inline_after_opt_a]: 1.469e-05
[cconv]: 2.503e-05
[loop_unroll]: 0.00041449
[jit_opt_after_cconv]: 0.00013396, [1]
    [Cycle 1]: 0.0001275, [11]
        [c_1]: 2.051e-05
        [parameter_eliminate]: 2.66999e-06
        [updatestate_depend_eliminate]: 5.24998e-06
        [updatestate_assign_eliminate]: 2.31998e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [cse]: 1.723e-05
        [call_graph_tuple_transform]: 1.684e-05
        [tuple_list_get_item_eliminator]: 5.17e-06
        [none_parameter_eliminate]: 1.19e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 5.27999e-06
[remove_dup_value]: 1.494e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 4.89003e-06
[add_recomputation]: 4.554e-05
[cse_after_recomputation]: 1.999e-05, [1]
    [Cycle 1]: 1.46e-05, [1]
        [cse]: 8.89e-06
[auto_monad_reorder]: 1.493e-05
[get_jit_bprop_graph]: 1.44998e-06
[rewriter_after_jit_bprop_graph]: 4.76002e-06
[opt_after_jit_grad]: 0.00044759
[symbol_engine_optimizer]: 7.092e-05, [1]
    [Cycle 1]: 6.468e-05, [6]
        [build]: 3.30003e-06
        [elim_shapecalc]: 7.39002e-06
        [elim_not_effective]: 1.216e-05
        [opt_reshape]: 5.50001e-06
        [fold_const_symbol]: 8.29998e-06
        [renormalize]: 4.99975e-07
[validate]: 3.151e-05
Sums
    bootstrap                            :     0.000342s :  3.82%
    type_inference                       :     0.005080s : 56.78%
    event_method                         :     0.000017s :  0.19%
    auto_monad                           :     0.000064s :  0.72%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000038s :  0.42%
    rewriter_before_opt_a                :     0.000072s :  0.81%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000062s :  0.70%
    jit_opt_a.loop_unroll                :     0.000035s :  0.39%
    jit_opt_a.a_1                        :     0.000630s :  7.05%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000010s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.18%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000622s :  6.95%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.20%
    jit_opt_a.cse                        :     0.000038s :  0.43%
    jit_opt_a.replace_applicator         :     0.000016s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.10%
    rewriter_after_opt_a                 :     0.000032s :  0.35%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000491s :  5.49%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000025s :  0.28%
    loop_unroll                          :     0.000414s :  4.63%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000046s :  0.51%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.17%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000448s :  5.00%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000032s :  0.35%


Time group info:
------[substitution.]   0.000204    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.30% :     0.000005s :      3: substitution.graph_param_transform
84.50% :     0.000172s :      4: substitution.inline
 1.55% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.10% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.97% :     0.000004s :      2: substitution.replace_old_param
 6.00% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005004     2
81.49% :     0.004077s :      1: type_inference.infer
18.51% :     0.000926s :      1: type_inference.specialize
------[replace.]   0.000067     6
66.60% :     0.000044s :      4: replace.inline
33.40% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000181     6
93.85% :     0.000170s :      4: match.inline
 6.15% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.53% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.47% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.58% :     0.000003s :     10: predicate.arithmetic_simplify
 1.11% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.26% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.46% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.41% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.28% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.79% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.75% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.83% :     0.000002s :     10: predicate.reduce_eliminate
 1.62% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.57% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.45% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.83% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.23% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.39% :     0.000009s :     53: predicate.switch_simplify
 1.08% :     0.000001s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.09% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000674    12
52.84% :     0.000356s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.11% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.06% :     0.000290s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014609    72
 0.33% :     0.000048s :      1: add_recomputation
 0.46% :     0.000068s :      1: auto_monad
 0.12% :     0.000017s :      1: auto_monad_reorder
 2.50% :     0.000366s :      1: bootstrap
 0.19% :     0.000028s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000021s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
37.44% :     0.005470s :      1: jit_opt_a
 0.93% :     0.000137s :      1: jit_opt_after_cconv
 0.37% :     0.000054s :      1: jit_opt_b
 2.89% :     0.000422s :      1: loop_unroll
 3.42% :     0.000500s :      1: mutable_eliminate
 5.73% :     0.000837s :     26: opt.transform.jit_opt_a
 0.30% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000028s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.12% :     0.000456s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.28% :     0.000041s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.18% :     0.000319s :      1: renormalize.infer
 2.02% :     0.000295s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000034s :      1: rewriter_after_opt_a
 0.52% :     0.000075s :      1: rewriter_before_opt_a
 0.50% :     0.000073s :      1: symbol_engine_optimizer
34.89% :     0.005097s :      1: type_inference


TotalTime = 0.0182403, [33]
[bootstrap]: 0.00033637
[type_inference]: 0.004707
[event_method]: 1.61e-05
[auto_monad]: 5.4e-05
[graph_reusing]: 4.74e-06
[pre_auto_parallel]: 2.19001e-06
[py_interpret_to_execute]: 4.119e-05
[rewriter_before_opt_a]: 6.792e-05
[expand_dump_flag]: 2.77002e-06
[jit_opt_a]: 0.00532205, [2]
    [Cycle 1]: 0.00148599, [27]
        [switch_simplify]: 5.031e-05
        [loop_unroll]: 2.902e-05
        [a_1]: 0.00051192
        [with_stream_mark]: 1.622e-05
        [recompute_prepare]: 7.33e-06
        [updatestate_depend_eliminate]: 3.13e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.87001e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 5.26002e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 2.44001e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.03e-05
        [j_node_and_user_rematch]: 8.57e-06
        [meta_fg_expand]: 1.72001e-06
        [replace_old_param]: 7.70998e-06
        [inline_without_move]: 5.30999e-06
        [renormalize]: 0.00060355
        [add_forward_monad_depend]: 4.61002e-06
        [auto_monad_grad]: 1.20999e-06
        [auto_monad_eliminator]: 9.83002e-06
        [cse]: 1.542e-05
        [replace_applicator]: 1.164e-05
    [Cycle 2]: 0.00031371, [27]
        [switch_simplify]: 5.51e-06
        [loop_unroll]: 4.84003e-06
        [a_1]: 8.948e-05
        [with_stream_mark]: 8.72e-06
        [recompute_prepare]: 4.60001e-06
        [updatestate_depend_eliminate]: 2.51998e-06
        [updatestate_assign_eliminate]: 1.99999e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 9.10019e-07
        [specialize_transform]: 4.98001e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 4.95001e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.30999e-06
        [merge_forward]: 2.30002e-06
        [cell_reuse_recompute_pass]: 2.16e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.161e-05
        [j_node_and_user_rematch]: 7.85e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 1.224e-05
        [inline_without_move]: 4.60001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 4.83001e-06
        [cse]: 9.77999e-06
        [replace_applicator]: 4.99e-06
[py_interpret_to_execute_after_opt_a]: 8.13001e-06
[rewriter_after_opt_a]: 3.147e-05
[convert_after_rewriter]: 6.28e-06
[order_py_execute_after_rewriter]: 4.13999e-06
[mutable_eliminate]: 0.00051991
[jit_opt_b]: 4.9e-05, [1]
    [Cycle 1]: 4.279e-05, [2]
        [frontend_op_eliminate]: 1.727e-05
        [inline_after_opt_a]: 1.418e-05
[cconv]: 1.549e-05
[loop_unroll]: 0.00047625
[jit_opt_after_cconv]: 0.00013179, [1]
    [Cycle 1]: 0.00012599, [11]
        [c_1]: 1.965e-05
        [parameter_eliminate]: 2.44999e-06
        [updatestate_depend_eliminate]: 5.53002e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.01998e-06
        [cse]: 1.581e-05
        [call_graph_tuple_transform]: 1.693e-05
        [tuple_list_get_item_eliminator]: 5.10999e-06
        [none_parameter_eliminate]: 8.70001e-07
        [renormalize]: 5.89993e-07
        [switch_simplify]: 5.19998e-06
[remove_dup_value]: 6.31e-06
[partial_unused_args_eliminate]: 1.83002e-06
[environ_conv]: 3.44001e-06
[add_recomputation]: 3.767e-05
[cse_after_recomputation]: 1.967e-05, [1]
    [Cycle 1]: 1.431e-05, [1]
        [cse]: 8.71997e-06
[auto_monad_reorder]: 9.57999e-06
[get_jit_bprop_graph]: 1.48002e-06
[rewriter_after_jit_bprop_graph]: 4.47998e-06
[opt_after_jit_grad]: 0.00045578
[symbol_engine_optimizer]: 9.59e-05, [1]
    [Cycle 1]: 6.294e-05, [6]
        [build]: 2.51e-06
        [elim_shapecalc]: 6.94999e-06
        [elim_not_effective]: 1.209e-05
        [opt_reshape]: 5.61e-06
        [fold_const_symbol]: 8.63001e-06
        [renormalize]: 3.10014e-07
[validate]: 3.204e-05
[backend_pass]: 1.12999e-06
[task_emit]: 0.00559196
[execute]: 7.64002e-06
Sums
    bootstrap                            :     0.000336s :  2.38%
    type_inference                       :     0.004707s : 33.30%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000054s :  0.38%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000041s :  0.29%
    rewriter_before_opt_a                :     0.000068s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000601s :  4.26%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000604s :  4.27%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.10%
    jit_opt_a.cse                        :     0.000025s :  0.18%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000520s :  3.68%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000015s :  0.11%
    loop_unroll                          :     0.000476s :  3.37%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000006s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000003s :  0.02%
    add_recomputation                    :     0.000038s :  0.27%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000010s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000456s :  3.22%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005592s : 39.57%
    execute                              :     0.000008s :  0.05%


TotalTime = 0.0298715, [33]
[bootstrap]: 0.00022289
[type_inference]: 0.0101109
[event_method]: 4.109e-05
[auto_monad]: 8.203e-05
[graph_reusing]: 6.43e-06
[pre_auto_parallel]: 2.55002e-06
[py_interpret_to_execute]: 3.759e-05
[rewriter_before_opt_a]: 8.81e-05
[expand_dump_flag]: 2.68e-06
[jit_opt_a]: 0.0109775, [3]
    [Cycle 1]: 0.0054449, [27]
        [switch_simplify]: 6.008e-05
        [loop_unroll]: 3.898e-05
        [a_1]: 0.00085806
        [with_stream_mark]: 2.335e-05
        [recompute_prepare]: 1.806e-05
        [updatestate_depend_eliminate]: 8.42e-06
        [updatestate_assign_eliminate]: 6.02001e-06
        [updatestate_loads_eliminate]: 6.07999e-06
        [parameter_eliminate]: 1.86998e-06
        [specialize_transform]: 1.36e-05
        [updatestate_useless_node_eliminater]: 1.236e-05
        [accelerated_algorithm]: 1.301e-05
        [meta_shard_fg_expand]: 2.88e-06
        [get_grad_eliminate_]: 1.24e-05
        [merge_forward]: 7.4e-06
        [cell_reuse_recompute_pass]: 8.00006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.55e-05
        [j_node_and_user_rematch]: 2.332e-05
        [meta_fg_expand]: 0.00142292
        [replace_old_param]: 5.338e-05
        [inline_without_move]: 5.073e-05
        [renormalize]: 0.00230937
        [add_forward_monad_depend]: 9.93002e-06
        [auto_monad_grad]: 5.91998e-06
        [auto_monad_eliminator]: 4.953e-05
        [cse]: 0.00014034
        [replace_applicator]: 6.844e-05
    [Cycle 2]: 0.00198306, [27]
        [switch_simplify]: 3.794e-05
        [loop_unroll]: 3.525e-05
        [a_1]: 0.00104118
        [with_stream_mark]: 1.382e-05
        [recompute_prepare]: 8.33001e-06
        [updatestate_depend_eliminate]: 4.02e-06
        [updatestate_assign_eliminate]: 3.67002e-06
        [updatestate_loads_eliminate]: 3.28e-06
        [parameter_eliminate]: 1.48002e-06
        [specialize_transform]: 6.61999e-06
        [updatestate_useless_node_eliminater]: 6.51999e-06
        [accelerated_algorithm]: 5.99999e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 6.46e-06
        [merge_forward]: 3.65e-06
        [cell_reuse_recompute_pass]: 8.50006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.358e-05
        [j_node_and_user_rematch]: 1.108e-05
        [meta_fg_expand]: 4.233e-05
        [replace_old_param]: 9.82999e-06
        [inline_without_move]: 6.49001e-06
        [renormalize]: 0.00051802
        [add_forward_monad_depend]: 4.32e-06
        [auto_monad_grad]: 1.84e-06
        [auto_monad_eliminator]: 1.255e-05
        [cse]: 2.133e-05
        [replace_applicator]: 1.318e-05
    [Cycle 3]: 0.00037131, [27]
        [switch_simplify]: 7.02002e-06
        [loop_unroll]: 6.29999e-06
        [a_1]: 0.0001186
        [with_stream_mark]: 9.52999e-06
        [recompute_prepare]: 6.23e-06
        [updatestate_depend_eliminate]: 3.48999e-06
        [updatestate_assign_eliminate]: 2.68998e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 9.20001e-07
        [specialize_transform]: 6.44999e-06
        [updatestate_useless_node_eliminater]: 5.89e-06
        [accelerated_algorithm]: 5.78997e-06
        [meta_shard_fg_expand]: 1.62001e-06
        [get_grad_eliminate_]: 6.02001e-06
        [merge_forward]: 3.26999e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.43e-05
        [j_node_and_user_rematch]: 1.043e-05
        [meta_fg_expand]: 2.09999e-06
        [replace_old_param]: 8.43999e-06
        [inline_without_move]: 5.96998e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 6.72002e-06
        [cse]: 1.398e-05
        [replace_applicator]: 6.42001e-06
[py_interpret_to_execute_after_opt_a]: 1.199e-05
[rewriter_after_opt_a]: 3.693e-05
[convert_after_rewriter]: 7.61999e-06
[order_py_execute_after_rewriter]: 5.92999e-06
[mutable_eliminate]: 0.00056307
[jit_opt_b]: 5.494e-05, [1]
    [Cycle 1]: 4.808e-05, [2]
        [frontend_op_eliminate]: 2.008e-05
        [inline_after_opt_a]: 1.731e-05
[cconv]: 2.278e-05
[loop_unroll]: 0.00045842
[jit_opt_after_cconv]: 0.00015483, [1]
    [Cycle 1]: 0.00014775, [11]
        [c_1]: 2.61e-05
        [parameter_eliminate]: 2.46998e-06
        [updatestate_depend_eliminate]: 6.39999e-06
        [updatestate_assign_eliminate]: 2.86e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [cse]: 2.17e-05
        [call_graph_tuple_transform]: 2.011e-05
        [tuple_list_get_item_eliminator]: 6.59999e-06
        [none_parameter_eliminate]: 1.99999e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 6.09001e-06
[remove_dup_value]: 1.665e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 5.54998e-06
[add_recomputation]: 4.938e-05
[cse_after_recomputation]: 3.423e-05, [1]
    [Cycle 1]: 2.83e-05, [1]
        [cse]: 2.252e-05
[auto_monad_reorder]: 1.997e-05
[get_jit_bprop_graph]: 1.69e-06
[rewriter_after_jit_bprop_graph]: 5.35001e-06
[opt_after_jit_grad]: 0.00046868
[symbol_engine_optimizer]: 8.193e-05, [1]
    [Cycle 1]: 7.524e-05, [6]
        [build]: 3.63999e-06
        [elim_shapecalc]: 9.15001e-06
        [elim_not_effective]: 1.519e-05
        [opt_reshape]: 7.33999e-06
        [fold_const_symbol]: 1.141e-05
        [renormalize]: 4.40021e-07
[validate]: 3.324e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00606946
[execute]: 4.70999e-06
Sums
    bootstrap                            :     0.000223s :  0.86%
    type_inference                       :     0.010111s : 39.01%
    event_method                         :     0.000041s :  0.16%
    auto_monad                           :     0.000082s :  0.32%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000038s :  0.15%
    rewriter_before_opt_a                :     0.000088s :  0.34%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000105s :  0.41%
    jit_opt_a.loop_unroll                :     0.000081s :  0.31%
    jit_opt_a.a_1                        :     0.002018s :  7.78%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000027s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000025s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.10%
    jit_opt_a.merge_forward              :     0.000014s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000053s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000045s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001467s :  5.66%
    jit_opt_a.replace_old_param          :     0.000072s :  0.28%
    jit_opt_a.inline_without_move        :     0.000063s :  0.24%
    jit_opt_a.renormalize                :     0.002827s : 10.91%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000069s :  0.27%
    jit_opt_a.cse                        :     0.000176s :  0.68%
    jit_opt_a.replace_applicator         :     0.000088s :  0.34%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.05%
    rewriter_after_opt_a                 :     0.000037s :  0.14%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000563s :  2.17%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.07%
    cconv                                :     0.000023s :  0.09%
    loop_unroll                          :     0.000458s :  1.77%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000017s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000049s :  0.19%
    cse_after_recomputation.cse          :     0.000023s :  0.09%
    auto_monad_reorder                   :     0.000020s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000469s :  1.81%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006069s : 23.42%
    execute                              :     0.000005s :  0.02%


Time group info:
------[substitution.]   0.000180    23
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 1.91% :     0.000003s :      3: substitution.graph_param_transform
84.32% :     0.000152s :      4: substitution.inline
 1.44% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.08% :     0.000004s :      4: substitution.remove_not_recompute_node
 4.52% :     0.000008s :      2: substitution.replace_old_param
 4.28% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004648     2
82.25% :     0.003822s :      1: type_inference.infer
17.75% :     0.000825s :      1: type_inference.specialize
------[replace.]   0.000064     6
66.81% :     0.000043s :      4: replace.inline
33.19% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000156     6
95.78% :     0.000150s :      4: match.inline
 4.22% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.59% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.51% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 4.16% :     0.000005s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.49% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.36% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.22% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 4.77% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.83% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.10% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.58% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000002s :     15: predicate.partial_eliminate
 1.30% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.43% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.58% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.55% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.64% :     0.000002s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.50% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.10% :     0.000008s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.48% :     0.000002s :     10: predicate.transpose_eliminate
 1.28% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.18% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.90% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.88% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000613    12
49.50% :     0.000303s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.40% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.09% :     0.000276s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019749    76
 0.21% :     0.000041s :      1: add_recomputation
 0.29% :     0.000057s :      1: auto_monad
 0.06% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.76% :     0.000347s :      1: bootstrap
 0.09% :     0.000018s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
26.96% :     0.005325s :      1: jit_opt_a
 0.68% :     0.000135s :      1: jit_opt_after_cconv
 0.26% :     0.000052s :      1: jit_opt_b
 2.45% :     0.000484s :      1: loop_unroll
 2.67% :     0.000527s :      1: mutable_eliminate
 4.08% :     0.000805s :     26: opt.transform.jit_opt_a
 0.22% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.34% :     0.000463s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.22% :     0.000044s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000008s :      1: remove_dup_value
 1.60% :     0.000316s :      1: renormalize.infer
 1.42% :     0.000281s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000035s :      1: rewriter_after_opt_a
 0.36% :     0.000071s :      1: rewriter_before_opt_a
 0.50% :     0.000098s :      1: symbol_engine_optimizer
28.37% :     0.005603s :      1: task_emit
23.88% :     0.004716s :      1: type_inference
 0.25% :     0.000048s :      1: validate


Time group info:
------[substitution.]   0.000507   110
 0.48% :     0.000002s :      3: substitution.elim_not_effective
 0.31% :     0.000002s :      3: substitution.fold_const_symbol
 1.11% :     0.000006s :      4: substitution.graph_param_transform
57.66% :     0.000292s :     12: substitution.inline
 3.10% :     0.000016s :      2: substitution.inline_without_move
 1.76% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.16% :     0.000011s :      7: substitution.minmaximum_grad
 1.33% :     0.000007s :      5: substitution.partial_eliminate
 2.22% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.45% :     0.000028s :     10: substitution.replace_applicator
 1.53% :     0.000008s :      7: substitution.replace_old_param
 0.46% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.19% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
10.80% :     0.000055s :      7: substitution.tuple_list_get_item_depend_reorder
 7.44% :     0.000038s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010046     2
90.00% :     0.009042s :      1: type_inference.infer
10.00% :     0.001005s :      1: type_inference.specialize
------[replace.]   0.000144    19
63.24% :     0.000091s :     12: replace.inline
36.76% :     0.000053s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000303    19
94.11% :     0.000285s :     12: match.inline
 5.89% :     0.000018s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000370  2568
 1.58% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.44% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.40% :     0.000005s :     43: predicate.addn_check_dump
 1.89% :     0.000007s :     43: predicate.addn_zero_filter
 2.05% :     0.000008s :     43: predicate.arithmetic_simplify
 1.63% :     0.000006s :     43: predicate.cast_eliminate
 0.17% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.39% :     0.000005s :     43: predicate.compare_switch_simplify
 1.68% :     0.000006s :     43: predicate.depend_value_elim
 1.40% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.47% :     0.000005s :     43: predicate.dict_get_item_eliminator
 1.55% :     0.000006s :     43: predicate.dict_set_item_eliminator
 0.31% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.18% :     0.000001s :      4: predicate.elim_not_effective
 0.18% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000005s :     43: predicate.environ_add_const_eliminate
 1.44% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.50% :     0.000006s :     43: predicate.environ_get_depend_swap
 1.48% :     0.000005s :     43: predicate.environ_get_eliminate
 1.44% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.11% :     0.000000s :      4: predicate.fold_const_symbol
 0.89% :     0.000003s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.33% :     0.000016s :     70: predicate.inline
 1.86% :     0.000007s :     40: predicate.inline_without_move
 0.38% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.92% :     0.000003s :     19: predicate.less_batch_normalization
 1.78% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.95% :     0.000007s :     54: predicate.load_eliminater
 0.53% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.19% :     0.000012s :     85: predicate.loop_unroll_before_grad
 1.75% :     0.000006s :     47: predicate.make_slice_get_slice_eliminator
 1.48% :     0.000005s :     43: predicate.merge_addn
 1.50% :     0.000006s :     43: predicate.minmaximum_grad
 0.47% :     0.000002s :      4: predicate.mutable_eliminate
 0.19% :     0.000001s :      4: predicate.opt_reshape
 2.37% :     0.000009s :     54: predicate.partial_eliminate
 1.59% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.98% :     0.000007s :     43: predicate.reduce_eliminate
 1.77% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.51% :     0.000009s :     96: predicate.replace_applicator
 0.97% :     0.000004s :     40: predicate.replace_old_param
 0.13% :     0.000000s :      4: predicate.reset_defer_inline
 1.68% :     0.000006s :     43: predicate.reshape_eliminate
 1.53% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.31% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.54% :     0.000006s :     43: predicate.same_eliminate
 0.49% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.45% :     0.000002s :      8: predicate.special_op_eliminate
 0.89% :     0.000003s :     19: predicate.specialize_transform
 1.75% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000006s :     43: predicate.stack_unstack_eliminate
 0.19% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.80% :     0.000010s :     62: predicate.switch_defer_inline
 2.50% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.31% :     0.000023s :    151: predicate.switch_simplify
 1.53% :     0.000006s :     43: predicate.tile_eliminate
 1.53% :     0.000006s :     43: predicate.transpose_eliminate
 1.81% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.66% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.15% :     0.000012s :     58: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000007s :     43: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000006s :     50: predicate.tuple_to_list_eliminator_
 1.98% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000010s :     73: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000007s :     43: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.25% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001440    29
57.63% :     0.000830s :     13: func_graph_cloner_run.FuncGraphClonerGraph
42.37% :     0.000610s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.035446    91
 0.15% :     0.000052s :      1: add_recomputation
 0.24% :     0.000086s :      1: auto_monad
 0.06% :     0.000023s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.67% :     0.000238s :      1: bootstrap
 0.07% :     0.000025s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000037s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.13% :     0.000045s :      1: event_method
 0.02% :     0.000008s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
30.98% :     0.010981s :      1: jit_opt_a
 0.44% :     0.000158s :      1: jit_opt_after_cconv
 0.16% :     0.000058s :      1: jit_opt_b
 1.32% :     0.000467s :      1: loop_unroll
 1.61% :     0.000572s :      1: mutable_eliminate
 7.38% :     0.002616s :     39: opt.transform.jit_opt_a
 0.16% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000031s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.35% :     0.000477s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000041s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 3.98% :     0.001411s :      2: renormalize.infer
 3.96% :     0.001404s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000040s :      1: rewriter_after_opt_a
 0.26% :     0.000091s :      1: rewriter_before_opt_a
 0.24% :     0.000085s :      1: symbol_engine_optimizer
17.15% :     0.006079s :      1: task_emit
28.56% :     0.010122s :      1: type_inference
 0.15% :     0.000055s :      1: validate


TotalTime = 0.0219803, [33]
[bootstrap]: 0.00037078
[type_inference]: 0.00559682
[event_method]: 2.018e-05
[auto_monad]: 6.88e-05
[graph_reusing]: 6.31e-06
[pre_auto_parallel]: 3.44001e-06
[py_interpret_to_execute]: 4.263e-05
[rewriter_before_opt_a]: 7.804e-05
[expand_dump_flag]: 3.36001e-06
[jit_opt_a]: 0.00611634, [2]
    [Cycle 1]: 0.00186283, [27]
        [switch_simplify]: 6.048e-05
        [loop_unroll]: 2.879e-05
        [a_1]: 0.0005837
        [with_stream_mark]: 2.583e-05
        [recompute_prepare]: 9.32999e-06
        [updatestate_depend_eliminate]: 4.73001e-06
        [updatestate_assign_eliminate]: 2.76e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 6.74999e-06
        [updatestate_useless_node_eliminater]: 5.64998e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 2.59001e-06
        [get_grad_eliminate_]: 5.60001e-06
        [merge_forward]: 4.78001e-06
        [cell_reuse_recompute_pass]: 1.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.621e-05
        [j_node_and_user_rematch]: 9.99001e-06
        [meta_fg_expand]: 2.27999e-06
        [replace_old_param]: 9.38002e-06
        [inline_without_move]: 5.14e-06
        [renormalize]: 0.00079287
        [add_forward_monad_depend]: 7.16999e-06
        [auto_monad_grad]: 2.66e-06
        [auto_monad_eliminator]: 1.758e-05
        [cse]: 2.986e-05
        [replace_applicator]: 1.658e-05
    [Cycle 2]: 0.00035339, [27]
        [switch_simplify]: 5.98002e-06
        [loop_unroll]: 4.99e-06
        [a_1]: 9.092e-05
        [with_stream_mark]: 1.543e-05
        [recompute_prepare]: 6.51e-06
        [updatestate_depend_eliminate]: 2.77002e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [parameter_eliminate]: 1.45001e-06
        [specialize_transform]: 4.92999e-06
        [updatestate_useless_node_eliminater]: 4.57998e-06
        [accelerated_algorithm]: 6.24001e-06
        [meta_shard_fg_expand]: 1.70001e-06
        [get_grad_eliminate_]: 4.77e-06
        [merge_forward]: 2.94001e-06
        [cell_reuse_recompute_pass]: 2.65002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.64e-05
        [j_node_and_user_rematch]: 8.55001e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 8.18001e-06
        [inline_without_move]: 5.21998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.79998e-06
        [auto_monad_grad]: 1.10999e-06
        [auto_monad_eliminator]: 8.07998e-06
        [cse]: 1.194e-05
        [replace_applicator]: 5.62001e-06
[py_interpret_to_execute_after_opt_a]: 1.461e-05
[rewriter_after_opt_a]: 3.66e-05
[convert_after_rewriter]: 7.59002e-06
[order_py_execute_after_rewriter]: 4.82998e-06
[mutable_eliminate]: 0.00071833
[jit_opt_b]: 5.423e-05, [1]
    [Cycle 1]: 4.584e-05, [2]
        [frontend_op_eliminate]: 1.712e-05
        [inline_after_opt_a]: 1.603e-05
[cconv]: 8.543e-05
[loop_unroll]: 0.00052141
[jit_opt_after_cconv]: 0.00016209, [1]
    [Cycle 1]: 0.0001543, [11]
        [c_1]: 2.099e-05
        [parameter_eliminate]: 5.12999e-06
        [updatestate_depend_eliminate]: 9.02999e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.47001e-06
        [cse]: 2.701e-05
        [call_graph_tuple_transform]: 2.199e-05
        [tuple_list_get_item_eliminator]: 5.05001e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 8.2e-07
        [switch_simplify]: 5.04998e-06
[remove_dup_value]: 1.561e-05
[partial_unused_args_eliminate]: 2.06e-06
[environ_conv]: 5.66998e-06
[add_recomputation]: 5.452e-05
[cse_after_recomputation]: 2.395e-05, [1]
    [Cycle 1]: 1.721e-05, [1]
        [cse]: 1.111e-05
[auto_monad_reorder]: 1.627e-05
[get_jit_bprop_graph]: 2.62001e-06
[rewriter_after_jit_bprop_graph]: 8.38001e-06
[opt_after_jit_grad]: 0.00057974
[symbol_engine_optimizer]: 8.338e-05, [1]
    [Cycle 1]: 7.518e-05, [6]
        [build]: 5.12e-06
        [elim_shapecalc]: 9.00999e-06
        [elim_not_effective]: 1.56e-05
        [opt_reshape]: 5.77001e-06
        [fold_const_symbol]: 9.39998e-06
        [renormalize]: 9.29984e-07
[validate]: 4.782e-05
[backend_pass]: 1.64e-06
[task_emit]: 0.00689192
[execute]: 1.15e-05
Sums
    bootstrap                            :     0.000371s :  2.14%
    type_inference                       :     0.005597s : 32.35%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000069s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.25%
    rewriter_before_opt_a                :     0.000078s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000675s :  3.90%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000793s :  4.58%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.15%
    jit_opt_a.cse                        :     0.000042s :  0.24%
    jit_opt_a.replace_applicator         :     0.000022s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.08%
    rewriter_after_opt_a                 :     0.000037s :  0.21%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000718s :  4.15%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000085s :  0.49%
    loop_unroll                          :     0.000521s :  3.01%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000055s :  0.32%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000580s :  3.35%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000048s :  0.28%
    backend_pass                         :     0.000002s :  0.01%
    task_emit                            :     0.006892s : 39.84%
    execute                              :     0.000012s :  0.07%


TotalTime = 0.012038, [30]
[bootstrap]: 0.00027598
[type_inference]: 0.00421118
[event_method]: 1.728e-05
[auto_monad]: 6.163e-05
[graph_reusing]: 4.93001e-06
[pre_auto_parallel]: 2.26e-06
[py_interpret_to_execute]: 3.354e-05
[rewriter_before_opt_a]: 6.912e-05
[expand_dump_flag]: 2.92002e-06
[jit_opt_a]: 0.00527545, [2]
    [Cycle 1]: 0.00158709, [27]
        [switch_simplify]: 5.344e-05
        [loop_unroll]: 2.85e-05
        [a_1]: 0.00052221
        [with_stream_mark]: 1.651e-05
        [recompute_prepare]: 7.20998e-06
        [updatestate_depend_eliminate]: 3.40003e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 2.00002e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 1.83002e-06
        [get_grad_eliminate_]: 5.35999e-06
        [merge_forward]: 3.63999e-06
        [cell_reuse_recompute_pass]: 1.16002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.308e-05
        [j_node_and_user_rematch]: 8.62e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 8.35001e-06
        [inline_without_move]: 5.74999e-06
        [renormalize]: 0.00065725
        [add_forward_monad_depend]: 4.92e-06
        [auto_monad_grad]: 2.65002e-06
        [auto_monad_eliminator]: 1.335e-05
        [cse]: 2.503e-05
        [replace_applicator]: 1.134e-05
    [Cycle 2]: 0.00031979, [27]
        [switch_simplify]: 5.49998e-06
        [loop_unroll]: 4.95001e-06
        [a_1]: 8.912e-05
        [with_stream_mark]: 9.00999e-06
        [recompute_prepare]: 4.75001e-06
        [updatestate_depend_eliminate]: 2.51e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 1.02998e-06
        [specialize_transform]: 4.70001e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.37999e-06
        [meta_shard_fg_expand]: 1.24998e-06
        [get_grad_eliminate_]: 4.82998e-06
        [merge_forward]: 2.48998e-06
        [cell_reuse_recompute_pass]: 1.89e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.265e-05
        [j_node_and_user_rematch]: 7.35e-06
        [meta_fg_expand]: 1.64998e-06
        [replace_old_param]: 7.6e-06
        [inline_without_move]: 4.74e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.05001e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 5.86e-06
        [cse]: 1.026e-05
        [replace_applicator]: 5.07e-06
[py_interpret_to_execute_after_opt_a]: 9.51003e-06
[rewriter_after_opt_a]: 3.115e-05
[convert_after_rewriter]: 6.73e-06
[order_py_execute_after_rewriter]: 4.82e-06
[mutable_eliminate]: 0.0005457
[jit_opt_b]: 4.958e-05, [1]
    [Cycle 1]: 4.265e-05, [2]
        [frontend_op_eliminate]: 1.652e-05
        [inline_after_opt_a]: 1.469e-05
[cconv]: 2.648e-05
[loop_unroll]: 0.00041728
[jit_opt_after_cconv]: 0.00015217, [1]
    [Cycle 1]: 0.00014487, [11]
        [c_1]: 2.043e-05
        [parameter_eliminate]: 2.71999e-06
        [updatestate_depend_eliminate]: 5.15999e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 2.876e-05
        [call_graph_tuple_transform]: 1.981e-05
        [tuple_list_get_item_eliminator]: 5.27001e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 5.45001e-06
[remove_dup_value]: 1.39e-05
[partial_unused_args_eliminate]: 2.43e-06
[environ_conv]: 5.15999e-06
[add_recomputation]: 4.945e-05
[cse_after_recomputation]: 2.014e-05, [1]
    [Cycle 1]: 1.455e-05, [1]
        [cse]: 9.17999e-06
[auto_monad_reorder]: 1.422e-05
[get_jit_bprop_graph]: 1.64e-06
[rewriter_after_jit_bprop_graph]: 5.12999e-06
[opt_after_jit_grad]: 0.00045761
[symbol_engine_optimizer]: 7.337e-05, [1]
    [Cycle 1]: 6.686e-05, [6]
        [build]: 4.13001e-06
        [elim_shapecalc]: 7.95e-06
        [elim_not_effective]: 1.263e-05
        [opt_reshape]: 6.00002e-06
        [fold_const_symbol]: 8.17e-06
        [renormalize]: 5.3001e-07
[validate]: 3.57e-05
Sums
    bootstrap                            :     0.000276s :  3.41%
    type_inference                       :     0.004211s : 51.97%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000062s :  0.76%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.41%
    rewriter_before_opt_a                :     0.000069s :  0.85%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000059s :  0.73%
    jit_opt_a.loop_unroll                :     0.000033s :  0.41%
    jit_opt_a.a_1                        :     0.000611s :  7.55%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000657s :  8.11%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.24%
    jit_opt_a.cse                        :     0.000035s :  0.44%
    jit_opt_a.replace_applicator         :     0.000016s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000031s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000546s :  6.74%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000026s :  0.33%
    loop_unroll                          :     0.000417s :  5.15%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000029s :  0.35%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000049s :  0.61%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000014s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000458s :  5.65%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000036s :  0.44%


Time group info:
------[substitution.]   0.000247    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.25% :     0.000006s :      3: substitution.graph_param_transform
84.61% :     0.000209s :      4: substitution.inline
 1.81% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.35% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.32% :     0.000006s :      2: substitution.replace_old_param
 5.28% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005514     2
82.18% :     0.004532s :      1: type_inference.infer
17.82% :     0.000983s :      1: type_inference.specialize
------[replace.]   0.000073     6
67.22% :     0.000049s :      4: replace.inline
32.78% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000218     6
94.53% :     0.000206s :      4: match.inline
 5.47% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.27% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.14% :     0.000003s :     10: predicate.arithmetic_simplify
 1.65% :     0.000002s :     10: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.17% :     0.000001s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.02% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.17% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.83% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 6.27% :     0.000008s :     22: predicate.inline
 1.37% :     0.000002s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.85% :     0.000001s :      6: predicate.less_batch_normalization
 1.73% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 2.05% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.02% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.77% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 2.61% :     0.000003s :      3: predicate.mutable_eliminate
 0.84% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     15: predicate.partial_eliminate
 1.03% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.39% :     0.000002s :     10: predicate.reduce_eliminate
 1.30% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.57% :     0.000001s :      3: predicate.reset_defer_inline
 1.15% :     0.000001s :     10: predicate.reshape_eliminate
 1.10% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.09% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 0.85% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.30% :     0.000003s :     16: predicate.switch_defer_inline
 1.90% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.86% :     0.000010s :     53: predicate.switch_simplify
 1.08% :     0.000001s :     10: predicate.tile_eliminate
 1.06% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.00% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.84% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.59% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.37% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000743    12
53.91% :     0.000401s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.06% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.03% :     0.000290s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023794    76
 0.25% :     0.000058s :      1: add_recomputation
 0.31% :     0.000073s :      1: auto_monad
 0.09% :     0.000021s :      1: auto_monad_reorder
 0.03% :     0.000008s :      1: backend_pass
 1.65% :     0.000393s :      1: bootstrap
 0.38% :     0.000090s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000026s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000025s :      1: event_method
 0.07% :     0.000017s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.72% :     0.006120s :      1: jit_opt_a
 0.70% :     0.000166s :      1: jit_opt_after_cconv
 0.24% :     0.000057s :      1: jit_opt_b
 2.25% :     0.000534s :      1: loop_unroll
 3.08% :     0.000733s :      1: mutable_eliminate
 3.80% :     0.000904s :     26: opt.transform.jit_opt_a
 0.21% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.07% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.50% :     0.000595s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000045s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 1.84% :     0.000438s :      1: renormalize.infer
 1.45% :     0.000345s :      1: renormalize.specialize
 0.05% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000040s :      1: rewriter_after_opt_a
 0.34% :     0.000082s :      1: rewriter_before_opt_a
 0.36% :     0.000086s :      1: symbol_engine_optimizer
29.06% :     0.006915s :      1: task_emit
23.60% :     0.005616s :      1: type_inference
 0.54% :     0.000129s :      1: validate


Time group info:
------[substitution.]   0.000198    23
 0.95% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.93% :     0.000006s :      3: substitution.graph_param_transform
84.20% :     0.000167s :      4: substitution.inline
 1.43% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.20% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.91% :     0.000004s :      2: substitution.replace_old_param
 5.74% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004154     2
79.82% :     0.003316s :      1: type_inference.infer
20.18% :     0.000839s :      1: type_inference.specialize
------[replace.]   0.000065     6
67.51% :     0.000044s :      4: replace.inline
32.49% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000175     6
94.08% :     0.000164s :      4: match.inline
 5.92% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.23% :     0.000001s :     10: predicate.addn_zero_filter
 2.03% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.33% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.24% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.39% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.22% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.15% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.48% :     0.000006s :     22: predicate.inline
 1.10% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.79% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.95% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.60% :     0.000002s :      3: predicate.mutable_eliminate
 0.75% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.29% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.59% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.67% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.70% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.81% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.09% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000594    12
48.78% :     0.000290s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.16% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.07% :     0.000268s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013563    72
 0.38% :     0.000052s :      1: add_recomputation
 0.48% :     0.000065s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.13% :     0.000289s :      1: bootstrap
 0.21% :     0.000029s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.15% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
38.91% :     0.005278s :      1: jit_opt_a
 1.14% :     0.000155s :      1: jit_opt_after_cconv
 0.38% :     0.000052s :      1: jit_opt_b
 3.13% :     0.000425s :      1: loop_unroll
 4.09% :     0.000555s :      1: mutable_eliminate
 6.01% :     0.000816s :     26: opt.transform.jit_opt_a
 0.35% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.43% :     0.000465s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000036s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.59% :     0.000352s :      1: renormalize.infer
 2.20% :     0.000298s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000034s :      1: rewriter_after_opt_a
 0.53% :     0.000072s :      1: rewriter_before_opt_a
 0.56% :     0.000076s :      1: symbol_engine_optimizer
31.13% :     0.004223s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.139.335 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dcaf8b0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.139.922 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0127401, [30]
[bootstrap]: 0.00028637
[type_inference]: 0.00466238
[event_method]: 1.878e-05
[auto_monad]: 6.322e-05
[graph_reusing]: 5.94e-06
[pre_auto_parallel]: 2.64001e-06
[py_interpret_to_execute]: 4.079e-05
[rewriter_before_opt_a]: 6.755e-05
[expand_dump_flag]: 3.03e-06
[jit_opt_a]: 0.005578, [2]
    [Cycle 1]: 0.00154269, [27]
        [switch_simplify]: 5.073e-05
        [loop_unroll]: 3.004e-05
        [a_1]: 0.00052756
        [with_stream_mark]: 1.429e-05
        [recompute_prepare]: 6.98998e-06
        [updatestate_depend_eliminate]: 2.96999e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 1.09003e-06
        [specialize_transform]: 6.31998e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.69e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 5.33002e-06
        [merge_forward]: 2.69999e-06
        [cell_reuse_recompute_pass]: 8.60018e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.187e-05
        [j_node_and_user_rematch]: 8.78001e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 8.03999e-06
        [inline_without_move]: 5.74e-06
        [renormalize]: 0.00063235
        [add_forward_monad_depend]: 5.07999e-06
        [auto_monad_grad]: 1.47999e-06
        [auto_monad_eliminator]: 1.063e-05
        [cse]: 1.646e-05
        [replace_applicator]: 1.182e-05
    [Cycle 2]: 0.00031576, [27]
        [switch_simplify]: 5.59e-06
        [loop_unroll]: 5.20999e-06
        [a_1]: 8.933e-05
        [with_stream_mark]: 8.89998e-06
        [recompute_prepare]: 4.80001e-06
        [updatestate_depend_eliminate]: 2.52001e-06
        [updatestate_assign_eliminate]: 2.06998e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 9.19972e-07
        [specialize_transform]: 5.02e-06
        [updatestate_useless_node_eliminater]: 4.97999e-06
        [accelerated_algorithm]: 5.02e-06
        [meta_shard_fg_expand]: 1.06002e-06
        [get_grad_eliminate_]: 4.53999e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.44998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.173e-05
        [j_node_and_user_rematch]: 7.58001e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 6.94999e-06
        [inline_without_move]: 4.68001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 5.02e-06
        [cse]: 1.037e-05
        [replace_applicator]: 4.83001e-06
[py_interpret_to_execute_after_opt_a]: 8.79998e-06
[rewriter_after_opt_a]: 2.621e-05
[convert_after_rewriter]: 8.22998e-06
[order_py_execute_after_rewriter]: 4.21001e-06
[mutable_eliminate]: 0.00052517
[jit_opt_b]: 4.929e-05, [1]
    [Cycle 1]: 4.259e-05, [2]
        [frontend_op_eliminate]: 1.621e-05
        [inline_after_opt_a]: 1.511e-05
[cconv]: 1.669e-05
[loop_unroll]: 0.00042069
[jit_opt_after_cconv]: 0.00013402, [1]
    [Cycle 1]: 0.00012759, [11]
        [c_1]: 2.003e-05
        [parameter_eliminate]: 2.49001e-06
        [updatestate_depend_eliminate]: 5.14e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [cse]: 1.736e-05
        [call_graph_tuple_transform]: 1.635e-05
        [tuple_list_get_item_eliminator]: 5.37999e-06
        [none_parameter_eliminate]: 9.20001e-07
        [renormalize]: 4.10015e-07
        [switch_simplify]: 5.20001e-06
[remove_dup_value]: 7.1e-06
[partial_unused_args_eliminate]: 1.44e-06
[environ_conv]: 3.78001e-06
[add_recomputation]: 3.557e-05
[cse_after_recomputation]: 2.016e-05, [1]
    [Cycle 1]: 1.482e-05, [1]
        [cse]: 9.25001e-06
[auto_monad_reorder]: 9.31e-06
[get_jit_bprop_graph]: 1.49e-06
[rewriter_after_jit_bprop_graph]: 5.72999e-06
[opt_after_jit_grad]: 0.00046708
[symbol_engine_optimizer]: 7.27e-05, [1]
    [Cycle 1]: 6.541e-05, [6]
        [build]: 2.97002e-06
        [elim_shapecalc]: 7.18e-06
        [elim_not_effective]: 1.251e-05
        [opt_reshape]: 5.96e-06
        [fold_const_symbol]: 8.58001e-06
        [renormalize]: 4.50003e-07
[validate]: 2.933e-05
Sums
    bootstrap                            :     0.000286s :  3.39%
    type_inference                       :     0.004662s : 55.13%
    event_method                         :     0.000019s :  0.22%
    auto_monad                           :     0.000063s :  0.75%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000041s :  0.48%
    rewriter_before_opt_a                :     0.000068s :  0.80%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000056s :  0.67%
    jit_opt_a.loop_unroll                :     0.000035s :  0.42%
    jit_opt_a.a_1                        :     0.000617s :  7.29%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000005s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.18%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000632s :  7.48%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.19%
    jit_opt_a.cse                        :     0.000027s :  0.32%
    jit_opt_a.replace_applicator         :     0.000017s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.10%
    rewriter_after_opt_a                 :     0.000026s :  0.31%
    convert_after_rewriter               :     0.000008s :  0.10%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000525s :  6.21%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000017s :  0.20%
    loop_unroll                          :     0.000421s :  4.97%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000007s :  0.08%
    partial_unused_args_eliminate        :     0.000001s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000036s :  0.42%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000009s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000467s :  5.52%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000029s :  0.35%


Time group info:
------[substitution.]   0.000185    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 1.84% :     0.000003s :      3: substitution.graph_param_transform
86.82% :     0.000161s :      4: substitution.inline
 1.36% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.19% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.63% :     0.000003s :      2: substitution.replace_old_param
 4.87% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004551     2
79.55% :     0.003620s :      1: type_inference.infer
20.45% :     0.000931s :      1: type_inference.specialize
------[replace.]   0.000067     6
65.35% :     0.000044s :      4: replace.inline
34.65% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000166     6
95.23% :     0.000158s :      4: match.inline
 4.77% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.58% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.55% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.24% :     0.000003s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.32% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.38% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.24% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 4.85% :     0.000006s :     22: predicate.inline
 1.36% :     0.000002s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.49% :     0.000002s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.90% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000002s :     15: predicate.partial_eliminate
 1.30% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.60% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     18: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 2.23% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.60% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.47% :     0.000002s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.55% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000630    12
47.49% :     0.000299s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.05% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.46% :     0.000293s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014255    72
 0.27% :     0.000038s :      1: add_recomputation
 0.47% :     0.000067s :      1: auto_monad
 0.08% :     0.000012s :      1: auto_monad_reorder
 2.13% :     0.000304s :      1: bootstrap
 0.14% :     0.000019s :      1: cconv
 0.07% :     0.000011s :      1: convert_after_rewriter
 0.16% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.16% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
39.15% :     0.005581s :      1: jit_opt_a
 0.96% :     0.000137s :      1: jit_opt_after_cconv
 0.36% :     0.000052s :      1: jit_opt_b
 3.01% :     0.000429s :      1: loop_unroll
 3.74% :     0.000533s :      1: mutable_eliminate
 5.75% :     0.000819s :     26: opt.transform.jit_opt_a
 0.31% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.33% :     0.000475s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000043s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000010s :      1: remove_dup_value
 2.30% :     0.000327s :      1: renormalize.infer
 2.10% :     0.000299s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000029s :      1: rewriter_after_opt_a
 0.49% :     0.000070s :      1: rewriter_before_opt_a
 0.53% :     0.000075s :      1: symbol_engine_optimizer
32.81% :     0.004677s :      1: type_inference


TotalTime = 0.0318293, [33]
[bootstrap]: 0.00020271
[type_inference]: 0.00984494
[event_method]: 4.319e-05
[auto_monad]: 9.564e-05
[graph_reusing]: 7.08998e-06
[pre_auto_parallel]: 2.86999e-06
[py_interpret_to_execute]: 2.754e-05
[rewriter_before_opt_a]: 9.566e-05
[expand_dump_flag]: 3.26001e-06
[jit_opt_a]: 0.012114, [3]
    [Cycle 1]: 0.00606232, [27]
        [switch_simplify]: 7.383e-05
        [loop_unroll]: 4.08e-05
        [a_1]: 0.00091036
        [with_stream_mark]: 2.919e-05
        [recompute_prepare]: 2.107e-05
        [updatestate_depend_eliminate]: 8.43001e-06
        [updatestate_assign_eliminate]: 6.68e-06
        [updatestate_loads_eliminate]: 6.28002e-06
        [parameter_eliminate]: 2.89001e-06
        [specialize_transform]: 1.466e-05
        [updatestate_useless_node_eliminater]: 1.315e-05
        [accelerated_algorithm]: 1.351e-05
        [meta_shard_fg_expand]: 3.56999e-06
        [get_grad_eliminate_]: 1.244e-05
        [merge_forward]: 7.97998e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.866e-05
        [j_node_and_user_rematch]: 2.452e-05
        [meta_fg_expand]: 0.00163069
        [replace_old_param]: 6.139e-05
        [inline_without_move]: 5.594e-05
        [renormalize]: 0.00254557
        [add_forward_monad_depend]: 1.176e-05
        [auto_monad_grad]: 5.89e-06
        [auto_monad_eliminator]: 5.55e-05
        [cse]: 0.00015485
        [replace_applicator]: 8.194e-05
    [Cycle 2]: 0.00226973, [27]
        [switch_simplify]: 4.11e-05
        [loop_unroll]: 3.706e-05
        [a_1]: 0.00116341
        [with_stream_mark]: 1.986e-05
        [recompute_prepare]: 1.016e-05
        [updatestate_depend_eliminate]: 3.93001e-06
        [updatestate_assign_eliminate]: 4e-06
        [updatestate_loads_eliminate]: 3.21001e-06
        [parameter_eliminate]: 2.07999e-06
        [specialize_transform]: 7.53e-06
        [updatestate_useless_node_eliminater]: 6.79001e-06
        [accelerated_algorithm]: 2.182e-05
        [meta_shard_fg_expand]: 3.08e-06
        [get_grad_eliminate_]: 7.81001e-06
        [merge_forward]: 4.57e-06
        [cell_reuse_recompute_pass]: 1.48002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.778e-05
        [j_node_and_user_rematch]: 1.213e-05
        [meta_fg_expand]: 4.945e-05
        [replace_old_param]: 1.225e-05
        [inline_without_move]: 7.08e-06
        [renormalize]: 0.00060342
        [add_forward_monad_depend]: 5.32999e-06
        [auto_monad_grad]: 1.90001e-06
        [auto_monad_eliminator]: 1.384e-05
        [cse]: 2.711e-05
        [replace_applicator]: 1.33e-05
    [Cycle 3]: 0.00037637, [27]
        [switch_simplify]: 6.61999e-06
        [loop_unroll]: 6.36e-06
        [a_1]: 0.00012659
        [with_stream_mark]: 9.14e-06
        [recompute_prepare]: 6.16998e-06
        [updatestate_depend_eliminate]: 3.58e-06
        [updatestate_assign_eliminate]: 2.73998e-06
        [updatestate_loads_eliminate]: 2.46998e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 6.42001e-06
        [updatestate_useless_node_eliminater]: 5.89e-06
        [accelerated_algorithm]: 6.06e-06
        [meta_shard_fg_expand]: 1.70001e-06
        [get_grad_eliminate_]: 5.71e-06
        [merge_forward]: 2.89999e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.459e-05
        [j_node_and_user_rematch]: 1.03e-05
        [meta_fg_expand]: 2.25002e-06
        [replace_old_param]: 8.58001e-06
        [inline_without_move]: 6.12001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 7.05998e-06
        [cse]: 1.494e-05
        [replace_applicator]: 6.07999e-06
[py_interpret_to_execute_after_opt_a]: 1.214e-05
[rewriter_after_opt_a]: 4.121e-05
[convert_after_rewriter]: 7.45003e-06
[order_py_execute_after_rewriter]: 5.69e-06
[mutable_eliminate]: 0.00059485
[jit_opt_b]: 5.842e-05, [1]
    [Cycle 1]: 5.126e-05, [2]
        [frontend_op_eliminate]: 2.024e-05
        [inline_after_opt_a]: 1.937e-05
[cconv]: 2.334e-05
[loop_unroll]: 0.00042333
[jit_opt_after_cconv]: 0.00016669, [1]
    [Cycle 1]: 0.00016011, [11]
        [c_1]: 2.65e-05
        [parameter_eliminate]: 2.64999e-06
        [updatestate_depend_eliminate]: 6.49999e-06
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [cse]: 2.282e-05
        [call_graph_tuple_transform]: 2.206e-05
        [tuple_list_get_item_eliminator]: 6.73e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 7.21999e-06
[remove_dup_value]: 1.642e-05
[partial_unused_args_eliminate]: 2.64999e-06
[environ_conv]: 5.61e-06
[add_recomputation]: 5.33e-05
[cse_after_recomputation]: 2.689e-05, [1]
    [Cycle 1]: 2.118e-05, [1]
        [cse]: 1.537e-05
[auto_monad_reorder]: 1.953e-05
[get_jit_bprop_graph]: 1.99999e-06
[rewriter_after_jit_bprop_graph]: 4.87e-06
[opt_after_jit_grad]: 0.00047041
[symbol_engine_optimizer]: 8.352e-05, [1]
    [Cycle 1]: 7.737e-05, [6]
        [build]: 4.59998e-06
        [elim_shapecalc]: 9.56e-06
        [elim_not_effective]: 1.648e-05
        [opt_reshape]: 7.35e-06
        [fold_const_symbol]: 1.124e-05
        [renormalize]: 5.20027e-07
[validate]: 3.851e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00714054
[execute]: 4.87998e-06
Sums
    bootstrap                            :     0.000203s :  0.73%
    type_inference                       :     0.009845s : 35.68%
    event_method                         :     0.000043s :  0.16%
    auto_monad                           :     0.000096s :  0.35%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.10%
    rewriter_before_opt_a                :     0.000096s :  0.35%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000122s :  0.44%
    jit_opt_a.loop_unroll                :     0.000084s :  0.31%
    jit_opt_a.a_1                        :     0.002200s :  7.98%
    jit_opt_a.with_stream_mark           :     0.000058s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000037s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.02%
    jit_opt_a.specialize_transform       :     0.000029s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000041s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000026s :  0.09%
    jit_opt_a.merge_forward              :     0.000015s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000061s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000047s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001682s :  6.10%
    jit_opt_a.replace_old_param          :     0.000082s :  0.30%
    jit_opt_a.inline_without_move        :     0.000069s :  0.25%
    jit_opt_a.renormalize                :     0.003149s : 11.41%
    jit_opt_a.add_forward_monad_depend   :     0.000018s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000076s :  0.28%
    jit_opt_a.cse                        :     0.000197s :  0.71%
    jit_opt_a.replace_applicator         :     0.000101s :  0.37%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000041s :  0.15%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000595s :  2.16%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.07%
    cconv                                :     0.000023s :  0.08%
    loop_unroll                          :     0.000423s :  1.53%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000053s :  0.19%
    cse_after_recomputation.cse          :     0.000015s :  0.06%
    auto_monad_reorder                   :     0.000020s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000470s :  1.71%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007141s : 25.88%
    execute                              :     0.000005s :  0.02%


Time group info:
------[substitution.]   0.000545   110
 0.44% :     0.000002s :      3: substitution.elim_not_effective
 0.37% :     0.000002s :      3: substitution.fold_const_symbol
 1.02% :     0.000006s :      4: substitution.graph_param_transform
64.19% :     0.000350s :     12: substitution.inline
 3.40% :     0.000019s :      2: substitution.inline_without_move
 1.88% :     0.000010s :     14: substitution.j_node_and_user_rematch
 1.93% :     0.000010s :      7: substitution.minmaximum_grad
 1.26% :     0.000007s :      5: substitution.partial_eliminate
 2.19% :     0.000012s :     14: substitution.remove_not_recompute_node
 5.62% :     0.000031s :     10: substitution.replace_applicator
 1.51% :     0.000008s :      7: substitution.replace_old_param
 0.63% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.17% :     0.000023s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.98% :     0.000016s :      7: substitution.tuple_list_get_item_depend_reorder
 8.41% :     0.000046s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009776     2
89.01% :     0.008701s :      1: type_inference.infer
10.99% :     0.001074s :      1: type_inference.specialize
------[replace.]   0.000164    19
63.24% :     0.000104s :     12: replace.inline
36.76% :     0.000060s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000365    19
93.48% :     0.000342s :     12: match.inline
 6.52% :     0.000024s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000406  2568
 1.66% :     0.000007s :     43: predicate.accumulaten_eliminater
 0.43% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.43% :     0.000006s :     43: predicate.addn_check_dump
 1.62% :     0.000007s :     43: predicate.addn_zero_filter
 2.07% :     0.000008s :     43: predicate.arithmetic_simplify
 1.60% :     0.000006s :     43: predicate.cast_eliminate
 0.17% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.43% :     0.000006s :     43: predicate.compare_switch_simplify
 1.58% :     0.000006s :     43: predicate.depend_value_elim
 1.53% :     0.000006s :     43: predicate.dict_get_item_const_eliminator
 1.45% :     0.000006s :     43: predicate.dict_get_item_eliminator
 1.58% :     0.000006s :     43: predicate.dict_set_item_eliminator
 0.28% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      4: predicate.elim_not_effective
 0.19% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.46% :     0.000006s :     43: predicate.environ_add_const_eliminate
 1.47% :     0.000006s :     43: predicate.environ_get_add_eliminate
 1.45% :     0.000006s :     43: predicate.environ_get_depend_swap
 1.39% :     0.000006s :     43: predicate.environ_get_eliminate
 1.40% :     0.000006s :     43: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.83% :     0.000003s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.49% :     0.000018s :     70: predicate.inline
 1.84% :     0.000007s :     40: predicate.inline_without_move
 0.35% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.93% :     0.000004s :     19: predicate.less_batch_normalization
 1.68% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.92% :     0.000008s :     54: predicate.load_eliminater
 0.41% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.18% :     0.000013s :     85: predicate.loop_unroll_before_grad
 1.69% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.42% :     0.000006s :     43: predicate.merge_addn
 1.49% :     0.000006s :     43: predicate.minmaximum_grad
 0.54% :     0.000002s :      4: predicate.mutable_eliminate
 0.22% :     0.000001s :      4: predicate.opt_reshape
 2.25% :     0.000009s :     54: predicate.partial_eliminate
 1.82% :     0.000007s :     43: predicate.print_const_string_wrapper
 2.01% :     0.000008s :     43: predicate.reduce_eliminate
 1.81% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.59% :     0.000011s :     96: predicate.replace_applicator
 0.96% :     0.000004s :     40: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.67% :     0.000007s :     43: predicate.reshape_eliminate
 1.48% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.30% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.53% :     0.000006s :     43: predicate.same_eliminate
 0.52% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.33% :     0.000001s :      8: predicate.special_op_eliminate
 0.95% :     0.000004s :     19: predicate.specialize_transform
 1.69% :     0.000007s :     43: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000006s :     43: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.82% :     0.000011s :     62: predicate.switch_defer_inline
 2.50% :     0.000010s :     62: predicate.switch_layer_defer_inline
 6.30% :     0.000026s :    151: predicate.switch_simplify
 1.45% :     0.000006s :     43: predicate.tile_eliminate
 1.44% :     0.000006s :     43: predicate.transpose_eliminate
 1.85% :     0.000008s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.67% :     0.000007s :     43: predicate.tuple_list_get_item_depend_reorder
 3.46% :     0.000014s :     58: predicate.tuple_list_get_item_eliminator
 1.92% :     0.000008s :     43: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000007s :     50: predicate.tuple_to_list_eliminator_
 2.09% :     0.000008s :     54: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000012s :     73: predicate.updatestate_useless_node_eliminater
 1.95% :     0.000008s :     43: predicate.value_based_eliminate
 0.14% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.20% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001495    29
58.56% :     0.000876s :     13: func_graph_cloner_run.FuncGraphClonerGraph
41.44% :     0.000620s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.037990    91
 0.15% :     0.000056s :      1: add_recomputation
 0.26% :     0.000100s :      1: auto_monad
 0.06% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.57% :     0.000216s :      1: bootstrap
 0.07% :     0.000026s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000029s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.13% :     0.000048s :      1: event_method
 0.02% :     0.000008s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
31.89% :     0.012117s :      1: jit_opt_a
 0.45% :     0.000170s :      1: jit_opt_after_cconv
 0.16% :     0.000061s :      1: jit_opt_b
 1.14% :     0.000432s :      1: loop_unroll
 1.59% :     0.000603s :      1: mutable_eliminate
 7.57% :     0.002877s :     39: opt.transform.jit_opt_a
 0.16% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000033s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000041s :      4: opt.transform.symbol_engine_opt
 1.26% :     0.000479s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000030s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 4.36% :     0.001657s :      2: renormalize.infer
 3.88% :     0.001474s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000044s :      1: rewriter_after_opt_a
 0.26% :     0.000099s :      1: rewriter_before_opt_a
 0.23% :     0.000086s :      1: symbol_engine_optimizer
18.82% :     0.007151s :      1: task_emit
25.95% :     0.009858s :      1: type_inference
 0.16% :     0.000062s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.179.842 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dcafa60> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.180.502 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0124982, [30]
[bootstrap]: 0.00028858
[type_inference]: 0.00462213
[event_method]: 1.832e-05
[auto_monad]: 6.584e-05
[graph_reusing]: 5.89e-06
[pre_auto_parallel]: 2.34001e-06
[py_interpret_to_execute]: 3.507e-05
[rewriter_before_opt_a]: 7.065e-05
[expand_dump_flag]: 3.65998e-06
[jit_opt_a]: 0.00531923, [2]
    [Cycle 1]: 0.00153905, [27]
        [switch_simplify]: 5.601e-05
        [loop_unroll]: 2.879e-05
        [a_1]: 0.00051517
        [with_stream_mark]: 1.764e-05
        [recompute_prepare]: 7.07997e-06
        [updatestate_depend_eliminate]: 3.71001e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.83e-06
        [parameter_eliminate]: 1.95001e-06
        [specialize_transform]: 5.96003e-06
        [updatestate_useless_node_eliminater]: 5.02999e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 1.87001e-06
        [get_grad_eliminate_]: 5.22999e-06
        [merge_forward]: 3.73001e-06
        [cell_reuse_recompute_pass]: 1.52999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.495e-05
        [j_node_and_user_rematch]: 1.013e-05
        [meta_fg_expand]: 2.26998e-06
        [replace_old_param]: 8.45001e-06
        [inline_without_move]: 5.42001e-06
        [renormalize]: 0.00060794
        [add_forward_monad_depend]: 4.97999e-06
        [auto_monad_grad]: 2.26e-06
        [auto_monad_eliminator]: 1.261e-05
        [cse]: 2.634e-05
        [replace_applicator]: 1.15e-05
    [Cycle 2]: 0.00031679, [27]
        [switch_simplify]: 5.35999e-06
        [loop_unroll]: 5.06997e-06
        [a_1]: 9.037e-05
        [with_stream_mark]: 8.69e-06
        [recompute_prepare]: 4.87998e-06
        [updatestate_depend_eliminate]: 2.36998e-06
        [updatestate_assign_eliminate]: 1.92001e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.07998e-06
        [specialize_transform]: 4.89003e-06
        [updatestate_useless_node_eliminater]: 4.73001e-06
        [accelerated_algorithm]: 5.67999e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 4.43001e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.161e-05
        [j_node_and_user_rematch]: 7.92998e-06
        [meta_fg_expand]: 1.51998e-06
        [replace_old_param]: 7.4e-06
        [inline_without_move]: 4.76002e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 7.80012e-07
        [auto_monad_eliminator]: 4.89e-06
        [cse]: 9.64e-06
        [replace_applicator]: 4.92999e-06
[py_interpret_to_execute_after_opt_a]: 1.031e-05
[rewriter_after_opt_a]: 3.261e-05
[convert_after_rewriter]: 6.36e-06
[order_py_execute_after_rewriter]: 5.24998e-06
[mutable_eliminate]: 0.00053159
[jit_opt_b]: 5.051e-05, [1]
    [Cycle 1]: 4.39e-05, [2]
        [frontend_op_eliminate]: 1.644e-05
        [inline_after_opt_a]: 1.491e-05
[cconv]: 2.405e-05
[loop_unroll]: 0.00041615
[jit_opt_after_cconv]: 0.00013353, [1]
    [Cycle 1]: 0.00012711, [11]
        [c_1]: 2.03e-05
        [parameter_eliminate]: 2.22999e-06
        [updatestate_depend_eliminate]: 5.15999e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.573e-05
        [call_graph_tuple_transform]: 1.78e-05
        [tuple_list_get_item_eliminator]: 5.20999e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.42001e-06
[remove_dup_value]: 1.447e-05
[partial_unused_args_eliminate]: 2.05002e-06
[environ_conv]: 5.42001e-06
[add_recomputation]: 4.718e-05
[cse_after_recomputation]: 1.979e-05, [1]
    [Cycle 1]: 1.445e-05, [1]
        [cse]: 9.10999e-06
[auto_monad_reorder]: 1.4e-05
[get_jit_bprop_graph]: 1.40001e-06
[rewriter_after_jit_bprop_graph]: 4.84003e-06
[opt_after_jit_grad]: 0.00044581
[symbol_engine_optimizer]: 9.13e-05, [1]
    [Cycle 1]: 8.513e-05, [6]
        [build]: 3.60998e-06
        [elim_shapecalc]: 8.27e-06
        [elim_not_effective]: 1.26e-05
        [opt_reshape]: 5.52999e-06
        [fold_const_symbol]: 9.10999e-06
        [renormalize]: 4.2998e-07
[validate]: 3.426e-05
Sums
    bootstrap                            :     0.000289s :  3.42%
    type_inference                       :     0.004622s : 54.77%
    event_method                         :     0.000018s :  0.22%
    auto_monad                           :     0.000066s :  0.78%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.42%
    rewriter_before_opt_a                :     0.000071s :  0.84%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000061s :  0.73%
    jit_opt_a.loop_unroll                :     0.000034s :  0.40%
    jit_opt_a.a_1                        :     0.000606s :  7.18%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000608s :  7.20%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.21%
    jit_opt_a.cse                        :     0.000036s :  0.43%
    jit_opt_a.replace_applicator         :     0.000016s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000033s :  0.39%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000532s :  6.30%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000024s :  0.28%
    loop_unroll                          :     0.000416s :  4.93%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000047s :  0.56%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000014s :  0.17%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000446s :  5.28%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000034s :  0.41%


Time group info:
------[substitution.]   0.000190    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.55% :     0.000005s :      3: substitution.graph_param_transform
83.16% :     0.000158s :      4: substitution.inline
 1.81% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.34% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.07% :     0.000004s :      2: substitution.replace_old_param
 6.39% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004551     2
78.97% :     0.003594s :      1: type_inference.infer
21.03% :     0.000957s :      1: type_inference.specialize
------[replace.]   0.000065     6
69.05% :     0.000045s :      4: replace.inline
30.95% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000166     6
93.40% :     0.000155s :      4: match.inline
 6.60% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.15% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 1.94% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.36% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 5.38% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.72% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000000s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.33% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.85% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.91% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.83% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.91% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000672    12
52.68% :     0.000354s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.20% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.12% :     0.000290s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013971    72
 0.36% :     0.000050s :      1: add_recomputation
 0.50% :     0.000069s :      1: auto_monad
 0.12% :     0.000017s :      1: auto_monad_reorder
 2.16% :     0.000302s :      1: bootstrap
 0.19% :     0.000027s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.16% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000023s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.10% :     0.005323s :      1: jit_opt_a
 0.98% :     0.000136s :      1: jit_opt_after_cconv
 0.38% :     0.000053s :      1: jit_opt_b
 3.03% :     0.000424s :      1: loop_unroll
 3.91% :     0.000547s :      1: mutable_eliminate
 5.82% :     0.000813s :     26: opt.transform.jit_opt_a
 0.33% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.25% :     0.000454s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000038s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.16% :     0.000302s :      1: renormalize.infer
 2.14% :     0.000299s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000035s :      1: rewriter_after_opt_a
 0.52% :     0.000073s :      1: rewriter_before_opt_a
 0.68% :     0.000094s :      1: symbol_engine_optimizer
33.21% :     0.004639s :      1: type_inference


TotalTime = 0.0223386, [33]
[bootstrap]: 0.00032973
[type_inference]: 0.00471357
[event_method]: 1.718e-05
[auto_monad]: 6.353e-05
[graph_reusing]: 5.64e-06
[pre_auto_parallel]: 2.76e-06
[py_interpret_to_execute]: 3.687e-05
[rewriter_before_opt_a]: 7.535e-05
[expand_dump_flag]: 3.23e-06
[jit_opt_a]: 0.00822673, [2]
    [Cycle 1]: 0.0016952, [27]
        [switch_simplify]: 5.886e-05
        [loop_unroll]: 2.847e-05
        [a_1]: 0.00055431
        [with_stream_mark]: 2.245e-05
        [recompute_prepare]: 7.35e-06
        [updatestate_depend_eliminate]: 3.86999e-06
        [updatestate_assign_eliminate]: 3.11001e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [parameter_eliminate]: 1.84998e-06
        [specialize_transform]: 5.47001e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.10999e-06
        [meta_shard_fg_expand]: 2.43e-06
        [get_grad_eliminate_]: 4.82998e-06
        [merge_forward]: 3.81001e-06
        [cell_reuse_recompute_pass]: 1.00001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.454e-05
        [j_node_and_user_rematch]: 8.79e-06
        [meta_fg_expand]: 2.54999e-06
        [replace_old_param]: 8.18001e-06
        [inline_without_move]: 4.96002e-06
        [renormalize]: 0.00070555
        [add_forward_monad_depend]: 6.01e-06
        [auto_monad_grad]: 2.36e-06
        [auto_monad_eliminator]: 1.283e-05
        [cse]: 2.866e-05
        [replace_applicator]: 1.144e-05
    [Cycle 2]: 0.00286953, [27]
        [switch_simplify]: 5.32001e-06
        [loop_unroll]: 4.63001e-06
        [a_1]: 8.616e-05
        [with_stream_mark]: 9.36e-06
        [recompute_prepare]: 4.78001e-06
        [updatestate_depend_eliminate]: 2.44001e-06
        [updatestate_assign_eliminate]: 1.96998e-06
        [updatestate_loads_eliminate]: 1.79998e-06
        [parameter_eliminate]: 8.80013e-07
        [specialize_transform]: 5.00999e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 4.65001e-06
        [meta_shard_fg_expand]: 1.13001e-06
        [get_grad_eliminate_]: 4.3e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.217e-05
        [j_node_and_user_rematch]: 7.55e-06
        [meta_fg_expand]: 1.41002e-06
        [replace_old_param]: 7.83001e-06
        [inline_without_move]: 4.55001e-06
        [renormalize]: 3.09985e-07
        [add_forward_monad_depend]: 6.01e-06
        [auto_monad_grad]: 2.12001e-06
        [auto_monad_eliminator]: 1.531e-05
        [cse]: 2.106e-05
        [replace_applicator]: 1.465e-05
[py_interpret_to_execute_after_opt_a]: 1.484e-05
[rewriter_after_opt_a]: 4.023e-05
[convert_after_rewriter]: 6.59001e-06
[order_py_execute_after_rewriter]: 5.10001e-06
[mutable_eliminate]: 0.00074969
[jit_opt_b]: 5.218e-05, [1]
    [Cycle 1]: 4.383e-05, [2]
        [frontend_op_eliminate]: 1.579e-05
        [inline_after_opt_a]: 1.572e-05
[cconv]: 2.973e-05
[loop_unroll]: 0.00042406
[jit_opt_after_cconv]: 0.00021246, [1]
    [Cycle 1]: 0.00020566, [11]
        [c_1]: 8.338e-05
        [parameter_eliminate]: 3.36999e-06
        [updatestate_depend_eliminate]: 6.91999e-06
        [updatestate_assign_eliminate]: 2.63e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 2.082e-05
        [call_graph_tuple_transform]: 1.943e-05
        [tuple_list_get_item_eliminator]: 5.72001e-06
        [none_parameter_eliminate]: 1.46998e-06
        [renormalize]: 1.22e-06
        [switch_simplify]: 5.54e-06
[remove_dup_value]: 1.495e-05
[partial_unused_args_eliminate]: 2.16998e-06
[environ_conv]: 5.37001e-06
[add_recomputation]: 5.431e-05
[cse_after_recomputation]: 2.046e-05, [1]
    [Cycle 1]: 1.494e-05, [1]
        [cse]: 8.92e-06
[auto_monad_reorder]: 1.484e-05
[get_jit_bprop_graph]: 2.43998e-06
[rewriter_after_jit_bprop_graph]: 7.21001e-06
[opt_after_jit_grad]: 0.00049707
[symbol_engine_optimizer]: 7.827e-05, [1]
    [Cycle 1]: 7.13e-05, [6]
        [build]: 4.24997e-06
        [elim_shapecalc]: 7.78001e-06
        [elim_not_effective]: 1.352e-05
        [opt_reshape]: 5.96e-06
        [fold_const_symbol]: 9.82001e-06
        [renormalize]: 9.89996e-07
[validate]: 4.338e-05
[backend_pass]: 1.02998e-06
[task_emit]: 0.00635091
[execute]: 9.81998e-06
Sums
    bootstrap                            :     0.000330s :  2.13%
    type_inference                       :     0.004714s : 30.40%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000064s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.24%
    rewriter_before_opt_a                :     0.000075s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.41%
    jit_opt_a.loop_unroll                :     0.000033s :  0.21%
    jit_opt_a.a_1                        :     0.000640s :  4.13%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000706s :  4.55%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.18%
    jit_opt_a.cse                        :     0.000050s :  0.32%
    jit_opt_a.replace_applicator         :     0.000026s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.10%
    rewriter_after_opt_a                 :     0.000040s :  0.26%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000750s :  4.83%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000030s :  0.19%
    loop_unroll                          :     0.000424s :  2.73%
    jit_opt_after_cconv.c_1              :     0.000083s :  0.54%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000015s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000054s :  0.35%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.05%
    opt_after_jit_grad                   :     0.000497s :  3.21%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.28%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006351s : 40.95%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000233    23
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.52% :     0.000006s :      3: substitution.graph_param_transform
85.78% :     0.000200s :      4: substitution.inline
 1.26% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.86% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000005s :      2: substitution.replace_old_param
 5.10% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004636     2
82.73% :     0.003836s :      1: type_inference.infer
17.27% :     0.000801s :      1: type_inference.specialize
------[replace.]   0.000068     6
67.73% :     0.000046s :      4: replace.inline
32.27% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000208     6
94.87% :     0.000197s :      4: match.inline
 5.13% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000173   691
 0.86% :     0.000001s :     10: predicate.accumulaten_eliminater
 0.79% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.73% :     0.000001s :     10: predicate.addn_check_dump
 0.85% :     0.000001s :     10: predicate.addn_zero_filter
 1.16% :     0.000002s :     10: predicate.arithmetic_simplify
 0.82% :     0.000001s :     10: predicate.cast_eliminate
 0.26% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.72% :     0.000001s :     10: predicate.compare_switch_simplify
 0.79% :     0.000001s :     10: predicate.depend_value_elim
 0.75% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 0.83% :     0.000001s :     10: predicate.dict_get_item_eliminator
 0.78% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.60% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.23% :     0.000000s :      3: predicate.elim_not_effective
 0.34% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.77% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.73% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.68% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.75% :     0.000001s :     10: predicate.environ_get_eliminate
 0.71% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.16% :     0.000000s :      3: predicate.fold_const_symbol
 0.58% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 3.84% :     0.000007s :     22: predicate.inline
 0.56% :     0.000001s :      6: predicate.inline_without_move
 0.25% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.63% :     0.000001s :      6: predicate.less_batch_normalization
 1.10% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.29% :     0.000002s :     15: predicate.load_eliminater
 0.87% :     0.000002s :      3: predicate.loop_unroll_after_grad
 2.54% :     0.000004s :     34: predicate.loop_unroll_before_grad
36.76% :     0.000064s :     13: predicate.make_slice_get_slice_eliminator
 0.72% :     0.000001s :     10: predicate.merge_addn
 0.69% :     0.000001s :     10: predicate.minmaximum_grad
 1.03% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 1.45% :     0.000003s :     15: predicate.partial_eliminate
 0.75% :     0.000001s :     10: predicate.print_const_string_wrapper
 0.98% :     0.000002s :     10: predicate.reduce_eliminate
 1.04% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.32% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.31% :     0.000002s :     18: predicate.replace_applicator
 0.44% :     0.000001s :      6: predicate.replace_old_param
 0.20% :     0.000000s :      3: predicate.reset_defer_inline
 1.00% :     0.000002s :     10: predicate.reshape_eliminate
 0.79% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.59% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.80% :     0.000001s :     10: predicate.same_eliminate
 0.35% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.60% :     0.000001s :      6: predicate.special_op_eliminate
 0.60% :     0.000001s :      6: predicate.specialize_transform
 0.94% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 0.80% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.29% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 1.51% :     0.000003s :     16: predicate.switch_defer_inline
 1.35% :     0.000002s :     16: predicate.switch_layer_defer_inline
 5.23% :     0.000009s :     53: predicate.switch_simplify
 0.84% :     0.000001s :     10: predicate.tile_eliminate
 0.76% :     0.000001s :     10: predicate.transpose_eliminate
 0.86% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.79% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 2.57% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.01% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 0.95% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.12% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 1.86% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.29% :     0.000002s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.53% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000591    12
50.45% :     0.000298s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.41% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.14% :     0.000243s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024062    76
 0.24% :     0.000058s :      1: add_recomputation
 0.28% :     0.000067s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.46% :     0.000352s :      1: bootstrap
 0.14% :     0.000033s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
34.21% :     0.008231s :      1: jit_opt_a
 0.89% :     0.000215s :      1: jit_opt_after_cconv
 0.23% :     0.000055s :      1: jit_opt_b
 1.80% :     0.000432s :      1: loop_unroll
 3.16% :     0.000760s :      1: mutable_eliminate
 3.55% :     0.000854s :     26: opt.transform.jit_opt_a
 0.46% :     0.000111s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.10% :     0.000505s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000039s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 1.64% :     0.000394s :      1: renormalize.infer
 1.26% :     0.000304s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000046s :      1: rewriter_after_opt_a
 0.33% :     0.000079s :      1: rewriter_before_opt_a
 0.33% :     0.000081s :      1: symbol_engine_optimizer
26.47% :     0.006369s :      1: task_emit
19.66% :     0.004730s :      1: type_inference
 0.27% :     0.000066s :      1: validate


TotalTime = 0.0171843, [33]
[bootstrap]: 0.00031289
[type_inference]: 0.0043127
[event_method]: 1.506e-05
[auto_monad]: 3.998e-05
[graph_reusing]: 3.51999e-06
[pre_auto_parallel]: 1.62999e-06
[py_interpret_to_execute]: 3.099e-05
[rewriter_before_opt_a]: 6.232e-05
[expand_dump_flag]: 1.87999e-06
[jit_opt_a]: 0.00494028, [2]
    [Cycle 1]: 0.00139453, [27]
        [switch_simplify]: 4.458e-05
        [loop_unroll]: 2.907e-05
        [a_1]: 0.00045098
        [with_stream_mark]: 1.071e-05
        [recompute_prepare]: 6.46e-06
        [updatestate_depend_eliminate]: 2.78e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 5.61e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 1.02e-06
        [get_grad_eliminate_]: 4.93001e-06
        [merge_forward]: 2.26998e-06
        [cell_reuse_recompute_pass]: 8.30012e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.048e-05
        [j_node_and_user_rematch]: 7.91001e-06
        [meta_fg_expand]: 1.90001e-06
        [replace_old_param]: 7.78001e-06
        [inline_without_move]: 5.24e-06
        [renormalize]: 0.00053092
        [add_forward_monad_depend]: 3.53e-06
        [auto_monad_grad]: 2.16e-06
        [auto_monad_eliminator]: 1.009e-05
        [cse]: 1.784e-05
        [replace_applicator]: 7.281e-05
    [Cycle 2]: 0.00031454, [27]
        [switch_simplify]: 6.44001e-06
        [loop_unroll]: 5.12e-06
        [a_1]: 8.953e-05
        [with_stream_mark]: 8.33999e-06
        [recompute_prepare]: 4.75001e-06
        [updatestate_depend_eliminate]: 2.43e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.76998e-06
        [parameter_eliminate]: 8.59989e-07
        [specialize_transform]: 5.09e-06
        [updatestate_useless_node_eliminater]: 4.52e-06
        [accelerated_algorithm]: 4.73001e-06
        [meta_shard_fg_expand]: 9.80013e-07
        [get_grad_eliminate_]: 4.4e-06
        [merge_forward]: 2.22001e-06
        [cell_reuse_recompute_pass]: 1.91e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.133e-05
        [j_node_and_user_rematch]: 7.46001e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 8.75999e-06
        [inline_without_move]: 4.55001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.55999e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 5.80002e-06
        [cse]: 1.167e-05
        [replace_applicator]: 4.53001e-06
[py_interpret_to_execute_after_opt_a]: 8.40999e-06
[rewriter_after_opt_a]: 2.954e-05
[convert_after_rewriter]: 5.89999e-06
[order_py_execute_after_rewriter]: 4.43001e-06
[mutable_eliminate]: 0.00048
[jit_opt_b]: 4.562e-05, [1]
    [Cycle 1]: 3.984e-05, [2]
        [frontend_op_eliminate]: 1.488e-05
        [inline_after_opt_a]: 1.411e-05
[cconv]: 1.764e-05
[loop_unroll]: 0.0004096
[jit_opt_after_cconv]: 0.00012854, [1]
    [Cycle 1]: 0.0001227, [11]
        [c_1]: 1.899e-05
        [parameter_eliminate]: 1.97001e-06
        [updatestate_depend_eliminate]: 4.35e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [cse]: 1.559e-05
        [call_graph_tuple_transform]: 1.717e-05
        [tuple_list_get_item_eliminator]: 5.10999e-06
        [none_parameter_eliminate]: 1.22e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 5.19998e-06
[remove_dup_value]: 9.83998e-06
[partial_unused_args_eliminate]: 2.01998e-06
[environ_conv]: 4.12e-06
[add_recomputation]: 3.702e-05
[cse_after_recomputation]: 1.99e-05, [1]
    [Cycle 1]: 1.45e-05, [1]
        [cse]: 8.49002e-06
[auto_monad_reorder]: 1.189e-05
[get_jit_bprop_graph]: 2.35002e-06
[rewriter_after_jit_bprop_graph]: 4e-06
[opt_after_jit_grad]: 0.00045678
[symbol_engine_optimizer]: 6.736e-05, [1]
    [Cycle 1]: 6.221e-05, [6]
        [build]: 2.60997e-06
        [elim_shapecalc]: 7.11001e-06
        [elim_not_effective]: 1.17e-05
        [opt_reshape]: 5.86e-06
        [fold_const_symbol]: 8.2e-06
        [renormalize]: 3.50003e-07
[validate]: 2.648e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.00551809
[execute]: 6.19999e-06
Sums
    bootstrap                            :     0.000313s :  2.33%
    type_inference                       :     0.004313s : 32.15%
    event_method                         :     0.000015s :  0.11%
    auto_monad                           :     0.000040s :  0.30%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000031s :  0.23%
    rewriter_before_opt_a                :     0.000062s :  0.46%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000051s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000541s :  4.03%
    jit_opt_a.with_stream_mark           :     0.000019s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000015s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000531s :  3.96%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.12%
    jit_opt_a.cse                        :     0.000030s :  0.22%
    jit_opt_a.replace_applicator         :     0.000077s :  0.58%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000030s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000480s :  3.58%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000018s :  0.13%
    loop_unroll                          :     0.000410s :  3.05%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000010s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000037s :  0.28%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000457s :  3.41%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000026s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005518s : 41.14%
    execute                              :     0.000006s :  0.05%


Time group info:
------[substitution.]   0.000142    23
 1.24% :     0.000002s :      2: substitution.elim_not_effective
 0.98% :     0.000001s :      2: substitution.fold_const_symbol
 3.26% :     0.000005s :      3: substitution.graph_param_transform
82.59% :     0.000117s :      4: substitution.inline
 1.63% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.73% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.21% :     0.000003s :      2: substitution.replace_old_param
 5.35% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004261     2
81.86% :     0.003488s :      1: type_inference.infer
18.14% :     0.000773s :      1: type_inference.specialize
------[replace.]   0.000058     6
64.57% :     0.000038s :      4: replace.inline
35.43% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000121     6
94.62% :     0.000115s :      4: match.inline
 5.38% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.68% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.46% :     0.000002s :     10: predicate.addn_zero_filter
 1.86% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.04% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 4.70% :     0.000005s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.37% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.44% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.07% :     0.000002s :     15: predicate.partial_eliminate
 1.43% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.56% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.99% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.62% :     0.000002s :     10: predicate.transpose_eliminate
 1.34% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.61% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.29% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000534    12
46.45% :     0.000248s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.09% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.46% :     0.000259s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018609    76
 0.21% :     0.000040s :      1: add_recomputation
 0.23% :     0.000043s :      1: auto_monad
 0.08% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.75% :     0.000326s :      1: bootstrap
 0.11% :     0.000021s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.10% :     0.000018s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
26.56% :     0.004943s :      1: jit_opt_a
 0.71% :     0.000131s :      1: jit_opt_after_cconv
 0.26% :     0.000048s :      1: jit_opt_b
 2.24% :     0.000417s :      1: loop_unroll
 2.62% :     0.000487s :      1: mutable_eliminate
 4.27% :     0.000794s :     26: opt.transform.jit_opt_a
 0.23% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.49% :     0.000463s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000012s :      1: remove_dup_value
 1.56% :     0.000290s :      1: renormalize.infer
 1.26% :     0.000234s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000032s :      1: rewriter_after_opt_a
 0.35% :     0.000065s :      1: rewriter_before_opt_a
 0.38% :     0.000070s :      1: symbol_engine_optimizer
29.70% :     0.005528s :      1: task_emit
23.22% :     0.004321s :      1: type_inference
 0.24% :     0.000044s :      1: validate


TotalTime = 0.0309909, [33]
[bootstrap]: 0.00023711
[type_inference]: 0.0106324
[event_method]: 4.651e-05
[auto_monad]: 8.671e-05
[graph_reusing]: 6.24999e-06
[pre_auto_parallel]: 2.22001e-06
[py_interpret_to_execute]: 2.408e-05
[rewriter_before_opt_a]: 8.985e-05
[expand_dump_flag]: 2.64001e-06
[jit_opt_a]: 0.011018, [3]
    [Cycle 1]: 0.00534963, [27]
        [switch_simplify]: 5.904e-05
        [loop_unroll]: 4.293e-05
        [a_1]: 0.000886
        [with_stream_mark]: 2.328e-05
        [recompute_prepare]: 1.782e-05
        [updatestate_depend_eliminate]: 7.56999e-06
        [updatestate_assign_eliminate]: 6.43e-06
        [updatestate_loads_eliminate]: 5.47001e-06
        [parameter_eliminate]: 2.48e-06
        [specialize_transform]: 1.316e-05
        [updatestate_useless_node_eliminater]: 1.239e-05
        [accelerated_algorithm]: 1.293e-05
        [meta_shard_fg_expand]: 3.26001e-06
        [get_grad_eliminate_]: 1.298e-05
        [merge_forward]: 7.66999e-06
        [cell_reuse_recompute_pass]: 7.59988e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.404e-05
        [j_node_and_user_rematch]: 2.323e-05
        [meta_fg_expand]: 0.00142473
        [replace_old_param]: 4.854e-05
        [inline_without_move]: 5.127e-05
        [renormalize]: 0.00220408
        [add_forward_monad_depend]: 9.15999e-06
        [auto_monad_grad]: 5.68002e-06
        [auto_monad_eliminator]: 4.593e-05
        [cse]: 0.00014133
        [replace_applicator]: 6.664e-05
    [Cycle 2]: 0.00204713, [27]
        [switch_simplify]: 3.912e-05
        [loop_unroll]: 3.645e-05
        [a_1]: 0.00108068
        [with_stream_mark]: 1.498e-05
        [recompute_prepare]: 9.67999e-06
        [updatestate_depend_eliminate]: 3.85e-06
        [updatestate_assign_eliminate]: 3.53e-06
        [updatestate_loads_eliminate]: 2.60002e-06
        [parameter_eliminate]: 1.49e-06
        [specialize_transform]: 6.79001e-06
        [updatestate_useless_node_eliminater]: 6.09999e-06
        [accelerated_algorithm]: 6.49999e-06
        [meta_shard_fg_expand]: 2.17001e-06
        [get_grad_eliminate_]: 6.17001e-06
        [merge_forward]: 2.96001e-06
        [cell_reuse_recompute_pass]: 9.50007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.35e-05
        [j_node_and_user_rematch]: 9.92999e-06
        [meta_fg_expand]: 4.613e-05
        [replace_old_param]: 1.054e-05
        [inline_without_move]: 5.92001e-06
        [renormalize]: 0.00053215
        [add_forward_monad_depend]: 4.57e-06
        [auto_monad_grad]: 2.03997e-06
        [auto_monad_eliminator]: 1.151e-05
        [cse]: 2.314e-05
        [replace_applicator]: 1.415e-05
    [Cycle 3]: 0.00037103, [27]
        [switch_simplify]: 6.94999e-06
        [loop_unroll]: 6.04999e-06
        [a_1]: 0.00012053
        [with_stream_mark]: 9.69e-06
        [recompute_prepare]: 6.04999e-06
        [updatestate_depend_eliminate]: 3.79002e-06
        [updatestate_assign_eliminate]: 2.64999e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [parameter_eliminate]: 1.27999e-06
        [specialize_transform]: 6.49999e-06
        [updatestate_useless_node_eliminater]: 5.85002e-06
        [accelerated_algorithm]: 6.03002e-06
        [meta_shard_fg_expand]: 1.50001e-06
        [get_grad_eliminate_]: 5.54e-06
        [merge_forward]: 3.01001e-06
        [cell_reuse_recompute_pass]: 1.79998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.541e-05
        [j_node_and_user_rematch]: 1.006e-05
        [meta_fg_expand]: 2.17001e-06
        [replace_old_param]: 8.28999e-06
        [inline_without_move]: 5.68002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 6.79001e-06
        [cse]: 1.406e-05
        [replace_applicator]: 6.21998e-06
[py_interpret_to_execute_after_opt_a]: 1.229e-05
[rewriter_after_opt_a]: 3.508e-05
[convert_after_rewriter]: 6.92002e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.00060189
[jit_opt_b]: 5.743e-05, [1]
    [Cycle 1]: 5.007e-05, [2]
        [frontend_op_eliminate]: 1.951e-05
        [inline_after_opt_a]: 1.905e-05
[cconv]: 2.319e-05
[loop_unroll]: 0.00042467
[jit_opt_after_cconv]: 0.00015757, [1]
    [Cycle 1]: 0.00015022, [11]
        [c_1]: 2.636e-05
        [parameter_eliminate]: 3.81001e-06
        [updatestate_depend_eliminate]: 5.42999e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.96001e-06
        [cse]: 2.335e-05
        [call_graph_tuple_transform]: 2.125e-05
        [tuple_list_get_item_eliminator]: 6.33998e-06
        [none_parameter_eliminate]: 1.48002e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 6.26998e-06
[remove_dup_value]: 1.437e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 5.07999e-06
[add_recomputation]: 4.785e-05
[cse_after_recomputation]: 2.869e-05, [1]
    [Cycle 1]: 2.247e-05, [1]
        [cse]: 1.581e-05
[auto_monad_reorder]: 1.742e-05
[get_jit_bprop_graph]: 1.87001e-06
[rewriter_after_jit_bprop_graph]: 5.46e-06
[opt_after_jit_grad]: 0.00047419
[symbol_engine_optimizer]: 8.295e-05, [1]
    [Cycle 1]: 7.645e-05, [6]
        [build]: 3.96001e-06
        [elim_shapecalc]: 9.00999e-06
        [elim_not_effective]: 1.585e-05
        [opt_reshape]: 7.22002e-06
        [fold_const_symbol]: 1.124e-05
        [renormalize]: 3.69997e-07
[validate]: 3.367e-05
[backend_pass]: 7.09988e-07
[task_emit]: 0.00658298
[execute]: 6.54999e-06
Sums
    bootstrap                            :     0.000237s :  0.88%
    type_inference                       :     0.010632s : 39.45%
    event_method                         :     0.000047s :  0.17%
    auto_monad                           :     0.000087s :  0.32%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000024s :  0.09%
    rewriter_before_opt_a                :     0.000090s :  0.33%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000105s :  0.39%
    jit_opt_a.loop_unroll                :     0.000085s :  0.32%
    jit_opt_a.a_1                        :     0.002087s :  7.75%
    jit_opt_a.with_stream_mark           :     0.000048s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000034s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000025s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.09%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000053s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000043s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001473s :  5.47%
    jit_opt_a.replace_old_param          :     0.000067s :  0.25%
    jit_opt_a.inline_without_move        :     0.000063s :  0.23%
    jit_opt_a.renormalize                :     0.002736s : 10.15%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000064s :  0.24%
    jit_opt_a.cse                        :     0.000179s :  0.66%
    jit_opt_a.replace_applicator         :     0.000087s :  0.32%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.05%
    rewriter_after_opt_a                 :     0.000035s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000602s :  2.23%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.07%
    cconv                                :     0.000023s :  0.09%
    loop_unroll                          :     0.000425s :  1.58%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000014s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000048s :  0.18%
    cse_after_recomputation.cse          :     0.000016s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000474s :  1.76%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006583s : 24.43%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000476   110
 0.51% :     0.000002s :      3: substitution.elim_not_effective
 0.33% :     0.000002s :      3: substitution.fold_const_symbol
 1.16% :     0.000005s :      4: substitution.graph_param_transform
63.24% :     0.000301s :     12: substitution.inline
 3.40% :     0.000016s :      2: substitution.inline_without_move
 1.89% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.28% :     0.000011s :      7: substitution.minmaximum_grad
 1.31% :     0.000006s :      5: substitution.partial_eliminate
 2.20% :     0.000010s :     14: substitution.remove_not_recompute_node
 5.63% :     0.000027s :     10: substitution.replace_applicator
 1.67% :     0.000008s :      7: substitution.replace_old_param
 0.51% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.53% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.83% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 8.50% :     0.000040s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010557     2
90.04% :     0.009505s :      1: type_inference.infer
 9.96% :     0.001052s :      1: type_inference.specialize
------[replace.]   0.000146    19
63.57% :     0.000093s :     12: replace.inline
36.43% :     0.000053s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000314    19
93.35% :     0.000293s :     12: match.inline
 6.65% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000384  2568
 1.59% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.38% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.37% :     0.000005s :     43: predicate.addn_check_dump
 1.81% :     0.000007s :     43: predicate.addn_zero_filter
 2.16% :     0.000008s :     43: predicate.arithmetic_simplify
 1.48% :     0.000006s :     43: predicate.cast_eliminate
 0.23% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.49% :     0.000006s :     43: predicate.compare_switch_simplify
 1.64% :     0.000006s :     43: predicate.depend_value_elim
 1.48% :     0.000006s :     43: predicate.dict_get_item_const_eliminator
 1.46% :     0.000006s :     43: predicate.dict_get_item_eliminator
 1.54% :     0.000006s :     43: predicate.dict_set_item_eliminator
 0.28% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      4: predicate.elim_not_effective
 0.19% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.48% :     0.000006s :     43: predicate.environ_add_const_eliminate
 1.38% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.59% :     0.000006s :     43: predicate.environ_get_depend_swap
 1.45% :     0.000006s :     43: predicate.environ_get_eliminate
 1.55% :     0.000006s :     43: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.76% :     0.000003s :     19: predicate.get_grad_eliminate
 0.11% :     0.000000s :      4: predicate.graph_param_transform
 4.29% :     0.000016s :     70: predicate.inline
 1.73% :     0.000007s :     40: predicate.inline_without_move
 0.35% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.89% :     0.000003s :     19: predicate.less_batch_normalization
 1.73% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.88% :     0.000007s :     54: predicate.load_eliminater
 0.45% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.35% :     0.000013s :     85: predicate.loop_unroll_before_grad
 1.77% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.66% :     0.000006s :     43: predicate.merge_addn
 1.76% :     0.000007s :     43: predicate.minmaximum_grad
 0.49% :     0.000002s :      4: predicate.mutable_eliminate
 0.20% :     0.000001s :      4: predicate.opt_reshape
 2.24% :     0.000009s :     54: predicate.partial_eliminate
 1.43% :     0.000005s :     43: predicate.print_const_string_wrapper
 1.94% :     0.000007s :     43: predicate.reduce_eliminate
 1.75% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.40% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.51% :     0.000010s :     96: predicate.replace_applicator
 0.90% :     0.000003s :     40: predicate.replace_old_param
 0.10% :     0.000000s :      4: predicate.reset_defer_inline
 1.67% :     0.000006s :     43: predicate.reshape_eliminate
 1.50% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.28% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.47% :     0.000006s :     43: predicate.same_eliminate
 0.52% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.39% :     0.000001s :      8: predicate.special_op_eliminate
 0.83% :     0.000003s :     19: predicate.specialize_transform
 1.64% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000005s :     43: predicate.stack_unstack_eliminate
 0.16% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.96% :     0.000011s :     62: predicate.switch_defer_inline
 2.58% :     0.000010s :     62: predicate.switch_layer_defer_inline
 6.23% :     0.000024s :    151: predicate.switch_simplify
 1.48% :     0.000006s :     43: predicate.tile_eliminate
 1.50% :     0.000006s :     43: predicate.transpose_eliminate
 1.77% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.64% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.25% :     0.000012s :     58: predicate.tuple_list_get_item_eliminator
 2.23% :     0.000009s :     43: predicate.tuple_list_set_item_eliminator
 1.73% :     0.000007s :     50: predicate.tuple_to_list_eliminator_
 1.88% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000011s :     73: predicate.updatestate_useless_node_eliminater
 2.07% :     0.000008s :     43: predicate.value_based_eliminate
 0.15% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.27% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001438    29
60.63% :     0.000872s :     13: func_graph_cloner_run.FuncGraphClonerGraph
39.37% :     0.000566s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.036540    91
 0.14% :     0.000052s :      1: add_recomputation
 0.27% :     0.000097s :      1: auto_monad
 0.05% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.71% :     0.000260s :      1: bootstrap
 0.07% :     0.000026s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000031s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.14% :     0.000051s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
30.16% :     0.011021s :      1: jit_opt_a
 0.44% :     0.000160s :      1: jit_opt_after_cconv
 0.16% :     0.000060s :      1: jit_opt_b
 1.18% :     0.000432s :      1: loop_unroll
 1.67% :     0.000609s :      1: mutable_eliminate
 7.35% :     0.002684s :     39: opt.transform.jit_opt_a
 0.16% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000032s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000040s :      4: opt.transform.symbol_engine_opt
 1.32% :     0.000481s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000027s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000024s :      1: remove_dup_value
 3.89% :     0.001422s :      2: renormalize.infer
 3.56% :     0.001301s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000038s :      1: rewriter_after_opt_a
 0.25% :     0.000093s :      1: rewriter_before_opt_a
 0.23% :     0.000086s :      1: symbol_engine_optimizer
18.06% :     0.006598s :      1: task_emit
29.13% :     0.010642s :      1: type_inference
 0.15% :     0.000054s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.217.772 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dceb160> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.218.304 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0124409, [30]
[bootstrap]: 0.00027891
[type_inference]: 0.00459094
[event_method]: 1.714e-05
[auto_monad]: 6.472e-05
[graph_reusing]: 6.02001e-06
[pre_auto_parallel]: 2.31998e-06
[py_interpret_to_execute]: 3.734e-05
[rewriter_before_opt_a]: 7.106e-05
[expand_dump_flag]: 3.55998e-06
[jit_opt_a]: 0.00533707, [2]
    [Cycle 1]: 0.00165037, [27]
        [switch_simplify]: 5.474e-05
        [loop_unroll]: 2.976e-05
        [a_1]: 0.00054084
        [with_stream_mark]: 1.614e-05
        [recompute_prepare]: 7.05998e-06
        [updatestate_depend_eliminate]: 3.30998e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.71999e-06
        [parameter_eliminate]: 1.82999e-06
        [specialize_transform]: 6.21e-06
        [updatestate_useless_node_eliminater]: 4.91002e-06
        [accelerated_algorithm]: 5.84e-06
        [meta_shard_fg_expand]: 1.98002e-06
        [get_grad_eliminate_]: 5.24998e-06
        [merge_forward]: 3.86999e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.457e-05
        [j_node_and_user_rematch]: 8.92999e-06
        [meta_fg_expand]: 2.00002e-06
        [replace_old_param]: 8.67e-06
        [inline_without_move]: 5.14998e-06
        [renormalize]: 0.00069417
        [add_forward_monad_depend]: 5.30999e-06
        [auto_monad_grad]: 2.49999e-06
        [auto_monad_eliminator]: 1.382e-05
        [cse]: 2.493e-05
        [replace_applicator]: 1.237e-05
    [Cycle 2]: 0.00031837, [27]
        [switch_simplify]: 5.44e-06
        [loop_unroll]: 4.94998e-06
        [a_1]: 8.959e-05
        [with_stream_mark]: 9.02e-06
        [recompute_prepare]: 5.35999e-06
        [updatestate_depend_eliminate]: 2.69001e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 4.87e-06
        [updatestate_useless_node_eliminater]: 4.71002e-06
        [accelerated_algorithm]: 5.34998e-06
        [meta_shard_fg_expand]: 1.05001e-06
        [get_grad_eliminate_]: 4.43001e-06
        [merge_forward]: 2.49001e-06
        [cell_reuse_recompute_pass]: 1.47999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.227e-05
        [j_node_and_user_rematch]: 7.97998e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 7.18e-06
        [inline_without_move]: 4.53999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 4.94e-06
        [cse]: 1.082e-05
        [replace_applicator]: 4.74e-06
[py_interpret_to_execute_after_opt_a]: 1.076e-05
[rewriter_after_opt_a]: 3.318e-05
[convert_after_rewriter]: 9.12999e-06
[order_py_execute_after_rewriter]: 4.72e-06
[mutable_eliminate]: 0.00052588
[jit_opt_b]: 4.999e-05, [1]
    [Cycle 1]: 4.198e-05, [2]
        [frontend_op_eliminate]: 1.659e-05
        [inline_after_opt_a]: 1.425e-05
[cconv]: 2.497e-05
[loop_unroll]: 0.0004133
[jit_opt_after_cconv]: 0.0001376, [1]
    [Cycle 1]: 0.00013164, [11]
        [c_1]: 2.007e-05
        [parameter_eliminate]: 2.63e-06
        [updatestate_depend_eliminate]: 5.33002e-06
        [updatestate_assign_eliminate]: 2.48002e-06
        [updatestate_loads_eliminate]: 2.08002e-06
        [cse]: 1.738e-05
        [call_graph_tuple_transform]: 1.848e-05
        [tuple_list_get_item_eliminator]: 5.18002e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 5.25001e-06
[remove_dup_value]: 1.396e-05
[partial_unused_args_eliminate]: 2.68998e-06
[environ_conv]: 5.47001e-06
[add_recomputation]: 4.589e-05
[cse_after_recomputation]: 1.908e-05, [1]
    [Cycle 1]: 1.41e-05, [1]
        [cse]: 8.62e-06
[auto_monad_reorder]: 1.435e-05
[get_jit_bprop_graph]: 1.40001e-06
[rewriter_after_jit_bprop_graph]: 4.92e-06
[opt_after_jit_grad]: 0.00044855
[symbol_engine_optimizer]: 7.217e-05, [1]
    [Cycle 1]: 6.598e-05, [6]
        [build]: 3.45003e-06
        [elim_shapecalc]: 7.78001e-06
        [elim_not_effective]: 1.224e-05
        [opt_reshape]: 5.56e-06
        [fold_const_symbol]: 8.82999e-06
        [renormalize]: 6.29982e-07
[validate]: 3.268e-05
Sums
    bootstrap                            :     0.000279s :  3.28%
    type_inference                       :     0.004591s : 53.97%
    event_method                         :     0.000017s :  0.20%
    auto_monad                           :     0.000065s :  0.76%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000037s :  0.44%
    rewriter_before_opt_a                :     0.000071s :  0.84%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000060s :  0.71%
    jit_opt_a.loop_unroll                :     0.000035s :  0.41%
    jit_opt_a.a_1                        :     0.000630s :  7.41%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000694s :  8.16%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.22%
    jit_opt_a.cse                        :     0.000036s :  0.42%
    jit_opt_a.replace_applicator         :     0.000017s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.13%
    rewriter_after_opt_a                 :     0.000033s :  0.39%
    convert_after_rewriter               :     0.000009s :  0.11%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000526s :  6.18%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.17%
    cconv                                :     0.000025s :  0.29%
    loop_unroll                          :     0.000413s :  4.86%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.16%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000046s :  0.54%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000014s :  0.17%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000449s :  5.27%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000033s :  0.38%


Time group info:
------[substitution.]   0.000210    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.40% :     0.000005s :      3: substitution.graph_param_transform
84.81% :     0.000178s :      4: substitution.inline
 1.33% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.34% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000004s :      2: substitution.replace_old_param
 5.68% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004526     2
77.62% :     0.003513s :      1: type_inference.infer
22.38% :     0.001013s :      1: type_inference.specialize
------[replace.]   0.000070     6
65.79% :     0.000046s :      4: replace.inline
34.21% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000186     6
94.20% :     0.000176s :      4: match.inline
 5.80% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.44% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 2.18% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.30% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.40% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000001s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.77% :     0.000002s :      3: predicate.mutable_eliminate
 0.60% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     15: predicate.partial_eliminate
 1.26% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.65% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.48% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.69% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.75% :     0.000009s :     53: predicate.switch_simplify
 1.28% :     0.000001s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.64% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 2.22% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000692    12
46.41% :     0.000321s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.07% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.51% :     0.000329s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014030    72
 0.35% :     0.000049s :      1: add_recomputation
 0.48% :     0.000068s :      1: auto_monad
 0.12% :     0.000017s :      1: auto_monad_reorder
 2.10% :     0.000295s :      1: bootstrap
 0.20% :     0.000027s :      1: cconv
 0.08% :     0.000011s :      1: convert_after_rewriter
 0.15% :     0.000021s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.15% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.07% :     0.005341s :      1: jit_opt_a
 1.00% :     0.000140s :      1: jit_opt_after_cconv
 0.38% :     0.000053s :      1: jit_opt_b
 3.00% :     0.000421s :      1: loop_unroll
 3.81% :     0.000534s :      1: mutable_eliminate
 5.98% :     0.000839s :     26: opt.transform.jit_opt_a
 0.32% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.25% :     0.000456s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.28% :     0.000040s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.61% :     0.000366s :      1: renormalize.infer
 2.29% :     0.000321s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000036s :      1: rewriter_after_opt_a
 0.53% :     0.000074s :      1: rewriter_before_opt_a
 0.53% :     0.000075s :      1: symbol_engine_optimizer
32.81% :     0.004603s :      1: type_inference


TotalTime = 0.0179461, [33]
[bootstrap]: 0.00031431
[type_inference]: 0.00451747
[event_method]: 1.585e-05
[auto_monad]: 5.526e-05
[graph_reusing]: 4.71002e-06
[pre_auto_parallel]: 2.22999e-06
[py_interpret_to_execute]: 3.066e-05
[rewriter_before_opt_a]: 6.494e-05
[expand_dump_flag]: 2.59999e-06
[jit_opt_a]: 0.00498106, [2]
    [Cycle 1]: 0.00142263, [27]
        [switch_simplify]: 4.546e-05
        [loop_unroll]: 2.799e-05
        [a_1]: 0.00047254
        [with_stream_mark]: 1.13e-05
        [recompute_prepare]: 6.83998e-06
        [updatestate_depend_eliminate]: 2.98e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [parameter_eliminate]: 1.56002e-06
        [specialize_transform]: 5.76e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 5.13002e-06
        [meta_shard_fg_expand]: 1.73002e-06
        [get_grad_eliminate_]: 4.83001e-06
        [merge_forward]: 2.76e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.139e-05
        [j_node_and_user_rematch]: 8.54998e-06
        [meta_fg_expand]: 2.29999e-06
        [replace_old_param]: 7.43999e-06
        [inline_without_move]: 4.98001e-06
        [renormalize]: 0.00057692
        [add_forward_monad_depend]: 4.65001e-06
        [auto_monad_grad]: 2.18002e-06
        [auto_monad_eliminator]: 1.21e-05
        [cse]: 2.187e-05
        [replace_applicator]: 1.154e-05
    [Cycle 2]: 0.00031122, [27]
        [switch_simplify]: 5.20999e-06
        [loop_unroll]: 4.92999e-06
        [a_1]: 8.897e-05
        [with_stream_mark]: 8.89e-06
        [recompute_prepare]: 4.82e-06
        [updatestate_depend_eliminate]: 2.49001e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.60001e-06
        [updatestate_useless_node_eliminater]: 4.60001e-06
        [accelerated_algorithm]: 4.92e-06
        [meta_shard_fg_expand]: 1.17999e-06
        [get_grad_eliminate_]: 4.36002e-06
        [merge_forward]: 2.71999e-06
        [cell_reuse_recompute_pass]: 1.97001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.182e-05
        [j_node_and_user_rematch]: 7.68999e-06
        [meta_fg_expand]: 1.40999e-06
        [replace_old_param]: 6.84999e-06
        [inline_without_move]: 4.53001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 5.39e-06
        [cse]: 9.96e-06
        [replace_applicator]: 4.68999e-06
[py_interpret_to_execute_after_opt_a]: 8.84e-06
[rewriter_after_opt_a]: 2.993e-05
[convert_after_rewriter]: 6.09999e-06
[order_py_execute_after_rewriter]: 4.87e-06
[mutable_eliminate]: 0.00053376
[jit_opt_b]: 5.005e-05, [1]
    [Cycle 1]: 4.336e-05, [2]
        [frontend_op_eliminate]: 1.68e-05
        [inline_after_opt_a]: 1.482e-05
[cconv]: 2.208e-05
[loop_unroll]: 0.00051604
[jit_opt_after_cconv]: 0.00013023, [1]
    [Cycle 1]: 0.00012441, [11]
        [c_1]: 1.941e-05
        [parameter_eliminate]: 2.04e-06
        [updatestate_depend_eliminate]: 4.28001e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [cse]: 1.63e-05
        [call_graph_tuple_transform]: 1.655e-05
        [tuple_list_get_item_eliminator]: 5.46998e-06
        [none_parameter_eliminate]: 1.14e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 4.99e-06
[remove_dup_value]: 1.245e-05
[partial_unused_args_eliminate]: 2.39001e-06
[environ_conv]: 4.81002e-06
[add_recomputation]: 4.408e-05
[cse_after_recomputation]: 2.13e-05, [1]
    [Cycle 1]: 1.556e-05, [1]
        [cse]: 9.52001e-06
[auto_monad_reorder]: 1.375e-05
[get_jit_bprop_graph]: 1.82999e-06
[rewriter_after_jit_bprop_graph]: 4.49002e-06
[opt_after_jit_grad]: 0.00046257
[symbol_engine_optimizer]: 7.347e-05, [1]
    [Cycle 1]: 6.747e-05, [6]
        [build]: 3.44001e-06
        [elim_shapecalc]: 7.78001e-06
        [elim_not_effective]: 1.217e-05
        [opt_reshape]: 5.47001e-06
        [fold_const_symbol]: 8.84e-06
        [renormalize]: 4.19997e-07
[validate]: 3.092e-05
[backend_pass]: 1.00001e-06
[task_emit]: 0.00579363
[execute]: 6.16998e-06
Sums
    bootstrap                            :     0.000314s :  2.23%
    type_inference                       :     0.004517s : 31.98%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000055s :  0.39%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.22%
    rewriter_before_opt_a                :     0.000065s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000051s :  0.36%
    jit_opt_a.loop_unroll                :     0.000033s :  0.23%
    jit_opt_a.a_1                        :     0.000562s :  3.98%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000577s :  4.09%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.12%
    jit_opt_a.cse                        :     0.000032s :  0.23%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000030s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000534s :  3.78%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000022s :  0.16%
    loop_unroll                          :     0.000516s :  3.65%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000044s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000463s :  3.27%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005794s : 41.02%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000160    23
 1.15% :     0.000002s :      2: substitution.elim_not_effective
 1.04% :     0.000002s :      2: substitution.fold_const_symbol
 2.92% :     0.000005s :      3: substitution.graph_param_transform
81.80% :     0.000131s :      4: substitution.inline
 1.78% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.46% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.86% :     0.000003s :      2: substitution.replace_old_param
 6.99% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004459     2
82.78% :     0.003691s :      1: type_inference.infer
17.22% :     0.000768s :      1: type_inference.specialize
------[replace.]   0.000061     6
68.14% :     0.000041s :      4: replace.inline
31.86% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000139     6
92.71% :     0.000129s :      4: match.inline
 7.29% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.57% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.46% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.57% :     0.000002s :     10: predicate.addn_zero_filter
 1.81% :     0.000002s :     10: predicate.arithmetic_simplify
 1.34% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.77% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.46% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.33% :     0.000000s :      3: predicate.graph_param_transform
 5.00% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.10% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.18% :     0.000001s :     10: predicate.minmaximum_grad
 1.46% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000002s :     15: predicate.partial_eliminate
 1.41% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.87% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.21% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.37% :     0.000008s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.37% :     0.000002s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.71% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000569    12
52.98% :     0.000301s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.13% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.89% :     0.000244s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019370    76
 0.25% :     0.000048s :      1: add_recomputation
 0.30% :     0.000058s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.68% :     0.000326s :      1: bootstrap
 0.13% :     0.000025s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
25.73% :     0.004984s :      1: jit_opt_a
 0.69% :     0.000133s :      1: jit_opt_after_cconv
 0.27% :     0.000053s :      1: jit_opt_b
 2.70% :     0.000524s :      1: loop_unroll
 2.80% :     0.000542s :      1: mutable_eliminate
 3.88% :     0.000751s :     26: opt.transform.jit_opt_a
 0.22% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.43% :     0.000470s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.56% :     0.000302s :      1: renormalize.infer
 1.38% :     0.000268s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.35% :     0.000068s :      1: rewriter_before_opt_a
 0.39% :     0.000076s :      1: symbol_engine_optimizer
29.98% :     0.005808s :      1: task_emit
23.37% :     0.004526s :      1: type_inference
 0.25% :     0.000048s :      1: validate


TotalTime = 0.0186072, [33]
[bootstrap]: 0.00026684
[type_inference]: 0.00439022
[event_method]: 1.637e-05
[auto_monad]: 6.03e-05
[graph_reusing]: 5.92001e-06
[pre_auto_parallel]: 2.34999e-06
[py_interpret_to_execute]: 3.309e-05
[rewriter_before_opt_a]: 6.92e-05
[expand_dump_flag]: 3.14999e-06
[jit_opt_a]: 0.00537131, [2]
    [Cycle 1]: 0.00157002, [27]
        [switch_simplify]: 5.506e-05
        [loop_unroll]: 2.991e-05
        [a_1]: 0.00053827
        [with_stream_mark]: 2.001e-05
        [recompute_prepare]: 7.46001e-06
        [updatestate_depend_eliminate]: 3.58e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.78e-06
        [parameter_eliminate]: 1.81998e-06
        [specialize_transform]: 5.66e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.35001e-06
        [meta_shard_fg_expand]: 2.02001e-06
        [get_grad_eliminate_]: 5.02e-06
        [merge_forward]: 3.55e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.396e-05
        [j_node_and_user_rematch]: 8.74003e-06
        [meta_fg_expand]: 2.12001e-06
        [replace_old_param]: 8.42e-06
        [inline_without_move]: 5.10999e-06
        [renormalize]: 0.0006089
        [add_forward_monad_depend]: 5.22e-06
        [auto_monad_grad]: 2.47001e-06
        [auto_monad_eliminator]: 1.249e-05
        [cse]: 2.777e-05
        [replace_applicator]: 1.117e-05
    [Cycle 2]: 0.00039692, [27]
        [switch_simplify]: 5.46e-06
        [loop_unroll]: 4.77e-06
        [a_1]: 8.93e-05
        [with_stream_mark]: 9.42001e-06
        [recompute_prepare]: 5.12999e-06
        [updatestate_depend_eliminate]: 2.94001e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 4.93001e-06
        [updatestate_useless_node_eliminater]: 4.71002e-06
        [accelerated_algorithm]: 5.60001e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.48001e-06
        [merge_forward]: 2.61999e-06
        [cell_reuse_recompute_pass]: 1.65001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.298e-05
        [j_node_and_user_rematch]: 7.51999e-06
        [meta_fg_expand]: 1.92999e-06
        [replace_old_param]: 6.61e-06
        [inline_without_move]: 7.4e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.79e-06
        [auto_monad_grad]: 1.05001e-06
        [auto_monad_eliminator]: 5.54e-06
        [cse]: 1.149e-05
        [replace_applicator]: 4.87e-06
[py_interpret_to_execute_after_opt_a]: 1.116e-05
[rewriter_after_opt_a]: 3.553e-05
[convert_after_rewriter]: 7.15998e-06
[order_py_execute_after_rewriter]: 4.99998e-06
[mutable_eliminate]: 0.00057688
[jit_opt_b]: 4.868e-05, [1]
    [Cycle 1]: 4.153e-05, [2]
        [frontend_op_eliminate]: 1.582e-05
        [inline_after_opt_a]: 1.425e-05
[cconv]: 2.318e-05
[loop_unroll]: 0.00041654
[jit_opt_after_cconv]: 0.00013511, [1]
    [Cycle 1]: 0.00012835, [11]
        [c_1]: 1.922e-05
        [parameter_eliminate]: 2.39001e-06
        [updatestate_depend_eliminate]: 4.95999e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.727e-05
        [call_graph_tuple_transform]: 1.837e-05
        [tuple_list_get_item_eliminator]: 4.89e-06
        [none_parameter_eliminate]: 1.69e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 4.95001e-06
[remove_dup_value]: 1.293e-05
[partial_unused_args_eliminate]: 2.07001e-06
[environ_conv]: 4.99e-06
[add_recomputation]: 4.861e-05
[cse_after_recomputation]: 2.05e-05, [1]
    [Cycle 1]: 1.5e-05, [1]
        [cse]: 9.24e-06
[auto_monad_reorder]: 1.387e-05
[get_jit_bprop_graph]: 1.87001e-06
[rewriter_after_jit_bprop_graph]: 5.56998e-06
[opt_after_jit_grad]: 0.00045248
[symbol_engine_optimizer]: 7.266e-05, [1]
    [Cycle 1]: 6.604e-05, [6]
        [build]: 3.51001e-06
        [elim_shapecalc]: 7.08e-06
        [elim_not_effective]: 1.255e-05
        [opt_reshape]: 5.62999e-06
        [fold_const_symbol]: 8.15e-06
        [renormalize]: 7.79983e-07
[validate]: 3.452e-05
[backend_pass]: 9.10019e-07
[task_emit]: 0.00624623
[execute]: 8.42998e-06
Sums
    bootstrap                            :     0.000267s :  1.84%
    type_inference                       :     0.004390s : 30.24%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000060s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.23%
    rewriter_before_opt_a                :     0.000069s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.42%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000628s :  4.32%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000013s :  0.09%
    jit_opt_a.renormalize                :     0.000609s :  4.20%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000039s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.08%
    rewriter_after_opt_a                 :     0.000036s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000577s :  3.97%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000023s :  0.16%
    loop_unroll                          :     0.000417s :  2.87%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000452s :  3.12%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000035s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006246s : 43.03%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000205    23
 1.12% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.86% :     0.000006s :      3: substitution.graph_param_transform
83.72% :     0.000172s :      4: substitution.inline
 1.49% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.10% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.85% :     0.000004s :      2: substitution.replace_old_param
 6.28% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004328     2
82.25% :     0.003559s :      1: type_inference.infer
17.75% :     0.000768s :      1: type_inference.specialize
------[replace.]   0.000068     6
66.73% :     0.000045s :      4: replace.inline
33.27% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000181     6
93.48% :     0.000169s :      4: match.inline
 6.52% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.29% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.17% :     0.000001s :     10: predicate.addn_check_dump
 1.58% :     0.000002s :     10: predicate.addn_zero_filter
 2.23% :     0.000003s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.59% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.48% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.19% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.65% :     0.000006s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.37% :     0.000002s :      6: predicate.less_batch_normalization
 1.72% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.00% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.14% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.25% :     0.000001s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.75% :     0.000002s :     10: predicate.reshape_eliminate
 1.33% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.60% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.62% :     0.000003s :     16: predicate.switch_defer_inline
 2.25% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.03% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.53% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.82% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.47% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000555    12
50.97% :     0.000283s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.50% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.53% :     0.000236s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020143    76
 0.26% :     0.000052s :      1: add_recomputation
 0.32% :     0.000064s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.42% :     0.000285s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.69% :     0.005375s :      1: jit_opt_a
 0.68% :     0.000138s :      1: jit_opt_after_cconv
 0.25% :     0.000051s :      1: jit_opt_b
 2.10% :     0.000424s :      1: loop_unroll
 2.90% :     0.000585s :      1: mutable_eliminate
 4.15% :     0.000835s :     26: opt.transform.jit_opt_a
 0.22% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.28% :     0.000460s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000036s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.55% :     0.000312s :      1: renormalize.infer
 1.44% :     0.000290s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000039s :      1: rewriter_after_opt_a
 0.36% :     0.000073s :      1: rewriter_before_opt_a
 0.37% :     0.000075s :      1: symbol_engine_optimizer
31.09% :     0.006263s :      1: task_emit
21.87% :     0.004405s :      1: type_inference
 0.27% :     0.000053s :      1: validate


TotalTime = 0.0115903, [30]
[bootstrap]: 0.00029315
[type_inference]: 0.00430115
[event_method]: 1.669e-05
[auto_monad]: 6.047e-05
[graph_reusing]: 5.98002e-06
[pre_auto_parallel]: 2.16e-06
[py_interpret_to_execute]: 3.302e-05
[rewriter_before_opt_a]: 7.216e-05
[expand_dump_flag]: 3.18e-06
[jit_opt_a]: 0.00482251, [2]
    [Cycle 1]: 0.001359, [27]
        [switch_simplify]: 4.448e-05
        [loop_unroll]: 2.837e-05
        [a_1]: 0.00047914
        [with_stream_mark]: 1.315e-05
        [recompute_prepare]: 7.04001e-06
        [updatestate_depend_eliminate]: 2.74999e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [parameter_eliminate]: 1.07998e-06
        [specialize_transform]: 5.69e-06
        [updatestate_useless_node_eliminater]: 5.10999e-06
        [accelerated_algorithm]: 5.72999e-06
        [meta_shard_fg_expand]: 1.41002e-06
        [get_grad_eliminate_]: 5.45001e-06
        [merge_forward]: 2.78e-06
        [cell_reuse_recompute_pass]: 9.5999e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.164e-05
        [j_node_and_user_rematch]: 9.67001e-06
        [meta_fg_expand]: 2.02001e-06
        [replace_old_param]: 8.26002e-06
        [inline_without_move]: 5.74e-06
        [renormalize]: 0.00051026
        [add_forward_monad_depend]: 4.64998e-06
        [auto_monad_grad]: 1.43002e-06
        [auto_monad_eliminator]: 1.121e-05
        [cse]: 1.751e-05
        [replace_applicator]: 1.138e-05
    [Cycle 2]: 0.00031035, [27]
        [switch_simplify]: 5.38002e-06
        [loop_unroll]: 4.84e-06
        [a_1]: 9.004e-05
        [with_stream_mark]: 8.70999e-06
        [recompute_prepare]: 5.14998e-06
        [updatestate_depend_eliminate]: 2.49999e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 4.80001e-06
        [updatestate_useless_node_eliminater]: 4.73001e-06
        [accelerated_algorithm]: 4.85999e-06
        [meta_shard_fg_expand]: 1.20001e-06
        [get_grad_eliminate_]: 4.53999e-06
        [merge_forward]: 2.48998e-06
        [cell_reuse_recompute_pass]: 1.34998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.21e-05
        [j_node_and_user_rematch]: 7.68001e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 6.84001e-06
        [inline_without_move]: 5.08002e-06
        [renormalize]: 1.09983e-07
        [add_forward_monad_depend]: 1.05999e-06
        [auto_monad_grad]: 8.79983e-07
        [auto_monad_eliminator]: 4.52e-06
        [cse]: 9.25999e-06
        [replace_applicator]: 5.44998e-06
[py_interpret_to_execute_after_opt_a]: 8.29002e-06
[rewriter_after_opt_a]: 2.549e-05
[convert_after_rewriter]: 6.02001e-06
[order_py_execute_after_rewriter]: 4.50001e-06
[mutable_eliminate]: 0.00050714
[jit_opt_b]: 4.793e-05, [1]
    [Cycle 1]: 4.111e-05, [2]
        [frontend_op_eliminate]: 1.575e-05
        [inline_after_opt_a]: 1.431e-05
[cconv]: 1.843e-05
[loop_unroll]: 0.00045051
[jit_opt_after_cconv]: 0.0001315, [1]
    [Cycle 1]: 0.00012558, [11]
        [c_1]: 2.028e-05
        [parameter_eliminate]: 2.38998e-06
        [updatestate_depend_eliminate]: 5.39e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [cse]: 1.585e-05
        [call_graph_tuple_transform]: 1.675e-05
        [tuple_list_get_item_eliminator]: 5.35999e-06
        [none_parameter_eliminate]: 9.89996e-07
        [renormalize]: 5.89993e-07
        [switch_simplify]: 5.15001e-06
[remove_dup_value]: 9.12001e-06
[partial_unused_args_eliminate]: 1.51002e-06
[environ_conv]: 3.95998e-06
[add_recomputation]: 3.575e-05
[cse_after_recomputation]: 1.976e-05, [1]
    [Cycle 1]: 1.432e-05, [1]
        [cse]: 8.91002e-06
[auto_monad_reorder]: 9.57001e-06
[get_jit_bprop_graph]: 1.60999e-06
[rewriter_after_jit_bprop_graph]: 4.82998e-06
[opt_after_jit_grad]: 0.00044645
[symbol_engine_optimizer]: 7.016e-05, [1]
    [Cycle 1]: 6.428e-05, [6]
        [build]: 2.83e-06
        [elim_shapecalc]: 7.63001e-06
        [elim_not_effective]: 1.197e-05
        [opt_reshape]: 5.49e-06
        [fold_const_symbol]: 8.62e-06
        [renormalize]: 4.50003e-07
[validate]: 2.646e-05
Sums
    bootstrap                            :     0.000293s :  3.71%
    type_inference                       :     0.004301s : 54.43%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000060s :  0.77%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.42%
    rewriter_before_opt_a                :     0.000072s :  0.91%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000050s :  0.63%
    jit_opt_a.loop_unroll                :     0.000033s :  0.42%
    jit_opt_a.a_1                        :     0.000569s :  7.20%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000010s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000510s :  6.46%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.20%
    jit_opt_a.cse                        :     0.000027s :  0.34%
    jit_opt_a.replace_applicator         :     0.000017s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.10%
    rewriter_after_opt_a                 :     0.000025s :  0.32%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000507s :  6.42%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.18%
    cconv                                :     0.000018s :  0.23%
    loop_unroll                          :     0.000451s :  5.70%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000009s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000036s :  0.45%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000010s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000446s :  5.65%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000026s :  0.33%


Time group info:
------[substitution.]   0.000162    23
 0.89% :     0.000001s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.52% :     0.000004s :      3: substitution.graph_param_transform
83.87% :     0.000136s :      4: substitution.inline
 1.99% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.58% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.86% :     0.000003s :      2: substitution.replace_old_param
 5.63% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004241     2
80.06% :     0.003396s :      1: type_inference.infer
19.94% :     0.000846s :      1: type_inference.specialize
------[replace.]   0.000060     6
64.91% :     0.000039s :      4: replace.inline
35.09% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000141     6
94.31% :     0.000133s :      4: match.inline
 5.69% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.58% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.29% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.89% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.09% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.55% :     0.000001s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 5.22% :     0.000006s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.54% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.00% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.53% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000002s :     15: predicate.partial_eliminate
 1.29% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.85% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.57% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.51% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.12% :     0.000008s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.29% :     0.000001s :     10: predicate.transpose_eliminate
 1.28% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.68% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000589    12
47.97% :     0.000282s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.08% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.95% :     0.000276s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012929    72
 0.30% :     0.000039s :      1: add_recomputation
 0.49% :     0.000064s :      1: auto_monad
 0.09% :     0.000012s :      1: auto_monad_reorder
 2.36% :     0.000305s :      1: bootstrap
 0.16% :     0.000021s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000006s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
37.32% :     0.004825s :      1: jit_opt_a
 1.04% :     0.000134s :      1: jit_opt_after_cconv
 0.39% :     0.000050s :      1: jit_opt_b
 3.55% :     0.000458s :      1: loop_unroll
 3.99% :     0.000515s :      1: mutable_eliminate
 5.91% :     0.000765s :     26: opt.transform.jit_opt_a
 0.34% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000024s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.51% :     0.000454s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000003s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.28% :     0.000036s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000012s :      1: remove_dup_value
 1.98% :     0.000256s :      1: renormalize.infer
 1.91% :     0.000247s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000028s :      1: rewriter_after_opt_a
 0.58% :     0.000075s :      1: rewriter_before_opt_a
 0.56% :     0.000073s :      1: symbol_engine_optimizer
33.35% :     0.004312s :      1: type_inference


TotalTime = 0.0306569, [33]
[bootstrap]: 0.00017077
[type_inference]: 0.00969262
[event_method]: 4.581e-05
[auto_monad]: 9.058e-05
[graph_reusing]: 6.63e-06
[pre_auto_parallel]: 2.91e-06
[py_interpret_to_execute]: 2.44e-05
[rewriter_before_opt_a]: 9.935e-05
[expand_dump_flag]: 2.94999e-06
[jit_opt_a]: 0.0116051, [3]
    [Cycle 1]: 0.00576872, [27]
        [switch_simplify]: 6.48e-05
        [loop_unroll]: 3.938e-05
        [a_1]: 0.00084741
        [with_stream_mark]: 3.03e-05
        [recompute_prepare]: 1.907e-05
        [updatestate_depend_eliminate]: 8.59002e-06
        [updatestate_assign_eliminate]: 6.88e-06
        [updatestate_loads_eliminate]: 6.49001e-06
        [parameter_eliminate]: 3.58999e-06
        [specialize_transform]: 1.364e-05
        [updatestate_useless_node_eliminater]: 1.184e-05
        [accelerated_algorithm]: 1.202e-05
        [meta_shard_fg_expand]: 4.09002e-06
        [get_grad_eliminate_]: 1.153e-05
        [merge_forward]: 8.59e-06
        [cell_reuse_recompute_pass]: 1.09998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.649e-05
        [j_node_and_user_rematch]: 2.433e-05
        [meta_fg_expand]: 0.00151451
        [replace_old_param]: 5.814e-05
        [inline_without_move]: 5.477e-05
        [renormalize]: 0.00248839
        [add_forward_monad_depend]: 1.212e-05
        [auto_monad_grad]: 6.33e-06
        [auto_monad_eliminator]: 4.868e-05
        [cse]: 0.00015343
        [replace_applicator]: 7.114e-05
    [Cycle 2]: 0.00202556, [27]
        [switch_simplify]: 3.598e-05
        [loop_unroll]: 3.442e-05
        [a_1]: 0.00106244
        [with_stream_mark]: 1.519e-05
        [recompute_prepare]: 7.92e-06
        [updatestate_depend_eliminate]: 4.43001e-06
        [updatestate_assign_eliminate]: 4.02e-06
        [updatestate_loads_eliminate]: 3.19001e-06
        [parameter_eliminate]: 1.76e-06
        [specialize_transform]: 7.46001e-06
        [updatestate_useless_node_eliminater]: 6.31e-06
        [accelerated_algorithm]: 6.25002e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 6.73e-06
        [merge_forward]: 4.08999e-06
        [cell_reuse_recompute_pass]: 1.30001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.696e-05
        [j_node_and_user_rematch]: 1.08e-05
        [meta_fg_expand]: 4.433e-05
        [replace_old_param]: 1.139e-05
        [inline_without_move]: 7.29001e-06
        [renormalize]: 0.00051183
        [add_forward_monad_depend]: 5.10999e-06
        [auto_monad_grad]: 1.87001e-06
        [auto_monad_eliminator]: 1.449e-05
        [cse]: 2.494e-05
        [replace_applicator]: 1.586e-05
    [Cycle 3]: 0.00038691, [27]
        [switch_simplify]: 8.28001e-06
        [loop_unroll]: 6.77002e-06
        [a_1]: 0.00012353
        [with_stream_mark]: 1.024e-05
        [recompute_prepare]: 6.31998e-06
        [updatestate_depend_eliminate]: 3.86001e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 7.03e-06
        [updatestate_useless_node_eliminater]: 6.81001e-06
        [accelerated_algorithm]: 5.62999e-06
        [meta_shard_fg_expand]: 1.24e-06
        [get_grad_eliminate_]: 5.54e-06
        [merge_forward]: 3.17002e-06
        [cell_reuse_recompute_pass]: 2.34999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.634e-05
        [j_node_and_user_rematch]: 1.071e-05
        [meta_fg_expand]: 2.22001e-06
        [replace_old_param]: 7.93999e-06
        [inline_without_move]: 5.84999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.66998e-06
        [auto_monad_grad]: 9.89996e-07
        [auto_monad_eliminator]: 6.83e-06
        [cse]: 1.475e-05
        [replace_applicator]: 6.39999e-06
[py_interpret_to_execute_after_opt_a]: 1.216e-05
[rewriter_after_opt_a]: 3.768e-05
[convert_after_rewriter]: 7.56999e-06
[order_py_execute_after_rewriter]: 6.36998e-06
[mutable_eliminate]: 0.00065091
[jit_opt_b]: 5.731e-05, [1]
    [Cycle 1]: 4.96e-05, [2]
        [frontend_op_eliminate]: 1.955e-05
        [inline_after_opt_a]: 1.807e-05
[cconv]: 2.375e-05
[loop_unroll]: 0.00043
[jit_opt_after_cconv]: 0.00015665, [1]
    [Cycle 1]: 0.0001495, [11]
        [c_1]: 2.582e-05
        [parameter_eliminate]: 2.64001e-06
        [updatestate_depend_eliminate]: 6.49001e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.65997e-06
        [cse]: 2.421e-05
        [call_graph_tuple_transform]: 2.069e-05
        [tuple_list_get_item_eliminator]: 6.25002e-06
        [none_parameter_eliminate]: 1.40999e-06
        [renormalize]: 8.09989e-07
        [switch_simplify]: 5.95002e-06
[remove_dup_value]: 2.228e-05
[partial_unused_args_eliminate]: 2.24999e-06
[environ_conv]: 5.49e-06
[add_recomputation]: 5.071e-05
[cse_after_recomputation]: 2.799e-05, [1]
    [Cycle 1]: 2.113e-05, [1]
        [cse]: 1.505e-05
[auto_monad_reorder]: 1.806e-05
[get_jit_bprop_graph]: 1.39998e-06
[rewriter_after_jit_bprop_graph]: 5.35999e-06
[opt_after_jit_grad]: 0.00045707
[symbol_engine_optimizer]: 7.959e-05, [1]
    [Cycle 1]: 7.341e-05, [6]
        [build]: 3.66001e-06
        [elim_shapecalc]: 9.00001e-06
        [elim_not_effective]: 1.572e-05
        [opt_reshape]: 6.63e-06
        [fold_const_symbol]: 1.096e-05
        [renormalize]: 4.19997e-07
[validate]: 3.538e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.00664578
[execute]: 5.86998e-06
Sums
    bootstrap                            :     0.000171s :  0.65%
    type_inference                       :     0.009693s : 36.66%
    event_method                         :     0.000046s :  0.17%
    auto_monad                           :     0.000091s :  0.34%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000024s :  0.09%
    rewriter_before_opt_a                :     0.000099s :  0.38%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000109s :  0.41%
    jit_opt_a.loop_unroll                :     0.000081s :  0.30%
    jit_opt_a.a_1                        :     0.002033s :  7.69%
    jit_opt_a.with_stream_mark           :     0.000056s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000014s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.02%
    jit_opt_a.specialize_transform       :     0.000028s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.09%
    jit_opt_a.merge_forward              :     0.000016s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000060s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000046s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001561s :  5.90%
    jit_opt_a.replace_old_param          :     0.000077s :  0.29%
    jit_opt_a.inline_without_move        :     0.000068s :  0.26%
    jit_opt_a.renormalize                :     0.003000s : 11.35%
    jit_opt_a.add_forward_monad_depend   :     0.000019s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000070s :  0.26%
    jit_opt_a.cse                        :     0.000193s :  0.73%
    jit_opt_a.replace_applicator         :     0.000093s :  0.35%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.05%
    rewriter_after_opt_a                 :     0.000038s :  0.14%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000651s :  2.46%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.07%
    cconv                                :     0.000024s :  0.09%
    loop_unroll                          :     0.000430s :  1.63%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000022s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000051s :  0.19%
    cse_after_recomputation.cse          :     0.000015s :  0.06%
    auto_monad_reorder                   :     0.000018s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000457s :  1.73%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006646s : 25.13%
    execute                              :     0.000006s :  0.02%


Time group info:
------[substitution.]   0.000510   110
 0.55% :     0.000003s :      3: substitution.elim_not_effective
 0.38% :     0.000002s :      3: substitution.fold_const_symbol
 1.17% :     0.000006s :      4: substitution.graph_param_transform
63.35% :     0.000323s :     12: substitution.inline
 3.75% :     0.000019s :      2: substitution.inline_without_move
 1.86% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.32% :     0.000012s :      7: substitution.minmaximum_grad
 1.45% :     0.000007s :      5: substitution.partial_eliminate
 2.42% :     0.000012s :     14: substitution.remove_not_recompute_node
 5.57% :     0.000028s :     10: substitution.replace_applicator
 1.52% :     0.000008s :      7: substitution.replace_old_param
 0.47% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.33% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.82% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 8.03% :     0.000041s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009627     2
88.40% :     0.008511s :      1: type_inference.infer
11.60% :     0.001116s :      1: type_inference.specialize
------[replace.]   0.000144    19
63.67% :     0.000092s :     12: replace.inline
36.33% :     0.000052s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000337    19
93.95% :     0.000316s :     12: match.inline
 6.05% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000375  2568
 1.57% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.45% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.41% :     0.000005s :     43: predicate.addn_check_dump
 1.76% :     0.000007s :     43: predicate.addn_zero_filter
 1.98% :     0.000007s :     43: predicate.arithmetic_simplify
 1.61% :     0.000006s :     43: predicate.cast_eliminate
 0.18% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.41% :     0.000005s :     43: predicate.compare_switch_simplify
 1.57% :     0.000006s :     43: predicate.depend_value_elim
 1.39% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.61% :     0.000006s :     43: predicate.dict_get_item_eliminator
 1.53% :     0.000006s :     43: predicate.dict_set_item_eliminator
 0.31% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      4: predicate.elim_not_effective
 0.22% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000005s :     43: predicate.environ_add_const_eliminate
 1.41% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.46% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.48% :     0.000006s :     43: predicate.environ_get_eliminate
 1.44% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.83% :     0.000003s :     19: predicate.get_grad_eliminate
 0.09% :     0.000000s :      4: predicate.graph_param_transform
 4.38% :     0.000016s :     70: predicate.inline
 1.93% :     0.000007s :     40: predicate.inline_without_move
 0.34% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.93% :     0.000003s :     19: predicate.less_batch_normalization
 1.77% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.85% :     0.000007s :     54: predicate.load_eliminater
 0.54% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.46% :     0.000013s :     85: predicate.loop_unroll_before_grad
 1.79% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.42% :     0.000005s :     43: predicate.merge_addn
 1.47% :     0.000006s :     43: predicate.minmaximum_grad
 0.50% :     0.000002s :      4: predicate.mutable_eliminate
 0.18% :     0.000001s :      4: predicate.opt_reshape
 2.32% :     0.000009s :     54: predicate.partial_eliminate
 1.58% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.93% :     0.000007s :     43: predicate.reduce_eliminate
 1.74% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.57% :     0.000010s :     96: predicate.replace_applicator
 1.05% :     0.000004s :     40: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.78% :     0.000007s :     43: predicate.reshape_eliminate
 1.58% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.28% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.63% :     0.000006s :     43: predicate.same_eliminate
 0.46% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.44% :     0.000002s :      8: predicate.special_op_eliminate
 0.91% :     0.000003s :     19: predicate.specialize_transform
 1.77% :     0.000007s :     43: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000005s :     43: predicate.stack_unstack_eliminate
 0.18% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.83% :     0.000011s :     62: predicate.switch_defer_inline
 2.43% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.30% :     0.000024s :    151: predicate.switch_simplify
 1.52% :     0.000006s :     43: predicate.tile_eliminate
 1.48% :     0.000006s :     43: predicate.transpose_eliminate
 1.90% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.67% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.03% :     0.000011s :     58: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000007s :     43: predicate.tuple_list_set_item_eliminator
 1.93% :     0.000007s :     50: predicate.tuple_to_list_eliminator_
 1.86% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000011s :     73: predicate.updatestate_useless_node_eliminater
 1.90% :     0.000007s :     43: predicate.value_based_eliminate
 0.14% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.21% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001385    29
58.75% :     0.000814s :     13: func_graph_cloner_run.FuncGraphClonerGraph
41.25% :     0.000571s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.036440    91
 0.15% :     0.000054s :      1: add_recomputation
 0.26% :     0.000094s :      1: auto_monad
 0.06% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.50% :     0.000181s :      1: bootstrap
 0.07% :     0.000026s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000030s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.14% :     0.000050s :      1: event_method
 0.02% :     0.000009s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
31.86% :     0.011608s :      1: jit_opt_a
 0.44% :     0.000159s :      1: jit_opt_after_cconv
 0.17% :     0.000060s :      1: jit_opt_b
 1.20% :     0.000438s :      1: loop_unroll
 1.81% :     0.000659s :      1: mutable_eliminate
 7.28% :     0.002653s :     39: opt.transform.jit_opt_a
 0.15% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000031s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.28% :     0.000465s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000027s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000025s :      1: remove_dup_value
 4.43% :     0.001616s :      2: renormalize.infer
 3.76% :     0.001369s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000041s :      1: rewriter_after_opt_a
 0.28% :     0.000103s :      1: rewriter_before_opt_a
 0.23% :     0.000082s :      1: symbol_engine_optimizer
18.26% :     0.006654s :      1: task_emit
26.63% :     0.009704s :      1: type_inference
 0.15% :     0.000055s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.255.237 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc41040> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.255.763 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0162864, [33]
[bootstrap]: 0.00029462
[type_inference]: 0.00417525
[event_method]: 1.596e-05
[auto_monad]: 4.454e-05
[graph_reusing]: 4.46002e-06
[pre_auto_parallel]: 1.79e-06
[py_interpret_to_execute]: 2.912e-05
[rewriter_before_opt_a]: 6.268e-05
[expand_dump_flag]: 2.31998e-06
[jit_opt_a]: 0.00473298, [2]
    [Cycle 1]: 0.00134617, [27]
        [switch_simplify]: 4.145e-05
        [loop_unroll]: 2.808e-05
        [a_1]: 0.00049782
        [with_stream_mark]: 1.23e-05
        [recompute_prepare]: 7.29001e-06
        [updatestate_depend_eliminate]: 2.83998e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 5.67001e-06
        [updatestate_useless_node_eliminater]: 4.91002e-06
        [accelerated_algorithm]: 5.24e-06
        [meta_shard_fg_expand]: 1.50001e-06
        [get_grad_eliminate_]: 4.95999e-06
        [merge_forward]: 2.75002e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.133e-05
        [j_node_and_user_rematch]: 8.32e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 7.34002e-06
        [inline_without_move]: 5.10999e-06
        [renormalize]: 0.00048868
        [add_forward_monad_depend]: 5.00999e-06
        [auto_monad_grad]: 1.97999e-06
        [auto_monad_eliminator]: 9.70002e-06
        [cse]: 1.597e-05
        [replace_applicator]: 1.274e-05
    [Cycle 2]: 0.00030644, [27]
        [switch_simplify]: 5.44e-06
        [loop_unroll]: 4.88001e-06
        [a_1]: 8.728e-05
        [with_stream_mark]: 8.08999e-06
        [recompute_prepare]: 5.02999e-06
        [updatestate_depend_eliminate]: 2.36e-06
        [updatestate_assign_eliminate]: 1.91e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 8.50006e-07
        [specialize_transform]: 4.71002e-06
        [updatestate_useless_node_eliminater]: 4.86002e-06
        [accelerated_algorithm]: 4.73001e-06
        [meta_shard_fg_expand]: 1.08001e-06
        [get_grad_eliminate_]: 4.25e-06
        [merge_forward]: 2.34999e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.151e-05
        [j_node_and_user_rematch]: 7.36001e-06
        [meta_fg_expand]: 1.45001e-06
        [replace_old_param]: 6.64001e-06
        [inline_without_move]: 4.57e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 6.50005e-07
        [auto_monad_eliminator]: 4.80001e-06
        [cse]: 9.02e-06
        [replace_applicator]: 4.69998e-06
[py_interpret_to_execute_after_opt_a]: 9.12999e-06
[rewriter_after_opt_a]: 2.599e-05
[convert_after_rewriter]: 5.87001e-06
[order_py_execute_after_rewriter]: 4.29997e-06
[mutable_eliminate]: 0.00048464
[jit_opt_b]: 4.79e-05, [1]
    [Cycle 1]: 4.165e-05, [2]
        [frontend_op_eliminate]: 1.582e-05
        [inline_after_opt_a]: 1.457e-05
[cconv]: 1.757e-05
[loop_unroll]: 0.00041079
[jit_opt_after_cconv]: 0.00013153, [1]
    [Cycle 1]: 0.00012516, [11]
        [c_1]: 1.98e-05
        [parameter_eliminate]: 2.61999e-06
        [updatestate_depend_eliminate]: 5.04e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [cse]: 1.602e-05
        [call_graph_tuple_transform]: 1.658e-05
        [tuple_list_get_item_eliminator]: 5.29998e-06
        [none_parameter_eliminate]: 1.12e-06
        [renormalize]: 1.03001e-06
        [switch_simplify]: 5.35001e-06
[remove_dup_value]: 9.20999e-06
[partial_unused_args_eliminate]: 1.64e-06
[environ_conv]: 3.76001e-06
[add_recomputation]: 3.448e-05
[cse_after_recomputation]: 1.865e-05, [1]
    [Cycle 1]: 1.371e-05, [1]
        [cse]: 8.23001e-06
[auto_monad_reorder]: 1.005e-05
[get_jit_bprop_graph]: 1.92001e-06
[rewriter_after_jit_bprop_graph]: 3.85e-06
[opt_after_jit_grad]: 0.00044243
[symbol_engine_optimizer]: 6.887e-05, [1]
    [Cycle 1]: 6.254e-05, [6]
        [build]: 2.64001e-06
        [elim_shapecalc]: 7.24001e-06
        [elim_not_effective]: 1.153e-05
        [opt_reshape]: 5.57999e-06
        [fold_const_symbol]: 8.20999e-06
        [renormalize]: 5.59987e-07
[validate]: 2.659e-05
[backend_pass]: 8.29983e-07
[task_emit]: 0.00499558
[execute]: 5.62001e-06
Sums
    bootstrap                            :     0.000295s :  2.33%
    type_inference                       :     0.004175s : 32.98%
    event_method                         :     0.000016s :  0.13%
    auto_monad                           :     0.000045s :  0.35%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000029s :  0.23%
    rewriter_before_opt_a                :     0.000063s :  0.50%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000047s :  0.37%
    jit_opt_a.loop_unroll                :     0.000033s :  0.26%
    jit_opt_a.a_1                        :     0.000585s :  4.62%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000489s :  3.86%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000025s :  0.20%
    jit_opt_a.replace_applicator         :     0.000017s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000026s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000485s :  3.83%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000018s :  0.14%
    loop_unroll                          :     0.000411s :  3.25%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000009s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000034s :  0.27%
    cse_after_recomputation.cse          :     0.000008s :  0.07%
    auto_monad_reorder                   :     0.000010s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000442s :  3.50%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000027s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.004996s : 39.46%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000159    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.84% :     0.000001s :      2: substitution.fold_const_symbol
 2.51% :     0.000004s :      3: substitution.graph_param_transform
83.95% :     0.000134s :      4: substitution.inline
 1.59% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.44% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.59% :     0.000003s :      2: substitution.replace_old_param
 6.05% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004125     2
80.90% :     0.003337s :      1: type_inference.infer
19.10% :     0.000788s :      1: type_inference.specialize
------[replace.]   0.000058     6
65.80% :     0.000038s :      4: replace.inline
34.20% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000140     6
93.92% :     0.000131s :      4: match.inline
 6.08% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.30% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.52% :     0.000002s :     10: predicate.addn_zero_filter
 1.96% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000001s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.13% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.36% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.35% :     0.000000s :      3: predicate.graph_param_transform
 5.39% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.93% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.34% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.87% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.60% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000002s :     15: predicate.partial_eliminate
 1.42% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.74% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.57% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.22% :     0.000008s :     53: predicate.switch_simplify
 1.28% :     0.000001s :     10: predicate.tile_eliminate
 1.60% :     0.000002s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.93% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.00% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000562    12
42.82% :     0.000241s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.50% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
52.68% :     0.000296s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017650    76
 0.21% :     0.000037s :      1: add_recomputation
 0.27% :     0.000048s :      1: auto_monad
 0.07% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.71% :     0.000303s :      1: bootstrap
 0.11% :     0.000020s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.83% :     0.004736s :      1: jit_opt_a
 0.76% :     0.000134s :      1: jit_opt_after_cconv
 0.29% :     0.000050s :      1: jit_opt_b
 2.37% :     0.000418s :      1: loop_unroll
 2.79% :     0.000493s :      1: mutable_eliminate
 4.37% :     0.000771s :     26: opt.transform.jit_opt_a
 0.25% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.55% :     0.000450s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000032s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000012s :      1: remove_dup_value
 1.41% :     0.000248s :      1: renormalize.infer
 1.32% :     0.000234s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000029s :      1: rewriter_after_opt_a
 0.37% :     0.000065s :      1: rewriter_before_opt_a
 0.40% :     0.000071s :      1: symbol_engine_optimizer
28.35% :     0.005004s :      1: task_emit
23.71% :     0.004185s :      1: type_inference
 0.24% :     0.000042s :      1: validate


TotalTime = 0.0115544, [30]
[bootstrap]: 0.00023476
[type_inference]: 0.00384171
[event_method]: 1.659e-05
[auto_monad]: 6.123e-05
[graph_reusing]: 5.15001e-06
[pre_auto_parallel]: 2.36998e-06
[py_interpret_to_execute]: 3.252e-05
[rewriter_before_opt_a]: 6.836e-05
[expand_dump_flag]: 2.79999e-06
[jit_opt_a]: 0.00522801, [2]
    [Cycle 1]: 0.00157916, [27]
        [switch_simplify]: 5.386e-05
        [loop_unroll]: 3.084e-05
        [a_1]: 0.00056977
        [with_stream_mark]: 1.6e-05
        [recompute_prepare]: 7.92998e-06
        [updatestate_depend_eliminate]: 3.46001e-06
        [updatestate_assign_eliminate]: 3.64002e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [parameter_eliminate]: 2.09e-06
        [specialize_transform]: 6.60002e-06
        [updatestate_useless_node_eliminater]: 5.37999e-06
        [accelerated_algorithm]: 6.39001e-06
        [meta_shard_fg_expand]: 2.10002e-06
        [get_grad_eliminate_]: 5.96e-06
        [merge_forward]: 4.03001e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.57e-05
        [j_node_and_user_rematch]: 1.046e-05
        [meta_fg_expand]: 2.30002e-06
        [replace_old_param]: 1.017e-05
        [inline_without_move]: 6.16e-06
        [renormalize]: 0.00058069
        [add_forward_monad_depend]: 5.13002e-06
        [auto_monad_grad]: 2.33002e-06
        [auto_monad_eliminator]: 1.357e-05
        [cse]: 2.766e-05
        [replace_applicator]: 1.05e-05
    [Cycle 2]: 0.0003306, [27]
        [switch_simplify]: 5.79999e-06
        [loop_unroll]: 5.06997e-06
        [a_1]: 0.00010003
        [with_stream_mark]: 8.52e-06
        [recompute_prepare]: 5.47001e-06
        [updatestate_depend_eliminate]: 2.61e-06
        [updatestate_assign_eliminate]: 2.19999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.52999e-06
        [specialize_transform]: 4.97999e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 1.14003e-06
        [get_grad_eliminate_]: 4.80999e-06
        [merge_forward]: 2.32001e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.233e-05
        [j_node_and_user_rematch]: 8.13001e-06
        [meta_fg_expand]: 1.97999e-06
        [replace_old_param]: 7.66999e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.08001e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 4.90999e-06
        [cse]: 1.079e-05
        [replace_applicator]: 4.80001e-06
[py_interpret_to_execute_after_opt_a]: 9.04e-06
[rewriter_after_opt_a]: 2.947e-05
[convert_after_rewriter]: 7.87e-06
[order_py_execute_after_rewriter]: 5.00999e-06
[mutable_eliminate]: 0.0004909
[jit_opt_b]: 5e-05, [1]
    [Cycle 1]: 4.319e-05, [2]
        [frontend_op_eliminate]: 1.662e-05
        [inline_after_opt_a]: 1.48e-05
[cconv]: 2.428e-05
[loop_unroll]: 0.00043039
[jit_opt_after_cconv]: 0.00013917, [1]
    [Cycle 1]: 0.00013171, [11]
        [c_1]: 2.202e-05
        [parameter_eliminate]: 2.54999e-06
        [updatestate_depend_eliminate]: 4.95999e-06
        [updatestate_assign_eliminate]: 2.36998e-06
        [updatestate_loads_eliminate]: 2.03002e-06
        [cse]: 1.646e-05
        [call_graph_tuple_transform]: 1.833e-05
        [tuple_list_get_item_eliminator]: 5.24e-06
        [none_parameter_eliminate]: 1.29e-06
        [renormalize]: 9.50007e-07
        [switch_simplify]: 5.59e-06
[remove_dup_value]: 1.575e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 5.14e-06
[add_recomputation]: 4.586e-05
[cse_after_recomputation]: 1.974e-05, [1]
    [Cycle 1]: 1.449e-05, [1]
        [cse]: 8.97e-06
[auto_monad_reorder]: 1.414e-05
[get_jit_bprop_graph]: 1.37999e-06
[rewriter_after_jit_bprop_graph]: 5.23002e-06
[opt_after_jit_grad]: 0.00050646
[symbol_engine_optimizer]: 7.254e-05, [1]
    [Cycle 1]: 6.584e-05, [6]
        [build]: 3.19001e-06
        [elim_shapecalc]: 7.45e-06
        [elim_not_effective]: 1.248e-05
        [opt_reshape]: 5.72999e-06
        [fold_const_symbol]: 8.64e-06
        [renormalize]: 4.40021e-07
[validate]: 3.086e-05
Sums
    bootstrap                            :     0.000235s :  3.06%
    type_inference                       :     0.003842s : 50.07%
    event_method                         :     0.000017s :  0.22%
    auto_monad                           :     0.000061s :  0.80%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.42%
    rewriter_before_opt_a                :     0.000068s :  0.89%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000060s :  0.78%
    jit_opt_a.loop_unroll                :     0.000036s :  0.47%
    jit_opt_a.a_1                        :     0.000670s :  8.73%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.05%
    jit_opt_a.specialize_transform       :     0.000012s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.14%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.37%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.24%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.06%
    jit_opt_a.replace_old_param          :     0.000018s :  0.23%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000581s :  7.57%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.24%
    jit_opt_a.cse                        :     0.000038s :  0.50%
    jit_opt_a.replace_applicator         :     0.000015s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.12%
    rewriter_after_opt_a                 :     0.000029s :  0.38%
    convert_after_rewriter               :     0.000008s :  0.10%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000491s :  6.40%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000024s :  0.32%
    loop_unroll                          :     0.000430s :  5.61%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.29%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000016s :  0.21%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000046s :  0.60%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000014s :  0.18%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000506s :  6.60%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000031s :  0.40%


Time group info:
------[substitution.]   0.000211    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.40% :     0.000005s :      3: substitution.graph_param_transform
84.84% :     0.000179s :      4: substitution.inline
 1.53% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.19% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000004s :      2: substitution.replace_old_param
 5.74% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003788     2
78.49% :     0.002973s :      1: type_inference.infer
21.51% :     0.000815s :      1: type_inference.specialize
------[replace.]   0.000074     6
61.47% :     0.000045s :      4: replace.inline
38.53% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000188     6
94.15% :     0.000177s :      4: match.inline
 5.85% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.32% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.39% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.25% :     0.000002s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 1.99% :     0.000003s :     10: predicate.arithmetic_simplify
 1.27% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.53% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.28% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.10% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.35% :     0.000007s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000002s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.62% :     0.000002s :     15: predicate.load_eliminater
 1.56% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.83% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.22% :     0.000002s :     10: predicate.merge_addn
 1.20% :     0.000002s :     10: predicate.minmaximum_grad
 1.46% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.34% :     0.000003s :     15: predicate.partial_eliminate
 1.41% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.89% :     0.000002s :     10: predicate.reduce_eliminate
 1.69% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.51% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.59% :     0.000002s :     10: predicate.reshape_eliminate
 1.36% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.58% :     0.000003s :     16: predicate.switch_defer_inline
 2.40% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.23% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000002s :     10: predicate.tile_eliminate
 1.19% :     0.000002s :     10: predicate.transpose_eliminate
 1.26% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.32% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000525    12
46.10% :     0.000242s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.36% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.54% :     0.000255s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013085    72
 0.37% :     0.000049s :      1: add_recomputation
 0.49% :     0.000064s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 1.87% :     0.000245s :      1: bootstrap
 0.20% :     0.000027s :      1: cconv
 0.08% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
39.98% :     0.005231s :      1: jit_opt_a
 1.08% :     0.000142s :      1: jit_opt_after_cconv
 0.40% :     0.000053s :      1: jit_opt_b
 3.35% :     0.000439s :      1: loop_unroll
 3.82% :     0.000499s :      1: mutable_eliminate
 6.76% :     0.000885s :     26: opt.transform.jit_opt_a
 0.36% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000025s :      4: opt.transform.jit_opt_b
 0.10% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.93% :     0.000515s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000035s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.14% :     0.000018s :      1: remove_dup_value
 2.15% :     0.000282s :      1: renormalize.infer
 2.23% :     0.000291s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000032s :      1: rewriter_after_opt_a
 0.54% :     0.000071s :      1: rewriter_before_opt_a
 0.57% :     0.000075s :      1: symbol_engine_optimizer
29.44% :     0.003853s :      1: type_inference


TotalTime = 0.0180985, [33]
[bootstrap]: 0.00030591
[type_inference]: 0.00437709
[event_method]: 1.655e-05
[auto_monad]: 6.08e-05
[graph_reusing]: 5.58002e-06
[pre_auto_parallel]: 2.17999e-06
[py_interpret_to_execute]: 3.495e-05
[rewriter_before_opt_a]: 6.771e-05
[expand_dump_flag]: 2.89999e-06
[jit_opt_a]: 0.00511406, [2]
    [Cycle 1]: 0.00154303, [27]
        [switch_simplify]: 4.727e-05
        [loop_unroll]: 2.833e-05
        [a_1]: 0.00049883
        [with_stream_mark]: 1.533e-05
        [recompute_prepare]: 6.61999e-06
        [updatestate_depend_eliminate]: 2.81999e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 5.42001e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 5.32001e-06
        [meta_shard_fg_expand]: 1.84e-06
        [get_grad_eliminate_]: 6.652e-05
        [merge_forward]: 4.22e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.338e-05
        [j_node_and_user_rematch]: 8.85001e-06
        [meta_fg_expand]: 2.18002e-06
        [replace_old_param]: 8.05e-06
        [inline_without_move]: 5.30999e-06
        [renormalize]: 0.00058909
        [add_forward_monad_depend]: 5.88002e-06
        [auto_monad_grad]: 2.01e-06
        [auto_monad_eliminator]: 1.337e-05
        [cse]: 1.955e-05
        [replace_applicator]: 1.215e-05
    [Cycle 2]: 0.00032573, [27]
        [switch_simplify]: 5.07e-06
        [loop_unroll]: 4.94998e-06
        [a_1]: 9.486e-05
        [with_stream_mark]: 9.42999e-06
        [recompute_prepare]: 4.90999e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 1.97999e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 9.49978e-07
        [specialize_transform]: 5.09998e-06
        [updatestate_useless_node_eliminater]: 4.71002e-06
        [accelerated_algorithm]: 5.29998e-06
        [meta_shard_fg_expand]: 1.18001e-06
        [get_grad_eliminate_]: 4.84998e-06
        [merge_forward]: 2.91e-06
        [cell_reuse_recompute_pass]: 1.65001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.276e-05
        [j_node_and_user_rematch]: 8.35001e-06
        [meta_fg_expand]: 1.55001e-06
        [replace_old_param]: 6.89001e-06
        [inline_without_move]: 4.48999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.28002e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 5.34e-06
        [cse]: 1.078e-05
        [replace_applicator]: 4.60001e-06
[py_interpret_to_execute_after_opt_a]: 1.014e-05
[rewriter_after_opt_a]: 3.272e-05
[convert_after_rewriter]: 6.86001e-06
[order_py_execute_after_rewriter]: 4.77e-06
[mutable_eliminate]: 0.00059987
[jit_opt_b]: 4.789e-05, [1]
    [Cycle 1]: 4.14e-05, [2]
        [frontend_op_eliminate]: 1.568e-05
        [inline_after_opt_a]: 1.434e-05
[cconv]: 2.424e-05
[loop_unroll]: 0.00043627
[jit_opt_after_cconv]: 0.00013873, [1]
    [Cycle 1]: 0.00013202, [11]
        [c_1]: 2.079e-05
        [parameter_eliminate]: 3.19001e-06
        [updatestate_depend_eliminate]: 5.67001e-06
        [updatestate_assign_eliminate]: 2.49999e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.936e-05
        [call_graph_tuple_transform]: 1.748e-05
        [tuple_list_get_item_eliminator]: 4.87e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 3.80009e-07
        [switch_simplify]: 5.06002e-06
[remove_dup_value]: 1.104e-05
[partial_unused_args_eliminate]: 2.04e-06
[environ_conv]: 4.63999e-06
[add_recomputation]: 4.481e-05
[cse_after_recomputation]: 2.128e-05, [1]
    [Cycle 1]: 1.548e-05, [1]
        [cse]: 9.35001e-06
[auto_monad_reorder]: 1.328e-05
[get_jit_bprop_graph]: 1.52999e-06
[rewriter_after_jit_bprop_graph]: 7.64002e-06
[opt_after_jit_grad]: 0.00058083
[symbol_engine_optimizer]: 7.674e-05, [1]
    [Cycle 1]: 6.908e-05, [6]
        [build]: 4.27e-06
        [elim_shapecalc]: 7.03e-06
        [elim_not_effective]: 1.32e-05
        [opt_reshape]: 5.36998e-06
        [fold_const_symbol]: 8.92999e-06
        [renormalize]: 7.2e-07
[validate]: 4.313e-05
[backend_pass]: 8.80013e-07
[task_emit]: 0.00578927
[execute]: 7.96001e-06
Sums
    bootstrap                            :     0.000306s :  2.15%
    type_inference                       :     0.004377s : 30.74%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000061s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.25%
    rewriter_before_opt_a                :     0.000068s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000052s :  0.37%
    jit_opt_a.loop_unroll                :     0.000033s :  0.23%
    jit_opt_a.a_1                        :     0.000594s :  4.17%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000071s :  0.50%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000589s :  4.14%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000030s :  0.21%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000600s :  4.21%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000436s :  3.06%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000011s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000045s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000013s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000581s :  4.08%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.30%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005789s : 40.66%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000183    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000001s :      2: substitution.fold_const_symbol
 2.56% :     0.000005s :      3: substitution.graph_param_transform
83.81% :     0.000153s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.45% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.73% :     0.000003s :      2: substitution.replace_old_param
 6.13% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004318     2
80.89% :     0.003493s :      1: type_inference.infer
19.11% :     0.000825s :      1: type_inference.specialize
------[replace.]   0.000066     6
66.12% :     0.000044s :      4: replace.inline
33.88% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000161     6
93.75% :     0.000151s :      4: match.inline
 6.25% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.83% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.43% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 1.94% :     0.000002s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.70% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000001s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.40% :     0.000002s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.59% :     0.000002s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.15% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      6: predicate.less_batch_normalization
 1.78% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.93% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.68% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000002s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.48% :     0.000001s :      3: predicate.reset_defer_inline
 1.62% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.25% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.78% :     0.000001s :      6: predicate.special_op_eliminate
 1.22% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.25% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.64% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.26% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.39% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.77% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000569    12
45.50% :     0.000259s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.28% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.22% :     0.000280s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019642    76
 0.24% :     0.000048s :      1: add_recomputation
 0.33% :     0.000064s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.63% :     0.000320s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.07% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.05% :     0.005118s :      1: jit_opt_a
 0.72% :     0.000141s :      1: jit_opt_after_cconv
 0.26% :     0.000050s :      1: jit_opt_b
 2.27% :     0.000445s :      1: loop_unroll
 3.10% :     0.000610s :      1: mutable_eliminate
 4.35% :     0.000854s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.01% :     0.000590s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000038s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000013s :      1: remove_dup_value
 1.60% :     0.000314s :      1: renormalize.infer
 1.36% :     0.000268s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000036s :      1: rewriter_after_opt_a
 0.36% :     0.000070s :      1: rewriter_before_opt_a
 0.40% :     0.000079s :      1: symbol_engine_optimizer
29.55% :     0.005805s :      1: task_emit
22.34% :     0.004388s :      1: type_inference
 0.34% :     0.000066s :      1: validate


TotalTime = 0.0167485, [33]
[bootstrap]: 0.00028708
[type_inference]: 0.00421431
[event_method]: 1.587e-05
[auto_monad]: 4.611e-05
[graph_reusing]: 3.76001e-06
[pre_auto_parallel]: 1.72001e-06
[py_interpret_to_execute]: 2.945e-05
[rewriter_before_opt_a]: 6.333e-05
[expand_dump_flag]: 2.09e-06
[jit_opt_a]: 0.00484659, [2]
    [Cycle 1]: 0.00127071, [27]
        [switch_simplify]: 4.299e-05
        [loop_unroll]: 2.803e-05
        [a_1]: 0.00046042
        [with_stream_mark]: 1.139e-05
        [recompute_prepare]: 6.76999e-06
        [updatestate_depend_eliminate]: 2.86e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [parameter_eliminate]: 1.16002e-06
        [specialize_transform]: 5.69e-06
        [updatestate_useless_node_eliminater]: 5.22999e-06
        [accelerated_algorithm]: 5.34e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.90001e-06
        [merge_forward]: 2.47001e-06
        [cell_reuse_recompute_pass]: 8.60018e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.086e-05
        [j_node_and_user_rematch]: 8.79e-06
        [meta_fg_expand]: 1.81998e-06
        [replace_old_param]: 7.48999e-06
        [inline_without_move]: 5.14e-06
        [renormalize]: 0.0004544
        [add_forward_monad_depend]: 4.79e-06
        [auto_monad_grad]: 1.15999e-06
        [auto_monad_eliminator]: 9.41e-06
        [cse]: 1.57e-05
        [replace_applicator]: 1.129e-05
    [Cycle 2]: 0.00034163, [27]
        [switch_simplify]: 5.22e-06
        [loop_unroll]: 4.78001e-06
        [a_1]: 8.731e-05
        [with_stream_mark]: 3.462e-05
        [recompute_prepare]: 6.06e-06
        [updatestate_depend_eliminate]: 2.74999e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [parameter_eliminate]: 1.32e-06
        [specialize_transform]: 5.05001e-06
        [updatestate_useless_node_eliminater]: 4.65999e-06
        [accelerated_algorithm]: 4.99998e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.50999e-06
        [merge_forward]: 2.42001e-06
        [cell_reuse_recompute_pass]: 1.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.152e-05
        [j_node_and_user_rematch]: 7.23e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 6.94999e-06
        [inline_without_move]: 4.55001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.35999e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 5.44998e-06
        [cse]: 1.082e-05
        [replace_applicator]: 4.77e-06
[py_interpret_to_execute_after_opt_a]: 7.76001e-06
[rewriter_after_opt_a]: 2.675e-05
[convert_after_rewriter]: 5.64e-06
[order_py_execute_after_rewriter]: 4.84e-06
[mutable_eliminate]: 0.00046424
[jit_opt_b]: 4.623e-05, [1]
    [Cycle 1]: 3.992e-05, [2]
        [frontend_op_eliminate]: 1.495e-05
        [inline_after_opt_a]: 1.4e-05
[cconv]: 1.812e-05
[loop_unroll]: 0.0004299
[jit_opt_after_cconv]: 0.00012931, [1]
    [Cycle 1]: 0.00012325, [11]
        [c_1]: 1.997e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 5.05001e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 1.544e-05
        [call_graph_tuple_transform]: 1.622e-05
        [tuple_list_get_item_eliminator]: 5.27001e-06
        [none_parameter_eliminate]: 1.07e-06
        [renormalize]: 5.49975e-07
        [switch_simplify]: 5.02e-06
[remove_dup_value]: 9.07999e-06
[partial_unused_args_eliminate]: 1.89e-06
[environ_conv]: 3.9e-06
[add_recomputation]: 3.574e-05
[cse_after_recomputation]: 1.825e-05, [1]
    [Cycle 1]: 1.338e-05, [1]
        [cse]: 8.12003e-06
[auto_monad_reorder]: 1.158e-05
[get_jit_bprop_graph]: 1.64e-06
[rewriter_after_jit_bprop_graph]: 4.43999e-06
[opt_after_jit_grad]: 0.0004524
[symbol_engine_optimizer]: 6.856e-05, [1]
    [Cycle 1]: 6.235e-05, [6]
        [build]: 2.89001e-06
        [elim_shapecalc]: 7.09001e-06
        [elim_not_effective]: 1.192e-05
        [opt_reshape]: 5.29e-06
        [fold_const_symbol]: 8.17e-06
        [renormalize]: 4.39992e-07
[validate]: 2.869e-05
[backend_pass]: 7.7e-07
[task_emit]: 0.00529476
[execute]: 6.51e-06
Sums
    bootstrap                            :     0.000287s :  2.22%
    type_inference                       :     0.004214s : 32.52%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000046s :  0.36%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000029s :  0.23%
    rewriter_before_opt_a                :     0.000063s :  0.49%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000048s :  0.37%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000548s :  4.23%
    jit_opt_a.with_stream_mark           :     0.000046s :  0.36%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000454s :  3.51%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000027s :  0.20%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000027s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000464s :  3.58%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000018s :  0.14%
    loop_unroll                          :     0.000430s :  3.32%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000009s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000036s :  0.28%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000452s :  3.49%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005295s : 40.85%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000146    23
 1.26% :     0.000002s :      2: substitution.elim_not_effective
 0.87% :     0.000001s :      2: substitution.fold_const_symbol
 2.69% :     0.000004s :      3: substitution.graph_param_transform
83.25% :     0.000122s :      4: substitution.inline
 1.89% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.80% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.15% :     0.000003s :      2: substitution.replace_old_param
 5.09% :     0.000007s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004165     2
82.08% :     0.003419s :      1: type_inference.infer
17.92% :     0.000746s :      1: type_inference.specialize
------[replace.]   0.000057     6
66.60% :     0.000038s :      4: replace.inline
33.40% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000126     6
94.92% :     0.000119s :      4: match.inline
 5.08% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.60% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.51% :     0.000002s :     10: predicate.addn_zero_filter
 1.91% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.31% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.18% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.18% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.17% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000000s :      3: predicate.opt_reshape
 2.12% :     0.000002s :     15: predicate.partial_eliminate
 1.63% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.82% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.75% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.44% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.27% :     0.000008s :     53: predicate.switch_simplify
 1.30% :     0.000001s :     10: predicate.tile_eliminate
 1.35% :     0.000002s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.29% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000487    12
49.37% :     0.000240s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.24% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.38% :     0.000221s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018043    76
 0.21% :     0.000039s :      1: add_recomputation
 0.27% :     0.000049s :      1: auto_monad
 0.08% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.65% :     0.000298s :      1: bootstrap
 0.12% :     0.000021s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000020s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
26.88% :     0.004849s :      1: jit_opt_a
 0.73% :     0.000132s :      1: jit_opt_after_cconv
 0.27% :     0.000049s :      1: jit_opt_b
 2.42% :     0.000437s :      1: loop_unroll
 2.62% :     0.000472s :      1: mutable_eliminate
 4.09% :     0.000738s :     26: opt.transform.jit_opt_a
 0.24% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.55% :     0.000460s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000011s :      1: remove_dup_value
 1.21% :     0.000219s :      1: renormalize.infer
 1.27% :     0.000229s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000029s :      1: rewriter_after_opt_a
 0.37% :     0.000066s :      1: rewriter_before_opt_a
 0.39% :     0.000071s :      1: symbol_engine_optimizer
29.40% :     0.005305s :      1: task_emit
23.41% :     0.004224s :      1: type_inference
 0.25% :     0.000044s :      1: validate


TotalTime = 0.0281892, [33]
[bootstrap]: 0.00020626
[type_inference]: 0.00965295
[event_method]: 4.459e-05
[auto_monad]: 7.711e-05
[graph_reusing]: 5.57001e-06
[pre_auto_parallel]: 2.51998e-06
[py_interpret_to_execute]: 2.478e-05
[rewriter_before_opt_a]: 8.533e-05
[expand_dump_flag]: 2.54001e-06
[jit_opt_a]: 0.0104627, [3]
    [Cycle 1]: 0.00516486, [27]
        [switch_simplify]: 5.981e-05
        [loop_unroll]: 4.04e-05
        [a_1]: 0.00083083
        [with_stream_mark]: 2.038e-05
        [recompute_prepare]: 1.828e-05
        [updatestate_depend_eliminate]: 7.38e-06
        [updatestate_assign_eliminate]: 5.67001e-06
        [updatestate_loads_eliminate]: 5.24998e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 1.279e-05
        [updatestate_useless_node_eliminater]: 1.177e-05
        [accelerated_algorithm]: 1.185e-05
        [meta_shard_fg_expand]: 2.56e-06
        [get_grad_eliminate_]: 1.165e-05
        [merge_forward]: 7.07997e-06
        [cell_reuse_recompute_pass]: 7.39994e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.306e-05
        [j_node_and_user_rematch]: 2.141e-05
        [meta_fg_expand]: 0.00136007
        [replace_old_param]: 4.895e-05
        [inline_without_move]: 5.038e-05
        [renormalize]: 0.00216349
        [add_forward_monad_depend]: 8.1e-06
        [auto_monad_grad]: 4.15999e-06
        [auto_monad_eliminator]: 4.501e-05
        [cse]: 0.0001375
        [replace_applicator]: 6.153e-05
    [Cycle 2]: 0.00184237, [27]
        [switch_simplify]: 3.473e-05
        [loop_unroll]: 3.29e-05
        [a_1]: 0.00101514
        [with_stream_mark]: 1.097e-05
        [recompute_prepare]: 7.60998e-06
        [updatestate_depend_eliminate]: 3.45e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.38998e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 6.32001e-06
        [updatestate_useless_node_eliminater]: 5.92001e-06
        [accelerated_algorithm]: 5.66e-06
        [meta_shard_fg_expand]: 1.64998e-06
        [get_grad_eliminate_]: 5.85002e-06
        [merge_forward]: 3.13e-06
        [cell_reuse_recompute_pass]: 7.60017e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.252e-05
        [j_node_and_user_rematch]: 9.53002e-06
        [meta_fg_expand]: 3.243e-05
        [replace_old_param]: 9.09e-06
        [inline_without_move]: 6.15002e-06
        [renormalize]: 0.00044222
        [add_forward_monad_depend]: 4.05e-06
        [auto_monad_grad]: 1.14003e-06
        [auto_monad_eliminator]: 1.111e-05
        [cse]: 1.891e-05
        [replace_applicator]: 1.138e-05
    [Cycle 3]: 0.00035998, [27]
        [switch_simplify]: 6.33e-06
        [loop_unroll]: 5.89e-06
        [a_1]: 0.00011686
        [with_stream_mark]: 8.64e-06
        [recompute_prepare]: 5.97001e-06
        [updatestate_depend_eliminate]: 3.28e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.48998e-06
        [parameter_eliminate]: 1.25999e-06
        [specialize_transform]: 6.11998e-06
        [updatestate_useless_node_eliminater]: 5.61e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 5.71e-06
        [merge_forward]: 2.79999e-06
        [cell_reuse_recompute_pass]: 1.64998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.434e-05
        [j_node_and_user_rematch]: 9.52999e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 8.1e-06
        [inline_without_move]: 5.57999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 7.29982e-07
        [auto_monad_eliminator]: 6.61e-06
        [cse]: 1.346e-05
        [replace_applicator]: 6.14999e-06
[py_interpret_to_execute_after_opt_a]: 9.00001e-06
[rewriter_after_opt_a]: 2.877e-05
[convert_after_rewriter]: 6.33998e-06
[order_py_execute_after_rewriter]: 4.52e-06
[mutable_eliminate]: 0.00049502
[jit_opt_b]: 5.344e-05, [1]
    [Cycle 1]: 4.684e-05, [2]
        [frontend_op_eliminate]: 1.802e-05
        [inline_after_opt_a]: 1.73e-05
[cconv]: 1.512e-05
[loop_unroll]: 0.00043214
[jit_opt_after_cconv]: 0.00014722, [1]
    [Cycle 1]: 0.00014034, [11]
        [c_1]: 2.484e-05
        [parameter_eliminate]: 2.32001e-06
        [updatestate_depend_eliminate]: 6.29001e-06
        [updatestate_assign_eliminate]: 2.67001e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [cse]: 2.186e-05
        [call_graph_tuple_transform]: 1.803e-05
        [tuple_list_get_item_eliminator]: 6.01e-06
        [none_parameter_eliminate]: 8.79983e-07
        [renormalize]: 4.19997e-07
        [switch_simplify]: 6.02999e-06
[remove_dup_value]: 9.56e-06
[partial_unused_args_eliminate]: 1.47999e-06
[environ_conv]: 4.38001e-06
[add_recomputation]: 4.368e-05
[cse_after_recomputation]: 2.742e-05, [1]
    [Cycle 1]: 2.119e-05, [1]
        [cse]: 1.515e-05
[auto_monad_reorder]: 1.306e-05
[get_jit_bprop_graph]: 1.28002e-06
[rewriter_after_jit_bprop_graph]: 4.80001e-06
[opt_after_jit_grad]: 0.00045659
[symbol_engine_optimizer]: 7.459e-05, [1]
    [Cycle 1]: 6.851e-05, [6]
        [build]: 2.76e-06
        [elim_shapecalc]: 8.46002e-06
        [elim_not_effective]: 1.374e-05
        [opt_reshape]: 6.33e-06
        [fold_const_symbol]: 9.87001e-06
        [renormalize]: 5.19998e-07
[validate]: 2.796e-05
[backend_pass]: 6.79982e-07
[task_emit]: 0.00558329
[execute]: 4.43999e-06
Sums
    bootstrap                            :     0.000206s :  0.85%
    type_inference                       :     0.009653s : 39.64%
    event_method                         :     0.000045s :  0.18%
    auto_monad                           :     0.000077s :  0.32%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000025s :  0.10%
    rewriter_before_opt_a                :     0.000085s :  0.35%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000101s :  0.41%
    jit_opt_a.loop_unroll                :     0.000079s :  0.33%
    jit_opt_a.a_1                        :     0.001963s :  8.06%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000025s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000023s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.10%
    jit_opt_a.merge_forward              :     0.000013s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000050s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000040s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001394s :  5.73%
    jit_opt_a.replace_old_param          :     0.000066s :  0.27%
    jit_opt_a.inline_without_move        :     0.000062s :  0.26%
    jit_opt_a.renormalize                :     0.002606s : 10.70%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000006s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000063s :  0.26%
    jit_opt_a.cse                        :     0.000170s :  0.70%
    jit_opt_a.replace_applicator         :     0.000079s :  0.32%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.04%
    rewriter_after_opt_a                 :     0.000029s :  0.12%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000495s :  2.03%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.07%
    cconv                                :     0.000015s :  0.06%
    loop_unroll                          :     0.000432s :  1.77%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000010s :  0.04%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000044s :  0.18%
    cse_after_recomputation.cse          :     0.000015s :  0.06%
    auto_monad_reorder                   :     0.000013s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000457s :  1.88%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000028s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005583s : 22.93%
    execute                              :     0.000004s :  0.02%


Time group info:
------[substitution.]   0.000423   110
 0.38% :     0.000002s :      3: substitution.elim_not_effective
 0.29% :     0.000001s :      3: substitution.fold_const_symbol
 0.92% :     0.000004s :      4: substitution.graph_param_transform
62.91% :     0.000266s :     12: substitution.inline
 3.70% :     0.000016s :      2: substitution.inline_without_move
 1.85% :     0.000008s :     14: substitution.j_node_and_user_rematch
 2.29% :     0.000010s :      7: substitution.minmaximum_grad
 1.46% :     0.000006s :      5: substitution.partial_eliminate
 2.36% :     0.000010s :     14: substitution.remove_not_recompute_node
 5.72% :     0.000024s :     10: substitution.replace_applicator
 1.64% :     0.000007s :      7: substitution.replace_old_param
 0.58% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.75% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.10% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 8.05% :     0.000034s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009589     2
88.70% :     0.008505s :      1: type_inference.infer
11.30% :     0.001084s :      1: type_inference.specialize
------[replace.]   0.000138    19
63.03% :     0.000087s :     12: replace.inline
36.97% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000277    19
93.80% :     0.000260s :     12: match.inline
 6.20% :     0.000017s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000363  2568
 1.65% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.52% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.48% :     0.000005s :     43: predicate.addn_check_dump
 1.71% :     0.000006s :     43: predicate.addn_zero_filter
 1.95% :     0.000007s :     43: predicate.arithmetic_simplify
 1.52% :     0.000005s :     43: predicate.cast_eliminate
 0.17% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.46% :     0.000005s :     43: predicate.compare_switch_simplify
 1.68% :     0.000006s :     43: predicate.depend_value_elim
 1.61% :     0.000006s :     43: predicate.dict_get_item_const_eliminator
 1.49% :     0.000005s :     43: predicate.dict_get_item_eliminator
 1.44% :     0.000005s :     43: predicate.dict_set_item_eliminator
 0.33% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.14% :     0.000000s :      4: predicate.elim_not_effective
 0.19% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.50% :     0.000005s :     43: predicate.environ_add_const_eliminate
 1.45% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.49% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.54% :     0.000006s :     43: predicate.environ_get_eliminate
 1.46% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.88% :     0.000003s :     19: predicate.get_grad_eliminate
 0.11% :     0.000000s :      4: predicate.graph_param_transform
 4.32% :     0.000016s :     70: predicate.inline
 1.84% :     0.000007s :     40: predicate.inline_without_move
 0.36% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.80% :     0.000003s :     19: predicate.less_batch_normalization
 1.92% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.95% :     0.000007s :     54: predicate.load_eliminater
 0.51% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.22% :     0.000012s :     85: predicate.loop_unroll_before_grad
 1.78% :     0.000006s :     47: predicate.make_slice_get_slice_eliminator
 1.51% :     0.000005s :     43: predicate.merge_addn
 1.57% :     0.000006s :     43: predicate.minmaximum_grad
 0.59% :     0.000002s :      4: predicate.mutable_eliminate
 0.19% :     0.000001s :      4: predicate.opt_reshape
 2.37% :     0.000009s :     54: predicate.partial_eliminate
 1.53% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.85% :     0.000007s :     43: predicate.reduce_eliminate
 1.85% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.49% :     0.000009s :     96: predicate.replace_applicator
 0.89% :     0.000003s :     40: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.78% :     0.000006s :     43: predicate.reshape_eliminate
 1.63% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.27% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.52% :     0.000006s :     43: predicate.same_eliminate
 0.46% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.34% :     0.000001s :      8: predicate.special_op_eliminate
 0.90% :     0.000003s :     19: predicate.specialize_transform
 1.70% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000005s :     43: predicate.stack_unstack_eliminate
 0.18% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.70% :     0.000010s :     62: predicate.switch_defer_inline
 2.52% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.25% :     0.000023s :    151: predicate.switch_simplify
 1.49% :     0.000005s :     43: predicate.tile_eliminate
 1.50% :     0.000005s :     43: predicate.transpose_eliminate
 1.82% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.71% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 2.90% :     0.000011s :     58: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000007s :     43: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000006s :     50: predicate.tuple_to_list_eliminator_
 1.98% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000011s :     73: predicate.updatestate_useless_node_eliminater
 1.87% :     0.000007s :     43: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.29% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001343    29
59.43% :     0.000798s :     13: func_graph_cloner_run.FuncGraphClonerGraph
40.57% :     0.000545s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.033447    91
 0.14% :     0.000047s :      1: add_recomputation
 0.24% :     0.000081s :      1: auto_monad
 0.05% :     0.000016s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.66% :     0.000219s :      1: bootstrap
 0.05% :     0.000018s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000030s :      1: cse_after_recomputation
 0.02% :     0.000006s :      1: environ_conv
 0.15% :     0.000049s :      1: event_method
 0.02% :     0.000008s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000003s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
31.29% :     0.010465s :      1: jit_opt_a
 0.45% :     0.000150s :      1: jit_opt_after_cconv
 0.17% :     0.000056s :      1: jit_opt_b
 1.32% :     0.000440s :      1: loop_unroll
 1.50% :     0.000503s :      1: mutable_eliminate
 7.56% :     0.002529s :     39: opt.transform.jit_opt_a
 0.15% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000029s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.39% :     0.000464s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000028s :      1: py_interpret_to_execute
 0.03% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000012s :      1: remove_dup_value
 4.06% :     0.001358s :      2: renormalize.infer
 3.70% :     0.001237s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000032s :      1: rewriter_after_opt_a
 0.26% :     0.000088s :      1: rewriter_before_opt_a
 0.23% :     0.000077s :      1: symbol_engine_optimizer
16.72% :     0.005593s :      1: task_emit
28.90% :     0.009665s :      1: type_inference
 0.14% :     0.000046s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.290.709 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc41550> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.291.288 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0123088, [30]
[bootstrap]: 0.00026425
[type_inference]: 0.00439998
[event_method]: 1.772e-05
[auto_monad]: 6.222e-05
[graph_reusing]: 6.17999e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 3.389e-05
[rewriter_before_opt_a]: 9.451e-05
[expand_dump_flag]: 3.09001e-06
[jit_opt_a]: 0.00543097, [2]
    [Cycle 1]: 0.00162839, [27]
        [switch_simplify]: 5.578e-05
        [loop_unroll]: 3.094e-05
        [a_1]: 0.0005678
        [with_stream_mark]: 1.637e-05
        [recompute_prepare]: 7.40003e-06
        [updatestate_depend_eliminate]: 3.53999e-06
        [updatestate_assign_eliminate]: 3.04001e-06
        [updatestate_loads_eliminate]: 2.74999e-06
        [parameter_eliminate]: 1.81003e-06
        [specialize_transform]: 6.19001e-06
        [updatestate_useless_node_eliminater]: 5.32999e-06
        [accelerated_algorithm]: 5.52001e-06
        [meta_shard_fg_expand]: 2.26998e-06
        [get_grad_eliminate_]: 5.59998e-06
        [merge_forward]: 3.60003e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.412e-05
        [j_node_and_user_rematch]: 9.49e-06
        [meta_fg_expand]: 2.59999e-06
        [replace_old_param]: 8.53001e-06
        [inline_without_move]: 5.56998e-06
        [renormalize]: 0.00064377
        [add_forward_monad_depend]: 4.82e-06
        [auto_monad_grad]: 2.99001e-06
        [auto_monad_eliminator]: 1.375e-05
        [cse]: 2.638e-05
        [replace_applicator]: 1.052e-05
    [Cycle 2]: 0.00032059, [27]
        [switch_simplify]: 5.53002e-06
        [loop_unroll]: 5.12999e-06
        [a_1]: 9.199e-05
        [with_stream_mark]: 8.98002e-06
        [recompute_prepare]: 5.03002e-06
        [updatestate_depend_eliminate]: 2.54001e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.04998e-06
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 5.12e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.58001e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.259e-05
        [j_node_and_user_rematch]: 7.71999e-06
        [meta_fg_expand]: 1.72001e-06
        [replace_old_param]: 6.86999e-06
        [inline_without_move]: 4.70001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 4.84e-06
        [cse]: 1.018e-05
        [replace_applicator]: 5.00001e-06
[py_interpret_to_execute_after_opt_a]: 9.52999e-06
[rewriter_after_opt_a]: 3.059e-05
[convert_after_rewriter]: 6.78e-06
[order_py_execute_after_rewriter]: 5.55001e-06
[mutable_eliminate]: 0.00049539
[jit_opt_b]: 4.964e-05, [1]
    [Cycle 1]: 4.326e-05, [2]
        [frontend_op_eliminate]: 1.655e-05
        [inline_after_opt_a]: 1.511e-05
[cconv]: 2.339e-05
[loop_unroll]: 0.00041319
[jit_opt_after_cconv]: 0.00013582, [1]
    [Cycle 1]: 0.00012949, [11]
        [c_1]: 2.073e-05
        [parameter_eliminate]: 2.30002e-06
        [updatestate_depend_eliminate]: 5.14e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.671e-05
        [call_graph_tuple_transform]: 1.771e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.61998e-06
        [renormalize]: 6.60017e-07
        [switch_simplify]: 5.55001e-06
[remove_dup_value]: 1.421e-05
[partial_unused_args_eliminate]: 2.33002e-06
[environ_conv]: 5.29e-06
[add_recomputation]: 4.555e-05
[cse_after_recomputation]: 1.954e-05, [1]
    [Cycle 1]: 1.397e-05, [1]
        [cse]: 8.54e-06
[auto_monad_reorder]: 1.469e-05
[get_jit_bprop_graph]: 1.55999e-06
[rewriter_after_jit_bprop_graph]: 4.77e-06
[opt_after_jit_grad]: 0.00045173
[symbol_engine_optimizer]: 7.433e-05, [1]
    [Cycle 1]: 6.815e-05, [6]
        [build]: 3.46001e-06
        [elim_shapecalc]: 7.75e-06
        [elim_not_effective]: 1.322e-05
        [opt_reshape]: 5.86e-06
        [fold_const_symbol]: 8.67e-06
        [renormalize]: 4.10015e-07
[validate]: 3.168e-05
Sums
    bootstrap                            :     0.000264s :  3.20%
    type_inference                       :     0.004400s : 53.23%
    event_method                         :     0.000018s :  0.21%
    auto_monad                           :     0.000062s :  0.75%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.41%
    rewriter_before_opt_a                :     0.000095s :  1.14%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000061s :  0.74%
    jit_opt_a.loop_unroll                :     0.000036s :  0.44%
    jit_opt_a.a_1                        :     0.000660s :  7.98%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000644s :  7.79%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.22%
    jit_opt_a.cse                        :     0.000037s :  0.44%
    jit_opt_a.replace_applicator         :     0.000016s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000031s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.07%
    mutable_eliminate                    :     0.000495s :  5.99%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000023s :  0.28%
    loop_unroll                          :     0.000413s :  5.00%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000046s :  0.55%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000452s :  5.46%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.38%


Time group info:
------[substitution.]   0.000197    23
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.69% :     0.000005s :      3: substitution.graph_param_transform
83.46% :     0.000165s :      4: substitution.inline
 1.38% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.71% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.63% :     0.000003s :      2: substitution.replace_old_param
 6.49% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004343     2
78.57% :     0.003412s :      1: type_inference.infer
21.43% :     0.000931s :      1: type_inference.specialize
------[replace.]   0.000072     6
66.80% :     0.000048s :      4: replace.inline
33.20% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000174     6
93.28% :     0.000162s :      4: match.inline
 6.72% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.76% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.38% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.18% :     0.000001s :     10: predicate.addn_zero_filter
 2.19% :     0.000003s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000001s :     10: predicate.depend_value_elim
 1.28% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.17% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.46% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.65% :     0.000007s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.46% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.94% :     0.000002s :     10: predicate.reduce_eliminate
 1.64% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.76% :     0.000002s :     10: predicate.reshape_eliminate
 1.10% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.99% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.46% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.66% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.39% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.24% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.33% :     0.000009s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.04% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.43% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 2.07% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.07% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.02% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000631    12
47.52% :     0.000300s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.75% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.73% :     0.000301s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013884    72
 0.35% :     0.000048s :      1: add_recomputation
 0.47% :     0.000066s :      1: auto_monad
 0.12% :     0.000017s :      1: auto_monad_reorder
 1.98% :     0.000276s :      1: bootstrap
 0.19% :     0.000026s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.15% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
39.14% :     0.005435s :      1: jit_opt_a
 1.00% :     0.000139s :      1: jit_opt_after_cconv
 0.38% :     0.000052s :      1: jit_opt_b
 3.03% :     0.000421s :      1: loop_unroll
 3.62% :     0.000503s :      1: mutable_eliminate
 6.27% :     0.000871s :     26: opt.transform.jit_opt_a
 0.33% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.31% :     0.000460s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000036s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.28% :     0.000316s :      1: renormalize.infer
 2.31% :     0.000320s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000034s :      1: rewriter_after_opt_a
 0.71% :     0.000098s :      1: rewriter_before_opt_a
 0.55% :     0.000077s :      1: symbol_engine_optimizer
31.78% :     0.004412s :      1: type_inference


TotalTime = 0.0186375, [33]
[bootstrap]: 0.00027963
[type_inference]: 0.00455073
[event_method]: 1.714e-05
[auto_monad]: 6.232e-05
[graph_reusing]: 5.69e-06
[pre_auto_parallel]: 2.36e-06
[py_interpret_to_execute]: 3.254e-05
[rewriter_before_opt_a]: 7.2e-05
[expand_dump_flag]: 3.2e-06
[jit_opt_a]: 0.00534198, [2]
    [Cycle 1]: 0.00164322, [27]
        [switch_simplify]: 5.668e-05
        [loop_unroll]: 2.875e-05
        [a_1]: 0.00052608
        [with_stream_mark]: 2.02e-05
        [recompute_prepare]: 7.1e-06
        [updatestate_depend_eliminate]: 3.81001e-06
        [updatestate_assign_eliminate]: 2.82002e-06
        [updatestate_loads_eliminate]: 2.53003e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 5.81e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 1.82001e-06
        [get_grad_eliminate_]: 5.88998e-06
        [merge_forward]: 3.41001e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.432e-05
        [j_node_and_user_rematch]: 9.20999e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 9.46e-06
        [inline_without_move]: 5.77001e-06
        [renormalize]: 0.0006405
        [add_forward_monad_depend]: 5.79e-06
        [auto_monad_grad]: 2.34999e-06
        [auto_monad_eliminator]: 1.24e-05
        [cse]: 2.633e-05
        [replace_applicator]: 1.158e-05
    [Cycle 2]: 0.00030977, [27]
        [switch_simplify]: 5.39e-06
        [loop_unroll]: 4.64998e-06
        [a_1]: 8.708e-05
        [with_stream_mark]: 8.75001e-06
        [recompute_prepare]: 4.75001e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.77999e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 5.07999e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 5.22999e-06
        [meta_shard_fg_expand]: 1.40999e-06
        [get_grad_eliminate_]: 4.44998e-06
        [merge_forward]: 2.45002e-06
        [cell_reuse_recompute_pass]: 1.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.25e-05
        [j_node_and_user_rematch]: 7.74002e-06
        [meta_fg_expand]: 1.49998e-06
        [replace_old_param]: 6.56e-06
        [inline_without_move]: 4.45e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.16002e-06
        [auto_monad_grad]: 8.80013e-07
        [auto_monad_eliminator]: 4.95999e-06
        [cse]: 9.46e-06
        [replace_applicator]: 4.85001e-06
[py_interpret_to_execute_after_opt_a]: 1.125e-05
[rewriter_after_opt_a]: 3.424e-05
[convert_after_rewriter]: 6.89999e-06
[order_py_execute_after_rewriter]: 4.68999e-06
[mutable_eliminate]: 0.00055348
[jit_opt_b]: 4.891e-05, [1]
    [Cycle 1]: 4.114e-05, [2]
        [frontend_op_eliminate]: 1.642e-05
        [inline_after_opt_a]: 1.392e-05
[cconv]: 2.421e-05
[loop_unroll]: 0.00040176
[jit_opt_after_cconv]: 0.00013266, [1]
    [Cycle 1]: 0.000126, [11]
        [c_1]: 1.989e-05
        [parameter_eliminate]: 2.32001e-06
        [updatestate_depend_eliminate]: 4.92999e-06
        [updatestate_assign_eliminate]: 2.31998e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.56e-05
        [call_graph_tuple_transform]: 1.757e-05
        [tuple_list_get_item_eliminator]: 5.15999e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 5.66998e-06
[remove_dup_value]: 1.339e-05
[partial_unused_args_eliminate]: 2.52001e-06
[environ_conv]: 5.18002e-06
[add_recomputation]: 5.019e-05
[cse_after_recomputation]: 2.033e-05, [1]
    [Cycle 1]: 1.503e-05, [1]
        [cse]: 8.60999e-06
[auto_monad_reorder]: 1.426e-05
[get_jit_bprop_graph]: 1.87001e-06
[rewriter_after_jit_bprop_graph]: 5.21002e-06
[opt_after_jit_grad]: 0.00044616
[symbol_engine_optimizer]: 7.149e-05, [1]
    [Cycle 1]: 6.468e-05, [6]
        [build]: 3.57002e-06
        [elim_shapecalc]: 7.7e-06
        [elim_not_effective]: 1.206e-05
        [opt_reshape]: 5.36998e-06
        [fold_const_symbol]: 8.28001e-06
        [renormalize]: 5.19998e-07
[validate]: 3.571e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00616733
[execute]: 9.02999e-06
Sums
    bootstrap                            :     0.000280s :  1.92%
    type_inference                       :     0.004551s : 31.20%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000062s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.22%
    rewriter_before_opt_a                :     0.000072s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.43%
    jit_opt_a.loop_unroll                :     0.000033s :  0.23%
    jit_opt_a.a_1                        :     0.000613s :  4.20%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000641s :  4.39%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.12%
    jit_opt_a.cse                        :     0.000036s :  0.25%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.08%
    rewriter_after_opt_a                 :     0.000034s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000553s :  3.80%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000402s :  2.75%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000050s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000446s :  3.06%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006167s : 42.29%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000204    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.71% :     0.000006s :      3: substitution.graph_param_transform
83.57% :     0.000171s :      4: substitution.inline
 1.69% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.37% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.04% :     0.000004s :      2: substitution.replace_old_param
 5.98% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004489     2
81.63% :     0.003664s :      1: type_inference.infer
18.37% :     0.000825s :      1: type_inference.specialize
------[replace.]   0.000066     6
67.08% :     0.000045s :      4: replace.inline
32.92% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000179     6
93.77% :     0.000168s :      4: match.inline
 6.23% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.70% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.12% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 1.90% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.68% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000001s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.42% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.40% :     0.000002s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.19% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.30% :     0.000001s :      3: predicate.mutable_eliminate
 0.44% :     0.000000s :      3: predicate.opt_reshape
 2.40% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000001s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.14% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.57% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.68% :     0.000008s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.30% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.12% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.87% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000568    12
53.07% :     0.000301s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.97% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.96% :     0.000233s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020192    76
 0.26% :     0.000053s :      1: add_recomputation
 0.32% :     0.000066s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.47% :     0.000296s :      1: bootstrap
 0.13% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.48% :     0.005346s :      1: jit_opt_a
 0.67% :     0.000136s :      1: jit_opt_after_cconv
 0.25% :     0.000051s :      1: jit_opt_b
 2.03% :     0.000409s :      1: loop_unroll
 2.78% :     0.000562s :      1: mutable_eliminate
 4.07% :     0.000821s :     26: opt.transform.jit_opt_a
 0.22% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.25% :     0.000454s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000035s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.69% :     0.000341s :      1: renormalize.infer
 1.45% :     0.000292s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000037s :      1: rewriter_after_opt_a
 0.37% :     0.000075s :      1: rewriter_before_opt_a
 0.37% :     0.000074s :      1: symbol_engine_optimizer
30.64% :     0.006186s :      1: task_emit
22.62% :     0.004567s :      1: type_inference
 0.27% :     0.000055s :      1: validate


TotalTime = 0.0157894, [33]
[bootstrap]: 0.00025815
[type_inference]: 0.00405306
[event_method]: 1.631e-05
[auto_monad]: 5.265e-05
[graph_reusing]: 4.76002e-06
[pre_auto_parallel]: 1.89e-06
[py_interpret_to_execute]: 2.748e-05
[rewriter_before_opt_a]: 6.317e-05
[expand_dump_flag]: 2.68e-06
[jit_opt_a]: 0.00457594, [2]
    [Cycle 1]: 0.00126912, [27]
        [switch_simplify]: 4.054e-05
        [loop_unroll]: 2.769e-05
        [a_1]: 0.00044334
        [with_stream_mark]: 1.091e-05
        [recompute_prepare]: 6.99001e-06
        [updatestate_depend_eliminate]: 2.78003e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [parameter_eliminate]: 9.00007e-07
        [specialize_transform]: 5.64e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.09e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.89e-06
        [merge_forward]: 2.42001e-06
        [cell_reuse_recompute_pass]: 9.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.102e-05
        [j_node_and_user_rematch]: 8.15e-06
        [meta_fg_expand]: 1.82001e-06
        [replace_old_param]: 7.88001e-06
        [inline_without_move]: 4.92e-06
        [renormalize]: 0.00047193
        [add_forward_monad_depend]: 4.70001e-06
        [auto_monad_grad]: 1.45001e-06
        [auto_monad_eliminator]: 1.027e-05
        [cse]: 1.624e-05
        [replace_applicator]: 1.083e-05
    [Cycle 2]: 0.00030697, [27]
        [switch_simplify]: 5.14e-06
        [loop_unroll]: 4.68001e-06
        [a_1]: 8.704e-05
        [with_stream_mark]: 8.27e-06
        [recompute_prepare]: 4.89e-06
        [updatestate_depend_eliminate]: 2.26e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.80001e-06
        [parameter_eliminate]: 9.99979e-07
        [specialize_transform]: 4.68001e-06
        [updatestate_useless_node_eliminater]: 4.64998e-06
        [accelerated_algorithm]: 4.85001e-06
        [meta_shard_fg_expand]: 9.70002e-07
        [get_grad_eliminate_]: 4.54002e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 1.57999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.142e-05
        [j_node_and_user_rematch]: 7.82e-06
        [meta_fg_expand]: 1.57999e-06
        [replace_old_param]: 6.96001e-06
        [inline_without_move]: 4.43999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.11002e-06
        [auto_monad_grad]: 6.60017e-07
        [auto_monad_eliminator]: 4.80999e-06
        [cse]: 1.037e-05
        [replace_applicator]: 4.58999e-06
[py_interpret_to_execute_after_opt_a]: 9.59e-06
[rewriter_after_opt_a]: 2.599e-05
[convert_after_rewriter]: 5.76e-06
[order_py_execute_after_rewriter]: 4.50001e-06
[mutable_eliminate]: 0.00047495
[jit_opt_b]: 4.752e-05, [1]
    [Cycle 1]: 4.119e-05, [2]
        [frontend_op_eliminate]: 1.572e-05
        [inline_after_opt_a]: 1.408e-05
[cconv]: 1.643e-05
[loop_unroll]: 0.00041107
[jit_opt_after_cconv]: 0.00012863, [1]
    [Cycle 1]: 0.00012227, [11]
        [c_1]: 1.955e-05
        [parameter_eliminate]: 2.21e-06
        [updatestate_depend_eliminate]: 4.97e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 1.526e-05
        [call_graph_tuple_transform]: 1.563e-05
        [tuple_list_get_item_eliminator]: 5.27999e-06
        [none_parameter_eliminate]: 1.07e-06
        [renormalize]: 3.19997e-07
        [switch_simplify]: 5.05001e-06
[remove_dup_value]: 8.29998e-06
[partial_unused_args_eliminate]: 1.79e-06
[environ_conv]: 3.26001e-06
[add_recomputation]: 3.447e-05
[cse_after_recomputation]: 1.901e-05, [1]
    [Cycle 1]: 1.384e-05, [1]
        [cse]: 8.37e-06
[auto_monad_reorder]: 9.96998e-06
[get_jit_bprop_graph]: 1.37e-06
[rewriter_after_jit_bprop_graph]: 4.22e-06
[opt_after_jit_grad]: 0.00042986
[symbol_engine_optimizer]: 6.953e-05, [1]
    [Cycle 1]: 6.378e-05, [6]
        [build]: 2.67001e-06
        [elim_shapecalc]: 7.46001e-06
        [elim_not_effective]: 1.245e-05
        [opt_reshape]: 5.39998e-06
        [fold_const_symbol]: 8.25999e-06
        [renormalize]: 5.10016e-07
[validate]: 2.532e-05
[backend_pass]: 7.30011e-07
[task_emit]: 0.00483452
[execute]: 5.02e-06
Sums
    bootstrap                            :     0.000258s :  2.11%
    type_inference                       :     0.004053s : 33.11%
    event_method                         :     0.000016s :  0.13%
    auto_monad                           :     0.000053s :  0.43%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000027s :  0.22%
    rewriter_before_opt_a                :     0.000063s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000046s :  0.37%
    jit_opt_a.loop_unroll                :     0.000032s :  0.26%
    jit_opt_a.a_1                        :     0.000530s :  4.33%
    jit_opt_a.with_stream_mark           :     0.000019s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000009s :  0.08%
    jit_opt_a.renormalize                :     0.000472s :  3.86%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.12%
    jit_opt_a.cse                        :     0.000027s :  0.22%
    jit_opt_a.replace_applicator         :     0.000015s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.08%
    rewriter_after_opt_a                 :     0.000026s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000475s :  3.88%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.12%
    cconv                                :     0.000016s :  0.13%
    loop_unroll                          :     0.000411s :  3.36%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000008s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000003s :  0.03%
    add_recomputation                    :     0.000034s :  0.28%
    cse_after_recomputation.cse          :     0.000008s :  0.07%
    auto_monad_reorder                   :     0.000010s :  0.08%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000430s :  3.51%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000025s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.004835s : 39.49%
    execute                              :     0.000005s :  0.04%


Time group info:
------[substitution.]   0.000137    23
 1.33% :     0.000002s :      2: substitution.elim_not_effective
 0.92% :     0.000001s :      2: substitution.fold_const_symbol
 2.67% :     0.000004s :      3: substitution.graph_param_transform
81.91% :     0.000112s :      4: substitution.inline
 1.84% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.82% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.19% :     0.000003s :      2: substitution.replace_old_param
 6.32% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004002     2
82.27% :     0.003292s :      1: type_inference.infer
17.73% :     0.000710s :      1: type_inference.specialize
------[replace.]   0.000056     6
66.08% :     0.000037s :      4: replace.inline
33.92% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000118     6
93.51% :     0.000110s :      4: match.inline
 6.49% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000108   691
 1.43% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.43% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000001s :     10: predicate.addn_zero_filter
 1.90% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.42% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.53% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.09% :     0.000005s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.79% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.88% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.00% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.73% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000002s :     15: predicate.partial_eliminate
 1.50% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.35% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.48% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000008s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.30% :     0.000001s :     10: predicate.transpose_eliminate
 1.53% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.12% :     0.000003s :     18: predicate.tuple_list_get_item_eliminator
 1.38% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.80% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000462    12
48.39% :     0.000224s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.66% :     0.000022s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.95% :     0.000217s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017082    76
 0.22% :     0.000037s :      1: add_recomputation
 0.33% :     0.000056s :      1: auto_monad
 0.07% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.57% :     0.000269s :      1: bootstrap
 0.11% :     0.000019s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.12% :     0.000020s :      1: event_method
 0.05% :     0.000008s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.80% :     0.004578s :      1: jit_opt_a
 0.77% :     0.000131s :      1: jit_opt_after_cconv
 0.29% :     0.000050s :      1: jit_opt_b
 2.45% :     0.000419s :      1: loop_unroll
 2.83% :     0.000484s :      1: mutable_eliminate
 4.18% :     0.000715s :     26: opt.transform.jit_opt_a
 0.25% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.56% :     0.000437s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000030s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000011s :      1: remove_dup_value
 1.24% :     0.000212s :      1: renormalize.infer
 1.49% :     0.000254s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000029s :      1: rewriter_after_opt_a
 0.39% :     0.000066s :      1: rewriter_before_opt_a
 0.42% :     0.000072s :      1: symbol_engine_optimizer
28.35% :     0.004843s :      1: task_emit
23.78% :     0.004063s :      1: type_inference
 0.23% :     0.000040s :      1: validate


TotalTime = 0.0279391, [33]
[bootstrap]: 0.00019226
[type_inference]: 0.0090342
[event_method]: 3.956e-05
[auto_monad]: 6.891e-05
[graph_reusing]: 4.76002e-06
[pre_auto_parallel]: 2.04e-06
[py_interpret_to_execute]: 2.026e-05
[rewriter_before_opt_a]: 8.043e-05
[expand_dump_flag]: 2.32999e-06
[jit_opt_a]: 0.0101791, [3]
    [Cycle 1]: 0.00500073, [27]
        [switch_simplify]: 5.169e-05
        [loop_unroll]: 3.876e-05
        [a_1]: 0.00077293
        [with_stream_mark]: 1.825e-05
        [recompute_prepare]: 1.72e-05
        [updatestate_depend_eliminate]: 6.78998e-06
        [updatestate_assign_eliminate]: 5.76e-06
        [updatestate_loads_eliminate]: 5.22e-06
        [parameter_eliminate]: 1.70001e-06
        [specialize_transform]: 1.24e-05
        [updatestate_useless_node_eliminater]: 1.166e-05
        [accelerated_algorithm]: 1.161e-05
        [meta_shard_fg_expand]: 2.56e-06
        [get_grad_eliminate_]: 1.171e-05
        [merge_forward]: 7.28e-06
        [cell_reuse_recompute_pass]: 7.60017e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.322e-05
        [j_node_and_user_rematch]: 2.193e-05
        [meta_fg_expand]: 0.00135389
        [replace_old_param]: 4.933e-05
        [inline_without_move]: 4.946e-05
        [renormalize]: 0.00209977
        [add_forward_monad_depend]: 7.68001e-06
        [auto_monad_grad]: 4.13001e-06
        [auto_monad_eliminator]: 4.324e-05
        [cse]: 0.00013409
        [replace_applicator]: 6.184e-05
    [Cycle 2]: 0.00190426, [27]
        [switch_simplify]: 3.447e-05
        [loop_unroll]: 3.405e-05
        [a_1]: 0.00104092
        [with_stream_mark]: 1.108e-05
        [recompute_prepare]: 8.28001e-06
        [updatestate_depend_eliminate]: 3.58e-06
        [updatestate_assign_eliminate]: 3.04999e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 6.49999e-06
        [updatestate_useless_node_eliminater]: 5.79e-06
        [accelerated_algorithm]: 5.86998e-06
        [meta_shard_fg_expand]: 1.50001e-06
        [get_grad_eliminate_]: 5.81e-06
        [merge_forward]: 2.92002e-06
        [cell_reuse_recompute_pass]: 7.7e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.235e-05
        [j_node_and_user_rematch]: 9.53002e-06
        [meta_fg_expand]: 3.14e-05
        [replace_old_param]: 9.61998e-06
        [inline_without_move]: 6.46999e-06
        [renormalize]: 0.00046692
        [add_forward_monad_depend]: 4.42e-06
        [auto_monad_grad]: 1.15001e-06
        [auto_monad_eliminator]: 1.204e-05
        [cse]: 2.107e-05
        [replace_applicator]: 1.418e-05
    [Cycle 3]: 0.00037676, [27]
        [switch_simplify]: 7.76001e-06
        [loop_unroll]: 6.09001e-06
        [a_1]: 0.00012186
        [with_stream_mark]: 9.51e-06
        [recompute_prepare]: 6.80998e-06
        [updatestate_depend_eliminate]: 3.62002e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 6.24001e-06
        [updatestate_useless_node_eliminater]: 6.29001e-06
        [accelerated_algorithm]: 5.69999e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 5.90002e-06
        [merge_forward]: 2.98e-06
        [cell_reuse_recompute_pass]: 2.06e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.554e-05
        [j_node_and_user_rematch]: 1.057e-05
        [meta_fg_expand]: 1.89e-06
        [replace_old_param]: 7.92e-06
        [inline_without_move]: 5.66e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 6.60017e-07
        [auto_monad_eliminator]: 6.80002e-06
        [cse]: 1.354e-05
        [replace_applicator]: 5.94e-06
[py_interpret_to_execute_after_opt_a]: 1.087e-05
[rewriter_after_opt_a]: 3.33e-05
[convert_after_rewriter]: 8.19002e-06
[order_py_execute_after_rewriter]: 5.08002e-06
[mutable_eliminate]: 0.00054478
[jit_opt_b]: 5.342e-05, [1]
    [Cycle 1]: 4.71e-05, [2]
        [frontend_op_eliminate]: 1.85e-05
        [inline_after_opt_a]: 1.726e-05
[cconv]: 1.787e-05
[loop_unroll]: 0.00042828
[jit_opt_after_cconv]: 0.00014986, [1]
    [Cycle 1]: 0.00014387, [11]
        [c_1]: 2.501e-05
        [parameter_eliminate]: 2.87002e-06
        [updatestate_depend_eliminate]: 6.03002e-06
        [updatestate_assign_eliminate]: 3.36999e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [cse]: 2.179e-05
        [call_graph_tuple_transform]: 1.945e-05
        [tuple_list_get_item_eliminator]: 6.49999e-06
        [none_parameter_eliminate]: 1.05001e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 6.17001e-06
[remove_dup_value]: 1.215e-05
[partial_unused_args_eliminate]: 1.64998e-06
[environ_conv]: 4.37e-06
[add_recomputation]: 4.869e-05
[cse_after_recomputation]: 2.601e-05, [1]
    [Cycle 1]: 2.056e-05, [1]
        [cse]: 1.492e-05
[auto_monad_reorder]: 1.579e-05
[get_jit_bprop_graph]: 1.50999e-06
[rewriter_after_jit_bprop_graph]: 5.64e-06
[opt_after_jit_grad]: 0.00046651
[symbol_engine_optimizer]: 7.86e-05, [1]
    [Cycle 1]: 7.261e-05, [6]
        [build]: 3.33e-06
        [elim_shapecalc]: 9.51e-06
        [elim_not_effective]: 1.488e-05
        [opt_reshape]: 6.78e-06
        [fold_const_symbol]: 1.034e-05
        [renormalize]: 4.90021e-07
[validate]: 3.119e-05
[backend_pass]: 1.05001e-06
[task_emit]: 0.00618729
[execute]: 6.91001e-06
Sums
    bootstrap                            :     0.000192s :  0.79%
    type_inference                       :     0.009034s : 37.16%
    event_method                         :     0.000040s :  0.16%
    auto_monad                           :     0.000069s :  0.28%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000020s :  0.08%
    rewriter_before_opt_a                :     0.000080s :  0.33%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000094s :  0.39%
    jit_opt_a.loop_unroll                :     0.000079s :  0.32%
    jit_opt_a.a_1                        :     0.001936s :  7.96%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000025s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000023s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.10%
    jit_opt_a.merge_forward              :     0.000013s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000051s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001387s :  5.71%
    jit_opt_a.replace_old_param          :     0.000067s :  0.28%
    jit_opt_a.inline_without_move        :     0.000062s :  0.25%
    jit_opt_a.renormalize                :     0.002567s : 10.56%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000006s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000062s :  0.26%
    jit_opt_a.cse                        :     0.000169s :  0.69%
    jit_opt_a.replace_applicator         :     0.000082s :  0.34%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.04%
    rewriter_after_opt_a                 :     0.000033s :  0.14%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000545s :  2.24%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.07%
    cconv                                :     0.000018s :  0.07%
    loop_unroll                          :     0.000428s :  1.76%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000012s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000049s :  0.20%
    cse_after_recomputation.cse          :     0.000015s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000467s :  1.92%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006187s : 25.45%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000443   110
 0.51% :     0.000002s :      3: substitution.elim_not_effective
 0.39% :     0.000002s :      3: substitution.fold_const_symbol
 1.14% :     0.000005s :      4: substitution.graph_param_transform
57.42% :     0.000254s :     12: substitution.inline
 3.44% :     0.000015s :      2: substitution.inline_without_move
 1.81% :     0.000008s :     14: substitution.j_node_and_user_rematch
 2.18% :     0.000010s :      7: substitution.minmaximum_grad
 1.38% :     0.000006s :      5: substitution.partial_eliminate
 2.42% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.40% :     0.000024s :     10: substitution.replace_applicator
 1.44% :     0.000006s :      7: substitution.replace_old_param
 0.51% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.28% :     0.000019s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.06% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
14.62% :     0.000065s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008983     2
89.69% :     0.008057s :      1: type_inference.infer
10.31% :     0.000926s :      1: type_inference.specialize
------[replace.]   0.000134    19
62.18% :     0.000083s :     12: replace.inline
37.82% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000265    19
93.63% :     0.000248s :     12: match.inline
 6.37% :     0.000017s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000364  2568
 1.61% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.47% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.42% :     0.000005s :     43: predicate.addn_check_dump
 1.61% :     0.000006s :     43: predicate.addn_zero_filter
 2.04% :     0.000007s :     43: predicate.arithmetic_simplify
 1.51% :     0.000005s :     43: predicate.cast_eliminate
 0.17% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.41% :     0.000005s :     43: predicate.compare_switch_simplify
 1.73% :     0.000006s :     43: predicate.depend_value_elim
 1.46% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.56% :     0.000006s :     43: predicate.dict_get_item_eliminator
 1.46% :     0.000005s :     43: predicate.dict_set_item_eliminator
 0.34% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.15% :     0.000001s :      4: predicate.elim_not_effective
 0.22% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.63% :     0.000006s :     43: predicate.environ_add_const_eliminate
 1.46% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.48% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.50% :     0.000005s :     43: predicate.environ_get_eliminate
 1.44% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.88% :     0.000003s :     19: predicate.get_grad_eliminate
 0.12% :     0.000000s :      4: predicate.graph_param_transform
 4.39% :     0.000016s :     70: predicate.inline
 1.77% :     0.000006s :     40: predicate.inline_without_move
 0.37% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.85% :     0.000003s :     19: predicate.less_batch_normalization
 1.77% :     0.000006s :     50: predicate.list_to_tuple_eliminator_
 1.97% :     0.000007s :     54: predicate.load_eliminater
 0.54% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.21% :     0.000012s :     85: predicate.loop_unroll_before_grad
 1.93% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.41% :     0.000005s :     43: predicate.merge_addn
 1.55% :     0.000006s :     43: predicate.minmaximum_grad
 0.50% :     0.000002s :      4: predicate.mutable_eliminate
 0.20% :     0.000001s :      4: predicate.opt_reshape
 2.29% :     0.000008s :     54: predicate.partial_eliminate
 1.52% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.99% :     0.000007s :     43: predicate.reduce_eliminate
 1.91% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.51% :     0.000009s :     96: predicate.replace_applicator
 0.91% :     0.000003s :     40: predicate.replace_old_param
 0.13% :     0.000000s :      4: predicate.reset_defer_inline
 1.69% :     0.000006s :     43: predicate.reshape_eliminate
 1.57% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.28% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.52% :     0.000006s :     43: predicate.same_eliminate
 0.49% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.40% :     0.000001s :      8: predicate.special_op_eliminate
 0.86% :     0.000003s :     19: predicate.specialize_transform
 1.73% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.57% :     0.000006s :     43: predicate.stack_unstack_eliminate
 0.16% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.76% :     0.000010s :     62: predicate.switch_defer_inline
 2.64% :     0.000010s :     62: predicate.switch_layer_defer_inline
 6.14% :     0.000022s :    151: predicate.switch_simplify
 1.55% :     0.000006s :     43: predicate.tile_eliminate
 1.49% :     0.000005s :     43: predicate.transpose_eliminate
 1.85% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.67% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 2.94% :     0.000011s :     58: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000007s :     43: predicate.tuple_list_set_item_eliminator
 1.87% :     0.000007s :     50: predicate.tuple_to_list_eliminator_
 1.93% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000010s :     73: predicate.updatestate_useless_node_eliminater
 1.83% :     0.000007s :     43: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.21% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001284    29
59.89% :     0.000769s :     13: func_graph_cloner_run.FuncGraphClonerGraph
40.11% :     0.000515s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.033136    91
 0.16% :     0.000052s :      1: add_recomputation
 0.22% :     0.000072s :      1: auto_monad
 0.06% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.63% :     0.000207s :      1: bootstrap
 0.06% :     0.000021s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000028s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.13% :     0.000043s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
30.73% :     0.010182s :      1: jit_opt_a
 0.46% :     0.000153s :      1: jit_opt_after_cconv
 0.17% :     0.000056s :      1: jit_opt_b
 1.31% :     0.000436s :      1: loop_unroll
 1.67% :     0.000553s :      1: mutable_eliminate
 7.55% :     0.002502s :     39: opt.transform.jit_opt_a
 0.16% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000030s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.43% :     0.000475s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000023s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000015s :      1: remove_dup_value
 4.05% :     0.001342s :      2: renormalize.infer
 3.66% :     0.001213s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000036s :      1: rewriter_after_opt_a
 0.25% :     0.000083s :      1: rewriter_before_opt_a
 0.24% :     0.000081s :      1: symbol_engine_optimizer
18.70% :     0.006198s :      1: task_emit
27.29% :     0.009042s :      1: type_inference
 0.15% :     0.000049s :      1: validate


TotalTime = 0.0110968, [30]
[bootstrap]: 0.0003188
[type_inference]: 0.00386454
[event_method]: 1.584e-05
[auto_monad]: 5.59e-05
[graph_reusing]: 4.30999e-06
[pre_auto_parallel]: 2.20002e-06
[py_interpret_to_execute]: 2.957e-05
[rewriter_before_opt_a]: 6.514e-05
[expand_dump_flag]: 1.91e-06
[jit_opt_a]: 0.00475872, [2]
    [Cycle 1]: 0.0014321, [27]
        [switch_simplify]: 4.558e-05
        [loop_unroll]: 3.007e-05
        [a_1]: 0.00049933
        [with_stream_mark]: 1.209e-05
        [recompute_prepare]: 7.06001e-06
        [updatestate_depend_eliminate]: 3.00002e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [parameter_eliminate]: 1.23002e-06
        [specialize_transform]: 6.30002e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 5.82001e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 5.50001e-06
        [merge_forward]: 2.73e-06
        [cell_reuse_recompute_pass]: 1.14998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.239e-05
        [j_node_and_user_rematch]: 9.51e-06
        [meta_fg_expand]: 2.06998e-06
        [replace_old_param]: 8.67e-06
        [inline_without_move]: 5.72001e-06
        [renormalize]: 0.00053253
        [add_forward_monad_depend]: 4.45999e-06
        [auto_monad_grad]: 1.67999e-06
        [auto_monad_eliminator]: 9.89001e-06
        [cse]: 1.588e-05
        [replace_applicator]: 1.057e-05
    [Cycle 2]: 0.00031788, [27]
        [switch_simplify]: 5.42999e-06
        [loop_unroll]: 5.03002e-06
        [a_1]: 9.094e-05
        [with_stream_mark]: 7.8e-06
        [recompute_prepare]: 5.07e-06
        [updatestate_depend_eliminate]: 2.45002e-06
        [updatestate_assign_eliminate]: 2.03002e-06
        [updatestate_loads_eliminate]: 1.98002e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 5.35001e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 1.22999e-06
        [get_grad_eliminate_]: 4.63999e-06
        [merge_forward]: 2.22999e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.196e-05
        [j_node_and_user_rematch]: 7.65e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 7.06001e-06
        [inline_without_move]: 4.83001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 4.77e-06
        [cse]: 9.85002e-06
        [replace_applicator]: 5.15001e-06
[py_interpret_to_execute_after_opt_a]: 8.97999e-06
[rewriter_after_opt_a]: 2.553e-05
[convert_after_rewriter]: 5.46002e-06
[order_py_execute_after_rewriter]: 4.83001e-06
[mutable_eliminate]: 0.00049816
[jit_opt_b]: 5.18e-05, [1]
    [Cycle 1]: 4.426e-05, [2]
        [frontend_op_eliminate]: 1.661e-05
        [inline_after_opt_a]: 1.593e-05
[cconv]: 1.961e-05
[loop_unroll]: 0.00041814
[jit_opt_after_cconv]: 0.00013869, [1]
    [Cycle 1]: 0.00013219, [11]
        [c_1]: 2.058e-05
        [parameter_eliminate]: 2.34001e-06
        [updatestate_depend_eliminate]: 5.81998e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.902e-05
        [call_graph_tuple_transform]: 1.695e-05
        [tuple_list_get_item_eliminator]: 5.48002e-06
        [none_parameter_eliminate]: 1.05999e-06
        [renormalize]: 3.29979e-07
        [switch_simplify]: 5.40001e-06
[remove_dup_value]: 1.227e-05
[partial_unused_args_eliminate]: 2.07999e-06
[environ_conv]: 4.12998e-06
[add_recomputation]: 3.689e-05
[cse_after_recomputation]: 2.07e-05, [1]
    [Cycle 1]: 1.532e-05, [1]
        [cse]: 9.15999e-06
[auto_monad_reorder]: 1.214e-05
[get_jit_bprop_graph]: 1.38002e-06
[rewriter_after_jit_bprop_graph]: 5.01002e-06
[opt_after_jit_grad]: 0.00045365
[symbol_engine_optimizer]: 7.441e-05, [1]
    [Cycle 1]: 6.782e-05, [6]
        [build]: 3.85998e-06
        [elim_shapecalc]: 7.43e-06
        [elim_not_effective]: 1.314e-05
        [opt_reshape]: 6.24999e-06
        [fold_const_symbol]: 8.67e-06
        [renormalize]: 4.00003e-07
[validate]: 2.952e-05
Sums
    bootstrap                            :     0.000319s :  4.25%
    type_inference                       :     0.003865s : 51.50%
    event_method                         :     0.000016s :  0.21%
    auto_monad                           :     0.000056s :  0.74%
    graph_reusing                        :     0.000004s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000030s :  0.39%
    rewriter_before_opt_a                :     0.000065s :  0.87%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000051s :  0.68%
    jit_opt_a.loop_unroll                :     0.000035s :  0.47%
    jit_opt_a.a_1                        :     0.000590s :  7.87%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.21%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000533s :  7.10%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.20%
    jit_opt_a.cse                        :     0.000026s :  0.34%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.12%
    rewriter_after_opt_a                 :     0.000026s :  0.34%
    convert_after_rewriter               :     0.000005s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000498s :  6.64%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.21%
    cconv                                :     0.000020s :  0.26%
    loop_unroll                          :     0.000418s :  5.57%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000019s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000012s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000037s :  0.49%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000012s :  0.16%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000454s :  6.05%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.18%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000030s :  0.39%


Time group info:
------[substitution.]   0.000170    23
 1.20% :     0.000002s :      2: substitution.elim_not_effective
 0.79% :     0.000001s :      2: substitution.fold_const_symbol
 2.56% :     0.000004s :      3: substitution.graph_param_transform
83.43% :     0.000141s :      4: substitution.inline
 1.89% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.62% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.98% :     0.000003s :      2: substitution.replace_old_param
 5.52% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003763     2
79.87% :     0.003006s :      1: type_inference.infer
20.13% :     0.000758s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.08% :     0.000042s :      4: replace.inline
32.92% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000147     6
94.40% :     0.000139s :      4: match.inline
 5.60% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.36% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000001s :     10: predicate.addn_check_dump
 1.43% :     0.000002s :     10: predicate.addn_zero_filter
 1.91% :     0.000002s :     10: predicate.arithmetic_simplify
 1.40% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.39% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.70% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.36% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 4.91% :     0.000006s :     22: predicate.inline
 1.04% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.11% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.69% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000002s :     15: predicate.partial_eliminate
 1.24% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.63% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.59% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.47% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.60% :     0.000009s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.40% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.07% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000521    12
46.34% :     0.000241s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.78% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.88% :     0.000250s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012486    72
 0.32% :     0.000040s :      1: add_recomputation
 0.47% :     0.000059s :      1: auto_monad
 0.12% :     0.000015s :      1: auto_monad_reorder
 2.64% :     0.000330s :      1: bootstrap
 0.18% :     0.000022s :      1: cconv
 0.08% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000019s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
38.14% :     0.004762s :      1: jit_opt_a
 1.13% :     0.000141s :      1: jit_opt_after_cconv
 0.44% :     0.000054s :      1: jit_opt_b
 3.41% :     0.000426s :      1: loop_unroll
 4.06% :     0.000506s :      1: mutable_eliminate
 6.33% :     0.000790s :     26: opt.transform.jit_opt_a
 0.36% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.70% :     0.000462s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000032s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000015s :      1: remove_dup_value
 2.09% :     0.000261s :      1: renormalize.infer
 2.12% :     0.000265s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000029s :      1: rewriter_after_opt_a
 0.55% :     0.000068s :      1: rewriter_before_opt_a
 0.62% :     0.000077s :      1: symbol_engine_optimizer
31.04% :     0.003875s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.326.119 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc418b0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.326.661 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0173718, [33]
[bootstrap]: 0.00026375
[type_inference]: 0.00414272
[event_method]: 1.676e-05
[auto_monad]: 5.302e-05
[graph_reusing]: 4.82998e-06
[pre_auto_parallel]: 1.87999e-06
[py_interpret_to_execute]: 3.03e-05
[rewriter_before_opt_a]: 6.527e-05
[expand_dump_flag]: 2.28002e-06
[jit_opt_a]: 0.00518277, [2]
    [Cycle 1]: 0.00149019, [27]
        [switch_simplify]: 5.265e-05
        [loop_unroll]: 3.024e-05
        [a_1]: 0.00050986
        [with_stream_mark]: 1.453e-05
        [recompute_prepare]: 7.18e-06
        [updatestate_depend_eliminate]: 3.08998e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [parameter_eliminate]: 1.75001e-06
        [specialize_transform]: 5.59e-06
        [updatestate_useless_node_eliminater]: 4.71002e-06
        [accelerated_algorithm]: 5.50001e-06
        [meta_shard_fg_expand]: 1.76998e-06
        [get_grad_eliminate_]: 4.82998e-06
        [merge_forward]: 2.96001e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.258e-05
        [j_node_and_user_rematch]: 8.60001e-06
        [meta_fg_expand]: 2.07999e-06
        [replace_old_param]: 7.58001e-06
        [inline_without_move]: 4.90999e-06
        [renormalize]: 0.0005957
        [add_forward_monad_depend]: 6.02999e-06
        [auto_monad_grad]: 2.31e-06
        [auto_monad_eliminator]: 1.226e-05
        [cse]: 1.697e-05
        [replace_applicator]: 1.134e-05
    [Cycle 2]: 0.0003097, [27]
        [switch_simplify]: 5.47999e-06
        [loop_unroll]: 4.90001e-06
        [a_1]: 8.71e-05
        [with_stream_mark]: 8.1e-06
        [recompute_prepare]: 5.02999e-06
        [updatestate_depend_eliminate]: 2.53998e-06
        [updatestate_assign_eliminate]: 1.97999e-06
        [updatestate_loads_eliminate]: 1.96998e-06
        [parameter_eliminate]: 8.99978e-07
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.52998e-06
        [accelerated_algorithm]: 4.87998e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 4.58999e-06
        [merge_forward]: 2.51e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.154e-05
        [j_node_and_user_rematch]: 7.36999e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 7.13998e-06
        [inline_without_move]: 4.60999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.50001e-06
        [auto_monad_grad]: 6.09987e-07
        [auto_monad_eliminator]: 5.16002e-06
        [cse]: 9.29e-06
        [replace_applicator]: 4.74998e-06
[py_interpret_to_execute_after_opt_a]: 8.30999e-06
[rewriter_after_opt_a]: 2.884e-05
[convert_after_rewriter]: 6.13002e-06
[order_py_execute_after_rewriter]: 4.55001e-06
[mutable_eliminate]: 0.00049534
[jit_opt_b]: 4.839e-05, [1]
    [Cycle 1]: 4.156e-05, [2]
        [frontend_op_eliminate]: 1.594e-05
        [inline_after_opt_a]: 1.421e-05
[cconv]: 2.377e-05
[loop_unroll]: 0.00041255
[jit_opt_after_cconv]: 0.00013232, [1]
    [Cycle 1]: 0.00012601, [11]
        [c_1]: 1.955e-05
        [parameter_eliminate]: 2.53e-06
        [updatestate_depend_eliminate]: 5.03002e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.703e-05
        [call_graph_tuple_transform]: 1.707e-05
        [tuple_list_get_item_eliminator]: 4.94003e-06
        [none_parameter_eliminate]: 1.34e-06
        [renormalize]: 6.39993e-07
        [switch_simplify]: 5.08002e-06
[remove_dup_value]: 1.097e-05
[partial_unused_args_eliminate]: 2.49001e-06
[environ_conv]: 4.52998e-06
[add_recomputation]: 4.187e-05
[cse_after_recomputation]: 1.942e-05, [1]
    [Cycle 1]: 1.408e-05, [1]
        [cse]: 8.41002e-06
[auto_monad_reorder]: 1.297e-05
[get_jit_bprop_graph]: 1.57001e-06
[rewriter_after_jit_bprop_graph]: 4.62998e-06
[opt_after_jit_grad]: 0.00044496
[symbol_engine_optimizer]: 6.931e-05, [1]
    [Cycle 1]: 6.324e-05, [6]
        [build]: 2.79001e-06
        [elim_shapecalc]: 6.94001e-06
        [elim_not_effective]: 1.134e-05
        [opt_reshape]: 5.77001e-06
        [fold_const_symbol]: 8.12998e-06
        [renormalize]: 7.09988e-07
[validate]: 2.992e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00562535
[execute]: 7.48999e-06
Sums
    bootstrap                            :     0.000264s :  1.96%
    type_inference                       :     0.004143s : 30.85%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000053s :  0.39%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000030s :  0.23%
    rewriter_before_opt_a                :     0.000065s :  0.49%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.43%
    jit_opt_a.loop_unroll                :     0.000035s :  0.26%
    jit_opt_a.a_1                        :     0.000597s :  4.45%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000596s :  4.44%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000026s :  0.20%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000029s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000495s :  3.69%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000413s :  3.07%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000011s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000042s :  0.31%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000013s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000445s :  3.31%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000030s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005625s : 41.90%
    execute                              :     0.000007s :  0.06%


Time group info:
------[substitution.]   0.000193    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.49% :     0.000005s :      3: substitution.graph_param_transform
84.60% :     0.000163s :      4: substitution.inline
 1.50% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.47% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.62% :     0.000003s :      2: substitution.replace_old_param
 5.71% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004088     2
80.95% :     0.003309s :      1: type_inference.infer
19.05% :     0.000779s :      1: type_inference.specialize
------[replace.]   0.000060     6
66.78% :     0.000040s :      4: replace.inline
33.22% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000171     6
94.18% :     0.000161s :      4: match.inline
 5.82% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.52% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000001s :     10: predicate.addn_zero_filter
 2.17% :     0.000002s :     10: predicate.arithmetic_simplify
 1.48% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000001s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.65% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.56% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.27% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.04% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.96% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000000s :      3: predicate.opt_reshape
 2.10% :     0.000002s :     15: predicate.partial_eliminate
 1.38% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.29% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.50% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.31% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.83% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.22% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.83% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000522    12
44.31% :     0.000231s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.99% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.71% :     0.000254s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018867    76
 0.24% :     0.000045s :      1: add_recomputation
 0.30% :     0.000056s :      1: auto_monad
 0.08% :     0.000015s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.45% :     0.000273s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
27.49% :     0.005186s :      1: jit_opt_a
 0.71% :     0.000135s :      1: jit_opt_after_cconv
 0.27% :     0.000051s :      1: jit_opt_b
 2.23% :     0.000421s :      1: loop_unroll
 2.67% :     0.000503s :      1: mutable_eliminate
 4.23% :     0.000797s :     26: opt.transform.jit_opt_a
 0.23% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.40% :     0.000452s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000013s :      1: remove_dup_value
 1.69% :     0.000319s :      1: renormalize.infer
 1.43% :     0.000269s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000032s :      1: rewriter_after_opt_a
 0.36% :     0.000068s :      1: rewriter_before_opt_a
 0.38% :     0.000072s :      1: symbol_engine_optimizer
29.87% :     0.005636s :      1: task_emit
22.02% :     0.004155s :      1: type_inference
 0.25% :     0.000047s :      1: validate


TotalTime = 0.0118625, [30]
[bootstrap]: 0.00026951
[type_inference]: 0.00417844
[event_method]: 1.683e-05
[auto_monad]: 4.47e-05
[graph_reusing]: 3.66999e-06
[pre_auto_parallel]: 2.02999e-06
[py_interpret_to_execute]: 3.327e-05
[rewriter_before_opt_a]: 6.666e-05
[expand_dump_flag]: 2.24001e-06
[jit_opt_a]: 0.00510334, [2]
    [Cycle 1]: 0.00157592, [27]
        [switch_simplify]: 5.09e-05
        [loop_unroll]: 2.94e-05
        [a_1]: 0.00051211
        [with_stream_mark]: 1.28e-05
        [recompute_prepare]: 7.54002e-06
        [updatestate_depend_eliminate]: 2.94001e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [parameter_eliminate]: 1.34998e-06
        [specialize_transform]: 5.79999e-06
        [updatestate_useless_node_eliminater]: 5.51e-06
        [accelerated_algorithm]: 6.28e-06
        [meta_shard_fg_expand]: 1.45001e-06
        [get_grad_eliminate_]: 5.97999e-06
        [merge_forward]: 2.54999e-06
        [cell_reuse_recompute_pass]: 8.89995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.208e-05
        [j_node_and_user_rematch]: 9.12999e-06
        [meta_fg_expand]: 2.14999e-06
        [replace_old_param]: 8.92999e-06
        [inline_without_move]: 5.67999e-06
        [renormalize]: 0.00066699
        [add_forward_monad_depend]: 4.77998e-06
        [auto_monad_grad]: 2.26e-06
        [auto_monad_eliminator]: 1.348e-05
        [cse]: 2.51e-05
        [replace_applicator]: 1.177e-05
    [Cycle 2]: 0.00032767, [27]
        [switch_simplify]: 5.94e-06
        [loop_unroll]: 5.04e-06
        [a_1]: 9.67e-05
        [with_stream_mark]: 8.76002e-06
        [recompute_prepare]: 5.07e-06
        [updatestate_depend_eliminate]: 2.53998e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 4.99e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 4.99998e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.58999e-06
        [merge_forward]: 2.26998e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.222e-05
        [j_node_and_user_rematch]: 8e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 7.87e-06
        [inline_without_move]: 4.81002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 8.79983e-07
        [auto_monad_eliminator]: 5.15001e-06
        [cse]: 9.74999e-06
        [replace_applicator]: 5.64998e-06
[py_interpret_to_execute_after_opt_a]: 1.029e-05
[rewriter_after_opt_a]: 3.13e-05
[convert_after_rewriter]: 8.19998e-06
[order_py_execute_after_rewriter]: 5.54e-06
[mutable_eliminate]: 0.00057931
[jit_opt_b]: 5.104e-05, [1]
    [Cycle 1]: 4.377e-05, [2]
        [frontend_op_eliminate]: 1.682e-05
        [inline_after_opt_a]: 1.512e-05
[cconv]: 2.222e-05
[loop_unroll]: 0.00043298
[jit_opt_after_cconv]: 0.00014224, [1]
    [Cycle 1]: 0.00013564, [11]
        [c_1]: 2.196e-05
        [parameter_eliminate]: 2.49001e-06
        [updatestate_depend_eliminate]: 5.46e-06
        [updatestate_assign_eliminate]: 2.31998e-06
        [updatestate_loads_eliminate]: 2.33002e-06
        [cse]: 1.919e-05
        [call_graph_tuple_transform]: 1.865e-05
        [tuple_list_get_item_eliminator]: 5.44998e-06
        [none_parameter_eliminate]: 1.48002e-06
        [renormalize]: 5.10016e-07
        [switch_simplify]: 5.42999e-06
[remove_dup_value]: 1.399e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 4.90001e-06
[add_recomputation]: 4.743e-05
[cse_after_recomputation]: 1.998e-05, [1]
    [Cycle 1]: 1.455e-05, [1]
        [cse]: 8.59998e-06
[auto_monad_reorder]: 1.457e-05
[get_jit_bprop_graph]: 1.40999e-06
[rewriter_after_jit_bprop_graph]: 5.74999e-06
[opt_after_jit_grad]: 0.00047542
[symbol_engine_optimizer]: 7.353e-05, [1]
    [Cycle 1]: 6.731e-05, [6]
        [build]: 3.37002e-06
        [elim_shapecalc]: 8.38999e-06
        [elim_not_effective]: 1.295e-05
        [opt_reshape]: 5.66998e-06
        [fold_const_symbol]: 8.74e-06
        [renormalize]: 4.60015e-07
[validate]: 3.423e-05
Sums
    bootstrap                            :     0.000270s :  3.33%
    type_inference                       :     0.004178s : 51.61%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000045s :  0.55%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.41%
    rewriter_before_opt_a                :     0.000067s :  0.82%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000057s :  0.70%
    jit_opt_a.loop_unroll                :     0.000034s :  0.43%
    jit_opt_a.a_1                        :     0.000609s :  7.52%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.13%
    jit_opt_a.merge_forward              :     0.000005s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000667s :  8.24%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.23%
    jit_opt_a.cse                        :     0.000035s :  0.43%
    jit_opt_a.replace_applicator         :     0.000017s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.13%
    rewriter_after_opt_a                 :     0.000031s :  0.39%
    convert_after_rewriter               :     0.000008s :  0.10%
    order_py_execute_after_rewriter      :     0.000006s :  0.07%
    mutable_eliminate                    :     0.000579s :  7.16%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000022s :  0.27%
    loop_unroll                          :     0.000433s :  5.35%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000019s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000047s :  0.59%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.18%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000475s :  5.87%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000034s :  0.42%


Time group info:
------[substitution.]   0.000179    23
 1.22% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.95% :     0.000005s :      3: substitution.graph_param_transform
84.83% :     0.000152s :      4: substitution.inline
 1.72% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.94% :     0.000003s :      2: substitution.replace_old_param
 4.42% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004123     2
78.46% :     0.003235s :      1: type_inference.infer
21.54% :     0.000888s :      1: type_inference.specialize
------[replace.]   0.000066     6
67.52% :     0.000045s :      4: replace.inline
32.48% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000156     6
95.69% :     0.000149s :      4: match.inline
 4.31% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.43% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 2.17% :     0.000003s :     10: predicate.arithmetic_simplify
 1.16% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.44% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.53% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.76% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.38% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.14% :     0.000001s :      6: predicate.get_grad_eliminate
 0.40% :     0.000000s :      3: predicate.graph_param_transform
 4.96% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.20% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.82% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     15: predicate.partial_eliminate
 1.33% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.68% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.08% :     0.000010s :     53: predicate.switch_simplify
 1.29% :     0.000002s :     10: predicate.tile_eliminate
 1.31% :     0.000002s :     10: predicate.transpose_eliminate
 1.53% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000609    12
47.10% :     0.000287s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.21% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.69% :     0.000284s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013405    72
 0.38% :     0.000050s :      1: add_recomputation
 0.36% :     0.000048s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.12% :     0.000284s :      1: bootstrap
 0.18% :     0.000025s :      1: cconv
 0.08% :     0.000011s :      1: convert_after_rewriter
 0.16% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
38.10% :     0.005107s :      1: jit_opt_a
 1.08% :     0.000145s :      1: jit_opt_after_cconv
 0.40% :     0.000054s :      1: jit_opt_b
 3.29% :     0.000440s :      1: loop_unroll
 4.39% :     0.000589s :      1: mutable_eliminate
 6.10% :     0.000817s :     26: opt.transform.jit_opt_a
 0.36% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.61% :     0.000483s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000036s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.58% :     0.000346s :      1: renormalize.infer
 2.34% :     0.000314s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000034s :      1: rewriter_after_opt_a
 0.52% :     0.000069s :      1: rewriter_before_opt_a
 0.57% :     0.000076s :      1: symbol_engine_optimizer
31.27% :     0.004191s :      1: type_inference


TotalTime = 0.0196179, [33]
[bootstrap]: 0.00029402
[type_inference]: 0.00450114
[event_method]: 1.682e-05
[auto_monad]: 6.348e-05
[graph_reusing]: 5.37999e-06
[pre_auto_parallel]: 3.11999e-06
[py_interpret_to_execute]: 3.326e-05
[rewriter_before_opt_a]: 7.24e-05
[expand_dump_flag]: 3.01001e-06
[jit_opt_a]: 0.00548931, [2]
    [Cycle 1]: 0.00177694, [27]
        [switch_simplify]: 5.989e-05
        [loop_unroll]: 2.887e-05
        [a_1]: 0.00054722
        [with_stream_mark]: 2.113e-05
        [recompute_prepare]: 9.34e-06
        [updatestate_depend_eliminate]: 3.71999e-06
        [updatestate_assign_eliminate]: 3.03e-06
        [updatestate_loads_eliminate]: 2.64001e-06
        [parameter_eliminate]: 1.84998e-06
        [specialize_transform]: 7.46001e-06
        [updatestate_useless_node_eliminater]: 5.34e-06
        [accelerated_algorithm]: 6.38e-06
        [meta_shard_fg_expand]: 2.39999e-06
        [get_grad_eliminate_]: 5.89999e-06
        [merge_forward]: 3.53999e-06
        [cell_reuse_recompute_pass]: 1.38002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.517e-05
        [j_node_and_user_rematch]: 9.57001e-06
        [meta_fg_expand]: 2.63998e-06
        [replace_old_param]: 1.025e-05
        [inline_without_move]: 6.17999e-06
        [renormalize]: 0.00075233
        [add_forward_monad_depend]: 8.05999e-06
        [auto_monad_grad]: 2.68998e-06
        [auto_monad_eliminator]: 1.791e-05
        [cse]: 3.261e-05
        [replace_applicator]: 1.649e-05
    [Cycle 2]: 0.00034516, [27]
        [switch_simplify]: 5.96998e-06
        [loop_unroll]: 4.75001e-06
        [a_1]: 9.231e-05
        [with_stream_mark]: 1.32e-05
        [recompute_prepare]: 5.59e-06
        [updatestate_depend_eliminate]: 2.69001e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [parameter_eliminate]: 1.35001e-06
        [specialize_transform]: 5.64998e-06
        [updatestate_useless_node_eliminater]: 5.09998e-06
        [accelerated_algorithm]: 5.05001e-06
        [meta_shard_fg_expand]: 2.17001e-06
        [get_grad_eliminate_]: 4.99998e-06
        [merge_forward]: 2.59999e-06
        [cell_reuse_recompute_pass]: 2.29001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.484e-05
        [j_node_and_user_rematch]: 8.88002e-06
        [meta_fg_expand]: 1.84e-06
        [replace_old_param]: 8.73001e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.94e-06
        [auto_monad_grad]: 1.08001e-06
        [auto_monad_eliminator]: 6.44001e-06
        [cse]: 1.3e-05
        [replace_applicator]: 5.32999e-06
[py_interpret_to_execute_after_opt_a]: 1.377e-05
[rewriter_after_opt_a]: 3.803e-05
[convert_after_rewriter]: 9.08002e-06
[order_py_execute_after_rewriter]: 5.23002e-06
[mutable_eliminate]: 0.0006828
[jit_opt_b]: 5.324e-05, [1]
    [Cycle 1]: 4.376e-05, [2]
        [frontend_op_eliminate]: 1.701e-05
        [inline_after_opt_a]: 1.505e-05
[cconv]: 2.873e-05
[loop_unroll]: 0.00052584
[jit_opt_after_cconv]: 0.00015453, [1]
    [Cycle 1]: 0.0001467, [11]
        [c_1]: 1.999e-05
        [parameter_eliminate]: 3.20002e-06
        [updatestate_depend_eliminate]: 6.46e-06
        [updatestate_assign_eliminate]: 2.64999e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [cse]: 2.677e-05
        [call_graph_tuple_transform]: 2.106e-05
        [tuple_list_get_item_eliminator]: 5.24e-06
        [none_parameter_eliminate]: 1.43002e-06
        [renormalize]: 8.10018e-07
        [switch_simplify]: 5.52001e-06
[remove_dup_value]: 1.514e-05
[partial_unused_args_eliminate]: 2.22001e-06
[environ_conv]: 6.01e-06
[add_recomputation]: 5.232e-05
[cse_after_recomputation]: 2.303e-05, [1]
    [Cycle 1]: 1.643e-05, [1]
        [cse]: 9.85002e-06
[auto_monad_reorder]: 1.632e-05
[get_jit_bprop_graph]: 2.26998e-06
[rewriter_after_jit_bprop_graph]: 7.77e-06
[opt_after_jit_grad]: 0.00052113
[symbol_engine_optimizer]: 7.819e-05, [1]
    [Cycle 1]: 7.08e-05, [6]
        [build]: 5.11002e-06
        [elim_shapecalc]: 8.07e-06
        [elim_not_effective]: 1.311e-05
        [opt_reshape]: 5.57001e-06
        [fold_const_symbol]: 8.33001e-06
        [renormalize]: 7.99977e-07
[validate]: 3.86e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.0066339
[execute]: 9.19e-06
Sums
    bootstrap                            :     0.000294s :  1.89%
    type_inference                       :     0.004501s : 28.87%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000063s :  0.41%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.21%
    rewriter_before_opt_a                :     0.000072s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.42%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000640s :  4.10%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.12%
    jit_opt_a.inline_without_move        :     0.000012s :  0.07%
    jit_opt_a.renormalize                :     0.000752s :  4.83%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.16%
    jit_opt_a.cse                        :     0.000046s :  0.29%
    jit_opt_a.replace_applicator         :     0.000022s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.09%
    rewriter_after_opt_a                 :     0.000038s :  0.24%
    convert_after_rewriter               :     0.000009s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000683s :  4.38%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000029s :  0.18%
    loop_unroll                          :     0.000526s :  3.37%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000015s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000052s :  0.34%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000521s :  3.34%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006634s : 42.55%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000217    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.82% :     0.000002s :      2: substitution.fold_const_symbol
 2.59% :     0.000006s :      3: substitution.graph_param_transform
82.63% :     0.000179s :      4: substitution.inline
 2.10% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.05% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.76% :     0.000006s :      2: substitution.replace_old_param
 6.22% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004429     2
80.26% :     0.003555s :      1: type_inference.infer
19.74% :     0.000874s :      1: type_inference.specialize
------[replace.]   0.000071     6
65.66% :     0.000047s :      4: replace.inline
34.34% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000189     6
93.56% :     0.000177s :      4: match.inline
 6.44% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.49% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.01% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000002s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.73% :     0.000001s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.16% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.63% :     0.000007s :     22: predicate.inline
 1.22% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.47% :     0.000002s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.83% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 2.34% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     15: predicate.partial_eliminate
 1.24% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 1.16% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.22% :     0.000003s :     18: predicate.replace_applicator
 0.86% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.63% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.62% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.26% :     0.000001s :      6: predicate.specialize_transform
 1.22% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.56% :     0.000009s :     53: predicate.switch_simplify
 1.10% :     0.000001s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.73% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000581    12
44.45% :     0.000258s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.26% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.29% :     0.000286s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021346    76
 0.26% :     0.000056s :      1: add_recomputation
 0.31% :     0.000067s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.46% :     0.000311s :      1: bootstrap
 0.15% :     0.000032s :      1: cconv
 0.06% :     0.000012s :      1: convert_after_rewriter
 0.12% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
25.73% :     0.005493s :      1: jit_opt_a
 0.74% :     0.000157s :      1: jit_opt_after_cconv
 0.26% :     0.000056s :      1: jit_opt_b
 2.51% :     0.000535s :      1: loop_unroll
 3.26% :     0.000695s :      1: mutable_eliminate
 4.07% :     0.000869s :     26: opt.transform.jit_opt_a
 0.22% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.48% :     0.000530s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000036s :      1: py_interpret_to_execute
 0.08% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 1.74% :     0.000370s :      1: renormalize.infer
 1.74% :     0.000372s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000041s :      1: rewriter_after_opt_a
 0.35% :     0.000075s :      1: rewriter_before_opt_a
 0.38% :     0.000081s :      1: symbol_engine_optimizer
31.16% :     0.006651s :      1: task_emit
21.17% :     0.004519s :      1: type_inference
 0.28% :     0.000060s :      1: validate


TotalTime = 0.0294035, [33]
[bootstrap]: 0.00018182
[type_inference]: 0.00979383
[event_method]: 4.381e-05
[auto_monad]: 9.327e-05
[graph_reusing]: 7.05e-06
[pre_auto_parallel]: 2.51998e-06
[py_interpret_to_execute]: 2.601e-05
[rewriter_before_opt_a]: 9.042e-05
[expand_dump_flag]: 3.09999e-06
[jit_opt_a]: 0.0109759, [3]
    [Cycle 1]: 0.00544359, [27]
        [switch_simplify]: 6.288e-05
        [loop_unroll]: 3.803e-05
        [a_1]: 0.00085854
        [with_stream_mark]: 2.697e-05
        [recompute_prepare]: 1.936e-05
        [updatestate_depend_eliminate]: 7.48999e-06
        [updatestate_assign_eliminate]: 6.26e-06
        [updatestate_loads_eliminate]: 6.17999e-06
        [parameter_eliminate]: 2.91999e-06
        [specialize_transform]: 1.352e-05
        [updatestate_useless_node_eliminater]: 1.156e-05
        [accelerated_algorithm]: 1.188e-05
        [meta_shard_fg_expand]: 3.63999e-06
        [get_grad_eliminate_]: 1.185e-05
        [merge_forward]: 8.43001e-06
        [cell_reuse_recompute_pass]: 8.79983e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.716e-05
        [j_node_and_user_rematch]: 2.351e-05
        [meta_fg_expand]: 0.00141697
        [replace_old_param]: 5.366e-05
        [inline_without_move]: 5.105e-05
        [renormalize]: 0.00230263
        [add_forward_monad_depend]: 8.60001e-06
        [auto_monad_grad]: 4.37998e-06
        [auto_monad_eliminator]: 4.478e-05
        [cse]: 0.00013895
        [replace_applicator]: 6.449e-05
    [Cycle 2]: 0.00187806, [27]
        [switch_simplify]: 3.57e-05
        [loop_unroll]: 3.347e-05
        [a_1]: 0.00100254
        [with_stream_mark]: 1.286e-05
        [recompute_prepare]: 8.25e-06
        [updatestate_depend_eliminate]: 3.58e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [parameter_eliminate]: 1.42999e-06
        [specialize_transform]: 6.49001e-06
        [updatestate_useless_node_eliminater]: 6.17001e-06
        [accelerated_algorithm]: 5.89999e-06
        [meta_shard_fg_expand]: 1.77001e-06
        [get_grad_eliminate_]: 5.94999e-06
        [merge_forward]: 3.08e-06
        [cell_reuse_recompute_pass]: 7.90023e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.328e-05
        [j_node_and_user_rematch]: 1.031e-05
        [meta_fg_expand]: 3.864e-05
        [replace_old_param]: 9.76003e-06
        [inline_without_move]: 6.21998e-06
        [renormalize]: 0.00046747
        [add_forward_monad_depend]: 3.98001e-06
        [auto_monad_grad]: 1.14e-06
        [auto_monad_eliminator]: 1.219e-05
        [cse]: 2.086e-05
        [replace_applicator]: 1.394e-05
    [Cycle 3]: 0.00036173, [27]
        [switch_simplify]: 6.37001e-06
        [loop_unroll]: 5.72999e-06
        [a_1]: 0.00011674
        [with_stream_mark]: 8.50001e-06
        [recompute_prepare]: 5.92001e-06
        [updatestate_depend_eliminate]: 3.3e-06
        [updatestate_assign_eliminate]: 2.73998e-06
        [updatestate_loads_eliminate]: 2.50002e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 6.36e-06
        [updatestate_useless_node_eliminater]: 5.52999e-06
        [accelerated_algorithm]: 5.77001e-06
        [meta_shard_fg_expand]: 1.39998e-06
        [get_grad_eliminate_]: 5.72999e-06
        [merge_forward]: 2.97002e-06
        [cell_reuse_recompute_pass]: 1.28002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.503e-05
        [j_node_and_user_rematch]: 1.045e-05
        [meta_fg_expand]: 2.12999e-06
        [replace_old_param]: 8.34002e-06
        [inline_without_move]: 5.61e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.33002e-06
        [auto_monad_grad]: 1.04e-06
        [auto_monad_eliminator]: 7.07002e-06
        [cse]: 1.361e-05
        [replace_applicator]: 5.97999e-06
[py_interpret_to_execute_after_opt_a]: 9.66e-06
[rewriter_after_opt_a]: 3.14e-05
[convert_after_rewriter]: 7.09001e-06
[order_py_execute_after_rewriter]: 5.34998e-06
[mutable_eliminate]: 0.00056216
[jit_opt_b]: 5.404e-05, [1]
    [Cycle 1]: 4.686e-05, [2]
        [frontend_op_eliminate]: 1.843e-05
        [inline_after_opt_a]: 1.761e-05
[cconv]: 1.685e-05
[loop_unroll]: 0.00043116
[jit_opt_after_cconv]: 0.00019484, [1]
    [Cycle 1]: 0.00018777, [11]
        [c_1]: 2.491e-05
        [parameter_eliminate]: 2.06e-06
        [updatestate_depend_eliminate]: 5.47001e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 4.463e-05
        [cse]: 2.357e-05
        [call_graph_tuple_transform]: 2.073e-05
        [tuple_list_get_item_eliminator]: 5.86e-06
        [none_parameter_eliminate]: 1.02998e-06
        [renormalize]: 5.09986e-07
        [switch_simplify]: 6.48e-06
[remove_dup_value]: 1.22e-05
[partial_unused_args_eliminate]: 1.97001e-06
[environ_conv]: 4.91002e-06
[add_recomputation]: 4.321e-05
[cse_after_recomputation]: 3.333e-05, [1]
    [Cycle 1]: 2.133e-05, [1]
        [cse]: 1.523e-05
[auto_monad_reorder]: 1.661e-05
[get_jit_bprop_graph]: 1.24e-06
[rewriter_after_jit_bprop_graph]: 4.88001e-06
[opt_after_jit_grad]: 0.00048031
[symbol_engine_optimizer]: 8.005e-05, [1]
    [Cycle 1]: 7.373e-05, [6]
        [build]: 3.75e-06
        [elim_shapecalc]: 9.71e-06
        [elim_not_effective]: 1.48e-05
        [opt_reshape]: 6.87002e-06
        [fold_const_symbol]: 1.046e-05
        [renormalize]: 4.69998e-07
[validate]: 3.683e-05
[backend_pass]: 6.50005e-07
[task_emit]: 0.00597207
[execute]: 5.48002e-06
Sums
    bootstrap                            :     0.000182s :  0.72%
    type_inference                       :     0.009794s : 38.66%
    event_method                         :     0.000044s :  0.17%
    auto_monad                           :     0.000093s :  0.37%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.10%
    rewriter_before_opt_a                :     0.000090s :  0.36%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000105s :  0.41%
    jit_opt_a.loop_unroll                :     0.000077s :  0.30%
    jit_opt_a.a_1                        :     0.001978s :  7.81%
    jit_opt_a.with_stream_mark           :     0.000048s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000034s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.09%
    jit_opt_a.merge_forward              :     0.000014s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000055s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000044s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001458s :  5.75%
    jit_opt_a.replace_old_param          :     0.000072s :  0.28%
    jit_opt_a.inline_without_move        :     0.000063s :  0.25%
    jit_opt_a.renormalize                :     0.002770s : 10.94%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000064s :  0.25%
    jit_opt_a.cse                        :     0.000173s :  0.68%
    jit_opt_a.replace_applicator         :     0.000084s :  0.33%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.04%
    rewriter_after_opt_a                 :     0.000031s :  0.12%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000562s :  2.22%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.07%
    cconv                                :     0.000017s :  0.07%
    loop_unroll                          :     0.000431s :  1.70%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000045s :  0.18%
    jit_opt_after_cconv.cse              :     0.000024s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000012s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000043s :  0.17%
    cse_after_recomputation.cse          :     0.000015s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000480s :  1.90%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005972s : 23.58%
    execute                              :     0.000005s :  0.02%


Time group info:
------[substitution.]   0.000469   110
 0.40% :     0.000002s :      3: substitution.elim_not_effective
 0.32% :     0.000002s :      3: substitution.fold_const_symbol
 1.13% :     0.000005s :      4: substitution.graph_param_transform
63.52% :     0.000298s :     12: substitution.inline
 3.60% :     0.000017s :      2: substitution.inline_without_move
 1.85% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.28% :     0.000011s :      7: substitution.minmaximum_grad
 1.51% :     0.000007s :      5: substitution.partial_eliminate
 2.24% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.39% :     0.000025s :     10: substitution.replace_applicator
 1.45% :     0.000007s :      7: substitution.replace_old_param
 0.62% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.40% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.90% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 8.40% :     0.000039s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009728     2
89.24% :     0.008681s :      1: type_inference.infer
10.76% :     0.001047s :      1: type_inference.specialize
------[replace.]   0.000137    19
63.23% :     0.000087s :     12: replace.inline
36.77% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000311    19
93.58% :     0.000291s :     12: match.inline
 6.42% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000403  2568
 1.58% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.44% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.26% :     0.000005s :     43: predicate.addn_check_dump
 1.54% :     0.000006s :     43: predicate.addn_zero_filter
11.89% :     0.000048s :     43: predicate.arithmetic_simplify
 1.38% :     0.000006s :     43: predicate.cast_eliminate
 0.14% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.27% :     0.000005s :     43: predicate.compare_switch_simplify
 1.39% :     0.000006s :     43: predicate.depend_value_elim
 1.37% :     0.000006s :     43: predicate.dict_get_item_const_eliminator
 1.33% :     0.000005s :     43: predicate.dict_get_item_eliminator
 1.43% :     0.000006s :     43: predicate.dict_set_item_eliminator
 0.21% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      4: predicate.elim_not_effective
 0.20% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.38% :     0.000006s :     43: predicate.environ_add_const_eliminate
 1.30% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.30% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.33% :     0.000005s :     43: predicate.environ_get_eliminate
 1.37% :     0.000006s :     43: predicate.environ_get_set_eliminate
 0.08% :     0.000000s :      4: predicate.fold_const_symbol
 0.87% :     0.000004s :     19: predicate.get_grad_eliminate
 0.09% :     0.000000s :      4: predicate.graph_param_transform
 3.92% :     0.000016s :     70: predicate.inline
 1.62% :     0.000007s :     40: predicate.inline_without_move
 0.33% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.78% :     0.000003s :     19: predicate.less_batch_normalization
 1.59% :     0.000006s :     50: predicate.list_to_tuple_eliminator_
 1.72% :     0.000007s :     54: predicate.load_eliminater
 0.43% :     0.000002s :      4: predicate.loop_unroll_after_grad
 2.82% :     0.000011s :     85: predicate.loop_unroll_before_grad
 1.68% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.35% :     0.000005s :     43: predicate.merge_addn
 1.32% :     0.000005s :     43: predicate.minmaximum_grad
 0.52% :     0.000002s :      4: predicate.mutable_eliminate
 0.17% :     0.000001s :      4: predicate.opt_reshape
 2.11% :     0.000009s :     54: predicate.partial_eliminate
 1.40% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.66% :     0.000007s :     43: predicate.reduce_eliminate
 1.66% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.36% :     0.000001s :     19: predicate.remove_not_recompute_node
 2.29% :     0.000009s :     96: predicate.replace_applicator
 0.86% :     0.000003s :     40: predicate.replace_old_param
 0.11% :     0.000000s :      4: predicate.reset_defer_inline
 1.52% :     0.000006s :     43: predicate.reshape_eliminate
 1.46% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.33% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.53% :     0.000006s :     43: predicate.same_eliminate
 0.40% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.39% :     0.000002s :      8: predicate.special_op_eliminate
 0.81% :     0.000003s :     19: predicate.specialize_transform
 1.50% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000005s :     43: predicate.stack_unstack_eliminate
 0.15% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.42% :     0.000010s :     62: predicate.switch_defer_inline
 2.22% :     0.000009s :     62: predicate.switch_layer_defer_inline
 5.57% :     0.000022s :    151: predicate.switch_simplify
 1.34% :     0.000005s :     43: predicate.tile_eliminate
 1.37% :     0.000006s :     43: predicate.transpose_eliminate
 1.64% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.51% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 2.80% :     0.000011s :     58: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000007s :     43: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000007s :     50: predicate.tuple_to_list_eliminator_
 1.79% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000010s :     73: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000006s :     43: predicate.value_based_eliminate
 0.14% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.20% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001298    29
59.41% :     0.000771s :     13: func_graph_cloner_run.FuncGraphClonerGraph
40.59% :     0.000527s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.034865    91
 0.13% :     0.000046s :      1: add_recomputation
 0.28% :     0.000097s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.55% :     0.000190s :      1: bootstrap
 0.06% :     0.000020s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000035s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.14% :     0.000048s :      1: event_method
 0.03% :     0.000009s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000003s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
31.49% :     0.010979s :      1: jit_opt_a
 0.57% :     0.000198s :      1: jit_opt_after_cconv
 0.16% :     0.000057s :      1: jit_opt_b
 1.26% :     0.000439s :      1: loop_unroll
 1.63% :     0.000570s :      1: mutable_eliminate
 7.35% :     0.002564s :     39: opt.transform.jit_opt_a
 0.16% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000030s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.40% :     0.000488s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000029s :      1: py_interpret_to_execute
 0.03% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000015s :      1: remove_dup_value
 4.27% :     0.001490s :      2: renormalize.infer
 3.63% :     0.001267s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000034s :      1: rewriter_after_opt_a
 0.27% :     0.000093s :      1: rewriter_before_opt_a
 0.24% :     0.000083s :      1: symbol_engine_optimizer
17.16% :     0.005981s :      1: task_emit
28.13% :     0.009808s :      1: type_inference
 0.16% :     0.000057s :      1: validate


TotalTime = 0.017268, [33]
[bootstrap]: 0.00029177
[type_inference]: 0.00424165
[event_method]: 1.569e-05
[auto_monad]: 5.771e-05
[graph_reusing]: 4.92e-06
[pre_auto_parallel]: 2.02999e-06
[py_interpret_to_execute]: 2.927e-05
[rewriter_before_opt_a]: 6.673e-05
[expand_dump_flag]: 2.41e-06
[jit_opt_a]: 0.00513704, [2]
    [Cycle 1]: 0.00139691, [27]
        [switch_simplify]: 4.592e-05
        [loop_unroll]: 2.861e-05
        [a_1]: 0.00049322
        [with_stream_mark]: 1.269e-05
        [recompute_prepare]: 6.93e-06
        [updatestate_depend_eliminate]: 2.79001e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 5.64e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.25001e-06
        [meta_shard_fg_expand]: 1.67001e-06
        [get_grad_eliminate_]: 1.424e-05
        [merge_forward]: 3.11001e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.195e-05
        [j_node_and_user_rematch]: 8.84003e-06
        [meta_fg_expand]: 2.11e-06
        [replace_old_param]: 8.13999e-06
        [inline_without_move]: 5.20001e-06
        [renormalize]: 0.00053261
        [add_forward_monad_depend]: 4.40999e-06
        [auto_monad_grad]: 1.50001e-06
        [auto_monad_eliminator]: 1.119e-05
        [cse]: 1.374e-05
        [replace_applicator]: 1.086e-05
    [Cycle 2]: 0.00030798, [27]
        [switch_simplify]: 5.35001e-06
        [loop_unroll]: 4.77998e-06
        [a_1]: 8.694e-05
        [with_stream_mark]: 8.38999e-06
        [recompute_prepare]: 4.63999e-06
        [updatestate_depend_eliminate]: 2.58e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.91003e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 4.72998e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 4.98001e-06
        [meta_shard_fg_expand]: 1.06997e-06
        [get_grad_eliminate_]: 4.39002e-06
        [merge_forward]: 2.31998e-06
        [cell_reuse_recompute_pass]: 1.78997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.165e-05
        [j_node_and_user_rematch]: 7.21999e-06
        [meta_fg_expand]: 1.45001e-06
        [replace_old_param]: 1.17e-05
        [inline_without_move]: 4.66002e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 4.77998e-06
        [cse]: 9.27001e-06
        [replace_applicator]: 4.72e-06
[py_interpret_to_execute_after_opt_a]: 9.30001e-06
[rewriter_after_opt_a]: 2.744e-05
[convert_after_rewriter]: 6.38998e-06
[order_py_execute_after_rewriter]: 4.41002e-06
[mutable_eliminate]: 0.00048451
[jit_opt_b]: 4.845e-05, [1]
    [Cycle 1]: 4.135e-05, [2]
        [frontend_op_eliminate]: 1.558e-05
        [inline_after_opt_a]: 1.475e-05
[cconv]: 2.199e-05
[loop_unroll]: 0.00040578
[jit_opt_after_cconv]: 0.00013112, [1]
    [Cycle 1]: 0.0001248, [11]
        [c_1]: 1.954e-05
        [parameter_eliminate]: 2.53e-06
        [updatestate_depend_eliminate]: 4.57e-06
        [updatestate_assign_eliminate]: 2.20002e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 1.653e-05
        [call_graph_tuple_transform]: 1.779e-05
        [tuple_list_get_item_eliminator]: 4.99e-06
        [none_parameter_eliminate]: 1.26997e-06
        [renormalize]: 5.50004e-07
        [switch_simplify]: 4.95001e-06
[remove_dup_value]: 1.04e-05
[partial_unused_args_eliminate]: 2.13002e-06
[environ_conv]: 4.46002e-06
[add_recomputation]: 4e-05
[cse_after_recomputation]: 1.9e-05, [1]
    [Cycle 1]: 1.381e-05, [1]
        [cse]: 8.10999e-06
[auto_monad_reorder]: 1.345e-05
[get_jit_bprop_graph]: 1.57001e-06
[rewriter_after_jit_bprop_graph]: 4.25e-06
[opt_after_jit_grad]: 0.00044489
[symbol_engine_optimizer]: 6.81e-05, [1]
    [Cycle 1]: 6.123e-05, [6]
        [build]: 2.69001e-06
        [elim_shapecalc]: 7.03e-06
        [elim_not_effective]: 1.165e-05
        [opt_reshape]: 5.32001e-06
        [fold_const_symbol]: 8.62998e-06
        [renormalize]: 3.9002e-07
[validate]: 3.162e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00546381
[execute]: 6.12999e-06
Sums
    bootstrap                            :     0.000292s :  2.20%
    type_inference                       :     0.004242s : 31.92%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000058s :  0.43%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.22%
    rewriter_before_opt_a                :     0.000067s :  0.50%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000051s :  0.39%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000580s :  4.37%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000019s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000020s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000533s :  4.01%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.12%
    jit_opt_a.cse                        :     0.000023s :  0.17%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000027s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000485s :  3.65%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000022s :  0.17%
    loop_unroll                          :     0.000406s :  3.05%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000010s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000040s :  0.30%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000013s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000445s :  3.35%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005464s : 41.12%
    execute                              :     0.000006s :  0.05%


Time group info:
------[substitution.]   0.000186    23
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000001s :      2: substitution.fold_const_symbol
 2.58% :     0.000005s :      3: substitution.graph_param_transform
82.32% :     0.000153s :      4: substitution.inline
 1.38% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.18% :     0.000004s :      4: substitution.remove_not_recompute_node
 4.56% :     0.000008s :      2: substitution.replace_old_param
 5.34% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004187     2
81.43% :     0.003410s :      1: type_inference.infer
18.57% :     0.000777s :      1: type_inference.specialize
------[replace.]   0.000060     6
66.85% :     0.000040s :      4: replace.inline
33.15% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000160     6
94.44% :     0.000151s :      4: match.inline
 5.56% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.21% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.48% :     0.000002s :     10: predicate.addn_zero_filter
 2.08% :     0.000002s :     10: predicate.arithmetic_simplify
 1.35% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000001s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.11% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000001s :      6: predicate.less_batch_normalization
 1.62% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.33% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.21% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.08% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.42% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000002s :     15: predicate.partial_eliminate
 1.60% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.02% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.34% :     0.000008s :     53: predicate.switch_simplify
 1.46% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000507    12
49.35% :     0.000250s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.86% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.80% :     0.000227s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018682    76
 0.23% :     0.000043s :      1: add_recomputation
 0.33% :     0.000061s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.62% :     0.000303s :      1: bootstrap
 0.13% :     0.000025s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
27.51% :     0.005140s :      1: jit_opt_a
 0.72% :     0.000134s :      1: jit_opt_after_cconv
 0.27% :     0.000051s :      1: jit_opt_b
 2.21% :     0.000413s :      1: loop_unroll
 2.63% :     0.000492s :      1: mutable_eliminate
 4.20% :     0.000785s :     26: opt.transform.jit_opt_a
 0.24% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.42% :     0.000451s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000013s :      1: remove_dup_value
 1.50% :     0.000280s :      1: renormalize.infer
 1.32% :     0.000246s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000030s :      1: rewriter_after_opt_a
 0.37% :     0.000069s :      1: rewriter_before_opt_a
 0.38% :     0.000071s :      1: symbol_engine_optimizer
29.30% :     0.005474s :      1: task_emit
22.75% :     0.004249s :      1: type_inference
 0.26% :     0.000049s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.362.842 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc5ca60> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.363.389 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0119066, [30]
[bootstrap]: 0.00028016
[type_inference]: 0.0044461
[event_method]: 1.728e-05
[auto_monad]: 6.269e-05
[graph_reusing]: 5.32999e-06
[pre_auto_parallel]: 2.68e-06
[py_interpret_to_execute]: 3.394e-05
[rewriter_before_opt_a]: 7.292e-05
[expand_dump_flag]: 2.79999e-06
[jit_opt_a]: 0.0050136, [2]
    [Cycle 1]: 0.0014661, [27]
        [switch_simplify]: 5.234e-05
        [loop_unroll]: 2.99e-05
        [a_1]: 0.00051031
        [with_stream_mark]: 1.438e-05
        [recompute_prepare]: 6.85998e-06
        [updatestate_depend_eliminate]: 3.49001e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.76999e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 6.06e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 5.59998e-06
        [meta_shard_fg_expand]: 1.95001e-06
        [get_grad_eliminate_]: 5.20001e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 9.89996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.52e-05
        [j_node_and_user_rematch]: 9.30001e-06
        [meta_fg_expand]: 2.34999e-06
        [replace_old_param]: 8.37e-06
        [inline_without_move]: 5.33002e-06
        [renormalize]: 0.00055247
        [add_forward_monad_depend]: 5.09003e-06
        [auto_monad_grad]: 1.91998e-06
        [auto_monad_eliminator]: 1.38e-05
        [cse]: 2.388e-05
        [replace_applicator]: 1.105e-05
    [Cycle 2]: 0.00032019, [27]
        [switch_simplify]: 5.76e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 9.041e-05
        [with_stream_mark]: 8.12e-06
        [recompute_prepare]: 5.22e-06
        [updatestate_depend_eliminate]: 2.63e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 5.07e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 1.16002e-06
        [get_grad_eliminate_]: 4.70999e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.226e-05
        [j_node_and_user_rematch]: 7.73001e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 7.12997e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 5.29e-06
        [cse]: 9.67999e-06
        [replace_applicator]: 5.51e-06
[py_interpret_to_execute_after_opt_a]: 8.76002e-06
[rewriter_after_opt_a]: 3.129e-05
[convert_after_rewriter]: 7.05e-06
[order_py_execute_after_rewriter]: 7.43e-06
[mutable_eliminate]: 0.00046668
[jit_opt_b]: 4.891e-05, [1]
    [Cycle 1]: 4.267e-05, [2]
        [frontend_op_eliminate]: 1.662e-05
        [inline_after_opt_a]: 1.491e-05
[cconv]: 2.643e-05
[loop_unroll]: 0.00040412
[jit_opt_after_cconv]: 0.00013538, [1]
    [Cycle 1]: 0.0001293, [11]
        [c_1]: 2.077e-05
        [parameter_eliminate]: 2.54999e-06
        [updatestate_depend_eliminate]: 5.06997e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 1.583e-05
        [call_graph_tuple_transform]: 1.792e-05
        [tuple_list_get_item_eliminator]: 5.39e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 5.60016e-07
        [switch_simplify]: 5.52001e-06
[remove_dup_value]: 1.29e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 4.97e-06
[add_recomputation]: 4.377e-05
[cse_after_recomputation]: 1.956e-05, [1]
    [Cycle 1]: 1.405e-05, [1]
        [cse]: 8.47e-06
[auto_monad_reorder]: 3.556e-05
[get_jit_bprop_graph]: 1.76003e-06
[rewriter_after_jit_bprop_graph]: 5.17999e-06
[opt_after_jit_grad]: 0.00044734
[symbol_engine_optimizer]: 7.229e-05, [1]
    [Cycle 1]: 6.571e-05, [6]
        [build]: 3.24001e-06
        [elim_shapecalc]: 7.7e-06
        [elim_not_effective]: 1.255e-05
        [opt_reshape]: 5.62999e-06
        [fold_const_symbol]: 8.64e-06
        [renormalize]: 5.50004e-07
[validate]: 3.039e-05
Sums
    bootstrap                            :     0.000280s :  3.45%
    type_inference                       :     0.004446s : 54.73%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000063s :  0.77%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.42%
    rewriter_before_opt_a                :     0.000073s :  0.90%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000058s :  0.72%
    jit_opt_a.loop_unroll                :     0.000035s :  0.43%
    jit_opt_a.a_1                        :     0.000601s :  7.39%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000553s :  6.80%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.23%
    jit_opt_a.cse                        :     0.000034s :  0.41%
    jit_opt_a.replace_applicator         :     0.000017s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000031s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000007s :  0.09%
    mutable_eliminate                    :     0.000467s :  5.74%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000026s :  0.33%
    loop_unroll                          :     0.000404s :  4.97%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000013s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000044s :  0.54%
    cse_after_recomputation.cse          :     0.000008s :  0.10%
    auto_monad_reorder                   :     0.000036s :  0.44%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000447s :  5.51%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000030s :  0.37%


Time group info:
------[substitution.]   0.000181    23
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.82% :     0.000005s :      3: substitution.graph_param_transform
82.88% :     0.000150s :      4: substitution.inline
 1.44% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.68% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.04% :     0.000004s :      2: substitution.replace_old_param
 6.45% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004388     2
79.48% :     0.003487s :      1: type_inference.infer
20.52% :     0.000900s :      1: type_inference.specialize
------[replace.]   0.000065     6
67.25% :     0.000044s :      4: replace.inline
32.75% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000158     6
93.37% :     0.000148s :      4: match.inline
 6.63% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.23% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 2.09% :     0.000002s :     10: predicate.arithmetic_simplify
 1.10% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.08% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.50% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.49% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.52% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.43% :     0.000003s :     15: predicate.partial_eliminate
 1.27% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     10: predicate.reduce_eliminate
 1.63% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.67% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.07% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.62% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.57% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.62% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.94% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000623    12
50.98% :     0.000318s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.63% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.39% :     0.000277s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013327    72
 0.35% :     0.000047s :      1: add_recomputation
 0.49% :     0.000066s :      1: auto_monad
 0.29% :     0.000039s :      1: auto_monad_reorder
 2.21% :     0.000295s :      1: bootstrap
 0.22% :     0.000029s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.15% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
37.64% :     0.005017s :      1: jit_opt_a
 1.04% :     0.000138s :      1: jit_opt_after_cconv
 0.39% :     0.000052s :      1: jit_opt_b
 3.09% :     0.000412s :      1: loop_unroll
 3.56% :     0.000474s :      1: mutable_eliminate
 6.06% :     0.000807s :     26: opt.transform.jit_opt_a
 0.35% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.42% :     0.000455s :      1: opt_after_jit_grad
 0.07% :     0.000010s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000037s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000015s :      1: remove_dup_value
 1.99% :     0.000265s :      1: renormalize.infer
 2.10% :     0.000280s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000034s :      1: rewriter_after_opt_a
 0.57% :     0.000076s :      1: rewriter_before_opt_a
 0.56% :     0.000075s :      1: symbol_engine_optimizer
33.44% :     0.004456s :      1: type_inference


TotalTime = 0.0163025, [33]
[bootstrap]: 0.00026233
[type_inference]: 0.00390527
[event_method]: 2.193e-05
[auto_monad]: 4.462e-05
[graph_reusing]: 3.81001e-06
[pre_auto_parallel]: 1.63002e-06
[py_interpret_to_execute]: 2.713e-05
[rewriter_before_opt_a]: 6.195e-05
[expand_dump_flag]: 2.03002e-06
[jit_opt_a]: 0.00466506, [2]
    [Cycle 1]: 0.00126098, [27]
        [switch_simplify]: 4.205e-05
        [loop_unroll]: 2.832e-05
        [a_1]: 0.00045227
        [with_stream_mark]: 1.247e-05
        [recompute_prepare]: 6.95002e-06
        [updatestate_depend_eliminate]: 2.76e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [parameter_eliminate]: 1.17999e-06
        [specialize_transform]: 5.81e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 6.01e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.098e-05
        [j_node_and_user_rematch]: 8.18999e-06
        [meta_fg_expand]: 1.88002e-06
        [replace_old_param]: 7.63001e-06
        [inline_without_move]: 4.93001e-06
        [renormalize]: 0.0004525
        [add_forward_monad_depend]: 4.97999e-06
        [auto_monad_grad]: 1.45999e-06
        [auto_monad_eliminator]: 1.116e-05
        [cse]: 1.663e-05
        [replace_applicator]: 1.026e-05
    [Cycle 2]: 0.00030329, [27]
        [switch_simplify]: 5.47999e-06
        [loop_unroll]: 4.60999e-06
        [a_1]: 8.598e-05
        [with_stream_mark]: 8.13999e-06
        [recompute_prepare]: 4.64002e-06
        [updatestate_depend_eliminate]: 2.43e-06
        [updatestate_assign_eliminate]: 1.94999e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.70999e-06
        [updatestate_useless_node_eliminater]: 4.55001e-06
        [accelerated_algorithm]: 4.79998e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.38999e-06
        [merge_forward]: 2.16e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.143e-05
        [j_node_and_user_rematch]: 7.25e-06
        [meta_fg_expand]: 1.48002e-06
        [replace_old_param]: 1.119e-05
        [inline_without_move]: 4.52e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.04998e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 4.84e-06
        [cse]: 9.87001e-06
        [replace_applicator]: 4.85001e-06
[py_interpret_to_execute_after_opt_a]: 7.73999e-06
[rewriter_after_opt_a]: 3.106e-05
[convert_after_rewriter]: 5.84999e-06
[order_py_execute_after_rewriter]: 4.82e-06
[mutable_eliminate]: 0.00049358
[jit_opt_b]: 4.636e-05, [1]
    [Cycle 1]: 4.036e-05, [2]
        [frontend_op_eliminate]: 1.534e-05
        [inline_after_opt_a]: 1.394e-05
[cconv]: 1.964e-05
[loop_unroll]: 0.0004401
[jit_opt_after_cconv]: 0.00012972, [1]
    [Cycle 1]: 0.00012356, [11]
        [c_1]: 1.938e-05
        [parameter_eliminate]: 2.36e-06
        [updatestate_depend_eliminate]: 4.98001e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.01998e-06
        [cse]: 1.494e-05
        [call_graph_tuple_transform]: 1.701e-05
        [tuple_list_get_item_eliminator]: 5.25001e-06
        [none_parameter_eliminate]: 1.29998e-06
        [renormalize]: 5.19998e-07
        [switch_simplify]: 5.02e-06
[remove_dup_value]: 1.267e-05
[partial_unused_args_eliminate]: 2.23998e-06
[environ_conv]: 4.94e-06
[add_recomputation]: 4.116e-05
[cse_after_recomputation]: 1.936e-05, [1]
    [Cycle 1]: 1.404e-05, [1]
        [cse]: 8.37e-06
[auto_monad_reorder]: 1.4e-05
[get_jit_bprop_graph]: 1.50001e-06
[rewriter_after_jit_bprop_graph]: 5.12e-06
[opt_after_jit_grad]: 0.00044685
[symbol_engine_optimizer]: 6.854e-05, [1]
    [Cycle 1]: 6.255e-05, [6]
        [build]: 2.84001e-06
        [elim_shapecalc]: 7.06001e-06
        [elim_not_effective]: 1.172e-05
        [opt_reshape]: 5.32001e-06
        [fold_const_symbol]: 8.45001e-06
        [renormalize]: 3.69997e-07
[validate]: 2.847e-05
[backend_pass]: 8.79983e-07
[task_emit]: 0.00530566
[execute]: 6.07999e-06
Sums
    bootstrap                            :     0.000262s :  2.07%
    type_inference                       :     0.003905s : 30.86%
    event_method                         :     0.000022s :  0.17%
    auto_monad                           :     0.000045s :  0.35%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000027s :  0.21%
    rewriter_before_opt_a                :     0.000062s :  0.49%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000048s :  0.38%
    jit_opt_a.loop_unroll                :     0.000033s :  0.26%
    jit_opt_a.a_1                        :     0.000538s :  4.25%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000015s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.15%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000453s :  3.58%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.13%
    jit_opt_a.cse                        :     0.000026s :  0.21%
    jit_opt_a.replace_applicator         :     0.000015s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.25%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000494s :  3.90%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000020s :  0.16%
    loop_unroll                          :     0.000440s :  3.48%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000041s :  0.33%
    cse_after_recomputation.cse          :     0.000008s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000447s :  3.53%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005306s : 41.93%
    execute                              :     0.000006s :  0.05%


Time group info:
------[substitution.]   0.000150    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.92% :     0.000001s :      2: substitution.fold_const_symbol
 3.05% :     0.000005s :      3: substitution.graph_param_transform
80.44% :     0.000121s :      4: substitution.inline
 1.65% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.55% :     0.000004s :      4: substitution.remove_not_recompute_node
 4.82% :     0.000007s :      2: substitution.replace_old_param
 5.50% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003849     2
81.97% :     0.003155s :      1: type_inference.infer
18.03% :     0.000694s :      1: type_inference.specialize
------[replace.]   0.000057     6
66.08% :     0.000037s :      4: replace.inline
33.92% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000126     6
94.24% :     0.000119s :      4: match.inline
 5.76% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000108   691
 1.48% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.46% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000001s :     10: predicate.addn_zero_filter
 2.11% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000001s :     10: predicate.depend_value_elim
 1.24% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.45% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.14% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000001s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.29% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.83% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.68% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000002s :     15: predicate.partial_eliminate
 1.40% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000001s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.57% :     0.000008s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.71% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.83% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.07% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000464    12
48.01% :     0.000223s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.47% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.52% :     0.000216s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017584    76
 0.25% :     0.000044s :      1: add_recomputation
 0.27% :     0.000048s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.55% :     0.000273s :      1: bootstrap
 0.13% :     0.000022s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000026s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
26.55% :     0.004668s :      1: jit_opt_a
 0.75% :     0.000132s :      1: jit_opt_after_cconv
 0.28% :     0.000049s :      1: jit_opt_b
 2.55% :     0.000448s :      1: loop_unroll
 2.85% :     0.000501s :      1: mutable_eliminate
 4.14% :     0.000729s :     26: opt.transform.jit_opt_a
 0.25% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.58% :     0.000454s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000003s :      1: pre_auto_parallel
 0.17% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.26% :     0.000222s :      1: renormalize.infer
 1.28% :     0.000225s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000034s :      1: rewriter_after_opt_a
 0.37% :     0.000065s :      1: rewriter_before_opt_a
 0.40% :     0.000071s :      1: symbol_engine_optimizer
30.23% :     0.005315s :      1: task_emit
22.27% :     0.003916s :      1: type_inference
 0.26% :     0.000046s :      1: validate


TotalTime = 0.02665, [33]
[bootstrap]: 0.00028366
[type_inference]: 0.00592208
[event_method]: 1.831e-05
[auto_monad]: 6.733e-05
[graph_reusing]: 6.02999e-06
[pre_auto_parallel]: 2.54999e-06
[py_interpret_to_execute]: 3.748e-05
[rewriter_before_opt_a]: 7.491e-05
[expand_dump_flag]: 3.27997e-06
[jit_opt_a]: 0.0061906, [2]
    [Cycle 1]: 0.00200957, [27]
        [switch_simplify]: 6.572e-05
        [loop_unroll]: 2.92e-05
        [a_1]: 0.00067938
        [with_stream_mark]: 2.475e-05
        [recompute_prepare]: 7.41999e-06
        [updatestate_depend_eliminate]: 4.2e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.69001e-06
        [parameter_eliminate]: 1.77001e-06
        [specialize_transform]: 6.28e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 2.19001e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 4.2e-06
        [cell_reuse_recompute_pass]: 1.53002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.517e-05
        [j_node_and_user_rematch]: 9.34998e-06
        [meta_fg_expand]: 2.14999e-06
        [replace_old_param]: 9.38002e-06
        [inline_without_move]: 5.97001e-06
        [renormalize]: 0.00085576
        [add_forward_monad_depend]: 8.43001e-06
        [auto_monad_grad]: 2.44999e-06
        [auto_monad_eliminator]: 1.58e-05
        [cse]: 3.041e-05
        [replace_applicator]: 1.435e-05
    [Cycle 2]: 0.00032379, [27]
        [switch_simplify]: 5.60001e-06
        [loop_unroll]: 4.83001e-06
        [a_1]: 9.167e-05
        [with_stream_mark]: 9.79e-06
        [recompute_prepare]: 5.48002e-06
        [updatestate_depend_eliminate]: 2.47001e-06
        [updatestate_assign_eliminate]: 2.51998e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.74002e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 5.67001e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 2.88998e-06
        [cell_reuse_recompute_pass]: 2.42001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.536e-05
        [j_node_and_user_rematch]: 7.87e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 7.3e-06
        [inline_without_move]: 6.98998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.45999e-06
        [auto_monad_grad]: 8.60018e-07
        [auto_monad_eliminator]: 4.88001e-06
        [cse]: 9.81e-06
        [replace_applicator]: 5.12e-06
[py_interpret_to_execute_after_opt_a]: 1.338e-05
[rewriter_after_opt_a]: 3.677e-05
[convert_after_rewriter]: 6.91001e-06
[order_py_execute_after_rewriter]: 5.61e-06
[mutable_eliminate]: 0.00070704
[jit_opt_b]: 5.058e-05, [1]
    [Cycle 1]: 4.302e-05, [2]
        [frontend_op_eliminate]: 1.638e-05
        [inline_after_opt_a]: 1.492e-05
[cconv]: 2.878e-05
[loop_unroll]: 0.00041688
[jit_opt_after_cconv]: 0.00014204, [1]
    [Cycle 1]: 0.00013543, [11]
        [c_1]: 2.106e-05
        [parameter_eliminate]: 2.87002e-06
        [updatestate_depend_eliminate]: 6.91999e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [cse]: 1.903e-05
        [call_graph_tuple_transform]: 1.886e-05
        [tuple_list_get_item_eliminator]: 4.89e-06
        [none_parameter_eliminate]: 1.42999e-06
        [renormalize]: 5.59987e-07
        [switch_simplify]: 5.17e-06
[remove_dup_value]: 1.414e-05
[partial_unused_args_eliminate]: 2.07001e-06
[environ_conv]: 5.17e-06
[add_recomputation]: 4.787e-05
[cse_after_recomputation]: 2.003e-05, [1]
    [Cycle 1]: 1.457e-05, [1]
        [cse]: 8.89998e-06
[auto_monad_reorder]: 1.441e-05
[get_jit_bprop_graph]: 1.86e-06
[rewriter_after_jit_bprop_graph]: 6.08002e-06
[opt_after_jit_grad]: 0.00044916
[symbol_engine_optimizer]: 0.0001115, [1]
    [Cycle 1]: 0.00010461, [6]
        [build]: 4.05e-06
        [elim_shapecalc]: 1.017e-05
        [elim_not_effective]: 1.31e-05
        [opt_reshape]: 5.74999e-06
        [fold_const_symbol]: 8.70001e-06
        [renormalize]: 4.80009e-07
[validate]: 4.014e-05
[backend_pass]: 9.49978e-07
[task_emit]: 0.0116931
[execute]: 1.057e-05
Sums
    bootstrap                            :     0.000284s :  1.28%
    type_inference                       :     0.005922s : 26.78%
    event_method                         :     0.000018s :  0.08%
    auto_monad                           :     0.000067s :  0.30%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000037s :  0.17%
    rewriter_before_opt_a                :     0.000075s :  0.34%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000071s :  0.32%
    jit_opt_a.loop_unroll                :     0.000034s :  0.15%
    jit_opt_a.a_1                        :     0.000771s :  3.49%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.08%
    jit_opt_a.inline_without_move        :     0.000013s :  0.06%
    jit_opt_a.renormalize                :     0.000856s :  3.87%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.09%
    jit_opt_a.cse                        :     0.000040s :  0.18%
    jit_opt_a.replace_applicator         :     0.000019s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.06%
    rewriter_after_opt_a                 :     0.000037s :  0.17%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000707s :  3.20%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.07%
    cconv                                :     0.000029s :  0.13%
    loop_unroll                          :     0.000417s :  1.89%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000014s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000048s :  0.22%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000014s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000449s :  2.03%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000040s :  0.18%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011693s : 52.87%
    execute                              :     0.000011s :  0.05%


Time group info:
------[substitution.]   0.000329    23
 0.65% :     0.000002s :      2: substitution.elim_not_effective
 0.40% :     0.000001s :      2: substitution.fold_const_symbol
 1.71% :     0.000006s :      3: substitution.graph_param_transform
89.14% :     0.000293s :      4: substitution.inline
 1.05% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.49% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.47% :     0.000005s :      2: substitution.replace_old_param
 4.09% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005850     2
84.00% :     0.004914s :      1: type_inference.infer
16.00% :     0.000936s :      1: type_inference.specialize
------[replace.]   0.000075     6
65.43% :     0.000049s :      4: replace.inline
34.57% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000303     6
95.91% :     0.000291s :      4: match.inline
 4.09% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.30% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.80% :     0.000002s :     10: predicate.addn_zero_filter
 2.38% :     0.000003s :     10: predicate.arithmetic_simplify
 1.15% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.56% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.72% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.76% :     0.000007s :     22: predicate.inline
 1.10% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.56% :     0.000002s :      6: predicate.less_batch_normalization
 2.02% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.60% :     0.000002s :     15: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.88% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 2.05% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000000s :      3: predicate.opt_reshape
 2.42% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.76% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.20% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     10: predicate.same_eliminate
 0.84% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.17% :     0.000010s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.85% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.58% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000731    12
49.89% :     0.000365s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.27% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.84% :     0.000313s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.028606    76
 0.18% :     0.000051s :      1: add_recomputation
 0.25% :     0.000071s :      1: auto_monad
 0.06% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.03% :     0.000296s :      1: bootstrap
 0.11% :     0.000032s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000022s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
21.65% :     0.006194s :      1: jit_opt_a
 0.51% :     0.000145s :      1: jit_opt_after_cconv
 0.19% :     0.000053s :      1: jit_opt_b
 1.49% :     0.000425s :      1: loop_unroll
 2.51% :     0.000718s :      1: mutable_eliminate
 3.49% :     0.000999s :     26: opt.transform.jit_opt_a
 0.16% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000025s :      4: opt.transform.jit_opt_b
 0.04% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.60% :     0.000458s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.14% :     0.000040s :      1: py_interpret_to_execute
 0.05% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000016s :      1: remove_dup_value
 1.55% :     0.000444s :      1: renormalize.infer
 1.41% :     0.000404s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000040s :      1: rewriter_after_opt_a
 0.27% :     0.000078s :      1: rewriter_before_opt_a
 0.40% :     0.000114s :      1: symbol_engine_optimizer
40.94% :     0.011713s :      1: task_emit
20.77% :     0.005941s :      1: type_inference
 0.24% :     0.000068s :      1: validate


TotalTime = 0.0275645, [33]
[bootstrap]: 0.00018115
[type_inference]: 0.00914333
[event_method]: 3.835e-05
[auto_monad]: 7.696e-05
[graph_reusing]: 5.52999e-06
[pre_auto_parallel]: 2.50002e-06
[py_interpret_to_execute]: 2.092e-05
[rewriter_before_opt_a]: 8.198e-05
[expand_dump_flag]: 2.41998e-06
[jit_opt_a]: 0.0102123, [3]
    [Cycle 1]: 0.00503964, [27]
        [switch_simplify]: 5.579e-05
        [loop_unroll]: 3.769e-05
        [a_1]: 0.00077604
        [with_stream_mark]: 2.117e-05
        [recompute_prepare]: 1.776e-05
        [updatestate_depend_eliminate]: 6.89999e-06
        [updatestate_assign_eliminate]: 5.52001e-06
        [updatestate_loads_eliminate]: 5.61e-06
        [parameter_eliminate]: 2.06998e-06
        [specialize_transform]: 1.284e-05
        [updatestate_useless_node_eliminater]: 1.148e-05
        [accelerated_algorithm]: 1.19e-05
        [meta_shard_fg_expand]: 2.73e-06
        [get_grad_eliminate_]: 1.139e-05
        [merge_forward]: 7.16999e-06
        [cell_reuse_recompute_pass]: 7.30011e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.34e-05
        [j_node_and_user_rematch]: 2.162e-05
        [meta_fg_expand]: 0.00138916
        [replace_old_param]: 5.032e-05
        [inline_without_move]: 5.011e-05
        [renormalize]: 0.0020754
        [add_forward_monad_depend]: 7.34002e-06
        [auto_monad_grad]: 4.29997e-06
        [auto_monad_eliminator]: 4.331e-05
        [cse]: 0.0001318
        [replace_applicator]: 6.189e-05
    [Cycle 2]: 0.00183911, [27]
        [switch_simplify]: 3.414e-05
        [loop_unroll]: 3.286e-05
        [a_1]: 0.00099878
        [with_stream_mark]: 1.177e-05
        [recompute_prepare]: 7.5e-06
        [updatestate_depend_eliminate]: 3.28e-06
        [updatestate_assign_eliminate]: 2.84999e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 1.29998e-06
        [specialize_transform]: 6.51e-06
        [updatestate_useless_node_eliminater]: 5.93002e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 1.67001e-06
        [get_grad_eliminate_]: 5.96e-06
        [merge_forward]: 3.31999e-06
        [cell_reuse_recompute_pass]: 7.99977e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.194e-05
        [j_node_and_user_rematch]: 9.44998e-06
        [meta_fg_expand]: 3.273e-05
        [replace_old_param]: 9.15001e-06
        [inline_without_move]: 6.07001e-06
        [renormalize]: 0.00045203
        [add_forward_monad_depend]: 3.90998e-06
        [auto_monad_grad]: 1.12e-06
        [auto_monad_eliminator]: 1.122e-05
        [cse]: 1.973e-05
        [replace_applicator]: 1.223e-05
    [Cycle 3]: 0.000363, [27]
        [switch_simplify]: 6.77002e-06
        [loop_unroll]: 5.77001e-06
        [a_1]: 0.00011461
        [with_stream_mark]: 8.75001e-06
        [recompute_prepare]: 5.84e-06
        [updatestate_depend_eliminate]: 3.25e-06
        [updatestate_assign_eliminate]: 2.58998e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 8.40024e-07
        [specialize_transform]: 5.90002e-06
        [updatestate_useless_node_eliminater]: 5.56998e-06
        [accelerated_algorithm]: 5.61e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 6.11e-06
        [merge_forward]: 3.09001e-06
        [cell_reuse_recompute_pass]: 1.55001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.55e-05
        [j_node_and_user_rematch]: 9.73002e-06
        [meta_fg_expand]: 1.86003e-06
        [replace_old_param]: 8.42e-06
        [inline_without_move]: 5.61e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.21002e-06
        [auto_monad_grad]: 7.29982e-07
        [auto_monad_eliminator]: 7.15003e-06
        [cse]: 1.401e-05
        [replace_applicator]: 6.63e-06
[py_interpret_to_execute_after_opt_a]: 9.63997e-06
[rewriter_after_opt_a]: 3.024e-05
[convert_after_rewriter]: 6.36998e-06
[order_py_execute_after_rewriter]: 4.73001e-06
[mutable_eliminate]: 0.00050145
[jit_opt_b]: 5.388e-05, [1]
    [Cycle 1]: 4.674e-05, [2]
        [frontend_op_eliminate]: 1.851e-05
        [inline_after_opt_a]: 1.74e-05
[cconv]: 1.766e-05
[loop_unroll]: 0.00040731
[jit_opt_after_cconv]: 0.00015095, [1]
    [Cycle 1]: 0.0001447, [11]
        [c_1]: 2.405e-05
        [parameter_eliminate]: 2.24001e-06
        [updatestate_depend_eliminate]: 5.62001e-06
        [updatestate_assign_eliminate]: 2.87002e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [cse]: 2.129e-05
        [call_graph_tuple_transform]: 1.809e-05
        [tuple_list_get_item_eliminator]: 6.24001e-06
        [none_parameter_eliminate]: 8.70001e-07
        [renormalize]: 4.50003e-07
        [switch_simplify]: 6.96999e-06
[remove_dup_value]: 1.12e-05
[partial_unused_args_eliminate]: 1.81998e-06
[environ_conv]: 3.86999e-06
[add_recomputation]: 3.879e-05
[cse_after_recomputation]: 2.489e-05, [1]
    [Cycle 1]: 1.935e-05, [1]
        [cse]: 1.381e-05
[auto_monad_reorder]: 1.38e-05
[get_jit_bprop_graph]: 1.60999e-06
[rewriter_after_jit_bprop_graph]: 4.86002e-06
[opt_after_jit_grad]: 0.00044795
[symbol_engine_optimizer]: 7.474e-05, [1]
    [Cycle 1]: 6.883e-05, [6]
        [build]: 2.73e-06
        [elim_shapecalc]: 8.70999e-06
        [elim_not_effective]: 1.396e-05
        [opt_reshape]: 6.35997e-06
        [fold_const_symbol]: 9.97999e-06
        [renormalize]: 4.7998e-07
[validate]: 2.936e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00576524
[execute]: 4.75999e-06
Sums
    bootstrap                            :     0.000181s :  0.76%
    type_inference                       :     0.009143s : 38.37%
    event_method                         :     0.000038s :  0.16%
    auto_monad                           :     0.000077s :  0.32%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000021s :  0.09%
    rewriter_before_opt_a                :     0.000082s :  0.34%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000097s :  0.41%
    jit_opt_a.loop_unroll                :     0.000076s :  0.32%
    jit_opt_a.a_1                        :     0.001889s :  7.93%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000031s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000013s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000025s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000023s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.10%
    jit_opt_a.merge_forward              :     0.000014s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000051s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000041s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001424s :  5.97%
    jit_opt_a.replace_old_param          :     0.000068s :  0.28%
    jit_opt_a.inline_without_move        :     0.000062s :  0.26%
    jit_opt_a.renormalize                :     0.002527s : 10.61%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000006s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000062s :  0.26%
    jit_opt_a.cse                        :     0.000166s :  0.69%
    jit_opt_a.replace_applicator         :     0.000081s :  0.34%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.04%
    rewriter_after_opt_a                 :     0.000030s :  0.13%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000501s :  2.10%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.07%
    cconv                                :     0.000018s :  0.07%
    loop_unroll                          :     0.000407s :  1.71%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000011s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000039s :  0.16%
    cse_after_recomputation.cse          :     0.000014s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000448s :  1.88%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005765s : 24.19%
    execute                              :     0.000005s :  0.02%


Time group info:
------[substitution.]   0.000417   110
 0.50% :     0.000002s :      3: substitution.elim_not_effective
 0.36% :     0.000001s :      3: substitution.fold_const_symbol
 1.00% :     0.000004s :      4: substitution.graph_param_transform
62.03% :     0.000259s :     12: substitution.inline
 3.86% :     0.000016s :      2: substitution.inline_without_move
 1.87% :     0.000008s :     14: substitution.j_node_and_user_rematch
 2.36% :     0.000010s :      7: substitution.minmaximum_grad
 1.57% :     0.000007s :      5: substitution.partial_eliminate
 2.49% :     0.000010s :     14: substitution.remove_not_recompute_node
 5.82% :     0.000024s :     10: substitution.replace_applicator
 1.65% :     0.000007s :      7: substitution.replace_old_param
 0.60% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.70% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.15% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 8.04% :     0.000034s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009085     2
89.91% :     0.008168s :      1: type_inference.infer
10.09% :     0.000916s :      1: type_inference.specialize
------[replace.]   0.000133    19
63.73% :     0.000084s :     12: replace.inline
36.27% :     0.000048s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000269    19
93.82% :     0.000252s :     12: match.inline
 6.18% :     0.000017s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000352  2568
 1.66% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.47% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.44% :     0.000005s :     43: predicate.addn_check_dump
 1.67% :     0.000006s :     43: predicate.addn_zero_filter
 2.09% :     0.000007s :     43: predicate.arithmetic_simplify
 1.70% :     0.000006s :     43: predicate.cast_eliminate
 0.17% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.44% :     0.000005s :     43: predicate.compare_switch_simplify
 1.54% :     0.000005s :     43: predicate.depend_value_elim
 1.46% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.52% :     0.000005s :     43: predicate.dict_get_item_eliminator
 1.48% :     0.000005s :     43: predicate.dict_set_item_eliminator
 0.32% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.14% :     0.000001s :      4: predicate.elim_not_effective
 0.18% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.50% :     0.000005s :     43: predicate.environ_add_const_eliminate
 1.52% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.51% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.56% :     0.000006s :     43: predicate.environ_get_eliminate
 1.50% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.82% :     0.000003s :     19: predicate.get_grad_eliminate
 0.14% :     0.000001s :      4: predicate.graph_param_transform
 4.25% :     0.000015s :     70: predicate.inline
 1.83% :     0.000006s :     40: predicate.inline_without_move
 0.36% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.85% :     0.000003s :     19: predicate.less_batch_normalization
 1.79% :     0.000006s :     50: predicate.list_to_tuple_eliminator_
 1.93% :     0.000007s :     54: predicate.load_eliminater
 0.51% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.14% :     0.000011s :     85: predicate.loop_unroll_before_grad
 1.77% :     0.000006s :     47: predicate.make_slice_get_slice_eliminator
 1.44% :     0.000005s :     43: predicate.merge_addn
 1.51% :     0.000005s :     43: predicate.minmaximum_grad
 0.49% :     0.000002s :      4: predicate.mutable_eliminate
 0.20% :     0.000001s :      4: predicate.opt_reshape
 2.35% :     0.000008s :     54: predicate.partial_eliminate
 1.53% :     0.000005s :     43: predicate.print_const_string_wrapper
 2.11% :     0.000007s :     43: predicate.reduce_eliminate
 1.83% :     0.000006s :     50: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.57% :     0.000009s :     96: predicate.replace_applicator
 0.93% :     0.000003s :     40: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.70% :     0.000006s :     43: predicate.reshape_eliminate
 1.54% :     0.000005s :     43: predicate.row_tensor_add_zeros_like
 0.33% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.52% :     0.000005s :     43: predicate.same_eliminate
 0.47% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.36% :     0.000001s :      8: predicate.special_op_eliminate
 0.86% :     0.000003s :     19: predicate.specialize_transform
 1.64% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000005s :     43: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.71% :     0.000010s :     62: predicate.switch_defer_inline
 2.50% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.28% :     0.000022s :    151: predicate.switch_simplify
 1.49% :     0.000005s :     43: predicate.tile_eliminate
 1.53% :     0.000005s :     43: predicate.transpose_eliminate
 1.81% :     0.000006s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.78% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.02% :     0.000011s :     58: predicate.tuple_list_get_item_eliminator
 1.87% :     0.000007s :     43: predicate.tuple_list_set_item_eliminator
 1.84% :     0.000006s :     50: predicate.tuple_to_list_eliminator_
 1.91% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000010s :     73: predicate.updatestate_useless_node_eliminater
 1.91% :     0.000007s :     43: predicate.value_based_eliminate
 0.15% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.29% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001263    29
60.27% :     0.000761s :     13: func_graph_cloner_run.FuncGraphClonerGraph
39.73% :     0.000502s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.032646    91
 0.13% :     0.000042s :      1: add_recomputation
 0.25% :     0.000081s :      1: auto_monad
 0.05% :     0.000016s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.59% :     0.000193s :      1: bootstrap
 0.06% :     0.000020s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000027s :      1: cse_after_recomputation
 0.02% :     0.000006s :      1: environ_conv
 0.13% :     0.000042s :      1: event_method
 0.02% :     0.000008s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
31.29% :     0.010215s :      1: jit_opt_a
 0.47% :     0.000154s :      1: jit_opt_after_cconv
 0.17% :     0.000056s :      1: jit_opt_b
 1.27% :     0.000415s :      1: loop_unroll
 1.56% :     0.000509s :      1: mutable_eliminate
 7.50% :     0.002450s :     39: opt.transform.jit_opt_a
 0.16% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000030s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.40% :     0.000456s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000023s :      1: py_interpret_to_execute
 0.04% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000014s :      1: remove_dup_value
 4.08% :     0.001331s :      2: renormalize.infer
 3.63% :     0.001185s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000033s :      1: rewriter_after_opt_a
 0.26% :     0.000085s :      1: rewriter_before_opt_a
 0.24% :     0.000077s :      1: symbol_engine_optimizer
17.69% :     0.005776s :      1: task_emit
28.04% :     0.009154s :      1: type_inference
 0.15% :     0.000047s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.397.100 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dbf30d0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.397.645 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0119293, [30]
[bootstrap]: 0.00029723
[type_inference]: 0.00417074
[event_method]: 1.705e-05
[auto_monad]: 6.196e-05
[graph_reusing]: 5.24e-06
[pre_auto_parallel]: 2.19001e-06
[py_interpret_to_execute]: 3.307e-05
[rewriter_before_opt_a]: 6.959e-05
[expand_dump_flag]: 3.18e-06
[jit_opt_a]: 0.00519046, [2]
    [Cycle 1]: 0.00154057, [27]
        [switch_simplify]: 4.908e-05
        [loop_unroll]: 3.015e-05
        [a_1]: 0.00052385
        [with_stream_mark]: 1.704e-05
        [recompute_prepare]: 7.03998e-06
        [updatestate_depend_eliminate]: 3.3e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 2.00002e-06
        [specialize_transform]: 6.26998e-06
        [updatestate_useless_node_eliminater]: 5.18002e-06
        [accelerated_algorithm]: 6.33e-06
        [meta_shard_fg_expand]: 1.91998e-06
        [get_grad_eliminate_]: 5.66e-06
        [merge_forward]: 3.23e-06
        [cell_reuse_recompute_pass]: 8.59989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.536e-05
        [j_node_and_user_rematch]: 9.94999e-06
        [meta_fg_expand]: 2.45002e-06
        [replace_old_param]: 8.45001e-06
        [inline_without_move]: 5.50001e-06
        [renormalize]: 0.00061472
        [add_forward_monad_depend]: 5.15001e-06
        [auto_monad_grad]: 2.26e-06
        [auto_monad_eliminator]: 1.258e-05
        [cse]: 2.192e-05
        [replace_applicator]: 1.146e-05
    [Cycle 2]: 0.00032272, [27]
        [switch_simplify]: 5.61998e-06
        [loop_unroll]: 5.12e-06
        [a_1]: 9.14e-05
        [with_stream_mark]: 9.24e-06
        [recompute_prepare]: 5.19998e-06
        [updatestate_depend_eliminate]: 2.54001e-06
        [updatestate_assign_eliminate]: 1.94999e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 4.94e-06
        [updatestate_useless_node_eliminater]: 4.86002e-06
        [accelerated_algorithm]: 5.72001e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.54002e-06
        [merge_forward]: 2.58e-06
        [cell_reuse_recompute_pass]: 1.93997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.245e-05
        [j_node_and_user_rematch]: 8.15e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 7.16999e-06
        [inline_without_move]: 4.66002e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 5.08002e-06
        [cse]: 9.87999e-06
        [replace_applicator]: 5.07e-06
[py_interpret_to_execute_after_opt_a]: 9.43002e-06
[rewriter_after_opt_a]: 3.235e-05
[convert_after_rewriter]: 6.75002e-06
[order_py_execute_after_rewriter]: 5.03002e-06
[mutable_eliminate]: 0.0005335
[jit_opt_b]: 5.342e-05, [1]
    [Cycle 1]: 4.59e-05, [2]
        [frontend_op_eliminate]: 1.796e-05
        [inline_after_opt_a]: 1.557e-05
[cconv]: 2.495e-05
[loop_unroll]: 0.00041748
[jit_opt_after_cconv]: 0.00015324, [1]
    [Cycle 1]: 0.00014679, [11]
        [c_1]: 2.127e-05
        [parameter_eliminate]: 2.54999e-06
        [updatestate_depend_eliminate]: 5.14e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 1.778e-05
        [call_graph_tuple_transform]: 1.88e-05
        [tuple_list_get_item_eliminator]: 5.87001e-06
        [none_parameter_eliminate]: 1.31002e-06
        [renormalize]: 1.159e-05
        [switch_simplify]: 5.81e-06
[remove_dup_value]: 1.398e-05
[partial_unused_args_eliminate]: 2.40002e-06
[environ_conv]: 5.24998e-06
[add_recomputation]: 4.581e-05
[cse_after_recomputation]: 2.05e-05, [1]
    [Cycle 1]: 1.487e-05, [1]
        [cse]: 9.31e-06
[auto_monad_reorder]: 1.404e-05
[get_jit_bprop_graph]: 1.69998e-06
[rewriter_after_jit_bprop_graph]: 4.59002e-06
[opt_after_jit_grad]: 0.0004727
[symbol_engine_optimizer]: 7.363e-05, [1]
    [Cycle 1]: 6.624e-05, [6]
        [build]: 3.26001e-06
        [elim_shapecalc]: 7.36001e-06
        [elim_not_effective]: 1.275e-05
        [opt_reshape]: 5.64e-06
        [fold_const_symbol]: 8.82e-06
        [renormalize]: 5.60016e-07
[validate]: 3.193e-05
Sums
    bootstrap                            :     0.000297s :  3.70%
    type_inference                       :     0.004171s : 51.88%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000062s :  0.77%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.41%
    rewriter_before_opt_a                :     0.000070s :  0.87%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000055s :  0.68%
    jit_opt_a.loop_unroll                :     0.000035s :  0.44%
    jit_opt_a.a_1                        :     0.000615s :  7.65%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000615s :  7.65%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.22%
    jit_opt_a.cse                        :     0.000032s :  0.40%
    jit_opt_a.replace_applicator         :     0.000017s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.12%
    rewriter_after_opt_a                 :     0.000032s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000534s :  6.64%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.19%
    cconv                                :     0.000025s :  0.31%
    loop_unroll                          :     0.000417s :  5.19%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000018s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000012s :  0.14%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000046s :  0.57%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000014s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000473s :  5.88%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000032s :  0.40%


Time group info:
------[substitution.]   0.000190    23
 1.13% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.67% :     0.000005s :      3: substitution.graph_param_transform
83.55% :     0.000159s :      4: substitution.inline
 1.70% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.35% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000004s :      2: substitution.replace_old_param
 5.99% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004118     2
80.15% :     0.003301s :      1: type_inference.infer
19.85% :     0.000817s :      1: type_inference.specialize
------[replace.]   0.000064     6
67.52% :     0.000043s :      4: replace.inline
32.48% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000167     6
93.85% :     0.000156s :      4: match.inline
 6.15% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.37% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000002s :     10: predicate.addn_zero_filter
 2.10% :     0.000002s :     10: predicate.arithmetic_simplify
 1.74% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.20% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.25% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.55% :     0.000002s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.33% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.08% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.93% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 1.27% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.73% :     0.000002s :     10: predicate.reduce_eliminate
 1.61% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.11% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.14% :     0.000008s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.65% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.61% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000577    12
47.25% :     0.000273s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.41% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.33% :     0.000273s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013429    72
 0.36% :     0.000049s :      1: add_recomputation
 0.48% :     0.000065s :      1: auto_monad
 0.12% :     0.000017s :      1: auto_monad_reorder
 2.29% :     0.000308s :      1: bootstrap
 0.21% :     0.000028s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.67% :     0.005193s :      1: jit_opt_a
 1.16% :     0.000156s :      1: jit_opt_after_cconv
 0.42% :     0.000056s :      1: jit_opt_b
 3.17% :     0.000425s :      1: loop_unroll
 4.04% :     0.000542s :      1: mutable_eliminate
 6.12% :     0.000822s :     26: opt.transform.jit_opt_a
 0.36% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000027s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.58% :     0.000481s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000036s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.27% :     0.000305s :      1: renormalize.infer
 2.25% :     0.000302s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000035s :      1: rewriter_after_opt_a
 0.54% :     0.000072s :      1: rewriter_before_opt_a
 0.57% :     0.000076s :      1: symbol_engine_optimizer
31.14% :     0.004182s :      1: type_inference


TotalTime = 0.0173842, [33]
[bootstrap]: 0.00027035
[type_inference]: 0.00416179
[event_method]: 1.621e-05
[auto_monad]: 5.71e-05
[graph_reusing]: 5.991e-05
[pre_auto_parallel]: 2.73998e-06
[py_interpret_to_execute]: 2.953e-05
[rewriter_before_opt_a]: 6.878e-05
[expand_dump_flag]: 3.11999e-06
[jit_opt_a]: 0.00498251, [2]
    [Cycle 1]: 0.00134546, [27]
        [switch_simplify]: 4.743e-05
        [loop_unroll]: 2.825e-05
        [a_1]: 0.00048947
        [with_stream_mark]: 1.454e-05
        [recompute_prepare]: 9.50001e-06
        [updatestate_depend_eliminate]: 3.08e-06
        [updatestate_assign_eliminate]: 2.69001e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.61998e-06
        [specialize_transform]: 7.06999e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.66e-06
        [meta_shard_fg_expand]: 2.01998e-06
        [get_grad_eliminate_]: 5.15999e-06
        [merge_forward]: 3.68e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.308e-05
        [j_node_and_user_rematch]: 8.74e-06
        [meta_fg_expand]: 1.99999e-06
        [replace_old_param]: 8.07e-06
        [inline_without_move]: 5.22e-06
        [renormalize]: 0.00046987
        [add_forward_monad_depend]: 4.77998e-06
        [auto_monad_grad]: 2.06998e-06
        [auto_monad_eliminator]: 1.184e-05
        [cse]: 1.703e-05
        [replace_applicator]: 1.208e-05
    [Cycle 2]: 0.00031415, [27]
        [switch_simplify]: 5.94999e-06
        [loop_unroll]: 5.30001e-06
        [a_1]: 9.174e-05
        [with_stream_mark]: 8.25e-06
        [recompute_prepare]: 4.76002e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 1.86e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 9.99979e-07
        [specialize_transform]: 4.78001e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 4.68999e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.35999e-06
        [merge_forward]: 2.24999e-06
        [cell_reuse_recompute_pass]: 1.69e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.18e-05
        [j_node_and_user_rematch]: 7.36001e-06
        [meta_fg_expand]: 1.41998e-06
        [replace_old_param]: 7.26999e-06
        [inline_without_move]: 4.62998e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 1.00001e-06
        [auto_monad_eliminator]: 4.63001e-06
        [cse]: 9.29998e-06
        [replace_applicator]: 4.99e-06
[py_interpret_to_execute_after_opt_a]: 9.39e-06
[rewriter_after_opt_a]: 3.207e-05
[convert_after_rewriter]: 6.96999e-06
[order_py_execute_after_rewriter]: 5.37999e-06
[mutable_eliminate]: 0.00053198
[jit_opt_b]: 4.716e-05, [1]
    [Cycle 1]: 4.081e-05, [2]
        [frontend_op_eliminate]: 1.534e-05
        [inline_after_opt_a]: 1.431e-05
[cconv]: 2.256e-05
[loop_unroll]: 0.00041708
[jit_opt_after_cconv]: 0.00013, [1]
    [Cycle 1]: 0.00012376, [11]
        [c_1]: 2.003e-05
        [parameter_eliminate]: 2.76999e-06
        [updatestate_depend_eliminate]: 4.89998e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [cse]: 1.451e-05
        [call_graph_tuple_transform]: 1.72e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.13001e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 5.12999e-06
[remove_dup_value]: 1.173e-05
[partial_unused_args_eliminate]: 2.61e-06
[environ_conv]: 4.47e-06
[add_recomputation]: 4.425e-05
[cse_after_recomputation]: 1.925e-05, [1]
    [Cycle 1]: 1.41e-05, [1]
        [cse]: 8.19002e-06
[auto_monad_reorder]: 1.422e-05
[get_jit_bprop_graph]: 1.40999e-06
[rewriter_after_jit_bprop_graph]: 5.95002e-06
[opt_after_jit_grad]: 0.00047249
[symbol_engine_optimizer]: 6.906e-05, [1]
    [Cycle 1]: 6.258e-05, [6]
        [build]: 3.06999e-06
        [elim_shapecalc]: 7.18998e-06
        [elim_not_effective]: 1.154e-05
        [opt_reshape]: 5.20999e-06
        [fold_const_symbol]: 8.26002e-06
        [renormalize]: 5.10016e-07
[validate]: 3.083e-05
[backend_pass]: 8.10018e-07
[task_emit]: 0.00567106
[execute]: 7e-06
Sums
    bootstrap                            :     0.000270s :  2.00%
    type_inference                       :     0.004162s : 30.84%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000057s :  0.42%
    graph_reusing                        :     0.000060s :  0.44%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.22%
    rewriter_before_opt_a                :     0.000069s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000581s :  4.31%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000470s :  3.48%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.12%
    jit_opt_a.cse                        :     0.000026s :  0.20%
    jit_opt_a.replace_applicator         :     0.000017s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000532s :  3.94%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000417s :  3.09%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000044s :  0.33%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000472s :  3.50%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000031s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005671s : 42.03%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000163    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.79% :     0.000001s :      2: substitution.fold_const_symbol
 2.95% :     0.000005s :      3: substitution.graph_param_transform
81.77% :     0.000133s :      4: substitution.inline
 1.76% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.50% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.26% :     0.000004s :      2: substitution.replace_old_param
 6.87% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004104     2
82.55% :     0.003388s :      1: type_inference.infer
17.45% :     0.000716s :      1: type_inference.specialize
------[replace.]   0.000060     6
62.24% :     0.000037s :      4: replace.inline
37.76% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000141     6
92.80% :     0.000131s :      4: match.inline
 7.20% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.59% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.46% :     0.000002s :     10: predicate.addn_zero_filter
 2.11% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.60% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.66% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.38% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.63% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000001s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.53% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000002s :     15: predicate.partial_eliminate
 1.54% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.59% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.52% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.54% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.59% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.41% :     0.000008s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.09% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.50% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.05% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000467    12
48.45% :     0.000226s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.45% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.10% :     0.000215s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018733    76
 0.25% :     0.000047s :      1: add_recomputation
 0.32% :     0.000060s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.51% :     0.000282s :      1: bootstrap
 0.13% :     0.000025s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.33% :     0.000063s :      1: graph_reusing
26.61% :     0.004985s :      1: jit_opt_a
 0.71% :     0.000133s :      1: jit_opt_after_cconv
 0.26% :     0.000050s :      1: jit_opt_b
 2.27% :     0.000425s :      1: loop_unroll
 2.88% :     0.000540s :      1: mutable_eliminate
 4.17% :     0.000781s :     26: opt.transform.jit_opt_a
 0.24% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.56% :     0.000480s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000014s :      1: remove_dup_value
 1.27% :     0.000237s :      1: renormalize.infer
 1.21% :     0.000226s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000035s :      1: rewriter_after_opt_a
 0.38% :     0.000071s :      1: rewriter_before_opt_a
 0.38% :     0.000071s :      1: symbol_engine_optimizer
30.33% :     0.005682s :      1: task_emit
22.27% :     0.004172s :      1: type_inference
 0.26% :     0.000048s :      1: validate


TotalTime = 0.0290586, [33]
[bootstrap]: 0.00017161
[type_inference]: 0.0092126
[event_method]: 4.079e-05
[auto_monad]: 8.132e-05
[graph_reusing]: 6.46999e-06
[pre_auto_parallel]: 2.15002e-06
[py_interpret_to_execute]: 2.169e-05
[rewriter_before_opt_a]: 8.329e-05
[expand_dump_flag]: 2.88998e-06
[jit_opt_a]: 0.0107319, [3]
    [Cycle 1]: 0.00537614, [27]
        [switch_simplify]: 5.508e-05
        [loop_unroll]: 3.945e-05
        [a_1]: 0.00078133
        [with_stream_mark]: 2.301e-05
        [recompute_prepare]: 1.725e-05
        [updatestate_depend_eliminate]: 7.01001e-06
        [updatestate_assign_eliminate]: 5.73997e-06
        [updatestate_loads_eliminate]: 5.39e-06
        [parameter_eliminate]: 2.44999e-06
        [specialize_transform]: 1.263e-05
        [updatestate_useless_node_eliminater]: 1.14e-05
        [accelerated_algorithm]: 1.193e-05
        [meta_shard_fg_expand]: 3.02002e-06
        [get_grad_eliminate_]: 1.144e-05
        [merge_forward]: 6.59001e-06
        [cell_reuse_recompute_pass]: 8.89995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.359e-05
        [j_node_and_user_rematch]: 2.197e-05
        [meta_fg_expand]: 0.00141369
        [replace_old_param]: 5.428e-05
        [inline_without_move]: 5.177e-05
        [renormalize]: 0.00233531
        [add_forward_monad_depend]: 1.123e-05
        [auto_monad_grad]: 5.56e-06
        [auto_monad_eliminator]: 4.915e-05
        [cse]: 0.00014248
        [replace_applicator]: 6.842e-05
    [Cycle 2]: 0.00197208, [27]
        [switch_simplify]: 3.421e-05
        [loop_unroll]: 3.395e-05
        [a_1]: 0.00100812
        [with_stream_mark]: 1.721e-05
        [recompute_prepare]: 8.28001e-06
        [updatestate_depend_eliminate]: 3.8e-06
        [updatestate_assign_eliminate]: 3.8e-06
        [updatestate_loads_eliminate]: 2.93e-06
        [parameter_eliminate]: 2.06998e-06
        [specialize_transform]: 6.46e-06
        [updatestate_useless_node_eliminater]: 6.19001e-06
        [accelerated_algorithm]: 5.92001e-06
        [meta_shard_fg_expand]: 2.63998e-06
        [get_grad_eliminate_]: 6.28e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.598e-05
        [j_node_and_user_rematch]: 1.052e-05
        [meta_fg_expand]: 4.019e-05
        [replace_old_param]: 9.86e-06
        [inline_without_move]: 6.09001e-06
        [renormalize]: 0.0005359
        [add_forward_monad_depend]: 4.77e-06
        [auto_monad_grad]: 2.43e-06
        [auto_monad_eliminator]: 1.306e-05
        [cse]: 2.327e-05
        [replace_applicator]: 1.318e-05
    [Cycle 3]: 0.00035493, [27]
        [switch_simplify]: 6.38e-06
        [loop_unroll]: 6.26e-06
        [a_1]: 0.00011515
        [with_stream_mark]: 9.17001e-06
        [recompute_prepare]: 5.87999e-06
        [updatestate_depend_eliminate]: 3.58e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.40002e-06
        [parameter_eliminate]: 9.10019e-07
        [specialize_transform]: 5.91998e-06
        [updatestate_useless_node_eliminater]: 5.79e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 1.21002e-06
        [get_grad_eliminate_]: 5.63997e-06
        [merge_forward]: 3.21001e-06
        [cell_reuse_recompute_pass]: 1.60001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.476e-05
        [j_node_and_user_rematch]: 9.21002e-06
        [meta_fg_expand]: 1.80001e-06
        [replace_old_param]: 7.68999e-06
        [inline_without_move]: 5.62001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.08001e-06
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 6.39001e-06
        [cse]: 1.331e-05
        [replace_applicator]: 5.77001e-06
[py_interpret_to_execute_after_opt_a]: 1.329e-05
[rewriter_after_opt_a]: 3.929e-05
[convert_after_rewriter]: 7.79002e-06
[order_py_execute_after_rewriter]: 5.35001e-06
[mutable_eliminate]: 0.00068615
[jit_opt_b]: 5.632e-05, [1]
    [Cycle 1]: 4.77e-05, [2]
        [frontend_op_eliminate]: 1.924e-05
        [inline_after_opt_a]: 1.74e-05
[cconv]: 2.503e-05
[loop_unroll]: 0.00042188
[jit_opt_after_cconv]: 0.00015448, [1]
    [Cycle 1]: 0.00014832, [11]
        [c_1]: 2.524e-05
        [parameter_eliminate]: 2.53e-06
        [updatestate_depend_eliminate]: 5.57001e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [cse]: 2.146e-05
        [call_graph_tuple_transform]: 1.986e-05
        [tuple_list_get_item_eliminator]: 1.082e-05
        [none_parameter_eliminate]: 1.86e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 6.61999e-06
[remove_dup_value]: 1.648e-05
[partial_unused_args_eliminate]: 2.24001e-06
[environ_conv]: 5.35999e-06
[add_recomputation]: 4.909e-05
[cse_after_recomputation]: 2.481e-05, [1]
    [Cycle 1]: 1.924e-05, [1]
        [cse]: 1.38e-05
[auto_monad_reorder]: 1.868e-05
[get_jit_bprop_graph]: 1.72001e-06
[rewriter_after_jit_bprop_graph]: 6.06e-06
[opt_after_jit_grad]: 0.000449
[symbol_engine_optimizer]: 7.776e-05, [1]
    [Cycle 1]: 7.133e-05, [6]
        [build]: 3.98999e-06
        [elim_shapecalc]: 8.94998e-06
        [elim_not_effective]: 1.394e-05
        [opt_reshape]: 6.61e-06
        [fold_const_symbol]: 1.082e-05
        [renormalize]: 5.09986e-07
[validate]: 3.26e-05
[backend_pass]: 8.80013e-07
[task_emit]: 0.00642311
[execute]: 6.50002e-06
Sums
    bootstrap                            :     0.000172s :  0.68%
    type_inference                       :     0.009213s : 36.46%
    event_method                         :     0.000041s :  0.16%
    auto_monad                           :     0.000081s :  0.32%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000022s :  0.09%
    rewriter_before_opt_a                :     0.000083s :  0.33%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000096s :  0.38%
    jit_opt_a.loop_unroll                :     0.000080s :  0.32%
    jit_opt_a.a_1                        :     0.001905s :  7.54%
    jit_opt_a.with_stream_mark           :     0.000049s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000031s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000025s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.09%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000054s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001456s :  5.76%
    jit_opt_a.replace_old_param          :     0.000072s :  0.28%
    jit_opt_a.inline_without_move        :     0.000063s :  0.25%
    jit_opt_a.renormalize                :     0.002871s : 11.36%
    jit_opt_a.add_forward_monad_depend   :     0.000017s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000069s :  0.27%
    jit_opt_a.cse                        :     0.000179s :  0.71%
    jit_opt_a.replace_applicator         :     0.000087s :  0.35%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.05%
    rewriter_after_opt_a                 :     0.000039s :  0.16%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000686s :  2.72%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.07%
    cconv                                :     0.000025s :  0.10%
    loop_unroll                          :     0.000422s :  1.67%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000011s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000016s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000049s :  0.19%
    cse_after_recomputation.cse          :     0.000014s :  0.05%
    auto_monad_reorder                   :     0.000019s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000449s :  1.78%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000033s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006423s : 25.42%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000456   110
 0.48% :     0.000002s :      3: substitution.elim_not_effective
 0.36% :     0.000002s :      3: substitution.fold_const_symbol
 1.25% :     0.000006s :      4: substitution.graph_param_transform
61.86% :     0.000282s :     12: substitution.inline
 3.71% :     0.000017s :      2: substitution.inline_without_move
 2.02% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.27% :     0.000010s :      7: substitution.minmaximum_grad
 1.36% :     0.000006s :      5: substitution.partial_eliminate
 2.40% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.89% :     0.000027s :     10: substitution.replace_applicator
 1.63% :     0.000007s :      7: substitution.replace_old_param
 0.56% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.81% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.03% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 8.36% :     0.000038s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009156     2
89.90% :     0.008231s :      1: type_inference.infer
10.10% :     0.000925s :      1: type_inference.specialize
------[replace.]   0.000131    19
62.90% :     0.000083s :     12: replace.inline
37.10% :     0.000049s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000294    19
93.75% :     0.000275s :     12: match.inline
 6.25% :     0.000018s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000359  2568
 1.71% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.42% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.42% :     0.000005s :     43: predicate.addn_check_dump
 1.90% :     0.000007s :     43: predicate.addn_zero_filter
 2.04% :     0.000007s :     43: predicate.arithmetic_simplify
 1.72% :     0.000006s :     43: predicate.cast_eliminate
 0.17% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.43% :     0.000005s :     43: predicate.compare_switch_simplify
 1.57% :     0.000006s :     43: predicate.depend_value_elim
 1.47% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.47% :     0.000005s :     43: predicate.dict_get_item_eliminator
 1.51% :     0.000005s :     43: predicate.dict_set_item_eliminator
 0.29% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      4: predicate.elim_not_effective
 0.21% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000005s :     43: predicate.environ_add_const_eliminate
 1.45% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.44% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.44% :     0.000005s :     43: predicate.environ_get_eliminate
 1.46% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.95% :     0.000003s :     19: predicate.get_grad_eliminate
 0.16% :     0.000001s :      4: predicate.graph_param_transform
 4.24% :     0.000015s :     70: predicate.inline
 1.80% :     0.000006s :     40: predicate.inline_without_move
 0.36% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.83% :     0.000003s :     19: predicate.less_batch_normalization
 1.91% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.90% :     0.000007s :     54: predicate.load_eliminater
 0.45% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.36% :     0.000012s :     85: predicate.loop_unroll_before_grad
 1.91% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.41% :     0.000005s :     43: predicate.merge_addn
 1.47% :     0.000005s :     43: predicate.minmaximum_grad
 0.67% :     0.000002s :      4: predicate.mutable_eliminate
 0.20% :     0.000001s :      4: predicate.opt_reshape
 2.29% :     0.000008s :     54: predicate.partial_eliminate
 1.53% :     0.000006s :     43: predicate.print_const_string_wrapper
 2.01% :     0.000007s :     43: predicate.reduce_eliminate
 1.87% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.61% :     0.000009s :     96: predicate.replace_applicator
 1.05% :     0.000004s :     40: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.82% :     0.000007s :     43: predicate.reshape_eliminate
 1.51% :     0.000005s :     43: predicate.row_tensor_add_zeros_like
 0.33% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.58% :     0.000006s :     43: predicate.same_eliminate
 0.46% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.43% :     0.000002s :      8: predicate.special_op_eliminate
 0.87% :     0.000003s :     19: predicate.specialize_transform
 1.59% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000005s :     43: predicate.stack_unstack_eliminate
 0.18% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.61% :     0.000009s :     62: predicate.switch_defer_inline
 2.50% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.18% :     0.000022s :    151: predicate.switch_simplify
 1.48% :     0.000005s :     43: predicate.tile_eliminate
 1.49% :     0.000005s :     43: predicate.transpose_eliminate
 1.75% :     0.000006s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.59% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.08% :     0.000011s :     58: predicate.tuple_list_get_item_eliminator
 1.87% :     0.000007s :     43: predicate.tuple_list_set_item_eliminator
 1.82% :     0.000007s :     50: predicate.tuple_to_list_eliminator_
 1.89% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000010s :     73: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000006s :     43: predicate.value_based_eliminate
 0.17% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.23% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001287    29
61.06% :     0.000786s :     13: func_graph_cloner_run.FuncGraphClonerGraph
38.94% :     0.000501s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.034543    91
 0.15% :     0.000052s :      1: add_recomputation
 0.25% :     0.000085s :      1: auto_monad
 0.06% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.53% :     0.000183s :      1: bootstrap
 0.08% :     0.000028s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000027s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.13% :     0.000045s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
31.08% :     0.010736s :      1: jit_opt_a
 0.46% :     0.000157s :      1: jit_opt_after_cconv
 0.17% :     0.000059s :      1: jit_opt_b
 1.24% :     0.000429s :      1: loop_unroll
 2.01% :     0.000694s :      1: mutable_eliminate
 7.19% :     0.002484s :     39: opt.transform.jit_opt_a
 0.17% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000030s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.32% :     0.000456s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000024s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000019s :      1: remove_dup_value
 4.54% :     0.001568s :      2: renormalize.infer
 3.73% :     0.001289s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000042s :      1: rewriter_after_opt_a
 0.25% :     0.000086s :      1: rewriter_before_opt_a
 0.23% :     0.000080s :      1: symbol_engine_optimizer
18.63% :     0.006434s :      1: task_emit
26.70% :     0.009224s :      1: type_inference
 0.15% :     0.000052s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.435.118 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dbf34c0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.435.677 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0185307, [30]
[bootstrap]: 0.00033104
[type_inference]: 0.00602352
[event_method]: 4.219e-05
[auto_monad]: 0.00011444
[graph_reusing]: 7.75e-06
[pre_auto_parallel]: 3.70998e-06
[py_interpret_to_execute]: 0.00012831
[rewriter_before_opt_a]: 0.00011615
[expand_dump_flag]: 4.53001e-06
[jit_opt_a]: 0.00839261, [2]
    [Cycle 1]: 0.00223337, [27]
        [switch_simplify]: 7.604e-05
        [loop_unroll]: 3.282e-05
        [a_1]: 0.00069678
        [with_stream_mark]: 2.926e-05
        [recompute_prepare]: 1.025e-05
        [updatestate_depend_eliminate]: 4.38999e-06
        [updatestate_assign_eliminate]: 3.11001e-06
        [updatestate_loads_eliminate]: 3.03998e-06
        [parameter_eliminate]: 3.08998e-06
        [specialize_transform]: 6.17001e-06
        [updatestate_useless_node_eliminater]: 5.40001e-06
        [accelerated_algorithm]: 6.28e-06
        [meta_shard_fg_expand]: 2.53e-06
        [get_grad_eliminate_]: 6.29999e-06
        [merge_forward]: 4.43001e-06
        [cell_reuse_recompute_pass]: 1.60999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.916e-05
        [j_node_and_user_rematch]: 9.89001e-06
        [meta_fg_expand]: 2.19001e-06
        [replace_old_param]: 1.11e-05
        [inline_without_move]: 5.57999e-06
        [renormalize]: 0.0009719
        [add_forward_monad_depend]: 1.011e-05
        [auto_monad_grad]: 2.81999e-06
        [auto_monad_eliminator]: 2.222e-05
        [cse]: 3.736e-05
        [replace_applicator]: 2.386e-05
    [Cycle 2]: 0.00038177, [27]
        [switch_simplify]: 5.51e-06
        [loop_unroll]: 5.32999e-06
        [a_1]: 9.948e-05
        [with_stream_mark]: 1.827e-05
        [recompute_prepare]: 5.86e-06
        [updatestate_depend_eliminate]: 3.88001e-06
        [updatestate_assign_eliminate]: 2.30002e-06
        [updatestate_loads_eliminate]: 2.53003e-06
        [parameter_eliminate]: 1.62999e-06
        [specialize_transform]: 5.74e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.98998e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 5.07999e-06
        [merge_forward]: 4.32e-06
        [cell_reuse_recompute_pass]: 3.25998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.052e-05
        [j_node_and_user_rematch]: 1e-05
        [meta_fg_expand]: 2.05002e-06
        [replace_old_param]: 8.98002e-06
        [inline_without_move]: 5.19e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.64e-06
        [auto_monad_grad]: 1.70001e-06
        [auto_monad_eliminator]: 7.51001e-06
        [cse]: 1.479e-05
        [replace_applicator]: 6.18002e-06
[py_interpret_to_execute_after_opt_a]: 1.888e-05
[rewriter_after_opt_a]: 4.375e-05
[convert_after_rewriter]: 8.99e-06
[order_py_execute_after_rewriter]: 5.89e-06
[mutable_eliminate]: 0.00102671
[jit_opt_b]: 6.955e-05, [1]
    [Cycle 1]: 5.635e-05, [2]
        [frontend_op_eliminate]: 2.08e-05
        [inline_after_opt_a]: 2.135e-05
[cconv]: 4.319e-05
[loop_unroll]: 0.0006705
[jit_opt_after_cconv]: 0.00020037, [1]
    [Cycle 1]: 0.00019019, [11]
        [c_1]: 2.278e-05
        [parameter_eliminate]: 6.79999e-06
        [updatestate_depend_eliminate]: 1.117e-05
        [updatestate_assign_eliminate]: 3.53999e-06
        [updatestate_loads_eliminate]: 3.38999e-06
        [cse]: 3.77e-05
        [call_graph_tuple_transform]: 3.072e-05
        [tuple_list_get_item_eliminator]: 6.68998e-06
        [none_parameter_eliminate]: 1.90001e-06
        [renormalize]: 8.29983e-07
        [switch_simplify]: 6.20002e-06
[remove_dup_value]: 1.746e-05
[partial_unused_args_eliminate]: 3.38e-06
[environ_conv]: 7.02002e-06
[add_recomputation]: 6.225e-05
[cse_after_recomputation]: 2.828e-05, [1]
    [Cycle 1]: 2.015e-05, [1]
        [cse]: 1.217e-05
[auto_monad_reorder]: 1.926e-05
[get_jit_bprop_graph]: 2.09999e-06
[rewriter_after_jit_bprop_graph]: 1.095e-05
[opt_after_jit_grad]: 0.00068803
[symbol_engine_optimizer]: 9.203e-05, [1]
    [Cycle 1]: 8.241e-05, [6]
        [build]: 8.04002e-06
        [elim_shapecalc]: 8.90001e-06
        [elim_not_effective]: 1.867e-05
        [opt_reshape]: 6.54999e-06
        [fold_const_symbol]: 9.12001e-06
        [renormalize]: 1.03001e-06
[validate]: 5.516e-05
Sums
    bootstrap                            :     0.000331s :  2.77%
    type_inference                       :     0.006024s : 50.38%
    event_method                         :     0.000042s :  0.35%
    auto_monad                           :     0.000114s :  0.96%
    graph_reusing                        :     0.000008s :  0.06%
    pre_auto_parallel                    :     0.000004s :  0.03%
    py_interpret_to_execute              :     0.000128s :  1.07%
    rewriter_before_opt_a                :     0.000116s :  0.97%
    expand_dump_flag                     :     0.000005s :  0.04%
    jit_opt_a.switch_simplify            :     0.000082s :  0.68%
    jit_opt_a.loop_unroll                :     0.000038s :  0.32%
    jit_opt_a.a_1                        :     0.000796s :  6.66%
    jit_opt_a.with_stream_mark           :     0.000048s :  0.40%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.10%
    jit_opt_a.merge_forward              :     0.000009s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000020s :  0.17%
    jit_opt_a.inline_without_move        :     0.000011s :  0.09%
    jit_opt_a.renormalize                :     0.000972s :  8.13%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000030s :  0.25%
    jit_opt_a.cse                        :     0.000052s :  0.44%
    jit_opt_a.replace_applicator         :     0.000030s :  0.25%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.16%
    rewriter_after_opt_a                 :     0.000044s :  0.37%
    convert_after_rewriter               :     0.000009s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.001027s :  8.59%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000021s :  0.18%
    cconv                                :     0.000043s :  0.36%
    loop_unroll                          :     0.000670s :  5.61%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.06%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.09%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000038s :  0.32%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000031s :  0.26%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000017s :  0.15%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000007s :  0.06%
    add_recomputation                    :     0.000062s :  0.52%
    cse_after_recomputation.cse          :     0.000012s :  0.10%
    auto_monad_reorder                   :     0.000019s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.09%
    opt_after_jit_grad                   :     0.000688s :  5.75%
    symbol_engine_optimizer.build        :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000055s :  0.46%


Time group info:
------[substitution.]   0.000312    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 2.20% :     0.000007s :      3: substitution.graph_param_transform
86.02% :     0.000269s :      4: substitution.inline
 1.63% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.07% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.16% :     0.000007s :      2: substitution.replace_old_param
 4.69% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005846     2
61.26% :     0.003582s :      1: type_inference.infer
38.74% :     0.002265s :      1: type_inference.specialize
------[replace.]   0.000086     6
68.13% :     0.000058s :      4: replace.inline
31.87% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000278     6
95.23% :     0.000265s :      4: match.inline
 4.77% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000150   691
 0.97% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.41% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.88% :     0.000001s :     10: predicate.addn_check_dump
 1.60% :     0.000002s :     10: predicate.addn_zero_filter
 2.88% :     0.000004s :     10: predicate.arithmetic_simplify
 1.11% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.90% :     0.000001s :     10: predicate.compare_switch_simplify
 1.53% :     0.000002s :     10: predicate.depend_value_elim
 0.99% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.27% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.59% :     0.000001s :      3: predicate.elim_not_effective
 0.83% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.01% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     10: predicate.environ_get_eliminate
 1.05% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000001s :      3: predicate.graph_param_transform
 6.51% :     0.000010s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.51% :     0.000002s :      6: predicate.less_batch_normalization
 2.09% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.43% :     0.000002s :     15: predicate.load_eliminater
 2.42% :     0.000004s :      3: predicate.loop_unroll_after_grad
 3.69% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.85% :     0.000001s :     10: predicate.merge_addn
 0.87% :     0.000001s :     10: predicate.minmaximum_grad
 4.24% :     0.000006s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 1.97% :     0.000003s :     15: predicate.partial_eliminate
 1.05% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.36% :     0.000002s :     10: predicate.reduce_eliminate
 1.13% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.34% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000001s :      3: predicate.reset_defer_inline
 1.16% :     0.000002s :     10: predicate.reshape_eliminate
 1.09% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.46% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 0.81% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000004s :     16: predicate.switch_defer_inline
 1.81% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.09% :     0.000011s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000002s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.78% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.28% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.44% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.90% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000925    12
35.58% :     0.000329s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.58% :     0.000061s :      1: func_graph_cloner_run.FuncGraphClonerNode
57.84% :     0.000535s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020634    72
 0.33% :     0.000067s :      1: add_recomputation
 0.58% :     0.000120s :      1: auto_monad
 0.11% :     0.000022s :      1: auto_monad_reorder
 1.67% :     0.000345s :      1: bootstrap
 0.23% :     0.000046s :      1: cconv
 0.06% :     0.000012s :      1: convert_after_rewriter
 0.15% :     0.000031s :      1: cse_after_recomputation
 0.05% :     0.000011s :      1: environ_conv
 0.24% :     0.000050s :      1: event_method
 0.04% :     0.000008s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000011s :      1: graph_reusing
40.69% :     0.008397s :      1: jit_opt_a
 0.99% :     0.000204s :      1: jit_opt_after_cconv
 0.36% :     0.000073s :      1: jit_opt_b
 3.33% :     0.000687s :      1: loop_unroll
 5.10% :     0.001053s :      1: mutable_eliminate
 5.15% :     0.001062s :     26: opt.transform.jit_opt_a
 0.30% :     0.000063s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000033s :      4: opt.transform.jit_opt_b
 0.11% :     0.000022s :      1: opt.transform.loop_unroll_optimizer
 0.16% :     0.000033s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000040s :      4: opt.transform.symbol_engine_opt
 3.40% :     0.000702s :      1: opt_after_jit_grad
 0.04% :     0.000009s :      1: order_py_execute_after_rewriter
 0.03% :     0.000006s :      1: partial_unused_args_eliminate
 0.03% :     0.000007s :      1: pre_auto_parallel
 0.70% :     0.000144s :      1: py_interpret_to_execute
 0.11% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000021s :      1: remove_dup_value
 2.60% :     0.000536s :      1: renormalize.infer
 2.05% :     0.000423s :      1: renormalize.specialize
 0.07% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000048s :      1: rewriter_after_opt_a
 0.61% :     0.000125s :      1: rewriter_before_opt_a
 0.46% :     0.000095s :      1: symbol_engine_optimizer
29.35% :     0.006056s :      1: type_inference


TotalTime = 0.0199136, [33]
[bootstrap]: 0.0003155
[type_inference]: 0.0046794
[event_method]: 1.692e-05
[auto_monad]: 6.101e-05
[graph_reusing]: 6.34001e-06
[pre_auto_parallel]: 2.66e-06
[py_interpret_to_execute]: 3.105e-05
[rewriter_before_opt_a]: 7.111e-05
[expand_dump_flag]: 3.29001e-06
[jit_opt_a]: 0.00552402, [2]
    [Cycle 1]: 0.00173515, [27]
        [switch_simplify]: 5.499e-05
        [loop_unroll]: 2.908e-05
        [a_1]: 0.00056479
        [with_stream_mark]: 2.054e-05
        [recompute_prepare]: 8.55001e-06
        [updatestate_depend_eliminate]: 3.90998e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 6.48003e-06
        [updatestate_useless_node_eliminater]: 4.96002e-06
        [accelerated_algorithm]: 5.24998e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 5.33002e-06
        [merge_forward]: 4.48001e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.456e-05
        [j_node_and_user_rematch]: 9.30001e-06
        [meta_fg_expand]: 2.29001e-06
        [replace_old_param]: 9.47001e-06
        [inline_without_move]: 5.33002e-06
        [renormalize]: 0.00073216
        [add_forward_monad_depend]: 9.97001e-06
        [auto_monad_grad]: 2.22001e-06
        [auto_monad_eliminator]: 1.473e-05
        [cse]: 2.838e-05
        [replace_applicator]: 1.205e-05
    [Cycle 2]: 0.00030957, [27]
        [switch_simplify]: 5.94e-06
        [loop_unroll]: 5.09998e-06
        [a_1]: 8.922e-05
        [with_stream_mark]: 9.41998e-06
        [recompute_prepare]: 4.72998e-06
        [updatestate_depend_eliminate]: 2.52001e-06
        [updatestate_assign_eliminate]: 1.99999e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 4.74e-06
        [updatestate_useless_node_eliminater]: 4.65999e-06
        [accelerated_algorithm]: 5.02e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.78e-06
        [cell_reuse_recompute_pass]: 1.62001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.219e-05
        [j_node_and_user_rematch]: 8.45001e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 7.55e-06
        [inline_without_move]: 4.81002e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 6.60017e-07
        [auto_monad_eliminator]: 5.04998e-06
        [cse]: 9.34e-06
        [replace_applicator]: 4.68001e-06
[py_interpret_to_execute_after_opt_a]: 1.024e-05
[rewriter_after_opt_a]: 3.361e-05
[convert_after_rewriter]: 6.28e-06
[order_py_execute_after_rewriter]: 4.66002e-06
[mutable_eliminate]: 0.00058058
[jit_opt_b]: 5.206e-05, [1]
    [Cycle 1]: 4.453e-05, [2]
        [frontend_op_eliminate]: 1.534e-05
        [inline_after_opt_a]: 1.816e-05
[cconv]: 2.323e-05
[loop_unroll]: 0.00041352
[jit_opt_after_cconv]: 0.00013856, [1]
    [Cycle 1]: 0.00013145, [11]
        [c_1]: 2.035e-05
        [parameter_eliminate]: 2.34999e-06
        [updatestate_depend_eliminate]: 5.66e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.825e-05
        [call_graph_tuple_transform]: 1.91e-05
        [tuple_list_get_item_eliminator]: 5.17999e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 5.40022e-07
        [switch_simplify]: 5.24998e-06
[remove_dup_value]: 1.513e-05
[partial_unused_args_eliminate]: 2.12001e-06
[environ_conv]: 5.05001e-06
[add_recomputation]: 4.605e-05
[cse_after_recomputation]: 1.985e-05, [1]
    [Cycle 1]: 1.463e-05, [1]
        [cse]: 8.94998e-06
[auto_monad_reorder]: 1.465e-05
[get_jit_bprop_graph]: 2.14999e-06
[rewriter_after_jit_bprop_graph]: 4.99998e-06
[opt_after_jit_grad]: 0.00044437
[symbol_engine_optimizer]: 6.982e-05, [1]
    [Cycle 1]: 6.366e-05, [6]
        [build]: 3.47997e-06
        [elim_shapecalc]: 7.50998e-06
        [elim_not_effective]: 1.201e-05
        [opt_reshape]: 5.39e-06
        [fold_const_symbol]: 8.35001e-06
        [renormalize]: 3.30008e-07
[validate]: 3.433e-05
[backend_pass]: 8.60018e-07
[task_emit]: 0.00705882
[execute]: 8.56002e-06
Sums
    bootstrap                            :     0.000315s :  1.99%
    type_inference                       :     0.004679s : 29.58%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000061s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.20%
    rewriter_before_opt_a                :     0.000071s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000654s :  4.13%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000732s :  4.63%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000038s :  0.24%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000034s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000581s :  3.67%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.11%
    cconv                                :     0.000023s :  0.15%
    loop_unroll                          :     0.000414s :  2.61%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000046s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000444s :  2.81%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007059s : 44.62%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000208    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.25% :     0.000005s :      3: substitution.graph_param_transform
83.87% :     0.000175s :      4: substitution.inline
 1.94% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.17% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.02% :     0.000004s :      2: substitution.replace_old_param
 6.19% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004612     2
81.64% :     0.003766s :      1: type_inference.infer
18.36% :     0.000847s :      1: type_inference.specialize
------[replace.]   0.000068     6
63.07% :     0.000043s :      4: replace.inline
36.93% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000184     6
93.55% :     0.000172s :      4: match.inline
 6.45% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.56% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000001s :     10: predicate.addn_zero_filter
 2.12% :     0.000002s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.41% :     0.000002s :     10: predicate.compare_switch_simplify
 1.30% :     0.000001s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.58% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.46% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.18% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.16% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000001s :      6: predicate.less_batch_normalization
 1.83% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 2.36% :     0.000003s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000001s :     10: predicate.same_eliminate
 0.73% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.51% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.85% :     0.000009s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.36% :     0.000002s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.06% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.32% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000612    12
49.33% :     0.000302s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.54% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.13% :     0.000270s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021605    76
 0.23% :     0.000049s :      1: add_recomputation
 0.30% :     0.000064s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.55% :     0.000335s :      1: bootstrap
 0.12% :     0.000026s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.58% :     0.005527s :      1: jit_opt_a
 0.65% :     0.000141s :      1: jit_opt_after_cconv
 0.25% :     0.000055s :      1: jit_opt_b
 1.95% :     0.000421s :      1: loop_unroll
 2.73% :     0.000590s :      1: mutable_eliminate
 4.00% :     0.000864s :     26: opt.transform.jit_opt_a
 0.22% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.09% :     0.000452s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.15% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 1.73% :     0.000374s :      1: renormalize.infer
 1.62% :     0.000350s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000037s :      1: rewriter_after_opt_a
 0.34% :     0.000074s :      1: rewriter_before_opt_a
 0.34% :     0.000072s :      1: symbol_engine_optimizer
32.76% :     0.007078s :      1: task_emit
21.73% :     0.004695s :      1: type_inference
 0.25% :     0.000053s :      1: validate


TotalTime = 0.0172632, [33]
[bootstrap]: 0.00025916
[type_inference]: 0.00397463
[event_method]: 1.636e-05
[auto_monad]: 6.156e-05
[graph_reusing]: 5.09e-06
[pre_auto_parallel]: 2.24001e-06
[py_interpret_to_execute]: 2.817e-05
[rewriter_before_opt_a]: 6.871e-05
[expand_dump_flag]: 3.40998e-06
[jit_opt_a]: 0.00505904, [2]
    [Cycle 1]: 0.00138356, [27]
        [switch_simplify]: 4.839e-05
        [loop_unroll]: 2.797e-05
        [a_1]: 0.0004903
        [with_stream_mark]: 1.544e-05
        [recompute_prepare]: 6.59001e-06
        [updatestate_depend_eliminate]: 3.28998e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 5.54e-06
        [updatestate_useless_node_eliminater]: 4.79998e-06
        [accelerated_algorithm]: 5.35001e-06
        [meta_shard_fg_expand]: 1.98002e-06
        [get_grad_eliminate_]: 4.86002e-06
        [merge_forward]: 3.21001e-06
        [cell_reuse_recompute_pass]: 1.07998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.343e-05
        [j_node_and_user_rematch]: 8.49002e-06
        [meta_fg_expand]: 2.05002e-06
        [replace_old_param]: 7.58001e-06
        [inline_without_move]: 4.82998e-06
        [renormalize]: 0.00051218
        [add_forward_monad_depend]: 6.07001e-06
        [auto_monad_grad]: 1.77999e-06
        [auto_monad_eliminator]: 1.214e-05
        [cse]: 2.129e-05
        [replace_applicator]: 1.062e-05
    [Cycle 2]: 0.00031959, [27]
        [switch_simplify]: 5.34003e-06
        [loop_unroll]: 4.79e-06
        [a_1]: 8.64e-05
        [with_stream_mark]: 8.02e-06
        [recompute_prepare]: 4.70999e-06
        [updatestate_depend_eliminate]: 2.39001e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.77e-06
        [updatestate_useless_node_eliminater]: 4.71997e-06
        [accelerated_algorithm]: 4.79e-06
        [meta_shard_fg_expand]: 1.39998e-06
        [get_grad_eliminate_]: 4.55001e-06
        [merge_forward]: 2.41e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.194e-05
        [j_node_and_user_rematch]: 7.4e-06
        [meta_fg_expand]: 1.39e-06
        [replace_old_param]: 6.93998e-06
        [inline_without_move]: 4.42e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.21002e-06
        [auto_monad_grad]: 7.49977e-07
        [auto_monad_eliminator]: 4.80001e-06
        [cse]: 1.081e-05
        [replace_applicator]: 5.63002e-06
[py_interpret_to_execute_after_opt_a]: 9.78002e-06
[rewriter_after_opt_a]: 3.061e-05
[convert_after_rewriter]: 6.66e-06
[order_py_execute_after_rewriter]: 5.69999e-06
[mutable_eliminate]: 0.0005175
[jit_opt_b]: 4.561e-05, [1]
    [Cycle 1]: 3.936e-05, [2]
        [frontend_op_eliminate]: 1.461e-05
        [inline_after_opt_a]: 1.366e-05
[cconv]: 2.372e-05
[loop_unroll]: 0.00041303
[jit_opt_after_cconv]: 0.00013141, [1]
    [Cycle 1]: 0.00012531, [11]
        [c_1]: 1.901e-05
        [parameter_eliminate]: 2.49001e-06
        [updatestate_depend_eliminate]: 5.03002e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [cse]: 1.644e-05
        [call_graph_tuple_transform]: 1.719e-05
        [tuple_list_get_item_eliminator]: 5.19998e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 5.10016e-07
        [switch_simplify]: 4.92999e-06
[remove_dup_value]: 1.172e-05
[partial_unused_args_eliminate]: 2.27001e-06
[environ_conv]: 4.95001e-06
[add_recomputation]: 4.636e-05
[cse_after_recomputation]: 1.929e-05, [1]
    [Cycle 1]: 1.397e-05, [1]
        [cse]: 8.38001e-06
[auto_monad_reorder]: 1.135e-05
[get_jit_bprop_graph]: 1.54e-06
[rewriter_after_jit_bprop_graph]: 4.63001e-06
[opt_after_jit_grad]: 0.00044654
[symbol_engine_optimizer]: 6.884e-05, [1]
    [Cycle 1]: 6.256e-05, [6]
        [build]: 3.12002e-06
        [elim_shapecalc]: 7.08998e-06
        [elim_not_effective]: 1.203e-05
        [opt_reshape]: 5.40001e-06
        [fold_const_symbol]: 7.95e-06
        [renormalize]: 4.09986e-07
[validate]: 3.134e-05
[backend_pass]: 7.59988e-07
[task_emit]: 0.00577857
[execute]: 6.53e-06
Sums
    bootstrap                            :     0.000259s :  1.94%
    type_inference                       :     0.003975s : 29.79%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000062s :  0.46%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.21%
    rewriter_before_opt_a                :     0.000069s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000054s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000577s :  4.32%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000512s :  3.84%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000032s :  0.24%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000031s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000517s :  3.88%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000413s :  3.10%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000046s :  0.35%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000011s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000447s :  3.35%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005779s : 43.31%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000180    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.74% :     0.000005s :      3: substitution.graph_param_transform
83.22% :     0.000150s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.41% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.76% :     0.000003s :      2: substitution.replace_old_param
 6.55% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003924     2
82.47% :     0.003236s :      1: type_inference.infer
17.53% :     0.000688s :      1: type_inference.specialize
------[replace.]   0.000058     6
67.15% :     0.000039s :      4: replace.inline
32.85% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000158     6
93.29% :     0.000147s :      4: match.inline
 6.71% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.72% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.23% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000001s :     10: predicate.addn_zero_filter
 2.04% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000001s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.65% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.29% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.71% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.34% :     0.000003s :     15: predicate.partial_eliminate
 1.52% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.38% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.91% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.53% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.64% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.63% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.66% :     0.000008s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.37% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000475    12
47.88% :     0.000227s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.72% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.39% :     0.000216s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018645    76
 0.26% :     0.000049s :      1: add_recomputation
 0.35% :     0.000065s :      1: auto_monad
 0.07% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.45% :     0.000270s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
27.15% :     0.005062s :      1: jit_opt_a
 0.72% :     0.000134s :      1: jit_opt_after_cconv
 0.26% :     0.000048s :      1: jit_opt_b
 2.26% :     0.000421s :      1: loop_unroll
 2.82% :     0.000526s :      1: mutable_eliminate
 4.12% :     0.000769s :     26: opt.transform.jit_opt_a
 0.23% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000022s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.44% :     0.000454s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000031s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000014s :      1: remove_dup_value
 1.49% :     0.000279s :      1: renormalize.infer
 1.22% :     0.000227s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000033s :      1: rewriter_after_opt_a
 0.38% :     0.000071s :      1: rewriter_before_opt_a
 0.38% :     0.000071s :      1: symbol_engine_optimizer
31.05% :     0.005789s :      1: task_emit
21.37% :     0.003984s :      1: type_inference
 0.26% :     0.000049s :      1: validate


TotalTime = 0.0277232, [33]
[bootstrap]: 0.00018786
[type_inference]: 0.00931062
[event_method]: 3.784e-05
[auto_monad]: 8.82e-05
[graph_reusing]: 6.54001e-06
[pre_auto_parallel]: 2.11e-06
[py_interpret_to_execute]: 2.273e-05
[rewriter_before_opt_a]: 8.732e-05
[expand_dump_flag]: 2.91999e-06
[jit_opt_a]: 0.00999345, [3]
    [Cycle 1]: 0.00492971, [27]
        [switch_simplify]: 5.757e-05
        [loop_unroll]: 3.883e-05
        [a_1]: 0.00079413
        [with_stream_mark]: 2.289e-05
        [recompute_prepare]: 1.743e-05
        [updatestate_depend_eliminate]: 6.77002e-06
        [updatestate_assign_eliminate]: 6.28e-06
        [updatestate_loads_eliminate]: 5.94999e-06
        [parameter_eliminate]: 2.89999e-06
        [specialize_transform]: 1.269e-05
        [updatestate_useless_node_eliminater]: 1.136e-05
        [accelerated_algorithm]: 1.148e-05
        [meta_shard_fg_expand]: 2.89001e-06
        [get_grad_eliminate_]: 1.158e-05
        [merge_forward]: 7.83999e-06
        [cell_reuse_recompute_pass]: 9.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.514e-05
        [j_node_and_user_rematch]: 2.232e-05
        [meta_fg_expand]: 0.00129254
        [replace_old_param]: 5.162e-05
        [inline_without_move]: 5.024e-05
        [renormalize]: 0.00198953
        [add_forward_monad_depend]: 9.84999e-06
        [auto_monad_grad]: 5.46e-06
        [auto_monad_eliminator]: 4.487e-05
        [cse]: 0.00016922
        [replace_applicator]: 6.33e-05
    [Cycle 2]: 0.00174763, [27]
        [switch_simplify]: 3.449e-05
        [loop_unroll]: 3.367e-05
        [a_1]: 0.0009677
        [with_stream_mark]: 1.079e-05
        [recompute_prepare]: 7.96001e-06
        [updatestate_depend_eliminate]: 3.46999e-06
        [updatestate_assign_eliminate]: 2.89999e-06
        [updatestate_loads_eliminate]: 2.41998e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 6.56e-06
        [updatestate_useless_node_eliminater]: 5.75001e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 6.14001e-06
        [merge_forward]: 2.90002e-06
        [cell_reuse_recompute_pass]: 8.30012e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.186e-05
        [j_node_and_user_rematch]: 9.57999e-06
        [meta_fg_expand]: 2.874e-05
        [replace_old_param]: 9.12999e-06
        [inline_without_move]: 6.38003e-06
        [renormalize]: 0.00040145
        [add_forward_monad_depend]: 3.95e-06
        [auto_monad_grad]: 1.20999e-06
        [auto_monad_eliminator]: 1.064e-05
        [cse]: 1.818e-05
        [replace_applicator]: 1.139e-05
    [Cycle 3]: 0.00034951, [27]
        [switch_simplify]: 6.31e-06
        [loop_unroll]: 5.81e-06
        [a_1]: 0.00011464
        [with_stream_mark]: 8.50999e-06
        [recompute_prepare]: 5.84e-06
        [updatestate_depend_eliminate]: 3.21001e-06
        [updatestate_assign_eliminate]: 2.65002e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 8.80013e-07
        [specialize_transform]: 5.86e-06
        [updatestate_useless_node_eliminater]: 5.44e-06
        [accelerated_algorithm]: 5.39998e-06
        [meta_shard_fg_expand]: 1.33002e-06
        [get_grad_eliminate_]: 5.69e-06
        [merge_forward]: 2.72001e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.372e-05
        [j_node_and_user_rematch]: 9.27001e-06
        [meta_fg_expand]: 1.82001e-06
        [replace_old_param]: 7.8e-06
        [inline_without_move]: 5.62999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.08001e-06
        [auto_monad_grad]: 7.10017e-07
        [auto_monad_eliminator]: 6.98e-06
        [cse]: 1.308e-05
        [replace_applicator]: 5.70001e-06
[py_interpret_to_execute_after_opt_a]: 8.55001e-06
[rewriter_after_opt_a]: 3.162e-05
[convert_after_rewriter]: 6.79999e-06
[order_py_execute_after_rewriter]: 5.61e-06
[mutable_eliminate]: 0.00045134
[jit_opt_b]: 5.146e-05, [1]
    [Cycle 1]: 4.523e-05, [2]
        [frontend_op_eliminate]: 1.804e-05
        [inline_after_opt_a]: 1.678e-05
[cconv]: 2.014e-05
[loop_unroll]: 0.00039777
[jit_opt_after_cconv]: 0.00014458, [1]
    [Cycle 1]: 0.0001386, [11]
        [c_1]: 2.414e-05
        [parameter_eliminate]: 2.29001e-06
        [updatestate_depend_eliminate]: 5.44998e-06
        [updatestate_assign_eliminate]: 2.56998e-06
        [updatestate_loads_eliminate]: 2.43998e-06
        [cse]: 2.079e-05
        [call_graph_tuple_transform]: 1.978e-05
        [tuple_list_get_item_eliminator]: 5.87001e-06
        [none_parameter_eliminate]: 1.19998e-06
        [renormalize]: 4.7998e-07
        [switch_simplify]: 6.34001e-06
[remove_dup_value]: 1.406e-05
[partial_unused_args_eliminate]: 1.72001e-06
[environ_conv]: 5.18002e-06
[add_recomputation]: 4.41e-05
[cse_after_recomputation]: 2.481e-05, [1]
    [Cycle 1]: 1.948e-05, [1]
        [cse]: 1.39e-05
[auto_monad_reorder]: 1.69e-05
[get_jit_bprop_graph]: 1.32e-06
[rewriter_after_jit_bprop_graph]: 4.58999e-06
[opt_after_jit_grad]: 0.00045217
[symbol_engine_optimizer]: 7.676e-05, [1]
    [Cycle 1]: 7.105e-05, [6]
        [build]: 3.38999e-06
        [elim_shapecalc]: 8.73001e-06
        [elim_not_effective]: 1.422e-05
        [opt_reshape]: 6.81999e-06
        [fold_const_symbol]: 1.007e-05
        [renormalize]: 5.10016e-07
[validate]: 3.177e-05
[backend_pass]: 7.80012e-07
[task_emit]: 0.00600708
[execute]: 6.26998e-06
Sums
    bootstrap                            :     0.000188s :  0.78%
    type_inference                       :     0.009311s : 38.76%
    event_method                         :     0.000038s :  0.16%
    auto_monad                           :     0.000088s :  0.37%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000023s :  0.09%
    rewriter_before_opt_a                :     0.000087s :  0.36%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000098s :  0.41%
    jit_opt_a.loop_unroll                :     0.000078s :  0.33%
    jit_opt_a.a_1                        :     0.001876s :  7.81%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000031s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000013s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000025s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000023s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.10%
    jit_opt_a.merge_forward              :     0.000013s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000051s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000041s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001323s :  5.51%
    jit_opt_a.replace_old_param          :     0.000069s :  0.29%
    jit_opt_a.inline_without_move        :     0.000062s :  0.26%
    jit_opt_a.renormalize                :     0.002391s :  9.95%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000062s :  0.26%
    jit_opt_a.cse                        :     0.000200s :  0.83%
    jit_opt_a.replace_applicator         :     0.000080s :  0.33%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.04%
    rewriter_after_opt_a                 :     0.000032s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000451s :  1.88%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.07%
    cconv                                :     0.000020s :  0.08%
    loop_unroll                          :     0.000398s :  1.66%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000044s :  0.18%
    cse_after_recomputation.cse          :     0.000014s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000452s :  1.88%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006007s : 25.00%
    execute                              :     0.000006s :  0.03%


Time group info:
------[substitution.]   0.000416   110
 0.50% :     0.000002s :      3: substitution.elim_not_effective
 0.37% :     0.000002s :      3: substitution.fold_const_symbol
 1.32% :     0.000005s :      4: substitution.graph_param_transform
62.19% :     0.000259s :     12: substitution.inline
 3.58% :     0.000015s :      2: substitution.inline_without_move
 2.00% :     0.000008s :     14: substitution.j_node_and_user_rematch
 2.30% :     0.000010s :      7: substitution.minmaximum_grad
 1.40% :     0.000006s :      5: substitution.partial_eliminate
 2.48% :     0.000010s :     14: substitution.remove_not_recompute_node
 5.77% :     0.000024s :     10: substitution.replace_applicator
 1.54% :     0.000006s :      7: substitution.replace_old_param
 0.58% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.43% :     0.000018s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.99% :     0.000012s :      7: substitution.tuple_list_get_item_depend_reorder
 8.54% :     0.000036s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009238     2
90.05% :     0.008319s :      1: type_inference.infer
 9.95% :     0.000919s :      1: type_inference.specialize
------[replace.]   0.000132    19
63.59% :     0.000084s :     12: replace.inline
36.41% :     0.000048s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000270    19
93.18% :     0.000252s :     12: match.inline
 6.82% :     0.000018s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000353  2568
 1.67% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.45% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.45% :     0.000005s :     43: predicate.addn_check_dump
 1.70% :     0.000006s :     43: predicate.addn_zero_filter
 2.08% :     0.000007s :     43: predicate.arithmetic_simplify
 1.62% :     0.000006s :     43: predicate.cast_eliminate
 0.20% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.46% :     0.000005s :     43: predicate.compare_switch_simplify
 1.71% :     0.000006s :     43: predicate.depend_value_elim
 1.45% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.49% :     0.000005s :     43: predicate.dict_get_item_eliminator
 1.46% :     0.000005s :     43: predicate.dict_set_item_eliminator
 0.34% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      4: predicate.elim_not_effective
 0.21% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.46% :     0.000005s :     43: predicate.environ_add_const_eliminate
 1.49% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.44% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.54% :     0.000005s :     43: predicate.environ_get_eliminate
 1.46% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.84% :     0.000003s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.28% :     0.000015s :     70: predicate.inline
 1.87% :     0.000007s :     40: predicate.inline_without_move
 0.35% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.84% :     0.000003s :     19: predicate.less_batch_normalization
 1.90% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.95% :     0.000007s :     54: predicate.load_eliminater
 0.52% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.47% :     0.000012s :     85: predicate.loop_unroll_before_grad
 1.73% :     0.000006s :     47: predicate.make_slice_get_slice_eliminator
 1.51% :     0.000005s :     43: predicate.merge_addn
 1.47% :     0.000005s :     43: predicate.minmaximum_grad
 0.50% :     0.000002s :      4: predicate.mutable_eliminate
 0.22% :     0.000001s :      4: predicate.opt_reshape
 2.40% :     0.000008s :     54: predicate.partial_eliminate
 1.50% :     0.000005s :     43: predicate.print_const_string_wrapper
 1.82% :     0.000006s :     43: predicate.reduce_eliminate
 1.86% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.50% :     0.000009s :     96: predicate.replace_applicator
 0.91% :     0.000003s :     40: predicate.replace_old_param
 0.13% :     0.000000s :      4: predicate.reset_defer_inline
 1.67% :     0.000006s :     43: predicate.reshape_eliminate
 1.55% :     0.000005s :     43: predicate.row_tensor_add_zeros_like
 0.31% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.57% :     0.000006s :     43: predicate.same_eliminate
 0.47% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.37% :     0.000001s :      8: predicate.special_op_eliminate
 0.90% :     0.000003s :     19: predicate.specialize_transform
 1.69% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000005s :     43: predicate.stack_unstack_eliminate
 0.16% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.68% :     0.000009s :     62: predicate.switch_defer_inline
 2.56% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.38% :     0.000023s :    151: predicate.switch_simplify
 1.56% :     0.000006s :     43: predicate.tile_eliminate
 1.57% :     0.000006s :     43: predicate.transpose_eliminate
 1.79% :     0.000006s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.71% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.14% :     0.000011s :     58: predicate.tuple_list_get_item_eliminator
 1.79% :     0.000006s :     43: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000006s :     50: predicate.tuple_to_list_eliminator_
 1.93% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000010s :     73: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000006s :     43: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.24% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001232    29
61.21% :     0.000754s :     13: func_graph_cloner_run.FuncGraphClonerGraph
38.79% :     0.000478s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.032674    91
 0.14% :     0.000047s :      1: add_recomputation
 0.28% :     0.000092s :      1: auto_monad
 0.06% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.61% :     0.000199s :      1: bootstrap
 0.07% :     0.000023s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000027s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.13% :     0.000042s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
30.59% :     0.009996s :      1: jit_opt_a
 0.45% :     0.000147s :      1: jit_opt_after_cconv
 0.17% :     0.000054s :      1: jit_opt_b
 1.24% :     0.000405s :      1: loop_unroll
 1.41% :     0.000459s :      1: mutable_eliminate
 7.47% :     0.002441s :     39: opt.transform.jit_opt_a
 0.16% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000029s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.41% :     0.000460s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.08% :     0.000025s :      1: py_interpret_to_execute
 0.03% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000016s :      1: remove_dup_value
 3.78% :     0.001235s :      2: renormalize.infer
 3.50% :     0.001143s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000034s :      1: rewriter_after_opt_a
 0.27% :     0.000090s :      1: rewriter_before_opt_a
 0.24% :     0.000079s :      1: symbol_engine_optimizer
18.41% :     0.006017s :      1: task_emit
28.53% :     0.009322s :      1: type_inference
 0.15% :     0.000050s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.470.404 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dbf3940> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.470.935 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0223496, [30]
[bootstrap]: 0.00038073
[type_inference]: 0.00503658
[event_method]: 1.88e-05
[auto_monad]: 6.985e-05
[graph_reusing]: 6.36998e-06
[pre_auto_parallel]: 3.33e-06
[py_interpret_to_execute]: 3.71e-05
[rewriter_before_opt_a]: 8.159e-05
[expand_dump_flag]: 3.16999e-06
[jit_opt_a]: 0.0145163, [2]
    [Cycle 1]: 0.00168198, [27]
        [switch_simplify]: 5.725e-05
        [loop_unroll]: 3.028e-05
        [a_1]: 0.00053263
        [with_stream_mark]: 1.697e-05
        [recompute_prepare]: 7.60998e-06
        [updatestate_depend_eliminate]: 2.99001e-06
        [updatestate_assign_eliminate]: 2.40002e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [parameter_eliminate]: 1.61998e-06
        [specialize_transform]: 5.84999e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 6.00002e-06
        [meta_shard_fg_expand]: 1.69998e-06
        [get_grad_eliminate_]: 5.19e-06
        [merge_forward]: 2.68e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.211e-05
        [j_node_and_user_rematch]: 9.20999e-06
        [meta_fg_expand]: 1.89999e-06
        [replace_old_param]: 8.22998e-06
        [inline_without_move]: 5.35001e-06
        [renormalize]: 0.00071992
        [add_forward_monad_depend]: 7.18998e-06
        [auto_monad_grad]: 2.22999e-06
        [auto_monad_eliminator]: 1.437e-05
        [cse]: 2.529e-05
        [replace_applicator]: 1.417e-05
    [Cycle 2]: 0.00037315, [27]
        [switch_simplify]: 5.97999e-06
        [loop_unroll]: 5.51e-06
        [a_1]: 0.00011402
        [with_stream_mark]: 1.4e-05
        [recompute_prepare]: 5.89e-06
        [updatestate_depend_eliminate]: 3.48e-06
        [updatestate_assign_eliminate]: 2.69001e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.88997e-06
        [specialize_transform]: 5.17999e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 5.59e-06
        [meta_shard_fg_expand]: 2.00002e-06
        [get_grad_eliminate_]: 4.32e-06
        [merge_forward]: 3.28e-06
        [cell_reuse_recompute_pass]: 2.66999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.42e-05
        [j_node_and_user_rematch]: 8.48999e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 7.82e-06
        [inline_without_move]: 4.53999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.52001e-06
        [auto_monad_grad]: 9.60019e-07
        [auto_monad_eliminator]: 5.62999e-06
        [cse]: 1.362e-05
        [replace_applicator]: 5.79e-06
[py_interpret_to_execute_after_opt_a]: 1.195e-05
[rewriter_after_opt_a]: 3.337e-05
[convert_after_rewriter]: 6.36e-06
[order_py_execute_after_rewriter]: 4.67998e-06
[mutable_eliminate]: 0.0006434
[jit_opt_b]: 5.017e-05, [1]
    [Cycle 1]: 4.339e-05, [2]
        [frontend_op_eliminate]: 1.71e-05
        [inline_after_opt_a]: 1.491e-05
[cconv]: 2.2e-05
[loop_unroll]: 0.00044446
[jit_opt_after_cconv]: 0.00014599, [1]
    [Cycle 1]: 0.00013953, [11]
        [c_1]: 2.12e-05
        [parameter_eliminate]: 2.69999e-06
        [updatestate_depend_eliminate]: 5.84999e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [cse]: 2.156e-05
        [call_graph_tuple_transform]: 1.968e-05
        [tuple_list_get_item_eliminator]: 5.15001e-06
        [none_parameter_eliminate]: 1.15999e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 5.57001e-06
[remove_dup_value]: 9.47999e-06
[partial_unused_args_eliminate]: 1.93002e-06
[environ_conv]: 4.13999e-06
[add_recomputation]: 3.752e-05
[cse_after_recomputation]: 2.056e-05, [1]
    [Cycle 1]: 1.498e-05, [1]
        [cse]: 8.99998e-06
[auto_monad_reorder]: 1.167e-05
[get_jit_bprop_graph]: 2.26998e-06
[rewriter_after_jit_bprop_graph]: 5.45001e-06
[opt_after_jit_grad]: 0.00045283
[symbol_engine_optimizer]: 7.314e-05, [1]
    [Cycle 1]: 6.671e-05, [6]
        [build]: 3.81001e-06
        [elim_shapecalc]: 7.46001e-06
        [elim_not_effective]: 1.246e-05
        [opt_reshape]: 5.64e-06
        [fold_const_symbol]: 8.40001e-06
        [renormalize]: 5.19998e-07
[validate]: 3.221e-05
Sums
    bootstrap                            :     0.000381s :  4.10%
    type_inference                       :     0.005037s : 54.29%
    event_method                         :     0.000019s :  0.20%
    auto_monad                           :     0.000070s :  0.75%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.04%
    py_interpret_to_execute              :     0.000037s :  0.40%
    rewriter_before_opt_a                :     0.000082s :  0.88%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000063s :  0.68%
    jit_opt_a.loop_unroll                :     0.000036s :  0.39%
    jit_opt_a.a_1                        :     0.000647s :  6.97%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.17%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000720s :  7.76%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.22%
    jit_opt_a.cse                        :     0.000039s :  0.42%
    jit_opt_a.replace_applicator         :     0.000020s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.13%
    rewriter_after_opt_a                 :     0.000033s :  0.36%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000643s :  6.93%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000022s :  0.24%
    loop_unroll                          :     0.000444s :  4.79%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000009s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000038s :  0.40%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000012s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000453s :  4.88%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000032s :  0.35%


Time group info:
------[substitution.]   0.000204    23
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.67% :     0.000005s :      3: substitution.graph_param_transform
84.18% :     0.000171s :      4: substitution.inline
 1.88% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.21% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.95% :     0.000004s :      2: substitution.replace_old_param
 5.80% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004963     2
82.86% :     0.004113s :      1: type_inference.infer
17.14% :     0.000850s :      1: type_inference.specialize
------[replace.]   0.000068     6
64.02% :     0.000043s :      4: replace.inline
35.98% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000180     6
94.07% :     0.000169s :      4: match.inline
 5.93% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   691
 1.19% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.01% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.11% :     0.000002s :     10: predicate.addn_zero_filter
 1.76% :     0.000002s :     10: predicate.arithmetic_simplify
 1.03% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.00% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.67% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.98% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.03% :     0.000001s :     10: predicate.environ_get_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.79% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 4.44% :     0.000006s :     22: predicate.inline
 0.74% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.30% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.46% :     0.000002s :     15: predicate.load_eliminater
 1.24% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.52% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.55% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.91% :     0.000001s :     10: predicate.merge_addn
 0.89% :     0.000001s :     10: predicate.minmaximum_grad
 1.31% :     0.000002s :      3: predicate.mutable_eliminate
 0.38% :     0.000001s :      3: predicate.opt_reshape
 2.04% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.30% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.43% :     0.000001s :      6: predicate.replace_old_param
 0.19% :     0.000000s :      3: predicate.reset_defer_inline
 1.08% :     0.000001s :     10: predicate.reshape_eliminate
16.36% :     0.000023s :     10: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 0.86% :     0.000001s :      6: predicate.specialize_transform
 1.19% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.08% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.13% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.43% :     0.000009s :     53: predicate.switch_simplify
 1.02% :     0.000001s :     10: predicate.tile_eliminate
 0.93% :     0.000001s :     10: predicate.transpose_eliminate
 1.20% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.97% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.26% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.32% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.24% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.37% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.42% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.35% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000625    12
50.76% :     0.000317s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.48% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.75% :     0.000273s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023989    72
 0.17% :     0.000040s :      1: add_recomputation
 0.31% :     0.000075s :      1: auto_monad
 0.06% :     0.000015s :      1: auto_monad_reorder
 1.68% :     0.000402s :      1: bootstrap
 0.10% :     0.000025s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
60.53% :     0.014520s :      1: jit_opt_a
 0.62% :     0.000149s :      1: jit_opt_after_cconv
 0.22% :     0.000053s :      1: jit_opt_b
 1.88% :     0.000452s :      1: loop_unroll
 2.72% :     0.000652s :      1: mutable_eliminate
 3.60% :     0.000863s :     26: opt.transform.jit_opt_a
 0.20% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.92% :     0.000461s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000040s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000012s :      1: remove_dup_value
 1.50% :     0.000359s :      1: renormalize.infer
 1.47% :     0.000352s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000036s :      1: rewriter_after_opt_a
 0.36% :     0.000085s :      1: rewriter_before_opt_a
 0.32% :     0.000076s :      1: symbol_engine_optimizer
21.07% :     0.005054s :      1: type_inference


TotalTime = 0.0171692, [33]
[bootstrap]: 0.00026605
[type_inference]: 0.00400746
[event_method]: 1.645e-05
[auto_monad]: 5.647e-05
[graph_reusing]: 5.15001e-06
[pre_auto_parallel]: 2.11e-06
[py_interpret_to_execute]: 4.709e-05
[rewriter_before_opt_a]: 6.791e-05
[expand_dump_flag]: 3.00998e-06
[jit_opt_a]: 0.00500373, [2]
    [Cycle 1]: 0.00130755, [27]
        [switch_simplify]: 4.851e-05
        [loop_unroll]: 2.858e-05
        [a_1]: 0.00047634
        [with_stream_mark]: 1.583e-05
        [recompute_prepare]: 6.39001e-06
        [updatestate_depend_eliminate]: 3.20998e-06
        [updatestate_assign_eliminate]: 2.71e-06
        [updatestate_loads_eliminate]: 2.31998e-06
        [parameter_eliminate]: 1.60999e-06
        [specialize_transform]: 5.61003e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 1.87001e-06
        [get_grad_eliminate_]: 4.74e-06
        [merge_forward]: 3.21999e-06
        [cell_reuse_recompute_pass]: 1.07998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.258e-05
        [j_node_and_user_rematch]: 8.08001e-06
        [meta_fg_expand]: 2.29001e-06
        [replace_old_param]: 8.18999e-06
        [inline_without_move]: 4.94e-06
        [renormalize]: 0.00044951
        [add_forward_monad_depend]: 5.28002e-06
        [auto_monad_grad]: 1.86e-06
        [auto_monad_eliminator]: 1.177e-05
        [cse]: 2.189e-05
        [replace_applicator]: 1.043e-05
    [Cycle 2]: 0.00030138, [27]
        [switch_simplify]: 5.42999e-06
        [loop_unroll]: 4.94e-06
        [a_1]: 8.549e-05
        [with_stream_mark]: 8.28001e-06
        [recompute_prepare]: 4.90001e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 9.20001e-07
        [specialize_transform]: 4.77e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 4.33999e-06
        [meta_shard_fg_expand]: 1.06002e-06
        [get_grad_eliminate_]: 4.4e-06
        [merge_forward]: 2.26e-06
        [cell_reuse_recompute_pass]: 1.32999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.212e-05
        [j_node_and_user_rematch]: 7.45998e-06
        [meta_fg_expand]: 1.32999e-06
        [replace_old_param]: 6.92002e-06
        [inline_without_move]: 4.67e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.59998e-06
        [auto_monad_grad]: 7.10017e-07
        [auto_monad_eliminator]: 4.81002e-06
        [cse]: 9.79e-06
        [replace_applicator]: 4.94998e-06
[py_interpret_to_execute_after_opt_a]: 8.52e-06
[rewriter_after_opt_a]: 3.067e-05
[convert_after_rewriter]: 6.61e-06
[order_py_execute_after_rewriter]: 4.63999e-06
[mutable_eliminate]: 0.00050812
[jit_opt_b]: 4.498e-05, [1]
    [Cycle 1]: 3.882e-05, [2]
        [frontend_op_eliminate]: 1.412e-05
        [inline_after_opt_a]: 1.368e-05
[cconv]: 2.379e-05
[loop_unroll]: 0.00041488
[jit_opt_after_cconv]: 0.00013194, [1]
    [Cycle 1]: 0.00012592, [11]
        [c_1]: 1.908e-05
        [parameter_eliminate]: 2.50002e-06
        [updatestate_depend_eliminate]: 5.00999e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 1.98997e-06
        [cse]: 1.695e-05
        [call_graph_tuple_transform]: 1.693e-05
        [tuple_list_get_item_eliminator]: 5.25999e-06
        [none_parameter_eliminate]: 1.93002e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 4.97e-06
[remove_dup_value]: 1.178e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 4.87e-06
[add_recomputation]: 4.23e-05
[cse_after_recomputation]: 2.03e-05, [1]
    [Cycle 1]: 1.506e-05, [1]
        [cse]: 8.80001e-06
[auto_monad_reorder]: 1.364e-05
[get_jit_bprop_graph]: 1.40999e-06
[rewriter_after_jit_bprop_graph]: 5.51998e-06
[opt_after_jit_grad]: 0.00045018
[symbol_engine_optimizer]: 6.833e-05, [1]
    [Cycle 1]: 6.255e-05, [6]
        [build]: 3.24001e-06
        [elim_shapecalc]: 7.16999e-06
        [elim_not_effective]: 1.186e-05
        [opt_reshape]: 5.17e-06
        [fold_const_symbol]: 7.88001e-06
        [renormalize]: 7.30011e-07
[validate]: 3.05e-05
[backend_pass]: 6.89994e-07
[task_emit]: 0.00568011
[execute]: 6.64999e-06
Sums
    bootstrap                            :     0.000266s :  2.01%
    type_inference                       :     0.004007s : 30.33%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000056s :  0.43%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000047s :  0.36%
    rewriter_before_opt_a                :     0.000068s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000562s :  4.25%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000450s :  3.40%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000032s :  0.24%
    jit_opt_a.replace_applicator         :     0.000015s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000508s :  3.85%
    jit_opt_b.frontend_op_eliminate      :     0.000014s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000415s :  3.14%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000042s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000450s :  3.41%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000030s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005680s : 42.99%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000168    23
 1.07% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.91% :     0.000005s :      3: substitution.graph_param_transform
82.80% :     0.000139s :      4: substitution.inline
 1.58% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.33% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000003s :      2: substitution.replace_old_param
 6.51% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003955     2
82.01% :     0.003243s :      1: type_inference.infer
17.99% :     0.000711s :      1: type_inference.specialize
------[replace.]   0.000057     6
67.44% :     0.000039s :      4: replace.inline
32.56% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000147     6
93.30% :     0.000137s :      4: match.inline
 6.70% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.63% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.48% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 1.89% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.24% :     0.000001s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.52% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.41% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.33% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.46% :     0.000002s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.01% :     0.000002s :     15: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.05% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.94% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000002s :     15: predicate.partial_eliminate
 1.52% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.38% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.81% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000001s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.53% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.58% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.55% :     0.000008s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.24% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.83% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.17% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000461    12
48.03% :     0.000221s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.41% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.56% :     0.000214s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018474    76
 0.24% :     0.000045s :      1: add_recomputation
 0.32% :     0.000060s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.49% :     0.000276s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000036s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
27.10% :     0.005006s :      1: jit_opt_a
 0.73% :     0.000135s :      1: jit_opt_after_cconv
 0.26% :     0.000047s :      1: jit_opt_b
 2.29% :     0.000423s :      1: loop_unroll
 2.80% :     0.000517s :      1: mutable_eliminate
 4.08% :     0.000754s :     26: opt.transform.jit_opt_a
 0.23% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000022s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.48% :     0.000459s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000050s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 1.27% :     0.000235s :      1: renormalize.infer
 1.13% :     0.000208s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.38% :     0.000071s :      1: rewriter_before_opt_a
 0.38% :     0.000071s :      1: symbol_engine_optimizer
30.80% :     0.005689s :      1: task_emit
21.75% :     0.004018s :      1: type_inference
 0.26% :     0.000048s :      1: validate


TotalTime = 0.0197993, [33]
[bootstrap]: 0.00030863
[type_inference]: 0.00520716
[event_method]: 1.802e-05
[auto_monad]: 6.412e-05
[graph_reusing]: 6.39999e-06
[pre_auto_parallel]: 2.61999e-06
[py_interpret_to_execute]: 3.623e-05
[rewriter_before_opt_a]: 7.28e-05
[expand_dump_flag]: 3.28e-06
[jit_opt_a]: 0.00557159, [2]
    [Cycle 1]: 0.00167281, [27]
        [switch_simplify]: 5.5e-05
        [loop_unroll]: 2.894e-05
        [a_1]: 0.00055276
        [with_stream_mark]: 1.508e-05
        [recompute_prepare]: 7.2e-06
        [updatestate_depend_eliminate]: 2.89999e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 9.00007e-07
        [specialize_transform]: 6.29001e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 1.54998e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 3.08e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.183e-05
        [j_node_and_user_rematch]: 9.09e-06
        [meta_fg_expand]: 2.11998e-06
        [replace_old_param]: 8e-06
        [inline_without_move]: 5.27001e-06
        [renormalize]: 0.00072091
        [add_forward_monad_depend]: 5.88002e-06
        [auto_monad_grad]: 2.38002e-06
        [auto_monad_eliminator]: 1.388e-05
        [cse]: 1.992e-05
        [replace_applicator]: 1.2e-05
    [Cycle 2]: 0.00032135, [27]
        [switch_simplify]: 5.84e-06
        [loop_unroll]: 4.85001e-06
        [a_1]: 9.041e-05
        [with_stream_mark]: 9.75002e-06
        [recompute_prepare]: 5.09e-06
        [updatestate_depend_eliminate]: 2.54999e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 8.00006e-07
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 1.59e-06
        [get_grad_eliminate_]: 4.43001e-06
        [merge_forward]: 2.54999e-06
        [cell_reuse_recompute_pass]: 2.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.337e-05
        [j_node_and_user_rematch]: 8.07e-06
        [meta_fg_expand]: 1.42999e-06
        [replace_old_param]: 1.041e-05
        [inline_without_move]: 4.58001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 1.39998e-06
        [auto_monad_eliminator]: 4.72e-06
        [cse]: 1.036e-05
        [replace_applicator]: 4.72e-06
[py_interpret_to_execute_after_opt_a]: 1.284e-05
[rewriter_after_opt_a]: 3.355e-05
[convert_after_rewriter]: 6.29999e-06
[order_py_execute_after_rewriter]: 7.38e-06
[mutable_eliminate]: 0.000721
[jit_opt_b]: 5.028e-05, [1]
    [Cycle 1]: 4.354e-05, [2]
        [frontend_op_eliminate]: 1.723e-05
        [inline_after_opt_a]: 1.453e-05
[cconv]: 2.779e-05
[loop_unroll]: 0.00043801
[jit_opt_after_cconv]: 0.00015427, [1]
    [Cycle 1]: 0.00014722, [11]
        [c_1]: 2.23e-05
        [parameter_eliminate]: 3.73001e-06
        [updatestate_depend_eliminate]: 6.53003e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 2.47001e-06
        [cse]: 2.281e-05
        [call_graph_tuple_transform]: 2.048e-05
        [tuple_list_get_item_eliminator]: 5.29e-06
        [none_parameter_eliminate]: 1.69998e-06
        [renormalize]: 7.50006e-07
        [switch_simplify]: 5.82999e-06
[remove_dup_value]: 3.167e-05
[partial_unused_args_eliminate]: 2.54999e-06
[environ_conv]: 4.90999e-06
[add_recomputation]: 6.034e-05
[cse_after_recomputation]: 2.518e-05, [1]
    [Cycle 1]: 1.884e-05, [1]
        [cse]: 1.131e-05
[auto_monad_reorder]: 1.552e-05
[get_jit_bprop_graph]: 2.11e-06
[rewriter_after_jit_bprop_graph]: 7.89002e-06
[opt_after_jit_grad]: 0.00049504
[symbol_engine_optimizer]: 8.383e-05, [1]
    [Cycle 1]: 7.621e-05, [6]
        [build]: 4.4e-06
        [elim_shapecalc]: 8.94e-06
        [elim_not_effective]: 1.38e-05
        [opt_reshape]: 6.34001e-06
        [fold_const_symbol]: 1.096e-05
        [renormalize]: 5.89993e-07
[validate]: 4.292e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.0060583
[execute]: 7.78999e-06
Sums
    bootstrap                            :     0.000309s :  1.98%
    type_inference                       :     0.005207s : 33.39%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000064s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.23%
    rewriter_before_opt_a                :     0.000073s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000643s :  4.12%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000721s :  4.62%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000030s :  0.19%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000034s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000007s :  0.05%
    mutable_eliminate                    :     0.000721s :  4.62%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000028s :  0.18%
    loop_unroll                          :     0.000438s :  2.81%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000032s :  0.20%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000060s :  0.39%
    cse_after_recomputation.cse          :     0.000011s :  0.07%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000495s :  3.17%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.28%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006058s : 38.85%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000225    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000002s :      2: substitution.fold_const_symbol
 2.62% :     0.000006s :      3: substitution.graph_param_transform
85.58% :     0.000192s :      4: substitution.inline
 1.43% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.98% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.50% :     0.000003s :      2: substitution.replace_old_param
 5.29% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005136     2
81.23% :     0.004172s :      1: type_inference.infer
18.77% :     0.000964s :      1: type_inference.specialize
------[replace.]   0.000073     6
66.53% :     0.000049s :      4: replace.inline
33.47% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     6
94.65% :     0.000190s :      4: match.inline
 5.35% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.56% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.47% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 2.66% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.60% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.34% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.01% :     0.000006s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.55% :     0.000002s :      6: predicate.less_batch_normalization
 1.68% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.62% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.78% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.07% :     0.000002s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.73% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.89% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.25% :     0.000001s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.57% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.75% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.26% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.77% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000655    12
48.39% :     0.000317s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.10% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.51% :     0.000298s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021477    76
 0.30% :     0.000063s :      1: add_recomputation
 0.31% :     0.000067s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.52% :     0.000327s :      1: bootstrap
 0.14% :     0.000030s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000028s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.96% :     0.005575s :      1: jit_opt_a
 0.73% :     0.000157s :      1: jit_opt_after_cconv
 0.25% :     0.000053s :      1: jit_opt_b
 2.08% :     0.000447s :      1: loop_unroll
 3.39% :     0.000729s :      1: mutable_eliminate
 3.98% :     0.000855s :     26: opt.transform.jit_opt_a
 0.23% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.35% :     0.000504s :      1: opt_after_jit_grad
 0.04% :     0.000010s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000039s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.16% :     0.000035s :      1: remove_dup_value
 1.77% :     0.000379s :      1: renormalize.infer
 1.56% :     0.000334s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000037s :      1: rewriter_after_opt_a
 0.36% :     0.000076s :      1: rewriter_before_opt_a
 0.40% :     0.000086s :      1: symbol_engine_optimizer
28.27% :     0.006072s :      1: task_emit
24.32% :     0.005224s :      1: type_inference
 0.29% :     0.000063s :      1: validate


TotalTime = 0.0284998, [33]
[bootstrap]: 0.00018148
[type_inference]: 0.00915973
[event_method]: 3.783e-05
[auto_monad]: 8.055e-05
[graph_reusing]: 6.24001e-06
[pre_auto_parallel]: 2.34999e-06
[py_interpret_to_execute]: 2.116e-05
[rewriter_before_opt_a]: 8.472e-05
[expand_dump_flag]: 2.68e-06
[jit_opt_a]: 0.0103795, [3]
    [Cycle 1]: 0.00512051, [27]
        [switch_simplify]: 5.75e-05
        [loop_unroll]: 3.713e-05
        [a_1]: 0.00078141
        [with_stream_mark]: 2.539e-05
        [recompute_prepare]: 1.795e-05
        [updatestate_depend_eliminate]: 7.08e-06
        [updatestate_assign_eliminate]: 6.16e-06
        [updatestate_loads_eliminate]: 5.61e-06
        [parameter_eliminate]: 2.48e-06
        [specialize_transform]: 1.309e-05
        [updatestate_useless_node_eliminater]: 1.187e-05
        [accelerated_algorithm]: 1.181e-05
        [meta_shard_fg_expand]: 3.24001e-06
        [get_grad_eliminate_]: 1.191e-05
        [merge_forward]: 6.87002e-06
        [cell_reuse_recompute_pass]: 1.11997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.609e-05
        [j_node_and_user_rematch]: 2.193e-05
        [meta_fg_expand]: 0.00140878
        [replace_old_param]: 5.296e-05
        [inline_without_move]: 5.008e-05
        [renormalize]: 0.00209167
        [add_forward_monad_depend]: 9.72999e-06
        [auto_monad_grad]: 5.44e-06
        [auto_monad_eliminator]: 4.65e-05
        [cse]: 0.00013649
        [replace_applicator]: 6.507e-05
    [Cycle 2]: 0.00190431, [27]
        [switch_simplify]: 3.35e-05
        [loop_unroll]: 3.318e-05
        [a_1]: 0.0010273
        [with_stream_mark]: 1.444e-05
        [recompute_prepare]: 8.87e-06
        [updatestate_depend_eliminate]: 4.37e-06
        [updatestate_assign_eliminate]: 2.85998e-06
        [updatestate_loads_eliminate]: 2.67001e-06
        [parameter_eliminate]: 1.30001e-06
        [specialize_transform]: 6.76999e-06
        [updatestate_useless_node_eliminater]: 6.93e-06
        [accelerated_algorithm]: 6.17001e-06
        [meta_shard_fg_expand]: 1.92001e-06
        [get_grad_eliminate_]: 5.67001e-06
        [merge_forward]: 3.03e-06
        [cell_reuse_recompute_pass]: 6.89994e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.515e-05
        [j_node_and_user_rematch]: 1.052e-05
        [meta_fg_expand]: 3.528e-05
        [replace_old_param]: 1.025e-05
        [inline_without_move]: 6.13998e-06
        [renormalize]: 0.00046324
        [add_forward_monad_depend]: 4.4e-06
        [auto_monad_grad]: 1.10001e-06
        [auto_monad_eliminator]: 1.218e-05
        [cse]: 2.217e-05
        [replace_applicator]: 1.265e-05
    [Cycle 3]: 0.00035749, [27]
        [switch_simplify]: 6.44999e-06
        [loop_unroll]: 5.61e-06
        [a_1]: 0.00011423
        [with_stream_mark]: 8.59e-06
        [recompute_prepare]: 5.79e-06
        [updatestate_depend_eliminate]: 3.36001e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [parameter_eliminate]: 8.89995e-07
        [specialize_transform]: 6.01e-06
        [updatestate_useless_node_eliminater]: 5.50001e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 1.21997e-06
        [get_grad_eliminate_]: 6.28e-06
        [merge_forward]: 3.03e-06
        [cell_reuse_recompute_pass]: 1.59998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.546e-05
        [j_node_and_user_rematch]: 1.033e-05
        [meta_fg_expand]: 2.09999e-06
        [replace_old_param]: 8.35999e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.31998e-06
        [auto_monad_grad]: 7.10017e-07
        [auto_monad_eliminator]: 6.14001e-06
        [cse]: 1.367e-05
        [replace_applicator]: 5.67999e-06
[py_interpret_to_execute_after_opt_a]: 1.302e-05
[rewriter_after_opt_a]: 3.807e-05
[convert_after_rewriter]: 8.00999e-06
[order_py_execute_after_rewriter]: 5.27999e-06
[mutable_eliminate]: 0.00059766
[jit_opt_b]: 5.675e-05, [1]
    [Cycle 1]: 4.883e-05, [2]
        [frontend_op_eliminate]: 1.842e-05
        [inline_after_opt_a]: 1.799e-05
[cconv]: 2.237e-05
[loop_unroll]: 0.00042655
[jit_opt_after_cconv]: 0.00015703, [1]
    [Cycle 1]: 0.00015058, [11]
        [c_1]: 2.526e-05
        [parameter_eliminate]: 2.52001e-06
        [updatestate_depend_eliminate]: 5.62001e-06
        [updatestate_assign_eliminate]: 3.45998e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [cse]: 2.46e-05
        [call_graph_tuple_transform]: 2.156e-05
        [tuple_list_get_item_eliminator]: 6.84999e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 6.26e-06
[remove_dup_value]: 1.492e-05
[partial_unused_args_eliminate]: 2.84999e-06
[environ_conv]: 1.068e-05
[add_recomputation]: 5.088e-05
[cse_after_recomputation]: 2.527e-05, [1]
    [Cycle 1]: 1.972e-05, [1]
        [cse]: 1.401e-05
[auto_monad_reorder]: 1.74e-05
[get_jit_bprop_graph]: 2.05002e-06
[rewriter_after_jit_bprop_graph]: 5.13002e-06
[opt_after_jit_grad]: 0.0004636
[symbol_engine_optimizer]: 7.787e-05, [1]
    [Cycle 1]: 7.175e-05, [6]
        [build]: 3.75998e-06
        [elim_shapecalc]: 8.33001e-06
        [elim_not_effective]: 1.464e-05
        [opt_reshape]: 6.54001e-06
        [fold_const_symbol]: 1.024e-05
        [renormalize]: 4.89992e-07
[validate]: 3.323e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.006329
[execute]: 6.37001e-06
Sums
    bootstrap                            :     0.000181s :  0.73%
    type_inference                       :     0.009160s : 37.03%
    event_method                         :     0.000038s :  0.15%
    auto_monad                           :     0.000081s :  0.33%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000021s :  0.09%
    rewriter_before_opt_a                :     0.000085s :  0.34%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000097s :  0.39%
    jit_opt_a.loop_unroll                :     0.000076s :  0.31%
    jit_opt_a.a_1                        :     0.001923s :  7.77%
    jit_opt_a.with_stream_mark           :     0.000048s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000023s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.10%
    jit_opt_a.merge_forward              :     0.000013s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000057s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000043s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001446s :  5.85%
    jit_opt_a.replace_old_param          :     0.000072s :  0.29%
    jit_opt_a.inline_without_move        :     0.000062s :  0.25%
    jit_opt_a.renormalize                :     0.002555s : 10.33%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000065s :  0.26%
    jit_opt_a.cse                        :     0.000172s :  0.70%
    jit_opt_a.replace_applicator         :     0.000083s :  0.34%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.05%
    rewriter_after_opt_a                 :     0.000038s :  0.15%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000598s :  2.42%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.07%
    cconv                                :     0.000022s :  0.09%
    loop_unroll                          :     0.000427s :  1.72%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000011s :  0.04%
    add_recomputation                    :     0.000051s :  0.21%
    cse_after_recomputation.cse          :     0.000014s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000464s :  1.87%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006329s : 25.59%
    execute                              :     0.000006s :  0.03%


Time group info:
------[substitution.]   0.000446   110
 0.51% :     0.000002s :      3: substitution.elim_not_effective
 0.34% :     0.000002s :      3: substitution.fold_const_symbol
 1.20% :     0.000005s :      4: substitution.graph_param_transform
62.38% :     0.000278s :     12: substitution.inline
 3.62% :     0.000016s :      2: substitution.inline_without_move
 1.90% :     0.000008s :     14: substitution.j_node_and_user_rematch
 2.54% :     0.000011s :      7: substitution.minmaximum_grad
 1.37% :     0.000006s :      5: substitution.partial_eliminate
 2.57% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.63% :     0.000025s :     10: substitution.replace_applicator
 1.53% :     0.000007s :      7: substitution.replace_old_param
 0.62% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.37% :     0.000019s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.96% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 8.47% :     0.000038s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009099     2
89.77% :     0.008168s :      1: type_inference.infer
10.23% :     0.000931s :      1: type_inference.specialize
------[replace.]   0.000133    19
63.51% :     0.000084s :     12: replace.inline
36.49% :     0.000048s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000291    19
93.43% :     0.000272s :     12: match.inline
 6.57% :     0.000019s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000360  2568
 1.56% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.42% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.45% :     0.000005s :     43: predicate.addn_check_dump
 1.80% :     0.000006s :     43: predicate.addn_zero_filter
 2.00% :     0.000007s :     43: predicate.arithmetic_simplify
 1.62% :     0.000006s :     43: predicate.cast_eliminate
 0.16% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.42% :     0.000005s :     43: predicate.compare_switch_simplify
 1.66% :     0.000006s :     43: predicate.depend_value_elim
 1.51% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.49% :     0.000005s :     43: predicate.dict_get_item_eliminator
 1.46% :     0.000005s :     43: predicate.dict_set_item_eliminator
 0.28% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      4: predicate.elim_not_effective
 0.20% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.57% :     0.000006s :     43: predicate.environ_add_const_eliminate
 1.42% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.43% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.47% :     0.000005s :     43: predicate.environ_get_eliminate
 1.49% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.87% :     0.000003s :     19: predicate.get_grad_eliminate
 0.18% :     0.000001s :      4: predicate.graph_param_transform
 4.64% :     0.000017s :     70: predicate.inline
 1.79% :     0.000006s :     40: predicate.inline_without_move
 0.35% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.99% :     0.000004s :     19: predicate.less_batch_normalization
 1.85% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.98% :     0.000007s :     54: predicate.load_eliminater
 0.49% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.05% :     0.000011s :     85: predicate.loop_unroll_before_grad
 1.79% :     0.000006s :     47: predicate.make_slice_get_slice_eliminator
 1.48% :     0.000005s :     43: predicate.merge_addn
 1.53% :     0.000006s :     43: predicate.minmaximum_grad
 0.59% :     0.000002s :      4: predicate.mutable_eliminate
 0.19% :     0.000001s :      4: predicate.opt_reshape
 2.37% :     0.000009s :     54: predicate.partial_eliminate
 1.59% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.71% :     0.000006s :     43: predicate.reduce_eliminate
 1.82% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.49% :     0.000009s :     96: predicate.replace_applicator
 0.94% :     0.000003s :     40: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.70% :     0.000006s :     43: predicate.reshape_eliminate
 1.60% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.27% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.80% :     0.000006s :     43: predicate.same_eliminate
 0.46% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.36% :     0.000001s :      8: predicate.special_op_eliminate
 0.89% :     0.000003s :     19: predicate.specialize_transform
 1.59% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000005s :     43: predicate.stack_unstack_eliminate
 0.16% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.67% :     0.000010s :     62: predicate.switch_defer_inline
 2.53% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.25% :     0.000022s :    151: predicate.switch_simplify
 1.51% :     0.000005s :     43: predicate.tile_eliminate
 1.45% :     0.000005s :     43: predicate.transpose_eliminate
 1.83% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.69% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.04% :     0.000011s :     58: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000006s :     43: predicate.tuple_list_set_item_eliminator
 1.81% :     0.000006s :     50: predicate.tuple_to_list_eliminator_
 1.92% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000011s :     73: predicate.updatestate_useless_node_eliminater
 2.04% :     0.000007s :     43: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.22% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001245    29
60.86% :     0.000758s :     13: func_graph_cloner_run.FuncGraphClonerGraph
39.14% :     0.000487s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.033688    91
 0.16% :     0.000054s :      1: add_recomputation
 0.25% :     0.000085s :      1: auto_monad
 0.06% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.58% :     0.000194s :      1: bootstrap
 0.07% :     0.000025s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000028s :      1: cse_after_recomputation
 0.04% :     0.000013s :      1: environ_conv
 0.13% :     0.000042s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
30.82% :     0.010383s :      1: jit_opt_a
 0.48% :     0.000160s :      1: jit_opt_after_cconv
 0.18% :     0.000059s :      1: jit_opt_b
 1.29% :     0.000434s :      1: loop_unroll
 1.80% :     0.000606s :      1: mutable_eliminate
 7.42% :     0.002500s :     39: opt.transform.jit_opt_a
 0.17% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000030s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.40% :     0.000471s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000024s :      1: py_interpret_to_execute
 0.05% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 4.06% :     0.001368s :      2: renormalize.infer
 3.49% :     0.001174s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000041s :      1: rewriter_after_opt_a
 0.26% :     0.000087s :      1: rewriter_before_opt_a
 0.24% :     0.000080s :      1: symbol_engine_optimizer
18.81% :     0.006338s :      1: task_emit
27.23% :     0.009172s :      1: type_inference
 0.16% :     0.000053s :      1: validate


TotalTime = 0.0162265, [33]
[bootstrap]: 0.00028816
[type_inference]: 0.00424412
[event_method]: 1.68e-05
[auto_monad]: 5.912e-05
[graph_reusing]: 5.02999e-06
[pre_auto_parallel]: 2.03002e-06
[py_interpret_to_execute]: 4.587e-05
[rewriter_before_opt_a]: 6.706e-05
[expand_dump_flag]: 2.78998e-06
[jit_opt_a]: 0.00474664, [2]
    [Cycle 1]: 0.00130806, [27]
        [switch_simplify]: 4.182e-05
        [loop_unroll]: 2.82e-05
        [a_1]: 0.00045703
        [with_stream_mark]: 1.304e-05
        [recompute_prepare]: 6.65002e-06
        [updatestate_depend_eliminate]: 3.04001e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 2.03002e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 6.04001e-06
        [updatestate_useless_node_eliminater]: 4.70999e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 1.45001e-06
        [get_grad_eliminate_]: 4.87998e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 8.80013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.08e-05
        [j_node_and_user_rematch]: 8.10999e-06
        [meta_fg_expand]: 1.84998e-06
        [replace_old_param]: 7.48999e-06
        [inline_without_move]: 4.92e-06
        [renormalize]: 0.00049414
        [add_forward_monad_depend]: 4.89e-06
        [auto_monad_grad]: 1.30999e-06
        [auto_monad_eliminator]: 1.067e-05
        [cse]: 1.503e-05
        [replace_applicator]: 1.13e-05
    [Cycle 2]: 0.00030649, [27]
        [switch_simplify]: 5.37001e-06
        [loop_unroll]: 4.60999e-06
        [a_1]: 8.65e-05
        [with_stream_mark]: 7.84997e-06
        [recompute_prepare]: 4.82e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.75001e-06
        [parameter_eliminate]: 9.09989e-07
        [specialize_transform]: 4.68001e-06
        [updatestate_useless_node_eliminater]: 4.63999e-06
        [accelerated_algorithm]: 4.87e-06
        [meta_shard_fg_expand]: 1.05001e-06
        [get_grad_eliminate_]: 4.35e-06
        [merge_forward]: 2.19001e-06
        [cell_reuse_recompute_pass]: 2.04999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.162e-05
        [j_node_and_user_rematch]: 7.41999e-06
        [meta_fg_expand]: 1.39e-06
        [replace_old_param]: 6.54001e-06
        [inline_without_move]: 8.79e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 5.79981e-07
        [auto_monad_eliminator]: 4.80001e-06
        [cse]: 9.39998e-06
        [replace_applicator]: 4.87e-06
[py_interpret_to_execute_after_opt_a]: 7.61999e-06
[rewriter_after_opt_a]: 2.821e-05
[convert_after_rewriter]: 5.62001e-06
[order_py_execute_after_rewriter]: 4.22e-06
[mutable_eliminate]: 0.00048278
[jit_opt_b]: 4.572e-05, [1]
    [Cycle 1]: 3.948e-05, [2]
        [frontend_op_eliminate]: 1.468e-05
        [inline_after_opt_a]: 1.347e-05
[cconv]: 1.78e-05
[loop_unroll]: 0.00040596
[jit_opt_after_cconv]: 0.00012717, [1]
    [Cycle 1]: 0.00012095, [11]
        [c_1]: 1.965e-05
        [parameter_eliminate]: 2.37999e-06
        [updatestate_depend_eliminate]: 5.05999e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.485e-05
        [call_graph_tuple_transform]: 1.586e-05
        [tuple_list_get_item_eliminator]: 4.97e-06
        [none_parameter_eliminate]: 7.89994e-07
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.00001e-06
[remove_dup_value]: 8.65001e-06
[partial_unused_args_eliminate]: 1.77001e-06
[environ_conv]: 4.18001e-06
[add_recomputation]: 3.728e-05
[cse_after_recomputation]: 1.896e-05, [1]
    [Cycle 1]: 1.404e-05, [1]
        [cse]: 8.19998e-06
[auto_monad_reorder]: 9.77999e-06
[get_jit_bprop_graph]: 1.37e-06
[rewriter_after_jit_bprop_graph]: 4.58999e-06
[opt_after_jit_grad]: 0.00046209
[symbol_engine_optimizer]: 6.893e-05, [1]
    [Cycle 1]: 6.267e-05, [6]
        [build]: 2.35002e-06
        [elim_shapecalc]: 6.98e-06
        [elim_not_effective]: 1.195e-05
        [opt_reshape]: 5.27001e-06
        [fold_const_symbol]: 8.07e-06
        [renormalize]: 5.50004e-07
[validate]: 2.461e-05
[backend_pass]: 1.02998e-06
[task_emit]: 0.00481641
[execute]: 4.37e-06
Sums
    bootstrap                            :     0.000288s :  2.30%
    type_inference                       :     0.004244s : 33.81%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000059s :  0.47%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000046s :  0.37%
    rewriter_before_opt_a                :     0.000067s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000047s :  0.38%
    jit_opt_a.loop_unroll                :     0.000033s :  0.26%
    jit_opt_a.a_1                        :     0.000544s :  4.33%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.11%
    jit_opt_a.inline_without_move        :     0.000014s :  0.11%
    jit_opt_a.renormalize                :     0.000494s :  3.94%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.12%
    jit_opt_a.cse                        :     0.000024s :  0.19%
    jit_opt_a.replace_applicator         :     0.000016s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000028s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000483s :  3.85%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000013s :  0.11%
    cconv                                :     0.000018s :  0.14%
    loop_unroll                          :     0.000406s :  3.23%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000009s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000037s :  0.30%
    cse_after_recomputation.cse          :     0.000008s :  0.07%
    auto_monad_reorder                   :     0.000010s :  0.08%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000462s :  3.68%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000025s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.004816s : 38.37%
    execute                              :     0.000004s :  0.03%


Time group info:
------[substitution.]   0.000153    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.82% :     0.000001s :      2: substitution.fold_const_symbol
 2.58% :     0.000004s :      3: substitution.graph_param_transform
83.42% :     0.000128s :      4: substitution.inline
 1.56% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.50% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000003s :      2: substitution.replace_old_param
 6.25% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004190     2
83.40% :     0.003494s :      1: type_inference.infer
16.60% :     0.000696s :      1: type_inference.specialize
------[replace.]   0.000055     6
65.79% :     0.000036s :      4: replace.inline
34.21% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000134     6
93.68% :     0.000125s :      4: match.inline
 6.32% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000108   691
 1.31% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000001s :     10: predicate.addn_zero_filter
 2.23% :     0.000002s :     10: predicate.arithmetic_simplify
 1.39% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000001s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.37% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.04% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.36% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.31% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.34% :     0.000001s :      6: predicate.less_batch_normalization
 1.62% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.33% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.17% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000002s :     15: predicate.partial_eliminate
 1.36% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.43% :     0.000002s :     10: predicate.reduce_eliminate
 1.57% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.08% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000001s :     10: predicate.reshape_eliminate
 1.29% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.53% :     0.000002s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.20% :     0.000001s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.53% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000008s :     53: predicate.switch_simplify
 1.24% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.45% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000466    12
47.63% :     0.000222s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.26% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.11% :     0.000215s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017555    76
 0.23% :     0.000040s :      1: add_recomputation
 0.35% :     0.000062s :      1: auto_monad
 0.07% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.70% :     0.000299s :      1: bootstrap
 0.12% :     0.000020s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.04% :     0.000008s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
27.05% :     0.004749s :      1: jit_opt_a
 0.74% :     0.000130s :      1: jit_opt_after_cconv
 0.28% :     0.000048s :      1: jit_opt_b
 2.35% :     0.000413s :      1: loop_unroll
 2.79% :     0.000490s :      1: mutable_eliminate
 4.18% :     0.000733s :     26: opt.transform.jit_opt_a
 0.24% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.67% :     0.000469s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.28% :     0.000049s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000011s :      1: remove_dup_value
 1.51% :     0.000265s :      1: renormalize.infer
 1.27% :     0.000223s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000031s :      1: rewriter_after_opt_a
 0.40% :     0.000070s :      1: rewriter_before_opt_a
 0.41% :     0.000071s :      1: symbol_engine_optimizer
27.48% :     0.004824s :      1: task_emit
24.23% :     0.004254s :      1: type_inference
 0.23% :     0.000040s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.506.440 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc5cd30> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.506.957 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0141981, [30]
[bootstrap]: 0.00032446
[type_inference]: 0.00501573
[event_method]: 1.982e-05
[auto_monad]: 6.631e-05
[graph_reusing]: 6.27001e-06
[pre_auto_parallel]: 2.75997e-06
[py_interpret_to_execute]: 3.612e-05
[rewriter_before_opt_a]: 7.489e-05
[expand_dump_flag]: 3.28e-06
[jit_opt_a]: 0.00624869, [2]
    [Cycle 1]: 0.00176858, [27]
        [switch_simplify]: 5.897e-05
        [loop_unroll]: 2.94e-05
        [a_1]: 0.00056617
        [with_stream_mark]: 2.2e-05
        [recompute_prepare]: 9.61998e-06
        [updatestate_depend_eliminate]: 3.91001e-06
        [updatestate_assign_eliminate]: 2.83998e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [parameter_eliminate]: 2.07999e-06
        [specialize_transform]: 6.34999e-06
        [updatestate_useless_node_eliminater]: 5.14998e-06
        [accelerated_algorithm]: 5.51998e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 5.07e-06
        [merge_forward]: 3.51999e-06
        [cell_reuse_recompute_pass]: 1.90001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.588e-05
        [j_node_and_user_rematch]: 1.068e-05
        [meta_fg_expand]: 2.53e-06
        [replace_old_param]: 9.91e-06
        [inline_without_move]: 5.42001e-06
        [renormalize]: 0.0006794
        [add_forward_monad_depend]: 8.47e-06
        [auto_monad_grad]: 2.59999e-06
        [auto_monad_eliminator]: 1.607e-05
        [cse]: 2.896e-05
        [replace_applicator]: 1.556e-05
    [Cycle 2]: 0.00034599, [27]
        [switch_simplify]: 5.74999e-06
        [loop_unroll]: 4.89e-06
        [a_1]: 8.999e-05
        [with_stream_mark]: 1.447e-05
        [recompute_prepare]: 6.46999e-06
        [updatestate_depend_eliminate]: 3.6e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [parameter_eliminate]: 1.35999e-06
        [specialize_transform]: 5.34e-06
        [updatestate_useless_node_eliminater]: 4.68999e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 1.49998e-06
        [get_grad_eliminate_]: 4.55001e-06
        [merge_forward]: 3.6e-06
        [cell_reuse_recompute_pass]: 2.25002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.629e-05
        [j_node_and_user_rematch]: 8.48999e-06
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 8.35001e-06
        [inline_without_move]: 4.90999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.61998e-06
        [auto_monad_grad]: 1.09e-06
        [auto_monad_eliminator]: 6.28e-06
        [cse]: 1.209e-05
        [replace_applicator]: 6.46e-06
[py_interpret_to_execute_after_opt_a]: 1.401e-05
[rewriter_after_opt_a]: 3.932e-05
[convert_after_rewriter]: 7.71999e-06
[order_py_execute_after_rewriter]: 4.97e-06
[mutable_eliminate]: 0.00069259
[jit_opt_b]: 5.589e-05, [1]
    [Cycle 1]: 4.707e-05, [2]
        [frontend_op_eliminate]: 1.671e-05
        [inline_after_opt_a]: 1.686e-05
[cconv]: 2.771e-05
[loop_unroll]: 0.00047282
[jit_opt_after_cconv]: 0.00015122, [1]
    [Cycle 1]: 0.00014409, [11]
        [c_1]: 2.088e-05
        [parameter_eliminate]: 3.36999e-06
        [updatestate_depend_eliminate]: 6.40002e-06
        [updatestate_assign_eliminate]: 2.84999e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 2.327e-05
        [call_graph_tuple_transform]: 2.085e-05
        [tuple_list_get_item_eliminator]: 5.40999e-06
        [none_parameter_eliminate]: 1.72999e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 4.90999e-06
[remove_dup_value]: 1.517e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 4.53999e-06
[add_recomputation]: 4.957e-05
[cse_after_recomputation]: 2.253e-05, [1]
    [Cycle 1]: 1.655e-05, [1]
        [cse]: 9.54999e-06
[auto_monad_reorder]: 1.762e-05
[get_jit_bprop_graph]: 2.51998e-06
[rewriter_after_jit_bprop_graph]: 6.14001e-06
[opt_after_jit_grad]: 0.00051046
[symbol_engine_optimizer]: 8.042e-05, [1]
    [Cycle 1]: 7.326e-05, [6]
        [build]: 5.31002e-06
        [elim_shapecalc]: 8.65999e-06
        [elim_not_effective]: 1.459e-05
        [opt_reshape]: 5.46e-06
        [fold_const_symbol]: 9.02e-06
        [renormalize]: 8.59989e-07
[validate]: 3.809e-05
Sums
    bootstrap                            :     0.000324s :  3.46%
    type_inference                       :     0.005016s : 53.45%
    event_method                         :     0.000020s :  0.21%
    auto_monad                           :     0.000066s :  0.71%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.38%
    rewriter_before_opt_a                :     0.000075s :  0.80%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000065s :  0.69%
    jit_opt_a.loop_unroll                :     0.000034s :  0.37%
    jit_opt_a.a_1                        :     0.000656s :  6.99%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.39%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000018s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000679s :  7.24%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.11%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.24%
    jit_opt_a.cse                        :     0.000041s :  0.44%
    jit_opt_a.replace_applicator         :     0.000022s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.15%
    rewriter_after_opt_a                 :     0.000039s :  0.42%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000693s :  7.38%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.18%
    cconv                                :     0.000028s :  0.30%
    loop_unroll                          :     0.000473s :  5.04%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000050s :  0.53%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000018s :  0.19%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000510s :  5.44%
    symbol_engine_optimizer.build        :     0.000005s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000038s :  0.41%


Time group info:
------[substitution.]   0.000235    23
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000002s :      2: substitution.fold_const_symbol
 2.75% :     0.000006s :      3: substitution.graph_param_transform
83.86% :     0.000197s :      4: substitution.inline
 1.84% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.33% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.42% :     0.000006s :      2: substitution.replace_old_param
 5.29% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004946     2
82.07% :     0.004059s :      1: type_inference.infer
17.93% :     0.000887s :      1: type_inference.specialize
------[replace.]   0.000067     6
67.33% :     0.000045s :      4: replace.inline
32.67% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000205     6
94.54% :     0.000194s :      4: match.inline
 5.46% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.23% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.13% :     0.000001s :     10: predicate.addn_zero_filter
 2.25% :     0.000003s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.36% :     0.000000s :      3: predicate.graph_param_transform
 7.18% :     0.000009s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.18% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.61% :     0.000002s :     15: predicate.load_eliminater
 1.83% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.86% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.29% :     0.000002s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 2.16% :     0.000003s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.47% :     0.000003s :     15: predicate.partial_eliminate
 1.09% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.32% :     0.000002s :     10: predicate.reduce_eliminate
 1.78% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.81% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.06% :     0.000003s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.12% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.63% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.21% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.32% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.55% :     0.000009s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.08% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.29% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.37% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.67% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000639    12
50.29% :     0.000321s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.42% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.29% :     0.000270s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015822    72
 0.33% :     0.000053s :      1: add_recomputation
 0.44% :     0.000070s :      1: auto_monad
 0.13% :     0.000020s :      1: auto_monad_reorder
 2.18% :     0.000345s :      1: bootstrap
 0.19% :     0.000030s :      1: cconv
 0.08% :     0.000012s :      1: convert_after_rewriter
 0.16% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
39.52% :     0.006252s :      1: jit_opt_a
 0.97% :     0.000154s :      1: jit_opt_after_cconv
 0.37% :     0.000059s :      1: jit_opt_b
 3.04% :     0.000481s :      1: loop_unroll
 4.44% :     0.000703s :      1: mutable_eliminate
 5.60% :     0.000885s :     26: opt.transform.jit_opt_a
 0.30% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.29% :     0.000520s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000039s :      1: py_interpret_to_execute
 0.10% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.20% :     0.000348s :      1: renormalize.infer
 2.04% :     0.000322s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000043s :      1: rewriter_after_opt_a
 0.49% :     0.000078s :      1: rewriter_before_opt_a
 0.52% :     0.000083s :      1: symbol_engine_optimizer
31.79% :     0.005030s :      1: type_inference


TotalTime = 0.0166914, [33]
[bootstrap]: 0.00028954
[type_inference]: 0.00396582
[event_method]: 1.641e-05
[auto_monad]: 5.671e-05
[graph_reusing]: 5.59998e-06
[pre_auto_parallel]: 2.16e-06
[py_interpret_to_execute]: 2.711e-05
[rewriter_before_opt_a]: 6.826e-05
[expand_dump_flag]: 3.09001e-06
[jit_opt_a]: 0.00473847, [2]
    [Cycle 1]: 0.00129571, [27]
        [switch_simplify]: 5.663e-05
        [loop_unroll]: 2.737e-05
        [a_1]: 0.00046755
        [with_stream_mark]: 1.562e-05
        [recompute_prepare]: 6.70998e-06
        [updatestate_depend_eliminate]: 3.3e-06
        [updatestate_assign_eliminate]: 2.86e-06
        [updatestate_loads_eliminate]: 2.28002e-06
        [parameter_eliminate]: 1.74e-06
        [specialize_transform]: 5.61e-06
        [updatestate_useless_node_eliminater]: 4.77998e-06
        [accelerated_algorithm]: 5.48002e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 4.81002e-06
        [merge_forward]: 3.39001e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.317e-05
        [j_node_and_user_rematch]: 8.23999e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 7.61999e-06
        [inline_without_move]: 4.82998e-06
        [renormalize]: 0.00044226
        [add_forward_monad_depend]: 5.18002e-06
        [auto_monad_grad]: 1.61998e-06
        [auto_monad_eliminator]: 1.246e-05
        [cse]: 2.372e-05
        [replace_applicator]: 1.057e-05
    [Cycle 2]: 0.00035898, [27]
        [switch_simplify]: 4.95999e-06
        [loop_unroll]: 4.62e-06
        [a_1]: 0.00013075
        [with_stream_mark]: 8.75001e-06
        [recompute_prepare]: 5.54e-06
        [updatestate_depend_eliminate]: 2.64999e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 1.96003e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 4.89998e-06
        [updatestate_useless_node_eliminater]: 4.61002e-06
        [accelerated_algorithm]: 4.91002e-06
        [meta_shard_fg_expand]: 1.02e-06
        [get_grad_eliminate_]: 4.58999e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.63002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.189e-05
        [j_node_and_user_rematch]: 7.98001e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 7.18e-06
        [inline_without_move]: 4.62998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.31002e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 5.24998e-06
        [cse]: 1.078e-05
        [replace_applicator]: 5.45001e-06
[py_interpret_to_execute_after_opt_a]: 8.28001e-06
[rewriter_after_opt_a]: 3.056e-05
[convert_after_rewriter]: 6.07999e-06
[order_py_execute_after_rewriter]: 5.03002e-06
[mutable_eliminate]: 0.00046559
[jit_opt_b]: 4.57e-05, [1]
    [Cycle 1]: 3.948e-05, [2]
        [frontend_op_eliminate]: 1.505e-05
        [inline_after_opt_a]: 1.378e-05
[cconv]: 1.971e-05
[loop_unroll]: 0.00040026
[jit_opt_after_cconv]: 0.00012814, [1]
    [Cycle 1]: 0.00012194, [11]
        [c_1]: 1.915e-05
        [parameter_eliminate]: 2.49001e-06
        [updatestate_depend_eliminate]: 4.65999e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [cse]: 1.438e-05
        [call_graph_tuple_transform]: 1.649e-05
        [tuple_list_get_item_eliminator]: 5.34e-06
        [none_parameter_eliminate]: 1.30001e-06
        [renormalize]: 6.89994e-07
        [switch_simplify]: 5.54e-06
[remove_dup_value]: 1.276e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 4.95999e-06
[add_recomputation]: 4.418e-05
[cse_after_recomputation]: 1.897e-05, [1]
    [Cycle 1]: 1.364e-05, [1]
        [cse]: 8.06001e-06
[auto_monad_reorder]: 1.346e-05
[get_jit_bprop_graph]: 1.42999e-06
[rewriter_after_jit_bprop_graph]: 4.48001e-06
[opt_after_jit_grad]: 0.00045855
[symbol_engine_optimizer]: 6.978e-05, [1]
    [Cycle 1]: 6.351e-05, [6]
        [build]: 2.89999e-06
        [elim_shapecalc]: 7.01999e-06
        [elim_not_effective]: 1.211e-05
        [opt_reshape]: 5.58002e-06
        [fold_const_symbol]: 8.42e-06
        [renormalize]: 6.30011e-07
[validate]: 2.918e-05
[backend_pass]: 6.50005e-07
[task_emit]: 0.0055776
[execute]: 6.57002e-06
Sums
    bootstrap                            :     0.000290s :  2.22%
    type_inference                       :     0.003966s : 30.37%
    event_method                         :     0.000016s :  0.13%
    auto_monad                           :     0.000057s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000027s :  0.21%
    rewriter_before_opt_a                :     0.000068s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.47%
    jit_opt_a.loop_unroll                :     0.000032s :  0.24%
    jit_opt_a.a_1                        :     0.000598s :  4.58%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000442s :  3.39%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.14%
    jit_opt_a.cse                        :     0.000035s :  0.26%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000466s :  3.56%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000020s :  0.15%
    loop_unroll                          :     0.000400s :  3.06%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000014s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000044s :  0.34%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000013s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000459s :  3.51%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000029s :  0.22%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005578s : 42.71%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000162    23
 1.10% :     0.000002s :      2: substitution.elim_not_effective
 0.81% :     0.000001s :      2: substitution.fold_const_symbol
 2.70% :     0.000004s :      3: substitution.graph_param_transform
84.64% :     0.000137s :      4: substitution.inline
 1.80% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.67% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.93% :     0.000003s :      2: substitution.replace_old_param
 4.34% :     0.000007s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003914     2
83.12% :     0.003253s :      1: type_inference.infer
16.88% :     0.000661s :      1: type_inference.specialize
------[replace.]   0.000057     6
66.88% :     0.000038s :      4: replace.inline
33.12% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000141     6
95.76% :     0.000135s :      4: match.inline
 4.24% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.67% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.54% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000001s :     10: predicate.addn_zero_filter
 1.87% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000001s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.42% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000001s :     10: predicate.environ_get_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.67% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000001s :      6: predicate.less_batch_normalization
 1.62% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.61% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     15: predicate.partial_eliminate
 1.37% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.40% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.28% :     0.000001s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.97% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.60% :     0.000002s :     10: predicate.same_eliminate
 0.67% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.44% :     0.000008s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.48% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000448    12
48.40% :     0.000217s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.96% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.65% :     0.000204s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018025    76
 0.26% :     0.000047s :      1: add_recomputation
 0.33% :     0.000060s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.67% :     0.000301s :      1: bootstrap
 0.12% :     0.000022s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.30% :     0.004741s :      1: jit_opt_a
 0.73% :     0.000131s :      1: jit_opt_after_cconv
 0.27% :     0.000048s :      1: jit_opt_b
 2.26% :     0.000407s :      1: loop_unroll
 2.63% :     0.000474s :      1: mutable_eliminate
 4.36% :     0.000786s :     26: opt.transform.jit_opt_a
 0.24% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.58% :     0.000466s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.26% :     0.000228s :      1: renormalize.infer
 1.15% :     0.000208s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000033s :      1: rewriter_after_opt_a
 0.39% :     0.000071s :      1: rewriter_before_opt_a
 0.40% :     0.000072s :      1: symbol_engine_optimizer
30.99% :     0.005586s :      1: task_emit
22.07% :     0.003978s :      1: type_inference
 0.26% :     0.000047s :      1: validate


TotalTime = 0.0274009, [33]
[bootstrap]: 0.00019645
[type_inference]: 0.00932733
[event_method]: 3.778e-05
[auto_monad]: 6.595e-05
[graph_reusing]: 4.72e-06
[pre_auto_parallel]: 1.94e-06
[py_interpret_to_execute]: 2.232e-05
[rewriter_before_opt_a]: 7.893e-05
[expand_dump_flag]: 1.96998e-06
[jit_opt_a]: 0.0100301, [3]
    [Cycle 1]: 0.0048689, [27]
        [switch_simplify]: 5.796e-05
        [loop_unroll]: 3.713e-05
        [a_1]: 0.00077467
        [with_stream_mark]: 1.908e-05
        [recompute_prepare]: 1.672e-05
        [updatestate_depend_eliminate]: 6.76e-06
        [updatestate_assign_eliminate]: 5.53997e-06
        [updatestate_loads_eliminate]: 5.45001e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 1.275e-05
        [updatestate_useless_node_eliminater]: 1.163e-05
        [accelerated_algorithm]: 1.179e-05
        [meta_shard_fg_expand]: 2.39001e-06
        [get_grad_eliminate_]: 1.13e-05
        [merge_forward]: 6.56e-06
        [cell_reuse_recompute_pass]: 7.50006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.277e-05
        [j_node_and_user_rematch]: 2.179e-05
        [meta_fg_expand]: 0.00136456
        [replace_old_param]: 4.78e-05
        [inline_without_move]: 4.828e-05
        [renormalize]: 0.00194866
        [add_forward_monad_depend]: 8.10999e-06
        [auto_monad_grad]: 3.83001e-06
        [auto_monad_eliminator]: 4.103e-05
        [cse]: 0.00012889
        [replace_applicator]: 6.043e-05
    [Cycle 2]: 0.00176699, [27]
        [switch_simplify]: 3.425e-05
        [loop_unroll]: 3.288e-05
        [a_1]: 0.00097957
        [with_stream_mark]: 1.1e-05
        [recompute_prepare]: 7.35e-06
        [updatestate_depend_eliminate]: 3.42002e-06
        [updatestate_assign_eliminate]: 2.92002e-06
        [updatestate_loads_eliminate]: 2.41998e-06
        [parameter_eliminate]: 1.40001e-06
        [specialize_transform]: 6.26998e-06
        [updatestate_useless_node_eliminater]: 5.84e-06
        [accelerated_algorithm]: 5.59e-06
        [meta_shard_fg_expand]: 1.45999e-06
        [get_grad_eliminate_]: 5.94e-06
        [merge_forward]: 2.83e-06
        [cell_reuse_recompute_pass]: 7.39994e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.192e-05
        [j_node_and_user_rematch]: 9.25001e-06
        [meta_fg_expand]: 2.77e-05
        [replace_old_param]: 9.39e-06
        [inline_without_move]: 5.86998e-06
        [renormalize]: 0.00040821
        [add_forward_monad_depend]: 4.17e-06
        [auto_monad_grad]: 1.24e-06
        [auto_monad_eliminator]: 1.087e-05
        [cse]: 1.909e-05
        [replace_applicator]: 1.227e-05
    [Cycle 3]: 0.00034995, [27]
        [switch_simplify]: 6.49999e-06
        [loop_unroll]: 5.61e-06
        [a_1]: 0.00011395
        [with_stream_mark]: 8.33001e-06
        [recompute_prepare]: 5.54e-06
        [updatestate_depend_eliminate]: 3.08e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 5.95002e-06
        [updatestate_useless_node_eliminater]: 5.61003e-06
        [accelerated_algorithm]: 5.48002e-06
        [meta_shard_fg_expand]: 1.27999e-06
        [get_grad_eliminate_]: 5.58002e-06
        [merge_forward]: 2.83e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.36e-05
        [j_node_and_user_rematch]: 9.54e-06
        [meta_fg_expand]: 1.86998e-06
        [replace_old_param]: 8.07998e-06
        [inline_without_move]: 5.47001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.98998e-06
        [cse]: 1.354e-05
        [replace_applicator]: 5.77001e-06
[py_interpret_to_execute_after_opt_a]: 9.02e-06
[rewriter_after_opt_a]: 2.866e-05
[convert_after_rewriter]: 6.89999e-06
[order_py_execute_after_rewriter]: 4.53001e-06
[mutable_eliminate]: 0.00047476
[jit_opt_b]: 5.177e-05, [1]
    [Cycle 1]: 4.549e-05, [2]
        [frontend_op_eliminate]: 1.749e-05
        [inline_after_opt_a]: 1.695e-05
[cconv]: 1.51e-05
[loop_unroll]: 0.00041194
[jit_opt_after_cconv]: 0.00014136, [1]
    [Cycle 1]: 0.00013557, [11]
        [c_1]: 2.407e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 5.37001e-06
        [updatestate_assign_eliminate]: 3.24001e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [cse]: 2.018e-05
        [call_graph_tuple_transform]: 1.77e-05
        [tuple_list_get_item_eliminator]: 6.04999e-06
        [none_parameter_eliminate]: 7.7e-07
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.85002e-06
[remove_dup_value]: 1.085e-05
[partial_unused_args_eliminate]: 6.79999e-06
[environ_conv]: 4.37998e-06
[add_recomputation]: 4.055e-05
[cse_after_recomputation]: 2.539e-05, [1]
    [Cycle 1]: 2.01e-05, [1]
        [cse]: 1.437e-05
[auto_monad_reorder]: 1.246e-05
[get_jit_bprop_graph]: 1.51002e-06
[rewriter_after_jit_bprop_graph]: 4.59998e-06
[opt_after_jit_grad]: 0.00044806
[symbol_engine_optimizer]: 7.321e-05, [1]
    [Cycle 1]: 6.756e-05, [6]
        [build]: 2.78e-06
        [elim_shapecalc]: 8.58001e-06
        [elim_not_effective]: 1.375e-05
        [opt_reshape]: 6.39999e-06
        [fold_const_symbol]: 9.82001e-06
        [renormalize]: 4.09986e-07
[validate]: 2.891e-05
[backend_pass]: 7.7e-07
[task_emit]: 0.00565449
[execute]: 5.51e-06
Sums
    bootstrap                            :     0.000196s :  0.83%
    type_inference                       :     0.009327s : 39.47%
    event_method                         :     0.000038s :  0.16%
    auto_monad                           :     0.000066s :  0.28%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000022s :  0.09%
    rewriter_before_opt_a                :     0.000079s :  0.33%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000099s :  0.42%
    jit_opt_a.loop_unroll                :     0.000076s :  0.32%
    jit_opt_a.a_1                        :     0.001868s :  7.91%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000030s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000013s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000011s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000025s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000023s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.10%
    jit_opt_a.merge_forward              :     0.000012s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000048s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000041s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001394s :  5.90%
    jit_opt_a.replace_old_param          :     0.000065s :  0.28%
    jit_opt_a.inline_without_move        :     0.000060s :  0.25%
    jit_opt_a.renormalize                :     0.002357s :  9.97%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000006s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000058s :  0.24%
    jit_opt_a.cse                        :     0.000162s :  0.68%
    jit_opt_a.replace_applicator         :     0.000078s :  0.33%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.04%
    rewriter_after_opt_a                 :     0.000029s :  0.12%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000475s :  2.01%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.07%
    cconv                                :     0.000015s :  0.06%
    loop_unroll                          :     0.000412s :  1.74%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000011s :  0.05%
    partial_unused_args_eliminate        :     0.000007s :  0.03%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000041s :  0.17%
    cse_after_recomputation.cse          :     0.000014s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000448s :  1.90%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005654s : 23.93%
    execute                              :     0.000006s :  0.02%


Time group info:
------[substitution.]   0.000411   110
 0.43% :     0.000002s :      3: substitution.elim_not_effective
 0.40% :     0.000002s :      3: substitution.fold_const_symbol
 0.99% :     0.000004s :      4: substitution.graph_param_transform
61.99% :     0.000255s :     12: substitution.inline
 3.57% :     0.000015s :      2: substitution.inline_without_move
 2.06% :     0.000008s :     14: substitution.j_node_and_user_rematch
 2.57% :     0.000011s :      7: substitution.minmaximum_grad
 1.31% :     0.000005s :      5: substitution.partial_eliminate
 2.47% :     0.000010s :     14: substitution.remove_not_recompute_node
 5.60% :     0.000023s :     10: substitution.replace_applicator
 1.62% :     0.000007s :      7: substitution.replace_old_param
 0.57% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.63% :     0.000019s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.09% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 8.70% :     0.000036s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009268     2
89.88% :     0.008330s :      1: type_inference.infer
10.12% :     0.000938s :      1: type_inference.specialize
------[replace.]   0.000131    19
64.26% :     0.000084s :     12: replace.inline
35.74% :     0.000047s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000268    19
92.93% :     0.000249s :     12: match.inline
 7.07% :     0.000019s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000351  2568
 1.64% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.54% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.54% :     0.000005s :     43: predicate.addn_check_dump
 1.68% :     0.000006s :     43: predicate.addn_zero_filter
 2.10% :     0.000007s :     43: predicate.arithmetic_simplify
 1.52% :     0.000005s :     43: predicate.cast_eliminate
 0.17% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.51% :     0.000005s :     43: predicate.compare_switch_simplify
 1.60% :     0.000006s :     43: predicate.depend_value_elim
 1.52% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.54% :     0.000005s :     43: predicate.dict_get_item_eliminator
 1.54% :     0.000005s :     43: predicate.dict_set_item_eliminator
 0.31% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      4: predicate.elim_not_effective
 0.22% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.56% :     0.000005s :     43: predicate.environ_add_const_eliminate
 1.47% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.46% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.51% :     0.000005s :     43: predicate.environ_get_eliminate
 1.47% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.81% :     0.000003s :     19: predicate.get_grad_eliminate
 0.11% :     0.000000s :      4: predicate.graph_param_transform
 4.20% :     0.000015s :     70: predicate.inline
 1.85% :     0.000006s :     40: predicate.inline_without_move
 0.36% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.84% :     0.000003s :     19: predicate.less_batch_normalization
 1.80% :     0.000006s :     50: predicate.list_to_tuple_eliminator_
 2.09% :     0.000007s :     54: predicate.load_eliminater
 0.51% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.26% :     0.000011s :     85: predicate.loop_unroll_before_grad
 1.77% :     0.000006s :     47: predicate.make_slice_get_slice_eliminator
 1.47% :     0.000005s :     43: predicate.merge_addn
 1.57% :     0.000005s :     43: predicate.minmaximum_grad
 0.55% :     0.000002s :      4: predicate.mutable_eliminate
 0.20% :     0.000001s :      4: predicate.opt_reshape
 2.33% :     0.000008s :     54: predicate.partial_eliminate
 1.64% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.88% :     0.000007s :     43: predicate.reduce_eliminate
 1.80% :     0.000006s :     50: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :     19: predicate.remove_not_recompute_node
 2.52% :     0.000009s :     96: predicate.replace_applicator
 0.94% :     0.000003s :     40: predicate.replace_old_param
 0.13% :     0.000000s :      4: predicate.reset_defer_inline
 1.69% :     0.000006s :     43: predicate.reshape_eliminate
 1.57% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.26% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.57% :     0.000005s :     43: predicate.same_eliminate
 0.47% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.34% :     0.000001s :      8: predicate.special_op_eliminate
 0.89% :     0.000003s :     19: predicate.specialize_transform
 1.69% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000005s :     43: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.77% :     0.000010s :     62: predicate.switch_defer_inline
 2.55% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.31% :     0.000022s :    151: predicate.switch_simplify
 1.55% :     0.000005s :     43: predicate.tile_eliminate
 1.51% :     0.000005s :     43: predicate.transpose_eliminate
 1.80% :     0.000006s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.65% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.01% :     0.000011s :     58: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000006s :     43: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000006s :     50: predicate.tuple_to_list_eliminator_
 1.94% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000010s :     73: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000006s :     43: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.23% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001232    29
61.37% :     0.000756s :     13: func_graph_cloner_run.FuncGraphClonerGraph
38.63% :     0.000476s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.032300    91
 0.13% :     0.000043s :      1: add_recomputation
 0.22% :     0.000070s :      1: auto_monad
 0.05% :     0.000015s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.65% :     0.000209s :      1: bootstrap
 0.05% :     0.000018s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000028s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.13% :     0.000041s :      1: event_method
 0.03% :     0.000009s :      1: execute
 0.01% :     0.000004s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
31.06% :     0.010033s :      1: jit_opt_a
 0.45% :     0.000144s :      1: jit_opt_after_cconv
 0.17% :     0.000054s :      1: jit_opt_b
 1.30% :     0.000420s :      1: loop_unroll
 1.49% :     0.000483s :      1: mutable_eliminate
 7.50% :     0.002421s :     39: opt.transform.jit_opt_a
 0.16% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000028s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.41% :     0.000456s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000009s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.08% :     0.000025s :      1: py_interpret_to_execute
 0.04% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000013s :      1: remove_dup_value
 3.75% :     0.001212s :      2: renormalize.infer
 3.51% :     0.001134s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000031s :      1: rewriter_after_opt_a
 0.25% :     0.000082s :      1: rewriter_before_opt_a
 0.23% :     0.000076s :      1: symbol_engine_optimizer
17.53% :     0.005663s :      1: task_emit
28.91% :     0.009338s :      1: type_inference
 0.15% :     0.000047s :      1: validate


TotalTime = 0.0211351, [33]
[bootstrap]: 0.000289
[type_inference]: 0.00475688
[event_method]: 2.009e-05
[auto_monad]: 6.762e-05
[graph_reusing]: 5.99e-06
[pre_auto_parallel]: 2.46998e-06
[py_interpret_to_execute]: 3.931e-05
[rewriter_before_opt_a]: 7.453e-05
[expand_dump_flag]: 3.25998e-06
[jit_opt_a]: 0.00597666, [2]
    [Cycle 1]: 0.0020413, [27]
        [switch_simplify]: 6.472e-05
        [loop_unroll]: 2.966e-05
        [a_1]: 0.00067488
        [with_stream_mark]: 2.369e-05
        [recompute_prepare]: 8.94003e-06
        [updatestate_depend_eliminate]: 3.73001e-06
        [updatestate_assign_eliminate]: 3.36999e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 5.77001e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 6.24001e-06
        [meta_shard_fg_expand]: 2.55002e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 3.52997e-06
        [cell_reuse_recompute_pass]: 1.80001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.522e-05
        [j_node_and_user_rematch]: 9.30001e-06
        [meta_fg_expand]: 2.16e-06
        [replace_old_param]: 9.54e-06
        [inline_without_move]: 5.94e-06
        [renormalize]: 0.00089038
        [add_forward_monad_depend]: 7.08e-06
        [auto_monad_grad]: 2.89001e-06
        [auto_monad_eliminator]: 1.568e-05
        [cse]: 2.831e-05
        [replace_applicator]: 1.535e-05
    [Cycle 2]: 0.00034436, [27]
        [switch_simplify]: 5.76e-06
        [loop_unroll]: 5.30999e-06
        [a_1]: 9.603e-05
        [with_stream_mark]: 1.273e-05
        [recompute_prepare]: 4.70001e-06
        [updatestate_depend_eliminate]: 3.26001e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 1.96998e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 4.87e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 1.40001e-06
        [get_grad_eliminate_]: 4.33999e-06
        [merge_forward]: 3.33998e-06
        [cell_reuse_recompute_pass]: 3.4e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.599e-05
        [j_node_and_user_rematch]: 8.97e-06
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 9.19998e-06
        [inline_without_move]: 5.17e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.69e-06
        [auto_monad_grad]: 1.24998e-06
        [auto_monad_eliminator]: 5.62999e-06
        [cse]: 1.237e-05
        [replace_applicator]: 5.72001e-06
[py_interpret_to_execute_after_opt_a]: 1.884e-05
[rewriter_after_opt_a]: 3.892e-05
[convert_after_rewriter]: 6.93e-06
[order_py_execute_after_rewriter]: 5.05999e-06
[mutable_eliminate]: 0.00077165
[jit_opt_b]: 5.885e-05, [1]
    [Cycle 1]: 4.979e-05, [2]
        [frontend_op_eliminate]: 1.78e-05
        [inline_after_opt_a]: 1.78e-05
[cconv]: 3.337e-05
[loop_unroll]: 0.00049792
[jit_opt_after_cconv]: 0.00015894, [1]
    [Cycle 1]: 0.00015204, [11]
        [c_1]: 2.316e-05
        [parameter_eliminate]: 3.83001e-06
        [updatestate_depend_eliminate]: 5.59e-06
        [updatestate_assign_eliminate]: 3.2e-06
        [updatestate_loads_eliminate]: 2.33002e-06
        [cse]: 2.742e-05
        [call_graph_tuple_transform]: 2.104e-05
        [tuple_list_get_item_eliminator]: 5.88998e-06
        [none_parameter_eliminate]: 1.34e-06
        [renormalize]: 5.09986e-07
        [switch_simplify]: 5.38002e-06
[remove_dup_value]: 1.424e-05
[partial_unused_args_eliminate]: 2.29999e-06
[environ_conv]: 4.66002e-06
[add_recomputation]: 4.92e-05
[cse_after_recomputation]: 2.128e-05, [1]
    [Cycle 1]: 1.536e-05, [1]
        [cse]: 9.37001e-06
[auto_monad_reorder]: 1.43e-05
[get_jit_bprop_graph]: 2.17001e-06
[rewriter_after_jit_bprop_graph]: 6.07999e-06
[opt_after_jit_grad]: 0.00057006
[symbol_engine_optimizer]: 8.212e-05, [1]
    [Cycle 1]: 7.361e-05, [6]
        [build]: 4.32e-06
        [elim_shapecalc]: 8.15999e-06
        [elim_not_effective]: 1.444e-05
        [opt_reshape]: 6.54999e-06
        [fold_const_symbol]: 8.99e-06
        [renormalize]: 5.19998e-07
[validate]: 3.83e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00727027
[execute]: 8.24002e-06
Sums
    bootstrap                            :     0.000289s :  1.71%
    type_inference                       :     0.004757s : 28.19%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000068s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.23%
    rewriter_before_opt_a                :     0.000075s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000070s :  0.42%
    jit_opt_a.loop_unroll                :     0.000035s :  0.21%
    jit_opt_a.a_1                        :     0.000771s :  4.57%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000890s :  5.28%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.13%
    jit_opt_a.cse                        :     0.000041s :  0.24%
    jit_opt_a.replace_applicator         :     0.000021s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.11%
    rewriter_after_opt_a                 :     0.000039s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000772s :  4.57%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.11%
    cconv                                :     0.000033s :  0.20%
    loop_unroll                          :     0.000498s :  2.95%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000570s :  3.38%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007270s : 43.08%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000327    23
 0.68% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000002s :      2: substitution.fold_const_symbol
 1.84% :     0.000006s :      3: substitution.graph_param_transform
88.02% :     0.000288s :      4: substitution.inline
 1.36% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.55% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.67% :     0.000005s :      2: substitution.replace_old_param
 4.36% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004680     2
79.43% :     0.003718s :      1: type_inference.infer
20.57% :     0.000963s :      1: type_inference.specialize
------[replace.]   0.000080     6
66.49% :     0.000053s :      4: replace.inline
33.51% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000298     6
95.63% :     0.000285s :      4: match.inline
 4.37% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.66% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.14% :     0.000001s :     10: predicate.addn_zero_filter
 1.95% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000001s :     10: predicate.compare_switch_simplify
 1.82% :     0.000002s :     10: predicate.depend_value_elim
 0.99% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.04% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.36% :     0.000002s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 5.41% :     0.000007s :     22: predicate.inline
 1.04% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.63% :     0.000002s :      6: predicate.less_batch_normalization
 1.75% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.17% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.96% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.27% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.96% :     0.000004s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.38% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.41% :     0.000002s :     10: predicate.reduce_eliminate
 1.32% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.45% :     0.000002s :     18: predicate.replace_applicator
 0.86% :     0.000001s :      6: predicate.replace_old_param
 0.59% :     0.000001s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.19% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.81% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.20% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.61% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.04% :     0.000010s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.06% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.27% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000735    12
38.60% :     0.000284s :      5: func_graph_cloner_run.FuncGraphClonerGraph
16.82% :     0.000124s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.58% :     0.000327s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023136    76
 0.23% :     0.000052s :      1: add_recomputation
 0.31% :     0.000071s :      1: auto_monad
 0.07% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.31% :     0.000303s :      1: bootstrap
 0.16% :     0.000036s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.85% :     0.005981s :      1: jit_opt_a
 0.70% :     0.000162s :      1: jit_opt_after_cconv
 0.27% :     0.000062s :      1: jit_opt_b
 2.19% :     0.000508s :      1: loop_unroll
 3.40% :     0.000787s :      1: mutable_eliminate
 4.33% :     0.001001s :     26: opt.transform.jit_opt_a
 0.23% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000029s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.50% :     0.000579s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000042s :      1: py_interpret_to_execute
 0.09% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.97% :     0.000456s :      1: renormalize.infer
 1.84% :     0.000425s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000042s :      1: rewriter_after_opt_a
 0.34% :     0.000078s :      1: rewriter_before_opt_a
 0.37% :     0.000085s :      1: symbol_engine_optimizer
31.48% :     0.007283s :      1: task_emit
20.64% :     0.004776s :      1: type_inference
 0.25% :     0.000057s :      1: validate


TotalTime = 0.013849, [30]
[bootstrap]: 0.00032272
[type_inference]: 0.00457794
[event_method]: 1.837e-05
[auto_monad]: 7.741e-05
[graph_reusing]: 5.91e-06
[pre_auto_parallel]: 2.84001e-06
[py_interpret_to_execute]: 3.663e-05
[rewriter_before_opt_a]: 7.254e-05
[expand_dump_flag]: 3.83001e-06
[jit_opt_a]: 0.00634858, [2]
    [Cycle 1]: 0.0017717, [27]
        [switch_simplify]: 6.199e-05
        [loop_unroll]: 2.951e-05
        [a_1]: 0.00058557
        [with_stream_mark]: 2.146e-05
        [recompute_prepare]: 8.92e-06
        [updatestate_depend_eliminate]: 3.44001e-06
        [updatestate_assign_eliminate]: 3.53e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 2.31e-06
        [specialize_transform]: 6.01998e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 5.59998e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 4.95001e-06
        [merge_forward]: 3.53e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.416e-05
        [j_node_and_user_rematch]: 9.47999e-06
        [meta_fg_expand]: 2.32001e-06
        [replace_old_param]: 9.97001e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 0.00073524
        [add_forward_monad_depend]: 6.31e-06
        [auto_monad_grad]: 2.34999e-06
        [auto_monad_eliminator]: 1.52e-05
        [cse]: 2.681e-05
        [replace_applicator]: 1.356e-05
    [Cycle 2]: 0.00036647, [27]
        [switch_simplify]: 5.47001e-06
        [loop_unroll]: 4.91002e-06
        [a_1]: 0.00012205
        [with_stream_mark]: 1.129e-05
        [recompute_prepare]: 6.68e-06
        [updatestate_depend_eliminate]: 3.04999e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 1.11997e-06
        [specialize_transform]: 5.75001e-06
        [updatestate_useless_node_eliminater]: 4.86002e-06
        [accelerated_algorithm]: 5.15001e-06
        [meta_shard_fg_expand]: 1.22999e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 3.16001e-06
        [cell_reuse_recompute_pass]: 1.76998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.475e-05
        [j_node_and_user_rematch]: 8.82999e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 8.55001e-06
        [inline_without_move]: 4.56002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.41002e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 5.00001e-06
        [cse]: 1.215e-05
        [replace_applicator]: 4.88001e-06
[py_interpret_to_execute_after_opt_a]: 1.265e-05
[rewriter_after_opt_a]: 3.685e-05
[convert_after_rewriter]: 6.73e-06
[order_py_execute_after_rewriter]: 7.1e-06
[mutable_eliminate]: 0.00071731
[jit_opt_b]: 5.132e-05, [1]
    [Cycle 1]: 4.322e-05, [2]
        [frontend_op_eliminate]: 1.61e-05
        [inline_after_opt_a]: 1.513e-05
[cconv]: 2.708e-05
[loop_unroll]: 0.00044438
[jit_opt_after_cconv]: 0.00015313, [1]
    [Cycle 1]: 0.00014595, [11]
        [c_1]: 2.091e-05
        [parameter_eliminate]: 2.49001e-06
        [updatestate_depend_eliminate]: 7.11999e-06
        [updatestate_assign_eliminate]: 2.78998e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [cse]: 2.327e-05
        [call_graph_tuple_transform]: 2.077e-05
        [tuple_list_get_item_eliminator]: 5.91e-06
        [none_parameter_eliminate]: 1.72999e-06
        [renormalize]: 1.09e-06
        [switch_simplify]: 5.81003e-06
[remove_dup_value]: 1.571e-05
[partial_unused_args_eliminate]: 2.56e-06
[environ_conv]: 6.33e-06
[add_recomputation]: 4.851e-05
[cse_after_recomputation]: 2.41e-05, [1]
    [Cycle 1]: 1.762e-05, [1]
        [cse]: 1.074e-05
[auto_monad_reorder]: 1.524e-05
[get_jit_bprop_graph]: 2.13002e-06
[rewriter_after_jit_bprop_graph]: 7.36001e-06
[opt_after_jit_grad]: 0.0005127
[symbol_engine_optimizer]: 7.531e-05, [1]
    [Cycle 1]: 6.82e-05, [6]
        [build]: 3.86999e-06
        [elim_shapecalc]: 7.53999e-06
        [elim_not_effective]: 1.368e-05
        [opt_reshape]: 5.35999e-06
        [fold_const_symbol]: 8.43999e-06
        [renormalize]: 5.79981e-07
[validate]: 3.874e-05
Sums
    bootstrap                            :     0.000323s :  3.57%
    type_inference                       :     0.004578s : 50.69%
    event_method                         :     0.000018s :  0.20%
    auto_monad                           :     0.000077s :  0.86%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000037s :  0.41%
    rewriter_before_opt_a                :     0.000073s :  0.80%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000067s :  0.75%
    jit_opt_a.loop_unroll                :     0.000034s :  0.38%
    jit_opt_a.a_1                        :     0.000708s :  7.84%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.36%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000735s :  8.14%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.22%
    jit_opt_a.cse                        :     0.000039s :  0.43%
    jit_opt_a.replace_applicator         :     0.000018s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.14%
    rewriter_after_opt_a                 :     0.000037s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000007s :  0.08%
    mutable_eliminate                    :     0.000717s :  7.94%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000027s :  0.30%
    loop_unroll                          :     0.000444s :  4.92%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.26%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000016s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000049s :  0.54%
    cse_after_recomputation.cse          :     0.000011s :  0.12%
    auto_monad_reorder                   :     0.000015s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.08%
    opt_after_jit_grad                   :     0.000513s :  5.68%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.43%


Time group info:
------[substitution.]   0.000229    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.45% :     0.000006s :      3: substitution.graph_param_transform
83.76% :     0.000192s :      4: substitution.inline
 1.55% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.19% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.43% :     0.000006s :      2: substitution.replace_old_param
 6.07% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004508     2
80.71% :     0.003638s :      1: type_inference.infer
19.29% :     0.000870s :      1: type_inference.specialize
------[replace.]   0.000073     6
65.16% :     0.000048s :      4: replace.inline
34.84% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000203     6
93.66% :     0.000190s :      4: match.inline
 6.34% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.22% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.03% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.50% :     0.000002s :     10: predicate.addn_zero_filter
 1.88% :     0.000002s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.22% :     0.000001s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      3: predicate.elim_not_effective
 0.81% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.40% :     0.000002s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.51% :     0.000001s :      3: predicate.graph_param_transform
 6.69% :     0.000008s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      6: predicate.less_batch_normalization
 1.71% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.92% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.85% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.70% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.69% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.04% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.24% :     0.000001s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.21% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.55% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.69% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.27% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.08% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.48% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000574    12
45.78% :     0.000263s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.56% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.66% :     0.000268s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015573    72
 0.33% :     0.000052s :      1: add_recomputation
 0.53% :     0.000082s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.16% :     0.000337s :      1: bootstrap
 0.19% :     0.000030s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
40.79% :     0.006352s :      1: jit_opt_a
 1.00% :     0.000156s :      1: jit_opt_after_cconv
 0.35% :     0.000054s :      1: jit_opt_b
 2.92% :     0.000454s :      1: loop_unroll
 4.66% :     0.000726s :      1: mutable_eliminate
 5.98% :     0.000931s :     26: opt.transform.jit_opt_a
 0.32% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.36% :     0.000522s :      1: opt_after_jit_grad
 0.06% :     0.000009s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000039s :      1: py_interpret_to_execute
 0.10% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000019s :      1: remove_dup_value
 2.59% :     0.000404s :      1: renormalize.infer
 2.08% :     0.000324s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000040s :      1: rewriter_after_opt_a
 0.49% :     0.000076s :      1: rewriter_before_opt_a
 0.50% :     0.000078s :      1: symbol_engine_optimizer
29.49% :     0.004593s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.541.364 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc418b0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.541.912 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0172758, [33]
[bootstrap]: 0.00036221
[type_inference]: 0.00405568
[event_method]: 1.544e-05
[auto_monad]: 5.982e-05
[graph_reusing]: 5.11002e-06
[pre_auto_parallel]: 2.09e-06
[py_interpret_to_execute]: 2.62e-05
[rewriter_before_opt_a]: 6.781e-05
[expand_dump_flag]: 2.90998e-06
[jit_opt_a]: 0.00498535, [2]
    [Cycle 1]: 0.00140825, [27]
        [switch_simplify]: 5.093e-05
        [loop_unroll]: 3.097e-05
        [a_1]: 0.00054869
        [with_stream_mark]: 1.47e-05
        [recompute_prepare]: 7.62002e-06
        [updatestate_depend_eliminate]: 4.05e-06
        [updatestate_assign_eliminate]: 3.19001e-06
        [updatestate_loads_eliminate]: 2.41998e-06
        [parameter_eliminate]: 1.45001e-06
        [specialize_transform]: 6.53998e-06
        [updatestate_useless_node_eliminater]: 5.01002e-06
        [accelerated_algorithm]: 5.92001e-06
        [meta_shard_fg_expand]: 1.97999e-06
        [get_grad_eliminate_]: 5.61e-06
        [merge_forward]: 3.71001e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.352e-05
        [j_node_and_user_rematch]: 8.97e-06
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 8.99e-06
        [inline_without_move]: 5.44998e-06
        [renormalize]: 0.00045328
        [add_forward_monad_depend]: 5.43002e-06
        [auto_monad_grad]: 1.82999e-06
        [auto_monad_eliminator]: 1.289e-05
        [cse]: 2.236e-05
        [replace_applicator]: 1.092e-05
    [Cycle 2]: 0.00030379, [27]
        [switch_simplify]: 5.07e-06
        [loop_unroll]: 4.79e-06
        [a_1]: 8.745e-05
        [with_stream_mark]: 7.76001e-06
        [recompute_prepare]: 4.60999e-06
        [updatestate_depend_eliminate]: 2.64001e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 4.79e-06
        [updatestate_useless_node_eliminater]: 4.50001e-06
        [accelerated_algorithm]: 4.87998e-06
        [meta_shard_fg_expand]: 9.99979e-07
        [get_grad_eliminate_]: 4.37998e-06
        [merge_forward]: 2.23002e-06
        [cell_reuse_recompute_pass]: 1.49998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.156e-05
        [j_node_and_user_rematch]: 7.29001e-06
        [meta_fg_expand]: 1.41002e-06
        [replace_old_param]: 6.63e-06
        [inline_without_move]: 4.35e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 6.90023e-07
        [auto_monad_eliminator]: 4.82e-06
        [cse]: 9.00001e-06
        [replace_applicator]: 4.88001e-06
[py_interpret_to_execute_after_opt_a]: 8.07998e-06
[rewriter_after_opt_a]: 3.009e-05
[convert_after_rewriter]: 6.42001e-06
[order_py_execute_after_rewriter]: 5.17999e-06
[mutable_eliminate]: 0.00046632
[jit_opt_b]: 4.704e-05, [1]
    [Cycle 1]: 4.084e-05, [2]
        [frontend_op_eliminate]: 1.582e-05
        [inline_after_opt_a]: 1.414e-05
[cconv]: 2.344e-05
[loop_unroll]: 0.00041571
[jit_opt_after_cconv]: 0.00012891, [1]
    [Cycle 1]: 0.00012271, [11]
        [c_1]: 1.917e-05
        [parameter_eliminate]: 2.37999e-06
        [updatestate_depend_eliminate]: 4.95999e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [cse]: 1.509e-05
        [call_graph_tuple_transform]: 1.74e-05
        [tuple_list_get_item_eliminator]: 4.87e-06
        [none_parameter_eliminate]: 8.89995e-07
        [renormalize]: 5.10016e-07
        [switch_simplify]: 5.24e-06
[remove_dup_value]: 1.259e-05
[partial_unused_args_eliminate]: 1.91e-06
[environ_conv]: 4.68999e-06
[add_recomputation]: 4.315e-05
[cse_after_recomputation]: 1.853e-05, [1]
    [Cycle 1]: 1.338e-05, [1]
        [cse]: 8.03001e-06
[auto_monad_reorder]: 1.332e-05
[get_jit_bprop_graph]: 1.43002e-06
[rewriter_after_jit_bprop_graph]: 4.33999e-06
[opt_after_jit_grad]: 0.00043868
[symbol_engine_optimizer]: 6.779e-05, [1]
    [Cycle 1]: 6.2e-05, [6]
        [build]: 2.91999e-06
        [elim_shapecalc]: 7.17002e-06
        [elim_not_effective]: 1.186e-05
        [opt_reshape]: 5.29e-06
        [fold_const_symbol]: 8.08001e-06
        [renormalize]: 5.00004e-07
[validate]: 2.867e-05
[backend_pass]: 7.80012e-07
[task_emit]: 0.00575407
[execute]: 6.64001e-06
Sums
    bootstrap                            :     0.000362s :  2.69%
    type_inference                       :     0.004056s : 30.17%
    event_method                         :     0.000015s :  0.11%
    auto_monad                           :     0.000060s :  0.44%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.19%
    rewriter_before_opt_a                :     0.000068s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.42%
    jit_opt_a.loop_unroll                :     0.000036s :  0.27%
    jit_opt_a.a_1                        :     0.000636s :  4.73%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000453s :  3.37%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000031s :  0.23%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000030s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000466s :  3.47%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000416s :  3.09%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000043s :  0.32%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000013s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000439s :  3.26%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000029s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005754s : 42.80%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000221    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.40% :     0.000005s :      3: substitution.graph_param_transform
85.92% :     0.000190s :      4: substitution.inline
 1.29% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.91% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.57% :     0.000003s :      2: substitution.replace_old_param
 5.46% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004004     2
82.28% :     0.003294s :      1: type_inference.infer
17.72% :     0.000710s :      1: type_inference.specialize
------[replace.]   0.000059     6
66.77% :     0.000039s :      4: replace.inline
33.23% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000198     6
94.50% :     0.000187s :      4: match.inline
 5.50% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.32% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.44% :     0.000002s :     10: predicate.addn_zero_filter
 2.07% :     0.000002s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000001s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.08% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.13% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.97% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.19% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.14% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.57% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000000s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.57% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.77% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.04% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.44% :     0.000002s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.52% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.68% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.32% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.45% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.74% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.80% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.10% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000459    12
49.69% :     0.000228s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.78% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.53% :     0.000205s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018671    76
 0.25% :     0.000046s :      1: add_recomputation
 0.34% :     0.000063s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 2.00% :     0.000373s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.71% :     0.004987s :      1: jit_opt_a
 0.71% :     0.000132s :      1: jit_opt_after_cconv
 0.27% :     0.000050s :      1: jit_opt_b
 2.27% :     0.000424s :      1: loop_unroll
 2.54% :     0.000474s :      1: mutable_eliminate
 4.49% :     0.000838s :     26: opt.transform.jit_opt_a
 0.23% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.39% :     0.000446s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000029s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.24% :     0.000232s :      1: renormalize.infer
 1.15% :     0.000215s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000033s :      1: rewriter_after_opt_a
 0.38% :     0.000070s :      1: rewriter_before_opt_a
 0.38% :     0.000070s :      1: symbol_engine_optimizer
30.87% :     0.005764s :      1: task_emit
21.77% :     0.004066s :      1: type_inference
 0.24% :     0.000045s :      1: validate


TotalTime = 0.0126369, [30]
[bootstrap]: 0.0003387
[type_inference]: 0.00444213
[event_method]: 1.745e-05
[auto_monad]: 6.341e-05
[graph_reusing]: 5.65001e-06
[pre_auto_parallel]: 2.74999e-06
[py_interpret_to_execute]: 3.475e-05
[rewriter_before_opt_a]: 7.187e-05
[expand_dump_flag]: 3.03e-06
[jit_opt_a]: 0.00558881, [2]
    [Cycle 1]: 0.00170741, [27]
        [switch_simplify]: 5.764e-05
        [loop_unroll]: 2.925e-05
        [a_1]: 0.00060192
        [with_stream_mark]: 2.107e-05
        [recompute_prepare]: 8.55001e-06
        [updatestate_depend_eliminate]: 3.35998e-06
        [updatestate_assign_eliminate]: 2.79999e-06
        [updatestate_loads_eliminate]: 2.76e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 5.74999e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 5.32001e-06
        [meta_shard_fg_expand]: 2.02999e-06
        [get_grad_eliminate_]: 5.40999e-06
        [merge_forward]: 3.51999e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.356e-05
        [j_node_and_user_rematch]: 8.72998e-06
        [meta_fg_expand]: 2.37001e-06
        [replace_old_param]: 8.81002e-06
        [inline_without_move]: 5.16002e-06
        [renormalize]: 0.00067625
        [add_forward_monad_depend]: 5.98002e-06
        [auto_monad_grad]: 2.07999e-06
        [auto_monad_eliminator]: 1.243e-05
        [cse]: 3.066e-05
        [replace_applicator]: 1.174e-05
    [Cycle 2]: 0.00031526, [27]
        [switch_simplify]: 5.39e-06
        [loop_unroll]: 5.14998e-06
        [a_1]: 8.934e-05
        [with_stream_mark]: 9.19e-06
        [recompute_prepare]: 5.37001e-06
        [updatestate_depend_eliminate]: 2.32999e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 8.99978e-07
        [specialize_transform]: 5.00999e-06
        [updatestate_useless_node_eliminater]: 4.84003e-06
        [accelerated_algorithm]: 5.07e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 2.42001e-06
        [cell_reuse_recompute_pass]: 1.43002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.188e-05
        [j_node_and_user_rematch]: 7.93999e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 7.09001e-06
        [inline_without_move]: 4.74998e-06
        [renormalize]: 1.10012e-07
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 4.74e-06
        [cse]: 9.81e-06
        [replace_applicator]: 4.94e-06
[py_interpret_to_execute_after_opt_a]: 1.056e-05
[rewriter_after_opt_a]: 3.489e-05
[convert_after_rewriter]: 6.66e-06
[order_py_execute_after_rewriter]: 5.35999e-06
[mutable_eliminate]: 0.00055005
[jit_opt_b]: 4.778e-05, [1]
    [Cycle 1]: 4.094e-05, [2]
        [frontend_op_eliminate]: 1.519e-05
        [inline_after_opt_a]: 1.456e-05
[cconv]: 2.547e-05
[loop_unroll]: 0.00040767
[jit_opt_after_cconv]: 0.00013254, [1]
    [Cycle 1]: 0.00012642, [11]
        [c_1]: 2.005e-05
        [parameter_eliminate]: 2.54001e-06
        [updatestate_depend_eliminate]: 5.72001e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [cse]: 1.628e-05
        [call_graph_tuple_transform]: 1.682e-05
        [tuple_list_get_item_eliminator]: 5.02999e-06
        [none_parameter_eliminate]: 1.02e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.40001e-06
[remove_dup_value]: 1.055e-05
[partial_unused_args_eliminate]: 2.19001e-06
[environ_conv]: 4.87e-06
[add_recomputation]: 4.69e-05
[cse_after_recomputation]: 2.09e-05, [1]
    [Cycle 1]: 1.501e-05, [1]
        [cse]: 9.07999e-06
[auto_monad_reorder]: 1.552e-05
[get_jit_bprop_graph]: 1.52001e-06
[rewriter_after_jit_bprop_graph]: 5.71e-06
[opt_after_jit_grad]: 0.00046095
[symbol_engine_optimizer]: 7.193e-05, [1]
    [Cycle 1]: 6.563e-05, [6]
        [build]: 3.71999e-06
        [elim_shapecalc]: 6.93e-06
        [elim_not_effective]: 1.304e-05
        [opt_reshape]: 6.17001e-06
        [fold_const_symbol]: 8.25999e-06
        [renormalize]: 4.80009e-07
[validate]: 3.676e-05
Sums
    bootstrap                            :     0.000339s :  3.99%
    type_inference                       :     0.004442s : 52.28%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000063s :  0.75%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.41%
    rewriter_before_opt_a                :     0.000072s :  0.85%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000063s :  0.74%
    jit_opt_a.loop_unroll                :     0.000034s :  0.40%
    jit_opt_a.a_1                        :     0.000691s :  8.14%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.36%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000676s :  7.96%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.20%
    jit_opt_a.cse                        :     0.000040s :  0.48%
    jit_opt_a.replace_applicator         :     0.000017s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000035s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000550s :  6.47%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000025s :  0.30%
    loop_unroll                          :     0.000408s :  4.80%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000011s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000047s :  0.55%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000461s :  5.43%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000037s :  0.43%


Time group info:
------[substitution.]   0.000209    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 1.87% :     0.000004s :      3: substitution.graph_param_transform
85.48% :     0.000179s :      4: substitution.inline
 1.40% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.95% :     0.000004s :      2: substitution.replace_old_param
 5.77% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004348     2
82.24% :     0.003576s :      1: type_inference.infer
17.76% :     0.000772s :      1: type_inference.specialize
------[replace.]   0.000067     6
68.18% :     0.000046s :      4: replace.inline
31.82% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000188     6
94.16% :     0.000177s :      4: match.inline
 5.84% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.16% :     0.000001s :     10: predicate.accumulaten_eliminater
 0.96% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 1.94% :     0.000002s :     10: predicate.arithmetic_simplify
 1.13% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
11.90% :     0.000014s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.81% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 2.10% :     0.000003s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.25% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.22% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.19% :     0.000003s :     16: predicate.switch_defer_inline
 1.99% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.06% :     0.000009s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.36% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.35% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.96% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.59% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.41% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000565    12
48.27% :     0.000273s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.39% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.34% :     0.000250s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014261    72
 0.35% :     0.000050s :      1: add_recomputation
 0.47% :     0.000067s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 2.48% :     0.000354s :      1: bootstrap
 0.20% :     0.000028s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000022s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.21% :     0.005592s :      1: jit_opt_a
 0.95% :     0.000135s :      1: jit_opt_after_cconv
 0.35% :     0.000050s :      1: jit_opt_b
 2.91% :     0.000416s :      1: loop_unroll
 3.91% :     0.000558s :      1: mutable_eliminate
 6.32% :     0.000901s :     26: opt.transform.jit_opt_a
 0.31% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000023s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.29% :     0.000469s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000037s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000013s :      1: remove_dup_value
 2.65% :     0.000378s :      1: renormalize.infer
 2.03% :     0.000290s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000038s :      1: rewriter_after_opt_a
 0.52% :     0.000075s :      1: rewriter_before_opt_a
 0.52% :     0.000075s :      1: symbol_engine_optimizer
31.26% :     0.004458s :      1: type_inference


TotalTime = 0.029125, [33]
[bootstrap]: 0.00016916
[type_inference]: 0.00908234
[event_method]: 3.831e-05
[auto_monad]: 8.153e-05
[graph_reusing]: 5.99999e-06
[pre_auto_parallel]: 2.32001e-06
[py_interpret_to_execute]: 2.223e-05
[rewriter_before_opt_a]: 8.374e-05
[expand_dump_flag]: 3.14001e-06
[jit_opt_a]: 0.0107227, [3]
    [Cycle 1]: 0.00535087, [27]
        [switch_simplify]: 5.743e-05
        [loop_unroll]: 3.798e-05
        [a_1]: 0.00077934
        [with_stream_mark]: 2.398e-05
        [recompute_prepare]: 1.754e-05
        [updatestate_depend_eliminate]: 6.86001e-06
        [updatestate_assign_eliminate]: 5.96e-06
        [updatestate_loads_eliminate]: 5.54e-06
        [parameter_eliminate]: 2.46e-06
        [specialize_transform]: 1.27e-05
        [updatestate_useless_node_eliminater]: 1.139e-05
        [accelerated_algorithm]: 1.211e-05
        [meta_shard_fg_expand]: 2.93e-06
        [get_grad_eliminate_]: 1.149e-05
        [merge_forward]: 6.83e-06
        [cell_reuse_recompute_pass]: 9.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.43e-05
        [j_node_and_user_rematch]: 2.114e-05
        [meta_fg_expand]: 0.00142402
        [replace_old_param]: 5.2e-05
        [inline_without_move]: 8.667e-05
        [renormalize]: 0.00227447
        [add_forward_monad_depend]: 1.017e-05
        [auto_monad_grad]: 5.46e-06
        [auto_monad_eliminator]: 4.719e-05
        [cse]: 0.00014014
        [replace_applicator]: 6.822e-05
    [Cycle 2]: 0.00194307, [27]
        [switch_simplify]: 3.468e-05
        [loop_unroll]: 3.294e-05
        [a_1]: 0.00099678
        [with_stream_mark]: 1.301e-05
        [recompute_prepare]: 7.8e-06
        [updatestate_depend_eliminate]: 4.12e-06
        [updatestate_assign_eliminate]: 3.07002e-06
        [updatestate_loads_eliminate]: 2.83003e-06
        [parameter_eliminate]: 1.57999e-06
        [specialize_transform]: 6.28e-06
        [updatestate_useless_node_eliminater]: 5.94e-06
        [accelerated_algorithm]: 5.97001e-06
        [meta_shard_fg_expand]: 2.14999e-06
        [get_grad_eliminate_]: 6.01998e-06
        [merge_forward]: 3.2e-06
        [cell_reuse_recompute_pass]: 9.10019e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.372e-05
        [j_node_and_user_rematch]: 1.016e-05
        [meta_fg_expand]: 3.322e-05
        [replace_old_param]: 1.006e-05
        [inline_without_move]: 5.71003e-06
        [renormalize]: 0.00054022
        [add_forward_monad_depend]: 4.43999e-06
        [auto_monad_grad]: 1.45999e-06
        [auto_monad_eliminator]: 1.258e-05
        [cse]: 2.376e-05
        [replace_applicator]: 1.246e-05
    [Cycle 3]: 0.00035316, [27]
        [switch_simplify]: 6.44001e-06
        [loop_unroll]: 5.86e-06
        [a_1]: 0.0001143
        [with_stream_mark]: 8.95999e-06
        [recompute_prepare]: 5.90002e-06
        [updatestate_depend_eliminate]: 3.56999e-06
        [updatestate_assign_eliminate]: 2.54001e-06
        [updatestate_loads_eliminate]: 2.47001e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 5.97001e-06
        [updatestate_useless_node_eliminater]: 5.47001e-06
        [accelerated_algorithm]: 5.53002e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 5.46e-06
        [merge_forward]: 2.81e-06
        [cell_reuse_recompute_pass]: 1.45001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.409e-05
        [j_node_and_user_rematch]: 9.54999e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 8.03999e-06
        [inline_without_move]: 5.32999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 9.20001e-07
        [auto_monad_grad]: 5.69999e-07
        [auto_monad_eliminator]: 6.58e-06
        [cse]: 1.38e-05
        [replace_applicator]: 5.97999e-06
[py_interpret_to_execute_after_opt_a]: 1.247e-05
[rewriter_after_opt_a]: 3.697e-05
[convert_after_rewriter]: 7.68999e-06
[order_py_execute_after_rewriter]: 5.75001e-06
[mutable_eliminate]: 0.00061416
[jit_opt_b]: 5.248e-05, [1]
    [Cycle 1]: 4.534e-05, [2]
        [frontend_op_eliminate]: 1.745e-05
        [inline_after_opt_a]: 1.705e-05
[cconv]: 2.309e-05
[loop_unroll]: 0.00041583
[jit_opt_after_cconv]: 0.0001585, [1]
    [Cycle 1]: 0.00015234, [11]
        [c_1]: 2.464e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 5.96998e-06
        [updatestate_assign_eliminate]: 3.19001e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [cse]: 2.28e-05
        [call_graph_tuple_transform]: 2.009e-05
        [tuple_list_get_item_eliminator]: 6.51e-06
        [none_parameter_eliminate]: 1.30001e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 6.60002e-06
[remove_dup_value]: 1.566e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 5.15001e-06
[add_recomputation]: 4.99e-05
[cse_after_recomputation]: 2.609e-05, [1]
    [Cycle 1]: 2.002e-05, [1]
        [cse]: 1.394e-05
[auto_monad_reorder]: 1.693e-05
[get_jit_bprop_graph]: 1.76e-06
[rewriter_after_jit_bprop_graph]: 5.59e-06
[opt_after_jit_grad]: 0.00046129
[symbol_engine_optimizer]: 7.587e-05, [1]
    [Cycle 1]: 6.984e-05, [6]
        [build]: 3.97e-06
        [elim_shapecalc]: 8.40001e-06
        [elim_not_effective]: 1.404e-05
        [opt_reshape]: 6.34999e-06
        [fold_const_symbol]: 1.011e-05
        [renormalize]: 4.19997e-07
[validate]: 3.42e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00669958
[execute]: 7.93001e-06
Sums
    bootstrap                            :     0.000169s :  0.67%
    type_inference                       :     0.009082s : 35.92%
    event_method                         :     0.000038s :  0.15%
    auto_monad                           :     0.000082s :  0.32%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000022s :  0.09%
    rewriter_before_opt_a                :     0.000084s :  0.33%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000099s :  0.39%
    jit_opt_a.loop_unroll                :     0.000077s :  0.30%
    jit_opt_a.a_1                        :     0.001890s :  7.48%
    jit_opt_a.with_stream_mark           :     0.000046s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000031s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000025s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.09%
    jit_opt_a.merge_forward              :     0.000013s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000052s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000041s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001459s :  5.77%
    jit_opt_a.replace_old_param          :     0.000070s :  0.28%
    jit_opt_a.inline_without_move        :     0.000098s :  0.39%
    jit_opt_a.renormalize                :     0.002815s : 11.13%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000066s :  0.26%
    jit_opt_a.cse                        :     0.000178s :  0.70%
    jit_opt_a.replace_applicator         :     0.000087s :  0.34%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.05%
    rewriter_after_opt_a                 :     0.000037s :  0.15%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000614s :  2.43%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.07%
    cconv                                :     0.000023s :  0.09%
    loop_unroll                          :     0.000416s :  1.64%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000050s :  0.20%
    cse_after_recomputation.cse          :     0.000014s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000461s :  1.82%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006700s : 26.50%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000475   110
 0.44% :     0.000002s :      3: substitution.elim_not_effective
 0.35% :     0.000002s :      3: substitution.fold_const_symbol
 1.10% :     0.000005s :      4: substitution.graph_param_transform
57.29% :     0.000272s :     12: substitution.inline
10.76% :     0.000051s :      2: substitution.inline_without_move
 1.70% :     0.000008s :     14: substitution.j_node_and_user_rematch
 2.30% :     0.000011s :      7: substitution.minmaximum_grad
 1.37% :     0.000006s :      5: substitution.partial_eliminate
 2.23% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.68% :     0.000027s :     10: substitution.replace_applicator
 1.51% :     0.000007s :      7: substitution.replace_old_param
 0.51% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.06% :     0.000019s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.93% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.77% :     0.000037s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009028     2
90.05% :     0.008130s :      1: type_inference.infer
 9.95% :     0.000898s :      1: type_inference.specialize
------[replace.]   0.000135    19
64.58% :     0.000087s :     12: replace.inline
35.42% :     0.000048s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000284    19
93.63% :     0.000266s :     12: match.inline
 6.37% :     0.000018s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000355  2568
 1.63% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.41% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.45% :     0.000005s :     43: predicate.addn_check_dump
 1.88% :     0.000007s :     43: predicate.addn_zero_filter
 1.94% :     0.000007s :     43: predicate.arithmetic_simplify
 1.66% :     0.000006s :     43: predicate.cast_eliminate
 0.16% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.46% :     0.000005s :     43: predicate.compare_switch_simplify
 1.66% :     0.000006s :     43: predicate.depend_value_elim
 1.48% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.49% :     0.000005s :     43: predicate.dict_get_item_eliminator
 1.62% :     0.000006s :     43: predicate.dict_set_item_eliminator
 0.29% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.14% :     0.000000s :      4: predicate.elim_not_effective
 0.20% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.51% :     0.000005s :     43: predicate.environ_add_const_eliminate
 1.48% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.47% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.74% :     0.000006s :     43: predicate.environ_get_eliminate
 1.45% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.82% :     0.000003s :     19: predicate.get_grad_eliminate
 0.15% :     0.000001s :      4: predicate.graph_param_transform
 4.31% :     0.000015s :     70: predicate.inline
 1.81% :     0.000006s :     40: predicate.inline_without_move
 0.36% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.92% :     0.000003s :     19: predicate.less_batch_normalization
 1.85% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.91% :     0.000007s :     54: predicate.load_eliminater
 0.52% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.19% :     0.000011s :     85: predicate.loop_unroll_before_grad
 1.79% :     0.000006s :     47: predicate.make_slice_get_slice_eliminator
 1.49% :     0.000005s :     43: predicate.merge_addn
 1.51% :     0.000005s :     43: predicate.minmaximum_grad
 0.48% :     0.000002s :      4: predicate.mutable_eliminate
 0.19% :     0.000001s :      4: predicate.opt_reshape
 2.30% :     0.000008s :     54: predicate.partial_eliminate
 1.72% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.79% :     0.000006s :     43: predicate.reduce_eliminate
 1.76% :     0.000006s :     50: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.54% :     0.000009s :     96: predicate.replace_applicator
 0.91% :     0.000003s :     40: predicate.replace_old_param
 0.13% :     0.000000s :      4: predicate.reset_defer_inline
 1.91% :     0.000007s :     43: predicate.reshape_eliminate
 1.60% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.28% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.65% :     0.000006s :     43: predicate.same_eliminate
 0.46% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.37% :     0.000001s :      8: predicate.special_op_eliminate
 0.85% :     0.000003s :     19: predicate.specialize_transform
 1.59% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000005s :     43: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.69% :     0.000010s :     62: predicate.switch_defer_inline
 2.58% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.14% :     0.000022s :    151: predicate.switch_simplify
 1.51% :     0.000005s :     43: predicate.tile_eliminate
 1.52% :     0.000005s :     43: predicate.transpose_eliminate
 1.79% :     0.000006s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.66% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.00% :     0.000011s :     58: predicate.tuple_list_get_item_eliminator
 1.84% :     0.000007s :     43: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000006s :     50: predicate.tuple_to_list_eliminator_
 1.97% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000010s :     73: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000006s :     43: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.23% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001338    29
58.10% :     0.000777s :     13: func_graph_cloner_run.FuncGraphClonerGraph
41.90% :     0.000560s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.034561    91
 0.15% :     0.000053s :      1: add_recomputation
 0.25% :     0.000086s :      1: auto_monad
 0.06% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.53% :     0.000182s :      1: bootstrap
 0.07% :     0.000026s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000029s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.12% :     0.000042s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
31.03% :     0.010726s :      1: jit_opt_a
 0.47% :     0.000161s :      1: jit_opt_after_cconv
 0.16% :     0.000055s :      1: jit_opt_b
 1.22% :     0.000423s :      1: loop_unroll
 1.80% :     0.000623s :      1: mutable_eliminate
 7.23% :     0.002498s :     39: opt.transform.jit_opt_a
 0.16% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000028s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.36% :     0.000469s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000025s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 4.14% :     0.001430s :      2: renormalize.infer
 3.97% :     0.001372s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000040s :      1: rewriter_after_opt_a
 0.25% :     0.000087s :      1: rewriter_before_opt_a
 0.23% :     0.000078s :      1: symbol_engine_optimizer
19.42% :     0.006713s :      1: task_emit
26.31% :     0.009094s :      1: type_inference
 0.16% :     0.000055s :      1: validate


TotalTime = 0.0193704, [33]
[bootstrap]: 0.00037737
[type_inference]: 0.00504961
[event_method]: 1.754e-05
[auto_monad]: 6.37e-05
[graph_reusing]: 5.10001e-06
[pre_auto_parallel]: 2.63e-06
[py_interpret_to_execute]: 3.435e-05
[rewriter_before_opt_a]: 7.171e-05
[expand_dump_flag]: 3.53e-06
[jit_opt_a]: 0.00565464, [2]
    [Cycle 1]: 0.00174597, [27]
        [switch_simplify]: 5.499e-05
        [loop_unroll]: 0.00015256
        [a_1]: 0.00053705
        [with_stream_mark]: 1.413e-05
        [recompute_prepare]: 7.63001e-06
        [updatestate_depend_eliminate]: 2.94999e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 5.06002e-06
        [accelerated_algorithm]: 5.91e-06
        [meta_shard_fg_expand]: 1.26002e-06
        [get_grad_eliminate_]: 5.28002e-06
        [merge_forward]: 2.34999e-06
        [cell_reuse_recompute_pass]: 1.02998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.189e-05
        [j_node_and_user_rematch]: 8.48001e-06
        [meta_fg_expand]: 2.14999e-06
        [replace_old_param]: 8.12998e-06
        [inline_without_move]: 5.42001e-06
        [renormalize]: 0.00069227
        [add_forward_monad_depend]: 4.44002e-06
        [auto_monad_grad]: 1.49e-06
        [auto_monad_eliminator]: 1.018e-05
        [cse]: 1.648e-05
        [replace_applicator]: 1.179e-05
    [Cycle 2]: 0.00031222, [27]
        [switch_simplify]: 5.52999e-06
        [loop_unroll]: 4.80001e-06
        [a_1]: 8.944e-05
        [with_stream_mark]: 8.14002e-06
        [recompute_prepare]: 4.95001e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 4.72e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 4.87998e-06
        [meta_shard_fg_expand]: 9.70002e-07
        [get_grad_eliminate_]: 4.37998e-06
        [merge_forward]: 2.24001e-06
        [cell_reuse_recompute_pass]: 1.52999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.194e-05
        [j_node_and_user_rematch]: 7.77e-06
        [meta_fg_expand]: 1.70001e-06
        [replace_old_param]: 7.09001e-06
        [inline_without_move]: 4.52e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.08001e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 4.50999e-06
        [cse]: 1.009e-05
        [replace_applicator]: 4.82998e-06
[py_interpret_to_execute_after_opt_a]: 8.13001e-06
[rewriter_after_opt_a]: 3.216e-05
[convert_after_rewriter]: 5.46998e-06
[order_py_execute_after_rewriter]: 3.77998e-06
[mutable_eliminate]: 0.00050542
[jit_opt_b]: 5.125e-05, [1]
    [Cycle 1]: 4.492e-05, [2]
        [frontend_op_eliminate]: 1.91e-05
        [inline_after_opt_a]: 1.482e-05
[cconv]: 1.467e-05
[loop_unroll]: 0.00040629
[jit_opt_after_cconv]: 0.00012789, [1]
    [Cycle 1]: 0.00012178, [11]
        [c_1]: 1.99e-05
        [parameter_eliminate]: 1.48002e-06
        [updatestate_depend_eliminate]: 4.69002e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [cse]: 1.586e-05
        [call_graph_tuple_transform]: 1.64e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.33002e-06
        [renormalize]: 3.39991e-07
        [switch_simplify]: 5.29998e-06
[remove_dup_value]: 1.351e-05
[partial_unused_args_eliminate]: 1.97001e-06
[environ_conv]: 7.85e-06
[add_recomputation]: 4.32e-05
[cse_after_recomputation]: 2.208e-05, [1]
    [Cycle 1]: 1.648e-05, [1]
        [cse]: 9.73998e-06
[auto_monad_reorder]: 1.429e-05
[get_jit_bprop_graph]: 1.98002e-06
[rewriter_after_jit_bprop_graph]: 6.54001e-06
[opt_after_jit_grad]: 0.00046106
[symbol_engine_optimizer]: 7.006e-05, [1]
    [Cycle 1]: 6.339e-05, [6]
        [build]: 2.51998e-06
        [elim_shapecalc]: 7.50998e-06
        [elim_not_effective]: 1.194e-05
        [opt_reshape]: 5.67001e-06
        [fold_const_symbol]: 8.2e-06
        [renormalize]: 6.10016e-07
[validate]: 4.157e-05
[backend_pass]: 1.04998e-06
[task_emit]: 0.00604623
[execute]: 6.86999e-06
Sums
    bootstrap                            :     0.000377s :  2.49%
    type_inference                       :     0.005050s : 33.28%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000064s :  0.42%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.23%
    rewriter_before_opt_a                :     0.000072s :  0.47%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.40%
    jit_opt_a.loop_unroll                :     0.000157s :  1.04%
    jit_opt_a.a_1                        :     0.000626s :  4.13%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000692s :  4.56%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.10%
    jit_opt_a.cse                        :     0.000027s :  0.18%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000032s :  0.21%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000505s :  3.33%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000015s :  0.10%
    loop_unroll                          :     0.000406s :  2.68%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000008s :  0.05%
    add_recomputation                    :     0.000043s :  0.28%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000461s :  3.04%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006046s : 39.85%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000192    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.62% :     0.000005s :      3: substitution.graph_param_transform
85.90% :     0.000165s :      4: substitution.inline
 1.40% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.38% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.70% :     0.000003s :      2: substitution.replace_old_param
 4.46% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004952     2
81.61% :     0.004041s :      1: type_inference.infer
18.39% :     0.000911s :      1: type_inference.specialize
------[replace.]   0.000073     6
65.07% :     0.000047s :      4: replace.inline
34.93% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000170     6
95.61% :     0.000163s :      4: match.inline
 4.39% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000232   691
 0.74% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.68% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.53% :     0.000001s :     10: predicate.addn_check_dump
 0.82% :     0.000002s :     10: predicate.addn_zero_filter
 1.35% :     0.000003s :     10: predicate.arithmetic_simplify
 0.72% :     0.000002s :     10: predicate.cast_eliminate
 0.27% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.54% :     0.000001s :     10: predicate.compare_switch_simplify
 0.61% :     0.000001s :     10: predicate.depend_value_elim
 0.53% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 0.65% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.58% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.38% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.15% :     0.000000s :      3: predicate.elim_not_effective
 0.37% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.60% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.54% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.56% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.57% :     0.000001s :     10: predicate.environ_get_eliminate
 0.61% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.11% :     0.000000s :      3: predicate.fold_const_symbol
 0.50% :     0.000001s :      6: predicate.get_grad_eliminate
 0.13% :     0.000000s :      3: predicate.graph_param_transform
 2.51% :     0.000006s :     22: predicate.inline
 0.46% :     0.000001s :      6: predicate.inline_without_move
 0.18% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.74% :     0.000002s :      6: predicate.less_batch_normalization
 0.74% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.06% :     0.000002s :     15: predicate.load_eliminater
 0.55% :     0.000001s :      3: predicate.loop_unroll_after_grad
52.97% :     0.000123s :     34: predicate.loop_unroll_before_grad
 0.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.55% :     0.000001s :     10: predicate.merge_addn
 0.54% :     0.000001s :     10: predicate.minmaximum_grad
 0.63% :     0.000001s :      3: predicate.mutable_eliminate
 0.21% :     0.000000s :      3: predicate.opt_reshape
 1.05% :     0.000002s :     15: predicate.partial_eliminate
 0.88% :     0.000002s :     10: predicate.print_const_string_wrapper
 0.80% :     0.000002s :     10: predicate.reduce_eliminate
 0.72% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.24% :     0.000001s :      6: predicate.remove_not_recompute_node
 0.84% :     0.000002s :     18: predicate.replace_applicator
 0.25% :     0.000001s :      6: predicate.replace_old_param
 0.14% :     0.000000s :      3: predicate.reset_defer_inline
 0.77% :     0.000002s :     10: predicate.reshape_eliminate
 0.61% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.48% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.63% :     0.000001s :     10: predicate.same_eliminate
 0.25% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.55% :     0.000001s :      6: predicate.special_op_eliminate
 0.50% :     0.000001s :      6: predicate.specialize_transform
 0.63% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 0.54% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.18% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 1.26% :     0.000003s :     16: predicate.switch_defer_inline
 1.02% :     0.000002s :     16: predicate.switch_layer_defer_inline
 3.75% :     0.000009s :     53: predicate.switch_simplify
 0.71% :     0.000002s :     10: predicate.tile_eliminate
 0.56% :     0.000001s :     10: predicate.transpose_eliminate
 0.77% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.55% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 1.69% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 0.90% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 0.68% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 0.80% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 1.45% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 0.85% :     0.000002s :     10: predicate.value_based_eliminate
 0.15% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.27% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000620    12
48.14% :     0.000299s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.92% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.93% :     0.000291s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021110    76
 0.22% :     0.000047s :      1: add_recomputation
 0.32% :     0.000067s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.89% :     0.000398s :      1: bootstrap
 0.08% :     0.000017s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000011s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
26.80% :     0.005658s :      1: jit_opt_a
 0.62% :     0.000130s :      1: jit_opt_after_cconv
 0.25% :     0.000054s :      1: jit_opt_b
 1.95% :     0.000412s :      1: loop_unroll
 2.43% :     0.000513s :      1: mutable_eliminate
 4.53% :     0.000956s :     26: opt.transform.jit_opt_a
 0.21% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000027s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000010s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.22% :     0.000469s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000037s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.71% :     0.000362s :      1: renormalize.infer
 1.53% :     0.000324s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000035s :      1: rewriter_after_opt_a
 0.35% :     0.000074s :      1: rewriter_before_opt_a
 0.34% :     0.000072s :      1: symbol_engine_optimizer
28.69% :     0.006057s :      1: task_emit
23.99% :     0.005064s :      1: type_inference
 0.28% :     0.000060s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.577.544 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc41160> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.578.138 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0183853, [33]
[bootstrap]: 0.00027929
[type_inference]: 0.00430445
[event_method]: 1.738e-05
[auto_monad]: 5.968e-05
[graph_reusing]: 5.17e-06
[pre_auto_parallel]: 2.43e-06
[py_interpret_to_execute]: 2.878e-05
[rewriter_before_opt_a]: 7.139e-05
[expand_dump_flag]: 3.14999e-06
[jit_opt_a]: 0.00535213, [2]
    [Cycle 1]: 0.00139078, [27]
        [switch_simplify]: 4.818e-05
        [loop_unroll]: 2.798e-05
        [a_1]: 0.00048713
        [with_stream_mark]: 1.636e-05
        [recompute_prepare]: 7.11001e-06
        [updatestate_depend_eliminate]: 3.37002e-06
        [updatestate_assign_eliminate]: 2.84001e-06
        [updatestate_loads_eliminate]: 2.39001e-06
        [parameter_eliminate]: 1.64e-06
        [specialize_transform]: 5.71998e-06
        [updatestate_useless_node_eliminater]: 4.77998e-06
        [accelerated_algorithm]: 5.56e-06
        [meta_shard_fg_expand]: 2.19999e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 3.41999e-06
        [cell_reuse_recompute_pass]: 1.08001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.359e-05
        [j_node_and_user_rematch]: 8.41002e-06
        [meta_fg_expand]: 2.17999e-06
        [replace_old_param]: 8.40001e-06
        [inline_without_move]: 5.19998e-06
        [renormalize]: 0.00051767
        [add_forward_monad_depend]: 5.00999e-06
        [auto_monad_grad]: 2.17001e-06
        [auto_monad_eliminator]: 1.208e-05
        [cse]: 1.839e-05
        [replace_applicator]: 1.165e-05
    [Cycle 2]: 0.00032851, [27]
        [switch_simplify]: 5.52001e-06
        [loop_unroll]: 5.07999e-06
        [a_1]: 9.429e-05
        [with_stream_mark]: 8.82e-06
        [recompute_prepare]: 5.66e-06
        [updatestate_depend_eliminate]: 2.57001e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 5.22999e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 4.84003e-06
        [meta_shard_fg_expand]: 1.17999e-06
        [get_grad_eliminate_]: 4.65999e-06
        [merge_forward]: 2.37001e-06
        [cell_reuse_recompute_pass]: 1.67001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.258e-05
        [j_node_and_user_rematch]: 7.85e-06
        [meta_fg_expand]: 1.42e-06
        [replace_old_param]: 7.68001e-06
        [inline_without_move]: 4.82998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.42e-06
        [auto_monad_grad]: 9.5999e-07
        [auto_monad_eliminator]: 5.50001e-06
        [cse]: 9.87999e-06
        [replace_applicator]: 9.82999e-06
[py_interpret_to_execute_after_opt_a]: 1.012e-05
[rewriter_after_opt_a]: 3.37e-05
[convert_after_rewriter]: 7.18e-06
[order_py_execute_after_rewriter]: 5.12e-06
[mutable_eliminate]: 0.00051258
[jit_opt_b]: 5.015e-05, [1]
    [Cycle 1]: 4.295e-05, [2]
        [frontend_op_eliminate]: 1.596e-05
        [inline_after_opt_a]: 1.507e-05
[cconv]: 2.512e-05
[loop_unroll]: 0.0004488
[jit_opt_after_cconv]: 0.00014367, [1]
    [Cycle 1]: 0.00013628, [11]
        [c_1]: 2.092e-05
        [parameter_eliminate]: 3.29001e-06
        [updatestate_depend_eliminate]: 5.60001e-06
        [updatestate_assign_eliminate]: 2.51998e-06
        [updatestate_loads_eliminate]: 2.33002e-06
        [cse]: 1.838e-05
        [call_graph_tuple_transform]: 1.869e-05
        [tuple_list_get_item_eliminator]: 5.24998e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 8.59989e-07
        [switch_simplify]: 5.22e-06
[remove_dup_value]: 1.261e-05
[partial_unused_args_eliminate]: 2.53003e-06
[environ_conv]: 5.81998e-06
[add_recomputation]: 5.005e-05
[cse_after_recomputation]: 2.148e-05, [1]
    [Cycle 1]: 1.543e-05, [1]
        [cse]: 9.04e-06
[auto_monad_reorder]: 1.441e-05
[get_jit_bprop_graph]: 1.55001e-06
[rewriter_after_jit_bprop_graph]: 6.46e-06
[opt_after_jit_grad]: 0.00048225
[symbol_engine_optimizer]: 7.263e-05, [1]
    [Cycle 1]: 6.661e-05, [6]
        [build]: 3.23e-06
        [elim_shapecalc]: 7.53999e-06
        [elim_not_effective]: 1.299e-05
        [opt_reshape]: 5.62999e-06
        [fold_const_symbol]: 8.23001e-06
        [renormalize]: 5.8001e-07
[validate]: 3.194e-05
[backend_pass]: 1.19e-06
[task_emit]: 0.00613094
[execute]: 7.73999e-06
Sums
    bootstrap                            :     0.000279s :  1.97%
    type_inference                       :     0.004304s : 30.38%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000060s :  0.42%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.20%
    rewriter_before_opt_a                :     0.000071s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.38%
    jit_opt_a.loop_unroll                :     0.000033s :  0.23%
    jit_opt_a.a_1                        :     0.000581s :  4.10%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000518s :  3.65%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000028s :  0.20%
    jit_opt_a.replace_applicator         :     0.000021s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000513s :  3.62%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000025s :  0.18%
    loop_unroll                          :     0.000449s :  3.17%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000050s :  0.35%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.05%
    opt_after_jit_grad                   :     0.000482s :  3.40%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006131s : 43.27%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000179    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 3.00% :     0.000005s :      3: substitution.graph_param_transform
82.60% :     0.000148s :      4: substitution.inline
 1.51% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.68% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.35% :     0.000004s :      2: substitution.replace_old_param
 6.10% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004243     2
83.18% :     0.003529s :      1: type_inference.infer
16.82% :     0.000714s :      1: type_inference.specialize
------[replace.]   0.000057     6
66.37% :     0.000038s :      4: replace.inline
33.63% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000155     6
93.66% :     0.000146s :      4: match.inline
 6.34% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.89% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.54% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000001s :     10: predicate.addn_zero_filter
 2.09% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000001s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.52% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.33% :     0.000000s :      3: predicate.graph_param_transform
 5.65% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000001s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.23% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.82% :     0.000002s :      3: predicate.mutable_eliminate
 0.59% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.52% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.90% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.25% :     0.000001s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.49% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 1.13% :     0.000001s :      6: predicate.specialize_transform
 1.54% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.47% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.25% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.17% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.73% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000504    12
45.86% :     0.000231s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.70% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.43% :     0.000239s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019801    76
 0.27% :     0.000053s :      1: add_recomputation
 0.32% :     0.000063s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.46% :     0.000290s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
27.04% :     0.005355s :      1: jit_opt_a
 0.74% :     0.000146s :      1: jit_opt_after_cconv
 0.27% :     0.000053s :      1: jit_opt_b
 2.31% :     0.000458s :      1: loop_unroll
 2.63% :     0.000521s :      1: mutable_eliminate
 3.96% :     0.000784s :     26: opt.transform.jit_opt_a
 0.24% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.48% :     0.000491s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000031s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.32% :     0.000262s :      1: renormalize.infer
 1.26% :     0.000249s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000037s :      1: rewriter_after_opt_a
 0.38% :     0.000074s :      1: rewriter_before_opt_a
 0.38% :     0.000075s :      1: symbol_engine_optimizer
31.04% :     0.006145s :      1: task_emit
21.81% :     0.004318s :      1: type_inference
 0.26% :     0.000052s :      1: validate

.
TotalTime = 0.0165191, [33]
[bootstrap]: 0.00026519
[type_inference]: 0.00393285
[event_method]: 1.539e-05
[auto_monad]: 4.272e-05
[graph_reusing]: 3.47002e-06
[pre_auto_parallel]: 1.72999e-06
[py_interpret_to_execute]: 2.442e-05
[rewriter_before_opt_a]: 6.941e-05
[expand_dump_flag]: 2.45002e-06
[jit_opt_a]: 0.00452071, [2]
    [Cycle 1]: 0.00121801, [27]
        [switch_simplify]: 4.215e-05
        [loop_unroll]: 2.79e-05
        [a_1]: 0.00044935
        [with_stream_mark]: 1.193e-05
        [recompute_prepare]: 6.51999e-06
        [updatestate_depend_eliminate]: 2.73e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.11002e-06
        [specialize_transform]: 5.67999e-06
        [updatestate_useless_node_eliminater]: 4.65999e-06
        [accelerated_algorithm]: 5.21002e-06
        [meta_shard_fg_expand]: 1.31002e-06
        [get_grad_eliminate_]: 4.75999e-06
        [merge_forward]: 2.42001e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.036e-05
        [j_node_and_user_rematch]: 8.09997e-06
        [meta_fg_expand]: 1.72001e-06
        [replace_old_param]: 7.24001e-06
        [inline_without_move]: 5.17e-06
        [renormalize]: 0.00041922
        [add_forward_monad_depend]: 4.52998e-06
        [auto_monad_grad]: 1.34998e-06
        [auto_monad_eliminator]: 9.97001e-06
        [cse]: 1.496e-05
        [replace_applicator]: 1.083e-05
    [Cycle 2]: 0.00030801, [27]
        [switch_simplify]: 5.37001e-06
        [loop_unroll]: 4.73001e-06
        [a_1]: 8.631e-05
        [with_stream_mark]: 8.02e-06
        [recompute_prepare]: 4.85001e-06
        [updatestate_depend_eliminate]: 2.53e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.96003e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 4.72e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 4.92e-06
        [meta_shard_fg_expand]: 1.05001e-06
        [get_grad_eliminate_]: 4.37998e-06
        [merge_forward]: 2.54001e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.202e-05
        [j_node_and_user_rematch]: 7.16999e-06
        [meta_fg_expand]: 1.47999e-06
        [replace_old_param]: 7.3e-06
        [inline_without_move]: 4.58001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 1.05999e-06
        [auto_monad_eliminator]: 5.39e-06
        [cse]: 9.98002e-06
        [replace_applicator]: 6.56e-06
[py_interpret_to_execute_after_opt_a]: 7.86001e-06
[rewriter_after_opt_a]: 2.515e-05
[convert_after_rewriter]: 5.56002e-06
[order_py_execute_after_rewriter]: 3.87002e-06
[mutable_eliminate]: 0.00046247
[jit_opt_b]: 4.91e-05, [1]
    [Cycle 1]: 4.321e-05, [2]
        [frontend_op_eliminate]: 1.716e-05
        [inline_after_opt_a]: 1.432e-05
[cconv]: 1.624e-05
[loop_unroll]: 0.0004027
[jit_opt_after_cconv]: 0.00012667, [1]
    [Cycle 1]: 0.00012076, [11]
        [c_1]: 1.903e-05
        [parameter_eliminate]: 2.45002e-06
        [updatestate_depend_eliminate]: 4.98001e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [cse]: 1.516e-05
        [call_graph_tuple_transform]: 1.564e-05
        [tuple_list_get_item_eliminator]: 5.11997e-06
        [none_parameter_eliminate]: 9.09989e-07
        [renormalize]: 4.69998e-07
        [switch_simplify]: 4.99e-06
[remove_dup_value]: 8.05999e-06
[partial_unused_args_eliminate]: 2.14999e-06
[environ_conv]: 4.74e-06
[add_recomputation]: 4.14e-05
[cse_after_recomputation]: 1.943e-05, [1]
    [Cycle 1]: 1.414e-05, [1]
        [cse]: 8.58001e-06
[auto_monad_reorder]: 1.477e-05
[get_jit_bprop_graph]: 1.50999e-06
[rewriter_after_jit_bprop_graph]: 4.67e-06
[opt_after_jit_grad]: 0.00043723
[symbol_engine_optimizer]: 6.791e-05, [1]
    [Cycle 1]: 6.182e-05, [6]
        [build]: 3.36999e-06
        [elim_shapecalc]: 6.98e-06
        [elim_not_effective]: 1.143e-05
        [opt_reshape]: 5.24e-06
        [fold_const_symbol]: 7.85e-06
        [renormalize]: 4.49974e-07
[validate]: 3.811e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.00573037
[execute]: 6.84001e-06
Sums
    bootstrap                            :     0.000265s :  2.04%
    type_inference                       :     0.003933s : 30.29%
    event_method                         :     0.000015s :  0.12%
    auto_monad                           :     0.000043s :  0.33%
    graph_reusing                        :     0.000003s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000024s :  0.19%
    rewriter_before_opt_a                :     0.000069s :  0.53%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000048s :  0.37%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000536s :  4.13%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000015s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000419s :  3.23%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.12%
    jit_opt_a.cse                        :     0.000025s :  0.19%
    jit_opt_a.replace_applicator         :     0.000017s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000025s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000462s :  3.56%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000016s :  0.13%
    loop_unroll                          :     0.000403s :  3.10%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000008s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000041s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000437s :  3.37%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.29%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005730s : 44.14%
    execute                              :     0.000007s :  0.05%


TotalTime = 0.0288003, [33]
[bootstrap]: 0.00021781
[type_inference]: 0.0101518
[event_method]: 3.976e-05
[auto_monad]: 9.199e-05
[graph_reusing]: 6.74999e-06
[pre_auto_parallel]: 2.88e-06
[py_interpret_to_execute]: 2.635e-05
[rewriter_before_opt_a]: 9.034e-05
[expand_dump_flag]: 3.11001e-06
[jit_opt_a]: 0.010335, [3]
    [Cycle 1]: 0.00503719, [27]
        [switch_simplify]: 5.768e-05
        [loop_unroll]: 3.693e-05
        [a_1]: 0.00080872
        [with_stream_mark]: 2.228e-05
        [recompute_prepare]: 1.773e-05
        [updatestate_depend_eliminate]: 7.22997e-06
        [updatestate_assign_eliminate]: 6.12001e-06
        [updatestate_loads_eliminate]: 6.65002e-06
        [parameter_eliminate]: 3.13e-06
        [specialize_transform]: 1.334e-05
        [updatestate_useless_node_eliminater]: 1.214e-05
        [accelerated_algorithm]: 1.232e-05
        [meta_shard_fg_expand]: 3.16001e-06
        [get_grad_eliminate_]: 1.275e-05
        [merge_forward]: 8.08999e-06
        [cell_reuse_recompute_pass]: 9.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.594e-05
        [j_node_and_user_rematch]: 2.309e-05
        [meta_fg_expand]: 0.00138264
        [replace_old_param]: 4.996e-05
        [inline_without_move]: 4.939e-05
        [renormalize]: 0.00202103
        [add_forward_monad_depend]: 9.97001e-06
        [auto_monad_grad]: 4.88001e-06
        [auto_monad_eliminator]: 4.619e-05
        [cse]: 0.00013982
        [replace_applicator]: 6.052e-05
    [Cycle 2]: 0.00183086, [27]
        [switch_simplify]: 3.45e-05
        [loop_unroll]: 3.264e-05
        [a_1]: 0.00103025
        [with_stream_mark]: 1.164e-05
        [recompute_prepare]: 7.33e-06
        [updatestate_depend_eliminate]: 3.48999e-06
        [updatestate_assign_eliminate]: 3.04999e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [parameter_eliminate]: 1.42e-06
        [specialize_transform]: 6.39001e-06
        [updatestate_useless_node_eliminater]: 5.96998e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 1.52001e-06
        [get_grad_eliminate_]: 5.94e-06
        [merge_forward]: 2.91999e-06
        [cell_reuse_recompute_pass]: 7.60017e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.243e-05
        [j_node_and_user_rematch]: 9.24e-06
        [meta_fg_expand]: 3.027e-05
        [replace_old_param]: 9.28002e-06
        [inline_without_move]: 5.82999e-06
        [renormalize]: 0.00041842
        [add_forward_monad_depend]: 3.73001e-06
        [auto_monad_grad]: 1.17999e-06
        [auto_monad_eliminator]: 1.058e-05
        [cse]: 1.827e-05
        [replace_applicator]: 1.216e-05
    [Cycle 3]: 0.00035125, [27]
        [switch_simplify]: 6.29999e-06
        [loop_unroll]: 5.52999e-06
        [a_1]: 0.00011494
        [with_stream_mark]: 8.57998e-06
        [recompute_prepare]: 5.72001e-06
        [updatestate_depend_eliminate]: 3.24001e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 9.90025e-07
        [specialize_transform]: 6.06998e-06
        [updatestate_useless_node_eliminater]: 5.56e-06
        [accelerated_algorithm]: 5.70001e-06
        [meta_shard_fg_expand]: 1.19e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 2.84999e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.334e-05
        [j_node_and_user_rematch]: 9.44998e-06
        [meta_fg_expand]: 1.86e-06
        [replace_old_param]: 8.02e-06
        [inline_without_move]: 5.59e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.17999e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 6.29999e-06
        [cse]: 1.295e-05
        [replace_applicator]: 5.67001e-06
[py_interpret_to_execute_after_opt_a]: 9.43002e-06
[rewriter_after_opt_a]: 3.525e-05
[convert_after_rewriter]: 7.75e-06
[order_py_execute_after_rewriter]: 5.87999e-06
[mutable_eliminate]: 0.0004738
[jit_opt_b]: 5.389e-05, [1]
    [Cycle 1]: 4.757e-05, [2]
        [frontend_op_eliminate]: 1.882e-05
        [inline_after_opt_a]: 1.813e-05
[cconv]: 2.075e-05
[loop_unroll]: 0.00040654
[jit_opt_after_cconv]: 0.00015317, [1]
    [Cycle 1]: 0.00014637, [11]
        [c_1]: 2.421e-05
        [parameter_eliminate]: 2.40002e-06
        [updatestate_depend_eliminate]: 5.76003e-06
        [updatestate_assign_eliminate]: 2.84999e-06
        [updatestate_loads_eliminate]: 2.62001e-06
        [cse]: 2.081e-05
        [call_graph_tuple_transform]: 1.864e-05
        [tuple_list_get_item_eliminator]: 6.13002e-06
        [none_parameter_eliminate]: 1.69998e-06
        [renormalize]: 3.50003e-07
        [switch_simplify]: 1.232e-05
[remove_dup_value]: 1.513e-05
[partial_unused_args_eliminate]: 1.52001e-06
[environ_conv]: 3.63999e-06
[add_recomputation]: 4.499e-05
[cse_after_recomputation]: 2.527e-05, [1]
    [Cycle 1]: 1.997e-05, [1]
        [cse]: 1.41e-05
[auto_monad_reorder]: 1.2e-05
[get_jit_bprop_graph]: 1.37999e-06
[rewriter_after_jit_bprop_graph]: 4.05e-06
[opt_after_jit_grad]: 0.00044659
[symbol_engine_optimizer]: 7.41e-05, [1]
    [Cycle 1]: 6.816e-05, [6]
        [build]: 2.78e-06
        [elim_shapecalc]: 8.30999e-06
        [elim_not_effective]: 1.327e-05
        [opt_reshape]: 6.47001e-06
        [fold_const_symbol]: 9.59e-06
        [renormalize]: 6.80011e-07
[validate]: 2.834e-05
[backend_pass]: 7.29982e-07
[task_emit]: 0.00582205
[execute]: 4.59998e-06
Sums
    bootstrap                            :     0.000218s :  0.87%
    type_inference                       :     0.010152s : 40.71%
    event_method                         :     0.000040s :  0.16%
    auto_monad                           :     0.000092s :  0.37%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.11%
    rewriter_before_opt_a                :     0.000090s :  0.36%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000098s :  0.39%
    jit_opt_a.loop_unroll                :     0.000075s :  0.30%
    jit_opt_a.a_1                        :     0.001954s :  7.83%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000031s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.10%
    jit_opt_a.merge_forward              :     0.000014s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000052s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001415s :  5.67%
    jit_opt_a.replace_old_param          :     0.000067s :  0.27%
    jit_opt_a.inline_without_move        :     0.000061s :  0.24%
    jit_opt_a.renormalize                :     0.002440s :  9.78%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000063s :  0.25%
    jit_opt_a.cse                        :     0.000171s :  0.69%
    jit_opt_a.replace_applicator         :     0.000078s :  0.31%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.04%
    rewriter_after_opt_a                 :     0.000035s :  0.14%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000474s :  1.90%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.07%
    cconv                                :     0.000021s :  0.08%
    loop_unroll                          :     0.000407s :  1.63%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000012s :  0.05%
    remove_dup_value                     :     0.000015s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.01%
    add_recomputation                    :     0.000045s :  0.18%
    cse_after_recomputation.cse          :     0.000014s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.02%
    opt_after_jit_grad                   :     0.000447s :  1.79%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000028s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005822s : 23.35%
    execute                              :     0.000005s :  0.02%


Time group info:
------[substitution.]   0.000143    23
 1.28% :     0.000002s :      2: substitution.elim_not_effective
 0.83% :     0.000001s :      2: substitution.fold_const_symbol
 2.34% :     0.000003s :      3: substitution.graph_param_transform
83.70% :     0.000120s :      4: substitution.inline
 1.69% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.73% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.28% :     0.000003s :      2: substitution.replace_old_param
 5.14% :     0.000007s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003888     2
83.00% :     0.003227s :      1: type_inference.infer
17.00% :     0.000661s :      1: type_inference.specialize
------[replace.]   0.000054     6
66.02% :     0.000036s :      4: replace.inline
33.98% :     0.000018s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000124     6
94.90% :     0.000118s :      4: match.inline
 5.10% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000108   691
 1.53% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.81% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 1.85% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000001s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.61% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.16% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.50% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.51% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000000s :      3: predicate.opt_reshape
 2.14% :     0.000002s :     15: predicate.partial_eliminate
 1.39% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000001s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.05% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 1.13% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.41% :     0.000008s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.34% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.79% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.98% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000436    12
48.69% :     0.000212s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.03% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.28% :     0.000197s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017765    76
 0.25% :     0.000044s :      1: add_recomputation
 0.26% :     0.000046s :      1: auto_monad
 0.10% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.56% :     0.000277s :      1: bootstrap
 0.11% :     0.000019s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000005s :      1: graph_reusing
25.46% :     0.004523s :      1: jit_opt_a
 0.73% :     0.000129s :      1: jit_opt_after_cconv
 0.29% :     0.000052s :      1: jit_opt_b
 2.31% :     0.000410s :      1: loop_unroll
 2.65% :     0.000470s :      1: mutable_eliminate
 4.07% :     0.000722s :     26: opt.transform.jit_opt_a
 0.23% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.50% :     0.000445s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000027s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000010s :      1: remove_dup_value
 1.14% :     0.000202s :      1: renormalize.infer
 1.19% :     0.000212s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000028s :      1: rewriter_after_opt_a
 0.41% :     0.000073s :      1: rewriter_before_opt_a
 0.40% :     0.000070s :      1: symbol_engine_optimizer
32.31% :     0.005740s :      1: task_emit
22.19% :     0.003942s :      1: type_inference
 0.31% :     0.000055s :      1: validate


Time group info:
------[substitution.]   0.000447   110
 0.34% :     0.000002s :      3: substitution.elim_not_effective
 0.27% :     0.000001s :      3: substitution.fold_const_symbol
 1.11% :     0.000005s :      4: substitution.graph_param_transform
63.18% :     0.000282s :     12: substitution.inline
 3.37% :     0.000015s :      2: substitution.inline_without_move
 1.94% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.30% :     0.000010s :      7: substitution.minmaximum_grad
 1.35% :     0.000006s :      5: substitution.partial_eliminate
 2.36% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.38% :     0.000024s :     10: substitution.replace_applicator
 1.47% :     0.000007s :      7: substitution.replace_old_param
 0.53% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.55% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.14% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 8.71% :     0.000039s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010082     2
89.72% :     0.009046s :      1: type_inference.infer
10.28% :     0.001036s :      1: type_inference.specialize
------[replace.]   0.000138    19
63.61% :     0.000088s :     12: replace.inline
36.39% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000297    19
92.98% :     0.000276s :     12: match.inline
 7.02% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000357  2568
 1.69% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.44% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.46% :     0.000005s :     43: predicate.addn_check_dump
 1.68% :     0.000006s :     43: predicate.addn_zero_filter
 2.03% :     0.000007s :     43: predicate.arithmetic_simplify
 1.64% :     0.000006s :     43: predicate.cast_eliminate
 0.25% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.51% :     0.000005s :     43: predicate.compare_switch_simplify
 1.58% :     0.000006s :     43: predicate.depend_value_elim
 1.44% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.51% :     0.000005s :     43: predicate.dict_get_item_eliminator
 1.49% :     0.000005s :     43: predicate.dict_set_item_eliminator
 0.38% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.14% :     0.000000s :      4: predicate.elim_not_effective
 0.19% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.50% :     0.000005s :     43: predicate.environ_add_const_eliminate
 1.46% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.46% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.51% :     0.000005s :     43: predicate.environ_get_eliminate
 1.48% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.80% :     0.000003s :     19: predicate.get_grad_eliminate
 0.11% :     0.000000s :      4: predicate.graph_param_transform
 4.32% :     0.000015s :     70: predicate.inline
 1.80% :     0.000006s :     40: predicate.inline_without_move
 0.36% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.89% :     0.000003s :     19: predicate.less_batch_normalization
 1.83% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.94% :     0.000007s :     54: predicate.load_eliminater
 0.47% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.14% :     0.000011s :     85: predicate.loop_unroll_before_grad
 1.80% :     0.000006s :     47: predicate.make_slice_get_slice_eliminator
 1.47% :     0.000005s :     43: predicate.merge_addn
 1.57% :     0.000006s :     43: predicate.minmaximum_grad
 0.57% :     0.000002s :      4: predicate.mutable_eliminate
 0.19% :     0.000001s :      4: predicate.opt_reshape
 2.42% :     0.000009s :     54: predicate.partial_eliminate
 1.57% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.86% :     0.000007s :     43: predicate.reduce_eliminate
 1.81% :     0.000006s :     50: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.55% :     0.000009s :     96: predicate.replace_applicator
 0.92% :     0.000003s :     40: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.76% :     0.000006s :     43: predicate.reshape_eliminate
 1.56% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.29% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.68% :     0.000006s :     43: predicate.same_eliminate
 0.48% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.45% :     0.000002s :      8: predicate.special_op_eliminate
 0.88% :     0.000003s :     19: predicate.specialize_transform
 1.68% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.54% :     0.000005s :     43: predicate.stack_unstack_eliminate
 0.16% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.71% :     0.000010s :     62: predicate.switch_defer_inline
 2.58% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.25% :     0.000022s :    151: predicate.switch_simplify
 1.47% :     0.000005s :     43: predicate.tile_eliminate
 1.50% :     0.000005s :     43: predicate.transpose_eliminate
 1.75% :     0.000006s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.61% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.08% :     0.000011s :     58: predicate.tuple_list_get_item_eliminator
 1.84% :     0.000007s :     43: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000006s :     50: predicate.tuple_to_list_eliminator_
 1.94% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000010s :     73: predicate.updatestate_useless_node_eliminater
 1.83% :     0.000007s :     43: predicate.value_based_eliminate
 0.15% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.25% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001378    29
60.26% :     0.000830s :     13: func_graph_cloner_run.FuncGraphClonerGraph
39.74% :     0.000548s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.033880    91
 0.14% :     0.000048s :      1: add_recomputation
 0.28% :     0.000096s :      1: auto_monad
 0.04% :     0.000014s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.69% :     0.000233s :      1: bootstrap
 0.07% :     0.000023s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000027s :      1: cse_after_recomputation
 0.02% :     0.000006s :      1: environ_conv
 0.13% :     0.000045s :      1: event_method
 0.02% :     0.000008s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
30.51% :     0.010338s :      1: jit_opt_a
 0.46% :     0.000156s :      1: jit_opt_after_cconv
 0.17% :     0.000057s :      1: jit_opt_b
 1.22% :     0.000413s :      1: loop_unroll
 1.42% :     0.000482s :      1: mutable_eliminate
 7.42% :     0.002515s :     39: opt.transform.jit_opt_a
 0.17% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000030s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.34% :     0.000454s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000003s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000029s :      1: py_interpret_to_execute
 0.03% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 3.63% :     0.001230s :      2: renormalize.infer
 3.53% :     0.001196s :      2: renormalize.specialize
 0.02% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000038s :      1: rewriter_after_opt_a
 0.27% :     0.000093s :      1: rewriter_before_opt_a
 0.23% :     0.000077s :      1: symbol_engine_optimizer
17.21% :     0.005830s :      1: task_emit
30.01% :     0.010167s :      1: type_inference
 0.14% :     0.000046s :      1: validate


TotalTime = 0.0212017, [33]
[bootstrap]: 0.00027557
[type_inference]: 0.00560481
[event_method]: 1.922e-05
[auto_monad]: 6.511e-05
[graph_reusing]: 6.96001e-06
[pre_auto_parallel]: 3.26001e-06
[py_interpret_to_execute]: 4.636e-05
[rewriter_before_opt_a]: 8.185e-05
[expand_dump_flag]: 4.23999e-06
[jit_opt_a]: 0.00592083, [2]
    [Cycle 1]: 0.00188672, [27]
        [switch_simplify]: 6.638e-05
        [loop_unroll]: 3.158e-05
        [a_1]: 0.00057844
        [with_stream_mark]: 2.079e-05
        [recompute_prepare]: 8.30999e-06
        [updatestate_depend_eliminate]: 3.88001e-06
        [updatestate_assign_eliminate]: 2.99001e-06
        [updatestate_loads_eliminate]: 2.48002e-06
        [parameter_eliminate]: 2.34999e-06
        [specialize_transform]: 6.34001e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 5.77001e-06
        [meta_shard_fg_expand]: 2.58e-06
        [get_grad_eliminate_]: 5.47999e-06
        [merge_forward]: 3.62002e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.501e-05
        [j_node_and_user_rematch]: 8.97e-06
        [meta_fg_expand]: 2.46e-06
        [replace_old_param]: 9.47999e-06
        [inline_without_move]: 6.16e-06
        [renormalize]: 0.00084531
        [add_forward_monad_depend]: 5.97999e-06
        [auto_monad_grad]: 2.86999e-06
        [auto_monad_eliminator]: 1.377e-05
        [cse]: 2.794e-05
        [replace_applicator]: 1.343e-05
    [Cycle 2]: 0.00031135, [27]
        [switch_simplify]: 5.77999e-06
        [loop_unroll]: 4.85001e-06
        [a_1]: 8.967e-05
        [with_stream_mark]: 9.27001e-06
        [recompute_prepare]: 4.67e-06
        [updatestate_depend_eliminate]: 2.44001e-06
        [updatestate_assign_eliminate]: 1.93002e-06
        [updatestate_loads_eliminate]: 1.83002e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 5.17e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.36998e-06
        [meta_shard_fg_expand]: 1.04998e-06
        [get_grad_eliminate_]: 4.42998e-06
        [merge_forward]: 2.45002e-06
        [cell_reuse_recompute_pass]: 1.31002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.202e-05
        [j_node_and_user_rematch]: 7.77998e-06
        [meta_fg_expand]: 1.40999e-06
        [replace_old_param]: 7.29001e-06
        [inline_without_move]: 4.50999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 5.09998e-06
        [cse]: 9.34998e-06
        [replace_applicator]: 4.66002e-06
[py_interpret_to_execute_after_opt_a]: 1.057e-05
[rewriter_after_opt_a]: 3.583e-05
[convert_after_rewriter]: 6.84999e-06
[order_py_execute_after_rewriter]: 4.78001e-06
[mutable_eliminate]: 0.0006214
[jit_opt_b]: 5.308e-05, [1]
    [Cycle 1]: 4.559e-05, [2]
        [frontend_op_eliminate]: 1.909e-05
        [inline_after_opt_a]: 1.489e-05
[cconv]: 2.502e-05
[loop_unroll]: 0.00043312
[jit_opt_after_cconv]: 0.00014383, [1]
    [Cycle 1]: 0.00013683, [11]
        [c_1]: 2.175e-05
        [parameter_eliminate]: 2.29001e-06
        [updatestate_depend_eliminate]: 5.22999e-06
        [updatestate_assign_eliminate]: 2.43998e-06
        [updatestate_loads_eliminate]: 2.64999e-06
        [cse]: 2.1e-05
        [call_graph_tuple_transform]: 2.015e-05
        [tuple_list_get_item_eliminator]: 5.02e-06
        [none_parameter_eliminate]: 1.76e-06
        [renormalize]: 5.29981e-07
        [switch_simplify]: 5.25999e-06
[remove_dup_value]: 1.392e-05
[partial_unused_args_eliminate]: 1.89e-06
[environ_conv]: 5.22999e-06
[add_recomputation]: 5.152e-05
[cse_after_recomputation]: 1.932e-05, [1]
    [Cycle 1]: 1.436e-05, [1]
        [cse]: 8.93002e-06
[auto_monad_reorder]: 1.426e-05
[get_jit_bprop_graph]: 1.87999e-06
[rewriter_after_jit_bprop_graph]: 6.12999e-06
[opt_after_jit_grad]: 0.00048852
[symbol_engine_optimizer]: 7.951e-05, [1]
    [Cycle 1]: 7.132e-05, [6]
        [build]: 3.76999e-06
        [elim_shapecalc]: 7.47002e-06
        [elim_not_effective]: 1.438e-05
        [opt_reshape]: 6.41e-06
        [fold_const_symbol]: 9.02999e-06
        [renormalize]: 6.10016e-07
[validate]: 4.243e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.00688361
[execute]: 9.33002e-06
Sums
    bootstrap                            :     0.000276s :  1.64%
    type_inference                       :     0.005605s : 33.29%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000065s :  0.39%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000046s :  0.28%
    rewriter_before_opt_a                :     0.000082s :  0.49%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000072s :  0.43%
    jit_opt_a.loop_unroll                :     0.000036s :  0.22%
    jit_opt_a.a_1                        :     0.000668s :  3.97%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000845s :  5.02%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.11%
    jit_opt_a.cse                        :     0.000037s :  0.22%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000036s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000621s :  3.69%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000025s :  0.15%
    loop_unroll                          :     0.000433s :  2.57%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000052s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000489s :  2.90%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006884s : 40.89%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000238    23
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000002s :      2: substitution.fold_const_symbol
 2.40% :     0.000006s :      3: substitution.graph_param_transform
85.63% :     0.000204s :      4: substitution.inline
 1.26% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.03% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.77% :     0.000004s :      2: substitution.replace_old_param
 5.35% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005512     2
81.82% :     0.004510s :      1: type_inference.infer
18.18% :     0.001002s :      1: type_inference.specialize
------[replace.]   0.000076     6
67.24% :     0.000051s :      4: replace.inline
32.76% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000213     6
94.52% :     0.000201s :      4: match.inline
 5.48% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.62% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.43% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.48% :     0.000002s :     10: predicate.addn_zero_filter
 1.79% :     0.000002s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.27% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.94% :     0.000007s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000001s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.75% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 2.10% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.45% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.37% :     0.000002s :     10: predicate.reduce_eliminate
 1.75% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.48% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.89% :     0.000009s :     53: predicate.switch_simplify
 1.32% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.20% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.90% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000668    12
46.01% :     0.000308s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.55% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.44% :     0.000324s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023046    76
 0.23% :     0.000054s :      1: add_recomputation
 0.30% :     0.000069s :      1: auto_monad
 0.07% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.27% :     0.000293s :      1: bootstrap
 0.12% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.71% :     0.005925s :      1: jit_opt_a
 0.64% :     0.000147s :      1: jit_opt_after_cconv
 0.24% :     0.000056s :      1: jit_opt_b
 1.92% :     0.000441s :      1: loop_unroll
 2.74% :     0.000631s :      1: mutable_eliminate
 3.88% :     0.000895s :     26: opt.transform.jit_opt_a
 0.21% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.16% :     0.000498s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000049s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.94% :     0.000446s :      1: renormalize.infer
 1.70% :     0.000391s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000039s :      1: rewriter_after_opt_a
 0.37% :     0.000086s :      1: rewriter_before_opt_a
 0.36% :     0.000082s :      1: symbol_engine_optimizer
29.95% :     0.006901s :      1: task_emit
24.39% :     0.005622s :      1: type_inference
 0.28% :     0.000064s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.614.327 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dd27f70> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.614.846 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0142648, [30]
[bootstrap]: 0.00033011
[type_inference]: 0.00462621
[event_method]: 1.998e-05
[auto_monad]: 6.462e-05
[graph_reusing]: 5.94e-06
[pre_auto_parallel]: 2.39001e-06
[py_interpret_to_execute]: 3.465e-05
[rewriter_before_opt_a]: 7.437e-05
[expand_dump_flag]: 2.86999e-06
[jit_opt_a]: 0.00672614, [2]
    [Cycle 1]: 0.00170196, [27]
        [switch_simplify]: 5.898e-05
        [loop_unroll]: 2.948e-05
        [a_1]: 0.00054071
        [with_stream_mark]: 2.032e-05
        [recompute_prepare]: 9.41e-06
        [updatestate_depend_eliminate]: 3.79002e-06
        [updatestate_assign_eliminate]: 2.90002e-06
        [updatestate_loads_eliminate]: 2.74001e-06
        [parameter_eliminate]: 1.80001e-06
        [specialize_transform]: 5.88998e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.59e-06
        [meta_shard_fg_expand]: 2.31e-06
        [get_grad_eliminate_]: 5.14e-06
        [merge_forward]: 3.35e-06
        [cell_reuse_recompute_pass]: 1.43002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.492e-05
        [j_node_and_user_rematch]: 1.094e-05
        [meta_fg_expand]: 2.02001e-06
        [replace_old_param]: 1.096e-05
        [inline_without_move]: 6.09001e-06
        [renormalize]: 0.00069628
        [add_forward_monad_depend]: 7.27002e-06
        [auto_monad_grad]: 2.42001e-06
        [auto_monad_eliminator]: 1.504e-05
        [cse]: 2.993e-05
        [replace_applicator]: 1.565e-05
    [Cycle 2]: 0.00033101, [27]
        [switch_simplify]: 5.30999e-06
        [loop_unroll]: 5.17999e-06
        [a_1]: 8.858e-05
        [with_stream_mark]: 1.036e-05
        [recompute_prepare]: 5.37001e-06
        [updatestate_depend_eliminate]: 2.89001e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 5.52999e-06
        [updatestate_useless_node_eliminater]: 4.60999e-06
        [accelerated_algorithm]: 5.34998e-06
        [meta_shard_fg_expand]: 1.43002e-06
        [get_grad_eliminate_]: 4.33999e-06
        [merge_forward]: 2.71e-06
        [cell_reuse_recompute_pass]: 3.17002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.501e-05
        [j_node_and_user_rematch]: 8.40999e-06
        [meta_fg_expand]: 1.84e-06
        [replace_old_param]: 8.09002e-06
        [inline_without_move]: 4.89e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.86e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 5.56998e-06
        [cse]: 1.12e-05
        [replace_applicator]: 5.52001e-06
[py_interpret_to_execute_after_opt_a]: 1.332e-05
[rewriter_after_opt_a]: 3.663e-05
[convert_after_rewriter]: 7e-06
[order_py_execute_after_rewriter]: 4.96997e-06
[mutable_eliminate]: 0.00067488
[jit_opt_b]: 5.149e-05, [1]
    [Cycle 1]: 4.408e-05, [2]
        [frontend_op_eliminate]: 1.663e-05
        [inline_after_opt_a]: 1.547e-05
[cconv]: 2.631e-05
[loop_unroll]: 0.00046761
[jit_opt_after_cconv]: 0.00015603, [1]
    [Cycle 1]: 0.00014864, [11]
        [c_1]: 2.108e-05
        [parameter_eliminate]: 3.3e-06
        [updatestate_depend_eliminate]: 6.91999e-06
        [updatestate_assign_eliminate]: 2.74001e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [cse]: 2.363e-05
        [call_graph_tuple_transform]: 2.134e-05
        [tuple_list_get_item_eliminator]: 5.49e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 1.14e-06
        [switch_simplify]: 5.82999e-06
[remove_dup_value]: 1.501e-05
[partial_unused_args_eliminate]: 2.06e-06
[environ_conv]: 6.18998e-06
[add_recomputation]: 5.15e-05
[cse_after_recomputation]: 2.333e-05, [1]
    [Cycle 1]: 1.723e-05, [1]
        [cse]: 9.71e-06
[auto_monad_reorder]: 1.522e-05
[get_jit_bprop_graph]: 1.62999e-06
[rewriter_after_jit_bprop_graph]: 7e-06
[opt_after_jit_grad]: 0.00050411
[symbol_engine_optimizer]: 7.917e-05, [1]
    [Cycle 1]: 7.213e-05, [6]
        [build]: 4.92999e-06
        [elim_shapecalc]: 7.71001e-06
        [elim_not_effective]: 1.529e-05
        [opt_reshape]: 5.89e-06
        [fold_const_symbol]: 8.60999e-06
        [renormalize]: 7.7e-07
[validate]: 4.073e-05
Sums
    bootstrap                            :     0.000330s :  3.69%
    type_inference                       :     0.004626s : 51.75%
    event_method                         :     0.000020s :  0.22%
    auto_monad                           :     0.000065s :  0.72%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.39%
    rewriter_before_opt_a                :     0.000074s :  0.83%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000064s :  0.72%
    jit_opt_a.loop_unroll                :     0.000035s :  0.39%
    jit_opt_a.a_1                        :     0.000629s :  7.04%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.34%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.21%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000696s :  7.79%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.23%
    jit_opt_a.cse                        :     0.000041s :  0.46%
    jit_opt_a.replace_applicator         :     0.000021s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.15%
    rewriter_after_opt_a                 :     0.000037s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000675s :  7.55%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000026s :  0.29%
    loop_unroll                          :     0.000468s :  5.23%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000024s :  0.26%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000052s :  0.58%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.08%
    opt_after_jit_grad                   :     0.000504s :  5.64%
    symbol_engine_optimizer.build        :     0.000005s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.46%


Time group info:
------[substitution.]   0.000214    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.57% :     0.000006s :      3: substitution.graph_param_transform
82.57% :     0.000177s :      4: substitution.inline
 1.90% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.63% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.50% :     0.000005s :      2: substitution.replace_old_param
 6.22% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004555     2
80.69% :     0.003676s :      1: type_inference.infer
19.31% :     0.000880s :      1: type_inference.specialize
------[replace.]   0.000069     6
67.75% :     0.000047s :      4: replace.inline
32.25% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000187     6
93.45% :     0.000175s :      4: match.inline
 6.55% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.20% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.84% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.52% :     0.000002s :     10: predicate.addn_zero_filter
 2.58% :     0.000003s :     10: predicate.arithmetic_simplify
 1.13% :     0.000001s :     10: predicate.cast_eliminate
 0.66% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.49% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.16% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.08% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 6.20% :     0.000007s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.54% :     0.000002s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.77% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.11% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.12% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.97% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.85% :     0.000002s :     18: predicate.replace_applicator
 0.88% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.24% :     0.000001s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.12% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000009s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.76% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.38% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.67% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000606    12
50.56% :     0.000306s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.16% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.29% :     0.000268s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015870    72
 0.35% :     0.000055s :      1: add_recomputation
 0.43% :     0.000068s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 2.19% :     0.000348s :      1: bootstrap
 0.18% :     0.000029s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
42.41% :     0.006730s :      1: jit_opt_a
 1.00% :     0.000159s :      1: jit_opt_after_cconv
 0.34% :     0.000054s :      1: jit_opt_b
 3.01% :     0.000478s :      1: loop_unroll
 4.32% :     0.000685s :      1: mutable_eliminate
 5.39% :     0.000855s :     26: opt.transform.jit_opt_a
 0.32% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000026s :      4: opt.transform.jit_opt_b
 0.10% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.23% :     0.000513s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000038s :      1: py_interpret_to_execute
 0.10% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.27% :     0.000361s :      1: renormalize.infer
 2.06% :     0.000327s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000041s :      1: rewriter_after_opt_a
 0.49% :     0.000078s :      1: rewriter_before_opt_a
 0.52% :     0.000082s :      1: symbol_engine_optimizer
29.27% :     0.004645s :      1: type_inference


TotalTime = 0.0180025, [33]
[bootstrap]: 0.0002668
[type_inference]: 0.00429886
[event_method]: 1.697e-05
[auto_monad]: 6.585e-05
[graph_reusing]: 5.59e-06
[pre_auto_parallel]: 2.94999e-06
[py_interpret_to_execute]: 3.092e-05
[rewriter_before_opt_a]: 7.59e-05
[expand_dump_flag]: 3.00998e-06
[jit_opt_a]: 0.00498021, [2]
    [Cycle 1]: 0.0013735, [27]
        [switch_simplify]: 4.89e-05
        [loop_unroll]: 2.82e-05
        [a_1]: 0.00048863
        [with_stream_mark]: 1.564e-05
        [recompute_prepare]: 6.77002e-06
        [updatestate_depend_eliminate]: 3.58999e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.41998e-06
        [parameter_eliminate]: 2.19999e-06
        [specialize_transform]: 5.50001e-06
        [updatestate_useless_node_eliminater]: 4.92999e-06
        [accelerated_algorithm]: 5.86e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 5.11997e-06
        [merge_forward]: 4.1e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.371e-05
        [j_node_and_user_rematch]: 8.75001e-06
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 7.9e-06
        [inline_without_move]: 5.07e-06
        [renormalize]: 0.00047362
        [add_forward_monad_depend]: 5.37999e-06
        [auto_monad_grad]: 1.76e-06
        [auto_monad_eliminator]: 1.35e-05
        [cse]: 4.187e-05
        [replace_applicator]: 1.167e-05
    [Cycle 2]: 0.00031411, [27]
        [switch_simplify]: 5.35999e-06
        [loop_unroll]: 5.10999e-06
        [a_1]: 8.785e-05
        [with_stream_mark]: 8.35999e-06
        [recompute_prepare]: 5.34e-06
        [updatestate_depend_eliminate]: 2.43e-06
        [updatestate_assign_eliminate]: 1.97999e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 1.26997e-06
        [specialize_transform]: 4.85001e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 4.95001e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.35999e-06
        [merge_forward]: 2.58998e-06
        [cell_reuse_recompute_pass]: 1.67999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.231e-05
        [j_node_and_user_rematch]: 7.41001e-06
        [meta_fg_expand]: 1.45001e-06
        [replace_old_param]: 7.33e-06
        [inline_without_move]: 4.57e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.47001e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 5.50001e-06
        [cse]: 1.055e-05
        [replace_applicator]: 5.22e-06
[py_interpret_to_execute_after_opt_a]: 8.67e-06
[rewriter_after_opt_a]: 3.363e-05
[convert_after_rewriter]: 6.84999e-06
[order_py_execute_after_rewriter]: 4.76002e-06
[mutable_eliminate]: 0.00049037
[jit_opt_b]: 4.713e-05, [1]
    [Cycle 1]: 4.06e-05, [2]
        [frontend_op_eliminate]: 1.518e-05
        [inline_after_opt_a]: 1.411e-05
[cconv]: 2.342e-05
[loop_unroll]: 0.00041212
[jit_opt_after_cconv]: 0.00013266, [1]
    [Cycle 1]: 0.00012647, [11]
        [c_1]: 1.996e-05
        [parameter_eliminate]: 2.58e-06
        [updatestate_depend_eliminate]: 5.15001e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.588e-05
        [call_graph_tuple_transform]: 1.797e-05
        [tuple_list_get_item_eliminator]: 5.08002e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.06002e-06
[remove_dup_value]: 1.318e-05
[partial_unused_args_eliminate]: 2.34999e-06
[environ_conv]: 5.02999e-06
[add_recomputation]: 4.547e-05
[cse_after_recomputation]: 2.023e-05, [1]
    [Cycle 1]: 1.477e-05, [1]
        [cse]: 8.87e-06
[auto_monad_reorder]: 1.481e-05
[get_jit_bprop_graph]: 1.48002e-06
[rewriter_after_jit_bprop_graph]: 4.62e-06
[opt_after_jit_grad]: 0.00044687
[symbol_engine_optimizer]: 7.055e-05, [1]
    [Cycle 1]: 6.423e-05, [6]
        [build]: 3.9e-06
        [elim_shapecalc]: 6.81001e-06
        [elim_not_effective]: 1.215e-05
        [opt_reshape]: 5.24998e-06
        [fold_const_symbol]: 7.98999e-06
        [renormalize]: 4.99975e-07
[validate]: 3.246e-05
[backend_pass]: 8.2e-07
[task_emit]: 0.00625312
[execute]: 7.53999e-06
Sums
    bootstrap                            :     0.000267s :  1.89%
    type_inference                       :     0.004299s : 30.40%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000066s :  0.47%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.22%
    rewriter_before_opt_a                :     0.000076s :  0.54%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.38%
    jit_opt_a.loop_unroll                :     0.000033s :  0.24%
    jit_opt_a.a_1                        :     0.000576s :  4.08%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000474s :  3.35%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000052s :  0.37%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000034s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000490s :  3.47%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000412s :  2.91%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000045s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000447s :  3.16%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006253s : 44.22%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000175    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 3.01% :     0.000005s :      3: substitution.graph_param_transform
82.69% :     0.000145s :      4: substitution.inline
 1.73% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.33% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.12% :     0.000004s :      2: substitution.replace_old_param
 6.37% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004237     2
82.03% :     0.003475s :      1: type_inference.infer
17.97% :     0.000762s :      1: type_inference.specialize
------[replace.]   0.000061     6
67.91% :     0.000042s :      4: replace.inline
32.09% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000152     6
93.41% :     0.000142s :      4: match.inline
 6.59% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.48% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000001s :     10: predicate.addn_zero_filter
 1.88% :     0.000002s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.39% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000000s :      3: predicate.graph_param_transform
 5.78% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.53% :     0.000002s :      6: predicate.less_batch_normalization
 1.77% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 1.77% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.09% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.60% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000002s :     15: predicate.partial_eliminate
 1.39% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.72% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000001s :     10: predicate.same_eliminate
 0.60% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.32% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.67% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.01% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000489    12
48.80% :     0.000239s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.28% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.92% :     0.000225s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019352    76
 0.25% :     0.000048s :      1: add_recomputation
 0.36% :     0.000069s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.44% :     0.000278s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.75% :     0.004983s :      1: jit_opt_a
 0.70% :     0.000135s :      1: jit_opt_after_cconv
 0.26% :     0.000050s :      1: jit_opt_b
 2.17% :     0.000420s :      1: loop_unroll
 2.58% :     0.000499s :      1: mutable_eliminate
 4.00% :     0.000773s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.35% :     0.000455s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.25% :     0.000242s :      1: renormalize.infer
 1.16% :     0.000225s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000036s :      1: rewriter_after_opt_a
 0.41% :     0.000079s :      1: rewriter_before_opt_a
 0.38% :     0.000073s :      1: symbol_engine_optimizer
32.37% :     0.006264s :      1: task_emit
22.28% :     0.004312s :      1: type_inference
 0.27% :     0.000051s :      1: validate


TotalTime = 0.0308197, [33]
[bootstrap]: 0.00019746
[type_inference]: 0.00957899
[event_method]: 3.813e-05
[auto_monad]: 7.488e-05
[graph_reusing]: 4.97e-06
[pre_auto_parallel]: 2.33002e-06
[py_interpret_to_execute]: 2.247e-05
[rewriter_before_opt_a]: 8.504e-05
[expand_dump_flag]: 2.99001e-06
[jit_opt_a]: 0.0108273, [3]
    [Cycle 1]: 0.00539311, [27]
        [switch_simplify]: 5.56e-05
        [loop_unroll]: 3.74e-05
        [a_1]: 0.00077422
        [with_stream_mark]: 2.036e-05
        [recompute_prepare]: 1.788e-05
        [updatestate_depend_eliminate]: 6.93e-06
        [updatestate_assign_eliminate]: 5.70001e-06
        [updatestate_loads_eliminate]: 5.19e-06
        [parameter_eliminate]: 2.02999e-06
        [specialize_transform]: 1.238e-05
        [updatestate_useless_node_eliminater]: 1.209e-05
        [accelerated_algorithm]: 1.224e-05
        [meta_shard_fg_expand]: 2.33002e-06
        [get_grad_eliminate_]: 1.158e-05
        [merge_forward]: 6.69999e-06
        [cell_reuse_recompute_pass]: 7.2e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.554e-05
        [j_node_and_user_rematch]: 2.539e-05
        [meta_fg_expand]: 0.00146732
        [replace_old_param]: 5.245e-05
        [inline_without_move]: 5.106e-05
        [renormalize]: 0.0022794
        [add_forward_monad_depend]: 1.095e-05
        [auto_monad_grad]: 6.04999e-06
        [auto_monad_eliminator]: 4.754e-05
        [cse]: 0.00017681
        [replace_applicator]: 6.71e-05
    [Cycle 2]: 0.00198513, [27]
        [switch_simplify]: 3.502e-05
        [loop_unroll]: 3.394e-05
        [a_1]: 0.00100431
        [with_stream_mark]: 1.443e-05
        [recompute_prepare]: 8.59e-06
        [updatestate_depend_eliminate]: 3.51001e-06
        [updatestate_assign_eliminate]: 3.45e-06
        [updatestate_loads_eliminate]: 2.89001e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 6.79999e-06
        [updatestate_useless_node_eliminater]: 6.14999e-06
        [accelerated_algorithm]: 6.41e-06
        [meta_shard_fg_expand]: 2.02001e-06
        [get_grad_eliminate_]: 5.91e-06
        [merge_forward]: 3.75e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.489e-05
        [j_node_and_user_rematch]: 1.059e-05
        [meta_fg_expand]: 3.577e-05
        [replace_old_param]: 1.051e-05
        [inline_without_move]: 5.99999e-06
        [renormalize]: 0.00054948
        [add_forward_monad_depend]: 5.84e-06
        [auto_monad_grad]: 1.94e-06
        [auto_monad_eliminator]: 1.518e-05
        [cse]: 2.488e-05
        [replace_applicator]: 1.56e-05
    [Cycle 3]: 0.00036176, [27]
        [switch_simplify]: 6.17001e-06
        [loop_unroll]: 5.56998e-06
        [a_1]: 0.00011601
        [with_stream_mark]: 9.90002e-06
        [recompute_prepare]: 5.81998e-06
        [updatestate_depend_eliminate]: 3.56001e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.40002e-06
        [parameter_eliminate]: 1.45999e-06
        [specialize_transform]: 6.48e-06
        [updatestate_useless_node_eliminater]: 5.67001e-06
        [accelerated_algorithm]: 5.61e-06
        [meta_shard_fg_expand]: 1.40001e-06
        [get_grad_eliminate_]: 5.66998e-06
        [merge_forward]: 3.14999e-06
        [cell_reuse_recompute_pass]: 1.55999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.487e-05
        [j_node_and_user_rematch]: 1.028e-05
        [meta_fg_expand]: 2.17999e-06
        [replace_old_param]: 8.30999e-06
        [inline_without_move]: 5.62001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.06002e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 6.78998e-06
        [cse]: 1.439e-05
        [replace_applicator]: 6.02999e-06
[py_interpret_to_execute_after_opt_a]: 1.43e-05
[rewriter_after_opt_a]: 3.602e-05
[convert_after_rewriter]: 7.89002e-06
[order_py_execute_after_rewriter]: 5.27999e-06
[mutable_eliminate]: 0.00073371
[jit_opt_b]: 5.89e-05, [1]
    [Cycle 1]: 5.101e-05, [2]
        [frontend_op_eliminate]: 1.941e-05
        [inline_after_opt_a]: 1.884e-05
[cconv]: 3.088e-05
[loop_unroll]: 0.00044147
[jit_opt_after_cconv]: 0.00019925, [1]
    [Cycle 1]: 0.00019127, [11]
        [c_1]: 2.685e-05
        [parameter_eliminate]: 1.88e-05
        [updatestate_depend_eliminate]: 8.15e-06
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 3.5e-06
        [cse]: 2.823e-05
        [call_graph_tuple_transform]: 3.585e-05
        [tuple_list_get_item_eliminator]: 6.54999e-06
        [none_parameter_eliminate]: 1.37999e-06
        [renormalize]: 5.99975e-07
        [switch_simplify]: 6.36e-06
[remove_dup_value]: 1.786e-05
[partial_unused_args_eliminate]: 2.64001e-06
[environ_conv]: 5.57999e-06
[add_recomputation]: 5.387e-05
[cse_after_recomputation]: 2.666e-05, [1]
    [Cycle 1]: 2.022e-05, [1]
        [cse]: 1.433e-05
[auto_monad_reorder]: 1.754e-05
[get_jit_bprop_graph]: 1.89999e-06
[rewriter_after_jit_bprop_graph]: 6.37001e-06
[opt_after_jit_grad]: 0.00049145
[symbol_engine_optimizer]: 7.938e-05, [1]
    [Cycle 1]: 7.255e-05, [6]
        [build]: 3.73001e-06
        [elim_shapecalc]: 8.25e-06
        [elim_not_effective]: 1.462e-05
        [opt_reshape]: 7.12002e-06
        [fold_const_symbol]: 1.115e-05
        [renormalize]: 5.00004e-07
[validate]: 4.091e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00750505
[execute]: 8.50001e-06
Sums
    bootstrap                            :     0.000197s :  0.73%
    type_inference                       :     0.009579s : 35.55%
    event_method                         :     0.000038s :  0.14%
    auto_monad                           :     0.000075s :  0.28%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000022s :  0.08%
    rewriter_before_opt_a                :     0.000085s :  0.32%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000097s :  0.36%
    jit_opt_a.loop_unroll                :     0.000077s :  0.29%
    jit_opt_a.a_1                        :     0.001895s :  7.03%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000010s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.09%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000055s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000046s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001505s :  5.59%
    jit_opt_a.replace_old_param          :     0.000071s :  0.26%
    jit_opt_a.inline_without_move        :     0.000063s :  0.23%
    jit_opt_a.renormalize                :     0.002829s : 10.50%
    jit_opt_a.add_forward_monad_depend   :     0.000018s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000070s :  0.26%
    jit_opt_a.cse                        :     0.000216s :  0.80%
    jit_opt_a.replace_applicator         :     0.000089s :  0.33%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.05%
    rewriter_after_opt_a                 :     0.000036s :  0.13%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000734s :  2.72%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.07%
    cconv                                :     0.000031s :  0.11%
    loop_unroll                          :     0.000441s :  1.64%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000019s :  0.07%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000036s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000018s :  0.07%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000054s :  0.20%
    cse_after_recomputation.cse          :     0.000014s :  0.05%
    auto_monad_reorder                   :     0.000018s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000491s :  1.82%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007505s : 27.85%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000439   110
 0.57% :     0.000003s :      3: substitution.elim_not_effective
 0.47% :     0.000002s :      3: substitution.fold_const_symbol
 1.48% :     0.000007s :      4: substitution.graph_param_transform
60.93% :     0.000267s :     12: substitution.inline
 3.69% :     0.000016s :      2: substitution.inline_without_move
 1.99% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.53% :     0.000011s :      7: substitution.minmaximum_grad
 1.37% :     0.000006s :      5: substitution.partial_eliminate
 2.51% :     0.000011s :     14: substitution.remove_not_recompute_node
 6.07% :     0.000027s :     10: substitution.replace_applicator
 1.74% :     0.000008s :      7: substitution.replace_old_param
 0.56% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.60% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.04% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 8.46% :     0.000037s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009523     2
89.76% :     0.008548s :      1: type_inference.infer
10.24% :     0.000975s :      1: type_inference.specialize
------[replace.]   0.000134    19
63.41% :     0.000085s :     12: replace.inline
36.59% :     0.000049s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000279    19
93.60% :     0.000261s :     12: match.inline
 6.40% :     0.000018s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000357  2568
 1.62% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.45% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.43% :     0.000005s :     43: predicate.addn_check_dump
 1.74% :     0.000006s :     43: predicate.addn_zero_filter
 1.89% :     0.000007s :     43: predicate.arithmetic_simplify
 1.54% :     0.000005s :     43: predicate.cast_eliminate
 0.18% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.43% :     0.000005s :     43: predicate.compare_switch_simplify
 1.53% :     0.000005s :     43: predicate.depend_value_elim
 1.46% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.49% :     0.000005s :     43: predicate.dict_get_item_eliminator
 1.49% :     0.000005s :     43: predicate.dict_set_item_eliminator
 0.39% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      4: predicate.elim_not_effective
 0.20% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000005s :     43: predicate.environ_add_const_eliminate
 1.51% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.43% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.50% :     0.000005s :     43: predicate.environ_get_eliminate
 1.44% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.92% :     0.000003s :     19: predicate.get_grad_eliminate
 0.16% :     0.000001s :      4: predicate.graph_param_transform
 4.26% :     0.000015s :     70: predicate.inline
 1.86% :     0.000007s :     40: predicate.inline_without_move
 0.38% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.88% :     0.000003s :     19: predicate.less_batch_normalization
 1.82% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.92% :     0.000007s :     54: predicate.load_eliminater
 0.60% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.16% :     0.000011s :     85: predicate.loop_unroll_before_grad
 1.81% :     0.000006s :     47: predicate.make_slice_get_slice_eliminator
 1.49% :     0.000005s :     43: predicate.merge_addn
 1.54% :     0.000005s :     43: predicate.minmaximum_grad
 0.72% :     0.000003s :      4: predicate.mutable_eliminate
 0.30% :     0.000001s :      4: predicate.opt_reshape
 2.31% :     0.000008s :     54: predicate.partial_eliminate
 1.50% :     0.000005s :     43: predicate.print_const_string_wrapper
 1.79% :     0.000006s :     43: predicate.reduce_eliminate
 1.79% :     0.000006s :     50: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.65% :     0.000009s :     96: predicate.replace_applicator
 1.01% :     0.000004s :     40: predicate.replace_old_param
 0.14% :     0.000000s :      4: predicate.reset_defer_inline
 1.83% :     0.000007s :     43: predicate.reshape_eliminate
 1.54% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.29% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.52% :     0.000005s :     43: predicate.same_eliminate
 0.47% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.32% :     0.000001s :      8: predicate.special_op_eliminate
 0.85% :     0.000003s :     19: predicate.specialize_transform
 1.70% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000005s :     43: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.75% :     0.000010s :     62: predicate.switch_defer_inline
 2.55% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.20% :     0.000022s :    151: predicate.switch_simplify
 1.46% :     0.000005s :     43: predicate.tile_eliminate
 1.52% :     0.000005s :     43: predicate.transpose_eliminate
 1.82% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.73% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.06% :     0.000011s :     58: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000006s :     43: predicate.tuple_list_set_item_eliminator
 1.75% :     0.000006s :     50: predicate.tuple_to_list_eliminator_
 2.22% :     0.000008s :     54: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000011s :     73: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000006s :     43: predicate.value_based_eliminate
 0.15% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.21% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001432    29
59.53% :     0.000852s :     13: func_graph_cloner_run.FuncGraphClonerGraph
40.47% :     0.000580s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.036278    91
 0.16% :     0.000057s :      1: add_recomputation
 0.22% :     0.000079s :      1: auto_monad
 0.05% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.58% :     0.000212s :      1: bootstrap
 0.09% :     0.000034s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.08% :     0.000029s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.12% :     0.000042s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
29.86% :     0.010831s :      1: jit_opt_a
 0.56% :     0.000202s :      1: jit_opt_after_cconv
 0.17% :     0.000062s :      1: jit_opt_b
 1.24% :     0.000450s :      1: loop_unroll
 2.05% :     0.000744s :      1: mutable_eliminate
 6.84% :     0.002481s :     39: opt.transform.jit_opt_a
 0.20% :     0.000071s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000031s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.38% :     0.000500s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000025s :      1: py_interpret_to_execute
 0.05% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000020s :      1: remove_dup_value
 3.96% :     0.001438s :      2: renormalize.infer
 3.79% :     0.001376s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000039s :      1: rewriter_after_opt_a
 0.24% :     0.000088s :      1: rewriter_before_opt_a
 0.23% :     0.000082s :      1: symbol_engine_optimizer
20.72% :     0.007518s :      1: task_emit
26.44% :     0.009590s :      1: type_inference
 0.18% :     0.000065s :      1: validate


TotalTime = 0.0218713, [33]
[bootstrap]: 0.00030765
[type_inference]: 0.00561014
[event_method]: 1.831e-05
[auto_monad]: 6.434e-05
[graph_reusing]: 6.17001e-06
[pre_auto_parallel]: 2.74999e-06
[py_interpret_to_execute]: 3.889e-05
[rewriter_before_opt_a]: 6.964e-05
[expand_dump_flag]: 3.85e-06
[jit_opt_a]: 0.00595208, [2]
    [Cycle 1]: 0.00185344, [27]
        [switch_simplify]: 6.21e-05
        [loop_unroll]: 3.008e-05
        [a_1]: 0.0005857
        [with_stream_mark]: 1.998e-05
        [recompute_prepare]: 9.57999e-06
        [updatestate_depend_eliminate]: 3.30998e-06
        [updatestate_assign_eliminate]: 3.3e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [parameter_eliminate]: 1.76e-06
        [specialize_transform]: 6.38e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 6.98e-06
        [meta_shard_fg_expand]: 2.27001e-06
        [get_grad_eliminate_]: 5.46e-06
        [merge_forward]: 3.76999e-06
        [cell_reuse_recompute_pass]: 1.35001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.503e-05
        [j_node_and_user_rematch]: 9.29e-06
        [meta_fg_expand]: 2.46998e-06
        [replace_old_param]: 9.56998e-06
        [inline_without_move]: 5.14e-06
        [renormalize]: 0.00081177
        [add_forward_monad_depend]: 7.71999e-06
        [auto_monad_grad]: 2.64999e-06
        [auto_monad_eliminator]: 1.502e-05
        [cse]: 2.474e-05
        [replace_applicator]: 1.485e-05
    [Cycle 2]: 0.00034342, [27]
        [switch_simplify]: 6.16e-06
        [loop_unroll]: 5.02e-06
        [a_1]: 9.325e-05
        [with_stream_mark]: 1.141e-05
        [recompute_prepare]: 5.14998e-06
        [updatestate_depend_eliminate]: 2.37001e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [parameter_eliminate]: 1.42999e-06
        [specialize_transform]: 4.61002e-06
        [updatestate_useless_node_eliminater]: 4.91002e-06
        [accelerated_algorithm]: 5.72999e-06
        [meta_shard_fg_expand]: 1.55001e-06
        [get_grad_eliminate_]: 4.46002e-06
        [merge_forward]: 2.75002e-06
        [cell_reuse_recompute_pass]: 2.48002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.457e-05
        [j_node_and_user_rematch]: 8.86002e-06
        [meta_fg_expand]: 1.87999e-06
        [replace_old_param]: 7.71999e-06
        [inline_without_move]: 1.443e-05
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.55001e-06
        [auto_monad_grad]: 1.09e-06
        [auto_monad_eliminator]: 6.22001e-06
        [cse]: 1.044e-05
        [replace_applicator]: 6.16e-06
[py_interpret_to_execute_after_opt_a]: 1.119e-05
[rewriter_after_opt_a]: 3.626e-05
[convert_after_rewriter]: 6.81001e-06
[order_py_execute_after_rewriter]: 4.92e-06
[mutable_eliminate]: 0.00079603
[jit_opt_b]: 5.723e-05, [1]
    [Cycle 1]: 4.77e-05, [2]
        [frontend_op_eliminate]: 1.831e-05
        [inline_after_opt_a]: 1.725e-05
[cconv]: 2.897e-05
[loop_unroll]: 0.00050554
[jit_opt_after_cconv]: 0.00016383, [1]
    [Cycle 1]: 0.00015654, [11]
        [c_1]: 2.332e-05
        [parameter_eliminate]: 4.71002e-06
        [updatestate_depend_eliminate]: 6.94001e-06
        [updatestate_assign_eliminate]: 2.55002e-06
        [updatestate_loads_eliminate]: 2.83e-06
        [cse]: 2.338e-05
        [call_graph_tuple_transform]: 2.31e-05
        [tuple_list_get_item_eliminator]: 5.71e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 6.43e-06
[remove_dup_value]: 1.395e-05
[partial_unused_args_eliminate]: 2.09e-06
[environ_conv]: 6.04999e-06
[add_recomputation]: 5.206e-05
[cse_after_recomputation]: 2.399e-05, [1]
    [Cycle 1]: 1.836e-05, [1]
        [cse]: 1.012e-05
[auto_monad_reorder]: 1.489e-05
[get_jit_bprop_graph]: 2.68e-06
[rewriter_after_jit_bprop_graph]: 7.95998e-06
[opt_after_jit_grad]: 0.00061827
[symbol_engine_optimizer]: 8.825e-05, [1]
    [Cycle 1]: 7.809e-05, [6]
        [build]: 4.17e-06
        [elim_shapecalc]: 1.044e-05
        [elim_not_effective]: 1.553e-05
        [opt_reshape]: 6.58e-06
        [fold_const_symbol]: 1.026e-05
        [renormalize]: 1.05001e-06
[validate]: 4.352e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00707082
[execute]: 9.24e-06
Sums
    bootstrap                            :     0.000308s :  1.76%
    type_inference                       :     0.005610s : 32.16%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000064s :  0.37%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.22%
    rewriter_before_opt_a                :     0.000070s :  0.40%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.20%
    jit_opt_a.a_1                        :     0.000679s :  3.89%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000020s :  0.11%
    jit_opt_a.renormalize                :     0.000812s :  4.65%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.12%
    jit_opt_a.cse                        :     0.000035s :  0.20%
    jit_opt_a.replace_applicator         :     0.000021s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000036s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000796s :  4.56%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.10%
    cconv                                :     0.000029s :  0.17%
    loop_unroll                          :     0.000506s :  2.90%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000052s :  0.30%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000618s :  3.54%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007071s : 40.53%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000235    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.96% :     0.000002s :      2: substitution.fold_const_symbol
 2.67% :     0.000006s :      3: substitution.graph_param_transform
83.42% :     0.000196s :      4: substitution.inline
 1.56% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.25% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.30% :     0.000005s :      2: substitution.replace_old_param
 6.00% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005536     2
82.38% :     0.004560s :      1: type_inference.infer
17.62% :     0.000976s :      1: type_inference.specialize
------[replace.]   0.000082     6
64.46% :     0.000053s :      4: replace.inline
35.54% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000207     6
93.68% :     0.000193s :      4: match.inline
 6.32% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.13% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.61% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.19% :     0.000001s :     10: predicate.addn_zero_filter
 1.92% :     0.000002s :     10: predicate.arithmetic_simplify
 1.41% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.51% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.07% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 6.30% :     0.000008s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.66% :     0.000002s :      6: predicate.less_batch_normalization
 1.77% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.95% :     0.000002s :     15: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.92% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 2.38% :     0.000003s :      3: predicate.mutable_eliminate
 0.86% :     0.000001s :      3: predicate.opt_reshape
 1.92% :     0.000002s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.52% :     0.000002s :     18: predicate.replace_applicator
 0.86% :     0.000001s :      6: predicate.replace_old_param
 0.44% :     0.000001s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.77% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.38% :     0.000002s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.60% :     0.000009s :     53: predicate.switch_simplify
 1.09% :     0.000001s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.70% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.24% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.32% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.66% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.01% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.02% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000681    12
46.44% :     0.000316s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.56% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.99% :     0.000320s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023727    76
 0.24% :     0.000057s :      1: add_recomputation
 0.29% :     0.000068s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.37% :     0.000326s :      1: bootstrap
 0.13% :     0.000031s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000026s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
25.10% :     0.005956s :      1: jit_opt_a
 0.70% :     0.000167s :      1: jit_opt_after_cconv
 0.25% :     0.000060s :      1: jit_opt_b
 2.17% :     0.000515s :      1: loop_unroll
 3.41% :     0.000810s :      1: mutable_eliminate
 3.86% :     0.000917s :     26: opt.transform.jit_opt_a
 0.23% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000028s :      4: opt.transform.jit_opt_b
 0.07% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000039s :      4: opt.transform.symbol_engine_opt
 2.65% :     0.000628s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000042s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.82% :     0.000433s :      1: renormalize.infer
 1.56% :     0.000371s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000040s :      1: rewriter_after_opt_a
 0.31% :     0.000073s :      1: rewriter_before_opt_a
 0.38% :     0.000091s :      1: symbol_engine_optimizer
29.87% :     0.007088s :      1: task_emit
23.70% :     0.005623s :      1: type_inference
 0.29% :     0.000069s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.654.263 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dcafe50> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.654.960 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0206955, [33]
[bootstrap]: 0.00033647
[type_inference]: 0.00507329
[event_method]: 1.873e-05
[auto_monad]: 6.328e-05
[graph_reusing]: 5.47999e-06
[pre_auto_parallel]: 2.44001e-06
[py_interpret_to_execute]: 3.289e-05
[rewriter_before_opt_a]: 7.315e-05
[expand_dump_flag]: 3.76001e-06
[jit_opt_a]: 0.00616288, [2]
    [Cycle 1]: 0.00172345, [27]
        [switch_simplify]: 6.215e-05
        [loop_unroll]: 2.921e-05
        [a_1]: 0.00056127
        [with_stream_mark]: 2.317e-05
        [recompute_prepare]: 6.80998e-06
        [updatestate_depend_eliminate]: 3.41001e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 6.08002e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 1.90001e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 3.73001e-06
        [cell_reuse_recompute_pass]: 1.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.374e-05
        [j_node_and_user_rematch]: 8.48999e-06
        [meta_fg_expand]: 1.99999e-06
        [replace_old_param]: 9.34998e-06
        [inline_without_move]: 5.21002e-06
        [renormalize]: 0.00072081
        [add_forward_monad_depend]: 5.77999e-06
        [auto_monad_grad]: 2.23002e-06
        [auto_monad_eliminator]: 1.329e-05
        [cse]: 2.853e-05
        [replace_applicator]: 1.205e-05
    [Cycle 2]: 0.00031301, [27]
        [switch_simplify]: 5.45001e-06
        [loop_unroll]: 5.25999e-06
        [a_1]: 8.794e-05
        [with_stream_mark]: 8.85001e-06
        [recompute_prepare]: 4.65999e-06
        [updatestate_depend_eliminate]: 2.44001e-06
        [updatestate_assign_eliminate]: 2.05002e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.07999e-06
        [updatestate_useless_node_eliminater]: 4.65999e-06
        [accelerated_algorithm]: 5.00999e-06
        [meta_shard_fg_expand]: 1.24e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 1.87999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.218e-05
        [j_node_and_user_rematch]: 7.63999e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 7.3e-06
        [inline_without_move]: 4.48999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.33002e-06
        [auto_monad_grad]: 1.07e-06
        [auto_monad_eliminator]: 5.26998e-06
        [cse]: 9.54e-06
        [replace_applicator]: 4.89e-06
[py_interpret_to_execute_after_opt_a]: 1.014e-05
[rewriter_after_opt_a]: 3.627e-05
[convert_after_rewriter]: 6.41e-06
[order_py_execute_after_rewriter]: 4.90001e-06
[mutable_eliminate]: 0.00062283
[jit_opt_b]: 4.875e-05, [1]
    [Cycle 1]: 4.11e-05, [2]
        [frontend_op_eliminate]: 1.552e-05
        [inline_after_opt_a]: 1.388e-05
[cconv]: 2.522e-05
[loop_unroll]: 0.00043511
[jit_opt_after_cconv]: 0.00013615, [1]
    [Cycle 1]: 0.00012997, [11]
        [c_1]: 1.967e-05
        [parameter_eliminate]: 2.56998e-06
        [updatestate_depend_eliminate]: 5.46998e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 1.801e-05
        [call_graph_tuple_transform]: 1.805e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 4.86002e-06
[remove_dup_value]: 2.644e-05
[partial_unused_args_eliminate]: 2.41998e-06
[environ_conv]: 5.68002e-06
[add_recomputation]: 5.178e-05
[cse_after_recomputation]: 2.104e-05, [1]
    [Cycle 1]: 1.531e-05, [1]
        [cse]: 8.89e-06
[auto_monad_reorder]: 1.47e-05
[get_jit_bprop_graph]: 1.82001e-06
[rewriter_after_jit_bprop_graph]: 5.86e-06
[opt_after_jit_grad]: 0.00047091
[symbol_engine_optimizer]: 7.208e-05, [1]
    [Cycle 1]: 6.52e-05, [6]
        [build]: 4.05e-06
        [elim_shapecalc]: 7.56999e-06
        [elim_not_effective]: 1.169e-05
        [opt_reshape]: 5.66e-06
        [fold_const_symbol]: 8.41002e-06
        [renormalize]: 6.10016e-07
[validate]: 3.662e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00667288
[execute]: 8.57e-06
Sums
    bootstrap                            :     0.000336s :  2.11%
    type_inference                       :     0.005073s : 31.81%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000063s :  0.40%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.21%
    rewriter_before_opt_a                :     0.000073s :  0.46%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.42%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000649s :  4.07%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000721s :  4.52%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000038s :  0.24%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000036s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000623s :  3.91%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000025s :  0.16%
    loop_unroll                          :     0.000435s :  2.73%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000026s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000052s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000471s :  2.95%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006673s : 41.84%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000233    23
 0.77% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 2.21% :     0.000005s :      3: substitution.graph_param_transform
86.07% :     0.000200s :      4: substitution.inline
 1.28% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.97% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000005s :      2: substitution.replace_old_param
 5.12% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005003     2
83.94% :     0.004199s :      1: type_inference.infer
16.06% :     0.000804s :      1: type_inference.specialize
------[replace.]   0.000066     6
67.33% :     0.000045s :      4: replace.inline
32.67% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000209     6
94.84% :     0.000198s :      4: match.inline
 5.16% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.53% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.62% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 2.07% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000001s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.47% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.08% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.12% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.61% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.48% :     0.000002s :      6: predicate.less_batch_normalization
 1.69% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.83% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.20% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.76% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.47% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.37% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.86% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.26% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.52% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.23% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.70% :     0.000009s :     53: predicate.switch_simplify
 1.43% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.21% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.73% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.82% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000597    12
50.79% :     0.000303s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.49% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.73% :     0.000243s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022372    76
 0.24% :     0.000055s :      1: add_recomputation
 0.30% :     0.000067s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.59% :     0.000355s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
27.56% :     0.006166s :      1: jit_opt_a
 0.62% :     0.000139s :      1: jit_opt_after_cconv
 0.23% :     0.000051s :      1: jit_opt_b
 1.98% :     0.000444s :      1: loop_unroll
 2.82% :     0.000631s :      1: mutable_eliminate
 3.85% :     0.000862s :     26: opt.transform.jit_opt_a
 0.20% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.14% :     0.000479s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000030s :      1: remove_dup_value
 1.82% :     0.000407s :      1: renormalize.infer
 1.37% :     0.000306s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000039s :      1: rewriter_after_opt_a
 0.34% :     0.000076s :      1: rewriter_before_opt_a
 0.33% :     0.000075s :      1: symbol_engine_optimizer
29.90% :     0.006689s :      1: task_emit
22.74% :     0.005087s :      1: type_inference
 0.25% :     0.000056s :      1: validate


TotalTime = 0.0223704, [33]
[bootstrap]: 0.0003488
[type_inference]: 0.00608255
[event_method]: 1.893e-05
[auto_monad]: 6.554e-05
[graph_reusing]: 5.59e-06
[pre_auto_parallel]: 2.71e-06
[py_interpret_to_execute]: 3.864e-05
[rewriter_before_opt_a]: 7.664e-05
[expand_dump_flag]: 3.25e-06
[jit_opt_a]: 0.00624677, [2]
    [Cycle 1]: 0.00195693, [27]
        [switch_simplify]: 6.639e-05
        [loop_unroll]: 2.878e-05
        [a_1]: 0.00060947
        [with_stream_mark]: 2.556e-05
        [recompute_prepare]: 7.55e-06
        [updatestate_depend_eliminate]: 4.1e-06
        [updatestate_assign_eliminate]: 3.11001e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 1.85001e-06
        [specialize_transform]: 6.12001e-06
        [updatestate_useless_node_eliminater]: 4.72e-06
        [accelerated_algorithm]: 5.61e-06
        [meta_shard_fg_expand]: 1.93997e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 3.38999e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.482e-05
        [j_node_and_user_rematch]: 9.39e-06
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 9.82001e-06
        [inline_without_move]: 5.49998e-06
        [renormalize]: 0.00089021
        [add_forward_monad_depend]: 6.51e-06
        [auto_monad_grad]: 2.84999e-06
        [auto_monad_eliminator]: 1.359e-05
        [cse]: 2.736e-05
        [replace_applicator]: 1.209e-05
    [Cycle 2]: 0.00032601, [27]
        [switch_simplify]: 5.56e-06
        [loop_unroll]: 4.62e-06
        [a_1]: 8.975e-05
        [with_stream_mark]: 1.16e-05
        [recompute_prepare]: 5.21002e-06
        [updatestate_depend_eliminate]: 2.58998e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.83002e-06
        [specialize_transform]: 5.22e-06
        [updatestate_useless_node_eliminater]: 4.68001e-06
        [accelerated_algorithm]: 5.19998e-06
        [meta_shard_fg_expand]: 1.33002e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 2.46998e-06
        [cell_reuse_recompute_pass]: 1.99e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.325e-05
        [j_node_and_user_rematch]: 7.65e-06
        [meta_fg_expand]: 1.83002e-06
        [replace_old_param]: 7.23999e-06
        [inline_without_move]: 4.89e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 9.09989e-07
        [auto_monad_eliminator]: 4.95999e-06
        [cse]: 1.049e-05
        [replace_applicator]: 5.22999e-06
[py_interpret_to_execute_after_opt_a]: 1.146e-05
[rewriter_after_opt_a]: 3.793e-05
[convert_after_rewriter]: 6.88998e-06
[order_py_execute_after_rewriter]: 4.86002e-06
[mutable_eliminate]: 0.00077583
[jit_opt_b]: 5.58e-05, [1]
    [Cycle 1]: 4.78e-05, [2]
        [frontend_op_eliminate]: 2.019e-05
        [inline_after_opt_a]: 1.576e-05
[cconv]: 2.88e-05
[loop_unroll]: 0.00047559
[jit_opt_after_cconv]: 0.00015298, [1]
    [Cycle 1]: 0.00014654, [11]
        [c_1]: 2.119e-05
        [parameter_eliminate]: 3.45e-06
        [updatestate_depend_eliminate]: 6.63e-06
        [updatestate_assign_eliminate]: 2.55002e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [cse]: 2.526e-05
        [call_graph_tuple_transform]: 2.044e-05
        [tuple_list_get_item_eliminator]: 5.08002e-06
        [none_parameter_eliminate]: 1.73002e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 5.87001e-06
[remove_dup_value]: 1.453e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 5.44998e-06
[add_recomputation]: 5.35e-05
[cse_after_recomputation]: 2.108e-05, [1]
    [Cycle 1]: 1.487e-05, [1]
        [cse]: 8.59e-06
[auto_monad_reorder]: 1.542e-05
[get_jit_bprop_graph]: 2.27001e-06
[rewriter_after_jit_bprop_graph]: 7.60998e-06
[opt_after_jit_grad]: 0.00051027
[symbol_engine_optimizer]: 7.817e-05, [1]
    [Cycle 1]: 7.043e-05, [6]
        [build]: 4.87e-06
        [elim_shapecalc]: 8.2e-06
        [elim_not_effective]: 1.409e-05
        [opt_reshape]: 5.76003e-06
        [fold_const_symbol]: 8.79003e-06
        [renormalize]: 6.19999e-07
[validate]: 4.406e-05
[backend_pass]: 8.79983e-07
[task_emit]: 0.00692869
[execute]: 8.77e-06
Sums
    bootstrap                            :     0.000349s :  1.97%
    type_inference                       :     0.006083s : 34.29%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000066s :  0.37%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.22%
    rewriter_before_opt_a                :     0.000077s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000072s :  0.41%
    jit_opt_a.loop_unroll                :     0.000033s :  0.19%
    jit_opt_a.a_1                        :     0.000699s :  3.94%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000890s :  5.02%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.10%
    jit_opt_a.cse                        :     0.000038s :  0.21%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000038s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000776s :  4.37%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000029s :  0.16%
    loop_unroll                          :     0.000476s :  2.68%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000053s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000510s :  2.88%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.25%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006929s : 39.06%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000264    23
 0.67% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000002s :      2: substitution.fold_const_symbol
 2.16% :     0.000006s :      3: substitution.graph_param_transform
86.57% :     0.000229s :      4: substitution.inline
 1.38% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.97% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.68% :     0.000004s :      2: substitution.replace_old_param
 5.00% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006009     2
81.54% :     0.004900s :      1: type_inference.infer
18.46% :     0.001109s :      1: type_inference.specialize
------[replace.]   0.000079     6
66.07% :     0.000052s :      4: replace.inline
33.93% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000238     6
94.91% :     0.000226s :      4: match.inline
 5.09% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.25% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 2.00% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000001s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.19% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.19% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.19% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000000s :      3: predicate.graph_param_transform
 6.31% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.65% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.92% :     0.000002s :     15: predicate.load_eliminater
 1.49% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.81% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.98% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.24% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.41% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.81% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.23% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.67% :     0.000009s :     53: predicate.switch_simplify
 1.31% :     0.000002s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.94% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000793    12
51.26% :     0.000406s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.06% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.68% :     0.000330s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024290    76
 0.24% :     0.000057s :      1: add_recomputation
 0.28% :     0.000069s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.53% :     0.000371s :      1: bootstrap
 0.13% :     0.000031s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
25.73% :     0.006250s :      1: jit_opt_a
 0.64% :     0.000155s :      1: jit_opt_after_cconv
 0.24% :     0.000059s :      1: jit_opt_b
 1.99% :     0.000484s :      1: loop_unroll
 3.24% :     0.000788s :      1: mutable_eliminate
 3.79% :     0.000921s :     26: opt.transform.jit_opt_a
 0.20% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000029s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.14% :     0.000520s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000041s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.07% :     0.000502s :      1: renormalize.infer
 1.56% :     0.000380s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000041s :      1: rewriter_after_opt_a
 0.33% :     0.000079s :      1: rewriter_before_opt_a
 0.33% :     0.000081s :      1: symbol_engine_optimizer
28.61% :     0.006949s :      1: task_emit
25.13% :     0.006104s :      1: type_inference
 0.27% :     0.000066s :      1: validate


TotalTime = 0.0335096, [33]
[bootstrap]: 0.00021618
[type_inference]: 0.0115351
[event_method]: 4.485e-05
[auto_monad]: 9.824e-05
[graph_reusing]: 7.51999e-06
[pre_auto_parallel]: 3.26001e-06
[py_interpret_to_execute]: 3.067e-05
[rewriter_before_opt_a]: 0.00010531
[expand_dump_flag]: 3.71999e-06
[jit_opt_a]: 0.0122158, [3]
    [Cycle 1]: 0.00604197, [27]
        [switch_simplify]: 6.452e-05
        [loop_unroll]: 3.778e-05
        [a_1]: 0.00084671
        [with_stream_mark]: 3.308e-05
        [recompute_prepare]: 2.144e-05
        [updatestate_depend_eliminate]: 8.49998e-06
        [updatestate_assign_eliminate]: 6.10002e-06
        [updatestate_loads_eliminate]: 6.64001e-06
        [parameter_eliminate]: 2.88e-06
        [specialize_transform]: 1.348e-05
        [updatestate_useless_node_eliminater]: 1.216e-05
        [accelerated_algorithm]: 1.278e-05
        [meta_shard_fg_expand]: 3.66001e-06
        [get_grad_eliminate_]: 1.184e-05
        [merge_forward]: 8.54e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.933e-05
        [j_node_and_user_rematch]: 2.428e-05
        [meta_fg_expand]: 0.0017164
        [replace_old_param]: 5.835e-05
        [inline_without_move]: 5.408e-05
        [renormalize]: 0.00255503
        [add_forward_monad_depend]: 9.27999e-06
        [auto_monad_grad]: 5.71003e-06
        [auto_monad_eliminator]: 4.834e-05
        [cse]: 0.0001556
        [replace_applicator]: 6.804e-05
    [Cycle 2]: 0.00213453, [27]
        [switch_simplify]: 3.959e-05
        [loop_unroll]: 3.577e-05
        [a_1]: 0.00110432
        [with_stream_mark]: 1.488e-05
        [recompute_prepare]: 9.16998e-06
        [updatestate_depend_eliminate]: 3.75e-06
        [updatestate_assign_eliminate]: 3.36001e-06
        [updatestate_loads_eliminate]: 2.98e-06
        [parameter_eliminate]: 1.67999e-06
        [specialize_transform]: 7.16001e-06
        [updatestate_useless_node_eliminater]: 5.96e-06
        [accelerated_algorithm]: 6.83e-06
        [meta_shard_fg_expand]: 2.65997e-06
        [get_grad_eliminate_]: 6.12999e-06
        [merge_forward]: 3.76999e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.532e-05
        [j_node_and_user_rematch]: 1.013e-05
        [meta_fg_expand]: 5.42e-05
        [replace_old_param]: 9.40001e-06
        [inline_without_move]: 6.38e-06
        [renormalize]: 0.00057562
        [add_forward_monad_depend]: 4.26001e-06
        [auto_monad_grad]: 1.71e-06
        [auto_monad_eliminator]: 1.241e-05
        [cse]: 2.369e-05
        [replace_applicator]: 1.346e-05
    [Cycle 3]: 0.00035573, [27]
        [switch_simplify]: 6.28e-06
        [loop_unroll]: 5.84999e-06
        [a_1]: 0.00011632
        [with_stream_mark]: 9.67999e-06
        [recompute_prepare]: 5.59e-06
        [updatestate_depend_eliminate]: 3.37002e-06
        [updatestate_assign_eliminate]: 2.76e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 5.97999e-06
        [updatestate_useless_node_eliminater]: 5.68002e-06
        [accelerated_algorithm]: 5.48002e-06
        [meta_shard_fg_expand]: 1.37999e-06
        [get_grad_eliminate_]: 5.37999e-06
        [merge_forward]: 2.86e-06
        [cell_reuse_recompute_pass]: 1.20999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.407e-05
        [j_node_and_user_rematch]: 9.41003e-06
        [meta_fg_expand]: 1.92999e-06
        [replace_old_param]: 7.78001e-06
        [inline_without_move]: 5.47001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 6.28e-06
        [cse]: 1.369e-05
        [replace_applicator]: 5.56998e-06
[py_interpret_to_execute_after_opt_a]: 1.258e-05
[rewriter_after_opt_a]: 3.856e-05
[convert_after_rewriter]: 7.59002e-06
[order_py_execute_after_rewriter]: 5.69999e-06
[mutable_eliminate]: 0.00063445
[jit_opt_b]: 5.445e-05, [1]
    [Cycle 1]: 4.727e-05, [2]
        [frontend_op_eliminate]: 1.866e-05
        [inline_after_opt_a]: 1.699e-05
[cconv]: 2.394e-05
[loop_unroll]: 0.0004147
[jit_opt_after_cconv]: 0.0001528, [1]
    [Cycle 1]: 0.00014624, [11]
        [c_1]: 2.452e-05
        [parameter_eliminate]: 2.36e-06
        [updatestate_depend_eliminate]: 6.84999e-06
        [updatestate_assign_eliminate]: 2.66999e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [cse]: 2.452e-05
        [call_graph_tuple_transform]: 2.173e-05
        [tuple_list_get_item_eliminator]: 5.87999e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 5.81e-06
[remove_dup_value]: 2.316e-05
[partial_unused_args_eliminate]: 2.62001e-06
[environ_conv]: 5.65001e-06
[add_recomputation]: 5.269e-05
[cse_after_recomputation]: 2.703e-05, [1]
    [Cycle 1]: 2.112e-05, [1]
        [cse]: 1.54e-05
[auto_monad_reorder]: 1.929e-05
[get_jit_bprop_graph]: 1.89e-06
[rewriter_after_jit_bprop_graph]: 5.11002e-06
[opt_after_jit_grad]: 0.00047249
[symbol_engine_optimizer]: 8.064e-05, [1]
    [Cycle 1]: 7.391e-05, [6]
        [build]: 3.78999e-06
        [elim_shapecalc]: 9.37001e-06
        [elim_not_effective]: 1.64e-05
        [opt_reshape]: 6.73e-06
        [fold_const_symbol]: 1.003e-05
        [renormalize]: 4.89992e-07
[validate]: 4.059e-05
[backend_pass]: 1.05001e-06
[task_emit]: 0.00696536
[execute]: 7.35e-06
Sums
    bootstrap                            :     0.000216s :  0.75%
    type_inference                       :     0.011535s : 39.75%
    event_method                         :     0.000045s :  0.15%
    auto_monad                           :     0.000098s :  0.34%
    graph_reusing                        :     0.000008s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000031s :  0.11%
    rewriter_before_opt_a                :     0.000105s :  0.36%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000110s :  0.38%
    jit_opt_a.loop_unroll                :     0.000079s :  0.27%
    jit_opt_a.a_1                        :     0.002067s :  7.12%
    jit_opt_a.with_stream_mark           :     0.000058s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000036s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.02%
    jit_opt_a.specialize_transform       :     0.000027s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000025s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.08%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000059s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000044s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001773s :  6.11%
    jit_opt_a.replace_old_param          :     0.000076s :  0.26%
    jit_opt_a.inline_without_move        :     0.000066s :  0.23%
    jit_opt_a.renormalize                :     0.003131s : 10.79%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000067s :  0.23%
    jit_opt_a.cse                        :     0.000193s :  0.67%
    jit_opt_a.replace_applicator         :     0.000087s :  0.30%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.04%
    rewriter_after_opt_a                 :     0.000039s :  0.13%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000634s :  2.19%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.06%
    cconv                                :     0.000024s :  0.08%
    loop_unroll                          :     0.000415s :  1.43%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000023s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000053s :  0.18%
    cse_after_recomputation.cse          :     0.000015s :  0.05%
    auto_monad_reorder                   :     0.000019s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000472s :  1.63%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000041s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006965s : 24.01%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000517   110
 0.42% :     0.000002s :      3: substitution.elim_not_effective
 0.31% :     0.000002s :      3: substitution.fold_const_symbol
 1.06% :     0.000005s :      4: substitution.graph_param_transform
63.59% :     0.000329s :     12: substitution.inline
 3.59% :     0.000019s :      2: substitution.inline_without_move
 2.03% :     0.000010s :     14: substitution.j_node_and_user_rematch
 2.23% :     0.000011s :      7: substitution.minmaximum_grad
 1.28% :     0.000007s :      5: substitution.partial_eliminate
 2.19% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.23% :     0.000027s :     10: substitution.replace_applicator
 1.54% :     0.000008s :      7: substitution.replace_old_param
 0.69% :     0.000004s :      1: substitution.set_cell_output_no_recompute
 4.32% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.88% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 8.67% :     0.000045s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011450     2
88.15% :     0.010093s :      1: type_inference.infer
11.85% :     0.001357s :      1: type_inference.specialize
------[replace.]   0.000146    19
64.56% :     0.000094s :     12: replace.inline
35.44% :     0.000052s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000346    19
93.20% :     0.000322s :     12: match.inline
 6.80% :     0.000024s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000375  2568
 1.73% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.52% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.43% :     0.000005s :     43: predicate.addn_check_dump
 1.92% :     0.000007s :     43: predicate.addn_zero_filter
 1.95% :     0.000007s :     43: predicate.arithmetic_simplify
 1.61% :     0.000006s :     43: predicate.cast_eliminate
 0.18% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.41% :     0.000005s :     43: predicate.compare_switch_simplify
 1.55% :     0.000006s :     43: predicate.depend_value_elim
 1.69% :     0.000006s :     43: predicate.dict_get_item_const_eliminator
 1.45% :     0.000005s :     43: predicate.dict_get_item_eliminator
 1.45% :     0.000005s :     43: predicate.dict_set_item_eliminator
 0.33% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      4: predicate.elim_not_effective
 0.23% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.44% :     0.000005s :     43: predicate.environ_add_const_eliminate
 1.41% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.47% :     0.000006s :     43: predicate.environ_get_depend_swap
 1.45% :     0.000005s :     43: predicate.environ_get_eliminate
 1.45% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.79% :     0.000003s :     19: predicate.get_grad_eliminate
 0.14% :     0.000001s :      4: predicate.graph_param_transform
 4.44% :     0.000017s :     70: predicate.inline
 1.76% :     0.000007s :     40: predicate.inline_without_move
 0.36% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.92% :     0.000003s :     19: predicate.less_batch_normalization
 1.74% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 2.04% :     0.000008s :     54: predicate.load_eliminater
 0.46% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.31% :     0.000012s :     85: predicate.loop_unroll_before_grad
 1.76% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.46% :     0.000005s :     43: predicate.merge_addn
 1.47% :     0.000006s :     43: predicate.minmaximum_grad
 0.50% :     0.000002s :      4: predicate.mutable_eliminate
 0.18% :     0.000001s :      4: predicate.opt_reshape
 2.29% :     0.000009s :     54: predicate.partial_eliminate
 1.54% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.81% :     0.000007s :     43: predicate.reduce_eliminate
 1.80% :     0.000007s :     50: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.41% :     0.000009s :     96: predicate.replace_applicator
 0.95% :     0.000004s :     40: predicate.replace_old_param
 0.11% :     0.000000s :      4: predicate.reset_defer_inline
 1.82% :     0.000007s :     43: predicate.reshape_eliminate
 1.54% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.28% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.93% :     0.000007s :     43: predicate.same_eliminate
 0.46% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.43% :     0.000002s :      8: predicate.special_op_eliminate
 0.82% :     0.000003s :     19: predicate.specialize_transform
 1.60% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.55% :     0.000006s :     43: predicate.stack_unstack_eliminate
 0.15% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 3.02% :     0.000011s :     62: predicate.switch_defer_inline
 2.58% :     0.000010s :     62: predicate.switch_layer_defer_inline
 6.04% :     0.000023s :    151: predicate.switch_simplify
 1.45% :     0.000005s :     43: predicate.tile_eliminate
 1.49% :     0.000006s :     43: predicate.transpose_eliminate
 1.76% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.63% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.17% :     0.000012s :     58: predicate.tuple_list_get_item_eliminator
 1.87% :     0.000007s :     43: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000006s :     50: predicate.tuple_to_list_eliminator_
 2.02% :     0.000008s :     54: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000011s :     73: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000007s :     43: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.27% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001684    29
55.71% :     0.000938s :     13: func_graph_cloner_run.FuncGraphClonerGraph
44.29% :     0.000746s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.039439    91
 0.14% :     0.000056s :      1: add_recomputation
 0.26% :     0.000102s :      1: auto_monad
 0.06% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.58% :     0.000228s :      1: bootstrap
 0.07% :     0.000026s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000029s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.13% :     0.000050s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
30.98% :     0.012219s :      1: jit_opt_a
 0.39% :     0.000155s :      1: jit_opt_after_cconv
 0.14% :     0.000057s :      1: jit_opt_b
 1.07% :     0.000423s :      1: loop_unroll
 1.63% :     0.000643s :      1: mutable_eliminate
 6.78% :     0.002676s :     39: opt.transform.jit_opt_a
 0.14% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000029s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.22% :     0.000481s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000033s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000026s :      1: remove_dup_value
 4.05% :     0.001595s :      2: renormalize.infer
 3.86% :     0.001521s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000042s :      1: rewriter_after_opt_a
 0.28% :     0.000109s :      1: rewriter_before_opt_a
 0.21% :     0.000083s :      1: symbol_engine_optimizer
17.70% :     0.006979s :      1: task_emit
29.29% :     0.011553s :      1: type_inference
 0.16% :     0.000062s :      1: validate


TotalTime = 0.0183927, [33]
[bootstrap]: 0.0002995
[type_inference]: 0.00448446
[event_method]: 1.75e-05
[auto_monad]: 6.696e-05
[graph_reusing]: 5.52999e-06
[pre_auto_parallel]: 2.36e-06
[py_interpret_to_execute]: 2.735e-05
[rewriter_before_opt_a]: 6.478e-05
[expand_dump_flag]: 3.08e-06
[jit_opt_a]: 0.00518042, [2]
    [Cycle 1]: 0.0013831, [27]
        [switch_simplify]: 4.476e-05
        [loop_unroll]: 2.888e-05
        [a_1]: 0.00047615
        [with_stream_mark]: 1.482e-05
        [recompute_prepare]: 7.43e-06
        [updatestate_depend_eliminate]: 2.99999e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.86003e-06
        [parameter_eliminate]: 1.27999e-06
        [specialize_transform]: 5.72999e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 1.52999e-06
        [get_grad_eliminate_]: 4.97e-06
        [merge_forward]: 2.53003e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.108e-05
        [j_node_and_user_rematch]: 8.28001e-06
        [meta_fg_expand]: 1.88002e-06
        [replace_old_param]: 7.13e-06
        [inline_without_move]: 5.02e-06
        [renormalize]: 0.00053121
        [add_forward_monad_depend]: 5.39e-06
        [auto_monad_grad]: 1.86e-06
        [auto_monad_eliminator]: 1.184e-05
        [cse]: 1.728e-05
        [replace_applicator]: 1.135e-05
    [Cycle 2]: 0.00031867, [27]
        [switch_simplify]: 5.43002e-06
        [loop_unroll]: 4.89e-06
        [a_1]: 8.875e-05
        [with_stream_mark]: 9.60001e-06
        [recompute_prepare]: 4.90001e-06
        [updatestate_depend_eliminate]: 2.69999e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [parameter_eliminate]: 1.11002e-06
        [specialize_transform]: 5.01997e-06
        [updatestate_useless_node_eliminater]: 4.67998e-06
        [accelerated_algorithm]: 4.87e-06
        [meta_shard_fg_expand]: 1.09998e-06
        [get_grad_eliminate_]: 4.39002e-06
        [merge_forward]: 2.32999e-06
        [cell_reuse_recompute_pass]: 2.16e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.233e-05
        [j_node_and_user_rematch]: 8.62998e-06
        [meta_fg_expand]: 1.52999e-06
        [replace_old_param]: 7.75e-06
        [inline_without_move]: 4.44998e-06
        [renormalize]: 1.19995e-07
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 6.50005e-07
        [auto_monad_eliminator]: 5.49e-06
        [cse]: 9.89999e-06
        [replace_applicator]: 9.82001e-06
[py_interpret_to_execute_after_opt_a]: 1.008e-05
[rewriter_after_opt_a]: 3.273e-05
[convert_after_rewriter]: 6.71999e-06
[order_py_execute_after_rewriter]: 5.02999e-06
[mutable_eliminate]: 0.00060604
[jit_opt_b]: 4.803e-05, [1]
    [Cycle 1]: 4.134e-05, [2]
        [frontend_op_eliminate]: 1.5e-05
        [inline_after_opt_a]: 1.431e-05
[cconv]: 2.428e-05
[loop_unroll]: 0.00043208
[jit_opt_after_cconv]: 0.00013613, [1]
    [Cycle 1]: 0.00012939, [11]
        [c_1]: 1.919e-05
        [parameter_eliminate]: 2.46998e-06
        [updatestate_depend_eliminate]: 5.30999e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.76e-05
        [call_graph_tuple_transform]: 1.789e-05
        [tuple_list_get_item_eliminator]: 5.12999e-06
        [none_parameter_eliminate]: 1.63002e-06
        [renormalize]: 7.59988e-07
        [switch_simplify]: 5.39998e-06
[remove_dup_value]: 1.17e-05
[partial_unused_args_eliminate]: 2.18002e-06
[environ_conv]: 5.46e-06
[add_recomputation]: 4.845e-05
[cse_after_recomputation]: 1.937e-05, [1]
    [Cycle 1]: 1.408e-05, [1]
        [cse]: 8.65001e-06
[auto_monad_reorder]: 1.221e-05
[get_jit_bprop_graph]: 2.11e-06
[rewriter_after_jit_bprop_graph]: 5.56e-06
[opt_after_jit_grad]: 0.00047337
[symbol_engine_optimizer]: 7.409e-05, [1]
    [Cycle 1]: 6.708e-05, [6]
        [build]: 3.93999e-06
        [elim_shapecalc]: 7.44002e-06
        [elim_not_effective]: 1.319e-05
        [opt_reshape]: 5.79e-06
        [fold_const_symbol]: 8.62e-06
        [renormalize]: 4.59986e-07
[validate]: 3.557e-05
[backend_pass]: 1.25001e-06
[task_emit]: 0.00604711
[execute]: 8.38999e-06
Sums
    bootstrap                            :     0.000300s :  2.09%
    type_inference                       :     0.004484s : 31.30%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000067s :  0.47%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000027s :  0.19%
    rewriter_before_opt_a                :     0.000065s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000050s :  0.35%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000565s :  3.94%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000531s :  3.71%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.12%
    jit_opt_a.cse                        :     0.000027s :  0.19%
    jit_opt_a.replace_applicator         :     0.000021s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000606s :  4.23%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000432s :  3.02%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000048s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000473s :  3.30%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006047s : 42.21%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000167    23
 1.17% :     0.000002s :      2: substitution.elim_not_effective
 0.89% :     0.000001s :      2: substitution.fold_const_symbol
 3.19% :     0.000005s :      3: substitution.graph_param_transform
82.74% :     0.000138s :      4: substitution.inline
 1.89% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.43% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.17% :     0.000004s :      2: substitution.replace_old_param
 5.52% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004427     2
83.88% :     0.003713s :      1: type_inference.infer
16.12% :     0.000714s :      1: type_inference.specialize
------[replace.]   0.000059     6
65.52% :     0.000038s :      4: replace.inline
34.48% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000144     6
94.33% :     0.000136s :      4: match.inline
 5.67% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.48% :     0.000002s :     10: predicate.addn_zero_filter
 2.17% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.34% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.62% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.12% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.99% :     0.000002s :      3: predicate.mutable_eliminate
 0.68% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000002s :     15: predicate.partial_eliminate
 1.34% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.83% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.42% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.48% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 1.12% :     0.000001s :      6: predicate.specialize_transform
 1.69% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.75% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.28% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000523    12
49.31% :     0.000258s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.11% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.59% :     0.000233s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019792    76
 0.26% :     0.000052s :      1: add_recomputation
 0.35% :     0.000070s :      1: auto_monad
 0.07% :     0.000015s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.57% :     0.000311s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.20% :     0.005185s :      1: jit_opt_a
 0.70% :     0.000139s :      1: jit_opt_after_cconv
 0.26% :     0.000051s :      1: jit_opt_b
 2.22% :     0.000440s :      1: loop_unroll
 3.11% :     0.000616s :      1: mutable_eliminate
 3.86% :     0.000763s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.43% :     0.000481s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000014s :      1: remove_dup_value
 1.44% :     0.000284s :      1: renormalize.infer
 1.22% :     0.000241s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000036s :      1: rewriter_after_opt_a
 0.34% :     0.000067s :      1: rewriter_before_opt_a
 0.39% :     0.000077s :      1: symbol_engine_optimizer
30.64% :     0.006065s :      1: task_emit
22.71% :     0.004495s :      1: type_inference
 0.28% :     0.000056s :      1: validate


TotalTime = 0.0120791, [30]
[bootstrap]: 0.0002674
[type_inference]: 0.00398086
[event_method]: 1.661e-05
[auto_monad]: 6.049e-05
[graph_reusing]: 5.70001e-06
[pre_auto_parallel]: 2.76e-06
[py_interpret_to_execute]: 2.905e-05
[rewriter_before_opt_a]: 7.278e-05
[expand_dump_flag]: 2.84001e-06
[jit_opt_a]: 0.00549612, [2]
    [Cycle 1]: 0.00168282, [27]
        [switch_simplify]: 5.703e-05
        [loop_unroll]: 2.891e-05
        [a_1]: 0.00054456
        [with_stream_mark]: 1.901e-05
        [recompute_prepare]: 7.56001e-06
        [updatestate_depend_eliminate]: 3.72998e-06
        [updatestate_assign_eliminate]: 2.88e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 1.75001e-06
        [specialize_transform]: 6.01998e-06
        [updatestate_useless_node_eliminater]: 5.25999e-06
        [accelerated_algorithm]: 5.47999e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 3.67998e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.475e-05
        [j_node_and_user_rematch]: 9.34998e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 9.72999e-06
        [inline_without_move]: 5.44998e-06
        [renormalize]: 0.00070296
        [add_forward_monad_depend]: 5.84e-06
        [auto_monad_grad]: 2.85002e-06
        [auto_monad_eliminator]: 1.408e-05
        [cse]: 3.084e-05
        [replace_applicator]: 1.136e-05
    [Cycle 2]: 0.00032608, [27]
        [switch_simplify]: 5.30999e-06
        [loop_unroll]: 4.88001e-06
        [a_1]: 8.748e-05
        [with_stream_mark]: 8.67e-06
        [recompute_prepare]: 5.27999e-06
        [updatestate_depend_eliminate]: 2.69999e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 5.22e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.19e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 4.89998e-06
        [merge_forward]: 3.04999e-06
        [cell_reuse_recompute_pass]: 1.84e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.365e-05
        [j_node_and_user_rematch]: 8.62998e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 7.81001e-06
        [inline_without_move]: 4.94e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.27999e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 5.27001e-06
        [cse]: 1.101e-05
        [replace_applicator]: 5.62001e-06
[py_interpret_to_execute_after_opt_a]: 1.056e-05
[rewriter_after_opt_a]: 3.422e-05
[convert_after_rewriter]: 6.68e-06
[order_py_execute_after_rewriter]: 5.83002e-06
[mutable_eliminate]: 0.00060737
[jit_opt_b]: 5.265e-05, [1]
    [Cycle 1]: 4.55e-05, [2]
        [frontend_op_eliminate]: 1.726e-05
        [inline_after_opt_a]: 1.621e-05
[cconv]: 2.558e-05
[loop_unroll]: 0.00041293
[jit_opt_after_cconv]: 0.00013921, [1]
    [Cycle 1]: 0.00013267, [11]
        [c_1]: 2.017e-05
        [parameter_eliminate]: 3.03e-06
        [updatestate_depend_eliminate]: 5.41998e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 1.815e-05
        [call_graph_tuple_transform]: 1.892e-05
        [tuple_list_get_item_eliminator]: 4.97e-06
        [none_parameter_eliminate]: 1.50999e-06
        [renormalize]: 6.89994e-07
        [switch_simplify]: 5.30001e-06
[remove_dup_value]: 1.354e-05
[partial_unused_args_eliminate]: 2.59001e-06
[environ_conv]: 5.00999e-06
[add_recomputation]: 4.783e-05
[cse_after_recomputation]: 2.036e-05, [1]
    [Cycle 1]: 1.451e-05, [1]
        [cse]: 8.80001e-06
[auto_monad_reorder]: 1.544e-05
[get_jit_bprop_graph]: 1.79998e-06
[rewriter_after_jit_bprop_graph]: 5.27999e-06
[opt_after_jit_grad]: 0.00046227
[symbol_engine_optimizer]: 7.534e-05, [1]
    [Cycle 1]: 6.806e-05, [6]
        [build]: 3.40998e-06
        [elim_shapecalc]: 7.65998e-06
        [elim_not_effective]: 1.364e-05
        [opt_reshape]: 5.61e-06
        [fold_const_symbol]: 9.12999e-06
        [renormalize]: 4.90021e-07
[validate]: 3.521e-05
Sums
    bootstrap                            :     0.000267s :  3.34%
    type_inference                       :     0.003981s : 49.70%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000060s :  0.76%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000029s :  0.36%
    rewriter_before_opt_a                :     0.000073s :  0.91%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000062s :  0.78%
    jit_opt_a.loop_unroll                :     0.000034s :  0.42%
    jit_opt_a.a_1                        :     0.000632s :  7.89%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000018s :  0.22%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000703s :  8.78%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.24%
    jit_opt_a.cse                        :     0.000042s :  0.52%
    jit_opt_a.replace_applicator         :     0.000017s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.13%
    rewriter_after_opt_a                 :     0.000034s :  0.43%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.07%
    mutable_eliminate                    :     0.000607s :  7.58%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.20%
    cconv                                :     0.000026s :  0.32%
    loop_unroll                          :     0.000413s :  5.16%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000018s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000048s :  0.60%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000462s :  5.77%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000035s :  0.44%


Time group info:
------[substitution.]   0.000207    23
 0.94% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.39% :     0.000005s :      3: substitution.graph_param_transform
83.96% :     0.000174s :      4: substitution.inline
 1.75% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.30% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.39% :     0.000005s :      2: substitution.replace_old_param
 5.63% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003922     2
80.35% :     0.003151s :      1: type_inference.infer
19.65% :     0.000771s :      1: type_inference.specialize
------[replace.]   0.000066     6
68.87% :     0.000045s :      4: replace.inline
31.13% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000181     6
94.20% :     0.000171s :      4: match.inline
 5.80% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 1.91% :     0.000002s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000002s :     10: predicate.depend_value_elim
 1.20% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.24% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 6.15% :     0.000007s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 2.01% :     0.000002s :      3: predicate.mutable_eliminate
 0.67% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.57% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.43% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.23% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.25% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.76% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.31% :     0.000002s :     10: predicate.transpose_eliminate
 1.56% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.73% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.39% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.86% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000522    12
47.74% :     0.000249s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.88% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.38% :     0.000237s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013687    72
 0.37% :     0.000051s :      1: add_recomputation
 0.47% :     0.000064s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 2.04% :     0.000280s :      1: bootstrap
 0.21% :     0.000029s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000021s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
40.18% :     0.005499s :      1: jit_opt_a
 1.04% :     0.000142s :      1: jit_opt_after_cconv
 0.41% :     0.000055s :      1: jit_opt_b
 3.08% :     0.000421s :      1: loop_unroll
 4.50% :     0.000616s :      1: mutable_eliminate
 6.19% :     0.000847s :     26: opt.transform.jit_opt_a
 0.34% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000027s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.43% :     0.000470s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000032s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.92% :     0.000400s :      1: renormalize.infer
 2.16% :     0.000295s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000037s :      1: rewriter_after_opt_a
 0.55% :     0.000076s :      1: rewriter_before_opt_a
 0.57% :     0.000078s :      1: symbol_engine_optimizer
29.19% :     0.003995s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.697.543 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dcafee0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.698.209 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0194947, [33]
[bootstrap]: 0.00036642
[type_inference]: 0.00469105
[event_method]: 1.813e-05
[auto_monad]: 6.383e-05
[graph_reusing]: 5.74e-06
[pre_auto_parallel]: 2.63e-06
[py_interpret_to_execute]: 3.117e-05
[rewriter_before_opt_a]: 7.599e-05
[expand_dump_flag]: 3.11001e-06
[jit_opt_a]: 0.00591439, [2]
    [Cycle 1]: 0.00169682, [27]
        [switch_simplify]: 5.724e-05
        [loop_unroll]: 2.992e-05
        [a_1]: 0.0005314
        [with_stream_mark]: 2.268e-05
        [recompute_prepare]: 8.70999e-06
        [updatestate_depend_eliminate]: 3.51999e-06
        [updatestate_assign_eliminate]: 2.94001e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 6.01e-06
        [updatestate_useless_node_eliminater]: 5.21998e-06
        [accelerated_algorithm]: 7.13e-06
        [meta_shard_fg_expand]: 2.45002e-06
        [get_grad_eliminate_]: 5.85002e-06
        [merge_forward]: 4.38001e-06
        [cell_reuse_recompute_pass]: 2.48998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.669e-05
        [j_node_and_user_rematch]: 9.69999e-06
        [meta_fg_expand]: 2.53998e-06
        [replace_old_param]: 1.03e-05
        [inline_without_move]: 5.59e-06
        [renormalize]: 0.00070874
        [add_forward_monad_depend]: 6.12001e-06
        [auto_monad_grad]: 2.86999e-06
        [auto_monad_eliminator]: 1.42e-05
        [cse]: 3.019e-05
        [replace_applicator]: 1.28e-05
    [Cycle 2]: 0.00032926, [27]
        [switch_simplify]: 6.16e-06
        [loop_unroll]: 5.05001e-06
        [a_1]: 9.088e-05
        [with_stream_mark]: 1.016e-05
        [recompute_prepare]: 5.22e-06
        [updatestate_depend_eliminate]: 2.61e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 1.21002e-06
        [specialize_transform]: 5.22e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 5.77001e-06
        [meta_shard_fg_expand]: 1.60001e-06
        [get_grad_eliminate_]: 4.84e-06
        [merge_forward]: 2.96001e-06
        [cell_reuse_recompute_pass]: 2.87002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.282e-05
        [j_node_and_user_rematch]: 8.47e-06
        [meta_fg_expand]: 1.44998e-06
        [replace_old_param]: 7.23e-06
        [inline_without_move]: 5.00999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.34998e-06
        [auto_monad_grad]: 1.02e-06
        [auto_monad_eliminator]: 5.46998e-06
        [cse]: 9.60001e-06
        [replace_applicator]: 4.68999e-06
[py_interpret_to_execute_after_opt_a]: 1.002e-05
[rewriter_after_opt_a]: 3.434e-05
[convert_after_rewriter]: 6.76e-06
[order_py_execute_after_rewriter]: 5.35999e-06
[mutable_eliminate]: 0.00057608
[jit_opt_b]: 4.694e-05, [1]
    [Cycle 1]: 4.056e-05, [2]
        [frontend_op_eliminate]: 1.544e-05
        [inline_after_opt_a]: 1.405e-05
[cconv]: 2.65e-05
[loop_unroll]: 0.00041794
[jit_opt_after_cconv]: 0.00013415, [1]
    [Cycle 1]: 0.00012723, [11]
        [c_1]: 1.958e-05
        [parameter_eliminate]: 2.49001e-06
        [updatestate_depend_eliminate]: 5.30999e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.98997e-06
        [cse]: 1.747e-05
        [call_graph_tuple_transform]: 1.758e-05
        [tuple_list_get_item_eliminator]: 4.94e-06
        [none_parameter_eliminate]: 1.71998e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 5.02999e-06
[remove_dup_value]: 1.314e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 5.07e-06
[add_recomputation]: 4.84e-05
[cse_after_recomputation]: 2.005e-05, [1]
    [Cycle 1]: 1.47e-05, [1]
        [cse]: 9.00999e-06
[auto_monad_reorder]: 1.548e-05
[get_jit_bprop_graph]: 1.99e-06
[rewriter_after_jit_bprop_graph]: 5.23002e-06
[opt_after_jit_grad]: 0.00045576
[symbol_engine_optimizer]: 7.208e-05, [1]
    [Cycle 1]: 6.502e-05, [6]
        [build]: 3.78001e-06
        [elim_shapecalc]: 7.78999e-06
        [elim_not_effective]: 1.208e-05
        [opt_reshape]: 5.51e-06
        [fold_const_symbol]: 8.2e-06
        [renormalize]: 6.40022e-07
[validate]: 3.246e-05
[backend_pass]: 1.12999e-06
[task_emit]: 0.0061835
[execute]: 7.69002e-06
Sums
    bootstrap                            :     0.000366s :  2.44%
    type_inference                       :     0.004691s : 31.30%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000064s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.21%
    rewriter_before_opt_a                :     0.000076s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.42%
    jit_opt_a.loop_unroll                :     0.000035s :  0.23%
    jit_opt_a.a_1                        :     0.000622s :  4.15%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000709s :  4.73%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000040s :  0.27%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000576s :  3.84%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000026s :  0.18%
    loop_unroll                          :     0.000418s :  2.79%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000048s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000456s :  3.04%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006183s : 41.26%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000200    23
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.56% :     0.000005s :      3: substitution.graph_param_transform
82.85% :     0.000166s :      4: substitution.inline
 1.90% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.30% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.48% :     0.000005s :      2: substitution.replace_old_param
 6.34% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004616     2
82.44% :     0.003806s :      1: type_inference.infer
17.56% :     0.000811s :      1: type_inference.specialize
------[replace.]   0.000066     6
65.99% :     0.000044s :      4: replace.inline
34.01% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000175     6
93.38% :     0.000163s :      4: match.inline
 6.62% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.88% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.56% :     0.000002s :     10: predicate.addn_zero_filter
 2.17% :     0.000003s :     10: predicate.arithmetic_simplify
 1.15% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000001s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.52% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.38% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 6.17% :     0.000007s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.73% :     0.000002s :      6: predicate.less_batch_normalization
 2.28% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.25% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.64% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.35% :     0.000003s :     15: predicate.partial_eliminate
 1.40% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 1.21% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 1.99% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.36% :     0.000009s :     53: predicate.switch_simplify
 1.50% :     0.000002s :     10: predicate.tile_eliminate
 1.07% :     0.000001s :     10: predicate.transpose_eliminate
 1.27% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.07% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.55% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.94% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000548    12
48.40% :     0.000265s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.51% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.09% :     0.000247s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021146    76
 0.24% :     0.000051s :      1: add_recomputation
 0.32% :     0.000067s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.83% :     0.000387s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000023s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.98% :     0.005917s :      1: jit_opt_a
 0.65% :     0.000137s :      1: jit_opt_after_cconv
 0.23% :     0.000050s :      1: jit_opt_b
 2.01% :     0.000426s :      1: loop_unroll
 2.77% :     0.000585s :      1: mutable_eliminate
 3.98% :     0.000842s :     26: opt.transform.jit_opt_a
 0.21% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.20% :     0.000465s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 1.85% :     0.000392s :      1: renormalize.infer
 1.45% :     0.000307s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000037s :      1: rewriter_after_opt_a
 0.37% :     0.000079s :      1: rewriter_before_opt_a
 0.35% :     0.000075s :      1: symbol_engine_optimizer
29.29% :     0.006195s :      1: task_emit
22.26% :     0.004707s :      1: type_inference
 0.26% :     0.000055s :      1: validate


TotalTime = 0.0128091, [30]
[bootstrap]: 0.00030266
[type_inference]: 0.00435159
[event_method]: 1.688e-05
[auto_monad]: 5.039e-05
[graph_reusing]: 3.76999e-06
[pre_auto_parallel]: 1.82999e-06
[py_interpret_to_execute]: 2.81e-05
[rewriter_before_opt_a]: 6.457e-05
[expand_dump_flag]: 2.46998e-06
[jit_opt_a]: 0.00581293, [2]
    [Cycle 1]: 0.00149454, [27]
        [switch_simplify]: 4.971e-05
        [loop_unroll]: 2.802e-05
        [a_1]: 0.00051259
        [with_stream_mark]: 1.551e-05
        [recompute_prepare]: 7.75e-06
        [updatestate_depend_eliminate]: 2.94999e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 1.91998e-06
        [parameter_eliminate]: 1.38002e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 5.40999e-06
        [accelerated_algorithm]: 5.24e-06
        [meta_shard_fg_expand]: 1.78002e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 2.73998e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.232e-05
        [j_node_and_user_rematch]: 8.35001e-06
        [meta_fg_expand]: 1.84998e-06
        [replace_old_param]: 7.63001e-06
        [inline_without_move]: 5.14998e-06
        [renormalize]: 0.00059748
        [add_forward_monad_depend]: 5.41002e-06
        [auto_monad_grad]: 1.41002e-06
        [auto_monad_eliminator]: 9.86998e-06
        [cse]: 1.65e-05
        [replace_applicator]: 1.263e-05
    [Cycle 2]: 0.00041805, [27]
        [switch_simplify]: 5.90002e-06
        [loop_unroll]: 4.62998e-06
        [a_1]: 0.00011366
        [with_stream_mark]: 1.089e-05
        [recompute_prepare]: 6.11e-06
        [updatestate_depend_eliminate]: 2.64001e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 5.32001e-06
        [updatestate_useless_node_eliminater]: 5.17999e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 1.19e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 2.07001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.332e-05
        [j_node_and_user_rematch]: 7.21999e-06
        [meta_fg_expand]: 1.99e-06
        [replace_old_param]: 7.71999e-06
        [inline_without_move]: 4.92e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.45001e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 6.01e-06
        [cse]: 1.292e-05
        [replace_applicator]: 6.31e-06
[py_interpret_to_execute_after_opt_a]: 1.035e-05
[rewriter_after_opt_a]: 2.992e-05
[convert_after_rewriter]: 5.56998e-06
[order_py_execute_after_rewriter]: 3.88001e-06
[mutable_eliminate]: 0.00059532
[jit_opt_b]: 5.093e-05, [1]
    [Cycle 1]: 4.341e-05, [2]
        [frontend_op_eliminate]: 1.675e-05
        [inline_after_opt_a]: 1.517e-05
[cconv]: 2.46e-05
[loop_unroll]: 0.00042941
[jit_opt_after_cconv]: 0.0001434, [1]
    [Cycle 1]: 0.00013708, [11]
        [c_1]: 2.003e-05
        [parameter_eliminate]: 3.33e-06
        [updatestate_depend_eliminate]: 5.86e-06
        [updatestate_assign_eliminate]: 2.55002e-06
        [updatestate_loads_eliminate]: 2.26998e-06
        [cse]: 2.079e-05
        [call_graph_tuple_transform]: 1.974e-05
        [tuple_list_get_item_eliminator]: 5.69e-06
        [none_parameter_eliminate]: 1.41002e-06
        [renormalize]: 9.20001e-07
        [switch_simplify]: 4.90999e-06
[remove_dup_value]: 1.296e-05
[partial_unused_args_eliminate]: 2.01e-06
[environ_conv]: 5.24998e-06
[add_recomputation]: 4.203e-05
[cse_after_recomputation]: 2.048e-05, [1]
    [Cycle 1]: 1.47e-05, [1]
        [cse]: 8.69e-06
[auto_monad_reorder]: 1.331e-05
[get_jit_bprop_graph]: 1.77001e-06
[rewriter_after_jit_bprop_graph]: 5.96998e-06
[opt_after_jit_grad]: 0.00049896
[symbol_engine_optimizer]: 7.549e-05, [1]
    [Cycle 1]: 6.822e-05, [6]
        [build]: 4.24002e-06
        [elim_shapecalc]: 7.58999e-06
        [elim_not_effective]: 1.396e-05
        [opt_reshape]: 5.43002e-06
        [fold_const_symbol]: 8.09002e-06
        [renormalize]: 7.2e-07
[validate]: 3.63e-05
Sums
    bootstrap                            :     0.000303s :  3.66%
    type_inference                       :     0.004352s : 52.60%
    event_method                         :     0.000017s :  0.20%
    auto_monad                           :     0.000050s :  0.61%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.34%
    rewriter_before_opt_a                :     0.000065s :  0.78%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000056s :  0.67%
    jit_opt_a.loop_unroll                :     0.000033s :  0.39%
    jit_opt_a.a_1                        :     0.000626s :  7.57%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.11%
    jit_opt_a.merge_forward              :     0.000005s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000598s :  7.22%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.19%
    jit_opt_a.cse                        :     0.000029s :  0.36%
    jit_opt_a.replace_applicator         :     0.000019s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.13%
    rewriter_after_opt_a                 :     0.000030s :  0.36%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000595s :  7.20%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000025s :  0.30%
    loop_unroll                          :     0.000429s :  5.19%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000021s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000013s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000042s :  0.51%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000013s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000499s :  6.03%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000036s :  0.44%


Time group info:
------[substitution.]   0.000189    23
 1.24% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.79% :     0.000005s :      3: substitution.graph_param_transform
84.81% :     0.000160s :      4: substitution.inline
 1.26% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.24% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.15% :     0.000004s :      2: substitution.replace_old_param
 4.87% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004293     2
81.43% :     0.003495s :      1: type_inference.infer
18.57% :     0.000797s :      1: type_inference.specialize
------[replace.]   0.000068     6
68.35% :     0.000047s :      4: replace.inline
31.65% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000166     6
95.07% :     0.000158s :      4: match.inline
 4.93% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.31% :     0.000003s :     10: predicate.arithmetic_simplify
 1.15% :     0.000001s :     10: predicate.cast_eliminate
 0.58% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.48% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.74% :     0.000001s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.33% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.43% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.80% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000002s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.48% :     0.000001s :      3: predicate.reset_defer_inline
 1.62% :     0.000002s :     10: predicate.reshape_eliminate
 1.53% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.53% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.22% :     0.000008s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.31% :     0.000001s :     10: predicate.transpose_eliminate
 1.69% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.24% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.88% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.83% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000622    12
54.13% :     0.000337s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.32% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.55% :     0.000252s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014294    72
 0.31% :     0.000045s :      1: add_recomputation
 0.38% :     0.000054s :      1: auto_monad
 0.11% :     0.000016s :      1: auto_monad_reorder
 2.21% :     0.000315s :      1: bootstrap
 0.19% :     0.000027s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.16% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000021s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
40.69% :     0.005817s :      1: jit_opt_a
 1.02% :     0.000146s :      1: jit_opt_after_cconv
 0.37% :     0.000054s :      1: jit_opt_b
 3.06% :     0.000438s :      1: loop_unroll
 4.24% :     0.000606s :      1: mutable_eliminate
 5.80% :     0.000829s :     26: opt.transform.jit_opt_a
 0.33% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.56% :     0.000508s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.21% :     0.000031s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000015s :      1: remove_dup_value
 2.16% :     0.000308s :      1: renormalize.infer
 1.98% :     0.000283s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000033s :      1: rewriter_after_opt_a
 0.47% :     0.000067s :      1: rewriter_before_opt_a
 0.54% :     0.000078s :      1: symbol_engine_optimizer
30.53% :     0.004364s :      1: type_inference


TotalTime = 0.0335159, [33]
[bootstrap]: 0.00026156
[type_inference]: 0.010857
[event_method]: 4.144e-05
[auto_monad]: 7.406e-05
[graph_reusing]: 5.32999e-06
[pre_auto_parallel]: 2.32999e-06
[py_interpret_to_execute]: 2.223e-05
[rewriter_before_opt_a]: 9.008e-05
[expand_dump_flag]: 2.45002e-06
[jit_opt_a]: 0.0118312, [3]
    [Cycle 1]: 0.00597473, [27]
        [switch_simplify]: 6.647e-05
        [loop_unroll]: 5.323e-05
        [a_1]: 0.00084499
        [with_stream_mark]: 2.198e-05
        [recompute_prepare]: 1.792e-05
        [updatestate_depend_eliminate]: 6.99001e-06
        [updatestate_assign_eliminate]: 5.85002e-06
        [updatestate_loads_eliminate]: 5.20001e-06
        [parameter_eliminate]: 2.21e-06
        [specialize_transform]: 1.313e-05
        [updatestate_useless_node_eliminater]: 1.206e-05
        [accelerated_algorithm]: 1.206e-05
        [meta_shard_fg_expand]: 2.94999e-06
        [get_grad_eliminate_]: 1.266e-05
        [merge_forward]: 7.03e-06
        [cell_reuse_recompute_pass]: 7.7e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.331e-05
        [j_node_and_user_rematch]: 2.238e-05
        [meta_fg_expand]: 0.00149048
        [replace_old_param]: 5.212e-05
        [inline_without_move]: 5.068e-05
        [renormalize]: 0.00274323
        [add_forward_monad_depend]: 1.127e-05
        [auto_monad_grad]: 6.13002e-06
        [auto_monad_eliminator]: 5.056e-05
        [cse]: 0.00015182
        [replace_applicator]: 7.403e-05
    [Cycle 2]: 0.00210762, [27]
        [switch_simplify]: 3.549e-05
        [loop_unroll]: 3.406e-05
        [a_1]: 0.00105798
        [with_stream_mark]: 1.469e-05
        [recompute_prepare]: 8.23999e-06
        [updatestate_depend_eliminate]: 4.63999e-06
        [updatestate_assign_eliminate]: 3.55e-06
        [updatestate_loads_eliminate]: 3.18e-06
        [parameter_eliminate]: 1.71998e-06
        [specialize_transform]: 6.83e-06
        [updatestate_useless_node_eliminater]: 5.87001e-06
        [accelerated_algorithm]: 6.48e-06
        [meta_shard_fg_expand]: 3.11001e-06
        [get_grad_eliminate_]: 6.58e-06
        [merge_forward]: 3.95e-06
        [cell_reuse_recompute_pass]: 1.16997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.51e-05
        [j_node_and_user_rematch]: 1.018e-05
        [meta_fg_expand]: 5.863e-05
        [replace_old_param]: 1.017e-05
        [inline_without_move]: 6.36998e-06
        [renormalize]: 0.00060777
        [add_forward_monad_depend]: 4.57e-06
        [auto_monad_grad]: 1.50999e-06
        [auto_monad_eliminator]: 1.245e-05
        [cse]: 2.047e-05
        [replace_applicator]: 1.22e-05
    [Cycle 3]: 0.00039107, [27]
        [switch_simplify]: 6.75002e-06
        [loop_unroll]: 6.09001e-06
        [a_1]: 0.00013689
        [with_stream_mark]: 9.99001e-06
        [recompute_prepare]: 6.28998e-06
        [updatestate_depend_eliminate]: 3.7e-06
        [updatestate_assign_eliminate]: 2.59999e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 1.26002e-06
        [specialize_transform]: 7.01999e-06
        [updatestate_useless_node_eliminater]: 5.61e-06
        [accelerated_algorithm]: 5.54998e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 5.49e-06
        [merge_forward]: 3.25e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.517e-05
        [j_node_and_user_rematch]: 9.86e-06
        [meta_fg_expand]: 2.52001e-06
        [replace_old_param]: 9.37001e-06
        [inline_without_move]: 6.06e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.05001e-06
        [auto_monad_grad]: 1.08001e-06
        [auto_monad_eliminator]: 7.71001e-06
        [cse]: 1.52e-05
        [replace_applicator]: 5.76e-06
[py_interpret_to_execute_after_opt_a]: 1.325e-05
[rewriter_after_opt_a]: 3.788e-05
[convert_after_rewriter]: 9.02e-06
[order_py_execute_after_rewriter]: 5.64e-06
[mutable_eliminate]: 0.00084696
[jit_opt_b]: 5.73e-05, [1]
    [Cycle 1]: 4.887e-05, [2]
        [frontend_op_eliminate]: 1.923e-05
        [inline_after_opt_a]: 1.787e-05
[cconv]: 2.634e-05
[loop_unroll]: 0.00045293
[jit_opt_after_cconv]: 0.00018062, [1]
    [Cycle 1]: 0.0001735, [11]
        [c_1]: 2.527e-05
        [parameter_eliminate]: 3.86001e-06
        [updatestate_depend_eliminate]: 7.35998e-06
        [updatestate_assign_eliminate]: 3.70998e-06
        [updatestate_loads_eliminate]: 2.89001e-06
        [cse]: 2.915e-05
        [call_graph_tuple_transform]: 2.223e-05
        [tuple_list_get_item_eliminator]: 1.888e-05
        [none_parameter_eliminate]: 1.81998e-06
        [renormalize]: 8.00006e-07
        [switch_simplify]: 6.69999e-06
[remove_dup_value]: 1.715e-05
[partial_unused_args_eliminate]: 2.24999e-06
[environ_conv]: 5.59998e-06
[add_recomputation]: 5.515e-05
[cse_after_recomputation]: 2.7e-05, [1]
    [Cycle 1]: 2.021e-05, [1]
        [cse]: 1.38e-05
[auto_monad_reorder]: 1.795e-05
[get_jit_bprop_graph]: 2.11e-06
[rewriter_after_jit_bprop_graph]: 7.2e-06
[opt_after_jit_grad]: 0.00051194
[symbol_engine_optimizer]: 8.523e-05, [1]
    [Cycle 1]: 7.751e-05, [6]
        [build]: 4.22e-06
        [elim_shapecalc]: 1.009e-05
        [elim_not_effective]: 1.556e-05
        [opt_reshape]: 7.05e-06
        [fold_const_symbol]: 1.04e-05
        [renormalize]: 4.59986e-07
[validate]: 4.331e-05
[backend_pass]: 1.02e-06
[task_emit]: 0.00768631
[execute]: 9.01002e-06
Sums
    bootstrap                            :     0.000262s :  0.89%
    type_inference                       :     0.010857s : 37.02%
    event_method                         :     0.000041s :  0.14%
    auto_monad                           :     0.000074s :  0.25%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000022s :  0.08%
    rewriter_before_opt_a                :     0.000090s :  0.31%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000109s :  0.37%
    jit_opt_a.loop_unroll                :     0.000093s :  0.32%
    jit_opt_a.a_1                        :     0.002040s :  6.96%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000027s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.08%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000054s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.001552s :  5.29%
    jit_opt_a.replace_old_param          :     0.000072s :  0.24%
    jit_opt_a.inline_without_move        :     0.000063s :  0.22%
    jit_opt_a.renormalize                :     0.003351s : 11.43%
    jit_opt_a.add_forward_monad_depend   :     0.000017s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000071s :  0.24%
    jit_opt_a.cse                        :     0.000187s :  0.64%
    jit_opt_a.replace_applicator         :     0.000092s :  0.31%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.05%
    rewriter_after_opt_a                 :     0.000038s :  0.13%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000847s :  2.89%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000026s :  0.09%
    loop_unroll                          :     0.000453s :  1.54%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000019s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000017s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000055s :  0.19%
    cse_after_recomputation.cse          :     0.000014s :  0.05%
    auto_monad_reorder                   :     0.000018s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000512s :  1.75%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000043s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007686s : 26.21%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000501   110
 0.52% :     0.000003s :      3: substitution.elim_not_effective
 0.31% :     0.000002s :      3: substitution.fold_const_symbol
 1.17% :     0.000006s :      4: substitution.graph_param_transform
63.42% :     0.000318s :     12: substitution.inline
 3.29% :     0.000016s :      2: substitution.inline_without_move
 1.84% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.25% :     0.000011s :      7: substitution.minmaximum_grad
 1.37% :     0.000007s :      5: substitution.partial_eliminate
 2.32% :     0.000012s :     14: substitution.remove_not_recompute_node
 6.03% :     0.000030s :     10: substitution.replace_applicator
 1.53% :     0.000008s :      7: substitution.replace_old_param
 0.49% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.33% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.08% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 8.05% :     0.000040s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010792     2
89.04% :     0.009609s :      1: type_inference.infer
10.96% :     0.001182s :      1: type_inference.specialize
------[replace.]   0.000144    19
64.14% :     0.000093s :     12: replace.inline
35.86% :     0.000052s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000330    19
94.33% :     0.000311s :     12: match.inline
 5.67% :     0.000019s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000391  2568
 1.56% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.49% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.47% :     0.000006s :     43: predicate.addn_check_dump
 1.66% :     0.000007s :     43: predicate.addn_zero_filter
 2.08% :     0.000008s :     43: predicate.arithmetic_simplify
 1.52% :     0.000006s :     43: predicate.cast_eliminate
 0.17% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.32% :     0.000005s :     43: predicate.compare_switch_simplify
 1.77% :     0.000007s :     43: predicate.depend_value_elim
 1.41% :     0.000006s :     43: predicate.dict_get_item_const_eliminator
 1.40% :     0.000005s :     43: predicate.dict_get_item_eliminator
 1.38% :     0.000005s :     43: predicate.dict_set_item_eliminator
 0.27% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      4: predicate.elim_not_effective
 0.19% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000005s :     43: predicate.environ_add_const_eliminate
 1.35% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.33% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.42% :     0.000006s :     43: predicate.environ_get_eliminate
 1.38% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.92% :     0.000004s :     19: predicate.get_grad_eliminate
 0.13% :     0.000001s :      4: predicate.graph_param_transform
 4.21% :     0.000016s :     70: predicate.inline
 1.71% :     0.000007s :     40: predicate.inline_without_move
 0.33% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.91% :     0.000004s :     19: predicate.less_batch_normalization
 1.77% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.97% :     0.000008s :     54: predicate.load_eliminater
 0.68% :     0.000003s :      4: predicate.loop_unroll_after_grad
 3.13% :     0.000012s :     85: predicate.loop_unroll_before_grad
 1.64% :     0.000006s :     47: predicate.make_slice_get_slice_eliminator
 1.33% :     0.000005s :     43: predicate.merge_addn
 1.41% :     0.000006s :     43: predicate.minmaximum_grad
 0.60% :     0.000002s :      4: predicate.mutable_eliminate
 0.19% :     0.000001s :      4: predicate.opt_reshape
 5.79% :     0.000023s :     54: predicate.partial_eliminate
 1.55% :     0.000006s :     43: predicate.print_const_string_wrapper
 1.98% :     0.000008s :     43: predicate.reduce_eliminate
 1.66% :     0.000006s :     50: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.38% :     0.000009s :     96: predicate.replace_applicator
 0.84% :     0.000003s :     40: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.61% :     0.000006s :     43: predicate.reshape_eliminate
 1.54% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.30% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.50% :     0.000006s :     43: predicate.same_eliminate
 0.44% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.47% :     0.000002s :      8: predicate.special_op_eliminate
 0.82% :     0.000003s :     19: predicate.specialize_transform
 1.57% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000005s :     43: predicate.stack_unstack_eliminate
 0.15% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.74% :     0.000011s :     62: predicate.switch_defer_inline
 2.46% :     0.000010s :     62: predicate.switch_layer_defer_inline
 5.76% :     0.000023s :    151: predicate.switch_simplify
 1.40% :     0.000005s :     43: predicate.tile_eliminate
 1.39% :     0.000005s :     43: predicate.transpose_eliminate
 1.74% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.56% :     0.000006s :     43: predicate.tuple_list_get_item_depend_reorder
 3.06% :     0.000012s :     58: predicate.tuple_list_get_item_eliminator
 2.00% :     0.000008s :     43: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000006s :     50: predicate.tuple_to_list_eliminator_
 1.84% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 3.04% :     0.000012s :     73: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000007s :     43: predicate.value_based_eliminate
 0.14% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.21% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001695    29
59.09% :     0.001002s :     13: func_graph_cloner_run.FuncGraphClonerGraph
40.91% :     0.000693s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.039657    91
 0.15% :     0.000058s :      1: add_recomputation
 0.20% :     0.000078s :      1: auto_monad
 0.05% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.72% :     0.000286s :      1: bootstrap
 0.07% :     0.000029s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.07% :     0.000029s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.12% :     0.000046s :      1: event_method
 0.04% :     0.000014s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
29.85% :     0.011836s :      1: jit_opt_a
 0.46% :     0.000183s :      1: jit_opt_after_cconv
 0.15% :     0.000060s :      1: jit_opt_b
 1.16% :     0.000461s :      1: loop_unroll
 2.16% :     0.000856s :      1: mutable_eliminate
 6.69% :     0.002655s :     39: opt.transform.jit_opt_a
 0.18% :     0.000070s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000030s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000040s :      4: opt.transform.symbol_engine_opt
 1.31% :     0.000521s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.06% :     0.000025s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000020s :      1: remove_dup_value
 4.65% :     0.001844s :      2: renormalize.infer
 3.76% :     0.001491s :      2: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000041s :      1: rewriter_after_opt_a
 0.23% :     0.000093s :      1: rewriter_before_opt_a
 0.22% :     0.000088s :      1: symbol_engine_optimizer
19.43% :     0.007705s :      1: task_emit
27.41% :     0.010871s :      1: type_inference
 0.17% :     0.000068s :      1: validate


TotalTime = 0.0230832, [33]
[bootstrap]: 0.00030451
[type_inference]: 0.00508342
[event_method]: 1.658e-05
[auto_monad]: 6.66e-05
[graph_reusing]: 5.29998e-06
[pre_auto_parallel]: 2.58998e-06
[py_interpret_to_execute]: 3.682e-05
[rewriter_before_opt_a]: 6.981e-05
[expand_dump_flag]: 3.48e-06
[jit_opt_a]: 0.00622545, [2]
    [Cycle 1]: 0.00185456, [27]
        [switch_simplify]: 5.883e-05
        [loop_unroll]: 2.852e-05
        [a_1]: 0.00056703
        [with_stream_mark]: 1.761e-05
        [recompute_prepare]: 7.82e-06
        [updatestate_depend_eliminate]: 3.00002e-06
        [updatestate_assign_eliminate]: 3.22002e-06
        [updatestate_loads_eliminate]: 2.63998e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 6.69999e-06
        [updatestate_useless_node_eliminater]: 5.38002e-06
        [accelerated_algorithm]: 5.77999e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 5.27001e-06
        [merge_forward]: 3.76001e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.366e-05
        [j_node_and_user_rematch]: 9.13002e-06
        [meta_fg_expand]: 2.19001e-06
        [replace_old_param]: 9.16002e-06
        [inline_without_move]: 5.34998e-06
        [renormalize]: 0.00085787
        [add_forward_monad_depend]: 6.56e-06
        [auto_monad_grad]: 2.11998e-06
        [auto_monad_eliminator]: 1.392e-05
        [cse]: 2.714e-05
        [replace_applicator]: 1.497e-05
    [Cycle 2]: 0.00033777, [27]
        [switch_simplify]: 6.06e-06
        [loop_unroll]: 4.63999e-06
        [a_1]: 9.196e-05
        [with_stream_mark]: 1.046e-05
        [recompute_prepare]: 5.60001e-06
        [updatestate_depend_eliminate]: 2.64999e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 1.56998e-06
        [specialize_transform]: 5.02e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 5.84e-06
        [meta_shard_fg_expand]: 1.78002e-06
        [get_grad_eliminate_]: 4.59998e-06
        [merge_forward]: 3.04001e-06
        [cell_reuse_recompute_pass]: 2.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.498e-05
        [j_node_and_user_rematch]: 8.27998e-06
        [meta_fg_expand]: 1.70001e-06
        [replace_old_param]: 8.13999e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 1.10001e-06
        [auto_monad_eliminator]: 4.95999e-06
        [cse]: 1.119e-05
        [replace_applicator]: 5.06997e-06
[py_interpret_to_execute_after_opt_a]: 9.22001e-06
[rewriter_after_opt_a]: 3.602e-05
[convert_after_rewriter]: 7.03e-06
[order_py_execute_after_rewriter]: 4.48999e-06
[mutable_eliminate]: 0.00058497
[jit_opt_b]: 5.806e-05, [1]
    [Cycle 1]: 4.993e-05, [2]
        [frontend_op_eliminate]: 2.257e-05
        [inline_after_opt_a]: 1.581e-05
[cconv]: 2.612e-05
[loop_unroll]: 0.00044881
[jit_opt_after_cconv]: 0.00015222, [1]
    [Cycle 1]: 0.0001449, [11]
        [c_1]: 2.184e-05
        [parameter_eliminate]: 3.2e-06
        [updatestate_depend_eliminate]: 7.05e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 2.56e-05
        [call_graph_tuple_transform]: 1.928e-05
        [tuple_list_get_item_eliminator]: 4.98001e-06
        [none_parameter_eliminate]: 1.76e-06
        [renormalize]: 1.00001e-06
        [switch_simplify]: 5.58002e-06
[remove_dup_value]: 1.335e-05
[partial_unused_args_eliminate]: 2.29999e-06
[environ_conv]: 5.37999e-06
[add_recomputation]: 4.899e-05
[cse_after_recomputation]: 2.132e-05, [1]
    [Cycle 1]: 1.553e-05, [1]
        [cse]: 9.47001e-06
[auto_monad_reorder]: 1.515e-05
[get_jit_bprop_graph]: 2.07999e-06
[rewriter_after_jit_bprop_graph]: 6.06e-06
[opt_after_jit_grad]: 0.00050007
[symbol_engine_optimizer]: 7.464e-05, [1]
    [Cycle 1]: 6.671e-05, [6]
        [build]: 3.25e-06
        [elim_shapecalc]: 7.86001e-06
        [elim_not_effective]: 1.269e-05
        [opt_reshape]: 5.82999e-06
        [fold_const_symbol]: 9.00001e-06
        [renormalize]: 5.10016e-07
[validate]: 4.278e-05
[backend_pass]: 1.35999e-06
[task_emit]: 0.00899837
[execute]: 4.47e-06
Sums
    bootstrap                            :     0.000305s :  1.65%
    type_inference                       :     0.005083s : 27.59%
    event_method                         :     0.000017s :  0.09%
    auto_monad                           :     0.000067s :  0.36%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000037s :  0.20%
    rewriter_before_opt_a                :     0.000070s :  0.38%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.35%
    jit_opt_a.loop_unroll                :     0.000033s :  0.18%
    jit_opt_a.a_1                        :     0.000659s :  3.58%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000858s :  4.66%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.10%
    jit_opt_a.cse                        :     0.000038s :  0.21%
    jit_opt_a.replace_applicator         :     0.000020s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.05%
    rewriter_after_opt_a                 :     0.000036s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000585s :  3.17%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000026s :  0.14%
    loop_unroll                          :     0.000449s :  2.44%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000013s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.27%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000500s :  2.71%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008998s : 48.83%
    execute                              :     0.000004s :  0.02%


Time group info:
------[substitution.]   0.000218    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.60% :     0.000006s :      3: substitution.graph_param_transform
84.05% :     0.000183s :      4: substitution.inline
 1.68% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.19% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.36% :     0.000005s :      2: substitution.replace_old_param
 5.75% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005016     2
80.48% :     0.004037s :      1: type_inference.infer
19.52% :     0.000979s :      1: type_inference.specialize
------[replace.]   0.000073     6
66.41% :     0.000049s :      4: replace.inline
33.59% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     6
94.04% :     0.000181s :      4: match.inline
 5.96% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.77% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.28% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 2.21% :     0.000003s :     10: predicate.arithmetic_simplify
 1.49% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.47% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000001s :     10: predicate.environ_get_eliminate
 1.38% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.75% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000001s :      6: predicate.less_batch_normalization
 2.03% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.30% :     0.000003s :     15: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.92% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.30% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.43% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.88% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.85% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.46% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.39% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.90% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000656    12
47.38% :     0.000311s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.11% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.51% :     0.000312s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024926    76
 0.21% :     0.000052s :      1: add_recomputation
 0.28% :     0.000070s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.29% :     0.000321s :      1: bootstrap
 0.12% :     0.000029s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000020s :      1: event_method
 0.03% :     0.000008s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
24.99% :     0.006229s :      1: jit_opt_a
 0.62% :     0.000155s :      1: jit_opt_after_cconv
 0.24% :     0.000061s :      1: jit_opt_b
 1.84% :     0.000458s :      1: loop_unroll
 2.39% :     0.000595s :      1: mutable_eliminate
 3.52% :     0.000878s :     26: opt.transform.jit_opt_a
 0.19% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000031s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.04% :     0.000508s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000040s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000016s :      1: remove_dup_value
 1.90% :     0.000472s :      1: renormalize.infer
 1.51% :     0.000376s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000039s :      1: rewriter_after_opt_a
 0.29% :     0.000073s :      1: rewriter_before_opt_a
 0.31% :     0.000078s :      1: symbol_engine_optimizer
36.15% :     0.009010s :      1: task_emit
20.44% :     0.005096s :      1: type_inference
 0.26% :     0.000065s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.740.411 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc39c10> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.741.052 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0210121, [33]
[bootstrap]: 0.00049905
[type_inference]: 0.0048322
[event_method]: 1.904e-05
[auto_monad]: 6.254e-05
[graph_reusing]: 6.04999e-06
[pre_auto_parallel]: 3.10998e-06
[py_interpret_to_execute]: 3.086e-05
[rewriter_before_opt_a]: 7.121e-05
[expand_dump_flag]: 3.58e-06
[jit_opt_a]: 0.00607011, [2]
    [Cycle 1]: 0.00159569, [27]
        [switch_simplify]: 5.688e-05
        [loop_unroll]: 3.14e-05
        [a_1]: 0.00053228
        [with_stream_mark]: 1.884e-05
        [recompute_prepare]: 7.87003e-06
        [updatestate_depend_eliminate]: 3.93999e-06
        [updatestate_assign_eliminate]: 2.83998e-06
        [updatestate_loads_eliminate]: 2.52001e-06
        [parameter_eliminate]: 1.82999e-06
        [specialize_transform]: 6.17001e-06
        [updatestate_useless_node_eliminater]: 5.46998e-06
        [accelerated_algorithm]: 5.42999e-06
        [meta_shard_fg_expand]: 2.30002e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 3.55e-06
        [cell_reuse_recompute_pass]: 2.03002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.455e-05
        [j_node_and_user_rematch]: 9.17999e-06
        [meta_fg_expand]: 2.11e-06
        [replace_old_param]: 8.84e-06
        [inline_without_move]: 5.28002e-06
        [renormalize]: 0.00061421
        [add_forward_monad_depend]: 5.89999e-06
        [auto_monad_grad]: 2.93998e-06
        [auto_monad_eliminator]: 1.561e-05
        [cse]: 3.028e-05
        [replace_applicator]: 1.441e-05
    [Cycle 2]: 0.00034087, [27]
        [switch_simplify]: 5.30999e-06
        [loop_unroll]: 4.99998e-06
        [a_1]: 8.964e-05
        [with_stream_mark]: 1.209e-05
        [recompute_prepare]: 5.59e-06
        [updatestate_depend_eliminate]: 3.11001e-06
        [updatestate_assign_eliminate]: 2.14999e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 4.96002e-06
        [updatestate_useless_node_eliminater]: 4.78001e-06
        [accelerated_algorithm]: 5.69999e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 4.62e-06
        [merge_forward]: 2.81e-06
        [cell_reuse_recompute_pass]: 3.10002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.557e-05
        [j_node_and_user_rematch]: 8.28001e-06
        [meta_fg_expand]: 1.53002e-06
        [replace_old_param]: 8e-06
        [inline_without_move]: 4.37e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.81e-06
        [auto_monad_grad]: 1.24998e-06
        [auto_monad_eliminator]: 7.81001e-06
        [cse]: 1.229e-05
        [replace_applicator]: 5.56e-06
[py_interpret_to_execute_after_opt_a]: 1.341e-05
[rewriter_after_opt_a]: 3.66e-05
[convert_after_rewriter]: 6.74001e-06
[order_py_execute_after_rewriter]: 5.67999e-06
[mutable_eliminate]: 0.00072145
[jit_opt_b]: 5.264e-05, [1]
    [Cycle 1]: 4.445e-05, [2]
        [frontend_op_eliminate]: 1.722e-05
        [inline_after_opt_a]: 1.477e-05
[cconv]: 2.794e-05
[loop_unroll]: 0.00050758
[jit_opt_after_cconv]: 0.00015013, [1]
    [Cycle 1]: 0.00014321, [11]
        [c_1]: 1.977e-05
        [parameter_eliminate]: 3.4e-06
        [updatestate_depend_eliminate]: 7.86001e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [cse]: 2.238e-05
        [call_graph_tuple_transform]: 2.017e-05
        [tuple_list_get_item_eliminator]: 5.07999e-06
        [none_parameter_eliminate]: 1.71e-06
        [renormalize]: 8.09989e-07
        [switch_simplify]: 5.26998e-06
[remove_dup_value]: 1.38e-05
[partial_unused_args_eliminate]: 2.34001e-06
[environ_conv]: 6.09001e-06
[add_recomputation]: 5.136e-05
[cse_after_recomputation]: 2.148e-05, [1]
    [Cycle 1]: 1.588e-05, [1]
        [cse]: 9.44998e-06
[auto_monad_reorder]: 1.575e-05
[get_jit_bprop_graph]: 1.95001e-06
[rewriter_after_jit_bprop_graph]: 5.96e-06
[opt_after_jit_grad]: 0.0005665
[symbol_engine_optimizer]: 7.417e-05, [1]
    [Cycle 1]: 6.711e-05, [6]
        [build]: 4.75001e-06
        [elim_shapecalc]: 7.9e-06
        [elim_not_effective]: 1.333e-05
        [opt_reshape]: 5.15001e-06
        [fold_const_symbol]: 8.10999e-06
        [renormalize]: 6.09987e-07
[validate]: 3.636e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00688003
[execute]: 8.18999e-06
Sums
    bootstrap                            :     0.000499s :  3.07%
    type_inference                       :     0.004832s : 29.75%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000063s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.19%
    rewriter_before_opt_a                :     0.000071s :  0.44%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.38%
    jit_opt_a.loop_unroll                :     0.000036s :  0.22%
    jit_opt_a.a_1                        :     0.000622s :  3.83%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000614s :  3.78%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.14%
    jit_opt_a.cse                        :     0.000043s :  0.26%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000037s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000721s :  4.44%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000028s :  0.17%
    loop_unroll                          :     0.000508s :  3.13%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000051s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000567s :  3.49%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006880s : 42.36%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000203    23
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.56% :     0.000005s :      3: substitution.graph_param_transform
83.18% :     0.000169s :      4: substitution.inline
 1.68% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.13% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.51% :     0.000005s :      2: substitution.replace_old_param
 6.42% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004757     2
83.49% :     0.003971s :      1: type_inference.infer
16.51% :     0.000785s :      1: type_inference.specialize
------[replace.]   0.000065     6
66.50% :     0.000043s :      4: replace.inline
33.50% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000178     6
93.45% :     0.000166s :      4: match.inline
 6.55% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.78% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.43% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 1.73% :     0.000002s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.24% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.04% :     0.000001s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 5.56% :     0.000007s :     22: predicate.inline
 1.06% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.45% :     0.000002s :      6: predicate.less_batch_normalization
 1.90% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 1.10% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.28% :     0.000002s :     10: predicate.minmaximum_grad
 2.08% :     0.000003s :      3: predicate.mutable_eliminate
 0.38% :     0.000000s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     15: predicate.partial_eliminate
 1.24% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.60% :     0.000003s :     10: predicate.reduce_eliminate
 2.27% :     0.000003s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.45% :     0.000001s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.07% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.33% :     0.000002s :      6: predicate.special_op_eliminate
 1.14% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.14% :     0.000009s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.30% :     0.000002s :     10: predicate.transpose_eliminate
 1.19% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.04% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000546    12
51.19% :     0.000279s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.84% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.97% :     0.000235s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022581    76
 0.24% :     0.000055s :      1: add_recomputation
 0.29% :     0.000066s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 2.28% :     0.000516s :      1: bootstrap
 0.14% :     0.000031s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.90% :     0.006074s :      1: jit_opt_a
 0.68% :     0.000153s :      1: jit_opt_after_cconv
 0.24% :     0.000055s :      1: jit_opt_b
 2.30% :     0.000519s :      1: loop_unroll
 3.25% :     0.000734s :      1: mutable_eliminate
 3.72% :     0.000840s :     26: opt.transform.jit_opt_a
 0.21% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.55% :     0.000577s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.15% :     0.000033s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.43% :     0.000323s :      1: renormalize.infer
 1.26% :     0.000284s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000039s :      1: rewriter_after_opt_a
 0.33% :     0.000074s :      1: rewriter_before_opt_a
 0.34% :     0.000077s :      1: symbol_engine_optimizer
30.52% :     0.006892s :      1: task_emit
21.46% :     0.004846s :      1: type_inference
 0.26% :     0.000059s :      1: validate


TotalTime = 0.0153636, [30]
[bootstrap]: 0.00032526
[type_inference]: 0.00462804
[event_method]: 1.879e-05
[auto_monad]: 6.512e-05
[graph_reusing]: 5.72999e-06
[pre_auto_parallel]: 2.56e-06
[py_interpret_to_execute]: 3.102e-05
[rewriter_before_opt_a]: 7.169e-05
[expand_dump_flag]: 2.99999e-06
[jit_opt_a]: 0.00660224, [2]
    [Cycle 1]: 0.00166259, [27]
        [switch_simplify]: 5.852e-05
        [loop_unroll]: 2.991e-05
        [a_1]: 0.00052448
        [with_stream_mark]: 1.918e-05
        [recompute_prepare]: 7.61001e-06
        [updatestate_depend_eliminate]: 3.73999e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.83998e-06
        [parameter_eliminate]: 2.24999e-06
        [specialize_transform]: 5.66003e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 2.25002e-06
        [get_grad_eliminate_]: 5.27001e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 1.28002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.469e-05
        [j_node_and_user_rematch]: 9.52999e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 8.92e-06
        [inline_without_move]: 5.05001e-06
        [renormalize]: 0.00070108
        [add_forward_monad_depend]: 5.99999e-06
        [auto_monad_grad]: 2.16998e-06
        [auto_monad_eliminator]: 1.469e-05
        [cse]: 2.79e-05
        [replace_applicator]: 1.334e-05
    [Cycle 2]: 0.00031623, [27]
        [switch_simplify]: 5.66e-06
        [loop_unroll]: 5.37001e-06
        [a_1]: 8.941e-05
        [with_stream_mark]: 9.93998e-06
        [recompute_prepare]: 5.71998e-06
        [updatestate_depend_eliminate]: 2.27999e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 4.74e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 5.12999e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 4.12e-06
        [merge_forward]: 2.81e-06
        [cell_reuse_recompute_pass]: 2.40002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.284e-05
        [j_node_and_user_rematch]: 7.33999e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 8.23999e-06
        [inline_without_move]: 4.99e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.24998e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 5.05999e-06
        [cse]: 9.67999e-06
        [replace_applicator]: 4.80001e-06
[py_interpret_to_execute_after_opt_a]: 1.011e-05
[rewriter_after_opt_a]: 3.639e-05
[convert_after_rewriter]: 6.88e-06
[order_py_execute_after_rewriter]: 4.72998e-06
[mutable_eliminate]: 0.00068345
[jit_opt_b]: 5.558e-05, [1]
    [Cycle 1]: 4.837e-05, [2]
        [frontend_op_eliminate]: 2.036e-05
        [inline_after_opt_a]: 1.608e-05
[cconv]: 3.007e-05
[loop_unroll]: 0.00044693
[jit_opt_after_cconv]: 0.0001463, [1]
    [Cycle 1]: 0.0001395, [11]
        [c_1]: 2.037e-05
        [parameter_eliminate]: 3.07002e-06
        [updatestate_depend_eliminate]: 5.22e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [cse]: 2.168e-05
        [call_graph_tuple_transform]: 2.033e-05
        [tuple_list_get_item_eliminator]: 5.96e-06
        [none_parameter_eliminate]: 1.51002e-06
        [renormalize]: 5.3001e-07
        [switch_simplify]: 5.07999e-06
[remove_dup_value]: 1.477e-05
[partial_unused_args_eliminate]: 2.26998e-06
[environ_conv]: 1.359e-05
[add_recomputation]: 6.357e-05
[cse_after_recomputation]: 4.356e-05, [1]
    [Cycle 1]: 3.393e-05, [1]
        [cse]: 2.097e-05
[auto_monad_reorder]: 1.872e-05
[get_jit_bprop_graph]: 2.21e-06
[rewriter_after_jit_bprop_graph]: 1.071e-05
[opt_after_jit_grad]: 0.00066721
[symbol_engine_optimizer]: 7.813e-05, [1]
    [Cycle 1]: 7.022e-05, [6]
        [build]: 5.20999e-06
        [elim_shapecalc]: 7.73999e-06
        [elim_not_effective]: 1.453e-05
        [opt_reshape]: 6.01e-06
        [fold_const_symbol]: 8.70999e-06
        [renormalize]: 6.59988e-07
[validate]: 4.438e-05
Sums
    bootstrap                            :     0.000325s :  3.58%
    type_inference                       :     0.004628s : 50.93%
    event_method                         :     0.000019s :  0.21%
    auto_monad                           :     0.000065s :  0.72%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000031s :  0.34%
    rewriter_before_opt_a                :     0.000072s :  0.79%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000064s :  0.71%
    jit_opt_a.loop_unroll                :     0.000035s :  0.39%
    jit_opt_a.a_1                        :     0.000614s :  6.76%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000010s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000701s :  7.72%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.22%
    jit_opt_a.cse                        :     0.000038s :  0.41%
    jit_opt_a.replace_applicator         :     0.000018s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.11%
    rewriter_after_opt_a                 :     0.000036s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000683s :  7.52%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.18%
    cconv                                :     0.000030s :  0.33%
    loop_unroll                          :     0.000447s :  4.92%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000022s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000014s :  0.15%
    add_recomputation                    :     0.000064s :  0.70%
    cse_after_recomputation.cse          :     0.000021s :  0.23%
    auto_monad_reorder                   :     0.000019s :  0.21%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.12%
    opt_after_jit_grad                   :     0.000667s :  7.34%
    symbol_engine_optimizer.build        :     0.000005s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.49%


Time group info:
------[substitution.]   0.000200    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000001s :      2: substitution.fold_const_symbol
 2.89% :     0.000006s :      3: substitution.graph_param_transform
83.14% :     0.000167s :      4: substitution.inline
 1.34% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.57% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.46% :     0.000005s :      2: substitution.replace_old_param
 5.82% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004560     2
82.24% :     0.003750s :      1: type_inference.infer
17.76% :     0.000810s :      1: type_inference.specialize
------[replace.]   0.000067     6
68.48% :     0.000046s :      4: replace.inline
31.52% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000175     6
93.93% :     0.000164s :      4: match.inline
 6.07% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.24% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 1.92% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.16% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.04% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000000s :      3: predicate.graph_param_transform
 6.24% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.68% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.72% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.73% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.26% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.81% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.86% :     0.000002s :     18: predicate.replace_applicator
 0.92% :     0.000001s :      6: predicate.replace_old_param
 0.54% :     0.000001s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.40% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.08% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.40% :     0.000002s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.00% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.62% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.07% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.27% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.37% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.38% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000605    12
51.14% :     0.000309s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.85% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.02% :     0.000266s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016944    72
 0.40% :     0.000067s :      1: add_recomputation
 0.41% :     0.000069s :      1: auto_monad
 0.13% :     0.000022s :      1: auto_monad_reorder
 2.04% :     0.000345s :      1: bootstrap
 0.19% :     0.000033s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.27% :     0.000046s :      1: cse_after_recomputation
 0.14% :     0.000024s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.98% :     0.006605s :      1: jit_opt_a
 0.88% :     0.000149s :      1: jit_opt_after_cconv
 0.34% :     0.000058s :      1: jit_opt_b
 2.69% :     0.000456s :      1: loop_unroll
 4.09% :     0.000694s :      1: mutable_eliminate
 4.89% :     0.000829s :     26: opt.transform.jit_opt_a
 0.29% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000030s :      4: opt.transform.jit_opt_b
 0.08% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.99% :     0.000676s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 6.10% :     0.001034s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000033s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.15% :     0.000364s :      1: renormalize.infer
 1.94% :     0.000329s :      1: renormalize.specialize
 0.08% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000040s :      1: rewriter_after_opt_a
 0.44% :     0.000074s :      1: rewriter_before_opt_a
 0.47% :     0.000080s :      1: symbol_engine_optimizer
27.41% :     0.004645s :      1: type_inference


TotalTime = 0.0197231, [33]
[bootstrap]: 0.00029698
[type_inference]: 0.00468896
[event_method]: 1.714e-05
[auto_monad]: 6.489e-05
[graph_reusing]: 5.43002e-06
[pre_auto_parallel]: 2.17001e-06
[py_interpret_to_execute]: 4.247e-05
[rewriter_before_opt_a]: 7.064e-05
[expand_dump_flag]: 3.03998e-06
[jit_opt_a]: 0.00615557, [2]
    [Cycle 1]: 0.00226918, [27]
        [switch_simplify]: 5.885e-05
        [loop_unroll]: 2.878e-05
        [a_1]: 0.00092389
        [with_stream_mark]: 1.958e-05
        [recompute_prepare]: 8.12e-06
        [updatestate_depend_eliminate]: 3.21999e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 2.29999e-06
        [specialize_transform]: 6.31998e-06
        [updatestate_useless_node_eliminater]: 5.22999e-06
        [accelerated_algorithm]: 5.43002e-06
        [meta_shard_fg_expand]: 2.34999e-06
        [get_grad_eliminate_]: 5.20001e-06
        [merge_forward]: 3.47997e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.527e-05
        [j_node_and_user_rematch]: 9.17001e-06
        [meta_fg_expand]: 9.49e-06
        [replace_old_param]: 1.013e-05
        [inline_without_move]: 5.30001e-06
        [renormalize]: 0.0009028
        [add_forward_monad_depend]: 6.03998e-06
        [auto_monad_grad]: 2.39999e-06
        [auto_monad_eliminator]: 1.335e-05
        [cse]: 2.208e-05
        [replace_applicator]: 1.241e-05
    [Cycle 2]: 0.00032008, [27]
        [switch_simplify]: 5.39e-06
        [loop_unroll]: 5.42999e-06
        [a_1]: 8.901e-05
        [with_stream_mark]: 1.09e-05
        [recompute_prepare]: 5.17999e-06
        [updatestate_depend_eliminate]: 2.45002e-06
        [updatestate_assign_eliminate]: 1.91998e-06
        [updatestate_loads_eliminate]: 1.78002e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 4.86002e-06
        [updatestate_useless_node_eliminater]: 4.68001e-06
        [accelerated_algorithm]: 5.20001e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 4.72e-06
        [merge_forward]: 2.83998e-06
        [cell_reuse_recompute_pass]: 1.57999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.423e-05
        [j_node_and_user_rematch]: 7.90998e-06
        [meta_fg_expand]: 1.60001e-06
        [replace_old_param]: 1.225e-05
        [inline_without_move]: 4.48999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 9.09989e-07
        [auto_monad_grad]: 8.59989e-07
        [auto_monad_eliminator]: 4.77e-06
        [cse]: 1.044e-05
        [replace_applicator]: 4.95999e-06
[py_interpret_to_execute_after_opt_a]: 1.231e-05
[rewriter_after_opt_a]: 3.51e-05
[convert_after_rewriter]: 7.35e-06
[order_py_execute_after_rewriter]: 5.10001e-06
[mutable_eliminate]: 0.00069427
[jit_opt_b]: 5.178e-05, [1]
    [Cycle 1]: 4.34e-05, [2]
        [frontend_op_eliminate]: 1.744e-05
        [inline_after_opt_a]: 1.496e-05
[cconv]: 2.637e-05
[loop_unroll]: 0.00042688
[jit_opt_after_cconv]: 0.0001443, [1]
    [Cycle 1]: 0.00013768, [11]
        [c_1]: 2.109e-05
        [parameter_eliminate]: 2.88003e-06
        [updatestate_depend_eliminate]: 5.72999e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.01003e-06
        [cse]: 2.095e-05
        [call_graph_tuple_transform]: 2.086e-05
        [tuple_list_get_item_eliminator]: 5.07e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 8.99978e-07
        [switch_simplify]: 5.22e-06
[remove_dup_value]: 1.406e-05
[partial_unused_args_eliminate]: 2.12001e-06
[environ_conv]: 5.17999e-06
[add_recomputation]: 4.405e-05
[cse_after_recomputation]: 2.07e-05, [1]
    [Cycle 1]: 1.482e-05, [1]
        [cse]: 9.00001e-06
[auto_monad_reorder]: 1.454e-05
[get_jit_bprop_graph]: 1.75001e-06
[rewriter_after_jit_bprop_graph]: 6.01e-06
[opt_after_jit_grad]: 0.00053575
[symbol_engine_optimizer]: 7.082e-05, [1]
    [Cycle 1]: 6.43e-05, [6]
        [build]: 3.06999e-06
        [elim_shapecalc]: 7.95e-06
        [elim_not_effective]: 1.18e-05
        [opt_reshape]: 5.51e-06
        [fold_const_symbol]: 8.20999e-06
        [renormalize]: 1.10999e-06
[validate]: 3.535e-05
[backend_pass]: 8.00006e-07
[task_emit]: 0.00601483
[execute]: 6.78e-06
Sums
    bootstrap                            :     0.000297s :  1.91%
    type_inference                       :     0.004689s : 30.16%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000065s :  0.42%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.27%
    rewriter_before_opt_a                :     0.000071s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.001013s :  6.52%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000011s :  0.07%
    jit_opt_a.replace_old_param          :     0.000022s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000903s :  5.81%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000033s :  0.21%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000035s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000694s :  4.47%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000026s :  0.17%
    loop_unroll                          :     0.000427s :  2.75%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000044s :  0.28%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000536s :  3.45%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000035s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006015s : 38.69%
    execute                              :     0.000007s :  0.04%


Time group info:
------[substitution.]   0.000561    23
 0.32% :     0.000002s :      2: substitution.elim_not_effective
 0.21% :     0.000001s :      2: substitution.fold_const_symbol
 0.98% :     0.000005s :      3: substitution.graph_param_transform
92.72% :     0.000520s :      4: substitution.inline
 0.55% :     0.000003s :      4: substitution.j_node_and_user_rematch
 0.96% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.81% :     0.000010s :      2: substitution.replace_old_param
 2.45% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004624     2
80.44% :     0.003719s :      1: type_inference.infer
19.56% :     0.000905s :      1: type_inference.specialize
------[replace.]   0.000091     6
67.27% :     0.000061s :      4: replace.inline
32.73% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000529     6
97.63% :     0.000517s :      4: match.inline
 2.37% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 2.12% :     0.000002s :     10: predicate.arithmetic_simplify
 1.48% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.60% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.53% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.01% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.95% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.49% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000000s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.36% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.84% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.52% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.60% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.35% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.16% :     0.000008s :     53: predicate.switch_simplify
 1.45% :     0.000002s :     10: predicate.tile_eliminate
 1.35% :     0.000002s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.31% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 2.09% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.13% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000667    12
45.86% :     0.000306s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.94% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.20% :     0.000308s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021959    76
 0.21% :     0.000047s :      1: add_recomputation
 0.31% :     0.000068s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.42% :     0.000311s :      1: bootstrap
 0.13% :     0.000029s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000020s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
28.05% :     0.006160s :      1: jit_opt_a
 0.67% :     0.000148s :      1: jit_opt_after_cconv
 0.25% :     0.000054s :      1: jit_opt_b
 1.98% :     0.000435s :      1: loop_unroll
 3.20% :     0.000703s :      1: mutable_eliminate
 5.62% :     0.001235s :     26: opt.transform.jit_opt_a
 0.22% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.48% :     0.000544s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.21% :     0.000045s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.24% :     0.000493s :      1: renormalize.infer
 1.82% :     0.000400s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000038s :      1: rewriter_after_opt_a
 0.34% :     0.000074s :      1: rewriter_before_opt_a
 0.33% :     0.000073s :      1: symbol_engine_optimizer
27.46% :     0.006031s :      1: task_emit
21.41% :     0.004702s :      1: type_inference
 0.25% :     0.000054s :      1: validate


TotalTime = 0.0358999, [33]
[bootstrap]: 0.00024392
[type_inference]: 0.0118747
[event_method]: 6.191e-05
[auto_monad]: 0.00010373
[graph_reusing]: 6.98e-06
[pre_auto_parallel]: 3.08e-06
[py_interpret_to_execute]: 3.076e-05
[rewriter_before_opt_a]: 0.00010184
[expand_dump_flag]: 3.09001e-06
[jit_opt_a]: 0.013367, [3]
    [Cycle 1]: 0.00654852, [27]
        [switch_simplify]: 7.514e-05
        [loop_unroll]: 3.842e-05
        [a_1]: 0.00088115
        [with_stream_mark]: 3.681e-05
        [recompute_prepare]: 2.376e-05
        [updatestate_depend_eliminate]: 9.41e-06
        [updatestate_assign_eliminate]: 6.78998e-06
        [updatestate_loads_eliminate]: 6.40002e-06
        [parameter_eliminate]: 2.69999e-06
        [specialize_transform]: 1.367e-05
        [updatestate_useless_node_eliminater]: 1.213e-05
        [accelerated_algorithm]: 1.343e-05
        [meta_shard_fg_expand]: 4.37998e-06
        [get_grad_eliminate_]: 1.264e-05
        [merge_forward]: 8.57e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.787e-05
        [j_node_and_user_rematch]: 2.465e-05
        [meta_fg_expand]: 0.00180237
        [replace_old_param]: 6.354e-05
        [inline_without_move]: 5.443e-05
        [renormalize]: 0.00287683
        [add_forward_monad_depend]: 1.191e-05
        [auto_monad_grad]: 6.43003e-06
        [auto_monad_eliminator]: 5.233e-05
        [cse]: 0.00015492
        [replace_applicator]: 7.551e-05
    [Cycle 2]: 0.0024327, [27]
        [switch_simplify]: 3.555e-05
        [loop_unroll]: 3.385e-05
        [a_1]: 0.00117446
        [with_stream_mark]: 2.21e-05
        [recompute_prepare]: 1.091e-05
        [updatestate_depend_eliminate]: 4.42e-06
        [updatestate_assign_eliminate]: 3.93001e-06
        [updatestate_loads_eliminate]: 3.41999e-06
        [parameter_eliminate]: 2.11998e-06
        [specialize_transform]: 6.84001e-06
        [updatestate_useless_node_eliminater]: 6.96001e-06
        [accelerated_algorithm]: 7.46999e-06
        [meta_shard_fg_expand]: 3.08998e-06
        [get_grad_eliminate_]: 6.94001e-06
        [merge_forward]: 4.30999e-06
        [cell_reuse_recompute_pass]: 1.33002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.913e-05
        [j_node_and_user_rematch]: 1.003e-05
        [meta_fg_expand]: 6.819e-05
        [replace_old_param]: 1.577e-05
        [inline_without_move]: 7.33999e-06
        [renormalize]: 0.00073081
        [add_forward_monad_depend]: 5.94999e-06
        [auto_monad_grad]: 1.99999e-06
        [auto_monad_eliminator]: 1.607e-05
        [cse]: 2.666e-05
        [replace_applicator]: 1.607e-05
    [Cycle 3]: 0.00039264, [27]
        [switch_simplify]: 7.45998e-06
        [loop_unroll]: 6.11e-06
        [a_1]: 0.0001222
        [with_stream_mark]: 1.01e-05
        [recompute_prepare]: 6.29999e-06
        [updatestate_depend_eliminate]: 3.6e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.62001e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 7.16001e-06
        [updatestate_useless_node_eliminater]: 6.32001e-06
        [accelerated_algorithm]: 6.34999e-06
        [meta_shard_fg_expand]: 1.69e-06
        [get_grad_eliminate_]: 6.32001e-06
        [merge_forward]: 3.63e-06
        [cell_reuse_recompute_pass]: 2.71e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.552e-05
        [j_node_and_user_rematch]: 1.059e-05
        [meta_fg_expand]: 2.07001e-06
        [replace_old_param]: 1.102e-05
        [inline_without_move]: 5.68002e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 1.25999e-06
        [auto_monad_eliminator]: 8.29998e-06
        [cse]: 1.624e-05
        [replace_applicator]: 6.76e-06
[py_interpret_to_execute_after_opt_a]: 1.43e-05
[rewriter_after_opt_a]: 3.972e-05
[convert_after_rewriter]: 7.88999e-06
[order_py_execute_after_rewriter]: 5.72999e-06
[mutable_eliminate]: 0.00075339
[jit_opt_b]: 5.93e-05, [1]
    [Cycle 1]: 5.081e-05, [2]
        [frontend_op_eliminate]: 2.086e-05
        [inline_after_opt_a]: 1.851e-05
[cconv]: 2.512e-05
[loop_unroll]: 0.00048465
[jit_opt_after_cconv]: 0.00017592, [1]
    [Cycle 1]: 0.00016802, [11]
        [c_1]: 2.687e-05
        [parameter_eliminate]: 3.58e-06
        [updatestate_depend_eliminate]: 7.1e-06
        [updatestate_assign_eliminate]: 3.26001e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [cse]: 2.614e-05
        [call_graph_tuple_transform]: 2.315e-05
        [tuple_list_get_item_eliminator]: 7.04001e-06
        [none_parameter_eliminate]: 1.50999e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 6.89001e-06
[remove_dup_value]: 1.691e-05
[partial_unused_args_eliminate]: 2.54001e-06
[environ_conv]: 5.87001e-06
[add_recomputation]: 5.432e-05
[cse_after_recomputation]: 2.761e-05, [1]
    [Cycle 1]: 2.108e-05, [1]
        [cse]: 1.465e-05
[auto_monad_reorder]: 1.955e-05
[get_jit_bprop_graph]: 1.81e-06
[rewriter_after_jit_bprop_graph]: 6.42001e-06
[opt_after_jit_grad]: 0.00058499
[symbol_engine_optimizer]: 8.636e-05, [1]
    [Cycle 1]: 7.885e-05, [6]
        [build]: 4.68999e-06
        [elim_shapecalc]: 9.97999e-06
        [elim_not_effective]: 1.666e-05
        [opt_reshape]: 6.85002e-06
        [fold_const_symbol]: 1.065e-05
        [renormalize]: 5.00004e-07
[validate]: 3.965e-05
[backend_pass]: 1.07e-06
[task_emit]: 0.00745888
[execute]: 7.25e-06
Sums
    bootstrap                            :     0.000244s :  0.79%
    type_inference                       :     0.011875s : 38.32%
    event_method                         :     0.000062s :  0.20%
    auto_monad                           :     0.000104s :  0.33%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000031s :  0.10%
    rewriter_before_opt_a                :     0.000102s :  0.33%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000118s :  0.38%
    jit_opt_a.loop_unroll                :     0.000078s :  0.25%
    jit_opt_a.a_1                        :     0.002178s :  7.03%
    jit_opt_a.with_stream_mark           :     0.000069s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000041s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.02%
    jit_opt_a.specialize_transform       :     0.000028s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000027s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000009s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000026s :  0.08%
    jit_opt_a.merge_forward              :     0.000017s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000063s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000045s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001873s :  6.04%
    jit_opt_a.replace_old_param          :     0.000090s :  0.29%
    jit_opt_a.inline_without_move        :     0.000067s :  0.22%
    jit_opt_a.renormalize                :     0.003608s : 11.64%
    jit_opt_a.add_forward_monad_depend   :     0.000019s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000010s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000077s :  0.25%
    jit_opt_a.cse                        :     0.000198s :  0.64%
    jit_opt_a.replace_applicator         :     0.000098s :  0.32%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.05%
    rewriter_after_opt_a                 :     0.000040s :  0.13%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000753s :  2.43%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.06%
    cconv                                :     0.000025s :  0.08%
    loop_unroll                          :     0.000485s :  1.56%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000017s :  0.05%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000054s :  0.18%
    cse_after_recomputation.cse          :     0.000015s :  0.05%
    auto_monad_reorder                   :     0.000020s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000585s :  1.89%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007459s : 24.07%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000590   110
 0.50% :     0.000003s :      3: substitution.elim_not_effective
 0.31% :     0.000002s :      3: substitution.fold_const_symbol
 1.13% :     0.000007s :      4: substitution.graph_param_transform
65.92% :     0.000389s :     12: substitution.inline
 3.28% :     0.000019s :      2: substitution.inline_without_move
 1.69% :     0.000010s :     14: substitution.j_node_and_user_rematch
 2.19% :     0.000013s :      7: substitution.minmaximum_grad
 1.15% :     0.000007s :      5: substitution.partial_eliminate
 1.92% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.06% :     0.000030s :     10: substitution.replace_applicator
 1.80% :     0.000011s :      7: substitution.replace_old_param
 0.55% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.08% :     0.000024s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.65% :     0.000016s :      7: substitution.tuple_list_get_item_depend_reorder
 7.77% :     0.000046s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011760     2
88.58% :     0.010416s :      1: type_inference.infer
11.42% :     0.001343s :      1: type_inference.specialize
------[replace.]   0.000166    19
63.21% :     0.000105s :     12: replace.inline
36.79% :     0.000061s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000405    19
94.16% :     0.000381s :     12: match.inline
 5.84% :     0.000024s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000389  2568
 1.54% :     0.000006s :     43: predicate.accumulaten_eliminater
 0.51% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.32% :     0.000005s :     43: predicate.addn_check_dump
 1.72% :     0.000007s :     43: predicate.addn_zero_filter
 2.18% :     0.000008s :     43: predicate.arithmetic_simplify
 1.61% :     0.000006s :     43: predicate.cast_eliminate
 0.16% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.32% :     0.000005s :     43: predicate.compare_switch_simplify
 1.49% :     0.000006s :     43: predicate.depend_value_elim
 1.34% :     0.000005s :     43: predicate.dict_get_item_const_eliminator
 1.46% :     0.000006s :     43: predicate.dict_get_item_eliminator
 1.40% :     0.000005s :     43: predicate.dict_set_item_eliminator
 0.29% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      4: predicate.elim_not_effective
 0.20% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000006s :     43: predicate.environ_add_const_eliminate
 1.33% :     0.000005s :     43: predicate.environ_get_add_eliminate
 1.37% :     0.000005s :     43: predicate.environ_get_depend_swap
 1.43% :     0.000006s :     43: predicate.environ_get_eliminate
 1.33% :     0.000005s :     43: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.96% :     0.000004s :     19: predicate.get_grad_eliminate
 0.14% :     0.000001s :      4: predicate.graph_param_transform
 4.48% :     0.000017s :     70: predicate.inline
 1.99% :     0.000008s :     40: predicate.inline_without_move
 0.35% :     0.000001s :     19: predicate.j_node_and_user_rematch
 1.06% :     0.000004s :     19: predicate.less_batch_normalization
 1.72% :     0.000007s :     50: predicate.list_to_tuple_eliminator_
 1.84% :     0.000007s :     54: predicate.load_eliminater
 0.45% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.13% :     0.000012s :     85: predicate.loop_unroll_before_grad
 1.72% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.41% :     0.000005s :     43: predicate.merge_addn
 1.39% :     0.000005s :     43: predicate.minmaximum_grad
 0.63% :     0.000002s :      4: predicate.mutable_eliminate
 0.20% :     0.000001s :      4: predicate.opt_reshape
 2.20% :     0.000009s :     54: predicate.partial_eliminate
 2.07% :     0.000008s :     43: predicate.print_const_string_wrapper
 2.04% :     0.000008s :     43: predicate.reduce_eliminate
 1.64% :     0.000006s :     50: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.73% :     0.000011s :     96: predicate.replace_applicator
 1.16% :     0.000005s :     40: predicate.replace_old_param
 0.16% :     0.000001s :      4: predicate.reset_defer_inline
 1.93% :     0.000008s :     43: predicate.reshape_eliminate
 1.42% :     0.000006s :     43: predicate.row_tensor_add_zeros_like
 0.41% :     0.000002s :      4: predicate.row_tensor_eliminate
 1.65% :     0.000006s :     43: predicate.same_eliminate
 0.44% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.37% :     0.000001s :      8: predicate.special_op_eliminate
 0.97% :     0.000004s :     19: predicate.specialize_transform
 1.51% :     0.000006s :     43: predicate.split_environ_get_set_with_tuple_value
 1.48% :     0.000006s :     43: predicate.stack_unstack_eliminate
 0.16% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.95% :     0.000011s :     62: predicate.switch_defer_inline
 2.40% :     0.000009s :     62: predicate.switch_layer_defer_inline
 6.54% :     0.000025s :    151: predicate.switch_simplify
 1.37% :     0.000005s :     43: predicate.tile_eliminate
 1.38% :     0.000005s :     43: predicate.transpose_eliminate
 1.87% :     0.000007s :     43: predicate.tuple_list_convert_item_index_to_positive
 1.93% :     0.000007s :     43: predicate.tuple_list_get_item_depend_reorder
 3.08% :     0.000012s :     58: predicate.tuple_list_get_item_eliminator
 1.89% :     0.000007s :     43: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000006s :     50: predicate.tuple_to_list_eliminator_
 2.15% :     0.000008s :     54: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000011s :     73: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000007s :     43: predicate.value_based_eliminate
 0.14% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.22% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001832    29
58.81% :     0.001077s :     13: func_graph_cloner_run.FuncGraphClonerGraph
41.19% :     0.000755s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042479    91
 0.14% :     0.000058s :      1: add_recomputation
 0.25% :     0.000108s :      1: auto_monad
 0.05% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.61% :     0.000261s :      1: bootstrap
 0.07% :     0.000028s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000030s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.16% :     0.000068s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
31.48% :     0.013372s :      1: jit_opt_a
 0.42% :     0.000179s :      1: jit_opt_after_cconv
 0.15% :     0.000062s :      1: jit_opt_b
 1.17% :     0.000495s :      1: loop_unroll
 1.80% :     0.000763s :      1: mutable_eliminate
 6.67% :     0.002832s :     39: opt.transform.jit_opt_a
 0.14% :     0.000060s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000033s :      4: opt.transform.jit_opt_b
 0.04% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000041s :      4: opt.transform.symbol_engine_opt
 1.40% :     0.000594s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000034s :      1: py_interpret_to_execute
 0.04% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000020s :      1: remove_dup_value
 4.47% :     0.001897s :      2: renormalize.infer
 3.98% :     0.001692s :      2: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000044s :      1: rewriter_after_opt_a
 0.25% :     0.000106s :      1: rewriter_before_opt_a
 0.21% :     0.000089s :      1: symbol_engine_optimizer
17.59% :     0.007472s :      1: task_emit
28.00% :     0.011896s :      1: type_inference
 0.15% :     0.000063s :      1: validate


TotalTime = 0.0201106, [33]
[bootstrap]: 0.00032264
[type_inference]: 0.00511879
[event_method]: 1.736e-05
[auto_monad]: 6.234e-05
[graph_reusing]: 6.24001e-06
[pre_auto_parallel]: 2.26998e-06
[py_interpret_to_execute]: 3.067e-05
[rewriter_before_opt_a]: 6.795e-05
[expand_dump_flag]: 3.08998e-06
[jit_opt_a]: 0.00569682, [2]
    [Cycle 1]: 0.00165001, [27]
        [switch_simplify]: 5.613e-05
        [loop_unroll]: 3.112e-05
        [a_1]: 0.00055363
        [with_stream_mark]: 2.077e-05
        [recompute_prepare]: 8.25e-06
        [updatestate_depend_eliminate]: 3.36999e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.95001e-06
        [specialize_transform]: 6.78e-06
        [updatestate_useless_node_eliminater]: 5.56002e-06
        [accelerated_algorithm]: 5.76998e-06
        [meta_shard_fg_expand]: 2.12999e-06
        [get_grad_eliminate_]: 5.57999e-06
        [merge_forward]: 4.17003e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.239e-05
        [j_node_and_user_rematch]: 8.64e-06
        [meta_fg_expand]: 2.25002e-06
        [replace_old_param]: 7.65e-06
        [inline_without_move]: 5.00999e-06
        [renormalize]: 0.00066422
        [add_forward_monad_depend]: 6.33998e-06
        [auto_monad_grad]: 2.96001e-06
        [auto_monad_eliminator]: 1.428e-05
        [cse]: 1.859e-05
        [replace_applicator]: 1.474e-05
    [Cycle 2]: 0.0003452, [27]
        [switch_simplify]: 6.14001e-06
        [loop_unroll]: 5.37001e-06
        [a_1]: 9.763e-05
        [with_stream_mark]: 1.109e-05
        [recompute_prepare]: 5.26002e-06
        [updatestate_depend_eliminate]: 2.68998e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.04998e-06
        [specialize_transform]: 5.23002e-06
        [updatestate_useless_node_eliminater]: 5.39e-06
        [accelerated_algorithm]: 5.05001e-06
        [meta_shard_fg_expand]: 1.54e-06
        [get_grad_eliminate_]: 5.24e-06
        [merge_forward]: 2.91999e-06
        [cell_reuse_recompute_pass]: 3.55998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.39e-05
        [j_node_and_user_rematch]: 8.39998e-06
        [meta_fg_expand]: 1.65001e-06
        [replace_old_param]: 1.248e-05
        [inline_without_move]: 5.00001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.67999e-06
        [auto_monad_grad]: 8.80013e-07
        [auto_monad_eliminator]: 5.66e-06
        [cse]: 1.081e-05
        [replace_applicator]: 5.31002e-06
[py_interpret_to_execute_after_opt_a]: 1.1e-05
[rewriter_after_opt_a]: 3.73e-05
[convert_after_rewriter]: 6.93e-06
[order_py_execute_after_rewriter]: 4.72e-06
[mutable_eliminate]: 0.00069506
[jit_opt_b]: 4.798e-05, [1]
    [Cycle 1]: 4.07e-05, [2]
        [frontend_op_eliminate]: 1.497e-05
        [inline_after_opt_a]: 1.457e-05
[cconv]: 2.535e-05
[loop_unroll]: 0.00044653
[jit_opt_after_cconv]: 0.0001444, [1]
    [Cycle 1]: 0.00013766, [11]
        [c_1]: 2.008e-05
        [parameter_eliminate]: 3.4e-06
        [updatestate_depend_eliminate]: 6.17999e-06
        [updatestate_assign_eliminate]: 2.84999e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.963e-05
        [call_graph_tuple_transform]: 1.951e-05
        [tuple_list_get_item_eliminator]: 5.57001e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 4.7998e-07
        [switch_simplify]: 5.00001e-06
[remove_dup_value]: 1.31e-05
[partial_unused_args_eliminate]: 2.46e-06
[environ_conv]: 5.24e-06
[add_recomputation]: 4.629e-05
[cse_after_recomputation]: 1.972e-05, [1]
    [Cycle 1]: 1.446e-05, [1]
        [cse]: 8.22998e-06
[auto_monad_reorder]: 1.494e-05
[get_jit_bprop_graph]: 2.29999e-06
[rewriter_after_jit_bprop_graph]: 6.20002e-06
[opt_after_jit_grad]: 0.00055426
[symbol_engine_optimizer]: 7.234e-05, [1]
    [Cycle 1]: 6.58e-05, [6]
        [build]: 3.46999e-06
        [elim_shapecalc]: 7.54002e-06
        [elim_not_effective]: 1.286e-05
        [opt_reshape]: 5.51e-06
        [fold_const_symbol]: 9.01002e-06
        [renormalize]: 2.9002e-07
[validate]: 3.457e-05
[backend_pass]: 1.14e-06
[task_emit]: 0.00638621
[execute]: 9.45001e-06
Sums
    bootstrap                            :     0.000323s :  2.04%
    type_inference                       :     0.005119s : 32.41%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000062s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000031s :  0.19%
    rewriter_before_opt_a                :     0.000068s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.39%
    jit_opt_a.loop_unroll                :     0.000036s :  0.23%
    jit_opt_a.a_1                        :     0.000651s :  4.12%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.13%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000664s :  4.21%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000029s :  0.19%
    jit_opt_a.replace_applicator         :     0.000020s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000695s :  4.40%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000025s :  0.16%
    loop_unroll                          :     0.000447s :  2.83%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000046s :  0.29%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000554s :  3.51%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006386s : 40.44%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000216    23
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000002s :      2: substitution.fold_const_symbol
 2.41% :     0.000005s :      3: substitution.graph_param_transform
82.67% :     0.000179s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.01% :     0.000004s :      4: substitution.remove_not_recompute_node
 3.75% :     0.000008s :      2: substitution.replace_old_param
 5.93% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005053     2
83.72% :     0.004230s :      1: type_inference.infer
16.28% :     0.000823s :      1: type_inference.specialize
------[replace.]   0.000066     6
66.78% :     0.000044s :      4: replace.inline
33.22% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000188     6
93.79% :     0.000176s :      4: match.inline
 6.21% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.79% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.56% :     0.000002s :     10: predicate.addn_zero_filter
 2.17% :     0.000003s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.32% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.45% :     0.000001s :      3: predicate.elim_not_effective
 0.42% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.55% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.05% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.67% :     0.000007s :     22: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.89% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.96% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.78% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.15% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 1.37% :     0.000002s :      3: predicate.mutable_eliminate
 0.39% :     0.000000s :      3: predicate.opt_reshape
 1.96% :     0.000002s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.79% :     0.000003s :     10: predicate.reduce_eliminate
 1.75% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.42% :     0.000001s :      3: predicate.reset_defer_inline
 1.82% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.44% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.51% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.33% :     0.000009s :     53: predicate.switch_simplify
 1.23% :     0.000002s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.24% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.21% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.82% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.86% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000559    12
48.22% :     0.000270s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.19% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.58% :     0.000249s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021753    76
 0.23% :     0.000049s :      1: add_recomputation
 0.30% :     0.000065s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.54% :     0.000335s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.21% :     0.005700s :      1: jit_opt_a
 0.68% :     0.000147s :      1: jit_opt_after_cconv
 0.23% :     0.000051s :      1: jit_opt_b
 2.09% :     0.000455s :      1: loop_unroll
 3.24% :     0.000705s :      1: mutable_eliminate
 4.02% :     0.000873s :     26: opt.transform.jit_opt_a
 0.22% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000023s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.59% :     0.000562s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 1.70% :     0.000371s :      1: renormalize.infer
 1.31% :     0.000284s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000040s :      1: rewriter_after_opt_a
 0.33% :     0.000071s :      1: rewriter_before_opt_a
 0.34% :     0.000075s :      1: symbol_engine_optimizer
29.44% :     0.006403s :      1: task_emit
23.59% :     0.005132s :      1: type_inference
 0.25% :     0.000054s :      1: validate


TotalTime = 0.0123294, [30]
[bootstrap]: 0.00032146
[type_inference]: 0.00427888
[event_method]: 1.758e-05
[auto_monad]: 6.359e-05
[graph_reusing]: 5.74e-06
[pre_auto_parallel]: 2.58e-06
[py_interpret_to_execute]: 3.309e-05
[rewriter_before_opt_a]: 7.227e-05
[expand_dump_flag]: 2.81999e-06
[jit_opt_a]: 0.00532354, [2]
    [Cycle 1]: 0.00157167, [27]
        [switch_simplify]: 5.841e-05
        [loop_unroll]: 3.138e-05
        [a_1]: 0.00051625
        [with_stream_mark]: 1.822e-05
        [recompute_prepare]: 7.29001e-06
        [updatestate_depend_eliminate]: 3.98001e-06
        [updatestate_assign_eliminate]: 3.06999e-06
        [updatestate_loads_eliminate]: 2.89999e-06
        [parameter_eliminate]: 1.74e-06
        [specialize_transform]: 6.24001e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 3.38e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.44e-05
        [j_node_and_user_rematch]: 8.95001e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 8.59998e-06
        [inline_without_move]: 5.52001e-06
        [renormalize]: 0.00062649
        [add_forward_monad_depend]: 5.24998e-06
        [auto_monad_grad]: 2.36998e-06
        [auto_monad_eliminator]: 1.339e-05
        [cse]: 2.692e-05
        [replace_applicator]: 1.381e-05
    [Cycle 2]: 0.00031202, [27]
        [switch_simplify]: 5.46002e-06
        [loop_unroll]: 4.80001e-06
        [a_1]: 8.651e-05
        [with_stream_mark]: 9.87001e-06
        [recompute_prepare]: 5.02999e-06
        [updatestate_depend_eliminate]: 2.59001e-06
        [updatestate_assign_eliminate]: 1.92999e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 9.30013e-07
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.02e-06
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.34002e-06
        [merge_forward]: 2.49001e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.181e-05
        [j_node_and_user_rematch]: 7.8e-06
        [meta_fg_expand]: 1.40001e-06
        [replace_old_param]: 7.03e-06
        [inline_without_move]: 4.75001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 5.07e-06
        [cse]: 1.023e-05
        [replace_applicator]: 5.46998e-06
[py_interpret_to_execute_after_opt_a]: 1.08e-05
[rewriter_after_opt_a]: 3.447e-05
[convert_after_rewriter]: 7e-06
[order_py_execute_after_rewriter]: 5.04e-06
[mutable_eliminate]: 0.00062352
[jit_opt_b]: 4.847e-05, [1]
    [Cycle 1]: 4.166e-05, [2]
        [frontend_op_eliminate]: 1.572e-05
        [inline_after_opt_a]: 1.457e-05
[cconv]: 2.513e-05
[loop_unroll]: 0.00042623
[jit_opt_after_cconv]: 0.00013844, [1]
    [Cycle 1]: 0.00013223, [11]
        [c_1]: 2.05e-05
        [parameter_eliminate]: 2.45002e-06
        [updatestate_depend_eliminate]: 4.57e-06
        [updatestate_assign_eliminate]: 2.67001e-06
        [updatestate_loads_eliminate]: 2.21003e-06
        [cse]: 1.832e-05
        [call_graph_tuple_transform]: 1.943e-05
        [tuple_list_get_item_eliminator]: 5.34e-06
        [none_parameter_eliminate]: 1.27999e-06
        [renormalize]: 8.10018e-07
        [switch_simplify]: 5.48002e-06
[remove_dup_value]: 1.478e-05
[partial_unused_args_eliminate]: 2.22999e-06
[environ_conv]: 5.39e-06
[add_recomputation]: 4.922e-05
[cse_after_recomputation]: 1.974e-05, [1]
    [Cycle 1]: 1.417e-05, [1]
        [cse]: 8.59998e-06
[auto_monad_reorder]: 1.519e-05
[get_jit_bprop_graph]: 2.19001e-06
[rewriter_after_jit_bprop_graph]: 5.30001e-06
[opt_after_jit_grad]: 0.00046501
[symbol_engine_optimizer]: 0.00010303, [1]
    [Cycle 1]: 9.606e-05, [6]
        [build]: 3.75e-06
        [elim_shapecalc]: 8.03999e-06
        [elim_not_effective]: 1.231e-05
        [opt_reshape]: 6.98998e-06
        [fold_const_symbol]: 3.401e-05
        [renormalize]: 7.80012e-07
[validate]: 3.965e-05
Sums
    bootstrap                            :     0.000321s :  3.87%
    type_inference                       :     0.004279s : 51.45%
    event_method                         :     0.000018s :  0.21%
    auto_monad                           :     0.000064s :  0.76%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.40%
    rewriter_before_opt_a                :     0.000072s :  0.87%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000064s :  0.77%
    jit_opt_a.loop_unroll                :     0.000036s :  0.44%
    jit_opt_a.a_1                        :     0.000603s :  7.25%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.34%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000627s :  7.53%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.22%
    jit_opt_a.cse                        :     0.000037s :  0.45%
    jit_opt_a.replace_applicator         :     0.000019s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.13%
    rewriter_after_opt_a                 :     0.000034s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000624s :  7.50%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000025s :  0.30%
    loop_unroll                          :     0.000426s :  5.13%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000018s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000015s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000049s :  0.59%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000465s :  5.59%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000034s :  0.41%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.48%


Time group info:
------[substitution.]   0.000214    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
11.20% :     0.000024s :      2: substitution.fold_const_symbol
 2.85% :     0.000006s :      3: substitution.graph_param_transform
74.20% :     0.000158s :      4: substitution.inline
 1.53% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.05% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.78% :     0.000004s :      2: substitution.replace_old_param
 5.55% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004220     2
79.49% :     0.003355s :      1: type_inference.infer
20.51% :     0.000866s :      1: type_inference.specialize
------[replace.]   0.000065     6
67.74% :     0.000044s :      4: replace.inline
32.26% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000167     6
93.55% :     0.000156s :      4: match.inline
 6.45% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.32% :     0.000001s :     10: predicate.accumulaten_eliminater
 0.96% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 1.96% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000001s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.29% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 6.06% :     0.000007s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.17% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.70% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.41% :     0.000002s :     10: predicate.minmaximum_grad
 1.47% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.88% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000001s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.16% :     0.000001s :      6: predicate.specialize_transform
 1.21% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.80% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.04% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.14% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000595    12
45.40% :     0.000270s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.83% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.77% :     0.000296s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013854    72
 0.38% :     0.000052s :      1: add_recomputation
 0.48% :     0.000067s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 2.43% :     0.000337s :      1: bootstrap
 0.20% :     0.000028s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000022s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.45% :     0.005327s :      1: jit_opt_a
 1.02% :     0.000141s :      1: jit_opt_after_cconv
 0.37% :     0.000051s :      1: jit_opt_b
 3.13% :     0.000434s :      1: loop_unroll
 4.56% :     0.000631s :      1: mutable_eliminate
 5.88% :     0.000815s :     26: opt.transform.jit_opt_a
 0.34% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.21% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.42% :     0.000058s :      4: opt.transform.symbol_engine_opt
 3.42% :     0.000473s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000036s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.26% :     0.000313s :      1: renormalize.infer
 2.21% :     0.000307s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000037s :      1: rewriter_after_opt_a
 0.54% :     0.000075s :      1: rewriter_before_opt_a
 0.77% :     0.000106s :      1: symbol_engine_optimizer
30.98% :     0.004292s :      1: type_inference


TotalTime = 0.0183025, [33]
[bootstrap]: 0.00032068
[type_inference]: 0.00458616
[event_method]: 1.671e-05
[auto_monad]: 6.323e-05
[graph_reusing]: 5.37001e-06
[pre_auto_parallel]: 2.61999e-06
[py_interpret_to_execute]: 3.093e-05
[rewriter_before_opt_a]: 7.209e-05
[expand_dump_flag]: 3.19001e-06
[jit_opt_a]: 0.00536163, [2]
    [Cycle 1]: 0.00148468, [27]
        [switch_simplify]: 5.225e-05
        [loop_unroll]: 2.793e-05
        [a_1]: 0.00050446
        [with_stream_mark]: 1.709e-05
        [recompute_prepare]: 6.61e-06
        [updatestate_depend_eliminate]: 3.43e-06
        [updatestate_assign_eliminate]: 2.76999e-06
        [updatestate_loads_eliminate]: 2.64001e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 5.54998e-06
        [updatestate_useless_node_eliminater]: 4.71002e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 1.87999e-06
        [get_grad_eliminate_]: 4.67998e-06
        [merge_forward]: 3.32002e-06
        [cell_reuse_recompute_pass]: 1.10999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.329e-05
        [j_node_and_user_rematch]: 8.48999e-06
        [meta_fg_expand]: 1.97999e-06
        [replace_old_param]: 8.92e-06
        [inline_without_move]: 4.80001e-06
        [renormalize]: 0.00057301
        [add_forward_monad_depend]: 5.27001e-06
        [auto_monad_grad]: 2.26998e-06
        [auto_monad_eliminator]: 1.356e-05
        [cse]: 3.003e-05
        [replace_applicator]: 1.072e-05
    [Cycle 2]: 0.00031669, [27]
        [switch_simplify]: 5.17e-06
        [loop_unroll]: 4.57998e-06
        [a_1]: 8.725e-05
        [with_stream_mark]: 8.80999e-06
        [recompute_prepare]: 4.87e-06
        [updatestate_depend_eliminate]: 2.51e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.04998e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 4.91002e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 4.83001e-06
        [merge_forward]: 2.33002e-06
        [cell_reuse_recompute_pass]: 1.79e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.221e-05
        [j_node_and_user_rematch]: 8.56002e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 6.94999e-06
        [inline_without_move]: 4.77e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.37999e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.59e-06
        [cse]: 1.06e-05
        [replace_applicator]: 5.69e-06
[py_interpret_to_execute_after_opt_a]: 9.09e-06
[rewriter_after_opt_a]: 3.523e-05
[convert_after_rewriter]: 7.08e-06
[order_py_execute_after_rewriter]: 4.79e-06
[mutable_eliminate]: 0.0005008
[jit_opt_b]: 4.823e-05, [1]
    [Cycle 1]: 4.17e-05, [2]
        [frontend_op_eliminate]: 1.596e-05
        [inline_after_opt_a]: 1.446e-05
[cconv]: 2.403e-05
[loop_unroll]: 0.0004146
[jit_opt_after_cconv]: 0.00017732, [1]
    [Cycle 1]: 0.00017086, [11]
        [c_1]: 2.027e-05
        [parameter_eliminate]: 2.37999e-06
        [updatestate_depend_eliminate]: 5.17e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.11998e-06
        [cse]: 1.705e-05
        [call_graph_tuple_transform]: 1.775e-05
        [tuple_list_get_item_eliminator]: 5.24e-06
        [none_parameter_eliminate]: 4.114e-05
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.65001e-06
[remove_dup_value]: 1.398e-05
[partial_unused_args_eliminate]: 2.53998e-06
[environ_conv]: 5.13002e-06
[add_recomputation]: 4.978e-05
[cse_after_recomputation]: 2.162e-05, [1]
    [Cycle 1]: 1.581e-05, [1]
        [cse]: 9.74e-06
[auto_monad_reorder]: 1.47e-05
[get_jit_bprop_graph]: 1.50999e-06
[rewriter_after_jit_bprop_graph]: 5.02999e-06
[opt_after_jit_grad]: 0.00046313
[symbol_engine_optimizer]: 7.414e-05, [1]
    [Cycle 1]: 6.767e-05, [6]
        [build]: 3.78001e-06
        [elim_shapecalc]: 7.37002e-06
        [elim_not_effective]: 1.304e-05
        [opt_reshape]: 6.02999e-06
        [fold_const_symbol]: 9.08002e-06
        [renormalize]: 4.80009e-07
[validate]: 3.324e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.00573432
[execute]: 7.33e-06
Sums
    bootstrap                            :     0.000321s :  2.27%
    type_inference                       :     0.004586s : 32.42%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000063s :  0.45%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.22%
    rewriter_before_opt_a                :     0.000072s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.41%
    jit_opt_a.loop_unroll                :     0.000033s :  0.23%
    jit_opt_a.a_1                        :     0.000592s :  4.18%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000573s :  4.05%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.14%
    jit_opt_a.cse                        :     0.000041s :  0.29%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000035s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000501s :  3.54%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000415s :  2.93%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000041s :  0.29%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000050s :  0.35%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000463s :  3.27%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005734s : 40.53%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000191    23
 1.12% :     0.000002s :      2: substitution.elim_not_effective
 0.77% :     0.000001s :      2: substitution.fold_const_symbol
 2.57% :     0.000005s :      3: substitution.graph_param_transform
83.41% :     0.000159s :      4: substitution.inline
 1.60% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.38% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.31% :     0.000004s :      2: substitution.replace_old_param
 5.84% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004521     2
83.27% :     0.003765s :      1: type_inference.infer
16.73% :     0.000756s :      1: type_inference.specialize
------[replace.]   0.000061     6
67.70% :     0.000042s :      4: replace.inline
32.30% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000167     6
94.00% :     0.000157s :      4: match.inline
 6.00% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.63% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.28% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 2.12% :     0.000002s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.12% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.32% :     0.000006s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :      6: predicate.less_batch_normalization
 1.81% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.79% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.36% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.46% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.68% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.03% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.47% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.54% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.66% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.25% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.49% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.25% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.84% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.82% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.18% :     0.000003s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000531    12
51.46% :     0.000273s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.99% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.55% :     0.000226s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019767    76
 0.27% :     0.000053s :      1: add_recomputation
 0.34% :     0.000067s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.72% :     0.000340s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
27.14% :     0.005365s :      1: jit_opt_a
 0.91% :     0.000180s :      1: jit_opt_after_cconv
 0.26% :     0.000051s :      1: jit_opt_b
 2.14% :     0.000423s :      1: loop_unroll
 2.58% :     0.000510s :      1: mutable_eliminate
 4.00% :     0.000791s :     26: opt.transform.jit_opt_a
 0.23% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.39% :     0.000472s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.58% :     0.000312s :      1: renormalize.infer
 1.28% :     0.000253s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000038s :      1: rewriter_after_opt_a
 0.38% :     0.000075s :      1: rewriter_before_opt_a
 0.39% :     0.000077s :      1: symbol_engine_optimizer
29.07% :     0.005745s :      1: task_emit
23.26% :     0.004597s :      1: type_inference
 0.27% :     0.000054s :      1: validate


TotalTime = 0.0132794, [30]
[bootstrap]: 0.00030595
[type_inference]: 0.00480568
[event_method]: 1.797e-05
[auto_monad]: 5.166e-05
[graph_reusing]: 4.63001e-06
[pre_auto_parallel]: 2.22999e-06
[py_interpret_to_execute]: 3.43e-05
[rewriter_before_opt_a]: 7.032e-05
[expand_dump_flag]: 1.80001e-06
[jit_opt_a]: 0.00564001, [2]
    [Cycle 1]: 0.00169571, [27]
        [switch_simplify]: 5.899e-05
        [loop_unroll]: 2.785e-05
        [a_1]: 0.00053547
        [with_stream_mark]: 2.15e-05
        [recompute_prepare]: 7.04001e-06
        [updatestate_depend_eliminate]: 3.79002e-06
        [updatestate_assign_eliminate]: 2.77002e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 1.75001e-06
        [specialize_transform]: 5.83002e-06
        [updatestate_useless_node_eliminater]: 4.78001e-06
        [accelerated_algorithm]: 5.16002e-06
        [meta_shard_fg_expand]: 1.96998e-06
        [get_grad_eliminate_]: 5.32999e-06
        [merge_forward]: 3.47002e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.385e-05
        [j_node_and_user_rematch]: 8.77999e-06
        [meta_fg_expand]: 2.29999e-06
        [replace_old_param]: 9.05001e-06
        [inline_without_move]: 5.32001e-06
        [renormalize]: 0.00072529
        [add_forward_monad_depend]: 5.92001e-06
        [auto_monad_grad]: 2.51e-06
        [auto_monad_eliminator]: 1.468e-05
        [cse]: 3.076e-05
        [replace_applicator]: 1.287e-05
    [Cycle 2]: 0.00032049, [27]
        [switch_simplify]: 5.61998e-06
        [loop_unroll]: 5.29e-06
        [a_1]: 8.785e-05
        [with_stream_mark]: 9.66e-06
        [recompute_prepare]: 5.40999e-06
        [updatestate_depend_eliminate]: 2.54999e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 8.80013e-07
        [specialize_transform]: 5.29e-06
        [updatestate_useless_node_eliminater]: 4.80999e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 1.26002e-06
        [get_grad_eliminate_]: 4.46002e-06
        [merge_forward]: 2.58998e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.301e-05
        [j_node_and_user_rematch]: 7.9e-06
        [meta_fg_expand]: 1.68002e-06
        [replace_old_param]: 7.49002e-06
        [inline_without_move]: 4.68001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.30999e-06
        [auto_monad_grad]: 1.05999e-06
        [auto_monad_eliminator]: 5.00001e-06
        [cse]: 1.081e-05
        [replace_applicator]: 6.54999e-06
[py_interpret_to_execute_after_opt_a]: 1.194e-05
[rewriter_after_opt_a]: 3.806e-05
[convert_after_rewriter]: 6.74999e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00066623
[jit_opt_b]: 5.238e-05, [1]
    [Cycle 1]: 4.406e-05, [2]
        [frontend_op_eliminate]: 1.693e-05
        [inline_after_opt_a]: 1.443e-05
[cconv]: 2.633e-05
[loop_unroll]: 0.00044739
[jit_opt_after_cconv]: 0.00019753, [1]
    [Cycle 1]: 0.00018963, [11]
        [c_1]: 2.173e-05
        [parameter_eliminate]: 3.91999e-06
        [updatestate_depend_eliminate]: 6.71999e-06
        [updatestate_assign_eliminate]: 2.53998e-06
        [updatestate_loads_eliminate]: 2.11998e-06
        [cse]: 2.4e-05
        [call_graph_tuple_transform]: 2.246e-05
        [tuple_list_get_item_eliminator]: 5.08002e-06
        [none_parameter_eliminate]: 1.56998e-06
        [renormalize]: 6.89994e-07
        [switch_simplify]: 6.04999e-06
[remove_dup_value]: 1.591e-05
[partial_unused_args_eliminate]: 2.29999e-06
[environ_conv]: 6.14999e-06
[add_recomputation]: 5.139e-05
[cse_after_recomputation]: 2.234e-05, [1]
    [Cycle 1]: 1.64e-05, [1]
        [cse]: 9.22001e-06
[auto_monad_reorder]: 1.555e-05
[get_jit_bprop_graph]: 2.41e-06
[rewriter_after_jit_bprop_graph]: 7.08e-06
[opt_after_jit_grad]: 0.00049219
[symbol_engine_optimizer]: 7.26e-05, [1]
    [Cycle 1]: 6.631e-05, [6]
        [build]: 3.81999e-06
        [elim_shapecalc]: 7.21999e-06
        [elim_not_effective]: 1.312e-05
        [opt_reshape]: 5.39e-06
        [fold_const_symbol]: 8.47998e-06
        [renormalize]: 9.40025e-07
[validate]: 3.778e-05
Sums
    bootstrap                            :     0.000306s :  3.39%
    type_inference                       :     0.004806s : 53.22%
    event_method                         :     0.000018s :  0.20%
    auto_monad                           :     0.000052s :  0.57%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.38%
    rewriter_before_opt_a                :     0.000070s :  0.78%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.72%
    jit_opt_a.loop_unroll                :     0.000033s :  0.37%
    jit_opt_a.a_1                        :     0.000623s :  6.90%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.18%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000725s :  8.03%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.22%
    jit_opt_a.cse                        :     0.000042s :  0.46%
    jit_opt_a.replace_applicator         :     0.000019s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.13%
    rewriter_after_opt_a                 :     0.000038s :  0.42%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000666s :  7.38%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.16%
    cconv                                :     0.000026s :  0.29%
    loop_unroll                          :     0.000447s :  4.95%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.27%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.25%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000016s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000051s :  0.57%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.08%
    opt_after_jit_grad                   :     0.000492s :  5.45%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000038s :  0.42%


Time group info:
------[substitution.]   0.000222    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.85% :     0.000006s :      3: substitution.graph_param_transform
84.08% :     0.000186s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.01% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.94% :     0.000004s :      2: substitution.replace_old_param
 5.98% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004744     2
80.14% :     0.003802s :      1: type_inference.infer
19.86% :     0.000942s :      1: type_inference.specialize
------[replace.]   0.000068     6
67.61% :     0.000046s :      4: replace.inline
32.39% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000197     6
93.79% :     0.000184s :      4: match.inline
 6.21% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.47% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.04% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 1.82% :     0.000002s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000001s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.73% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.58% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.30% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 2.30% :     0.000003s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000002s :     15: predicate.partial_eliminate
 1.34% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.55% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.38% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.85% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.48% :     0.000002s :     10: predicate.transpose_eliminate
 1.32% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.07% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000652    12
50.79% :     0.000331s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.20% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.01% :     0.000274s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014900    72
 0.36% :     0.000054s :      1: add_recomputation
 0.37% :     0.000055s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.16% :     0.000323s :      1: bootstrap
 0.19% :     0.000029s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000024s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.15% :     0.000022s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
37.87% :     0.005643s :      1: jit_opt_a
 1.35% :     0.000201s :      1: jit_opt_after_cconv
 0.37% :     0.000055s :      1: jit_opt_b
 3.06% :     0.000455s :      1: loop_unroll
 4.53% :     0.000675s :      1: mutable_eliminate
 5.61% :     0.000835s :     26: opt.transform.jit_opt_a
 0.34% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.35% :     0.000499s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000038s :      1: py_interpret_to_execute
 0.10% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000018s :      1: remove_dup_value
 2.69% :     0.000401s :      1: renormalize.infer
 2.13% :     0.000317s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000041s :      1: rewriter_after_opt_a
 0.49% :     0.000073s :      1: rewriter_before_opt_a
 0.50% :     0.000075s :      1: symbol_engine_optimizer
32.34% :     0.004818s :      1: type_inference


TotalTime = 0.025004, [33]
[bootstrap]: 0.00037332
[type_inference]: 0.00477778
[event_method]: 1.766e-05
[auto_monad]: 7.037e-05
[graph_reusing]: 6.33998e-06
[pre_auto_parallel]: 2.79001e-06
[py_interpret_to_execute]: 4.615e-05
[rewriter_before_opt_a]: 8.056e-05
[expand_dump_flag]: 4.37e-06
[jit_opt_a]: 0.00616842, [2]
    [Cycle 1]: 0.00207641, [27]
        [switch_simplify]: 6.297e-05
        [loop_unroll]: 2.939e-05
        [a_1]: 0.00061512
        [with_stream_mark]: 2.241e-05
        [recompute_prepare]: 9.25001e-06
        [updatestate_depend_eliminate]: 4.13001e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 2.41998e-06
        [specialize_transform]: 6.78998e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 5.96e-06
        [meta_shard_fg_expand]: 2.09999e-06
        [get_grad_eliminate_]: 5.52001e-06
        [merge_forward]: 3.98001e-06
        [cell_reuse_recompute_pass]: 1.42e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.596e-05
        [j_node_and_user_rematch]: 1.042e-05
        [meta_fg_expand]: 2.21e-06
        [replace_old_param]: 9.97999e-06
        [inline_without_move]: 5.12999e-06
        [renormalize]: 0.00095841
        [add_forward_monad_depend]: 1.018e-05
        [auto_monad_grad]: 3.24001e-06
        [auto_monad_eliminator]: 1.739e-05
        [cse]: 2.834e-05
        [replace_applicator]: 2.213e-05
    [Cycle 2]: 0.00037334, [27]
        [switch_simplify]: 6.85002e-06
        [loop_unroll]: 5.67999e-06
        [a_1]: 0.00010622
        [with_stream_mark]: 1.458e-05
        [recompute_prepare]: 5.70001e-06
        [updatestate_depend_eliminate]: 3.57997e-06
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 2.19999e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 5.46e-06
        [updatestate_useless_node_eliminater]: 4.92999e-06
        [accelerated_algorithm]: 5.19e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 4.90001e-06
        [merge_forward]: 3.76001e-06
        [cell_reuse_recompute_pass]: 3.58e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.679e-05
        [j_node_and_user_rematch]: 9.52001e-06
        [meta_fg_expand]: 2.17001e-06
        [replace_old_param]: 8.32e-06
        [inline_without_move]: 5.08002e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.71998e-06
        [auto_monad_grad]: 1.09998e-06
        [auto_monad_eliminator]: 6.49999e-06
        [cse]: 1.27e-05
        [replace_applicator]: 5.44e-06
[py_interpret_to_execute_after_opt_a]: 1.486e-05
[rewriter_after_opt_a]: 3.743e-05
[convert_after_rewriter]: 7.12997e-06
[order_py_execute_after_rewriter]: 5.22e-06
[mutable_eliminate]: 0.00256182
[jit_opt_b]: 7.356e-05, [1]
    [Cycle 1]: 6.14e-05, [2]
        [frontend_op_eliminate]: 2.366e-05
        [inline_after_opt_a]: 2.297e-05
[cconv]: 4.234e-05
[loop_unroll]: 0.0007785
[jit_opt_after_cconv]: 0.00020039, [1]
    [Cycle 1]: 0.00019104, [11]
        [c_1]: 2.653e-05
        [parameter_eliminate]: 6.64001e-06
        [updatestate_depend_eliminate]: 1.127e-05
        [updatestate_assign_eliminate]: 3.51001e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [cse]: 3.843e-05
        [call_graph_tuple_transform]: 2.8e-05
        [tuple_list_get_item_eliminator]: 6.42001e-06
        [none_parameter_eliminate]: 1.65001e-06
        [renormalize]: 1.22e-06
        [switch_simplify]: 6.28e-06
[remove_dup_value]: 1.525e-05
[partial_unused_args_eliminate]: 2.66999e-06
[environ_conv]: 7.36999e-06
[add_recomputation]: 6.048e-05
[cse_after_recomputation]: 2.911e-05, [1]
    [Cycle 1]: 2.206e-05, [1]
        [cse]: 1.188e-05
[auto_monad_reorder]: 1.771e-05
[get_jit_bprop_graph]: 3.10002e-06
[rewriter_after_jit_bprop_graph]: 1.051e-05
[opt_after_jit_grad]: 0.00061996
[symbol_engine_optimizer]: 0.00014477, [1]
    [Cycle 1]: 0.00013625, [6]
        [build]: 4.22e-06
        [elim_shapecalc]: 9.30001e-06
        [elim_not_effective]: 1.613e-05
        [opt_reshape]: 7.8e-06
        [fold_const_symbol]: 1.006e-05
        [renormalize]: 1.01997e-06
[validate]: 4.761e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00850784
[execute]: 7.77e-06
Sums
    bootstrap                            :     0.000373s :  1.82%
    type_inference                       :     0.004778s : 23.33%
    event_method                         :     0.000018s :  0.09%
    auto_monad                           :     0.000070s :  0.34%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000046s :  0.23%
    rewriter_before_opt_a                :     0.000081s :  0.39%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000070s :  0.34%
    jit_opt_a.loop_unroll                :     0.000035s :  0.17%
    jit_opt_a.a_1                        :     0.000721s :  3.52%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000958s :  4.68%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.12%
    jit_opt_a.cse                        :     0.000041s :  0.20%
    jit_opt_a.replace_applicator         :     0.000028s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.18%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.002562s : 12.51%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000023s :  0.11%
    cconv                                :     0.000042s :  0.21%
    loop_unroll                          :     0.000778s :  3.80%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000038s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.07%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000060s :  0.30%
    cse_after_recomputation.cse          :     0.000012s :  0.06%
    auto_monad_reorder                   :     0.000018s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.05%
    opt_after_jit_grad                   :     0.000620s :  3.03%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000048s :  0.23%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008508s : 41.54%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000265    23
 0.98% :     0.000003s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.54% :     0.000007s :      3: substitution.graph_param_transform
85.13% :     0.000226s :      4: substitution.inline
 1.93% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.08% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000005s :      2: substitution.replace_old_param
 4.84% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004697     2
78.94% :     0.003708s :      1: type_inference.infer
21.06% :     0.000989s :      1: type_inference.specialize
------[replace.]   0.000081     6
65.27% :     0.000053s :      4: replace.inline
34.73% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000235     6
94.98% :     0.000223s :      4: match.inline
 5.02% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   691
 1.03% :     0.000001s :     10: predicate.accumulaten_eliminater
 2.02% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.92% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.23% :     0.000002s :     10: predicate.cast_eliminate
 0.61% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.89% :     0.000001s :     10: predicate.compare_switch_simplify
 1.08% :     0.000001s :     10: predicate.depend_value_elim
 0.91% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.11% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.00% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.99% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     10: predicate.environ_get_eliminate
 0.87% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.12% :     0.000002s :      6: predicate.get_grad_eliminate
 0.43% :     0.000001s :      3: predicate.graph_param_transform
 5.80% :     0.000008s :     22: predicate.inline
 0.70% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :      6: predicate.less_batch_normalization
 1.30% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000003s :     15: predicate.load_eliminater
 1.85% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.89% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.92% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 3.98% :     0.000005s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     15: predicate.partial_eliminate
 0.98% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.43% :     0.000002s :     10: predicate.reduce_eliminate
 1.26% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.51% :     0.000001s :      3: predicate.reset_defer_inline
 1.63% :     0.000002s :     10: predicate.reshape_eliminate
 1.04% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.25% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000002s :     10: predicate.same_eliminate
 0.76% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 0.82% :     0.000001s :      6: predicate.specialize_transform
 1.19% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.32% :     0.000003s :     16: predicate.switch_defer_inline
 1.99% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.65% :     0.000010s :     53: predicate.switch_simplify
 1.20% :     0.000002s :     10: predicate.tile_eliminate
 1.02% :     0.000001s :     10: predicate.transpose_eliminate
 1.22% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.52% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.93% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.40% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.91% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000691    12
43.55% :     0.000301s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.56% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.89% :     0.000338s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027087    76
 0.24% :     0.000065s :      1: add_recomputation
 0.28% :     0.000075s :      1: auto_monad
 0.08% :     0.000021s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.42% :     0.000385s :      1: bootstrap
 0.17% :     0.000045s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000032s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.08% :     0.000022s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.02% :     0.000007s :      1: expand_dump_flag
 0.02% :     0.000006s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
22.79% :     0.006172s :      1: jit_opt_a
 0.76% :     0.000205s :      1: jit_opt_after_cconv
 0.29% :     0.000078s :      1: jit_opt_b
 2.94% :     0.000795s :      1: loop_unroll
 9.55% :     0.002587s :      1: mutable_eliminate
 3.55% :     0.000961s :     26: opt.transform.jit_opt_a
 0.23% :     0.000064s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000037s :      4: opt.transform.jit_opt_b
 0.07% :     0.000020s :      1: opt.transform.loop_unroll_optimizer
 0.13% :     0.000036s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000039s :      4: opt.transform.symbol_engine_opt
 2.33% :     0.000631s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000049s :      1: py_interpret_to_execute
 0.07% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 1.88% :     0.000508s :      1: renormalize.infer
 1.62% :     0.000440s :      1: renormalize.specialize
 0.05% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000041s :      1: rewriter_after_opt_a
 0.31% :     0.000084s :      1: rewriter_before_opt_a
 0.55% :     0.000148s :      1: symbol_engine_optimizer
31.48% :     0.008526s :      1: task_emit
17.70% :     0.004793s :      1: type_inference
 0.26% :     0.000071s :      1: validate


TotalTime = 0.0183662, [33]
[bootstrap]: 0.00032371
[type_inference]: 0.00448826
[event_method]: 1.687e-05
[auto_monad]: 6.55e-05
[graph_reusing]: 5.65001e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 3.194e-05
[rewriter_before_opt_a]: 7.051e-05
[expand_dump_flag]: 3.22002e-06
[jit_opt_a]: 0.00535201, [2]
    [Cycle 1]: 0.00144984, [27]
        [switch_simplify]: 5.574e-05
        [loop_unroll]: 2.912e-05
        [a_1]: 0.00050608
        [with_stream_mark]: 1.567e-05
        [recompute_prepare]: 6.65998e-06
        [updatestate_depend_eliminate]: 3.16999e-06
        [updatestate_assign_eliminate]: 3.11999e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [parameter_eliminate]: 2.22999e-06
        [specialize_transform]: 5.62001e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.52999e-06
        [meta_shard_fg_expand]: 1.84e-06
        [get_grad_eliminate_]: 4.88001e-06
        [merge_forward]: 3.49001e-06
        [cell_reuse_recompute_pass]: 1.05999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.354e-05
        [j_node_and_user_rematch]: 8.60999e-06
        [meta_fg_expand]: 2.22001e-06
        [replace_old_param]: 8.48999e-06
        [inline_without_move]: 5.19998e-06
        [renormalize]: 0.00053427
        [add_forward_monad_depend]: 5.10001e-06
        [auto_monad_grad]: 2.12999e-06
        [auto_monad_eliminator]: 1.327e-05
        [cse]: 2.766e-05
        [replace_applicator]: 1.11e-05
    [Cycle 2]: 0.00032138, [27]
        [switch_simplify]: 5.51e-06
        [loop_unroll]: 4.99998e-06
        [a_1]: 8.72e-05
        [with_stream_mark]: 8.42998e-06
        [recompute_prepare]: 4.62998e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 4.75001e-06
        [updatestate_useless_node_eliminater]: 5.54998e-06
        [accelerated_algorithm]: 5.02999e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.69998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.204e-05
        [j_node_and_user_rematch]: 7.56001e-06
        [meta_fg_expand]: 1.42999e-06
        [replace_old_param]: 6.75998e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 8.10018e-07
        [auto_monad_eliminator]: 5.19e-06
        [cse]: 1.078e-05
        [replace_applicator]: 4.99e-06
[py_interpret_to_execute_after_opt_a]: 8.47e-06
[rewriter_after_opt_a]: 3.124e-05
[convert_after_rewriter]: 6.99001e-06
[order_py_execute_after_rewriter]: 5.35001e-06
[mutable_eliminate]: 0.00048681
[jit_opt_b]: 4.958e-05, [1]
    [Cycle 1]: 4.281e-05, [2]
        [frontend_op_eliminate]: 1.708e-05
        [inline_after_opt_a]: 1.433e-05
[cconv]: 2.355e-05
[loop_unroll]: 0.00040898
[jit_opt_after_cconv]: 0.00013345, [1]
    [Cycle 1]: 0.00012717, [11]
        [c_1]: 1.925e-05
        [parameter_eliminate]: 2.36e-06
        [updatestate_depend_eliminate]: 5.20999e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 1.731e-05
        [call_graph_tuple_transform]: 1.762e-05
        [tuple_list_get_item_eliminator]: 5.35001e-06
        [none_parameter_eliminate]: 1.45001e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 5.15999e-06
[remove_dup_value]: 1.316e-05
[partial_unused_args_eliminate]: 2.25002e-06
[environ_conv]: 4.67e-06
[add_recomputation]: 4.859e-05
[cse_after_recomputation]: 1.959e-05, [1]
    [Cycle 1]: 1.434e-05, [1]
        [cse]: 8.70999e-06
[auto_monad_reorder]: 1.435e-05
[get_jit_bprop_graph]: 1.53002e-06
[rewriter_after_jit_bprop_graph]: 4.67e-06
[opt_after_jit_grad]: 0.0005085
[symbol_engine_optimizer]: 7.141e-05, [1]
    [Cycle 1]: 6.498e-05, [6]
        [build]: 3.65003e-06
        [elim_shapecalc]: 7.26001e-06
        [elim_not_effective]: 1.304e-05
        [opt_reshape]: 5.54e-06
        [fold_const_symbol]: 8.29998e-06
        [renormalize]: 5.10016e-07
[validate]: 3.241e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00593172
[execute]: 9.14998e-06
Sums
    bootstrap                            :     0.000324s :  2.28%
    type_inference                       :     0.004488s : 31.63%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000065s :  0.46%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.23%
    rewriter_before_opt_a                :     0.000071s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.43%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000593s :  4.18%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000534s :  3.77%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000038s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000487s :  3.43%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000409s :  2.88%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000509s :  3.58%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005932s : 41.80%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000189    23
 1.12% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.68% :     0.000005s :      3: substitution.graph_param_transform
83.49% :     0.000157s :      4: substitution.inline
 1.65% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.80% :     0.000003s :      2: substitution.replace_old_param
 6.33% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004426     2
81.20% :     0.003594s :      1: type_inference.infer
18.80% :     0.000832s :      1: type_inference.specialize
------[replace.]   0.000062     6
68.51% :     0.000043s :      4: replace.inline
31.49% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000166     6
93.39% :     0.000155s :      4: match.inline
 6.61% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.38% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 2.00% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.61% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000001s :     10: predicate.depend_value_elim
 1.22% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.17% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.58% :     0.000001s :      3: predicate.graph_param_transform
 5.52% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.41% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.75% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.32% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.45% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.00% :     0.000009s :     53: predicate.switch_simplify
 1.51% :     0.000002s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.74% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.10% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.02% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.89% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000539    12
51.25% :     0.000276s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.79% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.96% :     0.000232s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019798    76
 0.26% :     0.000051s :      1: add_recomputation
 0.35% :     0.000069s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.71% :     0.000339s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.07% :     0.000013s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.05% :     0.005354s :      1: jit_opt_a
 0.69% :     0.000136s :      1: jit_opt_after_cconv
 0.26% :     0.000052s :      1: jit_opt_b
 2.11% :     0.000417s :      1: loop_unroll
 2.50% :     0.000495s :      1: mutable_eliminate
 4.02% :     0.000796s :     26: opt.transform.jit_opt_a
 0.22% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.61% :     0.000517s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000035s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.42% :     0.000282s :      1: renormalize.infer
 1.24% :     0.000245s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000034s :      1: rewriter_after_opt_a
 0.37% :     0.000073s :      1: rewriter_before_opt_a
 0.37% :     0.000074s :      1: symbol_engine_optimizer
30.03% :     0.005946s :      1: task_emit
22.74% :     0.004503s :      1: type_inference
 0.25% :     0.000050s :      1: validate


TotalTime = 0.0139194, [30]
[bootstrap]: 0.00030542
[type_inference]: 0.00495806
[event_method]: 1.928e-05
[auto_monad]: 6.698e-05
[graph_reusing]: 6.33e-06
[pre_auto_parallel]: 2.43e-06
[py_interpret_to_execute]: 3.426e-05
[rewriter_before_opt_a]: 7.931e-05
[expand_dump_flag]: 3.23e-06
[jit_opt_a]: 0.00609949, [2]
    [Cycle 1]: 0.00179233, [27]
        [switch_simplify]: 6.286e-05
        [loop_unroll]: 3.135e-05
        [a_1]: 0.00057916
        [with_stream_mark]: 2.298e-05
        [recompute_prepare]: 8.97e-06
        [updatestate_depend_eliminate]: 3.69002e-06
        [updatestate_assign_eliminate]: 2.92002e-06
        [updatestate_loads_eliminate]: 2.89001e-06
        [parameter_eliminate]: 2.07001e-06
        [specialize_transform]: 7.03998e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 5.84e-06
        [meta_shard_fg_expand]: 2.12001e-06
        [get_grad_eliminate_]: 5.92999e-06
        [merge_forward]: 3.68e-06
        [cell_reuse_recompute_pass]: 1.49998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.664e-05
        [j_node_and_user_rematch]: 9.52999e-06
        [meta_fg_expand]: 2.56998e-06
        [replace_old_param]: 1.087e-05
        [inline_without_move]: 5.86e-06
        [renormalize]: 0.00071692
        [add_forward_monad_depend]: 6.79999e-06
        [auto_monad_grad]: 2.48e-06
        [auto_monad_eliminator]: 1.722e-05
        [cse]: 3.184e-05
        [replace_applicator]: 1.571e-05
    [Cycle 2]: 0.00040248, [27]
        [switch_simplify]: 5.86998e-06
        [loop_unroll]: 5.56e-06
        [a_1]: 9.686e-05
        [with_stream_mark]: 1.277e-05
        [recompute_prepare]: 6.66e-06
        [updatestate_depend_eliminate]: 3.25e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 4.384e-05
        [parameter_eliminate]: 1.33002e-06
        [specialize_transform]: 5.86e-06
        [updatestate_useless_node_eliminater]: 5.56998e-06
        [accelerated_algorithm]: 4.94003e-06
        [meta_shard_fg_expand]: 1.76e-06
        [get_grad_eliminate_]: 4.63999e-06
        [merge_forward]: 3.66999e-06
        [cell_reuse_recompute_pass]: 2.14999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.893e-05
        [j_node_and_user_rematch]: 8.55001e-06
        [meta_fg_expand]: 2.32001e-06
        [replace_old_param]: 7.46999e-06
        [inline_without_move]: 5.06002e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.92001e-06
        [auto_monad_grad]: 1.03001e-06
        [auto_monad_eliminator]: 8.10999e-06
        [cse]: 1.25e-05
        [replace_applicator]: 6.01e-06
[py_interpret_to_execute_after_opt_a]: 1.437e-05
[rewriter_after_opt_a]: 4.053e-05
[convert_after_rewriter]: 7.48e-06
[order_py_execute_after_rewriter]: 5.51e-06
[mutable_eliminate]: 0.00067472
[jit_opt_b]: 5.377e-05, [1]
    [Cycle 1]: 4.526e-05, [2]
        [frontend_op_eliminate]: 1.654e-05
        [inline_after_opt_a]: 1.633e-05
[cconv]: 3.19e-05
[loop_unroll]: 0.00047564
[jit_opt_after_cconv]: 0.00015653, [1]
    [Cycle 1]: 0.0001493, [11]
        [c_1]: 2.04e-05
        [parameter_eliminate]: 3.83001e-06
        [updatestate_depend_eliminate]: 8.48999e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [cse]: 2.792e-05
        [call_graph_tuple_transform]: 2.067e-05
        [tuple_list_get_item_eliminator]: 5.20001e-06
        [none_parameter_eliminate]: 1.46998e-06
        [renormalize]: 5.60016e-07
        [switch_simplify]: 5.05001e-06
[remove_dup_value]: 1.426e-05
[partial_unused_args_eliminate]: 2.66e-06
[environ_conv]: 6.39001e-06
[add_recomputation]: 5.231e-05
[cse_after_recomputation]: 2.16e-05, [1]
    [Cycle 1]: 1.613e-05, [1]
        [cse]: 1.006e-05
[auto_monad_reorder]: 1.557e-05
[get_jit_bprop_graph]: 2.14e-06
[rewriter_after_jit_bprop_graph]: 6.92002e-06
[opt_after_jit_grad]: 0.0004652
[symbol_engine_optimizer]: 6.974e-05, [1]
    [Cycle 1]: 6.366e-05, [6]
        [build]: 3.36001e-06
        [elim_shapecalc]: 7.44002e-06
        [elim_not_effective]: 1.079e-05
        [opt_reshape]: 5.78002e-06
        [fold_const_symbol]: 8.97999e-06
        [renormalize]: 3.89991e-07
[validate]: 3.917e-05
Sums
    bootstrap                            :     0.000305s :  3.26%
    type_inference                       :     0.004958s : 52.90%
    event_method                         :     0.000019s :  0.21%
    auto_monad                           :     0.000067s :  0.71%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.37%
    rewriter_before_opt_a                :     0.000079s :  0.85%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000069s :  0.73%
    jit_opt_a.loop_unroll                :     0.000037s :  0.39%
    jit_opt_a.a_1                        :     0.000676s :  7.21%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.38%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000047s :  0.50%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000013s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.38%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000018s :  0.20%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000717s :  7.65%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.27%
    jit_opt_a.cse                        :     0.000044s :  0.47%
    jit_opt_a.replace_applicator         :     0.000022s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.15%
    rewriter_after_opt_a                 :     0.000041s :  0.43%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000675s :  7.20%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.17%
    cconv                                :     0.000032s :  0.34%
    loop_unroll                          :     0.000476s :  5.08%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.09%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000028s :  0.30%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.15%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000052s :  0.56%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000465s :  4.96%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000039s :  0.42%


Time group info:
------[substitution.]   0.000235    23
 0.70% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000002s :      2: substitution.fold_const_symbol
 2.50% :     0.000006s :      3: substitution.graph_param_transform
82.39% :     0.000193s :      4: substitution.inline
 1.54% :     0.000004s :      4: substitution.j_node_and_user_rematch
 3.63% :     0.000009s :      4: substitution.remove_not_recompute_node
 2.25% :     0.000005s :      2: substitution.replace_old_param
 6.30% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004884     2
79.27% :     0.003871s :      1: type_inference.infer
20.73% :     0.001012s :      1: type_inference.specialize
------[replace.]   0.000073     6
68.56% :     0.000050s :      4: replace.inline
31.44% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000204     6
93.34% :     0.000191s :      4: match.inline
 6.66% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.60% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.57% :     0.000002s :     10: predicate.addn_zero_filter
 2.03% :     0.000003s :     10: predicate.arithmetic_simplify
 1.21% :     0.000002s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.13% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.69% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.74% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.20% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000001s :      3: predicate.graph_param_transform
 6.16% :     0.000008s :     22: predicate.inline
 1.13% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 1.84% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 2.24% :     0.000003s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.44% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.60% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.59% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.08% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.72% :     0.000010s :     53: predicate.switch_simplify
 1.02% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000002s :     10: predicate.transpose_eliminate
 1.24% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.48% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.28% :     0.000003s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000627    12
49.07% :     0.000308s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.12% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.81% :     0.000275s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015603    72
 0.35% :     0.000055s :      1: add_recomputation
 0.46% :     0.000072s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.07% :     0.000322s :      1: bootstrap
 0.23% :     0.000036s :      1: cconv
 0.07% :     0.000011s :      1: convert_after_rewriter
 0.15% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
39.12% :     0.006104s :      1: jit_opt_a
 1.02% :     0.000159s :      1: jit_opt_after_cconv
 0.36% :     0.000057s :      1: jit_opt_b
 3.10% :     0.000484s :      1: loop_unroll
 4.41% :     0.000689s :      1: mutable_eliminate
 5.87% :     0.000915s :     26: opt.transform.jit_opt_a
 0.31% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000026s :      4: opt.transform.jit_opt_b
 0.10% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000018s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.03% :     0.000473s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000037s :      1: py_interpret_to_execute
 0.11% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 2.36% :     0.000368s :      1: renormalize.infer
 2.18% :     0.000340s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000044s :      1: rewriter_after_opt_a
 0.53% :     0.000082s :      1: rewriter_before_opt_a
 0.46% :     0.000072s :      1: symbol_engine_optimizer
31.88% :     0.004974s :      1: type_inference


TotalTime = 0.0187208, [33]
[bootstrap]: 0.00032527
[type_inference]: 0.00453955
[event_method]: 1.858e-05
[auto_monad]: 6.082e-05
[graph_reusing]: 5.64998e-06
[pre_auto_parallel]: 2.44001e-06
[py_interpret_to_execute]: 2.983e-05
[rewriter_before_opt_a]: 7.027e-05
[expand_dump_flag]: 2.76e-06
[jit_opt_a]: 0.00561982, [2]
    [Cycle 1]: 0.00149099, [27]
        [switch_simplify]: 5.282e-05
        [loop_unroll]: 2.789e-05
        [a_1]: 0.00051054
        [with_stream_mark]: 1.807e-05
        [recompute_prepare]: 6.88998e-06
        [updatestate_depend_eliminate]: 3.16999e-06
        [updatestate_assign_eliminate]: 3.29001e-06
        [updatestate_loads_eliminate]: 2.71999e-06
        [parameter_eliminate]: 2.02001e-06
        [specialize_transform]: 5.53002e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 6.15002e-06
        [meta_shard_fg_expand]: 2.21e-06
        [get_grad_eliminate_]: 5.52999e-06
        [merge_forward]: 3.65998e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.382e-05
        [j_node_and_user_rematch]: 8.54e-06
        [meta_fg_expand]: 2.23002e-06
        [replace_old_param]: 8.49998e-06
        [inline_without_move]: 4.87998e-06
        [renormalize]: 0.00056724
        [add_forward_monad_depend]: 5.59e-06
        [auto_monad_grad]: 2.54999e-06
        [auto_monad_eliminator]: 1.494e-05
        [cse]: 2.816e-05
        [replace_applicator]: 1.222e-05
    [Cycle 2]: 0.00031202, [27]
        [switch_simplify]: 5.12999e-06
        [loop_unroll]: 4.74998e-06
        [a_1]: 8.694e-05
        [with_stream_mark]: 8.95999e-06
        [recompute_prepare]: 5.09003e-06
        [updatestate_depend_eliminate]: 2.33998e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 9.99979e-07
        [specialize_transform]: 4.70999e-06
        [updatestate_useless_node_eliminater]: 4.53001e-06
        [accelerated_algorithm]: 4.90001e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 4.44002e-06
        [merge_forward]: 2.32001e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.207e-05
        [j_node_and_user_rematch]: 7.98001e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 7.35e-06
        [inline_without_move]: 4.47998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.30999e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 5.17999e-06
        [cse]: 1.032e-05
        [replace_applicator]: 5.10001e-06
[py_interpret_to_execute_after_opt_a]: 1.045e-05
[rewriter_after_opt_a]: 3.511e-05
[convert_after_rewriter]: 6.96999e-06
[order_py_execute_after_rewriter]: 4.82998e-06
[mutable_eliminate]: 0.00055368
[jit_opt_b]: 4.764e-05, [1]
    [Cycle 1]: 4.084e-05, [2]
        [frontend_op_eliminate]: 1.56e-05
        [inline_after_opt_a]: 1.402e-05
[cconv]: 2.625e-05
[loop_unroll]: 0.00041415
[jit_opt_after_cconv]: 0.00013357, [1]
    [Cycle 1]: 0.00012734, [11]
        [c_1]: 1.922e-05
        [parameter_eliminate]: 2.36e-06
        [updatestate_depend_eliminate]: 5.52001e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.709e-05
        [call_graph_tuple_transform]: 1.8e-05
        [tuple_list_get_item_eliminator]: 4.90999e-06
        [none_parameter_eliminate]: 1.51998e-06
        [renormalize]: 8.80013e-07
        [switch_simplify]: 5.04e-06
[remove_dup_value]: 1.324e-05
[partial_unused_args_eliminate]: 2.18998e-06
[environ_conv]: 5.11002e-06
[add_recomputation]: 4.707e-05
[cse_after_recomputation]: 1.967e-05, [1]
    [Cycle 1]: 1.411e-05, [1]
        [cse]: 8.59e-06
[auto_monad_reorder]: 1.497e-05
[get_jit_bprop_graph]: 1.79e-06
[rewriter_after_jit_bprop_graph]: 5.57999e-06
[opt_after_jit_grad]: 0.0004813
[symbol_engine_optimizer]: 7.185e-05, [1]
    [Cycle 1]: 6.55e-05, [6]
        [build]: 4.32998e-06
        [elim_shapecalc]: 7.33e-06
        [elim_not_effective]: 1.264e-05
        [opt_reshape]: 5.40001e-06
        [fold_const_symbol]: 8.24002e-06
        [renormalize]: 4.89992e-07
[validate]: 3.352e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00591173
[execute]: 8.84e-06
Sums
    bootstrap                            :     0.000325s :  2.27%
    type_inference                       :     0.004540s : 31.72%
    event_method                         :     0.000019s :  0.13%
    auto_monad                           :     0.000061s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.21%
    rewriter_before_opt_a                :     0.000070s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.23%
    jit_opt_a.a_1                        :     0.000597s :  4.17%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000567s :  3.96%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.14%
    jit_opt_a.cse                        :     0.000038s :  0.27%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000554s :  3.87%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000026s :  0.18%
    loop_unroll                          :     0.000414s :  2.89%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000481s :  3.36%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005912s : 41.31%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000195    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.98% :     0.000006s :      3: substitution.graph_param_transform
83.50% :     0.000162s :      4: substitution.inline
 1.54% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.02% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000004s :      2: substitution.replace_old_param
 6.29% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004473     2
82.97% :     0.003711s :      1: type_inference.infer
17.03% :     0.000762s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.50% :     0.000042s :      4: replace.inline
32.50% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000171     6
93.48% :     0.000160s :      4: match.inline
 6.52% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.49% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.41% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000001s :     10: predicate.addn_zero_filter
 2.12% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.16% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.38% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.82% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.92% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.95% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000002s :     15: predicate.partial_eliminate
 1.28% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.90% :     0.000009s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.27% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.45% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.54% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000527    12
51.09% :     0.000269s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.45% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.46% :     0.000224s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020186    76
 0.25% :     0.000050s :      1: add_recomputation
 0.32% :     0.000064s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.69% :     0.000342s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000023s :      1: event_method
 0.07% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.86% :     0.005623s :      1: jit_opt_a
 0.67% :     0.000136s :      1: jit_opt_after_cconv
 0.25% :     0.000050s :      1: jit_opt_b
 2.09% :     0.000422s :      1: loop_unroll
 2.78% :     0.000562s :      1: mutable_eliminate
 3.95% :     0.000798s :     26: opt.transform.jit_opt_a
 0.22% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.43% :     0.000490s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.53% :     0.000309s :      1: renormalize.infer
 1.24% :     0.000251s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000038s :      1: rewriter_after_opt_a
 0.36% :     0.000073s :      1: rewriter_before_opt_a
 0.37% :     0.000074s :      1: symbol_engine_optimizer
29.36% :     0.005926s :      1: task_emit
22.57% :     0.004556s :      1: type_inference
 0.27% :     0.000055s :      1: validate


TotalTime = 0.013641, [30]
[bootstrap]: 0.00030659
[type_inference]: 0.00450184
[event_method]: 1.666e-05
[auto_monad]: 7.21e-05
[graph_reusing]: 5.85002e-06
[pre_auto_parallel]: 2.76999e-06
[py_interpret_to_execute]: 3.466e-05
[rewriter_before_opt_a]: 7.743e-05
[expand_dump_flag]: 3.13e-06
[jit_opt_a]: 0.00619964, [2]
    [Cycle 1]: 0.00174801, [27]
        [switch_simplify]: 5.533e-05
        [loop_unroll]: 2.883e-05
        [a_1]: 0.00055635
        [with_stream_mark]: 2.241e-05
        [recompute_prepare]: 8.01001e-06
        [updatestate_depend_eliminate]: 4.07e-06
        [updatestate_assign_eliminate]: 3.06001e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 1.77001e-06
        [specialize_transform]: 6.04001e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 5.60001e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 5.04998e-06
        [merge_forward]: 3.68999e-06
        [cell_reuse_recompute_pass]: 1.29003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.512e-05
        [j_node_and_user_rematch]: 9.31002e-06
        [meta_fg_expand]: 2.32001e-06
        [replace_old_param]: 9.00999e-06
        [inline_without_move]: 5.47001e-06
        [renormalize]: 0.000748
        [add_forward_monad_depend]: 7.70998e-06
        [auto_monad_grad]: 2.97002e-06
        [auto_monad_eliminator]: 1.499e-05
        [cse]: 2.864e-05
        [replace_applicator]: 1.455e-05
    [Cycle 2]: 0.00032361, [27]
        [switch_simplify]: 5.62001e-06
        [loop_unroll]: 4.70001e-06
        [a_1]: 8.798e-05
        [with_stream_mark]: 1.125e-05
        [recompute_prepare]: 5.52999e-06
        [updatestate_depend_eliminate]: 2.66999e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 4.90001e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 5.07e-06
        [meta_shard_fg_expand]: 1.50999e-06
        [get_grad_eliminate_]: 4.32e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.73002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.44e-05
        [j_node_and_user_rematch]: 7.65e-06
        [meta_fg_expand]: 1.45001e-06
        [replace_old_param]: 1.105e-05
        [inline_without_move]: 4.47e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.18001e-06
        [auto_monad_grad]: 9.40025e-07
        [auto_monad_eliminator]: 4.95001e-06
        [cse]: 1.069e-05
        [replace_applicator]: 6.13002e-06
[py_interpret_to_execute_after_opt_a]: 1.396e-05
[rewriter_after_opt_a]: 3.482e-05
[convert_after_rewriter]: 7.05002e-06
[order_py_execute_after_rewriter]: 4.75999e-06
[mutable_eliminate]: 0.0007021
[jit_opt_b]: 5.165e-05, [1]
    [Cycle 1]: 4.287e-05, [2]
        [frontend_op_eliminate]: 1.609e-05
        [inline_after_opt_a]: 1.504e-05
[cconv]: 3.269e-05
[loop_unroll]: 0.00048135
[jit_opt_after_cconv]: 0.00015797, [1]
    [Cycle 1]: 0.00015016, [11]
        [c_1]: 2.061e-05
        [parameter_eliminate]: 4.72998e-06
        [updatestate_depend_eliminate]: 8.82e-06
        [updatestate_assign_eliminate]: 3.03e-06
        [updatestate_loads_eliminate]: 2.39001e-06
        [cse]: 2.535e-05
        [call_graph_tuple_transform]: 2.152e-05
        [tuple_list_get_item_eliminator]: 5.24998e-06
        [none_parameter_eliminate]: 1.87001e-06
        [renormalize]: 9.79984e-07
        [switch_simplify]: 5.32999e-06
[remove_dup_value]: 1.575e-05
[partial_unused_args_eliminate]: 2.27001e-06
[environ_conv]: 5.40999e-06
[add_recomputation]: 5.208e-05
[cse_after_recomputation]: 2.138e-05, [1]
    [Cycle 1]: 1.54e-05, [1]
        [cse]: 9.47999e-06
[auto_monad_reorder]: 1.602e-05
[get_jit_bprop_graph]: 2.26998e-06
[rewriter_after_jit_bprop_graph]: 6.96999e-06
[opt_after_jit_grad]: 0.00050823
[symbol_engine_optimizer]: 7.548e-05, [1]
    [Cycle 1]: 6.803e-05, [6]
        [build]: 4.38001e-06
        [elim_shapecalc]: 8.08001e-06
        [elim_not_effective]: 1.377e-05
        [opt_reshape]: 5.44e-06
        [fold_const_symbol]: 8.72e-06
        [renormalize]: 5.50004e-07
[validate]: 4.283e-05
Sums
    bootstrap                            :     0.000307s :  3.44%
    type_inference                       :     0.004502s : 50.52%
    event_method                         :     0.000017s :  0.19%
    auto_monad                           :     0.000072s :  0.81%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.39%
    rewriter_before_opt_a                :     0.000077s :  0.87%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000061s :  0.68%
    jit_opt_a.loop_unroll                :     0.000034s :  0.38%
    jit_opt_a.a_1                        :     0.000644s :  7.23%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.38%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000020s :  0.23%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000748s :  8.40%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.22%
    jit_opt_a.cse                        :     0.000039s :  0.44%
    jit_opt_a.replace_applicator         :     0.000021s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.16%
    rewriter_after_opt_a                 :     0.000035s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000702s :  7.88%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000033s :  0.37%
    loop_unroll                          :     0.000481s :  5.40%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.10%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000025s :  0.28%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000016s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000052s :  0.58%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.08%
    opt_after_jit_grad                   :     0.000508s :  5.70%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.48%


Time group info:
------[substitution.]   0.000226    23
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 2.59% :     0.000006s :      3: substitution.graph_param_transform
83.05% :     0.000188s :      4: substitution.inline
 1.49% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.25% :     0.000005s :      4: substitution.remove_not_recompute_node
 3.68% :     0.000008s :      2: substitution.replace_old_param
 5.53% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004431     2
80.65% :     0.003574s :      1: type_inference.infer
19.35% :     0.000857s :      1: type_inference.specialize
------[replace.]   0.000069     6
66.86% :     0.000046s :      4: replace.inline
33.14% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000197     6
94.21% :     0.000186s :      4: match.inline
 5.79% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.47% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.23% :     0.000003s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.19% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.09% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 6.46% :     0.000008s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.73% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.96% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.22% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.80% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.83% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.02% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.10% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.46% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.76% :     0.000009s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.03% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.07% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.35% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.66% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000600    12
46.30% :     0.000278s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.36% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.35% :     0.000272s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015304    72
 0.36% :     0.000055s :      1: add_recomputation
 0.50% :     0.000076s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 2.08% :     0.000319s :      1: bootstrap
 0.23% :     0.000036s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000021s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
40.54% :     0.006204s :      1: jit_opt_a
 1.05% :     0.000161s :      1: jit_opt_after_cconv
 0.35% :     0.000054s :      1: jit_opt_b
 3.21% :     0.000491s :      1: loop_unroll
 4.67% :     0.000715s :      1: mutable_eliminate
 5.63% :     0.000862s :     26: opt.transform.jit_opt_a
 0.32% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.38% :     0.000518s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000038s :      1: py_interpret_to_execute
 0.11% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000018s :      1: remove_dup_value
 2.65% :     0.000405s :      1: renormalize.infer
 2.18% :     0.000334s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000038s :      1: rewriter_after_opt_a
 0.53% :     0.000081s :      1: rewriter_before_opt_a
 0.51% :     0.000078s :      1: symbol_engine_optimizer
29.54% :     0.004521s :      1: type_inference


TotalTime = 0.0173445, [33]
[bootstrap]: 0.00028378
[type_inference]: 0.00433789
[event_method]: 1.715e-05
[auto_monad]: 6.066e-05
[graph_reusing]: 5.56002e-06
[pre_auto_parallel]: 2.31e-06
[py_interpret_to_execute]: 2.747e-05
[rewriter_before_opt_a]: 7.128e-05
[expand_dump_flag]: 2.99001e-06
[jit_opt_a]: 0.00490944, [2]
    [Cycle 1]: 0.00133072, [27]
        [switch_simplify]: 4.805e-05
        [loop_unroll]: 2.835e-05
        [a_1]: 0.00048774
        [with_stream_mark]: 1.437e-05
        [recompute_prepare]: 7.23e-06
        [updatestate_depend_eliminate]: 3.76999e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [parameter_eliminate]: 1.91998e-06
        [specialize_transform]: 6.12001e-06
        [updatestate_useless_node_eliminater]: 4.79998e-06
        [accelerated_algorithm]: 5.12e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 4.89998e-06
        [merge_forward]: 3.48e-06
        [cell_reuse_recompute_pass]: 1.11002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.359e-05
        [j_node_and_user_rematch]: 8.97e-06
        [meta_fg_expand]: 2.01998e-06
        [replace_old_param]: 8.23999e-06
        [inline_without_move]: 5.02999e-06
        [renormalize]: 0.00045532
        [add_forward_monad_depend]: 5.29e-06
        [auto_monad_grad]: 1.89999e-06
        [auto_monad_eliminator]: 1.245e-05
        [cse]: 2.557e-05
        [replace_applicator]: 1.065e-05
    [Cycle 2]: 0.00031316, [27]
        [switch_simplify]: 5.14e-06
        [loop_unroll]: 4.75001e-06
        [a_1]: 8.547e-05
        [with_stream_mark]: 8.22998e-06
        [recompute_prepare]: 4.80001e-06
        [updatestate_depend_eliminate]: 2.43e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.77999e-06
        [parameter_eliminate]: 9.99979e-07
        [specialize_transform]: 4.68999e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 4.69998e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.39002e-06
        [merge_forward]: 2.27999e-06
        [cell_reuse_recompute_pass]: 1.43002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.139e-05
        [j_node_and_user_rematch]: 7.51001e-06
        [meta_fg_expand]: 1.41998e-06
        [replace_old_param]: 7.03e-06
        [inline_without_move]: 4.43999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 5.19e-06
        [cse]: 1.729e-05
        [replace_applicator]: 5.79e-06
[py_interpret_to_execute_after_opt_a]: 8.49002e-06
[rewriter_after_opt_a]: 3.234e-05
[convert_after_rewriter]: 6.68e-06
[order_py_execute_after_rewriter]: 4.56002e-06
[mutable_eliminate]: 0.00056214
[jit_opt_b]: 4.723e-05, [1]
    [Cycle 1]: 4.066e-05, [2]
        [frontend_op_eliminate]: 1.589e-05
        [inline_after_opt_a]: 1.362e-05
[cconv]: 2.324e-05
[loop_unroll]: 0.00040048
[jit_opt_after_cconv]: 0.00012989, [1]
    [Cycle 1]: 0.00012384, [11]
        [c_1]: 1.99e-05
        [parameter_eliminate]: 2.32001e-06
        [updatestate_depend_eliminate]: 4.97e-06
        [updatestate_assign_eliminate]: 2.20002e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 1.572e-05
        [call_graph_tuple_transform]: 1.71e-05
        [tuple_list_get_item_eliminator]: 4.88001e-06
        [none_parameter_eliminate]: 1.67001e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 4.85999e-06
[remove_dup_value]: 1.373e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 4.50999e-06
[add_recomputation]: 4.441e-05
[cse_after_recomputation]: 1.884e-05, [1]
    [Cycle 1]: 1.36e-05, [1]
        [cse]: 8.22e-06
[auto_monad_reorder]: 1.46e-05
[get_jit_bprop_graph]: 1.40999e-06
[rewriter_after_jit_bprop_graph]: 4.89998e-06
[opt_after_jit_grad]: 0.00043789
[symbol_engine_optimizer]: 6.882e-05, [1]
    [Cycle 1]: 6.263e-05, [6]
        [build]: 3.14001e-06
        [elim_shapecalc]: 7.16001e-06
        [elim_not_effective]: 1.224e-05
        [opt_reshape]: 5.03002e-06
        [fold_const_symbol]: 8.21002e-06
        [renormalize]: 3.80009e-07
[validate]: 2.933e-05
[backend_pass]: 8.49977e-07
[task_emit]: 0.00558712
[execute]: 7.18e-06
Sums
    bootstrap                            :     0.000284s :  2.10%
    type_inference                       :     0.004338s : 32.09%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000061s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000027s :  0.20%
    rewriter_before_opt_a                :     0.000071s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.39%
    jit_opt_a.loop_unroll                :     0.000033s :  0.24%
    jit_opt_a.a_1                        :     0.000573s :  4.24%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000455s :  3.37%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000043s :  0.32%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000032s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000562s :  4.16%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000400s :  2.96%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000044s :  0.33%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000438s :  3.24%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005587s : 41.33%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000178    23
 1.11% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.98% :     0.000005s :      3: substitution.graph_param_transform
82.89% :     0.000147s :      4: substitution.inline
 1.77% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.31% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.02% :     0.000004s :      2: substitution.replace_old_param
 6.25% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004271     2
82.97% :     0.003543s :      1: type_inference.infer
17.03% :     0.000728s :      1: type_inference.specialize
------[replace.]   0.000060     6
67.31% :     0.000040s :      4: replace.inline
32.69% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000155     6
93.54% :     0.000145s :      4: match.inline
 6.46% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.58% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.52% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 2.10% :     0.000002s :     10: predicate.arithmetic_simplify
 1.19% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.04% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.36% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.67% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.21% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.95% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.56% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000000s :      3: predicate.opt_reshape
 2.08% :     0.000002s :     15: predicate.partial_eliminate
 1.28% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.31% :     0.000001s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.83% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.01% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.44% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.26% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.51% :     0.000008s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.99% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000495    12
51.77% :     0.000256s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.80% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.44% :     0.000210s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018669    76
 0.25% :     0.000047s :      1: add_recomputation
 0.34% :     0.000064s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.62% :     0.000302s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.31% :     0.004912s :      1: jit_opt_a
 0.71% :     0.000133s :      1: jit_opt_after_cconv
 0.27% :     0.000050s :      1: jit_opt_b
 2.18% :     0.000408s :      1: loop_unroll
 3.05% :     0.000570s :      1: mutable_eliminate
 4.11% :     0.000767s :     26: opt.transform.jit_opt_a
 0.23% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.39% :     0.000445s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 1.25% :     0.000233s :      1: renormalize.infer
 1.15% :     0.000215s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000035s :      1: rewriter_after_opt_a
 0.40% :     0.000074s :      1: rewriter_before_opt_a
 0.38% :     0.000071s :      1: symbol_engine_optimizer
29.97% :     0.005596s :      1: task_emit
23.30% :     0.004350s :      1: type_inference
 0.25% :     0.000046s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.918.624 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc39d30> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:17.919.743 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0155634, [30]
[bootstrap]: 0.00037554
[type_inference]: 0.0054678
[event_method]: 2.152e-05
[auto_monad]: 7.197e-05
[graph_reusing]: 6.16998e-06
[pre_auto_parallel]: 3.14001e-06
[py_interpret_to_execute]: 4.065e-05
[rewriter_before_opt_a]: 8.042e-05
[expand_dump_flag]: 4.12e-06
[jit_opt_a]: 0.00660109, [2]
    [Cycle 1]: 0.00211706, [27]
        [switch_simplify]: 6.787e-05
        [loop_unroll]: 2.954e-05
        [a_1]: 0.00062597
        [with_stream_mark]: 2.84e-05
        [recompute_prepare]: 1.122e-05
        [updatestate_depend_eliminate]: 4.70001e-06
        [updatestate_assign_eliminate]: 3.35e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [parameter_eliminate]: 2.07999e-06
        [specialize_transform]: 6.56999e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 5.77001e-06
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 5.49998e-06
        [merge_forward]: 3.76001e-06
        [cell_reuse_recompute_pass]: 2.39001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.627e-05
        [j_node_and_user_rematch]: 1.014e-05
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 1.084e-05
        [inline_without_move]: 5.78002e-06
        [renormalize]: 0.00095504
        [add_forward_monad_depend]: 1.018e-05
        [auto_monad_grad]: 3.13998e-06
        [auto_monad_eliminator]: 2.001e-05
        [cse]: 3.119e-05
        [replace_applicator]: 2.103e-05
    [Cycle 2]: 0.00038326, [27]
        [switch_simplify]: 6.39999e-06
        [loop_unroll]: 5.78997e-06
        [a_1]: 9.846e-05
        [with_stream_mark]: 1.472e-05
        [recompute_prepare]: 5.47999e-06
        [updatestate_depend_eliminate]: 3.62002e-06
        [updatestate_assign_eliminate]: 3.10998e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [parameter_eliminate]: 1.52999e-06
        [specialize_transform]: 5.99e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 5.72001e-06
        [meta_shard_fg_expand]: 1.75001e-06
        [get_grad_eliminate_]: 4.95001e-06
        [merge_forward]: 3.3e-06
        [cell_reuse_recompute_pass]: 3.62002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.024e-05
        [j_node_and_user_rematch]: 9.42001e-06
        [meta_fg_expand]: 1.77001e-06
        [replace_old_param]: 9.56e-06
        [inline_without_move]: 4.65001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 3.11001e-06
        [auto_monad_grad]: 1.80001e-06
        [auto_monad_eliminator]: 8.70001e-06
        [cse]: 1.391e-05
        [replace_applicator]: 7.01001e-06
[py_interpret_to_execute_after_opt_a]: 1.553e-05
[rewriter_after_opt_a]: 4.294e-05
[convert_after_rewriter]: 8.50001e-06
[order_py_execute_after_rewriter]: 6.09001e-06
[mutable_eliminate]: 0.0008722
[jit_opt_b]: 6.222e-05, [1]
    [Cycle 1]: 5.206e-05, [2]
        [frontend_op_eliminate]: 1.978e-05
        [inline_after_opt_a]: 1.83e-05
[cconv]: 3.849e-05
[loop_unroll]: 0.00057809
[jit_opt_after_cconv]: 0.00017394, [1]
    [Cycle 1]: 0.00016619, [11]
        [c_1]: 2.36e-05
        [parameter_eliminate]: 5.96e-06
        [updatestate_depend_eliminate]: 8.73001e-06
        [updatestate_assign_eliminate]: 3.06999e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [cse]: 3.129e-05
        [call_graph_tuple_transform]: 2.39e-05
        [tuple_list_get_item_eliminator]: 5.91003e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 7.59988e-07
        [switch_simplify]: 5.62999e-06
[remove_dup_value]: 1.527e-05
[partial_unused_args_eliminate]: 2.29999e-06
[environ_conv]: 6.36e-06
[add_recomputation]: 5.424e-05
[cse_after_recomputation]: 2.229e-05, [1]
    [Cycle 1]: 1.615e-05, [1]
        [cse]: 9.54999e-06
[auto_monad_reorder]: 1.544e-05
[get_jit_bprop_graph]: 3.07002e-06
[rewriter_after_jit_bprop_graph]: 9.41003e-06
[opt_after_jit_grad]: 0.00061266
[symbol_engine_optimizer]: 8.382e-05, [1]
    [Cycle 1]: 7.47e-05, [6]
        [build]: 4.96002e-06
        [elim_shapecalc]: 8.62998e-06
        [elim_not_effective]: 1.615e-05
        [opt_reshape]: 5.97001e-06
        [fold_const_symbol]: 9.10999e-06
        [renormalize]: 1.07998e-06
[validate]: 4.329e-05
Sums
    bootstrap                            :     0.000376s :  3.49%
    type_inference                       :     0.005468s : 50.88%
    event_method                         :     0.000022s :  0.20%
    auto_monad                           :     0.000072s :  0.67%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000041s :  0.38%
    rewriter_before_opt_a                :     0.000080s :  0.75%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000074s :  0.69%
    jit_opt_a.loop_unroll                :     0.000035s :  0.33%
    jit_opt_a.a_1                        :     0.000724s :  6.74%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.40%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000020s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000955s :  8.89%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.12%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000029s :  0.27%
    jit_opt_a.cse                        :     0.000045s :  0.42%
    jit_opt_a.replace_applicator         :     0.000028s :  0.26%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.14%
    rewriter_after_opt_a                 :     0.000043s :  0.40%
    convert_after_rewriter               :     0.000009s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000872s :  8.12%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.17%
    cconv                                :     0.000038s :  0.36%
    loop_unroll                          :     0.000578s :  5.38%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000031s :  0.29%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000015s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000054s :  0.50%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.14%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.09%
    opt_after_jit_grad                   :     0.000613s :  5.70%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.40%


Time group info:
------[substitution.]   0.000278    23
 1.11% :     0.000003s :      2: substitution.elim_not_effective
 0.44% :     0.000001s :      2: substitution.fold_const_symbol
 2.11% :     0.000006s :      3: substitution.graph_param_transform
85.07% :     0.000237s :      4: substitution.inline
 1.51% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.04% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.50% :     0.000007s :      2: substitution.replace_old_param
 5.21% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005381     2
81.40% :     0.004380s :      1: type_inference.infer
18.60% :     0.001001s :      1: type_inference.specialize
------[replace.]   0.000080     6
67.83% :     0.000054s :      4: replace.inline
32.17% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000247     6
94.60% :     0.000234s :      4: match.inline
 5.40% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.55% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.56% :     0.000002s :     10: predicate.addn_zero_filter
 2.29% :     0.000003s :     10: predicate.arithmetic_simplify
 1.27% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.08% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.03% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.22% :     0.000002s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 6.67% :     0.000009s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.83% :     0.000002s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.52% :     0.000002s :     15: predicate.load_eliminater
 1.60% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.11% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 1.86% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.00% :     0.000003s :     15: predicate.partial_eliminate
 1.03% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.40% :     0.000002s :     10: predicate.reduce_eliminate
 1.71% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.69% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.80% :     0.000001s :      6: predicate.replace_old_param
 0.44% :     0.000001s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.23% :     0.000002s :      6: predicate.special_op_eliminate
 1.24% :     0.000002s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.07% :     0.000003s :     16: predicate.switch_defer_inline
 2.01% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.09% :     0.000010s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 0.97% :     0.000001s :     10: predicate.transpose_eliminate
 1.26% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.08% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.59% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.43% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.97% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000739    12
47.30% :     0.000350s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.66% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.04% :     0.000333s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017563    72
 0.33% :     0.000057s :      1: add_recomputation
 0.43% :     0.000076s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 2.25% :     0.000395s :      1: bootstrap
 0.24% :     0.000041s :      1: cconv
 0.08% :     0.000013s :      1: convert_after_rewriter
 0.14% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.16% :     0.000027s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
37.61% :     0.006605s :      1: jit_opt_a
 1.01% :     0.000177s :      1: jit_opt_after_cconv
 0.37% :     0.000065s :      1: jit_opt_b
 3.35% :     0.000588s :      1: loop_unroll
 5.08% :     0.000892s :      1: mutable_eliminate
 5.56% :     0.000977s :     26: opt.transform.jit_opt_a
 0.31% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000031s :      4: opt.transform.jit_opt_b
 0.10% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.13% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.55% :     0.000624s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000006s :      1: pre_auto_parallel
 0.25% :     0.000044s :      1: py_interpret_to_execute
 0.11% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000018s :      1: remove_dup_value
 3.00% :     0.000527s :      1: renormalize.infer
 2.37% :     0.000416s :      1: renormalize.specialize
 0.07% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000047s :      1: rewriter_after_opt_a
 0.48% :     0.000085s :      1: rewriter_before_opt_a
 0.49% :     0.000086s :      1: symbol_engine_optimizer
31.27% :     0.005492s :      1: type_inference


TotalTime = 0.0128444, [30]
[bootstrap]: 0.00034469
[type_inference]: 0.0044809
[event_method]: 1.618e-05
[auto_monad]: 6.284e-05
[graph_reusing]: 5.64e-06
[pre_auto_parallel]: 2.64001e-06
[py_interpret_to_execute]: 2.913e-05
[rewriter_before_opt_a]: 6.707e-05
[expand_dump_flag]: 3.01999e-06
[jit_opt_a]: 0.00553359, [2]
    [Cycle 1]: 0.00154398, [27]
        [switch_simplify]: 4.906e-05
        [loop_unroll]: 2.975e-05
        [a_1]: 0.00050458
        [with_stream_mark]: 1.566e-05
        [recompute_prepare]: 8.17e-06
        [updatestate_depend_eliminate]: 3.11001e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 6.01998e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 5.61e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 5.10001e-06
        [merge_forward]: 3.68e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.397e-05
        [j_node_and_user_rematch]: 8.85999e-06
        [meta_fg_expand]: 2.01e-06
        [replace_old_param]: 8.1e-06
        [inline_without_move]: 5.64e-06
        [renormalize]: 0.00062684
        [add_forward_monad_depend]: 6.81001e-06
        [auto_monad_grad]: 2.60997e-06
        [auto_monad_eliminator]: 1.434e-05
        [cse]: 2.523e-05
        [replace_applicator]: 1.43e-05
    [Cycle 2]: 0.00033174, [27]
        [switch_simplify]: 5.79e-06
        [loop_unroll]: 4.55999e-06
        [a_1]: 9.071e-05
        [with_stream_mark]: 1.11e-05
        [recompute_prepare]: 4.78001e-06
        [updatestate_depend_eliminate]: 3.01001e-06
        [updatestate_assign_eliminate]: 2.38002e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 4.76002e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 4.78001e-06
        [meta_shard_fg_expand]: 1.71998e-06
        [get_grad_eliminate_]: 4.67e-06
        [merge_forward]: 2.70002e-06
        [cell_reuse_recompute_pass]: 2.06e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.367e-05
        [j_node_and_user_rematch]: 8.36002e-06
        [meta_fg_expand]: 1.43002e-06
        [replace_old_param]: 8.68001e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 1.25999e-06
        [auto_monad_eliminator]: 4.99e-06
        [cse]: 1.122e-05
        [replace_applicator]: 5.42999e-06
[py_interpret_to_execute_after_opt_a]: 1.168e-05
[rewriter_after_opt_a]: 3.578e-05
[convert_after_rewriter]: 6.83998e-06
[order_py_execute_after_rewriter]: 4.70001e-06
[mutable_eliminate]: 0.00068913
[jit_opt_b]: 5.315e-05, [1]
    [Cycle 1]: 4.604e-05, [2]
        [frontend_op_eliminate]: 1.928e-05
        [inline_after_opt_a]: 1.509e-05
[cconv]: 2.754e-05
[loop_unroll]: 0.00043195
[jit_opt_after_cconv]: 0.00015328, [1]
    [Cycle 1]: 0.00013484, [11]
        [c_1]: 1.999e-05
        [parameter_eliminate]: 3.66001e-06
        [updatestate_depend_eliminate]: 5.57001e-06
        [updatestate_assign_eliminate]: 2.66999e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [cse]: 1.879e-05
        [call_graph_tuple_transform]: 1.92e-05
        [tuple_list_get_item_eliminator]: 5.77999e-06
        [none_parameter_eliminate]: 1.67999e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 5.19e-06
[remove_dup_value]: 1.43e-05
[partial_unused_args_eliminate]: 3.03e-06
[environ_conv]: 6.19001e-06
[add_recomputation]: 4.777e-05
[cse_after_recomputation]: 2.204e-05, [1]
    [Cycle 1]: 1.575e-05, [1]
        [cse]: 9.72999e-06
[auto_monad_reorder]: 1.584e-05
[get_jit_bprop_graph]: 2.11e-06
[rewriter_after_jit_bprop_graph]: 6.00002e-06
[opt_after_jit_grad]: 0.00048772
[symbol_engine_optimizer]: 7.458e-05, [1]
    [Cycle 1]: 6.724e-05, [6]
        [build]: 3.43999e-06
        [elim_shapecalc]: 7.36001e-06
        [elim_not_effective]: 1.345e-05
        [opt_reshape]: 5.95002e-06
        [fold_const_symbol]: 8.95001e-06
        [renormalize]: 4.00003e-07
[validate]: 3.811e-05
Sums
    bootstrap                            :     0.000345s :  4.01%
    type_inference                       :     0.004481s : 52.12%
    event_method                         :     0.000016s :  0.19%
    auto_monad                           :     0.000063s :  0.73%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000029s :  0.34%
    rewriter_before_opt_a                :     0.000067s :  0.78%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000055s :  0.64%
    jit_opt_a.loop_unroll                :     0.000034s :  0.40%
    jit_opt_a.a_1                        :     0.000595s :  6.92%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.20%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000627s :  7.29%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.22%
    jit_opt_a.cse                        :     0.000036s :  0.42%
    jit_opt_a.replace_applicator         :     0.000020s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.14%
    rewriter_after_opt_a                 :     0.000036s :  0.42%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000689s :  8.02%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000028s :  0.32%
    loop_unroll                          :     0.000432s :  5.02%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000019s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.04%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000048s :  0.56%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000488s :  5.67%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.44%


Time group info:
------[substitution.]   0.000186    23
 1.19% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.79% :     0.000005s :      3: substitution.graph_param_transform
81.99% :     0.000153s :      4: substitution.inline
 1.99% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.56% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.48% :     0.000005s :      2: substitution.replace_old_param
 6.35% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004417     2
80.85% :     0.003571s :      1: type_inference.infer
19.15% :     0.000846s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.00% :     0.000042s :      4: replace.inline
33.00% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000161     6
93.36% :     0.000150s :      4: match.inline
 6.64% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.43% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.21% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.35% :     0.000003s :     10: predicate.arithmetic_simplify
 1.29% :     0.000001s :     10: predicate.cast_eliminate
 0.63% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.11% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.72% :     0.000001s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.76% :     0.000007s :     22: predicate.inline
 1.15% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.95% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.40% :     0.000002s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000002s :     15: predicate.partial_eliminate
 1.49% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.98% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000001s :     10: predicate.reshape_eliminate
 1.54% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.70% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.81% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000009s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.25% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.44% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 2.08% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.86% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000583    12
51.08% :     0.000298s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.94% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.98% :     0.000256s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014335    72
 0.35% :     0.000051s :      1: add_recomputation
 0.46% :     0.000066s :      1: auto_monad
 0.13% :     0.000019s :      1: auto_monad_reorder
 2.51% :     0.000360s :      1: bootstrap
 0.21% :     0.000030s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000024s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.14% :     0.000020s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.62% :     0.005537s :      1: jit_opt_a
 1.10% :     0.000157s :      1: jit_opt_after_cconv
 0.39% :     0.000056s :      1: jit_opt_b
 3.07% :     0.000440s :      1: loop_unroll
 4.87% :     0.000698s :      1: mutable_eliminate
 5.60% :     0.000802s :     26: opt.transform.jit_opt_a
 0.32% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000028s :      4: opt.transform.jit_opt_b
 0.09% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.46% :     0.000497s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.04% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000032s :      1: py_interpret_to_execute
 0.10% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.31% :     0.000331s :      1: renormalize.infer
 2.01% :     0.000289s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000039s :      1: rewriter_after_opt_a
 0.49% :     0.000070s :      1: rewriter_before_opt_a
 0.54% :     0.000077s :      1: symbol_engine_optimizer
31.34% :     0.004493s :      1: type_inference


TotalTime = 0.0211914, [33]
[bootstrap]: 0.0003252
[type_inference]: 0.00530358
[event_method]: 1.917e-05
[auto_monad]: 6.667e-05
[graph_reusing]: 5.94e-06
[pre_auto_parallel]: 2.36e-06
[py_interpret_to_execute]: 4.152e-05
[rewriter_before_opt_a]: 7.222e-05
[expand_dump_flag]: 3.04001e-06
[jit_opt_a]: 0.00588502, [2]
    [Cycle 1]: 0.00189893, [27]
        [switch_simplify]: 6.277e-05
        [loop_unroll]: 2.88e-05
        [a_1]: 0.00055783
        [with_stream_mark]: 1.8e-05
        [recompute_prepare]: 9.01002e-06
        [updatestate_depend_eliminate]: 3.43e-06
        [updatestate_assign_eliminate]: 3.46001e-06
        [updatestate_loads_eliminate]: 2.76e-06
        [parameter_eliminate]: 1.87001e-06
        [specialize_transform]: 5.88002e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 5.75001e-06
        [meta_shard_fg_expand]: 2.61999e-06
        [get_grad_eliminate_]: 6.21e-06
        [merge_forward]: 3.33e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.52e-05
        [j_node_and_user_rematch]: 9.09998e-06
        [meta_fg_expand]: 2.64999e-06
        [replace_old_param]: 1.055e-05
        [inline_without_move]: 5.58997e-06
        [renormalize]: 0.00088266
        [add_forward_monad_depend]: 7.06001e-06
        [auto_monad_grad]: 2.71e-06
        [auto_monad_eliminator]: 1.522e-05
        [cse]: 2.877e-05
        [replace_applicator]: 1.648e-05
    [Cycle 2]: 0.0003605, [27]
        [switch_simplify]: 6.16998e-06
        [loop_unroll]: 4.89998e-06
        [a_1]: 9.195e-05
        [with_stream_mark]: 1.244e-05
        [recompute_prepare]: 5.69e-06
        [updatestate_depend_eliminate]: 2.71e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 4.92999e-06
        [updatestate_useless_node_eliminater]: 4.89998e-06
        [accelerated_algorithm]: 5.20001e-06
        [meta_shard_fg_expand]: 1.77999e-06
        [get_grad_eliminate_]: 4.74002e-06
        [merge_forward]: 3.35003e-06
        [cell_reuse_recompute_pass]: 2.39001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.39e-05
        [j_node_and_user_rematch]: 9.84001e-06
        [meta_fg_expand]: 2.07001e-06
        [replace_old_param]: 8.27998e-06
        [inline_without_move]: 4.97e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 2.12001e-06
        [auto_monad_grad]: 1.72001e-06
        [auto_monad_eliminator]: 7.68001e-06
        [cse]: 1.33e-05
        [replace_applicator]: 7.53e-06
[py_interpret_to_execute_after_opt_a]: 1.473e-05
[rewriter_after_opt_a]: 3.616e-05
[convert_after_rewriter]: 8.54e-06
[order_py_execute_after_rewriter]: 5.42999e-06
[mutable_eliminate]: 0.00068023
[jit_opt_b]: 5.378e-05, [1]
    [Cycle 1]: 4.597e-05, [2]
        [frontend_op_eliminate]: 1.859e-05
        [inline_after_opt_a]: 1.554e-05
[cconv]: 2.884e-05
[loop_unroll]: 0.00049782
[jit_opt_after_cconv]: 0.00015912, [1]
    [Cycle 1]: 0.00015096, [11]
        [c_1]: 2.208e-05
        [parameter_eliminate]: 3.63999e-06
        [updatestate_depend_eliminate]: 6.81001e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 2.86999e-06
        [cse]: 2.538e-05
        [call_graph_tuple_transform]: 2.109e-05
        [tuple_list_get_item_eliminator]: 5.79e-06
        [none_parameter_eliminate]: 1.71e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 5.77999e-06
[remove_dup_value]: 1.5e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 5.59e-06
[add_recomputation]: 4.884e-05
[cse_after_recomputation]: 2.183e-05, [1]
    [Cycle 1]: 1.54e-05, [1]
        [cse]: 9.20999e-06
[auto_monad_reorder]: 1.507e-05
[get_jit_bprop_graph]: 2.03997e-06
[rewriter_after_jit_bprop_graph]: 7.76001e-06
[opt_after_jit_grad]: 0.0005324
[symbol_engine_optimizer]: 8.08e-05, [1]
    [Cycle 1]: 7.288e-05, [6]
        [build]: 4.80999e-06
        [elim_shapecalc]: 8.26002e-06
        [elim_not_effective]: 1.412e-05
        [opt_reshape]: 6.02001e-06
        [fold_const_symbol]: 1.012e-05
        [renormalize]: 8.80013e-07
[validate]: 4.097e-05
[backend_pass]: 1.13001e-06
[task_emit]: 0.006984
[execute]: 7.27002e-06
Sums
    bootstrap                            :     0.000325s :  1.92%
    type_inference                       :     0.005304s : 31.38%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000067s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.25%
    rewriter_before_opt_a                :     0.000072s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000069s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000650s :  3.84%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000883s :  5.22%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.14%
    jit_opt_a.cse                        :     0.000042s :  0.25%
    jit_opt_a.replace_applicator         :     0.000024s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.09%
    rewriter_after_opt_a                 :     0.000036s :  0.21%
    convert_after_rewriter               :     0.000009s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000680s :  4.02%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000029s :  0.17%
    loop_unroll                          :     0.000498s :  2.95%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000025s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000049s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000532s :  3.15%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006984s : 41.32%
    execute                              :     0.000007s :  0.04%


Time group info:
------[substitution.]   0.000218    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 1.02% :     0.000002s :      2: substitution.fold_const_symbol
 2.86% :     0.000006s :      3: substitution.graph_param_transform
82.73% :     0.000180s :      4: substitution.inline
 1.82% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.09% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.48% :     0.000005s :      2: substitution.replace_old_param
 6.13% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005227     2
80.81% :     0.004224s :      1: type_inference.infer
19.19% :     0.001003s :      1: type_inference.specialize
------[replace.]   0.000079     6
65.85% :     0.000052s :      4: replace.inline
34.15% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000190     6
93.55% :     0.000178s :      4: match.inline
 6.45% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.13% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.42% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000002s :     10: predicate.addn_zero_filter
 2.08% :     0.000003s :     10: predicate.arithmetic_simplify
 1.28% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.04% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.73% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.50% :     0.000001s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.38% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.32% :     0.000002s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.77% :     0.000007s :     22: predicate.inline
 1.20% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.78% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.60% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 1.86% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000000s :      3: predicate.opt_reshape
 1.94% :     0.000002s :     15: predicate.partial_eliminate
 1.02% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.31% :     0.000003s :     18: predicate.replace_applicator
 0.95% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.56% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.20% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.66% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.73% :     0.000002s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.17% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.36% :     0.000010s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.56% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.31% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.13% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000694    12
48.29% :     0.000335s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.38% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.33% :     0.000328s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023071    76
 0.22% :     0.000052s :      1: add_recomputation
 0.30% :     0.000070s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.47% :     0.000339s :      1: bootstrap
 0.14% :     0.000032s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.52% :     0.005888s :      1: jit_opt_a
 0.70% :     0.000162s :      1: jit_opt_after_cconv
 0.25% :     0.000057s :      1: jit_opt_b
 2.20% :     0.000508s :      1: loop_unroll
 3.00% :     0.000692s :      1: mutable_eliminate
 3.82% :     0.000881s :     26: opt.transform.jit_opt_a
 0.22% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000028s :      4: opt.transform.jit_opt_b
 0.07% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.35% :     0.000543s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000045s :      1: py_interpret_to_execute
 0.08% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.85% :     0.000427s :      1: renormalize.infer
 1.94% :     0.000447s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000041s :      1: rewriter_after_opt_a
 0.33% :     0.000076s :      1: rewriter_before_opt_a
 0.36% :     0.000083s :      1: symbol_engine_optimizer
30.33% :     0.006997s :      1: task_emit
23.05% :     0.005318s :      1: type_inference
 0.27% :     0.000063s :      1: validate


TotalTime = 0.05337, [33]
[bootstrap]: 0.00036886
[type_inference]: 0.0258744
[event_method]: 6.022e-05
[auto_monad]: 0.00011065
[graph_reusing]: 7.23e-06
[pre_auto_parallel]: 2.99999e-06
[py_interpret_to_execute]: 2.846e-05
[rewriter_before_opt_a]: 0.00010476
[expand_dump_flag]: 3.28998e-06
[jit_opt_a]: 0.0125944, [3]
    [Cycle 1]: 0.00603856, [27]
        [switch_simplify]: 0.00012772
        [loop_unroll]: 4.273e-05
        [a_1]: 0.0008449
        [with_stream_mark]: 2.293e-05
        [recompute_prepare]: 1.794e-05
        [updatestate_depend_eliminate]: 7.65e-06
        [updatestate_assign_eliminate]: 6.34999e-06
        [updatestate_loads_eliminate]: 5.63002e-06
        [parameter_eliminate]: 2.79001e-06
        [specialize_transform]: 1.277e-05
        [updatestate_useless_node_eliminater]: 1.173e-05
        [accelerated_algorithm]: 1.187e-05
        [meta_shard_fg_expand]: 3.65998e-06
        [get_grad_eliminate_]: 1.181e-05
        [merge_forward]: 7.88001e-06
        [cell_reuse_recompute_pass]: 8.89995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.527e-05
        [j_node_and_user_rematch]: 2.341e-05
        [meta_fg_expand]: 0.00175288
        [replace_old_param]: 5.196e-05
        [inline_without_move]: 5.026e-05
        [renormalize]: 0.00250408
        [add_forward_monad_depend]: 1.66e-05
        [auto_monad_grad]: 6.59001e-06
        [auto_monad_eliminator]: 4.941e-05
        [cse]: 0.00015019
        [replace_applicator]: 6.633e-05
    [Cycle 2]: 0.00211515, [27]
        [switch_simplify]: 3.527e-05
        [loop_unroll]: 3.362e-05
        [a_1]: 0.00105144
        [with_stream_mark]: 1.477e-05
        [recompute_prepare]: 8.09002e-06
        [updatestate_depend_eliminate]: 3.49001e-06
        [updatestate_assign_eliminate]: 3.33e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 1.51998e-06
        [specialize_transform]: 6.44001e-06
        [updatestate_useless_node_eliminater]: 5.71e-06
        [accelerated_algorithm]: 5.93002e-06
        [meta_shard_fg_expand]: 1.83002e-06
        [get_grad_eliminate_]: 6.39999e-06
        [merge_forward]: 2.98e-06
        [cell_reuse_recompute_pass]: 1.00999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.501e-05
        [j_node_and_user_rematch]: 9.87001e-06
        [meta_fg_expand]: 0.00010802
        [replace_old_param]: 1.058e-05
        [inline_without_move]: 6.23002e-06
        [renormalize]: 0.00057501
        [add_forward_monad_depend]: 4.53999e-06
        [auto_monad_grad]: 1.36998e-06
        [auto_monad_eliminator]: 1.26e-05
        [cse]: 2.336e-05
        [replace_applicator]: 1.316e-05
    [Cycle 3]: 0.00036147, [27]
        [switch_simplify]: 6.29001e-06
        [loop_unroll]: 5.64e-06
        [a_1]: 0.00011719
        [with_stream_mark]: 9.00999e-06
        [recompute_prepare]: 6.10002e-06
        [updatestate_depend_eliminate]: 3.24001e-06
        [updatestate_assign_eliminate]: 2.63998e-06
        [updatestate_loads_eliminate]: 2.66e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 5.95002e-06
        [updatestate_useless_node_eliminater]: 5.99e-06
        [accelerated_algorithm]: 5.74999e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 5.54e-06
        [merge_forward]: 2.93998e-06
        [cell_reuse_recompute_pass]: 1.42e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.544e-05
        [j_node_and_user_rematch]: 1.029e-05
        [meta_fg_expand]: 1.87999e-06
        [replace_old_param]: 8.11002e-06
        [inline_without_move]: 5.62999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 9.39996e-07
        [auto_monad_grad]: 6.09987e-07
        [auto_monad_eliminator]: 6.07999e-06
        [cse]: 1.327e-05
        [replace_applicator]: 6.12001e-06
[py_interpret_to_execute_after_opt_a]: 1.237e-05
[rewriter_after_opt_a]: 3.783e-05
[convert_after_rewriter]: 8.03999e-06
[order_py_execute_after_rewriter]: 5.46e-06
[mutable_eliminate]: 0.00060098
[jit_opt_b]: 5.648e-05, [1]
    [Cycle 1]: 4.764e-05, [2]
        [frontend_op_eliminate]: 1.854e-05
        [inline_after_opt_a]: 1.73e-05
[cconv]: 2.454e-05
[loop_unroll]: 0.00042335
[jit_opt_after_cconv]: 0.0001551, [1]
    [Cycle 1]: 0.00014814, [11]
        [c_1]: 2.581e-05
        [parameter_eliminate]: 2.59999e-06
        [updatestate_depend_eliminate]: 6.11e-06
        [updatestate_assign_eliminate]: 2.79999e-06
        [updatestate_loads_eliminate]: 3.05998e-06
        [cse]: 2.311e-05
        [call_graph_tuple_transform]: 2.109e-05
        [tuple_list_get_item_eliminator]: 6.09999e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 6.39993e-07
        [switch_simplify]: 6.24999e-06
[remove_dup_value]: 1.628e-05
[partial_unused_args_eliminate]: 2.22001e-06
[environ_conv]: 5.56e-06
[add_recomputation]: 7.069e-05
[cse_after_recomputation]: 2.662e-05, [1]
    [Cycle 1]: 2.08e-05, [1]
        [cse]: 1.44e-05
[auto_monad_reorder]: 1.849e-05
[get_jit_bprop_graph]: 1.52999e-06
[rewriter_after_jit_bprop_graph]: 6.12999e-06
[opt_after_jit_grad]: 0.00047706
[symbol_engine_optimizer]: 0.00013892, [1]
    [Cycle 1]: 0.00013203, [6]
        [build]: 3.98001e-06
        [elim_shapecalc]: 1.004e-05
        [elim_not_effective]: 1.55e-05
        [opt_reshape]: 7.68001e-06
        [fold_const_symbol]: 1.175e-05
        [renormalize]: 6.50005e-07
[validate]: 4.534e-05
[backend_pass]: 1.03001e-06
[task_emit]: 0.0118172
[execute]: 9.89001e-06
Sums
    bootstrap                            :     0.000369s :  0.76%
    type_inference                       :     0.025874s : 53.46%
    event_method                         :     0.000060s :  0.12%
    auto_monad                           :     0.000111s :  0.23%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.06%
    rewriter_before_opt_a                :     0.000105s :  0.22%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000169s :  0.35%
    jit_opt_a.loop_unroll                :     0.000082s :  0.17%
    jit_opt_a.a_1                        :     0.002014s :  4.16%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000025s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.05%
    jit_opt_a.merge_forward              :     0.000014s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000056s :  0.12%
    jit_opt_a.j_node_and_user_rematch    :     0.000044s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.001863s :  3.85%
    jit_opt_a.replace_old_param          :     0.000071s :  0.15%
    jit_opt_a.inline_without_move        :     0.000062s :  0.13%
    jit_opt_a.renormalize                :     0.003079s :  6.36%
    jit_opt_a.add_forward_monad_depend   :     0.000022s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000068s :  0.14%
    jit_opt_a.cse                        :     0.000187s :  0.39%
    jit_opt_a.replace_applicator         :     0.000086s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.03%
    rewriter_after_opt_a                 :     0.000038s :  0.08%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000601s :  1.24%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.04%
    cconv                                :     0.000025s :  0.05%
    loop_unroll                          :     0.000423s :  0.87%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.05%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.05%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.01%
    remove_dup_value                     :     0.000016s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000006s :  0.01%
    add_recomputation                    :     0.000071s :  0.15%
    cse_after_recomputation.cse          :     0.000014s :  0.03%
    auto_monad_reorder                   :     0.000018s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000477s :  0.99%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.03%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.09%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011817s : 24.42%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000485   113
 0.49% :     0.000002s :      3: substitution.elim_not_effective
 0.42% :     0.000002s :      3: substitution.fold_const_symbol
 1.21% :     0.000006s :      4: substitution.graph_param_transform
63.04% :     0.000305s :     14: substitution.inline
 3.14% :     0.000015s :      2: substitution.inline_without_move
 2.06% :     0.000010s :     14: substitution.j_node_and_user_rematch
 2.20% :     0.000011s :      7: substitution.minmaximum_grad
 1.34% :     0.000006s :      5: substitution.partial_eliminate
 2.32% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.23% :     0.000025s :     10: substitution.replace_applicator
 1.55% :     0.000007s :      7: substitution.replace_old_param
 0.47% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.61% :     0.000008s :      1: substitution.switch_simplify
 4.26% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.80% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.86% :     0.000038s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.025775     2
93.72% :     0.024156s :      1: type_inference.infer
 6.28% :     0.001619s :      1: type_inference.specialize
------[replace.]   0.000178    22
54.10% :     0.000096s :     14: replace.inline
17.54% :     0.000031s :      1: replace.switch_simplify
28.36% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000325    22
91.66% :     0.000298s :     14: match.inline
 2.19% :     0.000007s :      1: match.switch_simplify
 6.15% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000368  2634
 1.63% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.43% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.42% :     0.000005s :     44: predicate.addn_check_dump
 1.85% :     0.000007s :     44: predicate.addn_zero_filter
 2.04% :     0.000008s :     44: predicate.arithmetic_simplify
 1.51% :     0.000006s :     44: predicate.cast_eliminate
 0.16% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.46% :     0.000005s :     44: predicate.compare_switch_simplify
 1.53% :     0.000006s :     44: predicate.depend_value_elim
 1.47% :     0.000005s :     44: predicate.dict_get_item_const_eliminator
 1.48% :     0.000005s :     44: predicate.dict_get_item_eliminator
 1.48% :     0.000005s :     44: predicate.dict_set_item_eliminator
 0.37% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      4: predicate.elim_not_effective
 0.23% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.48% :     0.000005s :     44: predicate.environ_add_const_eliminate
 1.46% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.46% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.50% :     0.000006s :     44: predicate.environ_get_eliminate
 1.43% :     0.000005s :     44: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.83% :     0.000003s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.30% :     0.000016s :     73: predicate.inline
 1.78% :     0.000007s :     40: predicate.inline_without_move
 0.36% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.85% :     0.000003s :     19: predicate.less_batch_normalization
 1.86% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.90% :     0.000007s :     55: predicate.load_eliminater
 0.52% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.32% :     0.000012s :     91: predicate.loop_unroll_before_grad
 1.87% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.44% :     0.000005s :     44: predicate.merge_addn
 1.49% :     0.000006s :     44: predicate.minmaximum_grad
 0.61% :     0.000002s :      4: predicate.mutable_eliminate
 0.21% :     0.000001s :      4: predicate.opt_reshape
 2.40% :     0.000009s :     55: predicate.partial_eliminate
 1.53% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.88% :     0.000007s :     44: predicate.reduce_eliminate
 1.80% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.58% :     0.000009s :     97: predicate.replace_applicator
 0.88% :     0.000003s :     40: predicate.replace_old_param
 0.11% :     0.000000s :      4: predicate.reset_defer_inline
 1.81% :     0.000007s :     44: predicate.reshape_eliminate
 1.53% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.27% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.52% :     0.000006s :     44: predicate.same_eliminate
 0.47% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.36% :     0.000001s :      8: predicate.special_op_eliminate
 0.86% :     0.000003s :     19: predicate.specialize_transform
 1.68% :     0.000006s :     44: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000005s :     44: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.80% :     0.000010s :     65: predicate.switch_defer_inline
 2.69% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.59% :     0.000024s :    162: predicate.switch_simplify
 1.46% :     0.000005s :     44: predicate.tile_eliminate
 1.47% :     0.000005s :     44: predicate.transpose_eliminate
 1.82% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.70% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 3.10% :     0.000011s :     59: predicate.tuple_list_get_item_eliminator
 1.92% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000006s :     51: predicate.tuple_to_list_eliminator_
 1.90% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000007s :     44: predicate.value_based_eliminate
 0.14% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.21% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.002003    32
56.94% :     0.001141s :     14: func_graph_cloner_run.FuncGraphClonerGraph
43.06% :     0.000863s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.059230    91
 0.12% :     0.000074s :      1: add_recomputation
 0.19% :     0.000115s :      1: auto_monad
 0.04% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.67% :     0.000398s :      1: bootstrap
 0.05% :     0.000027s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.05% :     0.000029s :      1: cse_after_recomputation
 0.01% :     0.000008s :      1: environ_conv
 0.11% :     0.000066s :      1: event_method
 0.03% :     0.000016s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
21.27% :     0.012598s :      1: jit_opt_a
 0.27% :     0.000158s :      1: jit_opt_after_cconv
 0.10% :     0.000059s :      1: jit_opt_b
 0.73% :     0.000431s :      1: loop_unroll
 1.03% :     0.000609s :      1: mutable_eliminate
 4.50% :     0.002666s :     39: opt.transform.jit_opt_a
 0.09% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000030s :      4: opt.transform.jit_opt_b
 0.02% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000041s :      4: opt.transform.symbol_engine_opt
 0.82% :     0.000486s :      1: opt_after_jit_grad
 0.01% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.05% :     0.000031s :      1: py_interpret_to_execute
 0.02% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000019s :      1: remove_dup_value
 2.66% :     0.001578s :      2: renormalize.infer
 2.51% :     0.001486s :      2: renormalize.specialize
 0.01% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000041s :      1: rewriter_after_opt_a
 0.18% :     0.000107s :      1: rewriter_before_opt_a
 0.24% :     0.000142s :      1: symbol_engine_optimizer
19.99% :     0.011838s :      1: task_emit
43.71% :     0.025890s :      1: type_inference
 0.15% :     0.000086s :      1: validate


TotalTime = 0.0149289, [30]
[bootstrap]: 0.00029074
[type_inference]: 0.00448059
[event_method]: 1.768e-05
[auto_monad]: 6.155e-05
[graph_reusing]: 6.69001e-06
[pre_auto_parallel]: 2.71999e-06
[py_interpret_to_execute]: 3.267e-05
[rewriter_before_opt_a]: 7.322e-05
[expand_dump_flag]: 3.48999e-06
[jit_opt_a]: 0.00750961, [2]
    [Cycle 1]: 0.00177743, [27]
        [switch_simplify]: 6.428e-05
        [loop_unroll]: 2.931e-05
        [a_1]: 0.00055956
        [with_stream_mark]: 2.179e-05
        [recompute_prepare]: 7.24001e-06
        [updatestate_depend_eliminate]: 4.2e-06
        [updatestate_assign_eliminate]: 3.61999e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 2.51e-06
        [specialize_transform]: 5.56998e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 5.79999e-06
        [meta_shard_fg_expand]: 1.99999e-06
        [get_grad_eliminate_]: 5.47999e-06
        [merge_forward]: 3.50998e-06
        [cell_reuse_recompute_pass]: 1.38002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.382e-05
        [j_node_and_user_rematch]: 9.23002e-06
        [meta_fg_expand]: 2.27999e-06
        [replace_old_param]: 8.93002e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.00076943
        [add_forward_monad_depend]: 7.18998e-06
        [auto_monad_grad]: 2.91e-06
        [auto_monad_eliminator]: 1.472e-05
        [cse]: 2.999e-05
        [replace_applicator]: 1.344e-05
    [Cycle 2]: 0.0003278, [27]
        [switch_simplify]: 5.70001e-06
        [loop_unroll]: 4.84e-06
        [a_1]: 9.231e-05
        [with_stream_mark]: 1.163e-05
        [recompute_prepare]: 5.24e-06
        [updatestate_depend_eliminate]: 2.86e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.37999e-06
        [specialize_transform]: 4.91002e-06
        [updatestate_useless_node_eliminater]: 5.02999e-06
        [accelerated_algorithm]: 5.92001e-06
        [meta_shard_fg_expand]: 1.64e-06
        [get_grad_eliminate_]: 4.58001e-06
        [merge_forward]: 2.51e-06
        [cell_reuse_recompute_pass]: 2.63e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.315e-05
        [j_node_and_user_rematch]: 7.33e-06
        [meta_fg_expand]: 1.55999e-06
        [replace_old_param]: 7.8e-06
        [inline_without_move]: 4.89e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.35001e-06
        [auto_monad_grad]: 1.11997e-06
        [auto_monad_eliminator]: 4.50001e-06
        [cse]: 1.222e-05
        [replace_applicator]: 4.97e-06
[py_interpret_to_execute_after_opt_a]: 1.375e-05
[rewriter_after_opt_a]: 4.038e-05
[convert_after_rewriter]: 7.61001e-06
[order_py_execute_after_rewriter]: 5.43002e-06
[mutable_eliminate]: 0.00080822
[jit_opt_b]: 5.288e-05, [1]
    [Cycle 1]: 4.505e-05, [2]
        [frontend_op_eliminate]: 1.717e-05
        [inline_after_opt_a]: 1.485e-05
[cconv]: 3.097e-05
[loop_unroll]: 0.00045994
[jit_opt_after_cconv]: 0.00013782, [1]
    [Cycle 1]: 0.00013124, [11]
        [c_1]: 1.944e-05
        [parameter_eliminate]: 3.13e-06
        [updatestate_depend_eliminate]: 5.04998e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.43002e-06
        [cse]: 1.828e-05
        [call_graph_tuple_transform]: 1.9e-05
        [tuple_list_get_item_eliminator]: 5.04998e-06
        [none_parameter_eliminate]: 1.50999e-06
        [renormalize]: 7.59988e-07
        [switch_simplify]: 5.15001e-06
[remove_dup_value]: 1.526e-05
[partial_unused_args_eliminate]: 2.40002e-06
[environ_conv]: 5.48002e-06
[add_recomputation]: 5.103e-05
[cse_after_recomputation]: 2.129e-05, [1]
    [Cycle 1]: 1.513e-05, [1]
        [cse]: 8.84e-06
[auto_monad_reorder]: 1.479e-05
[get_jit_bprop_graph]: 1.87001e-06
[rewriter_after_jit_bprop_graph]: 8.08001e-06
[opt_after_jit_grad]: 0.00047308
[symbol_engine_optimizer]: 7.228e-05, [1]
    [Cycle 1]: 6.574e-05, [6]
        [build]: 3.02002e-06
        [elim_shapecalc]: 7.80998e-06
        [elim_not_effective]: 1.235e-05
        [opt_reshape]: 6.04999e-06
        [fold_const_symbol]: 8.74998e-06
        [renormalize]: 6.10016e-07
[validate]: 3.994e-05
Sums
    bootstrap                            :     0.000291s :  3.26%
    type_inference                       :     0.004481s : 50.21%
    event_method                         :     0.000018s :  0.20%
    auto_monad                           :     0.000062s :  0.69%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.37%
    rewriter_before_opt_a                :     0.000073s :  0.82%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000070s :  0.78%
    jit_opt_a.loop_unroll                :     0.000034s :  0.38%
    jit_opt_a.a_1                        :     0.000652s :  7.30%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.37%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000010s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000770s :  8.62%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.22%
    jit_opt_a.cse                        :     0.000042s :  0.47%
    jit_opt_a.replace_applicator         :     0.000018s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.15%
    rewriter_after_opt_a                 :     0.000040s :  0.45%
    convert_after_rewriter               :     0.000008s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000808s :  9.06%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000031s :  0.35%
    loop_unroll                          :     0.000460s :  5.15%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000018s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000051s :  0.57%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.09%
    opt_after_jit_grad                   :     0.000473s :  5.30%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.45%


Time group info:
------[substitution.]   0.000229    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000002s :      2: substitution.fold_const_symbol
 2.45% :     0.000006s :      3: substitution.graph_param_transform
85.03% :     0.000195s :      4: substitution.inline
 1.46% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.05% :     0.000005s :      2: substitution.replace_old_param
 5.40% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004413     2
81.24% :     0.003585s :      1: type_inference.infer
18.76% :     0.000828s :      1: type_inference.specialize
------[replace.]   0.000067     6
67.72% :     0.000045s :      4: replace.inline
32.28% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000204     6
94.44% :     0.000193s :      4: match.inline
 5.56% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.20% :     0.000001s :     10: predicate.accumulaten_eliminater
 0.95% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 2.19% :     0.000003s :     10: predicate.arithmetic_simplify
 1.29% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.67% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.39% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.08% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 6.04% :     0.000007s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.64% :     0.000002s :      6: predicate.less_batch_normalization
 2.11% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.95% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 1.60% :     0.000002s :      3: predicate.mutable_eliminate
 0.75% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.90% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.05% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.61% :     0.000002s :     10: predicate.reshape_eliminate
 1.38% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.23% :     0.000009s :     53: predicate.switch_simplify
 1.47% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.55% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 2.15% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.64% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.89% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000587    12
48.39% :     0.000284s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.23% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.38% :     0.000254s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016618    72
 0.32% :     0.000054s :      1: add_recomputation
 0.39% :     0.000065s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 1.82% :     0.000302s :      1: bootstrap
 0.21% :     0.000034s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.13% :     0.000022s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
45.29% :     0.007526s :      1: jit_opt_a
 0.85% :     0.000140s :      1: jit_opt_after_cconv
 0.34% :     0.000056s :      1: jit_opt_b
 2.82% :     0.000468s :      1: loop_unroll
 4.93% :     0.000819s :      1: mutable_eliminate
 5.24% :     0.000871s :     26: opt.transform.jit_opt_a
 0.27% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.89% :     0.000481s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000035s :      1: py_interpret_to_execute
 0.10% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.55% :     0.000424s :      1: renormalize.infer
 2.03% :     0.000338s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000043s :      1: rewriter_after_opt_a
 0.46% :     0.000076s :      1: rewriter_before_opt_a
 0.45% :     0.000075s :      1: symbol_engine_optimizer
27.07% :     0.004499s :      1: type_inference


TotalTime = 0.0180696, [33]
[bootstrap]: 0.00030364
[type_inference]: 0.00454335
[event_method]: 1.724e-05
[auto_monad]: 6.064e-05
[graph_reusing]: 5.41998e-06
[pre_auto_parallel]: 2.43e-06
[py_interpret_to_execute]: 2.954e-05
[rewriter_before_opt_a]: 7.053e-05
[expand_dump_flag]: 3.07002e-06
[jit_opt_a]: 0.00522038, [2]
    [Cycle 1]: 0.00147957, [27]
        [switch_simplify]: 5.067e-05
        [loop_unroll]: 2.825e-05
        [a_1]: 0.00050502
        [with_stream_mark]: 1.837e-05
        [recompute_prepare]: 7.1e-06
        [updatestate_depend_eliminate]: 3.92002e-06
        [updatestate_assign_eliminate]: 3.16001e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 6.07999e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 5.24e-06
        [meta_shard_fg_expand]: 2.75002e-06
        [get_grad_eliminate_]: 5.34998e-06
        [merge_forward]: 3.58e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.398e-05
        [j_node_and_user_rematch]: 8.94e-06
        [meta_fg_expand]: 2.11e-06
        [replace_old_param]: 8.50001e-06
        [inline_without_move]: 5.32999e-06
        [renormalize]: 0.00053303
        [add_forward_monad_depend]: 5.69e-06
        [auto_monad_grad]: 1.94999e-06
        [auto_monad_eliminator]: 1.528e-05
        [cse]: 2.664e-05
        [replace_applicator]: 1.123e-05
    [Cycle 2]: 0.00030582, [27]
        [switch_simplify]: 5.17999e-06
        [loop_unroll]: 4.60999e-06
        [a_1]: 8.685e-05
        [with_stream_mark]: 8.87e-06
        [recompute_prepare]: 4.68001e-06
        [updatestate_depend_eliminate]: 2.34999e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.14998e-06
        [specialize_transform]: 4.65001e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 4.62e-06
        [meta_shard_fg_expand]: 1.03001e-06
        [get_grad_eliminate_]: 4.34002e-06
        [merge_forward]: 2.73e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.166e-05
        [j_node_and_user_rematch]: 7.31999e-06
        [meta_fg_expand]: 1.39e-06
        [replace_old_param]: 6.63e-06
        [inline_without_move]: 4.57998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.24998e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 4.66002e-06
        [cse]: 9.24e-06
        [replace_applicator]: 4.79998e-06
[py_interpret_to_execute_after_opt_a]: 1.045e-05
[rewriter_after_opt_a]: 3.228e-05
[convert_after_rewriter]: 6.46e-06
[order_py_execute_after_rewriter]: 4.79998e-06
[mutable_eliminate]: 0.00051471
[jit_opt_b]: 4.839e-05, [1]
    [Cycle 1]: 4.081e-05, [2]
        [frontend_op_eliminate]: 1.552e-05
        [inline_after_opt_a]: 1.388e-05
[cconv]: 2.702e-05
[loop_unroll]: 0.00042203
[jit_opt_after_cconv]: 0.00013427, [1]
    [Cycle 1]: 0.00012783, [11]
        [c_1]: 1.986e-05
        [parameter_eliminate]: 2.76e-06
        [updatestate_depend_eliminate]: 4.98001e-06
        [updatestate_assign_eliminate]: 2.23998e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [cse]: 1.701e-05
        [call_graph_tuple_transform]: 1.793e-05
        [tuple_list_get_item_eliminator]: 5.03002e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.15001e-06
[remove_dup_value]: 1.257e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 4.99e-06
[add_recomputation]: 4.837e-05
[cse_after_recomputation]: 2.227e-05, [1]
    [Cycle 1]: 1.599e-05, [1]
        [cse]: 9.34998e-06
[auto_monad_reorder]: 1.454e-05
[get_jit_bprop_graph]: 1.72001e-06
[rewriter_after_jit_bprop_graph]: 4.73001e-06
[opt_after_jit_grad]: 0.00044489
[symbol_engine_optimizer]: 6.794e-05, [1]
    [Cycle 1]: 6.191e-05, [6]
        [build]: 3.41999e-06
        [elim_shapecalc]: 6.63e-06
        [elim_not_effective]: 1.153e-05
        [opt_reshape]: 5.34e-06
        [fold_const_symbol]: 7.80998e-06
        [renormalize]: 5.00004e-07
[validate]: 3.298e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00576468
[execute]: 7.54002e-06
Sums
    bootstrap                            :     0.000304s :  2.17%
    type_inference                       :     0.004543s : 32.40%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000061s :  0.43%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.21%
    rewriter_before_opt_a                :     0.000071s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.23%
    jit_opt_a.a_1                        :     0.000592s :  4.22%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000533s :  3.80%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.14%
    jit_opt_a.cse                        :     0.000036s :  0.26%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000515s :  3.67%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000027s :  0.19%
    loop_unroll                          :     0.000422s :  3.01%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000048s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000445s :  3.17%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000033s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005765s : 41.11%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000191    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.81% :     0.000005s :      3: substitution.graph_param_transform
83.87% :     0.000160s :      4: substitution.inline
 1.70% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.36% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.77% :     0.000003s :      2: substitution.replace_old_param
 5.99% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004480     2
83.14% :     0.003725s :      1: type_inference.infer
16.86% :     0.000755s :      1: type_inference.specialize
------[replace.]   0.000061     6
68.09% :     0.000042s :      4: replace.inline
31.91% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000168     6
93.85% :     0.000158s :      4: match.inline
 6.15% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 1.89% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000001s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 5.70% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.28% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.90% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.66% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.28% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.79% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.58% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.01% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.13% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.27% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.44% :     0.000009s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.06% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000541    12
52.05% :     0.000282s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.57% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.38% :     0.000224s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019491    76
 0.26% :     0.000051s :      1: add_recomputation
 0.33% :     0.000064s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.64% :     0.000319s :      1: bootstrap
 0.15% :     0.000030s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.80% :     0.005224s :      1: jit_opt_a
 0.70% :     0.000137s :      1: jit_opt_after_cconv
 0.26% :     0.000051s :      1: jit_opt_b
 2.20% :     0.000429s :      1: loop_unroll
 2.69% :     0.000524s :      1: mutable_eliminate
 4.05% :     0.000789s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000028s :      4: opt.transform.symbol_engine_opt
 2.32% :     0.000453s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000032s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.49% :     0.000289s :      1: renormalize.infer
 1.21% :     0.000236s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000035s :      1: rewriter_after_opt_a
 0.38% :     0.000073s :      1: rewriter_before_opt_a
 0.36% :     0.000070s :      1: symbol_engine_optimizer
29.63% :     0.005775s :      1: task_emit
23.39% :     0.004558s :      1: type_inference
 0.27% :     0.000053s :      1: validate


TotalTime = 0.0314879, [33]
[bootstrap]: 0.0003088
[type_inference]: 0.00651819
[event_method]: 2.355e-05
[auto_monad]: 5.146e-05
[graph_reusing]: 4.37e-06
[pre_auto_parallel]: 2.23998e-06
[py_interpret_to_execute]: 4.077e-05
[rewriter_before_opt_a]: 6.62e-05
[expand_dump_flag]: 3.33998e-06
[jit_opt_a]: 0.00667732, [2]
    [Cycle 1]: 0.00180676, [27]
        [switch_simplify]: 5.777e-05
        [loop_unroll]: 2.873e-05
        [a_1]: 0.00054091
        [with_stream_mark]: 1.696e-05
        [recompute_prepare]: 8.32e-06
        [updatestate_depend_eliminate]: 3.30003e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.08002e-06
        [parameter_eliminate]: 1.54998e-06
        [specialize_transform]: 6.36e-06
        [updatestate_useless_node_eliminater]: 5.31998e-06
        [accelerated_algorithm]: 5.91998e-06
        [meta_shard_fg_expand]: 1.59e-06
        [get_grad_eliminate_]: 5.23002e-06
        [merge_forward]: 2.59001e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.138e-05
        [j_node_and_user_rematch]: 9.94999e-06
        [meta_fg_expand]: 2.13998e-06
        [replace_old_param]: 9.63002e-06
        [inline_without_move]: 5.77999e-06
        [renormalize]: 0.00084342
        [add_forward_monad_depend]: 5.41998e-06
        [auto_monad_grad]: 2.09e-06
        [auto_monad_eliminator]: 1.142e-05
        [cse]: 2.126e-05
        [replace_applicator]: 1.497e-05
    [Cycle 2]: 0.0003398, [27]
        [switch_simplify]: 5.54998e-06
        [loop_unroll]: 5.12999e-06
        [a_1]: 9.623e-05
        [with_stream_mark]: 9.94001e-06
        [recompute_prepare]: 5.23002e-06
        [updatestate_depend_eliminate]: 2.76e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 1.64e-06
        [get_grad_eliminate_]: 4.98001e-06
        [merge_forward]: 2.34999e-06
        [cell_reuse_recompute_pass]: 2.94999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.444e-05
        [j_node_and_user_rematch]: 7.88001e-06
        [meta_fg_expand]: 1.86e-06
        [replace_old_param]: 7.45998e-06
        [inline_without_move]: 4.67998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.64998e-06
        [auto_monad_grad]: 7.10017e-07
        [auto_monad_eliminator]: 5.69e-06
        [cse]: 9.92001e-06
        [replace_applicator]: 5.76e-06
[py_interpret_to_execute_after_opt_a]: 1.276e-05
[rewriter_after_opt_a]: 2.916e-05
[convert_after_rewriter]: 5.57999e-06
[order_py_execute_after_rewriter]: 5.05001e-06
[mutable_eliminate]: 0.00078938
[jit_opt_b]: 5.487e-05, [1]
    [Cycle 1]: 4.67e-05, [2]
        [frontend_op_eliminate]: 1.852e-05
        [inline_after_opt_a]: 1.55e-05
[cconv]: 2.044e-05
[loop_unroll]: 0.00045889
[jit_opt_after_cconv]: 0.00014847, [1]
    [Cycle 1]: 0.00014079, [11]
        [c_1]: 2.196e-05
        [parameter_eliminate]: 2.89999e-06
        [updatestate_depend_eliminate]: 6.40002e-06
        [updatestate_assign_eliminate]: 2.39999e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 2.152e-05
        [call_graph_tuple_transform]: 1.961e-05
        [tuple_list_get_item_eliminator]: 5.96e-06
        [none_parameter_eliminate]: 9.09989e-07
        [renormalize]: 4.39992e-07
        [switch_simplify]: 6.19001e-06
[remove_dup_value]: 9.74999e-06
[partial_unused_args_eliminate]: 1.89999e-06
[environ_conv]: 3.50003e-06
[add_recomputation]: 3.591e-05
[cse_after_recomputation]: 2.006e-05, [1]
    [Cycle 1]: 1.461e-05, [1]
        [cse]: 8.75001e-06
[auto_monad_reorder]: 1.1e-05
[get_jit_bprop_graph]: 2.18998e-06
[rewriter_after_jit_bprop_graph]: 7.72002e-06
[opt_after_jit_grad]: 0.00048201
[symbol_engine_optimizer]: 7.778e-05, [1]
    [Cycle 1]: 6.968e-05, [6]
        [build]: 3.24001e-06
        [elim_shapecalc]: 7.88001e-06
        [elim_not_effective]: 1.357e-05
        [opt_reshape]: 6.29001e-06
        [fold_const_symbol]: 9.46998e-06
        [renormalize]: 5.8001e-07
[validate]: 3.242e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.0153513
[execute]: 8.97e-06
Sums
    bootstrap                            :     0.000309s :  1.17%
    type_inference                       :     0.006518s : 24.78%
    event_method                         :     0.000024s :  0.09%
    auto_monad                           :     0.000051s :  0.20%
    graph_reusing                        :     0.000004s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.15%
    rewriter_before_opt_a                :     0.000066s :  0.25%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000063s :  0.24%
    jit_opt_a.loop_unroll                :     0.000034s :  0.13%
    jit_opt_a.a_1                        :     0.000637s :  2.42%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.10%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.10%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.07%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.06%
    jit_opt_a.inline_without_move        :     0.000010s :  0.04%
    jit_opt_a.renormalize                :     0.000843s :  3.21%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.07%
    jit_opt_a.cse                        :     0.000031s :  0.12%
    jit_opt_a.replace_applicator         :     0.000021s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.05%
    rewriter_after_opt_a                 :     0.000029s :  0.11%
    convert_after_rewriter               :     0.000006s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000789s :  3.00%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.06%
    cconv                                :     0.000020s :  0.08%
    loop_unroll                          :     0.000459s :  1.74%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000010s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.01%
    add_recomputation                    :     0.000036s :  0.14%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000011s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.03%
    opt_after_jit_grad                   :     0.000482s :  1.83%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.015351s : 58.36%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000199    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 1.00% :     0.000002s :      2: substitution.fold_const_symbol
 2.31% :     0.000005s :      3: substitution.graph_param_transform
85.03% :     0.000170s :      4: substitution.inline
 1.84% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.11% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.24% :     0.000004s :      2: substitution.replace_old_param
 4.64% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006445     2
84.71% :     0.005459s :      1: type_inference.infer
15.29% :     0.000986s :      1: type_inference.specialize
------[replace.]   0.000076     6
66.03% :     0.000050s :      4: replace.inline
33.97% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000175     6
95.31% :     0.000167s :      4: match.inline
 4.69% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.60% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 2.19% :     0.000003s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000001s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.62% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.08% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.04% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :      6: predicate.less_batch_normalization
 1.80% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.92% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.86% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.19% :     0.000003s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 1.96% :     0.000002s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.82% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.90% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 1.14% :     0.000001s :      6: predicate.specialize_transform
 1.19% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.56% :     0.000003s :     16: predicate.switch_defer_inline
 2.38% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.46% :     0.000009s :     53: predicate.switch_simplify
 1.10% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.66% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.46% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.58% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.37% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.02% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000674    12
44.74% :     0.000302s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.02% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.24% :     0.000339s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.033295    76
 0.12% :     0.000039s :      1: add_recomputation
 0.16% :     0.000055s :      1: auto_monad
 0.04% :     0.000013s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.99% :     0.000329s :      1: bootstrap
 0.07% :     0.000023s :      1: cconv
 0.02% :     0.000008s :      1: convert_after_rewriter
 0.07% :     0.000022s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.08% :     0.000028s :      1: event_method
 0.04% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
20.06% :     0.006680s :      1: jit_opt_a
 0.45% :     0.000151s :      1: jit_opt_after_cconv
 0.17% :     0.000057s :      1: jit_opt_b
 1.41% :     0.000468s :      1: loop_unroll
 2.41% :     0.000801s :      1: mutable_eliminate
 2.57% :     0.000857s :     26: opt.transform.jit_opt_a
 0.15% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000028s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.47% :     0.000490s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000044s :      1: py_interpret_to_execute
 0.05% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000012s :      1: remove_dup_value
 1.28% :     0.000426s :      1: renormalize.infer
 1.23% :     0.000409s :      1: renormalize.specialize
 0.03% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000033s :      1: rewriter_after_opt_a
 0.21% :     0.000069s :      1: rewriter_before_opt_a
 0.24% :     0.000081s :      1: symbol_engine_optimizer
46.16% :     0.015371s :      1: task_emit
19.62% :     0.006534s :      1: type_inference
 0.15% :     0.000050s :      1: validate


TotalTime = 0.0124267, [30]
[bootstrap]: 0.00033725
[type_inference]: 0.0042644
[event_method]: 1.618e-05
[auto_monad]: 6.15e-05
[graph_reusing]: 6.03998e-06
[pre_auto_parallel]: 2.56998e-06
[py_interpret_to_execute]: 2.954e-05
[rewriter_before_opt_a]: 7.05e-05
[expand_dump_flag]: 3.17002e-06
[jit_opt_a]: 0.00481298, [2]
    [Cycle 1]: 0.001323, [27]
        [switch_simplify]: 4.909e-05
        [loop_unroll]: 2.878e-05
        [a_1]: 0.00048578
        [with_stream_mark]: 1.579e-05
        [recompute_prepare]: 6.69999e-06
        [updatestate_depend_eliminate]: 3.35998e-06
        [updatestate_assign_eliminate]: 3.38999e-06
        [updatestate_loads_eliminate]: 2.60002e-06
        [parameter_eliminate]: 1.97999e-06
        [specialize_transform]: 6.26e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.43002e-06
        [meta_shard_fg_expand]: 2.07001e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 3.47002e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.376e-05
        [j_node_and_user_rematch]: 9.05999e-06
        [meta_fg_expand]: 2.24999e-06
        [replace_old_param]: 8.26002e-06
        [inline_without_move]: 5.14e-06
        [renormalize]: 0.00043736
        [add_forward_monad_depend]: 4.89e-06
        [auto_monad_grad]: 2.29999e-06
        [auto_monad_eliminator]: 1.245e-05
        [cse]: 2.569e-05
        [replace_applicator]: 1.057e-05
    [Cycle 2]: 0.00030642, [27]
        [switch_simplify]: 5.67001e-06
        [loop_unroll]: 4.90001e-06
        [a_1]: 8.757e-05
        [with_stream_mark]: 8.12998e-06
        [recompute_prepare]: 4.79e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 2.10002e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 8.99978e-07
        [specialize_transform]: 4.77998e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 4.74998e-06
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.24002e-06
        [merge_forward]: 2.41e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.158e-05
        [j_node_and_user_rematch]: 7.73001e-06
        [meta_fg_expand]: 1.45001e-06
        [replace_old_param]: 6.84001e-06
        [inline_without_move]: 4.59998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 4.66002e-06
        [cse]: 9.57001e-06
        [replace_applicator]: 4.67e-06
[py_interpret_to_execute_after_opt_a]: 7.88999e-06
[rewriter_after_opt_a]: 3.202e-05
[convert_after_rewriter]: 6.81999e-06
[order_py_execute_after_rewriter]: 4.73001e-06
[mutable_eliminate]: 0.00130277
[jit_opt_b]: 5.411e-05, [1]
    [Cycle 1]: 4.699e-05, [2]
        [frontend_op_eliminate]: 1.954e-05
        [inline_after_opt_a]: 1.543e-05
[cconv]: 2.506e-05
[loop_unroll]: 0.00043571
[jit_opt_after_cconv]: 0.00013472, [1]
    [Cycle 1]: 0.00012854, [11]
        [c_1]: 2.061e-05
        [parameter_eliminate]: 1.70001e-06
        [updatestate_depend_eliminate]: 4.22e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 1.705e-05
        [call_graph_tuple_transform]: 1.825e-05
        [tuple_list_get_item_eliminator]: 5.63002e-06
        [none_parameter_eliminate]: 1.61002e-06
        [renormalize]: 6.69999e-07
        [switch_simplify]: 5.09998e-06
[remove_dup_value]: 1.327e-05
[partial_unused_args_eliminate]: 2.12001e-06
[environ_conv]: 4.50001e-06
[add_recomputation]: 4.777e-05
[cse_after_recomputation]: 2.036e-05, [1]
    [Cycle 1]: 1.432e-05, [1]
        [cse]: 8.52e-06
[auto_monad_reorder]: 1.5e-05
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 5.09e-06
[opt_after_jit_grad]: 0.00044435
[symbol_engine_optimizer]: 6.933e-05, [1]
    [Cycle 1]: 6.313e-05, [6]
        [build]: 3.25e-06
        [elim_shapecalc]: 7.4e-06
        [elim_not_effective]: 1.181e-05
        [opt_reshape]: 5.27999e-06
        [fold_const_symbol]: 8.2e-06
        [renormalize]: 5.19998e-07
[validate]: 3.196e-05
Sums
    bootstrap                            :     0.000337s :  3.88%
    type_inference                       :     0.004264s : 49.11%
    event_method                         :     0.000016s :  0.19%
    auto_monad                           :     0.000062s :  0.71%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000030s :  0.34%
    rewriter_before_opt_a                :     0.000071s :  0.81%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000055s :  0.63%
    jit_opt_a.loop_unroll                :     0.000034s :  0.39%
    jit_opt_a.a_1                        :     0.000573s :  6.60%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.17%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000437s :  5.04%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.20%
    jit_opt_a.cse                        :     0.000035s :  0.41%
    jit_opt_a.replace_applicator         :     0.000015s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.09%
    rewriter_after_opt_a                 :     0.000032s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.001303s : 15.00%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000025s :  0.29%
    loop_unroll                          :     0.000436s :  5.02%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000013s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000048s :  0.55%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.17%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000444s :  5.12%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000032s :  0.37%


Time group info:
------[substitution.]   0.000174    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.78% :     0.000001s :      2: substitution.fold_const_symbol
 2.91% :     0.000005s :      3: substitution.graph_param_transform
81.95% :     0.000142s :      4: substitution.inline
 1.63% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.43% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.99% :     0.000003s :      2: substitution.replace_old_param
 7.28% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004204     2
81.51% :     0.003427s :      1: type_inference.infer
18.49% :     0.000777s :      1: type_inference.specialize
------[replace.]   0.000060     6
67.16% :     0.000040s :      4: replace.inline
32.84% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000152     6
92.38% :     0.000140s :      4: match.inline
 7.62% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.12% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000001s :     10: predicate.addn_zero_filter
 1.79% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.59% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.86% :     0.000002s :     15: predicate.load_eliminater
 1.25% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.23% :     0.000001s :     10: predicate.minmaximum_grad
 2.35% :     0.000003s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.42% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.40% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000001s :     10: predicate.reshape_eliminate
 1.29% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.76% :     0.000009s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.22% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.29% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000534    12
53.01% :     0.000283s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.89% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.10% :     0.000225s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013695    72
 0.37% :     0.000051s :      1: add_recomputation
 0.47% :     0.000065s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.62% :     0.000358s :      1: bootstrap
 0.20% :     0.000028s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
35.16% :     0.004816s :      1: jit_opt_a
 1.00% :     0.000137s :      1: jit_opt_after_cconv
 0.41% :     0.000057s :      1: jit_opt_b
 3.23% :     0.000442s :      1: loop_unroll
 9.59% :     0.001313s :      1: mutable_eliminate
 5.62% :     0.000769s :     26: opt.transform.jit_opt_a
 0.34% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000028s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.29% :     0.000451s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000032s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000016s :      1: remove_dup_value
 1.55% :     0.000212s :      1: renormalize.infer
 1.60% :     0.000219s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000035s :      1: rewriter_after_opt_a
 0.54% :     0.000073s :      1: rewriter_before_opt_a
 0.52% :     0.000072s :      1: symbol_engine_optimizer
31.22% :     0.004276s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.801.37 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dcaf8b0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.808.39 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0382316, [33]
[bootstrap]: 0.00028065
[type_inference]: 0.013936
[event_method]: 6.046e-05
[auto_monad]: 0.00010688
[graph_reusing]: 7.51999e-06
[pre_auto_parallel]: 3.38e-06
[py_interpret_to_execute]: 3.118e-05
[rewriter_before_opt_a]: 0.00010518
[expand_dump_flag]: 3.6e-06
[jit_opt_a]: 0.0125451, [3]
    [Cycle 1]: 0.00640842, [27]
        [switch_simplify]: 0.00014608
        [loop_unroll]: 4.265e-05
        [a_1]: 0.00093211
        [with_stream_mark]: 3.247e-05
        [recompute_prepare]: 1.823e-05
        [updatestate_depend_eliminate]: 8.17e-06
        [updatestate_assign_eliminate]: 6.69999e-06
        [updatestate_loads_eliminate]: 6.24001e-06
        [parameter_eliminate]: 2.69999e-06
        [specialize_transform]: 1.319e-05
        [updatestate_useless_node_eliminater]: 1.187e-05
        [accelerated_algorithm]: 1.383e-05
        [meta_shard_fg_expand]: 3.3e-06
        [get_grad_eliminate_]: 1.326e-05
        [merge_forward]: 8.37e-06
        [cell_reuse_recompute_pass]: 1.07998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.74e-05
        [j_node_and_user_rematch]: 2.447e-05
        [meta_fg_expand]: 0.00171555
        [replace_old_param]: 6.117e-05
        [inline_without_move]: 5.364e-05
        [renormalize]: 0.00272406
        [add_forward_monad_depend]: 1.105e-05
        [auto_monad_grad]: 5.94e-06
        [auto_monad_eliminator]: 4.957e-05
        [cse]: 0.00015565
        [replace_applicator]: 8.505e-05
    [Cycle 2]: 0.00208183, [27]
        [switch_simplify]: 3.492e-05
        [loop_unroll]: 3.268e-05
        [a_1]: 0.0010281
        [with_stream_mark]: 1.452e-05
        [recompute_prepare]: 7.87e-06
        [updatestate_depend_eliminate]: 3.58e-06
        [updatestate_assign_eliminate]: 3.63e-06
        [updatestate_loads_eliminate]: 2.87002e-06
        [parameter_eliminate]: 1.79998e-06
        [specialize_transform]: 7.26001e-06
        [updatestate_useless_node_eliminater]: 6.13998e-06
        [accelerated_algorithm]: 7.61999e-06
        [meta_shard_fg_expand]: 2.49001e-06
        [get_grad_eliminate_]: 5.44998e-06
        [merge_forward]: 4.12e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.588e-05
        [j_node_and_user_rematch]: 1.118e-05
        [meta_fg_expand]: 0.00013212
        [replace_old_param]: 1.031e-05
        [inline_without_move]: 5.89999e-06
        [renormalize]: 0.00054171
        [add_forward_monad_depend]: 4.33001e-06
        [auto_monad_grad]: 1.94999e-06
        [auto_monad_eliminator]: 1.202e-05
        [cse]: 2.122e-05
        [replace_applicator]: 1.249e-05
    [Cycle 3]: 0.00035347, [27]
        [switch_simplify]: 6.25002e-06
        [loop_unroll]: 5.74e-06
        [a_1]: 0.00011399
        [with_stream_mark]: 9.09e-06
        [recompute_prepare]: 5.59e-06
        [updatestate_depend_eliminate]: 3.81999e-06
        [updatestate_assign_eliminate]: 2.59999e-06
        [updatestate_loads_eliminate]: 2.46998e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 6.11e-06
        [updatestate_useless_node_eliminater]: 5.54e-06
        [accelerated_algorithm]: 5.59e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 5.23002e-06
        [merge_forward]: 2.93e-06
        [cell_reuse_recompute_pass]: 1.82999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.448e-05
        [j_node_and_user_rematch]: 9.84001e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 8.07e-06
        [inline_without_move]: 5.63002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 6.29001e-06
        [cse]: 1.304e-05
        [replace_applicator]: 5.66e-06
[py_interpret_to_execute_after_opt_a]: 1.306e-05
[rewriter_after_opt_a]: 4.033e-05
[convert_after_rewriter]: 8.12e-06
[order_py_execute_after_rewriter]: 5.30999e-06
[mutable_eliminate]: 0.00089883
[jit_opt_b]: 5.754e-05, [1]
    [Cycle 1]: 4.925e-05, [2]
        [frontend_op_eliminate]: 1.975e-05
        [inline_after_opt_a]: 1.729e-05
[cconv]: 3.918e-05
[loop_unroll]: 0.00050338
[jit_opt_after_cconv]: 0.00020834, [1]
    [Cycle 1]: 0.00020093, [11]
        [c_1]: 4.971e-05
        [parameter_eliminate]: 4.78001e-06
        [updatestate_depend_eliminate]: 1.001e-05
        [updatestate_assign_eliminate]: 4.05e-06
        [updatestate_loads_eliminate]: 2.83998e-06
        [cse]: 3.697e-05
        [call_graph_tuple_transform]: 2.492e-05
        [tuple_list_get_item_eliminator]: 6.83e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 8.59989e-07
        [switch_simplify]: 6.79001e-06
[remove_dup_value]: 1.794e-05
[partial_unused_args_eliminate]: 2.15002e-06
[environ_conv]: 7.80998e-06
[add_recomputation]: 8.301e-05
[cse_after_recomputation]: 3.287e-05, [1]
    [Cycle 1]: 2.596e-05, [1]
        [cse]: 1.793e-05
[auto_monad_reorder]: 2.141e-05
[get_jit_bprop_graph]: 2.32999e-06
[rewriter_after_jit_bprop_graph]: 8.1e-06
[opt_after_jit_grad]: 0.00060025
[symbol_engine_optimizer]: 0.00010137, [1]
    [Cycle 1]: 9.302e-05, [6]
        [build]: 7.64002e-06
        [elim_shapecalc]: 1.022e-05
        [elim_not_effective]: 1.879e-05
        [opt_reshape]: 9.96e-06
        [fold_const_symbol]: 1.353e-05
        [renormalize]: 8.70001e-07
[validate]: 5.6e-05
[backend_pass]: 1.19e-06
[task_emit]: 0.00818331
[execute]: 9.05999e-06
Sums
    bootstrap                            :     0.000281s :  0.83%
    type_inference                       :     0.013936s : 41.41%
    event_method                         :     0.000060s :  0.18%
    auto_monad                           :     0.000107s :  0.32%
    graph_reusing                        :     0.000008s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000031s :  0.09%
    rewriter_before_opt_a                :     0.000105s :  0.31%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000187s :  0.56%
    jit_opt_a.loop_unroll                :     0.000081s :  0.24%
    jit_opt_a.a_1                        :     0.002074s :  6.16%
    jit_opt_a.with_stream_mark           :     0.000056s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000027s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000027s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.07%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000058s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000045s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.001850s :  5.50%
    jit_opt_a.replace_old_param          :     0.000080s :  0.24%
    jit_opt_a.inline_without_move        :     0.000065s :  0.19%
    jit_opt_a.renormalize                :     0.003266s :  9.70%
    jit_opt_a.add_forward_monad_depend   :     0.000017s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000068s :  0.20%
    jit_opt_a.cse                        :     0.000190s :  0.56%
    jit_opt_a.replace_applicator         :     0.000103s :  0.31%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.04%
    rewriter_after_opt_a                 :     0.000040s :  0.12%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000899s :  2.67%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.05%
    cconv                                :     0.000039s :  0.12%
    loop_unroll                          :     0.000503s :  1.50%
    jit_opt_after_cconv.c_1              :     0.000050s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000037s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000018s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000008s :  0.02%
    add_recomputation                    :     0.000083s :  0.25%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000021s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000600s :  1.78%
    symbol_engine_optimizer.build        :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000056s :  0.17%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008183s : 24.32%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000562   113
 0.44% :     0.000002s :      3: substitution.elim_not_effective
 0.49% :     0.000003s :      3: substitution.fold_const_symbol
 1.08% :     0.000006s :      4: substitution.graph_param_transform
61.60% :     0.000346s :     14: substitution.inline
 3.37% :     0.000019s :      2: substitution.inline_without_move
 1.80% :     0.000010s :     14: substitution.j_node_and_user_rematch
 2.04% :     0.000011s :      7: substitution.minmaximum_grad
 1.17% :     0.000007s :      5: substitution.partial_eliminate
 1.98% :     0.000011s :     14: substitution.remove_not_recompute_node
 4.99% :     0.000028s :     10: substitution.replace_applicator
 1.46% :     0.000008s :      7: substitution.replace_old_param
 0.40% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.63% :     0.000009s :      1: substitution.switch_simplify
 3.85% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.55% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
11.16% :     0.000063s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.013845     2
89.11% :     0.012337s :      1: type_inference.infer
10.89% :     0.001507s :      1: type_inference.specialize
------[replace.]   0.000195    22
51.15% :     0.000100s :     14: replace.inline
21.66% :     0.000042s :      1: replace.switch_simplify
27.19% :     0.000053s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000392    22
86.69% :     0.000340s :     14: match.inline
 2.20% :     0.000009s :      1: match.switch_simplify
11.10% :     0.000043s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000377  2634
 1.60% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.75% :     0.000003s :      4: predicate.ad_related_special_op_eliminate
 1.43% :     0.000005s :     44: predicate.addn_check_dump
 1.65% :     0.000006s :     44: predicate.addn_zero_filter
 1.87% :     0.000007s :     44: predicate.arithmetic_simplify
 1.51% :     0.000006s :     44: predicate.cast_eliminate
 0.18% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.38% :     0.000005s :     44: predicate.compare_switch_simplify
 1.53% :     0.000006s :     44: predicate.depend_value_elim
 1.41% :     0.000005s :     44: predicate.dict_get_item_const_eliminator
 1.46% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.42% :     0.000005s :     44: predicate.dict_set_item_eliminator
 0.39% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      4: predicate.elim_not_effective
 0.26% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000005s :     44: predicate.environ_add_const_eliminate
 1.39% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.41% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.52% :     0.000006s :     44: predicate.environ_get_eliminate
 1.54% :     0.000006s :     44: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.81% :     0.000003s :     19: predicate.get_grad_eliminate
 0.11% :     0.000000s :      4: predicate.graph_param_transform
 4.44% :     0.000017s :     73: predicate.inline
 1.86% :     0.000007s :     40: predicate.inline_without_move
 0.34% :     0.000001s :     19: predicate.j_node_and_user_rematch
 1.08% :     0.000004s :     19: predicate.less_batch_normalization
 1.78% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.91% :     0.000007s :     55: predicate.load_eliminater
 0.72% :     0.000003s :      4: predicate.loop_unroll_after_grad
 3.25% :     0.000012s :     91: predicate.loop_unroll_before_grad
 1.72% :     0.000006s :     48: predicate.make_slice_get_slice_eliminator
 1.42% :     0.000005s :     44: predicate.merge_addn
 1.46% :     0.000006s :     44: predicate.minmaximum_grad
 0.81% :     0.000003s :      4: predicate.mutable_eliminate
 0.31% :     0.000001s :      4: predicate.opt_reshape
 2.32% :     0.000009s :     55: predicate.partial_eliminate
 1.52% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.81% :     0.000007s :     44: predicate.reduce_eliminate
 1.77% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.45% :     0.000009s :     97: predicate.replace_applicator
 1.07% :     0.000004s :     40: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.86% :     0.000007s :     44: predicate.reshape_eliminate
 1.60% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.32% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.62% :     0.000006s :     44: predicate.same_eliminate
 0.48% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.43% :     0.000002s :      8: predicate.special_op_eliminate
 0.80% :     0.000003s :     19: predicate.specialize_transform
 1.64% :     0.000006s :     44: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000005s :     44: predicate.stack_unstack_eliminate
 0.18% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.74% :     0.000010s :     65: predicate.switch_defer_inline
 2.48% :     0.000009s :     65: predicate.switch_layer_defer_inline
 6.46% :     0.000024s :    162: predicate.switch_simplify
 1.47% :     0.000006s :     44: predicate.tile_eliminate
 1.52% :     0.000006s :     44: predicate.transpose_eliminate
 1.81% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.59% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 3.17% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 1.89% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.82% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000007s :     44: predicate.value_based_eliminate
 0.15% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.24% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001811    32
61.12% :     0.001107s :     14: func_graph_cloner_run.FuncGraphClonerGraph
38.88% :     0.000704s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.044447    91
 0.19% :     0.000087s :      1: add_recomputation
 0.25% :     0.000112s :      1: auto_monad
 0.05% :     0.000024s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.69% :     0.000308s :      1: bootstrap
 0.10% :     0.000042s :      1: cconv
 0.02% :     0.000011s :      1: convert_after_rewriter
 0.08% :     0.000035s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.15% :     0.000066s :      1: event_method
 0.03% :     0.000014s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
28.23% :     0.012549s :      1: jit_opt_a
 0.48% :     0.000212s :      1: jit_opt_after_cconv
 0.14% :     0.000060s :      1: jit_opt_b
 1.16% :     0.000516s :      1: loop_unroll
 2.05% :     0.000910s :      1: mutable_eliminate
 6.25% :     0.002779s :     39: opt.transform.jit_opt_a
 0.19% :     0.000085s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000031s :      4: opt.transform.jit_opt_b
 0.04% :     0.000020s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000048s :      4: opt.transform.symbol_engine_opt
 1.38% :     0.000611s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000034s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000020s :      1: remove_dup_value
 4.07% :     0.001809s :      2: renormalize.infer
 3.24% :     0.001439s :      2: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000043s :      1: rewriter_after_opt_a
 0.24% :     0.000108s :      1: rewriter_before_opt_a
 0.23% :     0.000104s :      1: symbol_engine_optimizer
18.45% :     0.008199s :      1: task_emit
31.40% :     0.013957s :      1: type_inference
 0.20% :     0.000087s :      1: validate


TotalTime = 0.013761, [30]
[bootstrap]: 0.00039642
[type_inference]: 0.00498063
[event_method]: 1.772e-05
[auto_monad]: 7.062e-05
[graph_reusing]: 6.76999e-06
[pre_auto_parallel]: 2.66e-06
[py_interpret_to_execute]: 4.131e-05
[rewriter_before_opt_a]: 7.796e-05
[expand_dump_flag]: 3.43999e-06
[jit_opt_a]: 0.00596238, [2]
    [Cycle 1]: 0.00188604, [27]
        [switch_simplify]: 6.311e-05
        [loop_unroll]: 3.207e-05
        [a_1]: 0.00058317
        [with_stream_mark]: 2.236e-05
        [recompute_prepare]: 7.85e-06
        [updatestate_depend_eliminate]: 3.91001e-06
        [updatestate_assign_eliminate]: 3.12002e-06
        [updatestate_loads_eliminate]: 2.47001e-06
        [parameter_eliminate]: 1.90001e-06
        [specialize_transform]: 6.31e-06
        [updatestate_useless_node_eliminater]: 5.39e-06
        [accelerated_algorithm]: 5.47999e-06
        [meta_shard_fg_expand]: 2.09999e-06
        [get_grad_eliminate_]: 5.37001e-06
        [merge_forward]: 3.46001e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.44e-05
        [j_node_and_user_rematch]: 9.61998e-06
        [meta_fg_expand]: 2.22001e-06
        [replace_old_param]: 8.77e-06
        [inline_without_move]: 5.11997e-06
        [renormalize]: 0.00084318
        [add_forward_monad_depend]: 5.56e-06
        [auto_monad_grad]: 2.14999e-06
        [auto_monad_eliminator]: 1.453e-05
        [cse]: 3.075e-05
        [replace_applicator]: 1.363e-05
    [Cycle 2]: 0.00032073, [27]
        [switch_simplify]: 5.74e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 8.938e-05
        [with_stream_mark]: 9.39998e-06
        [recompute_prepare]: 5.09e-06
        [updatestate_depend_eliminate]: 2.68e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.80001e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 4.78001e-06
        [updatestate_useless_node_eliminater]: 4.72e-06
        [accelerated_algorithm]: 5.32001e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.24999e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.34e-05
        [j_node_and_user_rematch]: 8.17998e-06
        [meta_fg_expand]: 1.82001e-06
        [replace_old_param]: 7.56999e-06
        [inline_without_move]: 4.68001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 5.08002e-06
        [cse]: 1.036e-05
        [replace_applicator]: 4.99e-06
[py_interpret_to_execute_after_opt_a]: 1.263e-05
[rewriter_after_opt_a]: 3.68e-05
[convert_after_rewriter]: 6.98e-06
[order_py_execute_after_rewriter]: 4.86002e-06
[mutable_eliminate]: 0.00060203
[jit_opt_b]: 5.232e-05, [1]
    [Cycle 1]: 4.494e-05, [2]
        [frontend_op_eliminate]: 1.709e-05
        [inline_after_opt_a]: 1.584e-05
[cconv]: 2.663e-05
[loop_unroll]: 0.00041972
[jit_opt_after_cconv]: 0.00014202, [1]
    [Cycle 1]: 0.00013589, [11]
        [c_1]: 2.066e-05
        [parameter_eliminate]: 2.68e-06
        [updatestate_depend_eliminate]: 5.69999e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 1.997e-05
        [call_graph_tuple_transform]: 2.011e-05
        [tuple_list_get_item_eliminator]: 5.37999e-06
        [none_parameter_eliminate]: 1.34e-06
        [renormalize]: 4.7998e-07
        [switch_simplify]: 5.35001e-06
[remove_dup_value]: 1.447e-05
[partial_unused_args_eliminate]: 2.48e-06
[environ_conv]: 5.82999e-06
[add_recomputation]: 4.993e-05
[cse_after_recomputation]: 2.021e-05, [1]
    [Cycle 1]: 1.464e-05, [1]
        [cse]: 9.12999e-06
[auto_monad_reorder]: 1.424e-05
[get_jit_bprop_graph]: 2.63e-06
[rewriter_after_jit_bprop_graph]: 7.97e-06
[opt_after_jit_grad]: 0.00047733
[symbol_engine_optimizer]: 7.533e-05, [1]
    [Cycle 1]: 6.872e-05, [6]
        [build]: 3.67002e-06
        [elim_shapecalc]: 7.13998e-06
        [elim_not_effective]: 1.309e-05
        [opt_reshape]: 5.71e-06
        [fold_const_symbol]: 1.018e-05
        [renormalize]: 6.69999e-07
[validate]: 3.557e-05
Sums
    bootstrap                            :     0.000396s :  4.22%
    type_inference                       :     0.004981s : 53.04%
    event_method                         :     0.000018s :  0.19%
    auto_monad                           :     0.000071s :  0.75%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000041s :  0.44%
    rewriter_before_opt_a                :     0.000078s :  0.83%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000069s :  0.73%
    jit_opt_a.loop_unroll                :     0.000037s :  0.39%
    jit_opt_a.a_1                        :     0.000673s :  7.16%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.34%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.17%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000843s :  8.98%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.21%
    jit_opt_a.cse                        :     0.000041s :  0.44%
    jit_opt_a.replace_applicator         :     0.000019s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.13%
    rewriter_after_opt_a                 :     0.000037s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000602s :  6.41%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.17%
    cconv                                :     0.000027s :  0.28%
    loop_unroll                          :     0.000420s :  4.47%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000050s :  0.53%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000014s :  0.15%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.08%
    opt_after_jit_grad                   :     0.000477s :  5.08%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000036s :  0.38%


Time group info:
------[substitution.]   0.000245    23
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.36% :     0.000006s :      3: substitution.graph_param_transform
86.61% :     0.000213s :      4: substitution.inline
 1.39% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.76% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.61% :     0.000004s :      2: substitution.replace_old_param
 4.98% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004907     2
80.80% :     0.003965s :      1: type_inference.infer
19.20% :     0.000942s :      1: type_inference.specialize
------[replace.]   0.000074     6
68.88% :     0.000051s :      4: replace.inline
31.12% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000221     6
94.95% :     0.000210s :      4: match.inline
 5.05% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.68% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000001s :     10: predicate.addn_zero_filter
 1.84% :     0.000002s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.39% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.38% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.43% :     0.000000s :      3: predicate.graph_param_transform
 5.31% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.20% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.18% :     0.000001s :     10: predicate.minmaximum_grad
 1.98% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.30% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.68% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.35% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.58% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.30% :     0.000010s :     53: predicate.switch_simplify
 1.33% :     0.000002s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.57% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.68% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000694    12
48.99% :     0.000340s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.36% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.65% :     0.000303s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015550    72
 0.34% :     0.000053s :      1: add_recomputation
 0.48% :     0.000075s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 2.68% :     0.000417s :      1: bootstrap
 0.19% :     0.000029s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.14% :     0.000022s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
38.36% :     0.005966s :      1: jit_opt_a
 0.93% :     0.000145s :      1: jit_opt_after_cconv
 0.35% :     0.000055s :      1: jit_opt_b
 2.75% :     0.000428s :      1: loop_unroll
 3.93% :     0.000611s :      1: mutable_eliminate
 5.75% :     0.000894s :     26: opt.transform.jit_opt_a
 0.31% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.13% :     0.000486s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000044s :      1: py_interpret_to_execute
 0.10% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 3.03% :     0.000471s :      1: renormalize.infer
 2.34% :     0.000364s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000040s :      1: rewriter_after_opt_a
 0.52% :     0.000081s :      1: rewriter_before_opt_a
 0.50% :     0.000078s :      1: symbol_engine_optimizer
32.16% :     0.005000s :      1: type_inference


TotalTime = 0.0410926, [33]
[bootstrap]: 0.00029027
[type_inference]: 0.0043651
[event_method]: 1.711e-05
[auto_monad]: 6.368e-05
[graph_reusing]: 6.21e-06
[pre_auto_parallel]: 2.64999e-06
[py_interpret_to_execute]: 3.038e-05
[rewriter_before_opt_a]: 7.116e-05
[expand_dump_flag]: 3.4e-06
[jit_opt_a]: 0.0281883, [2]
    [Cycle 1]: 0.00185307, [27]
        [switch_simplify]: 6.458e-05
        [loop_unroll]: 2.871e-05
        [a_1]: 0.00062156
        [with_stream_mark]: 2.48e-05
        [recompute_prepare]: 8.06001e-06
        [updatestate_depend_eliminate]: 3.76999e-06
        [updatestate_assign_eliminate]: 3.21999e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 2.03002e-06
        [specialize_transform]: 6.06e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.40001e-06
        [meta_shard_fg_expand]: 2.02999e-06
        [get_grad_eliminate_]: 5.51e-06
        [merge_forward]: 3.97998e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.456e-05
        [j_node_and_user_rematch]: 8.47998e-06
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 9.67999e-06
        [inline_without_move]: 5.22999e-06
        [renormalize]: 0.00075973
        [add_forward_monad_depend]: 6.98e-06
        [auto_monad_grad]: 2.69001e-06
        [auto_monad_eliminator]: 1.508e-05
        [cse]: 2.907e-05
        [replace_applicator]: 1.465e-05
    [Cycle 2]: 0.00031623, [27]
        [switch_simplify]: 5.28002e-06
        [loop_unroll]: 4.82e-06
        [a_1]: 8.771e-05
        [with_stream_mark]: 1.054e-05
        [recompute_prepare]: 4.72e-06
        [updatestate_depend_eliminate]: 2.41e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.57001e-06
        [specialize_transform]: 4.95999e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 4.74e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 4.33001e-06
        [merge_forward]: 2.39001e-06
        [cell_reuse_recompute_pass]: 2.16998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.223e-05
        [j_node_and_user_rematch]: 7.80998e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 7.09001e-06
        [inline_without_move]: 6.66e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 5.44998e-06
        [cse]: 1.017e-05
        [replace_applicator]: 4.72998e-06
[py_interpret_to_execute_after_opt_a]: 1.246e-05
[rewriter_after_opt_a]: 3.714e-05
[convert_after_rewriter]: 6.96001e-06
[order_py_execute_after_rewriter]: 4.73001e-06
[mutable_eliminate]: 0.00062642
[jit_opt_b]: 4.985e-05, [1]
    [Cycle 1]: 4.255e-05, [2]
        [frontend_op_eliminate]: 1.628e-05
        [inline_after_opt_a]: 1.471e-05
[cconv]: 2.926e-05
[loop_unroll]: 0.00044799
[jit_opt_after_cconv]: 0.0001422, [1]
    [Cycle 1]: 0.00013565, [11]
        [c_1]: 1.947e-05
        [parameter_eliminate]: 3.14001e-06
        [updatestate_depend_eliminate]: 6.66999e-06
        [updatestate_assign_eliminate]: 2.36998e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [cse]: 2.052e-05
        [call_graph_tuple_transform]: 1.938e-05
        [tuple_list_get_item_eliminator]: 4.97999e-06
        [none_parameter_eliminate]: 1.53002e-06
        [renormalize]: 1.20001e-06
        [switch_simplify]: 4.92999e-06
[remove_dup_value]: 1.395e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 5.14e-06
[add_recomputation]: 4.885e-05
[cse_after_recomputation]: 2.058e-05, [1]
    [Cycle 1]: 1.495e-05, [1]
        [cse]: 8.85001e-06
[auto_monad_reorder]: 1.503e-05
[get_jit_bprop_graph]: 2.02999e-06
[rewriter_after_jit_bprop_graph]: 5.65001e-06
[opt_after_jit_grad]: 0.00051467
[symbol_engine_optimizer]: 7.049e-05, [1]
    [Cycle 1]: 6.333e-05, [6]
        [build]: 3.5e-06
        [elim_shapecalc]: 6.85002e-06
        [elim_not_effective]: 1.208e-05
        [opt_reshape]: 5.17e-06
        [fold_const_symbol]: 7.95998e-06
        [renormalize]: 9.5999e-07
[validate]: 3.653e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00575854
[execute]: 8.72998e-06
Sums
    bootstrap                            :     0.000290s :  2.01%
    type_inference                       :     0.004365s : 30.21%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000064s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.21%
    rewriter_before_opt_a                :     0.000071s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000070s :  0.48%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000709s :  4.91%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.12%
    jit_opt_a.inline_without_move        :     0.000012s :  0.08%
    jit_opt_a.renormalize                :     0.000760s :  5.26%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.14%
    jit_opt_a.cse                        :     0.000039s :  0.27%
    jit_opt_a.replace_applicator         :     0.000019s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.09%
    rewriter_after_opt_a                 :     0.000037s :  0.26%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000626s :  4.34%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000029s :  0.20%
    loop_unroll                          :     0.000448s :  3.10%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000049s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000515s :  3.56%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000037s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005759s : 39.85%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000266    23
 0.66% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 2.06% :     0.000005s :      3: substitution.graph_param_transform
87.81% :     0.000234s :      4: substitution.inline
 1.10% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.51% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.93% :     0.000005s :      2: substitution.replace_old_param
 4.47% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004299     2
82.21% :     0.003534s :      1: type_inference.infer
17.79% :     0.000765s :      1: type_inference.specialize
------[replace.]   0.000072     6
69.42% :     0.000050s :      4: replace.inline
30.58% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000242     6
95.56% :     0.000231s :      4: match.inline
 4.44% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.54% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.36% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 1.95% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.62% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.19% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.40% :     0.000000s :      3: predicate.graph_param_transform
 5.86% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      6: predicate.less_batch_normalization
 1.62% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.02% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.76% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 2.35% :     0.000003s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000002s :     15: predicate.partial_eliminate
 1.24% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.39% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.84% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.32% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.45% :     0.000002s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.56% :     0.000009s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.25% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.73% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.69% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000568    12
48.02% :     0.000273s :      5: func_graph_cloner_run.FuncGraphClonerGraph
10.22% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.77% :     0.000237s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042883    76
 0.12% :     0.000052s :      1: add_recomputation
 0.16% :     0.000067s :      1: auto_monad
 0.04% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.72% :     0.000307s :      1: bootstrap
 0.08% :     0.000032s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.05% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.05% :     0.000021s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
65.74% :     0.028192s :      1: jit_opt_a
 0.34% :     0.000145s :      1: jit_opt_after_cconv
 0.12% :     0.000052s :      1: jit_opt_b
 1.06% :     0.000456s :      1: loop_unroll
 1.49% :     0.000637s :      1: mutable_eliminate
 2.16% :     0.000927s :     26: opt.transform.jit_opt_a
 0.11% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000024s :      4: opt.transform.jit_opt_b
 0.03% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000029s :      4: opt.transform.symbol_engine_opt
 1.22% :     0.000525s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.08% :     0.000033s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000016s :      1: remove_dup_value
 1.02% :     0.000437s :      1: renormalize.infer
 0.73% :     0.000314s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000040s :      1: rewriter_after_opt_a
 0.17% :     0.000074s :      1: rewriter_before_opt_a
 0.17% :     0.000073s :      1: symbol_engine_optimizer
13.45% :     0.005769s :      1: task_emit
10.21% :     0.004379s :      1: type_inference
 0.14% :     0.000058s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.186.206 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dcaf670> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.186.805 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0515925, [33]
[bootstrap]: 0.00027223
[type_inference]: 0.0041618
[event_method]: 1.54e-05
[auto_monad]: 5.446e-05
[graph_reusing]: 4.01001e-06
[pre_auto_parallel]: 1.84e-06
[py_interpret_to_execute]: 3.14e-05
[rewriter_before_opt_a]: 6.62e-05
[expand_dump_flag]: 2.42001e-06
[jit_opt_a]: 0.00488855, [2]
    [Cycle 1]: 0.00156304, [27]
        [switch_simplify]: 0.00020349
        [loop_unroll]: 2.885e-05
        [a_1]: 0.00048432
        [with_stream_mark]: 1.103e-05
        [recompute_prepare]: 1.335e-05
        [updatestate_depend_eliminate]: 3.16999e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [parameter_eliminate]: 1.32999e-06
        [specialize_transform]: 6.01e-06
        [updatestate_useless_node_eliminater]: 4.96002e-06
        [accelerated_algorithm]: 5.35001e-06
        [meta_shard_fg_expand]: 1.44998e-06
        [get_grad_eliminate_]: 5.15999e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 1.05999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.158e-05
        [j_node_and_user_rematch]: 8.70999e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 8.28999e-06
        [inline_without_move]: 5.39998e-06
        [renormalize]: 0.00055311
        [add_forward_monad_depend]: 3.81999e-06
        [auto_monad_grad]: 2.02999e-06
        [auto_monad_eliminator]: 9.99999e-06
        [cse]: 1.723e-05
        [replace_applicator]: 1.061e-05
    [Cycle 2]: 0.00031061, [27]
        [switch_simplify]: 5.91e-06
        [loop_unroll]: 4.72998e-06
        [a_1]: 8.856e-05
        [with_stream_mark]: 8.37e-06
        [recompute_prepare]: 4.77998e-06
        [updatestate_depend_eliminate]: 2.37999e-06
        [updatestate_assign_eliminate]: 1.97999e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 9.30013e-07
        [specialize_transform]: 4.66997e-06
        [updatestate_useless_node_eliminater]: 4.80999e-06
        [accelerated_algorithm]: 4.97e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.3e-06
        [merge_forward]: 2.19999e-06
        [cell_reuse_recompute_pass]: 1.50001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.196e-05
        [j_node_and_user_rematch]: 7.63001e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 6.88998e-06
        [inline_without_move]: 4.67998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.11997e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 5.10999e-06
        [cse]: 9.07999e-06
        [replace_applicator]: 4.94003e-06
[py_interpret_to_execute_after_opt_a]: 7.39002e-06
[rewriter_after_opt_a]: 2.772e-05
[convert_after_rewriter]: 5.65001e-06
[order_py_execute_after_rewriter]: 4.37998e-06
[mutable_eliminate]: 0.00047181
[jit_opt_b]: 4.888e-05, [1]
    [Cycle 1]: 4.245e-05, [2]
        [frontend_op_eliminate]: 1.612e-05
        [inline_after_opt_a]: 1.505e-05
[cconv]: 1.935e-05
[loop_unroll]: 0.00040865
[jit_opt_after_cconv]: 0.00012958, [1]
    [Cycle 1]: 0.00012356, [11]
        [c_1]: 2.098e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 4.15999e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 1.96998e-06
        [cse]: 1.575e-05
        [call_graph_tuple_transform]: 1.56e-05
        [tuple_list_get_item_eliminator]: 5.09e-06
        [none_parameter_eliminate]: 1.60001e-06
        [renormalize]: 5.00004e-07
        [switch_simplify]: 5.00999e-06
[remove_dup_value]: 7.73999e-06
[partial_unused_args_eliminate]: 1.89e-06
[environ_conv]: 4.27998e-06
[add_recomputation]: 3.536e-05
[cse_after_recomputation]: 1.927e-05, [1]
    [Cycle 1]: 1.405e-05, [1]
        [cse]: 8.23999e-06
[auto_monad_reorder]: 1.087e-05
[get_jit_bprop_graph]: 1.80001e-06
[rewriter_after_jit_bprop_graph]: 3.87998e-06
[opt_after_jit_grad]: 0.00043754
[symbol_engine_optimizer]: 6.77e-05, [1]
    [Cycle 1]: 6.169e-05, [6]
        [build]: 2.61999e-06
        [elim_shapecalc]: 6.79999e-06
        [elim_not_effective]: 1.142e-05
        [opt_reshape]: 5.29e-06
        [fold_const_symbol]: 8.40999e-06
        [renormalize]: 4.80009e-07
[validate]: 2.662e-05
[backend_pass]: 8.49977e-07
[task_emit]: 0.0401603
[execute]: 9.69999e-06
Sums
    bootstrap                            :     0.000272s :  0.57%
    type_inference                       :     0.004162s :  8.67%
    event_method                         :     0.000015s :  0.03%
    auto_monad                           :     0.000054s :  0.11%
    graph_reusing                        :     0.000004s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.00%
    py_interpret_to_execute              :     0.000031s :  0.07%
    rewriter_before_opt_a                :     0.000066s :  0.14%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000209s :  0.44%
    jit_opt_a.loop_unroll                :     0.000034s :  0.07%
    jit_opt_a.a_1                        :     0.000573s :  1.19%
    jit_opt_a.with_stream_mark           :     0.000019s :  0.04%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.00%
    jit_opt_a.specialize_transform       :     0.000011s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.02%
    jit_opt_a.merge_forward              :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.05%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.01%
    jit_opt_a.replace_old_param          :     0.000015s :  0.03%
    jit_opt_a.inline_without_move        :     0.000010s :  0.02%
    jit_opt_a.renormalize                :     0.000553s :  1.15%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.03%
    jit_opt_a.cse                        :     0.000026s :  0.05%
    jit_opt_a.replace_applicator         :     0.000016s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.02%
    rewriter_after_opt_a                 :     0.000028s :  0.06%
    convert_after_rewriter               :     0.000006s :  0.01%
    order_py_execute_after_rewriter      :     0.000004s :  0.01%
    mutable_eliminate                    :     0.000472s :  0.98%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.03%
    cconv                                :     0.000019s :  0.04%
    loop_unroll                          :     0.000409s :  0.85%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.04%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000016s :  0.03%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.01%
    remove_dup_value                     :     0.000008s :  0.02%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000004s :  0.01%
    add_recomputation                    :     0.000035s :  0.07%
    cse_after_recomputation.cse          :     0.000008s :  0.02%
    auto_monad_reorder                   :     0.000011s :  0.02%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.01%
    opt_after_jit_grad                   :     0.000438s :  0.91%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.02%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.06%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.040160s : 83.65%
    execute                              :     0.000010s :  0.02%


Time group info:
------[substitution.]   0.000164    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.78% :     0.000001s :      2: substitution.fold_const_symbol
 2.22% :     0.000004s :      3: substitution.graph_param_transform
83.72% :     0.000137s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.59% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.25% :     0.000004s :      2: substitution.replace_old_param
 5.90% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004110     2
80.10% :     0.003293s :      1: type_inference.infer
19.90% :     0.000818s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.35% :     0.000042s :      4: replace.inline
32.65% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000143     6
94.02% :     0.000135s :      4: match.inline
 5.98% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 1.94% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000001s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000001s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.58% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.33% :     0.000000s :      3: predicate.graph_param_transform
 5.01% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.78% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.90% :     0.000002s :     15: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.66% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.16% :     0.000001s :     10: predicate.minmaximum_grad
 1.47% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000000s :      3: predicate.opt_reshape
 2.15% :     0.000002s :     15: predicate.partial_eliminate
 1.27% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.83% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.52% :     0.000001s :      3: predicate.reset_defer_inline
 1.69% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.32% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.93% :     0.000009s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.43% :     0.000002s :     10: predicate.transpose_eliminate
 1.35% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.41% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000550    12
45.03% :     0.000248s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.14% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.83% :     0.000274s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.053172    76
 0.07% :     0.000038s :      1: add_recomputation
 0.11% :     0.000058s :      1: auto_monad
 0.03% :     0.000013s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.54% :     0.000285s :      1: bootstrap
 0.04% :     0.000022s :      1: cconv
 0.01% :     0.000008s :      1: convert_after_rewriter
 0.04% :     0.000021s :      1: cse_after_recomputation
 0.01% :     0.000007s :      1: environ_conv
 0.04% :     0.000019s :      1: event_method
 0.03% :     0.000015s :      1: execute
 0.01% :     0.000004s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000006s :      1: graph_reusing
 9.20% :     0.004891s :      1: jit_opt_a
 0.25% :     0.000132s :      1: jit_opt_after_cconv
 0.10% :     0.000051s :      1: jit_opt_b
 0.78% :     0.000416s :      1: loop_unroll
 0.90% :     0.000479s :      1: mutable_eliminate
 1.75% :     0.000931s :     26: opt.transform.jit_opt_a
 0.08% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000025s :      4: opt.transform.jit_opt_b
 0.02% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.04% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.05% :     0.000029s :      4: opt.transform.symbol_engine_opt
 0.84% :     0.000444s :      1: opt_after_jit_grad
 0.01% :     0.000006s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.06% :     0.000034s :      1: py_interpret_to_execute
 0.02% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000010s :      1: remove_dup_value
 0.52% :     0.000277s :      1: renormalize.infer
 0.51% :     0.000270s :      1: renormalize.specialize
 0.01% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.06% :     0.000031s :      1: rewriter_after_opt_a
 0.13% :     0.000069s :      1: rewriter_before_opt_a
 0.13% :     0.000070s :      1: symbol_engine_optimizer
75.56% :     0.040179s :      1: task_emit
 7.85% :     0.004172s :      1: type_inference
 0.08% :     0.000044s :      1: validate


TotalTime = 0.0364269, [33]
[bootstrap]: 0.00024822
[type_inference]: 0.0129522
[event_method]: 5.88e-05
[auto_monad]: 0.00010457
[graph_reusing]: 7.23999e-06
[pre_auto_parallel]: 2.98998e-06
[py_interpret_to_execute]: 2.77e-05
[rewriter_before_opt_a]: 0.00010274
[expand_dump_flag]: 3.6e-06
[jit_opt_a]: 0.0121739, [3]
    [Cycle 1]: 0.00618959, [27]
        [switch_simplify]: 0.00012245
        [loop_unroll]: 4.183e-05
        [a_1]: 0.00085678
        [with_stream_mark]: 2.339e-05
        [recompute_prepare]: 1.69e-05
        [updatestate_depend_eliminate]: 7.49002e-06
        [updatestate_assign_eliminate]: 6.81999e-06
        [updatestate_loads_eliminate]: 6.07001e-06
        [parameter_eliminate]: 2.94999e-06
        [specialize_transform]: 1.272e-05
        [updatestate_useless_node_eliminater]: 1.167e-05
        [accelerated_algorithm]: 1.216e-05
        [meta_shard_fg_expand]: 3.7e-06
        [get_grad_eliminate_]: 1.136e-05
        [merge_forward]: 7.51999e-06
        [cell_reuse_recompute_pass]: 1.20999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.538e-05
        [j_node_and_user_rematch]: 2.201e-05
        [meta_fg_expand]: 0.0014965
        [replace_old_param]: 5.695e-05
        [inline_without_move]: 5.181e-05
        [renormalize]: 0.00285696
        [add_forward_monad_depend]: 1.234e-05
        [auto_monad_grad]: 6.22001e-06
        [auto_monad_eliminator]: 5.064e-05
        [cse]: 0.00016842
        [replace_applicator]: 7.558e-05
    [Cycle 2]: 0.00224653, [27]
        [switch_simplify]: 3.614e-05
        [loop_unroll]: 3.376e-05
        [a_1]: 0.00110787
        [with_stream_mark]: 1.803e-05
        [recompute_prepare]: 1.018e-05
        [updatestate_depend_eliminate]: 4.19997e-06
        [updatestate_assign_eliminate]: 3.83001e-06
        [updatestate_loads_eliminate]: 2.59001e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 7.06001e-06
        [updatestate_useless_node_eliminater]: 7.08998e-06
        [accelerated_algorithm]: 7.38999e-06
        [meta_shard_fg_expand]: 1.89999e-06
        [get_grad_eliminate_]: 6.50002e-06
        [merge_forward]: 4.00998e-06
        [cell_reuse_recompute_pass]: 9.80013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.441e-05
        [j_node_and_user_rematch]: 1.044e-05
        [meta_fg_expand]: 0.00011342
        [replace_old_param]: 1.243e-05
        [inline_without_move]: 6.04001e-06
        [renormalize]: 0.00060933
        [add_forward_monad_depend]: 5.27001e-06
        [auto_monad_grad]: 1.55999e-06
        [auto_monad_eliminator]: 1.303e-05
        [cse]: 2.528e-05
        [replace_applicator]: 1.521e-05
    [Cycle 3]: 0.00036478, [27]
        [switch_simplify]: 6.97002e-06
        [loop_unroll]: 5.73997e-06
        [a_1]: 0.00011565
        [with_stream_mark]: 1.025e-05
        [recompute_prepare]: 5.99999e-06
        [updatestate_depend_eliminate]: 3.66001e-06
        [updatestate_assign_eliminate]: 2.67001e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 1.35001e-06
        [specialize_transform]: 6.04001e-06
        [updatestate_useless_node_eliminater]: 5.82001e-06
        [accelerated_algorithm]: 5.66e-06
        [meta_shard_fg_expand]: 1.59e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 2.99001e-06
        [cell_reuse_recompute_pass]: 1.48002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.56e-05
        [j_node_and_user_rematch]: 1.029e-05
        [meta_fg_expand]: 1.84998e-06
        [replace_old_param]: 8.33999e-06
        [inline_without_move]: 5.77001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 5.79e-06
        [cse]: 1.397e-05
        [replace_applicator]: 6.29999e-06
[py_interpret_to_execute_after_opt_a]: 1.39e-05
[rewriter_after_opt_a]: 3.936e-05
[convert_after_rewriter]: 7.56001e-06
[order_py_execute_after_rewriter]: 6.04001e-06
[mutable_eliminate]: 0.00074137
[jit_opt_b]: 5.732e-05, [1]
    [Cycle 1]: 4.924e-05, [2]
        [frontend_op_eliminate]: 1.927e-05
        [inline_after_opt_a]: 1.909e-05
[cconv]: 2.708e-05
[loop_unroll]: 0.00047429
[jit_opt_after_cconv]: 0.00016826, [1]
    [Cycle 1]: 0.00016115, [11]
        [c_1]: 2.699e-05
        [parameter_eliminate]: 3.7e-06
        [updatestate_depend_eliminate]: 7.39002e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.74001e-06
        [cse]: 2.946e-05
        [call_graph_tuple_transform]: 2.3e-05
        [tuple_list_get_item_eliminator]: 5.92001e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 9.70002e-07
        [switch_simplify]: 6.44999e-06
[remove_dup_value]: 1.703e-05
[partial_unused_args_eliminate]: 2.02001e-06
[environ_conv]: 5.74e-06
[add_recomputation]: 5.695e-05
[cse_after_recomputation]: 4.591e-05, [1]
    [Cycle 1]: 2.641e-05, [1]
        [cse]: 1.762e-05
[auto_monad_reorder]: 1.935e-05
[get_jit_bprop_graph]: 1.94e-06
[rewriter_after_jit_bprop_graph]: 6.41998e-06
[opt_after_jit_grad]: 0.00053919
[symbol_engine_optimizer]: 8.692e-05, [1]
    [Cycle 1]: 8.058e-05, [6]
        [build]: 5.88002e-06
        [elim_shapecalc]: 8.18001e-06
        [elim_not_effective]: 1.635e-05
        [opt_reshape]: 7.66001e-06
        [fold_const_symbol]: 1.112e-05
        [renormalize]: 9.39996e-07
[validate]: 4.32e-05
[backend_pass]: 1.04e-06
[task_emit]: 0.00815603
[execute]: 8.59002e-06
Sums
    bootstrap                            :     0.000248s :  0.77%
    type_inference                       :     0.012952s : 40.23%
    event_method                         :     0.000059s :  0.18%
    auto_monad                           :     0.000105s :  0.32%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.09%
    rewriter_before_opt_a                :     0.000103s :  0.32%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000166s :  0.51%
    jit_opt_a.loop_unroll                :     0.000081s :  0.25%
    jit_opt_a.a_1                        :     0.002080s :  6.46%
    jit_opt_a.with_stream_mark           :     0.000052s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000025s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.07%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000055s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000043s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.001612s :  5.01%
    jit_opt_a.replace_old_param          :     0.000078s :  0.24%
    jit_opt_a.inline_without_move        :     0.000064s :  0.20%
    jit_opt_a.renormalize                :     0.003466s : 10.77%
    jit_opt_a.add_forward_monad_depend   :     0.000019s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000069s :  0.22%
    jit_opt_a.cse                        :     0.000208s :  0.65%
    jit_opt_a.replace_applicator         :     0.000097s :  0.30%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.04%
    rewriter_after_opt_a                 :     0.000039s :  0.12%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000741s :  2.30%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.06%
    cconv                                :     0.000027s :  0.08%
    loop_unroll                          :     0.000474s :  1.47%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000017s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000057s :  0.18%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000019s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000539s :  1.67%
    symbol_engine_optimizer.build        :     0.000006s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008156s : 25.33%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000522   113
 0.45% :     0.000002s :      3: substitution.elim_not_effective
 0.39% :     0.000002s :      3: substitution.fold_const_symbol
 1.12% :     0.000006s :      4: substitution.graph_param_transform
63.32% :     0.000330s :     14: substitution.inline
 3.35% :     0.000018s :      2: substitution.inline_without_move
 1.71% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.14% :     0.000011s :      7: substitution.minmaximum_grad
 1.36% :     0.000007s :      5: substitution.partial_eliminate
 2.10% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.58% :     0.000029s :     10: substitution.replace_applicator
 1.57% :     0.000008s :      7: substitution.replace_old_param
 0.41% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.44% :     0.000007s :      1: substitution.switch_simplify
 4.20% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.93% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 7.93% :     0.000041s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012874     2
89.86% :     0.011568s :      1: type_inference.infer
10.14% :     0.001305s :      1: type_inference.specialize
------[replace.]   0.000187    22
54.38% :     0.000102s :     14: replace.inline
17.85% :     0.000033s :      1: replace.switch_simplify
27.77% :     0.000052s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000350    22
92.04% :     0.000322s :     14: match.inline
 1.97% :     0.000007s :      1: match.switch_simplify
 5.99% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000379  2634
 1.69% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.52% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.41% :     0.000005s :     44: predicate.addn_check_dump
 2.06% :     0.000008s :     44: predicate.addn_zero_filter
 1.89% :     0.000007s :     44: predicate.arithmetic_simplify
 1.46% :     0.000006s :     44: predicate.cast_eliminate
 0.17% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.46% :     0.000006s :     44: predicate.compare_switch_simplify
 1.55% :     0.000006s :     44: predicate.depend_value_elim
 1.44% :     0.000005s :     44: predicate.dict_get_item_const_eliminator
 1.50% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.48% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.36% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.28% :     0.000001s :      4: predicate.elim_not_effective
 0.19% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.42% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.43% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.51% :     0.000006s :     44: predicate.environ_get_eliminate
 1.44% :     0.000005s :     44: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.77% :     0.000003s :     19: predicate.get_grad_eliminate
 0.09% :     0.000000s :      4: predicate.graph_param_transform
 4.43% :     0.000017s :     73: predicate.inline
 1.73% :     0.000007s :     40: predicate.inline_without_move
 0.34% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.94% :     0.000004s :     19: predicate.less_batch_normalization
 1.82% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 2.03% :     0.000008s :     55: predicate.load_eliminater
 0.47% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.22% :     0.000012s :     91: predicate.loop_unroll_before_grad
 1.84% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.47% :     0.000006s :     44: predicate.merge_addn
 1.45% :     0.000006s :     44: predicate.minmaximum_grad
 0.59% :     0.000002s :      4: predicate.mutable_eliminate
 0.21% :     0.000001s :      4: predicate.opt_reshape
 2.31% :     0.000009s :     55: predicate.partial_eliminate
 1.48% :     0.000006s :     44: predicate.print_const_string_wrapper
 2.03% :     0.000008s :     44: predicate.reduce_eliminate
 1.81% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.60% :     0.000010s :     97: predicate.replace_applicator
 0.97% :     0.000004s :     40: predicate.replace_old_param
 0.14% :     0.000001s :      4: predicate.reset_defer_inline
 2.01% :     0.000008s :     44: predicate.reshape_eliminate
 1.56% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.27% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.56% :     0.000006s :     44: predicate.same_eliminate
 0.46% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.39% :     0.000001s :      8: predicate.special_op_eliminate
 0.84% :     0.000003s :     19: predicate.specialize_transform
 1.63% :     0.000006s :     44: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.16% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.80% :     0.000011s :     65: predicate.switch_defer_inline
 2.57% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.48% :     0.000025s :    162: predicate.switch_simplify
 1.44% :     0.000005s :     44: predicate.tile_eliminate
 1.48% :     0.000006s :     44: predicate.transpose_eliminate
 1.79% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.59% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 3.09% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.90% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000007s :     44: predicate.value_based_eliminate
 0.15% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.20% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001709    32
59.50% :     0.001017s :     14: func_graph_cloner_run.FuncGraphClonerGraph
40.50% :     0.000692s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042775    91
 0.14% :     0.000060s :      1: add_recomputation
 0.26% :     0.000109s :      1: auto_monad
 0.05% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.63% :     0.000269s :      1: bootstrap
 0.07% :     0.000030s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000049s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.15% :     0.000064s :      1: event_method
 0.03% :     0.000014s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
28.47% :     0.012177s :      1: jit_opt_a
 0.40% :     0.000171s :      1: jit_opt_after_cconv
 0.14% :     0.000060s :      1: jit_opt_b
 1.13% :     0.000484s :      1: loop_unroll
 1.76% :     0.000753s :      1: mutable_eliminate
 6.42% :     0.002748s :     39: opt.transform.jit_opt_a
 0.14% :     0.000059s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000031s :      4: opt.transform.jit_opt_b
 0.03% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000040s :      4: opt.transform.symbol_engine_opt
 1.29% :     0.000550s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000030s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 3.82% :     0.001633s :      2: renormalize.infer
 4.25% :     0.001818s :      2: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000042s :      1: rewriter_after_opt_a
 0.25% :     0.000106s :      1: rewriter_before_opt_a
 0.21% :     0.000089s :      1: symbol_engine_optimizer
19.11% :     0.008173s :      1: task_emit
30.31% :     0.012966s :      1: type_inference
 0.16% :     0.000067s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.276.975 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dd27dc0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.277.617 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0182149, [33]
[bootstrap]: 0.00029354
[type_inference]: 0.00426031
[event_method]: 1.572e-05
[auto_monad]: 6.119e-05
[graph_reusing]: 5.79e-06
[pre_auto_parallel]: 2.36e-06
[py_interpret_to_execute]: 2.756e-05
[rewriter_before_opt_a]: 7.088e-05
[expand_dump_flag]: 3.04999e-06
[jit_opt_a]: 0.00505777, [2]
    [Cycle 1]: 0.00141868, [27]
        [switch_simplify]: 5.235e-05
        [loop_unroll]: 2.896e-05
        [a_1]: 0.00051813
        [with_stream_mark]: 1.601e-05
        [recompute_prepare]: 7.41999e-06
        [updatestate_depend_eliminate]: 3.52997e-06
        [updatestate_assign_eliminate]: 2.80002e-06
        [updatestate_loads_eliminate]: 2.58003e-06
        [parameter_eliminate]: 1.92999e-06
        [specialize_transform]: 5.74e-06
        [updatestate_useless_node_eliminater]: 4.73001e-06
        [accelerated_algorithm]: 5.62001e-06
        [meta_shard_fg_expand]: 1.97999e-06
        [get_grad_eliminate_]: 5.05999e-06
        [merge_forward]: 3.58e-06
        [cell_reuse_recompute_pass]: 1.12999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.332e-05
        [j_node_and_user_rematch]: 8.55999e-06
        [meta_fg_expand]: 2.12999e-06
        [replace_old_param]: 8.74998e-06
        [inline_without_move]: 5.05001e-06
        [renormalize]: 0.00048952
        [add_forward_monad_depend]: 9.14e-06
        [auto_monad_grad]: 1.93002e-06
        [auto_monad_eliminator]: 1.298e-05
        [cse]: 2.865e-05
        [replace_applicator]: 1.047e-05
    [Cycle 2]: 0.00031693, [27]
        [switch_simplify]: 5.50001e-06
        [loop_unroll]: 4.67998e-06
        [a_1]: 8.788e-05
        [with_stream_mark]: 9.38002e-06
        [recompute_prepare]: 5.47999e-06
        [updatestate_depend_eliminate]: 2.66e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 4.87998e-06
        [updatestate_useless_node_eliminater]: 4.52e-06
        [accelerated_algorithm]: 4.90001e-06
        [meta_shard_fg_expand]: 1.13001e-06
        [get_grad_eliminate_]: 4.22e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.96e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.307e-05
        [j_node_and_user_rematch]: 8.2e-06
        [meta_fg_expand]: 1.47999e-06
        [replace_old_param]: 6.53e-06
        [inline_without_move]: 4.81002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.51998e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 5.23002e-06
        [cse]: 1.028e-05
        [replace_applicator]: 4.99e-06
[py_interpret_to_execute_after_opt_a]: 9.17001e-06
[rewriter_after_opt_a]: 3.675e-05
[convert_after_rewriter]: 6.78e-06
[order_py_execute_after_rewriter]: 4.84e-06
[mutable_eliminate]: 0.00055459
[jit_opt_b]: 4.975e-05, [1]
    [Cycle 1]: 4.298e-05, [2]
        [frontend_op_eliminate]: 1.672e-05
        [inline_after_opt_a]: 1.421e-05
[cconv]: 2.455e-05
[loop_unroll]: 0.00041735
[jit_opt_after_cconv]: 0.00013134, [1]
    [Cycle 1]: 0.00012482, [11]
        [c_1]: 1.961e-05
        [parameter_eliminate]: 2.22999e-06
        [updatestate_depend_eliminate]: 5.40999e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 1.655e-05
        [call_graph_tuple_transform]: 1.674e-05
        [tuple_list_get_item_eliminator]: 5.28002e-06
        [none_parameter_eliminate]: 1.27999e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 4.90999e-06
[remove_dup_value]: 1.303e-05
[partial_unused_args_eliminate]: 2.25002e-06
[environ_conv]: 4.89998e-06
[add_recomputation]: 4.702e-05
[cse_after_recomputation]: 2.012e-05, [1]
    [Cycle 1]: 1.492e-05, [1]
        [cse]: 9.18002e-06
[auto_monad_reorder]: 1.496e-05
[get_jit_bprop_graph]: 1.62001e-06
[rewriter_after_jit_bprop_graph]: 4.48001e-06
[opt_after_jit_grad]: 0.00050375
[symbol_engine_optimizer]: 7.086e-05, [1]
    [Cycle 1]: 6.473e-05, [6]
        [build]: 3.53999e-06
        [elim_shapecalc]: 7.63999e-06
        [elim_not_effective]: 1.211e-05
        [opt_reshape]: 5.14998e-06
        [fold_const_symbol]: 8.13999e-06
        [renormalize]: 3.69997e-07
[validate]: 3.064e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00628369
[execute]: 7.93001e-06
Sums
    bootstrap                            :     0.000294s :  2.05%
    type_inference                       :     0.004260s : 29.76%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000061s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.19%
    rewriter_before_opt_a                :     0.000071s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000606s :  4.23%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000490s :  3.42%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000039s :  0.27%
    jit_opt_a.replace_applicator         :     0.000015s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000037s :  0.26%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000555s :  3.87%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000417s :  2.92%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000047s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000504s :  3.52%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006284s : 43.89%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000188    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.57% :     0.000005s :      3: substitution.graph_param_transform
83.34% :     0.000156s :      4: substitution.inline
 1.64% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.41% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.89% :     0.000004s :      2: substitution.replace_old_param
 6.48% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004200     2
82.36% :     0.003459s :      1: type_inference.infer
17.64% :     0.000741s :      1: type_inference.specialize
------[replace.]   0.000061     6
67.17% :     0.000041s :      4: replace.inline
32.83% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000165     6
93.28% :     0.000154s :      4: match.inline
 6.72% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.57% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.30% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000001s :     10: predicate.addn_zero_filter
 1.93% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000001s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000001s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.44% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.59% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.47% :     0.000002s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.92% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.89% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.34% :     0.000003s :     15: predicate.partial_eliminate
 1.44% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.35% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.56% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.80% :     0.000009s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.25% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.06% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.96% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000482    12
48.76% :     0.000235s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.48% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.75% :     0.000221s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019614    76
 0.25% :     0.000050s :      1: add_recomputation
 0.33% :     0.000064s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.57% :     0.000307s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.80% :     0.005060s :      1: jit_opt_a
 0.68% :     0.000134s :      1: jit_opt_after_cconv
 0.27% :     0.000052s :      1: jit_opt_b
 2.17% :     0.000426s :      1: loop_unroll
 2.87% :     0.000562s :      1: mutable_eliminate
 4.11% :     0.000807s :     26: opt.transform.jit_opt_a
 0.22% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.61% :     0.000512s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.23% :     0.000242s :      1: renormalize.infer
 1.22% :     0.000240s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000040s :      1: rewriter_after_opt_a
 0.38% :     0.000074s :      1: rewriter_before_opt_a
 0.37% :     0.000073s :      1: symbol_engine_optimizer
32.09% :     0.006294s :      1: task_emit
21.78% :     0.004271s :      1: type_inference
 0.25% :     0.000048s :      1: validate


TotalTime = 0.0190299, [33]
[bootstrap]: 0.00030068
[type_inference]: 0.00465941
[event_method]: 1.728e-05
[auto_monad]: 6.256e-05
[graph_reusing]: 5.74e-06
[pre_auto_parallel]: 2.59001e-06
[py_interpret_to_execute]: 3.349e-05
[rewriter_before_opt_a]: 7.333e-05
[expand_dump_flag]: 3.33e-06
[jit_opt_a]: 0.0051955, [2]
    [Cycle 1]: 0.00163735, [27]
        [switch_simplify]: 5.102e-05
        [loop_unroll]: 2.884e-05
        [a_1]: 0.00051104
        [with_stream_mark]: 4.419e-05
        [recompute_prepare]: 7.43999e-06
        [updatestate_depend_eliminate]: 4.07e-06
        [updatestate_assign_eliminate]: 3.10998e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 5.62001e-06
        [updatestate_useless_node_eliminater]: 4.78001e-06
        [accelerated_algorithm]: 5.44e-06
        [meta_shard_fg_expand]: 2.32001e-06
        [get_grad_eliminate_]: 5.46002e-06
        [merge_forward]: 3.69002e-06
        [cell_reuse_recompute_pass]: 1.03001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.309e-05
        [j_node_and_user_rematch]: 9.15001e-06
        [meta_fg_expand]: 2.37001e-06
        [replace_old_param]: 9.38002e-06
        [inline_without_move]: 5.02e-06
        [renormalize]: 0.00068458
        [add_forward_monad_depend]: 5.69999e-06
        [auto_monad_grad]: 2.65002e-06
        [auto_monad_eliminator]: 1.327e-05
        [cse]: 3.098e-05
        [replace_applicator]: 1.162e-05
    [Cycle 2]: 0.00031476, [27]
        [switch_simplify]: 5.56002e-06
        [loop_unroll]: 4.80999e-06
        [a_1]: 8.863e-05
        [with_stream_mark]: 9.33002e-06
        [recompute_prepare]: 4.90999e-06
        [updatestate_depend_eliminate]: 2.88e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 9.29984e-07
        [specialize_transform]: 4.79998e-06
        [updatestate_useless_node_eliminater]: 4.96002e-06
        [accelerated_algorithm]: 5.00001e-06
        [meta_shard_fg_expand]: 1.35001e-06
        [get_grad_eliminate_]: 4.56002e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.62001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.161e-05
        [j_node_and_user_rematch]: 7.45998e-06
        [meta_fg_expand]: 1.41002e-06
        [replace_old_param]: 1.238e-05
        [inline_without_move]: 4.66002e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.06002e-06
        [auto_monad_grad]: 8.89995e-07
        [auto_monad_eliminator]: 4.79998e-06
        [cse]: 1.086e-05
        [replace_applicator]: 4.62998e-06
[py_interpret_to_execute_after_opt_a]: 1.021e-05
[rewriter_after_opt_a]: 3.517e-05
[convert_after_rewriter]: 6.98998e-06
[order_py_execute_after_rewriter]: 4.94e-06
[mutable_eliminate]: 0.00049642
[jit_opt_b]: 4.866e-05, [1]
    [Cycle 1]: 4.144e-05, [2]
        [frontend_op_eliminate]: 1.54e-05
        [inline_after_opt_a]: 1.392e-05
[cconv]: 2.518e-05
[loop_unroll]: 0.00041216
[jit_opt_after_cconv]: 0.00013587, [1]
    [Cycle 1]: 0.00012941, [11]
        [c_1]: 1.993e-05
        [parameter_eliminate]: 2.58e-06
        [updatestate_depend_eliminate]: 5.27001e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 1.738e-05
        [call_graph_tuple_transform]: 1.773e-05
        [tuple_list_get_item_eliminator]: 5.10001e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 6.40022e-07
        [switch_simplify]: 5.02e-06
[remove_dup_value]: 1.445e-05
[partial_unused_args_eliminate]: 2.20002e-06
[environ_conv]: 4.90001e-06
[add_recomputation]: 5.089e-05
[cse_after_recomputation]: 1.944e-05, [1]
    [Cycle 1]: 1.427e-05, [1]
        [cse]: 8.79e-06
[auto_monad_reorder]: 1.431e-05
[get_jit_bprop_graph]: 1.94e-06
[rewriter_after_jit_bprop_graph]: 4.34997e-06
[opt_after_jit_grad]: 0.00044922
[symbol_engine_optimizer]: 7.189e-05, [1]
    [Cycle 1]: 6.551e-05, [6]
        [build]: 4.17003e-06
        [elim_shapecalc]: 7.06999e-06
        [elim_not_effective]: 1.253e-05
        [opt_reshape]: 5.25001e-06
        [fold_const_symbol]: 8.55999e-06
        [renormalize]: 4.59986e-07
[validate]: 3.725e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.00657867
[execute]: 9.42001e-06
Sums
    bootstrap                            :     0.000301s :  1.98%
    type_inference                       :     0.004659s : 30.76%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000063s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.22%
    rewriter_before_opt_a                :     0.000073s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.37%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000600s :  3.96%
    jit_opt_a.with_stream_mark           :     0.000054s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.14%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000685s :  4.52%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000042s :  0.28%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000496s :  3.28%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000412s :  2.72%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000051s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000449s :  2.97%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006579s : 43.43%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000202    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.67% :     0.000005s :      3: substitution.graph_param_transform
81.42% :     0.000165s :      4: substitution.inline
 1.60% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.06% :     0.000004s :      4: substitution.remove_not_recompute_node
 4.66% :     0.000009s :      2: substitution.replace_old_param
 5.93% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004596     2
81.30% :     0.003737s :      1: type_inference.infer
18.70% :     0.000859s :      1: type_inference.specialize
------[replace.]   0.000063     6
66.15% :     0.000042s :      4: replace.inline
33.85% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000173     6
93.75% :     0.000162s :      4: match.inline
 6.25% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.28% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.29% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.06% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000001s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.36% :     0.000000s :      3: predicate.graph_param_transform
 5.71% :     0.000006s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.93% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 2.14% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000002s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.57% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.20% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.75% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.03% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.39% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000620    12
49.10% :     0.000304s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.79% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.10% :     0.000279s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020612    76
 0.26% :     0.000054s :      1: add_recomputation
 0.32% :     0.000066s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.55% :     0.000319s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.22% :     0.005199s :      1: jit_opt_a
 0.67% :     0.000139s :      1: jit_opt_after_cconv
 0.25% :     0.000051s :      1: jit_opt_b
 2.04% :     0.000421s :      1: loop_unroll
 2.45% :     0.000505s :      1: mutable_eliminate
 3.91% :     0.000805s :     26: opt.transform.jit_opt_a
 0.22% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.22% :     0.000457s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.33% :     0.000068s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.81% :     0.000373s :      1: renormalize.infer
 1.48% :     0.000304s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000038s :      1: rewriter_after_opt_a
 0.37% :     0.000077s :      1: rewriter_before_opt_a
 0.36% :     0.000074s :      1: symbol_engine_optimizer
32.00% :     0.006596s :      1: task_emit
22.67% :     0.004673s :      1: type_inference
 0.28% :     0.000058s :      1: validate


TotalTime = 0.0360326, [33]
[bootstrap]: 0.00023137
[type_inference]: 0.01295
[event_method]: 8.112e-05
[auto_monad]: 0.00010588
[graph_reusing]: 6.98e-06
[pre_auto_parallel]: 3.5e-06
[py_interpret_to_execute]: 2.774e-05
[rewriter_before_opt_a]: 0.00010122
[expand_dump_flag]: 3.55e-06
[jit_opt_a]: 0.0128886, [3]
    [Cycle 1]: 0.00625865, [27]
        [switch_simplify]: 0.00013811
        [loop_unroll]: 4.22e-05
        [a_1]: 0.00090652
        [with_stream_mark]: 2.923e-05
        [recompute_prepare]: 1.833e-05
        [updatestate_depend_eliminate]: 8.70001e-06
        [updatestate_assign_eliminate]: 6.30002e-06
        [updatestate_loads_eliminate]: 5.87001e-06
        [parameter_eliminate]: 2.86e-06
        [specialize_transform]: 1.321e-05
        [updatestate_useless_node_eliminater]: 1.159e-05
        [accelerated_algorithm]: 1.23e-05
        [meta_shard_fg_expand]: 3.26001e-06
        [get_grad_eliminate_]: 1.188e-05
        [merge_forward]: 7.9e-06
        [cell_reuse_recompute_pass]: 9.5999e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.645e-05
        [j_node_and_user_rematch]: 2.35e-05
        [meta_fg_expand]: 0.00161674
        [replace_old_param]: 5.775e-05
        [inline_without_move]: 5.462e-05
        [renormalize]: 0.00272043
        [add_forward_monad_depend]: 1.089e-05
        [auto_monad_grad]: 5.94e-06
        [auto_monad_eliminator]: 4.852e-05
        [cse]: 0.00015681
        [replace_applicator]: 6.744e-05
    [Cycle 2]: 0.00224918, [27]
        [switch_simplify]: 3.532e-05
        [loop_unroll]: 3.361e-05
        [a_1]: 0.00105789
        [with_stream_mark]: 1.931e-05
        [recompute_prepare]: 9.58997e-06
        [updatestate_depend_eliminate]: 3.95998e-06
        [updatestate_assign_eliminate]: 4.13001e-06
        [updatestate_loads_eliminate]: 3.09999e-06
        [parameter_eliminate]: 1.77001e-06
        [specialize_transform]: 7.38e-06
        [updatestate_useless_node_eliminater]: 6.51e-06
        [accelerated_algorithm]: 6.43e-06
        [meta_shard_fg_expand]: 3.09999e-06
        [get_grad_eliminate_]: 5.80002e-06
        [merge_forward]: 3.61001e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.736e-05
        [j_node_and_user_rematch]: 1.112e-05
        [meta_fg_expand]: 0.00012842
        [replace_old_param]: 1.243e-05
        [inline_without_move]: 6.54001e-06
        [renormalize]: 0.00064837
        [add_forward_monad_depend]: 4.63001e-06
        [auto_monad_grad]: 2.11e-06
        [auto_monad_eliminator]: 1.317e-05
        [cse]: 2.781e-05
        [replace_applicator]: 1.46e-05
    [Cycle 3]: 0.0003618, [27]
        [switch_simplify]: 6.96999e-06
        [loop_unroll]: 6.34999e-06
        [a_1]: 0.00011755
        [with_stream_mark]: 1.012e-05
        [recompute_prepare]: 5.78997e-06
        [updatestate_depend_eliminate]: 3.45e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.46998e-06
        [parameter_eliminate]: 8.10018e-07
        [specialize_transform]: 5.92999e-06
        [updatestate_useless_node_eliminater]: 5.86998e-06
        [accelerated_algorithm]: 5.51998e-06
        [meta_shard_fg_expand]: 1.37999e-06
        [get_grad_eliminate_]: 5.50001e-06
        [merge_forward]: 2.74999e-06
        [cell_reuse_recompute_pass]: 1.47999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.562e-05
        [j_node_and_user_rematch]: 9.67001e-06
        [meta_fg_expand]: 1.86e-06
        [replace_old_param]: 8.03999e-06
        [inline_without_move]: 5.64e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 9.90025e-07
        [auto_monad_grad]: 7.80012e-07
        [auto_monad_eliminator]: 6.11e-06
        [cse]: 1.36e-05
        [replace_applicator]: 5.72999e-06
[py_interpret_to_execute_after_opt_a]: 1.202e-05
[rewriter_after_opt_a]: 3.992e-05
[convert_after_rewriter]: 7.68999e-06
[order_py_execute_after_rewriter]: 5.39e-06
[mutable_eliminate]: 0.00073333
[jit_opt_b]: 5.626e-05, [1]
    [Cycle 1]: 4.771e-05, [2]
        [frontend_op_eliminate]: 1.88e-05
        [inline_after_opt_a]: 1.764e-05
[cconv]: 2.551e-05
[loop_unroll]: 0.00046108
[jit_opt_after_cconv]: 0.00015706, [1]
    [Cycle 1]: 0.00015037, [11]
        [c_1]: 2.609e-05
        [parameter_eliminate]: 2.63e-06
        [updatestate_depend_eliminate]: 6.06998e-06
        [updatestate_assign_eliminate]: 2.74999e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [cse]: 2.409e-05
        [call_graph_tuple_transform]: 2.129e-05
        [tuple_list_get_item_eliminator]: 6.02999e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 6.24001e-06
[remove_dup_value]: 1.702e-05
[partial_unused_args_eliminate]: 2.49999e-06
[environ_conv]: 5.71e-06
[add_recomputation]: 5.252e-05
[cse_after_recomputation]: 3.579e-05, [1]
    [Cycle 1]: 2.209e-05, [1]
        [cse]: 1.581e-05
[auto_monad_reorder]: 1.878e-05
[get_jit_bprop_graph]: 2.32001e-06
[rewriter_after_jit_bprop_graph]: 5.39e-06
[opt_after_jit_grad]: 0.00052832
[symbol_engine_optimizer]: 8.137e-05, [1]
    [Cycle 1]: 7.482e-05, [6]
        [build]: 3.96001e-06
        [elim_shapecalc]: 8.88002e-06
        [elim_not_effective]: 1.547e-05
        [opt_reshape]: 7.23e-06
        [fold_const_symbol]: 1.017e-05
        [renormalize]: 5.59987e-07
[validate]: 3.982e-05
[backend_pass]: 1.02e-06
[task_emit]: 0.00712681
[execute]: 8.68001e-06
Sums
    bootstrap                            :     0.000231s :  0.74%
    type_inference                       :     0.012950s : 41.57%
    event_method                         :     0.000081s :  0.26%
    auto_monad                           :     0.000106s :  0.34%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.09%
    rewriter_before_opt_a                :     0.000101s :  0.32%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000180s :  0.58%
    jit_opt_a.loop_unroll                :     0.000082s :  0.26%
    jit_opt_a.a_1                        :     0.002082s :  6.68%
    jit_opt_a.with_stream_mark           :     0.000059s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000034s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000027s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.07%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000059s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000044s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.001747s :  5.61%
    jit_opt_a.replace_old_param          :     0.000078s :  0.25%
    jit_opt_a.inline_without_move        :     0.000067s :  0.21%
    jit_opt_a.renormalize                :     0.003369s : 10.81%
    jit_opt_a.add_forward_monad_depend   :     0.000017s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000068s :  0.22%
    jit_opt_a.cse                        :     0.000198s :  0.64%
    jit_opt_a.replace_applicator         :     0.000088s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000040s :  0.13%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000733s :  2.35%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000026s :  0.08%
    loop_unroll                          :     0.000461s :  1.48%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000017s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000053s :  0.17%
    cse_after_recomputation.cse          :     0.000016s :  0.05%
    auto_monad_reorder                   :     0.000019s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000528s :  1.70%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007127s : 22.88%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000547   113
 0.45% :     0.000002s :      3: substitution.elim_not_effective
 0.29% :     0.000002s :      3: substitution.fold_const_symbol
 1.20% :     0.000007s :      4: substitution.graph_param_transform
64.56% :     0.000353s :     14: substitution.inline
 3.56% :     0.000019s :      2: substitution.inline_without_move
 1.81% :     0.000010s :     14: substitution.j_node_and_user_rematch
 2.02% :     0.000011s :      7: substitution.minmaximum_grad
 1.13% :     0.000006s :      5: substitution.partial_eliminate
 2.23% :     0.000012s :     14: substitution.remove_not_recompute_node
 5.10% :     0.000028s :     10: substitution.replace_applicator
 1.48% :     0.000008s :      7: substitution.replace_old_param
 0.39% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.81% :     0.000010s :      1: substitution.switch_simplify
 3.96% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.57% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.44% :     0.000041s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012873     2
88.54% :     0.011397s :      1: type_inference.infer
11.46% :     0.001476s :      1: type_inference.specialize
------[replace.]   0.000189    22
53.49% :     0.000101s :     14: replace.inline
18.21% :     0.000034s :      1: replace.switch_simplify
28.30% :     0.000054s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000375    22
92.13% :     0.000346s :     14: match.inline
 2.48% :     0.000009s :      1: match.switch_simplify
 5.38% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000376  2634
 1.62% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.52% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.38% :     0.000005s :     44: predicate.addn_check_dump
 1.70% :     0.000006s :     44: predicate.addn_zero_filter
 1.88% :     0.000007s :     44: predicate.arithmetic_simplify
 1.53% :     0.000006s :     44: predicate.cast_eliminate
 0.15% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.38% :     0.000005s :     44: predicate.compare_switch_simplify
 1.58% :     0.000006s :     44: predicate.depend_value_elim
 1.42% :     0.000005s :     44: predicate.dict_get_item_const_eliminator
 1.47% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.54% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.34% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      4: predicate.elim_not_effective
 0.19% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.54% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.44% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.41% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.44% :     0.000005s :     44: predicate.environ_get_eliminate
 1.54% :     0.000006s :     44: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.84% :     0.000003s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.49% :     0.000017s :     73: predicate.inline
 1.83% :     0.000007s :     40: predicate.inline_without_move
 0.33% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.97% :     0.000004s :     19: predicate.less_batch_normalization
 1.71% :     0.000006s :     51: predicate.list_to_tuple_eliminator_
 1.96% :     0.000007s :     55: predicate.load_eliminater
 0.54% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.24% :     0.000012s :     91: predicate.loop_unroll_before_grad
 1.72% :     0.000006s :     48: predicate.make_slice_get_slice_eliminator
 1.42% :     0.000005s :     44: predicate.merge_addn
 1.48% :     0.000006s :     44: predicate.minmaximum_grad
 0.57% :     0.000002s :      4: predicate.mutable_eliminate
 0.21% :     0.000001s :      4: predicate.opt_reshape
 2.33% :     0.000009s :     55: predicate.partial_eliminate
 1.73% :     0.000007s :     44: predicate.print_const_string_wrapper
 1.85% :     0.000007s :     44: predicate.reduce_eliminate
 1.79% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.49% :     0.000009s :     97: predicate.replace_applicator
 1.04% :     0.000004s :     40: predicate.replace_old_param
 0.11% :     0.000000s :      4: predicate.reset_defer_inline
 1.83% :     0.000007s :     44: predicate.reshape_eliminate
 1.52% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.35% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.70% :     0.000006s :     44: predicate.same_eliminate
 0.45% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.32% :     0.000001s :      8: predicate.special_op_eliminate
 0.81% :     0.000003s :     19: predicate.specialize_transform
 1.65% :     0.000006s :     44: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000005s :     44: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.74% :     0.000010s :     65: predicate.switch_defer_inline
 2.57% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.70% :     0.000025s :    162: predicate.switch_simplify
 1.47% :     0.000006s :     44: predicate.tile_eliminate
 1.59% :     0.000006s :     44: predicate.transpose_eliminate
 1.82% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.81% :     0.000007s :     44: predicate.tuple_list_get_item_depend_reorder
 3.08% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 2.00% :     0.000008s :     44: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000006s :     51: predicate.tuple_to_list_eliminator_
 1.83% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000010s :     74: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000007s :     44: predicate.value_based_eliminate
 0.14% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.25% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001769    32
58.88% :     0.001042s :     14: func_graph_cloner_run.FuncGraphClonerGraph
41.12% :     0.000728s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042286    91
 0.13% :     0.000056s :      1: add_recomputation
 0.26% :     0.000111s :      1: auto_monad
 0.05% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.60% :     0.000253s :      1: bootstrap
 0.07% :     0.000028s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000038s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.20% :     0.000086s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
30.49% :     0.012893s :      1: jit_opt_a
 0.38% :     0.000160s :      1: jit_opt_after_cconv
 0.14% :     0.000059s :      1: jit_opt_b
 1.11% :     0.000469s :      1: loop_unroll
 1.76% :     0.000743s :      1: mutable_eliminate
 6.53% :     0.002762s :     39: opt.transform.jit_opt_a
 0.13% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000030s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.27% :     0.000537s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000030s :      1: py_interpret_to_execute
 0.03% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 4.43% :     0.001875s :      2: renormalize.infer
 3.49% :     0.001477s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000043s :      1: rewriter_after_opt_a
 0.25% :     0.000104s :      1: rewriter_before_opt_a
 0.20% :     0.000084s :      1: symbol_engine_optimizer
16.88% :     0.007138s :      1: task_emit
30.66% :     0.012966s :      1: type_inference
 0.15% :     0.000062s :      1: validate


TotalTime = 0.0535697, [30]
[bootstrap]: 0.00029132
[type_inference]: 0.0411575
[event_method]: 1.782e-05
[auto_monad]: 6.445e-05
[graph_reusing]: 5.80002e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 3.546e-05
[rewriter_before_opt_a]: 7.596e-05
[expand_dump_flag]: 3.36999e-06
[jit_opt_a]: 0.0056011, [2]
    [Cycle 1]: 0.0016191, [27]
        [switch_simplify]: 5.527e-05
        [loop_unroll]: 2.97e-05
        [a_1]: 0.00053993
        [with_stream_mark]: 2.249e-05
        [recompute_prepare]: 7.55003e-06
        [updatestate_depend_eliminate]: 3.43999e-06
        [updatestate_assign_eliminate]: 2.88998e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [parameter_eliminate]: 1.76e-06
        [specialize_transform]: 5.82999e-06
        [updatestate_useless_node_eliminater]: 5.30999e-06
        [accelerated_algorithm]: 5.29e-06
        [meta_shard_fg_expand]: 2.31e-06
        [get_grad_eliminate_]: 4.92999e-06
        [merge_forward]: 4.08999e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.504e-05
        [j_node_and_user_rematch]: 9.19e-06
        [meta_fg_expand]: 2.42001e-06
        [replace_old_param]: 8.92999e-06
        [inline_without_move]: 5.20001e-06
        [renormalize]: 0.00063838
        [add_forward_monad_depend]: 1.193e-05
        [auto_monad_grad]: 2.34999e-06
        [auto_monad_eliminator]: 1.36e-05
        [cse]: 2.794e-05
        [replace_applicator]: 1.08e-05
    [Cycle 2]: 0.00031531, [27]
        [switch_simplify]: 5.61e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 8.787e-05
        [with_stream_mark]: 9.84001e-06
        [recompute_prepare]: 5.12999e-06
        [updatestate_depend_eliminate]: 2.68e-06
        [updatestate_assign_eliminate]: 2.11998e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 5.02999e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.01997e-06
        [meta_shard_fg_expand]: 1.33002e-06
        [get_grad_eliminate_]: 4.45e-06
        [merge_forward]: 3.17997e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.335e-05
        [j_node_and_user_rematch]: 7.90998e-06
        [meta_fg_expand]: 1.57999e-06
        [replace_old_param]: 7.21001e-06
        [inline_without_move]: 4.77998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.14998e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 4.66002e-06
        [cse]: 9.46998e-06
        [replace_applicator]: 4.79998e-06
[py_interpret_to_execute_after_opt_a]: 1.103e-05
[rewriter_after_opt_a]: 3.454e-05
[convert_after_rewriter]: 8.70001e-06
[order_py_execute_after_rewriter]: 5.21002e-06
[mutable_eliminate]: 0.00053608
[jit_opt_b]: 4.753e-05, [1]
    [Cycle 1]: 4.052e-05, [2]
        [frontend_op_eliminate]: 1.517e-05
        [inline_after_opt_a]: 1.45e-05
[cconv]: 2.475e-05
[loop_unroll]: 0.0004005
[jit_opt_after_cconv]: 0.00013344, [1]
    [Cycle 1]: 0.00012753, [11]
        [c_1]: 2.047e-05
        [parameter_eliminate]: 2.59999e-06
        [updatestate_depend_eliminate]: 5.27001e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.646e-05
        [call_graph_tuple_transform]: 1.761e-05
        [tuple_list_get_item_eliminator]: 5.10999e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 5.20999e-06
[remove_dup_value]: 1.416e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 4.73001e-06
[add_recomputation]: 4.943e-05
[cse_after_recomputation]: 1.924e-05, [1]
    [Cycle 1]: 1.409e-05, [1]
        [cse]: 8.62e-06
[auto_monad_reorder]: 1.518e-05
[get_jit_bprop_graph]: 1.91e-06
[rewriter_after_jit_bprop_graph]: 5.05001e-06
[opt_after_jit_grad]: 0.00466061
[symbol_engine_optimizer]: 9.65e-05, [1]
    [Cycle 1]: 8.487e-05, [6]
        [build]: 8.00999e-06
        [elim_shapecalc]: 8.57e-06
        [elim_not_effective]: 2.228e-05
        [opt_reshape]: 6.19001e-06
        [fold_const_symbol]: 8.94e-06
        [renormalize]: 9.79984e-07
[validate]: 4.904e-05
Sums
    bootstrap                            :     0.000291s :  0.59%
    type_inference                       :     0.041158s : 83.50%
    event_method                         :     0.000018s :  0.04%
    auto_monad                           :     0.000064s :  0.13%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000035s :  0.07%
    rewriter_before_opt_a                :     0.000076s :  0.15%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000061s :  0.12%
    jit_opt_a.loop_unroll                :     0.000035s :  0.07%
    jit_opt_a.a_1                        :     0.000628s :  1.27%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.02%
    jit_opt_a.merge_forward              :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.01%
    jit_opt_a.replace_old_param          :     0.000016s :  0.03%
    jit_opt_a.inline_without_move        :     0.000010s :  0.02%
    jit_opt_a.renormalize                :     0.000638s :  1.30%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.04%
    jit_opt_a.cse                        :     0.000037s :  0.08%
    jit_opt_a.replace_applicator         :     0.000016s :  0.03%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.02%
    rewriter_after_opt_a                 :     0.000035s :  0.07%
    convert_after_rewriter               :     0.000009s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000536s :  1.09%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.03%
    cconv                                :     0.000025s :  0.05%
    loop_unroll                          :     0.000400s :  0.81%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.04%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000016s :  0.03%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.04%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.01%
    remove_dup_value                     :     0.000014s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000005s :  0.01%
    add_recomputation                    :     0.000049s :  0.10%
    cse_after_recomputation.cse          :     0.000009s :  0.02%
    auto_monad_reorder                   :     0.000015s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.004661s :  9.46%
    symbol_engine_optimizer.build        :     0.000008s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000022s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.02%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000049s :  0.10%


Time group info:
------[substitution.]   0.000218    23
 1.27% :     0.000003s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.43% :     0.000005s :      3: substitution.graph_param_transform
84.59% :     0.000185s :      4: substitution.inline
 1.29% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.38% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.98% :     0.000004s :      2: substitution.replace_old_param
 5.49% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.041089     2
97.95% :     0.040248s :      1: type_inference.infer
 2.05% :     0.000840s :      1: type_inference.specialize
------[replace.]   0.000065     6
66.62% :     0.000043s :      4: replace.inline
33.38% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000193     6
94.33% :     0.000182s :      4: match.inline
 5.67% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.19% :     0.000001s :     10: predicate.accumulaten_eliminater
 3.38% :     0.000004s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 1.71% :     0.000002s :     10: predicate.arithmetic_simplify
 1.15% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.64% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 2.46% :     0.000003s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 6.04% :     0.000007s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.86% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.67% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     15: predicate.partial_eliminate
 1.03% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.51% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.13% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.63% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.32% :     0.000009s :     53: predicate.switch_simplify
 1.10% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.84% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000580    12
50.19% :     0.000291s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.53% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.28% :     0.000245s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.055144    72
 0.09% :     0.000052s :      1: add_recomputation
 0.12% :     0.000068s :      1: auto_monad
 0.03% :     0.000018s :      1: auto_monad_reorder
 0.56% :     0.000311s :      1: bootstrap
 0.05% :     0.000027s :      1: cconv
 0.02% :     0.000011s :      1: convert_after_rewriter
 0.04% :     0.000021s :      1: cse_after_recomputation
 0.01% :     0.000007s :      1: environ_conv
 0.04% :     0.000022s :      1: event_method
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000008s :      1: graph_reusing
10.16% :     0.005605s :      1: jit_opt_a
 0.25% :     0.000136s :      1: jit_opt_after_cconv
 0.09% :     0.000050s :      1: jit_opt_b
 0.74% :     0.000408s :      1: loop_unroll
 0.99% :     0.000544s :      1: mutable_eliminate
 1.51% :     0.000835s :     26: opt.transform.jit_opt_a
 0.08% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000023s :      4: opt.transform.jit_opt_b
 0.02% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000083s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000042s :      4: opt.transform.symbol_engine_opt
 8.49% :     0.004683s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000038s :      1: py_interpret_to_execute
 0.02% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000017s :      1: remove_dup_value
 0.64% :     0.000354s :      1: renormalize.infer
 0.50% :     0.000277s :      1: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.07% :     0.000037s :      1: rewriter_after_opt_a
 0.14% :     0.000079s :      1: rewriter_before_opt_a
 0.18% :     0.000100s :      1: symbol_engine_optimizer
74.66% :     0.041172s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.343.271 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc41040> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.343.912 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0179817, [33]
[bootstrap]: 0.00026614
[type_inference]: 0.00422384
[event_method]: 1.702e-05
[auto_monad]: 6.438e-05
[graph_reusing]: 6.28998e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 2.877e-05
[rewriter_before_opt_a]: 7.143e-05
[expand_dump_flag]: 3.22002e-06
[jit_opt_a]: 0.00531348, [2]
    [Cycle 1]: 0.00142051, [27]
        [switch_simplify]: 5.342e-05
        [loop_unroll]: 2.813e-05
        [a_1]: 0.00049235
        [with_stream_mark]: 1.837e-05
        [recompute_prepare]: 7.78999e-06
        [updatestate_depend_eliminate]: 3.73001e-06
        [updatestate_assign_eliminate]: 3.11999e-06
        [updatestate_loads_eliminate]: 2.88e-06
        [parameter_eliminate]: 1.77999e-06
        [specialize_transform]: 5.47001e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 2.00002e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 3.98999e-06
        [cell_reuse_recompute_pass]: 1.31998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.285e-05
        [j_node_and_user_rematch]: 8.90999e-06
        [meta_fg_expand]: 2.07001e-06
        [replace_old_param]: 8.08001e-06
        [inline_without_move]: 5.37999e-06
        [renormalize]: 0.00051219
        [add_forward_monad_depend]: 6.23998e-06
        [auto_monad_grad]: 2.44999e-06
        [auto_monad_eliminator]: 1.259e-05
        [cse]: 2.943e-05
        [replace_applicator]: 1.187e-05
    [Cycle 2]: 0.00033018, [27]
        [switch_simplify]: 5.24e-06
        [loop_unroll]: 4.68999e-06
        [a_1]: 8.689e-05
        [with_stream_mark]: 9.35001e-06
        [recompute_prepare]: 4.76002e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 5.28002e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 4.75001e-06
        [meta_shard_fg_expand]: 9.60019e-07
        [get_grad_eliminate_]: 4.32e-06
        [merge_forward]: 2.66e-06
        [cell_reuse_recompute_pass]: 1.60999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.249e-05
        [j_node_and_user_rematch]: 7.39002e-06
        [meta_fg_expand]: 1.66002e-06
        [replace_old_param]: 1.046e-05
        [inline_without_move]: 4.70999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 5.40001e-06
        [cse]: 1.046e-05
        [replace_applicator]: 5.00999e-06
[py_interpret_to_execute_after_opt_a]: 1.027e-05
[rewriter_after_opt_a]: 3.608e-05
[convert_after_rewriter]: 7.3e-06
[order_py_execute_after_rewriter]: 5.48002e-06
[mutable_eliminate]: 0.00057138
[jit_opt_b]: 4.7e-05, [1]
    [Cycle 1]: 4.054e-05, [2]
        [frontend_op_eliminate]: 1.583e-05
        [inline_after_opt_a]: 1.379e-05
[cconv]: 2.463e-05
[loop_unroll]: 0.00041394
[jit_opt_after_cconv]: 0.00013313, [1]
    [Cycle 1]: 0.00012714, [11]
        [c_1]: 1.913e-05
        [parameter_eliminate]: 2.43998e-06
        [updatestate_depend_eliminate]: 5.23002e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [cse]: 1.695e-05
        [call_graph_tuple_transform]: 1.834e-05
        [tuple_list_get_item_eliminator]: 5.05001e-06
        [none_parameter_eliminate]: 1.27999e-06
        [renormalize]: 4.90021e-07
        [switch_simplify]: 5.04e-06
[remove_dup_value]: 1.274e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 4.62e-06
[add_recomputation]: 5.059e-05
[cse_after_recomputation]: 1.99e-05, [1]
    [Cycle 1]: 1.458e-05, [1]
        [cse]: 8.77e-06
[auto_monad_reorder]: 1.48e-05
[get_jit_bprop_graph]: 2.11998e-06
[rewriter_after_jit_bprop_graph]: 4.99e-06
[opt_after_jit_grad]: 0.00046438
[symbol_engine_optimizer]: 6.846e-05, [1]
    [Cycle 1]: 6.218e-05, [6]
        [build]: 3.41999e-06
        [elim_shapecalc]: 7e-06
        [elim_not_effective]: 1.204e-05
        [opt_reshape]: 5.17999e-06
        [fold_const_symbol]: 7.96001e-06
        [renormalize]: 4.59986e-07
[validate]: 3.102e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00587623
[execute]: 7.53e-06
Sums
    bootstrap                            :     0.000266s :  1.92%
    type_inference                       :     0.004224s : 30.54%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000064s :  0.47%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.21%
    rewriter_before_opt_a                :     0.000071s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.42%
    jit_opt_a.loop_unroll                :     0.000033s :  0.24%
    jit_opt_a.a_1                        :     0.000579s :  4.19%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.13%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000512s :  3.70%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000040s :  0.29%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.26%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000571s :  4.13%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.18%
    loop_unroll                          :     0.000414s :  2.99%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000051s :  0.37%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000464s :  3.36%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005876s : 42.48%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000185    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 3.12% :     0.000006s :      3: substitution.graph_param_transform
81.07% :     0.000150s :      4: substitution.inline
 1.83% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.25% :     0.000004s :      4: substitution.remove_not_recompute_node
 3.78% :     0.000007s :      2: substitution.replace_old_param
 6.27% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004158     2
81.23% :     0.003377s :      1: type_inference.infer
18.77% :     0.000780s :      1: type_inference.specialize
------[replace.]   0.000060     6
68.28% :     0.000041s :      4: replace.inline
31.72% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000158     6
93.28% :     0.000147s :      4: match.inline
 6.72% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.29% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 2.00% :     0.000002s :     10: predicate.arithmetic_simplify
 1.31% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.47% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.52% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000002s :     15: predicate.partial_eliminate
 1.34% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.77% :     0.000002s :     10: predicate.reshape_eliminate
 1.40% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.01% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.21% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.73% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.02% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.93% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000495    12
49.10% :     0.000243s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.32% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.59% :     0.000226s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019379    76
 0.28% :     0.000053s :      1: add_recomputation
 0.35% :     0.000068s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.44% :     0.000278s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.43% :     0.005316s :      1: jit_opt_a
 0.70% :     0.000136s :      1: jit_opt_after_cconv
 0.26% :     0.000049s :      1: jit_opt_b
 2.17% :     0.000421s :      1: loop_unroll
 2.99% :     0.000580s :      1: mutable_eliminate
 4.04% :     0.000783s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.44% :     0.000472s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000031s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.36% :     0.000264s :      1: renormalize.infer
 1.24% :     0.000241s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000039s :      1: rewriter_after_opt_a
 0.38% :     0.000074s :      1: rewriter_before_opt_a
 0.37% :     0.000071s :      1: symbol_engine_optimizer
30.38% :     0.005886s :      1: task_emit
21.86% :     0.004237s :      1: type_inference
 0.26% :     0.000050s :      1: validate


TotalTime = 0.0345266, [33]
[bootstrap]: 0.00025597
[type_inference]: 0.0132546
[event_method]: 0.00010097
[auto_monad]: 0.00010403
[graph_reusing]: 7.28e-06
[pre_auto_parallel]: 3.14001e-06
[py_interpret_to_execute]: 3.136e-05
[rewriter_before_opt_a]: 0.00010312
[expand_dump_flag]: 3.35e-06
[jit_opt_a]: 0.0115955, [3]
    [Cycle 1]: 0.00598793, [27]
        [switch_simplify]: 0.00012503
        [loop_unroll]: 4.522e-05
        [a_1]: 0.00088422
        [with_stream_mark]: 2.768e-05
        [recompute_prepare]: 1.74e-05
        [updatestate_depend_eliminate]: 7.26999e-06
        [updatestate_assign_eliminate]: 5.74e-06
        [updatestate_loads_eliminate]: 5.46e-06
        [parameter_eliminate]: 2.39999e-06
        [specialize_transform]: 1.275e-05
        [updatestate_useless_node_eliminater]: 1.167e-05
        [accelerated_algorithm]: 1.249e-05
        [meta_shard_fg_expand]: 3.43e-06
        [get_grad_eliminate_]: 1.225e-05
        [merge_forward]: 7.98001e-06
        [cell_reuse_recompute_pass]: 8.99978e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.65e-05
        [j_node_and_user_rematch]: 2.28e-05
        [meta_fg_expand]: 0.00155517
        [replace_old_param]: 5.427e-05
        [inline_without_move]: 5.123e-05
        [renormalize]: 0.00262399
        [add_forward_monad_depend]: 9.32999e-06
        [auto_monad_grad]: 5.90002e-06
        [auto_monad_eliminator]: 4.681e-05
        [cse]: 0.00014168
        [replace_applicator]: 6.446e-05
    [Cycle 2]: 0.00194471, [27]
        [switch_simplify]: 3.544e-05
        [loop_unroll]: 3.361e-05
        [a_1]: 0.00099418
        [with_stream_mark]: 1.364e-05
        [recompute_prepare]: 8.38999e-06
        [updatestate_depend_eliminate]: 3.72002e-06
        [updatestate_assign_eliminate]: 3.14999e-06
        [updatestate_loads_eliminate]: 2.74999e-06
        [parameter_eliminate]: 1.42e-06
        [specialize_transform]: 6.53003e-06
        [updatestate_useless_node_eliminater]: 6.06e-06
        [accelerated_algorithm]: 5.91e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 5.69999e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.378e-05
        [j_node_and_user_rematch]: 9.62001e-06
        [meta_fg_expand]: 9.263e-05
        [replace_old_param]: 1.009e-05
        [inline_without_move]: 6.21e-06
        [renormalize]: 0.00049393
        [add_forward_monad_depend]: 4.50001e-06
        [auto_monad_grad]: 1.33002e-06
        [auto_monad_eliminator]: 1.143e-05
        [cse]: 1.969e-05
        [replace_applicator]: 1.127e-05
    [Cycle 3]: 0.00036717, [27]
        [switch_simplify]: 6.31e-06
        [loop_unroll]: 6.66e-06
        [a_1]: 0.00011466
        [with_stream_mark]: 9.02999e-06
        [recompute_prepare]: 5.81e-06
        [updatestate_depend_eliminate]: 3.66001e-06
        [updatestate_assign_eliminate]: 2.61999e-06
        [updatestate_loads_eliminate]: 2.56998e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 5.84999e-06
        [updatestate_useless_node_eliminater]: 5.66e-06
        [accelerated_algorithm]: 5.61e-06
        [meta_shard_fg_expand]: 1.52001e-06
        [get_grad_eliminate_]: 5.69e-06
        [merge_forward]: 2.86999e-06
        [cell_reuse_recompute_pass]: 1.41002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.468e-05
        [j_node_and_user_rematch]: 9.56998e-06
        [meta_fg_expand]: 1.77001e-06
        [replace_old_param]: 8.17998e-06
        [inline_without_move]: 5.47001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.42e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 7.08e-06
        [cse]: 1.415e-05
        [replace_applicator]: 5.64e-06
[py_interpret_to_execute_after_opt_a]: 1.236e-05
[rewriter_after_opt_a]: 3.969e-05
[convert_after_rewriter]: 7.76001e-06
[order_py_execute_after_rewriter]: 5.07e-06
[mutable_eliminate]: 0.00058647
[jit_opt_b]: 5.36e-05, [1]
    [Cycle 1]: 4.684e-05, [2]
        [frontend_op_eliminate]: 1.849e-05
        [inline_after_opt_a]: 1.693e-05
[cconv]: 2.246e-05
[loop_unroll]: 0.00041633
[jit_opt_after_cconv]: 0.00015228, [1]
    [Cycle 1]: 0.00014542, [11]
        [c_1]: 2.569e-05
        [parameter_eliminate]: 2.71999e-06
        [updatestate_depend_eliminate]: 5.72001e-06
        [updatestate_assign_eliminate]: 3.6e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [cse]: 2.227e-05
        [call_graph_tuple_transform]: 1.977e-05
        [tuple_list_get_item_eliminator]: 6.44999e-06
        [none_parameter_eliminate]: 1.42999e-06
        [renormalize]: 3.80009e-07
        [switch_simplify]: 5.96998e-06
[remove_dup_value]: 2.617e-05
[partial_unused_args_eliminate]: 2.19999e-06
[environ_conv]: 5.91e-06
[add_recomputation]: 5.094e-05
[cse_after_recomputation]: 2.611e-05, [1]
    [Cycle 1]: 2.033e-05, [1]
        [cse]: 1.427e-05
[auto_monad_reorder]: 1.843e-05
[get_jit_bprop_graph]: 1.83002e-06
[rewriter_after_jit_bprop_graph]: 5.29e-06
[opt_after_jit_grad]: 0.00045882
[symbol_engine_optimizer]: 7.914e-05, [1]
    [Cycle 1]: 7.275e-05, [6]
        [build]: 3.4e-06
        [elim_shapecalc]: 9.02e-06
        [elim_not_effective]: 1.543e-05
        [opt_reshape]: 6.42001e-06
        [fold_const_symbol]: 1.1e-05
        [renormalize]: 4.7998e-07
[validate]: 3.839e-05
[backend_pass]: 9.79984e-07
[task_emit]: 0.00682596
[execute]: 9.99001e-06
Sums
    bootstrap                            :     0.000256s :  0.84%
    type_inference                       :     0.013255s : 43.56%
    event_method                         :     0.000101s :  0.33%
    auto_monad                           :     0.000104s :  0.34%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000031s :  0.10%
    rewriter_before_opt_a                :     0.000103s :  0.34%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000167s :  0.55%
    jit_opt_a.loop_unroll                :     0.000085s :  0.28%
    jit_opt_a.a_1                        :     0.001993s :  6.55%
    jit_opt_a.with_stream_mark           :     0.000050s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000025s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.08%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000055s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.001650s :  5.42%
    jit_opt_a.replace_old_param          :     0.000073s :  0.24%
    jit_opt_a.inline_without_move        :     0.000063s :  0.21%
    jit_opt_a.renormalize                :     0.003118s : 10.25%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000065s :  0.21%
    jit_opt_a.cse                        :     0.000176s :  0.58%
    jit_opt_a.replace_applicator         :     0.000081s :  0.27%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000040s :  0.13%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000586s :  1.93%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.06%
    cconv                                :     0.000022s :  0.07%
    loop_unroll                          :     0.000416s :  1.37%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000026s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000051s :  0.17%
    cse_after_recomputation.cse          :     0.000014s :  0.05%
    auto_monad_reorder                   :     0.000018s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000459s :  1.51%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006826s : 22.43%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.000489   113
 0.56% :     0.000003s :      3: substitution.elim_not_effective
 0.36% :     0.000002s :      3: substitution.fold_const_symbol
 1.12% :     0.000005s :      4: substitution.graph_param_transform
63.90% :     0.000312s :     14: substitution.inline
 3.46% :     0.000017s :      2: substitution.inline_without_move
 1.84% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.02% :     0.000010s :      7: substitution.minmaximum_grad
 1.51% :     0.000007s :      5: substitution.partial_eliminate
 2.25% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.17% :     0.000025s :     10: substitution.replace_applicator
 1.44% :     0.000007s :      7: substitution.replace_old_param
 0.50% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.75% :     0.000009s :      1: substitution.switch_simplify
 4.04% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.74% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 7.34% :     0.000036s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.013177     2
87.03% :     0.011468s :      1: type_inference.infer
12.97% :     0.001709s :      1: type_inference.specialize
------[replace.]   0.000180    22
53.44% :     0.000096s :     14: replace.inline
18.54% :     0.000033s :      1: replace.switch_simplify
28.02% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000331    22
92.17% :     0.000305s :     14: match.inline
 2.41% :     0.000008s :      1: match.switch_simplify
 5.42% :     0.000018s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000367  2634
 1.63% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.50% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.51% :     0.000006s :     44: predicate.addn_check_dump
 1.65% :     0.000006s :     44: predicate.addn_zero_filter
 1.99% :     0.000007s :     44: predicate.arithmetic_simplify
 1.54% :     0.000006s :     44: predicate.cast_eliminate
 0.18% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.42% :     0.000005s :     44: predicate.compare_switch_simplify
 1.54% :     0.000006s :     44: predicate.depend_value_elim
 1.49% :     0.000005s :     44: predicate.dict_get_item_const_eliminator
 1.50% :     0.000005s :     44: predicate.dict_get_item_eliminator
 1.50% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.33% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      4: predicate.elim_not_effective
 0.26% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.59% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.46% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.50% :     0.000006s :     44: predicate.environ_get_depend_swap
 1.48% :     0.000005s :     44: predicate.environ_get_eliminate
 1.45% :     0.000005s :     44: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.79% :     0.000003s :     19: predicate.get_grad_eliminate
 0.11% :     0.000000s :      4: predicate.graph_param_transform
 4.39% :     0.000016s :     73: predicate.inline
 1.78% :     0.000007s :     40: predicate.inline_without_move
 0.36% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.86% :     0.000003s :     19: predicate.less_batch_normalization
 1.78% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.93% :     0.000007s :     55: predicate.load_eliminater
 0.49% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.47% :     0.000013s :     91: predicate.loop_unroll_before_grad
 1.78% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.47% :     0.000005s :     44: predicate.merge_addn
 1.53% :     0.000006s :     44: predicate.minmaximum_grad
 0.66% :     0.000002s :      4: predicate.mutable_eliminate
 0.19% :     0.000001s :      4: predicate.opt_reshape
 2.22% :     0.000008s :     55: predicate.partial_eliminate
 1.53% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.78% :     0.000007s :     44: predicate.reduce_eliminate
 1.78% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.51% :     0.000009s :     97: predicate.replace_applicator
 0.90% :     0.000003s :     40: predicate.replace_old_param
 0.13% :     0.000000s :      4: predicate.reset_defer_inline
 1.80% :     0.000007s :     44: predicate.reshape_eliminate
 1.55% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.40% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.57% :     0.000006s :     44: predicate.same_eliminate
 0.47% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.41% :     0.000002s :      8: predicate.special_op_eliminate
 0.87% :     0.000003s :     19: predicate.specialize_transform
 1.62% :     0.000006s :     44: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000005s :     44: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.77% :     0.000010s :     65: predicate.switch_defer_inline
 2.53% :     0.000009s :     65: predicate.switch_layer_defer_inline
 6.80% :     0.000025s :    162: predicate.switch_simplify
 1.48% :     0.000005s :     44: predicate.tile_eliminate
 1.50% :     0.000006s :     44: predicate.transpose_eliminate
 1.77% :     0.000006s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.65% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 3.01% :     0.000011s :     59: predicate.tuple_list_get_item_eliminator
 1.86% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.81% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.89% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000010s :     74: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000006s :     44: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.25% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001853    32
58.72% :     0.001088s :     14: func_graph_cloner_run.FuncGraphClonerGraph
41.28% :     0.000765s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.040403    91
 0.13% :     0.000054s :      1: add_recomputation
 0.27% :     0.000108s :      1: auto_monad
 0.05% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.69% :     0.000279s :      1: bootstrap
 0.06% :     0.000025s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000028s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.26% :     0.000107s :      1: event_method
 0.04% :     0.000015s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
28.71% :     0.011599s :      1: jit_opt_a
 0.38% :     0.000155s :      1: jit_opt_after_cconv
 0.14% :     0.000056s :      1: jit_opt_b
 1.05% :     0.000424s :      1: loop_unroll
 1.47% :     0.000595s :      1: mutable_eliminate
 6.55% :     0.002644s :     39: opt.transform.jit_opt_a
 0.14% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000030s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.15% :     0.000466s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000035s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000030s :      1: remove_dup_value
 4.25% :     0.001719s :      2: renormalize.infer
 3.43% :     0.001386s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000042s :      1: rewriter_after_opt_a
 0.26% :     0.000106s :      1: rewriter_before_opt_a
 0.20% :     0.000082s :      1: symbol_engine_optimizer
16.93% :     0.006841s :      1: task_emit
32.84% :     0.013269s :      1: type_inference
 0.15% :     0.000061s :      1: validate

.
TotalTime = 0.0121904, [30]
[bootstrap]: 0.00029558
[type_inference]: 0.00421954
[event_method]: 1.701e-05
[auto_monad]: 6.374e-05
[graph_reusing]: 5.71998e-06
[pre_auto_parallel]: 2.68e-06
[py_interpret_to_execute]: 2.954e-05
[rewriter_before_opt_a]: 7.216e-05
[expand_dump_flag]: 3.44001e-06
[jit_opt_a]: 0.00547109, [2]
    [Cycle 1]: 0.00158566, [27]
        [switch_simplify]: 6.883e-05
        [loop_unroll]: 3.03e-05
        [a_1]: 0.00055494
        [with_stream_mark]: 2.032e-05
        [recompute_prepare]: 7.53e-06
        [updatestate_depend_eliminate]: 3.8e-06
        [updatestate_assign_eliminate]: 2.94001e-06
        [updatestate_loads_eliminate]: 2.81e-06
        [parameter_eliminate]: 1.72001e-06
        [specialize_transform]: 6.01998e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.74e-06
        [meta_shard_fg_expand]: 2.01998e-06
        [get_grad_eliminate_]: 5.61e-06
        [merge_forward]: 4.05e-06
        [cell_reuse_recompute_pass]: 1.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.456e-05
        [j_node_and_user_rematch]: 9.12999e-06
        [meta_fg_expand]: 2.84999e-06
        [replace_old_param]: 9.39998e-06
        [inline_without_move]: 5.32999e-06
        [renormalize]: 0.00058415
        [add_forward_monad_depend]: 5.61e-06
        [auto_monad_grad]: 2.24001e-06
        [auto_monad_eliminator]: 1.338e-05
        [cse]: 3.11e-05
        [replace_applicator]: 1.104e-05
    [Cycle 2]: 0.00032753, [27]
        [switch_simplify]: 5.90002e-06
        [loop_unroll]: 4.89e-06
        [a_1]: 9.24e-05
        [with_stream_mark]: 8.63001e-06
        [recompute_prepare]: 5.79e-06
        [updatestate_depend_eliminate]: 2.45002e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 9.49978e-07
        [specialize_transform]: 5.10999e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.77e-06
        [merge_forward]: 2.41e-06
        [cell_reuse_recompute_pass]: 1.32999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.27e-05
        [j_node_and_user_rematch]: 7.93001e-06
        [meta_fg_expand]: 1.51002e-06
        [replace_old_param]: 7.24001e-06
        [inline_without_move]: 4.90999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.54e-06
        [auto_monad_grad]: 6.79982e-07
        [auto_monad_eliminator]: 5.16998e-06
        [cse]: 1.054e-05
        [replace_applicator]: 5.19998e-06
[py_interpret_to_execute_after_opt_a]: 9.85002e-06
[rewriter_after_opt_a]: 3.34e-05
[convert_after_rewriter]: 8.50001e-06
[order_py_execute_after_rewriter]: 5.25999e-06
[mutable_eliminate]: 0.00048152
[jit_opt_b]: 4.91e-05, [1]
    [Cycle 1]: 4.237e-05, [2]
        [frontend_op_eliminate]: 1.596e-05
        [inline_after_opt_a]: 1.477e-05
[cconv]: 2.436e-05
[loop_unroll]: 0.00040632
[jit_opt_after_cconv]: 0.00014782, [1]
    [Cycle 1]: 0.00014145, [11]
        [c_1]: 2.05e-05
        [parameter_eliminate]: 2.29001e-06
        [updatestate_depend_eliminate]: 5.54e-06
        [updatestate_assign_eliminate]: 2.28002e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.623e-05
        [call_graph_tuple_transform]: 1.783e-05
        [tuple_list_get_item_eliminator]: 1.61e-05
        [none_parameter_eliminate]: 1.74e-06
        [renormalize]: 3.10014e-07
        [switch_simplify]: 5.54e-06
[remove_dup_value]: 1.519e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 5.09e-06
[add_recomputation]: 4.969e-05
[cse_after_recomputation]: 2.086e-05, [1]
    [Cycle 1]: 1.533e-05, [1]
        [cse]: 9.88002e-06
[auto_monad_reorder]: 1.453e-05
[get_jit_bprop_graph]: 1.96e-06
[rewriter_after_jit_bprop_graph]: 5.31998e-06
[opt_after_jit_grad]: 0.0004561
[symbol_engine_optimizer]: 7.437e-05, [1]
    [Cycle 1]: 6.784e-05, [6]
        [build]: 3.85e-06
        [elim_shapecalc]: 7.26999e-06
        [elim_not_effective]: 1.333e-05
        [opt_reshape]: 5.84999e-06
        [fold_const_symbol]: 9.00999e-06
        [renormalize]: 3.80009e-07
[validate]: 3.422e-05
Sums
    bootstrap                            :     0.000296s :  3.67%
    type_inference                       :     0.004220s : 52.40%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000064s :  0.79%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000030s :  0.37%
    rewriter_before_opt_a                :     0.000072s :  0.90%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000075s :  0.93%
    jit_opt_a.loop_unroll                :     0.000035s :  0.44%
    jit_opt_a.a_1                        :     0.000647s :  8.04%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.36%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000584s :  7.25%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.23%
    jit_opt_a.cse                        :     0.000042s :  0.52%
    jit_opt_a.replace_applicator         :     0.000016s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000033s :  0.41%
    convert_after_rewriter               :     0.000009s :  0.11%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000482s :  5.98%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000024s :  0.30%
    loop_unroll                          :     0.000406s :  5.05%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000016s :  0.20%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000015s :  0.19%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000050s :  0.62%
    cse_after_recomputation.cse          :     0.000010s :  0.12%
    auto_monad_reorder                   :     0.000015s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000456s :  5.66%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.42%

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.387.683 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc41430> should be wrapped by jit only once.

Time group info:
------[substitution.]   0.000208    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.28% :     0.000005s :      3: substitution.graph_param_transform
84.66% :     0.000176s :      4: substitution.inline
 1.30% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.14% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000004s :      2: substitution.replace_old_param
 5.88% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004158     2
81.20% :     0.003376s :      1: type_inference.infer
18.80% :     0.000781s :      1: type_inference.specialize
------[replace.]   0.000065     6
68.07% :     0.000044s :      4: replace.inline
31.93% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000184     6
94.06% :     0.000173s :      4: match.inline
 5.94% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.30% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.20% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 1.88% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.80% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.13% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.95% :     0.000007s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.72% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.63% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.61% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.86% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000002s :     10: predicate.reshape_eliminate
 1.32% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.57% :     0.000003s :     16: predicate.switch_defer_inline
 2.32% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.15% :     0.000010s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.25% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.86% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000554    12
51.05% :     0.000283s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.97% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.98% :     0.000233s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013713    72
 0.38% :     0.000052s :      1: add_recomputation
 0.49% :     0.000067s :      1: auto_monad
 0.12% :     0.000017s :      1: auto_monad_reorder
 2.30% :     0.000315s :      1: bootstrap
 0.20% :     0.000027s :      1: cconv
 0.08% :     0.000011s :      1: convert_after_rewriter
 0.17% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
39.92% :     0.005474s :      1: jit_opt_a
 1.10% :     0.000151s :      1: jit_opt_after_cconv
 0.38% :     0.000052s :      1: jit_opt_b
 3.02% :     0.000414s :      1: loop_unroll
 3.57% :     0.000489s :      1: mutable_eliminate
 6.36% :     0.000872s :     26: opt.transform.jit_opt_a
 0.41% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000024s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.38% :     0.000464s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000032s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000018s :      1: remove_dup_value
 2.30% :     0.000315s :      1: renormalize.infer
 1.90% :     0.000261s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000036s :      1: rewriter_after_opt_a
 0.55% :     0.000075s :      1: rewriter_before_opt_a
 0.56% :     0.000077s :      1: symbol_engine_optimizer
30.85% :     0.004231s :      1: type_inference


TotalTime = 0.0173333, [33]
[bootstrap]: 0.00026043
[type_inference]: 0.00411766
[event_method]: 1.523e-05
[auto_monad]: 4.08e-05
[graph_reusing]: 3.46999e-06
[pre_auto_parallel]: 1.80001e-06
[py_interpret_to_execute]: 3.021e-05
[rewriter_before_opt_a]: 6.287e-05
[expand_dump_flag]: 2.21e-06
[jit_opt_a]: 0.00475801, [2]
    [Cycle 1]: 0.00141889, [27]
        [switch_simplify]: 4.679e-05
        [loop_unroll]: 2.912e-05
        [a_1]: 0.000472
        [with_stream_mark]: 1.261e-05
        [recompute_prepare]: 7.11999e-06
        [updatestate_depend_eliminate]: 3.09001e-06
        [updatestate_assign_eliminate]: 2.13998e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 5.57001e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 5.30001e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 2.34999e-06
        [cell_reuse_recompute_pass]: 1.21002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.19e-05
        [j_node_and_user_rematch]: 8.36002e-06
        [meta_fg_expand]: 1.86e-06
        [replace_old_param]: 7.62998e-06
        [inline_without_move]: 5.39e-06
        [renormalize]: 0.00058132
        [add_forward_monad_depend]: 4.33001e-06
        [auto_monad_grad]: 1.17e-06
        [auto_monad_eliminator]: 9.52001e-06
        [cse]: 1.56e-05
        [replace_applicator]: 1.067e-05
    [Cycle 2]: 0.00031228, [27]
        [switch_simplify]: 5.30999e-06
        [loop_unroll]: 4.74e-06
        [a_1]: 8.853e-05
        [with_stream_mark]: 8.40001e-06
        [recompute_prepare]: 5.15999e-06
        [updatestate_depend_eliminate]: 2.47001e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 4.70999e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 5.02999e-06
        [meta_shard_fg_expand]: 1.14998e-06
        [get_grad_eliminate_]: 4.34997e-06
        [merge_forward]: 2.39999e-06
        [cell_reuse_recompute_pass]: 1.52999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.224e-05
        [j_node_and_user_rematch]: 7.73999e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 6.83998e-06
        [inline_without_move]: 4.57e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 6.39993e-07
        [auto_monad_eliminator]: 5.07e-06
        [cse]: 9.80002e-06
        [replace_applicator]: 4.68999e-06
[py_interpret_to_execute_after_opt_a]: 8.05e-06
[rewriter_after_opt_a]: 2.467e-05
[convert_after_rewriter]: 5.14998e-06
[order_py_execute_after_rewriter]: 4e-06
[mutable_eliminate]: 0.00047013
[jit_opt_b]: 5.171e-05, [1]
    [Cycle 1]: 4.503e-05, [2]
        [frontend_op_eliminate]: 1.879e-05
        [inline_after_opt_a]: 1.501e-05
[cconv]: 1.543e-05
[loop_unroll]: 0.00041173
[jit_opt_after_cconv]: 0.00016556, [1]
    [Cycle 1]: 0.00013106, [11]
        [c_1]: 2.183e-05
        [parameter_eliminate]: 2.55002e-06
        [updatestate_depend_eliminate]: 5.47001e-06
        [updatestate_assign_eliminate]: 2.54001e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.64e-05
        [call_graph_tuple_transform]: 1.59e-05
        [tuple_list_get_item_eliminator]: 5.27001e-06
        [none_parameter_eliminate]: 7.79983e-07
        [renormalize]: 3.10014e-07
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 6.49001e-06
[partial_unused_args_eliminate]: 1.39e-06
[environ_conv]: 3.61001e-06
[add_recomputation]: 3.247e-05
[cse_after_recomputation]: 2.753e-05, [1]
    [Cycle 1]: 1.375e-05, [1]
        [cse]: 8.42e-06
[auto_monad_reorder]: 9.17001e-06
[get_jit_bprop_graph]: 1.50001e-06
[rewriter_after_jit_bprop_graph]: 4.45999e-06
[opt_after_jit_grad]: 0.00045631
[symbol_engine_optimizer]: 6.817e-05, [1]
    [Cycle 1]: 6.207e-05, [6]
        [build]: 2.53003e-06
        [elim_shapecalc]: 6.98998e-06
        [elim_not_effective]: 1.185e-05
        [opt_reshape]: 5.46e-06
        [fold_const_symbol]: 8.00999e-06
        [renormalize]: 4.39992e-07
[validate]: 2.661e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.00605632
[execute]: 8.42998e-06
Sums
    bootstrap                            :     0.000260s :  1.90%
    type_inference                       :     0.004118s : 30.07%
    event_method                         :     0.000015s :  0.11%
    auto_monad                           :     0.000041s :  0.30%
    graph_reusing                        :     0.000003s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000030s :  0.22%
    rewriter_before_opt_a                :     0.000063s :  0.46%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000052s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000561s :  4.09%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000014s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000581s :  4.25%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000025s :  0.19%
    jit_opt_a.replace_applicator         :     0.000015s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000025s :  0.18%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000470s :  3.43%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000015s :  0.11%
    loop_unroll                          :     0.000412s :  3.01%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000006s :  0.05%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000032s :  0.24%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000009s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000456s :  3.33%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.19%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006056s : 44.22%
    execute                              :     0.000008s :  0.06%

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.392.804 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

Time group info:
------[substitution.]   0.000155    23
 0.91% :     0.000001s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.34% :     0.000004s :      3: substitution.graph_param_transform
84.62% :     0.000131s :      4: substitution.inline
 1.65% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.97% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.69% :     0.000003s :      2: substitution.replace_old_param
 5.13% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004072     2
81.98% :     0.003338s :      1: type_inference.infer
18.02% :     0.000734s :      1: type_inference.specialize
------[replace.]   0.000061     6
65.56% :     0.000040s :      4: replace.inline
34.44% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000136     6
94.95% :     0.000129s :      4: match.inline
 5.05% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.31% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.39% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000001s :     10: predicate.addn_zero_filter
 2.13% :     0.000002s :     10: predicate.arithmetic_simplify
 1.61% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.60% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.40% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.44% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.18% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.89% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.22% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.63% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000002s :     15: predicate.partial_eliminate
 1.23% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.81% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.66% :     0.000002s :     10: predicate.reshape_eliminate
 1.29% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.47% :     0.000008s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.20% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.07% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.48% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000555    12
51.14% :     0.000284s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.91% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.95% :     0.000244s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018775    76
 0.19% :     0.000035s :      1: add_recomputation
 0.24% :     0.000044s :      1: auto_monad
 0.06% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.45% :     0.000272s :      1: bootstrap
 0.10% :     0.000018s :      1: cconv
 0.04% :     0.000007s :      1: convert_after_rewriter
 0.16% :     0.000030s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.07% :     0.000013s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
25.36% :     0.004761s :      1: jit_opt_a
 0.90% :     0.000168s :      1: jit_opt_after_cconv
 0.29% :     0.000054s :      1: jit_opt_b
 2.23% :     0.000419s :      1: loop_unroll
 2.55% :     0.000479s :      1: mutable_eliminate
 4.02% :     0.000756s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000027s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.47% :     0.000464s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000033s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000009s :      1: remove_dup_value
 1.64% :     0.000308s :      1: renormalize.infer
 1.42% :     0.000267s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000027s :      1: rewriter_after_opt_a
 0.35% :     0.000066s :      1: rewriter_before_opt_a
 0.38% :     0.000071s :      1: symbol_engine_optimizer
32.32% :     0.006069s :      1: task_emit
21.99% :     0.004130s :      1: type_inference
 0.23% :     0.000043s :      1: validate


TotalTime = 0.0195459, [33]
[bootstrap]: 0.00030231
[type_inference]: 0.00453712
[event_method]: 1.806e-05
[auto_monad]: 6.293e-05
[graph_reusing]: 5.52999e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 3.063e-05
[rewriter_before_opt_a]: 7.073e-05
[expand_dump_flag]: 3.6e-06
[jit_opt_a]: 0.00552694, [2]
    [Cycle 1]: 0.00161756, [27]
        [switch_simplify]: 5.607e-05
        [loop_unroll]: 2.879e-05
        [a_1]: 0.00052184
        [with_stream_mark]: 1.846e-05
        [recompute_prepare]: 7.53e-06
        [updatestate_depend_eliminate]: 3.58e-06
        [updatestate_assign_eliminate]: 3.11001e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [parameter_eliminate]: 1.88002e-06
        [specialize_transform]: 5.74e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 1.97999e-06
        [get_grad_eliminate_]: 4.87e-06
        [merge_forward]: 3.75e-06
        [cell_reuse_recompute_pass]: 2.12001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.489e-05
        [j_node_and_user_rematch]: 9.39e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 8.60999e-06
        [inline_without_move]: 5.01997e-06
        [renormalize]: 0.00063405
        [add_forward_monad_depend]: 6.62002e-06
        [auto_monad_grad]: 2.20002e-06
        [auto_monad_eliminator]: 1.483e-05
        [cse]: 4.401e-05
        [replace_applicator]: 1.389e-05
    [Cycle 2]: 0.00034386, [27]
        [switch_simplify]: 5.59e-06
        [loop_unroll]: 4.89e-06
        [a_1]: 9.983e-05
        [with_stream_mark]: 1.14e-05
        [recompute_prepare]: 5.62001e-06
        [updatestate_depend_eliminate]: 3.03e-06
        [updatestate_assign_eliminate]: 2.20002e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 5.20001e-06
        [updatestate_useless_node_eliminater]: 5.02999e-06
        [accelerated_algorithm]: 5.32001e-06
        [meta_shard_fg_expand]: 1.62999e-06
        [get_grad_eliminate_]: 4.52998e-06
        [merge_forward]: 2.61e-06
        [cell_reuse_recompute_pass]: 3.24001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.396e-05
        [j_node_and_user_rematch]: 7.68999e-06
        [meta_fg_expand]: 1.59998e-06
        [replace_old_param]: 7.68001e-06
        [inline_without_move]: 4.47998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.69e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 5.83997e-06
        [cse]: 1.159e-05
        [replace_applicator]: 5.37001e-06
[py_interpret_to_execute_after_opt_a]: 1.096e-05
[rewriter_after_opt_a]: 4.464e-05
[convert_after_rewriter]: 7.17002e-06
[order_py_execute_after_rewriter]: 5.07999e-06
[mutable_eliminate]: 0.00067148
[jit_opt_b]: 4.812e-05, [1]
    [Cycle 1]: 4.104e-05, [2]
        [frontend_op_eliminate]: 1.604e-05
        [inline_after_opt_a]: 1.394e-05
[cconv]: 2.54e-05
[loop_unroll]: 0.00046203
[jit_opt_after_cconv]: 0.00015418, [1]
    [Cycle 1]: 0.00014785, [11]
        [c_1]: 2.814e-05
        [parameter_eliminate]: 2.94999e-06
        [updatestate_depend_eliminate]: 6.44001e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 1.975e-05
        [call_graph_tuple_transform]: 1.865e-05
        [tuple_list_get_item_eliminator]: 5.03002e-06
        [none_parameter_eliminate]: 1.50999e-06
        [renormalize]: 8.39995e-07
        [switch_simplify]: 5.16002e-06
[remove_dup_value]: 1.41e-05
[partial_unused_args_eliminate]: 2.23998e-06
[environ_conv]: 5.94999e-06
[add_recomputation]: 5.375e-05
[cse_after_recomputation]: 1.973e-05, [1]
    [Cycle 1]: 1.419e-05, [1]
        [cse]: 8.50001e-06
[auto_monad_reorder]: 1.607e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 5.81e-06
[opt_after_jit_grad]: 0.00052511
[symbol_engine_optimizer]: 7.344e-05, [1]
    [Cycle 1]: 6.672e-05, [6]
        [build]: 4.55001e-06
        [elim_shapecalc]: 7.6e-06
        [elim_not_effective]: 1.3e-05
        [opt_reshape]: 5.64e-06
        [fold_const_symbol]: 8.11002e-06
        [renormalize]: 4.59986e-07
[validate]: 3.812e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00658671
[execute]: 7.19001e-06
Sums
    bootstrap                            :     0.000302s :  1.97%
    type_inference                       :     0.004537s : 29.57%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000063s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.20%
    rewriter_before_opt_a                :     0.000071s :  0.46%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000622s :  4.05%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000009s :  0.06%
    jit_opt_a.renormalize                :     0.000634s :  4.13%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.13%
    jit_opt_a.cse                        :     0.000056s :  0.36%
    jit_opt_a.replace_applicator         :     0.000019s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000045s :  0.29%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000671s :  4.38%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000462s :  3.01%
    jit_opt_after_cconv.c_1              :     0.000028s :  0.18%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000054s :  0.35%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000525s :  3.42%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006587s : 42.93%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000201    23
 0.95% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.75% :     0.000006s :      3: substitution.graph_param_transform
83.12% :     0.000167s :      4: substitution.inline
 1.59% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.55% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.48% :     0.000005s :      2: substitution.replace_old_param
 5.98% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004471     2
82.85% :     0.003704s :      1: type_inference.infer
17.15% :     0.000767s :      1: type_inference.specialize
------[replace.]   0.000065     6
68.08% :     0.000044s :      4: replace.inline
31.92% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000175     6
93.71% :     0.000164s :      4: match.inline
 6.29% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.70% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000001s :     10: predicate.addn_zero_filter
 2.10% :     0.000002s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.61% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.93% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.37% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.89% :     0.000002s :     15: predicate.load_eliminater
 1.77% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.78% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.56% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     15: predicate.partial_eliminate
 1.40% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.68% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.80% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.43% :     0.000009s :     53: predicate.switch_simplify
 1.31% :     0.000002s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.07% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.07% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.95% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000540    12
51.32% :     0.000277s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.43% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.25% :     0.000234s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021131    76
 0.27% :     0.000057s :      1: add_recomputation
 0.31% :     0.000066s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.50% :     0.000317s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.06% :     0.000012s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.08% :     0.000016s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.17% :     0.005530s :      1: jit_opt_a
 0.74% :     0.000157s :      1: jit_opt_after_cconv
 0.24% :     0.000051s :      1: jit_opt_b
 2.23% :     0.000470s :      1: loop_unroll
 3.22% :     0.000681s :      1: mutable_eliminate
 3.94% :     0.000833s :     26: opt.transform.jit_opt_a
 0.26% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.53% :     0.000534s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.65% :     0.000349s :      1: renormalize.infer
 1.31% :     0.000277s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000048s :      1: rewriter_after_opt_a
 0.35% :     0.000073s :      1: rewriter_before_opt_a
 0.36% :     0.000076s :      1: symbol_engine_optimizer
31.24% :     0.006602s :      1: task_emit
21.53% :     0.004549s :      1: type_inference
 0.28% :     0.000059s :      1: validate


TotalTime = 0.0388959, [33]
[bootstrap]: 0.00024642
[type_inference]: 0.015843
[event_method]: 0.00017985
[auto_monad]: 0.00013252
[graph_reusing]: 7.76001e-06
[pre_auto_parallel]: 3.53e-06
[py_interpret_to_execute]: 3.195e-05
[rewriter_before_opt_a]: 0.0001076
[expand_dump_flag]: 3.55e-06
[jit_opt_a]: 0.012774, [3]
    [Cycle 1]: 0.0065072, [27]
        [switch_simplify]: 0.00015018
        [loop_unroll]: 4.361e-05
        [a_1]: 0.00093177
        [with_stream_mark]: 3.418e-05
        [recompute_prepare]: 2.015e-05
        [updatestate_depend_eliminate]: 8.32e-06
        [updatestate_assign_eliminate]: 6.27001e-06
        [updatestate_loads_eliminate]: 6.21e-06
        [parameter_eliminate]: 3.31001e-06
        [specialize_transform]: 1.388e-05
        [updatestate_useless_node_eliminater]: 1.186e-05
        [accelerated_algorithm]: 1.331e-05
        [meta_shard_fg_expand]: 4.43001e-06
        [get_grad_eliminate_]: 1.245e-05
        [merge_forward]: 8.06001e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.757e-05
        [j_node_and_user_rematch]: 2.311e-05
        [meta_fg_expand]: 0.00167772
        [replace_old_param]: 5.368e-05
        [inline_without_move]: 5.26e-05
        [renormalize]: 0.00287688
        [add_forward_monad_depend]: 9.67001e-06
        [auto_monad_grad]: 5.62001e-06
        [auto_monad_eliminator]: 4.794e-05
        [cse]: 0.00016787
        [replace_applicator]: 6.678e-05
    [Cycle 2]: 0.00210891, [27]
        [switch_simplify]: 3.521e-05
        [loop_unroll]: 3.338e-05
        [a_1]: 0.00101093
        [with_stream_mark]: 1.345e-05
        [recompute_prepare]: 7.65e-06
        [updatestate_depend_eliminate]: 3.99002e-06
        [updatestate_assign_eliminate]: 3.14001e-06
        [updatestate_loads_eliminate]: 2.93e-06
        [parameter_eliminate]: 1.49e-06
        [specialize_transform]: 6.39999e-06
        [updatestate_useless_node_eliminater]: 6.11998e-06
        [accelerated_algorithm]: 6.01998e-06
        [meta_shard_fg_expand]: 1.97999e-06
        [get_grad_eliminate_]: 5.99e-06
        [merge_forward]: 3.26999e-06
        [cell_reuse_recompute_pass]: 7.10017e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.424e-05
        [j_node_and_user_rematch]: 1.016e-05
        [meta_fg_expand]: 0.00015326
        [replace_old_param]: 1.215e-05
        [inline_without_move]: 6.38e-06
        [renormalize]: 0.00056002
        [add_forward_monad_depend]: 4.57e-06
        [auto_monad_grad]: 1.38002e-06
        [auto_monad_eliminator]: 1.303e-05
        [cse]: 2.459e-05
        [replace_applicator]: 1.321e-05
    [Cycle 3]: 0.00036996, [27]
        [switch_simplify]: 6.77002e-06
        [loop_unroll]: 6.31e-06
        [a_1]: 0.00012354
        [with_stream_mark]: 9.15001e-06
        [recompute_prepare]: 5.73997e-06
        [updatestate_depend_eliminate]: 3.45998e-06
        [updatestate_assign_eliminate]: 2.76e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 6.04001e-06
        [updatestate_useless_node_eliminater]: 5.67999e-06
        [accelerated_algorithm]: 5.74999e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 5.61e-06
        [merge_forward]: 3.21999e-06
        [cell_reuse_recompute_pass]: 1.53002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.451e-05
        [j_node_and_user_rematch]: 9.61998e-06
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 7.94002e-06
        [inline_without_move]: 5.56e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.05001e-06
        [auto_monad_grad]: 7.29982e-07
        [auto_monad_eliminator]: 6.21e-06
        [cse]: 1.365e-05
        [replace_applicator]: 5.74e-06
[py_interpret_to_execute_after_opt_a]: 1.295e-05
[rewriter_after_opt_a]: 4.027e-05
[convert_after_rewriter]: 8.42e-06
[order_py_execute_after_rewriter]: 5.71998e-06
[mutable_eliminate]: 0.00068378
[jit_opt_b]: 5.653e-05, [1]
    [Cycle 1]: 4.898e-05, [2]
        [frontend_op_eliminate]: 1.913e-05
        [inline_after_opt_a]: 1.79e-05
[cconv]: 2.43e-05
[loop_unroll]: 0.00043977
[jit_opt_after_cconv]: 0.00016174, [1]
    [Cycle 1]: 0.00015418, [11]
        [c_1]: 2.677e-05
        [parameter_eliminate]: 2.63e-06
        [updatestate_depend_eliminate]: 6.19001e-06
        [updatestate_assign_eliminate]: 3.54002e-06
        [updatestate_loads_eliminate]: 3.21001e-06
        [cse]: 2.649e-05
        [call_graph_tuple_transform]: 2.079e-05
        [tuple_list_get_item_eliminator]: 6.79999e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 7.40023e-07
        [switch_simplify]: 6.34999e-06
[remove_dup_value]: 1.735e-05
[partial_unused_args_eliminate]: 2.23002e-06
[environ_conv]: 5.67001e-06
[add_recomputation]: 5.667e-05
[cse_after_recomputation]: 2.601e-05, [1]
    [Cycle 1]: 2.041e-05, [1]
        [cse]: 1.485e-05
[auto_monad_reorder]: 1.958e-05
[get_jit_bprop_graph]: 2.20002e-06
[rewriter_after_jit_bprop_graph]: 5.67999e-06
[opt_after_jit_grad]: 0.00049061
[symbol_engine_optimizer]: 8.113e-05, [1]
    [Cycle 1]: 7.473e-05, [6]
        [build]: 4.45e-06
        [elim_shapecalc]: 9.08002e-06
        [elim_not_effective]: 1.562e-05
        [opt_reshape]: 6.69001e-06
        [fold_const_symbol]: 1.081e-05
        [renormalize]: 4.30009e-07
[validate]: 3.974e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00714033
[execute]: 7.75e-06
Sums
    bootstrap                            :     0.000246s :  0.72%
    type_inference                       :     0.015843s : 46.25%
    event_method                         :     0.000180s :  0.53%
    auto_monad                           :     0.000133s :  0.39%
    graph_reusing                        :     0.000008s :  0.02%
    pre_auto_parallel                    :     0.000004s :  0.01%
    py_interpret_to_execute              :     0.000032s :  0.09%
    rewriter_before_opt_a                :     0.000108s :  0.31%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000192s :  0.56%
    jit_opt_a.loop_unroll                :     0.000083s :  0.24%
    jit_opt_a.a_1                        :     0.002066s :  6.03%
    jit_opt_a.with_stream_mark           :     0.000057s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000034s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000025s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.07%
    jit_opt_a.merge_forward              :     0.000015s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000056s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000043s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.001833s :  5.35%
    jit_opt_a.replace_old_param          :     0.000074s :  0.22%
    jit_opt_a.inline_without_move        :     0.000065s :  0.19%
    jit_opt_a.renormalize                :     0.003437s : 10.03%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000067s :  0.20%
    jit_opt_a.cse                        :     0.000206s :  0.60%
    jit_opt_a.replace_applicator         :     0.000086s :  0.25%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.04%
    rewriter_after_opt_a                 :     0.000040s :  0.12%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000684s :  2.00%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.05%
    cconv                                :     0.000024s :  0.07%
    loop_unroll                          :     0.000440s :  1.28%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000017s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000057s :  0.17%
    cse_after_recomputation.cse          :     0.000015s :  0.04%
    auto_monad_reorder                   :     0.000020s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000491s :  1.43%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000040s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007140s : 20.84%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000546   113
 0.46% :     0.000003s :      3: substitution.elim_not_effective
 0.29% :     0.000002s :      3: substitution.fold_const_symbol
 1.08% :     0.000006s :      4: substitution.graph_param_transform
64.56% :     0.000352s :     14: substitution.inline
 3.32% :     0.000018s :      2: substitution.inline_without_move
 1.82% :     0.000010s :     14: substitution.j_node_and_user_rematch
 1.88% :     0.000010s :      7: substitution.minmaximum_grad
 1.30% :     0.000007s :      5: substitution.partial_eliminate
 2.00% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.24% :     0.000029s :     10: substitution.replace_applicator
 1.50% :     0.000008s :      7: substitution.replace_old_param
 0.51% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.68% :     0.000009s :      1: substitution.switch_simplify
 4.03% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.85% :     0.000016s :      7: substitution.tuple_list_get_item_depend_reorder
 7.46% :     0.000041s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.015750     2
88.97% :     0.014012s :      1: type_inference.infer
11.03% :     0.001738s :      1: type_inference.specialize
------[replace.]   0.000202    22
51.28% :     0.000104s :     14: replace.inline
22.83% :     0.000046s :      1: replace.switch_simplify
25.89% :     0.000052s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000375    22
92.04% :     0.000345s :     14: match.inline
 2.24% :     0.000008s :      1: match.switch_simplify
 5.72% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000375  2634
 1.72% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.51% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.39% :     0.000005s :     44: predicate.addn_check_dump
 1.76% :     0.000007s :     44: predicate.addn_zero_filter
 1.98% :     0.000007s :     44: predicate.arithmetic_simplify
 1.54% :     0.000006s :     44: predicate.cast_eliminate
 0.16% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.38% :     0.000005s :     44: predicate.compare_switch_simplify
 1.61% :     0.000006s :     44: predicate.depend_value_elim
 1.58% :     0.000006s :     44: predicate.dict_get_item_const_eliminator
 1.53% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.50% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.35% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      4: predicate.elim_not_effective
 0.20% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000005s :     44: predicate.environ_add_const_eliminate
 1.42% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.42% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.54% :     0.000006s :     44: predicate.environ_get_eliminate
 1.40% :     0.000005s :     44: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 1.00% :     0.000004s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.32% :     0.000016s :     73: predicate.inline
 1.81% :     0.000007s :     40: predicate.inline_without_move
 0.34% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.89% :     0.000003s :     19: predicate.less_batch_normalization
 1.85% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.92% :     0.000007s :     55: predicate.load_eliminater
 0.47% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.29% :     0.000012s :     91: predicate.loop_unroll_before_grad
 1.72% :     0.000006s :     48: predicate.make_slice_get_slice_eliminator
 1.56% :     0.000006s :     44: predicate.merge_addn
 1.48% :     0.000006s :     44: predicate.minmaximum_grad
 0.62% :     0.000002s :      4: predicate.mutable_eliminate
 0.18% :     0.000001s :      4: predicate.opt_reshape
 2.37% :     0.000009s :     55: predicate.partial_eliminate
 1.52% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.94% :     0.000007s :     44: predicate.reduce_eliminate
 1.72% :     0.000006s :     51: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.44% :     0.000009s :     97: predicate.replace_applicator
 1.02% :     0.000004s :     40: predicate.replace_old_param
 0.11% :     0.000000s :      4: predicate.reset_defer_inline
 1.75% :     0.000007s :     44: predicate.reshape_eliminate
 1.56% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.27% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.49% :     0.000006s :     44: predicate.same_eliminate
 0.44% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.34% :     0.000001s :      8: predicate.special_op_eliminate
 0.82% :     0.000003s :     19: predicate.specialize_transform
 1.67% :     0.000006s :     44: predicate.split_environ_get_set_with_tuple_value
 1.52% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.21% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.74% :     0.000010s :     65: predicate.switch_defer_inline
 2.54% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.55% :     0.000025s :    162: predicate.switch_simplify
 1.58% :     0.000006s :     44: predicate.tile_eliminate
 1.51% :     0.000006s :     44: predicate.transpose_eliminate
 1.83% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.64% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 3.18% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 1.94% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.97% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000006s :     44: predicate.value_based_eliminate
 0.15% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.23% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001998    32
61.10% :     0.001221s :     14: func_graph_cloner_run.FuncGraphClonerGraph
38.90% :     0.000777s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.045206    91
 0.13% :     0.000060s :      1: add_recomputation
 0.31% :     0.000138s :      1: auto_monad
 0.05% :     0.000023s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.58% :     0.000261s :      1: bootstrap
 0.06% :     0.000027s :      1: cconv
 0.02% :     0.000011s :      1: convert_after_rewriter
 0.08% :     0.000038s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.42% :     0.000189s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
28.26% :     0.012777s :      1: jit_opt_a
 0.37% :     0.000165s :      1: jit_opt_after_cconv
 0.13% :     0.000059s :      1: jit_opt_b
 0.99% :     0.000449s :      1: loop_unroll
 1.53% :     0.000692s :      1: mutable_eliminate
 6.09% :     0.002751s :     39: opt.transform.jit_opt_a
 0.13% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000031s :      4: opt.transform.jit_opt_b
 0.03% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.10% :     0.000499s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000035s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000020s :      1: remove_dup_value
 4.46% :     0.002016s :      2: renormalize.infer
 3.11% :     0.001406s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000043s :      1: rewriter_after_opt_a
 0.24% :     0.000110s :      1: rewriter_before_opt_a
 0.19% :     0.000084s :      1: symbol_engine_optimizer
15.83% :     0.007155s :      1: task_emit
35.10% :     0.015867s :      1: type_inference
 0.14% :     0.000064s :      1: validate


TotalTime = 0.0301236, [30]
[bootstrap]: 0.0002993
[type_inference]: 0.00444821
[event_method]: 1.97e-05
[auto_monad]: 7.185e-05
[graph_reusing]: 6.49999e-06
[pre_auto_parallel]: 3.39001e-06
[py_interpret_to_execute]: 3.553e-05
[rewriter_before_opt_a]: 7.673e-05
[expand_dump_flag]: 3.63999e-06
[jit_opt_a]: 0.0225943, [2]
    [Cycle 1]: 0.0184165, [27]
        [switch_simplify]: 5.917e-05
        [loop_unroll]: 3.014e-05
        [a_1]: 0.0169108
        [with_stream_mark]: 3.404e-05
        [recompute_prepare]: 1.532e-05
        [updatestate_depend_eliminate]: 4.43001e-06
        [updatestate_assign_eliminate]: 3.24001e-06
        [updatestate_loads_eliminate]: 2.89999e-06
        [parameter_eliminate]: 2.89999e-06
        [specialize_transform]: 7.34002e-06
        [updatestate_useless_node_eliminater]: 5.81e-06
        [accelerated_algorithm]: 6.52001e-06
        [meta_shard_fg_expand]: 2.74999e-06
        [get_grad_eliminate_]: 5.51e-06
        [merge_forward]: 5.25999e-06
        [cell_reuse_recompute_pass]: 4.13001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.769e-05
        [j_node_and_user_rematch]: 1.33e-05
        [meta_fg_expand]: 2.86e-06
        [replace_old_param]: 1.165e-05
        [inline_without_move]: 6.13998e-06
        [renormalize]: 0.00092113
        [add_forward_monad_depend]: 8.93002e-06
        [auto_monad_grad]: 2.49001e-06
        [auto_monad_eliminator]: 1.883e-05
        [cse]: 3.453e-05
        [replace_applicator]: 1.643e-05
    [Cycle 2]: 0.00034122, [27]
        [switch_simplify]: 5.91e-06
        [loop_unroll]: 5.52999e-06
        [a_1]: 9.582e-05
        [with_stream_mark]: 1.284e-05
        [recompute_prepare]: 5.20001e-06
        [updatestate_depend_eliminate]: 3.04001e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.61998e-06
        [specialize_transform]: 4.95999e-06
        [updatestate_useless_node_eliminater]: 5.34998e-06
        [accelerated_algorithm]: 5.56e-06
        [meta_shard_fg_expand]: 1.21997e-06
        [get_grad_eliminate_]: 4.75001e-06
        [merge_forward]: 3.5e-06
        [cell_reuse_recompute_pass]: 1.68002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.479e-05
        [j_node_and_user_rematch]: 8.90999e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 8.70001e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 5.25999e-06
        [cse]: 1.093e-05
        [replace_applicator]: 5.14e-06
[py_interpret_to_execute_after_opt_a]: 1.458e-05
[rewriter_after_opt_a]: 3.801e-05
[convert_after_rewriter]: 7.03e-06
[order_py_execute_after_rewriter]: 7.45998e-06
[mutable_eliminate]: 0.00084279
[jit_opt_b]: 5.308e-05, [1]
    [Cycle 1]: 4.494e-05, [2]
        [frontend_op_eliminate]: 1.718e-05
        [inline_after_opt_a]: 1.558e-05
[cconv]: 2.819e-05
[loop_unroll]: 0.0004314
[jit_opt_after_cconv]: 0.00015054, [1]
    [Cycle 1]: 0.00014401, [11]
        [c_1]: 2.295e-05
        [parameter_eliminate]: 3.3e-06
        [updatestate_depend_eliminate]: 6.56999e-06
        [updatestate_assign_eliminate]: 2.30002e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [cse]: 1.987e-05
        [call_graph_tuple_transform]: 2.095e-05
        [tuple_list_get_item_eliminator]: 5.76998e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 9.00007e-07
        [switch_simplify]: 5.60001e-06
[remove_dup_value]: 1.474e-05
[partial_unused_args_eliminate]: 2.61999e-06
[environ_conv]: 4.92999e-06
[add_recomputation]: 5.059e-05
[cse_after_recomputation]: 2.522e-05, [1]
    [Cycle 1]: 1.84e-05, [1]
        [cse]: 1.122e-05
[auto_monad_reorder]: 1.592e-05
[get_jit_bprop_graph]: 2.21e-06
[rewriter_after_jit_bprop_graph]: 6.97002e-06
[opt_after_jit_grad]: 0.00051225
[symbol_engine_optimizer]: 7.869e-05, [1]
    [Cycle 1]: 7.14e-05, [6]
        [build]: 4.4e-06
        [elim_shapecalc]: 8.33001e-06
        [elim_not_effective]: 1.473e-05
        [opt_reshape]: 6.11e-06
        [fold_const_symbol]: 8.67998e-06
        [renormalize]: 7.00005e-07
[validate]: 4.023e-05
Sums
    bootstrap                            :     0.000299s :  1.17%
    type_inference                       :     0.004448s : 17.42%
    event_method                         :     0.000020s :  0.08%
    auto_monad                           :     0.000072s :  0.28%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000036s :  0.14%
    rewriter_before_opt_a                :     0.000077s :  0.30%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000065s :  0.25%
    jit_opt_a.loop_unroll                :     0.000036s :  0.14%
    jit_opt_a.a_1                        :     0.017007s : 66.58%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000009s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000022s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.08%
    jit_opt_a.inline_without_move        :     0.000011s :  0.04%
    jit_opt_a.renormalize                :     0.000921s :  3.61%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.09%
    jit_opt_a.cse                        :     0.000045s :  0.18%
    jit_opt_a.replace_applicator         :     0.000022s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.06%
    rewriter_after_opt_a                 :     0.000038s :  0.15%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.03%
    mutable_eliminate                    :     0.000843s :  3.30%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.06%
    cconv                                :     0.000028s :  0.11%
    loop_unroll                          :     0.000431s :  1.69%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000015s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000051s :  0.20%
    cse_after_recomputation.cse          :     0.000011s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.03%
    opt_after_jit_grad                   :     0.000512s :  2.01%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.16%


Time group info:
------[substitution.]   0.016433    23
 0.01% :     0.000002s :      2: substitution.elim_not_effective
 0.01% :     0.000001s :      2: substitution.fold_const_symbol
 0.04% :     0.000006s :      3: substitution.graph_param_transform
99.72% :     0.016387s :      4: substitution.inline
 0.04% :     0.000006s :      4: substitution.j_node_and_user_rematch
 0.03% :     0.000005s :      4: substitution.remove_not_recompute_node
 0.05% :     0.000008s :      2: substitution.replace_old_param
 0.11% :     0.000019s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004370     2
80.86% :     0.003534s :      1: type_inference.infer
19.14% :     0.000836s :      1: type_inference.specialize
------[replace.]   0.000140     6
67.88% :     0.000095s :      4: replace.inline
32.12% :     0.000045s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.016399     6
99.89% :     0.016382s :      4: match.inline
 0.11% :     0.000017s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000143   691
 1.21% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.28% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.85% :     0.000001s :     10: predicate.addn_check_dump
 1.51% :     0.000002s :     10: predicate.addn_zero_filter
 3.13% :     0.000004s :     10: predicate.arithmetic_simplify
 1.08% :     0.000002s :     10: predicate.cast_eliminate
 0.60% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.90% :     0.000001s :     10: predicate.compare_switch_simplify
 1.67% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.19% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.63% :     0.000001s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000002s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 6.82% :     0.000010s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.55% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.56% :     0.000002s :      6: predicate.less_batch_normalization
 1.88% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.88% :     0.000003s :     15: predicate.load_eliminater
 1.04% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.49% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.07% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 0.92% :     0.000001s :     10: predicate.minmaximum_grad
 1.40% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.07% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.02% :     0.000003s :     10: predicate.reduce_eliminate
 1.78% :     0.000003s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.47% :     0.000002s :     18: predicate.replace_applicator
 0.41% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000002s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.07% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.01% :     0.000004s :     16: predicate.switch_defer_inline
 2.10% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.55% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000002s :     10: predicate.tile_eliminate
 1.12% :     0.000002s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.12% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 2.09% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.79% :     0.000003s :     15: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000003s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.016785    12
 1.74% :     0.000292s :      5: func_graph_cloner_run.FuncGraphClonerGraph
96.61% :     0.016216s :      1: func_graph_cloner_run.FuncGraphClonerNode
 1.65% :     0.000277s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.048338    72
 0.21% :     0.000099s :      1: add_recomputation
 0.16% :     0.000075s :      1: auto_monad
 0.04% :     0.000019s :      1: auto_monad_reorder
 0.66% :     0.000318s :      1: bootstrap
 0.06% :     0.000031s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.06% :     0.000028s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.05% :     0.000024s :      1: event_method
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
46.75% :     0.022598s :      1: jit_opt_a
 0.32% :     0.000153s :      1: jit_opt_after_cconv
 0.11% :     0.000056s :      1: jit_opt_b
 0.91% :     0.000440s :      1: loop_unroll
 1.76% :     0.000852s :      1: mutable_eliminate
35.67% :     0.017243s :     26: opt.transform.jit_opt_a
 0.11% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000026s :      4: opt.transform.jit_opt_b
 0.02% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000034s :      4: opt.transform.symbol_engine_opt
 1.08% :     0.000522s :      1: opt_after_jit_grad
 0.02% :     0.000010s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.08% :     0.000039s :      1: py_interpret_to_execute
 0.04% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000017s :      1: remove_dup_value
 1.05% :     0.000507s :      1: renormalize.infer
 0.83% :     0.000403s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000042s :      1: rewriter_after_opt_a
 0.17% :     0.000080s :      1: rewriter_before_opt_a
 0.17% :     0.000081s :      1: symbol_engine_optimizer
 9.24% :     0.004466s :      1: type_inference


TotalTime = 0.0191973, [33]
[bootstrap]: 0.00031769
[type_inference]: 0.00462098
[event_method]: 1.824e-05
[auto_monad]: 6.582e-05
[graph_reusing]: 5.97999e-06
[pre_auto_parallel]: 2.54001e-06
[py_interpret_to_execute]: 3.226e-05
[rewriter_before_opt_a]: 7.25e-05
[expand_dump_flag]: 2.91e-06
[jit_opt_a]: 0.00567286, [2]
    [Cycle 1]: 0.00150921, [27]
        [switch_simplify]: 5.26e-05
        [loop_unroll]: 2.893e-05
        [a_1]: 0.00051331
        [with_stream_mark]: 1.738e-05
        [recompute_prepare]: 6.92002e-06
        [updatestate_depend_eliminate]: 3.75998e-06
        [updatestate_assign_eliminate]: 3.26001e-06
        [updatestate_loads_eliminate]: 2.66e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 5.67001e-06
        [updatestate_useless_node_eliminater]: 4.66997e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 1.82999e-06
        [get_grad_eliminate_]: 5.32999e-06
        [merge_forward]: 3.51999e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.392e-05
        [j_node_and_user_rematch]: 8.97e-06
        [meta_fg_expand]: 2.01e-06
        [replace_old_param]: 8.61002e-06
        [inline_without_move]: 5.22e-06
        [renormalize]: 0.00058219
        [add_forward_monad_depend]: 5.14e-06
        [auto_monad_grad]: 2.63998e-06
        [auto_monad_eliminator]: 1.336e-05
        [cse]: 2.909e-05
        [replace_applicator]: 1.087e-05
    [Cycle 2]: 0.00030948, [27]
        [switch_simplify]: 5.58002e-06
        [loop_unroll]: 4.68999e-06
        [a_1]: 8.736e-05
        [with_stream_mark]: 8.37e-06
        [recompute_prepare]: 4.72998e-06
        [updatestate_depend_eliminate]: 2.72001e-06
        [updatestate_assign_eliminate]: 1.93002e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 4.62e-06
        [updatestate_useless_node_eliminater]: 4.63001e-06
        [accelerated_algorithm]: 4.87998e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.23999e-06
        [merge_forward]: 2.27999e-06
        [cell_reuse_recompute_pass]: 1.62001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.192e-05
        [j_node_and_user_rematch]: 7.56999e-06
        [meta_fg_expand]: 1.52999e-06
        [replace_old_param]: 7.01999e-06
        [inline_without_move]: 4.52003e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 8.10018e-07
        [auto_monad_eliminator]: 5.35001e-06
        [cse]: 1.09e-05
        [replace_applicator]: 4.80001e-06
[py_interpret_to_execute_after_opt_a]: 1.09e-05
[rewriter_after_opt_a]: 3.509e-05
[convert_after_rewriter]: 6.66e-06
[order_py_execute_after_rewriter]: 4.62998e-06
[mutable_eliminate]: 0.00055794
[jit_opt_b]: 4.833e-05, [1]
    [Cycle 1]: 4.143e-05, [2]
        [frontend_op_eliminate]: 1.612e-05
        [inline_after_opt_a]: 1.423e-05
[cconv]: 2.577e-05
[loop_unroll]: 0.00041295
[jit_opt_after_cconv]: 0.00015338, [1]
    [Cycle 1]: 0.00014695, [11]
        [c_1]: 1.932e-05
        [parameter_eliminate]: 3.04001e-06
        [updatestate_depend_eliminate]: 6.17999e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 1.773e-05
        [call_graph_tuple_transform]: 1.856e-05
        [tuple_list_get_item_eliminator]: 5.30999e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 5.20027e-07
        [switch_simplify]: 5.42001e-06
[remove_dup_value]: 1.367e-05
[partial_unused_args_eliminate]: 2.26998e-06
[environ_conv]: 5.25999e-06
[add_recomputation]: 4.825e-05
[cse_after_recomputation]: 2.056e-05, [1]
    [Cycle 1]: 1.503e-05, [1]
        [cse]: 9.07001e-06
[auto_monad_reorder]: 1.457e-05
[get_jit_bprop_graph]: 1.76e-06
[rewriter_after_jit_bprop_graph]: 5.45001e-06
[opt_after_jit_grad]: 0.00075838
[symbol_engine_optimizer]: 8.174e-05, [1]
    [Cycle 1]: 7.528e-05, [6]
        [build]: 3.71001e-06
        [elim_shapecalc]: 8.45001e-06
        [elim_not_effective]: 1.316e-05
        [opt_reshape]: 5.53002e-06
        [fold_const_symbol]: 7.99002e-06
        [renormalize]: 7.09988e-07
[validate]: 3.523e-05
[backend_pass]: 1.30999e-06
[task_emit]: 0.00594442
[execute]: 8.95999e-06
Sums
    bootstrap                            :     0.000318s :  2.16%
    type_inference                       :     0.004621s : 31.37%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000066s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.22%
    rewriter_before_opt_a                :     0.000072s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000601s :  4.08%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000582s :  3.95%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000040s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000558s :  3.79%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000026s :  0.17%
    loop_unroll                          :     0.000413s :  2.80%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000048s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000758s :  5.15%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005944s : 40.35%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000195    23
 1.12% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.61% :     0.000005s :      3: substitution.graph_param_transform
83.78% :     0.000164s :      4: substitution.inline
 1.68% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.11% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.97% :     0.000004s :      2: substitution.replace_old_param
 6.08% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004553     2
82.13% :     0.003739s :      1: type_inference.infer
17.87% :     0.000813s :      1: type_inference.specialize
------[replace.]   0.000062     6
68.34% :     0.000042s :      4: replace.inline
31.66% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000172     6
93.73% :     0.000161s :      4: match.inline
 6.27% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.55% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.69% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 2.04% :     0.000002s :     10: predicate.arithmetic_simplify
 1.16% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000001s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.44% :     0.000001s :      3: predicate.graph_param_transform
 4.98% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.89% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.44% :     0.000003s :     15: predicate.partial_eliminate
 1.38% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.43% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.43% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.10% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.57% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.54% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.31% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.65% :     0.000009s :     53: predicate.switch_simplify
 1.50% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.31% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.88% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 2.08% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000571    12
53.67% :     0.000306s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.21% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.11% :     0.000229s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020687    76
 0.25% :     0.000051s :      1: add_recomputation
 0.33% :     0.000069s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.59% :     0.000329s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.43% :     0.005675s :      1: jit_opt_a
 0.75% :     0.000156s :      1: jit_opt_after_cconv
 0.25% :     0.000051s :      1: jit_opt_b
 2.03% :     0.000421s :      1: loop_unroll
 2.75% :     0.000568s :      1: mutable_eliminate
 3.87% :     0.000800s :     26: opt.transform.jit_opt_a
 0.22% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.71% :     0.000768s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000035s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.56% :     0.000323s :      1: renormalize.infer
 1.22% :     0.000252s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000038s :      1: rewriter_after_opt_a
 0.36% :     0.000075s :      1: rewriter_before_opt_a
 0.41% :     0.000084s :      1: symbol_engine_optimizer
28.81% :     0.005959s :      1: task_emit
22.41% :     0.004635s :      1: type_inference
 0.27% :     0.000056s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.459.596 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc41c10> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.460.206 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0202486, [33]
[bootstrap]: 0.00029207
[type_inference]: 0.00443865
[event_method]: 1.805e-05
[auto_monad]: 6.468e-05
[graph_reusing]: 6.51999e-06
[pre_auto_parallel]: 2.82002e-06
[py_interpret_to_execute]: 2.946e-05
[rewriter_before_opt_a]: 7.703e-05
[expand_dump_flag]: 3.30998e-06
[jit_opt_a]: 0.00702878, [2]
    [Cycle 1]: 0.00166546, [27]
        [switch_simplify]: 5.162e-05
        [loop_unroll]: 2.846e-05
        [a_1]: 0.00052848
        [with_stream_mark]: 1.876e-05
        [recompute_prepare]: 7.3e-06
        [updatestate_depend_eliminate]: 3.53e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.76999e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 6.10002e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.39998e-06
        [meta_shard_fg_expand]: 2.02001e-06
        [get_grad_eliminate_]: 4.99e-06
        [merge_forward]: 3.57997e-06
        [cell_reuse_recompute_pass]: 2.18002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.355e-05
        [j_node_and_user_rematch]: 9.33002e-06
        [meta_fg_expand]: 2.48e-06
        [replace_old_param]: 8.63001e-06
        [inline_without_move]: 5.05999e-06
        [renormalize]: 0.00072145
        [add_forward_monad_depend]: 5.67001e-06
        [auto_monad_grad]: 1.91998e-06
        [auto_monad_eliminator]: 1.351e-05
        [cse]: 2.923e-05
        [replace_applicator]: 1.163e-05
    [Cycle 2]: 0.0003101, [27]
        [switch_simplify]: 5.17999e-06
        [loop_unroll]: 4.97999e-06
        [a_1]: 8.908e-05
        [with_stream_mark]: 9.00999e-06
        [recompute_prepare]: 4.74e-06
        [updatestate_depend_eliminate]: 2.54001e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.80001e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 4.70001e-06
        [updatestate_useless_node_eliminater]: 4.68999e-06
        [accelerated_algorithm]: 5.05999e-06
        [meta_shard_fg_expand]: 1.30001e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.20002e-06
        [cell_reuse_recompute_pass]: 1.82999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.14e-05
        [j_node_and_user_rematch]: 8.01001e-06
        [meta_fg_expand]: 1.58002e-06
        [replace_old_param]: 7.16999e-06
        [inline_without_move]: 8.64003e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 6.60017e-07
        [auto_monad_eliminator]: 4.55999e-06
        [cse]: 9.76e-06
        [replace_applicator]: 4.64002e-06
[py_interpret_to_execute_after_opt_a]: 8.79e-06
[rewriter_after_opt_a]: 3.501e-05
[convert_after_rewriter]: 7.36001e-06
[order_py_execute_after_rewriter]: 5.09e-06
[mutable_eliminate]: 0.00055811
[jit_opt_b]: 4.749e-05, [1]
    [Cycle 1]: 4.089e-05, [2]
        [frontend_op_eliminate]: 1.562e-05
        [inline_after_opt_a]: 1.431e-05
[cconv]: 2.451e-05
[loop_unroll]: 0.00041054
[jit_opt_after_cconv]: 0.00013293, [1]
    [Cycle 1]: 0.00012669, [11]
        [c_1]: 2.014e-05
        [parameter_eliminate]: 2.19001e-06
        [updatestate_depend_eliminate]: 5.04e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [cse]: 1.67e-05
        [call_graph_tuple_transform]: 1.747e-05
        [tuple_list_get_item_eliminator]: 5.24998e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 5.17999e-06
[remove_dup_value]: 1.32e-05
[partial_unused_args_eliminate]: 1.98002e-06
[environ_conv]: 5.10999e-06
[add_recomputation]: 5.008e-05
[cse_after_recomputation]: 1.986e-05, [1]
    [Cycle 1]: 1.473e-05, [1]
        [cse]: 8.84e-06
[auto_monad_reorder]: 1.598e-05
[get_jit_bprop_graph]: 1.57001e-06
[rewriter_after_jit_bprop_graph]: 4.63999e-06
[opt_after_jit_grad]: 0.00048328
[symbol_engine_optimizer]: 6.738e-05, [1]
    [Cycle 1]: 6.128e-05, [6]
        [build]: 3.65998e-06
        [elim_shapecalc]: 6.48e-06
        [elim_not_effective]: 1.073e-05
        [opt_reshape]: 5.35001e-06
        [fold_const_symbol]: 8.12e-06
        [renormalize]: 4.00003e-07
[validate]: 3.349e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00616597
[execute]: 7.80998e-06
Sums
    bootstrap                            :     0.000292s :  2.00%
    type_inference                       :     0.004439s : 30.36%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000065s :  0.44%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.20%
    rewriter_before_opt_a                :     0.000077s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.39%
    jit_opt_a.loop_unroll                :     0.000033s :  0.23%
    jit_opt_a.a_1                        :     0.000618s :  4.22%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000014s :  0.09%
    jit_opt_a.renormalize                :     0.000722s :  4.94%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000039s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000035s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000558s :  3.82%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000411s :  2.81%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000483s :  3.31%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006166s : 42.18%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000199    23
 0.90% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.60% :     0.000005s :      3: substitution.graph_param_transform
83.77% :     0.000167s :      4: substitution.inline
 1.55% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.12% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.15% :     0.000004s :      2: substitution.replace_old_param
 6.29% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004369     2
81.66% :     0.003568s :      1: type_inference.infer
18.34% :     0.000801s :      1: type_inference.specialize
------[replace.]   0.000075     6
71.64% :     0.000054s :      4: replace.inline
28.36% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000176     6
93.49% :     0.000164s :      4: match.inline
 6.51% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.43% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.11% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 2.28% :     0.000003s :     10: predicate.arithmetic_simplify
 1.37% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.12% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.46% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000001s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.86% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.04% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.16% :     0.000001s :     10: predicate.minmaximum_grad
 1.28% :     0.000001s :      3: predicate.mutable_eliminate
 0.43% :     0.000000s :      3: predicate.opt_reshape
 2.41% :     0.000003s :     15: predicate.partial_eliminate
 1.55% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.38% :     0.000000s :      3: predicate.reset_defer_inline
 1.56% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.58% :     0.000009s :     53: predicate.switch_simplify
 1.38% :     0.000002s :     10: predicate.tile_eliminate
 1.48% :     0.000002s :     10: predicate.transpose_eliminate
 1.57% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.88% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000497    12
45.70% :     0.000227s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.28% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.03% :     0.000239s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021889    76
 0.24% :     0.000053s :      1: add_recomputation
 0.31% :     0.000068s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.41% :     0.000309s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
32.13% :     0.007032s :      1: jit_opt_a
 0.62% :     0.000136s :      1: jit_opt_after_cconv
 0.23% :     0.000050s :      1: jit_opt_b
 1.91% :     0.000419s :      1: loop_unroll
 2.59% :     0.000567s :      1: mutable_eliminate
 3.75% :     0.000821s :     26: opt.transform.jit_opt_a
 0.21% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000018s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000028s :      4: opt.transform.symbol_engine_opt
 2.24% :     0.000490s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.15% :     0.000032s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.94% :     0.000425s :      1: renormalize.infer
 1.32% :     0.000288s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000038s :      1: rewriter_after_opt_a
 0.36% :     0.000080s :      1: rewriter_before_opt_a
 0.32% :     0.000070s :      1: symbol_engine_optimizer
28.21% :     0.006176s :      1: task_emit
20.34% :     0.004453s :      1: type_inference
 0.25% :     0.000054s :      1: validate


TotalTime = 0.0231876, [33]
[bootstrap]: 0.000287
[type_inference]: 0.00467882
[event_method]: 1.724e-05
[auto_monad]: 6.639e-05
[graph_reusing]: 6.20002e-06
[pre_auto_parallel]: 2.86999e-06
[py_interpret_to_execute]: 3.962e-05
[rewriter_before_opt_a]: 7.494e-05
[expand_dump_flag]: 3.28e-06
[jit_opt_a]: 0.00838562, [2]
    [Cycle 1]: 0.00221301, [27]
        [switch_simplify]: 9.415e-05
        [loop_unroll]: 3.008e-05
        [a_1]: 0.00065169
        [with_stream_mark]: 2.689e-05
        [recompute_prepare]: 8.42e-06
        [updatestate_depend_eliminate]: 5.21002e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.91e-06
        [parameter_eliminate]: 2.29001e-06
        [specialize_transform]: 6.24001e-06
        [updatestate_useless_node_eliminater]: 5.51e-06
        [accelerated_algorithm]: 6.41e-06
        [meta_shard_fg_expand]: 2.39999e-06
        [get_grad_eliminate_]: 5.46e-06
        [merge_forward]: 3.58999e-06
        [cell_reuse_recompute_pass]: 2.02999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.531e-05
        [j_node_and_user_rematch]: 9.89999e-06
        [meta_fg_expand]: 2.43e-06
        [replace_old_param]: 1.045e-05
        [inline_without_move]: 5.42001e-06
        [renormalize]: 0.00102204
        [add_forward_monad_depend]: 7.28e-06
        [auto_monad_grad]: 2.74999e-06
        [auto_monad_eliminator]: 1.876e-05
        [cse]: 3.212e-05
        [replace_applicator]: 1.605e-05
    [Cycle 2]: 0.00035229, [27]
        [switch_simplify]: 5.42999e-06
        [loop_unroll]: 5.15001e-06
        [a_1]: 9.368e-05
        [with_stream_mark]: 1.423e-05
        [recompute_prepare]: 5.45001e-06
        [updatestate_depend_eliminate]: 3.11999e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.42e-06
        [specialize_transform]: 5.05001e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.75001e-06
        [merge_forward]: 3.28998e-06
        [cell_reuse_recompute_pass]: 2.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.542e-05
        [j_node_and_user_rematch]: 7.81001e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 8.60999e-06
        [inline_without_move]: 5.27001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.82999e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 6.44999e-06
        [cse]: 1.333e-05
        [replace_applicator]: 1.451e-05
[py_interpret_to_execute_after_opt_a]: 1.389e-05
[rewriter_after_opt_a]: 3.879e-05
[convert_after_rewriter]: 7.80998e-06
[order_py_execute_after_rewriter]: 4.66002e-06
[mutable_eliminate]: 0.00073886
[jit_opt_b]: 5.742e-05, [1]
    [Cycle 1]: 4.868e-05, [2]
        [frontend_op_eliminate]: 2.162e-05
        [inline_after_opt_a]: 1.487e-05
[cconv]: 3.198e-05
[loop_unroll]: 0.00048872
[jit_opt_after_cconv]: 0.00015547, [1]
    [Cycle 1]: 0.00014823, [11]
        [c_1]: 2.107e-05
        [parameter_eliminate]: 3.54002e-06
        [updatestate_depend_eliminate]: 7.95e-06
        [updatestate_assign_eliminate]: 2.41998e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [cse]: 2.544e-05
        [call_graph_tuple_transform]: 2.13e-05
        [tuple_list_get_item_eliminator]: 5.27001e-06
        [none_parameter_eliminate]: 1.48002e-06
        [renormalize]: 9.69972e-07
        [switch_simplify]: 5.10999e-06
[remove_dup_value]: 1.438e-05
[partial_unused_args_eliminate]: 2.29001e-06
[environ_conv]: 5.15999e-06
[add_recomputation]: 5.498e-05
[cse_after_recomputation]: 2.575e-05, [1]
    [Cycle 1]: 1.875e-05, [1]
        [cse]: 1.132e-05
[auto_monad_reorder]: 1.592e-05
[get_jit_bprop_graph]: 2.16e-06
[rewriter_after_jit_bprop_graph]: 7.8e-06
[opt_after_jit_grad]: 0.00054088
[symbol_engine_optimizer]: 8.323e-05, [1]
    [Cycle 1]: 7.492e-05, [6]
        [build]: 5.37001e-06
        [elim_shapecalc]: 9.65002e-06
        [elim_not_effective]: 1.454e-05
        [opt_reshape]: 5.54e-06
        [fold_const_symbol]: 9.02e-06
        [renormalize]: 5.39992e-07
[validate]: 4.586e-05
[backend_pass]: 1.13001e-06
[task_emit]: 0.00702184
[execute]: 1.011e-05
Sums
    bootstrap                            :     0.000287s :  1.72%
    type_inference                       :     0.004679s : 28.10%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000066s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.24%
    rewriter_before_opt_a                :     0.000075s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000100s :  0.60%
    jit_opt_a.loop_unroll                :     0.000035s :  0.21%
    jit_opt_a.a_1                        :     0.000745s :  4.48%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001022s :  6.14%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.15%
    jit_opt_a.cse                        :     0.000045s :  0.27%
    jit_opt_a.replace_applicator         :     0.000031s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000039s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000739s :  4.44%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000032s :  0.19%
    loop_unroll                          :     0.000489s :  2.94%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000055s :  0.33%
    cse_after_recomputation.cse          :     0.000011s :  0.07%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000541s :  3.25%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.28%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007022s : 42.18%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000284    23
 0.63% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000002s :      2: substitution.fold_const_symbol
 2.34% :     0.000007s :      3: substitution.graph_param_transform
86.73% :     0.000246s :      4: substitution.inline
 1.35% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.74% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.23% :     0.000006s :      2: substitution.replace_old_param
 4.41% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004527     2
81.34% :     0.003682s :      1: type_inference.infer
18.66% :     0.000845s :      1: type_inference.specialize
------[replace.]   0.000084     6
66.90% :     0.000056s :      4: replace.inline
33.10% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000255     6
95.49% :     0.000244s :      4: match.inline
 4.51% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   691
 1.09% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.57% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.16% :     0.000002s :     10: predicate.addn_zero_filter
 3.24% :     0.000004s :     10: predicate.arithmetic_simplify
 1.14% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.93% :     0.000001s :     10: predicate.compare_switch_simplify
 1.73% :     0.000002s :     10: predicate.depend_value_elim
 0.92% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.49% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     10: predicate.environ_get_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 6.22% :     0.000008s :     22: predicate.inline
 1.27% :     0.000002s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.63% :     0.000002s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.53% :     0.000002s :     15: predicate.load_eliminater
 2.01% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.64% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.32% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.93% :     0.000001s :     10: predicate.merge_addn
 1.34% :     0.000002s :     10: predicate.minmaximum_grad
 2.45% :     0.000003s :      3: predicate.mutable_eliminate
 0.36% :     0.000000s :      3: predicate.opt_reshape
 1.94% :     0.000003s :     15: predicate.partial_eliminate
 1.01% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.28% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.20% :     0.000002s :      6: predicate.special_op_eliminate
 0.87% :     0.000001s :      6: predicate.specialize_transform
 1.19% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.06% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.93% :     0.000004s :     16: predicate.switch_defer_inline
 1.77% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.58% :     0.000010s :     53: predicate.switch_simplify
 1.47% :     0.000002s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.57% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.32% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.44% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.84% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000645    12
47.77% :     0.000308s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 9.11% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.12% :     0.000278s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025337    76
 0.23% :     0.000059s :      1: add_recomputation
 0.28% :     0.000070s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.21% :     0.000307s :      1: bootstrap
 0.14% :     0.000035s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000028s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000021s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
33.11% :     0.008390s :      1: jit_opt_a
 0.63% :     0.000159s :      1: jit_opt_after_cconv
 0.24% :     0.000061s :      1: jit_opt_b
 1.98% :     0.000502s :      1: loop_unroll
 2.97% :     0.000752s :      1: mutable_eliminate
 4.01% :     0.001016s :     26: opt.transform.jit_opt_a
 0.19% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000030s :      4: opt.transform.jit_opt_b
 0.06% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.19% :     0.000556s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000042s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 1.93% :     0.000490s :      1: renormalize.infer
 2.06% :     0.000522s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000042s :      1: rewriter_after_opt_a
 0.31% :     0.000078s :      1: rewriter_before_opt_a
 0.34% :     0.000086s :      1: symbol_engine_optimizer
27.79% :     0.007042s :      1: task_emit
18.54% :     0.004697s :      1: type_inference
 0.29% :     0.000073s :      1: validate


TotalTime = 0.0381156, [33]
[bootstrap]: 0.00026605
[type_inference]: 0.0153999
[event_method]: 6.698e-05
[auto_monad]: 0.00010439
[graph_reusing]: 7.22002e-06
[pre_auto_parallel]: 2.73e-06
[py_interpret_to_execute]: 3.051e-05
[rewriter_before_opt_a]: 0.00010078
[expand_dump_flag]: 3.65e-06
[jit_opt_a]: 0.011881, [3]
    [Cycle 1]: 0.00605227, [27]
        [switch_simplify]: 0.00010826
        [loop_unroll]: 4.207e-05
        [a_1]: 0.00085161
        [with_stream_mark]: 1.948e-05
        [recompute_prepare]: 1.749e-05
        [updatestate_depend_eliminate]: 6.79999e-06
        [updatestate_assign_eliminate]: 5.66e-06
        [updatestate_loads_eliminate]: 5.22999e-06
        [parameter_eliminate]: 2.07999e-06
        [specialize_transform]: 1.284e-05
        [updatestate_useless_node_eliminater]: 1.149e-05
        [accelerated_algorithm]: 1.234e-05
        [meta_shard_fg_expand]: 2.86999e-06
        [get_grad_eliminate_]: 1.209e-05
        [merge_forward]: 6.89999e-06
        [cell_reuse_recompute_pass]: 7.39994e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.34e-05
        [j_node_and_user_rematch]: 2.184e-05
        [meta_fg_expand]: 0.00156543
        [replace_old_param]: 5.069e-05
        [inline_without_move]: 5.22e-05
        [renormalize]: 0.00275716
        [add_forward_monad_depend]: 9.84999e-06
        [auto_monad_grad]: 5.15999e-06
        [auto_monad_eliminator]: 4.682e-05
        [cse]: 0.00014229
        [replace_applicator]: 7.077e-05
    [Cycle 2]: 0.0020401, [27]
        [switch_simplify]: 3.592e-05
        [loop_unroll]: 3.369e-05
        [a_1]: 0.00100307
        [with_stream_mark]: 1.364e-05
        [recompute_prepare]: 7.68999e-06
        [updatestate_depend_eliminate]: 3.71001e-06
        [updatestate_assign_eliminate]: 3.73001e-06
        [updatestate_loads_eliminate]: 2.93e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 6.48e-06
        [updatestate_useless_node_eliminater]: 5.76e-06
        [accelerated_algorithm]: 6.07999e-06
        [meta_shard_fg_expand]: 2.31e-06
        [get_grad_eliminate_]: 5.99e-06
        [merge_forward]: 3.52002e-06
        [cell_reuse_recompute_pass]: 7.7e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.354e-05
        [j_node_and_user_rematch]: 9.83998e-06
        [meta_fg_expand]: 0.00010182
        [replace_old_param]: 1.079e-05
        [inline_without_move]: 6.84999e-06
        [renormalize]: 0.00056401
        [add_forward_monad_depend]: 4.48999e-06
        [auto_monad_grad]: 1.50999e-06
        [auto_monad_eliminator]: 1.24e-05
        [cse]: 2.046e-05
        [replace_applicator]: 1.251e-05
    [Cycle 3]: 0.00040098, [27]
        [switch_simplify]: 6.88e-06
        [loop_unroll]: 6.09001e-06
        [a_1]: 0.00013097
        [with_stream_mark]: 1.064e-05
        [recompute_prepare]: 7.3e-06
        [updatestate_depend_eliminate]: 3.5e-06
        [updatestate_assign_eliminate]: 2.79999e-06
        [updatestate_loads_eliminate]: 2.58998e-06
        [parameter_eliminate]: 9.90025e-07
        [specialize_transform]: 6.64001e-06
        [updatestate_useless_node_eliminater]: 6.76999e-06
        [accelerated_algorithm]: 6.15002e-06
        [meta_shard_fg_expand]: 1.50001e-06
        [get_grad_eliminate_]: 5.92999e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 1.77001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.743e-05
        [j_node_and_user_rematch]: 1.062e-05
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 9.54e-06
        [inline_without_move]: 6.73e-06
        [renormalize]: 1.09983e-07
        [add_forward_monad_depend]: 1.32999e-06
        [auto_monad_grad]: 8.49977e-07
        [auto_monad_eliminator]: 7.5e-06
        [cse]: 1.471e-05
        [replace_applicator]: 6.29999e-06
[py_interpret_to_execute_after_opt_a]: 1.487e-05
[rewriter_after_opt_a]: 4.242e-05
[convert_after_rewriter]: 8.18001e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.00078197
[jit_opt_b]: 7.547e-05, [1]
    [Cycle 1]: 6.66e-05, [2]
        [frontend_op_eliminate]: 2.21e-05
        [inline_after_opt_a]: 1.955e-05
[cconv]: 3.135e-05
[loop_unroll]: 0.00047681
[jit_opt_after_cconv]: 0.00017631, [1]
    [Cycle 1]: 0.00016863, [11]
        [c_1]: 2.854e-05
        [parameter_eliminate]: 3.27002e-06
        [updatestate_depend_eliminate]: 8.25e-06
        [updatestate_assign_eliminate]: 4.32e-06
        [updatestate_loads_eliminate]: 2.62001e-06
        [cse]: 2.899e-05
        [call_graph_tuple_transform]: 2.592e-05
        [tuple_list_get_item_eliminator]: 6.93e-06
        [none_parameter_eliminate]: 1.53002e-06
        [renormalize]: 4.2998e-07
        [switch_simplify]: 6.51999e-06
[remove_dup_value]: 1.986e-05
[partial_unused_args_eliminate]: 2.19001e-06
[environ_conv]: 5.99999e-06
[add_recomputation]: 6.06e-05
[cse_after_recomputation]: 3.016e-05, [1]
    [Cycle 1]: 2.441e-05, [1]
        [cse]: 1.715e-05
[auto_monad_reorder]: 2.239e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 6.84999e-06
[opt_after_jit_grad]: 0.00052414
[symbol_engine_optimizer]: 9.053e-05, [1]
    [Cycle 1]: 8.246e-05, [6]
        [build]: 4.57998e-06
        [elim_shapecalc]: 1.099e-05
        [elim_not_effective]: 1.718e-05
        [opt_reshape]: 8.95001e-06
        [fold_const_symbol]: 1.139e-05
        [renormalize]: 4.39992e-07
[validate]: 4.851e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00753854
[execute]: 1.025e-05
Sums
    bootstrap                            :     0.000266s :  0.79%
    type_inference                       :     0.015400s : 45.49%
    event_method                         :     0.000067s :  0.20%
    auto_monad                           :     0.000104s :  0.31%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000031s :  0.09%
    rewriter_before_opt_a                :     0.000101s :  0.30%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000151s :  0.45%
    jit_opt_a.loop_unroll                :     0.000082s :  0.24%
    jit_opt_a.a_1                        :     0.001986s :  5.87%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000026s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000025s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.07%
    jit_opt_a.merge_forward              :     0.000014s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000054s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.001670s :  4.93%
    jit_opt_a.replace_old_param          :     0.000071s :  0.21%
    jit_opt_a.inline_without_move        :     0.000066s :  0.19%
    jit_opt_a.renormalize                :     0.003321s :  9.81%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000067s :  0.20%
    jit_opt_a.cse                        :     0.000177s :  0.52%
    jit_opt_a.replace_applicator         :     0.000090s :  0.26%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.04%
    rewriter_after_opt_a                 :     0.000042s :  0.13%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000782s :  2.31%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.06%
    cconv                                :     0.000031s :  0.09%
    loop_unroll                          :     0.000477s :  1.41%
    jit_opt_after_cconv.c_1              :     0.000029s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000020s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000061s :  0.18%
    cse_after_recomputation.cse          :     0.000017s :  0.05%
    auto_monad_reorder                   :     0.000022s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000524s :  1.55%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000049s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007539s : 22.27%
    execute                              :     0.000010s :  0.03%


Time group info:
------[substitution.]   0.000479   113
 0.56% :     0.000003s :      3: substitution.elim_not_effective
 0.32% :     0.000002s :      3: substitution.fold_const_symbol
 1.33% :     0.000006s :      4: substitution.graph_param_transform
62.98% :     0.000302s :     14: substitution.inline
 3.55% :     0.000017s :      2: substitution.inline_without_move
 1.82% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.07% :     0.000010s :      7: substitution.minmaximum_grad
 1.20% :     0.000006s :      5: substitution.partial_eliminate
 2.25% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.55% :     0.000027s :     10: substitution.replace_applicator
 1.60% :     0.000008s :      7: substitution.replace_old_param
 0.48% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.62% :     0.000008s :      1: substitution.switch_simplify
 4.17% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.97% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.51% :     0.000036s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.015307     2
79.12% :     0.012111s :      1: type_inference.infer
20.88% :     0.003196s :      1: type_inference.specialize
------[replace.]   0.000169    22
54.83% :     0.000093s :     14: replace.inline
16.03% :     0.000027s :      1: replace.switch_simplify
29.14% :     0.000049s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000319    22
92.38% :     0.000294s :     14: match.inline
 2.26% :     0.000007s :      1: match.switch_simplify
 5.36% :     0.000017s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000374  2634
 1.59% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.54% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.40% :     0.000005s :     44: predicate.addn_check_dump
 1.72% :     0.000006s :     44: predicate.addn_zero_filter
 2.14% :     0.000008s :     44: predicate.arithmetic_simplify
 1.66% :     0.000006s :     44: predicate.cast_eliminate
 0.16% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.42% :     0.000005s :     44: predicate.compare_switch_simplify
 1.57% :     0.000006s :     44: predicate.depend_value_elim
 1.44% :     0.000005s :     44: predicate.dict_get_item_const_eliminator
 1.56% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.49% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.36% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      4: predicate.elim_not_effective
 0.21% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.55% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.52% :     0.000006s :     44: predicate.environ_get_add_eliminate
 1.40% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.44% :     0.000005s :     44: predicate.environ_get_eliminate
 1.43% :     0.000005s :     44: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.79% :     0.000003s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.28% :     0.000016s :     73: predicate.inline
 1.86% :     0.000007s :     40: predicate.inline_without_move
 0.34% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.94% :     0.000004s :     19: predicate.less_batch_normalization
 1.80% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.93% :     0.000007s :     55: predicate.load_eliminater
 0.53% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.37% :     0.000013s :     91: predicate.loop_unroll_before_grad
 1.78% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.42% :     0.000005s :     44: predicate.merge_addn
 1.45% :     0.000005s :     44: predicate.minmaximum_grad
 0.77% :     0.000003s :      4: predicate.mutable_eliminate
 0.21% :     0.000001s :      4: predicate.opt_reshape
 2.28% :     0.000009s :     55: predicate.partial_eliminate
 1.51% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.82% :     0.000007s :     44: predicate.reduce_eliminate
 1.80% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.46% :     0.000009s :     97: predicate.replace_applicator
 0.92% :     0.000003s :     40: predicate.replace_old_param
 0.13% :     0.000000s :      4: predicate.reset_defer_inline
 1.71% :     0.000006s :     44: predicate.reshape_eliminate
 1.51% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.28% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.55% :     0.000006s :     44: predicate.same_eliminate
 0.46% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.41% :     0.000002s :      8: predicate.special_op_eliminate
 0.84% :     0.000003s :     19: predicate.specialize_transform
 1.68% :     0.000006s :     44: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.18% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.72% :     0.000010s :     65: predicate.switch_defer_inline
 2.55% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.53% :     0.000024s :    162: predicate.switch_simplify
 1.50% :     0.000006s :     44: predicate.tile_eliminate
 1.50% :     0.000006s :     44: predicate.transpose_eliminate
 1.82% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.72% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 3.08% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 1.89% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.90% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000010s :     74: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000007s :     44: predicate.value_based_eliminate
 0.18% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.37% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001913    32
55.87% :     0.001069s :     14: func_graph_cloner_run.FuncGraphClonerGraph
44.13% :     0.000844s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.044188    91
 0.14% :     0.000064s :      1: add_recomputation
 0.25% :     0.000109s :      1: auto_monad
 0.06% :     0.000025s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.66% :     0.000293s :      1: bootstrap
 0.08% :     0.000034s :      1: cconv
 0.02% :     0.000011s :      1: convert_after_rewriter
 0.07% :     0.000032s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.16% :     0.000072s :      1: event_method
 0.04% :     0.000016s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
26.90% :     0.011885s :      1: jit_opt_a
 0.41% :     0.000181s :      1: jit_opt_after_cconv
 0.18% :     0.000078s :      1: jit_opt_b
 1.10% :     0.000486s :      1: loop_unroll
 1.80% :     0.000794s :      1: mutable_eliminate
 5.96% :     0.002632s :     39: opt.transform.jit_opt_a
 0.15% :     0.000065s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000034s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000045s :      4: opt.transform.symbol_engine_opt
 1.21% :     0.000534s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000033s :      1: py_interpret_to_execute
 0.04% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000023s :      1: remove_dup_value
 3.82% :     0.001688s :      2: renormalize.infer
 3.66% :     0.001618s :      2: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000046s :      1: rewriter_after_opt_a
 0.24% :     0.000104s :      1: rewriter_before_opt_a
 0.21% :     0.000094s :      1: symbol_engine_optimizer
17.11% :     0.007563s :      1: task_emit
34.90% :     0.015420s :      1: type_inference
 0.17% :     0.000073s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.507.467 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc5cdc0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.508.005 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0157447, [30]
[bootstrap]: 0.00039155
[type_inference]: 0.00552312
[event_method]: 1.977e-05
[auto_monad]: 0.00010016
[graph_reusing]: 6.18998e-06
[pre_auto_parallel]: 2.93e-06
[py_interpret_to_execute]: 3.64e-05
[rewriter_before_opt_a]: 7.462e-05
[expand_dump_flag]: 3.81999e-06
[jit_opt_a]: 0.00690762, [2]
    [Cycle 1]: 0.0021419, [27]
        [switch_simplify]: 6.31e-05
        [loop_unroll]: 3.007e-05
        [a_1]: 0.00063796
        [with_stream_mark]: 3.158e-05
        [recompute_prepare]: 1.248e-05
        [updatestate_depend_eliminate]: 4.68001e-06
        [updatestate_assign_eliminate]: 3.13e-06
        [updatestate_loads_eliminate]: 3.01999e-06
        [parameter_eliminate]: 2.98998e-06
        [specialize_transform]: 7.70998e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 6.01e-06
        [meta_shard_fg_expand]: 2.22999e-06
        [get_grad_eliminate_]: 5.88002e-06
        [merge_forward]: 4.89998e-06
        [cell_reuse_recompute_pass]: 1.82999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.742e-05
        [j_node_and_user_rematch]: 9.94001e-06
        [meta_fg_expand]: 2.74999e-06
        [replace_old_param]: 1.171e-05
        [inline_without_move]: 6.30002e-06
        [renormalize]: 0.00095358
        [add_forward_monad_depend]: 9.46e-06
        [auto_monad_grad]: 3.16999e-06
        [auto_monad_eliminator]: 1.885e-05
        [cse]: 3.459e-05
        [replace_applicator]: 1.9e-05
    [Cycle 2]: 0.00035238, [27]
        [switch_simplify]: 6.13998e-06
        [loop_unroll]: 4.95001e-06
        [a_1]: 9.492e-05
        [with_stream_mark]: 1.369e-05
        [recompute_prepare]: 6.02999e-06
        [updatestate_depend_eliminate]: 3.18998e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 1.33002e-06
        [specialize_transform]: 5.49e-06
        [updatestate_useless_node_eliminater]: 5.35999e-06
        [accelerated_algorithm]: 5.90002e-06
        [meta_shard_fg_expand]: 1.30001e-06
        [get_grad_eliminate_]: 4.63999e-06
        [merge_forward]: 2.89999e-06
        [cell_reuse_recompute_pass]: 3.00998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.49e-05
        [j_node_and_user_rematch]: 8.42e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 8.28999e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.74998e-06
        [auto_monad_grad]: 1.30999e-06
        [auto_monad_eliminator]: 5.96e-06
        [cse]: 1.291e-05
        [replace_applicator]: 5.18002e-06
[py_interpret_to_execute_after_opt_a]: 1.487e-05
[rewriter_after_opt_a]: 4.035e-05
[convert_after_rewriter]: 7.46999e-06
[order_py_execute_after_rewriter]: 7.6e-06
[mutable_eliminate]: 0.00082442
[jit_opt_b]: 6.153e-05, [1]
    [Cycle 1]: 5.036e-05, [2]
        [frontend_op_eliminate]: 1.884e-05
        [inline_after_opt_a]: 1.81e-05
[cconv]: 3.692e-05
[loop_unroll]: 0.0005239
[jit_opt_after_cconv]: 0.00016686, [1]
    [Cycle 1]: 0.00015917, [11]
        [c_1]: 2.126e-05
        [parameter_eliminate]: 5.07e-06
        [updatestate_depend_eliminate]: 9.15999e-06
        [updatestate_assign_eliminate]: 3.04001e-06
        [updatestate_loads_eliminate]: 2.64001e-06
        [cse]: 2.806e-05
        [call_graph_tuple_transform]: 2.331e-05
        [tuple_list_get_item_eliminator]: 5.72001e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 8.2e-07
        [switch_simplify]: 5.81998e-06
[remove_dup_value]: 1.531e-05
[partial_unused_args_eliminate]: 2.43e-06
[environ_conv]: 5.94999e-06
[add_recomputation]: 5.613e-05
[cse_after_recomputation]: 2.173e-05, [1]
    [Cycle 1]: 1.587e-05, [1]
        [cse]: 9.47999e-06
[auto_monad_reorder]: 1.545e-05
[get_jit_bprop_graph]: 1.82999e-06
[rewriter_after_jit_bprop_graph]: 6.94001e-06
[opt_after_jit_grad]: 0.0005403
[symbol_engine_optimizer]: 8.07e-05, [1]
    [Cycle 1]: 7.228e-05, [6]
        [build]: 5.02e-06
        [elim_shapecalc]: 8.10999e-06
        [elim_not_effective]: 1.394e-05
        [opt_reshape]: 6.29999e-06
        [fold_const_symbol]: 9.42999e-06
        [renormalize]: 9.99979e-07
[validate]: 4.379e-05
Sums
    bootstrap                            :     0.000392s :  3.68%
    type_inference                       :     0.005523s : 51.92%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000100s :  0.94%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.34%
    rewriter_before_opt_a                :     0.000075s :  0.70%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000069s :  0.65%
    jit_opt_a.loop_unroll                :     0.000035s :  0.33%
    jit_opt_a.a_1                        :     0.000733s :  6.89%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.43%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000013s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.10%
    jit_opt_a.merge_forward              :     0.000008s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000020s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000954s :  8.97%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.11%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.23%
    jit_opt_a.cse                        :     0.000047s :  0.45%
    jit_opt_a.replace_applicator         :     0.000024s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.14%
    rewriter_after_opt_a                 :     0.000040s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000008s :  0.07%
    mutable_eliminate                    :     0.000824s :  7.75%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.17%
    cconv                                :     0.000037s :  0.35%
    loop_unroll                          :     0.000524s :  4.93%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.09%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000028s :  0.26%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000015s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000056s :  0.53%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000540s :  5.08%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.41%


Time group info:
------[substitution.]   0.000273    23
 0.68% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.37% :     0.000006s :      3: substitution.graph_param_transform
85.87% :     0.000235s :      4: substitution.inline
 1.47% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.80% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.16% :     0.000006s :      2: substitution.replace_old_param
 5.10% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005450     2
82.38% :     0.004489s :      1: type_inference.infer
17.62% :     0.000960s :      1: type_inference.specialize
------[replace.]   0.000081     6
67.84% :     0.000055s :      4: replace.inline
32.16% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000245     6
94.83% :     0.000232s :      4: match.inline
 5.17% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.25% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.43% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.12% :     0.000001s :     10: predicate.addn_zero_filter
 2.34% :     0.000003s :     10: predicate.arithmetic_simplify
 1.10% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.08% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 6.09% :     0.000008s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :      6: predicate.less_batch_normalization
 2.04% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.58% :     0.000002s :     15: predicate.load_eliminater
 1.67% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.59% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.65% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 2.54% :     0.000003s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 1.97% :     0.000003s :     15: predicate.partial_eliminate
 1.09% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.88% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.72% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.48% :     0.000002s :     18: predicate.replace_applicator
 0.79% :     0.000001s :      6: predicate.replace_old_param
 0.57% :     0.000001s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.19% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.01% :     0.000001s :     10: predicate.same_eliminate
 0.84% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.22% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.07% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.88% :     0.000004s :     16: predicate.switch_defer_inline
 1.93% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.07% :     0.000011s :     53: predicate.switch_simplify
 1.06% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000002s :     10: predicate.transpose_eliminate
 1.23% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.42% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 2.17% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.37% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000709    12
48.34% :     0.000343s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.21% :     0.000058s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.45% :     0.000308s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017733    72
 0.34% :     0.000059s :      1: add_recomputation
 0.59% :     0.000105s :      1: auto_monad
 0.10% :     0.000018s :      1: auto_monad_reorder
 2.32% :     0.000411s :      1: bootstrap
 0.22% :     0.000040s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.14% :     0.000026s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.98% :     0.006912s :      1: jit_opt_a
 0.96% :     0.000170s :      1: jit_opt_after_cconv
 0.36% :     0.000065s :      1: jit_opt_b
 3.01% :     0.000534s :      1: loop_unroll
 4.73% :     0.000839s :      1: mutable_eliminate
 5.49% :     0.000974s :     26: opt.transform.jit_opt_a
 0.30% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000029s :      4: opt.transform.jit_opt_b
 0.08% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.10% :     0.000549s :      1: opt_after_jit_grad
 0.06% :     0.000010s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000039s :      1: py_interpret_to_execute
 0.10% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000018s :      1: remove_dup_value
 3.07% :     0.000545s :      1: renormalize.infer
 2.24% :     0.000398s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000044s :      1: rewriter_after_opt_a
 0.44% :     0.000078s :      1: rewriter_before_opt_a
 0.47% :     0.000083s :      1: symbol_engine_optimizer
31.26% :     0.005543s :      1: type_inference


TotalTime = 0.0191437, [33]
[bootstrap]: 0.00031777
[type_inference]: 0.00475945
[event_method]: 1.724e-05
[auto_monad]: 6.182e-05
[graph_reusing]: 5.92001e-06
[pre_auto_parallel]: 2.37999e-06
[py_interpret_to_execute]: 3e-05
[rewriter_before_opt_a]: 7.043e-05
[expand_dump_flag]: 3.03e-06
[jit_opt_a]: 0.0053064, [2]
    [Cycle 1]: 0.00151804, [27]
        [switch_simplify]: 5.405e-05
        [loop_unroll]: 2.975e-05
        [a_1]: 0.00051177
        [with_stream_mark]: 1.513e-05
        [recompute_prepare]: 8.20999e-06
        [updatestate_depend_eliminate]: 3.91999e-06
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 2.41998e-06
        [specialize_transform]: 5.87999e-06
        [updatestate_useless_node_eliminater]: 5.10999e-06
        [accelerated_algorithm]: 6.07001e-06
        [meta_shard_fg_expand]: 2.47001e-06
        [get_grad_eliminate_]: 5.19e-06
        [merge_forward]: 3.98001e-06
        [cell_reuse_recompute_pass]: 1.96e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.457e-05
        [j_node_and_user_rematch]: 9.22001e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 9.54e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 0.00057307
        [add_forward_monad_depend]: 5.87001e-06
        [auto_monad_grad]: 2.31e-06
        [auto_monad_eliminator]: 1.431e-05
        [cse]: 2.928e-05
        [replace_applicator]: 1.271e-05
    [Cycle 2]: 0.00033613, [27]
        [switch_simplify]: 5.81998e-06
        [loop_unroll]: 5.17999e-06
        [a_1]: 9.276e-05
        [with_stream_mark]: 9.81e-06
        [recompute_prepare]: 5.66998e-06
        [updatestate_depend_eliminate]: 2.79001e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 9.30013e-07
        [specialize_transform]: 5.09e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.51998e-06
        [meta_shard_fg_expand]: 1.55001e-06
        [get_grad_eliminate_]: 4.60999e-06
        [merge_forward]: 2.65002e-06
        [cell_reuse_recompute_pass]: 2.29001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.377e-05
        [j_node_and_user_rematch]: 8.28999e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 7.85e-06
        [inline_without_move]: 5.01997e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 1.02e-06
        [auto_monad_eliminator]: 5.96e-06
        [cse]: 1.064e-05
        [replace_applicator]: 5.10999e-06
[py_interpret_to_execute_after_opt_a]: 1.036e-05
[rewriter_after_opt_a]: 4.507e-05
[convert_after_rewriter]: 7.91001e-06
[order_py_execute_after_rewriter]: 5.82999e-06
[mutable_eliminate]: 0.00059524
[jit_opt_b]: 5.927e-05, [1]
    [Cycle 1]: 4.387e-05, [2]
        [frontend_op_eliminate]: 1.605e-05
        [inline_after_opt_a]: 1.541e-05
[cconv]: 2.508e-05
[loop_unroll]: 0.00045389
[jit_opt_after_cconv]: 0.0001373, [1]
    [Cycle 1]: 0.00013064, [11]
        [c_1]: 2.049e-05
        [parameter_eliminate]: 2.11e-06
        [updatestate_depend_eliminate]: 4.85001e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [cse]: 1.683e-05
        [call_graph_tuple_transform]: 1.852e-05
        [tuple_list_get_item_eliminator]: 5.42001e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 4.2998e-07
        [switch_simplify]: 5.50001e-06
[remove_dup_value]: 1.28e-05
[partial_unused_args_eliminate]: 2.31e-06
[environ_conv]: 4.87e-06
[add_recomputation]: 4.845e-05
[cse_after_recomputation]: 2.067e-05, [1]
    [Cycle 1]: 1.511e-05, [1]
        [cse]: 9.05999e-06
[auto_monad_reorder]: 1.456e-05
[get_jit_bprop_graph]: 2.27999e-06
[rewriter_after_jit_bprop_graph]: 4.46002e-06
[opt_after_jit_grad]: 0.00047625
[symbol_engine_optimizer]: 0.00035356, [1]
    [Cycle 1]: 0.00034607, [6]
        [build]: 3.43e-06
        [elim_shapecalc]: 7.98001e-06
        [elim_not_effective]: 0.00027334
        [opt_reshape]: 8.05e-06
        [fold_const_symbol]: 1.009e-05
        [renormalize]: 4.39992e-07
[validate]: 3.786e-05
[backend_pass]: 1.00999e-06
[task_emit]: 0.00604397
[execute]: 6.73998e-06
Sums
    bootstrap                            :     0.000318s :  2.11%
    type_inference                       :     0.004759s : 31.63%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000062s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.20%
    rewriter_before_opt_a                :     0.000070s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000060s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.23%
    jit_opt_a.a_1                        :     0.000605s :  4.02%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000573s :  3.81%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000040s :  0.27%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000045s :  0.30%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000595s :  3.96%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000454s :  3.02%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000048s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000476s :  3.17%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000273s :  1.82%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006044s : 40.17%
    execute                              :     0.000007s :  0.04%


Time group info:
------[substitution.]   0.000182    23
 1.31% :     0.000002s :      2: substitution.elim_not_effective
 0.81% :     0.000001s :      2: substitution.fold_const_symbol
 3.07% :     0.000006s :      3: substitution.graph_param_transform
80.86% :     0.000147s :      4: substitution.inline
 1.76% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.68% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.40% :     0.000004s :      2: substitution.replace_old_param
 7.10% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004689     2
81.95% :     0.003843s :      1: type_inference.infer
18.05% :     0.000846s :      1: type_inference.specialize
------[replace.]   0.000063     6
65.48% :     0.000041s :      4: replace.inline
34.52% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000156     6
92.39% :     0.000144s :      4: match.inline
 7.61% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.43% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.57% :     0.000002s :     10: predicate.addn_zero_filter
 2.11% :     0.000003s :     10: predicate.arithmetic_simplify
 1.27% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.22% :     0.000001s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.42% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      3: predicate.elim_not_effective
 0.76% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.20% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.71% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.56% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.33% :     0.000002s :      3: predicate.mutable_eliminate
 0.84% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.71% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.82% :     0.000001s :      6: predicate.replace_old_param
 0.49% :     0.000001s :      3: predicate.reset_defer_inline
 1.57% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.05% :     0.000010s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.28% :     0.000002s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.07% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.91% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000602    12
46.32% :     0.000279s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 3.91% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.76% :     0.000300s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020904    76
 0.25% :     0.000051s :      1: add_recomputation
 0.31% :     0.000065s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.61% :     0.000337s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.07% :     0.000015s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.40% :     0.005309s :      1: jit_opt_a
 0.67% :     0.000140s :      1: jit_opt_after_cconv
 0.30% :     0.000062s :      1: jit_opt_b
 2.22% :     0.000463s :      1: loop_unroll
 2.89% :     0.000604s :      1: mutable_eliminate
 3.91% :     0.000818s :     26: opt.transform.jit_opt_a
 0.22% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 1.41% :     0.000295s :      4: opt.transform.symbol_engine_opt
 2.32% :     0.000485s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.43% :     0.000299s :      1: renormalize.infer
 1.27% :     0.000266s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000051s :      1: rewriter_after_opt_a
 0.35% :     0.000073s :      1: rewriter_before_opt_a
 1.71% :     0.000357s :      1: symbol_engine_optimizer
28.96% :     0.006054s :      1: task_emit
22.83% :     0.004771s :      1: type_inference
 0.27% :     0.000056s :      1: validate


TotalTime = 0.0244768, [33]
[bootstrap]: 0.00038317
[type_inference]: 0.00612623
[event_method]: 2.035e-05
[auto_monad]: 7.779e-05
[graph_reusing]: 6.23e-06
[pre_auto_parallel]: 3.01001e-06
[py_interpret_to_execute]: 5.095e-05
[rewriter_before_opt_a]: 8.361e-05
[expand_dump_flag]: 3.26999e-06
[jit_opt_a]: 0.00743022, [2]
    [Cycle 1]: 0.00246435, [27]
        [switch_simplify]: 6.866e-05
        [loop_unroll]: 2.996e-05
        [a_1]: 0.00067651
        [with_stream_mark]: 3.086e-05
        [recompute_prepare]: 1.546e-05
        [updatestate_depend_eliminate]: 5.00001e-06
        [updatestate_assign_eliminate]: 3.06001e-06
        [updatestate_loads_eliminate]: 3.08998e-06
        [parameter_eliminate]: 2.04999e-06
        [specialize_transform]: 8.34002e-06
        [updatestate_useless_node_eliminater]: 6.01e-06
        [accelerated_algorithm]: 6.46e-06
        [meta_shard_fg_expand]: 2.41e-06
        [get_grad_eliminate_]: 5.64998e-06
        [merge_forward]: 4.24997e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.223e-05
        [j_node_and_user_rematch]: 1.106e-05
        [meta_fg_expand]: 2.79999e-06
        [replace_old_param]: 1.303e-05
        [inline_without_move]: 6.78e-06
        [renormalize]: 0.0011824
        [add_forward_monad_depend]: 1.169e-05
        [auto_monad_grad]: 3.5e-06
        [auto_monad_eliminator]: 2.258e-05
        [cse]: 3.597e-05
        [replace_applicator]: 2.736e-05
    [Cycle 2]: 0.0004504, [27]
        [switch_simplify]: 7.68999e-06
        [loop_unroll]: 6.17999e-06
        [a_1]: 0.00010688
        [with_stream_mark]: 2.031e-05
        [recompute_prepare]: 6.90998e-06
        [updatestate_depend_eliminate]: 3.9e-06
        [updatestate_assign_eliminate]: 3.33e-06
        [updatestate_loads_eliminate]: 2.91e-06
        [parameter_eliminate]: 2.66999e-06
        [specialize_transform]: 5.72001e-06
        [updatestate_useless_node_eliminater]: 5.52001e-06
        [accelerated_algorithm]: 6.16e-06
        [meta_shard_fg_expand]: 2.59999e-06
        [get_grad_eliminate_]: 6.45002e-06
        [merge_forward]: 5.12e-06
        [cell_reuse_recompute_pass]: 4.90999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.344e-05
        [j_node_and_user_rematch]: 9.76998e-06
        [meta_fg_expand]: 2.16e-06
        [replace_old_param]: 1.068e-05
        [inline_without_move]: 5.39998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.78e-06
        [auto_monad_grad]: 1.86e-06
        [auto_monad_eliminator]: 8.06001e-06
        [cse]: 2.658e-05
        [replace_applicator]: 7.72002e-06
[py_interpret_to_execute_after_opt_a]: 2.108e-05
[rewriter_after_opt_a]: 4.509e-05
[convert_after_rewriter]: 9.75002e-06
[order_py_execute_after_rewriter]: 5.87999e-06
[mutable_eliminate]: 0.00092235
[jit_opt_b]: 6.63e-05, [1]
    [Cycle 1]: 5.577e-05, [2]
        [frontend_op_eliminate]: 2.21e-05
        [inline_after_opt_a]: 1.943e-05
[cconv]: 3.898e-05
[loop_unroll]: 0.00048221
[jit_opt_after_cconv]: 0.00017347, [1]
    [Cycle 1]: 0.00016618, [11]
        [c_1]: 2.303e-05
        [parameter_eliminate]: 5.14998e-06
        [updatestate_depend_eliminate]: 1.065e-05
        [updatestate_assign_eliminate]: 2.99001e-06
        [updatestate_loads_eliminate]: 2.98998e-06
        [cse]: 3.136e-05
        [call_graph_tuple_transform]: 2.413e-05
        [tuple_list_get_item_eliminator]: 5.71003e-06
        [none_parameter_eliminate]: 1.56002e-06
        [renormalize]: 9.39996e-07
        [switch_simplify]: 5.44e-06
[remove_dup_value]: 1.529e-05
[partial_unused_args_eliminate]: 2.29999e-06
[environ_conv]: 6.05002e-06
[add_recomputation]: 5.627e-05
[cse_after_recomputation]: 2.291e-05, [1]
    [Cycle 1]: 1.648e-05, [1]
        [cse]: 9.61e-06
[auto_monad_reorder]: 1.514e-05
[get_jit_bprop_graph]: 3.14999e-06
[rewriter_after_jit_bprop_graph]: 8.38001e-06
[opt_after_jit_grad]: 0.00049975
[symbol_engine_optimizer]: 8.856e-05, [1]
    [Cycle 1]: 8.088e-05, [6]
        [build]: 5.81e-06
        [elim_shapecalc]: 8.27e-06
        [elim_not_effective]: 1.576e-05
        [opt_reshape]: 6.59001e-06
        [fold_const_symbol]: 1.009e-05
        [renormalize]: 1.12e-06
[validate]: 0.00010364
[backend_pass]: 1.49998e-06
[task_emit]: 0.00743868
[execute]: 8.90999e-06
Sums
    bootstrap                            :     0.000383s :  2.00%
    type_inference                       :     0.006126s : 31.98%
    event_method                         :     0.000020s :  0.11%
    auto_monad                           :     0.000078s :  0.41%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000051s :  0.27%
    rewriter_before_opt_a                :     0.000084s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.40%
    jit_opt_a.loop_unroll                :     0.000036s :  0.19%
    jit_opt_a.a_1                        :     0.000783s :  4.09%
    jit_opt_a.with_stream_mark           :     0.000051s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000014s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.06%
    jit_opt_a.merge_forward              :     0.000009s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000046s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000024s :  0.12%
    jit_opt_a.inline_without_move        :     0.000012s :  0.06%
    jit_opt_a.renormalize                :     0.001182s :  6.17%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000031s :  0.16%
    jit_opt_a.cse                        :     0.000063s :  0.33%
    jit_opt_a.replace_applicator         :     0.000035s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000021s :  0.11%
    rewriter_after_opt_a                 :     0.000045s :  0.24%
    convert_after_rewriter               :     0.000010s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000922s :  4.81%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.10%
    cconv                                :     0.000039s :  0.20%
    loop_unroll                          :     0.000482s :  2.52%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000031s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000056s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.08%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000500s :  2.61%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000104s :  0.54%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007439s : 38.83%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000297    23
 0.98% :     0.000003s :      2: substitution.elim_not_effective
 0.42% :     0.000001s :      2: substitution.fold_const_symbol
 2.29% :     0.000007s :      3: substitution.graph_param_transform
84.63% :     0.000251s :      4: substitution.inline
 1.56% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.28% :     0.000007s :      4: substitution.remove_not_recompute_node
 2.72% :     0.000008s :      2: substitution.replace_old_param
 5.10% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006036     2
82.31% :     0.004969s :      1: type_inference.infer
17.69% :     0.001068s :      1: type_inference.specialize
------[replace.]   0.000096     6
63.52% :     0.000061s :      4: replace.inline
36.48% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000262     6
94.62% :     0.000248s :      4: match.inline
 5.38% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   691
 1.25% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.89% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 2.57% :     0.000004s :     10: predicate.arithmetic_simplify
 1.58% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.89% :     0.000001s :     10: predicate.compare_switch_simplify
 1.16% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.07% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.05% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.72% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.60% :     0.000001s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.02% :     0.000001s :     10: predicate.environ_get_eliminate
 1.21% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.13% :     0.000002s :      6: predicate.get_grad_eliminate
 0.44% :     0.000001s :      3: predicate.graph_param_transform
 5.44% :     0.000007s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.65% :     0.000002s :      6: predicate.less_batch_normalization
 1.81% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.49% :     0.000002s :     15: predicate.load_eliminater
 1.54% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.55% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.10% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.92% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 3.05% :     0.000004s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 1.98% :     0.000003s :     15: predicate.partial_eliminate
 1.08% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.80% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.67% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.99% :     0.000003s :     18: predicate.replace_applicator
 0.41% :     0.000001s :      6: predicate.replace_old_param
 0.67% :     0.000001s :      3: predicate.reset_defer_inline
 1.97% :     0.000003s :     10: predicate.reshape_eliminate
 1.09% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.72% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000002s :     10: predicate.same_eliminate
 0.86% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.50% :     0.000002s :      6: predicate.specialize_transform
 1.25% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.31% :     0.000003s :     16: predicate.switch_defer_inline
 1.71% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.17% :     0.000010s :     53: predicate.switch_simplify
 1.18% :     0.000002s :     10: predicate.tile_eliminate
 1.03% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.24% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.40% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.33% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000796    12
46.66% :     0.000372s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.92% :     0.000063s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.42% :     0.000362s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026846    76
 0.22% :     0.000060s :      1: add_recomputation
 0.31% :     0.000082s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.48% :     0.000399s :      1: bootstrap
 0.16% :     0.000042s :      1: cconv
 0.05% :     0.000013s :      1: convert_after_rewriter
 0.09% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.10% :     0.000026s :      1: event_method
 0.05% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
27.70% :     0.007435s :      1: jit_opt_a
 0.66% :     0.000177s :      1: jit_opt_after_cconv
 0.26% :     0.000070s :      1: jit_opt_b
 1.83% :     0.000491s :      1: loop_unroll
 3.49% :     0.000938s :      1: mutable_eliminate
 3.96% :     0.001064s :     26: opt.transform.jit_opt_a
 0.20% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000033s :      4: opt.transform.jit_opt_b
 0.05% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000029s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.90% :     0.000509s :      1: opt_after_jit_grad
 0.03% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000054s :      1: py_interpret_to_execute
 0.09% :     0.000024s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.42% :     0.000649s :      1: renormalize.infer
 1.93% :     0.000518s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000049s :      1: rewriter_after_opt_a
 0.32% :     0.000087s :      1: rewriter_before_opt_a
 0.34% :     0.000091s :      1: symbol_engine_optimizer
27.78% :     0.007458s :      1: task_emit
22.91% :     0.006150s :      1: type_inference
 0.49% :     0.000131s :      1: validate


TotalTime = 0.0353037, [33]
[bootstrap]: 0.00018064
[type_inference]: 0.0134635
[event_method]: 6.457e-05
[auto_monad]: 0.00010865
[graph_reusing]: 7.61001e-06
[pre_auto_parallel]: 3.38e-06
[py_interpret_to_execute]: 2.889e-05
[rewriter_before_opt_a]: 0.00010544
[expand_dump_flag]: 3.45e-06
[jit_opt_a]: 0.0122739, [3]
    [Cycle 1]: 0.00626096, [27]
        [switch_simplify]: 0.00012442
        [loop_unroll]: 4.327e-05
        [a_1]: 0.00089072
        [with_stream_mark]: 2.956e-05
        [recompute_prepare]: 1.909e-05
        [updatestate_depend_eliminate]: 8.43001e-06
        [updatestate_assign_eliminate]: 7.06001e-06
        [updatestate_loads_eliminate]: 6.61e-06
        [parameter_eliminate]: 2.70997e-06
        [specialize_transform]: 1.381e-05
        [updatestate_useless_node_eliminater]: 1.278e-05
        [accelerated_algorithm]: 1.333e-05
        [meta_shard_fg_expand]: 3.91999e-06
        [get_grad_eliminate_]: 1.242e-05
        [merge_forward]: 9.18002e-06
        [cell_reuse_recompute_pass]: 1.11002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.711e-05
        [j_node_and_user_rematch]: 2.343e-05
        [meta_fg_expand]: 0.0014793
        [replace_old_param]: 5.923e-05
        [inline_without_move]: 5.272e-05
        [renormalize]: 0.00291254
        [add_forward_monad_depend]: 1.082e-05
        [auto_monad_grad]: 6.62002e-06
        [auto_monad_eliminator]: 4.863e-05
        [cse]: 0.00016305
        [replace_applicator]: 6.603e-05
    [Cycle 2]: 0.00209978, [27]
        [switch_simplify]: 3.999e-05
        [loop_unroll]: 3.607e-05
        [a_1]: 0.00105262
        [with_stream_mark]: 1.191e-05
        [recompute_prepare]: 7.64002e-06
        [updatestate_depend_eliminate]: 3.61999e-06
        [updatestate_assign_eliminate]: 3.21999e-06
        [updatestate_loads_eliminate]: 2.91999e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 7.45e-06
        [updatestate_useless_node_eliminater]: 6.18002e-06
        [accelerated_algorithm]: 6.32001e-06
        [meta_shard_fg_expand]: 2.96999e-06
        [get_grad_eliminate_]: 5.84e-06
        [merge_forward]: 3.43e-06
        [cell_reuse_recompute_pass]: 7.39994e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.563e-05
        [j_node_and_user_rematch]: 1.088e-05
        [meta_fg_expand]: 0.00015087
        [replace_old_param]: 1.095e-05
        [inline_without_move]: 6.22001e-06
        [renormalize]: 0.00050742
        [add_forward_monad_depend]: 4.92e-06
        [auto_monad_grad]: 1.44e-06
        [auto_monad_eliminator]: 1.206e-05
        [cse]: 2.145e-05
        [replace_applicator]: 1.352e-05
    [Cycle 3]: 0.00037502, [27]
        [switch_simplify]: 7.28999e-06
        [loop_unroll]: 5.98998e-06
        [a_1]: 0.00011971
        [with_stream_mark]: 8.97e-06
        [recompute_prepare]: 6.34999e-06
        [updatestate_depend_eliminate]: 3.51999e-06
        [updatestate_assign_eliminate]: 2.69001e-06
        [updatestate_loads_eliminate]: 2.55002e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 6.68003e-06
        [updatestate_useless_node_eliminater]: 5.66e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 2.04e-06
        [get_grad_eliminate_]: 6.23e-06
        [merge_forward]: 2.96999e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.449e-05
        [j_node_and_user_rematch]: 9.87001e-06
        [meta_fg_expand]: 2.04999e-06
        [replace_old_param]: 8.42e-06
        [inline_without_move]: 5.91e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 6.72002e-06
        [cse]: 1.358e-05
        [replace_applicator]: 6.20002e-06
[py_interpret_to_execute_after_opt_a]: 1.384e-05
[rewriter_after_opt_a]: 4.422e-05
[convert_after_rewriter]: 8.23999e-06
[order_py_execute_after_rewriter]: 5.45001e-06
[mutable_eliminate]: 0.00057181
[jit_opt_b]: 5.588e-05, [1]
    [Cycle 1]: 4.815e-05, [2]
        [frontend_op_eliminate]: 1.882e-05
        [inline_after_opt_a]: 1.745e-05
[cconv]: 2.249e-05
[loop_unroll]: 0.00042698
[jit_opt_after_cconv]: 0.00015371, [1]
    [Cycle 1]: 0.00014688, [11]
        [c_1]: 2.676e-05
        [parameter_eliminate]: 2.36e-06
        [updatestate_depend_eliminate]: 5.47001e-06
        [updatestate_assign_eliminate]: 3.44001e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [cse]: 2.186e-05
        [call_graph_tuple_transform]: 2.007e-05
        [tuple_list_get_item_eliminator]: 6.43998e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 6.39999e-06
[remove_dup_value]: 1.609e-05
[partial_unused_args_eliminate]: 2.11e-06
[environ_conv]: 5.03002e-06
[add_recomputation]: 4.969e-05
[cse_after_recomputation]: 3.415e-05, [1]
    [Cycle 1]: 2.136e-05, [1]
        [cse]: 1.529e-05
[auto_monad_reorder]: 1.916e-05
[get_jit_bprop_graph]: 1.81998e-06
[rewriter_after_jit_bprop_graph]: 5.41002e-06
[opt_after_jit_grad]: 0.00044672
[symbol_engine_optimizer]: 7.863e-05, [1]
    [Cycle 1]: 7.253e-05, [6]
        [build]: 3.78999e-06
        [elim_shapecalc]: 8.59e-06
        [elim_not_effective]: 1.502e-05
        [opt_reshape]: 6.81001e-06
        [fold_const_symbol]: 1.054e-05
        [renormalize]: 3.69997e-07
[validate]: 3.385e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00687082
[execute]: 7.01001e-06
Sums
    bootstrap                            :     0.000181s :  0.58%
    type_inference                       :     0.013463s : 43.48%
    event_method                         :     0.000065s :  0.21%
    auto_monad                           :     0.000109s :  0.35%
    graph_reusing                        :     0.000008s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000029s :  0.09%
    rewriter_before_opt_a                :     0.000105s :  0.34%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000172s :  0.55%
    jit_opt_a.loop_unroll                :     0.000085s :  0.28%
    jit_opt_a.a_1                        :     0.002063s :  6.66%
    jit_opt_a.with_stream_mark           :     0.000050s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000028s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000025s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000009s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.08%
    jit_opt_a.merge_forward              :     0.000016s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000057s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000044s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.001632s :  5.27%
    jit_opt_a.replace_old_param          :     0.000079s :  0.25%
    jit_opt_a.inline_without_move        :     0.000065s :  0.21%
    jit_opt_a.renormalize                :     0.003420s : 11.04%
    jit_opt_a.add_forward_monad_depend   :     0.000017s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000067s :  0.22%
    jit_opt_a.cse                        :     0.000198s :  0.64%
    jit_opt_a.replace_applicator         :     0.000086s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.04%
    rewriter_after_opt_a                 :     0.000044s :  0.14%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000572s :  1.85%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.06%
    cconv                                :     0.000022s :  0.07%
    loop_unroll                          :     0.000427s :  1.38%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000050s :  0.16%
    cse_after_recomputation.cse          :     0.000015s :  0.05%
    auto_monad_reorder                   :     0.000019s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000447s :  1.44%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006871s : 22.19%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000506   113
 0.53% :     0.000003s :      3: substitution.elim_not_effective
 0.30% :     0.000002s :      3: substitution.fold_const_symbol
 1.11% :     0.000006s :      4: substitution.graph_param_transform
63.22% :     0.000320s :     14: substitution.inline
 3.35% :     0.000017s :      2: substitution.inline_without_move
 1.85% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.17% :     0.000011s :      7: substitution.minmaximum_grad
 1.28% :     0.000006s :      5: substitution.partial_eliminate
 2.20% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.27% :     0.000027s :     10: substitution.replace_applicator
 1.45% :     0.000007s :      7: substitution.replace_old_param
 0.49% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.58% :     0.000008s :      1: substitution.switch_simplify
 4.46% :     0.000023s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.95% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 7.80% :     0.000039s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.013393     2
88.92% :     0.011908s :      1: type_inference.infer
11.08% :     0.001484s :      1: type_inference.specialize
------[replace.]   0.000179    22
53.01% :     0.000095s :     14: replace.inline
18.01% :     0.000032s :      1: replace.switch_simplify
28.99% :     0.000052s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000340    22
91.80% :     0.000312s :     14: match.inline
 2.18% :     0.000007s :      1: match.switch_simplify
 6.03% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000384  2634
 1.56% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.51% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.42% :     0.000005s :     44: predicate.addn_check_dump
 1.73% :     0.000007s :     44: predicate.addn_zero_filter
 1.94% :     0.000007s :     44: predicate.arithmetic_simplify
 1.64% :     0.000006s :     44: predicate.cast_eliminate
 0.17% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.41% :     0.000005s :     44: predicate.compare_switch_simplify
 1.54% :     0.000006s :     44: predicate.depend_value_elim
 1.43% :     0.000006s :     44: predicate.dict_get_item_const_eliminator
 1.49% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.52% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.27% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      4: predicate.elim_not_effective
 0.19% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.55% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.44% :     0.000006s :     44: predicate.environ_get_add_eliminate
 1.42% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.50% :     0.000006s :     44: predicate.environ_get_eliminate
 1.44% :     0.000006s :     44: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.88% :     0.000003s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.40% :     0.000017s :     73: predicate.inline
 1.74% :     0.000007s :     40: predicate.inline_without_move
 0.34% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.89% :     0.000003s :     19: predicate.less_batch_normalization
 1.78% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.94% :     0.000007s :     55: predicate.load_eliminater
 0.47% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.44% :     0.000013s :     91: predicate.loop_unroll_before_grad
 1.80% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.43% :     0.000006s :     44: predicate.merge_addn
 1.48% :     0.000006s :     44: predicate.minmaximum_grad
 0.54% :     0.000002s :      4: predicate.mutable_eliminate
 0.22% :     0.000001s :      4: predicate.opt_reshape
 2.34% :     0.000009s :     55: predicate.partial_eliminate
 1.47% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.82% :     0.000007s :     44: predicate.reduce_eliminate
 1.77% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.51% :     0.000010s :     97: predicate.replace_applicator
 0.97% :     0.000004s :     40: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.80% :     0.000007s :     44: predicate.reshape_eliminate
 1.52% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.25% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.50% :     0.000006s :     44: predicate.same_eliminate
 0.50% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.41% :     0.000002s :      8: predicate.special_op_eliminate
 0.92% :     0.000004s :     19: predicate.specialize_transform
 1.68% :     0.000006s :     44: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.90% :     0.000011s :     65: predicate.switch_defer_inline
 2.63% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.91% :     0.000027s :    162: predicate.switch_simplify
 1.48% :     0.000006s :     44: predicate.tile_eliminate
 1.46% :     0.000006s :     44: predicate.transpose_eliminate
 1.79% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.60% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 3.07% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 1.89% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.89% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000007s :     44: predicate.value_based_eliminate
 0.15% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.21% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001716    32
58.11% :     0.000997s :     14: func_graph_cloner_run.FuncGraphClonerGraph
41.89% :     0.000719s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.041587    91
 0.13% :     0.000053s :      1: add_recomputation
 0.27% :     0.000113s :      1: auto_monad
 0.05% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.47% :     0.000195s :      1: bootstrap
 0.06% :     0.000025s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000036s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.17% :     0.000070s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
29.52% :     0.012278s :      1: jit_opt_a
 0.38% :     0.000157s :      1: jit_opt_after_cconv
 0.14% :     0.000059s :      1: jit_opt_b
 1.04% :     0.000434s :      1: loop_unroll
 1.40% :     0.000580s :      1: mutable_eliminate
 6.59% :     0.002740s :     39: opt.transform.jit_opt_a
 0.14% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000030s :      4: opt.transform.jit_opt_b
 0.03% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.09% :     0.000453s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000032s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000018s :      1: remove_dup_value
 4.74% :     0.001969s :      2: renormalize.infer
 3.45% :     0.001436s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000047s :      1: rewriter_after_opt_a
 0.26% :     0.000109s :      1: rewriter_before_opt_a
 0.19% :     0.000081s :      1: symbol_engine_optimizer
16.55% :     0.006881s :      1: task_emit
32.41% :     0.013477s :      1: type_inference
 0.13% :     0.000053s :      1: validate


TotalTime = 0.0148448, [30]
[bootstrap]: 0.00035553
[type_inference]: 0.00518474
[event_method]: 1.954e-05
[auto_monad]: 6.654e-05
[graph_reusing]: 5.97999e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 3.9e-05
[rewriter_before_opt_a]: 7.646e-05
[expand_dump_flag]: 3.16001e-06
[jit_opt_a]: 0.00651338, [2]
    [Cycle 1]: 0.00196824, [27]
        [switch_simplify]: 6.915e-05
        [loop_unroll]: 3.031e-05
        [a_1]: 0.00063645
        [with_stream_mark]: 2.662e-05
        [recompute_prepare]: 1.064e-05
        [updatestate_depend_eliminate]: 4.02998e-06
        [updatestate_assign_eliminate]: 2.89001e-06
        [updatestate_loads_eliminate]: 3.02002e-06
        [parameter_eliminate]: 2.06e-06
        [specialize_transform]: 6.40002e-06
        [updatestate_useless_node_eliminater]: 5.31998e-06
        [accelerated_algorithm]: 6.31e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 5.97999e-06
        [merge_forward]: 4.47e-06
        [cell_reuse_recompute_pass]: 1.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.614e-05
        [j_node_and_user_rematch]: 1.08e-05
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 1.169e-05
        [inline_without_move]: 5.77001e-06
        [renormalize]: 0.00082635
        [add_forward_monad_depend]: 8.59e-06
        [auto_monad_grad]: 2.98e-06
        [auto_monad_eliminator]: 1.562e-05
        [cse]: 3.102e-05
        [replace_applicator]: 1.603e-05
    [Cycle 2]: 0.00034995, [27]
        [switch_simplify]: 5.71e-06
        [loop_unroll]: 5.30001e-06
        [a_1]: 9.447e-05
        [with_stream_mark]: 1.281e-05
        [recompute_prepare]: 5.14e-06
        [updatestate_depend_eliminate]: 3.3e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.97001e-06
        [specialize_transform]: 6.02001e-06
        [updatestate_useless_node_eliminater]: 5.49e-06
        [accelerated_algorithm]: 5.96e-06
        [meta_shard_fg_expand]: 2.07001e-06
        [get_grad_eliminate_]: 4.95999e-06
        [merge_forward]: 3.66999e-06
        [cell_reuse_recompute_pass]: 2.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.593e-05
        [j_node_and_user_rematch]: 8.90001e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 7.21999e-06
        [inline_without_move]: 5.25001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.39998e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 5.82001e-06
        [cse]: 1.138e-05
        [replace_applicator]: 5.51e-06
[py_interpret_to_execute_after_opt_a]: 1.51e-05
[rewriter_after_opt_a]: 3.822e-05
[convert_after_rewriter]: 7.06001e-06
[order_py_execute_after_rewriter]: 5.35999e-06
[mutable_eliminate]: 0.00080919
[jit_opt_b]: 5.47e-05, [1]
    [Cycle 1]: 4.586e-05, [2]
        [frontend_op_eliminate]: 1.872e-05
        [inline_after_opt_a]: 1.541e-05
[cconv]: 3.366e-05
[loop_unroll]: 0.00049195
[jit_opt_after_cconv]: 0.00015896, [1]
    [Cycle 1]: 0.00015177, [11]
        [c_1]: 2.273e-05
        [parameter_eliminate]: 3.68999e-06
        [updatestate_depend_eliminate]: 6.56999e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [cse]: 2.441e-05
        [call_graph_tuple_transform]: 2.261e-05
        [tuple_list_get_item_eliminator]: 5.62001e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.63997e-06
[remove_dup_value]: 1.519e-05
[partial_unused_args_eliminate]: 2.73e-06
[environ_conv]: 6.09999e-06
[add_recomputation]: 5.34e-05
[cse_after_recomputation]: 2.132e-05, [1]
    [Cycle 1]: 1.463e-05, [1]
        [cse]: 8.54e-06
[auto_monad_reorder]: 1.593e-05
[get_jit_bprop_graph]: 3.02002e-06
[rewriter_after_jit_bprop_graph]: 8.30999e-06
[opt_after_jit_grad]: 0.00049928
[symbol_engine_optimizer]: 7.968e-05, [1]
    [Cycle 1]: 7.263e-05, [6]
        [build]: 4.77e-06
        [elim_shapecalc]: 8.52998e-06
        [elim_not_effective]: 1.407e-05
        [opt_reshape]: 6.14999e-06
        [fold_const_symbol]: 9.59999e-06
        [renormalize]: 7.00005e-07
[validate]: 4.336e-05
Sums
    bootstrap                            :     0.000356s :  3.56%
    type_inference                       :     0.005185s : 51.96%
    event_method                         :     0.000020s :  0.20%
    auto_monad                           :     0.000067s :  0.67%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.39%
    rewriter_before_opt_a                :     0.000076s :  0.77%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000075s :  0.75%
    jit_opt_a.loop_unroll                :     0.000036s :  0.36%
    jit_opt_a.a_1                        :     0.000731s :  7.33%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.40%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000826s :  8.28%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.21%
    jit_opt_a.cse                        :     0.000042s :  0.42%
    jit_opt_a.replace_applicator         :     0.000022s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.15%
    rewriter_after_opt_a                 :     0.000038s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000809s :  8.11%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.15%
    cconv                                :     0.000034s :  0.34%
    loop_unroll                          :     0.000492s :  4.93%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000015s :  0.15%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000053s :  0.54%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.16%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.08%
    opt_after_jit_grad                   :     0.000499s :  5.00%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.43%


Time group info:
------[substitution.]   0.000268    23
 0.71% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000002s :      2: substitution.fold_const_symbol
 2.25% :     0.000006s :      3: substitution.graph_param_transform
85.48% :     0.000229s :      4: substitution.inline
 1.84% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.82% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.32% :     0.000006s :      2: substitution.replace_old_param
 4.90% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005117     2
82.26% :     0.004209s :      1: type_inference.infer
17.74% :     0.000908s :      1: type_inference.specialize
------[replace.]   0.000081     6
69.02% :     0.000056s :      4: replace.inline
30.98% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000239     6
94.94% :     0.000227s :      4: match.inline
 5.06% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.22% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.32% :     0.000003s :     10: predicate.arithmetic_simplify
 1.17% :     0.000002s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 0.99% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.01% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000001s :      3: predicate.graph_param_transform
 5.58% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.66% :     0.000002s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.61% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.02% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 1.22% :     0.000002s :     10: predicate.minmaximum_grad
 2.27% :     0.000003s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     15: predicate.partial_eliminate
 1.38% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.56% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.85% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.53% :     0.000002s :     18: predicate.replace_applicator
 0.92% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.37% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000002s :     10: predicate.same_eliminate
 0.74% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.01% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 1.96% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.68% :     0.000010s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.85% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.29% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.05% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000680    12
49.56% :     0.000337s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.98% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.46% :     0.000289s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016672    72
 0.34% :     0.000056s :      1: add_recomputation
 0.42% :     0.000070s :      1: auto_monad
 0.11% :     0.000019s :      1: auto_monad_reorder
 2.23% :     0.000372s :      1: bootstrap
 0.22% :     0.000037s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.04% :     0.000006s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.10% :     0.006518s :      1: jit_opt_a
 0.97% :     0.000162s :      1: jit_opt_after_cconv
 0.34% :     0.000057s :      1: jit_opt_b
 3.01% :     0.000502s :      1: loop_unroll
 4.92% :     0.000820s :      1: mutable_eliminate
 5.84% :     0.000973s :     26: opt.transform.jit_opt_a
 0.31% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000028s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.05% :     0.000508s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000042s :      1: py_interpret_to_execute
 0.11% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.63% :     0.000438s :      1: renormalize.infer
 2.27% :     0.000379s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000042s :      1: rewriter_after_opt_a
 0.48% :     0.000079s :      1: rewriter_before_opt_a
 0.49% :     0.000082s :      1: symbol_engine_optimizer
31.18% :     0.005199s :      1: type_inference


TotalTime = 0.0172826, [33]
[bootstrap]: 0.00029648
[type_inference]: 0.00423183
[event_method]: 1.634e-05
[auto_monad]: 6.105e-05
[graph_reusing]: 4.72e-06
[pre_auto_parallel]: 2.22999e-06
[py_interpret_to_execute]: 2.702e-05
[rewriter_before_opt_a]: 6.751e-05
[expand_dump_flag]: 3.11999e-06
[jit_opt_a]: 0.00511677, [2]
    [Cycle 1]: 0.00139241, [27]
        [switch_simplify]: 4.582e-05
        [loop_unroll]: 2.88e-05
        [a_1]: 0.00049649
        [with_stream_mark]: 1.573e-05
        [recompute_prepare]: 6.89999e-06
        [updatestate_depend_eliminate]: 3.28e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [parameter_eliminate]: 1.44e-06
        [specialize_transform]: 6.22001e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.33002e-06
        [meta_shard_fg_expand]: 1.71e-06
        [get_grad_eliminate_]: 5.32001e-06
        [merge_forward]: 2.98e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.346e-05
        [j_node_and_user_rematch]: 8.89e-06
        [meta_fg_expand]: 2.01998e-06
        [replace_old_param]: 8.63001e-06
        [inline_without_move]: 5.43002e-06
        [renormalize]: 0.00050999
        [add_forward_monad_depend]: 5.40001e-06
        [auto_monad_grad]: 2.19001e-06
        [auto_monad_eliminator]: 1.241e-05
        [cse]: 1.889e-05
        [replace_applicator]: 1.022e-05
    [Cycle 2]: 0.00029981, [27]
        [switch_simplify]: 5.16998e-06
        [loop_unroll]: 4.55999e-06
        [a_1]: 8.755e-05
        [with_stream_mark]: 7.87e-06
        [recompute_prepare]: 4.58999e-06
        [updatestate_depend_eliminate]: 2.39001e-06
        [updatestate_assign_eliminate]: 1.99999e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 9.00007e-07
        [specialize_transform]: 4.58001e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 4.60001e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.3e-06
        [merge_forward]: 2.22001e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.173e-05
        [j_node_and_user_rematch]: 7.21001e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 6.79999e-06
        [inline_without_move]: 4.57e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.21997e-06
        [auto_monad_grad]: 6.19999e-07
        [auto_monad_eliminator]: 4.72e-06
        [cse]: 9.34e-06
        [replace_applicator]: 4.61002e-06
[py_interpret_to_execute_after_opt_a]: 8.44998e-06
[rewriter_after_opt_a]: 3.066e-05
[convert_after_rewriter]: 6.73e-06
[order_py_execute_after_rewriter]: 4.78001e-06
[mutable_eliminate]: 0.00055618
[jit_opt_b]: 5.339e-05, [1]
    [Cycle 1]: 4.085e-05, [2]
        [frontend_op_eliminate]: 1.591e-05
        [inline_after_opt_a]: 1.401e-05
[cconv]: 2.498e-05
[loop_unroll]: 0.00044804
[jit_opt_after_cconv]: 0.00013323, [1]
    [Cycle 1]: 0.00012701, [11]
        [c_1]: 1.97e-05
        [parameter_eliminate]: 2.39001e-06
        [updatestate_depend_eliminate]: 5.14e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.647e-05
        [call_graph_tuple_transform]: 1.786e-05
        [tuple_list_get_item_eliminator]: 5.06997e-06
        [none_parameter_eliminate]: 1.46998e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 5.14e-06
[remove_dup_value]: 1.323e-05
[partial_unused_args_eliminate]: 2.15002e-06
[environ_conv]: 4.69002e-06
[add_recomputation]: 4.568e-05
[cse_after_recomputation]: 2.153e-05, [1]
    [Cycle 1]: 1.583e-05, [1]
        [cse]: 9.46e-06
[auto_monad_reorder]: 1.398e-05
[get_jit_bprop_graph]: 1.69998e-06
[rewriter_after_jit_bprop_graph]: 3.78999e-06
[opt_after_jit_grad]: 0.00044106
[symbol_engine_optimizer]: 6.776e-05, [1]
    [Cycle 1]: 6.196e-05, [6]
        [build]: 3.01001e-06
        [elim_shapecalc]: 7.13e-06
        [elim_not_effective]: 1.176e-05
        [opt_reshape]: 5.37001e-06
        [fold_const_symbol]: 8.01001e-06
        [renormalize]: 4.39992e-07
[validate]: 2.977e-05
[backend_pass]: 8.40024e-07
[task_emit]: 0.00536945
[execute]: 7.12002e-06
Sums
    bootstrap                            :     0.000296s :  2.23%
    type_inference                       :     0.004232s : 31.82%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000061s :  0.46%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000027s :  0.20%
    rewriter_before_opt_a                :     0.000068s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000051s :  0.38%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000584s :  4.39%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000510s :  3.84%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000028s :  0.21%
    jit_opt_a.replace_applicator         :     0.000015s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000556s :  4.18%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000025s :  0.19%
    loop_unroll                          :     0.000448s :  3.37%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000046s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000441s :  3.32%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000030s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005369s : 40.38%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000175    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000001s :      2: substitution.fold_const_symbol
 3.14% :     0.000005s :      3: substitution.graph_param_transform
81.79% :     0.000143s :      4: substitution.inline
 1.61% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.57% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000003s :      2: substitution.replace_old_param
 7.08% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004172     2
82.38% :     0.003437s :      1: type_inference.infer
17.62% :     0.000735s :      1: type_inference.specialize
------[replace.]   0.000059     6
66.71% :     0.000039s :      4: replace.inline
33.29% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000152     6
92.57% :     0.000140s :      4: match.inline
 7.43% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.60% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.12% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.24% :     0.000003s :     10: predicate.arithmetic_simplify
 1.19% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.60% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.06% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.02% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.04% :     0.000001s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.02% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.31% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.42% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     15: predicate.partial_eliminate
 1.73% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.92% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.51% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.28% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.66% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.54% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.61% :     0.000003s :     16: predicate.switch_defer_inline
 1.99% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.11% :     0.000008s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.48% :     0.000002s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.57% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.66% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.89% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.65% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000485    12
47.09% :     0.000228s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.89% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.02% :     0.000228s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018675    76
 0.26% :     0.000049s :      1: add_recomputation
 0.34% :     0.000064s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.63% :     0.000305s :      1: bootstrap
 0.15% :     0.000028s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
27.41% :     0.005119s :      1: jit_opt_a
 0.73% :     0.000136s :      1: jit_opt_after_cconv
 0.30% :     0.000056s :      1: jit_opt_b
 2.44% :     0.000456s :      1: loop_unroll
 3.02% :     0.000565s :      1: mutable_eliminate
 4.16% :     0.000776s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.40% :     0.000449s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000029s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.44% :     0.000269s :      1: renormalize.infer
 1.25% :     0.000233s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.38% :     0.000070s :      1: rewriter_before_opt_a
 0.38% :     0.000070s :      1: symbol_engine_optimizer
28.81% :     0.005379s :      1: task_emit
22.71% :     0.004242s :      1: type_inference
 0.26% :     0.000048s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.566.581 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6c072040> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.567.198 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0128487, [30]
[bootstrap]: 0.00030552
[type_inference]: 0.0043198
[event_method]: 1.717e-05
[auto_monad]: 6.352e-05
[graph_reusing]: 4.84e-06
[pre_auto_parallel]: 2.20002e-06
[py_interpret_to_execute]: 2.805e-05
[rewriter_before_opt_a]: 6.666e-05
[expand_dump_flag]: 2.97002e-06
[jit_opt_a]: 0.00576528, [2]
    [Cycle 1]: 0.0017823, [27]
        [switch_simplify]: 5.721e-05
        [loop_unroll]: 3.047e-05
        [a_1]: 0.00058524
        [with_stream_mark]: 1.963e-05
        [recompute_prepare]: 1.018e-05
        [updatestate_depend_eliminate]: 3.33e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 2.13002e-06
        [specialize_transform]: 6.59999e-06
        [updatestate_useless_node_eliminater]: 5.39e-06
        [accelerated_algorithm]: 6.48998e-06
        [meta_shard_fg_expand]: 2.34999e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 3.58999e-06
        [cell_reuse_recompute_pass]: 2.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.599e-05
        [j_node_and_user_rematch]: 9.32001e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 9.31e-06
        [inline_without_move]: 5.97999e-06
        [renormalize]: 0.00074824
        [add_forward_monad_depend]: 7.03e-06
        [auto_monad_grad]: 3.03e-06
        [auto_monad_eliminator]: 1.46e-05
        [cse]: 2.543e-05
        [replace_applicator]: 1.562e-05
    [Cycle 2]: 0.00034068, [27]
        [switch_simplify]: 6.11998e-06
        [loop_unroll]: 5.09e-06
        [a_1]: 8.966e-05
        [with_stream_mark]: 1.22e-05
        [recompute_prepare]: 5.89999e-06
        [updatestate_depend_eliminate]: 3.09001e-06
        [updatestate_assign_eliminate]: 2.14999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.42999e-06
        [specialize_transform]: 5.39e-06
        [updatestate_useless_node_eliminater]: 5.48002e-06
        [accelerated_algorithm]: 5.27001e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 4.84e-06
        [merge_forward]: 3.04999e-06
        [cell_reuse_recompute_pass]: 2.21998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.514e-05
        [j_node_and_user_rematch]: 9.86998e-06
        [meta_fg_expand]: 1.69998e-06
        [replace_old_param]: 7.65e-06
        [inline_without_move]: 4.90001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.35999e-06
        [auto_monad_grad]: 1.27e-06
        [auto_monad_eliminator]: 5.87001e-06
        [cse]: 1.118e-05
        [replace_applicator]: 6.14001e-06
[py_interpret_to_execute_after_opt_a]: 1.218e-05
[rewriter_after_opt_a]: 3.478e-05
[convert_after_rewriter]: 8.15999e-06
[order_py_execute_after_rewriter]: 5.30999e-06
[mutable_eliminate]: 0.00064335
[jit_opt_b]: 5.182e-05, [1]
    [Cycle 1]: 4.47e-05, [2]
        [frontend_op_eliminate]: 1.745e-05
        [inline_after_opt_a]: 1.542e-05
[cconv]: 5.846e-05
[loop_unroll]: 0.00045227
[jit_opt_after_cconv]: 0.00014344, [1]
    [Cycle 1]: 0.00013657, [11]
        [c_1]: 2.098e-05
        [parameter_eliminate]: 3.95e-06
        [updatestate_depend_eliminate]: 6.14001e-06
        [updatestate_assign_eliminate]: 3.14001e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [cse]: 1.889e-05
        [call_graph_tuple_transform]: 1.834e-05
        [tuple_list_get_item_eliminator]: 5.25999e-06
        [none_parameter_eliminate]: 1.94999e-06
        [renormalize]: 5.29981e-07
        [switch_simplify]: 5.25001e-06
[remove_dup_value]: 1.484e-05
[partial_unused_args_eliminate]: 2.46e-06
[environ_conv]: 4.58001e-06
[add_recomputation]: 4.829e-05
[cse_after_recomputation]: 1.976e-05, [1]
    [Cycle 1]: 1.443e-05, [1]
        [cse]: 8.72e-06
[auto_monad_reorder]: 1.559e-05
[get_jit_bprop_graph]: 2.29999e-06
[rewriter_after_jit_bprop_graph]: 6.52001e-06
[opt_after_jit_grad]: 0.00047363
[symbol_engine_optimizer]: 7.209e-05, [1]
    [Cycle 1]: 6.57e-05, [6]
        [build]: 3.28e-06
        [elim_shapecalc]: 7.51999e-06
        [elim_not_effective]: 1.258e-05
        [opt_reshape]: 5.51e-06
        [fold_const_symbol]: 8.45999e-06
        [renormalize]: 7.09988e-07
[validate]: 3.857e-05
Sums
    bootstrap                            :     0.000306s :  3.54%
    type_inference                       :     0.004320s : 50.11%
    event_method                         :     0.000017s :  0.20%
    auto_monad                           :     0.000064s :  0.74%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000028s :  0.33%
    rewriter_before_opt_a                :     0.000067s :  0.77%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000063s :  0.73%
    jit_opt_a.loop_unroll                :     0.000036s :  0.41%
    jit_opt_a.a_1                        :     0.000675s :  7.83%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.37%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.19%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.36%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.20%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000748s :  8.68%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.24%
    jit_opt_a.cse                        :     0.000037s :  0.42%
    jit_opt_a.replace_applicator         :     0.000022s :  0.25%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.14%
    rewriter_after_opt_a                 :     0.000035s :  0.40%
    convert_after_rewriter               :     0.000008s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000643s :  7.46%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000058s :  0.68%
    loop_unroll                          :     0.000452s :  5.25%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000048s :  0.56%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.08%
    opt_after_jit_grad                   :     0.000474s :  5.49%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.45%


Time group info:
------[substitution.]   0.000241    23
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.04% :     0.000005s :      3: substitution.graph_param_transform
82.43% :     0.000198s :      4: substitution.inline
 1.87% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.09% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.98% :     0.000005s :      2: substitution.replace_old_param
 8.31% :     0.000020s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004260     2
81.10% :     0.003455s :      1: type_inference.infer
18.90% :     0.000805s :      1: type_inference.specialize
------[replace.]   0.000076     6
67.19% :     0.000051s :      4: replace.inline
32.81% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000214     6
91.28% :     0.000196s :      4: match.inline
 8.72% :     0.000019s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.25% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.12% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.66% :     0.000002s :     10: predicate.addn_zero_filter
 2.28% :     0.000003s :     10: predicate.arithmetic_simplify
 1.11% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.56% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.40% :     0.000000s :      3: predicate.graph_param_transform
 5.61% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :      6: predicate.less_batch_normalization
 1.87% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.21% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.11% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.77% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 1.73% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.07% :     0.000002s :     15: predicate.partial_eliminate
 1.06% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.67% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.14% :     0.000003s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.28% :     0.000002s :     10: predicate.reshape_eliminate
 1.46% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.26% :     0.000002s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.12% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.29% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.59% :     0.000010s :     53: predicate.switch_simplify
 1.08% :     0.000001s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.56% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.02% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.04% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000619    12
49.23% :     0.000305s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.72% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.05% :     0.000266s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014559    72
 0.35% :     0.000051s :      1: add_recomputation
 0.46% :     0.000067s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 2.20% :     0.000321s :      1: bootstrap
 0.43% :     0.000062s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.14% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
39.62% :     0.005769s :      1: jit_opt_a
 1.00% :     0.000146s :      1: jit_opt_after_cconv
 0.37% :     0.000054s :      1: jit_opt_b
 3.16% :     0.000460s :      1: loop_unroll
 4.48% :     0.000652s :      1: mutable_eliminate
 6.22% :     0.000905s :     26: opt.transform.jit_opt_a
 0.32% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000027s :      4: opt.transform.jit_opt_b
 0.09% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.31% :     0.000482s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.21% :     0.000031s :      1: py_interpret_to_execute
 0.10% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.73% :     0.000398s :      1: renormalize.infer
 2.34% :     0.000341s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000038s :      1: rewriter_after_opt_a
 0.48% :     0.000069s :      1: rewriter_before_opt_a
 0.51% :     0.000075s :      1: symbol_engine_optimizer
29.75% :     0.004331s :      1: type_inference


TotalTime = 0.0222989, [33]
[bootstrap]: 0.00032803
[type_inference]: 0.00499242
[event_method]: 1.982e-05
[auto_monad]: 6.752e-05
[graph_reusing]: 5.84e-06
[pre_auto_parallel]: 2.78e-06
[py_interpret_to_execute]: 4.373e-05
[rewriter_before_opt_a]: 7.707e-05
[expand_dump_flag]: 2.98e-06
[jit_opt_a]: 0.00648794, [2]
    [Cycle 1]: 0.00171986, [27]
        [switch_simplify]: 5.97e-05
        [loop_unroll]: 2.853e-05
        [a_1]: 0.00056398
        [with_stream_mark]: 2.187e-05
        [recompute_prepare]: 8.13001e-06
        [updatestate_depend_eliminate]: 3.88999e-06
        [updatestate_assign_eliminate]: 3.31001e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [parameter_eliminate]: 2.24001e-06
        [specialize_transform]: 6.37001e-06
        [updatestate_useless_node_eliminater]: 5.39e-06
        [accelerated_algorithm]: 5.99999e-06
        [meta_shard_fg_expand]: 2.19001e-06
        [get_grad_eliminate_]: 5.45001e-06
        [merge_forward]: 3.69002e-06
        [cell_reuse_recompute_pass]: 1.41002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.534e-05
        [j_node_and_user_rematch]: 9.09e-06
        [meta_fg_expand]: 2.47001e-06
        [replace_old_param]: 9.13002e-06
        [inline_without_move]: 5.96003e-06
        [renormalize]: 0.00070905
        [add_forward_monad_depend]: 4.91002e-06
        [auto_monad_grad]: 2.31e-06
        [auto_monad_eliminator]: 1.408e-05
        [cse]: 2.939e-05
        [replace_applicator]: 1.333e-05
    [Cycle 2]: 0.00121228, [27]
        [switch_simplify]: 5.60001e-06
        [loop_unroll]: 5.17999e-06
        [a_1]: 9.066e-05
        [with_stream_mark]: 9.89001e-06
        [recompute_prepare]: 4.84e-06
        [updatestate_depend_eliminate]: 2.74001e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 5.07e-06
        [updatestate_useless_node_eliminater]: 4.62e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 1.58002e-06
        [get_grad_eliminate_]: 4.42998e-06
        [merge_forward]: 2.61e-06
        [cell_reuse_recompute_pass]: 1.54e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.249e-05
        [j_node_and_user_rematch]: 8.11002e-06
        [meta_fg_expand]: 1.54998e-06
        [replace_old_param]: 7.51001e-06
        [inline_without_move]: 4.53999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 1.17e-06
        [auto_monad_eliminator]: 5.35999e-06
        [cse]: 9.88002e-06
        [replace_applicator]: 0.00088753
[py_interpret_to_execute_after_opt_a]: 1.369e-05
[rewriter_after_opt_a]: 3.976e-05
[convert_after_rewriter]: 7.38e-06
[order_py_execute_after_rewriter]: 5.24998e-06
[mutable_eliminate]: 0.00063606
[jit_opt_b]: 5.08e-05, [1]
    [Cycle 1]: 4.28e-05, [2]
        [frontend_op_eliminate]: 1.577e-05
        [inline_after_opt_a]: 1.539e-05
[cconv]: 2.662e-05
[loop_unroll]: 0.00045707
[jit_opt_after_cconv]: 0.00016011, [1]
    [Cycle 1]: 0.0001534, [11]
        [c_1]: 2.205e-05
        [parameter_eliminate]: 4.14002e-06
        [updatestate_depend_eliminate]: 7.42002e-06
        [updatestate_assign_eliminate]: 2.62001e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [cse]: 2.834e-05
        [call_graph_tuple_transform]: 2.155e-05
        [tuple_list_get_item_eliminator]: 5.37999e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 8.99978e-07
        [switch_simplify]: 5.44e-06
[remove_dup_value]: 1.538e-05
[partial_unused_args_eliminate]: 2.26998e-06
[environ_conv]: 5.20999e-06
[add_recomputation]: 5.242e-05
[cse_after_recomputation]: 2.149e-05, [1]
    [Cycle 1]: 1.556e-05, [1]
        [cse]: 9.47001e-06
[auto_monad_reorder]: 0.00017017
[get_jit_bprop_graph]: 2.36e-06
[rewriter_after_jit_bprop_graph]: 7.31999e-06
[opt_after_jit_grad]: 0.00052419
[symbol_engine_optimizer]: 8.669e-05, [1]
    [Cycle 1]: 7.832e-05, [6]
        [build]: 5.14998e-06
        [elim_shapecalc]: 9.22999e-06
        [elim_not_effective]: 1.584e-05
        [opt_reshape]: 7.23e-06
        [fold_const_symbol]: 9.91e-06
        [renormalize]: 7.7e-07
[validate]: 4.396e-05
[backend_pass]: 8.10018e-07
[task_emit]: 0.00770228
[execute]: 7.97e-06
Sums
    bootstrap                            :     0.000328s :  1.81%
    type_inference                       :     0.004992s : 27.61%
    event_method                         :     0.000020s :  0.11%
    auto_monad                           :     0.000068s :  0.37%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000044s :  0.24%
    rewriter_before_opt_a                :     0.000077s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.36%
    jit_opt_a.loop_unroll                :     0.000034s :  0.19%
    jit_opt_a.a_1                        :     0.000655s :  3.62%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.000709s :  3.92%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.11%
    jit_opt_a.cse                        :     0.000039s :  0.22%
    jit_opt_a.replace_applicator         :     0.000901s :  4.98%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000040s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000636s :  3.52%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000027s :  0.15%
    loop_unroll                          :     0.000457s :  2.53%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000052s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000170s :  0.94%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000524s :  2.90%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007702s : 42.60%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000208    23
 1.13% :     0.000002s :      2: substitution.elim_not_effective
 1.03% :     0.000002s :      2: substitution.fold_const_symbol
 2.88% :     0.000006s :      3: substitution.graph_param_transform
83.32% :     0.000173s :      4: substitution.inline
 1.70% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.11% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000004s :      2: substitution.replace_old_param
 5.95% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004920     2
80.41% :     0.003956s :      1: type_inference.infer
19.59% :     0.000964s :      1: type_inference.specialize
------[replace.]   0.000071     6
66.63% :     0.000047s :      4: replace.inline
33.37% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000182     6
93.79% :     0.000171s :      4: match.inline
 6.21% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   691
 1.06% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.01% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 1.78% :     0.000002s :     10: predicate.arithmetic_simplify
 1.18% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.11% :     0.000002s :     10: predicate.depend_value_elim
 0.94% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.11% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.95% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.03% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.01% :     0.000001s :     10: predicate.environ_get_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.08% :     0.000007s :     22: predicate.inline
 0.76% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.67% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.14% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.37% :     0.000005s :     34: predicate.loop_unroll_before_grad
14.60% :     0.000020s :     13: predicate.make_slice_get_slice_eliminator
 1.28% :     0.000002s :     10: predicate.merge_addn
 0.93% :     0.000001s :     10: predicate.minmaximum_grad
 1.21% :     0.000002s :      3: predicate.mutable_eliminate
 0.36% :     0.000000s :      3: predicate.opt_reshape
 1.97% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.26% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.19% :     0.000002s :     10: predicate.reshape_eliminate
 1.05% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.62% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.03% :     0.000001s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.23% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.08% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.13% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.82% :     0.000009s :     53: predicate.switch_simplify
 1.00% :     0.000001s :     10: predicate.tile_eliminate
 1.07% :     0.000001s :     10: predicate.transpose_eliminate
 1.14% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.60% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.31% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.48% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.31% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.52% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000636    12
49.50% :     0.000315s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.03% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.47% :     0.000289s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024866    76
 0.22% :     0.000056s :      1: add_recomputation
 0.29% :     0.000071s :      1: auto_monad
 0.71% :     0.000176s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.41% :     0.000350s :      1: bootstrap
 0.12% :     0.000029s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.11% :     0.006493s :      1: jit_opt_a
 0.66% :     0.000163s :      1: jit_opt_after_cconv
 0.22% :     0.000053s :      1: jit_opt_b
 1.87% :     0.000465s :      1: loop_unroll
 2.59% :     0.000644s :      1: mutable_eliminate
 7.05% :     0.001754s :     26: opt.transform.jit_opt_a
 0.20% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000038s :      4: opt.transform.symbol_engine_opt
 2.15% :     0.000534s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000047s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 1.46% :     0.000363s :      1: renormalize.infer
 1.36% :     0.000338s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000043s :      1: rewriter_after_opt_a
 0.32% :     0.000080s :      1: rewriter_before_opt_a
 0.36% :     0.000089s :      1: symbol_engine_optimizer
31.03% :     0.007717s :      1: task_emit
20.15% :     0.005011s :      1: type_inference
 0.27% :     0.000068s :      1: validate


TotalTime = 0.0174427, [33]
[bootstrap]: 0.00029797
[type_inference]: 0.00432729
[event_method]: 1.492e-05
[auto_monad]: 4.78e-05
[graph_reusing]: 3.99002e-06
[pre_auto_parallel]: 2.06998e-06
[py_interpret_to_execute]: 2.624e-05
[rewriter_before_opt_a]: 6.291e-05
[expand_dump_flag]: 2.39999e-06
[jit_opt_a]: 0.0049242, [2]
    [Cycle 1]: 0.00125544, [27]
        [switch_simplify]: 4.349e-05
        [loop_unroll]: 2.86e-05
        [a_1]: 0.00045994
        [with_stream_mark]: 1.151e-05
        [recompute_prepare]: 6.34999e-06
        [updatestate_depend_eliminate]: 3.00002e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.01998e-06
        [parameter_eliminate]: 1.32999e-06
        [specialize_transform]: 5.82001e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 5.16002e-06
        [merge_forward]: 2.91e-06
        [cell_reuse_recompute_pass]: 7.90023e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.152e-05
        [j_node_and_user_rematch]: 8.48001e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 8.08999e-06
        [inline_without_move]: 4.99e-06
        [renormalize]: 0.00044911
        [add_forward_monad_depend]: 3.69002e-06
        [auto_monad_grad]: 1.15001e-06
        [auto_monad_eliminator]: 8.87e-06
        [cse]: 1.48e-05
        [replace_applicator]: 9.16998e-06
    [Cycle 2]: 0.00029926, [27]
        [switch_simplify]: 5.31002e-06
        [loop_unroll]: 4.48001e-06
        [a_1]: 8.676e-05
        [with_stream_mark]: 6.36e-06
        [recompute_prepare]: 4.52e-06
        [updatestate_depend_eliminate]: 2.05002e-06
        [updatestate_assign_eliminate]: 1.84998e-06
        [updatestate_loads_eliminate]: 1.72001e-06
        [parameter_eliminate]: 7.10017e-07
        [specialize_transform]: 4.62998e-06
        [updatestate_useless_node_eliminater]: 4.55001e-06
        [accelerated_algorithm]: 4.79e-06
        [meta_shard_fg_expand]: 9.30013e-07
        [get_grad_eliminate_]: 4.34002e-06
        [merge_forward]: 2.11e-06
        [cell_reuse_recompute_pass]: 1.32999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.038e-05
        [j_node_and_user_rematch]: 7.43999e-06
        [meta_fg_expand]: 1.40999e-06
        [replace_old_param]: 6.98998e-06
        [inline_without_move]: 4.66002e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 6.50005e-07
        [auto_monad_eliminator]: 5.00001e-06
        [cse]: 9.96998e-06
        [replace_applicator]: 4.85001e-06
[py_interpret_to_execute_after_opt_a]: 8.33999e-06
[rewriter_after_opt_a]: 2.981e-05
[convert_after_rewriter]: 6.56e-06
[order_py_execute_after_rewriter]: 5.00001e-06
[mutable_eliminate]: 0.00051119
[jit_opt_b]: 4.63e-05, [1]
    [Cycle 1]: 4.004e-05, [2]
        [frontend_op_eliminate]: 1.509e-05
        [inline_after_opt_a]: 1.401e-05
[cconv]: 1.952e-05
[loop_unroll]: 0.00041013
[jit_opt_after_cconv]: 0.00012505, [1]
    [Cycle 1]: 0.00011857, [11]
        [c_1]: 1.896e-05
        [parameter_eliminate]: 1.89e-06
        [updatestate_depend_eliminate]: 4.03001e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [cse]: 1.416e-05
        [call_graph_tuple_transform]: 1.594e-05
        [tuple_list_get_item_eliminator]: 4.96997e-06
        [none_parameter_eliminate]: 1.26997e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.18002e-06
[remove_dup_value]: 1.244e-05
[partial_unused_args_eliminate]: 2.04e-06
[environ_conv]: 4.52e-06
[add_recomputation]: 4.595e-05
[cse_after_recomputation]: 1.865e-05, [1]
    [Cycle 1]: 1.364e-05, [1]
        [cse]: 8.35999e-06
[auto_monad_reorder]: 1.464e-05
[get_jit_bprop_graph]: 1.75001e-06
[rewriter_after_jit_bprop_graph]: 4.43999e-06
[opt_after_jit_grad]: 0.00049812
[symbol_engine_optimizer]: 6.836e-05, [1]
    [Cycle 1]: 6.253e-05, [6]
        [build]: 3.06001e-06
        [elim_shapecalc]: 7.15998e-06
        [elim_not_effective]: 1.198e-05
        [opt_reshape]: 5.17e-06
        [fold_const_symbol]: 8.27003e-06
        [renormalize]: 5.00004e-07
[validate]: 3.066e-05
[backend_pass]: 8.99978e-07
[task_emit]: 0.00569231
[execute]: 7.82e-06
Sums
    bootstrap                            :     0.000298s :  2.20%
    type_inference                       :     0.004327s : 31.98%
    event_method                         :     0.000015s :  0.11%
    auto_monad                           :     0.000048s :  0.35%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.19%
    rewriter_before_opt_a                :     0.000063s :  0.46%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000049s :  0.36%
    jit_opt_a.loop_unroll                :     0.000033s :  0.24%
    jit_opt_a.a_1                        :     0.000547s :  4.04%
    jit_opt_a.with_stream_mark           :     0.000018s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000449s :  3.32%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.10%
    jit_opt_a.cse                        :     0.000025s :  0.18%
    jit_opt_a.replace_applicator         :     0.000014s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000030s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000511s :  3.78%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000020s :  0.14%
    loop_unroll                          :     0.000410s :  3.03%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000014s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000046s :  0.34%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000498s :  3.68%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000031s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005692s : 42.07%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000152    23
 1.37% :     0.000002s :      2: substitution.elim_not_effective
 0.81% :     0.000001s :      2: substitution.fold_const_symbol
 3.13% :     0.000005s :      3: substitution.graph_param_transform
84.01% :     0.000127s :      4: substitution.inline
 1.77% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.56% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.97% :     0.000003s :      2: substitution.replace_old_param
 4.39% :     0.000007s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004280     2
83.33% :     0.003566s :      1: type_inference.infer
16.67% :     0.000714s :      1: type_inference.specialize
------[replace.]   0.000055     6
66.88% :     0.000037s :      4: replace.inline
33.12% :     0.000018s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000131     6
95.69% :     0.000125s :      4: match.inline
 4.31% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.51% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000001s :     10: predicate.addn_zero_filter
 2.36% :     0.000003s :     10: predicate.arithmetic_simplify
 1.30% :     0.000001s :     10: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.19% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000001s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.42% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.38% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.17% :     0.000006s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.40% :     0.000002s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.13% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.26% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.29% :     0.000001s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.37% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.54% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.43% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.31% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.74% :     0.000009s :     53: predicate.switch_simplify
 1.37% :     0.000002s :     10: predicate.tile_eliminate
 1.41% :     0.000002s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.19% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000485    12
49.35% :     0.000239s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.38% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.26% :     0.000219s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018722    76
 0.26% :     0.000049s :      1: add_recomputation
 0.27% :     0.000051s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.67% :     0.000312s :      1: bootstrap
 0.12% :     0.000022s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000018s :      1: event_method
 0.07% :     0.000012s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
26.31% :     0.004927s :      1: jit_opt_a
 0.68% :     0.000128s :      1: jit_opt_after_cconv
 0.26% :     0.000049s :      1: jit_opt_b
 2.22% :     0.000416s :      1: loop_unroll
 2.77% :     0.000518s :      1: mutable_eliminate
 3.91% :     0.000732s :     26: opt.transform.jit_opt_a
 0.22% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000010s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.69% :     0.000504s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000029s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.23% :     0.000231s :      1: renormalize.infer
 1.14% :     0.000213s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000032s :      1: rewriter_after_opt_a
 0.35% :     0.000066s :      1: rewriter_before_opt_a
 0.38% :     0.000071s :      1: symbol_engine_optimizer
30.47% :     0.005706s :      1: task_emit
23.17% :     0.004337s :      1: type_inference
 0.26% :     0.000048s :      1: validate


TotalTime = 0.0330205, [33]
[bootstrap]: 0.00025549
[type_inference]: 0.0130987
[event_method]: 7.179e-05
[auto_monad]: 0.00011266
[graph_reusing]: 7.26999e-06
[pre_auto_parallel]: 3.48e-06
[py_interpret_to_execute]: 2.989e-05
[rewriter_before_opt_a]: 0.00010963
[expand_dump_flag]: 3.33998e-06
[jit_opt_a]: 0.0114373, [3]
    [Cycle 1]: 0.0057845, [27]
        [switch_simplify]: 0.00012216
        [loop_unroll]: 4.34e-05
        [a_1]: 0.00092044
        [with_stream_mark]: 2.428e-05
        [recompute_prepare]: 1.879e-05
        [updatestate_depend_eliminate]: 8.48001e-06
        [updatestate_assign_eliminate]: 6.82002e-06
        [updatestate_loads_eliminate]: 6.19001e-06
        [parameter_eliminate]: 3.16999e-06
        [specialize_transform]: 1.36e-05
        [updatestate_useless_node_eliminater]: 1.22e-05
        [accelerated_algorithm]: 1.256e-05
        [meta_shard_fg_expand]: 3.8e-06
        [get_grad_eliminate_]: 1.229e-05
        [merge_forward]: 7.82e-06
        [cell_reuse_recompute_pass]: 9.50007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.627e-05
        [j_node_and_user_rematch]: 2.354e-05
        [meta_fg_expand]: 0.00139647
        [replace_old_param]: 5.144e-05
        [inline_without_move]: 5.251e-05
        [renormalize]: 0.002532
        [add_forward_monad_depend]: 9.71e-06
        [auto_monad_grad]: 5.31002e-06
        [auto_monad_eliminator]: 4.699e-05
        [cse]: 0.00015615
        [replace_applicator]: 6.109e-05
    [Cycle 2]: 0.00191822, [27]
        [switch_simplify]: 3.876e-05
        [loop_unroll]: 3.721e-05
        [a_1]: 0.00099314
        [with_stream_mark]: 9.74e-06
        [recompute_prepare]: 7.48999e-06
        [updatestate_depend_eliminate]: 3.53e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.60002e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 6.68998e-06
        [updatestate_useless_node_eliminater]: 6.33e-06
        [accelerated_algorithm]: 6.05002e-06
        [meta_shard_fg_expand]: 1.58002e-06
        [get_grad_eliminate_]: 5.97999e-06
        [merge_forward]: 3.08998e-06
        [cell_reuse_recompute_pass]: 8.70001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.214e-05
        [j_node_and_user_rematch]: 9.61e-06
        [meta_fg_expand]: 8.226e-05
        [replace_old_param]: 1.025e-05
        [inline_without_move]: 6.05002e-06
        [renormalize]: 0.00048181
        [add_forward_monad_depend]: 4.39002e-06
        [auto_monad_grad]: 1.28002e-06
        [auto_monad_eliminator]: 1.087e-05
        [cse]: 1.927e-05
        [replace_applicator]: 1.117e-05
    [Cycle 3]: 0.00036466, [27]
        [switch_simplify]: 6.76999e-06
        [loop_unroll]: 5.91e-06
        [a_1]: 0.00011742
        [with_stream_mark]: 9.02999e-06
        [recompute_prepare]: 5.90002e-06
        [updatestate_depend_eliminate]: 3.41001e-06
        [updatestate_assign_eliminate]: 2.74999e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 6.29999e-06
        [updatestate_useless_node_eliminater]: 5.86e-06
        [accelerated_algorithm]: 5.86998e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 6.49999e-06
        [merge_forward]: 2.86e-06
        [cell_reuse_recompute_pass]: 1.42e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.413e-05
        [j_node_and_user_rematch]: 1.003e-05
        [meta_fg_expand]: 1.94e-06
        [replace_old_param]: 7.99002e-06
        [inline_without_move]: 5.72999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 6.83e-06
        [cse]: 1.408e-05
        [replace_applicator]: 5.85002e-06
[py_interpret_to_execute_after_opt_a]: 9.67001e-06
[rewriter_after_opt_a]: 3.176e-05
[convert_after_rewriter]: 6.59001e-06
[order_py_execute_after_rewriter]: 5.47999e-06
[mutable_eliminate]: 0.00050167
[jit_opt_b]: 5.342e-05, [1]
    [Cycle 1]: 4.776e-05, [2]
        [frontend_op_eliminate]: 1.909e-05
        [inline_after_opt_a]: 1.768e-05
[cconv]: 7.484e-05
[loop_unroll]: 0.00040152
[jit_opt_after_cconv]: 0.00014272, [1]
    [Cycle 1]: 0.00013688, [11]
        [c_1]: 2.456e-05
        [parameter_eliminate]: 1.89e-06
        [updatestate_depend_eliminate]: 4.61002e-06
        [updatestate_assign_eliminate]: 2.79999e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [cse]: 2.085e-05
        [call_graph_tuple_transform]: 1.778e-05
        [tuple_list_get_item_eliminator]: 6.43e-06
        [none_parameter_eliminate]: 7.90023e-07
        [renormalize]: 4.39992e-07
        [switch_simplify]: 6.71e-06
[remove_dup_value]: 9.87001e-06
[partial_unused_args_eliminate]: 1.40001e-06
[environ_conv]: 3.45998e-06
[add_recomputation]: 3.827e-05
[cse_after_recomputation]: 2.402e-05, [1]
    [Cycle 1]: 1.907e-05, [1]
        [cse]: 1.326e-05
[auto_monad_reorder]: 1.35e-05
[get_jit_bprop_graph]: 1.38002e-06
[rewriter_after_jit_bprop_graph]: 4.64998e-06
[opt_after_jit_grad]: 0.0004632
[symbol_engine_optimizer]: 7.729e-05, [1]
    [Cycle 1]: 7.168e-05, [6]
        [build]: 2.69001e-06
        [elim_shapecalc]: 8.65001e-06
        [elim_not_effective]: 1.451e-05
        [opt_reshape]: 7.21999e-06
        [fold_const_symbol]: 1.052e-05
        [renormalize]: 4.00003e-07
[validate]: 2.917e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00576979
[execute]: 4.33001e-06
Sums
    bootstrap                            :     0.000255s :  0.89%
    type_inference                       :     0.013099s : 45.39%
    event_method                         :     0.000072s :  0.25%
    auto_monad                           :     0.000113s :  0.39%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000030s :  0.10%
    rewriter_before_opt_a                :     0.000110s :  0.38%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000168s :  0.58%
    jit_opt_a.loop_unroll                :     0.000087s :  0.30%
    jit_opt_a.a_1                        :     0.002031s :  7.04%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000027s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.09%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000053s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000043s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001481s :  5.13%
    jit_opt_a.replace_old_param          :     0.000070s :  0.24%
    jit_opt_a.inline_without_move        :     0.000064s :  0.22%
    jit_opt_a.renormalize                :     0.003014s : 10.44%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000065s :  0.22%
    jit_opt_a.cse                        :     0.000189s :  0.66%
    jit_opt_a.replace_applicator         :     0.000078s :  0.27%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.03%
    rewriter_after_opt_a                 :     0.000032s :  0.11%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000502s :  1.74%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000075s :  0.26%
    loop_unroll                          :     0.000402s :  1.39%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000010s :  0.03%
    partial_unused_args_eliminate        :     0.000001s :  0.00%
    environ_conv                         :     0.000003s :  0.01%
    add_recomputation                    :     0.000038s :  0.13%
    cse_after_recomputation.cse          :     0.000013s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000463s :  1.61%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.10%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005770s : 20.00%
    execute                              :     0.000004s :  0.02%


Time group info:
------[substitution.]   0.000469   113
 0.40% :     0.000002s :      3: substitution.elim_not_effective
 0.33% :     0.000002s :      3: substitution.fold_const_symbol
 0.90% :     0.000004s :      4: substitution.graph_param_transform
62.30% :     0.000292s :     14: substitution.inline
 3.60% :     0.000017s :      2: substitution.inline_without_move
 1.76% :     0.000008s :     14: substitution.j_node_and_user_rematch
 2.28% :     0.000011s :      7: substitution.minmaximum_grad
 1.30% :     0.000006s :      5: substitution.partial_eliminate
 2.17% :     0.000010s :     14: substitution.remove_not_recompute_node
 5.28% :     0.000025s :     10: substitution.replace_applicator
 1.49% :     0.000007s :      7: substitution.replace_old_param
 0.51% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.70% :     0.000008s :      1: substitution.switch_simplify
 4.47% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.06% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 8.44% :     0.000040s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.013003     2
88.64% :     0.011526s :      1: type_inference.infer
11.36% :     0.001478s :      1: type_inference.specialize
------[replace.]   0.000177    22
53.64% :     0.000095s :     14: replace.inline
17.97% :     0.000032s :      1: replace.switch_simplify
28.39% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000311    22
91.25% :     0.000284s :     14: match.inline
 2.37% :     0.000007s :      1: match.switch_simplify
 6.38% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000377  2634
 1.61% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.51% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.48% :     0.000006s :     44: predicate.addn_check_dump
 1.64% :     0.000006s :     44: predicate.addn_zero_filter
 2.10% :     0.000008s :     44: predicate.arithmetic_simplify
 1.49% :     0.000006s :     44: predicate.cast_eliminate
 0.18% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.42% :     0.000005s :     44: predicate.compare_switch_simplify
 1.58% :     0.000006s :     44: predicate.depend_value_elim
 1.43% :     0.000005s :     44: predicate.dict_get_item_const_eliminator
 1.53% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.49% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.29% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      4: predicate.elim_not_effective
 0.20% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.46% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.43% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.43% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.46% :     0.000006s :     44: predicate.environ_get_eliminate
 1.42% :     0.000005s :     44: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.96% :     0.000004s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.33% :     0.000016s :     73: predicate.inline
 1.76% :     0.000007s :     40: predicate.inline_without_move
 0.37% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.92% :     0.000003s :     19: predicate.less_batch_normalization
 1.78% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.91% :     0.000007s :     55: predicate.load_eliminater
 0.49% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.57% :     0.000013s :     91: predicate.loop_unroll_before_grad
 1.65% :     0.000006s :     48: predicate.make_slice_get_slice_eliminator
 1.44% :     0.000005s :     44: predicate.merge_addn
 1.47% :     0.000006s :     44: predicate.minmaximum_grad
 0.46% :     0.000002s :      4: predicate.mutable_eliminate
 0.22% :     0.000001s :      4: predicate.opt_reshape
 2.36% :     0.000009s :     55: predicate.partial_eliminate
 1.54% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.95% :     0.000007s :     44: predicate.reduce_eliminate
 1.86% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.42% :     0.000009s :     97: predicate.replace_applicator
 0.90% :     0.000003s :     40: predicate.replace_old_param
 0.10% :     0.000000s :      4: predicate.reset_defer_inline
 1.69% :     0.000006s :     44: predicate.reshape_eliminate
 1.55% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.30% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.66% :     0.000006s :     44: predicate.same_eliminate
 0.48% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.44% :     0.000002s :      8: predicate.special_op_eliminate
 0.88% :     0.000003s :     19: predicate.specialize_transform
 1.62% :     0.000006s :     44: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.89% :     0.000011s :     65: predicate.switch_defer_inline
 2.61% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.83% :     0.000026s :    162: predicate.switch_simplify
 1.48% :     0.000006s :     44: predicate.tile_eliminate
 1.45% :     0.000005s :     44: predicate.transpose_eliminate
 1.87% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.64% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 3.02% :     0.000011s :     59: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.85% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000007s :     44: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.23% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001788    32
62.96% :     0.001126s :     14: func_graph_cloner_run.FuncGraphClonerGraph
37.04% :     0.000662s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038826    91
 0.11% :     0.000041s :      1: add_recomputation
 0.31% :     0.000118s :      1: auto_monad
 0.04% :     0.000016s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.73% :     0.000284s :      1: bootstrap
 0.20% :     0.000078s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.07% :     0.000026s :      1: cse_after_recomputation
 0.01% :     0.000005s :      1: environ_conv
 0.20% :     0.000077s :      1: event_method
 0.02% :     0.000008s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
29.47% :     0.011441s :      1: jit_opt_a
 0.37% :     0.000145s :      1: jit_opt_after_cconv
 0.14% :     0.000056s :      1: jit_opt_b
 1.05% :     0.000408s :      1: loop_unroll
 1.31% :     0.000509s :      1: mutable_eliminate
 6.92% :     0.002685s :     39: opt.transform.jit_opt_a
 0.14% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000031s :      4: opt.transform.jit_opt_b
 0.03% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.21% :     0.000471s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000003s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000032s :      1: py_interpret_to_execute
 0.03% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000012s :      1: remove_dup_value
 4.26% :     0.001656s :      2: renormalize.infer
 3.46% :     0.001345s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000035s :      1: rewriter_after_opt_a
 0.29% :     0.000113s :      1: rewriter_before_opt_a
 0.21% :     0.000080s :      1: symbol_engine_optimizer
14.88% :     0.005779s :      1: task_emit
33.80% :     0.013122s :      1: type_inference
 0.12% :     0.000047s :      1: validate


TotalTime = 0.0227418, [33]
[bootstrap]: 0.00032702
[type_inference]: 0.0058768
[event_method]: 2.344e-05
[auto_monad]: 7.05e-05
[graph_reusing]: 5.98998e-06
[pre_auto_parallel]: 2.53003e-06
[py_interpret_to_execute]: 4.938e-05
[rewriter_before_opt_a]: 8.32e-05
[expand_dump_flag]: 3.18e-06
[jit_opt_a]: 0.00640039, [2]
    [Cycle 1]: 0.00216211, [27]
        [switch_simplify]: 7.04e-05
        [loop_unroll]: 3.298e-05
        [a_1]: 0.00073305
        [with_stream_mark]: 2.81e-05
        [recompute_prepare]: 1.116e-05
        [updatestate_depend_eliminate]: 4.86002e-06
        [updatestate_assign_eliminate]: 2.84999e-06
        [updatestate_loads_eliminate]: 2.63998e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 7.46999e-06
        [updatestate_useless_node_eliminater]: 5.96998e-06
        [accelerated_algorithm]: 6.28e-06
        [meta_shard_fg_expand]: 2.56998e-06
        [get_grad_eliminate_]: 5.91998e-06
        [merge_forward]: 3.56001e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.626e-05
        [j_node_and_user_rematch]: 1.051e-05
        [meta_fg_expand]: 3.54002e-06
        [replace_old_param]: 1.224e-05
        [inline_without_move]: 5.64e-06
        [renormalize]: 0.00088831
        [add_forward_monad_depend]: 8.60999e-06
        [auto_monad_grad]: 2.64001e-06
        [auto_monad_eliminator]: 1.877e-05
        [cse]: 3.62e-05
        [replace_applicator]: 1.978e-05
    [Cycle 2]: 0.00038923, [27]
        [switch_simplify]: 7.78999e-06
        [loop_unroll]: 5.56002e-06
        [a_1]: 9.954e-05
        [with_stream_mark]: 1.558e-05
        [recompute_prepare]: 6.09001e-06
        [updatestate_depend_eliminate]: 3.4e-06
        [updatestate_assign_eliminate]: 2.72001e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 5.64e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 6.02001e-06
        [meta_shard_fg_expand]: 2.21998e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 3.85e-06
        [cell_reuse_recompute_pass]: 2.36e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.728e-05
        [j_node_and_user_rematch]: 9.37001e-06
        [meta_fg_expand]: 1.70001e-06
        [replace_old_param]: 8.06001e-06
        [inline_without_move]: 1.656e-05
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 2.71e-06
        [auto_monad_grad]: 1.39e-06
        [auto_monad_eliminator]: 6.81999e-06
        [cse]: 1.36e-05
        [replace_applicator]: 6.63e-06
[py_interpret_to_execute_after_opt_a]: 1.589e-05
[rewriter_after_opt_a]: 4.258e-05
[convert_after_rewriter]: 7.15e-06
[order_py_execute_after_rewriter]: 5.56e-06
[mutable_eliminate]: 0.00076995
[jit_opt_b]: 5.804e-05, [1]
    [Cycle 1]: 4.873e-05, [2]
        [frontend_op_eliminate]: 1.875e-05
        [inline_after_opt_a]: 1.675e-05
[cconv]: 3.031e-05
[loop_unroll]: 0.00056849
[jit_opt_after_cconv]: 0.00017008, [1]
    [Cycle 1]: 0.00016207, [11]
        [c_1]: 2.357e-05
        [parameter_eliminate]: 4.3e-06
        [updatestate_depend_eliminate]: 9.29998e-06
        [updatestate_assign_eliminate]: 3.34001e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [cse]: 2.903e-05
        [call_graph_tuple_transform]: 2.403e-05
        [tuple_list_get_item_eliminator]: 5.84e-06
        [none_parameter_eliminate]: 1.43002e-06
        [renormalize]: 6.29982e-07
        [switch_simplify]: 6.06e-06
[remove_dup_value]: 1.596e-05
[partial_unused_args_eliminate]: 2.71999e-06
[environ_conv]: 6.56e-06
[add_recomputation]: 5.566e-05
[cse_after_recomputation]: 2.488e-05, [1]
    [Cycle 1]: 1.788e-05, [1]
        [cse]: 1.02e-05
[auto_monad_reorder]: 1.508e-05
[get_jit_bprop_graph]: 1.71002e-06
[rewriter_after_jit_bprop_graph]: 8.27998e-06
[opt_after_jit_grad]: 0.00089323
[symbol_engine_optimizer]: 9.278e-05, [1]
    [Cycle 1]: 8.287e-05, [6]
        [build]: 6.49001e-06
        [elim_shapecalc]: 9.94001e-06
        [elim_not_effective]: 1.829e-05
        [opt_reshape]: 6.51999e-06
        [fold_const_symbol]: 1.022e-05
        [renormalize]: 6.50005e-07
[validate]: 4.66e-05
[backend_pass]: 1.04e-06
[task_emit]: 0.00681081
[execute]: 7.97003e-06
Sums
    bootstrap                            :     0.000327s :  1.80%
    type_inference                       :     0.005877s : 32.36%
    event_method                         :     0.000023s :  0.13%
    auto_monad                           :     0.000070s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000049s :  0.27%
    rewriter_before_opt_a                :     0.000083s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000078s :  0.43%
    jit_opt_a.loop_unroll                :     0.000039s :  0.21%
    jit_opt_a.a_1                        :     0.000833s :  4.59%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000020s :  0.11%
    jit_opt_a.inline_without_move        :     0.000022s :  0.12%
    jit_opt_a.renormalize                :     0.000888s :  4.89%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.14%
    jit_opt_a.cse                        :     0.000050s :  0.27%
    jit_opt_a.replace_applicator         :     0.000026s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.09%
    rewriter_after_opt_a                 :     0.000043s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000770s :  4.24%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000030s :  0.17%
    loop_unroll                          :     0.000568s :  3.13%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000056s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000893s :  4.92%
    symbol_engine_optimizer.build        :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000047s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006811s : 37.51%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000337    23
 0.61% :     0.000002s :      2: substitution.elim_not_effective
 0.38% :     0.000001s :      2: substitution.fold_const_symbol
 1.92% :     0.000006s :      3: substitution.graph_param_transform
87.89% :     0.000296s :      4: substitution.inline
 1.31% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.61% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.97% :     0.000007s :      2: substitution.replace_old_param
 4.30% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005784     2
80.71% :     0.004668s :      1: type_inference.infer
19.29% :     0.001116s :      1: type_inference.specialize
------[replace.]   0.000097     6
65.23% :     0.000064s :      4: replace.inline
34.77% :     0.000034s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000306     6
95.63% :     0.000293s :      4: match.inline
 4.37% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000142   691
 1.02% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.64% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.86% :     0.000001s :     10: predicate.addn_check_dump
 1.11% :     0.000002s :     10: predicate.addn_zero_filter
 2.59% :     0.000004s :     10: predicate.arithmetic_simplify
 1.77% :     0.000003s :     10: predicate.cast_eliminate
 0.58% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.86% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000002s :     10: predicate.depend_value_elim
 0.88% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.51% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.35% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.17% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      3: predicate.elim_not_effective
 0.90% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.04% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.02% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 6.08% :     0.000009s :     22: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.71% :     0.000002s :      6: predicate.less_batch_normalization
 1.96% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.55% :     0.000002s :     15: predicate.load_eliminater
 1.74% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.88% :     0.000001s :     10: predicate.merge_addn
 0.88% :     0.000001s :     10: predicate.minmaximum_grad
 1.76% :     0.000003s :      3: predicate.mutable_eliminate
 0.36% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.35% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.97% :     0.000003s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 0.97% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.04% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.07% :     0.000002s :     10: predicate.same_eliminate
 0.42% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.16% :     0.000002s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.01% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 0.95% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.45% :     0.000005s :     16: predicate.switch_defer_inline
 1.94% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.77% :     0.000013s :     53: predicate.switch_simplify
 0.87% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000002s :     10: predicate.transpose_eliminate
 1.78% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.87% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.32% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.20% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.38% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.46% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000002s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.52% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000909    12
49.39% :     0.000449s :      5: func_graph_cloner_run.FuncGraphClonerGraph
11.80% :     0.000107s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.81% :     0.000353s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024858    76
 0.24% :     0.000059s :      1: add_recomputation
 0.30% :     0.000074s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.39% :     0.000346s :      1: bootstrap
 0.13% :     0.000033s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.11% :     0.000028s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
25.77% :     0.006405s :      1: jit_opt_a
 0.70% :     0.000173s :      1: jit_opt_after_cconv
 0.24% :     0.000061s :      1: jit_opt_b
 2.34% :     0.000582s :      1: loop_unroll
 3.15% :     0.000782s :      1: mutable_eliminate
 4.43% :     0.001100s :     26: opt.transform.jit_opt_a
 0.23% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000029s :      4: opt.transform.jit_opt_b
 0.07% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000041s :      4: opt.transform.symbol_engine_opt
 3.66% :     0.000910s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000052s :      1: py_interpret_to_execute
 0.08% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 1.82% :     0.000452s :      1: renormalize.infer
 1.71% :     0.000425s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000046s :      1: rewriter_after_opt_a
 0.35% :     0.000087s :      1: rewriter_before_opt_a
 0.38% :     0.000096s :      1: symbol_engine_optimizer
27.45% :     0.006824s :      1: task_emit
23.72% :     0.005897s :      1: type_inference
 0.28% :     0.000070s :      1: validate


TotalTime = 0.0147153, [30]
[bootstrap]: 0.00033731
[type_inference]: 0.005645
[event_method]: 1.793e-05
[auto_monad]: 7.042e-05
[graph_reusing]: 6.43003e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 3.701e-05
[rewriter_before_opt_a]: 7.195e-05
[expand_dump_flag]: 4.02e-06
[jit_opt_a]: 0.00594589, [2]
    [Cycle 1]: 0.00170972, [27]
        [switch_simplify]: 5.454e-05
        [loop_unroll]: 3.094e-05
        [a_1]: 0.00057266
        [with_stream_mark]: 2.01e-05
        [recompute_prepare]: 7.36001e-06
        [updatestate_depend_eliminate]: 3.21001e-06
        [updatestate_assign_eliminate]: 3.00002e-06
        [updatestate_loads_eliminate]: 2.83e-06
        [parameter_eliminate]: 1.78002e-06
        [specialize_transform]: 6.05002e-06
        [updatestate_useless_node_eliminater]: 4.97999e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 2.04999e-06
        [get_grad_eliminate_]: 5.14e-06
        [merge_forward]: 3.33998e-06
        [cell_reuse_recompute_pass]: 1.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.384e-05
        [j_node_and_user_rematch]: 8.83001e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 8.42e-06
        [inline_without_move]: 5.22e-06
        [renormalize]: 0.00071539
        [add_forward_monad_depend]: 5.76003e-06
        [auto_monad_grad]: 2.93998e-06
        [auto_monad_eliminator]: 1.38e-05
        [cse]: 2.639e-05
        [replace_applicator]: 1.163e-05
    [Cycle 2]: 0.00032424, [27]
        [switch_simplify]: 5.46e-06
        [loop_unroll]: 5.31002e-06
        [a_1]: 9.038e-05
        [with_stream_mark]: 1.163e-05
        [recompute_prepare]: 5.10001e-06
        [updatestate_depend_eliminate]: 2.69999e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 4.95001e-06
        [updatestate_useless_node_eliminater]: 5.27999e-06
        [accelerated_algorithm]: 4.85001e-06
        [meta_shard_fg_expand]: 1.36002e-06
        [get_grad_eliminate_]: 4.39998e-06
        [merge_forward]: 3.11001e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.312e-05
        [j_node_and_user_rematch]: 8.38001e-06
        [meta_fg_expand]: 1.58002e-06
        [replace_old_param]: 7.39002e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 4.94e-06
        [cse]: 1.025e-05
        [replace_applicator]: 5.19998e-06
[py_interpret_to_execute_after_opt_a]: 1.273e-05
[rewriter_after_opt_a]: 3.57e-05
[convert_after_rewriter]: 9.34998e-06
[order_py_execute_after_rewriter]: 4.89998e-06
[mutable_eliminate]: 0.00069396
[jit_opt_b]: 5.308e-05, [1]
    [Cycle 1]: 4.513e-05, [2]
        [frontend_op_eliminate]: 1.733e-05
        [inline_after_opt_a]: 1.587e-05
[cconv]: 3.125e-05
[loop_unroll]: 0.00056077
[jit_opt_after_cconv]: 0.00016342, [1]
    [Cycle 1]: 0.0001547, [11]
        [c_1]: 2.126e-05
        [parameter_eliminate]: 3.85998e-06
        [updatestate_depend_eliminate]: 8.23999e-06
        [updatestate_assign_eliminate]: 2.72001e-06
        [updatestate_loads_eliminate]: 2.79001e-06
        [cse]: 2.531e-05
        [call_graph_tuple_transform]: 2.291e-05
        [tuple_list_get_item_eliminator]: 5.30001e-06
        [none_parameter_eliminate]: 2.16998e-06
        [renormalize]: 1.05001e-06
        [switch_simplify]: 5.59e-06
[remove_dup_value]: 1.585e-05
[partial_unused_args_eliminate]: 2.35002e-06
[environ_conv]: 6.07001e-06
[add_recomputation]: 5.039e-05
[cse_after_recomputation]: 2.331e-05, [1]
    [Cycle 1]: 1.676e-05, [1]
        [cse]: 1.04e-05
[auto_monad_reorder]: 1.678e-05
[get_jit_bprop_graph]: 1.95001e-06
[rewriter_after_jit_bprop_graph]: 6.92002e-06
[opt_after_jit_grad]: 0.00058006
[symbol_engine_optimizer]: 7.971e-05, [1]
    [Cycle 1]: 7.298e-05, [6]
        [build]: 4.32e-06
        [elim_shapecalc]: 8.60001e-06
        [elim_not_effective]: 1.482e-05
        [opt_reshape]: 6.26998e-06
        [fold_const_symbol]: 9.14e-06
        [renormalize]: 6.60017e-07
[validate]: 3.902e-05
Sums
    bootstrap                            :     0.000337s :  3.31%
    type_inference                       :     0.005645s : 55.36%
    event_method                         :     0.000018s :  0.18%
    auto_monad                           :     0.000070s :  0.69%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.36%
    rewriter_before_opt_a                :     0.000072s :  0.71%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000060s :  0.59%
    jit_opt_a.loop_unroll                :     0.000036s :  0.36%
    jit_opt_a.a_1                        :     0.000663s :  6.50%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000715s :  7.02%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.18%
    jit_opt_a.cse                        :     0.000037s :  0.36%
    jit_opt_a.replace_applicator         :     0.000017s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.12%
    rewriter_after_opt_a                 :     0.000036s :  0.35%
    convert_after_rewriter               :     0.000009s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000694s :  6.81%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.16%
    cconv                                :     0.000031s :  0.31%
    loop_unroll                          :     0.000561s :  5.50%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000025s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000016s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000050s :  0.49%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000017s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000580s :  5.69%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.38%


Time group info:
------[substitution.]   0.000239    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.48% :     0.000006s :      3: substitution.graph_param_transform
86.26% :     0.000206s :      4: substitution.inline
 1.43% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.90% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.62% :     0.000004s :      2: substitution.replace_old_param
 4.92% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005568     2
82.83% :     0.004612s :      1: type_inference.infer
17.17% :     0.000956s :      1: type_inference.specialize
------[replace.]   0.000071     6
68.22% :     0.000048s :      4: replace.inline
31.78% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000214     6
95.01% :     0.000203s :      4: match.inline
 4.99% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.24% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 2.36% :     0.000003s :     10: predicate.arithmetic_simplify
 1.32% :     0.000002s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.66% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.16% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.40% :     0.000000s :      3: predicate.graph_param_transform
 6.25% :     0.000008s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.07% :     0.000001s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 2.19% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.45% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.14% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 2.03% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.07% :     0.000002s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.87% :     0.000002s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.50% :     0.000001s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 1.99% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.69% :     0.000009s :     53: predicate.switch_simplify
 1.07% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.58% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000694    12
48.94% :     0.000340s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.64% :     0.000046s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.41% :     0.000308s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016370    72
 0.33% :     0.000054s :      1: add_recomputation
 0.46% :     0.000075s :      1: auto_monad
 0.12% :     0.000020s :      1: auto_monad_reorder
 2.20% :     0.000361s :      1: bootstrap
 0.21% :     0.000035s :      1: cconv
 0.07% :     0.000012s :      1: convert_after_rewriter
 0.16% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.13% :     0.000022s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
36.34% :     0.005949s :      1: jit_opt_a
 1.02% :     0.000167s :      1: jit_opt_after_cconv
 0.34% :     0.000056s :      1: jit_opt_b
 3.50% :     0.000572s :      1: loop_unroll
 4.31% :     0.000705s :      1: mutable_eliminate
 5.32% :     0.000871s :     26: opt.transform.jit_opt_a
 0.31% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000026s :      4: opt.transform.jit_opt_b
 0.11% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.60% :     0.000589s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000040s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.35% :     0.000385s :      1: renormalize.infer
 1.97% :     0.000322s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000039s :      1: rewriter_after_opt_a
 0.46% :     0.000075s :      1: rewriter_before_opt_a
 0.50% :     0.000082s :      1: symbol_engine_optimizer
34.57% :     0.005658s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.615.216 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6c0724c0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.615.766 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0182245, [33]
[bootstrap]: 0.00033544
[type_inference]: 0.00446068
[event_method]: 1.696e-05
[auto_monad]: 6.303e-05
[graph_reusing]: 6.33998e-06
[pre_auto_parallel]: 2.41998e-06
[py_interpret_to_execute]: 2.923e-05
[rewriter_before_opt_a]: 7.149e-05
[expand_dump_flag]: 3.14999e-06
[jit_opt_a]: 0.0053055, [2]
    [Cycle 1]: 0.00141156, [27]
        [switch_simplify]: 5.12e-05
        [loop_unroll]: 2.813e-05
        [a_1]: 0.00049894
        [with_stream_mark]: 1.662e-05
        [recompute_prepare]: 6.98e-06
        [updatestate_depend_eliminate]: 3.23e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [parameter_eliminate]: 1.72001e-06
        [specialize_transform]: 5.45001e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 4.95999e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 4.81002e-06
        [merge_forward]: 3.48e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.334e-05
        [j_node_and_user_rematch]: 8.32e-06
        [meta_fg_expand]: 2.10002e-06
        [replace_old_param]: 8.23999e-06
        [inline_without_move]: 5.36998e-06
        [renormalize]: 0.00049424
        [add_forward_monad_depend]: 2.449e-05
        [auto_monad_grad]: 2.42001e-06
        [auto_monad_eliminator]: 1.366e-05
        [cse]: 2.701e-05
        [replace_applicator]: 1.075e-05
    [Cycle 2]: 0.0003153, [27]
        [switch_simplify]: 5.44e-06
        [loop_unroll]: 4.57e-06
        [a_1]: 9.144e-05
        [with_stream_mark]: 8.82e-06
        [recompute_prepare]: 4.90999e-06
        [updatestate_depend_eliminate]: 2.54001e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 1.16002e-06
        [specialize_transform]: 4.89998e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 4.72e-06
        [meta_shard_fg_expand]: 1.01002e-06
        [get_grad_eliminate_]: 4.27e-06
        [merge_forward]: 2.32999e-06
        [cell_reuse_recompute_pass]: 1.46998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.204e-05
        [j_node_and_user_rematch]: 7.9e-06
        [meta_fg_expand]: 1.52999e-06
        [replace_old_param]: 6.98998e-06
        [inline_without_move]: 4.55001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 5.10999e-06
        [cse]: 1.023e-05
        [replace_applicator]: 5.00999e-06
[py_interpret_to_execute_after_opt_a]: 8.92999e-06
[rewriter_after_opt_a]: 3.225e-05
[convert_after_rewriter]: 6.46999e-06
[order_py_execute_after_rewriter]: 4.50999e-06
[mutable_eliminate]: 0.00048319
[jit_opt_b]: 4.72e-05, [1]
    [Cycle 1]: 4.085e-05, [2]
        [frontend_op_eliminate]: 1.511e-05
        [inline_after_opt_a]: 1.436e-05
[cconv]: 2.251e-05
[loop_unroll]: 0.00039628
[jit_opt_after_cconv]: 0.0001293, [1]
    [Cycle 1]: 0.00012308, [11]
        [c_1]: 1.935e-05
        [parameter_eliminate]: 2.61999e-06
        [updatestate_depend_eliminate]: 4.75001e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [cse]: 1.529e-05
        [call_graph_tuple_transform]: 1.678e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.31998e-06
        [renormalize]: 3.7998e-07
        [switch_simplify]: 5.24e-06
[remove_dup_value]: 1.306e-05
[partial_unused_args_eliminate]: 2.24999e-06
[environ_conv]: 5.26998e-06
[add_recomputation]: 4.599e-05
[cse_after_recomputation]: 2.002e-05, [1]
    [Cycle 1]: 1.451e-05, [1]
        [cse]: 8.97e-06
[auto_monad_reorder]: 1.521e-05
[get_jit_bprop_graph]: 1.32e-06
[rewriter_after_jit_bprop_graph]: 4.88001e-06
[opt_after_jit_grad]: 0.00043228
[symbol_engine_optimizer]: 6.837e-05, [1]
    [Cycle 1]: 6.269e-05, [6]
        [build]: 3.13e-06
        [elim_shapecalc]: 7.41001e-06
        [elim_not_effective]: 1.13e-05
        [opt_reshape]: 5.05999e-06
        [fold_const_symbol]: 8.58001e-06
        [renormalize]: 4.09986e-07
[validate]: 3.107e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00594637
[execute]: 8.05999e-06
Sums
    bootstrap                            :     0.000335s :  2.39%
    type_inference                       :     0.004461s : 31.75%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000063s :  0.45%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.21%
    rewriter_before_opt_a                :     0.000071s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.23%
    jit_opt_a.a_1                        :     0.000590s :  4.20%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000494s :  3.52%
    jit_opt_a.add_forward_monad_depend   :     0.000026s :  0.18%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000037s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000032s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000483s :  3.44%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000023s :  0.16%
    loop_unroll                          :     0.000396s :  2.82%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000046s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000432s :  3.08%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005946s : 42.33%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000183    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.81% :     0.000001s :      2: substitution.fold_const_symbol
 2.71% :     0.000005s :      3: substitution.graph_param_transform
81.36% :     0.000149s :      4: substitution.inline
 1.73% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.42% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.26% :     0.000004s :      2: substitution.replace_old_param
 7.62% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004393     2
81.77% :     0.003592s :      1: type_inference.infer
18.23% :     0.000801s :      1: type_inference.specialize
------[replace.]   0.000061     6
66.69% :     0.000041s :      4: replace.inline
33.31% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000159     6
91.92% :     0.000146s :      4: match.inline
 8.08% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.65% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.54% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000001s :     10: predicate.addn_zero_filter
 2.10% :     0.000002s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.49% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.60% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.83% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.50% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.97% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.81% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.77% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.44% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.68% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.82% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.61% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.58% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.58% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.67% :     0.000002s :     10: predicate.transpose_eliminate
 1.28% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.90% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000517    12
51.07% :     0.000264s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.34% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.58% :     0.000225s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019603    76
 0.25% :     0.000049s :      1: add_recomputation
 0.34% :     0.000066s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.90% :     0.000373s :      1: bootstrap
 0.13% :     0.000025s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000003s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
27.08% :     0.005308s :      1: jit_opt_a
 0.67% :     0.000132s :      1: jit_opt_after_cconv
 0.25% :     0.000050s :      1: jit_opt_b
 2.06% :     0.000404s :      1: loop_unroll
 2.51% :     0.000492s :      1: mutable_eliminate
 4.01% :     0.000787s :     26: opt.transform.jit_opt_a
 0.22% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.25% :     0.000441s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.26% :     0.000247s :      1: renormalize.infer
 1.23% :     0.000240s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000035s :      1: rewriter_after_opt_a
 0.38% :     0.000074s :      1: rewriter_before_opt_a
 0.36% :     0.000071s :      1: symbol_engine_optimizer
30.40% :     0.005960s :      1: task_emit
22.82% :     0.004474s :      1: type_inference
 0.24% :     0.000047s :      1: validate


TotalTime = 0.0137178, [30]
[bootstrap]: 0.00031551
[type_inference]: 0.00458722
[event_method]: 1.785e-05
[auto_monad]: 6.428e-05
[graph_reusing]: 5.81998e-06
[pre_auto_parallel]: 2.46998e-06
[py_interpret_to_execute]: 3.564e-05
[rewriter_before_opt_a]: 6.999e-05
[expand_dump_flag]: 3.02002e-06
[jit_opt_a]: 0.00601861, [2]
    [Cycle 1]: 0.00188676, [27]
        [switch_simplify]: 5.239e-05
        [loop_unroll]: 2.946e-05
        [a_1]: 0.00062164
        [with_stream_mark]: 2.283e-05
        [recompute_prepare]: 9.28002e-06
        [updatestate_depend_eliminate]: 3.09001e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 6.01998e-06
        [updatestate_useless_node_eliminater]: 5.37001e-06
        [accelerated_algorithm]: 6.02999e-06
        [meta_shard_fg_expand]: 2.29001e-06
        [get_grad_eliminate_]: 5.62001e-06
        [merge_forward]: 4.25e-06
        [cell_reuse_recompute_pass]: 2.83003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.608e-05
        [j_node_and_user_rematch]: 1.032e-05
        [meta_fg_expand]: 2.53e-06
        [replace_old_param]: 1.021e-05
        [inline_without_move]: 5.35999e-06
        [renormalize]: 0.00081117
        [add_forward_monad_depend]: 7.24001e-06
        [auto_monad_grad]: 2.99999e-06
        [auto_monad_eliminator]: 1.639e-05
        [cse]: 2.562e-05
        [replace_applicator]: 1.66e-05
    [Cycle 2]: 0.00036411, [27]
        [switch_simplify]: 6.20002e-06
        [loop_unroll]: 5.76998e-06
        [a_1]: 9.129e-05
        [with_stream_mark]: 1.429e-05
        [recompute_prepare]: 6.81001e-06
        [updatestate_depend_eliminate]: 3.57002e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 1.93997e-06
        [parameter_eliminate]: 1.56998e-06
        [specialize_transform]: 6.21998e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 2.04e-06
        [get_grad_eliminate_]: 5.07999e-06
        [merge_forward]: 3.28e-06
        [cell_reuse_recompute_pass]: 1.80001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.682e-05
        [j_node_and_user_rematch]: 9.43002e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 8.69e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 2.57001e-06
        [auto_monad_grad]: 1.37e-06
        [auto_monad_eliminator]: 8.21002e-06
        [cse]: 1.415e-05
        [replace_applicator]: 7.82998e-06
[py_interpret_to_execute_after_opt_a]: 1.495e-05
[rewriter_after_opt_a]: 4.179e-05
[convert_after_rewriter]: 7.5e-06
[order_py_execute_after_rewriter]: 5.32999e-06
[mutable_eliminate]: 0.00077816
[jit_opt_b]: 5.645e-05, [1]
    [Cycle 1]: 4.676e-05, [2]
        [frontend_op_eliminate]: 1.788e-05
        [inline_after_opt_a]: 1.567e-05
[cconv]: 3.065e-05
[loop_unroll]: 0.00050656
[jit_opt_after_cconv]: 0.00014996, [1]
    [Cycle 1]: 0.00014284, [11]
        [c_1]: 2.1e-05
        [parameter_eliminate]: 4.17e-06
        [updatestate_depend_eliminate]: 6.32001e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [cse]: 2.198e-05
        [call_graph_tuple_transform]: 2.082e-05
        [tuple_list_get_item_eliminator]: 5.35001e-06
        [none_parameter_eliminate]: 1.92001e-06
        [renormalize]: 5.59987e-07
        [switch_simplify]: 5.50001e-06
[remove_dup_value]: 1.559e-05
[partial_unused_args_eliminate]: 2.12001e-06
[environ_conv]: 5.61e-06
[add_recomputation]: 5.122e-05
[cse_after_recomputation]: 2.187e-05, [1]
    [Cycle 1]: 1.597e-05, [1]
        [cse]: 9.30001e-06
[auto_monad_reorder]: 1.556e-05
[get_jit_bprop_graph]: 2.35002e-06
[rewriter_after_jit_bprop_graph]: 5.53002e-06
[opt_after_jit_grad]: 0.00052597
[symbol_engine_optimizer]: 7.867e-05, [1]
    [Cycle 1]: 7.08e-05, [6]
        [build]: 4.70001e-06
        [elim_shapecalc]: 8.25e-06
        [elim_not_effective]: 1.464e-05
        [opt_reshape]: 5.93998e-06
        [fold_const_symbol]: 8.80999e-06
        [renormalize]: 6.19999e-07
[validate]: 4.313e-05
Sums
    bootstrap                            :     0.000316s :  3.40%
    type_inference                       :     0.004587s : 49.47%
    event_method                         :     0.000018s :  0.19%
    auto_monad                           :     0.000064s :  0.69%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.38%
    rewriter_before_opt_a                :     0.000070s :  0.75%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000059s :  0.63%
    jit_opt_a.loop_unroll                :     0.000035s :  0.38%
    jit_opt_a.a_1                        :     0.000713s :  7.69%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.40%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.12%
    jit_opt_a.merge_forward              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000019s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000811s :  8.75%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.11%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.27%
    jit_opt_a.cse                        :     0.000040s :  0.43%
    jit_opt_a.replace_applicator         :     0.000024s :  0.26%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.16%
    rewriter_after_opt_a                 :     0.000042s :  0.45%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000778s :  8.39%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.17%
    cconv                                :     0.000031s :  0.33%
    loop_unroll                          :     0.000507s :  5.46%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000016s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000051s :  0.55%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000526s :  5.67%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.47%


Time group info:
------[substitution.]   0.000236    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000001s :      2: substitution.fold_const_symbol
 2.24% :     0.000005s :      3: substitution.graph_param_transform
84.38% :     0.000199s :      4: substitution.inline
 1.93% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.10% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.50% :     0.000006s :      2: substitution.replace_old_param
 5.47% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004524     2
81.62% :     0.003692s :      1: type_inference.infer
18.38% :     0.000831s :      1: type_inference.specialize
------[replace.]   0.000077     6
62.77% :     0.000048s :      4: replace.inline
37.23% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000208     6
94.45% :     0.000196s :      4: match.inline
 5.55% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.25% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.63% :     0.000002s :     10: predicate.addn_zero_filter
 2.12% :     0.000003s :     10: predicate.arithmetic_simplify
 1.12% :     0.000001s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.71% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.24% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.18% :     0.000001s :      6: predicate.get_grad_eliminate
 0.59% :     0.000001s :      3: predicate.graph_param_transform
 5.81% :     0.000007s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :      6: predicate.less_batch_normalization
 2.15% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 2.38% :     0.000003s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 1.94% :     0.000002s :     15: predicate.partial_eliminate
 1.02% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.74% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.98% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     10: predicate.reshape_eliminate
 1.09% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 1.71% :     0.000002s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.49% :     0.000009s :     53: predicate.switch_simplify
 1.02% :     0.000001s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.82% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.85% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000593    12
48.02% :     0.000285s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.07% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.91% :     0.000260s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015527    72
 0.36% :     0.000055s :      1: add_recomputation
 0.44% :     0.000068s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 2.12% :     0.000330s :      1: bootstrap
 0.22% :     0.000034s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000021s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.79% :     0.006023s :      1: jit_opt_a
 0.99% :     0.000153s :      1: jit_opt_after_cconv
 0.38% :     0.000060s :      1: jit_opt_b
 3.32% :     0.000516s :      1: loop_unroll
 5.11% :     0.000793s :      1: mutable_eliminate
 6.08% :     0.000944s :     26: opt.transform.jit_opt_a
 0.32% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000026s :      4: opt.transform.jit_opt_b
 0.10% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.45% :     0.000535s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000038s :      1: py_interpret_to_execute
 0.12% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000018s :      1: remove_dup_value
 2.86% :     0.000443s :      1: renormalize.infer
 2.29% :     0.000356s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.29% :     0.000046s :      1: rewriter_after_opt_a
 0.47% :     0.000073s :      1: rewriter_before_opt_a
 0.86% :     0.000134s :      1: symbol_engine_optimizer
29.62% :     0.004599s :      1: type_inference


TotalTime = 0.0211198, [33]
[bootstrap]: 0.00031204
[type_inference]: 0.00510232
[event_method]: 1.809e-05
[auto_monad]: 6.26e-05
[graph_reusing]: 6.58e-06
[pre_auto_parallel]: 2.42001e-06
[py_interpret_to_execute]: 3.799e-05
[rewriter_before_opt_a]: 7.494e-05
[expand_dump_flag]: 2.94999e-06
[jit_opt_a]: 0.00575344, [2]
    [Cycle 1]: 0.00180018, [27]
        [switch_simplify]: 6.301e-05
        [loop_unroll]: 2.891e-05
        [a_1]: 0.00055663
        [with_stream_mark]: 1.797e-05
        [recompute_prepare]: 7.92e-06
        [updatestate_depend_eliminate]: 3.46001e-06
        [updatestate_assign_eliminate]: 2.79999e-06
        [updatestate_loads_eliminate]: 2.48998e-06
        [parameter_eliminate]: 1.71002e-06
        [specialize_transform]: 5.75001e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.94999e-06
        [meta_shard_fg_expand]: 2.16998e-06
        [get_grad_eliminate_]: 5.44e-06
        [merge_forward]: 3.76999e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.512e-05
        [j_node_and_user_rematch]: 9.31e-06
        [meta_fg_expand]: 2.21e-06
        [replace_old_param]: 8.49998e-06
        [inline_without_move]: 5.61998e-06
        [renormalize]: 0.00078774
        [add_forward_monad_depend]: 5.84999e-06
        [auto_monad_grad]: 2.37001e-06
        [auto_monad_eliminator]: 1.444e-05
        [cse]: 2.878e-05
        [replace_applicator]: 1.213e-05
    [Cycle 2]: 0.00033224, [27]
        [switch_simplify]: 5.96998e-06
        [loop_unroll]: 4.79e-06
        [a_1]: 9.379e-05
        [with_stream_mark]: 1.255e-05
        [recompute_prepare]: 4.95999e-06
        [updatestate_depend_eliminate]: 2.69999e-06
        [updatestate_assign_eliminate]: 2.40002e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 4.84998e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 4.92999e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 4.61002e-06
        [merge_forward]: 2.90998e-06
        [cell_reuse_recompute_pass]: 1.70001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.439e-05
        [j_node_and_user_rematch]: 8.60001e-06
        [meta_fg_expand]: 1.99e-06
        [replace_old_param]: 7.78001e-06
        [inline_without_move]: 4.69998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.84e-06
        [auto_monad_grad]: 7.29982e-07
        [auto_monad_eliminator]: 4.93001e-06
        [cse]: 1.132e-05
        [replace_applicator]: 5.56998e-06
[py_interpret_to_execute_after_opt_a]: 1.002e-05
[rewriter_after_opt_a]: 3.614e-05
[convert_after_rewriter]: 7.24001e-06
[order_py_execute_after_rewriter]: 5.39e-06
[mutable_eliminate]: 0.00072621
[jit_opt_b]: 5.878e-05, [1]
    [Cycle 1]: 5.011e-05, [2]
        [frontend_op_eliminate]: 1.689e-05
        [inline_after_opt_a]: 1.577e-05
[cconv]: 3.288e-05
[loop_unroll]: 0.00046026
[jit_opt_after_cconv]: 0.0001609, [1]
    [Cycle 1]: 0.00015344, [11]
        [c_1]: 2.205e-05
        [parameter_eliminate]: 4.50999e-06
        [updatestate_depend_eliminate]: 8.99998e-06
        [updatestate_assign_eliminate]: 2.89001e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [cse]: 2.683e-05
        [call_graph_tuple_transform]: 2.112e-05
        [tuple_list_get_item_eliminator]: 5.32999e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 9.80013e-07
        [switch_simplify]: 5.73997e-06
[remove_dup_value]: 1.522e-05
[partial_unused_args_eliminate]: 2.82002e-06
[environ_conv]: 5.99999e-06
[add_recomputation]: 5.433e-05
[cse_after_recomputation]: 2.193e-05, [1]
    [Cycle 1]: 1.563e-05, [1]
        [cse]: 9.54999e-06
[auto_monad_reorder]: 1.487e-05
[get_jit_bprop_graph]: 2.28998e-06
[rewriter_after_jit_bprop_graph]: 6.13998e-06
[opt_after_jit_grad]: 0.00049742
[symbol_engine_optimizer]: 7.96e-05, [1]
    [Cycle 1]: 7.243e-05, [6]
        [build]: 4.92e-06
        [elim_shapecalc]: 7.56001e-06
        [elim_not_effective]: 1.507e-05
        [opt_reshape]: 6.52001e-06
        [fold_const_symbol]: 9.04998e-06
        [renormalize]: 7.90023e-07
[validate]: 4.441e-05
[backend_pass]: 8.80013e-07
[task_emit]: 0.00726661
[execute]: 7.56001e-06
Sums
    bootstrap                            :     0.000312s :  1.85%
    type_inference                       :     0.005102s : 30.32%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000063s :  0.37%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000038s :  0.23%
    rewriter_before_opt_a                :     0.000075s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000069s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000650s :  3.86%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000788s :  4.68%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000040s :  0.24%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000036s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000726s :  4.32%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000033s :  0.20%
    loop_unroll                          :     0.000460s :  2.73%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000054s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000497s :  2.96%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007267s : 43.18%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000218    23
 1.20% :     0.000003s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.85% :     0.000006s :      3: substitution.graph_param_transform
83.38% :     0.000182s :      4: substitution.inline
 1.65% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.44% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.04% :     0.000004s :      2: substitution.replace_old_param
 5.86% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005027     2
81.50% :     0.004097s :      1: type_inference.infer
18.50% :     0.000930s :      1: type_inference.specialize
------[replace.]   0.000077     6
69.24% :     0.000053s :      4: replace.inline
30.76% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000191     6
93.85% :     0.000179s :      4: match.inline
 6.15% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.27% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.87% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 2.52% :     0.000003s :     10: predicate.arithmetic_simplify
 1.26% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.15% :     0.000001s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.52% :     0.000001s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.62% :     0.000007s :     22: predicate.inline
 1.02% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000002s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.62% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.66% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 2.76% :     0.000003s :      3: predicate.mutable_eliminate
 0.58% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000003s :     15: predicate.partial_eliminate
 1.08% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.88% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.21% :     0.000001s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.02% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.84% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.80% :     0.000003s :     16: predicate.switch_defer_inline
 1.96% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.74% :     0.000009s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.86% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.24% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.59% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000668    12
51.49% :     0.000344s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.46% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.04% :     0.000294s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022888    76
 0.25% :     0.000058s :      1: add_recomputation
 0.29% :     0.000066s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.44% :     0.000329s :      1: bootstrap
 0.16% :     0.000036s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.16% :     0.005758s :      1: jit_opt_a
 0.71% :     0.000164s :      1: jit_opt_after_cconv
 0.27% :     0.000062s :      1: jit_opt_b
 2.06% :     0.000471s :      1: loop_unroll
 3.24% :     0.000741s :      1: mutable_eliminate
 3.80% :     0.000870s :     26: opt.transform.jit_opt_a
 0.22% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000507s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000041s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.82% :     0.000416s :      1: renormalize.infer
 1.59% :     0.000363s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000040s :      1: rewriter_after_opt_a
 0.34% :     0.000078s :      1: rewriter_before_opt_a
 0.36% :     0.000082s :      1: symbol_engine_optimizer
31.82% :     0.007282s :      1: task_emit
22.37% :     0.005119s :      1: type_inference
 0.30% :     0.000068s :      1: validate


TotalTime = 0.0334681, [33]
[bootstrap]: 0.00019396
[type_inference]: 0.0126907
[event_method]: 6.13e-05
[auto_monad]: 0.00011045
[graph_reusing]: 8.15e-06
[pre_auto_parallel]: 3.16999e-06
[py_interpret_to_execute]: 2.983e-05
[rewriter_before_opt_a]: 0.00010495
[expand_dump_flag]: 3.47002e-06
[jit_opt_a]: 0.0113681, [3]
    [Cycle 1]: 0.00574959, [27]
        [switch_simplify]: 0.00012529
        [loop_unroll]: 4.447e-05
        [a_1]: 0.00091997
        [with_stream_mark]: 2.699e-05
        [recompute_prepare]: 1.785e-05
        [updatestate_depend_eliminate]: 7.97998e-06
        [updatestate_assign_eliminate]: 6.25002e-06
        [updatestate_loads_eliminate]: 7.03e-06
        [parameter_eliminate]: 3.06999e-06
        [specialize_transform]: 1.328e-05
        [updatestate_useless_node_eliminater]: 1.263e-05
        [accelerated_algorithm]: 1.26e-05
        [meta_shard_fg_expand]: 3.63999e-06
        [get_grad_eliminate_]: 1.233e-05
        [merge_forward]: 8.53001e-06
        [cell_reuse_recompute_pass]: 8.60018e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.618e-05
        [j_node_and_user_rematch]: 2.311e-05
        [meta_fg_expand]: 0.00142833
        [replace_old_param]: 5.225e-05
        [inline_without_move]: 5.145e-05
        [renormalize]: 0.00245761
        [add_forward_monad_depend]: 9.54e-06
        [auto_monad_grad]: 5.79e-06
        [auto_monad_eliminator]: 4.852e-05
        [cse]: 0.00015578
        [replace_applicator]: 6.658e-05
    [Cycle 2]: 0.00193105, [27]
        [switch_simplify]: 3.634e-05
        [loop_unroll]: 3.441e-05
        [a_1]: 0.00105182
        [with_stream_mark]: 1.086e-05
        [recompute_prepare]: 7.68999e-06
        [updatestate_depend_eliminate]: 3.7e-06
        [updatestate_assign_eliminate]: 3.18e-06
        [updatestate_loads_eliminate]: 2.87002e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 6.56e-06
        [updatestate_useless_node_eliminater]: 6.19001e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 2.16e-06
        [get_grad_eliminate_]: 6.13998e-06
        [merge_forward]: 3.61999e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.384e-05
        [j_node_and_user_rematch]: 1.004e-05
        [meta_fg_expand]: 7.728e-05
        [replace_old_param]: 1.022e-05
        [inline_without_move]: 6.21998e-06
        [renormalize]: 0.00043957
        [add_forward_monad_depend]: 3.98999e-06
        [auto_monad_grad]: 1.14e-06
        [auto_monad_eliminator]: 1.056e-05
        [cse]: 1.953e-05
        [replace_applicator]: 1.206e-05
    [Cycle 3]: 0.00036623, [27]
        [switch_simplify]: 6.89999e-06
        [loop_unroll]: 6.17999e-06
        [a_1]: 0.00011971
        [with_stream_mark]: 8.54998e-06
        [recompute_prepare]: 6.23e-06
        [updatestate_depend_eliminate]: 3.96001e-06
        [updatestate_assign_eliminate]: 3.02002e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 9.29984e-07
        [specialize_transform]: 6.24999e-06
        [updatestate_useless_node_eliminater]: 6.21e-06
        [accelerated_algorithm]: 5.85002e-06
        [meta_shard_fg_expand]: 1.27999e-06
        [get_grad_eliminate_]: 5.87001e-06
        [merge_forward]: 3.06999e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.542e-05
        [j_node_and_user_rematch]: 9.99001e-06
        [meta_fg_expand]: 2.01e-06
        [replace_old_param]: 8.22e-06
        [inline_without_move]: 5.79e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.08001e-06
        [auto_monad_grad]: 6.90023e-07
        [auto_monad_eliminator]: 6.17999e-06
        [cse]: 1.363e-05
        [replace_applicator]: 6.04999e-06
[py_interpret_to_execute_after_opt_a]: 1.055e-05
[rewriter_after_opt_a]: 3.952e-05
[convert_after_rewriter]: 7.21001e-06
[order_py_execute_after_rewriter]: 5.67001e-06
[mutable_eliminate]: 0.00051094
[jit_opt_b]: 5.413e-05, [1]
    [Cycle 1]: 4.703e-05, [2]
        [frontend_op_eliminate]: 1.912e-05
        [inline_after_opt_a]: 1.715e-05
[cconv]: 1.995e-05
[loop_unroll]: 0.00045723
[jit_opt_after_cconv]: 0.00015544, [1]
    [Cycle 1]: 0.00014839, [11]
        [c_1]: 2.571e-05
        [parameter_eliminate]: 2.44001e-06
        [updatestate_depend_eliminate]: 6.54999e-06
        [updatestate_assign_eliminate]: 3.14001e-06
        [updatestate_loads_eliminate]: 2.65002e-06
        [cse]: 2.242e-05
        [call_graph_tuple_transform]: 2.174e-05
        [tuple_list_get_item_eliminator]: 6.52001e-06
        [none_parameter_eliminate]: 1.57999e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 6.42001e-06
[remove_dup_value]: 1.6e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 5.27001e-06
[add_recomputation]: 4.903e-05
[cse_after_recomputation]: 3.095e-05, [1]
    [Cycle 1]: 1.932e-05, [1]
        [cse]: 1.367e-05
[auto_monad_reorder]: 1.918e-05
[get_jit_bprop_graph]: 1.54e-06
[rewriter_after_jit_bprop_graph]: 4.80999e-06
[opt_after_jit_grad]: 0.00045569
[symbol_engine_optimizer]: 8.001e-05, [1]
    [Cycle 1]: 7.347e-05, [6]
        [build]: 3.9e-06
        [elim_shapecalc]: 8.80001e-06
        [elim_not_effective]: 1.501e-05
        [opt_reshape]: 6.71999e-06
        [fold_const_symbol]: 1.056e-05
        [renormalize]: 4.80009e-07
[validate]: 3.408e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.00674659
[execute]: 7.38999e-06
Sums
    bootstrap                            :     0.000194s :  0.66%
    type_inference                       :     0.012691s : 43.18%
    event_method                         :     0.000061s :  0.21%
    auto_monad                           :     0.000110s :  0.38%
    graph_reusing                        :     0.000008s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000030s :  0.10%
    rewriter_before_opt_a                :     0.000105s :  0.36%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000169s :  0.57%
    jit_opt_a.loop_unroll                :     0.000085s :  0.29%
    jit_opt_a.a_1                        :     0.002092s :  7.12%
    jit_opt_a.with_stream_mark           :     0.000046s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.08%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000055s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000043s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001508s :  5.13%
    jit_opt_a.replace_old_param          :     0.000071s :  0.24%
    jit_opt_a.inline_without_move        :     0.000063s :  0.22%
    jit_opt_a.renormalize                :     0.002897s :  9.86%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000065s :  0.22%
    jit_opt_a.cse                        :     0.000189s :  0.64%
    jit_opt_a.replace_applicator         :     0.000085s :  0.29%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.04%
    rewriter_after_opt_a                 :     0.000040s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000511s :  1.74%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.06%
    cconv                                :     0.000020s :  0.07%
    loop_unroll                          :     0.000457s :  1.56%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000049s :  0.17%
    cse_after_recomputation.cse          :     0.000014s :  0.05%
    auto_monad_reorder                   :     0.000019s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000456s :  1.55%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006747s : 22.96%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000504   113
 0.53% :     0.000003s :      3: substitution.elim_not_effective
 0.32% :     0.000002s :      3: substitution.fold_const_symbol
 1.28% :     0.000006s :      4: substitution.graph_param_transform
61.17% :     0.000308s :     14: substitution.inline
 3.20% :     0.000016s :      2: substitution.inline_without_move
 1.81% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.10% :     0.000011s :      7: substitution.minmaximum_grad
 1.27% :     0.000006s :      5: substitution.partial_eliminate
 2.22% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.31% :     0.000027s :     10: substitution.replace_applicator
 1.48% :     0.000007s :      7: substitution.replace_old_param
 0.45% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.70% :     0.000009s :      1: substitution.switch_simplify
 3.95% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 5.11% :     0.000026s :      7: substitution.tuple_list_get_item_depend_reorder
 8.09% :     0.000041s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012618     2
89.26% :     0.011264s :      1: type_inference.infer
10.74% :     0.001355s :      1: type_inference.specialize
------[replace.]   0.000182    22
53.44% :     0.000097s :     14: replace.inline
17.68% :     0.000032s :      1: replace.switch_simplify
28.89% :     0.000052s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000329    22
91.15% :     0.000300s :     14: match.inline
 2.43% :     0.000008s :      1: match.switch_simplify
 6.43% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000398  2634
 1.64% :     0.000007s :     44: predicate.accumulaten_eliminater
 0.47% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.54% :     0.000006s :     44: predicate.addn_check_dump
 1.70% :     0.000007s :     44: predicate.addn_zero_filter
 1.89% :     0.000008s :     44: predicate.arithmetic_simplify
 1.46% :     0.000006s :     44: predicate.cast_eliminate
 0.17% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.50% :     0.000006s :     44: predicate.compare_switch_simplify
 1.65% :     0.000007s :     44: predicate.depend_value_elim
 1.53% :     0.000006s :     44: predicate.dict_get_item_const_eliminator
 1.44% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.47% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.30% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      4: predicate.elim_not_effective
 0.22% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.50% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.40% :     0.000006s :     44: predicate.environ_get_add_eliminate
 1.39% :     0.000006s :     44: predicate.environ_get_depend_swap
 1.44% :     0.000006s :     44: predicate.environ_get_eliminate
 1.44% :     0.000006s :     44: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.82% :     0.000003s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.40% :     0.000018s :     73: predicate.inline
 1.71% :     0.000007s :     40: predicate.inline_without_move
 0.33% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.89% :     0.000004s :     19: predicate.less_batch_normalization
 1.90% :     0.000008s :     51: predicate.list_to_tuple_eliminator_
 2.01% :     0.000008s :     55: predicate.load_eliminater
 0.49% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.21% :     0.000013s :     91: predicate.loop_unroll_before_grad
 1.72% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.59% :     0.000006s :     44: predicate.merge_addn
 1.50% :     0.000006s :     44: predicate.minmaximum_grad
 0.51% :     0.000002s :      4: predicate.mutable_eliminate
 0.18% :     0.000001s :      4: predicate.opt_reshape
 2.28% :     0.000009s :     55: predicate.partial_eliminate
 1.70% :     0.000007s :     44: predicate.print_const_string_wrapper
 1.89% :     0.000008s :     44: predicate.reduce_eliminate
 1.81% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.39% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.52% :     0.000010s :     97: predicate.replace_applicator
 0.86% :     0.000003s :     40: predicate.replace_old_param
 0.11% :     0.000000s :      4: predicate.reset_defer_inline
 1.79% :     0.000007s :     44: predicate.reshape_eliminate
 1.66% :     0.000007s :     44: predicate.row_tensor_add_zeros_like
 0.36% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.62% :     0.000006s :     44: predicate.same_eliminate
 0.46% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.40% :     0.000002s :      8: predicate.special_op_eliminate
 0.89% :     0.000004s :     19: predicate.specialize_transform
 1.67% :     0.000007s :     44: predicate.split_environ_get_set_with_tuple_value
 1.52% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.88% :     0.000011s :     65: predicate.switch_defer_inline
 2.65% :     0.000011s :     65: predicate.switch_layer_defer_inline
 6.60% :     0.000026s :    162: predicate.switch_simplify
 1.53% :     0.000006s :     44: predicate.tile_eliminate
 1.44% :     0.000006s :     44: predicate.transpose_eliminate
 1.80% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.59% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 3.02% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.84% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 2.00% :     0.000008s :     55: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000007s :     44: predicate.value_based_eliminate
 0.15% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.26% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001658    32
61.10% :     0.001013s :     14: func_graph_cloner_run.FuncGraphClonerGraph
38.90% :     0.000645s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.039244    91
 0.13% :     0.000052s :      1: add_recomputation
 0.29% :     0.000115s :      1: auto_monad
 0.06% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.53% :     0.000208s :      1: bootstrap
 0.06% :     0.000022s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000033s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.17% :     0.000066s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000011s :      1: graph_reusing
28.98% :     0.011372s :      1: jit_opt_a
 0.40% :     0.000158s :      1: jit_opt_after_cconv
 0.14% :     0.000056s :      1: jit_opt_b
 1.18% :     0.000464s :      1: loop_unroll
 1.32% :     0.000517s :      1: mutable_eliminate
 7.01% :     0.002750s :     39: opt.transform.jit_opt_a
 0.15% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000030s :      4: opt.transform.jit_opt_b
 0.03% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.18% :     0.000464s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000032s :      1: py_interpret_to_execute
 0.03% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 3.95% :     0.001549s :      2: renormalize.infer
 3.40% :     0.001335s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000042s :      1: rewriter_after_opt_a
 0.27% :     0.000108s :      1: rewriter_before_opt_a
 0.21% :     0.000082s :      1: symbol_engine_optimizer
17.21% :     0.006756s :      1: task_emit
32.37% :     0.012703s :      1: type_inference
 0.13% :     0.000053s :      1: validate


TotalTime = 0.0158193, [33]
[bootstrap]: 0.00026435
[type_inference]: 0.00399389
[event_method]: 1.608e-05
[auto_monad]: 5.615e-05
[graph_reusing]: 5.12999e-06
[pre_auto_parallel]: 1.99e-06
[py_interpret_to_execute]: 2.365e-05
[rewriter_before_opt_a]: 0.00011295
[expand_dump_flag]: 2.78998e-06
[jit_opt_a]: 0.00451224, [2]
    [Cycle 1]: 0.0012367, [27]
        [switch_simplify]: 4.039e-05
        [loop_unroll]: 2.785e-05
        [a_1]: 0.00044832
        [with_stream_mark]: 1.077e-05
        [recompute_prepare]: 6.54001e-06
        [updatestate_depend_eliminate]: 2.64999e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 5.61003e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 5.01002e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.94998e-06
        [merge_forward]: 2.37001e-06
        [cell_reuse_recompute_pass]: 7.49977e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.016e-05
        [j_node_and_user_rematch]: 8.08001e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 7.71999e-06
        [inline_without_move]: 5.06997e-06
        [renormalize]: 0.00044522
        [add_forward_monad_depend]: 5.12e-06
        [auto_monad_grad]: 1.10999e-06
        [auto_monad_eliminator]: 9.32001e-06
        [cse]: 1.396e-05
        [replace_applicator]: 1.043e-05
    [Cycle 2]: 0.00030521, [27]
        [switch_simplify]: 5.20001e-06
        [loop_unroll]: 4.68999e-06
        [a_1]: 8.704e-05
        [with_stream_mark]: 8.75001e-06
        [recompute_prepare]: 4.63999e-06
        [updatestate_depend_eliminate]: 2.48998e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 4.77998e-06
        [updatestate_useless_node_eliminater]: 4.66002e-06
        [accelerated_algorithm]: 4.72e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.33002e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.136e-05
        [j_node_and_user_rematch]: 7.24001e-06
        [meta_fg_expand]: 1.44003e-06
        [replace_old_param]: 6.74001e-06
        [inline_without_move]: 4.48001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.02e-06
        [auto_monad_grad]: 7.10017e-07
        [auto_monad_eliminator]: 5.07e-06
        [cse]: 9.19e-06
        [replace_applicator]: 4.73001e-06
[py_interpret_to_execute_after_opt_a]: 7.82e-06
[rewriter_after_opt_a]: 2.425e-05
[convert_after_rewriter]: 8.32e-06
[order_py_execute_after_rewriter]: 3.98999e-06
[mutable_eliminate]: 0.00045717
[jit_opt_b]: 4.583e-05, [1]
    [Cycle 1]: 3.98e-05, [2]
        [frontend_op_eliminate]: 1.491e-05
        [inline_after_opt_a]: 1.393e-05
[cconv]: 1.66e-05
[loop_unroll]: 0.00044037
[jit_opt_after_cconv]: 0.00012725, [1]
    [Cycle 1]: 0.00012124, [11]
        [c_1]: 1.887e-05
        [parameter_eliminate]: 2.53998e-06
        [updatestate_depend_eliminate]: 4.42e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [cse]: 1.525e-05
        [call_graph_tuple_transform]: 1.565e-05
        [tuple_list_get_item_eliminator]: 5.10001e-06
        [none_parameter_eliminate]: 1.14998e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 4.99e-06
[remove_dup_value]: 7.61999e-06
[partial_unused_args_eliminate]: 1.54e-06
[environ_conv]: 3.74002e-06
[add_recomputation]: 3.568e-05
[cse_after_recomputation]: 1.917e-05, [1]
    [Cycle 1]: 1.433e-05, [1]
        [cse]: 7.82e-06
[auto_monad_reorder]: 1.054e-05
[get_jit_bprop_graph]: 1.47999e-06
[rewriter_after_jit_bprop_graph]: 5.07999e-06
[opt_after_jit_grad]: 0.00044224
[symbol_engine_optimizer]: 6.648e-05, [1]
    [Cycle 1]: 6.07e-05, [6]
        [build]: 2.60002e-06
        [elim_shapecalc]: 6.86999e-06
        [elim_not_effective]: 1.127e-05
        [opt_reshape]: 5.72001e-06
        [fold_const_symbol]: 7.87003e-06
        [renormalize]: 5.19998e-07
[validate]: 2.605e-05
[backend_pass]: 6.89994e-07
[task_emit]: 0.00491714
[execute]: 4.20999e-06
Sums
    bootstrap                            :     0.000264s :  2.15%
    type_inference                       :     0.003994s : 32.44%
    event_method                         :     0.000016s :  0.13%
    auto_monad                           :     0.000056s :  0.46%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000024s :  0.19%
    rewriter_before_opt_a                :     0.000113s :  0.92%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000046s :  0.37%
    jit_opt_a.loop_unroll                :     0.000033s :  0.26%
    jit_opt_a.a_1                        :     0.000535s :  4.35%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000015s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000445s :  3.62%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.12%
    jit_opt_a.cse                        :     0.000023s :  0.19%
    jit_opt_a.replace_applicator         :     0.000015s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000024s :  0.20%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000457s :  3.71%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000017s :  0.13%
    loop_unroll                          :     0.000440s :  3.58%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000008s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000036s :  0.29%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000011s :  0.09%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000442s :  3.59%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000026s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.004917s : 39.93%
    execute                              :     0.000004s :  0.03%


Time group info:
------[substitution.]   0.000140    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.81% :     0.000004s :      3: substitution.graph_param_transform
83.25% :     0.000117s :      4: substitution.inline
 1.84% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.70% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.05% :     0.000003s :      2: substitution.replace_old_param
 5.58% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003944     2
82.93% :     0.003271s :      1: type_inference.infer
17.07% :     0.000673s :      1: type_inference.specialize
------[replace.]   0.000055     6
66.09% :     0.000037s :      4: replace.inline
33.91% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000121     6
94.36% :     0.000115s :      4: match.inline
 5.64% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.66% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 2.61% :     0.000003s :     10: predicate.arithmetic_simplify
 1.29% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.47% :     0.000001s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.51% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.05% :     0.000006s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.99% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.29% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.11% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.22% :     0.000001s :      3: predicate.mutable_eliminate
 0.79% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000002s :     15: predicate.partial_eliminate
 1.48% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.10% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.50% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.30% :     0.000008s :     53: predicate.switch_simplify
 1.28% :     0.000001s :     10: predicate.tile_eliminate
 1.45% :     0.000002s :     10: predicate.transpose_eliminate
 1.32% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.25% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.94% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.90% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.87% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000469    12
43.32% :     0.000203s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.03% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.65% :     0.000242s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017085    76
 0.23% :     0.000039s :      1: add_recomputation
 0.35% :     0.000059s :      1: auto_monad
 0.08% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.61% :     0.000275s :      1: bootstrap
 0.11% :     0.000019s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.05% :     0.000008s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.43% :     0.004515s :      1: jit_opt_a
 0.76% :     0.000130s :      1: jit_opt_after_cconv
 0.28% :     0.000049s :      1: jit_opt_b
 2.62% :     0.000447s :      1: loop_unroll
 2.72% :     0.000464s :      1: mutable_eliminate
 4.20% :     0.000718s :     26: opt.transform.jit_opt_a
 0.24% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.63% :     0.000449s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000026s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000010s :      1: remove_dup_value
 1.19% :     0.000204s :      1: renormalize.infer
 1.38% :     0.000236s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000027s :      1: rewriter_after_opt_a
 0.68% :     0.000116s :      1: rewriter_before_opt_a
 0.40% :     0.000069s :      1: symbol_engine_optimizer
28.84% :     0.004927s :      1: task_emit
23.43% :     0.004003s :      1: type_inference
 0.24% :     0.000041s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.657.265 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6c072940> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.657.811 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0173236, [33]
[bootstrap]: 0.00026602
[type_inference]: 0.00397821
[event_method]: 1.622e-05
[auto_monad]: 6.379e-05
[graph_reusing]: 5.54998e-06
[pre_auto_parallel]: 2.51e-06
[py_interpret_to_execute]: 2.731e-05
[rewriter_before_opt_a]: 7.018e-05
[expand_dump_flag]: 3.50998e-06
[jit_opt_a]: 0.00499912, [2]
    [Cycle 1]: 0.00132501, [27]
        [switch_simplify]: 7.028e-05
        [loop_unroll]: 2.874e-05
        [a_1]: 0.00047261
        [with_stream_mark]: 1.394e-05
        [recompute_prepare]: 6.84001e-06
        [updatestate_depend_eliminate]: 3.35e-06
        [updatestate_assign_eliminate]: 2.67001e-06
        [updatestate_loads_eliminate]: 2.68998e-06
        [parameter_eliminate]: 1.57999e-06
        [specialize_transform]: 6.01e-06
        [updatestate_useless_node_eliminater]: 4.84998e-06
        [accelerated_algorithm]: 5.00999e-06
        [meta_shard_fg_expand]: 1.84e-06
        [get_grad_eliminate_]: 4.85001e-06
        [merge_forward]: 3.67998e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.375e-05
        [j_node_and_user_rematch]: 8.62e-06
        [meta_fg_expand]: 2.07001e-06
        [replace_old_param]: 9.07999e-06
        [inline_without_move]: 4.99e-06
        [renormalize]: 0.00044207
        [add_forward_monad_depend]: 5.17999e-06
        [auto_monad_grad]: 1.70001e-06
        [auto_monad_eliminator]: 1.244e-05
        [cse]: 2.735e-05
        [replace_applicator]: 1.081e-05
    [Cycle 2]: 0.0002992, [27]
        [switch_simplify]: 5.39e-06
        [loop_unroll]: 4.70001e-06
        [a_1]: 8.682e-05
        [with_stream_mark]: 8.59e-06
        [recompute_prepare]: 4.71002e-06
        [updatestate_depend_eliminate]: 2.50997e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 8.59989e-07
        [specialize_transform]: 4.65001e-06
        [updatestate_useless_node_eliminater]: 4.59998e-06
        [accelerated_algorithm]: 4.44998e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.57998e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 1.63002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.13e-05
        [j_node_and_user_rematch]: 7.56999e-06
        [meta_fg_expand]: 1.51002e-06
        [replace_old_param]: 6.66999e-06
        [inline_without_move]: 4.55999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 4.79e-06
        [cse]: 8.90999e-06
        [replace_applicator]: 4.80001e-06
[py_interpret_to_execute_after_opt_a]: 6.38998e-06
[rewriter_after_opt_a]: 3.18e-05
[convert_after_rewriter]: 6.81001e-06
[order_py_execute_after_rewriter]: 4.90001e-06
[mutable_eliminate]: 0.00048828
[jit_opt_b]: 4.697e-05, [1]
    [Cycle 1]: 4.026e-05, [2]
        [frontend_op_eliminate]: 1.538e-05
        [inline_after_opt_a]: 1.383e-05
[cconv]: 2.304e-05
[loop_unroll]: 0.0004053
[jit_opt_after_cconv]: 0.0001316, [1]
    [Cycle 1]: 0.00012548, [11]
        [c_1]: 1.918e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 5.23002e-06
        [updatestate_assign_eliminate]: 2.70997e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [cse]: 1.765e-05
        [call_graph_tuple_transform]: 1.646e-05
        [tuple_list_get_item_eliminator]: 4.95001e-06
        [none_parameter_eliminate]: 1.39998e-06
        [renormalize]: 3.60014e-07
        [switch_simplify]: 5.05001e-06
[remove_dup_value]: 1.214e-05
[partial_unused_args_eliminate]: 2.13002e-06
[environ_conv]: 4.82998e-06
[add_recomputation]: 4.819e-05
[cse_after_recomputation]: 1.963e-05, [1]
    [Cycle 1]: 1.443e-05, [1]
        [cse]: 8.15e-06
[auto_monad_reorder]: 1.478e-05
[get_jit_bprop_graph]: 1.35001e-06
[rewriter_after_jit_bprop_graph]: 4.53999e-06
[opt_after_jit_grad]: 0.00043634
[symbol_engine_optimizer]: 6.863e-05, [1]
    [Cycle 1]: 6.284e-05, [6]
        [build]: 3.68e-06
        [elim_shapecalc]: 6.88998e-06
        [elim_not_effective]: 1.154e-05
        [opt_reshape]: 5.25999e-06
        [fold_const_symbol]: 7.93001e-06
        [renormalize]: 2.79979e-07
[validate]: 3.172e-05
[backend_pass]: 9.49978e-07
[task_emit]: 0.00592306
[execute]: 6.30002e-06
Sums
    bootstrap                            :     0.000266s :  1.99%
    type_inference                       :     0.003978s : 29.69%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000064s :  0.48%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000027s :  0.20%
    rewriter_before_opt_a                :     0.000070s :  0.52%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000076s :  0.56%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000559s :  4.18%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000009s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000442s :  3.30%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000036s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000006s :  0.05%
    rewriter_after_opt_a                 :     0.000032s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000488s :  3.64%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000405s :  3.03%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000048s :  0.36%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000436s :  3.26%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005923s : 44.21%
    execute                              :     0.000006s :  0.05%


Time group info:
------[substitution.]   0.000165    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000001s :      2: substitution.fold_const_symbol
 2.90% :     0.000005s :      3: substitution.graph_param_transform
81.72% :     0.000135s :      4: substitution.inline
 1.83% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.48% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.23% :     0.000004s :      2: substitution.replace_old_param
 7.02% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003918     2
79.74% :     0.003124s :      1: type_inference.infer
20.26% :     0.000794s :      1: type_inference.specialize
------[replace.]   0.000059     6
68.05% :     0.000040s :      4: replace.inline
31.95% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000143     6
92.60% :     0.000132s :      4: match.inline
 7.40% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000001s :     10: predicate.addn_zero_filter
 2.27% :     0.000002s :     10: predicate.arithmetic_simplify
 1.49% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000001s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.50% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.49% :     0.000001s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.23% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.15% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.58% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000002s :     15: predicate.partial_eliminate
 1.33% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 1.11% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.71% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.32% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.42% :     0.000008s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.35% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.75% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000444    12
47.24% :     0.000210s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.62% :     0.000021s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.14% :     0.000214s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018622    76
 0.27% :     0.000051s :      1: add_recomputation
 0.36% :     0.000067s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.49% :     0.000278s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000003s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.86% :     0.005002s :      1: jit_opt_a
 0.72% :     0.000134s :      1: jit_opt_after_cconv
 0.27% :     0.000050s :      1: jit_opt_b
 2.22% :     0.000413s :      1: loop_unroll
 2.67% :     0.000497s :      1: mutable_eliminate
 4.05% :     0.000754s :     26: opt.transform.jit_opt_a
 0.23% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.38% :     0.000444s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000030s :      1: py_interpret_to_execute
 0.05% :     0.000009s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 1.27% :     0.000236s :      1: renormalize.infer
 1.07% :     0.000199s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000035s :      1: rewriter_after_opt_a
 0.39% :     0.000073s :      1: rewriter_before_opt_a
 0.38% :     0.000071s :      1: symbol_engine_optimizer
31.86% :     0.005933s :      1: task_emit
21.43% :     0.003990s :      1: type_inference
 0.27% :     0.000050s :      1: validate


TotalTime = 0.0150127, [30]
[bootstrap]: 0.00032326
[type_inference]: 0.00503127
[event_method]: 1.882e-05
[auto_monad]: 6.61e-05
[graph_reusing]: 6.61e-06
[pre_auto_parallel]: 3.08998e-06
[py_interpret_to_execute]: 0.00010472
[rewriter_before_opt_a]: 7.861e-05
[expand_dump_flag]: 3.98001e-06
[jit_opt_a]: 0.00668647, [2]
    [Cycle 1]: 0.00200191, [27]
        [switch_simplify]: 7.035e-05
        [loop_unroll]: 3.191e-05
        [a_1]: 0.00061867
        [with_stream_mark]: 2.668e-05
        [recompute_prepare]: 8.94e-06
        [updatestate_depend_eliminate]: 4.12e-06
        [updatestate_assign_eliminate]: 2.79999e-06
        [updatestate_loads_eliminate]: 2.81e-06
        [parameter_eliminate]: 1.97001e-06
        [specialize_transform]: 6.31e-06
        [updatestate_useless_node_eliminater]: 5.39e-06
        [accelerated_algorithm]: 6.29999e-06
        [meta_shard_fg_expand]: 2.42001e-06
        [get_grad_eliminate_]: 5.57999e-06
        [merge_forward]: 3.75e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.622e-05
        [j_node_and_user_rematch]: 9.41e-06
        [meta_fg_expand]: 2.58e-06
        [replace_old_param]: 1.114e-05
        [inline_without_move]: 5.49e-06
        [renormalize]: 0.00089066
        [add_forward_monad_depend]: 6.78e-06
        [auto_monad_grad]: 2.59999e-06
        [auto_monad_eliminator]: 1.525e-05
        [cse]: 3.218e-05
        [replace_applicator]: 1.36e-05
    [Cycle 2]: 0.00034348, [27]
        [switch_simplify]: 6.21e-06
        [loop_unroll]: 5.18002e-06
        [a_1]: 9.234e-05
        [with_stream_mark]: 1.082e-05
        [recompute_prepare]: 5.40999e-06
        [updatestate_depend_eliminate]: 2.79999e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 5.02999e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.74999e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 5.02e-06
        [merge_forward]: 2.78998e-06
        [cell_reuse_recompute_pass]: 2.06e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.4e-05
        [j_node_and_user_rematch]: 8.2e-06
        [meta_fg_expand]: 1.77001e-06
        [replace_old_param]: 8.10999e-06
        [inline_without_move]: 4.74002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.74998e-06
        [auto_monad_grad]: 9.79984e-07
        [auto_monad_eliminator]: 6.58e-06
        [cse]: 1.257e-05
        [replace_applicator]: 5.81e-06
[py_interpret_to_execute_after_opt_a]: 1.655e-05
[rewriter_after_opt_a]: 3.986e-05
[convert_after_rewriter]: 7.87998e-06
[order_py_execute_after_rewriter]: 5.12e-06
[mutable_eliminate]: 0.00084926
[jit_opt_b]: 6.088e-05, [1]
    [Cycle 1]: 4.62e-05, [2]
        [frontend_op_eliminate]: 1.791e-05
        [inline_after_opt_a]: 1.528e-05
[cconv]: 2.929e-05
[loop_unroll]: 0.00053674
[jit_opt_after_cconv]: 0.00015686, [1]
    [Cycle 1]: 0.00014863, [11]
        [c_1]: 2.214e-05
        [parameter_eliminate]: 3.10002e-06
        [updatestate_depend_eliminate]: 7.65998e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [cse]: 2.411e-05
        [call_graph_tuple_transform]: 2.045e-05
        [tuple_list_get_item_eliminator]: 5.39998e-06
        [none_parameter_eliminate]: 2.39999e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 5.51e-06
[remove_dup_value]: 1.474e-05
[partial_unused_args_eliminate]: 2.23002e-06
[environ_conv]: 5.84e-06
[add_recomputation]: 5.421e-05
[cse_after_recomputation]: 2.199e-05, [1]
    [Cycle 1]: 1.634e-05, [1]
        [cse]: 1.034e-05
[auto_monad_reorder]: 1.656e-05
[get_jit_bprop_graph]: 2.17999e-06
[rewriter_after_jit_bprop_graph]: 9.06002e-06
[opt_after_jit_grad]: 0.00052479
[symbol_engine_optimizer]: 8.949e-05, [1]
    [Cycle 1]: 8.161e-05, [6]
        [build]: 5.94999e-06
        [elim_shapecalc]: 9.69999e-06
        [elim_not_effective]: 1.643e-05
        [opt_reshape]: 6.93e-06
        [fold_const_symbol]: 9.59999e-06
        [renormalize]: 8.89995e-07
[validate]: 4.715e-05
Sums
    bootstrap                            :     0.000323s :  3.23%
    type_inference                       :     0.005031s : 50.25%
    event_method                         :     0.000019s :  0.19%
    auto_monad                           :     0.000066s :  0.66%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000105s :  1.05%
    rewriter_before_opt_a                :     0.000079s :  0.79%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000077s :  0.76%
    jit_opt_a.loop_unroll                :     0.000037s :  0.37%
    jit_opt_a.a_1                        :     0.000711s :  7.10%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.37%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000891s :  8.90%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.22%
    jit_opt_a.cse                        :     0.000045s :  0.45%
    jit_opt_a.replace_applicator         :     0.000019s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.17%
    rewriter_after_opt_a                 :     0.000040s :  0.40%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000849s :  8.48%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.15%
    cconv                                :     0.000029s :  0.29%
    loop_unroll                          :     0.000537s :  5.36%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000015s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000054s :  0.54%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000017s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.09%
    opt_after_jit_grad                   :     0.000525s :  5.24%
    symbol_engine_optimizer.build        :     0.000006s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000047s :  0.47%


Time group info:
------[substitution.]   0.000266    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.15% :     0.000006s :      3: substitution.graph_param_transform
85.65% :     0.000228s :      4: substitution.inline
 1.32% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.95% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.20% :     0.000006s :      2: substitution.replace_old_param
 5.37% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004956     2
81.98% :     0.004063s :      1: type_inference.infer
18.02% :     0.000893s :      1: type_inference.specialize
------[replace.]   0.000077     6
67.98% :     0.000052s :      4: replace.inline
32.02% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000238     6
94.50% :     0.000225s :      4: match.inline
 5.50% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.19% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.15% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000002s :     10: predicate.addn_zero_filter
 2.41% :     0.000003s :     10: predicate.arithmetic_simplify
 1.30% :     0.000002s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.84% :     0.000002s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.05% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.04% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.55% :     0.000001s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.30% :     0.000002s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.87% :     0.000008s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.70% :     0.000002s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.87% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.75% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.27% :     0.000002s :     10: predicate.minmaximum_grad
 2.35% :     0.000003s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.73% :     0.000002s :     10: predicate.reduce_eliminate
 1.35% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.14% :     0.000001s :     10: predicate.reshape_eliminate
 1.09% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.20% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.53% :     0.000011s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.02% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.79% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000674    12
48.72% :     0.000328s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.87% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.42% :     0.000292s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016903    72
 0.34% :     0.000057s :      1: add_recomputation
 0.41% :     0.000069s :      1: auto_monad
 0.11% :     0.000019s :      1: auto_monad_reorder
 2.04% :     0.000345s :      1: bootstrap
 0.19% :     0.000032s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000024s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
39.58% :     0.006691s :      1: jit_opt_a
 0.94% :     0.000160s :      1: jit_opt_after_cconv
 0.38% :     0.000064s :      1: jit_opt_b
 3.24% :     0.000547s :      1: loop_unroll
 5.08% :     0.000859s :      1: mutable_eliminate
 5.62% :     0.000949s :     26: opt.transform.jit_opt_a
 0.29% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000027s :      4: opt.transform.jit_opt_b
 0.09% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000038s :      4: opt.transform.symbol_engine_opt
 3.16% :     0.000534s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.64% :     0.000109s :      1: py_interpret_to_execute
 0.11% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.98% :     0.000504s :      1: renormalize.infer
 2.23% :     0.000377s :      1: renormalize.specialize
 0.07% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000044s :      1: rewriter_after_opt_a
 0.48% :     0.000081s :      1: rewriter_before_opt_a
 0.55% :     0.000092s :      1: symbol_engine_optimizer
29.87% :     0.005048s :      1: type_inference


TotalTime = 0.0232036, [33]
[bootstrap]: 0.00035992
[type_inference]: 0.00530884
[event_method]: 2.154e-05
[auto_monad]: 6.588e-05
[graph_reusing]: 5.66e-06
[pre_auto_parallel]: 2.77002e-06
[py_interpret_to_execute]: 5.021e-05
[rewriter_before_opt_a]: 0.00012388
[expand_dump_flag]: 3.68e-06
[jit_opt_a]: 0.00608738, [2]
    [Cycle 1]: 0.00184906, [27]
        [switch_simplify]: 5.719e-05
        [loop_unroll]: 3.017e-05
        [a_1]: 0.00055941
        [with_stream_mark]: 1.773e-05
        [recompute_prepare]: 7.40003e-06
        [updatestate_depend_eliminate]: 3.73001e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 2.13998e-06
        [parameter_eliminate]: 1.39998e-06
        [specialize_transform]: 5.89999e-06
        [updatestate_useless_node_eliminater]: 4.86997e-06
        [accelerated_algorithm]: 5.77001e-06
        [meta_shard_fg_expand]: 2.12999e-06
        [get_grad_eliminate_]: 5.54e-06
        [merge_forward]: 2.89001e-06
        [cell_reuse_recompute_pass]: 1.12999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.333e-05
        [j_node_and_user_rematch]: 9.66e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 1.007e-05
        [inline_without_move]: 5.62001e-06
        [renormalize]: 0.00085745
        [add_forward_monad_depend]: 5.54e-06
        [auto_monad_grad]: 1.69e-06
        [auto_monad_eliminator]: 1.228e-05
        [cse]: 2.085e-05
        [replace_applicator]: 1.342e-05
    [Cycle 2]: 0.00033989, [27]
        [switch_simplify]: 6.32001e-06
        [loop_unroll]: 4.65001e-06
        [a_1]: 9.334e-05
        [with_stream_mark]: 1.149e-05
        [recompute_prepare]: 5.34998e-06
        [updatestate_depend_eliminate]: 3.11001e-06
        [updatestate_assign_eliminate]: 2.14999e-06
        [updatestate_loads_eliminate]: 2.19999e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 5.19e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 1.52999e-06
        [get_grad_eliminate_]: 4.33001e-06
        [merge_forward]: 2.92002e-06
        [cell_reuse_recompute_pass]: 2.07999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.543e-05
        [j_node_and_user_rematch]: 8.27998e-06
        [meta_fg_expand]: 1.74998e-06
        [replace_old_param]: 7.78001e-06
        [inline_without_move]: 4.58001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 6.82002e-06
        [cse]: 1.214e-05
        [replace_applicator]: 1.312e-05
[py_interpret_to_execute_after_opt_a]: 1.14e-05
[rewriter_after_opt_a]: 3.138e-05
[convert_after_rewriter]: 5.84e-06
[order_py_execute_after_rewriter]: 4.25999e-06
[mutable_eliminate]: 0.00070693
[jit_opt_b]: 6.219e-05, [1]
    [Cycle 1]: 5.422e-05, [2]
        [frontend_op_eliminate]: 2.662e-05
        [inline_after_opt_a]: 1.512e-05
[cconv]: 2.372e-05
[loop_unroll]: 0.00053229
[jit_opt_after_cconv]: 0.00021783, [1]
    [Cycle 1]: 0.000209, [11]
        [c_1]: 2.468e-05
        [parameter_eliminate]: 4.72998e-06
        [updatestate_depend_eliminate]: 7.08998e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.48998e-06
        [cse]: 3.073e-05
        [call_graph_tuple_transform]: 2.157e-05
        [tuple_list_get_item_eliminator]: 7.98999e-06
        [none_parameter_eliminate]: 1.34e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 6.24001e-06
[remove_dup_value]: 1.279e-05
[partial_unused_args_eliminate]: 1.94999e-06
[environ_conv]: 5.23002e-06
[add_recomputation]: 4.666e-05
[cse_after_recomputation]: 2.914e-05, [1]
    [Cycle 1]: 2.189e-05, [1]
        [cse]: 1.318e-05
[auto_monad_reorder]: 1.36e-05
[get_jit_bprop_graph]: 1.94e-06
[rewriter_after_jit_bprop_graph]: 1.045e-05
[opt_after_jit_grad]: 0.00066065
[symbol_engine_optimizer]: 8.3e-05, [1]
    [Cycle 1]: 7.46e-05, [6]
        [build]: 5.52001e-06
        [elim_shapecalc]: 7.71001e-06
        [elim_not_effective]: 1.611e-05
        [opt_reshape]: 6.60997e-06
        [fold_const_symbol]: 8.88002e-06
        [renormalize]: 3.39991e-07
[validate]: 4.074e-05
[backend_pass]: 1.26997e-06
[task_emit]: 0.00841983
[execute]: 7.86001e-06
Sums
    bootstrap                            :     0.000360s :  1.94%
    type_inference                       :     0.005309s : 28.57%
    event_method                         :     0.000022s :  0.12%
    auto_monad                           :     0.000066s :  0.35%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000050s :  0.27%
    rewriter_before_opt_a                :     0.000124s :  0.67%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.34%
    jit_opt_a.loop_unroll                :     0.000035s :  0.19%
    jit_opt_a.a_1                        :     0.000653s :  3.51%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000858s :  4.62%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.10%
    jit_opt_a.cse                        :     0.000033s :  0.18%
    jit_opt_a.replace_applicator         :     0.000027s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.17%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000707s :  3.80%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000024s :  0.13%
    loop_unroll                          :     0.000532s :  2.86%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000031s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000013s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000047s :  0.25%
    cse_after_recomputation.cse          :     0.000013s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.06%
    opt_after_jit_grad                   :     0.000661s :  3.56%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000041s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008420s : 45.32%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000216    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.63% :     0.000006s :      3: substitution.graph_param_transform
84.63% :     0.000183s :      4: substitution.inline
 1.71% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.61% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.16% :     0.000005s :      2: substitution.replace_old_param
 4.70% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005210     2
80.09% :     0.004172s :      1: type_inference.infer
19.91% :     0.001037s :      1: type_inference.specialize
------[replace.]   0.000073     6
68.19% :     0.000050s :      4: replace.inline
31.81% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000189     6
95.27% :     0.000180s :      4: match.inline
 4.73% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.73% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 2.53% :     0.000003s :     10: predicate.arithmetic_simplify
 1.07% :     0.000001s :     10: predicate.cast_eliminate
 0.59% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.16% :     0.000001s :     10: predicate.depend_value_elim
 1.40% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.53% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.04% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000001s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.06% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.24% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.47% :     0.000001s :      3: predicate.graph_param_transform
 5.00% :     0.000006s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 2.22% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.66% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 2.38% :     0.000003s :      3: predicate.mutable_eliminate
 0.65% :     0.000001s :      3: predicate.opt_reshape
 1.95% :     0.000002s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.16% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.55% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.87% :     0.000004s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.83% :     0.000010s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.66% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.00% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.92% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000692    12
43.85% :     0.000303s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.01% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.14% :     0.000354s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025074    76
 0.20% :     0.000051s :      1: add_recomputation
 0.28% :     0.000070s :      1: auto_monad
 0.07% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.50% :     0.000377s :      1: bootstrap
 0.11% :     0.000026s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.13% :     0.000031s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000026s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
24.29% :     0.006092s :      1: jit_opt_a
 0.88% :     0.000221s :      1: jit_opt_after_cconv
 0.26% :     0.000065s :      1: jit_opt_b
 2.18% :     0.000546s :      1: loop_unroll
 2.86% :     0.000718s :      1: mutable_eliminate
 3.51% :     0.000880s :     26: opt.transform.jit_opt_a
 0.22% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000035s :      4: opt.transform.jit_opt_b
 0.07% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.68% :     0.000672s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000053s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000015s :      1: remove_dup_value
 1.81% :     0.000454s :      1: renormalize.infer
 1.58% :     0.000395s :      1: renormalize.specialize
 0.05% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000034s :      1: rewriter_after_opt_a
 0.52% :     0.000129s :      1: rewriter_before_opt_a
 0.34% :     0.000086s :      1: symbol_engine_optimizer
33.66% :     0.008439s :      1: task_emit
21.26% :     0.005332s :      1: type_inference
 0.25% :     0.000061s :      1: validate


TotalTime = 0.0350381, [33]
[bootstrap]: 0.00021327
[type_inference]: 0.0122271
[event_method]: 6.648e-05
[auto_monad]: 9.53e-05
[graph_reusing]: 6.17999e-06
[pre_auto_parallel]: 2.56e-06
[py_interpret_to_execute]: 2.618e-05
[rewriter_before_opt_a]: 9.871e-05
[expand_dump_flag]: 2.76e-06
[jit_opt_a]: 0.0112609, [3]
    [Cycle 1]: 0.0056362, [27]
        [switch_simplify]: 0.00010428
        [loop_unroll]: 4.511e-05
        [a_1]: 0.00085812
        [with_stream_mark]: 2.218e-05
        [recompute_prepare]: 1.682e-05
        [updatestate_depend_eliminate]: 7.58001e-06
        [updatestate_assign_eliminate]: 5.88998e-06
        [updatestate_loads_eliminate]: 5.62999e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 1.274e-05
        [updatestate_useless_node_eliminater]: 1.219e-05
        [accelerated_algorithm]: 1.202e-05
        [meta_shard_fg_expand]: 2.93e-06
        [get_grad_eliminate_]: 1.19e-05
        [merge_forward]: 7.40998e-06
        [cell_reuse_recompute_pass]: 7.39994e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.343e-05
        [j_node_and_user_rematch]: 2.224e-05
        [meta_fg_expand]: 0.00141812
        [replace_old_param]: 5.2e-05
        [inline_without_move]: 5.28e-05
        [renormalize]: 0.00247057
        [add_forward_monad_depend]: 1.077e-05
        [auto_monad_grad]: 6.40997e-06
        [auto_monad_eliminator]: 4.831e-05
        [cse]: 0.00014258
        [replace_applicator]: 7.347e-05
    [Cycle 2]: 0.00213805, [27]
        [switch_simplify]: 3.78e-05
        [loop_unroll]: 3.409e-05
        [a_1]: 0.00104638
        [with_stream_mark]: 1.485e-05
        [recompute_prepare]: 8.02e-06
        [updatestate_depend_eliminate]: 4.01001e-06
        [updatestate_assign_eliminate]: 3.4e-06
        [updatestate_loads_eliminate]: 2.93998e-06
        [parameter_eliminate]: 2.00002e-06
        [specialize_transform]: 6.33e-06
        [updatestate_useless_node_eliminater]: 6.36e-06
        [accelerated_algorithm]: 6.79001e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 6.28e-06
        [merge_forward]: 3.24001e-06
        [cell_reuse_recompute_pass]: 9.5999e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.631e-05
        [j_node_and_user_rematch]: 1.041e-05
        [meta_fg_expand]: 8.472e-05
        [replace_old_param]: 1.229e-05
        [inline_without_move]: 6.10002e-06
        [renormalize]: 0.00060244
        [add_forward_monad_depend]: 5.20001e-06
        [auto_monad_grad]: 1.86998e-06
        [auto_monad_eliminator]: 1.414e-05
        [cse]: 2.472e-05
        [replace_applicator]: 1.469e-05
    [Cycle 3]: 0.00037164, [27]
        [switch_simplify]: 6.90002e-06
        [loop_unroll]: 6.01e-06
        [a_1]: 0.00012084
        [with_stream_mark]: 1.053e-05
        [recompute_prepare]: 6.57002e-06
        [updatestate_depend_eliminate]: 3.48999e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 6.04001e-06
        [updatestate_useless_node_eliminater]: 5.88002e-06
        [accelerated_algorithm]: 5.92001e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 5.79e-06
        [merge_forward]: 2.86999e-06
        [cell_reuse_recompute_pass]: 1.96e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.545e-05
        [j_node_and_user_rematch]: 1.026e-05
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 8.37998e-06
        [inline_without_move]: 5.67001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 7.2e-06
        [cse]: 1.371e-05
        [replace_applicator]: 5.82999e-06
[py_interpret_to_execute_after_opt_a]: 1.254e-05
[rewriter_after_opt_a]: 3.991e-05
[convert_after_rewriter]: 8.82999e-06
[order_py_execute_after_rewriter]: 5.74e-06
[mutable_eliminate]: 0.00065561
[jit_opt_b]: 5.632e-05, [1]
    [Cycle 1]: 4.895e-05, [2]
        [frontend_op_eliminate]: 1.98e-05
        [inline_after_opt_a]: 1.722e-05
[cconv]: 2.523e-05
[loop_unroll]: 0.00042466
[jit_opt_after_cconv]: 0.00016095, [1]
    [Cycle 1]: 0.00015427, [11]
        [c_1]: 2.686e-05
        [parameter_eliminate]: 3.2e-06
        [updatestate_depend_eliminate]: 6.51999e-06
        [updatestate_assign_eliminate]: 3.53999e-06
        [updatestate_loads_eliminate]: 2.74999e-06
        [cse]: 2.526e-05
        [call_graph_tuple_transform]: 2.006e-05
        [tuple_list_get_item_eliminator]: 6.90998e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 4.90021e-07
        [switch_simplify]: 7.19001e-06
[remove_dup_value]: 1.596e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 6.02999e-06
[add_recomputation]: 5.113e-05
[cse_after_recomputation]: 2.5e-05, [1]
    [Cycle 1]: 1.939e-05, [1]
        [cse]: 1.354e-05
[auto_monad_reorder]: 2.647e-05
[get_jit_bprop_graph]: 2.02001e-06
[rewriter_after_jit_bprop_graph]: 5.61998e-06
[opt_after_jit_grad]: 0.00045559
[symbol_engine_optimizer]: 8.046e-05, [1]
    [Cycle 1]: 7.451e-05, [6]
        [build]: 3.3e-06
        [elim_shapecalc]: 9.19e-06
        [elim_not_effective]: 1.647e-05
        [opt_reshape]: 6.83e-06
        [fold_const_symbol]: 1.054e-05
        [renormalize]: 4.69998e-07
[validate]: 3.73e-05
[backend_pass]: 9.79984e-07
[task_emit]: 0.00873593
[execute]: 7.98001e-06
Sums
    bootstrap                            :     0.000213s :  0.68%
    type_inference                       :     0.012227s : 39.26%
    event_method                         :     0.000066s :  0.21%
    auto_monad                           :     0.000095s :  0.31%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.08%
    rewriter_before_opt_a                :     0.000099s :  0.32%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000149s :  0.48%
    jit_opt_a.loop_unroll                :     0.000085s :  0.27%
    jit_opt_a.a_1                        :     0.002025s :  6.50%
    jit_opt_a.with_stream_mark           :     0.000048s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000031s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000025s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000025s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.08%
    jit_opt_a.merge_forward              :     0.000014s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000055s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000043s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.001505s :  4.83%
    jit_opt_a.replace_old_param          :     0.000073s :  0.23%
    jit_opt_a.inline_without_move        :     0.000065s :  0.21%
    jit_opt_a.renormalize                :     0.003073s :  9.87%
    jit_opt_a.add_forward_monad_depend   :     0.000017s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000070s :  0.22%
    jit_opt_a.cse                        :     0.000181s :  0.58%
    jit_opt_a.replace_applicator         :     0.000094s :  0.30%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.04%
    rewriter_after_opt_a                 :     0.000040s :  0.13%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000656s :  2.10%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.06%
    cconv                                :     0.000025s :  0.08%
    loop_unroll                          :     0.000425s :  1.36%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000016s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000051s :  0.16%
    cse_after_recomputation.cse          :     0.000014s :  0.04%
    auto_monad_reorder                   :     0.000026s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000456s :  1.46%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008736s : 28.05%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000480   113
 0.58% :     0.000003s :      3: substitution.elim_not_effective
 0.33% :     0.000002s :      3: substitution.fold_const_symbol
 1.14% :     0.000005s :      4: substitution.graph_param_transform
61.98% :     0.000297s :     14: substitution.inline
 3.52% :     0.000017s :      2: substitution.inline_without_move
 1.93% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.40% :     0.000012s :      7: substitution.minmaximum_grad
 1.26% :     0.000006s :      5: substitution.partial_eliminate
 2.25% :     0.000011s :     14: substitution.remove_not_recompute_node
 6.08% :     0.000029s :     10: substitution.replace_applicator
 1.54% :     0.000007s :      7: substitution.replace_old_param
 0.49% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.47% :     0.000007s :      1: substitution.switch_simplify
 4.05% :     0.000019s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.84% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 8.13% :     0.000039s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012161     2
89.32% :     0.010862s :      1: type_inference.infer
10.68% :     0.001299s :      1: type_inference.specialize
------[replace.]   0.000171    22
56.21% :     0.000096s :     14: replace.inline
14.02% :     0.000024s :      1: replace.switch_simplify
29.77% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000314    22
92.05% :     0.000289s :     14: match.inline
 2.06% :     0.000006s :      1: match.switch_simplify
 5.90% :     0.000019s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000384  2634
 1.56% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.40% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.37% :     0.000005s :     44: predicate.addn_check_dump
 1.77% :     0.000007s :     44: predicate.addn_zero_filter
 1.93% :     0.000007s :     44: predicate.arithmetic_simplify
 1.52% :     0.000006s :     44: predicate.cast_eliminate
 0.24% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.40% :     0.000005s :     44: predicate.compare_switch_simplify
 1.54% :     0.000006s :     44: predicate.depend_value_elim
 1.48% :     0.000006s :     44: predicate.dict_get_item_const_eliminator
 1.42% :     0.000005s :     44: predicate.dict_get_item_eliminator
 1.62% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.26% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      4: predicate.elim_not_effective
 0.24% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.46% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.39% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.38% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.61% :     0.000006s :     44: predicate.environ_get_eliminate
 1.47% :     0.000006s :     44: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.91% :     0.000003s :     19: predicate.get_grad_eliminate
 0.13% :     0.000000s :      4: predicate.graph_param_transform
 4.24% :     0.000016s :     73: predicate.inline
 1.74% :     0.000007s :     40: predicate.inline_without_move
 0.36% :     0.000001s :     19: predicate.j_node_and_user_rematch
 1.01% :     0.000004s :     19: predicate.less_batch_normalization
 1.89% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.87% :     0.000007s :     55: predicate.load_eliminater
 0.53% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.41% :     0.000013s :     91: predicate.loop_unroll_before_grad
 1.76% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.48% :     0.000006s :     44: predicate.merge_addn
 1.55% :     0.000006s :     44: predicate.minmaximum_grad
 0.57% :     0.000002s :      4: predicate.mutable_eliminate
 0.17% :     0.000001s :      4: predicate.opt_reshape
 2.25% :     0.000009s :     55: predicate.partial_eliminate
 1.47% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.93% :     0.000007s :     44: predicate.reduce_eliminate
 1.78% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.56% :     0.000010s :     97: predicate.replace_applicator
 1.11% :     0.000004s :     40: predicate.replace_old_param
 0.11% :     0.000000s :      4: predicate.reset_defer_inline
 1.66% :     0.000006s :     44: predicate.reshape_eliminate
 1.56% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.35% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.57% :     0.000006s :     44: predicate.same_eliminate
 0.47% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.37% :     0.000001s :      8: predicate.special_op_eliminate
 0.85% :     0.000003s :     19: predicate.specialize_transform
 1.84% :     0.000007s :     44: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.81% :     0.000011s :     65: predicate.switch_defer_inline
 2.58% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.53% :     0.000025s :    162: predicate.switch_simplify
 1.43% :     0.000005s :     44: predicate.tile_eliminate
 1.55% :     0.000006s :     44: predicate.transpose_eliminate
 1.80% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.67% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 3.08% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 1.92% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.88% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.98% :     0.000008s :     44: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.23% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001619    32
59.46% :     0.000963s :     14: func_graph_cloner_run.FuncGraphClonerGraph
40.54% :     0.000656s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.040916    91
 0.13% :     0.000054s :      1: add_recomputation
 0.24% :     0.000099s :      1: auto_monad
 0.07% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.56% :     0.000228s :      1: bootstrap
 0.07% :     0.000028s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.07% :     0.000027s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.17% :     0.000071s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
27.53% :     0.011264s :      1: jit_opt_a
 0.40% :     0.000164s :      1: jit_opt_after_cconv
 0.14% :     0.000059s :      1: jit_opt_b
 1.06% :     0.000432s :      1: loop_unroll
 1.62% :     0.000664s :      1: mutable_eliminate
 6.54% :     0.002676s :     39: opt.transform.jit_opt_a
 0.14% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000031s :      4: opt.transform.jit_opt_b
 0.03% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000040s :      4: opt.transform.symbol_engine_opt
 1.13% :     0.000463s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000029s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 4.10% :     0.001678s :      2: renormalize.infer
 3.38% :     0.001381s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000043s :      1: rewriter_after_opt_a
 0.25% :     0.000101s :      1: rewriter_before_opt_a
 0.20% :     0.000083s :      1: symbol_engine_optimizer
21.39% :     0.008752s :      1: task_emit
29.91% :     0.012239s :      1: type_inference
 0.14% :     0.000058s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.705.071 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6c072dc0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.705.808 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0199292, [33]
[bootstrap]: 0.00038285
[type_inference]: 0.00484241
[event_method]: 1.802e-05
[auto_monad]: 6.385e-05
[graph_reusing]: 5.89999e-06
[pre_auto_parallel]: 2.59001e-06
[py_interpret_to_execute]: 3.182e-05
[rewriter_before_opt_a]: 7.187e-05
[expand_dump_flag]: 2.99001e-06
[jit_opt_a]: 0.00598581, [2]
    [Cycle 1]: 0.0016039, [27]
        [switch_simplify]: 5.606e-05
        [loop_unroll]: 2.868e-05
        [a_1]: 0.00053721
        [with_stream_mark]: 2.202e-05
        [recompute_prepare]: 7.08e-06
        [updatestate_depend_eliminate]: 3.28e-06
        [updatestate_assign_eliminate]: 2.76e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [parameter_eliminate]: 2.00002e-06
        [specialize_transform]: 5.86e-06
        [updatestate_useless_node_eliminater]: 4.72e-06
        [accelerated_algorithm]: 5.32999e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 4.88001e-06
        [merge_forward]: 3.36999e-06
        [cell_reuse_recompute_pass]: 1.45001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.294e-05
        [j_node_and_user_rematch]: 8.42e-06
        [meta_fg_expand]: 2.04999e-06
        [replace_old_param]: 8.72e-06
        [inline_without_move]: 4.90999e-06
        [renormalize]: 0.00064311
        [add_forward_monad_depend]: 5.24998e-06
        [auto_monad_grad]: 2.39001e-06
        [auto_monad_eliminator]: 1.39e-05
        [cse]: 2.807e-05
        [replace_applicator]: 1.199e-05
    [Cycle 2]: 0.00031099, [27]
        [switch_simplify]: 5.37001e-06
        [loop_unroll]: 4.77e-06
        [a_1]: 8.684e-05
        [with_stream_mark]: 8.56002e-06
        [recompute_prepare]: 4.72e-06
        [updatestate_depend_eliminate]: 2.44001e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 9.60019e-07
        [specialize_transform]: 4.57998e-06
        [updatestate_useless_node_eliminater]: 4.58999e-06
        [accelerated_algorithm]: 4.97e-06
        [meta_shard_fg_expand]: 1.09998e-06
        [get_grad_eliminate_]: 4.70001e-06
        [merge_forward]: 2.15002e-06
        [cell_reuse_recompute_pass]: 2.24001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.201e-05
        [j_node_and_user_rematch]: 7.67998e-06
        [meta_fg_expand]: 1.42e-06
        [replace_old_param]: 7.26001e-06
        [inline_without_move]: 4.57e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.09998e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 4.88001e-06
        [cse]: 1.013e-05
        [replace_applicator]: 4.95001e-06
[py_interpret_to_execute_after_opt_a]: 1.02e-05
[rewriter_after_opt_a]: 3.578e-05
[convert_after_rewriter]: 7.15998e-06
[order_py_execute_after_rewriter]: 5.00001e-06
[mutable_eliminate]: 0.00057552
[jit_opt_b]: 4.872e-05, [1]
    [Cycle 1]: 4.112e-05, [2]
        [frontend_op_eliminate]: 1.522e-05
        [inline_after_opt_a]: 1.397e-05
[cconv]: 2.433e-05
[loop_unroll]: 0.00041351
[jit_opt_after_cconv]: 0.0001338, [1]
    [Cycle 1]: 0.00012747, [11]
        [c_1]: 1.963e-05
        [parameter_eliminate]: 3.04001e-06
        [updatestate_depend_eliminate]: 4.93001e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.692e-05
        [call_graph_tuple_transform]: 1.743e-05
        [tuple_list_get_item_eliminator]: 4.92999e-06
        [none_parameter_eliminate]: 1.71e-06
        [renormalize]: 5.59987e-07
        [switch_simplify]: 5.23002e-06
[remove_dup_value]: 1.264e-05
[partial_unused_args_eliminate]: 2.06998e-06
[environ_conv]: 4.68999e-06
[add_recomputation]: 4.79e-05
[cse_after_recomputation]: 1.948e-05, [1]
    [Cycle 1]: 1.42e-05, [1]
        [cse]: 8.50999e-06
[auto_monad_reorder]: 1.455e-05
[get_jit_bprop_graph]: 1.67001e-06
[rewriter_after_jit_bprop_graph]: 4.98001e-06
[opt_after_jit_grad]: 0.00044898
[symbol_engine_optimizer]: 7.004e-05, [1]
    [Cycle 1]: 6.37e-05, [6]
        [build]: 3.56999e-06
        [elim_shapecalc]: 7.9e-06
        [elim_not_effective]: 1.155e-05
        [opt_reshape]: 5.63002e-06
        [fold_const_symbol]: 8.05e-06
        [renormalize]: 4.39992e-07
[validate]: 3.651e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00639988
[execute]: 8.48001e-06
Sums
    bootstrap                            :     0.000383s :  2.51%
    type_inference                       :     0.004842s : 31.74%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000064s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.21%
    rewriter_before_opt_a                :     0.000072s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.22%
    jit_opt_a.a_1                        :     0.000624s :  4.09%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000009s :  0.06%
    jit_opt_a.renormalize                :     0.000643s :  4.22%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000038s :  0.25%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000576s :  3.77%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000024s :  0.16%
    loop_unroll                          :     0.000414s :  2.71%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000048s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000449s :  2.94%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006400s : 41.94%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000217    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.34% :     0.000005s :      3: substitution.graph_param_transform
85.38% :     0.000186s :      4: substitution.inline
 1.41% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.58% :     0.000003s :      2: substitution.replace_old_param
 5.91% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004773     2
83.39% :     0.003980s :      1: type_inference.infer
16.61% :     0.000793s :      1: type_inference.specialize
------[replace.]   0.000061     6
68.02% :     0.000041s :      4: replace.inline
31.98% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000195     6
93.96% :     0.000183s :      4: match.inline
 6.04% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.51% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.49% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 2.09% :     0.000002s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.53% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.93% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.54% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.63% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000001s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.89% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.46% :     0.000002s :      3: predicate.mutable_eliminate
 0.79% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.47% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.02% :     0.000002s :     18: predicate.replace_applicator
 0.95% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.50% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 2.01% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.48% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.51% :     0.000002s :     10: predicate.transpose_eliminate
 1.27% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.74% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.27% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.64% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000560    12
49.57% :     0.000277s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.90% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.53% :     0.000238s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021496    76
 0.24% :     0.000051s :      1: add_recomputation
 0.31% :     0.000067s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.86% :     0.000400s :      1: bootstrap
 0.13% :     0.000027s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.86% :     0.005989s :      1: jit_opt_a
 0.64% :     0.000137s :      1: jit_opt_after_cconv
 0.24% :     0.000051s :      1: jit_opt_b
 1.96% :     0.000421s :      1: loop_unroll
 2.72% :     0.000584s :      1: mutable_eliminate
 3.84% :     0.000826s :     26: opt.transform.jit_opt_a
 0.21% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000023s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.13% :     0.000457s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000035s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 1.66% :     0.000357s :      1: renormalize.infer
 1.30% :     0.000279s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000039s :      1: rewriter_after_opt_a
 0.35% :     0.000075s :      1: rewriter_before_opt_a
 0.34% :     0.000072s :      1: symbol_engine_optimizer
29.84% :     0.006414s :      1: task_emit
22.61% :     0.004859s :      1: type_inference
 0.26% :     0.000057s :      1: validate


TotalTime = 0.0151977, [30]
[bootstrap]: 0.00039891
[type_inference]: 0.00559595
[event_method]: 2.178e-05
[auto_monad]: 6.576e-05
[graph_reusing]: 6.61e-06
[pre_auto_parallel]: 3.7e-06
[py_interpret_to_execute]: 3.81e-05
[rewriter_before_opt_a]: 7.548e-05
[expand_dump_flag]: 3.16999e-06
[jit_opt_a]: 0.00639286, [2]
    [Cycle 1]: 0.00183992, [27]
        [switch_simplify]: 6.316e-05
        [loop_unroll]: 3.285e-05
        [a_1]: 0.00057233
        [with_stream_mark]: 2.245e-05
        [recompute_prepare]: 9.76e-06
        [updatestate_depend_eliminate]: 3.98001e-06
        [updatestate_assign_eliminate]: 3.00002e-06
        [updatestate_loads_eliminate]: 2.78003e-06
        [parameter_eliminate]: 2.04999e-06
        [specialize_transform]: 7.65998e-06
        [updatestate_useless_node_eliminater]: 5.59e-06
        [accelerated_algorithm]: 6.58998e-06
        [meta_shard_fg_expand]: 2.48e-06
        [get_grad_eliminate_]: 6.36e-06
        [merge_forward]: 4.13999e-06
        [cell_reuse_recompute_pass]: 1.86e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.623e-05
        [j_node_and_user_rematch]: 1.031e-05
        [meta_fg_expand]: 2.78e-06
        [replace_old_param]: 9.25999e-06
        [inline_without_move]: 5.59e-06
        [renormalize]: 0.00076939
        [add_forward_monad_depend]: 9.45001e-06
        [auto_monad_grad]: 2.66e-06
        [auto_monad_eliminator]: 1.878e-05
        [cse]: 3.331e-05
        [replace_applicator]: 1.849e-05
    [Cycle 2]: 0.00037429, [27]
        [switch_simplify]: 5.92999e-06
        [loop_unroll]: 5.40001e-06
        [a_1]: 9.303e-05
        [with_stream_mark]: 1.465e-05
        [recompute_prepare]: 6.63e-06
        [updatestate_depend_eliminate]: 3.68e-06
        [updatestate_assign_eliminate]: 2.58998e-06
        [updatestate_loads_eliminate]: 2.28998e-06
        [parameter_eliminate]: 1.48002e-06
        [specialize_transform]: 4.96002e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 5.27001e-06
        [meta_shard_fg_expand]: 1.99999e-06
        [get_grad_eliminate_]: 4.72e-06
        [merge_forward]: 4.17e-06
        [cell_reuse_recompute_pass]: 3.8e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.959e-05
        [j_node_and_user_rematch]: 9.18002e-06
        [meta_fg_expand]: 1.80001e-06
        [replace_old_param]: 9.45001e-06
        [inline_without_move]: 5.47999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 2.90998e-06
        [auto_monad_grad]: 1.87001e-06
        [auto_monad_eliminator]: 8.93002e-06
        [cse]: 1.513e-05
        [replace_applicator]: 6.12999e-06
[py_interpret_to_execute_after_opt_a]: 1.611e-05
[rewriter_after_opt_a]: 4.257e-05
[convert_after_rewriter]: 7.68001e-06
[order_py_execute_after_rewriter]: 5.87999e-06
[mutable_eliminate]: 0.00076594
[jit_opt_b]: 6.002e-05, [1]
    [Cycle 1]: 5.054e-05, [2]
        [frontend_op_eliminate]: 1.832e-05
        [inline_after_opt_a]: 1.878e-05
[cconv]: 3.544e-05
[loop_unroll]: 0.0005178
[jit_opt_after_cconv]: 0.00016479, [1]
    [Cycle 1]: 0.00015708, [11]
        [c_1]: 2.236e-05
        [parameter_eliminate]: 3.92998e-06
        [updatestate_depend_eliminate]: 1.025e-05
        [updatestate_assign_eliminate]: 2.75002e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [cse]: 2.828e-05
        [call_graph_tuple_transform]: 2.291e-05
        [tuple_list_get_item_eliminator]: 5.26002e-06
        [none_parameter_eliminate]: 1.77001e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 5.66e-06
[remove_dup_value]: 1.513e-05
[partial_unused_args_eliminate]: 2.29999e-06
[environ_conv]: 5.96e-06
[add_recomputation]: 5.472e-05
[cse_after_recomputation]: 2.264e-05, [1]
    [Cycle 1]: 1.637e-05, [1]
        [cse]: 1.004e-05
[auto_monad_reorder]: 1.637e-05
[get_jit_bprop_graph]: 2.61999e-06
[rewriter_after_jit_bprop_graph]: 6.93e-06
[opt_after_jit_grad]: 0.00052099
[symbol_engine_optimizer]: 8.214e-05, [1]
    [Cycle 1]: 7.354e-05, [6]
        [build]: 4.67e-06
        [elim_shapecalc]: 9.03002e-06
        [elim_not_effective]: 1.605e-05
        [opt_reshape]: 5.69e-06
        [fold_const_symbol]: 8.77e-06
        [renormalize]: 6.19999e-07
[validate]: 4.147e-05
Sums
    bootstrap                            :     0.000399s :  3.85%
    type_inference                       :     0.005596s : 54.04%
    event_method                         :     0.000022s :  0.21%
    auto_monad                           :     0.000066s :  0.64%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000004s :  0.04%
    py_interpret_to_execute              :     0.000038s :  0.37%
    rewriter_before_opt_a                :     0.000075s :  0.73%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000069s :  0.67%
    jit_opt_a.loop_unroll                :     0.000038s :  0.37%
    jit_opt_a.a_1                        :     0.000665s :  6.43%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.36%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.18%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000769s :  7.43%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.12%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.27%
    jit_opt_a.cse                        :     0.000048s :  0.47%
    jit_opt_a.replace_applicator         :     0.000025s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.16%
    rewriter_after_opt_a                 :     0.000043s :  0.41%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000766s :  7.40%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.18%
    cconv                                :     0.000035s :  0.34%
    loop_unroll                          :     0.000518s :  5.00%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.10%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000028s :  0.27%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000015s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000055s :  0.53%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.16%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000521s :  5.03%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.40%


Time group info:
------[substitution.]   0.000222    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.66% :     0.000006s :      3: substitution.graph_param_transform
82.69% :     0.000184s :      4: substitution.inline
 1.83% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.41% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.71% :     0.000006s :      2: substitution.replace_old_param
 6.06% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005521     2
83.49% :     0.004609s :      1: type_inference.infer
16.51% :     0.000912s :      1: type_inference.specialize
------[replace.]   0.000076     6
67.81% :     0.000052s :      4: replace.inline
32.19% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000193     6
93.66% :     0.000181s :      4: match.inline
 6.34% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.61% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.84% :     0.000002s :     10: predicate.addn_zero_filter
 2.57% :     0.000003s :     10: predicate.arithmetic_simplify
 1.08% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.57% :     0.000002s :     10: predicate.depend_value_elim
 0.98% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.14% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.13% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      3: predicate.elim_not_effective
 0.76% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.43% :     0.000001s :      3: predicate.graph_param_transform
 6.30% :     0.000008s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000002s :      6: predicate.less_batch_normalization
 1.33% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.69% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.36% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.09% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 2.35% :     0.000003s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.46% :     0.000003s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.89% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.52% :     0.000001s :      3: predicate.reset_defer_inline
 1.24% :     0.000002s :     10: predicate.reshape_eliminate
 1.40% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.07% :     0.000001s :     10: predicate.same_eliminate
 0.76% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.21% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.55% :     0.000003s :     16: predicate.switch_defer_inline
 1.92% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.49% :     0.000010s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.18% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.31% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.44% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.41% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000676    12
52.31% :     0.000353s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.93% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.75% :     0.000289s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016953    72
 0.34% :     0.000058s :      1: add_recomputation
 0.41% :     0.000069s :      1: auto_monad
 0.11% :     0.000019s :      1: auto_monad_reorder
 2.49% :     0.000422s :      1: bootstrap
 0.23% :     0.000039s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.15% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
37.73% :     0.006397s :      1: jit_opt_a
 0.99% :     0.000168s :      1: jit_opt_after_cconv
 0.37% :     0.000063s :      1: jit_opt_b
 3.12% :     0.000529s :      1: loop_unroll
 4.62% :     0.000783s :      1: mutable_eliminate
 5.38% :     0.000912s :     26: opt.transform.jit_opt_a
 0.31% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000029s :      4: opt.transform.jit_opt_b
 0.10% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.13% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.14% :     0.000532s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000006s :      1: pre_auto_parallel
 0.24% :     0.000041s :      1: py_interpret_to_execute
 0.11% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000018s :      1: remove_dup_value
 2.47% :     0.000419s :      1: renormalize.infer
 2.02% :     0.000342s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000047s :      1: rewriter_after_opt_a
 0.46% :     0.000079s :      1: rewriter_before_opt_a
 0.50% :     0.000085s :      1: symbol_engine_optimizer
33.12% :     0.005614s :      1: type_inference


TotalTime = 0.021488, [33]
[bootstrap]: 0.00032006
[type_inference]: 0.00530013
[event_method]: 1.747e-05
[auto_monad]: 6.516e-05
[graph_reusing]: 5.48002e-06
[pre_auto_parallel]: 3.13e-06
[py_interpret_to_execute]: 4.146e-05
[rewriter_before_opt_a]: 7.743e-05
[expand_dump_flag]: 3.09999e-06
[jit_opt_a]: 0.00603645, [2]
    [Cycle 1]: 0.00192242, [27]
        [switch_simplify]: 5.777e-05
        [loop_unroll]: 2.942e-05
        [a_1]: 0.00061289
        [with_stream_mark]: 1.86e-05
        [recompute_prepare]: 1.025e-05
        [updatestate_depend_eliminate]: 3.66001e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 1.96998e-06
        [parameter_eliminate]: 1.60999e-06
        [specialize_transform]: 6.49999e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 5.75001e-06
        [meta_shard_fg_expand]: 2.11e-06
        [get_grad_eliminate_]: 5.10001e-06
        [merge_forward]: 3.35003e-06
        [cell_reuse_recompute_pass]: 1.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.325e-05
        [j_node_and_user_rematch]: 1.004e-05
        [meta_fg_expand]: 2.51e-06
        [replace_old_param]: 9.63002e-06
        [inline_without_move]: 5.30999e-06
        [renormalize]: 0.00087351
        [add_forward_monad_depend]: 5.94999e-06
        [auto_monad_grad]: 2.74999e-06
        [auto_monad_eliminator]: 1.33e-05
        [cse]: 2.267e-05
        [replace_applicator]: 1.591e-05
    [Cycle 2]: 0.00034232, [27]
        [switch_simplify]: 5.90002e-06
        [loop_unroll]: 5.46002e-06
        [a_1]: 9.81e-05
        [with_stream_mark]: 1.219e-05
        [recompute_prepare]: 5.11997e-06
        [updatestate_depend_eliminate]: 2.98e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.75001e-06
        [specialize_transform]: 5.52001e-06
        [updatestate_useless_node_eliminater]: 4.77998e-06
        [accelerated_algorithm]: 5.29e-06
        [meta_shard_fg_expand]: 1.40999e-06
        [get_grad_eliminate_]: 4.67e-06
        [merge_forward]: 3.09001e-06
        [cell_reuse_recompute_pass]: 2.05002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.49e-05
        [j_node_and_user_rematch]: 8.07e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 7.83999e-06
        [inline_without_move]: 4.56002e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.65001e-06
        [auto_monad_grad]: 1.37999e-06
        [auto_monad_eliminator]: 5.44e-06
        [cse]: 1.108e-05
        [replace_applicator]: 5.40999e-06
[py_interpret_to_execute_after_opt_a]: 1.307e-05
[rewriter_after_opt_a]: 3.532e-05
[convert_after_rewriter]: 6.94999e-06
[order_py_execute_after_rewriter]: 4.73001e-06
[mutable_eliminate]: 0.00071158
[jit_opt_b]: 5.532e-05, [1]
    [Cycle 1]: 4.748e-05, [2]
        [frontend_op_eliminate]: 1.817e-05
        [inline_after_opt_a]: 1.747e-05
[cconv]: 3.05e-05
[loop_unroll]: 0.00051219
[jit_opt_after_cconv]: 0.00016554, [1]
    [Cycle 1]: 0.0001579, [11]
        [c_1]: 2.311e-05
        [parameter_eliminate]: 4.90999e-06
        [updatestate_depend_eliminate]: 6.56999e-06
        [updatestate_assign_eliminate]: 3.17002e-06
        [updatestate_loads_eliminate]: 2.84001e-06
        [cse]: 2.513e-05
        [call_graph_tuple_transform]: 2.263e-05
        [tuple_list_get_item_eliminator]: 5.89e-06
        [none_parameter_eliminate]: 2.07999e-06
        [renormalize]: 4.7998e-07
        [switch_simplify]: 6.16998e-06
[remove_dup_value]: 6.66e-05
[partial_unused_args_eliminate]: 1.94e-06
[environ_conv]: 5.27001e-06
[add_recomputation]: 5.039e-05
[cse_after_recomputation]: 2.504e-05, [1]
    [Cycle 1]: 1.768e-05, [1]
        [cse]: 1.102e-05
[auto_monad_reorder]: 1.474e-05
[get_jit_bprop_graph]: 1.94999e-06
[rewriter_after_jit_bprop_graph]: 8.54998e-06
[opt_after_jit_grad]: 0.00060586
[symbol_engine_optimizer]: 8.653e-05, [1]
    [Cycle 1]: 7.837e-05, [6]
        [build]: 3.97998e-06
        [elim_shapecalc]: 9.67999e-06
        [elim_not_effective]: 1.768e-05
        [opt_reshape]: 5.91998e-06
        [fold_const_symbol]: 9.56e-06
        [renormalize]: 1.12e-06
[validate]: 4.207e-05
[backend_pass]: 1.20999e-06
[task_emit]: 0.00692932
[execute]: 7.57998e-06
Sums
    bootstrap                            :     0.000320s :  1.88%
    type_inference                       :     0.005300s : 31.09%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000065s :  0.38%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000041s :  0.24%
    rewriter_before_opt_a                :     0.000077s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.37%
    jit_opt_a.loop_unroll                :     0.000035s :  0.20%
    jit_opt_a.a_1                        :     0.000711s :  4.17%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000874s :  5.12%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.11%
    jit_opt_a.cse                        :     0.000034s :  0.20%
    jit_opt_a.replace_applicator         :     0.000021s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000035s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000712s :  4.17%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.10%
    cconv                                :     0.000030s :  0.18%
    loop_unroll                          :     0.000512s :  3.00%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000025s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000067s :  0.39%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.30%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000606s :  3.55%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000042s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006929s : 40.65%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000253    23
 1.26% :     0.000003s :      2: substitution.elim_not_effective
 0.61% :     0.000002s :      2: substitution.fold_const_symbol
 2.26% :     0.000006s :      3: substitution.graph_param_transform
85.26% :     0.000216s :      4: substitution.inline
 1.41% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.75% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.12% :     0.000005s :      2: substitution.replace_old_param
 5.32% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005227     2
80.24% :     0.004194s :      1: type_inference.infer
19.76% :     0.001033s :      1: type_inference.specialize
------[replace.]   0.000086     6
62.58% :     0.000054s :      4: replace.inline
37.42% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000224     6
94.61% :     0.000212s :      4: match.inline
 5.39% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.15% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.42% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.18% :     0.000001s :     10: predicate.addn_zero_filter
 2.29% :     0.000003s :     10: predicate.arithmetic_simplify
 2.00% :     0.000003s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.07% :     0.000001s :     10: predicate.depend_value_elim
 0.96% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.51% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.73% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.83% :     0.000002s :     10: predicate.environ_get_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 6.23% :     0.000008s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000002s :      6: predicate.less_batch_normalization
 1.73% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.23% :     0.000003s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.26% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.14% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 2.06% :     0.000003s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.38% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.45% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.45% :     0.000001s :      3: predicate.reset_defer_inline
 2.01% :     0.000003s :     10: predicate.reshape_eliminate
 1.38% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.93% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.58% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 1.90% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.00% :     0.000009s :     53: predicate.switch_simplify
 1.04% :     0.000001s :     10: predicate.tile_eliminate
 1.26% :     0.000002s :     10: predicate.transpose_eliminate
 1.25% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.09% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.36% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.33% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000718    12
45.61% :     0.000328s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.14% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.25% :     0.000339s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023424    76
 0.23% :     0.000054s :      1: add_recomputation
 0.29% :     0.000069s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.42% :     0.000333s :      1: bootstrap
 0.14% :     0.000033s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000027s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
25.79% :     0.006042s :      1: jit_opt_a
 0.72% :     0.000168s :      1: jit_opt_after_cconv
 0.25% :     0.000059s :      1: jit_opt_b
 2.24% :     0.000526s :      1: loop_unroll
 3.10% :     0.000725s :      1: mutable_eliminate
 3.99% :     0.000936s :     26: opt.transform.jit_opt_a
 0.23% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000029s :      4: opt.transform.jit_opt_b
 0.07% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000039s :      4: opt.transform.symbol_engine_opt
 2.63% :     0.000617s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000044s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.30% :     0.000070s :      1: remove_dup_value
 2.03% :     0.000475s :      1: renormalize.infer
 1.66% :     0.000389s :      1: renormalize.specialize
 0.05% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000039s :      1: rewriter_after_opt_a
 0.35% :     0.000081s :      1: rewriter_before_opt_a
 0.38% :     0.000089s :      1: symbol_engine_optimizer
29.65% :     0.006946s :      1: task_emit
22.69% :     0.005315s :      1: type_inference
 0.29% :     0.000068s :      1: validate


TotalTime = 0.0174067, [33]
[bootstrap]: 0.0003273
[type_inference]: 0.00430228
[event_method]: 1.682e-05
[auto_monad]: 4.354e-05
[graph_reusing]: 4.48999e-06
[pre_auto_parallel]: 1.60999e-06
[py_interpret_to_execute]: 2.583e-05
[rewriter_before_opt_a]: 6.135e-05
[expand_dump_flag]: 2.55002e-06
[jit_opt_a]: 0.00471768, [2]
    [Cycle 1]: 0.00131558, [27]
        [switch_simplify]: 4.204e-05
        [loop_unroll]: 2.824e-05
        [a_1]: 0.00045701
        [with_stream_mark]: 1.399e-05
        [recompute_prepare]: 6.69001e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 5.66998e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 6.21e-06
        [meta_shard_fg_expand]: 1.52999e-06
        [get_grad_eliminate_]: 5.01002e-06
        [merge_forward]: 3.06001e-06
        [cell_reuse_recompute_pass]: 8.59989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.311e-05
        [j_node_and_user_rematch]: 8.69e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 7.25e-06
        [inline_without_move]: 5.42001e-06
        [renormalize]: 0.00047714
        [add_forward_monad_depend]: 6.02999e-06
        [auto_monad_grad]: 1.75001e-06
        [auto_monad_eliminator]: 1.353e-05
        [cse]: 2.612e-05
        [replace_applicator]: 1.041e-05
    [Cycle 2]: 0.00030567, [27]
        [switch_simplify]: 5.25999e-06
        [loop_unroll]: 4.51002e-06
        [a_1]: 8.671e-05
        [with_stream_mark]: 8e-06
        [recompute_prepare]: 4.78001e-06
        [updatestate_depend_eliminate]: 2.41998e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 4.81002e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 5.05999e-06
        [meta_shard_fg_expand]: 1.13001e-06
        [get_grad_eliminate_]: 4.52998e-06
        [merge_forward]: 2.32999e-06
        [cell_reuse_recompute_pass]: 2.06e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.165e-05
        [j_node_and_user_rematch]: 7.55998e-06
        [meta_fg_expand]: 1.32999e-06
        [replace_old_param]: 6.91999e-06
        [inline_without_move]: 4.42998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.35999e-06
        [auto_monad_grad]: 6.29982e-07
        [auto_monad_eliminator]: 4.68001e-06
        [cse]: 8.77e-06
        [replace_applicator]: 4.80999e-06
[py_interpret_to_execute_after_opt_a]: 8.48001e-06
[rewriter_after_opt_a]: 3.281e-05
[convert_after_rewriter]: 7.58001e-06
[order_py_execute_after_rewriter]: 5.19998e-06
[mutable_eliminate]: 0.00059141
[jit_opt_b]: 4.716e-05, [1]
    [Cycle 1]: 4.035e-05, [2]
        [frontend_op_eliminate]: 1.47e-05
        [inline_after_opt_a]: 1.458e-05
[cconv]: 2.08e-05
[loop_unroll]: 0.00042463
[jit_opt_after_cconv]: 0.00012522, [1]
    [Cycle 1]: 0.00011933, [11]
        [c_1]: 1.916e-05
        [parameter_eliminate]: 1.86998e-06
        [updatestate_depend_eliminate]: 4e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 1.96998e-06
        [cse]: 1.489e-05
        [call_graph_tuple_transform]: 1.616e-05
        [tuple_list_get_item_eliminator]: 5.02999e-06
        [none_parameter_eliminate]: 1.34e-06
        [renormalize]: 3.19997e-07
        [switch_simplify]: 5.03002e-06
[remove_dup_value]: 1.297e-05
[partial_unused_args_eliminate]: 2.28002e-06
[environ_conv]: 4.61002e-06
[add_recomputation]: 4.595e-05
[cse_after_recomputation]: 1.985e-05, [1]
    [Cycle 1]: 1.433e-05, [1]
        [cse]: 8.70999e-06
[auto_monad_reorder]: 1.494e-05
[get_jit_bprop_graph]: 1.77001e-06
[rewriter_after_jit_bprop_graph]: 4.75001e-06
[opt_after_jit_grad]: 0.00044852
[symbol_engine_optimizer]: 6.847e-05, [1]
    [Cycle 1]: 6.225e-05, [6]
        [build]: 2.81999e-06
        [elim_shapecalc]: 7.19001e-06
        [elim_not_effective]: 1.117e-05
        [opt_reshape]: 5.27001e-06
        [fold_const_symbol]: 8.05e-06
        [renormalize]: 3.50003e-07
[validate]: 3.07e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.005805
[execute]: 7.15e-06
Sums
    bootstrap                            :     0.000327s :  2.38%
    type_inference                       :     0.004302s : 31.29%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000044s :  0.32%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.19%
    rewriter_before_opt_a                :     0.000061s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000047s :  0.34%
    jit_opt_a.loop_unroll                :     0.000033s :  0.24%
    jit_opt_a.a_1                        :     0.000544s :  3.95%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000014s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000477s :  3.47%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000035s :  0.25%
    jit_opt_a.replace_applicator         :     0.000015s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.24%
    convert_after_rewriter               :     0.000008s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000591s :  4.30%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000021s :  0.15%
    loop_unroll                          :     0.000425s :  3.09%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000046s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000449s :  3.26%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005805s : 42.22%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000148    23
 1.27% :     0.000002s :      2: substitution.elim_not_effective
 0.85% :     0.000001s :      2: substitution.fold_const_symbol
 3.31% :     0.000005s :      3: substitution.graph_param_transform
80.41% :     0.000119s :      4: substitution.inline
 2.31% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.72% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.67% :     0.000002s :      2: substitution.replace_old_param
 7.45% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004246     2
83.52% :     0.003547s :      1: type_inference.infer
16.48% :     0.000700s :      1: type_inference.specialize
------[replace.]   0.000055     6
65.95% :     0.000036s :      4: replace.inline
34.05% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000126     6
92.15% :     0.000116s :      4: match.inline
 7.85% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.60% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.39% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 2.43% :     0.000003s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.51% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.46% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.73% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.38% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.13% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 4.92% :     0.000005s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.78% :     0.000002s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.23% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.93% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.69% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.48% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000002s :     15: predicate.partial_eliminate
 1.70% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.66% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.67% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.85% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.71% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.29% :     0.000008s :     53: predicate.switch_simplify
 1.33% :     0.000001s :     10: predicate.tile_eliminate
 1.29% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.80% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.19% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.88% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.51% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000462    12
49.20% :     0.000227s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.49% :     0.000021s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.31% :     0.000214s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018711    76
 0.26% :     0.000049s :      1: add_recomputation
 0.25% :     0.000047s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.84% :     0.000344s :      1: bootstrap
 0.12% :     0.000023s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
25.23% :     0.004721s :      1: jit_opt_a
 0.68% :     0.000128s :      1: jit_opt_after_cconv
 0.26% :     0.000049s :      1: jit_opt_b
 2.31% :     0.000431s :      1: loop_unroll
 3.20% :     0.000598s :      1: mutable_eliminate
 3.91% :     0.000732s :     26: opt.transform.jit_opt_a
 0.23% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.44% :     0.000456s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000003s :      1: pre_auto_parallel
 0.15% :     0.000028s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.38% :     0.000258s :      1: renormalize.infer
 1.14% :     0.000213s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000036s :      1: rewriter_after_opt_a
 0.34% :     0.000064s :      1: rewriter_before_opt_a
 0.38% :     0.000071s :      1: symbol_engine_optimizer
31.08% :     0.005814s :      1: task_emit
23.04% :     0.004311s :      1: type_inference
 0.27% :     0.000051s :      1: validate


TotalTime = 0.0336175, [33]
[bootstrap]: 0.00021236
[type_inference]: 0.0136834
[event_method]: 8.105e-05
[auto_monad]: 0.00011126
[graph_reusing]: 7.88999e-06
[pre_auto_parallel]: 3.31001e-06
[py_interpret_to_execute]: 3.152e-05
[rewriter_before_opt_a]: 0.00010813
[expand_dump_flag]: 3.5e-06
[jit_opt_a]: 0.0112464, [3]
    [Cycle 1]: 0.00557989, [27]
        [switch_simplify]: 0.00011892
        [loop_unroll]: 4.408e-05
        [a_1]: 0.00085767
        [with_stream_mark]: 2.347e-05
        [recompute_prepare]: 1.731e-05
        [updatestate_depend_eliminate]: 7.59002e-06
        [updatestate_assign_eliminate]: 6.60002e-06
        [updatestate_loads_eliminate]: 6.16e-06
        [parameter_eliminate]: 2.36e-06
        [specialize_transform]: 1.278e-05
        [updatestate_useless_node_eliminater]: 1.206e-05
        [accelerated_algorithm]: 1.231e-05
        [meta_shard_fg_expand]: 3.78999e-06
        [get_grad_eliminate_]: 1.176e-05
        [merge_forward]: 8.58001e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.711e-05
        [j_node_and_user_rematch]: 2.291e-05
        [meta_fg_expand]: 0.00147674
        [replace_old_param]: 5.11e-05
        [inline_without_move]: 5.064e-05
        [renormalize]: 0.00234144
        [add_forward_monad_depend]: 7.64002e-06
        [auto_monad_grad]: 4.2e-06
        [auto_monad_eliminator]: 4.335e-05
        [cse]: 0.00014276
        [replace_applicator]: 6.548e-05
    [Cycle 2]: 0.00204962, [27]
        [switch_simplify]: 3.931e-05
        [loop_unroll]: 3.651e-05
        [a_1]: 0.00110458
        [with_stream_mark]: 1.291e-05
        [recompute_prepare]: 7.31999e-06
        [updatestate_depend_eliminate]: 3.5e-06
        [updatestate_assign_eliminate]: 2.89999e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [parameter_eliminate]: 8.49977e-07
        [specialize_transform]: 6.44999e-06
        [updatestate_useless_node_eliminater]: 6.40002e-06
        [accelerated_algorithm]: 6.06e-06
        [meta_shard_fg_expand]: 1.71e-06
        [get_grad_eliminate_]: 7.13e-06
        [merge_forward]: 3.25998e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.265e-05
        [j_node_and_user_rematch]: 9.88998e-06
        [meta_fg_expand]: 8.858e-05
        [replace_old_param]: 1.016e-05
        [inline_without_move]: 6.40002e-06
        [renormalize]: 0.00049107
        [add_forward_monad_depend]: 4.58001e-06
        [auto_monad_grad]: 1.23002e-06
        [auto_monad_eliminator]: 1.031e-05
        [cse]: 2.066e-05
        [replace_applicator]: 1.059e-05
    [Cycle 3]: 0.00036246, [27]
        [switch_simplify]: 6.55002e-06
        [loop_unroll]: 6.21998e-06
        [a_1]: 0.00011716
        [with_stream_mark]: 9.12001e-06
        [recompute_prepare]: 6.04999e-06
        [updatestate_depend_eliminate]: 3.6e-06
        [updatestate_assign_eliminate]: 2.65002e-06
        [updatestate_loads_eliminate]: 2.55002e-06
        [parameter_eliminate]: 1.01997e-06
        [specialize_transform]: 6.05002e-06
        [updatestate_useless_node_eliminater]: 6.10002e-06
        [accelerated_algorithm]: 5.92001e-06
        [meta_shard_fg_expand]: 1.31002e-06
        [get_grad_eliminate_]: 5.82001e-06
        [merge_forward]: 2.83998e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.4e-05
        [j_node_and_user_rematch]: 9.79999e-06
        [meta_fg_expand]: 2.01e-06
        [replace_old_param]: 8.38999e-06
        [inline_without_move]: 5.77999e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.45001e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 6.16e-06
        [cse]: 1.352e-05
        [replace_applicator]: 5.89e-06
[py_interpret_to_execute_after_opt_a]: 9.57999e-06
[rewriter_after_opt_a]: 3.096e-05
[convert_after_rewriter]: 6.89999e-06
[order_py_execute_after_rewriter]: 4.80999e-06
[mutable_eliminate]: 0.00058022
[jit_opt_b]: 5.378e-05, [1]
    [Cycle 1]: 4.787e-05, [2]
        [frontend_op_eliminate]: 1.972e-05
        [inline_after_opt_a]: 1.682e-05
[cconv]: 1.549e-05
[loop_unroll]: 0.00042267
[jit_opt_after_cconv]: 0.0001539, [1]
    [Cycle 1]: 0.00014762, [11]
        [c_1]: 2.671e-05
        [parameter_eliminate]: 2.61999e-06
        [updatestate_depend_eliminate]: 5.21998e-06
        [updatestate_assign_eliminate]: 3.4e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [cse]: 2.266e-05
        [call_graph_tuple_transform]: 1.971e-05
        [tuple_list_get_item_eliminator]: 6.46e-06
        [none_parameter_eliminate]: 9.09989e-07
        [renormalize]: 4.49974e-07
        [switch_simplify]: 6.47001e-06
[remove_dup_value]: 1.189e-05
[partial_unused_args_eliminate]: 2.26998e-06
[environ_conv]: 4.08001e-06
[add_recomputation]: 4.279e-05
[cse_after_recomputation]: 2.784e-05, [1]
    [Cycle 1]: 2.161e-05, [1]
        [cse]: 1.535e-05
[auto_monad_reorder]: 1.636e-05
[get_jit_bprop_graph]: 1.72999e-06
[rewriter_after_jit_bprop_graph]: 4.62e-06
[opt_after_jit_grad]: 0.00053455
[symbol_engine_optimizer]: 8.1e-05, [1]
    [Cycle 1]: 7.492e-05, [6]
        [build]: 3.3e-06
        [elim_shapecalc]: 9.05001e-06
        [elim_not_effective]: 1.503e-05
        [opt_reshape]: 7.38e-06
        [fold_const_symbol]: 1.112e-05
        [renormalize]: 5.50004e-07
[validate]: 3.21e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.0058828
[execute]: 4.86997e-06
Sums
    bootstrap                            :     0.000212s :  0.72%
    type_inference                       :     0.013683s : 46.25%
    event_method                         :     0.000081s :  0.27%
    auto_monad                           :     0.000111s :  0.38%
    graph_reusing                        :     0.000008s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000032s :  0.11%
    rewriter_before_opt_a                :     0.000108s :  0.37%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000165s :  0.56%
    jit_opt_a.loop_unroll                :     0.000087s :  0.29%
    jit_opt_a.a_1                        :     0.002079s :  7.03%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000031s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000025s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.08%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000054s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000043s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.001567s :  5.30%
    jit_opt_a.replace_old_param          :     0.000070s :  0.24%
    jit_opt_a.inline_without_move        :     0.000063s :  0.21%
    jit_opt_a.renormalize                :     0.002833s :  9.57%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000006s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000060s :  0.20%
    jit_opt_a.cse                        :     0.000177s :  0.60%
    jit_opt_a.replace_applicator         :     0.000082s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.03%
    rewriter_after_opt_a                 :     0.000031s :  0.10%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000580s :  1.96%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.06%
    cconv                                :     0.000015s :  0.05%
    loop_unroll                          :     0.000423s :  1.43%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000012s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.01%
    add_recomputation                    :     0.000043s :  0.14%
    cse_after_recomputation.cse          :     0.000015s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000535s :  1.81%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005883s : 19.88%
    execute                              :     0.000005s :  0.02%


Time group info:
------[substitution.]   0.000472   113
 0.49% :     0.000002s :      3: substitution.elim_not_effective
 0.46% :     0.000002s :      3: substitution.fold_const_symbol
 0.95% :     0.000004s :      4: substitution.graph_param_transform
62.23% :     0.000294s :     14: substitution.inline
 3.26% :     0.000015s :      2: substitution.inline_without_move
 1.87% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.38% :     0.000011s :      7: substitution.minmaximum_grad
 1.42% :     0.000007s :      5: substitution.partial_eliminate
 2.39% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.58% :     0.000026s :     10: substitution.replace_applicator
 1.43% :     0.000007s :      7: substitution.replace_old_param
 0.47% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.52% :     0.000007s :      1: substitution.switch_simplify
 4.25% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.06% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 8.23% :     0.000039s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.013602     2
89.02% :     0.012109s :      1: type_inference.infer
10.98% :     0.001493s :      1: type_inference.specialize
------[replace.]   0.000176    22
53.67% :     0.000094s :     14: replace.inline
18.04% :     0.000032s :      1: replace.switch_simplify
28.29% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000313    22
91.31% :     0.000286s :     14: match.inline
 2.10% :     0.000007s :      1: match.switch_simplify
 6.60% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000450  2634
 1.31% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.36% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.29% :     0.000006s :     44: predicate.addn_check_dump
 1.55% :     0.000007s :     44: predicate.addn_zero_filter
16.85% :     0.000076s :     44: predicate.arithmetic_simplify
 1.34% :     0.000006s :     44: predicate.cast_eliminate
 0.15% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.18% :     0.000005s :     44: predicate.compare_switch_simplify
 1.40% :     0.000006s :     44: predicate.depend_value_elim
 1.16% :     0.000005s :     44: predicate.dict_get_item_const_eliminator
 1.24% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.21% :     0.000005s :     44: predicate.dict_set_item_eliminator
 0.22% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.14% :     0.000001s :      4: predicate.elim_not_effective
 0.18% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.22% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.17% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.20% :     0.000005s :     44: predicate.environ_get_eliminate
 1.16% :     0.000005s :     44: predicate.environ_get_set_eliminate
 0.08% :     0.000000s :      4: predicate.fold_const_symbol
 0.78% :     0.000003s :     19: predicate.get_grad_eliminate
 0.09% :     0.000000s :      4: predicate.graph_param_transform
 3.65% :     0.000016s :     73: predicate.inline
 1.53% :     0.000007s :     40: predicate.inline_without_move
 0.30% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.82% :     0.000004s :     19: predicate.less_batch_normalization
 1.58% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.56% :     0.000007s :     55: predicate.load_eliminater
 0.40% :     0.000002s :      4: predicate.loop_unroll_after_grad
 2.95% :     0.000013s :     91: predicate.loop_unroll_before_grad
 1.54% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000005s :     44: predicate.merge_addn
 1.24% :     0.000006s :     44: predicate.minmaximum_grad
 0.49% :     0.000002s :      4: predicate.mutable_eliminate
 0.16% :     0.000001s :      4: predicate.opt_reshape
 1.99% :     0.000009s :     55: predicate.partial_eliminate
 1.38% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.68% :     0.000008s :     44: predicate.reduce_eliminate
 1.47% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.34% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.08% :     0.000009s :     97: predicate.replace_applicator
 0.78% :     0.000004s :     40: predicate.replace_old_param
 0.10% :     0.000000s :      4: predicate.reset_defer_inline
 1.54% :     0.000007s :     44: predicate.reshape_eliminate
 1.29% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.30% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.44% :     0.000006s :     44: predicate.same_eliminate
 0.41% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.38% :     0.000002s :      8: predicate.special_op_eliminate
 0.73% :     0.000003s :     19: predicate.specialize_transform
 1.44% :     0.000007s :     44: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.15% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.38% :     0.000011s :     65: predicate.switch_defer_inline
 2.15% :     0.000010s :     65: predicate.switch_layer_defer_inline
 5.56% :     0.000025s :    162: predicate.switch_simplify
 1.25% :     0.000006s :     44: predicate.tile_eliminate
 1.36% :     0.000006s :     44: predicate.transpose_eliminate
 1.55% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 2.56% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000008s :     44: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.57% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.39% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000007s :     44: predicate.value_based_eliminate
 0.12% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.20% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001651    32
61.13% :     0.001009s :     14: func_graph_cloner_run.FuncGraphClonerGraph
38.87% :     0.000642s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.039305    91
 0.12% :     0.000046s :      1: add_recomputation
 0.29% :     0.000115s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.58% :     0.000228s :      1: bootstrap
 0.05% :     0.000018s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000031s :      1: cse_after_recomputation
 0.02% :     0.000006s :      1: environ_conv
 0.22% :     0.000085s :      1: event_method
 0.02% :     0.000008s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
28.62% :     0.011249s :      1: jit_opt_a
 0.41% :     0.000163s :      1: jit_opt_after_cconv
 0.14% :     0.000056s :      1: jit_opt_b
 1.09% :     0.000429s :      1: loop_unroll
 1.49% :     0.000587s :      1: mutable_eliminate
 6.94% :     0.002728s :     39: opt.transform.jit_opt_a
 0.14% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000031s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.38% :     0.000543s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000034s :      1: py_interpret_to_execute
 0.03% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000015s :      1: remove_dup_value
 3.80% :     0.001495s :      2: renormalize.infer
 3.37% :     0.001325s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000034s :      1: rewriter_after_opt_a
 0.28% :     0.000111s :      1: rewriter_before_opt_a
 0.21% :     0.000083s :      1: symbol_engine_optimizer
14.99% :     0.005892s :      1: task_emit
34.86% :     0.013700s :      1: type_inference
 0.13% :     0.000052s :      1: validate


TotalTime = 0.0120428, [30]
[bootstrap]: 0.00029374
[type_inference]: 0.00413983
[event_method]: 1.675e-05
[auto_monad]: 6.408e-05
[graph_reusing]: 6.39999e-06
[pre_auto_parallel]: 2.45002e-06
[py_interpret_to_execute]: 3.029e-05
[rewriter_before_opt_a]: 7.121e-05
[expand_dump_flag]: 2.93998e-06
[jit_opt_a]: 0.00531141, [2]
    [Cycle 1]: 0.00159797, [27]
        [switch_simplify]: 5.56e-05
        [loop_unroll]: 3.025e-05
        [a_1]: 0.00053096
        [with_stream_mark]: 1.936e-05
        [recompute_prepare]: 7.33e-06
        [updatestate_depend_eliminate]: 3.36999e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 2.01e-06
        [specialize_transform]: 6.05002e-06
        [updatestate_useless_node_eliminater]: 5.07999e-06
        [accelerated_algorithm]: 5.27999e-06
        [meta_shard_fg_expand]: 1.97001e-06
        [get_grad_eliminate_]: 5.14998e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 1.00001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.464e-05
        [j_node_and_user_rematch]: 9.33002e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 9.10999e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 0.00063147
        [add_forward_monad_depend]: 5.59998e-06
        [auto_monad_grad]: 2.34001e-06
        [auto_monad_eliminator]: 1.294e-05
        [cse]: 2.87e-05
        [replace_applicator]: 1.153e-05
    [Cycle 2]: 0.0003211, [27]
        [switch_simplify]: 5.62001e-06
        [loop_unroll]: 5.03002e-06
        [a_1]: 8.917e-05
        [with_stream_mark]: 9.01998e-06
        [recompute_prepare]: 5.36998e-06
        [updatestate_depend_eliminate]: 2.41e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 4.87e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 4.97e-06
        [meta_shard_fg_expand]: 1.16002e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.53e-06
        [cell_reuse_recompute_pass]: 1.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.263e-05
        [j_node_and_user_rematch]: 8.25e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 7.20003e-06
        [inline_without_move]: 4.93001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.35999e-06
        [auto_monad_grad]: 1.07e-06
        [auto_monad_eliminator]: 5.89e-06
        [cse]: 1.157e-05
        [replace_applicator]: 4.93001e-06
[py_interpret_to_execute_after_opt_a]: 1.009e-05
[rewriter_after_opt_a]: 3.587e-05
[convert_after_rewriter]: 7.3e-06
[order_py_execute_after_rewriter]: 7.64002e-06
[mutable_eliminate]: 0.00054893
[jit_opt_b]: 5.167e-05, [1]
    [Cycle 1]: 4.398e-05, [2]
        [frontend_op_eliminate]: 1.697e-05
        [inline_after_opt_a]: 1.518e-05
[cconv]: 2.56e-05
[loop_unroll]: 0.00040513
[jit_opt_after_cconv]: 0.00013529, [1]
    [Cycle 1]: 0.00012868, [11]
        [c_1]: 2.053e-05
        [parameter_eliminate]: 2.51e-06
        [updatestate_depend_eliminate]: 5.25001e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.666e-05
        [call_graph_tuple_transform]: 1.734e-05
        [tuple_list_get_item_eliminator]: 5.42999e-06
        [none_parameter_eliminate]: 1.51998e-06
        [renormalize]: 3.39991e-07
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 1.378e-05
[partial_unused_args_eliminate]: 2.14999e-06
[environ_conv]: 4.75999e-06
[add_recomputation]: 4.617e-05
[cse_after_recomputation]: 1.987e-05, [1]
    [Cycle 1]: 1.462e-05, [1]
        [cse]: 8.80001e-06
[auto_monad_reorder]: 1.532e-05
[get_jit_bprop_graph]: 2.12999e-06
[rewriter_after_jit_bprop_graph]: 4.60001e-06
[opt_after_jit_grad]: 0.0004956
[symbol_engine_optimizer]: 7.3e-05, [1]
    [Cycle 1]: 6.634e-05, [6]
        [build]: 3.51999e-06
        [elim_shapecalc]: 7.61999e-06
        [elim_not_effective]: 1.345e-05
        [opt_reshape]: 5.42999e-06
        [fold_const_symbol]: 8.54e-06
        [renormalize]: 4.89992e-07
[validate]: 3.333e-05
Sums
    bootstrap                            :     0.000294s :  3.64%
    type_inference                       :     0.004140s : 51.32%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000064s :  0.79%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000030s :  0.38%
    rewriter_before_opt_a                :     0.000071s :  0.88%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000061s :  0.76%
    jit_opt_a.loop_unroll                :     0.000035s :  0.44%
    jit_opt_a.a_1                        :     0.000620s :  7.69%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000632s :  7.83%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.23%
    jit_opt_a.cse                        :     0.000040s :  0.50%
    jit_opt_a.replace_applicator         :     0.000016s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.13%
    rewriter_after_opt_a                 :     0.000036s :  0.44%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000008s :  0.09%
    mutable_eliminate                    :     0.000549s :  6.80%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000026s :  0.32%
    loop_unroll                          :     0.000405s :  5.02%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000046s :  0.57%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000496s :  6.14%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000033s :  0.41%


Time group info:
------[substitution.]   0.000206    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.46% :     0.000005s :      3: substitution.graph_param_transform
84.60% :     0.000174s :      4: substitution.inline
 1.43% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.14% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.91% :     0.000004s :      2: substitution.replace_old_param
 5.87% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004079     2
78.58% :     0.003206s :      1: type_inference.infer
21.42% :     0.000874s :      1: type_inference.specialize
------[replace.]   0.000066     6
67.49% :     0.000045s :      4: replace.inline
32.51% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000182     6
94.05% :     0.000172s :      4: match.inline
 5.95% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.44% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 1.91% :     0.000002s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.79% :     0.000001s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.92% :     0.000007s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.23% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.20% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.17% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.70% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000000s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.23% :     0.000001s :      6: predicate.special_op_eliminate
 1.12% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.99% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.74% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.11% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000547    12
47.43% :     0.000260s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.53% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.04% :     0.000247s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013562    72
 0.36% :     0.000049s :      1: add_recomputation
 0.50% :     0.000068s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 2.29% :     0.000310s :      1: bootstrap
 0.21% :     0.000028s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
39.18% :     0.005314s :      1: jit_opt_a
 1.02% :     0.000138s :      1: jit_opt_after_cconv
 0.40% :     0.000054s :      1: jit_opt_b
 3.04% :     0.000413s :      1: loop_unroll
 4.12% :     0.000559s :      1: mutable_eliminate
 6.12% :     0.000830s :     26: opt.transform.jit_opt_a
 0.33% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.71% :     0.000503s :      1: opt_after_jit_grad
 0.07% :     0.000010s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000033s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.52% :     0.000342s :      1: renormalize.infer
 2.07% :     0.000281s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.29% :     0.000039s :      1: rewriter_after_opt_a
 0.54% :     0.000074s :      1: rewriter_before_opt_a
 0.56% :     0.000076s :      1: symbol_engine_optimizer
30.62% :     0.004153s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.766.725 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6c078280> should be wrapped by jit only once.

TotalTime = 0.0192846, [33]
[bootstrap]: 0.00029303
[type_inference]: 0.00467665
[event_method]: 1.92e-05
[auto_monad]: 6.602e-05
[graph_reusing]: 5.91998e-06
[pre_auto_parallel]: 2.54999e-06
[py_interpret_to_execute]: 3.319e-05
[rewriter_before_opt_a]: 7.235e-05
[expand_dump_flag]: 2.96999e-06
[jit_opt_a]: 0.00620136, [2]
    [Cycle 1]: 0.00161914, [27]
        [switch_simplify]: 5.392e-05
        [loop_unroll]: 2.96e-05
        [a_1]: 0.0005976
        [with_stream_mark]: 1.932e-05
        [recompute_prepare]: 7.30003e-06
        [updatestate_depend_eliminate]: 3.57002e-06
        [updatestate_assign_eliminate]: 3.04999e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [parameter_eliminate]: 1.76e-06
        [specialize_transform]: 5.57001e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.84e-06
        [get_grad_eliminate_]: 4.81002e-06
        [merge_forward]: 3.33998e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.289e-05
        [j_node_and_user_rematch]: 8.32e-06
        [meta_fg_expand]: 2.21e-06
        [replace_old_param]: 8.45001e-06
        [inline_without_move]: 5.02999e-06
        [renormalize]: 0.00060095
        [add_forward_monad_depend]: 5.37001e-06
        [auto_monad_grad]: 2.07001e-06
        [auto_monad_eliminator]: 1.388e-05
        [cse]: 2.722e-05
        [replace_applicator]: 1.074e-05
    [Cycle 2]: 0.00030658, [27]
        [switch_simplify]: 5.17e-06
        [loop_unroll]: 4.73001e-06
        [a_1]: 8.655e-05
        [with_stream_mark]: 8.48999e-06
        [recompute_prepare]: 4.58001e-06
        [updatestate_depend_eliminate]: 2.53e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.81003e-06
        [parameter_eliminate]: 1.06997e-06
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.60999e-06
        [accelerated_algorithm]: 4.76002e-06
        [meta_shard_fg_expand]: 1.04003e-06
        [get_grad_eliminate_]: 4.22998e-06
        [merge_forward]: 2.19001e-06
        [cell_reuse_recompute_pass]: 1.54998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.18e-05
        [j_node_and_user_rematch]: 8.04002e-06
        [meta_fg_expand]: 1.39998e-06
        [replace_old_param]: 7.03e-06
        [inline_without_move]: 4.51002e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 4.99e-06
        [cse]: 1.008e-05
        [replace_applicator]: 4.79e-06
[py_interpret_to_execute_after_opt_a]: 1.076e-05
[rewriter_after_opt_a]: 3.247e-05
[convert_after_rewriter]: 6.62002e-06
[order_py_execute_after_rewriter]: 4.53999e-06
[mutable_eliminate]: 0.00051761
[jit_opt_b]: 4.667e-05, [1]
    [Cycle 1]: 4.014e-05, [2]
        [frontend_op_eliminate]: 1.511e-05
        [inline_after_opt_a]: 1.395e-05
[cconv]: 2.31e-05
[loop_unroll]: 0.00039755
[jit_opt_after_cconv]: 0.00012731, [1]
    [Cycle 1]: 0.00012111, [11]
        [c_1]: 1.893e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 4.75001e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [cse]: 1.537e-05
        [call_graph_tuple_transform]: 1.67e-05
        [tuple_list_get_item_eliminator]: 4.84998e-06
        [none_parameter_eliminate]: 1.37e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.00999e-06
[remove_dup_value]: 1.226e-05
[partial_unused_args_eliminate]: 2.61999e-06
[environ_conv]: 5.27999e-06
[add_recomputation]: 4.728e-05
[cse_after_recomputation]: 1.907e-05, [1]
    [Cycle 1]: 1.383e-05, [1]
        [cse]: 8.43999e-06
[auto_monad_reorder]: 1.452e-05
[get_jit_bprop_graph]: 2.29001e-06
[rewriter_after_jit_bprop_graph]: 4.42998e-06
[opt_after_jit_grad]: 0.00044709
[symbol_engine_optimizer]: 6.849e-05, [1]
    [Cycle 1]: 6.227e-05, [6]
        [build]: 3.29001e-06
        [elim_shapecalc]: 7.26001e-06
        [elim_not_effective]: 1.181e-05
        [opt_reshape]: 5.25999e-06
        [fold_const_symbol]: 8.1e-06
        [renormalize]: 4.80009e-07
[validate]: 3.101e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00590034
[execute]: 7.88999e-06
Sums
    bootstrap                            :     0.000293s :  2.03%
    type_inference                       :     0.004677s : 32.42%
    event_method                         :     0.000019s :  0.13%
    auto_monad                           :     0.000066s :  0.46%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.23%
    rewriter_before_opt_a                :     0.000072s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000684s :  4.74%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000601s :  4.17%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000037s :  0.26%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000518s :  3.59%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000023s :  0.16%
    loop_unroll                          :     0.000398s :  2.76%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000012s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.33%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000447s :  3.10%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005900s : 40.90%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000277    23
 0.66% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 1.78% :     0.000005s :      3: substitution.graph_param_transform
88.41% :     0.000245s :      4: substitution.inline
 1.24% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.43% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.39% :     0.000004s :      2: substitution.replace_old_param
 4.65% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004603     2
83.24% :     0.003832s :      1: type_inference.infer
16.76% :     0.000771s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.43% :     0.000042s :      4: replace.inline
32.57% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000255     6
95.36% :     0.000243s :      4: match.inline
 4.64% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000001s :     10: predicate.addn_zero_filter
 1.93% :     0.000002s :     10: predicate.arithmetic_simplify
 1.34% :     0.000001s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 2.09% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.06% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.54% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.33% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.80% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.03% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.52% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.56% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.55% :     0.000008s :     53: predicate.switch_simplify
 1.24% :     0.000001s :     10: predicate.tile_eliminate
 1.35% :     0.000002s :     10: predicate.transpose_eliminate
 1.32% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.09% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.87% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000599    12
44.01% :     0.000264s :      5: func_graph_cloner_run.FuncGraphClonerGraph
17.85% :     0.000107s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.14% :     0.000228s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020884    76
 0.24% :     0.000050s :      1: add_recomputation
 0.33% :     0.000070s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.48% :     0.000308s :      1: bootstrap
 0.12% :     0.000025s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000024s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
29.71% :     0.006204s :      1: jit_opt_a
 0.62% :     0.000130s :      1: jit_opt_after_cconv
 0.24% :     0.000049s :      1: jit_opt_b
 1.94% :     0.000405s :      1: loop_unroll
 2.52% :     0.000526s :      1: mutable_eliminate
 4.23% :     0.000884s :     26: opt.transform.jit_opt_a
 0.20% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000023s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000035s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.18% :     0.000455s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 1.63% :     0.000341s :      1: renormalize.infer
 1.21% :     0.000252s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000035s :      1: rewriter_after_opt_a
 0.36% :     0.000075s :      1: rewriter_before_opt_a
 0.34% :     0.000071s :      1: symbol_engine_optimizer
28.31% :     0.005912s :      1: task_emit
22.46% :     0.004691s :      1: type_inference
 0.24% :     0.000050s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.767.361 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0125036, [30]
[bootstrap]: 0.0003118
[type_inference]: 0.00426099
[event_method]: 1.677e-05
[auto_monad]: 6.49e-05
[graph_reusing]: 5.66e-06
[pre_auto_parallel]: 2.54999e-06
[py_interpret_to_execute]: 2.926e-05
[rewriter_before_opt_a]: 6.817e-05
[expand_dump_flag]: 3.08e-06
[jit_opt_a]: 0.00561289, [2]
    [Cycle 1]: 0.00169559, [27]
        [switch_simplify]: 5.513e-05
        [loop_unroll]: 3.242e-05
        [a_1]: 0.00054753
        [with_stream_mark]: 1.789e-05
        [recompute_prepare]: 7.42998e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 5.96e-06
        [updatestate_useless_node_eliminater]: 5.34e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 1.97999e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 3.58999e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.454e-05
        [j_node_and_user_rematch]: 9.88998e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 9.03002e-06
        [inline_without_move]: 5.49e-06
        [renormalize]: 0.00072312
        [add_forward_monad_depend]: 6.63998e-06
        [auto_monad_grad]: 2.43e-06
        [auto_monad_eliminator]: 1.409e-05
        [cse]: 2.632e-05
        [replace_applicator]: 1.248e-05
    [Cycle 2]: 0.00032864, [27]
        [switch_simplify]: 6.29999e-06
        [loop_unroll]: 5.19e-06
        [a_1]: 9.159e-05
        [with_stream_mark]: 1e-05
        [recompute_prepare]: 5.25001e-06
        [updatestate_depend_eliminate]: 2.83e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 9.30013e-07
        [specialize_transform]: 5.56e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.42999e-06
        [get_grad_eliminate_]: 4.67e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 1.54e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.234e-05
        [j_node_and_user_rematch]: 7.85e-06
        [meta_fg_expand]: 1.99e-06
        [replace_old_param]: 7.46999e-06
        [inline_without_move]: 1.035e-05
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.29998e-06
        [auto_monad_grad]: 9.30013e-07
        [auto_monad_eliminator]: 4.92e-06
        [cse]: 1.015e-05
        [replace_applicator]: 5.22999e-06
[py_interpret_to_execute_after_opt_a]: 1.058e-05
[rewriter_after_opt_a]: 3.895e-05
[convert_after_rewriter]: 7.29001e-06
[order_py_execute_after_rewriter]: 5.78997e-06
[mutable_eliminate]: 0.000547
[jit_opt_b]: 5.057e-05, [1]
    [Cycle 1]: 4.34e-05, [2]
        [frontend_op_eliminate]: 1.645e-05
        [inline_after_opt_a]: 1.496e-05
[cconv]: 2.379e-05
[loop_unroll]: 0.00048153
[jit_opt_after_cconv]: 0.0001377, [1]
    [Cycle 1]: 0.00013144, [11]
        [c_1]: 2.072e-05
        [parameter_eliminate]: 2.46998e-06
        [updatestate_depend_eliminate]: 5.52999e-06
        [updatestate_assign_eliminate]: 2.45002e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.626e-05
        [call_graph_tuple_transform]: 1.803e-05
        [tuple_list_get_item_eliminator]: 5.25999e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 5.79981e-07
        [switch_simplify]: 5.47001e-06
[remove_dup_value]: 1.316e-05
[partial_unused_args_eliminate]: 2.25002e-06
[environ_conv]: 5.32001e-06
[add_recomputation]: 4.45e-05
[cse_after_recomputation]: 2.057e-05, [1]
    [Cycle 1]: 1.51e-05, [1]
        [cse]: 9.21002e-06
[auto_monad_reorder]: 1.507e-05
[get_jit_bprop_graph]: 1.86e-06
[rewriter_after_jit_bprop_graph]: 4.87e-06
[opt_after_jit_grad]: 0.00044944
[symbol_engine_optimizer]: 7.516e-05, [1]
    [Cycle 1]: 6.796e-05, [6]
        [build]: 3.31999e-06
        [elim_shapecalc]: 8.23001e-06
        [elim_not_effective]: 1.312e-05
        [opt_reshape]: 5.55001e-06
        [fold_const_symbol]: 8.94e-06
        [renormalize]: 9.20001e-07
[validate]: 3.47e-05
Sums
    bootstrap                            :     0.000312s :  3.73%
    type_inference                       :     0.004261s : 51.02%
    event_method                         :     0.000017s :  0.20%
    auto_monad                           :     0.000065s :  0.78%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000029s :  0.35%
    rewriter_before_opt_a                :     0.000068s :  0.82%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000061s :  0.74%
    jit_opt_a.loop_unroll                :     0.000038s :  0.45%
    jit_opt_a.a_1                        :     0.000639s :  7.65%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.20%
    jit_opt_a.inline_without_move        :     0.000016s :  0.19%
    jit_opt_a.renormalize                :     0.000723s :  8.66%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.23%
    jit_opt_a.cse                        :     0.000036s :  0.44%
    jit_opt_a.replace_applicator         :     0.000018s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.13%
    rewriter_after_opt_a                 :     0.000039s :  0.47%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000006s :  0.07%
    mutable_eliminate                    :     0.000547s :  6.55%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000024s :  0.28%
    loop_unroll                          :     0.000482s :  5.77%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000013s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000045s :  0.53%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000449s :  5.38%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000035s :  0.42%


Time group info:
------[substitution.]   0.000217    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.40% :     0.000005s :      3: substitution.graph_param_transform
85.24% :     0.000185s :      4: substitution.inline
 1.27% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.09% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.91% :     0.000004s :      2: substitution.replace_old_param
 5.63% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004204     2
81.18% :     0.003413s :      1: type_inference.infer
18.82% :     0.000791s :      1: type_inference.specialize
------[replace.]   0.000069     6
68.68% :     0.000047s :      4: replace.inline
31.32% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000193     6
94.26% :     0.000182s :      4: match.inline
 5.74% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.17% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.44% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.12% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 1.13% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.65% :     0.000007s :     22: predicate.inline
 1.05% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.21% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.37% :     0.000002s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.57% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.82% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.63% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.66% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.34% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000573    12
45.24% :     0.000259s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.56% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.20% :     0.000270s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014146    72
 0.33% :     0.000047s :      1: add_recomputation
 0.48% :     0.000068s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.29% :     0.000324s :      1: bootstrap
 0.19% :     0.000026s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.14% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
39.70% :     0.005616s :      1: jit_opt_a
 0.99% :     0.000141s :      1: jit_opt_after_cconv
 0.38% :     0.000053s :      1: jit_opt_b
 3.46% :     0.000490s :      1: loop_unroll
 3.93% :     0.000555s :      1: mutable_eliminate
 6.08% :     0.000861s :     26: opt.transform.jit_opt_a
 0.33% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.23% :     0.000457s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.23% :     0.000032s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000016s :      1: remove_dup_value
 2.76% :     0.000391s :      1: renormalize.infer
 2.29% :     0.000325s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.30% :     0.000042s :      1: rewriter_after_opt_a
 0.50% :     0.000071s :      1: rewriter_before_opt_a
 0.55% :     0.000078s :      1: symbol_engine_optimizer
30.20% :     0.004272s :      1: type_inference


TotalTime = 0.0329568, [33]
[bootstrap]: 0.0002252
[type_inference]: 0.0124629
[event_method]: 6.053e-05
[auto_monad]: 0.00011182
[graph_reusing]: 7.68999e-06
[pre_auto_parallel]: 3.55e-06
[py_interpret_to_execute]: 3.063e-05
[rewriter_before_opt_a]: 0.00011454
[expand_dump_flag]: 3.76999e-06
[jit_opt_a]: 0.0110728, [3]
    [Cycle 1]: 0.00543559, [27]
        [switch_simplify]: 0.0001204
        [loop_unroll]: 4.404e-05
        [a_1]: 0.00086637
        [with_stream_mark]: 2.296e-05
        [recompute_prepare]: 1.749e-05
        [updatestate_depend_eliminate]: 8.08001e-06
        [updatestate_assign_eliminate]: 6.75002e-06
        [updatestate_loads_eliminate]: 6.17999e-06
        [parameter_eliminate]: 2.55002e-06
        [specialize_transform]: 1.329e-05
        [updatestate_useless_node_eliminater]: 1.207e-05
        [accelerated_algorithm]: 1.247e-05
        [meta_shard_fg_expand]: 4.18001e-06
        [get_grad_eliminate_]: 1.208e-05
        [merge_forward]: 8.06001e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.709e-05
        [j_node_and_user_rematch]: 2.305e-05
        [meta_fg_expand]: 0.00140136
        [replace_old_param]: 5.367e-05
        [inline_without_move]: 5.185e-05
        [renormalize]: 0.00222946
        [add_forward_monad_depend]: 8.93002e-06
        [auto_monad_grad]: 5.56e-06
        [auto_monad_eliminator]: 4.803e-05
        [cse]: 0.0001628
        [replace_applicator]: 6.432e-05
    [Cycle 2]: 0.00194864, [27]
        [switch_simplify]: 3.933e-05
        [loop_unroll]: 3.553e-05
        [a_1]: 0.00101141
        [with_stream_mark]: 1.309e-05
        [recompute_prepare]: 7.93001e-06
        [updatestate_depend_eliminate]: 4.20999e-06
        [updatestate_assign_eliminate]: 3.22002e-06
        [updatestate_loads_eliminate]: 2.63003e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 6.33002e-06
        [updatestate_useless_node_eliminater]: 6.13998e-06
        [accelerated_algorithm]: 6.02999e-06
        [meta_shard_fg_expand]: 1.72999e-06
        [get_grad_eliminate_]: 6.00002e-06
        [merge_forward]: 3.38999e-06
        [cell_reuse_recompute_pass]: 9.50007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.357e-05
        [j_node_and_user_rematch]: 1.08e-05
        [meta_fg_expand]: 7.518e-05
        [replace_old_param]: 1.038e-05
        [inline_without_move]: 6.16998e-06
        [renormalize]: 0.00048361
        [add_forward_monad_depend]: 4.23001e-06
        [auto_monad_grad]: 1.27999e-06
        [auto_monad_eliminator]: 1.159e-05
        [cse]: 1.934e-05
        [replace_applicator]: 1.205e-05
    [Cycle 3]: 0.00036708, [27]
        [switch_simplify]: 6.73998e-06
        [loop_unroll]: 5.92999e-06
        [a_1]: 0.00011841
        [with_stream_mark]: 8.85001e-06
        [recompute_prepare]: 6.39999e-06
        [updatestate_depend_eliminate]: 3.68999e-06
        [updatestate_assign_eliminate]: 2.72001e-06
        [updatestate_loads_eliminate]: 2.64999e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 6.18998e-06
        [updatestate_useless_node_eliminater]: 6.11e-06
        [accelerated_algorithm]: 5.47999e-06
        [meta_shard_fg_expand]: 1.33002e-06
        [get_grad_eliminate_]: 5.69e-06
        [merge_forward]: 2.93e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.424e-05
        [j_node_and_user_rematch]: 1.01e-05
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 8.38001e-06
        [inline_without_move]: 5.92999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 6.33998e-06
        [cse]: 1.423e-05
        [replace_applicator]: 5.79e-06
[py_interpret_to_execute_after_opt_a]: 9.58997e-06
[rewriter_after_opt_a]: 3.901e-05
[convert_after_rewriter]: 8.60001e-06
[order_py_execute_after_rewriter]: 5.91e-06
[mutable_eliminate]: 0.00050589
[jit_opt_b]: 5.453e-05, [1]
    [Cycle 1]: 4.836e-05, [2]
        [frontend_op_eliminate]: 1.901e-05
        [inline_after_opt_a]: 1.774e-05
[cconv]: 2.234e-05
[loop_unroll]: 0.00040722
[jit_opt_after_cconv]: 0.0001512, [1]
    [Cycle 1]: 0.00014488, [11]
        [c_1]: 2.603e-05
        [parameter_eliminate]: 2.44999e-06
        [updatestate_depend_eliminate]: 5.93998e-06
        [updatestate_assign_eliminate]: 3.30998e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [cse]: 2.19e-05
        [call_graph_tuple_transform]: 1.993e-05
        [tuple_list_get_item_eliminator]: 6.48003e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 6.12999e-06
[remove_dup_value]: 1.625e-05
[partial_unused_args_eliminate]: 1.98002e-06
[environ_conv]: 5.00999e-06
[add_recomputation]: 5.071e-05
[cse_after_recomputation]: 2.492e-05, [1]
    [Cycle 1]: 1.931e-05, [1]
        [cse]: 1.366e-05
[auto_monad_reorder]: 2.645e-05
[get_jit_bprop_graph]: 1.54e-06
[rewriter_after_jit_bprop_graph]: 5.56e-06
[opt_after_jit_grad]: 0.00048846
[symbol_engine_optimizer]: 8.098e-05, [1]
    [Cycle 1]: 7.487e-05, [6]
        [build]: 3.6e-06
        [elim_shapecalc]: 8.60001e-06
        [elim_not_effective]: 1.521e-05
        [opt_reshape]: 6.93e-06
        [fold_const_symbol]: 1.171e-05
        [renormalize]: 3.00002e-07
[validate]: 3.287e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00671587
[execute]: 6.91999e-06
Sums
    bootstrap                            :     0.000225s :  0.78%
    type_inference                       :     0.012463s : 43.20%
    event_method                         :     0.000061s :  0.21%
    auto_monad                           :     0.000112s :  0.39%
    graph_reusing                        :     0.000008s :  0.03%
    pre_auto_parallel                    :     0.000004s :  0.01%
    py_interpret_to_execute              :     0.000031s :  0.11%
    rewriter_before_opt_a                :     0.000115s :  0.40%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000166s :  0.58%
    jit_opt_a.loop_unroll                :     0.000085s :  0.30%
    jit_opt_a.a_1                        :     0.001996s :  6.92%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.08%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000055s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000044s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001479s :  5.12%
    jit_opt_a.replace_old_param          :     0.000072s :  0.25%
    jit_opt_a.inline_without_move        :     0.000064s :  0.22%
    jit_opt_a.renormalize                :     0.002713s :  9.40%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000066s :  0.23%
    jit_opt_a.cse                        :     0.000196s :  0.68%
    jit_opt_a.replace_applicator         :     0.000082s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.03%
    rewriter_after_opt_a                 :     0.000039s :  0.14%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000506s :  1.75%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000022s :  0.08%
    loop_unroll                          :     0.000407s :  1.41%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000051s :  0.18%
    cse_after_recomputation.cse          :     0.000014s :  0.05%
    auto_monad_reorder                   :     0.000026s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000488s :  1.69%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006716s : 23.28%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000471   113
 0.53% :     0.000002s :      3: substitution.elim_not_effective
 0.37% :     0.000002s :      3: substitution.fold_const_symbol
 1.18% :     0.000006s :      4: substitution.graph_param_transform
62.68% :     0.000295s :     14: substitution.inline
 3.43% :     0.000016s :      2: substitution.inline_without_move
 2.02% :     0.000010s :     14: substitution.j_node_and_user_rematch
 2.07% :     0.000010s :      7: substitution.minmaximum_grad
 1.28% :     0.000006s :      5: substitution.partial_eliminate
 2.31% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.41% :     0.000025s :     10: substitution.replace_applicator
 1.75% :     0.000008s :      7: substitution.replace_old_param
 0.47% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.66% :     0.000008s :      1: substitution.switch_simplify
 4.08% :     0.000019s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.87% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.89% :     0.000037s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012384     2
89.87% :     0.011130s :      1: type_inference.infer
10.13% :     0.001254s :      1: type_inference.specialize
------[replace.]   0.000178    22
53.15% :     0.000095s :     14: replace.inline
18.01% :     0.000032s :      1: replace.switch_simplify
28.84% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000315    22
91.27% :     0.000287s :     14: match.inline
 2.30% :     0.000007s :      1: match.switch_simplify
 6.44% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000378  2634
 1.54% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.45% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.43% :     0.000005s :     44: predicate.addn_check_dump
 1.64% :     0.000006s :     44: predicate.addn_zero_filter
 2.07% :     0.000008s :     44: predicate.arithmetic_simplify
 1.49% :     0.000006s :     44: predicate.cast_eliminate
 0.16% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.41% :     0.000005s :     44: predicate.compare_switch_simplify
 1.55% :     0.000006s :     44: predicate.depend_value_elim
 1.47% :     0.000006s :     44: predicate.dict_get_item_const_eliminator
 1.54% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.46% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.29% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      4: predicate.elim_not_effective
 0.22% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.46% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.39% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.54% :     0.000006s :     44: predicate.environ_get_depend_swap
 1.43% :     0.000005s :     44: predicate.environ_get_eliminate
 1.47% :     0.000006s :     44: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.85% :     0.000003s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.48% :     0.000017s :     73: predicate.inline
 1.83% :     0.000007s :     40: predicate.inline_without_move
 0.36% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.84% :     0.000003s :     19: predicate.less_batch_normalization
 1.80% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.92% :     0.000007s :     55: predicate.load_eliminater
 0.48% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.57% :     0.000013s :     91: predicate.loop_unroll_before_grad
 1.74% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.43% :     0.000005s :     44: predicate.merge_addn
 1.48% :     0.000006s :     44: predicate.minmaximum_grad
 0.48% :     0.000002s :      4: predicate.mutable_eliminate
 0.18% :     0.000001s :      4: predicate.opt_reshape
 2.37% :     0.000009s :     55: predicate.partial_eliminate
 1.55% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.81% :     0.000007s :     44: predicate.reduce_eliminate
 1.77% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.40% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.54% :     0.000010s :     97: predicate.replace_applicator
 0.93% :     0.000004s :     40: predicate.replace_old_param
 0.11% :     0.000000s :      4: predicate.reset_defer_inline
 1.65% :     0.000006s :     44: predicate.reshape_eliminate
 1.55% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.31% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.61% :     0.000006s :     44: predicate.same_eliminate
 0.49% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.35% :     0.000001s :      8: predicate.special_op_eliminate
 0.85% :     0.000003s :     19: predicate.specialize_transform
 1.72% :     0.000007s :     44: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.16% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.91% :     0.000011s :     65: predicate.switch_defer_inline
 2.61% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.73% :     0.000025s :    162: predicate.switch_simplify
 1.47% :     0.000006s :     44: predicate.tile_eliminate
 1.46% :     0.000006s :     44: predicate.transpose_eliminate
 1.94% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.72% :     0.000007s :     44: predicate.tuple_list_get_item_depend_reorder
 3.17% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.82% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.87% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000007s :     44: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.24% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001566    32
62.50% :     0.000979s :     14: func_graph_cloner_run.FuncGraphClonerGraph
37.50% :     0.000587s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038442    91
 0.14% :     0.000054s :      1: add_recomputation
 0.30% :     0.000117s :      1: auto_monad
 0.08% :     0.000029s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.62% :     0.000239s :      1: bootstrap
 0.07% :     0.000025s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.07% :     0.000027s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.17% :     0.000066s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
28.81% :     0.011076s :      1: jit_opt_a
 0.40% :     0.000154s :      1: jit_opt_after_cconv
 0.15% :     0.000057s :      1: jit_opt_b
 1.08% :     0.000415s :      1: loop_unroll
 1.34% :     0.000514s :      1: mutable_eliminate
 6.90% :     0.002653s :     39: opt.transform.jit_opt_a
 0.14% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000030s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.29% :     0.000495s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.09% :     0.000034s :      1: py_interpret_to_execute
 0.03% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 3.65% :     0.001402s :      2: renormalize.infer
 3.37% :     0.001297s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000042s :      1: rewriter_after_opt_a
 0.31% :     0.000117s :      1: rewriter_before_opt_a
 0.22% :     0.000083s :      1: symbol_engine_optimizer
17.50% :     0.006727s :      1: task_emit
32.46% :     0.012479s :      1: type_inference
 0.14% :     0.000053s :      1: validate


TotalTime = 0.0157503, [33]
[bootstrap]: 0.00027989
[type_inference]: 0.00395569
[event_method]: 1.614e-05
[auto_monad]: 5.08e-05
[graph_reusing]: 4.27e-06
[pre_auto_parallel]: 1.76e-06
[py_interpret_to_execute]: 2.781e-05
[rewriter_before_opt_a]: 6.444e-05
[expand_dump_flag]: 3.01999e-06
[jit_opt_a]: 0.00461695, [2]
    [Cycle 1]: 0.00122224, [27]
        [switch_simplify]: 3.957e-05
        [loop_unroll]: 2.836e-05
        [a_1]: 0.00047227
        [with_stream_mark]: 1.01e-05
        [recompute_prepare]: 6.39999e-06
        [updatestate_depend_eliminate]: 2.61e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 5.57001e-06
        [updatestate_useless_node_eliminater]: 4.97999e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 1.17999e-06
        [get_grad_eliminate_]: 4.84e-06
        [merge_forward]: 2.39999e-06
        [cell_reuse_recompute_pass]: 1.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.05e-05
        [j_node_and_user_rematch]: 8.35001e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 7.51001e-06
        [inline_without_move]: 5.04e-06
        [renormalize]: 0.00041418
        [add_forward_monad_depend]: 4.29002e-06
        [auto_monad_grad]: 1.09998e-06
        [auto_monad_eliminator]: 8.50001e-06
        [cse]: 1.346e-05
        [replace_applicator]: 1.092e-05
    [Cycle 2]: 0.00030747, [27]
        [switch_simplify]: 5.10001e-06
        [loop_unroll]: 4.71002e-06
        [a_1]: 8.734e-05
        [with_stream_mark]: 7.88001e-06
        [recompute_prepare]: 4.60001e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 2.15002e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.04003e-06
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.11002e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 2.11e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.127e-05
        [j_node_and_user_rematch]: 7.5e-06
        [meta_fg_expand]: 1.58002e-06
        [replace_old_param]: 7.03998e-06
        [inline_without_move]: 4.36002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 7.79983e-07
        [auto_monad_eliminator]: 4.87e-06
        [cse]: 9.75002e-06
        [replace_applicator]: 4.86002e-06
[py_interpret_to_execute_after_opt_a]: 7.06999e-06
[rewriter_after_opt_a]: 2.273e-05
[convert_after_rewriter]: 5.72999e-06
[order_py_execute_after_rewriter]: 4.05e-06
[mutable_eliminate]: 0.00047237
[jit_opt_b]: 4.662e-05, [1]
    [Cycle 1]: 4.044e-05, [2]
        [frontend_op_eliminate]: 1.575e-05
        [inline_after_opt_a]: 1.365e-05
[cconv]: 1.494e-05
[loop_unroll]: 0.00039958
[jit_opt_after_cconv]: 0.00014935, [1]
    [Cycle 1]: 0.00014361, [11]
        [c_1]: 3.443e-05
        [parameter_eliminate]: 2.19001e-06
        [updatestate_depend_eliminate]: 4.25e-06
        [updatestate_assign_eliminate]: 7.43e-06
        [updatestate_loads_eliminate]: 2.30002e-06
        [cse]: 1.456e-05
        [call_graph_tuple_transform]: 1.567e-05
        [tuple_list_get_item_eliminator]: 5.22999e-06
        [none_parameter_eliminate]: 1.05999e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 5.10001e-06
[remove_dup_value]: 6.77002e-06
[partial_unused_args_eliminate]: 1.44e-06
[environ_conv]: 3.38999e-06
[add_recomputation]: 3.414e-05
[cse_after_recomputation]: 1.934e-05, [1]
    [Cycle 1]: 1.426e-05, [1]
        [cse]: 8.14997e-06
[auto_monad_reorder]: 8.95999e-06
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 5.04e-06
[opt_after_jit_grad]: 0.00045039
[symbol_engine_optimizer]: 6.7e-05, [1]
    [Cycle 1]: 6.118e-05, [6]
        [build]: 2.39999e-06
        [elim_shapecalc]: 7.15998e-06
        [elim_not_effective]: 1.143e-05
        [opt_reshape]: 5.23002e-06
        [fold_const_symbol]: 8.18999e-06
        [renormalize]: 5.00004e-07
[validate]: 2.36e-05
[backend_pass]: 7.29982e-07
[task_emit]: 0.00480107
[execute]: 1.662e-05
Sums
    bootstrap                            :     0.000280s :  2.31%
    type_inference                       :     0.003956s : 32.64%
    event_method                         :     0.000016s :  0.13%
    auto_monad                           :     0.000051s :  0.42%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.23%
    rewriter_before_opt_a                :     0.000064s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000045s :  0.37%
    jit_opt_a.loop_unroll                :     0.000033s :  0.27%
    jit_opt_a.a_1                        :     0.000560s :  4.62%
    jit_opt_a.with_stream_mark           :     0.000018s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000009s :  0.08%
    jit_opt_a.renormalize                :     0.000414s :  3.42%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.11%
    jit_opt_a.cse                        :     0.000023s :  0.19%
    jit_opt_a.replace_applicator         :     0.000016s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.06%
    rewriter_after_opt_a                 :     0.000023s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000472s :  3.90%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000015s :  0.12%
    loop_unroll                          :     0.000400s :  3.30%
    jit_opt_after_cconv.c_1              :     0.000034s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000007s :  0.06%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000007s :  0.06%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000003s :  0.03%
    add_recomputation                    :     0.000034s :  0.28%
    cse_after_recomputation.cse          :     0.000008s :  0.07%
    auto_monad_reorder                   :     0.000009s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000450s :  3.72%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000024s :  0.19%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.004801s : 39.61%
    execute                              :     0.000017s :  0.14%


Time group info:
------[substitution.]   0.000144    23
 0.84% :     0.000001s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.79% :     0.000004s :      3: substitution.graph_param_transform
84.63% :     0.000122s :      4: substitution.inline
 1.69% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.61% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.76% :     0.000003s :      2: substitution.replace_old_param
 5.00% :     0.000007s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003908     2
81.10% :     0.003170s :      1: type_inference.infer
18.90% :     0.000739s :      1: type_inference.specialize
------[replace.]   0.000057     6
66.59% :     0.000038s :      4: replace.inline
33.41% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000126     6
95.08% :     0.000120s :      4: match.inline
 4.92% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.48% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.02% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.05% :     0.000003s :     10: predicate.arithmetic_simplify
 1.44% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.05% :     0.000001s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.02% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.73% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.41% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.99% :     0.000001s :     10: predicate.environ_get_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.80% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 4.31% :     0.000005s :     22: predicate.inline
 0.76% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000002s :      6: predicate.less_batch_normalization
 1.37% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.60% :     0.000002s :     15: predicate.load_eliminater
 1.05% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.53% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 1.26% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
12.99% :     0.000016s :     15: predicate.partial_eliminate
 1.37% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.32% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.44% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.53% :     0.000002s :     10: predicate.reshape_eliminate
 1.06% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.10% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.06% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.11% :     0.000003s :     16: predicate.switch_defer_inline
 1.87% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.40% :     0.000008s :     53: predicate.switch_simplify
 1.02% :     0.000001s :     10: predicate.tile_eliminate
 1.39% :     0.000002s :     10: predicate.transpose_eliminate
 1.20% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.04% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.14% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.49% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.50% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000504    12
41.46% :     0.000209s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.91% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
53.63% :     0.000271s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017027    76
 0.22% :     0.000037s :      1: add_recomputation
 0.32% :     0.000054s :      1: auto_monad
 0.07% :     0.000011s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.70% :     0.000290s :      1: bootstrap
 0.10% :     0.000017s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.13% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.12% :     0.000021s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
27.13% :     0.004619s :      1: jit_opt_a
 0.89% :     0.000152s :      1: jit_opt_after_cconv
 0.29% :     0.000049s :      1: jit_opt_b
 2.39% :     0.000406s :      1: loop_unroll
 2.83% :     0.000481s :      1: mutable_eliminate
 4.36% :     0.000743s :     26: opt.transform.jit_opt_a
 0.34% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.69% :     0.000458s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000030s :      1: py_interpret_to_execute
 0.05% :     0.000009s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000009s :      1: remove_dup_value
 1.20% :     0.000205s :      1: renormalize.infer
 1.20% :     0.000204s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000027s :      1: rewriter_after_opt_a
 0.40% :     0.000067s :      1: rewriter_before_opt_a
 0.41% :     0.000069s :      1: symbol_engine_optimizer
28.30% :     0.004819s :      1: task_emit
23.28% :     0.003964s :      1: type_inference
 0.23% :     0.000038s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.808.174 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc5cee0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.808.816 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0249745, [33]
[bootstrap]: 0.00028245
[type_inference]: 0.00493893
[event_method]: 1.758e-05
[auto_monad]: 6.609e-05
[graph_reusing]: 6.53e-06
[pre_auto_parallel]: 2.37001e-06
[py_interpret_to_execute]: 4.109e-05
[rewriter_before_opt_a]: 7.302e-05
[expand_dump_flag]: 3.10002e-06
[jit_opt_a]: 0.0061927, [2]
    [Cycle 1]: 0.00204926, [27]
        [switch_simplify]: 6.548e-05
        [loop_unroll]: 2.93e-05
        [a_1]: 0.00058582
        [with_stream_mark]: 2.422e-05
        [recompute_prepare]: 7.35e-06
        [updatestate_depend_eliminate]: 3.66999e-06
        [updatestate_assign_eliminate]: 3.35e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 6.34001e-06
        [updatestate_useless_node_eliminater]: 4.73001e-06
        [accelerated_algorithm]: 5.99999e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 5.52999e-06
        [merge_forward]: 4.63999e-06
        [cell_reuse_recompute_pass]: 1.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.537e-05
        [j_node_and_user_rematch]: 9.97001e-06
        [meta_fg_expand]: 2.18998e-06
        [replace_old_param]: 9.39998e-06
        [inline_without_move]: 6.44999e-06
        [renormalize]: 0.00100301
        [add_forward_monad_depend]: 7.11999e-06
        [auto_monad_grad]: 2.79999e-06
        [auto_monad_eliminator]: 1.571e-05
        [cse]: 2.922e-05
        [replace_applicator]: 1.409e-05
    [Cycle 2]: 0.0003315, [27]
        [switch_simplify]: 5.82999e-06
        [loop_unroll]: 4.94e-06
        [a_1]: 9.105e-05
        [with_stream_mark]: 1.106e-05
        [recompute_prepare]: 5.09998e-06
        [updatestate_depend_eliminate]: 3.07002e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.01998e-06
        [parameter_eliminate]: 1.30999e-06
        [specialize_transform]: 4.92999e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.64998e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 3.16999e-06
        [cell_reuse_recompute_pass]: 2.48998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.537e-05
        [j_node_and_user_rematch]: 8.84e-06
        [meta_fg_expand]: 1.59998e-06
        [replace_old_param]: 8.23999e-06
        [inline_without_move]: 4.97999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 1.29e-06
        [auto_monad_eliminator]: 5.28002e-06
        [cse]: 1.064e-05
        [replace_applicator]: 5.12999e-06
[py_interpret_to_execute_after_opt_a]: 1.306e-05
[rewriter_after_opt_a]: 3.904e-05
[convert_after_rewriter]: 7.61999e-06
[order_py_execute_after_rewriter]: 5.22e-06
[mutable_eliminate]: 0.00079163
[jit_opt_b]: 5.574e-05, [1]
    [Cycle 1]: 4.657e-05, [2]
        [frontend_op_eliminate]: 1.816e-05
        [inline_after_opt_a]: 1.596e-05
[cconv]: 0.00345211
[loop_unroll]: 0.00092355
[jit_opt_after_cconv]: 0.0002092, [1]
    [Cycle 1]: 0.00019901, [11]
        [c_1]: 2.75e-05
        [parameter_eliminate]: 7.50998e-06
        [updatestate_depend_eliminate]: 1.341e-05
        [updatestate_assign_eliminate]: 3.36999e-06
        [updatestate_loads_eliminate]: 3.37002e-06
        [cse]: 4.249e-05
        [call_graph_tuple_transform]: 3.038e-05
        [tuple_list_get_item_eliminator]: 7e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 7.89994e-07
        [switch_simplify]: 6.71999e-06
[remove_dup_value]: 1.785e-05
[partial_unused_args_eliminate]: 2.76999e-06
[environ_conv]: 6.16e-06
[add_recomputation]: 6.473e-05
[cse_after_recomputation]: 2.441e-05, [1]
    [Cycle 1]: 1.777e-05, [1]
        [cse]: 1.143e-05
[auto_monad_reorder]: 1.592e-05
[get_jit_bprop_graph]: 3.11999e-06
[rewriter_after_jit_bprop_graph]: 1.454e-05
[opt_after_jit_grad]: 0.00057194
[symbol_engine_optimizer]: 8.454e-05, [1]
    [Cycle 1]: 7.564e-05, [6]
        [build]: 5.34e-06
        [elim_shapecalc]: 8.23001e-06
        [elim_not_effective]: 1.553e-05
        [opt_reshape]: 6.58e-06
        [fold_const_symbol]: 9.17999e-06
        [renormalize]: 7.7e-07
[validate]: 5.204e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.00672623
[execute]: 9.65002e-06
Sums
    bootstrap                            :     0.000282s :  1.38%
    type_inference                       :     0.004939s : 24.13%
    event_method                         :     0.000018s :  0.09%
    auto_monad                           :     0.000066s :  0.32%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.20%
    rewriter_before_opt_a                :     0.000073s :  0.36%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000071s :  0.35%
    jit_opt_a.loop_unroll                :     0.000034s :  0.17%
    jit_opt_a.a_1                        :     0.000677s :  3.31%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001003s :  4.90%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.10%
    jit_opt_a.cse                        :     0.000040s :  0.19%
    jit_opt_a.replace_applicator         :     0.000019s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.06%
    rewriter_after_opt_a                 :     0.000039s :  0.19%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000792s :  3.87%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.003452s : 16.86%
    loop_unroll                          :     0.000924s :  4.51%
    jit_opt_after_cconv.c_1              :     0.000028s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000013s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000042s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000030s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000018s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000065s :  0.32%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.08%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000015s :  0.07%
    opt_after_jit_grad                   :     0.000572s :  2.79%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000052s :  0.25%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006726s : 32.86%
    execute                              :     0.000010s :  0.05%


Time group info:
------[substitution.]   0.000253    23
 1.13% :     0.000003s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.71% :     0.000007s :      3: substitution.graph_param_transform
84.87% :     0.000215s :      4: substitution.inline
 1.88% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.81% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.97% :     0.000005s :      2: substitution.replace_old_param
 5.13% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004868     2
78.88% :     0.003840s :      1: type_inference.infer
21.12% :     0.001028s :      1: type_inference.specialize
------[replace.]   0.000073     6
67.58% :     0.000049s :      4: replace.inline
32.42% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000224     6
94.66% :     0.000212s :      4: match.inline
 5.34% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.18% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.71% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000002s :     10: predicate.addn_zero_filter
 2.14% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000001s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.06% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.34% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.45% :     0.000001s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.41% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.03% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000001s :      3: predicate.graph_param_transform
 6.35% :     0.000008s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.70% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 1.96% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.78% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.32% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 2.19% :     0.000003s :      3: predicate.mutable_eliminate
 0.60% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.91% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.47% :     0.000001s :      3: predicate.reset_defer_inline
 1.28% :     0.000002s :     10: predicate.reshape_eliminate
 1.35% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.11% :     0.000001s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.18% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.68% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 1.93% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.80% :     0.000010s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.22% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.09% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.37% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.33% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.67% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.38% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.83% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.004101    12
90.48% :     0.003710s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 1.28% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
 8.24% :     0.000338s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027019    76
 0.25% :     0.000068s :      1: add_recomputation
 0.26% :     0.000069s :      1: auto_monad
 0.07% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.09% :     0.000293s :      1: bootstrap
12.87% :     0.003476s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.08% :     0.000022s :      1: event_method
 0.06% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
22.93% :     0.006197s :      1: jit_opt_a
 0.79% :     0.000212s :      1: jit_opt_after_cconv
 0.22% :     0.000058s :      1: jit_opt_b
 3.47% :     0.000938s :      1: loop_unroll
 2.97% :     0.000803s :      1: mutable_eliminate
 3.35% :     0.000904s :     26: opt.transform.jit_opt_a
 0.25% :     0.000068s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000027s :      4: opt.transform.jit_opt_b
 0.09% :     0.000025s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.16% :     0.000583s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000044s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000021s :      1: remove_dup_value
 2.04% :     0.000550s :      1: renormalize.infer
 1.64% :     0.000442s :      1: renormalize.specialize
 0.06% :     0.000017s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000042s :      1: rewriter_after_opt_a
 0.28% :     0.000076s :      1: rewriter_before_opt_a
 0.32% :     0.000087s :      1: symbol_engine_optimizer
24.97% :     0.006746s :      1: task_emit
18.34% :     0.004955s :      1: type_inference
 0.28% :     0.000075s :      1: validate


TotalTime = 0.017839, [33]
[bootstrap]: 0.00030667
[type_inference]: 0.00426483
[event_method]: 1.795e-05
[auto_monad]: 4.85e-05
[graph_reusing]: 3.75e-06
[pre_auto_parallel]: 2.65002e-06
[py_interpret_to_execute]: 2.783e-05
[rewriter_before_opt_a]: 6.269e-05
[expand_dump_flag]: 1.99999e-06
[jit_opt_a]: 0.00505903, [2]
    [Cycle 1]: 0.00124717, [27]
        [switch_simplify]: 4.128e-05
        [loop_unroll]: 2.795e-05
        [a_1]: 0.00043867
        [with_stream_mark]: 1.014e-05
        [recompute_prepare]: 6.66e-06
        [updatestate_depend_eliminate]: 2.69001e-06
        [updatestate_assign_eliminate]: 1.91998e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 5.62001e-06
        [updatestate_useless_node_eliminater]: 4.72998e-06
        [accelerated_algorithm]: 5.09003e-06
        [meta_shard_fg_expand]: 1.16002e-06
        [get_grad_eliminate_]: 4.65999e-06
        [merge_forward]: 2.18998e-06
        [cell_reuse_recompute_pass]: 9.30013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.049e-05
        [j_node_and_user_rematch]: 7.90998e-06
        [meta_fg_expand]: 1.73002e-06
        [replace_old_param]: 8.04002e-06
        [inline_without_move]: 5.00999e-06
        [renormalize]: 0.00045208
        [add_forward_monad_depend]: 4.58001e-06
        [auto_monad_grad]: 2.22001e-06
        [auto_monad_eliminator]: 1.335e-05
        [cse]: 2.708e-05
        [replace_applicator]: 9.41e-06
    [Cycle 2]: 0.00030781, [27]
        [switch_simplify]: 5.07e-06
        [loop_unroll]: 4.52e-06
        [a_1]: 8.565e-05
        [with_stream_mark]: 7.43e-06
        [recompute_prepare]: 4.70999e-06
        [updatestate_depend_eliminate]: 2.36998e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.57e-06
        [updatestate_useless_node_eliminater]: 4.64002e-06
        [accelerated_algorithm]: 4.39998e-06
        [meta_shard_fg_expand]: 1.14998e-06
        [get_grad_eliminate_]: 4.27e-06
        [merge_forward]: 1.305e-05
        [cell_reuse_recompute_pass]: 1.71998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.124e-05
        [j_node_and_user_rematch]: 7.26001e-06
        [meta_fg_expand]: 1.45999e-06
        [replace_old_param]: 6.48e-06
        [inline_without_move]: 4.45e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 8.79983e-07
        [auto_monad_eliminator]: 4.82e-06
        [cse]: 9.74999e-06
        [replace_applicator]: 4.65999e-06
[py_interpret_to_execute_after_opt_a]: 7.31999e-06
[rewriter_after_opt_a]: 3.019e-05
[convert_after_rewriter]: 7.17002e-06
[order_py_execute_after_rewriter]: 5.12e-06
[mutable_eliminate]: 0.00046478
[jit_opt_b]: 4.476e-05, [1]
    [Cycle 1]: 3.863e-05, [2]
        [frontend_op_eliminate]: 1.38e-05
        [inline_after_opt_a]: 1.395e-05
[cconv]: 1.843e-05
[loop_unroll]: 0.00040665
[jit_opt_after_cconv]: 0.00012687, [1]
    [Cycle 1]: 0.00012066, [11]
        [c_1]: 1.925e-05
        [parameter_eliminate]: 1.86e-06
        [updatestate_depend_eliminate]: 4.72e-06
        [updatestate_assign_eliminate]: 2.40002e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [cse]: 1.513e-05
        [call_graph_tuple_transform]: 1.584e-05
        [tuple_list_get_item_eliminator]: 4.96002e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 4.95001e-06
[remove_dup_value]: 1.242e-05
[partial_unused_args_eliminate]: 2.54001e-06
[environ_conv]: 4.65001e-06
[add_recomputation]: 4.458e-05
[cse_after_recomputation]: 1.99e-05, [1]
    [Cycle 1]: 1.473e-05, [1]
        [cse]: 9.00001e-06
[auto_monad_reorder]: 1.496e-05
[get_jit_bprop_graph]: 1.42999e-06
[rewriter_after_jit_bprop_graph]: 4.73001e-06
[opt_after_jit_grad]: 0.00043667
[symbol_engine_optimizer]: 6.734e-05, [1]
    [Cycle 1]: 6.138e-05, [6]
        [build]: 2.92002e-06
        [elim_shapecalc]: 6.94999e-06
        [elim_not_effective]: 1.152e-05
        [opt_reshape]: 5.30999e-06
        [fold_const_symbol]: 8.09002e-06
        [renormalize]: 3.89991e-07
[validate]: 2.762e-05
[backend_pass]: 1.17e-06
[task_emit]: 0.00611342
[execute]: 7.45e-06
Sums
    bootstrap                            :     0.000307s :  2.22%
    type_inference                       :     0.004265s : 30.92%
    event_method                         :     0.000018s :  0.13%
    auto_monad                           :     0.000049s :  0.35%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.20%
    rewriter_before_opt_a                :     0.000063s :  0.45%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000046s :  0.34%
    jit_opt_a.loop_unroll                :     0.000032s :  0.24%
    jit_opt_a.a_1                        :     0.000524s :  3.80%
    jit_opt_a.with_stream_mark           :     0.000018s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000009s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000015s :  0.11%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000015s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000452s :  3.28%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000037s :  0.27%
    jit_opt_a.replace_applicator         :     0.000014s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.05%
    rewriter_after_opt_a                 :     0.000030s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000465s :  3.37%
    jit_opt_b.frontend_op_eliminate      :     0.000014s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000018s :  0.13%
    loop_unroll                          :     0.000407s :  2.95%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000045s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000437s :  3.17%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006113s : 44.33%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000136    23
 1.32% :     0.000002s :      2: substitution.elim_not_effective
 0.91% :     0.000001s :      2: substitution.fold_const_symbol
 3.69% :     0.000005s :      3: substitution.graph_param_transform
81.88% :     0.000111s :      4: substitution.inline
 1.72% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.92% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.40% :     0.000003s :      2: substitution.replace_old_param
 5.16% :     0.000007s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004210     2
82.98% :     0.003493s :      1: type_inference.infer
17.02% :     0.000717s :      1: type_inference.specialize
------[replace.]   0.000054     6
66.58% :     0.000036s :      4: replace.inline
33.42% :     0.000018s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000115     6
94.73% :     0.000109s :      4: match.inline
 5.27% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000107   691
 1.56% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000001s :     10: predicate.addn_zero_filter
 1.96% :     0.000002s :     10: predicate.arithmetic_simplify
 1.38% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.20% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.63% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.29% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.25% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.22% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.22% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.22% :     0.000001s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000002s :     15: predicate.partial_eliminate
 1.36% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.73% :     0.000002s :     10: predicate.reduce_eliminate
 1.57% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.58% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.39% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.51% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.46% :     0.000008s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.37% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.41% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.60% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000475    12
49.46% :     0.000235s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.56% :     0.000022s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.98% :     0.000219s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019093    76
 0.25% :     0.000047s :      1: add_recomputation
 0.27% :     0.000052s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.70% :     0.000324s :      1: bootstrap
 0.11% :     0.000021s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
26.51% :     0.005061s :      1: jit_opt_a
 0.68% :     0.000129s :      1: jit_opt_after_cconv
 0.25% :     0.000047s :      1: jit_opt_b
 2.17% :     0.000414s :      1: loop_unroll
 2.48% :     0.000473s :      1: mutable_eliminate
 3.70% :     0.000706s :     26: opt.transform.jit_opt_a
 0.22% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000022s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.32% :     0.000443s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000030s :      1: py_interpret_to_execute
 0.05% :     0.000009s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.28% :     0.000245s :      1: renormalize.infer
 1.05% :     0.000201s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000033s :      1: rewriter_after_opt_a
 0.34% :     0.000065s :      1: rewriter_before_opt_a
 0.36% :     0.000070s :      1: symbol_engine_optimizer
32.07% :     0.006123s :      1: task_emit
22.41% :     0.004278s :      1: type_inference
 0.24% :     0.000046s :      1: validate


TotalTime = 0.0326421, [33]
[bootstrap]: 0.00023846
[type_inference]: 0.0129191
[event_method]: 5.89e-05
[auto_monad]: 0.00010994
[graph_reusing]: 7.3e-06
[pre_auto_parallel]: 3.23e-06
[py_interpret_to_execute]: 2.941e-05
[rewriter_before_opt_a]: 0.00010782
[expand_dump_flag]: 3.34001e-06
[jit_opt_a]: 0.0111464, [3]
    [Cycle 1]: 0.00531754, [27]
        [switch_simplify]: 0.00012511
        [loop_unroll]: 4.711e-05
        [a_1]: 0.00090194
        [with_stream_mark]: 2.571e-05
        [recompute_prepare]: 1.955e-05
        [updatestate_depend_eliminate]: 8.10999e-06
        [updatestate_assign_eliminate]: 6.95002e-06
        [updatestate_loads_eliminate]: 6.61999e-06
        [parameter_eliminate]: 2.76e-06
        [specialize_transform]: 1.395e-05
        [updatestate_useless_node_eliminater]: 1.294e-05
        [accelerated_algorithm]: 1.324e-05
        [meta_shard_fg_expand]: 3.46999e-06
        [get_grad_eliminate_]: 1.228e-05
        [merge_forward]: 8.80999e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.732e-05
        [j_node_and_user_rematch]: 2.422e-05
        [meta_fg_expand]: 0.00132209
        [replace_old_param]: 4.855e-05
        [inline_without_move]: 5.209e-05
        [renormalize]: 0.00217779
        [add_forward_monad_depend]: 7.73001e-06
        [auto_monad_grad]: 4.66002e-06
        [auto_monad_eliminator]: 4.436e-05
        [cse]: 0.00014012
        [replace_applicator]: 6.102e-05
    [Cycle 2]: 0.0019495, [27]
        [switch_simplify]: 3.732e-05
        [loop_unroll]: 3.512e-05
        [a_1]: 0.0010543
        [with_stream_mark]: 9.96e-06
        [recompute_prepare]: 8.55001e-06
        [updatestate_depend_eliminate]: 3.4e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.61999e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 7.28999e-06
        [updatestate_useless_node_eliminater]: 6.79999e-06
        [accelerated_algorithm]: 6.38998e-06
        [meta_shard_fg_expand]: 1.68002e-06
        [get_grad_eliminate_]: 6.09999e-06
        [merge_forward]: 3.12002e-06
        [cell_reuse_recompute_pass]: 8.39995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.288e-05
        [j_node_and_user_rematch]: 1.081e-05
        [meta_fg_expand]: 6.985e-05
        [replace_old_param]: 1.03e-05
        [inline_without_move]: 6.48998e-06
        [renormalize]: 0.00045833
        [add_forward_monad_depend]: 3.87002e-06
        [auto_monad_grad]: 1.07e-06
        [auto_monad_eliminator]: 1.069e-05
        [cse]: 1.914e-05
        [replace_applicator]: 1.254e-05
    [Cycle 3]: 0.0004358, [27]
        [switch_simplify]: 6.83e-06
        [loop_unroll]: 6.26998e-06
        [a_1]: 0.00017689
        [with_stream_mark]: 9.34e-06
        [recompute_prepare]: 6.53e-06
        [updatestate_depend_eliminate]: 3.58999e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.61999e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 6.69999e-06
        [updatestate_useless_node_eliminater]: 6.36998e-06
        [accelerated_algorithm]: 6.12999e-06
        [meta_shard_fg_expand]: 1.43002e-06
        [get_grad_eliminate_]: 5.98998e-06
        [merge_forward]: 2.96999e-06
        [cell_reuse_recompute_pass]: 1.47999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.453e-05
        [j_node_and_user_rematch]: 1.027e-05
        [meta_fg_expand]: 2.07999e-06
        [replace_old_param]: 8.51002e-06
        [inline_without_move]: 6.23e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 7.31999e-06
        [cse]: 1.566e-05
        [replace_applicator]: 6.32001e-06
[py_interpret_to_execute_after_opt_a]: 9.36998e-06
[rewriter_after_opt_a]: 3.172e-05
[convert_after_rewriter]: 6.41e-06
[order_py_execute_after_rewriter]: 5.44998e-06
[mutable_eliminate]: 0.00048539
[jit_opt_b]: 5.515e-05, [1]
    [Cycle 1]: 4.876e-05, [2]
        [frontend_op_eliminate]: 1.986e-05
        [inline_after_opt_a]: 1.793e-05
[cconv]: 1.544e-05
[loop_unroll]: 0.00041403
[jit_opt_after_cconv]: 0.00016214, [1]
    [Cycle 1]: 0.00015592, [11]
        [c_1]: 2.642e-05
        [parameter_eliminate]: 2.46998e-06
        [updatestate_depend_eliminate]: 5.58002e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [cse]: 2.842e-05
        [call_graph_tuple_transform]: 2.167e-05
        [tuple_list_get_item_eliminator]: 6.84999e-06
        [none_parameter_eliminate]: 1.12e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 6.93e-06
[remove_dup_value]: 1.196e-05
[partial_unused_args_eliminate]: 1.77999e-06
[environ_conv]: 4.57e-06
[add_recomputation]: 4.308e-05
[cse_after_recomputation]: 2.502e-05, [1]
    [Cycle 1]: 1.987e-05, [1]
        [cse]: 1.406e-05
[auto_monad_reorder]: 1.369e-05
[get_jit_bprop_graph]: 1.26997e-06
[rewriter_after_jit_bprop_graph]: 4.63999e-06
[opt_after_jit_grad]: 0.00045248
[symbol_engine_optimizer]: 7.782e-05, [1]
    [Cycle 1]: 7.187e-05, [6]
        [build]: 3.03998e-06
        [elim_shapecalc]: 8.67998e-06
        [elim_not_effective]: 1.45e-05
        [opt_reshape]: 6.88e-06
        [fold_const_symbol]: 1.103e-05
        [renormalize]: 5.10016e-07
[validate]: 3.168e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.00596094
[execute]: 4.94003e-06
Sums
    bootstrap                            :     0.000238s :  0.84%
    type_inference                       :     0.012919s : 45.45%
    event_method                         :     0.000059s :  0.21%
    auto_monad                           :     0.000110s :  0.39%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000029s :  0.10%
    rewriter_before_opt_a                :     0.000108s :  0.38%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000169s :  0.60%
    jit_opt_a.loop_unroll                :     0.000089s :  0.31%
    jit_opt_a.a_1                        :     0.002133s :  7.51%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000035s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000028s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000026s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.09%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000055s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000045s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001394s :  4.90%
    jit_opt_a.replace_old_param          :     0.000067s :  0.24%
    jit_opt_a.inline_without_move        :     0.000065s :  0.23%
    jit_opt_a.renormalize                :     0.002636s :  9.27%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000062s :  0.22%
    jit_opt_a.cse                        :     0.000175s :  0.62%
    jit_opt_a.replace_applicator         :     0.000080s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.03%
    rewriter_after_opt_a                 :     0.000032s :  0.11%
    convert_after_rewriter               :     0.000006s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000485s :  1.71%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000015s :  0.05%
    loop_unroll                          :     0.000414s :  1.46%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000012s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000043s :  0.15%
    cse_after_recomputation.cse          :     0.000014s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000452s :  1.59%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005961s : 20.97%
    execute                              :     0.000005s :  0.02%


Time group info:
------[substitution.]   0.000466   113
 0.48% :     0.000002s :      3: substitution.elim_not_effective
 0.30% :     0.000001s :      3: substitution.fold_const_symbol
 1.21% :     0.000006s :      4: substitution.graph_param_transform
63.13% :     0.000294s :     14: substitution.inline
 3.18% :     0.000015s :      2: substitution.inline_without_move
 1.94% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.13% :     0.000010s :      7: substitution.minmaximum_grad
 1.23% :     0.000006s :      5: substitution.partial_eliminate
 2.41% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.21% :     0.000024s :     10: substitution.replace_applicator
 1.37% :     0.000006s :      7: substitution.replace_old_param
 0.57% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.64% :     0.000008s :      1: substitution.switch_simplify
 4.10% :     0.000019s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.01% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 8.08% :     0.000038s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012845     2
89.43% :     0.011487s :      1: type_inference.infer
10.57% :     0.001358s :      1: type_inference.specialize
------[replace.]   0.000180    22
53.23% :     0.000096s :     14: replace.inline
18.85% :     0.000034s :      1: replace.switch_simplify
27.92% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000313    22
91.31% :     0.000286s :     14: match.inline
 2.23% :     0.000007s :      1: match.switch_simplify
 6.46% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000398  2634
 1.57% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.44% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.44% :     0.000006s :     44: predicate.addn_check_dump
 1.71% :     0.000007s :     44: predicate.addn_zero_filter
 1.96% :     0.000008s :     44: predicate.arithmetic_simplify
 1.50% :     0.000006s :     44: predicate.cast_eliminate
 0.18% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.44% :     0.000006s :     44: predicate.compare_switch_simplify
 1.56% :     0.000006s :     44: predicate.depend_value_elim
 1.47% :     0.000006s :     44: predicate.dict_get_item_const_eliminator
 1.57% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.51% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.24% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      4: predicate.elim_not_effective
 0.24% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.43% :     0.000006s :     44: predicate.environ_get_add_eliminate
 1.39% :     0.000006s :     44: predicate.environ_get_depend_swap
 1.46% :     0.000006s :     44: predicate.environ_get_eliminate
 1.43% :     0.000006s :     44: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.83% :     0.000003s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.33% :     0.000017s :     73: predicate.inline
 1.81% :     0.000007s :     40: predicate.inline_without_move
 0.35% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.87% :     0.000003s :     19: predicate.less_batch_normalization
 1.76% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.95% :     0.000008s :     55: predicate.load_eliminater
 0.47% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.41% :     0.000014s :     91: predicate.loop_unroll_before_grad
 1.78% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.46% :     0.000006s :     44: predicate.merge_addn
 1.54% :     0.000006s :     44: predicate.minmaximum_grad
 0.51% :     0.000002s :      4: predicate.mutable_eliminate
 0.20% :     0.000001s :      4: predicate.opt_reshape
 2.35% :     0.000009s :     55: predicate.partial_eliminate
 1.50% :     0.000006s :     44: predicate.print_const_string_wrapper
 2.11% :     0.000008s :     44: predicate.reduce_eliminate
 1.76% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.51% :     0.000010s :     97: predicate.replace_applicator
 0.86% :     0.000003s :     40: predicate.replace_old_param
 0.10% :     0.000000s :      4: predicate.reset_defer_inline
 1.77% :     0.000007s :     44: predicate.reshape_eliminate
 1.53% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.34% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.61% :     0.000006s :     44: predicate.same_eliminate
 0.46% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.36% :     0.000001s :      8: predicate.special_op_eliminate
 0.88% :     0.000003s :     19: predicate.specialize_transform
 1.75% :     0.000007s :     44: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.18% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.82% :     0.000011s :     65: predicate.switch_defer_inline
 2.55% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.64% :     0.000026s :    162: predicate.switch_simplify
 1.50% :     0.000006s :     44: predicate.tile_eliminate
 1.46% :     0.000006s :     44: predicate.transpose_eliminate
 1.80% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.77% :     0.000007s :     44: predicate.tuple_list_get_item_depend_reorder
 3.23% :     0.000013s :     59: predicate.tuple_list_get_item_eliminator
 1.83% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.91% :     0.000008s :     55: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000012s :     74: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000007s :     44: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.23% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001649    32
60.08% :     0.000991s :     14: func_graph_cloner_run.FuncGraphClonerGraph
39.92% :     0.000658s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038205    91
 0.12% :     0.000046s :      1: add_recomputation
 0.30% :     0.000115s :      1: auto_monad
 0.04% :     0.000016s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.69% :     0.000262s :      1: bootstrap
 0.05% :     0.000018s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.07% :     0.000027s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.17% :     0.000064s :      1: event_method
 0.02% :     0.000008s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
29.18% :     0.011149s :      1: jit_opt_a
 0.43% :     0.000165s :      1: jit_opt_after_cconv
 0.15% :     0.000058s :      1: jit_opt_b
 1.10% :     0.000421s :      1: loop_unroll
 1.29% :     0.000494s :      1: mutable_eliminate
 7.33% :     0.002799s :     39: opt.transform.jit_opt_a
 0.15% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000032s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.20% :     0.000459s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000032s :      1: py_interpret_to_execute
 0.03% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000014s :      1: remove_dup_value
 3.60% :     0.001375s :      2: renormalize.infer
 3.27% :     0.001249s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000035s :      1: rewriter_after_opt_a
 0.29% :     0.000111s :      1: rewriter_before_opt_a
 0.21% :     0.000081s :      1: symbol_engine_optimizer
15.63% :     0.005971s :      1: task_emit
33.85% :     0.012932s :      1: type_inference
 0.13% :     0.000050s :      1: validate


TotalTime = 0.012753, [30]
[bootstrap]: 0.00029702
[type_inference]: 0.00454416
[event_method]: 1.696e-05
[auto_monad]: 6.577e-05
[graph_reusing]: 5.97999e-06
[pre_auto_parallel]: 2.73998e-06
[py_interpret_to_execute]: 3.115e-05
[rewriter_before_opt_a]: 7.281e-05
[expand_dump_flag]: 3.36999e-06
[jit_opt_a]: 0.0053252, [2]
    [Cycle 1]: 0.00153354, [27]
        [switch_simplify]: 5.68e-05
        [loop_unroll]: 3.058e-05
        [a_1]: 0.00051014
        [with_stream_mark]: 1.692e-05
        [recompute_prepare]: 7.1e-06
        [updatestate_depend_eliminate]: 3.42002e-06
        [updatestate_assign_eliminate]: 3.37997e-06
        [updatestate_loads_eliminate]: 2.49999e-06
        [parameter_eliminate]: 2.07999e-06
        [specialize_transform]: 5.94999e-06
        [updatestate_useless_node_eliminater]: 5.64998e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 5.40001e-06
        [merge_forward]: 3.76001e-06
        [cell_reuse_recompute_pass]: 1.05999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.482e-05
        [j_node_and_user_rematch]: 9.59e-06
        [meta_fg_expand]: 2.58e-06
        [replace_old_param]: 8.87999e-06
        [inline_without_move]: 5.31998e-06
        [renormalize]: 0.00061067
        [add_forward_monad_depend]: 5.52001e-06
        [auto_monad_grad]: 2.35002e-06
        [auto_monad_eliminator]: 1.305e-05
        [cse]: 2.018e-05
        [replace_applicator]: 1.172e-05
    [Cycle 2]: 0.00032049, [27]
        [switch_simplify]: 5.71998e-06
        [loop_unroll]: 5.19e-06
        [a_1]: 9.036e-05
        [with_stream_mark]: 8.1e-06
        [recompute_prepare]: 5.12999e-06
        [updatestate_depend_eliminate]: 2.57001e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 5.42001e-06
        [updatestate_useless_node_eliminater]: 5.12999e-06
        [accelerated_algorithm]: 5.07e-06
        [meta_shard_fg_expand]: 1.07998e-06
        [get_grad_eliminate_]: 4.53001e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.178e-05
        [j_node_and_user_rematch]: 7.88999e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 7.28e-06
        [inline_without_move]: 4.96002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 8.59989e-07
        [auto_monad_eliminator]: 5.21002e-06
        [cse]: 1.077e-05
        [replace_applicator]: 5.07e-06
[py_interpret_to_execute_after_opt_a]: 9.17999e-06
[rewriter_after_opt_a]: 3.247e-05
[convert_after_rewriter]: 6.93e-06
[order_py_execute_after_rewriter]: 5.34e-06
[mutable_eliminate]: 0.00084499
[jit_opt_b]: 5.933e-05, [1]
    [Cycle 1]: 5.13e-05, [2]
        [frontend_op_eliminate]: 1.976e-05
        [inline_after_opt_a]: 1.848e-05
[cconv]: 2.422e-05
[loop_unroll]: 0.00043459
[jit_opt_after_cconv]: 0.0001401, [1]
    [Cycle 1]: 0.00013332, [11]
        [c_1]: 2.121e-05
        [parameter_eliminate]: 2.27999e-06
        [updatestate_depend_eliminate]: 5.20001e-06
        [updatestate_assign_eliminate]: 2.43998e-06
        [updatestate_loads_eliminate]: 2.23002e-06
        [cse]: 1.82e-05
        [call_graph_tuple_transform]: 1.855e-05
        [tuple_list_get_item_eliminator]: 5.52001e-06
        [none_parameter_eliminate]: 1.51002e-06
        [renormalize]: 5.39992e-07
        [switch_simplify]: 5.52001e-06
[remove_dup_value]: 1.407e-05
[partial_unused_args_eliminate]: 2.40002e-06
[environ_conv]: 5.94999e-06
[add_recomputation]: 4.636e-05
[cse_after_recomputation]: 2.138e-05, [1]
    [Cycle 1]: 1.543e-05, [1]
        [cse]: 9.72001e-06
[auto_monad_reorder]: 1.509e-05
[get_jit_bprop_graph]: 1.96998e-06
[rewriter_after_jit_bprop_graph]: 6.49001e-06
[opt_after_jit_grad]: 0.00043783
[symbol_engine_optimizer]: 7.376e-05, [1]
    [Cycle 1]: 6.679e-05, [6]
        [build]: 3.45998e-06
        [elim_shapecalc]: 7.64002e-06
        [elim_not_effective]: 1.257e-05
        [opt_reshape]: 5.89e-06
        [fold_const_symbol]: 9.19e-06
        [renormalize]: 4.50003e-07
[validate]: 3.563e-05
Sums
    bootstrap                            :     0.000297s :  3.41%
    type_inference                       :     0.004544s : 52.20%
    event_method                         :     0.000017s :  0.19%
    auto_monad                           :     0.000066s :  0.76%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000031s :  0.36%
    rewriter_before_opt_a                :     0.000073s :  0.84%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000063s :  0.72%
    jit_opt_a.loop_unroll                :     0.000036s :  0.41%
    jit_opt_a.a_1                        :     0.000601s :  6.90%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000611s :  7.02%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.21%
    jit_opt_a.cse                        :     0.000031s :  0.36%
    jit_opt_a.replace_applicator         :     0.000017s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000032s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000845s :  9.71%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.21%
    cconv                                :     0.000024s :  0.28%
    loop_unroll                          :     0.000435s :  4.99%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000018s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000014s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000046s :  0.53%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000438s :  5.03%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000036s :  0.41%


Time group info:
------[substitution.]   0.000184    23
 1.13% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.88% :     0.000005s :      3: substitution.graph_param_transform
82.89% :     0.000153s :      4: substitution.inline
 1.66% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.36% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.17% :     0.000004s :      2: substitution.replace_old_param
 6.18% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004482     2
82.12% :     0.003681s :      1: type_inference.infer
17.88% :     0.000801s :      1: type_inference.specialize
------[replace.]   0.000063     6
66.58% :     0.000042s :      4: replace.inline
33.42% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000160     6
93.62% :     0.000150s :      4: match.inline
 6.38% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.21% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.01% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.44% :     0.000002s :     10: predicate.addn_zero_filter
 2.19% :     0.000003s :     10: predicate.arithmetic_simplify
 1.32% :     0.000002s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.50% :     0.000007s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.69% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 2.10% :     0.000003s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.34% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.17% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 1.14% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.46% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.94% :     0.000010s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.55% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.87% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.85% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000564    12
53.29% :     0.000300s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.62% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.09% :     0.000237s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014240    72
 0.35% :     0.000049s :      1: add_recomputation
 0.49% :     0.000069s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.21% :     0.000314s :      1: bootstrap
 0.19% :     0.000027s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000024s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.15% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
37.42% :     0.005328s :      1: jit_opt_a
 1.00% :     0.000143s :      1: jit_opt_after_cconv
 0.44% :     0.000062s :      1: jit_opt_b
 3.10% :     0.000442s :      1: loop_unroll
 6.00% :     0.000854s :      1: mutable_eliminate
 5.70% :     0.000812s :     26: opt.transform.jit_opt_a
 0.33% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000030s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.13% :     0.000446s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000034s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.18% :     0.000311s :      1: renormalize.infer
 2.06% :     0.000293s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000036s :      1: rewriter_after_opt_a
 0.53% :     0.000076s :      1: rewriter_before_opt_a
 0.54% :     0.000076s :      1: symbol_engine_optimizer
32.01% :     0.004558s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.862.672 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc41280> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.863.253 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0181196, [33]
[bootstrap]: 0.00036367
[type_inference]: 0.00449003
[event_method]: 1.77e-05
[auto_monad]: 6.515e-05
[graph_reusing]: 5.72001e-06
[pre_auto_parallel]: 3.31001e-06
[py_interpret_to_execute]: 2.842e-05
[rewriter_before_opt_a]: 7.473e-05
[expand_dump_flag]: 3.46999e-06
[jit_opt_a]: 0.00493546, [2]
    [Cycle 1]: 0.00138784, [27]
        [switch_simplify]: 4.901e-05
        [loop_unroll]: 2.878e-05
        [a_1]: 0.00052515
        [with_stream_mark]: 1.484e-05
        [recompute_prepare]: 7.48999e-06
        [updatestate_depend_eliminate]: 3.38e-06
        [updatestate_assign_eliminate]: 2.86999e-06
        [updatestate_loads_eliminate]: 2.88998e-06
        [parameter_eliminate]: 1.92999e-06
        [specialize_transform]: 6.28e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 5.91003e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 5.47001e-06
        [merge_forward]: 3.58e-06
        [cell_reuse_recompute_pass]: 1.26002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.484e-05
        [j_node_and_user_rematch]: 8.89003e-06
        [meta_fg_expand]: 2.18002e-06
        [replace_old_param]: 9.21998e-06
        [inline_without_move]: 5.62999e-06
        [renormalize]: 0.00044888
        [add_forward_monad_depend]: 5.34e-06
        [auto_monad_grad]: 1.68002e-06
        [auto_monad_eliminator]: 1.368e-05
        [cse]: 2.923e-05
        [replace_applicator]: 1.277e-05
    [Cycle 2]: 0.0003127, [27]
        [switch_simplify]: 5.29998e-06
        [loop_unroll]: 4.72e-06
        [a_1]: 8.723e-05
        [with_stream_mark]: 8.57e-06
        [recompute_prepare]: 4.70999e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 4.82998e-06
        [meta_shard_fg_expand]: 1.06002e-06
        [get_grad_eliminate_]: 4.46002e-06
        [merge_forward]: 2.50002e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.246e-05
        [j_node_and_user_rematch]: 7.84002e-06
        [meta_fg_expand]: 1.46002e-06
        [replace_old_param]: 7.61999e-06
        [inline_without_move]: 4.48999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.42001e-06
        [cse]: 1.009e-05
        [replace_applicator]: 4.92e-06
[py_interpret_to_execute_after_opt_a]: 8.18999e-06
[rewriter_after_opt_a]: 3.252e-05
[convert_after_rewriter]: 7.51001e-06
[order_py_execute_after_rewriter]: 5.40999e-06
[mutable_eliminate]: 0.00051065
[jit_opt_b]: 4.777e-05, [1]
    [Cycle 1]: 4.063e-05, [2]
        [frontend_op_eliminate]: 1.543e-05
        [inline_after_opt_a]: 1.379e-05
[cconv]: 2.359e-05
[loop_unroll]: 0.00041339
[jit_opt_after_cconv]: 0.00013341, [1]
    [Cycle 1]: 0.000127, [11]
        [c_1]: 2.014e-05
        [parameter_eliminate]: 2.44001e-06
        [updatestate_depend_eliminate]: 5.15001e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [cse]: 1.678e-05
        [call_graph_tuple_transform]: 1.742e-05
        [tuple_list_get_item_eliminator]: 5.39e-06
        [none_parameter_eliminate]: 1.55001e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 5.04998e-06
[remove_dup_value]: 1.331e-05
[partial_unused_args_eliminate]: 2.29001e-06
[environ_conv]: 4.64002e-06
[add_recomputation]: 4.757e-05
[cse_after_recomputation]: 1.944e-05, [1]
    [Cycle 1]: 1.43e-05, [1]
        [cse]: 8.78001e-06
[auto_monad_reorder]: 1.466e-05
[get_jit_bprop_graph]: 1.39998e-06
[rewriter_after_jit_bprop_graph]: 5.32999e-06
[opt_after_jit_grad]: 0.00043776
[symbol_engine_optimizer]: 6.859e-05, [1]
    [Cycle 1]: 6.254e-05, [6]
        [build]: 3.03998e-06
        [elim_shapecalc]: 6.97002e-06
        [elim_not_effective]: 1.195e-05
        [opt_reshape]: 5.16998e-06
        [fold_const_symbol]: 8.03001e-06
        [renormalize]: 4.30009e-07
[validate]: 3.213e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.00611192
[execute]: 6.78e-06
Sums
    bootstrap                            :     0.000364s :  2.54%
    type_inference                       :     0.004490s : 31.40%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000065s :  0.46%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.20%
    rewriter_before_opt_a                :     0.000075s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.38%
    jit_opt_a.loop_unroll                :     0.000033s :  0.23%
    jit_opt_a.a_1                        :     0.000612s :  4.28%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000449s :  3.14%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000039s :  0.27%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000511s :  3.57%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.16%
    loop_unroll                          :     0.000413s :  2.89%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000048s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000438s :  3.06%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006112s : 42.74%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000206    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.44% :     0.000005s :      3: substitution.graph_param_transform
84.38% :     0.000174s :      4: substitution.inline
 1.32% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.70% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.12% :     0.000004s :      2: substitution.replace_old_param
 5.57% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004424     2
82.82% :     0.003664s :      1: type_inference.infer
17.18% :     0.000760s :      1: type_inference.specialize
------[replace.]   0.000060     6
67.30% :     0.000040s :      4: replace.inline
32.70% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000182     6
94.27% :     0.000172s :      4: match.inline
 5.73% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.60% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.53% :     0.000002s :     10: predicate.addn_zero_filter
 1.94% :     0.000002s :     10: predicate.arithmetic_simplify
 1.32% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.19% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.33% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.22% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.34% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.86% :     0.000002s :     15: predicate.load_eliminater
 1.33% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.87% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.49% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.43% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.86% :     0.000002s :     18: predicate.replace_applicator
 0.83% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.07% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.26% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.36% :     0.000008s :     53: predicate.switch_simplify
 1.24% :     0.000001s :     10: predicate.tile_eliminate
 1.31% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.66% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000508    12
52.61% :     0.000267s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.26% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.13% :     0.000214s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019487    76
 0.26% :     0.000050s :      1: add_recomputation
 0.35% :     0.000069s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.97% :     0.000383s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.06% :     0.000012s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.34% :     0.004938s :      1: jit_opt_a
 0.70% :     0.000136s :      1: jit_opt_after_cconv
 0.26% :     0.000050s :      1: jit_opt_b
 2.16% :     0.000422s :      1: loop_unroll
 2.66% :     0.000519s :      1: mutable_eliminate
 4.19% :     0.000816s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.29% :     0.000446s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000031s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.13% :     0.000221s :      1: renormalize.infer
 1.14% :     0.000221s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000036s :      1: rewriter_after_opt_a
 0.40% :     0.000078s :      1: rewriter_before_opt_a
 0.36% :     0.000071s :      1: symbol_engine_optimizer
31.41% :     0.006122s :      1: task_emit
23.10% :     0.004501s :      1: type_inference
 0.26% :     0.000051s :      1: validate


TotalTime = 0.0174041, [33]
[bootstrap]: 0.00028371
[type_inference]: 0.0043066
[event_method]: 1.614e-05
[auto_monad]: 0.00015729
[graph_reusing]: 5.41998e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 3.431e-05
[rewriter_before_opt_a]: 6.808e-05
[expand_dump_flag]: 3.36999e-06
[jit_opt_a]: 0.00508834, [2]
    [Cycle 1]: 0.00159354, [27]
        [switch_simplify]: 5.334e-05
        [loop_unroll]: 2.863e-05
        [a_1]: 0.00053538
        [with_stream_mark]: 1.582e-05
        [recompute_prepare]: 7.4e-06
        [updatestate_depend_eliminate]: 3.17997e-06
        [updatestate_assign_eliminate]: 2.66999e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [parameter_eliminate]: 2.12001e-06
        [specialize_transform]: 6.48998e-06
        [updatestate_useless_node_eliminater]: 5.56e-06
        [accelerated_algorithm]: 5.85002e-06
        [meta_shard_fg_expand]: 2.19999e-06
        [get_grad_eliminate_]: 6.01e-06
        [merge_forward]: 2.99999e-06
        [cell_reuse_recompute_pass]: 1.50999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.297e-05
        [j_node_and_user_rematch]: 9.41998e-06
        [meta_fg_expand]: 1.86998e-06
        [replace_old_param]: 9.94999e-06
        [inline_without_move]: 5.68002e-06
        [renormalize]: 0.00065488
        [add_forward_monad_depend]: 5.53002e-06
        [auto_monad_grad]: 2.06998e-06
        [auto_monad_eliminator]: 1.258e-05
        [cse]: 1.648e-05
        [replace_applicator]: 1.213e-05
    [Cycle 2]: 0.00031736, [27]
        [switch_simplify]: 5.57999e-06
        [loop_unroll]: 5.53002e-06
        [a_1]: 9.139e-05
        [with_stream_mark]: 9.82999e-06
        [recompute_prepare]: 4.81002e-06
        [updatestate_depend_eliminate]: 2.34999e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.28002e-06
        [specialize_transform]: 4.87998e-06
        [updatestate_useless_node_eliminater]: 4.71002e-06
        [accelerated_algorithm]: 4.95001e-06
        [meta_shard_fg_expand]: 9.70002e-07
        [get_grad_eliminate_]: 4.33999e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.184e-05
        [j_node_and_user_rematch]: 7.82e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 7.26001e-06
        [inline_without_move]: 4.94998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.21002e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 4.62e-06
        [cse]: 9.47999e-06
        [replace_applicator]: 4.80999e-06
[py_interpret_to_execute_after_opt_a]: 8.77e-06
[rewriter_after_opt_a]: 3.092e-05
[convert_after_rewriter]: 6.63e-06
[order_py_execute_after_rewriter]: 4.97999e-06
[mutable_eliminate]: 0.0005644
[jit_opt_b]: 5.1e-05, [1]
    [Cycle 1]: 4.465e-05, [2]
        [frontend_op_eliminate]: 1.835e-05
        [inline_after_opt_a]: 1.508e-05
[cconv]: 2.26e-05
[loop_unroll]: 0.00040833
[jit_opt_after_cconv]: 0.00013315, [1]
    [Cycle 1]: 0.00012674, [11]
        [c_1]: 2.036e-05
        [parameter_eliminate]: 2.54001e-06
        [updatestate_depend_eliminate]: 5.17e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 1.673e-05
        [call_graph_tuple_transform]: 1.733e-05
        [tuple_list_get_item_eliminator]: 5.19e-06
        [none_parameter_eliminate]: 1.72001e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 4.94e-06
[remove_dup_value]: 1.219e-05
[partial_unused_args_eliminate]: 1.97999e-06
[environ_conv]: 4.58001e-06
[add_recomputation]: 4.17e-05
[cse_after_recomputation]: 1.989e-05, [1]
    [Cycle 1]: 1.436e-05, [1]
        [cse]: 8.92999e-06
[auto_monad_reorder]: 1.453e-05
[get_jit_bprop_graph]: 1.79998e-06
[rewriter_after_jit_bprop_graph]: 4.45e-06
[opt_after_jit_grad]: 0.00044468
[symbol_engine_optimizer]: 6.91e-05, [1]
    [Cycle 1]: 6.284e-05, [6]
        [build]: 3.38999e-06
        [elim_shapecalc]: 6.74999e-06
        [elim_not_effective]: 1.216e-05
        [opt_reshape]: 5.54e-06
        [fold_const_symbol]: 8.24998e-06
        [renormalize]: 4.00003e-07
[validate]: 3.028e-05
[backend_pass]: 8.49977e-07
[task_emit]: 0.0053678
[execute]: 6.16e-06
Sums
    bootstrap                            :     0.000284s :  2.08%
    type_inference                       :     0.004307s : 31.57%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000157s :  1.15%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.25%
    rewriter_before_opt_a                :     0.000068s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.43%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000627s :  4.59%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.13%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.000655s :  4.80%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000026s :  0.19%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000564s :  4.14%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000408s :  2.99%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000042s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000445s :  3.26%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000030s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005368s : 39.35%
    execute                              :     0.000006s :  0.05%


Time group info:
------[substitution.]   0.000199    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.47% :     0.000005s :      3: substitution.graph_param_transform
83.91% :     0.000167s :      4: substitution.inline
 1.58% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.02% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.15% :     0.000004s :      2: substitution.replace_old_param
 6.23% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004253     2
80.50% :     0.003423s :      1: type_inference.infer
19.50% :     0.000829s :      1: type_inference.specialize
------[replace.]   0.000070     6
64.96% :     0.000045s :      4: replace.inline
35.04% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000176     6
93.51% :     0.000165s :      4: match.inline
 6.49% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.29% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.21% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000001s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.39% :     0.000006s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.68% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.00% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.93% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.43% :     0.000002s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.64% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.45% :     0.000003s :     15: predicate.partial_eliminate
 1.23% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.83% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.69% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.94% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.32% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.25% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.00% :     0.000003s :     16: predicate.switch_defer_inline
 2.34% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.82% :     0.000009s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.26% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.55% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.84% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.07% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000588    12
48.37% :     0.000284s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.40% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.23% :     0.000272s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018994    76
 0.23% :     0.000045s :      1: add_recomputation
 0.86% :     0.000164s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.58% :     0.000299s :      1: bootstrap
 0.13% :     0.000025s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.80% :     0.005091s :      1: jit_opt_a
 0.71% :     0.000136s :      1: jit_opt_after_cconv
 0.28% :     0.000054s :      1: jit_opt_b
 2.19% :     0.000416s :      1: loop_unroll
 3.02% :     0.000573s :      1: mutable_eliminate
 4.40% :     0.000835s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.38% :     0.000452s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 1.79% :     0.000340s :      1: renormalize.infer
 1.62% :     0.000308s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.37% :     0.000071s :      1: rewriter_before_opt_a
 0.38% :     0.000071s :      1: symbol_engine_optimizer
28.32% :     0.005380s :      1: task_emit
22.73% :     0.004318s :      1: type_inference
 0.25% :     0.000048s :      1: validate


TotalTime = 0.0122682, [30]
[bootstrap]: 0.00027011
[type_inference]: 0.00416067
[event_method]: 1.684e-05
[auto_monad]: 6.461e-05
[graph_reusing]: 5.92001e-06
[pre_auto_parallel]: 2.45997e-06
[py_interpret_to_execute]: 2.938e-05
[rewriter_before_opt_a]: 7.844e-05
[expand_dump_flag]: 3.01001e-06
[jit_opt_a]: 0.00544819, [2]
    [Cycle 1]: 0.00167903, [27]
        [switch_simplify]: 5.588e-05
        [loop_unroll]: 3.298e-05
        [a_1]: 0.00054278
        [with_stream_mark]: 1.73e-05
        [recompute_prepare]: 7.50998e-06
        [updatestate_depend_eliminate]: 3.28e-06
        [updatestate_assign_eliminate]: 3.06999e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 1.92001e-06
        [specialize_transform]: 6.09999e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.19003e-06
        [meta_shard_fg_expand]: 2.21e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 3.49001e-06
        [cell_reuse_recompute_pass]: 1.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.413e-05
        [j_node_and_user_rematch]: 9.37001e-06
        [meta_fg_expand]: 2.29999e-06
        [replace_old_param]: 9.31e-06
        [inline_without_move]: 5.34998e-06
        [renormalize]: 0.0007052
        [add_forward_monad_depend]: 5.61e-06
        [auto_monad_grad]: 2.09999e-06
        [auto_monad_eliminator]: 1.439e-05
        [cse]: 3.042e-05
        [replace_applicator]: 1.279e-05
    [Cycle 2]: 0.0003459, [27]
        [switch_simplify]: 5.69e-06
        [loop_unroll]: 5.49e-06
        [a_1]: 9.701e-05
        [with_stream_mark]: 9.91e-06
        [recompute_prepare]: 5.62999e-06
        [updatestate_depend_eliminate]: 3.01999e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 5.34e-06
        [updatestate_useless_node_eliminater]: 5.54998e-06
        [accelerated_algorithm]: 6.00002e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 5.12e-06
        [merge_forward]: 2.99999e-06
        [cell_reuse_recompute_pass]: 2.26e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.44e-05
        [j_node_and_user_rematch]: 7.95e-06
        [meta_fg_expand]: 1.89e-06
        [replace_old_param]: 8e-06
        [inline_without_move]: 5.10001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.38002e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 5.67999e-06
        [cse]: 1.207e-05
        [replace_applicator]: 5.43002e-06
[py_interpret_to_execute_after_opt_a]: 9.85002e-06
[rewriter_after_opt_a]: 3.715e-05
[convert_after_rewriter]: 7.1e-06
[order_py_execute_after_rewriter]: 4.93001e-06
[mutable_eliminate]: 0.00061471
[jit_opt_b]: 5.137e-05, [1]
    [Cycle 1]: 4.445e-05, [2]
        [frontend_op_eliminate]: 1.701e-05
        [inline_after_opt_a]: 1.581e-05
[cconv]: 2.626e-05
[loop_unroll]: 0.00041039
[jit_opt_after_cconv]: 0.00018738, [1]
    [Cycle 1]: 0.00018098, [11]
        [c_1]: 2.105e-05
        [parameter_eliminate]: 2.91999e-06
        [updatestate_depend_eliminate]: 6.61e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 1.98002e-06
        [cse]: 1.785e-05
        [call_graph_tuple_transform]: 1.899e-05
        [tuple_list_get_item_eliminator]: 5.92999e-06
        [none_parameter_eliminate]: 1.43002e-06
        [renormalize]: 6.39993e-07
        [switch_simplify]: 5.37001e-06
[remove_dup_value]: 1.43e-05
[partial_unused_args_eliminate]: 2.36998e-06
[environ_conv]: 4.91002e-06
[add_recomputation]: 5.051e-05
[cse_after_recomputation]: 2.057e-05, [1]
    [Cycle 1]: 1.505e-05, [1]
        [cse]: 9.34e-06
[auto_monad_reorder]: 1.603e-05
[get_jit_bprop_graph]: 1.86998e-06
[rewriter_after_jit_bprop_graph]: 5.36998e-06
[opt_after_jit_grad]: 0.0004516
[symbol_engine_optimizer]: 7.554e-05, [1]
    [Cycle 1]: 6.91e-05, [6]
        [build]: 4.86997e-06
        [elim_shapecalc]: 8.02e-06
        [elim_not_effective]: 1.214e-05
        [opt_reshape]: 5.62001e-06
        [fold_const_symbol]: 9.00999e-06
        [renormalize]: 4.09986e-07
[validate]: 3.706e-05
Sums
    bootstrap                            :     0.000270s :  3.28%
    type_inference                       :     0.004161s : 50.59%
    event_method                         :     0.000017s :  0.20%
    auto_monad                           :     0.000065s :  0.79%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000029s :  0.36%
    rewriter_before_opt_a                :     0.000078s :  0.95%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000062s :  0.75%
    jit_opt_a.loop_unroll                :     0.000038s :  0.47%
    jit_opt_a.a_1                        :     0.000640s :  7.78%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000705s :  8.58%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.24%
    jit_opt_a.cse                        :     0.000042s :  0.52%
    jit_opt_a.replace_applicator         :     0.000018s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000037s :  0.45%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000615s :  7.47%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.19%
    cconv                                :     0.000026s :  0.32%
    loop_unroll                          :     0.000410s :  4.99%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000051s :  0.61%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000452s :  5.49%
    symbol_engine_optimizer.build        :     0.000005s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.45%


Time group info:
------[substitution.]   0.000210    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.74% :     0.000006s :      3: substitution.graph_param_transform
83.34% :     0.000175s :      4: substitution.inline
 1.72% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.30% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.25% :     0.000005s :      2: substitution.replace_old_param
 6.10% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004106     2
78.00% :     0.003202s :      1: type_inference.infer
22.00% :     0.000903s :      1: type_inference.specialize
------[replace.]   0.000069     6
67.35% :     0.000046s :      4: replace.inline
32.65% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000184     6
93.69% :     0.000172s :      4: match.inline
 6.31% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.32% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.12% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.16% :     0.000003s :     10: predicate.arithmetic_simplify
 1.19% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.53% :     0.000007s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000001s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.45% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.16% :     0.000001s :     10: predicate.minmaximum_grad
 1.75% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.44% :     0.000003s :     15: predicate.partial_eliminate
 1.31% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.03% :     0.000002s :     10: predicate.reduce_eliminate
 1.36% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.67% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.65% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.11% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.46% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.29% :     0.000002s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.41% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.50% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.81% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000569    12
44.75% :     0.000255s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.07% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.18% :     0.000274s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013889    72
 0.38% :     0.000053s :      1: add_recomputation
 0.49% :     0.000068s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 2.02% :     0.000281s :      1: bootstrap
 0.21% :     0.000029s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.15% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
39.25% :     0.005452s :      1: jit_opt_a
 1.37% :     0.000190s :      1: jit_opt_after_cconv
 0.39% :     0.000054s :      1: jit_opt_b
 3.01% :     0.000418s :      1: loop_unroll
 4.49% :     0.000624s :      1: mutable_eliminate
 6.19% :     0.000860s :     26: opt.transform.jit_opt_a
 0.34% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.31% :     0.000459s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.23% :     0.000032s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.71% :     0.000376s :      1: renormalize.infer
 2.31% :     0.000321s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.30% :     0.000041s :      1: rewriter_after_opt_a
 0.59% :     0.000082s :      1: rewriter_before_opt_a
 0.56% :     0.000078s :      1: symbol_engine_optimizer
30.04% :     0.004172s :      1: type_inference


TotalTime = 0.0176543, [33]
[bootstrap]: 0.00030022
[type_inference]: 0.00440908
[event_method]: 1.577e-05
[auto_monad]: 5.04e-05
[graph_reusing]: 5.05999e-06
[pre_auto_parallel]: 1.95001e-06
[py_interpret_to_execute]: 2.882e-05
[rewriter_before_opt_a]: 6.203e-05
[expand_dump_flag]: 2.73e-06
[jit_opt_a]: 0.00486356, [2]
    [Cycle 1]: 0.00126816, [27]
        [switch_simplify]: 4.143e-05
        [loop_unroll]: 2.78e-05
        [a_1]: 0.00044837
        [with_stream_mark]: 1.119e-05
        [recompute_prepare]: 6.71999e-06
        [updatestate_depend_eliminate]: 2.99999e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [parameter_eliminate]: 1.75001e-06
        [specialize_transform]: 5.70001e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 5.25001e-06
        [meta_shard_fg_expand]: 1.53002e-06
        [get_grad_eliminate_]: 5.00999e-06
        [merge_forward]: 2.41998e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.092e-05
        [j_node_and_user_rematch]: 8.33001e-06
        [meta_fg_expand]: 2.19001e-06
        [replace_old_param]: 7.30998e-06
        [inline_without_move]: 5.09998e-06
        [renormalize]: 0.00047478
        [add_forward_monad_depend]: 4.12e-06
        [auto_monad_grad]: 1.05001e-06
        [auto_monad_eliminator]: 8.07e-06
        [cse]: 1.386e-05
        [replace_applicator]: 9.44e-06
    [Cycle 2]: 0.0003031, [27]
        [switch_simplify]: 5.44e-06
        [loop_unroll]: 4.53999e-06
        [a_1]: 8.64e-05
        [with_stream_mark]: 7.55998e-06
        [recompute_prepare]: 4.60001e-06
        [updatestate_depend_eliminate]: 2.36e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 9.00007e-07
        [specialize_transform]: 4.70999e-06
        [updatestate_useless_node_eliminater]: 4.57998e-06
        [accelerated_algorithm]: 4.72e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 4.32e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.094e-05
        [j_node_and_user_rematch]: 7.35e-06
        [meta_fg_expand]: 1.39e-06
        [replace_old_param]: 6.49999e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 4.75001e-06
        [cse]: 9.42999e-06
        [replace_applicator]: 4.56002e-06
[py_interpret_to_execute_after_opt_a]: 9.91e-06
[rewriter_after_opt_a]: 3.087e-05
[convert_after_rewriter]: 7.9e-06
[order_py_execute_after_rewriter]: 4.83001e-06
[mutable_eliminate]: 0.00047703
[jit_opt_b]: 4.53e-05, [1]
    [Cycle 1]: 3.893e-05, [2]
        [frontend_op_eliminate]: 1.428e-05
        [inline_after_opt_a]: 1.393e-05
[cconv]: 1.952e-05
[loop_unroll]: 0.00040542
[jit_opt_after_cconv]: 0.00012867, [1]
    [Cycle 1]: 0.00012265, [11]
        [c_1]: 1.905e-05
        [parameter_eliminate]: 2.02001e-06
        [updatestate_depend_eliminate]: 4.48999e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [cse]: 1.589e-05
        [call_graph_tuple_transform]: 1.642e-05
        [tuple_list_get_item_eliminator]: 5.04998e-06
        [none_parameter_eliminate]: 1.38002e-06
        [renormalize]: 3.80009e-07
        [switch_simplify]: 5.12999e-06
[remove_dup_value]: 1.323e-05
[partial_unused_args_eliminate]: 2.11998e-06
[environ_conv]: 5.17e-06
[add_recomputation]: 4.669e-05
[cse_after_recomputation]: 1.877e-05, [1]
    [Cycle 1]: 1.352e-05, [1]
        [cse]: 8.21002e-06
[auto_monad_reorder]: 1.435e-05
[get_jit_bprop_graph]: 1.34e-06
[rewriter_after_jit_bprop_graph]: 4.70999e-06
[opt_after_jit_grad]: 0.00043599
[symbol_engine_optimizer]: 6.678e-05, [1]
    [Cycle 1]: 6.091e-05, [6]
        [build]: 2.98e-06
        [elim_shapecalc]: 6.64001e-06
        [elim_not_effective]: 1.094e-05
        [opt_reshape]: 5.56998e-06
        [fold_const_symbol]: 8.56002e-06
        [renormalize]: 4.19997e-07
[validate]: 2.874e-05
[backend_pass]: 1.12e-06
[task_emit]: 0.0059809
[execute]: 7.31999e-06
Sums
    bootstrap                            :     0.000300s :  2.17%
    type_inference                       :     0.004409s : 31.89%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000050s :  0.36%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000029s :  0.21%
    rewriter_before_opt_a                :     0.000062s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000047s :  0.34%
    jit_opt_a.loop_unroll                :     0.000032s :  0.23%
    jit_opt_a.a_1                        :     0.000535s :  3.87%
    jit_opt_a.with_stream_mark           :     0.000019s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000475s :  3.43%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.09%
    jit_opt_a.cse                        :     0.000023s :  0.17%
    jit_opt_a.replace_applicator         :     0.000014s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000031s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000477s :  3.45%
    jit_opt_b.frontend_op_eliminate      :     0.000014s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000020s :  0.14%
    loop_unroll                          :     0.000405s :  2.93%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.34%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000436s :  3.15%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005981s : 43.26%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000142    23
 1.23% :     0.000002s :      2: substitution.elim_not_effective
 0.90% :     0.000001s :      2: substitution.fold_const_symbol
 3.40% :     0.000005s :      3: substitution.graph_param_transform
82.56% :     0.000117s :      4: substitution.inline
 1.76% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.67% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.72% :     0.000002s :      2: substitution.replace_old_param
 5.76% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004358     2
83.21% :     0.003626s :      1: type_inference.infer
16.79% :     0.000732s :      1: type_inference.specialize
------[replace.]   0.000057     6
65.40% :     0.000037s :      4: replace.inline
34.60% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000122     6
94.20% :     0.000115s :      4: match.inline
 5.80% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000105   691
 1.57% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.98% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.16% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000001s :     10: predicate.addn_zero_filter
 1.90% :     0.000002s :     10: predicate.arithmetic_simplify
 1.31% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000001s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.48% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 4.98% :     0.000005s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000001s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.88% :     0.000002s :     15: predicate.load_eliminater
 1.28% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.09% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.28% :     0.000001s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000002s :     15: predicate.partial_eliminate
 1.49% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.64% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 1.15% :     0.000001s :      6: predicate.specialize_transform
 1.64% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.38% :     0.000008s :     53: predicate.switch_simplify
 1.34% :     0.000001s :     10: predicate.tile_eliminate
 1.53% :     0.000002s :     10: predicate.transpose_eliminate
 1.41% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.36% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.38% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.80% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000492    12
50.93% :     0.000251s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.29% :     0.000021s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.78% :     0.000221s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018941    76
 0.26% :     0.000049s :      1: add_recomputation
 0.28% :     0.000053s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.64% :     0.000311s :      1: bootstrap
 0.12% :     0.000022s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000003s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
25.69% :     0.004866s :      1: jit_opt_a
 0.69% :     0.000131s :      1: jit_opt_after_cconv
 0.25% :     0.000048s :      1: jit_opt_b
 2.18% :     0.000412s :      1: loop_unroll
 2.55% :     0.000483s :      1: mutable_eliminate
 3.78% :     0.000717s :     26: opt.transform.jit_opt_a
 0.23% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000022s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000010s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000017s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.33% :     0.000442s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000031s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.39% :     0.000263s :      1: renormalize.infer
 1.09% :     0.000207s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.34% :     0.000065s :      1: rewriter_before_opt_a
 0.36% :     0.000069s :      1: symbol_engine_optimizer
31.63% :     0.005991s :      1: task_emit
23.32% :     0.004417s :      1: type_inference
 0.24% :     0.000046s :      1: validate


TotalTime = 0.0317984, [33]
[bootstrap]: 0.00020804
[type_inference]: 0.0122564
[event_method]: 6.086e-05
[auto_monad]: 0.00011642
[graph_reusing]: 8.57998e-06
[pre_auto_parallel]: 2.88998e-06
[py_interpret_to_execute]: 2.973e-05
[rewriter_before_opt_a]: 0.00010828
[expand_dump_flag]: 3.68e-06
[jit_opt_a]: 0.0114241, [3]
    [Cycle 1]: 0.00574079, [27]
        [switch_simplify]: 0.00013409
        [loop_unroll]: 4.659e-05
        [a_1]: 0.00089897
        [with_stream_mark]: 2.538e-05
        [recompute_prepare]: 2e-05
        [updatestate_depend_eliminate]: 8.00999e-06
        [updatestate_assign_eliminate]: 6.62002e-06
        [updatestate_loads_eliminate]: 7.19001e-06
        [parameter_eliminate]: 2.53e-06
        [specialize_transform]: 1.305e-05
        [updatestate_useless_node_eliminater]: 1.259e-05
        [accelerated_algorithm]: 1.263e-05
        [meta_shard_fg_expand]: 4.08001e-06
        [get_grad_eliminate_]: 1.226e-05
        [merge_forward]: 8.35001e-06
        [cell_reuse_recompute_pass]: 1.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.716e-05
        [j_node_and_user_rematch]: 2.324e-05
        [meta_fg_expand]: 0.00160283
        [replace_old_param]: 5.675e-05
        [inline_without_move]: 5.316e-05
        [renormalize]: 0.00225547
        [add_forward_monad_depend]: 8.47998e-06
        [auto_monad_grad]: 5.15001e-06
        [auto_monad_eliminator]: 4.568e-05
        [cse]: 0.00016846
        [replace_applicator]: 6.307e-05
    [Cycle 2]: 0.00191848, [27]
        [switch_simplify]: 3.621e-05
        [loop_unroll]: 3.412e-05
        [a_1]: 0.00103461
        [with_stream_mark]: 1.116e-05
        [recompute_prepare]: 8.42998e-06
        [updatestate_depend_eliminate]: 3.70998e-06
        [updatestate_assign_eliminate]: 3.03998e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 6.44999e-06
        [updatestate_useless_node_eliminater]: 6.33e-06
        [accelerated_algorithm]: 6.34999e-06
        [meta_shard_fg_expand]: 1.58002e-06
        [get_grad_eliminate_]: 6.28e-06
        [merge_forward]: 3.14999e-06
        [cell_reuse_recompute_pass]: 7.59988e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.239e-05
        [j_node_and_user_rematch]: 1.016e-05
        [meta_fg_expand]: 7.541e-05
        [replace_old_param]: 1.016e-05
        [inline_without_move]: 6.19001e-06
        [renormalize]: 0.00045175
        [add_forward_monad_depend]: 3.82998e-06
        [auto_monad_grad]: 1.07998e-06
        [auto_monad_eliminator]: 9.96e-06
        [cse]: 1.684e-05
        [replace_applicator]: 1.048e-05
    [Cycle 3]: 0.00036365, [27]
        [switch_simplify]: 6.66e-06
        [loop_unroll]: 6.11998e-06
        [a_1]: 0.0001195
        [with_stream_mark]: 8.43999e-06
        [recompute_prepare]: 5.96998e-06
        [updatestate_depend_eliminate]: 3.51999e-06
        [updatestate_assign_eliminate]: 2.70997e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 8.2e-07
        [specialize_transform]: 6.00002e-06
        [updatestate_useless_node_eliminater]: 5.88998e-06
        [accelerated_algorithm]: 5.84e-06
        [meta_shard_fg_expand]: 1.29003e-06
        [get_grad_eliminate_]: 5.58002e-06
        [merge_forward]: 3.01999e-06
        [cell_reuse_recompute_pass]: 1.36998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.428e-05
        [j_node_and_user_rematch]: 9.67999e-06
        [meta_fg_expand]: 1.99e-06
        [replace_old_param]: 8.17998e-06
        [inline_without_move]: 5.89e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.06997e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 6.59001e-06
        [cse]: 1.386e-05
        [replace_applicator]: 6.09001e-06
[py_interpret_to_execute_after_opt_a]: 9.30001e-06
[rewriter_after_opt_a]: 2.96e-05
[convert_after_rewriter]: 6.48003e-06
[order_py_execute_after_rewriter]: 4.63001e-06
[mutable_eliminate]: 0.00046778
[jit_opt_b]: 5.367e-05, [1]
    [Cycle 1]: 4.759e-05, [2]
        [frontend_op_eliminate]: 1.921e-05
        [inline_after_opt_a]: 1.747e-05
[cconv]: 1.505e-05
[loop_unroll]: 0.00040973
[jit_opt_after_cconv]: 0.00014902, [1]
    [Cycle 1]: 0.00014261, [11]
        [c_1]: 2.651e-05
        [parameter_eliminate]: 2.21e-06
        [updatestate_depend_eliminate]: 5.13002e-06
        [updatestate_assign_eliminate]: 2.84001e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [cse]: 2.043e-05
        [call_graph_tuple_transform]: 1.886e-05
        [tuple_list_get_item_eliminator]: 6.60002e-06
        [none_parameter_eliminate]: 8.80013e-07
        [renormalize]: 4.50003e-07
        [switch_simplify]: 6.15002e-06
[remove_dup_value]: 9.87999e-06
[partial_unused_args_eliminate]: 1.39e-06
[environ_conv]: 3.85998e-06
[add_recomputation]: 3.745e-05
[cse_after_recomputation]: 2.427e-05, [1]
    [Cycle 1]: 1.927e-05, [1]
        [cse]: 1.354e-05
[auto_monad_reorder]: 1.285e-05
[get_jit_bprop_graph]: 1.46998e-06
[rewriter_after_jit_bprop_graph]: 4.85999e-06
[opt_after_jit_grad]: 0.00046978
[symbol_engine_optimizer]: 7.712e-05, [1]
    [Cycle 1]: 7.152e-05, [6]
        [build]: 2.88e-06
        [elim_shapecalc]: 9.17001e-06
        [elim_not_effective]: 1.451e-05
        [opt_reshape]: 6.80998e-06
        [fold_const_symbol]: 1.041e-05
        [renormalize]: 4.00003e-07
[validate]: 2.817e-05
[backend_pass]: 6.89994e-07
[task_emit]: 0.00556604
[execute]: 4.44002e-06
Sums
    bootstrap                            :     0.000208s :  0.75%
    type_inference                       :     0.012256s : 44.38%
    event_method                         :     0.000061s :  0.22%
    auto_monad                           :     0.000116s :  0.42%
    graph_reusing                        :     0.000009s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000030s :  0.11%
    rewriter_before_opt_a                :     0.000108s :  0.39%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000177s :  0.64%
    jit_opt_a.loop_unroll                :     0.000087s :  0.31%
    jit_opt_a.a_1                        :     0.002053s :  7.43%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000034s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000025s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.09%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000054s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000043s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001680s :  6.08%
    jit_opt_a.replace_old_param          :     0.000075s :  0.27%
    jit_opt_a.inline_without_move        :     0.000065s :  0.24%
    jit_opt_a.renormalize                :     0.002707s :  9.80%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000062s :  0.23%
    jit_opt_a.cse                        :     0.000199s :  0.72%
    jit_opt_a.replace_applicator         :     0.000080s :  0.29%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.03%
    rewriter_after_opt_a                 :     0.000030s :  0.11%
    convert_after_rewriter               :     0.000006s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000468s :  1.69%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.06%
    cconv                                :     0.000015s :  0.05%
    loop_unroll                          :     0.000410s :  1.48%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000010s :  0.04%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000004s :  0.01%
    add_recomputation                    :     0.000037s :  0.14%
    cse_after_recomputation.cse          :     0.000014s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000470s :  1.70%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.10%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005566s : 20.15%
    execute                              :     0.000004s :  0.02%


Time group info:
------[substitution.]   0.000484   113
 0.34% :     0.000002s :      3: substitution.elim_not_effective
 0.27% :     0.000001s :      3: substitution.fold_const_symbol
 0.84% :     0.000004s :      4: substitution.graph_param_transform
63.20% :     0.000306s :     14: substitution.inline
 3.57% :     0.000017s :      2: substitution.inline_without_move
 1.75% :     0.000008s :     14: substitution.j_node_and_user_rematch
 2.16% :     0.000010s :      7: substitution.minmaximum_grad
 1.34% :     0.000006s :      5: substitution.partial_eliminate
 2.25% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.21% :     0.000025s :     10: substitution.replace_applicator
 1.75% :     0.000008s :      7: substitution.replace_old_param
 0.49% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.75% :     0.000008s :      1: substitution.switch_simplify
 4.33% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.79% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 7.97% :     0.000039s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012171     2
89.11% :     0.010846s :      1: type_inference.infer
10.89% :     0.001325s :      1: type_inference.specialize
------[replace.]   0.000182    22
53.46% :     0.000097s :     14: replace.inline
18.95% :     0.000034s :      1: replace.switch_simplify
27.60% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000326    22
91.31% :     0.000297s :     14: match.inline
 2.36% :     0.000008s :      1: match.switch_simplify
 6.33% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000387  2634
 1.47% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.39% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.35% :     0.000005s :     44: predicate.addn_check_dump
 1.69% :     0.000007s :     44: predicate.addn_zero_filter
 2.23% :     0.000009s :     44: predicate.arithmetic_simplify
 1.55% :     0.000006s :     44: predicate.cast_eliminate
 0.20% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.38% :     0.000005s :     44: predicate.compare_switch_simplify
 1.59% :     0.000006s :     44: predicate.depend_value_elim
 1.42% :     0.000005s :     44: predicate.dict_get_item_const_eliminator
 1.62% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.73% :     0.000007s :     44: predicate.dict_set_item_eliminator
 0.28% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      4: predicate.elim_not_effective
 0.19% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.41% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.38% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.49% :     0.000006s :     44: predicate.environ_get_eliminate
 1.62% :     0.000006s :     44: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.81% :     0.000003s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.48% :     0.000017s :     73: predicate.inline
 1.78% :     0.000007s :     40: predicate.inline_without_move
 0.35% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.95% :     0.000004s :     19: predicate.less_batch_normalization
 1.78% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.84% :     0.000007s :     55: predicate.load_eliminater
 0.43% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.44% :     0.000013s :     91: predicate.loop_unroll_before_grad
 1.80% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.41% :     0.000005s :     44: predicate.merge_addn
 1.47% :     0.000006s :     44: predicate.minmaximum_grad
 0.42% :     0.000002s :      4: predicate.mutable_eliminate
 0.17% :     0.000001s :      4: predicate.opt_reshape
 2.31% :     0.000009s :     55: predicate.partial_eliminate
 1.43% :     0.000006s :     44: predicate.print_const_string_wrapper
 2.08% :     0.000008s :     44: predicate.reduce_eliminate
 1.75% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.38% :     0.000009s :     97: predicate.replace_applicator
 0.99% :     0.000004s :     40: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.79% :     0.000007s :     44: predicate.reshape_eliminate
 1.54% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.27% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.55% :     0.000006s :     44: predicate.same_eliminate
 0.45% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.43% :     0.000002s :      8: predicate.special_op_eliminate
 0.80% :     0.000003s :     19: predicate.specialize_transform
 1.82% :     0.000007s :     44: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000005s :     44: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.81% :     0.000011s :     65: predicate.switch_defer_inline
 2.57% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.60% :     0.000026s :    162: predicate.switch_simplify
 1.45% :     0.000006s :     44: predicate.tile_eliminate
 1.41% :     0.000005s :     44: predicate.transpose_eliminate
 1.93% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.71% :     0.000007s :     44: predicate.tuple_list_get_item_depend_reorder
 3.37% :     0.000013s :     59: predicate.tuple_list_get_item_eliminator
 1.90% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.75% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.83% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000007s :     44: predicate.value_based_eliminate
 0.14% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.29% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001559    32
63.04% :     0.000982s :     14: func_graph_cloner_run.FuncGraphClonerGraph
36.96% :     0.000576s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.037346    91
 0.11% :     0.000040s :      1: add_recomputation
 0.32% :     0.000121s :      1: auto_monad
 0.04% :     0.000015s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.60% :     0.000223s :      1: bootstrap
 0.05% :     0.000018s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.07% :     0.000026s :      1: cse_after_recomputation
 0.02% :     0.000006s :      1: environ_conv
 0.18% :     0.000066s :      1: event_method
 0.02% :     0.000008s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000011s :      1: graph_reusing
30.60% :     0.011427s :      1: jit_opt_a
 0.41% :     0.000152s :      1: jit_opt_after_cconv
 0.15% :     0.000056s :      1: jit_opt_b
 1.12% :     0.000417s :      1: loop_unroll
 1.27% :     0.000474s :      1: mutable_eliminate
 7.29% :     0.002723s :     39: opt.transform.jit_opt_a
 0.15% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000031s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.28% :     0.000477s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000003s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000032s :      1: py_interpret_to_execute
 0.03% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000012s :      1: remove_dup_value
 3.96% :     0.001478s :      2: renormalize.infer
 3.26% :     0.001216s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000032s :      1: rewriter_after_opt_a
 0.30% :     0.000111s :      1: rewriter_before_opt_a
 0.21% :     0.000080s :      1: symbol_engine_optimizer
14.93% :     0.005575s :      1: task_emit
32.86% :     0.012274s :      1: type_inference
 0.12% :     0.000045s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.900.813 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc418b0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.901.355 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0197433, [33]
[bootstrap]: 0.00027621
[type_inference]: 0.00590681
[event_method]: 1.789e-05
[auto_monad]: 6.318e-05
[graph_reusing]: 5.20999e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 3.839e-05
[rewriter_before_opt_a]: 6.98e-05
[expand_dump_flag]: 2.91e-06
[jit_opt_a]: 0.00523883, [2]
    [Cycle 1]: 0.00157443, [27]
        [switch_simplify]: 5.115e-05
        [loop_unroll]: 2.85e-05
        [a_1]: 0.00052637
        [with_stream_mark]: 1.61e-05
        [recompute_prepare]: 7.36999e-06
        [updatestate_depend_eliminate]: 3.63e-06
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 1.77001e-06
        [specialize_transform]: 6.13002e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 5.30001e-06
        [meta_shard_fg_expand]: 2.55997e-06
        [get_grad_eliminate_]: 5.27001e-06
        [merge_forward]: 3.85e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.457e-05
        [j_node_and_user_rematch]: 8.57e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 9.13002e-06
        [inline_without_move]: 5.47001e-06
        [renormalize]: 0.00064722
        [add_forward_monad_depend]: 5.72001e-06
        [auto_monad_grad]: 2.14999e-06
        [auto_monad_eliminator]: 1.325e-05
        [cse]: 2.299e-05
        [replace_applicator]: 1.14e-05
    [Cycle 2]: 0.00030891, [27]
        [switch_simplify]: 5.34998e-06
        [loop_unroll]: 4.63001e-06
        [a_1]: 8.807e-05
        [with_stream_mark]: 8.37e-06
        [recompute_prepare]: 4.89e-06
        [updatestate_depend_eliminate]: 2.49001e-06
        [updatestate_assign_eliminate]: 1.92999e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 4.72998e-06
        [updatestate_useless_node_eliminater]: 4.77998e-06
        [accelerated_algorithm]: 4.94e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.33001e-06
        [merge_forward]: 2.10002e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.174e-05
        [j_node_and_user_rematch]: 7.87e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 6.83e-06
        [inline_without_move]: 4.52998e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 4.79002e-06
        [cse]: 9.19998e-06
        [replace_applicator]: 4.82e-06
[py_interpret_to_execute_after_opt_a]: 8.90999e-06
[rewriter_after_opt_a]: 3.06e-05
[convert_after_rewriter]: 6.54001e-06
[order_py_execute_after_rewriter]: 5.16002e-06
[mutable_eliminate]: 0.00047955
[jit_opt_b]: 4.709e-05, [1]
    [Cycle 1]: 4.076e-05, [2]
        [frontend_op_eliminate]: 1.538e-05
        [inline_after_opt_a]: 1.376e-05
[cconv]: 2.316e-05
[loop_unroll]: 0.00040422
[jit_opt_after_cconv]: 0.00020332, [1]
    [Cycle 1]: 0.00019704, [11]
        [c_1]: 2.021e-05
        [parameter_eliminate]: 2.44999e-06
        [updatestate_depend_eliminate]: 4.87e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [cse]: 1.61e-05
        [call_graph_tuple_transform]: 1.873e-05
        [tuple_list_get_item_eliminator]: 5.20999e-06
        [none_parameter_eliminate]: 1.87001e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.37001e-06
[remove_dup_value]: 1.28e-05
[partial_unused_args_eliminate]: 2.09999e-06
[environ_conv]: 4.95999e-06
[add_recomputation]: 4.524e-05
[cse_after_recomputation]: 1.941e-05, [1]
    [Cycle 1]: 1.393e-05, [1]
        [cse]: 8.46002e-06
[auto_monad_reorder]: 1.44e-05
[get_jit_bprop_graph]: 1.55999e-06
[rewriter_after_jit_bprop_graph]: 4.53001e-06
[opt_after_jit_grad]: 0.00044201
[symbol_engine_optimizer]: 6.949e-05, [1]
    [Cycle 1]: 6.292e-05, [6]
        [build]: 3.16001e-06
        [elim_shapecalc]: 7.19001e-06
        [elim_not_effective]: 1.18e-05
        [opt_reshape]: 5.46e-06
        [fold_const_symbol]: 8.32998e-06
        [renormalize]: 4.19997e-07
[validate]: 3.174e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.00606898
[execute]: 7.61001e-06
Sums
    bootstrap                            :     0.000276s :  1.75%
    type_inference                       :     0.005907s : 37.53%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000063s :  0.40%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.24%
    rewriter_before_opt_a                :     0.000070s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.36%
    jit_opt_a.loop_unroll                :     0.000033s :  0.21%
    jit_opt_a.a_1                        :     0.000614s :  3.90%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000647s :  4.11%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.11%
    jit_opt_a.cse                        :     0.000032s :  0.20%
    jit_opt_a.replace_applicator         :     0.000016s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.19%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000480s :  3.05%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000023s :  0.15%
    loop_unroll                          :     0.000404s :  2.57%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000045s :  0.29%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000442s :  2.81%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006069s : 38.56%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000187    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000001s :      2: substitution.fold_const_symbol
 2.63% :     0.000005s :      3: substitution.graph_param_transform
83.38% :     0.000156s :      4: substitution.inline
 1.39% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.38% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.19% :     0.000004s :      2: substitution.replace_old_param
 6.28% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005839     2
83.09% :     0.004851s :      1: type_inference.infer
16.91% :     0.000987s :      1: type_inference.specialize
------[replace.]   0.000064     6
67.16% :     0.000043s :      4: replace.inline
32.84% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000164     6
93.48% :     0.000154s :      4: match.inline
 6.52% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.25% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.74% :     0.000002s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.43% :     0.000002s :     10: predicate.cast_eliminate
 0.68% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.32% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.67% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.61% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.69% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.91% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000001s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.26% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.71% :     0.000009s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.20% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000628    12
47.22% :     0.000297s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.71% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.06% :     0.000302s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021299    76
 0.23% :     0.000048s :      1: add_recomputation
 0.31% :     0.000066s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.35% :     0.000288s :      1: bootstrap
 0.12% :     0.000026s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
24.61% :     0.005242s :      1: jit_opt_a
 0.97% :     0.000206s :      1: jit_opt_after_cconv
 0.23% :     0.000050s :      1: jit_opt_b
 1.93% :     0.000412s :      1: loop_unroll
 2.29% :     0.000488s :      1: mutable_eliminate
 3.82% :     0.000814s :     26: opt.transform.jit_opt_a
 0.22% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000023s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.11% :     0.000450s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000041s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 1.60% :     0.000341s :      1: renormalize.infer
 1.40% :     0.000299s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000034s :      1: rewriter_after_opt_a
 0.34% :     0.000073s :      1: rewriter_before_opt_a
 0.34% :     0.000072s :      1: symbol_engine_optimizer
28.54% :     0.006079s :      1: task_emit
27.81% :     0.005923s :      1: type_inference
 0.23% :     0.000048s :      1: validate


TotalTime = 0.0113327, [30]
[bootstrap]: 0.0002742
[type_inference]: 0.00402067
[event_method]: 1.616e-05
[auto_monad]: 5.844e-05
[graph_reusing]: 4.86002e-06
[pre_auto_parallel]: 2.17999e-06
[py_interpret_to_execute]: 2.944e-05
[rewriter_before_opt_a]: 0.00010869
[expand_dump_flag]: 3.17002e-06
[jit_opt_a]: 0.00489944, [2]
    [Cycle 1]: 0.00137571, [27]
        [switch_simplify]: 4.503e-05
        [loop_unroll]: 3.007e-05
        [a_1]: 0.00049108
        [with_stream_mark]: 1.413e-05
        [recompute_prepare]: 7.08998e-06
        [updatestate_depend_eliminate]: 2.84999e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [parameter_eliminate]: 1.49e-06
        [specialize_transform]: 6.02001e-06
        [updatestate_useless_node_eliminater]: 5.40999e-06
        [accelerated_algorithm]: 5.73002e-06
        [meta_shard_fg_expand]: 1.62001e-06
        [get_grad_eliminate_]: 5.14e-06
        [merge_forward]: 2.70002e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.216e-05
        [j_node_and_user_rematch]: 9.10001e-06
        [meta_fg_expand]: 2.29001e-06
        [replace_old_param]: 7.66999e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 0.00051254
        [add_forward_monad_depend]: 4.73001e-06
        [auto_monad_grad]: 1.69998e-06
        [auto_monad_eliminator]: 1.161e-05
        [cse]: 1.613e-05
        [replace_applicator]: 1.086e-05
    [Cycle 2]: 0.00034263, [27]
        [switch_simplify]: 5.57999e-06
        [loop_unroll]: 4.78001e-06
        [a_1]: 8.942e-05
        [with_stream_mark]: 8.71002e-06
        [recompute_prepare]: 4.90001e-06
        [updatestate_depend_eliminate]: 2.53e-06
        [updatestate_assign_eliminate]: 2.13002e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 5.07e-06
        [updatestate_useless_node_eliminater]: 4.99998e-06
        [accelerated_algorithm]: 4.99e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.47001e-06
        [cell_reuse_recompute_pass]: 1.24998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.199e-05
        [j_node_and_user_rematch]: 7.91001e-06
        [meta_fg_expand]: 1.72001e-06
        [replace_old_param]: 7.66999e-06
        [inline_without_move]: 5.20001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 5.11002e-06
        [cse]: 1.038e-05
        [replace_applicator]: 6.81999e-06
[py_interpret_to_execute_after_opt_a]: 9.61998e-06
[rewriter_after_opt_a]: 3.049e-05
[convert_after_rewriter]: 5.88002e-06
[order_py_execute_after_rewriter]: 4.90999e-06
[mutable_eliminate]: 0.00047612
[jit_opt_b]: 5.07e-05, [1]
    [Cycle 1]: 4.382e-05, [2]
        [frontend_op_eliminate]: 1.698e-05
        [inline_after_opt_a]: 1.485e-05
[cconv]: 2.204e-05
[loop_unroll]: 0.00039826
[jit_opt_after_cconv]: 0.00013306, [1]
    [Cycle 1]: 0.00012675, [11]
        [c_1]: 2.016e-05
        [parameter_eliminate]: 2.49999e-06
        [updatestate_depend_eliminate]: 5.37001e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.633e-05
        [call_graph_tuple_transform]: 1.742e-05
        [tuple_list_get_item_eliminator]: 5.40001e-06
        [none_parameter_eliminate]: 1.02998e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.56e-06
[remove_dup_value]: 1.177e-05
[partial_unused_args_eliminate]: 2.01e-06
[environ_conv]: 4.31002e-06
[add_recomputation]: 3.824e-05
[cse_after_recomputation]: 1.955e-05, [1]
    [Cycle 1]: 1.4e-05, [1]
        [cse]: 8.45999e-06
[auto_monad_reorder]: 1.252e-05
[get_jit_bprop_graph]: 1.82001e-06
[rewriter_after_jit_bprop_graph]: 4.35e-06
[opt_after_jit_grad]: 0.00043173
[symbol_engine_optimizer]: 7.253e-05, [1]
    [Cycle 1]: 6.642e-05, [6]
        [build]: 3.03998e-06
        [elim_shapecalc]: 8.06001e-06
        [elim_not_effective]: 1.237e-05
        [opt_reshape]: 5.72001e-06
        [fold_const_symbol]: 8.48001e-06
        [renormalize]: 3.59985e-07
[validate]: 2.898e-05
Sums
    bootstrap                            :     0.000274s :  3.62%
    type_inference                       :     0.004021s : 53.04%
    event_method                         :     0.000016s :  0.21%
    auto_monad                           :     0.000058s :  0.77%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000029s :  0.39%
    rewriter_before_opt_a                :     0.000109s :  1.43%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000051s :  0.67%
    jit_opt_a.loop_unroll                :     0.000035s :  0.46%
    jit_opt_a.a_1                        :     0.000581s :  7.66%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.20%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000513s :  6.76%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.22%
    jit_opt_a.cse                        :     0.000027s :  0.35%
    jit_opt_a.replace_applicator         :     0.000018s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.13%
    rewriter_after_opt_a                 :     0.000030s :  0.40%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000476s :  6.28%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.20%
    cconv                                :     0.000022s :  0.29%
    loop_unroll                          :     0.000398s :  5.25%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000012s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000038s :  0.50%
    cse_after_recomputation.cse          :     0.000008s :  0.11%
    auto_monad_reorder                   :     0.000013s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.06%
    opt_after_jit_grad                   :     0.000432s :  5.70%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.38%


Time group info:
------[substitution.]   0.000172    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.69% :     0.000005s :      3: substitution.graph_param_transform
84.08% :     0.000145s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.30% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.97% :     0.000003s :      2: substitution.replace_old_param
 5.67% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003968     2
81.37% :     0.003228s :      1: type_inference.infer
18.63% :     0.000739s :      1: type_inference.specialize
------[replace.]   0.000062     6
66.90% :     0.000042s :      4: replace.inline
33.10% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000151     6
94.30% :     0.000142s :      4: match.inline
 5.70% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.27% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 1.96% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000001s :     10: predicate.cast_eliminate
 0.79% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.41% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.42% :     0.000006s :     22: predicate.inline
 1.08% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000001s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.28% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.51% :     0.000002s :      3: predicate.mutable_eliminate
 0.59% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000002s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.33% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.60% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.61% :     0.000003s :     16: predicate.switch_defer_inline
 2.50% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.66% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.28% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.46% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.15% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000527    12
49.90% :     0.000263s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.35% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.75% :     0.000231s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012686    72
 0.32% :     0.000041s :      1: add_recomputation
 0.49% :     0.000062s :      1: auto_monad
 0.12% :     0.000015s :      1: auto_monad_reorder
 2.26% :     0.000286s :      1: bootstrap
 0.19% :     0.000025s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
38.64% :     0.004902s :      1: jit_opt_a
 1.07% :     0.000136s :      1: jit_opt_after_cconv
 0.42% :     0.000053s :      1: jit_opt_b
 3.20% :     0.000406s :      1: loop_unroll
 3.82% :     0.000484s :      1: mutable_eliminate
 6.14% :     0.000779s :     26: opt.transform.jit_opt_a
 0.35% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.46% :     0.000439s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000032s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000014s :      1: remove_dup_value
 2.04% :     0.000259s :      1: renormalize.infer
 1.95% :     0.000248s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000033s :      1: rewriter_after_opt_a
 0.89% :     0.000113s :      1: rewriter_before_opt_a
 0.59% :     0.000075s :      1: symbol_engine_optimizer
31.78% :     0.004031s :      1: type_inference


TotalTime = 0.0171666, [33]
[bootstrap]: 0.00024603
[type_inference]: 0.00398439
[event_method]: 1.74e-05
[auto_monad]: 6.126e-05
[graph_reusing]: 5.29e-06
[pre_auto_parallel]: 2.69999e-06
[py_interpret_to_execute]: 2.879e-05
[rewriter_before_opt_a]: 6.845e-05
[expand_dump_flag]: 3.11001e-06
[jit_opt_a]: 0.00467441, [2]
    [Cycle 1]: 0.00127369, [27]
        [switch_simplify]: 4.831e-05
        [loop_unroll]: 2.821e-05
        [a_1]: 0.00046909
        [with_stream_mark]: 1.51e-05
        [recompute_prepare]: 7.18998e-06
        [updatestate_depend_eliminate]: 3.51001e-06
        [updatestate_assign_eliminate]: 2.74999e-06
        [updatestate_loads_eliminate]: 2.37001e-06
        [parameter_eliminate]: 1.69e-06
        [specialize_transform]: 5.49998e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 4.97e-06
        [meta_shard_fg_expand]: 1.94999e-06
        [get_grad_eliminate_]: 4.84e-06
        [merge_forward]: 3.55e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.315e-05
        [j_node_and_user_rematch]: 8.82e-06
        [meta_fg_expand]: 2.17999e-06
        [replace_old_param]: 8.08999e-06
        [inline_without_move]: 4.93001e-06
        [renormalize]: 0.0004226
        [add_forward_monad_depend]: 5.51e-06
        [auto_monad_grad]: 2.51e-06
        [auto_monad_eliminator]: 1.236e-05
        [cse]: 2.312e-05
        [replace_applicator]: 1.109e-05
    [Cycle 2]: 0.00030449, [27]
        [switch_simplify]: 5.10001e-06
        [loop_unroll]: 4.67998e-06
        [a_1]: 8.917e-05
        [with_stream_mark]: 8.74998e-06
        [recompute_prepare]: 4.83001e-06
        [updatestate_depend_eliminate]: 2.63e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 4.59998e-06
        [updatestate_useless_node_eliminater]: 4.70999e-06
        [accelerated_algorithm]: 4.72998e-06
        [meta_shard_fg_expand]: 1.15999e-06
        [get_grad_eliminate_]: 4.35e-06
        [merge_forward]: 2.50002e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.193e-05
        [j_node_and_user_rematch]: 7.18e-06
        [meta_fg_expand]: 1.35999e-06
        [replace_old_param]: 6.89001e-06
        [inline_without_move]: 4.60999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 4.33001e-06
        [cse]: 9.32001e-06
        [replace_applicator]: 5.02999e-06
[py_interpret_to_execute_after_opt_a]: 9.27001e-06
[rewriter_after_opt_a]: 3.013e-05
[convert_after_rewriter]: 6.42001e-06
[order_py_execute_after_rewriter]: 4.58999e-06
[mutable_eliminate]: 0.00047492
[jit_opt_b]: 4.554e-05, [1]
    [Cycle 1]: 3.903e-05, [2]
        [frontend_op_eliminate]: 1.432e-05
        [inline_after_opt_a]: 1.365e-05
[cconv]: 2.4e-05
[loop_unroll]: 0.00040872
[jit_opt_after_cconv]: 0.00012991, [1]
    [Cycle 1]: 0.0001238, [11]
        [c_1]: 1.965e-05
        [parameter_eliminate]: 2.54001e-06
        [updatestate_depend_eliminate]: 4.82e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.596e-05
        [call_graph_tuple_transform]: 1.666e-05
        [tuple_list_get_item_eliminator]: 5.09003e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 5.10999e-06
[remove_dup_value]: 1.236e-05
[partial_unused_args_eliminate]: 2.02999e-06
[environ_conv]: 4.97e-06
[add_recomputation]: 4.685e-05
[cse_after_recomputation]: 2.007e-05, [1]
    [Cycle 1]: 1.462e-05, [1]
        [cse]: 8.89003e-06
[auto_monad_reorder]: 1.462e-05
[get_jit_bprop_graph]: 1.39998e-06
[rewriter_after_jit_bprop_graph]: 4.60999e-06
[opt_after_jit_grad]: 0.00045
[symbol_engine_optimizer]: 0.00015229, [1]
    [Cycle 1]: 0.00014573, [6]
        [build]: 3.91999e-06
        [elim_shapecalc]: 6.91999e-06
        [elim_not_effective]: 1.178e-05
        [opt_reshape]: 5.39e-06
        [fold_const_symbol]: 8.17e-06
        [renormalize]: 4.20026e-07
[validate]: 3.147e-05
[backend_pass]: 9.79984e-07
[task_emit]: 0.00602829
[execute]: 8.17e-06
Sums
    bootstrap                            :     0.000246s :  1.83%
    type_inference                       :     0.003984s : 29.63%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000061s :  0.46%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.21%
    rewriter_before_opt_a                :     0.000068s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.24%
    jit_opt_a.a_1                        :     0.000558s :  4.15%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000423s :  3.14%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.12%
    jit_opt_a.cse                        :     0.000032s :  0.24%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000030s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000475s :  3.53%
    jit_opt_b.frontend_op_eliminate      :     0.000014s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000409s :  3.04%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.35%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000450s :  3.35%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006028s : 44.83%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000163    23
 1.10% :     0.000002s :      2: substitution.elim_not_effective
 0.78% :     0.000001s :      2: substitution.fold_const_symbol
 3.11% :     0.000005s :      3: substitution.graph_param_transform
81.47% :     0.000133s :      4: substitution.inline
 1.62% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.64% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.12% :     0.000003s :      2: substitution.replace_old_param
 7.14% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003929     2
81.62% :     0.003207s :      1: type_inference.infer
18.38% :     0.000722s :      1: type_inference.specialize
------[replace.]   0.000058     6
65.99% :     0.000038s :      4: replace.inline
34.01% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000141     6
92.55% :     0.000131s :      4: match.inline
 7.45% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.43% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.18% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000001s :     10: predicate.addn_zero_filter
 2.09% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000001s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.39% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.41% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.26% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.54% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.20% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.44% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000002s :     15: predicate.partial_eliminate
 1.42% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.54% :     0.000002s :     10: predicate.reshape_eliminate
 1.29% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.69% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.63% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.56% :     0.000008s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.31% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.80% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000450    12
48.11% :     0.000217s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.45% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.44% :     0.000209s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018443    76
 0.27% :     0.000050s :      1: add_recomputation
 0.35% :     0.000065s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.39% :     0.000256s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.12% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
25.36% :     0.004677s :      1: jit_opt_a
 0.72% :     0.000132s :      1: jit_opt_after_cconv
 0.26% :     0.000048s :      1: jit_opt_b
 2.25% :     0.000416s :      1: loop_unroll
 2.62% :     0.000482s :      1: mutable_eliminate
 4.07% :     0.000751s :     26: opt.transform.jit_opt_a
 0.24% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000022s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.48% :     0.000458s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000031s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.18% :     0.000217s :      1: renormalize.infer
 1.08% :     0.000198s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000033s :      1: rewriter_after_opt_a
 0.39% :     0.000071s :      1: rewriter_before_opt_a
 0.84% :     0.000155s :      1: symbol_engine_optimizer
32.74% :     0.006038s :      1: task_emit
21.66% :     0.003994s :      1: type_inference
 0.26% :     0.000049s :      1: validate


TotalTime = 0.0326709, [33]
[bootstrap]: 0.00019903
[type_inference]: 0.0117113
[event_method]: 6.258e-05
[auto_monad]: 0.00010158
[graph_reusing]: 7.01001e-06
[pre_auto_parallel]: 2.76e-06
[py_interpret_to_execute]: 2.895e-05
[rewriter_before_opt_a]: 0.00010046
[expand_dump_flag]: 3.21001e-06
[jit_opt_a]: 0.0112045, [3]
    [Cycle 1]: 0.0057786, [27]
        [switch_simplify]: 0.00012497
        [loop_unroll]: 4.563e-05
        [a_1]: 0.00095523
        [with_stream_mark]: 2.952e-05
        [recompute_prepare]: 1.767e-05
        [updatestate_depend_eliminate]: 7.9e-06
        [updatestate_assign_eliminate]: 6.67002e-06
        [updatestate_loads_eliminate]: 5.52999e-06
        [parameter_eliminate]: 2.54001e-06
        [specialize_transform]: 1.328e-05
        [updatestate_useless_node_eliminater]: 1.202e-05
        [accelerated_algorithm]: 1.189e-05
        [meta_shard_fg_expand]: 3.61001e-06
        [get_grad_eliminate_]: 1.2e-05
        [merge_forward]: 8.61002e-06
        [cell_reuse_recompute_pass]: 1.36002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.634e-05
        [j_node_and_user_rematch]: 2.229e-05
        [meta_fg_expand]: 0.00148998
        [replace_old_param]: 5.007e-05
        [inline_without_move]: 5.055e-05
        [renormalize]: 0.0024072
        [add_forward_monad_depend]: 9.17001e-06
        [auto_monad_grad]: 5.97999e-06
        [auto_monad_eliminator]: 4.871e-05
        [cse]: 0.00014462
        [replace_applicator]: 6.31e-05
    [Cycle 2]: 0.00191517, [27]
        [switch_simplify]: 3.721e-05
        [loop_unroll]: 3.354e-05
        [a_1]: 0.00101999
        [with_stream_mark]: 1.144e-05
        [recompute_prepare]: 7.89002e-06
        [updatestate_depend_eliminate]: 3.84002e-06
        [updatestate_assign_eliminate]: 3.33e-06
        [updatestate_loads_eliminate]: 2.82002e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 7.01001e-06
        [updatestate_useless_node_eliminater]: 6.09999e-06
        [accelerated_algorithm]: 5.74e-06
        [meta_shard_fg_expand]: 1.70001e-06
        [get_grad_eliminate_]: 5.60001e-06
        [merge_forward]: 3.56001e-06
        [cell_reuse_recompute_pass]: 7.7e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.412e-05
        [j_node_and_user_rematch]: 1.013e-05
        [meta_fg_expand]: 7.118e-05
        [replace_old_param]: 1.02e-05
        [inline_without_move]: 6.47001e-06
        [renormalize]: 0.00045745
        [add_forward_monad_depend]: 4.09002e-06
        [auto_monad_grad]: 1.07998e-06
        [auto_monad_eliminator]: 1.158e-05
        [cse]: 1.945e-05
        [replace_applicator]: 1.209e-05
    [Cycle 3]: 0.00037259, [27]
        [switch_simplify]: 7.06001e-06
        [loop_unroll]: 6.05002e-06
        [a_1]: 0.00011936
        [with_stream_mark]: 9.82999e-06
        [recompute_prepare]: 6.11e-06
        [updatestate_depend_eliminate]: 3.66999e-06
        [updatestate_assign_eliminate]: 2.78998e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 1.02998e-06
        [specialize_transform]: 6.59999e-06
        [updatestate_useless_node_eliminater]: 6.29001e-06
        [accelerated_algorithm]: 6.13998e-06
        [meta_shard_fg_expand]: 1.40999e-06
        [get_grad_eliminate_]: 5.97001e-06
        [merge_forward]: 3.05998e-06
        [cell_reuse_recompute_pass]: 1.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.491e-05
        [j_node_and_user_rematch]: 1.004e-05
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 8.38999e-06
        [inline_without_move]: 6.12001e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 6.99976e-07
        [auto_monad_eliminator]: 6.84001e-06
        [cse]: 1.427e-05
        [replace_applicator]: 6.03002e-06
[py_interpret_to_execute_after_opt_a]: 1.211e-05
[rewriter_after_opt_a]: 3.942e-05
[convert_after_rewriter]: 8.11002e-06
[order_py_execute_after_rewriter]: 5.45001e-06
[mutable_eliminate]: 0.00053828
[jit_opt_b]: 6.488e-05, [1]
    [Cycle 1]: 5.784e-05, [2]
        [frontend_op_eliminate]: 2.753e-05
        [inline_after_opt_a]: 1.812e-05
[cconv]: 2.345e-05
[loop_unroll]: 0.00043317
[jit_opt_after_cconv]: 0.00015676, [1]
    [Cycle 1]: 0.00015069, [11]
        [c_1]: 2.631e-05
        [parameter_eliminate]: 2.47001e-06
        [updatestate_depend_eliminate]: 6.64999e-06
        [updatestate_assign_eliminate]: 3.11001e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [cse]: 2.369e-05
        [call_graph_tuple_transform]: 2.13e-05
        [tuple_list_get_item_eliminator]: 6.31998e-06
        [none_parameter_eliminate]: 1.38002e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 7e-06
[remove_dup_value]: 1.666e-05
[partial_unused_args_eliminate]: 2.74999e-06
[environ_conv]: 5.52001e-06
[add_recomputation]: 5.772e-05
[cse_after_recomputation]: 2.727e-05, [1]
    [Cycle 1]: 2.194e-05, [1]
        [cse]: 1.59e-05
[auto_monad_reorder]: 1.88e-05
[get_jit_bprop_graph]: 1.82001e-06
[rewriter_after_jit_bprop_graph]: 5.23002e-06
[opt_after_jit_grad]: 0.00050659
[symbol_engine_optimizer]: 8.227e-05, [1]
    [Cycle 1]: 7.551e-05, [6]
        [build]: 3.93999e-06
        [elim_shapecalc]: 9.22001e-06
        [elim_not_effective]: 1.641e-05
        [opt_reshape]: 7.16999e-06
        [fold_const_symbol]: 1.005e-05
        [renormalize]: 6.50005e-07
[validate]: 3.544e-05
[backend_pass]: 9.49978e-07
[task_emit]: 0.00700062
[execute]: 7.43e-06
Sums
    bootstrap                            :     0.000199s :  0.69%
    type_inference                       :     0.011711s : 40.73%
    event_method                         :     0.000063s :  0.22%
    auto_monad                           :     0.000102s :  0.35%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000029s :  0.10%
    rewriter_before_opt_a                :     0.000100s :  0.35%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000169s :  0.59%
    jit_opt_a.loop_unroll                :     0.000085s :  0.30%
    jit_opt_a.a_1                        :     0.002095s :  7.28%
    jit_opt_a.with_stream_mark           :     0.000051s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000027s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.08%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000055s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001563s :  5.44%
    jit_opt_a.replace_old_param          :     0.000069s :  0.24%
    jit_opt_a.inline_without_move        :     0.000063s :  0.22%
    jit_opt_a.renormalize                :     0.002865s :  9.96%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000067s :  0.23%
    jit_opt_a.cse                        :     0.000178s :  0.62%
    jit_opt_a.replace_applicator         :     0.000081s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000039s :  0.14%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000538s :  1.87%
    jit_opt_b.frontend_op_eliminate      :     0.000028s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000023s :  0.08%
    loop_unroll                          :     0.000433s :  1.51%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000017s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000058s :  0.20%
    cse_after_recomputation.cse          :     0.000016s :  0.06%
    auto_monad_reorder                   :     0.000019s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000507s :  1.76%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007001s : 24.35%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000536   113
 0.42% :     0.000002s :      3: substitution.elim_not_effective
 0.30% :     0.000002s :      3: substitution.fold_const_symbol
 1.20% :     0.000006s :      4: substitution.graph_param_transform
66.95% :     0.000359s :     14: substitution.inline
 2.94% :     0.000016s :      2: substitution.inline_without_move
 1.69% :     0.000009s :     14: substitution.j_node_and_user_rematch
 1.80% :     0.000010s :      7: substitution.minmaximum_grad
 1.27% :     0.000007s :      5: substitution.partial_eliminate
 2.12% :     0.000011s :     14: substitution.remove_not_recompute_node
 4.64% :     0.000025s :     10: substitution.replace_applicator
 1.25% :     0.000007s :      7: substitution.replace_old_param
 0.44% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.48% :     0.000008s :      1: substitution.switch_simplify
 3.82% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.58% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.09% :     0.000038s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011641     2
89.15% :     0.010378s :      1: type_inference.infer
10.85% :     0.001263s :      1: type_inference.specialize
------[replace.]   0.000184    22
53.00% :     0.000097s :     14: replace.inline
19.10% :     0.000035s :      1: replace.switch_simplify
27.90% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000378    22
92.94% :     0.000351s :     14: match.inline
 1.94% :     0.000007s :      1: match.switch_simplify
 5.11% :     0.000019s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000386  2634
 1.51% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.57% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.43% :     0.000006s :     44: predicate.addn_check_dump
 1.61% :     0.000006s :     44: predicate.addn_zero_filter
 1.91% :     0.000007s :     44: predicate.arithmetic_simplify
 1.45% :     0.000006s :     44: predicate.cast_eliminate
 0.15% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.46% :     0.000006s :     44: predicate.compare_switch_simplify
 1.64% :     0.000006s :     44: predicate.depend_value_elim
 1.45% :     0.000006s :     44: predicate.dict_get_item_const_eliminator
 1.64% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.46% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.34% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      4: predicate.elim_not_effective
 0.31% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.44% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.38% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.36% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.44% :     0.000006s :     44: predicate.environ_get_eliminate
 1.36% :     0.000005s :     44: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.86% :     0.000003s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.54% :     0.000018s :     73: predicate.inline
 1.69% :     0.000007s :     40: predicate.inline_without_move
 0.36% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.86% :     0.000003s :     19: predicate.less_batch_normalization
 1.82% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.87% :     0.000007s :     55: predicate.load_eliminater
 0.53% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.47% :     0.000013s :     91: predicate.loop_unroll_before_grad
 1.78% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.45% :     0.000006s :     44: predicate.merge_addn
 1.58% :     0.000006s :     44: predicate.minmaximum_grad
 0.65% :     0.000003s :      4: predicate.mutable_eliminate
 0.21% :     0.000001s :      4: predicate.opt_reshape
 2.30% :     0.000009s :     55: predicate.partial_eliminate
 1.49% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.91% :     0.000007s :     44: predicate.reduce_eliminate
 1.73% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.40% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.41% :     0.000009s :     97: predicate.replace_applicator
 0.89% :     0.000003s :     40: predicate.replace_old_param
 0.11% :     0.000000s :      4: predicate.reset_defer_inline
 1.72% :     0.000007s :     44: predicate.reshape_eliminate
 1.54% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.30% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.52% :     0.000006s :     44: predicate.same_eliminate
 0.46% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.40% :     0.000002s :      8: predicate.special_op_eliminate
 0.86% :     0.000003s :     19: predicate.specialize_transform
 1.74% :     0.000007s :     44: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000005s :     44: predicate.stack_unstack_eliminate
 0.16% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.83% :     0.000011s :     65: predicate.switch_defer_inline
 2.61% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.91% :     0.000027s :    162: predicate.switch_simplify
 1.42% :     0.000005s :     44: predicate.tile_eliminate
 1.78% :     0.000007s :     44: predicate.transpose_eliminate
 1.89% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.75% :     0.000007s :     44: predicate.tuple_list_get_item_depend_reorder
 2.90% :     0.000011s :     59: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.90% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000007s :     44: predicate.value_based_eliminate
 0.15% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.28% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001479    32
61.87% :     0.000915s :     14: func_graph_cloner_run.FuncGraphClonerGraph
38.13% :     0.000564s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038424    91
 0.16% :     0.000061s :      1: add_recomputation
 0.28% :     0.000107s :      1: auto_monad
 0.06% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.55% :     0.000212s :      1: bootstrap
 0.07% :     0.000026s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.08% :     0.000030s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.18% :     0.000068s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
29.17% :     0.011208s :      1: jit_opt_a
 0.41% :     0.000159s :      1: jit_opt_after_cconv
 0.18% :     0.000068s :      1: jit_opt_b
 1.15% :     0.000441s :      1: loop_unroll
 1.42% :     0.000546s :      1: mutable_eliminate
 7.15% :     0.002748s :     39: opt.transform.jit_opt_a
 0.15% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000039s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000040s :      4: opt.transform.symbol_engine_opt
 1.34% :     0.000516s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000032s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 4.06% :     0.001562s :      2: renormalize.infer
 3.35% :     0.001288s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000043s :      1: rewriter_after_opt_a
 0.27% :     0.000104s :      1: rewriter_before_opt_a
 0.22% :     0.000085s :      1: symbol_engine_optimizer
18.26% :     0.007016s :      1: task_emit
30.51% :     0.011724s :      1: type_inference
 0.15% :     0.000058s :      1: validate


TotalTime = 0.0126313, [30]
[bootstrap]: 0.00026514
[type_inference]: 0.00423695
[event_method]: 1.956e-05
[auto_monad]: 6.655e-05
[graph_reusing]: 5.72001e-06
[pre_auto_parallel]: 2.68e-06
[py_interpret_to_execute]: 3.344e-05
[rewriter_before_opt_a]: 7.53e-05
[expand_dump_flag]: 3.25e-06
[jit_opt_a]: 0.00572339, [2]
    [Cycle 1]: 0.00170473, [27]
        [switch_simplify]: 5.995e-05
        [loop_unroll]: 3.08e-05
        [a_1]: 0.00054703
        [with_stream_mark]: 1.878e-05
        [recompute_prepare]: 7.63999e-06
        [updatestate_depend_eliminate]: 4.05e-06
        [updatestate_assign_eliminate]: 3.20998e-06
        [updatestate_loads_eliminate]: 2.78998e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 5.79999e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 2.23998e-06
        [get_grad_eliminate_]: 5.46002e-06
        [merge_forward]: 3.72998e-06
        [cell_reuse_recompute_pass]: 1.84e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.375e-05
        [j_node_and_user_rematch]: 9.72001e-06
        [meta_fg_expand]: 2.16e-06
        [replace_old_param]: 9.02999e-06
        [inline_without_move]: 5.24e-06
        [renormalize]: 0.00065449
        [add_forward_monad_depend]: 6.51e-06
        [auto_monad_grad]: 2.64999e-06
        [auto_monad_eliminator]: 1.312e-05
        [cse]: 2.723e-05
        [replace_applicator]: 1.307e-05
    [Cycle 2]: 0.00032375, [27]
        [switch_simplify]: 5.79999e-06
        [loop_unroll]: 4.84003e-06
        [a_1]: 9.068e-05
        [with_stream_mark]: 1.016e-05
        [recompute_prepare]: 5.29e-06
        [updatestate_depend_eliminate]: 2.79999e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 4.87e-06
        [updatestate_useless_node_eliminater]: 4.91002e-06
        [accelerated_algorithm]: 5.03002e-06
        [meta_shard_fg_expand]: 1.44998e-06
        [get_grad_eliminate_]: 4.58001e-06
        [merge_forward]: 2.60002e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.255e-05
        [j_node_and_user_rematch]: 8.53001e-06
        [meta_fg_expand]: 1.55001e-06
        [replace_old_param]: 7.98999e-06
        [inline_without_move]: 4.79e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 4.89e-06
        [cse]: 1.063e-05
        [replace_applicator]: 5.08002e-06
[py_interpret_to_execute_after_opt_a]: 1.109e-05
[rewriter_after_opt_a]: 3.545e-05
[convert_after_rewriter]: 7.58999e-06
[order_py_execute_after_rewriter]: 5.15999e-06
[mutable_eliminate]: 0.00063568
[jit_opt_b]: 5.32e-05, [1]
    [Cycle 1]: 4.496e-05, [2]
        [frontend_op_eliminate]: 1.699e-05
        [inline_after_opt_a]: 1.575e-05
[cconv]: 2.701e-05
[loop_unroll]: 0.00042656
[jit_opt_after_cconv]: 0.00014439, [1]
    [Cycle 1]: 0.00013773, [11]
        [c_1]: 2.076e-05
        [parameter_eliminate]: 2.86e-06
        [updatestate_depend_eliminate]: 6.24999e-06
        [updatestate_assign_eliminate]: 2.50002e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 2.016e-05
        [call_graph_tuple_transform]: 1.903e-05
        [tuple_list_get_item_eliminator]: 5.76e-06
        [none_parameter_eliminate]: 1.46998e-06
        [renormalize]: 6.60017e-07
        [switch_simplify]: 5.38002e-06
[remove_dup_value]: 1.408e-05
[partial_unused_args_eliminate]: 2.61e-06
[environ_conv]: 5.73002e-06
[add_recomputation]: 4.801e-05
[cse_after_recomputation]: 2.075e-05, [1]
    [Cycle 1]: 1.462e-05, [1]
        [cse]: 8.74e-06
[auto_monad_reorder]: 1.45e-05
[get_jit_bprop_graph]: 1.87999e-06
[rewriter_after_jit_bprop_graph]: 4.83001e-06
[opt_after_jit_grad]: 0.0004625
[symbol_engine_optimizer]: 7.66e-05, [1]
    [Cycle 1]: 6.904e-05, [6]
        [build]: 3.91999e-06
        [elim_shapecalc]: 8e-06
        [elim_not_effective]: 1.309e-05
        [opt_reshape]: 5.97999e-06
        [fold_const_symbol]: 8.96002e-06
        [renormalize]: 9.20001e-07
[validate]: 3.401e-05
Sums
    bootstrap                            :     0.000265s :  3.20%
    type_inference                       :     0.004237s : 51.14%
    event_method                         :     0.000020s :  0.24%
    auto_monad                           :     0.000067s :  0.80%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.40%
    rewriter_before_opt_a                :     0.000075s :  0.91%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000066s :  0.79%
    jit_opt_a.loop_unroll                :     0.000036s :  0.43%
    jit_opt_a.a_1                        :     0.000638s :  7.70%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000655s :  7.90%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.22%
    jit_opt_a.cse                        :     0.000038s :  0.46%
    jit_opt_a.replace_applicator         :     0.000018s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.13%
    rewriter_after_opt_a                 :     0.000035s :  0.43%
    convert_after_rewriter               :     0.000008s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000636s :  7.67%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.19%
    cconv                                :     0.000027s :  0.33%
    loop_unroll                          :     0.000427s :  5.15%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000048s :  0.58%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000014s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000463s :  5.58%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000034s :  0.41%


Time group info:
------[substitution.]   0.000219    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.60% :     0.000006s :      3: substitution.graph_param_transform
83.53% :     0.000183s :      4: substitution.inline
 1.84% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.15% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.00% :     0.000004s :      2: substitution.replace_old_param
 6.25% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004168     2
79.87% :     0.003329s :      1: type_inference.infer
20.13% :     0.000839s :      1: type_inference.specialize
------[replace.]   0.000068     6
67.59% :     0.000046s :      4: replace.inline
32.41% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000193     6
93.49% :     0.000180s :      4: match.inline
 6.51% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.22% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.66% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.19% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.27% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.39% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.25% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.58% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.65% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.38% :     0.000003s :     15: predicate.partial_eliminate
 1.27% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.80% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.82% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.66% :     0.000002s :     10: predicate.reshape_eliminate
 1.53% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.63% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.56% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.48% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.86% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.92% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000546    12
47.07% :     0.000257s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.46% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.46% :     0.000248s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014197    72
 0.36% :     0.000051s :      1: add_recomputation
 0.49% :     0.000070s :      1: auto_monad
 0.12% :     0.000017s :      1: auto_monad_reorder
 1.95% :     0.000277s :      1: bootstrap
 0.21% :     0.000030s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.17% :     0.000024s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
40.34% :     0.005726s :      1: jit_opt_a
 1.04% :     0.000147s :      1: jit_opt_after_cconv
 0.39% :     0.000056s :      1: jit_opt_b
 3.06% :     0.000435s :      1: loop_unroll
 4.55% :     0.000646s :      1: mutable_eliminate
 6.01% :     0.000853s :     26: opt.transform.jit_opt_a
 0.33% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.32% :     0.000471s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000036s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.39% :     0.000339s :      1: renormalize.infer
 2.17% :     0.000308s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000038s :      1: rewriter_after_opt_a
 0.55% :     0.000078s :      1: rewriter_before_opt_a
 0.56% :     0.000080s :      1: symbol_engine_optimizer
29.95% :     0.004252s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.941.288 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc41310> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.941.864 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0201529, [33]
[bootstrap]: 0.0003627
[type_inference]: 0.00466734
[event_method]: 1.893e-05
[auto_monad]: 6.801e-05
[graph_reusing]: 5.14e-06
[pre_auto_parallel]: 2.47001e-06
[py_interpret_to_execute]: 4.334e-05
[rewriter_before_opt_a]: 7.374e-05
[expand_dump_flag]: 3.04999e-06
[jit_opt_a]: 0.00572108, [2]
    [Cycle 1]: 0.00175555, [27]
        [switch_simplify]: 5.589e-05
        [loop_unroll]: 2.849e-05
        [a_1]: 0.00055757
        [with_stream_mark]: 1.894e-05
        [recompute_prepare]: 6.66e-06
        [updatestate_depend_eliminate]: 3.58e-06
        [updatestate_assign_eliminate]: 2.84001e-06
        [updatestate_loads_eliminate]: 2.66999e-06
        [parameter_eliminate]: 2.19001e-06
        [specialize_transform]: 6.33e-06
        [updatestate_useless_node_eliminater]: 5.24998e-06
        [accelerated_algorithm]: 6.36e-06
        [meta_shard_fg_expand]: 2.14999e-06
        [get_grad_eliminate_]: 5.70001e-06
        [merge_forward]: 3.51999e-06
        [cell_reuse_recompute_pass]: 1.35001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.398e-05
        [j_node_and_user_rematch]: 9.22999e-06
        [meta_fg_expand]: 2.07999e-06
        [replace_old_param]: 9.12001e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.00078368
        [add_forward_monad_depend]: 5.69999e-06
        [auto_monad_grad]: 2.78e-06
        [auto_monad_eliminator]: 1.297e-05
        [cse]: 1.84e-05
        [replace_applicator]: 1.165e-05
    [Cycle 2]: 0.00032508, [27]
        [switch_simplify]: 5.69e-06
        [loop_unroll]: 5.27001e-06
        [a_1]: 8.781e-05
        [with_stream_mark]: 9.14e-06
        [recompute_prepare]: 4.87998e-06
        [updatestate_depend_eliminate]: 2.99001e-06
        [updatestate_assign_eliminate]: 2.05002e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 4.73001e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 5.00999e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 4.57998e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 2.16998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.292e-05
        [j_node_and_user_rematch]: 7.83999e-06
        [meta_fg_expand]: 1.82001e-06
        [replace_old_param]: 8.31002e-06
        [inline_without_move]: 4.60001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.50001e-06
        [auto_monad_grad]: 9.49978e-07
        [auto_monad_eliminator]: 4.75001e-06
        [cse]: 1.117e-05
        [replace_applicator]: 4.72998e-06
[py_interpret_to_execute_after_opt_a]: 1.208e-05
[rewriter_after_opt_a]: 3.425e-05
[convert_after_rewriter]: 7.25998e-06
[order_py_execute_after_rewriter]: 5.15999e-06
[mutable_eliminate]: 0.0006277
[jit_opt_b]: 5.723e-05, [1]
    [Cycle 1]: 4.958e-05, [2]
        [frontend_op_eliminate]: 2.231e-05
        [inline_after_opt_a]: 1.55e-05
[cconv]: 2.879e-05
[loop_unroll]: 0.00049011
[jit_opt_after_cconv]: 0.0001495, [1]
    [Cycle 1]: 0.00014288, [11]
        [c_1]: 2.136e-05
        [parameter_eliminate]: 3.56001e-06
        [updatestate_depend_eliminate]: 6.14999e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [cse]: 2.462e-05
        [call_graph_tuple_transform]: 1.98e-05
        [tuple_list_get_item_eliminator]: 5.05999e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 8.60018e-07
        [switch_simplify]: 5.26002e-06
[remove_dup_value]: 1.434e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 5.54e-06
[add_recomputation]: 4.861e-05
[cse_after_recomputation]: 2.141e-05, [1]
    [Cycle 1]: 1.604e-05, [1]
        [cse]: 9.97001e-06
[auto_monad_reorder]: 1.622e-05
[get_jit_bprop_graph]: 2.32999e-06
[rewriter_after_jit_bprop_graph]: 5.82001e-06
[opt_after_jit_grad]: 0.00048592
[symbol_engine_optimizer]: 7.86e-05, [1]
    [Cycle 1]: 7.094e-05, [6]
        [build]: 3.98001e-06
        [elim_shapecalc]: 8.15e-06
        [elim_not_effective]: 1.417e-05
        [opt_reshape]: 6.23e-06
        [fold_const_symbol]: 8.77e-06
        [renormalize]: 1.03001e-06
[validate]: 3.953e-05
[backend_pass]: 8.2e-07
[task_emit]: 0.00679497
[execute]: 8.33999e-06
Sums
    bootstrap                            :     0.000363s :  2.29%
    type_inference                       :     0.004667s : 29.45%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000068s :  0.43%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000043s :  0.27%
    rewriter_before_opt_a                :     0.000074s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.21%
    jit_opt_a.a_1                        :     0.000645s :  4.07%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000784s :  4.95%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.11%
    jit_opt_a.cse                        :     0.000030s :  0.19%
    jit_opt_a.replace_applicator         :     0.000016s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000034s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000628s :  3.96%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000029s :  0.18%
    loop_unroll                          :     0.000490s :  3.09%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.00
TotalTime = 0.0176137, [33]
[bootstrap]: 0.0003511
[type_inference]: 0.00453196
[event_method]: 1.739e-05
[auto_monad]: 6.19e-05
[graph_reusing]: 6.14999e-06
[pre_auto_parallel]: 2.53e-06
[py_interpret_to_execute]: 2.897e-05
[rewriter_before_opt_a]: 6.982e-05
[expand_dump_flag]: 3.18998e-06
[jit_opt_a]: 0.00484508, [2]
    [Cycle 1]: 0.00134136, [27]
        [switch_simplify]: 4.821e-05
        [loop_unroll]: 2.843e-05
        [a_1]: 0.00049179
        [with_stream_mark]: 1.478e-05
        [recompute_prepare]: 7.43e-06
        [updatestate_depend_eliminate]: 3.22002e-06
        [updatestate_assign_eliminate]: 3.18998e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 1.93997e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.46998e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 5.08002e-06
        [merge_forward]: 3.49001e-06
        [cell_reuse_recompute_pass]: 1.12999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.301e-05
        [j_node_and_user_rematch]: 8.17e-06
        [meta_fg_expand]: 2.16998e-06
        [replace_old_param]: 7.93999e-06
        [inline_without_move]: 4.94003e-06
        [renormalize]: 0.00046034
        [add_forward_monad_depend]: 5.17e-06
        [auto_monad_grad]: 1.55999e-06
        [auto_monad_eliminator]: 1.274e-05
        [cse]: 2.193e-05
        [replace_applicator]: 1.058e-05
    [Cycle 2]: 0.00030411, [27]
        [switch_simplify]: 5.29e-06
        [loop_unroll]: 4.75001e-06
        [a_1]: 8.819e-05
        [with_stream_mark]: 7.74002e-06
        [recompute_prepare]: 5.34e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 1.04998e-06
        [specialize_transform]: 4.76002e-06
        [updatestate_useless_node_eliminater]: 4.50001e-06
        [accelerated_algorithm]: 4.58999e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.24002e-06
        [merge_forward]: 2.24001e-06
        [cell_reuse_recompute_pass]: 1.81e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.155e-05
        [j_node_and_user_rematch]: 7.38e-06
        [meta_fg_expand]: 1.42999e-06
        [replace_old_param]: 6.50002e-06
        [inline_without_move]: 4.63999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.09003e-06
        [auto_monad_grad]: 6.19999e-07
        [auto_monad_eliminator]: 4.38999e-06
        [cse]: 9.33002e-06
        [replace_applicator]: 5.79999e-06
[py_interpret_to_execute_after_opt_a]: 7.61001e-06
[rewriter_after_opt_a]: 3.409e-05
[convert_after_rewriter]: 7.66999e-06
[order_py_execute_after_rewriter]: 5.11002e-06
[mutable_eliminate]: 0.00048335
[jit_opt_b]: 4.675e-05, [1]
    [Cycle 1]: 4e-05, [2]
        [frontend_op_eliminate]: 1.531e-05
        [inline_after_opt_a]: 1.351e-05
[cconv]: 2.436e-05
[loop_unroll]: 0.00042349
[jit_opt_after_cconv]: 0.00013481, [1]
    [Cycle 1]: 0.00012813, [11]
        [c_1]: 2.056e-05
        [parameter_eliminate]: 2.84001e-06
        [updatestate_depend_eliminate]: 5.39998e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [cse]: 1.636e-05
        [call_graph_tuple_transform]: 1.736e-05
        [tuple_list_get_item_eliminator]: 5.14998e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 4.7998e-07
        [switch_simplify]: 5.14e-06
[remove_dup_value]: 1.365e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 4.72e-06
[add_recomputation]: 4.773e-05
[cse_after_recomputation]: 2.057e-05, [1]
    [Cycle 1]: 1.477e-05, [1]
        [cse]: 8.60001e-06
[auto_monad_reorder]: 1.449e-05
[get_jit_bprop_graph]: 1.44e-06
[rewriter_after_jit_bprop_graph]: 4.87e-06
[opt_after_jit_grad]: 0.00044545
[symbol_engine_optimizer]: 6.942e-05, [1]
    [Cycle 1]: 6.352e-05, [6]
        [build]: 3.15998e-06
        [elim_shapecalc]: 7.4e-06
  0014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000049s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000486s :  3.07%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006795s : 42.88%
    execute                              :     0.000008s :  0.05%

      [elim_not_effective]: 1.178e-05
        [opt_reshape]: 5.44e-06
        [fold_const_symbol]: 8.04002e-06
        [renormalize]: 3.69997e-07
[validate]: 2.796e-05
[backend_pass]: 1.14e-06
[task_emit]: 0.00569147
[execute]: 6.44001e-06
Sums
    bootstrap                            :     0.000351s :  2.54%
    type_inference                       :     0.004532s : 32.73%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000062s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.21%
    rewriter_before_opt_a                :     0.000070s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.39%
    jit_opt_a.loop_unroll                :     0.000033s :  0.24%
    jit_opt_a.a_1                        :     0.000580s :  4.19%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000014s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000460s :  3.33%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.12%
    jit_opt_a.cse                        :     0.000031s :  0.23%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000034s :  0.25%
    convert_after_rewriter               :     0.000008s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000483s :  3.49%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000423s :  3.06%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000048s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000445s :  3.22%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005691s : 41.11%
    execute                              :     0.000006s :  0.05%


Time group info:
------[substitution.]   0.000175    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.86% :     0.000005s :      3: substitution.graph_param_transform
82.80% :     0.000145s :      4: substitution.inline
 1.44% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.28% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.95% :     0.000003s :      2: substitution.replace_old_param
 6.84% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004471     2
83.30% :     0.003725s :      1: type_inference.infer
16.70% :     0.000747s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.50% :     0.000042s :      4: replace.inline
32.50% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000154     6
92.89% :     0.000143s :      4: match.inline
 7.11% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.57% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 1.96% :     0.000002s :     10: predicate.arithmetic_simplify
 1.19% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.53% :     0.000002s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.35% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.45% :     0.000002s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.99% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.96% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.57% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     15: predicate.partial_eliminate
 1.35% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 1.78% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.14% :     0.000001s :      3:
Time group info:
------[substitution.]   0.000226    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.59% :     0.000006s :      3: substitution.graph_param_transform
85.10% :     0.000192s :      4: substitution.inline
 1.53% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.24% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.99% :     0.000004s :      2: substitution.replace_old_param
 5.15% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004589     2
78.95% :     0.003623s :      1: type_inference.infer
21.05% :     0.000966s :      1: type_inference.specialize
------[replace.]   0.000073     6
66.19% :     0.000048s :      4: replace.inline
33.81% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     6
94.71% :     0.000190s :      4: match.inline
 5.29% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.26% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.67% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 1.97% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000001s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.17% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.81% :     0.000007s :     22: predicate.inline
 1.04% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 2.36% :     0.000003s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.40% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.38% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.89% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.43% :     0.000008s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.29% :     0.000001s :     10: predicate.transpose_eliminate
 1.28% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.84% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000518    12
51.40% :     0.000266s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.17% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.43% :     0.000225s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018954    76
 0.27% :     0.000051s :      1: add_recomputation
 0.34% :     0.000065s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.94% :     0.000367s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.57% :     0.004847s :      1: jit_opt_a
 0.72% :     0.000137s :      1: jit_opt_after_cconv
 0.26% :     0.000049s :      1: jit_opt_b
 2.28% :     0.000432s :      1: loop_unroll
 2.59% :     0.000492s :      1: mutable_eliminate
 4.08% :     0.000773s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.39% :     0.000453s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000031s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.15% :     0.000217s :      1: renormalize.infer
 1.24% :     0.000236s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000037s :      1: rewriter_after_opt_a
 0.38% :     0.000072s :      1: rewriter_before_opt_a
 0.38% :     0.000072s :      1: symbol_engine_optimizer
30.08% :     0.005702s :       predicate.row_tensor_eliminate
 1.47% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.80% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.73% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.17% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.30% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.39% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000637    12
46.46% :     0.000296s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.19% :     0.000046s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.35% :     0.000295s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021859    76
 0.24% :     0.000051s :      1: add_recomputation
 0.33% :     0.000072s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.72% :     0.000376s :      1: bootstrap
 0.14% :     0.000031s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
26.19% :     0.005724s :      1: jit_opt_a
 0.70% :     0.000152s :      1: jit_opt_after_cconv
 0.27% :     0.000060s :      1: jit_opt_b
 2.28% :     0.000499s :      1: loop_unroll
 2.92% :     0.000638s :      1: mutable_eliminate
 3.91% :     0.000855s :     26: opt.transform.jit_opt_a
 0.22% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000031s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.27% :     0.000496s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.21% :     0.000046s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.86% :     0.000407s :      1: renormalize.infer
 1.69% :     0.000369s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000038s :      1: rewriter_after_opt_a
 0.35% :     0.000077s :      1: rewriter_before_opt_a
 0.37% :     0.000081s :      1: symbol_engine_optimizer
31.15% :     0.006808s :      1: task_emit
23.98% :     0.004544s :      1: type_inference
 0.23% :     0.000044s :      1: validate

1: task_emit
21.42% :     0.004682s :      1: type_inference
 0.29% :     0.000063s :      1: validate


TotalTime = 0.0124921, [30]
[bootstrap]: 0.00029487
[type_inference]: 0.00451764
[event_method]: 6.711e-05
[auto_monad]: 7.302e-05
[graph_reusing]: 5.38002e-06
[pre_auto_parallel]: 2.17001e-06
[py_interpret_to_execute]: 3.637e-05
[rewriter_before_opt_a]: 7.942e-05
[expand_dump_flag]: 3.55e-06
[jit_opt_a]: 0.00537784, [2]
    [Cycle 1]: 0.00162894, [27]
        [switch_simplify]: 5.296e-05
        [loop_unroll]: 3.008e-05
        [a_1]: 0.00062096
        [with_stream_mark]: 1.882e-05
        [recompute_prepare]: 7.48e-06
        [updatestate_depend_eliminate]: 3.59002e-06
        [updatestate_assign_eliminate]: 3.25e-06
        [updatestate_loads_eliminate]: 2.98e-06
        [parameter_eliminate]: 2.09999e-06
        [specialize_transform]: 5.56e-06
        [updatestate_useless_node_eliminater]: 5.44998e-06
        [accelerated_algorithm]: 5.38002e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 5.22999e-06
        [merge_forward]: 4.18999e-06
        [cell_reuse_recompute_pass]: 1.14003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.425e-05
        [j_node_and_user_rematch]: 9.46998e-06
        [meta_fg_expand]: 2.69999e-06
        [replace_old_param]: 8.60001e-06
        [inline_without_move]: 5.24998e-06
        [renormalize]: 0.00058491
        [add_forward_monad_depend]: 5.21998e-06
        [auto_monad_grad]: 2.41998e-06
        [auto_monad_eliminator]: 1.31e-05
        [cse]: 2.685e-05
        [replace_applicator]: 1.105e-05
    [Cycle 2]: 0.00031985, [27]
        [switch_simplify]: 5.98002e-06
        [loop_unroll]: 5.07e-06
        [a_1]: 8.934e-05
        [with_stream_mark]: 9.20999e-06
        [recompute_prepare]: 5.17999e-06
        [updatestate_depend_eliminate]: 2.83998e-06
        [updatestate_assign_eliminate]: 2.13998e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 4.96002e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 5.28002e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.52998e-06
        [merge_forward]: 2.23002e-06
        [cell_reuse_recompute_pass]: 1.41002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.258e-05
        [j_node_and_user_rematch]: 7.81001e-06
        [meta_fg_expand]: 1.72001e-06
        [replace_old_param]: 6.63998e-06
        [inline_without_move]: 1.033e-05
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 5.17e-06
        [cse]: 1.03e-05
        [replace_applicator]: 5.02999e-06
[py_interpret_to_execute_after_opt_a]: 9.54e-06
[rewriter_after_opt_a]: 3.618e-05
[convert_after_rewriter]: 6.91001e-06
[order_py_execute_after_rewriter]: 5.45001e-06
[mutable_eliminate]: 0.00048295
[jit_opt_b]: 4.941e-05, [1]
    [Cycle 1]: 4.246e-05, [2]
        [frontend_op_eliminate]: 1.53e-05
        [inline_after_opt_a]: 1.49e-05
[cconv]: 2.409e-05
[loop_unroll]: 0.0004111
[jit_opt_after_cconv]: 0.00013853, [1]
    [Cycle 1]: 0.00013175, [11]
        [c_1]: 2.127e-05
        [parameter_eliminate]: 2.21e-06
        [updatestate_depend_eliminate]: 5.27001e-06
        [updatestate_assign_eliminate]: 2.43002e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 1.743e-05
        [call_graph_tuple_transform]: 1.809e-05
        [tuple_list_get_item_eliminator]: 5.36998e-06
        [none_parameter_eliminate]: 1.60001e-06
        [renormalize]: 2.69996e-07
        [switch_simplify]: 5.39e-06
[remove_dup_value]: 1.408e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 5.47001e-06
[add_recomputation]: 4.767e-05
[cse_after_recomputation]: 2.225e-05, [1]
    [Cycle 1]: 1.58e-05, [1]
        [cse]: 9.57001e-06
[auto_monad_reorder]: 1.561e-05
[get_jit_bprop_graph]: 1.82001e-06
[rewriter_after_jit_bprop_graph]: 5.72001e-06
[opt_after_jit_grad]: 0.00046145
[symbol_engine_optimizer]: 8.012e-05, [1]
    [Cycle 1]: 7.346e-05, [6]
        [build]: 4.24997e-06
        [elim_shapecalc]: 7.93999e-06
        [elim_not_effective]: 1.459e-05
        [opt_reshape]: 6.59999e-06
        [fold_const_symbol]: 9.38002e-06
        [renormalize]: 8.2e-07
[validate]: 3.738e-05
Sums
    bootstrap                            :     0.000295s :  3.48%
    type_inference                       :     0.004518s : 53.30%
    event_method                         :     0.000067s :  0.79%
    auto_monad                           :     0.000073s :  0.86%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.43%
    rewriter_before_opt_a                :     0.000079s :  0.94%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000059s :  0.70%
    jit_opt_a.loop_unroll                :     0.000035s :  0.41%
    jit_opt_a.a_1                        :     0.000710s :  8.38%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.18%
    jit_opt_a.inline_without_move        :     0.000016s :  0.18%
    jit_opt_a.renormalize                :     0.000585s :  6.90%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.22%
    jit_opt_a.cse                        :     0.000037s :  0.44%
    jit_opt_a.replace_applicator         :     0.000016s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.11%
    rewriter_after_opt_a                 :     0.000036s :  0.43%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000483s :  5.70%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000024s :  0.28%
    loop_unroll                          :     0.000411s :  4.85%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000048s :  0.56%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000461s :  5.44%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000037s :  0.44%


Time group info:
------[substitution.]   0.000198    23
 1.20% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.64% :     0.000005s :      3: substitution.graph_param_transform
83.90% :     0.000166s :      4: substitution.inline
 1.72% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.19% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.82% :     0.000004s :      2: substitution.replace_old_param
 5.81% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004456     2
81.71% :     0.003641s :      1: type_inference.infer
18.29% :     0.000815s :      1: type_inference.specialize
------[replace.]   0.000065     6
67.08% :     0.000043s :      4: replace.inline
32.92% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000174     6
94.10% :     0.000164s :      4: match.inline
 5.90% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.29% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.39% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000001s :     10: predicate.addn_zero_filter
 2.09% :     0.000002s :     10: predicate.arithmetic_simplify
 1.39% :     0.000002s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.47% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.64% :     0.000006s :     22: predicate.inline
 1.02% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.37% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.50% :     0.000002s :      3: predicate.mutable_eliminate
 0.59% :     0.000001s :      3: predicate.opt_reshape
 2.36% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.77% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.82% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000001s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.53% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.80% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.06% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.73% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000579    12
50.66% :     0.000293s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.27% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.07% :     0.000249s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014051    72
 0.36% :     0.000051s :      1: add_recomputation
 0.55% :     0.000078s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 2.22% :     0.000312s :      1: bootstrap
 0.19% :     0.000027s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.51% :     0.000072s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.30% :     0.005381s :      1: jit_opt_a
 1.01% :     0.000141s :      1: jit_opt_after_cconv
 0.37% :     0.000052s :      1: jit_opt_b
 2.98% :     0.000418s :      1: loop_unroll
 3.50% :     0.000492s :      1: mutable_eliminate
 6.56% :     0.000921s :     26: opt.transform.jit_opt_a
 0.33% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.34% :     0.000470s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.28% :     0.000039s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.17% :     0.000305s :      1: renormalize.infer
 1.94% :     0.000273s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000039s :      1: rewriter_after_opt_a
 0.59% :     0.000082s :      1: rewriter_before_opt_a
 0.59% :     0.000083s :      1: symbol_engine_optimizer
32.26% :     0.004533s :      1: type_inference


TotalTime = 0.0321168, [33]
[bootstrap]: 0.00017797
[type_inference]: 0.0115596
[event_method]: 6.066e-05
[auto_monad]: 0.0001048
[graph_reusing]: 6.76e-06
[pre_auto_parallel]: 3.31001e-06
[py_interpret_to_execute]: 2.752e-05
[rewriter_before_opt_a]: 0.00011026
[expand_dump_flag]: 3.89002e-06
[jit_opt_a]: 0.0109669, [3]
    [Cycle 1]: 0.00544242, [27]
        [switch_simplify]: 0.00012588
        [loop_unroll]: 4.389e-05
        [a_1]: 0.00088072
        [with_stream_mark]: 2.248e-05
        [recompute_prepare]: 1.73e-05
        [updatestate_depend_eliminate]: 7.55998e-06
        [updatestate_assign_eliminate]: 6.29999e-06
        [updatestate_loads_eliminate]: 5.57001e-06
        [parameter_eliminate]: 2.29001e-06
        [specialize_transform]: 1.268e-05
        [updatestate_useless_node_eliminater]: 1.195e-05
        [accelerated_algorithm]: 1.224e-05
        [meta_shard_fg_expand]: 3.64002e-06
        [get_grad_eliminate_]: 1.211e-05
        [merge_forward]: 8.27998e-06
        [cell_reuse_recompute_pass]: 9.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.617e-05
        [j_node_and_user_rematch]: 2.485e-05
        [meta_fg_expand]: 0.00139897
        [replace_old_param]: 5.102e-05
        [inline_without_move]: 5.194e-05
        [renormalize]: 0.00225021
        [add_forward_monad_depend]: 9.44e-06
        [auto_monad_grad]: 6.06e-06
        [auto_monad_eliminator]: 4.677e-05
        [cse]: 0.00014332
        [replace_applicator]: 6.18e-05
    [Cycle 2]: 0.00198065, [27]
        [switch_simplify]: 3.68e-05
        [loop_unroll]: 3.405e-05
        [a_1]: 0.00101012
        [with_stream_mark]: 1.241e-05
        [recompute_prepare]: 8.25999e-06
        [updatestate_depend_eliminate]: 3.92002e-06
        [updatestate_assign_eliminate]: 3.07002e-06
        [updatestate_loads_eliminate]: 2.64999e-06
        [parameter_eliminate]: 1.43002e-06
        [specialize_transform]: 6.63998e-06
        [updatestate_useless_node_eliminater]: 6.11e-06
        [accelerated_algorithm]: 6.12001e-06
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 5.88002e-06
        [merge_forward]: 3.09999e-06
        [cell_reuse_recompute_pass]: 7.59988e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.36e-05
        [j_node_and_user_rematch]: 9.95002e-06
        [meta_fg_expand]: 6.66e-05
        [replace_old_param]: 1.029e-05
        [inline_without_move]: 6.01e-06
        [renormalize]: 0.00051856
        [add_forward_monad_depend]: 4.72e-06
        [auto_monad_grad]: 1.54998e-06
        [auto_monad_eliminator]: 1.322e-05
        [cse]: 2.538e-05
        [replace_applicator]: 1.38e-05
    [Cycle 3]: 0.0003801, [27]
        [switch_simplify]: 7.5e-06
        [loop_unroll]: 6.49001e-06
        [a_1]: 0.00012175
        [with_stream_mark]: 1.044e-05
        [recompute_prepare]: 6.32001e-06
        [updatestate_depend_eliminate]: 3.55e-06
        [updatestate_assign_eliminate]: 2.86999e-06
        [updatestate_loads_eliminate]: 2.73998e-06
        [parameter_eliminate]: 1.04998e-06
        [specialize_transform]: 6.23e-06
        [updatestate_useless_node_eliminater]: 6.16e-06
        [accelerated_algorithm]: 5.86e-06
        [meta_shard_fg_expand]: 1.56002e-06
        [get_grad_eliminate_]: 5.62999e-06
        [merge_forward]: 3.36001e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.519e-05
        [j_node_and_user_rematch]: 1.008e-05
        [meta_fg_expand]: 2.27999e-06
        [replace_old_param]: 8.22e-06
        [inline_without_move]: 6.09001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.45001e-06
        [auto_monad_grad]: 8.60018e-07
        [auto_monad_eliminator]: 6.64999e-06
        [cse]: 1.418e-05
        [replace_applicator]: 5.88002e-06
[py_interpret_to_execute_after_opt_a]: 1.237e-05
[rewriter_after_opt_a]: 3.85e-05
[convert_after_rewriter]: 7.15e-06
[order_py_execute_after_rewriter]: 6.54999e-06
[mutable_eliminate]: 0.00061456
[jit_opt_b]: 5.595e-05, [1]
    [Cycle 1]: 4.909e-05, [2]
        [frontend_op_eliminate]: 1.97e-05
        [inline_after_opt_a]: 1.836e-05
[cconv]: 2.274e-05
[loop_unroll]: 0.00042935
[jit_opt_after_cconv]: 0.00015485, [1]
    [Cycle 1]: 0.00014802, [11]
        [c_1]: 2.715e-05
        [parameter_eliminate]: 2.59001e-06
        [updatestate_depend_eliminate]: 5.52001e-06
        [updatestate_assign_eliminate]: 2.79999e-06
        [updatestate_loads_eliminate]: 2.55002e-06
        [cse]: 2.249e-05
        [call_graph_tuple_transform]: 2.034e-05
        [tuple_list_get_item_eliminator]: 6.37001e-06
        [none_parameter_eliminate]: 1.24e-06
        [renormalize]: 5.60016e-07
        [switch_simplify]: 6.77002e-06
[remove_dup_value]: 1.631e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 5.17999e-06
[add_recomputation]: 4.979e-05
[cse_after_recomputation]: 2.551e-05, [1]
    [Cycle 1]: 2.024e-05, [1]
        [cse]: 1.425e-05
[auto_monad_reorder]: 1.798e-05
[get_jit_bprop_graph]: 1.59998e-06
[rewriter_after_jit_bprop_graph]: 5.37001e-06
[opt_after_jit_grad]: 0.00046661
[symbol_engine_optimizer]: 8.086e-05, [1]
    [Cycle 1]: 7.43e-05, [6]
        [build]: 3.78001e-06
        [elim_shapecalc]: 8.69998e-06
        [elim_not_effective]: 1.535e-05
        [opt_reshape]: 6.76999e-06
        [fold_const_symbol]: 1.115e-05
        [renormalize]: 4.60015e-07
[validate]: 3.639e-05
[backend_pass]: 9.79984e-07
[task_emit]: 0.00684043
[execute]: 7.56999e-06
Sums
    bootstrap                            :     0.000178s :  0.63%
    type_inference                       :     0.011560s : 41.04%
    event_method                         :     0.000061s :  0.22%
    auto_monad                           :     0.000105s :  0.37%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.10%
    rewriter_before_opt_a                :     0.000110s :  0.39%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000170s :  0.60%
    jit_opt_a.loop_unroll                :     0.000084s :  0.30%
    jit_opt_a.a_1                        :     0.002013s :  7.14%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.08%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000055s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000045s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001468s :  5.21%
    jit_opt_a.replace_old_param          :     0.000070s :  0.25%
    jit_opt_a.inline_without_move        :     0.000064s :  0.23%
    jit_opt_a.renormalize                :     0.002769s :  9.83%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000067s :  0.24%
    jit_opt_a.cse                        :     0.000183s :  0.65%
    jit_opt_a.replace_applicator         :     0.000081s :  0.29%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000038s :  0.14%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000615s :  2.18%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.07%
    cconv                                :     0.000023s :  0.08%
    loop_unroll                          :     0.000429s :  1.52%
    jit_opt_after_cconv.c_1              :     0.000027s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000050s :  0.18%
    cse_after_recomputation.cse          :     0.000014s :  0.05%
    auto_monad_reorder                   :     0.000018s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000467s :  1.66%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006840s : 24.28%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000460   113
 0.56% :     0.000003s :      3: substitution.elim_not_effective
 0.35% :     0.000002s :      3: substitution.fold_const_symbol
 1.27% :     0.000006s :      4: substitution.graph_param_transform
62.66% :     0.000288s :     14: substitution.inline
 3.52% :     0.000016s :      2: substitution.inline_without_move
 1.87% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.12% :     0.000010s :      7: substitution.minmaximum_grad
 1.39% :     0.000006s :      5: substitution.partial_eliminate
 2.20% :     0.000010s :     14: substitution.remove_not_recompute_node
 5.20% :     0.000024s :     10: substitution.replace_applicator
 1.53% :     0.000007s :      7: substitution.replace_old_param
 0.48% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.63% :     0.000007s :      1: substitution.switch_simplify
 4.19% :     0.000019s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.90% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 8.14% :     0.000037s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011492     2
89.17% :     0.010247s :      1: type_inference.infer
10.83% :     0.001245s :      1: type_inference.specialize
------[replace.]   0.000169    22
54.82% :     0.000092s :     14: replace.inline
15.15% :     0.000026s :      1: replace.switch_simplify
30.03% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000306    22
91.44% :     0.000280s :     14: match.inline
 2.20% :     0.000007s :      1: match.switch_simplify
 6.36% :     0.000019s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000389  2634
 1.65% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.56% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.47% :     0.000006s :     44: predicate.addn_check_dump
 1.73% :     0.000007s :     44: predicate.addn_zero_filter
 1.98% :     0.000008s :     44: predicate.arithmetic_simplify
 1.49% :     0.000006s :     44: predicate.cast_eliminate
 0.27% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.44% :     0.000006s :     44: predicate.compare_switch_simplify
 1.59% :     0.000006s :     44: predicate.depend_value_elim
 1.50% :     0.000006s :     44: predicate.dict_get_item_const_eliminator
 1.74% :     0.000007s :     44: predicate.dict_get_item_eliminator
 1.49% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.31% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      4: predicate.elim_not_effective
 0.21% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.41% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.49% :     0.000006s :     44: predicate.environ_get_depend_swap
 1.49% :     0.000006s :     44: predicate.environ_get_eliminate
 1.46% :     0.000006s :     44: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.81% :     0.000003s :     19: predicate.get_grad_eliminate
 0.15% :     0.000001s :      4: predicate.graph_param_transform
 4.22% :     0.000016s :     73: predicate.inline
 1.66% :     0.000006s :     40: predicate.inline_without_move
 0.37% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.89% :     0.000003s :     19: predicate.less_batch_normalization
 1.75% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.95% :     0.000008s :     55: predicate.load_eliminater
 0.45% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.32% :     0.000013s :     91: predicate.loop_unroll_before_grad
 1.90% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.52% :     0.000006s :     44: predicate.merge_addn
 1.47% :     0.000006s :     44: predicate.minmaximum_grad
 0.56% :     0.000002s :      4: predicate.mutable_eliminate
 0.20% :     0.000001s :      4: predicate.opt_reshape
 2.28% :     0.000009s :     55: predicate.partial_eliminate
 1.50% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.94% :     0.000008s :     44: predicate.reduce_eliminate
 1.74% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.36% :     0.000009s :     97: predicate.replace_applicator
 0.87% :     0.000003s :     40: predicate.replace_old_param
 0.11% :     0.000000s :      4: predicate.reset_defer_inline
 1.63% :     0.000006s :     44: predicate.reshape_eliminate
 1.61% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.29% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.54% :     0.000006s :     44: predicate.same_eliminate
 0.47% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.36% :     0.000001s :      8: predicate.special_op_eliminate
 0.84% :     0.000003s :     19: predicate.specialize_transform
 1.84% :     0.000007s :     44: predicate.split_environ_get_set_with_tuple_value
 1.62% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.89% :     0.000011s :     65: predicate.switch_defer_inline
 2.56% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.55% :     0.000026s :    162: predicate.switch_simplify
 1.50% :     0.000006s :     44: predicate.tile_eliminate
 1.45% :     0.000006s :     44: predicate.transpose_eliminate
 1.84% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.81% :     0.000007s :     44: predicate.tuple_list_get_item_depend_reorder
 3.12% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.75% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.84% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000007s :     44: predicate.value_based_eliminate
 0.14% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.22% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001452    32
60.16% :     0.000874s :     14: func_graph_cloner_run.FuncGraphClonerGraph
39.84% :     0.000578s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.037677    91
 0.14% :     0.000053s :      1: add_recomputation
 0.29% :     0.000110s :      1: auto_monad
 0.05% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.50% :     0.000189s :      1: bootstrap
 0.07% :     0.000026s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000028s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.17% :     0.000066s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
29.12% :     0.010970s :      1: jit_opt_a
 0.42% :     0.000158s :      1: jit_opt_after_cconv
 0.16% :     0.000059s :      1: jit_opt_b
 1.16% :     0.000439s :      1: loop_unroll
 1.66% :     0.000624s :      1: mutable_eliminate
 7.08% :     0.002668s :     39: opt.transform.jit_opt_a
 0.15% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000032s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.26% :     0.000474s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000031s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 4.01% :     0.001509s :      2: renormalize.infer
 3.31% :     0.001246s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000041s :      1: rewriter_after_opt_a
 0.30% :     0.000113s :      1: rewriter_before_opt_a
 0.22% :     0.000084s :      1: symbol_engine_optimizer
18.18% :     0.006850s :      1: task_emit
30.71% :     0.011572s :      1: type_inference
 0.15% :     0.000057s :      1: validate


TotalTime = 0.016235, [33]
[bootstrap]: 0.00032638
[type_inference]: 0.00399493
[event_method]: 1.803e-05
[auto_monad]: 5.863e-05
[graph_reusing]: 5.29e-06
[pre_auto_parallel]: 2.19999e-06
[py_interpret_to_execute]: 2.69e-05
[rewriter_before_opt_a]: 6.628e-05
[expand_dump_flag]: 2.71e-06
[jit_opt_a]: 0.00482092, [2]
    [Cycle 1]: 0.00135161, [27]
        [switch_simplify]: 4.265e-05
        [loop_unroll]: 2.939e-05
        [a_1]: 0.00052088
        [with_stream_mark]: 1.228e-05
        [recompute_prepare]: 7.18998e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 6.41998e-06
        [updatestate_useless_node_eliminater]: 5.66e-06
        [accelerated_algorithm]: 6.17001e-06
        [meta_shard_fg_expand]: 1.71998e-06
        [get_grad_eliminate_]: 5.62999e-06
        [merge_forward]: 2.84999e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.267e-05
        [j_node_and_user_rematch]: 9.24e-06
        [meta_fg_expand]: 2.34001e-06
        [replace_old_param]: 8.49002e-06
        [inline_without_move]: 5.89999e-06
        [renormalize]: 0.00045609
        [add_forward_monad_depend]: 4.62e-06
        [auto_monad_grad]: 1.05999e-06
        [auto_monad_eliminator]: 1.018e-05
        [cse]: 1.52e-05
        [replace_applicator]: 1.071e-05
    [Cycle 2]: 0.00031671, [27]
        [switch_simplify]: 5.36998e-06
        [loop_unroll]: 5.06002e-06
        [a_1]: 8.952e-05
        [with_stream_mark]: 8.55999e-06
        [recompute_prepare]: 4.82e-06
        [updatestate_depend_eliminate]: 2.34001e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.83002e-06
        [parameter_eliminate]: 1.09003e-06
        [specialize_transform]: 5.47999e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 4.82998e-06
        [meta_shard_fg_expand]: 1.18001e-06
        [get_grad_eliminate_]: 4.53999e-06
        [merge_forward]: 2.39999e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.201e-05
        [j_node_and_user_rematch]: 7.79002e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 7.18e-06
        [inline_without_move]: 4.68999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 5.59998e-06
        [cse]: 9.59e-06
        [replace_applicator]: 5.09e-06
[py_interpret_to_execute_after_opt_a]: 7.53e-06
[rewriter_after_opt_a]: 2.697e-05
[convert_after_rewriter]: 5.49e-06
[order_py_execute_after_rewriter]: 4.26001e-06
[mutable_eliminate]: 0.0004878
[jit_opt_b]: 4.684e-05, [1]
    [Cycle 1]: 4.063e-05, [2]
        [frontend_op_eliminate]: 1.527e-05
        [inline_after_opt_a]: 1.446e-05
[cconv]: 1.676e-05
[loop_unroll]: 0.00042124
[jit_opt_after_cconv]: 0.00013548, [1]
    [Cycle 1]: 0.00012918, [11]
        [c_1]: 2.067e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 5.37001e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [cse]: 1.655e-05
        [call_graph_tuple_transform]: 1.653e-05
        [tuple_list_get_item_eliminator]: 5.76998e-06
        [none_parameter_eliminate]: 8.70001e-07
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.28002e-06
[remove_dup_value]: 7.11001e-06
[partial_unused_args_eliminate]: 1.55999e-06
[environ_conv]: 3.20998e-06
[add_recomputation]: 3.398e-05
[cse_after_recomputation]: 1.926e-05, [1]
    [Cycle 1]: 1.405e-05, [1]
        [cse]: 8.60001e-06
[auto_monad_reorder]: 9.09998e-06
[get_jit_bprop_graph]: 1.42e-06
[rewriter_after_jit_bprop_graph]: 4.03999e-06
[opt_after_jit_grad]: 0.00045804
[symbol_engine_optimizer]: 7.296e-05, [1]
    [Cycle 1]: 6.624e-05, [6]
        [build]: 3.09999e-06
        [elim_shapecalc]: 7.81001e-06
        [elim_not_effective]: 1.223e-05
        [opt_reshape]: 5.38002e-06
        [fold_const_symbol]: 9.20001e-06
        [renormalize]: 5.20027e-07
[validate]: 2.559e-05
[backend_pass]: 6.39993e-07
[task_emit]: 0.00495249
[execute]: 4.77e-06
Sums
    bootstrap                            :     0.000326s :  2.61%
    type_inference                       :     0.003995s : 31.92%
    event_method                         :     0.000018s :  0.14%
    auto_monad                           :     0.000059s :  0.47%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000027s :  0.21%
    rewriter_before_opt_a                :     0.000066s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000048s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.28%
    jit_opt_a.a_1                        :     0.000610s :  4.88%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.13%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.000456s :  3.64%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.13%
    jit_opt_a.cse                        :     0.000025s :  0.20%
    jit_opt_a.replace_applicator         :     0.000016s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000027s :  0.22%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000488s :  3.90%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.12%
    cconv                                :     0.000017s :  0.13%
    loop_unroll                          :     0.000421s :  3.37%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.17%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000007s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000003s :  0.03%
    add_recomputation                    :     0.000034s :  0.27%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000009s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000458s :  3.66%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000026s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.004952s : 39.57%
    execute                              :     0.000005s :  0.04%


Time group info:
------[substitution.]   0.000157    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.79% :     0.000001s :      2: substitution.fold_const_symbol
 2.52% :     0.000004s :      3: substitution.graph_param_transform
82.21% :     0.000129s :      4: substitution.inline
 1.77% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.76% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.11% :     0.000003s :      2: substitution.replace_old_param
 6.75% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003940     2
81.13% :     0.003197s :      1: type_inference.infer
18.87% :     0.000744s :      1: type_inference.specialize
------[replace.]   0.000101     6
79.09% :     0.000080s :      4: replace.inline
20.91% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000136     6
93.05% :     0.000127s :      4: match.inline
 6.95% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.56% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.53% :     0.000002s :     10: predicate.addn_zero_filter
 2.19% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.05% :     0.000006s :     22: predicate.inline
 1.07% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000001s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.91% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.16% :     0.000001s :     10: predicate.minmaximum_grad
 1.86% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000002s :     15: predicate.partial_eliminate
 1.40% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 1.20% :     0.000001s :      6: predicate.specialize_transform
 1.70% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.57% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.46% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.42% :     0.000002s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.54% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     10: predicate.value_based_eliminate
 0.43% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000467    12
46.49% :     0.000217s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.66% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.86% :     0.000224s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017608    76
 0.21% :     0.000037s :      1: add_recomputation
 0.35% :     0.000062s :      1: auto_monad
 0.07% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.91% :     0.000337s :      1: bootstrap
 0.11% :     0.000020s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.12% :     0.000022s :      1: event_method
 0.05% :     0.000008s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
27.39% :     0.004823s :      1: jit_opt_a
 0.78% :     0.000138s :      1: jit_opt_after_cconv
 0.28% :     0.000049s :      1: jit_opt_b
 2.43% :     0.000429s :      1: loop_unroll
 2.82% :     0.000497s :      1: mutable_eliminate
 4.58% :     0.000807s :     26: opt.transform.jit_opt_a
 0.26% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.65% :     0.000466s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000009s :      1: remove_dup_value
 1.29% :     0.000228s :      1: renormalize.infer
 1.26% :     0.000222s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000030s :      1: rewriter_after_opt_a
 0.39% :     0.000069s :      1: rewriter_before_opt_a
 0.43% :     0.000076s :      1: symbol_engine_optimizer
28.18% :     0.004962s :      1: task_emit
22.76% :     0.004008s :      1: type_inference
 0.24% :     0.000041s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.981.227 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dd27f70> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:18.981.803 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0132474, [30]
[bootstrap]: 0.00028656
[type_inference]: 0.0044002
[event_method]: 1.921e-05
[auto_monad]: 6.586e-05
[graph_reusing]: 6.40997e-06
[pre_auto_parallel]: 2.88e-06
[py_interpret_to_execute]: 3.538e-05
[rewriter_before_opt_a]: 7.675e-05
[expand_dump_flag]: 3.09001e-06
[jit_opt_a]: 0.00599392, [2]
    [Cycle 1]: 0.00183863, [27]
        [switch_simplify]: 6.54e-05
        [loop_unroll]: 3.614e-05
        [a_1]: 0.00057198
        [with_stream_mark]: 1.98e-05
        [recompute_prepare]: 7.39002e-06
        [updatestate_depend_eliminate]: 4.02e-06
        [updatestate_assign_eliminate]: 3.04001e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 2.49001e-06
        [specialize_transform]: 5.82001e-06
        [updatestate_useless_node_eliminater]: 5.46e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 2.24999e-06
        [get_grad_eliminate_]: 5.69e-06
        [merge_forward]: 3.94002e-06
        [cell_reuse_recompute_pass]: 1.79998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.504e-05
        [j_node_and_user_rematch]: 1.004e-05
        [meta_fg_expand]: 2.53998e-06
        [replace_old_param]: 8.61002e-06
        [inline_without_move]: 5.08002e-06
        [renormalize]: 0.00080098
        [add_forward_monad_depend]: 8.12e-06
        [auto_monad_grad]: 3.10998e-06
        [auto_monad_eliminator]: 1.728e-05
        [cse]: 2.804e-05
        [replace_applicator]: 1.399e-05
    [Cycle 2]: 0.00033868, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 5.18002e-06
        [a_1]: 9.308e-05
        [with_stream_mark]: 1.176e-05
        [recompute_prepare]: 5.46998e-06
        [updatestate_depend_eliminate]: 2.82002e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 1.44998e-06
        [specialize_transform]: 5.23002e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 5.26998e-06
        [meta_shard_fg_expand]: 1.52001e-06
        [get_grad_eliminate_]: 4.82e-06
        [merge_forward]: 3.21999e-06
        [cell_reuse_recompute_pass]: 1.99e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.434e-05
        [j_node_and_user_rematch]: 8.23001e-06
        [meta_fg_expand]: 1.74998e-06
        [replace_old_param]: 8.80001e-06
        [inline_without_move]: 4.70999e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.26002e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 5.14998e-06
        [cse]: 1.127e-05
        [replace_applicator]: 5.39998e-06
[py_interpret_to_execute_after_opt_a]: 1.314e-05
[rewriter_after_opt_a]: 3.711e-05
[convert_after_rewriter]: 7.21001e-06
[order_py_execute_after_rewriter]: 4.79e-06
[mutable_eliminate]: 0.000714
[jit_opt_b]: 5.466e-05, [1]
    [Cycle 1]: 4.632e-05, [2]
        [frontend_op_eliminate]: 1.755e-05
        [inline_after_opt_a]: 1.636e-05
[cconv]: 2.93e-05
[loop_unroll]: 0.00043755
[jit_opt_after_cconv]: 0.00014926, [1]
    [Cycle 1]: 0.00014253, [11]
        [c_1]: 2.085e-05
        [parameter_eliminate]: 3.35003e-06
        [updatestate_depend_eliminate]: 7.47002e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 2.186e-05
        [call_graph_tuple_transform]: 2.015e-05
        [tuple_list_get_item_eliminator]: 6.16e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 6.40022e-07
        [switch_simplify]: 5.43002e-06
[remove_dup_value]: 1.651e-05
[partial_unused_args_eliminate]: 3.31999e-06
[environ_conv]: 6.24001e-06
[add_recomputation]: 5.206e-05
[cse_after_recomputation]: 2.109e-05, [1]
    [Cycle 1]: 1.535e-05, [1]
        [cse]: 9.35001e-06
[auto_monad_reorder]: 1.647e-05
[get_jit_bprop_graph]: 2.19999e-06
[rewriter_after_jit_bprop_graph]: 5.30001e-06
[opt_after_jit_grad]: 0.00048119
[symbol_engine_optimizer]: 7.782e-05, [1]
    [Cycle 1]: 7.077e-05, [6]
        [build]: 4.80999e-06
        [elim_shapecalc]: 8.02998e-06
        [elim_not_effective]: 1.339e-05
        [opt_reshape]: 5.96e-06
        [fold_const_symbol]: 9.05999e-06
        [renormalize]: 5.89993e-07
[validate]: 4.154e-05
Sums
    bootstrap                            :     0.000287s :  3.25%
    type_inference                       :     0.004400s : 49.91%
    event_method                         :     0.000019s :  0.22%
    auto_monad                           :     0.000066s :  0.75%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.40%
    rewriter_before_opt_a                :     0.000077s :  0.87%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000071s :  0.81%
    jit_opt_a.loop_unroll                :     0.000041s :  0.47%
    jit_opt_a.a_1                        :     0.000665s :  7.54%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.36%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.12%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000801s :  9.09%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.11%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.25%
    jit_opt_a.cse                        :     0.000039s :  0.45%
    jit_opt_a.replace_applicator         :     0.000019s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.15%
    rewriter_after_opt_a                 :     0.000037s :  0.42%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000714s :  8.10%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.19%
    cconv                                :     0.000029s :  0.33%
    loop_unroll                          :     0.000438s :  4.96%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000017s :  0.19%
    partial_unused_args_eliminate        :     0.000003s :  0.04%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000052s :  0.59%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000481s :  5.46%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000042s :  0.47%


Time group info:
------[substitution.]   0.000232    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.60% :     0.000006s :      3: substitution.graph_param_transform
84.65% :     0.000196s :      4: substitution.inline
 1.83% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.35% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.97% :     0.000005s :      2: substitution.replace_old_param
 5.22% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004333     2
81.03% :     0.003511s :      1: type_inference.infer
18.97% :     0.000822s :      1: type_inference.specialize
------[replace.]   0.000071     6
67.60% :     0.000048s :      4: replace.inline
32.40% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000204     6
94.74% :     0.000193s :      4: match.inline
 5.26% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.00% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 2.05% :     0.000003s :     10: predicate.arithmetic_simplify
 1.23% :     0.000002s :     10: predicate.cast_eliminate
 0.55% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.61% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.31% :     0.000002s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.66% :     0.000007s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.84% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.79% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 2.03% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.36% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.84% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.25% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.48% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.00% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.33% :     0.000010s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.30% :     0.000002s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.17% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.12% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000600    12
49.28% :     0.000296s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.21% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.51% :     0.000255s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014995    72
 0.37% :     0.000056s :      1: add_recomputation
 0.46% :     0.000070s :      1: auto_monad
 0.13% :     0.000019s :      1: auto_monad_reorder
 2.03% :     0.000305s :      1: bootstrap
 0.21% :     0.000032s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
40.00% :     0.005998s :      1: jit_opt_a
 1.02% :     0.000152s :      1: jit_opt_after_cconv
 0.38% :     0.000057s :      1: jit_opt_b
 2.97% :     0.000446s :      1: loop_unroll
 4.84% :     0.000726s :      1: mutable_eliminate
 5.98% :     0.000897s :     26: opt.transform.jit_opt_a
 0.33% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000027s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.26% :     0.000490s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.04% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000038s :      1: py_interpret_to_execute
 0.10% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000019s :      1: remove_dup_value
 2.95% :     0.000443s :      1: renormalize.infer
 2.33% :     0.000350s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000041s :      1: rewriter_after_opt_a
 0.53% :     0.000080s :      1: rewriter_before_opt_a
 0.54% :     0.000080s :      1: symbol_engine_optimizer
29.46% :     0.004417s :      1: type_inference


TotalTime = 0.0191127, [33]
[bootstrap]: 0.00027998
[type_inference]: 0.00464198
[event_method]: 1.74e-05
[auto_monad]: 6.52e-05
[graph_reusing]: 5.71e-06
[pre_auto_parallel]: 2.27999e-06
[py_interpret_to_execute]: 3.176e-05
[rewriter_before_opt_a]: 7.107e-05
[expand_dump_flag]: 3.78001e-06
[jit_opt_a]: 0.00569795, [2]
    [Cycle 1]: 0.00146867, [27]
        [switch_simplify]: 5.103e-05
        [loop_unroll]: 2.936e-05
        [a_1]: 0.00050934
        [with_stream_mark]: 1.48e-05
        [recompute_prepare]: 7.37002e-06
        [updatestate_depend_eliminate]: 3.28998e-06
        [updatestate_assign_eliminate]: 3.48e-06
        [updatestate_loads_eliminate]: 2.59001e-06
        [parameter_eliminate]: 1.88997e-06
        [specialize_transform]: 5.97999e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 2.41e-06
        [get_grad_eliminate_]: 5.15001e-06
        [merge_forward]: 3.38999e-06
        [cell_reuse_recompute_pass]: 1.44998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.383e-05
        [j_node_and_user_rematch]: 8.73001e-06
        [meta_fg_expand]: 2.41e-06
        [replace_old_param]: 8.40001e-06
        [inline_without_move]: 5.22e-06
        [renormalize]: 0.0005403
        [add_forward_monad_depend]: 5.77001e-06
        [auto_monad_grad]: 1.78002e-06
        [auto_monad_eliminator]: 1.396e-05
        [cse]: 2.965e-05
        [replace_applicator]: 1.261e-05
    [Cycle 2]: 0.00031883, [27]
        [switch_simplify]: 5.49e-06
        [loop_unroll]: 4.89998e-06
        [a_1]: 8.911e-05
        [with_stream_mark]: 1.012e-05
        [recompute_prepare]: 5.27001e-06
        [updatestate_depend_eliminate]: 2.54001e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 5.19003e-06
        [updatestate_useless_node_eliminater]: 4.71002e-06
        [accelerated_algorithm]: 5.09003e-06
        [meta_shard_fg_expand]: 1.02e-06
        [get_grad_eliminate_]: 4.53001e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 1.58002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.224e-05
        [j_node_and_user_rematch]: 7.51999e-06
        [meta_fg_expand]: 1.55999e-06
        [replace_old_param]: 7.43e-06
        [inline_without_move]: 4.80001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 5.44e-06
        [cse]: 1.074e-05
        [replace_applicator]: 4.80999e-06
[py_interpret_to_execute_after_opt_a]: 1.029e-05
[rewriter_after_opt_a]: 3.246e-05
[convert_after_rewriter]: 7.23e-06
[order_py_execute_after_rewriter]: 5.92001e-06
[mutable_eliminate]: 0.00056289
[jit_opt_b]: 5.036e-05, [1]
    [Cycle 1]: 4.33e-05, [2]
        [frontend_op_eliminate]: 1.687e-05
        [inline_after_opt_a]: 1.43e-05
[cconv]: 2.555e-05
[loop_unroll]: 0.00042512
[jit_opt_after_cconv]: 0.00013725, [1]
    [Cycle 1]: 0.00013095, [11]
        [c_1]: 2.013e-05
        [parameter_eliminate]: 2.93e-06
        [updatestate_depend_eliminate]: 5.40999e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.789e-05
        [call_graph_tuple_transform]: 1.725e-05
        [tuple_list_get_item_eliminator]: 5.02999e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.21998e-06
[remove_dup_value]: 1.331e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 5.19e-06
[add_recomputation]: 4.897e-05
[cse_after_recomputation]: 2.117e-05, [1]
    [Cycle 1]: 1.566e-05, [1]
        [cse]: 9.49e-06
[auto_monad_reorder]: 1.593e-05
[get_jit_bprop_graph]: 1.50999e-06
[rewriter_after_jit_bprop_graph]: 5.24e-06
[opt_after_jit_grad]: 0.00043341
[symbol_engine_optimizer]: 7.227e-05, [1]
    [Cycle 1]: 6.617e-05, [6]
        [build]: 3.46999e-06
        [elim_shapecalc]: 7.21999e-06
        [elim_not_effective]: 1.324e-05
        [opt_reshape]: 5.30999e-06
        [fold_const_symbol]: 8.12e-06
        [renormalize]: 5.8001e-07
[validate]: 3.395e-05
[backend_pass]: 1.10001e-06
[task_emit]: 0.0061851
[execute]: 7.13e-06
Sums
    bootstrap                            :     0.000280s :  1.92%
    type_inference                       :     0.004642s : 31.80%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000065s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.22%
    rewriter_before_opt_a                :     0.000071s :  0.49%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000057s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000598s :  4.10%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000540s :  3.70%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000040s :  0.28%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000563s :  3.86%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000026s :  0.18%
    loop_unroll                          :     0.000425s :  2.91%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000049s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000016s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000433s :  2.97%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006185s : 42.37%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000178    23
 1.14% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.85% :     0.000005s :      3: substitution.graph_param_transform
82.54% :     0.000147s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.33% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.34% :     0.000004s :      2: substitution.replace_old_param
 6.55% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004569     2
80.78% :     0.003691s :      1: type_inference.infer
19.22% :     0.000878s :      1: type_inference.specialize
------[replace.]   0.000065     6
67.44% :     0.000044s :      4: replace.inline
32.56% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000155     6
93.22% :     0.000144s :      4: match.inline
 6.78% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.32% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.39% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000001s :     10: predicate.addn_zero_filter
 2.01% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.57% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.52% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.15% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.16% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.52% :     0.000002s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.98% :     0.000002s :     15: predicate.load_eliminater
 1.38% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.69% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     15: predicate.partial_eliminate
 1.30% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.77% :     0.000002s :     10: predicate.reduce_eliminate
 1.62% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.56% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.54% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.33% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.42% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.60% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.87% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000548    12
47.51% :     0.000260s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.49% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.00% :     0.000263s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020560    76
 0.25% :     0.000052s :      1: add_recomputation
 0.33% :     0.000068s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.42% :     0.000291s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.73% :     0.005701s :      1: jit_opt_a
 0.68% :     0.000140s :      1: jit_opt_after_cconv
 0.26% :     0.000053s :      1: jit_opt_b
 2.11% :     0.000433s :      1: loop_unroll
 2.78% :     0.000572s :      1: mutable_eliminate
 3.90% :     0.000801s :     26: opt.transform.jit_opt_a
 0.22% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.15% :     0.000441s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.14% :     0.000029s :      1: remove_dup_value
 1.34% :     0.000275s :      1: renormalize.infer
 1.26% :     0.000258s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000035s :      1: rewriter_after_opt_a
 0.36% :     0.000074s :      1: rewriter_before_opt_a
 0.37% :     0.000075s :      1: symbol_engine_optimizer
30.12% :     0.006194s :      1: task_emit
22.64% :     0.004655s :      1: type_inference
 0.26% :     0.000054s :      1: validate


TotalTime = 0.0325831, [33]
[bootstrap]: 0.0002316
[type_inference]: 0.0125929
[event_method]: 5.65e-05
[auto_monad]: 9.426e-05
[graph_reusing]: 6.91001e-06
[pre_auto_parallel]: 2.37001e-06
[py_interpret_to_execute]: 2.711e-05
[rewriter_before_opt_a]: 9.732e-05
[expand_dump_flag]: 3.10998e-06
[jit_opt_a]: 0.0109945, [3]
    [Cycle 1]: 0.00544044, [27]
        [switch_simplify]: 0.00010539
        [loop_unroll]: 4.332e-05
        [a_1]: 0.00085138
        [with_stream_mark]: 2.232e-05
        [recompute_prepare]: 1.752e-05
        [updatestate_depend_eliminate]: 7.28999e-06
        [updatestate_assign_eliminate]: 5.52999e-06
        [updatestate_loads_eliminate]: 5.44e-06
        [parameter_eliminate]: 1.97999e-06
        [specialize_transform]: 1.294e-05
        [updatestate_useless_node_eliminater]: 1.22e-05
        [accelerated_algorithm]: 1.21e-05
        [meta_shard_fg_expand]: 2.77002e-06
        [get_grad_eliminate_]: 1.168e-05
        [merge_forward]: 7.4e-06
        [cell_reuse_recompute_pass]: 9.20001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.401e-05
        [j_node_and_user_rematch]: 2.325e-05
        [meta_fg_expand]: 0.00139876
        [replace_old_param]: 4.975e-05
        [inline_without_move]: 5.077e-05
        [renormalize]: 0.00232305
        [add_forward_monad_depend]: 9.20999e-06
        [auto_monad_grad]: 5.71e-06
        [auto_monad_eliminator]: 4.457e-05
        [cse]: 0.00013854
        [replace_applicator]: 6.278e-05
    [Cycle 2]: 0.00202662, [27]
        [switch_simplify]: 3.626e-05
        [loop_unroll]: 3.394e-05
        [a_1]: 0.00101075
        [with_stream_mark]: 1.177e-05
        [recompute_prepare]: 7.95e-06
        [updatestate_depend_eliminate]: 3.81001e-06
        [updatestate_assign_eliminate]: 3.06999e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 7.26999e-06
        [updatestate_useless_node_eliminater]: 6.82002e-06
        [accelerated_algorithm]: 6.00002e-06
        [meta_shard_fg_expand]: 1.73002e-06
        [get_grad_eliminate_]: 5.66e-06
        [merge_forward]: 3.18998e-06
        [cell_reuse_recompute_pass]: 8.2e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.286e-05
        [j_node_and_user_rematch]: 1.066e-05
        [meta_fg_expand]: 7.202e-05
        [replace_old_param]: 1.019e-05
        [inline_without_move]: 5.97999e-06
        [renormalize]: 0.00056958
        [add_forward_monad_depend]: 4.57e-06
        [auto_monad_grad]: 1.26002e-06
        [auto_monad_eliminator]: 1.187e-05
        [cse]: 2.132e-05
        [replace_applicator]: 1.24e-05
    [Cycle 3]: 0.00038074, [27]
        [switch_simplify]: 6.79001e-06
        [loop_unroll]: 5.98002e-06
        [a_1]: 0.0001194
        [with_stream_mark]: 9.15999e-06
        [recompute_prepare]: 5.99999e-06
        [updatestate_depend_eliminate]: 3.55e-06
        [updatestate_assign_eliminate]: 3.05998e-06
        [updatestate_loads_eliminate]: 3.11001e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 6.51e-06
        [updatestate_useless_node_eliminater]: 6.46e-06
        [accelerated_algorithm]: 6.19001e-06
        [meta_shard_fg_expand]: 1.40001e-06
        [get_grad_eliminate_]: 5.87999e-06
        [merge_forward]: 3.08998e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.721e-05
        [j_node_and_user_rematch]: 1.059e-05
        [meta_fg_expand]: 1.92001e-06
        [replace_old_param]: 8.13001e-06
        [inline_without_move]: 5.80002e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 6.93998e-06
        [cse]: 1.502e-05
        [replace_applicator]: 6.33998e-06
[py_interpret_to_execute_after_opt_a]: 1.246e-05
[rewriter_after_opt_a]: 3.812e-05
[convert_after_rewriter]: 7.46999e-06
[order_py_execute_after_rewriter]: 6.24001e-06
[mutable_eliminate]: 0.00053938
[jit_opt_b]: 5.464e-05, [1]
    [Cycle 1]: 4.796e-05, [2]
        [frontend_op_eliminate]: 1.898e-05
        [inline_after_opt_a]: 1.84e-05
[cconv]: 2.073e-05
[loop_unroll]: 0.00041611
[jit_opt_after_cconv]: 0.0001528, [1]
    [Cycle 1]: 0.00014681, [11]
        [c_1]: 2.536e-05
        [parameter_eliminate]: 2.43998e-06
        [updatestate_depend_eliminate]: 5.77999e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.66999e-06
        [cse]: 2.147e-05
        [call_graph_tuple_transform]: 2.092e-05
        [tuple_list_get_item_eliminator]: 6.32001e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 6.86999e-06
[remove_dup_value]: 1.375e-05
[partial_unused_args_eliminate]: 2.19001e-06
[environ_conv]: 4.72e-06
[add_recomputation]: 4.369e-05
[cse_after_recomputation]: 2.681e-05, [1]
    [Cycle 1]: 2.145e-05, [1]
        [cse]: 1.536e-05
[auto_monad_reorder]: 1.871e-05
[get_jit_bprop_graph]: 1.28002e-06
[rewriter_after_jit_bprop_graph]: 4.61002e-06
[opt_after_jit_grad]: 0.00046054
[symbol_engine_optimizer]: 7.907e-05, [1]
    [Cycle 1]: 7.292e-05, [6]
        [build]: 3.62002e-06
        [elim_shapecalc]: 9.11998e-06
        [elim_not_effective]: 1.452e-05
        [opt_reshape]: 7.01999e-06
        [fold_const_symbol]: 1.065e-05
        [renormalize]: 4.10015e-07
[validate]: 3.409e-05
[backend_pass]: 7.30011e-07
[task_emit]: 0.00633651
[execute]: 6.58e-06
Sums
    bootstrap                            :     0.000232s :  0.81%
    type_inference                       :     0.012593s : 43.93%
    event_method                         :     0.000056s :  0.20%
    auto_monad                           :     0.000094s :  0.33%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000027s :  0.09%
    rewriter_before_opt_a                :     0.000097s :  0.34%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000148s :  0.52%
    jit_opt_a.loop_unroll                :     0.000083s :  0.29%
    jit_opt_a.a_1                        :     0.001982s :  6.91%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000031s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000027s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.08%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000054s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000044s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001473s :  5.14%
    jit_opt_a.replace_old_param          :     0.000068s :  0.24%
    jit_opt_a.inline_without_move        :     0.000063s :  0.22%
    jit_opt_a.renormalize                :     0.002893s : 10.09%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000063s :  0.22%
    jit_opt_a.cse                        :     0.000175s :  0.61%
    jit_opt_a.replace_applicator         :     0.000082s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000038s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000539s :  1.88%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000021s :  0.07%
    loop_unroll                          :     0.000416s :  1.45%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000014s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000044s :  0.15%
    cse_after_recomputation.cse          :     0.000015s :  0.05%
    auto_monad_reorder                   :     0.000019s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000461s :  1.61%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006337s : 22.11%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000457   113
 0.45% :     0.000002s :      3: substitution.elim_not_effective
 0.35% :     0.000002s :      3: substitution.fold_const_symbol
 1.17% :     0.000005s :      4: substitution.graph_param_transform
62.41% :     0.000285s :     14: substitution.inline
 3.54% :     0.000016s :      2: substitution.inline_without_move
 2.08% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.17% :     0.000010s :      7: substitution.minmaximum_grad
 1.39% :     0.000006s :      5: substitution.partial_eliminate
 2.29% :     0.000010s :     14: substitution.remove_not_recompute_node
 5.30% :     0.000024s :     10: substitution.replace_applicator
 1.44% :     0.000007s :      7: substitution.replace_old_param
 0.46% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.74% :     0.000008s :      1: substitution.switch_simplify
 4.20% :     0.000019s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.09% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.90% :     0.000036s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012523     2
90.20% :     0.011296s :      1: type_inference.infer
 9.80% :     0.001228s :      1: type_inference.specialize
------[replace.]   0.000167    22
55.40% :     0.000092s :     14: replace.inline
15.37% :     0.000026s :      1: replace.switch_simplify
29.23% :     0.000049s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000303    22
91.52% :     0.000278s :     14: match.inline
 2.43% :     0.000007s :      1: match.switch_simplify
 6.05% :     0.000018s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000380  2634
 1.54% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.46% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.42% :     0.000005s :     44: predicate.addn_check_dump
 1.74% :     0.000007s :     44: predicate.addn_zero_filter
 2.03% :     0.000008s :     44: predicate.arithmetic_simplify
 1.70% :     0.000006s :     44: predicate.cast_eliminate
 0.19% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.40% :     0.000005s :     44: predicate.compare_switch_simplify
 1.57% :     0.000006s :     44: predicate.depend_value_elim
 1.53% :     0.000006s :     44: predicate.dict_get_item_const_eliminator
 1.53% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.61% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.35% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      4: predicate.elim_not_effective
 0.22% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.48% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.42% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.44% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.57% :     0.000006s :     44: predicate.environ_get_eliminate
 1.42% :     0.000005s :     44: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.84% :     0.000003s :     19: predicate.get_grad_eliminate
 0.13% :     0.000000s :      4: predicate.graph_param_transform
 4.23% :     0.000016s :     73: predicate.inline
 1.68% :     0.000006s :     40: predicate.inline_without_move
 0.35% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.90% :     0.000003s :     19: predicate.less_batch_normalization
 1.77% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.90% :     0.000007s :     55: predicate.load_eliminater
 0.49% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.43% :     0.000013s :     91: predicate.loop_unroll_before_grad
 1.88% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.48% :     0.000006s :     44: predicate.merge_addn
 1.49% :     0.000006s :     44: predicate.minmaximum_grad
 0.51% :     0.000002s :      4: predicate.mutable_eliminate
 0.23% :     0.000001s :      4: predicate.opt_reshape
 2.30% :     0.000009s :     55: predicate.partial_eliminate
 1.56% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.99% :     0.000008s :     44: predicate.reduce_eliminate
 1.78% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.43% :     0.000009s :     97: predicate.replace_applicator
 0.95% :     0.000004s :     40: predicate.replace_old_param
 0.10% :     0.000000s :      4: predicate.reset_defer_inline
 1.76% :     0.000007s :     44: predicate.reshape_eliminate
 1.71% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.39% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.63% :     0.000006s :     44: predicate.same_eliminate
 0.47% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.36% :     0.000001s :      8: predicate.special_op_eliminate
 0.87% :     0.000003s :     19: predicate.specialize_transform
 1.69% :     0.000006s :     44: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.79% :     0.000011s :     65: predicate.switch_defer_inline
 2.58% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.46% :     0.000025s :    162: predicate.switch_simplify
 1.47% :     0.000006s :     44: predicate.tile_eliminate
 1.46% :     0.000006s :     44: predicate.transpose_eliminate
 1.76% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.69% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 2.97% :     0.000011s :     59: predicate.tuple_list_get_item_eliminator
 1.95% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.88% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000007s :     44: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.23% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001574    32
63.12% :     0.000993s :     14: func_graph_cloner_run.FuncGraphClonerGraph
36.88% :     0.000580s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038217    91
 0.12% :     0.000047s :      1: add_recomputation
 0.26% :     0.000098s :      1: auto_monad
 0.06% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.66% :     0.000253s :      1: bootstrap
 0.06% :     0.000023s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000029s :      1: cse_after_recomputation
 0.03% :     0.000013s :      1: environ_conv
 0.16% :     0.000060s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
28.78% :     0.010998s :      1: jit_opt_a
 0.41% :     0.000155s :      1: jit_opt_after_cconv
 0.15% :     0.000058s :      1: jit_opt_b
 1.11% :     0.000423s :      1: loop_unroll
 1.43% :     0.000548s :      1: mutable_eliminate
 6.84% :     0.002614s :     39: opt.transform.jit_opt_a
 0.15% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000031s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.23% :     0.000469s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.08% :     0.000030s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000016s :      1: remove_dup_value
 4.21% :     0.001609s :      2: renormalize.infer
 3.32% :     0.001269s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000041s :      1: rewriter_after_opt_a
 0.26% :     0.000100s :      1: rewriter_before_opt_a
 0.21% :     0.000082s :      1: symbol_engine_optimizer
16.60% :     0.006346s :      1: task_emit
32.98% :     0.012602s :      1: type_inference
 0.14% :     0.000054s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.219.31 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dcafe50> should be wrapped by jit only once.

TotalTime = 0.0277939, [33]
[bootstrap]: 0.00456447
[type_inference]: 0.00588253
[event_method]: 1.879e-05
[auto_monad]: 6.661e-05
[graph_reusing]: 7.00998e-06
[pre_auto_parallel]: 2.53003e-06
[py_interpret_to_execute]: 3.902e-05
[rewriter_before_opt_a]: 7.445e-05
[expand_dump_flag]: 3.00002e-06
[jit_opt_a]: 0.00705287, [2]
    [Cycle 1]: 0.00296761, [27]
        [switch_simplify]: 5.928e-05
        [loop_unroll]: 2.936e-05
        [a_1]: 0.00056804
        [with_stream_mark]: 1.933e-05
        [recompute_prepare]: 6.81999e-06
        [updatestate_depend_eliminate]: 3.8e-06
        [updatestate_assign_eliminate]: 3.33e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 4.99003e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 2.02999e-06
        [get_grad_eliminate_]: 5.16998e-06
        [merge_forward]: 3.73001e-06
        [cell_reuse_recompute_pass]: 1.18001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 3.161e-05
        [j_node_and_user_rematch]: 1.214e-05
        [meta_fg_expand]: 4.21001e-06
        [replace_old_param]: 1.113e-05
        [inline_without_move]: 5.93002e-06
        [renormalize]: 0.00097073
        [add_forward_monad_depend]: 7.83001e-06
        [auto_monad_grad]: 2.86e-06
        [auto_monad_eliminator]: 1.72e-05
        [cse]: 3.036e-05
        [replace_applicator]: 1.773e-05
    [Cycle 2]: 0.00040316, [27]
        [switch_simplify]: 6.81001e-06
        [loop_unroll]: 5.34e-06
        [a_1]: 0.00010314
        [with_stream_mark]: 1.482e-05
        [recompute_prepare]: 6.15002e-06
        [updatestate_depend_eliminate]: 4.1e-06
        [updatestate_assign_eliminate]: 2.83998e-06
        [updatestate_loads_eliminate]: 2.69999e-06
        [parameter_eliminate]: 2.43e-06
        [specialize_transform]: 4.81997e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 5.86e-06
        [meta_shard_fg_expand]: 2.29001e-06
        [get_grad_eliminate_]: 4.76002e-06
        [merge_forward]: 4.63999e-06
        [cell_reuse_recompute_pass]: 3.21999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.802e-05
        [j_node_and_user_rematch]: 8.71002e-06
        [meta_fg_expand]: 1.92001e-06
        [replace_old_param]: 9.22001e-06
        [inline_without_move]: 5.02999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.52001e-06
        [auto_monad_grad]: 1.53002e-06
        [auto_monad_eliminator]: 7.09001e-06
        [cse]: 1.57e-05
        [replace_applicator]: 7.72998e-06
[py_interpret_to_execute_after_opt_a]: 1.641e-05
[rewriter_after_opt_a]: 4.08e-05
[convert_after_rewriter]: 7.61999e-06
[order_py_execute_after_rewriter]: 5.46e-06
[mutable_eliminate]: 0.00084396
[jit_opt_b]: 6.3e-05, [1]
    [Cycle 1]: 5.187e-05, [2]
        [frontend_op_eliminate]: 2.176e-05
        [inline_after_opt_a]: 1.779e-05
[cconv]: 3.812e-05
[loop_unroll]: 0.00055912
[jit_opt_after_cconv]: 0.00017677, [1]
    [Cycle 1]: 0.0001689, [11]
        [c_1]: 2.219e-05
        [parameter_eliminate]: 4.37e-06
        [updatestate_depend_eliminate]: 7.73999e-06
        [updatestate_assign_eliminate]: 3.5e-06
        [updatestate_loads_eliminate]: 2.74001e-06
        [cse]: 3.528e-05
        [call_graph_tuple_transform]: 2.514e-05
        [tuple_list_get_item_eliminator]: 6.16e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 1.02e-06
        [switch_simplify]: 6.14001e-06
[remove_dup_value]: 1.685e-05
[partial_unused_args_eliminate]: 2.63e-06
[environ_conv]: 6.49001e-06
[add_recomputation]: 5.646e-05
[cse_after_recomputation]: 2.394e-05, [1]
    [Cycle 1]: 1.733e-05, [1]
        [cse]: 1.047e-05
[auto_monad_reorder]: 1.687e-05
[get_jit_bprop_graph]: 2.29999e-06
[rewriter_after_jit_bprop_graph]: 1.023e-05
[opt_after_jit_grad]: 0.00054114
[symbol_engine_optimizer]: 8.302e-05, [1]
    [Cycle 1]: 7.427e-05, [6]
        [build]: 4.83001e-06
        [elim_shapecalc]: 9.49999e-06
        [elim_not_effective]: 1.466e-05
        [opt_reshape]: 7.25e-06
        [fold_const_symbol]: 9.07999e-06
        [renormalize]: 5.50004e-07
[validate]: 4.761e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00725434
[execute]: 8.25999e-06
Sums
    bootstrap                            :     0.004564s : 20.34%
    type_inference                       :     0.005883s : 26.22%
    event_method                         :     0.000019s :  0.08%
    auto_monad                           :     0.000067s :  0.30%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.17%
    rewriter_before_opt_a                :     0.000074s :  0.33%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000066s :  0.29%
    jit_opt_a.loop_unroll                :     0.000035s :  0.15%
    jit_opt_a.a_1                        :     0.000671s :  2.99%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.04%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000050s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.03%
    jit_opt_a.replace_old_param          :     0.000020s :  0.09%
    jit_opt_a.inline_without_move        :     0.000011s :  0.05%
    jit_opt_a.renormalize                :     0.000971s :  4.33%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.11%
    jit_opt_a.cse                        :     0.000046s :  0.21%
    jit_opt_a.replace_applicator         :     0.000025s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.07%
    rewriter_after_opt_a                 :     0.000041s :  0.18%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000844s :  3.76%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.08%
    cconv                                :     0.000038s :  0.17%
    loop_unroll                          :     0.000559s :  2.49%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000035s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000056s :  0.25%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.05%
    opt_after_jit_grad                   :     0.000541s :  2.41%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000048s :  0.21%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007254s : 32.33%
    execute                              :     0.000008s :  0.04%

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.224.60 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

Time group info:
------[substitution.]   0.000238    23
 0.77% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000002s :      2: substitution.fold_const_symbol
 2.95% :     0.000007s :      3: substitution.graph_param_transform
81.80% :     0.000195s :      4: substitution.inline
 2.08% :     0.000005s :      4: substitution.j_node_and_user_rematch
 3.51% :     0.000008s :      4: substitution.remove_not_recompute_node
 2.75% :     0.000007s :      2: substitution.replace_old_param
 5.48% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005803     2
82.48% :     0.004786s :      1: type_inference.infer
17.52% :     0.001017s :      1: type_inference.specialize
------[replace.]   0.000072     6
67.62% :     0.000048s :      4: replace.inline
32.38% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000204     6
94.10% :     0.000192s :      4: match.inline
 5.90% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.63% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.66% :     0.000002s :     10: predicate.addn_zero_filter
 2.25% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000002s :     10: predicate.depend_value_elim
 1.40% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.04% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.76% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     10: predicate.environ_get_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.51% :     0.000007s :     22: predicate.inline
 1.02% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.96% :     0.000001s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 1.85% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.62% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 2.17% :     0.000003s :      3: predicate.mutable_eliminate
 0.67% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000003s :     15: predicate.partial_eliminate
 1.56% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     10: predicate.reduce_eliminate
 1.56% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.91% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.82% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.22% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.20% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 1.90% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.91% :     0.000010s :     53: predicate.switch_simplify
 1.01% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.51% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.17% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000783    12
50.73% :     0.000397s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.57% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.69% :     0.000342s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.029810    76
 0.20% :     0.000060s :      1: add_recomputation
 0.23% :     0.000070s :      1: auto_monad
 0.06% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
15.40% :     0.004591s :      1: bootstrap
 0.14% :     0.000041s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000026s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.08% :     0.000024s :      1: event_method
 0.04% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
23.68% :     0.007058s :      1: jit_opt_a
 0.60% :     0.000180s :      1: jit_opt_after_cconv
 0.22% :     0.000066s :      1: jit_opt_b
 1.91% :     0.000571s :      1: loop_unroll
 2.89% :     0.000860s :      1: mutable_eliminate
 3.08% :     0.000918s :     26: opt.transform.jit_opt_a
 0.19% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000032s :      4: opt.transform.jit_opt_b
 0.05% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000030s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.86% :     0.000553s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.14% :     0.000042s :      1: py_interpret_to_execute
 0.07% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 1.61% :     0.000479s :      1: renormalize.infer
 1.62% :     0.000482s :      1: renormalize.specialize
 0.04% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000045s :      1: rewriter_after_opt_a
 0.26% :     0.000078s :      1: rewriter_before_opt_a
 0.29% :     0.000086s :      1: symbol_engine_optimizer
24.39% :     0.007271s :      1: task_emit
19.79% :     0.005900s :      1: type_inference
 0.23% :     0.000070s :      1: validate


TotalTime = 0.0135222, [30]
[bootstrap]: 0.00035832
[type_inference]: 0.0046876
[event_method]: 1.842e-05
[auto_monad]: 7.039e-05
[graph_reusing]: 5.77001e-06
[pre_auto_parallel]: 2.53e-06
[py_interpret_to_execute]: 3.643e-05
[rewriter_before_opt_a]: 7.596e-05
[expand_dump_flag]: 3.46999e-06
[jit_opt_a]: 0.00597072, [2]
    [Cycle 1]: 0.00172075, [27]
        [switch_simplify]: 5.907e-05
        [loop_unroll]: 3.049e-05
        [a_1]: 0.00054529
        [with_stream_mark]: 2.002e-05
        [recompute_prepare]: 7.88001e-06
        [updatestate_depend_eliminate]: 3.76999e-06
        [updatestate_assign_eliminate]: 3.09001e-06
        [updatestate_loads_eliminate]: 2.93998e-06
        [parameter_eliminate]: 1.88002e-06
        [specialize_transform]: 5.80002e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.64e-06
        [meta_shard_fg_expand]: 2.11e-06
        [get_grad_eliminate_]: 5.27999e-06
        [merge_forward]: 4.00998e-06
        [cell_reuse_recompute_pass]: 1.38002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.522e-05
        [j_node_and_user_rematch]: 8.92e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 9.23002e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.00072844
        [add_forward_monad_depend]: 6.50997e-06
        [auto_monad_grad]: 2.68003e-06
        [auto_monad_eliminator]: 1.623e-05
        [cse]: 2.889e-05
        [replace_applicator]: 1.316e-05
    [Cycle 2]: 0.00033579, [27]
        [switch_simplify]: 5.30999e-06
        [loop_unroll]: 5.17e-06
        [a_1]: 9.225e-05
        [with_stream_mark]: 1.207e-05
        [recompute_prepare]: 5.06002e-06
        [updatestate_depend_eliminate]: 2.69001e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.57999e-06
        [specialize_transform]: 5.44e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 5.16002e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 4.52998e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 2.14999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.335e-05
        [j_node_and_user_rematch]: 8.2e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 8.36002e-06
        [inline_without_move]: 4.82e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.41998e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 4.95001e-06
        [cse]: 1.068e-05
        [replace_applicator]: 5.51002e-06
[py_interpret_to_execute_after_opt_a]: 1.133e-05
[rewriter_after_opt_a]: 3.508e-05
[convert_after_rewriter]: 6.81001e-06
[order_py_execute_after_rewriter]: 4.75999e-06
[mutable_eliminate]: 0.00066959
[jit_opt_b]: 5.273e-05, [1]
    [Cycle 1]: 4.542e-05, [2]
        [frontend_op_eliminate]: 1.628e-05
        [inline_after_opt_a]: 1.622e-05
[cconv]: 2.751e-05
[loop_unroll]: 0.00043352
[jit_opt_after_cconv]: 0.00014406, [1]
    [Cycle 1]: 0.00013676, [11]
        [c_1]: 2.031e-05
        [parameter_eliminate]: 3.21001e-06
        [updatestate_depend_eliminate]: 5.25001e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 1.982e-05
        [call_graph_tuple_transform]: 2.021e-05
        [tuple_list_get_item_eliminator]: 6.24001e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 6.49976e-07
        [switch_simplify]: 5.52999e-06
[remove_dup_value]: 1.444e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 5.40999e-06
[add_recomputation]: 4.869e-05
[cse_after_recomputation]: 2.181e-05, [1]
    [Cycle 1]: 1.57e-05, [1]
        [cse]: 9.60001e-06
[auto_monad_reorder]: 1.506e-05
[get_jit_bprop_graph]: 1.84e-06
[rewriter_after_jit_bprop_graph]: 5.44e-06
[opt_after_jit_grad]: 0.00050136
[symbol_engine_optimizer]: 7.585e-05, [1]
    [Cycle 1]: 6.871e-05, [6]
        [build]: 4.15e-06
        [elim_shapecalc]: 7.95e-06
        [elim_not_effective]: 1.336e-05
        [opt_reshape]: 5.72001e-06
        [fold_const_symbol]: 9.02e-06
        [renormalize]: 1.02e-06
[validate]: 3.506e-05
Sums
    bootstrap                            :     0.000358s :  3.98%
    type_inference                       :     0.004688s : 52.06%
    event_method                         :     0.000018s :  0.20%
    auto_monad                           :     0.000070s :  0.78%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.40%
    rewriter_before_opt_a                :     0.000076s :  0.84%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000064s :  0.72%
    jit_opt_a.loop_unroll                :     0.000036s :  0.40%
    jit_opt_a.a_1                        :     0.000638s :  7.08%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.36%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000729s :  8.09%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.24%
    jit_opt_a.cse                        :     0.000040s :  0.44%
    jit_opt_a.replace_applicator         :     0.000019s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.13%
    rewriter_after_opt_a                 :     0.000035s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000670s :  7.44%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.18%
    cconv                                :     0.000028s :  0.31%
    loop_unroll                          :     0.000434s :  4.82%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000014s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000049s :  0.54%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000501s :  5.57%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000035s :  0.39%


Time group info:
------[substitution.]   0.000212    23
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.69% :     0.000006s :      3: substitution.graph_param_transform
83.23% :     0.000176s :      4: substitution.inline
 1.68% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.39% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.40% :     0.000005s :      2: substitution.replace_old_param
 6.00% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004614     2
81.36% :     0.003754s :      1: type_inference.infer
18.64% :     0.000860s :      1: type_inference.specialize
------[replace.]   0.000072     6
65.53% :     0.000047s :      4: replace.inline
34.47% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000185     6
93.77% :     0.000174s :      4: match.inline
 6.23% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.47% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 2.23% :     0.000003s :     10: predicate.arithmetic_simplify
 1.62% :     0.000002s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000001s :     10: predicate.depend_value_elim
 1.23% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.28% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.66% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.80% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.45% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.90% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.53% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000001s :     10: predicate.same_eliminate
 0.69% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.55% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 2.01% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.64% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.22% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.51% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000650    12
54.54% :     0.000354s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.14% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.32% :     0.000262s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015162    72
 0.34% :     0.000052s :      1: add_recomputation
 0.49% :     0.000074s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.45% :     0.000372s :      1: bootstrap
 0.20% :     0.000030s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.40% :     0.005974s :      1: jit_opt_a
 0.97% :     0.000148s :      1: jit_opt_after_cconv
 0.37% :     0.000056s :      1: jit_opt_b
 2.91% :     0.000442s :      1: loop_unroll
 4.48% :     0.000679s :      1: mutable_eliminate
 5.63% :     0.000854s :     26: opt.transform.jit_opt_a
 0.32% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.38% :     0.000512s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000040s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 2.15% :     0.000326s :      1: renormalize.infer
 2.60% :     0.000394s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000038s :      1: rewriter_after_opt_a
 0.52% :     0.000079s :      1: rewriter_before_opt_a
 0.52% :     0.000079s :      1: symbol_engine_optimizer
31.03% :     0.004704s :      1: type_inference


TotalTime = 0.0179104, [33]
[bootstrap]: 0.00032179
[type_inference]: 0.00442309
[event_method]: 1.822e-05
[auto_monad]: 6.333e-05
[graph_reusing]: 5.34e-06
[pre_auto_parallel]: 2.43002e-06
[py_interpret_to_execute]: 3.054e-05
[rewriter_before_opt_a]: 7.074e-05
[expand_dump_flag]: 2.427e-05
[jit_opt_a]: 0.00508031, [2]
    [Cycle 1]: 0.00140902, [27]
        [switch_simplify]: 4.912e-05
        [loop_unroll]: 3.01e-05
        [a_1]: 0.0005184
        [with_stream_mark]: 1.507e-05
        [recompute_prepare]: 7.26001e-06
        [updatestate_depend_eliminate]: 3.37997e-06
        [updatestate_assign_eliminate]: 2.78998e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 5.76998e-06
        [updatestate_useless_node_eliminater]: 4.92999e-06
        [accelerated_algorithm]: 5.15999e-06
        [meta_shard_fg_expand]: 2.11e-06
        [get_grad_eliminate_]: 4.90999e-06
        [merge_forward]: 3.39001e-06
        [cell_reuse_recompute_pass]: 1.11002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.379e-05
        [j_node_and_user_rematch]: 8.32998e-06
        [meta_fg_expand]: 2.32001e-06
        [replace_old_param]: 7.82e-06
        [inline_without_move]: 4.91002e-06
        [renormalize]: 0.00050188
        [add_forward_monad_depend]: 5.07999e-06
        [auto_monad_grad]: 1.97001e-06
        [auto_monad_eliminator]: 1.226e-05
        [cse]: 2.04e-05
        [replace_applicator]: 1.091e-05
    [Cycle 2]: 0.00031125, [27]
        [switch_simplify]: 5.57001e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 8.805e-05
        [with_stream_mark]: 7.97e-06
        [recompute_prepare]: 4.87e-06
        [updatestate_depend_eliminate]: 2.43998e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.83002e-06
        [parameter_eliminate]: 1.02998e-06
        [specialize_transform]: 4.79e-06
        [updatestate_useless_node_eliminater]: 4.57e-06
        [accelerated_algorithm]: 4.98001e-06
        [meta_shard_fg_expand]: 1.07998e-06
        [get_grad_eliminate_]: 4.57e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 1.54e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.174e-05
        [j_node_and_user_rematch]: 7.92e-06
        [meta_fg_expand]: 1.45001e-06
        [replace_old_param]: 6.73e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.14998e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 5.05001e-06
        [cse]: 1.007e-05
        [replace_applicator]: 4.75999e-06
[py_interpret_to_execute_after_opt_a]: 8.92e-06
[rewriter_after_opt_a]: 3.065e-05
[convert_after_rewriter]: 7.00002e-06
[order_py_execute_after_rewriter]: 5.51998e-06
[mutable_eliminate]: 0.00048095
[jit_opt_b]: 4.795e-05, [1]
    [Cycle 1]: 4.161e-05, [2]
        [frontend_op_eliminate]: 1.624e-05
        [inline_after_opt_a]: 1.438e-05
[cconv]: 2.446e-05
[loop_unroll]: 0.00040901
[jit_opt_after_cconv]: 0.00013329, [1]
    [Cycle 1]: 0.00012705, [11]
        [c_1]: 1.994e-05
        [parameter_eliminate]: 2.49001e-06
        [updatestate_depend_eliminate]: 4.79e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.572e-05
        [call_graph_tuple_transform]: 1.811e-05
        [tuple_list_get_item_eliminator]: 5.41002e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 3.30008e-07
        [switch_simplify]: 5.39e-06
[remove_dup_value]: 1.293e-05
[partial_unused_args_eliminate]: 2.14999e-06
[environ_conv]: 5.57001e-06
[add_recomputation]: 4.488e-05
[cse_after_recomputation]: 1.947e-05, [1]
    [Cycle 1]: 1.423e-05, [1]
        [cse]: 8.67e-06
[auto_monad_reorder]: 1.394e-05
[get_jit_bprop_graph]: 1.41002e-06
[rewriter_after_jit_bprop_graph]: 4.58999e-06
[opt_after_jit_grad]: 0.00044971
[symbol_engine_optimizer]: 7.017e-05, [1]
    [Cycle 1]: 6.357e-05, [6]
        [build]: 2.76e-06
        [elim_shapecalc]: 7.33e-06
        [elim_not_effective]: 1.213e-05
        [opt_reshape]: 5.81e-06
        [fold_const_symbol]: 8.15e-06
        [renormalize]: 4.60015e-07
[validate]: 3.021e-05
[backend_pass]: 7.89994e-07
[task_emit]: 0.00588912
[execute]: 6.78998e-06
Sums
    bootstrap                            :     0.000322s :  2.30%
    type_inference                       :     0.004423s : 31.62%
    event_method                         :     0.000018s :  0.13%
    auto_monad                           :     0.000063s :  0.45%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.22%
    rewriter_before_opt_a                :     0.000071s :  0.51%
    expand_dump_flag                     :     0.000024s :  0.17%
    jit_opt_a.switch_simplify            :     0.000055s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.25%
    jit_opt_a.a_1                        :     0.000606s :  4.34%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000502s :  3.59%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.12%
    jit_opt_a.cse                        :     0.000030s :  0.22%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000481s :  3.44%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000409s :  2.92%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000045s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000450s :  3.22%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000030s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005889s : 42.10%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000180    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.91% :     0.000005s :      3: substitution.graph_param_transform
83.28% :     0.000150s :      4: substitution.inline
 1.51% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.40% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.77% :     0.000003s :      2: substitution.replace_old_param
 6.45% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004364     2
80.16% :     0.003498s :      1: type_inference.infer
19.84% :     0.000866s :      1: type_inference.specialize
------[replace.]   0.000062     6
68.18% :     0.000042s :      4: replace.inline
31.82% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000159     6
93.38% :     0.000148s :      4: match.inline
 6.62% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000001s :     10: predicate.addn_zero_filter
 2.09% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.65% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.33% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.76% :     0.000007s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000001s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.20% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.50% :     0.000002s :      3: predicate.mutable_eliminate
 0.71% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     15: predicate.partial_eliminate
 1.28% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     10: predicate.reduce_eliminate
 1.64% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.26% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.79% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.54% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.70% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000532    12
46.35% :     0.000247s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.47% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.18% :     0.000257s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019319    76
 0.25% :     0.000048s :      1: add_recomputation
 0.34% :     0.000067s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.72% :     0.000333s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.14% :     0.000027s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.31% :     0.005083s :      1: jit_opt_a
 0.70% :     0.000136s :      1: jit_opt_after_cconv
 0.26% :     0.000050s :      1: jit_opt_b
 2.16% :     0.000417s :      1: loop_unroll
 2.53% :     0.000489s :      1: mutable_eliminate
 4.15% :     0.000802s :     26: opt.transform.jit_opt_a
 0.24% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.37% :     0.000458s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.33% :     0.000258s :      1: renormalize.infer
 1.23% :     0.000237s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000033s :      1: rewriter_after_opt_a
 0.38% :     0.000073s :      1: rewriter_before_opt_a
 0.37% :     0.000072s :      1: symbol_engine_optimizer
30.54% :     0.005899s :      1: task_emit
22.95% :     0.004434s :      1: type_inference
 0.24% :     0.000047s :      1: validate


TotalTime = 0.0335465, [33]
[bootstrap]: 0.00018242
[type_inference]: 0.0119949
[event_method]: 6.174e-05
[auto_monad]: 0.00012969
[graph_reusing]: 7.4e-06
[pre_auto_parallel]: 2.98e-06
[py_interpret_to_execute]: 2.948e-05
[rewriter_before_opt_a]: 0.00010453
[expand_dump_flag]: 2.99999e-06
[jit_opt_a]: 0.011797, [3]
    [Cycle 1]: 0.00578773, [27]
        [switch_simplify]: 0.00013268
        [loop_unroll]: 4.407e-05
        [a_1]: 0.0008867
        [with_stream_mark]: 3.167e-05
        [recompute_prepare]: 1.916e-05
        [updatestate_depend_eliminate]: 7.41999e-06
        [updatestate_assign_eliminate]: 6.56e-06
        [updatestate_loads_eliminate]: 5.72999e-06
        [parameter_eliminate]: 2.71999e-06
        [specialize_transform]: 1.298e-05
        [updatestate_useless_node_eliminater]: 1.184e-05
        [accelerated_algorithm]: 1.246e-05
        [meta_shard_fg_expand]: 3.55e-06
        [get_grad_eliminate_]: 1.184e-05
        [merge_forward]: 8.22998e-06
        [cell_reuse_recompute_pass]: 9.50007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.673e-05
        [j_node_and_user_rematch]: 2.236e-05
        [meta_fg_expand]: 0.00152753
        [replace_old_param]: 5.366e-05
        [inline_without_move]: 6.885e-05
        [renormalize]: 0.00238243
        [add_forward_monad_depend]: 8.94e-06
        [auto_monad_grad]: 6.10002e-06
        [auto_monad_eliminator]: 4.829e-05
        [cse]: 0.00015057
        [replace_applicator]: 6.328e-05
    [Cycle 2]: 0.00205572, [27]
        [switch_simplify]: 3.511e-05
        [loop_unroll]: 3.327e-05
        [a_1]: 0.00105724
        [with_stream_mark]: 1.474e-05
        [recompute_prepare]: 7.70998e-06
        [updatestate_depend_eliminate]: 3.80998e-06
        [updatestate_assign_eliminate]: 3.4e-06
        [updatestate_loads_eliminate]: 2.79999e-06
        [parameter_eliminate]: 1.62001e-06
        [specialize_transform]: 6.43e-06
        [updatestate_useless_node_eliminater]: 6.11998e-06
        [accelerated_algorithm]: 6.26e-06
        [meta_shard_fg_expand]: 2.09e-06
        [get_grad_eliminate_]: 6.25002e-06
        [merge_forward]: 4.17998e-06
        [cell_reuse_recompute_pass]: 9.29984e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.609e-05
        [j_node_and_user_rematch]: 9.76998e-06
        [meta_fg_expand]: 8.171e-05
        [replace_old_param]: 1.006e-05
        [inline_without_move]: 6.23e-06
        [renormalize]: 0.00051591
        [add_forward_monad_depend]: 4.76002e-06
        [auto_monad_grad]: 1.52001e-06
        [auto_monad_eliminator]: 1.407e-05
        [cse]: 4.165e-05
        [replace_applicator]: 1.341e-05
    [Cycle 3]: 0.00036551, [27]
        [switch_simplify]: 6.93e-06
        [loop_unroll]: 6.33998e-06
        [a_1]: 0.00011723
        [with_stream_mark]: 8.84998e-06
        [recompute_prepare]: 5.89e-06
        [updatestate_depend_eliminate]: 3.9e-06
        [updatestate_assign_eliminate]: 2.71e-06
        [updatestate_loads_eliminate]: 2.53003e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 6.24999e-06
        [updatestate_useless_node_eliminater]: 6.12001e-06
        [accelerated_algorithm]: 6.26998e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 5.57001e-06
        [merge_forward]: 3.07002e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.466e-05
        [j_node_and_user_rematch]: 9.75002e-06
        [meta_fg_expand]: 2.19001e-06
        [replace_old_param]: 8.45999e-06
        [inline_without_move]: 5.87999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.17999e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 6.71e-06
        [cse]: 1.383e-05
        [replace_applicator]: 6.09001e-06
[py_interpret_to_execute_after_opt_a]: 1.005e-05
[rewriter_after_opt_a]: 4.158e-05
[convert_after_rewriter]: 8.05e-06
[order_py_execute_after_rewriter]: 5.29003e-06
[mutable_eliminate]: 0.0006978
[jit_opt_b]: 5.792e-05, [1]
    [Cycle 1]: 4.967e-05, [2]
        [frontend_op_eliminate]: 2.055e-05
        [inline_after_opt_a]: 1.773e-05
[cconv]: 2.354e-05
[loop_unroll]: 0.00042663
[jit_opt_after_cconv]: 0.00015438, [1]
    [Cycle 1]: 0.00014789, [11]
        [c_1]: 2.547e-05
        [parameter_eliminate]: 2.37999e-06
        [updatestate_depend_eliminate]: 5.72001e-06
        [updatestate_assign_eliminate]: 2.86999e-06
        [updatestate_loads_eliminate]: 2.74999e-06
        [cse]: 2.181e-05
        [call_graph_tuple_transform]: 2.15e-05
        [tuple_list_get_item_eliminator]: 6.74001e-06
        [none_parameter_eliminate]: 1.54998e-06
        [renormalize]: 4.10015e-07
        [switch_simplify]: 6.51e-06
[remove_dup_value]: 1.717e-05
[partial_unused_args_eliminate]: 2.56e-06
[environ_conv]: 5.15001e-06
[add_recomputation]: 4.861e-05
[cse_after_recomputation]: 3.162e-05, [1]
    [Cycle 1]: 2.044e-05, [1]
        [cse]: 1.479e-05
[auto_monad_reorder]: 1.786e-05
[get_jit_bprop_graph]: 1.44e-06
[rewriter_after_jit_bprop_graph]: 5.76e-06
[opt_after_jit_grad]: 0.00048213
[symbol_engine_optimizer]: 8.059e-05, [1]
    [Cycle 1]: 7.395e-05, [6]
        [build]: 4.22998e-06
        [elim_shapecalc]: 8.95001e-06
        [elim_not_effective]: 1.489e-05
        [opt_reshape]: 6.74999e-06
        [fold_const_symbol]: 1.122e-05
        [renormalize]: 4.89992e-07
[validate]: 3.662e-05
[backend_pass]: 8.99978e-07
[task_emit]: 0.0068791
[execute]: 6.94999e-06
Sums
    bootstrap                            :     0.000182s :  0.63%
    type_inference                       :     0.011995s : 41.15%
    event_method                         :     0.000062s :  0.21%
    auto_monad                           :     0.000130s :  0.44%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000029s :  0.10%
    rewriter_before_opt_a                :     0.000105s :  0.36%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000175s :  0.60%
    jit_opt_a.loop_unroll                :     0.000084s :  0.29%
    jit_opt_a.a_1                        :     0.002061s :  7.07%
    jit_opt_a.with_stream_mark           :     0.000055s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000025s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.08%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000057s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.001611s :  5.53%
    jit_opt_a.replace_old_param          :     0.000072s :  0.25%
    jit_opt_a.inline_without_move        :     0.000081s :  0.28%
    jit_opt_a.renormalize                :     0.002898s :  9.94%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000069s :  0.24%
    jit_opt_a.cse                        :     0.000206s :  0.71%
    jit_opt_a.replace_applicator         :     0.000083s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.03%
    rewriter_after_opt_a                 :     0.000042s :  0.14%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000698s :  2.39%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000024s :  0.08%
    loop_unroll                          :     0.000427s :  1.46%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000017s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000049s :  0.17%
    cse_after_recomputation.cse          :     0.000015s :  0.05%
    auto_monad_reorder                   :     0.000018s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000482s :  1.65%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006879s : 23.60%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000507   113
 0.44% :     0.000002s :      3: substitution.elim_not_effective
 0.32% :     0.000002s :      3: substitution.fold_const_symbol
 1.31% :     0.000007s :      4: substitution.graph_param_transform
63.95% :     0.000324s :     14: substitution.inline
 3.44% :     0.000017s :      2: substitution.inline_without_move
 1.71% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.14% :     0.000011s :      7: substitution.minmaximum_grad
 1.30% :     0.000007s :      5: substitution.partial_eliminate
 2.20% :     0.000011s :     14: substitution.remove_not_recompute_node
 4.96% :     0.000025s :     10: substitution.replace_applicator
 1.55% :     0.000008s :      7: substitution.replace_old_param
 0.45% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.56% :     0.000008s :      1: substitution.switch_simplify
 4.01% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.70% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.97% :     0.000040s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011921     2
89.68% :     0.010691s :      1: type_inference.infer
10.32% :     0.001230s :      1: type_inference.specialize
------[replace.]   0.000186    22
53.05% :     0.000099s :     14: replace.inline
19.90% :     0.000037s :      1: replace.switch_simplify
27.05% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000345    22
91.70% :     0.000317s :     14: match.inline
 2.11% :     0.000007s :      1: match.switch_simplify
 6.19% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000412  2634
 1.44% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.39% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 8.75% :     0.000036s :     44: predicate.addn_check_dump
 1.53% :     0.000006s :     44: predicate.addn_zero_filter
 1.82% :     0.000008s :     44: predicate.arithmetic_simplify
 1.45% :     0.000006s :     44: predicate.cast_eliminate
 0.23% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.25% :     0.000005s :     44: predicate.compare_switch_simplify
 1.50% :     0.000006s :     44: predicate.depend_value_elim
 1.44% :     0.000006s :     44: predicate.dict_get_item_const_eliminator
 1.55% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.40% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.30% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      4: predicate.elim_not_effective
 0.18% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.29% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.30% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.42% :     0.000006s :     44: predicate.environ_get_eliminate
 1.29% :     0.000005s :     44: predicate.environ_get_set_eliminate
 0.08% :     0.000000s :      4: predicate.fold_const_symbol
 0.78% :     0.000003s :     19: predicate.get_grad_eliminate
 0.09% :     0.000000s :      4: predicate.graph_param_transform
 4.23% :     0.000017s :     73: predicate.inline
 1.77% :     0.000007s :     40: predicate.inline_without_move
 0.32% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.99% :     0.000004s :     19: predicate.less_batch_normalization
 1.69% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.89% :     0.000008s :     55: predicate.load_eliminater
 0.49% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.02% :     0.000012s :     91: predicate.loop_unroll_before_grad
 1.57% :     0.000006s :     48: predicate.make_slice_get_slice_eliminator
 1.36% :     0.000006s :     44: predicate.merge_addn
 1.33% :     0.000005s :     44: predicate.minmaximum_grad
 0.51% :     0.000002s :      4: predicate.mutable_eliminate
 0.17% :     0.000001s :      4: predicate.opt_reshape
 2.14% :     0.000009s :     55: predicate.partial_eliminate
 1.42% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.91% :     0.000008s :     44: predicate.reduce_eliminate
 1.65% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.38% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.30% :     0.000009s :     97: predicate.replace_applicator
 0.82% :     0.000003s :     40: predicate.replace_old_param
 0.09% :     0.000000s :      4: predicate.reset_defer_inline
 1.51% :     0.000006s :     44: predicate.reshape_eliminate
 1.38% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.36% :     0.000002s :      4: predicate.row_tensor_eliminate
 1.41% :     0.000006s :     44: predicate.same_eliminate
 0.40% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.41% :     0.000002s :      8: predicate.special_op_eliminate
 0.80% :     0.000003s :     19: predicate.specialize_transform
 1.66% :     0.000007s :     44: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.15% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.56% :     0.000011s :     65: predicate.switch_defer_inline
 2.37% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.09% :     0.000025s :    162: predicate.switch_simplify
 1.35% :     0.000006s :     44: predicate.tile_eliminate
 1.33% :     0.000005s :     44: predicate.transpose_eliminate
 1.61% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.51% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 2.91% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.73% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.66% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000007s :     44: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.23% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001487    32
60.84% :     0.000904s :     14: func_graph_cloner_run.FuncGraphClonerGraph
39.16% :     0.000582s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.039312    91
 0.13% :     0.000051s :      1: add_recomputation
 0.34% :     0.000135s :      1: auto_monad
 0.05% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.50% :     0.000195s :      1: bootstrap
 0.07% :     0.000026s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000034s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.17% :     0.000067s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
30.02% :     0.011800s :      1: jit_opt_a
 0.40% :     0.000157s :      1: jit_opt_after_cconv
 0.15% :     0.000061s :      1: jit_opt_b
 1.10% :     0.000434s :      1: loop_unroll
 1.80% :     0.000707s :      1: mutable_eliminate
 6.97% :     0.002740s :     39: opt.transform.jit_opt_a
 0.14% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000032s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.25% :     0.000490s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000032s :      1: py_interpret_to_execute
 0.03% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000020s :      1: remove_dup_value
 4.00% :     0.001573s :      2: renormalize.infer
 3.33% :     0.001310s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000044s :      1: rewriter_after_opt_a
 0.27% :     0.000107s :      1: rewriter_before_opt_a
 0.21% :     0.000083s :      1: symbol_engine_optimizer
17.52% :     0.006889s :      1: task_emit
30.55% :     0.012011s :      1: type_inference
 0.15% :     0.000058s :      1: validate


TotalTime = 0.0207883, [33]
[bootstrap]: 0.00028763
[type_inference]: 0.00455354
[event_method]: 1.755e-05
[auto_monad]: 6.32e-05
[graph_reusing]: 5.76003e-06
[pre_auto_parallel]: 2.48998e-06
[py_interpret_to_execute]: 3.477e-05
[rewriter_before_opt_a]: 7.088e-05
[expand_dump_flag]: 3.68999e-06
[jit_opt_a]: 0.00608653, [2]
    [Cycle 1]: 0.0020557, [27]
        [switch_simplify]: 5.892e-05
        [loop_unroll]: 2.972e-05
        [a_1]: 0.0005901
        [with_stream_mark]: 2.116e-05
        [recompute_prepare]: 8.65001e-06
        [updatestate_depend_eliminate]: 4.53001e-06
        [updatestate_assign_eliminate]: 3.25e-06
        [updatestate_loads_eliminate]: 2.94001e-06
        [parameter_eliminate]: 2.03997e-06
        [specialize_transform]: 6.22001e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 5.97001e-06
        [meta_shard_fg_expand]: 2.39001e-06
        [get_grad_eliminate_]: 5.40001e-06
        [merge_forward]: 3.72002e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.592e-05
        [j_node_and_user_rematch]: 9.71e-06
        [meta_fg_expand]: 2.23002e-06
        [replace_old_param]: 9.19e-06
        [inline_without_move]: 5.39e-06
        [renormalize]: 0.00093703
        [add_forward_monad_depend]: 8.44998e-06
        [auto_monad_grad]: 2.56e-06
        [auto_monad_eliminator]: 1.594e-05
        [cse]: 3.078e-05
        [replace_applicator]: 1.576e-05
    [Cycle 2]: 0.00034005, [27]
        [switch_simplify]: 5.99e-06
        [loop_unroll]: 5.59e-06
        [a_1]: 9.485e-05
        [with_stream_mark]: 1.229e-05
        [recompute_prepare]: 5.05001e-06
        [updatestate_depend_eliminate]: 2.95002e-06
        [updatestate_assign_eliminate]: 2.86e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [parameter_eliminate]: 1.37999e-06
        [specialize_transform]: 4.84998e-06
        [updatestate_useless_node_eliminater]: 4.74998e-06
        [accelerated_algorithm]: 5.81e-06
        [meta_shard_fg_expand]: 2.02001e-06
        [get_grad_eliminate_]: 4.84998e-06
        [merge_forward]: 3.41001e-06
        [cell_reuse_recompute_pass]: 1.96e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.394e-05
        [j_node_and_user_rematch]: 8.59e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 8.08999e-06
        [inline_without_move]: 4.63999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.63002e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 5.26998e-06
        [cse]: 1.094e-05
        [replace_applicator]: 5.24e-06
[py_interpret_to_execute_after_opt_a]: 1.18e-05
[rewriter_after_opt_a]: 4.037e-05
[convert_after_rewriter]: 7.68999e-06
[order_py_execute_after_rewriter]: 4.84003e-06
[mutable_eliminate]: 0.00069446
[jit_opt_b]: 5.328e-05, [1]
    [Cycle 1]: 4.5e-05, [2]
        [frontend_op_eliminate]: 1.739e-05
        [inline_after_opt_a]: 1.537e-05
[cconv]: 2.779e-05
[loop_unroll]: 0.00045731
[jit_opt_after_cconv]: 0.00015049, [1]
    [Cycle 1]: 0.00014336, [11]
        [c_1]: 2.117e-05
        [parameter_eliminate]: 2.63e-06
        [updatestate_depend_eliminate]: 6.04001e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.56998e-06
        [cse]: 2.37e-05
        [call_graph_tuple_transform]: 2.113e-05
        [tuple_list_get_item_eliminator]: 5.80002e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.61e-06
[remove_dup_value]: 1.502e-05
[partial_unused_args_eliminate]: 2.99001e-06
[environ_conv]: 6.62002e-06
[add_recomputation]: 5.509e-05
[cse_after_recomputation]: 2.157e-05, [1]
    [Cycle 1]: 1.584e-05, [1]
        [cse]: 9.41003e-06
[auto_monad_reorder]: 1.557e-05
[get_jit_bprop_graph]: 2.53998e-06
[rewriter_after_jit_bprop_graph]: 8.2e-06
[opt_after_jit_grad]: 0.0005338
[symbol_engine_optimizer]: 7.955e-05, [1]
    [Cycle 1]: 7.148e-05, [6]
        [build]: 4.88001e-06
        [elim_shapecalc]: 8.08001e-06
        [elim_not_effective]: 1.372e-05
        [opt_reshape]: 5.94999e-06
        [fold_const_symbol]: 9.20001e-06
        [renormalize]: 5.89993e-07
[validate]: 4.45e-05
[backend_pass]: 1.22999e-06
[task_emit]: 0.00712839
[execute]: 7.91001e-06
Sums
    bootstrap                            :     0.000288s :  1.76%
    type_inference                       :     0.004554s : 27.92%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000063s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.21%
    rewriter_before_opt_a                :     0.000071s :  0.43%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000685s :  4.20%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000937s :  5.75%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.13%
    jit_opt_a.cse                        :     0.000042s :  0.26%
    jit_opt_a.replace_applicator         :     0.000021s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000040s :  0.25%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000694s :  4.26%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000028s :  0.17%
    loop_unroll                          :     0.000457s :  2.80%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000007s :  0.04%
    add_recomputation                    :     0.000055s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000534s :  3.27%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007128s : 43.71%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000255    23
 0.80% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.39% :     0.000006s :      3: substitution.graph_param_transform
85.19% :     0.000217s :      4: substitution.inline
 1.76% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.10% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.23% :     0.000006s :      2: substitution.replace_old_param
 5.05% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004492     2
80.68% :     0.003624s :      1: type_inference.infer
19.32% :     0.000868s :      1: type_inference.specialize
------[replace.]   0.000075     6
69.07% :     0.000052s :      4: replace.inline
30.93% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000226     6
94.81% :     0.000215s :      4: match.inline
 5.19% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.13% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.58% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.72% :     0.000002s :     10: predicate.addn_zero_filter
 2.30% :     0.000003s :     10: predicate.arithmetic_simplify
 1.62% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000001s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.10% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 6.11% :     0.000007s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.18% :     0.000001s :      6: predicate.less_batch_normalization
 1.74% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.56% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 1.66% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.92% :     0.000002s :     18: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.72% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.09% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.21% :     0.000001s :      6: predicate.special_op_eliminate
 1.13% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.09% :     0.000003s :     16: predicate.switch_defer_inline
 2.01% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.45% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.41% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.06% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000647    12
45.00% :     0.000291s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.85% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.15% :     0.000305s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022735    76
 0.26% :     0.000058s :      1: add_recomputation
 0.29% :     0.000066s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.32% :     0.000301s :      1: bootstrap
 0.13% :     0.000030s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.79% :     0.006091s :      1: jit_opt_a
 0.67% :     0.000153s :      1: jit_opt_after_cconv
 0.25% :     0.000056s :      1: jit_opt_b
 2.05% :     0.000467s :      1: loop_unroll
 3.10% :     0.000704s :      1: mutable_eliminate
 3.99% :     0.000908s :     26: opt.transform.jit_opt_a
 0.22% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.39% :     0.000543s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 2.20% :     0.000501s :      1: renormalize.infer
 1.87% :     0.000425s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000043s :      1: rewriter_after_opt_a
 0.32% :     0.000074s :      1: rewriter_before_opt_a
 0.36% :     0.000083s :      1: symbol_engine_optimizer
31.42% :     0.007144s :      1: task_emit
20.09% :     0.004568s :      1: type_inference
 0.64% :     0.000146s :      1: validate


TotalTime = 0.0229255, [33]
[bootstrap]: 0.00032638
[type_inference]: 0.0046733
[event_method]: 1.685e-05
[auto_monad]: 6.379e-05
[graph_reusing]: 5.30001e-06
[pre_auto_parallel]: 2.63e-06
[py_interpret_to_execute]: 3.072e-05
[rewriter_before_opt_a]: 6.832e-05
[expand_dump_flag]: 3.66001e-06
[jit_opt_a]: 0.01006, [2]
    [Cycle 1]: 0.0061351, [27]
        [switch_simplify]: 4.772e-05
        [loop_unroll]: 2.895e-05
        [a_1]: 0.00504547
        [with_stream_mark]: 1.971e-05
        [recompute_prepare]: 8.65001e-06
        [updatestate_depend_eliminate]: 3.21999e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 6.24999e-06
        [updatestate_useless_node_eliminater]: 5.07999e-06
        [accelerated_algorithm]: 5.44998e-06
        [meta_shard_fg_expand]: 2.01998e-06
        [get_grad_eliminate_]: 4.94e-06
        [merge_forward]: 2.84001e-06
        [cell_reuse_recompute_pass]: 7.06001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.341e-05
        [j_node_and_user_rematch]: 8.82e-06
        [meta_fg_expand]: 2.19001e-06
        [replace_old_param]: 8.89e-06
        [inline_without_move]: 5.22e-06
        [renormalize]: 0.00067382
        [add_forward_monad_depend]: 5.39e-06
        [auto_monad_grad]: 1.71e-06
        [auto_monad_eliminator]: 1.193e-05
        [cse]: 2.229e-05
        [replace_applicator]: 1.263e-05
    [Cycle 2]: 0.00032732, [27]
        [switch_simplify]: 6.02999e-06
        [loop_unroll]: 5.30999e-06
        [a_1]: 9.181e-05
        [with_stream_mark]: 9.02999e-06
        [recompute_prepare]: 5.49e-06
        [updatestate_depend_eliminate]: 2.89999e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 5.28002e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.01002e-06
        [get_grad_eliminate_]: 4.99e-06
        [merge_forward]: 2.50997e-06
        [cell_reuse_recompute_pass]: 1.52001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.19e-05
        [j_node_and_user_rematch]: 8.65999e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 7.43e-06
        [inline_without_move]: 4.90001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 6.59988e-07
        [auto_monad_eliminator]: 5.57999e-06
        [cse]: 1.068e-05
        [replace_applicator]: 6.39001e-06
[py_interpret_to_execute_after_opt_a]: 9.17001e-06
[rewriter_after_opt_a]: 3.376e-05
[convert_after_rewriter]: 6.88e-06
[order_py_execute_after_rewriter]: 4.58999e-06
[mutable_eliminate]: 0.00073099
[jit_opt_b]: 5.23e-05, [1]
    [Cycle 1]: 4.401e-05, [2]
        [frontend_op_eliminate]: 1.721e-05
        [inline_after_opt_a]: 1.489e-05
[cconv]: 2.037e-05
[loop_unroll]: 0.00042386
[jit_opt_after_cconv]: 0.00013513, [1]
    [Cycle 1]: 0.00012854, [11]
        [c_1]: 2.009e-05
        [parameter_eliminate]: 2.57001e-06
        [updatestate_depend_eliminate]: 5.20999e-06
        [updatestate_assign_eliminate]: 2.47001e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [cse]: 1.726e-05
        [call_graph_tuple_transform]: 1.741e-05
        [tuple_list_get_item_eliminator]: 5.34e-06
        [none_parameter_eliminate]: 1.42999e-06
        [renormalize]: 4.2998e-07
        [switch_simplify]: 5.31002e-06
[remove_dup_value]: 1.063e-05
[partial_unused_args_eliminate]: 1.87999e-06
[environ_conv]: 4.07998e-06
[add_recomputation]: 4.769e-05
[cse_after_recomputation]: 2.158e-05, [1]
    [Cycle 1]: 1.554e-05, [1]
        [cse]: 9.86e-06
[auto_monad_reorder]: 1.283e-05
[get_jit_bprop_graph]: 1.40999e-06
[rewriter_after_jit_bprop_graph]: 5.01997e-06
[opt_after_jit_grad]: 0.00049784
[symbol_engine_optimizer]: 7.251e-05, [1]
    [Cycle 1]: 6.62e-05, [6]
        [build]: 3.43999e-06
        [elim_shapecalc]: 7.55e-06
        [elim_not_effective]: 1.243e-05
        [opt_reshape]: 6.01e-06
        [fold_const_symbol]: 8.85001e-06
        [renormalize]: 4.39992e-07
[validate]: 3.4e-05
[backend_pass]: 1.02998e-06
[task_emit]: 0.00536209
[execute]: 5.89e-06
Sums
    bootstrap                            :     0.000326s :  1.74%
    type_inference                       :     0.004673s : 24.94%
    event_method                         :     0.000017s :  0.09%
    auto_monad                           :     0.000064s :  0.34%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000031s :  0.16%
    rewriter_before_opt_a                :     0.000068s :  0.36%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.29%
    jit_opt_a.loop_unroll                :     0.000034s :  0.18%
    jit_opt_a.a_1                        :     0.005137s : 27.42%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000009s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.09%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000674s :  3.60%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.09%
    jit_opt_a.cse                        :     0.000033s :  0.18%
    jit_opt_a.replace_applicator         :     0.000019s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.05%
    rewriter_after_opt_a                 :     0.000034s :  0.18%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000731s :  3.90%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.08%
    cconv                                :     0.000020s :  0.11%
    loop_unroll                          :     0.000424s :  2.26%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000011s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000048s :  0.25%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000498s :  2.66%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.18%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005362s : 28.62%
    execute                              :     0.000006s :  0.03%


Time group info:
------[substitution.]   0.004674    23
 0.04% :     0.000002s :      2: substitution.elim_not_effective
 0.03% :     0.000001s :      2: substitution.fold_const_symbol
 0.10% :     0.000005s :      3: substitution.graph_param_transform
 2.87% :     0.000134s :      4: substitution.inline
 0.06% :     0.000003s :      4: substitution.j_node_and_user_rematch
 0.09% :     0.000004s :      4: substitution.remove_not_recompute_node
 0.09% :     0.000004s :      2: substitution.replace_old_param
96.71% :     0.004520s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004616     2
83.51% :     0.003855s :      1: type_inference.infer
16.49% :     0.000761s :      1: type_inference.specialize
------[replace.]   0.000084     6
47.08% :     0.000039s :      4: replace.inline
52.92% :     0.000044s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.004650     6
 2.84% :     0.000132s :      4: match.inline
97.16% :     0.004518s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000164   691
 1.18% :     0.000002s :     10: predicate.accumulaten_eliminater
25.22% :     0.000041s :      3: predicate.ad_related_special_op_eliminate
 0.74% :     0.000001s :     10: predicate.addn_check_dump
 1.08% :     0.000002s :     10: predicate.addn_zero_filter
 1.82% :     0.000003s :     10: predicate.arithmetic_simplify
 1.04% :     0.000002s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.74% :     0.000001s :     10: predicate.compare_switch_simplify
 1.01% :     0.000002s :     10: predicate.depend_value_elim
 0.79% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.39% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.89% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.24% :     0.000000s :      3: predicate.elim_not_effective
 0.37% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.80% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.82% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.83% :     0.000001s :     10: predicate.environ_get_eliminate
 0.77% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.69% :     0.000001s :      6: predicate.get_grad_eliminate
 0.19% :     0.000000s :      3: predicate.graph_param_transform
 3.69% :     0.000006s :     22: predicate.inline
 0.66% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.79% :     0.000001s :      6: predicate.less_batch_normalization
 1.11% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.24% :     0.000002s :     15: predicate.load_eliminater
 0.93% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.06% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.67% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.79% :     0.000001s :     10: predicate.merge_addn
 0.76% :     0.000001s :     10: predicate.minmaximum_grad
 1.20% :     0.000002s :      3: predicate.mutable_eliminate
 0.65% :     0.000001s :      3: predicate.opt_reshape
 1.66% :     0.000003s :     15: predicate.partial_eliminate
 1.02% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.36% :     0.000002s :     10: predicate.reduce_eliminate
 1.29% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.32% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.26% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.19% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.02% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000002s :     10: predicate.same_eliminate
 0.40% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000002s :      6: predicate.special_op_eliminate
 0.73% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.29% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.07% :     0.000003s :     16: predicate.switch_defer_inline
 1.52% :     0.000003s :     16: predicate.switch_layer_defer_inline
 5.26% :     0.000009s :     53: predicate.switch_simplify
 0.91% :     0.000002s :     10: predicate.tile_eliminate
 0.88% :     0.000001s :     10: predicate.transpose_eliminate
 0.97% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.87% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 2.50% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.35% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.17% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.22% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.36% :     0.000002s :     10: predicate.value_based_eliminate
 0.24% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.52% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000547    12
49.61% :     0.000271s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.27% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.12% :     0.000247s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.029089    76
 0.18% :     0.000051s :      1: add_recomputation
 0.23% :     0.000067s :      1: auto_monad
 0.05% :     0.000015s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.16% :     0.000338s :      1: bootstrap
 0.08% :     0.000024s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.07% :     0.000020s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
34.60% :     0.010064s :      1: jit_opt_a
 0.47% :     0.000138s :      1: jit_opt_after_cconv
 0.19% :     0.000055s :      1: jit_opt_b
 1.48% :     0.000432s :      1: loop_unroll
 2.54% :     0.000739s :      1: mutable_eliminate
18.37% :     0.005343s :     26: opt.transform.jit_opt_a
 0.15% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000026s :      4: opt.transform.jit_opt_b
 0.04% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.21% :     0.000060s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000031s :      4: opt.transform.symbol_engine_opt
 1.74% :     0.000506s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000033s :      1: py_interpret_to_execute
 0.04% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000013s :      1: remove_dup_value
 1.26% :     0.000366s :      1: renormalize.infer
 1.03% :     0.000299s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000037s :      1: rewriter_after_opt_a
 0.24% :     0.000071s :      1: rewriter_before_opt_a
 0.26% :     0.000075s :      1: symbol_engine_optimizer
18.47% :     0.005372s :      1: task_emit
16.11% :     0.004686s :      1: type_inference
 0.18% :     0.000053s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.638.04 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dcafb80> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.643.83 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0155904, [30]
[bootstrap]: 0.00031882
[type_inference]: 0.00505564
[event_method]: 1.796e-05
[auto_monad]: 6.594e-05
[graph_reusing]: 5.74e-06
[pre_auto_parallel]: 2.74999e-06
[py_interpret_to_execute]: 3.724e-05
[rewriter_before_opt_a]: 0.0001574
[expand_dump_flag]: 3.26999e-06
[jit_opt_a]: 0.00635684, [2]
    [Cycle 1]: 0.00176545, [27]
        [switch_simplify]: 6.073e-05
        [loop_unroll]: 3.188e-05
        [a_1]: 0.0005751
        [with_stream_mark]: 2.113e-05
        [recompute_prepare]: 8.64998e-06
        [updatestate_depend_eliminate]: 3.61999e-06
        [updatestate_assign_eliminate]: 2.90002e-06
        [updatestate_loads_eliminate]: 2.72001e-06
        [parameter_eliminate]: 2.01998e-06
        [specialize_transform]: 5.86998e-06
        [updatestate_useless_node_eliminater]: 5.37001e-06
        [accelerated_algorithm]: 5.66e-06
        [meta_shard_fg_expand]: 1.90001e-06
        [get_grad_eliminate_]: 5.49998e-06
        [merge_forward]: 3.53999e-06
        [cell_reuse_recompute_pass]: 1.56998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.445e-05
        [j_node_and_user_rematch]: 8.82e-06
        [meta_fg_expand]: 2.35002e-06
        [replace_old_param]: 9.25001e-06
        [inline_without_move]: 5.86e-06
        [renormalize]: 0.00074002
        [add_forward_monad_depend]: 6.18998e-06
        [auto_monad_grad]: 2.74999e-06
        [auto_monad_eliminator]: 1.48e-05
        [cse]: 2.757e-05
        [replace_applicator]: 1.26e-05
    [Cycle 2]: 0.00034264, [27]
        [switch_simplify]: 6.51999e-06
        [loop_unroll]: 4.86002e-06
        [a_1]: 9.086e-05
        [with_stream_mark]: 1.142e-05
        [recompute_prepare]: 5.72001e-06
        [updatestate_depend_eliminate]: 3.61001e-06
        [updatestate_assign_eliminate]: 2.13998e-06
        [updatestate_loads_eliminate]: 2.34999e-06
        [parameter_eliminate]: 1.28002e-06
        [specialize_transform]: 5.99e-06
        [updatestate_useless_node_eliminater]: 5.30001e-06
        [accelerated_algorithm]: 5.62001e-06
        [meta_shard_fg_expand]: 1.35001e-06
        [get_grad_eliminate_]: 4.87998e-06
        [merge_forward]: 2.91999e-06
        [cell_reuse_recompute_pass]: 1.98002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.442e-05
        [j_node_and_user_rematch]: 8.61002e-06
        [meta_fg_expand]: 1.87999e-06
        [replace_old_param]: 8.53001e-06
        [inline_without_move]: 5.35999e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.52999e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 5.37001e-06
        [cse]: 1.157e-05
        [replace_applicator]: 6.02001e-06
[py_interpret_to_execute_after_opt_a]: 1.793e-05
[rewriter_after_opt_a]: 3.828e-05
[convert_after_rewriter]: 8e-06
[order_py_execute_after_rewriter]: 5.98002e-06
[mutable_eliminate]: 0.00163834
[jit_opt_b]: 6.44e-05, [1]
    [Cycle 1]: 5.378e-05, [2]
        [frontend_op_eliminate]: 2.067e-05
        [inline_after_opt_a]: 2.043e-05
[cconv]: 3.939e-05
[loop_unroll]: 0.00059812
[jit_opt_after_cconv]: 0.0001798, [1]
    [Cycle 1]: 0.00017215, [11]
        [c_1]: 2.176e-05
        [parameter_eliminate]: 6.36e-06
        [updatestate_depend_eliminate]: 1.07e-05
        [updatestate_assign_eliminate]: 3.71001e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [cse]: 3.42e-05
        [call_graph_tuple_transform]: 2.501e-05
        [tuple_list_get_item_eliminator]: 6.09999e-06
        [none_parameter_eliminate]: 1.67001e-06
        [renormalize]: 8.70001e-07
        [switch_simplify]: 5.69999e-06
[remove_dup_value]: 1.572e-05
[partial_unused_args_eliminate]: 2.51998e-06
[environ_conv]: 5.72001e-06
[add_recomputation]: 5.294e-05
[cse_after_recomputation]: 2.355e-05, [1]
    [Cycle 1]: 1.71e-05, [1]
        [cse]: 1.052e-05
[auto_monad_reorder]: 1.737e-05
[get_jit_bprop_graph]: 2.79001e-06
[rewriter_after_jit_bprop_graph]: 9.42001e-06
[opt_after_jit_grad]: 0.00052458
[symbol_engine_optimizer]: 8.39e-05, [1]
    [Cycle 1]: 7.534e-05, [6]
        [build]: 5.24e-06
        [elim_shapecalc]: 7.87e-06
        [elim_not_effective]: 1.572e-05
        [opt_reshape]: 6.46e-06
        [fold_const_symbol]: 9.86e-06
        [renormalize]: 5.50004e-07
[validate]: 4.619e-05
Sums
    bootstrap                            :     0.000319s :  2.98%
    type_inference                       :     0.005056s : 47.21%
    event_method                         :     0.000018s :  0.17%
    auto_monad                           :     0.000066s :  0.62%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000037s :  0.35%
    rewriter_before_opt_a                :     0.000157s :  1.47%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000067s :  0.63%
    jit_opt_a.loop_unroll                :     0.000037s :  0.34%
    jit_opt_a.a_1                        :     0.000666s :  6.22%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.17%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000740s :  6.91%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.19%
    jit_opt_a.cse                        :     0.000039s :  0.37%
    jit_opt_a.replace_applicator         :     0.000019s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.17%
    rewriter_after_opt_a                 :     0.000038s :  0.36%
    convert_after_rewriter               :     0.000008s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.001638s : 15.30%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.19%
    cconv                                :     0.000039s :  0.37%
    loop_unroll                          :     0.000598s :  5.59%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.20%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.10%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000034s :  0.32%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000016s :  0.15%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000053s :  0.49%
    cse_after_recomputation.cse          :     0.000011s :  0.10%
    auto_monad_reorder                   :     0.000017s :  0.16%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.09%
    opt_after_jit_grad                   :     0.000525s :  4.90%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000046s :  0.43%


Time group info:
------[substitution.]   0.000231    23
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000002s :      2: substitution.fold_const_symbol
 2.66% :     0.000006s :      3: substitution.graph_param_transform
84.92% :     0.000196s :      4: substitution.inline
 1.35% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.88% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.15% :     0.000005s :      2: substitution.replace_old_param
 5.44% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004993     2
81.59% :     0.004073s :      1: type_inference.infer
18.41% :     0.000919s :      1: type_inference.specialize
------[replace.]   0.000069     6
67.53% :     0.000047s :      4: replace.inline
32.47% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000205     6
94.41% :     0.000194s :      4: match.inline
 5.59% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.20% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.13% :     0.000001s :     10: predicate.addn_zero_filter
 2.39% :     0.000003s :     10: predicate.arithmetic_simplify
 1.08% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.93% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 0.98% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.99% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000002s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 6.07% :     0.000008s :     22: predicate.inline
 1.20% :     0.000002s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 1.51% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.78% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 3.35% :     0.000004s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     15: predicate.partial_eliminate
 1.08% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.50% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.39% :     0.000001s :      3: predicate.reset_defer_inline
 1.58% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.31% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.30% :     0.000002s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.99% :     0.000004s :     16: predicate.switch_defer_inline
 2.39% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.40% :     0.000010s :     53: predicate.switch_simplify
 1.06% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000002s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.09% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.20% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.34% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.24% :     0.000002s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000625    12
49.28% :     0.000308s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.88% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.83% :     0.000268s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017296    72
 0.32% :     0.000056s :      1: add_recomputation
 0.40% :     0.000069s :      1: auto_monad
 0.12% :     0.000020s :      1: auto_monad_reorder
 1.92% :     0.000331s :      1: bootstrap
 0.25% :     0.000043s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.12% :     0.000022s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
36.77% :     0.006360s :      1: jit_opt_a
 1.06% :     0.000183s :      1: jit_opt_after_cconv
 0.39% :     0.000068s :      1: jit_opt_b
 3.52% :     0.000609s :      1: loop_unroll
 9.58% :     0.001658s :      1: mutable_eliminate
 5.15% :     0.000891s :     26: opt.transform.jit_opt_a
 0.32% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000032s :      4: opt.transform.jit_opt_b
 0.09% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.15% :     0.000025s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.09% :     0.000535s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000040s :      1: py_interpret_to_execute
 0.12% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.31% :     0.000399s :      1: renormalize.infer
 1.92% :     0.000332s :      1: renormalize.specialize
 0.07% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000041s :      1: rewriter_after_opt_a
 0.93% :     0.000162s :      1: rewriter_before_opt_a
 0.50% :     0.000087s :      1: symbol_engine_optimizer
29.31% :     0.005069s :      1: type_inference


TotalTime = 0.021989, [33]
[bootstrap]: 0.00034182
[type_inference]: 0.00543074
[event_method]: 1.857e-05
[auto_monad]: 7.219e-05
[graph_reusing]: 6.65998e-06
[pre_auto_parallel]: 2.41998e-06
[py_interpret_to_execute]: 4.276e-05
[rewriter_before_opt_a]: 7.644e-05
[expand_dump_flag]: 3.21001e-06
[jit_opt_a]: 0.00646066, [2]
    [Cycle 1]: 0.00209487, [27]
        [switch_simplify]: 6.798e-05
        [loop_unroll]: 2.953e-05
        [a_1]: 0.00060367
        [with_stream_mark]: 2.304e-05
        [recompute_prepare]: 8.52998e-06
        [updatestate_depend_eliminate]: 4e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 2.55002e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 6.07001e-06
        [updatestate_useless_node_eliminater]: 5.48002e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 2.32999e-06
        [get_grad_eliminate_]: 5.73002e-06
        [merge_forward]: 4.28001e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.46e-05
        [j_node_and_user_rematch]: 9.56003e-06
        [meta_fg_expand]: 2.21e-06
        [replace_old_param]: 9.59999e-06
        [inline_without_move]: 5.72001e-06
        [renormalize]: 0.00101593
        [add_forward_monad_depend]: 6.88e-06
        [auto_monad_grad]: 2.69001e-06
        [auto_monad_eliminator]: 1.682e-05
        [cse]: 3.085e-05
        [replace_applicator]: 1.615e-05
    [Cycle 2]: 0.00034062, [27]
        [switch_simplify]: 5.77001e-06
        [loop_unroll]: 5.01997e-06
        [a_1]: 9.08e-05
        [with_stream_mark]: 1.323e-05
        [recompute_prepare]: 5.49998e-06
        [updatestate_depend_eliminate]: 3.31001e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 1.45001e-06
        [specialize_transform]: 4.95001e-06
        [updatestate_useless_node_eliminater]: 4.68999e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 1.69e-06
        [get_grad_eliminate_]: 4.77998e-06
        [merge_forward]: 2.88e-06
        [cell_reuse_recompute_pass]: 1.83002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.633e-05
        [j_node_and_user_rematch]: 8.62998e-06
        [meta_fg_expand]: 1.71998e-06
        [replace_old_param]: 8.37e-06
        [inline_without_move]: 5.32001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 1.10001e-06
        [auto_monad_eliminator]: 5.74e-06
        [cse]: 1.085e-05
        [replace_applicator]: 5.02e-06
[py_interpret_to_execute_after_opt_a]: 1.511e-05
[rewriter_after_opt_a]: 3.949e-05
[convert_after_rewriter]: 7.04001e-06
[order_py_execute_after_rewriter]: 5.57001e-06
[mutable_eliminate]: 0.00074361
[jit_opt_b]: 5.457e-05, [1]
    [Cycle 1]: 4.562e-05, [2]
        [frontend_op_eliminate]: 1.799e-05
        [inline_after_opt_a]: 1.504e-05
[cconv]: 2.858e-05
[loop_unroll]: 0.00046073
[jit_opt_after_cconv]: 0.00015109, [1]
    [Cycle 1]: 0.00014318, [11]
        [c_1]: 2.015e-05
        [parameter_eliminate]: 3.28e-06
        [updatestate_depend_eliminate]: 7.75e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 2.285e-05
        [call_graph_tuple_transform]: 1.962e-05
        [tuple_list_get_item_eliminator]: 5.78997e-06
        [none_parameter_eliminate]: 1.65001e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 5.49e-06
[remove_dup_value]: 1.434e-05
[partial_unused_args_eliminate]: 2.04999e-06
[environ_conv]: 5.62001e-06
[add_recomputation]: 5.511e-05
[cse_after_recomputation]: 2.171e-05, [1]
    [Cycle 1]: 1.504e-05, [1]
        [cse]: 8.99e-06
[auto_monad_reorder]: 1.533e-05
[get_jit_bprop_graph]: 2.16e-06
[rewriter_after_jit_bprop_graph]: 6.93e-06
[opt_after_jit_grad]: 0.00053503
[symbol_engine_optimizer]: 8.19e-05, [1]
    [Cycle 1]: 7.35e-05, [6]
        [build]: 5.37001e-06
        [elim_shapecalc]: 8.42998e-06
        [elim_not_effective]: 1.546e-05
        [opt_reshape]: 5.86e-06
        [fold_const_symbol]: 9.24e-06
        [renormalize]: 7.7e-07
[validate]: 4.2e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00702365
[execute]: 7.54002e-06
Sums
    bootstrap                            :     0.000342s :  1.97%
    type_inference                       :     0.005431s : 31.37%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000072s :  0.42%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000043s :  0.25%
    rewriter_before_opt_a                :     0.000076s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.43%
    jit_opt_a.loop_unroll                :     0.000035s :  0.20%
    jit_opt_a.a_1                        :     0.000694s :  4.01%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001016s :  5.87%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.13%
    jit_opt_a.cse                        :     0.000042s :  0.24%
    jit_opt_a.replace_applicator         :     0.000021s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.09%
    rewriter_after_opt_a                 :     0.000039s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000744s :  4.30%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000029s :  0.17%
    loop_unroll                          :     0.000461s :  2.66%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000055s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000535s :  3.09%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007024s : 40.57%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000257    23
 1.09% :     0.000003s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 2.29% :     0.000006s :      3: substitution.graph_param_transform
85.53% :     0.000220s :      4: substitution.inline
 1.58% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.89% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.94% :     0.000005s :      2: substitution.replace_old_param
 5.20% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005354     2
81.18% :     0.004347s :      1: type_inference.infer
18.82% :     0.001007s :      1: type_inference.specialize
------[replace.]   0.000079     6
69.55% :     0.000055s :      4: replace.inline
30.45% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000229     6
94.65% :     0.000217s :      4: match.inline
 5.35% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.51% :     0.000002s :     10: predicate.addn_zero_filter
 2.21% :     0.000003s :     10: predicate.arithmetic_simplify
 1.46% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000002s :     10: predicate.depend_value_elim
 1.26% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.06% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.72% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.68% :     0.000007s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.96% :     0.000001s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.88% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.84% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 2.18% :     0.000003s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.39% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.69% :     0.000002s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.20% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.79% :     0.000009s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.95% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000743    12
46.86% :     0.000348s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.25% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.90% :     0.000341s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024039    76
 0.24% :     0.000058s :      1: add_recomputation
 0.31% :     0.000076s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.50% :     0.000360s :      1: bootstrap
 0.13% :     0.000031s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.04% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
26.89% :     0.006465s :      1: jit_opt_a
 0.64% :     0.000154s :      1: jit_opt_after_cconv
 0.24% :     0.000057s :      1: jit_opt_b
 1.95% :     0.000470s :      1: loop_unroll
 3.13% :     0.000753s :      1: mutable_eliminate
 3.86% :     0.000927s :     26: opt.transform.jit_opt_a
 0.20% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.27% :     0.000545s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000045s :      1: py_interpret_to_execute
 0.07% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.15% :     0.000517s :      1: renormalize.infer
 2.03% :     0.000489s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000043s :      1: rewriter_after_opt_a
 0.33% :     0.000079s :      1: rewriter_before_opt_a
 0.35% :     0.000085s :      1: symbol_engine_optimizer
29.27% :     0.007036s :      1: task_emit
22.66% :     0.005446s :      1: type_inference
 0.27% :     0.000065s :      1: validate


TotalTime = 0.0330208, [33]
[bootstrap]: 0.00020762
[type_inference]: 0.0124671
[event_method]: 7.46e-05
[auto_monad]: 8.609e-05
[graph_reusing]: 5.37001e-06
[pre_auto_parallel]: 2.27001e-06
[py_interpret_to_execute]: 2.441e-05
[rewriter_before_opt_a]: 9.551e-05
[expand_dump_flag]: 2.02999e-06
[jit_opt_a]: 0.0112582, [3]
    [Cycle 1]: 0.00558806, [27]
        [switch_simplify]: 0.0001313
        [loop_unroll]: 4.679e-05
        [a_1]: 0.00091199
        [with_stream_mark]: 2.998e-05
        [recompute_prepare]: 1.873e-05
        [updatestate_depend_eliminate]: 8.88002e-06
        [updatestate_assign_eliminate]: 6.69001e-06
        [updatestate_loads_eliminate]: 6.33e-06
        [parameter_eliminate]: 2.81e-06
        [specialize_transform]: 1.445e-05
        [updatestate_useless_node_eliminater]: 1.285e-05
        [accelerated_algorithm]: 1.341e-05
        [meta_shard_fg_expand]: 3.65e-06
        [get_grad_eliminate_]: 1.199e-05
        [merge_forward]: 8.52e-06
        [cell_reuse_recompute_pass]: 9.29984e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.907e-05
        [j_node_and_user_rematch]: 2.401e-05
        [meta_fg_expand]: 0.00142855
        [replace_old_param]: 4.995e-05
        [inline_without_move]: 5.058e-05
        [renormalize]: 0.00228428
        [add_forward_monad_depend]: 9.07001e-06
        [auto_monad_grad]: 6.04999e-06
        [auto_monad_eliminator]: 4.631e-05
        [cse]: 0.0001487
        [replace_applicator]: 6.007e-05
    [Cycle 2]: 0.00189038, [27]
        [switch_simplify]: 3.512e-05
        [loop_unroll]: 3.288e-05
        [a_1]: 0.00098995
        [with_stream_mark]: 1.206e-05
        [recompute_prepare]: 7.61999e-06
        [updatestate_depend_eliminate]: 3.23998e-06
        [updatestate_assign_eliminate]: 2.94001e-06
        [updatestate_loads_eliminate]: 2.75002e-06
        [parameter_eliminate]: 1.23002e-06
        [specialize_transform]: 6.66e-06
        [updatestate_useless_node_eliminater]: 5.79e-06
        [accelerated_algorithm]: 6.14001e-06
        [meta_shard_fg_expand]: 1.41002e-06
        [get_grad_eliminate_]: 5.49e-06
        [merge_forward]: 3.18e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.263e-05
        [j_node_and_user_rematch]: 1.015e-05
        [meta_fg_expand]: 6.849e-05
        [replace_old_param]: 9.91e-06
        [inline_without_move]: 5.83997e-06
        [renormalize]: 0.0004687
        [add_forward_monad_depend]: 4.11001e-06
        [auto_monad_grad]: 1.12e-06
        [auto_monad_eliminator]: 1.218e-05
        [cse]: 2.004e-05
        [replace_applicator]: 1.154e-05
    [Cycle 3]: 0.00035785, [27]
        [switch_simplify]: 6.34999e-06
        [loop_unroll]: 6.01998e-06
        [a_1]: 0.00011462
        [with_stream_mark]: 8.89e-06
        [recompute_prepare]: 6.05002e-06
        [updatestate_depend_eliminate]: 3.24001e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 2.51998e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 6.41998e-06
        [updatestate_useless_node_eliminater]: 5.87001e-06
        [accelerated_algorithm]: 5.70001e-06
        [meta_shard_fg_expand]: 1.26997e-06
        [get_grad_eliminate_]: 5.76003e-06
        [merge_forward]: 2.92002e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.431e-05
        [j_node_and_user_rematch]: 9.57999e-06
        [meta_fg_expand]: 1.89e-06
        [replace_old_param]: 8.21002e-06
        [inline_without_move]: 5.87999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 9.09989e-07
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 6.49001e-06
        [cse]: 1.314e-05
        [replace_applicator]: 6.13002e-06
[py_interpret_to_execute_after_opt_a]: 1.165e-05
[rewriter_after_opt_a]: 3.853e-05
[convert_after_rewriter]: 7.33e-06
[order_py_execute_after_rewriter]: 5.74999e-06
[mutable_eliminate]: 0.00053017
[jit_opt_b]: 5.288e-05, [1]
    [Cycle 1]: 4.547e-05, [2]
        [frontend_op_eliminate]: 1.74e-05
        [inline_after_opt_a]: 1.676e-05
[cconv]: 2.051e-05
[loop_unroll]: 0.00042162
[jit_opt_after_cconv]: 0.00015005, [1]
    [Cycle 1]: 0.00014343, [11]
        [c_1]: 2.452e-05
        [parameter_eliminate]: 2.27001e-06
        [updatestate_depend_eliminate]: 6.11e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.62001e-06
        [cse]: 2.089e-05
        [call_graph_tuple_transform]: 1.916e-05
        [tuple_list_get_item_eliminator]: 6.17001e-06
        [none_parameter_eliminate]: 1.92001e-06
        [renormalize]: 5.00004e-07
        [switch_simplify]: 6.66e-06
[remove_dup_value]: 1.557e-05
[partial_unused_args_eliminate]: 2.26998e-06
[environ_conv]: 5.55001e-06
[add_recomputation]: 4.857e-05
[cse_after_recomputation]: 2.47e-05, [1]
    [Cycle 1]: 1.921e-05, [1]
        [cse]: 1.356e-05
[auto_monad_reorder]: 2.431e-05
[get_jit_bprop_graph]: 1.61998e-06
[rewriter_after_jit_bprop_graph]: 5.37001e-06
[opt_after_jit_grad]: 0.00049361
[symbol_engine_optimizer]: 8.119e-05, [1]
    [Cycle 1]: 7.483e-05, [6]
        [build]: 3.6e-06
        [elim_shapecalc]: 8.66997e-06
        [elim_not_effective]: 1.583e-05
        [opt_reshape]: 6.61e-06
        [fold_const_symbol]: 1.104e-05
        [renormalize]: 5.00004e-07
[validate]: 3.544e-05
[backend_pass]: 1.02998e-06
[task_emit]: 0.00662296
[execute]: 6.96999e-06
Sums
    bootstrap                            :     0.000208s :  0.72%
    type_inference                       :     0.012467s : 43.28%
    event_method                         :     0.000075s :  0.26%
    auto_monad                           :     0.000086s :  0.30%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000024s :  0.08%
    rewriter_before_opt_a                :     0.000096s :  0.33%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000173s :  0.60%
    jit_opt_a.loop_unroll                :     0.000086s :  0.30%
    jit_opt_a.a_1                        :     0.002017s :  7.00%
    jit_opt_a.with_stream_mark           :     0.000051s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000028s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000025s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.08%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000056s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000044s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001499s :  5.20%
    jit_opt_a.replace_old_param          :     0.000068s :  0.24%
    jit_opt_a.inline_without_move        :     0.000062s :  0.22%
    jit_opt_a.renormalize                :     0.002753s :  9.56%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000065s :  0.23%
    jit_opt_a.cse                        :     0.000182s :  0.63%
    jit_opt_a.replace_applicator         :     0.000078s :  0.27%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000039s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000530s :  1.84%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.06%
    cconv                                :     0.000021s :  0.07%
    loop_unroll                          :     0.000422s :  1.46%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000016s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000049s :  0.17%
    cse_after_recomputation.cse          :     0.000014s :  0.05%
    auto_monad_reorder                   :     0.000024s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000494s :  1.71%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006623s : 22.99%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000496   113
 0.50% :     0.000002s :      3: substitution.elim_not_effective
 0.47% :     0.000002s :      3: substitution.fold_const_symbol
 0.99% :     0.000005s :      4: substitution.graph_param_transform
64.38% :     0.000319s :     14: substitution.inline
 3.21% :     0.000016s :      2: substitution.inline_without_move
 1.84% :     0.000009s :     14: substitution.j_node_and_user_rematch
 1.93% :     0.000010s :      7: substitution.minmaximum_grad
 1.34% :     0.000007s :      5: substitution.partial_eliminate
 2.24% :     0.000011s :     14: substitution.remove_not_recompute_node
 4.95% :     0.000025s :     10: substitution.replace_applicator
 1.46% :     0.000007s :      7: substitution.replace_old_param
 0.59% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.54% :     0.000008s :      1: substitution.switch_simplify
 3.98% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.73% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.85% :     0.000039s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012390     2
90.15% :     0.011170s :      1: type_inference.infer
 9.85% :     0.001221s :      1: type_inference.specialize
------[replace.]   0.000184    22
53.35% :     0.000098s :     14: replace.inline
19.45% :     0.000036s :      1: replace.switch_simplify
27.20% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000339    22
91.91% :     0.000312s :     14: match.inline
 2.06% :     0.000007s :      1: match.switch_simplify
 6.03% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000377  2634
 1.64% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.47% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.42% :     0.000005s :     44: predicate.addn_check_dump
 1.63% :     0.000006s :     44: predicate.addn_zero_filter
 1.96% :     0.000007s :     44: predicate.arithmetic_simplify
 1.56% :     0.000006s :     44: predicate.cast_eliminate
 0.17% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.39% :     0.000005s :     44: predicate.compare_switch_simplify
 1.61% :     0.000006s :     44: predicate.depend_value_elim
 1.44% :     0.000005s :     44: predicate.dict_get_item_const_eliminator
 1.52% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.49% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.29% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      4: predicate.elim_not_effective
 0.21% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.46% :     0.000005s :     44: predicate.environ_add_const_eliminate
 1.43% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.41% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.51% :     0.000006s :     44: predicate.environ_get_eliminate
 1.41% :     0.000005s :     44: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.80% :     0.000003s :     19: predicate.get_grad_eliminate
 0.16% :     0.000001s :      4: predicate.graph_param_transform
 4.49% :     0.000017s :     73: predicate.inline
 1.74% :     0.000007s :     40: predicate.inline_without_move
 0.36% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.94% :     0.000004s :     19: predicate.less_batch_normalization
 1.77% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.94% :     0.000007s :     55: predicate.load_eliminater
 0.57% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.47% :     0.000013s :     91: predicate.loop_unroll_before_grad
 1.90% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.48% :     0.000006s :     44: predicate.merge_addn
 1.53% :     0.000006s :     44: predicate.minmaximum_grad
 0.53% :     0.000002s :      4: predicate.mutable_eliminate
 0.22% :     0.000001s :      4: predicate.opt_reshape
 2.30% :     0.000009s :     55: predicate.partial_eliminate
 1.88% :     0.000007s :     44: predicate.print_const_string_wrapper
 1.94% :     0.000007s :     44: predicate.reduce_eliminate
 1.80% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.40% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.37% :     0.000009s :     97: predicate.replace_applicator
 0.85% :     0.000003s :     40: predicate.replace_old_param
 0.11% :     0.000000s :      4: predicate.reset_defer_inline
 1.66% :     0.000006s :     44: predicate.reshape_eliminate
 1.55% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.29% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.56% :     0.000006s :     44: predicate.same_eliminate
 0.45% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.40% :     0.000002s :      8: predicate.special_op_eliminate
 0.93% :     0.000004s :     19: predicate.specialize_transform
 1.66% :     0.000006s :     44: predicate.split_environ_get_set_with_tuple_value
 1.59% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.16% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.81% :     0.000011s :     65: predicate.switch_defer_inline
 2.58% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.59% :     0.000025s :    162: predicate.switch_simplify
 1.55% :     0.000006s :     44: predicate.tile_eliminate
 1.49% :     0.000006s :     44: predicate.transpose_eliminate
 1.79% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.60% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 3.18% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.80% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.89% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000006s :     44: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.22% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001476    32
62.23% :     0.000918s :     14: func_graph_cloner_run.FuncGraphClonerGraph
37.77% :     0.000557s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038568    91
 0.13% :     0.000051s :      1: add_recomputation
 0.24% :     0.000091s :      1: auto_monad
 0.07% :     0.000027s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.57% :     0.000220s :      1: bootstrap
 0.06% :     0.000023s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000027s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.21% :     0.000081s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.01% :     0.000004s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
29.20% :     0.011261s :      1: jit_opt_a
 0.40% :     0.000153s :      1: jit_opt_after_cconv
 0.14% :     0.000056s :      1: jit_opt_b
 1.11% :     0.000429s :      1: loop_unroll
 1.40% :     0.000539s :      1: mutable_eliminate
 6.93% :     0.002672s :     39: opt.transform.jit_opt_a
 0.14% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000028s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.30% :     0.000502s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000027s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 3.90% :     0.001503s :      2: renormalize.infer
 3.21% :     0.001236s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000041s :      1: rewriter_after_opt_a
 0.25% :     0.000098s :      1: rewriter_before_opt_a
 0.22% :     0.000084s :      1: symbol_engine_optimizer
17.20% :     0.006633s :      1: task_emit
32.36% :     0.012481s :      1: type_inference
 0.15% :     0.000058s :      1: validate


TotalTime = 0.0141238, [30]
[bootstrap]: 0.00032019
[type_inference]: 0.00467052
[event_method]: 1.841e-05
[auto_monad]: 6.624e-05
[graph_reusing]: 5.71e-06
[pre_auto_parallel]: 2.81e-06
[py_interpret_to_execute]: 3.517e-05
[rewriter_before_opt_a]: 7.114e-05
[expand_dump_flag]: 2.94999e-06
[jit_opt_a]: 0.00606753, [2]
    [Cycle 1]: 0.0019612, [27]
        [switch_simplify]: 5.192e-05
        [loop_unroll]: 3.03e-05
        [a_1]: 0.00080402
        [with_stream_mark]: 1.942e-05
        [recompute_prepare]: 8.48999e-06
        [updatestate_depend_eliminate]: 3.16001e-06
        [updatestate_assign_eliminate]: 2.43002e-06
        [updatestate_loads_eliminate]: 2.25002e-06
        [parameter_eliminate]: 1.36002e-06
        [specialize_transform]: 6.42001e-06
        [updatestate_useless_node_eliminater]: 5.60001e-06
        [accelerated_algorithm]: 6.04001e-06
        [meta_shard_fg_expand]: 2.22999e-06
        [get_grad_eliminate_]: 5.37001e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.335e-05
        [j_node_and_user_rematch]: 9.47999e-06
        [meta_fg_expand]: 2.01003e-06
        [replace_old_param]: 9.32001e-06
        [inline_without_move]: 5.51e-06
        [renormalize]: 0.00073812
        [add_forward_monad_depend]: 6.17999e-06
        [auto_monad_grad]: 3.26999e-06
        [auto_monad_eliminator]: 1.292e-05
        [cse]: 2.125e-05
        [replace_applicator]: 1.238e-05
    [Cycle 2]: 0.0004029, [27]
        [switch_simplify]: 6.74999e-06
        [loop_unroll]: 6.02001e-06
        [a_1]: 9.607e-05
        [with_stream_mark]: 1.153e-05
        [recompute_prepare]: 5.32999e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [parameter_eliminate]: 1.75001e-06
        [specialize_transform]: 5.56e-06
        [updatestate_useless_node_eliminater]: 5.31002e-06
        [accelerated_algorithm]: 6.07999e-06
        [meta_shard_fg_expand]: 1.58002e-06
        [get_grad_eliminate_]: 5.26002e-06
        [merge_forward]: 2.76999e-06
        [cell_reuse_recompute_pass]: 1.95001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.487e-05
        [j_node_and_user_rematch]: 8.67998e-06
        [meta_fg_expand]: 1.72999e-06
        [replace_old_param]: 8.07e-06
        [inline_without_move]: 1.171e-05
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 2.51e-06
        [auto_monad_grad]: 1.34e-06
        [auto_monad_eliminator]: 8.32e-06
        [cse]: 1.5e-05
        [replace_applicator]: 7.82998e-06
[py_interpret_to_execute_after_opt_a]: 3.031e-05
[rewriter_after_opt_a]: 3.981e-05
[convert_after_rewriter]: 7.43999e-06
[order_py_execute_after_rewriter]: 5.89e-06
[mutable_eliminate]: 0.00097038
[jit_opt_b]: 6.228e-05, [1]
    [Cycle 1]: 5.205e-05, [2]
        [frontend_op_eliminate]: 1.892e-05
        [inline_after_opt_a]: 1.919e-05
[cconv]: 3.527e-05
[loop_unroll]: 0.00054173
[jit_opt_after_cconv]: 0.00016489, [1]
    [Cycle 1]: 0.00015715, [11]
        [c_1]: 2.2e-05
        [parameter_eliminate]: 3.93001e-06
        [updatestate_depend_eliminate]: 7.11999e-06
        [updatestate_assign_eliminate]: 2.89999e-06
        [updatestate_loads_eliminate]: 2.78998e-06
        [cse]: 2.746e-05
        [call_graph_tuple_transform]: 2.436e-05
        [tuple_list_get_item_eliminator]: 5.66998e-06
        [none_parameter_eliminate]: 1.77999e-06
        [renormalize]: 8.2e-07
        [switch_simplify]: 5.64998e-06
[remove_dup_value]: 1.428e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 5.25001e-06
[add_recomputation]: 5.147e-05
[cse_after_recomputation]: 2.237e-05, [1]
    [Cycle 1]: 1.615e-05, [1]
        [cse]: 9.95002e-06
[auto_monad_reorder]: 1.581e-05
[get_jit_bprop_graph]: 2.14999e-06
[rewriter_after_jit_bprop_graph]: 8.02e-06
[opt_after_jit_grad]: 0.00057551
[symbol_engine_optimizer]: 7.905e-05, [1]
    [Cycle 1]: 7.067e-05, [6]
        [build]: 5.29e-06
        [elim_shapecalc]: 7.68001e-06
        [elim_not_effective]: 1.357e-05
        [opt_reshape]: 6.26e-06
        [fold_const_symbol]: 8.89e-06
        [renormalize]: 6.00005e-07
[validate]: 4e-05
Sums
    bootstrap                            :     0.000320s :  3.28%
    type_inference                       :     0.004671s : 47.83%
    event_method                         :     0.000018s :  0.19%
    auto_monad                           :     0.000066s :  0.68%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.36%
    rewriter_before_opt_a                :     0.000071s :  0.73%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000059s :  0.60%
    jit_opt_a.loop_unroll                :     0.000036s :  0.37%
    jit_opt_a.a_1                        :     0.000900s :  9.22%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.18%
    jit_opt_a.inline_without_move        :     0.000017s :  0.18%
    jit_opt_a.renormalize                :     0.000738s :  7.56%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.22%
    jit_opt_a.cse                        :     0.000036s :  0.37%
    jit_opt_a.replace_applicator         :     0.000020s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000030s :  0.31%
    rewriter_after_opt_a                 :     0.000040s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000970s :  9.94%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.20%
    cconv                                :     0.000035s :  0.36%
    loop_unroll                          :     0.000542s :  5.55%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000027s :  0.28%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.25%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000014s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000051s :  0.53%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.08%
    opt_after_jit_grad                   :     0.000576s :  5.89%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.41%


Time group info:
------[substitution.]   0.000394    23
 0.62% :     0.000002s :      2: substitution.elim_not_effective
 0.31% :     0.000001s :      2: substitution.fold_const_symbol
 1.79% :     0.000007s :      3: substitution.graph_param_transform
90.70% :     0.000358s :      4: substitution.inline
 0.84% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.22% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.21% :     0.000005s :      2: substitution.replace_old_param
 3.32% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004604     2
81.61% :     0.003757s :      1: type_inference.infer
18.39% :     0.000847s :      1: type_inference.specialize
------[replace.]   0.000102     6
74.28% :     0.000076s :      4: replace.inline
25.72% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000367     6
96.82% :     0.000355s :      4: match.inline
 3.18% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.23% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000002s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.30% :     0.000000s :      3: predicate.fold_const_symbol
 1.13% :     0.000001s :      6: predicate.get_grad_eliminate
 0.57% :     0.000001s :      3: predicate.graph_param_transform
 5.01% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.78% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.35% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000002s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 2.52% :     0.000003s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.04% :     0.000003s :     15: predicate.partial_eliminate
 1.33% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.32% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.57% :     0.000002s :     10: predicate.reshape_eliminate
 1.32% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.55% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.70% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.30% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.41% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.85% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000609    12
49.77% :     0.000303s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.94% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.29% :     0.000264s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016060    72
 0.34% :     0.000055s :      1: add_recomputation
 0.44% :     0.000070s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.08% :     0.000333s :      1: bootstrap
 0.24% :     0.000039s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
37.81% :     0.006072s :      1: jit_opt_a
 1.05% :     0.000168s :      1: jit_opt_after_cconv
 0.41% :     0.000066s :      1: jit_opt_b
 3.43% :     0.000551s :      1: loop_unroll
 6.14% :     0.000986s :      1: mutable_eliminate
 7.01% :     0.001127s :     26: opt.transform.jit_opt_a
 0.34% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000030s :      4: opt.transform.jit_opt_b
 0.09% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.13% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.65% :     0.000586s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000038s :      1: py_interpret_to_execute
 0.21% :     0.000033s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 2.45% :     0.000393s :      1: renormalize.infer
 2.10% :     0.000338s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000043s :      1: rewriter_after_opt_a
 0.46% :     0.000074s :      1: rewriter_before_opt_a
 0.51% :     0.000082s :      1: symbol_engine_optimizer
29.18% :     0.004686s :      1: type_inference


TotalTime = 0.0264418, [33]
[bootstrap]: 0.0003291
[type_inference]: 0.00442475
[event_method]: 1.684e-05
[auto_monad]: 7.663e-05
[graph_reusing]: 5.71998e-06
[pre_auto_parallel]: 2.44001e-06
[py_interpret_to_execute]: 3.117e-05
[rewriter_before_opt_a]: 7.277e-05
[expand_dump_flag]: 3.68e-06
[jit_opt_a]: 0.0133807, [2]
    [Cycle 1]: 0.00908822, [27]
        [switch_simplify]: 4.767e-05
        [loop_unroll]: 3.545e-05
        [a_1]: 0.00752269
        [with_stream_mark]: 2.125e-05
        [recompute_prepare]: 1.188e-05
        [updatestate_depend_eliminate]: 3.9e-06
        [updatestate_assign_eliminate]: 2.94001e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 2.46998e-06
        [specialize_transform]: 6.85002e-06
        [updatestate_useless_node_eliminater]: 6.53e-06
        [accelerated_algorithm]: 6.72002e-06
        [meta_shard_fg_expand]: 3.23e-06
        [get_grad_eliminate_]: 6.33e-06
        [merge_forward]: 4.25e-06
        [cell_reuse_recompute_pass]: 2.19001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.609e-05
        [j_node_and_user_rematch]: 1.134e-05
        [meta_fg_expand]: 3.25e-06
        [replace_old_param]: 3.116e-05
        [inline_without_move]: 6.56999e-06
        [renormalize]: 0.00103437
        [add_forward_monad_depend]: 8.33999e-06
        [auto_monad_grad]: 2.76999e-06
        [auto_monad_eliminator]: 1.629e-05
        [cse]: 2.349e-05
        [replace_applicator]: 1.673e-05
    [Cycle 2]: 0.00037979, [27]
        [switch_simplify]: 7.03998e-06
        [loop_unroll]: 6.00002e-06
        [a_1]: 0.00011166
        [with_stream_mark]: 1.255e-05
        [recompute_prepare]: 5.90002e-06
        [updatestate_depend_eliminate]: 3.45998e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [parameter_eliminate]: 1.35999e-06
        [specialize_transform]: 5.89999e-06
        [updatestate_useless_node_eliminater]: 5.19998e-06
        [accelerated_algorithm]: 6.81001e-06
        [meta_shard_fg_expand]: 1.80001e-06
        [get_grad_eliminate_]: 5.07e-06
        [merge_forward]: 3.23e-06
        [cell_reuse_recompute_pass]: 3.04999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.498e-05
        [j_node_and_user_rematch]: 9.19e-06
        [meta_fg_expand]: 2.04999e-06
        [replace_old_param]: 8.41002e-06
        [inline_without_move]: 4.94e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.29998e-06
        [auto_monad_grad]: 6.79982e-07
        [auto_monad_eliminator]: 5.63002e-06
        [cse]: 1.212e-05
        [replace_applicator]: 5.46002e-06
[py_interpret_to_execute_after_opt_a]: 1.363e-05
[rewriter_after_opt_a]: 3.632e-05
[convert_after_rewriter]: 7.33e-06
[order_py_execute_after_rewriter]: 4.74e-06
[mutable_eliminate]: 0.00091329
[jit_opt_b]: 5.511e-05, [1]
    [Cycle 1]: 4.662e-05, [2]
        [frontend_op_eliminate]: 1.798e-05
        [inline_after_opt_a]: 1.626e-05
[cconv]: 2.671e-05
[loop_unroll]: 0.0004286
[jit_opt_after_cconv]: 0.00014943, [1]
    [Cycle 1]: 0.00014208, [11]
        [c_1]: 2.228e-05
        [parameter_eliminate]: 2.68e-06
        [updatestate_depend_eliminate]: 5.71998e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 2.103e-05
        [call_graph_tuple_transform]: 2.06e-05
        [tuple_list_get_item_eliminator]: 5.73002e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 3.99974e-07
        [switch_simplify]: 5.79e-06
[remove_dup_value]: 1.474e-05
[partial_unused_args_eliminate]: 2.20002e-06
[environ_conv]: 5.30999e-06
[add_recomputation]: 5.045e-05
[cse_after_recomputation]: 2.143e-05, [1]
    [Cycle 1]: 1.571e-05, [1]
        [cse]: 9.90002e-06
[auto_monad_reorder]: 1.527e-05
[get_jit_bprop_graph]: 2.32001e-06
[rewriter_after_jit_bprop_graph]: 5.76e-06
[opt_after_jit_grad]: 0.00048612
[symbol_engine_optimizer]: 0.00013945, [1]
    [Cycle 1]: 0.00013141, [6]
        [build]: 5.39e-06
        [elim_shapecalc]: 8.57e-06
        [elim_not_effective]: 1.458e-05
        [opt_reshape]: 6.21e-06
        [fold_const_symbol]: 1.036e-05
        [renormalize]: 5.00004e-07
[validate]: 4.382e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.00547432
[execute]: 4.53999e-06
Sums
    bootstrap                            :     0.000329s :  1.51%
    type_inference                       :     0.004425s : 20.31%
    event_method                         :     0.000017s :  0.08%
    auto_monad                           :     0.000077s :  0.35%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000031s :  0.14%
    rewriter_before_opt_a                :     0.000073s :  0.33%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.25%
    jit_opt_a.loop_unroll                :     0.000041s :  0.19%
    jit_opt_a.a_1                        :     0.007634s : 35.04%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000014s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000007s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.02%
    jit_opt_a.replace_old_param          :     0.000040s :  0.18%
    jit_opt_a.inline_without_move        :     0.000012s :  0.05%
    jit_opt_a.renormalize                :     0.001034s :  4.75%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.10%
    jit_opt_a.cse                        :     0.000036s :  0.16%
    jit_opt_a.replace_applicator         :     0.000022s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.06%
    rewriter_after_opt_a                 :     0.000036s :  0.17%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000913s :  4.19%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.07%
    cconv                                :     0.000027s :  0.12%
    loop_unroll                          :     0.000429s :  1.97%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000050s :  0.23%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000486s :  2.23%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.20%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005474s : 25.13%
    execute                              :     0.000005s :  0.02%


Time group info:
------[substitution.]   0.000251    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000002s :      2: substitution.fold_const_symbol
 2.38% :     0.000006s :      3: substitution.graph_param_transform
82.10% :     0.000206s :      4: substitution.inline
 1.70% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.93% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000005s :      2: substitution.replace_old_param
 8.20% :     0.000021s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004363     2
83.16% :     0.003629s :      1: type_inference.infer
16.84% :     0.000735s :      1: type_inference.specialize
------[replace.]   0.000124     6
62.75% :     0.000078s :      4: replace.inline
37.25% :     0.000046s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000220     6
91.43% :     0.000202s :      4: match.inline
 8.57% :     0.000019s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000164   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.98% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.80% :     0.000001s :     10: predicate.addn_check_dump
 1.44% :     0.000002s :     10: predicate.addn_zero_filter
 3.12% :     0.000005s :     10: predicate.arithmetic_simplify
 1.75% :     0.000003s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.81% :     0.000001s :     10: predicate.compare_switch_simplify
 1.53% :     0.000003s :     10: predicate.depend_value_elim
 0.91% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.23% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.06% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000002s :     10: predicate.environ_get_eliminate
 0.98% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.17% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 5.24% :     0.000009s :     22: predicate.inline
 0.76% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000002s :      6: predicate.less_batch_normalization
 1.69% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.90% :     0.000003s :     15: predicate.load_eliminater
 1.00% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.77% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.91% :     0.000001s :     10: predicate.merge_addn
 0.83% :     0.000001s :     10: predicate.minmaximum_grad
 1.63% :     0.000003s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 1.94% :     0.000003s :     15: predicate.partial_eliminate
 1.65% :     0.000003s :     10: predicate.print_const_string_wrapper
 1.82% :     0.000003s :     10: predicate.reduce_eliminate
 1.83% :     0.000003s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.25% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.05% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 0.82% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.32% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.57% :     0.000004s :     16: predicate.switch_defer_inline
 7.90% :     0.000013s :     16: predicate.switch_layer_defer_inline
 5.86% :     0.000010s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.28% :     0.000002s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.49% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 2.07% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.20% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.42% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.88% :     0.000003s :     10: predicate.value_based_eliminate
 0.34% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000610    12
40.80% :     0.000249s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 9.37% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.82% :     0.000304s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.035480    76
 0.15% :     0.000053s :      1: add_recomputation
 0.23% :     0.000080s :      1: auto_monad
 0.05% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.98% :     0.000346s :      1: bootstrap
 0.08% :     0.000029s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000024s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.06% :     0.000020s :      1: event_method
 0.02% :     0.000008s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
37.73% :     0.013385s :      1: jit_opt_a
 0.43% :     0.000152s :      1: jit_opt_after_cconv
 0.16% :     0.000058s :      1: jit_opt_b
 1.23% :     0.000436s :      1: loop_unroll
 2.60% :     0.000923s :      1: mutable_eliminate
22.24% :     0.007889s :     26: opt.transform.jit_opt_a
 0.14% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000027s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.40% :     0.000495s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.09% :     0.000034s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000017s :      1: remove_dup_value
 1.55% :     0.000549s :      1: renormalize.infer
 1.34% :     0.000475s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000040s :      1: rewriter_after_opt_a
 0.21% :     0.000075s :      1: rewriter_before_opt_a
 0.40% :     0.000142s :      1: symbol_engine_optimizer
15.46% :     0.005484s :      1: task_emit
12.50% :     0.004435s :      1: type_inference
 0.19% :     0.000066s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.105.056 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc39c10> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.105.699 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0161318, [30]
[bootstrap]: 0.0003024
[type_inference]: 0.00531961
[event_method]: 2.197e-05
[auto_monad]: 7.604e-05
[graph_reusing]: 6.29999e-06
[pre_auto_parallel]: 2.98e-06
[py_interpret_to_execute]: 4.365e-05
[rewriter_before_opt_a]: 7.977e-05
[expand_dump_flag]: 4.23001e-06
[jit_opt_a]: 0.00713106, [2]
    [Cycle 1]: 0.00215764, [27]
        [switch_simplify]: 7.348e-05
        [loop_unroll]: 3.525e-05
        [a_1]: 0.00067287
        [with_stream_mark]: 3.013e-05
        [recompute_prepare]: 1.172e-05
        [updatestate_depend_eliminate]: 4.60999e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 2.69001e-06
        [parameter_eliminate]: 2.69999e-06
        [specialize_transform]: 7.18e-06
        [updatestate_useless_node_eliminater]: 6.11998e-06
        [accelerated_algorithm]: 7.92e-06
        [meta_shard_fg_expand]: 2.88e-06
        [get_grad_eliminate_]: 6.23998e-06
        [merge_forward]: 4.13999e-06
        [cell_reuse_recompute_pass]: 2.12001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.718e-05
        [j_node_and_user_rematch]: 1.016e-05
        [meta_fg_expand]: 2.49001e-06
        [replace_old_param]: 1.137e-05
        [inline_without_move]: 5.39e-06
        [renormalize]: 0.00091673
        [add_forward_monad_depend]: 1.123e-05
        [auto_monad_grad]: 3.23998e-06
        [auto_monad_eliminator]: 1.974e-05
        [cse]: 3.361e-05
        [replace_applicator]: 2.04e-05
    [Cycle 2]: 0.00038202, [27]
        [switch_simplify]: 6.23e-06
        [loop_unroll]: 5.71e-06
        [a_1]: 9.534e-05
        [with_stream_mark]: 1.705e-05
        [recompute_prepare]: 5.49998e-06
        [updatestate_depend_eliminate]: 3.73999e-06
        [updatestate_assign_eliminate]: 3.04001e-06
        [updatestate_loads_eliminate]: 2.32001e-06
        [parameter_eliminate]: 1.55001e-06
        [specialize_transform]: 5.43002e-06
        [updatestate_useless_node_eliminater]: 4.86997e-06
        [accelerated_algorithm]: 5.44e-06
        [meta_shard_fg_expand]: 1.76003e-06
        [get_grad_eliminate_]: 5.02e-06
        [merge_forward]: 3.91999e-06
        [cell_reuse_recompute_pass]: 2.80002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.99e-05
        [j_node_and_user_rematch]: 8.87e-06
        [meta_fg_expand]: 2.02999e-06
        [replace_old_param]: 9.87999e-06
        [inline_without_move]: 5.61e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 2.67001e-06
        [auto_monad_grad]: 1.12e-06
        [auto_monad_eliminator]: 7.18998e-06
        [cse]: 1.319e-05
        [replace_applicator]: 6.38e-06
[py_interpret_to_execute_after_opt_a]: 1.828e-05
[rewriter_after_opt_a]: 4.508e-05
[convert_after_rewriter]: 8.48999e-06
[order_py_execute_after_rewriter]: 6.58998e-06
[mutable_eliminate]: 0.00099805
[jit_opt_b]: 6.317e-05, [1]
    [Cycle 1]: 5.195e-05, [2]
        [frontend_op_eliminate]: 1.842e-05
        [inline_after_opt_a]: 1.837e-05
[cconv]: 3.922e-05
[loop_unroll]: 0.0005943
[jit_opt_after_cconv]: 0.00019463, [1]
    [Cycle 1]: 0.00018607, [11]
        [c_1]: 2.255e-05
        [parameter_eliminate]: 7.43e-06
        [updatestate_depend_eliminate]: 1.195e-05
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [cse]: 3.92e-05
        [call_graph_tuple_transform]: 2.838e-05
        [tuple_list_get_item_eliminator]: 5.59998e-06
        [none_parameter_eliminate]: 2.08998e-06
        [renormalize]: 1.15999e-06
        [switch_simplify]: 5.85002e-06
[remove_dup_value]: 1.853e-05
[partial_unused_args_eliminate]: 2.83e-06
[environ_conv]: 6.56e-06
[add_recomputation]: 6.443e-05
[cse_after_recomputation]: 2.786e-05, [1]
    [Cycle 1]: 2.025e-05, [1]
        [cse]: 1.188e-05
[auto_monad_reorder]: 1.866e-05
[get_jit_bprop_graph]: 2.25002e-06
[rewriter_after_jit_bprop_graph]: 1.057e-05
[opt_after_jit_grad]: 0.00064346
[symbol_engine_optimizer]: 8.909e-05, [1]
    [Cycle 1]: 7.947e-05, [6]
        [build]: 6.09999e-06
        [elim_shapecalc]: 9.19e-06
        [elim_not_effective]: 1.775e-05
        [opt_reshape]: 6.69999e-06
        [fold_const_symbol]: 9.46998e-06
        [renormalize]: 9.29984e-07
[validate]: 5.117e-05
Sums
    bootstrap                            :     0.000302s :  2.80%
    type_inference                       :     0.005320s : 49.32%
    event_method                         :     0.000022s :  0.20%
    auto_monad                           :     0.000076s :  0.71%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000044s :  0.40%
    rewriter_before_opt_a                :     0.000080s :  0.74%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000080s :  0.74%
    jit_opt_a.loop_unroll                :     0.000041s :  0.38%
    jit_opt_a.a_1                        :     0.000768s :  7.12%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.44%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000013s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.10%
    jit_opt_a.merge_forward              :     0.000008s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000021s :  0.20%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000917s :  8.50%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.13%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.25%
    jit_opt_a.cse                        :     0.000047s :  0.43%
    jit_opt_a.replace_applicator         :     0.000027s :  0.25%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.17%
    rewriter_after_opt_a                 :     0.000045s :  0.42%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000007s :  0.06%
    mutable_eliminate                    :     0.000998s :  9.25%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.17%
    cconv                                :     0.000039s :  0.36%
    loop_unroll                          :     0.000594s :  5.51%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000012s :  0.11%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000039s :  0.36%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.26%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000019s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000007s :  0.06%
    add_recomputation                    :     0.000064s :  0.60%
    cse_after_recomputation.cse          :     0.000012s :  0.11%
    auto_monad_reorder                   :     0.000019s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.10%
    opt_after_jit_grad                   :     0.000643s :  5.97%
    symbol_engine_optimizer.build        :     0.000006s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000051s :  0.47%


Time group info:
------[substitution.]   0.000302    23
 0.73% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000002s :      2: substitution.fold_const_symbol
 2.50% :     0.000008s :      3: substitution.graph_param_transform
85.26% :     0.000258s :      4: substitution.inline
 1.21% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.15% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.37% :     0.000007s :      2: substitution.replace_old_param
 5.13% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005160     2
80.91% :     0.004175s :      1: type_inference.infer
19.09% :     0.000985s :      1: type_inference.specialize
------[replace.]   0.000085     6
65.82% :     0.000056s :      4: replace.inline
34.18% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000268     6
94.78% :     0.000254s :      4: match.inline
 5.22% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000139   691
 1.24% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.67% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 1.89% :     0.000003s :     10: predicate.arithmetic_simplify
 1.25% :     0.000002s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.24% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.02% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.29% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.56% :     0.000001s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     10: predicate.environ_get_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.10% :     0.000002s :      6: predicate.get_grad_eliminate
 0.40% :     0.000001s :      3: predicate.graph_param_transform
 7.05% :     0.000010s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.60% :     0.000002s :      6: predicate.less_batch_normalization
 1.31% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.49% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.20% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 0.92% :     0.000001s :     10: predicate.minmaximum_grad
 2.39% :     0.000003s :      3: predicate.mutable_eliminate
 0.66% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     10: predicate.reduce_eliminate
 1.28% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.74% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.41% :     0.000001s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.32% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.10% :     0.000002s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.14% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.49% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.12% :     0.000003s :     16: predicate.switch_defer_inline
 1.91% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.54% :     0.000010s :     53: predicate.switch_simplify
 1.36% :     0.000002s :     10: predicate.tile_eliminate
 1.36% :     0.000002s :     10: predicate.transpose_eliminate
 1.23% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.03% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.20% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.42% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.63% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.51% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000761    12
49.77% :     0.000379s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.74% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.49% :     0.000323s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018143    72
 0.37% :     0.000068s :      1: add_recomputation
 0.44% :     0.000080s :      1: auto_monad
 0.12% :     0.000022s :      1: auto_monad_reorder
 1.76% :     0.000319s :      1: bootstrap
 0.23% :     0.000043s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.17% :     0.000030s :      1: cse_after_recomputation
 0.06% :     0.000010s :      1: environ_conv
 0.15% :     0.000027s :      1: event_method
 0.04% :     0.000007s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
39.33% :     0.007136s :      1: jit_opt_a
 1.09% :     0.000198s :      1: jit_opt_after_cconv
 0.37% :     0.000067s :      1: jit_opt_b
 3.36% :     0.000609s :      1: loop_unroll
 5.63% :     0.001021s :      1: mutable_eliminate
 5.69% :     0.001032s :     26: opt.transform.jit_opt_a
 0.32% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000029s :      4: opt.transform.jit_opt_b
 0.10% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.15% :     0.000027s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000040s :      4: opt.transform.symbol_engine_opt
 3.64% :     0.000660s :      1: opt_after_jit_grad
 0.05% :     0.000009s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000046s :      1: py_interpret_to_execute
 0.12% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000021s :      1: remove_dup_value
 2.69% :     0.000488s :      1: renormalize.infer
 2.29% :     0.000415s :      1: renormalize.specialize
 0.08% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000049s :      1: rewriter_after_opt_a
 0.46% :     0.000084s :      1: rewriter_before_opt_a
 0.51% :     0.000092s :      1: symbol_engine_optimizer
29.44% :     0.005342s :      1: type_inference


TotalTime = 0.0339355, [33]
[bootstrap]: 0.0002488
[type_inference]: 0.0127999
[event_method]: 5.401e-05
[auto_monad]: 0.00010524
[graph_reusing]: 7.76001e-06
[pre_auto_parallel]: 2.84999e-06
[py_interpret_to_execute]: 2.876e-05
[rewriter_before_opt_a]: 0.00010087
[expand_dump_flag]: 3.34001e-06
[jit_opt_a]: 0.0104151, [3]
    [Cycle 1]: 0.00516048, [27]
        [switch_simplify]: 0.00011737
        [loop_unroll]: 4.192e-05
        [a_1]: 0.00083865
        [with_stream_mark]: 2.337e-05
        [recompute_prepare]: 1.862e-05
        [updatestate_depend_eliminate]: 7.43999e-06
        [updatestate_assign_eliminate]: 6.38998e-06
        [updatestate_loads_eliminate]: 6.16e-06
        [parameter_eliminate]: 2.59001e-06
        [specialize_transform]: 1.297e-05
        [updatestate_useless_node_eliminater]: 1.177e-05
        [accelerated_algorithm]: 1.198e-05
        [meta_shard_fg_expand]: 3.11999e-06
        [get_grad_eliminate_]: 1.207e-05
        [merge_forward]: 8.28999e-06
        [cell_reuse_recompute_pass]: 9.89996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.592e-05
        [j_node_and_user_rematch]: 2.302e-05
        [meta_fg_expand]: 0.00132638
        [replace_old_param]: 4.935e-05
        [inline_without_move]: 4.982e-05
        [renormalize]: 0.00209836
        [add_forward_monad_depend]: 9.37001e-06
        [auto_monad_grad]: 5.40001e-06
        [auto_monad_eliminator]: 4.638e-05
        [cse]: 0.0001414
        [replace_applicator]: 6.445e-05
    [Cycle 2]: 0.00183144, [27]
        [switch_simplify]: 3.495e-05
        [loop_unroll]: 3.286e-05
        [a_1]: 0.00097324
        [with_stream_mark]: 1.155e-05
        [recompute_prepare]: 7.68001e-06
        [updatestate_depend_eliminate]: 3.22002e-06
        [updatestate_assign_eliminate]: 3.14001e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 6.63003e-06
        [updatestate_useless_node_eliminater]: 6.32001e-06
        [accelerated_algorithm]: 5.99e-06
        [meta_shard_fg_expand]: 1.45001e-06
        [get_grad_eliminate_]: 5.84999e-06
        [merge_forward]: 2.90002e-06
        [cell_reuse_recompute_pass]: 8.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.279e-05
        [j_node_and_user_rematch]: 9.79999e-06
        [meta_fg_expand]: 5.978e-05
        [replace_old_param]: 9.44998e-06
        [inline_without_move]: 5.82999e-06
        [renormalize]: 0.00042735
        [add_forward_monad_depend]: 4.02e-06
        [auto_monad_grad]: 1.16002e-06
        [auto_monad_eliminator]: 1.125e-05
        [cse]: 1.876e-05
        [replace_applicator]: 1.193e-05
    [Cycle 3]: 0.00036837, [27]
        [switch_simplify]: 6.73e-06
        [loop_unroll]: 5.95002e-06
        [a_1]: 0.00012125
        [with_stream_mark]: 8.60001e-06
        [recompute_prepare]: 6.04001e-06
        [updatestate_depend_eliminate]: 3.45998e-06
        [updatestate_assign_eliminate]: 2.56998e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 6.19001e-06
        [updatestate_useless_node_eliminater]: 5.66e-06
        [accelerated_algorithm]: 5.56e-06
        [meta_shard_fg_expand]: 1.16002e-06
        [get_grad_eliminate_]: 5.46e-06
        [merge_forward]: 2.90998e-06
        [cell_reuse_recompute_pass]: 1.56998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.53e-05
        [j_node_and_user_rematch]: 1.035e-05
        [meta_fg_expand]: 1.90001e-06
        [replace_old_param]: 8.80999e-06
        [inline_without_move]: 5.74999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 6.18998e-06
        [cse]: 1.379e-05
        [replace_applicator]: 5.94999e-06
[py_interpret_to_execute_after_opt_a]: 1.012e-05
[rewriter_after_opt_a]: 3.752e-05
[convert_after_rewriter]: 7.45e-06
[order_py_execute_after_rewriter]: 6.29001e-06
[mutable_eliminate]: 0.00045974
[jit_opt_b]: 5.357e-05, [1]
    [Cycle 1]: 4.704e-05, [2]
        [frontend_op_eliminate]: 1.862e-05
        [inline_after_opt_a]: 1.756e-05
[cconv]: 2.078e-05
[loop_unroll]: 0.0004048
[jit_opt_after_cconv]: 0.00015121, [1]
    [Cycle 1]: 0.00014514, [11]
        [c_1]: 2.554e-05
        [parameter_eliminate]: 2.33002e-06
        [updatestate_depend_eliminate]: 5.62001e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [cse]: 2.125e-05
        [call_graph_tuple_transform]: 2.067e-05
        [tuple_list_get_item_eliminator]: 6.33998e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 6.81999e-06
[remove_dup_value]: 1.625e-05
[partial_unused_args_eliminate]: 2.39001e-06
[environ_conv]: 5.27999e-06
[add_recomputation]: 4.978e-05
[cse_after_recomputation]: 3.03e-05, [1]
    [Cycle 1]: 2.485e-05, [1]
        [cse]: 1.855e-05
[auto_monad_reorder]: 1.856e-05
[get_jit_bprop_graph]: 1.37999e-06
[rewriter_after_jit_bprop_graph]: 5.08002e-06
[opt_after_jit_grad]: 0.00045592
[symbol_engine_optimizer]: 7.708e-05, [1]
    [Cycle 1]: 7.114e-05, [6]
        [build]: 3.4e-06
        [elim_shapecalc]: 8.77e-06
        [elim_not_effective]: 1.438e-05
        [opt_reshape]: 6.59001e-06
        [fold_const_symbol]: 1.038e-05
        [renormalize]: 5.50004e-07
[validate]: 3.397e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00810421
[execute]: 7.87e-06
Sums
    bootstrap                            :     0.000249s :  0.83%
    type_inference                       :     0.012800s : 42.54%
    event_method                         :     0.000054s :  0.18%
    auto_monad                           :     0.000105s :  0.35%
    graph_reusing                        :     0.000008s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000029s :  0.10%
    rewriter_before_opt_a                :     0.000101s :  0.34%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000159s :  0.53%
    jit_opt_a.loop_unroll                :     0.000081s :  0.27%
    jit_opt_a.a_1                        :     0.001933s :  6.43%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.08%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000054s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000043s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.001388s :  4.61%
    jit_opt_a.replace_old_param          :     0.000068s :  0.22%
    jit_opt_a.inline_without_move        :     0.000061s :  0.20%
    jit_opt_a.renormalize                :     0.002526s :  8.39%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000064s :  0.21%
    jit_opt_a.cse                        :     0.000174s :  0.58%
    jit_opt_a.replace_applicator         :     0.000082s :  0.27%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.03%
    rewriter_after_opt_a                 :     0.000038s :  0.12%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000460s :  1.53%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000021s :  0.07%
    loop_unroll                          :     0.000405s :  1.35%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000016s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000050s :  0.17%
    cse_after_recomputation.cse          :     0.000019s :  0.06%
    auto_monad_reorder                   :     0.000019s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000456s :  1.52%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008104s : 26.94%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000443   113
 0.50% :     0.000002s :      3: substitution.elim_not_effective
 0.42% :     0.000002s :      3: substitution.fold_const_symbol
 1.29% :     0.000006s :      4: substitution.graph_param_transform
61.71% :     0.000273s :     14: substitution.inline
 3.46% :     0.000015s :      2: substitution.inline_without_move
 2.01% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.11% :     0.000009s :      7: substitution.minmaximum_grad
 1.37% :     0.000006s :      5: substitution.partial_eliminate
 2.52% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.50% :     0.000024s :     10: substitution.replace_applicator
 1.56% :     0.000007s :      7: substitution.replace_old_param
 0.54% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.71% :     0.000008s :      1: substitution.switch_simplify
 4.21% :     0.000019s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.90% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 8.19% :     0.000036s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012725     2
91.04% :     0.011585s :      1: type_inference.infer
 8.96% :     0.001140s :      1: type_inference.specialize
------[replace.]   0.000170    22
53.49% :     0.000091s :     14: replace.inline
18.94% :     0.000032s :      1: replace.switch_simplify
27.58% :     0.000047s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000292    22
91.02% :     0.000266s :     14: match.inline
 2.39% :     0.000007s :      1: match.switch_simplify
 6.60% :     0.000019s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000365  2634
 1.63% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.54% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.46% :     0.000005s :     44: predicate.addn_check_dump
 1.63% :     0.000006s :     44: predicate.addn_zero_filter
 2.07% :     0.000008s :     44: predicate.arithmetic_simplify
 1.54% :     0.000006s :     44: predicate.cast_eliminate
 0.16% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.46% :     0.000005s :     44: predicate.compare_switch_simplify
 1.58% :     0.000006s :     44: predicate.depend_value_elim
 1.49% :     0.000005s :     44: predicate.dict_get_item_const_eliminator
 1.50% :     0.000005s :     44: predicate.dict_get_item_eliminator
 1.51% :     0.000005s :     44: predicate.dict_set_item_eliminator
 0.33% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      4: predicate.elim_not_effective
 0.21% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000005s :     44: predicate.environ_add_const_eliminate
 1.45% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.45% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.51% :     0.000006s :     44: predicate.environ_get_eliminate
 1.48% :     0.000005s :     44: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.79% :     0.000003s :     19: predicate.get_grad_eliminate
 0.15% :     0.000001s :      4: predicate.graph_param_transform
 4.35% :     0.000016s :     73: predicate.inline
 1.79% :     0.000007s :     40: predicate.inline_without_move
 0.36% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.86% :     0.000003s :     19: predicate.less_batch_normalization
 1.79% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.94% :     0.000007s :     55: predicate.load_eliminater
 0.58% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.43% :     0.000013s :     91: predicate.loop_unroll_before_grad
 1.78% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.45% :     0.000005s :     44: predicate.merge_addn
 1.47% :     0.000005s :     44: predicate.minmaximum_grad
 0.59% :     0.000002s :      4: predicate.mutable_eliminate
 0.22% :     0.000001s :      4: predicate.opt_reshape
 2.31% :     0.000008s :     55: predicate.partial_eliminate
 1.52% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.89% :     0.000007s :     44: predicate.reduce_eliminate
 1.86% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.53% :     0.000009s :     97: predicate.replace_applicator
 0.95% :     0.000003s :     40: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.74% :     0.000006s :     44: predicate.reshape_eliminate
 1.53% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.28% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.52% :     0.000006s :     44: predicate.same_eliminate
 0.44% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.31% :     0.000001s :      8: predicate.special_op_eliminate
 0.88% :     0.000003s :     19: predicate.specialize_transform
 1.64% :     0.000006s :     44: predicate.split_environ_get_set_with_tuple_value
 1.55% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.16% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.78% :     0.000010s :     65: predicate.switch_defer_inline
 2.60% :     0.000009s :     65: predicate.switch_layer_defer_inline
 6.60% :     0.000024s :    162: predicate.switch_simplify
 1.47% :     0.000005s :     44: predicate.tile_eliminate
 1.55% :     0.000006s :     44: predicate.transpose_eliminate
 1.81% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.67% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 3.08% :     0.000011s :     59: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000006s :     44: predicate.tuple_list_set_item_eliminator
 1.83% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.92% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000010s :     74: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000006s :     44: predicate.value_based_eliminate
 0.17% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.35% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001471    32
63.58% :     0.000935s :     14: func_graph_cloner_run.FuncGraphClonerGraph
36.42% :     0.000536s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.039154    91
 0.14% :     0.000053s :      1: add_recomputation
 0.28% :     0.000110s :      1: auto_monad
 0.05% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.71% :     0.000277s :      1: bootstrap
 0.06% :     0.000023s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000033s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.15% :     0.000060s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
26.61% :     0.010418s :      1: jit_opt_a
 0.39% :     0.000154s :      1: jit_opt_after_cconv
 0.14% :     0.000056s :      1: jit_opt_b
 1.05% :     0.000412s :      1: loop_unroll
 1.20% :     0.000469s :      1: mutable_eliminate
 6.56% :     0.002567s :     39: opt.transform.jit_opt_a
 0.14% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000030s :      4: opt.transform.jit_opt_b
 0.03% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.18% :     0.000464s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000031s :      1: py_interpret_to_execute
 0.03% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 3.45% :     0.001351s :      2: renormalize.infer
 2.97% :     0.001162s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000040s :      1: rewriter_after_opt_a
 0.27% :     0.000104s :      1: rewriter_before_opt_a
 0.20% :     0.000080s :      1: symbol_engine_optimizer
20.73% :     0.008116s :      1: task_emit
32.73% :     0.012815s :      1: type_inference
 0.14% :     0.000055s :      1: validate


TotalTime = 0.0223547, [33]
[bootstrap]: 0.00036986
[type_inference]: 0.00561498
[event_method]: 1.899e-05
[auto_monad]: 6.663e-05
[graph_reusing]: 6.71999e-06
[pre_auto_parallel]: 3.24001e-06
[py_interpret_to_execute]: 4.379e-05
[rewriter_before_opt_a]: 7.531e-05
[expand_dump_flag]: 3.01001e-06
[jit_opt_a]: 0.00663328, [2]
    [Cycle 1]: 0.00221851, [27]
        [switch_simplify]: 6.892e-05
        [loop_unroll]: 2.94e-05
        [a_1]: 0.00061503
        [with_stream_mark]: 2.448e-05
        [recompute_prepare]: 8.99e-06
        [updatestate_depend_eliminate]: 4.52998e-06
        [updatestate_assign_eliminate]: 3.38e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [parameter_eliminate]: 1.97001e-06
        [specialize_transform]: 6.32001e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 6.07001e-06
        [meta_shard_fg_expand]: 2.74001e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 3.61001e-06
        [cell_reuse_recompute_pass]: 1.68002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.601e-05
        [j_node_and_user_rematch]: 9.41e-06
        [meta_fg_expand]: 2.29999e-06
        [replace_old_param]: 1.043e-05
        [inline_without_move]: 5.47001e-06
        [renormalize]: 0.00108918
        [add_forward_monad_depend]: 8.59998e-06
        [auto_monad_grad]: 2.73998e-06
        [auto_monad_eliminator]: 1.941e-05
        [cse]: 3.173e-05
        [replace_applicator]: 1.882e-05
    [Cycle 2]: 0.00036054, [27]
        [switch_simplify]: 6.31e-06
        [loop_unroll]: 5.30001e-06
        [a_1]: 0.00010078
        [with_stream_mark]: 1.393e-05
        [recompute_prepare]: 5.15001e-06
        [updatestate_depend_eliminate]: 3.27002e-06
        [updatestate_assign_eliminate]: 2.58998e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.46002e-06
        [specialize_transform]: 4.94998e-06
        [updatestate_useless_node_eliminater]: 5.02999e-06
        [accelerated_algorithm]: 5.88998e-06
        [meta_shard_fg_expand]: 2.39001e-06
        [get_grad_eliminate_]: 4.42998e-06
        [merge_forward]: 3.2e-06
        [cell_reuse_recompute_pass]: 2.73998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.861e-05
        [j_node_and_user_rematch]: 8.47e-06
        [meta_fg_expand]: 2.68003e-06
        [replace_old_param]: 8.28001e-06
        [inline_without_move]: 4.63001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 1.47999e-06
        [auto_monad_eliminator]: 5.86003e-06
        [cse]: 1.152e-05
        [replace_applicator]: 5.19e-06
[py_interpret_to_execute_after_opt_a]: 1.559e-05
[rewriter_after_opt_a]: 3.847e-05
[convert_after_rewriter]: 7.25998e-06
[order_py_execute_after_rewriter]: 5.66e-06
[mutable_eliminate]: 0.00084339
[jit_opt_b]: 5.483e-05, [1]
    [Cycle 1]: 4.618e-05, [2]
        [frontend_op_eliminate]: 1.796e-05
        [inline_after_opt_a]: 1.632e-05
[cconv]: 3.548e-05
[loop_unroll]: 0.00054193
[jit_opt_after_cconv]: 0.00017385, [1]
    [Cycle 1]: 0.00016571, [11]
        [c_1]: 2.406e-05
        [parameter_eliminate]: 5.14e-06
        [updatestate_depend_eliminate]: 1.014e-05
        [updatestate_assign_eliminate]: 2.53998e-06
        [updatestate_loads_eliminate]: 3.14999e-06
        [cse]: 2.986e-05
        [call_graph_tuple_transform]: 2.569e-05
        [tuple_list_get_item_eliminator]: 5.22e-06
        [none_parameter_eliminate]: 1.74998e-06
        [renormalize]: 1.05999e-06
        [switch_simplify]: 5.74e-06
[remove_dup_value]: 1.598e-05
[partial_unused_args_eliminate]: 2.43e-06
[environ_conv]: 6.14999e-06
[add_recomputation]: 5.704e-05
[cse_after_recomputation]: 2.284e-05, [1]
    [Cycle 1]: 1.61e-05, [1]
        [cse]: 9.80002e-06
[auto_monad_reorder]: 1.543e-05
[get_jit_bprop_graph]: 2.46e-06
[rewriter_after_jit_bprop_graph]: 8.64998e-06
[opt_after_jit_grad]: 0.00053654
[symbol_engine_optimizer]: 8.209e-05, [1]
    [Cycle 1]: 7.414e-05, [6]
        [build]: 5.87999e-06
        [elim_shapecalc]: 8.12998e-06
        [elim_not_effective]: 1.494e-05
        [opt_reshape]: 6.06e-06
        [fold_const_symbol]: 9.45001e-06
        [renormalize]: 1.12e-06
[validate]: 4.587e-05
[backend_pass]: 8.59989e-07
[task_emit]: 0.00675508
[execute]: 8.43999e-06
Sums
    bootstrap                            :     0.000370s :  2.10%
    type_inference                       :     0.005615s : 31.92%
    event_method                         :     0.000019s :  0.11%
    auto_monad                           :     0.000067s :  0.38%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000044s :  0.25%
    rewriter_before_opt_a                :     0.000075s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000075s :  0.43%
    jit_opt_a.loop_unroll                :     0.000035s :  0.20%
    jit_opt_a.a_1                        :     0.000716s :  4.07%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001089s :  6.19%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.14%
    jit_opt_a.cse                        :     0.000043s :  0.25%
    jit_opt_a.replace_applicator         :     0.000024s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.09%
    rewriter_after_opt_a                 :     0.000038s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000843s :  4.79%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000035s :  0.20%
    loop_unroll                          :     0.000542s :  3.08%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000030s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000057s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000537s :  3.05%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000046s :  0.26%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006755s : 38.40%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000257    23
 1.14% :     0.000003s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.62% :     0.000007s :      3: substitution.graph_param_transform
83.93% :     0.000216s :      4: substitution.inline
 1.59% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.39% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.30% :     0.000006s :      2: substitution.replace_old_param
 5.47% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005534     2
80.18% :     0.004437s :      1: type_inference.infer
19.82% :     0.001097s :      1: type_inference.specialize
------[replace.]   0.000086     6
65.34% :     0.000056s :      4: replace.inline
34.66% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000226     6
94.27% :     0.000213s :      4: match.inline
 5.73% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.47% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.66% :     0.000002s :     10: predicate.addn_zero_filter
 2.68% :     0.000003s :     10: predicate.arithmetic_simplify
 1.45% :     0.000002s :     10: predicate.cast_eliminate
 0.56% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.53% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.30% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000002s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 6.18% :     0.000008s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.34% :     0.000002s :      6: predicate.less_batch_normalization
 1.96% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.76% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 1.94% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.36% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.69% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.97% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.23% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 1.89% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.20% :     0.000010s :     53: predicate.switch_simplify
 1.23% :     0.000002s :     10: predicate.tile_eliminate
 1.06% :     0.000001s :     10: predicate.transpose_eliminate
 1.69% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.68% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000781    12
44.96% :     0.000351s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.78% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.26% :     0.000392s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024512    76
 0.25% :     0.000060s :      1: add_recomputation
 0.29% :     0.000070s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.59% :     0.000391s :      1: bootstrap
 0.16% :     0.000039s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
27.08% :     0.006637s :      1: jit_opt_a
 0.72% :     0.000176s :      1: jit_opt_after_cconv
 0.24% :     0.000058s :      1: jit_opt_b
 2.25% :     0.000552s :      1: loop_unroll
 3.49% :     0.000855s :      1: mutable_eliminate
 3.90% :     0.000956s :     26: opt.transform.jit_opt_a
 0.23% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.23% :     0.000546s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000047s :      1: py_interpret_to_execute
 0.07% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000019s :      1: remove_dup_value
 2.37% :     0.000582s :      1: renormalize.infer
 2.03% :     0.000497s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000045s :      1: rewriter_after_opt_a
 0.32% :     0.000078s :      1: rewriter_before_opt_a
 0.35% :     0.000085s :      1: symbol_engine_optimizer
27.63% :     0.006773s :      1: task_emit
22.99% :     0.005635s :      1: type_inference
 0.28% :     0.000069s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.150.824 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6c072280> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.151.529 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0148828, [30]
[bootstrap]: 0.00042313
[type_inference]: 0.00519509
[event_method]: 1.994e-05
[auto_monad]: 7.304e-05
[graph_reusing]: 5.96e-06
[pre_auto_parallel]: 2.49001e-06
[py_interpret_to_execute]: 3.83e-05
[rewriter_before_opt_a]: 8.322e-05
[expand_dump_flag]: 3.53e-06
[jit_opt_a]: 0.00643254, [2]
    [Cycle 1]: 0.00192405, [27]
        [switch_simplify]: 6.763e-05
        [loop_unroll]: 3.247e-05
        [a_1]: 0.0006313
        [with_stream_mark]: 2.621e-05
        [recompute_prepare]: 8.69e-06
        [updatestate_depend_eliminate]: 4.52e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.60002e-06
        [parameter_eliminate]: 2.93e-06
        [specialize_transform]: 7.28999e-06
        [updatestate_useless_node_eliminater]: 5.69e-06
        [accelerated_algorithm]: 5.58002e-06
        [meta_shard_fg_expand]: 2.84999e-06
        [get_grad_eliminate_]: 5.86e-06
        [merge_forward]: 4.05e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.488e-05
        [j_node_and_user_rematch]: 9.66998e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 1.035e-05
        [inline_without_move]: 5.61998e-06
        [renormalize]: 0.00078729
        [add_forward_monad_depend]: 9.02e-06
        [auto_monad_grad]: 3.31001e-06
        [auto_monad_eliminator]: 1.866e-05
        [cse]: 3.396e-05
        [replace_applicator]: 1.538e-05
    [Cycle 2]: 0.00037186, [27]
        [switch_simplify]: 6.07999e-06
        [loop_unroll]: 5.94999e-06
        [a_1]: 9.343e-05
        [with_stream_mark]: 1.347e-05
        [recompute_prepare]: 5.67001e-06
        [updatestate_depend_eliminate]: 3.71001e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [parameter_eliminate]: 1.54e-06
        [specialize_transform]: 5.40001e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 5.92999e-06
        [meta_shard_fg_expand]: 1.66e-06
        [get_grad_eliminate_]: 4.79e-06
        [merge_forward]: 3.46999e-06
        [cell_reuse_recompute_pass]: 2.29001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.594e-05
        [j_node_and_user_rematch]: 8.85999e-06
        [meta_fg_expand]: 1.72001e-06
        [replace_old_param]: 8.46002e-06
        [inline_without_move]: 4.94e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.47001e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 1.877e-05
        [cse]: 1.568e-05
        [replace_applicator]: 6.01e-06
[py_interpret_to_execute_after_opt_a]: 1.424e-05
[rewriter_after_opt_a]: 4.189e-05
[convert_after_rewriter]: 7.94002e-06
[order_py_execute_after_rewriter]: 5.56998e-06
[mutable_eliminate]: 0.00079561
[jit_opt_b]: 5.421e-05, [1]
    [Cycle 1]: 4.638e-05, [2]
        [frontend_op_eliminate]: 1.758e-05
        [inline_after_opt_a]: 1.652e-05
[cconv]: 3.335e-05
[loop_unroll]: 0.00049498
[jit_opt_after_cconv]: 0.00016669, [1]
    [Cycle 1]: 0.00015848, [11]
        [c_1]: 2.238e-05
        [parameter_eliminate]: 4.52e-06
        [updatestate_depend_eliminate]: 8.08999e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.28998e-06
        [cse]: 2.715e-05
        [call_graph_tuple_transform]: 2.353e-05
        [tuple_list_get_item_eliminator]: 6.21e-06
        [none_parameter_eliminate]: 2.11e-06
        [renormalize]: 8.09989e-07
        [switch_simplify]: 5.59e-06
[remove_dup_value]: 1.489e-05
[partial_unused_args_eliminate]: 2.57001e-06
[environ_conv]: 5.79e-06
[add_recomputation]: 5.48e-05
[cse_after_recomputation]: 2.332e-05, [1]
    [Cycle 1]: 1.688e-05, [1]
        [cse]: 1.013e-05
[auto_monad_reorder]: 1.573e-05
[get_jit_bprop_graph]: 2.22999e-06
[rewriter_after_jit_bprop_graph]: 7.8e-06
[opt_after_jit_grad]: 0.0005312
[symbol_engine_optimizer]: 8.418e-05, [1]
    [Cycle 1]: 7.62e-05, [6]
        [build]: 5.55001e-06
        [elim_shapecalc]: 8.90999e-06
        [elim_not_effective]: 1.513e-05
        [opt_reshape]: 6.16998e-06
        [fold_const_symbol]: 9.29998e-06
        [renormalize]: 8.00006e-07
[validate]: 4.848e-05
Sums
    bootstrap                            :     0.000423s :  4.20%
    type_inference                       :     0.005195s : 51.51%
    event_method                         :     0.000020s :  0.20%
    auto_monad                           :     0.000073s :  0.72%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.38%
    rewriter_before_opt_a                :     0.000083s :  0.83%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000074s :  0.73%
    jit_opt_a.loop_unroll                :     0.000038s :  0.38%
    jit_opt_a.a_1                        :     0.000725s :  7.19%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.39%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000013s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000008s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000787s :  7.81%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000037s :  0.37%
    jit_opt_a.cse                        :     0.000050s :  0.49%
    jit_opt_a.replace_applicator         :     0.000021s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.14%
    rewriter_after_opt_a                 :     0.000042s :  0.42%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000796s :  7.89%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.16%
    cconv                                :     0.000033s :  0.33%
    loop_unroll                          :     0.000495s :  4.91%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000027s :  0.27%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000015s :  0.15%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000055s :  0.54%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.08%
    opt_after_jit_grad                   :     0.000531s :  5.27%
    symbol_engine_optimizer.build        :     0.000006s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000048s :  0.48%


Time group info:
------[substitution.]   0.000262    23
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.52% :     0.000007s :      3: substitution.graph_param_transform
85.30% :     0.000224s :      4: substitution.inline
 1.36% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.01% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.07% :     0.000005s :      2: substitution.replace_old_param
 5.45% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005117     2
82.29% :     0.004211s :      1: type_inference.infer
17.71% :     0.000906s :      1: type_inference.specialize
------[replace.]   0.000082     6
67.01% :     0.000055s :      4: replace.inline
32.99% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000234     6
94.39% :     0.000221s :      4: match.inline
 5.61% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.24% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.41% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000002s :     10: predicate.addn_zero_filter
 2.74% :     0.000004s :     10: predicate.arithmetic_simplify
 1.11% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.05% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.00% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.11% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000001s :      3: predicate.graph_param_transform
 6.06% :     0.000008s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.71% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.10% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 0.95% :     0.000001s :     10: predicate.minmaximum_grad
 1.76% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     15: predicate.partial_eliminate
 1.08% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.80% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.83% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.25% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.11% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000001s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.56% :     0.000002s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.02% :     0.000004s :     16: predicate.switch_defer_inline
 2.01% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.87% :     0.000010s :     53: predicate.switch_simplify
 1.32% :     0.000002s :     10: predicate.tile_eliminate
 1.25% :     0.000002s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.47% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.64% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000681    12
50.30% :     0.000342s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.98% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.72% :     0.000284s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016685    72
 0.35% :     0.000058s :      1: add_recomputation
 0.46% :     0.000077s :      1: auto_monad
 0.11% :     0.000019s :      1: auto_monad_reorder
 2.63% :     0.000439s :      1: bootstrap
 0.22% :     0.000036s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.14% :     0.000024s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.58% :     0.006437s :      1: jit_opt_a
 1.02% :     0.000170s :      1: jit_opt_after_cconv
 0.35% :     0.000058s :      1: jit_opt_b
 3.03% :     0.000505s :      1: loop_unroll
 4.84% :     0.000808s :      1: mutable_eliminate
 5.79% :     0.000966s :     26: opt.transform.jit_opt_a
 0.32% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000027s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.25% :     0.000542s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000042s :      1: py_interpret_to_execute
 0.10% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.50% :     0.000417s :      1: renormalize.infer
 2.16% :     0.000361s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000046s :      1: rewriter_after_opt_a
 0.52% :     0.000087s :      1: rewriter_before_opt_a
 0.52% :     0.000087s :      1: symbol_engine_optimizer
31.24% :     0.005213s :      1: type_inference


TotalTime = 0.0194674, [33]
[bootstrap]: 0.00029221
[type_inference]: 0.00482903
[event_method]: 1.702e-05
[auto_monad]: 6.608e-05
[graph_reusing]: 6.21998e-06
[pre_auto_parallel]: 2.83e-06
[py_interpret_to_execute]: 3.661e-05
[rewriter_before_opt_a]: 7.082e-05
[expand_dump_flag]: 3.85e-06
[jit_opt_a]: 0.00563337, [2]
    [Cycle 1]: 0.00170772, [27]
        [switch_simplify]: 5.737e-05
        [loop_unroll]: 2.842e-05
        [a_1]: 0.00053004
        [with_stream_mark]: 1.682e-05
        [recompute_prepare]: 7.00002e-06
        [updatestate_depend_eliminate]: 3.52002e-06
        [updatestate_assign_eliminate]: 3.20002e-06
        [updatestate_loads_eliminate]: 2.53003e-06
        [parameter_eliminate]: 1.73002e-06
        [specialize_transform]: 5.63002e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.10001e-06
        [meta_shard_fg_expand]: 2.23998e-06
        [get_grad_eliminate_]: 5.42999e-06
        [merge_forward]: 3.45e-06
        [cell_reuse_recompute_pass]: 1.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.351e-05
        [j_node_and_user_rematch]: 8.40001e-06
        [meta_fg_expand]: 2.22001e-06
        [replace_old_param]: 8.22e-06
        [inline_without_move]: 5.26002e-06
        [renormalize]: 0.00076442
        [add_forward_monad_depend]: 5.81998e-06
        [auto_monad_grad]: 2.46e-06
        [auto_monad_eliminator]: 1.444e-05
        [cse]: 2.516e-05
        [replace_applicator]: 1.202e-05
    [Cycle 2]: 0.00030585, [27]
        [switch_simplify]: 5.70001e-06
        [loop_unroll]: 4.77e-06
        [a_1]: 8.845e-05
        [with_stream_mark]: 8.45001e-06
        [recompute_prepare]: 4.71002e-06
        [updatestate_depend_eliminate]: 2.69001e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.04003e-06
        [specialize_transform]: 4.60999e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 4.79e-06
        [meta_shard_fg_expand]: 1.26997e-06
        [get_grad_eliminate_]: 4.38999e-06
        [merge_forward]: 2.45002e-06
        [cell_reuse_recompute_pass]: 1.64998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.19e-05
        [j_node_and_user_rematch]: 7.92003e-06
        [meta_fg_expand]: 1.35999e-06
        [replace_old_param]: 7.06999e-06
        [inline_without_move]: 4.57e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.03001e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 5.13002e-06
        [cse]: 9.45001e-06
        [replace_applicator]: 4.75001e-06
[py_interpret_to_execute_after_opt_a]: 9.38002e-06
[rewriter_after_opt_a]: 3.25e-05
[convert_after_rewriter]: 6.49999e-06
[order_py_execute_after_rewriter]: 4.79002e-06
[mutable_eliminate]: 0.00053525
[jit_opt_b]: 4.854e-05, [1]
    [Cycle 1]: 4.192e-05, [2]
        [frontend_op_eliminate]: 1.603e-05
        [inline_after_opt_a]: 1.496e-05
[cconv]: 2.375e-05
[loop_unroll]: 0.00041063
[jit_opt_after_cconv]: 0.00014949, [1]
    [Cycle 1]: 0.00014278, [11]
        [c_1]: 2.062e-05
        [parameter_eliminate]: 2.72001e-06
        [updatestate_depend_eliminate]: 6.11e-06
        [updatestate_assign_eliminate]: 2.30002e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [cse]: 1.87e-05
        [call_graph_tuple_transform]: 1.887e-05
        [tuple_list_get_item_eliminator]: 5.12e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 5.17999e-06
[remove_dup_value]: 1.351e-05
[partial_unused_args_eliminate]: 8.877e-05
[environ_conv]: 6.28e-06
[add_recomputation]: 5.29e-05
[cse_after_recomputation]: 2.72e-05, [1]
    [Cycle 1]: 2.048e-05, [1]
        [cse]: 1.308e-05
[auto_monad_reorder]: 1.654e-05
[get_jit_bprop_graph]: 1.79998e-06
[rewriter_after_jit_bprop_graph]: 6.84001e-06
[opt_after_jit_grad]: 0.00054025
[symbol_engine_optimizer]: 8.124e-05, [1]
    [Cycle 1]: 7.434e-05, [6]
        [build]: 4.02e-06
        [elim_shapecalc]: 7.80998e-06
        [elim_not_effective]: 1.29e-05
        [opt_reshape]: 1.196e-05
        [fold_const_symbol]: 8.64e-06
        [renormalize]: 6.09987e-07
[validate]: 4.007e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00618143
[execute]: 6.86001e-06
Sums
    bootstrap                            :     0.000292s :  1.92%
    type_inference                       :     0.004829s : 31.74%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000066s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.24%
    rewriter_before_opt_a                :     0.000071s :  0.47%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000063s :  0.41%
    jit_opt_a.loop_unroll                :     0.000033s :  0.22%
    jit_opt_a.a_1                        :     0.000618s :  4.06%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000764s :  5.02%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000035s :  0.23%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000535s :  3.52%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000024s :  0.16%
    loop_unroll                          :     0.000411s :  2.70%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000089s :  0.58%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000053s :  0.35%
    cse_after_recomputation.cse          :     0.000013s :  0.09%
    auto_monad_reorder                   :     0.000017s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000540s :  3.55%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000012s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006181s : 40.62%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000206    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.90% :     0.000006s :      3: substitution.graph_param_transform
83.92% :     0.000173s :      4: substitution.inline
 1.34% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.07% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.84% :     0.000004s :      2: substitution.replace_old_param
 6.31% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004755     2
77.87% :     0.003702s :      1: type_inference.infer
22.13% :     0.001052s :      1: type_inference.specialize
------[replace.]   0.000072     6
66.00% :     0.000048s :      4: replace.inline
34.00% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000182     6
93.48% :     0.000170s :      4: match.inline
 6.52% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.62% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.42% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 2.26% :     0.000003s :     10: predicate.arithmetic_simplify
 1.30% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.45% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.76% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.26% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 5.74% :     0.000007s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.78% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.08% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.71% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.20% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.73% :     0.000009s :     53: predicate.switch_simplify
 1.27% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.32% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.91% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000625    12
45.57% :     0.000285s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.69% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.74% :     0.000311s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021157    76
 0.26% :     0.000056s :      1: add_recomputation
 0.33% :     0.000070s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.44% :     0.000306s :      1: bootstrap
 0.12% :     0.000026s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000029s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.65% :     0.005637s :      1: jit_opt_a
 0.72% :     0.000152s :      1: jit_opt_after_cconv
 0.24% :     0.000051s :      1: jit_opt_b
 1.98% :     0.000418s :      1: loop_unroll
 2.57% :     0.000544s :      1: mutable_eliminate
 3.89% :     0.000823s :     26: opt.transform.jit_opt_a
 0.22% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000038s :      4: opt.transform.symbol_engine_opt
 2.60% :     0.000551s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.44% :     0.000094s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000039s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.97% :     0.000418s :      1: renormalize.infer
 1.60% :     0.000340s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000035s :      1: rewriter_after_opt_a
 0.35% :     0.000075s :      1: rewriter_before_opt_a
 0.40% :     0.000084s :      1: symbol_engine_optimizer
29.27% :     0.006194s :      1: task_emit
22.97% :     0.004859s :      1: type_inference
 0.30% :     0.000062s :      1: validate


TotalTime = 0.0330639, [33]
[bootstrap]: 0.00023245
[type_inference]: 0.0124857
[event_method]: 5.867e-05
[auto_monad]: 0.00010306
[graph_reusing]: 6.91999e-06
[pre_auto_parallel]: 2.82002e-06
[py_interpret_to_execute]: 2.671e-05
[rewriter_before_opt_a]: 0.00010307
[expand_dump_flag]: 3.13e-06
[jit_opt_a]: 0.0110104, [3]
    [Cycle 1]: 0.00527597, [27]
        [switch_simplify]: 0.00011654
        [loop_unroll]: 4.186e-05
        [a_1]: 0.00087327
        [with_stream_mark]: 2.335e-05
        [recompute_prepare]: 1.689e-05
        [updatestate_depend_eliminate]: 1.676e-05
        [updatestate_assign_eliminate]: 6.64001e-06
        [updatestate_loads_eliminate]: 5.79999e-06
        [parameter_eliminate]: 2.95002e-06
        [specialize_transform]: 1.327e-05
        [updatestate_useless_node_eliminater]: 1.254e-05
        [accelerated_algorithm]: 1.204e-05
        [meta_shard_fg_expand]: 3.78999e-06
        [get_grad_eliminate_]: 1.166e-05
        [merge_forward]: 7.97e-06
        [cell_reuse_recompute_pass]: 9.50007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.588e-05
        [j_node_and_user_rematch]: 2.241e-05
        [meta_fg_expand]: 0.00132632
        [replace_old_param]: 4.854e-05
        [inline_without_move]: 4.883e-05
        [renormalize]: 0.00209457
        [add_forward_monad_depend]: 9.42999e-06
        [auto_monad_grad]: 4.79e-06
        [auto_monad_eliminator]: 5.452e-05
        [cse]: 0.00019064
        [replace_applicator]: 6.213e-05
    [Cycle 2]: 0.00199593, [27]
        [switch_simplify]: 3.509e-05
        [loop_unroll]: 3.283e-05
        [a_1]: 0.00106779
        [with_stream_mark]: 1.262e-05
        [recompute_prepare]: 7.92e-06
        [updatestate_depend_eliminate]: 3.83999e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 2.526e-05
        [updatestate_useless_node_eliminater]: 6.34001e-06
        [accelerated_algorithm]: 6.09001e-06
        [meta_shard_fg_expand]: 1.60999e-06
        [get_grad_eliminate_]: 5.59e-06
        [merge_forward]: 3.18998e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 3.363e-05
        [j_node_and_user_rematch]: 9.85002e-06
        [meta_fg_expand]: 6.535e-05
        [replace_old_param]: 1.002e-05
        [inline_without_move]: 5.99e-06
        [renormalize]: 0.00045499
        [add_forward_monad_depend]: 4.50999e-06
        [auto_monad_grad]: 1.29e-06
        [auto_monad_eliminator]: 1.162e-05
        [cse]: 2.049e-05
        [replace_applicator]: 1.168e-05
    [Cycle 3]: 0.00035586, [27]
        [switch_simplify]: 6.31e-06
        [loop_unroll]: 5.82001e-06
        [a_1]: 0.00011443
        [with_stream_mark]: 8.75999e-06
        [recompute_prepare]: 5.72999e-06
        [updatestate_depend_eliminate]: 3.35e-06
        [updatestate_assign_eliminate]: 3.26001e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 5.99e-06
        [updatestate_useless_node_eliminater]: 5.59e-06
        [accelerated_algorithm]: 5.84999e-06
        [meta_shard_fg_expand]: 1.29998e-06
        [get_grad_eliminate_]: 5.59e-06
        [merge_forward]: 2.76e-06
        [cell_reuse_recompute_pass]: 1.43002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.42e-05
        [j_node_and_user_rematch]: 9.46e-06
        [meta_fg_expand]: 1.84998e-06
        [replace_old_param]: 8.02e-06
        [inline_without_move]: 5.79999e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.59e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 6.34001e-06
        [cse]: 1.335e-05
        [replace_applicator]: 5.77999e-06
[py_interpret_to_execute_after_opt_a]: 1.024e-05
[rewriter_after_opt_a]: 3.743e-05
[convert_after_rewriter]: 7.87e-06
[order_py_execute_after_rewriter]: 5.34e-06
[mutable_eliminate]: 0.00051891
[jit_opt_b]: 5.458e-05, [1]
    [Cycle 1]: 4.772e-05, [2]
        [frontend_op_eliminate]: 1.851e-05
        [inline_after_opt_a]: 1.774e-05
[cconv]: 2.086e-05
[loop_unroll]: 0.00044566
[jit_opt_after_cconv]: 0.00015018, [1]
    [Cycle 1]: 0.00014341, [11]
        [c_1]: 2.529e-05
        [parameter_eliminate]: 2.36998e-06
        [updatestate_depend_eliminate]: 6.06e-06
        [updatestate_assign_eliminate]: 2.84001e-06
        [updatestate_loads_eliminate]: 2.57001e-06
        [cse]: 2.148e-05
        [call_graph_tuple_transform]: 2.058e-05
        [tuple_list_get_item_eliminator]: 5.94999e-06
        [none_parameter_eliminate]: 1.73002e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 6.23e-06
[remove_dup_value]: 1.599e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 5.40999e-06
[add_recomputation]: 5.905e-05
[cse_after_recomputation]: 2.696e-05, [1]
    [Cycle 1]: 2.126e-05, [1]
        [cse]: 1.532e-05
[auto_monad_reorder]: 1.778e-05
[get_jit_bprop_graph]: 1.34e-06
[rewriter_after_jit_bprop_graph]: 5.10001e-06
[opt_after_jit_grad]: 0.00048593
[symbol_engine_optimizer]: 8.369e-05, [1]
    [Cycle 1]: 7.71e-05, [6]
        [build]: 3.61999e-06
        [elim_shapecalc]: 9.32999e-06
        [elim_not_effective]: 1.54e-05
        [opt_reshape]: 7.23e-06
        [fold_const_symbol]: 1.19e-05
        [renormalize]: 4.50003e-07
[validate]: 3.57e-05
[backend_pass]: 9.10019e-07
[task_emit]: 0.00683547
[execute]: 8.47e-06
Sums
    bootstrap                            :     0.000232s :  0.80%
    type_inference                       :     0.012486s : 43.23%
    event_method                         :     0.000059s :  0.20%
    auto_monad                           :     0.000103s :  0.36%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000027s :  0.09%
    rewriter_before_opt_a                :     0.000103s :  0.36%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000158s :  0.55%
    jit_opt_a.loop_unroll                :     0.000081s :  0.28%
    jit_opt_a.a_1                        :     0.002055s :  7.12%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000031s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000024s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000045s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.08%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000074s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.001394s :  4.82%
    jit_opt_a.replace_old_param          :     0.000067s :  0.23%
    jit_opt_a.inline_without_move        :     0.000061s :  0.21%
    jit_opt_a.renormalize                :     0.002550s :  8.83%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000072s :  0.25%
    jit_opt_a.cse                        :     0.000224s :  0.78%
    jit_opt_a.replace_applicator         :     0.000080s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.04%
    rewriter_after_opt_a                 :     0.000037s :  0.13%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000519s :  1.80%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000021s :  0.07%
    loop_unroll                          :     0.000446s :  1.54%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000059s :  0.20%
    cse_after_recomputation.cse          :     0.000015s :  0.05%
    auto_monad_reorder                   :     0.000018s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000486s :  1.68%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006835s : 23.67%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000473   113
 0.49% :     0.000002s :      3: substitution.elim_not_effective
 0.44% :     0.000002s :      3: substitution.fold_const_symbol
 1.26% :     0.000006s :      4: substitution.graph_param_transform
63.98% :     0.000302s :     14: substitution.inline
 3.10% :     0.000015s :      2: substitution.inline_without_move
 1.76% :     0.000008s :     14: substitution.j_node_and_user_rematch
 1.97% :     0.000009s :      7: substitution.minmaximum_grad
 1.28% :     0.000006s :      5: substitution.partial_eliminate
 2.17% :     0.000010s :     14: substitution.remove_not_recompute_node
 5.28% :     0.000025s :     10: substitution.replace_applicator
 1.46% :     0.000007s :      7: substitution.replace_old_param
 0.46% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.54% :     0.000007s :      1: substitution.switch_simplify
 3.99% :     0.000019s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.74% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 8.09% :     0.000038s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012412     2
90.39% :     0.011219s :      1: type_inference.infer
 9.61% :     0.001193s :      1: type_inference.specialize
------[replace.]   0.000177    22
53.83% :     0.000095s :     14: replace.inline
17.84% :     0.000032s :      1: replace.switch_simplify
28.34% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000322    22
91.48% :     0.000295s :     14: match.inline
 2.07% :     0.000007s :      1: match.switch_simplify
 6.45% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000407  2634
 1.50% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.41% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.36% :     0.000006s :     44: predicate.addn_check_dump
 1.56% :     0.000006s :     44: predicate.addn_zero_filter
 1.73% :     0.000007s :     44: predicate.arithmetic_simplify
 1.39% :     0.000006s :     44: predicate.cast_eliminate
 0.14% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.32% :     0.000005s :     44: predicate.compare_switch_simplify
 1.51% :     0.000006s :     44: predicate.depend_value_elim
 5.41% :     0.000022s :     44: predicate.dict_get_item_const_eliminator
 1.46% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.37% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.28% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      4: predicate.elim_not_effective
 0.21% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000005s :     44: predicate.environ_add_const_eliminate
 1.34% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.33% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.35% :     0.000005s :     44: predicate.environ_get_eliminate
 1.33% :     0.000005s :     44: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.75% :     0.000003s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.00% :     0.000016s :     73: predicate.inline
 1.58% :     0.000006s :     40: predicate.inline_without_move
 0.31% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.81% :     0.000003s :     19: predicate.less_batch_normalization
 1.65% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.75% :     0.000007s :     55: predicate.load_eliminater
 0.49% :     0.000002s :      4: predicate.loop_unroll_after_grad
 2.96% :     0.000012s :     91: predicate.loop_unroll_before_grad
 1.65% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.35% :     0.000006s :     44: predicate.merge_addn
 1.35% :     0.000005s :     44: predicate.minmaximum_grad
 0.51% :     0.000002s :      4: predicate.mutable_eliminate
 0.20% :     0.000001s :      4: predicate.opt_reshape
 2.14% :     0.000009s :     55: predicate.partial_eliminate
 1.43% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.71% :     0.000007s :     44: predicate.reduce_eliminate
 1.62% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.37% :     0.000001s :     19: predicate.remove_not_recompute_node
 2.20% :     0.000009s :     97: predicate.replace_applicator
 0.79% :     0.000003s :     40: predicate.replace_old_param
 0.10% :     0.000000s :      4: predicate.reset_defer_inline
 6.82% :     0.000028s :     44: predicate.reshape_eliminate
 1.50% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.26% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.40% :     0.000006s :     44: predicate.same_eliminate
 0.41% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.30% :     0.000001s :      8: predicate.special_op_eliminate
 0.78% :     0.000003s :     19: predicate.specialize_transform
 1.53% :     0.000006s :     44: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.15% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.52% :     0.000010s :     65: predicate.switch_defer_inline
 2.35% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.02% :     0.000025s :    162: predicate.switch_simplify
 1.43% :     0.000006s :     44: predicate.tile_eliminate
 1.38% :     0.000006s :     44: predicate.transpose_eliminate
 1.60% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.44% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 2.84% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.70% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.59% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000007s :     44: predicate.value_based_eliminate
 0.13% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.25% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001444    32
62.61% :     0.000904s :     14: func_graph_cloner_run.FuncGraphClonerGraph
37.39% :     0.000540s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038459    91
 0.16% :     0.000062s :      1: add_recomputation
 0.28% :     0.000108s :      1: auto_monad
 0.05% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.65% :     0.000249s :      1: bootstrap
 0.06% :     0.000023s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000029s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.17% :     0.000064s :      1: event_method
 0.03% :     0.000013s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
28.64% :     0.011013s :      1: jit_opt_a
 0.40% :     0.000153s :      1: jit_opt_after_cconv
 0.15% :     0.000057s :      1: jit_opt_b
 1.18% :     0.000453s :      1: loop_unroll
 1.37% :     0.000526s :      1: mutable_eliminate
 7.07% :     0.002720s :     39: opt.transform.jit_opt_a
 0.14% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000030s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000041s :      4: opt.transform.symbol_engine_opt
 1.29% :     0.000494s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000029s :      1: py_interpret_to_execute
 0.03% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 3.52% :     0.001354s :      2: renormalize.infer
 3.07% :     0.001182s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000040s :      1: rewriter_after_opt_a
 0.28% :     0.000106s :      1: rewriter_before_opt_a
 0.22% :     0.000086s :      1: symbol_engine_optimizer
17.81% :     0.006849s :      1: task_emit
32.50% :     0.012499s :      1: type_inference
 0.15% :     0.000057s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.192.542 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6c072700> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.193.232 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0140346, [30]
[bootstrap]: 0.00030741
[type_inference]: 0.00431584
[event_method]: 1.71e-05
[auto_monad]: 6.284e-05
[graph_reusing]: 6.02999e-06
[pre_auto_parallel]: 3.02002e-06
[py_interpret_to_execute]: 3.126e-05
[rewriter_before_opt_a]: 7.159e-05
[expand_dump_flag]: 3.25e-06
[jit_opt_a]: 0.00679931, [2]
    [Cycle 1]: 0.00271667, [27]
        [switch_simplify]: 5.919e-05
        [loop_unroll]: 3.011e-05
        [a_1]: 0.00055185
        [with_stream_mark]: 2.052e-05
        [recompute_prepare]: 7.6e-06
        [updatestate_depend_eliminate]: 3.48e-06
        [updatestate_assign_eliminate]: 3.11001e-06
        [updatestate_loads_eliminate]: 2.89001e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 5.94999e-06
        [updatestate_useless_node_eliminater]: 5.10999e-06
        [accelerated_algorithm]: 5.31998e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 5.26002e-06
        [merge_forward]: 3.63999e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.575e-05
        [j_node_and_user_rematch]: 9.14e-06
        [meta_fg_expand]: 2.32999e-06
        [replace_old_param]: 8.55001e-06
        [inline_without_move]: 5.30001e-06
        [renormalize]: 0.00169627
        [add_forward_monad_depend]: 8.72e-06
        [auto_monad_grad]: 2.93998e-06
        [auto_monad_eliminator]: 1.85e-05
        [cse]: 3.233e-05
        [replace_applicator]: 1.899e-05
    [Cycle 2]: 0.00035158, [27]
        [switch_simplify]: 6.56999e-06
        [loop_unroll]: 5.30001e-06
        [a_1]: 9.516e-05
        [with_stream_mark]: 1.403e-05
        [recompute_prepare]: 5.07999e-06
        [updatestate_depend_eliminate]: 3.18e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 1.45999e-06
        [specialize_transform]: 5.67999e-06
        [updatestate_useless_node_eliminater]: 5.07999e-06
        [accelerated_algorithm]: 5.55001e-06
        [meta_shard_fg_expand]: 2.34999e-06
        [get_grad_eliminate_]: 4.54998e-06
        [merge_forward]: 3.39001e-06
        [cell_reuse_recompute_pass]: 3.10002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.752e-05
        [j_node_and_user_rematch]: 8.43999e-06
        [meta_fg_expand]: 2.01998e-06
        [replace_old_param]: 8.25999e-06
        [inline_without_move]: 4.74e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 5.59e-06
        [cse]: 1.072e-05
        [replace_applicator]: 5.62999e-06
[py_interpret_to_execute_after_opt_a]: 1.388e-05
[rewriter_after_opt_a]: 3.611e-05
[convert_after_rewriter]: 6.84999e-06
[order_py_execute_after_rewriter]: 5.64998e-06
[mutable_eliminate]: 0.00075183
[jit_opt_b]: 5.382e-05, [1]
    [Cycle 1]: 4.547e-05, [2]
        [frontend_op_eliminate]: 1.706e-05
        [inline_after_opt_a]: 1.567e-05
[cconv]: 2.639e-05
[loop_unroll]: 0.00049308
[jit_opt_after_cconv]: 0.00015037, [1]
    [Cycle 1]: 0.00014313, [11]
        [c_1]: 2.169e-05
        [parameter_eliminate]: 3.4e-06
        [updatestate_depend_eliminate]: 6.39999e-06
        [updatestate_assign_eliminate]: 2.54001e-06
        [updatestate_loads_eliminate]: 2.23998e-06
        [cse]: 2.203e-05
        [call_graph_tuple_transform]: 1.976e-05
        [tuple_list_get_item_eliminator]: 5.64e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 7.50006e-07
        [switch_simplify]: 5.60001e-06
[remove_dup_value]: 1.412e-05
[partial_unused_args_eliminate]: 2.18002e-06
[environ_conv]: 5.04998e-06
[add_recomputation]: 4.796e-05
[cse_after_recomputation]: 2.267e-05, [1]
    [Cycle 1]: 1.64e-05, [1]
        [cse]: 9.99001e-06
[auto_monad_reorder]: 1.512e-05
[get_jit_bprop_graph]: 1.86e-06
[rewriter_after_jit_bprop_graph]: 6.21998e-06
[opt_after_jit_grad]: 0.00048068
[symbol_engine_optimizer]: 7.39e-05, [1]
    [Cycle 1]: 6.738e-05, [6]
        [build]: 4.04997e-06
        [elim_shapecalc]: 7.82e-06
        [elim_not_effective]: 1.246e-05
        [opt_reshape]: 6.07001e-06
        [fold_const_symbol]: 8.89e-06
        [renormalize]: 8.09989e-07
[validate]: 3.861e-05
Sums
    bootstrap                            :     0.000307s :  3.17%
    type_inference                       :     0.004316s : 44.51%
    event_method                         :     0.000017s :  0.18%
    auto_monad                           :     0.000063s :  0.65%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000031s :  0.32%
    rewriter_before_opt_a                :     0.000072s :  0.74%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000066s :  0.68%
    jit_opt_a.loop_unroll                :     0.000035s :  0.37%
    jit_opt_a.a_1                        :     0.000647s :  6.67%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.36%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.17%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.001696s : 17.49%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.25%
    jit_opt_a.cse                        :     0.000043s :  0.44%
    jit_opt_a.replace_applicator         :     0.000025s :  0.25%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.14%
    rewriter_after_opt_a                 :     0.000036s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000752s :  7.75%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.16%
    cconv                                :     0.000026s :  0.27%
    loop_unroll                          :     0.000493s :  5.08%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000014s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000048s :  0.49%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000481s :  4.96%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.40%


Time group info:
------[substitution.]   0.000220    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.66% :     0.000006s :      3: substitution.graph_param_transform
84.80% :     0.000187s :      4: substitution.inline
 1.48% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.21% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000005s :      2: substitution.replace_old_param
 5.34% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004254     2
80.65% :     0.003431s :      1: type_inference.infer
19.35% :     0.000823s :      1: type_inference.specialize
------[replace.]   0.000067     6
65.96% :     0.000044s :      4: replace.inline
34.04% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000195     6
94.51% :     0.000184s :      4: match.inline
 5.49% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.21% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.06% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000001s :     10: predicate.addn_zero_filter
 2.12% :     0.000003s :     10: predicate.arithmetic_simplify
 1.19% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.17% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.58% :     0.000007s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.86% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 2.00% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.06% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.86% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.84% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.43% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.53% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.65% :     0.000009s :     53: predicate.switch_simplify
 1.43% :     0.000002s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.53% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.18% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.79% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.98% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000565    12
47.16% :     0.000267s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.54% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.29% :     0.000256s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016660    72
 0.30% :     0.000051s :      1: add_recomputation
 0.40% :     0.000067s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 1.92% :     0.000319s :      1: bootstrap
 0.18% :     0.000029s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.12% :     0.000021s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
40.84% :     0.006803s :      1: jit_opt_a
 0.92% :     0.000153s :      1: jit_opt_after_cconv
 0.34% :     0.000057s :      1: jit_opt_b
 3.01% :     0.000502s :      1: loop_unroll
 4.58% :     0.000763s :      1: mutable_eliminate
 5.24% :     0.000873s :     26: opt.transform.jit_opt_a
 0.29% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.93% :     0.000489s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000034s :      1: py_interpret_to_execute
 0.10% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.27% :     0.000379s :      1: renormalize.infer
 7.85% :     0.001307s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000039s :      1: rewriter_after_opt_a
 0.45% :     0.000074s :      1: rewriter_before_opt_a
 0.46% :     0.000077s :      1: symbol_engine_optimizer
25.99% :     0.004329s :      1: type_inference


TotalTime = 0.0212889, [33]
[bootstrap]: 0.0003164
[type_inference]: 0.00534491
[event_method]: 1.729e-05
[auto_monad]: 6.358e-05
[graph_reusing]: 5.38997e-06
[pre_auto_parallel]: 2.83e-06
[py_interpret_to_execute]: 3.661e-05
[rewriter_before_opt_a]: 7.232e-05
[expand_dump_flag]: 2.76e-06
[jit_opt_a]: 0.00582376, [2]
    [Cycle 1]: 0.00190334, [27]
        [switch_simplify]: 6.057e-05
        [loop_unroll]: 2.868e-05
        [a_1]: 0.00059444
        [with_stream_mark]: 2.223e-05
        [recompute_prepare]: 7.48e-06
        [updatestate_depend_eliminate]: 3.95e-06
        [updatestate_assign_eliminate]: 3.70998e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 2.39999e-06
        [specialize_transform]: 5.79999e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.36998e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 3.69002e-06
        [cell_reuse_recompute_pass]: 1.93997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.618e-05
        [j_node_and_user_rematch]: 9.29998e-06
        [meta_fg_expand]: 2.11e-06
        [replace_old_param]: 9.47999e-06
        [inline_without_move]: 5.30001e-06
        [renormalize]: 0.00086028
        [add_forward_monad_depend]: 6.44001e-06
        [auto_monad_grad]: 2.12001e-06
        [auto_monad_eliminator]: 1.465e-05
        [cse]: 2.776e-05
        [replace_applicator]: 1.347e-05
    [Cycle 2]: 0.00033307, [27]
        [switch_simplify]: 5.37999e-06
        [loop_unroll]: 4.71002e-06
        [a_1]: 9.241e-05
        [with_stream_mark]: 1.209e-05
        [recompute_prepare]: 5.37001e-06
        [updatestate_depend_eliminate]: 3.01001e-06
        [updatestate_assign_eliminate]: 2.74001e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 5.19998e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 5.70001e-06
        [meta_shard_fg_expand]: 1.83002e-06
        [get_grad_eliminate_]: 4.76002e-06
        [merge_forward]: 2.63e-06
        [cell_reuse_recompute_pass]: 2.46e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.549e-05
        [j_node_and_user_rematch]: 8.51002e-06
        [meta_fg_expand]: 1.76e-06
        [replace_old_param]: 7.81001e-06
        [inline_without_move]: 4.76002e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.17999e-06
        [auto_monad_grad]: 1.01002e-06
        [auto_monad_eliminator]: 5.71e-06
        [cse]: 1.078e-05
        [replace_applicator]: 5.57999e-06
[py_interpret_to_execute_after_opt_a]: 1.034e-05
[rewriter_after_opt_a]: 3.733e-05
[convert_after_rewriter]: 7.4e-06
[order_py_execute_after_rewriter]: 4.53999e-06
[mutable_eliminate]: 0.00073846
[jit_opt_b]: 5.263e-05, [1]
    [Cycle 1]: 4.443e-05, [2]
        [frontend_op_eliminate]: 1.634e-05
        [inline_after_opt_a]: 1.567e-05
[cconv]: 3.287e-05
[loop_unroll]: 0.00048837
[jit_opt_after_cconv]: 0.00015119, [1]
    [Cycle 1]: 0.00014461, [11]
        [c_1]: 2.193e-05
        [parameter_eliminate]: 3.17002e-06
        [updatestate_depend_eliminate]: 6.46e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [cse]: 2.379e-05
        [call_graph_tuple_transform]: 2.052e-05
        [tuple_list_get_item_eliminator]: 5.45001e-06
        [none_parameter_eliminate]: 1.66002e-06
        [renormalize]: 3.29979e-07
        [switch_simplify]: 5.97001e-06
[remove_dup_value]: 1.403e-05
[partial_unused_args_eliminate]: 2.50002e-06
[environ_conv]: 7.59002e-06
[add_recomputation]: 6.974e-05
[cse_after_recomputation]: 2.292e-05, [1]
    [Cycle 1]: 1.671e-05, [1]
        [cse]: 1.08e-05
[auto_monad_reorder]: 1.754e-05
[get_jit_bprop_graph]: 1.90001e-06
[rewriter_after_jit_bprop_graph]: 7.90998e-06
[opt_after_jit_grad]: 0.00052418
[symbol_engine_optimizer]: 8.004e-05, [1]
    [Cycle 1]: 7.328e-05, [6]
        [build]: 5.46e-06
        [elim_shapecalc]: 8.84998e-06
        [elim_not_effective]: 1.361e-05
        [opt_reshape]: 5.62999e-06
        [fold_const_symbol]: 9.51e-06
        [renormalize]: 6.69999e-07
[validate]: 4.637e-05
[backend_pass]: 8.99978e-07
[task_emit]: 0.0070553
[execute]: 8.47998e-06
Sums
    bootstrap                            :     0.000316s :  1.85%
    type_inference                       :     0.005345s : 31.33%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000064s :  0.37%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.21%
    rewriter_before_opt_a                :     0.000072s :  0.42%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.39%
    jit_opt_a.loop_unroll                :     0.000033s :  0.20%
    jit_opt_a.a_1                        :     0.000687s :  4.03%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000860s :  5.04%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000039s :  0.23%
    jit_opt_a.replace_applicator         :     0.000019s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000037s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000738s :  4.33%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000033s :  0.19%
    loop_unroll                          :     0.000488s :  2.86%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000008s :  0.04%
    add_recomputation                    :     0.000070s :  0.41%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000018s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000524s :  3.07%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007055s : 41.36%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000242    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.79% :     0.000002s :      2: substitution.fold_const_symbol
 2.25% :     0.000005s :      3: substitution.graph_param_transform
85.57% :     0.000207s :      4: substitution.inline
 1.31% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.24% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.79% :     0.000004s :      2: substitution.replace_old_param
 5.20% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005272     2
81.44% :     0.004294s :      1: type_inference.infer
18.56% :     0.000978s :      1: type_inference.specialize
------[replace.]   0.000075     6
67.01% :     0.000050s :      4: replace.inline
32.99% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000215     6
94.65% :     0.000204s :      4: match.inline
 5.35% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.52% :     0.000002s :     10: predicate.addn_zero_filter
 2.78% :     0.000003s :     10: predicate.arithmetic_simplify
 1.14% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.44% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.19% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 6.26% :     0.000008s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :      6: predicate.less_batch_normalization
 1.70% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.74% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.20% :     0.000001s :     10: predicate.minmaximum_grad
 1.39% :     0.000002s :      3: predicate.mutable_eliminate
 0.39% :     0.000000s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.69% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.23% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.94% :     0.000010s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.54% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.59% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.94% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.52% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000635    12
45.90% :     0.000291s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.21% :     0.000046s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.89% :     0.000298s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023164    76
 0.31% :     0.000073s :      1: add_recomputation
 0.29% :     0.000067s :      1: auto_monad
 0.09% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.42% :     0.000330s :      1: bootstrap
 0.15% :     0.000036s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
25.16% :     0.005828s :      1: jit_opt_a
 0.67% :     0.000154s :      1: jit_opt_after_cconv
 0.26% :     0.000061s :      1: jit_opt_b
 2.15% :     0.000498s :      1: loop_unroll
 3.26% :     0.000756s :      1: mutable_eliminate
 3.91% :     0.000907s :     26: opt.transform.jit_opt_a
 0.22% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.31% :     0.000535s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000039s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.14% :     0.000495s :      1: renormalize.infer
 1.54% :     0.000357s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000040s :      1: rewriter_after_opt_a
 0.33% :     0.000076s :      1: rewriter_before_opt_a
 0.36% :     0.000083s :      1: symbol_engine_optimizer
30.52% :     0.007070s :      1: task_emit
23.13% :     0.005358s :      1: type_inference
 0.29% :     0.000067s :      1: validate


TotalTime = 0.0135081, [30]
[bootstrap]: 0.00035516
[type_inference]: 0.00488951
[event_method]: 1.893e-05
[auto_monad]: 5.649e-05
[graph_reusing]: 5.49998e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 3.451e-05
[rewriter_before_opt_a]: 7.127e-05
[expand_dump_flag]: 2.71e-06
[jit_opt_a]: 0.00590502, [2]
    [Cycle 1]: 0.00162495, [27]
        [switch_simplify]: 5.135e-05
        [loop_unroll]: 3.063e-05
        [a_1]: 0.00052044
        [with_stream_mark]: 1.334e-05
        [recompute_prepare]: 7.33999e-06
        [updatestate_depend_eliminate]: 3.28e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 1.52999e-06
        [specialize_transform]: 6.05002e-06
        [updatestate_useless_node_eliminater]: 5.11002e-06
        [accelerated_algorithm]: 5.67001e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 2.59999e-06
        [cell_reuse_recompute_pass]: 1.33002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.279e-05
        [j_node_and_user_rematch]: 9.27999e-06
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 8.86002e-06
        [inline_without_move]: 5.47001e-06
        [renormalize]: 0.00069109
        [add_forward_monad_depend]: 5.99e-06
        [auto_monad_grad]: 2.22001e-06
        [auto_monad_eliminator]: 1.382e-05
        [cse]: 2.039e-05
        [replace_applicator]: 1.271e-05
    [Cycle 2]: 0.00033885, [27]
        [switch_simplify]: 5.63997e-06
        [loop_unroll]: 5.54998e-06
        [a_1]: 9.453e-05
        [with_stream_mark]: 9.71998e-06
        [recompute_prepare]: 5.57999e-06
        [updatestate_depend_eliminate]: 2.66e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 5.49998e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 5.42001e-06
        [meta_shard_fg_expand]: 1.11002e-06
        [get_grad_eliminate_]: 4.58999e-06
        [merge_forward]: 2.32999e-06
        [cell_reuse_recompute_pass]: 2.12001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.31e-05
        [j_node_and_user_rematch]: 8.22e-06
        [meta_fg_expand]: 2.07999e-06
        [replace_old_param]: 1.4e-05
        [inline_without_move]: 4.99003e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.74998e-06
        [auto_monad_grad]: 6.50005e-07
        [auto_monad_eliminator]: 5.56002e-06
        [cse]: 1.14e-05
        [replace_applicator]: 5.15999e-06
[py_interpret_to_execute_after_opt_a]: 1.042e-05
[rewriter_after_opt_a]: 3.167e-05
[convert_after_rewriter]: 7.45e-06
[order_py_execute_after_rewriter]: 5.40999e-06
[mutable_eliminate]: 0.00061505
[jit_opt_b]: 5.176e-05, [1]
    [Cycle 1]: 4.532e-05, [2]
        [frontend_op_eliminate]: 1.825e-05
        [inline_after_opt_a]: 1.557e-05
[cconv]: 2.351e-05
[loop_unroll]: 0.00043882
[jit_opt_after_cconv]: 0.00014658, [1]
    [Cycle 1]: 0.00013948, [11]
        [c_1]: 2.25e-05
        [parameter_eliminate]: 2.79999e-06
        [updatestate_depend_eliminate]: 6.07999e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [cse]: 2.02e-05
        [call_graph_tuple_transform]: 1.905e-05
        [tuple_list_get_item_eliminator]: 5.62999e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 5.90022e-07
        [switch_simplify]: 5.42001e-06
[remove_dup_value]: 1.186e-05
[partial_unused_args_eliminate]: 1.92999e-06
[environ_conv]: 4.12e-06
[add_recomputation]: 3.913e-05
[cse_after_recomputation]: 2.029e-05, [1]
    [Cycle 1]: 1.514e-05, [1]
        [cse]: 9.49e-06
[auto_monad_reorder]: 1.284e-05
[get_jit_bprop_graph]: 1.44998e-06
[rewriter_after_jit_bprop_graph]: 5.97999e-06
[opt_after_jit_grad]: 0.00046177
[symbol_engine_optimizer]: 7.513e-05, [1]
    [Cycle 1]: 6.774e-05, [6]
        [build]: 3.53e-06
        [elim_shapecalc]: 7.71001e-06
        [elim_not_effective]: 1.273e-05
        [opt_reshape]: 5.93002e-06
        [fold_const_symbol]: 9.13002e-06
        [renormalize]: 4.2998e-07
[validate]: 3.12e-05
Sums
    bootstrap                            :     0.000355s :  3.96%
    type_inference                       :     0.004890s : 54.46%
    event_method                         :     0.000019s :  0.21%
    auto_monad                           :     0.000056s :  0.63%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.38%
    rewriter_before_opt_a                :     0.000071s :  0.79%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000057s :  0.63%
    jit_opt_a.loop_unroll                :     0.000036s :  0.40%
    jit_opt_a.a_1                        :     0.000615s :  6.85%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000023s :  0.25%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000691s :  7.70%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.22%
    jit_opt_a.cse                        :     0.000032s :  0.35%
    jit_opt_a.replace_applicator         :     0.000018s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000032s :  0.35%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000615s :  6.85%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.17%
    cconv                                :     0.000024s :  0.26%
    loop_unroll                          :     0.000439s :  4.89%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000012s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000039s :  0.44%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000013s :  0.14%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000462s :  5.14%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.35%


Time group info:
------[substitution.]   0.000190    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000001s :      2: substitution.fold_const_symbol
 2.74% :     0.000005s :      3: substitution.graph_param_transform
80.34% :     0.000152s :      4: substitution.inline
 1.78% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.51% :     0.000005s :      4: substitution.remove_not_recompute_node
 5.34% :     0.000010s :      2: substitution.replace_old_param
 5.49% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004823     2
81.08% :     0.003910s :      1: type_inference.infer
18.92% :     0.000913s :      1: type_inference.specialize
------[replace.]   0.000067     6
65.91% :     0.000044s :      4: replace.inline
34.09% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000159     6
94.21% :     0.000150s :      4: match.inline
 5.79% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.25% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.09% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 2.43% :     0.000003s :     10: predicate.arithmetic_simplify
 1.26% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.66% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.61% :     0.000007s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.86% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.89% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.20% :     0.000001s :     10: predicate.minmaximum_grad
 1.44% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000002s :     15: predicate.partial_eliminate
 1.34% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.98% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.54% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.68% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.59% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.14% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 2.02% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.45% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000626    12
50.53% :     0.000317s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.55% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.92% :     0.000281s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015094    72
 0.28% :     0.000042s :      1: add_recomputation
 0.40% :     0.000060s :      1: auto_monad
 0.10% :     0.000015s :      1: auto_monad_reorder
 2.45% :     0.000370s :      1: bootstrap
 0.17% :     0.000026s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.15% :     0.005909s :      1: jit_opt_a
 0.99% :     0.000149s :      1: jit_opt_after_cconv
 0.36% :     0.000054s :      1: jit_opt_b
 2.96% :     0.000447s :      1: loop_unroll
 4.13% :     0.000623s :      1: mutable_eliminate
 5.51% :     0.000832s :     26: opt.transform.jit_opt_a
 0.32% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000027s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.11% :     0.000469s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000037s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000015s :      1: remove_dup_value
 2.10% :     0.000317s :      1: renormalize.infer
 2.43% :     0.000367s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000035s :      1: rewriter_after_opt_a
 0.49% :     0.000074s :      1: rewriter_before_opt_a
 0.51% :     0.000078s :      1: symbol_engine_optimizer
32.51% :     0.004907s :      1: type_inference


TotalTime = 0.033856, [33]
[bootstrap]: 0.00026254
[type_inference]: 0.0129513
[event_method]: 5.388e-05
[auto_monad]: 0.00010796
[graph_reusing]: 7.31001e-06
[pre_auto_parallel]: 3.16999e-06
[py_interpret_to_execute]: 2.926e-05
[rewriter_before_opt_a]: 0.00010369
[expand_dump_flag]: 3.41999e-06
[jit_opt_a]: 0.0112177, [3]
    [Cycle 1]: 0.00547992, [27]
        [switch_simplify]: 0.0001233
        [loop_unroll]: 4.19e-05
        [a_1]: 0.00091048
        [with_stream_mark]: 2.903e-05
        [recompute_prepare]: 1.906e-05
        [updatestate_depend_eliminate]: 7.77e-06
        [updatestate_assign_eliminate]: 6.93998e-06
        [updatestate_loads_eliminate]: 6.33e-06
        [parameter_eliminate]: 2.79999e-06
        [specialize_transform]: 1.272e-05
        [updatestate_useless_node_eliminater]: 1.18e-05
        [accelerated_algorithm]: 1.195e-05
        [meta_shard_fg_expand]: 3.38999e-06
        [get_grad_eliminate_]: 1.202e-05
        [merge_forward]: 7.81001e-06
        [cell_reuse_recompute_pass]: 9.49978e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.586e-05
        [j_node_and_user_rematch]: 2.228e-05
        [meta_fg_expand]: 0.00141576
        [replace_old_param]: 5.011e-05
        [inline_without_move]: 5.262e-05
        [renormalize]: 0.00221931
        [add_forward_monad_depend]: 9.06002e-06
        [auto_monad_grad]: 5.78002e-06
        [auto_monad_eliminator]: 4.601e-05
        [cse]: 0.00014285
        [replace_applicator]: 6.201e-05
    [Cycle 2]: 0.00190853, [27]
        [switch_simplify]: 3.528e-05
        [loop_unroll]: 3.321e-05
        [a_1]: 0.00096715
        [with_stream_mark]: 1.062e-05
        [recompute_prepare]: 7.34002e-06
        [updatestate_depend_eliminate]: 3.56999e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 6.44001e-06
        [updatestate_useless_node_eliminater]: 5.97001e-06
        [accelerated_algorithm]: 5.77999e-06
        [meta_shard_fg_expand]: 1.53002e-06
        [get_grad_eliminate_]: 5.54e-06
        [merge_forward]: 2.91999e-06
        [cell_reuse_recompute_pass]: 7.89994e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.296e-05
        [j_node_and_user_rematch]: 9.41998e-06
        [meta_fg_expand]: 0.00014942
        [replace_old_param]: 1.173e-05
        [inline_without_move]: 6.28e-06
        [renormalize]: 0.00043005
        [add_forward_monad_depend]: 4.44998e-06
        [auto_monad_grad]: 2.19999e-06
        [auto_monad_eliminator]: 1.08e-05
        [cse]: 1.93e-05
        [replace_applicator]: 1.271e-05
    [Cycle 3]: 0.0003542, [27]
        [switch_simplify]: 6.81999e-06
        [loop_unroll]: 5.62001e-06
        [a_1]: 0.00011401
        [with_stream_mark]: 8.62e-06
        [recompute_prepare]: 5.67999e-06
        [updatestate_depend_eliminate]: 3.38e-06
        [updatestate_assign_eliminate]: 2.72001e-06
        [updatestate_loads_eliminate]: 2.55002e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 6.02999e-06
        [updatestate_useless_node_eliminater]: 5.78002e-06
        [accelerated_algorithm]: 5.55001e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 5.71e-06
        [merge_forward]: 2.79001e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.438e-05
        [j_node_and_user_rematch]: 9.81e-06
        [meta_fg_expand]: 1.77999e-06
        [replace_old_param]: 8.11002e-06
        [inline_without_move]: 5.58002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.00001e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 5.90002e-06
        [cse]: 1.341e-05
        [replace_applicator]: 5.82001e-06
[py_interpret_to_execute_after_opt_a]: 1.316e-05
[rewriter_after_opt_a]: 3.816e-05
[convert_after_rewriter]: 7.6e-06
[order_py_execute_after_rewriter]: 5.66e-06
[mutable_eliminate]: 0.00052608
[jit_opt_b]: 5.304e-05, [1]
    [Cycle 1]: 4.605e-05, [2]
        [frontend_op_eliminate]: 1.78e-05
        [inline_after_opt_a]: 1.744e-05
[cconv]: 2.222e-05
[loop_unroll]: 0.0004256
[jit_opt_after_cconv]: 0.00015415, [1]
    [Cycle 1]: 0.00014798, [11]
        [c_1]: 2.457e-05
        [parameter_eliminate]: 2.58e-06
        [updatestate_depend_eliminate]: 5.81e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.68003e-06
        [cse]: 2.626e-05
        [call_graph_tuple_transform]: 2.023e-05
        [tuple_list_get_item_eliminator]: 6.21998e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 6.41e-06
[remove_dup_value]: 1.622e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 5.18002e-06
[add_recomputation]: 5.21e-05
[cse_after_recomputation]: 2.729e-05, [1]
    [Cycle 1]: 2.148e-05, [1]
        [cse]: 1.502e-05
[auto_monad_reorder]: 1.854e-05
[get_jit_bprop_graph]: 1.79e-06
[rewriter_after_jit_bprop_graph]: 5.32999e-06
[opt_after_jit_grad]: 0.00045553
[symbol_engine_optimizer]: 7.662e-05, [1]
    [Cycle 1]: 7.076e-05, [6]
        [build]: 3.89002e-06
        [elim_shapecalc]: 8.28999e-06
        [elim_not_effective]: 1.42e-05
        [opt_reshape]: 6.32001e-06
        [fold_const_symbol]: 1.036e-05
        [renormalize]: 4.30009e-07
[validate]: 3.605e-05
[backend_pass]: 9.79984e-07
[task_emit]: 0.00694589
[execute]: 7.15e-06
Sums
    bootstrap                            :     0.000263s :  0.89%
    type_inference                       :     0.012951s : 43.80%
    event_method                         :     0.000054s :  0.18%
    auto_monad                           :     0.000108s :  0.37%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000029s :  0.10%
    rewriter_before_opt_a                :     0.000104s :  0.35%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000165s :  0.56%
    jit_opt_a.loop_unroll                :     0.000081s :  0.27%
    jit_opt_a.a_1                        :     0.001992s :  6.74%
    jit_opt_a.with_stream_mark           :     0.000048s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000025s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000023s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.08%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000053s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.001567s :  5.30%
    jit_opt_a.replace_old_param          :     0.000070s :  0.24%
    jit_opt_a.inline_without_move        :     0.000064s :  0.22%
    jit_opt_a.renormalize                :     0.002649s :  8.96%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000063s :  0.21%
    jit_opt_a.cse                        :     0.000176s :  0.59%
    jit_opt_a.replace_applicator         :     0.000081s :  0.27%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.04%
    rewriter_after_opt_a                 :     0.000038s :  0.13%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000526s :  1.78%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.06%
    cconv                                :     0.000022s :  0.08%
    loop_unroll                          :     0.000426s :  1.44%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000052s :  0.18%
    cse_after_recomputation.cse          :     0.000015s :  0.05%
    auto_monad_reorder                   :     0.000019s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000456s :  1.54%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006946s : 23.49%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000474   113
 0.45% :     0.000002s :      3: substitution.elim_not_effective
 0.33% :     0.000002s :      3: substitution.fold_const_symbol
 1.21% :     0.000006s :      4: substitution.graph_param_transform
63.80% :     0.000302s :     14: substitution.inline
 3.39% :     0.000016s :      2: substitution.inline_without_move
 1.82% :     0.000009s :     14: substitution.j_node_and_user_rematch
 1.95% :     0.000009s :      7: substitution.minmaximum_grad
 1.18% :     0.000006s :      5: substitution.partial_eliminate
 2.28% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.06% :     0.000024s :     10: substitution.replace_applicator
 1.51% :     0.000007s :      7: substitution.replace_old_param
 0.54% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.69% :     0.000008s :      1: substitution.switch_simplify
 4.00% :     0.000019s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.75% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 8.04% :     0.000038s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012854     2
89.36% :     0.011486s :      1: type_inference.infer
10.64% :     0.001368s :      1: type_inference.specialize
------[replace.]   0.000196    22
58.85% :     0.000115s :     14: replace.inline
17.30% :     0.000034s :      1: replace.switch_simplify
23.85% :     0.000047s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000323    22
91.27% :     0.000295s :     14: match.inline
 2.31% :     0.000007s :      1: match.switch_simplify
 6.42% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000368  2634
 1.64% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.39% :     0.000001s :      4: predicate.ad_related_special_op_eliminate
 1.46% :     0.000005s :     44: predicate.addn_check_dump
 1.70% :     0.000006s :     44: predicate.addn_zero_filter
 1.90% :     0.000007s :     44: predicate.arithmetic_simplify
 1.50% :     0.000006s :     44: predicate.cast_eliminate
 0.17% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.48% :     0.000005s :     44: predicate.compare_switch_simplify
 1.61% :     0.000006s :     44: predicate.depend_value_elim
 1.44% :     0.000005s :     44: predicate.dict_get_item_const_eliminator
 1.48% :     0.000005s :     44: predicate.dict_get_item_eliminator
 1.56% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.31% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      4: predicate.elim_not_effective
 0.18% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.50% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.45% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.43% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.48% :     0.000005s :     44: predicate.environ_get_eliminate
 1.59% :     0.000006s :     44: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.86% :     0.000003s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.42% :     0.000016s :     73: predicate.inline
 1.86% :     0.000007s :     40: predicate.inline_without_move
 0.34% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.86% :     0.000003s :     19: predicate.less_batch_normalization
 1.90% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.92% :     0.000007s :     55: predicate.load_eliminater
 0.61% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.29% :     0.000012s :     91: predicate.loop_unroll_before_grad
 1.77% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.45% :     0.000005s :     44: predicate.merge_addn
 1.52% :     0.000006s :     44: predicate.minmaximum_grad
 0.57% :     0.000002s :      4: predicate.mutable_eliminate
 0.20% :     0.000001s :      4: predicate.opt_reshape
 2.33% :     0.000009s :     55: predicate.partial_eliminate
 1.54% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.80% :     0.000007s :     44: predicate.reduce_eliminate
 1.76% :     0.000006s :     51: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.59% :     0.000010s :     97: predicate.replace_applicator
 0.89% :     0.000003s :     40: predicate.replace_old_param
 0.11% :     0.000000s :      4: predicate.reset_defer_inline
 1.65% :     0.000006s :     44: predicate.reshape_eliminate
 1.72% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.27% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.61% :     0.000006s :     44: predicate.same_eliminate
 0.45% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.35% :     0.000001s :      8: predicate.special_op_eliminate
 0.86% :     0.000003s :     19: predicate.specialize_transform
 1.66% :     0.000006s :     44: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.16% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.81% :     0.000010s :     65: predicate.switch_defer_inline
 2.64% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.66% :     0.000024s :    162: predicate.switch_simplify
 1.50% :     0.000006s :     44: predicate.tile_eliminate
 1.49% :     0.000005s :     44: predicate.transpose_eliminate
 1.77% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.61% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 3.16% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 1.86% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.91% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000007s :     44: predicate.value_based_eliminate
 0.14% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.22% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001529    32
61.19% :     0.000936s :     14: func_graph_cloner_run.FuncGraphClonerGraph
38.81% :     0.000593s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.039251    91
 0.14% :     0.000055s :      1: add_recomputation
 0.29% :     0.000113s :      1: auto_monad
 0.05% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.74% :     0.000292s :      1: bootstrap
 0.06% :     0.000025s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000030s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.15% :     0.000059s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
28.59% :     0.011221s :      1: jit_opt_a
 0.40% :     0.000157s :      1: jit_opt_after_cconv
 0.14% :     0.000056s :      1: jit_opt_b
 1.10% :     0.000433s :      1: loop_unroll
 1.36% :     0.000535s :      1: mutable_eliminate
 6.71% :     0.002633s :     39: opt.transform.jit_opt_a
 0.14% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000029s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000036s :      4: opt.transform.symbol_engine_opt
 1.18% :     0.000463s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000032s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 3.74% :     0.001466s :      2: renormalize.infer
 2.98% :     0.001170s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000041s :      1: rewriter_after_opt_a
 0.27% :     0.000106s :      1: rewriter_before_opt_a
 0.20% :     0.000079s :      1: symbol_engine_optimizer
17.72% :     0.006957s :      1: task_emit
33.04% :     0.012970s :      1: type_inference
 0.14% :     0.000056s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.234.255 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6c072b80> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.234.823 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.025337, [33]
[bootstrap]: 0.00034568
[type_inference]: 0.00839754
[event_method]: 1.628e-05
[auto_monad]: 6.882e-05
[graph_reusing]: 5.14e-06
[pre_auto_parallel]: 2.54999e-06
[py_interpret_to_execute]: 4.746e-05
[rewriter_before_opt_a]: 0.00014037
[expand_dump_flag]: 3.38e-06
[jit_opt_a]: 0.00597455, [2]
    [Cycle 1]: 0.0019472, [27]
        [switch_simplify]: 6.003e-05
        [loop_unroll]: 3.149e-05
        [a_1]: 0.00061877
        [with_stream_mark]: 1.825e-05
        [recompute_prepare]: 7.98001e-06
        [updatestate_depend_eliminate]: 3.16999e-06
        [updatestate_assign_eliminate]: 3.03e-06
        [updatestate_loads_eliminate]: 2.93e-06
        [parameter_eliminate]: 2.12001e-06
        [specialize_transform]: 6.48e-06
        [updatestate_useless_node_eliminater]: 5.71998e-06
        [accelerated_algorithm]: 5.86998e-06
        [meta_shard_fg_expand]: 2.12999e-06
        [get_grad_eliminate_]: 5.44e-06
        [merge_forward]: 2.95998e-06
        [cell_reuse_recompute_pass]: 1.02998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.408e-05
        [j_node_and_user_rematch]: 9.18002e-06
        [meta_fg_expand]: 2.56e-06
        [replace_old_param]: 8.94e-06
        [inline_without_move]: 5.33002e-06
        [renormalize]: 0.00089597
        [add_forward_monad_depend]: 8.01001e-06
        [auto_monad_grad]: 3.04999e-06
        [auto_monad_eliminator]: 1.565e-05
        [cse]: 2.076e-05
        [replace_applicator]: 1.507e-05
    [Cycle 2]: 0.00033979, [27]
        [switch_simplify]: 5.71e-06
        [loop_unroll]: 5.45001e-06
        [a_1]: 9.605e-05
        [with_stream_mark]: 1.068e-05
        [recompute_prepare]: 5.68002e-06
        [updatestate_depend_eliminate]: 3.11999e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.01998e-06
        [parameter_eliminate]: 1.44e-06
        [specialize_transform]: 4.83001e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.29998e-06
        [meta_shard_fg_expand]: 1.57001e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.74999e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.296e-05
        [j_node_and_user_rematch]: 8.65999e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 7.61001e-06
        [inline_without_move]: 4.76002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.51002e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.86e-06
        [cse]: 1.046e-05
        [replace_applicator]: 5.58002e-06
[py_interpret_to_execute_after_opt_a]: 1.177e-05
[rewriter_after_opt_a]: 3.554e-05
[convert_after_rewriter]: 7.25e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00065616
[jit_opt_b]: 5.363e-05, [1]
    [Cycle 1]: 4.577e-05, [2]
        [frontend_op_eliminate]: 1.776e-05
        [inline_after_opt_a]: 1.617e-05
[cconv]: 2.789e-05
[loop_unroll]: 0.00044243
[jit_opt_after_cconv]: 0.00014778, [1]
    [Cycle 1]: 0.00014085, [11]
        [c_1]: 2.169e-05
        [parameter_eliminate]: 3.49001e-06
        [updatestate_depend_eliminate]: 7.38e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 2.23002e-06
        [cse]: 2.004e-05
        [call_graph_tuple_transform]: 2.01e-05
        [tuple_list_get_item_eliminator]: 5.35001e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 7.59988e-07
        [switch_simplify]: 5.53002e-06
[remove_dup_value]: 2.385e-05
[partial_unused_args_eliminate]: 2.73e-06
[environ_conv]: 5.56002e-06
[add_recomputation]: 5.018e-05
[cse_after_recomputation]: 2.259e-05, [1]
    [Cycle 1]: 1.585e-05, [1]
        [cse]: 9.34998e-06
[auto_monad_reorder]: 1.564e-05
[get_jit_bprop_graph]: 1.87001e-06
[rewriter_after_jit_bprop_graph]: 5.42999e-06
[opt_after_jit_grad]: 0.00049643
[symbol_engine_optimizer]: 7.941e-05, [1]
    [Cycle 1]: 7.25e-05, [6]
        [build]: 3.85e-06
        [elim_shapecalc]: 7.51001e-06
        [elim_not_effective]: 1.497e-05
        [opt_reshape]: 6.33998e-06
        [fold_const_symbol]: 9.97001e-06
        [renormalize]: 1.08001e-06
[validate]: 4.307e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.0079789
[execute]: 7.75e-06
Sums
    bootstrap                            :     0.000346s :  1.64%
    type_inference                       :     0.008398s : 39.96%
    event_method                         :     0.000016s :  0.08%
    auto_monad                           :     0.000069s :  0.33%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000047s :  0.23%
    rewriter_before_opt_a                :     0.000140s :  0.67%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.31%
    jit_opt_a.loop_unroll                :     0.000037s :  0.18%
    jit_opt_a.a_1                        :     0.000715s :  3.40%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.06%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.05%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.05%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.08%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.08%
    jit_opt_a.inline_without_move        :     0.000010s :  0.05%
    jit_opt_a.renormalize                :     0.000896s :  4.26%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.10%
    jit_opt_a.cse                        :     0.000031s :  0.15%
    jit_opt_a.replace_applicator         :     0.000021s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.06%
    rewriter_after_opt_a                 :     0.000036s :  0.17%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000656s :  3.12%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.08%
    cconv                                :     0.000028s :  0.13%
    loop_unroll                          :     0.000442s :  2.11%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000024s :  0.11%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000050s :  0.24%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000016s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000496s :  2.36%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.20%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007979s : 37.96%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000265    23
 1.11% :     0.000003s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.05% :     0.000005s :      3: substitution.graph_param_transform
86.71% :     0.000230s :      4: substitution.inline
 1.30% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.72% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.59% :     0.000004s :      2: substitution.replace_old_param
 5.02% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008324     2
49.20% :     0.004095s :      1: type_inference.infer
50.80% :     0.004229s :      1: type_inference.specialize
------[replace.]   0.000075     6
67.50% :     0.000051s :      4: replace.inline
32.50% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000239     6
94.93% :     0.000227s :      4: match.inline
 5.07% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.49% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000002s :     10: predicate.addn_zero_filter
 2.54% :     0.000003s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.49% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.10% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 5.75% :     0.000007s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.77% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.95% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.92% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.69% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.03% :     0.000003s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.29% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.50% :     0.000009s :     53: predicate.switch_simplify
 1.42% :     0.000002s :     10: predicate.tile_eliminate
 1.35% :     0.000002s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.93% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.05% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000715    12
42.83% :     0.000306s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.38% :     0.000046s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.78% :     0.000363s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027274    76
 0.19% :     0.000053s :      1: add_recomputation
 0.26% :     0.000072s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.32% :     0.000361s :      1: bootstrap
 0.11% :     0.000031s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.07% :     0.000020s :      1: event_method
 0.04% :     0.000011s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
21.92% :     0.005978s :      1: jit_opt_a
 0.55% :     0.000151s :      1: jit_opt_after_cconv
 0.21% :     0.000056s :      1: jit_opt_b
 1.65% :     0.000451s :      1: loop_unroll
 2.44% :     0.000666s :      1: mutable_eliminate
 3.44% :     0.000937s :     26: opt.transform.jit_opt_a
 0.18% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000028s :      4: opt.transform.jit_opt_b
 0.04% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000035s :      4: opt.transform.symbol_engine_opt
 1.85% :     0.000505s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000050s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000027s :      1: remove_dup_value
 1.83% :     0.000500s :      1: renormalize.infer
 1.42% :     0.000387s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000039s :      1: rewriter_after_opt_a
 0.53% :     0.000144s :      1: rewriter_before_opt_a
 0.30% :     0.000083s :      1: symbol_engine_optimizer
29.31% :     0.007993s :      1: task_emit
30.83% :     0.008410s :      1: type_inference
 0.24% :     0.000065s :      1: validate


TotalTime = 0.0141173, [30]
[bootstrap]: 0.00032685
[type_inference]: 0.00486544
[event_method]: 1.893e-05
[auto_monad]: 6.788e-05
[graph_reusing]: 6.09001e-06
[pre_auto_parallel]: 2.76999e-06
[py_interpret_to_execute]: 3.616e-05
[rewriter_before_opt_a]: 7.752e-05
[expand_dump_flag]: 3.5e-06
[jit_opt_a]: 0.00609441, [2]
    [Cycle 1]: 0.00176275, [27]
        [switch_simplify]: 5.994e-05
        [loop_unroll]: 3.294e-05
        [a_1]: 0.00061277
        [with_stream_mark]: 1.908e-05
        [recompute_prepare]: 7.36999e-06
        [updatestate_depend_eliminate]: 4.03001e-06
        [updatestate_assign_eliminate]: 2.88e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 2.26e-06
        [specialize_transform]: 6.24001e-06
        [updatestate_useless_node_eliminater]: 5.17999e-06
        [accelerated_algorithm]: 5.87001e-06
        [meta_shard_fg_expand]: 2.07999e-06
        [get_grad_eliminate_]: 5.56e-06
        [merge_forward]: 3.60998e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.453e-05
        [j_node_and_user_rematch]: 9.97001e-06
        [meta_fg_expand]: 2.67001e-06
        [replace_old_param]: 9.24998e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 0.00069644
        [add_forward_monad_depend]: 5.74e-06
        [auto_monad_grad]: 1.92999e-06
        [auto_monad_eliminator]: 1.457e-05
        [cse]: 2.872e-05
        [replace_applicator]: 1.233e-05
    [Cycle 2]: 0.0003432, [27]
        [switch_simplify]: 5.55001e-06
        [loop_unroll]: 5.07999e-06
        [a_1]: 9.492e-05
        [with_stream_mark]: 1.111e-05
        [recompute_prepare]: 5.03002e-06
        [updatestate_depend_eliminate]: 2.43998e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [parameter_eliminate]: 1.54e-06
        [specialize_transform]: 5.24e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 5.33002e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 4.90001e-06
        [merge_forward]: 3.25e-06
        [cell_reuse_recompute_pass]: 1.78002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.401e-05
        [j_node_and_user_rematch]: 8.75001e-06
        [meta_fg_expand]: 1.85001e-06
        [replace_old_param]: 8.10999e-06
        [inline_without_move]: 4.86002e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.54e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 5.72001e-06
        [cse]: 1.17e-05
        [replace_applicator]: 5.15999e-06
[py_interpret_to_execute_after_opt_a]: 1.093e-05
[rewriter_after_opt_a]: 3.816e-05
[convert_after_rewriter]: 7.58001e-06
[order_py_execute_after_rewriter]: 7.51999e-06
[mutable_eliminate]: 0.00074182
[jit_opt_b]: 5.77e-05, [1]
    [Cycle 1]: 4.913e-05, [2]
        [frontend_op_eliminate]: 1.931e-05
        [inline_after_opt_a]: 1.727e-05
[cconv]: 3.078e-05
[loop_unroll]: 0.00051485
[jit_opt_after_cconv]: 0.00017389, [1]
    [Cycle 1]: 0.00016552, [11]
        [c_1]: 2.359e-05
        [parameter_eliminate]: 3.86001e-06
        [updatestate_depend_eliminate]: 7.31999e-06
        [updatestate_assign_eliminate]: 3.41001e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [cse]: 3.232e-05
        [call_graph_tuple_transform]: 2.361e-05
        [tuple_list_get_item_eliminator]: 5.51e-06
        [none_parameter_eliminate]: 1.69998e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 5.96e-06
[remove_dup_value]: 1.572e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 5.80002e-06
[add_recomputation]: 5.384e-05
[cse_after_recomputation]: 2.316e-05, [1]
    [Cycle 1]: 1.717e-05, [1]
        [cse]: 9.97001e-06
[auto_monad_reorder]: 1.564e-05
[get_jit_bprop_graph]: 2.63998e-06
[rewriter_after_jit_bprop_graph]: 7.66001e-06
[opt_after_jit_grad]: 0.0005854
[symbol_engine_optimizer]: 8.927e-05, [1]
    [Cycle 1]: 8.124e-05, [6]
        [build]: 6.41998e-06
        [elim_shapecalc]: 9.25999e-06
        [elim_not_effective]: 1.703e-05
        [opt_reshape]: 6.99001e-06
        [fold_const_symbol]: 1.048e-05
        [renormalize]: 8.59989e-07
[validate]: 4.146e-05
Sums
    bootstrap                            :     0.000327s :  3.44%
    type_inference                       :     0.004865s : 51.26%
    event_method                         :     0.000019s :  0.20%
    auto_monad                           :     0.000068s :  0.72%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.38%
    rewriter_before_opt_a                :     0.000078s :  0.82%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000065s :  0.69%
    jit_opt_a.loop_unroll                :     0.000038s :  0.40%
    jit_opt_a.a_1                        :     0.000708s :  7.46%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.18%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000697s :  7.34%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.21%
    jit_opt_a.cse                        :     0.000040s :  0.43%
    jit_opt_a.replace_applicator         :     0.000017s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000038s :  0.40%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000008s :  0.08%
    mutable_eliminate                    :     0.000742s :  7.82%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.18%
    cconv                                :     0.000031s :  0.32%
    loop_unroll                          :     0.000515s :  5.42%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000032s :  0.34%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.25%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000016s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000054s :  0.57%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.16%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.08%
    opt_after_jit_grad                   :     0.000585s :  6.17%
    symbol_engine_optimizer.build        :     0.000006s :  0.07%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.18%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.44%


Time group info:
------[substitution.]   0.000217    23
 1.23% :     0.000003s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.80% :     0.000006s :      3: substitution.graph_param_transform
83.21% :     0.000181s :      4: substitution.inline
 1.62% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.19% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.29% :     0.000005s :      2: substitution.replace_old_param
 5.99% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004796     2
81.94% :     0.003930s :      1: type_inference.infer
18.06% :     0.000866s :      1: type_inference.specialize
------[replace.]   0.000074     6
66.76% :     0.000049s :      4: replace.inline
33.24% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000190     6
93.73% :     0.000178s :      4: match.inline
 6.27% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   691
 1.25% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.75% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.43% :     0.000002s :     10: predicate.addn_zero_filter
 2.12% :     0.000003s :     10: predicate.arithmetic_simplify
 1.22% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.46% :     0.000002s :     10: predicate.environ_get_eliminate
 1.13% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.71% :     0.000008s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.68% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.30% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.09% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000002s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 2.02% :     0.000003s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.73% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.67% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.54% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.80% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.23% :     0.000002s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.22% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.42% :     0.000010s :     53: predicate.switch_simplify
 1.19% :     0.000002s :     10: predicate.tile_eliminate
 1.21% :     0.000002s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.58% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.43% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000600    12
51.45% :     0.000309s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.09% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.46% :     0.000261s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015818    72
 0.36% :     0.000057s :      1: add_recomputation
 0.45% :     0.000071s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.18% :     0.000345s :      1: bootstrap
 0.21% :     0.000033s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.55% :     0.006098s :      1: jit_opt_a
 1.12% :     0.000177s :      1: jit_opt_after_cconv
 0.38% :     0.000061s :      1: jit_opt_b
 3.32% :     0.000525s :      1: loop_unroll
 4.78% :     0.000756s :      1: mutable_eliminate
 5.88% :     0.000930s :     26: opt.transform.jit_opt_a
 0.35% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000030s :      4: opt.transform.jit_opt_b
 0.09% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000040s :      4: opt.transform.symbol_engine_opt
 3.77% :     0.000596s :      1: opt_after_jit_grad
 0.06% :     0.000010s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000039s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000019s :      1: remove_dup_value
 2.38% :     0.000376s :      1: renormalize.infer
 1.98% :     0.000313s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000041s :      1: rewriter_after_opt_a
 0.51% :     0.000081s :      1: rewriter_before_opt_a
 0.58% :     0.000092s :      1: symbol_engine_optimizer
30.85% :     0.004880s :      1: type_inference


TotalTime = 0.0326005, [33]
[bootstrap]: 0.00020738
[type_inference]: 0.0119473
[event_method]: 7.306e-05
[auto_monad]: 0.00010374
[graph_reusing]: 7.71999e-06
[pre_auto_parallel]: 2.85002e-06
[py_interpret_to_execute]: 2.558e-05
[rewriter_before_opt_a]: 0.00010748
[expand_dump_flag]: 3.48e-06
[jit_opt_a]: 0.0107583, [3]
    [Cycle 1]: 0.00531487, [27]
        [switch_simplify]: 0.0001177
        [loop_unroll]: 4.238e-05
        [a_1]: 0.00082554
        [with_stream_mark]: 2.244e-05
        [recompute_prepare]: 1.715e-05
        [updatestate_depend_eliminate]: 7.63999e-06
        [updatestate_assign_eliminate]: 6.23e-06
        [updatestate_loads_eliminate]: 5.66e-06
        [parameter_eliminate]: 2.36e-06
        [specialize_transform]: 1.258e-05
        [updatestate_useless_node_eliminater]: 1.16e-05
        [accelerated_algorithm]: 1.189e-05
        [meta_shard_fg_expand]: 3.45e-06
        [get_grad_eliminate_]: 1.129e-05
        [merge_forward]: 7.78001e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.615e-05
        [j_node_and_user_rematch]: 2.206e-05
        [meta_fg_expand]: 0.00133542
        [replace_old_param]: 9.281e-05
        [inline_without_move]: 5.094e-05
        [renormalize]: 0.00218465
        [add_forward_monad_depend]: 8.92e-06
        [auto_monad_grad]: 5.07e-06
        [auto_monad_eliminator]: 4.571e-05
        [cse]: 0.00016116
        [replace_applicator]: 6.07e-05
    [Cycle 2]: 0.00195432, [27]
        [switch_simplify]: 3.517e-05
        [loop_unroll]: 3.281e-05
        [a_1]: 0.00103576
        [with_stream_mark]: 1.256e-05
        [recompute_prepare]: 7.68999e-06
        [updatestate_depend_eliminate]: 3.51999e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.55002e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 7.65e-06
        [updatestate_useless_node_eliminater]: 5.92001e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 5.63002e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 7.59988e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.357e-05
        [j_node_and_user_rematch]: 9.77001e-06
        [meta_fg_expand]: 5.81e-05
        [replace_old_param]: 1.812e-05
        [inline_without_move]: 6.86999e-06
        [renormalize]: 0.00048933
        [add_forward_monad_depend]: 4.02e-06
        [auto_monad_grad]: 1.10999e-06
        [auto_monad_eliminator]: 1.073e-05
        [cse]: 1.978e-05
        [replace_applicator]: 1.16e-05
    [Cycle 3]: 0.0003507, [27]
        [switch_simplify]: 6.43e-06
        [loop_unroll]: 6.17001e-06
        [a_1]: 0.00011499
        [with_stream_mark]: 7.93999e-06
        [recompute_prepare]: 5.67999e-06
        [updatestate_depend_eliminate]: 3.38e-06
        [updatestate_assign_eliminate]: 2.57001e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 5.95002e-06
        [updatestate_useless_node_eliminater]: 5.72999e-06
        [accelerated_algorithm]: 5.81003e-06
        [meta_shard_fg_expand]: 1.22999e-06
        [get_grad_eliminate_]: 5.47001e-06
        [merge_forward]: 2.78e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.409e-05
        [j_node_and_user_rematch]: 9.48002e-06
        [meta_fg_expand]: 1.82001e-06
        [replace_old_param]: 7.89002e-06
        [inline_without_move]: 5.72001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.05001e-06
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 5.99999e-06
        [cse]: 1.285e-05
        [replace_applicator]: 5.61e-06
[py_interpret_to_execute_after_opt_a]: 1.01e-05
[rewriter_after_opt_a]: 3.734e-05
[convert_after_rewriter]: 7.31999e-06
[order_py_execute_after_rewriter]: 5.63997e-06
[mutable_eliminate]: 0.00048175
[jit_opt_b]: 5.308e-05, [1]
    [Cycle 1]: 4.645e-05, [2]
        [frontend_op_eliminate]: 1.835e-05
        [inline_after_opt_a]: 1.718e-05
[cconv]: 2.057e-05
[loop_unroll]: 0.00044782
[jit_opt_after_cconv]: 0.00015044, [1]
    [Cycle 1]: 0.00014425, [11]
        [c_1]: 2.601e-05
        [parameter_eliminate]: 2.36998e-06
        [updatestate_depend_eliminate]: 5.71e-06
        [updatestate_assign_eliminate]: 2.92002e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [cse]: 2.149e-05
        [call_graph_tuple_transform]: 2.022e-05
        [tuple_list_get_item_eliminator]: 6.06998e-06
        [none_parameter_eliminate]: 1.68997e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 6.12001e-06
[remove_dup_value]: 1.599e-05
[partial_unused_args_eliminate]: 2.50002e-06
[environ_conv]: 5.24998e-06
[add_recomputation]: 4.815e-05
[cse_after_recomputation]: 2.428e-05, [1]
    [Cycle 1]: 1.901e-05, [1]
        [cse]: 1.349e-05
[auto_monad_reorder]: 1.754e-05
[get_jit_bprop_graph]: 1.98997e-06
[rewriter_after_jit_bprop_graph]: 5.40999e-06
[opt_after_jit_grad]: 0.00054306
[symbol_engine_optimizer]: 7.723e-05, [1]
    [Cycle 1]: 7.149e-05, [6]
        [build]: 3.87002e-06
        [elim_shapecalc]: 8.63001e-06
        [elim_not_effective]: 1.486e-05
        [opt_reshape]: 7e-06
        [fold_const_symbol]: 1.035e-05
        [renormalize]: 4.30009e-07
[validate]: 3.367e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.00716677
[execute]: 8.97e-06
Sums
    bootstrap                            :     0.000207s :  0.72%
    type_inference                       :     0.011947s : 41.66%
    event_method                         :     0.000073s :  0.25%
    auto_monad                           :     0.000104s :  0.36%
    graph_reusing                        :     0.000008s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.09%
    rewriter_before_opt_a                :     0.000107s :  0.37%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000159s :  0.56%
    jit_opt_a.loop_unroll                :     0.000081s :  0.28%
    jit_opt_a.a_1                        :     0.001976s :  6.89%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000031s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000023s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000023s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000022s :  0.08%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000054s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000041s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.001395s :  4.87%
    jit_opt_a.replace_old_param          :     0.000119s :  0.41%
    jit_opt_a.inline_without_move        :     0.000064s :  0.22%
    jit_opt_a.renormalize                :     0.002674s :  9.32%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000062s :  0.22%
    jit_opt_a.cse                        :     0.000194s :  0.68%
    jit_opt_a.replace_applicator         :     0.000078s :  0.27%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.04%
    rewriter_after_opt_a                 :     0.000037s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000482s :  1.68%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.06%
    cconv                                :     0.000021s :  0.07%
    loop_unroll                          :     0.000448s :  1.56%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000048s :  0.17%
    cse_after_recomputation.cse          :     0.000013s :  0.05%
    auto_monad_reorder                   :     0.000018s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000543s :  1.89%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007167s : 24.99%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000465   113
 0.46% :     0.000002s :      3: substitution.elim_not_effective
 0.36% :     0.000002s :      3: substitution.fold_const_symbol
 1.27% :     0.000006s :      4: substitution.graph_param_transform
59.51% :     0.000277s :     14: substitution.inline
 3.45% :     0.000016s :      2: substitution.inline_without_move
 1.84% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.20% :     0.000010s :      7: substitution.minmaximum_grad
 1.23% :     0.000006s :      5: substitution.partial_eliminate
 2.19% :     0.000010s :     14: substitution.remove_not_recompute_node
 5.35% :     0.000025s :     10: substitution.replace_applicator
 1.53% :     0.000007s :      7: substitution.replace_old_param
 0.48% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.69% :     0.000008s :      1: substitution.switch_simplify
 3.88% :     0.000018s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.80% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
11.76% :     0.000055s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011881     2
90.14% :     0.010709s :      1: type_inference.infer
 9.86% :     0.001172s :      1: type_inference.specialize
------[replace.]   0.000172    22
52.48% :     0.000090s :     14: replace.inline
18.98% :     0.000033s :      1: replace.switch_simplify
28.54% :     0.000049s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000296    22
90.96% :     0.000270s :     14: match.inline
 2.46% :     0.000007s :      1: match.switch_simplify
 6.58% :     0.000019s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000366  2634
 1.58% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.44% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.43% :     0.000005s :     44: predicate.addn_check_dump
 1.70% :     0.000006s :     44: predicate.addn_zero_filter
 2.01% :     0.000007s :     44: predicate.arithmetic_simplify
 1.60% :     0.000006s :     44: predicate.cast_eliminate
 0.18% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.43% :     0.000005s :     44: predicate.compare_switch_simplify
 1.55% :     0.000006s :     44: predicate.depend_value_elim
 1.61% :     0.000006s :     44: predicate.dict_get_item_const_eliminator
 1.50% :     0.000005s :     44: predicate.dict_get_item_eliminator
 1.59% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.35% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.17% :     0.000001s :      4: predicate.elim_not_effective
 0.21% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.58% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.48% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.56% :     0.000006s :     44: predicate.environ_get_depend_swap
 1.53% :     0.000006s :     44: predicate.environ_get_eliminate
 1.46% :     0.000005s :     44: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.80% :     0.000003s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.36% :     0.000016s :     73: predicate.inline
 1.93% :     0.000007s :     40: predicate.inline_without_move
 0.35% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.84% :     0.000003s :     19: predicate.less_batch_normalization
 1.82% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.99% :     0.000007s :     55: predicate.load_eliminater
 0.53% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.42% :     0.000013s :     91: predicate.loop_unroll_before_grad
 1.80% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.46% :     0.000005s :     44: predicate.merge_addn
 1.52% :     0.000006s :     44: predicate.minmaximum_grad
 0.52% :     0.000002s :      4: predicate.mutable_eliminate
 0.19% :     0.000001s :      4: predicate.opt_reshape
 2.29% :     0.000008s :     55: predicate.partial_eliminate
 1.65% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.78% :     0.000006s :     44: predicate.reduce_eliminate
 1.78% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.44% :     0.000009s :     97: predicate.replace_applicator
 0.91% :     0.000003s :     40: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.62% :     0.000006s :     44: predicate.reshape_eliminate
 1.57% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.25% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.54% :     0.000006s :     44: predicate.same_eliminate
 0.44% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.34% :     0.000001s :      8: predicate.special_op_eliminate
 0.85% :     0.000003s :     19: predicate.specialize_transform
 1.59% :     0.000006s :     44: predicate.split_environ_get_set_with_tuple_value
 1.48% :     0.000005s :     44: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.85% :     0.000010s :     65: predicate.switch_defer_inline
 2.60% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.65% :     0.000024s :    162: predicate.switch_simplify
 1.56% :     0.000006s :     44: predicate.tile_eliminate
 1.48% :     0.000005s :     44: predicate.transpose_eliminate
 1.80% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.61% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 3.01% :     0.000011s :     59: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000006s :     44: predicate.tuple_list_set_item_eliminator
 1.80% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.93% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000006s :     44: predicate.value_based_eliminate
 0.14% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.26% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001446    32
62.63% :     0.000906s :     14: func_graph_cloner_run.FuncGraphClonerGraph
37.37% :     0.000540s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038052    91
 0.13% :     0.000051s :      1: add_recomputation
 0.28% :     0.000108s :      1: auto_monad
 0.05% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.59% :     0.000223s :      1: bootstrap
 0.06% :     0.000023s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000026s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.21% :     0.000079s :      1: event_method
 0.04% :     0.000014s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
28.28% :     0.010761s :      1: jit_opt_a
 0.40% :     0.000153s :      1: jit_opt_after_cconv
 0.15% :     0.000056s :      1: jit_opt_b
 1.20% :     0.000456s :      1: loop_unroll
 1.29% :     0.000490s :      1: mutable_eliminate
 6.98% :     0.002656s :     39: opt.transform.jit_opt_a
 0.15% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000030s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000037s :      4: opt.transform.symbol_engine_opt
 1.45% :     0.000551s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000028s :      1: py_interpret_to_execute
 0.03% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 3.79% :     0.001442s :      2: renormalize.infer
 3.20% :     0.001219s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000040s :      1: rewriter_after_opt_a
 0.29% :     0.000110s :      1: rewriter_before_opt_a
 0.21% :     0.000080s :      1: symbol_engine_optimizer
18.87% :     0.007182s :      1: task_emit
31.43% :     0.011959s :      1: type_inference
 0.14% :     0.000055s :      1: validate


TotalTime = 0.0210168, [33]
[bootstrap]: 0.00033276
[type_inference]: 0.00501766
[event_method]: 1.724e-05
[auto_monad]: 6.96e-05
[graph_reusing]: 5.22999e-06
[pre_auto_parallel]: 3.03998e-06
[py_interpret_to_execute]: 4.376e-05
[rewriter_before_opt_a]: 7.201e-05
[expand_dump_flag]: 2.94001e-06
[jit_opt_a]: 0.00588785, [2]
    [Cycle 1]: 0.00183387, [27]
        [switch_simplify]: 5.567e-05
        [loop_unroll]: 2.897e-05
        [a_1]: 0.00057176
        [with_stream_mark]: 1.911e-05
        [recompute_prepare]: 7.31001e-06
        [updatestate_depend_eliminate]: 3.38999e-06
        [updatestate_assign_eliminate]: 2.89999e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 2.11e-06
        [specialize_transform]: 5.59e-06
        [updatestate_useless_node_eliminater]: 4.71002e-06
        [accelerated_algorithm]: 5.10999e-06
        [meta_shard_fg_expand]: 3.2e-06
        [get_grad_eliminate_]: 4.89e-06
        [merge_forward]: 3.54002e-06
        [cell_reuse_recompute_pass]: 1.28002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.341e-05
        [j_node_and_user_rematch]: 8.08001e-06
        [meta_fg_expand]: 2.02001e-06
        [replace_old_param]: 9.27001e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.00084536
        [add_forward_monad_depend]: 5.62001e-06
        [auto_monad_grad]: 2.98998e-06
        [auto_monad_eliminator]: 1.406e-05
        [cse]: 2.117e-05
        [replace_applicator]: 1.349e-05
    [Cycle 2]: 0.00032006, [27]
        [switch_simplify]: 5.67001e-06
        [loop_unroll]: 5.07e-06
        [a_1]: 8.83e-05
        [with_stream_mark]: 1.005e-05
        [recompute_prepare]: 4.87998e-06
        [updatestate_depend_eliminate]: 2.75997e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 9.20001e-07
        [specialize_transform]: 4.53999e-06
        [updatestate_useless_node_eliminater]: 4.52e-06
        [accelerated_algorithm]: 4.85001e-06
        [meta_shard_fg_expand]: 1.06997e-06
        [get_grad_eliminate_]: 4.43001e-06
        [merge_forward]: 2.62001e-06
        [cell_reuse_recompute_pass]: 2.69001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.403e-05
        [j_node_and_user_rematch]: 7.68001e-06
        [meta_fg_expand]: 1.34e-06
        [replace_old_param]: 1.286e-05
        [inline_without_move]: 4.69998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.42e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 4.92e-06
        [cse]: 1.021e-05
        [replace_applicator]: 4.73001e-06
[py_interpret_to_execute_after_opt_a]: 1.106e-05
[rewriter_after_opt_a]: 3.52e-05
[convert_after_rewriter]: 5.022e-05
[order_py_execute_after_rewriter]: 1.149e-05
[mutable_eliminate]: 0.0006462
[jit_opt_b]: 5.035e-05, [1]
    [Cycle 1]: 4.337e-05, [2]
        [frontend_op_eliminate]: 1.693e-05
        [inline_after_opt_a]: 1.434e-05
[cconv]: 2.663e-05
[loop_unroll]: 0.00044463
[jit_opt_after_cconv]: 0.00015371, [1]
    [Cycle 1]: 0.00014614, [11]
        [c_1]: 2.148e-05
        [parameter_eliminate]: 3.45e-06
        [updatestate_depend_eliminate]: 6.83e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.20002e-06
        [cse]: 2.316e-05
        [call_graph_tuple_transform]: 2.152e-05
        [tuple_list_get_item_eliminator]: 5.47999e-06
        [none_parameter_eliminate]: 1.46002e-06
        [renormalize]: 7.2e-07
        [switch_simplify]: 5.67001e-06
[remove_dup_value]: 1.3e-05
[partial_unused_args_eliminate]: 2.59999e-06
[environ_conv]: 5.61e-06
[add_recomputation]: 5.26e-05
[cse_after_recomputation]: 2.239e-05, [1]
    [Cycle 1]: 1.568e-05, [1]
        [cse]: 9.82999e-06
[auto_monad_reorder]: 1.579e-05
[get_jit_bprop_graph]: 2.16e-06
[rewriter_after_jit_bprop_graph]: 7.81001e-06
[opt_after_jit_grad]: 0.00051167
[symbol_engine_optimizer]: 8.095e-05, [1]
    [Cycle 1]: 7.258e-05, [6]
        [build]: 3.9e-06
        [elim_shapecalc]: 7.66001e-06
        [elim_not_effective]: 1.479e-05
        [opt_reshape]: 6.47001e-06
        [fold_const_symbol]: 8.99998e-06
        [renormalize]: 9.20001e-07
[validate]: 4.055e-05
[backend_pass]: 1.24e-06
[task_emit]: 0.00716666
[execute]: 7.25e-06
Sums
    bootstrap                            :     0.000333s :  2.00%
    type_inference                       :     0.005018s : 30.11%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000070s :  0.42%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000044s :  0.26%
    rewriter_before_opt_a                :     0.000072s :  0.43%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.37%
    jit_opt_a.loop_unroll                :     0.000034s :  0.20%
    jit_opt_a.a_1                        :     0.000660s :  3.96%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000022s :  0.13%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000845s :  5.07%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.11%
    jit_opt_a.cse                        :     0.000031s :  0.19%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.21%
    convert_after_rewriter               :     0.000050s :  0.30%
    order_py_execute_after_rewriter      :     0.000011s :  0.07%
    mutable_eliminate                    :     0.000646s :  3.88%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000027s :  0.16%
    loop_unroll                          :     0.000445s :  2.67%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000053s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000512s :  3.07%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007167s : 43.00%
    execute                              :     0.000007s :  0.04%


Time group info:
------[substitution.]   0.000242    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000002s :      2: substitution.fold_const_symbol
 2.29% :     0.000006s :      3: substitution.graph_param_transform
82.99% :     0.000201s :      4: substitution.inline
 1.23% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.88% :     0.000005s :      4: substitution.remove_not_recompute_node
 4.19% :     0.000010s :      2: substitution.replace_old_param
 5.80% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004949     2
81.69% :     0.004043s :      1: type_inference.infer
18.31% :     0.000906s :      1: type_inference.specialize
------[replace.]   0.000076     6
66.24% :     0.000050s :      4: replace.inline
33.76% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000211     6
93.89% :     0.000198s :      4: match.inline
 6.11% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.47% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.77% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 2.36% :     0.000003s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.67% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.55% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.34% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.37% :     0.000002s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.44% :     0.000006s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      6: predicate.less_batch_normalization
 1.74% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.32% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.61% :     0.000002s :      3: predicate.mutable_eliminate
 0.65% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.43% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.73% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.24% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.55% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.25% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.58% :     0.000002s :     10: predicate.transpose_eliminate
 1.30% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.16% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000656    12
48.21% :     0.000316s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.92% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.87% :     0.000294s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022840    76
 0.25% :     0.000056s :      1: add_recomputation
 0.32% :     0.000073s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.51% :     0.000346s :      1: bootstrap
 0.13% :     0.000030s :      1: cconv
 0.24% :     0.000054s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
25.79% :     0.005892s :      1: jit_opt_a
 0.68% :     0.000156s :      1: jit_opt_after_cconv
 0.23% :     0.000053s :      1: jit_opt_b
 1.99% :     0.000454s :      1: loop_unroll
 2.87% :     0.000655s :      1: mutable_eliminate
 3.82% :     0.000872s :     26: opt.transform.jit_opt_a
 0.22% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.28% :     0.000521s :      1: opt_after_jit_grad
 0.06% :     0.000014s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000046s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.10% :     0.000479s :      1: renormalize.infer
 1.57% :     0.000358s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000038s :      1: rewriter_after_opt_a
 0.33% :     0.000075s :      1: rewriter_before_opt_a
 0.36% :     0.000083s :      1: symbol_engine_optimizer
31.44% :     0.007180s :      1: task_emit
22.02% :     0.005030s :      1: type_inference
 0.28% :     0.000064s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.274.750 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6c07b0d0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.275.347 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0144856, [30]
[bootstrap]: 0.0003056
[type_inference]: 0.00468181
[event_method]: 2.003e-05
[auto_monad]: 7.133e-05
[graph_reusing]: 6.50002e-06
[pre_auto_parallel]: 2.59001e-06
[py_interpret_to_execute]: 4.14e-05
[rewriter_before_opt_a]: 7.856e-05
[expand_dump_flag]: 3.34001e-06
[jit_opt_a]: 0.00664537, [2]
    [Cycle 1]: 0.00216568, [27]
        [switch_simplify]: 6.822e-05
        [loop_unroll]: 3.194e-05
        [a_1]: 0.00062923
        [with_stream_mark]: 2.596e-05
        [recompute_prepare]: 8.66002e-06
        [updatestate_depend_eliminate]: 4.25999e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 2.67001e-06
        [parameter_eliminate]: 2.63998e-06
        [specialize_transform]: 6.69999e-06
        [updatestate_useless_node_eliminater]: 5.58997e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 2.53e-06
        [get_grad_eliminate_]: 5.76e-06
        [merge_forward]: 4.03001e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.548e-05
        [j_node_and_user_rematch]: 1.075e-05
        [meta_fg_expand]: 2.53e-06
        [replace_old_param]: 9.90002e-06
        [inline_without_move]: 6.19999e-06
        [renormalize]: 0.00098724
        [add_forward_monad_depend]: 1.036e-05
        [auto_monad_grad]: 2.98998e-06
        [auto_monad_eliminator]: 2.116e-05
        [cse]: 3.643e-05
        [replace_applicator]: 2.293e-05
    [Cycle 2]: 0.00038444, [27]
        [switch_simplify]: 6.84999e-06
        [loop_unroll]: 5.42001e-06
        [a_1]: 0.0001017
        [with_stream_mark]: 1.526e-05
        [recompute_prepare]: 5.22e-06
        [updatestate_depend_eliminate]: 3.97e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 3.41001e-06
        [parameter_eliminate]: 1.47001e-06
        [specialize_transform]: 5.44998e-06
        [updatestate_useless_node_eliminater]: 4.91002e-06
        [accelerated_algorithm]: 6.48e-06
        [meta_shard_fg_expand]: 2.16e-06
        [get_grad_eliminate_]: 5.40999e-06
        [merge_forward]: 3.72002e-06
        [cell_reuse_recompute_pass]: 3.56999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.925e-05
        [j_node_and_user_rematch]: 9.38002e-06
        [meta_fg_expand]: 2.01998e-06
        [replace_old_param]: 1.088e-05
        [inline_without_move]: 5.32001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.71002e-06
        [auto_monad_grad]: 1.29e-06
        [auto_monad_eliminator]: 6.43998e-06
        [cse]: 1.407e-05
        [replace_applicator]: 5.74e-06
[py_interpret_to_execute_after_opt_a]: 1.647e-05
[rewriter_after_opt_a]: 4.336e-05
[convert_after_rewriter]: 7.85e-06
[order_py_execute_after_rewriter]: 5.24e-06
[mutable_eliminate]: 0.00081363
[jit_opt_b]: 5.675e-05, [1]
    [Cycle 1]: 4.774e-05, [2]
        [frontend_op_eliminate]: 1.766e-05
        [inline_after_opt_a]: 1.769e-05
[cconv]: 2.988e-05
[loop_unroll]: 0.00049935
[jit_opt_after_cconv]: 0.00016297, [1]
    [Cycle 1]: 0.00015623, [11]
        [c_1]: 2.231e-05
        [parameter_eliminate]: 4.57e-06
        [updatestate_depend_eliminate]: 6.81999e-06
        [updatestate_assign_eliminate]: 3.48999e-06
        [updatestate_loads_eliminate]: 2.28998e-06
        [cse]: 2.825e-05
        [call_graph_tuple_transform]: 2.368e-05
        [tuple_list_get_item_eliminator]: 5.67001e-06
        [none_parameter_eliminate]: 1.74e-06
        [renormalize]: 9.89996e-07
        [switch_simplify]: 5.22e-06
[remove_dup_value]: 1.593e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 5.89999e-06
[add_recomputation]: 5.175e-05
[cse_after_recomputation]: 2.221e-05, [1]
    [Cycle 1]: 1.612e-05, [1]
        [cse]: 9.82999e-06
[auto_monad_reorder]: 1.545e-05
[get_jit_bprop_graph]: 2.14999e-06
[rewriter_after_jit_bprop_graph]: 7.17002e-06
[opt_after_jit_grad]: 0.00053747
[symbol_engine_optimizer]: 8.55e-05, [1]
    [Cycle 1]: 7.601e-05, [6]
        [build]: 5.30001e-06
        [elim_shapecalc]: 8.57e-06
        [elim_not_effective]: 1.624e-05
        [opt_reshape]: 6.47001e-06
        [fold_const_symbol]: 9.45001e-06
        [renormalize]: 8.40024e-07
[validate]: 4.52e-05
Sums
    bootstrap                            :     0.000306s :  3.15%
    type_inference                       :     0.004682s : 48.29%
    event_method                         :     0.000020s :  0.21%
    auto_monad                           :     0.000071s :  0.74%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000041s :  0.43%
    rewriter_before_opt_a                :     0.000079s :  0.81%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000075s :  0.77%
    jit_opt_a.loop_unroll                :     0.000037s :  0.39%
    jit_opt_a.a_1                        :     0.000731s :  7.54%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.43%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.12%
    jit_opt_a.merge_forward              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.36%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000021s :  0.21%
    jit_opt_a.inline_without_move        :     0.000012s :  0.12%
    jit_opt_a.renormalize                :     0.000987s : 10.18%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.12%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000028s :  0.28%
    jit_opt_a.cse                        :     0.000050s :  0.52%
    jit_opt_a.replace_applicator         :     0.000029s :  0.30%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.17%
    rewriter_after_opt_a                 :     0.000043s :  0.45%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000814s :  8.39%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.18%
    cconv                                :     0.000030s :  0.31%
    loop_unroll                          :     0.000499s :  5.15%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000028s :  0.29%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000016s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000052s :  0.53%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000537s :  5.54%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000045s :  0.47%


Time group info:
------[substitution.]   0.000273    23
 0.90% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000002s :      2: substitution.fold_const_symbol
 2.25% :     0.000006s :      3: substitution.graph_param_transform
84.38% :     0.000231s :      4: substitution.inline
 1.56% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.94% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.63% :     0.000007s :      2: substitution.replace_old_param
 5.76% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004612     2
80.62% :     0.003718s :      1: type_inference.infer
19.38% :     0.000894s :      1: type_inference.specialize
------[replace.]   0.000083     6
68.02% :     0.000057s :      4: replace.inline
31.98% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000242     6
94.01% :     0.000228s :      4: match.inline
 5.99% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   691
 1.18% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.73% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000002s :     10: predicate.addn_zero_filter
 3.05% :     0.000004s :     10: predicate.arithmetic_simplify
 1.29% :     0.000002s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     10: predicate.environ_get_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.40% :     0.000001s :      3: predicate.graph_param_transform
 6.23% :     0.000008s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.59% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.57% :     0.000002s :     15: predicate.load_eliminater
 1.66% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.82% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.74% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.95% :     0.000001s :     10: predicate.merge_addn
 0.95% :     0.000001s :     10: predicate.minmaximum_grad
 1.73% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     15: predicate.partial_eliminate
 1.03% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.92% :     0.000003s :     18: predicate.replace_applicator
 0.97% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.09% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000002s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.23% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.03% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.63% :     0.000010s :     53: predicate.switch_simplify
 1.19% :     0.000002s :     10: predicate.tile_eliminate
 1.29% :     0.000002s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.53% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.73% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.86% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.44% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.35% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000653    12
46.17% :     0.000301s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.85% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.98% :     0.000300s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016507    72
 0.33% :     0.000055s :      1: add_recomputation
 0.45% :     0.000075s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 1.96% :     0.000323s :      1: bootstrap
 0.20% :     0.000032s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
40.29% :     0.006650s :      1: jit_opt_a
 1.00% :     0.000166s :      1: jit_opt_after_cconv
 0.36% :     0.000060s :      1: jit_opt_b
 3.09% :     0.000510s :      1: loop_unroll
 5.01% :     0.000827s :      1: mutable_eliminate
 5.96% :     0.000983s :     26: opt.transform.jit_opt_a
 0.32% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000028s :      4: opt.transform.jit_opt_b
 0.08% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.32% :     0.000548s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000045s :      1: py_interpret_to_execute
 0.12% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 3.05% :     0.000504s :      1: renormalize.infer
 2.85% :     0.000471s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.29% :     0.000047s :      1: rewriter_after_opt_a
 0.49% :     0.000081s :      1: rewriter_before_opt_a
 0.54% :     0.000088s :      1: symbol_engine_optimizer
28.47% :     0.004700s :      1: type_inference


TotalTime = 0.0196461, [33]
[bootstrap]: 0.00032969
[type_inference]: 0.00502375
[event_method]: 1.825e-05
[auto_monad]: 5.751e-05
[graph_reusing]: 5.35999e-06
[pre_auto_parallel]: 3.46001e-06
[py_interpret_to_execute]: 4.158e-05
[rewriter_before_opt_a]: 6.933e-05
[expand_dump_flag]: 2.36e-06
[jit_opt_a]: 0.00571778, [2]
    [Cycle 1]: 0.00176429, [27]
        [switch_simplify]: 5.232e-05
        [loop_unroll]: 0.00011659
        [a_1]: 0.00052688
        [with_stream_mark]: 1.829e-05
        [recompute_prepare]: 8.97e-06
        [updatestate_depend_eliminate]: 3.04001e-06
        [updatestate_assign_eliminate]: 2.19999e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.42e-06
        [specialize_transform]: 6.53e-06
        [updatestate_useless_node_eliminater]: 5.79e-06
        [accelerated_algorithm]: 5.96998e-06
        [meta_shard_fg_expand]: 1.45999e-06
        [get_grad_eliminate_]: 5.71998e-06
        [merge_forward]: 3.2e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.209e-05
        [j_node_and_user_rematch]: 8.25999e-06
        [meta_fg_expand]: 2.13998e-06
        [replace_old_param]: 7.88999e-06
        [inline_without_move]: 5.28002e-06
        [renormalize]: 0.00072509
        [add_forward_monad_depend]: 5.32999e-06
        [auto_monad_grad]: 2.04e-06
        [auto_monad_eliminator]: 1.07e-05
        [cse]: 2.025e-05
        [replace_applicator]: 1.379e-05
    [Cycle 2]: 0.00033593, [27]
        [switch_simplify]: 5.91e-06
        [loop_unroll]: 5.67001e-06
        [a_1]: 9.107e-05
        [with_stream_mark]: 1.119e-05
        [recompute_prepare]: 5.08002e-06
        [updatestate_depend_eliminate]: 2.71e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 9.90025e-07
        [specialize_transform]: 5.59e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.39998e-06
        [meta_shard_fg_expand]: 1.40999e-06
        [get_grad_eliminate_]: 4.69998e-06
        [merge_forward]: 2.64999e-06
        [cell_reuse_recompute_pass]: 2.07999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.294e-05
        [j_node_and_user_rematch]: 8e-06
        [meta_fg_expand]: 1.30999e-06
        [replace_old_param]: 8.32e-06
        [inline_without_move]: 4.84e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.47001e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 5.83997e-06
        [cse]: 1.198e-05
        [replace_applicator]: 4.89e-06
[py_interpret_to_execute_after_opt_a]: 1.149e-05
[rewriter_after_opt_a]: 2.95e-05
[convert_after_rewriter]: 5.67001e-06
[order_py_execute_after_rewriter]: 7.62998e-06
[mutable_eliminate]: 0.00066947
[jit_opt_b]: 4.952e-05, [1]
    [Cycle 1]: 4.251e-05, [2]
        [frontend_op_eliminate]: 1.617e-05
        [inline_after_opt_a]: 1.475e-05
[cconv]: 1.836e-05
[loop_unroll]: 0.00044699
[jit_opt_after_cconv]: 0.00014191, [1]
    [Cycle 1]: 0.00013556, [11]
        [c_1]: 2.088e-05
        [parameter_eliminate]: 3.03998e-06
        [updatestate_depend_eliminate]: 6.04999e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [cse]: 2.091e-05
        [call_graph_tuple_transform]: 1.838e-05
        [tuple_list_get_item_eliminator]: 4.99e-06
        [none_parameter_eliminate]: 9.39996e-07
        [renormalize]: 7.30011e-07
        [switch_simplify]: 5.67999e-06
[remove_dup_value]: 7.08998e-06
[partial_unused_args_eliminate]: 1.60001e-06
[environ_conv]: 3.46999e-06
[add_recomputation]: 3.547e-05
[cse_after_recomputation]: 2.03e-05, [1]
    [Cycle 1]: 1.472e-05, [1]
        [cse]: 9.07001e-06
[auto_monad_reorder]: 1.176e-05
[get_jit_bprop_graph]: 2.02001e-06
[rewriter_after_jit_bprop_graph]: 7.41001e-06
[opt_after_jit_grad]: 0.000513
[symbol_engine_optimizer]: 7.495e-05, [1]
    [Cycle 1]: 6.77e-05, [6]
        [build]: 4.2e-06
        [elim_shapecalc]: 7.92e-06
        [elim_not_effective]: 1.325e-05
        [opt_reshape]: 5.69e-06
        [fold_const_symbol]: 9.02e-06
        [renormalize]: 6.39993e-07
[validate]: 4.021e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00605181
[execute]: 8.18999e-06
Sums
    bootstrap                            :     0.000330s :  2.14%
    type_inference                       :     0.005024s : 32.67%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000058s :  0.37%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000042s :  0.27%
    rewriter_before_opt_a                :     0.000069s :  0.45%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.38%
    jit_opt_a.loop_unroll                :     0.000122s :  0.79%
    jit_opt_a.a_1                        :     0.000618s :  4.02%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000725s :  4.72%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.11%
    jit_opt_a.cse                        :     0.000032s :  0.21%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000029s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000008s :  0.05%
    mutable_eliminate                    :     0.000669s :  4.35%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000018s :  0.12%
    loop_unroll                          :     0.000447s :  2.91%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000007s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000003s :  0.02%
    add_recomputation                    :     0.000035s :  0.23%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.05%
    opt_after_jit_grad                   :     0.000513s :  3.34%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006052s : 39.35%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000190    23
 1.06% :     0.000002s :      2: substitution.elim_not_effective
 0.80% :     0.000002s :      2: substitution.fold_const_symbol
 1.92% :     0.000004s :      3: substitution.graph_param_transform
85.38% :     0.000162s :      4: substitution.inline
 1.49% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.95% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.30% :     0.000004s :      2: substitution.replace_old_param
 5.11% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004905     2
81.71% :     0.004008s :      1: type_inference.infer
18.29% :     0.000897s :      1: type_inference.specialize
------[replace.]   0.000069     6
65.19% :     0.000045s :      4: replace.inline
34.81% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000168     6
95.12% :     0.000160s :      4: match.inline
 4.88% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000201   691
 0.76% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.69% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.65% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000003s :     10: predicate.addn_zero_filter
 1.82% :     0.000004s :     10: predicate.arithmetic_simplify
 0.70% :     0.000001s :     10: predicate.cast_eliminate
 0.22% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.60% :     0.000001s :     10: predicate.compare_switch_simplify
 1.00% :     0.000002s :     10: predicate.depend_value_elim
 0.73% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 0.83% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.71% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.56% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.17% :     0.000000s :      3: predicate.elim_not_effective
 0.35% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.68% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.59% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.64% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.70% :     0.000001s :     10: predicate.environ_get_eliminate
 0.66% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.12% :     0.000000s :      3: predicate.fold_const_symbol
 0.71% :     0.000001s :      6: predicate.get_grad_eliminate
 0.14% :     0.000000s :      3: predicate.graph_param_transform
 3.48% :     0.000007s :     22: predicate.inline
 0.53% :     0.000001s :      6: predicate.inline_without_move
 0.21% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.76% :     0.000002s :      6: predicate.less_batch_normalization
 0.93% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.02% :     0.000002s :     15: predicate.load_eliminater
 0.87% :     0.000002s :      3: predicate.loop_unroll_after_grad
43.28% :     0.000087s :     34: predicate.loop_unroll_before_grad
 1.07% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.63% :     0.000001s :     10: predicate.merge_addn
 0.63% :     0.000001s :     10: predicate.minmaximum_grad
 1.36% :     0.000003s :      3: predicate.mutable_eliminate
 0.27% :     0.000001s :      3: predicate.opt_reshape
 1.32% :     0.000003s :     15: predicate.partial_eliminate
 0.64% :     0.000001s :     10: predicate.print_const_string_wrapper
 0.99% :     0.000002s :     10: predicate.reduce_eliminate
 0.93% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.31% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.10% :     0.000002s :     18: predicate.replace_applicator
 0.31% :     0.000001s :      6: predicate.replace_old_param
 0.18% :     0.000000s :      3: predicate.reset_defer_inline
 0.71% :     0.000001s :     10: predicate.reshape_eliminate
 0.73% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.53% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.78% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.55% :     0.000001s :      6: predicate.special_op_eliminate
 0.63% :     0.000001s :      6: predicate.specialize_transform
 0.83% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 0.73% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.24% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 1.53% :     0.000003s :     16: predicate.switch_defer_inline
 1.22% :     0.000002s :     16: predicate.switch_layer_defer_inline
 4.24% :     0.000009s :     53: predicate.switch_simplify
 0.65% :     0.000001s :     10: predicate.tile_eliminate
 0.72% :     0.000001s :     10: predicate.transpose_eliminate
 0.87% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.68% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 1.87% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 0.86% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 0.84% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 0.96% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 1.62% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.02% :     0.000002s :     10: predicate.value_based_eliminate
 0.20% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.42% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000616    12
48.25% :     0.000297s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.38% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.36% :     0.000285s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021388    76
 0.18% :     0.000039s :      1: add_recomputation
 0.29% :     0.000062s :      1: auto_monad
 0.07% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.60% :     0.000343s :      1: bootstrap
 0.10% :     0.000021s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.75% :     0.005722s :      1: jit_opt_a
 0.68% :     0.000145s :      1: jit_opt_after_cconv
 0.24% :     0.000052s :      1: jit_opt_b
 2.13% :     0.000455s :      1: loop_unroll
 3.18% :     0.000680s :      1: mutable_eliminate
 4.28% :     0.000916s :     26: opt.transform.jit_opt_a
 0.22% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.44% :     0.000521s :      1: opt_after_jit_grad
 0.05% :     0.000010s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.03% :     0.000006s :      1: pre_auto_parallel
 0.21% :     0.000044s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000010s :      1: remove_dup_value
 1.81% :     0.000387s :      1: renormalize.infer
 1.55% :     0.000331s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000033s :      1: rewriter_after_opt_a
 0.34% :     0.000073s :      1: rewriter_before_opt_a
 0.36% :     0.000077s :      1: symbol_engine_optimizer
28.38% :     0.006071s :      1: task_emit
23.57% :     0.005041s :      1: type_inference
 0.28% :     0.000061s :      1: validate


TotalTime = 0.0329017, [33]
[bootstrap]: 0.00023317
[type_inference]: 0.0123159
[event_method]: 5.531e-05
[auto_monad]: 0.00010327
[graph_reusing]: 6.40002e-06
[pre_auto_parallel]: 2.73e-06
[py_interpret_to_execute]: 2.687e-05
[rewriter_before_opt_a]: 0.00010128
[expand_dump_flag]: 3.09999e-06
[jit_opt_a]: 0.0111592, [3]
    [Cycle 1]: 0.00552238, [27]
        [switch_simplify]: 0.00012035
        [loop_unroll]: 4.221e-05
        [a_1]: 0.00084634
        [with_stream_mark]: 2.295e-05
        [recompute_prepare]: 1.736e-05
        [updatestate_depend_eliminate]: 7.38999e-06
        [updatestate_assign_eliminate]: 6.14001e-06
        [updatestate_loads_eliminate]: 5.79e-06
        [parameter_eliminate]: 2.71999e-06
        [specialize_transform]: 1.244e-05
        [updatestate_useless_node_eliminater]: 1.262e-05
        [accelerated_algorithm]: 1.262e-05
        [meta_shard_fg_expand]: 3.97998e-06
        [get_grad_eliminate_]: 1.168e-05
        [merge_forward]: 8.52e-06
        [cell_reuse_recompute_pass]: 1.13001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.524e-05
        [j_node_and_user_rematch]: 2.249e-05
        [meta_fg_expand]: 0.00151607
        [replace_old_param]: 5.117e-05
        [inline_without_move]: 5.091e-05
        [renormalize]: 0.00220883
        [add_forward_monad_depend]: 9.10999e-06
        [auto_monad_grad]: 5.14998e-06
        [auto_monad_eliminator]: 6.757e-05
        [cse]: 0.00014479
        [replace_applicator]: 6.303e-05
    [Cycle 2]: 0.00190789, [27]
        [switch_simplify]: 4.597e-05
        [loop_unroll]: 3.391e-05
        [a_1]: 0.0009826
        [with_stream_mark]: 1.144e-05
        [recompute_prepare]: 7.6e-06
        [updatestate_depend_eliminate]: 3.83999e-06
        [updatestate_assign_eliminate]: 2.88e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 6.83998e-06
        [updatestate_useless_node_eliminater]: 6.17999e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 1.65001e-06
        [get_grad_eliminate_]: 5.52001e-06
        [merge_forward]: 3.11001e-06
        [cell_reuse_recompute_pass]: 7.7e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.345e-05
        [j_node_and_user_rematch]: 1.004e-05
        [meta_fg_expand]: 6.558e-05
        [replace_old_param]: 9.97999e-06
        [inline_without_move]: 6.07999e-06
        [renormalize]: 0.00045267
        [add_forward_monad_depend]: 4.32e-06
        [auto_monad_grad]: 1.09998e-06
        [auto_monad_eliminator]: 1.107e-05
        [cse]: 1.925e-05
        [replace_applicator]: 1.234e-05
    [Cycle 3]: 0.00037546, [27]
        [switch_simplify]: 6.61e-06
        [loop_unroll]: 2.353e-05
        [a_1]: 0.00011602
        [with_stream_mark]: 8.57998e-06
        [recompute_prepare]: 6.09999e-06
        [updatestate_depend_eliminate]: 3.36001e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 6.26e-06
        [updatestate_useless_node_eliminater]: 5.89e-06
        [accelerated_algorithm]: 5.56e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 5.55001e-06
        [merge_forward]: 2.98e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.427e-05
        [j_node_and_user_rematch]: 9.44e-06
        [meta_fg_expand]: 1.82001e-06
        [replace_old_param]: 7.88999e-06
        [inline_without_move]: 5.77001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.08001e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 6.46e-06
        [cse]: 1.331e-05
        [replace_applicator]: 5.89e-06
[py_interpret_to_execute_after_opt_a]: 1.025e-05
[rewriter_after_opt_a]: 3.735e-05
[convert_after_rewriter]: 7.82998e-06
[order_py_execute_after_rewriter]: 5.61e-06
[mutable_eliminate]: 0.00050711
[jit_opt_b]: 5.284e-05, [1]
    [Cycle 1]: 4.621e-05, [2]
        [frontend_op_eliminate]: 1.741e-05
        [inline_after_opt_a]: 1.778e-05
[cconv]: 2.137e-05
[loop_unroll]: 0.00044726
[jit_opt_after_cconv]: 0.0001494, [1]
    [Cycle 1]: 0.00014259, [11]
        [c_1]: 2.496e-05
        [parameter_eliminate]: 2.39001e-06
        [updatestate_depend_eliminate]: 6.07999e-06
        [updatestate_assign_eliminate]: 2.66999e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [cse]: 2.129e-05
        [call_graph_tuple_transform]: 2.008e-05
        [tuple_list_get_item_eliminator]: 6.16e-06
        [none_parameter_eliminate]: 1.69e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 6.07999e-06
[remove_dup_value]: 1.552e-05
[partial_unused_args_eliminate]: 1.96998e-06
[environ_conv]: 5.61e-06
[add_recomputation]: 5.06e-05
[cse_after_recomputation]: 2.397e-05, [1]
    [Cycle 1]: 1.871e-05, [1]
        [cse]: 1.307e-05
[auto_monad_reorder]: 1.847e-05
[get_jit_bprop_graph]: 1.66e-06
[rewriter_after_jit_bprop_graph]: 5.60001e-06
[opt_after_jit_grad]: 0.00052002
[symbol_engine_optimizer]: 7.827e-05, [1]
    [Cycle 1]: 7.2e-05, [6]
        [build]: 3.68999e-06
        [elim_shapecalc]: 8.27998e-06
        [elim_not_effective]: 1.521e-05
        [opt_reshape]: 6.86001e-06
        [fold_const_symbol]: 1.066e-05
        [renormalize]: 4.19997e-07
[validate]: 3.316e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00668728
[execute]: 7.46001e-06
Sums
    bootstrap                            :     0.000233s :  0.81%
    type_inference                       :     0.012316s : 42.89%
    event_method                         :     0.000055s :  0.19%
    auto_monad                           :     0.000103s :  0.36%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000027s :  0.09%
    rewriter_before_opt_a                :     0.000101s :  0.35%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000173s :  0.60%
    jit_opt_a.loop_unroll                :     0.000100s :  0.35%
    jit_opt_a.a_1                        :     0.001945s :  6.77%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000031s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.08%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000053s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001583s :  5.51%
    jit_opt_a.replace_old_param          :     0.000069s :  0.24%
    jit_opt_a.inline_without_move        :     0.000063s :  0.22%
    jit_opt_a.renormalize                :     0.002662s :  9.27%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000085s :  0.30%
    jit_opt_a.cse                        :     0.000177s :  0.62%
    jit_opt_a.replace_applicator         :     0.000081s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.04%
    rewriter_after_opt_a                 :     0.000037s :  0.13%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000507s :  1.77%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000021s :  0.07%
    loop_unroll                          :     0.000447s :  1.56%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000051s :  0.18%
    cse_after_recomputation.cse          :     0.000013s :  0.05%
    auto_monad_reorder                   :     0.000018s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000520s :  1.81%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006687s : 23.29%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000451   113
 0.52% :     0.000002s :      3: substitution.elim_not_effective
 0.35% :     0.000002s :      3: substitution.fold_const_symbol
 1.16% :     0.000005s :      4: substitution.graph_param_transform
62.41% :     0.000281s :     14: substitution.inline
 3.49% :     0.000016s :      2: substitution.inline_without_move
 1.95% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.06% :     0.000009s :      7: substitution.minmaximum_grad
 1.32% :     0.000006s :      5: substitution.partial_eliminate
 2.29% :     0.000010s :     14: substitution.remove_not_recompute_node
 5.40% :     0.000024s :     10: substitution.replace_applicator
 1.55% :     0.000007s :      7: substitution.replace_old_param
 0.49% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.63% :     0.000007s :      1: substitution.switch_simplify
 4.07% :     0.000018s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.02% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 8.31% :     0.000037s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012244     2
90.47% :     0.011077s :      1: type_inference.infer
 9.53% :     0.001166s :      1: type_inference.specialize
------[replace.]   0.000173    22
53.34% :     0.000092s :     14: replace.inline
18.85% :     0.000033s :      1: replace.switch_simplify
27.81% :     0.000048s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000301    22
91.15% :     0.000274s :     14: match.inline
 2.24% :     0.000007s :      1: match.switch_simplify
 6.61% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000391  2634
 1.53% :     0.000006s :     44: predicate.accumulaten_eliminater
 6.95% :     0.000027s :      4: predicate.ad_related_special_op_eliminate
 1.35% :     0.000005s :     44: predicate.addn_check_dump
 1.52% :     0.000006s :     44: predicate.addn_zero_filter
 1.88% :     0.000007s :     44: predicate.arithmetic_simplify
 1.44% :     0.000006s :     44: predicate.cast_eliminate
 0.16% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.35% :     0.000005s :     44: predicate.compare_switch_simplify
 1.45% :     0.000006s :     44: predicate.depend_value_elim
 1.35% :     0.000005s :     44: predicate.dict_get_item_const_eliminator
 1.38% :     0.000005s :     44: predicate.dict_get_item_eliminator
 1.39% :     0.000005s :     44: predicate.dict_set_item_eliminator
 0.32% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      4: predicate.elim_not_effective
 0.18% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000005s :     44: predicate.environ_add_const_eliminate
 1.34% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.39% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.57% :     0.000006s :     44: predicate.environ_get_eliminate
 1.34% :     0.000005s :     44: predicate.environ_get_set_eliminate
 0.08% :     0.000000s :      4: predicate.fold_const_symbol
 0.75% :     0.000003s :     19: predicate.get_grad_eliminate
 0.09% :     0.000000s :      4: predicate.graph_param_transform
 4.13% :     0.000016s :     73: predicate.inline
 1.77% :     0.000007s :     40: predicate.inline_without_move
 0.32% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.79% :     0.000003s :     19: predicate.less_batch_normalization
 1.67% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.80% :     0.000007s :     55: predicate.load_eliminater
 0.47% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.18% :     0.000012s :     91: predicate.loop_unroll_before_grad
 1.63% :     0.000006s :     48: predicate.make_slice_get_slice_eliminator
 1.32% :     0.000005s :     44: predicate.merge_addn
 1.42% :     0.000006s :     44: predicate.minmaximum_grad
 0.54% :     0.000002s :      4: predicate.mutable_eliminate
 0.20% :     0.000001s :      4: predicate.opt_reshape
 2.17% :     0.000009s :     55: predicate.partial_eliminate
 1.46% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.84% :     0.000007s :     44: predicate.reduce_eliminate
 1.77% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.37% :     0.000009s :     97: predicate.replace_applicator
 0.83% :     0.000003s :     40: predicate.replace_old_param
 0.10% :     0.000000s :      4: predicate.reset_defer_inline
 1.60% :     0.000006s :     44: predicate.reshape_eliminate
 1.44% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.23% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.43% :     0.000006s :     44: predicate.same_eliminate
 0.43% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.38% :     0.000001s :      8: predicate.special_op_eliminate
 0.82% :     0.000003s :     19: predicate.specialize_transform
 1.51% :     0.000006s :     44: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.62% :     0.000010s :     65: predicate.switch_defer_inline
 2.46% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.11% :     0.000024s :    162: predicate.switch_simplify
 1.41% :     0.000006s :     44: predicate.tile_eliminate
 1.37% :     0.000005s :     44: predicate.transpose_eliminate
 1.69% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.62% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 2.98% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000006s :     44: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.78% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.69% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000006s :     44: predicate.value_based_eliminate
 0.15% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.21% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001495    32
62.00% :     0.000927s :     14: func_graph_cloner_run.FuncGraphClonerGraph
38.00% :     0.000568s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038316    91
 0.14% :     0.000054s :      1: add_recomputation
 0.28% :     0.000108s :      1: auto_monad
 0.06% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.66% :     0.000255s :      1: bootstrap
 0.06% :     0.000024s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000026s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.16% :     0.000060s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
29.13% :     0.011162s :      1: jit_opt_a
 0.40% :     0.000152s :      1: jit_opt_after_cconv
 0.14% :     0.000055s :      1: jit_opt_b
 1.19% :     0.000455s :      1: loop_unroll
 1.34% :     0.000515s :      1: mutable_eliminate
 6.81% :     0.002611s :     39: opt.transform.jit_opt_a
 0.14% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000029s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000050s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000038s :      4: opt.transform.symbol_engine_opt
 1.38% :     0.000528s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000029s :      1: py_interpret_to_execute
 0.03% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 3.74% :     0.001432s :      2: renormalize.infer
 3.18% :     0.001217s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000040s :      1: rewriter_after_opt_a
 0.27% :     0.000104s :      1: rewriter_before_opt_a
 0.21% :     0.000081s :      1: symbol_engine_optimizer
17.48% :     0.006697s :      1: task_emit
32.18% :     0.012331s :      1: type_inference
 0.14% :     0.000053s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.314.600 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6c07b4c0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.315.137 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0149669, [30]
[bootstrap]: 0.00033034
[type_inference]: 0.00538973
[event_method]: 1.916e-05
[auto_monad]: 7.327e-05
[graph_reusing]: 6.59999e-06
[pre_auto_parallel]: 2.78e-06
[py_interpret_to_execute]: 4.401e-05
[rewriter_before_opt_a]: 7.889e-05
[expand_dump_flag]: 3.33e-06
[jit_opt_a]: 0.00642254, [2]
    [Cycle 1]: 0.00193686, [27]
        [switch_simplify]: 6.772e-05
        [loop_unroll]: 3.177e-05
        [a_1]: 0.00059603
        [with_stream_mark]: 2.114e-05
        [recompute_prepare]: 7.89002e-06
        [updatestate_depend_eliminate]: 3.43e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 3.23e-06
        [parameter_eliminate]: 1.85001e-06
        [specialize_transform]: 6.25002e-06
        [updatestate_useless_node_eliminater]: 5.82001e-06
        [accelerated_algorithm]: 6.05002e-06
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 5.99999e-06
        [merge_forward]: 3.53e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.606e-05
        [j_node_and_user_rematch]: 1.024e-05
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 1.112e-05
        [inline_without_move]: 6.01e-06
        [renormalize]: 0.00085076
        [add_forward_monad_depend]: 6.93e-06
        [auto_monad_grad]: 3.3e-06
        [auto_monad_eliminator]: 1.584e-05
        [cse]: 3.229e-05
        [replace_applicator]: 1.418e-05
    [Cycle 2]: 0.00036047, [27]
        [switch_simplify]: 6.07999e-06
        [loop_unroll]: 5.61003e-06
        [a_1]: 9.341e-05
        [with_stream_mark]: 1.165e-05
        [recompute_prepare]: 5.50001e-06
        [updatestate_depend_eliminate]: 2.69999e-06
        [updatestate_assign_eliminate]: 1.97999e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.72001e-06
        [specialize_transform]: 5.24e-06
        [updatestate_useless_node_eliminater]: 5.61e-06
        [accelerated_algorithm]: 6.04001e-06
        [meta_shard_fg_expand]: 1.36002e-06
        [get_grad_eliminate_]: 5.47999e-06
        [merge_forward]: 3.03998e-06
        [cell_reuse_recompute_pass]: 1.62999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.744e-05
        [j_node_and_user_rematch]: 9.04998e-06
        [meta_fg_expand]: 1.77999e-06
        [replace_old_param]: 9.45001e-06
        [inline_without_move]: 5.87999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.66e-06
        [auto_monad_grad]: 1.50999e-06
        [auto_monad_eliminator]: 5.76998e-06
        [cse]: 1.301e-05
        [replace_applicator]: 6.70002e-06
[py_interpret_to_execute_after_opt_a]: 1.261e-05
[rewriter_after_opt_a]: 3.681e-05
[convert_after_rewriter]: 7.26999e-06
[order_py_execute_after_rewriter]: 5.75001e-06
[mutable_eliminate]: 0.000764
[jit_opt_b]: 5.539e-05, [1]
    [Cycle 1]: 4.689e-05, [2]
        [frontend_op_eliminate]: 1.769e-05
        [inline_after_opt_a]: 1.67e-05
[cconv]: 2.872e-05
[loop_unroll]: 0.00052673
[jit_opt_after_cconv]: 0.00017019, [1]
    [Cycle 1]: 0.00016269, [11]
        [c_1]: 2.295e-05
        [parameter_eliminate]: 3.67002e-06
        [updatestate_depend_eliminate]: 9.71998e-06
        [updatestate_assign_eliminate]: 2.71999e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [cse]: 2.878e-05
        [call_graph_tuple_transform]: 2.361e-05
        [tuple_list_get_item_eliminator]: 5.62001e-06
        [none_parameter_eliminate]: 1.66002e-06
        [renormalize]: 1.04e-06
        [switch_simplify]: 5.51e-06
[remove_dup_value]: 1.492e-05
[partial_unused_args_eliminate]: 2.22999e-06
[environ_conv]: 6.14999e-06
[add_recomputation]: 5.202e-05
[cse_after_recomputation]: 2.337e-05, [1]
    [Cycle 1]: 1.705e-05, [1]
        [cse]: 1.051e-05
[auto_monad_reorder]: 1.543e-05
[get_jit_bprop_graph]: 2.12001e-06
[rewriter_after_jit_bprop_graph]: 9.18002e-06
[opt_after_jit_grad]: 0.0005408
[symbol_engine_optimizer]: 8.432e-05, [1]
    [Cycle 1]: 7.539e-05, [6]
        [build]: 5.52999e-06
        [elim_shapecalc]: 9.12001e-06
        [elim_not_effective]: 1.489e-05
        [opt_reshape]: 6.26e-06
        [fold_const_symbol]: 1.016e-05
        [renormalize]: 4.39992e-07
[validate]: 3.9e-05
Sums
    bootstrap                            :     0.000330s :  3.25%
    type_inference                       :     0.005390s : 52.95%
    event_method                         :     0.000019s :  0.19%
    auto_monad                           :     0.000073s :  0.72%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000044s :  0.43%
    rewriter_before_opt_a                :     0.000079s :  0.78%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000074s :  0.73%
    jit_opt_a.loop_unroll                :     0.000037s :  0.37%
    jit_opt_a.a_1                        :     0.000689s :  6.77%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000021s :  0.20%
    jit_opt_a.inline_without_move        :     0.000012s :  0.12%
    jit_opt_a.renormalize                :     0.000851s :  8.36%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.21%
    jit_opt_a.cse                        :     0.000045s :  0.45%
    jit_opt_a.replace_applicator         :     0.000021s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.12%
    rewriter_after_opt_a                 :     0.000037s :  0.36%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000764s :  7.51%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.16%
    cconv                                :     0.000029s :  0.28%
    loop_unroll                          :     0.000527s :  5.18%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.10%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000029s :  0.28%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000015s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000052s :  0.51%
    cse_after_recomputation.cse          :     0.000011s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.09%
    opt_after_jit_grad                   :     0.000541s :  5.31%
    symbol_engine_optimizer.build        :     0.000006s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000039s :  0.38%


Time group info:
------[substitution.]   0.000240    23
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000002s :      2: substitution.fold_const_symbol
 2.60% :     0.000006s :      3: substitution.graph_param_transform
83.93% :     0.000202s :      4: substitution.inline
 1.51% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.54% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.64% :     0.000006s :      2: substitution.replace_old_param
 5.29% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005314     2
82.18% :     0.004367s :      1: type_inference.infer
17.82% :     0.000947s :      1: type_inference.specialize
------[replace.]   0.000076     6
66.98% :     0.000051s :      4: replace.inline
33.02% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000210     6
94.53% :     0.000199s :      4: match.inline
 5.47% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.12% :     0.000001s :     10: predicate.addn_zero_filter
 1.99% :     0.000003s :     10: predicate.arithmetic_simplify
 1.07% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.41% :     0.000001s :      3: predicate.graph_param_transform
 5.75% :     0.000007s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000002s :      6: predicate.less_batch_normalization
 2.09% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.76% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.25% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 2.11% :     0.000003s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     15: predicate.partial_eliminate
 1.06% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.34% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.23% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.37% :     0.000011s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.16% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.61% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000662    12
51.27% :     0.000339s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.80% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.94% :     0.000284s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016818    72
 0.33% :     0.000056s :      1: add_recomputation
 0.46% :     0.000077s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 2.07% :     0.000348s :      1: bootstrap
 0.19% :     0.000031s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
38.21% :     0.006426s :      1: jit_opt_a
 1.03% :     0.000174s :      1: jit_opt_after_cconv
 0.35% :     0.000058s :      1: jit_opt_b
 3.20% :     0.000538s :      1: loop_unroll
 4.61% :     0.000775s :      1: mutable_eliminate
 5.55% :     0.000933s :     26: opt.transform.jit_opt_a
 0.32% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000027s :      4: opt.transform.jit_opt_b
 0.11% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.27% :     0.000550s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000047s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.52% :     0.000425s :      1: renormalize.infer
 2.48% :     0.000418s :      1: renormalize.specialize
 0.07% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000040s :      1: rewriter_after_opt_a
 0.49% :     0.000082s :      1: rewriter_before_opt_a
 0.52% :     0.000087s :      1: symbol_engine_optimizer
32.16% :     0.005408s :      1: type_inference


TotalTime = 0.0327031, [33]
[bootstrap]: 0.00020551
[type_inference]: 0.0120551
[event_method]: 5.595e-05
[auto_monad]: 0.00010557
[graph_reusing]: 7.75e-06
[pre_auto_parallel]: 2.83e-06
[py_interpret_to_execute]: 2.687e-05
[rewriter_before_opt_a]: 0.00010254
[expand_dump_flag]: 3.3e-06
[jit_opt_a]: 0.0110274, [3]
    [Cycle 1]: 0.0055344, [27]
        [switch_simplify]: 0.00014693
        [loop_unroll]: 4.212e-05
        [a_1]: 0.00091775
        [with_stream_mark]: 2.381e-05
        [recompute_prepare]: 1.782e-05
        [updatestate_depend_eliminate]: 8.3e-06
        [updatestate_assign_eliminate]: 6.19001e-06
        [updatestate_loads_eliminate]: 6.44999e-06
        [parameter_eliminate]: 2.96001e-06
        [specialize_transform]: 1.362e-05
        [updatestate_useless_node_eliminater]: 1.193e-05
        [accelerated_algorithm]: 1.192e-05
        [meta_shard_fg_expand]: 3.6e-06
        [get_grad_eliminate_]: 1.161e-05
        [merge_forward]: 8.54002e-06
        [cell_reuse_recompute_pass]: 8.80013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.654e-05
        [j_node_and_user_rematch]: 2.217e-05
        [meta_fg_expand]: 0.00140823
        [replace_old_param]: 5.09e-05
        [inline_without_move]: 5.163e-05
        [renormalize]: 0.00227345
        [add_forward_monad_depend]: 9.20999e-06
        [auto_monad_grad]: 4.92e-06
        [auto_monad_eliminator]: 4.699e-05
        [cse]: 0.00014648
        [replace_applicator]: 6.137e-05
    [Cycle 2]: 0.00190789, [27]
        [switch_simplify]: 3.595e-05
        [loop_unroll]: 3.371e-05
        [a_1]: 0.00100414
        [with_stream_mark]: 1.12e-05
        [recompute_prepare]: 8.03001e-06
        [updatestate_depend_eliminate]: 3.53e-06
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 2.40002e-06
        [parameter_eliminate]: 1.06002e-06
        [specialize_transform]: 6.74999e-06
        [updatestate_useless_node_eliminater]: 6.09999e-06
        [accelerated_algorithm]: 5.82999e-06
        [meta_shard_fg_expand]: 1.66e-06
        [get_grad_eliminate_]: 2.761e-05
        [merge_forward]: 3.31001e-06
        [cell_reuse_recompute_pass]: 8.30012e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.308e-05
        [j_node_and_user_rematch]: 9.62001e-06
        [meta_fg_expand]: 6.405e-05
        [replace_old_param]: 1.005e-05
        [inline_without_move]: 6.05002e-06
        [renormalize]: 0.00045549
        [add_forward_monad_depend]: 3.81999e-06
        [auto_monad_grad]: 1.24e-06
        [auto_monad_eliminator]: 1.09e-05
        [cse]: 1.915e-05
        [replace_applicator]: 1.144e-05
    [Cycle 3]: 0.00035817, [27]
        [switch_simplify]: 6.48e-06
        [loop_unroll]: 6.03998e-06
        [a_1]: 0.00011576
        [with_stream_mark]: 8.73001e-06
        [recompute_prepare]: 5.70001e-06
        [updatestate_depend_eliminate]: 3.4e-06
        [updatestate_assign_eliminate]: 2.54001e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 6.26e-06
        [updatestate_useless_node_eliminater]: 5.71998e-06
        [accelerated_algorithm]: 5.67999e-06
        [meta_shard_fg_expand]: 1.29998e-06
        [get_grad_eliminate_]: 5.54e-06
        [merge_forward]: 2.83e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.432e-05
        [j_node_and_user_rematch]: 9.87001e-06
        [meta_fg_expand]: 1.89999e-06
        [replace_old_param]: 8.18999e-06
        [inline_without_move]: 5.59e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 6.33e-06
        [cse]: 1.338e-05
        [replace_applicator]: 6.19001e-06
[py_interpret_to_execute_after_opt_a]: 1.006e-05
[rewriter_after_opt_a]: 3.572e-05
[convert_after_rewriter]: 7.43999e-06
[order_py_execute_after_rewriter]: 5.48002e-06
[mutable_eliminate]: 0.00050832
[jit_opt_b]: 7.636e-05, [1]
    [Cycle 1]: 6.963e-05, [2]
        [frontend_op_eliminate]: 1.788e-05
        [inline_after_opt_a]: 3.982e-05
[cconv]: 2.05e-05
[loop_unroll]: 0.00049228
[jit_opt_after_cconv]: 0.00015013, [1]
    [Cycle 1]: 0.0001436, [11]
        [c_1]: 2.607e-05
        [parameter_eliminate]: 2.52001e-06
        [updatestate_depend_eliminate]: 6.38e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [cse]: 2.12e-05
        [call_graph_tuple_transform]: 1.988e-05
        [tuple_list_get_item_eliminator]: 6.33998e-06
        [none_parameter_eliminate]: 1.59998e-06
        [renormalize]: 4.60015e-07
        [switch_simplify]: 6.25002e-06
[remove_dup_value]: 1.558e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 5.60001e-06
[add_recomputation]: 5.669e-05
[cse_after_recomputation]: 2.58e-05, [1]
    [Cycle 1]: 2.023e-05, [1]
        [cse]: 1.414e-05
[auto_monad_reorder]: 1.828e-05
[get_jit_bprop_graph]: 1.29e-06
[rewriter_after_jit_bprop_graph]: 2.348e-05
[opt_after_jit_grad]: 0.00052179
[symbol_engine_optimizer]: 8.208e-05, [1]
    [Cycle 1]: 7.591e-05, [6]
        [build]: 4.00998e-06
        [elim_shapecalc]: 9.77001e-06
        [elim_not_effective]: 1.548e-05
        [opt_reshape]: 6.78998e-06
        [fold_const_symbol]: 1.141e-05
        [renormalize]: 5.50004e-07
[validate]: 3.491e-05
[backend_pass]: 1.09e-06
[task_emit]: 0.00681997
[execute]: 6.79999e-06
Sums
    bootstrap                            :     0.000206s :  0.72%
    type_inference                       :     0.012055s : 41.98%
    event_method                         :     0.000056s :  0.19%
    auto_monad                           :     0.000106s :  0.37%
    graph_reusing                        :     0.000008s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000027s :  0.09%
    rewriter_before_opt_a                :     0.000103s :  0.36%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000189s :  0.66%
    jit_opt_a.loop_unroll                :     0.000082s :  0.29%
    jit_opt_a.a_1                        :     0.002038s :  7.10%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000027s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000023s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000045s :  0.16%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000054s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001474s :  5.13%
    jit_opt_a.replace_old_param          :     0.000069s :  0.24%
    jit_opt_a.inline_without_move        :     0.000063s :  0.22%
    jit_opt_a.renormalize                :     0.002729s :  9.50%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000064s :  0.22%
    jit_opt_a.cse                        :     0.000179s :  0.62%
    jit_opt_a.replace_applicator         :     0.000079s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.04%
    rewriter_after_opt_a                 :     0.000036s :  0.12%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000508s :  1.77%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000040s :  0.14%
    cconv                                :     0.000021s :  0.07%
    loop_unroll                          :     0.000492s :  1.71%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000057s :  0.20%
    cse_after_recomputation.cse          :     0.000014s :  0.05%
    auto_monad_reorder                   :     0.000018s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000023s :  0.08%
    opt_after_jit_grad                   :     0.000522s :  1.82%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006820s : 23.75%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000526   113
 0.45% :     0.000002s :      3: substitution.elim_not_effective
 0.30% :     0.000002s :      3: substitution.fold_const_symbol
 1.04% :     0.000005s :      4: substitution.graph_param_transform
63.97% :     0.000337s :     14: substitution.inline
 3.13% :     0.000016s :      2: substitution.inline_without_move
 1.64% :     0.000009s :     14: substitution.j_node_and_user_rematch
 1.72% :     0.000009s :      7: substitution.minmaximum_grad
 1.23% :     0.000006s :      5: substitution.partial_eliminate
 2.11% :     0.000011s :     14: substitution.remove_not_recompute_node
 4.60% :     0.000024s :     10: substitution.replace_applicator
 1.28% :     0.000007s :      7: substitution.replace_old_param
 0.45% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.38% :     0.000007s :      1: substitution.switch_simplify
 3.58% :     0.000019s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.49% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
10.62% :     0.000056s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011981     2
90.33% :     0.010822s :      1: type_inference.infer
 9.67% :     0.001158s :      1: type_inference.specialize
------[replace.]   0.000179    22
52.22% :     0.000093s :     14: replace.inline
19.28% :     0.000035s :      1: replace.switch_simplify
28.50% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000375    22
87.90% :     0.000329s :     14: match.inline
 1.80% :     0.000007s :      1: match.switch_simplify
10.30% :     0.000039s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000391  2634
 1.51% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.48% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.40% :     0.000005s :     44: predicate.addn_check_dump
 1.63% :     0.000006s :     44: predicate.addn_zero_filter
 1.83% :     0.000007s :     44: predicate.arithmetic_simplify
 1.52% :     0.000006s :     44: predicate.cast_eliminate
 0.15% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.34% :     0.000005s :     44: predicate.compare_switch_simplify
 1.53% :     0.000006s :     44: predicate.depend_value_elim
 1.34% :     0.000005s :     44: predicate.dict_get_item_const_eliminator
 1.38% :     0.000005s :     44: predicate.dict_get_item_eliminator
 1.36% :     0.000005s :     44: predicate.dict_set_item_eliminator
 0.33% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      4: predicate.elim_not_effective
 0.23% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.38% :     0.000005s :     44: predicate.environ_add_const_eliminate
 1.33% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.36% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.48% :     0.000006s :     44: predicate.environ_get_eliminate
 1.35% :     0.000005s :     44: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 6.20% :     0.000024s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.20% :     0.000016s :     73: predicate.inline
 1.66% :     0.000006s :     40: predicate.inline_without_move
 0.33% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.80% :     0.000003s :     19: predicate.less_batch_normalization
 1.71% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.85% :     0.000007s :     55: predicate.load_eliminater
 0.49% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.21% :     0.000013s :     91: predicate.loop_unroll_before_grad
 1.68% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.44% :     0.000006s :     44: predicate.merge_addn
 1.39% :     0.000005s :     44: predicate.minmaximum_grad
 0.56% :     0.000002s :      4: predicate.mutable_eliminate
 0.18% :     0.000001s :      4: predicate.opt_reshape
 2.21% :     0.000009s :     55: predicate.partial_eliminate
 1.46% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.72% :     0.000007s :     44: predicate.reduce_eliminate
 1.68% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.36% :     0.000009s :     97: predicate.replace_applicator
 0.88% :     0.000003s :     40: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.57% :     0.000006s :     44: predicate.reshape_eliminate
 1.44% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.25% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.72% :     0.000007s :     44: predicate.same_eliminate
 0.44% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.37% :     0.000001s :      8: predicate.special_op_eliminate
 0.84% :     0.000003s :     19: predicate.specialize_transform
 1.70% :     0.000007s :     44: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000005s :     44: predicate.stack_unstack_eliminate
 0.15% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.65% :     0.000010s :     65: predicate.switch_defer_inline
 2.44% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.21% :     0.000024s :    162: predicate.switch_simplify
 1.41% :     0.000006s :     44: predicate.tile_eliminate
 1.36% :     0.000005s :     44: predicate.transpose_eliminate
 1.76% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.56% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 3.06% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.81% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000010s :     74: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000007s :     44: predicate.value_based_eliminate
 0.14% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.18% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001469    32
60.82% :     0.000894s :     14: func_graph_cloner_run.FuncGraphClonerGraph
39.18% :     0.000576s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038286    91
 0.16% :     0.000060s :      1: add_recomputation
 0.29% :     0.000110s :      1: auto_monad
 0.05% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.57% :     0.000220s :      1: bootstrap
 0.06% :     0.000023s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.07% :     0.000028s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.16% :     0.000061s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
28.81% :     0.011030s :      1: jit_opt_a
 0.40% :     0.000153s :      1: jit_opt_after_cconv
 0.21% :     0.000079s :      1: jit_opt_b
 1.31% :     0.000500s :      1: loop_unroll
 1.35% :     0.000517s :      1: mutable_eliminate
 7.12% :     0.002724s :     39: opt.transform.jit_opt_a
 0.14% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000029s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000040s :      4: opt.transform.symbol_engine_opt
 1.39% :     0.000530s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000029s :      1: py_interpret_to_execute
 0.03% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 3.86% :     0.001478s :      2: renormalize.infer
 3.23% :     0.001238s :      2: renormalize.specialize
 0.07% :     0.000026s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000039s :      1: rewriter_after_opt_a
 0.28% :     0.000105s :      1: rewriter_before_opt_a
 0.22% :     0.000085s :      1: symbol_engine_optimizer
17.84% :     0.006829s :      1: task_emit
31.53% :     0.012070s :      1: type_inference
 0.14% :     0.000054s :      1: validate


TotalTime = 0.015097, [30]
[bootstrap]: 0.00035512
[type_inference]: 0.00530006
[event_method]: 2.046e-05
[auto_monad]: 7.084e-05
[graph_reusing]: 6.39001e-06
[pre_auto_parallel]: 3.25e-06
[py_interpret_to_execute]: 4.043e-05
[rewriter_before_opt_a]: 7.679e-05
[expand_dump_flag]: 3.14999e-06
[jit_opt_a]: 0.00642664, [2]
    [Cycle 1]: 0.00194316, [27]
        [switch_simplify]: 6.507e-05
        [loop_unroll]: 3.216e-05
        [a_1]: 0.00059358
        [with_stream_mark]: 2.202e-05
        [recompute_prepare]: 8.55999e-06
        [updatestate_depend_eliminate]: 3.93999e-06
        [updatestate_assign_eliminate]: 3.35003e-06
        [updatestate_loads_eliminate]: 2.69001e-06
        [parameter_eliminate]: 2.46e-06
        [specialize_transform]: 6.01998e-06
        [updatestate_useless_node_eliminater]: 5.11002e-06
        [accelerated_algorithm]: 5.82999e-06
        [meta_shard_fg_expand]: 2.12001e-06
        [get_grad_eliminate_]: 5.65001e-06
        [merge_forward]: 3.75e-06
        [cell_reuse_recompute_pass]: 1.78997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.584e-05
        [j_node_and_user_rematch]: 1.044e-05
        [meta_fg_expand]: 2.90998e-06
        [replace_old_param]: 9.23002e-06
        [inline_without_move]: 5.55001e-06
        [renormalize]: 0.0008396
        [add_forward_monad_depend]: 8.95999e-06
        [auto_monad_grad]: 2.30002e-06
        [auto_monad_eliminator]: 1.959e-05
        [cse]: 3.365e-05
        [replace_applicator]: 1.869e-05
    [Cycle 2]: 0.00038361, [27]
        [switch_simplify]: 6.56e-06
        [loop_unroll]: 5.27999e-06
        [a_1]: 0.00010056
        [with_stream_mark]: 1.581e-05
        [recompute_prepare]: 6.16e-06
        [updatestate_depend_eliminate]: 3.38e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [parameter_eliminate]: 1.36002e-06
        [specialize_transform]: 5.82999e-06
        [updatestate_useless_node_eliminater]: 5.47001e-06
        [accelerated_algorithm]: 6.46999e-06
        [meta_shard_fg_expand]: 1.67001e-06
        [get_grad_eliminate_]: 4.95001e-06
        [merge_forward]: 4.25999e-06
        [cell_reuse_recompute_pass]: 3.12002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.862e-05
        [j_node_and_user_rematch]: 8.89998e-06
        [meta_fg_expand]: 1.94e-06
        [replace_old_param]: 9.09e-06
        [inline_without_move]: 4.99998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 2.34001e-06
        [auto_monad_grad]: 1.59e-06
        [auto_monad_eliminator]: 7.55998e-06
        [cse]: 1.523e-05
        [replace_applicator]: 6.18002e-06
[py_interpret_to_execute_after_opt_a]: 1.653e-05
[rewriter_after_opt_a]: 4.266e-05
[convert_after_rewriter]: 7.36999e-06
[order_py_execute_after_rewriter]: 5.37999e-06
[mutable_eliminate]: 0.00085212
[jit_opt_b]: 5.764e-05, [1]
    [Cycle 1]: 4.846e-05, [2]
        [frontend_op_eliminate]: 1.809e-05
        [inline_after_opt_a]: 1.751e-05
[cconv]: 2.846e-05
[loop_unroll]: 0.00052738
[jit_opt_after_cconv]: 0.00017063, [1]
    [Cycle 1]: 0.0001623, [11]
        [c_1]: 2.246e-05
        [parameter_eliminate]: 4.24002e-06
        [updatestate_depend_eliminate]: 7.4e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [cse]: 3.094e-05
        [call_graph_tuple_transform]: 2.432e-05
        [tuple_list_get_item_eliminator]: 5.69e-06
        [none_parameter_eliminate]: 1.65001e-06
        [renormalize]: 8.59989e-07
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 1.568e-05
[partial_unused_args_eliminate]: 2.11e-06
[environ_conv]: 5.84999e-06
[add_recomputation]: 5.453e-05
[cse_after_recomputation]: 2.341e-05, [1]
    [Cycle 1]: 1.747e-05, [1]
        [cse]: 1.025e-05
[auto_monad_reorder]: 1.698e-05
[get_jit_bprop_graph]: 2.49001e-06
[rewriter_after_jit_bprop_graph]: 8.90001e-06
[opt_after_jit_grad]: 0.00061566
[symbol_engine_optimizer]: 8.971e-05, [1]
    [Cycle 1]: 8.111e-05, [6]
        [build]: 5.60001e-06
        [elim_shapecalc]: 8.25e-06
        [elim_not_effective]: 1.603e-05
        [opt_reshape]: 7.1e-06
        [fold_const_symbol]: 1.079e-05
        [renormalize]: 9.70002e-07
[validate]: 4.353e-05
Sums
    bootstrap                            :     0.000355s :  3.45%
    type_inference                       :     0.005300s : 51.42%
    event_method                         :     0.000020s :  0.20%
    auto_monad                           :     0.000071s :  0.69%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000040s :  0.39%
    rewriter_before_opt_a                :     0.000077s :  0.74%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000072s :  0.69%
    jit_opt_a.loop_unroll                :     0.000037s :  0.36%
    jit_opt_a.a_1                        :     0.000694s :  6.73%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.37%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.10%
    jit_opt_a.merge_forward              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000018s :  0.18%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000840s :  8.15%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.11%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.26%
    jit_opt_a.cse                        :     0.000049s :  0.47%
    jit_opt_a.replace_applicator         :     0.000025s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.16%
    rewriter_after_opt_a                 :     0.000043s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000852s :  8.27%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.17%
    cconv                                :     0.000028s :  0.28%
    loop_unroll                          :     0.000527s :  5.12%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000031s :  0.30%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000016s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000055s :  0.53%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000017s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.09%
    opt_after_jit_grad                   :     0.000616s :  5.97%
    symbol_engine_optimizer.build        :     0.000006s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.42%


Time group info:
------[substitution.]   0.000237    23
 0.90% :     0.000002s :      2: substitution.elim_not_effective
 0.85% :     0.000002s :      2: substitution.fold_const_symbol
 2.72% :     0.000006s :      3: substitution.graph_param_transform
83.04% :     0.000197s :      4: substitution.inline
 1.80% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.35% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.19% :     0.000005s :      2: substitution.replace_old_param
 6.16% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005228     2
81.82% :     0.004278s :      1: type_inference.infer
18.18% :     0.000951s :      1: type_inference.specialize
------[replace.]   0.000080     6
67.01% :     0.000053s :      4: replace.inline
32.99% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000207     6
93.74% :     0.000194s :      4: match.inline
 6.26% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.16% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.51% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.53% :     0.000002s :     10: predicate.addn_zero_filter
 2.33% :     0.000003s :     10: predicate.arithmetic_simplify
 1.60% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.56% :     0.000002s :     10: predicate.depend_value_elim
 0.98% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.07% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.14% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.45% :     0.000001s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.03% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.22% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     10: predicate.environ_get_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 6.21% :     0.000008s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.37% :     0.000002s :      6: predicate.less_batch_normalization
 1.37% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.60% :     0.000002s :     15: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.83% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 1.77% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     15: predicate.partial_eliminate
 0.98% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.79% :     0.000002s :     10: predicate.reduce_eliminate
 1.63% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.77% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.38% :     0.000000s :      3: predicate.reset_defer_inline
 1.19% :     0.000002s :     10: predicate.reshape_eliminate
 1.39% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.09% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.93% :     0.000001s :      6: predicate.special_op_eliminate
 1.14% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.53% :     0.000010s :     53: predicate.switch_simplify
 1.58% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.58% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.38% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000648    12
50.96% :     0.000330s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.84% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.20% :     0.000286s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016938    72
 0.34% :     0.000058s :      1: add_recomputation
 0.44% :     0.000075s :      1: auto_monad
 0.12% :     0.000020s :      1: auto_monad_reorder
 2.20% :     0.000373s :      1: bootstrap
 0.19% :     0.000032s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
37.96% :     0.006431s :      1: jit_opt_a
 1.03% :     0.000174s :      1: jit_opt_after_cconv
 0.36% :     0.000060s :      1: jit_opt_b
 3.18% :     0.000538s :      1: loop_unroll
 5.09% :     0.000862s :      1: mutable_eliminate
 5.53% :     0.000937s :     26: opt.transform.jit_opt_a
 0.32% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000028s :      4: opt.transform.jit_opt_b
 0.09% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000039s :      4: opt.transform.symbol_engine_opt
 3.72% :     0.000630s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000044s :      1: py_interpret_to_execute
 0.11% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.23% :     0.000378s :      1: renormalize.infer
 2.66% :     0.000451s :      1: renormalize.specialize
 0.07% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000047s :      1: rewriter_after_opt_a
 0.47% :     0.000080s :      1: rewriter_before_opt_a
 0.55% :     0.000093s :      1: symbol_engine_optimizer
31.40% :     0.005319s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.354.954 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dcafee0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.355.551 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0244081, [33]
[bootstrap]: 0.00036157
[type_inference]: 0.00557531
[event_method]: 1.974e-05
[auto_monad]: 7.28e-05
[graph_reusing]: 6.61999e-06
[pre_auto_parallel]: 2.79999e-06
[py_interpret_to_execute]: 4.927e-05
[rewriter_before_opt_a]: 7.695e-05
[expand_dump_flag]: 3.18e-06
[jit_opt_a]: 0.00714547, [2]
    [Cycle 1]: 0.00236055, [27]
        [switch_simplify]: 7.561e-05
        [loop_unroll]: 3.309e-05
        [a_1]: 0.00068003
        [with_stream_mark]: 2.516e-05
        [recompute_prepare]: 1.373e-05
        [updatestate_depend_eliminate]: 4.29002e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 3.65e-06
        [specialize_transform]: 7.3e-06
        [updatestate_useless_node_eliminater]: 6.34999e-06
        [accelerated_algorithm]: 6.07999e-06
        [meta_shard_fg_expand]: 2.86999e-06
        [get_grad_eliminate_]: 5.87001e-06
        [merge_forward]: 3.79002e-06
        [cell_reuse_recompute_pass]: 1.91998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.847e-05
        [j_node_and_user_rematch]: 1.139e-05
        [meta_fg_expand]: 3.26999e-06
        [replace_old_param]: 1.059e-05
        [inline_without_move]: 6.39999e-06
        [renormalize]: 0.00111857
        [add_forward_monad_depend]: 1.06e-05
        [auto_monad_grad]: 3.11001e-06
        [auto_monad_eliminator]: 1.671e-05
        [cse]: 3.149e-05
        [replace_applicator]: 2.414e-05
    [Cycle 2]: 0.0004154, [27]
        [switch_simplify]: 6.57002e-06
        [loop_unroll]: 5.34e-06
        [a_1]: 0.00010943
        [with_stream_mark]: 1.796e-05
        [recompute_prepare]: 7.06999e-06
        [updatestate_depend_eliminate]: 3.42002e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.83e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 6.69999e-06
        [updatestate_useless_node_eliminater]: 5.47999e-06
        [accelerated_algorithm]: 6.23e-06
        [meta_shard_fg_expand]: 2.13002e-06
        [get_grad_eliminate_]: 5.34998e-06
        [merge_forward]: 4.01001e-06
        [cell_reuse_recompute_pass]: 2.50002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.826e-05
        [j_node_and_user_rematch]: 9.62001e-06
        [meta_fg_expand]: 2.44001e-06
        [replace_old_param]: 9.02999e-06
        [inline_without_move]: 4.98001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 2.63998e-06
        [auto_monad_grad]: 1.74e-06
        [auto_monad_eliminator]: 7.94002e-06
        [cse]: 1.707e-05
        [replace_applicator]: 8.25e-06
[py_interpret_to_execute_after_opt_a]: 1.473e-05
[rewriter_after_opt_a]: 4.467e-05
[convert_after_rewriter]: 9.86e-06
[order_py_execute_after_rewriter]: 5.46002e-06
[mutable_eliminate]: 0.00092555
[jit_opt_b]: 6.921e-05, [1]
    [Cycle 1]: 5.886e-05, [2]
        [frontend_op_eliminate]: 2.322e-05
        [inline_after_opt_a]: 2.046e-05
[cconv]: 4.277e-05
[loop_unroll]: 0.00074705
[jit_opt_after_cconv]: 0.00021053, [1]
    [Cycle 1]: 0.00019956, [11]
        [c_1]: 2.798e-05
        [parameter_eliminate]: 6.58e-06
        [updatestate_depend_eliminate]: 9.06002e-06
        [updatestate_assign_eliminate]: 3.24001e-06
        [updatestate_loads_eliminate]: 3.22002e-06
        [cse]: 3.443e-05
        [call_graph_tuple_transform]: 3.321e-05
        [tuple_list_get_item_eliminator]: 7.51999e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 1.46002e-06
        [switch_simplify]: 7.08e-06
[remove_dup_value]: 1.811e-05
[partial_unused_args_eliminate]: 2.92002e-06
[environ_conv]: 9.19e-06
[add_recomputation]: 5.739e-05
[cse_after_recomputation]: 2.928e-05, [1]
    [Cycle 1]: 2.096e-05, [1]
        [cse]: 1.256e-05
[auto_monad_reorder]: 1.737e-05
[get_jit_bprop_graph]: 2.51e-06
[rewriter_after_jit_bprop_graph]: 8.40001e-06
[opt_after_jit_grad]: 0.00074662
[symbol_engine_optimizer]: 0.00011907, [1]
    [Cycle 1]: 0.00011001, [6]
        [build]: 4.84003e-06
        [elim_shapecalc]: 9.66e-06
        [elim_not_effective]: 4.055e-05
        [opt_reshape]: 8.25999e-06
        [fold_const_symbol]: 1.138e-05
        [renormalize]: 7.49977e-07
[validate]: 5.13e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00767967
[execute]: 7.98999e-06
Sums
    bootstrap                            :     0.000362s :  1.88%
    type_inference                       :     0.005575s : 28.99%
    event_method                         :     0.000020s :  0.10%
    auto_monad                           :     0.000073s :  0.38%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000049s :  0.26%
    rewriter_before_opt_a                :     0.000077s :  0.40%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000082s :  0.43%
    jit_opt_a.loop_unroll                :     0.000038s :  0.20%
    jit_opt_a.a_1                        :     0.000789s :  4.11%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.03%
    jit_opt_a.specialize_transform       :     0.000014s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.03%
    jit_opt_a.replace_old_param          :     0.000020s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001119s :  5.82%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.13%
    jit_opt_a.cse                        :     0.000049s :  0.25%
    jit_opt_a.replace_applicator         :     0.000032s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.08%
    rewriter_after_opt_a                 :     0.000045s :  0.23%
    convert_after_rewriter               :     0.000010s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000926s :  4.81%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.11%
    cconv                                :     0.000043s :  0.22%
    loop_unroll                          :     0.000747s :  3.88%
    jit_opt_after_cconv.c_1              :     0.000028s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000034s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000033s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.04%
    remove_dup_value                     :     0.000018s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000009s :  0.05%
    add_recomputation                    :     0.000057s :  0.30%
    cse_after_recomputation.cse          :     0.000013s :  0.07%
    auto_monad_reorder                   :     0.000017s :  0.09%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000747s :  3.88%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000041s :  0.21%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000051s :  0.27%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007680s : 39.94%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000280    23
 1.31% :     0.000004s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 2.98% :     0.000008s :      3: substitution.graph_param_transform
83.77% :     0.000235s :      4: substitution.inline
 1.63% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.06% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.10% :     0.000006s :      2: substitution.replace_old_param
 5.69% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005496     2
80.16% :     0.004405s :      1: type_inference.infer
19.84% :     0.001090s :      1: type_inference.specialize
------[replace.]   0.000094     6
62.76% :     0.000059s :      4: replace.inline
37.24% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000246     6
94.08% :     0.000231s :      4: match.inline
 5.92% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000151   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.49% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.82% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 2.85% :     0.000004s :     10: predicate.arithmetic_simplify
 1.75% :     0.000003s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.82% :     0.000001s :     10: predicate.compare_switch_simplify
 1.53% :     0.000002s :     10: predicate.depend_value_elim
 1.24% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.12% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.98% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.08% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.78% :     0.000001s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.82% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.81% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     10: predicate.environ_get_eliminate
 1.31% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.34% :     0.000001s :      3: predicate.graph_param_transform
 4.77% :     0.000007s :     22: predicate.inline
 0.78% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000002s :      6: predicate.less_batch_normalization
 1.79% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000003s :     15: predicate.load_eliminater
 1.93% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.87% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.46% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 0.85% :     0.000001s :     10: predicate.merge_addn
 0.82% :     0.000001s :     10: predicate.minmaximum_grad
 2.39% :     0.000004s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 1.85% :     0.000003s :     15: predicate.partial_eliminate
 1.33% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000003s :     18: predicate.replace_applicator
 0.98% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.26% :     0.000002s :     10: predicate.reshape_eliminate
 1.06% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.08% :     0.000002s :      3: predicate.row_tensor_eliminate
 2.01% :     0.000003s :     10: predicate.same_eliminate
 0.73% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000002s :      6: predicate.special_op_eliminate
 1.27% :     0.000002s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000004s :     16: predicate.switch_defer_inline
 1.62% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.61% :     0.000010s :     53: predicate.switch_simplify
 1.06% :     0.000002s :     10: predicate.tile_eliminate
 1.33% :     0.000002s :     10: predicate.transpose_eliminate
 1.25% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.88% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.82% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 2.60% :     0.000004s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.27% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.10% :     0.000003s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.50% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000809    12
48.15% :     0.000390s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.92% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.93% :     0.000380s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026776    76
 0.23% :     0.000061s :      1: add_recomputation
 0.29% :     0.000077s :      1: auto_monad
 0.08% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.42% :     0.000380s :      1: bootstrap
 0.17% :     0.000046s :      1: cconv
 0.05% :     0.000013s :      1: convert_after_rewriter
 0.12% :     0.000031s :      1: cse_after_recomputation
 0.04% :     0.000012s :      1: environ_conv
 0.09% :     0.000024s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
26.70% :     0.007150s :      1: jit_opt_a
 0.80% :     0.000215s :      1: jit_opt_after_cconv
 0.27% :     0.000072s :      1: jit_opt_b
 2.86% :     0.000766s :      1: loop_unroll
 3.52% :     0.000943s :      1: mutable_eliminate
 3.96% :     0.001061s :     26: opt.transform.jit_opt_a
 0.27% :     0.000072s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000034s :      4: opt.transform.jit_opt_b
 0.09% :     0.000025s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000028s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000038s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000065s :      4: opt.transform.symbol_engine_opt
 2.85% :     0.000762s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000052s :      1: py_interpret_to_execute
 0.07% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000022s :      1: remove_dup_value
 2.16% :     0.000580s :      1: renormalize.infer
 1.97% :     0.000527s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000049s :      1: rewriter_after_opt_a
 0.30% :     0.000080s :      1: rewriter_before_opt_a
 0.46% :     0.000123s :      1: symbol_engine_optimizer
28.76% :     0.007699s :      1: task_emit
20.88% :     0.005590s :      1: type_inference
 0.29% :     0.000077s :      1: validate


TotalTime = 0.0153154, [30]
[bootstrap]: 0.00037423
[type_inference]: 0.00532345
[event_method]: 2.259e-05
[auto_monad]: 8.089e-05
[graph_reusing]: 6.58e-06
[pre_auto_parallel]: 2.39999e-06
[py_interpret_to_execute]: 4.721e-05
[rewriter_before_opt_a]: 7.749e-05
[expand_dump_flag]: 3.38e-06
[jit_opt_a]: 0.00664525, [2]
    [Cycle 1]: 0.00194271, [27]
        [switch_simplify]: 9.386e-05
        [loop_unroll]: 3.212e-05
        [a_1]: 0.00066363
        [with_stream_mark]: 2.013e-05
        [recompute_prepare]: 9.87001e-06
        [updatestate_depend_eliminate]: 4.3e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 2.63e-06
        [specialize_transform]: 6.96001e-06
        [updatestate_useless_node_eliminater]: 5.46e-06
        [accelerated_algorithm]: 6.36e-06
        [meta_shard_fg_expand]: 2.36e-06
        [get_grad_eliminate_]: 5.77001e-06
        [merge_forward]: 3.69002e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.791e-05
        [j_node_and_user_rematch]: 1.061e-05
        [meta_fg_expand]: 2.44001e-06
        [replace_old_param]: 9.46e-06
        [inline_without_move]: 5.72001e-06
        [renormalize]: 0.00074732
        [add_forward_monad_depend]: 7.63001e-06
        [auto_monad_grad]: 2.58998e-06
        [auto_monad_eliminator]: 1.839e-05
        [cse]: 2.971e-05
        [replace_applicator]: 1.644e-05
    [Cycle 2]: 0.00036997, [27]
        [switch_simplify]: 5.88002e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 9.449e-05
        [with_stream_mark]: 1.17e-05
        [recompute_prepare]: 4.88001e-06
        [updatestate_depend_eliminate]: 3.68e-06
        [updatestate_assign_eliminate]: 9.92999e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [parameter_eliminate]: 1.78002e-06
        [specialize_transform]: 5.61e-06
        [updatestate_useless_node_eliminater]: 5.86e-06
        [accelerated_algorithm]: 6.01e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 5.19e-06
        [merge_forward]: 2.89999e-06
        [cell_reuse_recompute_pass]: 2.83e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.685e-05
        [j_node_and_user_rematch]: 8.70001e-06
        [meta_fg_expand]: 1.45999e-06
        [replace_old_param]: 8.85999e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.62999e-06
        [auto_monad_grad]: 1.23002e-06
        [auto_monad_eliminator]: 6.59001e-06
        [cse]: 1.437e-05
        [replace_applicator]: 6.14001e-06
[py_interpret_to_execute_after_opt_a]: 1.264e-05
[rewriter_after_opt_a]: 3.991e-05
[convert_after_rewriter]: 7.88001e-06
[order_py_execute_after_rewriter]: 7.5e-06
[mutable_eliminate]: 0.00077773
[jit_opt_b]: 5.64e-05, [1]
    [Cycle 1]: 4.834e-05, [2]
        [frontend_op_eliminate]: 1.765e-05
        [inline_after_opt_a]: 1.702e-05
[cconv]: 6.733e-05
[loop_unroll]: 0.00054748
[jit_opt_after_cconv]: 0.00017557, [1]
    [Cycle 1]: 0.0001676, [11]
        [c_1]: 2.325e-05
        [parameter_eliminate]: 5.09003e-06
        [updatestate_depend_eliminate]: 9.66998e-06
        [updatestate_assign_eliminate]: 3.13e-06
        [updatestate_loads_eliminate]: 2.66e-06
        [cse]: 3.01e-05
        [call_graph_tuple_transform]: 2.365e-05
        [tuple_list_get_item_eliminator]: 6.59001e-06
        [none_parameter_eliminate]: 1.50999e-06
        [renormalize]: 9.5999e-07
        [switch_simplify]: 5.57001e-06
[remove_dup_value]: 1.609e-05
[partial_unused_args_eliminate]: 2.29001e-06
[environ_conv]: 5.04e-06
[add_recomputation]: 5.644e-05
[cse_after_recomputation]: 2.627e-05, [1]
    [Cycle 1]: 1.96e-05, [1]
        [cse]: 1.17e-05
[auto_monad_reorder]: 1.579e-05
[get_jit_bprop_graph]: 1.82001e-06
[rewriter_after_jit_bprop_graph]: 7.89002e-06
[opt_after_jit_grad]: 0.00056375
[symbol_engine_optimizer]: 8.439e-05, [1]
    [Cycle 1]: 7.568e-05, [6]
        [build]: 5.29e-06
        [elim_shapecalc]: 9.29e-06
        [elim_not_effective]: 1.596e-05
        [opt_reshape]: 5.91e-06
        [fold_const_symbol]: 9.64e-06
        [renormalize]: 6.59988e-07
[validate]: 4.415e-05
Sums
    bootstrap                            :     0.000374s :  3.64%
    type_inference                       :     0.005323s : 51.73%
    event_method                         :     0.000023s :  0.22%
    auto_monad                           :     0.000081s :  0.79%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000047s :  0.46%
    rewriter_before_opt_a                :     0.000077s :  0.75%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000100s :  0.97%
    jit_opt_a.loop_unroll                :     0.000037s :  0.36%
    jit_opt_a.a_1                        :     0.000758s :  7.37%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.13%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000013s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.18%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000747s :  7.26%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.24%
    jit_opt_a.cse                        :     0.000044s :  0.43%
    jit_opt_a.replace_applicator         :     0.000023s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.12%
    rewriter_after_opt_a                 :     0.000040s :  0.39%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000007s :  0.07%
    mutable_eliminate                    :     0.000778s :  7.56%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.17%
    cconv                                :     0.000067s :  0.65%
    loop_unroll                          :     0.000547s :  5.32%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.09%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000030s :  0.29%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000016s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000056s :  0.55%
    cse_after_recomputation.cse          :     0.000012s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.08%
    opt_after_jit_grad                   :     0.000564s :  5.48%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.43%


Time group info:
------[substitution.]   0.000243    23
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.56% :     0.000006s :      3: substitution.graph_param_transform
83.41% :     0.000203s :      4: substitution.inline
 1.61% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.67% :     0.000007s :      4: substitution.remove_not_recompute_node
 2.28% :     0.000006s :      2: substitution.replace_old_param
 6.10% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005184     2
82.35% :     0.004269s :      1: type_inference.infer
17.65% :     0.000915s :      1: type_inference.specialize
------[replace.]   0.000107     6
55.30% :     0.000059s :      4: replace.inline
44.70% :     0.000048s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000214     6
93.61% :     0.000200s :      4: match.inline
 6.39% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   691
 1.43% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.17% :     0.000002s :     10: predicate.addn_zero_filter
 1.72% :     0.000002s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.15% :     0.000002s :     10: predicate.depend_value_elim
 1.24% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.69% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.01% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.72% :     0.000002s :     10: predicate.environ_get_eliminate
 1.15% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 6.33% :     0.000009s :     22: predicate.inline
 0.77% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.46% :     0.000002s :      6: predicate.less_batch_normalization
 1.86% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 2.13% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.18% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.96% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 1.58% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.48% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 1.13% :     0.000002s :      6: predicate.specialize_transform
 1.66% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.08% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.55% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.21% :     0.000004s :     16: predicate.switch_defer_inline
 2.02% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.93% :     0.000011s :     53: predicate.switch_simplify
 1.15% :     0.000002s :     10: predicate.tile_eliminate
 1.15% :     0.000002s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.48% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.34% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.32% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.48% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.87% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000682    12
54.47% :     0.000372s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.23% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.29% :     0.000275s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017152    72
 0.35% :     0.000061s :      1: add_recomputation
 0.49% :     0.000085s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 2.30% :     0.000394s :      1: bootstrap
 0.41% :     0.000071s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000029s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000027s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
38.77% :     0.006650s :      1: jit_opt_a
 1.04% :     0.000179s :      1: jit_opt_after_cconv
 0.35% :     0.000059s :      1: jit_opt_b
 3.26% :     0.000559s :      1: loop_unroll
 4.59% :     0.000787s :      1: mutable_eliminate
 6.00% :     0.001028s :     26: opt.transform.jit_opt_a
 0.32% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000027s :      4: opt.transform.jit_opt_b
 0.11% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.37% :     0.000578s :      1: opt_after_jit_grad
 0.06% :     0.000011s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000050s :      1: py_interpret_to_execute
 0.10% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.23% :     0.000383s :      1: renormalize.infer
 2.07% :     0.000356s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000044s :      1: rewriter_after_opt_a
 0.47% :     0.000081s :      1: rewriter_before_opt_a
 0.51% :     0.000088s :      1: symbol_engine_optimizer
31.16% :     0.005345s :      1: type_inference


TotalTime = 0.032308, [33]
[bootstrap]: 0.0001925
[type_inference]: 0.0117149
[event_method]: 5.509e-05
[auto_monad]: 0.00013819
[graph_reusing]: 7.13e-06
[pre_auto_parallel]: 2.53998e-06
[py_interpret_to_execute]: 4.39e-05
[rewriter_before_opt_a]: 0.00010193
[expand_dump_flag]: 3.71999e-06
[jit_opt_a]: 0.0111754, [3]
    [Cycle 1]: 0.00554679, [27]
        [switch_simplify]: 0.00011562
        [loop_unroll]: 4.294e-05
        [a_1]: 0.00084733
        [with_stream_mark]: 2.453e-05
        [recompute_prepare]: 1.922e-05
        [updatestate_depend_eliminate]: 7.53999e-06
        [updatestate_assign_eliminate]: 6.03998e-06
        [updatestate_loads_eliminate]: 6.03002e-06
        [parameter_eliminate]: 2.62001e-06
        [specialize_transform]: 5.074e-05
        [updatestate_useless_node_eliminater]: 1.349e-05
        [accelerated_algorithm]: 1.214e-05
        [meta_shard_fg_expand]: 3.53999e-06
        [get_grad_eliminate_]: 1.196e-05
        [merge_forward]: 8e-06
        [cell_reuse_recompute_pass]: 1.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 4.381e-05
        [j_node_and_user_rematch]: 2.29e-05
        [meta_fg_expand]: 0.00154731
        [replace_old_param]: 5.174e-05
        [inline_without_move]: 5.183e-05
        [renormalize]: 0.00218161
        [add_forward_monad_depend]: 9.31e-06
        [auto_monad_grad]: 5.47001e-06
        [auto_monad_eliminator]: 4.607e-05
        [cse]: 0.00014247
        [replace_applicator]: 6.167e-05
    [Cycle 2]: 0.00211462, [27]
        [switch_simplify]: 3.628e-05
        [loop_unroll]: 3.326e-05
        [a_1]: 0.00118231
        [with_stream_mark]: 1.205e-05
        [recompute_prepare]: 7.87e-06
        [updatestate_depend_eliminate]: 3.85998e-06
        [updatestate_assign_eliminate]: 2.94001e-06
        [updatestate_loads_eliminate]: 2.47001e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 6.98e-06
        [updatestate_useless_node_eliminater]: 6.12001e-06
        [accelerated_algorithm]: 6.54001e-06
        [meta_shard_fg_expand]: 1.72001e-06
        [get_grad_eliminate_]: 5.75001e-06
        [merge_forward]: 3.20998e-06
        [cell_reuse_recompute_pass]: 9.5999e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.337e-05
        [j_node_and_user_rematch]: 9.83998e-06
        [meta_fg_expand]: 6.676e-05
        [replace_old_param]: 1.012e-05
        [inline_without_move]: 5.96e-06
        [renormalize]: 0.00045628
        [add_forward_monad_depend]: 4e-06
        [auto_monad_grad]: 1.20999e-06
        [auto_monad_eliminator]: 1.105e-05
        [cse]: 1.949e-05
        [replace_applicator]: 1.17e-05
    [Cycle 3]: 0.00035805, [27]
        [switch_simplify]: 6.36998e-06
        [loop_unroll]: 5.86e-06
        [a_1]: 0.00011566
        [with_stream_mark]: 8.95999e-06
        [recompute_prepare]: 5.86e-06
        [updatestate_depend_eliminate]: 3.36999e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 6.09001e-06
        [updatestate_useless_node_eliminater]: 5.77999e-06
        [accelerated_algorithm]: 5.81e-06
        [meta_shard_fg_expand]: 1.25001e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 2.82002e-06
        [cell_reuse_recompute_pass]: 1.50999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.41e-05
        [j_node_and_user_rematch]: 9.68002e-06
        [meta_fg_expand]: 1.89999e-06
        [replace_old_param]: 8.28999e-06
        [inline_without_move]: 5.81e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.26002e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 6.73e-06
        [cse]: 1.34e-05
        [replace_applicator]: 5.96e-06
[py_interpret_to_execute_after_opt_a]: 9.76998e-06
[rewriter_after_opt_a]: 3.788e-05
[convert_after_rewriter]: 7.04001e-06
[order_py_execute_after_rewriter]: 5.32999e-06
[mutable_eliminate]: 0.0005004
[jit_opt_b]: 5.287e-05, [1]
    [Cycle 1]: 4.674e-05, [2]
        [frontend_op_eliminate]: 1.906e-05
        [inline_after_opt_a]: 1.706e-05
[cconv]: 2.111e-05
[loop_unroll]: 0.00045006
[jit_opt_after_cconv]: 0.00014932, [1]
    [Cycle 1]: 0.00014338, [11]
        [c_1]: 2.485e-05
        [parameter_eliminate]: 2.50002e-06
        [updatestate_depend_eliminate]: 5.91e-06
        [updatestate_assign_eliminate]: 2.84001e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [cse]: 2.182e-05
        [call_graph_tuple_transform]: 1.995e-05
        [tuple_list_get_item_eliminator]: 5.97001e-06
        [none_parameter_eliminate]: 1.39998e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 6.04999e-06
[remove_dup_value]: 1.598e-05
[partial_unused_args_eliminate]: 2.73e-06
[environ_conv]: 5.25999e-06
[add_recomputation]: 4.822e-05
[cse_after_recomputation]: 3.151e-05, [1]
    [Cycle 1]: 1.881e-05, [1]
        [cse]: 1.314e-05
[auto_monad_reorder]: 1.709e-05
[get_jit_bprop_graph]: 1.84e-06
[rewriter_after_jit_bprop_graph]: 4.65999e-06
[opt_after_jit_grad]: 0.00051189
[symbol_engine_optimizer]: 9.92e-05, [1]
    [Cycle 1]: 9.295e-05, [6]
        [build]: 3.6e-06
        [elim_shapecalc]: 9.54999e-06
        [elim_not_effective]: 1.482e-05
        [opt_reshape]: 7.11999e-06
        [fold_const_symbol]: 2.872e-05
        [renormalize]: 4.39992e-07
[validate]: 3.486e-05
[backend_pass]: 3.367e-05
[task_emit]: 0.00664052
[execute]: 7.45e-06
Sums
    bootstrap                            :     0.000192s :  0.68%
    type_inference                       :     0.011715s : 41.35%
    event_method                         :     0.000055s :  0.19%
    auto_monad                           :     0.000138s :  0.49%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.15%
    rewriter_before_opt_a                :     0.000102s :  0.36%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000158s :  0.56%
    jit_opt_a.loop_unroll                :     0.000082s :  0.29%
    jit_opt_a.a_1                        :     0.002145s :  7.57%
    jit_opt_a.with_stream_mark           :     0.000046s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000064s :  0.23%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.08%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000071s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001616s :  5.70%
    jit_opt_a.replace_old_param          :     0.000070s :  0.25%
    jit_opt_a.inline_without_move        :     0.000064s :  0.22%
    jit_opt_a.renormalize                :     0.002638s :  9.31%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000064s :  0.23%
    jit_opt_a.cse                        :     0.000175s :  0.62%
    jit_opt_a.replace_applicator         :     0.000079s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.03%
    rewriter_after_opt_a                 :     0.000038s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000500s :  1.77%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.06%
    cconv                                :     0.000021s :  0.07%
    loop_unroll                          :     0.000450s :  1.59%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000048s :  0.17%
    cse_after_recomputation.cse          :     0.000013s :  0.05%
    auto_monad_reorder                   :     0.000017s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000512s :  1.81%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000029s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.12%
    backend_pass                         :     0.000034s :  0.12%
    task_emit                            :     0.006641s : 23.44%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000493   113
 0.44% :     0.000002s :      3: substitution.elim_not_effective
 0.34% :     0.000002s :      3: substitution.fold_const_symbol
 1.18% :     0.000006s :      4: substitution.graph_param_transform
60.53% :     0.000299s :     14: substitution.inline
 3.20% :     0.000016s :      2: substitution.inline_without_move
 1.76% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.17% :     0.000011s :      7: substitution.minmaximum_grad
 1.20% :     0.000006s :      5: substitution.partial_eliminate
 2.18% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.06% :     0.000025s :     10: substitution.replace_applicator
 1.40% :     0.000007s :      7: substitution.replace_old_param
 0.51% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.46% :     0.000007s :      1: substitution.switch_simplify
 7.82% :     0.000039s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.82% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.92% :     0.000039s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011651     2
90.12% :     0.010500s :      1: type_inference.infer
 9.88% :     0.001151s :      1: type_inference.specialize
------[replace.]   0.000183    22
54.22% :     0.000099s :     14: replace.inline
16.65% :     0.000030s :      1: replace.switch_simplify
29.13% :     0.000053s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000318    22
91.32% :     0.000291s :     14: match.inline
 2.07% :     0.000007s :      1: match.switch_simplify
 6.60% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000377  2634
 1.69% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.43% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.49% :     0.000006s :     44: predicate.addn_check_dump
 1.67% :     0.000006s :     44: predicate.addn_zero_filter
 1.98% :     0.000007s :     44: predicate.arithmetic_simplify
 1.56% :     0.000006s :     44: predicate.cast_eliminate
 0.15% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.49% :     0.000006s :     44: predicate.compare_switch_simplify
 1.59% :     0.000006s :     44: predicate.depend_value_elim
 1.41% :     0.000005s :     44: predicate.dict_get_item_const_eliminator
 1.45% :     0.000005s :     44: predicate.dict_get_item_eliminator
 1.46% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.31% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      4: predicate.elim_not_effective
 0.25% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.48% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.48% :     0.000006s :     44: predicate.environ_get_add_eliminate
 1.44% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.55% :     0.000006s :     44: predicate.environ_get_eliminate
 1.47% :     0.000006s :     44: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.84% :     0.000003s :     19: predicate.get_grad_eliminate
 0.11% :     0.000000s :      4: predicate.graph_param_transform
 4.36% :     0.000016s :     73: predicate.inline
 1.86% :     0.000007s :     40: predicate.inline_without_move
 0.35% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.87% :     0.000003s :     19: predicate.less_batch_normalization
 1.78% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.97% :     0.000007s :     55: predicate.load_eliminater
 0.44% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.31% :     0.000012s :     91: predicate.loop_unroll_before_grad
 1.74% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.55% :     0.000006s :     44: predicate.merge_addn
 1.57% :     0.000006s :     44: predicate.minmaximum_grad
 0.55% :     0.000002s :      4: predicate.mutable_eliminate
 0.20% :     0.000001s :      4: predicate.opt_reshape
 2.30% :     0.000009s :     55: predicate.partial_eliminate
 1.52% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.82% :     0.000007s :     44: predicate.reduce_eliminate
 1.82% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.37% :     0.000009s :     97: predicate.replace_applicator
 0.91% :     0.000003s :     40: predicate.replace_old_param
 0.12% :     0.000000s :      4: predicate.reset_defer_inline
 1.84% :     0.000007s :     44: predicate.reshape_eliminate
 1.57% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.25% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.68% :     0.000006s :     44: predicate.same_eliminate
 0.49% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.38% :     0.000001s :      8: predicate.special_op_eliminate
 0.91% :     0.000003s :     19: predicate.specialize_transform
 1.67% :     0.000006s :     44: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 3.02% :     0.000011s :     65: predicate.switch_defer_inline
 2.72% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.53% :     0.000025s :    162: predicate.switch_simplify
 1.57% :     0.000006s :     44: predicate.tile_eliminate
 1.49% :     0.000006s :     44: predicate.transpose_eliminate
 1.78% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.68% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 2.91% :     0.000011s :     59: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.86% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000007s :     44: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.23% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001522    32
62.14% :     0.000946s :     14: func_graph_cloner_run.FuncGraphClonerGraph
37.86% :     0.000576s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.037954    91
 0.14% :     0.000051s :      1: add_recomputation
 0.38% :     0.000144s :      1: auto_monad
 0.05% :     0.000020s :      1: auto_monad_reorder
 0.10% :     0.000037s :      1: backend_pass
 0.55% :     0.000208s :      1: bootstrap
 0.06% :     0.000024s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000034s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.16% :     0.000059s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
29.45% :     0.011178s :      1: jit_opt_a
 0.40% :     0.000152s :      1: jit_opt_after_cconv
 0.15% :     0.000055s :      1: jit_opt_b
 1.21% :     0.000458s :      1: loop_unroll
 1.34% :     0.000509s :      1: mutable_eliminate
 7.48% :     0.002839s :     39: opt.transform.jit_opt_a
 0.14% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000030s :      4: opt.transform.jit_opt_b
 0.03% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000047s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000057s :      4: opt.transform.symbol_engine_opt
 1.37% :     0.000520s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000047s :      1: py_interpret_to_execute
 0.03% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 3.67% :     0.001392s :      2: renormalize.infer
 3.25% :     0.001232s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000041s :      1: rewriter_after_opt_a
 0.28% :     0.000105s :      1: rewriter_before_opt_a
 0.27% :     0.000102s :      1: symbol_engine_optimizer
17.52% :     0.006651s :      1: task_emit
30.90% :     0.011727s :      1: type_inference
 0.15% :     0.000056s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.394.684 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dcaf820> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.395.238 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0220975, [33]
[bootstrap]: 0.00035872
[type_inference]: 0.00586335
[event_method]: 1.967e-05
[auto_monad]: 7.269e-05
[graph_reusing]: 6.20002e-06
[pre_auto_parallel]: 2.69999e-06
[py_interpret_to_execute]: 4.908e-05
[rewriter_before_opt_a]: 7.728e-05
[expand_dump_flag]: 4.28001e-06
[jit_opt_a]: 0.00641985, [2]
    [Cycle 1]: 0.00216402, [27]
        [switch_simplify]: 6.951e-05
        [loop_unroll]: 3.355e-05
        [a_1]: 0.00067192
        [with_stream_mark]: 1.995e-05
        [recompute_prepare]: 9.71e-06
        [updatestate_depend_eliminate]: 3.97e-06
        [updatestate_assign_eliminate]: 3.04999e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 7.21001e-06
        [updatestate_useless_node_eliminater]: 5.35001e-06
        [accelerated_algorithm]: 6.37001e-06
        [meta_shard_fg_expand]: 2.09e-06
        [get_grad_eliminate_]: 5.77001e-06
        [merge_forward]: 3.53999e-06
        [cell_reuse_recompute_pass]: 1.77999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.683e-05
        [j_node_and_user_rematch]: 9.54e-06
        [meta_fg_expand]: 2.53998e-06
        [replace_old_param]: 9.96e-06
        [inline_without_move]: 5.56e-06
        [renormalize]: 0.00099972
        [add_forward_monad_depend]: 7.71001e-06
        [auto_monad_grad]: 2.36998e-06
        [auto_monad_eliminator]: 1.693e-05
        [cse]: 3.082e-05
        [replace_applicator]: 1.636e-05
    [Cycle 2]: 0.00038987, [27]
        [switch_simplify]: 6.65002e-06
        [loop_unroll]: 5.37999e-06
        [a_1]: 0.00010287
        [with_stream_mark]: 1.532e-05
        [recompute_prepare]: 5.75001e-06
        [updatestate_depend_eliminate]: 3.40998e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 1.37999e-06
        [specialize_transform]: 5.39998e-06
        [updatestate_useless_node_eliminater]: 5.12999e-06
        [accelerated_algorithm]: 6.56e-06
        [meta_shard_fg_expand]: 1.32999e-06
        [get_grad_eliminate_]: 5.05999e-06
        [merge_forward]: 3.04001e-06
        [cell_reuse_recompute_pass]: 2.66999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.717e-05
        [j_node_and_user_rematch]: 9.60001e-06
        [meta_fg_expand]: 1.94999e-06
        [replace_old_param]: 8.90001e-06
        [inline_without_move]: 4.79e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.76e-06
        [auto_monad_grad]: 1.50001e-06
        [auto_monad_eliminator]: 7.18e-06
        [cse]: 1.41e-05
        [replace_applicator]: 6.32001e-06
[py_interpret_to_execute_after_opt_a]: 1.396e-05
[rewriter_after_opt_a]: 3.87e-05
[convert_after_rewriter]: 7.28e-06
[order_py_execute_after_rewriter]: 4.91002e-06
[mutable_eliminate]: 0.00078653
[jit_opt_b]: 6.084e-05, [1]
    [Cycle 1]: 5.257e-05, [2]
        [frontend_op_eliminate]: 2.112e-05
        [inline_after_opt_a]: 1.794e-05
[cconv]: 3.227e-05
[loop_unroll]: 0.00053196
[jit_opt_after_cconv]: 0.00016937, [1]
    [Cycle 1]: 0.00016157, [11]
        [c_1]: 2.396e-05
        [parameter_eliminate]: 4.55999e-06
        [updatestate_depend_eliminate]: 7.36999e-06
        [updatestate_assign_eliminate]: 3.06001e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [cse]: 2.584e-05
        [call_graph_tuple_transform]: 2.413e-05
        [tuple_list_get_item_eliminator]: 5.63002e-06
        [none_parameter_eliminate]: 1.87999e-06
        [renormalize]: 9.5999e-07
        [switch_simplify]: 5.95002e-06
[remove_dup_value]: 1.704e-05
[partial_unused_args_eliminate]: 2.58998e-06
[environ_conv]: 6.17999e-06
[add_recomputation]: 5.684e-05
[cse_after_recomputation]: 2.439e-05, [1]
    [Cycle 1]: 1.773e-05, [1]
        [cse]: 9.86998e-06
[auto_monad_reorder]: 1.727e-05
[get_jit_bprop_graph]: 2.94001e-06
[rewriter_after_jit_bprop_graph]: 8.72e-06
[opt_after_jit_grad]: 0.00059724
[symbol_engine_optimizer]: 8.919e-05, [1]
    [Cycle 1]: 7.943e-05, [6]
        [build]: 5.13002e-06
        [elim_shapecalc]: 9.10999e-06
        [elim_not_effective]: 1.724e-05
        [opt_reshape]: 6.81001e-06
        [fold_const_symbol]: 9.87999e-06
        [renormalize]: 9.50007e-07
[validate]: 4.347e-05
[backend_pass]: 1.17e-06
[task_emit]: 0.00646148
[execute]: 8.50999e-06
Sums
    bootstrap                            :     0.000359s :  2.05%
    type_inference                       :     0.005863s : 33.48%
    event_method                         :     0.000020s :  0.11%
    auto_monad                           :     0.000073s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000049s :  0.28%
    rewriter_before_opt_a                :     0.000077s :  0.44%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.43%
    jit_opt_a.loop_unroll                :     0.000039s :  0.22%
    jit_opt_a.a_1                        :     0.000775s :  4.42%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001000s :  5.71%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.14%
    jit_opt_a.cse                        :     0.000045s :  0.26%
    jit_opt_a.replace_applicator         :     0.000023s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000039s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000787s :  4.49%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.10%
    cconv                                :     0.000032s :  0.18%
    loop_unroll                          :     0.000532s :  3.04%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000017s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000057s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000597s :  3.41%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006461s : 36.90%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000254    23
 1.07% :     0.000003s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.43% :     0.000006s :      3: substitution.graph_param_transform
83.62% :     0.000213s :      4: substitution.inline
 1.62% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.72% :     0.000007s :      4: substitution.remove_not_recompute_node
 2.45% :     0.000006s :      2: substitution.replace_old_param
 5.59% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005783     2
81.96% :     0.004740s :      1: type_inference.infer
18.04% :     0.001043s :      1: type_inference.specialize
------[replace.]   0.000095     6
62.61% :     0.000060s :      4: replace.inline
37.39% :     0.000036s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000222     6
94.13% :     0.000209s :      4: match.inline
 5.87% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000140   691
 1.70% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.58% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.87% :     0.000001s :     10: predicate.addn_check_dump
 1.59% :     0.000002s :     10: predicate.addn_zero_filter
 2.80% :     0.000004s :     10: predicate.arithmetic_simplify
 1.42% :     0.000002s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.89% :     0.000001s :     10: predicate.compare_switch_simplify
 1.78% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.04% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.35% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.71% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.94% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.31% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.47% :     0.000002s :     10: predicate.environ_get_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000001s :      3: predicate.graph_param_transform
 4.67% :     0.000007s :     22: predicate.inline
 0.74% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :      6: predicate.less_batch_normalization
 1.18% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.87% :     0.000003s :     15: predicate.load_eliminater
 1.54% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.63% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.85% :     0.000004s :     13: predicate.make_slice_get_slice_eliminator
 0.86% :     0.000001s :     10: predicate.merge_addn
 0.92% :     0.000001s :     10: predicate.minmaximum_grad
 2.79% :     0.000004s :      3: predicate.mutable_eliminate
 0.87% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.73% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.86% :     0.000003s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.41% :     0.000001s :      3: predicate.reset_defer_inline
 1.29% :     0.000002s :     10: predicate.reshape_eliminate
 1.05% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.16% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000002s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.27% :     0.000002s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.52% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 0.99% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 1.67% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.83% :     0.000010s :     53: predicate.switch_simplify
 1.02% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000002s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.01% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.01% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.90% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.97% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.33% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.37% :     0.000002s :     10: predicate.value_based_eliminate
 0.27% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000778    12
49.51% :     0.000385s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.59% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.91% :     0.000357s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024250    76
 0.25% :     0.000060s :      1: add_recomputation
 0.32% :     0.000077s :      1: auto_monad
 0.09% :     0.000021s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 1.55% :     0.000377s :      1: bootstrap
 0.15% :     0.000035s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.05% :     0.000013s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.49% :     0.006424s :      1: jit_opt_a
 0.71% :     0.000173s :      1: jit_opt_after_cconv
 0.26% :     0.000064s :      1: jit_opt_b
 2.24% :     0.000543s :      1: loop_unroll
 3.29% :     0.000799s :      1: mutable_eliminate
 4.22% :     0.001024s :     26: opt.transform.jit_opt_a
 0.23% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000031s :      4: opt.transform.jit_opt_b
 0.07% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000039s :      4: opt.transform.symbol_engine_opt
 2.51% :     0.000609s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000052s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000019s :      1: remove_dup_value
 2.18% :     0.000528s :      1: renormalize.infer
 1.90% :     0.000460s :      1: renormalize.specialize
 0.04% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000043s :      1: rewriter_after_opt_a
 0.33% :     0.000080s :      1: rewriter_before_opt_a
 0.38% :     0.000093s :      1: symbol_engine_optimizer
26.72% :     0.006480s :      1: task_emit
24.24% :     0.005878s :      1: type_inference
 0.28% :     0.000068s :      1: validate


TotalTime = 0.0202322, [33]
[bootstrap]: 0.00032741
[type_inference]: 0.00535482
[event_method]: 1.83e-05
[auto_monad]: 6.58e-05
[graph_reusing]: 5.54998e-06
[pre_auto_parallel]: 2.83e-06
[py_interpret_to_execute]: 4.059e-05
[rewriter_before_opt_a]: 7.146e-05
[expand_dump_flag]: 3.16001e-06
[jit_opt_a]: 0.0058027, [2]
    [Cycle 1]: 0.00174571, [27]
        [switch_simplify]: 5.315e-05
        [loop_unroll]: 2.792e-05
        [a_1]: 0.00056027
        [with_stream_mark]: 2.098e-05
        [recompute_prepare]: 7.06999e-06
        [updatestate_depend_eliminate]: 3.6e-06
        [updatestate_assign_eliminate]: 2.71e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 6.02001e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.08002e-06
        [meta_shard_fg_expand]: 2.39999e-06
        [get_grad_eliminate_]: 5.49e-06
        [merge_forward]: 3.80998e-06
        [cell_reuse_recompute_pass]: 2.05002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.422e-05
        [j_node_and_user_rematch]: 9.67001e-06
        [meta_fg_expand]: 2.26998e-06
        [replace_old_param]: 9.14e-06
        [inline_without_move]: 5.10999e-06
        [renormalize]: 0.00076606
        [add_forward_monad_depend]: 5.87001e-06
        [auto_monad_grad]: 2.98e-06
        [auto_monad_eliminator]: 1.443e-05
        [cse]: 2.37e-05
        [replace_applicator]: 1.291e-05
    [Cycle 2]: 0.00031818, [27]
        [switch_simplify]: 5.30999e-06
        [loop_unroll]: 5.10001e-06
        [a_1]: 8.879e-05
        [with_stream_mark]: 9.95002e-06
        [recompute_prepare]: 4.75999e-06
        [updatestate_depend_eliminate]: 2.59999e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.58999e-06
        [accelerated_algorithm]: 5.10999e-06
        [meta_shard_fg_expand]: 1.16002e-06
        [get_grad_eliminate_]: 4.45e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 2.84001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.307e-05
        [j_node_and_user_rematch]: 7.65e-06
        [meta_fg_expand]: 1.53002e-06
        [replace_old_param]: 7.7e-06
        [inline_without_move]: 4.74e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 9.99979e-07
        [auto_monad_eliminator]: 4.52e-06
        [cse]: 1.011e-05
        [replace_applicator]: 5.29e-06
[py_interpret_to_execute_after_opt_a]: 1.156e-05
[rewriter_after_opt_a]: 3.466e-05
[convert_after_rewriter]: 7.01999e-06
[order_py_execute_after_rewriter]: 5.02e-06
[mutable_eliminate]: 0.00065288
[jit_opt_b]: 4.98e-05, [1]
    [Cycle 1]: 4.189e-05, [2]
        [frontend_op_eliminate]: 1.576e-05
        [inline_after_opt_a]: 1.473e-05
[cconv]: 2.888e-05
[loop_unroll]: 0.00049656
[jit_opt_after_cconv]: 0.00014384, [1]
    [Cycle 1]: 0.00013699, [11]
        [c_1]: 2.012e-05
        [parameter_eliminate]: 3.59002e-06
        [updatestate_depend_eliminate]: 6.02999e-06
        [updatestate_assign_eliminate]: 2.28002e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [cse]: 2.256e-05
        [call_graph_tuple_transform]: 1.991e-05
        [tuple_list_get_item_eliminator]: 5.10001e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 7.09988e-07
        [switch_simplify]: 5.15999e-06
[remove_dup_value]: 1.504e-05
[partial_unused_args_eliminate]: 1.99e-06
[environ_conv]: 5.07e-06
[add_recomputation]: 4.941e-05
[cse_after_recomputation]: 2.081e-05, [1]
    [Cycle 1]: 1.515e-05, [1]
        [cse]: 9.04e-06
[auto_monad_reorder]: 1.45e-05
[get_jit_bprop_graph]: 2.17001e-06
[rewriter_after_jit_bprop_graph]: 6.51e-06
[opt_after_jit_grad]: 0.00046148
[symbol_engine_optimizer]: 7.144e-05, [1]
    [Cycle 1]: 6.444e-05, [6]
        [build]: 3.53999e-06
        [elim_shapecalc]: 7.28e-06
        [elim_not_effective]: 1.186e-05
        [opt_reshape]: 5.32001e-06
        [fold_const_symbol]: 8.84e-06
        [renormalize]: 5.89993e-07
[validate]: 3.728e-05
[backend_pass]: 8.99978e-07
[task_emit]: 0.00621484
[execute]: 6.34001e-06
Sums
    bootstrap                            :     0.000327s :  2.06%
    type_inference                       :     0.005355s : 33.70%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000066s :  0.41%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000041s :  0.26%
    rewriter_before_opt_a                :     0.000071s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.37%
    jit_opt_a.loop_unroll                :     0.000033s :  0.21%
    jit_opt_a.a_1                        :     0.000649s :  4.09%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000766s :  4.82%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000034s :  0.21%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000653s :  4.11%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000029s :  0.18%
    loop_unroll                          :     0.000497s :  3.13%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000461s :  2.90%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006215s : 39.12%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000222    23
 0.81% :     0.000002s :      2: substitution.elim_not_effective
 0.80% :     0.000002s :      2: substitution.fold_const_symbol
 2.59% :     0.000006s :      3: substitution.graph_param_transform
84.55% :     0.000188s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.00% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000005s :      2: substitution.replace_old_param
 5.62% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005284     2
81.58% :     0.004311s :      1: type_inference.infer
18.42% :     0.000973s :      1: type_inference.specialize
------[replace.]   0.000074     6
62.97% :     0.000047s :      4: replace.inline
37.03% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000196     6
94.21% :     0.000185s :      4: match.inline
 5.79% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.21% :     0.000003s :     10: predicate.arithmetic_simplify
 1.32% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.75% :     0.000002s :     10: predicate.depend_value_elim
 1.44% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.15% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.49% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.81% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 2.15% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000002s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.36% :     0.000002s :     10: predicate.reduce_eliminate
 1.61% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.68% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.27% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.20% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.57% :     0.000004s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.69% :     0.000009s :     53: predicate.switch_simplify
 1.46% :     0.000002s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.27% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.17% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.35% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000667    12
49.95% :     0.000333s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.70% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.35% :     0.000289s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021952    76
 0.24% :     0.000052s :      1: add_recomputation
 0.32% :     0.000070s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.55% :     0.000340s :      1: bootstrap
 0.15% :     0.000032s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.04% :     0.000010s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.45% :     0.005806s :      1: jit_opt_a
 0.67% :     0.000147s :      1: jit_opt_after_cconv
 0.24% :     0.000052s :      1: jit_opt_b
 2.30% :     0.000506s :      1: loop_unroll
 3.02% :     0.000662s :      1: mutable_eliminate
 3.90% :     0.000856s :     26: opt.transform.jit_opt_a
 0.21% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.14% :     0.000469s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000044s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.94% :     0.000426s :      1: renormalize.infer
 1.51% :     0.000332s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000038s :      1: rewriter_after_opt_a
 0.34% :     0.000074s :      1: rewriter_before_opt_a
 0.34% :     0.000074s :      1: symbol_engine_optimizer
28.37% :     0.006228s :      1: task_emit
24.46% :     0.005370s :      1: type_inference
 0.26% :     0.000058s :      1: validate


TotalTime = 0.0135608, [30]
[bootstrap]: 0.00031649
[type_inference]: 0.00450963
[event_method]: 1.782e-05
[auto_monad]: 6.326e-05
[graph_reusing]: 5.44998e-06
[pre_auto_parallel]: 2.64001e-06
[py_interpret_to_execute]: 3.099e-05
[rewriter_before_opt_a]: 7.147e-05
[expand_dump_flag]: 3.10998e-06
[jit_opt_a]: 0.00582308, [2]
    [Cycle 1]: 0.00177633, [27]
        [switch_simplify]: 5.877e-05
        [loop_unroll]: 3.147e-05
        [a_1]: 0.00057304
        [with_stream_mark]: 2.075e-05
        [recompute_prepare]: 8e-06
        [updatestate_depend_eliminate]: 4.1e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.64001e-06
        [parameter_eliminate]: 1.91998e-06
        [specialize_transform]: 6.58e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 6.12999e-06
        [meta_shard_fg_expand]: 2.04e-06
        [get_grad_eliminate_]: 5.13002e-06
        [merge_forward]: 4.58999e-06
        [cell_reuse_recompute_pass]: 1.45001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.583e-05
        [j_node_and_user_rematch]: 9.27999e-06
        [meta_fg_expand]: 2.19999e-06
        [replace_old_param]: 1.175e-05
        [inline_without_move]: 5.72001e-06
        [renormalize]: 0.00071369
        [add_forward_monad_depend]: 7.36999e-06
        [auto_monad_grad]: 2.73e-06
        [auto_monad_eliminator]: 1.764e-05
        [cse]: 3.372e-05
        [replace_applicator]: 1.663e-05
    [Cycle 2]: 0.00037614, [27]
        [switch_simplify]: 5.85002e-06
        [loop_unroll]: 5.03002e-06
        [a_1]: 9.486e-05
        [with_stream_mark]: 1.394e-05
        [recompute_prepare]: 7.36001e-06
        [updatestate_depend_eliminate]: 3.14999e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [parameter_eliminate]: 1.95001e-06
        [specialize_transform]: 6.15002e-06
        [updatestate_useless_node_eliminater]: 5.57999e-06
        [accelerated_algorithm]: 6.61e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 3.08e-06
        [cell_reuse_recompute_pass]: 2.81e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.692e-05
        [j_node_and_user_rematch]: 9.30001e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 8.95001e-06
        [inline_without_move]: 4.97e-06
        [renormalize]: 1.10012e-07
        [add_forward_monad_depend]: 1.84e-06
        [auto_monad_grad]: 1.15999e-06
        [auto_monad_eliminator]: 8.03999e-06
        [cse]: 1.662e-05
        [replace_applicator]: 5.99999e-06
[py_interpret_to_execute_after_opt_a]: 1.433e-05
[rewriter_after_opt_a]: 5.432e-05
[convert_after_rewriter]: 7.76001e-06
[order_py_execute_after_rewriter]: 5.04e-06
[mutable_eliminate]: 0.00100713
[jit_opt_b]: 5.546e-05, [1]
    [Cycle 1]: 4.647e-05, [2]
        [frontend_op_eliminate]: 1.7e-05
        [inline_after_opt_a]: 1.656e-05
[cconv]: 2.851e-05
[loop_unroll]: 0.00047232
[jit_opt_after_cconv]: 0.00015798, [1]
    [Cycle 1]: 0.00015084, [11]
        [c_1]: 2.196e-05
        [parameter_eliminate]: 3.50998e-06
        [updatestate_depend_eliminate]: 6.91999e-06
        [updatestate_assign_eliminate]: 2.68003e-06
        [updatestate_loads_eliminate]: 2.56998e-06
        [cse]: 2.631e-05
        [call_graph_tuple_transform]: 2.118e-05
        [tuple_list_get_item_eliminator]: 5.76e-06
        [none_parameter_eliminate]: 1.92999e-06
        [renormalize]: 8.89995e-07
        [switch_simplify]: 5.51e-06
[remove_dup_value]: 1.522e-05
[partial_unused_args_eliminate]: 2.46e-06
[environ_conv]: 5.84e-06
[add_recomputation]: 4.8e-05
[cse_after_recomputation]: 2.156e-05, [1]
    [Cycle 1]: 1.579e-05, [1]
        [cse]: 9.82001e-06
[auto_monad_reorder]: 1.557e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 6.44999e-06
[opt_after_jit_grad]: 0.00049523
[symbol_engine_optimizer]: 7.665e-05, [1]
    [Cycle 1]: 7.002e-05, [6]
        [build]: 4.78001e-06
        [elim_shapecalc]: 7.8e-06
        [elim_not_effective]: 1.358e-05
        [opt_reshape]: 5.74e-06
        [fold_const_symbol]: 9.14e-06
        [renormalize]: 6.50005e-07
[validate]: 4.012e-05
Sums
    bootstrap                            :     0.000316s :  3.42%
    type_inference                       :     0.004510s : 48.80%
    event_method                         :     0.000018s :  0.19%
    auto_monad                           :     0.000063s :  0.68%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000031s :  0.34%
    rewriter_before_opt_a                :     0.000071s :  0.77%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000065s :  0.70%
    jit_opt_a.loop_unroll                :     0.000037s :  0.39%
    jit_opt_a.a_1                        :     0.000668s :  7.23%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.38%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000013s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000021s :  0.22%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000714s :  7.72%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.28%
    jit_opt_a.cse                        :     0.000050s :  0.54%
    jit_opt_a.replace_applicator         :     0.000023s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.16%
    rewriter_after_opt_a                 :     0.000054s :  0.59%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.001007s : 10.90%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.18%
    cconv                                :     0.000029s :  0.31%
    loop_unroll                          :     0.000472s :  5.11%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000026s :  0.28%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000048s :  0.52%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000495s :  5.36%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.43%


Time group info:
------[substitution.]   0.000230    23
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.44% :     0.000006s :      3: substitution.graph_param_transform
83.85% :     0.000193s :      4: substitution.inline
 1.71% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.12% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.63% :     0.000006s :      2: substitution.replace_old_param
 5.83% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004450     2
81.36% :     0.003621s :      1: type_inference.infer
18.64% :     0.000830s :      1: type_inference.specialize
------[replace.]   0.000073     6
66.60% :     0.000049s :      4: replace.inline
33.40% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000202     6
94.01% :     0.000190s :      4: match.inline
 5.99% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.18% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.43% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.55% :     0.000002s :     10: predicate.addn_zero_filter
 2.00% :     0.000003s :     10: predicate.arithmetic_simplify
 1.11% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.09% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.05% :     0.000001s :     10: predicate.environ_get_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.97% :     0.000008s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.67% :     0.000002s :      6: predicate.less_batch_normalization
 1.89% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.20% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 0.93% :     0.000001s :     10: predicate.minmaximum_grad
 2.38% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.36% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.58% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.71% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 1.24% :     0.000002s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.10% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.53% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.44% :     0.000009s :     53: predicate.switch_simplify
 1.10% :     0.000001s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.09% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.10% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.98% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000609    12
49.50% :     0.000301s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.07% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.42% :     0.000264s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015235    72
 0.33% :     0.000051s :      1: add_recomputation
 0.44% :     0.000066s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.19% :     0.000333s :      1: bootstrap
 0.20% :     0.000031s :      1: cconv
 0.07% :     0.000011s :      1: convert_after_rewriter
 0.16% :     0.000024s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.14% :     0.000022s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.25% :     0.005827s :      1: jit_opt_a
 1.05% :     0.000161s :      1: jit_opt_after_cconv
 0.38% :     0.000058s :      1: jit_opt_b
 3.16% :     0.000482s :      1: loop_unroll
 6.70% :     0.001021s :      1: mutable_eliminate
 5.92% :     0.000902s :     26: opt.transform.jit_opt_a
 0.33% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.14% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.30% :     0.000503s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000034s :      1: py_interpret_to_execute
 0.11% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000018s :      1: remove_dup_value
 2.43% :     0.000371s :      1: renormalize.infer
 2.19% :     0.000334s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.39% :     0.000060s :      1: rewriter_after_opt_a
 0.49% :     0.000074s :      1: rewriter_before_opt_a
 0.52% :     0.000079s :      1: symbol_engine_optimizer
29.68% :     0.004522s :      1: type_inference


TotalTime = 0.0228543, [33]
[bootstrap]: 0.00034587
[type_inference]: 0.00569249
[event_method]: 1.75e-05
[auto_monad]: 7.055e-05
[graph_reusing]: 6.09001e-06
[pre_auto_parallel]: 2.36998e-06
[py_interpret_to_execute]: 4.196e-05
[rewriter_before_opt_a]: 7.366e-05
[expand_dump_flag]: 3.55e-06
[jit_opt_a]: 0.00684565, [2]
    [Cycle 1]: 0.00212382, [27]
        [switch_simplify]: 6.814e-05
        [loop_unroll]: 3.169e-05
        [a_1]: 0.00061945
        [with_stream_mark]: 2.122e-05
        [recompute_prepare]: 7.96001e-06
        [updatestate_depend_eliminate]: 4.02e-06
        [updatestate_assign_eliminate]: 2.94001e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 6.58e-06
        [updatestate_useless_node_eliminater]: 5.64e-06
        [accelerated_algorithm]: 5.34998e-06
        [meta_shard_fg_expand]: 2.36e-06
        [get_grad_eliminate_]: 5.42999e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 1.18001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.553e-05
        [j_node_and_user_rematch]: 1.001e-05
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 1.065e-05
        [inline_without_move]: 5.51e-06
        [renormalize]: 0.0010456
        [add_forward_monad_depend]: 7.89002e-06
        [auto_monad_grad]: 3.01001e-06
        [auto_monad_eliminator]: 1.42e-05
        [cse]: 1.801e-05
        [replace_applicator]: 1.589e-05
    [Cycle 2]: 0.00034025, [27]
        [switch_simplify]: 5.75001e-06
        [loop_unroll]: 5.54e-06
        [a_1]: 9.473e-05
        [with_stream_mark]: 1.153e-05
        [recompute_prepare]: 5.15001e-06
        [updatestate_depend_eliminate]: 2.54001e-06
        [updatestate_assign_eliminate]: 2.52001e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 5.02999e-06
        [accelerated_algorithm]: 5.98002e-06
        [meta_shard_fg_expand]: 1.26997e-06
        [get_grad_eliminate_]: 5.02e-06
        [merge_forward]: 2.76e-06
        [cell_reuse_recompute_pass]: 1.74e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.402e-05
        [j_node_and_user_rematch]: 8.31002e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 1.542e-05
        [inline_without_move]: 5.53002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.75001e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 5.19e-06
        [cse]: 1.093e-05
        [replace_applicator]: 5.25001e-06
[py_interpret_to_execute_after_opt_a]: 1.372e-05
[rewriter_after_opt_a]: 3.778e-05
[convert_after_rewriter]: 6.79001e-06
[order_py_execute_after_rewriter]: 4.67e-06
[mutable_eliminate]: 0.00081749
[jit_opt_b]: 5.64e-05, [1]
    [Cycle 1]: 4.857e-05, [2]
        [frontend_op_eliminate]: 1.941e-05
        [inline_after_opt_a]: 1.675e-05
[cconv]: 3.037e-05
[loop_unroll]: 0.00045517
[jit_opt_after_cconv]: 0.00015417, [1]
    [Cycle 1]: 0.00014689, [11]
        [c_1]: 2.308e-05
        [parameter_eliminate]: 3.28e-06
        [updatestate_depend_eliminate]: 6.28e-06
        [updatestate_assign_eliminate]: 2.59999e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 2.445e-05
        [call_graph_tuple_transform]: 2.077e-05
        [tuple_list_get_item_eliminator]: 5.64998e-06
        [none_parameter_eliminate]: 1.76003e-06
        [renormalize]: 1.00999e-06
        [switch_simplify]: 5.37999e-06
[remove_dup_value]: 1.499e-05
[partial_unused_args_eliminate]: 2.11e-06
[environ_conv]: 5.34998e-06
[add_recomputation]: 4.927e-05
[cse_after_recomputation]: 2.201e-05, [1]
    [Cycle 1]: 1.605e-05, [1]
        [cse]: 1.01e-05
[auto_monad_reorder]: 1.621e-05
[get_jit_bprop_graph]: 1.97999e-06
[rewriter_after_jit_bprop_graph]: 7.54002e-06
[opt_after_jit_grad]: 0.00054392
[symbol_engine_optimizer]: 8.354e-05, [1]
    [Cycle 1]: 7.423e-05, [6]
        [build]: 4.79e-06
        [elim_shapecalc]: 8.78001e-06
        [elim_not_effective]: 1.433e-05
        [opt_reshape]: 6.96999e-06
        [fold_const_symbol]: 9.59e-06
        [renormalize]: 1.04003e-06
[validate]: 4.132e-05
[backend_pass]: 1.10999e-06
[task_emit]: 0.0071581
[execute]: 7.48999e-06
Sums
    bootstrap                            :     0.000346s :  1.94%
    type_inference                       :     0.005692s : 31.94%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000071s :  0.40%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000042s :  0.24%
    rewriter_before_opt_a                :     0.000074s :  0.41%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.41%
    jit_opt_a.loop_unroll                :     0.000037s :  0.21%
    jit_opt_a.a_1                        :     0.000714s :  4.01%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000026s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001046s :  5.87%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.11%
    jit_opt_a.cse                        :     0.000029s :  0.16%
    jit_opt_a.replace_applicator         :     0.000021s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000038s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000817s :  4.59%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000030s :  0.17%
    loop_unroll                          :     0.000455s :  2.55%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.28%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000544s :  3.05%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007158s : 40.16%
    execute                              :     0.000007s :  0.04%


TotalTime = 0.033254, [33]
[bootstrap]: 0.00020584
[type_inference]: 0.0120164
[event_method]: 5.86e-05
[auto_monad]: 0.00010565
[graph_reusing]: 7.23e-06
[pre_auto_parallel]: 2.97002e-06
[py_interpret_to_execute]: 2.708e-05
[rewriter_before_opt_a]: 0.00012855
[expand_dump_flag]: 3.48e-06
[jit_opt_a]: 0.0113902, [3]
    [Cycle 1]: 0.00561679, [27]
        [switch_simplify]: 0.00012516
        [loop_unroll]: 4.609e-05
        [a_1]: 0.00087693
        [with_stream_mark]: 2.529e-05
        [recompute_prepare]: 1.881e-05
        [updatestate_depend_eliminate]: 8.59e-06
        [updatestate_assign_eliminate]: 6.84999e-06
        [updatestate_loads_eliminate]: 6.61e-06
        [parameter_eliminate]: 2.54001e-06
        [specialize_transform]: 1.298e-05
        [updatestate_useless_node_eliminater]: 1.251e-05
        [accelerated_algorithm]: 1.265e-05
        [meta_shard_fg_expand]: 3.46999e-06
        [get_grad_eliminate_]: 1.212e-05
        [merge_forward]: 7.98999e-06
        [cell_reuse_recompute_pass]: 9.20001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.651e-05
        [j_node_and_user_rematch]: 2.24e-05
        [meta_fg_expand]: 0.00148127
        [replace_old_param]: 5.449e-05
        [inline_without_move]: 5.145e-05
        [renormalize]: 0.00230718
        [add_forward_monad_depend]: 9.22001e-06
        [auto_monad_grad]: 5.70001e-06
        [auto_monad_eliminator]: 4.746e-05
        [cse]: 0.00014729
        [replace_applicator]: 6.226e-05
    [Cycle 2]: 0.00209865, [27]
        [switch_simplify]: 3.581e-05
        [loop_unroll]: 3.393e-05
        [a_1]: 0.00108331
        [with_stream_mark]: 1.153e-05
        [recompute_prepare]: 7.56999e-06
        [updatestate_depend_eliminate]: 3.68e-06
        [updatestate_assign_eliminate]: 3.3e-06
        [updatestate_loads_eliminate]: 2.56998e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 7.20003e-06
        [updatestate_useless_node_eliminater]: 6.17999e-06
        [accelerated_algorithm]: 5.94999e-06
        [meta_shard_fg_expand]: 1.79e-06
        [get_grad_eliminate_]: 5.52999e-06
        [merge_forward]: 3.18e-06
        [cell_reuse_recompute_pass]: 7.49977e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.343e-05
        [j_node_and_user_rematch]: 9.96e-06
        [meta_fg_expand]: 6.888e-05
        [replace_old_param]: 9.86e-06
        [inline_without_move]: 6.33002e-06
        [renormalize]: 0.00057474
        [add_forward_monad_depend]: 4.15e-06
        [auto_monad_grad]: 1.32e-06
        [auto_monad_eliminator]: 1.265e-05
        [cse]: 2.018e-05
        [replace_applicator]: 1.293e-05
    [Cycle 3]: 0.00039561, [27]
        [switch_simplify]: 7.13e-06
        [loop_unroll]: 7.10002e-06
        [a_1]: 0.00012331
        [with_stream_mark]: 9.02e-06
        [recompute_prepare]: 5.96e-06
        [updatestate_depend_eliminate]: 3.49001e-06
        [updatestate_assign_eliminate]: 2.89999e-06
        [updatestate_loads_eliminate]: 2.52001e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 6.27001e-06
        [updatestate_useless_node_eliminater]: 6.23e-06
        [accelerated_algorithm]: 5.81e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 5.63002e-06
        [merge_forward]: 3.11001e-06
        [cell_reuse_recompute_pass]: 1.43002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.424e-05
        [j_node_and_user_rematch]: 1.023e-05
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 8.95999e-06
        [inline_without_move]: 6.06e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 6.64999e-06
        [cse]: 1.448e-05
        [replace_applicator]: 6.13998e-06
[py_interpret_to_execute_after_opt_a]: 1.166e-05
[rewriter_after_opt_a]: 4.028e-05
[convert_after_rewriter]: 8e-06
[order_py_execute_after_rewriter]: 5.84e-06
[mutable_eliminate]: 0.00058041
[jit_opt_b]: 5.49e-05, [1]
    [Cycle 1]: 4.85e-05, [2]
        [frontend_op_eliminate]: 1.942e-05
        [inline_after_opt_a]: 1.802e-05
[cconv]: 2.248e-05
[loop_unroll]: 0.00043005
[jit_opt_after_cconv]: 0.00015117, [1]
    [Cycle 1]: 0.00014474, [11]
        [c_1]: 2.594e-05
        [parameter_eliminate]: 2.53e-06
        [updatestate_depend_eliminate]: 5.99e-06
        [updatestate_assign_eliminate]: 2.84999e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [cse]: 2.175e-05
        [call_graph_tuple_transform]: 1.966e-05
        [tuple_list_get_item_eliminator]: 6.07001e-06
        [none_parameter_eliminate]: 1.58002e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 6.60997e-06
[remove_dup_value]: 1.589e-05
[partial_unused_args_eliminate]: 2.17001e-06
[environ_conv]: 4.85001e-06
[add_recomputation]: 5.053e-05
[cse_after_recomputation]: 3.246e-05, [1]
    [Cycle 1]: 2.685e-05, [1]
        [cse]: 2.051e-05
[auto_monad_reorder]: 1.894e-05
[get_jit_bprop_graph]: 1.87001e-06
[rewriter_after_jit_bprop_graph]: 4.90001e-06
[opt_after_jit_grad]: 0.00048384
[symbol_engine_optimizer]: 8.191e-05, [1]
    [Cycle 1]: 7.515e-05, [6]
        [build]: 4.16001e-06
        [elim_shapecalc]: 9.90002e-06
        [elim_not_effective]: 1.497e-05
        [opt_reshape]: 6.91999e-06
        [fold_const_symbol]: 1.03e-05
        [renormalize]: 4.89992e-07
[validate]: 3.514e-05
[backend_pass]: 1.02998e-06
[task_emit]: 0.00704488
[execute]: 7.18e-06
Sums
    bootstrap                            :     0.000206s :  0.71%
    type_inference                       :     0.012016s : 41.25%
    event_method                         :     0.000059s :  0.20%
    auto_monad                           :     0.000106s :  0.36%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000027s :  0.09%
    rewriter_before_opt_a                :     0.000129s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000168s :  0.58%
    jit_opt_a.loop_unroll                :     0.000087s :  0.30%
    jit_opt_a.a_1                        :     0.002084s :  7.15%
    jit_opt_a.with_stream_mark           :     0.000046s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000026s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.08%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000054s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000043s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001552s :  5.33%
    jit_opt_a.replace_old_param          :     0.000073s :  0.25%
    jit_opt_a.inline_without_move        :     0.000064s :  0.22%
    jit_opt_a.renormalize                :     0.002882s :  9.89%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000067s :  0.23%
    jit_opt_a.cse                        :     0.000182s :  0.62%
    jit_opt_a.replace_applicator         :     0.000081s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000040s :  0.14%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                 
Time group info:
------[substitution.]   0.000270    23
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000002s :      2: substitution.fold_const_symbol
 2.14% :     0.000006s :      3: substitution.graph_param_transform
83.80% :     0.000226s :      4: substitution.inline
 1.45% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.64% :     0.000004s :      4: substitution.remove_not_recompute_node
 4.52% :     0.000012s :      2: substitution.replace_old_param
 5.11% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005620     2
82.02% :     0.004609s :      1: type_inference.infer
17.98% :     0.001010s :      1: type_inference.specialize
------[replace.]   0.000084     6
66.90% :     0.000056s :      4: replace.inline
33.10% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000236     6
94.68% :     0.000224s :      4: match.inline
 5.32% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.54% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.93% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.18% :     0.000001s :     10: predicate.addn_zero_filter
 2.36% :     0.000003s :     10: predicate.arithmetic_simplify
 1.56% :     0.000002s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.16% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.07% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.06% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.30% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.14% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.71% :     0.000007s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.76% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.90% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 1.80% :     0.000002s :      3: predicate.mutable_eliminate
 0.57% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     15: predicate.partial_eliminate
 1.08% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     10: predicate.reduce_eliminate
 1.61% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.93% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.24% :     0.000002s :     10: predicate.reshape_eliminate
 1.10% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.08% :     0.000001s :      3:   :     0.000580s :  1.99%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000022s :  0.08%
    loop_unroll                          :     0.000430s :  1.48%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000016s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000051s :  0.17%
    cse_after_recomputation.cse          :     0.000021s :  0.07%
    auto_monad_reorder                   :     0.000019s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000484s :  1.66%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007045s : 24.18%
    execute                              :     0.000007s :  0.02%

 predicate.row_tensor_eliminate
 1.11% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.55% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.00% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.73% :     0.000010s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.35% :     0.000002s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.11% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.62% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000778    12
48.01% :     0.000373s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.32% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.67% :     0.000355s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024968    76
 0.21% :     0.000053s :      1: add_recomputation
 0.30% :     0.000074s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.46% :     0.000365s :      1: bootstrap
 0.13% :     0.000033s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.04% :     0.000011s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
27.43% :     0.006849s :      1: jit_opt_a
 0.63% :     0.000158s :      1: jit_opt_after_cconv
 0.24% :     0.000059s :      1: jit_opt_b
 1.86% :     0.000464s :      1: loop_unroll
 3.32% :     0.000828s :      1: mutable_eliminate
 3.83% :     0.000957s :     26: opt.transform.jit_opt_a
 0.20% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000029s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.22% :     0.000554s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000045s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.36% :     0.000589s :      1: renormalize.infer
 1.80% :     0.000449s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000041s :      1: rewriter_after_opt_a
 0.31% :     0.000077s :      1: rewriter_before_opt_a
 0.35% :     0.000087s :      1: symbol_engine_optimizer
28.72% :     0.007171s :      1: task_emit
22.85% :     0.005706s :      1: type_inference
 0.26% :     0.000065s :      1: validate


Time group info:
------[substitution.]   0.000493   113
 0.51% :     0.000003s :      3: substitution.elim_not_effective
 0.33% :     0.000002s :      3: substitution.fold_const_symbol
 1.05% :     0.000005s :      4: substitution.graph_param_transform
64.20% :     0.000316s :     14: substitution.inline
 3.32% :     0.000016s :      2: substitution.inline_without_move
 1.73% :     0.000009s :     14: substitution.j_node_and_user_rematch
 1.98% :     0.000010s :      7: substitution.minmaximum_grad
 1.24% :     0.000006s :      5: substitution.partial_eliminate
 2.16% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.11% :     0.000025s :     10: substitution.replace_applicator
 1.57% :     0.000008s :      7: substitution.replace_old_param
 0.45% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.64% :     0.000008s :      1: substitution.switch_simplify
 4.11% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.72% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 7.88% :     0.000039s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011946     2
89.77% :     0.010723s :      1: type_inference.infer
10.23% :     0.001222s :      1: type_inference.specialize
------[replace.]   0.000184    22
53.33% :     0.000098s :     14: replace.inline
18.93% :     0.000035s :      1: replace.switch_simplify
27.74% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000337    22
91.66% :     0.000308s :     14: match.inline
 2.23% :     0.000007s :      1: match.switch_simplify
 6.11% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000380  2634
 1.56% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.52% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.37% :     0.000005s :     44: predicate.addn_check_dump
 1.62% :     0.000006s :     44: predicate.addn_zero_filter
 2.02% :     0.000008s :     44: predicate.arithmetic_simplify
 1.48% :     0.000006s :     44: predicate.cast_eliminate
 0.16% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.37% :     0.000005s :     44: predicate.compare_switch_simplify
 1.52% :     0.000006s :     44: predicate.depend_value_elim
 1.39% :     0.000005s :     44: predicate.dict_get_item_const_eliminator
 1.46% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.41% :     0.000005s :     44: predicate.dict_set_item_eliminator
 0.31% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.14% :     0.000001s :      4: predicate.elim_not_effective
 0.22% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.50% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.39% :     0.000005s :     44: predicate.environ_get_add_eliminate
 1.40% :     0.000005s :     44: predicate.environ_get_depend_swap
 1.42% :     0.000005s :     44: predicate.environ_get_eliminate
 1.41% :     0.000005s :     44: predicate.environ_get_set_eliminate
 0.11% :     0.000000s :      4: predicate.fold_const_symbol
 0.83% :     0.000003s :     19: predicate.get_grad_eliminate
 0.11% :     0.000000s :      4: predicate.graph_param_transform
 4.48% :     0.000017s :     73: predicate.inline
 1.74% :     0.000007s :     40: predicate.inline_without_move
 0.34% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.84% :     0.000003s :     19: predicate.less_batch_normalization
 1.80% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.97% :     0.000008s :     55: predicate.load_eliminater
 0.55% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.62% :     0.000014s :     91: predicate.loop_unroll_before_grad
 1.72% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.37% :     0.000005s :     44: predicate.merge_addn
 1.48% :     0.000006s :     44: predicate.minmaximum_grad
 0.58% :     0.000002s :      4: predicate.mutable_eliminate
 0.18% :     0.000001s :      4: predicate.opt_reshape
 2.30% :     0.000009s :     55: predicate.partial_eliminate
 1.59% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.89% :     0.000007s :     44: predicate.reduce_eliminate
 1.83% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.40% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.44% :     0.000009s :     97: predicate.replace_applicator
 0.91% :     0.000003s :     40: predicate.replace_old_param
 0.11% :     0.000000s :      4: predicate.reset_defer_inline
 1.62% :     0.000006s :     44: predicate.reshape_eliminate
 1.52% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.29% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.55% :     0.000006s :     44: predicate.same_eliminate
 0.48% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.38% :     0.000001s :      8: predicate.special_op_eliminate
 0.89% :     0.000003s :     19: predicate.specialize_transform
 1.76% :     0.000007s :     44: predicate.split_environ_get_set_with_tuple_value
 1.59% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.18% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 3.00% :     0.000011s :     65: predicate.switch_defer_inline
 2.71% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.82% :     0.000026s :    162: predicate.switch_simplify
 1.49% :     0.000006s :     44: predicate.tile_eliminate
 1.46% :     0.000006s :     44: predicate.transpose_eliminate
 1.80% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.84% :     0.000007s :     44: predicate.tuple_list_get_item_depend_reorder
 3.15% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.80% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 2.00% :     0.000008s :     55: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000007s :     44: predicate.value_based_eliminate
 0.14% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.23% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001603    32
61.06% :     0.000979s :     14: func_graph_cloner_run.FuncGraphClonerGraph
38.94% :     0.000624s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.039008    91
 0.14% :     0.000054s :      1: add_recomputation
 0.28% :     0.000111s :      1: auto_monad
 0.06% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.57% :     0.000222s :      1: bootstrap
 0.06% :     0.000025s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000035s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.16% :     0.000064s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
29.21% :     0.011393s :      1: jit_opt_a
 0.39% :     0.000154s :      1: jit_opt_after_cconv
 0.15% :     0.000058s :      1: jit_opt_b
 1.12% :     0.000438s :      1: loop_unroll
 1.51% :     0.000589s :      1: mutable_eliminate
 7.03% :     0.002742s :     39: opt.transform.jit_opt_a
 0.14% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000031s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.32% :     0.000515s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000030s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 3.92% :     0.001528s :      2: renormalize.infer
 3.44% :     0.001341s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000043s :      1: rewriter_after_opt_a
 0.34% :     0.000133s :      1: rewriter_before_opt_a
 0.22% :     0.000085s :      1: symbol_engine_optimizer
18.08% :     0.007054s :      1: task_emit
30.84% :     0.012030s :      1: type_inference
 0.15% :     0.000058s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.435.214 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dd27d30> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.435.752 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0137776, [30]
[bootstrap]: 0.00033799
[type_inference]: 0.00508392
[event_method]: 1.965e-05
[auto_monad]: 7.04e-05
[graph_reusing]: 4.124e-05
[pre_auto_parallel]: 3.09001e-06
[py_interpret_to_execute]: 4.067e-05
[rewriter_before_opt_a]: 7.764e-05
[expand_dump_flag]: 4.23999e-06
[jit_opt_a]: 0.0058668, [2]
    [Cycle 1]: 0.00165156, [27]
        [switch_simplify]: 6.551e-05
        [loop_unroll]: 3.281e-05
        [a_1]: 0.00053715
        [with_stream_mark]: 1.965e-05
        [recompute_prepare]: 7.87998e-06
        [updatestate_depend_eliminate]: 4.12e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.64001e-06
        [parameter_eliminate]: 2.11998e-06
        [specialize_transform]: 5.85002e-06
        [updatestate_useless_node_eliminater]: 5.14998e-06
        [accelerated_algorithm]: 5.56998e-06
        [meta_shard_fg_expand]: 2.56998e-06
        [get_grad_eliminate_]: 5.37001e-06
        [merge_forward]: 4.21001e-06
        [cell_reuse_recompute_pass]: 1.98002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.495e-05
        [j_node_and_user_rematch]: 9.36e-06
        [meta_fg_expand]: 2.32001e-06
        [replace_old_param]: 9.52999e-06
        [inline_without_move]: 5.33002e-06
        [renormalize]: 0.00066378
        [add_forward_monad_depend]: 5.51e-06
        [auto_monad_grad]: 2.51e-06
        [auto_monad_eliminator]: 1.452e-05
        [cse]: 2.948e-05
        [replace_applicator]: 1.241e-05
    [Cycle 2]: 0.00033036, [27]
        [switch_simplify]: 5.84e-06
        [loop_unroll]: 4.99e-06
        [a_1]: 9.506e-05
        [with_stream_mark]: 9.05999e-06
        [recompute_prepare]: 5.27999e-06
        [updatestate_depend_eliminate]: 2.88e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 5.02e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 4.94e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.52998e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 1.82999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.311e-05
        [j_node_and_user_rematch]: 7.92e-06
        [meta_fg_expand]: 1.76e-06
        [replace_old_param]: 8.02e-06
        [inline_without_move]: 4.77998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.41002e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.17e-06
        [cse]: 1.074e-05
        [replace_applicator]: 5.29e-06
[py_interpret_to_execute_after_opt_a]: 1.047e-05
[rewriter_after_opt_a]: 3.528e-05
[convert_after_rewriter]: 7.44002e-06
[order_py_execute_after_rewriter]: 4.87e-06
[mutable_eliminate]: 0.00061595
[jit_opt_b]: 5.298e-05, [1]
    [Cycle 1]: 4.578e-05, [2]
        [frontend_op_eliminate]: 1.924e-05
        [inline_after_opt_a]: 1.486e-05
[cconv]: 2.434e-05
[loop_unroll]: 0.00041762
[jit_opt_after_cconv]: 0.00014058, [1]
    [Cycle 1]: 0.00013392, [11]
        [c_1]: 2.082e-05
        [parameter_eliminate]: 2.75997e-06
        [updatestate_depend_eliminate]: 5.25999e-06
        [updatestate_assign_eliminate]: 2.36998e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [cse]: 1.897e-05
        [call_graph_tuple_transform]: 1.782e-05
        [tuple_list_get_item_eliminator]: 5.80002e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 6.10016e-07
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 1.315e-05
[partial_unused_args_eliminate]: 2.42001e-06
[environ_conv]: 5.54e-06
[add_recomputation]: 4.79e-05
[cse_after_recomputation]: 2.1e-05, [1]
    [Cycle 1]: 1.564e-05, [1]
        [cse]: 9.55001e-06
[auto_monad_reorder]: 1.531e-05
[get_jit_bprop_graph]: 1.78002e-06
[rewriter_after_jit_bprop_graph]: 5.32001e-06
[opt_after_jit_grad]: 0.00047899
[symbol_engine_optimizer]: 0.00010884, [1]
    [Cycle 1]: 0.00010208, [6]
        [build]: 3.28e-06
        [elim_shapecalc]: 7.73001e-06
        [elim_not_effective]: 1.231e-05
        [opt_reshape]: 6.07001e-06
        [fold_const_symbol]: 8.84e-06
        [renormalize]: 7.59988e-07
[validate]: 3.654e-05
Sums
    bootstrap                            :     0.000338s :  3.65%
    type_inference                       :     0.005084s : 54.93%
    event_method                         :     0.000020s :  0.21%
    auto_monad                           :     0.000070s :  0.76%
    graph_reusing                        :     0.000041s :  0.45%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000041s :  0.44%
    rewriter_before_opt_a                :     0.000078s :  0.84%
    expand_dump_flag                     :     0.000004s :  0.05%
    jit_opt_a.switch_simplify            :     0.000071s :  0.77%
    jit_opt_a.loop_unroll                :     0.000038s :  0.41%
    jit_opt_a.a_1                        :     0.000632s :  6.83%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000664s :  7.17%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.21%
    jit_opt_a.cse                        :     0.000040s :  0.43%
    jit_opt_a.replace_applicator         :     0.000018s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.11%
    rewriter_after_opt_a                 :     0.000035s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000616s :  6.65%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000024s :  0.26%
    loop_unroll                          :     0.000418s :  4.51%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000013s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000048s :  0.52%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000479s :  5.18%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000037s :  0.39%


Time group info:
------[substitution.]   0.000203    23
 0.90% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.72% :     0.000006s :      3: substitution.graph_param_transform
83.57% :     0.000169s :      4: substitution.inline
 1.72% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.16% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.25% :     0.000005s :      2: substitution.replace_old_param
 6.04% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005009     2
82.12% :     0.004114s :      1: type_inference.infer
17.88% :     0.000895s :      1: type_inference.specialize
------[replace.]   0.000071     6
68.30% :     0.000049s :      4: replace.inline
31.70% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000178     6
93.73% :     0.000167s :      4: match.inline
 6.27% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.30% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.20% :     0.000001s :     10: predicate.addn_check_dump
 1.48% :     0.000002s :     10: predicate.addn_zero_filter
 2.67% :     0.000003s :     10: predicate.arithmetic_simplify
 1.28% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.42% :     0.000002s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.44% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      6: predicate.less_batch_normalization
 2.13% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 1.61% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.28% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.53% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.62% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.90% :     0.000002s :     10: predicate.reshape_eliminate
 1.33% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.13% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.17% :     0.000010s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.67% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000639    12
54.12% :     0.000346s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.27% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.61% :     0.000266s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015352    72
 0.33% :     0.000051s :      1: add_recomputation
 0.48% :     0.000074s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.33% :     0.000358s :      1: bootstrap
 0.18% :     0.000027s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000025s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.30% :     0.000046s :      1: graph_reusing
38.24% :     0.005871s :      1: jit_opt_a
 0.93% :     0.000143s :      1: jit_opt_after_cconv
 0.36% :     0.000056s :      1: jit_opt_b
 2.78% :     0.000426s :      1: loop_unroll
 4.07% :     0.000625s :      1: mutable_eliminate
 5.58% :     0.000856s :     26: opt.transform.jit_opt_a
 0.30% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000028s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.18% :     0.000488s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.04% :     0.000006s :      1: pre_auto_parallel
 0.28% :     0.000043s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000015s :      1: remove_dup_value
 2.17% :     0.000334s :      1: renormalize.infer
 2.10% :     0.000322s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000039s :      1: rewriter_after_opt_a
 0.53% :     0.000081s :      1: rewriter_before_opt_a
 0.73% :     0.000111s :      1: symbol_engine_optimizer
33.24% :     0.005103s :      1: type_inference


TotalTime = 0.021818, [33]
[bootstrap]: 0.00035099
[type_inference]: 0.00482909
[event_method]: 1.824e-05
[auto_monad]: 6.649e-05
[graph_reusing]: 5.32001e-06
[pre_auto_parallel]: 2.96001e-06
[py_interpret_to_execute]: 4.016e-05
[rewriter_before_opt_a]: 7.203e-05
[expand_dump_flag]: 3.71999e-06
[jit_opt_a]: 0.00652926, [2]
    [Cycle 1]: 0.00225702, [27]
        [switch_simplify]: 7.882e-05
        [loop_unroll]: 3.057e-05
        [a_1]: 0.00062789
        [with_stream_mark]: 2.222e-05
        [recompute_prepare]: 8.52e-06
        [updatestate_depend_eliminate]: 3.71999e-06
        [updatestate_assign_eliminate]: 3.25998e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 2.13002e-06
        [specialize_transform]: 6.81001e-06
        [updatestate_useless_node_eliminater]: 5.44e-06
        [accelerated_algorithm]: 6.12999e-06
        [meta_shard_fg_expand]: 2.26e-06
        [get_grad_eliminate_]: 5.24e-06
        [merge_forward]: 3.59002e-06
        [cell_reuse_recompute_pass]: 1.71998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.607e-05
        [j_node_and_user_rematch]: 1.043e-05
        [meta_fg_expand]: 2.46998e-06
        [replace_old_param]: 9.07999e-06
        [inline_without_move]: 5.55001e-06
        [renormalize]: 0.00113207
        [add_forward_monad_depend]: 7.41001e-06
        [auto_monad_grad]: 3.09001e-06
        [auto_monad_eliminator]: 1.662e-05
        [cse]: 2.342e-05
        [replace_applicator]: 1.907e-05
    [Cycle 2]: 0.00037782, [27]
        [switch_simplify]: 6.04999e-06
        [loop_unroll]: 5.45001e-06
        [a_1]: 0.00010202
        [with_stream_mark]: 1.323e-05
        [recompute_prepare]: 6.36998e-06
        [updatestate_depend_eliminate]: 3.39001e-06
        [updatestate_assign_eliminate]: 2.69999e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 5.99999e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 6.34999e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 4.94998e-06
        [merge_forward]: 2.86999e-06
        [cell_reuse_recompute_pass]: 3.4e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.739e-05
        [j_node_and_user_rematch]: 8.55001e-06
        [meta_fg_expand]: 2.11e-06
        [replace_old_param]: 9.52001e-06
        [inline_without_move]: 5.09e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.66e-06
        [auto_monad_grad]: 1.12e-06
        [auto_monad_eliminator]: 5.72001e-06
        [cse]: 1.144e-05
        [replace_applicator]: 5.54e-06
[py_interpret_to_execute_after_opt_a]: 1.513e-05
[rewriter_after_opt_a]: 3.9e-05
[convert_after_rewriter]: 7.45e-06
[order_py_execute_after_rewriter]: 4.70001e-06
[mutable_eliminate]: 0.00078134
[jit_opt_b]: 6.633e-05, [1]
    [Cycle 1]: 5.617e-05, [2]
        [frontend_op_eliminate]: 2.164e-05
        [inline_after_opt_a]: 2.005e-05
[cconv]: 3.455e-05
[loop_unroll]: 0.00053708
[jit_opt_after_cconv]: 0.00018995, [1]
    [Cycle 1]: 0.00018172, [11]
        [c_1]: 2.602e-05
        [parameter_eliminate]: 5.00001e-06
        [updatestate_depend_eliminate]: 9.85002e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [cse]: 3.337e-05
        [call_graph_tuple_transform]: 2.606e-05
        [tuple_list_get_item_eliminator]: 6.75002e-06
        [none_parameter_eliminate]: 1.86998e-06
        [renormalize]: 1.44e-06
        [switch_simplify]: 7.29001e-06
[remove_dup_value]: 1.856e-05
[partial_unused_args_eliminate]: 2.68003e-06
[environ_conv]: 6.33e-06
[add_recomputation]: 5.994e-05
[cse_after_recomputation]: 3.205e-05, [1]
    [Cycle 1]: 2.439e-05, [1]
        [cse]: 1.373e-05
[auto_monad_reorder]: 1.772e-05
[get_jit_bprop_graph]: 2.12999e-06
[rewriter_after_jit_bprop_graph]: 9.95002e-06
[opt_after_jit_grad]: 0.00071372
[symbol_engine_optimizer]: 9.831e-05, [1]
    [Cycle 1]: 8.931e-05, [6]
        [build]: 6.16e-06
        [elim_shapecalc]: 1.043e-05
        [elim_not_effective]: 1.848e-05
        [opt_reshape]: 7.95998e-06
        [fold_const_symbol]: 1.162e-05
        [renormalize]: 1.14e-06
[validate]: 4.874e-05
[backend_pass]: 1.10001e-06
[task_emit]: 0.00694314
[execute]: 9.04e-06
Sums
    bootstrap                            :     0.000351s :  2.04%
    type_inference                       :     0.004829s : 28.12%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000066s :  0.39%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.23%
    rewriter_before_opt_a                :     0.000072s :  0.42%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000085s :  0.49%
    jit_opt_a.loop_unroll                :     0.000036s :  0.21%
    jit_opt_a.a_1                        :     0.000730s :  4.25%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001132s :  6.59%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.13%
    jit_opt_a.cse                        :     0.000035s :  0.20%
    jit_opt_a.replace_applicator         :     0.000025s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.09%
    rewriter_after_opt_a                 :     0.000039s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000781s :  4.55%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.12%
    cconv                                :     0.000035s :  0.20%
    loop_unroll                          :     0.000537s :  3.13%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000033s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.04%
    remove_dup_value                     :     0.000019s :  0.11%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000060s :  0.35%
    cse_after_recomputation.cse          :     0.000014s :  0.08%
    auto_monad_reorder                   :     0.000018s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.06%
    opt_after_jit_grad                   :     0.000714s :  4.16%
    symbol_engine_optimizer.build        :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.11%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000049s :  0.28%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006943s : 40.43%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000260    23
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.95% :     0.000002s :      2: substitution.fold_const_symbol
 2.61% :     0.000007s :      3: substitution.graph_param_transform
84.54% :     0.000220s :      4: substitution.inline
 1.28% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.16% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.19% :     0.000006s :      2: substitution.replace_old_param
 5.39% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004761     2
80.18% :     0.003817s :      1: type_inference.infer
19.82% :     0.000943s :      1: type_inference.specialize
------[replace.]   0.000089     6
64.06% :     0.000057s :      4: replace.inline
35.94% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000230     6
94.43% :     0.000217s :      4: match.inline
 5.57% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.87% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.91% :     0.000001s :     10: predicate.addn_check_dump
 1.20% :     0.000002s :     10: predicate.addn_zero_filter
 3.05% :     0.000004s :     10: predicate.arithmetic_simplify
 1.29% :     0.000002s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.92% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 0.97% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.04% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.04% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.20% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.03% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.20% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.89% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000002s :     10: predicate.environ_get_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.79% :     0.000001s :      6: predicate.get_grad_eliminate
 0.47% :     0.000001s :      3: predicate.graph_param_transform
 5.20% :     0.000007s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.36% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.57% :     0.000002s :     15: predicate.load_eliminater
 1.93% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.95% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.60% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 1.43% :     0.000002s :     10: predicate.minmaximum_grad
 2.05% :     0.000003s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.99% :     0.000003s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.09% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.08% :     0.000001s :     10: predicate.same_eliminate
 0.81% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.06% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.74% :     0.000004s :     16: predicate.switch_defer_inline
 1.85% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.10% :     0.000011s :     53: predicate.switch_simplify
 1.12% :     0.000002s :     10: predicate.tile_eliminate
 1.15% :     0.000002s :     10: predicate.transpose_eliminate
 1.20% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.69% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.79% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.30% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.54% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.43% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000696    12
45.36% :     0.000316s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.26% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.39% :     0.000337s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024068    76
 0.27% :     0.000065s :      1: add_recomputation
 0.29% :     0.000070s :      1: auto_monad
 0.09% :     0.000021s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.52% :     0.000365s :      1: bootstrap
 0.16% :     0.000038s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000035s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
27.14% :     0.006533s :      1: jit_opt_a
 0.80% :     0.000194s :      1: jit_opt_after_cconv
 0.29% :     0.000070s :      1: jit_opt_b
 2.28% :     0.000549s :      1: loop_unroll
 3.31% :     0.000796s :      1: mutable_eliminate
 4.10% :     0.000986s :     26: opt.transform.jit_opt_a
 0.26% :     0.000062s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000033s :      4: opt.transform.jit_opt_b
 0.08% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000044s :      4: opt.transform.symbol_engine_opt
 3.05% :     0.000733s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000043s :      1: py_interpret_to_execute
 0.08% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000022s :      1: remove_dup_value
 2.67% :     0.000643s :      1: renormalize.infer
 1.98% :     0.000476s :      1: renormalize.specialize
 0.05% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000042s :      1: rewriter_after_opt_a
 0.31% :     0.000075s :      1: rewriter_before_opt_a
 0.42% :     0.000101s :      1: symbol_engine_optimizer
28.93% :     0.006963s :      1: task_emit
20.12% :     0.004842s :      1: type_inference
 0.31% :     0.000074s :      1: validate


TotalTime = 0.0325113, [33]
[bootstrap]: 0.00018981
[type_inference]: 0.0118924
[event_method]: 5.736e-05
[auto_monad]: 0.00010593
[graph_reusing]: 7.95e-06
[pre_auto_parallel]: 2.69999e-06
[py_interpret_to_execute]: 2.655e-05
[rewriter_before_opt_a]: 0.00010469
[expand_dump_flag]: 3.04999e-06
[jit_opt_a]: 0.0112428, [3]
    [Cycle 1]: 0.00565865, [27]
        [switch_simplify]: 0.00012359
        [loop_unroll]: 4.983e-05
        [a_1]: 0.00096425
        [with_stream_mark]: 2.369e-05
        [recompute_prepare]: 2.048e-05
        [updatestate_depend_eliminate]: 8.32e-06
        [updatestate_assign_eliminate]: 6.44999e-06
        [updatestate_loads_eliminate]: 6.39001e-06
        [parameter_eliminate]: 2.89999e-06
        [specialize_transform]: 1.452e-05
        [updatestate_useless_node_eliminater]: 1.397e-05
        [accelerated_algorithm]: 1.374e-05
        [meta_shard_fg_expand]: 4.06001e-06
        [get_grad_eliminate_]: 1.3e-05
        [merge_forward]: 7.73999e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.853e-05
        [j_node_and_user_rematch]: 2.614e-05
        [meta_fg_expand]: 0.00156885
        [replace_old_param]: 5.172e-05
        [inline_without_move]: 4.978e-05
        [renormalize]: 0.00218537
        [add_forward_monad_depend]: 8.99998e-06
        [auto_monad_grad]: 4.33999e-06
        [auto_monad_eliminator]: 4.615e-05
        [cse]: 0.00014681
        [replace_applicator]: 6.245e-05
    [Cycle 2]: 0.00194444, [27]
        [switch_simplify]: 3.627e-05
        [loop_unroll]: 4.68e-05
        [a_1]: 0.0010165
        [with_stream_mark]: 1.038e-05
        [recompute_prepare]: 8.10999e-06
        [updatestate_depend_eliminate]: 3.76001e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 1.09998e-06
        [specialize_transform]: 7.09001e-06
        [updatestate_useless_node_eliminater]: 6.01e-06
        [accelerated_algorithm]: 6.07999e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 5.64e-06
        [merge_forward]: 2.98e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.443e-05
        [j_node_and_user_rematch]: 1.039e-05
        [meta_fg_expand]: 6.394e-05
        [replace_old_param]: 1.032e-05
        [inline_without_move]: 6.06e-06
        [renormalize]: 0.00046514
        [add_forward_monad_depend]: 4.23999e-06
        [auto_monad_grad]: 1.09998e-06
        [auto_monad_eliminator]: 1.101e-05
        [cse]: 1.963e-05
        [replace_applicator]: 1.204e-05
    [Cycle 3]: 0.00036709, [27]
        [switch_simplify]: 6.90998e-06
        [loop_unroll]: 5.91003e-06
        [a_1]: 0.00011774
        [with_stream_mark]: 1.079e-05
        [recompute_prepare]: 6.04001e-06
        [updatestate_depend_eliminate]: 3.51999e-06
        [updatestate_assign_eliminate]: 2.63998e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 6.17001e-06
        [updatestate_useless_node_eliminater]: 5.79999e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 5.62001e-06
        [merge_forward]: 2.91e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.453e-05
        [j_node_and_user_rematch]: 9.49999e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 8.42e-06
        [inline_without_move]: 5.73002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.06997e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 6.46e-06
        [cse]: 1.369e-05
        [replace_applicator]: 6.12999e-06
[py_interpret_to_execute_after_opt_a]: 1.033e-05
[rewriter_after_opt_a]: 3.6e-05
[convert_after_rewriter]: 7.83001e-06
[order_py_execute_after_rewriter]: 6.14999e-06
[mutable_eliminate]: 0.00047981
[jit_opt_b]: 5.48e-05, [1]
    [Cycle 1]: 4.871e-05, [2]
        [frontend_op_eliminate]: 1.846e-05
        [inline_after_opt_a]: 1.842e-05
[cconv]: 2.11e-05
[loop_unroll]: 0.00042816
[jit_opt_after_cconv]: 0.00015109, [1]
    [Cycle 1]: 0.0001452, [11]
        [c_1]: 2.521e-05
        [parameter_eliminate]: 2.32999e-06
        [updatestate_depend_eliminate]: 5.92999e-06
        [updatestate_assign_eliminate]: 2.80997e-06
        [updatestate_loads_eliminate]: 3.08e-06
        [cse]: 2.217e-05
        [call_graph_tuple_transform]: 2.01e-05
        [tuple_list_get_item_eliminator]: 6.32001e-06
        [none_parameter_eliminate]: 1.33002e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 6.06e-06
[remove_dup_value]: 1.476e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 5.20001e-06
[add_recomputation]: 8.438e-05
[cse_after_recomputation]: 2.776e-05, [1]
    [Cycle 1]: 2.202e-05, [1]
        [cse]: 1.612e-05
[auto_monad_reorder]: 1.841e-05
[get_jit_bprop_graph]: 1.42e-06
[rewriter_after_jit_bprop_graph]: 4.55999e-06
[opt_after_jit_grad]: 0.00053236
[symbol_engine_optimizer]: 8.1e-05, [1]
    [Cycle 1]: 7.482e-05, [6]
        [build]: 3.53e-06
        [elim_shapecalc]: 9.96e-06
        [elim_not_effective]: 1.528e-05
        [opt_reshape]: 7.08998e-06
        [fold_const_symbol]: 1.073e-05
        [renormalize]: 3.50003e-07
[validate]: 3.482e-05
[backend_pass]: 9.49978e-07
[task_emit]: 0.0066856
[execute]: 6.79999e-06
Sums
    bootstrap                            :     0.000190s :  0.67%
    type_inference                       :     0.011892s : 41.80%
    event_method                         :     0.000057s :  0.20%
    auto_monad                           :     0.000106s :  0.37%
    graph_reusing                        :     0.000008s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000027s :  0.09%
    rewriter_before_opt_a                :     0.000105s :  0.37%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000167s :  0.59%
    jit_opt_a.loop_unroll                :     0.000103s :  0.36%
    jit_opt_a.a_1                        :     0.002098s :  7.38%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000035s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000028s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000026s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.09%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000057s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000046s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001635s :  5.75%
    jit_opt_a.replace_old_param          :     0.000070s :  0.25%
    jit_opt_a.inline_without_move        :     0.000062s :  0.22%
    jit_opt_a.renormalize                :     0.002651s :  9.32%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000006s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000064s :  0.22%
    jit_opt_a.cse                        :     0.000180s :  0.63%
    jit_opt_a.replace_applicator         :     0.000081s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.04%
    rewriter_after_opt_a                 :     0.000036s :  0.13%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000480s :  1.69%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.06%
    cconv                                :     0.000021s :  0.07%
    loop_unroll                          :     0.000428s :  1.50%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.02%
    remove_dup_value                     :     0.000015s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000084s :  0.30%
    cse_after_recomputation.cse          :     0.000016s :  0.06%
    auto_monad_reorder                   :     0.000018s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000532s :  1.87%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006686s : 23.50%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000466   113
 0.49% :     0.000002s :      3: substitution.elim_not_effective
 0.42% :     0.000002s :      3: substitution.fold_const_symbol
 1.17% :     0.000005s :      4: substitution.graph_param_transform
62.38% :     0.000291s :     14: substitution.inline
 3.24% :     0.000015s :      2: substitution.inline_without_move
 1.92% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.13% :     0.000010s :      7: substitution.minmaximum_grad
 1.30% :     0.000006s :      5: substitution.partial_eliminate
 2.50% :     0.000012s :     14: substitution.remove_not_recompute_node
 5.29% :     0.000025s :     10: substitution.replace_applicator
 1.46% :     0.000007s :      7: substitution.replace_old_param
 0.68% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.60% :     0.000007s :      1: substitution.switch_simplify
 4.17% :     0.000019s :      7: substitution.tuple_list_convert_item_index_to_positive
 3.04% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 8.19% :     0.000038s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011825     2
89.62% :     0.010597s :      1: type_inference.infer
10.38% :     0.001227s :      1: type_inference.specialize
------[replace.]   0.000177    22
54.97% :     0.000097s :     14: replace.inline
16.14% :     0.000029s :      1: replace.switch_simplify
28.89% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000309    22
91.59% :     0.000283s :     14: match.inline
 2.19% :     0.000007s :      1: match.switch_simplify
 6.22% :     0.000019s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000390  2634
 1.54% :     0.000006s :     44: predicate.accumulaten_eliminater
 0.44% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.41% :     0.000006s :     44: predicate.addn_check_dump
 1.79% :     0.000007s :     44: predicate.addn_zero_filter
 2.26% :     0.000009s :     44: predicate.arithmetic_simplify
 1.48% :     0.000006s :     44: predicate.cast_eliminate
 0.15% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.39% :     0.000005s :     44: predicate.compare_switch_simplify
 1.58% :     0.000006s :     44: predicate.depend_value_elim
 1.60% :     0.000006s :     44: predicate.dict_get_item_const_eliminator
 1.41% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.50% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.40% :     0.000002s :      4: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      4: predicate.elim_not_effective
 0.23% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.46% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.42% :     0.000006s :     44: predicate.environ_get_add_eliminate
 1.43% :     0.000006s :     44: predicate.environ_get_depend_swap
 1.54% :     0.000006s :     44: predicate.environ_get_eliminate
 1.41% :     0.000006s :     44: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      4: predicate.fold_const_symbol
 0.86% :     0.000003s :     19: predicate.get_grad_eliminate
 0.09% :     0.000000s :      4: predicate.graph_param_transform
 4.35% :     0.000017s :     73: predicate.inline
 1.74% :     0.000007s :     40: predicate.inline_without_move
 0.36% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.91% :     0.000004s :     19: predicate.less_batch_normalization
 1.73% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 2.00% :     0.000008s :     55: predicate.load_eliminater
 0.42% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.56% :     0.000014s :     91: predicate.loop_unroll_before_grad
 1.85% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.48% :     0.000006s :     44: predicate.merge_addn
 1.50% :     0.000006s :     44: predicate.minmaximum_grad
 0.51% :     0.000002s :      4: predicate.mutable_eliminate
 0.21% :     0.000001s :      4: predicate.opt_reshape
 2.30% :     0.000009s :     55: predicate.partial_eliminate
 1.54% :     0.000006s :     44: predicate.print_const_string_wrapper
 1.86% :     0.000007s :     44: predicate.reduce_eliminate
 1.75% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.41% :     0.000009s :     97: predicate.replace_applicator
 0.88% :     0.000003s :     40: predicate.replace_old_param
 0.13% :     0.000001s :      4: predicate.reset_defer_inline
 1.70% :     0.000007s :     44: predicate.reshape_eliminate
 1.58% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.23% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.51% :     0.000006s :     44: predicate.same_eliminate
 0.50% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.34% :     0.000001s :      8: predicate.special_op_eliminate
 0.87% :     0.000003s :     19: predicate.specialize_transform
 1.68% :     0.000007s :     44: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000006s :     44: predicate.stack_unstack_eliminate
 0.16% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.98% :     0.000012s :     65: predicate.switch_defer_inline
 2.55% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.81% :     0.000027s :    162: predicate.switch_simplify
 1.47% :     0.000006s :     44: predicate.tile_eliminate
 1.49% :     0.000006s :     44: predicate.transpose_eliminate
 1.81% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.70% :     0.000007s :     44: predicate.tuple_list_get_item_depend_reorder
 3.05% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.94% :     0.000008s :     55: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000011s :     74: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000007s :     44: predicate.value_based_eliminate
 0.16% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.21% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001454    32
61.53% :     0.000895s :     14: func_graph_cloner_run.FuncGraphClonerGraph
38.47% :     0.000559s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038071    91
 0.23% :     0.000088s :      1: add_recomputation
 0.29% :     0.000111s :      1: auto_monad
 0.06% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.53% :     0.000202s :      1: bootstrap
 0.06% :     0.000024s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000030s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.16% :     0.000062s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
29.54% :     0.011245s :      1: jit_opt_a
 0.40% :     0.000154s :      1: jit_opt_after_cconv
 0.15% :     0.000057s :      1: jit_opt_b
 1.14% :     0.000436s :      1: loop_unroll
 1.28% :     0.000488s :      1: mutable_eliminate
 7.30% :     0.002779s :     39: opt.transform.jit_opt_a
 0.14% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000030s :      4: opt.transform.jit_opt_b
 0.04% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000040s :      4: opt.transform.symbol_engine_opt
 1.42% :     0.000541s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000029s :      1: py_interpret_to_execute
 0.03% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000017s :      1: remove_dup_value
 3.67% :     0.001398s :      2: renormalize.infer
 3.25% :     0.001239s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000039s :      1: rewriter_after_opt_a
 0.28% :     0.000108s :      1: rewriter_before_opt_a
 0.22% :     0.000084s :      1: symbol_engine_optimizer
17.59% :     0.006695s :      1: task_emit
31.27% :     0.011904s :      1: type_inference
 0.15% :     0.000056s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.474.775 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc410d0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.475.283 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.01326, [30]
[bootstrap]: 0.00032407
[type_inference]: 0.00457023
[event_method]: 1.81e-05
[auto_monad]: 6.677e-05
[graph_reusing]: 6.25002e-06
[pre_auto_parallel]: 2.57001e-06
[py_interpret_to_execute]: 3.602e-05
[rewriter_before_opt_a]: 7.29e-05
[expand_dump_flag]: 2.99999e-06
[jit_opt_a]: 0.00559665, [2]
    [Cycle 1]: 0.00162965, [27]
        [switch_simplify]: 5.98e-05
        [loop_unroll]: 3.016e-05
        [a_1]: 0.00054098
        [with_stream_mark]: 1.819e-05
        [recompute_prepare]: 6.72002e-06
        [updatestate_depend_eliminate]: 3.80998e-06
        [updatestate_assign_eliminate]: 2.74999e-06
        [updatestate_loads_eliminate]: 3.38e-06
        [parameter_eliminate]: 2.24001e-06
        [specialize_transform]: 6.16e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 1.97999e-06
        [get_grad_eliminate_]: 5.61998e-06
        [merge_forward]: 3.60998e-06
        [cell_reuse_recompute_pass]: 1.49998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.419e-05
        [j_node_and_user_rematch]: 9.39998e-06
        [meta_fg_expand]: 2.84999e-06
        [replace_old_param]: 9.08002e-06
        [inline_without_move]: 5.22e-06
        [renormalize]: 0.00065122
        [add_forward_monad_depend]: 4.99e-06
        [auto_monad_grad]: 2.59999e-06
        [auto_monad_eliminator]: 1.337e-05
        [cse]: 2.86e-05
        [replace_applicator]: 1.29e-05
    [Cycle 2]: 0.00034191, [27]
        [switch_simplify]: 5.62001e-06
        [loop_unroll]: 4.81002e-06
        [a_1]: 9.19e-05
        [with_stream_mark]: 1.194e-05
        [recompute_prepare]: 5.32999e-06
        [updatestate_depend_eliminate]: 2.68e-06
        [updatestate_assign_eliminate]: 2.18002e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [parameter_eliminate]: 1.19998e-06
        [specialize_transform]: 5.25999e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 5.19998e-06
        [meta_shard_fg_expand]: 1.16002e-06
        [get_grad_eliminate_]: 4.46002e-06
        [merge_forward]: 2.95002e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.39e-05
        [j_node_and_user_rematch]: 8.28001e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 8.02e-06
        [inline_without_move]: 5.04998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.48002e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 5.81998e-06
        [cse]: 1.28e-05
        [replace_applicator]: 8.38001e-06
[py_interpret_to_execute_after_opt_a]: 1.147e-05
[rewriter_after_opt_a]: 3.654e-05
[convert_after_rewriter]: 6.91001e-06
[order_py_execute_after_rewriter]: 5.91e-06
[mutable_eliminate]: 0.00059968
[jit_opt_b]: 5.477e-05, [1]
    [Cycle 1]: 4.672e-05, [2]
        [frontend_op_eliminate]: 1.828e-05
        [inline_after_opt_a]: 1.54e-05
[cconv]: 2.834e-05
[loop_unroll]: 0.00052789
[jit_opt_after_cconv]: 0.00020652, [1]
    [Cycle 1]: 0.00019816, [11]
        [c_1]: 2.381e-05
        [parameter_eliminate]: 4.85999e-06
        [updatestate_depend_eliminate]: 8.02998e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 3.20998e-06
        [cse]: 2.85e-05
        [call_graph_tuple_transform]: 2.064e-05
        [tuple_list_get_item_eliminator]: 6.26e-06
        [none_parameter_eliminate]: 1.72999e-06
        [renormalize]: 1.90019e-07
        [switch_simplify]: 6.32001e-06
[remove_dup_value]: 1.563e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 6.02001e-06
[add_recomputation]: 5.359e-05
[cse_after_recomputation]: 2.592e-05, [1]
    [Cycle 1]: 1.906e-05, [1]
        [cse]: 1.069e-05
[auto_monad_reorder]: 1.579e-05
[get_jit_bprop_graph]: 2.12001e-06
[rewriter_after_jit_bprop_graph]: 8.64998e-06
[opt_after_jit_grad]: 0.00063073
[symbol_engine_optimizer]: 8.464e-05, [1]
    [Cycle 1]: 7.593e-05, [6]
        [build]: 4.45e-06
        [elim_shapecalc]: 8.87999e-06
        [elim_not_effective]: 1.506e-05
        [opt_reshape]: 6.80998e-06
        [fold_const_symbol]: 9.53002e-06
        [renormalize]: 1.14e-06
[validate]: 3.921e-05
Sums
    bootstrap                            :     0.000324s :  3.62%
    type_inference                       :     0.004570s : 51.01%
    event_method                         :     0.000018s :  0.20%
    auto_monad                           :     0.000067s :  0.75%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.40%
    rewriter_before_opt_a                :     0.000073s :  0.81%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000065s :  0.73%
    jit_opt_a.loop_unroll                :     0.000035s :  0.39%
    jit_opt_a.a_1                        :     0.000633s :  7.06%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.34%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000651s :  7.27%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.21%
    jit_opt_a.cse                        :     0.000041s :  0.46%
    jit_opt_a.replace_applicator         :     0.000021s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.13%
    rewriter_after_opt_a                 :     0.000037s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.07%
    mutable_eliminate                    :     0.000600s :  6.69%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000028s :  0.32%
    loop_unroll                          :     0.000528s :  5.89%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.09%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.cse              :     0.000029s :  0.32%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000016s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000054s :  0.60%
    cse_after_recomputation.cse          :     0.000011s :  0.12%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.10%
    opt_after_jit_grad                   :     0.000631s :  7.04%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.44%


Time group info:
------[substitution.]   0.000205    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.77% :     0.000002s :      2: substitution.fold_const_symbol
 2.69% :     0.000006s :      3: substitution.graph_param_transform
83.04% :     0.000171s :      4: substitution.inline
 1.58% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.22% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.30% :     0.000005s :      2: substitution.replace_old_param
 6.52% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004500     2
80.71% :     0.003632s :      1: type_inference.infer
19.29% :     0.000868s :      1: type_inference.specialize
------[replace.]   0.000071     6
67.79% :     0.000048s :      4: replace.inline
32.21% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000180     6
93.20% :     0.000168s :      4: match.inline
 6.80% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.26% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.78% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 1.85% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.50% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.10% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.07% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.39% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.81% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000000s :      3: predicate.graph_param_transform
 5.80% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.90% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.61% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.89% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.95% :     0.000002s :      3: predicate.mutable_eliminate
 0.60% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     15: predicate.partial_eliminate
 1.04% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.80% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 1.11% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.51% :     0.000002s :      6: predicate.special_op_eliminate
 1.27% :     0.000002s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.26% :     0.000009s :     53: predicate.switch_simplify
 1.09% :     0.000001s :     10: predicate.tile_eliminate
 1.76% :     0.000002s :     10: predicate.transpose_eliminate
 1.26% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.93% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.35% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000596    12
51.93% :     0.000309s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.14% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.93% :     0.000256s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014832    72
 0.39% :     0.000058s :      1: add_recomputation
 0.47% :     0.000070s :      1: auto_monad
 0.13% :     0.000019s :      1: auto_monad_reorder
 2.31% :     0.000343s :      1: bootstrap
 0.21% :     0.000031s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.19% :     0.000029s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.15% :     0.000022s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
37.76% :     0.005600s :      1: jit_opt_a
 1.41% :     0.000210s :      1: jit_opt_after_cconv
 0.39% :     0.000058s :      1: jit_opt_b
 3.64% :     0.000539s :      1: loop_unroll
 4.13% :     0.000612s :      1: mutable_eliminate
 5.75% :     0.000853s :     26: opt.transform.jit_opt_a
 0.35% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000027s :      4: opt.transform.jit_opt_b
 0.11% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.19% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000036s :      4: opt.transform.symbol_engine_opt
 4.33% :     0.000642s :      1: opt_after_jit_grad
 0.06% :     0.000009s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000039s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000018s :      1: remove_dup_value
 2.24% :     0.000332s :      1: renormalize.infer
 2.10% :     0.000311s :      1: renormalize.specialize
 0.07% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000039s :      1: rewriter_after_opt_a
 0.51% :     0.000076s :      1: rewriter_before_opt_a
 0.59% :     0.000087s :      1: symbol_engine_optimizer
30.93% :     0.004587s :      1: type_inference


TotalTime = 0.0209808, [33]
[bootstrap]: 0.00035301
[type_inference]: 0.0048053
[event_method]: 1.786e-05
[auto_monad]: 5.785e-05
[graph_reusing]: 5.24e-06
[pre_auto_parallel]: 2.26e-06
[py_interpret_to_execute]: 4.104e-05
[rewriter_before_opt_a]: 7.07e-05
[expand_dump_flag]: 2.76e-06
[jit_opt_a]: 0.00624117, [2]
    [Cycle 1]: 0.00224843, [27]
        [switch_simplify]: 5.771e-05
        [loop_unroll]: 3.267e-05
        [a_1]: 0.00069426
        [with_stream_mark]: 2.084e-05
        [recompute_prepare]: 1.036e-05
        [updatestate_depend_eliminate]: 3.93001e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.53002e-06
        [specialize_transform]: 7.93999e-06
        [updatestate_useless_node_eliminater]: 5.90002e-06
        [accelerated_algorithm]: 6.49001e-06
        [meta_shard_fg_expand]: 2.32999e-06
        [get_grad_eliminate_]: 5.89999e-06
        [merge_forward]: 4.50001e-06
        [cell_reuse_recompute_pass]: 1.91e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.601e-05
        [j_node_and_user_rematch]: 1.07e-05
        [meta_fg_expand]: 3.3e-06
        [replace_old_param]: 1.149e-05
        [inline_without_move]: 5.32999e-06
        [renormalize]: 0.00106348
        [add_forward_monad_depend]: 9.04e-06
        [auto_monad_grad]: 2.46998e-06
        [auto_monad_eliminator]: 2.045e-05
        [cse]: 2.512e-05
        [replace_applicator]: 2.067e-05
    [Cycle 2]: 0.00036643, [27]
        [switch_simplify]: 6.61999e-06
        [loop_unroll]: 5.60001e-06
        [a_1]: 0.00010187
        [with_stream_mark]: 1.584e-05
        [recompute_prepare]: 5.39e-06
        [updatestate_depend_eliminate]: 2.61999e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.73998e-06
        [parameter_eliminate]: 1.55999e-06
        [specialize_transform]: 5.10999e-06
        [updatestate_useless_node_eliminater]: 5.79999e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 2.06998e-06
        [get_grad_eliminate_]: 5.11002e-06
        [merge_forward]: 3.56001e-06
        [cell_reuse_recompute_pass]: 2.69001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.67e-05
        [j_node_and_user_rematch]: 8.72e-06
        [meta_fg_expand]: 2.04999e-06
        [replace_old_param]: 8.44002e-06
        [inline_without_move]: 4.97e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.82999e-06
        [auto_monad_grad]: 1.40001e-06
        [auto_monad_eliminator]: 5.77001e-06
        [cse]: 1.285e-05
        [replace_applicator]: 6.93e-06
[py_interpret_to_execute_after_opt_a]: 1.519e-05
[rewriter_after_opt_a]: 3.608e-05
[convert_after_rewriter]: 7.61001e-06
[order_py_execute_after_rewriter]: 5.35001e-06
[mutable_eliminate]: 0.00080947
[jit_opt_b]: 6.116e-05, [1]
    [Cycle 1]: 5.082e-05, [2]
        [frontend_op_eliminate]: 1.98e-05
        [inline_after_opt_a]: 1.722e-05
[cconv]: 2.506e-05
[loop_unroll]: 0.00057836
[jit_opt_after_cconv]: 0.00017675, [1]
    [Cycle 1]: 0.00016909, [11]
        [c_1]: 2.468e-05
        [parameter_eliminate]: 4.55999e-06
        [updatestate_depend_eliminate]: 7.35e-06
        [updatestate_assign_eliminate]: 3.52002e-06
        [updatestate_loads_eliminate]: 2.91999e-06
        [cse]: 3.051e-05
        [call_graph_tuple_transform]: 2.454e-05
        [tuple_list_get_item_eliminator]: 6.27001e-06
        [none_parameter_eliminate]: 1.70001e-06
        [renormalize]: 8.80013e-07
        [switch_simplify]: 6.43e-06
[remove_dup_value]: 1.369e-05
[partial_unused_args_eliminate]: 2.33002e-06
[environ_conv]: 5.49e-06
[add_recomputation]: 8.87e-05
[cse_after_recomputation]: 2.709e-05, [1]
    [Cycle 1]: 2.115e-05, [1]
        [cse]: 1.275e-05
[auto_monad_reorder]: 1.641e-05
[get_jit_bprop_graph]: 2.98998e-06
[rewriter_after_jit_bprop_graph]: 9.69999e-06
[opt_after_jit_grad]: 0.0005801
[symbol_engine_optimizer]: 8.516e-05, [1]
    [Cycle 1]: 7.715e-05, [6]
        [build]: 4.48001e-06
        [elim_shapecalc]: 8.62998e-06
        [elim_not_effective]: 1.629e-05
        [opt_reshape]: 6.23e-06
        [fold_const_symbol]: 9.07001e-06
        [renormalize]: 6.50005e-07
[validate]: 4.466e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00654806
[execute]: 7.62998e-06
Sums
    bootstrap                            :     0.000353s :  2.12%
    type_inference                       :     0.004805s : 28.85%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000058s :  0.35%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000041s :  0.25%
    rewriter_before_opt_a                :     0.000071s :  0.42%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.39%
    jit_opt_a.loop_unroll                :     0.000038s :  0.23%
    jit_opt_a.a_1                        :     0.000796s :  4.78%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000020s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001064s :  6.39%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.16%
    jit_opt_a.cse                        :     0.000038s :  0.23%
    jit_opt_a.replace_applicator         :     0.000028s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.09%
    rewriter_after_opt_a                 :     0.000036s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000809s :  4.86%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.10%
    cconv                                :     0.000025s :  0.15%
    loop_unroll                          :     0.000578s :  3.47%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000031s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000089s :  0.53%
    cse_after_recomputation.cse          :     0.000013s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.06%
    opt_after_jit_grad                   :     0.000580s :  3.48%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006548s : 39.32%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000255    23
 1.03% :     0.000003s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.69% :     0.000007s :      3: substitution.graph_param_transform
85.25% :     0.000217s :      4: substitution.inline
 1.38% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.13% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.04% :     0.000005s :      2: substitution.replace_old_param
 5.00% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004736     2
80.32% :     0.003803s :      1: type_inference.infer
19.68% :     0.000932s :      1: type_inference.specialize
------[replace.]   0.000106     6
58.98% :     0.000063s :      4: replace.inline
41.02% :     0.000044s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000226     6
94.93% :     0.000215s :      4: match.inline
 5.07% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   691
 1.09% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.48% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.89% :     0.000001s :     10: predicate.addn_check_dump
 1.64% :     0.000002s :     10: predicate.addn_zero_filter
 2.93% :     0.000004s :     10: predicate.arithmetic_simplify
 1.18% :     0.000002s :     10: predicate.cast_eliminate
 0.58% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.92% :     0.000003s :     10: predicate.depend_value_elim
 1.33% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.07% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.97% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.35% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000002s :     10: predicate.environ_get_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.15% :     0.000002s :      6: predicate.get_grad_eliminate
 0.45% :     0.000001s :      3: predicate.graph_param_transform
 5.02% :     0.000007s :     22: predicate.inline
 0.74% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.57% :     0.000002s :      6: predicate.less_batch_normalization
 1.35% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 2.51% :     0.000003s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 1.82% :     0.000003s :     15: predicate.partial_eliminate
 0.99% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.71% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.48% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.08% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.98% :     0.000001s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.29% :     0.000002s :      6: predicate.special_op_eliminate
 0.86% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.76% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.78% :     0.000004s :     16: predicate.switch_defer_inline
 1.89% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.17% :     0.000010s :     53: predicate.switch_simplify
 1.59% :     0.000002s :     10: predicate.tile_eliminate
 1.19% :     0.000002s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.64% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.88% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.92% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.53% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.42% :     0.000002s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000692    12
43.17% :     0.000299s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.77% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.06% :     0.000346s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023208    76
 0.40% :     0.000093s :      1: add_recomputation
 0.26% :     0.000061s :      1: auto_monad
 0.09% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.59% :     0.000370s :      1: bootstrap
 0.12% :     0.000028s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.13% :     0.000030s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000006s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
26.91% :     0.006245s :      1: jit_opt_a
 0.78% :     0.000181s :      1: jit_opt_after_cconv
 0.28% :     0.000064s :      1: jit_opt_b
 2.54% :     0.000589s :      1: loop_unroll
 3.55% :     0.000824s :      1: mutable_eliminate
 4.47% :     0.001037s :     26: opt.transform.jit_opt_a
 0.25% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000029s :      4: opt.transform.jit_opt_b
 0.07% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000022s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.55% :     0.000592s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000044s :      1: py_interpret_to_execute
 0.08% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.33% :     0.000541s :      1: renormalize.infer
 2.21% :     0.000512s :      1: renormalize.specialize
 0.05% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000040s :      1: rewriter_after_opt_a
 0.32% :     0.000074s :      1: rewriter_before_opt_a
 0.38% :     0.000089s :      1: symbol_engine_optimizer
28.29% :     0.006565s :      1: task_emit
20.77% :     0.004819s :      1: type_inference
 0.28% :     0.000066s :      1: validate


TotalTime = 0.03264, [33]
[bootstrap]: 0.0001863
[type_inference]: 0.0117938
[event_method]: 5.733e-05
[auto_monad]: 0.00010512
[graph_reusing]: 7.46999e-06
[pre_auto_parallel]: 2.98e-06
[py_interpret_to_execute]: 2.785e-05
[rewriter_before_opt_a]: 0.00010504
[expand_dump_flag]: 3.56999e-06
[jit_opt_a]: 0.0113566, [3]
    [Cycle 1]: 0.00570481, [27]
        [switch_simplify]: 0.00011976
        [loop_unroll]: 4.693e-05
        [a_1]: 0.00088067
        [with_stream_mark]: 2.256e-05
        [recompute_prepare]: 1.803e-05
        [updatestate_depend_eliminate]: 7.56001e-06
        [updatestate_assign_eliminate]: 6.26e-06
        [updatestate_loads_eliminate]: 5.92999e-06
        [parameter_eliminate]: 3.09001e-06
        [specialize_transform]: 1.31e-05
        [updatestate_useless_node_eliminater]: 1.257e-05
        [accelerated_algorithm]: 1.263e-05
        [meta_shard_fg_expand]: 3.63e-06
        [get_grad_eliminate_]: 1.203e-05
        [merge_forward]: 8.43999e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.677e-05
        [j_node_and_user_rematch]: 2.317e-05
        [meta_fg_expand]: 0.00150286
        [replace_old_param]: 5.59e-05
        [inline_without_move]: 5.436e-05
        [renormalize]: 0.00235893
        [add_forward_monad_depend]: 9.49e-06
        [auto_monad_grad]: 5.52001e-06
        [auto_monad_eliminator]: 4.999e-05
        [cse]: 0.00015894
        [replace_applicator]: 7.301e-05
    [Cycle 2]: 0.00209664, [27]
        [switch_simplify]: 4.161e-05
        [loop_unroll]: 3.892e-05
        [a_1]: 0.00112545
        [with_stream_mark]: 1.276e-05
        [recompute_prepare]: 8.75001e-06
        [updatestate_depend_eliminate]: 4.1e-06
        [updatestate_assign_eliminate]: 3.24001e-06
        [updatestate_loads_eliminate]: 3.01001e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 7.4e-06
        [updatestate_useless_node_eliminater]: 6.86001e-06
        [accelerated_algorithm]: 5.98998e-06
        [meta_shard_fg_expand]: 1.82001e-06
        [get_grad_eliminate_]: 5.76e-06
        [merge_forward]: 2.99999e-06
        [cell_reuse_recompute_pass]: 8.39995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.494e-05
        [j_node_and_user_rematch]: 1.052e-05
        [meta_fg_expand]: 7.253e-05
        [replace_old_param]: 1.112e-05
        [inline_without_move]: 6.56e-06
        [renormalize]: 0.00050385
        [add_forward_monad_depend]: 4.53001e-06
        [auto_monad_grad]: 1.05999e-06
        [auto_monad_eliminator]: 1.188e-05
        [cse]: 2.061e-05
        [replace_applicator]: 1.395e-05
    [Cycle 3]: 0.0003928, [27]
        [switch_simplify]: 7.11001e-06
        [loop_unroll]: 6.61999e-06
        [a_1]: 0.0001276
        [with_stream_mark]: 9.94999e-06
        [recompute_prepare]: 6.71e-06
        [updatestate_depend_eliminate]: 3.58999e-06
        [updatestate_assign_eliminate]: 3.06999e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [parameter_eliminate]: 1.26002e-06
        [specialize_transform]: 6.66e-06
        [updatestate_useless_node_eliminater]: 6.48e-06
        [accelerated_algorithm]: 6.29999e-06
        [meta_shard_fg_expand]: 1.45001e-06
        [get_grad_eliminate_]: 6.26998e-06
        [merge_forward]: 3.03998e-06
        [cell_reuse_recompute_pass]: 1.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.481e-05
        [j_node_and_user_rematch]: 9.88002e-06
        [meta_fg_expand]: 2.16e-06
        [replace_old_param]: 9.41e-06
        [inline_without_move]: 6.62002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.48002e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 7.1e-06
        [cse]: 1.462e-05
        [replace_applicator]: 5.97001e-06
[py_interpret_to_execute_after_opt_a]: 1.154e-05
[rewriter_after_opt_a]: 4.156e-05
[convert_after_rewriter]: 7.90998e-06
[order_py_execute_after_rewriter]: 6.21e-06
[mutable_eliminate]: 0.00059386
[jit_opt_b]: 5.9e-05, [1]
    [Cycle 1]: 5.15e-05, [2]
        [frontend_op_eliminate]: 2.047e-05
        [inline_after_opt_a]: 1.92e-05
[cconv]: 2.344e-05
[loop_unroll]: 0.00045682
[jit_opt_after_cconv]: 0.00016589, [1]
    [Cycle 1]: 0.00015902, [11]
        [c_1]: 2.855e-05
        [parameter_eliminate]: 2.71e-06
        [updatestate_depend_eliminate]: 6.91001e-06
        [updatestate_assign_eliminate]: 3.35e-06
        [updatestate_loads_eliminate]: 2.83e-06
        [cse]: 2.355e-05
        [call_graph_tuple_transform]: 2.119e-05
        [tuple_list_get_item_eliminator]: 6.41e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 7e-06
[remove_dup_value]: 1.675e-05
[partial_unused_args_eliminate]: 2.66999e-06
[environ_conv]: 5.59998e-06
[add_recomputation]: 5.392e-05
[cse_after_recomputation]: 3.595e-05, [1]
    [Cycle 1]: 2.988e-05, [1]
        [cse]: 2.285e-05
[auto_monad_reorder]: 2.034e-05
[get_jit_bprop_graph]: 2.09e-06
[rewriter_after_jit_bprop_graph]: 5.94999e-06
[opt_after_jit_grad]: 0.00051172
[symbol_engine_optimizer]: 8.142e-05, [1]
    [Cycle 1]: 7.446e-05, [6]
        [build]: 3.53e-06
        [elim_shapecalc]: 8.57e-06
        [elim_not_effective]: 1.509e-05
        [opt_reshape]: 6.71999e-06
        [fold_const_symbol]: 1.141e-05
        [renormalize]: 4.89992e-07
[validate]: 4.969e-05
[backend_pass]: 1.07998e-06
[task_emit]: 0.00663979
[execute]: 7.29001e-06
Sums
    bootstrap                            :     0.000186s :  0.65%
    type_inference                       :     0.011794s : 41.14%
    event_method                         :     0.000057s :  0.20%
    auto_monad                           :     0.000105s :  0.37%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.10%
    rewriter_before_opt_a                :     0.000105s :  0.37%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000168s :  0.59%
    jit_opt_a.loop_unroll                :     0.000092s :  0.32%
    jit_opt_a.a_1                        :     0.002134s :  7.44%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.02%
    jit_opt_a.specialize_transform       :     0.000027s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000025s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.08%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000057s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000044s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001578s :  5.50%
    jit_opt_a.replace_old_param          :     0.000076s :  0.27%
    jit_opt_a.inline_without_move        :     0.000068s :  0.24%
    jit_opt_a.renormalize                :     0.002863s :  9.99%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000069s :  0.24%
    jit_opt_a.cse                        :     0.000194s :  0.68%
    jit_opt_a.replace_applicator         :     0.000093s :  0.32%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000042s :  0.14%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000594s :  2.07%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.07%
    cconv                                :     0.000023s :  0.08%
    loop_unroll                          :     0.000457s :  1.59%
    jit_opt_after_cconv.c_1              :     0.000029s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000017s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000054s :  0.19%
    cse_after_recomputation.cse          :     0.000023s :  0.08%
    auto_monad_reorder                   :     0.000020s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000512s :  1.78%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000050s :  0.17%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006640s : 23.16%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000525   113
 0.44% :     0.000002s :      3: substitution.elim_not_effective
 0.31% :     0.000002s :      3: substitution.fold_const_symbol
 1.20% :     0.000006s :      4: substitution.graph_param_transform
65.47% :     0.000344s :     14: substitution.inline
 3.23% :     0.000017s :      2: substitution.inline_without_move
 1.62% :     0.000009s :     14: substitution.j_node_and_user_rematch
 2.00% :     0.000010s :      7: substitution.minmaximum_grad
 1.20% :     0.000006s :      5: substitution.partial_eliminate
 2.14% :     0.000011s :     14: substitution.remove_not_recompute_node
 5.25% :     0.000028s :     10: substitution.replace_applicator
 1.37% :     0.000007s :      7: substitution.replace_old_param
 0.40% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.42% :     0.000007s :      1: substitution.switch_simplify
 3.94% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.56% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 7.45% :     0.000039s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011725     2
89.99% :     0.010551s :      1: type_inference.infer
10.01% :     0.001174s :      1: type_inference.specialize
------[replace.]   0.000183    22
54.11% :     0.000099s :     14: replace.inline
16.60% :     0.000030s :      1: replace.switch_simplify
29.29% :     0.000054s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000362    22
92.65% :     0.000336s :     14: match.inline
 1.88% :     0.000007s :      1: match.switch_simplify
 5.47% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000396  2634
 1.72% :     0.000007s :     44: predicate.accumulaten_eliminater
 0.44% :     0.000002s :      4: predicate.ad_related_special_op_eliminate
 1.46% :     0.000006s :     44: predicate.addn_check_dump
 1.71% :     0.000007s :     44: predicate.addn_zero_filter
 1.95% :     0.000008s :     44: predicate.arithmetic_simplify
 1.48% :     0.000006s :     44: predicate.cast_eliminate
 0.20% :     0.000001s :      4: predicate.check_bprop_eliminate
 1.44% :     0.000006s :     44: predicate.compare_switch_simplify
 1.70% :     0.000007s :     44: predicate.depend_value_elim
 1.41% :     0.000006s :     44: predicate.dict_get_item_const_eliminator
 1.51% :     0.000006s :     44: predicate.dict_get_item_eliminator
 1.46% :     0.000006s :     44: predicate.dict_set_item_eliminator
 0.30% :     0.000001s :      4: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      4: predicate.elim_not_effective
 0.19% :     0.000001s :      4: predicate.elim_shapecalc_of_broadcastargs
 1.44% :     0.000006s :     44: predicate.environ_add_const_eliminate
 1.42% :     0.000006s :     44: predicate.environ_get_add_eliminate
 1.40% :     0.000006s :     44: predicate.environ_get_depend_swap
 1.52% :     0.000006s :     44: predicate.environ_get_eliminate
 1.38% :     0.000005s :     44: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      4: predicate.fold_const_symbol
 0.80% :     0.000003s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      4: predicate.graph_param_transform
 4.38% :     0.000017s :     73: predicate.inline
 1.70% :     0.000007s :     40: predicate.inline_without_move
 0.35% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.84% :     0.000003s :     19: predicate.less_batch_normalization
 1.72% :     0.000007s :     51: predicate.list_to_tuple_eliminator_
 1.86% :     0.000007s :     55: predicate.load_eliminater
 0.51% :     0.000002s :      4: predicate.loop_unroll_after_grad
 3.63% :     0.000014s :     91: predicate.loop_unroll_before_grad
 1.81% :     0.000007s :     48: predicate.make_slice_get_slice_eliminator
 1.58% :     0.000006s :     44: predicate.merge_addn
 1.45% :     0.000006s :     44: predicate.minmaximum_grad
 0.54% :     0.000002s :      4: predicate.mutable_eliminate
 0.17% :     0.000001s :      4: predicate.opt_reshape
 2.30% :     0.000009s :     55: predicate.partial_eliminate
 1.71% :     0.000007s :     44: predicate.print_const_string_wrapper
 1.84% :     0.000007s :     44: predicate.reduce_eliminate
 1.73% :     0.000007s :     51: predicate.redundant_stop_gradient_eliminater
 0.40% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.44% :     0.000010s :     97: predicate.replace_applicator
 0.93% :     0.000004s :     40: predicate.replace_old_param
 0.09% :     0.000000s :      4: predicate.reset_defer_inline
 1.64% :     0.000007s :     44: predicate.reshape_eliminate
 1.54% :     0.000006s :     44: predicate.row_tensor_add_zeros_like
 0.23% :     0.000001s :      4: predicate.row_tensor_eliminate
 1.67% :     0.000007s :     44: predicate.same_eliminate
 0.50% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.38% :     0.000002s :      8: predicate.special_op_eliminate
 0.83% :     0.000003s :     19: predicate.specialize_transform
 1.66% :     0.000007s :     44: predicate.split_environ_get_set_with_tuple_value
 1.67% :     0.000007s :     44: predicate.stack_unstack_eliminate
 0.17% :     0.000001s :      4: predicate.switch_call_monad_eliminater
 2.99% :     0.000012s :     65: predicate.switch_defer_inline
 2.64% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.74% :     0.000027s :    162: predicate.switch_simplify
 1.52% :     0.000006s :     44: predicate.tile_eliminate
 1.47% :     0.000006s :     44: predicate.transpose_eliminate
 1.83% :     0.000007s :     44: predicate.tuple_list_convert_item_index_to_positive
 1.63% :     0.000006s :     44: predicate.tuple_list_get_item_depend_reorder
 2.98% :     0.000012s :     59: predicate.tuple_list_get_item_eliminator
 1.84% :     0.000007s :     44: predicate.tuple_list_set_item_eliminator
 1.73% :     0.000007s :     51: predicate.tuple_to_list_eliminator_
 1.86% :     0.000007s :     55: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000012s :     74: predicate.updatestate_useless_node_eliminater
 1.89% :     0.000007s :     44: predicate.value_based_eliminate
 0.13% :     0.000001s :      4: predicate.virtual_view_grad_eliminate
 0.26% :     0.000001s :      4: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001541    32
61.29% :     0.000945s :     14: func_graph_cloner_run.FuncGraphClonerGraph
38.71% :     0.000597s :     18: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038459    91
 0.15% :     0.000058s :      1: add_recomputation
 0.29% :     0.000110s :      1: auto_monad
 0.06% :     0.000023s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.51% :     0.000198s :      1: bootstrap
 0.07% :     0.000026s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000039s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.16% :     0.000062s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
29.54% :     0.011359s :      1: jit_opt_a
 0.44% :     0.000169s :      1: jit_opt_after_cconv
 0.16% :     0.000062s :      1: jit_opt_b
 1.21% :     0.000465s :      1: loop_unroll
 1.56% :     0.000602s :      1: mutable_eliminate
 7.33% :     0.002820s :     39: opt.transform.jit_opt_a
 0.15% :     0.000060s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000032s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000039s :      4: opt.transform.symbol_engine_opt
 1.35% :     0.000520s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000030s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 3.84% :     0.001478s :      2: renormalize.infer
 3.56% :     0.001370s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000045s :      1: rewriter_after_opt_a
 0.28% :     0.000108s :      1: rewriter_before_opt_a
 0.22% :     0.000084s :      1: symbol_engine_optimizer
17.29% :     0.006651s :      1: task_emit
30.70% :     0.011806s :      1: type_inference
 0.20% :     0.000075s :      1: validate


TotalTime = 0.0135622, [30]
[bootstrap]: 0.00034283
[type_inference]: 0.00466082
[event_method]: 2.003e-05
[auto_monad]: 6.688e-05
[graph_reusing]: 5.72999e-06
[pre_auto_parallel]: 2.44999e-06
[py_interpret_to_execute]: 3.594e-05
[rewriter_before_opt_a]: 7.533e-05
[expand_dump_flag]: 3.25998e-06
[jit_opt_a]: 0.00594425, [2]
    [Cycle 1]: 0.00181313, [27]
        [switch_simplify]: 6.437e-05
        [loop_unroll]: 3.037e-05
        [a_1]: 0.00057118
        [with_stream_mark]: 2.462e-05
        [recompute_prepare]: 8.07998e-06
        [updatestate_depend_eliminate]: 4.13999e-06
        [updatestate_assign_eliminate]: 3.25e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 5.97999e-06
        [updatestate_useless_node_eliminater]: 2.096e-05
        [accelerated_algorithm]: 6.74001e-06
        [meta_shard_fg_expand]: 2.24999e-06
        [get_grad_eliminate_]: 5.59e-06
        [merge_forward]: 3.68e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.468e-05
        [j_node_and_user_rematch]: 8.84003e-06
        [meta_fg_expand]: 2.25002e-06
        [replace_old_param]: 9.04e-06
        [inline_without_move]: 5.59e-06
        [renormalize]: 0.0007611
        [add_forward_monad_depend]: 6.31998e-06
        [auto_monad_grad]: 2.40002e-06
        [auto_monad_eliminator]: 1.616e-05
        [cse]: 2.985e-05
        [replace_applicator]: 1.438e-05
    [Cycle 2]: 0.00033821, [27]
        [switch_simplify]: 5.79e-06
        [loop_unroll]: 5.09e-06
        [a_1]: 9.154e-05
        [with_stream_mark]: 1.149e-05
        [recompute_prepare]: 5.27001e-06
        [updatestate_depend_eliminate]: 2.74999e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.23002e-06
        [parameter_eliminate]: 1.16997e-06
        [specialize_transform]: 5.29e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 4.99e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.48999e-06
        [merge_forward]: 3.88001e-06
        [cell_reuse_recompute_pass]: 1.99e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.51e-05
        [j_node_and_user_rematch]: 8.16002e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 7.85e-06
        [inline_without_move]: 4.82998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.56998e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 5.67999e-06
        [cse]: 1.271e-05
        [replace_applicator]: 5.07e-06
[py_interpret_to_execute_after_opt_a]: 1.245e-05
[rewriter_after_opt_a]: 3.789e-05
[convert_after_rewriter]: 9.91998e-06
[order_py_execute_after_rewriter]: 5.01002e-06
[mutable_eliminate]: 0.00070696
[jit_opt_b]: 5.4e-05, [1]
    [Cycle 1]: 4.567e-05, [2]
        [frontend_op_eliminate]: 1.805e-05
        [inline_after_opt_a]: 1.569e-05
[cconv]: 3.1e-05
[loop_unroll]: 0.00045612
[jit_opt_after_cconv]: 0.00015568, [1]
    [Cycle 1]: 0.00014817, [11]
        [c_1]: 2.138e-05
        [parameter_eliminate]: 3.86999e-06
        [updatestate_depend_eliminate]: 6.88e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [cse]: 2.404e-05
        [call_graph_tuple_transform]: 2.14e-05
        [tuple_list_get_item_eliminator]: 5.43002e-06
        [none_parameter_eliminate]: 1.63002e-06
        [renormalize]: 7.2e-07
        [switch_simplify]: 5.30001e-06
[remove_dup_value]: 1.484e-05
[partial_unused_args_eliminate]: 2.82002e-06
[environ_conv]: 5.02999e-06
[add_recomputation]: 5.007e-05
[cse_after_recomputation]: 2.239e-05, [1]
    [Cycle 1]: 1.614e-05, [1]
        [cse]: 1.04e-05
[auto_monad_reorder]: 1.574e-05
[get_jit_bprop_graph]: 1.87999e-06
[rewriter_after_jit_bprop_graph]: 6.02999e-06
[opt_after_jit_grad]: 0.00050097
[symbol_engine_optimizer]: 7.823e-05, [1]
    [Cycle 1]: 7.05e-05, [6]
        [build]: 4.45e-06
        [elim_shapecalc]: 8.67998e-06
        [elim_not_effective]: 1.367e-05
        [opt_reshape]: 5.46e-06
        [fold_const_symbol]: 9.00001e-06
        [renormalize]: 7.7e-07
[validate]: 4.073e-05
Sums
    bootstrap                            :     0.000343s :  3.75%
    type_inference                       :     0.004661s : 51.01%
    event_method                         :     0.000020s :  0.22%
    auto_monad                           :     0.000067s :  0.73%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.39%
    rewriter_before_opt_a                :     0.000075s :  0.82%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000070s :  0.77%
    jit_opt_a.loop_unroll                :     0.000035s :  0.39%
    jit_opt_a.a_1                        :     0.000663s :  7.25%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.40%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.28%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.18%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000761s :  8.33%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.24%
    jit_opt_a.cse                        :     0.000043s :  0.47%
    jit_opt_a.replace_applicator         :     0.000019s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.14%
    rewriter_after_opt_a                 :     0.000038s :  0.41%
    convert_after_rewriter               :     0.000010s :  0.11%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000707s :  7.74%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.17%
    cconv                                :     0.000031s :  0.34%
    loop_unroll                          :     0.000456s :  4.99%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000024s :  0.26%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000050s :  0.55%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000501s :  5.48%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.45%


Time group info:
------[substitution.]   0.000237    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.43% :     0.000006s :      3: substitution.graph_param_transform
85.24% :     0.000202s :      4: substitution.inline
 1.37% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.04% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.11% :     0.000005s :      2: substitution.replace_old_param
 5.40% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004588     2
80.95% :     0.003714s :      1: type_inference.infer
19.05% :     0.000874s :      1: type_inference.specialize
------[replace.]   0.000071     6
69.18% :     0.000049s :      4: replace.inline
30.82% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000212     6
94.45% :     0.000200s :      4: match.inline
 5.55% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.32% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.47% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000001s :     10: predicate.addn_zero_filter
 2.81% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.47% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.15% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000000s :      3: predicate.graph_param_transform
 6.41% :     0.000008s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.67% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.65% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.92% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     15: predicate.partial_eliminate
 1.09% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.56% :     0.000001s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.37% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.15% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.12% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.01% :     0.000004s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.77% :     0.000009s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.86% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.47% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.53% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000648    12
50.56% :     0.000327s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.37% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.07% :     0.000272s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015278    72
 0.35% :     0.000053s :      1: add_recomputation
 0.46% :     0.000070s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 2.37% :     0.000362s :      1: bootstrap
 0.22% :     0.000034s :      1: cconv
 0.08% :     0.000013s :      1: convert_after_rewriter
 0.16% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.94% :     0.005948s :      1: jit_opt_a
 1.04% :     0.000158s :      1: jit_opt_after_cconv
 0.37% :     0.000057s :      1: jit_opt_b
 3.05% :     0.000465s :      1: loop_unroll
 4.71% :     0.000719s :      1: mutable_eliminate
 5.91% :     0.000902s :     26: opt.transform.jit_opt_a
 0.33% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000027s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.34% :     0.000511s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000039s :      1: py_interpret_to_execute
 0.10% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000018s :      1: remove_dup_value
 2.71% :     0.000415s :      1: renormalize.infer
 2.21% :     0.000338s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000041s :      1: rewriter_after_opt_a
 0.51% :     0.000078s :      1: rewriter_before_opt_a
 0.53% :     0.000081s :      1: symbol_engine_optimizer
30.63% :     0.004679s :      1: type_inference


TotalTime = 0.0205537, [33]
[bootstrap]: 0.00039255
[type_inference]: 0.00512269
[event_method]: 1.776e-05
[auto_monad]: 6.511e-05
[graph_reusing]: 6.38998e-06
[pre_auto_parallel]: 2.40002e-06
[py_interpret_to_execute]: 3.619e-05
[rewriter_before_opt_a]: 7.303e-05
[expand_dump_flag]: 3.06001e-06
[jit_opt_a]: 0.00576563, [2]
    [Cycle 1]: 0.00171395, [27]
        [switch_simplify]: 5.626e-05
        [loop_unroll]: 3.055e-05
        [a_1]: 0.00055626
        [with_stream_mark]: 1.854e-05
        [recompute_prepare]: 8.17998e-06
        [updatestate_depend_eliminate]: 3.48999e-06
        [updatestate_assign_eliminate]: 3.11999e-06
        [updatestate_loads_eliminate]: 2.55002e-06
        [parameter_eliminate]: 1.88997e-06
        [specialize_transform]: 5.91998e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 2.29001e-06
        [get_grad_eliminate_]: 7.48e-06
        [merge_forward]: 3.59002e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.4e-05
        [j_node_and_user_rematch]: 9.02e-06
        [meta_fg_expand]: 2.36e-06
        [replace_old_param]: 9.99001e-06
        [inline_without_move]: 5.64998e-06
        [renormalize]: 0.00072499
        [add_forward_monad_depend]: 5.44e-06
        [auto_monad_grad]: 2.54999e-06
        [auto_monad_eliminator]: 1.373e-05
        [cse]: 2.688e-05
        [replace_applicator]: 1.323e-05
    [Cycle 2]: 0.00032811, [27]
        [switch_simplify]: 5.98998e-06
        [loop_unroll]: 5.02e-06
        [a_1]: 9.344e-05
        [with_stream_mark]: 9.00001e-06
        [recompute_prepare]: 5.20999e-06
        [updatestate_depend_eliminate]: 2.53998e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 4.70999e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 5.53002e-06
        [meta_shard_fg_expand]: 1.40999e-06
        [get_grad_eliminate_]: 4.51002e-06
        [merge_forward]: 2.70002e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.347e-05
        [j_node_and_user_rematch]: 7.63001e-06
        [meta_fg_expand]: 1.82001e-06
        [replace_old_param]: 1.01e-05
        [inline_without_move]: 4.68001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 5.37001e-06
        [cse]: 1.1e-05
        [replace_applicator]: 4.97999e-06
[py_interpret_to_execute_after_opt_a]: 1.266e-05
[rewriter_after_opt_a]: 3.491e-05
[convert_after_rewriter]: 8.3e-06
[order_py_execute_after_rewriter]: 5.69e-06
[mutable_eliminate]: 0.00053811
[jit_opt_b]: 5.052e-05, [1]
    [Cycle 1]: 4.339e-05, [2]
        [frontend_op_eliminate]: 1.663e-05
        [inline_after_opt_a]: 1.489e-05
[cconv]: 2.513e-05
[loop_unroll]: 0.00041627
[jit_opt_after_cconv]: 0.00014999, [1]
    [Cycle 1]: 0.00014367, [11]
        [c_1]: 2.115e-05
        [parameter_eliminate]: 2.94001e-06
        [updatestate_depend_eliminate]: 5.77001e-06
        [updatestate_assign_eliminate]: 2.36998e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [cse]: 1.877e-05
        [call_graph_tuple_transform]: 2.563e-05
        [tuple_list_get_item_eliminator]: 5.87001e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.30001e-06
[remove_dup_value]: 1.483e-05
[partial_unused_args_eliminate]: 2.17001e-06
[environ_conv]: 5.29e-06
[add_recomputation]: 5.321e-05
[cse_after_recomputation]: 2.11e-05, [1]
    [Cycle 1]: 1.543e-05, [1]
        [cse]: 9.87999e-06
[auto_monad_reorder]: 1.511e-05
[get_jit_bprop_graph]: 2.19001e-06
[rewriter_after_jit_bprop_graph]: 5.10001e-06
[opt_after_jit_grad]: 0.0004557
[symbol_engine_optimizer]: 7.436e-05, [1]
    [Cycle 1]: 6.757e-05, [6]
        [build]: 4.29002e-06
        [elim_shapecalc]: 7.64002e-06
        [elim_not_effective]: 1.252e-05
        [opt_reshape]: 5.64998e-06
        [fold_const_symbol]: 9.21002e-06
        [renormalize]: 3.59985e-07
[validate]: 4.178e-05
[backend_pass]: 8.59989e-07
[task_emit]: 0.00689926
[execute]: 9.07001e-06
Sums
    bootstrap                            :     0.000393s :  2.42%
    type_inference                       :     0.005123s : 31.64%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000065s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000036s :  0.22%
    rewriter_before_opt_a                :     0.000073s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.38%
    jit_opt_a.loop_unroll                :     0.000036s :  0.22%
    jit_opt_a.a_1                        :     0.000650s :  4.01%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000020s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000725s :  4.48%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000038s :  0.23%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000035s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000538s :  3.32%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000025s :  0.16%
    loop_unroll                          :     0.000416s :  2.57%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000053s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000456s :  2.81%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000042s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006899s : 42.61%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000213    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.52% :     0.000005s :      3: substitution.graph_param_transform
84.18% :     0.000179s :      4: substitution.inline
 1.44% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.20% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.84% :     0.000004s :      2: substitution.replace_old_param
 6.31% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005052     2
81.89% :     0.004137s :      1: type_inference.infer
18.11% :     0.000915s :      1: type_inference.specialize
------[replace.]   0.000072     6
67.18% :     0.000048s :      4: replace.inline
32.82% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000189     6
93.48% :     0.000177s :      4: match.inline
 6.52% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.21% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.47% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.18% :     0.000001s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.53% :     0.000002s :     10: predicate.cast_eliminate
 0.59% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.34% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.34% :     0.000000s :      3: predicate.graph_param_transform
 5.34% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.96% :     0.000002s :     15: predicate.load_eliminater
 1.29% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.02% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.38% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.60% :     0.000009s :     53: predicate.switch_simplify
 1.42% :     0.000002s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.53% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.93% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.07% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000676    12
50.06% :     0.000338s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.47% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.47% :     0.000301s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022253    76
 0.25% :     0.000056s :      1: add_recomputation
 0.31% :     0.000068s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.85% :     0.000412s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.06% :     0.000014s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.93% :     0.005770s :      1: jit_opt_a
 0.69% :     0.000153s :      1: jit_opt_after_cconv
 0.24% :     0.000053s :      1: jit_opt_b
 1.91% :     0.000424s :      1: loop_unroll
 2.46% :     0.000547s :      1: mutable_eliminate
 3.90% :     0.000868s :     26: opt.transform.jit_opt_a
 0.25% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.08% :     0.000464s :      1: opt_after_jit_grad
 0.09% :     0.000020s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000039s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.68% :     0.000374s :      1: renormalize.infer
 1.54% :     0.000343s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000038s :      1: rewriter_after_opt_a
 0.34% :     0.000076s :      1: rewriter_before_opt_a
 0.35% :     0.000077s :      1: symbol_engine_optimizer
31.09% :     0.006918s :      1: task_emit
23.09% :     0.005139s :      1: type_inference
 0.27% :     0.000060s :      1: validate


TotalTime = 0.0205414, [33]
[bootstrap]: 0.00031642
[type_inference]: 0.00478526
[event_method]: 1.716e-05
[auto_monad]: 6.786e-05
[graph_reusing]: 5.71998e-06
[pre_auto_parallel]: 2.38002e-06
[py_interpret_to_execute]: 3.91e-05
[rewriter_before_opt_a]: 7.16e-05
[expand_dump_flag]: 3.54002e-06
[jit_opt_a]: 0.00612469, [2]
    [Cycle 1]: 0.0020945, [27]
        [switch_simplify]: 6.518e-05
        [loop_unroll]: 3.17e-05
        [a_1]: 0.00062003
        [with_stream_mark]: 2.075e-05
        [recompute_prepare]: 9.49e-06
        [updatestate_depend_eliminate]: 3.83001e-06
        [updatestate_assign_eliminate]: 2.94001e-06
        [updatestate_loads_eliminate]: 3.2e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 6.60002e-06
        [updatestate_useless_node_eliminater]: 5.44e-06
        [accelerated_algorithm]: 6.07999e-06
        [meta_shard_fg_expand]: 2.55002e-06
        [get_grad_eliminate_]: 5.97001e-06
        [merge_forward]: 4.03001e-06
        [cell_reuse_recompute_pass]: 2.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.598e-05
        [j_node_and_user_rematch]: 1.078e-05
        [meta_fg_expand]: 3.55e-06
        [replace_old_param]: 1.082e-05
        [inline_without_move]: 5.79e-06
        [renormalize]: 0.00099993
        [add_forward_monad_depend]: 8.27e-06
        [auto_monad_grad]: 2.85002e-06
        [auto_monad_eliminator]: 1.496e-05
        [cse]: 2.147e-05
        [replace_applicator]: 1.795e-05
    [Cycle 2]: 0.0003683, [27]
        [switch_simplify]: 6.34001e-06
        [loop_unroll]: 5.53002e-06
        [a_1]: 0.00010493
        [with_stream_mark]: 1.161e-05
        [recompute_prepare]: 6.49001e-06
        [updatestate_depend_eliminate]: 3.11001e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 1.28002e-06
        [specialize_transform]: 5.66998e-06
        [updatestate_useless_node_eliminater]: 4.84998e-06
        [accelerated_algorithm]: 5.99e-06
        [meta_shard_fg_expand]: 1.72999e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 3.41001e-06
        [cell_reuse_recompute_pass]: 2.11998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.524e-05
        [j_node_and_user_rematch]: 8.52e-06
        [meta_fg_expand]: 1.72999e-06
        [replace_old_param]: 8.92999e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 1.09983e-07
        [add_forward_monad_depend]: 1.75001e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 5.42999e-06
        [cse]: 1.154e-05
        [replace_applicator]: 6.02001e-06
[py_interpret_to_execute_after_opt_a]: 1.381e-05
[rewriter_after_opt_a]: 3.675e-05
[convert_after_rewriter]: 8.17e-06
[order_py_execute_after_rewriter]: 5.22e-06
[mutable_eliminate]: 0.0007322
[jit_opt_b]: 0.00013546, [1]
    [Cycle 1]: 0.0001253, [2]
        [frontend_op_eliminate]: 2.081e-05
        [inline_after_opt_a]: 9.028e-05
[cconv]: 2.624e-05
[loop_unroll]: 0.00048613
[jit_opt_after_cconv]: 0.000157, [1]
    [Cycle 1]: 0.00015028, [11]
        [c_1]: 2.285e-05
        [parameter_eliminate]: 3.26999e-06
        [updatestate_depend_eliminate]: 6.51999e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [cse]: 2.217e-05
        [call_graph_tuple_transform]: 2.373e-05
        [tuple_list_get_item_eliminator]: 5.58002e-06
        [none_parameter_eliminate]: 1.69e-06
        [renormalize]: 1.12e-06
        [switch_simplify]: 5.92999e-06
[remove_dup_value]: 1.384e-05
[partial_unused_args_eliminate]: 2.68e-06
[environ_conv]: 5.44e-06
[add_recomputation]: 4.897e-05
[cse_after_recomputation]: 2.132e-05, [1]
    [Cycle 1]: 1.549e-05, [1]
        [cse]: 9.25999e-06
[auto_monad_reorder]: 1.538e-05
[get_jit_bprop_graph]: 2.33998e-06
[rewriter_after_jit_bprop_graph]: 7.37997e-06
[opt_after_jit_grad]: 0.00047292
[symbol_engine_optimizer]: 8.106e-05, [1]
    [Cycle 1]: 7.354e-05, [6]
        [build]: 3.62998e-06
        [elim_shapecalc]: 8.58001e-06
        [elim_not_effective]: 1.498e-05
        [opt_reshape]: 6.65002e-06
        [fold_const_symbol]: 9.51e-06
        [renormalize]: 5.19998e-07
[validate]: 3.935e-05
[backend_pass]: 1.15999e-06
[task_emit]: 0.00657338
[execute]: 6.81001e-06
Sums
    bootstrap                            :     0.000316s :  1.95%
    type_inference                       :     0.004785s : 29.52%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000068s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.24%
    rewriter_before_opt_a                :     0.000072s :  0.44%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000072s :  0.44%
    jit_opt_a.loop_unroll                :     0.000037s :  0.23%
    jit_opt_a.a_1                        :     0.000725s :  4.47%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000006s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000020s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.001000s :  6.17%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000033s :  0.20%
    jit_opt_a.replace_applicator         :     0.000024s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.09%
    rewriter_after_opt_a                 :     0.000037s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000732s :  4.52%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000090s :  0.56%
    cconv                                :     0.000026s :  0.16%
    loop_unroll                          :     0.000486s :  3.00%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.05%
    opt_after_jit_grad                   :     0.000473s :  2.92%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006573s : 40.55%
    execute                              :     0.000007s :  0.04%


Time group info:
------[substitution.]   0.000249    23
 1.15% :     0.000003s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.61% :     0.000007s :      3: substitution.graph_param_transform
85.07% :     0.000212s :      4: substitution.inline
 1.53% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.78% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.02% :     0.000005s :      2: substitution.replace_old_param
 5.32% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004719     2
80.36% :     0.003792s :      1: type_inference.infer
19.64% :     0.000927s :      1: type_inference.specialize
------[replace.]   0.000086     6
65.04% :     0.000056s :      4: replace.inline
34.96% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000221     6
94.60% :     0.000209s :      4: match.inline
 5.40% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   691
 1.24% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.23% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 2.25% :     0.000003s :     10: predicate.arithmetic_simplify
 1.49% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.45% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.56% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.76% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.15% :     0.000002s :      6: predicate.get_grad_eliminate
 0.35% :     0.000000s :      3: predicate.graph_param_transform
 5.31% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.34% :     0.000002s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 1.79% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.18% :     0.000002s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.89% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.75% :     0.000002s :     10: predicate.same_eliminate
 0.69% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.57% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.56% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.64% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 1.97% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.75% :     0.000010s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.57% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.47% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.66% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000685    12
43.55% :     0.000299s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.82% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.62% :     0.000340s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022623    76
 0.23% :     0.000052s :      1: add_recomputation
 0.31% :     0.000071s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.46% :     0.000329s :      1: bootstrap
 0.13% :     0.000029s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.04% :     0.000010s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.09% :     0.006129s :      1: jit_opt_a
 0.71% :     0.000160s :      1: jit_opt_after_cconv
 0.63% :     0.000142s :      1: jit_opt_b
 2.19% :     0.000495s :      1: loop_unroll
 3.29% :     0.000744s :      1: mutable_eliminate
 4.27% :     0.000965s :     26: opt.transform.jit_opt_a
 0.24% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000031s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.13% :     0.000481s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000042s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.32% :     0.000525s :      1: renormalize.infer
 2.06% :     0.000465s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000045s :      1: rewriter_after_opt_a
 0.33% :     0.000075s :      1: rewriter_before_opt_a
 0.37% :     0.000084s :      1: symbol_engine_optimizer
29.12% :     0.006587s :      1: task_emit
21.21% :     0.004798s :      1: type_inference
 0.26% :     0.000059s :      1: validate


TotalTime = 0.0130886, [30]
[bootstrap]: 0.00033222
[type_inference]: 0.00454126
[event_method]: 1.723e-05
[auto_monad]: 6.287e-05
[graph_reusing]: 5.71998e-06
[pre_auto_parallel]: 2.67001e-06
[py_interpret_to_execute]: 3.274e-05
[rewriter_before_opt_a]: 7.172e-05
[expand_dump_flag]: 3.02002e-06
[jit_opt_a]: 0.00563144, [2]
    [Cycle 1]: 0.00169185, [27]
        [switch_simplify]: 6.249e-05
        [loop_unroll]: 3.116e-05
        [a_1]: 0.00056895
        [with_stream_mark]: 1.895e-05
        [recompute_prepare]: 7.38e-06
        [updatestate_depend_eliminate]: 3.50998e-06
        [updatestate_assign_eliminate]: 2.83998e-06
        [updatestate_loads_eliminate]: 2.67001e-06
        [parameter_eliminate]: 1.80001e-06
        [specialize_transform]: 6.04001e-06
        [updatestate_useless_node_eliminater]: 5.28002e-06
        [accelerated_algorithm]: 5.91e-06
        [meta_shard_fg_expand]: 2.07001e-06
        [get_grad_eliminate_]: 5.57999e-06
        [merge_forward]: 3.33e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.417e-05
        [j_node_and_user_rematch]: 9.39998e-06
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 9.37001e-06
        [inline_without_move]: 5.68002e-06
        [renormalize]: 0.00070075
        [add_forward_monad_depend]: 4.99998e-06
        [auto_monad_grad]: 2.82002e-06
        [auto_monad_eliminator]: 1.269e-05
        [cse]: 2.196e-05
        [replace_applicator]: 1.098e-05
    [Cycle 2]: 0.00031945, [27]
        [switch_simplify]: 5.62999e-06
        [loop_unroll]: 4.92999e-06
        [a_1]: 9.34e-05
        [with_stream_mark]: 9.10001e-06
        [recompute_prepare]: 5.14e-06
        [updatestate_depend_eliminate]: 2.36e-06
        [updatestate_assign_eliminate]: 2.00002e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 5.32001e-06
        [updatestate_useless_node_eliminater]: 5.10001e-06
        [accelerated_algorithm]: 5.00001e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 4.75999e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.46002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.235e-05
        [j_node_and_user_rematch]: 8.45999e-06
        [meta_fg_expand]: 1.74998e-06
        [replace_old_param]: 7.46999e-06
        [inline_without_move]: 4.75999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.18001e-06
        [auto_monad_grad]: 8.99978e-07
        [auto_monad_eliminator]: 4.99e-06
        [cse]: 1.047e-05
        [replace_applicator]: 5.14e-06
[py_interpret_to_execute_after_opt_a]: 1.102e-05
[rewriter_after_opt_a]: 3.483e-05
[convert_after_rewriter]: 7.35998e-06
[order_py_execute_after_rewriter]: 5.17e-06
[mutable_eliminate]: 0.00057422
[jit_opt_b]: 5.235e-05, [1]
    [Cycle 1]: 4.436e-05, [2]
        [frontend_op_eliminate]: 1.762e-05
        [inline_after_opt_a]: 1.569e-05
[cconv]: 2.624e-05
[loop_unroll]: 0.00046932
[jit_opt_after_cconv]: 0.00026959, [1]
    [Cycle 1]: 0.00026224, [11]
        [c_1]: 4.418e-05
        [parameter_eliminate]: 2.39001e-06
        [updatestate_depend_eliminate]: 5.67001e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [cse]: 1.965e-05
        [call_graph_tuple_transform]: 8.401e-05
        [tuple_list_get_item_eliminator]: 5.76e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 4.081e-05
[remove_dup_value]: 1.53e-05
[partial_unused_args_eliminate]: 2.55002e-06
[environ_conv]: 5.09e-06
[add_recomputation]: 4.897e-05
[cse_after_recomputation]: 2.185e-05, [1]
    [Cycle 1]: 1.571e-05, [1]
        [cse]: 1.005e-05
[auto_monad_reorder]: 1.535e-05
[get_jit_bprop_graph]: 1.92001e-06
[rewriter_after_jit_bprop_graph]: 6.29999e-06
[opt_after_jit_grad]: 0.00053067
[symbol_engine_optimizer]: 8.31e-05, [1]
    [Cycle 1]: 7.578e-05, [6]
        [build]: 4.08001e-06
        [elim_shapecalc]: 8.65999e-06
        [elim_not_effective]: 1.575e-05
        [opt_reshape]: 6.89999e-06
        [fold_const_symbol]: 9.37999e-06
        [renormalize]: 6.10016e-07
[validate]: 3.543e-05
Sums
    bootstrap                            :     0.000332s :  3.74%
    type_inference                       :     0.004541s : 51.08%
    event_method                         :     0.000017s :  0.19%
    auto_monad                           :     0.000063s :  0.71%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.37%
    rewriter_before_opt_a                :     0.000072s :  0.81%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000068s :  0.77%
    jit_opt_a.loop_unroll                :     0.000036s :  0.41%
    jit_opt_a.a_1                        :     0.000662s :  7.45%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000701s :  7.88%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.20%
    jit_opt_a.cse                        :     0.000032s :  0.36%
    jit_opt_a.replace_applicator         :     0.000016s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000035s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000574s :  6.46%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.18%
    cconv                                :     0.000026s :  0.30%
    loop_unroll                          :     0.000469s :  5.28%
    jit_opt_after_cconv.c_1              :     0.000044s :  0.50%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000084s :  0.95%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000041s :  0.46%
    remove_dup_value                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000049s :  0.55%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000531s :  5.97%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.18%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000035s :  0.40%


Time group info:
------[substitution.]   0.000223    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000002s :      2: substitution.fold_const_symbol
 2.30% :     0.000005s :      3: substitution.graph_param_transform
84.68% :     0.000189s :      4: substitution.inline
 1.66% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.03% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.81% :     0.000004s :      2: substitution.replace_old_param
 5.73% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004477     2
81.11% :     0.003631s :      1: type_inference.infer
18.89% :     0.000846s :      1: type_inference.specialize
------[replace.]   0.000072     6
67.00% :     0.000048s :      4: replace.inline
33.00% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000198     6
94.15% :     0.000186s :      4: match.inline
 5.85% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.52% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 2.17% :     0.000003s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.36% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.23% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.51% :     0.000001s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.22% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.89% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.86% :     0.000002s :     15: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.08% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.77% :     0.000002s :      3: predicate.mutable_eliminate
 0.61% :     0.000001s :      3: predicate.opt_reshape
 2.03% :     0.000002s :     15: predicate.partial_eliminate
 1.25% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.14% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.01% :     0.000010s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.32% :     0.000002s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.69% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.02% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000574    12
47.25% :     0.000271s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.95% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.80% :     0.000263s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014822    72
 0.35% :     0.000052s :      1: add_recomputation
 0.45% :     0.000066s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.32% :     0.000344s :      1: bootstrap
 0.20% :     0.000029s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.14% :     0.000021s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.02% :     0.005635s :      1: jit_opt_a
 1.84% :     0.000273s :      1: jit_opt_after_cconv
 0.37% :     0.000055s :      1: jit_opt_b
 3.22% :     0.000478s :      1: loop_unroll
 3.93% :     0.000583s :      1: mutable_eliminate
 5.95% :     0.000882s :     26: opt.transform.jit_opt_a
 0.91% :     0.000135s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000027s :      4: opt.transform.jit_opt_b
 0.09% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.18% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.65% :     0.000541s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000035s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000018s :      1: remove_dup_value
 2.65% :     0.000393s :      1: renormalize.infer
 2.02% :     0.000300s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000038s :      1: rewriter_after_opt_a
 0.50% :     0.000074s :      1: rewriter_before_opt_a
 0.58% :     0.000086s :      1: symbol_engine_optimizer
30.72% :     0.004553s :      1: type_inference


TotalTime = 0.0207867, [33]
[bootstrap]: 0.00034866
[type_inference]: 0.00498684
[event_method]: 1.65e-05
[auto_monad]: 4.511e-05
[graph_reusing]: 3.75e-06
[pre_auto_parallel]: 1.81003e-06
[py_interpret_to_execute]: 4.664e-05
[rewriter_before_opt_a]: 7.634e-05
[expand_dump_flag]: 3.11999e-06
[jit_opt_a]: 0.0062235, [2]
    [Cycle 1]: 0.00206591, [27]
        [switch_simplify]: 6.738e-05
        [loop_unroll]: 3.138e-05
        [a_1]: 0.00061986
        [with_stream_mark]: 1.921e-05
        [recompute_prepare]: 2.088e-05
        [updatestate_depend_eliminate]: 4.12998e-06
        [updatestate_assign_eliminate]: 3.64002e-06
        [updatestate_loads_eliminate]: 2.73003e-06
        [parameter_eliminate]: 2.39999e-06
        [specialize_transform]: 6.89999e-06
        [updatestate_useless_node_eliminater]: 5.91e-06
        [accelerated_algorithm]: 6.17001e-06
        [meta_shard_fg_expand]: 2.65002e-06
        [get_grad_eliminate_]: 5.50001e-06
        [merge_forward]: 4.07003e-06
        [cell_reuse_recompute_pass]: 1.69998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.516e-05
        [j_node_and_user_rematch]: 9.56e-06
        [meta_fg_expand]: 2.47001e-06
        [replace_old_param]: 9.83002e-06
        [inline_without_move]: 5.44e-06
        [renormalize]: 0.00096453
        [add_forward_monad_depend]: 6.31e-06
        [auto_monad_grad]: 2.56e-06
        [auto_monad_eliminator]: 1.454e-05
        [cse]: 2.703e-05
        [replace_applicator]: 1.354e-05
    [Cycle 2]: 0.0003312, [27]
        [switch_simplify]: 5.86e-06
        [loop_unroll]: 5.15001e-06
        [a_1]: 9.602e-05
        [with_stream_mark]: 1.014e-05
        [recompute_prepare]: 5.13002e-06
        [updatestate_depend_eliminate]: 2.65002e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 5.05001e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 4.70001e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 2.73e-06
        [cell_reuse_recompute_pass]: 2.08002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.319e-05
        [j_node_and_user_rematch]: 8.19998e-06
        [meta_fg_expand]: 1.80001e-06
        [replace_old_param]: 8.2e-06
        [inline_without_move]: 5.26002e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 8.80013e-07
        [auto_monad_eliminator]: 5.65001e-06
        [cse]: 1.046e-05
        [replace_applicator]: 4.74e-06
[py_interpret_to_execute_after_opt_a]: 1.2e-05
[rewriter_after_opt_a]: 3.798e-05
[convert_after_rewriter]: 7.09001e-06
[order_py_execute_after_rewriter]: 4.63999e-06
[mutable_eliminate]: 0.00068184
[jit_opt_b]: 7.112e-05, [1]
    [Cycle 1]: 4.611e-05, [2]
        [frontend_op_eliminate]: 1.884e-05
        [inline_after_opt_a]: 1.638e-05
[cconv]: 2.726e-05
[loop_unroll]: 0.0004554
[jit_opt_after_cconv]: 0.00015234, [1]
    [Cycle 1]: 0.00014486, [11]
        [c_1]: 2.216e-05
        [parameter_eliminate]: 2.61e-06
        [updatestate_depend_eliminate]: 6.05002e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.35002e-06
        [cse]: 2.422e-05
        [call_graph_tuple_transform]: 2.06e-05
        [tuple_list_get_item_eliminator]: 5.75001e-06
        [none_parameter_eliminate]: 1.34e-06
        [renormalize]: 8.80013e-07
        [switch_simplify]: 5.45001e-06
[remove_dup_value]: 1.43e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 5.97999e-06
[add_recomputation]: 5.293e-05
[cse_after_recomputation]: 2.3e-05, [1]
    [Cycle 1]: 1.713e-05, [1]
        [cse]: 1.09e-05
[auto_monad_reorder]: 1.495e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 6.10002e-06
[opt_after_jit_grad]: 0.00047638
[symbol_engine_optimizer]: 7.707e-05, [1]
    [Cycle 1]: 7.042e-05, [6]
        [build]: 4.3e-06
        [elim_shapecalc]: 7.73999e-06
        [elim_not_effective]: 1.407e-05
        [opt_reshape]: 6.01e-06
        [fold_const_symbol]: 9.49999e-06
        [renormalize]: 4.50003e-07
[validate]: 4.173e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00662912
[execute]: 1.009e-05
Sums
    bootstrap                            :     0.000349s :  2.14%
    type_inference                       :     0.004987s : 30.62%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000045s :  0.28%
    graph_reusing                        :     0.000004s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000047s :  0.29%
    rewriter_before_opt_a                :     0.000076s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000073s :  0.45%
    jit_opt_a.loop_unroll                :     0.000037s :  0.22%
    jit_opt_a.a_1                        :     0.000716s :  4.40%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000026s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000965s :  5.92%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000037s :  0.23%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000682s :  4.19%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000027s :  0.17%
    loop_unroll                          :     0.000455s :  2.80%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000053s :  0.33%
    cse_after_recomputation.cse          :     0.000011s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000476s :  2.93%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000042s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006629s : 40.71%
    execute                              :     0.000010s :  0.06%


Time group info:
------[substitution.]   0.000247    23
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000002s :      2: substitution.fold_const_symbol
 2.22% :     0.000005s :      3: substitution.graph_param_transform
85.06% :     0.000210s :      4: substitution.inline
 1.36% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.96% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.85% :     0.000005s :      2: substitution.replace_old_param
 5.97% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004922     2
80.97% :     0.003985s :      1: type_inference.infer
19.03% :     0.000936s :      1: type_inference.specialize
------[replace.]   0.000083     6
66.27% :     0.000055s :      4: replace.inline
33.73% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000221     6
93.86% :     0.000208s :      4: match.inline
 6.14% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000144   691
 1.14% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.04% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.89% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 2.29% :     0.000003s :     10: predicate.arithmetic_simplify
 1.29% :     0.000002s :     10: predicate.cast_eliminate
 0.29% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000002s :     10: predicate.depend_value_elim
 0.94% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.02% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.73% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.03% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     10: predicate.environ_get_eliminate
 1.11% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 4.91% :     0.000007s :     22: predicate.inline
 0.72% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.98% :     0.000001s :      6: predicate.less_batch_normalization
 1.70% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.43% :     0.000002s :     15: predicate.load_eliminater
 1.16% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.45% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.57% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.90% :     0.000001s :     10: predicate.merge_addn
 0.92% :     0.000001s :     10: predicate.minmaximum_grad
 1.32% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000003s :     15: predicate.partial_eliminate
 1.04% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.98% :     0.000003s :     10: predicate.reduce_eliminate
 1.66% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.38% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.38% :     0.000002s :     18: predicate.replace_applicator
 0.79% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.11% :     0.000002s :     10: predicate.same_eliminate
 9.14% :     0.000013s :      6: predicate.set_cell_output_no_recompute
 1.32% :     0.000002s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.13% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 1.86% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.03% :     0.000010s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000002s :     10: predicate.transpose_eliminate
 1.57% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.27% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.07% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.29% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.34% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.55% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000003s :     10: predicate.value_based_eliminate
 0.25% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000705    12
48.30% :     0.000341s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.58% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.12% :     0.000325s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022806    76
 0.25% :     0.000057s :      1: add_recomputation
 0.21% :     0.000048s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.62% :     0.000368s :      1: bootstrap
 0.13% :     0.000030s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000020s :      1: event_method
 0.07% :     0.000016s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
27.31% :     0.006228s :      1: jit_opt_a
 0.68% :     0.000155s :      1: jit_opt_after_cconv
 0.33% :     0.000074s :      1: jit_opt_b
 2.03% :     0.000464s :      1: loop_unroll
 3.04% :     0.000692s :      1: mutable_eliminate
 4.20% :     0.000959s :     26: opt.transform.jit_opt_a
 0.22% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000028s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.12% :     0.000484s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.22% :     0.000049s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.36% :     0.000539s :      1: renormalize.infer
 1.82% :     0.000416s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000041s :      1: rewriter_after_opt_a
 0.35% :     0.000080s :      1: rewriter_before_opt_a
 0.35% :     0.000080s :      1: symbol_engine_optimizer
29.14% :     0.006646s :      1: task_emit
21.92% :     0.005000s :      1: type_inference
 0.28% :     0.000063s :      1: validate


TotalTime = 0.017981, [33]
[bootstrap]: 0.00030466
[type_inference]: 0.00472737
[event_method]: 2.031e-05
[auto_monad]: 6.149e-05
[graph_reusing]: 5.50001e-06
[pre_auto_parallel]: 2.26998e-06
[py_interpret_to_execute]: 4.054e-05
[rewriter_before_opt_a]: 7.21e-05
[expand_dump_flag]: 2.67001e-06
[jit_opt_a]: 0.00537663, [2]
    [Cycle 1]: 0.00158825, [27]
        [switch_simplify]: 4.882e-05
        [loop_unroll]: 2.984e-05
        [a_1]: 0.00051875
        [with_stream_mark]: 1.26e-05
        [recompute_prepare]: 6.59001e-06
        [updatestate_depend_eliminate]: 3.16001e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.49e-06
        [updatestate_useless_node_eliminater]: 4.59998e-06
        [accelerated_algorithm]: 5.09998e-06
        [meta_shard_fg_expand]: 1.18001e-06
        [get_grad_eliminate_]: 5.30001e-06
        [merge_forward]: 2.39999e-06
        [cell_reuse_recompute_pass]: 8.00006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.077e-05
        [j_node_and_user_rematch]: 8.67e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 7.96001e-06
        [inline_without_move]: 4.90001e-06
        [renormalize]: 0.00069207
        [add_forward_monad_depend]: 4.53001e-06
        [auto_monad_grad]: 1.45999e-06
        [auto_monad_eliminator]: 1.042e-05
        [cse]: 1.512e-05
        [replace_applicator]: 1.217e-05
    [Cycle 2]: 0.00030882, [27]
        [switch_simplify]: 5.46998e-06
        [loop_unroll]: 4.70001e-06
        [a_1]: 8.695e-05
        [with_stream_mark]: 7.97e-06
        [recompute_prepare]: 5.14e-06
        [updatestate_depend_eliminate]: 2.37001e-06
        [updatestate_assign_eliminate]: 1.97999e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 8.59989e-07
        [specialize_transform]: 4.67e-06
        [updatestate_useless_node_eliminater]: 4.68001e-06
        [accelerated_algorithm]: 5.25001e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.39002e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 1.69998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.137e-05
        [j_node_and_user_rematch]: 7.55e-06
        [meta_fg_expand]: 1.44998e-06
        [replace_old_param]: 7.28999e-06
        [inline_without_move]: 4.65001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 6.50005e-07
        [auto_monad_eliminator]: 4.77998e-06
        [cse]: 9.39e-06
        [replace_applicator]: 5.37999e-06
[py_interpret_to_execute_after_opt_a]: 8.21002e-06
[rewriter_after_opt_a]: 2.732e-05
[convert_after_rewriter]: 5.34998e-06
[order_py_execute_after_rewriter]: 3.95e-06
[mutable_eliminate]: 0.00051217
[jit_opt_b]: 4.707e-05, [1]
    [Cycle 1]: 4.079e-05, [2]
        [frontend_op_eliminate]: 1.557e-05
        [inline_after_opt_a]: 1.394e-05
[cconv]: 1.634e-05
[loop_unroll]: 0.00040275
[jit_opt_after_cconv]: 0.00013071, [1]
    [Cycle 1]: 0.00012495, [11]
        [c_1]: 1.995e-05
        [parameter_eliminate]: 2.32001e-06
        [updatestate_depend_eliminate]: 5.50001e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.08002e-06
        [cse]: 1.599e-05
        [call_graph_tuple_transform]: 1.647e-05
        [tuple_list_get_item_eliminator]: 4.97e-06
        [none_parameter_eliminate]: 1.11997e-06
        [renormalize]: 5.00004e-07
        [switch_simplify]: 4.84998e-06
[remove_dup_value]: 6.53998e-06
[partial_unused_args_eliminate]: 1.54998e-06
[environ_conv]: 4.10998e-06
[add_recomputation]: 3.072e-05
[cse_after_recomputation]: 1.907e-05, [1]
    [Cycle 1]: 1.409e-05, [1]
        [cse]: 8.38001e-06
[auto_monad_reorder]: 9.03002e-06
[get_jit_bprop_graph]: 1.74998e-06
[rewriter_after_jit_bprop_graph]: 5.21002e-06
[opt_after_jit_grad]: 0.0004509
[symbol_engine_optimizer]: 6.948e-05, [1]
    [Cycle 1]: 6.369e-05, [6]
        [build]: 3.73999e-06
        [elim_shapecalc]: 6.86999e-06
        [elim_not_effective]: 1.208e-05
        [opt_reshape]: 5.27001e-06
        [fold_const_symbol]: 8.60001e-06
        [renormalize]: 3.89991e-07
[validate]: 2.938e-05
[backend_pass]: 7.30011e-07
[task_emit]: 0.00537832
[execute]: 6.09001e-06
Sums
    bootstrap                            :     0.000305s :  2.19%
    type_inference                       :     0.004727s : 34.00%
    event_method                         :     0.000020s :  0.15%
    auto_monad                           :     0.000061s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000041s :  0.29%
    rewriter_before_opt_a                :     0.000072s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.25%
    jit_opt_a.a_1                        :     0.000606s :  4.36%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000692s :  4.98%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000025s :  0.18%
    jit_opt_a.replace_applicator         :     0.000018s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000027s :  0.20%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000512s :  3.68%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000016s :  0.12%
    loop_unroll                          :     0.000403s :  2.90%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000007s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000031s :  0.22%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000009s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000451s :  3.24%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005378s : 38.69%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000196    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000001s :      2: substitution.fold_const_symbol
 2.02% :     0.000004s :      3: substitution.graph_param_transform
85.67% :     0.000168s :      4: substitution.inline
 1.28% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.01% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.93% :     0.000004s :      2: substitution.replace_old_param
 5.52% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004654     2
81.68% :     0.003801s :      1: type_inference.infer
18.32% :     0.000853s :      1: type_inference.specialize
------[replace.]   0.000067     6
65.97% :     0.000044s :      4: replace.inline
34.03% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000175     6
94.44% :     0.000165s :      4: match.inline
 5.56% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.27% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.23% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.59% :     0.000002s :     10: predicate.addn_zero_filter
 2.40% :     0.000003s :     10: predicate.arithmetic_simplify
 1.30% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.52% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.26% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.45% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.85% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.68% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.83% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.42% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.41% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.22% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.73% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.84% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000008s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.44% :     0.000002s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.20% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000562    12
48.47% :     0.000272s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.65% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.89% :     0.000258s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019567    76
 0.17% :     0.000034s :      1: add_recomputation
 0.33% :     0.000065s :      1: auto_monad
 0.06% :     0.000011s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.65% :     0.000323s :      1: bootstrap
 0.10% :     0.000019s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.12% :     0.000024s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.50% :     0.005380s :      1: jit_opt_a
 0.68% :     0.000133s :      1: jit_opt_after_cconv
 0.25% :     0.000049s :      1: jit_opt_b
 2.10% :     0.000410s :      1: loop_unroll
 2.66% :     0.000521s :      1: mutable_eliminate
 4.11% :     0.000804s :     26: opt.transform.jit_opt_a
 0.22% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.34% :     0.000459s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.22% :     0.000043s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000009s :      1: remove_dup_value
 1.97% :     0.000386s :      1: renormalize.infer
 1.53% :     0.000299s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000030s :      1: rewriter_after_opt_a
 0.39% :     0.000076s :      1: rewriter_before_opt_a
 0.37% :     0.000072s :      1: symbol_engine_optimizer
27.56% :     0.005392s :      1: task_emit
24.24% :     0.004744s :      1: type_inference
 0.24% :     0.000046s :      1: validate


TotalTime = 0.01884, [33]
[bootstrap]: 0.00032231
[type_inference]: 0.0047856
[event_method]: 1.655e-05
[auto_monad]: 6.266e-05
[graph_reusing]: 5.96e-06
[pre_auto_parallel]: 2.82002e-06
[py_interpret_to_execute]: 3.614e-05
[rewriter_before_opt_a]: 7.068e-05
[expand_dump_flag]: 3.33e-06
[jit_opt_a]: 0.00529591, [2]
    [Cycle 1]: 0.00156715, [27]
        [switch_simplify]: 5.577e-05
        [loop_unroll]: 2.885e-05
        [a_1]: 0.0005176
        [with_stream_mark]: 1.68e-05
        [recompute_prepare]: 7.29001e-06
        [updatestate_depend_eliminate]: 3.57997e-06
        [updatestate_assign_eliminate]: 2.84001e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [parameter_eliminate]: 1.66998e-06
        [specialize_transform]: 5.44e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 1.91998e-06
        [get_grad_eliminate_]: 5.27999e-06
        [merge_forward]: 3.79002e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.392e-05
        [j_node_and_user_rematch]: 8.56002e-06
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 8.50999e-06
        [inline_without_move]: 5.19998e-06
        [renormalize]: 0.00063565
        [add_forward_monad_depend]: 4.80001e-06
        [auto_monad_grad]: 2.58e-06
        [auto_monad_eliminator]: 1.257e-05
        [cse]: 2.561e-05
        [replace_applicator]: 1.156e-05
    [Cycle 2]: 0.00034973, [27]
        [switch_simplify]: 5.49998e-06
        [loop_unroll]: 4.95999e-06
        [a_1]: 8.607e-05
        [with_stream_mark]: 4.564e-05
        [recompute_prepare]: 5.51998e-06
        [updatestate_depend_eliminate]: 2.65002e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 9.00007e-07
        [specialize_transform]: 4.85001e-06
        [updatestate_useless_node_eliminater]: 5.08002e-06
        [accelerated_algorithm]: 4.74e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 2.33998e-06
        [cell_reuse_recompute_pass]: 1.81998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.169e-05
        [j_node_and_user_rematch]: 7.5e-06
        [meta_fg_expand]: 1.43002e-06
        [replace_old_param]: 6.81001e-06
        [inline_without_move]: 4.53999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.38002e-06
        [auto_monad_grad]: 8.30012e-07
        [auto_monad_eliminator]: 5.53002e-06
        [cse]: 1.116e-05
        [replace_applicator]: 4.89e-06
[py_interpret_to_execute_after_opt_a]: 1.014e-05
[rewriter_after_opt_a]: 3.376e-05
[convert_after_rewriter]: 6.86999e-06
[order_py_execute_after_rewriter]: 5.22e-06
[mutable_eliminate]: 0.00051841
[jit_opt_b]: 4.892e-05, [1]
    [Cycle 1]: 4.151e-05, [2]
        [frontend_op_eliminate]: 1.556e-05
        [inline_after_opt_a]: 1.47e-05
[cconv]: 2.453e-05
[loop_unroll]: 0.00040486
[jit_opt_after_cconv]: 0.00013191, [1]
    [Cycle 1]: 0.00012584, [11]
        [c_1]: 1.955e-05
        [parameter_eliminate]: 2.58e-06
        [updatestate_depend_eliminate]: 5.64e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 1.98002e-06
        [cse]: 1.607e-05
        [call_graph_tuple_transform]: 1.763e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 4.89998e-06
[remove_dup_value]: 1.385e-05
[partial_unused_args_eliminate]: 1.96e-06
[environ_conv]: 5.35999e-06
[add_recomputation]: 4.705e-05
[cse_after_recomputation]: 1.926e-05, [1]
    [Cycle 1]: 1.404e-05, [1]
        [cse]: 8.50001e-06
[auto_monad_reorder]: 1.491e-05
[get_jit_bprop_graph]: 2.04999e-06
[rewriter_after_jit_bprop_graph]: 4.77e-06
[opt_after_jit_grad]: 0.0004348
[symbol_engine_optimizer]: 6.883e-05, [1]
    [Cycle 1]: 6.28e-05, [6]
        [build]: 3.29001e-06
        [elim_shapecalc]: 6.93e-06
        [elim_not_effective]: 1.211e-05
        [opt_reshape]: 5.30001e-06
        [fold_const_symbol]: 8.36002e-06
        [renormalize]: 5.8001e-07
[validate]: 3.391e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.0061928
[execute]: 7.6e-06
Sums
    bootstrap                            :     0.000322s :  2.17%
    type_inference                       :     0.004786s : 32.22%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000063s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.24%
    rewriter_before_opt_a                :     0.000071s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000604s :  4.06%
    jit_opt_a.with_stream_mark           :     0.000062s :  0.42%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000636s :  4.28%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000037s :  0.25%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000518s :  3.49%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000405s :  2.73%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000435s :  2.93%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006193s : 41.69%
    execute                              :     0.000008s :  0.05%


TotalTime = 0.0181211, [33]
[bootstrap]: 0.00032074
[type_inference]: 0.00474583
[event_method]: 1.719e-05
[auto_monad]: 4.516e-05
[graph_reusing]: 4.22e-06
[pre_auto_parallel]: 2.17999e-06
[py_interpret_to_execute]: 4.127e-05
[rewriter_before_opt_a]: 6.71e-05
[expand_dump_flag]: 2.09999e-06
[jit_opt_a]: 0.00552921, [2]
    [Cycle 1]: 0.00179437, [27]
        [switch_simplify]: 5.375e-05
        [loop_unroll]: 3.082e-05
        [a_1]: 0.00053383
        [with_stream_mark]: 1.438e-05
        [recompute_prepare]: 6.94999e-06
        [updatestate_depend_eliminate]: 3.05998e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.32e-06
        [specialize_transform]: 6.43e-06
        [updatestate_useless_node_eliminater]: 5.30999e-06
        [accelerated_algorithm]: 6.07999e-06
        [meta_shard_fg_expand]: 1.45999e-06
        [get_grad_eliminate_]: 5.40999e-06
        [merge_forward]: 2.71e-06
        [cell_reuse_recompute_pass]: 9.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.143e-05
        [j_node_and_user_rematch]: 8.93002e-06
        [meta_fg_expand]: 2.02999e-06
        [replace_old_param]: 8.57e-06
        [inline_without_move]: 5.19e-06
        [renormalize]: 0.00087051
        [add_forward_monad_depend]: 5.37001e-06
        [auto_monad_grad]: 1.30999e-06
        [auto_monad_eliminator]: 9.64e-06
        [cse]: 1.658e-05
        [replace_applicator]: 1.165e-05
    [Cycle 2]: 0.0003273, [27]
        [switch_simplify]: 5.51998e-06
        [loop_unroll]: 5.02e-06
        [a_1]: 9.226e-05
        [with_stream_mark]: 9.36998e-06
        [recompute_prepare]: 5.17e-06
        [updatestate_depend_eliminate]: 2.51998e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 5.18002e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 5.36998e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.80999e-06
        [merge_forward]: 2.58003e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.273e-05
        [j_node_and_user_rematch]: 7.85998e-06
        [meta_fg_expand]: 1.69998e-06
        [replace_old_param]: 7.2e-06
        [inline_without_move]: 4.63999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.35999e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 5.46998e-06
        [cse]: 1.033e-05
        [replace_applicator]: 5.40001e-06
[py_interpret_to_execute_after_opt_a]: 9.76e-06
[rewriter_after_opt_a]: 2.843e-05
[convert_after_rewriter]: 6.02001e-06
[order_py_execute_after_rewriter]: 4.22e-06
[mutable_eliminate]: 0.0005482
[jit_opt_b]: 5.079e-05, [1]
    [Cycle 1]: 4.391e-05, [2]
        [frontend_op_eliminate]: 1.678e-05
        [inline_after_opt_a]: 1.534e-05
[cconv]: 1.689e-05
[loop_unroll]: 0.00042327
[jit_opt_after_cconv]: 0.00013858, [1]
    [Cycle 1]: 0.00013152, [11]
        [c_1]: 2.14e-05
        [parameter_eliminate]: 2.78998e-06
        [updatestate_depend_eliminate]: 5.79999e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.87e-05
        [call_graph_tuple_transform]: 1.722e-05
        [tuple_list_get_item_eliminator]: 5.41998e-06
        [none_parameter_eliminate]: 9.10019e-07
        [renormalize]: 4.00003e-07
        [switch_simplify]: 5.14e-06
[remove_dup_value]: 7.36001e-06
[partial_unused_args_eliminate]: 1.71e-06
[environ_conv]: 4.67e-06
[add_recomputation]: 3.222e-05
[cse_after_recomputation]: 2.086e-05, [1]
    [Cycle 1]: 1.53e-05, [1]
        [cse]: 9.72001e-06
[auto_monad_reorder]: 1.049e-05
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 5.39998e-06
[opt_after_jit_grad]: 0.00045956
[symbol_engine_optimizer]: 7.201e-05, [1]
    [Cycle 1]: 6.523e-05, [6]
        [build]: 2.83e-06
        [elim_shapecalc]: 7.46001e-06
        [elim_not_effective]: 1.252e-05
        [opt_reshape]: 5.81e-06
        [fold_const_symbol]: 8.25999e-06
        [renormalize]: 4.39992e-07
[validate]: 2.899e-05
[backend_pass]: 6.69999e-07
[task_emit]: 0.00527402
[execute]: 4.66002e-06
Sums
    bootstrap                            :     0.000321s :  2.27%
    type_inference                       :     0.004746s : 33.63%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000045s :  0.32%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000041s :  0.29%
    rewriter_before_opt_a                :     0.000067s :  0.48%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000059s :  0.42%
    jit_opt_a.loop_unroll                :     0.000036s :  0.25%
    jit_opt_a.a_1                        :     0.000626s :  4.44%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000871s :  6.17%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000027s :  0.19%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000028s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000548s :  3.88%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000017s :  0.12%
    loop_unroll                          :     0.000423s :  3.00%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000007s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000032s :  0.23%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000010s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000460s :  3.26%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.21%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005274s : 37.37%
    execute                              :     0.000005s :  0.03%


Time group info:
------[substitution.]   0.000198    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.83% :     0.000006s :      3: substitution.graph_param_transform
84.02% :     0.000167s :      4: substitution.inline
 1.46% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.20% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.77% :     0.000004s :      2: substitution.replace_old_param
 6.12% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004714     2
82.62% :     0.003895s :      1: type_inference.infer
17.38% :     0.000819s :      1: type_inference.specialize
------[replace.]   0.000067     6
67.60% :     0.000045s :      4: replace.inline
32.40% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000175     6
93.71% :     0.000164s :      4: match.inline
 6.29% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.33% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000001s :     10: predicate.addn_zero_filter
 2.24% :     0.000002s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000001s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.44% :     0.000000s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.33% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.02% :     0.000005s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.02% :     0.000001s :      6: predicate.less_batch_normalization
 1.92% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.70% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.55% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000002s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.82% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.84% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.62% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.81% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.42% :     0.000002s :     10: predicate.transpose_eliminate
 1.31% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.81% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000603    12
53.02% :     0.000320s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.36% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.61% :     0.000251s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020375    76
 0.25% :     0.000050s :      1: add_recomputation
 0.32% :     0.000066s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.68% :     0.000343s :      1: bootstrap
 0.13% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.01% :     0.005299s :      1: jit_opt_a
 0.66% :     0.000135s :      1: jit_opt_after_cconv
 0.25% :     0.000051s :      1: jit_opt_b
 2.02% :     0.000413s :      1: loop_unroll
 2.59% :     0.000527s :      1: mutable_eliminate
 3.97% :     0.000808s :     26: opt.transform.jit_opt_a
 0.22% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.17% :     0.000443s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000039s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.62% :     0.000331s :      1: renormalize.infer
 1.46% :     0.000298s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000037s :      1: rewriter_after_opt_a
 0.36% :     0.000073s :      1: rewriter_before_opt_a
 0.35% :     0.000071s :      1: symbol_engine_optimizer
30.47% :     0.006208s :      1: task_emit
23.57% :     0.004803s :      1: type_inference
 0.26% :     0.000053s :      1: validate


Time group info:
------[substitution.]   0.000189    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 2.08% :     0.000004s :      3: substitution.graph_param_transform
86.40% :     0.000163s :      4: substitution.inline
 1.50% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.32% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000004s :      2: substitution.replace_old_param
 4.18% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004688     2
80.91% :     0.003793s :      1: type_inference.infer
19.09% :     0.000895s :      1: type_inference.specialize
------[replace.]   0.000076     6
65.91% :     0.000050s :      4: replace.inline
34.09% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000167     6
95.97% :     0.000161s :      4: match.inline
 4.03% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.23% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.57% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000001s :     10: predicate.addn_zero_filter
 1.83% :     0.000002s :     10: predicate.arithmetic_simplify
 1.13% :     0.000001s :     10: predicate.cast_eliminate
 0.65% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.50% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.08% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.32% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.57% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.37% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.38% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.79% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.82% :     0.000002s :     10: predicate.reduce_eliminate
 1.75% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000001s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.62% :     0.000009s :     53: predicate.switch_simplify
 1.28% :     0.000002s :     10: predicate.tile_eliminate
 1.36% :     0.000002s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.38% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000642    12
47.24% :     0.000304s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.11% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.64% :     0.000306s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019920    76
 0.18% :     0.000035s :      1: add_recomputation
 0.24% :     0.000049s :      1: auto_monad
 0.07% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.68% :     0.000334s :      1: bootstrap
 0.10% :     0.000019s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.04% :     0.000008s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
27.77% :     0.005532s :      1: jit_opt_a
 0.71% :     0.000142s :      1: jit_opt_after_cconv
 0.27% :     0.000054s :      1: jit_opt_b
 2.17% :     0.000432s :      1: loop_unroll
 2.79% :     0.000557s :      1: mutable_eliminate
 4.19% :     0.000835s :     26: opt.transform.jit_opt_a
 0.23% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.35% :     0.000468s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.22% :     0.000044s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000010s :      1: remove_dup_value
 2.57% :     0.000511s :      1: renormalize.infer
 1.77% :     0.000352s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000031s :      1: rewriter_after_opt_a
 0.35% :     0.000070s :      1: rewriter_before_opt_a
 0.37% :     0.000075s :      1: symbol_engine_optimizer
26.53% :     0.005285s :      1: task_emit
23.89% :     0.004759s :      1: type_inference
 0.23% :     0.000046s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.653.208 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc413a0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.654.780 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0191338, [33]
[bootstrap]: 0.00030331
[type_inference]: 0.00437037
[event_method]: 1.748e-05
[auto_monad]: 6.211e-05
[graph_reusing]: 5.84e-06
[pre_auto_parallel]: 2.57001e-06
[py_interpret_to_execute]: 3.322e-05
[rewriter_before_opt_a]: 7.263e-05
[expand_dump_flag]: 3.09999e-06
[jit_opt_a]: 0.00540526, [2]
    [Cycle 1]: 0.00167185, [27]
        [switch_simplify]: 6.452e-05
        [loop_unroll]: 3.052e-05
        [a_1]: 0.00055705
        [with_stream_mark]: 2.157e-05
        [recompute_prepare]: 7.06999e-06
        [updatestate_depend_eliminate]: 3.85e-06
        [updatestate_assign_eliminate]: 3.32002e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [parameter_eliminate]: 2.27999e-06
        [specialize_transform]: 5.82001e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.56002e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 5.18002e-06
        [merge_forward]: 3.44001e-06
        [cell_reuse_recompute_pass]: 1.05999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.374e-05
        [j_node_and_user_rematch]: 8.54e-06
        [meta_fg_expand]: 2.25002e-06
        [replace_old_param]: 8.92e-06
        [inline_without_move]: 5.40999e-06
        [renormalize]: 0.00066853
        [add_forward_monad_depend]: 5.74e-06
        [auto_monad_grad]: 2.64999e-06
        [auto_monad_eliminator]: 1.375e-05
        [cse]: 2.8e-05
        [replace_applicator]: 1.367e-05
    [Cycle 2]: 0.0003196, [27]
        [switch_simplify]: 5.34998e-06
        [loop_unroll]: 4.57e-06
        [a_1]: 8.988e-05
        [with_stream_mark]: 9.19e-06
        [recompute_prepare]: 5.46e-06
        [updatestate_depend_eliminate]: 2.74001e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 1.34998e-06
        [specialize_transform]: 4.75001e-06
        [updatestate_useless_node_eliminater]: 4.68001e-06
        [accelerated_algorithm]: 5.39998e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.48999e-06
        [merge_forward]: 2.71999e-06
        [cell_reuse_recompute_pass]: 2.22001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.223e-05
        [j_node_and_user_rematch]: 7.71001e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 8e-06
        [inline_without_move]: 4.58999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.37999e-06
        [auto_monad_grad]: 1.12999e-06
        [auto_monad_eliminator]: 5.59998e-06
        [cse]: 1.033e-05
        [replace_applicator]: 4.68999e-06
[py_interpret_to_execute_after_opt_a]: 1.104e-05
[rewriter_after_opt_a]: 3.677e-05
[convert_after_rewriter]: 6.96001e-06
[order_py_execute_after_rewriter]: 4.87998e-06
[mutable_eliminate]: 0.00062699
[jit_opt_b]: 4.979e-05, [1]
    [Cycle 1]: 4.233e-05, [2]
        [frontend_op_eliminate]: 1.611e-05
        [inline_after_opt_a]: 1.495e-05
[cconv]: 2.718e-05
[loop_unroll]: 0.00046882
[jit_opt_after_cconv]: 0.00014786, [1]
    [Cycle 1]: 0.0001411, [11]
        [c_1]: 2.14e-05
        [parameter_eliminate]: 3.26001e-06
        [updatestate_depend_eliminate]: 6.78e-06
        [updatestate_assign_eliminate]: 2.50997e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [cse]: 2.134e-05
        [call_graph_tuple_transform]: 2.045e-05
        [tuple_list_get_item_eliminator]: 5.46e-06
        [none_parameter_eliminate]: 1.70001e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 5.04998e-06
[remove_dup_value]: 1.434e-05
[partial_unused_args_eliminate]: 2.27001e-06
[environ_conv]: 5.34e-06
[add_recomputation]: 5.621e-05
[cse_after_recomputation]: 2.326e-05, [1]
    [Cycle 1]: 1.722e-05, [1]
        [cse]: 1.121e-05
[auto_monad_reorder]: 1.576e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 5.77999e-06
[opt_after_jit_grad]: 0.0005064
[symbol_engine_optimizer]: 7.79e-05, [1]
    [Cycle 1]: 7.106e-05, [6]
        [build]: 4.70001e-06
        [elim_shapecalc]: 7.7e-06
        [elim_not_effective]: 1.544e-05
        [opt_reshape]: 5.76e-06
        [fold_const_symbol]: 8.50001e-06
        [renormalize]: 6.09987e-07
[validate]: 4.154e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00644056
[execute]: 7.71999e-06
Sums
    bootstrap                            :     0.000303s :  2.02%
    type_inference                       :     0.004370s : 29.09%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000062s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.22%
    rewriter_before_opt_a                :     0.000073s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000070s :  0.47%
    jit_opt_a.loop_unroll                :     0.000035s :  0.23%
    jit_opt_a.a_1                        :     0.000647s :  4.31%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000669s :  4.45%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000038s :  0.26%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000627s :  4.17%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000027s :  0.18%
    loop_unroll                          :     0.000469s :  3.12%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000056s :  0.37%
    cse_after_recomputation.cse          :     0.000011s :  0.07%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000506s :  3.37%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.28%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006441s : 42.87%
    execute                              :     0.000008s :  0.05%


TotalTime = 0.019096, [33]
[bootstrap]: 0.00029439
[type_inference]: 0.00456015
[event_method]: 1.673e-05
[auto_monad]: 4.424e-05
[graph_reusing]: 3.86001e-06
[pre_auto_parallel]: 1.82001e-06
[py_interpret_to_execute]: 3.93e-05
[rewriter_before_opt_a]: 6.55e-05
[expand_dump_flag]: 2.07001e-06
[jit_opt_a]: 0.00572567, [2]
    [Cycle 1]: 0.00183995, [27]
        [switch_simplify]: 5.486e-05
        [loop_unroll]: 3.004e-05
        [a_1]: 0.00054582
        [with_stream_mark]: 1.481e-05
        [recompute_prepare]: 8.60999e-06
        [updatestate_depend_eliminate]: 3.08e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 2.27001e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 6.79001e-06
        [updatestate_useless_node_eliminater]: 5.37001e-06
        [accelerated_algorithm]: 6.12001e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 6.20002e-06
        [merge_forward]: 2.89999e-06
        [cell_reuse_recompute_pass]: 1.06997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.161e-05
        [j_node_and_user_rematch]: 8.85001e-06
        [meta_fg_expand]: 2.06e-06
        [replace_old_param]: 9.15999e-06
        [inline_without_move]: 5.67001e-06
        [renormalize]: 0.00088563
        [add_forward_monad_depend]: 5.47001e-06
        [auto_monad_grad]: 1.59e-06
        [auto_monad_eliminator]: 1.168e-05
        [cse]: 1.928e-05
        [replace_applicator]: 1.287e-05
    [Cycle 2]: 0.00034766, [27]
        [switch_simplify]: 5.81e-06
        [loop_unroll]: 5.40001e-06
        [a_1]: 9.696e-05
        [with_stream_mark]: 9.95002e-06
        [recompute_prepare]: 5.05999e-06
        [updatestate_depend_eliminate]: 2.99001e-06
        [updatestate_assign_eliminate]: 2.33002e-06
        [updatestate_loads_eliminate]: 1.98002e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 5.51e-06
        [updatestate_useless_node_eliminater]: 5.34e-06
        [accelerated_algorithm]: 5.87001e-06
        [meta_shard_fg_expand]: 1.21997e-06
        [get_grad_eliminate_]: 4.92e-06
        [merge_forward]: 3.27997e-06
        [cell_reuse_recompute_pass]: 2.55002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.21e-05
        [j_node_and_user_rematch]: 8.35001e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 7.96001e-06
        [inline_without_move]: 1.66e-05
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.09998e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 5.42001e-06
        [cse]: 1.155e-05
        [replace_applicator]: 5.19e-06
[py_interpret_to_execute_after_opt_a]: 1.063e-05
[rewriter_after_opt_a]: 3.501e-05
[convert_after_rewriter]: 6.65002e-06
[order_py_execute_after_rewriter]: 4.90999e-06
[mutable_eliminate]: 0.00066998
[jit_opt_b]: 9.11e-05, [1]
    [Cycle 1]: 8.321e-05, [2]
        [frontend_op_eliminate]: 5.181e-05
        [inline_after_opt_a]: 1.763e-05
[cconv]: 2.575e-05
[loop_unroll]: 0.000556
[jit_opt_after_cconv]: 0.0001704, [1]
    [Cycle 1]: 0.0001619, [11]
        [c_1]: 2.598e-05
        [parameter_eliminate]: 4.15e-06
        [updatestate_depend_eliminate]: 7.04001e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 2.48002e-06
        [cse]: 2.89e-05
        [call_graph_tuple_transform]: 2.203e-05
        [tuple_list_get_item_eliminator]: 6.66999e-06
        [none_parameter_eliminate]: 1.55001e-06
        [renormalize]: 3.60014e-07
        [switch_simplify]: 5.91998e-06
[remove_dup_value]: 1.301e-05
[partial_unused_args_eliminate]: 2.51998e-06
[environ_conv]: 5.19e-06
[add_recomputation]: 4.462e-05
[cse_after_recomputation]: 2.43e-05, [1]
    [Cycle 1]: 1.723e-05, [1]
        [cse]: 1.033e-05
[auto_monad_reorder]: 1.448e-05
[get_jit_bprop_graph]: 2.04999e-06
[rewriter_after_jit_bprop_graph]: 9.52001e-06
[opt_after_jit_grad]: 0.0006519
[symbol_engine_optimizer]: 8.349e-05, [1]
    [Cycle 1]: 7.497e-05, [6]
        [build]: 3.65e-06
        [elim_shapecalc]: 8.85001e-06
        [elim_not_effective]: 1.521e-05
        [opt_reshape]: 7.21999e-06
        [fold_const_symbol]: 9.47999e-06
        [renormalize]: 8.30012e-07
[validate]: 3.709e-05
[backend_pass]: 1.08001e-06
[task_emit]: 0.00566134
[execute]: 4.93001e-06
Sums
    bootstrap                            :     0.000294s :  1.97%
    type_inference                       :     0.004560s : 30.57%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000044s :  0.30%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000039s :  0.26%
    rewriter_before_opt_a                :     0.000066s :  0.44%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000061s :  0.41%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000643s :  4.31%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000022s :  0.15%
    jit_opt_a.renormalize                :     0.000886s :  5.94%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.11%
    jit_opt_a.cse                        :     0.000031s :  0.21%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000670s :  4.49%
    jit_opt_b.frontend_op_eliminate      :     0.000052s :  0.35%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.12%
    cconv                                :     0.000026s :  0.17%
    loop_unroll                          :     0.000556s :  3.73%
    jit_opt_after_cconv.c_1              :     0.000026s :  0.17%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000029s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000045s :  0.30%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.06%
    opt_after_jit_grad                   :     0.000652s :  4.37%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000037s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005661s : 37.96%
    execute                              :     0.000005s :  0.03%


Time group info:
------[substitution.]   0.000226    23
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.51% :     0.000006s :      3: substitution.graph_param_transform
84.95% :     0.000192s :      4: substitution.inline
 1.37% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.98% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.31% :     0.000005s :      2: substitution.replace_old_param
 5.33% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004277     2
81.16% :     0.003471s :      1: type_inference.infer
18.84% :     0.000806s :      1: type_inference.specialize
------[replace.]   0.000069     6
67.50% :     0.000047s :      4: replace.inline
32.50% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     6
94.48% :     0.000189s :      4: match.inline
 5.52% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.28% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.53% :     0.000002s :     10: predicate.addn_zero_filter
 3.20% :     0.000004s :     10: predicate.arithmetic_simplify
 1.67% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.47% :     0.000001s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.15% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 6.27% :     0.000007s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.35% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.90% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.14% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.26% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.61% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.40% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     10: predicate.reduce_eliminate
 1.37% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.54% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.49% :     0.000001s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.22% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.18% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.57% :     0.000009s :     53: predicate.switch_simplify
 1.28% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.84% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.84% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.37% :     0.000002s :     10: predicate.value_based_eliminate
 0.44% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000554    12
49.13% :     0.000272s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.91% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.96% :     0.000238s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020715    76
 0.29% :     0.000061s :      1: add_recomputation
 0.32% :     0.000066s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.54% :     0.000320s :      1: bootstrap
 0.14% :     0.000030s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000026s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.11% :     0.005409s :      1: jit_opt_a
 0.73% :     0.000151s :      1: jit_opt_after_cconv
 0.25% :     0.000052s :      1: jit_opt_b
 2.31% :     0.000479s :      1: loop_unroll
 3.07% :     0.000636s :      1: mutable_eliminate
 4.18% :     0.000865s :     26: opt.transform.jit_opt_a
 0.24% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.49% :     0.000516s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.70% :     0.000352s :      1: renormalize.infer
 1.49% :     0.000308s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000040s :      1: rewriter_after_opt_a
 0.36% :     0.000075s :      1: rewriter_before_opt_a
 0.39% :     0.000081s :      1: symbol_engine_optimizer
31.17% :     0.006457s :      1: task_emit
21.17% :     0.004385s :      1: type_inference
 0.31% :     0.000065s :      1: validate


Time group info:
------[substitution.]   0.000200    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.93% :     0.000006s :      3: substitution.graph_param_transform
84.81% :     0.000169s :      4: substitution.inline
 1.61% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.27% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.25% :     0.000004s :      2: substitution.replace_old_param
 4.54% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004501     2
80.34% :     0.003616s :      1: type_inference.infer
19.66% :     0.000885s :      1: type_inference.specialize
------[replace.]   0.000077     6
65.68% :     0.000051s :      4: replace.inline
34.32% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000175     6
95.45% :     0.000167s :      4: match.inline
 4.55% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.23% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.87% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.20% :     0.000002s :     10: predicate.addn_zero_filter
 2.10% :     0.000003s :     10: predicate.arithmetic_simplify
 1.49% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.78% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.51% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.77% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.06% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.08% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.10% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.10% :     0.000006s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.10% :     0.000003s :     15: predicate.load_eliminater
 2.11% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.07% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.27% :     0.000002s :     10: predicate.minmaximum_grad
 1.95% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 1.99% :     0.000002s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.27% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.25% :     0.000002s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.94% :     0.000004s :     16: predicate.switch_defer_inline
 1.97% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.23% :     0.000009s :     53: predicate.switch_simplify
 1.67% :     0.000002s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.52% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.88% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.43% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000655    12
45.95% :     0.000301s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.93% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.12% :     0.000322s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021012    76
 0.23% :     0.000048s :      1: add_recomputation
 0.23% :     0.000048s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.47% :     0.000309s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.13% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.04% :     0.000009s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
27.27% :     0.005729s :      1: jit_opt_a
 0.83% :     0.000174s :      1: jit_opt_after_cconv
 0.45% :     0.000094s :      1: jit_opt_b
 2.70% :     0.000568s :      1: loop_unroll
 3.24% :     0.000681s :      1: mutable_eliminate
 4.15% :     0.000872s :     26: opt.transform.jit_opt_a
 0.27% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.30% :     0.000062s :      4: opt.transform.jit_opt_b
 0.09% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.15% :     0.000663s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000042s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.21% :     0.000464s :      1: renormalize.infer
 1.97% :     0.000414s :      1: renormalize.specialize
 0.06% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000038s :      1: rewriter_after_opt_a
 0.33% :     0.000069s :      1: rewriter_before_opt_a
 0.41% :     0.000086s :      1: symbol_engine_optimizer
27.00% :     0.005674s :      1: task_emit
21.76% :     0.004573s :      1: type_inference
 0.28% :     0.000058s :      1: validate


TotalTime = 0.0144566, [30]
[bootstrap]: 0.00033296
[type_inference]: 0.00518039
[event_method]: 2.209e-05
[auto_monad]: 7.203e-05
[graph_reusing]: 6.16e-06
[pre_auto_parallel]: 2.93e-06
[py_interpret_to_execute]: 3.963e-05
[rewriter_before_opt_a]: 7.934e-05
[expand_dump_flag]: 3.63999e-06
[jit_opt_a]: 0.00635875, [2]
    [Cycle 1]: 0.00192086, [27]
        [switch_simplify]: 6.484e-05
        [loop_unroll]: 3.056e-05
        [a_1]: 0.00061183
        [with_stream_mark]: 2.582e-05
        [recompute_prepare]: 8.35001e-06
        [updatestate_depend_eliminate]: 4.02e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.84001e-06
        [parameter_eliminate]: 1.73002e-06
        [specialize_transform]: 6.44999e-06
        [updatestate_useless_node_eliminater]: 5.07999e-06
        [accelerated_algorithm]: 5.84e-06
        [meta_shard_fg_expand]: 2.62001e-06
        [get_grad_eliminate_]: 5.37001e-06
        [merge_forward]: 3.54002e-06
        [cell_reuse_recompute_pass]: 1.60999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.518e-05
        [j_node_and_user_rematch]: 9.58002e-06
        [meta_fg_expand]: 2.53e-06
        [replace_old_param]: 1.028e-05
        [inline_without_move]: 5.59e-06
        [renormalize]: 0.00082853
        [add_forward_monad_depend]: 7.7e-06
        [auto_monad_grad]: 2.99999e-06
        [auto_monad_eliminator]: 1.577e-05
        [cse]: 2.752e-05
        [replace_applicator]: 1.662e-05
    [Cycle 2]: 0.00033111, [27]
        [switch_simplify]: 5.62001e-06
        [loop_unroll]: 5.09e-06
        [a_1]: 9.135e-05
        [with_stream_mark]: 1.191e-05
        [recompute_prepare]: 5.09e-06
        [updatestate_depend_eliminate]: 2.84999e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 5.15001e-06
        [updatestate_useless_node_eliminater]: 5.19998e-06
        [accelerated_algorithm]: 5.29e-06
        [meta_shard_fg_expand]: 1.60001e-06
        [get_grad_eliminate_]: 4.82e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 2.61e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.433e-05
        [j_node_and_user_rematch]: 8.33999e-06
        [meta_fg_expand]: 1.77001e-06
        [replace_old_param]: 7.98001e-06
        [inline_without_move]: 4.71002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.39998e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 4.87e-06
        [cse]: 1.097e-05
        [replace_applicator]: 5.07999e-06
[py_interpret_to_execute_after_opt_a]: 1.214e-05
[rewriter_after_opt_a]: 3.763e-05
[convert_after_rewriter]: 7.58001e-06
[order_py_execute_after_rewriter]: 5.17999e-06
[mutable_eliminate]: 0.00071959
[jit_opt_b]: 5.731e-05, [1]
    [Cycle 1]: 4.915e-05, [2]
        [frontend_op_eliminate]: 2.111e-05
        [inline_after_opt_a]: 1.573e-05
[cconv]: 2.649e-05
[loop_unroll]: 0.00044846
[jit_opt_after_cconv]: 0.00014841, [1]
    [Cycle 1]: 0.00014182, [11]
        [c_1]: 2.164e-05
        [parameter_eliminate]: 3.03e-06
        [updatestate_depend_eliminate]: 5.59e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.31998e-06
        [cse]: 2.228e-05
        [call_graph_tuple_transform]: 2.022e-05
        [tuple_list_get_item_eliminator]: 5.55001e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 6.29982e-07
        [switch_simplify]: 5.67999e-06
[remove_dup_value]: 1.43e-05
[partial_unused_args_eliminate]: 2.23002e-06
[environ_conv]: 5.52001e-06
[add_recomputation]: 4.973e-05
[cse_after_recomputation]: 2.324e-05, [1]
    [Cycle 1]: 1.725e-05, [1]
        [cse]: 9.78998e-06
[auto_monad_reorder]: 1.473e-05
[get_jit_bprop_graph]: 2.14999e-06
[rewriter_after_jit_bprop_graph]: 6.69999e-06
[opt_after_jit_grad]: 0.00048109
[symbol_engine_optimizer]: 7.661e-05, [1]
    [Cycle 1]: 6.979e-05, [6]
        [build]: 4.11001e-06
        [elim_shapecalc]: 8.11002e-06
        [elim_not_effective]: 1.326e-05
        [opt_reshape]: 5.79999e-06
        [fold_const_symbol]: 9.41e-06
        [renormalize]: 1.10999e-06
[validate]: 3.898e-05
Sums
    bootstrap                            :     0.000333s :  3.42%
    type_inference                       :     0.005180s : 53.22%
    event_method                         :     0.000022s :  0.23%
    auto_monad                           :     0.000072s :  0.74%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000040s :  0.41%
    rewriter_before_opt_a                :     0.000079s :  0.82%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000070s :  0.72%
    jit_opt_a.loop_unroll                :     0.000036s :  0.37%
    jit_opt_a.a_1                        :     0.000703s :  7.22%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.39%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000829s :  8.51%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.21%
    jit_opt_a.cse                        :     0.000038s :  0.40%
    jit_opt_a.replace_applicator         :     0.000022s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.12%
    rewriter_after_opt_a                 :     0.000038s :  0.39%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000720s :  7.39%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.16%
    cconv                                :     0.000026s :  0.27%
    loop_unroll                          :     0.000448s :  4.61%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000014s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000050s :  0.51%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000481s :  4.94%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.40%


Time group info:
------[substitution.]   0.000263    23
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.15% :     0.000006s :      3: substitution.graph_param_transform
85.51% :     0.000225s :      4: substitution.inline
 1.31% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.91% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.32% :     0.000006s :      2: substitution.replace_old_param
 5.50% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005101     2
80.90% :     0.004126s :      1: type_inference.infer
19.10% :     0.000974s :      1: type_inference.specialize
------[replace.]   0.000077     6
66.90% :     0.000052s :      4: replace.inline
33.10% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000236     6
94.35% :     0.000223s :      4: match.inline
 5.65% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.20% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.49% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 2.54% :     0.000003s :     10: predicate.arithmetic_simplify
 1.25% :     0.000002s :     10: predicate.cast_eliminate
 0.35% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.89% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.40% :     0.000000s :      3: predicate.graph_param_transform
 6.15% :     0.000007s :     22: predicate.inline
 1.12% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 2.13% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.88% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.02% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 1.74% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000000s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.35% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.81% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.45% :     0.000001s :      3: predicate.reset_defer_inline
 1.22% :     0.000001s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.55% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.02% :     0.000010s :     53: predicate.switch_simplify
 1.10% :     0.000001s :     10: predicate.tile_eliminate
 1.07% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.71% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.30% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.07% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000703    12
50.04% :     0.000352s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.59% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.37% :     0.000298s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016279    72
 0.32% :     0.000053s :      1: add_recomputation
 0.47% :     0.000076s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 2.14% :     0.000349s :      1: bootstrap
 0.18% :     0.000029s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000027s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
39.08% :     0.006362s :      1: jit_opt_a
 0.93% :     0.000151s :      1: jit_opt_after_cconv
 0.37% :     0.000060s :      1: jit_opt_b
 2.80% :     0.000456s :      1: loop_unroll
 4.48% :     0.000729s :      1: mutable_eliminate
 5.73% :     0.000933s :     26: opt.transform.jit_opt_a
 0.30% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000030s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.01% :     0.000490s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000043s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.74% :     0.000447s :      1: renormalize.infer
 2.29% :     0.000372s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000041s :      1: rewriter_after_opt_a
 0.50% :     0.000082s :      1: rewriter_before_opt_a
 0.49% :     0.000079s :      1: symbol_engine_optimizer
31.94% :     0.005199s :      1: type_inference


TotalTime = 0.0205448, [33]
[bootstrap]: 0.00030976
[type_inference]: 0.0049999
[event_method]: 1.801e-05
[auto_monad]: 6.467e-05
[graph_reusing]: 5.74999e-06
[pre_auto_parallel]: 2.81e-06
[py_interpret_to_execute]: 3.452e-05
[rewriter_before_opt_a]: 7.119e-05
[expand_dump_flag]: 3.27002e-06
[jit_opt_a]: 0.00604593, [2]
    [Cycle 1]: 0.00183869, [27]
        [switch_simplify]: 6.24e-05
        [loop_unroll]: 3.02e-05
        [a_1]: 0.00056686
        [with_stream_mark]: 2.365e-05
        [recompute_prepare]: 8.16002e-06
        [updatestate_depend_eliminate]: 4.13999e-06
        [updatestate_assign_eliminate]: 3.10002e-06
        [updatestate_loads_eliminate]: 2.60002e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 6.26e-06
        [updatestate_useless_node_eliminater]: 5.15999e-06
        [accelerated_algorithm]: 5.66e-06
        [meta_shard_fg_expand]: 2.39001e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 3.50998e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.361e-05
        [j_node_and_user_rematch]: 8.31002e-06
        [meta_fg_expand]: 2.06998e-06
        [replace_old_param]: 9.36e-06
        [inline_without_move]: 5.21002e-06
        [renormalize]: 0.0008161
        [add_forward_monad_depend]: 6.81001e-06
        [auto_monad_grad]: 2.93e-06
        [auto_monad_eliminator]: 1.476e-05
        [cse]: 3.025e-05
        [replace_applicator]: 1.284e-05
    [Cycle 2]: 0.00044415, [27]
        [switch_simplify]: 5.65001e-06
        [loop_unroll]: 5.30999e-06
        [a_1]: 9.676e-05
        [with_stream_mark]: 1.356e-05
        [recompute_prepare]: 5.61e-06
        [updatestate_depend_eliminate]: 3.01999e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.18002e-06
        [parameter_eliminate]: 1.55999e-06
        [specialize_transform]: 5.44998e-06
        [updatestate_useless_node_eliminater]: 8.455e-05
        [accelerated_algorithm]: 6.44999e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 4.84e-06
        [merge_forward]: 4.48999e-06
        [cell_reuse_recompute_pass]: 3.09001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.547e-05
        [j_node_and_user_rematch]: 8.54e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 8.31002e-06
        [inline_without_move]: 6.01e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.11e-06
        [auto_monad_grad]: 1.29e-06
        [auto_monad_eliminator]: 6.94001e-06
        [cse]: 1.654e-05
        [replace_applicator]: 5.59e-06
[py_interpret_to_execute_after_opt_a]: 1.679e-05
[rewriter_after_opt_a]: 3.854e-05
[convert_after_rewriter]: 6.53e-06
[order_py_execute_after_rewriter]: 5.52999e-06
[mutable_eliminate]: 0.00075473
[jit_opt_b]: 5.589e-05, [1]
    [Cycle 1]: 4.738e-05, [2]
        [frontend_op_eliminate]: 1.874e-05
        [inline_after_opt_a]: 1.696e-05
[cconv]: 2.798e-05
[loop_unroll]: 0.0004944
[jit_opt_after_cconv]: 0.00015401, [1]
    [Cycle 1]: 0.00014676, [11]
        [c_1]: 2.258e-05
        [parameter_eliminate]: 2.69001e-06
        [updatestate_depend_eliminate]: 6.33e-06
        [updatestate_assign_eliminate]: 2.48002e-06
        [updatestate_loads_eliminate]: 2.34999e-06
        [cse]: 2.134e-05
        [call_graph_tuple_transform]: 2.132e-05
        [tuple_list_get_item_eliminator]: 5.80002e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 6.69999e-07
        [switch_simplify]: 5.49e-06
[remove_dup_value]: 1.405e-05
[partial_unused_args_eliminate]: 2.73e-06
[environ_conv]: 5.37001e-06
[add_recomputation]: 5.041e-05
[cse_after_recomputation]: 2.24e-05, [1]
    [Cycle 1]: 1.589e-05, [1]
        [cse]: 9.44998e-06
[auto_monad_reorder]: 1.613e-05
[get_jit_bprop_graph]: 2.58e-06
[rewriter_after_jit_bprop_graph]: 7.48e-06
[opt_after_jit_grad]: 0.00057145
[symbol_engine_optimizer]: 8.194e-05, [1]
    [Cycle 1]: 7.27e-05, [6]
        [build]: 5.54998e-06
        [elim_shapecalc]: 8.51002e-06
        [elim_not_effective]: 1.493e-05
        [opt_reshape]: 6.36e-06
        [fold_const_symbol]: 8.92e-06
        [renormalize]: 8.80013e-07
[validate]: 4.373e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.00639075
[execute]: 7.82e-06
Sums
    bootstrap                            :     0.000310s :  1.92%
    type_inference                       :     0.005000s : 31.01%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000065s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.21%
    rewriter_before_opt_a                :     0.000071s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.42%
    jit_opt_a.loop_unroll                :     0.000036s :  0.22%
    jit_opt_a.a_1                        :     0.000664s :  4.12%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000090s :  0.56%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000816s :  5.06%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.13%
    jit_opt_a.cse                        :     0.000047s :  0.29%
    jit_opt_a.replace_applicator         :     0.000018s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.10%
    rewriter_after_opt_a                 :     0.000039s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000755s :  4.68%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.11%
    cconv                                :     0.000028s :  0.17%
    loop_unroll                          :     0.000494s :  3.07%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.05%
    opt_after_jit_grad                   :     0.000571s :  3.54%
    symbol_engine_optimizer.build        :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006391s : 39.63%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000236    23
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000002s :      2: substitution.fold_const_symbol
 2.58% :     0.000006s :      3: substitution.graph_param_transform
85.20% :     0.000201s :      4: substitution.inline
 1.32% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.01% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.05% :     0.000005s :      2: substitution.replace_old_param
 5.29% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004932     2
83.14% :     0.004100s :      1: type_inference.infer
16.86% :     0.000831s :      1: type_inference.specialize
------[replace.]   0.000071     6
67.52% :     0.000048s :      4: replace.inline
32.48% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000210     6
94.61% :     0.000198s :      4: match.inline
 5.39% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.15% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.73% :     0.000002s :     10: predicate.addn_zero_filter
 3.09% :     0.000004s :     10: predicate.arithmetic_simplify
 1.51% :     0.000002s :     10: predicate.cast_eliminate
 0.63% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.45% :     0.000002s :     10: predicate.depend_value_elim
 1.21% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.11% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.65% :     0.000002s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.82% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.53% :     0.000002s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.77% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 1.71% :     0.000002s :      3: predicate.mutable_eliminate
 0.59% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.33% :     0.000002s :     10: predicate.reduce_eliminate
 1.56% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.87% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.13% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.29% :     0.000002s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.18% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.09% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.42% :     0.000009s :     53: predicate.switch_simplify
 1.42% :     0.000002s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.13% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.98% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.58% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.37% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.49% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.37% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000632    12
48.58% :     0.000307s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.13% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.28% :     0.000273s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022446    76
 0.24% :     0.000053s :      1: add_recomputation
 0.31% :     0.000069s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.45% :     0.000325s :      1: bootstrap
 0.14% :     0.000031s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.96% :     0.006050s :      1: jit_opt_a
 0.70% :     0.000157s :      1: jit_opt_after_cconv
 0.26% :     0.000059s :      1: jit_opt_b
 2.24% :     0.000503s :      1: loop_unroll
 3.41% :     0.000765s :      1: mutable_eliminate
 4.32% :     0.000969s :     26: opt.transform.jit_opt_a
 0.23% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000028s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.59% :     0.000582s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000037s :      1: py_interpret_to_execute
 0.09% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.98% :     0.000444s :      1: renormalize.infer
 1.61% :     0.000361s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000042s :      1: rewriter_after_opt_a
 0.33% :     0.000074s :      1: rewriter_before_opt_a
 0.38% :     0.000084s :      1: symbol_engine_optimizer
28.53% :     0.006403s :      1: task_emit
22.35% :     0.005017s :      1: type_inference
 0.31% :     0.000070s :      1: validate


TotalTime = 0.0253964, [33]
[bootstrap]: 0.00058218
[type_inference]: 0.0101134
[event_method]: 1.884e-05
[auto_monad]: 5.426e-05
[graph_reusing]: 4.90999e-06
[pre_auto_parallel]: 2.02999e-06
[py_interpret_to_execute]: 4.617e-05
[rewriter_before_opt_a]: 7.151e-05
[expand_dump_flag]: 2.81999e-06
[jit_opt_a]: 0.00644726, [2]
    [Cycle 1]: 0.00208229, [27]
        [switch_simplify]: 5.948e-05
        [loop_unroll]: 3.232e-05
        [a_1]: 0.00064287
        [with_stream_mark]: 1.707e-05
        [recompute_prepare]: 8.46002e-06
        [updatestate_depend_eliminate]: 3.41999e-06
        [updatestate_assign_eliminate]: 2.20002e-06
        [updatestate_loads_eliminate]: 2.11998e-06
        [parameter_eliminate]: 1.38002e-06
        [specialize_transform]: 6.51e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 5.62001e-06
        [meta_shard_fg_expand]: 1.54e-06
        [get_grad_eliminate_]: 5.82999e-06
        [merge_forward]: 3.06999e-06
        [cell_reuse_recompute_pass]: 1.67999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.269e-05
        [j_node_and_user_rematch]: 1.009e-05
        [meta_fg_expand]: 1.89e-06
        [replace_old_param]: 9.59e-06
        [inline_without_move]: 6.48998e-06
        [renormalize]: 0.00099694
        [add_forward_monad_depend]: 7.52002e-06
        [auto_monad_grad]: 2.23002e-06
        [auto_monad_eliminator]: 1.233e-05
        [cse]: 1.962e-05
        [replace_applicator]: 1.412e-05
    [Cycle 2]: 0.00038542, [27]
        [switch_simplify]: 6.75002e-06
        [loop_unroll]: 6.64001e-06
        [a_1]: 0.00010661
        [with_stream_mark]: 1.389e-05
        [recompute_prepare]: 5.35001e-06
        [updatestate_depend_eliminate]: 2.96001e-06
        [updatestate_assign_eliminate]: 2.57001e-06
        [updatestate_loads_eliminate]: 2.40002e-06
        [parameter_eliminate]: 1.80001e-06
        [specialize_transform]: 5.78002e-06
        [updatestate_useless_node_eliminater]: 5.29998e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 1.62001e-06
        [get_grad_eliminate_]: 5.27001e-06
        [merge_forward]: 2.54001e-06
        [cell_reuse_recompute_pass]: 3.44001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.72e-05
        [j_node_and_user_rematch]: 8.82999e-06
        [meta_fg_expand]: 1.85001e-06
        [replace_old_param]: 8.27e-06
        [inline_without_move]: 1.819e-05
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.77001e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 6.34999e-06
        [cse]: 1.265e-05
        [replace_applicator]: 4.75999e-06
[py_interpret_to_execute_after_opt_a]: 1.245e-05
[rewriter_after_opt_a]: 3.796e-05
[convert_after_rewriter]: 7.57998e-06
[order_py_execute_after_rewriter]: 6.12999e-06
[mutable_eliminate]: 0.00067019
[jit_opt_b]: 5.544e-05, [1]
    [Cycle 1]: 4.776e-05, [2]
        [frontend_op_eliminate]: 1.838e-05
        [inline_after_opt_a]: 1.671e-05
[cconv]: 3.038e-05
[loop_unroll]: 0.00052299
[jit_opt_after_cconv]: 0.00016232, [1]
    [Cycle 1]: 0.00015524, [11]
        [c_1]: 2.373e-05
        [parameter_eliminate]: 4.46002e-06
        [updatestate_depend_eliminate]: 6.55002e-06
        [updatestate_assign_eliminate]: 2.92002e-06
        [updatestate_loads_eliminate]: 2.76e-06
        [cse]: 2.503e-05
        [call_graph_tuple_transform]: 2.073e-05
        [tuple_list_get_item_eliminator]: 6.05002e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 1.05001e-06
        [switch_simplify]: 6.61e-06
[remove_dup_value]: 1.186e-05
[partial_unused_args_eliminate]: 2.04999e-06
[environ_conv]: 4.82e-06
[add_recomputation]: 4.349e-05
[cse_after_recomputation]: 2.098e-05, [1]
    [Cycle 1]: 1.483e-05, [1]
        [cse]: 8.69998e-06
[auto_monad_reorder]: 1.295e-05
[get_jit_bprop_graph]: 1.59e-06
[rewriter_after_jit_bprop_graph]: 6.91001e-06
[opt_after_jit_grad]: 0.00049638
[symbol_engine_optimizer]: 7.706e-05, [1]
    [Cycle 1]: 6.86e-05, [6]
        [build]: 2.84001e-06
        [elim_shapecalc]: 7.91001e-06
        [elim_not_effective]: 1.341e-05
        [opt_reshape]: 6.68e-06
        [fold_const_symbol]: 8.80999e-06
        [renormalize]: 8.10018e-07
[validate]: 3.62e-05
[backend_pass]: 7.90023e-07
[task_emit]: 0.00561752
[execute]: 5.34e-06
Sums
    bootstrap                            :     0.000582s :  2.80%
    type_inference                       :     0.010113s : 48.71%
    event_method                         :     0.000019s :  0.09%
    auto_monad                           :     0.000054s :  0.26%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000046s :  0.22%
    rewriter_before_opt_a                :     0.000072s :  0.34%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000066s :  0.32%
    jit_opt_a.loop_unroll                :     0.000039s :  0.19%
    jit_opt_a.a_1                        :     0.000749s :  3.61%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.07%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.05%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.05%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.09%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.09%
    jit_opt_a.inline_without_move        :     0.000025s :  0.12%
    jit_opt_a.renormalize                :     0.000997s :  4.80%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.09%
    jit_opt_a.cse                        :     0.000032s :  0.16%
    jit_opt_a.replace_applicator         :     0.000019s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.06%
    rewriter_after_opt_a                 :     0.000038s :  0.18%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.03%
    mutable_eliminate                    :     0.000670s :  3.23%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.08%
    cconv                                :     0.000030s :  0.15%
    loop_unroll                          :     0.000523s :  2.52%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000012s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000043s :  0.21%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000013s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.03%
    opt_after_jit_grad                   :     0.000496s :  2.39%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.17%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005618s : 27.06%
    execute                              :     0.000005s :  0.03%


Time group info:
------[substitution.]   0.000216    23
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.67% :     0.000006s :      3: substitution.graph_param_transform
84.53% :     0.000183s :      4: substitution.inline
 1.41% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.41% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000004s :      2: substitution.replace_old_param
 5.61% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010044     2
89.83% :     0.009023s :      1: type_inference.infer
10.17% :     0.001022s :      1: type_inference.specialize
------[replace.]   0.000085     6
63.43% :     0.000054s :      4: replace.inline
36.57% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000191     6
94.28% :     0.000180s :      4: match.inline
 5.72% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000175   691
 1.05% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.81% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.76% :     0.000001s :     10: predicate.addn_check_dump
 0.96% :     0.000002s :     10: predicate.addn_zero_filter
 1.80% :     0.000003s :     10: predicate.arithmetic_simplify
 1.15% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.74% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
26.36% :     0.000046s :     10: predicate.dict_get_item_const_eliminator
 1.01% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.25% :     0.000000s :      3: predicate.elim_not_effective
 0.39% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.84% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.89% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.81% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.98% :     0.000002s :     10: predicate.environ_get_eliminate
 0.92% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.14% :     0.000000s :      3: predicate.fold_const_symbol
 0.65% :     0.000001s :      6: predicate.get_grad_eliminate
 0.17% :     0.000000s :      3: predicate.graph_param_transform
 3.37% :     0.000006s :     22: predicate.inline
 0.68% :     0.000001s :      6: predicate.inline_without_move
 0.28% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.93% :     0.000002s :      6: predicate.less_batch_normalization
 1.20% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.20% :     0.000002s :     15: predicate.load_eliminater
 1.28% :     0.000002s :      3: predicate.loop_unroll_after_grad
 2.89% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.67% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000002s :     10: predicate.merge_addn
 0.81% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000003s :      3: predicate.mutable_eliminate
 0.30% :     0.000001s :      3: predicate.opt_reshape
 1.55% :     0.000003s :     15: predicate.partial_eliminate
 0.96% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.19% :     0.000002s :     10: predicate.reduce_eliminate
 1.34% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.34% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.17% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.19% :     0.000000s :      3: predicate.reset_defer_inline
 0.98% :     0.000002s :     10: predicate.reshape_eliminate
 0.92% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.66% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.97% :     0.000002s :     10: predicate.same_eliminate
 0.40% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.61% :     0.000001s :      6: predicate.special_op_eliminate
 0.82% :     0.000001s :      6: predicate.specialize_transform
 0.89% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 0.93% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.27% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.21% :     0.000004s :     16: predicate.switch_defer_inline
 1.72% :     0.000003s :     16: predicate.switch_layer_defer_inline
 5.22% :     0.000009s :     53: predicate.switch_simplify
 0.93% :     0.000002s :     10: predicate.tile_eliminate
 0.93% :     0.000002s :     10: predicate.transpose_eliminate
 1.15% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.92% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 2.64% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.29% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.14% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.17% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.12% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.22% :     0.000002s :     10: predicate.value_based_eliminate
 0.23% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.52% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000750    12
44.68% :     0.000335s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.51% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.81% :     0.000381s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027498    76
 0.17% :     0.000047s :      1: add_recomputation
 0.21% :     0.000058s :      1: auto_monad
 0.06% :     0.000015s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 2.17% :     0.000596s :      1: bootstrap
 0.12% :     0.000034s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.08% :     0.000022s :      1: event_method
 0.03% :     0.000009s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
23.46% :     0.006452s :      1: jit_opt_a
 0.60% :     0.000165s :      1: jit_opt_after_cconv
 0.21% :     0.000058s :      1: jit_opt_b
 1.93% :     0.000532s :      1: loop_unroll
 2.48% :     0.000682s :      1: mutable_eliminate
 3.62% :     0.000995s :     26: opt.transform.jit_opt_a
 0.19% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000028s :      4: opt.transform.jit_opt_b
 0.06% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000033s :      4: opt.transform.symbol_engine_opt
 1.84% :     0.000505s :      1: opt_after_jit_grad
 0.03% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000049s :      1: py_interpret_to_execute
 0.05% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000014s :      1: remove_dup_value
 1.94% :     0.000534s :      1: renormalize.infer
 1.65% :     0.000453s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000041s :      1: rewriter_after_opt_a
 0.27% :     0.000074s :      1: rewriter_before_opt_a
 0.29% :     0.000080s :      1: symbol_engine_optimizer
20.47% :     0.005630s :      1: task_emit
36.83% :     0.010129s :      1: type_inference
 0.20% :     0.000055s :      1: validate


TotalTime = 0.0130593, [30]
[bootstrap]: 0.00036682
[type_inference]: 0.00477024
[event_method]: 1.834e-05
[auto_monad]: 6.785e-05
[graph_reusing]: 5.46e-06
[pre_auto_parallel]: 2.56e-06
[py_interpret_to_execute]: 3.429e-05
[rewriter_before_opt_a]: 7.298e-05
[expand_dump_flag]: 3.36999e-06
[jit_opt_a]: 0.00563738, [2]
    [Cycle 1]: 0.00168002, [27]
        [switch_simplify]: 5.624e-05
        [loop_unroll]: 3.016e-05
        [a_1]: 0.00054577
        [with_stream_mark]: 2.19e-05
        [recompute_prepare]: 7.01001e-06
        [updatestate_depend_eliminate]: 3.5e-06
        [updatestate_assign_eliminate]: 3.38e-06
        [updatestate_loads_eliminate]: 2.98e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 5.94999e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 2.09e-06
        [get_grad_eliminate_]: 5.47999e-06
        [merge_forward]: 4.2e-06
        [cell_reuse_recompute_pass]: 1.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.367e-05
        [j_node_and_user_rematch]: 9.91e-06
        [meta_fg_expand]: 2.43e-06
        [replace_old_param]: 8.94e-06
        [inline_without_move]: 5.27001e-06
        [renormalize]: 0.00070815
        [add_forward_monad_depend]: 5.27999e-06
        [auto_monad_grad]: 2.66999e-06
        [auto_monad_eliminator]: 1.401e-05
        [cse]: 2.645e-05
        [replace_applicator]: 1.127e-05
    [Cycle 2]: 0.00032892, [27]
        [switch_simplify]: 5.56998e-06
        [loop_unroll]: 4.89e-06
        [a_1]: 9.096e-05
        [with_stream_mark]: 9.72001e-06
        [recompute_prepare]: 4.78001e-06
        [updatestate_depend_eliminate]: 2.48e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.96998e-06
        [parameter_eliminate]: 1.52001e-06
        [specialize_transform]: 4.82998e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.15999e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 4.53001e-06
        [merge_forward]: 2.76e-06
        [cell_reuse_recompute_pass]: 1.35001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.278e-05
        [j_node_and_user_rematch]: 7.81001e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 7.08e-06
        [inline_without_move]: 4.69998e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.40999e-06
        [auto_monad_grad]: 1.00001e-06
        [auto_monad_eliminator]: 5.64e-06
        [cse]: 1.091e-05
        [replace_applicator]: 5.93002e-06
[py_interpret_to_execute_after_opt_a]: 1.074e-05
[rewriter_after_opt_a]: 3.617e-05
[convert_after_rewriter]: 7.68001e-06
[order_py_execute_after_rewriter]: 5.20001e-06
[mutable_eliminate]: 0.00052426
[jit_opt_b]: 5.062e-05, [1]
    [Cycle 1]: 4.33e-05, [2]
        [frontend_op_eliminate]: 1.694e-05
        [inline_after_opt_a]: 1.491e-05
[cconv]: 2.543e-05
[loop_unroll]: 0.00043935
[jit_opt_after_cconv]: 0.00013767, [1]
    [Cycle 1]: 0.00013117, [11]
        [c_1]: 2.11e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 5.40001e-06
        [updatestate_assign_eliminate]: 2.69001e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.676e-05
        [call_graph_tuple_transform]: 1.844e-05
        [tuple_list_get_item_eliminator]: 5.22999e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 5.24998e-06
[remove_dup_value]: 1.341e-05
[partial_unused_args_eliminate]: 2.39001e-06
[environ_conv]: 4.95001e-06
[add_recomputation]: 4.667e-05
[cse_after_recomputation]: 1.98e-05, [1]
    [Cycle 1]: 1.441e-05, [1]
        [cse]: 8.40999e-06
[auto_monad_reorder]: 1.566e-05
[get_jit_bprop_graph]: 1.89e-06
[rewriter_after_jit_bprop_graph]: 4.66002e-06
[opt_after_jit_grad]: 0.00045393
[symbol_engine_optimizer]: 7.186e-05, [1]
    [Cycle 1]: 6.502e-05, [6]
        [build]: 3.11999e-06
        [elim_shapecalc]: 7.41999e-06
        [elim_not_effective]: 1.234e-05
        [opt_reshape]: 5.76e-06
        [fold_const_symbol]: 8.59e-06
        [renormalize]: 9.20001e-07
[validate]: 3.485e-05
Sums
    bootstrap                            :     0.000367s :  4.15%
    type_inference                       :     0.004770s : 53.92%
    event_method                         :     0.000018s :  0.21%
    auto_monad                           :     0.000068s :  0.77%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.39%
    rewriter_before_opt_a                :     0.000073s :  0.82%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000062s :  0.70%
    jit_opt_a.loop_unroll                :     0.000035s :  0.40%
    jit_opt_a.a_1                        :     0.000637s :  7.20%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.36%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.18%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000708s :  8.00%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.22%
    jit_opt_a.cse                        :     0.000037s :  0.42%
    jit_opt_a.replace_applicator         :     0.000017s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000036s :  0.41%
    convert_after_rewriter               :     0.000008s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000524s :  5.93%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000025s :  0.29%
    loop_unroll                          :     0.000439s :  4.97%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000013s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000047s :  0.53%
    cse_after_recomputation.cse          :     0.000008s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000454s :  5.13%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000035s :  0.39%


Time group info:
------[substitution.]   0.000215    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.50% :     0.000005s :      3: substitution.graph_param_transform
85.30% :     0.000184s :      4: substitution.inline
 1.60% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.90% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.64% :     0.000004s :      2: substitution.replace_old_param
 5.59% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004703     2
81.13% :     0.003816s :      1: type_inference.infer
18.87% :     0.000887s :      1: type_inference.specialize
------[replace.]   0.000066     6
66.97% :     0.000044s :      4: replace.inline
33.03% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     6
94.35% :     0.000181s :      4: match.inline
 5.65% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.30% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.48% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 2.41% :     0.000003s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.65% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.26% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.41% :     0.000002s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.90% :     0.000007s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.19% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.28% :     0.000002s :     10: predicate.minmaximum_grad
 1.64% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.20% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.63% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.02% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.95% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000647    12
50.38% :     0.000326s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.93% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.69% :     0.000276s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014665    72
 0.34% :     0.000050s :      1: add_recomputation
 0.49% :     0.000071s :      1: auto_monad
 0.13% :     0.000019s :      1: auto_monad_reorder
 2.59% :     0.000380s :      1: bootstrap
 0.19% :     0.000028s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000023s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.46% :     0.005640s :      1: jit_opt_a
 0.96% :     0.000140s :      1: jit_opt_after_cconv
 0.36% :     0.000053s :      1: jit_opt_b
 3.05% :     0.000447s :      1: loop_unroll
 3.63% :     0.000532s :      1: mutable_eliminate
 5.77% :     0.000847s :     26: opt.transform.jit_opt_a
 0.32% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.15% :     0.000462s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000037s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000016s :      1: remove_dup_value
 2.64% :     0.000387s :      1: renormalize.infer
 2.13% :     0.000313s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000041s :      1: rewriter_after_opt_a
 0.52% :     0.000076s :      1: rewriter_before_opt_a
 0.51% :     0.000074s :      1: symbol_engine_optimizer
32.64% :     0.004787s :      1: type_inference


TotalTime = 0.0203004, [33]
[bootstrap]: 0.00029851
[type_inference]: 0.00460061
[event_method]: 1.674e-05
[auto_monad]: 6.333e-05
[graph_reusing]: 5.41998e-06
[pre_auto_parallel]: 2.19001e-06
[py_interpret_to_execute]: 0.00011711
[rewriter_before_opt_a]: 7.347e-05
[expand_dump_flag]: 3.01001e-06
[jit_opt_a]: 0.00589544, [2]
    [Cycle 1]: 0.00203465, [27]
        [switch_simplify]: 6.236e-05
        [loop_unroll]: 2.996e-05
        [a_1]: 0.00062364
        [with_stream_mark]: 1.726e-05
        [recompute_prepare]: 8.56002e-06
        [updatestate_depend_eliminate]: 3.35e-06
        [updatestate_assign_eliminate]: 3.09001e-06
        [updatestate_loads_eliminate]: 2.47001e-06
        [parameter_eliminate]: 1.82999e-06
        [specialize_transform]: 6.09001e-06
        [updatestate_useless_node_eliminater]: 5.37001e-06
        [accelerated_algorithm]: 5.68002e-06
        [meta_shard_fg_expand]: 2.12001e-06
        [get_grad_eliminate_]: 5.24e-06
        [merge_forward]: 3.78999e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.543e-05
        [j_node_and_user_rematch]: 9.64e-06
        [meta_fg_expand]: 2.15002e-06
        [replace_old_param]: 8.80999e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.0009629
        [add_forward_monad_depend]: 5.75001e-06
        [auto_monad_grad]: 2.12001e-06
        [auto_monad_eliminator]: 1.567e-05
        [cse]: 2.726e-05
        [replace_applicator]: 1.5e-05
    [Cycle 2]: 0.00034108, [27]
        [switch_simplify]: 5.95002e-06
        [loop_unroll]: 5.00999e-06
        [a_1]: 9.36e-05
        [with_stream_mark]: 1.091e-05
        [recompute_prepare]: 5.52001e-06
        [updatestate_depend_eliminate]: 2.79999e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.43002e-06
        [parameter_eliminate]: 1.56998e-06
        [specialize_transform]: 5.00001e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 5.45001e-06
        [meta_shard_fg_expand]: 1.77999e-06
        [get_grad_eliminate_]: 4.87e-06
        [merge_forward]: 3.08998e-06
        [cell_reuse_recompute_pass]: 2.07001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.332e-05
        [j_node_and_user_rematch]: 7.65998e-06
        [meta_fg_expand]: 1.81e-06
        [replace_old_param]: 7.93999e-06
        [inline_without_move]: 4.77998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 8.79983e-07
        [auto_monad_eliminator]: 5.26998e-06
        [cse]: 1.158e-05
        [replace_applicator]: 4.90999e-06
[py_interpret_to_execute_after_opt_a]: 1.11e-05
[rewriter_after_opt_a]: 3.804e-05
[convert_after_rewriter]: 7.05998e-06
[order_py_execute_after_rewriter]: 8.18001e-06
[mutable_eliminate]: 0.00062742
[jit_opt_b]: 5.604e-05, [1]
    [Cycle 1]: 4.729e-05, [2]
        [frontend_op_eliminate]: 1.995e-05
        [inline_after_opt_a]: 1.485e-05
[cconv]: 2.688e-05
[loop_unroll]: 0.00043208
[jit_opt_after_cconv]: 0.00014832, [1]
    [Cycle 1]: 0.00014046, [11]
        [c_1]: 2.205e-05
        [parameter_eliminate]: 3.3e-06
        [updatestate_depend_eliminate]: 5.62999e-06
        [updatestate_assign_eliminate]: 2.84001e-06
        [updatestate_loads_eliminate]: 2.20002e-06
        [cse]: 2.112e-05
        [call_graph_tuple_transform]: 1.961e-05
        [tuple_list_get_item_eliminator]: 5.44998e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 6.69999e-07
        [switch_simplify]: 5.52001e-06
[remove_dup_value]: 1.504e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 5.67001e-06
[add_recomputation]: 5.094e-05
[cse_after_recomputation]: 2.453e-05, [1]
    [Cycle 1]: 1.77e-05, [1]
        [cse]: 1.082e-05
[auto_monad_reorder]: 1.706e-05
[get_jit_bprop_graph]: 2.01998e-06
[rewriter_after_jit_bprop_graph]: 7.12002e-06
[opt_after_jit_grad]: 0.0005828
[symbol_engine_optimizer]: 7.95e-05, [1]
    [Cycle 1]: 7.128e-05, [6]
        [build]: 3.76999e-06
        [elim_shapecalc]: 8.94e-06
        [elim_not_effective]: 1.326e-05
        [opt_reshape]: 6.11e-06
        [fold_const_symbol]: 9.16002e-06
        [renormalize]: 5.00004e-07
[validate]: 3.854e-05
[backend_pass]: 1.07e-06
[task_emit]: 0.00681818
[execute]: 8.18999e-06
Sums
    bootstrap                            :     0.000299s :  1.85%
    type_inference                       :     0.004601s : 28.53%
    event_method                         :     0.000017s :  0.10%
    auto_monad                           :     0.000063s :  0.39%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000117s :  0.73%
    rewriter_before_opt_a                :     0.000073s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.42%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000717s :  4.45%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000963s :  5.97%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.13%
    jit_opt_a.cse                        :     0.000039s :  0.24%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000008s :  0.05%
    mutable_eliminate                    :     0.000627s :  3.89%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.09%
    cconv                                :     0.000027s :  0.17%
    loop_unroll                          :     0.000432s :  2.68%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000051s :  0.32%
    cse_after_recomputation.cse          :     0.000011s :  0.07%
    auto_monad_reorder                   :     0.000017s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000583s :  3.61%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006818s : 42.28%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000237    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000002s :      2: substitution.fold_const_symbol
 2.85% :     0.000007s :      3: substitution.graph_param_transform
84.76% :     0.000201s :      4: substitution.inline
 1.35% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.04% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.79% :     0.000004s :      2: substitution.replace_old_param
 5.71% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004536     2
79.87% :     0.003623s :      1: type_inference.infer
20.13% :     0.000913s :      1: type_inference.specialize
------[replace.]   0.000081     6
67.23% :     0.000055s :      4: replace.inline
32.77% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000211     6
94.09% :     0.000198s :      4: match.inline
 5.91% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.69% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.83% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.11% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.23% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.24% :     0.000002s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 4.99% :     0.000006s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.89% :     0.000002s :     15: predicate.load_eliminater
 1.13% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.67% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.38% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 1.69% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     10: predicate.reduce_eliminate
 1.57% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.56% :     0.000002s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.62% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.54% :     0.000009s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.86% :     0.000002s :     10: predicate.transpose_eliminate
 1.54% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.09% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.91% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.69% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.02% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000652    12
46.99% :     0.000306s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.37% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.64% :     0.000311s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022306    76
 0.25% :     0.000055s :      1: add_recomputation
 0.30% :     0.000066s :      1: auto_monad
 0.09% :     0.000020s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.40% :     0.000312s :      1: bootstrap
 0.13% :     0.000029s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000020s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.45% :     0.005899s :      1: jit_opt_a
 0.68% :     0.000151s :      1: jit_opt_after_cconv
 0.27% :     0.000059s :      1: jit_opt_b
 1.97% :     0.000440s :      1: loop_unroll
 2.86% :     0.000638s :      1: mutable_eliminate
 4.21% :     0.000939s :     26: opt.transform.jit_opt_a
 0.22% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000028s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.66% :     0.000593s :      1: opt_after_jit_grad
 0.05% :     0.000011s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.54% :     0.000121s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 2.40% :     0.000536s :      1: renormalize.infer
 1.87% :     0.000418s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000041s :      1: rewriter_after_opt_a
 0.34% :     0.000076s :      1: rewriter_before_opt_a
 0.37% :     0.000082s :      1: symbol_engine_optimizer
30.64% :     0.006836s :      1: task_emit
20.67% :     0.004612s :      1: type_inference
 0.27% :     0.000060s :      1: validate


TotalTime = 0.0186876, [33]
[bootstrap]: 0.00027569
[type_inference]: 0.00458122
[event_method]: 1.877e-05
[auto_monad]: 5.954e-05
[graph_reusing]: 6.21e-06
[pre_auto_parallel]: 2.22001e-06
[py_interpret_to_execute]: 3.12e-05
[rewriter_before_opt_a]: 6.826e-05
[expand_dump_flag]: 2.66e-06
[jit_opt_a]: 0.00536255, [2]
    [Cycle 1]: 0.00168195, [27]
        [switch_simplify]: 4.821e-05
        [loop_unroll]: 2.823e-05
        [a_1]: 0.00052051
        [with_stream_mark]: 1.872e-05
        [recompute_prepare]: 8.40001e-06
        [updatestate_depend_eliminate]: 3.65998e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 2.07999e-06
        [specialize_transform]: 6.15002e-06
        [updatestate_useless_node_eliminater]: 4.94998e-06
        [accelerated_algorithm]: 6.34001e-06
        [meta_shard_fg_expand]: 1.62999e-06
        [get_grad_eliminate_]: 5.78997e-06
        [merge_forward]: 3.17002e-06
        [cell_reuse_recompute_pass]: 7.60017e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.247e-05
        [j_node_and_user_rematch]: 8.89e-06
        [meta_fg_expand]: 2.02001e-06
        [replace_old_param]: 7.98999e-06
        [inline_without_move]: 5.59e-06
        [renormalize]: 0.00074921
        [add_forward_monad_depend]: 5.77999e-06
        [auto_monad_grad]: 2.04e-06
        [auto_monad_eliminator]: 1.476e-05
        [cse]: 2.254e-05
        [replace_applicator]: 1.416e-05
    [Cycle 2]: 0.00033358, [27]
        [switch_simplify]: 5.67999e-06
        [loop_unroll]: 4.76002e-06
        [a_1]: 9.009e-05
        [with_stream_mark]: 1.26e-05
        [recompute_prepare]: 5.07999e-06
        [updatestate_depend_eliminate]: 2.56e-06
        [updatestate_assign_eliminate]: 2.53998e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [parameter_eliminate]: 1.23002e-06
        [specialize_transform]: 4.88001e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 5.72001e-06
        [meta_shard_fg_expand]: 1.35001e-06
        [get_grad_eliminate_]: 4.23999e-06
        [merge_forward]: 2.72001e-06
        [cell_reuse_recompute_pass]: 1.53002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.263e-05
        [j_node_and_user_rematch]: 8.47e-06
        [meta_fg_expand]: 1.76e-06
        [replace_old_param]: 8.54002e-06
        [inline_without_move]: 4.63999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.69e-06
        [auto_monad_grad]: 9.5999e-07
        [auto_monad_eliminator]: 6.15002e-06
        [cse]: 1.199e-05
        [replace_applicator]: 5.86e-06
[py_interpret_to_execute_after_opt_a]: 1.151e-05
[rewriter_after_opt_a]: 3.286e-05
[convert_after_rewriter]: 7.66001e-06
[order_py_execute_after_rewriter]: 5.56e-06
[mutable_eliminate]: 0.00063352
[jit_opt_b]: 5.044e-05, [1]
    [Cycle 1]: 4.274e-05, [2]
        [frontend_op_eliminate]: 1.605e-05
        [inline_after_opt_a]: 1.523e-05
[cconv]: 2.602e-05
[loop_unroll]: 0.00043538
[jit_opt_after_cconv]: 0.0001468, [1]
    [Cycle 1]: 0.00014028, [11]
        [c_1]: 2.173e-05
        [parameter_eliminate]: 3.42002e-06
        [updatestate_depend_eliminate]: 6.78003e-06
        [updatestate_assign_eliminate]: 2.57001e-06
        [updatestate_loads_eliminate]: 2.26998e-06
        [cse]: 2.298e-05
        [call_graph_tuple_transform]: 1.828e-05
        [tuple_list_get_item_eliminator]: 5.30001e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 6.80011e-07
        [switch_simplify]: 4.94998e-06
[remove_dup_value]: 1.171e-05
[partial_unused_args_eliminate]: 2.04999e-06
[environ_conv]: 4.75001e-06
[add_recomputation]: 4.377e-05
[cse_after_recomputation]: 2.113e-05, [1]
    [Cycle 1]: 1.516e-05, [1]
        [cse]: 8.97e-06
[auto_monad_reorder]: 1.368e-05
[get_jit_bprop_graph]: 2.19999e-06
[rewriter_after_jit_bprop_graph]: 6.27001e-06
[opt_after_jit_grad]: 0.0005193
[symbol_engine_optimizer]: 7.465e-05, [1]
    [Cycle 1]: 6.762e-05, [6]
        [build]: 3.66001e-06
        [elim_shapecalc]: 7.87998e-06
        [elim_not_effective]: 1.308e-05
        [opt_reshape]: 6.08998e-06
        [fold_const_symbol]: 8.88002e-06
        [renormalize]: 6.69999e-07
[validate]: 3.733e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.00597788
[execute]: 6.02999e-06
Sums
    bootstrap                            :     0.000276s :  1.87%
    type_inference                       :     0.004581s : 31.13%
    event_method                         :     0.000019s :  0.13%
    auto_monad                           :     0.000060s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.21%
    rewriter_before_opt_a                :     0.000068s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.37%
    jit_opt_a.loop_unroll                :     0.000033s :  0.22%
    jit_opt_a.a_1                        :     0.000611s :  4.15%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000749s :  5.09%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.14%
    jit_opt_a.cse                        :     0.000035s :  0.23%
    jit_opt_a.replace_applicator         :     0.000020s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000033s :  0.22%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000634s :  4.30%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000026s :  0.18%
    loop_unroll                          :     0.000435s :  2.96%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000012s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000044s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000519s :  3.53%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005978s : 40.61%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000195    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.88% :     0.000002s :      2: substitution.fold_const_symbol
 2.78% :     0.000005s :      3: substitution.graph_param_transform
82.89% :     0.000161s :      4: substitution.inline
 1.90% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.92% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.57% :     0.000005s :      2: substitution.replace_old_param
 5.98% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004515     2
81.62% :     0.003686s :      1: type_inference.infer
18.38% :     0.000830s :      1: type_inference.specialize
------[replace.]   0.000072     6
65.84% :     0.000047s :      4: replace.inline
34.16% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000170     6
93.74% :     0.000159s :      4: match.inline
 6.26% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.69% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.57% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.59% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.77% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.35% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.18% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.85% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.29% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.63% :     0.000002s :      6: predicate.less_batch_normalization
 2.19% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.86% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.12% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 2.28% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000002s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.42% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.99% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.23% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 2.00% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.26% :     0.000008s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.33% :     0.000002s :     10: predicate.transpose_eliminate
 1.26% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.85% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.33% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.47% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000562    12
47.73% :     0.000268s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.40% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.86% :     0.000263s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020364    76
 0.23% :     0.000047s :      1: add_recomputation
 0.31% :     0.000063s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.43% :     0.000292s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.12% :     0.000024s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.35% :     0.005366s :      1: jit_opt_a
 0.74% :     0.000150s :      1: jit_opt_after_cconv
 0.26% :     0.000053s :      1: jit_opt_b
 2.18% :     0.000444s :      1: loop_unroll
 3.16% :     0.000643s :      1: mutable_eliminate
 4.01% :     0.000817s :     26: opt.transform.jit_opt_a
 0.23% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.60% :     0.000529s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000034s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000014s :      1: remove_dup_value
 2.02% :     0.000411s :      1: renormalize.infer
 1.62% :     0.000330s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000036s :      1: rewriter_after_opt_a
 0.35% :     0.000072s :      1: rewriter_before_opt_a
 0.38% :     0.000077s :      1: symbol_engine_optimizer
29.42% :     0.005992s :      1: task_emit
22.57% :     0.004597s :      1: type_inference
 0.28% :     0.000056s :      1: validate


TotalTime = 0.0141402, [30]
[bootstrap]: 0.00033369
[type_inference]: 0.00465318
[event_method]: 1.685e-05
[auto_monad]: 6.678e-05
[graph_reusing]: 5.93998e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 3.593e-05
[rewriter_before_opt_a]: 7.667e-05
[expand_dump_flag]: 3.41999e-06
[jit_opt_a]: 0.00623134, [2]
    [Cycle 1]: 0.00196954, [27]
        [switch_simplify]: 6.737e-05
        [loop_unroll]: 3.066e-05
        [a_1]: 0.00061207
        [with_stream_mark]: 2.464e-05
        [recompute_prepare]: 1.081e-05
        [updatestate_depend_eliminate]: 4.62e-06
        [updatestate_assign_eliminate]: 3.48999e-06
        [updatestate_loads_eliminate]: 2.81999e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 7.26001e-06
        [updatestate_useless_node_eliminater]: 5.07999e-06
        [accelerated_algorithm]: 5.67999e-06
        [meta_shard_fg_expand]: 2.74001e-06
        [get_grad_eliminate_]: 5.44e-06
        [merge_forward]: 4.05e-06
        [cell_reuse_recompute_pass]: 2.59001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.743e-05
        [j_node_and_user_rematch]: 1.161e-05
        [meta_fg_expand]: 2.62001e-06
        [replace_old_param]: 1.044e-05
        [inline_without_move]: 5.89e-06
        [renormalize]: 0.00084415
        [add_forward_monad_depend]: 9.02999e-06
        [auto_monad_grad]: 3.19001e-06
        [auto_monad_eliminator]: 1.679e-05
        [cse]: 3.047e-05
        [replace_applicator]: 2.139e-05
    [Cycle 2]: 0.00041769, [27]
        [switch_simplify]: 6.16998e-06
        [loop_unroll]: 5.25999e-06
        [a_1]: 9.525e-05
        [with_stream_mark]: 4.843e-05
        [recompute_prepare]: 8.32e-06
        [updatestate_depend_eliminate]: 3.06999e-06
        [updatestate_assign_eliminate]: 3.46999e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 6.68998e-06
        [updatestate_useless_node_eliminater]: 6.04001e-06
        [accelerated_algorithm]: 6.17001e-06
        [meta_shard_fg_expand]: 2.16998e-06
        [get_grad_eliminate_]: 5.87999e-06
        [merge_forward]: 3.38e-06
        [cell_reuse_recompute_pass]: 3.61999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.793e-05
        [j_node_and_user_rematch]: 8.66002e-06
        [meta_fg_expand]: 1.89999e-06
        [replace_old_param]: 8.79e-06
        [inline_without_move]: 5.00001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.33998e-06
        [auto_monad_grad]: 1.42e-06
        [auto_monad_eliminator]: 7.16999e-06
        [cse]: 1.387e-05
        [replace_applicator]: 7.26999e-06
[py_interpret_to_execute_after_opt_a]: 1.461e-05
[rewriter_after_opt_a]: 3.922e-05
[convert_after_rewriter]: 7.18998e-06
[order_py_execute_after_rewriter]: 5.24e-06
[mutable_eliminate]: 0.00077046
[jit_opt_b]: 5.833e-05, [1]
    [Cycle 1]: 4.845e-05, [2]
        [frontend_op_eliminate]: 1.833e-05
        [inline_after_opt_a]: 1.705e-05
[cconv]: 2.867e-05
[loop_unroll]: 0.00053453
[jit_opt_after_cconv]: 0.00016925, [1]
    [Cycle 1]: 0.00016251, [11]
        [c_1]: 2.359e-05
        [parameter_eliminate]: 4.22998e-06
        [updatestate_depend_eliminate]: 7.18e-06
        [updatestate_assign_eliminate]: 3.04999e-06
        [updatestate_loads_eliminate]: 2.47001e-06
        [cse]: 2.785e-05
        [call_graph_tuple_transform]: 2.451e-05
        [tuple_list_get_item_eliminator]: 5.91e-06
        [none_parameter_eliminate]: 1.88002e-06
        [renormalize]: 8.39995e-07
        [switch_simplify]: 7.15e-06
[remove_dup_value]: 1.538e-05
[partial_unused_args_eliminate]: 2.20002e-06
[environ_conv]: 5.89e-06
[add_recomputation]: 5.598e-05
[cse_after_recomputation]: 2.476e-05, [1]
    [Cycle 1]: 1.78e-05, [1]
        [cse]: 9.80002e-06
[auto_monad_reorder]: 1.606e-05
[get_jit_bprop_graph]: 2.52001e-06
[rewriter_after_jit_bprop_graph]: 8.15e-06
[opt_after_jit_grad]: 0.00062798
[symbol_engine_optimizer]: 8.6e-05, [1]
    [Cycle 1]: 7.663e-05, [6]
        [build]: 4.54002e-06
        [elim_shapecalc]: 9.17001e-06
        [elim_not_effective]: 1.575e-05
        [opt_reshape]: 6.33998e-06
        [fold_const_symbol]: 8.99e-06
        [renormalize]: 1.66e-06
[validate]: 4.335e-05
Sums
    bootstrap                            :     0.000334s :  3.47%
    type_inference                       :     0.004653s : 48.37%
    event_method                         :     0.000017s :  0.18%
    auto_monad                           :     0.000067s :  0.69%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.37%
    rewriter_before_opt_a                :     0.000077s :  0.80%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000074s :  0.76%
    jit_opt_a.loop_unroll                :     0.000036s :  0.37%
    jit_opt_a.a_1                        :     0.000707s :  7.35%
    jit_opt_a.with_stream_mark           :     0.000073s :  0.76%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.20%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000014s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.12%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.37%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000019s :  0.20%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000844s :  8.78%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.12%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.25%
    jit_opt_a.cse                        :     0.000044s :  0.46%
    jit_opt_a.replace_applicator         :     0.000029s :  0.30%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.15%
    rewriter_after_opt_a                 :     0.000039s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000770s :  8.01%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.18%
    cconv                                :     0.000029s :  0.30%
    loop_unroll                          :     0.000535s :  5.56%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000028s :  0.29%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.25%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.07%
    remove_dup_value                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000056s :  0.58%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.17%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.08%
    opt_after_jit_grad                   :     0.000628s :  6.53%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000002s :  0.02%
    validate                             :     0.000043s :  0.45%


Time group info:
------[substitution.]   0.000239    23
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.74% :     0.000007s :      3: substitution.graph_param_transform
83.48% :     0.000200s :      4: substitution.inline
 1.79% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.22% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.35% :     0.000006s :      2: substitution.replace_old_param
 6.08% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004583     2
80.25% :     0.003678s :      1: type_inference.infer
19.75% :     0.000905s :      1: type_inference.specialize
------[replace.]   0.000087     6
63.89% :     0.000056s :      4: replace.inline
36.11% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000209     6
93.84% :     0.000196s :      4: match.inline
 6.16% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.53% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.92% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.54% :     0.000003s :     10: predicate.arithmetic_simplify
 1.42% :     0.000002s :     10: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.15% :     0.000002s :     10: predicate.depend_value_elim
 0.93% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.15% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.99% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.03% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.04% :     0.000001s :     10: predicate.environ_get_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.10% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.86% :     0.000008s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 1.75% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.90% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 2.99% :     0.000004s :      3: predicate.mutable_eliminate
 0.67% :     0.000001s :      3: predicate.opt_reshape
 1.97% :     0.000003s :     15: predicate.partial_eliminate
 1.42% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.05% :     0.000003s :     18: predicate.replace_applicator
 0.84% :     0.000001s :      6: predicate.replace_old_param
 0.45% :     0.000001s :      3: predicate.reset_defer_inline
 1.22% :     0.000002s :     10: predicate.reshape_eliminate
 1.34% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.10% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 1.27% :     0.000002s :      6: predicate.specialize_transform
 1.16% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.09% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.06% :     0.000004s :     16: predicate.switch_defer_inline
 2.07% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.02% :     0.000009s :     53: predicate.switch_simplify
 1.15% :     0.000002s :     10: predicate.tile_eliminate
 1.24% :     0.000002s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.46% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000631    12
45.65% :     0.000288s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.85% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.50% :     0.000306s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016029    72
 0.38% :     0.000060s :      1: add_recomputation
 0.44% :     0.000070s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 2.16% :     0.000346s :      1: bootstrap
 0.20% :     0.000032s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000028s :      1: cse_after_recomputation
 0.06% :     0.000010s :      1: environ_conv
 0.13% :     0.000021s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.90% :     0.006236s :      1: jit_opt_a
 1.08% :     0.000172s :      1: jit_opt_after_cconv
 0.39% :     0.000062s :      1: jit_opt_b
 3.40% :     0.000545s :      1: loop_unroll
 4.89% :     0.000784s :      1: mutable_eliminate
 6.02% :     0.000965s :     26: opt.transform.jit_opt_a
 0.36% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000028s :      4: opt.transform.jit_opt_b
 0.11% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.15% :     0.000024s :      1: opt.transform.mutable_eliminate
 0.18% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000036s :      4: opt.transform.symbol_engine_opt
 4.00% :     0.000641s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000039s :      1: py_interpret_to_execute
 0.11% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.51% :     0.000403s :      1: renormalize.infer
 2.69% :     0.000431s :      1: renormalize.specialize
 0.07% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000043s :      1: rewriter_after_opt_a
 0.50% :     0.000080s :      1: rewriter_before_opt_a
 0.56% :     0.000089s :      1: symbol_engine_optimizer
29.13% :     0.004669s :      1: type_inference


TotalTime = 0.0206322, [33]
[bootstrap]: 0.00039497
[type_inference]: 0.00531402
[event_method]: 1.938e-05
[auto_monad]: 7.452e-05
[graph_reusing]: 5.20999e-06
[pre_auto_parallel]: 2.96001e-06
[py_interpret_to_execute]: 5.138e-05
[rewriter_before_opt_a]: 7.206e-05
[expand_dump_flag]: 3.23e-06
[jit_opt_a]: 0.00584991, [2]
    [Cycle 1]: 0.00188267, [27]
        [switch_simplify]: 0.00011338
        [loop_unroll]: 3.638e-05
        [a_1]: 0.00056753
        [with_stream_mark]: 1.811e-05
        [recompute_prepare]: 8.32e-06
        [updatestate_depend_eliminate]: 4.05e-06
        [updatestate_assign_eliminate]: 2.71e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [parameter_eliminate]: 1.50999e-06
        [specialize_transform]: 6.66999e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 5.81e-06
        [meta_shard_fg_expand]: 1.40999e-06
        [get_grad_eliminate_]: 5.44e-06
        [merge_forward]: 3.16999e-06
        [cell_reuse_recompute_pass]: 1.08001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.461e-05
        [j_node_and_user_rematch]: 9.03002e-06
        [meta_fg_expand]: 2.99999e-06
        [replace_old_param]: 8.95999e-06
        [inline_without_move]: 5.56e-06
        [renormalize]: 0.0008222
        [add_forward_monad_depend]: 5.25999e-06
        [auto_monad_grad]: 1.57999e-06
        [auto_monad_eliminator]: 1.156e-05
        [cse]: 1.864e-05
        [replace_applicator]: 1.277e-05
    [Cycle 2]: 0.00032982, [27]
        [switch_simplify]: 5.96e-06
        [loop_unroll]: 6.22001e-06
        [a_1]: 9.353e-05
        [with_stream_mark]: 1.095e-05
        [recompute_prepare]: 5.62999e-06
        [updatestate_depend_eliminate]: 2.64001e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 5.10999e-06
        [updatestate_useless_node_eliminater]: 5.16002e-06
        [accelerated_algorithm]: 4.63999e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 4.29997e-06
        [merge_forward]: 2.33998e-06
        [cell_reuse_recompute_pass]: 2.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.349e-05
        [j_node_and_user_rematch]: 8e-06
        [meta_fg_expand]: 1.65001e-06
        [replace_old_param]: 7.99002e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 5.67999e-06
        [cse]: 1.044e-05
        [replace_applicator]: 5.82999e-06
[py_interpret_to_execute_after_opt_a]: 1.113e-05
[rewriter_after_opt_a]: 3.043e-05
[convert_after_rewriter]: 5.74999e-06
[order_py_execute_after_rewriter]: 3.76999e-06
[mutable_eliminate]: 0.00063276
[jit_opt_b]: 5.269e-05, [1]
    [Cycle 1]: 4.564e-05, [2]
        [frontend_op_eliminate]: 1.719e-05
        [inline_after_opt_a]: 1.556e-05
[cconv]: 1.703e-05
[loop_unroll]: 0.000442
[jit_opt_after_cconv]: 0.00014216, [1]
    [Cycle 1]: 0.00013582, [11]
        [c_1]: 2.276e-05
        [parameter_eliminate]: 2.68e-06
        [updatestate_depend_eliminate]: 5.99999e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 1.972e-05
        [call_graph_tuple_transform]: 1.706e-05
        [tuple_list_get_item_eliminator]: 5.52001e-06
        [none_parameter_eliminate]: 1.00999e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 5.87999e-06
[remove_dup_value]: 7.56999e-06
[partial_unused_args_eliminate]: 1.37e-06
[environ_conv]: 4.2e-06
[add_recomputation]: 3.384e-05
[cse_after_recomputation]: 2.169e-05, [1]
    [Cycle 1]: 1.608e-05, [1]
        [cse]: 9.89001e-06
[auto_monad_reorder]: 9.24e-06
[get_jit_bprop_graph]: 1.72001e-06
[rewriter_after_jit_bprop_graph]: 6.81001e-06
[opt_after_jit_grad]: 0.00052618
[symbol_engine_optimizer]: 7.951e-05, [1]
    [Cycle 1]: 7.244e-05, [6]
        [build]: 4.32e-06
        [elim_shapecalc]: 8.30999e-06
        [elim_not_effective]: 1.414e-05
        [opt_reshape]: 6.24001e-06
        [fold_const_symbol]: 9.05001e-06
        [renormalize]: 4.09986e-07
[validate]: 4.175e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00654344
[execute]: 7.65e-06
Sums
    bootstrap                            :     0.000395s :  2.42%
    type_inference                       :     0.005314s : 32.51%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000075s :  0.46%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000051s :  0.31%
    rewriter_before_opt_a                :     0.000072s :  0.44%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000119s :  0.73%
    jit_opt_a.loop_unroll                :     0.000043s :  0.26%
    jit_opt_a.a_1                        :     0.000661s :  4.04%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000822s :  5.03%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.11%
    jit_opt_a.cse                        :     0.000029s :  0.18%
    jit_opt_a.replace_applicator         :     0.000019s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000030s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000633s :  3.87%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000017s :  0.10%
    loop_unroll                          :     0.000442s :  2.70%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000008s :  0.05%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000034s :  0.21%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000009s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000526s :  3.22%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000042s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006543s : 40.03%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000215    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.85% :     0.000002s :      2: substitution.fold_const_symbol
 1.86% :     0.000004s :      3: substitution.graph_param_transform
85.36% :     0.000183s :      4: substitution.inline
 1.30% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.09% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000004s :      2: substitution.replace_old_param
 5.71% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005206     2
80.33% :     0.004182s :      1: type_inference.infer
19.67% :     0.001024s :      1: type_inference.specialize
------[replace.]   0.000080     6
67.35% :     0.000054s :      4: replace.inline
32.65% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     6
94.35% :     0.000181s :      4: match.inline
 5.65% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000171   691
 0.89% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.96% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.72% :     0.000001s :     10: predicate.addn_check_dump
 0.95% :     0.000002s :     10: predicate.addn_zero_filter
 1.48% :     0.000003s :     10: predicate.arithmetic_simplify
 0.85% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.70% :     0.000001s :     10: predicate.compare_switch_simplify
 1.13% :     0.000002s :     10: predicate.depend_value_elim
 0.74% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 0.83% :     0.000001s :     10: predicate.dict_get_item_eliminator
 0.74% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.57% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.20% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.91% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.78% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.77% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.03% :     0.000002s :     10: predicate.environ_get_eliminate
 0.75% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.16% :     0.000000s :      3: predicate.fold_const_symbol
 0.63% :     0.000001s :      6: predicate.get_grad_eliminate
 0.19% :     0.000000s :      3: predicate.graph_param_transform
 3.63% :     0.000006s :     22: predicate.inline
 0.64% :     0.000001s :      6: predicate.inline_without_move
 0.25% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.72% :     0.000001s :      6: predicate.less_batch_normalization
 1.00% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.33% :     0.000002s :     15: predicate.load_eliminater
 0.92% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.05% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.47% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.70% :     0.000001s :     10: predicate.merge_addn
 0.75% :     0.000001s :     10: predicate.minmaximum_grad
 1.23% :     0.000002s :      3: predicate.mutable_eliminate
 0.33% :     0.000001s :      3: predicate.opt_reshape
 1.37% :     0.000002s :     15: predicate.partial_eliminate
 0.79% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.06% :     0.000002s :     10: predicate.reduce_eliminate
 0.97% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.34% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.16% :     0.000002s :     18: predicate.replace_applicator
 0.35% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.03% :     0.000002s :     10: predicate.reshape_eliminate
 0.83% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.82% :     0.000001s :     10: predicate.same_eliminate
 0.37% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000002s :      6: predicate.special_op_eliminate
 0.63% :     0.000001s :      6: predicate.specialize_transform
 0.92% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 0.77% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.27% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000004s :     16: predicate.switch_defer_inline
 1.45% :     0.000002s :     16: predicate.switch_layer_defer_inline
36.65% :     0.000063s :     53: predicate.switch_simplify
 0.83% :     0.000001s :     10: predicate.tile_eliminate
 0.87% :     0.000001s :     10: predicate.transpose_eliminate
 0.98% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 0.90% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 2.51% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 0.99% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.10% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 1.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 0.96% :     0.000002s :     10: predicate.value_based_eliminate
 0.21% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.50% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000696    12
47.85% :     0.000333s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.26% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.89% :     0.000326s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022496    76
 0.16% :     0.000037s :      1: add_recomputation
 0.35% :     0.000079s :      1: auto_monad
 0.05% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.85% :     0.000415s :      1: bootstrap
 0.09% :     0.000020s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.11% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.02% :     0.005854s :      1: jit_opt_a
 0.65% :     0.000146s :      1: jit_opt_after_cconv
 0.25% :     0.000055s :      1: jit_opt_b
 2.00% :     0.000451s :      1: loop_unroll
 2.86% :     0.000644s :      1: mutable_eliminate
 4.19% :     0.000942s :     26: opt.transform.jit_opt_a
 0.21% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.38% :     0.000534s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.01% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000054s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000010s :      1: remove_dup_value
 1.98% :     0.000445s :      1: renormalize.infer
 1.64% :     0.000369s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000033s :      1: rewriter_after_opt_a
 0.33% :     0.000075s :      1: rewriter_before_opt_a
 0.37% :     0.000082s :      1: symbol_engine_optimizer
29.15% :     0.006558s :      1: task_emit
23.70% :     0.005332s :      1: type_inference
 0.28% :     0.000062s :      1: validate


TotalTime = 0.013816, [30]
[bootstrap]: 0.00036201
[type_inference]: 0.00460835
[event_method]: 1.729e-05
[auto_monad]: 6.45e-05
[graph_reusing]: 6.68998e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 3.547e-05
[rewriter_before_opt_a]: 7.372e-05
[expand_dump_flag]: 2.97002e-06
[jit_opt_a]: 0.00630805, [2]
    [Cycle 1]: 0.00188234, [27]
        [switch_simplify]: 6.402e-05
        [loop_unroll]: 3.063e-05
        [a_1]: 0.00060297
        [with_stream_mark]: 2.119e-05
        [recompute_prepare]: 7.31999e-06
        [updatestate_depend_eliminate]: 3.48e-06
        [updatestate_assign_eliminate]: 2.87002e-06
        [updatestate_loads_eliminate]: 3.03e-06
        [parameter_eliminate]: 2.19999e-06
        [specialize_transform]: 6.49001e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 5.92001e-06
        [meta_shard_fg_expand]: 2.11e-06
        [get_grad_eliminate_]: 5.17e-06
        [merge_forward]: 3.64002e-06
        [cell_reuse_recompute_pass]: 1.36002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.532e-05
        [j_node_and_user_rematch]: 9.39e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 9.56003e-06
        [inline_without_move]: 5.48002e-06
        [renormalize]: 0.00082903
        [add_forward_monad_depend]: 5.75001e-06
        [auto_monad_grad]: 3.03e-06
        [auto_monad_eliminator]: 1.435e-05
        [cse]: 2.737e-05
        [replace_applicator]: 1.281e-05
    [Cycle 2]: 0.00032841, [27]
        [switch_simplify]: 5.52999e-06
        [loop_unroll]: 5.10001e-06
        [a_1]: 9.056e-05
        [with_stream_mark]: 1.035e-05
        [recompute_prepare]: 5.39e-06
        [updatestate_depend_eliminate]: 2.64999e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 5.22e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.24e-06
        [meta_shard_fg_expand]: 1.24e-06
        [get_grad_eliminate_]: 4.64002e-06
        [merge_forward]: 2.67001e-06
        [cell_reuse_recompute_pass]: 2.02999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.367e-05
        [j_node_and_user_rematch]: 8.40999e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 1.447e-05
        [inline_without_move]: 4.67e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.33002e-06
        [auto_monad_grad]: 1.17e-06
        [auto_monad_eliminator]: 5.19998e-06
        [cse]: 1.164e-05
        [replace_applicator]: 4.97999e-06
[py_interpret_to_execute_after_opt_a]: 1.442e-05
[rewriter_after_opt_a]: 3.756e-05
[convert_after_rewriter]: 6.88e-06
[order_py_execute_after_rewriter]: 5.42001e-06
[mutable_eliminate]: 0.00062225
[jit_opt_b]: 5.276e-05, [1]
    [Cycle 1]: 4.538e-05, [2]
        [frontend_op_eliminate]: 1.748e-05
        [inline_after_opt_a]: 1.559e-05
[cconv]: 2.546e-05
[loop_unroll]: 0.00044245
[jit_opt_after_cconv]: 0.00014729, [1]
    [Cycle 1]: 0.00014065, [11]
        [c_1]: 2.207e-05
        [parameter_eliminate]: 2.70002e-06
        [updatestate_depend_eliminate]: 5.40999e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 2.13002e-06
        [cse]: 2.118e-05
        [call_graph_tuple_transform]: 1.965e-05
        [tuple_list_get_item_eliminator]: 5.79999e-06
        [none_parameter_eliminate]: 1.85001e-06
        [renormalize]: 7.90023e-07
        [switch_simplify]: 5.59998e-06
[remove_dup_value]: 1.416e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 5.97001e-06
[add_recomputation]: 4.826e-05
[cse_after_recomputation]: 2.022e-05, [1]
    [Cycle 1]: 1.451e-05, [1]
        [cse]: 8.82999e-06
[auto_monad_reorder]: 1.565e-05
[get_jit_bprop_graph]: 2.00002e-06
[rewriter_after_jit_bprop_graph]: 6.26e-06
[opt_after_jit_grad]: 0.000512
[symbol_engine_optimizer]: 9.589e-05, [1]
    [Cycle 1]: 8.737e-05, [6]
        [build]: 5.13002e-06
        [elim_shapecalc]: 9.15999e-06
        [elim_not_effective]: 2.529e-05
        [opt_reshape]: 7.25e-06
        [fold_const_symbol]: 9.51e-06
        [renormalize]: 5.3001e-07
[validate]: 4.389e-05
Sums
    bootstrap                            :     0.000362s :  3.98%
    type_inference                       :     0.004608s : 50.72%
    event_method                         :     0.000017s :  0.19%
    auto_monad                           :     0.000064s :  0.71%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.39%
    rewriter_before_opt_a                :     0.000074s :  0.81%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000070s :  0.77%
    jit_opt_a.loop_unroll                :     0.000036s :  0.39%
    jit_opt_a.a_1                        :     0.000694s :  7.63%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000024s :  0.26%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000829s :  9.12%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.22%
    jit_opt_a.cse                        :     0.000039s :  0.43%
    jit_opt_a.replace_applicator         :     0.000018s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.16%
    rewriter_after_opt_a                 :     0.000038s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000622s :  6.85%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.17%
    cconv                                :     0.000025s :  0.28%
    loop_unroll                          :     0.000442s :  4.87%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000014s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000048s :  0.53%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000512s :  5.63%
    symbol_engine_optimizer.build        :     0.000005s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000025s :  0.28%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.48%


Time group info:
------[substitution.]   0.000252    23
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.52% :     0.000001s :      2: substitution.fold_const_symbol
 2.10% :     0.000005s :      3: substitution.graph_param_transform
86.26% :     0.000218s :      4: substitution.inline
 1.22% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.03% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.74% :     0.000004s :      2: substitution.replace_old_param
 5.37% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004539     2
79.96% :     0.003630s :      1: type_inference.infer
20.04% :     0.000910s :      1: type_inference.specialize
------[replace.]   0.000076     6
68.61% :     0.000052s :      4: replace.inline
31.39% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000227     6
94.55% :     0.000215s :      4: match.inline
 5.45% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.23% :     0.000002s :     10: predicate.addn_zero_filter
 2.29% :     0.000003s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.41% :     0.000002s :     10: predicate.depend_value_elim
 1.61% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.45% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.06% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.14% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.87% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.83% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.64% :     0.000007s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.34% :     0.000002s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.93% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.68% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.73% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 1.28% :     0.000002s :      6: predicate.specialize_transform
 1.28% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.10% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.77% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.75% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000002s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.97% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.92% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.86% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000681    12
44.05% :     0.000300s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.75% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.19% :     0.000328s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015637    72
 0.33% :     0.000051s :      1: add_recomputation
 0.43% :     0.000068s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 2.40% :     0.000375s :      1: bootstrap
 0.18% :     0.000028s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.14% :     0.000022s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
40.36% :     0.006312s :      1: jit_opt_a
 0.96% :     0.000150s :      1: jit_opt_after_cconv
 0.35% :     0.000055s :      1: jit_opt_b
 2.88% :     0.000451s :      1: loop_unroll
 4.27% :     0.000668s :      1: mutable_eliminate
 5.91% :     0.000925s :     26: opt.transform.jit_opt_a
 0.31% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.30% :     0.000047s :      4: opt.transform.symbol_engine_opt
 3.33% :     0.000521s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000038s :      1: py_interpret_to_execute
 0.11% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 2.85% :     0.000446s :      1: renormalize.infer
 2.40% :     0.000375s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000041s :      1: rewriter_after_opt_a
 0.49% :     0.000076s :      1: rewriter_before_opt_a
 0.63% :     0.000099s :      1: symbol_engine_optimizer
29.58% :     0.004626s :      1: type_inference


TotalTime = 0.0220097, [33]
[bootstrap]: 0.00033891
[type_inference]: 0.00516174
[event_method]: 1.821e-05
[auto_monad]: 6.444e-05
[graph_reusing]: 5.22999e-06
[pre_auto_parallel]: 2.76e-06
[py_interpret_to_execute]: 4.388e-05
[rewriter_before_opt_a]: 7.23e-05
[expand_dump_flag]: 2.99001e-06
[jit_opt_a]: 0.0063168, [2]
    [Cycle 1]: 0.00211051, [27]
        [switch_simplify]: 6.414e-05
        [loop_unroll]: 2.967e-05
        [a_1]: 0.00059472
        [with_stream_mark]: 2.298e-05
        [recompute_prepare]: 7.87003e-06
        [updatestate_depend_eliminate]: 3.33998e-06
        [updatestate_assign_eliminate]: 3.21999e-06
        [updatestate_loads_eliminate]: 2.43998e-06
        [parameter_eliminate]: 2.15002e-06
        [specialize_transform]: 6.19001e-06
        [updatestate_useless_node_eliminater]: 5.10001e-06
        [accelerated_algorithm]: 5.66e-06
        [meta_shard_fg_expand]: 2.43e-06
        [get_grad_eliminate_]: 5.81998e-06
        [merge_forward]: 3.65998e-06
        [cell_reuse_recompute_pass]: 1.72999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.359e-05
        [j_node_and_user_rematch]: 9.82999e-06
        [meta_fg_expand]: 2.60002e-06
        [replace_old_param]: 9.55001e-06
        [inline_without_move]: 5.61998e-06
        [renormalize]: 0.00105483
        [add_forward_monad_depend]: 7.71999e-06
        [auto_monad_grad]: 2.94999e-06
        [auto_monad_eliminator]: 1.628e-05
        [cse]: 2.864e-05
        [replace_applicator]: 1.827e-05
    [Cycle 2]: 0.00037075, [27]
        [switch_simplify]: 6.46e-06
        [loop_unroll]: 5.07999e-06
        [a_1]: 9.778e-05
        [with_stream_mark]: 1.341e-05
        [recompute_prepare]: 5.86e-06
        [updatestate_depend_eliminate]: 3.16999e-06
        [updatestate_assign_eliminate]: 2.52001e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.60001e-06
        [specialize_transform]: 5.30999e-06
        [updatestate_useless_node_eliminater]: 5.62999e-06
        [accelerated_algorithm]: 5.29e-06
        [meta_shard_fg_expand]: 1.82001e-06
        [get_grad_eliminate_]: 5.84999e-06
        [merge_forward]: 4.28999e-06
        [cell_reuse_recompute_pass]: 2.99999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.683e-05
        [j_node_and_user_rematch]: 9.69e-06
        [meta_fg_expand]: 1.74998e-06
        [replace_old_param]: 8.70001e-06
        [inline_without_move]: 5.66e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.10002e-06
        [auto_monad_grad]: 1.42e-06
        [auto_monad_eliminator]: 7.31001e-06
        [cse]: 1.461e-05
        [replace_applicator]: 7.03e-06
[py_interpret_to_execute_after_opt_a]: 1.413e-05
[rewriter_after_opt_a]: 4.471e-05
[convert_after_rewriter]: 7.80998e-06
[order_py_execute_after_rewriter]: 4.70999e-06
[mutable_eliminate]: 0.00074582
[jit_opt_b]: 7.178e-05, [1]
    [Cycle 1]: 6.426e-05, [2]
        [frontend_op_eliminate]: 1.838e-05
        [inline_after_opt_a]: 3.306e-05
[cconv]: 2.518e-05
[loop_unroll]: 0.00053059
[jit_opt_after_cconv]: 0.00016457, [1]
    [Cycle 1]: 0.00015649, [11]
        [c_1]: 2.257e-05
        [parameter_eliminate]: 3.63999e-06
        [updatestate_depend_eliminate]: 8.13999e-06
        [updatestate_assign_eliminate]: 3.27002e-06
        [updatestate_loads_eliminate]: 2.74001e-06
        [cse]: 2.505e-05
        [call_graph_tuple_transform]: 2.218e-05
        [tuple_list_get_item_eliminator]: 6.06998e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 9.39996e-07
        [switch_simplify]: 5.85002e-06
[remove_dup_value]: 1.465e-05
[partial_unused_args_eliminate]: 2.86e-06
[environ_conv]: 5.54e-06
[add_recomputation]: 5.397e-05
[cse_after_recomputation]: 2.622e-05, [1]
    [Cycle 1]: 1.906e-05, [1]
        [cse]: 1.103e-05
[auto_monad_reorder]: 1.797e-05
[get_jit_bprop_graph]: 2.37999e-06
[rewriter_after_jit_bprop_graph]: 7.6e-06
[opt_after_jit_grad]: 0.00055121
[symbol_engine_optimizer]: 8.2e-05, [1]
    [Cycle 1]: 7.374e-05, [6]
        [build]: 5.09998e-06
        [elim_shapecalc]: 9.60001e-06
        [elim_not_effective]: 1.412e-05
        [opt_reshape]: 6.11998e-06
        [fold_const_symbol]: 9.22001e-06
        [renormalize]: 7.60017e-07
[validate]: 4.188e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.00733108
[execute]: 7.73001e-06
Sums
    bootstrap                            :     0.000339s :  1.94%
    type_inference                       :     0.005162s : 29.49%
    event_method                         :     0.000018s :  0.10%
    auto_monad                           :     0.000064s :  0.37%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000044s :  0.25%
    rewriter_before_opt_a                :     0.000072s :  0.41%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000071s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.20%
    jit_opt_a.a_1                        :     0.000693s :  3.96%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.07%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001055s :  6.03%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.13%
    jit_opt_a.cse                        :     0.000043s :  0.25%
    jit_opt_a.replace_applicator         :     0.000025s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.08%
    rewriter_after_opt_a                 :     0.000045s :  0.26%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000746s :  4.26%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000033s :  0.19%
    cconv                                :     0.000025s :  0.14%
    loop_unroll                          :     0.000531s :  3.03%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000025s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000054s :  0.31%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000018s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000551s :  3.15%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007331s : 41.88%
    execute                              :     0.000008s :  0.04%


Time group info:
------[substitution.]   0.000252    23
 0.73% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000002s :      2: substitution.fold_const_symbol
 2.59% :     0.000007s :      3: substitution.graph_param_transform
84.77% :     0.000214s :      4: substitution.inline
 1.82% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.01% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.93% :     0.000005s :      2: substitution.replace_old_param
 5.43% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005090     2
81.21% :     0.004134s :      1: type_inference.infer
18.79% :     0.000957s :      1: type_inference.specialize
------[replace.]   0.000076     6
64.63% :     0.000049s :      4: replace.inline
35.37% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000224     6
94.41% :     0.000211s :      4: match.inline
 5.59% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   691
 1.05% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.93% :     0.000001s :     10: predicate.addn_check_dump
 1.19% :     0.000002s :     10: predicate.addn_zero_filter
 2.27% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000002s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.93% :     0.000001s :     10: predicate.compare_switch_simplify
 1.59% :     0.000002s :     10: predicate.depend_value_elim
 1.26% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.49% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.73% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 6.28% :     0.000008s :     22: predicate.inline
 1.12% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      6: predicate.less_batch_normalization
 1.34% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.17% :     0.000003s :     15: predicate.load_eliminater
 2.10% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.85% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.92% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 2.45% :     0.000003s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 1.93% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.88% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000002s :     10: predicate.reshape_eliminate
 1.38% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.33% :     0.000002s :      6: predicate.special_op_eliminate
 0.83% :     0.000001s :      6: predicate.specialize_transform
 1.18% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 1.87% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.58% :     0.000010s :     53: predicate.switch_simplify
 1.25% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.08% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.12% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.28% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.05% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000002s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000707    12
44.88% :     0.000317s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.22% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.90% :     0.000338s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024127    76
 0.24% :     0.000058s :      1: add_recomputation
 0.28% :     0.000068s :      1: auto_monad
 0.09% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.46% :     0.000352s :      1: bootstrap
 0.12% :     0.000028s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000029s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
26.20% :     0.006321s :      1: jit_opt_a
 0.70% :     0.000168s :      1: jit_opt_after_cconv
 0.31% :     0.000074s :      1: jit_opt_b
 2.24% :     0.000541s :      1: loop_unroll
 3.13% :     0.000755s :      1: mutable_eliminate
 3.86% :     0.000931s :     26: opt.transform.jit_opt_a
 0.22% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000044s :      4: opt.transform.jit_opt_b
 0.07% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.34% :     0.000564s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000047s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.51% :     0.000605s :      1: renormalize.infer
 1.82% :     0.000440s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000049s :      1: rewriter_after_opt_a
 0.31% :     0.000075s :      1: rewriter_before_opt_a
 0.35% :     0.000085s :      1: symbol_engine_optimizer
30.46% :     0.007349s :      1: task_emit
21.46% :     0.005177s :      1: type_inference
 0.26% :     0.000063s :      1: validate


TotalTime = 0.0143102, [30]
[bootstrap]: 0.00035077
[type_inference]: 0.00506079
[event_method]: 1.818e-05
[auto_monad]: 6.904e-05
[graph_reusing]: 6.28e-06
[pre_auto_parallel]: 2.62001e-06
[py_interpret_to_execute]: 3.87e-05
[rewriter_before_opt_a]: 7.832e-05
[expand_dump_flag]: 3.63e-06
[jit_opt_a]: 0.00617599, [2]
    [Cycle 1]: 0.00184875, [27]
        [switch_simplify]: 6.217e-05
        [loop_unroll]: 3e-05
        [a_1]: 0.00059853
        [with_stream_mark]: 2.004e-05
        [recompute_prepare]: 8.69998e-06
        [updatestate_depend_eliminate]: 3.66999e-06
        [updatestate_assign_eliminate]: 2.88998e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 6.04999e-06
        [updatestate_useless_node_eliminater]: 5.22999e-06
        [accelerated_algorithm]: 5.99999e-06
        [meta_shard_fg_expand]: 2.51e-06
        [get_grad_eliminate_]: 6.57002e-06
        [merge_forward]: 4.15e-06
        [cell_reuse_recompute_pass]: 1.65001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.705e-05
        [j_node_and_user_rematch]: 9.66e-06
        [meta_fg_expand]: 2.41e-06
        [replace_old_param]: 8.92e-06
        [inline_without_move]: 5.76998e-06
        [renormalize]: 0.0007857
        [add_forward_monad_depend]: 5.96998e-06
        [auto_monad_grad]: 2.37999e-06
        [auto_monad_eliminator]: 1.757e-05
        [cse]: 2.843e-05
        [replace_applicator]: 1.565e-05
    [Cycle 2]: 0.00034939, [27]
        [switch_simplify]: 6.12001e-06
        [loop_unroll]: 5.02e-06
        [a_1]: 9.4e-05
        [with_stream_mark]: 1.158e-05
        [recompute_prepare]: 5.09e-06
        [updatestate_depend_eliminate]: 3.26001e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [parameter_eliminate]: 1.04998e-06
        [specialize_transform]: 5.23002e-06
        [updatestate_useless_node_eliminater]: 5.37001e-06
        [accelerated_algorithm]: 5.64998e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 4.44998e-06
        [merge_forward]: 2.49001e-06
        [cell_reuse_recompute_pass]: 1.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.428e-05
        [j_node_and_user_rematch]: 8.59e-06
        [meta_fg_expand]: 1.77001e-06
        [replace_old_param]: 8.89e-06
        [inline_without_move]: 5.22e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 1.20999e-06
        [auto_monad_eliminator]: 6.02999e-06
        [cse]: 1.346e-05
        [replace_applicator]: 5.99e-06
[py_interpret_to_execute_after_opt_a]: 1.072e-05
[rewriter_after_opt_a]: 3.741e-05
[convert_after_rewriter]: 7.4e-06
[order_py_execute_after_rewriter]: 5.23002e-06
[mutable_eliminate]: 0.00069515
[jit_opt_b]: 6.41e-05, [1]
    [Cycle 1]: 5.573e-05, [2]
        [frontend_op_eliminate]: 2.509e-05
        [inline_after_opt_a]: 1.737e-05
[cconv]: 2.786e-05
[loop_unroll]: 0.00049855
[jit_opt_after_cconv]: 0.00016273, [1]
    [Cycle 1]: 0.00015493, [11]
        [c_1]: 2.271e-05
        [parameter_eliminate]: 3.51999e-06
        [updatestate_depend_eliminate]: 8.32e-06
        [updatestate_assign_eliminate]: 2.86e-06
        [updatestate_loads_eliminate]: 2.39001e-06
        [cse]: 2.484e-05
        [call_graph_tuple_transform]: 2.059e-05
        [tuple_list_get_item_eliminator]: 5.79e-06
        [none_parameter_eliminate]: 1.81e-06
        [renormalize]: 8.39995e-07
        [switch_simplify]: 5.84e-06
[remove_dup_value]: 1.534e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 6.07001e-06
[add_recomputation]: 4.836e-05
[cse_after_recomputation]: 2.366e-05, [1]
    [Cycle 1]: 1.669e-05, [1]
        [cse]: 9.62001e-06
[auto_monad_reorder]: 1.521e-05
[get_jit_bprop_graph]: 1.87001e-06
[rewriter_after_jit_bprop_graph]: 6.64001e-06
[opt_after_jit_grad]: 0.00051531
[symbol_engine_optimizer]: 0.00011712, [1]
    [Cycle 1]: 0.00010877, [6]
        [build]: 3.53e-06
        [elim_shapecalc]: 8.45999e-06
        [elim_not_effective]: 1.422e-05
        [opt_reshape]: 6.04001e-06
        [fold_const_symbol]: 4.265e-05
        [renormalize]: 5.69999e-07
[validate]: 4.316e-05
Sums
    bootstrap                            :     0.000351s :  3.62%
    type_inference                       :     0.005061s : 52.29%
    event_method                         :     0.000018s :  0.19%
    auto_monad                           :     0.000069s :  0.71%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000039s :  0.40%
    rewriter_before_opt_a                :     0.000078s :  0.81%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000068s :  0.71%
    jit_opt_a.loop_unroll                :     0.000035s :  0.36%
    jit_opt_a.a_1                        :     0.000693s :  7.16%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.18%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000786s :  8.12%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.24%
    jit_opt_a.cse                        :     0.000042s :  0.43%
    jit_opt_a.replace_applicator         :     0.000022s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.11%
    rewriter_after_opt_a                 :     0.000037s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000695s :  7.18%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.26%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.18%
    cconv                                :     0.000028s :  0.29%
    loop_unroll                          :     0.000499s :  5.15%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.09%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000025s :  0.26%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000048s :  0.50%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000515s :  5.32%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000043s :  0.44%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.45%


Time group info:
------[substitution.]   0.000214    23
 1.33% :     0.000003s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.57% :     0.000005s :      3: substitution.graph_param_transform
82.73% :     0.000177s :      4: substitution.inline
 1.77% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.45% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.48% :     0.000005s :      2: substitution.replace_old_param
 6.09% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004991     2
80.34% :     0.004010s :      1: type_inference.infer
19.66% :     0.000981s :      1: type_inference.specialize
------[replace.]   0.000085     6
61.52% :     0.000052s :      4: replace.inline
38.48% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000186     6
93.71% :     0.000175s :      4: match.inline
 6.29% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000145   691
 1.04% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.36% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.82% :     0.000001s :     10: predicate.addn_check_dump
 1.13% :     0.000002s :     10: predicate.addn_zero_filter
 2.23% :     0.000003s :     10: predicate.arithmetic_simplify
 1.13% :     0.000002s :     10: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.84% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 0.97% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.09% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.04% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.24% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.86% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.85% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.01% :     0.000001s :     10: predicate.environ_get_eliminate
 0.86% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.78% :     0.000001s :      6: predicate.get_grad_eliminate
 0.20% :     0.000000s :      3: predicate.graph_param_transform
 4.45% :     0.000006s :     22: predicate.inline
 1.15% :     0.000002s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000002s :      6: predicate.less_batch_normalization
 1.70% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.46% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.28% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.69% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 0.84% :     0.000001s :     10: predicate.minmaximum_grad
 1.62% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 1.66% :     0.000002s :     15: predicate.partial_eliminate
 0.96% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.31% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.46% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
13.56% :     0.000020s :     10: predicate.reshape_eliminate
 1.42% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.93% :     0.000001s :     10: predicate.same_eliminate
 0.41% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.84% :     0.000001s :      6: predicate.special_op_eliminate
 0.82% :     0.000001s :      6: predicate.specialize_transform
 1.07% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 0.91% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.32% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.15% :     0.000005s :     16: predicate.switch_defer_inline
 1.75% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.55% :     0.000011s :     53: predicate.switch_simplify
 1.15% :     0.000002s :     10: predicate.tile_eliminate
 1.18% :     0.000002s :     10: predicate.transpose_eliminate
 1.18% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.02% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.97% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.20% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.20% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.21% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.29% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000676    12
49.45% :     0.000334s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.38% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.17% :     0.000312s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016085    72
 0.33% :     0.000052s :      1: add_recomputation
 0.46% :     0.000074s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 2.29% :     0.000368s :      1: bootstrap
 0.19% :     0.000031s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
38.42% :     0.006179s :      1: jit_opt_a
 1.03% :     0.000166s :      1: jit_opt_after_cconv
 0.42% :     0.000068s :      1: jit_opt_b
 3.18% :     0.000511s :      1: loop_unroll
 4.40% :     0.000708s :      1: mutable_eliminate
 5.74% :     0.000924s :     26: opt.transform.jit_opt_a
 0.31% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000035s :      4: opt.transform.jit_opt_b
 0.10% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.26% :     0.000525s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000042s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.52% :     0.000405s :      1: renormalize.infer
 2.31% :     0.000372s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000042s :      1: rewriter_after_opt_a
 0.52% :     0.000083s :      1: rewriter_before_opt_a
 0.75% :     0.000120s :      1: symbol_engine_optimizer
31.57% :     0.005078s :      1: type_inference


TotalTime = 0.0195115, [33]
[bootstrap]: 0.00029305
[type_inference]: 0.00479983
[event_method]: 1.69e-05
[auto_monad]: 6.456e-05
[graph_reusing]: 5.72001e-06
[pre_auto_parallel]: 2.33002e-06
[py_interpret_to_execute]: 3.573e-05
[rewriter_before_opt_a]: 7.13e-05
[expand_dump_flag]: 3.09001e-06
[jit_opt_a]: 0.00522289, [2]
    [Cycle 1]: 0.00157184, [27]
        [switch_simplify]: 4.483e-05
        [loop_unroll]: 2.932e-05
        [a_1]: 0.00052518
        [with_stream_mark]: 1.502e-05
        [recompute_prepare]: 7.65e-06
        [updatestate_depend_eliminate]: 2.94999e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 6.47001e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 1.81e-06
        [get_grad_eliminate_]: 4.99998e-06
        [merge_forward]: 3.46001e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.273e-05
        [j_node_and_user_rematch]: 8.64e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 8.46997e-06
        [inline_without_move]: 4.99998e-06
        [renormalize]: 0.00066207
        [add_forward_monad_depend]: 6.23e-06
        [auto_monad_grad]: 2.61999e-06
        [auto_monad_eliminator]: 1.268e-05
        [cse]: 2.03e-05
        [replace_applicator]: 1.286e-05
    [Cycle 2]: 0.00032446, [27]
        [switch_simplify]: 5.52001e-06
        [loop_unroll]: 5.09e-06
        [a_1]: 8.855e-05
        [with_stream_mark]: 9.89999e-06
        [recompute_prepare]: 5.04e-06
        [updatestate_depend_eliminate]: 2.93e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 5.56e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 5.56998e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.63001e-06
        [merge_forward]: 2.60997e-06
        [cell_reuse_recompute_pass]: 1.74e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.321e-05
        [j_node_and_user_rematch]: 7.88001e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 7.82e-06
        [inline_without_move]: 5.16002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.47001e-06
        [auto_monad_grad]: 1.21002e-06
        [auto_monad_eliminator]: 6.24999e-06
        [cse]: 1.033e-05
        [replace_applicator]: 5.11002e-06
[py_interpret_to_execute_after_opt_a]: 1.177e-05
[rewriter_after_opt_a]: 3.686e-05
[convert_after_rewriter]: 6.79999e-06
[order_py_execute_after_rewriter]: 4.75999e-06
[mutable_eliminate]: 0.00064097
[jit_opt_b]: 4.927e-05, [1]
    [Cycle 1]: 4.146e-05, [2]
        [frontend_op_eliminate]: 1.612e-05
        [inline_after_opt_a]: 1.412e-05
[cconv]: 2.612e-05
[loop_unroll]: 0.00044405
[jit_opt_after_cconv]: 0.00015046, [1]
    [Cycle 1]: 0.00014399, [11]
        [c_1]: 2.118e-05
        [parameter_eliminate]: 3.59002e-06
        [updatestate_depend_eliminate]: 5.56e-06
        [updatestate_assign_eliminate]: 2.59999e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [cse]: 2.339e-05
        [call_graph_tuple_transform]: 2.141e-05
        [tuple_list_get_item_eliminator]: 5.26002e-06
        [none_parameter_eliminate]: 2.00002e-06
        [renormalize]: 9.89996e-07
        [switch_simplify]: 5.40999e-06
[remove_dup_value]: 1.405e-05
[partial_unused_args_eliminate]: 1.95001e-06
[environ_conv]: 4.94e-06
[add_recomputation]: 4.836e-05
[cse_after_recomputation]: 2.154e-05, [1]
    [Cycle 1]: 1.574e-05, [1]
        [cse]: 9.79e-06
[auto_monad_reorder]: 1.483e-05
[get_jit_bprop_graph]: 2.02001e-06
[rewriter_after_jit_bprop_graph]: 7.66999e-06
[opt_after_jit_grad]: 0.00053948
[symbol_engine_optimizer]: 7.733e-05, [1]
    [Cycle 1]: 6.992e-05, [6]
        [build]: 3.55e-06
        [elim_shapecalc]: 8.05999e-06
        [elim_not_effective]: 1.415e-05
        [opt_reshape]: 5.83002e-06
        [fold_const_symbol]: 9.45001e-06
        [renormalize]: 8.90024e-07
[validate]: 3.558e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00666223
[execute]: 6.65002e-06
Sums
    bootstrap                            :     0.000293s :  1.88%
    type_inference                       :     0.004800s : 30.77%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000065s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000036s :  0.23%
    rewriter_before_opt_a                :     0.000071s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000050s :  0.32%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000614s :  3.93%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000662s :  4.24%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000031s :  0.20%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000037s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000641s :  4.11%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000026s :  0.17%
    loop_unroll                          :     0.000444s :  2.85%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000048s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000539s :  3.46%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000036s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006662s : 42.71%
    execute                              :     0.000007s :  0.04%


Time group info:
------[substitution.]   0.000201    23
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000002s :      2: substitution.fold_const_symbol
 2.82% :     0.000006s :      3: substitution.graph_param_transform
83.63% :     0.000168s :      4: substitution.inline
 1.41% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.16% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.21% :     0.000004s :      2: substitution.replace_old_param
 6.14% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004738     2
81.05% :     0.003840s :      1: type_inference.infer
18.95% :     0.000898s :      1: type_inference.specialize
------[replace.]   0.000068     6
67.05% :     0.000046s :      4: replace.inline
32.95% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000177     6
93.61% :     0.000165s :      4: match.inline
 6.39% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.26% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.87% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.77% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.81% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 4.81% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.39% :     0.000002s :      6: predicate.less_batch_normalization
 1.66% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.57% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.29% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.61% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.79% :     0.000002s :     10: predicate.reduce_eliminate
 1.72% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000001s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.19% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.56% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.47% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.48% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.93% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000658    12
49.79% :     0.000328s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.10% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.12% :     0.000297s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021100    76
 0.24% :     0.000052s :      1: add_recomputation
 0.32% :     0.000068s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.44% :     0.000304s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
24.77% :     0.005226s :      1: jit_opt_a
 0.73% :     0.000153s :      1: jit_opt_after_cconv
 0.25% :     0.000052s :      1: jit_opt_b
 2.14% :     0.000452s :      1: loop_unroll
 3.08% :     0.000651s :      1: mutable_eliminate
 3.87% :     0.000816s :     26: opt.transform.jit_opt_a
 0.24% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.60% :     0.000548s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000038s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.65% :     0.000347s :      1: renormalize.infer
 1.45% :     0.000307s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000041s :      1: rewriter_after_opt_a
 0.35% :     0.000075s :      1: rewriter_before_opt_a
 0.38% :     0.000080s :      1: symbol_engine_optimizer
31.62% :     0.006673s :      1: task_emit
22.80% :     0.004811s :      1: type_inference
 0.27% :     0.000057s :      1: validate


TotalTime = 0.163508, [33]
[bootstrap]: 0.00046305
[type_inference]: 0.13412
[event_method]: 0.00025488
[auto_monad]: 0.00010661
[graph_reusing]: 7.83001e-06
[pre_auto_parallel]: 3.14999e-06
[py_interpret_to_execute]: 4.547e-05
[rewriter_before_opt_a]: 0.00010499
[expand_dump_flag]: 3.58e-06
[jit_opt_a]: 0.0261133, [3]
    [Cycle 1]: 0.0176914, [27]
        [switch_simplify]: 7.154e-05
        [loop_unroll]: 4.308e-05
        [a_1]: 0.00095214
        [with_stream_mark]: 2.632e-05
        [recompute_prepare]: 1.99e-05
        [updatestate_depend_eliminate]: 8.79003e-06
        [updatestate_assign_eliminate]: 6.61e-06
        [updatestate_loads_eliminate]: 6.86999e-06
        [parameter_eliminate]: 2.57001e-06
        [specialize_transform]: 1.478e-05
        [updatestate_useless_node_eliminater]: 1.378e-05
        [accelerated_algorithm]: 1.367e-05
        [meta_shard_fg_expand]: 4.34002e-06
        [get_grad_eliminate_]: 1.385e-05
        [merge_forward]: 9.09998e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.782e-05
        [j_node_and_user_rematch]: 2.478e-05
        [meta_fg_expand]: 0.00159043
        [replace_old_param]: 6.017e-05
        [inline_without_move]: 5.578e-05
        [renormalize]: 0.0141402
        [add_forward_monad_depend]: 1.724e-05
        [auto_monad_grad]: 6.44999e-06
        [auto_monad_eliminator]: 5.263e-05
        [cse]: 0.00020731
        [replace_applicator]: 8.015e-05
    [Cycle 2]: 0.00399308, [27]
        [switch_simplify]: 4.213e-05
        [loop_unroll]: 4.102e-05
        [a_1]: 0.00118527
        [with_stream_mark]: 1.509e-05
        [recompute_prepare]: 9.39998e-06
        [updatestate_depend_eliminate]: 4.94998e-06
        [updatestate_assign_eliminate]: 3.55003e-06
        [updatestate_loads_eliminate]: 3.95998e-06
        [parameter_eliminate]: 1.98002e-06
        [specialize_transform]: 8.33999e-06
        [updatestate_useless_node_eliminater]: 7.82998e-06
        [accelerated_algorithm]: 7.21999e-06
        [meta_shard_fg_expand]: 2.60002e-06
        [get_grad_eliminate_]: 7.51999e-06
        [merge_forward]: 4.08001e-06
        [cell_reuse_recompute_pass]: 8.60018e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.578e-05
        [j_node_and_user_rematch]: 1.187e-05
        [meta_fg_expand]: 6.108e-05
        [replace_old_param]: 7.96001e-06
        [inline_without_move]: 7.12002e-06
        [renormalize]: 0.0023426
        [add_forward_monad_depend]: 5.12e-06
        [auto_monad_grad]: 1.72001e-06
        [auto_monad_eliminator]: 8.95999e-06
        [cse]: 1.641e-05
        [replace_applicator]: 1.164e-05
    [Cycle 3]: 0.00023266, [27]
        [switch_simplify]: 4.58999e-06
        [loop_unroll]: 3.76999e-06
        [a_1]: 3.901e-05
        [with_stream_mark]: 8.57998e-06
        [recompute_prepare]: 3.8e-06
        [updatestate_depend_eliminate]: 2.01998e-06
        [updatestate_assign_eliminate]: 1.64e-06
        [updatestate_loads_eliminate]: 1.52001e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 3.56999e-06
        [updatestate_useless_node_eliminater]: 3.3e-06
        [accelerated_algorithm]: 3.11001e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 3.31999e-06
        [merge_forward]: 1.86998e-06
        [cell_reuse_recompute_pass]: 1.71e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.92001e-06
        [j_node_and_user_rematch]: 5.73002e-06
        [meta_fg_expand]: 1.45001e-06
        [replace_old_param]: 3.34001e-06
        [inline_without_move]: 3.05998e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.04998e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 3.57002e-06
        [cse]: 8.27e-06
        [replace_applicator]: 3.45e-06
[py_interpret_to_execute_after_opt_a]: 1.036e-05
[rewriter_after_opt_a]: 3.22e-05
[convert_after_rewriter]: 6.11e-06
[order_py_execute_after_rewriter]: 4.15999e-06
[mutable_eliminate]: 0.00068443
[jit_opt_b]: 4.72e-05, [1]
    [Cycle 1]: 3.843e-05, [2]
        [frontend_op_eliminate]: 1.303e-05
        [inline_after_opt_a]: 1.289e-05
[cconv]: 2.31e-05
[loop_unroll]: 0.00042522
[jit_opt_after_cconv]: 0.00011744, [1]
    [Cycle 1]: 0.00011023, [11]
        [c_1]: 1.294e-05
        [parameter_eliminate]: 2.58e-06
        [updatestate_depend_eliminate]: 4.54998e-06
        [updatestate_assign_eliminate]: 1.80001e-06
        [updatestate_loads_eliminate]: 1.44e-06
        [cse]: 1.554e-05
        [call_graph_tuple_transform]: 1.344e-05
        [tuple_list_get_item_eliminator]: 3.93001e-06
        [none_parameter_eliminate]: 1.37999e-06
        [renormalize]: 6.39993e-07
        [switch_simplify]: 4.08999e-06
[remove_dup_value]: 1.417e-05
[partial_unused_args_eliminate]: 1.94e-06
[environ_conv]: 4.55001e-06
[add_recomputation]: 4.345e-05
[cse_after_recomputation]: 1.919e-05, [1]
    [Cycle 1]: 1.374e-05, [1]
        [cse]: 7.61001e-06
[auto_monad_reorder]: 1.297e-05
[get_jit_bprop_graph]: 1.75001e-06
[rewriter_after_jit_bprop_graph]: 4.87998e-06
[opt_after_jit_grad]: 0.00044803
[symbol_engine_optimizer]: 6.275e-05, [1]
    [Cycle 1]: 5.653e-05, [6]
        [build]: 3.63999e-06
        [elim_shapecalc]: 5.99999e-06
        [elim_not_effective]: 9.95002e-06
        [opt_reshape]: 4.22e-06
        [fold_const_symbol]: 5.84e-06
        [renormalize]: 4.7998e-07
[validate]: 2.851e-05
[backend_pass]: 1.05001e-06
[task_emit]: 2.697e-05
[execute]: 1.24e-06
Sums
    bootstrap                            :     0.000463s :  0.29%
    type_inference                       :     0.134120s : 84.65%
    event_method                         :     0.000255s :  0.16%
    auto_monad                           :     0.000107s :  0.07%
    graph_reusing                        :     0.000008s :  0.00%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000045s :  0.03%
    rewriter_before_opt_a                :     0.000105s :  0.07%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000118s :  0.07%
    jit_opt_a.loop_unroll                :     0.000088s :  0.06%
    jit_opt_a.a_1                        :     0.002176s :  1.37%
    jit_opt_a.with_stream_mark           :     0.000050s :  0.03%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.02%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.00%
    jit_opt_a.specialize_transform       :     0.000027s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.02%
    jit_opt_a.merge_forward              :     0.000015s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.00%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000054s :  0.03%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.001653s :  1.04%
    jit_opt_a.replace_old_param          :     0.000071s :  0.05%
    jit_opt_a.inline_without_move        :     0.000066s :  0.04%
    jit_opt_a.renormalize                :     0.016483s : 10.40%
    jit_opt_a.add_forward_monad_depend   :     0.000023s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000065s :  0.04%
    jit_opt_a.cse                        :     0.000232s :  0.15%
    jit_opt_a.replace_applicator         :     0.000095s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.01%
    rewriter_after_opt_a                 :     0.000032s :  0.02%
    convert_after_rewriter               :     0.000006s :  0.00%
    order_py_execute_after_rewriter      :     0.000004s :  0.00%
    mutable_eliminate                    :     0.000684s :  0.43%
    jit_opt_b.frontend_op_eliminate      :     0.000013s :  0.01%
    jit_opt_b.inline_after_opt_a         :     0.000013s :  0.01%
    cconv                                :     0.000023s :  0.01%
    loop_unroll                          :     0.000425s :  0.27%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.01%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.00%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.cse              :     0.000016s :  0.01%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.01%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.00%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.00%
    remove_dup_value                     :     0.000014s :  0.01%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000005s :  0.00%
    add_recomputation                    :     0.000043s :  0.03%
    cse_after_recomputation.cse          :     0.000008s :  0.00%
    auto_monad_reorder                   :     0.000013s :  0.01%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.00%
    opt_after_jit_grad                   :     0.000448s :  0.28%
    symbol_engine_optimizer.build        :     0.000004s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.00%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.01%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.00%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.00%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.02%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000027s :  0.02%
    execute                              :     0.000001s :  0.00%


Time group info:
------[substitution.]   0.000531   102
 0.26% :     0.000001s :      1: substitution.elim_not_effective
 0.23% :     0.000001s :      1: substitution.fold_const_symbol
 0.70% :     0.000004s :      1: substitution.graph_param_transform
64.38% :     0.000342s :     12: substitution.inline
 3.09% :     0.000016s :      2: substitution.inline_without_move
 1.77% :     0.000009s :     12: substitution.j_node_and_user_rematch
 2.16% :     0.000011s :      7: substitution.minmaximum_grad
 1.37% :     0.000007s :      5: substitution.partial_eliminate
 1.79% :     0.000010s :     12: substitution.remove_not_recompute_node
 5.34% :     0.000028s :     10: substitution.replace_applicator
 1.04% :     0.000006s :      7: substitution.replace_old_param
 1.08% :     0.000006s :      1: substitution.reshape_eliminate
 0.47% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.62% :     0.000009s :      2: substitution.tile_eliminate
 3.97% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.70% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 8.03% :     0.000043s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.134011     2
97.85% :     0.131132s :      1: type_inference.infer
 2.15% :     0.002879s :      1: type_inference.specialize
------[replace.]   0.000158    19
65.10% :     0.000103s :     12: replace.inline
34.90% :     0.000055s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000357    19
93.80% :     0.000335s :     12: match.inline
 6.20% :     0.000022s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000366  2646
 1.68% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.22% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.48% :     0.000005s :     46: predicate.addn_check_dump
 1.82% :     0.000007s :     46: predicate.addn_zero_filter
 2.12% :     0.000008s :     46: predicate.arithmetic_simplify
 1.51% :     0.000006s :     46: predicate.cast_eliminate
 0.10% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.48% :     0.000005s :     46: predicate.compare_switch_simplify
 1.60% :     0.000006s :     46: predicate.depend_value_elim
 1.49% :     0.000005s :     46: predicate.dict_get_item_const_eliminator
 1.49% :     0.000005s :     46: predicate.dict_get_item_eliminator
 1.51% :     0.000006s :     46: predicate.dict_set_item_eliminator
 0.24% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.06% :     0.000000s :      1: predicate.elim_not_effective
 0.10% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.62% :     0.000006s :     46: predicate.environ_add_const_eliminate
 1.45% :     0.000005s :     46: predicate.environ_get_add_eliminate
 1.60% :     0.000006s :     46: predicate.environ_get_depend_swap
 1.54% :     0.000006s :     46: predicate.environ_get_eliminate
 1.59% :     0.000006s :     46: predicate.environ_get_set_eliminate
 0.03% :     0.000000s :      1: predicate.fold_const_symbol
 0.95% :     0.000003s :     19: predicate.get_grad_eliminate
 0.03% :     0.000000s :      1: predicate.graph_param_transform
 4.14% :     0.000015s :     67: predicate.inline
 1.89% :     0.000007s :     43: predicate.inline_without_move
 0.39% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.79% :     0.000003s :     19: predicate.less_batch_normalization
 1.94% :     0.000007s :     53: predicate.list_to_tuple_eliminator_
 1.85% :     0.000007s :     54: predicate.load_eliminater
 0.31% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.42% :     0.000012s :     95: predicate.loop_unroll_before_grad
 1.79% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.50% :     0.000005s :     46: predicate.merge_addn
 1.52% :     0.000006s :     46: predicate.minmaximum_grad
 0.36% :     0.000001s :      1: predicate.mutable_eliminate
 0.08% :     0.000000s :      1: predicate.opt_reshape
 2.35% :     0.000009s :     54: predicate.partial_eliminate
 1.59% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.90% :     0.000007s :     46: predicate.reduce_eliminate
 1.85% :     0.000007s :     53: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.83% :     0.000010s :    100: predicate.replace_applicator
 1.03% :     0.000004s :     43: predicate.replace_old_param
 0.07% :     0.000000s :      1: predicate.reset_defer_inline
 1.97% :     0.000007s :     46: predicate.reshape_eliminate
 1.55% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.19% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.75% :     0.000006s :     46: predicate.same_eliminate
 0.48% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.24% :     0.000001s :      2: predicate.special_op_eliminate
 0.86% :     0.000003s :     19: predicate.specialize_transform
 1.69% :     0.000006s :     46: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.07% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.88% :     0.000011s :     65: predicate.switch_defer_inline
 2.48% :     0.000009s :     65: predicate.switch_layer_defer_inline
 6.51% :     0.000024s :    161: predicate.switch_simplify
 1.61% :     0.000006s :     46: predicate.tile_eliminate
 1.53% :     0.000006s :     46: predicate.transpose_eliminate
 1.83% :     0.000007s :     46: predicate.tuple_list_convert_item_index_to_positive
 1.74% :     0.000006s :     46: predicate.tuple_list_get_item_depend_reorder
 3.21% :     0.000012s :     55: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000007s :     46: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000007s :     53: predicate.tuple_to_list_eliminator_
 1.79% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000010s :     73: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000007s :     46: predicate.value_based_eliminate
 0.06% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.13% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.004024    48
77.92% :     0.003136s :     32: func_graph_cloner_run.FuncGraphClonerGraph
22.08% :     0.000888s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.182845    91
 0.03% :     0.000047s :      1: add_recomputation
 0.06% :     0.000113s :      1: auto_monad
 0.01% :     0.000015s :      1: auto_monad_reorder
 0.00% :     0.000005s :      1: backend_pass
 0.28% :     0.000509s :      1: bootstrap
 0.01% :     0.000026s :      1: cconv
 0.00% :     0.000008s :      1: convert_after_rewriter
 0.01% :     0.000021s :      1: cse_after_recomputation
 0.00% :     0.000007s :      1: environ_conv
 0.14% :     0.000262s :      1: event_method
 0.00% :     0.000003s :      1: execute
 0.00% :     0.000007s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000011s :      1: graph_reusing
14.28% :     0.026117s :      1: jit_opt_a
 0.07% :     0.000120s :      1: jit_opt_after_cconv
 0.03% :     0.000050s :      1: jit_opt_b
 0.24% :     0.000433s :      1: loop_unroll
 0.38% :     0.000693s :      1: mutable_eliminate
 1.53% :     0.002800s :     39: opt.transform.jit_opt_a
 0.02% :     0.000031s :      4: opt.transform.jit_opt_after_cconv
 0.01% :     0.000019s :      4: opt.transform.jit_opt_b
 0.01% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.01% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.01% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.01% :     0.000022s :      4: opt.transform.symbol_engine_opt
 0.25% :     0.000456s :      1: opt_after_jit_grad
 0.00% :     0.000006s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000021s :      1: pre_auto_parallel
 0.03% :     0.000049s :      1: py_interpret_to_execute
 0.01% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.01% :     0.000017s :      1: remove_dup_value
 7.94% :     0.014525s :      2: renormalize.infer
 1.06% :     0.001941s :      2: renormalize.specialize
 0.00% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.02% :     0.000035s :      1: rewriter_after_opt_a
 0.06% :     0.000108s :      1: rewriter_before_opt_a
 0.04% :     0.000065s :      1: symbol_engine_optimizer
 0.02% :     0.000032s :      1: task_emit
73.36% :     0.134142s :      1: type_inference
 0.04% :     0.000065s :      1: validate


TotalTime = 0.0221179, [33]
[bootstrap]: 0.0003331
[type_inference]: 0.00525359
[event_method]: 1.953e-05
[auto_monad]: 6.782e-05
[graph_reusing]: 5.63002e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 4.701e-05
[rewriter_before_opt_a]: 7.87e-05
[expand_dump_flag]: 3.36001e-06
[jit_opt_a]: 0.0065148, [2]
    [Cycle 1]: 0.00219881, [27]
        [switch_simplify]: 6.782e-05
        [loop_unroll]: 2.986e-05
        [a_1]: 0.00065257
        [with_stream_mark]: 2.362e-05
        [recompute_prepare]: 9.47001e-06
        [updatestate_depend_eliminate]: 3.86999e-06
        [updatestate_assign_eliminate]: 3.06001e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [parameter_eliminate]: 1.66e-06
        [specialize_transform]: 6.21e-06
        [updatestate_useless_node_eliminater]: 5.08002e-06
        [accelerated_algorithm]: 3.284e-05
        [meta_shard_fg_expand]: 2.58998e-06
        [get_grad_eliminate_]: 5.57999e-06
        [merge_forward]: 3.85e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.555e-05
        [j_node_and_user_rematch]: 9.48997e-06
        [meta_fg_expand]: 2.14e-06
        [replace_old_param]: 1.001e-05
        [inline_without_move]: 5.56e-06
        [renormalize]: 0.00103721
        [add_forward_monad_depend]: 8.32e-06
        [auto_monad_grad]: 2.64999e-06
        [auto_monad_eliminator]: 1.774e-05
        [cse]: 3.144e-05
        [replace_applicator]: 1.721e-05
    [Cycle 2]: 0.00034146, [27]
        [switch_simplify]: 5.78997e-06
        [loop_unroll]: 4.82998e-06
        [a_1]: 9.64e-05
        [with_stream_mark]: 1.195e-05
        [recompute_prepare]: 5.09e-06
        [updatestate_depend_eliminate]: 2.63003e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 4.90001e-06
        [updatestate_useless_node_eliminater]: 4.77998e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 4.15e-06
        [merge_forward]: 2.73998e-06
        [cell_reuse_recompute_pass]: 2.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.484e-05
        [j_node_and_user_rematch]: 7.95998e-06
        [meta_fg_expand]: 1.50999e-06
        [replace_old_param]: 7.59002e-06
        [inline_without_move]: 4.78001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.54e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 5.48002e-06
        [cse]: 1.159e-05
        [replace_applicator]: 4.90001e-06
[py_interpret_to_execute_after_opt_a]: 1.17e-05
[rewriter_after_opt_a]: 3.777e-05
[convert_after_rewriter]: 6.92002e-06
[order_py_execute_after_rewriter]: 6.31998e-06
[mutable_eliminate]: 0.00074189
[jit_opt_b]: 5.332e-05, [1]
    [Cycle 1]: 4.539e-05, [2]
        [frontend_op_eliminate]: 1.742e-05
        [inline_after_opt_a]: 1.603e-05
[cconv]: 2.542e-05
[loop_unroll]: 0.00047157
[jit_opt_after_cconv]: 0.00015447, [1]
    [Cycle 1]: 0.00014686, [11]
        [c_1]: 2.224e-05
        [parameter_eliminate]: 3.62002e-06
        [updatestate_depend_eliminate]: 7.11001e-06
        [updatestate_assign_eliminate]: 2.39999e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [cse]: 2.339e-05
        [call_graph_tuple_transform]: 2.065e-05
        [tuple_list_get_item_eliminator]: 5.42999e-06
        [none_parameter_eliminate]: 1.58002e-06
        [renormalize]: 1.37e-06
        [switch_simplify]: 5.53002e-06
[remove_dup_value]: 1.44e-05
[partial_unused_args_eliminate]: 2.10002e-06
[environ_conv]: 5.82999e-06
[add_recomputation]: 5.098e-05
[cse_after_recomputation]: 2.263e-05, [1]
    [Cycle 1]: 1.684e-05, [1]
        [cse]: 9.72999e-06
[auto_monad_reorder]: 1.586e-05
[get_jit_bprop_graph]: 2.10002e-06
[rewriter_after_jit_bprop_graph]: 7.05998e-06
[opt_after_jit_grad]: 0.00053207
[symbol_engine_optimizer]: 8.141e-05, [1]
    [Cycle 1]: 7.33e-05, [6]
        [build]: 4.48001e-06
        [elim_shapecalc]: 8.70999e-06
        [elim_not_effective]: 1.436e-05
        [opt_reshape]: 6.29001e-06
        [fold_const_symbol]: 9.92999e-06
        [renormalize]: 6.39993e-07
[validate]: 0.00043304
[backend_pass]: 1.60001e-06
[task_emit]: 0.00687515
[execute]: 7.87e-06
Sums
    bootstrap                            :     0.000333s :  1.91%
    type_inference                       :     0.005254s : 30.07%
    event_method                         :     0.000020s :  0.11%
    auto_monad                           :     0.000068s :  0.39%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000047s :  0.27%
    rewriter_before_opt_a                :     0.000079s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.42%
    jit_opt_a.loop_unroll                :     0.000035s :  0.20%
    jit_opt_a.a_1                        :     0.000749s :  4.29%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000038s :  0.22%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000018s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.001037s :  5.94%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.13%
    jit_opt_a.cse                        :     0.000043s :  0.25%
    jit_opt_a.replace_applicator         :     0.000022s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000742s :  4.25%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.09%
    cconv                                :     0.000025s :  0.15%
    loop_unroll                          :     0.000472s :  2.70%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000051s :  0.29%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000532s :  3.04%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000433s :  2.48%
    backend_pass                         :     0.000002s :  0.01%
    task_emit                            :     0.006875s : 39.35%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000288    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000002s :      2: substitution.fold_const_symbol
 1.98% :     0.000006s :      3: substitution.graph_param_transform
82.83% :     0.000239s :      4: substitution.inline
 1.12% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.69% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000005s :      2: substitution.replace_old_param
 9.08% :     0.000026s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005180     2
81.45% :     0.004219s :      1: type_inference.infer
18.55% :     0.000961s :      1: type_inference.specialize
------[replace.]   0.000088     6
65.12% :     0.000058s :      4: replace.inline
34.88% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000261     6
90.59% :     0.000236s :      4: match.inline
 9.41% :     0.000025s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.67% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 2.05% :     0.000003s :     10: predicate.arithmetic_simplify
 1.43% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.93% :     0.000001s :     10: predicate.compare_switch_simplify
 1.67% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.11% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.94% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.40% :     0.000001s :      3: predicate.graph_param_transform
 5.53% :     0.000007s :     22: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.93% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.55% :     0.000002s :     15: predicate.load_eliminater
 1.68% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.52% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.28% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.01% :     0.000003s :      3: predicate.mutable_eliminate
 0.58% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.44% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.82% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 1.90% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.13% :     0.000010s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.04% :     0.000001s :     10: predicate.transpose_eliminate
 1.52% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.42% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.29% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.63% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000737    12
47.83% :     0.000353s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.64% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.53% :     0.000328s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024264    76
 0.22% :     0.000054s :      1: add_recomputation
 0.29% :     0.000071s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 1.43% :     0.000347s :      1: bootstrap
 0.12% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000024s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
26.87% :     0.006519s :      1: jit_opt_a
 0.65% :     0.000158s :      1: jit_opt_after_cconv
 0.23% :     0.000056s :      1: jit_opt_b
 1.98% :     0.000481s :      1: loop_unroll
 3.10% :     0.000752s :      1: mutable_eliminate
 4.15% :     0.001007s :     26: opt.transform.jit_opt_a
 0.21% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.24% :     0.000543s :      1: opt_after_jit_grad
 0.04% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.21% :     0.000050s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.35% :     0.000569s :      1: renormalize.infer
 1.89% :     0.000458s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000042s :      1: rewriter_after_opt_a
 0.34% :     0.000082s :      1: rewriter_before_opt_a
 0.35% :     0.000084s :      1: symbol_engine_optimizer
28.39% :     0.006889s :      1: task_emit
21.71% :     0.005268s :      1: type_inference
 1.89% :     0.000460s :      1: validate


TotalTime = 0.0165159, [30]
[bootstrap]: 0.00034217
[type_inference]: 0.0050166
[event_method]: 1.601e-05
[auto_monad]: 6.701e-05
[graph_reusing]: 6.67002e-06
[pre_auto_parallel]: 2.86e-06
[py_interpret_to_execute]: 3.954e-05
[rewriter_before_opt_a]: 7.61e-05
[expand_dump_flag]: 2.89001e-06
[jit_opt_a]: 0.00761475, [2]
    [Cycle 1]: 0.0025608, [27]
        [switch_simplify]: 8.935e-05
        [loop_unroll]: 3.109e-05
        [a_1]: 0.00078533
        [with_stream_mark]: 3.224e-05
        [recompute_prepare]: 1.216e-05
        [updatestate_depend_eliminate]: 4.65001e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.79001e-06
        [parameter_eliminate]: 2.73e-06
        [specialize_transform]: 7.82998e-06
        [updatestate_useless_node_eliminater]: 6.07999e-06
        [accelerated_algorithm]: 6.22001e-06
        [meta_shard_fg_expand]: 3.4e-06
        [get_grad_eliminate_]: 6.30002e-06
        [merge_forward]: 4.48999e-06
        [cell_reuse_recompute_pass]: 1.34998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.977e-05
        [j_node_and_user_rematch]: 1.286e-05
        [meta_fg_expand]: 2.82002e-06
        [replace_old_param]: 1.234e-05
        [inline_without_move]: 6.81001e-06
        [renormalize]: 0.00116906
        [add_forward_monad_depend]: 1.094e-05
        [auto_monad_grad]: 3.4e-06
        [auto_monad_eliminator]: 2.51e-05
        [cse]: 3.634e-05
        [replace_applicator]: 2.373e-05
    [Cycle 2]: 0.0004444, [27]
        [switch_simplify]: 7.28e-06
        [loop_unroll]: 5.39998e-06
        [a_1]: 0.00010815
        [with_stream_mark]: 1.939e-05
        [recompute_prepare]: 8.67998e-06
        [updatestate_depend_eliminate]: 4.39002e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.64999e-06
        [parameter_eliminate]: 2.19001e-06
        [specialize_transform]: 7.38e-06
        [updatestate_useless_node_eliminater]: 5.38002e-06
        [accelerated_algorithm]: 6.31e-06
        [meta_shard_fg_expand]: 2.32001e-06
        [get_grad_eliminate_]: 5.58002e-06
        [merge_forward]: 4.56002e-06
        [cell_reuse_recompute_pass]: 3.25e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.253e-05
        [j_node_and_user_rematch]: 1.036e-05
        [meta_fg_expand]: 3.11999e-06
        [replace_old_param]: 1.044e-05
        [inline_without_move]: 5.71998e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 3.56001e-06
        [auto_monad_grad]: 1.71e-06
        [auto_monad_eliminator]: 9.69e-06
        [cse]: 1.788e-05
        [replace_applicator]: 9.26002e-06
[py_interpret_to_execute_after_opt_a]: 1.943e-05
[rewriter_after_opt_a]: 4.937e-05
[convert_after_rewriter]: 9.56998e-06
[order_py_execute_after_rewriter]: 5.84999e-06
[mutable_eliminate]: 0.00098528
[jit_opt_b]: 7.113e-05, [1]
    [Cycle 1]: 5.956e-05, [2]
        [frontend_op_eliminate]: 2.206e-05
        [inline_after_opt_a]: 2.088e-05
[cconv]: 4.273e-05
[loop_unroll]: 0.00071399
[jit_opt_after_cconv]: 0.00020622, [1]
    [Cycle 1]: 0.00019461, [11]
        [c_1]: 2.538e-05
        [parameter_eliminate]: 6.83e-06
        [updatestate_depend_eliminate]: 1.27e-05
        [updatestate_assign_eliminate]: 3.21001e-06
        [updatestate_loads_eliminate]: 3.31999e-06
        [cse]: 3.722e-05
        [call_graph_tuple_transform]: 2.97e-05
        [tuple_list_get_item_eliminator]: 6.48e-06
        [none_parameter_eliminate]: 2.41998e-06
        [renormalize]: 1.30999e-06
        [switch_simplify]: 5.61e-06
[remove_dup_value]: 1.742e-05
[partial_unused_args_eliminate]: 3.97e-06
[environ_conv]: 7.45998e-06
[add_recomputation]: 6.192e-05
[cse_after_recomputation]: 2.995e-05, [1]
    [Cycle 1]: 2.178e-05, [1]
        [cse]: 1.22e-05
[auto_monad_reorder]: 1.799e-05
[get_jit_bprop_graph]: 2.94999e-06
[rewriter_after_jit_bprop_graph]: 1.083e-05
[opt_after_jit_grad]: 0.00070594
[symbol_engine_optimizer]: 9.272e-05, [1]
    [Cycle 1]: 8.354e-05, [6]
        [build]: 7.58999e-06
        [elim_shapecalc]: 8.64e-06
        [elim_not_effective]: 1.881e-05
        [opt_reshape]: 6.15002e-06
        [fold_const_symbol]: 1.003e-05
        [renormalize]: 4.50003e-07
[validate]: 5.282e-05
Sums
    bootstrap                            :     0.000342s :  3.07%
    type_inference                       :     0.005017s : 45.07%
    event_method                         :     0.000016s :  0.14%
    auto_monad                           :     0.000067s :  0.60%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000040s :  0.36%
    rewriter_before_opt_a                :     0.000076s :  0.68%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000097s :  0.87%
    jit_opt_a.loop_unroll                :     0.000036s :  0.33%
    jit_opt_a.a_1                        :     0.000893s :  8.03%
    jit_opt_a.with_stream_mark           :     0.000052s :  0.46%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.19%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.04%
    jit_opt_a.specialize_transform       :     0.000015s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000012s :  0.11%
    jit_opt_a.merge_forward              :     0.000009s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000042s :  0.38%
    jit_opt_a.j_node_and_user_rematch    :     0.000023s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.05%
    jit_opt_a.replace_old_param          :     0.000023s :  0.20%
    jit_opt_a.inline_without_move        :     0.000013s :  0.11%
    jit_opt_a.renormalize                :     0.001169s : 10.50%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.13%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000035s :  0.31%
    jit_opt_a.cse                        :     0.000054s :  0.49%
    jit_opt_a.replace_applicator         :     0.000033s :  0.30%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.17%
    rewriter_after_opt_a                 :     0.000049s :  0.44%
    convert_after_rewriter               :     0.000010s :  0.09%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000985s :  8.85%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000021s :  0.19%
    cconv                                :     0.000043s :  0.38%
    loop_unroll                          :     0.000714s :  6.41%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.06%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000013s :  0.11%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000037s :  0.33%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000030s :  0.27%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000017s :  0.16%
    partial_unused_args_eliminate        :     0.000004s :  0.04%
    environ_conv                         :     0.000007s :  0.07%
    add_recomputation                    :     0.000062s :  0.56%
    cse_after_recomputation.cse          :     0.000012s :  0.11%
    auto_monad_reorder                   :     0.000018s :  0.16%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.10%
    opt_after_jit_grad                   :     0.000706s :  6.34%
    symbol_engine_optimizer.build        :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000053s :  0.47%

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.827.690 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc395e0> should be wrapped by jit only once.

Time group info:
------[substitution.]   0.000377    23
 0.74% :     0.000003s :      2: substitution.elim_not_effective
 0.33% :     0.000001s :      2: substitution.fold_const_symbol
 2.01% :     0.000008s :      3: substitution.graph_param_transform
86.87% :     0.000327s :      4: substitution.inline
 1.40% :     0.000005s :      4: substitution.j_node_and_user_rematch
 1.95% :     0.000007s :      4: substitution.remove_not_recompute_node
 2.19% :     0.000008s :      2: substitution.replace_old_param
 4.50% :     0.000017s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004950     2
80.34% :     0.003976s :      1: type_inference.infer
19.66% :     0.000973s :      1: type_inference.specialize
------[replace.]   0.000101     6
67.51% :     0.000068s :      4: replace.inline
32.49% :     0.000033s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000339     6
95.41% :     0.000324s :      4: match.inline
 4.59% :     0.000016s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000153   691
 1.51% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.99% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.81% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 1.95% :     0.000003s :     10: predicate.arithmetic_simplify
 1.30% :     0.000002s :     10: predicate.cast_eliminate
 0.34% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.80% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.32% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000001s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.85% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.04% :     0.000002s :     10: predicate.environ_get_eliminate
 0.98% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.16% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000002s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 6.61% :     0.000010s :     22: predicate.inline
 1.17% :     0.000002s :      6: predicate.inline_without_move
 0.27% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.70% :     0.000003s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.42% :     0.000002s :     15: predicate.load_eliminater
 2.38% :     0.000004s :      3: predicate.loop_unroll_after_grad
 3.16% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.10% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.83% :     0.000001s :     10: predicate.merge_addn
 0.79% :     0.000001s :     10: predicate.minmaximum_grad
 3.51% :     0.000005s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.01% :     0.000003s :     15: predicate.partial_eliminate
 0.85% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000003s :     10: predicate.reduce_eliminate
 1.25% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.46% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.40% :     0.000001s :      3: predicate.reset_defer_inline
 1.28% :     0.000002s :     10: predicate.reshape_eliminate
 0.90% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.05% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.06% :     0.000002s :     10: predicate.same_eliminate
 0.70% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.13% :     0.000002s :      6: predicate.special_op_eliminate
 0.80% :     0.000001s :      6: predicate.specialize_transform
 1.79% :     0.000003s :     10: predicate.split_environ_get_set_with_tuple_value
 0.98% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.72% :     0.000004s :     16: predicate.switch_defer_inline
 1.53% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.27% :     0.000013s :     53: predicate.switch_simplify
 1.02% :     0.000002s :     10: predicate.tile_eliminate
 1.22% :     0.000002s :     10: predicate.transpose_eliminate
 2.05% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.48% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.14% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.26% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.53% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.27% :     0.000003s :     10: predicate.value_based_eliminate
 0.27% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.00% :     0.000002s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000749    12
44.12% :     0.000330s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.89% :     0.000067s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.00% :     0.000352s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018963    72
 0.35% :     0.000066s :      1: add_recomputation
 0.37% :     0.000070s :      1: auto_monad
 0.11% :     0.000022s :      1: auto_monad_reorder
 1.88% :     0.000357s :      1: bootstrap
 0.25% :     0.000047s :      1: cconv
 0.07% :     0.000013s :      1: convert_after_rewriter
 0.17% :     0.000032s :      1: cse_after_recomputation
 0.06% :     0.000011s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
40.18% :     0.007619s :      1: jit_opt_a
 1.11% :     0.000210s :      1: jit_opt_after_cconv
 0.39% :     0.000074s :      1: jit_opt_b
 3.84% :     0.000729s :      1: loop_unroll
 5.32% :     0.001009s :      1: mutable_eliminate
 6.27% :     0.001189s :     26: opt.transform.jit_opt_a
 0.33% :     0.000063s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000034s :      4: opt.transform.jit_opt_b
 0.12% :     0.000023s :      1: opt.transform.loop_unroll_optimizer
 0.17% :     0.000032s :      1: opt.transform.mutable_eliminate
 0.18% :     0.000034s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000040s :      4: opt.transform.symbol_engine_opt
 3.80% :     0.000721s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000006s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000042s :      1: py_interpret_to_execute
 0.12% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000020s :      1: remove_dup_value
 3.22% :     0.000610s :      1: renormalize.infer
 2.87% :     0.000543s :      1: renormalize.specialize
 0.07% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.29% :     0.000056s :      1: rewriter_after_opt_a
 0.42% :     0.000079s :      1: rewriter_before_opt_a
 0.51% :     0.000096s :      1: symbol_engine_optimizer
26.51% :     0.005028s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.828.337 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.02041, [33]
[bootstrap]: 0.00029955
[type_inference]: 0.00461755
[event_method]: 1.993e-05
[auto_monad]: 6.761e-05
[graph_reusing]: 7.18998e-06
[pre_auto_parallel]: 2.50002e-06
[py_interpret_to_execute]: 3.491e-05
[rewriter_before_opt_a]: 7.401e-05
[expand_dump_flag]: 3.41999e-06
[jit_opt_a]: 0.0061031, [2]
    [Cycle 1]: 0.00198238, [27]
        [switch_simplify]: 6.541e-05
        [loop_unroll]: 2.977e-05
        [a_1]: 0.00061149
        [with_stream_mark]: 2.534e-05
        [recompute_prepare]: 7.9e-06
        [updatestate_depend_eliminate]: 4.24002e-06
        [updatestate_assign_eliminate]: 3.04999e-06
        [updatestate_loads_eliminate]: 3.04999e-06
        [parameter_eliminate]: 1.73002e-06
        [specialize_transform]: 5.86998e-06
        [updatestate_useless_node_eliminater]: 5.34998e-06
        [accelerated_algorithm]: 5.96e-06
        [meta_shard_fg_expand]: 1.82999e-06
        [get_grad_eliminate_]: 5.44e-06
        [merge_forward]: 4.33001e-06
        [cell_reuse_recompute_pass]: 1.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.53e-05
        [j_node_and_user_rematch]: 9.72999e-06
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 1.106e-05
        [inline_without_move]: 5.97999e-06
        [renormalize]: 0.00088261
        [add_forward_monad_depend]: 6.91999e-06
        [auto_monad_grad]: 3.04001e-06
        [auto_monad_eliminator]: 1.773e-05
        [cse]: 3.031e-05
        [replace_applicator]: 1.636e-05
    [Cycle 2]: 0.00035139, [27]
        [switch_simplify]: 5.58002e-06
        [loop_unroll]: 5.17e-06
        [a_1]: 9.48e-05
        [with_stream_mark]: 1.281e-05
        [recompute_prepare]: 5.64e-06
        [updatestate_depend_eliminate]: 3.04999e-06
        [updatestate_assign_eliminate]: 2.37001e-06
        [updatestate_loads_eliminate]: 2.33002e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 5.19998e-06
        [updatestate_useless_node_eliminater]: 5.22999e-06
        [accelerated_algorithm]: 6.00002e-06
        [meta_shard_fg_expand]: 1.35999e-06
        [get_grad_eliminate_]: 4.78001e-06
        [merge_forward]: 2.53e-06
        [cell_reuse_recompute_pass]: 2.33998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.643e-05
        [j_node_and_user_rematch]: 9.13002e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 7.66001e-06
        [inline_without_move]: 4.77998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.60001e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 6.52001e-06
        [cse]: 1.29e-05
        [replace_applicator]: 7.05998e-06
[py_interpret_to_execute_after_opt_a]: 1.415e-05
[rewriter_after_opt_a]: 4.033e-05
[convert_after_rewriter]: 6.79001e-06
[order_py_execute_after_rewriter]: 4.84e-06
[mutable_eliminate]: 0.00073088
[jit_opt_b]: 5.319e-05, [1]
    [Cycle 1]: 4.509e-05, [2]
        [frontend_op_eliminate]: 1.726e-05
        [inline_after_opt_a]: 1.547e-05
[cconv]: 3.128e-05
[loop_unroll]: 0.00045797
[jit_opt_after_cconv]: 0.00015358, [1]
    [Cycle 1]: 0.00014669, [11]
        [c_1]: 2.06e-05
        [parameter_eliminate]: 3.13e-06
        [updatestate_depend_eliminate]: 6.96999e-06
        [updatestate_assign_eliminate]: 2.49999e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [cse]: 2.512e-05
        [call_graph_tuple_transform]: 2.299e-05
        [tuple_list_get_item_eliminator]: 5.37999e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 7.79983e-07
        [switch_simplify]: 5.30999e-06
[remove_dup_value]: 1.49e-05
[partial_unused_args_eliminate]: 2.69001e-06
[environ_conv]: 5.51e-06
[add_recomputation]: 5.267e-05
[cse_after_recomputation]: 2.005e-05, [1]
    [Cycle 1]: 1.465e-05, [1]
        [cse]: 9.04998e-06
[auto_monad_reorder]: 1.578e-05
[get_jit_bprop_graph]: 1.99e-06
[rewriter_after_jit_bprop_graph]: 6.07001e-06
[opt_after_jit_grad]: 0.00045769
[symbol_engine_optimizer]: 7.581e-05, [1]
    [Cycle 1]: 6.858e-05, [6]
        [build]: 4.75001e-06
        [elim_shapecalc]: 7.93001e-06
        [elim_not_effective]: 1.346e-05
        [opt_reshape]: 5.80002e-06
        [fold_const_symbol]: 9.04998e-06
        [renormalize]: 4.69998e-07
[validate]: 3.933e-05
[backend_pass]: 8.80013e-07
[task_emit]: 0.00676838
[execute]: 7.41001e-06
Sums
    bootstrap                            :     0.000300s :  1.87%
    type_inference                       :     0.004618s : 28.90%
    event_method                         :     0.000020s :  0.12%
    auto_monad                           :     0.000068s :  0.42%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.22%
    rewriter_before_opt_a                :     0.000074s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000071s :  0.44%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000706s :  4.42%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000883s :  5.52%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.15%
    jit_opt_a.cse                        :     0.000043s :  0.27%
    jit_opt_a.replace_applicator         :     0.000023s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.09%
    rewriter_after_opt_a                 :     0.000040s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000731s :  4.57%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000031s :  0.20%
    loop_unroll                          :     0.000458s :  2.87%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000053s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000458s :  2.86%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000039s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006768s : 42.36%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000258    23
 0.71% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 2.35% :     0.000006s :      3: substitution.graph_param_transform
85.63% :     0.000221s :      4: substitution.inline
 1.32% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.28% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.16% :     0.000006s :      2: substitution.replace_old_param
 5.09% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004551     2
81.03% :     0.003688s :      1: type_inference.infer
18.97% :     0.000864s :      1: type_inference.specialize
------[replace.]   0.000076     6
68.71% :     0.000052s :      4: replace.inline
31.29% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000231     6
94.87% :     0.000219s :      4: match.inline
 5.13% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.12% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.21% :     0.000002s :     10: predicate.addn_zero_filter
 2.48% :     0.000003s :     10: predicate.arithmetic_simplify
 1.16% :     0.000001s :     10: predicate.cast_eliminate
 0.67% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.41% :     0.000002s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.09% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000000s :      3: predicate.graph_param_transform
 5.86% :     0.000007s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.66% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.37% :     0.000003s :     15: predicate.load_eliminater
 1.24% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.79% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.74% :     0.000002s :      3: predicate.mutable_eliminate
 0.63% :     0.000001s :      3: predicate.opt_reshape
 2.37% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.20% :     0.000003s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.42% :     0.000001s :      3: predicate.reset_defer_inline
 1.29% :     0.000002s :     10: predicate.reshape_eliminate
 1.11% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.17% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.84% :     0.000004s :     16: predicate.switch_defer_inline
 2.05% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.65% :     0.000010s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.23% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.33% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.94% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.85% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000597    12
45.59% :     0.000272s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 9.21% :     0.000055s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.20% :     0.000270s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022346    76
 0.25% :     0.000056s :      1: add_recomputation
 0.32% :     0.000071s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.42% :     0.000317s :      1: bootstrap
 0.15% :     0.000034s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000025s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
27.33% :     0.006106s :      1: jit_opt_a
 0.70% :     0.000156s :      1: jit_opt_after_cconv
 0.25% :     0.000056s :      1: jit_opt_b
 2.10% :     0.000468s :      1: loop_unroll
 3.32% :     0.000743s :      1: mutable_eliminate
 4.22% :     0.000944s :     26: opt.transform.jit_opt_a
 0.23% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.08% :     0.000466s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000037s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 2.34% :     0.000524s :      1: renormalize.infer
 1.56% :     0.000349s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000044s :      1: rewriter_after_opt_a
 0.35% :     0.000077s :      1: rewriter_before_opt_a
 0.35% :     0.000078s :      1: symbol_engine_optimizer
30.38% :     0.006788s :      1: task_emit
20.74% :     0.004635s :      1: type_inference
 0.27% :     0.000059s :      1: validate


TotalTime = 0.02164, [33]
[bootstrap]: 0.00032862
[type_inference]: 0.00533565
[event_method]: 2.212e-05
[auto_monad]: 6.925e-05
[graph_reusing]: 7.06001e-06
[pre_auto_parallel]: 2.58e-06
[py_interpret_to_execute]: 5.167e-05
[rewriter_before_opt_a]: 7.877e-05
[expand_dump_flag]: 3.19001e-06
[jit_opt_a]: 0.00682675, [2]
    [Cycle 1]: 0.00224846, [27]
        [switch_simplify]: 7.306e-05
        [loop_unroll]: 3.001e-05
        [a_1]: 0.00070558
        [with_stream_mark]: 3.086e-05
        [recompute_prepare]: 1.098e-05
        [updatestate_depend_eliminate]: 4.01001e-06
        [updatestate_assign_eliminate]: 3.14999e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 2.43e-06
        [specialize_transform]: 6.16e-06
        [updatestate_useless_node_eliminater]: 5.10001e-06
        [accelerated_algorithm]: 6.43998e-06
        [meta_shard_fg_expand]: 2.27999e-06
        [get_grad_eliminate_]: 5.79e-06
        [merge_forward]: 3.55e-06
        [cell_reuse_recompute_pass]: 1.42999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.662e-05
        [j_node_and_user_rematch]: 9.97999e-06
        [meta_fg_expand]: 2.66e-06
        [replace_old_param]: 1.041e-05
        [inline_without_move]: 5.82999e-06
        [renormalize]: 0.00103459
        [add_forward_monad_depend]: 7.07002e-06
        [auto_monad_grad]: 2.17999e-06
        [auto_monad_eliminator]: 1.464e-05
        [cse]: 2.972e-05
        [replace_applicator]: 1.726e-05
    [Cycle 2]: 0.00035998, [27]
        [switch_simplify]: 6.11998e-06
        [loop_unroll]: 5.05999e-06
        [a_1]: 9.592e-05
        [with_stream_mark]: 1.309e-05
        [recompute_prepare]: 6.15002e-06
        [updatestate_depend_eliminate]: 3.06001e-06
        [updatestate_assign_eliminate]: 2.77002e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 5.47999e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.27999e-06
        [meta_shard_fg_expand]: 1.81003e-06
        [get_grad_eliminate_]: 4.58001e-06
        [merge_forward]: 2.79001e-06
        [cell_reuse_recompute_pass]: 2.41998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.484e-05
        [j_node_and_user_rematch]: 8.38001e-06
        [meta_fg_expand]: 2.07999e-06
        [replace_old_param]: 8.57e-06
        [inline_without_move]: 4.72998e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.76998e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 8.15999e-06
        [cse]: 1.344e-05
        [replace_applicator]: 5.42001e-06
[py_interpret_to_execute_after_opt_a]: 1.543e-05
[rewriter_after_opt_a]: 3.991e-05
[convert_after_rewriter]: 7.5e-06
[order_py_execute_after_rewriter]: 5.10999e-06
[mutable_eliminate]: 0.0007563
[jit_opt_b]: 5.92e-05, [1]
    [Cycle 1]: 5.038e-05, [2]
        [frontend_op_eliminate]: 1.691e-05
        [inline_after_opt_a]: 1.901e-05
[cconv]: 3.263e-05
[loop_unroll]: 0.00052756
[jit_opt_after_cconv]: 0.0001559, [1]
    [Cycle 1]: 0.00014892, [11]
        [c_1]: 2.321e-05
        [parameter_eliminate]: 4.33999e-06
        [updatestate_depend_eliminate]: 5.14e-06
        [updatestate_assign_eliminate]: 2.97002e-06
        [updatestate_loads_eliminate]: 2.27001e-06
        [cse]: 2.411e-05
        [call_graph_tuple_transform]: 2.121e-05
        [tuple_list_get_item_eliminator]: 5.47001e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 5.52999e-06
[remove_dup_value]: 1.477e-05
[partial_unused_args_eliminate]: 2.42001e-06
[environ_conv]: 6.41e-06
[add_recomputation]: 5.002e-05
[cse_after_recomputation]: 2.503e-05, [1]
    [Cycle 1]: 1.825e-05, [1]
        [cse]: 1.082e-05
[auto_monad_reorder]: 1.483e-05
[get_jit_bprop_graph]: 2.39001e-06
[rewriter_after_jit_bprop_graph]: 6.28e-06
[opt_after_jit_grad]: 0.00055283
[symbol_engine_optimizer]: 8.115e-05, [1]
    [Cycle 1]: 7.271e-05, [6]
        [build]: 4.15e-06
        [elim_shapecalc]: 8.22998e-06
        [elim_not_effective]: 1.512e-05
        [opt_reshape]: 6.59999e-06
        [fold_const_symbol]: 9.10999e-06
        [renormalize]: 5.00004e-07
[validate]: 4.085e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.00628097
[execute]: 6.06998e-06
Sums
    bootstrap                            :     0.000329s :  1.96%
    type_inference                       :     0.005336s : 31.90%
    event_method                         :     0.000022s :  0.13%
    auto_monad                           :     0.000069s :  0.41%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000052s :  0.31%
    rewriter_before_opt_a                :     0.000079s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000079s :  0.47%
    jit_opt_a.loop_unroll                :     0.000035s :  0.21%
    jit_opt_a.a_1                        :     0.000801s :  4.79%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001035s :  6.19%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.14%
    jit_opt_a.cse                        :     0.000043s :  0.26%
    jit_opt_a.replace_applicator         :     0.000023s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.09%
    rewriter_after_opt_a                 :     0.000040s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000756s :  4.52%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.11%
    cconv                                :     0.000033s :  0.20%
    loop_unroll                          :     0.000528s :  3.15%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000050s :  0.30%
    cse_after_recomputation.cse          :     0.000011s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000553s :  3.31%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006281s : 37.55%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000318    23
 0.65% :     0.000002s :      2: substitution.elim_not_effective
 0.44% :     0.000001s :      2: substitution.fold_const_symbol
 1.65% :     0.000005s :      3: substitution.graph_param_transform
88.16% :     0.000280s :      4: substitution.inline
 1.37% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.46% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.64% :     0.000005s :      2: substitution.replace_old_param
 4.63% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005260     2
80.35% :     0.004226s :      1: type_inference.infer
19.65% :     0.001034s :      1: type_inference.specialize
------[replace.]   0.000096     6
62.93% :     0.000061s :      4: replace.inline
37.07% :     0.000036s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000291     6
95.32% :     0.000277s :      4: match.inline
 4.68% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.15% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.36% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.68% :     0.000002s :     10: predicate.addn_zero_filter
 1.80% :     0.000002s :     10: predicate.arithmetic_simplify
 1.19% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.10% :     0.000001s :     10: predicate.dict_get_item_eliminator
 0.95% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.68% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.89% :     0.000008s :     22: predicate.inline
 0.82% :     0.000001s :      6: predicate.inline_without_move
 0.47% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.39% :     0.000002s :      6: predicate.less_batch_normalization
 1.32% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.60% :     0.000002s :     15: predicate.load_eliminater
 1.82% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.49% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 0.94% :     0.000001s :     10: predicate.minmaximum_grad
 1.79% :     0.000002s :      3: predicate.mutable_eliminate
 0.93% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.07% :     0.000003s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.93% :     0.000003s :     18: predicate.replace_applicator
 0.82% :     0.000001s :      6: predicate.replace_old_param
 0.49% :     0.000001s :      3: predicate.reset_defer_inline
 1.87% :     0.000002s :     10: predicate.reshape_eliminate
 1.09% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.05% :     0.000001s :     10: predicate.same_eliminate
 0.67% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.03% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.27% :     0.000003s :     16: predicate.switch_defer_inline
 1.99% :     0.000003s :     16: predicate.switch_layer_defer_inline
 9.00% :     0.000012s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.45% :     0.000002s :     10: predicate.transpose_eliminate
 1.52% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.03% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.38% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.08% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000780    12
46.95% :     0.000366s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.62% :     0.000059s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.42% :     0.000354s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023836    76
 0.22% :     0.000053s :      1: add_recomputation
 0.31% :     0.000073s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.45% :     0.000344s :      1: bootstrap
 0.15% :     0.000035s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000028s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.11% :     0.000026s :      1: event_method
 0.04% :     0.000009s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
28.66% :     0.006831s :      1: jit_opt_a
 0.67% :     0.000159s :      1: jit_opt_after_cconv
 0.26% :     0.000063s :      1: jit_opt_b
 2.25% :     0.000537s :      1: loop_unroll
 3.22% :     0.000768s :      1: mutable_eliminate
 4.39% :     0.001046s :     26: opt.transform.jit_opt_a
 0.22% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000028s :      4: opt.transform.jit_opt_b
 0.07% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.37% :     0.000564s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000054s :      1: py_interpret_to_execute
 0.08% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 2.39% :     0.000570s :      1: renormalize.infer
 1.91% :     0.000456s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000045s :      1: rewriter_after_opt_a
 0.35% :     0.000083s :      1: rewriter_before_opt_a
 0.35% :     0.000084s :      1: symbol_engine_optimizer
26.41% :     0.006294s :      1: task_emit
22.45% :     0.005352s :      1: type_inference
 0.26% :     0.000062s :      1: validate


TotalTime = 0.0192566, [33]
[bootstrap]: 0.00031351
[type_inference]: 0.00480853
[event_method]: 1.672e-05
[auto_monad]: 6.434e-05
[graph_reusing]: 5.34e-06
[pre_auto_parallel]: 2.42001e-06
[py_interpret_to_execute]: 3.337e-05
[rewriter_before_opt_a]: 7.221e-05
[expand_dump_flag]: 3.26999e-06
[jit_opt_a]: 0.00525953, [2]
    [Cycle 1]: 0.00156156, [27]
        [switch_simplify]: 5.363e-05
        [loop_unroll]: 3.029e-05
        [a_1]: 0.00052318
        [with_stream_mark]: 1.701e-05
        [recompute_prepare]: 6.87002e-06
        [updatestate_depend_eliminate]: 3.5e-06
        [updatestate_assign_eliminate]: 3.7e-06
        [updatestate_loads_eliminate]: 2.61999e-06
        [parameter_eliminate]: 2.51e-06
        [specialize_transform]: 5.61e-06
        [updatestate_useless_node_eliminater]: 5.54e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 2.09e-06
        [get_grad_eliminate_]: 5.39998e-06
        [merge_forward]: 3.85e-06
        [cell_reuse_recompute_pass]: 1.11002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.413e-05
        [j_node_and_user_rematch]: 9.04e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 8.79e-06
        [inline_without_move]: 5.87001e-06
        [renormalize]: 0.00062621
        [add_forward_monad_depend]: 5.63002e-06
        [auto_monad_grad]: 2.69999e-06
        [auto_monad_eliminator]: 1.29e-05
        [cse]: 2.558e-05
        [replace_applicator]: 1.235e-05
    [Cycle 2]: 0.00032307, [27]
        [switch_simplify]: 5.97999e-06
        [loop_unroll]: 5.07e-06
        [a_1]: 9.152e-05
        [with_stream_mark]: 8.53001e-06
        [recompute_prepare]: 4.85001e-06
        [updatestate_depend_eliminate]: 2.44001e-06
        [updatestate_assign_eliminate]: 2.05002e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 9.99979e-07
        [specialize_transform]: 4.95999e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.30001e-06
        [meta_shard_fg_expand]: 1.50999e-06
        [get_grad_eliminate_]: 4.53999e-06
        [merge_forward]: 2.34999e-06
        [cell_reuse_recompute_pass]: 1.84998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.295e-05
        [j_node_and_user_rematch]: 8.02e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 7.34002e-06
        [inline_without_move]: 4.74002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.40999e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 4.85999e-06
        [cse]: 1.046e-05
        [replace_applicator]: 5.13002e-06
[py_interpret_to_execute_after_opt_a]: 1.047e-05
[rewriter_after_opt_a]: 3.336e-05
[convert_after_rewriter]: 6.43e-06
[order_py_execute_after_rewriter]: 4.77998e-06
[mutable_eliminate]: 0.00053207
[jit_opt_b]: 5.106e-05, [1]
    [Cycle 1]: 4.376e-05, [2]
        [frontend_op_eliminate]: 1.699e-05
        [inline_after_opt_a]: 1.522e-05
[cconv]: 2.447e-05
[loop_unroll]: 0.00042618
[jit_opt_after_cconv]: 0.00023845, [1]
    [Cycle 1]: 0.00023156, [11]
        [c_1]: 2.194e-05
        [parameter_eliminate]: 2.59999e-06
        [updatestate_depend_eliminate]: 5.63002e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 1.869e-05
        [call_graph_tuple_transform]: 7.299e-05
        [tuple_list_get_item_eliminator]: 6.29001e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 5.77001e-06
[remove_dup_value]: 3.191e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 5.13002e-06
[add_recomputation]: 0.00025864
[cse_after_recomputation]: 2.991e-05, [1]
    [Cycle 1]: 2.263e-05, [1]
        [cse]: 1.363e-05
[auto_monad_reorder]: 1.626e-05
[get_jit_bprop_graph]: 2.19001e-06
[rewriter_after_jit_bprop_graph]: 5.27001e-06
[opt_after_jit_grad]: 0.00050144
[symbol_engine_optimizer]: 7.505e-05, [1]
    [Cycle 1]: 6.824e-05, [6]
        [build]: 3.26001e-06
        [elim_shapecalc]: 8.63001e-06
        [elim_not_effective]: 1.288e-05
        [opt_reshape]: 5.51e-06
        [fold_const_symbol]: 8.91997e-06
        [renormalize]: 7.00005e-07
[validate]: 3.423e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.00610896
[execute]: 7.05e-06
Sums
    bootstrap                            :     0.000314s :  2.07%
    type_inference                       :     0.004809s : 31.71%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000064s :  0.42%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.22%
    rewriter_before_opt_a                :     0.000072s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000060s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.23%
    jit_opt_a.a_1                        :     0.000615s :  4.05%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000626s :  4.13%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000036s :  0.24%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000532s :  3.51%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000024s :  0.16%
    loop_unroll                          :     0.000426s :  2.81%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000073s :  0.48%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000032s :  0.21%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000259s :  1.71%
    cse_after_recomputation.cse          :     0.000014s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000501s :  3.31%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006109s : 40.29%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000188    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.83% :     0.000002s :      2: substitution.fold_const_symbol
 3.33% :     0.000006s :      3: substitution.graph_param_transform
82.31% :     0.000154s :      4: substitution.inline
 1.62% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.35% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.10% :     0.000004s :      2: substitution.replace_old_param
 6.46% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004747     2
83.25% :     0.003952s :      1: type_inference.infer
16.75% :     0.000795s :      1: type_inference.specialize
------[replace.]   0.000067     6
67.51% :     0.000045s :      4: replace.inline
32.49% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000163     6
93.26% :     0.000152s :      4: match.inline
 6.74% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.21% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 2.44% :     0.000003s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.53% :     0.000001s :      3: predicate.graph_param_transform
 5.61% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.22% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.78% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.17% :     0.000001s :     10: predicate.minmaximum_grad
 1.56% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.59% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 1.18% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.25% :     0.000002s :     10: predicate.reshape_eliminate
 1.38% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.22% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.38% :     0.000010s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000002s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.20% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.29% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000562    12
51.90% :     0.000292s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.08% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.01% :     0.000242s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020884    76
 1.27% :     0.000265s :      1: add_recomputation
 0.32% :     0.000068s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.56% :     0.000326s :      1: bootstrap
 0.13% :     0.000027s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000032s :      1: cse_after_recomputation
 0.11% :     0.000023s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.20% :     0.005263s :      1: jit_opt_a
 1.16% :     0.000242s :      1: jit_opt_after_cconv
 0.26% :     0.000054s :      1: jit_opt_b
 2.08% :     0.000435s :      1: loop_unroll
 2.59% :     0.000541s :      1: mutable_eliminate
 3.95% :     0.000825s :     26: opt.transform.jit_opt_a
 0.49% :     0.000103s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.12% :     0.000025s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.44% :     0.000509s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.31% :     0.000065s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.17% :     0.000035s :      1: remove_dup_value
 1.52% :     0.000318s :      1: renormalize.infer
 1.44% :     0.000300s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000037s :      1: rewriter_after_opt_a
 0.36% :     0.000075s :      1: rewriter_before_opt_a
 0.39% :     0.000082s :      1: symbol_engine_optimizer
29.31% :     0.006122s :      1: task_emit
23.08% :     0.004821s :      1: type_inference
 0.26% :     0.000054s :      1: validate


TotalTime = 0.0121656, [30]
[bootstrap]: 0.00028277
[type_inference]: 0.00433659
[event_method]: 1.837e-05
[auto_monad]: 6.521e-05
[graph_reusing]: 5.66998e-06
[pre_auto_parallel]: 3.05998e-06
[py_interpret_to_execute]: 3.245e-05
[rewriter_before_opt_a]: 7.244e-05
[expand_dump_flag]: 3.25e-06
[jit_opt_a]: 0.00526997, [2]
    [Cycle 1]: 0.00153991, [27]
        [switch_simplify]: 5.347e-05
        [loop_unroll]: 2.929e-05
        [a_1]: 0.00053339
        [with_stream_mark]: 1.657e-05
        [recompute_prepare]: 6.88998e-06
        [updatestate_depend_eliminate]: 3.51001e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 2.02001e-06
        [specialize_transform]: 5.57999e-06
        [updatestate_useless_node_eliminater]: 4.96997e-06
        [accelerated_algorithm]: 5.91e-06
        [meta_shard_fg_expand]: 1.96e-06
        [get_grad_eliminate_]: 5.46002e-06
        [merge_forward]: 3.48e-06
        [cell_reuse_recompute_pass]: 1.13001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.409e-05
        [j_node_and_user_rematch]: 8.92e-06
        [meta_fg_expand]: 2.12999e-06
        [replace_old_param]: 9.20999e-06
        [inline_without_move]: 5.02999e-06
        [renormalize]: 0.00060408
        [add_forward_monad_depend]: 4.97e-06
        [auto_monad_grad]: 2.56998e-06
        [auto_monad_eliminator]: 1.298e-05
        [cse]: 2.28e-05
        [replace_applicator]: 1.248e-05
    [Cycle 2]: 0.0003341, [27]
        [switch_simplify]: 5.71e-06
        [loop_unroll]: 4.95001e-06
        [a_1]: 9.835e-05
        [with_stream_mark]: 1.11e-05
        [recompute_prepare]: 5.10999e-06
        [updatestate_depend_eliminate]: 2.41998e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.11002e-06
        [specialize_transform]: 5.27001e-06
        [updatestate_useless_node_eliminater]: 5.32999e-06
        [accelerated_algorithm]: 5.33002e-06
        [meta_shard_fg_expand]: 1.49998e-06
        [get_grad_eliminate_]: 4.87e-06
        [merge_forward]: 2.69001e-06
        [cell_reuse_recompute_pass]: 1.60999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.236e-05
        [j_node_and_user_rematch]: 7.50998e-06
        [meta_fg_expand]: 1.80001e-06
        [replace_old_param]: 7.18e-06
        [inline_without_move]: 4.54998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.37999e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 4.95999e-06
        [cse]: 1.034e-05
        [replace_applicator]: 4.81002e-06
[py_interpret_to_execute_after_opt_a]: 6.80998e-06
[rewriter_after_opt_a]: 2.982e-05
[convert_after_rewriter]: 7.31001e-06
[order_py_execute_after_rewriter]: 4.68001e-06
[mutable_eliminate]: 0.00053157
[jit_opt_b]: 5.02e-05, [1]
    [Cycle 1]: 4.375e-05, [2]
        [frontend_op_eliminate]: 1.665e-05
        [inline_after_opt_a]: 1.571e-05
[cconv]: 2.601e-05
[loop_unroll]: 0.00041447
[jit_opt_after_cconv]: 0.00013595, [1]
    [Cycle 1]: 0.00012967, [11]
        [c_1]: 2.055e-05
        [parameter_eliminate]: 2.58998e-06
        [updatestate_depend_eliminate]: 5.72999e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.698e-05
        [call_graph_tuple_transform]: 1.771e-05
        [tuple_list_get_item_eliminator]: 5.42999e-06
        [none_parameter_eliminate]: 1.33002e-06
        [renormalize]: 4.7998e-07
        [switch_simplify]: 5.14e-06
[remove_dup_value]: 1.394e-05
[partial_unused_args_eliminate]: 2.04e-06
[environ_conv]: 4.97999e-06
[add_recomputation]: 4.378e-05
[cse_after_recomputation]: 2.048e-05, [1]
    [Cycle 1]: 1.467e-05, [1]
        [cse]: 9.00001e-06
[auto_monad_reorder]: 1.516e-05
[get_jit_bprop_graph]: 1.39998e-06
[rewriter_after_jit_bprop_graph]: 4.93001e-06
[opt_after_jit_grad]: 0.00043982
[symbol_engine_optimizer]: 7.255e-05, [1]
    [Cycle 1]: 6.518e-05, [6]
        [build]: 3.18998e-06
        [elim_shapecalc]: 7.5e-06
        [elim_not_effective]: 1.271e-05
        [opt_reshape]: 5.15999e-06
        [fold_const_symbol]: 8.74e-06
        [renormalize]: 3.80009e-07
[validate]: 3.254e-05
Sums
    bootstrap                            :     0.000283s :  3.47%
    type_inference                       :     0.004337s : 53.20%
    event_method                         :     0.000018s :  0.23%
    auto_monad                           :     0.000065s :  0.80%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.04%
    py_interpret_to_execute              :     0.000032s :  0.40%
    rewriter_before_opt_a                :     0.000072s :  0.89%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000059s :  0.73%
    jit_opt_a.loop_unroll                :     0.000034s :  0.42%
    jit_opt_a.a_1                        :     0.000632s :  7.75%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.34%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000604s :  7.41%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.22%
    jit_opt_a.cse                        :     0.000033s :  0.41%
    jit_opt_a.replace_applicator         :     0.000017s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.08%
    rewriter_after_opt_a                 :     0.000030s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000532s :  6.52%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.19%
    cconv                                :     0.000026s :  0.32%
    loop_unroll                          :     0.000414s :  5.08%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000044s :  0.54%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.19%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000440s :  5.40%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.40%


Time group info:
------[substitution.]   0.000183    23
 1.15% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.79% :     0.000005s :      3: substitution.graph_param_transform
82.67% :     0.000151s :      4: substitution.inline
 1.51% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.63% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.97% :     0.000004s :      2: substitution.replace_old_param
 6.59% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004275     2
79.42% :     0.003395s :      1: type_inference.infer
20.58% :     0.000880s :      1: type_inference.specialize
------[replace.]   0.000062     6
65.56% :     0.000041s :      4: replace.inline
34.44% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000160     6
93.12% :     0.000149s :      4: match.inline
 6.88% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.53% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 2.03% :     0.000002s :     10: predicate.arithmetic_simplify
 1.66% :     0.000002s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.42% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.62% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.34% :     0.000002s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.34% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.28% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.96% :     0.000002s :      3: predicate.mutable_eliminate
 0.39% :     0.000000s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.36% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.51% :     0.000001s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.97% :     0.000008s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.11% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000002s :     10: predicate.value_based_eliminate
 0.43% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000584    12
48.48% :     0.000283s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.71% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.81% :     0.000273s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013664    72
 0.34% :     0.000047s :      1: add_recomputation
 0.50% :     0.000069s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 2.16% :     0.000295s :      1: bootstrap
 0.21% :     0.000029s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000022s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.59% :     0.005273s :      1: jit_opt_a
 1.02% :     0.000139s :      1: jit_opt_after_cconv
 0.39% :     0.000053s :      1: jit_opt_b
 3.09% :     0.000422s :      1: loop_unroll
 3.95% :     0.000540s :      1: mutable_eliminate
 6.14% :     0.000838s :     26: opt.transform.jit_opt_a
 0.33% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.28% :     0.000448s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.04% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000035s :      1: py_interpret_to_execute
 0.07% :     0.000009s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.29% :     0.000312s :      1: renormalize.infer
 2.08% :     0.000285s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000033s :      1: rewriter_after_opt_a
 0.55% :     0.000075s :      1: rewriter_before_opt_a
 0.55% :     0.000075s :      1: symbol_engine_optimizer
32.25% :     0.004406s :      1: type_inference


TotalTime = 0.018609, [33]
[bootstrap]: 0.00028966
[type_inference]: 0.00440023
[event_method]: 1.616e-05
[auto_monad]: 5.924e-05
[graph_reusing]: 5.64e-06
[pre_auto_parallel]: 2.44999e-06
[py_interpret_to_execute]: 3.621e-05
[rewriter_before_opt_a]: 6.791e-05
[expand_dump_flag]: 2.96999e-06
[jit_opt_a]: 0.00544584, [2]
    [Cycle 1]: 0.00176957, [27]
        [switch_simplify]: 4.593e-05
        [loop_unroll]: 2.917e-05
        [a_1]: 0.00052901
        [with_stream_mark]: 1.421e-05
        [recompute_prepare]: 7.6e-06
        [updatestate_depend_eliminate]: 3.07002e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [parameter_eliminate]: 1.12999e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 5.06997e-06
        [accelerated_algorithm]: 5.77001e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 5.61e-06
        [merge_forward]: 2.99999e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.261e-05
        [j_node_and_user_rematch]: 9.57999e-06
        [meta_fg_expand]: 3.23e-06
        [replace_old_param]: 8.95001e-06
        [inline_without_move]: 5.58002e-06
        [renormalize]: 0.00083919
        [add_forward_monad_depend]: 5.84999e-06
        [auto_monad_grad]: 2.61e-06
        [auto_monad_eliminator]: 1.19e-05
        [cse]: 2.206e-05
        [replace_applicator]: 1.404e-05
    [Cycle 2]: 0.00032795, [27]
        [switch_simplify]: 5.51998e-06
        [loop_unroll]: 5.15001e-06
        [a_1]: 9.191e-05
        [with_stream_mark]: 9.98002e-06
        [recompute_prepare]: 4.82e-06
        [updatestate_depend_eliminate]: 2.66e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.23002e-06
        [specialize_transform]: 5.04e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 5.32001e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.54002e-06
        [merge_forward]: 2.49001e-06
        [cell_reuse_recompute_pass]: 3.18e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.084e-05
        [j_node_and_user_rematch]: 7.82e-06
        [meta_fg_expand]: 1.91998e-06
        [replace_old_param]: 6.75002e-06
        [inline_without_move]: 4.55999e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 6.59988e-07
        [auto_monad_eliminator]: 5.02e-06
        [cse]: 9.97001e-06
        [replace_applicator]: 5.09e-06
[py_interpret_to_execute_after_opt_a]: 1.115e-05
[rewriter_after_opt_a]: 3.46e-05
[convert_after_rewriter]: 7.07002e-06
[order_py_execute_after_rewriter]: 5.54998e-06
[mutable_eliminate]: 0.00061821
[jit_opt_b]: 4.985e-05, [1]
    [Cycle 1]: 4.294e-05, [2]
        [frontend_op_eliminate]: 1.62e-05
        [inline_after_opt_a]: 1.5e-05
[cconv]: 2.343e-05
[loop_unroll]: 0.00073545
[jit_opt_after_cconv]: 0.00014263, [1]
    [Cycle 1]: 0.00013671, [11]
        [c_1]: 2.228e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 4.80001e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [cse]: 2.136e-05
        [call_graph_tuple_transform]: 1.885e-05
        [tuple_list_get_item_eliminator]: 5.33002e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 5.43002e-06
[remove_dup_value]: 1.245e-05
[partial_unused_args_eliminate]: 2.61e-06
[environ_conv]: 4.70999e-06
[add_recomputation]: 4.397e-05
[cse_after_recomputation]: 1.916e-05, [1]
    [Cycle 1]: 1.427e-05, [1]
        [cse]: 8.72e-06
[auto_monad_reorder]: 1.453e-05
[get_jit_bprop_graph]: 1.92001e-06
[rewriter_after_jit_bprop_graph]: 5.14e-06
[opt_after_jit_grad]: 0.00045892
[symbol_engine_optimizer]: 7.249e-05, [1]
    [Cycle 1]: 6.658e-05, [6]
        [build]: 3.79002e-06
        [elim_shapecalc]: 6.98998e-06
        [elim_not_effective]: 1.253e-05
        [opt_reshape]: 5.81e-06
        [fold_const_symbol]: 8.99003e-06
        [renormalize]: 5.29981e-07
[validate]: 3.407e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00579439
[execute]: 6.15002e-06
Sums
    bootstrap                            :     0.000290s :  1.97%
    type_inference                       :     0.004400s : 29.98%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000059s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.25%
    rewriter_before_opt_a                :     0.000068s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000051s :  0.35%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000621s :  4.23%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000839s :  5.72%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.12%
    jit_opt_a.cse                        :     0.000032s :  0.22%
    jit_opt_a.replace_applicator         :     0.000019s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.08%
    rewriter_after_opt_a                 :     0.000035s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000618s :  4.21%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000023s :  0.16%
    loop_unroll                          :     0.000735s :  5.01%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000044s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000459s :  3.13%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005794s : 39.48%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000207    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.82% :     0.000002s :      2: substitution.fold_const_symbol
 2.63% :     0.000005s :      3: substitution.graph_param_transform
80.94% :     0.000168s :      4: substitution.inline
 1.62% :     0.000003s :      4: substitution.j_node_and_user_rematch
 5.59% :     0.000012s :      4: substitution.remove_not_recompute_node
 1.87% :     0.000004s :      2: substitution.replace_old_param
 5.69% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004340     2
80.37% :     0.003488s :      1: type_inference.infer
19.63% :     0.000852s :      1: type_inference.specialize
------[replace.]   0.000070     6
64.30% :     0.000045s :      4: replace.inline
35.70% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000176     6
93.90% :     0.000165s :      4: match.inline
 6.10% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.31% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.07% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 2.29% :     0.000003s :     10: predicate.arithmetic_simplify
 1.72% :     0.000002s :     10: predicate.cast_eliminate
 0.66% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000001s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.25% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.20% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 4.82% :     0.000005s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      6: predicate.less_batch_normalization
 1.71% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.92% :     0.000002s :     15: predicate.load_eliminater
 1.59% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.26% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.76% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.86% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.08% :     0.000008s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.29% :     0.000001s :     10: predicate.transpose_eliminate
 1.80% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.49% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.94% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000593    12
46.36% :     0.000275s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.15% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.49% :     0.000282s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020380    76
 0.23% :     0.000047s :      1: add_recomputation
 0.31% :     0.000062s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.48% :     0.000302s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.09% :     0.000019s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.74% :     0.005450s :      1: jit_opt_a
 0.71% :     0.000145s :      1: jit_opt_after_cconv
 0.26% :     0.000052s :      1: jit_opt_b
 3.65% :     0.000745s :      1: loop_unroll
 3.07% :     0.000625s :      1: mutable_eliminate
 4.08% :     0.000831s :     26: opt.transform.jit_opt_a
 0.24% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.29% :     0.000466s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000039s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 2.16% :     0.000440s :      1: renormalize.infer
 1.92% :     0.000391s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000038s :      1: rewriter_after_opt_a
 0.35% :     0.000071s :      1: rewriter_before_opt_a
 0.37% :     0.000075s :      1: symbol_engine_optimizer
28.49% :     0.005807s :      1: task_emit
21.63% :     0.004409s :      1: type_inference
 0.25% :     0.000050s :      1: validate


TotalTime = 0.0184501, [33]
[bootstrap]: 0.00028321
[type_inference]: 0.00422222
[event_method]: 1.631e-05
[auto_monad]: 6.282e-05
[graph_reusing]: 5.77001e-06
[pre_auto_parallel]: 2.32001e-06
[py_interpret_to_execute]: 3.509e-05
[rewriter_before_opt_a]: 7.11e-05
[expand_dump_flag]: 2.93998e-06
[jit_opt_a]: 0.00520633, [2]
    [Cycle 1]: 0.00157542, [27]
        [switch_simplify]: 5.263e-05
        [loop_unroll]: 2.988e-05
        [a_1]: 0.00051202
        [with_stream_mark]: 1.775e-05
        [recompute_prepare]: 7.18e-06
        [updatestate_depend_eliminate]: 3.55e-06
        [updatestate_assign_eliminate]: 2.76e-06
        [updatestate_loads_eliminate]: 2.30002e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 6.01e-06
        [updatestate_useless_node_eliminater]: 5.25001e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 1.83002e-06
        [get_grad_eliminate_]: 5.15999e-06
        [merge_forward]: 3.53999e-06
        [cell_reuse_recompute_pass]: 1.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.395e-05
        [j_node_and_user_rematch]: 9.02e-06
        [meta_fg_expand]: 2.27999e-06
        [replace_old_param]: 8.40001e-06
        [inline_without_move]: 5.71e-06
        [renormalize]: 0.00065902
        [add_forward_monad_depend]: 5.58002e-06
        [auto_monad_grad]: 2.61999e-06
        [auto_monad_eliminator]: 1.279e-05
        [cse]: 1.932e-05
        [replace_applicator]: 1.231e-05
    [Cycle 2]: 0.00032425, [27]
        [switch_simplify]: 5.69e-06
        [loop_unroll]: 5.51998e-06
        [a_1]: 9.028e-05
        [with_stream_mark]: 9.42999e-06
        [recompute_prepare]: 4.90001e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 1.89999e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.22999e-06
        [specialize_transform]: 5.14998e-06
        [updatestate_useless_node_eliminater]: 5.10999e-06
        [accelerated_algorithm]: 5.14998e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.55001e-06
        [merge_forward]: 2.34999e-06
        [cell_reuse_recompute_pass]: 1.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.281e-05
        [j_node_and_user_rematch]: 8.43999e-06
        [meta_fg_expand]: 1.97999e-06
        [replace_old_param]: 7.6e-06
        [inline_without_move]: 4.74e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.27999e-06
        [auto_monad_grad]: 1.05001e-06
        [auto_monad_eliminator]: 4.85001e-06
        [cse]: 1.068e-05
        [replace_applicator]: 5.39e-06
[py_interpret_to_execute_after_opt_a]: 9.71e-06
[rewriter_after_opt_a]: 3.342e-05
[convert_after_rewriter]: 7.46999e-06
[order_py_execute_after_rewriter]: 4.72998e-06
[mutable_eliminate]: 0.00056284
[jit_opt_b]: 5.34e-05, [1]
    [Cycle 1]: 4.585e-05, [2]
        [frontend_op_eliminate]: 1.82e-05
        [inline_after_opt_a]: 1.55e-05
[cconv]: 2.78e-05
[loop_unroll]: 0.00043676
[jit_opt_after_cconv]: 0.00014231, [1]
    [Cycle 1]: 0.00013551, [11]
        [c_1]: 2.246e-05
        [parameter_eliminate]: 2.76e-06
        [updatestate_depend_eliminate]: 5.32999e-06
        [updatestate_assign_eliminate]: 2.38998e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [cse]: 1.856e-05
        [call_graph_tuple_transform]: 1.887e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 5.24e-06
[remove_dup_value]: 1.333e-05
[partial_unused_args_eliminate]: 2.06e-06
[environ_conv]: 5.03002e-06
[add_recomputation]: 4.567e-05
[cse_after_recomputation]: 1.945e-05, [1]
    [Cycle 1]: 1.419e-05, [1]
        [cse]: 8.78001e-06
[auto_monad_reorder]: 1.461e-05
[get_jit_bprop_graph]: 1.90001e-06
[rewriter_after_jit_bprop_graph]: 5.25999e-06
[opt_after_jit_grad]: 0.00045436
[symbol_engine_optimizer]: 7.148e-05, [1]
    [Cycle 1]: 6.48e-05, [6]
        [build]: 3.09999e-06
        [elim_shapecalc]: 7.44002e-06
        [elim_not_effective]: 1.241e-05
        [opt_reshape]: 5.61998e-06
        [fold_const_symbol]: 8.64e-06
        [renormalize]: 6.30011e-07
[validate]: 3.289e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00640815
[execute]: 8.32998e-06
Sums
    bootstrap                            :     0.000283s :  1.95%
    type_inference                       :     0.004222s : 29.00%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000063s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.24%
    rewriter_before_opt_a                :     0.000071s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000602s :  4.14%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000659s :  4.53%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000030s :  0.21%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000563s :  3.87%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000028s :  0.19%
    loop_unroll                          :     0.000437s :  3.00%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000046s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000454s :  3.12%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000033s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006408s : 44.02%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000185    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 3.10% :     0.000006s :      3: substitution.graph_param_transform
82.76% :     0.000153s :      4: substitution.inline
 1.69% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.30% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000004s :      2: substitution.replace_old_param
 6.33% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004166     2
81.30% :     0.003387s :      1: type_inference.infer
18.70% :     0.000779s :      1: type_inference.specialize
------[replace.]   0.000066     6
66.04% :     0.000043s :      4: replace.inline
33.96% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000161     6
93.40% :     0.000150s :      4: match.inline
 6.60% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.28% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.44% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 2.19% :     0.000003s :     10: predicate.arithmetic_simplify
 1.19% :     0.000001s :     10: predicate.cast_eliminate
 0.58% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.79% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.12% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.35% :     0.000000s :      3: predicate.graph_param_transform
 5.83% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.41% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.56% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.06% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.28% :     0.000001s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.13% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.70% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.30% :     0.000002s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.91% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.87% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000522    12
48.31% :     0.000252s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.60% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.10% :     0.000241s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020029    76
 0.24% :     0.000049s :      1: add_recomputation
 0.33% :     0.000066s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.48% :     0.000296s :      1: bootstrap
 0.15% :     0.000031s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.01% :     0.005210s :      1: jit_opt_a
 0.72% :     0.000145s :      1: jit_opt_after_cconv
 0.28% :     0.000056s :      1: jit_opt_b
 2.22% :     0.000445s :      1: loop_unroll
 2.85% :     0.000572s :      1: mutable_eliminate
 4.05% :     0.000811s :     26: opt.transform.jit_opt_a
 0.24% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.31% :     0.000463s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000038s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.79% :     0.000359s :      1: renormalize.infer
 1.46% :     0.000293s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000036s :      1: rewriter_after_opt_a
 0.37% :     0.000074s :      1: rewriter_before_opt_a
 0.37% :     0.000074s :      1: symbol_engine_optimizer
32.05% :     0.006419s :      1: task_emit
21.13% :     0.004232s :      1: type_inference
 0.26% :     0.000053s :      1: validate


TotalTime = 0.0137452, [30]
[bootstrap]: 0.00028478
[type_inference]: 0.00438958
[event_method]: 1.723e-05
[auto_monad]: 6.502e-05
[graph_reusing]: 5.35001e-06
[pre_auto_parallel]: 3.28e-06
[py_interpret_to_execute]: 3.479e-05
[rewriter_before_opt_a]: 8.648e-05
[expand_dump_flag]: 3.14001e-06
[jit_opt_a]: 0.00661335, [2]
    [Cycle 1]: 0.00281019, [27]
        [switch_simplify]: 5.662e-05
        [loop_unroll]: 3.049e-05
        [a_1]: 0.00166402
        [with_stream_mark]: 1.863e-05
        [recompute_prepare]: 8.08999e-06
        [updatestate_depend_eliminate]: 3.22002e-06
        [updatestate_assign_eliminate]: 3.24001e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 5.79e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.43002e-06
        [meta_shard_fg_expand]: 2.31e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 4.03999e-06
        [cell_reuse_recompute_pass]: 1.81e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.508e-05
        [j_node_and_user_rematch]: 9.59e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 1.009e-05
        [inline_without_move]: 5.44e-06
        [renormalize]: 0.000712
        [add_forward_monad_depend]: 5.74e-06
        [auto_monad_grad]: 2.59999e-06
        [auto_monad_eliminator]: 1.364e-05
        [cse]: 2.576e-05
        [replace_applicator]: 1.165e-05
    [Cycle 2]: 0.00031938, [27]
        [switch_simplify]: 5.69e-06
        [loop_unroll]: 4.82e-06
        [a_1]: 8.842e-05
        [with_stream_mark]: 9.15999e-06
        [recompute_prepare]: 4.72e-06
        [updatestate_depend_eliminate]: 2.38998e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 1.18001e-06
        [specialize_transform]: 4.67998e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.05999e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 4.51002e-06
        [merge_forward]: 2.26998e-06
        [cell_reuse_recompute_pass]: 1.69998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.324e-05
        [j_node_and_user_rematch]: 7.25998e-06
        [meta_fg_expand]: 1.73002e-06
        [replace_old_param]: 7.26999e-06
        [inline_without_move]: 4.77e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 1.09998e-06
        [auto_monad_eliminator]: 4.75001e-06
        [cse]: 1.151e-05
        [replace_applicator]: 4.85999e-06
[py_interpret_to_execute_after_opt_a]: 1.027e-05
[rewriter_after_opt_a]: 3.197e-05
[convert_after_rewriter]: 7.25998e-06
[order_py_execute_after_rewriter]: 7.97003e-06
[mutable_eliminate]: 0.00063055
[jit_opt_b]: 5.124e-05, [1]
    [Cycle 1]: 4.375e-05, [2]
        [frontend_op_eliminate]: 1.768e-05
        [inline_after_opt_a]: 1.455e-05
[cconv]: 2.541e-05
[loop_unroll]: 0.00042032
[jit_opt_after_cconv]: 0.00020438, [1]
    [Cycle 1]: 0.0001978, [11]
        [c_1]: 2.022e-05
        [parameter_eliminate]: 2.48998e-06
        [updatestate_depend_eliminate]: 5.20999e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [cse]: 8.265e-05
        [call_graph_tuple_transform]: 1.919e-05
        [tuple_list_get_item_eliminator]: 5.65001e-06
        [none_parameter_eliminate]: 1.86998e-06
        [renormalize]: 5.29981e-07
        [switch_simplify]: 5.31002e-06
[remove_dup_value]: 1.438e-05
[partial_unused_args_eliminate]: 2.22999e-06
[environ_conv]: 5.46e-06
[add_recomputation]: 4.735e-05
[cse_after_recomputation]: 2.11e-05, [1]
    [Cycle 1]: 1.522e-05, [1]
        [cse]: 9.81998e-06
[auto_monad_reorder]: 1.502e-05
[get_jit_bprop_graph]: 1.61998e-06
[rewriter_after_jit_bprop_graph]: 6.17999e-06
[opt_after_jit_grad]: 0.00046261
[symbol_engine_optimizer]: 7.365e-05, [1]
    [Cycle 1]: 6.645e-05, [6]
        [build]: 3.7e-06
        [elim_shapecalc]: 7.9e-06
        [elim_not_effective]: 1.252e-05
        [opt_reshape]: 5.29998e-06
        [fold_const_symbol]: 8.52998e-06
        [renormalize]: 6.40022e-07
[validate]: 3.682e-05
Sums
    bootstrap                            :     0.000285s :  2.94%
    type_inference                       :     0.004390s : 45.36%
    event_method                         :     0.000017s :  0.18%
    auto_monad                           :     0.000065s :  0.67%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.36%
    rewriter_before_opt_a                :     0.000086s :  0.89%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000062s :  0.64%
    jit_opt_a.loop_unroll                :     0.000035s :  0.36%
    jit_opt_a.a_1                        :     0.001752s : 18.11%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000010s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.18%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000712s :  7.36%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.19%
    jit_opt_a.cse                        :     0.000037s :  0.39%
    jit_opt_a.replace_applicator         :     0.000017s :  0.17%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.11%
    rewriter_after_opt_a                 :     0.000032s :  0.33%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000008s :  0.08%
    mutable_eliminate                    :     0.000631s :  6.52%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.15%
    cconv                                :     0.000025s :  0.26%
    loop_unroll                          :     0.000420s :  4.34%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000083s :  0.85%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000047s :  0.49%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000463s :  4.78%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000037s :  0.38%


Time group info:
------[substitution.]   0.001304    23
 0.14% :     0.000002s :      2: substitution.elim_not_effective
 0.10% :     0.000001s :      2: substitution.fold_const_symbol
 0.40% :     0.000005s :      3: substitution.graph_param_transform
97.32% :     0.001269s :      4: substitution.inline
 0.24% :     0.000003s :      4: substitution.j_node_and_user_rematch
 0.37% :     0.000005s :      4: substitution.remove_not_recompute_node
 0.35% :     0.000005s :      2: substitution.replace_old_param
 1.08% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004327     2
79.60% :     0.003444s :      1: type_inference.infer
20.40% :     0.000883s :      1: type_inference.specialize
------[replace.]   0.000087     6
64.13% :     0.000056s :      4: replace.inline
35.87% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.001278     6
99.01% :     0.001266s :      4: match.inline
 0.99% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.32% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.66% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000001s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.41% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.34% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.51% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.84% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.60% :     0.000002s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.60% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.45% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.17% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.33% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001748    12
18.89% :     0.000330s :      5: func_graph_cloner_run.FuncGraphClonerGraph
64.90% :     0.001135s :      1: func_graph_cloner_run.FuncGraphClonerNode
16.21% :     0.000283s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016475    72
 0.30% :     0.000050s :      1: add_recomputation
 0.41% :     0.000068s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 1.80% :     0.000297s :      1: bootstrap
 0.17% :     0.000028s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.13% :     0.000021s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
40.16% :     0.006617s :      1: jit_opt_a
 1.26% :     0.000207s :      1: jit_opt_after_cconv
 0.33% :     0.000054s :      1: jit_opt_b
 2.60% :     0.000428s :      1: loop_unroll
 3.88% :     0.000639s :      1: mutable_eliminate
11.92% :     0.001964s :     26: opt.transform.jit_opt_a
 0.28% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.86% :     0.000472s :      1: opt_after_jit_grad
 0.06% :     0.000010s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000037s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.24% :     0.000369s :      1: renormalize.infer
 2.04% :     0.000336s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000035s :      1: rewriter_after_opt_a
 0.55% :     0.000090s :      1: rewriter_before_opt_a
 0.46% :     0.000076s :      1: symbol_engine_optimizer
26.73% :     0.004404s :      1: type_inference


TotalTime = 0.0183651, [33]
[bootstrap]: 0.00029274
[type_inference]: 0.00460723
[event_method]: 1.664e-05
[auto_monad]: 6.059e-05
[graph_reusing]: 5.24e-06
[pre_auto_parallel]: 2.62001e-06
[py_interpret_to_execute]: 3.498e-05
[rewriter_before_opt_a]: 7.05e-05
[expand_dump_flag]: 3.03998e-06
[jit_opt_a]: 0.00541423, [2]
    [Cycle 1]: 0.00168737, [27]
        [switch_simplify]: 5.005e-05
        [loop_unroll]: 2.858e-05
        [a_1]: 0.00054003
        [with_stream_mark]: 1.691e-05
        [recompute_prepare]: 7.25e-06
        [updatestate_depend_eliminate]: 3.23998e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 1.50001e-06
        [specialize_transform]: 5.86003e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.56e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 5.35001e-06
        [merge_forward]: 2.71e-06
        [cell_reuse_recompute_pass]: 8.89995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.238e-05
        [j_node_and_user_rematch]: 8.82e-06
        [meta_fg_expand]: 2.12001e-06
        [replace_old_param]: 8.40001e-06
        [inline_without_move]: 5.45001e-06
        [renormalize]: 0.00075874
        [add_forward_monad_depend]: 5.51e-06
        [auto_monad_grad]: 2.27999e-06
        [auto_monad_eliminator]: 1.247e-05
        [cse]: 1.967e-05
        [replace_applicator]: 1.12e-05
    [Cycle 2]: 0.00031785, [27]
        [switch_simplify]: 5.51e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 8.947e-05
        [with_stream_mark]: 8.72998e-06
        [recompute_prepare]: 5.57999e-06
        [updatestate_depend_eliminate]: 2.49001e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 4.78001e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 5.02999e-06
        [meta_shard_fg_expand]: 1.28002e-06
        [get_grad_eliminate_]: 4.82998e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 2.22001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.289e-05
        [j_node_and_user_rematch]: 7.98999e-06
        [meta_fg_expand]: 1.64e-06
        [replace_old_param]: 6.86001e-06
        [inline_without_move]: 4.50001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.16997e-06
        [auto_monad_grad]: 6.70028e-07
        [auto_monad_eliminator]: 5.00999e-06
        [cse]: 9.64e-06
        [replace_applicator]: 5.46e-06
[py_interpret_to_execute_after_opt_a]: 1.044e-05
[rewriter_after_opt_a]: 3.257e-05
[convert_after_rewriter]: 7.09001e-06
[order_py_execute_after_rewriter]: 5.00001e-06
[mutable_eliminate]: 0.0005514
[jit_opt_b]: 4.864e-05, [1]
    [Cycle 1]: 4.184e-05, [2]
        [frontend_op_eliminate]: 1.618e-05
        [inline_after_opt_a]: 1.42e-05
[cconv]: 2.17e-05
[loop_unroll]: 0.00043608
[jit_opt_after_cconv]: 0.00022704, [1]
    [Cycle 1]: 0.00022007, [11]
        [c_1]: 5.999e-05
        [parameter_eliminate]: 3.2e-06
        [updatestate_depend_eliminate]: 5.64e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [cse]: 1.79e-05
        [call_graph_tuple_transform]: 5.944e-05
        [tuple_list_get_item_eliminator]: 6.06998e-06
        [none_parameter_eliminate]: 2.32999e-06
        [renormalize]: 3.50003e-07
        [switch_simplify]: 5.52001e-06
[remove_dup_value]: 1.33e-05
[partial_unused_args_eliminate]: 2.16998e-06
[environ_conv]: 4.80001e-06
[add_recomputation]: 4.511e-05
[cse_after_recomputation]: 2.156e-05, [1]
    [Cycle 1]: 1.589e-05, [1]
        [cse]: 1.031e-05
[auto_monad_reorder]: 1.319e-05
[get_jit_bprop_graph]: 1.86e-06
[rewriter_after_jit_bprop_graph]: 4.50999e-06
[opt_after_jit_grad]: 0.00046138
[symbol_engine_optimizer]: 7.292e-05, [1]
    [Cycle 1]: 6.568e-05, [6]
        [build]: 3.26001e-06
        [elim_shapecalc]: 7.77998e-06
        [elim_not_effective]: 1.197e-05
        [opt_reshape]: 5.52999e-06
        [fold_const_symbol]: 8.43999e-06
        [renormalize]: 3.89991e-07
[validate]: 3.429e-05
[backend_pass]: 7.89994e-07
[task_emit]: 0.00564539
[execute]: 6.39999e-06
Sums
    bootstrap                            :     0.000293s :  2.04%
    type_inference                       :     0.004607s : 32.08%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000061s :  0.42%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.24%
    rewriter_before_opt_a                :     0.000070s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000629s :  4.38%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000759s :  5.28%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.12%
    jit_opt_a.cse                        :     0.000029s :  0.20%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000551s :  3.84%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000022s :  0.15%
    loop_unroll                          :     0.000436s :  3.04%
    jit_opt_after_cconv.c_1              :     0.000060s :  0.42%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000059s :  0.41%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000045s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000013s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000461s :  3.21%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005645s : 39.31%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000207    23
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.66% :     0.000006s :      3: substitution.graph_param_transform
85.03% :     0.000176s :      4: substitution.inline
 1.40% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.43% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.47% :     0.000003s :      2: substitution.replace_old_param
 5.60% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004545     2
80.24% :     0.003647s :      1: type_inference.infer
19.76% :     0.000898s :      1: type_inference.specialize
------[replace.]   0.000072     6
65.83% :     0.000047s :      4: replace.inline
34.17% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000184     6
94.27% :     0.000174s :      4: match.inline
 5.73% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.25% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.48% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 2.02% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000001s :     10: predicate.cast_eliminate
 0.65% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.60% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000000s :      3: predicate.graph_param_transform
 5.42% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.75% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.60% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.93% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.26% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000000s :      3: predicate.opt_reshape
 2.05% :     0.000002s :     15: predicate.partial_eliminate
 1.25% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     10: predicate.reduce_eliminate
 1.75% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.73% :     0.000003s :     16: predicate.switch_defer_inline
 2.43% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.98% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.29% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.50% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000636    12
45.99% :     0.000293s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.73% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.28% :     0.000307s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020134    76
 0.24% :     0.000048s :      1: add_recomputation
 0.32% :     0.000064s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.52% :     0.000306s :      1: bootstrap
 0.12% :     0.000024s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.91% :     0.005418s :      1: jit_opt_a
 1.14% :     0.000230s :      1: jit_opt_after_cconv
 0.25% :     0.000051s :      1: jit_opt_b
 2.21% :     0.000445s :      1: loop_unroll
 2.78% :     0.000560s :      1: mutable_eliminate
 4.13% :     0.000832s :     26: opt.transform.jit_opt_a
 0.63% :     0.000127s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.33% :     0.000469s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000037s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.98% :     0.000399s :      1: renormalize.infer
 1.74% :     0.000351s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000036s :      1: rewriter_after_opt_a
 0.36% :     0.000073s :      1: rewriter_before_opt_a
 0.37% :     0.000075s :      1: symbol_engine_optimizer
28.10% :     0.005658s :      1: task_emit
22.95% :     0.004620s :      1: type_inference
 0.26% :     0.000052s :      1: validate


TotalTime = 0.0179685, [33]
[bootstrap]: 0.00027624
[type_inference]: 0.00421077
[event_method]: 1.565e-05
[auto_monad]: 4.326e-05
[graph_reusing]: 3.48e-06
[pre_auto_parallel]: 1.84e-06
[py_interpret_to_execute]: 2.84e-05
[rewriter_before_opt_a]: 6.378e-05
[expand_dump_flag]: 2.18002e-06
[jit_opt_a]: 0.0050522, [2]
    [Cycle 1]: 0.00158137, [27]
        [switch_simplify]: 5.793e-05
        [loop_unroll]: 3.035e-05
        [a_1]: 0.000531
        [with_stream_mark]: 1.86e-05
        [recompute_prepare]: 7.12002e-06
        [updatestate_depend_eliminate]: 3.48e-06
        [updatestate_assign_eliminate]: 3.51001e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [parameter_eliminate]: 1.83002e-06
        [specialize_transform]: 6.02001e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 5.29e-06
        [meta_shard_fg_expand]: 2.17001e-06
        [get_grad_eliminate_]: 5.37999e-06
        [merge_forward]: 3.93001e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.338e-05
        [j_node_and_user_rematch]: 9.24e-06
        [meta_fg_expand]: 2.73e-06
        [replace_old_param]: 8.36002e-06
        [inline_without_move]: 5.47001e-06
        [renormalize]: 0.00064332
        [add_forward_monad_depend]: 5.55001e-06
        [auto_monad_grad]: 2.59999e-06
        [auto_monad_eliminator]: 1.422e-05
        [cse]: 1.843e-05
        [replace_applicator]: 1.183e-05
    [Cycle 2]: 0.00032267, [27]
        [switch_simplify]: 6.01998e-06
        [loop_unroll]: 4.79e-06
        [a_1]: 9.081e-05
        [with_stream_mark]: 8.72e-06
        [recompute_prepare]: 4.93001e-06
        [updatestate_depend_eliminate]: 2.53e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 9.79984e-07
        [specialize_transform]: 5.00999e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.19e-06
        [meta_shard_fg_expand]: 1.02e-06
        [get_grad_eliminate_]: 4.65999e-06
        [merge_forward]: 2.39001e-06
        [cell_reuse_recompute_pass]: 1.56002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.263e-05
        [j_node_and_user_rematch]: 8.08999e-06
        [meta_fg_expand]: 1.82001e-06
        [replace_old_param]: 7.18e-06
        [inline_without_move]: 4.70001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 9.30013e-07
        [auto_monad_eliminator]: 4.76002e-06
        [cse]: 1.044e-05
        [replace_applicator]: 5.09e-06
[py_interpret_to_execute_after_opt_a]: 1.069e-05
[rewriter_after_opt_a]: 3.308e-05
[convert_after_rewriter]: 6.82002e-06
[order_py_execute_after_rewriter]: 5.34e-06
[mutable_eliminate]: 0.00055921
[jit_opt_b]: 6.547e-05, [1]
    [Cycle 1]: 5.785e-05, [2]
        [frontend_op_eliminate]: 1.699e-05
        [inline_after_opt_a]: 2.825e-05
[cconv]: 2.74e-05
[loop_unroll]: 0.00049846
[jit_opt_after_cconv]: 0.00014225, [1]
    [Cycle 1]: 0.00013575, [11]
        [c_1]: 2.19e-05
        [parameter_eliminate]: 2.98e-06
        [updatestate_depend_eliminate]: 6.00002e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 1.998e-05
        [call_graph_tuple_transform]: 1.848e-05
        [tuple_list_get_item_eliminator]: 5.37001e-06
        [none_parameter_eliminate]: 1.46998e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 5.15001e-06
[remove_dup_value]: 1.263e-05
[partial_unused_args_eliminate]: 2.43e-06
[environ_conv]: 5.35001e-06
[add_recomputation]: 4.562e-05
[cse_after_recomputation]: 2.028e-05, [1]
    [Cycle 1]: 1.484e-05, [1]
        [cse]: 9.02e-06
[auto_monad_reorder]: 1.608e-05
[get_jit_bprop_graph]: 2.12999e-06
[rewriter_after_jit_bprop_graph]: 5.22e-06
[opt_after_jit_grad]: 0.00046502
[symbol_engine_optimizer]: 7.233e-05, [1]
    [Cycle 1]: 6.561e-05, [6]
        [build]: 4.09997e-06
        [elim_shapecalc]: 7.13998e-06
        [elim_not_effective]: 1.265e-05
        [opt_reshape]: 5.46998e-06
        [fold_const_symbol]: 8.35001e-06
        [renormalize]: 5.20027e-07
[validate]: 5.329e-05
[backend_pass]: 1.55999e-06
[task_emit]: 0.00602592
[execute]: 8e-06
Sums
    bootstrap                            :     0.000276s :  1.94%
    type_inference                       :     0.004211s : 29.58%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000043s :  0.30%
    graph_reusing                        :     0.000003s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.20%
    rewriter_before_opt_a                :     0.000064s :  0.45%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.45%
    jit_opt_a.loop_unroll                :     0.000035s :  0.25%
    jit_opt_a.a_1                        :     0.000622s :  4.37%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000643s :  4.52%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000029s :  0.20%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.08%
    rewriter_after_opt_a                 :     0.000033s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000559s :  3.93%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000028s :  0.20%
    cconv                                :     0.000027s :  0.19%
    loop_unroll                          :     0.000498s :  3.50%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000046s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000465s :  3.27%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000053s :  0.37%
    backend_pass                         :     0.000002s :  0.01%
    task_emit                            :     0.006026s : 42.34%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000205    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.56% :     0.000005s :      3: substitution.graph_param_transform
84.44% :     0.000173s :      4: substitution.inline
 1.54% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.21% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.92% :     0.000004s :      2: substitution.replace_old_param
 5.75% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004159     2
81.92% :     0.003407s :      1: type_inference.infer
18.08% :     0.000752s :      1: type_inference.specialize
------[replace.]   0.000065     6
66.77% :     0.000044s :      4: replace.inline
33.23% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000182     6
94.13% :     0.000171s :      4: match.inline
 5.87% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.10% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.44% :     0.000002s :     10: predicate.addn_zero_filter
 1.79% :     0.000002s :     10: predicate.arithmetic_simplify
 1.11% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.17% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.08% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.01% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.69% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.08% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.36% :     0.000007s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000001s :      6: predicate.j_node_and_user_rematch
 0.94% :     0.000001s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.58% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.95% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.67% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.44% :     0.000003s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.29% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.17% :     0.000001s :     10: predicate.reshape_eliminate
 1.08% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.04% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.23% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.10% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.58% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.15% :     0.000003s :     16: predicate.switch_defer_inline
 1.93% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.86% :     0.000009s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.02% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
13.08% :     0.000017s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.29% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.33% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000530    12
47.39% :     0.000251s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.88% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.73% :     0.000237s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019570    76
 0.25% :     0.000049s :      1: add_recomputation
 0.24% :     0.000047s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 1.47% :     0.000288s :      1: bootstrap
 0.15% :     0.000030s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
25.83% :     0.005056s :      1: jit_opt_a
 0.74% :     0.000145s :      1: jit_opt_after_cconv
 0.35% :     0.000068s :      1: jit_opt_b
 2.59% :     0.000507s :      1: loop_unroll
 2.90% :     0.000568s :      1: mutable_eliminate
 4.26% :     0.000833s :     26: opt.transform.jit_opt_a
 0.24% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000038s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.42% :     0.000473s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000031s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.72% :     0.000337s :      1: renormalize.infer
 1.52% :     0.000298s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000036s :      1: rewriter_after_opt_a
 0.34% :     0.000067s :      1: rewriter_before_opt_a
 0.38% :     0.000075s :      1: symbol_engine_optimizer
30.86% :     0.006039s :      1: task_emit
21.57% :     0.004222s :      1: type_inference
 0.38% :     0.000075s :      1: validate


TotalTime = 0.0967147, [33]
[bootstrap]: 0.00022161
[type_inference]: 0.0729372
[event_method]: 0.00029336
[auto_monad]: 0.00010414
[graph_reusing]: 6.96001e-06
[pre_auto_parallel]: 3.26999e-06
[py_interpret_to_execute]: 4.193e-05
[rewriter_before_opt_a]: 0.00010718
[expand_dump_flag]: 3.3e-06
[jit_opt_a]: 0.020765, [3]
    [Cycle 1]: 0.0135757, [27]
        [switch_simplify]: 6.823e-05
        [loop_unroll]: 4.295e-05
        [a_1]: 0.00093227
        [with_stream_mark]: 2.623e-05
        [recompute_prepare]: 1.977e-05
        [updatestate_depend_eliminate]: 8.02e-06
        [updatestate_assign_eliminate]: 6.64999e-06
        [updatestate_loads_eliminate]: 6.16e-06
        [parameter_eliminate]: 2.59001e-06
        [specialize_transform]: 1.483e-05
        [updatestate_useless_node_eliminater]: 1.37e-05
        [accelerated_algorithm]: 1.351e-05
        [meta_shard_fg_expand]: 3.70998e-06
        [get_grad_eliminate_]: 1.339e-05
        [merge_forward]: 8.64e-06
        [cell_reuse_recompute_pass]: 1.00999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.787e-05
        [j_node_and_user_rematch]: 2.386e-05
        [meta_fg_expand]: 0.00149891
        [replace_old_param]: 5.961e-05
        [inline_without_move]: 5.492e-05
        [renormalize]: 0.0101332
        [add_forward_monad_depend]: 1.219e-05
        [auto_monad_grad]: 6.64999e-06
        [auto_monad_eliminator]: 5.427e-05
        [cse]: 0.00020789
        [replace_applicator]: 8.566e-05
    [Cycle 2]: 0.00358587, [27]
        [switch_simplify]: 4.314e-05
        [loop_unroll]: 4.172e-05
        [a_1]: 0.00119136
        [with_stream_mark]: 1.666e-05
        [recompute_prepare]: 1.01e-05
        [updatestate_depend_eliminate]: 5.61e-06
        [updatestate_assign_eliminate]: 4.4e-06
        [updatestate_loads_eliminate]: 3.89002e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 7.58999e-06
        [updatestate_useless_node_eliminater]: 9.30001e-06
        [accelerated_algorithm]: 7.88001e-06
        [meta_shard_fg_expand]: 2.84999e-06
        [get_grad_eliminate_]: 7.31001e-06
        [merge_forward]: 5.56998e-06
        [cell_reuse_recompute_pass]: 1.24998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.641e-05
        [j_node_and_user_rematch]: 1.198e-05
        [meta_fg_expand]: 6.037e-05
        [replace_old_param]: 8.24002e-06
        [inline_without_move]: 7.14001e-06
        [renormalize]: 0.00191246
        [add_forward_monad_depend]: 5.62001e-06
        [auto_monad_grad]: 1.62001e-06
        [auto_monad_eliminator]: 1.084e-05
        [cse]: 1.845e-05
        [replace_applicator]: 1.2e-05
    [Cycle 3]: 0.00023498, [27]
        [switch_simplify]: 4.17998e-06
        [loop_unroll]: 3.76999e-06
        [a_1]: 3.936e-05
        [with_stream_mark]: 8.64e-06
        [recompute_prepare]: 3.68999e-06
        [updatestate_depend_eliminate]: 2.16998e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.47001e-06
        [parameter_eliminate]: 1.47999e-06
        [specialize_transform]: 3.81001e-06
        [updatestate_useless_node_eliminater]: 3.28998e-06
        [accelerated_algorithm]: 3.7e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 3.28e-06
        [merge_forward]: 2.34999e-06
        [cell_reuse_recompute_pass]: 2.19001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.99001e-06
        [j_node_and_user_rematch]: 5.72999e-06
        [meta_fg_expand]: 1.22e-06
        [replace_old_param]: 3.29001e-06
        [inline_without_move]: 3.04999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.05999e-06
        [auto_monad_grad]: 5.89993e-07
        [auto_monad_eliminator]: 3.49001e-06
        [cse]: 7.90998e-06
        [replace_applicator]: 3.48e-06
[py_interpret_to_execute_after_opt_a]: 1.076e-05
[rewriter_after_opt_a]: 3.424e-05
[convert_after_rewriter]: 5.94999e-06
[order_py_execute_after_rewriter]: 4.33001e-06
[mutable_eliminate]: 0.00068424
[jit_opt_b]: 4.262e-05, [1]
    [Cycle 1]: 3.419e-05, [2]
        [frontend_op_eliminate]: 1.169e-05
        [inline_after_opt_a]: 1.197e-05
[cconv]: 2.368e-05
[loop_unroll]: 0.00042367
[jit_opt_after_cconv]: 0.00012037, [1]
    [Cycle 1]: 0.00011411, [11]
        [c_1]: 1.281e-05
        [parameter_eliminate]: 2.49999e-06
        [updatestate_depend_eliminate]: 4.69998e-06
        [updatestate_assign_eliminate]: 1.71998e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [cse]: 1.564e-05
        [call_graph_tuple_transform]: 1.4e-05
        [tuple_list_get_item_eliminator]: 4.72e-06
        [none_parameter_eliminate]: 1.41002e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 4e-06
[remove_dup_value]: 1.396e-05
[partial_unused_args_eliminate]: 1.92999e-06
[environ_conv]: 4.65001e-06
[add_recomputation]: 3.698e-05
[cse_after_recomputation]: 2.037e-05, [1]
    [Cycle 1]: 1.433e-05, [1]
        [cse]: 7.95e-06
[auto_monad_reorder]: 1.355e-05
[get_jit_bprop_graph]: 2.12999e-06
[rewriter_after_jit_bprop_graph]: 4.52e-06
[opt_after_jit_grad]: 0.0004568
[symbol_engine_optimizer]: 6.658e-05, [1]
    [Cycle 1]: 5.977e-05, [6]
        [build]: 4.74998e-06
        [elim_shapecalc]: 6.22001e-06
        [elim_not_effective]: 1.054e-05
        [opt_reshape]: 4.33001e-06
        [fold_const_symbol]: 6.42001e-06
        [renormalize]: 4.50003e-07
[validate]: 3.059e-05
[backend_pass]: 9.39996e-07
[task_emit]: 2.659e-05
[execute]: 1.29003e-06
Sums
    bootstrap                            :     0.000222s :  0.24%
    type_inference                       :     0.072937s : 78.83%
    event_method                         :     0.000293s :  0.32%
    auto_monad                           :     0.000104s :  0.11%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000042s :  0.05%
    rewriter_before_opt_a                :     0.000107s :  0.12%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000116s :  0.12%
    jit_opt_a.loop_unroll                :     0.000088s :  0.10%
    jit_opt_a.a_1                        :     0.002163s :  2.34%
    jit_opt_a.with_stream_mark           :     0.000052s :  0.06%
    jit_opt_a.recompute_prepare          :     0.000034s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.01%
    jit_opt_a.specialize_transform       :     0.000026s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000025s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.03%
    jit_opt_a.merge_forward              :     0.000017s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.00%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000054s :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.001561s :  1.69%
    jit_opt_a.replace_old_param          :     0.000071s :  0.08%
    jit_opt_a.inline_without_move        :     0.000065s :  0.07%
    jit_opt_a.renormalize                :     0.012046s : 13.02%
    jit_opt_a.add_forward_monad_depend   :     0.000019s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000069s :  0.07%
    jit_opt_a.cse                        :     0.000234s :  0.25%
    jit_opt_a.replace_applicator         :     0.000101s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.01%
    rewriter_after_opt_a                 :     0.000034s :  0.04%
    convert_after_rewriter               :     0.000006s :  0.01%
    order_py_execute_after_rewriter      :     0.000004s :  0.00%
    mutable_eliminate                    :     0.000684s :  0.74%
    jit_opt_b.frontend_op_eliminate      :     0.000012s :  0.01%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.01%
    cconv                                :     0.000024s :  0.03%
    loop_unroll                          :     0.000424s :  0.46%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.01%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000016s :  0.02%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.02%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.00%
    remove_dup_value                     :     0.000014s :  0.02%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000005s :  0.01%
    add_recomputation                    :     0.000037s :  0.04%
    cse_after_recomputation.cse          :     0.000008s :  0.01%
    auto_monad_reorder                   :     0.000014s :  0.01%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.00%
    opt_after_jit_grad                   :     0.000457s :  0.49%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.01%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.00%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.03%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000027s :  0.03%
    execute                              :     0.000001s :  0.00%


Time group info:
------[substitution.]   0.000529   102
 0.33% :     0.000002s :      1: substitution.elim_not_effective
 0.22% :     0.000001s :      1: substitution.fold_const_symbol
 0.77% :     0.000004s :      1: substitution.graph_param_transform
63.73% :     0.000337s :     12: substitution.inline
 2.87% :     0.000015s :      2: substitution.inline_without_move
 1.67% :     0.000009s :     12: substitution.j_node_and_user_rematch
 2.04% :     0.000011s :      7: substitution.minmaximum_grad
 1.37% :     0.000007s :      5: substitution.partial_eliminate
 1.80% :     0.000010s :     12: substitution.remove_not_recompute_node
 5.52% :     0.000029s :     10: substitution.replace_applicator
 0.92% :     0.000005s :      7: substitution.replace_old_param
 1.42% :     0.000008s :      1: substitution.reshape_eliminate
 0.46% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 1.91% :     0.000010s :      2: substitution.tile_eliminate
 3.92% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.87% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 8.19% :     0.000043s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.072833     2
95.45% :     0.069519s :      1: type_inference.infer
 4.55% :     0.003314s :      1: type_inference.specialize
------[replace.]   0.000154    19
64.06% :     0.000099s :     12: replace.inline
35.94% :     0.000055s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000352    19
93.80% :     0.000330s :     12: match.inline
 6.20% :     0.000022s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000375  2646
 1.62% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.25% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.44% :     0.000005s :     46: predicate.addn_check_dump
 1.72% :     0.000006s :     46: predicate.addn_zero_filter
 2.03% :     0.000008s :     46: predicate.arithmetic_simplify
 1.83% :     0.000007s :     46: predicate.cast_eliminate
 0.08% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.44% :     0.000005s :     46: predicate.compare_switch_simplify
 1.61% :     0.000006s :     46: predicate.depend_value_elim
 1.44% :     0.000005s :     46: predicate.dict_get_item_const_eliminator
 1.46% :     0.000005s :     46: predicate.dict_get_item_eliminator
 1.46% :     0.000005s :     46: predicate.dict_set_item_eliminator
 0.21% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.06% :     0.000000s :      1: predicate.elim_not_effective
 0.11% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000005s :     46: predicate.environ_add_const_eliminate
 1.41% :     0.000005s :     46: predicate.environ_get_add_eliminate
 1.45% :     0.000005s :     46: predicate.environ_get_depend_swap
 1.44% :     0.000005s :     46: predicate.environ_get_eliminate
 1.46% :     0.000005s :     46: predicate.environ_get_set_eliminate
 0.03% :     0.000000s :      1: predicate.fold_const_symbol
 0.91% :     0.000003s :     19: predicate.get_grad_eliminate
 0.04% :     0.000000s :      1: predicate.graph_param_transform
 4.14% :     0.000016s :     67: predicate.inline
 1.85% :     0.000007s :     43: predicate.inline_without_move
 0.40% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.98% :     0.000004s :     19: predicate.less_batch_normalization
 1.86% :     0.000007s :     53: predicate.list_to_tuple_eliminator_
 1.80% :     0.000007s :     54: predicate.load_eliminater
 0.31% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.37% :     0.000013s :     95: predicate.loop_unroll_before_grad
 1.73% :     0.000006s :     47: predicate.make_slice_get_slice_eliminator
 1.45% :     0.000005s :     46: predicate.merge_addn
 1.52% :     0.000006s :     46: predicate.minmaximum_grad
 0.41% :     0.000002s :      1: predicate.mutable_eliminate
 0.09% :     0.000000s :      1: predicate.opt_reshape
 2.34% :     0.000009s :     54: predicate.partial_eliminate
 1.93% :     0.000007s :     46: predicate.print_const_string_wrapper
 2.05% :     0.000008s :     46: predicate.reduce_eliminate
 1.81% :     0.000007s :     53: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.85% :     0.000011s :    100: predicate.replace_applicator
 0.93% :     0.000004s :     43: predicate.replace_old_param
 0.05% :     0.000000s :      1: predicate.reset_defer_inline
 1.76% :     0.000007s :     46: predicate.reshape_eliminate
 1.84% :     0.000007s :     46: predicate.row_tensor_add_zeros_like
 0.26% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.74% :     0.000007s :     46: predicate.same_eliminate
 0.46% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.33% :     0.000001s :      2: predicate.special_op_eliminate
 0.84% :     0.000003s :     19: predicate.specialize_transform
 1.74% :     0.000007s :     46: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.06% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.85% :     0.000011s :     65: predicate.switch_defer_inline
 2.69% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.47% :     0.000024s :    161: predicate.switch_simplify
 1.71% :     0.000006s :     46: predicate.tile_eliminate
 1.57% :     0.000006s :     46: predicate.transpose_eliminate
 2.01% :     0.000008s :     46: predicate.tuple_list_convert_item_index_to_positive
 1.67% :     0.000006s :     46: predicate.tuple_list_get_item_depend_reorder
 3.21% :     0.000012s :     55: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000007s :     46: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000007s :     53: predicate.tuple_to_list_eliminator_
 1.80% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000010s :     73: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000007s :     46: predicate.value_based_eliminate
 0.07% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.14% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.003688    48
74.39% :     0.002744s :     32: func_graph_cloner_run.FuncGraphClonerGraph
25.61% :     0.000945s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.111598    91
 0.04% :     0.000040s :      1: add_recomputation
 0.10% :     0.000112s :      1: auto_monad
 0.01% :     0.000016s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.21% :     0.000238s :      1: bootstrap
 0.02% :     0.000026s :      1: cconv
 0.01% :     0.000008s :      1: convert_after_rewriter
 0.02% :     0.000022s :      1: cse_after_recomputation
 0.01% :     0.000007s :      1: environ_conv
 0.27% :     0.000301s :      1: event_method
 0.00% :     0.000003s :      1: execute
 0.00% :     0.000005s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000010s :      1: graph_reusing
18.61% :     0.020769s :      1: jit_opt_a
 0.11% :     0.000123s :      1: jit_opt_after_cconv
 0.04% :     0.000045s :      1: jit_opt_b
 0.39% :     0.000431s :      1: loop_unroll
 0.62% :     0.000693s :      1: mutable_eliminate
 2.50% :     0.002788s :     39: opt.transform.jit_opt_a
 0.03% :     0.000031s :      4: opt.transform.jit_opt_after_cconv
 0.01% :     0.000017s :      4: opt.transform.jit_opt_b
 0.01% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.01% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.01% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.02% :     0.000024s :      4: opt.transform.symbol_engine_opt
 0.42% :     0.000464s :      1: opt_after_jit_grad
 0.01% :     0.000006s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.00% :     0.000005s :      1: pre_auto_parallel
 0.04% :     0.000045s :      1: py_interpret_to_execute
 0.01% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.01% :     0.000016s :      1: remove_dup_value
 9.00% :     0.010041s :      2: renormalize.infer
 1.78% :     0.001987s :      2: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.03% :     0.000037s :      1: rewriter_after_opt_a
 0.10% :     0.000110s :      1: rewriter_before_opt_a
 0.06% :     0.000069s :      1: symbol_engine_optimizer
 0.03% :     0.000031s :      1: task_emit
65.38% :     0.072961s :      1: type_inference
 0.04% :     0.000047s :      1: validate


TotalTime = 0.0129476, [30]
[bootstrap]: 0.00028981
[type_inference]: 0.00453528
[event_method]: 1.733e-05
[auto_monad]: 6.505e-05
[graph_reusing]: 5.74e-06
[pre_auto_parallel]: 2.63e-06
[py_interpret_to_execute]: 3.336e-05
[rewriter_before_opt_a]: 7.348e-05
[expand_dump_flag]: 3.63999e-06
[jit_opt_a]: 0.00573021, [2]
    [Cycle 1]: 0.0017345, [27]
        [switch_simplify]: 5.756e-05
        [loop_unroll]: 2.897e-05
        [a_1]: 0.00054259
        [with_stream_mark]: 1.706e-05
        [recompute_prepare]: 7.12002e-06
        [updatestate_depend_eliminate]: 3.7e-06
        [updatestate_assign_eliminate]: 3.36001e-06
        [updatestate_loads_eliminate]: 2.66999e-06
        [parameter_eliminate]: 1.92001e-06
        [specialize_transform]: 5.97999e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 5.61e-06
        [meta_shard_fg_expand]: 2.26998e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 3.36999e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.411e-05
        [j_node_and_user_rematch]: 9.05001e-06
        [meta_fg_expand]: 2.03002e-06
        [replace_old_param]: 9.51e-06
        [inline_without_move]: 5.94e-06
        [renormalize]: 0.00076942
        [add_forward_monad_depend]: 5.02999e-06
        [auto_monad_grad]: 2.56e-06
        [auto_monad_eliminator]: 1.427e-05
        [cse]: 2.01e-05
        [replace_applicator]: 1.243e-05
    [Cycle 2]: 0.00031797, [27]
        [switch_simplify]: 5.80002e-06
        [loop_unroll]: 5.00999e-06
        [a_1]: 8.839e-05
        [with_stream_mark]: 9.00001e-06
        [recompute_prepare]: 4.85001e-06
        [updatestate_depend_eliminate]: 2.41e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 5.13002e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.55999e-06
        [merge_forward]: 2.87002e-06
        [cell_reuse_recompute_pass]: 1.54e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.261e-05
        [j_node_and_user_rematch]: 7.98999e-06
        [meta_fg_expand]: 2.20002e-06
        [replace_old_param]: 1.336e-05
        [inline_without_move]: 4.75001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 5.02e-06
        [cse]: 9.83002e-06
        [replace_applicator]: 4.95001e-06
[py_interpret_to_execute_after_opt_a]: 1.103e-05
[rewriter_after_opt_a]: 3.327e-05
[convert_after_rewriter]: 6.80998e-06
[order_py_execute_after_rewriter]: 5.57999e-06
[mutable_eliminate]: 0.00056041
[jit_opt_b]: 8.154e-05, [1]
    [Cycle 1]: 7.46e-05, [2]
        [frontend_op_eliminate]: 1.724e-05
        [inline_after_opt_a]: 1.566e-05
[cconv]: 2.612e-05
[loop_unroll]: 0.00044944
[jit_opt_after_cconv]: 0.00013979, [1]
    [Cycle 1]: 0.00013272, [11]
        [c_1]: 2.012e-05
        [parameter_eliminate]: 2.88e-06
        [updatestate_depend_eliminate]: 5.52001e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.906e-05
        [call_graph_tuple_transform]: 1.859e-05
        [tuple_list_get_item_eliminator]: 5.42999e-06
        [none_parameter_eliminate]: 1.55001e-06
        [renormalize]: 4.60015e-07
        [switch_simplify]: 5.17e-06
[remove_dup_value]: 1.494e-05
[partial_unused_args_eliminate]: 2.44001e-06
[environ_conv]: 5.74999e-06
[add_recomputation]: 4.754e-05
[cse_after_recomputation]: 2.142e-05, [1]
    [Cycle 1]: 1.544e-05, [1]
        [cse]: 9.19e-06
[auto_monad_reorder]: 1.507e-05
[get_jit_bprop_graph]: 1.99999e-06
[rewriter_after_jit_bprop_graph]: 7.65e-06
[opt_after_jit_grad]: 0.00047795
[symbol_engine_optimizer]: 7.592e-05, [1]
    [Cycle 1]: 6.951e-05, [6]
        [build]: 3.65e-06
        [elim_shapecalc]: 8.22998e-06
        [elim_not_effective]: 1.351e-05
        [opt_reshape]: 6.07001e-06
        [fold_const_symbol]: 9.09998e-06
        [renormalize]: 6.40022e-07
[validate]: 3.796e-05
Sums
    bootstrap                            :     0.000290s :  3.35%
    type_inference                       :     0.004535s : 52.35%
    event_method                         :     0.000017s :  0.20%
    auto_monad                           :     0.000065s :  0.75%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.39%
    rewriter_before_opt_a                :     0.000073s :  0.85%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000063s :  0.73%
    jit_opt_a.loop_unroll                :     0.000034s :  0.39%
    jit_opt_a.a_1                        :     0.000631s :  7.28%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000023s :  0.26%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000769s :  8.88%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.22%
    jit_opt_a.cse                        :     0.000030s :  0.35%
    jit_opt_a.replace_applicator         :     0.000017s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.13%
    rewriter_after_opt_a                 :     0.000033s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000560s :  6.47%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.18%
    cconv                                :     0.000026s :  0.30%
    loop_unroll                          :     0.000449s :  5.19%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000048s :  0.55%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.09%
    opt_after_jit_grad                   :     0.000478s :  5.52%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000038s :  0.44%


Time group info:
------[substitution.]   0.000212    23
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.55% :     0.000005s :      3: substitution.graph_param_transform
84.44% :     0.000179s :      4: substitution.inline
 1.40% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.20% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.85% :     0.000004s :      2: substitution.replace_old_param
 6.06% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004472     2
80.26% :     0.003589s :      1: type_inference.infer
19.74% :     0.000883s :      1: type_inference.specialize
------[replace.]   0.000070     6
66.10% :     0.000046s :      4: replace.inline
33.90% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000188     6
93.77% :     0.000177s :      4: match.inline
 6.23% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.53% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.46% :     0.000002s :     10: predicate.addn_zero_filter
 2.01% :     0.000002s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000001s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.14% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.29% :     0.000006s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000001s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.89% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.76% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.60% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000002s :     15: predicate.partial_eliminate
 1.43% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.89% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.95% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000001s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.07% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.56% :     0.000009s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.47% :     0.000002s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.16% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000673    12
52.67% :     0.000354s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.19% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.14% :     0.000283s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014620    72
 0.35% :     0.000051s :      1: add_recomputation
 0.47% :     0.000068s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.08% :     0.000304s :      1: bootstrap
 0.20% :     0.000029s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000024s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.14% :     0.000021s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
39.22% :     0.005734s :      1: jit_opt_a
 0.97% :     0.000142s :      1: jit_opt_after_cconv
 0.58% :     0.000084s :      1: jit_opt_b
 3.13% :     0.000458s :      1: loop_unroll
 3.89% :     0.000569s :      1: mutable_eliminate
 5.80% :     0.000848s :     26: opt.transform.jit_opt_a
 0.32% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.33% :     0.000487s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000036s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.76% :     0.000404s :      1: renormalize.infer
 2.45% :     0.000358s :      1: renormalize.specialize
 0.07% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000036s :      1: rewriter_after_opt_a
 0.52% :     0.000076s :      1: rewriter_before_opt_a
 0.54% :     0.000078s :      1: symbol_engine_optimizer
31.10% :     0.004547s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.931.907 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c3680d0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:19.932.538 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0184684, [33]
[bootstrap]: 0.00026829
[type_inference]: 0.00458533
[event_method]: 1.907e-05
[auto_monad]: 5.607e-05
[graph_reusing]: 4.56002e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 4.012e-05
[rewriter_before_opt_a]: 7.006e-05
[expand_dump_flag]: 2.94999e-06
[jit_opt_a]: 0.0052934, [2]
    [Cycle 1]: 0.00181952, [27]
        [switch_simplify]: 4.773e-05
        [loop_unroll]: 2.893e-05
        [a_1]: 0.00055151
        [with_stream_mark]: 1.575e-05
        [recompute_prepare]: 7.97e-06
        [updatestate_depend_eliminate]: 2.88e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 1.39998e-06
        [specialize_transform]: 6.12999e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 5.43002e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 5.60001e-06
        [merge_forward]: 2.72001e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.134e-05
        [j_node_and_user_rematch]: 8.75999e-06
        [meta_fg_expand]: 1.97001e-06
        [replace_old_param]: 7.58001e-06
        [inline_without_move]: 5.10999e-06
        [renormalize]: 0.00084557
        [add_forward_monad_depend]: 5.97001e-06
        [auto_monad_grad]: 2.27999e-06
        [auto_monad_eliminator]: 1.42e-05
        [cse]: 2.158e-05
        [replace_applicator]: 1.533e-05
    [Cycle 2]: 0.00032232, [27]
        [switch_simplify]: 5.86003e-06
        [loop_unroll]: 4.85999e-06
        [a_1]: 9.31e-05
        [with_stream_mark]: 8.94e-06
        [recompute_prepare]: 5.07e-06
        [updatestate_depend_eliminate]: 2.56998e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.35001e-06
        [specialize_transform]: 4.90999e-06
        [updatestate_useless_node_eliminater]: 4.84998e-06
        [accelerated_algorithm]: 4.77998e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.36002e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 2.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.191e-05
        [j_node_and_user_rematch]: 7.50003e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 6.90998e-06
        [inline_without_move]: 4.58001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 5.61e-06
        [cse]: 1.071e-05
        [replace_applicator]: 4.94003e-06
[py_interpret_to_execute_after_opt_a]: 1.06e-05
[rewriter_after_opt_a]: 3.239e-05
[convert_after_rewriter]: 9.37001e-06
[order_py_execute_after_rewriter]: 4.48001e-06
[mutable_eliminate]: 0.00060915
[jit_opt_b]: 5.067e-05, [1]
    [Cycle 1]: 4.381e-05, [2]
        [frontend_op_eliminate]: 1.689e-05
        [inline_after_opt_a]: 1.439e-05
[cconv]: 2.307e-05
[loop_unroll]: 0.00043378
[jit_opt_after_cconv]: 0.00014472, [1]
    [Cycle 1]: 0.00013755, [11]
        [c_1]: 2.037e-05
        [parameter_eliminate]: 2.81e-06
        [updatestate_depend_eliminate]: 5.61e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.32001e-06
        [cse]: 2.118e-05
        [call_graph_tuple_transform]: 2.017e-05
        [tuple_list_get_item_eliminator]: 5.76e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 1.04e-06
        [switch_simplify]: 5.33002e-06
[remove_dup_value]: 1.217e-05
[partial_unused_args_eliminate]: 1.74e-06
[environ_conv]: 4.15e-06
[add_recomputation]: 4.102e-05
[cse_after_recomputation]: 2.12e-05, [1]
    [Cycle 1]: 1.502e-05, [1]
        [cse]: 8.86002e-06
[auto_monad_reorder]: 1.215e-05
[get_jit_bprop_graph]: 1.88002e-06
[rewriter_after_jit_bprop_graph]: 4.84998e-06
[opt_after_jit_grad]: 0.00046881
[symbol_engine_optimizer]: 7.479e-05, [1]
    [Cycle 1]: 6.796e-05, [6]
        [build]: 3.6e-06
        [elim_shapecalc]: 7.06999e-06
        [elim_not_effective]: 1.354e-05
        [opt_reshape]: 5.74e-06
        [fold_const_symbol]: 8.42998e-06
        [renormalize]: 5.19998e-07
[validate]: 3.569e-05
[backend_pass]: 9.60019e-07
[task_emit]: 0.00592733
[execute]: 5.51e-06
Sums
    bootstrap                            :     0.000268s :  1.83%
    type_inference                       :     0.004585s : 31.22%
    event_method                         :     0.000019s :  0.13%
    auto_monad                           :     0.000056s :  0.38%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.27%
    rewriter_before_opt_a                :     0.000070s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.36%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000645s :  4.39%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000014s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000846s :  5.76%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000032s :  0.22%
    jit_opt_a.replace_applicator         :     0.000020s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.22%
    convert_after_rewriter               :     0.000009s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000609s :  4.15%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000023s :  0.16%
    loop_unroll                          :     0.000434s :  2.95%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000041s :  0.28%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000469s :  3.19%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005927s : 40.35%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000183    23
 1.19% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.77% :     0.000005s :      3: substitution.graph_param_transform
84.48% :     0.000154s :      4: substitution.inline
 1.36% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.27% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.80% :     0.000003s :      2: substitution.replace_old_param
 5.45% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004433     2
79.51% :     0.003524s :      1: type_inference.infer
20.49% :     0.000908s :      1: type_inference.specialize
------[replace.]   0.000113     6
77.49% :     0.000087s :      4: replace.inline
22.51% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000161     6
94.47% :     0.000152s :      4: match.inline
 5.53% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.29% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 1.86% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000001s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000001s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.29% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 4.92% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.54% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.78% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.91% :     0.000002s :     15: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.19% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 2.09% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000002s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.60% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.86% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.46% :     0.000002s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.45% :     0.000009s :     53: predicate.switch_simplify
 1.32% :     0.000002s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.85% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.91% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.93% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000622    12
45.55% :     0.000283s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.27% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.18% :     0.000306s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020259    76
 0.22% :     0.000044s :      1: add_recomputation
 0.29% :     0.000059s :      1: auto_monad
 0.07% :     0.000015s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.39% :     0.000281s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.06% :     0.000012s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.11% :     0.000023s :      1: event_method
 0.04% :     0.000009s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
26.15% :     0.005297s :      1: jit_opt_a
 0.73% :     0.000148s :      1: jit_opt_after_cconv
 0.26% :     0.000053s :      1: jit_opt_b
 2.18% :     0.000442s :      1: loop_unroll
 3.05% :     0.000618s :      1: mutable_eliminate
 4.17% :     0.000844s :     26: opt.transform.jit_opt_a
 0.24% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.36% :     0.000477s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000043s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000014s :      1: remove_dup_value
 2.18% :     0.000441s :      1: renormalize.infer
 1.96% :     0.000396s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000035s :      1: rewriter_after_opt_a
 0.36% :     0.000073s :      1: rewriter_before_opt_a
 0.38% :     0.000078s :      1: symbol_engine_optimizer
29.33% :     0.005941s :      1: task_emit
22.69% :     0.004597s :      1: type_inference
 0.27% :     0.000054s :      1: validate


TotalTime = 0.0185236, [33]
[bootstrap]: 0.00026713
[type_inference]: 0.00423651
[event_method]: 1.606e-05
[auto_monad]: 6.382e-05
[graph_reusing]: 5.15001e-06
[pre_auto_parallel]: 2.31e-06
[py_interpret_to_execute]: 3.096e-05
[rewriter_before_opt_a]: 6.768e-05
[expand_dump_flag]: 3.25e-06
[jit_opt_a]: 0.00534072, [2]
    [Cycle 1]: 0.00173016, [27]
        [switch_simplify]: 5.669e-05
        [loop_unroll]: 3.027e-05
        [a_1]: 0.00055482
        [with_stream_mark]: 1.984e-05
        [recompute_prepare]: 1.019e-05
        [updatestate_depend_eliminate]: 3.43999e-06
        [updatestate_assign_eliminate]: 3.14999e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 1.81998e-06
        [specialize_transform]: 5.86003e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 5.40001e-06
        [meta_shard_fg_expand]: 2.36e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 3.31999e-06
        [cell_reuse_recompute_pass]: 1.32999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.483e-05
        [j_node_and_user_rematch]: 9.63002e-06
        [meta_fg_expand]: 2.22999e-06
        [replace_old_param]: 9.71998e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 0.00065298
        [add_forward_monad_depend]: 7.06999e-06
        [auto_monad_grad]: 2.52001e-06
        [auto_monad_eliminator]: 9.237e-05
        [cse]: 2.469e-05
        [replace_applicator]: 1.601e-05
    [Cycle 2]: 0.00035651, [27]
        [switch_simplify]: 5.65001e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 0.00011106
        [with_stream_mark]: 1.284e-05
        [recompute_prepare]: 6.11e-06
        [updatestate_depend_eliminate]: 2.82002e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.18002e-06
        [parameter_eliminate]: 1.69998e-06
        [specialize_transform]: 4.94003e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 5.17e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 4.3e-06
        [merge_forward]: 2.66e-06
        [cell_reuse_recompute_pass]: 1.99e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.411e-05
        [j_node_and_user_rematch]: 8.15e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 7.5e-06
        [inline_without_move]: 4.86002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.78002e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 5.56998e-06
        [cse]: 1.131e-05
        [replace_applicator]: 6.17001e-06
[py_interpret_to_execute_after_opt_a]: 1.253e-05
[rewriter_after_opt_a]: 3.35e-05
[convert_after_rewriter]: 6.83e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.00061779
[jit_opt_b]: 5.045e-05, [1]
    [Cycle 1]: 4.25e-05, [2]
        [frontend_op_eliminate]: 1.684e-05
        [inline_after_opt_a]: 1.437e-05
[cconv]: 2.477e-05
[loop_unroll]: 0.00043334
[jit_opt_after_cconv]: 0.00014389, [1]
    [Cycle 1]: 0.00013702, [11]
        [c_1]: 2.132e-05
        [parameter_eliminate]: 3.33e-06
        [updatestate_depend_eliminate]: 5.68997e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.46998e-06
        [cse]: 1.926e-05
        [call_graph_tuple_transform]: 1.916e-05
        [tuple_list_get_item_eliminator]: 5.51e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 5.79981e-07
        [switch_simplify]: 5.25999e-06
[remove_dup_value]: 1.468e-05
[partial_unused_args_eliminate]: 2.19999e-06
[environ_conv]: 5.64e-06
[add_recomputation]: 4.658e-05
[cse_after_recomputation]: 1.98e-05, [1]
    [Cycle 1]: 1.459e-05, [1]
        [cse]: 8.99e-06
[auto_monad_reorder]: 1.453e-05
[get_jit_bprop_graph]: 1.94e-06
[rewriter_after_jit_bprop_graph]: 5.82999e-06
[opt_after_jit_grad]: 0.00047666
[symbol_engine_optimizer]: 7.258e-05, [1]
    [Cycle 1]: 6.625e-05, [6]
        [build]: 3.61999e-06
        [elim_shapecalc]: 7.58001e-06
        [elim_not_effective]: 1.29e-05
        [opt_reshape]: 5.68997e-06
        [fold_const_symbol]: 8.25e-06
        [renormalize]: 9.80013e-07
[validate]: 3.559e-05
[backend_pass]: 1.06997e-06
[task_emit]: 0.00626762
[execute]: 7.82998e-06
Sums
    bootstrap                            :     0.000267s :  1.82%
    type_inference                       :     0.004237s : 28.90%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000064s :  0.44%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.21%
    rewriter_before_opt_a                :     0.000068s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.43%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000666s :  4.54%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000653s :  4.45%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000098s :  0.67%
    jit_opt_a.cse                        :     0.000036s :  0.25%
    jit_opt_a.replace_applicator         :     0.000022s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.09%
    rewriter_after_opt_a                 :     0.000033s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000618s :  4.21%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000433s :  2.96%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000015s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000047s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000477s :  3.25%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000036s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006268s : 42.76%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000207    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.64% :     0.000005s :      3: substitution.graph_param_transform
84.23% :     0.000174s :      4: substitution.inline
 1.42% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.30% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.22% :     0.000005s :      2: substitution.replace_old_param
 5.73% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004178     2
81.41% :     0.003401s :      1: type_inference.infer
18.59% :     0.000777s :      1: type_inference.specialize
------[replace.]   0.000072     6
65.34% :     0.000047s :      4: replace.inline
34.66% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000183     6
94.09% :     0.000172s :      4: match.inline
 5.91% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.17% :     0.000001s :     10: predicate.depend_value_elim
 1.23% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.69% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.49% :     0.000001s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.25% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.74% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.17% :     0.000003s :     15: predicate.load_eliminater
 1.28% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.00% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.48% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.12% :     0.000003s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.32% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.97% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000002s :     10: predicate.same_eliminate
 0.80% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.69% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.61% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.68% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.02% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.81% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.67% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000528    12
47.22% :     0.000249s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.13% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.65% :     0.000246s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020167    76
 0.25% :     0.000050s :      1: add_recomputation
 0.33% :     0.000067s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.38% :     0.000278s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.50% :     0.005344s :      1: jit_opt_a
 0.73% :     0.000147s :      1: jit_opt_after_cconv
 0.26% :     0.000053s :      1: jit_opt_b
 2.19% :     0.000442s :      1: loop_unroll
 3.11% :     0.000627s :      1: mutable_eliminate
 4.40% :     0.000886s :     26: opt.transform.jit_opt_a
 0.24% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.41% :     0.000485s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000034s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.65% :     0.000333s :      1: renormalize.infer
 1.55% :     0.000313s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000036s :      1: rewriter_after_opt_a
 0.35% :     0.000070s :      1: rewriter_before_opt_a
 0.37% :     0.000075s :      1: symbol_engine_optimizer
31.15% :     0.006282s :      1: task_emit
21.06% :     0.004247s :      1: type_inference
 0.28% :     0.000056s :      1: validate


TotalTime = 0.0134188, [30]
[bootstrap]: 0.00029381
[type_inference]: 0.00453596
[event_method]: 1.726e-05
[auto_monad]: 6.328e-05
[graph_reusing]: 5.42001e-06
[pre_auto_parallel]: 2.79001e-06
[py_interpret_to_execute]: 3.283e-05
[rewriter_before_opt_a]: 6.738e-05
[expand_dump_flag]: 2.98e-06
[jit_opt_a]: 0.00604114, [2]
    [Cycle 1]: 0.00189384, [27]
        [switch_simplify]: 5.807e-05
        [loop_unroll]: 2.897e-05
        [a_1]: 0.00056127
        [with_stream_mark]: 1.885e-05
        [recompute_prepare]: 8.51002e-06
        [updatestate_depend_eliminate]: 3.26999e-06
        [updatestate_assign_eliminate]: 3.14001e-06
        [updatestate_loads_eliminate]: 2.91999e-06
        [parameter_eliminate]: 1.91e-06
        [specialize_transform]: 6.11e-06
        [updatestate_useless_node_eliminater]: 5.72001e-06
        [accelerated_algorithm]: 5.74999e-06
        [meta_shard_fg_expand]: 2.42001e-06
        [get_grad_eliminate_]: 5.58002e-06
        [merge_forward]: 3.4e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.495e-05
        [j_node_and_user_rematch]: 9.27999e-06
        [meta_fg_expand]: 2.26998e-06
        [replace_old_param]: 1.024e-05
        [inline_without_move]: 6.36e-06
        [renormalize]: 0.00082629
        [add_forward_monad_depend]: 5.27999e-06
        [auto_monad_grad]: 2.72001e-06
        [auto_monad_eliminator]: 1.41e-05
        [cse]: 2.859e-05
        [replace_applicator]: 7.541e-05
    [Cycle 2]: 0.00034567, [27]
        [switch_simplify]: 8.49998e-06
        [loop_unroll]: 5.35999e-06
        [a_1]: 9.419e-05
        [with_stream_mark]: 1.087e-05
        [recompute_prepare]: 5.40999e-06
        [updatestate_depend_eliminate]: 2.73e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 1.76003e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 5.25001e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.18002e-06
        [meta_shard_fg_expand]: 1.64e-06
        [get_grad_eliminate_]: 4.72e-06
        [merge_forward]: 2.87002e-06
        [cell_reuse_recompute_pass]: 2.21e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.412e-05
        [j_node_and_user_rematch]: 7.98999e-06
        [meta_fg_expand]: 1.72999e-06
        [replace_old_param]: 6.98998e-06
        [inline_without_move]: 4.93001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.45999e-06
        [auto_monad_grad]: 1.07998e-06
        [auto_monad_eliminator]: 6.61e-06
        [cse]: 1.347e-05
        [replace_applicator]: 5.17e-06
[py_interpret_to_execute_after_opt_a]: 1.281e-05
[rewriter_after_opt_a]: 3.629e-05
[convert_after_rewriter]: 1.128e-05
[order_py_execute_after_rewriter]: 5.32001e-06
[mutable_eliminate]: 0.00064056
[jit_opt_b]: 5.105e-05, [1]
    [Cycle 1]: 4.344e-05, [2]
        [frontend_op_eliminate]: 1.689e-05
        [inline_after_opt_a]: 1.504e-05
[cconv]: 2.644e-05
[loop_unroll]: 0.00045152
[jit_opt_after_cconv]: 0.00014722, [1]
    [Cycle 1]: 0.00014013, [11]
        [c_1]: 2.155e-05
        [parameter_eliminate]: 3.20002e-06
        [updatestate_depend_eliminate]: 5.47999e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 2.092e-05
        [call_graph_tuple_transform]: 2.026e-05
        [tuple_list_get_item_eliminator]: 5.37999e-06
        [none_parameter_eliminate]: 1.79e-06
        [renormalize]: 4.20026e-07
        [switch_simplify]: 5.44998e-06
[remove_dup_value]: 1.53e-05
[partial_unused_args_eliminate]: 2.46e-06
[environ_conv]: 5.56e-06
[add_recomputation]: 4.843e-05
[cse_after_recomputation]: 2.177e-05, [1]
    [Cycle 1]: 1.552e-05, [1]
        [cse]: 9.01002e-06
[auto_monad_reorder]: 1.602e-05
[get_jit_bprop_graph]: 1.92999e-06
[rewriter_after_jit_bprop_graph]: 6.72002e-06
[opt_after_jit_grad]: 0.00055767
[symbol_engine_optimizer]: 7.965e-05, [1]
    [Cycle 1]: 7.128e-05, [6]
        [build]: 4.07998e-06
        [elim_shapecalc]: 7.51999e-06
        [elim_not_effective]: 1.437e-05
        [opt_reshape]: 6.63998e-06
        [fold_const_symbol]: 8.98002e-06
        [renormalize]: 1.05999e-06
[validate]: 4.355e-05
Sums
    bootstrap                            :     0.000294s :  3.26%
    type_inference                       :     0.004536s : 50.34%
    event_method                         :     0.000017s :  0.19%
    auto_monad                           :     0.000063s :  0.70%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.36%
    rewriter_before_opt_a                :     0.000067s :  0.75%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000067s :  0.74%
    jit_opt_a.loop_unroll                :     0.000034s :  0.38%
    jit_opt_a.a_1                        :     0.000655s :  7.27%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000826s :  9.17%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.23%
    jit_opt_a.cse                        :     0.000042s :  0.47%
    jit_opt_a.replace_applicator         :     0.000081s :  0.89%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.14%
    rewriter_after_opt_a                 :     0.000036s :  0.40%
    convert_after_rewriter               :     0.000011s :  0.13%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000641s :  7.11%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000026s :  0.29%
    loop_unroll                          :     0.000452s :  5.01%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000048s :  0.54%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000558s :  6.19%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.48%


Time group info:
------[substitution.]   0.000214    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.70% :     0.000006s :      3: substitution.graph_param_transform
84.06% :     0.000180s :      4: substitution.inline
 1.50% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.29% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000004s :      2: substitution.replace_old_param
 5.84% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004473     2
80.22% :     0.003588s :      1: type_inference.infer
19.78% :     0.000885s :      1: type_inference.specialize
------[replace.]   0.000078     6
67.95% :     0.000053s :      4: replace.inline
32.05% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000189     6
93.99% :     0.000178s :      4: match.inline
 6.01% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.29% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.92% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.47% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.60% :     0.000002s :     10: predicate.depend_value_elim
 1.23% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.51% :     0.000001s :      3: predicate.graph_param_transform
 5.85% :     0.000007s :     22: predicate.inline
 1.03% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.34% :     0.000002s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.57% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.76% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.13% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.68% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.62% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.29% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.26% :     0.000002s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.71% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.60% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.29% :     0.000002s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.64% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000626    12
48.31% :     0.000302s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.23% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.46% :     0.000291s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015233    72
 0.34% :     0.000052s :      1: add_recomputation
 0.44% :     0.000066s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 2.02% :     0.000307s :      1: bootstrap
 0.19% :     0.000029s :      1: cconv
 0.09% :     0.000013s :      1: convert_after_rewriter
 0.16% :     0.000024s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.14% :     0.000021s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.68% :     0.006045s :      1: jit_opt_a
 0.98% :     0.000150s :      1: jit_opt_after_cconv
 0.35% :     0.000054s :      1: jit_opt_b
 3.02% :     0.000460s :      1: loop_unroll
 4.26% :     0.000649s :      1: mutable_eliminate
 6.17% :     0.000939s :     26: opt.transform.jit_opt_a
 0.32% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.18% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.72% :     0.000567s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000036s :      1: py_interpret_to_execute
 0.10% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000018s :      1: remove_dup_value
 2.89% :     0.000440s :      1: renormalize.infer
 2.48% :     0.000378s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000039s :      1: rewriter_after_opt_a
 0.46% :     0.000070s :      1: rewriter_before_opt_a
 0.54% :     0.000082s :      1: symbol_engine_optimizer
29.85% :     0.004547s :      1: type_inference


TotalTime = 0.0199947, [33]
[bootstrap]: 0.00033288
[type_inference]: 0.00488852
[event_method]: 1.702e-05
[auto_monad]: 6.124e-05
[graph_reusing]: 5.15999e-06
[pre_auto_parallel]: 2.86999e-06
[py_interpret_to_execute]: 4.006e-05
[rewriter_before_opt_a]: 7.11e-05
[expand_dump_flag]: 2.74001e-06
[jit_opt_a]: 0.00590817, [2]
    [Cycle 1]: 0.00190452, [27]
        [switch_simplify]: 5.558e-05
        [loop_unroll]: 2.924e-05
        [a_1]: 0.00055523
        [with_stream_mark]: 1.44e-05
        [recompute_prepare]: 7.66001e-06
        [updatestate_depend_eliminate]: 2.76999e-06
        [updatestate_assign_eliminate]: 2.33002e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.27999e-06
        [specialize_transform]: 6.00002e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 5.72999e-06
        [meta_shard_fg_expand]: 1.72001e-06
        [get_grad_eliminate_]: 5.32999e-06
        [merge_forward]: 2.93998e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.155e-05
        [j_node_and_user_rematch]: 9.86e-06
        [meta_fg_expand]: 1.88002e-06
        [replace_old_param]: 9.29e-06
        [inline_without_move]: 5.30001e-06
        [renormalize]: 0.00093819
        [add_forward_monad_depend]: 6.04999e-06
        [auto_monad_grad]: 1.84998e-06
        [auto_monad_eliminator]: 1.385e-05
        [cse]: 2.066e-05
        [replace_applicator]: 1.441e-05
    [Cycle 2]: 0.00037019, [27]
        [switch_simplify]: 6.07001e-06
        [loop_unroll]: 4.98001e-06
        [a_1]: 0.00010198
        [with_stream_mark]: 1.327e-05
        [recompute_prepare]: 4.99e-06
        [updatestate_depend_eliminate]: 2.79001e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.47001e-06
        [specialize_transform]: 4.89998e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 5.85002e-06
        [meta_shard_fg_expand]: 2.04999e-06
        [get_grad_eliminate_]: 5.12999e-06
        [merge_forward]: 3.00002e-06
        [cell_reuse_recompute_pass]: 2.88e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.406e-05
        [j_node_and_user_rematch]: 8.94e-06
        [meta_fg_expand]: 2.01e-06
        [replace_old_param]: 2.105e-05
        [inline_without_move]: 5.09e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.77999e-06
        [auto_monad_grad]: 1.20999e-06
        [auto_monad_eliminator]: 5.82999e-06
        [cse]: 1.205e-05
        [replace_applicator]: 5.62999e-06
[py_interpret_to_execute_after_opt_a]: 1.528e-05
[rewriter_after_opt_a]: 3.461e-05
[convert_after_rewriter]: 6.40002e-06
[order_py_execute_after_rewriter]: 5.31002e-06
[mutable_eliminate]: 0.00065181
[jit_opt_b]: 5.533e-05, [1]
    [Cycle 1]: 4.747e-05, [2]
        [frontend_op_eliminate]: 1.873e-05
        [inline_after_opt_a]: 1.673e-05
[cconv]: 2.886e-05
[loop_unroll]: 0.00058404
[jit_opt_after_cconv]: 0.00015801, [1]
    [Cycle 1]: 0.00014925, [11]
        [c_1]: 2.166e-05
        [parameter_eliminate]: 3.92998e-06
        [updatestate_depend_eliminate]: 6.01998e-06
        [updatestate_assign_eliminate]: 2.59999e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 2.291e-05
        [call_graph_tuple_transform]: 2.12e-05
        [tuple_list_get_item_eliminator]: 6.17999e-06
        [none_parameter_eliminate]: 1.37999e-06
        [renormalize]: 7.89994e-07
        [switch_simplify]: 5.56e-06
[remove_dup_value]: 1.287e-05
[partial_unused_args_eliminate]: 2.61e-06
[environ_conv]: 5.74e-06
[add_recomputation]: 4.665e-05
[cse_after_recomputation]: 2.39e-05, [1]
    [Cycle 1]: 1.823e-05, [1]
        [cse]: 9.74e-06
[auto_monad_reorder]: 1.541e-05
[get_jit_bprop_graph]: 1.99e-06
[rewriter_after_jit_bprop_graph]: 5.05999e-06
[opt_after_jit_grad]: 0.00052079
[symbol_engine_optimizer]: 7.889e-05, [1]
    [Cycle 1]: 7.108e-05, [6]
        [build]: 3.45e-06
        [elim_shapecalc]: 8.59e-06
        [elim_not_effective]: 1.423e-05
        [opt_reshape]: 6.34999e-06
        [fold_const_symbol]: 8.54e-06
        [renormalize]: 1.10001e-06
[validate]: 3.82e-05
[backend_pass]: 7.00005e-07
[task_emit]: 0.00614618
[execute]: 7.35003e-06
Sums
    bootstrap                            :     0.000333s :  2.12%
    type_inference                       :     0.004889s : 31.12%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000061s :  0.39%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.25%
    rewriter_before_opt_a                :     0.000071s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000657s :  4.18%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000030s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000938s :  5.97%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000033s :  0.21%
    jit_opt_a.replace_applicator         :     0.000020s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.10%
    rewriter_after_opt_a                 :     0.000035s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000652s :  4.15%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.11%
    cconv                                :     0.000029s :  0.18%
    loop_unroll                          :     0.000584s :  3.72%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000047s :  0.30%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000521s :  3.31%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000038s :  0.24%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006146s : 39.12%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000217    23
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 2.51% :     0.000005s :      3: substitution.graph_param_transform
84.66% :     0.000184s :      4: substitution.inline
 1.96% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.89% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000004s :      2: substitution.replace_old_param
 5.70% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004820     2
79.58% :     0.003835s :      1: type_inference.infer
20.42% :     0.000984s :      1: type_inference.specialize
------[replace.]   0.000078     6
66.48% :     0.000052s :      4: replace.inline
33.52% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     6
94.14% :     0.000181s :      4: match.inline
 5.86% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 2.00% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.83% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 2.09% :     0.000003s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.98% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.04% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.59% :     0.000001s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.32% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.24% :     0.000002s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 4.99% :     0.000006s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :      6: predicate.less_batch_normalization
 1.62% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.64% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.09% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.18% :     0.000001s :     10: predicate.minmaximum_grad
 1.69% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 1.86% :     0.000002s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.88% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.55% :     0.000001s :      3: predicate.reset_defer_inline
 1.78% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.66% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.43% :     0.000009s :     53: predicate.switch_simplify
 1.54% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.77% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.40% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000697    12
49.13% :     0.000342s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.64% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.23% :     0.000322s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021936    76
 0.24% :     0.000052s :      1: add_recomputation
 0.30% :     0.000065s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.59% :     0.000348s :      1: bootstrap
 0.14% :     0.000031s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000026s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.95% :     0.005912s :      1: jit_opt_a
 0.73% :     0.000161s :      1: jit_opt_after_cconv
 0.26% :     0.000058s :      1: jit_opt_b
 2.71% :     0.000594s :      1: loop_unroll
 3.01% :     0.000661s :      1: mutable_eliminate
 4.04% :     0.000887s :     26: opt.transform.jit_opt_a
 0.23% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000028s :      4: opt.transform.jit_opt_b
 0.07% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.42% :     0.000531s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000043s :      1: py_interpret_to_execute
 0.08% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.28% :     0.000501s :      1: renormalize.infer
 1.96% :     0.000429s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000038s :      1: rewriter_after_opt_a
 0.34% :     0.000074s :      1: rewriter_before_opt_a
 0.37% :     0.000082s :      1: symbol_engine_optimizer
28.09% :     0.006162s :      1: task_emit
22.35% :     0.004903s :      1: type_inference
 0.27% :     0.000059s :      1: validate


TotalTime = 0.0195202, [33]
[bootstrap]: 0.00028944
[type_inference]: 0.00437174
[event_method]: 1.702e-05
[auto_monad]: 6.205e-05
[graph_reusing]: 5.62001e-06
[pre_auto_parallel]: 2.79999e-06
[py_interpret_to_execute]: 3.607e-05
[rewriter_before_opt_a]: 7.185e-05
[expand_dump_flag]: 3.25002e-06
[jit_opt_a]: 0.00555446, [2]
    [Cycle 1]: 0.00162327, [27]
        [switch_simplify]: 4.893e-05
        [loop_unroll]: 2.992e-05
        [a_1]: 0.00053856
        [with_stream_mark]: 1.717e-05
        [recompute_prepare]: 7.15e-06
        [updatestate_depend_eliminate]: 3.07002e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [parameter_eliminate]: 1.55999e-06
        [specialize_transform]: 6.10002e-06
        [updatestate_useless_node_eliminater]: 5.42001e-06
        [accelerated_algorithm]: 5.65001e-06
        [meta_shard_fg_expand]: 2.06998e-06
        [get_grad_eliminate_]: 4.97e-06
        [merge_forward]: 3.04999e-06
        [cell_reuse_recompute_pass]: 1.10999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.269e-05
        [j_node_and_user_rematch]: 8.43999e-06
        [meta_fg_expand]: 2.63e-06
        [replace_old_param]: 7.79002e-06
        [inline_without_move]: 5.12e-06
        [renormalize]: 0.00068295
        [add_forward_monad_depend]: 4.91002e-06
        [auto_monad_grad]: 2.77002e-06
        [auto_monad_eliminator]: 1.268e-05
        [cse]: 2.339e-05
        [replace_applicator]: 1.177e-05
    [Cycle 2]: 0.00032938, [27]
        [switch_simplify]: 5.79e-06
        [loop_unroll]: 5.06002e-06
        [a_1]: 9.001e-05
        [with_stream_mark]: 1.024e-05
        [recompute_prepare]: 5.15001e-06
        [updatestate_depend_eliminate]: 3.09001e-06
        [updatestate_assign_eliminate]: 2.23002e-06
        [updatestate_loads_eliminate]: 2.13998e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 5.35999e-06
        [updatestate_useless_node_eliminater]: 5.26998e-06
        [accelerated_algorithm]: 5.27999e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 4.99e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 1.74998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.277e-05
        [j_node_and_user_rematch]: 8.10999e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 7.2e-06
        [inline_without_move]: 4.70999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.10002e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 5.46998e-06
        [cse]: 1.041e-05
        [replace_applicator]: 5.61e-06
[py_interpret_to_execute_after_opt_a]: 1.113e-05
[rewriter_after_opt_a]: 3.383e-05
[convert_after_rewriter]: 6.69999e-06
[order_py_execute_after_rewriter]: 5.07e-06
[mutable_eliminate]: 0.00074228
[jit_opt_b]: 5.441e-05, [1]
    [Cycle 1]: 4.501e-05, [2]
        [frontend_op_eliminate]: 1.682e-05
        [inline_after_opt_a]: 1.597e-05
[cconv]: 3.285e-05
[loop_unroll]: 0.00053224
[jit_opt_after_cconv]: 0.00015406, [1]
    [Cycle 1]: 0.00014666, [11]
        [c_1]: 2.195e-05
        [parameter_eliminate]: 3.55998e-06
        [updatestate_depend_eliminate]: 6.52001e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.60002e-06
        [cse]: 2.354e-05
        [call_graph_tuple_transform]: 2.109e-05
        [tuple_list_get_item_eliminator]: 5.56998e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 7.80012e-07
        [switch_simplify]: 6.23002e-06
[remove_dup_value]: 1.431e-05
[partial_unused_args_eliminate]: 2.07999e-06
[environ_conv]: 5.33002e-06
[add_recomputation]: 4.919e-05
[cse_after_recomputation]: 2.127e-05, [1]
    [Cycle 1]: 1.541e-05, [1]
        [cse]: 9.41003e-06
[auto_monad_reorder]: 1.433e-05
[get_jit_bprop_graph]: 1.92999e-06
[rewriter_after_jit_bprop_graph]: 8.03001e-06
[opt_after_jit_grad]: 0.00052813
[symbol_engine_optimizer]: 7.626e-05, [1]
    [Cycle 1]: 6.933e-05, [6]
        [build]: 3.6e-06
        [elim_shapecalc]: 8.1e-06
        [elim_not_effective]: 1.335e-05
        [opt_reshape]: 5.79999e-06
        [fold_const_symbol]: 8.76002e-06
        [renormalize]: 1.37e-06
[validate]: 3.607e-05
[backend_pass]: 1.06997e-06
[task_emit]: 0.00655899
[execute]: 7.71001e-06
Sums
    bootstrap                            :     0.000289s :  1.89%
    type_inference                       :     0.004372s : 28.58%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000062s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.24%
    rewriter_before_opt_a                :     0.000072s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.36%
    jit_opt_a.loop_unroll                :     0.000035s :  0.23%
    jit_opt_a.a_1                        :     0.000629s :  4.11%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000683s :  4.47%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000034s :  0.22%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000742s :  4.85%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000033s :  0.21%
    loop_unroll                          :     0.000532s :  3.48%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000049s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000528s :  3.45%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000036s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006559s : 42.88%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000204    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.96% :     0.000006s :      3: substitution.graph_param_transform
84.30% :     0.000172s :      4: substitution.inline
 1.33% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.17% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.72% :     0.000004s :      2: substitution.replace_old_param
 5.97% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004312     2
82.19% :     0.003544s :      1: type_inference.infer
17.81% :     0.000768s :      1: type_inference.specialize
------[replace.]   0.000069     6
66.71% :     0.000046s :      4: replace.inline
33.29% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000180     6
93.97% :     0.000169s :      4: match.inline
 6.03% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.21% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.47% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000001s :     10: predicate.addn_zero_filter
 2.30% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.24% :     0.000001s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.54% :     0.000001s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.73% :     0.000007s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 2.12% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.17% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.28% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 2.27% :     0.000003s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000002s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.27% :     0.000001s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.17% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.61% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.67% :     0.000009s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.30% :     0.000002s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.39% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.40% :     0.000002s :     10: predicate.value_based_eliminate
 0.47% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000554    12
47.59% :     0.000264s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.38% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.03% :     0.000249s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021153    76
 0.25% :     0.000053s :      1: add_recomputation
 0.31% :     0.000066s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.43% :     0.000302s :      1: bootstrap
 0.17% :     0.000035s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.28% :     0.005558s :      1: jit_opt_a
 0.74% :     0.000157s :      1: jit_opt_after_cconv
 0.27% :     0.000057s :      1: jit_opt_b
 2.57% :     0.000543s :      1: loop_unroll
 3.58% :     0.000757s :      1: mutable_eliminate
 3.93% :     0.000831s :     26: opt.transform.jit_opt_a
 0.24% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.55% :     0.000539s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000039s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.74% :     0.000367s :      1: renormalize.infer
 1.45% :     0.000308s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000037s :      1: rewriter_after_opt_a
 0.35% :     0.000075s :      1: rewriter_before_opt_a
 0.37% :     0.000079s :      1: symbol_engine_optimizer
31.09% :     0.006576s :      1: task_emit
20.72% :     0.004383s :      1: type_inference
 0.27% :     0.000058s :      1: validate


TotalTime = 0.0134088, [30]
[bootstrap]: 0.0003024
[type_inference]: 0.00484196
[event_method]: 1.732e-05
[auto_monad]: 6.354e-05
[graph_reusing]: 6.66999e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 3.281e-05
[rewriter_before_opt_a]: 7.109e-05
[expand_dump_flag]: 3.46999e-06
[jit_opt_a]: 0.00567681, [2]
    [Cycle 1]: 0.00166406, [27]
        [switch_simplify]: 5.664e-05
        [loop_unroll]: 2.877e-05
        [a_1]: 0.00053436
        [with_stream_mark]: 1.476e-05
        [recompute_prepare]: 7.74002e-06
        [updatestate_depend_eliminate]: 3.23e-06
        [updatestate_assign_eliminate]: 3.76999e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 2.06e-06
        [specialize_transform]: 5.79999e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.34998e-06
        [meta_shard_fg_expand]: 2.02001e-06
        [get_grad_eliminate_]: 5.04e-06
        [merge_forward]: 3.98001e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.465e-05
        [j_node_and_user_rematch]: 9.37999e-06
        [meta_fg_expand]: 2.27999e-06
        [replace_old_param]: 9.82999e-06
        [inline_without_move]: 5.27001e-06
        [renormalize]: 0.00072975
        [add_forward_monad_depend]: 6.11e-06
        [auto_monad_grad]: 1.84e-06
        [auto_monad_eliminator]: 1.244e-05
        [cse]: 1.752e-05
        [replace_applicator]: 1.192e-05
    [Cycle 2]: 0.00032198, [27]
        [switch_simplify]: 5.29998e-06
        [loop_unroll]: 5.15999e-06
        [a_1]: 8.885e-05
        [with_stream_mark]: 1.018e-05
        [recompute_prepare]: 4.86002e-06
        [updatestate_depend_eliminate]: 2.24001e-06
        [updatestate_assign_eliminate]: 2.03002e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 8.79983e-07
        [specialize_transform]: 4.85001e-06
        [updatestate_useless_node_eliminater]: 4.83001e-06
        [accelerated_algorithm]: 5.01997e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 4.33001e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 1.57999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.284e-05
        [j_node_and_user_rematch]: 7.7e-06
        [meta_fg_expand]: 1.61998e-06
        [replace_old_param]: 7.53999e-06
        [inline_without_move]: 4.67e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 5.33002e-06
        [cse]: 1.002e-05
        [replace_applicator]: 5.09e-06
[py_interpret_to_execute_after_opt_a]: 8.23999e-06
[rewriter_after_opt_a]: 3.223e-05
[convert_after_rewriter]: 6.72002e-06
[order_py_execute_after_rewriter]: 4.78001e-06
[mutable_eliminate]: 0.00058471
[jit_opt_b]: 5.761e-05, [1]
    [Cycle 1]: 5.074e-05, [2]
        [frontend_op_eliminate]: 2.246e-05
        [inline_after_opt_a]: 1.599e-05
[cconv]: 2.764e-05
[loop_unroll]: 0.00050948
[jit_opt_after_cconv]: 0.00015036, [1]
    [Cycle 1]: 0.00014347, [11]
        [c_1]: 2.231e-05
        [parameter_eliminate]: 3.29001e-06
        [updatestate_depend_eliminate]: 5.76998e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 2.053e-05
        [call_graph_tuple_transform]: 2.122e-05
        [tuple_list_get_item_eliminator]: 5.75001e-06
        [none_parameter_eliminate]: 1.38002e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 6.24999e-06
[remove_dup_value]: 1.49e-05
[partial_unused_args_eliminate]: 2.29001e-06
[environ_conv]: 5.86e-06
[add_recomputation]: 4.864e-05
[cse_after_recomputation]: 1.99e-05, [1]
    [Cycle 1]: 1.406e-05, [1]
        [cse]: 8.66002e-06
[auto_monad_reorder]: 1.632e-05
[get_jit_bprop_graph]: 2.25002e-06
[rewriter_after_jit_bprop_graph]: 6.81001e-06
[opt_after_jit_grad]: 0.00059599
[symbol_engine_optimizer]: 7.546e-05, [1]
    [Cycle 1]: 6.783e-05, [6]
        [build]: 3.73001e-06
        [elim_shapecalc]: 7.57998e-06
        [elim_not_effective]: 1.294e-05
        [opt_reshape]: 6.46e-06
        [fold_const_symbol]: 8.72e-06
        [renormalize]: 7.10017e-07
[validate]: 4.21e-05
Sums
    bootstrap                            :     0.000302s :  3.31%
    type_inference                       :     0.004842s : 53.00%
    event_method                         :     0.000017s :  0.19%
    auto_monad                           :     0.000064s :  0.70%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.36%
    rewriter_before_opt_a                :     0.000071s :  0.78%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000062s :  0.68%
    jit_opt_a.loop_unroll                :     0.000034s :  0.37%
    jit_opt_a.a_1                        :     0.000623s :  6.82%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000730s :  7.99%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.19%
    jit_opt_a.cse                        :     0.000028s :  0.30%
    jit_opt_a.replace_applicator         :     0.000017s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.09%
    rewriter_after_opt_a                 :     0.000032s :  0.35%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000585s :  6.40%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.25%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.18%
    cconv                                :     0.000028s :  0.30%
    loop_unroll                          :     0.000509s :  5.58%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000049s :  0.53%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000596s :  6.52%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000042s :  0.46%


Time group info:
------[substitution.]   0.000206    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.73% :     0.000006s :      3: substitution.graph_param_transform
83.41% :     0.000172s :      4: substitution.inline
 1.25% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.42% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.38% :     0.000005s :      2: substitution.replace_old_param
 6.15% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004772     2
78.96% :     0.003768s :      1: type_inference.infer
21.04% :     0.001004s :      1: type_inference.specialize
------[replace.]   0.000073     6
67.65% :     0.000049s :      4: replace.inline
32.35% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000181     6
93.63% :     0.000169s :      4: match.inline
 6.37% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.56% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000002s :     10: predicate.addn_zero_filter
 2.03% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.70% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.33% :     0.000000s :      3: predicate.graph_param_transform
 5.55% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.46% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.86% :     0.000002s :     15: predicate.load_eliminater
 1.65% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.25% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.57% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000002s :     15: predicate.partial_eliminate
 1.25% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.10% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.56% :     0.000002s :     10: predicate.reshape_eliminate
 1.45% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.23% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.72% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.87% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.12% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.85% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000616    12
49.12% :     0.000302s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.84% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.04% :     0.000283s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015035    72
 0.35% :     0.000052s :      1: add_recomputation
 0.44% :     0.000067s :      1: auto_monad
 0.13% :     0.000019s :      1: auto_monad_reorder
 2.12% :     0.000318s :      1: bootstrap
 0.20% :     0.000031s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.14% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
37.78% :     0.005680s :      1: jit_opt_a
 1.02% :     0.000154s :      1: jit_opt_after_cconv
 0.40% :     0.000060s :      1: jit_opt_b
 3.44% :     0.000518s :      1: loop_unroll
 3.95% :     0.000594s :      1: mutable_eliminate
 5.54% :     0.000833s :     26: opt.transform.jit_opt_a
 0.35% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000032s :      4: opt.transform.jit_opt_b
 0.09% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000033s :      4: opt.transform.symbol_engine_opt
 4.03% :     0.000605s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000035s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000018s :      1: remove_dup_value
 2.56% :     0.000386s :      1: renormalize.infer
 2.24% :     0.000337s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000036s :      1: rewriter_after_opt_a
 0.49% :     0.000074s :      1: rewriter_before_opt_a
 0.52% :     0.000078s :      1: symbol_engine_optimizer
32.29% :     0.004856s :      1: type_inference


TotalTime = 0.0187047, [33]
[bootstrap]: 0.00031345
[type_inference]: 0.00467024
[event_method]: 1.667e-05
[auto_monad]: 5.07e-05
[graph_reusing]: 4.13001e-06
[pre_auto_parallel]: 2.44001e-06
[py_interpret_to_execute]: 3.578e-05
[rewriter_before_opt_a]: 6.834e-05
[expand_dump_flag]: 2.93003e-06
[jit_opt_a]: 0.00516035, [2]
    [Cycle 1]: 0.00161548, [27]
        [switch_simplify]: 4.759e-05
        [loop_unroll]: 2.912e-05
        [a_1]: 0.00048803
        [with_stream_mark]: 1.194e-05
        [recompute_prepare]: 6.76e-06
        [updatestate_depend_eliminate]: 3.06999e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.24998e-06
        [specialize_transform]: 5.73002e-06
        [updatestate_useless_node_eliminater]: 4.91002e-06
        [accelerated_algorithm]: 5.32001e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 5.31998e-06
        [merge_forward]: 2.27999e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.116e-05
        [j_node_and_user_rematch]: 8.99e-06
        [meta_fg_expand]: 1.85001e-06
        [replace_old_param]: 7.82998e-06
        [inline_without_move]: 5.09e-06
        [renormalize]: 0.00074257
        [add_forward_monad_depend]: 4.63999e-06
        [auto_monad_grad]: 2.02001e-06
        [auto_monad_eliminator]: 1.167e-05
        [cse]: 2.026e-05
        [replace_applicator]: 1.381e-05
    [Cycle 2]: 0.00034562, [27]
        [switch_simplify]: 5.82999e-06
        [loop_unroll]: 5.23002e-06
        [a_1]: 9.375e-05
        [with_stream_mark]: 1.091e-05
        [recompute_prepare]: 6.09001e-06
        [updatestate_depend_eliminate]: 2.48002e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 5.07e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 1.91998e-06
        [get_grad_eliminate_]: 4.60001e-06
        [merge_forward]: 2.65002e-06
        [cell_reuse_recompute_pass]: 2.41e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.341e-05
        [j_node_and_user_rematch]: 7.98001e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 7.68001e-06
        [inline_without_move]: 1.46e-05
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 9.30013e-07
        [auto_monad_eliminator]: 6.07999e-06
        [cse]: 1.197e-05
        [replace_applicator]: 5.02999e-06
[py_interpret_to_execute_after_opt_a]: 1.237e-05
[rewriter_after_opt_a]: 3.263e-05
[convert_after_rewriter]: 6.12999e-06
[order_py_execute_after_rewriter]: 4.62e-06
[mutable_eliminate]: 0.00072412
[jit_opt_b]: 5.645e-05, [1]
    [Cycle 1]: 4.858e-05, [2]
        [frontend_op_eliminate]: 1.909e-05
        [inline_after_opt_a]: 1.71e-05
[cconv]: 2.261e-05
[loop_unroll]: 0.00045357
[jit_opt_after_cconv]: 0.00015173, [1]
    [Cycle 1]: 0.00014372, [11]
        [c_1]: 2.232e-05
        [parameter_eliminate]: 3.03e-06
        [updatestate_depend_eliminate]: 6.33e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.52001e-06
        [cse]: 2.201e-05
        [call_graph_tuple_transform]: 1.799e-05
        [tuple_list_get_item_eliminator]: 5.30001e-06
        [none_parameter_eliminate]: 1.47999e-06
        [renormalize]: 6.80011e-07
        [switch_simplify]: 5.87999e-06
[remove_dup_value]: 1.221e-05
[partial_unused_args_eliminate]: 1.71998e-06
[environ_conv]: 5.02e-06
[add_recomputation]: 4.199e-05
[cse_after_recomputation]: 2.101e-05, [1]
    [Cycle 1]: 1.573e-05, [1]
        [cse]: 1.001e-05
[auto_monad_reorder]: 1.254e-05
[get_jit_bprop_graph]: 1.60999e-06
[rewriter_after_jit_bprop_graph]: 6.28002e-06
[opt_after_jit_grad]: 0.00051651
[symbol_engine_optimizer]: 7.672e-05, [1]
    [Cycle 1]: 6.967e-05, [6]
        [build]: 3.68e-06
        [elim_shapecalc]: 8.12e-06
        [elim_not_effective]: 1.338e-05
        [opt_reshape]: 6.51e-06
        [fold_const_symbol]: 8.2e-06
        [renormalize]: 6.49976e-07
[validate]: 3.887e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.00595369
[execute]: 7.93999e-06
Sums
    bootstrap                            :     0.000313s :  2.11%
    type_inference                       :     0.004670s : 31.40%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000051s :  0.34%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.24%
    rewriter_before_opt_a                :     0.000068s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.36%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000582s :  3.91%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000020s :  0.13%
    jit_opt_a.renormalize                :     0.000743s :  4.99%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000032s :  0.22%
    jit_opt_a.replace_applicator         :     0.000019s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000033s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000724s :  4.87%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.11%
    cconv                                :     0.000023s :  0.15%
    loop_unroll                          :     0.000454s :  3.05%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000012s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000042s :  0.28%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000013s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000517s :  3.47%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005954s : 40.03%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000167    23
 0.86% :     0.000001s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.62% :     0.000004s :      3: substitution.graph_param_transform
83.63% :     0.000139s :      4: substitution.inline
 1.82% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.46% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.14% :     0.000004s :      2: substitution.replace_old_param
 5.88% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004606     2
80.63% :     0.003714s :      1: type_inference.infer
19.37% :     0.000892s :      1: type_inference.specialize
------[replace.]   0.000062     6
63.94% :     0.000040s :      4: replace.inline
36.06% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000146     6
94.00% :     0.000137s :      4: match.inline
 6.00% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.54% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 1.79% :     0.000002s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000001s :     10: predicate.depend_value_elim
 1.29% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.73% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.30% :     0.000006s :     22: predicate.inline
 1.07% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.68% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.62% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.93% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.79% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.30% :     0.000002s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.86% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 1.91% :     0.000002s :     15: predicate.partial_eliminate
 1.33% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.82% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.51% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.11% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.78% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.40% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.62% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.45% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.81% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.81% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000605    12
46.29% :     0.000280s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.83% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.88% :     0.000296s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020347    76
 0.22% :     0.000045s :      1: add_recomputation
 0.27% :     0.000054s :      1: auto_monad
 0.07% :     0.000015s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.63% :     0.000331s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
25.38% :     0.005165s :      1: jit_opt_a
 0.76% :     0.000155s :      1: jit_opt_after_cconv
 0.29% :     0.000059s :      1: jit_opt_b
 2.28% :     0.000464s :      1: loop_unroll
 3.61% :     0.000735s :      1: mutable_eliminate
 3.91% :     0.000795s :     26: opt.transform.jit_opt_a
 0.24% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000029s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.58% :     0.000525s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000038s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000014s :      1: remove_dup_value
 1.90% :     0.000386s :      1: renormalize.infer
 1.72% :     0.000350s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000036s :      1: rewriter_after_opt_a
 0.35% :     0.000071s :      1: rewriter_before_opt_a
 0.39% :     0.000079s :      1: symbol_engine_optimizer
29.35% :     0.005972s :      1: task_emit
23.02% :     0.004684s :      1: type_inference
 0.28% :     0.000057s :      1: validate


TotalTime = 0.0189932, [33]
[bootstrap]: 0.00026625
[type_inference]: 0.00416942
[event_method]: 1.79e-05
[auto_monad]: 5.532e-05
[graph_reusing]: 4.48001e-06
[pre_auto_parallel]: 2.15002e-06
[py_interpret_to_execute]: 3.255e-05
[rewriter_before_opt_a]: 6.517e-05
[expand_dump_flag]: 2.47001e-06
[jit_opt_a]: 0.00540328, [2]
    [Cycle 1]: 0.00168108, [27]
        [switch_simplify]: 5.093e-05
        [loop_unroll]: 3.085e-05
        [a_1]: 0.00055514
        [with_stream_mark]: 1.64e-05
        [recompute_prepare]: 7.81001e-06
        [updatestate_depend_eliminate]: 3.3e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.23002e-06
        [specialize_transform]: 5.46e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 5.67001e-06
        [meta_shard_fg_expand]: 1.64e-06
        [get_grad_eliminate_]: 5.14e-06
        [merge_forward]: 3.81999e-06
        [cell_reuse_recompute_pass]: 9.80013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.152e-05
        [j_node_and_user_rematch]: 8.65001e-06
        [meta_fg_expand]: 2.67001e-06
        [replace_old_param]: 8.65001e-06
        [inline_without_move]: 5.32001e-06
        [renormalize]: 0.00072243
        [add_forward_monad_depend]: 5.32999e-06
        [auto_monad_grad]: 2.61e-06
        [auto_monad_eliminator]: 1.434e-05
        [cse]: 1.969e-05
        [replace_applicator]: 1.16e-05
    [Cycle 2]: 0.00032632, [27]
        [switch_simplify]: 6.17001e-06
        [loop_unroll]: 4.75999e-06
        [a_1]: 9.078e-05
        [with_stream_mark]: 9.27001e-06
        [recompute_prepare]: 5.10001e-06
        [updatestate_depend_eliminate]: 2.63e-06
        [updatestate_assign_eliminate]: 1.94999e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 4.52998e-06
        [merge_forward]: 2.78e-06
        [cell_reuse_recompute_pass]: 1.78002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.255e-05
        [j_node_and_user_rematch]: 8.13999e-06
        [meta_fg_expand]: 1.63002e-06
        [replace_old_param]: 7.93999e-06
        [inline_without_move]: 7.98001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 1.02e-06
        [auto_monad_eliminator]: 5.33002e-06
        [cse]: 1.16e-05
        [replace_applicator]: 4.87e-06
[py_interpret_to_execute_after_opt_a]: 1.183e-05
[rewriter_after_opt_a]: 3.665e-05
[convert_after_rewriter]: 7.82998e-06
[order_py_execute_after_rewriter]: 5.89e-06
[mutable_eliminate]: 0.00073339
[jit_opt_b]: 5.208e-05, [1]
    [Cycle 1]: 4.43e-05, [2]
        [frontend_op_eliminate]: 1.713e-05
        [inline_after_opt_a]: 1.567e-05
[cconv]: 2.994e-05
[loop_unroll]: 0.0005284
[jit_opt_after_cconv]: 0.00015571, [1]
    [Cycle 1]: 0.00014897, [11]
        [c_1]: 2.12e-05
        [parameter_eliminate]: 4.38999e-06
        [updatestate_depend_eliminate]: 6.36e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 2.67001e-06
        [cse]: 2.538e-05
        [call_graph_tuple_transform]: 2.076e-05
        [tuple_list_get_item_eliminator]: 5.62999e-06
        [none_parameter_eliminate]: 1.54998e-06
        [renormalize]: 9.5999e-07
        [switch_simplify]: 5.40001e-06
[remove_dup_value]: 1.48e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 5.62001e-06
[add_recomputation]: 4.861e-05
[cse_after_recomputation]: 2.121e-05, [1]
    [Cycle 1]: 1.533e-05, [1]
        [cse]: 9.72001e-06
[auto_monad_reorder]: 1.528e-05
[get_jit_bprop_graph]: 2.34001e-06
[rewriter_after_jit_bprop_graph]: 7.46001e-06
[opt_after_jit_grad]: 0.00050769
[symbol_engine_optimizer]: 7.848e-05, [1]
    [Cycle 1]: 7.047e-05, [6]
        [build]: 4.18001e-06
        [elim_shapecalc]: 7.52998e-06
        [elim_not_effective]: 1.481e-05
        [opt_reshape]: 6.23e-06
        [fold_const_symbol]: 8.77999e-06
        [renormalize]: 7.09988e-07
[validate]: 3.934e-05
[backend_pass]: 1.12e-06
[task_emit]: 0.00645199
[execute]: 9.75002e-06
Sums
    bootstrap                            :     0.000266s :  1.78%
    type_inference                       :     0.004169s : 27.83%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000055s :  0.37%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000033s :  0.22%
    rewriter_before_opt_a                :     0.000065s :  0.44%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.38%
    jit_opt_a.loop_unroll                :     0.000036s :  0.24%
    jit_opt_a.a_1                        :     0.000646s :  4.31%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000013s :  0.09%
    jit_opt_a.renormalize                :     0.000723s :  4.82%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.13%
    jit_opt_a.cse                        :     0.000031s :  0.21%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000037s :  0.24%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000733s :  4.90%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000030s :  0.20%
    loop_unroll                          :     0.000528s :  3.53%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000025s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000015s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000049s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.05%
    opt_after_jit_grad                   :     0.000508s :  3.39%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006452s : 43.07%
    execute                              :     0.000010s :  0.07%


Time group info:
------[substitution.]   0.000213    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.39% :     0.000005s :      3: substitution.graph_param_transform
85.25% :     0.000181s :      4: substitution.inline
 1.52% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.88% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.12% :     0.000005s :      2: substitution.replace_old_param
 5.40% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004119     2
81.35% :     0.003351s :      1: type_inference.infer
18.65% :     0.000768s :      1: type_inference.specialize
------[replace.]   0.000069     6
65.53% :     0.000045s :      4: replace.inline
34.47% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000189     6
94.54% :     0.000178s :      4: match.inline
 5.46% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.22% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.70% :     0.000003s :     10: predicate.arithmetic_simplify
 1.40% :     0.000002s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.04% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.60% :     0.000001s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.04% :     0.000006s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.57% :     0.000002s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.88% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.41% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.77% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.82% :     0.000002s :     10: predicate.reduce_eliminate
 1.72% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.93% :     0.000001s :      6: predicate.replace_old_param
 0.45% :     0.000001s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.02% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.63% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.48% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.56% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.30% :     0.000002s :     10: predicate.transpose_eliminate
 1.79% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.27% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.83% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.81% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000565    12
47.03% :     0.000266s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.99% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.98% :     0.000254s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020681    76
 0.25% :     0.000052s :      1: add_recomputation
 0.28% :     0.000059s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.35% :     0.000280s :      1: bootstrap
 0.16% :     0.000033s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.07% :     0.000015s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
26.14% :     0.005406s :      1: jit_opt_a
 0.77% :     0.000158s :      1: jit_opt_after_cconv
 0.27% :     0.000055s :      1: jit_opt_b
 2.60% :     0.000538s :      1: loop_unroll
 3.60% :     0.000744s :      1: mutable_eliminate
 4.14% :     0.000855s :     26: opt.transform.jit_opt_a
 0.24% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.50% :     0.000517s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000035s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.85% :     0.000382s :      1: renormalize.infer
 1.60% :     0.000331s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000039s :      1: rewriter_after_opt_a
 0.33% :     0.000068s :      1: rewriter_before_opt_a
 0.39% :     0.000081s :      1: symbol_engine_optimizer
31.28% :     0.006470s :      1: task_emit
20.22% :     0.004181s :      1: type_inference
 0.30% :     0.000062s :      1: validate


TotalTime = 0.0143997, [30]
[bootstrap]: 0.00034174
[type_inference]: 0.00497513
[event_method]: 1.837e-05
[auto_monad]: 6.457e-05
[graph_reusing]: 5.86e-06
[pre_auto_parallel]: 2.92002e-06
[py_interpret_to_execute]: 3.752e-05
[rewriter_before_opt_a]: 7.246e-05
[expand_dump_flag]: 3.5e-06
[jit_opt_a]: 0.006666, [2]
    [Cycle 1]: 0.00216834, [27]
        [switch_simplify]: 7.046e-05
        [loop_unroll]: 2.96e-05
        [a_1]: 0.0006148
        [with_stream_mark]: 1.796e-05
        [recompute_prepare]: 1.084e-05
        [updatestate_depend_eliminate]: 3.06999e-06
        [updatestate_assign_eliminate]: 2.51998e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 2.02999e-06
        [specialize_transform]: 6.73e-06
        [updatestate_useless_node_eliminater]: 6.66999e-06
        [accelerated_algorithm]: 7.04001e-06
        [meta_shard_fg_expand]: 2.20002e-06
        [get_grad_eliminate_]: 6.83e-06
        [merge_forward]: 3.55e-06
        [cell_reuse_recompute_pass]: 1.47999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.754e-05
        [j_node_and_user_rematch]: 1.133e-05
        [meta_fg_expand]: 3.19001e-06
        [replace_old_param]: 1.13e-05
        [inline_without_move]: 6.61999e-06
        [renormalize]: 0.0010818
        [add_forward_monad_depend]: 6.62002e-06
        [auto_monad_grad]: 1.45999e-06
        [auto_monad_eliminator]: 1.157e-05
        [cse]: 1.849e-05
        [replace_applicator]: 1.561e-05
    [Cycle 2]: 0.00033897, [27]
        [switch_simplify]: 6.19999e-06
        [loop_unroll]: 5.32001e-06
        [a_1]: 0.00010026
        [with_stream_mark]: 9.09998e-06
        [recompute_prepare]: 5.34998e-06
        [updatestate_depend_eliminate]: 2.59001e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 4.82e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 6.14001e-06
        [meta_shard_fg_expand]: 1.21997e-06
        [get_grad_eliminate_]: 4.48001e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.90001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.301e-05
        [j_node_and_user_rematch]: 7.78999e-06
        [meta_fg_expand]: 1.74998e-06
        [replace_old_param]: 1.121e-05
        [inline_without_move]: 4.89998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 4.67e-06
        [cse]: 1.089e-05
        [replace_applicator]: 5.02e-06
[py_interpret_to_execute_after_opt_a]: 1.235e-05
[rewriter_after_opt_a]: 2.74e-05
[convert_after_rewriter]: 5.74999e-06
[order_py_execute_after_rewriter]: 7.94002e-06
[mutable_eliminate]: 0.00067192
[jit_opt_b]: 5.15e-05, [1]
    [Cycle 1]: 4.473e-05, [2]
        [frontend_op_eliminate]: 1.73e-05
        [inline_after_opt_a]: 1.592e-05
[cconv]: 1.733e-05
[loop_unroll]: 0.00043082
[jit_opt_after_cconv]: 0.00013656, [1]
    [Cycle 1]: 0.00012958, [11]
        [c_1]: 2.09e-05
        [parameter_eliminate]: 3.4e-06
        [updatestate_depend_eliminate]: 4.65999e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 1.692e-05
        [call_graph_tuple_transform]: 1.766e-05
        [tuple_list_get_item_eliminator]: 5.20001e-06
        [none_parameter_eliminate]: 9.80013e-07
        [renormalize]: 4.00003e-07
        [switch_simplify]: 4.98001e-06
[remove_dup_value]: 9.99001e-06
[partial_unused_args_eliminate]: 2.36998e-06
[environ_conv]: 4.92e-06
[add_recomputation]: 4.339e-05
[cse_after_recomputation]: 2.034e-05, [1]
    [Cycle 1]: 1.491e-05, [1]
        [cse]: 8.96002e-06
[auto_monad_reorder]: 1.486e-05
[get_jit_bprop_graph]: 1.96998e-06
[rewriter_after_jit_bprop_graph]: 6.61e-06
[opt_after_jit_grad]: 0.00046215
[symbol_engine_optimizer]: 7.266e-05, [1]
    [Cycle 1]: 6.598e-05, [6]
        [build]: 2.79999e-06
        [elim_shapecalc]: 8.05e-06
        [elim_not_effective]: 1.27e-05
        [opt_reshape]: 6.18998e-06
        [fold_const_symbol]: 8.57e-06
        [renormalize]: 5.59987e-07
[validate]: 2.895e-05
Sums
    bootstrap                            :     0.000342s :  3.55%
    type_inference                       :     0.004975s : 51.68%
    event_method                         :     0.000018s :  0.19%
    auto_monad                           :     0.000065s :  0.67%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000038s :  0.39%
    rewriter_before_opt_a                :     0.000072s :  0.75%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000077s :  0.80%
    jit_opt_a.loop_unroll                :     0.000035s :  0.36%
    jit_opt_a.a_1                        :     0.000715s :  7.43%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000023s :  0.23%
    jit_opt_a.inline_without_move        :     0.000012s :  0.12%
    jit_opt_a.renormalize                :     0.001082s : 11.24%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.17%
    jit_opt_a.cse                        :     0.000029s :  0.31%
    jit_opt_a.replace_applicator         :     0.000021s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.13%
    rewriter_after_opt_a                 :     0.000027s :  0.28%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000008s :  0.08%
    mutable_eliminate                    :     0.000672s :  6.98%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.17%
    cconv                                :     0.000017s :  0.18%
    loop_unroll                          :     0.000431s :  4.48%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000010s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000043s :  0.45%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000462s :  4.80%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000029s :  0.30%


Time group info:
------[substitution.]   0.000259    23
 0.61% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 1.48% :     0.000004s :      3: substitution.graph_param_transform
86.81% :     0.000225s :      4: substitution.inline
 1.31% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.31% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.89% :     0.000005s :      2: substitution.replace_old_param
 5.05% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004903     2
80.67% :     0.003955s :      1: type_inference.infer
19.33% :     0.000948s :      1: type_inference.specialize
------[replace.]   0.000079     6
64.22% :     0.000051s :      4: replace.inline
35.78% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000234     6
94.86% :     0.000222s :      4: match.inline
 5.14% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.12% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.20% :     0.000001s :     10: predicate.addn_zero_filter
 2.44% :     0.000003s :     10: predicate.arithmetic_simplify
 1.27% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.69% :     0.000002s :     10: predicate.depend_value_elim
 1.58% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.64% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.73% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.74% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.28% :     0.000006s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.95% :     0.000002s :      6: predicate.less_batch_normalization
 1.37% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 1.19% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.74% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.16% :     0.000001s :     10: predicate.minmaximum_grad
 1.25% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.06% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.99% :     0.000002s :     10: predicate.reduce_eliminate
 1.96% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.14% :     0.000001s :      6: predicate.specialize_transform
 1.19% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.66% :     0.000003s :     16: predicate.switch_defer_inline
 1.90% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.16% :     0.000010s :     53: predicate.switch_simplify
 1.04% :     0.000001s :     10: predicate.tile_eliminate
 1.07% :     0.000001s :     10: predicate.transpose_eliminate
 1.66% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.37% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 2.13% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.86% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.57% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.99% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000707    12
46.02% :     0.000326s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.82% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.16% :     0.000334s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016499    72
 0.28% :     0.000047s :      1: add_recomputation
 0.41% :     0.000068s :      1: auto_monad
 0.10% :     0.000017s :      1: auto_monad_reorder
 2.21% :     0.000364s :      1: bootstrap
 0.12% :     0.000020s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000022s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
40.42% :     0.006669s :      1: jit_opt_a
 0.84% :     0.000139s :      1: jit_opt_after_cconv
 0.33% :     0.000054s :      1: jit_opt_b
 2.66% :     0.000438s :      1: loop_unroll
 4.13% :     0.000681s :      1: mutable_eliminate
 5.84% :     0.000963s :     26: opt.transform.jit_opt_a
 0.27% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.85% :     0.000469s :      1: opt_after_jit_grad
 0.06% :     0.000010s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000040s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000012s :      1: remove_dup_value
 3.69% :     0.000608s :      1: renormalize.infer
 2.81% :     0.000464s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000031s :      1: rewriter_after_opt_a
 0.46% :     0.000076s :      1: rewriter_before_opt_a
 0.46% :     0.000075s :      1: symbol_engine_optimizer
30.26% :     0.004992s :      1: type_inference


TotalTime = 0.0177263, [33]
[bootstrap]: 0.0002784
[type_inference]: 0.00436304
[event_method]: 1.799e-05
[auto_monad]: 5.8e-05
[graph_reusing]: 4.55001e-06
[pre_auto_parallel]: 2.28998e-06
[py_interpret_to_execute]: 3.713e-05
[rewriter_before_opt_a]: 7.407e-05
[expand_dump_flag]: 2.94001e-06
[jit_opt_a]: 0.00556685, [2]
    [Cycle 1]: 0.00148576, [27]
        [switch_simplify]: 4.78e-05
        [loop_unroll]: 3.136e-05
        [a_1]: 0.00051965
        [with_stream_mark]: 1.279e-05
        [recompute_prepare]: 7.1e-06
        [updatestate_depend_eliminate]: 3.24001e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [parameter_eliminate]: 1.16997e-06
        [specialize_transform]: 6.34999e-06
        [updatestate_useless_node_eliminater]: 5.44e-06
        [accelerated_algorithm]: 5.61998e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 5.35999e-06
        [merge_forward]: 2.83e-06
        [cell_reuse_recompute_pass]: 1.31998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.134e-05
        [j_node_and_user_rematch]: 9.29e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 8.15999e-06
        [inline_without_move]: 5.69999e-06
        [renormalize]: 0.00059451
        [add_forward_monad_depend]: 4.91997e-06
        [auto_monad_grad]: 1.64e-06
        [auto_monad_eliminator]: 9.71e-06
        [cse]: 1.628e-05
        [replace_applicator]: 1.034e-05
    [Cycle 2]: 0.00032201, [27]
        [switch_simplify]: 5.51002e-06
        [loop_unroll]: 4.99e-06
        [a_1]: 9.161e-05
        [with_stream_mark]: 8.77e-06
        [recompute_prepare]: 5.25001e-06
        [updatestate_depend_eliminate]: 2.85002e-06
        [updatestate_assign_eliminate]: 2.23998e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.99003e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.32999e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 4.79e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 2.00002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.172e-05
        [j_node_and_user_rematch]: 7.66999e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 7.06001e-06
        [inline_without_move]: 4.70001e-06
        [renormalize]: 1.30007e-07
        [add_forward_monad_depend]: 1.19998e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 4.99e-06
        [cse]: 8.75999e-06
        [replace_applicator]: 5.10001e-06
[py_interpret_to_execute_after_opt_a]: 8.50001e-06
[rewriter_after_opt_a]: 2.46e-05
[convert_after_rewriter]: 5.20001e-06
[order_py_execute_after_rewriter]: 3.96001e-06
[mutable_eliminate]: 0.00047538
[jit_opt_b]: 4.917e-05, [1]
    [Cycle 1]: 4.232e-05, [2]
        [frontend_op_eliminate]: 1.653e-05
        [inline_after_opt_a]: 1.482e-05
[cconv]: 1.565e-05
[loop_unroll]: 0.00040586
[jit_opt_after_cconv]: 0.00013459, [1]
    [Cycle 1]: 0.0001286, [11]
        [c_1]: 2.072e-05
        [parameter_eliminate]: 2.27001e-06
        [updatestate_depend_eliminate]: 4.33999e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 1.65e-05
        [call_graph_tuple_transform]: 1.699e-05
        [tuple_list_get_item_eliminator]: 5.40999e-06
        [none_parameter_eliminate]: 9.29984e-07
        [renormalize]: 5.19998e-07
        [switch_simplify]: 5.46998e-06
[remove_dup_value]: 6.81001e-06
[partial_unused_args_eliminate]: 1.92999e-06
[environ_conv]: 3.81999e-06
[add_recomputation]: 3.145e-05
[cse_after_recomputation]: 2.06e-05, [1]
    [Cycle 1]: 1.505e-05, [1]
        [cse]: 9.24998e-06
[auto_monad_reorder]: 1.034e-05
[get_jit_bprop_graph]: 1.65001e-06
[rewriter_after_jit_bprop_graph]: 5.14e-06
[opt_after_jit_grad]: 0.00048806
[symbol_engine_optimizer]: 7.328e-05, [1]
    [Cycle 1]: 6.63e-05, [6]
        [build]: 2.83e-06
        [elim_shapecalc]: 
TotalTime = 0.0198039, [33]
[bootstrap]: 0.0003058
[type_inference]: 0.00461635
[event_method]: 1.628e-05
[auto_monad]: 4.492e-05
[graph_reusing]: 3.67002e-06
[pre_auto_parallel]: 1.86e-06
[py_interpret_to_execute]: 3.719e-05
[rewriter_before_opt_a]: 6.633e-05
[expand_dump_flag]: 2.39001e-06
[jit_opt_a]: 0.0059309, [2]
    [Cycle 1]: 0.00210827, [27]
        [switch_simplify]: 5.901e-05
        [loop_unroll]: 3.038e-05
        [a_1]: 0.00057424
        [with_stream_mark]: 2.074e-05
        [recompute_prepare]: 1.016e-05
        [updatestate_depend_eliminate]: 3.61999e-06
        [updatestate_assign_eliminate]: 2.99001e-06
        [updatestate_loads_eliminate]: 2.46998e-06
        [parameter_eliminate]: 2.12999e-06
        [specialize_transform]: 6.51e-06
        [updatestate_useless_node_eliminater]: 5.35999e-06
        [accelerated_algorithm]: 6.55997e-06
        [meta_shard_fg_expand]: 2.60997e-06
        [get_grad_eliminate_]: 6.09999e-06
        [merge_forward]: 4.84003e-06
        [cell_reuse_recompute_pass]: 1.74e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.784e-05
        [j_node_and_user_rematch]: 1.073e-05
        [meta_fg_expand]: 3.13e-06
        [replace_old_param]: 1.195e-05
        [inline_without_move]: 5.91e-06
        [renormalize]: 0.00106084
        [add_forward_monad_depend]: 7.27002e-06
        [auto_monad_grad]: 3.30998e-06
        [auto_monad_eliminator]: 1.54e-05
        [cse]: 2.835e-05
        [replace_applicator]: 1.632e-05
    [Cycle 2]: 0.00032912, [27]
        [switch_simplify]: 5.56e-06
        [loop_unroll]: 5.34e-06
        [a_1]: 9.791e-05
        [with_stream_mark]: 1.258e-05
        [recompute_prepare]: 5.10001e-06
        [updatestate_depend_eliminate]: 3.46001e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.23002e-06
        [parameter_eliminate]: 1.59998e-06
        [specialize_transform]: 4.75001e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.20001e-06
        [meta_shard_fg_expand]: 1.55999e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.52001e-06
        [cell_reuse_recompute_pass]: 1.60999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.27e-05
        [j_node_and_user_rematch]: 7.71999e-06
        [meta_fg_expand]: 1.78997e-06
        [replace_old_param]: 6.94001e-06
        [inline_without_move]: 4.37e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 5.37999e-06
        [cse]: 9.71e-06
        [replace_applicator]: 4.97999e-06
[py_interpret_to_execute_after_opt_a]: 1.28e-05
[rewriter_after_opt_a]: 3.614e-05
[convert_after_rewriter]: 7.31001e-06
[order_py_execute_after_rewriter]: 5.24998e-06
[mutable_eliminate]: 0.00067357
[jit_opt_b]: 5.051e-05, [1]
    [Cycle 1]: 4.23e-05, [2]
        [frontend_op_eliminate]: 1.662e-05
        [inline_after_opt_a]: 1.416e-05
[cconv]: 2.351e-05
[loop_unroll]: 0.00043236
[jit_opt_after_cconv]: 0.00013376, [1]
    [Cycle 1]: 0.00012718, [11]
        [c_1]: 2.07e-05
        [parameter_eliminate]: 1.62999e-06
        [updatestate_depend_eliminate]: 4.42e-06
        [updatestate_assign_eliminate]: 2.10002e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.564e-05
        [call_graph_tuple_transform]: 1.712e-05
        [tuple_list_get_item_eliminator]: 5.22e-06
        [none_parameter_eliminate]: 1.19998e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 5.40001e-06
[remove_dup_value]: 1.284e-05
[partial_unused_args_eliminate]: 1.79998e-06
[environ_conv]: 4.48999e-06
[add_recomputation]: 4.366e-05
[cse_after_recomputation]: 1.994e-05, [1]
    [Cycle 1]: 1.441e-05, [1]
        [cse]: 8.47e-06
[auto_monad_reorder]: 1.429e-05
[get_jit_bprop_graph]: 1.81e-06
[rewriter_after_jit_bprop_graph]: 5.56998e-06
[opt_after_jit_grad]: 0.00044658
[symbol_engine_optimizer]: 7.2e-05, [1]
    [Cycle 1]: 6.61e-05, [6]
        [build]: 3.78001e-06
        [elim_shapecalc]: 7.48e-06
  8.01001e-06
        [elim_not_effective]: 1.189e-05
        [opt_reshape]: 5.64e-06
        [fold_const_symbol]: 9.12999e-06
        [renormalize]: 5.89993e-07
[validate]: 2.843e-05
[backend_pass]: 7.29982e-07
[task_emit]: 0.00531664
[execute]: 8.02e-06
Sums
    bootstrap                            :     0.000278s :  2.08%
    type_inference                       :     0.004363s : 32.63%
    event_method                         :     0.000018s :  0.13%
    auto_monad                           :     0.000058s :  0.43%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.28%
    rewriter_before_opt_a                :     0.000074s :  0.55%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.40%
    jit_opt_a.loop_unroll                :     0.000036s :  0.27%
    jit_opt_a.a_1                        :     0.000611s :  4.57%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000595s :  4.45%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000025s :  0.19%
    jit_opt_a.replace_applicator         :     0.000015s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000025s :  0.18%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000475s :  3.56%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000016s :  0.12%
    loop_unroll                          :     0.000406s :  3.04%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :      [elim_not_effective]: 1.322e-05
        [opt_reshape]: 5.74e-06
        [fold_const_symbol]: 8.53001e-06
        [renormalize]: 8.89995e-07
[validate]: 3.545e-05
[backend_pass]: 1.35999e-06
[task_emit]: 0.00653462
[execute]: 7.96001e-06
Sums
    bootstrap                            :     0.000306s :  1.95%
    type_inference                       :     0.004616s : 29.43%
    event_method                         :     0.000016s :  0.10%
    auto_monad                           :     0.000045s :  0.29%
    graph_reusing                        :     0.000004s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000037s :  0.24%
    rewriter_before_opt_a                :     0.000066s :  0.42%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.41%
    jit_opt_a.loop_unroll                :     0.000036s :  0.23%
    jit_opt_a.a_1                        :     0.000672s :  4.28%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.001061s :  6.76%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.13%
    jit_opt_a.cse                        :     0.000038s :  0.24%
    jit_opt_a.replace_applicator         :     0.000021s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000036s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000674s :  4.29%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000024s :  0.15%
    loop_unroll                          :     0.000432s :  2.76%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.0000     0.000007s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000031s :  0.24%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000010s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000488s :  3.65%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000028s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005317s : 39.77%
    execute                              :     0.000008s :  0.06%

13s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000044s :  0.28%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000447s :  2.85%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000035s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006535s : 41.65%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000183    23
 0.73% :     0.000001s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.07% :     0.000004s :      3: substitution.graph_param_transform
84.89% :     0.000155s :      4: substitution.inline
 1.61% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.03% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.65% :     0.000003s :      2: substitution.replace_old_param
 6.42% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004293     2
80.36% :     0.003450s :      1: type_inference.infer
19.64% :     0.000843s :      1: type_inference.specialize
------[replace.]   0.000066     6
66.76% :     0.000044s :      4: replace.inline
33.24% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000164     6
93.52% :     0.000153s :      4: match.inline
 6.48% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.76% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.30% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.21% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 2.49% :     0.000003s :     10: predicate.arithmetic_simplify
 1.25% :     0.000002s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     10: predicate.compare_switch_simplify
 1.53% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.39% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     10: predicate.environ_get_eliminate
 1.23% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 4.93% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.11% :     0.000003s :     15: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.33% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.20% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.25% :     0.000002s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.46% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000002s :     15: predicate.partial_eliminate
 1.38% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3:
Time group info:
------[substitution.]   0.000231    23
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.22% :     0.000005s :      3: substitution.graph_param_transform
84.31% :     0.000195s :      4: substitution.inline
 1.80% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.22% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.37% :     0.000005s :      2: substitution.replace_old_param
 5.79% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004554     2
80.68% :     0.003674s :      1: type_inference.infer
19.32% :     0.000880s :      1: type_inference.specialize
------[replace.]   0.000074     6
66.20% :     0.000049s :      4: replace.inline
33.80% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000205     6
93.99% :     0.000192s :      4: match.inline
 6.01% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.30% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.23% :     0.000002s :     10: predicate.addn_zero_filter
 1.76% :     0.000002s :     10: predicate.arithmetic_simplify
 2.14% :     0.000003s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.78% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.54% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.61% :     0.000002s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.11% :     0.000001s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 4.71% :     0.000006s :     22: predicate.inline
 1.06% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.95% :     0.000002s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.89% :     0.000002s :     15: predicate.load_eliminater
 1.12% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.79% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.28% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.36% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000000s :      3: predicate.opt_reshape
 1.88% :     0.000002s :     15: predicate.partial_eliminate
 1.43% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.67% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.77% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.62% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.58% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.63% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.56% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.44% :     0.000009s :     53: predicate.switch_simplify
 1.40% :     0.000002s :     10: predicate.tile_eliminate
 1.34% :     0.000002s :     10: predicate.transpose_eliminate
 1.26% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.37% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000520    12
44.47% :     0.000231s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.29% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.24% :     0.000256s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019234    76
 0.18% :     0.000034s :      1: add_recomputation
 0.32% :     0.000062s :      1: auto_monad
 0.07% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.53% :     0.000293s :      1: bootstrap
 0.09% :     0.000018s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.12% :     0.000023s :      1: event_method
 0.07% :     0.000013s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
28.96% :     0.005571s :      1: jit_opt_a
 0.71% :     0.000137s :      1: jit_opt_after_cconv
 0.27% :     0.000052s :      1: jit_opt_b
 2.14% :     0.000413s :      1: loop_unroll
 2.52% :     0.000484s :      1: mutable_eliminate
 4.23% :     0.000814s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.57% :     0.000495s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.21% :     0.000040s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000009s :      1: remove_dup_value
 1.64% :     0.000316s :      1: renormalize.infer
 1.41% :     0.000272s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000027s :      1: rewriter_after_opt_a
 0.40% :     0.000077s :      1: rewriter_before_opt_a
 0.40% :     0.000076s :      1: symbol_engine_optimizer
27.73% :     0.005334s :       predicate.row_tensor_eliminate
 1.58% :     0.000002s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 0.85% :     0.000001s :      6: predicate.specialize_transform
 1.18% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.80% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.00% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.64% :     0.000009s :     53: predicate.switch_simplify
 1.40% :     0.000002s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.85% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.36% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 2.15% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000674    12
43.67% :     0.000294s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.11% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.22% :     0.000332s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021854    76
 0.21% :     0.000047s :      1: add_recomputation
 0.22% :     0.000048s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.47% :     0.000321s :      1: bootstrap
 0.12% :     0.000026s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.09% :     0.000020s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
27.16% :     0.005935s :      1: jit_opt_a
 0.62% :     0.000136s :      1: jit_opt_after_cconv
 0.24% :     0.000053s :      1: jit_opt_b
 2.01% :     0.000438s :      1: loop_unroll
 3.11% :     0.000681s :      1: mutable_eliminate
 4.13% :     0.000902s :     26: opt.transform.jit_opt_a
 0.21% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.07% :     0.000453s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000040s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.78% :     0.000607s :      1: renormalize.infer
 2.03% :     0.000444s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000040s :      1: rewriter_after_opt_a
 0.32% :     0.000069s :      1: rewriter_before_opt_a
 0.34% :     0.000074s :      1: symbol_engine_optimizer
29.98% :     0.006553s :      1: task_emit
22.78% :     0.004382s :      1: type_inference
 0.23% :     0.000044s :      1: validate

1: task_emit
21.19% :     0.004630s :      1: type_inference
 0.25% :     0.000055s :      1: validate


TotalTime = 0.0929639, [33]
[bootstrap]: 0.00023389
[type_inference]: 0.0717984
[event_method]: 0.00027919
[auto_monad]: 0.00011393
[graph_reusing]: 7.80998e-06
[pre_auto_parallel]: 3.05002e-06
[py_interpret_to_execute]: 4.353e-05
[rewriter_before_opt_a]: 0.00014921
[expand_dump_flag]: 4.04002e-06
[jit_opt_a]: 0.0181862, [3]
    [Cycle 1]: 0.0115132, [27]
        [switch_simplify]: 6.403e-05
        [loop_unroll]: 4.299e-05
        [a_1]: 0.00091378
        [with_stream_mark]: 2.473e-05
        [recompute_prepare]: 1.977e-05
        [updatestate_depend_eliminate]: 7.91001e-06
        [updatestate_assign_eliminate]: 6.73e-06
        [updatestate_loads_eliminate]: 6.19001e-06
        [parameter_eliminate]: 2.63003e-06
        [specialize_transform]: 1.447e-05
        [updatestate_useless_node_eliminater]: 1.331e-05
        [accelerated_algorithm]: 1.4e-05
        [meta_shard_fg_expand]: 3.93001e-06
        [get_grad_eliminate_]: 1.317e-05
        [merge_forward]: 8.25999e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.717e-05
        [j_node_and_user_rematch]: 2.387e-05
        [meta_fg_expand]: 0.0014606
        [replace_old_param]: 5.804e-05
        [inline_without_move]: 5.469e-05
        [renormalize]: 0.00819683
        [add_forward_monad_depend]: 1.031e-05
        [auto_monad_grad]: 5.87001e-06
        [auto_monad_eliminator]: 5.078e-05
        [cse]: 0.00019522
        [replace_applicator]: 7.338e-05
    [Cycle 2]: 0.00324532, [27]
        [switch_simplify]: 4.315e-05
        [loop_unroll]: 4.113e-05
        [a_1]: 0.0011265
        [with_stream_mark]: 1.296e-05
        [recompute_prepare]: 9.42999e-06
        [updatestate_depend_eliminate]: 4.12e-06
        [updatestate_assign_eliminate]: 3.36999e-06
        [updatestate_loads_eliminate]: 3.37002e-06
        [parameter_eliminate]: 1.66002e-06
        [specialize_transform]: 7.98001e-06
        [updatestate_useless_node_eliminater]: 7.2e-06
        [accelerated_algorithm]: 7.18998e-06
        [meta_shard_fg_expand]: 2.69001e-06
        [get_grad_eliminate_]: 6.98e-06
        [merge_forward]: 3.85998e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.549e-05
        [j_node_and_user_rematch]: 1.081e-05
        [meta_fg_expand]: 5.271e-05
        [replace_old_param]: 7.83999e-06
        [inline_without_move]: 7.21001e-06
        [renormalize]: 0.00167377
        [add_forward_monad_depend]: 4.42998e-06
        [auto_monad_grad]: 1.19e-06
        [auto_monad_eliminator]: 9.52001e-06
        [cse]: 1.575e-05
        [replace_applicator]: 1.081e-05
    [Cycle 3]: 0.00022844, [27]
        [switch_simplify]: 4.58999e-06
        [loop_unroll]: 3.85e-06
        [a_1]: 3.766e-05
        [with_stream_mark]: 7.78001e-06
        [recompute_prepare]: 3.60998e-06
        [updatestate_depend_eliminate]: 2.14e-06
        [updatestate_assign_eliminate]: 1.66002e-06
        [updatestate_loads_eliminate]: 1.39e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 3.55e-06
        [updatestate_useless_node_eliminater]: 3.33e-06
        [accelerated_algorithm]: 3.25e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 3.06001e-06
        [merge_forward]: 2.05002e-06
        [cell_reuse_recompute_pass]: 1.60001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.96e-06
        [j_node_and_user_rematch]: 5.30999e-06
        [meta_fg_expand]: 1.29e-06
        [replace_old_param]: 3.08e-06
        [inline_without_move]: 2.98e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 9.20001e-07
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 3.4e-06
        [cse]: 7.58001e-06
        [replace_applicator]: 3.46999e-06
[py_interpret_to_execute_after_opt_a]: 9.79999e-06
[rewriter_after_opt_a]: 3.156e-05
[convert_after_rewriter]: 6.41e-06
[order_py_execute_after_rewriter]: 4.74e-06
[mutable_eliminate]: 0.00056983
[jit_opt_b]: 4.019e-05, [1]
    [Cycle 1]: 3.267e-05, [2]
        [frontend_op_eliminate]: 9.80002e-06
        [inline_after_opt_a]: 1.139e-05
[cconv]: 2.36e-05
[loop_unroll]: 0.00041866
[jit_opt_after_cconv]: 0.00011817, [1]
    [Cycle 1]: 0.00011203, [11]
        [c_1]: 1.275e-05
        [parameter_eliminate]: 2.83e-06
        [updatestate_depend_eliminate]: 4.49002e-06
        [updatestate_assign_eliminate]: 1.74998e-06
        [updatestate_loads_eliminate]: 1.77001e-06
        [cse]: 1.607e-05
        [call_graph_tuple_transform]: 1.382e-05
        [tuple_list_get_item_eliminator]: 5.10999e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 7.60017e-07
        [switch_simplify]: 4.12003e-06
[remove_dup_value]: 1.363e-05
[partial_unused_args_eliminate]: 1.96e-06
[environ_conv]: 4.62e-06
[add_recomputation]: 3.57e-05
[cse_after_recomputation]: 1.825e-05, [1]
    [Cycle 1]: 1.317e-05, [1]
        [cse]: 7.39002e-06
[auto_monad_reorder]: 1.327e-05
[get_jit_bprop_graph]: 2.35002e-06
[rewriter_after_jit_bprop_graph]: 4.72e-06
[opt_after_jit_grad]: 0.00050755
[symbol_engine_optimizer]: 6.677e-05, [1]
    [Cycle 1]: 6.086e-05, [6]
        [build]: 3.57002e-06
        [elim_shapecalc]: 6.58998e-06
        [elim_not_effective]: 1.201e-05
        [opt_reshape]: 3.88999e-06
        [fold_const_symbol]: 6.89001e-06
        [renormalize]: 5.19998e-07
[validate]: 2.889e-05
[backend_pass]: 8.50006e-07
[task_emit]: 2.722e-05
[execute]: 1.18001e-06
Sums
    bootstrap                            :     0.000234s :  0.26%
    type_inference                       :     0.071798s : 80.68%
    event_method                         :     0.000279s :  0.31%
    auto_monad                           :     0.000114s :  0.13%
    graph_reusing                        :     0.000008s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000044s :  0.05%
    rewriter_before_opt_a                :     0.000149s :  0.17%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000112s :  0.13%
    jit_opt_a.loop_unroll                :     0.000088s :  0.10%
    jit_opt_a.a_1                        :     0.002078s :  2.33%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.05%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000026s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000024s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000023s :  0.03%
    jit_opt_a.merge_forward              :     0.000014s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.00%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000053s :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000040s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.001515s :  1.70%
    jit_opt_a.replace_old_param          :     0.000069s :  0.08%
    jit_opt_a.inline_without_move        :     0.000065s :  0.07%
    jit_opt_a.renormalize                :     0.009871s : 11.09%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000064s :  0.07%
    jit_opt_a.cse                        :     0.000219s :  0.25%
    jit_opt_a.replace_applicator         :     0.000088s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.01%
    rewriter_after_opt_a                 :     0.000032s :  0.04%
    convert_after_rewriter               :     0.000006s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000570s :  0.64%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.01%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.01%
    cconv                                :     0.000024s :  0.03%
    loop_unroll                          :     0.000419s :  0.47%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.01%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000016s :  0.02%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.02%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.00%
    remove_dup_value                     :     0.000014s :  0.02%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000005s :  0.01%
    add_recomputation                    :     0.000036s :  0.04%
    cse_after_recomputation.cse          :     0.000007s :  0.01%
    auto_monad_reorder                   :     0.000013s :  0.01%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000508s :  0.57%
    symbol_engine_optimizer.build        :     0.000004s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.01%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.00%
    symbol_engine_optimizer.fold_const_symbol :     0.000007s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000029s :  0.03%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000027s :  0.03%
    execute                              :     0.000001s :  0.00%


Time group info:
------[substitution.]   0.000472   102
 0.63% :     0.000003s :      1: substitution.elim_not_effective
 0.25% :     0.000001s :      1: substitution.fold_const_symbol
 0.76% :     0.000004s :      1: substitution.graph_param_transform
62.71% :     0.000296s :     12: substitution.inline
 3.17% :     0.000015s :      2: substitution.inline_without_move
 1.63% :     0.000008s :     12: substitution.j_node_and_user_rematch
 2.06% :     0.000010s :      7: substitution.minmaximum_grad
 1.31% :     0.000006s :      5: substitution.partial_eliminate
 2.07% :     0.000010s :     12: substitution.remove_not_recompute_node
 5.59% :     0.000026s :     10: substitution.replace_applicator
 1.17% :     0.000006s :      7: substitution.replace_old_param
 1.29% :     0.000006s :      1: substitution.reshape_eliminate
 0.54% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.69% :     0.000008s :      2: substitution.tile_eliminate
 4.18% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.80% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 8.15% :     0.000038s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.071703     2
95.73% :     0.068641s :      1: type_inference.infer
 4.27% :     0.003063s :      1: type_inference.specialize
------[replace.]   0.000148    19
63.76% :     0.000095s :     12: replace.inline
36.24% :     0.000054s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000309    19
93.39% :     0.000289s :     12: match.inline
 6.61% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000361  2646
 1.61% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.40% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.46% :     0.000005s :     46: predicate.addn_check_dump
 1.71% :     0.000006s :     46: predicate.addn_zero_filter
 2.05% :     0.000007s :     46: predicate.arithmetic_simplify
 1.60% :     0.000006s :     46: predicate.cast_eliminate
 0.07% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.45% :     0.000005s :     46: predicate.compare_switch_simplify
 1.67% :     0.000006s :     46: predicate.depend_value_elim
 1.53% :     0.000006s :     46: predicate.dict_get_item_const_eliminator
 1.53% :     0.000006s :     46: predicate.dict_get_item_eliminator
 1.51% :     0.000005s :     46: predicate.dict_set_item_eliminator
 0.25% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.10% :     0.000000s :      1: predicate.elim_not_effective
 0.21% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.58% :     0.000006s :     46: predicate.environ_add_const_eliminate
 1.48% :     0.000005s :     46: predicate.environ_get_add_eliminate
 1.45% :     0.000005s :     46: predicate.environ_get_depend_swap
 1.52% :     0.000005s :     46: predicate.environ_get_eliminate
 1.49% :     0.000005s :     46: predicate.environ_get_set_eliminate
 0.02% :     0.000000s :      1: predicate.fold_const_symbol
 0.80% :     0.000003s :     19: predicate.get_grad_eliminate
 0.06% :     0.000000s :      1: predicate.graph_param_transform
 4.24% :     0.000015s :     67: predicate.inline
 1.90% :     0.000007s :     43: predicate.inline_without_move
 0.38% :     0.000001s :     19: predicate.j_node_and_user_rematch
 0.85% :     0.000003s :     19: predicate.less_batch_normalization
 1.79% :     0.000006s :     53: predicate.list_to_tuple_eliminator_
 1.87% :     0.000007s :     54: predicate.load_eliminater
 0.38% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.42% :     0.000012s :     95: predicate.loop_unroll_before_grad
 1.68% :     0.000006s :     47: predicate.make_slice_get_slice_eliminator
 1.46% :     0.000005s :     46: predicate.merge_addn
 1.54% :     0.000006s :     46: predicate.minmaximum_grad
 0.37% :     0.000001s :      1: predicate.mutable_eliminate
 0.11% :     0.000000s :      1: predicate.opt_reshape
 2.29% :     0.000008s :     54: predicate.partial_eliminate
 1.57% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.85% :     0.000007s :     46: predicate.reduce_eliminate
 1.81% :     0.000007s :     53: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.76% :     0.000010s :    100: predicate.replace_applicator
 1.02% :     0.000004s :     43: predicate.replace_old_param
 0.06% :     0.000000s :      1: predicate.reset_defer_inline
 1.88% :     0.000007s :     46: predicate.reshape_eliminate
 1.57% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.20% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.56% :     0.000006s :     46: predicate.same_eliminate
 0.48% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.17% :     0.000001s :      2: predicate.special_op_eliminate
 0.84% :     0.000003s :     19: predicate.specialize_transform
 1.73% :     0.000006s :     46: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000005s :     46: predicate.stack_unstack_eliminate
 0.06% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.86% :     0.000010s :     65: predicate.switch_defer_inline
 2.62% :     0.000009s :     65: predicate.switch_layer_defer_inline
 6.96% :     0.000025s :    161: predicate.switch_simplify
 1.55% :     0.000006s :     46: predicate.tile_eliminate
 1.48% :     0.000005s :     46: predicate.transpose_eliminate
 1.89% :     0.000007s :     46: predicate.tuple_list_convert_item_index_to_positive
 1.69% :     0.000006s :     46: predicate.tuple_list_get_item_depend_reorder
 3.12% :     0.000011s :     55: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000007s :     46: predicate.tuple_list_set_item_eliminator
 1.83% :     0.000007s :     53: predicate.tuple_to_list_eliminator_
 1.84% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000011s :     73: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000007s :     46: predicate.value_based_eliminate
 0.05% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.12% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.003315    46
75.83% :     0.002514s :     30: func_graph_cloner_run.FuncGraphClonerGraph
24.17% :     0.000801s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.105573    91
 0.04% :     0.000038s :      1: add_recomputation
 0.11% :     0.000121s :      1: auto_monad
 0.01% :     0.000016s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.24% :     0.000253s :      1: bootstrap
 0.02% :     0.000026s :      1: cconv
 0.01% :     0.000009s :      1: convert_after_rewriter
 0.02% :     0.000020s :      1: cse_after_recomputation
 0.01% :     0.000007s :      1: environ_conv
 0.27% :     0.000287s :      1: event_method
 0.00% :     0.000003s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000011s :      1: graph_reusing
17.23% :     0.018189s :      1: jit_opt_a
 0.11% :     0.000121s :      1: jit_opt_after_cconv
 0.04% :     0.000043s :      1: jit_opt_b
 0.40% :     0.000427s :      1: loop_unroll
 0.55% :     0.000578s :      1: mutable_eliminate
 2.54% :     0.002679s :     39: opt.transform.jit_opt_a
 0.03% :     0.000032s :      4: opt.transform.jit_opt_after_cconv
 0.01% :     0.000015s :      4: opt.transform.jit_opt_b
 0.01% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.01% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.02% :     0.000017s :      1: opt.transform.opt_after_jit_grad
 0.03% :     0.000027s :      4: opt.transform.symbol_engine_opt
 0.49% :     0.000517s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.00% :     0.000005s :      1: pre_auto_parallel
 0.04% :     0.000046s :      1: py_interpret_to_execute
 0.01% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000016s :      1: remove_dup_value
 7.71% :     0.008135s :      2: renormalize.infer
 1.63% :     0.001721s :      2: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.03% :     0.000035s :      1: rewriter_after_opt_a
 0.15% :     0.000153s :      1: rewriter_before_opt_a
 0.07% :     0.000069s :      1: symbol_engine_optimizer
 0.03% :     0.000031s :      1: task_emit
68.03% :     0.071816s :      1: type_inference
 0.04% :     0.000045s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:20.332.45 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c368550> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:20.340.99 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0197985, [33]
[bootstrap]: 0.0002914
[type_inference]: 0.00490397
[event_method]: 1.835e-05
[auto_monad]: 6.799e-05
[graph_reusing]: 5.96e-06
[pre_auto_parallel]: 2.37001e-06
[py_interpret_to_execute]: 3.776e-05
[rewriter_before_opt_a]: 7.326e-05
[expand_dump_flag]: 2.98e-06
[jit_opt_a]: 0.00553821, [2]
    [Cycle 1]: 0.00170475, [27]
        [switch_simplify]: 5.944e-05
        [loop_unroll]: 3.178e-05
        [a_1]: 0.00053697
        [with_stream_mark]: 1.985e-05
        [recompute_prepare]: 7.64002e-06
        [updatestate_depend_eliminate]: 3.5e-06
        [updatestate_assign_eliminate]: 2.96999e-06
        [updatestate_loads_eliminate]: 2.50002e-06
        [parameter_eliminate]: 1.98997e-06
        [specialize_transform]: 6.01e-06
        [updatestate_useless_node_eliminater]: 5.54e-06
        [accelerated_algorithm]: 6.06e-06
        [meta_shard_fg_expand]: 1.60001e-06
        [get_grad_eliminate_]: 5.67001e-06
        [merge_forward]: 3.38e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.49e-05
        [j_node_and_user_rematch]: 9.47001e-06
        [meta_fg_expand]: 1.96003e-06
        [replace_old_param]: 8.80001e-06
        [inline_without_move]: 5.77001e-06
        [renormalize]: 0.00071659
        [add_forward_monad_depend]: 7.13e-06
        [auto_monad_grad]: 2.99999e-06
        [auto_monad_eliminator]: 1.578e-05
        [cse]: 2.898e-05
        [replace_applicator]: 1.569e-05
    [Cycle 2]: 0.00037854, [27]
        [switch_simplify]: 6.44001e-06
        [loop_unroll]: 5.49e-06
        [a_1]: 0.00011799
        [with_stream_mark]: 1.207e-05
        [recompute_prepare]: 5.42001e-06
        [updatestate_depend_eliminate]: 2.86e-06
        [updatestate_assign_eliminate]: 2.41998e-06
        [updatestate_loads_eliminate]: 2.13998e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 5.51e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 5.99e-06
        [meta_shard_fg_expand]: 1.72001e-06
        [get_grad_eliminate_]: 5.10001e-06
        [merge_forward]: 2.93e-06
        [cell_reuse_recompute_pass]: 2.82002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.594e-05
        [j_node_and_user_rematch]: 9.37001e-06
        [meta_fg_expand]: 2.19001e-06
        [replace_old_param]: 8.82999e-06
        [inline_without_move]: 4.94998e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 1.64e-06
        [auto_monad_eliminator]: 6.39999e-06
        [cse]: 1.141e-05
        [replace_applicator]: 5.02999e-06
[py_interpret_to_execute_after_opt_a]: 1.355e-05
[rewriter_after_opt_a]: 3.777e-05
[convert_after_rewriter]: 7.46999e-06
[order_py_execute_after_rewriter]: 5.04e-06
[mutable_eliminate]: 0.00074424
[jit_opt_b]: 5.337e-05, [1]
    [Cycle 1]: 4.567e-05, [2]
        [frontend_op_eliminate]: 1.839e-05
        [inline_after_opt_a]: 1.584e-05
[cconv]: 2.614e-05
[loop_unroll]: 0.00049191
[jit_opt_after_cconv]: 0.0001564, [1]
    [Cycle 1]: 0.00014967, [11]
        [c_1]: 2.195e-05
        [parameter_eliminate]: 3.75e-06
        [updatestate_depend_eliminate]: 7.4e-06
        [updatestate_assign_eliminate]: 2.45002e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 2.346e-05
        [call_graph_tuple_transform]: 2.192e-05
        [tuple_list_get_item_eliminator]: 6.24999e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 6.29982e-07
        [switch_simplify]: 6.20002e-06
[remove_dup_value]: 1.333e-05
[partial_unused_args_eliminate]: 2.51e-06
[environ_conv]: 5.35001e-06
[add_recomputation]: 4.633e-05
[cse_after_recomputation]: 2.097e-05, [1]
    [Cycle 1]: 1.46e-05, [1]
        [cse]: 8.38999e-06
[auto_monad_reorder]: 1.489e-05
[get_jit_bprop_graph]: 1.97999e-06
[rewriter_after_jit_bprop_graph]: 7.66001e-06
[opt_after_jit_grad]: 0.00051469
[symbol_engine_optimizer]: 7.627e-05, [1]
    [Cycle 1]: 6.9e-05, [6]
        [build]: 4.03999e-06
        [elim_shapecalc]: 7.3e-06
        [elim_
TotalTime = 0.0195113, [33]
[bootstrap]: 0.00029063
[type_inference]: 0.00464884
[event_method]: 1.628e-05
[auto_monad]: 5.397e-05
[graph_reusing]: 5.99e-06
[pre_auto_parallel]: 2.24001e-06
[py_interpret_to_execute]: 4.403e-05
[rewriter_before_opt_a]: 6.778e-05
[expand_dump_flag]: 2.31e-06
[jit_opt_a]: 0.00558986, [2]
    [Cycle 1]: 0.00180775, [27]
        [switch_simplify]: 5.94e-05
        [loop_unroll]: 3.901e-05
        [a_1]: 0.00052834
        [with_stream_mark]: 2.023e-05
        [recompute_prepare]: 7.14001e-06
        [updatestate_depend_eliminate]: 3.53e-06
        [updatestate_assign_eliminate]: 2.72001e-06
        [updatestate_loads_eliminate]: 2.61999e-06
        [parameter_eliminate]: 1.74e-06
        [specialize_transform]: 6.04999e-06
        [updatestate_useless_node_eliminater]: 4.83001e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 2.56e-06
        [get_grad_eliminate_]: 5.49998e-06
        [merge_forward]: 3.70998e-06
        [cell_reuse_recompute_pass]: 1.81e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.435e-05
        [j_node_and_user_rematch]: 9.85002e-06
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 8.85999e-06
        [inline_without_move]: 5.19e-06
        [renormalize]: 0.00081542
        [add_forward_monad_depend]: 4.25999e-06
        [auto_monad_grad]: 1.22e-06
        [auto_monad_eliminator]: 7.95e-06
        [cse]: 1.44e-05
        [replace_applicator]: 5.555e-05
    [Cycle 2]: 0.00032955, [27]
        [switch_simplify]: 7.48e-06
        [loop_unroll]: 5.22999e-06
        [a_1]: 9.748e-05
        [with_stream_mark]: 8.03001e-06
        [recompute_prepare]: 4.77e-06
        [updatestate_depend_eliminate]: 2.36e-06
        [updatestate_assign_eliminate]: 2.01998e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 7.7e-07
        [specialize_transform]: 4.99998e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 1.18001e-06
        [get_grad_eliminate_]: 4.82e-06
        [merge_forward]: 2.46998e-06
        [cell_reuse_recompute_pass]: 2.82002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.144e-05
        [j_node_and_user_rematch]: 7.72998e-06
        [meta_fg_expand]: 1.64998e-06
        [replace_old_param]: 7.83001e-06
        [inline_without_move]: 4.52e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.55001e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 5.39998e-06
        [cse]: 1.122e-05
        [replace_applicator]: 5.42001e-06
[py_interpret_to_execute_after_opt_a]: 1.04e-05
[rewriter_after_opt_a]: 2.609e-05
[convert_after_rewriter]: 1.43e-05
[order_py_execute_after_rewriter]: 4.11001e-06
[mutable_eliminate]: 0.00068813
[jit_opt_b]: 5.429e-05, [1]
    [Cycle 1]: 4.609e-05, [2]
        [frontend_op_eliminate]: 1.788e-05
        [inline_after_opt_a]: 1.578e-05
[cconv]: 2.511e-05
[loop_unroll]: 0.0004929
[jit_opt_after_cconv]: 0.00015632, [1]
    [Cycle 1]: 0.00014885, [11]
        [c_1]: 2.276e-05
        [parameter_eliminate]: 3.59002e-06
        [updatestate_depend_eliminate]: 7.26999e-06
        [updatestate_assign_eliminate]: 2.86999e-06
        [updatestate_loads_eliminate]: 2.28998e-06
        [cse]: 2.357e-05
        [call_graph_tuple_transform]: 2.199e-05
        [tuple_list_get_item_eliminator]: 6.07999e-06
        [none_parameter_eliminate]: 1.24998e-06
        [renormalize]: 8.79983e-07
        [switch_simplify]: 5.46998e-06
[remove_dup_value]: 1.386e-05
[partial_unused_args_eliminate]: 2.02999e-06
[environ_conv]: 5.46e-06
[add_recomputation]: 4.638e-05
[cse_after_recomputation]: 2.043e-05, [1]
    [Cycle 1]: 1.446e-05, [1]
        [cse]: 8.2e-06
[auto_monad_reorder]: 1.538e-05
[get_jit_bprop_graph]: 2.08002e-06
[rewriter_after_jit_bprop_graph]: 7.13e-06
[opt_after_jit_grad]: 0.00051464
[symbol_engine_optimizer]: 7.703e-05, [1]
    [Cycle 1]: 6.955e-05, [6]
        [build]: 3.53e-06
        [elim_shapecalc]: 7.1e-06
        [elim_not_not_effective]: 1.388e-05
        [opt_reshape]: 6.01e-06
        [fold_const_symbol]: 9.49e-06
        [renormalize]: 4.89992e-07
[validate]: 3.692e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00635686
[execute]: 8.54e-06
Sums
    bootstrap                            :     0.000291s :  1.86%
    type_inference                       :     0.004904s : 31.24%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000068s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.24%
    rewriter_before_opt_a                :     0.000073s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.42%
    jit_opt_a.loop_unroll                :     0.000037s :  0.24%
    jit_opt_a.a_1                        :     0.000655s :  4.17%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000717s :  4.57%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.14%
    jit_opt_a.cse                        :     0.000040s :  0.26%
    jit_opt_a.replace_applicator         :     0.000021s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.09%
    rewriter_after_opt_a                 :     0.000038s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000744s :  4.74%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000026s :  0.17%
    loop_unroll                          :     0.000492s :  3.13%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.08%
    peffective]: 1.338e-05
        [opt_reshape]: 6.83998e-06
        [fold_const_symbol]: 9.09e-06
        [renormalize]: 5.10016e-07
[validate]: 3.561e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00635665
[execute]: 8.28001e-06
Sums
    bootstrap                            :     0.000291s :  1.88%
    type_inference                       :     0.004649s : 30.13%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000054s :  0.35%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.29%
    rewriter_before_opt_a                :     0.000068s :  0.44%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000067s :  0.43%
    jit_opt_a.loop_unroll                :     0.000044s :  0.29%
    jit_opt_a.a_1                        :     0.000626s :  4.06%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000816s :  5.28%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.09%
    jit_opt_a.cse                        :     0.000026s :  0.17%
    jit_opt_a.replace_applicator         :     0.000061s :  0.40%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000026s :  0.17%
    convert_after_rewriter               :     0.000014s :  0.09%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000688s :  4.46%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000025s :  0.16%
    loop_unroll                          :     0.000493s :  3.19%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
   artial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000046s :  0.30%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000515s :  3.28%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000037s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006357s : 40.49%
    execute                              :     0.000009s :  0.05%

 partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000046s :  0.30%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.05%
    opt_after_jit_grad                   :     0.000515s :  3.34%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006357s : 41.19%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000199    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 1.05% :     0.000002s :      2: substitution.fold_const_symbol
 3.24% :     0.000006s :      3: substitution.graph_param_transform
82.14% :     0.000163s :      4: substitution.inline
 2.04% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.15% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.41% :     0.000005s :      2: substitution.replace_old_param
 5.98% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004832     2
81.32% :     0.003929s :      1: type_inference.infer
18.68% :     0.000903s :      1: type_inference.specialize
------[replace.]   0.000069     6
64.90% :     0.000045s :      4: replace.inline
35.10% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000172     6
93.71% :     0.000161s :      4: match.inline
 6.29% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.46% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000002s :     10: predicate.addn_zero_filter
 2.02% :     0.000003s :     10: predicate.arithmetic_simplify
 1.52% :     0.000002s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.93% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.52% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.38% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.72% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.42% :     0.000001s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     10: predicate.environ_get_eliminate
 1.18% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.35% :     0.000000s :      3: predicate.graph_param_transform
 4.94% :     0.000006s :     22: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000002s :      6: predicate.less_batch_normalization
 1.68% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.51% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.20% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.28% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 0.93% :     0.000001s :     10: predicate.minmaximum_grad
 1.60% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.04% :     0.000003s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.84% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.26% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3:
Time group info:
------[substitution.]   0.000190    23
 0.95% :     0.000002s :      2: substitution.elim_not_effective
 0.91% :     0.000002s :      2: substitution.fold_const_symbol
 2.88% :     0.000005s :      3: substitution.graph_param_transform
82.36% :     0.000157s :      4: substitution.inline
 1.71% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.33% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.29% :     0.000004s :      2: substitution.replace_old_param
 6.56% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004594     2
80.27% :     0.003688s :      1: type_inference.infer
19.73% :     0.000907s :      1: type_inference.specialize
------[replace.]   0.000073     6
66.66% :     0.000049s :      4: replace.inline
33.34% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000166     6
93.15% :     0.000154s :      4: match.inline
 6.85% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.58% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.36% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.67% :     0.000002s :     10: predicate.addn_zero_filter
 2.42% :     0.000003s :     10: predicate.arithmetic_simplify
 1.41% :     0.000002s :     10: predicate.cast_eliminate
 0.59% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.50% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.45% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.38% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000002s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 4.47% :     0.000005s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.67% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.41% :     0.000003s :     15: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.28% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.19% :     0.000001s :     10: predicate.minmaximum_grad
 1.83% :     0.000002s :      3: predicate.mutable_eliminate
 0.57% :     0.000001s :      3: predicate.opt_reshape
 1.91% :     0.000002s :     15: predicate.partial_eliminate
 1.26% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.98% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.43% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.45% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.03% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.08% :     0.000001s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.69% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.26% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.76% :     0.000010s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.80% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.20% :     0.000003s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.92% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000632    12
48.79% :     0.000309s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.21% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.00% :     0.000291s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021521    76
 0.23% :     0.000050s :      1: add_recomputation
 0.33% :     0.000071s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.44% :     0.000311s :      1: bootstrap
 0.13% :     0.000029s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000023s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.75% :     0.005542s :      1: jit_opt_a
 0.74% :     0.000159s :      1: jit_opt_after_cconv
 0.26% :     0.000056s :      1: jit_opt_b
 2.33% :     0.000501s :      1: loop_unroll
 3.51% :     0.000755s :      1: mutable_eliminate
 4.11% :     0.000884s :     26: opt.transform.jit_opt_a
 0.24% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000028s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.43% :     0.000524s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000041s :      1: py_interpret_to_execute
 0.07% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.64% :     0.000354s :      1: renormalize.infer
 1.65% :     0.000356s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000041s :      1: rewriter_after_opt_a
 0.35% :     0.000076s :      1: rewriter_before_opt_a
 0.37% :     0.000079s :      1: symbol_engine_optimizer
29.62% :     0.006374s :       predicate.row_tensor_eliminate
 1.34% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.30% :     0.000009s :     53: predicate.switch_simplify
 1.28% :     0.000002s :     10: predicate.tile_eliminate
 1.27% :     0.000002s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.08% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.38% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000002s :     10: predicate.value_based_eliminate
 0.48% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000655    12
46.19% :     0.000303s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.97% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.85% :     0.000320s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021321    76
 0.23% :     0.000050s :      1: add_recomputation
 0.27% :     0.000058s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.41% :     0.000301s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.08% :     0.000017s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.09% :     0.000020s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.24% :     0.005594s :      1: jit_opt_a
 0.75% :     0.000159s :      1: jit_opt_after_cconv
 0.27% :     0.000057s :      1: jit_opt_b
 2.35% :     0.000501s :      1: loop_unroll
 3.28% :     0.000699s :      1: mutable_eliminate
 4.19% :     0.000894s :     26: opt.transform.jit_opt_a
 0.25% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000028s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.46% :     0.000524s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.22% :     0.000047s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 2.02% :     0.000431s :      1: renormalize.infer
 1.77% :     0.000377s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000029s :      1: rewriter_after_opt_a
 0.33% :     0.000070s :      1: rewriter_before_opt_a
 0.37% :     0.000079s :      1: symbol_engine_optimizer
29.90% :     0.006374s :      1: task_emit
22.85% :     0.004918s :      1: type_inference
 0.26% :     0.000056s :      1: validate

1: task_emit
21.85% :     0.004659s :      1: type_inference
 0.26% :     0.000055s :      1: validate


TotalTime = 0.0147011, [30]
[bootstrap]: 0.00031436
[type_inference]: 0.00482774
[event_method]: 1.868e-05
[auto_monad]: 6.357e-05
[graph_reusing]: 5.37999e-06
[pre_auto_parallel]: 2.58998e-06
[py_interpret_to_execute]: 3.775e-05
[rewriter_before_opt_a]: 7.249e-05
[expand_dump_flag]: 3.49001e-06
[jit_opt_a]: 0.00662513, [2]
    [Cycle 1]: 0.00214877, [27]
        [switch_simplify]: 6.68e-05
        [loop_unroll]: 3.032e-05
        [a_1]: 0.00061384
        [with_stream_mark]: 2.32e-05
        [recompute_prepare]: 8.43999e-06
        [updatestate_depend_eliminate]: 4.46002e-06
        [updatestate_assign_eliminate]: 3.18e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [parameter_eliminate]: 1.60001e-06
        [specialize_transform]: 6.37001e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 2.07e-05
        [meta_shard_fg_expand]: 2.29001e-06
        [get_grad_eliminate_]: 6.66e-06
        [merge_forward]: 4.28999e-06
        [cell_reuse_recompute_pass]: 9.79984e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.571e-05
        [j_node_and_user_rematch]: 1.015e-05
        [meta_fg_expand]: 2.65002e-06
        [replace_old_param]: 1.04e-05
        [inline_without_move]: 6.49999e-06
        [renormalize]: 0.00103871
        [add_forward_monad_depend]: 8.99998e-06
        [auto_monad_grad]: 3.39001e-06
        [auto_monad_eliminator]: 1.846e-05
        [cse]: 2.509e-05
        [replace_applicator]: 1.863e-05
    [Cycle 2]: 0.00037583, [27]
        [switch_simplify]: 6.28002e-06
        [loop_unroll]: 5.74e-06
        [a_1]: 9.921e-05
        [with_stream_mark]: 1.284e-05
        [recompute_prepare]: 6.58998e-06
        [updatestate_depend_eliminate]: 2.83998e-06
        [updatestate_assign_eliminate]: 2.76e-06
        [updatestate_loads_eliminate]: 2.28998e-06
        [parameter_eliminate]: 1.62001e-06
        [specialize_transform]: 5.15999e-06
        [updatestate_useless_node_eliminater]: 5.16998e-06
        [accelerated_algorithm]: 5.91998e-06
        [meta_shard_fg_expand]: 1.60001e-06
        [get_grad_eliminate_]: 4.56002e-06
        [merge_forward]: 3.93001e-06
        [cell_reuse_recompute_pass]: 2.92002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.735e-05
        [j_node_and_user_rematch]: 8.90001e-06
        [meta_fg_expand]: 2.05002e-06
        [replace_old_param]: 1.024e-05
        [inline_without_move]: 5.52001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 2.03997e-06
        [auto_monad_grad]: 1.52001e-06
        [auto_monad_eliminator]: 6.55002e-06
        [cse]: 1.438e-05
        [replace_applicator]: 6.09999e-06
[py_interpret_to_execute_after_opt_a]: 1.543e-05
[rewriter_after_opt_a]: 4.1e-05
[convert_after_rewriter]: 7.66999e-06
[order_py_execute_after_rewriter]: 5.62001e-06
[mutable_eliminate]: 0.00074754
[jit_opt_b]: 5.773e-05, [1]
    [Cycle 1]: 4.884e-05, [2]
        [frontend_op_eliminate]: 1.891e-05
        [inline_after_opt_a]: 1.585e-05
[cconv]: 3.301e-05
[loop_unroll]: 0.0005253
[jit_opt_after_cconv]: 0.00016926, [1]
    [Cycle 1]: 0.00016088, [11]
        [c_1]: 2.531e-05
        [parameter_eliminate]: 4.68001e-06
        [updatestate_depend_eliminate]: 8.26002e-06
        [updatestate_assign_eliminate]: 2.90998e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [cse]: 2.86e-05
        [call_graph_tuple_transform]: 2.213e-05
        [tuple_list_get_item_eliminator]: 5.35999e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 1.12999e-06
        [switch_simplify]: 5.50001e-06
[remove_dup_value]: 1.499e-05
[partial_unused_args_eliminate]: 2.14999e-06
[environ_conv]: 6.34001e-06
[add_recomputation]: 5.121e-05
[cse_after_recomputation]: 2.339e-05, [1]
    [Cycle 1]: 1.62e-05, [1]
        [cse]: 9.04998e-06
[auto_monad_reorder]: 1.633e-05
[get_jit_bprop_graph]: 2.68998e-06
[rewriter_after_jit_bprop_graph]: 7.82002e-06
[opt_after_jit_grad]: 0.00067891
[symbol_engine_optimizer]: 8.268e-05, [1]
    [Cycle 1]: 7.363e-05, [6]
        [build]: 4.18001e-06
        [elim_shapecalc]: 7.97e-06
        [elim_not_effective]: 1.548e-05
        [opt_reshape]: 6.29999e-06
        [fold_const_symbol]: 9.84999e-06
        [renormalize]: 8.60018e-07
[validate]: 4.044e-05
Sums
    bootstrap                            :     0.000314s :  3.16%
    type_inference                       :     0.004828s : 48.56%
    event_method                         :     0.000019s :  0.19%
    auto_monad                           :     0.000064s :  0.64%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000038s :  0.38%
    rewriter_before_opt_a                :     0.000072s :  0.73%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000073s :  0.74%
    jit_opt_a.loop_unroll                :     0.000036s :  0.36%
    jit_opt_a.a_1                        :     0.000713s :  7.17%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.36%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000027s :  0.27%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000021s :  0.21%
    jit_opt_a.inline_without_move        :     0.000012s :  0.12%
    jit_opt_a.renormalize                :     0.001039s : 10.45%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.11%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.25%
    jit_opt_a.cse                        :     0.000039s :  0.40%
    jit_opt_a.replace_applicator         :     0.000025s :  0.25%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.16%
    rewriter_after_opt_a                 :     0.000041s :  0.41%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000748s :  7.52%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.16%
    cconv                                :     0.000033s :  0.33%
    loop_unroll                          :     0.000525s :  5.28%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000029s :  0.29%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000015s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000051s :  0.52%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.16%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.08%
    opt_after_jit_grad                   :     0.000679s :  6.83%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.41%


Time group info:
------[substitution.]   0.000266    23
 1.02% :     0.000003s :      2: substitution.elim_not_effective
 0.50% :     0.000001s :      2: substitution.fold_const_symbol
 2.15% :     0.000006s :      3: substitution.graph_param_transform
85.26% :     0.000226s :      4: substitution.inline
 1.69% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.16% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.26% :     0.000006s :      2: substitution.replace_old_param
 4.97% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004756     2
79.49% :     0.003781s :      1: type_inference.infer
20.51% :     0.000975s :      1: type_inference.specialize
------[replace.]   0.000083     6
67.78% :     0.000056s :      4: replace.inline
32.22% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000235     6
94.84% :     0.000223s :      4: match.inline
 5.16% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.16% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.85% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 2.01% :     0.000003s :     10: predicate.addn_zero_filter
 2.60% :     0.000003s :     10: predicate.arithmetic_simplify
 1.12% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.41% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.17% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.04% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.57% :     0.000001s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.02% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.28% :     0.000002s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.87% :     0.000007s :     22: predicate.inline
 1.09% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.78% :     0.000002s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.63% :     0.000002s :     15: predicate.load_eliminater
 1.87% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 0.96% :     0.000001s :     10: predicate.minmaximum_grad
 1.70% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.00% :     0.000003s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.82% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.08% :     0.000001s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 0.86% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.54% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 1.88% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.96% :     0.000010s :     53: predicate.switch_simplify
 1.08% :     0.000001s :     10: predicate.tile_eliminate
 1.04% :     0.000001s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.70% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.45% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.47% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000708    12
44.54% :     0.000315s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.06% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.41% :     0.000343s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016771    72
 0.33% :     0.000055s :      1: add_recomputation
 0.40% :     0.000067s :      1: auto_monad
 0.12% :     0.000020s :      1: auto_monad_reorder
 1.96% :     0.000329s :      1: bootstrap
 0.22% :     0.000037s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000010s :      1: environ_conv
 0.13% :     0.000022s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
39.53% :     0.006630s :      1: jit_opt_a
 1.03% :     0.000172s :      1: jit_opt_after_cconv
 0.36% :     0.000060s :      1: jit_opt_b
 3.20% :     0.000537s :      1: loop_unroll
 4.54% :     0.000761s :      1: mutable_eliminate
 5.80% :     0.000973s :     26: opt.transform.jit_opt_a
 0.32% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000028s :      4: opt.transform.jit_opt_b
 0.09% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000036s :      4: opt.transform.symbol_engine_opt
 4.12% :     0.000691s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000041s :      1: py_interpret_to_execute
 0.11% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 3.28% :     0.000550s :      1: renormalize.infer
 2.85% :     0.000478s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000045s :      1: rewriter_after_opt_a
 0.45% :     0.000076s :      1: rewriter_before_opt_a
 0.51% :     0.000086s :      1: symbol_engine_optimizer
28.85% :     0.004839s :      1: type_inference


TotalTime = 0.0203487, [33]
[bootstrap]: 0.00028972
[type_inference]: 0.00488813
[event_method]: 1.891e-05
[auto_monad]: 6.771e-05
[graph_reusing]: 5.86003e-06
[pre_auto_parallel]: 2.53998e-06
[py_interpret_to_execute]: 4.698e-05
[rewriter_before_opt_a]: 7.834e-05
[expand_dump_flag]: 4.33999e-06
[jit_opt_a]: 0.00584476, [2]
    [Cycle 1]: 0.00191191, [27]
        [switch_simplify]: 5.543e-05
        [loop_unroll]: 2.949e-05
        [a_1]: 0.00056438
        [with_stream_mark]: 1.841e-05
        [recompute_prepare]: 8.57e-06
        [updatestate_depend_eliminate]: 3.39001e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.97002e-06
        [parameter_eliminate]: 2.09e-06
        [specialize_transform]: 6.52001e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 6.19001e-06
        [meta_shard_fg_expand]: 2.17999e-06
        [get_grad_eliminate_]: 5.46e-06
        [merge_forward]: 3.73999e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.613e-05
        [j_node_and_user_rematch]: 9.48002e-06
        [meta_fg_expand]: 2.31998e-06
        [replace_old_param]: 9.67001e-06
        [inline_without_move]: 5.71e-06
        [renormalize]: 0.0009022
        [add_forward_monad_depend]: 6.93998e-06
        [auto_monad_grad]: 2.66e-06
        [auto_monad_eliminator]: 1.478e-05
        [cse]: 2.777e-05
        [replace_applicator]: 1.512e-05
    [Cycle 2]: 0.00034534, [27]
        [switch_simplify]: 5.69e-06
        [loop_unroll]: 5.19998e-06
        [a_1]: 9.741e-05
        [with_stream_mark]: 1.187e-05
        [recompute_prepare]: 5.81e-06
        [updatestate_depend_eliminate]: 2.76999e-06
        [updatestate_assign_eliminate]: 2.39999e-06
        [updatestate_loads_eliminate]: 2.27001e-06
        [parameter_eliminate]: 1.35999e-06
        [specialize_transform]: 5.05001e-06
        [updatestate_useless_node_eliminater]: 4.96002e-06
        [accelerated_algorithm]: 6.07001e-06
        [meta_shard_fg_expand]: 1.67001e-06
        [get_grad_eliminate_]: 4.67998e-06
        [merge_forward]: 3.01001e-06
        [cell_reuse_recompute_pass]: 2.22001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.401e-05
        [j_node_and_user_rematch]: 7.53e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 8.54e-06
        [inline_without_move]: 4.84998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.26002e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 6.61e-06
        [cse]: 1.118e-05
        [replace_applicator]: 5.65001e-06
[py_interpret_to_execute_after_opt_a]: 1.121e-05
[rewriter_after_opt_a]: 3.793e-05
[convert_after_rewriter]: 7.66999e-06
[order_py_execute_after_rewriter]: 5.10001e-06
[mutable_eliminate]: 0.00064228
[jit_opt_b]: 5.533e-05, [1]
    [Cycle 1]: 4.62e-05, [2]
        [frontend_op_eliminate]: 1.776e-05
        [inline_after_opt_a]: 1.652e-05
[cconv]: 3.029e-05
[loop_unroll]: 0.00045566
[jit_opt_after_cconv]: 0.00015269, [1]
    [Cycle 1]: 0.00014537, [11]
        [c_1]: 2.199e-05
        [parameter_eliminate]: 3.43999e-06
        [updatestate_depend_eliminate]: 5.68002e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [cse]: 2.373e-05
        [call_graph_tuple_transform]: 2.031e-05
        [tuple_list_get_item_eliminator]: 6.26e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 6.60017e-07
        [switch_simplify]: 5.51998e-06
[remove_dup_value]: 1.274e-05
[partial_unused_args_eliminate]: 2.13998e-06
[environ_conv]: 5.57001e-06
[add_recomputation]: 4.94e-05
[cse_after_recomputation]: 2.115e-05, [1]
    [Cycle 1]: 1.52e-05, [1]
        [cse]: 9.17999e-06
[auto_monad_reorder]: 1.453e-05
[get_jit_bprop_graph]: 2.07001e-06
[rewriter_after_jit_bprop_graph]: 5.70001e-06
[opt_after_jit_grad]: 0.00048638
[symbol_engine_optimizer]: 7.826e-05, [1]
    [Cycle 1]: 7.003e-05, [6]
        [build]: 4.12003e-06
        [elim_shapecalc]: 7.69002e-06
        [elim_not_effective]: 1.52e-05
        [opt_reshape]: 5.66e-06
        [fold_const_symbol]: 9.37001e-06
        [renormalize]: 4.10015e-07
[validate]: 3.864e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00676136
[execute]: 7.05998e-06
Sums
    bootstrap                            :     0.000290s :  1.80%
    type_inference                       :     0.004888s : 30.33%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000068s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000047s :  0.29%
    rewriter_before_opt_a                :     0.000078s :  0.49%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000061s :  0.38%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000662s :  4.11%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000018s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000902s :  5.60%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.13%
    jit_opt_a.cse                        :     0.000039s :  0.24%
    jit_opt_a.replace_applicator         :     0.000021s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000038s :  0.24%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000642s :  3.99%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.10%
    cconv                                :     0.000030s :  0.19%
    loop_unroll                          :     0.000456s :  2.83%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000049s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000486s :  3.02%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000039s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006761s : 41.96%
    execute                              :     0.000007s :  0.04%


Time group info:
------[substitution.]   0.000227    23
 1.35% :     0.000003s :      2: substitution.elim_not_effective
 0.66% :     0.000002s :      2: substitution.fold_const_symbol
 2.68% :     0.000006s :      3: substitution.graph_param_transform
83.91% :     0.000190s :      4: substitution.inline
 1.17% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.27% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.59% :     0.000006s :      2: substitution.replace_old_param
 5.37% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004816     2
79.40% :     0.003824s :      1: type_inference.infer
20.60% :     0.000992s :      1: type_inference.specialize
------[replace.]   0.000075     6
65.22% :     0.000049s :      4: replace.inline
34.78% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000198     6
94.50% :     0.000188s :      4: match.inline
 5.50% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 1.99% :     0.000002s :     10: predicate.arithmetic_simplify
 1.52% :     0.000002s :     10: predicate.cast_eliminate
 0.65% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.74% :     0.000002s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.15% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.49% :     0.000001s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.40% :     0.000002s :     10: predicate.environ_get_eliminate
 1.28% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 4.89% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :      6: predicate.less_batch_normalization
 1.37% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.62% :     0.000002s :     15: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 1.65% :     0.000002s :      3: predicate.mutable_eliminate
 0.39% :     0.000000s :      3: predicate.opt_reshape
 2.08% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.37% :     0.000002s :     10: predicate.reduce_eliminate
 1.64% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.91% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.70% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.97% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 2.23% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.30% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.64% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.42% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000697    12
46.37% :     0.000323s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.03% :     0.000035s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.60% :     0.000339s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022241    76
 0.24% :     0.000052s :      1: add_recomputation
 0.32% :     0.000071s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.38% :     0.000306s :      1: bootstrap
 0.15% :     0.000033s :      1: cconv
 0.06% :     0.000014s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000007s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.30% :     0.005849s :      1: jit_opt_a
 0.70% :     0.000155s :      1: jit_opt_after_cconv
 0.26% :     0.000058s :      1: jit_opt_b
 2.09% :     0.000464s :      1: loop_unroll
 2.94% :     0.000653s :      1: mutable_eliminate
 3.97% :     0.000882s :     26: opt.transform.jit_opt_a
 0.23% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.23% :     0.000495s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000050s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.10% :     0.000467s :      1: renormalize.infer
 1.92% :     0.000427s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000041s :      1: rewriter_after_opt_a
 0.36% :     0.000081s :      1: rewriter_before_opt_a
 0.37% :     0.000081s :      1: symbol_engine_optimizer
30.46% :     0.006774s :      1: task_emit
22.05% :     0.004904s :      1: type_inference
 0.27% :     0.000059s :      1: validate


TotalTime = 0.0192724, [33]
[bootstrap]: 0.00029458
[type_inference]: 0.00480648
[event_method]: 1.864e-05
[auto_monad]: 5.908e-05
[graph_reusing]: 5.27999e-06
[pre_auto_parallel]: 2.12999e-06
[py_interpret_to_execute]: 3.532e-05
[rewriter_before_opt_a]: 7.085e-05
[expand_dump_flag]: 2.76999e-06
[jit_opt_a]: 0.00526316, [2]
    [Cycle 1]: 0.00159517, [27]
        [switch_simplify]: 5.835e-05
        [loop_unroll]: 3.157e-05
        [a_1]: 0.00053431
        [with_stream_mark]: 1.579e-05
        [recompute_prepare]: 6.98e-06
        [updatestate_depend_eliminate]: 3.51001e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [parameter_eliminate]: 1.28002e-06
        [specialize_transform]: 5.69e-06
        [updatestate_useless_node_eliminater]: 5.22999e-06
        [accelerated_algorithm]: 5.89999e-06
        [meta_shard_fg_expand]: 2.46e-06
        [get_grad_eliminate_]: 5.52999e-06
        [merge_forward]: 3.28998e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.167e-05
        [j_node_and_user_rematch]: 9.34e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 8.43001e-06
        [inline_without_move]: 5.49998e-06
        [renormalize]: 0.00064911
        [add_forward_monad_depend]: 5.04e-06
        [auto_monad_grad]: 1.96e-06
        [auto_monad_eliminator]: 1.365e-05
        [cse]: 2.051e-05
        [replace_applicator]: 1.316e-05
    [Cycle 2]: 0.000351, [27]
        [switch_simplify]: 5.78002e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 9.401e-05
        [with_stream_mark]: 9.37999e-06
        [recompute_prepare]: 5.29e-06
        [updatestate_depend_eliminate]: 2.43002e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 2.27999e-06
        [parameter_eliminate]: 1.32e-06
        [specialize_transform]: 5.65001e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 6.75002e-06
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 5.16002e-06
        [merge_forward]: 2.46998e-06
        [cell_reuse_recompute_pass]: 2.42001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.422e-05
        [j_node_and_user_rematch]: 9.31998e-06
        [meta_fg_expand]: 2.25002e-06
        [replace_old_param]: 1.569e-05
        [inline_without_move]: 5.49e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 5.75001e-06
        [cse]: 1.279e-05
        [replace_applicator]: 5.52001e-06
[py_interpret_to_execute_after_opt_a]: 1.406e-05
[rewriter_after_opt_a]: 5.187e-05
[convert_after_rewriter]: 8.33001e-06
[order_py_execute_after_rewriter]: 5.83002e-06
[mutable_eliminate]: 0.00062736
[jit_opt_b]: 5.365e-05, [1]
    [Cycle 1]: 4.589e-05, [2]
        [frontend_op_eliminate]: 1.787e-05
        [inline_after_opt_a]: 1.528e-05
[cconv]: 2.411e-05
[loop_unroll]: 0.00045004
[jit_opt_after_cconv]: 0.00014783, [1]
    [Cycle 1]: 0.00014098, [11]
        [c_1]: 2.182e-05
        [parameter_eliminate]: 2.79999e-06
        [updatestate_depend_eliminate]: 5.11997e-06
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [cse]: 2.191e-05
        [call_graph_tuple_transform]: 1.942e-05
        [tuple_list_get_item_eliminator]: 5.79999e-06
        [none_parameter_eliminate]: 1.32999e-06
        [renormalize]: 7.09988e-07
        [switch_simplify]: 5.51e-06
[remove_dup_value]: 1.191e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 4.97999e-06
[add_recomputation]: 4.666e-05
[cse_after_recomputation]: 2.061e-05, [1]
    [Cycle 1]: 1.509e-05, [1]
        [cse]: 9.39e-06
[auto_monad_reorder]: 1.534e-05
[get_jit_bprop_graph]: 1.42999e-06
[rewriter_after_jit_bprop_graph]: 6.30002e-06
[opt_after_jit_grad]: 0.00050629
[symbol_engine_optimizer]: 7.383e-05, [1]
    [Cycle 1]: 6.676e-05, [6]
        [build]: 3.49001e-06
        [elim_shapecalc]: 7.58999e-06
        [elim_not_effective]: 1.206e-05
        [opt_reshape]: 5.82001e-06
        [fold_const_symbol]: 8.89e-06
        [renormalize]: 6.69999e-07
[validate]: 3.56e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00639209
[execute]: 7.85e-06
Sums
    bootstrap                            :     0.000295s :  1.92%
    type_inference                       :     0.004806s : 31.34%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000059s :  0.39%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000035s :  0.23%
    rewriter_before_opt_a                :     0.000071s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.42%
    jit_opt_a.loop_unroll                :     0.000036s :  0.24%
    jit_opt_a.a_1                        :     0.000628s :  4.10%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000024s :  0.16%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000649s :  4.23%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000033s :  0.22%
    jit_opt_a.replace_applicator         :     0.000019s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.09%
    rewriter_after_opt_a                 :     0.000052s :  0.34%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000627s :  4.09%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000024s :  0.16%
    loop_unroll                          :     0.000450s :  2.93%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000012s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000047s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000506s :  3.30%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006392s : 41.68%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000208    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.80% :     0.000006s :      3: substitution.graph_param_transform
80.65% :     0.000168s :      4: substitution.inline
 1.49% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.24% :     0.000005s :      4: substitution.remove_not_recompute_node
 5.53% :     0.000012s :      2: substitution.replace_old_param
 5.74% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004739     2
81.11% :     0.003844s :      1: type_inference.infer
18.89% :     0.000895s :      1: type_inference.specialize
------[replace.]   0.000068     6
66.26% :     0.000045s :      4: replace.inline
33.74% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000176     6
93.93% :     0.000165s :      4: match.inline
 6.07% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.18% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.60% :     0.000002s :     10: predicate.addn_zero_filter
 2.05% :     0.000002s :     10: predicate.arithmetic_simplify
 1.55% :     0.000002s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.48% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.23% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.28% :     0.000006s :     22: predicate.inline
 1.04% :     0.000001s :      6: predicate.inline_without_move
 0.44% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.89% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.94% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.38% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.72% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.74% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000002s :     10: predicate.tile_eliminate
 1.28% :     0.000002s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.13% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 2.03% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.12% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.91% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000568    12
48.09% :     0.000273s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.62% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.28% :     0.000257s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020885    76
 0.24% :     0.000050s :      1: add_recomputation
 0.30% :     0.000062s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.49% :     0.000311s :      1: bootstrap
 0.13% :     0.000027s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
25.22% :     0.005267s :      1: jit_opt_a
 0.72% :     0.000151s :      1: jit_opt_after_cconv
 0.27% :     0.000056s :      1: jit_opt_b
 2.20% :     0.000458s :      1: loop_unroll
 3.05% :     0.000638s :      1: mutable_eliminate
 4.11% :     0.000858s :     26: opt.transform.jit_opt_a
 0.23% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.47% :     0.000516s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000038s :      1: py_interpret_to_execute
 0.08% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000014s :      1: remove_dup_value
 1.59% :     0.000331s :      1: renormalize.infer
 1.48% :     0.000310s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000056s :      1: rewriter_after_opt_a
 0.35% :     0.000074s :      1: rewriter_before_opt_a
 0.37% :     0.000077s :      1: symbol_engine_optimizer
30.66% :     0.006404s :      1: task_emit
23.09% :     0.004821s :      1: type_inference
 0.27% :     0.000056s :      1: validate


TotalTime = 0.0140077, [30]
[bootstrap]: 0.00032942
[type_inference]: 0.00466453
[event_method]: 1.684e-05
[auto_monad]: 6.914e-05
[graph_reusing]: 6.11998e-06
[pre_auto_parallel]: 2.27999e-06
[py_interpret_to_execute]: 3.686e-05
[rewriter_before_opt_a]: 7.056e-05
[expand_dump_flag]: 3.31001e-06
[jit_opt_a]: 0.00612379, [2]
    [Cycle 1]: 0.00191145, [27]
        [switch_simplify]: 6.219e-05
        [loop_unroll]: 3.038e-05
        [a_1]: 0.00056829
        [with_stream_mark]: 2.01e-05
        [recompute_prepare]: 8.08999e-06
        [updatestate_depend_eliminate]: 3.53999e-06
        [updatestate_assign_eliminate]: 3.3e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 1.92001e-06
        [specialize_transform]: 6.31998e-06
        [updatestate_useless_node_eliminater]: 5.50001e-06
        [accelerated_algorithm]: 5.74e-06
        [meta_shard_fg_expand]: 2.14999e-06
        [get_grad_eliminate_]: 5.87001e-06
        [merge_forward]: 4.02998e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.647e-05
        [j_node_and_user_rematch]: 9.79e-06
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 9.66e-06
        [inline_without_move]: 5.59e-06
        [renormalize]: 0.00088971
        [add_forward_monad_depend]: 7.63001e-06
        [auto_monad_grad]: 2.96001e-06
        [auto_monad_eliminator]: 1.501e-05
        [cse]: 2.083e-05
        [replace_applicator]: 1.605e-05
    [Cycle 2]: 0.00034554, [27]
        [switch_simplify]: 6.17999e-06
        [loop_unroll]: 4.82998e-06
        [a_1]: 9.208e-05
        [with_stream_mark]: 1.111e-05
        [recompute_prepare]: 5.12e-06
        [updatestate_depend_eliminate]: 2.58003e-06
        [updatestate_assign_eliminate]: 2.84999e-06
        [updatestate_loads_eliminate]: 1.77999e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 4.90999e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.57999e-06
        [meta_shard_fg_expand]: 1.57001e-06
        [get_grad_eliminate_]: 5.05999e-06
        [merge_forward]: 2.39999e-06
        [cell_reuse_recompute_pass]: 3.01999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.605e-05
        [j_node_and_user_rematch]: 8.76002e-06
        [meta_fg_expand]: 1.42999e-06
        [replace_old_param]: 8.12998e-06
        [inline_without_move]: 4.83001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.42999e-06
        [auto_monad_grad]: 1.09e-06
        [auto_monad_eliminator]: 6.09999e-06
        [cse]: 1.048e-05
        [replace_applicator]: 6.24001e-06
[py_interpret_to_execute_after_opt_a]: 1.501e-05
[rewriter_after_opt_a]: 3.657e-05
[convert_after_rewriter]: 7.97e-06
[order_py_execute_after_rewriter]: 5.17999e-06
[mutable_eliminate]: 0.00080641
[jit_opt_b]: 6.104e-05, [1]
    [Cycle 1]: 5.177e-05, [2]
        [frontend_op_eliminate]: 2.035e-05
        [inline_after_opt_a]: 1.866e-05
[cconv]: 3.064e-05
[loop_unroll]: 0.00054724
[jit_opt_after_cconv]: 0.00017165, [1]
    [Cycle 1]: 0.00016371, [11]
        [c_1]: 2.228e-05
        [parameter_eliminate]: 4.68001e-06
        [updatestate_depend_eliminate]: 7.71001e-06
        [updatestate_assign_eliminate]: 2.84001e-06
        [updatestate_loads_eliminate]: 2.52001e-06
        [cse]: 2.941e-05
        [call_graph_tuple_transform]: 2.531e-05
        [tuple_list_get_item_eliminator]: 6.11e-06
        [none_parameter_eliminate]: 1.54998e-06
        [renormalize]: 1.10001e-06
        [switch_simplify]: 6.34001e-06
[remove_dup_value]: 1.486e-05
[partial_unused_args_eliminate]: 2.53998e-06
[environ_conv]: 5.47001e-06
[add_recomputation]: 5.433e-05
[cse_after_recomputation]: 2.524e-05, [1]
    [Cycle 1]: 1.819e-05, [1]
        [cse]: 1.182e-05
[auto_monad_reorder]: 1.564e-05
[get_jit_bprop_graph]: 2.07001e-06
[rewriter_after_jit_bprop_graph]: 8.43999e-06
[opt_after_jit_grad]: 0.00055327
[symbol_engine_optimizer]: 8.593e-05, [1]
    [Cycle 1]: 7.788e-05, [6]
        [build]: 5.16998e-06
        [elim_shapecalc]: 8.70001e-06
        [elim_not_effective]: 1.627e-05
        [opt_reshape]: 7.03998e-06
        [fold_const_symbol]: 9.99001e-06
        [renormalize]: 9.99979e-07
[validate]: 4.549e-05
Sums
    bootstrap                            :     0.000329s :  3.47%
    type_inference                       :     0.004665s : 49.07%
    event_method                         :     0.000017s :  0.18%
    auto_monad                           :     0.000069s :  0.73%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.39%
    rewriter_before_opt_a                :     0.000071s :  0.74%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000068s :  0.72%
    jit_opt_a.loop_unroll                :     0.000035s :  0.37%
    jit_opt_a.a_1                        :     0.000660s :  6.95%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000890s :  9.36%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.22%
    jit_opt_a.cse                        :     0.000031s :  0.33%
    jit_opt_a.replace_applicator         :     0.000022s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.16%
    rewriter_after_opt_a                 :     0.000037s :  0.38%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000806s :  8.48%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.20%
    cconv                                :     0.000031s :  0.32%
    loop_unroll                          :     0.000547s :  5.76%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000029s :  0.31%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.27%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000054s :  0.57%
    cse_after_recomputation.cse          :     0.000012s :  0.12%
    auto_monad_reorder                   :     0.000016s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.09%
    opt_after_jit_grad                   :     0.000553s :  5.82%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000045s :  0.48%


Time group info:
------[substitution.]   0.000220    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 3.13% :     0.000007s :      3: substitution.graph_param_transform
82.20% :     0.000181s :      4: substitution.inline
 1.66% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.70% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.21% :     0.000005s :      2: substitution.replace_old_param
 6.52% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004595     2
79.77% :     0.003665s :      1: type_inference.infer
20.23% :     0.000930s :      1: type_inference.specialize
------[replace.]   0.000075     6
65.12% :     0.000049s :      4: replace.inline
34.88% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     6
93.03% :     0.000179s :      4: match.inline
 6.97% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.19% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 2.05% :     0.000002s :     10: predicate.arithmetic_simplify
 1.50% :     0.000002s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.46% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.49% :     0.000001s :      3: predicate.elim_not_effective
 0.74% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.35% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.14% :     0.000001s :      6: predicate.get_grad_eliminate
 0.43% :     0.000001s :      3: predicate.graph_param_transform
 5.29% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.92% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.88% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.81% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.83% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.45% :     0.000002s :      3: predicate.mutable_eliminate
 0.60% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.47% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.50% :     0.000001s :      3: predicate.reset_defer_inline
 1.72% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 1.18% :     0.000001s :      6: predicate.specialize_transform
 1.18% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.81% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.06% :     0.000001s :     10: predicate.transpose_eliminate
 1.55% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.98% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.35% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.02% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000665    12
46.57% :     0.000310s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.42% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.01% :     0.000319s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015857    72
 0.36% :     0.000057s :      1: add_recomputation
 0.46% :     0.000073s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 2.16% :     0.000343s :      1: bootstrap
 0.21% :     0.000033s :      1: cconv
 0.07% :     0.000011s :      1: convert_after_rewriter
 0.18% :     0.000028s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.13% :     0.000020s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.65% :     0.006128s :      1: jit_opt_a
 1.10% :     0.000175s :      1: jit_opt_after_cconv
 0.40% :     0.000064s :      1: jit_opt_b
 3.50% :     0.000555s :      1: loop_unroll
 5.16% :     0.000819s :      1: mutable_eliminate
 5.62% :     0.000892s :     26: opt.transform.jit_opt_a
 0.36% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000031s :      4: opt.transform.jit_opt_b
 0.10% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000038s :      4: opt.transform.symbol_engine_opt
 3.55% :     0.000564s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000040s :      1: py_interpret_to_execute
 0.11% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 2.81% :     0.000446s :      1: renormalize.infer
 2.74% :     0.000434s :      1: renormalize.specialize
 0.07% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000044s :      1: rewriter_after_opt_a
 0.47% :     0.000074s :      1: rewriter_before_opt_a
 0.56% :     0.000089s :      1: symbol_engine_optimizer
29.50% :     0.004677s :      1: type_inference


TotalTime = 0.0190363, [33]
[bootstrap]: 0.00028838
[type_inference]: 0.00463188
[event_method]: 1.812e-05
[auto_monad]: 6.297e-05
[graph_reusing]: 5.54e-06
[pre_auto_parallel]: 2.44001e-06
[py_interpret_to_execute]: 3.641e-05
[rewriter_before_opt_a]: 7.211e-05
[expand_dump_flag]: 3.33998e-06
[jit_opt_a]: 0.0053825, [2]
    [Cycle 1]: 0.00154016, [27]
        [switch_simplify]: 5.791e-05
        [loop_unroll]: 3.005e-05
        [a_1]: 0.00052708
        [with_stream_mark]: 1.97e-05
        [recompute_prepare]: 8.45001e-06
        [updatestate_depend_eliminate]: 3.75e-06
        [updatestate_assign_eliminate]: 3.11001e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 1.71e-06
        [specialize_transform]: 6.13998e-06
        [updatestate_useless_node_eliminater]: 5.30999e-06
        [accelerated_algorithm]: 6.38e-06
        [meta_shard_fg_expand]: 2.66e-06
        [get_grad_eliminate_]: 5.70001e-06
        [merge_forward]: 3.18e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.409e-05
        [j_node_and_user_rematch]: 1.004e-05
        [meta_fg_expand]: 2.54999e-06
        [replace_old_param]: 8.39998e-06
        [inline_without_move]: 5.22999e-06
        [renormalize]: 0.00058232
        [add_forward_monad_depend]: 5.04e-06
        [auto_monad_grad]: 1.35001e-06
        [auto_monad_eliminator]: 1.019e-05
        [cse]: 2.578e-05
        [replace_applicator]: 1.241e-05
    [Cycle 2]: 0.00032818, [27]
        [switch_simplify]: 5.49e-06
        [loop_unroll]: 4.95001e-06
        [a_1]: 8.999e-05
        [with_stream_mark]: 1.059e-05
        [recompute_prepare]: 5.61998e-06
        [updatestate_depend_eliminate]: 2.53e-06
        [updatestate_assign_eliminate]: 2.03997e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [parameter_eliminate]: 8.79983e-07
        [specialize_transform]: 5.05999e-06
        [updatestate_useless_node_eliminater]: 5.10999e-06
        [accelerated_algorithm]: 5.20001e-06
        [meta_shard_fg_expand]: 1.54e-06
        [get_grad_eliminate_]: 4.55001e-06
        [merge_forward]: 2.40002e-06
        [cell_reuse_recompute_pass]: 1.57999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.249e-05
        [j_node_and_user_rematch]: 8.27e-06
        [meta_fg_expand]: 1.52999e-06
        [replace_old_param]: 1.312e-05
        [inline_without_move]: 5.52001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 8.80013e-07
        [auto_monad_eliminator]: 5.27001e-06
        [cse]: 1.089e-05
        [replace_applicator]: 5.32001e-06
[py_interpret_to_execute_after_opt_a]: 1.149e-05
[rewriter_after_opt_a]: 3.635e-05
[convert_after_rewriter]: 6.45002e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00061216
[jit_opt_b]: 5.146e-05, [1]
    [Cycle 1]: 4.393e-05, [2]
        [frontend_op_eliminate]: 1.729e-05
        [inline_after_opt_a]: 1.511e-05
[cconv]: 2.56e-05
[loop_unroll]: 0.00047874
[jit_opt_after_cconv]: 0.00014207, [1]
    [Cycle 1]: 0.0001353, [11]
        [c_1]: 2.077e-05
        [parameter_eliminate]: 2.34001e-06
        [updatestate_depend_eliminate]: 6.02999e-06
        [updatestate_assign_eliminate]: 2.53998e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 1.84e-05
        [call_graph_tuple_transform]: 1.9e-05
        [tuple_list_get_item_eliminator]: 5.55001e-06
        [none_parameter_eliminate]: 1.54998e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 5.44e-06
[remove_dup_value]: 1.274e-05
[partial_unused_args_eliminate]: 2.32999e-06
[environ_conv]: 4.89998e-06
[add_recomputation]: 4.67e-05
[cse_after_recomputation]: 1.927e-05, [1]
    [Cycle 1]: 1.417e-05, [1]
        [cse]: 8.80001e-06
[auto_monad_reorder]: 1.544e-05
[get_jit_bprop_graph]: 1.87999e-06
[rewriter_after_jit_bprop_graph]: 5.37999e-06
[opt_after_jit_grad]: 0.00045979
[symbol_engine_optimizer]: 7.239e-05, [1]
    [Cycle 1]: 6.594e-05, [6]
        [build]: 3.65e-06
        [elim_shapecalc]: 7.28e-06
       
TotalTime = 0.0189698, [33]
[bootstrap]: 0.00031508
[type_inference]: 0.00489909
[event_method]: 1.757e-05
[auto_monad]: 4.458e-05
[graph_reusing]: 3.93999e-06
[pre_auto_parallel]: 1.82001e-06
[py_interpret_to_execute]: 3.608e-05
[rewriter_before_opt_a]: 6.637e-05
[expand_dump_flag]: 2.12999e-06
[jit_opt_a]: 0.00593116, [2]
    [Cycle 1]: 0.00183985, [27]
        [switch_simplify]: 5.865e-05
        [loop_unroll]: 2.997e-05
        [a_1]: 0.00055384
        [with_stream_mark]: 1.489e-05
        [recompute_prepare]: 7.78001e-06
        [updatestate_depend_eliminate]: 3.04001e-06
        [updatestate_assign_eliminate]: 2.11998e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 6.09001e-06
        [updatestate_useless_node_eliminater]: 5.40999e-06
        [accelerated_algorithm]: 5.94e-06
        [meta_shard_fg_expand]: 1.97001e-06
        [get_grad_eliminate_]: 5.59e-06
        [merge_forward]: 2.93e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.175e-05
        [j_node_and_user_rematch]: 9.33002e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 8.53001e-06
        [inline_without_move]: 5.54e-06
        [renormalize]: 0.00087774
        [add_forward_monad_depend]: 5.51002e-06
        [auto_monad_grad]: 1.02998e-06
        [auto_monad_eliminator]: 9.74e-06
        [cse]: 1.659e-05
        [replace_applicator]: 1.211e-05
    [Cycle 2]: 0.00033253, [27]
        [switch_simplify]: 5.62001e-06
        [loop_unroll]: 5.17e-06
        [a_1]: 9.51e-05
        [with_stream_mark]: 9.27001e-06
        [recompute_prepare]: 5.10999e-06
        [updatestate_depend_eliminate]: 2.76999e-06
        [updatestate_assign_eliminate]: 2.14999e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 4.98001e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 5.34998e-06
        [meta_shard_fg_expand]: 1.54998e-06
        [get_grad_eliminate_]: 4.58001e-06
        [merge_forward]: 2.20002e-06
        [cell_reuse_recompute_pass]: 2.53e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.231e-05
        [j_node_and_user_rematch]: 7.91001e-06
        [meta_fg_expand]: 1.72999e-06
        [replace_old_param]: 7.28e-06
        [inline_without_move]: 1.535e-05
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.04e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 4.92e-06
        [cse]: 1.014e-05
        [replace_applicator]: 5.29e-06
[py_interpret_to_execute_after_opt_a]: 1.062e-05
[rewriter_after_opt_a]: 2.82e-05
[convert_after_rewriter]: 5.66e-06
[order_py_execute_after_rewriter]: 4.1e-06
[mutable_eliminate]: 0.00064923
[jit_opt_b]: 4.773e-05, [1]
    [Cycle 1]: 4.138e-05, [2]
        [frontend_op_eliminate]: 1.561e-05
        [inline_after_opt_a]: 1.45e-05
[cconv]: 1.664e-05
[loop_unroll]: 0.00043255
[jit_opt_after_cconv]: 0.00014067, [1]
    [Cycle 1]: 0.00013365, [11]
        [c_1]: 2.106e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 5.12999e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.891e-05
        [call_graph_tuple_transform]: 1.844e-05
        [tuple_list_get_item_eliminator]: 5.37001e-06
        [none_parameter_eliminate]: 9.49978e-07
        [renormalize]: 5.50004e-07
        [switch_simplify]: 5.35001e-06
[remove_dup_value]: 7.3e-06
[partial_unused_args_eliminate]: 1.32999e-06
[environ_conv]: 3.73001e-06
[add_recomputation]: 3.532e-05
[cse_after_recomputation]: 2.065e-05, [1]
    [Cycle 1]: 1.521e-05, [1]
        [cse]: 8.80001e-06
[auto_monad_reorder]: 1.054e-05
[get_jit_bprop_graph]: 1.40001e-06
[rewriter_after_jit_bprop_graph]: 5.51e-06
[opt_after_jit_grad]: 0.00051286
[symbol_engine_optimizer]: 7.611e-05, [1]
    [Cycle 1]: 6.852e-05, [6]
        [build]: 2.98998e-06
        [elim_shapecalc]: 8.19002e-06 [elim_not_effective]: 1.291e-05
        [opt_reshape]: 5.68002e-06
        [fold_const_symbol]: 8.74e-06
        [renormalize]: 5.60016e-07
[validate]: 3.588e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.00627817
[execute]: 7.59002e-06
Sums
    bootstrap                            :     0.000288s :  1.93%
    type_inference                       :     0.004632s : 31.07%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000063s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.24%
    rewriter_before_opt_a                :     0.000072s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.43%
    jit_opt_a.loop_unroll                :     0.000035s :  0.23%
    jit_opt_a.a_1                        :     0.000617s :  4.14%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000022s :  0.14%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000582s :  3.91%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.10%
    jit_opt_a.cse                        :     0.000037s :  0.25%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.08%
    rewriter_after_opt_a                 :     0.000036s :  0.24%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000612s :  4.11%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000026s :  0.17%
    loop_unroll                          :     0.000479s :  3.21%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :
        [elim_not_effective]: 1.349e-05
        [opt_reshape]: 6.15002e-06
        [fold_const_symbol]: 8.85999e-06
        [renormalize]: 3.69997e-07
[validate]: 3.177e-05
[backend_pass]: 7.50006e-07
[task_emit]: 0.0054076
[execute]: 7.35e-06
Sums
    bootstrap                            :     0.000315s :  2.16%
    type_inference                       :     0.004899s : 33.55%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000045s :  0.31%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000036s :  0.25%
    rewriter_before_opt_a                :     0.000066s :  0.45%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000064s :  0.44%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000649s :  4.44%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000021s :  0.14%
    jit_opt_a.renormalize                :     0.000878s :  6.01%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.10%
    jit_opt_a.cse                        :     0.000027s :  0.18%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000028s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000649s :  4.45%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000017s :  0.11%
    loop_unroll                          :     0.000433s :  2.96%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.00  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000047s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000460s :  3.08%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006278s : 42.11%
    execute                              :     0.000008s :  0.05%

0007s :  0.05%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000035s :  0.24%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000011s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000513s :  3.51%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005408s : 37.03%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000201    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.62% :     0.000005s :      3: substitution.graph_param_transform
81.30% :     0.000164s :      4: substitution.inline
 1.59% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.11% :     0.000004s :      4: substitution.remove_not_recompute_node
 4.69% :     0.000009s :      2: substitution.replace_old_param
 6.01% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004564     2
81.99% :     0.003742s :      1: type_inference.infer
18.01% :     0.000822s :      1: type_inference.specialize
------[replace.]   0.000070     6
66.84% :     0.000047s :      4: replace.inline
33.16% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000172     6
93.66% :     0.000161s :      4: match.inline
 6.34% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.24% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 2.10% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.66% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.60% :     0.000007s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.51% :     0.000002s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.94% :     0.000002s :     15: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.22% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.63% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.27% :     0.000010s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.63% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.89% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.87% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.05% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.43% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000564    12
50.52% :     0.000285s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.41% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.07% :     0.000248s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020559    76
 0.24% :     0.000050s :      1: add_recomputation
 0.32% :     0.000067s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.48% :     0.000305s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000023s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.20% :     0.005386s :      1: jit_opt_a
 0.71% :     0.000145s :      1: jit_opt_after_cconv
 0.26% :     0.000054s :      1: jit_opt_b
 2.37% :     0.000487s :      1: loop_unroll
 3.03% :     0.000622s :      1: mutable_eliminate
 4.08% :     0.000839s :     26: opt.transform.jit_opt_a
 0.23% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.28% :     0.000468s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000039s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 1.45% :     0.000299s :      1: renormalize.infer
 1.34% :     0.000275s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000040s :      1: rewriter_after_opt_a
 0.37% :     0.000075s :      1: rewriter_before_opt_a
 0.37% :     0.000075s :      1: symbol_engine_optimizer
30.59% :     0.006289s :      
Time group info:
------[substitution.]   0.000201    23
 0.75% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 1.90% :     0.000004s :      3: substitution.graph_param_transform
86.58% :     0.000174s :      4: substitution.inline
 1.49% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.17% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.83% :     0.000004s :      2: substitution.replace_old_param
 4.57% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004840     2
81.54% :     0.003946s :      1: type_inference.infer
18.46% :     0.000893s :      1: type_inference.specialize
------[replace.]   0.000077     6
67.04% :     0.000052s :      4: replace.inline
32.96% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000179     6
95.48% :     0.000171s :      4: match.inline
 4.52% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.30% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.58% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.55% :     0.000002s :     10: predicate.addn_zero_filter
 2.57% :     0.000003s :     10: predicate.arithmetic_simplify
 1.25% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.22% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.38% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.72% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.50% :     0.000001s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.65% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.85% :     0.000002s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.33% :     0.000002s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 4.98% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.28% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.16% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.04% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     10: predicate.reduce_eliminate
 1.87% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.24% :     0.000002s :     10: predicate.reshape_eliminate
 1.32% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.63% :     0.000001s :      3:1: task_emit
22.60% :     0.004647s :      1: type_inference
 0.27% :     0.000056s :      1: validate

 predicate.row_tensor_eliminate
 1.32% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.65% :     0.000009s :     53: predicate.switch_simplify
 1.28% :     0.000002s :     10: predicate.tile_eliminate
 1.33% :     0.000002s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.34% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.10% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.92% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000638    12
46.44% :     0.000296s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.73% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.83% :     0.000312s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020822    76
 0.19% :     0.000039s :      1: add_recomputation
 0.23% :     0.000048s :      1: auto_monad
 0.06% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.58% :     0.000329s :      1: bootstrap
 0.09% :     0.000019s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
28.50% :     0.005935s :      1: jit_opt_a
 0.69% :     0.000144s :      1: jit_opt_after_cconv
 0.24% :     0.000050s :      1: jit_opt_b
 2.12% :     0.000441s :      1: loop_unroll
 3.15% :     0.000656s :      1: mutable_eliminate
 4.19% :     0.000873s :     26: opt.transform.jit_opt_a
 0.22% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.51% :     0.000522s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000039s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000010s :      1: remove_dup_value
 2.26% :     0.000471s :      1: renormalize.infer
 1.92% :     0.000399s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000031s :      1: rewriter_after_opt_a
 0.33% :     0.000069s :      1: rewriter_before_opt_a
 0.38% :     0.000079s :      1: symbol_engine_optimizer
26.03% :     0.005419s :      1: task_emit
23.59% :     0.004913s :      1: type_inference
 0.23% :     0.000049s :      1: validate


TotalTime = 0.0131991, [30]
[bootstrap]: 0.00033354
[type_inference]: 0.00473811
[event_method]: 1.788e-05
[auto_monad]: 6.464e-05
[graph_reusing]: 4.75999e-06
[pre_auto_parallel]: 2.66e-06
[py_interpret_to_execute]: 3.568e-05
[rewriter_before_opt_a]: 7.452e-05
[expand_dump_flag]: 3.09001e-06
[jit_opt_a]: 0.00580165, [2]
    [Cycle 1]: 0.00169324, [27]
        [switch_simplify]: 5.585e-05
        [loop_unroll]: 2.894e-05
        [a_1]: 0.00051775
        [with_stream_mark]: 1.6e-05
        [recompute_prepare]: 7.25e-06
        [updatestate_depend_eliminate]: 3.25e-06
        [updatestate_assign_eliminate]: 2.85998e-06
        [updatestate_loads_eliminate]: 2.73e-06
        [parameter_eliminate]: 2.04999e-06
        [specialize_transform]: 5.73002e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 2.14999e-06
        [get_grad_eliminate_]: 5.17999e-06
        [merge_forward]: 3.98999e-06
        [cell_reuse_recompute_pass]: 1.42e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.43e-05
        [j_node_and_user_rematch]: 8.81997e-06
        [meta_fg_expand]: 2.10002e-06
        [replace_old_param]: 8.45999e-06
        [inline_without_move]: 5.20001e-06
        [renormalize]: 0.00076095
        [add_forward_monad_depend]: 5.98002e-06
        [auto_monad_grad]: 2.79999e-06
        [auto_monad_eliminator]: 1.424e-05
        [cse]: 2.142e-05
        [replace_applicator]: 1.199e-05
    [Cycle 2]: 0.00031216, [27]
        [switch_simplify]: 5.47001e-06
        [loop_unroll]: 5.10001e-06
        [a_1]: 9.115e-05
        [with_stream_mark]: 8.67e-06
        [recompute_prepare]: 5.13002e-06
        [updatestate_depend_eliminate]: 2.59001e-06
        [updatestate_assign_eliminate]: 1.89e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 1.75001e-06
        [specialize_transform]: 4.90999e-06
        [updatestate_useless_node_eliminater]: 4.91002e-06
        [accelerated_algorithm]: 4.82e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.22e-06
        [merge_forward]: 2.32001e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.235e-05
        [j_node_and_user_rematch]: 7.96001e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 6.91001e-06
        [inline_without_move]: 4.84003e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 4.43001e-06
        [cse]: 9.79999e-06
        [replace_applicator]: 5.22999e-06
[py_interpret_to_execute_after_opt_a]: 1.036e-05
[rewriter_after_opt_a]: 3.293e-05
[convert_after_rewriter]: 7.41001e-06
[order_py_execute_after_rewriter]: 5.12999e-06
[mutable_eliminate]: 0.00057451
[jit_opt_b]: 4.928e-05, [1]
    [Cycle 1]: 4.208e-05, [2]
        [frontend_op_eliminate]: 1.597e-05
        [inline_after_opt_a]: 1.447e-05
[cconv]: 2.472e-05
[loop_unroll]: 0.00042663
[jit_opt_after_cconv]: 0.00014332, [1]
    [Cycle 1]: 0.00013643, [11]
        [c_1]: 2.153e-05
        [parameter_eliminate]: 2.93e-06
        [updatestate_depend_eliminate]: 5.89e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [cse]: 1.959e-05
        [call_graph_tuple_transform]: 1.915e-05
        [tuple_list_get_item_eliminator]: 5.17999e-06
        [none_parameter_eliminate]: 1.42999e-06
        [renormalize]: 9.09989e-07
        [switch_simplify]: 5.77999e-06
[remove_dup_value]: 1.49e-05
[partial_unused_args_eliminate]: 2.23002e-06
[environ_conv]: 5.58002e-06
[add_recomputation]: 4.485e-05
[cse_after_recomputation]: 2.12e-05, [1]
    [Cycle 1]: 1.5e-05, [1]
        [cse]: 9.32999e-06
[auto_monad_reorder]: 1.571e-05
[get_jit_bprop_graph]: 2.19999e-06
[rewriter_after_jit_bprop_graph]: 4.87e-06
[opt_after_jit_grad]: 0.00045227
[symbol_engine_optimizer]: 7.314e-05, [1]
    [Cycle 1]: 6.676e-05, [6]
        [build]: 3.42002e-06
        [elim_shapecalc]: 8.12e-06
        [elim_not_effective]: 1.299e-05
        [opt_reshape]: 5.48002e-06
        [fold_const_symbol]: 8.43999e-06
        [renormalize]: 3.30008e-07
[validate]: 3.644e-05
Sums
    bootstrap                            :     0.000334s :  3.78%
    type_inference                       :     0.004738s : 53.69%
    event_method                         :     0.000018s :  0.20%
    auto_monad                           :     0.000065s :  0.73%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.40%
    rewriter_before_opt_a                :     0.000075s :  0.84%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000061s :  0.69%
    jit_opt_a.loop_unroll                :     0.000034s :  0.39%
    jit_opt_a.a_1                        :     0.000609s :  6.90%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.17%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000761s :  8.62%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.21%
    jit_opt_a.cse                        :     0.000031s :  0.35%
    jit_opt_a.replace_applicator         :     0.000017s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000033s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000575s :  6.51%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.16%
    cconv                                :     0.000025s :  0.28%
    loop_unroll                          :     0.000427s :  4.83%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000020s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000045s :  0.51%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000452s :  5.12%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.41%


Time group info:
------[substitution.]   0.000194    23
 0.94% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.81% :     0.000005s :      3: substitution.graph_param_transform
83.58% :     0.000162s :      4: substitution.inline
 1.62% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.35% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.63% :     0.000003s :      2: substitution.replace_old_param
 6.40% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004669     2
79.48% :     0.003711s :      1: type_inference.infer
20.52% :     0.000958s :      1: type_inference.specialize
------[replace.]   0.000068     6
66.50% :     0.000045s :      4: replace.inline
33.50% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000171     6
93.44% :     0.000160s :      4: match.inline
 6.56% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.53% :     0.000002s :     10: predicate.addn_zero_filter
 1.75% :     0.000002s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.63% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.50% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.31% :     0.000006s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.33% :     0.000001s :      6: predicate.less_batch_normalization
 1.72% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.19% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.67% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.10% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.73% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.78% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.10% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000636    12
48.10% :     0.000306s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.53% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.37% :     0.000301s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014819    72
 0.33% :     0.000048s :      1: add_recomputation
 0.46% :     0.000068s :      1: auto_monad
 0.13% :     0.000019s :      1: auto_monad_reorder
 2.35% :     0.000348s :      1: bootstrap
 0.18% :     0.000027s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000022s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
39.18% :     0.005806s :      1: jit_opt_a
 0.99% :     0.000146s :      1: jit_opt_after_cconv
 0.35% :     0.000052s :      1: jit_opt_b
 2.94% :     0.000435s :      1: loop_unroll
 3.93% :     0.000583s :      1: mutable_eliminate
 5.50% :     0.000815s :     26: opt.transform.jit_opt_a
 0.33% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.11% :     0.000460s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000038s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.78% :     0.000412s :      1: renormalize.infer
 2.30% :     0.000341s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000036s :      1: rewriter_after_opt_a
 0.52% :     0.000077s :      1: rewriter_before_opt_a
 0.51% :     0.000076s :      1: symbol_engine_optimizer
32.05% :     0.004750s :      1: type_inference


TotalTime = 0.0183487, [33]
[bootstrap]: 0.00026633
[type_inference]: 0.00452381
[event_method]: 1.769e-05
[auto_monad]: 6.235e-05
[graph_reusing]: 5.61998e-06
[pre_auto_parallel]: 2.54001e-06
[py_interpret_to_execute]: 3.581e-05
[rewriter_before_opt_a]: 7.478e-05
[expand_dump_flag]: 3.09001e-06
[jit_opt_a]: 0.00506035, [2]
    [Cycle 1]: 0.00156518, [27]
        [switch_simplify]: 4.697e-05
        [loop_unroll]: 2.946e-05
        [a_1]: 0.00054663
        [with_stream_mark]: 1.913e-05
        [recompute_prepare]: 7.28999e-06
        [updatestate_depend_eliminate]: 3.50003e-06
        [updatestate_assign_eliminate]: 2.76e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.53002e-06
        [specialize_transform]: 5.62999e-06
        [updatestate_useless_node_eliminater]: 5.15999e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.96e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 3.2e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.373e-05
        [j_node_and_user_rematch]: 9.04e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 7.84002e-06
        [inline_without_move]: 4.86002e-06
        [renormalize]: 0.00060645
        [add_forward_monad_depend]: 5.51998e-06
        [auto_monad_grad]: 2.73e-06
        [auto_monad_eliminator]: 1.369e-05
        [cse]: 2.861e-05
        [replace_applicator]: 1.232e-05
    [Cycle 2]: 0.00032979, [27]
        [switch_simplify]: 5.97001e-06
        [loop_unroll]: 5.27001e-06
        [a_1]: 8.948e-05
        [with_stream_mark]: 1.131e-05
        [recompute_prepare]: 5.08002e-06
        [updatestate_depend_eliminate]: 2.80002e-06
        [updatestate_assign_eliminate]: 2.43998e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.00001e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.30001e-06
        [meta_shard_fg_expand]: 1.50999e-06
        [get_grad_eliminate_]: 4.78001e-06
        [merge_forward]: 3.11001e-06
        [cell_reuse_recompute_pass]: 2.96999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.329e-05
        [j_node_and_user_rematch]: 8.08001e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 8.42e-06
        [inline_without_move]: 4.82e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.60001e-06
        [auto_monad_grad]: 1.20999e-06
        [auto_monad_eliminator]: 5.47001e-06
        [cse]: 1.03e-05
        [replace_applicator]: 5.44e-06
[py_interpret_to_execute_after_opt_a]: 1.015e-05
[rewriter_after_opt_a]: 3.552e-05
[convert_after_rewriter]: 7.16001e-06
[order_py_execute_after_rewriter]: 5.39e-06
[mutable_eliminate]: 0.00053951
[jit_opt_b]: 5.098e-05, [1]
    [Cycle 1]: 4.359e-05, [2]
        [frontend_op_eliminate]: 1.668e-05
        [inline_after_opt_a]: 1.573e-05
[cconv]: 2.42e-05
[loop_unroll]: 0.00042524
[jit_opt_after_cconv]: 0.00014141, [1]
    [Cycle 1]: 0.00013484, [11]
        [c_1]: 2.082e-05
        [parameter_eliminate]: 2.59001e-06
        [updatestate_depend_eliminate]: 6.01998e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 2.021e-05
        [call_graph_tuple_transform]: 1.835e-05
        [tuple_list_get_item_eliminator]: 5.46998e-06
        [none_parameter_eliminate]: 1.67001e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 5.32999e-06
[remove_dup_value]: 1.382e-05
[partial_unused_args_eliminate]: 2.02999e-06
[environ_conv]: 4.96002e-06
[add_recomputation]: 4.829e-05
[cse_after_recomputation]: 2.012e-05, [1]
    [Cycle 1]: 1.455e-05, [1]
        [cse]: 8.70999e-06
[auto_monad_reorder]: 1.52e-05
[get_jit_bprop_graph]: 1.96e-06
[rewriter_after_jit_bprop_graph]: 6.06e-06
[opt_after_jit_grad]: 0.00046001
[symbol_engine_optimizer]: 7.122e-05, [1]
    [Cycle 1]: 6.482e-05, [6]
        [build]: 3.26001e-06
        [elim_shapecalc]: 7.71999e-06
        [elim_not_effective]: 1.239e-05
        [opt_reshape]: 5.47999e-06
        [fold_const_symbol]: 8.3e-06
        [renormalize]: 3.69997e-07
[validate]: 3.348e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.00618805
[execute]: 6.58998e-06
Sums
    bootstrap                            :     0.000266s :  1.83%
    type_inference                       :     0.004524s : 31.00%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000062s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000036s :  0.25%
    rewriter_before_opt_a                :     0.000075s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.36%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000636s :  4.36%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000607s :  4.16%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000039s :  0.27%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000540s :  3.70%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000425s :  2.91%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000048s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000460s :  3.15%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006188s : 42.41%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000173    23
 1.12% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000001s :      2: substitution.fold_const_symbol
 2.76% :     0.000005s :      3: substitution.graph_param_transform
82.64% :     0.000143s :      4: substitution.inline
 1.73% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.43% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.41% :     0.000004s :      2: substitution.replace_old_param
 6.14% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004424     2
81.52% :     0.003607s :      1: type_inference.infer
18.48% :     0.000818s :      1: type_inference.specialize
------[replace.]   0.000061     6
66.04% :     0.000040s :      4: replace.inline
33.96% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000149     6
93.72% :     0.000140s :      4: match.inline
 6.28% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.38% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 2.18% :     0.000003s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.35% :     0.000002s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.48% :     0.000006s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 2.15% :     0.000003s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.40% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.00% :     0.000002s :     18: predicate.replace_applicator
 0.80% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.05% :     0.000004s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.74% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.95% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000539    12
47.04% :     0.000253s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.87% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.09% :     0.000254s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019900    76
 0.26% :     0.000051s :      1: add_recomputation
 0.33% :     0.000066s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.41% :     0.000280s :      1: bootstrap
 0.13% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.44% :     0.005063s :      1: jit_opt_a
 0.72% :     0.000144s :      1: jit_opt_after_cconv
 0.27% :     0.000053s :      1: jit_opt_b
 2.18% :     0.000433s :      1: loop_unroll
 2.76% :     0.000549s :      1: mutable_eliminate
 4.21% :     0.000838s :     26: opt.transform.jit_opt_a
 0.23% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.35% :     0.000468s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000039s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.57% :     0.000313s :      1: renormalize.infer
 1.43% :     0.000285s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000039s :      1: rewriter_after_opt_a
 0.39% :     0.000078s :      1: rewriter_before_opt_a
 0.37% :     0.000074s :      1: symbol_engine_optimizer
31.15% :     0.006198s :      1: task_emit
22.79% :     0.004535s :      1: type_inference
 0.27% :     0.000053s :      1: validate


TotalTime = 0.0191027, [33]
[bootstrap]: 0.00028692
[type_inference]: 0.00474267
[event_method]: 1.693e-05
[auto_monad]: 5.438e-05
[graph_reusing]: 4.70001e-06
[pre_auto_parallel]: 2.01e-06
[py_interpret_to_execute]: 3.765e-05
[rewriter_before_opt_a]: 6.787e-05
[expand_dump_flag]: 2.67001e-06
[jit_opt_a]: 0.00573084, [2]
    [Cycle 1]: 0.00171231, [27]
        [switch_simplify]: 5.259e-05
        [loop_unroll]: 3.032e-05
        [a_1]: 0.00053667
        [with_stream_mark]: 1.45e-05
        [recompute_prepare]: 7.45998e-06
        [updatestate_depend_eliminate]: 2.94001e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.73002e-06
        [specialize_transform]: 5.76998e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 1.39998e-06
        [get_grad_eliminate_]: 5.20999e-06
        [merge_forward]: 2.86999e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.179e-05
        [j_node_and_user_rematch]: 8.80999e-06
        [meta_fg_expand]: 2.35002e-06
        [replace_old_param]: 8.70001e-06
        [inline_without_move]: 5.20001e-06
        [renormalize]: 0.00078162
        [add_forward_monad_depend]: 5.67999e-06
        [auto_monad_grad]: 1.57001e-06
        [auto_monad_eliminator]: 1.244e-05
        [cse]: 1.819e-05
        [replace_applicator]: 1.179e-05
    [Cycle 2]: 0.00032111, [27]
        [switch_simplify]: 5.59998e-06
        [loop_unroll]: 5.10999e-06
        [a_1]: 9.055e-05
        [with_stream_mark]: 8.75999e-06
        [recompute_prepare]: 5.14e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 5.24e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 5.09e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 4.67e-06
        [merge_forward]: 2.49001e-06
        [cell_reuse_recompute_pass]: 2.56e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.257e-05
        [j_node_and_user_rematch]: 7.58001e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 7.41001e-06
        [inline_without_move]: 4.52e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 5.44998e-06
        [cse]: 9.89999e-06
        [replace_applicator]: 4.75001e-06
[py_interpret_to_execute_after_opt_a]: 1.042e-05
[rewriter_after_opt_a]: 3.176e-05
[convert_after_rewriter]: 6.58e-06
[order_py_execute_after_rewriter]: 4.98001e-06
[mutable_eliminate]: 0.00062724
[jit_opt_b]: 5.009e-05, [1]
    [Cycle 1]: 4.348e-05, [2]
        [frontend_op_eliminate]: 1.689e-05
        [inline_after_opt_a]: 1.472e-05
[cconv]: 2.113e-05
[loop_unroll]: 0.00042332
[jit_opt_after_cconv]: 0.00013962, [1]
    [Cycle 1]: 0.00013317, [11]
        [c_1]: 2.05e-05
        [parameter_eliminate]: 2.70002e-06
        [updatestate_depend_eliminate]: 5.64e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 1.973e-05
        [call_graph_tuple_transform]: 1.813e-05
        [tuple_list_get_item_eliminator]: 5.42001e-06
        [none_parameter_eliminate]: 1.17999e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 5.30999e-06
[remove_dup_value]: 1.057e-05
[partial_unused_args_eliminate]: 2.07999e-06
[environ_conv]: 4.79002e-06
[add_recomputation]: 3.794e-05
[cse_after_recomputation]: 1.984e-05, [1]
    [Cycle 1]: 1.471e-05, [1]
        [cse]: 8.87999e-06
[auto_monad_reorder]: 1.252e-05
[get_jit_bprop_graph]: 1.55001e-06
[rewriter_after_jit_bprop_graph]: 5.79e-06
[opt_after_jit_grad]: 0.00046686
[symbol_engine_optimizer]: 7.695e-05, [1]
    [Cycle 1]: 7.014e-05, [6]
        [build]: 3.56001e-06
        [elim_shapecalc]: 8.27e-06
        [elim_not_effective]: 1.399e-05
        [opt_reshape]: 6.58e-06
        [fold_const_symbol]: 9.17001e-06
        [renormalize]: 5.19998e-07
[validate]: 3.326e-05
[backend_pass]: 7.59988e-07
[task_emit]: 0.00597104
[execute]: 5.82001e-06
Sums
    bootstrap                            :     0.000287s :  1.94%
    type_inference                       :     0.004743s : 32.02%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000054s :  0.37%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000038s :  0.25%
    rewriter_before_opt_a                :     0.000068s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000627s :  4.23%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000782s :  5.28%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000028s :  0.19%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000627s :  4.23%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000021s :  0.14%
    loop_unroll                          :     0.000423s :  2.86%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000011s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000038s :  0.26%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000013s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000467s :  3.15%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000033s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005971s : 40.32%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000201    23
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.15% :     0.000004s :      3: substitution.graph_param_transform
84.81% :     0.000170s :      4: substitution.inline
 1.41% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.13% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000004s :      2: substitution.replace_old_param
 5.90% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004680     2
79.98% :     0.003743s :      1: type_inference.infer
20.02% :     0.000937s :      1: type_inference.specialize
------[replace.]   0.000071     6
67.18% :     0.000048s :      4: replace.inline
32.82% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000178     6
93.97% :     0.000167s :      4: match.inline
 6.03% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.48% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000001s :     10: predicate.addn_zero_filter
 2.47% :     0.000003s :     10: predicate.arithmetic_simplify
 1.51% :     0.000002s :     10: predicate.cast_eliminate
 0.77% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 1.38% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.12% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.58% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.23% :     0.000006s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.49% :     0.000002s :      6: predicate.less_batch_normalization
 1.69% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.89% :     0.000002s :     15: predicate.load_eliminater
 1.48% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.88% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.10% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.23% :     0.000001s :     10: predicate.minmaximum_grad
 1.64% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000000s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     10: predicate.reduce_eliminate
 1.65% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.44% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.39% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.22% :     0.000003s :     16: predicate.switch_defer_inline
 1.99% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.36% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.60% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.70% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     10: predicate.value_based_eliminate
 0.47% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000626    12
46.63% :     0.000292s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.27% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.09% :     0.000301s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020823    76
 0.20% :     0.000041s :      1: add_recomputation
 0.28% :     0.000058s :      1: auto_monad
 0.07% :     0.000015s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.44% :     0.000301s :      1: bootstrap
 0.11% :     0.000024s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.04% :     0.000009s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
27.54% :     0.005734s :      1: jit_opt_a
 0.68% :     0.000142s :      1: jit_opt_after_cconv
 0.25% :     0.000052s :      1: jit_opt_b
 2.07% :     0.000432s :      1: loop_unroll
 3.05% :     0.000636s :      1: mutable_eliminate
 4.00% :     0.000832s :     26: opt.transform.jit_opt_a
 0.22% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.28% :     0.000475s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000040s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000013s :      1: remove_dup_value
 2.02% :     0.000421s :      1: renormalize.infer
 1.69% :     0.000353s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000035s :      1: rewriter_after_opt_a
 0.34% :     0.000071s :      1: rewriter_before_opt_a
 0.38% :     0.000079s :      1: symbol_engine_optimizer
28.73% :     0.005983s :      1: task_emit
22.83% :     0.004754s :      1: type_inference
 0.25% :     0.000052s :      1: validate


TotalTime = 0.0145261, [30]
[bootstrap]: 0.00031792
[type_inference]: 0.00441216
[event_method]: 1.806e-05
[auto_monad]: 5.822e-05
[graph_reusing]: 5.44e-06
[pre_auto_parallel]: 2.29999e-06
[py_interpret_to_execute]: 3.473e-05
[rewriter_before_opt_a]: 6.829e-05
[expand_dump_flag]: 2.96001e-06
[jit_opt_a]: 0.0063895, [2]
    [Cycle 1]: 0.00225523, [27]
        [switch_simplify]: 5.268e-05
        [loop_unroll]: 2.912e-05
        [a_1]: 0.00054331
        [with_stream_mark]: 1.41e-05
        [recompute_prepare]: 7.85e-06
        [updatestate_depend_eliminate]: 3.4e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 6.07999e-06
        [updatestate_useless_node_eliminater]: 5.35999e-06
        [accelerated_algorithm]: 5.47999e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 5.23002e-06
        [merge_forward]: 2.91e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.279e-05
        [j_node_and_user_rematch]: 9.65002e-06
        [meta_fg_expand]: 2.36e-06
        [replace_old_param]: 8.28001e-06
        [inline_without_move]: 5.59e-06
        [renormalize]: 0.00118454
        [add_forward_monad_depend]: 1.781e-05
        [auto_monad_grad]: 3.13e-06
        [auto_monad_eliminator]: 3.135e-05
        [cse]: 3.719e-05
        [replace_applicator]: 3.495e-05
    [Cycle 2]: 0.00048557, [27]
        [switch_simplify]: 8.35999e-06
        [loop_unroll]: 6.39001e-06
        [a_1]: 0.00013459
        [with_stream_mark]: 2.406e-05
        [recompute_prepare]: 6.76999e-06
        [updatestate_depend_eliminate]: 4.09997e-06
        [updatestate_assign_eliminate]: 2.98998e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [parameter_eliminate]: 2.37999e-06
        [specialize_transform]: 6.41998e-06
        [updatestate_useless_node_eliminater]: 6.87002e-06
        [accelerated_algorithm]: 9.00001e-06
        [meta_shard_fg_expand]: 2.84001e-06
        [get_grad_eliminate_]: 5.74e-06
        [merge_forward]: 4.75001e-06
        [cell_reuse_recompute_pass]: 4.11001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.983e-05
        [j_node_and_user_rematch]: 1.153e-05
        [meta_fg_expand]: 3.50998e-06
        [replace_old_param]: 1.322e-05
        [inline_without_move]: 5.25001e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.68002e-06
        [auto_monad_grad]: 1.07998e-06
        [auto_monad_eliminator]: 5.95002e-06
        [cse]: 1.304e-05
        [replace_applicator]: 5.30001e-06
[py_interpret_to_execute_after_opt_a]: 2.83e-05
[rewriter_after_opt_a]: 4.527e-05
[convert_after_rewriter]: 7.19001e-06
[order_py_execute_after_rewriter]: 1.245e-05
[mutable_eliminate]: 0.00107514
[jit_opt_b]: 7.565e-05, [1]
    [Cycle 1]: 6.283e-05, [2]
        [frontend_op_eliminate]: 2.313e-05
        [inline_after_opt_a]: 2.146e-05
[cconv]: 4.406e-05
[loop_unroll]: 0.00060882
[jit_opt_after_cconv]: 0.00019373, [1]
    [Cycle 1]: 0.00018609, [11]
        [c_1]: 2.37e-05
        [parameter_eliminate]: 5.67001e-06
        [updatestate_depend_eliminate]: 1.138e-05
        [updatestate_assign_eliminate]: 3.25e-06
        [updatestate_loads_eliminate]: 2.93e-06
        [cse]: 3.564e-05
        [call_graph_tuple_transform]: 2.81e-05
        [tuple_list_get_item_eliminator]: 6.23e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 1.20999e-06
        [switch_simplify]: 6.24999e-06
[remove_dup_value]: 1.901e-05
[partial_unused_args_eliminate]: 2.49001e-06
[environ_conv]: 6.45002e-06
[add_recomputation]: 6.033e-05
[cse_after_recomputation]: 2.607e-05, [1]
    [Cycle 1]: 1.89e-05, [1]
        [cse]: 1.123e-05
[auto_monad_reorder]: 1.761e-05
[get_jit_bprop_graph]: 2.73e-06
[rewriter_after_jit_bprop_graph]: 1.266e-05
[opt_after_jit_grad]: 0.00063123
[symbol_engine_optimizer]: 8.549e-05, [1]
    [Cycle 1]: 7.721e-05, [6]
        [build]: 4.74e-06
        [elim_shapecalc]: 9.14998e-06
        [elim_not_effective]: 1.62e-05
        [opt_reshape]: 6.16e-06
        [fold_const_symbol]: 1.014e-05
        [renormalize]: 5.50004e-07
[validate]: 5.051e-05
Sums
    bootstrap                            :     0.000318s :  3.14%
    type_inference                       :     0.004412s : 43.57%
    event_method                         :     0.000018s :  0.18%
    auto_monad                           :     0.000058s :  0.57%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.34%
    rewriter_before_opt_a                :     0.000068s :  0.67%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000061s :  0.60%
    jit_opt_a.loop_unroll                :     0.000036s :  0.35%
    jit_opt_a.a_1                        :     0.000678s :  6.69%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.38%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000014s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000043s :  0.42%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.06%
    jit_opt_a.replace_old_param          :     0.000022s :  0.21%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.001185s : 11.70%
    jit_opt_a.add_forward_monad_depend   :     0.000019s :  0.19%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000037s :  0.37%
    jit_opt_a.cse                        :     0.000050s :  0.50%
    jit_opt_a.replace_applicator         :     0.000040s :  0.40%
    py_interpret_to_execute_after_opt_a  :     0.000028s :  0.28%
    rewriter_after_opt_a                 :     0.000045s :  0.45%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000012s :  0.12%
    mutable_eliminate                    :     0.001075s : 10.62%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000021s :  0.21%
    cconv                                :     0.000044s :  0.44%
    loop_unroll                          :     0.000609s :  6.01%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.11%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000036s :  0.35%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.28%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000019s :  0.19%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000060s :  0.60%
    cse_after_recomputation.cse          :     0.000011s :  0.11%
    auto_monad_reorder                   :     0.000018s :  0.17%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000013s :  0.13%
    opt_after_jit_grad                   :     0.000631s :  6.23%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000051s :  0.50%


Time group info:
------[substitution.]   0.000212    23
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.86% :     0.000002s :      2: substitution.fold_const_symbol
 3.30% :     0.000007s :      3: substitution.graph_param_transform
80.81% :     0.000172s :      4: substitution.inline
 2.10% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.63% :     0.000006s :      4: substitution.remove_not_recompute_node
 3.87% :     0.000008s :      2: substitution.replace_old_param
 5.52% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004353     2
79.91% :     0.003478s :      1: type_inference.infer
20.09% :     0.000874s :      1: type_inference.specialize
------[replace.]   0.000075     6
66.62% :     0.000050s :      4: replace.inline
33.38% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000180     6
94.11% :     0.000169s :      4: match.inline
 5.89% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000140   691
 1.65% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.50% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.43% :     0.000002s :     10: predicate.addn_zero_filter
 2.35% :     0.000003s :     10: predicate.arithmetic_simplify
 1.22% :     0.000002s :     10: predicate.cast_eliminate
 0.62% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.89% :     0.000001s :     10: predicate.compare_switch_simplify
 1.10% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.27% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.82% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.02% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000002s :     10: predicate.environ_get_eliminate
 0.89% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.13% :     0.000002s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.56% :     0.000008s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 2.25% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.51% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.07% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.92% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 3.04% :     0.000004s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 1.86% :     0.000003s :     15: predicate.partial_eliminate
 1.03% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.59% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.13% :     0.000003s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.41% :     0.000001s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.02% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     10: predicate.same_eliminate
 0.42% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 0.80% :     0.000001s :      6: predicate.specialize_transform
 1.84% :     0.000003s :     10: predicate.split_environ_get_set_with_tuple_value
 0.96% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.20% :     0.000003s :     16: predicate.switch_defer_inline
 1.77% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.47% :     0.000009s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.22% :     0.000002s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.00% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.18% :     0.000003s :     10: predicate.value_based_eliminate
 0.28% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.27% :     0.000002s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000626    12
48.22% :     0.000302s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.24% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.54% :     0.000291s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016754    72
 0.38% :     0.000063s :      1: add_recomputation
 0.37% :     0.000062s :      1: auto_monad
 0.12% :     0.000020s :      1: auto_monad_reorder
 1.97% :     0.000331s :      1: bootstrap
 0.28% :     0.000047s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000029s :      1: cse_after_recomputation
 0.06% :     0.000010s :      1: environ_conv
 0.13% :     0.000022s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
38.17% :     0.006395s :      1: jit_opt_a
 1.18% :     0.000197s :      1: jit_opt_after_cconv
 0.47% :     0.000079s :      1: jit_opt_b
 3.70% :     0.000620s :      1: loop_unroll
 6.55% :     0.001097s :      1: mutable_eliminate
 5.61% :     0.000940s :     26: opt.transform.jit_opt_a
 0.36% :     0.000061s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000036s :      4: opt.transform.jit_opt_b
 0.10% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.16% :     0.000026s :      1: opt.transform.mutable_eliminate
 0.38% :     0.000064s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000038s :      4: opt.transform.symbol_engine_opt
 3.84% :     0.000644s :      1: opt_after_jit_grad
 0.09% :     0.000015s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.22% :     0.000037s :      1: py_interpret_to_execute
 0.19% :     0.000031s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000021s :      1: remove_dup_value
 2.62% :     0.000439s :      1: renormalize.infer
 4.35% :     0.000728s :      1: renormalize.specialize
 0.09% :     0.000015s :      1: rewriter_after_jit_bprop_graph
 0.30% :     0.000050s :      1: rewriter_after_opt_a
 0.42% :     0.000071s :      1: rewriter_before_opt_a
 0.53% :     0.000088s :      1: symbol_engine_optimizer
26.41% :     0.004425s :      1: type_inference


TotalTime = 0.10855, [33]
[bootstrap]: 0.00026796
[type_inference]: 0.0818781
[event_method]: 0.0002811
[auto_monad]: 0.0001032
[graph_reusing]: 7.43e-06
[pre_auto_parallel]: 3.05002e-06
[py_interpret_to_execute]: 4.439e-05
[rewriter_before_opt_a]: 0.00010733
[expand_dump_flag]: 4.09002e-06
[jit_opt_a]: 0.0233393, [3]
    [Cycle 1]: 0.0149417, [27]
        [switch_simplify]: 7.654e-05
        [loop_unroll]: 4.294e-05
        [a_1]: 0.00101707
        [with_stream_mark]: 2.924e-05
        [recompute_prepare]: 2.751e-05
        [updatestate_depend_eliminate]: 9.84001e-06
        [updatestate_assign_eliminate]: 7.43e-06
        [updatestate_loads_eliminate]: 6.89001e-06
        [parameter_eliminate]: 3.68999e-06
        [specialize_transform]: 1.509e-05
        [updatestate_useless_node_eliminater]: 1.424e-05
        [accelerated_algorithm]: 1.405e-05
        [meta_shard_fg_expand]: 3.8e-06
        [get_grad_eliminate_]: 1.418e-05
        [merge_forward]: 1.014e-05
        [cell_reuse_recompute_pass]: 1.59998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.871e-05
        [j_node_and_user_rematch]: 2.431e-05
        [meta_fg_expand]: 0.00180859
        [replace_old_param]: 6.537e-05
        [inline_without_move]: 6.063e-05
        [renormalize]: 0.0110672
        [add_forward_monad_depend]: 1.038e-05
        [auto_monad_grad]: 5.64998e-06
        [auto_monad_eliminator]: 5.564e-05
        [cse]: 0.00021689
        [replace_applicator]: 8.271e-05
    [Cycle 2]: 0.0041792, [27]
        [switch_simplify]: 4.348e-05
        [loop_unroll]: 4.286e-05
        [a_1]: 0.00124748
        [with_stream_mark]: 1.39e-05
        [recompute_prepare]: 1.18e-05
        [updatestate_depend_eliminate]: 5.39e-06
        [updatestate_assign_eliminate]: 3.97e-06
        [updatestate_loads_eliminate]: 3.51999e-06
        [parameter_eliminate]: 1.39e-06
        [specialize_transform]: 9.97999e-06
        [updatestate_useless_node_eliminater]: 8.62998e-06
        [accelerated_algorithm]: 1.031e-05
        [meta_shard_fg_expand]: 2.12999e-06
        [get_grad_eliminate_]: 8.98002e-06
        [merge_forward]: 4.30999e-06
        [cell_reuse_recompute_pass]: 9.20001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.729e-05
        [j_node_and_user_rematch]: 1.455e-05
        [meta_fg_expand]: 7.137e-05
        [replace_old_param]: 9.23002e-06
        [inline_without_move]: 8.92e-06
        [renormalize]: 0.00240894
        [add_forward_monad_depend]: 8.77e-06
        [auto_monad_grad]: 2.88998e-06
        [auto_monad_eliminator]: 1.386e-05
        [cse]: 2.306e-05
        [replace_applicator]: 1.901e-05
    [Cycle 3]: 0.00027032, [27]
        [switch_simplify]: 4.77998e-06
        [loop_unroll]: 4.40999e-06
        [a_1]: 5.241e-05
        [with_stream_mark]: 1.16e-05
        [recompute_prepare]: 4.37e-06
        [updatestate_depend_eliminate]: 2.12001e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 3.78999e-06
        [updatestate_useless_node_eliminater]: 3.22002e-06
        [accelerated_algorithm]: 4.43001e-06
        [meta_shard_fg_expand]: 1.47999e-06
        [get_grad_eliminate_]: 3.24001e-06
        [merge_forward]: 2.24999e-06
        [cell_reuse_recompute_pass]: 4.28001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.343e-05
        [j_node_and_user_rematch]: 7.87e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 3.43999e-06
        [inline_without_move]: 3.43999e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 9.39996e-07
        [auto_monad_grad]: 1.09e-06
        [auto_monad_eliminator]: 4.15e-06
        [cse]: 9.59999e-06
        [replace_applicator]: 4.04002e-06
[py_interpret_to_execute_after_opt_a]: 1.414e-05
[rewriter_after_opt_a]: 3.501e-05
[convert_after_rewriter]: 6.61e-06
[order_py_execute_after_rewriter]: 4.92999e-06
[mutable_eliminate]: 0.00080197
[jit_opt_b]: 4.56e-05, [1]
    [Cycle 1]: 3.672e-05, [2]
        [frontend_op_eliminate]: 1.196e-05
        [inline_after_opt_a]: 1.176e-05
[cconv]: 2.862e-05
[loop_unroll]: 0.00048362
[jit_opt_after_cconv]: 0.00013159, [1]
    [Cycle 1]: 0.00012359, [11]
        [c_1]: 1.361e-05
        [parameter_eliminate]: 3.16999e-06
        [updatestate_depend_eliminate]: 5.54e-06
        [updatestate_assign_eliminate]: 3.22002e-06
        [updatestate_loads_eliminate]: 1.57001e-06
        [cse]: 1.85e-05
        [call_graph_tuple_transform]: 1.545e-05
        [tuple_list_get_item_eliminator]: 4.00998e-06
        [none_parameter_eliminate]: 1.72999e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 3.86001e-06
[remove_dup_value]: 1.363e-05
[partial_unused_args_eliminate]: 2.06e-06
[environ_conv]: 6.16e-06
[add_recomputation]: 3.737e-05
[cse_after_recomputation]: 2.176e-05, [1]
    [Cycle 1]: 1.604e-05, [1]
        [cse]: 9.59999e-06
[auto_monad_reorder]: 1.351e-05
[get_jit_bprop_graph]: 1.91e-06
[rewriter_after_jit_bprop_graph]: 5.81998e-06
[opt_after_jit_grad]: 0.00050517
[symbol_engine_optimizer]: 6.941e-05, [1]
    [Cycle 1]: 6.211e-05, [6]
        [build]: 3.21999e-06
        [elim_shapecalc]: 7.88001e-06
        [elim_not_effective]: 1.127e-05
        [opt_reshape]: 4.23999e-06
        [fold_const_symbol]: 6.78998e-06
        [renormalize]: 4.59986e-07
[validate]: 3.379e-05
[backend_pass]: 8.2e-07
[task_emit]: 2.131e-05
[execute]: 1.32999e-06
Sums
    bootstrap                            :     0.000268s :  0.26%
    type_inference                       :     0.081878s : 78.92%
    event_method                         :     0.000281s :  0.27%
    auto_monad                           :     0.000103s :  0.10%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000044s :  0.04%
    rewriter_before_opt_a                :     0.000107s :  0.10%
    expand_dump_flag                     :     0.000004s :  0.00%
    jit_opt_a.switch_simplify            :     0.000125s :  0.12%
    jit_opt_a.loop_unroll                :     0.000090s :  0.09%
    jit_opt_a.a_1                        :     0.002317s :  2.23%
    jit_opt_a.with_stream_mark           :     0.000055s :  0.05%
    jit_opt_a.recompute_prepare          :     0.000044s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.01%
    jit_opt_a.specialize_transform       :     0.000029s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000029s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000026s :  0.03%
    jit_opt_a.merge_forward              :     0.000017s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000059s :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000047s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.001882s :  1.81%
    jit_opt_a.replace_old_param          :     0.000078s :  0.08%
    jit_opt_a.inline_without_move        :     0.000073s :  0.07%
    jit_opt_a.renormalize                :     0.013476s : 12.99%
    jit_opt_a.add_forward_monad_depend   :     0.000020s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000010s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000074s :  0.07%
    jit_opt_a.cse                        :     0.000250s :  0.24%
    jit_opt_a.replace_applicator         :     0.000106s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.01%
    rewriter_after_opt_a                 :     0.000035s :  0.03%
    convert_after_rewriter               :     0.000007s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.00%
    mutable_eliminate                    :     0.000802s :  0.77%
    jit_opt_b.frontend_op_eliminate      :     0.000012s :  0.01%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.01%
    cconv                                :     0.000029s :  0.03%
    loop_unroll                          :     0.000484s :  0.47%
    jit_opt_after_cconv.c_1              :     0.000014s :  0.01%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000018s :  0.02%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.01%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.00%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.00%
    remove_dup_value                     :     0.000014s :  0.01%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000006s :  0.01%
    add_recomputation                    :     0.000037s :  0.04%
    cse_after_recomputation.cse          :     0.000010s :  0.01%
    auto_monad_reorder                   :     0.000014s :  0.01%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000505s :  0.49%
    symbol_engine_optimizer.build        :     0.000003s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.01%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.00%
    symbol_engine_optimizer.fold_const_symbol :     0.000007s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.03%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000021s :  0.02%
    execute                              :     0.000001s :  0.00%


Time group info:
------[substitution.]   0.000570   105
 0.25% :     0.000001s :      1: substitution.elim_not_effective
 0.18% :     0.000001s :      1: substitution.fold_const_symbol
 0.80% :     0.000005s :      1: substitution.graph_param_transform
59.73% :     0.000340s :     12: substitution.inline
 3.78% :     0.000022s :      2: substitution.inline_without_move
 1.72% :     0.000010s :     13: substitution.j_node_and_user_rematch
 2.09% :     0.000012s :      7: substitution.minmaximum_grad
 1.30% :     0.000007s :      5: substitution.partial_eliminate
 1.90% :     0.000011s :     13: substitution.remove_not_recompute_node
 5.29% :     0.000030s :     10: substitution.replace_applicator
 1.02% :     0.000006s :      7: substitution.replace_old_param
 5.48% :     0.000031s :      2: substitution.reshape_eliminate
 0.61% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.86% :     0.000011s :      2: substitution.tile_eliminate
 4.13% :     0.000023s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.82% :     0.000016s :      7: substitution.tuple_list_get_item_depend_reorder
 7.05% :     0.000040s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.081779     2
96.99% :     0.079317s :      1: type_inference.infer
 3.01% :     0.002462s :      1: type_inference.specialize
------[replace.]   0.000156    19
62.64% :     0.000098s :     12: replace.inline
37.36% :     0.000058s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000353    19
94.01% :     0.000332s :     12: match.inline
 5.99% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000407  2758
 2.06% :     0.000008s :     48: predicate.accumulaten_eliminater
 0.40% :     0.000002s :      1: predicate.ad_related_special_op_eliminate
 1.41% :     0.000006s :     48: predicate.addn_check_dump
 1.88% :     0.000008s :     48: predicate.addn_zero_filter
 2.70% :     0.000011s :     48: predicate.arithmetic_simplify
 1.76% :     0.000007s :     48: predicate.cast_eliminate
 0.10% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.35% :     0.000006s :     48: predicate.compare_switch_simplify
 1.63% :     0.000007s :     48: predicate.depend_value_elim
 1.52% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.57% :     0.000006s :     48: predicate.dict_get_item_eliminator
 1.57% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.17% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.05% :     0.000000s :      1: predicate.elim_not_effective
 0.15% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.46% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.44% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.41% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.60% :     0.000006s :     48: predicate.environ_get_eliminate
 1.42% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.02% :     0.000000s :      1: predicate.fold_const_symbol
 0.85% :     0.000003s :     21: predicate.get_grad_eliminate
 0.04% :     0.000000s :      1: predicate.graph_param_transform
 4.03% :     0.000016s :     69: predicate.inline
 1.76% :     0.000007s :     45: predicate.inline_without_move
 0.37% :     0.000002s :     21: predicate.j_node_and_user_rematch
 1.24% :     0.000005s :     21: predicate.less_batch_normalization
 1.93% :     0.000008s :     55: predicate.list_to_tuple_eliminator_
 1.80% :     0.000007s :     56: predicate.load_eliminater
 0.28% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.29% :     0.000013s :     97: predicate.loop_unroll_before_grad
 1.81% :     0.000007s :     49: predicate.make_slice_get_slice_eliminator
 1.40% :     0.000006s :     48: predicate.merge_addn
 1.44% :     0.000006s :     48: predicate.minmaximum_grad
 0.38% :     0.000002s :      1: predicate.mutable_eliminate
 0.09% :     0.000000s :      1: predicate.opt_reshape
 2.16% :     0.000009s :     56: predicate.partial_eliminate
 1.64% :     0.000007s :     48: predicate.print_const_string_wrapper
 2.02% :     0.000008s :     48: predicate.reduce_eliminate
 1.83% :     0.000007s :     55: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.79% :     0.000011s :    104: predicate.replace_applicator
 1.07% :     0.000004s :     45: predicate.replace_old_param
 0.10% :     0.000000s :      1: predicate.reset_defer_inline
 1.78% :     0.000007s :     48: predicate.reshape_eliminate
 1.60% :     0.000007s :     48: predicate.row_tensor_add_zeros_like
 0.21% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.61% :     0.000007s :     48: predicate.same_eliminate
 0.52% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.15% :     0.000001s :      2: predicate.special_op_eliminate
 0.81% :     0.000003s :     21: predicate.specialize_transform
 1.69% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.65% :     0.000007s :     48: predicate.stack_unstack_eliminate
 0.06% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.66% :     0.000011s :     67: predicate.switch_defer_inline
 2.54% :     0.000010s :     67: predicate.switch_layer_defer_inline
 5.92% :     0.000024s :    165: predicate.switch_simplify
 1.68% :     0.000007s :     48: predicate.tile_eliminate
 1.47% :     0.000006s :     48: predicate.transpose_eliminate
 1.84% :     0.000007s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.67% :     0.000007s :     48: predicate.tuple_list_get_item_depend_reorder
 2.99% :     0.000012s :     57: predicate.tuple_list_get_item_eliminator
 2.26% :     0.000009s :     48: predicate.tuple_list_set_item_eliminator
 1.93% :     0.000008s :     55: predicate.tuple_to_list_eliminator_
 1.72% :     0.000007s :     56: predicate.updatestate_pure_node_eliminater
 2.66% :     0.000011s :     77: predicate.updatestate_useless_node_eliminater
 1.96% :     0.000008s :     48: predicate.value_based_eliminate
 0.08% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.10% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.003849    48
73.49% :     0.002829s :     32: func_graph_cloner_run.FuncGraphClonerGraph
26.51% :     0.001020s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.125151    91
 0.03% :     0.000040s :      1: add_recomputation
 0.09% :     0.000110s :      1: auto_monad
 0.01% :     0.000017s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.23% :     0.000288s :      1: bootstrap
 0.02% :     0.000031s :      1: cconv
 0.01% :     0.000009s :      1: convert_after_rewriter
 0.02% :     0.000024s :      1: cse_after_recomputation
 0.01% :     0.000008s :      1: environ_conv
 0.23% :     0.000287s :      1: event_method
 0.00% :     0.000003s :      1: execute
 0.00% :     0.000006s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000010s :      1: graph_reusing
18.65% :     0.023344s :      1: jit_opt_a
 0.11% :     0.000134s :      1: jit_opt_after_cconv
 0.04% :     0.000048s :      1: jit_opt_b
 0.39% :     0.000494s :      1: loop_unroll
 0.65% :     0.000814s :      1: mutable_eliminate
 2.40% :     0.003001s :     39: opt.transform.jit_opt_a
 0.03% :     0.000034s :      4: opt.transform.jit_opt_after_cconv
 0.01% :     0.000017s :      4: opt.transform.jit_opt_b
 0.01% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000078s :      1: opt.transform.mutable_eliminate
 0.01% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.02% :     0.000025s :      4: opt.transform.symbol_engine_opt
 0.41% :     0.000516s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000019s :      1: pre_auto_parallel
 0.04% :     0.000047s :      1: py_interpret_to_execute
 0.01% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.01% :     0.000016s :      1: remove_dup_value
 8.90% :     0.011141s :      2: renormalize.infer
 1.85% :     0.002317s :      2: renormalize.specialize
 0.01% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.03% :     0.000038s :      1: rewriter_after_opt_a
 0.09% :     0.000111s :      1: rewriter_before_opt_a
 0.06% :     0.000072s :      1: symbol_engine_optimizer
 0.02% :     0.000026s :      1: task_emit
65.44% :     0.081895s :      1: type_inference
 0.04% :     0.000052s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:20.150.860 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c368a60> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:20.151.581 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0189572, [33]
[bootstrap]: 0.00028517
[type_inference]: 0.00443185
[event_method]: 1.765e-05
[auto_monad]: 6.437e-05
[graph_reusing]: 6.04001e-06
[pre_auto_parallel]: 2.44001e-06
[py_interpret_to_execute]: 3.54e-05
[rewriter_before_opt_a]: 7.266e-05
[expand_dump_flag]: 3.23998e-06
[jit_opt_a]: 0.00528574, [2]
    [Cycle 1]: 0.00163572, [27]
        [switch_simplify]: 5.527e-05
        [loop_unroll]: 2.947e-05
        [a_1]: 0.00054787
        [with_stream_mark]: 2.119e-05
        [recompute_prepare]: 8.05e-06
        [updatestate_depend_eliminate]: 4.28001e-06
        [updatestate_assign_eliminate]: 3.19001e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [parameter_eliminate]: 2.17001e-06
        [specialize_transform]: 6.20002e-06
        [updatestate_useless_node_eliminater]: 5.17e-06
        [accelerated_algorithm]: 5.90002e-06
        [meta_shard_fg_expand]: 2.17999e-06
        [get_grad_eliminate_]: 6.11998e-06
        [merge_forward]: 3.85998e-06
        [cell_reuse_recompute_pass]: 1.39998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.546e-05
        [j_node_and_user_rematch]: 9.63002e-06
        [meta_fg_expand]: 2.46e-06
        [replace_old_param]: 8.95001e-06
        [inline_without_move]: 5.64998e-06
        [renormalize]: 0.00064511
        [add_forward_monad_depend]: 5.59e-06
        [auto_monad_grad]: 2.44999e-06
        [auto_monad_eliminator]: 1.568e-05
        [cse]: 2.831e-05
        [replace_applicator]: 1.355e-05
    [Cycle 2]: 0.00033064, [27]
        [switch_simplify]: 5.91e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 8.858e-05
        [with_stream_mark]: 1.02e-05
        [recompute_prepare]: 5.25999e-06
        [updatestate_depend_eliminate]: 2.93e-06
        [updatestate_assign_eliminate]: 2.18002e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 5.11002e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 5.25001e-06
        [meta_shard_fg_expand]: 1.40999e-06
        [get_grad_eliminate_]: 4.61002e-06
        [merge_forward]: 3.01001e-06
        [cell_reuse_recompute_pass]: 1.81e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.312e-05
        [j_node_and_user_rematch]: 7.71999e-06
        [meta_fg_expand]: 2.07001e-06
        [replace_old_param]: 1.325e-05
        [inline_without_move]: 4.97999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.88002e-06
        [auto_monad_grad]: 9.40025e-07
        [auto_monad_eliminator]: 5.43002e-06
        [cse]: 1.125e-05
        [replace_applicator]: 5.46998e-06
[py_interpret_to_execute_after_opt_a]: 1.034e-05
[rewriter_after_opt_a]: 3.432e-05
[convert_after_rewriter]: 7.65e-06
[order_py_execute_after_rewriter]: 5.23002e-06
[mutable_eliminate]: 0.00078786
[jit_opt_b]: 5.262e-05, [1]
    [Cycle 1]: 4.471e-05, [2]
        [frontend_op_eliminate]: 1.718e-05
        [inline_after_opt_a]: 1.53e-05
[cconv]: 2.479e-05
[loop_unroll]: 0.0004354
[jit_opt_after_cconv]: 0.00014395, [1]
    [Cycle 1]: 0.00013729, [11]
        [c_1]: 2.052e-05
        [parameter_eliminate]: 2.68e-06
        [updatestate_depend_eliminate]: 6.09999e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [cse]: 2.074e-05
        [call_graph_tuple_transform]: 1.933e-05
        [tuple_list_get_item_eliminator]: 5.57001e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 5.62999e-06
[remove_dup_value]: 1.392e-05
[partial_unused_args_eliminate]: 2.11998e-06
[environ_conv]: 4.96997e-06
[add_recomputation]: 4.961e-05
[cse_after_recomputation]: 2.06e-05, [1]
    [Cycle 1]: 1.501e-05, [1]
        [cse]: 8.97999e-06
[auto_monad_reorder]: 1.548e-05
[get_jit_bprop_graph]: 1.75001e-06
[rewriter_after_jit_bprop_graph]: 4.72e-06
[opt_after_jit_grad]: 0.0004543
[symbol_engine_optimizer]: 7.216e-05, [1]
    [Cycle 1]: 6.499e-05, [6]
        [build]: 3.28998e-06
        [elim_shapecalc]: 7.46001e-06
        [elim_not_effective]: 1.197e-05
        [opt_reshape]: 5.74999e-06
        [fold_const_symbol]: 8.54e-06
        [renormalize]: 7.10017e-07
[validate]: 3.482e-05
[backend_pass]: 1.22e-06
[task_emit]: 0.00635085
[execute]: 7.78999e-06
Sums
    bootstrap                            :     0.000285s :  1.90%
    type_inference                       :     0.004432s : 29.53%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000064s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.24%
    rewriter_before_opt_a                :     0.000073s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000636s :  4.24%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000022s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000645s :  4.30%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.14%
    jit_opt_a.cse                        :     0.000040s :  0.26%
    jit_opt_a.replace_applicator         :     0.000019s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000788s :  5.25%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000435s :  2.90%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000050s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000454s :  3.03%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006351s : 42.32%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000225    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.63% :     0.000006s :      3: substitution.graph_param_transform
81.94% :     0.000184s :      4: substitution.inline
 1.27% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.17% :     0.000005s :      4: substitution.remove_not_recompute_node
 4.51% :     0.000010s :      2: substitution.replace_old_param
 6.04% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004369     2
80.78% :     0.003529s :      1: type_inference.infer
19.22% :     0.000840s :      1: type_inference.specialize
------[replace.]   0.000071     6
67.44% :     0.000048s :      4: replace.inline
32.56% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000194     6
93.55% :     0.000181s :      4: match.inline
 6.45% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.35% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.47% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.35% :     0.000002s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.08% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000000s :      3: predicate.graph_param_transform
 5.26% :     0.000006s :     22: predicate.inline
 1.12% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.55% :     0.000002s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.92% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.51% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.37% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.95% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.49% :     0.000002s :      6: predicate.special_op_eliminate
 1.22% :     0.000001s :      6: predicate.specialize_transform
 1.25% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.76% :     0.000009s :     53: predicate.switch_simplify
 1.39% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.80% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.95% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000540    12
46.24% :     0.000250s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.66% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.11% :     0.000249s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020558    76
 0.26% :     0.000052s :      1: add_recomputation
 0.33% :     0.000068s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.43% :     0.000295s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.73% :     0.005289s :      1: jit_opt_a
 0.71% :     0.000147s :      1: jit_opt_after_cconv
 0.27% :     0.000055s :      1: jit_opt_b
 2.16% :     0.000443s :      1: loop_unroll
 3.89% :     0.000800s :      1: mutable_eliminate
 4.16% :     0.000855s :     26: opt.transform.jit_opt_a
 0.23% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.25% :     0.000463s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000038s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.67% :     0.000344s :      1: renormalize.infer
 1.43% :     0.000294s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000037s :      1: rewriter_after_opt_a
 0.37% :     0.000076s :      1: rewriter_before_opt_a
 0.36% :     0.000075s :      1: symbol_engine_optimizer
30.98% :     0.006368s :      1: task_emit
21.63% :     0.004447s :      1: type_inference
 0.32% :     0.000066s :      1: validate


TotalTime = 0.011086, [30]
[bootstrap]: 0.00027401
[type_inference]: 0.00394163
[event_method]: 1.668e-05
[auto_monad]: 6.077e-05
[graph_reusing]: 5.62999e-06
[pre_auto_parallel]: 2.30002e-06
[py_interpret_to_execute]: 2.861e-05
[rewriter_before_opt_a]: 7.095e-05
[expand_dump_flag]: 2.79999e-06
[jit_opt_a]: 0.00469348, [2]
    [Cycle 1]: 0.00136095, [27]
        [switch_simplify]: 4.768e-05
        [loop_unroll]: 2.88e-05
        [a_1]: 0.0004964
        [with_stream_mark]: 1.47e-05
        [recompute_prepare]: 6.71e-06
        [updatestate_depend_eliminate]: 3.41001e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 2.63998e-06
        [parameter_eliminate]: 1.97001e-06
        [specialize_transform]: 5.99999e-06
        [updatestate_useless_node_eliminater]: 5.24998e-06
        [accelerated_algorithm]: 5.86998e-06
        [meta_shard_fg_expand]: 2.01998e-06
        [get_grad_eliminate_]: 5.60001e-06
        [merge_forward]: 3.85998e-06
        [cell_reuse_recompute_pass]: 1.10999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.499e-05
        [j_node_and_user_rematch]: 8.52e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 9.17001e-06
        [inline_without_move]: 6.17001e-06
        [renormalize]: 0.00045729
        [add_forward_monad_depend]: 5.09998e-06
        [auto_monad_grad]: 1.96998e-06
        [auto_monad_eliminator]: 1.392e-05
        [cse]: 2.675e-05
        [replace_applicator]: 1.167e-05
    [Cycle 2]: 0.0003259, [27]
        [switch_simplify]: 5.97001e-06
        [loop_unroll]: 5.06002e-06
        [a_1]: 9.46e-05
        [with_stream_mark]: 8.03999e-06
        [recompute_prepare]: 5.25001e-06
        [updatestate_depend_eliminate]: 2.55002e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 8.50006e-07
        [specialize_transform]: 5.17999e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.02e-06
        [meta_shard_fg_expand]: 1.00999e-06
        [get_grad_eliminate_]: 4.58999e-06
        [merge_forward]: 2.39999e-06
        [cell_reuse_recompute_pass]: 1.11002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.197e-05
        [j_node_and_user_rematch]: 8.33001e-06
        [meta_fg_expand]: 1.49998e-06
        [replace_old_param]: 7e-06
        [inline_without_move]: 4.49002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.24003e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 5.24e-06
        [cse]: 1.047e-05
        [replace_applicator]: 5.76e-06
[py_interpret_to_execute_after_opt_a]: 7.86001e-06
[rewriter_after_opt_a]: 3.047e-05
[convert_after_rewriter]: 6.29001e-06
[order_py_execute_after_rewriter]: 4.67e-06
[mutable_eliminate]: 0.00046584
[jit_opt_b]: 4.954e-05, [1]
    [Cycle 1]: 4.318e-05, [2]
        [frontend_op_eliminate]: 1.68e-05
        [inline_after_opt_a]: 1.504e-05
[cconv]: 2.367e-05
[loop_unroll]: 0.00043829
[jit_opt_after_cconv]: 0.00013593, [1]
    [Cycle 1]: 0.00012952, [11]
        [c_1]: 2.027e-05
        [parameter_eliminate]: 2.81e-06
        [updatestate_depend_eliminate]: 5.30001e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.599e-05
        [call_graph_tuple_transform]: 1.832e-05
        [tuple_list_get_item_eliminator]: 5.29998e-06
        [none_parameter_eliminate]: 1.42999e-06
        [renormalize]: 4.10015e-07
        [switch_simplify]: 5.46e-06
[remove_dup_value]: 1.391e-05
[partial_unused_args_eliminate]: 2.22999e-06
[environ_conv]: 4.57e-06
[add_recomputation]: 4.506e-05
[cse_after_recomputation]: 1.999e-05, [1]
    [Cycle 1]: 1.442e-05, [1]
        [cse]: 8.82e-06
[auto_monad_reorder]: 1.457e-05
[get_jit_bprop_graph]: 1.49e-06
[rewriter_after_jit_bprop_graph]: 5.04e-06
[opt_after_jit_grad]: 0.00045551
[symbol_engine_optimizer]: 7.477e-05, [1]
    [Cycle 1]: 6.826e-05, [6]
        [build]: 3.63e-06
        [elim_shapecalc]: 7.31999e-06
        [elim_not_effective]: 1.25e-05
        [opt_reshape]: 5.82001e-06
        [fold_const_symbol]: 9.47001e-06
        [renormalize]: 4.39992e-07
[validate]: 2.853e-05
Sums
    bootstrap                            :     0.000274s :  3.65%
    type_inference                       :     0.003942s : 52.48%
    event_method                         :     0.000017s :  0.22%
    auto_monad                           :     0.000061s :  0.81%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000029s :  0.38%
    rewriter_before_opt_a                :     0.000071s :  0.94%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000054s :  0.71%
    jit_opt_a.loop_unroll                :     0.000034s :  0.45%
    jit_opt_a.a_1                        :     0.000591s :  7.87%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.14%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.36%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.22%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000457s :  6.09%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.26%
    jit_opt_a.cse                        :     0.000037s :  0.50%
    jit_opt_a.replace_applicator         :     0.000017s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.10%
    rewriter_after_opt_a                 :     0.000030s :  0.41%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000466s :  6.20%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.20%
    cconv                                :     0.000024s :  0.32%
    loop_unroll                          :     0.000438s :  5.84%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.19%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000045s :  0.60%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000015s :  0.19%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000456s :  6.06%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.13%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000029s :  0.38%


Time group info:
------[substitution.]   0.000171    23
 1.28% :     0.000002s :      2: substitution.elim_not_effective
 0.90% :     0.000002s :      2: substitution.fold_const_symbol
 3.10% :     0.000005s :      3: substitution.graph_param_transform
81.49% :     0.000139s :      4: substitution.inline
 1.62% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.87% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000004s :      2: substitution.replace_old_param
 6.67% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003877     2
80.81% :     0.003133s :      1: type_inference.infer
19.19% :     0.000744s :      1: type_inference.specialize
------[replace.]   0.000061     6
67.99% :     0.000041s :      4: replace.inline
32.01% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000147     6
93.08% :     0.000137s :      4: match.inline
 6.92% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.57% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.19% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 1.84% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000001s :     10: predicate.depend_value_elim
 1.29% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.67% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000002s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.22% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.39% :     0.000007s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.18% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.92% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.47% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     15: predicate.partial_eliminate
 1.44% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.66% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.60% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.52% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.36% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.16% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.26% :     0.000002s :     10: predicate.transpose_eliminate
 1.26% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.72% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.84% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.17% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.38% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000511    12
51.96% :     0.000266s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.90% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.14% :     0.000221s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012401    72
 0.39% :     0.000048s :      1: add_recomputation
 0.52% :     0.000064s :      1: auto_monad
 0.14% :     0.000017s :      1: auto_monad_reorder
 2.35% :     0.000292s :      1: bootstrap
 0.21% :     0.000026s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
37.87% :     0.004696s :      1: jit_opt_a
 1.12% :     0.000139s :      1: jit_opt_after_cconv
 0.42% :     0.000052s :      1: jit_opt_b
 3.60% :     0.000447s :      1: loop_unroll
 3.83% :     0.000474s :      1: mutable_eliminate
 6.40% :     0.000794s :     26: opt.transform.jit_opt_a
 0.37% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.74% :     0.000463s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000031s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000016s :      1: remove_dup_value
 1.82% :     0.000226s :      1: renormalize.infer
 1.81% :     0.000225s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000033s :      1: rewriter_after_opt_a
 0.59% :     0.000074s :      1: rewriter_before_opt_a
 0.62% :     0.000077s :      1: symbol_engine_optimizer
31.88% :     0.003953s :      1: type_inference


TotalTime = 0.0177278, [33]
[bootstrap]: 0.00027184
[type_inference]: 0.00419059
[event_method]: 1.754e-05
[auto_monad]: 6.7e-05
[graph_reusing]: 5.05999e-06
[pre_auto_parallel]: 2.26998e-06
[py_interpret_to_execute]: 3.459e-05
[rewriter_before_opt_a]: 6.993e-05
[expand_dump_flag]: 2.86e-06
[jit_opt_a]: 0.00519908, [2]
    [Cycle 1]: 0.00164015, [27]
        [switch_simplify]: 4.993e-05
        [loop_unroll]: 2.95e-05
        [a_1]: 0.0005145
        [with_stream_mark]: 1.7e-05
        [recompute_prepare]: 7.06999e-06
        [updatestate_depend_eliminate]: 3.26999e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.56998e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 5.06002e-06
        [accelerated_algorithm]: 5.08002e-06
        [meta_shard_fg_expand]: 2.19001e-06
        [get_grad_eliminate_]: 5.25001e-06
        [merge_forward]: 3.53e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.364e-05
        [j_node_and_user_rematch]: 8.59002e-06
        [meta_fg_expand]: 2.13998e-06
        [replace_old_param]: 8.64003e-06
        [inline_without_move]: 4.90999e-06
        [renormalize]: 0.00071497
        [add_forward_monad_depend]: 6.12001e-06
        [auto_monad_grad]: 2.41e-06
        [auto_monad_eliminator]: 1.317e-05
        [cse]: 1.91e-05
        [replace_applicator]: 1.283e-05
    [Cycle 2]: 0.0003191, [27]
        [switch_simplify]: 5.82999e-06
        [loop_unroll]: 4.89e-06
        [a_1]: 8.895e-05
        [with_stream_mark]: 9.24e-06
        [recompute_prepare]: 4.90001e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.75999e-06
        [updatestate_useless_node_eliminater]: 4.65001e-06
        [accelerated_algorithm]: 5.16002e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 4.62998e-06
        [merge_forward]: 2.39001e-06
        [cell_reuse_recompute_pass]: 1.50999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.269e-05
        [j_node_and_user_rematch]: 7.65998e-06
        [meta_fg_expand]: 1.35999e-06
        [replace_old_param]: 7.21001e-06
        [inline_without_move]: 4.58001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 4.97e-06
        [cse]: 1.02e-05
        [replace_applicator]: 5.74999e-06
[py_interpret_to_execute_after_opt_a]: 1.096e-05
[rewriter_after_opt_a]: 3.366e-05
[convert_after_rewriter]: 6.82002e-06
[order_py_execute_after_rewriter]: 4.72e-06
[mutable_eliminate]: 0.00056153
[jit_opt_b]: 4.968e-05, [1]
    [Cycle 1]: 4.228e-05, [2]
        [frontend_op_eliminate]: 1.572e-05
        [inline_after_opt_a]: 1.49e-05
[cconv]: 2.576e-05
[loop_unroll]: 0.00041551
[jit_opt_after_cconv]: 0.00013733, [1]
    [Cycle 1]: 0.00013114, [11]
        [c_1]: 2.093e-05
        [parameter_eliminate]: 2.44999e-06
        [updatestate_depend_eliminate]: 5.16002e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.08998e-06
        [cse]: 1.747e-05
        [call_graph_tuple_transform]: 1.827e-05
        [tuple_list_get_item_eliminator]: 5.37001e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 5.34e-06
[remove_dup_value]: 1.297e-05
[partial_unused_args_eliminate]: 2.39001e-06
[environ_conv]: 5.05001e-06
[add_recomputation]: 4.6e-05
[cse_after_recomputation]: 1.902e-05, [1]
    [Cycle 1]: 1.394e-05, [1]
        [cse]: 8.62e-06
[auto_monad_reorder]: 1.494e-05
[get_jit_bprop_graph]: 1.89999e-06
[rewriter_after_jit_bprop_graph]: 5.43002e-06
[opt_after_jit_grad]: 0.00045235
[symbol_engine_optimizer]: 7.113e-05, [1]
    [Cycle 1]: 6.5e-05, [6]
        [build]: 3.11999e-06
        [elim_shapecalc]: 7.77998e-06
        [elim_not_effective]: 1.22e-05
        [opt_reshape]: 5.97999e-06
        [fold_const_symbol]: 8.40001e-06
        [renormalize]: 4.30009e-07
[validate]: 3.402e-05
[backend_pass]: 8.79983e-07
[task_emit]: 0.00575166
[execute]: 8.23999e-06
Sums
    bootstrap                            :     0.000272s :  1.96%
    type_inference                       :     0.004191s : 30.19%
    event_method                         :     0.000018s :  0.13%
    auto_monad                           :     0.000067s :  0.48%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.25%
    rewriter_before_opt_a                :     0.000070s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.40%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000603s :  4.35%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000715s :  5.15%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000029s :  0.21%
    jit_opt_a.replace_applicator         :     0.000019s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.08%
    rewriter_after_opt_a                 :     0.000034s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000562s :  4.05%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000026s :  0.19%
    loop_unroll                          :     0.000416s :  2.99%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000046s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000452s :  3.26%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005752s : 41.44%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000189    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.59% :     0.000005s :      3: substitution.graph_param_transform
82.83% :     0.000157s :      4: substitution.inline
 1.42% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.53% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.20% :     0.000004s :      2: substitution.replace_old_param
 6.65% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004127     2
80.68% :     0.003330s :      1: type_inference.infer
19.32% :     0.000797s :      1: type_inference.specialize
------[replace.]   0.000065     6
67.18% :     0.000043s :      4: replace.inline
32.82% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000166     6
93.06% :     0.000154s :      4: match.inline
 6.94% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.31% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000001s :     10: predicate.addn_zero_filter
 2.19% :     0.000002s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.14% :     0.000001s :      6: predicate.get_grad_eliminate
 0.44% :     0.000001s :      3: predicate.graph_param_transform
 5.80% :     0.000007s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.66% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.48% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.52% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.73% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.89% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.33% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.56% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.59% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000537    12
47.02% :     0.000253s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.05% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.93% :     0.000252s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019349    76
 0.25% :     0.000049s :      1: add_recomputation
 0.36% :     0.000070s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.48% :     0.000287s :      1: bootstrap
 0.15% :     0.000028s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.89% :     0.005203s :      1: jit_opt_a
 0.72% :     0.000140s :      1: jit_opt_after_cconv
 0.27% :     0.000052s :      1: jit_opt_b
 2.19% :     0.000424s :      1: loop_unroll
 2.95% :     0.000570s :      1: mutable_eliminate
 4.17% :     0.000806s :     26: opt.transform.jit_opt_a
 0.24% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.38% :     0.000461s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000037s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.55% :     0.000299s :      1: renormalize.infer
 2.11% :     0.000407s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000037s :      1: rewriter_after_opt_a
 0.38% :     0.000073s :      1: rewriter_before_opt_a
 0.38% :     0.000074s :      1: symbol_engine_optimizer
29.80% :     0.005765s :      1: task_emit
21.73% :     0.004205s :      1: type_inference
 0.28% :     0.000054s :      1: validate


TotalTime = 0.0104378, [30]
[bootstrap]: 0.00023217
[type_inference]: 0.0035794
[event_method]: 1.55e-05
[auto_monad]: 4.959e-05
[graph_reusing]: 5.24e-06
[pre_auto_parallel]: 2.39001e-06
[py_interpret_to_execute]: 2.644e-05
[rewriter_before_opt_a]: 6.358e-05
[expand_dump_flag]: 2.17999e-06
[jit_opt_a]: 0.00458112, [2]
    [Cycle 1]: 0.00132847, [27]
        [switch_simplify]: 4.565e-05
        [loop_unroll]: 3.131e-05
        [a_1]: 0.00047796
        [with_stream_mark]: 1.206e-05
        [recompute_prepare]: 7.18998e-06
        [updatestate_depend_eliminate]: 2.99001e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [parameter_eliminate]: 1.66e-06
        [specialize_transform]: 6.54001e-06
        [updatestate_useless_node_eliminater]: 4.98001e-06
        [accelerated_algorithm]: 5.17999e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 5.35999e-06
        [merge_forward]: 2.59999e-06
        [cell_reuse_recompute_pass]: 9.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.201e-05
        [j_node_and_user_rematch]: 9.05999e-06
        [meta_fg_expand]: 1.94e-06
        [replace_old_param]: 8.26002e-06
        [inline_without_move]: 5.27999e-06
        [renormalize]: 0.00047747
        [add_forward_monad_depend]: 4.63001e-06
        [auto_monad_grad]: 1.53002e-06
        [auto_monad_eliminator]: 1.039e-05
        [cse]: 1.712e-05
        [replace_applicator]: 1.083e-05
    [Cycle 2]: 0.00031614, [27]
        [switch_simplify]: 5.09e-06
        [loop_unroll]: 4.76002e-06
        [a_1]: 9.115e-05
        [with_stream_mark]: 8.02998e-06
        [recompute_prepare]: 5.01002e-06
        [updatestate_depend_eliminate]: 2.24001e-06
        [updatestate_assign_eliminate]: 1.93997e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 9.20001e-07
        [specialize_transform]: 4.74e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 4.99e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 4.70999e-06
        [merge_forward]: 2.34999e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.183e-05
        [j_node_and_user_rematch]: 7.69002e-06
        [meta_fg_expand]: 1.52999e-06
        [replace_old_param]: 6.98e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 7.99977e-07
        [auto_monad_eliminator]: 4.65001e-06
        [cse]: 1.048e-05
        [replace_applicator]: 5.07e-06
[py_interpret_to_execute_after_opt_a]: 7.69002e-06
[rewriter_after_opt_a]: 2.563e-05
[convert_after_rewriter]: 6.02999e-06
[order_py_execute_after_rewriter]: 4.48001e-06
[mutable_eliminate]: 0.00045891
[jit_opt_b]: 4.853e-05, [1]
    [Cycle 1]: 4.234e-05, [2]
        [frontend_op_eliminate]: 1.619e-05
        [inline_after_opt_a]: 1.466e-05
[cconv]: 1.872e-05
[loop_unroll]: 0.00040657
[jit_opt_after_cconv]: 0.00013425, [1]
    [Cycle 1]: 0.00012813, [11]
        [c_1]: 2.047e-05
        [parameter_eliminate]: 2.46998e-06
        [updatestate_depend_eliminate]: 5.17e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.26998e-06
        [cse]: 1.605e-05
        [call_graph_tuple_transform]: 1.696e-05
        [tuple_list_get_item_eliminator]: 5.35001e-06
        [none_parameter_eliminate]: 1.20001e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 5.66e-06
[remove_dup_value]: 1.019e-05
[partial_unused_args_eliminate]: 1.79e-06
[environ_conv]: 3.82998e-06
[add_recomputation]: 3.485e-05
[cse_after_recomputation]: 2.005e-05, [1]
    [Cycle 1]: 1.468e-05, [1]
        [cse]: 8.65999e-06
[auto_monad_reorder]: 1.051e-05
[get_jit_bprop_graph]: 1.57001e-06
[rewriter_after_jit_bprop_graph]: 4.82e-06
[opt_after_jit_grad]: 0.00043791
[symbol_engine_optimizer]: 7.165e-05, [1]
    [Cycle 1]: 6.575e-05, [6]
        [build]: 2.75002e-06
        [elim_shapecalc]: 7.4e-06
        [elim_not_effective]: 1.268e-05
        [opt_reshape]: 5.91e-06
        [fold_const_symbol]: 8.83001e-06
        [renormalize]: 4.40021e-07
[validate]: 2.507e-05
Sums
    bootstrap                            :     0.000232s :  3.34%
    type_inference                       :     0.003579s : 51.43%
    event_method                         :     0.000016s :  0.22%
    auto_monad                           :     0.000050s :  0.71%
    graph_reusing                        :     0.000005s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000026s :  0.38%
    rewriter_before_opt_a                :     0.000064s :  0.91%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000051s :  0.73%
    jit_opt_a.loop_unroll                :     0.000036s :  0.52%
    jit_opt_a.a_1                        :     0.000569s :  8.18%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.18%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.24%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.22%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000478s :  6.86%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.22%
    jit_opt_a.cse                        :     0.000028s :  0.40%
    jit_opt_a.replace_applicator         :     0.000016s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000026s :  0.37%
    convert_after_rewriter               :     0.000006s :  0.09%
    order_py_execute_after_rewriter      :     0.000004s :  0.06%
    mutable_eliminate                    :     0.000459s :  6.59%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.21%
    cconv                                :     0.000019s :  0.27%
    loop_unroll                          :     0.000407s :  5.84%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.29%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.08%
    remove_dup_value                     :     0.000010s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000035s :  0.50%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000011s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000438s :  6.29%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.18%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.13%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000025s :  0.36%


Time group info:
------[substitution.]   0.000154    23
 1.20% :     0.000002s :      2: substitution.elim_not_effective
 0.90% :     0.000001s :      2: substitution.fold_const_symbol
 2.71% :     0.000004s :      3: substitution.graph_param_transform
83.07% :     0.000128s :      4: substitution.inline
 1.80% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.71% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.14% :     0.000003s :      2: substitution.replace_old_param
 5.46% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003534     2
80.15% :     0.002832s :      1: type_inference.infer
19.85% :     0.000701s :      1: type_inference.specialize
------[replace.]   0.000061     6
65.65% :     0.000040s :      4: replace.inline
34.35% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000132     6
94.50% :     0.000125s :      4: match.inline
 5.50% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 1.90% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000001s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.46% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.08% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.10% :     0.000006s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.80% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.75% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.48% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.79% :     0.000002s :     10: predicate.reduce_eliminate
 1.81% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.26% :     0.000001s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.79% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 2.51% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.86% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.51% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.41% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.18% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.83% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000467    12
48.45% :     0.000226s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.24% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.31% :     0.000216s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011755    72
 0.32% :     0.000038s :      1: add_recomputation
 0.45% :     0.000053s :      1: auto_monad
 0.11% :     0.000013s :      1: auto_monad_reorder
 2.08% :     0.000244s :      1: bootstrap
 0.18% :     0.000021s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.19% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000006s :      1: environ_conv
 0.16% :     0.000019s :      1: event_method
 0.04% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.07% :     0.000008s :      1: graph_reusing
38.99% :     0.004584s :      1: jit_opt_a
 1.16% :     0.000137s :      1: jit_opt_after_cconv
 0.43% :     0.000051s :      1: jit_opt_b
 3.52% :     0.000414s :      1: loop_unroll
 3.97% :     0.000467s :      1: mutable_eliminate
 6.53% :     0.000767s :     26: opt.transform.jit_opt_a
 0.38% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.27% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.79% :     0.000445s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.04% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000029s :      1: py_interpret_to_execute
 0.09% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000013s :      1: remove_dup_value
 2.12% :     0.000249s :      1: renormalize.infer
 1.89% :     0.000222s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000029s :      1: rewriter_after_opt_a
 0.56% :     0.000066s :      1: rewriter_before_opt_a
 0.63% :     0.000074s :      1: symbol_engine_optimizer
30.54% :     0.003590s :      1: type_inference


TotalTime = 0.0204064, [33]
[bootstrap]: 0.00030683
[type_inference]: 0.00495507
[event_method]: 1.807e-05
[auto_monad]: 6.534e-05
[graph_reusing]: 5.89999e-06
[pre_auto_parallel]: 2.56e-06
[py_interpret_to_execute]: 3.657e-05
[rewriter_before_opt_a]: 7.292e-05
[expand_dump_flag]: 3.09999e-06
[jit_opt_a]: 0.00551859, [2]
    [Cycle 1]: 0.00181978, [27]
        [switch_simplify]: 5.525e-05
        [loop_unroll]: 2.969e-05
        [a_1]: 0.00054232
        [with_stream_mark]: 1.819e-05
        [recompute_prepare]: 8.52e-06
        [updatestate_depend_eliminate]: 3.78001e-06
        [updatestate_assign_eliminate]: 2.91999e-06
        [updatestate_loads_eliminate]: 2.93998e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 6.24001e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 6.01998e-06
        [meta_shard_fg_expand]: 2.31e-06
        [get_grad_eliminate_]: 5.49e-06
        [merge_forward]: 3.78999e-06
        [cell_reuse_recompute_pass]: 1.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.469e-05
        [j_node_and_user_rematch]: 9.79e-06
        [meta_fg_expand]: 2.41998e-06
        [replace_old_param]: 9.57001e-06
        [inline_without_move]: 5.42999e-06
        [renormalize]: 0.00078173
        [add_forward_monad_depend]: 6.25997e-06
        [auto_monad_grad]: 2.83003e-06
        [auto_monad_eliminator]: 1.508e-05
        [cse]: 3.014e-05
        [replace_applicator]: 1.445e-05
    [Cycle 2]: 0.00034518, [27]
        [switch_simplify]: 5.91e-06
        [loop_unroll]: 5.17e-06
        [a_1]: 9.566e-05
        [with_stream_mark]: 1.166e-05
        [recompute_prepare]: 5.86003e-06
        [updatestate_depend_eliminate]: 2.84001e-06
        [updatestate_assign_eliminate]: 2.00002e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 1.30999e-06
        [specialize_transform]: 5.27001e-06
        [updatestate_useless_node_eliminater]: 4.81997e-06
        [accelerated_algorithm]: 5.74e-06
        [meta_shard_fg_expand]: 1.41998e-06
        [get_grad_eliminate_]: 4.75999e-06
        [merge_forward]: 2.91999e-06
        [cell_reuse_recompute_pass]: 2.17001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.497e-05
        [j_node_and_user_rematch]: 8.23001e-06
        [meta_fg_expand]: 1.67001e-06
        [replace_old_param]: 1.635e-05
        [inline_without_move]: 4.78001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.40999e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 5.27999e-06
        [cse]: 1.179e-05
        [replace_applicator]: 5.24e-06
[py_interpret_to_execute_after_opt_a]: 1.028e-05
[rewriter_after_opt_a]: 3.647e-05
[convert_after_rewriter]: 6.68e-06
[order_py_execute_after_rewriter]: 5.29998e-06
[mutable_eliminate]: 0.00061766
[jit_opt_b]: 5.694e-05, [1]
    [Cycle 1]: 4.869e-05, [2]
        [frontend_op_eliminate]: 1.932e-05
        [inline_after_opt_a]: 1.661e-05
[cconv]: 2.905e-05
[loop_unroll]: 0.00050635
[jit_opt_after_cconv]: 0.00015806, [1]
    [Cycle 1]: 0.00015045, [11]
        [c_1]: 2.26e-05
        [parameter_eliminate]: 3.14001e-06
        [updatestate_depend_eliminate]: 5.49998e-06
        [updatestate_assign_eliminate]: 2.71999e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [cse]: 2.278e-05
        [call_graph_tuple_transform]: 2.208e-05
        [tuple_list_get_item_eliminator]: 5.91e-06
        [none_parameter_eliminate]: 1.85001e-06
        [renormalize]: 9.5999e-07
        [switch_simplify]: 5.53002e-06
[remove_dup_value]: 1.38e-05
[partial_unused_args_eliminate]: 2.06e-06
[environ_conv]: 6.26e-06
[add_recomputation]: 4.886e-05
[cse_after_recomputation]: 2.475e-05, [1]
    [Cycle 1]: 1.756e-05, [1]
        [cse]: 1.109e-05
[auto_monad_reorder]: 1.592e-05
[get_jit_bprop_graph]: 1.89e-06
[rewriter_after_jit_bprop_graph]: 5.84999e-06
[opt_after_jit_grad]: 0.00050984
[symbol_engine_optimizer]: 8.559e-05, [1]
    [Cycle 1]: 7.692e-05, [6]
        [build]: 4.63999e-06
        [elim_shapecalc]: 8.94e-06
        [elim_not_effective]: 1.422e-05
        [opt_reshape]: 6.52001e-06
        [fold_const_symbol]: 1.006e-05
        [renormalize]: 5.3001e-07
[validate]: 3.609e-05
[backend_pass]: 9.30013e-07
[task_emit]: 0.00702059
[execute]: 7.66999e-06
Sums
    bootstrap                            :     0.000307s :  1.88%
    type_inference                       :     0.004955s : 30.30%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000065s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000037s :  0.22%
    rewriter_before_opt_a                :     0.000073s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.37%
    jit_opt_a.loop_unroll                :     0.000035s :  0.21%
    jit_opt_a.a_1                        :     0.000638s :  3.90%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000026s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000782s :  4.78%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.12%
    jit_opt_a.cse                        :     0.000042s :  0.26%
    jit_opt_a.replace_applicator         :     0.000020s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000036s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000618s :  3.78%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.10%
    cconv                                :     0.000029s :  0.18%
    loop_unroll                          :     0.000506s :  3.10%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000049s :  0.30%
    cse_after_recomputation.cse          :     0.000011s :  0.07%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000510s :  3.12%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.007021s : 42.93%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000215    23
 1.10% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.82% :     0.000006s :      3: substitution.graph_param_transform
80.67% :     0.000173s :      4: substitution.inline
 1.45% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.06% :     0.000004s :      4: substitution.remove_not_recompute_node
 5.67% :     0.000012s :      2: substitution.replace_old_param
 5.59% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004891     2
81.93% :     0.004007s :      1: type_inference.infer
18.07% :     0.000884s :      1: type_inference.specialize
------[replace.]   0.000068     6
65.65% :     0.000045s :      4: replace.inline
34.35% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000182     6
94.05% :     0.000171s :      4: match.inline
 5.95% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.27% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.99% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 1.82% :     0.000002s :     10: predicate.arithmetic_simplify
 1.15% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.74% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.80% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000002s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000000s :      3: predicate.graph_param_transform
 6.05% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.64% :     0.000002s :      6: predicate.less_batch_normalization
 1.83% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.24% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.47% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.36% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.10% :     0.000003s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.27% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.03% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.70% :     0.000009s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.75% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.57% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000628    12
49.94% :     0.000313s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.14% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.93% :     0.000276s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022169    76
 0.24% :     0.000054s :      1: add_recomputation
 0.31% :     0.000069s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.44% :     0.000319s :      1: bootstrap
 0.14% :     0.000032s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000027s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
24.91% :     0.005522s :      1: jit_opt_a
 0.73% :     0.000161s :      1: jit_opt_after_cconv
 0.27% :     0.000060s :      1: jit_opt_b
 2.33% :     0.000517s :      1: loop_unroll
 2.84% :     0.000629s :      1: mutable_eliminate
 3.90% :     0.000866s :     26: opt.transform.jit_opt_a
 0.24% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000029s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000036s :      4: opt.transform.symbol_engine_opt
 2.35% :     0.000520s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000039s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 1.86% :     0.000412s :      1: renormalize.infer
 1.63% :     0.000361s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000039s :      1: rewriter_after_opt_a
 0.34% :     0.000076s :      1: rewriter_before_opt_a
 0.40% :     0.000088s :      1: symbol_engine_optimizer
31.72% :     0.007033s :      1: task_emit
22.41% :     0.004968s :      1: type_inference
 0.25% :     0.000056s :      1: validate


TotalTime = 0.0191419, [33]
[bootstrap]: 0.00026924
[type_inference]: 0.00465514
[event_method]: 1.683e-05
[auto_monad]: 6.52e-05
[graph_reusing]: 5.87001e-06
[pre_auto_parallel]: 2.63e-06
[py_interpret_to_execute]: 3.338e-05
[rewriter_before_opt_a]: 7.799e-05
[expand_dump_flag]: 3.24001e-06
[jit_opt_a]: 0.00556251, [2]
    [Cycle 1]: 0.00157252, [27]
        [switch_simplify]: 5.583e-05
        [loop_unroll]: 3.133e-05
        [a_1]: 0.00052115
        [with_stream_mark]: 1.885e-05
        [recompute_prepare]: 8.57e-06
        [updatestate_depend_eliminate]: 3.50998e-06
        [updatestate_assign_eliminate]: 2.96999e-06
        [updatestate_loads_eliminate]: 2.21998e-06
        [parameter_eliminate]: 1.92001e-06
        [specialize_transform]: 6.04001e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 2.31e-06
        [get_grad_eliminate_]: 5.00001e-06
        [merge_forward]: 3.98999e-06
        [cell_reuse_recompute_pass]: 1.06002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.482e-05
        [j_node_and_user_rematch]: 1.017e-05
        [meta_fg_expand]: 2.16e-06
        [replace_old_param]: 8.12e-06
        [inline_without_move]: 5.18002e-06
        [renormalize]: 0.00062168
        [add_forward_monad_depend]: 5.77001e-06
        [auto_monad_grad]: 2.76e-06
        [auto_monad_eliminator]: 1.365e-05
        [cse]: 2.082e-05
        [replace_applicator]: 1.467e-05
    [Cycle 2]: 0.00033383, [27]
        [switch_simplify]: 6.28e-06
        [loop_unroll]: 5.46e-06
        [a_1]: 9.233e-05
        [with_stream_mark]: 1.063e-05
        [recompute_prepare]: 5.20999e-06
        [updatestate_depend_eliminate]: 2.69001e-06
        [updatestate_assign_eliminate]: 2.19999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 5.12999e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 5.60001e-06
        [meta_shard_fg_expand]: 1.78997e-06
        [get_grad_eliminate_]: 4.39998e-06
        [merge_forward]: 2.89999e-06
        [cell_reuse_recompute_pass]: 1.95001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.347e-05
        [j_node_and_user_rematch]: 8.14002e-06
        [meta_fg_expand]: 1.54998e-06
        [replace_old_param]: 7.38e-06
        [inline_without_move]: 4.52998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.43002e-06
        [auto_monad_grad]: 1.02e-06
        [auto_monad_eliminator]: 5.29e-06
        [cse]: 1e-05
        [replace_applicator]: 4.89e-06
[py_interpret_to_execute_after_opt_a]: 1.249e-05
[rewriter_after_opt_a]: 3.476e-05
[convert_after_rewriter]: 7.01001e-06
[order_py_execute_after_rewriter]: 4.86002e-06
[mutable_eliminate]: 0.00062033
[jit_opt_b]: 5.578e-05, [1]
    [Cycle 1]: 4.793e-05, [2]
        [frontend_op_eliminate]: 1.716e-05
        [inline_after_opt_a]: 1.646e-05
[cconv]: 2.666e-05
[loop_unroll]: 0.0004487
[jit_opt_after_cconv]: 0.00014408, [1]
    [Cycle 1]: 0.00013723, [11]
        [c_1]: 2.003e-05
        [parameter_eliminate]: 3.34001e-06
        [updatestate_depend_eliminate]: 6.81999e-06
        [updatestate_assign_eliminate]: 2.56998e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 2.146e-05
        [call_graph_tuple_transform]: 1.869e-05
        [tuple_list_get_item_eliminator]: 5.19e-06
        [none_parameter_eliminate]: 1.40999e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.21002e-06
[remove_dup_value]: 1.341e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 4.47e-06
[add_recomputation]: 4.65e-05
[cse_after_recomputation]: 2.112e-05, [1]
    [Cycle 1]: 1.588e-05, [1]
        [cse]: 9.82999e-06
[auto_monad_reorder]: 1.504e-05
[get_jit_bprop_graph]: 2.07999e-06
[rewriter_after_jit_bprop_graph]: 6.06e-06
[opt_after_jit_grad]: 0.00054094
[symbol_engine_optimizer]: 7.742e-05, [1]
    [Cycle 1]: 6.909e-05, [6]
        [build]: 4.60999e-06
        [elim_shapecalc]: 8.41002e-06
        [elim_not_effective]: 1.266e-05
        [opt_reshape]: 5.52999e-06
        [fold_const_symbol]: 8.64e-06
        [renormalize]: 5.69999e-07
[validate]: 3.869e-05
[backend_pass]: 1.12e-06
[task_emit]: 0.00612339
[execute]: 7.45e-06
Sums
    bootstrap                            :     0.000269s :  1.81%
    type_inference                       :     0.004655s : 31.32%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000065s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.22%
    rewriter_before_opt_a                :     0.000078s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.42%
    jit_opt_a.loop_unroll                :     0.000037s :  0.25%
    jit_opt_a.a_1                        :     0.000613s :  4.13%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000622s :  4.18%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000031s :  0.21%
    jit_opt_a.replace_applicator         :     0.000020s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000035s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000620s :  4.17%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000027s :  0.18%
    loop_unroll                          :     0.000449s :  3.02%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000047s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000541s :  3.64%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006123s : 41.19%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000193    23
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.79% :     0.000005s :      3: substitution.graph_param_transform
82.61% :     0.000159s :      4: substitution.inline
 1.95% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.54% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.98% :     0.000004s :      2: substitution.replace_old_param
 6.55% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004589     2
81.39% :     0.003735s :      1: type_inference.infer
18.61% :     0.000854s :      1: type_inference.specialize
------[replace.]   0.000064     6
66.39% :     0.000042s :      4: replace.inline
33.61% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000168     6
93.11% :     0.000157s :      4: match.inline
 6.89% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.62% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000002s :     10: predicate.addn_zero_filter
 2.36% :     0.000003s :     10: predicate.arithmetic_simplify
 1.56% :     0.000002s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.45% :     0.000002s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 5.47% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.56% :     0.000002s :      6: predicate.less_batch_normalization
 1.89% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.04% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.27% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.11% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000001s :     10: predicate.same_eliminate
 0.76% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.35% :     0.000002s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.29% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.38% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.74% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.07% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.75% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000560    12
50.26% :     0.000281s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.18% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.56% :     0.000249s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020704    76
 0.24% :     0.000050s :      1: add_recomputation
 0.34% :     0.000070s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.34% :     0.000278s :      1: bootstrap
 0.14% :     0.000029s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.88% :     0.005566s :      1: jit_opt_a
 0.71% :     0.000147s :      1: jit_opt_after_cconv
 0.28% :     0.000059s :      1: jit_opt_b
 2.21% :     0.000458s :      1: loop_unroll
 3.04% :     0.000629s :      1: mutable_eliminate
 4.02% :     0.000832s :     26: opt.transform.jit_opt_a
 0.22% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.68% :     0.000554s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000036s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.54% :     0.000318s :      1: renormalize.infer
 1.43% :     0.000296s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000038s :      1: rewriter_after_opt_a
 0.39% :     0.000081s :      1: rewriter_before_opt_a
 0.39% :     0.000080s :      1: symbol_engine_optimizer
29.64% :     0.006137s :      1: task_emit
22.54% :     0.004667s :      1: type_inference
 0.28% :     0.000058s :      1: validate


TotalTime = 0.0115149, [30]
[bootstrap]: 0.00027349
[type_inference]: 0.00410672
[event_method]: 1.659e-05
[auto_monad]: 5.979e-05
[graph_reusing]: 5.52001e-06
[pre_auto_parallel]: 2.69999e-06
[py_interpret_to_execute]: 3.022e-05
[rewriter_before_opt_a]: 7.136e-05
[expand_dump_flag]: 3.11999e-06
[jit_opt_a]: 0.00496655, [2]
    [Cycle 1]: 0.00136683, [27]
        [switch_simplify]: 4.961e-05
        [loop_unroll]: 2.919e-05
        [a_1]: 0.00050263
        [with_stream_mark]: 1.46e-05
        [recompute_prepare]: 7.33999e-06
        [updatestate_depend_eliminate]: 3.46001e-06
        [updatestate_assign_eliminate]: 3.41999e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 2.64001e-06
        [specialize_transform]: 5.81e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.57999e-06
        [meta_shard_fg_expand]: 2.68e-06
        [get_grad_eliminate_]: 5.35001e-06
        [merge_forward]: 3.33e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.34e-05
        [j_node_and_user_rematch]: 8.70001e-06
        [meta_fg_expand]: 2.28002e-06
        [replace_old_param]: 8.17998e-06
        [inline_without_move]: 5.16002e-06
        [renormalize]: 0.00046585
        [add_forward_monad_depend]: 4.60999e-06
        [auto_monad_grad]: 1.64e-06
        [auto_monad_eliminator]: 1.228e-05
        [cse]: 2.582e-05
        [replace_applicator]: 1.062e-05
    [Cycle 2]: 0.00031846, [27]
        [switch_simplify]: 5.49e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 9.182e-05
        [with_stream_mark]: 8.02e-06
        [recompute_prepare]: 5.20001e-06
        [updatestate_depend_eliminate]: 2.64001e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 8.39995e-07
        [specialize_transform]: 5.01002e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 4.96002e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.60001e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.67001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.176e-05
        [j_node_and_user_rematch]: 7.60998e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 7.31001e-06
        [inline_without_move]: 4.79e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 7.80012e-07
        [auto_monad_eliminator]: 4.78001e-06
        [cse]: 1.007e-05
        [replace_applicator]: 4.90001e-06
[py_interpret_to_execute_after_opt_a]: 8.2e-06
[rewriter_after_opt_a]: 3.037e-05
[convert_after_rewriter]: 6.29001e-06
[order_py_execute_after_rewriter]: 4.77e-06
[mutable_eliminate]: 0.00047099
[jit_opt_b]: 4.981e-05, [1]
    [Cycle 1]: 4.322e-05, [2]
        [frontend_op_eliminate]: 1.631e-05
        [inline_after_opt_a]: 1.493e-05
[cconv]: 2.304e-05
[loop_unroll]: 0.00041284
[jit_opt_after_cconv]: 0.00014961, [1]
    [Cycle 1]: 0.00014355, [11]
        [c_1]: 2.025e-05
        [parameter_eliminate]: 2.50002e-06
        [updatestate_depend_eliminate]: 5.44e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 2.86e-05
        [call_graph_tuple_transform]: 1.853e-05
        [tuple_list_get_item_eliminator]: 5.72001e-06
        [none_parameter_eliminate]: 1.72001e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 5.09998e-06
[remove_dup_value]: 1.455e-05
[partial_unused_args_eliminate]: 2.34001e-06
[environ_conv]: 5.05999e-06
[add_recomputation]: 4.719e-05
[cse_after_recomputation]: 2.039e-05, [1]
    [Cycle 1]: 1.461e-05, [1]
        [cse]: 8.82999e-06
[auto_monad_reorder]: 1.387e-05
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 5.20999e-06
[opt_after_jit_grad]: 0.0004543
[symbol_engine_optimizer]: 7.288e-05, [1]
    [Cycle 1]: 6.654e-05, [6]
        [build]: 3.47002e-06
        [elim_shapecalc]: 7.43e-06
        [elim_not_effective]: 1.226e-05
        [opt_reshape]: 6.06e-06
        [fold_const_symbol]: 8.77e-06
        [renormalize]: 4.89992e-07
[validate]: 2.96e-05
Sums
    bootstrap                            :     0.000273s :  3.56%
    type_inference                       :     0.004107s : 53.50%
    event_method                         :     0.000017s :  0.22%
    auto_monad                           :     0.000060s :  0.78%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.04%
    py_interpret_to_execute              :     0.000030s :  0.39%
    rewriter_before_opt_a                :     0.000071s :  0.93%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000055s :  0.72%
    jit_opt_a.loop_unroll                :     0.000034s :  0.44%
    jit_opt_a.a_1                        :     0.000594s :  7.74%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.05%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000466s :  6.07%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.22%
    jit_opt_a.cse                        :     0.000036s :  0.47%
    jit_opt_a.replace_applicator         :     0.000016s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000030s :  0.40%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000471s :  6.14%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000023s :  0.30%
    loop_unroll                          :     0.000413s :  5.38%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000029s :  0.37%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000015s :  0.19%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000047s :  0.61%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000014s :  0.18%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000454s :  5.92%
    symbol_engine_optimizer.build        :     0.000003s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000030s :  0.39%


Time group info:
------[substitution.]   0.000180    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.99% :     0.000005s :      3: substitution.graph_param_transform
83.16% :     0.000150s :      4: substitution.inline
 1.46% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.26% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.86% :     0.000003s :      2: substitution.replace_old_param
 6.49% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004046     2
79.48% :     0.003216s :      1: type_inference.infer
20.52% :     0.000830s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.17% :     0.000042s :      4: replace.inline
32.83% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000158     6
93.31% :     0.000147s :      4: match.inline
 6.69% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.66% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.48% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000001s :     10: predicate.addn_zero_filter
 2.35% :     0.000003s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000001s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.45% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.08% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.19% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.46% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.67% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.46% :     0.000006s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000001s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.51% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.05% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.69% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.64% :     0.000002s :      3: predicate.mutable_eliminate
 0.61% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.87% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.93% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.45% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.13% :     0.000001s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.48% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.60% :     0.000003s :     16: predicate.switch_defer_inline
 2.00% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.41% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.93% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000002s :     10: predicate.value_based_eliminate
 0.45% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000545    12
49.83% :     0.000272s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.01% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.16% :     0.000246s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012839    72
 0.39% :     0.000050s :      1: add_recomputation
 0.49% :     0.000063s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.26% :     0.000290s :      1: bootstrap
 0.20% :     0.000026s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.70% :     0.004969s :      1: jit_opt_a
 1.19% :     0.000153s :      1: jit_opt_after_cconv
 0.41% :     0.000052s :      1: jit_opt_b
 3.28% :     0.000421s :      1: loop_unroll
 3.73% :     0.000479s :      1: mutable_eliminate
 6.17% :     0.000793s :     26: opt.transform.jit_opt_a
 0.36% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.60% :     0.000462s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.04% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000033s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000017s :      1: remove_dup_value
 1.69% :     0.000217s :      1: renormalize.infer
 1.89% :     0.000242s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000033s :      1: rewriter_after_opt_a
 0.58% :     0.000074s :      1: rewriter_before_opt_a
 0.59% :     0.000076s :      1: symbol_engine_optimizer
32.07% :     0.004117s :      1: type_inference


TotalTime = 0.018271, [33]
[bootstrap]: 0.00026966
[type_inference]: 0.0044303
[event_method]: 1.672e-05
[auto_monad]: 6.367e-05
[graph_reusing]: 5.56998e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 3.432e-05
[rewriter_before_opt_a]: 7.623e-05
[expand_dump_flag]: 3.36001e-06
[jit_opt_a]: 0.00522222, [2]
    [Cycle 1]: 0.00146607, [27]
        [switch_simplify]: 4.91e-05
        [loop_unroll]: 2.993e-05
        [a_1]: 0.0004991
        [with_stream_mark]: 1.656e-05
        [recompute_prepare]: 7.11999e-06
        [updatestate_depend_eliminate]: 3.75e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [parameter_eliminate]: 2.47001e-06
        [specialize_transform]: 6.19999e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.29e-06
        [meta_shard_fg_expand]: 1.92001e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 3.4e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.428e-05
        [j_node_and_user_rematch]: 9.07001e-06
        [meta_fg_expand]: 2.61999e-06
        [replace_old_param]: 8.82e-06
        [inline_without_move]: 5.22e-06
        [renormalize]: 0.00055812
        [add_forward_monad_depend]: 5.43002e-06
        [auto_monad_grad]: 2.51e-06
        [auto_monad_eliminator]: 1.407e-05
        [cse]: 2.52e-05
        [replace_applicator]: 1.209e-05
    [Cycle 2]: 0.00031234, [27]
        [switch_simplify]: 5.38002e-06
        [loop_unroll]: 4.70999e-06
        [a_1]: 8.975e-05
        [with_stream_mark]: 9.14998e-06
        [recompute_prepare]: 4.77998e-06
        [updatestate_depend_eliminate]: 2.41998e-06
        [updatestate_assign_eliminate]: 1.99999e-06
        [updatestate_loads_eliminate]: 1.77001e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 4.77e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 4.92e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 4.37998e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.72999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.317e-05
        [j_node_and_user_rematch]: 7.97e-06
        [meta_fg_expand]: 1.61002e-06
        [replace_old_param]: 7.36001e-06
        [inline_without_move]: 4.63999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.39998e-06
        [auto_monad_grad]: 9.5999e-07
        [auto_monad_eliminator]: 4.84e-06
        [cse]: 9.87001e-06
        [replace_applicator]: 4.84e-06
[py_interpret_to_execute_after_opt_a]: 1.082e-05
[rewriter_after_opt_a]: 3.585e-05
[convert_after_rewriter]: 7.46999e-06
[order_py_execute_after_rewriter]: 4.78001e-06
[mutable_eliminate]: 0.00058305
[jit_opt_b]: 5.233e-05, [1]
    [Cycle 1]: 4.515e-05, [2]
        [frontend_op_eliminate]: 1.772e-05
        [inline_after_opt_a]: 1.537e-05
[cconv]: 2.509e-05
[loop_unroll]: 0.00042515
[jit_opt_after_cconv]: 0.00015996, [1]
    [Cycle 1]: 0.00015351, [11]
        [c_1]: 2.02e-05
        [parameter_eliminate]: 3.01999e-06
        [updatestate_depend_eliminate]: 5.57999e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.802e-05
        [call_graph_tuple_transform]: 1.842e-05
        [tuple_list_get_item_eliminator]: 5.59e-06
        [none_parameter_eliminate]: 1.87999e-06
        [renormalize]: 5.39992e-07
        [switch_simplify]: 5.37001e-06
[remove_dup_value]: 1.329e-05
[partial_unused_args_eliminate]: 2.66e-06
[environ_conv]: 4.90001e-06
[add_recomputation]: 4.636e-05
[cse_after_recomputation]: 2.003e-05, [1]
    [Cycle 1]: 1.471e-05, [1]
        [cse]: 9.22001e-06
[auto_monad_reorder]: 1.51e-05
[get_jit_bprop_graph]: 2.24999e-06
[rewriter_after_jit_bprop_graph]: 5.42001e-06
[opt_after_jit_grad]: 0.00045581
[symbol_engine_optimizer]: 7.248e-05, [1]
    [Cycle 1]: 6.642e-05, [6]
        [build]: 4.77e-06
        [elim_shapecalc]: 7.5e-06
        [elim_not_effective]: 1.227e-05
        [opt_reshape]: 5.51e-06
        [fold_const_symbol]: 8.12e-06
        [renormalize]: 4.90021e-07
[validate]: 3.342e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.00597396
[execute]: 8.10999e-06
Sums
    bootstrap                            :     0.000270s :  1.90%
    type_inference                       :     0.004430s : 31.15%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000064s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.24%
    rewriter_before_opt_a                :     0.000076s :  0.54%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.38%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000589s :  4.14%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000558s :  3.92%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000035s :  0.25%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.08%
    rewriter_after_opt_a                 :     0.000036s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000583s :  4.10%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000025s :  0.18%
    loop_unroll                          :     0.000425s :  2.99%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000046s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000456s :  3.21%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005974s : 42.01%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000180    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 3.04% :     0.000005s :      3: substitution.graph_param_transform
81.73% :     0.000147s :      4: substitution.inline
 2.05% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.60% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.24% :     0.000004s :      2: substitution.replace_old_param
 6.57% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004368     2
81.72% :     0.003569s :      1: type_inference.infer
18.28% :     0.000799s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.31% :     0.000042s :      4: replace.inline
32.69% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000156     6
93.06% :     0.000145s :      4: match.inline
 6.94% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.67% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.23% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.85% :     0.000002s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.17% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.18% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      6: predicate.less_batch_normalization
 1.77% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.19% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.90% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.61% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.79% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.05% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.79% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000530    12
50.13% :     0.000265s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.39% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.48% :     0.000236s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019726    76
 0.25% :     0.000049s :      1: add_recomputation
 0.34% :     0.000067s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.43% :     0.000283s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.49% :     0.005225s :      1: jit_opt_a
 0.83% :     0.000163s :      1: jit_opt_after_cconv
 0.28% :     0.000055s :      1: jit_opt_b
 2.20% :     0.000433s :      1: loop_unroll
 3.01% :     0.000594s :      1: mutable_eliminate
 4.01% :     0.000792s :     26: opt.transform.jit_opt_a
 0.23% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.35% :     0.000463s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000037s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.38% :     0.000272s :      1: renormalize.infer
 1.42% :     0.000279s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000039s :      1: rewriter_after_opt_a
 0.40% :     0.000080s :      1: rewriter_before_opt_a
 0.38% :     0.000075s :      1: symbol_engine_optimizer
30.34% :     0.005985s :      1: task_emit
22.53% :     0.004444s :      1: type_inference
 0.26% :     0.000051s :      1: validate


TotalTime = 0.0967252, [33]
[bootstrap]: 0.00023721
[type_inference]: 0.072842
[event_method]: 0.00027429
[auto_monad]: 9.808e-05
[graph_reusing]: 1.85e-05
[pre_auto_parallel]: 3.08e-06
[py_interpret_to_execute]: 2.855e-05
[rewriter_before_opt_a]: 0.00010479
[expand_dump_flag]: 3.46001e-06
[jit_opt_a]: 0.0207243, [3]
    [Cycle 1]: 0.0133301, [27]
        [switch_simplify]: 7.616e-05
        [loop_unroll]: 4.295e-05
        [a_1]: 0.00103219
        [with_stream_mark]: 2.971e-05
        [recompute_prepare]: 2.291e-05
        [updatestate_depend_eliminate]: 9.45001e-06
        [updatestate_assign_eliminate]: 7.48e-06
        [updatestate_loads_eliminate]: 6.71999e-06
        [parameter_eliminate]: 2.89001e-06
        [specialize_transform]: 1.488e-05
        [updatestate_useless_node_eliminater]: 1.413e-05
        [accelerated_algorithm]: 1.354e-05
        [meta_shard_fg_expand]: 3.66001e-06
        [get_grad_eliminate_]: 1.378e-05
        [merge_forward]: 9.56998e-06
        [cell_reuse_recompute_pass]: 1.16002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 3.028e-05
        [j_node_and_user_rematch]: 2.549e-05
        [meta_fg_expand]: 0.00177552
        [replace_old_param]: 7.042e-05
        [inline_without_move]: 6.187e-05
        [renormalize]: 0.00946473
        [add_forward_monad_depend]: 1.213e-05
        [auto_monad_grad]: 6.04999e-06
        [auto_monad_eliminator]: 5.741e-05
        [cse]: 0.00019993
        [replace_applicator]: 8.807e-05
    [Cycle 2]: 0.00353956, [27]
        [switch_simplify]: 4.266e-05
        [loop_unroll]: 4.108e-05
        [a_1]: 0.00120028
        [with_stream_mark]: 1.784e-05
        [recompute_prepare]: 1.026e-05
        [updatestate_depend_eliminate]: 5.09998e-06
        [updatestate_assign_eliminate]: 4.18999e-06
        [updatestate_loads_eliminate]: 3.89002e-06
        [parameter_eliminate]: 1.60001e-06
        [specialize_transform]: 8.50001e-06
        [updatestate_useless_node_eliminater]: 7.6e-06
        [accelerated_algorithm]: 8.05e-06
        [meta_shard_fg_expand]: 2.79999e-06
        [get_grad_eliminate_]: 7.38999e-06
        [merge_forward]: 4.66002e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.6e-05
        [j_node_and_user_rematch]: 1.197e-05
        [meta_fg_expand]: 6.703e-05
        [replace_old_param]: 9.32999e-06
        [inline_without_move]: 7.4e-06
        [renormalize]: 0.00184135
        [add_forward_monad_depend]: 7.7e-06
        [auto_monad_grad]: 2.81e-06
        [auto_monad_eliminator]: 1.206e-05
        [cse]: 2.103e-05
        [replace_applicator]: 1.533e-05
    [Cycle 3]: 0.00024998, [27]
        [switch_simplify]: 5.02e-06
        [loop_unroll]: 4.1e-06
        [a_1]: 4.033e-05
        [with_stream_mark]: 1.006e-05
        [recompute_prepare]: 3.83999e-06
        [updatestate_depend_eliminate]: 2.60002e-06
        [updatestate_assign_eliminate]: 1.81998e-06
        [updatestate_loads_eliminate]: 1.37e-06
        [parameter_eliminate]: 1.33002e-06
        [specialize_transform]: 3.68999e-06
        [updatestate_useless_node_eliminater]: 3.48e-06
        [accelerated_algorithm]: 3.85e-06
        [meta_shard_fg_expand]: 1.27999e-06
        [get_grad_eliminate_]: 3.26999e-06
        [merge_forward]: 2.48002e-06
        [cell_reuse_recompute_pass]: 2.82002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.15e-05
        [j_node_and_user_rematch]: 5.62999e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 3.43999e-06
        [inline_without_move]: 3.08998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.47999e-06
        [auto_monad_grad]: 8.10018e-07
        [auto_monad_eliminator]: 4e-06
        [cse]: 1e-05
        [replace_applicator]: 3.81999e-06
[py_interpret_to_execute_after_opt_a]: 1.243e-05
[rewriter_after_opt_a]: 3.451e-05
[convert_after_rewriter]: 6.68e-06
[order_py_execute_after_rewriter]: 4.42e-06
[mutable_eliminate]: 0.00071693
[jit_opt_b]: 4.224e-05, [1]
    [Cycle 1]: 3.464e-05, [2]
        [frontend_op_eliminate]: 1.129e-05
        [inline_after_opt_a]: 1.067e-05
[cconv]: 2.582e-05
[loop_unroll]: 0.00046698
[jit_opt_after_cconv]: 0.00013062, [1]
    [Cycle 1]: 0.00012285, [11]
        [c_1]: 1.335e-05
        [parameter_eliminate]: 3.53e-06
        [updatestate_depend_eliminate]: 6.17999e-06
        [updatestate_assign_eliminate]: 2.66999e-06
        [updatestate_loads_eliminate]: 1.50999e-06
        [cse]: 2.01e-05
        [call_graph_tuple_transform]: 1.396e-05
        [tuple_list_get_item_eliminator]: 3.55e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 3.83999e-06
[remove_dup_value]: 1.481e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 6.76999e-06
[add_recomputation]: 3.666e-05
[cse_after_recomputation]: 2.27e-05, [1]
    [Cycle 1]: 1.58e-05, [1]
        [cse]: 9.39e-06
[auto_monad_reorder]: 1.353e-05
[get_jit_bprop_graph]: 1.57001e-06
[rewriter_after_jit_bprop_graph]: 7.09001e-06
[opt_after_jit_grad]: 0.00050632
[symbol_engine_optimizer]: 6.77e-05, [1]
    [Cycle 1]: 6.032e-05, [6]
        [build]: 3.53999e-06
        [elim_shapecalc]: 7.19001e-06
        [elim_not_effective]: 1.044e-05
        [opt_reshape]: 4.47e-06
        [fold_const_symbol]: 6.36e-06
        [renormalize]: 3.7998e-07
[validate]: 4.846e-05
[backend_pass]: 1.20999e-06
[task_emit]: 2.61e-05
[execute]: 1.43002e-06
Sums
    bootstrap                            :     0.000237s :  0.26%
    type_inference                       :     0.072842s : 78.93%
    event_method                         :     0.000274s :  0.30%
    auto_monad                           :     0.000098s :  0.11%
    graph_reusing                        :     0.000018s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000029s :  0.03%
    rewriter_before_opt_a                :     0.000105s :  0.11%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000124s :  0.13%
    jit_opt_a.loop_unroll                :     0.000088s :  0.10%
    jit_opt_a.a_1                        :     0.002273s :  2.46%
    jit_opt_a.with_stream_mark           :     0.000058s :  0.06%
    jit_opt_a.recompute_prepare          :     0.000037s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.01%
    jit_opt_a.specialize_transform       :     0.000027s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.03%
    jit_opt_a.accelerated_algorithm      :     0.000025s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.03%
    jit_opt_a.merge_forward              :     0.000017s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000058s :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000043s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.001844s :  2.00%
    jit_opt_a.replace_old_param          :     0.000083s :  0.09%
    jit_opt_a.inline_without_move        :     0.000072s :  0.08%
    jit_opt_a.renormalize                :     0.011306s : 12.25%
    jit_opt_a.add_forward_monad_depend   :     0.000021s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000010s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000073s :  0.08%
    jit_opt_a.cse                        :     0.000231s :  0.25%
    jit_opt_a.replace_applicator         :     0.000107s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.01%
    rewriter_after_opt_a                 :     0.000035s :  0.04%
    convert_after_rewriter               :     0.000007s :  0.01%
    order_py_execute_after_rewriter      :     0.000004s :  0.00%
    mutable_eliminate                    :     0.000717s :  0.78%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.01%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.01%
    cconv                                :     0.000026s :  0.03%
    loop_unroll                          :     0.000467s :  0.51%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.01%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000020s :  0.02%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.02%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.00%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.00%
    remove_dup_value                     :     0.000015s :  0.02%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000007s :  0.01%
    add_recomputation                    :     0.000037s :  0.04%
    cse_after_recomputation.cse          :     0.000009s :  0.01%
    auto_monad_reorder                   :     0.000014s :  0.01%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.01%
    opt_after_jit_grad                   :     0.000506s :  0.55%
    symbol_engine_optimizer.build        :     0.000004s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.01%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.00%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000048s :  0.05%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000026s :  0.03%
    execute                              :     0.000001s :  0.00%


Time group info:
------[substitution.]   0.000618   102
 0.27% :     0.000002s :      1: substitution.elim_not_effective
 0.18% :     0.000001s :      1: substitution.fold_const_symbol
 0.60% :     0.000004s :      1: substitution.graph_param_transform
66.22% :     0.000409s :     12: substitution.inline
 3.46% :     0.000021s :      2: substitution.inline_without_move
 1.44% :     0.000009s :     12: substitution.j_node_and_user_rematch
 1.88% :     0.000012s :      7: substitution.minmaximum_grad
 1.22% :     0.000008s :      5: substitution.partial_eliminate
 1.61% :     0.000010s :     12: substitution.remove_not_recompute_node
 5.60% :     0.000035s :     10: substitution.replace_applicator
 1.01% :     0.000006s :      7: substitution.replace_old_param
 0.99% :     0.000006s :      1: substitution.reshape_eliminate
 0.44% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 1.64% :     0.000010s :      2: substitution.tile_eliminate
 3.71% :     0.000023s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.67% :     0.000017s :      7: substitution.tuple_list_get_item_depend_reorder
 7.05% :     0.000044s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.072752     2
97.04% :     0.070596s :      1: type_inference.infer
 2.96% :     0.002155s :      1: type_inference.specialize
------[replace.]   0.000165    19
64.97% :     0.000107s :     12: replace.inline
35.03% :     0.000058s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000424    19
94.73% :     0.000402s :     12: match.inline
 5.27% :     0.000022s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000389  2646
 1.61% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.30% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.40% :     0.000005s :     46: predicate.addn_check_dump
 1.67% :     0.000007s :     46: predicate.addn_zero_filter
 2.16% :     0.000008s :     46: predicate.arithmetic_simplify
 1.76% :     0.000007s :     46: predicate.cast_eliminate
 0.07% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.41% :     0.000005s :     46: predicate.compare_switch_simplify
 1.91% :     0.000007s :     46: predicate.depend_value_elim
 1.46% :     0.000006s :     46: predicate.dict_get_item_const_eliminator
 1.61% :     0.000006s :     46: predicate.dict_get_item_eliminator
 1.48% :     0.000006s :     46: predicate.dict_set_item_eliminator
 0.22% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.05% :     0.000000s :      1: predicate.elim_not_effective
 0.15% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.53% :     0.000006s :     46: predicate.environ_add_const_eliminate
 1.39% :     0.000005s :     46: predicate.environ_get_add_eliminate
 1.38% :     0.000005s :     46: predicate.environ_get_depend_swap
 1.54% :     0.000006s :     46: predicate.environ_get_eliminate
 1.37% :     0.000005s :     46: predicate.environ_get_set_eliminate
 0.03% :     0.000000s :      1: predicate.fold_const_symbol
 0.86% :     0.000003s :     19: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 4.12% :     0.000016s :     67: predicate.inline
 1.81% :     0.000007s :     43: predicate.inline_without_move
 0.39% :     0.000002s :     19: predicate.j_node_and_user_rematch
 1.13% :     0.000004s :     19: predicate.less_batch_normalization
 1.82% :     0.000007s :     53: predicate.list_to_tuple_eliminator_
 1.79% :     0.000007s :     54: predicate.load_eliminater
 0.35% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.40% :     0.000013s :     95: predicate.loop_unroll_before_grad
 1.83% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.51% :     0.000006s :     46: predicate.merge_addn
 1.49% :     0.000006s :     46: predicate.minmaximum_grad
 0.45% :     0.000002s :      1: predicate.mutable_eliminate
 0.12% :     0.000000s :      1: predicate.opt_reshape
 2.36% :     0.000009s :     54: predicate.partial_eliminate
 1.52% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.97% :     0.000008s :     46: predicate.reduce_eliminate
 1.97% :     0.000008s :     53: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.75% :     0.000011s :    100: predicate.replace_applicator
 1.16% :     0.000005s :     43: predicate.replace_old_param
 0.05% :     0.000000s :      1: predicate.reset_defer_inline
 1.81% :     0.000007s :     46: predicate.reshape_eliminate
 1.69% :     0.000007s :     46: predicate.row_tensor_add_zeros_like
 0.19% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.56% :     0.000006s :     46: predicate.same_eliminate
 0.47% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.24% :     0.000001s :      2: predicate.special_op_eliminate
 0.78% :     0.000003s :     19: predicate.specialize_transform
 1.64% :     0.000006s :     46: predicate.split_environ_get_set_with_tuple_value
 1.75% :     0.000007s :     46: predicate.stack_unstack_eliminate
 0.10% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.94% :     0.000011s :     65: predicate.switch_defer_inline
 2.44% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.34% :     0.000025s :    161: predicate.switch_simplify
 1.62% :     0.000006s :     46: predicate.tile_eliminate
 1.54% :     0.000006s :     46: predicate.transpose_eliminate
 1.75% :     0.000007s :     46: predicate.tuple_list_convert_item_index_to_positive
 1.74% :     0.000007s :     46: predicate.tuple_list_get_item_depend_reorder
 3.04% :     0.000012s :     55: predicate.tuple_list_get_item_eliminator
 1.97% :     0.000008s :     46: predicate.tuple_list_set_item_eliminator
 1.86% :     0.000007s :     53: predicate.tuple_to_list_eliminator_
 1.80% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.66% :     0.000010s :     73: predicate.updatestate_useless_node_eliminater
 1.90% :     0.000007s :     46: predicate.value_based_eliminate
 0.10% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.16% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.003079    42
73.89% :     0.002275s :     26: func_graph_cloner_run.FuncGraphClonerGraph
26.11% :     0.000804s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.111020    91
 0.04% :     0.000039s :      1: add_recomputation
 0.09% :     0.000104s :      1: auto_monad
 0.01% :     0.000016s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.22% :     0.000248s :      1: bootstrap
 0.03% :     0.000028s :      1: cconv
 0.01% :     0.000009s :      1: convert_after_rewriter
 0.02% :     0.000025s :      1: cse_after_recomputation
 0.01% :     0.000009s :      1: environ_conv
 0.25% :     0.000282s :      1: event_method
 0.00% :     0.000005s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000022s :      1: graph_reusing
18.67% :     0.020728s :      1: jit_opt_a
 0.12% :     0.000133s :      1: jit_opt_after_cconv
 0.04% :     0.000045s :      1: jit_opt_b
 0.43% :     0.000476s :      1: loop_unroll
 0.65% :     0.000726s :      1: mutable_eliminate
 2.64% :     0.002934s :     39: opt.transform.jit_opt_a
 0.03% :     0.000031s :      4: opt.transform.jit_opt_after_cconv
 0.01% :     0.000016s :      4: opt.transform.jit_opt_b
 0.01% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.01% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.02% :     0.000017s :      1: opt.transform.opt_after_jit_grad
 0.02% :     0.000024s :      4: opt.transform.symbol_engine_opt
 0.46% :     0.000516s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.00% :     0.000005s :      1: pre_auto_parallel
 0.03% :     0.000031s :      1: py_interpret_to_execute
 0.01% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.02% :     0.000017s :      1: remove_dup_value
 8.20% :     0.009101s :      2: renormalize.infer
 1.97% :     0.002185s :      2: renormalize.specialize
 0.01% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.03% :     0.000037s :      1: rewriter_after_opt_a
 0.10% :     0.000108s :      1: rewriter_before_opt_a
 0.06% :     0.000070s :      1: symbol_engine_optimizer
 0.03% :     0.000031s :      1: task_emit
65.63% :     0.072860s :      1: type_inference
 0.06% :     0.000066s :      1: validate


TotalTime = 0.0107983, [30]
[bootstrap]: 0.0002399
[type_inference]: 0.00372485
[event_method]: 1.613e-05
[auto_monad]: 5.468e-05
[graph_reusing]: 5.02999e-06
[pre_auto_parallel]: 2.08002e-06
[py_interpret_to_execute]: 2.671e-05
[rewriter_before_opt_a]: 6.575e-05
[expand_dump_flag]: 2.63998e-06
[jit_opt_a]: 0.00476328, [2]
    [Cycle 1]: 0.00130537, [27]
        [switch_simplify]: 4.444e-05
        [loop_unroll]: 2.893e-05
        [a_1]: 0.00048252
        [with_stream_mark]: 1.242e-05
        [recompute_prepare]: 7.13e-06
        [updatestate_depend_eliminate]: 3.28e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.39001e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 5.88998e-06
        [updatestate_useless_node_eliminater]: 4.96002e-06
        [accelerated_algorithm]: 5.17e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 5.19e-06
        [merge_forward]: 2.70002e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.225e-05
        [j_node_and_user_rematch]: 8.84e-06
        [meta_fg_expand]: 2.39999e-06
        [replace_old_param]: 7.97e-06
        [inline_without_move]: 5.27001e-06
        [renormalize]: 0.00045002
        [add_forward_monad_depend]: 5.08002e-06
        [auto_monad_grad]: 1.83002e-06
        [auto_monad_eliminator]: 1.195e-05
        [cse]: 1.885e-05
        [replace_applicator]: 1.078e-05
    [Cycle 2]: 0.0003365, [27]
        [switch_simplify]: 5.39998e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 9.008e-05
        [with_stream_mark]: 2.349e-05
        [recompute_prepare]: 5.44e-06
        [updatestate_depend_eliminate]: 2.74001e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 4.98001e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.40999e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.56002e-06
        [merge_forward]: 2.53e-06
        [cell_reuse_recompute_pass]: 1.13001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.22e-05
        [j_node_and_user_rematch]: 7.55e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 7.27997e-06
        [inline_without_move]: 4.69998e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.23002e-06
        [cse]: 1.118e-05
        [replace_applicator]: 4.92999e-06
[py_interpret_to_execute_after_opt_a]: 8.04997e-06
[rewriter_after_opt_a]: 2.708e-05
[convert_after_rewriter]: 5.86e-06
[order_py_execute_after_rewriter]: 6.96001e-06
[mutable_eliminate]: 0.00045328
[jit_opt_b]: 4.961e-05, [1]
    [Cycle 1]: 4.279e-05, [2]
        [frontend_op_eliminate]: 1.662e-05
        [inline_after_opt_a]: 1.498e-05
[cconv]: 2.179e-05
[loop_unroll]: 0.00040978
[jit_opt_after_cconv]: 0.00013373, [1]
    [Cycle 1]: 0.00012721, [11]
        [c_1]: 2.008e-05
        [parameter_eliminate]: 2.36e-06
        [updatestate_depend_eliminate]: 5.15999e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 1.555e-05
        [call_graph_tuple_transform]: 1.718e-05
        [tuple_list_get_item_eliminator]: 5.49e-06
        [none_parameter_eliminate]: 1.41998e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 5.30999e-06
[remove_dup_value]: 1.191e-05
[partial_unused_args_eliminate]: 1.94e-06
[environ_conv]: 4.57e-06
[add_recomputation]: 3.91e-05
[cse_after_recomputation]: 2.042e-05, [1]
    [Cycle 1]: 1.521e-05, [1]
        [cse]: 8.95999e-06
[auto_monad_reorder]: 1.375e-05
[get_jit_bprop_graph]: 1.42e-06
[rewriter_after_jit_bprop_graph]: 4.32e-06
[opt_after_jit_grad]: 0.00043825
[symbol_engine_optimizer]: 7.103e-05, [1]
    [Cycle 1]: 6.497e-05, [6]
        [build]: 3.29001e-06
        [elim_shapecalc]: 7.28e-06
        [elim_not_effective]: 1.231e-05
        [opt_reshape]: 5.61998e-06
        [fold_const_symbol]: 8.09002e-06
        [renormalize]: 4.60015e-07
[validate]: 2.768e-05
Sums
    bootstrap                            :     0.000240s :  3.36%
    type_inference                       :     0.003725s : 52.21%
    event_method                         :     0.000016s :  0.23%
    auto_monad                           :     0.000055s :  0.77%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000027s :  0.37%
    rewriter_before_opt_a                :     0.000066s :  0.92%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000050s :  0.70%
    jit_opt_a.loop_unroll                :     0.000034s :  0.47%
    jit_opt_a.a_1                        :     0.000573s :  8.03%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.50%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.18%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000450s :  6.31%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.24%
    jit_opt_a.cse                        :     0.000030s :  0.42%
    jit_opt_a.replace_applicator         :     0.000016s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000027s :  0.38%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000007s :  0.10%
    mutable_eliminate                    :     0.000453s :  6.35%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.21%
    cconv                                :     0.000022s :  0.31%
    loop_unroll                          :     0.000410s :  5.74%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000012s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000039s :  0.55%
    cse_after_recomputation.cse          :     0.000009s :  0.13%
    auto_monad_reorder                   :     0.000014s :  0.19%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.06%
    opt_after_jit_grad                   :     0.000438s :  6.14%
    symbol_engine_optimizer.build        :     0.000003s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000028s :  0.39%


Time group info:
------[substitution.]   0.000160    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000001s :      2: substitution.fold_const_symbol
 2.94% :     0.000005s :      3: substitution.graph_param_transform
82.35% :     0.000132s :      4: substitution.inline
 1.74% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.37% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.02% :     0.000003s :      2: substitution.replace_old_param
 6.73% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003674     2
80.00% :     0.002940s :      1: type_inference.infer
20.00% :     0.000735s :      1: type_inference.specialize
------[replace.]   0.000061     6
66.13% :     0.000040s :      4: replace.inline
33.87% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000140     6
92.98% :     0.000130s :      4: match.inline
 7.02% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.28% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.04% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 2.58% :     0.000003s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000001s :     10: predicate.depend_value_elim
 1.22% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.64% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.55% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.62% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.21% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.65% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000002s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     10: predicate.reduce_eliminate
 1.84% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 1.05% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.04% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.23% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.29% :     0.000008s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.61% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.82% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000491    12
48.20% :     0.000237s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.07% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.73% :     0.000230s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012082    72
 0.35% :     0.000042s :      1: add_recomputation
 0.48% :     0.000058s :      1: auto_monad
 0.14% :     0.000016s :      1: auto_monad_reorder
 2.08% :     0.000252s :      1: bootstrap
 0.20% :     0.000024s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.19% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
39.44% :     0.004766s :      1: jit_opt_a
 1.13% :     0.000137s :      1: jit_opt_after_cconv
 0.43% :     0.000052s :      1: jit_opt_b
 3.45% :     0.000417s :      1: loop_unroll
 3.82% :     0.000461s :      1: mutable_eliminate
 6.34% :     0.000766s :     26: opt.transform.jit_opt_a
 0.37% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.69% :     0.000446s :      1: opt_after_jit_grad
 0.08% :     0.000009s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000029s :      1: py_interpret_to_execute
 0.09% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000014s :      1: remove_dup_value
 1.75% :     0.000212s :      1: renormalize.infer
 1.92% :     0.000232s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000030s :      1: rewriter_after_opt_a
 0.56% :     0.000068s :      1: rewriter_before_opt_a
 0.61% :     0.000074s :      1: symbol_engine_optimizer
30.91% :     0.003735s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:20.256.014 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c368ee0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:20.256.710 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0191475, [33]
[bootstrap]: 0.00030718
[type_inference]: 0.00464557
[event_method]: 1.796e-05
[auto_monad]: 6.47e-05
[graph_reusing]: 5.49e-06
[pre_auto_parallel]: 2.73e-06
[py_interpret_to_execute]: 3.875e-05
[rewriter_before_opt_a]: 6.997e-05
[expand_dump_flag]: 3.53e-06
[jit_opt_a]: 0.00552788, [2]
    [Cycle 1]: 0.0018617, [27]
        [switch_simplify]: 5.393e-05
        [loop_unroll]: 3.06e-05
        [a_1]: 0.00057268
        [with_stream_mark]: 1.749e-05
        [recompute_prepare]: 8.05999e-06
        [updatestate_depend_eliminate]: 3.65e-06
        [updatestate_assign_eliminate]: 2.89001e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [parameter_eliminate]: 1.52001e-06
        [specialize_transform]: 6.54999e-06
        [updatestate_useless_node_eliminater]: 5.42001e-06
        [accelerated_algorithm]: 5.91e-06
        [meta_shard_fg_expand]: 2.44001e-06
        [get_grad_eliminate_]: 5.42001e-06
        [merge_forward]: 3.59002e-06
        [cell_reuse_recompute_pass]: 1.34998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.455e-05
        [j_node_and_user_rematch]: 9.25001e-06
        [meta_fg_expand]: 2.49001e-06
        [replace_old_param]: 9.92999e-06
        [inline_without_move]: 5.71998e-06
        [renormalize]: 0.00085229
        [add_forward_monad_depend]: 6.13002e-06
        [auto_monad_grad]: 2.30002e-06
        [auto_monad_eliminator]: 1.499e-05
        [cse]: 2.392e-05
        [replace_applicator]: 1.507e-05
    [Cycle 2]: 0.00035697, [27]
        [switch_simplify]: 6.32001e-06
        [loop_unroll]: 5.24e-06
        [a_1]: 9.93e-05
        [with_stream_mark]: 1.248e-05
        [recompute_prepare]: 5.40001e-06
        [updatestate_depend_eliminate]: 2.98e-06
        [updatestate_assign_eliminate]: 2.33998e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 5.45001e-06
        [updatestate_useless_node_eliminater]: 5.10001e-06
        [accelerated_algorithm]: 5.35001e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.46002e-06
        [merge_forward]: 3.2e-06
        [cell_reuse_recompute_pass]: 2.01998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.489e-05
        [j_node_and_user_rematch]: 8.52e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 7.56001e-06
        [inline_without_move]: 5.07e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.60999e-06
        [auto_monad_grad]: 1.03001e-06
        [auto_monad_eliminator]: 5.92001e-06
        [cse]: 1.21e-05
        [replace_applicator]: 5.32001e-06
[py_interpret_to_execute_after_opt_a]: 1.457e-05
[rewriter_after_opt_a]: 3.881e-05
[convert_after_rewriter]: 7.6e-06
[order_py_execute_after_rewriter]: 6.61999e-06
[mutable_eliminate]: 0.00063822
[jit_opt_b]: 5.833e-05, [1]
    [Cycle 1]: 4.987e-05, [2]
        [frontend_op_eliminate]: 2e-05
        [inline_after_opt_a]: 1.688e-05
[cconv]: 2.498e-05
[loop_unroll]: 0.0005388
[jit_opt_after_cconv]: 0.00015528, [1]
    [Cycle 1]: 0.00014819, [11]
        [c_1]: 2.259e-05
        [parameter_eliminate]: 3.8e-06
        [updatestate_depend_eliminate]: 6.59999e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [cse]: 2.315e-05
        [call_graph_tuple_transform]: 2.087e-05
        [tuple_list_get_item_eliminator]: 6.04999e-06
        [none_parameter_eliminate]: 1.58002e-06
        [renormalize]: 6.69999e-07
        [switch_simplify]: 5.46998e-06
[remove_dup_value]: 1.349e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 5.35999e-06
[add_recomputation]: 4.587e-05
[cse_after_recomputation]: 2.127e-05, [1]
    [Cycle 1]: 1.602e-05, [1]
        [cse]: 9.62001e-06
[auto_monad_reorder]: 1.388e-05
[get_jit_bprop_graph]: 2.03997e-06
[rewriter_after_jit_bprop_graph]: 6.12001e-06
[opt_after_jit_grad]: 0.0004768
[symbol_engine_optimizer]: 7.892e-05, [1]
    [Cycle 1]: 7.257e-05, [6]
        [build]: 4.25e-06
        [elim_shapecalc]: 8.13999e-06
        [elim_not_effective]: 1.302e-05
        [opt_reshape]: 6.73e-06
        [fold_const_symbol]: 9.47001e-06
        [renormalize]: 5.19998e-07
[validate]: 3.744e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.00606746
[execute]: 7.21999e-06
Sums
    bootstrap                            :     0.000307s :  2.02%
    type_inference                       :     0.004646s : 30.57%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000065s :  0.43%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.25%
    rewriter_before_opt_a                :     0.000070s :  0.46%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000060s :  0.40%
    jit_opt_a.loop_unroll                :     0.000036s :  0.24%
    jit_opt_a.a_1                        :     0.000672s :  4.42%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.07%
    jit_opt_a.renormalize                :     0.000852s :  5.61%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.14%
    jit_opt_a.cse                        :     0.000036s :  0.24%
    jit_opt_a.replace_applicator         :     0.000020s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.10%
    rewriter_after_opt_a                 :     0.000039s :  0.26%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000007s :  0.04%
    mutable_eliminate                    :     0.000638s :  4.20%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.11%
    cconv                                :     0.000025s :  0.16%
    loop_unroll                          :     0.000539s :  3.55%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000046s :  0.30%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000477s :  3.14%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006067s : 39.92%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000216    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000002s :      2: substitution.fold_const_symbol
 2.63% :     0.000006s :      3: substitution.graph_param_transform
83.90% :     0.000181s :      4: substitution.inline
 1.55% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.30% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.91% :     0.000004s :      2: substitution.replace_old_param
 6.00% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004583     2
80.58% :     0.003693s :      1: type_inference.infer
19.42% :     0.000890s :      1: type_inference.specialize
------[replace.]   0.000078     6
64.66% :     0.000051s :      4: replace.inline
35.34% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000190     6
93.87% :     0.000179s :      4: match.inline
 6.13% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.29% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.26% :     0.000002s :     10: predicate.addn_check_dump
 1.17% :     0.000001s :     10: predicate.addn_zero_filter
 2.11% :     0.000003s :     10: predicate.arithmetic_simplify
 1.83% :     0.000002s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.62% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.13% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000002s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 5.21% :     0.000007s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.60% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.68% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.41% :     0.000002s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.38% :     0.000003s :     15: predicate.partial_eliminate
 1.27% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 2.17% :     0.000003s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.67% :     0.000001s :      3: predicate.reset_defer_inline
 1.23% :     0.000002s :     10: predicate.reshape_eliminate
 1.32% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.49% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.58% :     0.000003s :     16: predicate.switch_defer_inline
 1.93% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.89% :     0.000010s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.27% :     0.000002s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.44% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.84% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000624    12
45.68% :     0.000285s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.21% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
49.11% :     0.000307s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021001    76
 0.23% :     0.000049s :      1: add_recomputation
 0.32% :     0.000068s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.53% :     0.000322s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.34% :     0.005532s :      1: jit_opt_a
 0.75% :     0.000158s :      1: jit_opt_after_cconv
 0.29% :     0.000061s :      1: jit_opt_b
 2.61% :     0.000549s :      1: loop_unroll
 3.09% :     0.000649s :      1: mutable_eliminate
 4.25% :     0.000892s :     26: opt.transform.jit_opt_a
 0.25% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000030s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000034s :      4: opt.transform.symbol_engine_opt
 2.31% :     0.000485s :      1: opt_after_jit_grad
 0.04% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000042s :      1: py_interpret_to_execute
 0.08% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 2.10% :     0.000442s :      1: renormalize.infer
 1.91% :     0.000402s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000042s :      1: rewriter_after_opt_a
 0.35% :     0.000073s :      1: rewriter_before_opt_a
 0.39% :     0.000081s :      1: symbol_engine_optimizer
28.95% :     0.006079s :      1: task_emit
22.17% :     0.004657s :      1: type_inference
 0.26% :     0.000055s :      1: validate


TotalTime = 0.017297, [33]
[bootstrap]: 0.00027674
[type_inference]: 0.0041943
[event_method]: 1.599e-05
[auto_monad]: 5.981e-05
[graph_reusing]: 5.24998e-06
[pre_auto_parallel]: 2.43e-06
[py_interpret_to_execute]: 2.722e-05
[rewriter_before_opt_a]: 6.75e-05
[expand_dump_flag]: 3.25e-06
[jit_opt_a]: 0.00474094, [2]
    [Cycle 1]: 0.00136387, [27]
        [switch_simplify]: 4.737e-05
        [loop_unroll]: 2.959e-05
        [a_1]: 0.00048469
        [with_stream_mark]: 1.516e-05
        [recompute_prepare]: 6.84001e-06
        [updatestate_depend_eliminate]: 3.24001e-06
        [updatestate_assign_eliminate]: 2.54001e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 5.52001e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 5.09e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 5.14e-06
        [merge_forward]: 3.73001e-06
        [cell_reuse_recompute_pass]: 1.06002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.312e-05
        [j_node_and_user_rematch]: 8.89e-06
        [meta_fg_expand]: 2.15002e-06
        [replace_old_param]: 8.06001e-06
        [inline_without_move]: 5.24998e-06
        [renormalize]: 0.00049561
        [add_forward_monad_depend]: 4.99e-06
        [auto_monad_grad]: 1.99999e-06
        [auto_monad_eliminator]: 1.347e-05
        [cse]: 2.05e-05
        [replace_applicator]: 1.121e-05
    [Cycle 2]: 0.00031175, [27]
        [switch_simplify]: 5.82001e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 8.666e-05
        [with_stream_mark]: 8.95999e-06
        [recompute_prepare]: 5.37001e-06
        [updatestate_depend_eliminate]: 2.37999e-06
        [updatestate_assign_eliminate]: 1.91998e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 9.49978e-07
        [specialize_transform]: 4.72e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 4.79002e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.42998e-06
        [merge_forward]: 2.28002e-06
        [cell_reuse_recompute_pass]: 1.62001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.247e-05
        [j_node_and_user_rematch]: 7.69002e-06
        [meta_fg_expand]: 1.56998e-06
        [replace_old_param]: 7.2e-06
        [inline_without_move]: 4.57e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 1.10001e-06
        [auto_monad_eliminator]: 4.67e-06
        [cse]: 9.82999e-06
        [replace_applicator]: 6.84001e-06
[py_interpret_to_execute_after_opt_a]: 8.78001e-06
[rewriter_after_opt_a]: 3.197e-05
[convert_after_rewriter]: 6.34999e-06
[order_py_execute_after_rewriter]: 4.80999e-06
[mutable_eliminate]: 0.00052681
[jit_opt_b]: 5.1e-05, [1]
    [Cycle 1]: 4.416e-05, [2]
        [frontend_op_eliminate]: 1.83e-05
        [inline_after_opt_a]: 1.455e-05
[cconv]: 2.326e-05
[loop_unroll]: 0.00040434
[jit_opt_after_cconv]: 0.00013209, [1]
    [Cycle 1]: 0.00012597, [11]
        [c_1]: 1.943e-05
        [parameter_eliminate]: 2.44999e-06
        [updatestate_depend_eliminate]: 5.11002e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.658e-05
        [call_graph_tuple_transform]: 1.737e-05
        [tuple_list_get_item_eliminator]: 5.25999e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 7.39994e-07
        [switch_simplify]: 5.18002e-06
[remove_dup_value]: 1.192e-05
[partial_unused_args_eliminate]: 1.95001e-06
[environ_conv]: 5.05001e-06
[add_recomputation]: 4.38e-05
[cse_after_recomputation]: 1.952e-05, [1]
    [Cycle 1]: 1.444e-05, [1]
        [cse]: 8.77e-06
[auto_monad_reorder]: 1.459e-05
[get_jit_bprop_graph]: 2.11e-06
[rewriter_after_jit_bprop_graph]: 4.68001e-06
[opt_after_jit_grad]: 0.00044377
[symbol_engine_optimizer]: 7.005e-05, [1]
    [Cycle 1]: 6.351e-05, [6]
        [build]: 3.15002e-06
        [elim_shapecalc]: 7.11001e-06
        [elim_not_effective]: 1.206e-05
        [opt_reshape]: 5.26002e-06
        [fold_const_symbol]: 8.2e-06
        [renormalize]: 5.00004e-07
[validate]: 3.21e-05
[backend_pass]: 8.79983e-07
[task_emit]: 0.00588134
[execute]: 7.96001e-06
Sums
    bootstrap                            :     0.000277s :  2.02%
    type_inference                       :     0.004194s : 30.68%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000060s :  0.44%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000027s :  0.20%
    rewriter_before_opt_a                :     0.000067s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.25%
    jit_opt_a.a_1                        :     0.000571s :  4.18%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000496s :  3.63%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000030s :  0.22%
    jit_opt_a.replace_applicator         :     0.000018s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000032s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000527s :  3.85%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000404s :  2.96%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000044s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000444s :  3.25%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005881s : 43.02%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000174    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.67% :     0.000005s :      3: substitution.graph_param_transform
82.35% :     0.000143s :      4: substitution.inline
 1.76% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.80% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.05% :     0.000004s :      2: substitution.replace_old_param
 6.63% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004137     2
82.08% :     0.003396s :      1: type_inference.infer
17.92% :     0.000741s :      1: type_inference.specialize
------[replace.]   0.000060     6
66.58% :     0.000040s :      4: replace.inline
33.42% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000151     6
93.08% :     0.000141s :      4: match.inline
 6.92% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.29% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.49% :     0.000002s :     10: predicate.addn_zero_filter
 1.97% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.51% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.46% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.79% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.92% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.27% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.57% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000002s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.36% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.19% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.95% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.60% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.42% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.90% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000492    12
49.34% :     0.000243s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.21% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.45% :     0.000224s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018665    76
 0.25% :     0.000047s :      1: add_recomputation
 0.34% :     0.000063s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.55% :     0.000289s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
25.41% :     0.004743s :      1: jit_opt_a
 0.72% :     0.000135s :      1: jit_opt_after_cconv
 0.29% :     0.000054s :      1: jit_opt_b
 2.21% :     0.000412s :      1: loop_unroll
 2.87% :     0.000535s :      1: mutable_eliminate
 4.12% :     0.000769s :     26: opt.transform.jit_opt_a
 0.24% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.42% :     0.000451s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000014s :      1: remove_dup_value
 1.30% :     0.000243s :      1: renormalize.infer
 1.31% :     0.000245s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000035s :      1: rewriter_after_opt_a
 0.38% :     0.000070s :      1: rewriter_before_opt_a
 0.39% :     0.000072s :      1: symbol_engine_optimizer
31.57% :     0.005893s :      1: task_emit
22.53% :     0.004205s :      1: type_inference
 0.26% :     0.000049s :      1: validate


TotalTime = 0.0104756, [30]
[bootstrap]: 0.00023137
[type_inference]: 0.00361747
[event_method]: 1.529e-05
[auto_monad]: 5.047e-05
[graph_reusing]: 4.72998e-06
[pre_auto_parallel]: 1.76e-06
[py_interpret_to_execute]: 2.604e-05
[rewriter_before_opt_a]: 6.511e-05
[expand_dump_flag]: 2.21e-06
[jit_opt_a]: 0.00451803, [2]
    [Cycle 1]: 0.00128422, [27]
        [switch_simplify]: 4.423e-05
        [loop_unroll]: 3.091e-05
        [a_1]: 0.00047845
        [with_stream_mark]: 1.193e-05
        [recompute_prepare]: 7.19001e-06
        [updatestate_depend_eliminate]: 2.84001e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 6.43998e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 1.47999e-06
        [get_grad_eliminate_]: 5.51e-06
        [merge_forward]: 2.53998e-06
        [cell_reuse_recompute_pass]: 8.79983e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.24e-05
        [j_node_and_user_rematch]: 9.59e-06
        [meta_fg_expand]: 2.07999e-06
        [replace_old_param]: 8.04002e-06
        [inline_without_move]: 5.55001e-06
        [renormalize]: 0.00043445
        [add_forward_monad_depend]: 4.28001e-06
        [auto_monad_grad]: 1.46998e-06
        [auto_monad_eliminator]: 1.059e-05
        [cse]: 1.769e-05
        [replace_applicator]: 1.059e-05
    [Cycle 2]: 0.00031971, [27]
        [switch_simplify]: 5.46e-06
        [loop_unroll]: 5.25999e-06
        [a_1]: 9.013e-05
        [with_stream_mark]: 8.42e-06
        [recompute_prepare]: 5.08002e-06
        [updatestate_depend_eliminate]: 2.55997e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 4.99998e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 5.29e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.93001e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.235e-05
        [j_node_and_user_rematch]: 8.05999e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 1.261e-05
        [inline_without_move]: 4.67998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 4.80999e-06
        [cse]: 9.71998e-06
        [replace_applicator]: 5.04e-06
[py_interpret_to_execute_after_opt_a]: 7.57998e-06
[rewriter_after_opt_a]: 2.597e-05
[convert_after_rewriter]: 6.25002e-06
[order_py_execute_after_rewriter]: 4.35999e-06
[mutable_eliminate]: 0.00045504
[jit_opt_b]: 4.884e-05, [1]
    [Cycle 1]: 4.248e-05, [2]
        [frontend_op_eliminate]: 1.608e-05
        [inline_after_opt_a]: 1.512e-05
[cconv]: 1.917e-05
[loop_unroll]: 0.00041323
[jit_opt_after_cconv]: 0.00013658, [1]
    [Cycle 1]: 0.00012976, [11]
        [c_1]: 2.051e-05
        [parameter_eliminate]: 2.53e-06
        [updatestate_depend_eliminate]: 5.30999e-06
        [updatestate_assign_eliminate]: 2.36998e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.589e-05
        [call_graph_tuple_transform]: 1.817e-05
        [tuple_list_get_item_eliminator]: 5.19e-06
        [none_parameter_eliminate]: 1.15001e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 5.32001e-06
[remove_dup_value]: 1.075e-05
[partial_unused_args_eliminate]: 1.81998e-06
[environ_conv]: 3.91001e-06
[add_recomputation]: 3.62e-05
[cse_after_recomputation]: 1.928e-05, [1]
    [Cycle 1]: 1.433e-05, [1]
        [cse]: 8.84e-06
[auto_monad_reorder]: 1.158e-05
[get_jit_bprop_graph]: 1.35001e-06
[rewriter_after_jit_bprop_graph]: 4.3e-06
[opt_after_jit_grad]: 0.00049293
[symbol_engine_optimizer]: 7.228e-05, [1]
    [Cycle 1]: 6.59e-05, [6]
        [build]: 3.35e-06
        [elim_shapecalc]: 7.2e-06
        [elim_not_effective]: 1.234e-05
        [opt_reshape]: 5.86e-06
        [fold_const_symbol]: 8.88002e-06
        [renormalize]: 3.99974e-07
[validate]: 2.696e-05
Sums
    bootstrap                            :     0.000231s :  3.29%
    type_inference                       :     0.003617s : 51.48%
    event_method                         :     0.000015s :  0.22%
    auto_monad                           :     0.000050s :  0.72%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000026s :  0.37%
    rewriter_before_opt_a                :     0.000065s :  0.93%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000050s :  0.71%
    jit_opt_a.loop_unroll                :     0.000036s :  0.51%
    jit_opt_a.a_1                        :     0.000569s :  8.09%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.16%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.15%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.25%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000021s :  0.29%
    jit_opt_a.inline_without_move        :     0.000010s :  0.15%
    jit_opt_a.renormalize                :     0.000435s :  6.18%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.22%
    jit_opt_a.cse                        :     0.000027s :  0.39%
    jit_opt_a.replace_applicator         :     0.000016s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000026s :  0.37%
    convert_after_rewriter               :     0.000006s :  0.09%
    order_py_execute_after_rewriter      :     0.000004s :  0.06%
    mutable_eliminate                    :     0.000455s :  6.48%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.22%
    cconv                                :     0.000019s :  0.27%
    loop_unroll                          :     0.000413s :  5.88%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.29%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.26%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.08%
    remove_dup_value                     :     0.000011s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000036s :  0.52%
    cse_after_recomputation.cse          :     0.000009s :  0.13%
    auto_monad_reorder                   :     0.000012s :  0.16%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.06%
    opt_after_jit_grad                   :     0.000493s :  7.02%
    symbol_engine_optimizer.build        :     0.000003s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.18%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.13%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000027s :  0.38%


Time group info:
------[substitution.]   0.000160    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.85% :     0.000001s :      2: substitution.fold_const_symbol
 3.42% :     0.000005s :      3: substitution.graph_param_transform
79.13% :     0.000127s :      4: substitution.inline
 2.16% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.50% :     0.000004s :      4: substitution.remove_not_recompute_node
 5.18% :     0.000008s :      2: substitution.replace_old_param
 5.70% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003571     2
80.16% :     0.002863s :      1: type_inference.infer
19.84% :     0.000709s :      1: type_inference.specialize
------[replace.]   0.000060     6
64.43% :     0.000039s :      4: replace.inline
35.57% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000132     6
93.98% :     0.000124s :      4: match.inline
 6.02% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 2.29% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000001s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.23% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.19% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.46% :     0.000002s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.52% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.78% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.90% :     0.000002s :     10: predicate.reduce_eliminate
 1.75% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.57% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.14% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 2.40% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.43% :     0.000009s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.48% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.70% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.83% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.15% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000470    12
48.12% :     0.000226s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.21% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.67% :     0.000219s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011753    72
 0.33% :     0.000039s :      1: add_recomputation
 0.46% :     0.000054s :      1: auto_monad
 0.12% :     0.000014s :      1: auto_monad_reorder
 2.06% :     0.000242s :      1: bootstrap
 0.18% :     0.000022s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000021s :      1: cse_after_recomputation
 0.05% :     0.000006s :      1: environ_conv
 0.16% :     0.000019s :      1: event_method
 0.04% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
38.46% :     0.004520s :      1: jit_opt_a
 1.19% :     0.000139s :      1: jit_opt_after_cconv
 0.44% :     0.000051s :      1: jit_opt_b
 3.58% :     0.000420s :      1: loop_unroll
 3.94% :     0.000463s :      1: mutable_eliminate
 6.58% :     0.000774s :     26: opt.transform.jit_opt_a
 0.39% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000025s :      4: opt.transform.jit_opt_b
 0.10% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000031s :      4: opt.transform.symbol_engine_opt
 4.26% :     0.000500s :      1: opt_after_jit_grad
 0.05% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000029s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000013s :      1: remove_dup_value
 1.74% :     0.000204s :      1: renormalize.infer
 1.91% :     0.000224s :      1: renormalize.specialize
 0.05% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000029s :      1: rewriter_after_opt_a
 0.58% :     0.000068s :      1: rewriter_before_opt_a
 0.64% :     0.000075s :      1: symbol_engine_optimizer
30.87% :     0.003628s :      1: type_inference


TotalTime = 0.0201843, [33]
[bootstrap]: 0.00029801
[type_inference]: 0.00468405
[event_method]: 1.772e-05
[auto_monad]: 6.69e-05
[graph_reusing]: 5.84999e-06
[pre_auto_parallel]: 2.14999e-06
[py_interpret_to_execute]: 4.406e-05
[rewriter_before_opt_a]: 7.627e-05
[expand_dump_flag]: 3.23e-06
[jit_opt_a]: 0.00613971, [2]
    [Cycle 1]: 0.00198898, [27]
        [switch_simplify]: 6.29e-05
        [loop_unroll]: 3.108e-05
        [a_1]: 0.00061062
        [with_stream_mark]: 2.123e-05
        [recompute_prepare]: 9.87001e-06
        [updatestate_depend_eliminate]: 3.82002e-06
        [updatestate_assign_eliminate]: 3.18e-06
        [updatestate_loads_eliminate]: 2.78e-06
        [parameter_eliminate]: 2.06e-06
        [specialize_transform]: 7.41999e-06
        [updatestate_useless_node_eliminater]: 5.85002e-06
        [accelerated_algorithm]: 6.36e-06
        [meta_shard_fg_expand]: 3.36001e-06
        [get_grad_eliminate_]: 5.84e-06
        [merge_forward]: 4.23999e-06
        [cell_reuse_recompute_pass]: 2.99999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.903e-05
        [j_node_and_user_rematch]: 1.092e-05
        [meta_fg_expand]: 2.49001e-06
        [replace_old_param]: 1.075e-05
        [inline_without_move]: 6.34001e-06
        [renormalize]: 0.00088986
        [add_forward_monad_depend]: 6.92002e-06
        [auto_monad_grad]: 3.07002e-06
        [auto_monad_eliminator]: 1.429e-05
        [cse]: 2.887e-05
        [replace_applicator]: 1.843e-05
    [Cycle 2]: 0.0003921, [27]
        [switch_simplify]: 5.92999e-06
        [loop_unroll]: 5.32999e-06
        [a_1]: 0.00010755
        [with_stream_mark]: 1.333e-05
        [recompute_prepare]: 5.12e-06
        [updatestate_depend_eliminate]: 2.59999e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 1.32e-06
        [specialize_transform]: 5.12e-06
        [updatestate_useless_node_eliminater]: 5.86e-06
        [accelerated_algorithm]: 6.71999e-06
        [meta_shard_fg_expand]: 2.78998e-06
        [get_grad_eliminate_]: 5.11002e-06
        [merge_forward]: 3.46999e-06
        [cell_reuse_recompute_pass]: 3.04001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.75e-05
        [j_node_and_user_rematch]: 9.44998e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 8.62e-06
        [inline_without_move]: 5.63997e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.02999e-06
        [auto_monad_grad]: 1.35999e-06
        [auto_monad_eliminator]: 7.48e-06
        [cse]: 1.23e-05
        [replace_applicator]: 6.76e-06
[py_interpret_to_execute_after_opt_a]: 1.275e-05
[rewriter_after_opt_a]: 3.971e-05
[convert_after_rewriter]: 8.72998e-06
[order_py_execute_after_rewriter]: 6.07001e-06
[mutable_eliminate]: 0.00074669
[jit_opt_b]: 5.739e-05, [1]
    [Cycle 1]: 4.918e-05, [2]
        [frontend_op_eliminate]: 1.93e-05
        [inline_after_opt_a]: 1.684e-05
[cconv]: 2.544e-05
[loop_unroll]: 0.0004583
[jit_opt_after_cconv]: 0.00015439, [1]
    [Cycle 1]: 0.00014737, [11]
        [c_1]: 2.347e-05
        [parameter_eliminate]: 3.37002e-06
        [updatestate_depend_eliminate]: 5.64998e-06
        [updatestate_assign_eliminate]: 2.94999e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 2.248e-05
        [call_graph_tuple_transform]: 1.87e-05
        [tuple_list_get_item_eliminator]: 5.75001e-06
        [none_parameter_eliminate]: 1.46998e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 5.88002e-06
[remove_dup_value]: 1.338e-05
[partial_unused_args_eliminate]: 2.09999e-06
[environ_conv]: 6.40002e-06
[add_recomputation]: 4.566e-05
[cse_after_recomputation]: 2.201e-05, [1]
    [Cycle 1]: 1.625e-05, [1]
        [cse]: 9.37999e-06
[auto_monad_reorder]: 1.531e-05
[get_jit_bprop_graph]: 1.54998e-06
[rewriter_after_jit_bprop_graph]: 6.12001e-06
[opt_after_jit_grad]: 0.00048195
[symbol_engine_optimizer]: 7.586e-05, [1]
    [Cycle 1]: 6.893e-05, [6]
        [build]: 3.74002e-06
        [elim_shapecalc]: 8.07e-06
        [elim_not_effective]: 1.358e-05
        [opt_reshape]: 6.46e-06
        [fold_const_symbol]: 8.94e-06
        [renormalize]: 9.09989e-07
[validate]: 3.792e-05
[backend_pass]: 1.36002e-06
[task_emit]: 0.00641487
[execute]: 7.18998e-06
Sums
    bootstrap                            :     0.000298s :  1.89%
    type_inference                       :     0.004684s : 29.71%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000067s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.28%
    rewriter_before_opt_a                :     0.000076s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000069s :  0.44%
    jit_opt_a.loop_unroll                :     0.000036s :  0.23%
    jit_opt_a.a_1                        :     0.000718s :  4.56%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000013s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.07%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000047s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.12%
    jit_opt_a.inline_without_move        :     0.000012s :  0.08%
    jit_opt_a.renormalize                :     0.000890s :  5.65%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.14%
    jit_opt_a.cse                        :     0.000041s :  0.26%
    jit_opt_a.replace_applicator         :     0.000025s :  0.16%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.08%
    rewriter_after_opt_a                 :     0.000040s :  0.25%
    convert_after_rewriter               :     0.000009s :  0.06%
    order_py_execute_after_rewriter      :     0.000006s :  0.04%
    mutable_eliminate                    :     0.000747s :  4.74%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.11%
    cconv                                :     0.000025s :  0.16%
    loop_unroll                          :     0.000458s :  2.91%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000046s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000482s :  3.06%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000038s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006415s : 40.69%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000255    23
 0.78% :     0.000002s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 2.25% :     0.000006s :      3: substitution.graph_param_transform
79.79% :     0.000203s :      4: substitution.inline
 1.61% :     0.000004s :      4: substitution.j_node_and_user_rematch
 7.36% :     0.000019s :      4: substitution.remove_not_recompute_node
 2.33% :     0.000006s :      2: substitution.replace_old_param
 5.39% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004615     2
80.75% :     0.003727s :      1: type_inference.infer
19.25% :     0.000888s :      1: type_inference.specialize
------[replace.]   0.000085     6
65.46% :     0.000056s :      4: replace.inline
34.54% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000212     6
94.09% :     0.000199s :      4: match.inline
 5.91% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.51% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.12% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.71% :     0.000002s :     10: predicate.addn_zero_filter
 1.81% :     0.000002s :     10: predicate.arithmetic_simplify
 2.00% :     0.000003s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.12% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.06% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.30% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.27% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.30% :     0.000007s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.56% :     0.000002s :      6: predicate.less_batch_normalization
 1.91% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.52% :     0.000002s :     15: predicate.load_eliminater
 1.25% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.68% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.23% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 1.89% :     0.000003s :      3: predicate.mutable_eliminate
 0.57% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.06% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.90% :     0.000003s :     10: predicate.reduce_eliminate
 1.68% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.72% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.85% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.22% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.08% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.22% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.08% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.74% :     0.000004s :     16: predicate.switch_defer_inline
 1.96% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.54% :     0.000010s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.85% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.86% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.23% :     0.000003s :     10: predicate.value_based_eliminate
 0.26% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.91% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000636    12
46.69% :     0.000297s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.40% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.91% :     0.000298s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022161    76
 0.22% :     0.000048s :      1: add_recomputation
 0.32% :     0.000072s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.40% :     0.000309s :      1: bootstrap
 0.13% :     0.000028s :      1: cconv
 0.05% :     0.000012s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.10% :     0.000022s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.73% :     0.006144s :      1: jit_opt_a
 0.71% :     0.000158s :      1: jit_opt_after_cconv
 0.27% :     0.000061s :      1: jit_opt_b
 2.11% :     0.000467s :      1: loop_unroll
 3.42% :     0.000758s :      1: mutable_eliminate
 4.42% :     0.000979s :     26: opt.transform.jit_opt_a
 0.23% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000029s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000490s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.21% :     0.000047s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 2.07% :     0.000460s :      1: renormalize.infer
 1.90% :     0.000420s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000043s :      1: rewriter_after_opt_a
 0.36% :     0.000079s :      1: rewriter_before_opt_a
 0.35% :     0.000078s :      1: symbol_engine_optimizer
29.01% :     0.006428s :      1: task_emit
21.20% :     0.004697s :      1: type_inference
 0.26% :     0.000057s :      1: validate


TotalTime = 0.0186164, [33]
[bootstrap]: 0.00031771
[type_inference]: 0.00472115
[event_method]: 1.576e-05
[auto_monad]: 6.079e-05
[graph_reusing]: 6.12999e-06
[pre_auto_parallel]: 2.50002e-06
[py_interpret_to_execute]: 2.929e-05
[rewriter_before_opt_a]: 6.878e-05
[expand_dump_flag]: 3.36001e-06
[jit_opt_a]: 0.00510796, [2]
    [Cycle 1]: 0.00155644, [27]
        [switch_simplify]: 5.015e-05
        [loop_unroll]: 2.865e-05
        [a_1]: 0.00051967
        [with_stream_mark]: 1.836e-05
        [recompute_prepare]: 7.05e-06
        [updatestate_depend_eliminate]: 2.97002e-06
        [updatestate_assign_eliminate]: 2.67001e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 2.24999e-06
        [specialize_transform]: 5.51e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.34e-06
        [meta_shard_fg_expand]: 1.84e-06
        [get_grad_eliminate_]: 4.94e-06
        [merge_forward]: 3.37002e-06
        [cell_reuse_recompute_pass]: 1.20999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.384e-05
        [j_node_and_user_rematch]: 8.80999e-06
        [meta_fg_expand]: 2.31998e-06
        [replace_old_param]: 8.46002e-06
        [inline_without_move]: 5.09e-06
        [renormalize]: 0.00064046
        [add_forward_monad_depend]: 5.21002e-06
        [auto_monad_grad]: 2.24001e-06
        [auto_monad_eliminator]: 1.189e-05
        [cse]: 2.432e-05
        [replace_applicator]: 1.185e-05
    [Cycle 2]: 0.00031272, [27]
        [switch_simplify]: 4.99003e-06
        [loop_unroll]: 5.02e-06
        [a_1]: 8.826e-05
        [with_stream_mark]: 9.19e-06
        [recompute_prepare]: 4.87e-06
        [updatestate_depend_eliminate]: 2.49999e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 9.20001e-07
        [specialize_transform]: 4.77998e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 4.98001e-06
        [meta_shard_fg_expand]: 1.47999e-06
        [get_grad_eliminate_]: 4.36002e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 1.61998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.243e-05
        [j_node_and_user_rematch]: 7.65998e-06
        [meta_fg_expand]: 1.95001e-06
        [replace_old_param]: 6.95002e-06
        [inline_without_move]: 4.77998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 4.77998e-06
        [cse]: 8.57e-06
        [replace_applicator]: 4.96002e-06
[py_interpret_to_execute_after_opt_a]: 1.013e-05
[rewriter_after_opt_a]: 3.376e-05
[convert_after_rewriter]: 6.42001e-06
[order_py_execute_after_rewriter]: 4.75999e-06
[mutable_eliminate]: 0.00055649
[jit_opt_b]: 5.122e-05, [1]
    [Cycle 1]: 4.362e-05, [2]
        [frontend_op_eliminate]: 1.647e-05
        [inline_after_opt_a]: 1.565e-05
[cconv]: 2.756e-05
[loop_unroll]: 0.00047988
[jit_opt_after_cconv]: 0.00016945, [1]
    [Cycle 1]: 0.00016224, [11]
        [c_1]: 3.7e-05
        [parameter_eliminate]: 2.74001e-06
        [updatestate_depend_eliminate]: 5.61e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 2.148e-05
        [call_graph_tuple_transform]: 2.176e-05
        [tuple_list_get_item_eliminator]: 5.61e-06
        [none_parameter_eliminate]: 1.69e-06
        [renormalize]: 9.70002e-07
        [switch_simplify]: 5.67999e-06
[remove_dup_value]: 1.443e-05
[partial_unused_args_eliminate]: 2.16998e-06
[environ_conv]: 5.72001e-06
[add_recomputation]: 4.816e-05
[cse_after_recomputation]: 2.258e-05, [1]
    [Cycle 1]: 1.616e-05, [1]
        [cse]: 9.36e-06
[auto_monad_reorder]: 1.616e-05
[get_jit_bprop_graph]: 1.82001e-06
[rewriter_after_jit_bprop_graph]: 6.33e-06
[opt_after_jit_grad]: 0.00051227
[symbol_engine_optimizer]: 7.185e-05, [1]
    [Cycle 1]: 6.553e-05, [6]
        [build]: 3.04999e-06
        [elim_shapecalc]: 7.23999e-06
        [elim_not_effective]: 1.333e-05
        [opt_reshape]: 5.37001e-06
        [fold_const_symbol]: 8.47998e-06
        [renormalize]: 1.24e-06
[validate]: 3.573e-05
[backend_pass]: 1.00999e-06
[task_emit]: 0.00601227
[execute]: 6.09001e-06
Sums
    bootstrap                            :     0.000318s :  2.15%
    type_inference                       :     0.004721s : 31.93%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000061s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.20%
    rewriter_before_opt_a                :     0.000069s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.37%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000608s :  4.11%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000641s :  4.33%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.11%
    jit_opt_a.cse                        :     0.000033s :  0.22%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000556s :  3.76%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000028s :  0.19%
    loop_unroll                          :     0.000480s :  3.25%
    jit_opt_after_cconv.c_1              :     0.000037s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.15%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000048s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000512s :  3.46%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000036s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006012s : 40.66%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000205    23
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.99% :     0.000006s :      3: substitution.graph_param_transform
84.34% :     0.000173s :      4: substitution.inline
 1.49% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.97% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.78% :     0.000004s :      2: substitution.replace_old_param
 5.89% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004663     2
83.12% :     0.003876s :      1: type_inference.infer
16.88% :     0.000787s :      1: type_inference.specialize
------[replace.]   0.000064     6
68.91% :     0.000044s :      4: replace.inline
31.09% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000181     6
93.94% :     0.000170s :      4: match.inline
 6.06% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.03% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000001s :     10: predicate.addn_zero_filter
 2.01% :     0.000002s :     10: predicate.arithmetic_simplify
 1.31% :     0.000001s :     10: predicate.cast_eliminate
 0.63% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.69% :     0.000001s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.35% :     0.000000s :      3: predicate.graph_param_transform
 5.29% :     0.000006s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.73% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.11% :     0.000002s :     15: predicate.load_eliminater
 2.54% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.57% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000002s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.49% :     0.000001s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.07% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.25% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.30% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.37% :     0.000008s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000575    12
50.11% :     0.000288s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.44% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.45% :     0.000244s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020192    76
 0.25% :     0.000051s :      1: add_recomputation
 0.32% :     0.000064s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.65% :     0.000333s :      1: bootstrap
 0.15% :     0.000031s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000019s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
25.31% :     0.005111s :      1: jit_opt_a
 0.85% :     0.000173s :      1: jit_opt_after_cconv
 0.27% :     0.000054s :      1: jit_opt_b
 2.43% :     0.000490s :      1: loop_unroll
 2.80% :     0.000566s :      1: mutable_eliminate
 3.99% :     0.000806s :     26: opt.transform.jit_opt_a
 0.33% :     0.000066s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.57% :     0.000519s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.16% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000017s :      1: remove_dup_value
 1.71% :     0.000345s :      1: renormalize.infer
 1.43% :     0.000289s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000037s :      1: rewriter_after_opt_a
 0.36% :     0.000072s :      1: rewriter_before_opt_a
 0.37% :     0.000074s :      1: symbol_engine_optimizer
29.83% :     0.006023s :      1: task_emit
23.42% :     0.004729s :      1: type_inference
 0.27% :     0.000055s :      1: validate


TotalTime = 0.0109392, [30]
[bootstrap]: 0.00025154
[type_inference]: 0.00386173
[event_method]: 1.613e-05
[auto_monad]: 5.821e-05
[graph_reusing]: 5.29998e-06
[pre_auto_parallel]: 2.28998e-06
[py_interpret_to_execute]: 2.913e-05
[rewriter_before_opt_a]: 6.888e-05
[expand_dump_flag]: 2.39001e-06
[jit_opt_a]: 0.00470708, [2]
    [Cycle 1]: 0.00134904, [27]
        [switch_simplify]: 4.792e-05
        [loop_unroll]: 3.019e-05
        [a_1]: 0.00049938
        [with_stream_mark]: 1.434e-05
        [recompute_prepare]: 7.26001e-06
        [updatestate_depend_eliminate]: 3.08e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 1.55999e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 6.02999e-06
        [meta_shard_fg_expand]: 1.86998e-06
        [get_grad_eliminate_]: 5.76e-06
        [merge_forward]: 3.34001e-06
        [cell_reuse_recompute_pass]: 1.09998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.328e-05
        [j_node_and_user_rematch]: 8.93002e-06
        [meta_fg_expand]: 2.12001e-06
        [replace_old_param]: 9.10999e-06
        [inline_without_move]: 5.79e-06
        [renormalize]: 0.00046148
        [add_forward_monad_depend]: 4.73001e-06
        [auto_monad_grad]: 2.31e-06
        [auto_monad_eliminator]: 1.245e-05
        [cse]: 2.028e-05
        [replace_applicator]: 1.096e-05
    [Cycle 2]: 0.00032591, [27]
        [switch_simplify]: 5.57999e-06
        [loop_unroll]: 5.05001e-06
        [a_1]: 9.231e-05
        [with_stream_mark]: 8.37e-06
        [recompute_prepare]: 5.37001e-06
        [updatestate_depend_eliminate]: 2.64999e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.96998e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 5.14e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.20001e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.71002e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 1.31002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.248e-05
        [j_node_and_user_rematch]: 8.22e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 7.53e-06
        [inline_without_move]: 4.70001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.24998e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 4.97e-06
        [cse]: 1.075e-05
        [replace_applicator]: 5.20999e-06
[py_interpret_to_execute_after_opt_a]: 7.93999e-06
[rewriter_after_opt_a]: 2.811e-05
[convert_after_rewriter]: 6.43998e-06
[order_py_execute_after_rewriter]: 5.29998e-06
[mutable_eliminate]: 0.0004585
[jit_opt_b]: 5.059e-05, [1]
    [Cycle 1]: 4.371e-05, [2]
        [frontend_op_eliminate]: 1.695e-05
        [inline_after_opt_a]: 1.49e-05
[cconv]: 2.361e-05
[loop_unroll]: 0.0004139
[jit_opt_after_cconv]: 0.00013558, [1]
    [Cycle 1]: 0.0001293, [11]
        [c_1]: 2.124e-05
        [parameter_eliminate]: 2.40997e-06
        [updatestate_depend_eliminate]: 5.05999e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [cse]: 1.556e-05
        [call_graph_tuple_transform]: 1.795e-05
        [tuple_list_get_item_eliminator]: 5.51998e-06
        [none_parameter_eliminate]: 1.42999e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 5.23002e-06
[remove_dup_value]: 1.21e-05
[partial_unused_args_eliminate]: 1.97001e-06
[environ_conv]: 4.60999e-06
[add_recomputation]: 4.124e-05
[cse_after_recomputation]: 2.026e-05, [1]
    [Cycle 1]: 1.512e-05, [1]
        [cse]: 8.93002e-06
[auto_monad_reorder]: 1.245e-05
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 4.36002e-06
[opt_after_jit_grad]: 0.00046041
[symbol_engine_optimizer]: 7.12e-05, [1]
    [Cycle 1]: 6.508e-05, [6]
        [build]: 3.71999e-06
        [elim_shapecalc]: 6.97002e-06
        [elim_not_effective]: 1.242e-05
        [opt_reshape]: 5.92001e-06
        [fold_const_symbol]: 8.49998e-06
        [renormalize]: 4.39992e-07
[validate]: 2.666e-05
Sums
    bootstrap                            :     0.000252s :  3.42%
    type_inference                       :     0.003862s : 52.48%
    event_method                         :     0.000016s :  0.22%
    auto_monad                           :     0.000058s :  0.79%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000029s :  0.40%
    rewriter_before_opt_a                :     0.000069s :  0.94%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000053s :  0.73%
    jit_opt_a.loop_unroll                :     0.000035s :  0.48%
    jit_opt_a.a_1                        :     0.000592s :  8.04%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.14%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.23%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000462s :  6.27%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.24%
    jit_opt_a.cse                        :     0.000031s :  0.42%
    jit_opt_a.replace_applicator         :     0.000016s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000028s :  0.38%
    convert_after_rewriter               :     0.000006s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000458s :  6.23%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.20%
    cconv                                :     0.000024s :  0.32%
    loop_unroll                          :     0.000414s :  5.62%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.29%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000012s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000041s :  0.56%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000012s :  0.17%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.06%
    opt_after_jit_grad                   :     0.000460s :  6.26%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000027s :  0.36%


Time group info:
------[substitution.]   0.000175    23
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 2.98% :     0.000005s :      3: substitution.graph_param_transform
81.78% :     0.000143s :      4: substitution.inline
 1.77% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.49% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.18% :     0.000004s :      2: substitution.replace_old_param
 7.08% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003807     2
80.18% :     0.003052s :      1: type_inference.infer
19.82% :     0.000755s :      1: type_inference.specialize
------[replace.]   0.000062     6
65.76% :     0.000041s :      4: replace.inline
34.24% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000151     6
92.61% :     0.000140s :      4: match.inline
 7.39% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.77% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.49% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000001s :     10: predicate.addn_zero_filter
 2.14% :     0.000003s :     10: predicate.arithmetic_simplify
 1.11% :     0.000001s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.53% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.68% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.16% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 4.85% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.41% :     0.000002s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 1.70% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000002s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.80% :     0.000002s :     10: predicate.reduce_eliminate
 1.77% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.84% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.20% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.27% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.28% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.93% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.07% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.97% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000509    12
49.25% :     0.000250s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.35% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.40% :     0.000231s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012274    72
 0.36% :     0.000044s :      1: add_recomputation
 0.50% :     0.000062s :      1: auto_monad
 0.12% :     0.000015s :      1: auto_monad_reorder
 2.15% :     0.000264s :      1: bootstrap
 0.21% :     0.000026s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.37% :     0.004710s :      1: jit_opt_a
 1.13% :     0.000138s :      1: jit_opt_after_cconv
 0.43% :     0.000053s :      1: jit_opt_b
 3.44% :     0.000422s :      1: loop_unroll
 3.80% :     0.000467s :      1: mutable_eliminate
 6.48% :     0.000795s :     26: opt.transform.jit_opt_a
 0.38% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000025s :      4: opt.transform.jit_opt_b
 0.10% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.13% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.82% :     0.000469s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000032s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000015s :      1: remove_dup_value
 1.73% :     0.000212s :      1: renormalize.infer
 1.98% :     0.000243s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000031s :      1: rewriter_after_opt_a
 0.58% :     0.000072s :      1: rewriter_before_opt_a
 0.60% :     0.000074s :      1: symbol_engine_optimizer
31.55% :     0.003873s :      1: type_inference


TotalTime = 0.0187794, [33]
[bootstrap]: 0.00030952
[type_inference]: 0.00448748
[event_method]: 1.64e-05
[auto_monad]: 5.683e-05
[graph_reusing]: 5.29998e-06
[pre_auto_parallel]: 2.02999e-06
[py_interpret_to_execute]: 3.76e-05
[rewriter_before_opt_a]: 6.933e-05
[expand_dump_flag]: 3.04001e-06
[jit_opt_a]: 0.00525177, [2]
    [Cycle 1]: 0.00165605, [27]
        [switch_simplify]: 5.021e-05
        [loop_unroll]: 2.924e-05
        [a_1]: 0.00051529
        [with_stream_mark]: 1.543e-05
        [recompute_prepare]: 6.74999e-06
        [updatestate_depend_eliminate]: 2.89001e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 1.40001e-06
        [specialize_transform]: 5.76998e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.64998e-06
        [meta_shard_fg_expand]: 2.26998e-06
        [get_grad_eliminate_]: 5.41002e-06
        [merge_forward]: 3.25e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.258e-05
        [j_node_and_user_rematch]: 9.76998e-06
        [meta_fg_expand]: 2.17001e-06
        [replace_old_param]: 8.03001e-06
        [inline_without_move]: 5.27999e-06
        [renormalize]: 0.00074537
        [add_forward_monad_depend]: 5.21002e-06
        [auto_monad_grad]: 2.02999e-06
        [auto_monad_eliminator]: 1.207e-05
        [cse]: 2.233e-05
        [replace_applicator]: 1.271e-05
    [Cycle 2]: 0.00036274, [27]
        [switch_simplify]: 5.32999e-06
        [loop_unroll]: 4.95001e-06
        [a_1]: 8.905e-05
        [with_stream_mark]: 9.24998e-06
        [recompute_prepare]: 4.74e-06
        [updatestate_depend_eliminate]: 2.49999e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 5.92999e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 5.49998e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 4.93001e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 1.98002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.28e-05
        [j_node_and_user_rematch]: 7.89002e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 7.47998e-06
        [inline_without_move]: 4.74998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.32999e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 5.29e-06
        [cse]: 1.208e-05
        [replace_applicator]: 4.85999e-06
[py_interpret_to_execute_after_opt_a]: 9.89001e-06
[rewriter_after_opt_a]: 4.191e-05
[convert_after_rewriter]: 6.28998e-06
[order_py_execute_after_rewriter]: 4.87e-06
[mutable_eliminate]: 0.00057965
[jit_opt_b]: 5.973e-05, [1]
    [Cycle 1]: 5.126e-05, [2]
        [frontend_op_eliminate]: 2.345e-05
        [inline_after_opt_a]: 1.519e-05
[cconv]: 2.568e-05
[loop_unroll]: 0.00044813
[jit_opt_after_cconv]: 0.00014907, [1]
    [Cycle 1]: 0.00014228, [11]
        [c_1]: 2.263e-05
        [parameter_eliminate]: 3.53e-06
        [updatestate_depend_eliminate]: 5.41998e-06
        [updatestate_assign_eliminate]: 2.48002e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [cse]: 2.276e-05
        [call_graph_tuple_transform]: 1.932e-05
        [tuple_list_get_item_eliminator]: 5.63002e-06
        [none_parameter_eliminate]: 1.74e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.32999e-06
[remove_dup_value]: 1.318e-05
[partial_unused_args_eliminate]: 2.07999e-06
[environ_conv]: 5.80002e-06
[add_recomputation]: 4.525e-05
[cse_after_recomputation]: 2.25e-05, [1]
    [Cycle 1]: 1.589e-05, [1]
        [cse]: 9.82999e-06
[auto_monad_reorder]: 1.414e-05
[get_jit_bprop_graph]: 2.35002e-06
[rewriter_after_jit_bprop_graph]: 6.21998e-06
[opt_after_jit_grad]: 0.00049495
[symbol_engine_optimizer]: 8.039e-05, [1]
    [Cycle 1]: 7.232e-05, [6]
        [build]: 4.27e-06
        [elim_shapecalc]: 8.40999e-06
        [elim_not_effective]: 1.444e-05
        [opt_reshape]: 6.31998e-06
        [fold_const_symbol]: 8.92e-06
        [renormalize]: 1.25001e-06
[validate]: 3.745e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.00627082
[execute]: 6.51e-06
Sums
    bootstrap                            :     0.000310s :  2.08%
    type_inference                       :     0.004487s : 30.15%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000057s :  0.38%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000038s :  0.25%
    rewriter_before_opt_a                :     0.000069s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.37%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000604s :  4.06%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000745s :  5.01%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.12%
    jit_opt_a.cse                        :     0.000034s :  0.23%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000042s :  0.28%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000580s :  3.89%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000026s :  0.17%
    loop_unroll                          :     0.000448s :  3.01%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000045s :  0.30%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000495s :  3.32%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000037s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006271s : 42.12%
    execute                              :     0.000007s :  0.04%


Time group info:
------[substitution.]   0.000189    23
 1.19% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.87% :     0.000005s :      3: substitution.graph_param_transform
83.12% :     0.000157s :      4: substitution.inline
 1.86% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.33% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000004s :      2: substitution.replace_old_param
 6.13% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004427     2
79.77% :     0.003532s :      1: type_inference.infer
20.23% :     0.000895s :      1: type_inference.specialize
------[replace.]   0.000069     6
67.61% :     0.000047s :      4: replace.inline
32.39% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000165     6
93.68% :     0.000155s :      4: match.inline
 6.32% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.27% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.50% :     0.000002s :     10: predicate.addn_zero_filter
 1.67% :     0.000002s :     10: predicate.arithmetic_simplify
 1.49% :     0.000002s :     10: predicate.cast_eliminate
 0.67% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.54% :     0.000002s :     10: predicate.depend_value_elim
 1.20% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.50% :     0.000001s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.17% :     0.000006s :     22: predicate.inline
 1.06% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.47% :     0.000002s :      6: predicate.less_batch_normalization
 1.65% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.89% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.95% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.12% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.65% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000002s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.53% :     0.000002s :     10: predicate.reduce_eliminate
 1.56% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.84% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.49% :     0.000001s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.13% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.20% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.55% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 2.43% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.25% :     0.000008s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.26% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.28% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.58% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000613    12
47.54% :     0.000291s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.91% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.54% :     0.000291s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020453    76
 0.24% :     0.000048s :      1: add_recomputation
 0.29% :     0.000060s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.59% :     0.000324s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
25.69% :     0.005255s :      1: jit_opt_a
 0.74% :     0.000152s :      1: jit_opt_after_cconv
 0.31% :     0.000063s :      1: jit_opt_b
 2.23% :     0.000456s :      1: loop_unroll
 2.89% :     0.000590s :      1: mutable_eliminate
 3.96% :     0.000809s :     26: opt.transform.jit_opt_a
 0.24% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000032s :      4: opt.transform.jit_opt_b
 0.07% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000035s :      4: opt.transform.symbol_engine_opt
 2.47% :     0.000505s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000040s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.90% :     0.000389s :      1: renormalize.infer
 1.71% :     0.000349s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000045s :      1: rewriter_after_opt_a
 0.35% :     0.000072s :      1: rewriter_before_opt_a
 0.41% :     0.000084s :      1: symbol_engine_optimizer
30.74% :     0.006287s :      1: task_emit
22.01% :     0.004501s :      1: type_inference
 0.29% :     0.000059s :      1: validate


TotalTime = 0.0111311, [30]
[bootstrap]: 0.00024634
[type_inference]: 0.00385434
[event_method]: 1.656e-05
[auto_monad]: 5.935e-05
[graph_reusing]: 5.81e-06
[pre_auto_parallel]: 2.21e-06
[py_interpret_to_execute]: 3.041e-05
[rewriter_before_opt_a]: 7.616e-05
[expand_dump_flag]: 3.03998e-06
[jit_opt_a]: 0.0048182, [2]
    [Cycle 1]: 0.0014124, [27]
        [switch_simplify]: 4.98e-05
        [loop_unroll]: 2.914e-05
        [a_1]: 0.00051599
        [with_stream_mark]: 1.455e-05
        [recompute_prepare]: 7.26001e-06
        [updatestate_depend_eliminate]: 3.26001e-06
        [updatestate_assign_eliminate]: 2.68998e-06
        [updatestate_loads_eliminate]: 2.38998e-06
        [parameter_eliminate]: 1.71998e-06
        [specialize_transform]: 6.21e-06
        [updatestate_useless_node_eliminater]: 5.79999e-06
        [accelerated_algorithm]: 6.02001e-06
        [meta_shard_fg_expand]: 2.01998e-06
        [get_grad_eliminate_]: 5.97999e-06
        [merge_forward]: 3.81001e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.495e-05
        [j_node_and_user_rematch]: 8.74998e-06
        [meta_fg_expand]: 2.42001e-06
        [replace_old_param]: 8.26002e-06
        [inline_without_move]: 5.28002e-06
        [renormalize]: 0.00048964
        [add_forward_monad_depend]: 5.10001e-06
        [auto_monad_grad]: 1.97001e-06
        [auto_monad_eliminator]: 1.261e-05
        [cse]: 2.424e-05
        [replace_applicator]: 1.209e-05
    [Cycle 2]: 0.00034152, [27]
        [switch_simplify]: 6.43e-06
        [loop_unroll]: 5.66e-06
        [a_1]: 9.643e-05
        [with_stream_mark]: 8.70999e-06
        [recompute_prepare]: 6.09001e-06
        [updatestate_depend_eliminate]: 2.60002e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 5.15999e-06
        [updatestate_useless_node_eliminater]: 5.34e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 1.24998e-06
        [get_grad_eliminate_]: 5.14e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.376e-05
        [j_node_and_user_rematch]: 8.19002e-06
        [meta_fg_expand]: 1.48002e-06
        [replace_old_param]: 7.38e-06
        [inline_without_move]: 4.79e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.05001e-06
        [cse]: 9.89001e-06
        [replace_applicator]: 5.77999e-06
[py_interpret_to_execute_after_opt_a]: 9.49999e-06
[rewriter_after_opt_a]: 3.157e-05
[convert_after_rewriter]: 7.7e-06
[order_py_execute_after_rewriter]: 6.30002e-06
[mutable_eliminate]: 0.00048149
[jit_opt_b]: 5.446e-05, [1]
    [Cycle 1]: 4.727e-05, [2]
        [frontend_op_eliminate]: 1.829e-05
        [inline_after_opt_a]: 1.679e-05
[cconv]: 2.335e-05
[loop_unroll]: 0.00043711
[jit_opt_after_cconv]: 0.00013977, [1]
    [Cycle 1]: 0.00013335, [11]
        [c_1]: 2.076e-05
        [parameter_eliminate]: 2.74001e-06
        [updatestate_depend_eliminate]: 5.54e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.667e-05
        [call_graph_tuple_transform]: 1.824e-05
        [tuple_list_get_item_eliminator]: 6.51e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 5.47999e-06
[remove_dup_value]: 1.303e-05
[partial_unused_args_eliminate]: 2.68e-06
[environ_conv]: 5.15999e-06
[add_recomputation]: 4.626e-05
[cse_after_recomputation]: 1.982e-05, [1]
    [Cycle 1]: 1.434e-05, [1]
        [cse]: 8.72998e-06
[auto_monad_reorder]: 1.42e-05
[get_jit_bprop_graph]: 1.44e-06
[rewriter_after_jit_bprop_graph]: 4.59002e-06
[opt_after_jit_grad]: 0.00046646
[symbol_engine_optimizer]: 7.146e-05, [1]
    [Cycle 1]: 6.505e-05, [6]
        [build]: 3.37002e-06
        [elim_shapecalc]: 7.44002e-06
        [elim_not_effective]: 1.19e-05
        [opt_reshape]: 5.58002e-06
        [fold_const_symbol]: 8.85999e-06
        [renormalize]: 4.19997e-07
[validate]: 2.768e-05
Sums
    bootstrap                            :     0.000246s :  3.29%
    type_inference                       :     0.003854s : 51.42%
    event_method                         :     0.000017s :  0.22%
    auto_monad                           :     0.000059s :  0.79%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000030s :  0.41%
    rewriter_before_opt_a                :     0.000076s :  1.02%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000056s :  0.75%
    jit_opt_a.loop_unroll                :     0.000035s :  0.46%
    jit_opt_a.a_1                        :     0.000612s :  8.17%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.18%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.15%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.15%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.38%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000490s :  6.53%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.24%
    jit_opt_a.cse                        :     0.000034s :  0.46%
    jit_opt_a.replace_applicator         :     0.000018s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.13%
    rewriter_after_opt_a                 :     0.000032s :  0.42%
    convert_after_rewriter               :     0.000008s :  0.10%
    order_py_execute_after_rewriter      :     0.000006s :  0.08%
    mutable_eliminate                    :     0.000481s :  6.42%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.24%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.22%
    cconv                                :     0.000023s :  0.31%
    loop_unroll                          :     0.000437s :  5.83%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.09%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000013s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.04%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000046s :  0.62%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000014s :  0.19%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000466s :  6.22%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000028s :  0.37%


Time group info:
------[substitution.]   0.000174    23
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000001s :      2: substitution.fold_const_symbol
 2.82% :     0.000005s :      3: substitution.graph_param_transform
82.22% :     0.000143s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.81% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000004s :      2: substitution.replace_old_param
 6.76% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003803     2
79.15% :     0.003010s :      1: type_inference.infer
20.85% :     0.000793s :      1: type_inference.specialize
------[replace.]   0.000064     6
65.36% :     0.000042s :      4: replace.inline
34.64% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000152     6
93.03% :     0.000141s :      4: match.inline
 6.97% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.55% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.06% :     0.000003s :     10: predicate.arithmetic_simplify
 1.12% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.66% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.33% :     0.000002s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.33% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.41% :     0.000002s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.55% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.65% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.63% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.53% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     10: predicate.same_eliminate
 0.63% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.64% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.24% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.24% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.87% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.86% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.13% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000519    12
47.59% :     0.000247s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.08% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.33% :     0.000246s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012514    72
 0.39% :     0.000049s :      1: add_recomputation
 0.50% :     0.000063s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.07% :     0.000258s :      1: bootstrap
 0.21% :     0.000026s :      1: cconv
 0.08% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.53% :     0.004821s :      1: jit_opt_a
 1.14% :     0.000143s :      1: jit_opt_after_cconv
 0.46% :     0.000057s :      1: jit_opt_b
 3.56% :     0.000445s :      1: loop_unroll
 3.91% :     0.000490s :      1: mutable_eliminate
 6.57% :     0.000822s :     26: opt.transform.jit_opt_a
 0.38% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000028s :      4: opt.transform.jit_opt_b
 0.10% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.79% :     0.000474s :      1: opt_after_jit_grad
 0.07% :     0.000009s :      1: order_py_execute_after_rewriter
 0.04% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000033s :      1: py_interpret_to_execute
 0.10% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000016s :      1: remove_dup_value
 1.88% :     0.000235s :      1: renormalize.infer
 1.98% :     0.000247s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000035s :      1: rewriter_after_opt_a
 0.63% :     0.000079s :      1: rewriter_before_opt_a
 0.59% :     0.000074s :      1: symbol_engine_optimizer
30.88% :     0.003864s :      1: type_inference


TotalTime = 0.0110207, [30]
[bootstrap]: 0.00025887
[type_inference]: 0.00388112
[event_method]: 1.659e-05
[auto_monad]: 5.273e-05
[graph_reusing]: 4.50001e-06
[pre_auto_parallel]: 2.29999e-06
[py_interpret_to_execute]: 2.991e-05
[rewriter_before_opt_a]: 8.639e-05
[expand_dump_flag]: 2.43002e-06
[jit_opt_a]: 0.00471715, [2]
    [Cycle 1]: 0.00143297, [27]
        [switch_simplify]: 4.54e-05
        [loop_unroll]: 3.058e-05
        [a_1]: 0.00049464
        [with_stream_mark]: 1.377e-05
        [recompute_prepare]: 7.21001e-06
        [updatestate_depend_eliminate]: 3.29001e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.19999e-06
        [parameter_eliminate]: 1.77999e-06
        [specialize_transform]: 6.01e-06
        [updatestate_useless_node_eliminater]: 5.19e-06
        [accelerated_algorithm]: 6.23998e-06
        [meta_shard_fg_expand]: 1.50001e-06
        [get_grad_eliminate_]: 5.50001e-06
        [merge_forward]: 2.53e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.294e-05
        [j_node_and_user_rematch]: 8.92e-06
        [meta_fg_expand]: 1.92001e-06
        [replace_old_param]: 7.91001e-06
        [inline_without_move]: 5.49e-06
        [renormalize]: 0.00055037
        [add_forward_monad_depend]: 4.65001e-06
        [auto_monad_grad]: 1.67001e-06
        [auto_monad_eliminator]: 1.101e-05
        [cse]: 1.974e-05
        [replace_applicator]: 1.069e-05
    [Cycle 2]: 0.00032263, [27]
        [switch_simplify]: 5.62001e-06
        [loop_unroll]: 4.80001e-06
        [a_1]: 9.121e-05
        [with_stream_mark]: 8.82e-06
        [recompute_prepare]: 4.91002e-06
        [updatestate_depend_eliminate]: 2.59001e-06
        [updatestate_assign_eliminate]: 2.28998e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 5.12e-06
        [updatestate_useless_node_eliminater]: 5.27999e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 1.03001e-06
        [get_grad_eliminate_]: 4.63999e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.55999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.194e-05
        [j_node_and_user_rematch]: 7.66001e-06
        [meta_fg_expand]: 1.71e-06
        [replace_old_param]: 7e-06
        [inline_without_move]: 4.64002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.30999e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 4.98001e-06
        [cse]: 1.058e-05
        [replace_applicator]: 5.10001e-06
[py_interpret_to_execute_after_opt_a]: 8.45999e-06
[rewriter_after_opt_a]: 2.815e-05
[convert_after_rewriter]: 7.2e-06
[order_py_execute_after_rewriter]: 4.55001e-06
[mutable_eliminate]: 0.00048219
[jit_opt_b]: 5.074e-05, [1]
    [Cycle 1]: 4.392e-05, [2]
        [frontend_op_eliminate]: 1.713e-05
        [inline_after_opt_a]: 1.494e-05
[cconv]: 1.989e-05
[loop_unroll]: 0.00042706
[jit_opt_after_cconv]: 0.00013467, [1]
    [Cycle 1]: 0.00012831, [11]
        [c_1]: 2.031e-05
        [parameter_eliminate]: 2.49999e-06
        [updatestate_depend_eliminate]: 5.27001e-06
        [updatestate_assign_eliminate]: 2.36998e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [cse]: 1.593e-05
        [call_graph_tuple_transform]: 1.741e-05
        [tuple_list_get_item_eliminator]: 5.44e-06
        [none_parameter_eliminate]: 1.30999e-06
        [renormalize]: 3.80009e-07
        [switch_simplify]: 5.75001e-06
[remove_dup_value]: 1.163e-05
[partial_unused_args_eliminate]: 2.01e-06
[environ_conv]: 4.18999e-06
[add_recomputation]: 4.094e-05
[cse_after_recomputation]: 2.114e-05, [1]
    [Cycle 1]: 1.552e-05, [1]
        [cse]: 9.66e-06
[auto_monad_reorder]: 1.33e-05
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 4.55999e-06
[opt_after_jit_grad]: 0.00045709
[symbol_engine_optimizer]: 7.183e-05, [1]
    [Cycle 1]: 6.547e-05, [6]
        [build]: 3.16001e-06
        [elim_shapecalc]: 7.73999e-06
        [elim_not_effective]: 1.269e-05
        [opt_reshape]: 5.88002e-06
        [fold_const_symbol]: 8.70999e-06
        [renormalize]: 3.19997e-07
[validate]: 2.735e-05
Sums
    bootstrap                            :     0.000259s :  3.45%
    type_inference                       :     0.003881s : 51.73%
    event_method                         :     0.000017s :  0.22%
    auto_monad                           :     0.000053s :  0.70%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000030s :  0.40%
    rewriter_before_opt_a                :     0.000086s :  1.15%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000051s :  0.68%
    jit_opt_a.loop_unroll                :     0.000035s :  0.47%
    jit_opt_a.a_1                        :     0.000586s :  7.81%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000550s :  7.34%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.21%
    jit_opt_a.cse                        :     0.000030s :  0.40%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000028s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.10%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000482s :  6.43%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.20%
    cconv                                :     0.000020s :  0.27%
    loop_unroll                          :     0.000427s :  5.69%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.08%
    remove_dup_value                     :     0.000012s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000041s :  0.55%
    cse_after_recomputation.cse          :     0.000010s :  0.13%
    auto_monad_reorder                   :     0.000013s :  0.18%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000457s :  6.09%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.36%


Time group info:
------[substitution.]   0.000169    23
 1.24% :     0.000002s :      2: substitution.elim_not_effective
 0.79% :     0.000001s :      2: substitution.fold_const_symbol
 2.67% :     0.000005s :      3: substitution.graph_param_transform
83.05% :     0.000140s :      4: substitution.inline
 1.72% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.50% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000003s :      2: substitution.replace_old_param
 6.07% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003829     2
80.58% :     0.003085s :      1: type_inference.infer
19.42% :     0.000744s :      1: type_inference.specialize
------[replace.]   0.000063     6
65.98% :     0.000042s :      4: replace.inline
34.02% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000147     6
93.81% :     0.000138s :      4: match.inline
 6.19% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.26% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.21% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.54% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.32% :     0.000002s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.29% :     0.000006s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.76% :     0.000002s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.25% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.71% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     10: predicate.reduce_eliminate
 1.57% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.57% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.56% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.63% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.60% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.83% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000535    12
50.08% :     0.000268s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.04% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.88% :     0.000240s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012428    72
 0.35% :     0.000044s :      1: add_recomputation
 0.45% :     0.000056s :      1: auto_monad
 0.13% :     0.000016s :      1: auto_monad_reorder
 2.18% :     0.000271s :      1: bootstrap
 0.18% :     0.000023s :      1: cconv
 0.08% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
37.98% :     0.004720s :      1: jit_opt_a
 1.11% :     0.000137s :      1: jit_opt_after_cconv
 0.43% :     0.000053s :      1: jit_opt_b
 3.49% :     0.000434s :      1: loop_unroll
 3.94% :     0.000490s :      1: mutable_eliminate
 6.32% :     0.000785s :     26: opt.transform.jit_opt_a
 0.37% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000026s :      4: opt.transform.jit_opt_b
 0.10% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.74% :     0.000465s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000032s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000014s :      1: remove_dup_value
 2.29% :     0.000285s :      1: renormalize.infer
 2.09% :     0.000259s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000031s :      1: rewriter_after_opt_a
 0.72% :     0.000090s :      1: rewriter_before_opt_a
 0.60% :     0.000074s :      1: symbol_engine_optimizer
31.31% :     0.003891s :      1: type_inference


TotalTime = 0.0227704, [33]
[bootstrap]: 0.00031404
[type_inference]: 0.00478611
[event_method]: 1.883e-05
[auto_monad]: 6.64e-05
[graph_reusing]: 5.42999e-06
[pre_auto_parallel]: 3.21999e-06
[py_interpret_to_execute]: 4.41e-05
[rewriter_before_opt_a]: 7.836e-05
[expand_dump_flag]: 3.65e-06
[jit_opt_a]: 0.00634631, [2]
    [Cycle 1]: 0.00218508, [27]
        [switch_simplify]: 6.721e-05
        [loop_unroll]: 3.048e-05
        [a_1]: 0.00061672
        [with_stream_mark]: 2.223e-05
        [recompute_prepare]: 1.054e-05
        [updatestate_depend_eliminate]: 4.4e-06
        [updatestate_assign_eliminate]: 3.28998e-06
        [updatestate_loads_eliminate]: 2.68e-06
        [parameter_eliminate]: 2.55002e-06
        [specialize_transform]: 7e-06
        [updatestate_useless_node_eliminater]: 5.59e-06
        [accelerated_algorithm]: 6.37001e-06
        [meta_shard_fg_expand]: 3.41999e-06
        [get_grad_eliminate_]: 5.69e-06
        [merge_forward]: 4.38999e-06
        [cell_reuse_recompute_pass]: 2.53998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.667e-05
        [j_node_and_user_rematch]: 1.022e-05
        [meta_fg_expand]: 3.05998e-06
        [replace_old_param]: 1.006e-05
        [inline_without_move]: 5.68002e-06
        [renormalize]: 0.00105202
        [add_forward_monad_depend]: 7.93001e-06
        [auto_monad_grad]: 3.03e-06
        [auto_monad_eliminator]: 1.622e-05
        [cse]: 3.185e-05
        [replace_applicator]: 1.8e-05
    [Cycle 2]: 0.00035565, [27]
        [switch_simplify]: 6.41e-06
        [loop_unroll]: 5.66e-06
        [a_1]: 9.98e-05
        [with_stream_mark]: 1.315e-05
        [recompute_prepare]: 5.22e-06
        [updatestate_depend_eliminate]: 3.18e-06
        [updatestate_assign_eliminate]: 2.28998e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 5.15999e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.64e-06
        [meta_shard_fg_expand]: 1.76998e-06
        [get_grad_eliminate_]: 4.53001e-06
        [merge_forward]: 2.64999e-06
        [cell_reuse_recompute_pass]: 4.01001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.681e-05
        [j_node_and_user_rematch]: 8.70001e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 8.89998e-06
        [inline_without_move]: 4.88001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.66e-06
        [auto_monad_grad]: 9.90025e-07
        [auto_monad_eliminator]: 5.69999e-06
        [cse]: 1.07e-05
        [replace_applicator]: 4.92e-06
[py_interpret_to_execute_after_opt_a]: 1.557e-05
[rewriter_after_opt_a]: 4.24e-05
[convert_after_rewriter]: 7.74002e-06
[order_py_execute_after_rewriter]: 5.09e-06
[mutable_eliminate]: 0.00082907
[jit_opt_b]: 6.017e-05, [1]
    [Cycle 1]: 5.037e-05, [2]
        [frontend_op_eliminate]: 2.159e-05
        [inline_after_opt_a]: 1.652e-05
[cconv]: 2.924e-05
[loop_unroll]: 0.00064595
[jit_opt_after_cconv]: 0.00017058, [1]
    [Cycle 1]: 0.0001624, [11]
        [c_1]: 2.386e-05
        [parameter_eliminate]: 4.20999e-06
        [updatestate_depend_eliminate]: 8.50999e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.23002e-06
        [cse]: 2.776e-05
        [call_graph_tuple_transform]: 2.505e-05
        [tuple_list_get_item_eliminator]: 5.66998e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 8.80013e-07
        [switch_simplify]: 6.02001e-06
[remove_dup_value]: 1.567e-05
[partial_unused_args_eliminate]: 2.17001e-06
[environ_conv]: 6.02999e-06
[add_recomputation]: 5.698e-05
[cse_after_recomputation]: 2.281e-05, [1]
    [Cycle 1]: 1.6e-05, [1]
        [cse]: 9.52999e-06
[auto_monad_reorder]: 1.611e-05
[get_jit_bprop_graph]: 2.01e-06
[rewriter_after_jit_bprop_graph]: 9.14e-06
[opt_after_jit_grad]: 0.00065001
[symbol_engine_optimizer]: 8.599e-05, [1]
    [Cycle 1]: 7.681e-05, [6]
        [build]: 4.55999e-06
        [elim_shapecalc]: 8.42e-06
        [elim_not_effective]: 1.633e-05
        [opt_reshape]: 7.23e-06
        [fold_const_symbol]: 9.25999e-06
        [renormalize]: 8.60018e-07
[validate]: 4.786e-05
[backend_pass]: 1.03001e-06
[task_emit]: 0.00812386
[execute]: 9.36002e-06
Sums
    bootstrap                            :     0.000314s :  1.72%
    type_inference                       :     0.004786s : 26.24%
    event_method                         :     0.000019s :  0.10%
    auto_monad                           :     0.000066s :  0.36%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000044s :  0.24%
    rewriter_before_opt_a                :     0.000078s :  0.43%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.40%
    jit_opt_a.loop_unroll                :     0.000036s :  0.20%
    jit_opt_a.a_1                        :     0.000717s :  3.93%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.10%
    jit_opt_a.inline_without_move        :     0.000011s :  0.06%
    jit_opt_a.renormalize                :     0.001052s :  5.77%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.12%
    jit_opt_a.cse                        :     0.000043s :  0.23%
    jit_opt_a.replace_applicator         :     0.000023s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.09%
    rewriter_after_opt_a                 :     0.000042s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000829s :  4.55%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.09%
    cconv                                :     0.000029s :  0.16%
    loop_unroll                          :     0.000646s :  3.54%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.03%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000057s :  0.31%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000016s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.05%
    opt_after_jit_grad                   :     0.000650s :  3.56%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000048s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.008124s : 44.54%
    execute                              :     0.000009s :  0.05%


Time group info:
------[substitution.]   0.000256    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.73% :     0.000007s :      3: substitution.graph_param_transform
84.14% :     0.000216s :      4: substitution.inline
 1.76% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.33% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.27% :     0.000006s :      2: substitution.replace_old_param
 5.28% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004704     2
78.35% :     0.003686s :      1: type_inference.infer
21.65% :     0.001018s :      1: type_inference.specialize
------[replace.]   0.000088     6
66.90% :     0.000059s :      4: replace.inline
33.10% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000226     6
94.53% :     0.000213s :      4: match.inline
 5.47% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   691
 1.17% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.03% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 2.31% :     0.000003s :     10: predicate.arithmetic_simplify
 1.48% :     0.000002s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.92% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000001s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.34% :     0.000002s :     10: predicate.environ_get_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 5.78% :     0.000008s :     22: predicate.inline
 0.74% :     0.000001s :      6: predicate.inline_without_move
 0.30% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.64% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.47% :     0.000002s :     15: predicate.load_eliminater
 1.82% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.87% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.92% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 2.00% :     0.000003s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000003s :     15: predicate.partial_eliminate
 1.06% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.36% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.96% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.25% :     0.000002s :      6: predicate.special_op_eliminate
 0.88% :     0.000001s :      6: predicate.specialize_transform
 1.17% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.84% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.65% :     0.000004s :     16: predicate.switch_defer_inline
 1.82% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.50% :     0.000010s :     53: predicate.switch_simplify
 1.10% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000002s :     10: predicate.transpose_eliminate
 1.70% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.16% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.38% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.66% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.93% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000694    12
43.61% :     0.000302s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.13% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.26% :     0.000349s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024902    76
 0.24% :     0.000060s :      1: add_recomputation
 0.28% :     0.000070s :      1: auto_monad
 0.08% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.33% :     0.000332s :      1: bootstrap
 0.13% :     0.000032s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000010s :      1: environ_conv
 0.09% :     0.000023s :      1: event_method
 0.06% :     0.000015s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
25.50% :     0.006351s :      1: jit_opt_a
 0.70% :     0.000174s :      1: jit_opt_after_cconv
 0.25% :     0.000063s :      1: jit_opt_b
 2.64% :     0.000657s :      1: loop_unroll
 3.38% :     0.000842s :      1: mutable_eliminate
 3.85% :     0.000959s :     26: opt.transform.jit_opt_a
 0.23% :     0.000057s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000030s :      4: opt.transform.jit_opt_b
 0.07% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.66% :     0.000663s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.19% :     0.000047s :      1: py_interpret_to_execute
 0.07% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 2.26% :     0.000564s :      1: renormalize.infer
 1.92% :     0.000478s :      1: renormalize.specialize
 0.05% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000047s :      1: rewriter_after_opt_a
 0.33% :     0.000082s :      1: rewriter_before_opt_a
 0.36% :     0.000089s :      1: symbol_engine_optimizer
32.71% :     0.008145s :      1: task_emit
19.28% :     0.004802s :      1: type_inference
 0.29% :     0.000073s :      1: validate


TotalTime = 0.0112911, [30]
[bootstrap]: 0.00025275
[type_inference]: 0.00388886
[event_method]: 1.648e-05
[auto_monad]: 5.716e-05
[graph_reusing]: 5.67001e-06
[pre_auto_parallel]: 2.48002e-06
[py_interpret_to_execute]: 3.163e-05
[rewriter_before_opt_a]: 6.88e-05
[expand_dump_flag]: 2.83e-06
[jit_opt_a]: 0.00494609, [2]
    [Cycle 1]: 0.00150721, [27]
        [switch_simplify]: 5.073e-05
        [loop_unroll]: 2.99e-05
        [a_1]: 0.00055348
        [with_stream_mark]: 1.571e-05
        [recompute_prepare]: 7.34002e-06
        [updatestate_depend_eliminate]: 3.21999e-06
        [updatestate_assign_eliminate]: 2.62001e-06
        [updatestate_loads_eliminate]: 2.78003e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 5.98998e-06
        [updatestate_useless_node_eliminater]: 5.27001e-06
        [accelerated_algorithm]: 5.79999e-06
        [meta_shard_fg_expand]: 1.94999e-06
        [get_grad_eliminate_]: 5.52999e-06
        [merge_forward]: 3.23998e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.377e-05
        [j_node_and_user_rematch]: 9.24998e-06
        [meta_fg_expand]: 2.43e-06
        [replace_old_param]: 8.40001e-06
        [inline_without_move]: 5.50001e-06
        [renormalize]: 0.0005363
        [add_forward_monad_depend]: 5.26002e-06
        [auto_monad_grad]: 2.09999e-06
        [auto_monad_eliminator]: 1.306e-05
        [cse]: 2.582e-05
        [replace_applicator]: 1.168e-05
    [Cycle 2]: 0.00034022, [27]
        [switch_simplify]: 5.56998e-06
        [loop_unroll]: 5.06002e-06
        [a_1]: 9.98e-05
        [with_stream_mark]: 8.91002e-06
        [recompute_prepare]: 5.22e-06
        [updatestate_depend_eliminate]: 2.56998e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 2.01998e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 5.92001e-06
        [updatestate_useless_node_eliminater]: 6.09999e-06
        [accelerated_algorithm]: 5.94999e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.60001e-06
        [merge_forward]: 2.54001e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.253e-05
        [j_node_and_user_rematch]: 8.74e-06
        [meta_fg_expand]: 1.57999e-06
        [replace_old_param]: 7.5e-06
        [inline_without_move]: 7.93001e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 5.32999e-06
        [cse]: 1.053e-05
        [replace_applicator]: 5.25999e-06
[py_interpret_to_execute_after_opt_a]: 9.05001e-06
[rewriter_after_opt_a]: 3.005e-05
[convert_after_rewriter]: 6.91001e-06
[order_py_execute_after_rewriter]: 5.30999e-06
[mutable_eliminate]: 0.00051955
[jit_opt_b]: 5.086e-05, [1]
    [Cycle 1]: 4.421e-05, [2]
        [frontend_op_eliminate]: 1.736e-05
        [inline_after_opt_a]: 1.494e-05
[cconv]: 2.378e-05
[loop_unroll]: 0.00041411
[jit_opt_after_cconv]: 0.00013734, [1]
    [Cycle 1]: 0.00013073, [11]
        [c_1]: 2.183e-05
        [parameter_eliminate]: 2.49999e-06
        [updatestate_depend_eliminate]: 5.32999e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 1.564e-05
        [call_graph_tuple_transform]: 1.779e-05
        [tuple_list_get_item_eliminator]: 5.56998e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 5.40999e-06
[remove_dup_value]: 1.36e-05
[partial_unused_args_eliminate]: 2.35002e-06
[environ_conv]: 5.14e-06
[add_recomputation]: 4.606e-05
[cse_after_recomputation]: 2.013e-05, [1]
    [Cycle 1]: 1.465e-05, [1]
        [cse]: 8.95999e-06
[auto_monad_reorder]: 1.427e-05
[get_jit_bprop_graph]: 1.54e-06
[rewriter_after_jit_bprop_graph]: 5.34e-06
[opt_after_jit_grad]: 0.00045393
[symbol_engine_optimizer]: 7.561e-05, [1]
    [Cycle 1]: 6.86e-05, [6]
        [build]: 3.66001e-06
        [elim_shapecalc]: 8.31002e-06
        [elim_not_effective]: 1.268e-05
        [opt_reshape]: 6.11e-06
        [fold_const_symbol]: 8.65999e-06
        [renormalize]: 6.99976e-07
[validate]: 2.953e-05
Sums
    bootstrap                            :     0.000253s :  3.32%
    type_inference                       :     0.003889s : 51.03%
    event_method                         :     0.000016s :  0.22%
    auto_monad                           :     0.000057s :  0.75%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.42%
    rewriter_before_opt_a                :     0.000069s :  0.90%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000056s :  0.74%
    jit_opt_a.loop_unroll                :     0.000035s :  0.46%
    jit_opt_a.a_1                        :     0.000653s :  8.57%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.15%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.24%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.21%
    jit_opt_a.inline_without_move        :     0.000013s :  0.18%
    jit_opt_a.renormalize                :     0.000536s :  7.04%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.24%
    jit_opt_a.cse                        :     0.000036s :  0.48%
    jit_opt_a.replace_applicator         :     0.000017s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.12%
    rewriter_after_opt_a                 :     0.000030s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000520s :  6.82%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.20%
    cconv                                :     0.000024s :  0.31%
    loop_unroll                          :     0.000414s :  5.43%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.29%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000046s :  0.60%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000014s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000454s :  5.96%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000030s :  0.39%


Time group info:
------[substitution.]   0.000194    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000001s :      2: substitution.fold_const_symbol
 2.36% :     0.000005s :      3: substitution.graph_param_transform
84.02% :     0.000163s :      4: substitution.inline
 1.58% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.41% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.80% :     0.000003s :      2: substitution.replace_old_param
 6.08% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003831     2
80.46% :     0.003083s :      1: type_inference.infer
19.54% :     0.000749s :      1: type_inference.specialize
------[replace.]   0.000069     6
67.79% :     0.000047s :      4: replace.inline
32.21% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000171     6
93.75% :     0.000160s :      4: match.inline
 6.25% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000130   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 2.93% :     0.000004s :     10: predicate.arithmetic_simplify
 1.17% :     0.000002s :     10: predicate.cast_eliminate
 0.55% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.50% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.26% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.03% :     0.000007s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :      6: predicate.less_batch_normalization
 1.68% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.12% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.77% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.79% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.67% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.84% :     0.000002s :     10: predicate.reduce_eliminate
 1.76% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.55% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.54% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.99% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.49% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.19% :     0.000002s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.57% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.12% :     0.000009s :     53: predicate.switch_simplify
 1.45% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.92% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.83% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.86% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000511    12
48.77% :     0.000249s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.08% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.15% :     0.000231s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012761    72
 0.38% :     0.000049s :      1: add_recomputation
 0.47% :     0.000060s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.08% :     0.000266s :      1: bootstrap
 0.21% :     0.000026s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.78% :     0.004949s :      1: jit_opt_a
 1.10% :     0.000140s :      1: jit_opt_after_cconv
 0.42% :     0.000053s :      1: jit_opt_b
 3.31% :     0.000422s :      1: loop_unroll
 4.14% :     0.000529s :      1: mutable_eliminate
 6.78% :     0.000865s :     26: opt.transform.jit_opt_a
 0.37% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.18% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.62% :     0.000461s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000034s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000016s :      1: remove_dup_value
 2.12% :     0.000270s :      1: renormalize.infer
 2.03% :     0.000259s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000033s :      1: rewriter_after_opt_a
 0.56% :     0.000072s :      1: rewriter_before_opt_a
 0.61% :     0.000078s :      1: symbol_engine_optimizer
30.56% :     0.003900s :      1: type_inference


TotalTime = 0.115809, [33]
[bootstrap]: 0.00021003
[type_inference]: 0.0859296
[event_method]: 0.00036102
[auto_monad]: 0.00011019
[graph_reusing]: 7.66999e-06
[pre_auto_parallel]: 3.04999e-06
[py_interpret_to_execute]: 3.626e-05
[rewriter_before_opt_a]: 0.00011484
[expand_dump_flag]: 3.33e-06
[jit_opt_a]: 0.0268773, [3]
    [Cycle 1]: 0.017622, [27]
        [switch_simplify]: 8.991e-05
        [loop_unroll]: 4.576e-05
        [a_1]: 0.00117857
        [with_stream_mark]: 4.128e-05
        [recompute_prepare]: 3.665e-05
        [updatestate_depend_eliminate]: 1.04e-05
        [updatestate_assign_eliminate]: 6.96001e-06
        [updatestate_loads_eliminate]: 6.90998e-06
        [parameter_eliminate]: 3.79002e-06
        [specialize_transform]: 1.802e-05
        [updatestate_useless_node_eliminater]: 1.363e-05
        [accelerated_algorithm]: 1.768e-05
        [meta_shard_fg_expand]: 3.2e-06
        [get_grad_eliminate_]: 1.531e-05
        [merge_forward]: 1.079e-05
        [cell_reuse_recompute_pass]: 1.63002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 3.61e-05
        [j_node_and_user_rematch]: 3.116e-05
        [meta_fg_expand]: 0.00229338
        [replace_old_param]: 7.189e-05
        [inline_without_move]: 6.371e-05
        [renormalize]: 0.0129222
        [add_forward_monad_depend]: 1.389e-05
        [auto_monad_grad]: 7.46001e-06
        [auto_monad_eliminator]: 6.057e-05
        [cse]: 0.00023016
        [replace_applicator]: 9.542e-05
    [Cycle 2]: 0.00397625, [27]
        [switch_simplify]: 4.353e-05
        [loop_unroll]: 4.173e-05
        [a_1]: 0.00125152
        [with_stream_mark]: 2.238e-05
        [recompute_prepare]: 1.338e-05
        [updatestate_depend_eliminate]: 4.95001e-06
        [updatestate_assign_eliminate]: 4.12003e-06
        [updatestate_loads_eliminate]: 4.45e-06
        [parameter_eliminate]: 1.79998e-06
        [specialize_transform]: 8.52e-06
        [updatestate_useless_node_eliminater]: 7.32002e-06
        [accelerated_algorithm]: 8.16002e-06
        [meta_shard_fg_expand]: 3.03998e-06
        [get_grad_eliminate_]: 9.17001e-06
        [merge_forward]: 4.47998e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.772e-05
        [j_node_and_user_rematch]: 1.326e-05
        [meta_fg_expand]: 7.945e-05
        [replace_old_param]: 1.042e-05
        [inline_without_move]: 7.77e-06
        [renormalize]: 0.00217284
        [add_forward_monad_depend]: 6.73998e-06
        [auto_monad_grad]: 2.15002e-06
        [auto_monad_eliminator]: 1.267e-05
        [cse]: 2.577e-05
        [replace_applicator]: 1.928e-05
    [Cycle 3]: 0.00026147, [27]
        [switch_simplify]: 5.00999e-06
        [loop_unroll]: 4.20999e-06
        [a_1]: 4.917e-05
        [with_stream_mark]: 1.072e-05
        [recompute_prepare]: 4.47998e-06
        [updatestate_depend_eliminate]: 2.76999e-06
        [updatestate_assign_eliminate]: 1.98997e-06
        [updatestate_loads_eliminate]: 1.76e-06
        [parameter_eliminate]: 1.37e-06
        [specialize_transform]: 3.76999e-06
        [updatestate_useless_node_eliminater]: 3.46001e-06
        [accelerated_algorithm]: 4.05e-06
        [meta_shard_fg_expand]: 1.81e-06
        [get_grad_eliminate_]: 3.63e-06
        [merge_forward]: 3.04001e-06
        [cell_reuse_recompute_pass]: 4.03999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.237e-05
        [j_node_and_user_rematch]: 6.25002e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 3.41999e-06
        [inline_without_move]: 3.00998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 6.39993e-07
        [auto_monad_eliminator]: 3.43e-06
        [cse]: 8.47998e-06
        [replace_applicator]: 3.88001e-06
[py_interpret_to_execute_after_opt_a]: 1.038e-05
[rewriter_after_opt_a]: 3.424e-05
[convert_after_rewriter]: 5.84e-06
[order_py_execute_after_rewriter]: 4.47e-06
[mutable_eliminate]: 0.00064422
[jit_opt_b]: 4.239e-05, [1]
    [Cycle 1]: 3.519e-05, [2]
        [frontend_op_eliminate]: 1.172e-05
        [inline_after_opt_a]: 1.203e-05
[cconv]: 2.233e-05
[loop_unroll]: 0.00042918
[jit_opt_after_cconv]: 0.00011995, [1]
    [Cycle 1]: 0.00011396, [11]
        [c_1]: 1.268e-05
        [parameter_eliminate]: 2.27999e-06
        [updatestate_depend_eliminate]: 4.07e-06
        [updatestate_assign_eliminate]: 1.71002e-06
        [updatestate_loads_eliminate]: 1.43002e-06
        [cse]: 1.925e-05
        [call_graph_tuple_transform]: 1.393e-05
        [tuple_list_get_item_eliminator]: 3.7e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 3.65998e-06
[remove_dup_value]: 1.44e-05
[partial_unused_args_eliminate]: 2.32999e-06
[environ_conv]: 5.22999e-06
[add_recomputation]: 3.569e-05
[cse_after_recomputation]: 1.88e-05, [1]
    [Cycle 1]: 1.347e-05, [1]
        [cse]: 7.91001e-06
[auto_monad_reorder]: 1.285e-05
[get_jit_bprop_graph]: 1.35999e-06
[rewriter_after_jit_bprop_graph]: 5.25999e-06
[opt_after_jit_grad]: 0.00044317
[symbol_engine_optimizer]: 6.251e-05, [1]
    [Cycle 1]: 5.647e-05, [6]
        [build]: 2.93e-06
        [elim_shapecalc]: 5.69e-06
        [elim_not_effective]: 8.47998e-06
        [opt_reshape]: 4.75001e-06
        [fold_const_symbol]: 6.36e-06
        [renormalize]: 6.39993e-07
[validate]: 2.708e-05
[backend_pass]: 9.00007e-07
[task_emit]: 2.464e-05
[execute]: 1.25001e-06
Sums
    bootstrap                            :     0.000210s :  0.19%
    type_inference                       :     0.085930s : 78.19%
    event_method                         :     0.000361s :  0.33%
    auto_monad                           :     0.000110s :  0.10%
    graph_reusing                        :     0.000008s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000036s :  0.03%
    rewriter_before_opt_a                :     0.000115s :  0.10%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000138s :  0.13%
    jit_opt_a.loop_unroll                :     0.000092s :  0.08%
    jit_opt_a.a_1                        :     0.002479s :  2.26%
    jit_opt_a.with_stream_mark           :     0.000074s :  0.07%
    jit_opt_a.recompute_prepare          :     0.000055s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000018s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000013s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000007s :  0.01%
    jit_opt_a.specialize_transform       :     0.000030s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000030s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000028s :  0.03%
    jit_opt_a.merge_forward              :     0.000018s :  0.02%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000007s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000066s :  0.06%
    jit_opt_a.j_node_and_user_rematch    :     0.000051s :  0.05%
    jit_opt_a.meta_fg_expand             :     0.002374s :  2.16%
    jit_opt_a.replace_old_param          :     0.000086s :  0.08%
    jit_opt_a.inline_without_move        :     0.000074s :  0.07%
    jit_opt_a.renormalize                :     0.015095s : 13.74%
    jit_opt_a.add_forward_monad_depend   :     0.000022s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000010s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000077s :  0.07%
    jit_opt_a.cse                        :     0.000264s :  0.24%
    jit_opt_a.replace_applicator         :     0.000119s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.01%
    rewriter_after_opt_a                 :     0.000034s :  0.03%
    convert_after_rewriter               :     0.000006s :  0.01%
    order_py_execute_after_rewriter      :     0.000004s :  0.00%
    mutable_eliminate                    :     0.000644s :  0.59%
    jit_opt_b.frontend_op_eliminate      :     0.000012s :  0.01%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.01%
    cconv                                :     0.000022s :  0.02%
    loop_unroll                          :     0.000429s :  0.39%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.01%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.00%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.cse              :     0.000019s :  0.02%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.01%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.00%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.00%
    remove_dup_value                     :     0.000014s :  0.01%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000005s :  0.00%
    add_recomputation                    :     0.000036s :  0.03%
    cse_after_recomputation.cse          :     0.000008s :  0.01%
    auto_monad_reorder                   :     0.000013s :  0.01%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.00%
    opt_after_jit_grad                   :     0.000443s :  0.40%
    symbol_engine_optimizer.build        :     0.000003s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000008s :  0.01%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.00%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000027s :  0.02%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000025s :  0.02%
    execute                              :     0.000001s :  0.00%


Time group info:
------[substitution.]   0.000697   102
 0.20% :     0.000001s :      1: substitution.elim_not_effective
 0.16% :     0.000001s :      1: substitution.fold_const_symbol
 0.65% :     0.000005s :      1: substitution.graph_param_transform
65.18% :     0.000454s :     12: substitution.inline
 3.17% :     0.000022s :      2: substitution.inline_without_move
 1.59% :     0.000011s :     12: substitution.j_node_and_user_rematch
 1.82% :     0.000013s :      7: substitution.minmaximum_grad
 1.48% :     0.000010s :      5: substitution.partial_eliminate
 1.74% :     0.000012s :     12: substitution.remove_not_recompute_node
 5.57% :     0.000039s :     10: substitution.replace_applicator
 0.94% :     0.000007s :      7: substitution.replace_old_param
 1.19% :     0.000008s :      1: substitution.reshape_eliminate
 0.94% :     0.000007s :      1: substitution.set_cell_output_no_recompute
 1.93% :     0.000013s :      2: substitution.tile_eliminate
 3.82% :     0.000027s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.30% :     0.000016s :      7: substitution.tuple_list_get_item_depend_reorder
 7.33% :     0.000051s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.085814     2
97.06% :     0.083287s :      1: type_inference.infer
 2.94% :     0.002527s :      1: type_inference.specialize
------[replace.]   0.000195    19
66.19% :     0.000129s :     12: replace.inline
33.81% :     0.000066s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000472    19
94.33% :     0.000445s :     12: match.inline
 5.67% :     0.000027s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000416  2646
 1.64% :     0.000007s :     46: predicate.accumulaten_eliminater
 0.23% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.33% :     0.000006s :     46: predicate.addn_check_dump
 1.65% :     0.000007s :     46: predicate.addn_zero_filter
 2.73% :     0.000011s :     46: predicate.arithmetic_simplify
 1.79% :     0.000007s :     46: predicate.cast_eliminate
 0.08% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.39% :     0.000006s :     46: predicate.compare_switch_simplify
 1.72% :     0.000007s :     46: predicate.depend_value_elim
 1.31% :     0.000005s :     46: predicate.dict_get_item_const_eliminator
 1.55% :     0.000006s :     46: predicate.dict_get_item_eliminator
 1.40% :     0.000006s :     46: predicate.dict_set_item_eliminator
 0.16% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.06% :     0.000000s :      1: predicate.elim_not_effective
 0.11% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.60% :     0.000007s :     46: predicate.environ_add_const_eliminate
 1.27% :     0.000005s :     46: predicate.environ_get_add_eliminate
 1.29% :     0.000005s :     46: predicate.environ_get_depend_swap
 1.74% :     0.000007s :     46: predicate.environ_get_eliminate
 1.31% :     0.000005s :     46: predicate.environ_get_set_eliminate
 0.03% :     0.000000s :      1: predicate.fold_const_symbol
 1.10% :     0.000005s :     19: predicate.get_grad_eliminate
 0.04% :     0.000000s :      1: predicate.graph_param_transform
 4.55% :     0.000019s :     67: predicate.inline
 1.83% :     0.000008s :     43: predicate.inline_without_move
 0.34% :     0.000001s :     19: predicate.j_node_and_user_rematch
 1.17% :     0.000005s :     19: predicate.less_batch_normalization
 2.27% :     0.000009s :     53: predicate.list_to_tuple_eliminator_
 1.80% :     0.000007s :     54: predicate.load_eliminater
 0.28% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.19% :     0.000013s :     95: predicate.loop_unroll_before_grad
 1.69% :     0.000007s :     47: predicate.make_slice_get_slice_eliminator
 1.42% :     0.000006s :     46: predicate.merge_addn
 1.34% :     0.000006s :     46: predicate.minmaximum_grad
 0.31% :     0.000001s :      1: predicate.mutable_eliminate
 0.10% :     0.000000s :      1: predicate.opt_reshape
 2.25% :     0.000009s :     54: predicate.partial_eliminate
 1.47% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.93% :     0.000008s :     46: predicate.reduce_eliminate
 1.91% :     0.000008s :     53: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000002s :     19: predicate.remove_not_recompute_node
 2.63% :     0.000011s :    100: predicate.replace_applicator
 1.15% :     0.000005s :     43: predicate.replace_old_param
 0.04% :     0.000000s :      1: predicate.reset_defer_inline
 1.81% :     0.000008s :     46: predicate.reshape_eliminate
 1.60% :     0.000007s :     46: predicate.row_tensor_add_zeros_like
 0.20% :     0.000001s :      1: predicate.row_tensor_eliminate
 2.16% :     0.000009s :     46: predicate.same_eliminate
 0.49% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.18% :     0.000001s :      2: predicate.special_op_eliminate
 0.81% :     0.000003s :     19: predicate.specialize_transform
 1.74% :     0.000007s :     46: predicate.split_environ_get_set_with_tuple_value
 1.57% :     0.000007s :     46: predicate.stack_unstack_eliminate
 0.06% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.82% :     0.000012s :     65: predicate.switch_defer_inline
 2.41% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.04% :     0.000025s :    161: predicate.switch_simplify
 1.70% :     0.000007s :     46: predicate.tile_eliminate
 1.83% :     0.000008s :     46: predicate.transpose_eliminate
 1.89% :     0.000008s :     46: predicate.tuple_list_convert_item_index_to_positive
 1.97% :     0.000008s :     46: predicate.tuple_list_get_item_depend_reorder
 3.17% :     0.000013s :     55: predicate.tuple_list_get_item_eliminator
 1.94% :     0.000008s :     46: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000007s :     53: predicate.tuple_to_list_eliminator_
 1.62% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.54% :     0.000011s :     73: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000007s :     46: predicate.value_based_eliminate
 0.05% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.19% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.004350    52
77.42% :     0.003368s :     36: func_graph_cloner_run.FuncGraphClonerGraph
22.58% :     0.000982s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.134160    91
 0.03% :     0.000038s :      1: add_recomputation
 0.09% :     0.000119s :      1: auto_monad
 0.01% :     0.000015s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.16% :     0.000221s :      1: bootstrap
 0.02% :     0.000025s :      1: cconv
 0.01% :     0.000008s :      1: convert_after_rewriter
 0.02% :     0.000021s :      1: cse_after_recomputation
 0.01% :     0.000007s :      1: environ_conv
 0.28% :     0.000370s :      1: event_method
 0.00% :     0.000003s :      1: execute
 0.00% :     0.000005s :      1: expand_dump_flag
 0.00% :     0.000003s :      1: get_jit_bprop_graph
 0.01% :     0.000010s :      1: graph_reusing
20.04% :     0.026881s :      1: jit_opt_a
 0.09% :     0.000123s :      1: jit_opt_after_cconv
 0.03% :     0.000045s :      1: jit_opt_b
 0.33% :     0.000438s :      1: loop_unroll
 0.49% :     0.000653s :      1: mutable_eliminate
 2.39% :     0.003210s :     39: opt.transform.jit_opt_a
 0.02% :     0.000031s :      4: opt.transform.jit_opt_after_cconv
 0.01% :     0.000017s :      4: opt.transform.jit_opt_b
 0.01% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.01% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.01% :     0.000014s :      1: opt.transform.opt_after_jit_grad
 0.02% :     0.000022s :      4: opt.transform.symbol_engine_opt
 0.34% :     0.000451s :      1: opt_after_jit_grad
 0.00% :     0.000007s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.00% :     0.000005s :      1: pre_auto_parallel
 0.03% :     0.000040s :      1: py_interpret_to_execute
 0.01% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.01% :     0.000017s :      1: remove_dup_value
 9.51% :     0.012759s :      2: renormalize.infer
 1.72% :     0.002314s :      2: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.03% :     0.000037s :      1: rewriter_after_opt_a
 0.09% :     0.000118s :      1: rewriter_before_opt_a
 0.05% :     0.000065s :      1: symbol_engine_optimizer
 0.02% :     0.000029s :      1: task_emit
64.06% :     0.085944s :      1: type_inference
 0.03% :     0.000043s :      1: validate


TotalTime = 0.016459, [33]
[bootstrap]: 0.00028958
[type_inference]: 0.00431675
[event_method]: 1.59e-05
[auto_monad]: 5.532e-05
[graph_reusing]: 4.28999e-06
[pre_auto_parallel]: 2.39001e-06
[py_interpret_to_execute]: 3.32e-05
[rewriter_before_opt_a]: 6.596e-05
[expand_dump_flag]: 2.66999e-06
[jit_opt_a]: 0.00476563, [2]
    [Cycle 1]: 0.00129676, [27]
        [switch_simplify]: 4.329e-05
        [loop_unroll]: 2.851e-05
        [a_1]: 0.00045697
        [with_stream_mark]: 1.199e-05
        [recompute_prepare]: 6.12001e-06
        [updatestate_depend_eliminate]: 2.57001e-06
        [updatestate_assign_eliminate]: 1.98002e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 7.7e-07
        [specialize_transform]: 5.48002e-06
        [updatestate_useless_node_eliminater]: 4.64998e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 9.79984e-07
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 2.19001e-06
        [cell_reuse_recompute_pass]: 9.60019e-07
        [cell_reuse_handle_not_recompute_node_pass]: 9.60001e-06
        [j_node_and_user_rematch]: 8.18999e-06
        [meta_fg_expand]: 1.60999e-06
        [replace_old_param]: 7.71999e-06
        [inline_without_move]: 4.93001e-06
        [renormalize]: 0.00049449
        [add_forward_monad_depend]: 4.32998e-06
        [auto_monad_grad]: 1.09e-06
        [auto_monad_eliminator]: 9.27001e-06
        [cse]: 1.507e-05
        [replace_applicator]: 1.082e-05
    [Cycle 2]: 0.00030763, [27]
        [switch_simplify]: 5.09e-06
        [loop_unroll]: 5.14003e-06
        [a_1]: 8.746e-05
        [with_stream_mark]: 8.52998e-06
        [recompute_prepare]: 4.59998e-06
        [updatestate_depend_eliminate]: 2.32001e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 4.75001e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 4.96997e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 4.62e-06
        [merge_forward]: 2.26998e-06
        [cell_reuse_recompute_pass]: 1.46002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.13e-05
        [j_node_and_user_rematch]: 7.24001e-06
        [meta_fg_expand]: 1.70001e-06
        [replace_old_param]: 7.21001e-06
        [inline_without_move]: 4.54002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 4.50001e-06
        [cse]: 9.64999e-06
        [replace_applicator]: 4.79e-06
[py_interpret_to_execute_after_opt_a]: 2.067e-05
[rewriter_after_opt_a]: 2.444e-05
[convert_after_rewriter]: 5.29e-06
[order_py_execute_after_rewriter]: 3.81999e-06
[mutable_eliminate]: 0.00045
[jit_opt_b]: 4.563e-05, [1]
    [Cycle 1]: 3.999e-05, [2]
        [frontend_op_eliminate]: 1.511e-05
        [inline_after_opt_a]: 1.403e-05
[cconv]: 1.573e-05
[loop_unroll]: 0.00037243
[jit_opt_after_cconv]: 0.0001254, [1]
    [Cycle 1]: 0.00012021, [11]
        [c_1]: 1.974e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 4.02e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.671e-05
        [call_graph_tuple_transform]: 1.473e-05
        [tuple_list_get_item_eliminator]: 5.24998e-06
        [none_parameter_eliminate]: 7.09988e-07
        [renormalize]: 3.39991e-07
        [switch_simplify]: 4.95001e-06
[remove_dup_value]: 6.19001e-06
[partial_unused_args_eliminate]: 1.38002e-06
[environ_conv]: 3.55e-06
[add_recomputation]: 3.004e-05
[cse_after_recomputation]: 1.89e-05, [1]
    [Cycle 1]: 1.39e-05, [1]
        [cse]: 8.29998e-06
[auto_monad_reorder]: 9.09e-06
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 4.32003e-06
[opt_after_jit_grad]: 0.00044564
[symbol_engine_optimizer]: 6.776e-05, [1]
    [Cycle 1]: 6.12e-05, [6]
        [build]: 2.26e-06
        [elim_shapecalc]: 
TotalTime = 0.0190149, [33]
[bootstrap]: 0.00040304
[type_inference]: 0.00494792
[event_method]: 1.713e-05
[auto_monad]: 6.16e-05
[graph_reusing]: 5.69999e-06
[pre_auto_parallel]: 2.59001e-06
[py_interpret_to_execute]: 3.945e-05
[rewriter_before_opt_a]: 7.279e-05
[expand_dump_flag]: 3.66001e-06
[jit_opt_a]: 0.00570492, [2]
    [Cycle 1]: 0.00171742, [27]
        [switch_simplify]: 5.101e-05
        [loop_unroll]: 2.881e-05
        [a_1]: 0.00054675
        [with_stream_mark]: 1.431e-05
        [recompute_prepare]: 6.88998e-06
        [updatestate_depend_eliminate]: 3.17002e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 6.21998e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.96e-06
        [meta_shard_fg_expand]: 1.45001e-06
        [get_grad_eliminate_]: 5.89e-06
        [merge_forward]: 2.42001e-06
        [cell_reuse_recompute_pass]: 8.79983e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.254e-05
        [j_node_and_user_rematch]: 9.22001e-06
        [meta_fg_expand]: 1.82999e-06
        [replace_old_param]: 8.79e-06
        [inline_without_move]: 5.55001e-06
        [renormalize]: 0.00079338
        [add_forward_monad_depend]: 4.28001e-06
        [auto_monad_grad]: 1.28002e-06
        [auto_monad_eliminator]: 9.64e-06
        [cse]: 1.611e-05
        [replace_applicator]: 1.003e-05
    [Cycle 2]: 0.00031501, [27]
        [switch_simplify]: 5.14e-06
        [loop_unroll]: 5.09e-06
        [a_1]: 9.094e-05
        [with_stream_mark]: 6.39999e-06
        [recompute_prepare]: 4.62998e-06
        [updatestate_depend_eliminate]: 2.40002e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 4.87e-06
        [updatestate_useless_node_eliminater]: 4.80999e-06
        [accelerated_algorithm]: 5.17999e-06
        [meta_shard_fg_expand]: 1.11002e-06
        [get_grad_eliminate_]: 4.28999e-06
        [merge_forward]: 2.30002e-06
        [cell_reuse_recompute_pass]: 2.40002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.118e-05
        [j_node_and_user_rematch]: 7.95998e-06
        [meta_fg_expand]: 1.48002e-06
        [replace_old_param]: 6.39999e-06
        [inline_without_move]: 4.61002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.19998e-06
        [auto_monad_grad]: 8.59989e-07
        [auto_monad_eliminator]: 4.89998e-06
        [cse]: 9.47001e-06
        [replace_applicator]: 4.89003e-06
[py_interpret_to_execute_after_opt_a]: 9.29e-06
[rewriter_after_opt_a]: 2.905e-05
[convert_after_rewriter]: 6.28e-06
[order_py_execute_after_rewriter]: 4.93001e-06
[mutable_eliminate]: 0.00050204
[jit_opt_b]: 4.813e-05, [1]
    [Cycle 1]: 4.191e-05, [2]
        [frontend_op_eliminate]: 1.653e-05
        [inline_after_opt_a]: 1.446e-05
[cconv]: 1.512e-05
[loop_unroll]: 0.00042843
[jit_opt_after_cconv]: 0.00013191, [1]
    [Cycle 1]: 0.00012558, [11]
        [c_1]: 2.003e-05
        [parameter_eliminate]: 2.37999e-06
        [updatestate_depend_eliminate]: 4.43001e-06
        [updatestate_assign_eliminate]: 2.62001e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [cse]: 1.626e-05
        [call_graph_tuple_transform]: 1.574e-05
        [tuple_list_get_item_eliminator]: 5.69e-06
        [none_parameter_eliminate]: 1.09003e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.22e-06
[remove_dup_value]: 6.93e-06
[partial_unused_args_eliminate]: 1.70001e-06
[environ_conv]: 3.82002e-06
[add_recomputation]: 3.573e-05
[cse_after_recomputation]: 1.992e-05, [1]
    [Cycle 1]: 1.451e-05, [1]
        [cse]: 8.66002e-06
[auto_monad_reorder]: 9.81e-06
[get_jit_bprop_graph]: 1.43002e-06
[rewriter_after_jit_bprop_graph]: 4.56002e-06
[opt_after_jit_grad]: 0.00043032
[symbol_engine_optimizer]: 6.922e-05, [1]
    [Cycle 1]: 6.282e-05, [6]
        [build]: 2.51e-06
        [elim_shapeca7.01999e-06
        [elim_not_effective]: 1.149e-05
        [opt_reshape]: 5.64998e-06
        [fold_const_symbol]: 7.75e-06
        [renormalize]: 5.09986e-07
[validate]: 2.42e-05
[backend_pass]: 6.10016e-07
[task_emit]: 0.00505114
[execute]: 6.79001e-06
Sums
    bootstrap                            :     0.000290s :  2.27%
    type_inference                       :     0.004317s : 33.86%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000055s :  0.43%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.26%
    rewriter_before_opt_a                :     0.000066s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000048s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.26%
    jit_opt_a.a_1                        :     0.000544s :  4.27%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000021s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000015s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.12%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000495s :  3.88%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.11%
    jit_opt_a.cse                        :     0.000025s :  0.19%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000021s :  0.16%
    rewriter_after_opt_a                 :     0.000024s :  0.19%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000450s :  3.53%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000016s :  0.12%
    loop_unroll                          :     0.000372s :  2.92%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                    lc]: 7.2e-06
        [elim_not_effective]: 1.169e-05
        [opt_reshape]: 5.57001e-06
        [fold_const_symbol]: 7.81001e-06
        [renormalize]: 6.89994e-07
[validate]: 2.727e-05
[backend_pass]: 7.59988e-07
[task_emit]: 0.00576706
[execute]: 6.78998e-06
Sums
    bootstrap                            :     0.000403s :  2.73%
    type_inference                       :     0.004948s : 33.54%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000062s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000039s :  0.27%
    rewriter_before_opt_a                :     0.000073s :  0.49%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000638s :  4.32%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000793s :  5.38%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.10%
    jit_opt_a.cse                        :     0.000026s :  0.17%
    jit_opt_a.replace_applicator         :     0.000015s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000029s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000502s :  3.40%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000015s :  0.10%
    loop_unroll                          :     0.000428s :  2.90%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                :     0.000006s :  0.05%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000030s :  0.24%
    cse_after_recomputation.cse          :     0.000008s :  0.07%
    auto_monad_reorder                   :     0.000009s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000446s :  3.50%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000024s :  0.19%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.005051s : 39.62%
    execute                              :     0.000007s :  0.05%

      :     0.000007s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000036s :  0.24%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000010s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000430s :  2.92%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000027s :  0.18%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005767s : 39.10%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000145    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.55% :     0.000004s :      3: substitution.graph_param_transform
84.17% :     0.000122s :      4: substitution.inline
 1.65% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.49% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.33% :     0.000003s :      2: substitution.replace_old_param
 5.19% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004258     2
82.14% :     0.003498s :      1: type_inference.infer
17.86% :     0.000761s :      1: type_inference.specialize
------[replace.]   0.000059     6
65.96% :     0.000039s :      4: replace.inline
34.04% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000126     6
94.84% :     0.000120s :      4: match.inline
 5.16% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000107   691
 1.34% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000001s :     10: predicate.addn_zero_filter
 2.10% :     0.000002s :     10: predicate.arithmetic_simplify
 1.37% :     0.000001s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.23% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.95% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.24% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.17% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.45% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.90% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.29% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.63% :     0.000003s :     16: predicate.switch_defer_inline
 2.29% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.67% :     0.000008s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.49% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.78% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.59% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000487    12
46.44% :     0.000226s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.89% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.67% :     0.000237s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017781    76
 0.18% :     0.000033s :      1: add_recomputation
 0.33% :     0.000059s :      1: auto_monad
 0.07% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.71% :     0.000304s :      1: bootstrap
 0.10% :     0.000018s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.11% :     0.000019s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.82% :     0.004768s :      1: jit_opt_a
 0.72% :     0.000128s :      1: jit_opt_after_cconv
 0.27% :     0.000048s :      1: jit_opt_b
 2.13% :     0.000378s :      1: loop_unroll
 2.57% :     0.000457s :      1: mutable_eliminate
 4.12% :     0.000732s :     26: opt.transform.jit_opt_a
 0.23% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000023s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.55% :     0.000453s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000036s :      1: py_interpret_to_execute
 0.13% :     0.000024s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000009s :      1: remove_dup_value
 1.40% :     0.000249s :      1: renormalize.infer
 1.35% :     0.000240s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000028s :      1: rewriter_after_opt_a
 0.39% :     0.000069s :      1: rewriter_before_opt_a
 0.40% :     0.000070s :      1: symbol_engine_optimizer
28.47% :     0.005061s :      1: task_emit
24.35% :     0.004330s :      1: type_inference
 0.22% :     0.000039s :      1: validate


Time group info:
------[substitution.]   0.000203    23
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.44% :     0.000001s :      2: substitution.fold_const_symbol
 1.76% :     0.000004s :      3: substitution.graph_param_transform
85.40% :     0.000173s :      4: substitution.inline
 1.60% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.28% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.59% :     0.000003s :      2: substitution.replace_old_param
 5.99% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004880     2
81.50% :     0.003977s :      1: type_inference.infer
18.50% :     0.000903s :      1: type_inference.specialize
------[replace.]   0.000076     6
65.35% :     0.000050s :      4: replace.inline
34.65% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000182     6
93.93% :     0.000171s :      4: match.inline
 6.07% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.48% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 2.38% :     0.000003s :     10: predicate.arithmetic_simplify
 1.36% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.41% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.47% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.41% :     0.000002s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 4.88% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.34% :     0.000002s :      6: predicate.less_batch_normalization
 1.82% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.93% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.79% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.40% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000002s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.91% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.63% :     0.000001s :      3:[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:20.380.765 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c3713a0> should be wrapped by jit only once.
 predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.28% :     0.000001s :      6: predicate.special_op_eliminate
 1.15% :     0.000001s :      6: predicate.specialize_transform
 1.62% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.64% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.17% :     0.000008s :     53: predicate.switch_simplify
 1.29% :     0.000001s :     10: predicate.tile_eliminate
 1.61% :     0.000002s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.43% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.06% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.85% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000666    12
49.00% :     0.000326s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.81% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.19% :     0.000308s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020734    76
 0.19% :     0.000039s :      1: add_recomputation
 0.32% :     0.000066s :      1: auto_monad
 0.06% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 2.04% :     0.000424s :      1: bootstrap
 0.08% :     0.000018s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.53% :     0.005709s :      1: jit_opt_a
 0.65% :     0.000134s :      1: jit_opt_after_cconv
 0.24% :     0.000050s :      1: jit_opt_b
 2.10% :     0.000435s :      1: loop_unroll
 2.45% :     0.000509s :      1: mutable_eliminate
 4.04% :     0.000838s :     26: opt.transform.jit_opt_a
 0.21% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000010s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.11% :     0.000438s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000043s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000009s :      1: remove_dup_value
 2.10% :     0.000435s :      1: renormalize.infer
 1.70% :     0.000352s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000032s :      1: rewriter_after_opt_a
 0.36% :     0.000076s :      1: rewriter_before_opt_a
 0.35% :     0.000072s :      1: symbol_engine_optimizer
27.87% :     0.005778s :      1: task_emit
23.93% :     0.004961s :      1: type_inference
 0.21% :     0.000044s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:20.381.526 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0114652, [30]
[bootstrap]: 0.00026981
[type_inference]: 0.00406806
[event_method]: 1.734e-05
[auto_monad]: 6.598e-05
[graph_reusing]: 4.95001e-06
[pre_auto_parallel]: 2.68e-06
[py_interpret_to_execute]: 3.191e-05
[rewriter_before_opt_a]: 7.266e-05
[expand_dump_flag]: 3.01001e-06
[jit_opt_a]: 0.00494105, [2]
    [Cycle 1]: 0.00146432, [27]
        [switch_simplify]: 5.22e-05
        [loop_unroll]: 4.435e-05
        [a_1]: 0.00052123
        [with_stream_mark]: 1.572e-05
        [recompute_prepare]: 7.12997e-06
        [updatestate_depend_eliminate]: 3.35e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 5.80002e-06
        [updatestate_useless_node_eliminater]: 5.22999e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 2.19999e-06
        [get_grad_eliminate_]: 5.17e-06
        [merge_forward]: 3.66001e-06
        [cell_reuse_recompute_pass]: 1.10999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.377e-05
        [j_node_and_user_rematch]: 8.80999e-06
        [meta_fg_expand]: 2.48e-06
        [replace_old_param]: 8.58001e-06
        [inline_without_move]: 5.32999e-06
        [renormalize]: 0.00051408
        [add_forward_monad_depend]: 4.71002e-06
        [auto_monad_grad]: 2.36998e-06
        [auto_monad_eliminator]: 1.411e-05
        [cse]: 3.092e-05
        [replace_applicator]: 1.114e-05
    [Cycle 2]: 0.00031833, [27]
        [switch_simplify]: 5.72001e-06
        [loop_unroll]: 5.10999e-06
        [a_1]: 8.899e-05
        [with_stream_mark]: 8.14002e-06
        [recompute_prepare]: 4.97999e-06
        [updatestate_depend_eliminate]: 2.39999e-06
        [updatestate_assign_eliminate]: 1.99999e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 9.29984e-07
        [specialize_transform]: 5.02e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.33002e-06
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 2.26998e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.169e-05
        [j_node_and_user_rematch]: 8.18999e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 6.88e-06
        [inline_without_move]: 4.75999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 5.29e-06
        [cse]: 1.06e-05
        [replace_applicator]: 5.07e-06
[py_interpret_to_execute_after_opt_a]: 8.47998e-06
[rewriter_after_opt_a]: 3.161e-05
[convert_after_rewriter]: 7.50998e-06
[order_py_execute_after_rewriter]: 4.84e-06
[mutable_eliminate]: 0.00048311
[jit_opt_b]: 5.101e-05, [1]
    [Cycle 1]: 4.395e-05, [2]
        [frontend_op_eliminate]: 1.692e-05
        [inline_after_opt_a]: 1.485e-05
[cconv]: 2.465e-05
[loop_unroll]: 0.0004118
[jit_opt_after_cconv]: 0.00013955, [1]
    [Cycle 1]: 0.00013349, [11]
        [c_1]: 2.163e-05
        [parameter_eliminate]: 2.44999e-06
        [updatestate_depend_eliminate]: 5.09e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.655e-05
        [call_graph_tuple_transform]: 1.777e-05
        [tuple_list_get_item_eliminator]: 5.47001e-06
        [none_parameter_eliminate]: 1.67001e-06
        [renormalize]: 5.00004e-07
        [switch_simplify]: 5.37001e-06
[remove_dup_value]: 1.386e-05
[partial_unused_args_eliminate]: 2.66e-06
[environ_conv]: 4.18001e-06
[add_recomputation]: 4.605e-05
[cse_after_recomputation]: 2.03e-05, [1]
    [Cycle 1]: 1.479e-05, [1]
        [cse]: 9.20001e-06
[auto_monad_reorder]: 1.453e-05
[get_jit_bprop_graph]: 1.51002e-06
[rewriter_after_jit_bprop_graph]: 4.55999e-06
[opt_after_jit_grad]: 0.00045006
[symbol_engine_optimizer]: 7.29e-05, [1]
    [Cycle 1]: 6.624e-05, [6]
        [build]: 3.44001e-06
        [elim_shapecalc]: 7.6e-06
        [elim_not_effective]: 1.251e-05
        [opt_reshape]: 5.88002e-06
        [fold_const_symbol]: 8.80001e-06
        [renormalize]: 4.59986e-07
[validate]: 3.244e-05
Sums
    bootstrap                            :     0.000270s :  3.49%
    type_inference                       :     0.004068s : 52.59%
    event_method                         :     0.000017s :  0.22%
    auto_monad                           :     0.000066s :  0.85%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.41%
    rewriter_before_opt_a                :     0.000073s :  0.94%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000058s :  0.75%
    jit_opt_a.loop_unroll                :     0.000049s :  0.64%
    jit_opt_a.a_1                        :     0.000610s :  7.89%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000514s :  6.65%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.25%
    jit_opt_a.cse                        :     0.000042s :  0.54%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000032s :  0.41%
    convert_after_rewriter               :     0.000008s :  0.10%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000483s :  6.25%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000025s :  0.32%
    loop_unroll                          :     0.000412s :  5.32%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.18%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000046s :  0.60%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000015s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000450s :  5.82%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000032s :  0.42%


Time group info:
------[substitution.]   0.000188    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000001s :      2: substitution.fold_const_symbol
 2.61% :     0.000005s :      3: substitution.graph_param_transform
83.84% :     0.000158s :      4: substitution.inline
 1.54% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000004s :      2: substitution.replace_old_param
 6.00% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004004     2
80.28% :     0.003214s :      1: type_inference.infer
19.72% :     0.000789s :      1: type_inference.specialize
------[replace.]   0.000067     6
67.03% :     0.000045s :      4: replace.inline
32.97% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000165     6
93.89% :     0.000155s :      4: match.inline
 6.11% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000001s :     10: predicate.addn_zero_filter
 1.82% :     0.000002s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.41% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.49% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.53% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.58% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     15: predicate.partial_eliminate
 1.27% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.26% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.39% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.63% :     0.000003s :     16: predicate.switch_defer_inline
 2.41% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.97% :     0.000009s :     53: predicate.switch_simplify
 1.31% :     0.000002s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.87% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000514    12
50.45% :     0.000259s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.70% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.86% :     0.000225s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012869    72
 0.38% :     0.000049s :      1: add_recomputation
 0.54% :     0.000069s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.21% :     0.000285s :      1: bootstrap
 0.21% :     0.000027s :      1: cconv
 0.08% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.17% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
38.41% :     0.004944s :      1: jit_opt_a
 1.11% :     0.000142s :      1: jit_opt_after_cconv
 0.42% :     0.000054s :      1: jit_opt_b
 3.26% :     0.000420s :      1: loop_unroll
 3.82% :     0.000492s :      1: mutable_eliminate
 6.43% :     0.000827s :     26: opt.transform.jit_opt_a
 0.36% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.56% :     0.000458s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.04% :     0.000005s :      1: partial_unused_args_eliminate
 0.04% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000035s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000017s :      1: remove_dup_value
 2.02% :     0.000259s :      1: renormalize.infer
 1.93% :     0.000248s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.29% :     0.000037s :      1: rewriter_after_opt_a
 0.59% :     0.000075s :      1: rewriter_before_opt_a
 0.59% :     0.000076s :      1: symbol_engine_optimizer
31.70% :     0.004080s :      1: type_inference


TotalTime = 0.0183923, [33]
[bootstrap]: 0.00026914
[type_inference]: 0.0041578
[event_method]: 1.575e-05
[auto_monad]: 6.127e-05
[graph_reusing]: 5.97001e-06
[pre_auto_parallel]: 2.49999e-06
[py_interpret_to_execute]: 3.312e-05
[rewriter_before_opt_a]: 7.117e-05
[expand_dump_flag]: 3.3e-06
[jit_opt_a]: 0.00515015, [2]
    [Cycle 1]: 0.00162353, [27]
        [switch_simplify]: 5.325e-05
        [loop_unroll]: 2.862e-05
        [a_1]: 0.00051148
        [with_stream_mark]: 1.501e-05
        [recompute_prepare]: 7.30998e-06
        [updatestate_depend_eliminate]: 3.24001e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 6.14999e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 5.35001e-06
        [meta_shard_fg_expand]: 2.37999e-06
        [get_grad_eliminate_]: 5.30001e-06
        [merge_forward]: 3.47002e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.543e-05
        [j_node_and_user_rematch]: 9.25001e-06
        [meta_fg_expand]: 2.73e-06
        [replace_old_param]: 8.33001e-06
        [inline_without_move]: 5.22e-06
        [renormalize]: 0.00069469
        [add_forward_monad_depend]: 5.60001e-06
        [auto_monad_grad]: 2.88998e-06
        [auto_monad_eliminator]: 1.272e-05
        [cse]: 2.796e-05
        [replace_applicator]: 1.313e-05
    [Cycle 2]: 0.00034339, [27]
        [switch_simplify]: 5.58997e-06
        [loop_unroll]: 4.59998e-06
        [a_1]: 0.00010053
        [with_stream_mark]: 9.65002e-06
        [recompute_prepare]: 5.62999e-06
        [updatestate_depend_eliminate]: 2.69999e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 4.79e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 5.12999e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 4.70999e-06
        [merge_forward]: 3.3e-06
        [cell_reuse_recompute_pass]: 1.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.292e-05
        [j_node_and_user_rematch]: 7.88001e-06
        [meta_fg_expand]: 1.64998e-06
        [replace_old_param]: 7.48999e-06
        [inline_without_move]: 4.70999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.42e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.92999e-06
        [cse]: 1.108e-05
        [replace_applicator]: 4.97999e-06
[py_interpret_to_execute_after_opt_a]: 1.083e-05
[rewriter_after_opt_a]: 3.539e-05
[convert_after_rewriter]: 7.33999e-06
[order_py_execute_after_rewriter]: 5.44e-06
[mutable_eliminate]: 0.0006299
[jit_opt_b]: 5.296e-05, [1]
    [Cycle 1]: 4.472e-05, [2]
        [frontend_op_eliminate]: 1.795e-05
        [inline_after_opt_a]: 1.574e-05
[cconv]: 2.628e-05
[loop_unroll]: 0.00043602
[jit_opt_after_cconv]: 0.00014733, [1]
    [Cycle 1]: 0.00014017, [11]
        [c_1]: 2.063e-05
        [parameter_eliminate]: 2.84001e-06
        [updatestate_depend_eliminate]: 5.57999e-06
        [updatestate_assign_eliminate]: 2.61e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [cse]: 2.336e-05
        [call_graph_tuple_transform]: 1.925e-05
        [tuple_list_get_item_eliminator]: 5.25999e-06
        [none_parameter_eliminate]: 1.25001e-06
        [renormalize]: 8.50006e-07
        [switch_simplify]: 5.22999e-06
[remove_dup_value]: 1.404e-05
[partial_unused_args_eliminate]: 2.02999e-06
[environ_conv]: 5.66e-06
[add_recomputation]: 5.008e-05
[cse_after_recomputation]: 2.121e-05, [1]
    [Cycle 1]: 1.577e-05, [1]
        [cse]: 9.29e-06
[auto_monad_reorder]: 1.508e-05
[get_jit_bprop_graph]: 2.03002e-06
[rewriter_after_jit_bprop_graph]: 6.08002e-06
[opt_after_jit_grad]: 0.00049469
[symbol_engine_optimizer]: 8.363e-05, [1]
    [Cycle 1]: 7.51e-05, [6]
        [build]: 3.52997e-06
        [elim_shapecalc]: 9.52001e-06
        [elim_not_effective]: 1.495e-05
        [opt_reshape]: 5.95002e-06
        [fold_const_symbol]: 1.002e-05
        [renormalize]: 5.40022e-07
[validate]: 3.91e-05
[backend_pass]: 1.13001e-06
[task_emit]: 0.00632292
[execute]: 7.28e-06
Sums
    bootstrap                            :     0.000269s :  1.85%
    type_inference                       :     0.004158s : 28.52%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000061s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.23%
    rewriter_before_opt_a                :     0.000071s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.23%
    jit_opt_a.a_1                        :     0.000612s :  4.20%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000695s :  4.77%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000039s :  0.27%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000630s :  4.32%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000026s :  0.18%
    loop_unroll                          :     0.000436s :  2.99%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000050s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000495s :  3.39%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006323s : 43.37%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000188    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 1.39% :     0.000003s :      2: substitution.fold_const_symbol
 3.22% :     0.000006s :      3: substitution.graph_param_transform
81.34% :     0.000153s :      4: substitution.inline
 1.66% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.69% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.98% :     0.000004s :      2: substitution.replace_old_param
 6.65% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004098     2
79.97% :     0.003277s :      1: type_inference.infer
20.03% :     0.000821s :      1: type_inference.specialize
------[replace.]   0.000066     6
67.54% :     0.000044s :      4: replace.inline
32.46% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     6
92.99% :     0.000150s :      4: match.inline
 7.01% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.30% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.53% :     0.000002s :     10: predicate.addn_zero_filter
 1.83% :     0.000002s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.69% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.24% :     0.000001s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.14% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.82% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.43% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000001s :      6: predicate.less_batch_normalization
 1.75% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.90% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.69% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.35% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.87% :     0.000002s :     18: predicate.replace_applicator
 0.95% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.07% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.58% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.77% :     0.000009s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.46% :     0.000002s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.39% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000564    12
46.89% :     0.000265s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.82% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.29% :     0.000272s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020013    76
 0.27% :     0.000053s :      1: add_recomputation
 0.32% :     0.000064s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.41% :     0.000281s :      1: bootstrap
 0.15% :     0.000029s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.75% :     0.005154s :      1: jit_opt_a
 0.75% :     0.000150s :      1: jit_opt_after_cconv
 0.28% :     0.000056s :      1: jit_opt_b
 2.22% :     0.000444s :      1: loop_unroll
 3.19% :     0.000639s :      1: mutable_eliminate
 4.09% :     0.000819s :     26: opt.transform.jit_opt_a
 0.24% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000037s :      4: opt.transform.symbol_engine_opt
 2.52% :     0.000505s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000036s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.75% :     0.000349s :      1: renormalize.infer
 1.69% :     0.000338s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000038s :      1: rewriter_after_opt_a
 0.37% :     0.000074s :      1: rewriter_before_opt_a
 0.43% :     0.000086s :      1: symbol_engine_optimizer
31.67% :     0.006338s :      1: task_emit
20.84% :     0.004171s :      1: type_inference
 0.30% :     0.000060s :      1: validate


TotalTime = 0.0117437, [30]
[bootstrap]: 0.00033149
[type_inference]: 0.00428312
[event_method]: 1.784e-05
[auto_monad]: 6.453e-05
[graph_reusing]: 6.24999e-06
[pre_auto_parallel]: 2.53e-06
[py_interpret_to_execute]: 3.54e-05
[rewriter_before_opt_a]: 7.4e-05
[expand_dump_flag]: 3.19001e-06
[jit_opt_a]: 0.00488843, [2]
    [Cycle 1]: 0.00141659, [27]
        [switch_simplify]: 5.021e-05
        [loop_unroll]: 2.933e-05
        [a_1]: 0.00050938
        [with_stream_mark]: 1.642e-05
        [recompute_prepare]: 7.06001e-06
        [updatestate_depend_eliminate]: 3.35e-06
        [updatestate_assign_eliminate]: 3.53999e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 1.82999e-06
        [specialize_transform]: 6.19001e-06
        [updatestate_useless_node_eliminater]: 5.37999e-06
        [accelerated_algorithm]: 5.50001e-06
        [meta_shard_fg_expand]: 2.28002e-06
        [get_grad_eliminate_]: 5.66e-06
        [merge_forward]: 3.79002e-06
        [cell_reuse_recompute_pass]: 1.21997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.448e-05
        [j_node_and_user_rematch]: 8.97e-06
        [meta_fg_expand]: 2.17999e-06
        [replace_old_param]: 8.35001e-06
        [inline_without_move]: 5.82001e-06
        [renormalize]: 0.00049523
        [add_forward_monad_depend]: 5.22e-06
        [auto_monad_grad]: 2.22999e-06
        [auto_monad_eliminator]: 1.34e-05
        [cse]: 2.771e-05
        [replace_applicator]: 1.167e-05
    [Cycle 2]: 0.00032536, [27]
        [switch_simplify]: 5.45001e-06
        [loop_unroll]: 4.95999e-06
        [a_1]: 9.131e-05
        [with_stream_mark]: 8.50999e-06
        [recompute_prepare]: 4.95999e-06
        [updatestate_depend_eliminate]: 2.56e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 2.18002e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 5.35999e-06
        [updatestate_useless_node_eliminater]: 4.89998e-06
        [accelerated_algorithm]: 5.42999e-06
        [meta_shard_fg_expand]: 1.28002e-06
        [get_grad_eliminate_]: 4.67e-06
        [merge_forward]: 2.38002e-06
        [cell_reuse_recompute_pass]: 1.30001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.212e-05
        [j_node_and_user_rematch]: 7.7e-06
        [meta_fg_expand]: 1.61002e-06
        [replace_old_param]: 7.26999e-06
        [inline_without_move]: 5.09e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.42e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.19e-06
        [cse]: 1.051e-05
        [replace_applicator]: 5.22999e-06
[py_interpret_to_execute_after_opt_a]: 8.15e-06
[rewriter_after_opt_a]: 3.163e-05
[convert_after_rewriter]: 7.03e-06
[order_py_execute_after_rewriter]: 4.99e-06
[mutable_eliminate]: 0.00052064
[jit_opt_b]: 4.897e-05, [1]
    [Cycle 1]: 4.251e-05, [2]
        [frontend_op_eliminate]: 1.618e-05
        [inline_after_opt_a]: 1.473e-05
[cconv]: 2.373e-05
[loop_unroll]: 0.00041979
[jit_opt_after_cconv]: 0.00014478, [1]
    [Cycle 1]: 0.00013843, [11]
        [c_1]: 2.061e-05
        [parameter_eliminate]: 2.56e-06
        [updatestate_depend_eliminate]: 5.14e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [cse]: 2.25e-05
        [call_graph_tuple_transform]: 1.901e-05
        [tuple_list_get_item_eliminator]: 5.52999e-06
        [none_parameter_eliminate]: 1.60001e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 5.90002e-06
[remove_dup_value]: 1.428e-05
[partial_unused_args_eliminate]: 2.51998e-06
[environ_conv]: 5.15001e-06
[add_recomputation]: 4.791e-05
[cse_after_recomputation]: 2.059e-05, [1]
    [Cycle 1]: 1.51e-05, [1]
        [cse]: 9.24e-06
[auto_monad_reorder]: 1.458e-05
[get_jit_bprop_graph]: 1.65001e-06
[rewriter_after_jit_bprop_graph]: 4.89e-06
[opt_after_jit_grad]: 0.00044464
[symbol_engine_optimizer]: 7.281e-05, [1]
    [Cycle 1]: 6.64e-05, [6]
        [build]: 3.45998e-06
        [elim_shapecalc]: 7.39002e-06
        [elim_not_effective]: 1.241e-05
        [opt_reshape]: 6.33998e-06
        [fold_const_symbol]: 8.48999e-06
        [renormalize]: 4.10015e-07
[validate]: 3.031e-05
Sums
    bootstrap                            :     0.000331s :  4.13%
    type_inference                       :     0.004283s : 53.40%
    event_method                         :     0.000018s :  0.22%
    auto_monad                           :     0.000065s :  0.80%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.44%
    rewriter_before_opt_a                :     0.000074s :  0.92%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000056s :  0.69%
    jit_opt_a.loop_unroll                :     0.000034s :  0.43%
    jit_opt_a.a_1                        :     0.000601s :  7.49%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.14%
    jit_opt_a.renormalize                :     0.000495s :  6.18%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.23%
    jit_opt_a.cse                        :     0.000038s :  0.48%
    jit_opt_a.replace_applicator         :     0.000017s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.10%
    rewriter_after_opt_a                 :     0.000032s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000521s :  6.49%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000024s :  0.30%
    loop_unroll                          :     0.000420s :  5.23%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000022s :  0.28%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000014s :  0.18%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000048s :  0.60%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000015s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000445s :  5.54%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000030s :  0.38%


Time group info:
------[substitution.]   0.000178    23
 1.06% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.68% :     0.000005s :      3: substitution.graph_param_transform
82.70% :     0.000147s :      4: substitution.inline
 1.69% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.42% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.10% :     0.000004s :      2: substitution.replace_old_param
 6.63% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004216     2
81.20% :     0.003423s :      1: type_inference.infer
18.80% :     0.000793s :      1: type_inference.specialize
------[replace.]   0.000065     6
67.46% :     0.000044s :      4: replace.inline
32.54% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000156     6
93.11% :     0.000145s :      4: match.inline
 6.89% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.59% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.19% :     0.000001s :     10: predicate.addn_zero_filter
 2.31% :     0.000003s :     10: predicate.arithmetic_simplify
 1.13% :     0.000001s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.22% :     0.000001s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.53% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.29% :     0.000002s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.26% :     0.000006s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.46% :     0.000002s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.62% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.19% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.77% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000003s :     15: predicate.partial_eliminate
 1.47% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.89% :     0.000002s :     10: predicate.reduce_eliminate
 1.62% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.65% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.46% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.59% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.39% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.28% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.33% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.92% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000548    12
53.19% :     0.000291s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.69% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.12% :     0.000231s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013107    72
 0.39% :     0.000051s :      1: add_recomputation
 0.52% :     0.000068s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.69% :     0.000352s :      1: bootstrap
 0.20% :     0.000027s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.07% :     0.000009s :      1: graph_reusing
37.32% :     0.004891s :      1: jit_opt_a
 1.13% :     0.000147s :      1: jit_opt_after_cconv
 0.40% :     0.000052s :      1: jit_opt_b
 3.26% :     0.000428s :      1: loop_unroll
 4.04% :     0.000529s :      1: mutable_eliminate
 6.14% :     0.000804s :     26: opt.transform.jit_opt_a
 0.36% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.46% :     0.000453s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.29% :     0.000038s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000017s :      1: remove_dup_value
 1.82% :     0.000239s :      1: renormalize.infer
 1.90% :     0.000250s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000037s :      1: rewriter_after_opt_a
 0.59% :     0.000077s :      1: rewriter_before_opt_a
 0.58% :     0.000076s :      1: symbol_engine_optimizer
32.76% :     0.004295s :      1: type_inference


TotalTime = 0.0197128, [33]
[bootstrap]: 0.00038744
[type_inference]: 0.00504304
[event_method]: 1.679e-05
[auto_monad]: 6.323e-05
[graph_reusing]: 5.60001e-06
[pre_auto_parallel]: 2.27999e-06
[py_interpret_to_execute]: 3.521e-05
[rewriter_before_opt_a]: 7.323e-05
[expand_dump_flag]: 3.24001e-06
[jit_opt_a]: 0.00554549, [2]
    [Cycle 1]: 0.00177131, [27]
        [switch_simplify]: 5.758e-05
        [loop_unroll]: 2.987e-05
        [a_1]: 0.00054826
        [with_stream_mark]: 1.753e-05
        [recompute_prepare]: 8.80999e-06
        [updatestate_depend_eliminate]: 3.65e-06
        [updatestate_assign_eliminate]: 2.88e-06
        [updatestate_loads_eliminate]: 2.81e-06
        [parameter_eliminate]: 2.26e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 5.84e-06
        [meta_shard_fg_expand]: 1.96e-06
        [get_grad_eliminate_]: 5.86e-06
        [merge_forward]: 3.88001e-06
        [cell_reuse_recompute_pass]: 1.53002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.621e-05
        [j_node_and_user_rematch]: 8.79e-06
        [meta_fg_expand]: 2.98e-06
        [replace_old_param]: 9.29998e-06
        [inline_without_move]: 5.51998e-06
        [renormalize]: 0.00078437
        [add_forward_monad_depend]: 5.74e-06
        [auto_monad_grad]: 2.34001e-06
        [auto_monad_eliminator]: 1.379e-05
        [cse]: 2.836e-05
        [replace_applicator]: 1.339e-05
    [Cycle 2]: 0.00032778, [27]
        [switch_simplify]: 5.49e-06
        [loop_unroll]: 4.84e-06
        [a_1]: 9.302e-05
        [with_stream_mark]: 9.74e-06
        [recompute_prepare]: 5.01002e-06
        [updatestate_depend_eliminate]: 2.79999e-06
        [updatestate_assign_eliminate]: 2.23002e-06
        [updatestate_loads_eliminate]: 1.72001e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 4.92e-06
        [updatestate_useless_node_eliminater]: 4.65001e-06
        [accelerated_algorithm]: 4.99e-06
        [meta_shard_fg_expand]: 1.20001e-06
        [get_grad_eliminate_]: 4.55001e-06
        [merge_forward]: 2.47001e-06
        [cell_reuse_recompute_pass]: 1.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.377e-05
        [j_node_and_user_rematch]: 7.67002e-06
        [meta_fg_expand]: 1.64e-06
        [replace_old_param]: 1.439e-05
        [inline_without_move]: 4.64002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.67001e-06
        [auto_monad_grad]: 8.60018e-07
        [auto_monad_eliminator]: 5.61998e-06
        [cse]: 1.042e-05
        [replace_applicator]: 4.97e-06
[py_interpret_to_execute_after_opt_a]: 1.125e-05
[rewriter_after_opt_a]: 3.455e-05
[convert_after_rewriter]: 7.08e-06
[order_py_execute_after_rewriter]: 5.47001e-06
[mutable_eliminate]: 0.00052867
[jit_opt_b]: 5.184e-05, [1]
    [Cycle 1]: 4.39e-05, [2]
        [frontend_op_eliminate]: 1.679e-05
        [inline_after_opt_a]: 1.515e-05
[cconv]: 2.591e-05
[loop_unroll]: 0.00045502
[jit_opt_after_cconv]: 0.00014449, [1]
    [Cycle 1]: 0.00013764, [11]
        [c_1]: 2.044e-05
        [parameter_eliminate]: 3.63e-06
        [updatestate_depend_eliminate]: 6.65998e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [cse]: 2.155e-05
        [call_graph_tuple_transform]: 1.96e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.67999e-06
        [renormalize]: 5.60016e-07
        [switch_simplify]: 5.22e-06
[remove_dup_value]: 1.395e-05
[partial_unused_args_eliminate]: 2.12001e-06
[environ_conv]: 5.62001e-06
[add_recomputation]: 4.85e-05
[cse_after_recomputation]: 2.001e-05, [1]
    [Cycle 1]: 1.477e-05, [1]
        [cse]: 8.77999e-06
[auto_monad_reorder]: 1.509e-05
[get_jit_bprop_graph]: 2.03997e-06
[rewriter_after_jit_bprop_graph]: 5.97001e-06
[opt_after_jit_grad]: 0.00046553
[symbol_engine_optimizer]: 7.569e-05, [1]
    [Cycle 1]: 6.768e-05, [6]
        [build]: 3.58e-06
        [elim_shapecalc]: 7.3e-06
        [elim_not_effective]: 1.35e-05
        [opt_reshape]: 5.96e-06
        [fold_const_symbol]: 8.66002e-06
        [renormalize]: 4.00003e-07
[validate]: 3.495e-05
[backend_pass]: 9.5999e-07
[task_emit]: 0.00631204
[execute]: 7.95e-06
Sums
    bootstrap                            :     0.000387s :  2.48%
    type_inference                       :     0.005043s : 32.34%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000063s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000035s :  0.23%
    rewriter_before_opt_a                :     0.000073s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.22%
    jit_opt_a.a_1                        :     0.000641s :  4.11%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.03%
    jit_opt_a.replace_old_param          :     0.000024s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000784s :  5.03%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000039s :  0.25%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000529s :  3.39%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000026s :  0.17%
    loop_unroll                          :     0.000455s :  2.92%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000022s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000048s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000466s :  2.99%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006312s : 40.48%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000211    23
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.56% :     0.000005s :      3: substitution.graph_param_transform
80.39% :     0.000170s :      4: substitution.inline
 1.43% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.18% :     0.000005s :      4: substitution.remove_not_recompute_node
 5.52% :     0.000012s :      2: substitution.replace_old_param
 6.41% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004973     2
81.30% :     0.004043s :      1: type_inference.infer
18.70% :     0.000930s :      1: type_inference.specialize
------[replace.]   0.000074     6
64.00% :     0.000048s :      4: replace.inline
36.00% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000179     6
93.06% :     0.000167s :      4: match.inline
 6.94% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.31% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.97% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 2.73% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.16% :     0.000001s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.71% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.33% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 4.87% :     0.000006s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.39% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.95% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.37% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 2.01% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.93% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.77% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.03% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.69% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.74% :     0.000002s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.22% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.55% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.73% :     0.000003s :     16: predicate.switch_defer_inline
 2.38% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000009s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.89% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.11% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000681    12
50.37% :     0.000343s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.49% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.14% :     0.000308s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021456    76
 0.24% :     0.000051s :      1: add_recomputation
 0.31% :     0.000066s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 2.12% :     0.000455s :      1: bootstrap
 0.13% :     0.000029s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.86% :     0.005549s :      1: jit_opt_a
 0.69% :     0.000147s :      1: jit_opt_after_cconv
 0.25% :     0.000054s :      1: jit_opt_b
 2.16% :     0.000463s :      1: loop_unroll
 2.51% :     0.000538s :      1: mutable_eliminate
 4.02% :     0.000863s :     26: opt.transform.jit_opt_a
 0.22% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000473s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000038s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.95% :     0.000417s :      1: renormalize.infer
 1.67% :     0.000359s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000037s :      1: rewriter_after_opt_a
 0.35% :     0.000076s :      1: rewriter_before_opt_a
 0.37% :     0.000079s :      1: symbol_engine_optimizer
29.50% :     0.006330s :      1: task_emit
23.57% :     0.005057s :      1: type_inference
 0.25% :     0.000055s :      1: validate


TotalTime = 0.0183163, [33]
[bootstrap]: 0.00030062
[type_inference]: 0.00447861
[event_method]: 1.712e-05
[auto_monad]: 5.592e-05
[graph_reusing]: 4.47998e-06
[pre_auto_parallel]: 2.25002e-06
[py_interpret_to_execute]: 3.779e-05
[rewriter_before_opt_a]: 6.825e-05
[expand_dump_flag]: 2.43e-06
[jit_opt_a]: 0.00526305, [2]
    [Cycle 1]: 0.00160932, [27]
        [switch_simplify]: 4.824e-05
        [loop_unroll]: 2.927e-05
        [a_1]: 0.00050489
        [with_stream_mark]: 1.264e-05
        [recompute_prepare]: 7.21999e-06
        [updatestate_depend_eliminate]: 3.08e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 5.98998e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 1.54998e-06
        [get_grad_eliminate_]: 5.26998e-06
        [merge_forward]: 2.90998e-06
        [cell_reuse_recompute_pass]: 8.00006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.153e-05
        [j_node_and_user_rematch]: 9.47001e-06
        [meta_fg_expand]: 1.76e-06
        [replace_old_param]: 8.66002e-06
        [inline_without_move]: 5.64e-06
        [renormalize]: 0.00072562
        [add_forward_monad_depend]: 5.69e-06
        [auto_monad_grad]: 1.98997e-06
        [auto_monad_eliminator]: 1.05e-05
        [cse]: 1.889e-05
        [replace_applicator]: 1.22e-05
    [Cycle 2]: 0.00032474, [27]
        [switch_simplify]: 5.18002e-06
        [loop_unroll]: 4.77998e-06
        [a_1]: 9.203e-05
        [with_stream_mark]: 8.87999e-06
        [recompute_prepare]: 5.02e-06
        [updatestate_depend_eliminate]: 2.37001e-06
        [updatestate_assign_eliminate]: 1.98002e-06
        [updatestate_loads_eliminate]: 1.81998e-06
        [parameter_eliminate]: 1.14998e-06
        [specialize_transform]: 4.80001e-06
        [updatestate_useless_node_eliminater]: 4.81002e-06
        [accelerated_algorithm]: 4.88001e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.30999e-06
        [merge_forward]: 2.74001e-06
        [cell_reuse_recompute_pass]: 1.87001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.268e-05
        [j_node_and_user_rematch]: 7.61999e-06
        [meta_fg_expand]: 1.70001e-06
        [replace_old_param]: 7.32002e-06
        [inline_without_move]: 4.49002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.66e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 4.88001e-06
        [cse]: 1.037e-05
        [replace_applicator]: 5.02999e-06
[py_interpret_to_execute_after_opt_a]: 1.013e-05
[rewriter_after_opt_a]: 3.299e-05
[convert_after_rewriter]: 6.16e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00061644
[jit_opt_b]: 5.266e-05, [1]
    [Cycle 1]: 4.488e-05, [2]
        [frontend_op_eliminate]: 1.659e-05
        [inline_after_opt_a]: 1.6e-05
[cconv]: 2.071e-05
[loop_unroll]: 0.00046384
[jit_opt_after_cconv]: 0.00014802, [1]
    [Cycle 1]: 0.00014017, [11]
        [c_1]: 2.177e-05
        [parameter_eliminate]: 3.53e-06
        [updatestate_depend_eliminate]: 6.06e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [cse]: 2.076e-05
        [call_graph_tuple_transform]: 1.946e-05
        [tuple_list_get_item_eliminator]: 5.80002e-06
        [none_parameter_eliminate]: 1.14e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 5.64998e-06
[remove_dup_value]: 9.59999e-06
[partial_unused_args_eliminate]: 2.62001e-06
[environ_conv]: 4.53001e-06
[add_recomputation]: 4.304e-05
[cse_after_recomputation]: 2.027e-05, [1]
    [Cycle 1]: 1.482e-05, [1]
        [cse]: 8.80001e-06
[auto_monad_reorder]: 1.231e-05
[get_jit_bprop_graph]: 2.09999e-06
[rewriter_after_jit_bprop_graph]: 5.45001e-06
[opt_after_jit_grad]: 0.00049809
[symbol_engine_optimizer]: 7.667e-05, [1]
    [Cycle 1]: 6.871e-05, [6]
        [build]: 3.14001e-06
        [elim_shapecalc]: 7.61999e-06
        [elim_not_effective]: 1.433e-05
        [opt_reshape]: 5.62001e-06
        [fold_const_symbol]: 8.22998e-06
        [renormalize]: 6.79982e-07
[validate]: 3.497e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00580812
[execute]: 5.17999e-06
Sums
    bootstrap                            :     0.000301s :  2.09%
    type_inference                       :     0.004479s : 31.16%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000056s :  0.39%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000038s :  0.26%
    rewriter_before_opt_a                :     0.000068s :  0.47%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.37%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000597s :  4.15%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000726s :  5.05%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000029s :  0.20%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.23%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000616s :  4.29%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000021s :  0.14%
    loop_unroll                          :     0.000464s :  3.23%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000010s :  0.07%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000043s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000012s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000498s :  3.47%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005808s : 40.41%
    execute                              :     0.000005s :  0.04%


Time group info:
------[substitution.]   0.000181    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000001s :      2: substitution.fold_const_symbol
 2.95% :     0.000005s :      3: substitution.graph_param_transform
82.78% :     0.000149s :      4: substitution.inline
 1.71% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.35% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.79% :     0.000003s :      2: substitution.replace_old_param
 6.64% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004414     2
80.30% :     0.003544s :      1: type_inference.infer
19.70% :     0.000870s :      1: type_inference.specialize
------[replace.]   0.000067     6
64.96% :     0.000043s :      4: replace.inline
35.04% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000158     6
93.13% :     0.000147s :      4: match.inline
 6.87% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.29% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.58% :     0.000002s :     10: predicate.addn_zero_filter
 1.88% :     0.000002s :     10: predicate.arithmetic_simplify
 1.47% :     0.000002s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.49% :     0.000001s :      3: predicate.elim_not_effective
 0.71% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.22% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.40% :     0.000000s :      3: predicate.graph_param_transform
 4.86% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.37% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.20% :     0.000003s :     15: predicate.load_eliminater
 1.85% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.00% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.17% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000001s :     10: predicate.merge_addn
 1.54% :     0.000002s :     10: predicate.minmaximum_grad
 1.92% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000002s :     15: predicate.partial_eliminate
 1.55% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.65% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.53% :     0.000001s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.29% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.77% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.46% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.26% :     0.000001s :     10: predicate.transpose_eliminate
 1.28% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.45% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.10% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000591    12
47.23% :     0.000279s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.93% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.84% :     0.000283s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019948    76
 0.23% :     0.000046s :      1: add_recomputation
 0.30% :     0.000059s :      1: auto_monad
 0.07% :     0.000015s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.59% :     0.000317s :      1: bootstrap
 0.12% :     0.000024s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
26.40% :     0.005266s :      1: jit_opt_a
 0.76% :     0.000151s :      1: jit_opt_after_cconv
 0.28% :     0.000055s :      1: jit_opt_b
 2.37% :     0.000473s :      1: loop_unroll
 3.14% :     0.000627s :      1: mutable_eliminate
 4.00% :     0.000797s :     26: opt.transform.jit_opt_a
 0.25% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.54% :     0.000507s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000041s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000012s :      1: remove_dup_value
 1.88% :     0.000375s :      1: renormalize.infer
 1.72% :     0.000343s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000036s :      1: rewriter_after_opt_a
 0.36% :     0.000071s :      1: rewriter_before_opt_a
 0.40% :     0.000079s :      1: symbol_engine_optimizer
29.18% :     0.005821s :      1: task_emit
22.52% :     0.004493s :      1: type_inference
 0.27% :     0.000053s :      1: validate


TotalTime = 0.0115595, [30]
[bootstrap]: 0.00027603
[type_inference]: 0.00420109
[event_method]: 1.639e-05
[auto_monad]: 6.399e-05
[graph_reusing]: 5.34e-06
[pre_auto_parallel]: 2.61999e-06
[py_interpret_to_execute]: 2.989e-05
[rewriter_before_opt_a]: 7.245e-05
[expand_dump_flag]: 3.18e-06
[jit_opt_a]: 0.00486509, [2]
    [Cycle 1]: 0.00142071, [27]
        [switch_simplify]: 5.074e-05
        [loop_unroll]: 2.965e-05
        [a_1]: 0.00051482
        [with_stream_mark]: 1.585e-05
        [recompute_prepare]: 7.15e-06
        [updatestate_depend_eliminate]: 3.56999e-06
        [updatestate_assign_eliminate]: 3.13e-06
        [updatestate_loads_eliminate]: 2.89001e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 5.82001e-06
        [updatestate_useless_node_eliminater]: 5.63002e-06
        [accelerated_algorithm]: 5.64998e-06
        [meta_shard_fg_expand]: 2.01998e-06
        [get_grad_eliminate_]: 5.76e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 1.02998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.432e-05
        [j_node_and_user_rematch]: 8.88002e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 9.20999e-06
        [inline_without_move]: 5.30999e-06
        [renormalize]: 0.00049186
        [add_forward_monad_depend]: 4.51002e-06
        [auto_monad_grad]: 2.43e-06
        [auto_monad_eliminator]: 1.31e-05
        [cse]: 2.701e-05
        [replace_applicator]: 1.037e-05
    [Cycle 2]: 0.00031962, [27]
        [switch_simplify]: 5.34e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 9.022e-05
        [with_stream_mark]: 8.03001e-06
        [recompute_prepare]: 5.11997e-06
        [updatestate_depend_eliminate]: 2.59001e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 1.98997e-06
        [parameter_eliminate]: 9.60019e-07
        [specialize_transform]: 4.90001e-06
        [updatestate_useless_node_eliminater]: 5.10001e-06
        [accelerated_algorithm]: 5.29998e-06
        [meta_shard_fg_expand]: 1.16002e-06
        [get_grad_eliminate_]: 4.61002e-06
        [merge_forward]: 2.32999e-06
        [cell_reuse_recompute_pass]: 1.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.175e-05
        [j_node_and_user_rematch]: 7.97e-06
        [meta_fg_expand]: 1.56002e-06
        [replace_old_param]: 6.66e-06
        [inline_without_move]: 4.63001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 7.79983e-07
        [auto_monad_eliminator]: 4.80001e-06
        [cse]: 1.04e-05
        [replace_applicator]: 5.62001e-06
[py_interpret_to_execute_after_opt_a]: 8.07e-06
[rewriter_after_opt_a]: 3.062e-05
[convert_after_rewriter]: 7.26001e-06
[order_py_execute_after_rewriter]: 5.57001e-06
[mutable_eliminate]: 0.00048485
[jit_opt_b]: 4.916e-05, [1]
    [Cycle 1]: 4.246e-05, [2]
        [frontend_op_eliminate]: 1.596e-05
        [inline_after_opt_a]: 1.472e-05
[cconv]: 2.412e-05
[loop_unroll]: 0.00044648
[jit_opt_after_cconv]: 0.0001371, [1]
    [Cycle 1]: 0.0001305, [11]
        [c_1]: 2.038e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 5.37001e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 1.68e-05
        [call_graph_tuple_transform]: 1.804e-05
        [tuple_list_get_item_eliminator]: 5.56e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 3.19997e-07
        [switch_simplify]: 5.33002e-06
[remove_dup_value]: 1.403e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 5.22999e-06
[add_recomputation]: 4.745e-05
[cse_after_recomputation]: 2.035e-05, [1]
    [Cycle 1]: 1.499e-05, [1]
        [cse]: 9.39e-06
[auto_monad_reorder]: 1.504e-05
[get_jit_bprop_graph]: 1.41998e-06
[rewriter_after_jit_bprop_graph]: 4.90001e-06
[opt_after_jit_grad]: 0.00045184
[symbol_engine_optimizer]: 7.22e-05, [1]
    [Cycle 1]: 6.591e-05, [6]
        [build]: 3.25998e-06
        [elim_shapecalc]: 7.61999e-06
        [elim_not_effective]: 1.214e-05
        [opt_reshape]: 5.58002e-06
        [fold_const_symbol]: 8.90999e-06
        [renormalize]: 3.69997e-07
[validate]: 3.154e-05
Sums
    bootstrap                            :     0.000276s :  3.51%
    type_inference                       :     0.004201s : 53.45%
    event_method                         :     0.000016s :  0.21%
    auto_monad                           :     0.000064s :  0.81%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000030s :  0.38%
    rewriter_before_opt_a                :     0.000072s :  0.92%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000056s :  0.71%
    jit_opt_a.loop_unroll                :     0.000035s :  0.44%
    jit_opt_a.a_1                        :     0.000605s :  7.70%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000492s :  6.26%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.23%
    jit_opt_a.cse                        :     0.000037s :  0.48%
    jit_opt_a.replace_applicator         :     0.000016s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.10%
    rewriter_after_opt_a                 :     0.000031s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000006s :  0.07%
    mutable_eliminate                    :     0.000485s :  6.17%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000024s :  0.31%
    loop_unroll                          :     0.000446s :  5.68%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000047s :  0.60%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000015s :  0.19%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000452s :  5.75%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.40%


Time group info:
------[substitution.]   0.000185    23
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.89% :     0.000002s :      2: substitution.fold_const_symbol
 2.76% :     0.000005s :      3: substitution.graph_param_transform
83.02% :     0.000154s :      4: substitution.inline
 1.58% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.29% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000004s :      2: substitution.replace_old_param
 6.42% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004138     2
81.28% :     0.003364s :      1: type_inference.infer
18.72% :     0.000774s :      1: type_inference.specialize
------[replace.]   0.000065     6
67.21% :     0.000043s :      4: replace.inline
32.79% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     6
93.42% :     0.000151s :      4: match.inline
 6.58% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.78% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.41% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 2.04% :     0.000002s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.42% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.05% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.13% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.46% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.08% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.08% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.30% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :      6: predicate.less_batch_normalization
 1.46% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.67% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.86% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.20% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 1.71% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     15: predicate.partial_eliminate
 1.38% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.06% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.61% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.61% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.16% :     0.000009s :     53: predicate.switch_simplify
 1.43% :     0.000002s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.71% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 2.02% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000592    12
56.43% :     0.000334s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.55% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.02% :     0.000231s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012919    72
 0.39% :     0.000050s :      1: add_recomputation
 0.52% :     0.000067s :      1: auto_monad
 0.14% :     0.000018s :      1: auto_monad_reorder
 2.28% :     0.000295s :      1: bootstrap
 0.21% :     0.000027s :      1: cconv
 0.08% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
37.68% :     0.004868s :      1: jit_opt_a
 1.08% :     0.000140s :      1: jit_opt_after_cconv
 0.40% :     0.000052s :      1: jit_opt_b
 3.52% :     0.000454s :      1: loop_unroll
 3.82% :     0.000493s :      1: mutable_eliminate
 6.26% :     0.000808s :     26: opt.transform.jit_opt_a
 0.36% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.56% :     0.000460s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.04% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000032s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000017s :      1: remove_dup_value
 1.89% :     0.000244s :      1: renormalize.infer
 1.87% :     0.000241s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000033s :      1: rewriter_after_opt_a
 0.58% :     0.000075s :      1: rewriter_before_opt_a
 0.58% :     0.000075s :      1: symbol_engine_optimizer
32.61% :     0.004213s :      1: type_inference


TotalTime = 0.113837, [33]
[bootstrap]: 0.00022678
[type_inference]: 0.084131
[event_method]: 0.00034116
[auto_monad]: 0.00010438
[graph_reusing]: 6.74999e-06
[pre_auto_parallel]: 3.18e-06
[py_interpret_to_execute]: 2.912e-05
[rewriter_before_opt_a]: 0.00010467
[expand_dump_flag]: 3.16999e-06
[jit_opt_a]: 0.0265444, [3]
    [Cycle 1]: 0.0179843, [27]
        [switch_simplify]: 7.591e-05
        [loop_unroll]: 4.324e-05
        [a_1]: 0.00098636
        [with_stream_mark]: 2.98e-05
        [recompute_prepare]: 2.414e-05
        [updatestate_depend_eliminate]: 9.25999e-06
        [updatestate_assign_eliminate]: 6.44999e-06
        [updatestate_loads_eliminate]: 6.20002e-06
        [parameter_eliminate]: 3.06999e-06
        [specialize_transform]: 1.521e-05
        [updatestate_useless_node_eliminater]: 1.385e-05
        [accelerated_algorithm]: 1.398e-05
        [meta_shard_fg_expand]: 3.43e-06
        [get_grad_eliminate_]: 1.412e-05
        [merge_forward]: 8.54e-06
        [cell_reuse_recompute_pass]: 1.44998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.98e-05
        [j_node_and_user_rematch]: 2.505e-05
        [meta_fg_expand]: 0.00191368
        [replace_old_param]: 6.629e-05
        [inline_without_move]: 6.059e-05
        [renormalize]: 0.013975
        [add_forward_monad_depend]: 1.318e-05
        [auto_monad_grad]: 6.59999e-06
        [auto_monad_eliminator]: 6.128e-05
        [cse]: 0.00025226
        [replace_applicator]: 8.648e-05
    [Cycle 2]: 0.00436307, [27]
        [switch_simplify]: 4.38e-05
        [loop_unroll]: 4.347e-05
        [a_1]: 0.00125653
        [with_stream_mark]: 1.467e-05
        [recompute_prepare]: 1.256e-05
        [updatestate_depend_eliminate]: 5.27001e-06
        [updatestate_assign_eliminate]: 4.52e-06
        [updatestate_loads_eliminate]: 3.48999e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 1.012e-05
        [updatestate_useless_node_eliminater]: 9.24e-06
        [accelerated_algorithm]: 1.05e-05
        [meta_shard_fg_expand]: 2.57001e-06
        [get_grad_eliminate_]: 8.96998e-06
        [merge_forward]: 4.22e-06
        [cell_reuse_recompute_pass]: 8.70001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.628e-05
        [j_node_and_user_rematch]: 1.377e-05
        [meta_fg_expand]: 7.076e-05
        [replace_old_param]: 1.08e-05
        [inline_without_move]: 8.87e-06
        [renormalize]: 0.0025921
        [add_forward_monad_depend]: 7.28e-06
        [auto_monad_grad]: 2.02999e-06
        [auto_monad_eliminator]: 1.177e-05
        [cse]: 1.838e-05
        [replace_applicator]: 1.506e-05
    [Cycle 3]: 0.00026569, [27]
        [switch_simplify]: 4.23999e-06
        [loop_unroll]: 4.50999e-06
        [a_1]: 4.996e-05
        [with_stream_mark]: 1.411e-05
        [recompute_prepare]: 4.38001e-06
        [updatestate_depend_eliminate]: 2.61999e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.49998e-06
        [parameter_eliminate]: 1.35999e-06
        [specialize_transform]: 3.93001e-06
        [updatestate_useless_node_eliminater]: 3.55998e-06
        [accelerated_algorithm]: 4.17e-06
        [meta_shard_fg_expand]: 1.92999e-06
        [get_grad_eliminate_]: 3.87998e-06
        [merge_forward]: 2.53e-06
        [cell_reuse_recompute_pass]: 3.41001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.293e-05
        [j_node_and_user_rematch]: 5.61998e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 3.46001e-06
        [inline_without_move]: 3.20998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 3.93999e-06
        [cse]: 8.42998e-06
        [replace_applicator]: 3.49001e-06
[py_interpret_to_execute_after_opt_a]: 1.246e-05
[rewriter_after_opt_a]: 3.344e-05
[convert_after_rewriter]: 5.96e-06
[order_py_execute_after_rewriter]: 4.92e-06
[mutable_eliminate]: 0.00071158
[jit_opt_b]: 4.289e-05, [1]
    [Cycle 1]: 3.571e-05, [2]
        [frontend_op_eliminate]: 1.082e-05
        [inline_after_opt_a]: 1.401e-05
[cconv]: 2.408e-05
[loop_unroll]: 0.00044997
[jit_opt_after_cconv]: 0.00013013, [1]
    [Cycle 1]: 0.0001228, [11]
        [c_1]: 1.362e-05
        [parameter_eliminate]: 3.85e-06
        [updatestate_depend_eliminate]: 5.05999e-06
        [updatestate_assign_eliminate]: 1.85001e-06
        [updatestate_loads_eliminate]: 1.66002e-06
        [cse]: 2.136e-05
        [call_graph_tuple_transform]: 1.461e-05
        [tuple_list_get_item_eliminator]: 4.04997e-06
        [none_parameter_eliminate]: 1.18001e-06
        [renormalize]: 6.60017e-07
        [switch_simplify]: 3.94002e-06
[remove_dup_value]: 1.237e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 5.20001e-06
[add_recomputation]: 3.634e-05
[cse_after_recomputation]: 2.086e-05, [1]
    [Cycle 1]: 1.474e-05, [1]
        [cse]: 8.46002e-06
[auto_monad_reorder]: 1.162e-05
[get_jit_bprop_graph]: 2.22001e-06
[rewriter_after_jit_bprop_graph]: 5.40001e-06
[opt_after_jit_grad]: 0.00048415
[symbol_engine_optimizer]: 6.536e-05, [1]
    [Cycle 1]: 5.916e-05, [6]
        [build]: 3.95e-06
        [elim_shapecalc]: 5.82001e-06
        [elim_not_effective]: 9.51e-06
        [opt_reshape]: 5.56e-06
        [fold_const_symbol]: 6.42001e-06
        [renormalize]: 3.4002e-07
[validate]: 3.447e-05
[backend_pass]: 9.5999e-07
[task_emit]: 2.278e-05
[execute]: 1.16997e-06
Sums
    bootstrap                            :     0.000227s :  0.21%
    type_inference                       :     0.084131s : 77.15%
    event_method                         :     0.000341s :  0.31%
    auto_monad                           :     0.000104s :  0.10%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000029s :  0.03%
    rewriter_before_opt_a                :     0.000105s :  0.10%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000124s :  0.11%
    jit_opt_a.loop_unroll                :     0.000091s :  0.08%
    jit_opt_a.a_1                        :     0.002293s :  2.10%
    jit_opt_a.with_stream_mark           :     0.000059s :  0.05%
    jit_opt_a.recompute_prepare          :     0.000041s :  0.04%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  0.02%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.01%
    jit_opt_a.specialize_transform       :     0.000029s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000027s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000029s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000027s :  0.02%
    jit_opt_a.merge_forward              :     0.000015s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000059s :  0.05%
    jit_opt_a.j_node_and_user_rematch    :     0.000044s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.001986s :  1.82%
    jit_opt_a.replace_old_param          :     0.000081s :  0.07%
    jit_opt_a.inline_without_move        :     0.000073s :  0.07%
    jit_opt_a.renormalize                :     0.016567s : 15.19%
    jit_opt_a.add_forward_monad_depend   :     0.000022s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000077s :  0.07%
    jit_opt_a.cse                        :     0.000279s :  0.26%
    jit_opt_a.replace_applicator         :     0.000105s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.01%
    rewriter_after_opt_a                 :     0.000033s :  0.03%
    convert_after_rewriter               :     0.000006s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.00%
    mutable_eliminate                    :     0.000712s :  0.65%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.01%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.01%
    cconv                                :     0.000024s :  0.02%
    loop_unroll                          :     0.000450s :  0.41%
    jit_opt_after_cconv.c_1              :     0.000014s :  0.01%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.00%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000021s :  0.02%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.01%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.00%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.00%
    remove_dup_value                     :     0.000012s :  0.01%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000005s :  0.00%
    add_recomputation                    :     0.000036s :  0.03%
    cse_after_recomputation.cse          :     0.000008s :  0.01%
    auto_monad_reorder                   :     0.000012s :  0.01%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.00%
    opt_after_jit_grad                   :     0.000484s :  0.44%
    symbol_engine_optimizer.build        :     0.000004s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.01%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.03%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000023s :  0.02%
    execute                              :     0.000001s :  0.00%


Time group info:
------[substitution.]   0.000566   105
 0.26% :     0.000001s :      1: substitution.elim_not_effective
 0.16% :     0.000001s :      1: substitution.fold_const_symbol
 0.85% :     0.000005s :      1: substitution.graph_param_transform
62.79% :     0.000356s :     12: substitution.inline
 3.72% :     0.000021s :      2: substitution.inline_without_move
 1.43% :     0.000008s :     13: substitution.j_node_and_user_rematch
 2.06% :     0.000012s :      7: substitution.minmaximum_grad
 1.54% :     0.000009s :      5: substitution.partial_eliminate
 1.81% :     0.000010s :     13: substitution.remove_not_recompute_node
 5.89% :     0.000033s :     10: substitution.replace_applicator
 0.92% :     0.000005s :      7: substitution.replace_old_param
 1.65% :     0.000009s :      2: substitution.reshape_eliminate
 0.71% :     0.000004s :      1: substitution.set_cell_output_no_recompute
 2.12% :     0.000012s :      2: substitution.tile_eliminate
 4.01% :     0.000023s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.58% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 7.50% :     0.000042s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.084030     2
96.98% :     0.081494s :      1: type_inference.infer
 3.02% :     0.002536s :      1: type_inference.specialize
------[replace.]   0.000160    19
62.59% :     0.000100s :     12: replace.inline
37.41% :     0.000060s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000370    19
93.89% :     0.000348s :     12: match.inline
 6.11% :     0.000023s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000400  2758
 1.94% :     0.000008s :     48: predicate.accumulaten_eliminater
 0.34% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.40% :     0.000006s :     48: predicate.addn_check_dump
 1.63% :     0.000007s :     48: predicate.addn_zero_filter
 2.55% :     0.000010s :     48: predicate.arithmetic_simplify
 1.68% :     0.000007s :     48: predicate.cast_eliminate
 0.07% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.46% :     0.000006s :     48: predicate.compare_switch_simplify
 2.05% :     0.000008s :     48: predicate.depend_value_elim
 1.33% :     0.000005s :     48: predicate.dict_get_item_const_eliminator
 1.58% :     0.000006s :     48: predicate.dict_get_item_eliminator
 1.49% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.20% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.05% :     0.000000s :      1: predicate.elim_not_effective
 0.10% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.35% :     0.000005s :     48: predicate.environ_get_add_eliminate
 1.43% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.52% :     0.000006s :     48: predicate.environ_get_eliminate
 1.41% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.03% :     0.000000s :      1: predicate.fold_const_symbol
 1.00% :     0.000004s :     21: predicate.get_grad_eliminate
 0.07% :     0.000000s :      1: predicate.graph_param_transform
 3.85% :     0.000015s :     69: predicate.inline
 1.80% :     0.000007s :     45: predicate.inline_without_move
 0.40% :     0.000002s :     21: predicate.j_node_and_user_rematch
 1.17% :     0.000005s :     21: predicate.less_batch_normalization
 2.11% :     0.000008s :     55: predicate.list_to_tuple_eliminator_
 1.97% :     0.000008s :     56: predicate.load_eliminater
 0.39% :     0.000002s :      1: predicate.loop_unroll_after_grad
 3.36% :     0.000013s :     97: predicate.loop_unroll_before_grad
 1.89% :     0.000008s :     49: predicate.make_slice_get_slice_eliminator
 1.36% :     0.000005s :     48: predicate.merge_addn
 1.47% :     0.000006s :     48: predicate.minmaximum_grad
 0.45% :     0.000002s :      1: predicate.mutable_eliminate
 0.13% :     0.000001s :      1: predicate.opt_reshape
 2.25% :     0.000009s :     56: predicate.partial_eliminate
 1.70% :     0.000007s :     48: predicate.print_const_string_wrapper
 1.92% :     0.000008s :     48: predicate.reduce_eliminate
 1.93% :     0.000008s :     55: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.76% :     0.000011s :    104: predicate.replace_applicator
 1.09% :     0.000004s :     45: predicate.replace_old_param
 0.08% :     0.000000s :      1: predicate.reset_defer_inline
 1.65% :     0.000007s :     48: predicate.reshape_eliminate
 1.59% :     0.000006s :     48: predicate.row_tensor_add_zeros_like
 0.18% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.48% :     0.000006s :     48: predicate.same_eliminate
 0.53% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.22% :     0.000001s :      2: predicate.special_op_eliminate
 0.84% :     0.000003s :     21: predicate.specialize_transform
 1.74% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.66% :     0.000007s :     48: predicate.stack_unstack_eliminate
 0.09% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.75% :     0.000011s :     67: predicate.switch_defer_inline
 2.37% :     0.000009s :     67: predicate.switch_layer_defer_inline
 6.19% :     0.000025s :    165: predicate.switch_simplify
 1.66% :     0.000007s :     48: predicate.tile_eliminate
 1.47% :     0.000006s :     48: predicate.transpose_eliminate
 1.89% :     0.000008s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.61% :     0.000006s :     48: predicate.tuple_list_get_item_depend_reorder
 2.98% :     0.000012s :     57: predicate.tuple_list_get_item_eliminator
 1.89% :     0.000008s :     48: predicate.tuple_list_set_item_eliminator
 1.81% :     0.000007s :     55: predicate.tuple_to_list_eliminator_
 1.90% :     0.000008s :     56: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000010s :     77: predicate.updatestate_useless_node_eliminater
 1.87% :     0.000007s :     48: predicate.value_based_eliminate
 0.05% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.20% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.004488    60
83.88% :     0.003764s :     44: func_graph_cloner_run.FuncGraphClonerGraph
16.12% :     0.000724s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.133433    91
 0.03% :     0.000039s :      1: add_recomputation
 0.08% :     0.000111s :      1: auto_monad
 0.01% :     0.000014s :      1: auto_monad_reorder
 0.00% :     0.000004s :      1: backend_pass
 0.18% :     0.000238s :      1: bootstrap
 0.02% :     0.000027s :      1: cconv
 0.01% :     0.000008s :      1: convert_after_rewriter
 0.02% :     0.000023s :      1: cse_after_recomputation
 0.01% :     0.000007s :      1: environ_conv
 0.26% :     0.000349s :      1: event_method
 0.00% :     0.000003s :      1: execute
 0.00% :     0.000005s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000009s :      1: graph_reusing
19.90% :     0.026548s :      1: jit_opt_a
 0.10% :     0.000133s :      1: jit_opt_after_cconv
 0.03% :     0.000045s :      1: jit_opt_b
 0.34% :     0.000458s :      1: loop_unroll
 0.54% :     0.000721s :      1: mutable_eliminate
 2.23% :     0.002972s :     39: opt.transform.jit_opt_a
 0.02% :     0.000033s :      4: opt.transform.jit_opt_after_cconv
 0.01% :     0.000017s :      4: opt.transform.jit_opt_b
 0.01% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.01% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.01% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.02% :     0.000023s :      4: opt.transform.symbol_engine_opt
 0.37% :     0.000493s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.00% :     0.000005s :      1: pre_auto_parallel
 0.02% :     0.000032s :      1: py_interpret_to_execute
 0.04% :     0.000048s :      1: py_interpret_to_execute_after_opt_a
 0.01% :     0.000015s :      1: remove_dup_value
10.82% :     0.014443s :      2: renormalize.infer
 1.58% :     0.002104s :      2: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.03% :     0.000037s :      1: rewriter_after_opt_a
 0.08% :     0.000108s :      1: rewriter_before_opt_a
 0.05% :     0.000068s :      1: symbol_engine_optimizer
 0.02% :     0.000028s :      1: task_emit
63.06% :     0.084148s :      1: type_inference
 0.04% :     0.000052s :      1: validate


TotalTime = 0.0198761, [33]
[bootstrap]: 0.00027898
[type_inference]: 0.00458761
[event_method]: 1.838e-05
[auto_monad]: 6.508e-05
[graph_reusing]: 5.67001e-06
[pre_auto_parallel]: 2.92002e-06
[py_interpret_to_execute]: 4.027e-05
[rewriter_before_opt_a]: 7.31e-05
[expand_dump_flag]: 3.51999e-06
[jit_opt_a]: 0.0057955, [2]
    [Cycle 1]: 0.00187338, [27]
        [switch_simplify]: 6.109e-05
        [loop_unroll]: 3.207e-05
        [a_1]: 0.00055275
        [with_stream_mark]: 1.701e-05
        [recompute_prepare]: 7.28e-06
        [updatestate_depend_eliminate]: 3.58999e-06
        [updatestate_assign_eliminate]: 2.80002e-06
        [updatestate_loads_eliminate]: 2.41998e-06
        [parameter_eliminate]: 1.66998e-06
        [specialize_transform]: 5.74e-06
        [updatestate_useless_node_eliminater]: 4.70999e-06
        [accelerated_algorithm]: 5.66e-06
        [meta_shard_fg_expand]: 1.90001e-06
        [get_grad_eliminate_]: 5.52001e-06
        [merge_forward]: 3.36999e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.548e-05
        [j_node_and_user_rematch]: 9.57999e-06
        [meta_fg_expand]: 2.28998e-06
        [replace_old_param]: 9.09e-06
        [inline_without_move]: 5.41002e-06
        [renormalize]: 0.00089566
        [add_forward_monad_depend]: 5.52001e-06
        [auto_monad_grad]: 1.83002e-06
        [auto_monad_eliminator]: 1.143e-05
        [cse]: 1.973e-05
        [replace_applicator]: 1.252e-05
    [Cycle 2]: 0.00032776, [27]
        [switch_simplify]: 6.22001e-06
        [loop_unroll]: 4.75001e-06
        [a_1]: 9.217e-05
        [with_stream_mark]: 1.073e-05
        [recompute_prepare]: 5.10001e-06
        [updatestate_depend_eliminate]: 3.22002e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [parameter_eliminate]: 1.39e-06
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.02e-06
        [meta_shard_fg_expand]: 1.11997e-06
        [get_grad_eliminate_]: 4.59002e-06
        [merge_forward]: 2.44999e-06
        [cell_reuse_recompute_pass]: 2.67001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.358e-05
        [j_node_and_user_rematch]: 8.08999e-06
        [meta_fg_expand]: 1.60001e-06
        [replace_old_param]: 7.06001e-06
        [inline_without_move]: 4.52003e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.05999e-06
        [auto_monad_grad]: 9.09989e-07
        [auto_monad_eliminator]: 5.32001e-06
        [cse]: 1.037e-05
        [replace_applicator]: 4.70999e-06
[py_interpret_to_execute_after_opt_a]: 1.239e-05
[rewriter_after_opt_a]: 4.047e-05
[convert_after_rewriter]: 7.18998e-06
[order_py_execute_after_rewriter]: 5.27001e-06
[mutable_eliminate]: 0.00065827
[jit_opt_b]: 5.265e-05, [1]
    [Cycle 1]: 4.432e-05, [2]
        [frontend_op_eliminate]: 1.661e-05
        [inline_after_opt_a]: 1.538e-05
[cconv]: 2.656e-05
[loop_unroll]: 0.00044182
[jit_opt_after_cconv]: 0.0001468, [1]
    [Cycle 1]: 0.00013906, [11]
        [c_1]: 2.209e-05
        [parameter_eliminate]: 2.78e-06
        [updatestate_depend_eliminate]: 6.25002e-06
        [updatestate_assign_eliminate]: 2.59001e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.987e-05
        [call_graph_tuple_transform]: 1.921e-05
        [tuple_list_get_item_eliminator]: 5.82999e-06
        [none_parameter_eliminate]: 1.41998e-06
        [renormalize]: 5.29981e-07
        [switch_simplify]: 5.56002e-06
[remove_dup_value]: 1.271e-05
[partial_unused_args_eliminate]: 2.89001e-06
[environ_conv]: 5.72999e-06
[add_recomputation]: 5.084e-05
[cse_after_recomputation]: 2.047e-05, [1]
    [Cycle 1]: 1.498e-05, [1]
        [cse]: 9.10999e-06
[auto_monad_reorder]: 1.544e-05
[get_jit_bprop_graph]: 2.24999e-06
[rewriter_after_jit_bprop_graph]: 5.59998e-06
[opt_after_jit_grad]: 0.00049651
[symbol_engine_optimizer]: 7.567e-05, [1]
    [Cycle 1]: 6.824e-05, [6]
        [build]: 4.3e-06
        [elim_shapecalc]: 7.2e-06
        [elim_not_effective]: 1.295e-05
        [opt_reshape]: 5.77001e-06
        [fold_const_symbol]: 9.20999e-06
        [renormalize]: 8.00006e-07
[validate]: 4.138e-05
[backend_pass]: 1.15999e-06
[task_emit]: 0.00666823
[execute]: 7.34002e-06
Sums
    bootstrap                            :     0.000279s :  1.78%
    type_inference                       :     0.004588s : 29.30%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000065s :  0.42%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000040s :  0.26%
    rewriter_before_opt_a                :     0.000073s :  0.47%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.43%
    jit_opt_a.loop_unroll                :     0.000037s :  0.24%
    jit_opt_a.a_1                        :     0.000645s :  4.12%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000896s :  5.72%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.11%
    jit_opt_a.cse                        :     0.000030s :  0.19%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000040s :  0.26%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000658s :  4.20%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000027s :  0.17%
    loop_unroll                          :     0.000442s :  2.82%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000051s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000497s :  3.17%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006668s : 42.59%
    execute                              :     0.000007s :  0.05%


TotalTime = 0.0110144, [30]
[bootstrap]: 0.00026275
[type_inference]: 0.00389858
[event_method]: 1.641e-05
[auto_monad]: 6.484e-05
[graph_reusing]: 5.62999e-06
[pre_auto_parallel]: 2.37999e-06
[py_interpret_to_execute]: 2.882e-05
[rewriter_before_opt_a]: 7.209e-05
[expand_dump_flag]: 2.80997e-06
[jit_opt_a]: 0.00470067, [2]
    [Cycle 1]: 0.00136295, [27]
        [switch_simplify]: 4.865e-05
        [loop_unroll]: 3.05e-05
        [a_1]: 0.00050154
        [with_stream_mark]: 1.517e-05
        [recompute_prepare]: 7.33e-06
        [updatestate_depend_eliminate]: 3.33e-06
        [updatestate_assign_eliminate]: 3.06999e-06
        [updatestate_loads_eliminate]: 2.98998e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 6.12999e-06
        [updatestate_useless_node_eliminater]: 5.30999e-06
        [accelerated_algorithm]: 5.86e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 5.47001e-06
        [merge_forward]: 3.46001e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.425e-05
        [j_node_and_user_rematch]: 8.94998e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 8.38999e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 0.00045954
        [add_forward_monad_depend]: 4.99998e-06
        [auto_monad_grad]: 2.01998e-06
        [auto_monad_eliminator]: 1.342e-05
        [cse]: 2.567e-05
        [replace_applicator]: 1.133e-05
    [Cycle 2]: 0.00032171, [27]
        [switch_simplify]: 5.82999e-06
        [loop_unroll]: 5.33002e-06
        [a_1]: 9.107e-05
        [with_stream_mark]: 8.2e-06
        [recompute_prepare]: 5.15999e-06
        [updatestate_depend_eliminate]: 2.53e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.98997e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 4.91997e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 5.32999e-06
        [meta_shard_fg_expand]: 1.06997e-06
        [get_grad_eliminate_]: 4.62e-06
        [merge_forward]: 2.43998e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.209e-05
        [j_node_and_user_rematch]: 7.8e-06
        [meta_fg_expand]: 1.56002e-06
        [replace_old_param]: 7.67002e-06
        [inline_without_move]: 4.89e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 5.05001e-06
        [cse]: 9.81e-06
        [replace_applicator]: 5.20999e-06
[py_interpret_to_execute_after_opt_a]: 8.33001e-06
[rewriter_after_opt_a]: 3.058e-05
[convert_after_rewriter]: 6.81999e-06
[order_py_execute_after_rewriter]: 5.89e-06
[mutable_eliminate]: 0.00046756
[jit_opt_b]: 5.158e-05, [1]
    [Cycle 1]: 4.533e-05, [2]
        [frontend_op_eliminate]: 1.73e-05
        [inline_after_opt_a]: 1.59e-05
[cconv]: 2.481e-05
[loop_unroll]: 0.00040831
[jit_opt_after_cconv]: 0.00013591, [1]
    [Cycle 1]: 0.00012976, [11]
        [c_1]: 2.081e-05
        [parameter_eliminate]: 2.64999e-06
        [updatestate_depend_eliminate]: 5.36002e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 1.621e-05
        [call_graph_tuple_transform]: 1.79e-05
        [tuple_list_get_item_eliminator]: 5.34e-06
        [none_parameter_eliminate]: 1.25001e-06
        [renormalize]: 4.2998e-07
        [switch_simplify]: 5.29998e-06
[remove_dup_value]: 1.404e-05
[partial_unused_args_eliminate]: 2.19001e-06
[environ_conv]: 5.61e-06
[add_recomputation]: 4.482e-05
[cse_after_recomputation]: 1.938e-05, [1]
    [Cycle 1]: 1.434e-05, [1]
        [cse]: 8.80001e-06
[auto_monad_reorder]: 1.413e-05
[get_jit_bprop_graph]: 1.35001e-06
[rewriter_after_jit_bprop_graph]: 4.82e-06
[opt_after_jit_grad]: 0.0004396
[symbol_engine_optimizer]: 7.081e-05, [1]
    [Cycle 1]: 6.486e-05, [6]
        [build]: 3.18e-06
        [elim_shapecalc]: 7.82e-06
        [elim_not_effective]: 1.247e-05
        [opt_reshape]: 5.92001e-06
        [fold_const_symbol]: 8.92e-06
        [renormalize]: 4.30009e-07
[validate]: 2.944e-05
Sums
    bootstrap                            :     0.000263s :  3.54%
    type_inference                       :     0.003899s : 52.49%
    event_method                         :     0.000016s :  0.22%
    auto_monad                           :     0.000065s :  0.87%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000029s :  0.39%
    rewriter_before_opt_a                :     0.000072s :  0.97%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000054s :  0.73%
    jit_opt_a.loop_unroll                :     0.000036s :  0.48%
    jit_opt_a.a_1                        :     0.000593s :  7.98%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.07%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.14%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.22%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000460s :  6.19%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.25%
    jit_opt_a.cse                        :     0.000035s :  0.48%
    jit_opt_a.replace_applicator         :     0.000017s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000031s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000006s :  0.08%
    mutable_eliminate                    :     0.000468s :  6.30%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.21%
    cconv                                :     0.000025s :  0.33%
    loop_unroll                          :     0.000408s :  5.50%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.19%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.08%
    add_recomputation                    :     0.000045s :  0.60%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000014s :  0.19%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000440s :  5.92%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000029s :  0.40%


Time group info:
------[substitution.]   0.000210    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.53% :     0.000005s :      3: substitution.graph_param_transform
83.84% :     0.000176s :      4: substitution.inline
 1.64% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.77% :     0.000006s :      4: substitution.remove_not_recompute_node
 1.76% :     0.000004s :      2: substitution.replace_old_param
 5.88% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004516     2
78.31% :     0.003537s :      1: type_inference.infer
21.69% :     0.000979s :      1: type_inference.specialize
------[replace.]   0.000075     6
67.39% :     0.000050s :      4: replace.inline
32.61% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000185     6
93.93% :     0.000174s :      4: match.inline
 6.07% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.19% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.20% :     0.000001s :     10: predicate.addn_zero_filter
 2.38% :     0.000003s :     10: predicate.arithmetic_simplify
 1.48% :     0.000002s :     10: predicate.cast_eliminate
 0.57% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000001s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.37% :     0.000002s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.00% :     0.000006s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000002s :      6: predicate.less_batch_normalization
 1.77% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.04% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.18% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.81% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.61% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.65% :     0.000002s :     10: predicate.reshape_eliminate
 1.39% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.20% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.58% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.92% :     0.000009s :     53: predicate.switch_simplify
 1.33% :     0.000002s :     10: predicate.tile_eliminate
 1.35% :     0.000002s :     10: predicate.transpose_eliminate
 1.73% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.78% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.48% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000663    12
43.02% :     0.000285s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.93% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
52.05% :     0.000345s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021737    76
 0.25% :     0.000054s :      1: add_recomputation
 0.31% :     0.000068s :      1: auto_monad
 0.08% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.33% :     0.000290s :      1: bootstrap
 0.13% :     0.000029s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000023s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.68% :     0.005799s :      1: jit_opt_a
 0.69% :     0.000150s :      1: jit_opt_after_cconv
 0.26% :     0.000055s :      1: jit_opt_b
 2.07% :     0.000450s :      1: loop_unroll
 3.07% :     0.000667s :      1: mutable_eliminate
 3.97% :     0.000863s :     26: opt.transform.jit_opt_a
 0.23% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.32% :     0.000505s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.20% :     0.000043s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 2.18% :     0.000473s :      1: renormalize.infer
 1.91% :     0.000414s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000044s :      1: rewriter_after_opt_a
 0.35% :     0.000076s :      1: rewriter_before_opt_a
 0.36% :     0.000078s :      1: symbol_engine_optimizer
30.75% :     0.006685s :      1: task_emit
21.18% :     0.004604s :      1: type_inference
 0.29% :     0.000063s :      1: validate


Time group info:
------[substitution.]   0.000178    23
 1.07% :     0.000002s :      2: substitution.elim_not_effective
 0.89% :     0.000002s :      2: substitution.fold_const_symbol
 2.81% :     0.000005s :      3: substitution.graph_param_transform
82.51% :     0.000147s :      4: substitution.inline
 1.55% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.42% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.07% :     0.000004s :      2: substitution.replace_old_param
 6.67% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003842     2
80.62% :     0.003097s :      1: type_inference.infer
19.38% :     0.000745s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.00% :     0.000042s :      4: replace.inline
33.00% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000155     6
93.05% :     0.000144s :      4: match.inline
 6.95% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.32% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.07% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000002s :     10: predicate.addn_zero_filter
 2.16% :     0.000003s :     10: predicate.arithmetic_simplify
 1.28% :     0.000002s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.50% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.19% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.47% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.39% :     0.000002s :      6: predicate.less_batch_normalization
 1.46% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.34% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.50% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000002s :     10: predicate.reshape_eliminate
 1.32% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.25% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.46% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.49% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.81% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.94% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000490    12
50.59% :     0.000248s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.36% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.06% :     0.000216s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012340    72
 0.39% :     0.000048s :      1: add_recomputation
 0.55% :     0.000068s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.24% :     0.000276s :      1: bootstrap
 0.22% :     0.000027s :      1: cconv
 0.24% :     0.000030s :      1: convert_after_rewriter
 0.18% :     0.000022s :      1: cse_after_recomputation
 0.07% :     0.000008s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.11% :     0.004703s :      1: jit_opt_a
 1.12% :     0.000139s :      1: jit_opt_after_cconv
 0.44% :     0.000054s :      1: jit_opt_b
 3.37% :     0.000416s :      1: loop_unroll
 3.85% :     0.000476s :      1: mutable_eliminate
 6.45% :     0.000796s :     26: opt.transform.jit_opt_a
 0.37% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.13% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.63% :     0.000447s :      1: opt_after_jit_grad
 0.07% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000031s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000017s :      1: remove_dup_value
 1.81% :     0.000223s :      1: renormalize.infer
 1.86% :     0.000229s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000033s :      1: rewriter_after_opt_a
 0.61% :     0.000075s :      1: rewriter_before_opt_a
 0.59% :     0.000073s :      1: symbol_engine_optimizer
31.67% :     0.003909s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:20.504.656 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c371820> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:20.505.543 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0114432, [30]
[bootstrap]: 0.00025589
[type_inference]: 0.00395603
[event_method]: 1.648e-05
[auto_monad]: 6.117e-05
[graph_reusing]: 5.77999e-06
[pre_auto_parallel]: 2.26e-06
[py_interpret_to_execute]: 3.069e-05
[rewriter_before_opt_a]: 7.096e-05
[expand_dump_flag]: 2.93e-06
[jit_opt_a]: 0.00481527, [2]
    [Cycle 1]: 0.00139042, [27]
        [switch_simplify]: 5.105e-05
        [loop_unroll]: 2.965e-05
        [a_1]: 0.0005059
        [with_stream_mark]: 1.589e-05
        [recompute_prepare]: 7.53999e-06
        [updatestate_depend_eliminate]: 3.08e-06
        [updatestate_assign_eliminate]: 3.26999e-06
        [updatestate_loads_eliminate]: 2.41998e-06
        [parameter_eliminate]: 2.58e-06
        [specialize_transform]: 5.94999e-06
        [updatestate_useless_node_eliminater]: 5.10999e-06
        [accelerated_algorithm]: 6.21998e-06
        [meta_shard_fg_expand]: 2.17999e-06
        [get_grad_eliminate_]: 5.27999e-06
        [merge_forward]: 3.69002e-06
        [cell_reuse_recompute_pass]: 1.00999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.373e-05
        [j_node_and_user_rematch]: 8.79e-06
        [meta_fg_expand]: 2.43998e-06
        [replace_old_param]: 8.88002e-06
        [inline_without_move]: 4.99e-06
        [renormalize]: 0.00047901
        [add_forward_monad_depend]: 4.82e-06
        [auto_monad_grad]: 2.43e-06
        [auto_monad_eliminator]: 1.251e-05
        [cse]: 2.395e-05
        [replace_applicator]: 1.125e-05
    [Cycle 2]: 0.00030948, [27]
        [switch_simplify]: 5.61e-06
        [loop_unroll]: 4.79e-06
        [a_1]: 8.751e-05
        [with_stream_mark]: 8.79e-06
        [recompute_prepare]: 5.02e-06
        [updatestate_depend_eliminate]: 2.62001e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 2.03997e-06
        [parameter_eliminate]: 1.06002e-06
        [specialize_transform]: 5.19e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 4.73001e-06
        [meta_shard_fg_expand]: 1.25001e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.202e-05
        [j_node_and_user_rematch]: 7.63001e-06
        [meta_fg_expand]: 1.72001e-06
        [replace_old_param]: 7.38999e-06
        [inline_without_move]: 5.10001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.51998e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 4.58999e-06
        [cse]: 9.32001e-06
        [replace_applicator]: 5.12e-06
[py_interpret_to_execute_after_opt_a]: 8.67e-06
[rewriter_after_opt_a]: 3.195e-05
[convert_after_rewriter]: 7.06999e-06
[order_py_execute_after_rewriter]: 4.99e-06
[mutable_eliminate]: 0.00048737
[jit_opt_b]: 5.191e-05, [1]
    [Cycle 1]: 4.414e-05, [2]
        [frontend_op_eliminate]: 1.697e-05
        [inline_after_opt_a]: 1.521e-05
[cconv]: 2.584e-05
[loop_unroll]: 0.00048423
[jit_opt_after_cconv]: 0.00016386, [1]
    [Cycle 1]: 0.00015556, [11]
        [c_1]: 2.273e-05
        [parameter_eliminate]: 3.65e-06
        [updatestate_depend_eliminate]: 7.74997e-06
        [updatestate_assign_eliminate]: 2.64999e-06
        [updatestate_loads_eliminate]: 2.11998e-06
        [cse]: 2.381e-05
        [call_graph_tuple_transform]: 2.247e-05
        [tuple_list_get_item_eliminator]: 6.41e-06
        [none_parameter_eliminate]: 1.64998e-06
        [renormalize]: 5.99975e-07
        [switch_simplify]: 6.19001e-06
[remove_dup_value]: 1.554e-05
[partial_unused_args_eliminate]: 2.78e-06
[environ_conv]: 6.10002e-06
[add_recomputation]: 5.272e-05
[cse_after_recomputation]: 2.345e-05, [1]
    [Cycle 1]: 1.765e-05, [1]
        [cse]: 1.054e-05
[auto_monad_reorder]: 1.501e-05
[get_jit_bprop_graph]: 1.67999e-06
[rewriter_after_jit_bprop_graph]: 8.04002e-06
[opt_after_jit_grad]: 0.00055675
[symbol_engine_optimizer]: 7.632e-05, [1]
    [Cycle 1]: 6.934e-05, [6]
        [build]: 4.20999e-06
        [elim_shapecalc]: 7.98999e-06
        [elim_not_effective]: 1.298e-05
        [opt_reshape]: 6.59999e-06
        [fold_const_symbol]: 9.29e-06
        [renormalize]: 4.09986e-07
[validate]: 3.751e-05
Sums
    bootstrap                            :     0.000256s :  3.30%
    type_inference                       :     0.003956s : 51.01%
    event_method                         :     0.000016s :  0.21%
    auto_monad                           :     0.000061s :  0.79%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000031s :  0.40%
    rewriter_before_opt_a                :     0.000071s :  0.91%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000057s :  0.73%
    jit_opt_a.loop_unroll                :     0.000034s :  0.44%
    jit_opt_a.a_1                        :     0.000593s :  7.65%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.05%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000479s :  6.18%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.22%
    jit_opt_a.cse                        :     0.000033s :  0.43%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000032s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000487s :  6.28%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.20%
    cconv                                :     0.000026s :  0.33%
    loop_unroll                          :     0.000484s :  6.24%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.29%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.10%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000024s :  0.31%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.29%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.08%
    remove_dup_value                     :     0.000016s :  0.20%
    partial_unused_args_eliminate        :     0.000003s :  0.04%
    environ_conv                         :     0.000006s :  0.08%
    add_recomputation                    :     0.000053s :  0.68%
    cse_after_recomputation.cse          :     0.000011s :  0.14%
    auto_monad_reorder                   :     0.000015s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.10%
    opt_after_jit_grad                   :     0.000557s :  7.18%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.09%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000038s :  0.48%


Time group info:
------[substitution.]   0.000179    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.90% :     0.000002s :      2: substitution.fold_const_symbol
 3.24% :     0.000006s :      3: substitution.graph_param_transform
82.74% :     0.000148s :      4: substitution.inline
 1.52% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.33% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.90% :     0.000003s :      2: substitution.replace_old_param
 6.34% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003896     2
80.66% :     0.003143s :      1: type_inference.infer
19.34% :     0.000753s :      1: type_inference.specialize
------[replace.]   0.000063     6
65.98% :     0.000042s :      4: replace.inline
34.02% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000155     6
93.47% :     0.000145s :      4: match.inline
 6.53% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.55% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.49% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 2.00% :     0.000002s :     10: predicate.arithmetic_simplify
 1.10% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.19% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.06% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.08% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.17% :     0.000001s :      6: predicate.get_grad_eliminate
 0.48% :     0.000001s :      3: predicate.graph_param_transform
 5.54% :     0.000007s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 2.01% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.90% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.82% :     0.000002s :      3: predicate.mutable_eliminate
 0.61% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.80% :     0.000002s :     10: predicate.reduce_eliminate
 1.86% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.82% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.32% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.66% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.39% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.67% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.24% :     0.000009s :     53: predicate.switch_simplify
 1.37% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.75% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000497    12
50.34% :     0.000250s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.45% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.21% :     0.000220s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012786    72
 0.44% :     0.000056s :      1: add_recomputation
 0.50% :     0.000064s :      1: auto_monad
 0.14% :     0.000018s :      1: auto_monad_reorder
 2.10% :     0.000269s :      1: bootstrap
 0.22% :     0.000029s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.20% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
37.68% :     0.004818s :      1: jit_opt_a
 1.31% :     0.000167s :      1: jit_opt_after_cconv
 0.43% :     0.000054s :      1: jit_opt_b
 3.86% :     0.000494s :      1: loop_unroll
 3.88% :     0.000496s :      1: mutable_eliminate
 6.23% :     0.000796s :     26: opt.transform.jit_opt_a
 0.41% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000026s :      4: opt.transform.jit_opt_b
 0.11% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.18% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000033s :      4: opt.transform.symbol_engine_opt
 4.43% :     0.000566s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.04% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000034s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.14% :     0.000018s :      1: remove_dup_value
 1.83% :     0.000234s :      1: renormalize.infer
 1.86% :     0.000237s :      1: renormalize.specialize
 0.08% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000035s :      1: rewriter_after_opt_a
 0.58% :     0.000074s :      1: rewriter_before_opt_a
 0.62% :     0.000079s :      1: symbol_engine_optimizer
31.03% :     0.003967s :      1: type_inference


TotalTime = 0.0173649, [33]
[bootstrap]: 0.00030815
[type_inference]: 0.00445031
[event_method]: 1.586e-05
[auto_monad]: 5.183e-05
[graph_reusing]: 4.18999e-06
[pre_auto_parallel]: 1.73002e-06
[py_interpret_to_execute]: 3.194e-05
[rewriter_before_opt_a]: 6.596e-05
[expand_dump_flag]: 2.51e-06
[jit_opt_a]: 0.0049492, [2]
    [Cycle 1]: 0.00144601, [27]
        [switch_simplify]: 4.507e-05
        [loop_unroll]: 2.9e-05
        [a_1]: 0.00048576
        [with_stream_mark]: 1.23e-05
        [recompute_prepare]: 7.23999e-06
        [updatestate_depend_eliminate]: 3.12997e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 1.39e-06
        [specialize_transform]: 5.87999e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 5.30999e-06
        [merge_forward]: 2.54999e-06
        [cell_reuse_recompute_pass]: 8.79983e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.118e-05
        [j_node_and_user_rematch]: 8.51002e-06
        [meta_fg_expand]: 1.72999e-06
        [replace_old_param]: 7.58999e-06
        [inline_without_move]: 5.02e-06
        [renormalize]: 0.00059249
        [add_forward_monad_depend]: 3.88999e-06
        [auto_monad_grad]: 1.22e-06
        [auto_monad_eliminator]: 1.018e-05
        [cse]: 1.733e-05
        [replace_applicator]: 1.102e-05
    [Cycle 2]: 0.000313, [27]
        [switch_simplify]: 5.55001e-06
        [loop_unroll]: 4.62e-06
        [a_1]: 8.811e-05
        [with_stream_mark]: 8.55999e-06
        [recompute_prepare]: 4.99e-06
        [updatestate_depend_eliminate]: 2.43e-06
        [updatestate_assign_eliminate]: 1.95001e-06
        [updatestate_loads_eliminate]: 1.91003e-06
        [parameter_eliminate]: 8.60018e-07
        [specialize_transform]: 4.60999e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 5.42999e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 4.32e-06
        [merge_forward]: 2.43998e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.215e-05
        [j_node_and_user_rematch]: 7.75e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 1.254e-05
        [inline_without_move]: 4.58999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.17999e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 5.00001e-06
        [cse]: 9.82001e-06
        [replace_applicator]: 4.99e-06
[py_interpret_to_execute_after_opt_a]: 8.11002e-06
[rewriter_after_opt_a]: 2.807e-05
[convert_after_rewriter]: 5.44998e-06
[order_py_execute_after_rewriter]: 4.50999e-06
[mutable_eliminate]: 0.000509
[jit_opt_b]: 4.875e-05, [1]
    [Cycle 1]: 4.174e-05, [2]
        [frontend_op_eliminate]: 1.605e-05
        [inline_after_opt_a]: 1.435e-05
[cconv]: 1.816e-05
[loop_unroll]: 0.00042449
[jit_opt_after_cconv]: 0.0001325, [1]
    [Cycle 1]: 0.00012604, [11]
        [c_1]: 2.039e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 5.22999e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [cse]: 1.697e-05
        [call_graph_tuple_transform]: 1.642e-05
        [tuple_list_get_item_eliminator]: 5.31002e-06
        [none_parameter_eliminate]: 1.10999e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 5.29e-06
[remove_dup_value]: 8.22003e-06
[partial_unused_args_eliminate]: 1.84e-06
[environ_conv]: 3.75e-06
[add_recomputation]: 3.366e-05
[cse_after_recomputation]: 1.918e-05, [1]
    [Cycle 1]: 1.41e-05, [1]
        [cse]: 8.20999e-06
[auto_monad_reorder]: 1.014e-05
[get_jit_bprop_graph]: 1.50999e-06
[rewriter_after_jit_bprop_graph]: 5.82001e-06
[opt_after_jit_grad]: 0.00048044
[symbol_engine_optimizer]: 7.103e-05, [1]
    [Cycle 1]: 6.447e-05, [6]
        [build]: 3.33998e-06
        [elim_shapecalc]: 7.68001e-06
        [elim_not_effective]: 1.193e-05
        [opt_reshape]: 5.70001e-06
        [fold_const_symbol]: 8.64998e-06
        [renormalize]: 4.30009e-07
[validate]: 2.823e-05
[backend_pass]: 8.2e-07
[task_emit]: 0.00544283
[execute]: 5.45001e-06
Sums
    bootstrap                            :     0.000308s :  2.27%
    type_inference                       :     0.004450s : 32.74%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000052s :  0.38%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000032s :  0.23%
    rewriter_before_opt_a                :     0.000066s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000051s :  0.37%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000574s :  4.22%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000020s :  0.15%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000593s :  4.36%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000027s :  0.20%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000028s :  0.21%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000509s :  3.74%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000018s :  0.13%
    loop_unroll                          :     0.000424s :  3.12%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000008s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000034s :  0.25%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000010s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000480s :  3.53%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005443s : 40.04%
    execute                              :     0.000005s :  0.04%


Time group info:
------[substitution.]   0.000169    23
 0.94% :     0.000002s :      2: substitution.elim_not_effective
 0.85% :     0.000001s :      2: substitution.fold_const_symbol
 2.05% :     0.000003s :      3: substitution.graph_param_transform
82.27% :     0.000139s :      4: substitution.inline
 1.53% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000004s :      4: substitution.remove_not_recompute_node
 4.70% :     0.000008s :      2: substitution.replace_old_param
 5.43% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004398     2
80.59% :     0.003544s :      1: type_inference.infer
19.41% :     0.000854s :      1: type_inference.specialize
------[replace.]   0.000065     6
66.30% :     0.000043s :      4: replace.inline
33.70% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000144     6
94.45% :     0.000136s :      4: match.inline
 5.55% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.35% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.28% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.54% :     0.000002s :     10: predicate.addn_zero_filter
 2.08% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.41% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.43% :     0.000000s :      3: predicate.graph_param_transform
 5.17% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000001s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.29% :     0.000001s :     10: predicate.minmaximum_grad
 1.42% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.20% :     0.000002s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.07% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.82% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000001s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.62% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.63% :     0.000008s :     53: predicate.switch_simplify
 1.28% :     0.000001s :     10: predicate.tile_eliminate
 1.31% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.48% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000587    12
48.49% :     0.000284s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.85% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.66% :     0.000274s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018829    76
 0.19% :     0.000036s :      1: add_recomputation
 0.29% :     0.000055s :      1: auto_monad
 0.07% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.70% :     0.000320s :      1: bootstrap
 0.11% :     0.000021s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
26.30% :     0.004952s :      1: jit_opt_a
 0.72% :     0.000135s :      1: jit_opt_after_cconv
 0.27% :     0.000051s :      1: jit_opt_b
 2.30% :     0.000432s :      1: loop_unroll
 2.75% :     0.000517s :      1: mutable_eliminate
 4.10% :     0.000772s :     26: opt.transform.jit_opt_a
 0.24% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.59% :     0.000489s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000035s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000011s :      1: remove_dup_value
 1.64% :     0.000309s :      1: renormalize.infer
 1.47% :     0.000277s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000031s :      1: rewriter_after_opt_a
 0.37% :     0.000069s :      1: rewriter_before_opt_a
 0.39% :     0.000073s :      1: symbol_engine_optimizer
28.96% :     0.005453s :      1: task_emit
23.69% :     0.004461s :      1: type_inference
 0.30% :     0.000057s :      1: validate


TotalTime = 0.011096, [30]
[bootstrap]: 0.00026509
[type_inference]: 0.00398
[event_method]: 1.605e-05
[auto_monad]: 6.187e-05
[graph_reusing]: 5.47999e-06
[pre_auto_parallel]: 2.43998e-06
[py_interpret_to_execute]: 3.112e-05
[rewriter_before_opt_a]: 7.07e-05
[expand_dump_flag]: 2.76999e-06
[jit_opt_a]: 0.00474918, [2]
    [Cycle 1]: 0.00135011, [27]
        [switch_simplify]: 4.937e-05
        [loop_unroll]: 3.122e-05
        [a_1]: 0.0004886
        [with_stream_mark]: 1.515e-05
        [recompute_prepare]: 6.68e-06
        [updatestate_depend_eliminate]: 3.11001e-06
        [updatestate_assign_eliminate]: 2.59001e-06
        [updatestate_loads_eliminate]: 2.64001e-06
        [parameter_eliminate]: 1.79998e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.67001e-06
        [meta_shard_fg_expand]: 1.90001e-06
        [get_grad_eliminate_]: 5.52999e-06
        [merge_forward]: 3.35e-06
        [cell_reuse_recompute_pass]: 9.90025e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.45e-05
        [j_node_and_user_rematch]: 8.58001e-06
        [meta_fg_expand]: 2.17001e-06
        [replace_old_param]: 8.28001e-06
        [inline_without_move]: 5.33002e-06
        [renormalize]: 0.00046898
        [add_forward_monad_depend]: 5.10001e-06
        [auto_monad_grad]: 2.01e-06
        [auto_monad_eliminator]: 1.247e-05
        [cse]: 2.172e-05
        [replace_applicator]: 1.069e-05
    [Cycle 2]: 0.00031928, [27]
        [switch_simplify]: 5.97999e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 9.089e-05
        [with_stream_mark]: 8.23999e-06
        [recompute_prepare]: 4.85999e-06
        [updatestate_depend_eliminate]: 2.53e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 4.88001e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.10999e-06
        [meta_shard_fg_expand]: 1.01002e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.71998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.208e-05
        [j_node_and_user_rematch]: 7.59002e-06
        [meta_fg_expand]: 1.64e-06
        [replace_old_param]: 6.98e-06
        [inline_without_move]: 4.80001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 7.49977e-07
        [auto_monad_eliminator]: 4.89e-06
        [cse]: 9.93998e-06
        [replace_applicator]: 5.26002e-06
[py_interpret_to_execute_after_opt_a]: 8.2e-06
[rewriter_after_opt_a]: 2.9e-05
[convert_after_rewriter]: 6.53e-06
[order_py_execute_after_rewriter]: 7.2e-06
[mutable_eliminate]: 0.0004561
[jit_opt_b]: 4.898e-05, [1]
    [Cycle 1]: 4.249e-05, [2]
        [frontend_op_eliminate]: 1.603e-05
        [inline_after_opt_a]: 1.502e-05
[cconv]: 2.412e-05
[loop_unroll]: 0.00040292
[jit_opt_after_cconv]: 0.00013376, [1]
    [Cycle 1]: 0.00012768, [11]
        [c_1]: 2.038e-05
        [parameter_eliminate]: 2.39001e-06
        [updatestate_depend_eliminate]: 5.32001e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 2.11998e-06
        [cse]: 1.581e-05
        [call_graph_tuple_transform]: 1.708e-05
        [tuple_list_get_item_eliminator]: 5.50001e-06
        [none_parameter_eliminate]: 1.27999e-06
        [renormalize]: 4.7998e-07
        [switch_simplify]: 5.54e-06
[remove_dup_value]: 1.288e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 4.60999e-06
[add_recomputation]: 4.303e-05
[cse_after_recomputation]: 1.894e-05, [1]
    [Cycle 1]: 1.375e-05, [1]
        [cse]: 8.25e-06
[auto_monad_reorder]: 1.473e-05
[get_jit_bprop_graph]: 1.34e-06
[rewriter_after_jit_bprop_graph]: 4.65001e-06
[opt_after_jit_grad]: 0.00044245
[symbol_engine_optimizer]: 7.184e-05, [1]
    [Cycle 1]: 6.567e-05, [6]
        [build]: 3.23998e-06
        [elim_shapecalc]: 7.17997e-06
        [elim_not_effective]: 1.255e-05
        [opt_reshape]: 5.82001e-06
        [fold_const_symbol]: 8.90999e-06
        [renormalize]: 5.00004e-07
[validate]: 2.906e-05
Sums
    bootstrap                            :     0.000265s :  3.55%
    type_inference                       :     0.003980s : 53.27%
    event_method                         :     0.000016s :  0.21%
    auto_monad                           :     0.000062s :  0.83%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000031s :  0.42%
    rewriter_before_opt_a                :     0.000071s :  0.95%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000055s :  0.74%
    jit_opt_a.loop_unroll                :     0.000036s :  0.48%
    jit_opt_a.a_1                        :     0.000579s :  7.76%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.36%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000469s :  6.28%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.23%
    jit_opt_a.cse                        :     0.000032s :  0.42%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000029s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000007s :  0.10%
    mutable_eliminate                    :     0.000456s :  6.10%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.20%
    cconv                                :     0.000024s :  0.32%
    loop_unroll                          :     0.000403s :  5.39%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000013s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000043s :  0.58%
    cse_after_recomputation.cse          :     0.000008s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.20%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000442s :  5.92%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000029s :  0.39%


Time group info:
------[substitution.]   0.000170    23
 1.42% :     0.000002s :      2: substitution.elim_not_effective
 0.77% :     0.000001s :      2: substitution.fold_const_symbol
 2.75% :     0.000005s :      3: substitution.graph_param_transform
82.43% :     0.000140s :      4: substitution.inline
 1.62% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.59% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000003s :      2: substitution.replace_old_param
 6.37% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003922     2
80.76% :     0.003167s :      1: type_inference.infer
19.24% :     0.000754s :      1: type_inference.specialize
------[replace.]   0.000059     6
65.81% :     0.000039s :      4: replace.inline
34.19% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000147     6
93.48% :     0.000137s :      4: match.inline
 6.52% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.73% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.07% :     0.000002s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.11% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.36% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.19% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.55% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.41% :     0.000002s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.49% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.47% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.65% :     0.000002s :     10: predicate.reduce_eliminate
 1.63% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.64% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.32% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.01% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.46% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.60% :     0.000009s :     53: predicate.switch_simplify
 1.27% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.95% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000497    12
50.73% :     0.000252s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.23% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.04% :     0.000219s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012413    72
 0.37% :     0.000046s :      1: add_recomputation
 0.53% :     0.000065s :      1: auto_monad
 0.14% :     0.000018s :      1: auto_monad_reorder
 2.24% :     0.000277s :      1: bootstrap
 0.22% :     0.000027s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000021s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000019s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.28% :     0.004752s :      1: jit_opt_a
 1.10% :     0.000137s :      1: jit_opt_after_cconv
 0.42% :     0.000052s :      1: jit_opt_b
 3.31% :     0.000411s :      1: loop_unroll
 3.74% :     0.000464s :      1: mutable_eliminate
 6.29% :     0.000780s :     26: opt.transform.jit_opt_a
 0.36% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.62% :     0.000450s :      1: opt_after_jit_grad
 0.08% :     0.000009s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000034s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000015s :      1: remove_dup_value
 1.84% :     0.000229s :      1: renormalize.infer
 1.88% :     0.000233s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000032s :      1: rewriter_after_opt_a
 0.59% :     0.000074s :      1: rewriter_before_opt_a
 0.60% :     0.000074s :      1: symbol_engine_optimizer
32.15% :     0.003991s :      1: type_inference


TotalTime = 0.0197497, [33]
[bootstrap]: 0.00032456
[type_inference]: 0.00517798
[event_method]: 1.87e-05
[auto_monad]: 6.854e-05
[graph_reusing]: 6.01e-06
[pre_auto_parallel]: 3.36999e-06
[py_interpret_to_execute]: 4.113e-05
[rewriter_before_opt_a]: 7.882e-05
[expand_dump_flag]: 3.66999e-06
[jit_opt_a]: 0.00580726, [2]
    [Cycle 1]: 0.00172565, [27]
        [switch_simplify]: 5.706e-05
        [loop_unroll]: 2.897e-05
        [a_1]: 0.00052666
        [with_stream_mark]: 1.675e-05
        [recompute_prepare]: 7.11999e-06
        [updatestate_depend_eliminate]: 3.96001e-06
        [updatestate_assign_eliminate]: 3.06999e-06
        [updatestate_loads_eliminate]: 2.72001e-06
        [parameter_eliminate]: 2.19999e-06
        [specialize_transform]: 5.86003e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 2.28998e-06
        [get_grad_eliminate_]: 4.99998e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.381e-05
        [j_node_and_user_rematch]: 9.02e-06
        [meta_fg_expand]: 2.02999e-06
        [replace_old_param]: 8.45999e-06
        [inline_without_move]: 5.48002e-06
        [renormalize]: 0.0007851
        [add_forward_monad_depend]: 5.96e-06
        [auto_monad_grad]: 2.26998e-06
        [auto_monad_eliminator]: 1.341e-05
        [cse]: 2.562e-05
        [replace_applicator]: 1.238e-05
    [Cycle 2]: 0.0003183, [27]
        [switch_simplify]: 5.69999e-06
        [loop_unroll]: 4.94998e-06
        [a_1]: 8.998e-05
        [with_stream_mark]: 9.55001e-06
        [recompute_prepare]: 4.90999e-06
        [updatestate_depend_eliminate]: 2.58003e-06
        [updatestate_assign_eliminate]: 2.10002e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 5.00001e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.02e-06
        [meta_shard_fg_expand]: 1.16002e-06
        [get_grad_eliminate_]: 4.46002e-06
        [merge_forward]: 2.66e-06
        [cell_reuse_recompute_pass]: 1.55999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.22e-05
        [j_node_and_user_rematch]: 7.63001e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 6.96001e-06
        [inline_without_move]: 4.50001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 5.43002e-06
        [cse]: 9.42001e-06
        [replace_applicator]: 5.33002e-06
[py_interpret_to_execute_after_opt_a]: 1.029e-05
[rewriter_after_opt_a]: 3.493e-05
[convert_after_rewriter]: 6.79001e-06
[order_py_execute_after_rewriter]: 4.85999e-06
[mutable_eliminate]: 0.00058218
[jit_opt_b]: 4.908e-05, [1]
    [Cycle 1]: 4.188e-05, [2]
        [frontend_op_eliminate]: 1.619e-05
        [inline_after_opt_a]: 1.431e-05
[cconv]: 2.61e-05
[loop_unroll]: 0.00045602
[jit_opt_after_cconv]: 0.00013722, [1]
    [Cycle 1]: 0.00013095, [11]
        [c_1]: 2.06e-05
        [parameter_eliminate]: 3.01001e-06
        [updatestate_depend_eliminate]: 5.54998e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.858e-05
        [call_graph_tuple_transform]: 1.809e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 5.20999e-06
[remove_dup_value]: 1.301e-05
[partial_unused_args_eliminate]: 2.11998e-06
[environ_conv]: 4.53999e-06
[add_recomputation]: 4.831e-05
[cse_after_recomputation]: 1.968e-05, [1]
    [Cycle 1]: 1.403e-05, [1]
        [cse]: 8.54e-06
[auto_monad_reorder]: 1.446e-05
[get_jit_bprop_graph]: 2.16e-06
[rewriter_after_jit_bprop_graph]: 5.81e-06
[opt_after_jit_grad]: 0.00046643
[symbol_engine_optimizer]: 7.264e-05, [1]
    [Cycle 1]: 6.671e-05, [6]
        [build]: 4.63999e-06
        [elim_shapecalc]: 7.66001e-06
        [elim_not_effective]: 1.249e-05
        [opt_reshape]: 5.35001e-06
        [fold_const_symbol]: 8.52e-06
        [renormalize]: 3.69997e-07
[validate]: 3.579e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00600629
[execute]: 7.99997e-06
Sums
    bootstrap                            :     0.000325s :  2.11%
    type_inference                       :     0.005178s : 33.68%
    event_method                         :     0.000019s :  0.12%
    auto_monad                           :     0.000069s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000041s :  0.27%
    rewriter_before_opt_a                :     0.000079s :  0.51%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000617s :  4.01%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000785s :  5.11%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000035s :  0.23%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000582s :  3.79%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000026s :  0.17%
    loop_unroll                          :     0.000456s :  2.97%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000048s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000466s :  3.03%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006006s : 39.07%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000205    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.48% :     0.000005s :      3: substitution.graph_param_transform
84.58% :     0.000174s :      4: substitution.inline
 1.61% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.09% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.97% :     0.000004s :      2: substitution.replace_old_param
 5.68% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005102     2
81.16% :     0.004141s :      1: type_inference.infer
18.84% :     0.000961s :      1: type_inference.specialize
------[replace.]   0.000067     6
66.50% :     0.000044s :      4: replace.inline
33.50% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000182     6
94.23% :     0.000171s :      4: match.inline
 5.77% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.75% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 1.97% :     0.000002s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000001s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.19% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000000s :      3: predicate.graph_param_transform
 5.62% :     0.000006s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.62% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.49% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.10% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.23% :     0.000001s :     10: predicate.minmaximum_grad
 1.49% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000002s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.92% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.84% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.13% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.15% :     0.000009s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.50% :     0.000002s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.75% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000686    12
49.20% :     0.000337s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.58% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.23% :     0.000310s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021451    76
 0.24% :     0.000051s :      1: add_recomputation
 0.34% :     0.000072s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.60% :     0.000343s :      1: bootstrap
 0.13% :     0.000029s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.11% :     0.000023s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.09% :     0.005811s :      1: jit_opt_a
 0.65% :     0.000140s :      1: jit_opt_after_cconv
 0.24% :     0.000052s :      1: jit_opt_b
 2.17% :     0.000464s :      1: loop_unroll
 2.75% :     0.000590s :      1: mutable_eliminate
 3.84% :     0.000824s :     26: opt.transform.jit_opt_a
 0.21% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000475s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.21% :     0.000045s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 1.96% :     0.000420s :      1: renormalize.infer
 1.67% :     0.000358s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000038s :      1: rewriter_after_opt_a
 0.38% :     0.000082s :      1: rewriter_before_opt_a
 0.35% :     0.000075s :      1: symbol_engine_optimizer
28.07% :     0.006022s :      1: task_emit
24.22% :     0.005197s :      1: type_inference
 0.26% :     0.000055s :      1: validate


TotalTime = 0.0111429, [30]
[bootstrap]: 0.00025475
[type_inference]: 0.00385403
[event_method]: 1.616e-05
[auto_monad]: 5.849e-05
[graph_reusing]: 5.71e-06
[pre_auto_parallel]: 2.44001e-06
[py_interpret_to_execute]: 2.993e-05
[rewriter_before_opt_a]: 6.919e-05
[expand_dump_flag]: 2.44999e-06
[jit_opt_a]: 0.00483573, [2]
    [Cycle 1]: 0.00146091, [27]
        [switch_simplify]: 5.134e-05
        [loop_unroll]: 4.458e-05
        [a_1]: 0.00051353
        [with_stream_mark]: 1.773e-05
        [recompute_prepare]: 7.41999e-06
        [updatestate_depend_eliminate]: 3.16001e-06
        [updatestate_assign_eliminate]: 2.71e-06
        [updatestate_loads_eliminate]: 2.52001e-06
        [parameter_eliminate]: 1.70001e-06
        [specialize_transform]: 5.90002e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.34e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 5.42999e-06
        [merge_forward]: 3.12002e-06
        [cell_reuse_recompute_pass]: 8.30012e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.387e-05
        [j_node_and_user_rematch]: 8.27998e-06
        [meta_fg_expand]: 2.38998e-06
        [replace_old_param]: 8.43001e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 0.000517
        [add_forward_monad_depend]: 5.37001e-06
        [auto_monad_grad]: 2.59001e-06
        [auto_monad_eliminator]: 1.255e-05
        [cse]: 2.615e-05
        [replace_applicator]: 1.11e-05
    [Cycle 2]: 0.00032558, [27]
        [switch_simplify]: 5.72999e-06
        [loop_unroll]: 5.04e-06
        [a_1]: 9.169e-05
        [with_stream_mark]: 8.89e-06
        [recompute_prepare]: 5.15001e-06
        [updatestate_depend_eliminate]: 2.55002e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [parameter_eliminate]: 9.99979e-07
        [specialize_transform]: 5.03002e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.41002e-06
        [meta_shard_fg_expand]: 1.09998e-06
        [get_grad_eliminate_]: 5.16002e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 1.45001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.252e-05
        [j_node_and_user_rematch]: 8.12998e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 1.332e-05
        [inline_without_move]: 4.84998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 5.47001e-06
        [cse]: 1.091e-05
        [replace_applicator]: 5.02e-06
[py_interpret_to_execute_after_opt_a]: 1.001e-05
[rewriter_after_opt_a]: 3.199e-05
[convert_after_rewriter]: 6.32001e-06
[order_py_execute_after_rewriter]: 7.5e-06
[mutable_eliminate]: 0.00051919
[jit_opt_b]: 5.002e-05, [1]
    [Cycle 1]: 4.344e-05, [2]
        [frontend_op_eliminate]: 1.641e-05
        [inline_after_opt_a]: 1.531e-05
[cconv]: 2.429e-05
[loop_unroll]: 0.00040943
[jit_opt_after_cconv]: 0.00013794, [1]
    [Cycle 1]: 0.00013088, [11]
        [c_1]: 2.049e-05
        [parameter_eliminate]: 2.59001e-06
        [updatestate_depend_eliminate]: 5.86003e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.19999e-06
        [cse]: 1.67e-05
        [call_graph_tuple_transform]: 1.786e-05
        [tuple_list_get_item_eliminator]: 5.57001e-06
        [none_parameter_eliminate]: 1.27e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 5.20001e-06
[remove_dup_value]: 1.358e-05
[partial_unused_args_eliminate]: 2.07999e-06
[environ_conv]: 5.02e-06
[add_recomputation]: 4.669e-05
[cse_after_recomputation]: 2.021e-05, [1]
    [Cycle 1]: 1.472e-05, [1]
        [cse]: 8.97e-06
[auto_monad_reorder]: 1.419e-05
[get_jit_bprop_graph]: 1.42999e-06
[rewriter_after_jit_bprop_graph]: 4.60999e-06
[opt_after_jit_grad]: 0.00044897
[symbol_engine_optimizer]: 7.18e-05, [1]
    [Cycle 1]: 6.571e-05, [6]
        [build]: 3.6e-06
        [elim_shapecalc]: 7.45e-06
        [elim_not_effective]: 1.233e-05
        [opt_reshape]: 5.81003e-06
        [fold_const_symbol]: 8.64003e-06
        [renormalize]: 4.19997e-07
[validate]: 3.24e-05
Sums
    bootstrap                            :     0.000255s :  3.38%
    type_inference                       :     0.003854s : 51.19%
    event_method                         :     0.000016s :  0.21%
    auto_monad                           :     0.000058s :  0.78%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000030s :  0.40%
    rewriter_before_opt_a                :     0.000069s :  0.92%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000057s :  0.76%
    jit_opt_a.loop_unroll                :     0.000050s :  0.66%
    jit_opt_a.a_1                        :     0.000605s :  8.04%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.14%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000022s :  0.29%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000517s :  6.87%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.24%
    jit_opt_a.cse                        :     0.000037s :  0.49%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.13%
    rewriter_after_opt_a                 :     0.000032s :  0.42%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000007s :  0.10%
    mutable_eliminate                    :     0.000519s :  6.90%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.20%
    cconv                                :     0.000024s :  0.32%
    loop_unroll                          :     0.000409s :  5.44%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000047s :  0.62%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000014s :  0.19%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000449s :  5.96%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000032s :  0.43%


Time group info:
------[substitution.]   0.000184    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.81% :     0.000005s :      3: substitution.graph_param_transform
83.63% :     0.000154s :      4: substitution.inline
 1.54% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.31% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000004s :      2: substitution.replace_old_param
 5.94% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003796     2
80.28% :     0.003047s :      1: type_inference.infer
19.72% :     0.000749s :      1: type_inference.specialize
------[replace.]   0.000064     6
66.83% :     0.000043s :      4: replace.inline
33.17% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000161     6
93.98% :     0.000151s :      4: match.inline
 6.02% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.70% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 2.77% :     0.000003s :     10: predicate.arithmetic_simplify
 1.07% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.14% :     0.000001s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.18% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.55% :     0.000007s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.16% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.51% :     0.000002s :      3: predicate.mutable_eliminate
 0.58% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000003s :     15: predicate.partial_eliminate
 1.58% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.83% :     0.000002s :     10: predicate.reduce_eliminate
 1.61% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.24% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.08% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.69% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.46% :     0.000003s :     16: predicate.switch_defer_inline
 2.38% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.57% :     0.000009s :     53: predicate.switch_simplify
 1.46% :     0.000002s :     10: predicate.tile_eliminate
 1.07% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.75% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.05% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000497    12
49.22% :     0.000245s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.08% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.71% :     0.000222s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012549    72
 0.39% :     0.000050s :      1: add_recomputation
 0.49% :     0.000062s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.13% :     0.000268s :      1: bootstrap
 0.21% :     0.000027s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.15% :     0.000019s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.56% :     0.004839s :      1: jit_opt_a
 1.12% :     0.000141s :      1: jit_opt_after_cconv
 0.42% :     0.000053s :      1: jit_opt_b
 3.33% :     0.000418s :      1: loop_unroll
 4.21% :     0.000528s :      1: mutable_eliminate
 6.61% :     0.000830s :     26: opt.transform.jit_opt_a
 0.36% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.64% :     0.000457s :      1: opt_after_jit_grad
 0.08% :     0.000010s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000033s :      1: py_interpret_to_execute
 0.10% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000016s :      1: remove_dup_value
 2.11% :     0.000264s :      1: renormalize.infer
 1.95% :     0.000245s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000035s :      1: rewriter_after_opt_a
 0.57% :     0.000072s :      1: rewriter_before_opt_a
 0.59% :     0.000075s :      1: symbol_engine_optimizer
30.79% :     0.003864s :      1: type_inference


TotalTime = 0.011287, [30]
[bootstrap]: 0.00026114
[type_inference]: 0.00392511
[event_method]: 1.814e-05
[auto_monad]: 6.241e-05
[graph_reusing]: 5.74e-06
[pre_auto_parallel]: 2.19001e-06
[py_interpret_to_execute]: 3.037e-05
[rewriter_before_opt_a]: 7.046e-05
[expand_dump_flag]: 2.74001e-06
[jit_opt_a]: 0.00486156, [2]
    [Cycle 1]: 0.00143582, [27]
        [switch_simplify]: 5.139e-05
        [loop_unroll]: 3.314e-05
        [a_1]: 0.00052361
        [with_stream_mark]: 1.592e-05
        [recompute_prepare]: 8.41002e-06
        [updatestate_depend_eliminate]: 3.58999e-06
        [updatestate_assign_eliminate]: 3.68999e-06
        [updatestate_loads_eliminate]: 2.70002e-06
        [parameter_eliminate]: 2.32001e-06
        [specialize_transform]: 6.29999e-06
        [updatestate_useless_node_eliminater]: 5.77001e-06
        [accelerated_algorithm]: 6.43998e-06
        [meta_shard_fg_expand]: 2.14e-06
        [get_grad_eliminate_]: 6.21e-06
        [merge_forward]: 4.07998e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.485e-05
        [j_node_and_user_rematch]: 1.002e-05
        [meta_fg_expand]: 2.61e-06
        [replace_old_param]: 9.25001e-06
        [inline_without_move]: 5.69e-06
        [renormalize]: 0.00048643
        [add_forward_monad_depend]: 5.06002e-06
        [auto_monad_grad]: 2.19001e-06
        [auto_monad_eliminator]: 1.231e-05
        [cse]: 2.385e-05
        [replace_applicator]: 1.149e-05
    [Cycle 2]: 0.00032397, [27]
        [switch_simplify]: 5.75001e-06
        [loop_unroll]: 5.37999e-06
        [a_1]: 9.249e-05
        [with_stream_mark]: 8.20999e-06
        [recompute_prepare]: 5.18002e-06
        [updatestate_depend_eliminate]: 2.48998e-06
        [updatestate_assign_eliminate]: 1.96998e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 9.30013e-07
        [specialize_transform]: 5.09e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.41e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.19e-05
        [j_node_and_user_rematch]: 7.77998e-06
        [meta_fg_expand]: 1.51998e-06
        [replace_old_param]: 7.21999e-06
        [inline_without_move]: 1.07e-05
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.33002e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 5.30001e-06
        [cse]: 1.032e-05
        [replace_applicator]: 5.19e-06
[py_interpret_to_execute_after_opt_a]: 8.42998e-06
[rewriter_after_opt_a]: 3.31e-05
[convert_after_rewriter]: 7.15e-06
[order_py_execute_after_rewriter]: 5.24e-06
[mutable_eliminate]: 0.00047547
[jit_opt_b]: 4.971e-05, [1]
    [Cycle 1]: 4.292e-05, [2]
        [frontend_op_eliminate]: 1.715e-05
        [inline_after_opt_a]: 1.468e-05
[cconv]: 2.515e-05
[loop_unroll]: 0.00047982
[jit_opt_after_cconv]: 0.00013748, [1]
    [Cycle 1]: 0.00013111, [11]
        [c_1]: 2.102e-05
        [parameter_eliminate]: 3.11001e-06
        [updatestate_depend_eliminate]: 5.22999e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.17001e-06
        [cse]: 1.598e-05
        [call_graph_tuple_transform]: 1.81e-05
        [tuple_list_get_item_eliminator]: 5.67999e-06
        [none_parameter_eliminate]: 1.37e-06
        [renormalize]: 4.60015e-07
        [switch_simplify]: 5.54e-06
[remove_dup_value]: 1.505e-05
[partial_unused_args_eliminate]: 2.17001e-06
[environ_conv]: 5.17e-06
[add_recomputation]: 4.604e-05
[cse_after_recomputation]: 2.072e-05, [1]
    [Cycle 1]: 1.529e-05, [1]
        [cse]: 9.45001e-06
[auto_monad_reorder]: 1.511e-05
[get_jit_bprop_graph]: 1.50001e-06
[rewriter_after_jit_bprop_graph]: 4.79998e-06
[opt_after_jit_grad]: 0.00045621
[symbol_engine_optimizer]: 7.346e-05, [1]
    [Cycle 1]: 6.724e-05, [6]
        [build]: 4.18001e-06
        [elim_shapecalc]: 7.44002e-06
        [elim_not_effective]: 1.265e-05
        [opt_reshape]: 5.81e-06
        [fold_const_symbol]: 9.08002e-06
        [renormalize]: 4.80009e-07
[validate]: 2.977e-05
Sums
    bootstrap                            :     0.000261s :  3.43%
    type_inference                       :     0.003925s : 51.50%
    event_method                         :     0.000018s :  0.24%
    auto_monad                           :     0.000062s :  0.82%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000030s :  0.40%
    rewriter_before_opt_a                :     0.000070s :  0.92%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000057s :  0.75%
    jit_opt_a.loop_unroll                :     0.000039s :  0.51%
    jit_opt_a.a_1                        :     0.000616s :  8.08%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.18%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.14%
    jit_opt_a.merge_forward              :     0.000006s :  0.09%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.22%
    jit_opt_a.inline_without_move        :     0.000016s :  0.22%
    jit_opt_a.renormalize                :     0.000487s :  6.38%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.23%
    jit_opt_a.cse                        :     0.000034s :  0.45%
    jit_opt_a.replace_applicator         :     0.000017s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000033s :  0.43%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000475s :  6.24%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000025s :  0.33%
    loop_unroll                          :     0.000480s :  6.30%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000015s :  0.20%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000046s :  0.60%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000015s :  0.20%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000456s :  5.99%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000030s :  0.39%


Time group info:
------[substitution.]   0.000179    23
 1.07% :     0.000002s :      2: substitution.elim_not_effective
 0.92% :     0.000002s :      2: substitution.fold_const_symbol
 2.98% :     0.000005s :      3: substitution.graph_param_transform
82.25% :     0.000147s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.61% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.24% :     0.000004s :      2: substitution.replace_old_param
 6.37% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003868     2
80.69% :     0.003121s :      1: type_inference.infer
19.31% :     0.000747s :      1: type_inference.specialize
------[replace.]   0.000066     6
66.74% :     0.000044s :      4: replace.inline
33.26% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000155     6
93.34% :     0.000145s :      4: match.inline
 6.66% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.51% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.44% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.21% :     0.000002s :     10: predicate.addn_zero_filter
 2.69% :     0.000003s :     10: predicate.arithmetic_simplify
 1.09% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.19% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.16% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.48% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.16% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.11% :     0.000006s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.44% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.41% :     0.000002s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.17% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.07% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.41% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.67% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.53% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.85% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.10% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.45% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.50% :     0.000009s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.06% :     0.000001s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.67% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.83% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.13% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.97% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000497    12
50.94% :     0.000253s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.36% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.70% :     0.000217s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012677    72
 0.39% :     0.000049s :      1: add_recomputation
 0.52% :     0.000066s :      1: auto_monad
 0.14% :     0.000018s :      1: auto_monad_reorder
 2.18% :     0.000277s :      1: bootstrap
 0.22% :     0.000028s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.17% :     0.000022s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.37% :     0.004864s :      1: jit_opt_a
 1.11% :     0.000140s :      1: jit_opt_after_cconv
 0.41% :     0.000052s :      1: jit_opt_b
 3.85% :     0.000488s :      1: loop_unroll
 3.82% :     0.000484s :      1: mutable_eliminate
 6.60% :     0.000837s :     26: opt.transform.jit_opt_a
 0.37% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.66% :     0.000464s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000033s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.14% :     0.000018s :      1: remove_dup_value
 1.88% :     0.000239s :      1: renormalize.infer
 1.89% :     0.000240s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000036s :      1: rewriter_after_opt_a
 0.58% :     0.000073s :      1: rewriter_before_opt_a
 0.60% :     0.000076s :      1: symbol_engine_optimizer
31.04% :     0.003935s :      1: type_inference


TotalTime = 0.0179161, [33]
[bootstrap]: 0.00027668
[type_inference]: 0.00418569
[event_method]: 1.757e-05
[auto_monad]: 5.255e-05
[graph_reusing]: 4.00998e-06
[pre_auto_parallel]: 1.79e-06
[py_interpret_to_execute]: 3.303e-05
[rewriter_before_opt_a]: 6.621e-05
[expand_dump_flag]: 2.31e-06
[jit_opt_a]: 0.00497463, [2]
    [Cycle 1]: 0.00159364, [27]
        [switch_simplify]: 4.403e-05
        [loop_unroll]: 2.817e-05
        [a_1]: 0.00054179
        [with_stream_mark]: 1.528e-05
        [recompute_prepare]: 8.35001e-06
        [updatestate_depend_eliminate]: 3.14001e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 6.42001e-06
        [updatestate_useless_node_eliminater]: 5.42999e-06
        [accelerated_algorithm]: 6.00002e-06
        [meta_shard_fg_expand]: 1.24998e-06
        [get_grad_eliminate_]: 5.40001e-06
        [merge_forward]: 2.44001e-06
        [cell_reuse_recompute_pass]: 1.00001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.203e-05
        [j_node_and_user_rematch]: 9.19998e-06
        [meta_fg_expand]: 2.25002e-06
        [replace_old_param]: 8.70999e-06
        [inline_without_move]: 5.15001e-06
        [renormalize]: 0.00064578
        [add_forward_monad_depend]: 6.51999e-06
        [auto_monad_grad]: 2.11003e-06
        [auto_monad_eliminator]: 1.401e-05
        [cse]: 2.929e-05
        [replace_applicator]: 1.428e-05
    [Cycle 2]: 0.00034264, [27]
        [switch_simplify]: 5.96998e-06
        [loop_unroll]: 5.17e-06
        [a_1]: 9.629e-05
        [with_stream_mark]: 9.97001e-06
        [recompute_prepare]: 6.02999e-06
        [updatestate_depend_eliminate]: 2.61e-06
        [updatestate_assign_eliminate]: 2.52001e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.44998e-06
        [specialize_transform]: 5.14998e-06
        [updatestate_useless_node_eliminater]: 5.10999e-06
        [accelerated_algorithm]: 5.05001e-06
        [meta_shard_fg_expand]: 1.57999e-06
        [get_grad_eliminate_]: 4.77e-06
        [merge_forward]: 2.32001e-06
        [cell_reuse_recompute_pass]: 1.99999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.353e-05
        [j_node_and_user_rematch]: 8.48999e-06
        [meta_fg_expand]: 1.52999e-06
        [replace_old_param]: 7.38e-06
        [inline_without_move]: 4.94998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.60001e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 6.06998e-06
        [cse]: 1.314e-05
        [replace_applicator]: 5.02e-06
[py_interpret_to_execute_after_opt_a]: 1.088e-05
[rewriter_after_opt_a]: 3.625e-05
[convert_after_rewriter]: 7.23999e-06
[order_py_execute_after_rewriter]: 5.24998e-06
[mutable_eliminate]: 0.00054674
[jit_opt_b]: 5.246e-05, [1]
    [Cycle 1]: 4.49e-05, [2]
        [frontend_op_eliminate]: 1.715e-05
        [inline_after_opt_a]: 1.551e-05
[cconv]: 2.415e-05
[loop_unroll]: 0.00045389
[jit_opt_after_cconv]: 0.00013725, [1]
    [Cycle 1]: 0.00013095, [11]
        [c_1]: 1.963e-05
        [parameter_eliminate]: 2.37999e-06
        [updatestate_depend_eliminate]: 5.35999e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.08998e-06
        [cse]: 2.056e-05
        [call_graph_tuple_transform]: 1.842e-05
        [tuple_list_get_item_eliminator]: 4.94e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 6.69999e-07
        [switch_simplify]: 5.12e-06
[remove_dup_value]: 1.196e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 5.40999e-06
[add_recomputation]: 5.024e-05
[cse_after_recomputation]: 1.925e-05, [1]
    [Cycle 1]: 1.41e-05, [1]
        [cse]: 8.48999e-06
[auto_monad_reorder]: 1.461e-05
[get_jit_bprop_graph]: 2.16998e-06
[rewriter_after_jit_bprop_graph]: 5.00999e-06
[opt_after_jit_grad]: 0.00045486
[symbol_engine_optimizer]: 7.182e-05, [1]
    [Cycle 1]: 6.524e-05, [6]
        [build]: 4.26001e-06
        [elim_shapecalc]: 7.43e-06
        [elim_not_effective]: 1.216e-05
        [opt_reshape]: 5.52001e-06
        [fold_const_symbol]: 8.45999e-06
        [renormalize]: 5.39992e-07
[validate]: 3.816e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.00615065
[execute]: 7.23e-06
Sums
    bootstrap                            :     0.000277s :  1.94%
    type_inference                       :     0.004186s : 29.32%
    event_method                         :     0.000018s :  0.12%
    auto_monad                           :     0.000053s :  0.37%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000033s :  0.23%
    rewriter_before_opt_a                :     0.000066s :  0.46%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000050s :  0.35%
    jit_opt_a.loop_unroll                :     0.000033s :  0.23%
    jit_opt_a.a_1                        :     0.000638s :  4.47%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000646s :  4.52%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.14%
    jit_opt_a.cse                        :     0.000042s :  0.30%
    jit_opt_a.replace_applicator         :     0.000019s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.08%
    rewriter_after_opt_a                 :     0.000036s :  0.25%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000547s :  3.83%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000454s :  3.18%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000050s :  0.35%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000455s :  3.19%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006151s : 43.09%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000167    23
 1.33% :     0.000002s :      2: substitution.elim_not_effective
 0.90% :     0.000002s :      2: substitution.fold_const_symbol
 3.69% :     0.000006s :      3: substitution.graph_param_transform
81.79% :     0.000137s :      4: substitution.inline
 1.95% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.63% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.23% :     0.000004s :      2: substitution.replace_old_param
 5.48% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004128     2
81.87% :     0.003379s :      1: type_inference.infer
18.13% :     0.000748s :      1: type_inference.specialize
------[replace.]   0.000059     6
63.36% :     0.000037s :      4: replace.inline
36.64% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000142     6
94.38% :     0.000134s :      4: match.inline
 5.62% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.46% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000001s :     10: predicate.addn_zero_filter
 1.92% :     0.000002s :     10: predicate.arithmetic_simplify
 1.31% :     0.000001s :     10: predicate.cast_eliminate
 0.64% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.54% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.30% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.23% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.36% :     0.000002s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.57% :     0.000006s :     22: predicate.inline
 1.04% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.41% :     0.000002s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.77% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000002s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.66% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.03% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.57% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.80% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 1.17% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.97% :     0.000008s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.50% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000523    12
46.86% :     0.000245s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.33% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.81% :     0.000240s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019509    76
 0.27% :     0.000053s :      1: add_recomputation
 0.29% :     0.000056s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.51% :     0.000295s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.06% :     0.000012s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
25.51% :     0.004978s :      1: jit_opt_a
 0.72% :     0.000140s :      1: jit_opt_after_cconv
 0.28% :     0.000055s :      1: jit_opt_b
 2.37% :     0.000462s :      1: loop_unroll
 2.85% :     0.000556s :      1: mutable_eliminate
 4.31% :     0.000842s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.37% :     0.000463s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000036s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000014s :      1: remove_dup_value
 1.79% :     0.000350s :      1: renormalize.infer
 1.47% :     0.000287s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000040s :      1: rewriter_after_opt_a
 0.35% :     0.000069s :      1: rewriter_before_opt_a
 0.38% :     0.000074s :      1: symbol_engine_optimizer
31.58% :     0.006161s :      1: task_emit
21.53% :     0.004199s :      1: type_inference
 0.30% :     0.000059s :      1: validate


TotalTime = 0.0990234, [33]
[bootstrap]: 0.00026314
[type_inference]: 0.0815991
[event_method]: 0.00028679
[auto_monad]: 0.00012219
[graph_reusing]: 6.98e-06
[pre_auto_parallel]: 2.95002e-06
[py_interpret_to_execute]: 4.356e-05
[rewriter_before_opt_a]: 0.00011219
[expand_dump_flag]: 3.36001e-06
[jit_opt_a]: 0.0144898, [2]
    [Cycle 1]: 0.0102925, [27]
        [switch_simplify]: 7.686e-05
        [loop_unroll]: 4.25e-05
        [a_1]: 0.00099876
        [with_stream_mark]: 3.094e-05
        [recompute_prepare]: 2.115e-05
        [updatestate_depend_eliminate]: 7.66001e-06
        [updatestate_assign_eliminate]: 6.46e-06
        [updatestate_loads_eliminate]: 5.69e-06
        [parameter_eliminate]: 2.77002e-06
        [specialize_transform]: 1.451e-05
        [updatestate_useless_node_eliminater]: 1.395e-05
        [accelerated_algorithm]: 1.349e-05
        [meta_shard_fg_expand]: 3.41999e-06
        [get_grad_eliminate_]: 1.368e-05
        [merge_forward]: 8.33001e-06
        [cell_reuse_recompute_pass]: 1.43002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.761e-05
        [j_node_and_user_rematch]: 2.302e-05
        [meta_fg_expand]: 0.00098413
        [replace_old_param]: 4.233e-05
        [inline_without_move]: 3.54e-05
        [renormalize]: 0.0076556
        [add_forward_monad_depend]: 6.24001e-06
        [auto_monad_grad]: 2.05002e-06
        [auto_monad_eliminator]: 9.72999e-06
        [cse]: 1.764e-05
        [replace_applicator]: 1.588e-05
    [Cycle 2]: 0.00034207, [27]
        [switch_simplify]: 4.96002e-06
        [loop_unroll]: 4.68999e-06
        [a_1]: 5.279e-05
        [with_stream_mark]: 1.11e-05
        [recompute_prepare]: 3.97e-06
        [updatestate_depend_eliminate]: 2.27999e-06
        [updatestate_assign_eliminate]: 1.93002e-06
        [updatestate_loads_eliminate]: 1.49e-06
        [parameter_eliminate]: 1.30999e-06
        [specialize_transform]: 4.01001e-06
        [updatestate_useless_node_eliminater]: 3.8e-06
        [accelerated_algorithm]: 3.90998e-06
        [meta_shard_fg_expand]: 1.77999e-06
        [get_grad_eliminate_]: 3.85998e-06
        [merge_forward]: 2.76e-06
        [cell_reuse_recompute_pass]: 3.7e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.114e-05
        [j_node_and_user_rematch]: 6.22001e-06
        [meta_fg_expand]: 7.057e-05
        [replace_old_param]: 4.4e-06
        [inline_without_move]: 3.40998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.35999e-06
        [auto_monad_grad]: 1.31002e-06
        [auto_monad_eliminator]: 4.4e-06
        [cse]: 9.50001e-06
        [replace_applicator]: 3.55e-06
[py_interpret_to_execute_after_opt_a]: 9.25001e-06
[rewriter_after_opt_a]: 2.957e-05
[convert_after_rewriter]: 5.06002e-06
[order_py_execute_after_rewriter]: 4.58001e-06
[mutable_eliminate]: 0.00060078
[jit_opt_b]: 4.123e-05, [1]
    [Cycle 1]: 3.466e-05, [2]
        [frontend_op_eliminate]: 1.132e-05
        [inline_after_opt_a]: 1.203e-05
[cconv]: 1.899e-05
[loop_unroll]: 0.00041996
[jit_opt_after_cconv]: 0.00011644, [1]
    [Cycle 1]: 0.00011023, [11]
        [c_1]: 1.168e-05
        [parameter_eliminate]: 2.90998e-06
        [updatestate_depend_eliminate]: 4.43001e-06
        [updatestate_assign_eliminate]: 1.70001e-06
        [updatestate_loads_eliminate]: 1.54e-06
        [cse]: 1.601e-05
        [call_graph_tuple_transform]: 1.293e-05
        [tuple_list_get_item_eliminator]: 3.79002e-06
        [none_parameter_eliminate]: 1.12999e-06
        [renormalize]: 6.69999e-07
        [switch_simplify]: 3.43999e-06
[remove_dup_value]: 9.76e-06
[partial_unused_args_eliminate]: 1.62999e-06
[environ_conv]: 3.47002e-06
[add_recomputation]: 3.161e-05
[cse_after_recomputation]: 1.956e-05, [1]
    [Cycle 1]: 1.4e-05, [1]
        [cse]: 8.13999e-06
[auto_monad_reorder]: 9.20999e-06
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 4.89e-06
[opt_after_jit_grad]: 0.00043738
[symbol_engine_optimizer]: 6.301e-05, [1]
    [Cycle 1]: 5.716e-05, [6]
        [build]: 2.42001e-06
        [elim_shapecalc]: 5.51e-06
        [elim_not_effective]: 9.20999e-06
        [opt_reshape]: 5.00001e-06
        [fold_const_symbol]: 5.70001e-06
        [renormalize]: 4.89992e-07
[validate]: 4.414e-05
[backend_pass]: 6.69999e-07
[task_emit]: 2.128e-05
[execute]: 1.29e-06
Sums
    bootstrap                            :     0.000263s :  0.28%
    type_inference                       :     0.081599s : 86.33%
    event_method                         :     0.000287s :  0.30%
    auto_monad                           :     0.000122s :  0.13%
    graph_reusing                        :     0.000007s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000044s :  0.05%
    rewriter_before_opt_a                :     0.000112s :  0.12%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000082s :  0.09%
    jit_opt_a.loop_unroll                :     0.000047s :  0.05%
    jit_opt_a.a_1                        :     0.001052s :  1.11%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.04%
    jit_opt_a.recompute_prepare          :     0.000025s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.00%
    jit_opt_a.specialize_transform       :     0.000019s :  0.02%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000018s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.02%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000018s :  0.02%
    jit_opt_a.merge_forward              :     0.000011s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.04%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.03%
    jit_opt_a.meta_fg_expand             :     0.001055s :  1.12%
    jit_opt_a.replace_old_param          :     0.000047s :  0.05%
    jit_opt_a.inline_without_move        :     0.000039s :  0.04%
    jit_opt_a.renormalize                :     0.007656s :  8.10%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.01%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.00%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.01%
    jit_opt_a.cse                        :     0.000027s :  0.03%
    jit_opt_a.replace_applicator         :     0.000019s :  0.02%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.01%
    rewriter_after_opt_a                 :     0.000030s :  0.03%
    convert_after_rewriter               :     0.000005s :  0.01%
    order_py_execute_after_rewriter      :     0.000005s :  0.00%
    mutable_eliminate                    :     0.000601s :  0.64%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.01%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.01%
    cconv                                :     0.000019s :  0.02%
    loop_unroll                          :     0.000420s :  0.44%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.01%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.00%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000016s :  0.02%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.01%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.00%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000003s :  0.00%
    remove_dup_value                     :     0.000010s :  0.01%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000003s :  0.00%
    add_recomputation                    :     0.000032s :  0.03%
    cse_after_recomputation.cse          :     0.000008s :  0.01%
    auto_monad_reorder                   :     0.000009s :  0.01%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000437s :  0.46%
    symbol_engine_optimizer.build        :     0.000002s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.01%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000044s :  0.05%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000021s :  0.02%
    execute                              :     0.000001s :  0.00%


Time group info:
------[substitution.]   0.000261    47
 0.56% :     0.000001s :      1: substitution.elim_not_effective
 0.27% :     0.000001s :      1: substitution.fold_const_symbol
 1.36% :     0.000004s :      1: substitution.graph_param_transform
66.82% :     0.000175s :      6: substitution.inline
 4.87% :     0.000013s :      1: substitution.inline_without_move
 2.62% :     0.000007s :      8: substitution.j_node_and_user_rematch
 1.51% :     0.000004s :      2: substitution.minmaximum_grad
 1.66% :     0.000004s :      3: substitution.partial_eliminate
 2.76% :     0.000007s :      8: substitution.remove_not_recompute_node
 2.32% :     0.000006s :      2: substitution.replace_applicator
 1.34% :     0.000004s :      4: substitution.replace_old_param
 1.02% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.34% :     0.000006s :      1: substitution.tile_eliminate
 2.66% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.77% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 6.11% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.081490     2
97.06% :     0.079091s :      1: type_inference.infer
 2.94% :     0.002399s :      1: type_inference.specialize
------[replace.]   0.000069     9
71.29% :     0.000049s :      6: replace.inline
 8.26% :     0.000006s :      1: replace.tile_eliminate
20.46% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000188     9
90.89% :     0.000171s :      6: match.inline
 2.94% :     0.000006s :      1: match.tile_eliminate
 6.17% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000187  1210
 1.56% :     0.000003s :     20: predicate.accumulaten_eliminater
 0.52% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.26% :     0.000002s :     20: predicate.addn_check_dump
 1.77% :     0.000003s :     20: predicate.addn_zero_filter
 1.91% :     0.000004s :     20: predicate.arithmetic_simplify
 1.51% :     0.000003s :     21: predicate.cast_eliminate
 0.13% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.18% :     0.000002s :     20: predicate.compare_switch_simplify
 1.60% :     0.000003s :     20: predicate.depend_value_elim
 1.56% :     0.000003s :     21: predicate.dict_get_item_const_eliminator
 1.58% :     0.000003s :     21: predicate.dict_get_item_eliminator
 1.34% :     0.000003s :     21: predicate.dict_set_item_eliminator
 0.44% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      1: predicate.elim_not_effective
 0.18% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.55% :     0.000003s :     21: predicate.environ_add_const_eliminate
 1.32% :     0.000002s :     21: predicate.environ_get_add_eliminate
 1.33% :     0.000002s :     21: predicate.environ_get_depend_swap
 1.43% :     0.000003s :     21: predicate.environ_get_eliminate
 1.32% :     0.000002s :     21: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.17% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.51% :     0.000008s :     31: predicate.inline
 1.85% :     0.000003s :     19: predicate.inline_without_move
 0.51% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :     12: predicate.less_batch_normalization
 1.77% :     0.000003s :     23: predicate.list_to_tuple_eliminator_
 1.74% :     0.000003s :     24: predicate.load_eliminater
 0.77% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.74% :     0.000007s :     50: predicate.loop_unroll_before_grad
 1.88% :     0.000004s :     22: predicate.make_slice_get_slice_eliminator
 1.26% :     0.000002s :     20: predicate.merge_addn
 1.30% :     0.000002s :     20: predicate.minmaximum_grad
 0.78% :     0.000001s :      1: predicate.mutable_eliminate
 0.19% :     0.000000s :      1: predicate.opt_reshape
 2.14% :     0.000004s :     24: predicate.partial_eliminate
 1.36% :     0.000003s :     20: predicate.print_const_string_wrapper
 1.96% :     0.000004s :     21: predicate.reduce_eliminate
 1.70% :     0.000003s :     23: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.65% :     0.000003s :     25: predicate.replace_applicator
 1.04% :     0.000002s :     19: predicate.replace_old_param
 0.07% :     0.000000s :      1: predicate.reset_defer_inline
 2.00% :     0.000004s :     21: predicate.reshape_eliminate
 1.49% :     0.000003s :     20: predicate.row_tensor_add_zeros_like
 0.44% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.42% :     0.000003s :     20: predicate.same_eliminate
 0.65% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.54% :     0.000001s :      2: predicate.special_op_eliminate
 1.17% :     0.000002s :     12: predicate.specialize_transform
 1.66% :     0.000003s :     20: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000003s :     20: predicate.stack_unstack_eliminate
 0.11% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.47% :     0.000005s :     29: predicate.switch_defer_inline
 2.30% :     0.000004s :     29: predicate.switch_layer_defer_inline
 6.72% :     0.000013s :     80: predicate.switch_simplify
 1.65% :     0.000003s :     21: predicate.tile_eliminate
 1.31% :     0.000002s :     20: predicate.transpose_eliminate
 1.87% :     0.000003s :     21: predicate.tuple_list_convert_item_index_to_positive
 1.58% :     0.000003s :     21: predicate.tuple_list_get_item_depend_reorder
 3.50% :     0.000007s :     25: predicate.tuple_list_get_item_eliminator
 2.22% :     0.000004s :     21: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000003s :     23: predicate.tuple_to_list_eliminator_
 1.55% :     0.000003s :     24: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000006s :     36: predicate.updatestate_useless_node_eliminater
 1.86% :     0.000003s :     20: predicate.value_based_eliminate
 0.11% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.26% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.002670    35
84.02% :     0.002244s :     26: func_graph_cloner_run.FuncGraphClonerGraph
15.98% :     0.000427s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.108150    76
 0.03% :     0.000035s :      1: add_recomputation
 0.12% :     0.000129s :      1: auto_monad
 0.01% :     0.000012s :      1: auto_monad_reorder
 0.00% :     0.000003s :      1: backend_pass
 0.27% :     0.000287s :      1: bootstrap
 0.02% :     0.000022s :      1: cconv
 0.01% :     0.000007s :      1: convert_after_rewriter
 0.02% :     0.000023s :      1: cse_after_recomputation
 0.01% :     0.000006s :      1: environ_conv
 0.27% :     0.000294s :      1: event_method
 0.00% :     0.000004s :      1: execute
 0.00% :     0.000005s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000010s :      1: graph_reusing
13.40% :     0.014494s :      1: jit_opt_a
 0.11% :     0.000119s :      1: jit_opt_after_cconv
 0.04% :     0.000044s :      1: jit_opt_b
 0.40% :     0.000427s :      1: loop_unroll
 0.56% :     0.000609s :      1: mutable_eliminate
 1.31% :     0.001417s :     26: opt.transform.jit_opt_a
 0.03% :     0.000029s :      4: opt.transform.jit_opt_after_cconv
 0.02% :     0.000017s :      4: opt.transform.jit_opt_b
 0.01% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.01% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.01% :     0.000013s :      1: opt.transform.opt_after_jit_grad
 0.02% :     0.000022s :      4: opt.transform.symbol_engine_opt
 0.41% :     0.000445s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.00% :     0.000005s :      1: pre_auto_parallel
 0.04% :     0.000047s :      1: py_interpret_to_execute
 0.01% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.01% :     0.000013s :      1: remove_dup_value
 6.58% :     0.007114s :      1: renormalize.infer
 0.49% :     0.000532s :      1: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.03% :     0.000032s :      1: rewriter_after_opt_a
 0.11% :     0.000117s :      1: rewriter_before_opt_a
 0.06% :     0.000065s :      1: symbol_engine_optimizer
 0.02% :     0.000025s :      1: task_emit
75.46% :     0.081615s :      1: type_inference
 0.05% :     0.000059s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:20.612.375 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c371ca0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:20.613.092 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0138331, [30]
[bootstrap]: 0.00025865
[type_inference]: 0.00577626
[event_method]: 1.8e-05
[auto_monad]: 4.66e-05
[graph_reusing]: 3.65998e-06
[pre_auto_parallel]: 2.29999e-06
[py_interpret_to_execute]: 4.068e-05
[rewriter_before_opt_a]: 6.964e-05
[expand_dump_flag]: 2.05002e-06
[jit_opt_a]: 0.00542396, [2]
    [Cycle 1]: 0.00161403, [27]
        [switch_simplify]: 5.038e-05
        [loop_unroll]: 3.091e-05
        [a_1]: 0.00052694
        [with_stream_mark]: 1.346e-05
        [recompute_prepare]: 7.43999e-06
        [updatestate_depend_eliminate]: 2.93998e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 6.56e-06
        [updatestate_useless_node_eliminater]: 5.08002e-06
        [accelerated_algorithm]: 5.71998e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 5.86e-06
        [merge_forward]: 2.66e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.206e-05
        [j_node_and_user_rematch]: 9.36998e-06
        [meta_fg_expand]: 2.16e-06
        [replace_old_param]: 8.55001e-06
        [inline_without_move]: 5.69e-06
        [renormalize]: 0.00070293
        [add_forward_monad_depend]: 4.31002e-06
        [auto_monad_grad]: 1.25001e-06
        [auto_monad_eliminator]: 9.41e-06
        [cse]: 1.677e-05
        [replace_applicator]: 1.145e-05
    [Cycle 2]: 0.00033868, [27]
        [switch_simplify]: 5.86998e-06
        [loop_unroll]: 5.32999e-06
        [a_1]: 9.573e-05
        [with_stream_mark]: 8.77e-06
        [recompute_prepare]: 5.10999e-06
        [updatestate_depend_eliminate]: 2.87002e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 9.60019e-07
        [specialize_transform]: 5.54e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.57999e-06
        [meta_shard_fg_expand]: 1.17e-06
        [get_grad_eliminate_]: 4.68999e-06
        [merge_forward]: 2.59001e-06
        [cell_reuse_recompute_pass]: 1.57999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.27e-05
        [j_node_and_user_rematch]: 8.28999e-06
        [meta_fg_expand]: 1.72999e-06
        [replace_old_param]: 7.63001e-06
        [inline_without_move]: 5.24998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.47999e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 5.50001e-06
        [cse]: 1.011e-05
        [replace_applicator]: 5.81e-06
[py_interpret_to_execute_after_opt_a]: 9.31998e-06
[rewriter_after_opt_a]: 2.627e-05
[convert_after_rewriter]: 5.87999e-06
[order_py_execute_after_rewriter]: 5.57999e-06
[mutable_eliminate]: 0.0005801
[jit_opt_b]: 5.542e-05, [1]
    [Cycle 1]: 4.852e-05, [2]
        [frontend_op_eliminate]: 1.955e-05
        [inline_after_opt_a]: 1.632e-05
[cconv]: 1.879e-05
[loop_unroll]: 0.00043377
[jit_opt_after_cconv]: 0.0001434, [1]
    [Cycle 1]: 0.00013638, [11]
        [c_1]: 2.345e-05
        [parameter_eliminate]: 2.47001e-06
        [updatestate_depend_eliminate]: 5.88002e-06
        [updatestate_assign_eliminate]: 2.41998e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.771e-05
        [call_graph_tuple_transform]: 1.839e-05
        [tuple_list_get_item_eliminator]: 5.87999e-06
        [none_parameter_eliminate]: 8.00006e-07
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.61e-06
[remove_dup_value]: 9.14e-06
[partial_unused_args_eliminate]: 2.71e-06
[environ_conv]: 3.37002e-06
[add_recomputation]: 3.207e-05
[cse_after_recomputation]: 2.118e-05, [1]
    [Cycle 1]: 1.578e-05, [1]
        [cse]: 9.82001e-06
[auto_monad_reorder]: 9.69e-06
[get_jit_bprop_graph]: 1.39e-06
[rewriter_after_jit_bprop_graph]: 5.81e-06
[opt_after_jit_grad]: 0.00052348
[symbol_engine_optimizer]: 8.466e-05, [1]
    [Cycle 1]: 7.674e-05, [6]
        [build]: 3.21001e-06
        [elim_shapecalc]: 9.72001e-06
        [elim_not_effective]: 1.41e-05
        [opt_reshape]: 6.48e-06
        [fold_const_symbol]: 1.08e-05
        [renormalize]: 8.50006e-07
[validate]: 3.503e-05
Sums
    bootstrap                            :     0.000259s :  2.65%
    type_inference                       :     0.005776s : 59.18%
    event_method                         :     0.000018s :  0.18%
    auto_monad                           :     0.000047s :  0.48%
    graph_reusing                        :     0.000004s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000041s :  0.42%
    rewriter_before_opt_a                :     0.000070s :  0.71%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.58%
    jit_opt_a.loop_unroll                :     0.000036s :  0.37%
    jit_opt_a.a_1                        :     0.000623s :  6.38%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.17%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000703s :  7.20%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.15%
    jit_opt_a.cse                        :     0.000027s :  0.28%
    jit_opt_a.replace_applicator         :     0.000017s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.10%
    rewriter_after_opt_a                 :     0.000026s :  0.27%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000580s :  5.94%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.17%
    cconv                                :     0.000019s :  0.19%
    loop_unroll                          :     0.000434s :  4.44%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.18%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000009s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000003s :  0.03%
    add_recomputation                    :     0.000032s :  0.33%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000010s :  0.10%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000523s :  5.36%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000035s :  0.36%


Time group info:
------[substitution.]   0.000183    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.91% :     0.000002s :      2: substitution.fold_const_symbol
 2.21% :     0.000004s :      3: substitution.graph_param_transform
85.64% :     0.000157s :      4: substitution.inline
 1.66% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.37% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.74% :     0.000003s :      2: substitution.replace_old_param
 4.63% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005716     2
83.16% :     0.004753s :      1: type_inference.infer
16.84% :     0.000963s :      1: type_inference.specialize
------[replace.]   0.000068     6
66.02% :     0.000045s :      4: replace.inline
33.98% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000161     6
95.50% :     0.000154s :      4: match.inline
 4.50% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.01% :     0.000002s :     10: predicate.arithmetic_simplify
 1.73% :     0.000002s :     10: predicate.cast_eliminate
 0.65% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.49% :     0.000002s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.92% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.42% :     0.000007s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.69% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.08% :     0.000003s :     15: predicate.load_eliminater
 1.28% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.93% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.74% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.68% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.74% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.86% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.46% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.30% :     0.000009s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.62% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.31% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001367    12
72.68% :     0.000993s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 2.71% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
24.61% :     0.000336s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015430    72
 0.23% :     0.000035s :      1: add_recomputation
 0.33% :     0.000050s :      1: auto_monad
 0.08% :     0.000012s :      1: auto_monad_reorder
 1.76% :     0.000271s :      1: bootstrap
 0.14% :     0.000021s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.14% :     0.000021s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
35.18% :     0.005428s :      1: jit_opt_a
 0.95% :     0.000146s :      1: jit_opt_after_cconv
 0.38% :     0.000059s :      1: jit_opt_b
 2.86% :     0.000442s :      1: loop_unroll
 3.83% :     0.000590s :      1: mutable_eliminate
 5.40% :     0.000833s :     26: opt.transform.jit_opt_a
 0.32% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000029s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.45% :     0.000533s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.28% :     0.000044s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000012s :      1: remove_dup_value
 2.32% :     0.000357s :      1: renormalize.infer
 2.19% :     0.000338s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000031s :      1: rewriter_after_opt_a
 0.47% :     0.000073s :      1: rewriter_before_opt_a
 0.57% :     0.000087s :      1: symbol_engine_optimizer
37.52% :     0.005789s :      1: type_inference


TotalTime = 0.0185818, [33]
[bootstrap]: 0.00026183
[type_inference]: 0.00450285
[event_method]: 1.856e-05
[auto_monad]: 6.488e-05
[graph_reusing]: 5.94e-06
[pre_auto_parallel]: 2.45002e-06
[py_interpret_to_execute]: 3.452e-05
[rewriter_before_opt_a]: 7.239e-05
[expand_dump_flag]: 3.44001e-06
[jit_opt_a]: 0.00511518, [2]
    [Cycle 1]: 0.0014774, [27]
        [switch_simplify]: 6.585e-05
        [loop_unroll]: 2.925e-05
        [a_1]: 0.00048914
        [with_stream_mark]: 1.693e-05
        [recompute_prepare]: 6.61e-06
        [updatestate_depend_eliminate]: 3.36001e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 1.95001e-06
        [specialize_transform]: 5.92001e-06
        [updatestate_useless_node_eliminater]: 5.32001e-06
        [accelerated_algorithm]: 5.03002e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 5.05001e-06
        [merge_forward]: 3.38999e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.377e-05
        [j_node_and_user_rematch]: 8.63001e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 8.82e-06
        [inline_without_move]: 5.02999e-06
        [renormalize]: 0.00056734
        [add_forward_monad_depend]: 5.14998e-06
        [auto_monad_grad]: 2.44001e-06
        [auto_monad_eliminator]: 1.346e-05
        [cse]: 2.55e-05
        [replace_applicator]: 1.11e-05
    [Cycle 2]: 0.0003094, [27]
        [switch_simplify]: 5.49998e-06
        [loop_unroll]: 4.99e-06
        [a_1]: 8.735e-05
        [with_stream_mark]: 8.45999e-06
        [recompute_prepare]: 4.74e-06
        [updatestate_depend_eliminate]: 2.47001e-06
        [updatestate_assign_eliminate]: 2.05002e-06
        [updatestate_loads_eliminate]: 1.83002e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 4.68999e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 4.56002e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 2.06e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.148e-05
        [j_node_and_user_rematch]: 7.5e-06
        [meta_fg_expand]: 1.55001e-06
        [replace_old_param]: 6.79001e-06
        [inline_without_move]: 4.80001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 4.72e-06
        [cse]: 9.69e-06
        [replace_applicator]: 4.80001e-06
[py_interpret_to_execute_after_opt_a]: 9.96998e-06
[rewriter_after_opt_a]: 3.345e-05
[convert_after_rewriter]: 6.93998e-06
[order_py_execute_after_rewriter]: 4.67e-06
[mutable_eliminate]: 0.00054802
[jit_opt_b]: 5.219e-05, [1]
    [Cycle 1]: 4.468e-05, [2]
        [frontend_op_eliminate]: 1.754e-05
        [inline_after_opt_a]: 1.484e-05
[cconv]: 2.375e-05
[loop_unroll]: 0.00042956
[jit_opt_after_cconv]: 0.00013841, [1]
    [Cycle 1]: 0.00013221, [11]
        [c_1]: 1.986e-05
        [parameter_eliminate]: 2.78e-06
        [updatestate_depend_eliminate]: 6.68998e-06
        [updatestate_assign_eliminate]: 2.64001e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [cse]: 1.95e-05
        [call_graph_tuple_transform]: 1.79e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.59998e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 5.24998e-06
[remove_dup_value]: 1.517e-05
[partial_unused_args_eliminate]: 2.15002e-06
[environ_conv]: 5.12999e-06
[add_recomputation]: 4.704e-05
[cse_after_recomputation]: 1.906e-05, [1]
    [Cycle 1]: 1.363e-05, [1]
        [cse]: 8.06001e-06
[auto_monad_reorder]: 1.41e-05
[get_jit_bprop_graph]: 2.01e-06
[rewriter_after_jit_bprop_graph]: 5.02999e-06
[opt_after_jit_grad]: 0.00044602
[symbol_engine_optimizer]: 6.82e-05, [1]
    [Cycle 1]: 6.205e-05, [6]
        [build]: 3.48e-06
        [elim_shapecalc]: 6.66999e-06
        [elim_not_effective]: 1.174e-05
        [opt_reshape]: 5.23002e-06
        [fold_const_symbol]: 8.27e-06
        [renormalize]: 3.80009e-07
[validate]: 3.538e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.0063792
[execute]: 8.83001e-06
Sums
    bootstrap                            :     0.000262s :  1.79%
    type_inference                       :     0.004503s : 30.73%
    event_method                         :     0.000019s :  0.13%
    auto_monad                           :     0.000065s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000035s :  0.24%
    rewriter_before_opt_a                :     0.000072s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000071s :  0.49%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.000576s :  3.93%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000567s :  3.87%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000035s :  0.24%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000548s :  3.74%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000024s :  0.16%
    loop_unroll                          :     0.000430s :  2.93%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000015s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000047s :  0.32%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000446s :  3.04%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006379s : 43.54%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000178    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000001s :      2: substitution.fold_const_symbol
 2.97% :     0.000005s :      3: substitution.graph_param_transform
82.83% :     0.000147s :      4: substitution.inline
 1.49% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.53% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.23% :     0.000004s :      2: substitution.replace_old_param
 6.13% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004436     2
80.82% :     0.003585s :      1: type_inference.infer
19.18% :     0.000851s :      1: type_inference.specialize
------[replace.]   0.000061     6
67.39% :     0.000041s :      4: replace.inline
32.61% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000155     6
93.59% :     0.000145s :      4: match.inline
 6.41% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.28% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.36% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000001s :     10: predicate.addn_zero_filter
 2.08% :     0.000002s :     10: predicate.arithmetic_simplify
 1.31% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000001s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.01% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.70% :     0.000006s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.64% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.39% :     0.000003s :     15: predicate.load_eliminater
 1.70% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.17% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 2.27% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000000s :      3: predicate.opt_reshape
 2.34% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.28% :     0.000001s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.16% :     0.000001s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.90% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.52% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.00% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000529    12
51.24% :     0.000271s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.06% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.71% :     0.000231s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020041    76
 0.25% :     0.000050s :      1: add_recomputation
 0.34% :     0.000069s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.39% :     0.000278s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000022s :      1: event_method
 0.07% :     0.000013s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.54% :     0.005118s :      1: jit_opt_a
 0.71% :     0.000141s :      1: jit_opt_after_cconv
 0.28% :     0.000055s :      1: jit_opt_b
 2.18% :     0.000438s :      1: loop_unroll
 2.79% :     0.000559s :      1: mutable_eliminate
 3.94% :     0.000789s :     26: opt.transform.jit_opt_a
 0.22% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.26% :     0.000454s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000038s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000017s :      1: remove_dup_value
 1.44% :     0.000288s :      1: renormalize.infer
 1.36% :     0.000272s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000036s :      1: rewriter_after_opt_a
 0.38% :     0.000075s :      1: rewriter_before_opt_a
 0.35% :     0.000071s :      1: symbol_engine_optimizer
31.92% :     0.006396s :      1: task_emit
22.57% :     0.004522s :      1: type_inference
 0.28% :     0.000056s :      1: validate


TotalTime = 0.011844, [30]
[bootstrap]: 0.00027791
[type_inference]: 0.00421212
[event_method]: 1.663e-05
[auto_monad]: 8.777e-05
[graph_reusing]: 6.21e-06
[pre_auto_parallel]: 2.38002e-06
[py_interpret_to_execute]: 3.483e-05
[rewriter_before_opt_a]: 7.699e-05
[expand_dump_flag]: 3.56001e-06
[jit_opt_a]: 0.00503191, [2]
    [Cycle 1]: 0.00151472, [27]
        [switch_simplify]: 5.121e-05
        [loop_unroll]: 3.098e-05
        [a_1]: 0.00052829
        [with_stream_mark]: 1.714e-05
        [recompute_prepare]: 7.88999e-06
        [updatestate_depend_eliminate]: 3.63e-06
        [updatestate_assign_eliminate]: 3.01999e-06
        [updatestate_loads_eliminate]: 2.74001e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 6.13998e-06
        [updatestate_useless_node_eliminater]: 5.31002e-06
        [accelerated_algorithm]: 5.40001e-06
        [meta_shard_fg_expand]: 1.92001e-06
        [get_grad_eliminate_]: 5.02e-06
        [merge_forward]: 3.85e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.424e-05
        [j_node_and_user_rematch]: 8.89003e-06
        [meta_fg_expand]: 2.59001e-06
        [replace_old_param]: 8.43001e-06
        [inline_without_move]: 5.52001e-06
        [renormalize]: 0.00050363
        [add_forward_monad_depend]: 5.65001e-06
        [auto_monad_grad]: 2.36e-06
        [auto_monad_eliminator]: 1.5e-05
        [cse]: 2.882e-05
        [replace_applicator]: 1.219e-05
    [Cycle 2]: 0.00033846, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 4.89998e-06
        [a_1]: 9.494e-05
        [with_stream_mark]: 8.40001e-06
        [recompute_prepare]: 5.37001e-06
        [updatestate_depend_eliminate]: 2.71e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 2.16003e-06
        [parameter_eliminate]: 1.19998e-06
        [specialize_transform]: 5.79999e-06
        [updatestate_useless_node_eliminater]: 5.77001e-06
        [accelerated_algorithm]: 6.19001e-06
        [meta_shard_fg_expand]: 1.35999e-06
        [get_grad_eliminate_]: 4.59002e-06
        [merge_forward]: 2.45002e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.224e-05
        [j_node_and_user_rematch]: 8.24002e-06
        [meta_fg_expand]: 1.52999e-06
        [replace_old_param]: 7.16999e-06
        [inline_without_move]: 4.65001e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 8.80013e-07
        [auto_monad_eliminator]: 5.87001e-06
        [cse]: 1.145e-05
        [replace_applicator]: 6.31998e-06
[py_interpret_to_execute_after_opt_a]: 9.04e-06
[rewriter_after_opt_a]: 3.149e-05
[convert_after_rewriter]: 7.06999e-06
[order_py_execute_after_rewriter]: 5.49e-06
[mutable_eliminate]: 0.00053316
[jit_opt_b]: 5.18e-05, [1]
    [Cycle 1]: 4.501e-05, [2]
        [frontend_op_eliminate]: 1.723e-05
        [inline_after_opt_a]: 1.601e-05
[cconv]: 2.659e-05
[loop_unroll]: 0.00043776
[jit_opt_after_cconv]: 0.00014262, [1]
    [Cycle 1]: 0.00013635, [11]
        [c_1]: 2.207e-05
        [parameter_eliminate]: 2.68e-06
        [updatestate_depend_eliminate]: 6.01998e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [cse]: 1.838e-05
        [call_graph_tuple_transform]: 1.902e-05
        [tuple_list_get_item_eliminator]: 5.55001e-06
        [none_parameter_eliminate]: 1.48002e-06
        [renormalize]: 5.39992e-07
        [switch_simplify]: 5.71e-06
[remove_dup_value]: 1.503e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 5.72999e-06
[add_recomputation]: 4.641e-05
[cse_after_recomputation]: 2.064e-05, [1]
    [Cycle 1]: 1.527e-05, [1]
        [cse]: 9.52001e-06
[auto_monad_reorder]: 1.501e-05
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 5.02e-06
[opt_after_jit_grad]: 0.00045329
[symbol_engine_optimizer]: 8.398e-05, [1]
    [Cycle 1]: 6.737e-05, [6]
        [build]: 3.50998e-06
        [elim_shapecalc]: 7.28e-06
        [elim_not_effective]: 1.293e-05
        [opt_reshape]: 6.03002e-06
        [fold_const_symbol]: 8.84998e-06
        [renormalize]: 4.39992e-07
[validate]: 3.319e-05
Sums
    bootstrap                            :     0.000278s :  3.47%
    type_inference                       :     0.004212s : 52.57%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000088s :  1.10%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.43%
    rewriter_before_opt_a                :     0.000077s :  0.96%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000057s :  0.71%
    jit_opt_a.loop_unroll                :     0.000036s :  0.45%
    jit_opt_a.a_1                        :     0.000623s :  7.78%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000504s :  6.29%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.26%
    jit_opt_a.cse                        :     0.000040s :  0.50%
    jit_opt_a.replace_applicator         :     0.000019s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000031s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000533s :  6.65%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.20%
    cconv                                :     0.000027s :  0.33%
    loop_unroll                          :     0.000438s :  5.46%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000018s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000015s :  0.19%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000046s :  0.58%
    cse_after_recomputation.cse          :     0.000010s :  0.12%
    auto_monad_reorder                   :     0.000015s :  0.19%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000453s :  5.66%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000033s :  0.41%


Time group info:
------[substitution.]   0.000187    23
 1.10% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.79% :     0.000005s :      3: substitution.graph_param_transform
83.03% :     0.000155s :      4: substitution.inline
 1.52% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.33% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.09% :     0.000004s :      2: substitution.replace_old_param
 6.40% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004152     2
81.12% :     0.003368s :      1: type_inference.infer
18.88% :     0.000784s :      1: type_inference.specialize
------[replace.]   0.000066     6
66.70% :     0.000044s :      4: replace.inline
33.30% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000164     6
93.37% :     0.000153s :      4: match.inline
 6.63% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.53% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.05% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.55% :     0.000002s :     10: predicate.addn_zero_filter
 2.17% :     0.000003s :     10: predicate.arithmetic_simplify
 1.13% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.51% :     0.000007s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.25% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.77% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000002s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.94% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     15: predicate.partial_eliminate
 1.53% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.93% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 1.14% :     0.000001s :      6: predicate.specialize_transform
 1.51% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.57% :     0.000003s :     16: predicate.switch_defer_inline
 1.93% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.23% :     0.000009s :     53: predicate.switch_simplify
 1.59% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.71% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000537    12
52.08% :     0.000280s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.32% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.60% :     0.000229s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013245    72
 0.37% :     0.000049s :      1: add_recomputation
 0.69% :     0.000092s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 2.23% :     0.000296s :      1: bootstrap
 0.22% :     0.000029s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000021s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.01% :     0.005034s :      1: jit_opt_a
 1.10% :     0.000146s :      1: jit_opt_after_cconv
 0.41% :     0.000054s :      1: jit_opt_b
 3.37% :     0.000447s :      1: loop_unroll
 4.09% :     0.000542s :      1: mutable_eliminate
 6.30% :     0.000834s :     26: opt.transform.jit_opt_a
 0.36% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000027s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.48% :     0.000461s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.28% :     0.000038s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000018s :      1: remove_dup_value
 1.83% :     0.000242s :      1: renormalize.infer
 1.92% :     0.000254s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000034s :      1: rewriter_after_opt_a
 0.60% :     0.000080s :      1: rewriter_before_opt_a
 0.65% :     0.000087s :      1: symbol_engine_optimizer
31.89% :     0.004224s :      1: type_inference


TotalTime = 0.0170665, [33]
[bootstrap]: 0.00025798
[type_inference]: 0.00394879
[event_method]: 1.667e-05
[auto_monad]: 6.151e-05
[graph_reusing]: 5.48002e-06
[pre_auto_parallel]: 2.43998e-06
[py_interpret_to_execute]: 2.885e-05
[rewriter_before_opt_a]: 6.904e-05
[expand_dump_flag]: 2.87002e-06
[jit_opt_a]: 0.00483168, [2]
    [Cycle 1]: 0.00146731, [27]
        [switch_simplify]: 5.02e-05
        [loop_unroll]: 2.873e-05
        [a_1]: 0.00049349
        [with_stream_mark]: 1.695e-05
        [recompute_prepare]: 6.94999e-06
        [updatestate_depend_eliminate]: 3.74002e-06
        [updatestate_assign_eliminate]: 2.96999e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 1.71e-06
        [specialize_transform]: 5.49e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 1.91998e-06
        [get_grad_eliminate_]: 4.83001e-06
        [merge_forward]: 3.31999e-06
        [cell_reuse_recompute_pass]: 9.79984e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.348e-05
        [j_node_and_user_rematch]: 8.29002e-06
        [meta_fg_expand]: 2.06003e-06
        [replace_old_param]: 8.42e-06
        [inline_without_move]: 5.14998e-06
        [renormalize]: 0.00058053
        [add_forward_monad_depend]: 5.07e-06
        [auto_monad_grad]: 2.18002e-06
        [auto_monad_eliminator]: 1.271e-05
        [cse]: 2.416e-05
        [replace_applicator]: 1.084e-05
    [Cycle 2]: 0.00030696, [27]
        [switch_simplify]: 5.61998e-06
        [loop_unroll]: 4.90999e-06
        [a_1]: 8.644e-05
        [with_stream_mark]: 8.97999e-06
        [recompute_prepare]: 4.70001e-06
        [updatestate_depend_eliminate]: 2.47001e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 8.59989e-07
        [specialize_transform]: 4.73001e-06
        [updatestate_useless_node_eliminater]: 4.60999e-06
        [accelerated_algorithm]: 4.83001e-06
        [meta_shard_fg_expand]: 1.28002e-06
        [get_grad_eliminate_]: 4.36002e-06
        [merge_forward]: 2.36998e-06
        [cell_reuse_recompute_pass]: 1.43002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.166e-05
        [j_node_and_user_rematch]: 7.76001e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 7e-06
        [inline_without_move]: 4.45e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 7.29982e-07
        [auto_monad_eliminator]: 4.91997e-06
        [cse]: 9.17999e-06
        [replace_applicator]: 4.85999e-06
[py_interpret_to_execute_after_opt_a]: 8.3e-06
[rewriter_after_opt_a]: 3.158e-05
[convert_after_rewriter]: 6.52001e-06
[order_py_execute_after_rewriter]: 4.91002e-06
[mutable_eliminate]: 0.0004579
[jit_opt_b]: 4.705e-05, [1]
    [Cycle 1]: 4.056e-05, [2]
        [frontend_op_eliminate]: 1.529e-05
        [inline_after_opt_a]: 1.43e-05
[cconv]: 2.338e-05
[loop_unroll]: 0.00039111
[jit_opt_after_cconv]: 0.00012912, [1]
    [Cycle 1]: 0.00012308, [11]
        [c_1]: 1.942e-05
        [parameter_eliminate]: 2.26998e-06
        [updatestate_depend_eliminate]: 5.21998e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.534e-05
        [call_graph_tuple_transform]: 1.669e-05
        [tuple_list_get_item_eliminator]: 4.95001e-06
        [none_parameter_eliminate]: 1.63002e-06
        [renormalize]: 4.49974e-07
        [switch_simplify]: 5.17e-06
[remove_dup_value]: 1.3e-05
[partial_unused_args_eliminate]: 2.22001e-06
[environ_conv]: 4.71002e-06
[add_recomputation]: 4.564e-05
[cse_after_recomputation]: 1.909e-05, [1]
    [Cycle 1]: 1.38e-05, [1]
        [cse]: 8.43999e-06
[auto_monad_reorder]: 1.511e-05
[get_jit_bprop_graph]: 2.06998e-06
[rewriter_after_jit_bprop_graph]: 4.62e-06
[opt_after_jit_grad]: 0.0004285
[symbol_engine_optimizer]: 6.714e-05, [1]
    [Cycle 1]: 6.105e-05, [6]
        [build]: 2.97002e-06
        [elim_shapecalc]: 6.81999e-06
        [elim_not_effective]: 1.153e-05
        [opt_reshape]: 5.17999e-06
        [fold_const_symbol]: 7.96001e-06
        [renormalize]: 3.60014e-07
[validate]: 3.105e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.00592006
[execute]: 7.5e-06
Sums
    bootstrap                            :     0.000258s :  1.92%
    type_inference                       :     0.003949s : 29.37%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000062s :  0.46%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.21%
    rewriter_before_opt_a                :     0.000069s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.42%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000580s :  4.31%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000581s :  4.32%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000033s :  0.25%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000032s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000458s :  3.41%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000391s :  2.91%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000015s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000046s :  0.34%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000429s :  3.19%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005920s : 44.04%
    execute                              :     0.000007s :  0.06%


Time group info:
------[substitution.]   0.000185    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.59% :     0.000005s :      3: substitution.graph_param_transform
83.11% :     0.000154s :      4: substitution.inline
 1.69% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.29% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.16% :     0.000004s :      2: substitution.replace_old_param
 6.46% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003892     2
82.01% :     0.003191s :      1: type_inference.infer
17.99% :     0.000700s :      1: type_inference.specialize
------[replace.]   0.000061     6
68.15% :     0.000041s :      4: replace.inline
31.85% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     6
93.33% :     0.000151s :      4: match.inline
 6.67% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000108   691
 1.35% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000001s :     10: predicate.addn_zero_filter
 1.97% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000001s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.32% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.91% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.29% :     0.000001s :      6: predicate.less_batch_normalization
 1.63% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 1.22% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.67% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000002s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.90% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.60% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.78% :     0.000008s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.31% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.08% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.39% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.80% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000489    12
48.20% :     0.000236s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.54% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.26% :     0.000221s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018519    76
 0.26% :     0.000049s :      1: add_recomputation
 0.35% :     0.000065s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.48% :     0.000275s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.11% :     0.004834s :      1: jit_opt_a
 0.71% :     0.000132s :      1: jit_opt_after_cconv
 0.27% :     0.000050s :      1: jit_opt_b
 2.15% :     0.000398s :      1: loop_unroll
 2.52% :     0.000466s :      1: mutable_eliminate
 4.19% :     0.000775s :     26: opt.transform.jit_opt_a
 0.23% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.35% :     0.000436s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.78% :     0.000330s :      1: renormalize.infer
 1.32% :     0.000244s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000035s :      1: rewriter_after_opt_a
 0.39% :     0.000072s :      1: rewriter_before_opt_a
 0.37% :     0.000069s :      1: symbol_engine_optimizer
32.02% :     0.005930s :      1: task_emit
21.39% :     0.003962s :      1: type_inference
 0.27% :     0.000049s :      1: validate


TotalTime = 0.0111745, [30]
[bootstrap]: 0.00026058
[type_inference]: 0.00398761
[event_method]: 1.638e-05
[auto_monad]: 6.218e-05
[graph_reusing]: 5.60001e-06
[pre_auto_parallel]: 2.29001e-06
[py_interpret_to_execute]: 3.194e-05
[rewriter_before_opt_a]: 7.019e-05
[expand_dump_flag]: 2.53998e-06
[jit_opt_a]: 0.00476382, [2]
    [Cycle 1]: 0.00137276, [27]
        [switch_simplify]: 5e-05
        [loop_unroll]: 2.979e-05
        [a_1]: 0.00050189
        [with_stream_mark]: 1.503e-05
        [recompute_prepare]: 7.03e-06
        [updatestate_depend_eliminate]: 3.54002e-06
        [updatestate_assign_eliminate]: 3.55e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 6.09001e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 5.58997e-06
        [meta_shard_fg_expand]: 2.16e-06
        [get_grad_eliminate_]: 5.26002e-06
        [merge_forward]: 3.21001e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.354e-05
        [j_node_and_user_rematch]: 8.83001e-06
        [meta_fg_expand]: 2.35002e-06
        [replace_old_param]: 8.22e-06
        [inline_without_move]: 5.10999e-06
        [renormalize]: 0.00047105
        [add_forward_monad_depend]: 5.21998e-06
        [auto_monad_grad]: 1.96998e-06
        [auto_monad_eliminator]: 1.308e-05
        [cse]: 2.527e-05
        [replace_applicator]: 1.069e-05
    [Cycle 2]: 0.0003196, [27]
        [switch_simplify]: 5.71998e-06
        [loop_unroll]: 5.30999e-06
        [a_1]: 8.931e-05
        [with_stream_mark]: 7.86001e-06
        [recompute_prepare]: 5.09e-06
        [updatestate_depend_eliminate]: 2.53e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 9.90025e-07
        [specialize_transform]: 5.25999e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.24e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.52998e-06
        [merge_forward]: 2.33002e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.21e-05
        [j_node_and_user_rematch]: 8.08001e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 6.98e-06
        [inline_without_move]: 4.75001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.18001e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 4.91002e-06
        [cse]: 1.028e-05
        [replace_applicator]: 5.35001e-06
[py_interpret_to_execute_after_opt_a]: 7.92e-06
[rewriter_after_opt_a]: 3.098e-05
[convert_after_rewriter]: 6.94001e-06
[order_py_execute_after_rewriter]: 5.28002e-06
[mutable_eliminate]: 0.00048115
[jit_opt_b]: 5.039e-05, [1]
    [Cycle 1]: 4.355e-05, [2]
        [frontend_op_eliminate]: 1.629e-05
        [inline_after_opt_a]: 1.542e-05
[cconv]: 3.185e-05
[loop_unroll]: 0.00041355
[jit_opt_after_cconv]: 0.00013694, [1]
    [Cycle 1]: 0.00013071, [11]
        [c_1]: 2.049e-05
        [parameter_eliminate]: 2.54999e-06
        [updatestate_depend_eliminate]: 5.34e-06
        [updatestate_assign_eliminate]: 2.36998e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 1.584e-05
        [call_graph_tuple_transform]: 1.867e-05
        [tuple_list_get_item_eliminator]: 5.62999e-06
        [none_parameter_eliminate]: 1.64e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 5.35001e-06
[remove_dup_value]: 1.34e-05
[partial_unused_args_eliminate]: 2.09e-06
[environ_conv]: 4.83001e-06
[add_recomputation]: 4.285e-05
[cse_after_recomputation]: 2.005e-05, [1]
    [Cycle 1]: 1.467e-05, [1]
        [cse]: 9.18002e-06
[auto_monad_reorder]: 1.443e-05
[get_jit_bprop_graph]: 1.42999e-06
[rewriter_after_jit_bprop_graph]: 4.66002e-06
[opt_after_jit_grad]: 0.00044205
[symbol_engine_optimizer]: 7.02e-05, [1]
    [Cycle 1]: 6.436e-05, [6]
        [build]: 3.66999e-06
        [elim_shapecalc]: 7.33e-06
        [elim_not_effective]: 1.237e-05
        [opt_reshape]: 5.47999e-06
        [fold_const_symbol]: 8.28001e-06
        [renormalize]: 4.09986e-07
[validate]: 3.083e-05
Sums
    bootstrap                            :     0.000261s :  3.45%
    type_inference                       :     0.003988s : 52.86%
    event_method                         :     0.000016s :  0.22%
    auto_monad                           :     0.000062s :  0.82%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.42%
    rewriter_before_opt_a                :     0.000070s :  0.93%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000056s :  0.74%
    jit_opt_a.loop_unroll                :     0.000035s :  0.47%
    jit_opt_a.a_1                        :     0.000591s :  7.84%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000471s :  6.25%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.24%
    jit_opt_a.cse                        :     0.000036s :  0.47%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.10%
    rewriter_after_opt_a                 :     0.000031s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000481s :  6.38%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.20%
    cconv                                :     0.000032s :  0.42%
    loop_unroll                          :     0.000414s :  5.48%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.25%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000013s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000043s :  0.57%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000014s :  0.19%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000442s :  5.86%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000031s :  0.41%


Time group info:
------[substitution.]   0.000178    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.99% :     0.000005s :      3: substitution.graph_param_transform
83.32% :     0.000148s :      4: substitution.inline
 1.59% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.30% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.92% :     0.000003s :      2: substitution.replace_old_param
 6.13% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003929     2
80.52% :     0.003164s :      1: type_inference.infer
19.48% :     0.000765s :      1: type_inference.specialize
------[replace.]   0.000064     6
67.71% :     0.000044s :      4: replace.inline
32.29% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000155     6
93.76% :     0.000146s :      4: match.inline
 6.24% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.10% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.46% :     0.000002s :     10: predicate.addn_zero_filter
 2.07% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.39% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.40% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.28% :     0.000001s :      6: predicate.less_batch_normalization
 1.69% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.15% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.26% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.18% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.48% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.90% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.74% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.56% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.58% :     0.000009s :     53: predicate.switch_simplify
 1.36% :     0.000002s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.01% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000500    12
50.30% :     0.000252s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.36% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.33% :     0.000222s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012502    72
 0.37% :     0.000046s :      1: add_recomputation
 0.53% :     0.000066s :      1: auto_monad
 0.14% :     0.000017s :      1: auto_monad_reorder
 2.20% :     0.000275s :      1: bootstrap
 0.28% :     0.000035s :      1: cconv
 0.09% :     0.000012s :      1: convert_after_rewriter
 0.18% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.13% :     0.004767s :      1: jit_opt_a
 1.12% :     0.000140s :      1: jit_opt_after_cconv
 0.42% :     0.000053s :      1: jit_opt_b
 3.37% :     0.000421s :      1: loop_unroll
 3.91% :     0.000489s :      1: mutable_eliminate
 6.34% :     0.000793s :     26: opt.transform.jit_opt_a
 0.37% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.60% :     0.000450s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.28% :     0.000035s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000016s :      1: remove_dup_value
 1.84% :     0.000230s :      1: renormalize.infer
 1.87% :     0.000234s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000034s :      1: rewriter_after_opt_a
 0.58% :     0.000073s :      1: rewriter_before_opt_a
 0.58% :     0.000073s :      1: symbol_engine_optimizer
32.00% :     0.004000s :      1: type_inference


TotalTime = 0.0195922, [33]
[bootstrap]: 0.00029899
[type_inference]: 0.00425373
[event_method]: 1.659e-05
[auto_monad]: 6.474e-05
[graph_reusing]: 5.63002e-06
[pre_auto_parallel]: 2.74999e-06
[py_interpret_to_execute]: 3.133e-05
[rewriter_before_opt_a]: 0.00010618
[expand_dump_flag]: 3.10998e-06
[jit_opt_a]: 0.00694259, [2]
    [Cycle 1]: 0.00336594, [27]
        [switch_simplify]: 5.221e-05
        [loop_unroll]: 2.859e-05
        [a_1]: 0.00235814
        [with_stream_mark]: 1.9e-05
        [recompute_prepare]: 8.19002e-06
        [updatestate_depend_eliminate]: 4.38999e-06
        [updatestate_assign_eliminate]: 2.82002e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [parameter_eliminate]: 2.37999e-06
        [specialize_transform]: 5.64998e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.90001e-06
        [get_grad_eliminate_]: 5.04e-06
        [merge_forward]: 3.83999e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.43e-05
        [j_node_and_user_rematch]: 8.80999e-06
        [meta_fg_expand]: 2.21e-06
        [replace_old_param]: 9.54999e-06
        [inline_without_move]: 4.95999e-06
        [renormalize]: 0.00058421
        [add_forward_monad_depend]: 5.46e-06
        [auto_monad_grad]: 2.17001e-06
        [auto_monad_eliminator]: 1.338e-05
        [cse]: 2.508e-05
        [replace_applicator]: 1.156e-05
    [Cycle 2]: 0.0003074, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 4.75999e-06
        [a_1]: 8.652e-05
        [with_stream_mark]: 7.73001e-06
        [recompute_prepare]: 4.65001e-06
        [updatestate_depend_eliminate]: 2.88998e-06
        [updatestate_assign_eliminate]: 2.05002e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 4.79e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 4.68001e-06
        [meta_shard_fg_expand]: 1.04998e-06
        [get_grad_eliminate_]: 4.45e-06
        [merge_forward]: 2.63e-06
        [cell_reuse_recompute_pass]: 1.72001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.142e-05
        [j_node_and_user_rematch]: 7.68001e-06
        [meta_fg_expand]: 1.37e-06
        [replace_old_param]: 7.1e-06
        [inline_without_move]: 4.50999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.15999e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 4.73001e-06
        [cse]: 9.86e-06
        [replace_applicator]: 4.65999e-06
[py_interpret_to_execute_after_opt_a]: 1.038e-05
[rewriter_after_opt_a]: 3.424e-05
[convert_after_rewriter]: 6.49001e-06
[order_py_execute_after_rewriter]: 4.62e-06
[mutable_eliminate]: 0.00054224
[jit_opt_b]: 4.845e-05, [1]
    [Cycle 1]: 4.137e-05, [2]
        [frontend_op_eliminate]: 1.554e-05
        [inline_after_opt_a]: 1.445e-05
[cconv]: 2.259e-05
[loop_unroll]: 0.00044954
[jit_opt_after_cconv]: 0.00013114, [1]
    [Cycle 1]: 0.00012472, [11]
        [c_1]: 2.009e-05
        [parameter_eliminate]: 2.73e-06
        [updatestate_depend_eliminate]: 5.54e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.08998e-06
        [cse]: 1.485e-05
        [call_graph_tuple_transform]: 1.661e-05
        [tuple_list_get_item_eliminator]: 5.22e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 5.09e-06
[remove_dup_value]: 1.202e-05
[partial_unused_args_eliminate]: 1.87001e-06
[environ_conv]: 4.49998e-06
[add_recomputation]: 4.473e-05
[cse_after_recomputation]: 1.975e-05, [1]
    [Cycle 1]: 1.426e-05, [1]
        [cse]: 8.79e-06
[auto_monad_reorder]: 1.392e-05
[get_jit_bprop_graph]: 2.27999e-06
[rewriter_after_jit_bprop_graph]: 5.12999e-06
[opt_after_jit_grad]: 0.00043154
[symbol_engine_optimizer]: 6.909e-05, [1]
    [Cycle 1]: 6.312e-05, [6]
        [build]: 3.43e-06
        [elim_shapecalc]: 7.11999e-06
        [elim_not_effective]: 1.165e-05
        [opt_reshape]: 5.28002e-06
        [fold_const_symbol]: 8.25e-06
        [renormalize]: 4.30009e-07
[validate]: 3.098e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.00578578
[execute]: 6.59999e-06
Sums
    bootstrap                            :     0.000299s :  1.90%
    type_inference                       :     0.004254s : 27.05%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000065s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.20%
    rewriter_before_opt_a                :     0.000106s :  0.68%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.37%
    jit_opt_a.loop_unroll                :     0.000033s :  0.21%
    jit_opt_a.a_1                        :     0.002445s : 15.54%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000017s :  0.11%
    jit_opt_a.inline_without_move        :     0.000009s :  0.06%
    jit_opt_a.renormalize                :     0.000584s :  3.72%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000035s :  0.22%
    jit_opt_a.replace_applicator         :     0.000016s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000542s :  3.45%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000023s :  0.14%
    loop_unroll                          :     0.000450s :  2.86%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000012s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000045s :  0.28%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000432s :  2.74%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005786s : 36.79%
    execute                              :     0.000007s :  0.04%


Time group info:
------[substitution.]   0.002027    23
 0.11% :     0.000002s :      2: substitution.elim_not_effective
 0.08% :     0.000002s :      2: substitution.fold_const_symbol
 0.25% :     0.000005s :      3: substitution.graph_param_transform
 7.93% :     0.000161s :      4: substitution.inline
 0.16% :     0.000003s :      4: substitution.j_node_and_user_rematch
 0.20% :     0.000004s :      4: substitution.remove_not_recompute_node
 0.22% :     0.000005s :      2: substitution.replace_old_param
91.06% :     0.001846s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004190     2
81.56% :     0.003418s :      1: type_inference.infer
18.44% :     0.000773s :      1: type_inference.specialize
------[replace.]   0.000072     6
59.57% :     0.000043s :      4: replace.inline
40.43% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.002002     6
 7.90% :     0.000158s :      4: match.inline
92.10% :     0.001844s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 2.27% :     0.000002s :     10: predicate.arithmetic_simplify
 1.16% :     0.000001s :     10: predicate.cast_eliminate
 0.58% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.42% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.81% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.59% :     0.000006s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.95% :     0.000001s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.79% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.57% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.65% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.87% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.53% :     0.000008s :     53: predicate.switch_simplify
 1.29% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.22% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.38% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000499    12
48.72% :     0.000243s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.97% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.31% :     0.000231s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022920    76
 0.21% :     0.000048s :      1: add_recomputation
 0.30% :     0.000068s :      1: auto_monad
 0.07% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.38% :     0.000316s :      1: bootstrap
 0.11% :     0.000025s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.09% :     0.000021s :      1: event_method
 0.04% :     0.000010s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
30.30% :     0.006945s :      1: jit_opt_a
 0.59% :     0.000134s :      1: jit_opt_after_cconv
 0.22% :     0.000051s :      1: jit_opt_b
 2.00% :     0.000457s :      1: loop_unroll
 2.40% :     0.000551s :      1: mutable_eliminate
11.55% :     0.002647s :     26: opt.transform.jit_opt_a
 0.19% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000029s :      4: opt.transform.symbol_engine_opt
 1.92% :     0.000439s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.15% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000014s :      1: remove_dup_value
 1.31% :     0.000300s :      1: renormalize.infer
 1.21% :     0.000277s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000037s :      1: rewriter_after_opt_a
 0.48% :     0.000110s :      1: rewriter_before_opt_a
 0.31% :     0.000072s :      1: symbol_engine_optimizer
25.29% :     0.005796s :      1: task_emit
18.62% :     0.004268s :      1: type_inference
 0.21% :     0.000049s :      1: validate


TotalTime = 0.0106681, [30]
[bootstrap]: 0.00025567
[type_inference]: 0.00376886
[event_method]: 1.617e-05
[auto_monad]: 5.859e-05
[graph_reusing]: 5.74e-06
[pre_auto_parallel]: 2.07001e-06
[py_interpret_to_execute]: 2.73e-05
[rewriter_before_opt_a]: 6.795e-05
[expand_dump_flag]: 2.54999e-06
[jit_opt_a]: 0.00454098, [2]
    [Cycle 1]: 0.00130425, [27]
        [switch_simplify]: 4.769e-05
        [loop_unroll]: 2.959e-05
        [a_1]: 0.00048736
        [with_stream_mark]: 1.183e-05
        [recompute_prepare]: 7.1e-06
        [updatestate_depend_eliminate]: 3.34001e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.93e-06
        [parameter_eliminate]: 1.69e-06
        [specialize_transform]: 5.94e-06
        [updatestate_useless_node_eliminater]: 5.19998e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 1.82999e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 4.11001e-06
        [cell_reuse_recompute_pass]: 1.14998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.374e-05
        [j_node_and_user_rematch]: 8.67998e-06
        [meta_fg_expand]: 2.23998e-06
        [replace_old_param]: 7.56999e-06
        [inline_without_move]: 5.28002e-06
        [renormalize]: 0.00043187
        [add_forward_monad_depend]: 4.84e-06
        [auto_monad_grad]: 1.84e-06
        [auto_monad_eliminator]: 1.231e-05
        [cse]: 2.214e-05
        [replace_applicator]: 1.119e-05
    [Cycle 2]: 0.00032155, [27]
        [switch_simplify]: 5.66003e-06
        [loop_unroll]: 5.30001e-06
        [a_1]: 9.079e-05
        [with_stream_mark]: 8.53001e-06
        [recompute_prepare]: 5.27001e-06
        [updatestate_depend_eliminate]: 2.54999e-06
        [updatestate_assign_eliminate]: 2.05002e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 1.04998e-06
        [specialize_transform]: 4.94e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.08001e-06
        [get_grad_eliminate_]: 4.57998e-06
        [merge_forward]: 2.36998e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.235e-05
        [j_node_and_user_rematch]: 8.18001e-06
        [meta_fg_expand]: 1.65001e-06
        [replace_old_param]: 7.31001e-06
        [inline_without_move]: 4.75999e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 5.02e-06
        [cse]: 1.036e-05
        [replace_applicator]: 5.30001e-06
[py_interpret_to_execute_after_opt_a]: 7.76001e-06
[rewriter_after_opt_a]: 2.819e-05
[convert_after_rewriter]: 6.58998e-06
[order_py_execute_after_rewriter]: 4.83001e-06
[mutable_eliminate]: 0.00046397
[jit_opt_b]: 4.998e-05, [1]
    [Cycle 1]: 4.271e-05, [2]
        [frontend_op_eliminate]: 1.68e-05
        [inline_after_opt_a]: 1.455e-05
[cconv]: 2.12e-05
[loop_unroll]: 0.00041078
[jit_opt_after_cconv]: 0.00013299, [1]
    [Cycle 1]: 0.00012668, [11]
        [c_1]: 2.037e-05
        [parameter_eliminate]: 2.22999e-06
        [updatestate_depend_eliminate]: 5.29e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [cse]: 1.526e-05
        [call_graph_tuple_transform]: 1.753e-05
        [tuple_list_get_item_eliminator]: 5.42001e-06
        [none_parameter_eliminate]: 1.32e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 5.45001e-06
[remove_dup_value]: 1.086e-05
[partial_unused_args_eliminate]: 2.06e-06
[environ_conv]: 4.76002e-06
[add_recomputation]: 4.046e-05
[cse_after_recomputation]: 2.061e-05, [1]
    [Cycle 1]: 1.461e-05, [1]
        [cse]: 9.05999e-06
[auto_monad_reorder]: 1.355e-05
[get_jit_bprop_graph]: 1.34998e-06
[rewriter_after_jit_bprop_graph]: 4.58999e-06
[opt_after_jit_grad]: 0.00044445
[symbol_engine_optimizer]: 7.123e-05, [1]
    [Cycle 1]: 6.479e-05, [6]
        [build]: 2.86e-06
        [elim_shapecalc]: 7.35e-06
        [elim_not_effective]: 1.257e-05
        [opt_reshape]: 5.44e-06
        [fold_const_symbol]: 9.10001e-06
        [renormalize]: 4.30009e-07
[validate]: 2.849e-05
Sums
    bootstrap                            :     0.000256s :  3.55%
    type_inference                       :     0.003769s : 52.32%
    event_method                         :     0.000016s :  0.22%
    auto_monad                           :     0.000059s :  0.81%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000027s :  0.38%
    rewriter_before_opt_a                :     0.000068s :  0.94%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000053s :  0.74%
    jit_opt_a.loop_unroll                :     0.000035s :  0.48%
    jit_opt_a.a_1                        :     0.000578s :  8.03%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.07%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.09%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.36%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000432s :  6.00%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.24%
    jit_opt_a.cse                        :     0.000032s :  0.45%
    jit_opt_a.replace_applicator         :     0.000016s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000028s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000464s :  6.44%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.20%
    cconv                                :     0.000021s :  0.29%
    loop_unroll                          :     0.000411s :  5.70%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000015s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.08%
    remove_dup_value                     :     0.000011s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000040s :  0.56%
    cse_after_recomputation.cse          :     0.000009s :  0.13%
    auto_monad_reorder                   :     0.000014s :  0.19%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000444s :  6.17%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.13%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000028s :  0.40%


Time group info:
------[substitution.]   0.000168    23
 1.22% :     0.000002s :      2: substitution.elim_not_effective
 0.78% :     0.000001s :      2: substitution.fold_const_symbol
 2.75% :     0.000005s :      3: substitution.graph_param_transform
82.56% :     0.000138s :      4: substitution.inline
 1.67% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.65% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.86% :     0.000003s :      2: substitution.replace_old_param
 6.52% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003716     2
80.54% :     0.002993s :      1: type_inference.infer
19.46% :     0.000723s :      1: type_inference.specialize
------[replace.]   0.000062     6
65.50% :     0.000040s :      4: replace.inline
34.50% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000146     6
93.27% :     0.000136s :      4: match.inline
 6.73% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.55% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.70% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.19% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.50% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.08% :     0.000006s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.43% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.58% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.53% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.56% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.47% :     0.000008s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.55% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.75% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.86% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000469    12
49.49% :     0.000232s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.43% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.08% :     0.000211s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011944    72
 0.36% :     0.000043s :      1: add_recomputation
 0.52% :     0.000062s :      1: auto_monad
 0.13% :     0.000016s :      1: auto_monad_reorder
 2.25% :     0.000269s :      1: bootstrap
 0.20% :     0.000024s :      1: cconv
 0.10% :     0.000011s :      1: convert_after_rewriter
 0.19% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.07% :     0.000008s :      1: graph_reusing
38.04% :     0.004544s :      1: jit_opt_a
 1.14% :     0.000136s :      1: jit_opt_after_cconv
 0.44% :     0.000053s :      1: jit_opt_b
 3.50% :     0.000418s :      1: loop_unroll
 3.96% :     0.000473s :      1: mutable_eliminate
 6.51% :     0.000778s :     26: opt.transform.jit_opt_a
 0.38% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.78% :     0.000452s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000030s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000013s :      1: remove_dup_value
 1.72% :     0.000206s :      1: renormalize.infer
 1.83% :     0.000219s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000031s :      1: rewriter_after_opt_a
 0.59% :     0.000071s :      1: rewriter_before_opt_a
 0.62% :     0.000074s :      1: symbol_engine_optimizer
31.64% :     0.003779s :      1: type_inference

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:20.708.394 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda84a8d30> should be wrapped by jit only once.

TotalTime = 0.106602, [33]
[bootstrap]: 0.00027213
[type_inference]: 0.0814904
[event_method]: 0.00031255
[auto_monad]: 9.879e-05
[graph_reusing]: 6.46e-06
[pre_auto_parallel]: 2.76999e-06
[py_interpret_to_execute]: 2.62e-05
[rewriter_before_opt_a]: 9.992e-05
[expand_dump_flag]: 2.98998e-06
[jit_opt_a]: 0.0221657, [3]
    [Cycle 1]: 0.015102, [27]
        [switch_simplify]: 6.635e-05
        [loop_unroll]: 4.294e-05
        [a_1]: 0.00090789
        [with_stream_mark]: 2.377e-05
        [recompute_prepare]: 1.979e-05
        [updatestate_depend_eliminate]: 8.15e-06
        [updatestate_assign_eliminate]: 6.16998e-06
        [updatestate_loads_eliminate]: 5.94e-06
        [parameter_eliminate]: 1.97999e-06
        [specialize_transform]: 1.397e-05
        [updatestate_useless_node_eliminater]: 1.305e-05
        [accelerated_algorithm]: 1.386e-05
        [meta_shard_fg_expand]: 3.41001e-06
        [get_grad_eliminate_]: 1.347e-05
        [merge_forward]: 8.27998e-06
        [cell_reuse_recompute_pass]: 8.2e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.601e-05
        [j_node_and_user_rematch]: 2.374e-05
        [meta_fg_expand]: 0.00151354
        [replace_old_param]: 6.011e-05
        [inline_without_move]: 5.561e-05
        [renormalize]: 0.0116702
        [add_forward_monad_depend]: 1.33e-05
        [auto_monad_grad]: 6.49999e-06
        [auto_monad_eliminator]: 5.545e-05
        [cse]: 0.0002292
        [replace_applicator]: 8.671e-05
    [Cycle 2]: 0.00357665, [27]
        [switch_simplify]: 4.266e-05
        [loop_unroll]: 4.007e-05
        [a_1]: 0.00116312
        [with_stream_mark]: 1.647e-05
        [recompute_prepare]: 9.62001e-06
        [updatestate_depend_eliminate]: 4.42e-06
        [updatestate_assign_eliminate]: 4.23999e-06
        [updatestate_loads_eliminate]: 3.82002e-06
        [parameter_eliminate]: 2.06998e-06
        [specialize_transform]: 8.35999e-06
        [updatestate_useless_node_eliminater]: 7.1e-06
        [accelerated_algorithm]: 8.72998e-06
        [meta_shard_fg_expand]: 2.65002e-06
        [get_grad_eliminate_]: 7.26001e-06
        [merge_forward]: 4.58999e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.701e-05
        [j_node_and_user_rematch]: 1.168e-05
        [meta_fg_expand]: 7.463e-05
        [replace_old_param]: 8.52e-06
        [inline_without_move]: 7.06001e-06
        [renormalize]: 0.00192597
        [add_forward_monad_depend]: 6.71e-06
        [auto_monad_grad]: 2.26e-06
        [auto_monad_eliminator]: 1.067e-05
        [cse]: 1.692e-05
        [replace_applicator]: 1.26e-05
    [Cycle 3]: 0.00024613, [27]
        [switch_simplify]: 4.48001e-06
        [loop_unroll]: 4e-06
        [a_1]: 4.301e-05
        [with_stream_mark]: 1.092e-05
        [recompute_prepare]: 3.66999e-06
        [updatestate_depend_eliminate]: 2.14999e-06
        [updatestate_assign_eliminate]: 1.67001e-06
        [updatestate_loads_eliminate]: 1.42e-06
        [parameter_eliminate]: 1.21997e-06
        [specialize_transform]: 3.70998e-06
        [updatestate_useless_node_eliminater]: 3.36001e-06
        [accelerated_algorithm]: 3.53999e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 3.46001e-06
        [merge_forward]: 2.17001e-06
        [cell_reuse_recompute_pass]: 2.02999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.129e-05
        [j_node_and_user_rematch]: 6.07999e-06
        [meta_fg_expand]: 1.27e-06
        [replace_old_param]: 3.3e-06
        [inline_without_move]: 3.31001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 3.78999e-06
        [cse]: 8.60001e-06
        [replace_applicator]: 3.58e-06
[py_interpret_to_execute_after_opt_a]: 9.96998e-06
[rewriter_after_opt_a]: 3.31e-05
[convert_after_rewriter]: 5.74999e-06
[order_py_execute_after_rewriter]: 4.08001e-06
[mutable_eliminate]: 0.00062781
[jit_opt_b]: 4.166e-05, [1]
    [Cycle 1]: 3.502e-05, [2]
        [frontend_op_eliminate]: 1.177e-05
        [inline_after_opt_a]: 1.25e-05
[cconv]: 2.464e-05
[loop_unroll]: 0.00042758
[jit_opt_after_cconv]: 0.00011998, [1]
    [Cycle 1]: 0.00011317, [11]
        [c_1]: 1.25e-05
        [parameter_eliminate]: 2.63998e-06
        [updatestate_depend_eliminate]: 5.44e-06
        [updatestate_assign_eliminate]: 1.76e-06
        [updatestate_loads_eliminate]: 1.54e-06
        [cse]: 1.693e-05
        [call_graph_tuple_transform]: 1.428e-05
        [tuple_list_get_item_eliminator]: 3.68e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 6.39993e-07
        [switch_simplify]: 3.61999e-06
[remove_dup_value]: 1.473e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 5.04e-06
[add_recomputation]: 3.268e-05
[cse_after_recomputation]: 1.859e-05, [1]
    [Cycle 1]: 1.335e-05, [1]
        [cse]: 7.38e-06
[auto_monad_reorder]: 1.276e-05
[get_jit_bprop_graph]: 2.07001e-06
[rewriter_after_jit_bprop_graph]: 4.90001e-06
[opt_after_jit_grad]: 0.00044681
[symbol_engine_optimizer]: 6.299e-05, [1]
    [Cycle 1]: 5.607e-05, [6]
        [build]: 3.04999e-06
        [elim_shapecalc]: 5.64e-06
        [elim_not_effective]: 1.003e-05
        [opt_reshape]: 4.43001e-06
        [fold_const_symbol]: 6.01e-06
        [renormalize]: 5.3001e-07
[validate]: 2.858e-05
[backend_pass]: 1.02e-06
[task_emit]: 2.154e-05
[execute]: 1.35001e-06
Sums
    bootstrap                            :     0.000272s :  0.27%
    type_inference                       :     0.081490s : 79.43%
    event_method                         :     0.000313s :  0.30%
    auto_monad                           :     0.000099s :  0.10%
    graph_reusing                        :     0.000006s :  0.01%
    pre_auto_parallel                    :     0.000003s :  0.00%
    py_interpret_to_execute              :     0.000026s :  0.03%
    rewriter_before_opt_a                :     0.000100s :  0.10%
    expand_dump_flag                     :     0.000003s :  0.00%
    jit_opt_a.switch_simplify            :     0.000113s :  0.11%
    jit_opt_a.loop_unroll                :     0.000087s :  0.08%
    jit_opt_a.a_1                        :     0.002114s :  2.06%
    jit_opt_a.with_stream_mark           :     0.000051s :  0.05%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.03%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.01%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.01%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.01%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000026s :  0.03%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000024s :  0.02%
    jit_opt_a.accelerated_algorithm      :     0.000026s :  0.03%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.01%
    jit_opt_a.get_grad_eliminate_        :     0.000024s :  0.02%
    jit_opt_a.merge_forward              :     0.000015s :  0.01%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.00%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000054s :  0.05%
    jit_opt_a.j_node_and_user_rematch    :     0.000042s :  0.04%
    jit_opt_a.meta_fg_expand             :     0.001589s :  1.55%
    jit_opt_a.replace_old_param          :     0.000072s :  0.07%
    jit_opt_a.inline_without_move        :     0.000066s :  0.06%
    jit_opt_a.renormalize                :     0.013596s : 13.25%
    jit_opt_a.add_forward_monad_depend   :     0.000021s :  0.02%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000070s :  0.07%
    jit_opt_a.cse                        :     0.000255s :  0.25%
    jit_opt_a.replace_applicator         :     0.000103s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.01%
    rewriter_after_opt_a                 :     0.000033s :  0.03%
    convert_after_rewriter               :     0.000006s :  0.01%
    order_py_execute_after_rewriter      :     0.000004s :  0.00%
    mutable_eliminate                    :     0.000628s :  0.61%
    jit_opt_b.frontend_op_eliminate      :     0.000012s :  0.01%
    jit_opt_b.inline_after_opt_a         :     0.000013s :  0.01%
    cconv                                :     0.000025s :  0.02%
    loop_unroll                          :     0.000428s :  0.42%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.01%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.00%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000017s :  0.02%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.01%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.00%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.00%
    remove_dup_value                     :     0.000015s :  0.01%
    partial_unused_args_eliminate        :     0.000002s :  0.00%
    environ_conv                         :     0.000005s :  0.00%
    add_recomputation                    :     0.000033s :  0.03%
    cse_after_recomputation.cse          :     0.000007s :  0.01%
    auto_monad_reorder                   :     0.000013s :  0.01%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.00%
    opt_after_jit_grad                   :     0.000447s :  0.44%
    symbol_engine_optimizer.build        :     0.000003s :  0.00%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.01%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.00%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000029s :  0.03%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000022s :  0.02%
    execute                              :     0.000001s :  0.00%


Time group info:
------[substitution.]   0.000514   102
 0.34% :     0.000002s :      1: substitution.elim_not_effective
 0.18% :     0.000001s :      1: substitution.fold_const_symbol
 0.84% :     0.000004s :      1: substitution.graph_param_transform
61.84% :     0.000318s :     12: substitution.inline
 3.32% :     0.000017s :      2: substitution.inline_without_move
 1.61% :     0.000008s :     12: substitution.j_node_and_user_rematch
 2.32% :     0.000012s :      7: substitution.minmaximum_grad
 1.56% :     0.000008s :      5: substitution.partial_eliminate
 1.90% :     0.000010s :     12: substitution.remove_not_recompute_node
 6.06% :     0.000031s :     10: substitution.replace_applicator
 0.94% :     0.000005s :      7: substitution.replace_old_param
 1.27% :     0.000007s :      1: substitution.reshape_eliminate
 0.47% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.11% :     0.000011s :      2: substitution.tile_eliminate
 4.18% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.88% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 8.18% :     0.000042s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.081404     2
97.02% :     0.078980s :      1: type_inference.infer
 2.98% :     0.002423s :      1: type_inference.specialize
------[replace.]   0.000152    19
63.82% :     0.000097s :     12: replace.inline
36.18% :     0.000055s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000330    19
94.11% :     0.000311s :     12: match.inline
 5.89% :     0.000019s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000374  2646
 1.84% :     0.000007s :     46: predicate.accumulaten_eliminater
 0.30% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.38% :     0.000005s :     46: predicate.addn_check_dump
 1.89% :     0.000007s :     46: predicate.addn_zero_filter
 2.36% :     0.000009s :     46: predicate.arithmetic_simplify
 1.83% :     0.000007s :     46: predicate.cast_eliminate
 0.10% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.40% :     0.000005s :     46: predicate.compare_switch_simplify
 1.53% :     0.000006s :     46: predicate.depend_value_elim
 1.53% :     0.000006s :     46: predicate.dict_get_item_const_eliminator
 1.56% :     0.000006s :     46: predicate.dict_get_item_eliminator
 1.50% :     0.000006s :     46: predicate.dict_set_item_eliminator
 0.24% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.06% :     0.000000s :      1: predicate.elim_not_effective
 0.11% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.61% :     0.000006s :     46: predicate.environ_add_const_eliminate
 1.48% :     0.000006s :     46: predicate.environ_get_add_eliminate
 1.45% :     0.000005s :     46: predicate.environ_get_depend_swap
 1.61% :     0.000006s :     46: predicate.environ_get_eliminate
 1.46% :     0.000005s :     46: predicate.environ_get_set_eliminate
 0.02% :     0.000000s :      1: predicate.fold_const_symbol
 0.80% :     0.000003s :     19: predicate.get_grad_eliminate
 0.05% :     0.000000s :      1: predicate.graph_param_transform
 4.09% :     0.000015s :     67: predicate.inline
 1.81% :     0.000007s :     43: predicate.inline_without_move
 0.36% :     0.000001s :     19: predicate.j_node_and_user_rematch
 1.08% :     0.000004s :     19: predicate.less_batch_normalization
 1.90% :     0.000007s :     53: predicate.list_to_tuple_eliminator_
 1.85% :     0.000007s :     54: predicate.load_eliminater
 0.35% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.31% :     0.000012s :     95: predicate.loop_unroll_before_grad
 1.70% :     0.000006s :     47: predicate.make_slice_get_slice_eliminator
 1.39% :     0.000005s :     46: predicate.merge_addn
 1.49% :     0.000006s :     46: predicate.minmaximum_grad
 0.35% :     0.000001s :      1: predicate.mutable_eliminate
 0.13% :     0.000000s :      1: predicate.opt_reshape
 2.27% :     0.000008s :     54: predicate.partial_eliminate
 1.53% :     0.000006s :     46: predicate.print_const_string_wrapper
 2.05% :     0.000008s :     46: predicate.reduce_eliminate
 1.82% :     0.000007s :     53: predicate.redundant_stop_gradient_eliminater
 0.40% :     0.000001s :     19: predicate.remove_not_recompute_node
 2.85% :     0.000011s :    100: predicate.replace_applicator
 0.99% :     0.000004s :     43: predicate.replace_old_param
 0.03% :     0.000000s :      1: predicate.reset_defer_inline
 1.78% :     0.000007s :     46: predicate.reshape_eliminate
 1.65% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.27% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.50% :     0.000006s :     46: predicate.same_eliminate
 0.49% :     0.000002s :     19: predicate.set_cell_output_no_recompute
 0.19% :     0.000001s :      2: predicate.special_op_eliminate
 0.80% :     0.000003s :     19: predicate.specialize_transform
 1.79% :     0.000007s :     46: predicate.split_environ_get_set_with_tuple_value
 1.57% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.06% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.79% :     0.000010s :     65: predicate.switch_defer_inline
 2.55% :     0.000010s :     65: predicate.switch_layer_defer_inline
 6.45% :     0.000024s :    161: predicate.switch_simplify
 1.67% :     0.000006s :     46: predicate.tile_eliminate
 1.52% :     0.000006s :     46: predicate.transpose_eliminate
 1.93% :     0.000007s :     46: predicate.tuple_list_convert_item_index_to_positive
 1.80% :     0.000007s :     46: predicate.tuple_list_get_item_depend_reorder
 2.96% :     0.000011s :     55: predicate.tuple_list_get_item_eliminator
 1.91% :     0.000007s :     46: predicate.tuple_list_set_item_eliminator
 1.88% :     0.000007s :     53: predicate.tuple_to_list_eliminator_
 1.77% :     0.000007s :     54: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000010s :     73: predicate.updatestate_useless_node_eliminater
 1.91% :     0.000007s :     46: predicate.value_based_eliminate
 0.05% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.12% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.004126    54
78.80% :     0.003251s :     38: func_graph_cloner_run.FuncGraphClonerGraph
21.20% :     0.000875s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.122989    91
 0.03% :     0.000035s :      1: add_recomputation
 0.09% :     0.000105s :      1: auto_monad
 0.01% :     0.000015s :      1: auto_monad_reorder
 0.00% :     0.000005s :      1: backend_pass
 0.23% :     0.000283s :      1: bootstrap
 0.02% :     0.000027s :      1: cconv
 0.01% :     0.000008s :      1: convert_after_rewriter
 0.02% :     0.000021s :      1: cse_after_recomputation
 0.01% :     0.000007s :      1: environ_conv
 0.26% :     0.000320s :      1: event_method
 0.00% :     0.000003s :      1: execute
 0.00% :     0.000005s :      1: expand_dump_flag
 0.00% :     0.000004s :      1: get_jit_bprop_graph
 0.01% :     0.000009s :      1: graph_reusing
18.03% :     0.022169s :      1: jit_opt_a
 0.10% :     0.000123s :      1: jit_opt_after_cconv
 0.04% :     0.000044s :      1: jit_opt_b
 0.35% :     0.000435s :      1: loop_unroll
 0.52% :     0.000636s :      1: mutable_eliminate
 2.23% :     0.002738s :     39: opt.transform.jit_opt_a
 0.02% :     0.000031s :      4: opt.transform.jit_opt_after_cconv
 0.01% :     0.000017s :      4: opt.transform.jit_opt_b
 0.01% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.01% :     0.000010s :      1: opt.transform.mutable_eliminate
 0.01% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.02% :     0.000023s :      4: opt.transform.symbol_engine_opt
 0.37% :     0.000455s :      1: opt_after_jit_grad
 0.00% :     0.000006s :      1: order_py_execute_after_rewriter
 0.00% :     0.000004s :      1: partial_unused_args_eliminate
 0.00% :     0.000005s :      1: pre_auto_parallel
 0.02% :     0.000029s :      1: py_interpret_to_execute
 0.01% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.01% :     0.000017s :      1: remove_dup_value
 9.16% :     0.011271s :      2: renormalize.infer
 1.88% :     0.002306s :      2: renormalize.specialize
 0.01% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.03% :     0.000036s :      1: rewriter_after_opt_a
 0.08% :     0.000103s :      1: rewriter_before_opt_a
 0.05% :     0.000066s :      1: symbol_engine_optimizer
 0.02% :     0.000026s :      1: task_emit
66.27% :     0.081503s :      1: type_inference
 0.04% :     0.000044s :      1: validate


TotalTime = 0.0106993, [30]
[bootstrap]: 0.00024705
[type_inference]: 0.003727
[event_method]: 1.608e-05
[auto_monad]: 5.621e-05
[graph_reusing]: 5.10999e-06
[pre_auto_parallel]: 1.94e-06
[py_interpret_to_execute]: 2.845e-05
[rewriter_before_opt_a]: 6.674e-05
[expand_dump_flag]: 3.02002e-06
[jit_opt_a]: 0.00456685, [2]
    [Cycle 1]: 0.00138245, [27]
        [switch_simplify]: 4.629e-05
        [loop_unroll]: 3.014e-05
        [a_1]: 0.00052542
        [with_stream_mark]: 1.524e-05
        [recompute_prepare]: 7.2e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 2.64999e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [parameter_eliminate]: 1.50999e-06
        [specialize_transform]: 5.87999e-06
        [updatestate_useless_node_eliminater]: 5.17999e-06
        [accelerated_algorithm]: 5.25999e-06
        [meta_shard_fg_expand]: 1.79998e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 2.78e-06
        [cell_reuse_recompute_pass]: 1.18001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.297e-05
        [j_node_and_user_rematch]: 8.69e-06
        [meta_fg_expand]: 2.68998e-06
        [replace_old_param]: 8.87e-06
        [inline_without_move]: 5.62999e-06
        [renormalize]: 0.00046615
        [add_forward_monad_depend]: 5.03002e-06
        [auto_monad_grad]: 1.52001e-06
        [auto_monad_eliminator]: 1.153e-05
        [cse]: 2.106e-05
        [replace_applicator]: 1.058e-05
    [Cycle 2]: 0.00032379, [27]
        [switch_simplify]: 6.04001e-06
        [loop_unroll]: 5.62999e-06
        [a_1]: 9.312e-05
        [with_stream_mark]: 8.56002e-06
        [recompute_prepare]: 4.92e-06
        [updatestate_depend_eliminate]: 2.56e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 5.12e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.19998e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.49001e-06
        [cell_reuse_recompute_pass]: 1.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.218e-05
        [j_node_and_user_rematch]: 7.76001e-06
        [meta_fg_expand]: 1.56998e-06
        [replace_old_param]: 7.30003e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 4.90001e-06
        [cse]: 1.058e-05
        [replace_applicator]: 5.10001e-06
[py_interpret_to_execute_after_opt_a]: 8.71002e-06
[rewriter_after_opt_a]: 3.902e-05
[convert_after_rewriter]: 6.17001e-06
[order_py_execute_after_rewriter]: 5.27001e-06
[mutable_eliminate]: 0.0004903
[jit_opt_b]: 4.987e-05, [1]
    [Cycle 1]: 4.289e-05, [2]
        [frontend_op_eliminate]: 1.617e-05
        [inline_after_opt_a]: 1.483e-05
[cconv]: 2.238e-05
[loop_unroll]: 0.00041376
[jit_opt_after_cconv]: 0.00013455, [1]
    [Cycle 1]: 0.00012789, [11]
        [c_1]: 2.032e-05
        [parameter_eliminate]: 2.45002e-06
        [updatestate_depend_eliminate]: 5.24e-06
        [updatestate_assign_eliminate]: 2.28998e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 1.59e-05
        [call_graph_tuple_transform]: 1.72e-05
        [tuple_list_get_item_eliminator]: 5.72001e-06
        [none_parameter_eliminate]: 1.02e-06
        [renormalize]: 5.00004e-07
        [switch_simplify]: 5.47999e-06
[remove_dup_value]: 1.276e-05
[partial_unused_args_eliminate]: 1.97001e-06
[environ_conv]: 4.65001e-06
[add_recomputation]: 4.048e-05
[cse_after_recomputation]: 2.16e-05, [1]
    [Cycle 1]: 1.595e-05, [1]
        [cse]: 1.002e-05
[auto_monad_reorder]: 1.324e-05
[get_jit_bprop_graph]: 1.44998e-06
[rewriter_after_jit_bprop_graph]: 4.68999e-06
[opt_after_jit_grad]: 0.00045971
[symbol_engine_optimizer]: 7.081e-05, [1]
    [Cycle 1]: 6.512e-05, [6]
        [build]: 3.17002e-06
        [elim_shapecalc]: 7.35998e-06
        [elim_not_effective]: 1.202e-05
        [opt_reshape]: 5.67999e-06
        [fold_const_symbol]: 8.70001e-06
        [renormalize]: 4.70027e-07
[validate]: 2.801e-05
Sums
    bootstrap                            :     0.000247s :  3.39%
    type_inference                       :     0.003727s : 51.17%
    event_method                         :     0.000016s :  0.22%
    auto_monad                           :     0.000056s :  0.77%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000028s :  0.39%
    rewriter_before_opt_a                :     0.000067s :  0.92%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000052s :  0.72%
    jit_opt_a.loop_unroll                :     0.000036s :  0.49%
    jit_opt_a.a_1                        :     0.000619s :  8.49%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.06%
    jit_opt_a.replace_old_param          :     0.000016s :  0.22%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000466s :  6.40%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.23%
    jit_opt_a.cse                        :     0.000032s :  0.43%
    jit_opt_a.replace_applicator         :     0.000016s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.12%
    rewriter_after_opt_a                 :     0.000039s :  0.54%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000490s :  6.73%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.20%
    cconv                                :     0.000022s :  0.31%
    loop_unroll                          :     0.000414s :  5.68%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.08%
    remove_dup_value                     :     0.000013s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000040s :  0.56%
    cse_after_recomputation.cse          :     0.000010s :  0.14%
    auto_monad_reorder                   :     0.000013s :  0.18%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000460s :  6.31%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000028s :  0.38%


Time group info:
------[substitution.]   0.000175    23
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.72% :     0.000005s :      3: substitution.graph_param_transform
80.08% :     0.000140s :      4: substitution.inline
 1.55% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.29% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.49% :     0.000004s :      2: substitution.replace_old_param
 9.11% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003677     2
80.02% :     0.002942s :      1: type_inference.infer
19.98% :     0.000735s :      1: type_inference.specialize
------[replace.]   0.000080     6
72.68% :     0.000058s :      4: replace.inline
27.32% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000152     6
90.34% :     0.000138s :      4: match.inline
 9.66% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000001s :     10: predicate.addn_zero_filter
 2.13% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.41% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.80% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.17% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.53% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000002s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.28% :     0.000001s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.51% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.81% :     0.000001s :      6: predicate.special_op_eliminate
 1.13% :     0.000001s :      6: predicate.specialize_transform
 1.56% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.63% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.51% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.88% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.91% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000471    12
49.45% :     0.000233s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.43% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.12% :     0.000212s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012050    72
 0.36% :     0.000044s :      1: add_recomputation
 0.50% :     0.000060s :      1: auto_monad
 0.13% :     0.000016s :      1: auto_monad_reorder
 2.15% :     0.000260s :      1: bootstrap
 0.21% :     0.000025s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.20% :     0.000024s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.17% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
37.92% :     0.004570s :      1: jit_opt_a
 1.14% :     0.000137s :      1: jit_opt_after_cconv
 0.44% :     0.000053s :      1: jit_opt_b
 3.49% :     0.000421s :      1: loop_unroll
 4.14% :     0.000499s :      1: mutable_eliminate
 6.78% :     0.000817s :     26: opt.transform.jit_opt_a
 0.38% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000025s :      4: opt.transform.jit_opt_b
 0.10% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.18% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.88% :     0.000467s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000031s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000015s :      1: remove_dup_value
 1.89% :     0.000228s :      1: renormalize.infer
 1.91% :     0.000230s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.35% :     0.000042s :      1: rewriter_after_opt_a
 0.58% :     0.000069s :      1: rewriter_before_opt_a
 0.61% :     0.000074s :      1: symbol_engine_optimizer
31.02% :     0.003737s :      1: type_inference


TotalTime = 0.0184456, [33]
[bootstrap]: 0.00026106
[type_inference]: 0.00397483
[event_method]: 1.583e-05
[auto_monad]: 6.06e-05
[graph_reusing]: 5.54e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 3.074e-05
[rewriter_before_opt_a]: 7.174e-05
[expand_dump_flag]: 3.35e-06
[jit_opt_a]: 0.00624699, [2]
    [Cycle 1]: 0.00283436, [27]
        [switch_simplify]: 4.586e-05
        [loop_unroll]: 2.902e-05
        [a_1]: 0.00193109
        [with_stream_mark]: 1.539e-05
        [recompute_prepare]: 7.6e-06
        [updatestate_depend_eliminate]: 2.95998e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 5.70001e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.32001e-06
        [meta_shard_fg_expand]: 1.35001e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 2.80002e-06
        [cell_reuse_recompute_pass]: 1.70001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.073e-05
        [j_node_and_user_rematch]: 8.79e-06
        [meta_fg_expand]: 1.78002e-06
        [replace_old_param]: 8.81002e-06
        [inline_without_move]: 5.02999e-06
        [renormalize]: 0.00052513
        [add_forward_monad_depend]: 4.16001e-06
        [auto_monad_grad]: 1.79998e-06
        [auto_monad_eliminator]: 9.79999e-06
        [cse]: 1.639e-05
        [replace_applicator]: 1.195e-05
    [Cycle 2]: 0.00030375, [27]
        [switch_simplify]: 5.29998e-06
        [loop_unroll]: 4.99e-06
        [a_1]: 8.772e-05
        [with_stream_mark]: 7.9e-06
        [recompute_prepare]: 4.85999e-06
        [updatestate_depend_eliminate]: 2.71999e-06
        [updatestate_assign_eliminate]: 1.92001e-06
        [updatestate_loads_eliminate]: 1.81998e-06
        [parameter_eliminate]: 1.09998e-06
        [specialize_transform]: 4.72998e-06
        [updatestate_useless_node_eliminater]: 4.65001e-06
        [accelerated_algorithm]: 4.58999e-06
        [meta_shard_fg_expand]: 1.18001e-06
        [get_grad_eliminate_]: 4.74e-06
        [merge_forward]: 2.16998e-06
        [cell_reuse_recompute_pass]: 2.29001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.187e-05
        [j_node_and_user_rematch]: 7.6e-06
        [meta_fg_expand]: 1.37e-06
        [replace_old_param]: 6.78e-06
        [inline_without_move]: 4.37998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.00001e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 4.56002e-06
        [cse]: 9.85002e-06
        [replace_applicator]: 4.81002e-06
[py_interpret_to_execute_after_opt_a]: 8.37998e-06
[rewriter_after_opt_a]: 2.644e-05
[convert_after_rewriter]: 5.62999e-06
[order_py_execute_after_rewriter]: 3.73001e-06
[mutable_eliminate]: 0.0005237
[jit_opt_b]: 4.663e-05, [1]
    [Cycle 1]: 4.076e-05, [2]
        [frontend_op_eliminate]: 1.542e-05
        [inline_after_opt_a]: 1.402e-05
[cconv]: 1.903e-05
[loop_unroll]: 0.00041247
[jit_opt_after_cconv]: 0.00013385, [1]
    [Cycle 1]: 0.00012775, [11]
        [c_1]: 2.037e-05
        [parameter_eliminate]: 2.69999e-06
        [updatestate_depend_eliminate]: 5.10999e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.785e-05
        [call_graph_tuple_transform]: 1.679e-05
        [tuple_list_get_item_eliminator]: 5.18002e-06
        [none_parameter_eliminate]: 1.05001e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 5.18002e-06
[remove_dup_value]: 5.541e-05
[partial_unused_args_eliminate]: 1.97999e-06
[environ_conv]: 4.02002e-06
[add_recomputation]: 3.85e-05
[cse_after_recomputation]: 2.139e-05, [1]
    [Cycle 1]: 1.601e-05, [1]
        [cse]: 9.89999e-06
[auto_monad_reorder]: 1.031e-05
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 4.77e-06
[opt_after_jit_grad]: 0.00046292
[symbol_engine_optimizer]: 6.818e-05, [1]
    [Cycle 1]: 6.23e-05, [6]
        [build]: 2.63e-06
        [elim_shapecalc]: 6.84001e-06
        [elim_not_effective]: 1.211e-05
        [opt_reshape]: 5.39e-06
        [fold_const_symbol]: 7.73999e-06
        [renormalize]: 4.69998e-07
[validate]: 2.781e-05
[backend_pass]: 7.39994e-07
[task_emit]: 0.00570697
[execute]: 8.36002e-06
Sums
    bootstrap                            :     0.000261s :  1.77%
    type_inference                       :     0.003975s : 26.92%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000061s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.21%
    rewriter_before_opt_a                :     0.000072s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000051s :  0.35%
    jit_opt_a.loop_unroll                :     0.000034s :  0.23%
    jit_opt_a.a_1                        :     0.002019s : 13.67%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000009s :  0.06%
    jit_opt_a.renormalize                :     0.000525s :  3.56%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.10%
    jit_opt_a.cse                        :     0.000026s :  0.18%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000026s :  0.18%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000524s :  3.55%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000019s :  0.13%
    loop_unroll                          :     0.000412s :  2.79%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000055s :  0.38%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000039s :  0.26%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000010s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000463s :  3.13%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.19%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005707s : 38.65%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.001600    23
 0.10% :     0.000002s :      2: substitution.elim_not_effective
 0.06% :     0.000001s :      2: substitution.fold_const_symbol
 0.25% :     0.000004s :      3: substitution.graph_param_transform
98.16% :     0.001570s :      4: substitution.inline
 0.20% :     0.000003s :      4: substitution.j_node_and_user_rematch
 0.28% :     0.000005s :      4: substitution.remove_not_recompute_node
 0.26% :     0.000004s :      2: substitution.replace_old_param
 0.69% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003922     2
81.62% :     0.003201s :      1: type_inference.infer
18.38% :     0.000721s :      1: type_inference.specialize
------[replace.]   0.000071     6
63.50% :     0.000045s :      4: replace.inline
36.50% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.001577     6
99.37% :     0.001567s :      4: match.inline
 0.63% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.54% :     0.000002s :     10: predicate.addn_zero_filter
 2.05% :     0.000002s :     10: predicate.arithmetic_simplify
 1.33% :     0.000001s :     10: predicate.cast_eliminate
 0.55% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.49% :     0.000002s :     10: predicate.depend_value_elim
 1.30% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.74% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.83% :     0.000007s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000001s :      6: predicate.less_batch_normalization
 1.75% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.66% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.44% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.65% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000002s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.34% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.53% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.84% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.23% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.46% :     0.000008s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.42% :     0.000002s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.41% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.39% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001919    12
11.86% :     0.000228s :      5: func_graph_cloner_run.FuncGraphClonerGraph
76.27% :     0.001464s :      1: func_graph_cloner_run.FuncGraphClonerNode
11.86% :     0.000228s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021285    76
 0.19% :     0.000041s :      1: add_recomputation
 0.30% :     0.000064s :      1: auto_monad
 0.06% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.30% :     0.000276s :      1: bootstrap
 0.10% :     0.000021s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.09% :     0.000019s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
29.36% :     0.006250s :      1: jit_opt_a
 0.64% :     0.000137s :      1: jit_opt_after_cconv
 0.23% :     0.000049s :      1: jit_opt_b
 1.98% :     0.000421s :      1: loop_unroll
 2.50% :     0.000533s :      1: mutable_eliminate
10.40% :     0.002214s :     26: opt.transform.jit_opt_a
 0.21% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000471s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000033s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.28% :     0.000059s :      1: remove_dup_value
 1.29% :     0.000274s :      1: renormalize.infer
 1.15% :     0.000245s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000029s :      1: rewriter_after_opt_a
 0.35% :     0.000075s :      1: rewriter_before_opt_a
 0.33% :     0.000071s :      1: symbol_engine_optimizer
26.87% :     0.005719s :      1: task_emit
18.73% :     0.003986s :      1: type_inference
 0.21% :     0.000044s :      1: validate


TotalTime = 0.0108703, [30]
[bootstrap]: 0.00025986
[type_inference]: 0.00386043
[event_method]: 1.618e-05
[auto_monad]: 7.648e-05
[graph_reusing]: 5.72999e-06
[pre_auto_parallel]: 2.14e-06
[py_interpret_to_execute]: 3.002e-05
[rewriter_before_opt_a]: 7.128e-05
[expand_dump_flag]: 2.74999e-06
[jit_opt_a]: 0.00461187, [2]
    [Cycle 1]: 0.00134522, [27]
        [switch_simplify]: 4.906e-05
        [loop_unroll]: 2.951e-05
        [a_1]: 0.00049312
        [with_stream_mark]: 1.383e-05
        [recompute_prepare]: 7.10998e-06
        [updatestate_depend_eliminate]: 3.24001e-06
        [updatestate_assign_eliminate]: 2.86999e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 2.27999e-06
        [specialize_transform]: 6.03998e-06
        [updatestate_useless_node_eliminater]: 5.08002e-06
        [accelerated_algorithm]: 5.70001e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 5.01002e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.436e-05
        [j_node_and_user_rematch]: 8.99998e-06
        [meta_fg_expand]: 2.30002e-06
        [replace_old_param]: 8.38001e-06
        [inline_without_move]: 5.17e-06
        [renormalize]: 0.00045737
        [add_forward_monad_depend]: 4.92999e-06
        [auto_monad_grad]: 1.68002e-06
        [auto_monad_eliminator]: 1.248e-05
        [cse]: 2.425e-05
        [replace_applicator]: 1.108e-05
    [Cycle 2]: 0.00031563, [27]
        [switch_simplify]: 5.46998e-06
        [loop_unroll]: 5.33002e-06
        [a_1]: 8.935e-05
        [with_stream_mark]: 8.17e-06
        [recompute_prepare]: 4.99e-06
        [updatestate_depend_eliminate]: 2.51e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.09998e-06
        [specialize_transform]: 5.16002e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 5.30001e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.79e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.185e-05
        [j_node_and_user_rematch]: 8.03999e-06
        [meta_fg_expand]: 1.52001e-06
        [replace_old_param]: 7.1e-06
        [inline_without_move]: 4.67e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 7.29982e-07
        [auto_monad_eliminator]: 5.14e-06
        [cse]: 1.007e-05
        [replace_applicator]: 5.07e-06
[py_interpret_to_execute_after_opt_a]: 1.065e-05
[rewriter_after_opt_a]: 2.925e-05
[convert_after_rewriter]: 6.66999e-06
[order_py_execute_after_rewriter]: 5.57001e-06
[mutable_eliminate]: 0.00046268
[jit_opt_b]: 4.79e-05, [1]
    [Cycle 1]: 4.14e-05, [2]
        [frontend_op_eliminate]: 1.577e-05
        [inline_after_opt_a]: 1.44e-05
[cconv]: 2.404e-05
[loop_unroll]: 0.00040778
[jit_opt_after_cconv]: 0.0001349, [1]
    [Cycle 1]: 0.00012843, [11]
        [c_1]: 2.084e-05
        [parameter_eliminate]: 2.27999e-06
        [updatestate_depend_eliminate]: 5.14e-06
        [updatestate_assign_eliminate]: 2.50002e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.559e-05
        [call_graph_tuple_transform]: 1.762e-05
        [tuple_list_get_item_eliminator]: 5.34998e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.43002e-06
[remove_dup_value]: 1.297e-05
[partial_unused_args_eliminate]: 2.00002e-06
[environ_conv]: 4.74002e-06
[add_recomputation]: 4.384e-05
[cse_after_recomputation]: 2.009e-05, [1]
    [Cycle 1]: 1.44e-05, [1]
        [cse]: 8.50001e-06
[auto_monad_reorder]: 1.459e-05
[get_jit_bprop_graph]: 1.42999e-06
[rewriter_after_jit_bprop_graph]: 4.73001e-06
[opt_after_jit_grad]: 0.00044705
[symbol_engine_optimizer]: 7.222e-05, [1]
    [Cycle 1]: 6.599e-05, [6]
        [build]: 3.26001e-06
        [elim_shapecalc]: 7.33999e-06
        [elim_not_effective]: 1.24e-05
        [opt_reshape]: 6.01e-06
        [fold_const_symbol]: 8.76997e-06
        [renormalize]: 4.69998e-07
[validate]: 2.917e-05
Sums
    bootstrap                            :     0.000260s :  3.52%
    type_inference                       :     0.003860s : 52.37%
    event_method                         :     0.000016s :  0.22%
    auto_monad                           :     0.000076s :  1.04%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000030s :  0.41%
    rewriter_before_opt_a                :     0.000071s :  0.97%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000055s :  0.74%
    jit_opt_a.loop_unroll                :     0.000035s :  0.47%
    jit_opt_a.a_1                        :     0.000582s :  7.90%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.05%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.36%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000457s :  6.21%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.24%
    jit_opt_a.cse                        :     0.000034s :  0.47%
    jit_opt_a.replace_applicator         :     0.000016s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.14%
    rewriter_after_opt_a                 :     0.000029s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000006s :  0.08%
    mutable_eliminate                    :     0.000463s :  6.28%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.20%
    cconv                                :     0.000024s :  0.33%
    loop_unroll                          :     0.000408s :  5.53%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000013s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000044s :  0.59%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000015s :  0.20%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000447s :  6.06%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000029s :  0.40%


Time group info:
------[substitution.]   0.000171    23
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000001s :      2: substitution.fold_const_symbol
 2.86% :     0.000005s :      3: substitution.graph_param_transform
82.84% :     0.000141s :      4: substitution.inline
 1.70% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.44% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.94% :     0.000003s :      2: substitution.replace_old_param
 6.44% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003802     2
80.40% :     0.003057s :      1: type_inference.infer
19.60% :     0.000745s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.01% :     0.000042s :      4: replace.inline
32.99% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000149     6
93.38% :     0.000139s :      4: match.inline
 6.62% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.13% :     0.000001s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.15% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.36% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.54% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.39% :     0.000002s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.37% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.17% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.64% :     0.000002s :      3: predicate.mutable_eliminate
 0.60% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.76% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.82% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.64% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.56% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.82% :     0.000009s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.57% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000481    12
49.97% :     0.000241s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.19% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.84% :     0.000216s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012176    72
 0.38% :     0.000047s :      1: add_recomputation
 0.66% :     0.000080s :      1: auto_monad
 0.14% :     0.000017s :      1: auto_monad_reorder
 2.24% :     0.000273s :      1: bootstrap
 0.22% :     0.000027s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.07% :     0.000008s :      1: graph_reusing
37.90% :     0.004615s :      1: jit_opt_a
 1.13% :     0.000138s :      1: jit_opt_after_cconv
 0.41% :     0.000050s :      1: jit_opt_b
 3.41% :     0.000415s :      1: loop_unroll
 3.86% :     0.000470s :      1: mutable_eliminate
 6.44% :     0.000784s :     26: opt.transform.jit_opt_a
 0.38% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000024s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.73% :     0.000454s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000033s :      1: py_interpret_to_execute
 0.11% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000015s :      1: remove_dup_value
 1.85% :     0.000225s :      1: renormalize.infer
 1.85% :     0.000225s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000032s :      1: rewriter_after_opt_a
 0.61% :     0.000074s :      1: rewriter_before_opt_a
 0.61% :     0.000075s :      1: symbol_engine_optimizer
31.80% :     0.003871s :      1: type_inference


TotalTime = 0.0170447, [33]
[bootstrap]: 0.0002732
[type_inference]: 0.00413994
[event_method]: 1.623e-05
[auto_monad]: 6.092e-05
[graph_reusing]: 5.92999e-06
[pre_auto_parallel]: 2.20002e-06
[py_interpret_to_execute]: 2.825e-05
[rewriter_before_opt_a]: 6.671e-05
[expand_dump_flag]: 2.77002e-06
[jit_opt_a]: 0.00486389, [2]
    [Cycle 1]: 0.0014135, [27]
        [switch_simplify]: 4.736e-05
        [loop_unroll]: 3.05e-05
        [a_1]: 0.00047242
        [with_stream_mark]: 1.262e-05
        [recompute_prepare]: 7.31001e-06
        [updatestate_depend_eliminate]: 2.94001e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.28002e-06
        [specialize_transform]: 6.04001e-06
        [updatestate_useless_node_eliminater]: 5.48002e-06
        [accelerated_algorithm]: 5.72001e-06
        [meta_shard_fg_expand]: 1.24e-06
        [get_grad_eliminate_]: 5.40001e-06
        [merge_forward]: 2.71e-06
        [cell_reuse_recompute_pass]: 1.09003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.19e-05
        [j_node_and_user_rematch]: 9.10001e-06
        [meta_fg_expand]: 1.60001e-06
        [replace_old_param]: 7.71001e-06
        [inline_without_move]: 4.92e-06
        [renormalize]: 0.00053016
        [add_forward_monad_depend]: 4.1e-06
        [auto_monad_grad]: 1.38002e-06
        [auto_monad_eliminator]: 9.07001e-06
        [cse]: 5.482e-05
        [replace_applicator]: 1.152e-05
    [Cycle 2]: 0.00030319, [27]
        [switch_simplify]: 5.47001e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 8.726e-05
        [with_stream_mark]: 8.55999e-06
        [recompute_prepare]: 4.92999e-06
        [updatestate_depend_eliminate]: 2.59001e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 9.29984e-07
        [specialize_transform]: 4.90999e-06
        [updatestate_useless_node_eliminater]: 4.69998e-06
        [accelerated_algorithm]: 4.47e-06
        [meta_shard_fg_expand]: 9.80013e-07
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.17001e-06
        [cell_reuse_recompute_pass]: 1.67001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.164e-05
        [j_node_and_user_rematch]: 7.8e-06
        [meta_fg_expand]: 1.26997e-06
        [replace_old_param]: 6.69001e-06
        [inline_without_move]: 4.42998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 9.5999e-07
        [auto_monad_eliminator]: 4.68999e-06
        [cse]: 9.61e-06
        [replace_applicator]: 5.19998e-06
[py_interpret_to_execute_after_opt_a]: 8.02e-06
[rewriter_after_opt_a]: 2.687e-05
[convert_after_rewriter]: 5.77999e-06
[order_py_execute_after_rewriter]: 4.05e-06
[mutable_eliminate]: 0.00049419
[jit_opt_b]: 4.695e-05, [1]
    [Cycle 1]: 4.049e-05, [2]
        [frontend_op_eliminate]: 1.603e-05
        [inline_after_opt_a]: 1.366e-05
[cconv]: 2.387e-05
[loop_unroll]: 0.00041117
[jit_opt_after_cconv]: 0.0001305, [1]
    [Cycle 1]: 0.00012452, [11]
        [c_1]: 2.026e-05
        [parameter_eliminate]: 2.79999e-06
        [updatestate_depend_eliminate]: 5.37001e-06
        [updatestate_assign_eliminate]: 2.19999e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [cse]: 1.61e-05
        [call_graph_tuple_transform]: 1.588e-05
        [tuple_list_get_item_eliminator]: 5.16002e-06
        [none_parameter_eliminate]: 9.80013e-07
        [renormalize]: 6.89994e-07
        [switch_simplify]: 5.32001e-06
[remove_dup_value]: 7.82e-06
[partial_unused_args_eliminate]: 1.67999e-06
[environ_conv]: 4.1e-06
[add_recomputation]: 4.847e-05
[cse_after_recomputation]: 2.122e-05, [1]
    [Cycle 1]: 1.577e-05, [1]
        [cse]: 9.67001e-06
[auto_monad_reorder]: 1.146e-05
[get_jit_bprop_graph]: 1.76e-06
[rewriter_after_jit_bprop_graph]: 5.44998e-06
[opt_after_jit_grad]: 0.00045481
[symbol_engine_optimizer]: 6.987e-05, [1]
    [Cycle 1]: 6.377e-05, [6]
        [build]: 3.03e-06
        [elim_shapecalc]: 6.91999e-06
        [elim_not_effective]: 1.256e-05
        [opt_reshape]: 5.64e-06
        [fold_const_symbol]: 8.31002e-06
        [renormalize]: 5.89993e-07
[validate]: 3.092e-05
[backend_pass]: 7.50006e-07
[task_emit]: 0.00558069
[execute]: 8.84e-06
Sums
    bootstrap                            :     0.000273s :  2.05%
    type_inference                       :     0.004140s : 31.06%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000061s :  0.46%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.21%
    rewriter_before_opt_a                :     0.000067s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.40%
    jit_opt_a.loop_unroll                :     0.000035s :  0.27%
    jit_opt_a.a_1                        :     0.000560s :  4.20%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000014s :  0.11%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000530s :  3.98%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.10%
    jit_opt_a.cse                        :     0.000064s :  0.48%
    jit_opt_a.replace_applicator         :     0.000017s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000027s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000494s :  3.71%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000411s :  3.08%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000008s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000048s :  0.36%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000011s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000455s :  3.41%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000031s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005581s : 41.87%
    execute                              :     0.000009s :  0.07%


Time group info:
------[substitution.]   0.000160    23
 0.82% :     0.000001s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.54% :     0.000004s :      3: substitution.graph_param_transform
84.24% :     0.000135s :      4: substitution.inline
 1.65% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.59% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.93% :     0.000003s :      2: substitution.replace_old_param
 5.53% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004088     2
82.57% :     0.003376s :      1: type_inference.infer
17.43% :     0.000712s :      1: type_inference.specialize
------[replace.]   0.000059     6
64.52% :     0.000038s :      4: replace.inline
35.48% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000140     6
94.43% :     0.000132s :      4: match.inline
 5.57% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.53% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000001s :     10: predicate.addn_zero_filter
 2.10% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.47% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.67% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000001s :      6: predicate.less_batch_normalization
 1.74% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.87% :     0.000002s :     15: predicate.load_eliminater
 1.66% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.40% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000002s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.87% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000001s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.19% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.59% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.89% :     0.000009s :     53: predicate.switch_simplify
 1.34% :     0.000001s :     10: predicate.tile_eliminate
 1.35% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.42% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.31% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000509    12
48.67% :     0.000248s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.74% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.59% :     0.000232s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018434    76
 0.28% :     0.000052s :      1: add_recomputation
 0.35% :     0.000064s :      1: auto_monad
 0.08% :     0.000014s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.54% :     0.000284s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.13% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.07% :     0.000013s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.40% :     0.004867s :      1: jit_opt_a
 0.72% :     0.000133s :      1: jit_opt_after_cconv
 0.27% :     0.000049s :      1: jit_opt_b
 2.27% :     0.000419s :      1: loop_unroll
 2.73% :     0.000503s :      1: mutable_eliminate
 4.12% :     0.000759s :     26: opt.transform.jit_opt_a
 0.24% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.51% :     0.000463s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000031s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000010s :      1: remove_dup_value
 1.53% :     0.000281s :      1: renormalize.infer
 1.32% :     0.000243s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000030s :      1: rewriter_after_opt_a
 0.38% :     0.000069s :      1: rewriter_before_opt_a
 0.40% :     0.000073s :      1: symbol_engine_optimizer
30.37% :     0.005599s :      1: task_emit
22.52% :     0.004151s :      1: type_inference
 0.26% :     0.000048s :      1: validate


TotalTime = 0.0471478, [33]
[bootstrap]: 0.0003244
[type_inference]: 0.01992
[event_method]: 4.567e-05
[auto_monad]: 8.739e-05
[graph_reusing]: 6.41e-06
[pre_auto_parallel]: 2.33998e-06
[py_interpret_to_execute]: 2.341e-05
[rewriter_before_opt_a]: 8.671e-05
[expand_dump_flag]: 3.38e-06
[jit_opt_a]: 0.0131584, [3]
    [Cycle 1]: 0.00620378, [27]
        [switch_simplify]: 6.909e-05
        [loop_unroll]: 3.796e-05
        [a_1]: 0.00087553
        [with_stream_mark]: 2.389e-05
        [recompute_prepare]: 1.802e-05
        [updatestate_depend_eliminate]: 7.5e-06
        [updatestate_assign_eliminate]: 5.64e-06
        [updatestate_loads_eliminate]: 5.31002e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 1.294e-05
        [updatestate_useless_node_eliminater]: 1.171e-05
        [accelerated_algorithm]: 1.249e-05
        [meta_shard_fg_expand]: 2.51998e-06
        [get_grad_eliminate_]: 1.202e-05
        [merge_forward]: 7.39002e-06
        [cell_reuse_recompute_pass]: 7.39994e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.309e-05
        [j_node_and_user_rematch]: 3.461e-05
        [meta_fg_expand]: 0.0015975
        [replace_old_param]: 5.682e-05
        [inline_without_move]: 5.201e-05
        [renormalize]: 0.00284384
        [add_forward_monad_depend]: 1.557e-05
        [auto_monad_grad]: 5.00999e-06
        [auto_monad_eliminator]: 4.999e-05
        [cse]: 0.00014805
        [replace_applicator]: 6.894e-05
    [Cycle 2]: 0.00231237, [27]
        [switch_simplify]: 3.725e-05
        [loop_unroll]: 3.662e-05
        [a_1]: 0.00115563
        [with_stream_mark]: 1.238e-05
        [recompute_prepare]: 9.79e-06
        [updatestate_depend_eliminate]: 4.63999e-06
        [updatestate_assign_eliminate]: 4.32998e-06
        [updatestate_loads_eliminate]: 3.65e-06
        [parameter_eliminate]: 1.32e-06
        [specialize_transform]: 8.58001e-06
        [updatestate_useless_node_eliminater]: 8.57e-06
        [accelerated_algorithm]: 8.74e-06
        [meta_shard_fg_expand]: 2.33998e-06
        [get_grad_eliminate_]: 8.07e-06
        [merge_forward]: 4.32e-06
        [cell_reuse_recompute_pass]: 1.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.706e-05
        [j_node_and_user_rematch]: 1.409e-05
        [meta_fg_expand]: 4.973e-05
        [replace_old_param]: 1.22e-05
        [inline_without_move]: 8.2e-06
        [renormalize]: 0.00067003
        [add_forward_monad_depend]: 6.13998e-06
        [auto_monad_grad]: 1.74e-06
        [auto_monad_eliminator]: 1.486e-05
        [cse]: 4.691e-05
        [replace_applicator]: 1.573e-05
    [Cycle 3]: 0.00049481, [27]
        [switch_simplify]: 8.65999e-06
        [loop_unroll]: 7.95998e-06
        [a_1]: 0.00019592
        [with_stream_mark]: 1.026e-05
        [recompute_prepare]: 8.22e-06
        [updatestate_depend_eliminate]: 4.38001e-06
        [updatestate_assign_eliminate]: 4.50999e-06
        [updatestate_loads_eliminate]: 3.58e-06
        [parameter_eliminate]: 1.14998e-06
        [specialize_transform]: 7.97998e-06
        [updatestate_useless_node_eliminater]: 7.55e-06
        [accelerated_algorithm]: 9.30001e-06
        [meta_shard_fg_expand]: 1.59e-06
        [get_grad_eliminate_]: 7.51999e-06
        [merge_forward]: 4.2e-06
        [cell_reuse_recompute_pass]: 2.06998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.638e-05
        [j_node_and_user_rematch]: 1.348e-05
        [meta_fg_expand]: 2.82002e-06
        [replace_old_param]: 1.031e-05
        [inline_without_move]: 7.65998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.54e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 1.066e-05
        [cse]: 2.415e-05
        [replace_applicator]: 8.09002e-06
[py_interpret_to_execute_after_opt_a]: 1.326e-05
[rewriter_after_opt_a]: 4.475e-05
[convert_after_rewriter]: 8.37998e-06
[order_py_execute_after_rewriter]: 6.01998e-06
[mutable_eliminate]: 0.00064002
[jit_opt_b]: 6.544e-05, [1]
    [Cycle 1]: 5.81e-05, [2]
        [frontend_op_eliminate]: 2.415e-05
        [inline_after_opt_a]: 2.274e-05
[cconv]: 2.156e-05
[loop_unroll]: 0.00043479
[jit_opt_after_cconv]: 0.00018717, [1]
    [Cycle 1]: 0.00018115, [11]
        [c_1]: 3.549e-05
        [parameter_eliminate]: 2.60002e-06
        [updatestate_depend_eliminate]: 6.97002e-06
        [updatestate_assign_eliminate]: 4.33001e-06
        [updatestate_loads_eliminate]: 3.86001e-06
        [cse]: 3.323e-05
        [call_graph_tuple_transform]: 2.55e-05
        [tuple_list_get_item_eliminator]: 8.26002e-06
        [none_parameter_eliminate]: 1.81e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 8.30999e-06
[remove_dup_value]: 2.158e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 1.645e-05
[add_recomputation]: 6.208e-05
[cse_after_recomputation]: 3.452e-05, [1]
    [Cycle 1]: 2.86e-05, [1]
        [cse]: 2.226e-05
[auto_monad_reorder]: 1.914e-05
[get_jit_bprop_graph]: 1.72999e-06
[rewriter_after_jit_bprop_graph]: 5.82001e-06
[opt_after_jit_grad]: 0.00051696
[symbol_engine_optimizer]: 9.174e-05, [1]
    [Cycle 1]: 8.606e-05, [6]
        [build]: 3.86999e-06
        [elim_shapecalc]: 1.185e-05
        [elim_not_effective]: 1.845e-05
        [opt_reshape]: 8.87999e-06
        [fold_const_symbol]: 1.523e-05
        [renormalize]: 4.59986e-07
[validate]: 4.463e-05
[backend_pass]: 7.80012e-07
[task_emit]: 0.011034
[execute]: 8.03001e-06
Sums
    bootstrap                            :     0.000324s :  0.77%
    type_inference                       :     0.019920s : 47.20%
    event_method                         :     0.000046s :  0.11%
    auto_monad                           :     0.000087s :  0.21%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000023s :  0.06%
    rewriter_before_opt_a                :     0.000087s :  0.21%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000115s :  0.27%
    jit_opt_a.loop_unroll                :     0.000083s :  0.20%
    jit_opt_a.a_1                        :     0.002227s :  5.28%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.11%
    jit_opt_a.recompute_prepare          :     0.000036s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000014s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000013s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000029s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000028s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000031s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000028s :  0.07%
    jit_opt_a.merge_forward              :     0.000016s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000057s :  0.13%
    jit_opt_a.j_node_and_user_rematch    :     0.000062s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001650s :  3.91%
    jit_opt_a.replace_old_param          :     0.000079s :  0.19%
    jit_opt_a.inline_without_move        :     0.000068s :  0.16%
    jit_opt_a.renormalize                :     0.003514s :  8.33%
    jit_opt_a.add_forward_monad_depend   :     0.000023s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000076s :  0.18%
    jit_opt_a.cse                        :     0.000219s :  0.52%
    jit_opt_a.replace_applicator         :     0.000093s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.03%
    rewriter_after_opt_a                 :     0.000045s :  0.11%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.01%
    mutable_eliminate                    :     0.000640s :  1.52%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000023s :  0.05%
    cconv                                :     0.000022s :  0.05%
    loop_unroll                          :     0.000435s :  1.03%
    jit_opt_after_cconv.c_1              :     0.000035s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000033s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000008s :  0.02%
    remove_dup_value                     :     0.000022s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000016s :  0.04%
    add_recomputation                    :     0.000062s :  0.15%
    cse_after_recomputation.cse          :     0.000022s :  0.05%
    auto_monad_reorder                   :     0.000019s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.01%
    opt_after_jit_grad                   :     0.000517s :  1.22%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000045s :  0.11%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011034s : 26.14%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000595   126
 5.94% :     0.000035s :      2: substitution.arithmetic_simplify
 0.43% :     0.000003s :      5: substitution.elim_not_effective
 0.38% :     0.000002s :      5: substitution.fold_const_symbol
 1.22% :     0.000007s :      6: substitution.graph_param_transform
56.02% :     0.000333s :     12: substitution.inline
 2.94% :     0.000018s :      2: substitution.inline_without_move
 3.67% :     0.000022s :     18: substitution.j_node_and_user_rematch
 2.01% :     0.000012s :      7: substitution.minmaximum_grad
 5.96% :     0.000035s :      5: substitution.partial_eliminate
 2.14% :     0.000013s :     18: substitution.remove_not_recompute_node
 4.59% :     0.000027s :     10: substitution.replace_applicator
 1.29% :     0.000008s :      7: substitution.replace_old_param
 0.46% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.73% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.56% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 6.67% :     0.000040s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.019840     2
93.73% :     0.018597s :      1: type_inference.infer
 6.27% :     0.001244s :      1: type_inference.specialize
------[replace.]   0.000143    19
64.52% :     0.000092s :     12: replace.inline
35.48% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000347    19
94.27% :     0.000327s :     12: match.inline
 5.73% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000409  2895
 1.57% :     0.000006s :     48: predicate.accumulaten_eliminater
 0.54% :     0.000002s :      6: predicate.ad_related_special_op_eliminate
 1.41% :     0.000006s :     48: predicate.addn_check_dump
 1.69% :     0.000007s :     48: predicate.addn_zero_filter
 2.13% :     0.000009s :     48: predicate.arithmetic_simplify
 1.70% :     0.000007s :     48: predicate.cast_eliminate
 0.25% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.44% :     0.000006s :     48: predicate.compare_switch_simplify
 1.64% :     0.000007s :     48: predicate.depend_value_elim
 1.47% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.51% :     0.000006s :     48: predicate.dict_get_item_eliminator
 1.55% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.33% :     0.000001s :      6: predicate.dumpgradient_eliminate
 0.17% :     0.000001s :      6: predicate.elim_not_effective
 0.26% :     0.000001s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.44% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.46% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.59% :     0.000007s :     48: predicate.environ_get_eliminate
 1.47% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.12% :     0.000000s :      6: predicate.fold_const_symbol
 0.89% :     0.000004s :     23: predicate.get_grad_eliminate
 0.18% :     0.000001s :      6: predicate.graph_param_transform
 4.32% :     0.000018s :     79: predicate.inline
 1.76% :     0.000007s :     44: predicate.inline_without_move
 0.40% :     0.000002s :     23: predicate.j_node_and_user_rematch
 1.09% :     0.000004s :     23: predicate.less_batch_normalization
 1.73% :     0.000007s :     55: predicate.list_to_tuple_eliminator_
 1.99% :     0.000008s :     61: predicate.load_eliminater
 0.56% :     0.000002s :      6: predicate.loop_unroll_after_grad
 3.00% :     0.000012s :     90: predicate.loop_unroll_before_grad
 1.92% :     0.000008s :     54: predicate.make_slice_get_slice_eliminator
 1.44% :     0.000006s :     48: predicate.merge_addn
 1.47% :     0.000006s :     48: predicate.minmaximum_grad
 0.56% :     0.000002s :      6: predicate.mutable_eliminate
 0.27% :     0.000001s :      6: predicate.opt_reshape
 2.27% :     0.000009s :     61: predicate.partial_eliminate
 1.54% :     0.000006s :     48: predicate.print_const_string_wrapper
 2.07% :     0.000008s :     48: predicate.reduce_eliminate
 1.90% :     0.000008s :     55: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.40% :     0.000010s :    108: predicate.replace_applicator
 0.85% :     0.000003s :     44: predicate.replace_old_param
 0.13% :     0.000001s :      6: predicate.reset_defer_inline
 1.62% :     0.000007s :     48: predicate.reshape_eliminate
 1.51% :     0.000006s :     48: predicate.row_tensor_add_zeros_like
 0.30% :     0.000001s :      6: predicate.row_tensor_eliminate
 1.59% :     0.000006s :     48: predicate.same_eliminate
 0.49% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.47% :     0.000002s :     12: predicate.special_op_eliminate
 0.91% :     0.000004s :     23: predicate.specialize_transform
 1.70% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.57% :     0.000006s :     48: predicate.stack_unstack_eliminate
 0.22% :     0.000001s :      6: predicate.switch_call_monad_eliminater
 2.58% :     0.000011s :     67: predicate.switch_defer_inline
 2.38% :     0.000010s :     67: predicate.switch_layer_defer_inline
 5.89% :     0.000024s :    163: predicate.switch_simplify
 1.64% :     0.000007s :     48: predicate.tile_eliminate
 1.49% :     0.000006s :     48: predicate.transpose_eliminate
 1.78% :     0.000007s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.61% :     0.000007s :     48: predicate.tuple_list_get_item_depend_reorder
 2.88% :     0.000012s :     67: predicate.tuple_list_get_item_eliminator
 1.90% :     0.000008s :     48: predicate.tuple_list_set_item_eliminator
 1.89% :     0.000008s :     55: predicate.tuple_to_list_eliminator_
 1.92% :     0.000008s :     61: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000012s :     84: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000007s :     48: predicate.value_based_eliminate
 0.19% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.26% :     0.000001s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001550    29
57.37% :     0.000889s :     13: func_graph_cloner_run.FuncGraphClonerGraph
42.63% :     0.000661s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.053733    91
 0.12% :     0.000065s :      1: add_recomputation
 0.17% :     0.000092s :      1: auto_monad
 0.04% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.63% :     0.000341s :      1: bootstrap
 0.04% :     0.000024s :      1: cconv
 0.02% :     0.000011s :      1: convert_after_rewriter
 0.07% :     0.000037s :      1: cse_after_recomputation
 0.03% :     0.000019s :      1: environ_conv
 0.09% :     0.000050s :      1: event_method
 0.02% :     0.000013s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
24.50% :     0.013163s :      1: jit_opt_a
 0.35% :     0.000190s :      1: jit_opt_after_cconv
 0.13% :     0.000068s :      1: jit_opt_b
 0.82% :     0.000442s :      1: loop_unroll
 1.21% :     0.000648s :      1: mutable_eliminate
 5.39% :     0.002894s :     39: opt.transform.jit_opt_a
 0.14% :     0.000074s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000041s :      4: opt.transform.jit_opt_b
 0.03% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.03% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000051s :      4: opt.transform.symbol_engine_opt
 0.98% :     0.000524s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.05% :     0.000026s :      1: py_interpret_to_execute
 0.03% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000024s :      1: remove_dup_value
 3.86% :     0.002072s :      2: renormalize.infer
 2.66% :     0.001428s :      2: renormalize.specialize
 0.01% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000048s :      1: rewriter_after_opt_a
 0.17% :     0.000089s :      1: rewriter_before_opt_a
 0.18% :     0.000095s :      1: symbol_engine_optimizer
20.57% :     0.011053s :      1: task_emit
37.10% :     0.019935s :      1: type_inference
 0.13% :     0.000068s :      1: validate


TotalTime = 0.0106836, [30]
[bootstrap]: 0.00024074
[type_inference]: 0.00386625
[event_method]: 1.698e-05
[auto_monad]: 5.103e-05
[graph_reusing]: 4.95999e-06
[pre_auto_parallel]: 2.14e-06
[py_interpret_to_execute]: 2.819e-05
[rewriter_before_opt_a]: 6.596e-05
[expand_dump_flag]: 2.44001e-06
[jit_opt_a]: 0.00447372, [2]
    [Cycle 1]: 0.00129388, [27]
        [switch_simplify]: 4.756e-05
        [loop_unroll]: 3.226e-05
        [a_1]: 0.00048684
        [with_stream_mark]: 1.21e-05
        [recompute_prepare]: 7.95e-06
        [updatestate_depend_eliminate]: 3.66999e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.37999e-06
        [specialize_transform]: 5.85002e-06
        [updatestate_useless_node_eliminater]: 5.08002e-06
        [accelerated_algorithm]: 5.61e-06
        [meta_shard_fg_expand]: 1.47999e-06
        [get_grad_eliminate_]: 5.62001e-06
        [merge_forward]: 2.59001e-06
        [cell_reuse_recompute_pass]: 9.89996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.349e-05
        [j_node_and_user_rematch]: 9.33002e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 8e-06
        [inline_without_move]: 5.54e-06
        [renormalize]: 0.0004254
        [add_forward_monad_depend]: 4.68001e-06
        [auto_monad_grad]: 1.96e-06
        [auto_monad_eliminator]: 1.058e-05
        [cse]: 1.618e-05
        [replace_applicator]: 1.085e-05
    [Cycle 2]: 0.00032157, [27]
        [switch_simplify]: 5.54998e-06
        [loop_unroll]: 5.04e-06
        [a_1]: 9.106e-05
        [with_stream_mark]: 8.15999e-06
        [recompute_prepare]: 5.05001e-06
        [updatestate_depend_eliminate]: 2.55002e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 8.80013e-07
        [specialize_transform]: 5.10001e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 5.38002e-06
        [meta_shard_fg_expand]: 1.23002e-06
        [get_grad_eliminate_]: 4.77998e-06
        [merge_forward]: 2.54001e-06
        [cell_reuse_recompute_pass]: 1.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.207e-05
        [j_node_and_user_rematch]: 8.02e-06
        [meta_fg_expand]: 1.43002e-06
        [replace_old_param]: 7.41999e-06
        [inline_without_move]: 4.88001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 4.70001e-06
        [cse]: 1.039e-05
        [replace_applicator]: 5.10001e-06
[py_interpret_to_execute_after_opt_a]: 7.31001e-06
[rewriter_after_opt_a]: 2.708e-05
[convert_after_rewriter]: 5.58997e-06
[order_py_execute_after_rewriter]: 4.63999e-06
[mutable_eliminate]: 0.00047929
[jit_opt_b]: 5.269e-05, [1]
    [Cycle 1]: 4.615e-05, [2]
        [frontend_op_eliminate]: 1.929e-05
        [inline_after_opt_a]: 1.456e-05
[cconv]: 1.881e-05
[loop_unroll]: 0.00041765
[jit_opt_after_cconv]: 0.00013558, [1]
    [Cycle 1]: 0.00012933, [11]
        [c_1]: 2.11e-05
        [parameter_eliminate]: 2.44001e-06
        [updatestate_depend_eliminate]: 5.22e-06
        [updatestate_assign_eliminate]: 2.50002e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 1.626e-05
        [call_graph_tuple_transform]: 1.664e-05
        [tuple_list_get_item_eliminator]: 5.40001e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 3.4002e-07
        [switch_simplify]: 5.34998e-06
[remove_dup_value]: 8.97e-06
[partial_unused_args_eliminate]: 1.84e-06
[environ_conv]: 4.53001e-06
[add_recomputation]: 3.532e-05
[cse_after_recomputation]: 1.965e-05, [1]
    [Cycle 1]: 1.455e-05, [1]
        [cse]: 8.92e-06
[auto_monad_reorder]: 1.099e-05
[get_jit_bprop_graph]: 1.35001e-06
[rewriter_after_jit_bprop_graph]: 4.57e-06
[opt_after_jit_grad]: 0.00044848
[symbol_engine_optimizer]: 7.097e-05, [1]
    [Cycle 1]: 6.475e-05, [6]
        [build]: 2.84001e-06
        [elim_shapecalc]: 7.5e-06
        [elim_not_effective]: 1.208e-05
        [opt_reshape]: 5.54998e-06
        [fold_const_symbol]: 8.47998e-06
        [renormalize]: 3.9002e-07
[validate]: 2.601e-05
Sums
    bootstrap                            :     0.000241s :  3.31%
    type_inference                       :     0.003866s : 53.14%
    event_method                         :     0.000017s :  0.23%
    auto_monad                           :     0.000051s :  0.70%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000028s :  0.39%
    rewriter_before_opt_a                :     0.000066s :  0.91%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000053s :  0.73%
    jit_opt_a.loop_unroll                :     0.000037s :  0.51%
    jit_opt_a.a_1                        :     0.000578s :  7.94%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.18%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.09%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.24%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000425s :  5.85%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.21%
    jit_opt_a.cse                        :     0.000027s :  0.37%
    jit_opt_a.replace_applicator         :     0.000016s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.10%
    rewriter_after_opt_a                 :     0.000027s :  0.37%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000479s :  6.59%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.27%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.20%
    cconv                                :     0.000019s :  0.26%
    loop_unroll                          :     0.000418s :  5.74%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.29%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000009s :  0.12%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000035s :  0.49%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000011s :  0.15%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000448s :  6.16%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000026s :  0.36%


Time group info:
------[substitution.]   0.000158    23
 1.06% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.55% :     0.000004s :      3: substitution.graph_param_transform
82.50% :     0.000130s :      4: substitution.inline
 1.77% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.66% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000003s :      2: substitution.replace_old_param
 6.63% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003814     2
80.41% :     0.003067s :      1: type_inference.infer
19.59% :     0.000747s :      1: type_inference.specialize
------[replace.]   0.000062     6
66.31% :     0.000041s :      4: replace.inline
33.69% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000137     6
93.17% :     0.000128s :      4: match.inline
 6.83% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.12% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.14% :     0.000001s :     10: predicate.addn_zero_filter
 2.12% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.41% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.08% :     0.000006s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.49% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.35% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.16% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.48% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000002s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.74% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.59% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 1.14% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.57% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.98% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.26% :     0.000001s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.51% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.12% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000478    12
49.24% :     0.000235s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.15% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.61% :     0.000218s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011965    72
 0.32% :     0.000038s :      1: add_recomputation
 0.45% :     0.000054s :      1: auto_monad
 0.11% :     0.000014s :      1: auto_monad_reorder
 2.10% :     0.000252s :      1: bootstrap
 0.18% :     0.000021s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.18% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.17% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
37.41% :     0.004476s :      1: jit_opt_a
 1.16% :     0.000138s :      1: jit_opt_after_cconv
 0.46% :     0.000055s :      1: jit_opt_b
 3.56% :     0.000426s :      1: loop_unroll
 4.07% :     0.000487s :      1: mutable_eliminate
 6.54% :     0.000783s :     26: opt.transform.jit_opt_a
 0.38% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.23% :     0.000028s :      4: opt.transform.jit_opt_b
 0.10% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.81% :     0.000456s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.04% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000031s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000011s :      1: remove_dup_value
 1.69% :     0.000202s :      1: renormalize.infer
 1.81% :     0.000217s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000030s :      1: rewriter_after_opt_a
 0.57% :     0.000069s :      1: rewriter_before_opt_a
 0.62% :     0.000074s :      1: symbol_engine_optimizer
32.41% :     0.003878s :      1: type_inference

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:20.772.116 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80cae50> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:20.772.779 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0179719, [33]
[bootstrap]: 0.00029221
[type_inference]: 0.00453312
[event_method]: 1.794e-05
[auto_monad]: 5.876e-05
[graph_reusing]: 4.75999e-06
[pre_auto_parallel]: 1.86998e-06
[py_interpret_to_execute]: 3.271e-05
[rewriter_before_opt_a]: 6.824e-05
[expand_dump_flag]: 2.84001e-06
[jit_opt_a]: 0.0050915, [2]
    [Cycle 1]: 0.00148387, [27]
        [switch_simplify]: 4.779e-05
        [loop_unroll]: 2.877e-05
        [a_1]: 0.00052911
        [with_stream_mark]: 1.475e-05
        [recompute_prepare]: 7.06001e-06
        [updatestate_depend_eliminate]: 3.48e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.83002e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 5.69e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.19998e-06
        [get_grad_eliminate_]: 5.50001e-06
        [merge_forward]: 2.56e-06
        [cell_reuse_recompute_pass]: 1.04998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.093e-05
        [j_node_and_user_rematch]: 8.37e-06
        [meta_fg_expand]: 1.78002e-06
        [replace_old_param]: 8.22e-06
        [inline_without_move]: 5.15001e-06
        [renormalize]: 0.00057401
        [add_forward_monad_depend]: 4.74e-06
        [auto_monad_grad]: 1.25999e-06
        [auto_monad_eliminator]: 1.072e-05
        [cse]: 1.977e-05
        [replace_applicator]: 1.12e-05
    [Cycle 2]: 0.000314, [27]
        [switch_simplify]: 5.57001e-06
        [loop_unroll]: 5.60001e-06
        [a_1]: 8.953e-05
        [with_stream_mark]: 9.10001e-06
        [recompute_prepare]: 4.82e-06
        [updatestate_depend_eliminate]: 2.34001e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.65001e-06
        [updatestate_useless_node_eliminater]: 4.61002e-06
        [accelerated_algorithm]: 4.95001e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 4.68999e-06
        [merge_forward]: 2.31998e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.179e-05
        [j_node_and_user_rematch]: 7.73999e-06
        [meta_fg_expand]: 1.44998e-06
        [replace_old_param]: 7.21001e-06
        [inline_without_move]: 4.53999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 4.83001e-06
        [cse]: 9.37001e-06
        [replace_applicator]: 6.91001e-06
[py_interpret_to_execute_after_opt_a]: 9.05001e-06
[rewriter_after_opt_a]: 2.922e-05
[convert_after_rewriter]: 6.09999e-06
[order_py_execute_after_rewriter]: 4.38999e-06
[mutable_eliminate]: 0.00056988
[jit_opt_b]: 5.446e-05, [1]
    [Cycle 1]: 4.727e-05, [2]
        [frontend_op_eliminate]: 1.876e-05
        [inline_after_opt_a]: 1.586e-05
[cconv]: 1.825e-05
[loop_unroll]: 0.00044237
[jit_opt_after_cconv]: 0.00014626, [1]
    [Cycle 1]: 0.00013931, [11]
        [c_1]: 2.117e-05
        [parameter_eliminate]: 2.88e-06
        [updatestate_depend_eliminate]: 5.78002e-06
        [updatestate_assign_eliminate]: 2.41998e-06
        [updatestate_loads_eliminate]: 2.19999e-06
        [cse]: 2.187e-05
        [call_graph_tuple_transform]: 1.854e-05
        [tuple_list_get_item_eliminator]: 5.84999e-06
        [none_parameter_eliminate]: 1.00999e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.43002e-06
[remove_dup_value]: 9.28002e-06
[partial_unused_args_eliminate]: 1.52001e-06
[environ_conv]: 4e-06
[add_recomputation]: 3.937e-05
[cse_after_recomputation]: 2.286e-05, [1]
    [Cycle 1]: 1.673e-05, [1]
        [cse]: 1.041e-05
[auto_monad_reorder]: 1.069e-05
[get_jit_bprop_graph]: 1.62001e-06
[rewriter_after_jit_bprop_graph]: 6.70998e-06
[opt_after_jit_grad]: 0.0004993
[symbol_engine_optimizer]: 7.548e-05, [1]
    [Cycle 1]: 6.787e-05, [6]
        [build]: 3.2e-06
        [elim_shapecalc]: 7.7e-06
        [elim_not_effective]: 1.32e-05
        [opt_reshape]: 5.76e-06
        [fold_const_symbol]: 8.65999e-06
        [renormalize]: 4.2998e-07
[validate]: 3.371e-05
[backend_pass]: 1.02998e-06
[task_emit]: 0.00567113
[execute]: 6.32001e-06
Sums
    bootstrap                            :     0.000292s :  2.08%
    type_inference                       :     0.004533s : 32.22%
    event_method                         :     0.000018s :  0.13%
    auto_monad                           :     0.000059s :  0.42%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000033s :  0.23%
    rewriter_before_opt_a                :     0.000068s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000619s :  4.40%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000574s :  4.08%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.11%
    jit_opt_a.cse                        :     0.000029s :  0.21%
    jit_opt_a.replace_applicator         :     0.000018s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000029s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000570s :  4.05%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000018s :  0.13%
    loop_unroll                          :     0.000442s :  3.14%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000009s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000039s :  0.28%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000011s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.05%
    opt_after_jit_grad                   :     0.000499s :  3.55%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005671s : 40.31%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000164    23
 1.07% :     0.000002s :      2: substitution.elim_not_effective
 0.95% :     0.000002s :      2: substitution.fold_const_symbol
 3.44% :     0.000006s :      3: substitution.graph_param_transform
83.15% :     0.000136s :      4: substitution.inline
 1.87% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.26% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000003s :      2: substitution.replace_old_param
 5.25% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004458     2
82.35% :     0.003671s :      1: type_inference.infer
17.65% :     0.000787s :      1: type_inference.specialize
------[replace.]   0.000063     6
65.66% :     0.000041s :      4: replace.inline
34.34% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000141     6
94.75% :     0.000134s :      4: match.inline
 5.25% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.80% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.65% :     0.000002s :     10: predicate.addn_zero_filter
 2.07% :     0.000002s :     10: predicate.arithmetic_simplify
 1.52% :     0.000002s :     10: predicate.cast_eliminate
 0.57% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000001s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 6.13% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.64% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.76% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000002s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.97% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.32% :     0.000002s :     10: predicate.reshape_eliminate
 1.52% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.89% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.68% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.30% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.09% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000516    12
48.52% :     0.000251s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.20% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.28% :     0.000239s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019477    76
 0.22% :     0.000042s :      1: add_recomputation
 0.32% :     0.000062s :      1: auto_monad
 0.07% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.59% :     0.000310s :      1: bootstrap
 0.11% :     0.000021s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.13% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.12% :     0.000022s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.16% :     0.005094s :      1: jit_opt_a
 0.76% :     0.000149s :      1: jit_opt_after_cconv
 0.29% :     0.000057s :      1: jit_opt_b
 2.31% :     0.000451s :      1: loop_unroll
 2.98% :     0.000581s :      1: mutable_eliminate
 4.19% :     0.000817s :     26: opt.transform.jit_opt_a
 0.24% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.63% :     0.000511s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000035s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000012s :      1: remove_dup_value
 1.52% :     0.000295s :      1: renormalize.infer
 1.40% :     0.000272s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000032s :      1: rewriter_after_opt_a
 0.37% :     0.000071s :      1: rewriter_before_opt_a
 0.40% :     0.000078s :      1: symbol_engine_optimizer
29.19% :     0.005685s :      1: task_emit
23.37% :     0.004551s :      1: type_inference
 0.27% :     0.000053s :      1: validate


TotalTime = 0.0106063, [30]
[bootstrap]: 0.00024302
[type_inference]: 0.00374159
[event_method]: 1.562e-05
[auto_monad]: 5.392e-05
[graph_reusing]: 4.80001e-06
[pre_auto_parallel]: 1.91e-06
[py_interpret_to_execute]: 2.817e-05
[rewriter_before_opt_a]: 6.404e-05
[expand_dump_flag]: 2.52001e-06
[jit_opt_a]: 0.00452431, [2]
    [Cycle 1]: 0.00133063, [27]
        [switch_simplify]: 4.625e-05
        [loop_unroll]: 2.906e-05
        [a_1]: 0.00049945
        [with_stream_mark]: 1.367e-05
        [recompute_prepare]: 6.97002e-06
        [updatestate_depend_eliminate]: 3.46001e-06
        [updatestate_assign_eliminate]: 3.20998e-06
        [updatestate_loads_eliminate]: 2.11998e-06
        [parameter_eliminate]: 1.28002e-06
        [specialize_transform]: 5.81e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.63002e-06
        [meta_shard_fg_expand]: 1.96e-06
        [get_grad_eliminate_]: 4.95999e-06
        [merge_forward]: 3.13998e-06
        [cell_reuse_recompute_pass]: 1.13001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.367e-05
        [j_node_and_user_rematch]: 9.39998e-06
        [meta_fg_expand]: 2.44001e-06
        [replace_old_param]: 9.36e-06
        [inline_without_move]: 5.76e-06
        [renormalize]: 0.00044591
        [add_forward_monad_depend]: 4.38001e-06
        [auto_monad_grad]: 1.87001e-06
        [auto_monad_eliminator]: 1.241e-05
        [cse]: 2.009e-05
        [replace_applicator]: 1.052e-05
    [Cycle 2]: 0.00031769, [27]
        [switch_simplify]: 5.87001e-06
        [loop_unroll]: 5.00001e-06
        [a_1]: 9.099e-05
        [with_stream_mark]: 8.3e-06
        [recompute_prepare]: 5.21002e-06
        [updatestate_depend_eliminate]: 2.51998e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 5.07e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 5.24e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.58002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.186e-05
        [j_node_and_user_rematch]: 7.66999e-06
        [meta_fg_expand]: 1.55001e-06
        [replace_old_param]: 6.89999e-06
        [inline_without_move]: 4.73001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 4.92e-06
        [cse]: 1.015e-05
        [replace_applicator]: 4.94e-06
[py_interpret_to_execute_after_opt_a]: 8e-06
[rewriter_after_opt_a]: 2.575e-05
[convert_after_rewriter]: 6.36998e-06
[order_py_execute_after_rewriter]: 5.10999e-06
[mutable_eliminate]: 0.00046951
[jit_opt_b]: 4.939e-05, [1]
    [Cycle 1]: 4.287e-05, [2]
        [frontend_op_eliminate]: 1.664e-05
        [inline_after_opt_a]: 1.442e-05
[cconv]: 2.219e-05
[loop_unroll]: 0.00041372
[jit_opt_after_cconv]: 0.0001328, [1]
    [Cycle 1]: 0.00012651, [11]
        [c_1]: 2.012e-05
        [parameter_eliminate]: 2.37999e-06
        [updatestate_depend_eliminate]: 5.29e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.579e-05
        [call_graph_tuple_transform]: 1.751e-05
        [tuple_list_get_item_eliminator]: 5.34e-06
        [none_parameter_eliminate]: 1.19e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 5.17e-06
[remove_dup_value]: 1.136e-05
[partial_unused_args_eliminate]: 2.48e-06
[environ_conv]: 4.75999e-06
[add_recomputation]: 3.924e-05
[cse_after_recomputation]: 2.01e-05, [1]
    [Cycle 1]: 1.463e-05, [1]
        [cse]: 8.85999e-06
[auto_monad_reorder]: 1.312e-05
[get_jit_bprop_graph]: 1.53002e-06
[rewriter_after_jit_bprop_graph]: 4.65001e-06
[opt_after_jit_grad]: 0.00044583
[symbol_engine_optimizer]: 7.263e-05, [1]
    [Cycle 1]: 6.654e-05, [6]
        [build]: 3.11999e-06
        [elim_shapecalc]: 7.95e-06
        [elim_not_effective]: 1.23e-05
        [opt_reshape]: 5.81998e-06
        [fold_const_symbol]: 8.75999e-06
        [renormalize]: 4.09986e-07
[validate]: 2.814e-05
Sums
    bootstrap                            :     0.000243s :  3.38%
    type_inference                       :     0.003742s : 52.08%
    event_method                         :     0.000016s :  0.22%
    auto_monad                           :     0.000054s :  0.75%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000028s :  0.39%
    rewriter_before_opt_a                :     0.000064s :  0.89%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000052s :  0.73%
    jit_opt_a.loop_unroll                :     0.000034s :  0.47%
    jit_opt_a.a_1                        :     0.000590s :  8.22%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.08%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.36%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.24%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.06%
    jit_opt_a.replace_old_param          :     0.000016s :  0.23%
    jit_opt_a.inline_without_move        :     0.000010s :  0.15%
    jit_opt_a.renormalize                :     0.000446s :  6.21%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.24%
    jit_opt_a.cse                        :     0.000030s :  0.42%
    jit_opt_a.replace_applicator         :     0.000015s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000026s :  0.36%
    convert_after_rewriter               :     0.000006s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000470s :  6.53%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.20%
    cconv                                :     0.000022s :  0.31%
    loop_unroll                          :     0.000414s :  5.76%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000011s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000039s :  0.55%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000013s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000446s :  6.21%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000028s :  0.39%


Time group info:
------[substitution.]   0.000178    23
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.67% :     0.000005s :      3: substitution.graph_param_transform
83.64% :     0.000149s :      4: substitution.inline
 1.63% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.40% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.98% :     0.000004s :      2: substitution.replace_old_param
 6.03% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003688     2
79.98% :     0.002950s :      1: type_inference.infer
20.02% :     0.000738s :      1: type_inference.specialize
------[replace.]   0.000060     6
66.80% :     0.000040s :      4: replace.inline
33.20% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000156     6
93.79% :     0.000146s :      4: match.inline
 6.21% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.68% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.24% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000002s :     10: predicate.addn_zero_filter
 2.19% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.16% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.22% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.42% :     0.000006s :     22: predicate.inline
 1.09% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.09% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.64% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.32% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.51% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.46% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.57% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.35% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.88% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000469    12
48.79% :     0.000229s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.29% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.92% :     0.000216s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011911    72
 0.35% :     0.000042s :      1: add_recomputation
 0.48% :     0.000057s :      1: auto_monad
 0.13% :     0.000016s :      1: auto_monad_reorder
 2.13% :     0.000254s :      1: bootstrap
 0.21% :     0.000025s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
38.00% :     0.004527s :      1: jit_opt_a
 1.14% :     0.000136s :      1: jit_opt_after_cconv
 0.44% :     0.000052s :      1: jit_opt_b
 3.54% :     0.000421s :      1: loop_unroll
 4.00% :     0.000477s :      1: mutable_eliminate
 6.63% :     0.000790s :     26: opt.transform.jit_opt_a
 0.37% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000025s :      4: opt.transform.jit_opt_b
 0.10% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.13% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.27% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.81% :     0.000454s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.04% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000031s :      1: py_interpret_to_execute
 0.09% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000014s :      1: remove_dup_value
 1.83% :     0.000218s :      1: renormalize.infer
 1.86% :     0.000221s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000029s :      1: rewriter_after_opt_a
 0.56% :     0.000067s :      1: rewriter_before_opt_a
 0.63% :     0.000075s :      1: symbol_engine_optimizer
31.51% :     0.003753s :      1: type_inference


TotalTime = 0.0174626, [33]
[bootstrap]: 0.0002922
[type_inference]: 0.00430163
[event_method]: 1.641e-05
[auto_monad]: 6.243e-05
[graph_reusing]: 5.90002e-06
[pre_auto_parallel]: 2.51998e-06
[py_interpret_to_execute]: 5.231e-05
[rewriter_before_opt_a]: 8.999e-05
[expand_dump_flag]: 3.11001e-06
[jit_opt_a]: 0.00492418, [2]
    [Cycle 1]: 0.00134167, [27]
        [switch_simplify]: 4.521e-05
        [loop_unroll]: 2.88e-05
        [a_1]: 0.00047387
        [with_stream_mark]: 1.232e-05
        [recompute_prepare]: 6.67002e-06
        [updatestate_depend_eliminate]: 3.08e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.38002e-06
        [specialize_transform]: 5.81e-06
        [updatestate_useless_node_eliminater]: 4.92999e-06
        [accelerated_algorithm]: 5.00999e-06
        [meta_shard_fg_expand]: 1.54e-06
        [get_grad_eliminate_]: 4.99e-06
        [merge_forward]: 2.29999e-06
        [cell_reuse_recompute_pass]: 7.50006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.041e-05
        [j_node_and_user_rematch]: 7.96001e-06
        [meta_fg_expand]: 1.66998e-06
        [replace_old_param]: 7.68001e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.00050394
        [add_forward_monad_depend]: 4.90001e-06
        [auto_monad_grad]: 1.18001e-06
        [auto_monad_eliminator]: 1.014e-05
        [cse]: 1.634e-05
        [replace_applicator]: 1.158e-05
    [Cycle 2]: 0.0003102, [27]
        [switch_simplify]: 5.51002e-06
        [loop_unroll]: 4.88001e-06
        [a_1]: 8.75e-05
        [with_stream_mark]: 7.98999e-06
        [recompute_prepare]: 4.79e-06
        [updatestate_depend_eliminate]: 2.53e-06
        [updatestate_assign_eliminate]: 2.10002e-06
        [updatestate_loads_eliminate]: 1.98997e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 4.87998e-06
        [updatestate_useless_node_eliminater]: 4.72e-06
        [accelerated_algorithm]: 5.05999e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 4.32003e-06
        [merge_forward]: 2.32999e-06
        [cell_reuse_recompute_pass]: 1.62999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.137e-05
        [j_node_and_user_rematch]: 7.63999e-06
        [meta_fg_expand]: 1.45001e-06
        [replace_old_param]: 6.96001e-06
        [inline_without_move]: 4.89e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 8.49977e-07
        [auto_monad_eliminator]: 4.79e-06
        [cse]: 1.004e-05
        [replace_applicator]: 4.81002e-06
[py_interpret_to_execute_after_opt_a]: 9.37001e-06
[rewriter_after_opt_a]: 2.561e-05
[convert_after_rewriter]: 5.67999e-06
[order_py_execute_after_rewriter]: 4.33999e-06
[mutable_eliminate]: 0.00048483
[jit_opt_b]: 4.828e-05, [1]
    [Cycle 1]: 4.174e-05, [2]
        [frontend_op_eliminate]: 1.58e-05
        [inline_after_opt_a]: 1.468e-05
[cconv]: 1.657e-05
[loop_unroll]: 0.00040526
[jit_opt_after_cconv]: 0.00012844, [1]
    [Cycle 1]: 0.00012227, [11]
        [c_1]: 1.926e-05
        [parameter_eliminate]: 2.49001e-06
        [updatestate_depend_eliminate]: 5.72999e-06
        [updatestate_assign_eliminate]: 2.20002e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [cse]: 1.605e-05
        [call_graph_tuple_transform]: 1.555e-05
        [tuple_list_get_item_eliminator]: 5.00001e-06
        [none_parameter_eliminate]: 9.39996e-07
        [renormalize]: 5.39992e-07
        [switch_simplify]: 5.07e-06
[remove_dup_value]: 6.51e-06
[partial_unused_args_eliminate]: 1.45001e-06
[environ_conv]: 3.47997e-06
[add_recomputation]: 3.678e-05
[cse_after_recomputation]: 1.918e-05, [1]
    [Cycle 1]: 1.44e-05, [1]
        [cse]: 8.95001e-06
[auto_monad_reorder]: 1.011e-05
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 4.77998e-06
[opt_after_jit_grad]: 0.00048659
[symbol_engine_optimizer]: 6.73e-05, [1]
    [Cycle 1]: 6.107e-05, [6]
        [build]: 2.61e-06
        [elim_shapecalc]: 6.88998e-06
        [elim_not_effective]: 1.158e-05
        [opt_reshape]: 5.14e-06
        [fold_const_symbol]: 7.48e-06
        [renormalize]: 5.19998e-07
[validate]: 2.859e-05
[backend_pass]: 7.39994e-07
[task_emit]: 0.00571782
[execute]: 9.06998e-06
Sums
    bootstrap                            :     0.000292s :  2.15%
    type_inference                       :     0.004302s : 31.60%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000062s :  0.46%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000052s :  0.38%
    rewriter_before_opt_a                :     0.000090s :  0.66%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000051s :  0.37%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000561s :  4.12%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000504s :  3.70%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000026s :  0.19%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000026s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000485s :  3.56%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000017s :  0.12%
    loop_unroll                          :     0.000405s :  2.98%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000007s :  0.05%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000003s :  0.03%
    add_recomputation                    :     0.000037s :  0.27%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000010s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000487s :  3.57%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000007s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000029s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005718s : 42.00%
    execute                              :     0.000009s :  0.07%


Time group info:
------[substitution.]   0.000158    23
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.30% :     0.000004s :      3: substitution.graph_param_transform
84.45% :     0.000134s :      4: substitution.inline
 1.45% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.43% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.31% :     0.000004s :      2: substitution.replace_old_param
 5.46% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004241     2
81.97% :     0.003476s :      1: type_inference.infer
18.03% :     0.000765s :      1: type_inference.specialize
------[replace.]   0.000062     6
65.97% :     0.000041s :      4: replace.inline
34.03% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000139     6
94.51% :     0.000131s :      4: match.inline
 5.49% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.30% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.74% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     10: predicate.addn_check_dump
 1.56% :     0.000002s :     10: predicate.addn_zero_filter
 2.06% :     0.000002s :     10: predicate.arithmetic_simplify
 1.44% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000001s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.50% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.02% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.97% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000000s :      3: predicate.opt_reshape
 2.42% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.69% :     0.000002s :     10: predicate.reduce_eliminate
 1.80% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.83% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.32% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.49% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.55% :     0.000003s :     16: predicate.switch_defer_inline
 2.27% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.67% :     0.000008s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.30% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.37% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000527    12
50.30% :     0.000265s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.19% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.51% :     0.000235s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018817    76
 0.21% :     0.000040s :      1: add_recomputation
 0.35% :     0.000066s :      1: auto_monad
 0.07% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.62% :     0.000305s :      1: bootstrap
 0.10% :     0.000019s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.07% :     0.000014s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.18% :     0.004927s :      1: jit_opt_a
 0.70% :     0.000131s :      1: jit_opt_after_cconv
 0.27% :     0.000051s :      1: jit_opt_b
 2.19% :     0.000413s :      1: loop_unroll
 2.62% :     0.000493s :      1: mutable_eliminate
 4.00% :     0.000753s :     26: opt.transform.jit_opt_a
 0.22% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000028s :      4: opt.transform.symbol_engine_opt
 2.63% :     0.000495s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.29% :     0.000055s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000009s :      1: remove_dup_value
 1.34% :     0.000251s :      1: renormalize.infer
 1.31% :     0.000246s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000029s :      1: rewriter_after_opt_a
 0.49% :     0.000093s :      1: rewriter_before_opt_a
 0.37% :     0.000070s :      1: symbol_engine_optimizer
30.48% :     0.005735s :      1: task_emit
22.94% :     0.004316s :      1: type_inference
 0.24% :     0.000045s :      1: validate


TotalTime = 0.0359517, [33]
[bootstrap]: 0.00021584
[type_inference]: 0.0117358
[event_method]: 5.24e-05
[auto_monad]: 9.75e-05
[graph_reusing]: 7.25e-06
[pre_auto_parallel]: 3.28e-06
[py_interpret_to_execute]: 3.294e-05
[rewriter_before_opt_a]: 9.708e-05
[expand_dump_flag]: 4.34002e-06
[jit_opt_a]: 0.0132616, [3]
    [Cycle 1]: 0.00629964, [27]
        [switch_simplify]: 8.551e-05
        [loop_unroll]: 4.591e-05
        [a_1]: 0.00092196
        [with_stream_mark]: 3.692e-05
        [recompute_prepare]: 2.035e-05
        [updatestate_depend_eliminate]: 9.14e-06
        [updatestate_assign_eliminate]: 6.53e-06
        [updatestate_loads_eliminate]: 6.49001e-06
        [parameter_eliminate]: 2.83e-06
        [specialize_transform]: 1.271e-05
        [updatestate_useless_node_eliminater]: 1.218e-05
        [accelerated_algorithm]: 1.249e-05
        [meta_shard_fg_expand]: 4.02e-06
        [get_grad_eliminate_]: 1.191e-05
        [merge_forward]: 9.41e-06
        [cell_reuse_recompute_pass]: 9.79984e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.767e-05
        [j_node_and_user_rematch]: 2.752e-05
        [meta_fg_expand]: 0.00158244
        [replace_old_param]: 5.713e-05
        [inline_without_move]: 5.412e-05
        [renormalize]: 0.00282828
        [add_forward_monad_depend]: 1.106e-05
        [auto_monad_grad]: 4.75001e-06
        [auto_monad_eliminator]: 5.573e-05
        [cse]: 0.00015997
        [replace_applicator]: 7.576e-05
    [Cycle 2]: 0.00243755, [27]
        [switch_simplify]: 3.846e-05
        [loop_unroll]: 3.661e-05
        [a_1]: 0.00116223
        [with_stream_mark]: 1.603e-05
        [recompute_prepare]: 9.56e-06
        [updatestate_depend_eliminate]: 6.49999e-06
        [updatestate_assign_eliminate]: 5.52001e-06
        [updatestate_loads_eliminate]: 4.42e-06
        [parameter_eliminate]: 2.32001e-06
        [specialize_transform]: 8.90999e-06
        [updatestate_useless_node_eliminater]: 7.95998e-06
        [accelerated_algorithm]: 8.85999e-06
        [meta_shard_fg_expand]: 2.86e-06
        [get_grad_eliminate_]: 9.02999e-06
        [merge_forward]: 4.89e-06
        [cell_reuse_recompute_pass]: 2.07999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.04e-05
        [j_node_and_user_rematch]: 1.421e-05
        [meta_fg_expand]: 4.988e-05
        [replace_old_param]: 1.403e-05
        [inline_without_move]: 8.27998e-06
        [renormalize]: 0.00075061
        [add_forward_monad_depend]: 6.09999e-06
        [auto_monad_grad]: 1.67999e-06
        [auto_monad_eliminator]: 1.74e-05
        [cse]: 5.384e-05
        [replace_applicator]: 1.724e-05
    [Cycle 3]: 0.0005074, [27]
        [switch_simplify]: 9.54e-06
        [loop_unroll]: 8.27e-06
        [a_1]: 0.00019863
        [with_stream_mark]: 1.214e-05
        [recompute_prepare]: 8.56002e-06
        [updatestate_depend_eliminate]: 4.73001e-06
        [updatestate_assign_eliminate]: 4.37e-06
        [updatestate_loads_eliminate]: 3.97998e-06
        [parameter_eliminate]: 1.70001e-06
        [specialize_transform]: 8.55001e-06
        [updatestate_useless_node_eliminater]: 7.73999e-06
        [accelerated_algorithm]: 8.82999e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 7.63001e-06
        [merge_forward]: 4.52e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.854e-05
        [j_node_and_user_rematch]: 1.393e-05
        [meta_fg_expand]: 2.91999e-06
        [replace_old_param]: 1.023e-05
        [inline_without_move]: 7.65e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.64998e-06
        [auto_monad_grad]: 8.10018e-07
        [auto_monad_eliminator]: 1.113e-05
        [cse]: 2.433e-05
        [replace_applicator]: 8.21002e-06
[py_interpret_to_execute_after_opt_a]: 1.432e-05
[rewriter_after_opt_a]: 5.118e-05
[convert_after_rewriter]: 8.72e-06
[order_py_execute_after_rewriter]: 6.56999e-06
[mutable_eliminate]: 0.00064792
[jit_opt_b]: 6.796e-05, [1]
    [Cycle 1]: 5.944e-05, [2]
        [frontend_op_eliminate]: 2.462e-05
        [inline_after_opt_a]: 2.352e-05
[cconv]: 2.61e-05
[loop_unroll]: 0.00043412
[jit_opt_after_cconv]: 0.0001937, [1]
    [Cycle 1]: 0.00018685, [11]
        [c_1]: 3.598e-05
        [parameter_eliminate]: 2.58e-06
        [updatestate_depend_eliminate]: 7.45998e-06
        [updatestate_assign_eliminate]: 4.38001e-06
        [updatestate_loads_eliminate]: 3.85998e-06
        [cse]: 3.389e-05
        [call_graph_tuple_transform]: 2.697e-05
        [tuple_list_get_item_eliminator]: 8.54e-06
        [none_parameter_eliminate]: 1.77001e-06
        [renormalize]: 5.19998e-07
        [switch_simplify]: 8.71002e-06
[remove_dup_value]: 2.547e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 8.37e-06
[add_recomputation]: 5.886e-05
[cse_after_recomputation]: 3.542e-05, [1]
    [Cycle 1]: 2.92e-05, [1]
        [cse]: 2.298e-05
[auto_monad_reorder]: 2.299e-05
[get_jit_bprop_graph]: 1.77001e-06
[rewriter_after_jit_bprop_graph]: 6.57002e-06
[opt_after_jit_grad]: 0.00048154
[symbol_engine_optimizer]: 9.514e-05, [1]
    [Cycle 1]: 8.828e-05, [6]
        [build]: 4.77e-06
        [elim_shapecalc]: 1.136e-05
        [elim_not_effective]: 1.905e-05
        [opt_reshape]: 9.09e-06
        [fold_const_symbol]: 1.506e-05
        [renormalize]: 4.69998e-07
[validate]: 4.707e-05
[backend_pass]: 8.2e-07
[task_emit]: 0.00793528
[execute]: 6.31e-06
Sums
    bootstrap                            :     0.000216s :  0.70%
    type_inference                       :     0.011736s : 37.79%
    event_method                         :     0.000052s :  0.17%
    auto_monad                           :     0.000098s :  0.31%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000033s :  0.11%
    rewriter_before_opt_a                :     0.000097s :  0.31%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000134s :  0.43%
    jit_opt_a.loop_unroll                :     0.000091s :  0.29%
    jit_opt_a.a_1                        :     0.002283s :  7.35%
    jit_opt_a.with_stream_mark           :     0.000065s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000038s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000020s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000016s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000015s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000007s :  0.02%
    jit_opt_a.specialize_transform       :     0.000030s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000028s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000030s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000009s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000029s :  0.09%
    jit_opt_a.merge_forward              :     0.000019s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000067s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000056s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001635s :  5.27%
    jit_opt_a.replace_old_param          :     0.000081s :  0.26%
    jit_opt_a.inline_without_move        :     0.000070s :  0.23%
    jit_opt_a.renormalize                :     0.003579s : 11.52%
    jit_opt_a.add_forward_monad_depend   :     0.000019s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000084s :  0.27%
    jit_opt_a.cse                        :     0.000238s :  0.77%
    jit_opt_a.replace_applicator         :     0.000101s :  0.33%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.05%
    rewriter_after_opt_a                 :     0.000051s :  0.16%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000648s :  2.09%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.08%
    cconv                                :     0.000026s :  0.08%
    loop_unroll                          :     0.000434s :  1.40%
    jit_opt_after_cconv.c_1              :     0.000036s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000034s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.03%
    remove_dup_value                     :     0.000025s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000008s :  0.03%
    add_recomputation                    :     0.000059s :  0.19%
    cse_after_recomputation.cse          :     0.000023s :  0.07%
    auto_monad_reorder                   :     0.000023s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000482s :  1.55%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000047s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007935s : 25.55%
    execute                              :     0.000006s :  0.02%


Time group info:
------[substitution.]   0.000605   126
 6.52% :     0.000039s :      2: substitution.arithmetic_simplify
 0.51% :     0.000003s :      5: substitution.elim_not_effective
 0.44% :     0.000003s :      5: substitution.fold_const_symbol
 1.23% :     0.000007s :      6: substitution.graph_param_transform
60.14% :     0.000364s :     12: substitution.inline
 2.98% :     0.000018s :      2: substitution.inline_without_move
 2.43% :     0.000015s :     18: substitution.j_node_and_user_rematch
 1.96% :     0.000012s :      7: substitution.minmaximum_grad
 1.24% :     0.000007s :      5: substitution.partial_eliminate
 2.22% :     0.000013s :     18: substitution.remove_not_recompute_node
 4.66% :     0.000028s :     10: substitution.replace_applicator
 1.40% :     0.000008s :      7: substitution.replace_old_param
 0.55% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.84% :     0.000023s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.49% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 7.38% :     0.000045s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011651     2
88.51% :     0.010313s :      1: type_inference.infer
11.49% :     0.001338s :      1: type_inference.specialize
------[replace.]   0.000157    19
66.66% :     0.000105s :     12: replace.inline
33.34% :     0.000052s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000380    19
94.07% :     0.000358s :     12: match.inline
 5.93% :     0.000023s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000428  2895
 1.47% :     0.000006s :     48: predicate.accumulaten_eliminater
 0.56% :     0.000002s :      6: predicate.ad_related_special_op_eliminate
 1.34% :     0.000006s :     48: predicate.addn_check_dump
 1.66% :     0.000007s :     48: predicate.addn_zero_filter
 2.40% :     0.000010s :     48: predicate.arithmetic_simplify
 1.63% :     0.000007s :     48: predicate.cast_eliminate
 0.26% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.35% :     0.000006s :     48: predicate.compare_switch_simplify
 1.62% :     0.000007s :     48: predicate.depend_value_elim
 1.44% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.47% :     0.000006s :     48: predicate.dict_get_item_eliminator
 1.46% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.38% :     0.000002s :      6: predicate.dumpgradient_eliminate
 0.14% :     0.000001s :      6: predicate.elim_not_effective
 0.24% :     0.000001s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.39% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.41% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.43% :     0.000006s :     48: predicate.environ_get_eliminate
 1.56% :     0.000007s :     48: predicate.environ_get_set_eliminate
 0.12% :     0.000000s :      6: predicate.fold_const_symbol
 0.92% :     0.000004s :     23: predicate.get_grad_eliminate
 0.12% :     0.000001s :      6: predicate.graph_param_transform
 4.33% :     0.000019s :     79: predicate.inline
 1.69% :     0.000007s :     44: predicate.inline_without_move
 0.40% :     0.000002s :     23: predicate.j_node_and_user_rematch
 1.03% :     0.000004s :     23: predicate.less_batch_normalization
 1.76% :     0.000008s :     55: predicate.list_to_tuple_eliminator_
 1.89% :     0.000008s :     61: predicate.load_eliminater
 0.58% :     0.000002s :      6: predicate.loop_unroll_after_grad
 4.28% :     0.000018s :     90: predicate.loop_unroll_before_grad
 1.86% :     0.000008s :     54: predicate.make_slice_get_slice_eliminator
 1.35% :     0.000006s :     48: predicate.merge_addn
 1.41% :     0.000006s :     48: predicate.minmaximum_grad
 0.73% :     0.000003s :      6: predicate.mutable_eliminate
 0.28% :     0.000001s :      6: predicate.opt_reshape
 2.28% :     0.000010s :     61: predicate.partial_eliminate
 1.38% :     0.000006s :     48: predicate.print_const_string_wrapper
 1.97% :     0.000008s :     48: predicate.reduce_eliminate
 1.79% :     0.000008s :     55: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.57% :     0.000011s :    108: predicate.replace_applicator
 1.01% :     0.000004s :     44: predicate.replace_old_param
 0.14% :     0.000001s :      6: predicate.reset_defer_inline
 1.71% :     0.000007s :     48: predicate.reshape_eliminate
 1.53% :     0.000007s :     48: predicate.row_tensor_add_zeros_like
 0.29% :     0.000001s :      6: predicate.row_tensor_eliminate
 1.50% :     0.000006s :     48: predicate.same_eliminate
 0.48% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.47% :     0.000002s :     12: predicate.special_op_eliminate
 0.87% :     0.000004s :     23: predicate.specialize_transform
 1.54% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000006s :     48: predicate.stack_unstack_eliminate
 0.21% :     0.000001s :      6: predicate.switch_call_monad_eliminater
 2.74% :     0.000012s :     67: predicate.switch_defer_inline
 2.28% :     0.000010s :     67: predicate.switch_layer_defer_inline
 5.93% :     0.000025s :    163: predicate.switch_simplify
 1.49% :     0.000006s :     48: predicate.tile_eliminate
 1.48% :     0.000006s :     48: predicate.transpose_eliminate
 1.80% :     0.000008s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.57% :     0.000007s :     48: predicate.tuple_list_get_item_depend_reorder
 3.07% :     0.000013s :     67: predicate.tuple_list_get_item_eliminator
 1.94% :     0.000008s :     48: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000008s :     55: predicate.tuple_to_list_eliminator_
 1.81% :     0.000008s :     61: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000012s :     84: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000007s :     48: predicate.value_based_eliminate
 0.19% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.29% :     0.000001s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001736    29
58.55% :     0.001017s :     13: func_graph_cloner_run.FuncGraphClonerGraph
41.45% :     0.000720s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042695    91
 0.14% :     0.000062s :      1: add_recomputation
 0.24% :     0.000102s :      1: auto_monad
 0.06% :     0.000026s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.54% :     0.000229s :      1: bootstrap
 0.07% :     0.000029s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000037s :      1: cse_after_recomputation
 0.03% :     0.000011s :      1: environ_conv
 0.14% :     0.000058s :      1: event_method
 0.02% :     0.000011s :      1: execute
 0.13% :     0.000056s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
31.07% :     0.013265s :      1: jit_opt_a
 0.46% :     0.000197s :      1: jit_opt_after_cconv
 0.16% :     0.000070s :      1: jit_opt_b
 1.04% :     0.000442s :      1: loop_unroll
 1.54% :     0.000658s :      1: mutable_eliminate
 7.00% :     0.002988s :     39: opt.transform.jit_opt_a
 0.18% :     0.000077s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000051s :      4: opt.transform.symbol_engine_opt
 1.15% :     0.000490s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000036s :      1: py_interpret_to_execute
 0.04% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000028s :      1: remove_dup_value
 4.59% :     0.001959s :      2: renormalize.infer
 3.75% :     0.001603s :      2: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000054s :      1: rewriter_after_opt_a
 0.23% :     0.000100s :      1: rewriter_before_opt_a
 0.23% :     0.000098s :      1: symbol_engine_optimizer
18.62% :     0.007950s :      1: task_emit
27.53% :     0.011755s :      1: type_inference
 0.17% :     0.000074s :      1: validate


TotalTime = 0.0108283, [30]
[bootstrap]: 0.00025483
[type_inference]: 0.00387189
[event_method]: 1.605e-05
[auto_monad]: 5.991e-05
[graph_reusing]: 4.90999e-06
[pre_auto_parallel]: 2.04e-06
[py_interpret_to_execute]: 3.167e-05
[rewriter_before_opt_a]: 6.845e-05
[expand_dump_flag]: 2.62001e-06
[jit_opt_a]: 0.00454766, [2]
    [Cycle 1]: 0.00134011, [27]
        [switch_simplify]: 4.583e-05
        [loop_unroll]: 2.892e-05
        [a_1]: 0.00050266
        [with_stream_mark]: 1.369e-05
        [recompute_prepare]: 6.89999e-06
        [updatestate_depend_eliminate]: 3.36999e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.39001e-06
        [parameter_eliminate]: 1.82001e-06
        [specialize_transform]: 5.81998e-06
        [updatestate_useless_node_eliminater]: 5.08002e-06
        [accelerated_algorithm]: 5.40999e-06
        [meta_shard_fg_expand]: 2.22001e-06
        [get_grad_eliminate_]: 6.06998e-06
        [merge_forward]: 3.33998e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.406e-05
        [j_node_and_user_rematch]: 8.47998e-06
        [meta_fg_expand]: 2.12001e-06
        [replace_old_param]: 8.57e-06
        [inline_without_move]: 5.15001e-06
        [renormalize]: 0.00044819
        [add_forward_monad_depend]: 4.50001e-06
        [auto_monad_grad]: 1.71998e-06
        [auto_monad_eliminator]: 1.202e-05
        [cse]: 1.767e-05
        [replace_applicator]: 1.091e-05
    [Cycle 2]: 0.00031157, [27]
        [switch_simplify]: 5.53997e-06
        [loop_unroll]: 4.85001e-06
        [a_1]: 9.009e-05
        [with_stream_mark]: 8.69e-06
        [recompute_prepare]: 4.99e-06
        [updatestate_depend_eliminate]: 2.46998e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 9.10019e-07
        [specialize_transform]: 4.84998e-06
        [updatestate_useless_node_eliminater]: 5.15999e-06
        [accelerated_algorithm]: 5.33002e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.51002e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.161e-05
        [j_node_and_user_rematch]: 7.73999e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 7.20998e-06
        [inline_without_move]: 4.85999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 6.90023e-07
        [auto_monad_eliminator]: 4.62998e-06
        [cse]: 1.014e-05
        [replace_applicator]: 4.99e-06
[py_interpret_to_execute_after_opt_a]: 8.37e-06
[rewriter_after_opt_a]: 2.694e-05
[convert_after_rewriter]: 6.51999e-06
[order_py_execute_after_rewriter]: 5.05001e-06
[mutable_eliminate]: 0.00047896
[jit_opt_b]: 5.012e-05, [1]
    [Cycle 1]: 4.353e-05, [2]
        [frontend_op_eliminate]: 1.676e-05
        [inline_after_opt_a]: 1.52e-05
[cconv]: 2.242e-05
[loop_unroll]: 0.00042187
[jit_opt_after_cconv]: 0.0001345, [1]
    [Cycle 1]: 0.00012826, [11]
        [c_1]: 2.091e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 5.59e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.587e-05
        [call_graph_tuple_transform]: 1.733e-05
        [tuple_list_get_item_eliminator]: 5.47999e-06
        [none_parameter_eliminate]: 1.08001e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.26002e-06
[remove_dup_value]: 1.259e-05
[partial_unused_args_eliminate]: 2.58e-06
[environ_conv]: 4.56002e-06
[add_recomputation]: 4.12e-05
[cse_after_recomputation]: 1.991e-05, [1]
    [Cycle 1]: 1.451e-05, [1]
        [cse]: 8.90999e-06
[auto_monad_reorder]: 1.279e-05
[get_jit_bprop_graph]: 1.39e-06
[rewriter_after_jit_bprop_graph]: 4.89e-06
[opt_after_jit_grad]: 0.00046205
[symbol_engine_optimizer]: 7.373e-05, [1]
    [Cycle 1]: 6.733e-05, [6]
        [build]: 3.83999e-06
        [elim_shapecalc]: 7.48e-06
        [elim_not_effective]: 1.291e-05
        [opt_reshape]: 5.54998e-06
        [fold_const_symbol]: 9.19e-06
        [renormalize]: 4.50003e-07
[validate]: 2.897e-05
Sums
    bootstrap                            :     0.000255s :  3.45%
    type_inference                       :     0.003872s : 52.44%
    event_method                         :     0.000016s :  0.22%
    auto_monad                           :     0.000060s :  0.81%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.43%
    rewriter_before_opt_a                :     0.000068s :  0.93%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000051s :  0.70%
    jit_opt_a.loop_unroll                :     0.000034s :  0.46%
    jit_opt_a.a_1                        :     0.000593s :  8.03%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.14%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000448s :  6.07%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.23%
    jit_opt_a.cse                        :     0.000028s :  0.38%
    jit_opt_a.replace_applicator         :     0.000016s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000027s :  0.36%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000479s :  6.49%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.21%
    cconv                                :     0.000022s :  0.30%
    loop_unroll                          :     0.000422s :  5.71%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000013s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000041s :  0.56%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000013s :  0.17%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000462s :  6.26%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000029s :  0.39%


Time group info:
------[substitution.]   0.000179    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.72% :     0.000005s :      3: substitution.graph_param_transform
84.27% :     0.000151s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.26% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.02% :     0.000004s :      2: substitution.replace_old_param
 5.48% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003820     2
80.00% :     0.003056s :      1: type_inference.infer
20.00% :     0.000764s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.41% :     0.000042s :      4: replace.inline
32.59% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000157     6
94.45% :     0.000149s :      4: match.inline
 5.55% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.28% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.47% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 1.94% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.49% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.70% :     0.000007s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.13% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.64% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.36% :     0.000003s :     15: predicate.partial_eliminate
 1.08% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.98% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.54% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.35% :     0.000009s :     53: predicate.switch_simplify
 1.37% :     0.000002s :     10: predicate.tile_eliminate
 1.23% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.80% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000495    12
50.52% :     0.000250s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.11% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.37% :     0.000220s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012144    72
 0.36% :     0.000044s :      1: add_recomputation
 0.52% :     0.000063s :      1: auto_monad
 0.13% :     0.000015s :      1: auto_monad_reorder
 2.19% :     0.000266s :      1: bootstrap
 0.21% :     0.000025s :      1: cconv
 0.08% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000019s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
37.47% :     0.004550s :      1: jit_opt_a
 1.13% :     0.000137s :      1: jit_opt_after_cconv
 0.43% :     0.000053s :      1: jit_opt_b
 3.54% :     0.000430s :      1: loop_unroll
 4.01% :     0.000487s :      1: mutable_eliminate
 6.50% :     0.000790s :     26: opt.transform.jit_opt_a
 0.38% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000025s :      4: opt.transform.jit_opt_b
 0.15% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.87% :     0.000470s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.04% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.28% :     0.000034s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000015s :      1: remove_dup_value
 1.77% :     0.000215s :      1: renormalize.infer
 1.87% :     0.000227s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000030s :      1: rewriter_after_opt_a
 0.59% :     0.000071s :      1: rewriter_before_opt_a
 0.63% :     0.000076s :      1: symbol_engine_optimizer
31.97% :     0.003883s :      1: type_inference

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:20.818.040 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e3310> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:20.818.707 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.018799, [33]
[bootstrap]: 0.00027551
[type_inference]: 0.00464332
[event_method]: 1.558e-05
[auto_monad]: 6.172e-05
[graph_reusing]: 5.63997e-06
[pre_auto_parallel]: 2.11e-06
[py_interpret_to_execute]: 3.511e-05
[rewriter_before_opt_a]: 6.907e-05
[expand_dump_flag]: 3.17002e-06
[jit_opt_a]: 0.00520975, [2]
    [Cycle 1]: 0.001671, [27]
        [switch_simplify]: 4.49e-05
        [loop_unroll]: 5.145e-05
        [a_1]: 0.00052821
        [with_stream_mark]: 1.618e-05
        [recompute_prepare]: 8.18001e-06
        [updatestate_depend_eliminate]: 3.48999e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 1.57999e-06
        [specialize_transform]: 5.75001e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.57999e-06
        [meta_shard_fg_expand]: 1.87001e-06
        [get_grad_eliminate_]: 5.24e-06
        [merge_forward]: 3.18998e-06
        [cell_reuse_recompute_pass]: 9.50007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.366e-05
        [j_node_and_user_rematch]: 8.65001e-06
        [meta_fg_expand]: 1.96e-06
        [replace_old_param]: 8.77999e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.00071195
        [add_forward_monad_depend]: 7.05998e-06
        [auto_monad_grad]: 3.03e-06
        [auto_monad_eliminator]: 1.354e-05
        [cse]: 2.118e-05
        [replace_applicator]: 1.508e-05
    [Cycle 2]: 0.00031741, [27]
        [switch_simplify]: 5.29e-06
        [loop_unroll]: 4.57e-06
        [a_1]: 8.855e-05
        [with_stream_mark]: 9.42001e-06
        [recompute_prepare]: 4.71002e-06
        [updatestate_depend_eliminate]: 2.63e-06
        [updatestate_assign_eliminate]: 2.05002e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 1.19998e-06
        [specialize_transform]: 4.71002e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 4.91002e-06
        [meta_shard_fg_expand]: 1.45999e-06
        [get_grad_eliminate_]: 4.26001e-06
        [merge_forward]: 2.85998e-06
        [cell_reuse_recompute_pass]: 2.32001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.284e-05
        [j_node_and_user_rematch]: 7.69002e-06
        [meta_fg_expand]: 1.76e-06
        [replace_old_param]: 7.06001e-06
        [inline_without_move]: 4.47e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 4.80001e-06
        [cse]: 1.085e-05
        [replace_applicator]: 4.83001e-06
[py_interpret_to_execute_after_opt_a]: 1.121e-05
[rewriter_after_opt_a]: 3.414e-05
[convert_after_rewriter]: 7.05e-06
[order_py_execute_after_rewriter]: 4.68999e-06
[mutable_eliminate]: 0.00067452
[jit_opt_b]: 5.018e-05, [1]
    [Cycle 1]: 4.293e-05, [2]
        [frontend_op_eliminate]: 1.601e-05
        [inline_after_opt_a]: 1.532e-05
[cconv]: 2.526e-05
[loop_unroll]: 0.00044145
[jit_opt_after_cconv]: 0.00014256, [1]
    [Cycle 1]: 0.00013655, [11]
        [c_1]: 2.072e-05
        [parameter_eliminate]: 3.48e-06
        [updatestate_depend_eliminate]: 6.12001e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [cse]: 2.107e-05
        [call_graph_tuple_transform]: 1.865e-05
        [tuple_list_get_item_eliminator]: 5.27999e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 5.17999e-06
[remove_dup_value]: 1.334e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 4.75999e-06
[add_recomputation]: 4.655e-05
[cse_after_recomputation]: 2.104e-05, [1]
    [Cycle 1]: 1.52e-05, [1]
        [cse]: 9.27001e-06
[auto_monad_reorder]: 1.494e-05
[get_jit_bprop_graph]: 1.92001e-06
[rewriter_after_jit_bprop_graph]: 5.19e-06
[opt_after_jit_grad]: 0.00047158
[symbol_engine_optimizer]: 7.252e-05, [1]
    [Cycle 1]: 6.642e-05, [6]
        [build]: 4.11001e-06
        [elim_shapecalc]: 7.68999e-06
        [elim_not_effective]: 1.277e-05
        [opt_reshape]: 5.54998e-06
        [fold_const_symbol]: 8.65001e-06
        [renormalize]: 7.30011e-07
[validate]: 3.771e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00619264
[execute]: 6.14001e-06
Sums
    bootstrap                            :     0.000276s :  1.84%
    type_inference                       :     0.004643s : 31.01%
    event_method                         :     0.000016s :  0.10%
    auto_monad                           :     0.000062s :  0.41%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000035s :  0.23%
    rewriter_before_opt_a                :     0.000069s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000050s :  0.34%
    jit_opt_a.loop_unroll                :     0.000056s :  0.37%
    jit_opt_a.a_1                        :     0.000617s :  4.12%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000712s :  4.76%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.12%
    jit_opt_a.cse                        :     0.000032s :  0.21%
    jit_opt_a.replace_applicator         :     0.000020s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000675s :  4.50%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000441s :  2.95%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000047s :  0.31%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000472s :  3.15%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006193s : 41.36%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000197    23
 0.96% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 3.04% :     0.000006s :      3: substitution.graph_param_transform
83.18% :     0.000164s :      4: substitution.inline
 1.39% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.55% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.97% :     0.000004s :      2: substitution.replace_old_param
 6.23% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004582     2
82.24% :     0.003768s :      1: type_inference.infer
17.76% :     0.000814s :      1: type_inference.specialize
------[replace.]   0.000067     6
67.06% :     0.000045s :      4: replace.inline
32.94% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000173     6
93.51% :     0.000161s :      4: match.inline
 6.49% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.96% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.47% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 2.15% :     0.000002s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.44% :     0.000002s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.88% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.41% :     0.000002s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.53% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000001s :      6: predicate.less_batch_normalization
 1.83% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.89% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.83% :     0.000002s :     18: predicate.replace_applicator
 0.71% :     0.000001s :      6: predicate.replace_old_param
 0.52% :     0.000001s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.63% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.49% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.34% :     0.000008s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.58% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.40% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000550    12
49.39% :     0.000272s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.33% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.28% :     0.000249s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020458    76
 0.24% :     0.000050s :      1: add_recomputation
 0.32% :     0.000065s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.43% :     0.000292s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.09% :     0.000019s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.48% :     0.005213s :      1: jit_opt_a
 0.71% :     0.000145s :      1: jit_opt_after_cconv
 0.26% :     0.000053s :      1: jit_opt_b
 2.19% :     0.000449s :      1: loop_unroll
 3.35% :     0.000685s :      1: mutable_eliminate
 4.10% :     0.000839s :     26: opt.transform.jit_opt_a
 0.23% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.35% :     0.000481s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000038s :      1: py_interpret_to_execute
 0.07% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.57% :     0.000320s :      1: renormalize.infer
 1.88% :     0.000385s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000038s :      1: rewriter_after_opt_a
 0.36% :     0.000073s :      1: rewriter_before_opt_a
 0.37% :     0.000075s :      1: symbol_engine_optimizer
30.35% :     0.006210s :      1: task_emit
22.74% :     0.004653s :      1: type_inference
 0.27% :     0.000056s :      1: validate


TotalTime = 0.0103494, [30]
[bootstrap]: 0.00023597
[type_inference]: 0.00365697
[event_method]: 1.536e-05
[auto_monad]: 5.854e-05
[graph_reusing]: 4.53999e-06
[pre_auto_parallel]: 1.82001e-06
[py_interpret_to_execute]: 2.808e-05
[rewriter_before_opt_a]: 6.332e-05
[expand_dump_flag]: 2.34999e-06
[jit_opt_a]: 0.00438069, [2]
    [Cycle 1]: 0.00124978, [27]
        [switch_simplify]: 4.49e-05
        [loop_unroll]: 2.891e-05
        [a_1]: 0.00046734
        [with_stream_mark]: 1.144e-05
        [recompute_prepare]: 6.93e-06
        [updatestate_depend_eliminate]: 2.77002e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.30002e-06
        [parameter_eliminate]: 1.39998e-06
        [specialize_transform]: 6.01998e-06
        [updatestate_useless_node_eliminater]: 4.97999e-06
        [accelerated_algorithm]: 5.21002e-06
        [meta_shard_fg_expand]: 1.37e-06
        [get_grad_eliminate_]: 4.90001e-06
        [merge_forward]: 2.79001e-06
        [cell_reuse_recompute_pass]: 8.79983e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.245e-05
        [j_node_and_user_rematch]: 9.02e-06
        [meta_fg_expand]: 1.82001e-06
        [replace_old_param]: 8.08999e-06
        [inline_without_move]: 5.01002e-06
        [renormalize]: 0.00042066
        [add_forward_monad_depend]: 4.22998e-06
        [auto_monad_grad]: 1.74e-06
        [auto_monad_eliminator]: 9.99999e-06
        [cse]: 1.65e-05
        [replace_applicator]: 1.004e-05
    [Cycle 2]: 0.00031236, [27]
        [switch_simplify]: 5.34e-06
        [loop_unroll]: 4.76002e-06
        [a_1]: 8.864e-05
        [with_stream_mark]: 7.99002e-06
        [recompute_prepare]: 4.84998e-06
        [updatestate_depend_eliminate]: 2.58003e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.91998e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 5.02e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 4.94e-06
        [meta_shard_fg_expand]: 9.89996e-07
        [get_grad_eliminate_]: 4.60999e-06
        [merge_forward]: 2.27001e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.243e-05
        [j_node_and_user_rematch]: 7.90998e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 6.57002e-06
        [inline_without_move]: 4.63001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.21002e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 4.75999e-06
        [cse]: 9.78998e-06
        [replace_applicator]: 5.16002e-06
[py_interpret_to_execute_after_opt_a]: 7.28e-06
[rewriter_after_opt_a]: 2.657e-05
[convert_after_rewriter]: 6.07999e-06
[order_py_execute_after_rewriter]: 4.92e-06
[mutable_eliminate]: 0.00046472
[jit_opt_b]: 4.85e-05, [1]
    [Cycle 1]: 4.203e-05, [2]
        [frontend_op_eliminate]: 1.599e-05
        [inline_after_opt_a]: 1.491e-05
[cconv]: 1.882e-05
[loop_unroll]: 0.00041608
[jit_opt_after_cconv]: 0.00013104, [1]
    [Cycle 1]: 0.00012492, [11]
        [c_1]: 2.004e-05
        [parameter_eliminate]: 2.37001e-06
        [updatestate_depend_eliminate]: 5.06997e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 2.18998e-06
        [cse]: 1.574e-05
        [call_graph_tuple_transform]: 1.643e-05
        [tuple_list_get_item_eliminator]: 5.43002e-06
        [none_parameter_eliminate]: 1.19e-06
        [renormalize]: 4.10015e-07
        [switch_simplify]: 5.07e-06
[remove_dup_value]: 9.14e-06
[partial_unused_args_eliminate]: 1.87999e-06
[environ_conv]: 4.34997e-06
[add_recomputation]: 3.402e-05
[cse_after_recomputation]: 1.89e-05, [1]
    [Cycle 1]: 1.377e-05, [1]
        [cse]: 8.3e-06
[auto_monad_reorder]: 1.267e-05
[get_jit_bprop_graph]: 1.37e-06
[rewriter_after_jit_bprop_graph]: 4.62e-06
[opt_after_jit_grad]: 0.00044499
[symbol_engine_optimizer]: 7.172e-05, [1]
    [Cycle 1]: 6.529e-05, [6]
        [build]: 2.91e-06
        [elim_shapecalc]: 7.35998e-06
        [elim_not_effective]: 1.282e-05
        [opt_reshape]: 5.71e-06
        [fold_const_symbol]: 8.54e-06
        [renormalize]: 4.30009e-07
[validate]: 2.533e-05
Sums
    bootstrap                            :     0.000236s :  3.37%
    type_inference                       :     0.003657s : 52.28%
    event_method                         :     0.000015s :  0.22%
    auto_monad                           :     0.000059s :  0.84%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000028s :  0.40%
    rewriter_before_opt_a                :     0.000063s :  0.91%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000050s :  0.72%
    jit_opt_a.loop_unroll                :     0.000034s :  0.48%
    jit_opt_a.a_1                        :     0.000556s :  7.95%
    jit_opt_a.with_stream_mark           :     0.000019s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.36%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.24%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000421s :  6.01%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.21%
    jit_opt_a.cse                        :     0.000026s :  0.38%
    jit_opt_a.replace_applicator         :     0.000015s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.10%
    rewriter_after_opt_a                 :     0.000027s :  0.38%
    convert_after_rewriter               :     0.000006s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000465s :  6.64%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.21%
    cconv                                :     0.000019s :  0.27%
    loop_unroll                          :     0.000416s :  5.95%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.29%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000009s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000034s :  0.49%
    cse_after_recomputation.cse          :     0.000008s :  0.12%
    auto_monad_reorder                   :     0.000013s :  0.18%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000445s :  6.36%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.18%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000025s :  0.36%


Time group info:
------[substitution.]   0.000155    23
 1.15% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.87% :     0.000004s :      3: substitution.graph_param_transform
83.10% :     0.000128s :      4: substitution.inline
 1.70% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.60% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.94% :     0.000003s :      2: substitution.replace_old_param
 5.90% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003611     2
80.58% :     0.002910s :      1: type_inference.infer
19.42% :     0.000701s :      1: type_inference.specialize
------[replace.]   0.000057     6
65.94% :     0.000038s :      4: replace.inline
34.06% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000134     6
94.06% :     0.000126s :      4: match.inline
 5.94% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.20% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000001s :     10: predicate.addn_zero_filter
 2.25% :     0.000003s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.45% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.39% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.83% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.23% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.19% :     0.000006s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.44% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.86% :     0.000002s :     15: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.11% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.02% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.64% :     0.000002s :      3: predicate.mutable_eliminate
 0.57% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000002s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.75% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.83% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.67% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.45% :     0.000008s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.51% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.36% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.92% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000458    12
48.63% :     0.000223s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.41% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.96% :     0.000210s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011588    72
 0.32% :     0.000037s :      1: add_recomputation
 0.54% :     0.000063s :      1: auto_monad
 0.13% :     0.000015s :      1: auto_monad_reorder
 2.13% :     0.000247s :      1: bootstrap
 0.18% :     0.000021s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.18% :     0.000021s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000019s :      1: event_method
 0.04% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000003s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
37.82% :     0.004383s :      1: jit_opt_a
 1.15% :     0.000134s :      1: jit_opt_after_cconv
 0.44% :     0.000051s :      1: jit_opt_b
 3.66% :     0.000424s :      1: loop_unroll
 4.08% :     0.000473s :      1: mutable_eliminate
 6.46% :     0.000749s :     26: opt.transform.jit_opt_a
 0.38% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000024s :      4: opt.transform.jit_opt_b
 0.10% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.27% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.91% :     0.000453s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000031s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000012s :      1: remove_dup_value
 1.71% :     0.000198s :      1: renormalize.infer
 1.87% :     0.000217s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000030s :      1: rewriter_after_opt_a
 0.57% :     0.000066s :      1: rewriter_before_opt_a
 0.64% :     0.000075s :      1: symbol_engine_optimizer
31.65% :     0.003667s :      1: type_inference


TotalTime = 0.0380924, [33]
[bootstrap]: 0.00021897
[type_inference]: 0.011315
[event_method]: 5.314e-05
[auto_monad]: 9.807e-05
[graph_reusing]: 7.45998e-06
[pre_auto_parallel]: 3.33e-06
[py_interpret_to_execute]: 3.754e-05
[rewriter_before_opt_a]: 9.854e-05
[expand_dump_flag]: 3.78001e-06
[jit_opt_a]: 0.0141091, [3]
    [Cycle 1]: 0.00685354, [27]
        [switch_simplify]: 8.197e-05
        [loop_unroll]: 4.018e-05
        [a_1]: 0.000893
        [with_stream_mark]: 3.377e-05
        [recompute_prepare]: 1.796e-05
        [updatestate_depend_eliminate]: 9.34998e-06
        [updatestate_assign_eliminate]: 7.3e-06
        [updatestate_loads_eliminate]: 6.28e-06
        [parameter_eliminate]: 4.17003e-06
        [specialize_transform]: 1.281e-05
        [updatestate_useless_node_eliminater]: 1.211e-05
        [accelerated_algorithm]: 1.258e-05
        [meta_shard_fg_expand]: 3.32002e-06
        [get_grad_eliminate_]: 1.181e-05
        [merge_forward]: 8.33999e-06
        [cell_reuse_recompute_pass]: 9.30013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.772e-05
        [j_node_and_user_rematch]: 2.314e-05
        [meta_fg_expand]: 0.00170075
        [replace_old_param]: 5.801e-05
        [inline_without_move]: 5.388e-05
        [renormalize]: 0.00322168
        [add_forward_monad_depend]: 1.51e-05
        [auto_monad_grad]: 6.69999e-06
        [auto_monad_eliminator]: 7.885e-05
        [cse]: 0.00019374
        [replace_applicator]: 8.462e-05
    [Cycle 2]: 0.00272458, [27]
        [switch_simplify]: 4.019e-05
        [loop_unroll]: 3.882e-05
        [a_1]: 0.00125132
        [with_stream_mark]: 2.108e-05
        [recompute_prepare]: 1.362e-05
        [updatestate_depend_eliminate]: 5.70001e-06
        [updatestate_assign_eliminate]: 5.42999e-06
        [updatestate_loads_eliminate]: 4.67e-06
        [parameter_eliminate]: 2.83e-06
        [specialize_transform]: 1.03e-05
        [updatestate_useless_node_eliminater]: 9.02e-06
        [accelerated_algorithm]: 9.15999e-06
        [meta_shard_fg_expand]: 3.18e-06
        [get_grad_eliminate_]: 9.51e-06
        [merge_forward]: 5.38002e-06
        [cell_reuse_recompute_pass]: 1.54998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.249e-05
        [j_node_and_user_rematch]: 1.592e-05
        [meta_fg_expand]: 6.675e-05
        [replace_old_param]: 1.514e-05
        [inline_without_move]: 9.25999e-06
        [renormalize]: 0.00083798
        [add_forward_monad_depend]: 6.98998e-06
        [auto_monad_grad]: 2.72001e-06
        [auto_monad_eliminator]: 2.211e-05
        [cse]: 6.217e-05
        [replace_applicator]: 5.814e-05
    [Cycle 3]: 0.00053385, [27]
        [switch_simplify]: 9.89999e-06
        [loop_unroll]: 8.57e-06
        [a_1]: 0.00020403
        [with_stream_mark]: 1.472e-05
        [recompute_prepare]: 8.48999e-06
        [updatestate_depend_eliminate]: 5.17999e-06
        [updatestate_assign_eliminate]: 5.10999e-06
        [updatestate_loads_eliminate]: 4.16001e-06
        [parameter_eliminate]: 1.27999e-06
        [specialize_transform]: 8.11002e-06
        [updatestate_useless_node_eliminater]: 8.66002e-06
        [accelerated_algorithm]: 1.035e-05
        [meta_shard_fg_expand]: 2.14e-06
        [get_grad_eliminate_]: 8.2e-06
        [merge_forward]: 4.88001e-06
        [cell_reuse_recompute_pass]: 3.06999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.955e-05
        [j_node_and_user_rematch]: 1.48e-05
        [meta_fg_expand]: 3.22002e-06
        [replace_old_param]: 1.152e-05
        [inline_without_move]: 7.88999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.81e-06
        [auto_monad_grad]: 1.10001e-06
        [auto_monad_eliminator]: 1.135e-05
        [cse]: 2.635e-05
        [replace_applicator]: 8.60999e-06
[py_interpret_to_execute_after_opt_a]: 1.706e-05
[rewriter_after_opt_a]: 5.103e-05
[convert_after_rewriter]: 8.96002e-06
[order_py_execute_after_rewriter]: 6.72002e-06
[mutable_eliminate]: 0.00088573
[jit_opt_b]: 7.58e-05, [1]
    [Cycle 1]: 6.652e-05, [2]
        [frontend_op_eliminate]: 2.811e-05
        [inline_after_opt_a]: 2.64e-05
[cconv]: 2.721e-05
[loop_unroll]: 0.0004818
[jit_opt_after_cconv]: 0.00022384, [1]
    [Cycle 1]: 0.00021624, [11]
        [c_1]: 3.738e-05
        [parameter_eliminate]: 3.58e-06
        [updatestate_depend_eliminate]: 1.125e-05
        [updatestate_assign_eliminate]: 4.69998e-06
        [updatestate_loads_eliminate]: 4.99e-06
        [cse]: 4.722e-05
        [call_graph_tuple_transform]: 3.058e-05
        [tuple_list_get_item_eliminator]: 8.76002e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 1.00999e-06
        [switch_simplify]: 8.87999e-06
[remove_dup_value]: 2.547e-05
[partial_unused_args_eliminate]: 2.91999e-06
[environ_conv]: 7.87998e-06
[add_recomputation]: 6.377e-05
[cse_after_recomputation]: 3.655e-05, [1]
    [Cycle 1]: 3.032e-05, [1]
        [cse]: 2.378e-05
[auto_monad_reorder]: 2.367e-05
[get_jit_bprop_graph]: 2.20002e-06
[rewriter_after_jit_bprop_graph]: 8.59e-06
[opt_after_jit_grad]: 0.00051417
[symbol_engine_optimizer]: 0.00010361, [1]
    [Cycle 1]: 9.622e-05, [6]
        [build]: 5.67999e-06
        [elim_shapecalc]: 1.238e-05
        [elim_not_effective]: 2.152e-05
        [opt_reshape]: 1.073e-05
        [fold_const_symbol]: 1.585e-05
        [renormalize]: 6.89994e-07
[validate]: 5.386e-05
[backend_pass]: 9.70002e-07
[task_emit]: 0.00928439
[execute]: 7.93001e-06
Sums
    bootstrap                            :     0.000219s :  0.66%
    type_inference                       :     0.011315s : 34.08%
    event_method                         :     0.000053s :  0.16%
    auto_monad                           :     0.000098s :  0.30%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000038s :  0.11%
    rewriter_before_opt_a                :     0.000099s :  0.30%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000132s :  0.40%
    jit_opt_a.loop_unroll                :     0.000088s :  0.26%
    jit_opt_a.a_1                        :     0.002348s :  7.07%
    jit_opt_a.with_stream_mark           :     0.000070s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000040s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000020s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000018s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000015s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000008s :  0.02%
    jit_opt_a.specialize_transform       :     0.000031s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000030s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000032s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000009s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000030s :  0.09%
    jit_opt_a.merge_forward              :     0.000019s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000070s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000054s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001771s :  5.33%
    jit_opt_a.replace_old_param          :     0.000085s :  0.26%
    jit_opt_a.inline_without_move        :     0.000071s :  0.21%
    jit_opt_a.renormalize                :     0.004060s : 12.23%
    jit_opt_a.add_forward_monad_depend   :     0.000024s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000011s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000112s :  0.34%
    jit_opt_a.cse                        :     0.000282s :  0.85%
    jit_opt_a.replace_applicator         :     0.000151s :  0.46%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.05%
    rewriter_after_opt_a                 :     0.000051s :  0.15%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000886s :  2.67%
    jit_opt_b.frontend_op_eliminate      :     0.000028s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.08%
    cconv                                :     0.000027s :  0.08%
    loop_unroll                          :     0.000482s :  1.45%
    jit_opt_after_cconv.c_1              :     0.000037s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.cse              :     0.000047s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000031s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.03%
    remove_dup_value                     :     0.000025s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000008s :  0.02%
    add_recomputation                    :     0.000064s :  0.19%
    cse_after_recomputation.cse          :     0.000024s :  0.07%
    auto_monad_reorder                   :     0.000024s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.03%
    opt_after_jit_grad                   :     0.000514s :  1.55%
    symbol_engine_optimizer.build        :     0.000006s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000022s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000011s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000016s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000054s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009284s : 27.97%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000647   126
 6.87% :     0.000044s :      2: substitution.arithmetic_simplify
 0.53% :     0.000003s :      5: substitution.elim_not_effective
 0.41% :     0.000003s :      5: substitution.fold_const_symbol
 1.19% :     0.000008s :      6: substitution.graph_param_transform
61.49% :     0.000398s :     12: substitution.inline
 2.74% :     0.000018s :      2: substitution.inline_without_move
 1.75% :     0.000011s :     18: substitution.j_node_and_user_rematch
 1.80% :     0.000012s :      7: substitution.minmaximum_grad
 1.22% :     0.000008s :      5: substitution.partial_eliminate
 2.21% :     0.000014s :     18: substitution.remove_not_recompute_node
 4.94% :     0.000032s :     10: substitution.replace_applicator
 1.49% :     0.000010s :      7: substitution.replace_old_param
 0.38% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 3.70% :     0.000024s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.51% :     0.000016s :      7: substitution.tuple_list_get_item_depend_reorder
 6.77% :     0.000044s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011233     2
88.35% :     0.009924s :      1: type_inference.infer
11.65% :     0.001309s :      1: type_inference.specialize
------[replace.]   0.000163    19
63.74% :     0.000104s :     12: replace.inline
36.26% :     0.000059s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000411    19
94.77% :     0.000390s :     12: match.inline
 5.23% :     0.000022s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000432  2895
 1.67% :     0.000007s :     48: predicate.accumulaten_eliminater
 0.67% :     0.000003s :      6: predicate.ad_related_special_op_eliminate
 1.34% :     0.000006s :     48: predicate.addn_check_dump
 1.66% :     0.000007s :     48: predicate.addn_zero_filter
 2.11% :     0.000009s :     48: predicate.arithmetic_simplify
 1.61% :     0.000007s :     48: predicate.cast_eliminate
 0.23% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.35% :     0.000006s :     48: predicate.compare_switch_simplify
 1.67% :     0.000007s :     48: predicate.depend_value_elim
 1.35% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.44% :     0.000006s :     48: predicate.dict_get_item_eliminator
 1.40% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.38% :     0.000002s :      6: predicate.dumpgradient_eliminate
 0.14% :     0.000001s :      6: predicate.elim_not_effective
 0.27% :     0.000001s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.48% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.33% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.42% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.55% :     0.000007s :     48: predicate.environ_get_eliminate
 1.40% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.11% :     0.000000s :      6: predicate.fold_const_symbol
 0.91% :     0.000004s :     23: predicate.get_grad_eliminate
 0.16% :     0.000001s :      6: predicate.graph_param_transform
 4.34% :     0.000019s :     79: predicate.inline
 1.79% :     0.000008s :     44: predicate.inline_without_move
 0.37% :     0.000002s :     23: predicate.j_node_and_user_rematch
 1.14% :     0.000005s :     23: predicate.less_batch_normalization
 1.75% :     0.000008s :     55: predicate.list_to_tuple_eliminator_
 1.94% :     0.000008s :     61: predicate.load_eliminater
 0.59% :     0.000003s :      6: predicate.loop_unroll_after_grad
 3.05% :     0.000013s :     90: predicate.loop_unroll_before_grad
 1.84% :     0.000008s :     54: predicate.make_slice_get_slice_eliminator
 1.59% :     0.000007s :     48: predicate.merge_addn
 1.40% :     0.000006s :     48: predicate.minmaximum_grad
 0.63% :     0.000003s :      6: predicate.mutable_eliminate
 0.30% :     0.000001s :      6: predicate.opt_reshape
 2.26% :     0.000010s :     61: predicate.partial_eliminate
 1.39% :     0.000006s :     48: predicate.print_const_string_wrapper
 1.73% :     0.000007s :     48: predicate.reduce_eliminate
 1.72% :     0.000007s :     55: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.61% :     0.000011s :    108: predicate.replace_applicator
 0.88% :     0.000004s :     44: predicate.replace_old_param
 0.15% :     0.000001s :      6: predicate.reset_defer_inline
 2.01% :     0.000009s :     48: predicate.reshape_eliminate
 1.46% :     0.000006s :     48: predicate.row_tensor_add_zeros_like
 0.31% :     0.000001s :      6: predicate.row_tensor_eliminate
 1.43% :     0.000006s :     48: predicate.same_eliminate
 0.51% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.54% :     0.000002s :     12: predicate.special_op_eliminate
 0.98% :     0.000004s :     23: predicate.specialize_transform
 1.67% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.56% :     0.000007s :     48: predicate.stack_unstack_eliminate
 0.21% :     0.000001s :      6: predicate.switch_call_monad_eliminater
 2.83% :     0.000012s :     67: predicate.switch_defer_inline
 2.28% :     0.000010s :     67: predicate.switch_layer_defer_inline
 5.94% :     0.000026s :    163: predicate.switch_simplify
 1.44% :     0.000006s :     48: predicate.tile_eliminate
 1.65% :     0.000007s :     48: predicate.transpose_eliminate
 1.67% :     0.000007s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.67% :     0.000007s :     48: predicate.tuple_list_get_item_depend_reorder
 3.67% :     0.000016s :     67: predicate.tuple_list_get_item_eliminator
 1.84% :     0.000008s :     48: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000007s :     55: predicate.tuple_to_list_eliminator_
 1.88% :     0.000008s :     61: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000013s :     84: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000008s :     48: predicate.value_based_eliminate
 0.20% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.28% :     0.000001s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001878    29
59.21% :     0.001112s :     13: func_graph_cloner_run.FuncGraphClonerGraph
40.79% :     0.000766s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.045446    91
 0.15% :     0.000067s :      1: add_recomputation
 0.22% :     0.000102s :      1: auto_monad
 0.06% :     0.000027s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.52% :     0.000236s :      1: bootstrap
 0.07% :     0.000030s :      1: cconv
 0.03% :     0.000012s :      1: convert_after_rewriter
 0.09% :     0.000039s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.13% :     0.000058s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
31.05% :     0.014113s :      1: jit_opt_a
 0.50% :     0.000227s :      1: jit_opt_after_cconv
 0.17% :     0.000078s :      1: jit_opt_b
 1.08% :     0.000491s :      1: loop_unroll
 1.98% :     0.000898s :      1: mutable_eliminate
 6.83% :     0.003105s :     39: opt.transform.jit_opt_a
 0.18% :     0.000082s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000047s :      4: opt.transform.jit_opt_b
 0.04% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000023s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000034s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000056s :      4: opt.transform.symbol_engine_opt
 1.15% :     0.000523s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000040s :      1: py_interpret_to_execute
 0.04% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000028s :      1: remove_dup_value
 4.64% :     0.002110s :      2: renormalize.infer
 4.24% :     0.001928s :      2: renormalize.specialize
 0.02% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000054s :      1: rewriter_after_opt_a
 0.22% :     0.000101s :      1: rewriter_before_opt_a
 0.23% :     0.000106s :      1: symbol_engine_optimizer
20.46% :     0.009299s :      1: task_emit
24.94% :     0.011334s :      1: type_inference
 0.18% :     0.000081s :      1: validate


TotalTime = 0.0107345, [30]
[bootstrap]: 0.00025181
[type_inference]: 0.00379812
[event_method]: 1.601e-05
[auto_monad]: 5.48e-05
[graph_reusing]: 5.39e-06
[pre_auto_parallel]: 2.04e-06
[py_interpret_to_execute]: 2.896e-05
[rewriter_before_opt_a]: 6.839e-05
[expand_dump_flag]: 2.61999e-06
[jit_opt_a]: 0.00457574, [2]
    [Cycle 1]: 0.00132525, [27]
        [switch_simplify]: 4.626e-05
        [loop_unroll]: 2.874e-05
        [a_1]: 0.00049095
        [with_stream_mark]: 1.392e-05
        [recompute_prepare]: 6.74999e-06
        [updatestate_depend_eliminate]: 3.14999e-06
        [updatestate_assign_eliminate]: 2.49001e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 5.82001e-06
        [updatestate_useless_node_eliminater]: 5.02999e-06
        [accelerated_algorithm]: 5.56e-06
        [meta_shard_fg_expand]: 1.65001e-06
        [get_grad_eliminate_]: 4.98001e-06
        [merge_forward]: 2.81e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.268e-05
        [j_node_and_user_rematch]: 8.70999e-06
        [meta_fg_expand]: 2.02001e-06
        [replace_old_param]: 8.25e-06
        [inline_without_move]: 5.20999e-06
        [renormalize]: 0.00044677
        [add_forward_monad_depend]: 5.00999e-06
        [auto_monad_grad]: 2.02001e-06
        [auto_monad_eliminator]: 1.193e-05
        [cse]: 2.76e-05
        [replace_applicator]: 1.133e-05
    [Cycle 2]: 0.00031857, [27]
        [switch_simplify]: 5.86e-06
        [loop_unroll]: 4.82e-06
        [a_1]: 9.066e-05
        [with_stream_mark]: 8.47e-06
        [recompute_prepare]: 5.27999e-06
        [updatestate_depend_eliminate]: 2.39999e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.83997e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 4.98001e-06
        [updatestate_useless_node_eliminater]: 4.67998e-06
        [accelerated_algorithm]: 5.01997e-06
        [meta_shard_fg_expand]: 9.89996e-07
        [get_grad_eliminate_]: 4.48001e-06
        [merge_forward]: 2.27001e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.173e-05
        [j_node_and_user_rematch]: 7.83999e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 9.69e-06
        [inline_without_move]: 4.79e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 4.90001e-06
        [cse]: 9.96998e-06
        [replace_applicator]: 5.07e-06
[py_interpret_to_execute_after_opt_a]: 6.85998e-06
[rewriter_after_opt_a]: 2.823e-05
[convert_after_rewriter]: 6.19001e-06
[order_py_execute_after_rewriter]: 4.68999e-06
[mutable_eliminate]: 0.00046712
[jit_opt_b]: 4.869e-05, [1]
    [Cycle 1]: 4.219e-05, [2]
        [frontend_op_eliminate]: 1.604e-05
        [inline_after_opt_a]: 1.436e-05
[cconv]: 2.145e-05
[loop_unroll]: 0.00041033
[jit_opt_after_cconv]: 0.00013401, [1]
    [Cycle 1]: 0.000127, [11]
        [c_1]: 2.12e-05
        [parameter_eliminate]: 2.37999e-06
        [updatestate_depend_eliminate]: 4.85999e-06
        [updatestate_assign_eliminate]: 2.35002e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [cse]: 1.619e-05
        [call_graph_tuple_transform]: 1.716e-05
        [tuple_list_get_item_eliminator]: 5.44e-06
        [none_parameter_eliminate]: 1.05999e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 5.59e-06
[remove_dup_value]: 1.279e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 4.4e-06
[add_recomputation]: 3.798e-05
[cse_after_recomputation]: 1.966e-05, [1]
    [Cycle 1]: 1.442e-05, [1]
        [cse]: 8.75999e-06
[auto_monad_reorder]: 1.429e-05
[get_jit_bprop_graph]: 1.41002e-06
[rewriter_after_jit_bprop_graph]: 4.87e-06
[opt_after_jit_grad]: 0.00045375
[symbol_engine_optimizer]: 7.101e-05, [1]
    [Cycle 1]: 6.466e-05, [6]
        [build]: 3.04001e-06
        [elim_shapecalc]: 7.03e-06
        [elim_not_effective]: 1.239e-05
        [opt_reshape]: 5.66e-06
        [fold_const_symbol]: 8.35999e-06
        [renormalize]: 4.59986e-07
[validate]: 2.78e-05
Sums
    bootstrap                            :     0.000252s :  3.47%
    type_inference                       :     0.003798s : 52.35%
    event_method                         :     0.000016s :  0.22%
    auto_monad                           :     0.000055s :  0.76%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000029s :  0.40%
    rewriter_before_opt_a                :     0.000068s :  0.94%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000052s :  0.72%
    jit_opt_a.loop_unroll                :     0.000034s :  0.46%
    jit_opt_a.a_1                        :     0.000582s :  8.02%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.13%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.05%
    jit_opt_a.replace_old_param          :     0.000018s :  0.25%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000447s :  6.16%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.23%
    jit_opt_a.cse                        :     0.000038s :  0.52%
    jit_opt_a.replace_applicator         :     0.000016s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.09%
    rewriter_after_opt_a                 :     0.000028s :  0.39%
    convert_after_rewriter               :     0.000006s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000467s :  6.44%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.20%
    cconv                                :     0.000021s :  0.30%
    loop_unroll                          :     0.000410s :  5.66%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.29%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.08%
    remove_dup_value                     :     0.000013s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000038s :  0.52%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000014s :  0.20%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000454s :  6.25%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000028s :  0.38%


Time group info:
------[substitution.]   0.000177    23
 1.13% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.59% :     0.000005s :      3: substitution.graph_param_transform
81.73% :     0.000145s :      4: substitution.inline
 1.85% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.35% :     0.000004s :      4: substitution.remove_not_recompute_node
 3.22% :     0.000006s :      2: substitution.replace_old_param
 6.40% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003748     2
79.30% :     0.002973s :      1: type_inference.infer
20.70% :     0.000776s :      1: type_inference.specialize
------[replace.]   0.000060     6
67.56% :     0.000041s :      4: replace.inline
32.44% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000153     6
93.28% :     0.000142s :      4: match.inline
 6.72% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.65% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.40% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 2.31% :     0.000003s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.45% :     0.000002s :     10: predicate.depend_value_elim
 1.31% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.11% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.47% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000000s :      3: predicate.graph_param_transform
 5.34% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.49% :     0.000002s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.29% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.00% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.47% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000002s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.86% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 1.12% :     0.000001s :      6: predicate.specialize_transform
 1.26% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.06% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.50% :     0.000009s :     53: predicate.switch_simplify
 1.58% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.72% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.99% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000498    12
50.23% :     0.000250s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.98% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.79% :     0.000223s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012025    72
 0.34% :     0.000041s :      1: add_recomputation
 0.49% :     0.000058s :      1: auto_monad
 0.14% :     0.000017s :      1: auto_monad_reorder
 2.21% :     0.000265s :      1: bootstrap
 0.20% :     0.000024s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.18% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000019s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.07% :     0.004578s :      1: jit_opt_a
 1.14% :     0.000137s :      1: jit_opt_after_cconv
 0.43% :     0.000051s :      1: jit_opt_b
 3.48% :     0.000419s :      1: loop_unroll
 3.95% :     0.000475s :      1: mutable_eliminate
 6.48% :     0.000779s :     26: opt.transform.jit_opt_a
 0.38% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000024s :      4: opt.transform.jit_opt_b
 0.09% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.84% :     0.000462s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000031s :      1: py_interpret_to_execute
 0.08% :     0.000009s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000015s :      1: remove_dup_value
 1.77% :     0.000213s :      1: renormalize.infer
 1.90% :     0.000228s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000031s :      1: rewriter_after_opt_a
 0.59% :     0.000071s :      1: rewriter_before_opt_a
 0.61% :     0.000073s :      1: symbol_engine_optimizer
31.68% :     0.003809s :      1: type_inference


TotalTime = 0.0192819, [33]
[bootstrap]: 0.0002774
[type_inference]: 0.00458608
[event_method]: 1.635e-05
[auto_monad]: 6.426e-05
[graph_reusing]: 5.67001e-06
[pre_auto_parallel]: 2.47001e-06
[py_interpret_to_execute]: 3.341e-05
[rewriter_before_opt_a]: 6.847e-05
[expand_dump_flag]: 3.09001e-06
[jit_opt_a]: 0.00569537, [2]
    [Cycle 1]: 0.00163474, [27]
        [switch_simplify]: 5.589e-05
        [loop_unroll]: 2.87e-05
        [a_1]: 0.00053697
        [with_stream_mark]: 2.041e-05
        [recompute_prepare]: 7.6e-06
        [updatestate_depend_eliminate]: 4.02e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 1.81998e-06
        [specialize_transform]: 6.12999e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 5.00001e-06
        [merge_forward]: 3.80998e-06
        [cell_reuse_recompute_pass]: 1.26002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.495e-05
        [j_node_and_user_rematch]: 9.71e-06
        [meta_fg_expand]: 2.17001e-06
        [replace_old_param]: 8.75999e-06
        [inline_without_move]: 5.35999e-06
        [renormalize]: 0.00066597
        [add_forward_monad_depend]: 5.10001e-06
        [auto_monad_grad]: 3.09999e-06
        [auto_monad_eliminator]: 1.392e-05
        [cse]: 2.673e-05
        [replace_applicator]: 1.181e-05
    [Cycle 2]: 0.00031301, [27]
        [switch_simplify]: 5.22e-06
        [loop_unroll]: 4.82998e-06
        [a_1]: 8.858e-05
        [with_stream_mark]: 9.29e-06
        [recompute_prepare]: 5.07e-06
        [updatestate_depend_eliminate]: 2.34999e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.83002e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 4.92999e-06
        [updatestate_useless_node_eliminater]: 4.88001e-06
        [accelerated_algorithm]: 5.05999e-06
        [meta_shard_fg_expand]: 1.13001e-06
        [get_grad_eliminate_]: 4.49002e-06
        [merge_forward]: 2.53e-06
        [cell_reuse_recompute_pass]: 1.32999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.195e-05
        [j_node_and_user_rematch]: 7.65e-06
        [meta_fg_expand]: 1.63002e-06
        [replace_old_param]: 6.79001e-06
        [inline_without_move]: 4.53999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.19003e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 4.79e-06
        [cse]: 1.086e-05
        [replace_applicator]: 5.11002e-06
[py_interpret_to_execute_after_opt_a]: 1.102e-05
[rewriter_after_opt_a]: 3.42e-05
[convert_after_rewriter]: 7.77002e-06
[order_py_execute_after_rewriter]: 4.92e-06
[mutable_eliminate]: 0.00070131
[jit_opt_b]: 5.033e-05, [1]
    [Cycle 1]: 4.267e-05, [2]
        [frontend_op_eliminate]: 1.571e-05
        [inline_after_opt_a]: 1.502e-05
[cconv]: 2.688e-05
[loop_unroll]: 0.00043916
[jit_opt_after_cconv]: 0.00014461, [1]
    [Cycle 1]: 0.00013798, [11]
        [c_1]: 2.12e-05
        [parameter_eliminate]: 3.18e-06
        [updatestate_depend_eliminate]: 5.94e-06
        [updatestate_assign_eliminate]: 2.48998e-06
        [updatestate_loads_eliminate]: 2.30002e-06
        [cse]: 1.936e-05
        [call_graph_tuple_transform]: 1.942e-05
        [tuple_list_get_item_eliminator]: 5.37001e-06
        [none_parameter_eliminate]: 1.43002e-06
        [renormalize]: 6.69999e-07
        [switch_simplify]: 5.54998e-06
[remove_dup_value]: 1.394e-05
[partial_unused_args_eliminate]: 2.11998e-06
[environ_conv]: 5.60001e-06
[add_recomputation]: 4.783e-05
[cse_after_recomputation]: 1.939e-05, [1]
    [Cycle 1]: 1.403e-05, [1]
        [cse]: 8.41002e-06
[auto_monad_reorder]: 1.557e-05
[get_jit_bprop_graph]: 1.99999e-06
[rewriter_after_jit_bprop_graph]: 6.34001e-06
[opt_after_jit_grad]: 0.00045977
[symbol_engine_optimizer]: 7.201e-05, [1]
    [Cycle 1]: 6.513e-05, [6]
        [build]: 3.69002e-06
        [elim_shapecalc]: 7.33999e-06
        [elim_not_effective]: 1.229e-05
        [opt_reshape]: 5.75001e-06
        [fold_const_symbol]: 8.60001e-06
        [renormalize]: 6.10016e-07
[validate]: 3.51e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.00622857
[execute]: 6.78e-06
Sums
    bootstrap                            :     0.000277s :  1.86%
    type_inference                       :     0.004586s : 30.72%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000064s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.22%
    rewriter_before_opt_a                :     0.000068s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.41%
    jit_opt_a.loop_unroll                :     0.000034s :  0.22%
    jit_opt_a.a_1                        :     0.000626s :  4.19%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000666s :  4.46%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000038s :  0.25%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000701s :  4.70%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000027s :  0.18%
    loop_unroll                          :     0.000439s :  2.94%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000048s :  0.32%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000460s :  3.08%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006229s : 41.72%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000214    23
 0.90% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.48% :     0.000005s :      3: substitution.graph_param_transform
84.15% :     0.000180s :      4: substitution.inline
 1.72% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.22% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.93% :     0.000004s :      2: substitution.replace_old_param
 6.00% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004523     2
82.67% :     0.003739s :      1: type_inference.infer
17.33% :     0.000784s :      1: type_inference.specialize
------[replace.]   0.000067     6
66.40% :     0.000044s :      4: replace.inline
33.60% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000190     6
93.81% :     0.000178s :      4: match.inline
 6.19% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.30% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.21% :     0.000001s :     10: predicate.addn_zero_filter
 2.02% :     0.000002s :     10: predicate.arithmetic_simplify
 1.16% :     0.000001s :     10: predicate.cast_eliminate
 0.82% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.42% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.45% :     0.000001s :      3: predicate.graph_param_transform
 6.40% :     0.000007s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.65% :     0.000002s :      6: predicate.less_batch_normalization
 1.76% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.56% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.11% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.08% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.71% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.36% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 2.12% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.27% :     0.000001s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 2.29% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.54% :     0.000009s :     53: predicate.switch_simplify
 1.40% :     0.000002s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.22% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.67% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000547    12
48.68% :     0.000266s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.26% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.05% :     0.000241s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020886    76
 0.24% :     0.000051s :      1: add_recomputation
 0.32% :     0.000068s :      1: auto_monad
 0.09% :     0.000019s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.40% :     0.000292s :      1: bootstrap
 0.14% :     0.000030s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.09% :     0.000020s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.29% :     0.005699s :      1: jit_opt_a
 0.71% :     0.000147s :      1: jit_opt_after_cconv
 0.25% :     0.000053s :      1: jit_opt_b
 2.14% :     0.000447s :      1: loop_unroll
 3.40% :     0.000710s :      1: mutable_eliminate
 3.98% :     0.000832s :     26: opt.transform.jit_opt_a
 0.23% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.25% :     0.000469s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.65% :     0.000344s :      1: renormalize.infer
 1.50% :     0.000314s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000038s :      1: rewriter_after_opt_a
 0.34% :     0.000071s :      1: rewriter_before_opt_a
 0.36% :     0.000075s :      1: symbol_engine_optimizer
29.90% :     0.006245s :      1: task_emit
22.01% :     0.004597s :      1: type_inference
 0.26% :     0.000055s :      1: validate

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:20.866.257 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e3790> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:20.866.872 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0102551, [30]
[bootstrap]: 0.00023422
[type_inference]: 0.0035787
[event_method]: 1.536e-05
[auto_monad]: 4.975e-05
[graph_reusing]: 5.04998e-06
[pre_auto_parallel]: 1.67999e-06
[py_interpret_to_execute]: 2.652e-05
[rewriter_before_opt_a]: 6.389e-05
[expand_dump_flag]: 2.49001e-06
[jit_opt_a]: 0.00436533, [2]
    [Cycle 1]: 0.00125892, [27]
        [switch_simplify]: 4.359e-05
        [loop_unroll]: 2.836e-05
        [a_1]: 0.00046749
        [with_stream_mark]: 1.25e-05
        [recompute_prepare]: 7.06999e-06
        [updatestate_depend_eliminate]: 2.99001e-06
        [updatestate_assign_eliminate]: 2.58e-06
        [updatestate_loads_eliminate]: 2.30002e-06
        [parameter_eliminate]: 1.68002e-06
        [specialize_transform]: 6.00002e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 6.12999e-06
        [meta_shard_fg_expand]: 1.59998e-06
        [get_grad_eliminate_]: 4.93001e-06
        [merge_forward]: 2.73e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.24e-05
        [j_node_and_user_rematch]: 8.94e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 8.05999e-06
        [inline_without_move]: 5.12e-06
        [renormalize]: 0.00041718
        [add_forward_monad_depend]: 4.08001e-06
        [auto_monad_grad]: 1.22e-06
        [auto_monad_eliminator]: 9.96e-06
        [cse]: 1.645e-05
        [replace_applicator]: 1.044e-05
    [Cycle 2]: 0.00031911, [27]
        [switch_simplify]: 5.45001e-06
        [loop_unroll]: 5.11002e-06
        [a_1]: 9.317e-05
        [with_stream_mark]: 8.05e-06
        [recompute_prepare]: 4.60999e-06
        [updatestate_depend_eliminate]: 2.37001e-06
        [updatestate_assign_eliminate]: 1.95001e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.02998e-06
        [specialize_transform]: 4.82998e-06
        [updatestate_useless_node_eliminater]: 4.78001e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 1.22999e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.18002e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.179e-05
        [j_node_and_user_rematch]: 7.63999e-06
        [meta_fg_expand]: 1.50001e-06
        [replace_old_param]: 6.96999e-06
        [inline_without_move]: 4.60999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 4.95001e-06
        [cse]: 1.03e-05
        [replace_applicator]: 4.89003e-06
[py_interpret_to_execute_after_opt_a]: 7.06001e-06
[rewriter_after_opt_a]: 2.587e-05
[convert_after_rewriter]: 7.95998e-06
[order_py_execute_after_rewriter]: 4.31002e-06
[mutable_eliminate]: 0.00046046
[jit_opt_b]: 4.722e-05, [1]
    [Cycle 1]: 4.073e-05, [2]
        [frontend_op_eliminate]: 1.499e-05
        [inline_after_opt_a]: 1.429e-05
[cconv]: 1.914e-05
[loop_unroll]: 0.00043391
[jit_opt_after_cconv]: 0.00013275, [1]
    [Cycle 1]: 0.00012648, [11]
        [c_1]: 1.989e-05
        [parameter_eliminate]: 2.13998e-06
        [updatestate_depend_eliminate]: 4.99998e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 1.526e-05
        [call_graph_tuple_transform]: 1.771e-05
        [tuple_list_get_item_eliminator]: 5.28002e-06
        [none_parameter_eliminate]: 8.39995e-07
        [renormalize]: 3.59985e-07
        [switch_simplify]: 5.29e-06
[remove_dup_value]: 1.059e-05
[partial_unused_args_eliminate]: 1.97999e-06
[environ_conv]: 4.38999e-06
[add_recomputation]: 3.503e-05
[cse_after_recomputation]: 1.919e-05, [1]
    [Cycle 1]: 1.426e-05, [1]
        [cse]: 8.50001e-06
[auto_monad_reorder]: 1.18e-05
[get_jit_bprop_graph]: 1.50999e-06
[rewriter_after_jit_bprop_graph]: 4.45999e-06
[opt_after_jit_grad]: 0.0004438
[symbol_engine_optimizer]: 6.961e-05, [1]
    [Cycle 1]: 6.37e-05, [6]
        [build]: 2.59999e-06
        [elim_shapecalc]: 7.18998e-06
        [elim_not_effective]: 1.212e-05
        [opt_reshape]: 5.90002e-06
        [fold_const_symbol]: 8.65001e-06
        [renormalize]: 4.39992e-07
[validate]: 2.601e-05
Sums
    bootstrap                            :     0.000234s :  3.38%
    type_inference                       :     0.003579s : 51.71%
    event_method                         :     0.000015s :  0.22%
    auto_monad                           :     0.000050s :  0.72%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000027s :  0.38%
    rewriter_before_opt_a                :     0.000064s :  0.92%
    expand_dump_flag                     :     0.000002s :  0.04%
    jit_opt_a.switch_simplify            :     0.000049s :  0.71%
    jit_opt_a.loop_unroll                :     0.000033s :  0.48%
    jit_opt_a.a_1                        :     0.000561s :  8.10%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.17%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.24%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.22%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000417s :  6.03%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.22%
    jit_opt_a.cse                        :     0.000027s :  0.39%
    jit_opt_a.replace_applicator         :     0.000015s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.10%
    rewriter_after_opt_a                 :     0.000026s :  0.37%
    convert_after_rewriter               :     0.000008s :  0.12%
    order_py_execute_after_rewriter      :     0.000004s :  0.06%
    mutable_eliminate                    :     0.000460s :  6.65%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.21%
    cconv                                :     0.000019s :  0.28%
    loop_unroll                          :     0.000434s :  6.27%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.29%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000015s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.26%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.08%
    remove_dup_value                     :     0.000011s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000035s :  0.51%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000012s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.06%
    opt_after_jit_grad                   :     0.000444s :  6.41%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.18%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.09%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000026s :  0.38%


Time group info:
------[substitution.]   0.000155    23
 1.13% :     0.000002s :      2: substitution.elim_not_effective
 0.83% :     0.000001s :      2: substitution.fold_const_symbol
 2.85% :     0.000004s :      3: substitution.graph_param_transform
82.54% :     0.000128s :      4: substitution.inline
 1.78% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.54% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.21% :     0.000003s :      2: substitution.replace_old_param
 6.12% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003534     2
80.67% :     0.002851s :      1: type_inference.infer
19.33% :     0.000683s :      1: type_inference.specialize
------[replace.]   0.000058     6
66.05% :     0.000038s :      4: replace.inline
33.95% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000133     6
93.71% :     0.000125s :      4: match.inline
 6.29% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.74% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.07% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.39% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.26% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.26% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000001s :     10: predicate.environ_get_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.56% :     0.000001s :      3: predicate.graph_param_transform
 5.30% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.50% :     0.000002s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.75% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.80% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.74% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.52% :     0.000008s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.90% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.86% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000447    12
48.75% :     0.000218s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.40% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.85% :     0.000205s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011493    72
 0.33% :     0.000038s :      1: add_recomputation
 0.46% :     0.000053s :      1: auto_monad
 0.12% :     0.000014s :      1: auto_monad_reorder
 2.13% :     0.000245s :      1: bootstrap
 0.19% :     0.000022s :      1: cconv
 0.09% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000021s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000019s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
38.00% :     0.004368s :      1: jit_opt_a
 1.18% :     0.000135s :      1: jit_opt_after_cconv
 0.44% :     0.000050s :      1: jit_opt_b
 3.84% :     0.000442s :      1: loop_unroll
 4.08% :     0.000469s :      1: mutable_eliminate
 6.54% :     0.000752s :     26: opt.transform.jit_opt_a
 0.39% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000023s :      4: opt.transform.jit_opt_b
 0.09% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.27% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.93% :     0.000452s :      1: opt_after_jit_grad
 0.06% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000029s :      1: py_interpret_to_execute
 0.08% :     0.000009s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000013s :      1: remove_dup_value
 1.74% :     0.000200s :      1: renormalize.infer
 1.84% :     0.000212s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000029s :      1: rewriter_after_opt_a
 0.58% :     0.000066s :      1: rewriter_before_opt_a
 0.63% :     0.000072s :      1: symbol_engine_optimizer
31.22% :     0.003588s :      1: type_inference


TotalTime = 0.0103311, [30]
[bootstrap]: 0.00024428
[type_inference]: 0.00362877
[event_method]: 1.729e-05
[auto_monad]: 5.243e-05
[graph_reusing]: 5.25999e-06
[pre_auto_parallel]: 1.84e-06
[py_interpret_to_execute]: 2.741e-05
[rewriter_before_opt_a]: 6.515e-05
[expand_dump_flag]: 2.39001e-06
[jit_opt_a]: 0.00439052, [2]
    [Cycle 1]: 0.00125791, [27]
        [switch_simplify]: 4.584e-05
        [loop_unroll]: 2.877e-05
        [a_1]: 0.00046896
        [with_stream_mark]: 1.279e-05
        [recompute_prepare]: 6.92002e-06
        [updatestate_depend_eliminate]: 3.09999e-06
        [updatestate_assign_eliminate]: 2.89001e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 5.56e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.62999e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 5.04e-06
        [merge_forward]: 2.86999e-06
        [cell_reuse_recompute_pass]: 7.89994e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.247e-05
        [j_node_and_user_rematch]: 8.65999e-06
        [meta_fg_expand]: 2.18002e-06
        [replace_old_param]: 7.61999e-06
        [inline_without_move]: 5.24e-06
        [renormalize]: 0.00041458
        [add_forward_monad_depend]: 4.51002e-06
        [auto_monad_grad]: 1.52001e-06
        [auto_monad_eliminator]: 1.097e-05
        [cse]: 1.958e-05
        [replace_applicator]: 1.046e-05
    [Cycle 2]: 0.00031138, [27]
        [switch_simplify]: 5.49e-06
        [loop_unroll]: 4.58001e-06
        [a_1]: 8.843e-05
        [with_stream_mark]: 7.82e-06
        [recompute_prepare]: 4.75999e-06
        [updatestate_depend_eliminate]: 2.37999e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 9.29984e-07
        [specialize_transform]: 5.10001e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 1.04003e-06
        [get_grad_eliminate_]: 4.45e-06
        [merge_forward]: 2.36998e-06
        [cell_reuse_recompute_pass]: 1.18001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.241e-05
        [j_node_and_user_rematch]: 7.35003e-06
        [meta_fg_expand]: 1.45001e-06
        [replace_old_param]: 6.66999e-06
        [inline_without_move]: 4.57998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 6.39993e-07
        [auto_monad_eliminator]: 4.90999e-06
        [cse]: 1.032e-05
        [replace_applicator]: 5.05999e-06
[py_interpret_to_execute_after_opt_a]: 7.89002e-06
[rewriter_after_opt_a]: 2.685e-05
[convert_after_rewriter]: 6.24001e-06
[order_py_execute_after_rewriter]: 5.21002e-06
[mutable_eliminate]: 0.00046177
[jit_opt_b]: 4.728e-05, [1]
    [Cycle 1]: 4.035e-05, [2]
        [frontend_op_eliminate]: 1.512e-05
        [inline_after_opt_a]: 1.427e-05
[cconv]: 2.01e-05
[loop_unroll]: 0.0004047
[jit_opt_after_cconv]: 0.00013041, [1]
    [Cycle 1]: 0.00012435, [11]
        [c_1]: 1.945e-05
        [parameter_eliminate]: 2.13002e-06
        [updatestate_depend_eliminate]: 5.04e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.98002e-06
        [cse]: 1.489e-05
        [call_graph_tuple_transform]: 1.741e-05
        [tuple_list_get_item_eliminator]: 5.32001e-06
        [none_parameter_eliminate]: 9.49978e-07
        [renormalize]: 4.39992e-07
        [switch_simplify]: 5.20001e-06
[remove_dup_value]: 1.219e-05
[partial_unused_args_eliminate]: 1.89999e-06
[environ_conv]: 4.16001e-06
[add_recomputation]: 3.933e-05
[cse_after_recomputation]: 2.007e-05, [1]
    [Cycle 1]: 1.463e-05, [1]
        [cse]: 8.82e-06
[auto_monad_reorder]: 1.276e-05
[get_jit_bprop_graph]: 1.42e-06
[rewriter_after_jit_bprop_graph]: 4.4e-06
[opt_after_jit_grad]: 0.0004403
[symbol_engine_optimizer]: 7.078e-05, [1]
    [Cycle 1]: 6.416e-05, [6]
        [build]: 2.93e-06
        [elim_shapecalc]: 7.56001e-06
        [elim_not_effective]: 1.199e-05
        [opt_reshape]: 5.29e-06
        [fold_const_symbol]: 8.40001e-06
        [renormalize]: 5.69999e-07
[validate]: 2.668e-05
Sums
    bootstrap                            :     0.000244s :  3.51%
    type_inference                       :     0.003629s : 52.11%
    event_method                         :     0.000017s :  0.25%
    auto_monad                           :     0.000052s :  0.75%
    graph_reusing                        :     0.000005s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000027s :  0.39%
    rewriter_before_opt_a                :     0.000065s :  0.94%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000051s :  0.74%
    jit_opt_a.loop_unroll                :     0.000033s :  0.48%
    jit_opt_a.a_1                        :     0.000557s :  8.00%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.36%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000014s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000415s :  5.95%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.23%
    jit_opt_a.cse                        :     0.000030s :  0.43%
    jit_opt_a.replace_applicator         :     0.000016s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000027s :  0.39%
    convert_after_rewriter               :     0.000006s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000462s :  6.63%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.20%
    cconv                                :     0.000020s :  0.29%
    loop_unroll                          :     0.000405s :  5.81%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000015s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.25%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000012s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000039s :  0.56%
    cse_after_recomputation.cse          :     0.000009s :  0.13%
    auto_monad_reorder                   :     0.000013s :  0.18%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.06%
    opt_after_jit_grad                   :     0.000440s :  6.32%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000027s :  0.38%


Time group info:
------[substitution.]   0.000159    23
 1.07% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 3.16% :     0.000005s :      3: substitution.graph_param_transform
82.09% :     0.000131s :      4: substitution.inline
 1.93% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.57% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.85% :     0.000003s :      2: substitution.replace_old_param
 6.63% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003579     2
80.49% :     0.002880s :      1: type_inference.infer
19.51% :     0.000698s :      1: type_inference.specialize
------[replace.]   0.000057     6
67.06% :     0.000038s :      4: replace.inline
32.94% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000138     6
93.11% :     0.000128s :      4: match.inline
 6.89% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.37% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000001s :     10: predicate.addn_zero_filter
 2.07% :     0.000002s :     10: predicate.arithmetic_simplify
 1.45% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.42% :     0.000002s :     10: predicate.depend_value_elim
 1.28% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.24% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000001s :     10: predicate.environ_get_eliminate
 1.22% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.05% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.55% :     0.000006s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.40% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.21% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.04% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.58% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000002s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.53% :     0.000008s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.33% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.54% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.92% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000468    12
51.25% :     0.000240s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.35% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.39% :     0.000203s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011560    72
 0.37% :     0.000042s :      1: add_recomputation
 0.48% :     0.000055s :      1: auto_monad
 0.13% :     0.000015s :      1: auto_monad_reorder
 2.23% :     0.000258s :      1: bootstrap
 0.20% :     0.000023s :      1: cconv
 0.08% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.18% :     0.000021s :      1: event_method
 0.04% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.07% :     0.000008s :      1: graph_reusing
38.00% :     0.004393s :      1: jit_opt_a
 1.15% :     0.000133s :      1: jit_opt_after_cconv
 0.43% :     0.000050s :      1: jit_opt_b
 3.56% :     0.000412s :      1: loop_unroll
 4.07% :     0.000471s :      1: mutable_eliminate
 6.48% :     0.000749s :     26: opt.transform.jit_opt_a
 0.38% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000024s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.87% :     0.000448s :      1: opt_after_jit_grad
 0.07% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000030s :      1: py_interpret_to_execute
 0.09% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000015s :      1: remove_dup_value
 1.72% :     0.000199s :      1: renormalize.infer
 1.80% :     0.000208s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000031s :      1: rewriter_after_opt_a
 0.59% :     0.000068s :      1: rewriter_before_opt_a
 0.63% :     0.000073s :      1: symbol_engine_optimizer
31.49% :     0.003640s :      1: type_inference


TotalTime = 0.0403928, [33]
[bootstrap]: 0.00022265
[type_inference]: 0.0117792
[event_method]: 5.763e-05
[auto_monad]: 0.00010101
[graph_reusing]: 7.4e-06
[pre_auto_parallel]: 3.17002e-06
[py_interpret_to_execute]: 4.398e-05
[rewriter_before_opt_a]: 0.00010439
[expand_dump_flag]: 3.16001e-06
[jit_opt_a]: 0.0154569, [3]
    [Cycle 1]: 0.00781717, [27]
        [switch_simplify]: 8.2e-05
        [loop_unroll]: 4.017e-05
        [a_1]: 0.00096456
        [with_stream_mark]: 3.398e-05
        [recompute_prepare]: 2.492e-05
        [updatestate_depend_eliminate]: 9.86e-06
        [updatestate_assign_eliminate]: 6.72002e-06
        [updatestate_loads_eliminate]: 7.21999e-06
        [parameter_eliminate]: 3.77998e-06
        [specialize_transform]: 1.55e-05
        [updatestate_useless_node_eliminater]: 1.288e-05
        [accelerated_algorithm]: 1.413e-05
        [meta_shard_fg_expand]: 5.02e-06
        [get_grad_eliminate_]: 1.288e-05
        [merge_forward]: 8.63001e-06
        [cell_reuse_recompute_pass]: 1.12999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.937e-05
        [j_node_and_user_rematch]: 2.491e-05
        [meta_fg_expand]: 0.00211938
        [replace_old_param]: 7.606e-05
        [inline_without_move]: 6.479e-05
        [renormalize]: 0.00357699
        [add_forward_monad_depend]: 1.776e-05
        [auto_monad_grad]: 8.28001e-06
        [auto_monad_eliminator]: 7.161e-05
        [cse]: 0.00019789
        [replace_applicator]: 9.669e-05
    [Cycle 2]: 0.00299767, [27]
        [switch_simplify]: 4.371e-05
        [loop_unroll]: 4.062e-05
        [a_1]: 0.00132976
        [with_stream_mark]: 2.24e-05
        [recompute_prepare]: 1.313e-05
        [updatestate_depend_eliminate]: 6.56999e-06
        [updatestate_assign_eliminate]: 5.71e-06
        [updatestate_loads_eliminate]: 4.93001e-06
        [parameter_eliminate]: 2.35002e-06
        [specialize_transform]: 9.27001e-06
        [updatestate_useless_node_eliminater]: 9.91e-06
        [accelerated_algorithm]: 9.56e-06
        [meta_shard_fg_expand]: 3.95e-06
        [get_grad_eliminate_]: 9.22999e-06
        [merge_forward]: 6.02001e-06
        [cell_reuse_recompute_pass]: 1.86e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.285e-05
        [j_node_and_user_rematch]: 1.536e-05
        [meta_fg_expand]: 6.779e-05
        [replace_old_param]: 1.682e-05
        [inline_without_move]: 9.84999e-06
        [renormalize]: 0.00102799
        [add_forward_monad_depend]: 7.58001e-06
        [auto_monad_grad]: 2.50002e-06
        [auto_monad_eliminator]: 2.369e-05
        [cse]: 6.769e-05
        [replace_applicator]: 2.324e-05
    [Cycle 3]: 0.00056717, [27]
        [switch_simplify]: 1.126e-05
        [loop_unroll]: 9.36002e-06
        [a_1]: 0.00021765
        [with_stream_mark]: 1.661e-05
        [recompute_prepare]: 9.92001e-06
        [updatestate_depend_eliminate]: 5.35001e-06
        [updatestate_assign_eliminate]: 5.10999e-06
        [updatestate_loads_eliminate]: 3.88001e-06
        [parameter_eliminate]: 1.34998e-06
        [specialize_transform]: 8.90999e-06
        [updatestate_useless_node_eliminater]: 8.85001e-06
        [accelerated_algorithm]: 9.67999e-06
        [meta_shard_fg_expand]: 2.17999e-06
        [get_grad_eliminate_]: 8.50001e-06
        [merge_forward]: 5.05999e-06
        [cell_reuse_recompute_pass]: 3.11999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.062e-05
        [j_node_and_user_rematch]: 1.512e-05
        [meta_fg_expand]: 3.41999e-06
        [replace_old_param]: 1.265e-05
        [inline_without_move]: 7.93999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 2.24001e-06
        [auto_monad_grad]: 1.86998e-06
        [auto_monad_eliminator]: 1.305e-05
        [cse]: 2.874e-05
        [replace_applicator]: 8.73001e-06
[py_interpret_to_execute_after_opt_a]: 1.953e-05
[rewriter_after_opt_a]: 5.415e-05
[convert_after_rewriter]: 9.81e-06
[order_py_execute_after_rewriter]: 6.72002e-06
[mutable_eliminate]: 0.00081956
[jit_opt_b]: 7.99e-05, [1]
    [Cycle 1]: 6.929e-05, [2]
        [frontend_op_eliminate]: 3.069e-05
        [inline_after_opt_a]: 2.547e-05
[cconv]: 2.763e-05
[loop_unroll]: 0.00048424
[jit_opt_after_cconv]: 0.00022465, [1]
    [Cycle 1]: 0.0002168, [11]
        [c_1]: 3.742e-05
        [parameter_eliminate]: 3.82998e-06
        [updatestate_depend_eliminate]: 9.40001e-06
        [updatestate_assign_eliminate]: 5.07999e-06
        [updatestate_loads_eliminate]: 4.28001e-06
        [cse]: 4.791e-05
        [call_graph_tuple_transform]: 3.068e-05
        [tuple_list_get_item_eliminator]: 9.27999e-06
        [none_parameter_eliminate]: 1.83002e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 8.77999e-06
[remove_dup_value]: 2.647e-05
[partial_unused_args_eliminate]: 2.07999e-06
[environ_conv]: 8.32998e-06
[add_recomputation]: 6.64e-05
[cse_after_recomputation]: 3.701e-05, [1]
    [Cycle 1]: 3.083e-05, [1]
        [cse]: 2.431e-05
[auto_monad_reorder]: 2.524e-05
[get_jit_bprop_graph]: 2.14e-06
[rewriter_after_jit_bprop_graph]: 7.80998e-06
[opt_after_jit_grad]: 0.00056044
[symbol_engine_optimizer]: 0.00010346, [1]
    [Cycle 1]: 9.61e-05, [6]
        [build]: 5.57001e-06
        [elim_shapecalc]: 1.255e-05
        [elim_not_effective]: 2.266e-05
        [opt_reshape]: 9.23002e-06
        [fold_const_symbol]: 1.543e-05
        [renormalize]: 8.89995e-07
[validate]: 5.576e-05
[backend_pass]: 1.04e-06
[task_emit]: 0.00973109
[execute]: 9.22999e-06
Sums
    bootstrap                            :     0.000223s :  0.63%
    type_inference                       :     0.011779s : 33.35%
    event_method                         :     0.000058s :  0.16%
    auto_monad                           :     0.000101s :  0.29%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000044s :  0.12%
    rewriter_before_opt_a                :     0.000104s :  0.30%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000137s :  0.39%
    jit_opt_a.loop_unroll                :     0.000090s :  0.26%
    jit_opt_a.a_1                        :     0.002512s :  7.11%
    jit_opt_a.with_stream_mark           :     0.000073s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000048s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000022s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000018s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000016s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000007s :  0.02%
    jit_opt_a.specialize_transform       :     0.000034s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000032s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000033s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000011s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000031s :  0.09%
    jit_opt_a.merge_forward              :     0.000020s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000073s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000055s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.002191s :  6.20%
    jit_opt_a.replace_old_param          :     0.000106s :  0.30%
    jit_opt_a.inline_without_move        :     0.000083s :  0.23%
    jit_opt_a.renormalize                :     0.004605s : 13.04%
    jit_opt_a.add_forward_monad_depend   :     0.000028s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000013s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000108s :  0.31%
    jit_opt_a.cse                        :     0.000294s :  0.83%
    jit_opt_a.replace_applicator         :     0.000129s :  0.36%
    py_interpret_to_execute_after_opt_a  :     0.000020s :  0.06%
    rewriter_after_opt_a                 :     0.000054s :  0.15%
    convert_after_rewriter               :     0.000010s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000820s :  2.32%
    jit_opt_b.frontend_op_eliminate      :     0.000031s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000025s :  0.07%
    cconv                                :     0.000028s :  0.08%
    loop_unroll                          :     0.000484s :  1.37%
    jit_opt_after_cconv.c_1              :     0.000037s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000048s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000031s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.02%
    remove_dup_value                     :     0.000026s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000008s :  0.02%
    add_recomputation                    :     0.000066s :  0.19%
    cse_after_recomputation.cse          :     0.000024s :  0.07%
    auto_monad_reorder                   :     0.000025s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.02%
    opt_after_jit_grad                   :     0.000560s :  1.59%
    symbol_engine_optimizer.build        :     0.000006s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000023s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000056s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009731s : 27.55%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000715   126
 7.11% :     0.000051s :      2: substitution.arithmetic_simplify
 0.51% :     0.000004s :      5: substitution.elim_not_effective
 0.36% :     0.000003s :      5: substitution.fold_const_symbol
 1.13% :     0.000008s :      6: substitution.graph_param_transform
60.34% :     0.000432s :     12: substitution.inline
 3.35% :     0.000024s :      2: substitution.inline_without_move
 1.77% :     0.000013s :     18: substitution.j_node_and_user_rematch
 1.91% :     0.000014s :      7: substitution.minmaximum_grad
 1.14% :     0.000008s :      5: substitution.partial_eliminate
 2.09% :     0.000015s :     18: substitution.remove_not_recompute_node
 5.20% :     0.000037s :     10: substitution.replace_applicator
 1.70% :     0.000012s :      7: substitution.replace_old_param
 0.47% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.78% :     0.000027s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.26% :     0.000016s :      7: substitution.tuple_list_get_item_depend_reorder
 6.87% :     0.000049s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011682     2
87.91% :     0.010270s :      1: type_inference.infer
12.09% :     0.001412s :      1: type_inference.specialize
------[replace.]   0.000182    19
67.19% :     0.000123s :     12: replace.inline
32.81% :     0.000060s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000448    19
94.70% :     0.000424s :     12: match.inline
 5.30% :     0.000024s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000462  2895
 1.61% :     0.000007s :     48: predicate.accumulaten_eliminater
 0.58% :     0.000003s :      6: predicate.ad_related_special_op_eliminate
 1.28% :     0.000006s :     48: predicate.addn_check_dump
 1.73% :     0.000008s :     48: predicate.addn_zero_filter
 2.12% :     0.000010s :     48: predicate.arithmetic_simplify
 1.58% :     0.000007s :     48: predicate.cast_eliminate
 0.27% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.28% :     0.000006s :     48: predicate.compare_switch_simplify
 1.64% :     0.000008s :     48: predicate.depend_value_elim
 1.28% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.97% :     0.000009s :     48: predicate.dict_get_item_eliminator
 1.37% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.34% :     0.000002s :      6: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      6: predicate.elim_not_effective
 0.28% :     0.000001s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.58% :     0.000007s :     48: predicate.environ_add_const_eliminate
 1.34% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.32% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.42% :     0.000007s :     48: predicate.environ_get_eliminate
 1.32% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      6: predicate.fold_const_symbol
 0.93% :     0.000004s :     23: predicate.get_grad_eliminate
 0.16% :     0.000001s :      6: predicate.graph_param_transform
 4.38% :     0.000020s :     79: predicate.inline
 1.99% :     0.000009s :     44: predicate.inline_without_move
 0.40% :     0.000002s :     23: predicate.j_node_and_user_rematch
 1.29% :     0.000006s :     23: predicate.less_batch_normalization
 1.60% :     0.000007s :     55: predicate.list_to_tuple_eliminator_
 2.00% :     0.000009s :     61: predicate.load_eliminater
 0.56% :     0.000003s :      6: predicate.loop_unroll_after_grad
 3.06% :     0.000014s :     90: predicate.loop_unroll_before_grad
 1.75% :     0.000008s :     54: predicate.make_slice_get_slice_eliminator
 1.39% :     0.000006s :     48: predicate.merge_addn
 1.34% :     0.000006s :     48: predicate.minmaximum_grad
 0.74% :     0.000003s :      6: predicate.mutable_eliminate
 0.24% :     0.000001s :      6: predicate.opt_reshape
 2.18% :     0.000010s :     61: predicate.partial_eliminate
 1.39% :     0.000006s :     48: predicate.print_const_string_wrapper
 1.79% :     0.000008s :     48: predicate.reduce_eliminate
 1.74% :     0.000008s :     55: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.54% :     0.000012s :    108: predicate.replace_applicator
 1.11% :     0.000005s :     44: predicate.replace_old_param
 0.12% :     0.000001s :      6: predicate.reset_defer_inline
 1.78% :     0.000008s :     48: predicate.reshape_eliminate
 1.55% :     0.000007s :     48: predicate.row_tensor_add_zeros_like
 0.39% :     0.000002s :      6: predicate.row_tensor_eliminate
 1.57% :     0.000007s :     48: predicate.same_eliminate
 0.52% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.54% :     0.000002s :     12: predicate.special_op_eliminate
 0.89% :     0.000004s :     23: predicate.specialize_transform
 1.57% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000007s :     48: predicate.stack_unstack_eliminate
 0.20% :     0.000001s :      6: predicate.switch_call_monad_eliminater
 2.66% :     0.000012s :     67: predicate.switch_defer_inline
 2.30% :     0.000011s :     67: predicate.switch_layer_defer_inline
 6.33% :     0.000029s :    163: predicate.switch_simplify
 1.47% :     0.000007s :     48: predicate.tile_eliminate
 1.47% :     0.000007s :     48: predicate.transpose_eliminate
 1.88% :     0.000009s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.55% :     0.000007s :     48: predicate.tuple_list_get_item_depend_reorder
 3.33% :     0.000015s :     67: predicate.tuple_list_get_item_eliminator
 1.91% :     0.000009s :     48: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000008s :     55: predicate.tuple_to_list_eliminator_
 1.79% :     0.000008s :     61: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000013s :     84: predicate.updatestate_useless_node_eliminater
 2.07% :     0.000010s :     48: predicate.value_based_eliminate
 0.21% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.33% :     0.000002s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001947    29
56.33% :     0.001097s :     13: func_graph_cloner_run.FuncGraphClonerGraph
43.67% :     0.000850s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.048492    91
 0.14% :     0.000070s :      1: add_recomputation
 0.22% :     0.000106s :      1: auto_monad
 0.06% :     0.000028s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.50% :     0.000242s :      1: bootstrap
 0.06% :     0.000030s :      1: cconv
 0.02% :     0.000012s :      1: convert_after_rewriter
 0.08% :     0.000039s :      1: cse_after_recomputation
 0.02% :     0.000011s :      1: environ_conv
 0.13% :     0.000064s :      1: event_method
 0.03% :     0.000013s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
31.88% :     0.015461s :      1: jit_opt_a
 0.47% :     0.000228s :      1: jit_opt_after_cconv
 0.17% :     0.000083s :      1: jit_opt_b
 1.02% :     0.000493s :      1: loop_unroll
 1.72% :     0.000832s :      1: mutable_eliminate
 6.81% :     0.003303s :     39: opt.transform.jit_opt_a
 0.17% :     0.000082s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000049s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000024s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000037s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000056s :      4: opt.transform.symbol_engine_opt
 1.18% :     0.000570s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.10% :     0.000047s :      1: py_interpret_to_execute
 0.05% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000029s :      1: remove_dup_value
 5.13% :     0.002486s :      2: renormalize.infer
 4.32% :     0.002095s :      2: renormalize.specialize
 0.02% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000058s :      1: rewriter_after_opt_a
 0.22% :     0.000108s :      1: rewriter_before_opt_a
 0.22% :     0.000106s :      1: symbol_engine_optimizer
20.11% :     0.009750s :      1: task_emit
24.34% :     0.011803s :      1: type_inference
 0.17% :     0.000085s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:20.914.776 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c375160> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:20.916.214 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:20.918.478 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e3ca0> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:20.919.110 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.021174, [33]
[bootstrap]: 0.00030499
[type_inference]: 0.00463508
[event_method]: 1.779e-05
[auto_monad]: 6.443e-05
[graph_reusing]: 6.46999e-06
[pre_auto_parallel]: 2.37999e-06
[py_interpret_to_execute]: 3.311e-05
[rewriter_before_opt_a]: 7.245e-05
[expand_dump_flag]: 2.91999e-06
[jit_opt_a]: 0.00675031, [2]
    [Cycle 1]: 0.00204966, [27]
        [switch_simplify]: 6.555e-05
        [loop_unroll]: 3.062e-05
        [a_1]: 0.00073542
        [with_stream_mark]: 2.878e-05
        [recompute_prepare]: 1.304e-05
        [updatestate_depend_eliminate]: 5.07e-06
        [updatestate_assign_eliminate]: 3.85e-06
        [updatestate_loads_eliminate]: 2.83003e-06
        [parameter_eliminate]: 3.2e-06
        [specialize_transform]: 7.7e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 6.06e-06
        [meta_shard_fg_expand]: 2.46998e-06
        [get_grad_eliminate_]: 5.15999e-06
        [merge_forward]: 5.46e-06
        [cell_reuse_recompute_pass]: 1.47999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.847e-05
        [j_node_and_user_rematch]: 9.47999e-06
        [meta_fg_expand]: 2.64001e-06
        [replace_old_param]: 1.054e-05
        [inline_without_move]: 5.22e-06
        [renormalize]: 0.00077383
        [add_forward_monad_depend]: 7.65e-06
        [auto_monad_grad]: 2.71e-06
        [auto_monad_eliminator]: 1.68e-05
        [cse]: 3.017e-05
        [replace_applicator]: 1.8e-05
    [Cycle 2]: 0.00035241, [27]
        [switch_simplify]: 5.68002e-06
        [loop_unroll]: 5.04003e-06
        [a_1]: 9.182e-05
        [with_stream_mark]: 1.305e-05
        [recompute_prepare]: 6.11e-06
        [updatestate_depend_eliminate]: 3.67002e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 1.23002e-06
        [specialize_transform]: 4.99e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 5.44998e-06
        [meta_shard_fg_expand]: 1.41998e-06
        [get_grad_eliminate_]: 4.54002e-06
        [merge_forward]: 2.69001e-06
        [cell_reuse_recompute_pass]: 2.07999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.652e-05
        [j_node_and_user_rematch]: 8.98002e-06
        [meta_fg_expand]: 1.76003e-06
        [replace_old_param]: 8.17e-06
        [inline_without_move]: 4.60001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.80001e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 6.99001e-06
        [cse]: 1.204e-05
        [replace_applicator]: 5.32999e-06
[py_interpret_to_execute_after_opt_a]: 1.444e-05
[rewriter_after_opt_a]: 3.715e-05
[convert_after_rewriter]: 8.17998e-06
[order_py_execute_after_rewriter]: 5.46e-06
[mutable_eliminate]: 0.00079186
[jit_opt_b]: 5.46e-05, [1]
    [Cycle 1]: 4.628e-05, [2]
        [frontend_op_eliminate]: 1.813e-05
        [inline_after_opt_a]: 1.586e-05
[cconv]: 2.939e-05
[loop_unroll]: 0.00050125
[jit_opt_after_cconv]: 0.000156, [1]
    [Cycle 1]: 0.00014856, [11]
        [c_1]: 2.193e-05
        [parameter_eliminate]: 3.12002e-06
        [updatestate_depend_eliminate]: 6.45002e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 2.69e-05
        [call_graph_tuple_transform]: 2.17e-05
        [tuple_list_get_item_eliminator]: 6.09999e-06
        [none_parameter_eliminate]: 1.66998e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 5.35999e-06
[remove_dup_value]: 1.486e-05
[partial_unused_args_eliminate]: 2.46e-06
[environ_conv]: 6.06e-06
[add_recomputation]: 5.45e-05
[cse_after_recomputation]: 2.257e-05, [1]
    [Cycle 1]: 1.628e-05, [1]
        [cse]: 1.017e-05
[auto_monad_reorder]: 8.089e-05
[get_jit_bprop_graph]: 2.06e-06
[rewriter_after_jit_bprop_graph]: 8.23001e-06
[opt_after_jit_grad]: 0.00050205
[symbol_engine_optimizer]: 7.623e-05, [1]
    [Cycle 1]: 6.898e-05, [6]
        [build]: 5.12e-06
        [elim_shapecalc]: 7.87e-06
        [elim_not_effective]: 1.255e-05
        [opt_reshape]: 5.67999e-06
        [fold_const_symbol]: 9.17999e-06
        [renormalize]: 1.10001e-06
[validate]: 4.441e-05
[backend_pass]: 8.89995e-07
[task_emit]: 0.00665332
[execute]: 7.97e-06
Sums
    bootstrap                            :     0.000305s :  1.89%
    type_inference                       :     0.004635s : 28.72%
    event_method                         :     0.000018s :  0.11%
    auto_monad                           :     0.000064s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000033s :  0.21%
    rewriter_before_opt_a                :     0.000072s :  0.45%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000071s :  0.44%
    jit_opt_a.loop_unroll                :     0.000036s :  0.22%
    jit_opt_a.a_1                        :     0.000827s :  5.13%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000019s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000008s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000045s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000019s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.06%
    jit_opt_a.renormalize                :     0.000774s :  4.79%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.15%
    jit_opt_a.cse                        :     0.000042s :  0.26%
    jit_opt_a.replace_applicator         :     0.000023s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.09%
    rewriter_after_opt_a                 :     0.000037s :  0.23%
    convert_after_rewriter               :     0.000008s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000792s :  4.91%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.10%
    cconv                                :     0.000029s :  0.18%
    loop_unroll                          :     0.000501s :  3.11%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.17%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000055s :  0.34%
    cse_after_recomputation.cse          :     0.000010s :  0.06%
    auto_monad_reorder                   :     0.000081s :  0.50%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.05%
    opt_after_jit_grad                   :     0.000502s :  3.11%
    symbol_engine_optimizer.build        :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.28%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006653s : 41.22%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000277    23
 0.66% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000002s :      2: substitution.fold_const_symbol
 2.48% :     0.000007s :      3: substitution.graph_param_transform
85.41% :     0.000236s :      4: substitution.inline
 1.42% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.97% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.10% :     0.000006s :      2: substitution.replace_old_param
 5.33% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004563     2
82.52% :     0.003765s :      1: type_inference.infer
17.48% :     0.000798s :      1: type_inference.specialize
------[replace.]   0.000085     6
66.34% :     0.000056s :      4: replace.inline
33.66% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000247     6
94.48% :     0.000234s :      4: match.inline
 5.52% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000146   691
 1.47% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.28% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.86% :     0.000001s :     10: predicate.addn_check_dump
 1.03% :     0.000002s :     10: predicate.addn_zero_filter
 2.35% :     0.000003s :     10: predicate.arithmetic_simplify
 1.37% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.85% :     0.000001s :     10: predicate.compare_switch_simplify
 1.69% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.03% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.96% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.87% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.87% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.04% :     0.000002s :     10: predicate.environ_get_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.17% :     0.000000s :      3: predicate.fold_const_symbol
 0.75% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.26% :     0.000008s :     22: predicate.inline
 0.75% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.04% :     0.000002s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.50% :     0.000002s :     15: predicate.load_eliminater
 7.05% :     0.000010s :      3: predicate.loop_unroll_after_grad
 3.29% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.53% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.89% :     0.000001s :     10: predicate.merge_addn
 0.89% :     0.000001s :     10: predicate.minmaximum_grad
 1.52% :     0.000002s :      3: predicate.mutable_eliminate
 0.40% :     0.000001s :      3: predicate.opt_reshape
 2.01% :     0.000003s :     15: predicate.partial_eliminate
 0.90% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     10: predicate.reduce_eliminate
 1.17% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.68% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.42% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 9.41% :     0.000014s :     10: predicate.reshape_eliminate
 1.00% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.56% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.00% :     0.000001s :     10: predicate.same_eliminate
 0.39% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.70% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.15% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 0.97% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 1.99% :     0.000003s :     16: predicate.switch_defer_inline
 1.60% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.20% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000002s :     10: predicate.tile_eliminate
 1.34% :     0.000002s :     10: predicate.transpose_eliminate
 1.03% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.05% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.74% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.16% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.33% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.30% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000003s :     10: predicate.value_based_eliminate
 0.34% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000586    12
46.36% :     0.000272s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 9.55% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.09% :     0.000258s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023145    76
 0.25% :     0.000058s :      1: add_recomputation
 0.29% :     0.000068s :      1: auto_monad
 0.36% :     0.000084s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.39% :     0.000321s :      1: bootstrap
 0.14% :     0.000032s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.11% :     0.000025s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.09% :     0.000022s :      1: event_method
 0.05% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
29.18% :     0.006754s :      1: jit_opt_a
 0.69% :     0.000159s :      1: jit_opt_after_cconv
 0.25% :     0.000057s :      1: jit_opt_b
 2.21% :     0.000511s :      1: loop_unroll
 3.47% :     0.000803s :      1: mutable_eliminate
 4.66% :     0.001078s :     26: opt.transform.jit_opt_a
 0.22% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000027s :      4: opt.transform.jit_opt_b
 0.09% :     0.000021s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.21% :     0.000512s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000036s :      1: py_interpret_to_execute
 0.07% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 1.83% :     0.000423s :      1: renormalize.infer
 1.47% :     0.000341s :      1: renormalize.specialize
 0.05% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000040s :      1: rewriter_after_opt_a
 0.33% :     0.000076s :      1: rewriter_before_opt_a
 0.34% :     0.000079s :      1: symbol_engine_optimizer
28.80% :     0.006666s :      1: task_emit
20.10% :     0.004651s :      1: type_inference
 0.30% :     0.000068s :      1: validate


TotalTime = 0.010689, [30]
[bootstrap]: 0.00023552
[type_inference]: 0.00373524
[event_method]: 1.681e-05
[auto_monad]: 5.493e-05
[graph_reusing]: 5.87001e-06
[pre_auto_parallel]: 1.90001e-06
[py_interpret_to_execute]: 2.581e-05
[rewriter_before_opt_a]: 6.604e-05
[expand_dump_flag]: 2.51e-06
[jit_opt_a]: 0.00458796, [2]
    [Cycle 1]: 0.00130641, [27]
        [switch_simplify]: 4.641e-05
        [loop_unroll]: 2.887e-05
        [a_1]: 0.0004872
        [with_stream_mark]: 1.449e-05
        [recompute_prepare]: 7.06001e-06
        [updatestate_depend_eliminate]: 3.08998e-06
        [updatestate_assign_eliminate]: 2.45002e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [parameter_eliminate]: 1.64e-06
        [specialize_transform]: 5.64e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 1.45001e-06
        [get_grad_eliminate_]: 4.90001e-06
        [merge_forward]: 3.30998e-06
        [cell_reuse_recompute_pass]: 1.02998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.259e-05
        [j_node_and_user_rematch]: 9.08002e-06
        [meta_fg_expand]: 1.96998e-06
        [replace_old_param]: 8.17003e-06
        [inline_without_move]: 5.14998e-06
        [renormalize]: 0.00043578
        [add_forward_monad_depend]: 4.79e-06
        [auto_monad_grad]: 1.74e-06
        [auto_monad_eliminator]: 1.264e-05
        [cse]: 2.279e-05
        [replace_applicator]: 1.172e-05
    [Cycle 2]: 0.00031373, [27]
        [switch_simplify]: 5.32001e-06
        [loop_unroll]: 4.79002e-06
        [a_1]: 8.899e-05
        [with_stream_mark]: 7.72998e-06
        [recompute_prepare]: 4.90001e-06
        [updatestate_depend_eliminate]: 2.39001e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 5.04998e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 4.95001e-06
        [meta_shard_fg_expand]: 1.29998e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.26e-06
        [cell_reuse_recompute_pass]: 1.54e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.249e-05
        [j_node_and_user_rematch]: 7.61999e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 6.94999e-06
        [inline_without_move]: 4.61002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.81003e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 4.81002e-06
        [cse]: 9.92999e-06
        [replace_applicator]: 4.96002e-06
[py_interpret_to_execute_after_opt_a]: 8.17998e-06
[rewriter_after_opt_a]: 2.815e-05
[convert_after_rewriter]: 5.99e-06
[order_py_execute_after_rewriter]: 4.62998e-06
[mutable_eliminate]: 0.00050906
[jit_opt_b]: 4.854e-05, [1]
    [Cycle 1]: 4.204e-05, [2]
        [frontend_op_eliminate]: 1.553e-05
        [inline_after_opt_a]: 1.497e-05
[cconv]: 2.199e-05
[loop_unroll]: 0.00041153
[jit_opt_after_cconv]: 0.00013064, [1]
    [Cycle 1]: 0.00012478, [11]
        [c_1]: 1.96e-05
        [parameter_eliminate]: 2.40002e-06
        [updatestate_depend_eliminate]: 5.04e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.541e-05
        [call_graph_tuple_transform]: 1.75e-05
        [tuple_list_get_item_eliminator]: 5.32001e-06
        [none_parameter_eliminate]: 9.30013e-07
        [renormalize]: 4.40021e-07
        [switch_simplify]: 4.88001e-06
[remove_dup_value]: 1.173e-05
[partial_unused_args_eliminate]: 2.64001e-06
[environ_conv]: 4.24997e-06
[add_recomputation]: 4.29e-05
[cse_after_recomputation]: 1.959e-05, [1]
    [Cycle 1]: 1.436e-05, [1]
        [cse]: 8.92999e-06
[auto_monad_reorder]: 1.181e-05
[get_jit_bprop_graph]: 1.54998e-06
[rewriter_after_jit_bprop_graph]: 4.61002e-06
[opt_after_jit_grad]: 0.00043781
[symbol_engine_optimizer]: 6.993e-05, [1]
    [Cycle 1]: 6.333e-05, [6]
        [build]: 3.18e-06
        [elim_shapecalc]: 7.03e-06
        [elim_not_effective]: 1.176e-05
        [opt_reshape]: 5.61003e-06
        [fold_const_symbol]: 8.15e-06
        [renormalize]: 6.79982e-07
[validate]: 3.04e-05
Sums
    bootstrap                            :     0.000236s :  3.28%
    type_inference                       :     0.003735s : 52.03%
    event_method                         :     0.000017s :  0.23%
    auto_monad                           :     0.000055s :  0.77%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000026s :  0.36%
    rewriter_before_opt_a                :     0.000066s :  0.92%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000052s :  0.72%
    jit_opt_a.loop_unroll                :     0.000034s :  0.47%
    jit_opt_a.a_1                        :     0.000576s :  8.03%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000436s :  6.07%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.24%
    jit_opt_a.cse                        :     0.000033s :  0.46%
    jit_opt_a.replace_applicator         :     0.000017s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000028s :  0.39%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000509s :  7.09%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.21%
    cconv                                :     0.000022s :  0.31%
    loop_unroll                          :     0.000412s :  5.73%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000015s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000012s :  0.16%
    partial_unused_args_eliminate        :     0.000003s :  0.04%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000043s :  0.60%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000012s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000438s :  6.10%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000030s :  0.42%


Time group info:
------[substitution.]   0.000170    23
 1.07% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.90% :     0.000005s :      3: substitution.graph_param_transform
82.54% :     0.000140s :      4: substitution.inline
 1.75% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.46% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000004s :      2: substitution.replace_old_param
 6.56% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003681     2
80.75% :     0.002973s :      1: type_inference.infer
19.25% :     0.000709s :      1: type_inference.specialize
------[replace.]   0.000058     6
66.48% :     0.000039s :      4: replace.inline
33.52% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000148     6
93.24% :     0.000138s :      4: match.inline
 6.76% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.65% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.44% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000001s :     10: predicate.addn_zero_filter
 2.33% :     0.000003s :     10: predicate.arithmetic_simplify
 1.12% :     0.000001s :     10: predicate.cast_eliminate
 0.35% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000001s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.17% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000000s :      3: predicate.graph_param_transform
 5.28% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.03% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.04% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.77% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.36% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.56% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.61% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.58% :     0.000009s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.77% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.75% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.99% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000459    12
49.57% :     0.000228s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.84% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.59% :     0.000205s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011956    72
 0.38% :     0.000046s :      1: add_recomputation
 0.49% :     0.000058s :      1: auto_monad
 0.12% :     0.000014s :      1: auto_monad_reorder
 2.07% :     0.000247s :      1: bootstrap
 0.20% :     0.000024s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.18% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.18% :     0.000021s :      1: event_method
 0.04% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.07% :     0.000008s :      1: graph_reusing
38.39% :     0.004590s :      1: jit_opt_a
 1.12% :     0.000133s :      1: jit_opt_after_cconv
 0.43% :     0.000051s :      1: jit_opt_b
 3.50% :     0.000419s :      1: loop_unroll
 4.33% :     0.000517s :      1: mutable_eliminate
 6.45% :     0.000771s :     26: opt.transform.jit_opt_a
 0.37% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000029s :      4: opt.transform.symbol_engine_opt
 3.72% :     0.000445s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.04% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000029s :      1: py_interpret_to_execute
 0.09% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000014s :      1: remove_dup_value
 1.82% :     0.000217s :      1: renormalize.infer
 1.77% :     0.000212s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000031s :      1: rewriter_after_opt_a
 0.58% :     0.000069s :      1: rewriter_before_opt_a
 0.61% :     0.000072s :      1: symbol_engine_optimizer
31.34% :     0.003747s :      1: type_inference


TotalTime = 0.0326356, [33]
[bootstrap]: 0.00045381
[type_inference]: 0.0192622
[event_method]: 5.49e-05
[auto_monad]: 9.915e-05
[graph_reusing]: 7.3e-06
[pre_auto_parallel]: 2.74001e-06
[py_interpret_to_execute]: 2.878e-05
[rewriter_before_opt_a]: 9.298e-05
[expand_dump_flag]: 3.09001e-06
[jit_opt_a]: 0.0103319, [2]
    [Cycle 1]: 0.00499068, [27]
        [switch_simplify]: 7.498e-05
        [loop_unroll]: 4.16e-05
        [a_1]: 0.00090702
        [with_stream_mark]: 4.013e-05
        [recompute_prepare]: 2.756e-05
        [updatestate_depend_eliminate]: 9.45001e-06
        [updatestate_assign_eliminate]: 6.68e-06
        [updatestate_loads_eliminate]: 6.79999e-06
        [parameter_eliminate]: 3.4e-06
        [specialize_transform]: 1.328e-05
        [updatestate_useless_node_eliminater]: 1.271e-05
        [accelerated_algorithm]: 1.23e-05
        [meta_shard_fg_expand]: 4.05998e-06
        [get_grad_eliminate_]: 1.199e-05
        [merge_forward]: 8.27e-06
        [cell_reuse_recompute_pass]: 1.20999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.749e-05
        [j_node_and_user_rematch]: 2.317e-05
        [meta_fg_expand]: 0.00165369
        [replace_old_param]: 5.123e-05
        [inline_without_move]: 5.119e-05
        [renormalize]: 0.00173125
        [add_forward_monad_depend]: 1.016e-05
        [auto_monad_grad]: 1.44e-06
        [auto_monad_eliminator]: 9.36e-06
        [cse]: 1.718e-05
        [replace_applicator]: 1.248e-05
    [Cycle 2]: 0.00029834, [27]
        [switch_simplify]: 4.67e-06
        [loop_unroll]: 3.73999e-06
        [a_1]: 4.809e-05
        [with_stream_mark]: 8.19998e-06
        [recompute_prepare]: 3.56999e-06
        [updatestate_depend_eliminate]: 2.07001e-06
        [updatestate_assign_eliminate]: 1.71998e-06
        [updatestate_loads_eliminate]: 1.46002e-06
        [parameter_eliminate]: 1.39998e-06
        [specialize_transform]: 3.46001e-06
        [updatestate_useless_node_eliminater]: 3.5e-06
        [accelerated_algorithm]: 3.38999e-06
        [meta_shard_fg_expand]: 1.25999e-06
        [get_grad_eliminate_]: 3.41001e-06
        [merge_forward]: 2.29999e-06
        [cell_reuse_recompute_pass]: 2.84001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.04e-05
        [j_node_and_user_rematch]: 6.45002e-06
        [meta_fg_expand]: 4.489e-05
        [replace_old_param]: 4.33999e-06
        [inline_without_move]: 3.66001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 6.60017e-07
        [auto_monad_eliminator]: 4.55001e-06
        [cse]: 9.46998e-06
        [replace_applicator]: 3.78001e-06
[py_interpret_to_execute_after_opt_a]: 9.37001e-06
[rewriter_after_opt_a]: 2.967e-05
[convert_after_rewriter]: 5.55001e-06
[order_py_execute_after_rewriter]: 3.75e-06
[mutable_eliminate]: 0.00061375
[jit_opt_b]: 4.387e-05, [1]
    [Cycle 1]: 3.7e-05, [2]
        [frontend_op_eliminate]: 1.25e-05
        [inline_after_opt_a]: 1.237e-05
[cconv]: 1.987e-05
[loop_unroll]: 0.00048275
[jit_opt_after_cconv]: 0.00012203, [1]
    [Cycle 1]: 0.00011548, [11]
        [c_1]: 1.319e-05
        [parameter_eliminate]: 2.68998e-06
        [updatestate_depend_eliminate]: 5.07e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 1.60999e-06
        [cse]: 1.95e-05
        [call_graph_tuple_transform]: 1.242e-05
        [tuple_list_get_item_eliminator]: 3.92002e-06
        [none_parameter_eliminate]: 1.15999e-06
        [renormalize]: 6.60017e-07
        [switch_simplify]: 4.08999e-06
[remove_dup_value]: 9.94999e-06
[partial_unused_args_eliminate]: 1.50999e-06
[environ_conv]: 4.21001e-06
[add_recomputation]: 3.641e-05
[cse_after_recomputation]: 2.255e-05, [1]
    [Cycle 1]: 1.654e-05, [1]
        [cse]: 9.50001e-06
[auto_monad_reorder]: 8.38001e-06
[get_jit_bprop_graph]: 1.57001e-06
[rewriter_after_jit_bprop_graph]: 6.51e-06
[opt_after_jit_grad]: 0.00050416
[symbol_engine_optimizer]: 6.878e-05, [1]
    [Cycle 1]: 6.164e-05, [6]
        [build]: 2.98e-06
        [elim_shapecalc]: 6.91001e-0
TotalTime = 0.0101781, [30]
[bootstrap]: 0.0002374
[type_inference]: 0.00357504
[event_method]: 1.523e-05
[auto_monad]: 5.218e-05
[graph_reusing]: 5.41002e-06
[pre_auto_parallel]: 1.92999e-06
[py_interpret_to_execute]: 2.455e-05
[rewriter_before_opt_a]: 6.553e-05
[expand_dump_flag]: 2.49001e-06
[jit_opt_a]: 0.00429657, [2]
    [Cycle 1]: 0.00121518, [27]
        [switch_simplify]: 4.23e-05
        [loop_unroll]: 2.82e-05
        [a_1]: 0.00045861
        [with_stream_mark]: 1.156e-05
        [recompute_prepare]: 6.67002e-06
        [updatestate_depend_eliminate]: 3.33e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.21998e-06
        [parameter_eliminate]: 1.45001e-06
        [specialize_transform]: 5.65001e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 5.14e-06
        [merge_forward]: 2.79999e-06
        [cell_reuse_recompute_pass]: 8.39995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.177e-05
        [j_node_and_user_rematch]: 8.21002e-06
        [meta_fg_expand]: 2.13002e-06
        [replace_old_param]: 7.46001e-06
        [inline_without_move]: 5.15001e-06
        [renormalize]: 0.00039751
        [add_forward_monad_depend]: 4.2e-06
        [auto_monad_grad]: 1.25001e-06
        [auto_monad_eliminator]: 1.073e-05
        [cse]: 1.686e-05
        [replace_applicator]: 1.083e-05
    [Cycle 2]: 0.00030944, [27]
        [switch_simplify]: 5.86998e-06
        [loop_unroll]: 4.75999e-06
        [a_1]: 8.8e-05
        [with_stream_mark]: 7.97e-06
        [recompute_prepare]: 4.96002e-06
        [updatestate_depend_eliminate]: 2.35002e-06
        [updatestate_assign_eliminate]: 2.00002e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 4.80999e-06
        [updatestate_useless_node_eliminater]: 4.71002e-06
        [accelerated_algorithm]: 4.90001e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.27998e-06
        [merge_forward]: 2.46998e-06
        [cell_reuse_recompute_pass]: 1.16002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.204e-05
        [j_node_and_user_rematch]: 7.35e-06
        [meta_fg_expand]: 1.43002e-06
        [replace_old_param]: 6.73e-06
        [inline_without_move]: 8.67e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.23002e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 4.87998e-06
        [cse]: 1.026e-05
        [replace_applicator]: 4.75999e-06
[py_interpret_to_execute_after_opt_a]: 7.59002e-06
[rewriter_after_opt_a]: 2.593e-05
[convert_after_rewriter]: 5.74999e-06
[order_py_execute_after_rewriter]: 4.27e-06
[mutable_eliminate]: 0.00045834
[jit_opt_b]: 4.732e-05, [1]
    [Cycle 1]: 4.09e-05, [2]
        [frontend_op_eliminate]: 1.529e-05
        [inline_after_opt_a]: 1.44e-05
[cconv]: 1.974e-05
[loop_unroll]: 0.00039951
[jit_opt_after_cconv]: 0.0001293, [1]
    [Cycle 1]: 0.00012313, [11]
        [c_1]: 1.995e-05
        [parameter_eliminate]: 2.32999e-06
        [updatestate_depend_eliminate]: 4.92e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [cse]: 1.542e-05
        [call_graph_tuple_transform]: 1.704e-05
        [tuple_list_get_item_eliminator]: 5.34998e-06
        [none_parameter_eliminate]: 7.90023e-07
        [renormalize]: 4.59986e-07
        [switch_simplify]: 5.07999e-06
[remove_dup_value]: 1.113e-05
[partial_unused_args_eliminate]: 1.72999e-06
[environ_conv]: 4.53999e-06
[add_recomputation]: 3.619e-05
[cse_after_recomputation]: 1.951e-05, [1]
    [Cycle 1]: 1.426e-05, [1]
        [cse]: 8.85001e-06
[auto_monad_reorder]: 1.133e-05
[get_jit_bprop_graph]: 1.50999e-06
[rewriter_after_jit_bprop_graph]: 4.10998e-06
[opt_after_jit_grad]: 0.00047577
[symbol_engine_optimizer]: 6.882e-05, [1]
    [Cycle 1]: 6.269e-05, [6]
        [build]: 2.72001e-06
        [elim_shapecalc]:6
        [elim_not_effective]: 1.086e-05
        [opt_reshape]: 4.82e-06
        [fold_const_symbol]: 6.78e-06
        [renormalize]: 4.10015e-07
[validate]: 2.731e-05
[backend_pass]: 8.70001e-07
[task_emit]: 3.266e-05
[execute]: 1.39e-06
Sums
    bootstrap                            :     0.000454s :  1.69%
    type_inference                       :     0.019262s : 71.63%
    event_method                         :     0.000055s :  0.20%
    auto_monad                           :     0.000099s :  0.37%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000029s :  0.11%
    rewriter_before_opt_a                :     0.000093s :  0.35%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000080s :  0.30%
    jit_opt_a.loop_unroll                :     0.000045s :  0.17%
    jit_opt_a.a_1                        :     0.000955s :  3.55%
    jit_opt_a.with_stream_mark           :     0.000048s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000031s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.06%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.06%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.06%
    jit_opt_a.merge_forward              :     0.000011s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000030s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.001699s :  6.32%
    jit_opt_a.replace_old_param          :     0.000056s :  0.21%
    jit_opt_a.inline_without_move        :     0.000055s :  0.20%
    jit_opt_a.renormalize                :     0.001731s :  6.44%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.05%
    jit_opt_a.cse                        :     0.000027s :  0.10%
    jit_opt_a.replace_applicator         :     0.000016s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.03%
    rewriter_after_opt_a                 :     0.000030s :  0.11%
    convert_after_rewriter               :     0.000006s :  0.02%
    order_py_execute_after_rewriter      :     0.000004s :  0.01%
    mutable_eliminate                    :     0.000614s :  2.28%
    jit_opt_b.frontend_op_eliminate      :     0.000013s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.05%
    cconv                                :     0.000020s :  0.07%
    loop_unroll                          :     0.000483s :  1.80%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.05%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000012s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000010s 7.42002e-06
        [elim_not_effective]: 1.168e-05
        [opt_reshape]: 5.15999e-06
        [fold_const_symbol]: 8.45999e-06
        [renormalize]: 4.50003e-07
[validate]: 2.503e-05
Sums
    bootstrap                            :     0.000237s :  3.45%
    type_inference                       :     0.003575s : 51.95%
    event_method                         :     0.000015s :  0.22%
    auto_monad                           :     0.000052s :  0.76%
    graph_reusing                        :     0.000005s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000025s :  0.36%
    rewriter_before_opt_a                :     0.000066s :  0.95%
    expand_dump_flag                     :     0.000002s :  0.04%
    jit_opt_a.switch_simplify            :     0.000048s :  0.70%
    jit_opt_a.loop_unroll                :     0.000033s :  0.48%
    jit_opt_a.a_1                        :     0.000547s :  7.94%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.04%
    jit_opt_a.specialize_transform       :     0.000010s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000014s :  0.21%
    jit_opt_a.inline_without_move        :     0.000014s :  0.20%
    jit_opt_a.renormalize                :     0.000398s :  5.78%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.23%
    jit_opt_a.cse                        :     0.000027s :  0.39%
    jit_opt_a.replace_applicator         :     0.000016s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000026s :  0.38%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000004s :  0.06%
    mutable_eliminate                    :     0.000458s :  6.66%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.21%
    cconv                                :     0.000020s :  0.29%
    loop_unroll                          :     0.000400s :  5.81%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.29%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000015s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.25%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000011s :  0.16%
    partial_unused_args_eliminate        :   :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000036s :  0.14%
    cse_after_recomputation.cse          :     0.000010s :  0.04%
    auto_monad_reorder                   :     0.000008s :  0.03%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000504s :  1.87%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000007s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.10%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000033s :  0.12%
    execute                              :     0.000001s :  0.01%

   0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000036s :  0.53%
    cse_after_recomputation.cse          :     0.000009s :  0.13%
    auto_monad_reorder                   :     0.000011s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.06%
    opt_after_jit_grad                   :     0.000476s :  6.91%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000025s :  0.36%


Time group info:
------[substitution.]   0.000149    23
 1.21% :     0.000002s :      2: substitution.elim_not_effective
 0.85% :     0.000001s :      2: substitution.fold_const_symbol
 2.98% :     0.000004s :      3: substitution.graph_param_transform
82.75% :     0.000123s :      4: substitution.inline
 1.68% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.50% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.76% :     0.000003s :      2: substitution.replace_old_param
 6.27% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003529     2
80.60% :     0.002844s :      1: type_inference.infer
19.40% :     0.000685s :      1: type_inference.specialize
------[replace.]   0.000057     6
66.61% :     0.000038s :      4: replace.inline
33.39% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000129     6
93.63% :     0.000121s :      4: match.inline
 6.37% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000108   691
 1.51% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.17% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000001s :     10: predicate.addn_zero_filter
 2.01% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000001s :     10: predicate.depend_value_elim
 1.32% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.22% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.40% :     0.000006s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000001s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.17% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.09% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.48% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000002s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.82% :     0.000002s :     18: predicate.replace_applicator
 0.79% :     0.000001s :      6: predicate.replace_old_param
 0.37% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.81% :     0.000001s :      6: predicate.special_op_eliminate
 1.17% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.61% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.63% :     0.000008s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.57% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.73% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.80% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000442    12
49.51% :     0.000219s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.56% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.93% :     0.000198s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011383    72
 0.34% :     0.000039s :      1: add_recomputation
 0.49% :     0.000056s :      1: auto_monad
 0.12% :     0.000014s :      1: auto_monad_reorder
 2.18% :     0.000248s :      1: bootstrap
 0.20% :     0.000022s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.19% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.17% :     0.000019s :      1: event_method
 0.04% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.07% :     0.000008s :      1: graph_reusing
37.77% :     0.004299s :      1: jit_opt_a
 1.16% :     0.000132s :      1: jit_opt_after_cconv
 0.44% :     0.000050s :      1: jit_opt_b
 3.57% :     0.000407s :      1: loop_unroll
 4.10% :     0.000467s :      1: mutable_eliminate
 6.49% :     0.000738s :     26: opt.transform.jit_opt_a
 0.39% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000024s :      4: opt.transform.jit_opt_b
 0.09% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000030s :      4: opt.transform.symbol_engine_opt
 4.25% :     0.000483s :      1: opt_after_jit_grad
 0.06% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000027s :      1: py_interpret_to_execute
 0.09% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000013s :      1: remove_dup_value
 1.66% :     0.000189s :      1: renormalize.infer
 1.77% :     0.000202s :      1: renormalize.specialize
 0.05% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000029s :      1: rewriter_after_opt_a
 0.60% :     0.000068s :      1: rewriter_before_opt_a
 0.63% :     0.000071s :      1: symbol_engine_optimizer
31.51% :     0.003586s :      1: type_inference


Time group info:
------[substitution.]   0.000291    50
 0.59% :     0.000002s :      1: substitution.elim_not_effective
 0.43% :     0.000001s :      1: substitution.fold_const_symbol
 0.98% :     0.000003s :      1: substitution.graph_param_transform
68.88% :     0.000201s :      6: substitution.inline
 5.63% :     0.000016s :      2: substitution.inline_without_move
 2.19% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.50% :     0.000004s :      2: substitution.minmaximum_grad
 1.42% :     0.000004s :      3: substitution.partial_eliminate
 2.40% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.34% :     0.000007s :      2: substitution.replace_applicator
 1.11% :     0.000003s :      5: substitution.replace_old_param
 1.94% :     0.000006s :      1: substitution.set_cell_output_no_recompute
 2.73% :     0.000008s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.97% :     0.000006s :      2: substitution.tuple_list_get_item_depend_reorder
 5.88% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.019164     2
93.54% :     0.017926s :      1: type_inference.infer
 6.46% :     0.001238s :      1: type_inference.specialize
------[replace.]   0.000072     8
78.74% :     0.000057s :      6: replace.inline
21.26% :     0.000015s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000209     8
94.25% :     0.000197s :      6: match.inline
 5.75% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000182  1109
 1.59% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.66% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.16% :     0.000002s :     18: predicate.addn_check_dump
 1.49% :     0.000003s :     18: predicate.addn_zero_filter
 2.37% :     0.000004s :     18: predicate.arithmetic_simplify
 1.53% :     0.000003s :     18: predicate.cast_eliminate
 0.15% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.25% :     0.000002s :     18: predicate.compare_switch_simplify
 1.83% :     0.000003s :     18: predicate.depend_value_elim
 1.37% :     0.000003s :     18: predicate.dict_get_item_const_eliminator
 1.39% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.41% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      1: predicate.elim_not_effective
 0.27% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.36% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.19% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     18: predicate.environ_get_eliminate
 1.18% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.10% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.31% :     0.000008s :     28: predicate.inline
 3.05% :     0.000006s :     33: predicate.inline_without_move
 0.50% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.42% :     0.000003s :     12: predicate.less_batch_normalization
 1.55% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.73% :     0.000003s :     21: predicate.load_eliminater
 0.86% :     0.000002s :      1: predicate.loop_unroll_after_grad
 3.63% :     0.000007s :     44: predicate.loop_unroll_before_grad
 1.68% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.41% :     0.000003s :     18: predicate.merge_addn
 1.33% :     0.000002s :     18: predicate.minmaximum_grad
 0.95% :     0.000002s :      1: predicate.mutable_eliminate
 0.25% :     0.000000s :      1: predicate.opt_reshape
 2.13% :     0.000004s :     21: predicate.partial_eliminate
 1.44% :     0.000003s :     18: predicate.print_const_string_wrapper
 1.77% :     0.000003s :     18: predicate.reduce_eliminate
 1.81% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.30% :     0.000002s :     22: predicate.replace_applicator
 1.61% :     0.000003s :     33: predicate.replace_old_param
 0.08% :     0.000000s :      1: predicate.reset_defer_inline
 1.55% :     0.000003s :     18: predicate.reshape_eliminate
 1.41% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     18: predicate.same_eliminate
 0.61% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.45% :     0.000001s :      2: predicate.special_op_eliminate
 1.08% :     0.000002s :     12: predicate.specialize_transform
 1.57% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.61% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.13% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 3.23% :     0.000006s :     26: predicate.switch_defer_inline
 2.21% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.92% :     0.000013s :     71: predicate.switch_simplify
 1.40% :     0.000003s :     18: predicate.tile_eliminate
 1.31% :     0.000002s :     18: predicate.transpose_eliminate
 1.56% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.42% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.66% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.56% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000003s :     18: predicate.value_based_eliminate
 0.11% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.36% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001522    22
77.97% :     0.001187s :     13: func_graph_cloner_run.FuncGraphClonerGraph
22.03% :     0.000335s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.035725    76
 0.11% :     0.000039s :      1: add_recomputation
 0.29% :     0.000103s :      1: auto_monad
 0.03% :     0.000011s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.33% :     0.000477s :      1: bootstrap
 0.06% :     0.000023s :      1: cconv
 0.02% :     0.000008s :      1: convert_after_rewriter
 0.07% :     0.000025s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.17% :     0.000060s :      1: event_method
 0.01% :     0.000004s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
28.93% :     0.010336s :      1: jit_opt_a
 0.35% :     0.000125s :      1: jit_opt_after_cconv
 0.13% :     0.000047s :      1: jit_opt_b
 1.38% :     0.000491s :      1: loop_unroll
 1.75% :     0.000625s :      1: mutable_eliminate
 3.74% :     0.001337s :     26: opt.transform.jit_opt_a
 0.08% :     0.000030s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000018s :      4: opt.transform.jit_opt_b
 0.03% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000017s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000026s :      4: opt.transform.symbol_engine_opt
 1.43% :     0.000512s :      1: opt_after_jit_grad
 0.02% :     0.000006s :      1: order_py_execute_after_rewriter
 0.01% :     0.000003s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000032s :      1: py_interpret_to_execute
 0.03% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000012s :      1: remove_dup_value
 3.73% :     0.001334s :      1: renormalize.infer
 1.09% :     0.000390s :      1: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.09% :     0.000032s :      1: rewriter_after_opt_a
 0.27% :     0.000096s :      1: rewriter_before_opt_a
 0.20% :     0.000072s :      1: symbol_engine_optimizer
 0.10% :     0.000037s :      1: task_emit
53.96% :     0.019277s :      1: type_inference
 0.12% :     0.000043s :      1: validate


TotalTime = 0.0185261, [33]
[bootstrap]: 0.00028807
[type_inference]: 0.00445603
[event_method]: 1.714e-05
[auto_monad]: 5.924e-05
[graph_reusing]: 5.02e-06
[pre_auto_parallel]: 2.89999e-06
[py_interpret_to_execute]: 3.05e-05
[rewriter_before_opt_a]: 6.968e-05
[expand_dump_flag]: 3.13e-06
[jit_opt_a]: 0.00525422, [2]
    [Cycle 1]: 0.00135471, [27]
        [switch_simplify]: 4.731e-05
        [loop_unroll]: 2.889e-05
        [a_1]: 0.00046917
        [with_stream_mark]: 1.302e-05
        [recompute_prepare]: 6.71e-06
        [updatestate_depend_eliminate]: 3.03e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 5.89e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 5.00999e-06
        [merge_forward]: 2.21e-06
        [cell_reuse_recompute_pass]: 1.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.154e-05
        [j_node_and_user_rematch]: 8.55999e-06
        [meta_fg_expand]: 1.74998e-06
        [replace_old_param]: 7.56001e-06
        [inline_without_move]: 5.22999e-06
        [renormalize]: 0.00052196
        [add_forward_monad_depend]: 4.08001e-06
        [auto_monad_grad]: 1.34e-06
        [auto_monad_eliminator]: 9.09e-06
        [cse]: 1.664e-05
        [replace_applicator]: 1.163e-05
    [Cycle 2]: 0.0003306, [27]
        [switch_simplify]: 5.44e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 8.709e-05
        [with_stream_mark]: 7.99002e-06
        [recompute_prepare]: 4.83001e-06
        [updatestate_depend_eliminate]: 2.53e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 2.63e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 5.52001e-06
        [updatestate_useless_node_eliminater]: 4.92999e-06
        [accelerated_algorithm]: 5.02999e-06
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.27999e-06
        [cell_reuse_recompute_pass]: 1.91e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.201e-05
        [j_node_and_user_rematch]: 7.51999e-06
        [meta_fg_expand]: 1.42e-06
        [replace_old_param]: 6.83998e-06
        [inline_without_move]: 4.47998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 5.34e-06
        [cse]: 1.084e-05
        [replace_applicator]: 4.74998e-06
[py_interpret_to_execute_after_opt_a]: 9.81e-06
[rewriter_after_opt_a]: 3.444e-05
[convert_after_rewriter]: 7.39002e-06
[order_py_execute_after_rewriter]: 4.76002e-06
[mutable_eliminate]: 0.00049202
[jit_opt_b]: 4.803e-05, [1]
    [Cycle 1]: 4.121e-05, [2]
        [frontend_op_eliminate]: 1.551e-05
        [inline_after_opt_a]: 1.445e-05
[cconv]: 2.099e-05
[loop_unroll]: 0.0004005
[jit_opt_after_cconv]: 0.00014277, [1]
    [Cycle 1]: 0.00013645, [11]
        [c_1]: 1.92e-05
        [parameter_eliminate]: 2.35002e-06
        [updatestate_depend_eliminate]: 5.04e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [cse]: 1.756e-05
        [call_graph_tuple_transform]: 1.791e-05
        [tuple_list_get_item_eliminator]: 4.97e-06
        [none_parameter_eliminate]: 1.47999e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 5.11002e-06
[remove_dup_value]: 1.267e-05
[partial_unused_args_eliminate]: 2.44999e-06
[environ_conv]: 4.87e-06
[add_recomputation]: 4.613e-05
[cse_after_recomputation]: 1.972e-05, [1]
    [Cycle 1]: 1.431e-05, [1]
        [cse]: 8.55999e-06
[auto_monad_reorder]: 1.538e-05
[get_jit_bprop_graph]: 1.94999e-06
[rewriter_after_jit_bprop_graph]: 5.89e-06
[opt_after_jit_grad]: 0.0004692
[symbol_engine_optimizer]: 7.013e-05, [1]
    [Cycle 1]: 6.413e-05, [6]
        [build]: 3.48e-06
        [elim_shapecalc]: 7.07997e-06
        [elim_not_effective]: 1.167e-05
        [opt_reshape]: 5.49998e-06
        [fold_const_symbol]: 8.90999e-06
        [renormalize]: 4.39992e-07
[validate]: 3.368e-05
[backend_pass]: 7.7e-07
[task_emit]: 0.00629842
[execute]: 8.27998e-06
Sums
    bootstrap                            :     0.000288s :  2.01%
    type_inference                       :     0.004456s : 31.05%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000059s :  0.41%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.21%
    rewriter_before_opt_a                :     0.000070s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.37%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000556s :  3.88%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000014s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000522s :  3.64%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.10%
    jit_opt_a.cse                        :     0.000027s :  0.19%
    jit_opt_a.replace_applicator         :     0.000016s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000492s :  3.43%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000021s :  0.15%
    loop_unroll                          :     0.000400s :  2.79%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000046s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000469s :  3.27%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006298s : 43.88%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000158    23
 1.14% :     0.000002s :      2: substitution.elim_not_effective
 0.93% :     0.000001s :      2: substitution.fold_const_symbol
 3.42% :     0.000005s :      3: substitution.graph_param_transform
83.59% :     0.000132s :      4: substitution.inline
 1.66% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.47% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.81% :     0.000003s :      2: substitution.replace_old_param
 4.98% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004396     2
81.46% :     0.003581s :      1: type_inference.infer
18.54% :     0.000815s :      1: type_inference.specialize
------[replace.]   0.000059     6
67.18% :     0.000040s :      4: replace.inline
32.82% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000136     6
94.93% :     0.000129s :      4: match.inline
 5.07% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.27% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.14% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000001s :     10: predicate.addn_zero_filter
 2.28% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000001s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000001s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.05% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.36% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.34% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.05% :     0.000001s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.32% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.21% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.19% :     0.000001s :     10: predicate.minmaximum_grad
 1.52% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000000s :      3: predicate.opt_reshape
 2.26% :     0.000002s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.88% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000001s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.13% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.65% :     0.000003s :     16: predicate.switch_defer_inline
 2.39% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.04% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.34% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.92% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000541    12
46.87% :     0.000254s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.94% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.19% :     0.000261s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019896    76
 0.25% :     0.000049s :      1: add_recomputation
 0.31% :     0.000062s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.51% :     0.000300s :      1: bootstrap
 0.12% :     0.000024s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.07% :     0.000013s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.43% :     0.005258s :      1: jit_opt_a
 0.73% :     0.000146s :      1: jit_opt_after_cconv
 0.25% :     0.000050s :      1: jit_opt_b
 2.05% :     0.000408s :      1: loop_unroll
 2.51% :     0.000500s :      1: mutable_eliminate
 3.77% :     0.000751s :     26: opt.transform.jit_opt_a
 0.22% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.40% :     0.000477s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.32% :     0.000263s :      1: renormalize.infer
 1.27% :     0.000253s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000037s :      1: rewriter_after_opt_a
 0.36% :     0.000073s :      1: rewriter_before_opt_a
 0.36% :     0.000072s :      1: symbol_engine_optimizer
31.75% :     0.006317s :      1: task_emit
22.46% :     0.004468s :      1: type_inference
 0.26% :     0.000052s :      1: validate


TotalTime = 0.0374605, [33]
[bootstrap]: 0.00023618
[type_inference]: 0.0114614
[event_method]: 6.044e-05
[auto_monad]: 0.00010596
[graph_reusing]: 7.35e-06
[pre_auto_parallel]: 3.52002e-06
[py_interpret_to_execute]: 3.289e-05
[rewriter_before_opt_a]: 0.00010364
[expand_dump_flag]: 3.91001e-06
[jit_opt_a]: 0.0143109, [3]
    [Cycle 1]: 0.00703441, [27]
        [switch_simplify]: 7.929e-05
        [loop_unroll]: 4.008e-05
        [a_1]: 0.00090847
        [with_stream_mark]: 2.859e-05
        [recompute_prepare]: 1.95e-05
        [updatestate_depend_eliminate]: 8.54998e-06
        [updatestate_assign_eliminate]: 6.51e-06
        [updatestate_loads_eliminate]: 6.17001e-06
        [parameter_eliminate]: 2.99999e-06
        [specialize_transform]: 1.296e-05
        [updatestate_useless_node_eliminater]: 1.228e-05
        [accelerated_algorithm]: 1.244e-05
        [meta_shard_fg_expand]: 4.15e-06
        [get_grad_eliminate_]: 1.174e-05
        [merge_forward]: 8.66002e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.583e-05
        [j_node_and_user_rematch]: 2.343e-05
        [meta_fg_expand]: 0.00176357
        [replace_old_param]: 6.42e-05
        [inline_without_move]: 5.901e-05
        [renormalize]: 0.00331136
        [add_forward_monad_depend]: 1.379e-05
        [auto_monad_grad]: 6.91001e-06
        [auto_monad_eliminator]: 6.267e-05
        [cse]: 0.00020399
        [replace_applicator]: 8.636e-05
    [Cycle 2]: 0.00256814, [27]
        [switch_simplify]: 3.906e-05
        [loop_unroll]: 3.761e-05
        [a_1]: 0.00119893
        [with_stream_mark]: 1.668e-05
        [recompute_prepare]: 1.057e-05
        [updatestate_depend_eliminate]: 5.05999e-06
        [updatestate_assign_eliminate]: 4.03999e-06
        [updatestate_loads_eliminate]: 3.6e-06
        [parameter_eliminate]: 1.38002e-06
        [specialize_transform]: 9.04998e-06
        [updatestate_useless_node_eliminater]: 8.57998e-06
        [accelerated_algorithm]: 8.42e-06
        [meta_shard_fg_expand]: 3.28e-06
        [get_grad_eliminate_]: 9.04e-06
        [merge_forward]: 4.08001e-06
        [cell_reuse_recompute_pass]: 1.72001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.76e-05
        [j_node_and_user_rematch]: 1.432e-05
        [meta_fg_expand]: 6.822e-05
        [replace_old_param]: 1.246e-05
        [inline_without_move]: 8.41002e-06
        [renormalize]: 0.00083165
        [add_forward_monad_depend]: 6.02999e-06
        [auto_monad_grad]: 1.82001e-06
        [auto_monad_eliminator]: 1.718e-05
        [cse]: 5.473e-05
        [replace_applicator]: 1.56e-05
    [Cycle 3]: 0.00063989, [27]
        [switch_simplify]: 8.88002e-06
        [loop_unroll]: 8.22e-06
        [a_1]: 0.00019436
        [with_stream_mark]: 1.07e-05
        [recompute_prepare]: 8.79003e-06
        [updatestate_depend_eliminate]: 5.17999e-06
        [updatestate_assign_eliminate]: 4.69002e-06
        [updatestate_loads_eliminate]: 3.89002e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 8.85001e-06
        [updatestate_useless_node_eliminater]: 8.05e-06
        [accelerated_algorithm]: 8.50001e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 7.92e-06
        [merge_forward]: 4.82e-06
        [cell_reuse_recompute_pass]: 2.20002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.875e-05
        [j_node_and_user_rematch]: 2.91e-05
        [meta_fg_expand]: 3.25998e-06
        [replace_old_param]: 2.683e-05
        [inline_without_move]: 2.583e-05
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.75001e-06
        [auto_monad_grad]: 1.02e-06
        [auto_monad_eliminator]: 1.132e-05
        [cse]: 2.691e-05
        [replace_applicator]: 8.17e-06
[py_interpret_to_execute_after_opt_a]: 1.363e-05
[rewriter_after_opt_a]: 4.216e-05
[convert_after_rewriter]: 7.88999e-06
[order_py_execute_after_rewriter]: 5.67001e-06
[mutable_eliminate]: 0.00067138
[jit_opt_b]: 6.966e-05, [1]
    [Cycle 1]: 6.299e-05, [2]
        [frontend_op_eliminate]: 2.75e-05
        [inline_after_opt_a]: 2.412e-05
[cconv]: 1.623e-05
[loop_unroll]: 0.00045132
[jit_opt_after_cconv]: 0.00019996, [1]
    [Cycle 1]: 0.00019286, [11]
        [c_1]: 3.591e-05
        [parameter_eliminate]: 3.00998e-06
        [updatestate_depend_eliminate]: 8.13001e-06
        [updatestate_assign_eliminate]: 4.07e-06
        [updatestate_loads_eliminate]: 3.79002e-06
        [cse]: 3.546e-05
        [call_graph_tuple_transform]: 2.8e-05
        [tuple_list_get_item_eliminator]: 9.49e-06
        [none_parameter_eliminate]: 1.29e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 9.68002e-06
[remove_dup_value]: 2.045e-05
[partial_unused_args_eliminate]: 1.70001e-06
[environ_conv]: 6.64999e-06
[add_recomputation]: 5.23e-05
[cse_after_recomputation]: 3.923e-05, [1]
    [Cycle 1]: 3.272e-05, [1]
        [cse]: 2.577e-05
[auto_monad_reorder]: 1.864e-05
[get_jit_bprop_graph]: 1.52001e-06
[rewriter_after_jit_bprop_graph]: 5.49e-06
[opt_after_jit_grad]: 0.00050883
[symbol_engine_optimizer]: 9.955e-05, [1]
    [Cycle 1]: 9.279e-05, [6]
        [build]: 5.42001e-06
        [elim_shapecalc]: 1.142e-05
        [elim_not_effective]: 2.16e-05
        [opt_reshape]: 9.34998e-06
        [fold_const_symbol]: 1.62e-05
        [renormalize]: 7.50006e-07
[validate]: 4.513e-05
[backend_pass]: 8.59989e-07
[task_emit]: 0.00862023
[execute]: 8.64e-06
Sums
    bootstrap                            :     0.000236s :  0.73%
    type_inference                       :     0.011461s : 35.34%
    event_method                         :     0.000060s :  0.19%
    auto_monad                           :     0.000106s :  0.33%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000004s :  0.01%
    py_interpret_to_execute              :     0.000033s :  0.10%
    rewriter_before_opt_a                :     0.000104s :  0.32%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000127s :  0.39%
    jit_opt_a.loop_unroll                :     0.000086s :  0.26%
    jit_opt_a.a_1                        :     0.002302s :  7.10%
    jit_opt_a.with_stream_mark           :     0.000056s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000039s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000019s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000014s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.02%
    jit_opt_a.specialize_transform       :     0.000031s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000029s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000029s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000009s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000029s :  0.09%
    jit_opt_a.merge_forward              :     0.000018s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000062s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000067s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.001835s :  5.66%
    jit_opt_a.replace_old_param          :     0.000103s :  0.32%
    jit_opt_a.inline_without_move        :     0.000093s :  0.29%
    jit_opt_a.renormalize                :     0.004143s : 12.78%
    jit_opt_a.add_forward_monad_depend   :     0.000022s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000010s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000091s :  0.28%
    jit_opt_a.cse                        :     0.000286s :  0.88%
    jit_opt_a.replace_applicator         :     0.000110s :  0.34%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.04%
    rewriter_after_opt_a                 :     0.000042s :  0.13%
    convert_after_rewriter               :     0.000008s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000671s :  2.07%
    jit_opt_b.frontend_op_eliminate      :     0.000027s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.07%
    cconv                                :     0.000016s :  0.05%
    loop_unroll                          :     0.000451s :  1.39%
    jit_opt_after_cconv.c_1              :     0.000036s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000035s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000010s :  0.03%
    remove_dup_value                     :     0.000020s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000052s :  0.16%
    cse_after_recomputation.cse          :     0.000026s :  0.08%
    auto_monad_reorder                   :     0.000019s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000509s :  1.57%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000022s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000016s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000045s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008620s : 26.58%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000653   126
 6.00% :     0.000039s :      2: substitution.arithmetic_simplify
 0.39% :     0.000003s :      5: substitution.elim_not_effective
 0.43% :     0.000003s :      5: substitution.fold_const_symbol
 0.89% :     0.000006s :      6: substitution.graph_param_transform
58.99% :     0.000385s :     12: substitution.inline
 3.50% :     0.000023s :      2: substitution.inline_without_move
 3.88% :     0.000025s :     18: substitution.j_node_and_user_rematch
 1.70% :     0.000011s :      7: substitution.minmaximum_grad
 1.07% :     0.000007s :      5: substitution.partial_eliminate
 2.01% :     0.000013s :     18: substitution.remove_not_recompute_node
 5.24% :     0.000034s :     10: substitution.replace_applicator
 3.82% :     0.000025s :      7: substitution.replace_old_param
 0.41% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.41% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.16% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 6.11% :     0.000040s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011359     2
87.57% :     0.009948s :      1: type_inference.infer
12.43% :     0.001412s :      1: type_inference.specialize
------[replace.]   0.000166    19
64.07% :     0.000107s :     12: replace.inline
35.93% :     0.000060s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000400    19
94.69% :     0.000378s :     12: match.inline
 5.31% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000427  2895
 1.68% :     0.000007s :     48: predicate.accumulaten_eliminater
 0.76% :     0.000003s :      6: predicate.ad_related_special_op_eliminate
 1.35% :     0.000006s :     48: predicate.addn_check_dump
 1.62% :     0.000007s :     48: predicate.addn_zero_filter
 2.14% :     0.000009s :     48: predicate.arithmetic_simplify
 1.80% :     0.000008s :     48: predicate.cast_eliminate
 0.22% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.33% :     0.000006s :     48: predicate.compare_switch_simplify
 1.60% :     0.000007s :     48: predicate.depend_value_elim
 1.45% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.62% :     0.000007s :     48: predicate.dict_get_item_eliminator
 1.45% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.30% :     0.000001s :      6: predicate.dumpgradient_eliminate
 0.16% :     0.000001s :      6: predicate.elim_not_effective
 0.29% :     0.000001s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.53% :     0.000007s :     48: predicate.environ_add_const_eliminate
 1.38% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.34% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.42% :     0.000006s :     48: predicate.environ_get_eliminate
 1.38% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.12% :     0.000001s :      6: predicate.fold_const_symbol
 0.94% :     0.000004s :     23: predicate.get_grad_eliminate
 0.12% :     0.000001s :      6: predicate.graph_param_transform
 4.31% :     0.000018s :     79: predicate.inline
 1.73% :     0.000007s :     44: predicate.inline_without_move
 0.39% :     0.000002s :     23: predicate.j_node_and_user_rematch
 0.97% :     0.000004s :     23: predicate.less_batch_normalization
 1.86% :     0.000008s :     55: predicate.list_to_tuple_eliminator_
 1.94% :     0.000008s :     61: predicate.load_eliminater
 0.74% :     0.000003s :      6: predicate.loop_unroll_after_grad
 3.12% :     0.000013s :     90: predicate.loop_unroll_before_grad
 1.93% :     0.000008s :     54: predicate.make_slice_get_slice_eliminator
 1.47% :     0.000006s :     48: predicate.merge_addn
 1.48% :     0.000006s :     48: predicate.minmaximum_grad
 0.61% :     0.000003s :      6: predicate.mutable_eliminate
 0.25% :     0.000001s :      6: predicate.opt_reshape
 2.24% :     0.000010s :     61: predicate.partial_eliminate
 1.40% :     0.000006s :     48: predicate.print_const_string_wrapper
 2.03% :     0.000009s :     48: predicate.reduce_eliminate
 1.73% :     0.000007s :     55: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.49% :     0.000011s :    108: predicate.replace_applicator
 0.90% :     0.000004s :     44: predicate.replace_old_param
 0.12% :     0.000001s :      6: predicate.reset_defer_inline
 1.65% :     0.000007s :     48: predicate.reshape_eliminate
 1.50% :     0.000006s :     48: predicate.row_tensor_add_zeros_like
 0.37% :     0.000002s :      6: predicate.row_tensor_eliminate
 1.45% :     0.000006s :     48: predicate.same_eliminate
 0.48% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.58% :     0.000002s :     12: predicate.special_op_eliminate
 0.92% :     0.000004s :     23: predicate.specialize_transform
 1.78% :     0.000008s :     48: predicate.split_environ_get_set_with_tuple_value
 1.54% :     0.000007s :     48: predicate.stack_unstack_eliminate
 0.20% :     0.000001s :      6: predicate.switch_call_monad_eliminater
 2.71% :     0.000012s :     67: predicate.switch_defer_inline
 2.23% :     0.000010s :     67: predicate.switch_layer_defer_inline
 5.92% :     0.000025s :    163: predicate.switch_simplify
 1.48% :     0.000006s :     48: predicate.tile_eliminate
 1.47% :     0.000006s :     48: predicate.transpose_eliminate
 1.88% :     0.000008s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.65% :     0.000007s :     48: predicate.tuple_list_get_item_depend_reorder
 3.21% :     0.000014s :     67: predicate.tuple_list_get_item_eliminator
 1.99% :     0.000008s :     48: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000008s :     55: predicate.tuple_to_list_eliminator_
 1.84% :     0.000008s :     61: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000012s :     84: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000008s :     48: predicate.value_based_eliminate
 0.19% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.31% :     0.000001s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001769    29
56.66% :     0.001002s :     13: func_graph_cloner_run.FuncGraphClonerGraph
43.34% :     0.000766s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.044852    91
 0.12% :     0.000055s :      1: add_recomputation
 0.25% :     0.000110s :      1: auto_monad
 0.05% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.56% :     0.000251s :      1: bootstrap
 0.04% :     0.000019s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000041s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.14% :     0.000065s :      1: event_method
 0.03% :     0.000013s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
31.92% :     0.014315s :      1: jit_opt_a
 0.45% :     0.000203s :      1: jit_opt_after_cconv
 0.16% :     0.000072s :      1: jit_opt_b
 1.03% :     0.000460s :      1: loop_unroll
 1.52% :     0.000681s :      1: mutable_eliminate
 6.82% :     0.003060s :     39: opt.transform.jit_opt_a
 0.18% :     0.000079s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000045s :      4: opt.transform.jit_opt_b
 0.04% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000035s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000055s :      4: opt.transform.symbol_engine_opt
 1.15% :     0.000518s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.08% :     0.000035s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000023s :      1: remove_dup_value
 4.94% :     0.002214s :      2: renormalize.infer
 4.26% :     0.001912s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000045s :      1: rewriter_after_opt_a
 0.24% :     0.000107s :      1: rewriter_before_opt_a
 0.23% :     0.000103s :      1: symbol_engine_optimizer
19.26% :     0.008639s :      1: task_emit
25.60% :     0.011484s :      1: type_inference
 0.15% :     0.000069s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:20.961.585 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dcaf8b0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:20.962.262 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:20.965.347 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e3940> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:20.965.986 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0101767, [30]
[bootstrap]: 0.00023142
[type_inference]: 0.00358464
[event_method]: 1.523e-05
[auto_monad]: 4.789e-05
[graph_reusing]: 4.55999e-06
[pre_auto_parallel]: 1.91998e-06
[py_interpret_to_execute]: 2.494e-05
[rewriter_before_opt_a]: 6.334e-05
[expand_dump_flag]: 2.92002e-06
[jit_opt_a]: 0.00432923, [2]
    [Cycle 1]: 0.00122043, [27]
        [switch_simplify]: 4.437e-05
        [loop_unroll]: 2.843e-05
        [a_1]: 0.00045868
        [with_stream_mark]: 1.225e-05
        [recompute_prepare]: 6.90002e-06
        [updatestate_depend_eliminate]: 3.2e-06
        [updatestate_assign_eliminate]: 2.45002e-06
        [updatestate_loads_eliminate]: 2.44999e-06
        [parameter_eliminate]: 1.37999e-06
        [specialize_transform]: 5.51998e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 1.71e-06
        [get_grad_eliminate_]: 4.91997e-06
        [merge_forward]: 2.69001e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.188e-05
        [j_node_and_user_rematch]: 8.2e-06
        [meta_fg_expand]: 1.96003e-06
        [replace_old_param]: 7.41001e-06
        [inline_without_move]: 5.34998e-06
        [renormalize]: 0.00039937
        [add_forward_monad_depend]: 4.83001e-06
        [auto_monad_grad]: 1.45001e-06
        [auto_monad_eliminator]: 1.096e-05
        [cse]: 1.756e-05
        [replace_applicator]: 1.079e-05
    [Cycle 2]: 0.00030843, [27]
        [switch_simplify]: 5.28002e-06
        [loop_unroll]: 4.58001e-06
        [a_1]: 8.824e-05
        [with_stream_mark]: 7.59002e-06
        [recompute_prepare]: 4.66002e-06
        [updatestate_depend_eliminate]: 2.60002e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 9.20001e-07
        [specialize_transform]: 4.67998e-06
        [updatestate_useless_node_eliminater]: 4.74998e-06
        [accelerated_algorithm]: 4.85999e-06
        [meta_shard_fg_expand]: 1.01002e-06
        [get_grad_eliminate_]: 4.34002e-06
        [merge_forward]: 2.34999e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.179e-05
        [j_node_and_user_rematch]: 7.39002e-06
        [meta_fg_expand]: 1.42e-06
        [replace_old_param]: 7.21001e-06
        [inline_without_move]: 4.60999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 4.74e-06
        [cse]: 9.72999e-06
        [replace_applicator]: 5.16998e-06
[py_interpret_to_execute_after_opt_a]: 7.06999e-06
[rewriter_after_opt_a]: 2.512e-05
[convert_after_rewriter]: 5.52999e-06
[order_py_execute_after_rewriter]: 4.60001e-06
[mutable_eliminate]: 0.00044882
[jit_opt_b]: 5.996e-05, [1]
    [Cycle 1]: 5.357e-05, [2]
        [frontend_op_eliminate]: 1.601e-05
        [inline_after_opt_a]: 1.474e-05
[cconv]: 1.898e-05
[loop_unroll]: 0.00040509
[jit_opt_after_cconv]: 0.00013327, [1]
    [Cycle 1]: 0.00012736, [11]
        [c_1]: 1.942e-05
        [parameter_eliminate]: 2.27999e-06
        [updatestate_depend_eliminate]: 4.90999e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [cse]: 1.521e-05
        [call_graph_tuple_transform]: 1.856e-05
        [tuple_list_get_item_eliminator]: 5.77999e-06
        [none_parameter_eliminate]: 9.60019e-07
        [renormalize]: 4.70027e-07
        [switch_simplify]: 5.51e-06
[remove_dup_value]: 9.96998e-06
[partial_unused_args_eliminate]: 1.99999e-06
[environ_conv]: 4.04002e-06
[add_recomputation]: 3.608e-05
[cse_after_recomputation]: 1.866e-05, [1]
    [Cycle 1]: 1.369e-05, [1]
        [cse]: 8.2e-06
[auto_monad_reorder]: 1.05e-05
[get_jit_bprop_graph]: 1.53002e-06
[rewriter_after_jit_bprop_graph]: 4.55999e-06
[opt_after_jit_grad]: 0.00043618
[symbol_engine_optimizer]: 6.948e-05, [1]
    [Cycle 1]: 6.301e-05, [6]
        [build]: 2.93998e-06
        [elim_shapecalc]: 7.2e-06
        [elim_not_effective]: 1.162e-05
        [opt_reshape]: 5.69999e-06
        [fold_const_symbol]: 8.17998e-06
        [renormalize]: 4.39992e-07
[validate]: 2.541e-05
Sums
    bootstrap                            :     0.000231s :  3.39%
    type_inference                       :     0.003585s : 52.44%
    event_method                         :     0.000015s :  0.22%
    auto_monad                           :     0.000048s :  0.70%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000025s :  0.36%
    rewriter_before_opt_a                :     0.000063s :  0.93%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000050s :  0.73%
    jit_opt_a.loop_unroll                :     0.000033s :  0.48%
    jit_opt_a.a_1                        :     0.000547s :  8.00%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000010s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.15%
    jit_opt_a.renormalize                :     0.000399s :  5.84%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.23%
    jit_opt_a.cse                        :     0.000027s :  0.40%
    jit_opt_a.replace_applicator         :     0.000016s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.10%
    rewriter_after_opt_a                 :     0.000025s :  0.37%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000449s :  6.57%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.22%
    cconv                                :     0.000019s :  0.28%
    loop_unroll                          :     0.000405s :  5.93%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000015s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.27%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.08%
    remove_dup_value                     :     0.000010s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000036s :  0.53%
    cse_after_recomputation.cse          :     0.000008s :  0.12%
    auto_monad_reorder                   :     0.000010s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000436s :  6.38%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000025s :  0.37%


Time group info:
------[substitution.]   0.000150    23
 1.19% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 3.15% :     0.000005s :      3: substitution.graph_param_transform
82.59% :     0.000124s :      4: substitution.inline
 1.65% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.48% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.99% :     0.000003s :      2: substitution.replace_old_param
 6.23% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003537     2
80.54% :     0.002849s :      1: type_inference.infer
19.46% :     0.000688s :      1: type_inference.specialize
------[replace.]   0.000057     6
66.28% :     0.000038s :      4: replace.inline
33.72% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000130     6
93.64% :     0.000122s :      4: match.inline
 6.36% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.51% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.49% :     0.000002s :     10: predicate.addn_zero_filter
 1.72% :     0.000002s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.20% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.39% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.43% :     0.000000s :      3: predicate.graph_param_transform
 5.39% :     0.000006s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000001s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.19% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.33% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.19% :     0.000001s :     10: predicate.merge_addn
 1.17% :     0.000001s :     10: predicate.minmaximum_grad
 1.52% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000002s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 1.60% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000001s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.16% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.79% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.56% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.64% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.62% :     0.000008s :     53: predicate.switch_simplify
 1.34% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.59% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.84% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000444    12
49.45% :     0.000220s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.42% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.12% :     0.000201s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011382    72
 0.34% :     0.000039s :      1: add_recomputation
 0.45% :     0.000051s :      1: auto_monad
 0.11% :     0.000013s :      1: auto_monad_reorder
 2.14% :     0.000243s :      1: bootstrap
 0.19% :     0.000022s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.18% :     0.000021s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000019s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
38.06% :     0.004332s :      1: jit_opt_a
 1.19% :     0.000136s :      1: jit_opt_after_cconv
 0.55% :     0.000063s :      1: jit_opt_b
 3.62% :     0.000412s :      1: loop_unroll
 4.01% :     0.000457s :      1: mutable_eliminate
 6.46% :     0.000735s :     26: opt.transform.jit_opt_a
 0.40% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000024s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.90% :     0.000444s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000028s :      1: py_interpret_to_execute
 0.08% :     0.000009s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000012s :      1: remove_dup_value
 1.67% :     0.000190s :      1: renormalize.infer
 1.79% :     0.000203s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000028s :      1: rewriter_after_opt_a
 0.58% :     0.000066s :      1: rewriter_before_opt_a
 0.63% :     0.000072s :      1: symbol_engine_optimizer
31.59% :     0.003595s :      1: type_inference


TotalTime = 0.0177913, [33]
[bootstrap]: 0.00024757
[type_inference]: 0.00423309
[event_method]: 1.673e-05
[auto_monad]: 6.343e-05
[graph_reusing]: 5.54e-06
[pre_auto_parallel]: 2.80997e-06
[py_interpret_to_execute]: 3.06e-05
[rewriter_before_opt_a]: 6.998e-05
[expand_dump_flag]: 3.63e-06
[jit_opt_a]: 0.00533638, [2]
    [Cycle 1]: 0.00155838, [27]
        [switch_simplify]: 4.895e-05
        [loop_unroll]: 2.909e-05
        [a_1]: 0.00052069
        [with_stream_mark]: 1.578e-05
        [recompute_prepare]: 7.31001e-06
        [updatestate_depend_eliminate]: 2.96999e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 1.62001e-06
        [specialize_transform]: 6.11e-06
        [updatestate_useless_node_eliminater]: 5.89999e-06
        [accelerated_algorithm]: 5.29e-06
        [meta_shard_fg_expand]: 2.19999e-06
        [get_grad_eliminate_]: 4.85001e-06
        [merge_forward]: 3.50003e-06
        [cell_reuse_recompute_pass]: 1.32999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.19e-05
        [j_node_and_user_rematch]: 9.94001e-06
        [meta_fg_expand]: 1.89e-06
        [replace_old_param]: 8.23999e-06
        [inline_without_move]: 5.01002e-06
        [renormalize]: 0.00063389
        [add_forward_monad_depend]: 6.04999e-06
        [auto_monad_grad]: 2.49999e-06
        [auto_monad_eliminator]: 1.376e-05
        [cse]: 2.138e-05
        [replace_applicator]: 1.406e-05
    [Cycle 2]: 0.00039178, [27]
        [switch_simplify]: 5.30999e-06
        [loop_unroll]: 5.10999e-06
        [a_1]: 8.775e-05
        [with_stream_mark]: 1.014e-05
        [recompute_prepare]: 5.46e-06
        [updatestate_depend_eliminate]: 2.43998e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 1.86998e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.76998e-06
        [updatestate_useless_node_eliminater]: 4.78001e-06
        [accelerated_algorithm]: 5.13002e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 5.48e-05
        [merge_forward]: 4.18999e-06
        [cell_reuse_recompute_pass]: 2.23998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.543e-05
        [j_node_and_user_rematch]: 8.95999e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 7.86001e-06
        [inline_without_move]: 4.87e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.14999e-06
        [auto_monad_grad]: 1.63002e-06
        [auto_monad_eliminator]: 6.02999e-06
        [cse]: 1.366e-05
        [replace_applicator]: 5.29e-06
[py_interpret_to_execute_after_opt_a]: 1.242e-05
[rewriter_after_opt_a]: 3.596e-05
[convert_after_rewriter]: 6.56999e-06
[order_py_execute_after_rewriter]: 4.98001e-06
[mutable_eliminate]: 0.00062565
[jit_opt_b]: 5.161e-05, [1]
    [Cycle 1]: 4.354e-05, [2]
        [frontend_op_eliminate]: 1.629e-05
        [inline_after_opt_a]: 1.548e-05
[cconv]: 2.427e-05
[loop_unroll]: 0.00046163
[jit_opt_after_cconv]: 0.00014416, [1]
    [Cycle 1]: 0.00013785, [11]
        [c_1]: 2.029e-05
        [parameter_eliminate]: 3.43e-06
        [updatestate_depend_eliminate]: 4.97e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.21998e-06
        [cse]: 2.192e-05
        [call_graph_tuple_transform]: 1.922e-05
        [tuple_list_get_item_eliminator]: 5.24e-06
        [none_parameter_eliminate]: 1.77001e-06
        [renormalize]: 7.2e-07
        [switch_simplify]: 5.22999e-06
[remove_dup_value]: 1.256e-05
[partial_unused_args_eliminate]: 2.04e-06
[environ_conv]: 5.08002e-06
[add_recomputation]: 4.505e-05
[cse_after_recomputation]: 2.119e-05, [1]
    [Cycle 1]: 1.556e-05, [1]
        [cse]: 8.97999e-06
[auto_monad_reorder]: 1.46e-05
[get_jit_bprop_graph]: 1.50999e-06
[rewriter_after_jit_bprop_graph]: 5.91003e-06
[opt_after_jit_grad]: 0.00050782
[symbol_engine_optimizer]: 7.48e-05, [1]
    [Cycle 1]: 6.769e-05, [6]
        [build]: 3.67998e-06
        [elim_shapecalc]: 7.62998e-06
        [elim_not_effective]: 1.418e-05
        [opt_reshape]: 5.40999e-06
        [fold_const_symbol]: 8.25e-06
        [renormalize]: 6.59988e-07
[validate]: 3.703e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.00548893
[execute]: 5.91998e-06
Sums
    bootstrap                            :     0.000248s :  1.79%
    type_inference                       :     0.004233s : 30.68%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000063s :  0.46%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.22%
    rewriter_before_opt_a                :     0.000070s :  0.51%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000054s :  0.39%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000608s :  4.41%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000012s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000060s :  0.43%
    jit_opt_a.merge_forward              :     0.000008s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000634s :  4.60%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.14%
    jit_opt_a.cse                        :     0.000035s :  0.25%
    jit_opt_a.replace_applicator         :     0.000019s :  0.14%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.09%
    rewriter_after_opt_a                 :     0.000036s :  0.26%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000626s :  4.54%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000462s :  3.35%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000045s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000508s :  3.68%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.27%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005489s : 39.79%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000197    23
 1.49% :     0.000003s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 2.61% :     0.000005s :      3: substitution.graph_param_transform
81.77% :     0.000161s :      4: substitution.inline
 2.23% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.19% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.29% :     0.000005s :      2: substitution.replace_old_param
 6.91% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004173     2
81.73% :     0.003410s :      1: type_inference.infer
18.27% :     0.000762s :      1: type_inference.specialize
------[replace.]   0.000068     6
66.56% :     0.000045s :      4: replace.inline
33.44% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000171     6
92.63% :     0.000158s :      4: match.inline
 7.37% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.25% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.62% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.43% :     0.000002s :     10: predicate.addn_zero_filter
 2.18% :     0.000003s :     10: predicate.arithmetic_simplify
 1.48% :     0.000002s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.29% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.16% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.32% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.17% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.51% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000002s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.36% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.91% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.88% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.73% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.71% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.49% :     0.000002s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.53% :     0.000003s :     16: predicate.switch_defer_inline
 2.27% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.97% :     0.000008s :     53: predicate.switch_simplify
 1.39% :     0.000002s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.32% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.42% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.83% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.26% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 2.02% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000522    12
48.49% :     0.000253s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.74% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.76% :     0.000239s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019405    76
 0.25% :     0.000048s :      1: add_recomputation
 0.35% :     0.000067s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.34% :     0.000259s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.52% :     0.005340s :      1: jit_opt_a
 0.76% :     0.000147s :      1: jit_opt_after_cconv
 0.28% :     0.000054s :      1: jit_opt_b
 2.42% :     0.000470s :      1: loop_unroll
 3.28% :     0.000637s :      1: mutable_eliminate
 4.47% :     0.000868s :     26: opt.transform.jit_opt_a
 0.24% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.67% :     0.000517s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000034s :      1: py_interpret_to_execute
 0.08% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.65% :     0.000320s :      1: renormalize.infer
 1.58% :     0.000307s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000040s :      1: rewriter_after_opt_a
 0.38% :     0.000073s :      1: rewriter_before_opt_a
 0.40% :     0.000077s :      1: symbol_engine_optimizer
28.34% :     0.005500s :      1: task_emit
21.88% :     0.004246s :      1: type_inference
 0.30% :     0.000057s :      1: validate


TotalTime = 0.0241149, [33]
[bootstrap]: 0.00023406
[type_inference]: 0.0117369
[event_method]: 6.327e-05
[auto_monad]: 0.00010348
[graph_reusing]: 7.97998e-06
[pre_auto_parallel]: 3.09999e-06
[py_interpret_to_execute]: 3.03e-05
[rewriter_before_opt_a]: 9.962e-05
[expand_dump_flag]: 3.78001e-06
[jit_opt_a]: 0.0095719, [2]
    [Cycle 1]: 0.00510728, [27]
        [switch_simplify]: 7.165e-05
        [loop_unroll]: 3.801e-05
        [a_1]: 0.00087892
        [with_stream_mark]: 3.112e-05
        [recompute_prepare]: 1.915e-05
        [updatestate_depend_eliminate]: 8.08999e-06
        [updatestate_assign_eliminate]: 6.31998e-06
        [updatestate_loads_eliminate]: 6.58e-06
        [parameter_eliminate]: 2.58e-06
        [specialize_transform]: 1.326e-05
        [updatestate_useless_node_eliminater]: 1.18e-05
        [accelerated_algorithm]: 1.259e-05
        [meta_shard_fg_expand]: 2.96999e-06
        [get_grad_eliminate_]: 1.167e-05
        [merge_forward]: 8.20999e-06
        [cell_reuse_recompute_pass]: 1.04998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.712e-05
        [j_node_and_user_rematch]: 2.322e-05
        [meta_fg_expand]: 0.00166018
        [replace_old_param]: 5.224e-05
        [inline_without_move]: 5.12e-05
        [renormalize]: 0.00190325
        [add_forward_monad_depend]: 6.38998e-06
        [auto_monad_grad]: 2.83e-06
        [auto_monad_eliminator]: 1.333e-05
        [cse]: 2.637e-05
        [replace_applicator]: 1.351e-05
    [Cycle 2]: 0.00030192, [27]
        [switch_simplify]: 4.92e-06
        [loop_unroll]: 4.14002e-06
        [a_1]: 4.927e-05
        [with_stream_mark]: 9.58002e-06
        [recompute_prepare]: 3.76999e-06
        [updatestate_depend_eliminate]: 2.48998e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.39e-06
        [specialize_transform]: 3.46001e-06
        [updatestate_useless_node_eliminater]: 3.33e-06
        [accelerated_algorithm]: 3.81001e-06
        [meta_shard_fg_expand]: 1.77001e-06
        [get_grad_eliminate_]: 3.78999e-06
        [merge_forward]: 2.46998e-06
        [cell_reuse_recompute_pass]: 2.89999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.107e-05
        [j_node_and_user_rematch]: 5.62999e-06
        [meta_fg_expand]: 4.307e-05
        [replace_old_param]: 3.98001e-06
        [inline_without_move]: 3.36001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 4.18001e-06
        [cse]: 1.069e-05
        [replace_applicator]: 3.75e-06
[py_interpret_to_execute_after_opt_a]: 1.198e-05
[rewriter_after_opt_a]: 3.762e-05
[convert_after_rewriter]: 6.33e-06
[order_py_execute_after_rewriter]: 4.32e-06
[mutable_eliminate]: 0.00067329
[jit_opt_b]: 4.076e-05, [1]
    [Cycle 1]: 3.3e-05, [2]
        [frontend_op_eliminate]: 1.015e-05
        [inline_after_opt_a]: 1.096e-05
[cconv]: 2.278e-05
[loop_unroll]: 0.00046478
[jit_opt_after_cconv]: 0.00013321, [1]
    [Cycle 1]: 0.00012623, [11]
        [c_1]: 1.162e-05
        [parameter_eliminate]: 3.06999e-06
        [updatestate_depend_eliminate]: 4.62e-06
        [updatestate_assign_eliminate]: 1.72001e-06
        [updatestate_loads_eliminate]: 1.66e-06
        [cse]: 1.857e-05
        [call_graph_tuple_transform]: 1.318e-05
        [tuple_list_get_item_eliminator]: 4.13001e-06
        [none_parameter_eliminate]: 1.57999e-06
        [renormalize]: 6.79982e-07
        [switch_simplify]: 3.96001e-06
[remove_dup_value]: 1.683e-05
[partial_unused_args_eliminate]: 2.32999e-06
[environ_conv]: 5.44998e-06
[add_recomputation]: 3.581e-05
[cse_after_recomputation]: 2.118e-05, [1]
    [Cycle 1]: 1.5e-05, [1]
        [cse]: 9.16002e-06
[auto_monad_reorder]: 1.344e-05
[get_jit_bprop_graph]: 1.82001e-06
[rewriter_after_jit_bprop_graph]: 5.20001e-06
[opt_after_jit_grad]: 0.00045151
[symbol_engine_optimizer]: 6.614e-05, [1]
    [Cycle 1]: 5.979e-05, [6]
        [build]: 3.05002e-06
        [elim_shapecalc]: 6.21e-06
        [elim_not_effective]: 1.038e-05
        [opt_reshape]: 5.18002e-06
        [fold_const_symbol]: 6.50997e-06
        [renormalize]: 4.00003e-07
[validate]: 2.893e-05
[backend_pass]: 9.00007e-07
[task_emit]: 2.474e-05
[execute]: 1.47001e-06
Sums
    bootstrap                            :     0.000234s :  1.21%
    type_inference                       :     0.011737s : 60.78%
    event_method                         :     0.000063s :  0.33%
    auto_monad                           :     0.000103s :  0.54%
    graph_reusing                        :     0.000008s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.16%
    rewriter_before_opt_a                :     0.000100s :  0.52%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000077s :  0.40%
    jit_opt_a.loop_unroll                :     0.000042s :  0.22%
    jit_opt_a.a_1                        :     0.000928s :  4.81%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000023s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.08%
    jit_opt_a.merge_forward              :     0.000011s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001703s :  8.82%
    jit_opt_a.replace_old_param          :     0.000056s :  0.29%
    jit_opt_a.inline_without_move        :     0.000055s :  0.28%
    jit_opt_a.renormalize                :     0.001903s :  9.86%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.09%
    jit_opt_a.cse                        :     0.000037s :  0.19%
    jit_opt_a.replace_applicator         :     0.000017s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.06%
    rewriter_after_opt_a                 :     0.000038s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000673s :  3.49%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.06%
    cconv                                :     0.000023s :  0.12%
    loop_unroll                          :     0.000465s :  2.41%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000017s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000036s :  0.19%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000452s :  2.34%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000007s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000025s :  0.13%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000285    50
 0.61% :     0.000002s :      1: substitution.elim_not_effective
 0.44% :     0.000001s :      1: substitution.fold_const_symbol
 1.46% :     0.000004s :      1: substitution.graph_param_transform
69.17% :     0.000197s :      6: substitution.inline
 5.84% :     0.000017s :      2: substitution.inline_without_move
 2.43% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.27% :     0.000004s :      2: substitution.minmaximum_grad
 1.50% :     0.000004s :      3: substitution.partial_eliminate
 2.60% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.23% :     0.000006s :      2: substitution.replace_applicator
 1.27% :     0.000004s :      5: substitution.replace_old_param
 0.92% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.98% :     0.000008s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.66% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.61% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011632     2
88.58% :     0.010303s :      1: type_inference.infer
11.42% :     0.001329s :      1: type_inference.specialize
------[replace.]   0.000070     8
78.04% :     0.000055s :      6: replace.inline
21.96% :     0.000015s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000205     8
94.58% :     0.000194s :      6: match.inline
 5.42% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000175  1109
 1.73% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.53% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.22% :     0.000002s :     18: predicate.addn_check_dump
 1.75% :     0.000003s :     18: predicate.addn_zero_filter
 2.55% :     0.000004s :     18: predicate.arithmetic_simplify
 1.57% :     0.000003s :     18: predicate.cast_eliminate
 0.17% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.34% :     0.000002s :     18: predicate.compare_switch_simplify
 1.38% :     0.000002s :     18: predicate.depend_value_elim
 1.22% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.67% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.44% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      1: predicate.elim_not_effective
 0.19% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.65% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.25% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.23% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.33% :     0.000002s :     18: predicate.environ_get_eliminate
 1.19% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.18% :     0.000002s :     12: predicate.get_grad_eliminate
 0.20% :     0.000000s :      1: predicate.graph_param_transform
 4.81% :     0.000008s :     28: predicate.inline
 3.16% :     0.000006s :     33: predicate.inline_without_move
 0.47% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.56% :     0.000003s :     12: predicate.less_batch_normalization
 1.60% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.67% :     0.000003s :     21: predicate.load_eliminater
 0.76% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.57% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.85% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.30% :     0.000002s :     18: predicate.merge_addn
 1.27% :     0.000002s :     18: predicate.minmaximum_grad
 0.77% :     0.000001s :      1: predicate.mutable_eliminate
 0.23% :     0.000000s :      1: predicate.opt_reshape
 2.11% :     0.000004s :     21: predicate.partial_eliminate
 1.32% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.92% :     0.000003s :     18: predicate.reduce_eliminate
 1.52% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.70% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.43% :     0.000002s :     22: predicate.replace_applicator
 1.56% :     0.000003s :     33: predicate.replace_old_param
 0.09% :     0.000000s :      1: predicate.reset_defer_inline
 1.55% :     0.000003s :     18: predicate.reshape_eliminate
 1.53% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.53% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.69% :     0.000003s :     18: predicate.same_eliminate
 0.66% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.46% :     0.000001s :      2: predicate.special_op_eliminate
 1.15% :     0.000002s :     12: predicate.specialize_transform
 1.88% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.55% :     0.000004s :     26: predicate.switch_defer_inline
 2.28% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.46% :     0.000011s :     71: predicate.switch_simplify
 1.39% :     0.000002s :     18: predicate.tile_eliminate
 1.27% :     0.000002s :     18: predicate.transpose_eliminate
 1.53% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.45% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.12% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.67% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.87% :     0.000003s :     18: predicate.value_based_eliminate
 0.10% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.27% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001341    22
72.58% :     0.000973s :     13: func_graph_cloner_run.FuncGraphClonerGraph
27.42% :     0.000368s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027372    76
 0.14% :     0.000039s :      1: add_recomputation
 0.40% :     0.000109s :      1: auto_monad
 0.06% :     0.000016s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.92% :     0.000251s :      1: bootstrap
 0.09% :     0.000025s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.25% :     0.000068s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.03% :     0.000007s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
34.98% :     0.009576s :      1: jit_opt_a
 0.50% :     0.000136s :      1: jit_opt_after_cconv
 0.16% :     0.000043s :      1: jit_opt_b
 1.73% :     0.000472s :      1: loop_unroll
 2.49% :     0.000682s :      1: mutable_eliminate
 4.74% :     0.001297s :     26: opt.transform.jit_opt_a
 0.11% :     0.000030s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000014s :      4: opt.transform.jit_opt_b
 0.04% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000025s :      4: opt.transform.symbol_engine_opt
 1.68% :     0.000459s :      1: opt_after_jit_grad
 0.02% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000033s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 5.41% :     0.001480s :      1: renormalize.infer
 1.51% :     0.000414s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000041s :      1: rewriter_after_opt_a
 0.38% :     0.000103s :      1: rewriter_before_opt_a
 0.25% :     0.000069s :      1: symbol_engine_optimizer
 0.11% :     0.000029s :      1: task_emit
42.93% :     0.011752s :      1: type_inference
 0.17% :     0.000047s :      1: validate


TotalTime = 0.0100239, [30]
[bootstrap]: 0.00022784
[type_inference]: 0.00351468
[event_method]: 1.512e-05
[auto_monad]: 4.534e-05
[graph_reusing]: 4.44002e-06
[pre_auto_parallel]: 1.63002e-06
[py_interpret_to_execute]: 2.348e-05
[rewriter_before_opt_a]: 6.232e-05
[expand_dump_flag]: 2.53e-06
[jit_opt_a]: 0.00424718, [2]
    [Cycle 1]: 0.00120811, [27]
        [switch_simplify]: 4.226e-05
        [loop_unroll]: 2.89e-05
        [a_1]: 0.00045667
        [with_stream_mark]: 1.184e-05
        [recompute_prepare]: 6.89001e-06
        [updatestate_depend_eliminate]: 2.96001e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 2.13998e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 5.58002e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 5.56e-06
        [meta_shard_fg_expand]: 1.19998e-06
        [get_grad_eliminate_]: 5.18002e-06
        [merge_forward]: 2.58998e-06
        [cell_reuse_recompute_pass]: 8.2e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.198e-05
        [j_node_and_user_rematch]: 8.18999e-06
        [meta_fg_expand]: 1.91998e-06
        [replace_old_param]: 7.3e-06
        [inline_without_move]: 5.13002e-06
        [renormalize]: 0.00039633
        [add_forward_monad_depend]: 4.03001e-06
        [auto_monad_grad]: 1.68002e-06
        [auto_monad_eliminator]: 1.074e-05
        [cse]: 1.558e-05
        [replace_applicator]: 1.04e-05
    [Cycle 2]: 0.00030825, [27]
        [switch_simplify]: 5.32001e-06
        [loop_unroll]: 4.81002e-06
        [a_1]: 8.7e-05
        [with_stream_mark]: 8.13001e-06
        [recompute_prepare]: 4.86002e-06
        [updatestate_depend_eliminate]: 2.57001e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 9.00007e-07
        [specialize_transform]: 4.67998e-06
        [updatestate_useless_node_eliminater]: 4.52e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.30999e-06
        [merge_forward]: 2.20002e-06
        [cell_reuse_recompute_pass]: 1.08001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.168e-05
        [j_node_and_user_rematch]: 7.24001e-06
        [meta_fg_expand]: 1.38002e-06
        [replace_old_param]: 6.81001e-06
        [inline_without_move]: 4.80001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 4.95001e-06
        [cse]: 9.87999e-06
        [replace_applicator]: 5.49e-06
[py_interpret_to_execute_after_opt_a]: 7.33999e-06
[rewriter_after_opt_a]: 2.453e-05
[convert_after_rewriter]: 5.89999e-06
[order_py_execute_after_rewriter]: 4.63999e-06
[mutable_eliminate]: 0.00047529
[jit_opt_b]: 4.708e-05, [1]
    [Cycle 1]: 4.084e-05, [2]
        [frontend_op_eliminate]: 1.533e-05
        [inline_after_opt_a]: 1.427e-05
[cconv]: 1.891e-05
[loop_unroll]: 0.0004021
[jit_opt_after_cconv]: 0.00013011, [1]
    [Cycle 1]: 0.00012409, [11]
        [c_1]: 1.97e-05
        [parameter_eliminate]: 2.24999e-06
        [updatestate_depend_eliminate]: 5.07999e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.544e-05
        [call_graph_tuple_transform]: 1.696e-05
        [tuple_list_get_item_eliminator]: 5.47001e-06
        [none_parameter_eliminate]: 9.39996e-07
        [renormalize]: 4.80009e-07
        [switch_simplify]: 5.14e-06
[remove_dup_value]: 9.65002e-06
[partial_unused_args_eliminate]: 1.86003e-06
[environ_conv]: 4.12e-06
[add_recomputation]: 3.455e-05
[cse_after_recomputation]: 1.938e-05, [1]
    [Cycle 1]: 1.417e-05, [1]
        [cse]: 8.67998e-06
[auto_monad_reorder]: 1.066e-05
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 4.57e-06
[opt_after_jit_grad]: 0.00044181
[symbol_engine_optimizer]: 6.847e-05, [1]
    [Cycle 1]: 6.248e-05, [6]
        [build]: 3.49001e-06
        [elim_shapecalc]: 6.57002e-06
        [elim_not_effective]: 1.108e-05
        [opt_reshape]: 5.31002e-06
        [fold_const_symbol]: 8.32998e-06
        [renormalize]: 4.50003e-07
[validate]: 2.436e-05
Sums
    bootstrap                            :     0.000228s :  3.37%
    type_inference                       :     0.003515s : 51.94%
    event_method                         :     0.000015s :  0.22%
    auto_monad                           :     0.000045s :  0.67%
    graph_reusing                        :     0.000004s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000023s :  0.35%
    rewriter_before_opt_a                :     0.000062s :  0.92%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000048s :  0.70%
    jit_opt_a.loop_unroll                :     0.000034s :  0.50%
    jit_opt_a.a_1                        :     0.000544s :  8.03%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000010s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.16%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000015s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.05%
    jit_opt_a.replace_old_param          :     0.000014s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.15%
    jit_opt_a.renormalize                :     0.000396s :  5.86%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.23%
    jit_opt_a.cse                        :     0.000025s :  0.38%
    jit_opt_a.replace_applicator         :     0.000016s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.11%
    rewriter_after_opt_a                 :     0.000025s :  0.36%
    convert_after_rewriter               :     0.000006s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000475s :  7.02%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.21%
    cconv                                :     0.000019s :  0.28%
    loop_unroll                          :     0.000402s :  5.94%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.29%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000015s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.25%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.08%
    remove_dup_value                     :     0.000010s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000035s :  0.51%
    cse_after_recomputation.cse          :     0.000009s :  0.13%
    auto_monad_reorder                   :     0.000011s :  0.16%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000442s :  6.53%
    symbol_engine_optimizer.build        :     0.000003s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000024s :  0.36%


Time group info:
------[substitution.]   0.000151    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 3.27% :     0.000005s :      3: substitution.graph_param_transform
82.55% :     0.000125s :      4: substitution.inline
 1.69% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.51% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.78% :     0.000003s :      2: substitution.replace_old_param
 6.46% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003471     2
80.98% :     0.002811s :      1: type_inference.infer
19.02% :     0.000660s :      1: type_inference.specialize
------[replace.]   0.000056     6
66.47% :     0.000037s :      4: replace.inline
33.53% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000131     6
93.38% :     0.000122s :      4: match.inline
 6.62% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000107   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.54% :     0.000002s :     10: predicate.addn_zero_filter
 2.00% :     0.000002s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.41% :     0.000002s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.30% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.28% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.44% :     0.000000s :      3: predicate.graph_param_transform
 5.44% :     0.000006s :     22: predicate.inline
 1.06% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.33% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.77% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000002s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.82% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000001s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.81% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.53% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.81% :     0.000008s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.32% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.16% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.58% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000437    12
49.14% :     0.000214s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.87% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.99% :     0.000196s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011219    72
 0.33% :     0.000037s :      1: add_recomputation
 0.43% :     0.000049s :      1: auto_monad
 0.12% :     0.000013s :      1: auto_monad_reorder
 2.12% :     0.000238s :      1: bootstrap
 0.19% :     0.000022s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.19% :     0.000021s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.17% :     0.000019s :      1: event_method
 0.04% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
37.88% :     0.004249s :      1: jit_opt_a
 1.18% :     0.000133s :      1: jit_opt_after_cconv
 0.44% :     0.000050s :      1: jit_opt_b
 3.65% :     0.000409s :      1: loop_unroll
 4.31% :     0.000484s :      1: mutable_eliminate
 6.52% :     0.000731s :     26: opt.transform.jit_opt_a
 0.39% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000024s :      4: opt.transform.jit_opt_b
 0.10% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000028s :      4: opt.transform.symbol_engine_opt
 4.00% :     0.000449s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.23% :     0.000026s :      1: py_interpret_to_execute
 0.09% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000012s :      1: remove_dup_value
 1.69% :     0.000189s :      1: renormalize.infer
 1.79% :     0.000201s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000027s :      1: rewriter_after_opt_a
 0.58% :     0.000065s :      1: rewriter_before_opt_a
 0.64% :     0.000071s :      1: symbol_engine_optimizer
31.41% :     0.003524s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:20.993.561 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c375310> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:20.994.226 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0357067, [33]
[bootstrap]: 0.00021175
[type_inference]: 0.0109604
[event_method]: 5.756e-05
[auto_monad]: 9.547e-05
[graph_reusing]: 6.62002e-06
[pre_auto_parallel]: 2.74001e-06
[py_interpret_to_execute]: 3.005e-05
[rewriter_before_opt_a]: 9.413e-05
[expand_dump_flag]: 3.2e-06
[jit_opt_a]: 0.0130075, [3]
    [Cycle 1]: 0.00636585, [27]
        [switch_simplify]: 7.054e-05
        [loop_unroll]: 8.891e-05
        [a_1]: 0.00084618
        [with_stream_mark]: 2.473e-05
        [recompute_prepare]: 1.832e-05
        [updatestate_depend_eliminate]: 7.21999e-06
        [updatestate_assign_eliminate]: 5.76998e-06
        [updatestate_loads_eliminate]: 5.23002e-06
        [parameter_eliminate]: 1.99e-06
        [specialize_transform]: 1.295e-05
        [updatestate_useless_node_eliminater]: 1.218e-05
        [accelerated_algorithm]: 1.26e-05
        [meta_shard_fg_expand]: 2.99999e-06
        [get_grad_eliminate_]: 1.228e-05
        [merge_forward]: 9.14998e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.586e-05
        [j_node_and_user_rematch]: 2.297e-05
        [meta_fg_expand]: 0.00156259
        [replace_old_param]: 5.66e-05
        [inline_without_move]: 5.49e-05
        [renormalize]: 0.00298629
        [add_forward_monad_depend]: 1.311e-05
        [auto_monad_grad]: 6.38998e-06
        [auto_monad_eliminator]: 5.529e-05
        [cse]: 0.00016229
        [replace_applicator]: 7.323e-05
    [Cycle 2]: 0.00266097, [27]
        [switch_simplify]: 3.903e-05
        [loop_unroll]: 3.723e-05
        [a_1]: 0.00120366
        [with_stream_mark]: 1.97e-05
        [recompute_prepare]: 1.101e-05
        [updatestate_depend_eliminate]: 6.93e-06
        [updatestate_assign_eliminate]: 4.33999e-06
        [updatestate_loads_eliminate]: 3.76001e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 8.74998e-06
        [updatestate_useless_node_eliminater]: 8.84e-06
        [accelerated_algorithm]: 8.88002e-06
        [meta_shard_fg_expand]: 2.80002e-06
        [get_grad_eliminate_]: 9.42001e-06
        [merge_forward]: 5.37001e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.941e-05
        [j_node_and_user_rematch]: 1.49e-05
        [meta_fg_expand]: 6.1e-05
        [replace_old_param]: 1.433e-05
        [inline_without_move]: 8.48999e-06
        [renormalize]: 0.00090457
        [add_forward_monad_depend]: 6.46e-06
        [auto_monad_grad]: 2.32999e-06
        [auto_monad_eliminator]: 1.816e-05
        [cse]: 5.661e-05
        [replace_applicator]: 1.804e-05
    [Cycle 3]: 0.00051073, [27]
        [switch_simplify]: 8.74e-06
        [loop_unroll]: 8.27e-06
        [a_1]: 0.00019982
        [with_stream_mark]: 1.211e-05
        [recompute_prepare]: 8.69e-06
        [updatestate_depend_eliminate]: 4.89e-06
        [updatestate_assign_eliminate]: 4.37998e-06
        [updatestate_loads_eliminate]: 3.65e-06
        [parameter_eliminate]: 1.24998e-06
        [specialize_transform]: 8.23999e-06
        [updatestate_useless_node_eliminater]: 7.89002e-06
        [accelerated_algorithm]: 9.56e-06
        [meta_shard_fg_expand]: 2.44001e-06
        [get_grad_eliminate_]: 7.92e-06
        [merge_forward]: 4.52998e-06
        [cell_reuse_recompute_pass]: 3.26001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.714e-05
        [j_node_and_user_rematch]: 1.418e-05
        [meta_fg_expand]: 3.64002e-06
        [replace_old_param]: 1.062e-05
        [inline_without_move]: 8.49002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.24001e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 1.075e-05
        [cse]: 2.514e-05
        [replace_applicator]: 8.56002e-06
[py_interpret_to_execute_after_opt_a]: 1.569e-05
[rewriter_after_opt_a]: 5.096e-05
[convert_after_rewriter]: 9.19e-06
[order_py_execute_after_rewriter]: 6.94999e-06
[mutable_eliminate]: 0.00075882
[jit_opt_b]: 7.381e-05, [1]
    [Cycle 1]: 6.538e-05, [2]
        [frontend_op_eliminate]: 2.774e-05
        [inline_after_opt_a]: 2.594e-05
[cconv]: 3.92e-05
[loop_unroll]: 0.00046644
[jit_opt_after_cconv]: 0.00020429, [1]
    [Cycle 1]: 0.0001968, [11]
        [c_1]: 3.68e-05
        [parameter_eliminate]: 2.43002e-06
        [updatestate_depend_eliminate]: 7.78001e-06
        [updatestate_assign_eliminate]: 4.89998e-06
        [updatestate_loads_eliminate]: 3.80998e-06
        [cse]: 3.781e-05
        [call_graph_tuple_transform]: 2.852e-05
        [tuple_list_get_item_eliminator]: 9.05999e-06
        [none_parameter_eliminate]: 1.37e-06
        [renormalize]: 9.50007e-07
        [switch_simplify]: 8.65999e-06
[remove_dup_value]: 2.517e-05
[partial_unused_args_eliminate]: 2.31e-06
[environ_conv]: 8.2e-06
[add_recomputation]: 5.915e-05
[cse_after_recomputation]: 3.715e-05, [1]
    [Cycle 1]: 3.047e-05, [1]
        [cse]: 2.398e-05
[auto_monad_reorder]: 2.257e-05
[get_jit_bprop_graph]: 1.99e-06
[rewriter_after_jit_bprop_graph]: 6.22001e-06
[opt_after_jit_grad]: 0.00048356
[symbol_engine_optimizer]: 9.655e-05, [1]
    [Cycle 1]: 8.942e-05, [6]
        [build]: 5.26998e-06
        [elim_shapecalc]: 1.148e-05
        [elim_not_effective]: 1.942e-05
        [opt_reshape]: 9.24e-06
        [fold_const_symbol]: 1.539e-05
        [renormalize]: 6.69999e-07
[validate]: 4.933e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00859183
[execute]: 7.00998e-06
Sums
    bootstrap                            :     0.000212s :  0.67%
    type_inference                       :     0.010960s : 34.91%
    event_method                         :     0.000058s :  0.18%
    auto_monad                           :     0.000095s :  0.30%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000030s :  0.10%
    rewriter_before_opt_a                :     0.000094s :  0.30%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000118s :  0.38%
    jit_opt_a.loop_unroll                :     0.000134s :  0.43%
    jit_opt_a.a_1                        :     0.002250s :  7.17%
    jit_opt_a.with_stream_mark           :     0.000057s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000038s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000019s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000014s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000013s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000030s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000029s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000031s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000030s :  0.09%
    jit_opt_a.merge_forward              :     0.000019s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000062s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000052s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001627s :  5.18%
    jit_opt_a.replace_old_param          :     0.000082s :  0.26%
    jit_opt_a.inline_without_move        :     0.000072s :  0.23%
    jit_opt_a.renormalize                :     0.003891s : 12.39%
    jit_opt_a.add_forward_monad_depend   :     0.000022s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000010s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000084s :  0.27%
    jit_opt_a.cse                        :     0.000244s :  0.78%
    jit_opt_a.replace_applicator         :     0.000100s :  0.32%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.05%
    rewriter_after_opt_a                 :     0.000051s :  0.16%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000759s :  2.42%
    jit_opt_b.frontend_op_eliminate      :     0.000028s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000026s :  0.08%
    cconv                                :     0.000039s :  0.12%
    loop_unroll                          :     0.000466s :  1.49%
    jit_opt_after_cconv.c_1              :     0.000037s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000038s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000029s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.03%
    remove_dup_value                     :     0.000025s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000008s :  0.03%
    add_recomputation                    :     0.000059s :  0.19%
    cse_after_recomputation.cse          :     0.000024s :  0.08%
    auto_monad_reorder                   :     0.000023s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000484s :  1.54%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000049s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008592s : 27.37%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000601   126
 7.00% :     0.000042s :      2: substitution.arithmetic_simplify
 0.54% :     0.000003s :      5: substitution.elim_not_effective
 0.40% :     0.000002s :      5: substitution.fold_const_symbol
 1.23% :     0.000007s :      6: substitution.graph_param_transform
60.73% :     0.000365s :     12: substitution.inline
 3.34% :     0.000020s :      2: substitution.inline_without_move
 1.75% :     0.000010s :     18: substitution.j_node_and_user_rematch
 1.87% :     0.000011s :      7: substitution.minmaximum_grad
 1.25% :     0.000008s :      5: substitution.partial_eliminate
 2.17% :     0.000013s :     18: substitution.remove_not_recompute_node
 4.85% :     0.000029s :     10: substitution.replace_applicator
 1.57% :     0.000009s :      7: substitution.replace_old_param
 0.50% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.68% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.36% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 6.75% :     0.000041s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010827     2
87.25% :     0.009447s :      1: type_inference.infer
12.75% :     0.001380s :      1: type_inference.specialize
------[replace.]   0.000155    19
63.22% :     0.000098s :     12: replace.inline
36.78% :     0.000057s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000377    19
94.98% :     0.000358s :     12: match.inline
 5.02% :     0.000019s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000419  2895
 1.61% :     0.000007s :     48: predicate.accumulaten_eliminater
 0.57% :     0.000002s :      6: predicate.ad_related_special_op_eliminate
 1.43% :     0.000006s :     48: predicate.addn_check_dump
 1.64% :     0.000007s :     48: predicate.addn_zero_filter
 2.17% :     0.000009s :     48: predicate.arithmetic_simplify
 1.76% :     0.000007s :     48: predicate.cast_eliminate
 0.21% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.36% :     0.000006s :     48: predicate.compare_switch_simplify
 1.72% :     0.000007s :     48: predicate.depend_value_elim
 1.43% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.58% :     0.000007s :     48: predicate.dict_get_item_eliminator
 1.44% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.31% :     0.000001s :      6: predicate.dumpgradient_eliminate
 0.18% :     0.000001s :      6: predicate.elim_not_effective
 0.26% :     0.000001s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.52% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.38% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.47% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.44% :     0.000006s :     48: predicate.environ_get_eliminate
 1.42% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.11% :     0.000000s :      6: predicate.fold_const_symbol
 1.03% :     0.000004s :     23: predicate.get_grad_eliminate
 0.14% :     0.000001s :      6: predicate.graph_param_transform
 4.12% :     0.000017s :     79: predicate.inline
 1.73% :     0.000007s :     44: predicate.inline_without_move
 0.38% :     0.000002s :     23: predicate.j_node_and_user_rematch
 1.12% :     0.000005s :     23: predicate.less_batch_normalization
 1.87% :     0.000008s :     55: predicate.list_to_tuple_eliminator_
 1.91% :     0.000008s :     61: predicate.load_eliminater
 0.63% :     0.000003s :      6: predicate.loop_unroll_after_grad
 3.14% :     0.000013s :     90: predicate.loop_unroll_before_grad
 1.79% :     0.000007s :     54: predicate.make_slice_get_slice_eliminator
 1.54% :     0.000006s :     48: predicate.merge_addn
 1.46% :     0.000006s :     48: predicate.minmaximum_grad
 0.67% :     0.000003s :      6: predicate.mutable_eliminate
 0.24% :     0.000001s :      6: predicate.opt_reshape
 2.23% :     0.000009s :     61: predicate.partial_eliminate
 1.47% :     0.000006s :     48: predicate.print_const_string_wrapper
 1.80% :     0.000008s :     48: predicate.reduce_eliminate
 1.81% :     0.000008s :     55: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.55% :     0.000011s :    108: predicate.replace_applicator
 0.84% :     0.000004s :     44: predicate.replace_old_param
 0.19% :     0.000001s :      6: predicate.reset_defer_inline
 1.73% :     0.000007s :     48: predicate.reshape_eliminate
 1.49% :     0.000006s :     48: predicate.row_tensor_add_zeros_like
 0.36% :     0.000002s :      6: predicate.row_tensor_eliminate
 1.53% :     0.000006s :     48: predicate.same_eliminate
 0.46% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.51% :     0.000002s :     12: predicate.special_op_eliminate
 0.90% :     0.000004s :     23: predicate.specialize_transform
 1.73% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000006s :     48: predicate.stack_unstack_eliminate
 0.22% :     0.000001s :      6: predicate.switch_call_monad_eliminater
 2.56% :     0.000011s :     67: predicate.switch_defer_inline
 2.34% :     0.000010s :     67: predicate.switch_layer_defer_inline
 5.76% :     0.000024s :    163: predicate.switch_simplify
 1.55% :     0.000006s :     48: predicate.tile_eliminate
 1.53% :     0.000006s :     48: predicate.transpose_eliminate
 2.07% :     0.000009s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.69% :     0.000007s :     48: predicate.tuple_list_get_item_depend_reorder
 2.96% :     0.000012s :     67: predicate.tuple_list_get_item_eliminator
 1.98% :     0.000008s :     48: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000007s :     55: predicate.tuple_to_list_eliminator_
 1.92% :     0.000008s :     61: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000013s :     84: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000007s :     48: predicate.value_based_eliminate
 0.20% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.36% :     0.000002s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001716    29
55.84% :     0.000958s :     13: func_graph_cloner_run.FuncGraphClonerGraph
44.16% :     0.000758s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042773    91
 0.15% :     0.000062s :      1: add_recomputation
 0.24% :     0.000101s :      1: auto_monad
 0.06% :     0.000025s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.53% :     0.000227s :      1: bootstrap
 0.10% :     0.000042s :      1: cconv
 0.03% :     0.000012s :      1: convert_after_rewriter
 0.09% :     0.000039s :      1: cse_after_recomputation
 0.02% :     0.000010s :      1: environ_conv
 0.15% :     0.000062s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
30.42% :     0.013012s :      1: jit_opt_a
 0.49% :     0.000207s :      1: jit_opt_after_cconv
 0.18% :     0.000077s :      1: jit_opt_b
 1.11% :     0.000474s :      1: loop_unroll
 1.80% :     0.000771s :      1: mutable_eliminate
 6.97% :     0.002982s :     39: opt.transform.jit_opt_a
 0.18% :     0.000079s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000047s :      4: opt.transform.jit_opt_b
 0.04% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000037s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000053s :      4: opt.transform.symbol_engine_opt
 1.15% :     0.000492s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000033s :      1: py_interpret_to_execute
 0.04% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000028s :      1: remove_dup_value
 4.98% :     0.002129s :      2: renormalize.infer
 4.08% :     0.001745s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000054s :      1: rewriter_after_opt_a
 0.23% :     0.000097s :      1: rewriter_before_opt_a
 0.23% :     0.000099s :      1: symbol_engine_optimizer
20.12% :     0.008604s :      1: task_emit
25.67% :     0.010978s :      1: type_inference
 0.18% :     0.000076s :      1: validate


TotalTime = 0.0171293, [33]
[bootstrap]: 0.00026609
[type_inference]: 0.00437343
[event_method]: 1.566e-05
[auto_monad]: 6.006e-05
[graph_reusing]: 5.82001e-06
[pre_auto_parallel]: 2.40002e-06
[py_interpret_to_execute]: 2.797e-05
[rewriter_before_opt_a]: 6.702e-05
[expand_dump_flag]: 3.11001e-06
[jit_opt_a]: 0.00486338, [2]
    [Cycle 1]: 0.00139772, [27]
        [switch_simplify]: 4.864e-05
        [loop_unroll]: 2.844e-05
        [a_1]: 0.00049129
        [with_stream_mark]: 1.449e-05
        [recompute_prepare]: 6.98e-06
        [updatestate_depend_eliminate]: 2.71999e-06
        [updatestate_assign_eliminate]: 3.36999e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [parameter_eliminate]: 1.48002e-06
        [specialize_transform]: 5.64e-06
        [updatestate_useless_node_eliminater]: 5.10001e-06
        [accelerated_algorithm]: 5.08002e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 5.32999e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 1.34998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.383e-05
        [j_node_and_user_rematch]: 8.48001e-06
        [meta_fg_expand]: 2.16e-06
        [replace_old_param]: 8.80001e-06
        [inline_without_move]: 5.09e-06
        [renormalize]: 0.00053066
        [add_forward_monad_depend]: 4.46002e-06
        [auto_monad_grad]: 2.26e-06
        [auto_monad_eliminator]: 1.254e-05
        [cse]: 1.47e-05
        [replace_applicator]: 1.064e-05
    [Cycle 2]: 0.00030828, [27]
        [switch_simplify]: 5.44e-06
        [loop_unroll]: 4.75001e-06
        [a_1]: 8.757e-05
        [with_stream_mark]: 8.86002e-06
        [recompute_prepare]: 4.84e-06
        [updatestate_depend_eliminate]: 2.34001e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 4.63001e-06
        [updatestate_useless_node_eliminater]: 4.63001e-06
        [accelerated_algorithm]: 4.75001e-06
        [meta_shard_fg_expand]: 1.02998e-06
        [get_grad_eliminate_]: 4.28001e-06
        [merge_forward]: 2.21e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.149e-05
        [j_node_and_user_rematch]: 7.32002e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 6.88e-06
        [inline_without_move]: 4.70999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 4.57998e-06
        [cse]: 1.02e-05
        [replace_applicator]: 5.27001e-06
[py_interpret_to_execute_after_opt_a]: 9.29e-06
[rewriter_after_opt_a]: 3.047e-05
[convert_after_rewriter]: 7.17002e-06
[order_py_execute_after_rewriter]: 4.59998e-06
[mutable_eliminate]: 0.00051999
[jit_opt_b]: 4.82e-05, [1]
    [Cycle 1]: 4.069e-05, [2]
        [frontend_op_eliminate]: 1.541e-05
        [inline_after_opt_a]: 1.395e-05
[cconv]: 2.424e-05
[loop_unroll]: 0.00041078
[jit_opt_after_cconv]: 0.0001326, [1]
    [Cycle 1]: 0.00012659, [11]
        [c_1]: 1.985e-05
        [parameter_eliminate]: 2.61999e-06
        [updatestate_depend_eliminate]: 4.27e-06
        [updatestate_assign_eliminate]: 2.14999e-06
        [updatestate_loads_eliminate]: 2.18002e-06
        [cse]: 1.739e-05
        [call_graph_tuple_transform]: 1.797e-05
        [tuple_list_get_item_eliminator]: 5.00999e-06
        [none_parameter_eliminate]: 1.55001e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 5.05999e-06
[remove_dup_value]: 1.38e-05
[partial_unused_args_eliminate]: 2.24999e-06
[environ_conv]: 5.14998e-06
[add_recomputation]: 4.355e-05
[cse_after_recomputation]: 1.967e-05, [1]
    [Cycle 1]: 1.46e-05, [1]
        [cse]: 8.75999e-06
[auto_monad_reorder]: 1.444e-05
[get_jit_bprop_graph]: 1.77001e-06
[rewriter_after_jit_bprop_graph]: 5.51e-06
[opt_after_jit_grad]: 0.00050089
[symbol_engine_optimizer]: 6.868e-05, [1]
    [Cycle 1]: 6.245e-05, [6]
        [build]: 2.94001e-06
        [elim_shapecalc]: 7.28e-06
        [elim_not_effective]: 1.152e-05
        [opt_reshape]: 5.08002e-06
        [fold_const_symbol]: 8.22e-06
        [renormalize]: 6.80011e-07
[validate]: 2.861e-05
[backend_pass]: 8.59989e-07
[task_emit]: 0.0053774
[execute]: 6.86001e-06
Sums
    bootstrap                            :     0.000266s :  1.98%
    type_inference                       :     0.004373s : 32.59%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000060s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.21%
    rewriter_before_opt_a                :     0.000067s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.40%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000579s :  4.31%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000531s :  3.96%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.13%
    jit_opt_a.cse                        :     0.000025s :  0.19%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000030s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000520s :  3.88%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.18%
    loop_unroll                          :     0.000411s :  3.06%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000044s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000501s :  3.73%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000029s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005377s : 40.07%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000183    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.98% :     0.000005s :      3: substitution.graph_param_transform
82.91% :     0.000151s :      4: substitution.inline
 1.51% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.59% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.14% :     0.000004s :      2: substitution.replace_old_param
 6.24% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004319     2
82.33% :     0.003556s :      1: type_inference.infer
17.67% :     0.000763s :      1: type_inference.specialize
------[replace.]   0.000061     6
68.03% :     0.000042s :      4: replace.inline
31.97% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000159     6
93.51% :     0.000149s :      4: match.inline
 6.49% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000108   691
 1.36% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.21% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000001s :     10: predicate.addn_zero_filter
 1.80% :     0.000002s :     10: predicate.arithmetic_simplify
 1.39% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000001s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.27% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.32% :     0.000001s :      6: predicate.get_grad_eliminate
 0.43% :     0.000000s :      3: predicate.graph_param_transform
 5.37% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.75% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.32% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.88% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.33% :     0.000001s :      3: predicate.mutable_eliminate
 0.45% :     0.000000s :      3: predicate.opt_reshape
 2.13% :     0.000002s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.70% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 1.10% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000001s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.46% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.02% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.22% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.27% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.51% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.44% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000513    12
49.35% :     0.000253s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.55% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.10% :     0.000231s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018535    76
 0.25% :     0.000046s :      1: add_recomputation
 0.34% :     0.000063s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.48% :     0.000275s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.26% :     0.004867s :      1: jit_opt_a
 0.73% :     0.000136s :      1: jit_opt_after_cconv
 0.27% :     0.000051s :      1: jit_opt_b
 2.25% :     0.000418s :      1: loop_unroll
 2.85% :     0.000528s :      1: mutable_eliminate
 4.18% :     0.000775s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.74% :     0.000509s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 1.44% :     0.000267s :      1: renormalize.infer
 1.39% :     0.000257s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000033s :      1: rewriter_after_opt_a
 0.38% :     0.000070s :      1: rewriter_before_opt_a
 0.38% :     0.000071s :      1: symbol_engine_optimizer
29.07% :     0.005388s :      1: task_emit
23.64% :     0.004382s :      1: type_inference
 0.25% :     0.000047s :      1: validate

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.107.83 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e33a0> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.113.52 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0101168, [30]
[bootstrap]: 0.00022867
[type_inference]: 0.00356441
[event_method]: 1.534e-05
[auto_monad]: 4.733e-05
[graph_reusing]: 4.78001e-06
[pre_auto_parallel]: 1.64998e-06
[py_interpret_to_execute]: 2.438e-05
[rewriter_before_opt_a]: 6.316e-05
[expand_dump_flag]: 2.40002e-06
[jit_opt_a]: 0.00431243, [2]
    [Cycle 1]: 0.00123165, [27]
        [switch_simplify]: 4.388e-05
        [loop_unroll]: 2.783e-05
        [a_1]: 0.00045512
        [with_stream_mark]: 1.264e-05
        [recompute_prepare]: 6.94999e-06
        [updatestate_depend_eliminate]: 3.04999e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 5.89999e-06
        [updatestate_useless_node_eliminater]: 4.89998e-06
        [accelerated_algorithm]: 5.29e-06
        [meta_shard_fg_expand]: 1.54e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 8.80013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.119e-05
        [j_node_and_user_rematch]: 9.11998e-06
        [meta_fg_expand]: 2.36998e-06
        [replace_old_param]: 7.85e-06
        [inline_without_move]: 5.54998e-06
        [renormalize]: 0.00041293
        [add_forward_monad_depend]: 4.70001e-06
        [auto_monad_grad]: 1.32e-06
        [auto_monad_eliminator]: 1.128e-05
        [cse]: 1.692e-05
        [replace_applicator]: 1.036e-05
    [Cycle 2]: 0.00030657, [27]
        [switch_simplify]: 5.46e-06
        [loop_unroll]: 4.63999e-06
        [a_1]: 8.768e-05
        [with_stream_mark]: 7.70998e-06
        [recompute_prepare]: 4.77e-06
        [updatestate_depend_eliminate]: 2.36e-06
        [updatestate_assign_eliminate]: 1.96003e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 4.90001e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 4.85001e-06
        [meta_shard_fg_expand]: 9.70002e-07
        [get_grad_eliminate_]: 4.55001e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.14998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.141e-05
        [j_node_and_user_rematch]: 7.59002e-06
        [meta_fg_expand]: 1.39998e-06
        [replace_old_param]: 6.87002e-06
        [inline_without_move]: 4.50001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.11002e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 4.87e-06
        [cse]: 9.81e-06
        [replace_applicator]: 4.94998e-06
[py_interpret_to_execute_after_opt_a]: 7.33e-06
[rewriter_after_opt_a]: 2.411e-05
[convert_after_rewriter]: 5.69e-06
[order_py_execute_after_rewriter]: 4.80001e-06
[mutable_eliminate]: 0.00045476
[jit_opt_b]: 4.699e-05, [1]
    [Cycle 1]: 4.085e-05, [2]
        [frontend_op_eliminate]: 1.533e-05
        [inline_after_opt_a]: 1.452e-05
[cconv]: 1.851e-05
[loop_unroll]: 0.00039974
[jit_opt_after_cconv]: 0.0001291, [1]
    [Cycle 1]: 0.0001235, [11]
        [c_1]: 1.999e-05
        [parameter_eliminate]: 2.39001e-06
        [updatestate_depend_eliminate]: 4.97999e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [cse]: 1.497e-05
        [call_graph_tuple_transform]: 1.621e-05
        [tuple_list_get_item_eliminator]: 5.32001e-06
        [none_parameter_eliminate]: 8.90024e-07
        [renormalize]: 4.89992e-07
        [switch_simplify]: 5.25001e-06
[remove_dup_value]: 9.46e-06
[partial_unused_args_eliminate]: 1.86e-06
[environ_conv]: 3.88001e-06
[add_recomputation]: 3.561e-05
[cse_after_recomputation]: 1.853e-05, [1]
    [Cycle 1]: 1.347e-05, [1]
        [cse]: 7.98001e-06
[auto_monad_reorder]: 1.103e-05
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 4.46002e-06
[opt_after_jit_grad]: 0.00043464
[symbol_engine_optimizer]: 6.891e-05, [1]
    [Cycle 1]: 6.306e-05, [6]
        [build]: 3.35998e-06
        [elim_shapecalc]: 7.17002e-06
        [elim_not_effective]: 1.119e-05
        [opt_reshape]: 5.56e-06
        [fold_const_symbol]: 8.40999e-06
        [renormalize]: 5.00004e-07
[validate]: 2.427e-05
Sums
    bootstrap                            :     0.000229s :  3.36%
    type_inference                       :     0.003564s : 52.33%
    event_method                         :     0.000015s :  0.23%
    auto_monad                           :     0.000047s :  0.69%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000024s :  0.36%
    rewriter_before_opt_a                :     0.000063s :  0.93%
    expand_dump_flag                     :     0.000002s :  0.04%
    jit_opt_a.switch_simplify            :     0.000049s :  0.72%
    jit_opt_a.loop_unroll                :     0.000032s :  0.48%
    jit_opt_a.a_1                        :     0.000543s :  7.97%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.25%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.06%
    jit_opt_a.replace_old_param          :     0.000015s :  0.22%
    jit_opt_a.inline_without_move        :     0.000010s :  0.15%
    jit_opt_a.renormalize                :     0.000413s :  6.06%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.24%
    jit_opt_a.cse                        :     0.000027s :  0.39%
    jit_opt_a.replace_applicator         :     0.000015s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.11%
    rewriter_after_opt_a                 :     0.000024s :  0.35%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000455s :  6.68%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.21%
    cconv                                :     0.000019s :  0.27%
    loop_unroll                          :     0.000400s :  5.87%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.29%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000015s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.08%
    remove_dup_value                     :     0.000009s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000036s :  0.52%
    cse_after_recomputation.cse          :     0.000008s :  0.12%
    auto_monad_reorder                   :     0.000011s :  0.16%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.07%
    opt_after_jit_grad                   :     0.000435s :  6.38%
    symbol_engine_optimizer.build        :     0.000003s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000024s :  0.36%


Time group info:
------[substitution.]   0.000148    23
 0.86% :     0.000001s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.75% :     0.000004s :      3: substitution.graph_param_transform
83.29% :     0.000123s :      4: substitution.inline
 1.70% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.54% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000003s :      2: substitution.replace_old_param
 6.15% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003475     2
80.58% :     0.002800s :      1: type_inference.infer
19.42% :     0.000675s :      1: type_inference.specialize
------[replace.]   0.000057     6
66.25% :     0.000038s :      4: replace.inline
33.75% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000129     6
93.81% :     0.000121s :      4: match.inline
 6.19% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000107   691
 1.36% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000001s :     10: predicate.addn_zero_filter
 1.82% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000001s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.48% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.26% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.41% :     0.000000s :      3: predicate.graph_param_transform
 5.38% :     0.000006s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000001s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.24% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.17% :     0.000001s :     10: predicate.minmaximum_grad
 1.47% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000002s :     15: predicate.partial_eliminate
 1.27% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.85% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000001s :     10: predicate.reshape_eliminate
 1.33% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.25% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.69% :     0.000008s :     53: predicate.switch_simplify
 1.30% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.49% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.80% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000450    12
48.01% :     0.000216s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.51% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.48% :     0.000209s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011329    72
 0.34% :     0.000039s :      1: add_recomputation
 0.44% :     0.000050s :      1: auto_monad
 0.12% :     0.000014s :      1: auto_monad_reorder
 2.14% :     0.000242s :      1: bootstrap
 0.19% :     0.000021s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.18% :     0.000021s :      1: cse_after_recomputation
 0.05% :     0.000006s :      1: environ_conv
 0.17% :     0.000019s :      1: event_method
 0.04% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
38.09% :     0.004315s :      1: jit_opt_a
 1.16% :     0.000132s :      1: jit_opt_after_cconv
 0.44% :     0.000050s :      1: jit_opt_b
 3.59% :     0.000407s :      1: loop_unroll
 4.08% :     0.000463s :      1: mutable_eliminate
 6.46% :     0.000732s :     26: opt.transform.jit_opt_a
 0.38% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000024s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000029s :      4: opt.transform.symbol_engine_opt
 3.90% :     0.000442s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.04% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000027s :      1: py_interpret_to_execute
 0.09% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000012s :      1: remove_dup_value
 1.72% :     0.000194s :      1: renormalize.infer
 1.88% :     0.000213s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000027s :      1: rewriter_after_opt_a
 0.58% :     0.000066s :      1: rewriter_before_opt_a
 0.63% :     0.000071s :      1: symbol_engine_optimizer
31.56% :     0.003576s :      1: type_inference


TotalTime = 0.0217578, [33]
[bootstrap]: 0.00021042
[type_inference]: 0.0104795
[event_method]: 5.042e-05
[auto_monad]: 9.316e-05
[graph_reusing]: 6.69999e-06
[pre_auto_parallel]: 2.65997e-06
[py_interpret_to_execute]: 2.667e-05
[rewriter_before_opt_a]: 0.00012084
[expand_dump_flag]: 3.21001e-06
[jit_opt_a]: 0.00859846, [2]
    [Cycle 1]: 0.00485494, [27]
        [switch_simplify]: 6.473e-05
        [loop_unroll]: 3.762e-05
        [a_1]: 0.00086171
        [with_stream_mark]: 2.523e-05
        [recompute_prepare]: 1.787e-05
        [updatestate_depend_eliminate]: 7.85e-06
        [updatestate_assign_eliminate]: 6.64001e-06
        [updatestate_loads_eliminate]: 6.37001e-06
        [parameter_eliminate]: 2.83e-06
        [specialize_transform]: 1.284e-05
        [updatestate_useless_node_eliminater]: 1.14e-05
        [accelerated_algorithm]: 1.249e-05
        [meta_shard_fg_expand]: 3.2e-06
        [get_grad_eliminate_]: 1.211e-05
        [merge_forward]: 8.17e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.656e-05
        [j_node_and_user_rematch]: 2.286e-05
        [meta_fg_expand]: 0.00145846
        [replace_old_param]: 5.057e-05
        [inline_without_move]: 5.118e-05
        [renormalize]: 0.00189522
        [add_forward_monad_depend]: 5.40999e-06
        [auto_monad_grad]: 2.85002e-06
        [auto_monad_eliminator]: 1.295e-05
        [cse]: 2.49e-05
        [replace_applicator]: 1.253e-05
    [Cycle 2]: 0.0002894, [27]
        [switch_simplify]: 4.47e-06
        [loop_unroll]: 3.9e-06
        [a_1]: 4.168e-05
        [with_stream_mark]: 9.34e-06
        [recompute_prepare]: 3.49001e-06
        [updatestate_depend_eliminate]: 2.37999e-06
        [updatestate_assign_eliminate]: 1.67999e-06
        [updatestate_loads_eliminate]: 1.74e-06
        [parameter_eliminate]: 1.55001e-06
        [specialize_transform]: 3.58999e-06
        [updatestate_useless_node_eliminater]: 3.45998e-06
        [accelerated_algorithm]: 3.35e-06
        [meta_shard_fg_expand]: 1.67999e-06
        [get_grad_eliminate_]: 3.21001e-06
        [merge_forward]: 2.81e-06
        [cell_reuse_recompute_pass]: 2.41998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.173e-05
        [j_node_and_user_rematch]: 5.66998e-06
        [meta_fg_expand]: 3.808e-05
        [replace_old_param]: 4.26001e-06
        [inline_without_move]: 3.28998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.49998e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 4.60001e-06
        [cse]: 1.093e-05
        [replace_applicator]: 3.69002e-06
[py_interpret_to_execute_after_opt_a]: 1.081e-05
[rewriter_after_opt_a]: 3.894e-05
[convert_after_rewriter]: 6.01998e-06
[order_py_execute_after_rewriter]: 4.75001e-06
[mutable_eliminate]: 0.0006275
[jit_opt_b]: 4.243e-05, [1]
    [Cycle 1]: 3.481e-05, [2]
        [frontend_op_eliminate]: 1.143e-05
        [inline_after_opt_a]: 1.145e-05
[cconv]: 2.209e-05
[loop_unroll]: 0.00044548
[jit_opt_after_cconv]: 0.00013996, [1]
    [Cycle 1]: 0.0001326, [11]
        [c_1]: 1.325e-05
        [parameter_eliminate]: 3.11001e-06
        [updatestate_depend_eliminate]: 5.57001e-06
        [updatestate_assign_eliminate]: 2.15002e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [cse]: 1.787e-05
        [call_graph_tuple_transform]: 1.402e-05
        [tuple_list_get_item_eliminator]: 3.70998e-06
        [none_parameter_eliminate]: 1.55999e-06
        [renormalize]: 8.39995e-07
        [switch_simplify]: 3.61999e-06
[remove_dup_value]: 1.494e-05
[partial_unused_args_eliminate]: 2.20002e-06
[environ_conv]: 4.79e-06
[add_recomputation]: 3.381e-05
[cse_after_recomputation]: 1.944e-05, [1]
    [Cycle 1]: 1.351e-05, [1]
        [cse]: 8.08999e-06
[auto_monad_reorder]: 1.362e-05
[get_jit_bprop_graph]: 1.69e-06
[rewriter_after_jit_bprop_graph]: 4.60999e-06
[opt_after_jit_grad]: 0.00044297
[symbol_engine_optimizer]: 6.283e-05, [1]
    [Cycle 1]: 5.655e-05, [6]
        [build]: 2.96999e-06
        [elim_shapecalc]: 5.67999e-06
        [elim_not_effective]: 9.42001e-06
        [opt_reshape]: 4.50999e-06
        [fold_const_symbol]: 6.12999e-06
        [renormalize]: 5.19998e-07
[validate]: 2.783e-05
[backend_pass]: 8.30012e-07
[task_emit]: 2.433e-05
[execute]: 1.35001e-06
Sums
    bootstrap                            :     0.000210s :  1.19%
    type_inference                       :     0.010480s : 59.27%
    event_method                         :     0.000050s :  0.29%
    auto_monad                           :     0.000093s :  0.53%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000027s :  0.15%
    rewriter_before_opt_a                :     0.000121s :  0.68%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000069s :  0.39%
    jit_opt_a.loop_unroll                :     0.000042s :  0.23%
    jit_opt_a.a_1                        :     0.000903s :  5.11%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000016s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.09%
    jit_opt_a.merge_forward              :     0.000011s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001497s :  8.46%
    jit_opt_a.replace_old_param          :     0.000055s :  0.31%
    jit_opt_a.inline_without_move        :     0.000054s :  0.31%
    jit_opt_a.renormalize                :     0.001895s : 10.72%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.10%
    jit_opt_a.cse                        :     0.000036s :  0.20%
    jit_opt_a.replace_applicator         :     0.000016s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000039s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000627s :  3.55%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.06%
    cconv                                :     0.000022s :  0.12%
    loop_unroll                          :     0.000445s :  2.52%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000034s :  0.19%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000443s :  2.51%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000028s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000024s :  0.14%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000259    50
 0.60% :     0.000002s :      1: substitution.elim_not_effective
 0.42% :     0.000001s :      1: substitution.fold_const_symbol
 1.52% :     0.000004s :      1: substitution.graph_param_transform
68.02% :     0.000176s :      6: substitution.inline
 6.36% :     0.000016s :      2: substitution.inline_without_move
 2.49% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.32% :     0.000003s :      2: substitution.minmaximum_grad
 1.68% :     0.000004s :      3: substitution.partial_eliminate
 2.87% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.33% :     0.000006s :      2: substitution.replace_applicator
 1.38% :     0.000004s :      5: substitution.replace_old_param
 0.89% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.67% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.71% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 5.74% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010409     2
89.56% :     0.009322s :      1: type_inference.infer
10.44% :     0.001087s :      1: type_inference.specialize
------[replace.]   0.000064     8
77.70% :     0.000050s :      6: replace.inline
22.30% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000183     8
94.42% :     0.000173s :      6: match.inline
 5.58% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000172  1109
 1.67% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.67% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.30% :     0.000002s :     18: predicate.addn_check_dump
 1.54% :     0.000003s :     18: predicate.addn_zero_filter
 2.30% :     0.000004s :     18: predicate.arithmetic_simplify
 1.48% :     0.000003s :     18: predicate.cast_eliminate
 0.20% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.37% :     0.000002s :     18: predicate.compare_switch_simplify
 1.72% :     0.000003s :     18: predicate.depend_value_elim
 1.23% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.60% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.27% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.52% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.21% :     0.000000s :      1: predicate.elim_not_effective
 0.30% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.24% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     18: predicate.environ_get_eliminate
 1.20% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.07% :     0.000000s :      1: predicate.fold_const_symbol
 1.14% :     0.000002s :     12: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 4.45% :     0.000008s :     28: predicate.inline
 3.18% :     0.000005s :     33: predicate.inline_without_move
 0.51% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.60% :     0.000003s :     12: predicate.less_batch_normalization
 1.55% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.66% :     0.000003s :     21: predicate.load_eliminater
 0.64% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.54% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.86% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.29% :     0.000002s :     18: predicate.merge_addn
 1.26% :     0.000002s :     18: predicate.minmaximum_grad
 0.59% :     0.000001s :      1: predicate.mutable_eliminate
 0.24% :     0.000000s :      1: predicate.opt_reshape
 2.00% :     0.000003s :     21: predicate.partial_eliminate
 1.28% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.68% :     0.000003s :     18: predicate.reduce_eliminate
 1.55% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.72% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.53% :     0.000003s :     22: predicate.replace_applicator
 1.74% :     0.000003s :     33: predicate.replace_old_param
 0.11% :     0.000000s :      1: predicate.reset_defer_inline
 1.51% :     0.000003s :     18: predicate.reshape_eliminate
 1.69% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.53% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.60% :     0.000003s :     18: predicate.same_eliminate
 0.60% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.46% :     0.000001s :      2: predicate.special_op_eliminate
 1.16% :     0.000002s :     12: predicate.specialize_transform
 1.82% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.63% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.15% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.46% :     0.000004s :     26: predicate.switch_defer_inline
 2.26% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.84% :     0.000012s :     71: predicate.switch_simplify
 1.42% :     0.000002s :     18: predicate.tile_eliminate
 1.32% :     0.000002s :     18: predicate.transpose_eliminate
 1.59% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.48% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.81% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.88% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.69% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.96% :     0.000003s :     18: predicate.value_based_eliminate
 0.13% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.31% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001142    22
73.82% :     0.000843s :     13: func_graph_cloner_run.FuncGraphClonerGraph
26.18% :     0.000299s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024975    76
 0.15% :     0.000037s :      1: add_recomputation
 0.39% :     0.000097s :      1: auto_monad
 0.06% :     0.000016s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.89% :     0.000222s :      1: bootstrap
 0.10% :     0.000025s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.22% :     0.000055s :      1: event_method
 0.01% :     0.000004s :      1: execute
 0.03% :     0.000007s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
34.44% :     0.008602s :      1: jit_opt_a
 0.57% :     0.000143s :      1: jit_opt_after_cconv
 0.18% :     0.000045s :      1: jit_opt_b
 1.81% :     0.000453s :      1: loop_unroll
 2.54% :     0.000635s :      1: mutable_eliminate
 5.05% :     0.001260s :     26: opt.transform.jit_opt_a
 0.13% :     0.000031s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000022s :      4: opt.transform.symbol_engine_opt
 1.80% :     0.000450s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000029s :      1: py_interpret_to_execute
 0.05% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 5.05% :     0.001262s :      1: renormalize.infer
 2.50% :     0.000625s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000042s :      1: rewriter_after_opt_a
 0.50% :     0.000125s :      1: rewriter_before_opt_a
 0.26% :     0.000066s :      1: symbol_engine_optimizer
 0.12% :     0.000029s :      1: task_emit
42.01% :     0.010492s :      1: type_inference
 0.18% :     0.000044s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.231.08 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c375700> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.237.57 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0172662, [33]
[bootstrap]: 0.00027165
[type_inference]: 0.00405566
[event_method]: 1.746e-05
[auto_monad]: 6.258e-05
[graph_reusing]: 5.60001e-06
[pre_auto_parallel]: 2.16998e-06
[py_interpret_to_execute]: 2.984e-05
[rewriter_before_opt_a]: 6.757e-05
[expand_dump_flag]: 3.37002e-06
[jit_opt_a]: 0.00490135, [2]
    [Cycle 1]: 0.00141325, [27]
        [switch_simplify]: 4.989e-05
        [loop_unroll]: 2.799e-05
        [a_1]: 0.00049477
        [with_stream_mark]: 1.686e-05
        [recompute_prepare]: 6.89999e-06
        [updatestate_depend_eliminate]: 3.55e-06
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 2.82002e-06
        [parameter_eliminate]: 1.76e-06
        [specialize_transform]: 5.63002e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 4.94998e-06
        [meta_shard_fg_expand]: 2.37999e-06
        [get_grad_eliminate_]: 5.24e-06
        [merge_forward]: 3.51999e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.374e-05
        [j_node_and_user_rematch]: 8.38999e-06
        [meta_fg_expand]: 2.01e-06
        [replace_old_param]: 8.13001e-06
        [inline_without_move]: 5.27999e-06
        [renormalize]: 0.00051672
        [add_forward_monad_depend]: 5.47999e-06
        [auto_monad_grad]: 2.01e-06
        [auto_monad_eliminator]: 1.333e-05
        [cse]: 2.655e-05
        [replace_applicator]: 1.107e-05
    [Cycle 2]: 0.00030959, [27]
        [switch_simplify]: 5.57999e-06
        [loop_unroll]: 5.35999e-06
        [a_1]: 8.676e-05
        [with_stream_mark]: 8.32e-06
        [recompute_prepare]: 4.75999e-06
        [updatestate_depend_eliminate]: 2.48002e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 1.81003e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 4.70999e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 4.90001e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 4.48999e-06
        [merge_forward]: 2.16998e-06
        [cell_reuse_recompute_pass]: 1.55999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.218e-05
        [j_node_and_user_rematch]: 7.72998e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 7.08998e-06
        [inline_without_move]: 4.50001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 7.79983e-07
        [auto_monad_eliminator]: 5.17e-06
        [cse]: 1.031e-05
        [replace_applicator]: 4.60001e-06
[py_interpret_to_execute_after_opt_a]: 7.62002e-06
[rewriter_after_opt_a]: 3.098e-05
[convert_after_rewriter]: 6.93e-06
[order_py_execute_after_rewriter]: 4.53999e-06
[mutable_eliminate]: 0.00050171
[jit_opt_b]: 4.762e-05, [1]
    [Cycle 1]: 4.106e-05, [2]
        [frontend_op_eliminate]: 1.564e-05
        [inline_after_opt_a]: 1.425e-05
[cconv]: 2.315e-05
[loop_unroll]: 0.00041086
[jit_opt_after_cconv]: 0.00013597, [1]
    [Cycle 1]: 0.00012961, [11]
        [c_1]: 2.014e-05
        [parameter_eliminate]: 2.34999e-06
        [updatestate_depend_eliminate]: 5.54e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 1.732e-05
        [call_graph_tuple_transform]: 1.817e-05
        [tuple_list_get_item_eliminator]: 5.18002e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 5.69999e-06
[remove_dup_value]: 1.283e-05
[partial_unused_args_eliminate]: 2.11e-06
[environ_conv]: 5.02999e-06
[add_recomputation]: 4.623e-05
[cse_after_recomputation]: 1.946e-05, [1]
    [Cycle 1]: 1.435e-05, [1]
        [cse]: 8.79e-06
[auto_monad_reorder]: 1.47e-05
[get_jit_bprop_graph]: 1.50001e-06
[rewriter_after_jit_bprop_graph]: 4.50999e-06
[opt_after_jit_grad]: 0.0004876
[symbol_engine_optimizer]: 7.073e-05, [1]
    [Cycle 1]: 6.426e-05, [6]
        [build]: 3.36001e-06
        [elim_shapecalc]: 7.25e-06
        [elim_not_effective]: 1.221e-05
        [opt_reshape]: 5.78002e-06
        [fold_const_symbol]: 8.25999e-06
        [renormalize]: 4.30009e-07
[validate]: 3.265e-05
[backend_pass]: 8.60018e-07
[task_emit]: 0.00578638
[execute]: 7.75e-06
Sums
    bootstrap                            :     0.000272s :  2.01%
    type_inference                       :     0.004056s : 30.03%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000063s :  0.46%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.22%
    rewriter_before_opt_a                :     0.000068s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000055s :  0.41%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000582s :  4.31%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000517s :  3.83%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.14%
    jit_opt_a.cse                        :     0.000037s :  0.27%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000502s :  3.71%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000411s :  3.04%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000046s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000488s :  3.61%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005786s : 42.84%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000181    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000001s :      2: substitution.fold_const_symbol
 2.93% :     0.000005s :      3: substitution.graph_param_transform
82.65% :     0.000149s :      4: substitution.inline
 1.62% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.45% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.98% :     0.000004s :      2: substitution.replace_old_param
 6.60% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003997     2
81.21% :     0.003246s :      1: type_inference.infer
18.79% :     0.000751s :      1: type_inference.specialize
------[replace.]   0.000063     6
66.77% :     0.000042s :      4: replace.inline
33.23% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000158     6
93.11% :     0.000147s :      4: match.inline
 6.89% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.32% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.46% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000001s :     10: predicate.addn_zero_filter
 2.00% :     0.000002s :     10: predicate.arithmetic_simplify
 1.42% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000001s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.13% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.08% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 5.52% :     0.000006s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.07% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.48% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.67% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.48% :     0.000002s :      3: predicate.mutable_eliminate
 0.57% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.36% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.82% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.02% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.88% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.66% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000489    12
47.79% :     0.000234s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.23% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.97% :     0.000230s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018664    76
 0.26% :     0.000049s :      1: add_recomputation
 0.35% :     0.000066s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.55% :     0.000290s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.27% :     0.004904s :      1: jit_opt_a
 0.74% :     0.000139s :      1: jit_opt_after_cconv
 0.27% :     0.000050s :      1: jit_opt_b
 2.25% :     0.000420s :      1: loop_unroll
 2.74% :     0.000512s :      1: mutable_eliminate
 4.16% :     0.000777s :     26: opt.transform.jit_opt_a
 0.25% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.66% :     0.000496s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000032s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.37% :     0.000256s :      1: renormalize.infer
 1.36% :     0.000253s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.38% :     0.000070s :      1: rewriter_before_opt_a
 0.39% :     0.000073s :      1: symbol_engine_optimizer
31.06% :     0.005797s :      1: task_emit
21.81% :     0.004070s :      1: type_inference
 0.28% :     0.000052s :      1: validate


TotalTime = 0.010099, [30]
[bootstrap]: 0.00022391
[type_inference]: 0.00342599
[event_method]: 1.49e-05
[auto_monad]: 4.831e-05
[graph_reusing]: 4.42998e-06
[pre_auto_parallel]: 2.31e-06
[py_interpret_to_execute]: 2.45e-05
[rewriter_before_opt_a]: 6.19e-05
[expand_dump_flag]: 2.22001e-06
[jit_opt_a]: 0.0044116, [2]
    [Cycle 1]: 0.00128483, [27]
        [switch_simplify]: 4.435e-05
        [loop_unroll]: 2.81e-05
        [a_1]: 0.000478
        [with_stream_mark]: 1.37e-05
        [recompute_prepare]: 6.87002e-06
        [updatestate_depend_eliminate]: 3.33e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [parameter_eliminate]: 1.17999e-06
        [specialize_transform]: 6.26998e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 5.66e-06
        [meta_shard_fg_expand]: 1.59e-06
        [get_grad_eliminate_]: 4.87e-06
        [merge_forward]: 2.71999e-06
        [cell_reuse_recompute_pass]: 1.01002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.248e-05
        [j_node_and_user_rematch]: 8.89e-06
        [meta_fg_expand]: 2.03997e-06
        [replace_old_param]: 8.43999e-06
        [inline_without_move]: 5.19998e-06
        [renormalize]: 0.00043455
        [add_forward_monad_depend]: 5.24e-06
        [auto_monad_grad]: 1.84e-06
        [auto_monad_eliminator]: 1.085e-05
        [cse]: 1.974e-05
        [replace_applicator]: 1.032e-05
    [Cycle 2]: 0.00030797, [27]
        [switch_simplify]: 5.40001e-06
        [loop_unroll]: 4.79e-06
        [a_1]: 8.81e-05
        [with_stream_mark]: 7.96001e-06
        [recompute_prepare]: 4.93001e-06
        [updatestate_depend_eliminate]: 2.36e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.83002e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 4.69998e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 5.01002e-06
        [meta_shard_fg_expand]: 9.80013e-07
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.36998e-06
        [cell_reuse_recompute_pass]: 1.21997e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.162e-05
        [j_node_and_user_rematch]: 7.57002e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 7.13998e-06
        [inline_without_move]: 4.46002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 5.22e-06
        [cse]: 9.54e-06
        [replace_applicator]: 4.91002e-06
[py_interpret_to_execute_after_opt_a]: 7.75998e-06
[rewriter_after_opt_a]: 2.56e-05
[convert_after_rewriter]: 6.59999e-06
[order_py_execute_after_rewriter]: 5.86e-06
[mutable_eliminate]: 0.00046577
[jit_opt_b]: 4.697e-05, [1]
    [Cycle 1]: 4.062e-05, [2]
        [frontend_op_eliminate]: 1.502e-05
        [inline_after_opt_a]: 1.474e-05
[cconv]: 2.134e-05
[loop_unroll]: 0.00040188
[jit_opt_after_cconv]: 0.00012868, [1]
    [Cycle 1]: 0.00012269, [11]
        [c_1]: 1.923e-05
        [parameter_eliminate]: 2.30002e-06
        [updatestate_depend_eliminate]: 4.90001e-06
        [updatestate_assign_eliminate]: 2.23002e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 1.483e-05
        [call_graph_tuple_transform]: 1.664e-05
        [tuple_list_get_item_eliminator]: 5.31002e-06
        [none_parameter_eliminate]: 9.89996e-07
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.23002e-06
[remove_dup_value]: 1.067e-05
[partial_unused_args_eliminate]: 1.95001e-06
[environ_conv]: 4.35999e-06
[add_recomputation]: 3.737e-05
[cse_after_recomputation]: 1.938e-05, [1]
    [Cycle 1]: 1.44e-05, [1]
        [cse]: 8.74998e-06
[auto_monad_reorder]: 1.213e-05
[get_jit_bprop_graph]: 1.39e-06
[rewriter_after_jit_bprop_graph]: 4.70001e-06
[opt_after_jit_grad]: 0.00043858
[symbol_engine_optimizer]: 6.837e-05, [1]
    [Cycle 1]: 6.234e-05, [6]
        [build]: 3.06001e-06
        [elim_shapecalc]: 7.10002e-06
        [elim_not_effective]: 1.142e-05
        [opt_reshape]: 5.67001e-06
        [fold_const_symbol]: 8.18999e-06
        [renormalize]: 4.39992e-07
[validate]: 2.676e-05
Sums
    bootstrap                            :     0.000224s :  3.32%
    type_inference                       :     0.003426s : 50.74%
    event_method                         :     0.000015s :  0.22%
    auto_monad                           :     0.000048s :  0.72%
    graph_reusing                        :     0.000004s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000025s :  0.36%
    rewriter_before_opt_a                :     0.000062s :  0.92%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000050s :  0.74%
    jit_opt_a.loop_unroll                :     0.000033s :  0.49%
    jit_opt_a.a_1                        :     0.000566s :  8.38%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.15%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.16%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.36%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.24%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.23%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000435s :  6.44%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.24%
    jit_opt_a.cse                        :     0.000029s :  0.43%
    jit_opt_a.replace_applicator         :     0.000015s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000026s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.10%
    order_py_execute_after_rewriter      :     0.000006s :  0.09%
    mutable_eliminate                    :     0.000466s :  6.90%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.22%
    cconv                                :     0.000021s :  0.32%
    loop_unroll                          :     0.000402s :  5.95%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000015s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.25%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.08%
    remove_dup_value                     :     0.000011s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000037s :  0.55%
    cse_after_recomputation.cse          :     0.000009s :  0.13%
    auto_monad_reorder                   :     0.000012s :  0.18%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000439s :  6.49%
    symbol_engine_optimizer.build        :     0.000003s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000027s :  0.40%


Time group info:
------[substitution.]   0.000158    23
 0.89% :     0.000001s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.75% :     0.000004s :      3: substitution.graph_param_transform
82.57% :     0.000130s :      4: substitution.inline
 1.83% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.53% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.21% :     0.000003s :      2: substitution.replace_old_param
 6.49% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003382     2
79.66% :     0.002694s :      1: type_inference.infer
20.34% :     0.000688s :      1: type_inference.specialize
------[replace.]   0.000059     6
67.42% :     0.000040s :      4: replace.inline
32.58% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000137     6
93.32% :     0.000128s :      4: match.inline
 6.68% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.12% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.26% :     0.000001s :     10: predicate.addn_zero_filter
 1.86% :     0.000002s :     10: predicate.arithmetic_simplify
 1.11% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000001s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.17% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.36% :     0.000000s :      3: predicate.graph_param_transform
 5.18% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.10% :     0.000001s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.72% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.86% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.46% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.00% :     0.000002s :     15: predicate.partial_eliminate
 1.09% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.47% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.24% :     0.000001s :     10: predicate.reshape_eliminate
 8.91% :     0.000011s :     10: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.11% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.98% :     0.000008s :     53: predicate.switch_simplify
 1.10% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.26% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.29% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.30% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000437    12
47.75% :     0.000209s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.52% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.73% :     0.000200s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011357    72
 0.35% :     0.000040s :      1: add_recomputation
 0.45% :     0.000051s :      1: auto_monad
 0.13% :     0.000015s :      1: auto_monad_reorder
 2.06% :     0.000234s :      1: bootstrap
 0.21% :     0.000024s :      1: cconv
 0.08% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000021s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000018s :      1: event_method
 0.04% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
38.87% :     0.004414s :      1: jit_opt_a
 1.16% :     0.000131s :      1: jit_opt_after_cconv
 0.44% :     0.000049s :      1: jit_opt_b
 3.60% :     0.000409s :      1: loop_unroll
 4.17% :     0.000474s :      1: mutable_eliminate
 6.67% :     0.000758s :     26: opt.transform.jit_opt_a
 0.38% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000024s :      4: opt.transform.jit_opt_b
 0.10% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000029s :      4: opt.transform.symbol_engine_opt
 3.92% :     0.000446s :      1: opt_after_jit_grad
 0.07% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.04% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000027s :      1: py_interpret_to_execute
 0.09% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000013s :      1: remove_dup_value
 1.95% :     0.000221s :      1: renormalize.infer
 1.83% :     0.000207s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000028s :      1: rewriter_after_opt_a
 0.57% :     0.000065s :      1: rewriter_before_opt_a
 0.62% :     0.000071s :      1: symbol_engine_optimizer
30.27% :     0.003437s :      1: type_inference


TotalTime = 0.0226118, [33]
[bootstrap]: 0.00020909
[type_inference]: 0.0103071
[event_method]: 5.364e-05
[auto_monad]: 9.789e-05
[graph_reusing]: 6.85002e-06
[pre_auto_parallel]: 2.48998e-06
[py_interpret_to_execute]: 2.715e-05
[rewriter_before_opt_a]: 9.222e-05
[expand_dump_flag]: 4.04997e-06
[jit_opt_a]: 0.00927968, [2]
    [Cycle 1]: 0.00506465, [27]
        [switch_simplify]: 6.65e-05
        [loop_unroll]: 3.783e-05
        [a_1]: 0.00091194
        [with_stream_mark]: 3.118e-05
        [recompute_prepare]: 2.231e-05
        [updatestate_depend_eliminate]: 8.95999e-06
        [updatestate_assign_eliminate]: 6.64999e-06
        [updatestate_loads_eliminate]: 6.48998e-06
        [parameter_eliminate]: 2.91e-06
        [specialize_transform]: 1.454e-05
        [updatestate_useless_node_eliminater]: 1.332e-05
        [accelerated_algorithm]: 1.226e-05
        [meta_shard_fg_expand]: 3.65e-06
        [get_grad_eliminate_]: 1.18e-05
        [merge_forward]: 7.88001e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.706e-05
        [j_node_and_user_rematch]: 2.32e-05
        [meta_fg_expand]: 0.00170272
        [replace_old_param]: 6.086e-05
        [inline_without_move]: 5.849e-05
        [renormalize]: 0.00174954
        [add_forward_monad_depend]: 6.61e-06
        [auto_monad_grad]: 2.79999e-06
        [auto_monad_eliminator]: 1.268e-05
        [cse]: 2.803e-05
        [replace_applicator]: 1.385e-05
    [Cycle 2]: 0.00030223, [27]
        [switch_simplify]: 4.48001e-06
        [loop_unroll]: 4.05e-06
        [a_1]: 4.412e-05
        [with_stream_mark]: 1.172e-05
        [recompute_prepare]: 3.66001e-06
        [updatestate_depend_eliminate]: 2.96999e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.91003e-06
        [specialize_transform]: 3.58999e-06
        [updatestate_useless_node_eliminater]: 3.44001e-06
        [accelerated_algorithm]: 3.58999e-06
        [meta_shard_fg_expand]: 1.72001e-06
        [get_grad_eliminate_]: 3.28e-06
        [merge_forward]: 3.28e-06
        [cell_reuse_recompute_pass]: 3.86999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.295e-05
        [j_node_and_user_rematch]: 5.76e-06
        [meta_fg_expand]: 3.907e-05
        [replace_old_param]: 4.62e-06
        [inline_without_move]: 3.41999e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.41998e-06
        [auto_monad_grad]: 1.55999e-06
        [auto_monad_eliminator]: 4.80001e-06
        [cse]: 1.06e-05
        [replace_applicator]: 3.8e-06
[py_interpret_to_execute_after_opt_a]: 1.232e-05
[rewriter_after_opt_a]: 3.6e-05
[convert_after_rewriter]: 6.33e-06
[order_py_execute_after_rewriter]: 4.3e-06
[mutable_eliminate]: 0.00074394
[jit_opt_b]: 5.286e-05, [1]
    [Cycle 1]: 4.482e-05, [2]
        [frontend_op_eliminate]: 1.007e-05
        [inline_after_opt_a]: 1.203e-05
[cconv]: 2.742e-05
[loop_unroll]: 0.00058834
[jit_opt_after_cconv]: 0.00013414, [1]
    [Cycle 1]: 0.00012663, [11]
        [c_1]: 1.339e-05
        [parameter_eliminate]: 3.30998e-06
        [updatestate_depend_eliminate]: 5.78997e-06
        [updatestate_assign_eliminate]: 2.18998e-06
        [updatestate_loads_eliminate]: 1.70001e-06
        [cse]: 2.259e-05
        [call_graph_tuple_transform]: 1.519e-05
        [tuple_list_get_item_eliminator]: 3.9e-06
        [none_parameter_eliminate]: 1.67001e-06
        [renormalize]: 7.30011e-07
        [switch_simplify]: 3.85e-06
[remove_dup_value]: 1.435e-05
[partial_unused_args_eliminate]: 2.58e-06
[environ_conv]: 5.70001e-06
[add_recomputation]: 3.64e-05
[cse_after_recomputation]: 1.987e-05, [1]
    [Cycle 1]: 1.383e-05, [1]
        [cse]: 7.77e-06
[auto_monad_reorder]: 1.374e-05
[get_jit_bprop_graph]: 2.34999e-06
[rewriter_after_jit_bprop_graph]: 5.92001e-06
[opt_after_jit_grad]: 0.00051668
[symbol_engine_optimizer]: 6.585e-05, [1]
    [Cycle 1]: 5.911e-05, [6]
        [build]: 3.74002e-06
        [elim_shapecalc]: 6.12001e-06
        [elim_not_effective]: 9.79e-06
        [opt_reshape]: 3.83001e-06
        [fold_const_symbol]: 6.26e-06
        [renormalize]: 5.10016e-07
[validate]: 3.152e-05
[backend_pass]: 9.5999e-07
[task_emit]: 2.236e-05
[execute]: 1.27e-06
Sums
    bootstrap                            :     0.000209s :  1.16%
    type_inference                       :     0.010307s : 57.13%
    event_method                         :     0.000054s :  0.30%
    auto_monad                           :     0.000098s :  0.54%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000027s :  0.15%
    rewriter_before_opt_a                :     0.000092s :  0.51%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000071s :  0.39%
    jit_opt_a.loop_unroll                :     0.000042s :  0.23%
    jit_opt_a.a_1                        :     0.000956s :  5.30%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000026s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.03%
    jit_opt_a.specialize_transform       :     0.000018s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000017s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.08%
    jit_opt_a.merge_forward              :     0.000011s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001742s :  9.65%
    jit_opt_a.replace_old_param          :     0.000065s :  0.36%
    jit_opt_a.inline_without_move        :     0.000062s :  0.34%
    jit_opt_a.renormalize                :     0.001750s :  9.70%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.10%
    jit_opt_a.cse                        :     0.000039s :  0.21%
    jit_opt_a.replace_applicator         :     0.000018s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000744s :  4.12%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.07%
    cconv                                :     0.000027s :  0.15%
    loop_unroll                          :     0.000588s :  3.26%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000036s :  0.20%
    cse_after_recomputation.cse          :     0.000008s :  0.04%
    auto_monad_reorder                   :     0.000014s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000517s :  2.86%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.17%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.000022s :  0.12%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000283    50
 0.54% :     0.000002s :      1: substitution.elim_not_effective
 0.39% :     0.000001s :      1: substitution.fold_const_symbol
 1.50% :     0.000004s :      1: substitution.graph_param_transform
65.81% :     0.000187s :      6: substitution.inline
 7.65% :     0.000022s :      2: substitution.inline_without_move
 2.48% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.06% :     0.000003s :      2: substitution.minmaximum_grad
 1.94% :     0.000006s :      3: substitution.partial_eliminate
 2.69% :     0.000008s :      9: substitution.remove_not_recompute_node
 2.61% :     0.000007s :      2: substitution.replace_applicator
 1.77% :     0.000005s :      5: substitution.replace_old_param
 1.23% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.89% :     0.000008s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.71% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.73% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010234     2
87.70% :     0.008975s :      1: type_inference.infer
12.30% :     0.001259s :      1: type_inference.specialize
------[replace.]   0.000072     8
78.41% :     0.000057s :      6: replace.inline
21.59% :     0.000016s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000194     8
94.40% :     0.000183s :      6: match.inline
 5.60% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000190  1109
 1.64% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.61% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.15% :     0.000002s :     18: predicate.addn_check_dump
 1.81% :     0.000003s :     18: predicate.addn_zero_filter
 2.33% :     0.000004s :     18: predicate.arithmetic_simplify
 1.40% :     0.000003s :     18: predicate.cast_eliminate
 0.16% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     18: predicate.compare_switch_simplify
 2.41% :     0.000005s :     18: predicate.depend_value_elim
 1.15% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.40% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.44% :     0.000003s :     18: predicate.dict_set_item_eliminator
 0.47% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.18% :     0.000000s :      1: predicate.elim_not_effective
 0.23% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.16% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.81% :     0.000003s :     18: predicate.environ_get_depend_swap
 1.22% :     0.000002s :     18: predicate.environ_get_eliminate
 1.14% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.12% :     0.000000s :      1: predicate.fold_const_symbol
 0.95% :     0.000002s :     12: predicate.get_grad_eliminate
 0.19% :     0.000000s :      1: predicate.graph_param_transform
 4.49% :     0.000009s :     28: predicate.inline
 3.00% :     0.000006s :     33: predicate.inline_without_move
 0.55% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.43% :     0.000003s :     12: predicate.less_batch_normalization
 1.69% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.56% :     0.000003s :     21: predicate.load_eliminater
 1.28% :     0.000002s :      1: predicate.loop_unroll_after_grad
 3.33% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.62% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000002s :     18: predicate.merge_addn
 1.28% :     0.000002s :     18: predicate.minmaximum_grad
 0.92% :     0.000002s :      1: predicate.mutable_eliminate
 0.13% :     0.000000s :      1: predicate.opt_reshape
 2.14% :     0.000004s :     21: predicate.partial_eliminate
 1.40% :     0.000003s :     18: predicate.print_const_string_wrapper
 1.60% :     0.000003s :     18: predicate.reduce_eliminate
 2.05% :     0.000004s :     20: predicate.redundant_stop_gradient_eliminater
 0.72% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.35% :     0.000003s :     22: predicate.replace_applicator
 1.73% :     0.000003s :     33: predicate.replace_old_param
 0.10% :     0.000000s :      1: predicate.reset_defer_inline
 1.42% :     0.000003s :     18: predicate.reshape_eliminate
 1.55% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.50% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.67% :     0.000003s :     18: predicate.same_eliminate
 0.71% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.28% :     0.000001s :      2: predicate.special_op_eliminate
 1.17% :     0.000002s :     12: predicate.specialize_transform
 1.55% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.69% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.21% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.85% :     0.000005s :     26: predicate.switch_defer_inline
 2.34% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.46% :     0.000012s :     71: predicate.switch_simplify
 1.69% :     0.000003s :     18: predicate.tile_eliminate
 1.23% :     0.000002s :     18: predicate.transpose_eliminate
 1.49% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 3.18% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.54% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.88% :     0.000004s :     18: predicate.value_based_eliminate
 0.11% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.17% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001253    22
66.59% :     0.000834s :     13: func_graph_cloner_run.FuncGraphClonerGraph
33.41% :     0.000419s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025775    76
 0.15% :     0.000039s :      1: add_recomputation
 0.40% :     0.000102s :      1: auto_monad
 0.06% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.86% :     0.000222s :      1: bootstrap
 0.12% :     0.000030s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.22% :     0.000058s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.03% :     0.000008s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
36.02% :     0.009284s :      1: jit_opt_a
 0.53% :     0.000137s :      1: jit_opt_after_cconv
 0.22% :     0.000056s :      1: jit_opt_b
 2.32% :     0.000599s :      1: loop_unroll
 2.93% :     0.000754s :      1: mutable_eliminate
 5.20% :     0.001341s :     26: opt.transform.jit_opt_a
 0.13% :     0.000033s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000015s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000017s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000023s :      4: opt.transform.symbol_engine_opt
 2.04% :     0.000526s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 5.18% :     0.001336s :      1: renormalize.infer
 1.57% :     0.000405s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000039s :      1: rewriter_after_opt_a
 0.37% :     0.000096s :      1: rewriter_before_opt_a
 0.27% :     0.000069s :      1: symbol_engine_optimizer
 0.10% :     0.000026s :      1: task_emit
40.04% :     0.010321s :      1: type_inference
 0.19% :     0.000049s :      1: validate


TotalTime = 0.0360415, [33]
[bootstrap]: 0.00018881
[type_inference]: 0.010084
[event_method]: 4.971e-05
[auto_monad]: 8.421e-05
[graph_reusing]: 6.83e-06
[pre_auto_parallel]: 2.59001e-06
[py_interpret_to_execute]: 2.377e-05
[rewriter_before_opt_a]: 8.774e-05
[expand_dump_flag]: 2.74001e-06
[jit_opt_a]: 0.0124302, [3]
    [Cycle 1]: 0.00598856, [27]
        [switch_simplify]: 5.934e-05
        [loop_unroll]: 3.904e-05
        [a_1]: 0.00082372
        [with_stream_mark]: 2.244e-05
        [recompute_prepare]: 1.78e-05
        [updatestate_depend_eliminate]: 7.16001e-06
        [updatestate_assign_eliminate]: 5.74e-06
        [updatestate_loads_eliminate]: 5.25001e-06
        [parameter_eliminate]: 2.24001e-06
        [specialize_transform]: 1.312e-05
        [updatestate_useless_node_eliminater]: 1.171e-05
        [accelerated_algorithm]: 1.21e-05
        [meta_shard_fg_expand]: 3.13998e-06
        [get_grad_eliminate_]: 1.224e-05
        [merge_forward]: 7.6e-06
        [cell_reuse_recompute_pass]: 7.40023e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.324e-05
        [j_node_and_user_rematch]: 2.387e-05
        [meta_fg_expand]: 0.00149951
        [replace_old_param]: 5.673e-05
        [inline_without_move]: 5.525e-05
        [renormalize]: 0.00276259
        [add_forward_monad_depend]: 1.182e-05
        [auto_monad_grad]: 6.84001e-06
        [auto_monad_eliminator]: 5.659e-05
        [cse]: 0.00016795
        [replace_applicator]: 7.652e-05
    [Cycle 2]: 0.00253248, [27]
        [switch_simplify]: 3.855e-05
        [loop_unroll]: 3.653e-05
        [a_1]: 0.00125126
        [with_stream_mark]: 1.786e-05
        [recompute_prepare]: 1.068e-05
        [updatestate_depend_eliminate]: 6.12999e-06
        [updatestate_assign_eliminate]: 5.00999e-06
        [updatestate_loads_eliminate]: 4.4e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 9.12999e-06
        [updatestate_useless_node_eliminater]: 8.59e-06
        [accelerated_algorithm]: 8.25999e-06
        [meta_shard_fg_expand]: 3.76001e-06
        [get_grad_eliminate_]: 9.18002e-06
        [merge_forward]: 5.44998e-06
        [cell_reuse_recompute_pass]: 1.80001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.074e-05
        [j_node_and_user_rematch]: 1.415e-05
        [meta_fg_expand]: 5.689e-05
        [replace_old_param]: 1.33e-05
        [inline_without_move]: 8.63001e-06
        [renormalize]: 0.00074054
        [add_forward_monad_depend]: 5.65001e-06
        [auto_monad_grad]: 1.82001e-06
        [auto_monad_eliminator]: 1.641e-05
        [cse]: 5.414e-05
        [replace_applicator]: 1.747e-05
    [Cycle 3]: 0.00055086, [27]
        [switch_simplify]: 9.76e-06
        [loop_unroll]: 8.33999e-06
        [a_1]: 0.00020447
        [with_stream_mark]: 1.25e-05
        [recompute_prepare]: 9.49e-06
        [updatestate_depend_eliminate]: 5.27999e-06
        [updatestate_assign_eliminate]: 5.40999e-06
        [updatestate_loads_eliminate]: 4.04002e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 9.86e-06
        [updatestate_useless_node_eliminater]: 8.67e-06
        [accelerated_algorithm]: 9.79e-06
        [meta_shard_fg_expand]: 2.44001e-06
        [get_grad_eliminate_]: 8.08001e-06
        [merge_forward]: 4.92999e-06
        [cell_reuse_recompute_pass]: 2.46e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.047e-05
        [j_node_and_user_rematch]: 1.473e-05
        [meta_fg_expand]: 3.68e-06
        [replace_old_param]: 1.202e-05
        [inline_without_move]: 8.48001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.14e-06
        [auto_monad_grad]: 1.14003e-06
        [auto_monad_eliminator]: 1.369e-05
        [cse]: 3.116e-05
        [replace_applicator]: 9.10999e-06
[py_interpret_to_execute_after_opt_a]: 1.954e-05
[rewriter_after_opt_a]: 5.575e-05
[convert_after_rewriter]: 9.33002e-06
[order_py_execute_after_rewriter]: 6.51999e-06
[mutable_eliminate]: 0.00082195
[jit_opt_b]: 7.591e-05, [1]
    [Cycle 1]: 6.784e-05, [2]
        [frontend_op_eliminate]: 2.819e-05
        [inline_after_opt_a]: 2.712e-05
[cconv]: 3.126e-05
[loop_unroll]: 0.00051531
[jit_opt_after_cconv]: 0.0002301, [1]
    [Cycle 1]: 0.00022125, [11]
        [c_1]: 3.784e-05
        [parameter_eliminate]: 4.64998e-06
        [updatestate_depend_eliminate]: 1.09e-05
        [updatestate_assign_eliminate]: 6.14001e-06
        [updatestate_loads_eliminate]: 4.81002e-06
        [cse]: 4.738e-05
        [call_graph_tuple_transform]: 3.208e-05
        [tuple_list_get_item_eliminator]: 8.75999e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 5.90022e-07
        [switch_simplify]: 9.81998e-06
[remove_dup_value]: 2.593e-05
[partial_unused_args_eliminate]: 2.84001e-06
[environ_conv]: 8.89e-06
[add_recomputation]: 6.66e-05
[cse_after_recomputation]: 4.388e-05, [1]
    [Cycle 1]: 3.633e-05, [1]
        [cse]: 2.651e-05
[auto_monad_reorder]: 2.496e-05
[get_jit_bprop_graph]: 2.08002e-06
[rewriter_after_jit_bprop_graph]: 7.08e-06
[opt_after_jit_grad]: 0.00056807
[symbol_engine_optimizer]: 0.00010834, [1]
    [Cycle 1]: 9.945e-05, [6]
        [build]: 5.67999e-06
        [elim_shapecalc]: 1.245e-05
        [elim_not_effective]: 2.338e-05
        [opt_reshape]: 9.68002e-06
        [fold_const_symbol]: 1.628e-05
        [renormalize]: 7.39994e-07
[validate]: 0.00012529
[backend_pass]: 1.39e-06
[task_emit]: 0.0100928
[execute]: 7.8e-06
Sums
    bootstrap                            :     0.000189s :  0.59%
    type_inference                       :     0.010084s : 31.70%
    event_method                         :     0.000050s :  0.16%
    auto_monad                           :     0.000084s :  0.26%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000024s :  0.07%
    rewriter_before_opt_a                :     0.000088s :  0.28%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000108s :  0.34%
    jit_opt_a.loop_unroll                :     0.000084s :  0.26%
    jit_opt_a.a_1                        :     0.002279s :  7.16%
    jit_opt_a.with_stream_mark           :     0.000053s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000038s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000019s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000016s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000014s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.02%
    jit_opt_a.specialize_transform       :     0.000032s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000029s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000030s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000009s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000030s :  0.09%
    jit_opt_a.merge_forward              :     0.000018s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000064s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000053s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001560s :  4.90%
    jit_opt_a.replace_old_param          :     0.000082s :  0.26%
    jit_opt_a.inline_without_move        :     0.000072s :  0.23%
    jit_opt_a.renormalize                :     0.003503s : 11.01%
    jit_opt_a.add_forward_monad_depend   :     0.000020s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000010s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000087s :  0.27%
    jit_opt_a.cse                        :     0.000253s :  0.80%
    jit_opt_a.replace_applicator         :     0.000103s :  0.32%
    py_interpret_to_execute_after_opt_a  :     0.000020s :  0.06%
    rewriter_after_opt_a                 :     0.000056s :  0.18%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000822s :  2.58%
    jit_opt_b.frontend_op_eliminate      :     0.000028s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000027s :  0.09%
    cconv                                :     0.000031s :  0.10%
    loop_unroll                          :     0.000515s :  1.62%
    jit_opt_after_cconv.c_1              :     0.000038s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.cse              :     0.000047s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000032s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000010s :  0.03%
    remove_dup_value                     :     0.000026s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000009s :  0.03%
    add_recomputation                    :     0.000067s :  0.21%
    cse_after_recomputation.cse          :     0.000027s :  0.08%
    auto_monad_reorder                   :     0.000025s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.02%
    opt_after_jit_grad                   :     0.000568s :  1.79%
    symbol_engine_optimizer.build        :     0.000006s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000023s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000016s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000125s :  0.39%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.010093s : 31.72%
    execute                              :     0.000008s :  0.02%


Time group info:
------[substitution.]   0.000567   126
 7.92% :     0.000045s :      2: substitution.arithmetic_simplify
 0.67% :     0.000004s :      5: substitution.elim_not_effective
 0.48% :     0.000003s :      5: substitution.fold_const_symbol
 1.53% :     0.000009s :      6: substitution.graph_param_transform
57.21% :     0.000324s :     12: substitution.inline
 3.44% :     0.000020s :      2: substitution.inline_without_move
 1.92% :     0.000011s :     18: substitution.j_node_and_user_rematch
 1.88% :     0.000011s :      7: substitution.minmaximum_grad
 1.34% :     0.000008s :      5: substitution.partial_eliminate
 2.47% :     0.000014s :     18: substitution.remove_not_recompute_node
 4.99% :     0.000028s :     10: substitution.replace_applicator
 1.54% :     0.000009s :      7: substitution.replace_old_param
 0.48% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.98% :     0.000023s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.65% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 7.49% :     0.000042s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010012     2
87.46% :     0.008757s :      1: type_inference.infer
12.54% :     0.001255s :      1: type_inference.specialize
------[replace.]   0.000146    19
63.62% :     0.000093s :     12: replace.inline
36.38% :     0.000053s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000338    19
94.08% :     0.000318s :     12: match.inline
 5.92% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000484  2895
 1.45% :     0.000007s :     48: predicate.accumulaten_eliminater
 0.60% :     0.000003s :      6: predicate.ad_related_special_op_eliminate
 1.20% :     0.000006s :     48: predicate.addn_check_dump
 1.46% :     0.000007s :     48: predicate.addn_zero_filter
 1.95% :     0.000009s :     48: predicate.arithmetic_simplify
 1.41% :     0.000007s :     48: predicate.cast_eliminate
 0.20% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.20% :     0.000006s :     48: predicate.compare_switch_simplify
 1.33% :     0.000006s :     48: predicate.depend_value_elim
 1.20% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.35% :     0.000007s :     48: predicate.dict_get_item_eliminator
 1.26% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.43% :     0.000002s :      6: predicate.dumpgradient_eliminate
 0.17% :     0.000001s :      6: predicate.elim_not_effective
 0.23% :     0.000001s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.31% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.24% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.28% :     0.000006s :     48: predicate.environ_get_eliminate
 1.22% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.10% :     0.000001s :      6: predicate.fold_const_symbol
 0.87% :     0.000004s :     23: predicate.get_grad_eliminate
 0.14% :     0.000001s :      6: predicate.graph_param_transform
 3.70% :     0.000018s :     79: predicate.inline
 1.54% :     0.000007s :     44: predicate.inline_without_move
 0.32% :     0.000002s :     23: predicate.j_node_and_user_rematch
 0.82% :     0.000004s :     23: predicate.less_batch_normalization
 1.58% :     0.000008s :     55: predicate.list_to_tuple_eliminator_
 1.71% :     0.000008s :     61: predicate.load_eliminater
 0.67% :     0.000003s :      6: predicate.loop_unroll_after_grad
 2.64% :     0.000013s :     90: predicate.loop_unroll_before_grad
 1.60% :     0.000008s :     54: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000006s :     48: predicate.merge_addn
 1.29% :     0.000006s :     48: predicate.minmaximum_grad
 0.95% :     0.000005s :      6: predicate.mutable_eliminate
 0.20% :     0.000001s :      6: predicate.opt_reshape
 2.07% :     0.000010s :     61: predicate.partial_eliminate
 1.24% :     0.000006s :     48: predicate.print_const_string_wrapper
14.11% :     0.000068s :     48: predicate.reduce_eliminate
 1.69% :     0.000008s :     55: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.26% :     0.000011s :    108: predicate.replace_applicator
 0.75% :     0.000004s :     44: predicate.replace_old_param
 0.16% :     0.000001s :      6: predicate.reset_defer_inline
 1.41% :     0.000007s :     48: predicate.reshape_eliminate
 1.29% :     0.000006s :     48: predicate.row_tensor_add_zeros_like
 0.36% :     0.000002s :      6: predicate.row_tensor_eliminate
 1.26% :     0.000006s :     48: predicate.same_eliminate
 0.42% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.50% :     0.000002s :     12: predicate.special_op_eliminate
 0.80% :     0.000004s :     23: predicate.specialize_transform
 1.43% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000006s :     48: predicate.stack_unstack_eliminate
 0.25% :     0.000001s :      6: predicate.switch_call_monad_eliminater
 2.22% :     0.000011s :     67: predicate.switch_defer_inline
 2.02% :     0.000010s :     67: predicate.switch_layer_defer_inline
 5.09% :     0.000025s :    163: predicate.switch_simplify
 1.35% :     0.000007s :     48: predicate.tile_eliminate
 1.26% :     0.000006s :     48: predicate.transpose_eliminate
 1.64% :     0.000008s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000007s :     48: predicate.tuple_list_get_item_depend_reorder
 2.68% :     0.000013s :     67: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000008s :     48: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000007s :     55: predicate.tuple_to_list_eliminator_
 1.67% :     0.000008s :     61: predicate.updatestate_pure_node_eliminater
 2.57% :     0.000012s :     84: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000008s :     48: predicate.value_based_eliminate
 0.18% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.31% :     0.000001s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001561    29
57.35% :     0.000895s :     13: func_graph_cloner_run.FuncGraphClonerGraph
42.65% :     0.000666s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.042716    91
 0.17% :     0.000071s :      1: add_recomputation
 0.21% :     0.000088s :      1: auto_monad
 0.06% :     0.000028s :      1: auto_monad_reorder
 0.01% :     0.000005s :      1: backend_pass
 0.47% :     0.000202s :      1: bootstrap
 0.08% :     0.000034s :      1: cconv
 0.03% :     0.000012s :      1: convert_after_rewriter
 0.11% :     0.000046s :      1: cse_after_recomputation
 0.03% :     0.000012s :      1: environ_conv
 0.13% :     0.000054s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
29.11% :     0.012434s :      1: jit_opt_a
 0.55% :     0.000234s :      1: jit_opt_after_cconv
 0.18% :     0.000079s :      1: jit_opt_b
 1.23% :     0.000527s :      1: loop_unroll
 1.95% :     0.000833s :      1: mutable_eliminate
 6.92% :     0.002956s :     39: opt.transform.jit_opt_a
 0.20% :     0.000085s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000048s :      4: opt.transform.jit_opt_b
 0.05% :     0.000022s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000026s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000037s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000058s :      4: opt.transform.symbol_engine_opt
 1.36% :     0.000579s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.06% :     0.000026s :      1: py_interpret_to_execute
 0.05% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000030s :      1: remove_dup_value
 4.57% :     0.001952s :      2: renormalize.infer
 3.60% :     0.001536s :      2: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000059s :      1: rewriter_after_opt_a
 0.21% :     0.000090s :      1: rewriter_before_opt_a
 0.26% :     0.000111s :      1: symbol_engine_optimizer
23.66% :     0.010106s :      1: task_emit
23.64% :     0.010099s :      1: type_inference
 0.37% :     0.000156s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.532.93 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c375b80> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.539.62 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0100565, [30]
[bootstrap]: 0.00023293
[type_inference]: 0.00355213
[event_method]: 1.523e-05
[auto_monad]: 4.596e-05
[graph_reusing]: 4.27e-06
[pre_auto_parallel]: 1.69e-06
[py_interpret_to_execute]: 2.49e-05
[rewriter_before_opt_a]: 6.322e-05
[expand_dump_flag]: 2.27999e-06
[jit_opt_a]: 0.00423286, [2]
    [Cycle 1]: 0.00119642, [27]
        [switch_simplify]: 4.326e-05
        [loop_unroll]: 2.837e-05
        [a_1]: 0.00045099
        [with_stream_mark]: 1.201e-05
        [recompute_prepare]: 6.76e-06
        [updatestate_depend_eliminate]: 3.26001e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 5.72001e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.25001e-06
        [meta_shard_fg_expand]: 1.24e-06
        [get_grad_eliminate_]: 4.94e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 8.70001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.172e-05
        [j_node_and_user_rematch]: 8.30999e-06
        [meta_fg_expand]: 1.82999e-06
        [replace_old_param]: 7.55e-06
        [inline_without_move]: 5.18002e-06
        [renormalize]: 0.00039164
        [add_forward_monad_depend]: 4.63999e-06
        [auto_monad_grad]: 1.35999e-06
        [auto_monad_eliminator]: 1.075e-05
        [cse]: 1.488e-05
        [replace_applicator]: 1.044e-05
    [Cycle 2]: 0.00030589, [27]
        [switch_simplify]: 5.11002e-06
        [loop_unroll]: 4.74998e-06
        [a_1]: 8.679e-05
        [with_stream_mark]: 7.92e-06
        [recompute_prepare]: 4.75001e-06
        [updatestate_depend_eliminate]: 2.61e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 8.59989e-07
        [specialize_transform]: 4.65999e-06
        [updatestate_useless_node_eliminater]: 4.58999e-06
        [accelerated_algorithm]: 5.26002e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.18999e-06
        [merge_forward]: 2.36e-06
        [cell_reuse_recompute_pass]: 1.42e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.14e-05
        [j_node_and_user_rematch]: 7.84002e-06
        [meta_fg_expand]: 1.46002e-06
        [replace_old_param]: 7.09001e-06
        [inline_without_move]: 4.67e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.02998e-06
        [auto_monad_grad]: 7.10017e-07
        [auto_monad_eliminator]: 4.53999e-06
        [cse]: 9.73002e-06
        [replace_applicator]: 4.97999e-06
[py_interpret_to_execute_after_opt_a]: 7.32002e-06
[rewriter_after_opt_a]: 2.504e-05
[convert_after_rewriter]: 5.89e-06
[order_py_execute_after_rewriter]: 4.12e-06
[mutable_eliminate]: 0.00045473
[jit_opt_b]: 4.694e-05, [1]
    [Cycle 1]: 4.053e-05, [2]
        [frontend_op_eliminate]: 1.509e-05
        [inline_after_opt_a]: 1.372e-05
[cconv]: 1.752e-05
[loop_unroll]: 0.00040522
[jit_opt_after_cconv]: 0.00012853, [1]
    [Cycle 1]: 0.00012236, [11]
        [c_1]: 1.949e-05
        [parameter_eliminate]: 2.27001e-06
        [updatestate_depend_eliminate]: 5.09e-06
        [updatestate_assign_eliminate]: 2.26998e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.49e-05
        [call_graph_tuple_transform]: 1.631e-05
        [tuple_list_get_item_eliminator]: 5.08002e-06
        [none_parameter_eliminate]: 1.00001e-06
        [renormalize]: 3.59985e-07
        [switch_simplify]: 5.20999e-06
[remove_dup_value]: 9.68002e-06
[partial_unused_args_eliminate]: 1.73002e-06
[environ_conv]: 3.68e-06
[add_recomputation]: 3.328e-05
[cse_after_recomputation]: 1.849e-05, [1]
    [Cycle 1]: 1.372e-05, [1]
        [cse]: 8.27998e-06
[auto_monad_reorder]: 1.059e-05
[get_jit_bprop_graph]: 1.30999e-06
[rewriter_after_jit_bprop_graph]: 4.18999e-06
[opt_after_jit_grad]: 0.0004691
[symbol_engine_optimizer]: 6.844e-05, [1]
    [Cycle 1]: 6.24e-05, [6]
        [build]: 2.84001e-06
        [elim_shapecalc]: 7.28e-06
        [elim_not_effective]: 1.158e-05
        [opt_reshape]: 5.30001e-06
        [fold_const_symbol]: 7.95998e-06
        [renormalize]: 4.59986e-07
[validate]: 2.459e-05
Sums
    bootstrap                            :     0.000233s :  3.42%
    type_inference                       :     0.003552s : 52.20%
    event_method                         :     0.000015s :  0.22%
    auto_monad                           :     0.000046s :  0.68%
    graph_reusing                        :     0.000004s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000025s :  0.37%
    rewriter_before_opt_a                :     0.000063s :  0.93%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000048s :  0.71%
    jit_opt_a.loop_unroll                :     0.000033s :  0.49%
    jit_opt_a.a_1                        :     0.000538s :  7.90%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.09%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000010s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.13%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.24%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.22%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000392s :  5.76%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.22%
    jit_opt_a.cse                        :     0.000025s :  0.36%
    jit_opt_a.replace_applicator         :     0.000015s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.11%
    rewriter_after_opt_a                 :     0.000025s :  0.37%
    convert_after_rewriter               :     0.000006s :  0.09%
    order_py_execute_after_rewriter      :     0.000004s :  0.06%
    mutable_eliminate                    :     0.000455s :  6.68%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.20%
    cconv                                :     0.000018s :  0.26%
    loop_unroll                          :     0.000405s :  5.96%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.29%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000015s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.08%
    remove_dup_value                     :     0.000010s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000033s :  0.49%
    cse_after_recomputation.cse          :     0.000008s :  0.12%
    auto_monad_reorder                   :     0.000011s :  0.16%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.06%
    opt_after_jit_grad                   :     0.000469s :  6.89%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000025s :  0.36%


Time group info:
------[substitution.]   0.000147    23
 0.92% :     0.000001s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.86% :     0.000004s :      3: substitution.graph_param_transform
82.96% :     0.000122s :      4: substitution.inline
 1.71% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.65% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.17% :     0.000003s :      2: substitution.replace_old_param
 6.05% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003507     2
80.93% :     0.002838s :      1: type_inference.infer
19.07% :     0.000669s :      1: type_inference.specialize
------[replace.]   0.000056     6
66.11% :     0.000037s :      4: replace.inline
33.89% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000127     6
93.94% :     0.000120s :      4: match.inline
 6.06% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000107   691
 1.33% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.23% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000001s :     10: predicate.addn_zero_filter
 1.85% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000001s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.04% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.73% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.40% :     0.000000s :      3: predicate.graph_param_transform
 5.55% :     0.000006s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.44% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.58% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.22% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000001s :     10: predicate.merge_addn
 1.16% :     0.000001s :     10: predicate.minmaximum_grad
 1.62% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000000s :      3: predicate.opt_reshape
 2.46% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.54% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.82% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.33% :     0.000001s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.80% :     0.000003s :     16: predicate.switch_defer_inline
 2.23% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.85% :     0.000008s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.40% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.38% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.43% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000436    12
49.31% :     0.000215s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.50% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.19% :     0.000197s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011244    72
 0.32% :     0.000036s :      1: add_recomputation
 0.44% :     0.000049s :      1: auto_monad
 0.12% :     0.000013s :      1: auto_monad_reorder
 2.16% :     0.000243s :      1: bootstrap
 0.18% :     0.000020s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.18% :     0.000021s :      1: cse_after_recomputation
 0.05% :     0.000006s :      1: environ_conv
 0.17% :     0.000019s :      1: event_method
 0.04% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000003s :      1: get_jit_bprop_graph
 0.06% :     0.000006s :      1: graph_reusing
37.67% :     0.004235s :      1: jit_opt_a
 1.16% :     0.000131s :      1: jit_opt_after_cconv
 0.44% :     0.000050s :      1: jit_opt_b
 3.66% :     0.000412s :      1: loop_unroll
 4.11% :     0.000462s :      1: mutable_eliminate
 6.46% :     0.000726s :     26: opt.transform.jit_opt_a
 0.38% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000023s :      4: opt.transform.jit_opt_b
 0.10% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000029s :      4: opt.transform.symbol_engine_opt
 4.24% :     0.000477s :      1: opt_after_jit_grad
 0.06% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000027s :      1: py_interpret_to_execute
 0.09% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000012s :      1: remove_dup_value
 1.65% :     0.000185s :      1: renormalize.infer
 1.78% :     0.000201s :      1: renormalize.specialize
 0.06% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000028s :      1: rewriter_after_opt_a
 0.59% :     0.000066s :      1: rewriter_before_opt_a
 0.63% :     0.000071s :      1: symbol_engine_optimizer
31.69% :     0.003563s :      1: type_inference

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.563.20 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80cad30> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.568.48 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0176368, [33]
[bootstrap]: 0.00030472
[type_inference]: 0.00439607
[event_method]: 1.594e-05
[auto_monad]: 4.197e-05
[graph_reusing]: 3.98001e-06
[pre_auto_parallel]: 1.87001e-06
[py_interpret_to_execute]: 2.64e-05
[rewriter_before_opt_a]: 6.215e-05
[expand_dump_flag]: 2.26e-06
[jit_opt_a]: 0.00474516, [2]
    [Cycle 1]: 0.00140411, [27]
        [switch_simplify]: 4.302e-05
        [loop_unroll]: 2.813e-05
        [a_1]: 0.00044669
        [with_stream_mark]: 1.209e-05
        [recompute_prepare]: 6.77002e-06
        [updatestate_depend_eliminate]: 2.93998e-06
        [updatestate_assign_eliminate]: 2.13002e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.31002e-06
        [specialize_transform]: 5.56e-06
        [updatestate_useless_node_eliminater]: 4.68999e-06
        [accelerated_algorithm]: 5.05999e-06
        [meta_shard_fg_expand]: 1.17999e-06
        [get_grad_eliminate_]: 4.95001e-06
        [merge_forward]: 2.34999e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.082e-05
        [j_node_and_user_rematch]: 8.22e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 7.45998e-06
        [inline_without_move]: 4.94e-06
        [renormalize]: 0.0005757
        [add_forward_monad_depend]: 6.46999e-06
        [auto_monad_grad]: 2.44001e-06
        [auto_monad_eliminator]: 1.372e-05
        [cse]: 2.755e-05
        [replace_applicator]: 1.397e-05
    [Cycle 2]: 0.00032185, [27]
        [switch_simplify]: 6.46999e-06
        [loop_unroll]: 5.46e-06
        [a_1]: 9.396e-05
        [with_stream_mark]: 8.56002e-06
        [recompute_prepare]: 4.94e-06
        [updatestate_depend_eliminate]: 2.48998e-06
        [updatestate_assign_eliminate]: 2.05002e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 4.72e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 4.82e-06
        [meta_shard_fg_expand]: 1.64998e-06
        [get_grad_eliminate_]: 4.39002e-06
        [merge_forward]: 2.21e-06
        [cell_reuse_recompute_pass]: 1.52001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.195e-05
        [j_node_and_user_rematch]: 7.5e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 7.06001e-06
        [inline_without_move]: 4.46002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.67999e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 5.19998e-06
        [cse]: 1.137e-05
        [replace_applicator]: 4.82998e-06
[py_interpret_to_execute_after_opt_a]: 9.19998e-06
[rewriter_after_opt_a]: 3.421e-05
[convert_after_rewriter]: 6.75998e-06
[order_py_execute_after_rewriter]: 4.70999e-06
[mutable_eliminate]: 0.00050718
[jit_opt_b]: 4.838e-05, [1]
    [Cycle 1]: 4.146e-05, [2]
        [frontend_op_eliminate]: 1.549e-05
        [inline_after_opt_a]: 1.464e-05
[cconv]: 2.325e-05
[loop_unroll]: 0.00042617
[jit_opt_after_cconv]: 0.00013614, [1]
    [Cycle 1]: 0.00012979, [11]
        [c_1]: 1.946e-05
        [parameter_eliminate]: 2.91e-06
        [updatestate_depend_eliminate]: 5.28002e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [cse]: 1.813e-05
        [call_graph_tuple_transform]: 1.845e-05
        [tuple_list_get_item_eliminator]: 5.25001e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 5.20001e-06
[remove_dup_value]: 1.413e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 4.82998e-06
[add_recomputation]: 4.689e-05
[cse_after_recomputation]: 1.996e-05, [1]
    [Cycle 1]: 1.449e-05, [1]
        [cse]: 8.97999e-06
[auto_monad_reorder]: 1.407e-05
[get_jit_bprop_graph]: 1.87001e-06
[rewriter_after_jit_bprop_graph]: 5.30001e-06
[opt_after_jit_grad]: 0.00043203
[symbol_engine_optimizer]: 6.938e-05, [1]
    [Cycle 1]: 6.304e-05, [6]
        [build]: 3.23e-06
        [elim_shapecalc]: 7.05002e-06
        [elim_not_effective]: 1.155e-05
        [opt_reshape]: 5.39e-06
        [fold_const_symbol]: 8.37e-06
        [renormalize]: 6.50005e-07
[validate]: 3.077e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00601243
[execute]: 7.77998e-06
Sums
    bootstrap                            :     0.000305s :  2.17%
    type_inference                       :     0.004396s : 31.30%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000042s :  0.30%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.19%
    rewriter_before_opt_a                :     0.000062s :  0.44%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000049s :  0.35%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000541s :  3.85%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000576s :  4.10%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.13%
    jit_opt_a.cse                        :     0.000039s :  0.28%
    jit_opt_a.replace_applicator         :     0.000019s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000507s :  3.61%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000426s :  3.03%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000047s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000432s :  3.08%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000031s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006012s : 42.80%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000143    23
 1.28% :     0.000002s :      2: substitution.elim_not_effective
 0.86% :     0.000001s :      2: substitution.fold_const_symbol
 3.80% :     0.000005s :      3: substitution.graph_param_transform
81.84% :     0.000117s :      4: substitution.inline
 1.77% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.66% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.22% :     0.000003s :      2: substitution.replace_old_param
 5.58% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004345     2
83.21% :     0.003616s :      1: type_inference.infer
16.79% :     0.000730s :      1: type_inference.specialize
------[replace.]   0.000057     6
64.53% :     0.000036s :      4: replace.inline
35.47% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000122     6
94.33% :     0.000115s :      4: match.inline
 5.67% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.12% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.17% :     0.000001s :     10: predicate.addn_check_dump
 1.45% :     0.000002s :     10: predicate.addn_zero_filter
 2.02% :     0.000002s :     10: predicate.arithmetic_simplify
 1.32% :     0.000001s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.21% :     0.000001s :     10: predicate.compare_switch_simplify
 1.58% :     0.000002s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.32% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.24% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.29% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.07% :     0.000006s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.72% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.92% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.23% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.22% :     0.000001s :     10: predicate.merge_addn
 1.16% :     0.000001s :     10: predicate.minmaximum_grad
 1.30% :     0.000001s :      3: predicate.mutable_eliminate
 0.45% :     0.000000s :      3: predicate.opt_reshape
 2.12% :     0.000002s :     15: predicate.partial_eliminate
 1.28% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.50% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.67% :     0.000003s :     16: predicate.switch_defer_inline
 2.30% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.73% :     0.000008s :     53: predicate.switch_simplify
 1.32% :     0.000001s :     10: predicate.tile_eliminate
 1.52% :     0.000002s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.30% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.80% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.48% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000505    12
49.19% :     0.000248s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.90% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.90% :     0.000232s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019047    76
 0.26% :     0.000050s :      1: add_recomputation
 0.24% :     0.000045s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.67% :     0.000318s :      1: bootstrap
 0.14% :     0.000026s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000006s :      1: graph_reusing
24.93% :     0.004748s :      1: jit_opt_a
 0.73% :     0.000139s :      1: jit_opt_after_cconv
 0.27% :     0.000051s :      1: jit_opt_b
 2.28% :     0.000434s :      1: loop_unroll
 2.71% :     0.000515s :      1: mutable_eliminate
 3.84% :     0.000732s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.31% :     0.000440s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000029s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000016s :      1: remove_dup_value
 1.40% :     0.000267s :      1: renormalize.infer
 1.59% :     0.000303s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000037s :      1: rewriter_after_opt_a
 0.34% :     0.000065s :      1: rewriter_before_opt_a
 0.38% :     0.000072s :      1: symbol_engine_optimizer
31.63% :     0.006024s :      1: task_emit
23.14% :     0.004408s :      1: type_inference
 0.26% :     0.000049s :      1: validate


TotalTime = 0.0209201, [33]
[bootstrap]: 0.0002054
[type_inference]: 0.0101004
[event_method]: 4.69e-05
[auto_monad]: 9.235e-05
[graph_reusing]: 6.89001e-06
[pre_auto_parallel]: 2.77002e-06
[py_interpret_to_execute]: 2.479e-05
[rewriter_before_opt_a]: 8.822e-05
[expand_dump_flag]: 3.41999e-06
[jit_opt_a]: 0.00825619, [2]
    [Cycle 1]: 0.00453906, [27]
        [switch_simplify]: 6.126e-05
        [loop_unroll]: 3.801e-05
        [a_1]: 0.00081191
        [with_stream_mark]: 2.717e-05
        [recompute_prepare]: 1.827e-05
        [updatestate_depend_eliminate]: 8.1e-06
        [updatestate_assign_eliminate]: 6.63e-06
        [updatestate_loads_eliminate]: 6.25002e-06
        [parameter_eliminate]: 2.85002e-06
        [specialize_transform]: 1.347e-05
        [updatestate_useless_node_eliminater]: 1.187e-05
        [accelerated_algorithm]: 1.3e-05
        [meta_shard_fg_expand]: 3.89002e-06
        [get_grad_eliminate_]: 4.018e-05
        [merge_forward]: 9.43002e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.729e-05
        [j_node_and_user_rematch]: 2.355e-05
        [meta_fg_expand]: 0.00149802
        [replace_old_param]: 5.269e-05
        [inline_without_move]: 5.21e-05
        [renormalize]: 0.00155639
        [add_forward_monad_depend]: 5.17e-06
        [auto_monad_grad]: 2.47001e-06
        [auto_monad_eliminator]: 1.191e-05
        [cse]: 2.571e-05
        [replace_applicator]: 1.033e-05
    [Cycle 2]: 0.0002694, [27]
        [switch_simplify]: 4.37e-06
        [loop_unroll]: 3.66999e-06
        [a_1]: 3.854e-05
        [with_stream_mark]: 8.03001e-06
        [recompute_prepare]: 3.37002e-06
        [updatestate_depend_eliminate]: 1.94e-06
        [updatestate_assign_eliminate]: 1.69e-06
        [updatestate_loads_eliminate]: 1.52001e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 3.41001e-06
        [updatestate_useless_node_eliminater]: 3.18998e-06
        [accelerated_algorithm]: 3.23998e-06
        [meta_shard_fg_expand]: 9.50007e-07
        [get_grad_eliminate_]: 3.64002e-06
        [merge_forward]: 1.90001e-06
        [cell_reuse_recompute_pass]: 2.27001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.36e-06
        [j_node_and_user_rematch]: 6.04001e-06
        [meta_fg_expand]: 3.354e-05
        [replace_old_param]: 4.49998e-06
        [inline_without_move]: 3.33e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 6.29982e-07
        [auto_monad_eliminator]: 3.84002e-06
        [cse]: 9.36e-06
        [replace_applicator]: 3.7e-06
[py_interpret_to_execute_after_opt_a]: 9.17999e-06
[rewriter_after_opt_a]: 3.332e-05
[convert_after_rewriter]: 5.78997e-06
[order_py_execute_after_rewriter]: 4.48999e-06
[mutable_eliminate]: 0.00059584
[jit_opt_b]: 4.865e-05, [1]
    [Cycle 1]: 4.174e-05, [2]
        [frontend_op_eliminate]: 9.69e-06
        [inline_after_opt_a]: 1.175e-05
[cconv]: 2.078e-05
[loop_unroll]: 0.00042046
[jit_opt_after_cconv]: 0.00011893, [1]
    [Cycle 1]: 0.00011195, [11]
        [c_1]: 1.14e-05
        [parameter_eliminate]: 2.84999e-06
        [updatestate_depend_eliminate]: 4.82e-06
        [updatestate_assign_eliminate]: 1.82001e-06
        [updatestate_loads_eliminate]: 1.66e-06
        [cse]: 1.722e-05
        [call_graph_tuple_transform]: 1.363e-05
        [tuple_list_get_item_eliminator]: 3.6e-06
        [none_parameter_eliminate]: 1.63002e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 3.70998e-06
[remove_dup_value]: 1.573e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 4.80001e-06
[add_recomputation]: 3.416e-05
[cse_after_recomputation]: 1.943e-05, [1]
    [Cycle 1]: 1.395e-05, [1]
        [cse]: 8.23999e-06
[auto_monad_reorder]: 1.325e-05
[get_jit_bprop_graph]: 1.86e-06
[rewriter_after_jit_bprop_graph]: 5.12e-06
[opt_after_jit_grad]: 0.00045179
[symbol_engine_optimizer]: 6.363e-05, [1]
    [Cycle 1]: 5.759e-05, [6]
        [build]: 3.16001e-06
        [elim_shapecalc]: 6.21e-06
        [elim_not_effective]: 9.93002e-06
        [opt_reshape]: 4.17003e-06
        [fold_const_symbol]: 5.86e-06
        [renormalize]: 4.19997e-07
[validate]: 2.817e-05
[backend_pass]: 9.89996e-07
[task_emit]: 2.248e-05
[execute]: 1.15999e-06
Sums
    bootstrap                            :     0.000205s :  1.22%
    type_inference                       :     0.010100s : 59.90%
    event_method                         :     0.000047s :  0.28%
    auto_monad                           :     0.000092s :  0.55%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000025s :  0.15%
    rewriter_before_opt_a                :     0.000088s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.39%
    jit_opt_a.loop_unroll                :     0.000042s :  0.25%
    jit_opt_a.a_1                        :     0.000850s :  5.04%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000044s :  0.26%
    jit_opt_a.merge_forward              :     0.000011s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000030s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001532s :  9.08%
    jit_opt_a.replace_old_param          :     0.000057s :  0.34%
    jit_opt_a.inline_without_move        :     0.000055s :  0.33%
    jit_opt_a.renormalize                :     0.001556s :  9.23%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.09%
    jit_opt_a.cse                        :     0.000035s :  0.21%
    jit_opt_a.replace_applicator         :     0.000014s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.05%
    rewriter_after_opt_a                 :     0.000033s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000596s :  3.53%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.07%
    cconv                                :     0.000021s :  0.12%
    loop_unroll                          :     0.000420s :  2.49%
    jit_opt_after_cconv.c_1              :     0.000011s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000034s :  0.20%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000452s :  2.68%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.17%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.000022s :  0.13%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000233    50
 0.62% :     0.000001s :      1: substitution.elim_not_effective
 0.41% :     0.000001s :      1: substitution.fold_const_symbol
 1.51% :     0.000004s :      1: substitution.graph_param_transform
63.62% :     0.000148s :      6: substitution.inline
 7.23% :     0.000017s :      2: substitution.inline_without_move
 2.79% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.44% :     0.000003s :      2: substitution.minmaximum_grad
 1.87% :     0.000004s :      3: substitution.partial_eliminate
 3.02% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.22% :     0.000005s :      2: substitution.replace_applicator
 2.15% :     0.000005s :      5: substitution.replace_old_param
 1.12% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.99% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.91% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 7.10% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010032     2
89.67% :     0.008996s :      1: type_inference.infer
10.33% :     0.001036s :      1: type_inference.specialize
------[replace.]   0.000059     8
77.99% :     0.000046s :      6: replace.inline
22.01% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000156     8
92.69% :     0.000145s :      6: match.inline
 7.31% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000168  1109
 1.98% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.64% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.31% :     0.000002s :     18: predicate.addn_check_dump
 1.62% :     0.000003s :     18: predicate.addn_zero_filter
 2.03% :     0.000003s :     18: predicate.arithmetic_simplify
 1.41% :     0.000002s :     18: predicate.cast_eliminate
 0.10% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.26% :     0.000002s :     18: predicate.compare_switch_simplify
 1.47% :     0.000002s :     18: predicate.depend_value_elim
 1.27% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.52% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.54% :     0.000003s :     18: predicate.dict_set_item_eliminator
 0.53% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.15% :     0.000000s :      1: predicate.elim_not_effective
 0.23% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.67% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.24% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.24% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     18: predicate.environ_get_eliminate
 1.25% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.07% :     0.000000s :      1: predicate.fold_const_symbol
 1.43% :     0.000002s :     12: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 4.48% :     0.000008s :     28: predicate.inline
 3.33% :     0.000006s :     33: predicate.inline_without_move
 0.51% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.27% :     0.000002s :     12: predicate.less_batch_normalization
 1.53% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.63% :     0.000003s :     21: predicate.load_eliminater
 0.86% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.58% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.78% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.26% :     0.000002s :     18: predicate.merge_addn
 1.34% :     0.000002s :     18: predicate.minmaximum_grad
 0.94% :     0.000002s :      1: predicate.mutable_eliminate
 0.26% :     0.000000s :      1: predicate.opt_reshape
 2.09% :     0.000004s :     21: predicate.partial_eliminate
 1.32% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.62% :     0.000003s :     18: predicate.reduce_eliminate
 1.70% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.47% :     0.000002s :     22: predicate.replace_applicator
 1.76% :     0.000003s :     33: predicate.replace_old_param
 0.12% :     0.000000s :      1: predicate.reset_defer_inline
 1.43% :     0.000002s :     18: predicate.reshape_eliminate
 1.47% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.51% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     18: predicate.same_eliminate
 0.60% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.60% :     0.000001s :      2: predicate.special_op_eliminate
 1.27% :     0.000002s :     12: predicate.specialize_transform
 1.69% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.64% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.15% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.46% :     0.000004s :     26: predicate.switch_defer_inline
 2.29% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.61% :     0.000011s :     71: predicate.switch_simplify
 1.41% :     0.000002s :     18: predicate.tile_eliminate
 1.26% :     0.000002s :     18: predicate.transpose_eliminate
 1.72% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.51% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.94% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.57% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.91% :     0.000003s :     18: predicate.value_based_eliminate
 0.11% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.33% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001112    22
73.55% :     0.000818s :     13: func_graph_cloner_run.FuncGraphClonerGraph
26.45% :     0.000294s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023772    76
 0.16% :     0.000037s :      1: add_recomputation
 0.41% :     0.000097s :      1: auto_monad
 0.07% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 0.92% :     0.000218s :      1: bootstrap
 0.10% :     0.000023s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.21% :     0.000051s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
34.74% :     0.008259s :      1: jit_opt_a
 0.51% :     0.000122s :      1: jit_opt_after_cconv
 0.21% :     0.000051s :      1: jit_opt_b
 1.80% :     0.000428s :      1: loop_unroll
 2.54% :     0.000604s :      1: mutable_eliminate
 5.19% :     0.001235s :     26: opt.transform.jit_opt_a
 0.12% :     0.000029s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000015s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000023s :      4: opt.transform.symbol_engine_opt
 1.93% :     0.000460s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000027s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 5.10% :     0.001212s :      1: renormalize.infer
 1.41% :     0.000336s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000036s :      1: rewriter_after_opt_a
 0.38% :     0.000091s :      1: rewriter_before_opt_a
 0.28% :     0.000066s :      1: symbol_engine_optimizer
 0.11% :     0.000026s :      1: task_emit
42.54% :     0.010112s :      1: type_inference
 0.20% :     0.000047s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.813.56 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c371040> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.820.15 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0338564, [33]
[bootstrap]: 0.0001909
[type_inference]: 0.0101278
[event_method]: 4.657e-05
[auto_monad]: 7.528e-05
[graph_reusing]: 4.99e-06
[pre_auto_parallel]: 2.26e-06
[py_interpret_to_execute]: 2.466e-05
[rewriter_before_opt_a]: 8.457e-05
[expand_dump_flag]: 2.54999e-06
[jit_opt_a]: 0.0119079, [3]
    [Cycle 1]: 0.00587026, [27]
        [switch_simplify]: 6.224e-05
        [loop_unroll]: 3.974e-05
        [a_1]: 0.0008049
        [with_stream_mark]: 2.244e-05
        [recompute_prepare]: 1.799e-05
        [updatestate_depend_eliminate]: 8.57998e-06
        [updatestate_assign_eliminate]: 6.61e-06
        [updatestate_loads_eliminate]: 5.37001e-06
        [parameter_eliminate]: 2.27999e-06
        [specialize_transform]: 1.379e-05
        [updatestate_useless_node_eliminater]: 1.19e-05
        [accelerated_algorithm]: 1.229e-05
        [meta_shard_fg_expand]: 3.50998e-06
        [get_grad_eliminate_]: 1.171e-05
        [merge_forward]: 7.23999e-06
        [cell_reuse_recompute_pass]: 9.89996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.355e-05
        [j_node_and_user_rematch]: 2.187e-05
        [meta_fg_expand]: 0.00142451
        [replace_old_param]: 5.09e-05
        [inline_without_move]: 5.317e-05
        [renormalize]: 0.00276249
        [add_forward_monad_depend]: 1.01e-05
        [auto_monad_grad]: 5.39e-06
        [auto_monad_eliminator]: 5.251e-05
        [cse]: 0.00015272
        [replace_applicator]: 6.809e-05
    [Cycle 2]: 0.00233975, [27]
        [switch_simplify]: 3.764e-05
        [loop_unroll]: 3.671e-05
        [a_1]: 0.00118061
        [with_stream_mark]: 1.574e-05
        [recompute_prepare]: 1.034e-05
        [updatestate_depend_eliminate]: 5.12e-06
        [updatestate_assign_eliminate]: 3.97e-06
        [updatestate_loads_eliminate]: 3.82002e-06
        [parameter_eliminate]: 1.83997e-06
        [specialize_transform]: 8.41002e-06
        [updatestate_useless_node_eliminater]: 8.20999e-06
        [accelerated_algorithm]: 8.40001e-06
        [meta_shard_fg_expand]: 2.79999e-06
        [get_grad_eliminate_]: 8.46002e-06
        [merge_forward]: 4.84003e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.777e-05
        [j_node_and_user_rematch]: 1.366e-05
        [meta_fg_expand]: 4.507e-05
        [replace_old_param]: 1.13e-05
        [inline_without_move]: 8.12998e-06
        [renormalize]: 0.00066637
        [add_forward_monad_depend]: 5.14e-06
        [auto_monad_grad]: 1.60001e-06
        [auto_monad_eliminator]: 1.608e-05
        [cse]: 4.848e-05
        [replace_applicator]: 1.595e-05
    [Cycle 3]: 0.00049054, [27]
        [switch_simplify]: 8.3e-06
        [loop_unroll]: 8.02e-06
        [a_1]: 0.00018937
        [with_stream_mark]: 1.164e-05
        [recompute_prepare]: 8.28999e-06
        [updatestate_depend_eliminate]: 4.57e-06
        [updatestate_assign_eliminate]: 4.42e-06
        [updatestate_loads_eliminate]: 3.63e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 8.15e-06
        [updatestate_useless_node_eliminater]: 7.87e-06
        [accelerated_algorithm]: 8.55999e-06
        [meta_shard_fg_expand]: 1.62001e-06
        [get_grad_eliminate_]: 7.7e-06
        [merge_forward]: 4.3e-06
        [cell_reuse_recompute_pass]: 1.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.654e-05
        [j_node_and_user_rematch]: 1.331e-05
        [meta_fg_expand]: 2.76999e-06
        [replace_old_param]: 1.018e-05
        [inline_without_move]: 7.83001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.74998e-06
        [auto_monad_grad]: 7.90023e-07
        [auto_monad_eliminator]: 1.088e-05
        [cse]: 2.477e-05
        [replace_applicator]: 8.45001e-06
[py_interpret_to_execute_after_opt_a]: 1.461e-05
[rewriter_after_opt_a]: 4.741e-05
[convert_after_rewriter]: 8.92999e-06
[order_py_execute_after_rewriter]: 6.99001e-06
[mutable_eliminate]: 0.00068343
[jit_opt_b]: 6.759e-05, [1]
    [Cycle 1]: 6.002e-05, [2]
        [frontend_op_eliminate]: 2.448e-05
        [inline_after_opt_a]: 2.389e-05
[cconv]: 2.484e-05
[loop_unroll]: 0.00043068
[jit_opt_after_cconv]: 0.00019296, [1]
    [Cycle 1]: 0.00018682, [11]
        [c_1]: 3.517e-05
        [parameter_eliminate]: 2.64999e-06
        [updatestate_depend_eliminate]: 8.13999e-06
        [updatestate_assign_eliminate]: 4.69998e-06
        [updatestate_loads_eliminate]: 4.35e-06
        [cse]: 3.416e-05
        [call_graph_tuple_transform]: 2.607e-05
        [tuple_list_get_item_eliminator]: 8.48999e-06
        [none_parameter_eliminate]: 1.59998e-06
        [renormalize]: 4.40021e-07
        [switch_simplify]: 8.57e-06
[remove_dup_value]: 2.179e-05
[partial_unused_args_eliminate]: 2.40002e-06
[environ_conv]: 6.84001e-06
[add_recomputation]: 5.38e-05
[cse_after_recomputation]: 3.371e-05, [1]
    [Cycle 1]: 2.794e-05, [1]
        [cse]: 2.206e-05
[auto_monad_reorder]: 2.03e-05
[get_jit_bprop_graph]: 1.79998e-06
[rewriter_after_jit_bprop_graph]: 5.51e-06
[opt_after_jit_grad]: 0.00073703
[symbol_engine_optimizer]: 9.638e-05, [1]
    [Cycle 1]: 8.972e-05, [6]
        [build]: 4.15999e-06
        [elim_shapecalc]: 1.194e-05
        [elim_not_effective]: 1.98e-05
        [opt_reshape]: 9.32999e-06
        [fold_const_symbol]: 1.446e-05
        [renormalize]: 6.69999e-07
[validate]: 4.587e-05
[backend_pass]: 9.89996e-07
[task_emit]: 0.00867649
[execute]: 7.56999e-06
Sums
    bootstrap                            :     0.000191s :  0.64%
    type_inference                       :     0.010128s : 33.94%
    event_method                         :     0.000047s :  0.16%
    auto_monad                           :     0.000075s :  0.25%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000025s :  0.08%
    rewriter_before_opt_a                :     0.000085s :  0.28%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000108s :  0.36%
    jit_opt_a.loop_unroll                :     0.000084s :  0.28%
    jit_opt_a.a_1                        :     0.002175s :  7.29%
    jit_opt_a.with_stream_mark           :     0.000050s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000037s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000018s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000013s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000030s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000028s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000029s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000028s :  0.09%
    jit_opt_a.merge_forward              :     0.000016s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000058s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000049s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001472s :  4.93%
    jit_opt_a.replace_old_param          :     0.000072s :  0.24%
    jit_opt_a.inline_without_move        :     0.000069s :  0.23%
    jit_opt_a.renormalize                :     0.003429s : 11.49%
    jit_opt_a.add_forward_monad_depend   :     0.000017s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000079s :  0.27%
    jit_opt_a.cse                        :     0.000226s :  0.76%
    jit_opt_a.replace_applicator         :     0.000092s :  0.31%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.05%
    rewriter_after_opt_a                 :     0.000047s :  0.16%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000683s :  2.29%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.08%
    cconv                                :     0.000025s :  0.08%
    loop_unroll                          :     0.000431s :  1.44%
    jit_opt_after_cconv.c_1              :     0.000035s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000034s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.03%
    remove_dup_value                     :     0.000022s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000054s :  0.18%
    cse_after_recomputation.cse          :     0.000022s :  0.07%
    auto_monad_reorder                   :     0.000020s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000737s :  2.47%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000020s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008676s : 29.07%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000515   126
 6.96% :     0.000036s :      2: substitution.arithmetic_simplify
 0.55% :     0.000003s :      5: substitution.elim_not_effective
 0.46% :     0.000002s :      5: substitution.fold_const_symbol
 1.33% :     0.000007s :      6: substitution.graph_param_transform
58.89% :     0.000303s :     12: substitution.inline
 3.36% :     0.000017s :      2: substitution.inline_without_move
 1.70% :     0.000009s :     18: substitution.j_node_and_user_rematch
 2.01% :     0.000010s :      7: substitution.minmaximum_grad
 1.27% :     0.000007s :      5: substitution.partial_eliminate
 2.50% :     0.000013s :     18: substitution.remove_not_recompute_node
 4.90% :     0.000025s :     10: substitution.replace_applicator
 1.45% :     0.000007s :      7: substitution.replace_old_param
 0.51% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.20% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.60% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 7.32% :     0.000038s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010067     2
88.94% :     0.008954s :      1: type_inference.infer
11.06% :     0.001113s :      1: type_inference.specialize
------[replace.]   0.000140    19
64.46% :     0.000090s :     12: replace.inline
35.54% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000314    19
94.32% :     0.000297s :     12: match.inline
 5.68% :     0.000018s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000408  2895
 1.64% :     0.000007s :     48: predicate.accumulaten_eliminater
 0.59% :     0.000002s :      6: predicate.ad_related_special_op_eliminate
 1.41% :     0.000006s :     48: predicate.addn_check_dump
 1.64% :     0.000007s :     48: predicate.addn_zero_filter
 1.94% :     0.000008s :     48: predicate.arithmetic_simplify
 1.54% :     0.000006s :     48: predicate.cast_eliminate
 0.25% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.43% :     0.000006s :     48: predicate.compare_switch_simplify
 1.67% :     0.000007s :     48: predicate.depend_value_elim
 1.47% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.51% :     0.000006s :     48: predicate.dict_get_item_eliminator
 1.45% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.43% :     0.000002s :      6: predicate.dumpgradient_eliminate
 0.16% :     0.000001s :      6: predicate.elim_not_effective
 0.32% :     0.000001s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.55% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.45% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.51% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.56% :     0.000006s :     48: predicate.environ_get_eliminate
 1.42% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.12% :     0.000000s :      6: predicate.fold_const_symbol
 0.94% :     0.000004s :     23: predicate.get_grad_eliminate
 0.14% :     0.000001s :      6: predicate.graph_param_transform
 4.28% :     0.000017s :     79: predicate.inline
 1.86% :     0.000008s :     44: predicate.inline_without_move
 0.39% :     0.000002s :     23: predicate.j_node_and_user_rematch
 1.03% :     0.000004s :     23: predicate.less_batch_normalization
 1.93% :     0.000008s :     55: predicate.list_to_tuple_eliminator_
 2.07% :     0.000008s :     61: predicate.load_eliminater
 0.52% :     0.000002s :      6: predicate.loop_unroll_after_grad
 3.05% :     0.000012s :     90: predicate.loop_unroll_before_grad
 1.77% :     0.000007s :     54: predicate.make_slice_get_slice_eliminator
 1.43% :     0.000006s :     48: predicate.merge_addn
 1.49% :     0.000006s :     48: predicate.minmaximum_grad
 0.62% :     0.000003s :      6: predicate.mutable_eliminate
 0.26% :     0.000001s :      6: predicate.opt_reshape
 2.35% :     0.000010s :     61: predicate.partial_eliminate
 1.54% :     0.000006s :     48: predicate.print_const_string_wrapper
 1.89% :     0.000008s :     48: predicate.reduce_eliminate
 1.78% :     0.000007s :     55: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.41% :     0.000010s :    108: predicate.replace_applicator
 0.85% :     0.000003s :     44: predicate.replace_old_param
 0.14% :     0.000001s :      6: predicate.reset_defer_inline
 1.62% :     0.000007s :     48: predicate.reshape_eliminate
 1.50% :     0.000006s :     48: predicate.row_tensor_add_zeros_like
 0.39% :     0.000002s :      6: predicate.row_tensor_eliminate
 1.45% :     0.000006s :     48: predicate.same_eliminate
 0.51% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.45% :     0.000002s :     12: predicate.special_op_eliminate
 0.90% :     0.000004s :     23: predicate.specialize_transform
 1.69% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000006s :     48: predicate.stack_unstack_eliminate
 0.23% :     0.000001s :      6: predicate.switch_call_monad_eliminater
 2.61% :     0.000011s :     67: predicate.switch_defer_inline
 2.40% :     0.000010s :     67: predicate.switch_layer_defer_inline
 6.03% :     0.000025s :    163: predicate.switch_simplify
 1.51% :     0.000006s :     48: predicate.tile_eliminate
 1.56% :     0.000006s :     48: predicate.transpose_eliminate
 1.93% :     0.000008s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.69% :     0.000007s :     48: predicate.tuple_list_get_item_depend_reorder
 3.18% :     0.000013s :     67: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000007s :     48: predicate.tuple_list_set_item_eliminator
 1.73% :     0.000007s :     55: predicate.tuple_to_list_eliminator_
 1.95% :     0.000008s :     61: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000012s :     84: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000007s :     48: predicate.value_based_eliminate
 0.20% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.30% :     0.000001s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001503    29
56.48% :     0.000849s :     13: func_graph_cloner_run.FuncGraphClonerGraph
43.52% :     0.000654s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.040286    91
 0.14% :     0.000057s :      1: add_recomputation
 0.20% :     0.000079s :      1: auto_monad
 0.06% :     0.000023s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.50% :     0.000203s :      1: bootstrap
 0.07% :     0.000028s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000036s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.13% :     0.000051s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
29.57% :     0.011911s :      1: jit_opt_a
 0.49% :     0.000196s :      1: jit_opt_after_cconv
 0.17% :     0.000070s :      1: jit_opt_b
 1.09% :     0.000438s :      1: loop_unroll
 1.72% :     0.000692s :      1: mutable_eliminate
 7.00% :     0.002820s :     39: opt.transform.jit_opt_a
 0.19% :     0.000075s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000033s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000052s :      4: opt.transform.symbol_engine_opt
 1.85% :     0.000747s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000027s :      1: py_interpret_to_execute
 0.04% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000025s :      1: remove_dup_value
 4.65% :     0.001874s :      2: renormalize.infer
 3.83% :     0.001541s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000050s :      1: rewriter_after_opt_a
 0.22% :     0.000087s :      1: rewriter_before_opt_a
 0.25% :     0.000099s :      1: symbol_engine_optimizer
21.57% :     0.008690s :      1: task_emit
25.17% :     0.010141s :      1: type_inference
 0.18% :     0.000072s :      1: validate


TotalTime = 0.0137528, [30]
[bootstrap]: 0.00029126
[type_inference]: 0.00645529
[event_method]: 1.632e-05
[auto_monad]: 6.302e-05
[graph_reusing]: 6.25002e-06
[pre_auto_parallel]: 2.49999e-06
[py_interpret_to_execute]: 2.772e-05
[rewriter_before_opt_a]: 6.949e-05
[expand_dump_flag]: 3.13e-06
[jit_opt_a]: 0.00472673, [2]
    [Cycle 1]: 0.00132439, [27]
        [switch_simplify]: 5.17e-05
        [loop_unroll]: 2.83e-05
        [a_1]: 0.00048938
        [with_stream_mark]: 1.387e-05
        [recompute_prepare]: 6.61e-06
        [updatestate_depend_eliminate]: 3.62002e-06
        [updatestate_assign_eliminate]: 2.86999e-06
        [updatestate_loads_eliminate]: 2.83e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 5.97001e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 5.83002e-06
        [meta_shard_fg_expand]: 1.81e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 3.53999e-06
        [cell_reuse_recompute_pass]: 1.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.373e-05
        [j_node_and_user_rematch]: 8.47e-06
        [meta_fg_expand]: 2.12001e-06
        [replace_old_param]: 8.95999e-06
        [inline_without_move]: 4.99e-06
        [renormalize]: 0.0004362
        [add_forward_monad_depend]: 4.67e-06
        [auto_monad_grad]: 1.86e-06
        [auto_monad_eliminator]: 1.235e-05
        [cse]: 2.738e-05
        [replace_applicator]: 1.046e-05
    [Cycle 2]: 0.00030877, [27]
        [switch_simplify]: 5.35999e-06
        [loop_unroll]: 4.74998e-06
        [a_1]: 8.742e-05
        [with_stream_mark]: 7.90998e-06
        [recompute_prepare]: 4.79002e-06
        [updatestate_depend_eliminate]: 2.29999e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 1.17999e-06
        [specialize_transform]: 4.76002e-06
        [updatestate_useless_node_eliminater]: 4.99e-06
        [accelerated_algorithm]: 4.68999e-06
        [meta_shard_fg_expand]: 1.02e-06
        [get_grad_eliminate_]: 4.18999e-06
        [merge_forward]: 2.29001e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.197e-05
        [j_node_and_user_rematch]: 7.78999e-06
        [meta_fg_expand]: 1.55999e-06
        [replace_old_param]: 1.084e-05
        [inline_without_move]: 4.84e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 4.70001e-06
        [cse]: 9.61e-06
        [replace_applicator]: 4.94e-06
[py_interpret_to_execute_after_opt_a]: 7.66999e-06
[rewriter_after_opt_a]: 3.186e-05
[convert_after_rewriter]: 7.06001e-06
[order_py_execute_after_rewriter]: 5.02e-06
[mutable_eliminate]: 0.0004903
[jit_opt_b]: 4.614e-05, [1]
    [Cycle 1]: 3.962e-05, [2]
        [frontend_op_eliminate]: 1.532e-05
        [inline_after_opt_a]: 1.365e-05
[cconv]: 2.36e-05
[loop_unroll]: 0.00045511
[jit_opt_after_cconv]: 0.00014684, [1]
    [Cycle 1]: 0.00014079, [11]
        [c_1]: 1.958e-05
        [parameter_eliminate]: 2.29001e-06
        [updatestate_depend_eliminate]: 5.09e-06
        [updatestate_assign_eliminate]: 2.54999e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.607e-05
        [call_graph_tuple_transform]: 1.756e-05
        [tuple_list_get_item_eliminator]: 5.24e-06
        [none_parameter_eliminate]: 1.67999e-06
        [renormalize]: 4.40021e-07
        [switch_simplify]: 5.10001e-06
[remove_dup_value]: 1.413e-05
[partial_unused_args_eliminate]: 2.61e-06
[environ_conv]: 5.03002e-06
[add_recomputation]: 5.309e-05
[cse_after_recomputation]: 1.913e-05, [1]
    [Cycle 1]: 1.403e-05, [1]
        [cse]: 8.47e-06
[auto_monad_reorder]: 1.856e-05
[get_jit_bprop_graph]: 1.48002e-06
[rewriter_after_jit_bprop_graph]: 4.18999e-06
[opt_after_jit_grad]: 0.00048129
[symbol_engine_optimizer]: 7.583e-05, [1]
    [Cycle 1]: 6.96e-05, [6]
        [build]: 3.28998e-06
        [elim_shapecalc]: 7.23999e-06
        [elim_not_effective]: 1.208e-05
        [opt_reshape]: 5.75001e-06
        [fold_const_symbol]: 9.04e-06
        [renormalize]: 4.90021e-07
[validate]: 2.944e-05
Sums
    bootstrap                            :     0.000291s :  2.89%
    type_inference                       :     0.006455s : 64.05%
    event_method                         :     0.000016s :  0.16%
    auto_monad                           :     0.000063s :  0.63%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.28%
    rewriter_before_opt_a                :     0.000069s :  0.69%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000057s :  0.57%
    jit_opt_a.loop_unroll                :     0.000033s :  0.33%
    jit_opt_a.a_1                        :     0.000577s :  5.72%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.09%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000020s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000436s :  4.33%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.17%
    jit_opt_a.cse                        :     0.000037s :  0.37%
    jit_opt_a.replace_applicator         :     0.000015s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.08%
    rewriter_after_opt_a                 :     0.000032s :  0.32%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000490s :  4.86%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.14%
    cconv                                :     0.000024s :  0.23%
    loop_unroll                          :     0.000455s :  4.52%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.17%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.14%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000053s :  0.53%
    cse_after_recomputation.cse          :     0.000008s :  0.08%
    auto_monad_reorder                   :     0.000019s :  0.18%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.04%
    opt_after_jit_grad                   :     0.000481s :  4.78%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.12%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.29%


Time group info:
------[substitution.]   0.000176    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000001s :      2: substitution.fold_const_symbol
 3.02% :     0.000005s :      3: substitution.graph_param_transform
80.26% :     0.000141s :      4: substitution.inline
 1.58% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.39% :     0.000004s :      4: substitution.remove_not_recompute_node
 4.42% :     0.000008s :      2: substitution.replace_old_param
 6.56% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.006393     2
88.36% :     0.005649s :      1: type_inference.infer
11.64% :     0.000744s :      1: type_inference.specialize
------[replace.]   0.000063     6
68.10% :     0.000043s :      4: replace.inline
31.90% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000149     6
93.06% :     0.000139s :      4: match.inline
 6.94% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 1.97% :     0.000002s :     10: predicate.arithmetic_simplify
 1.16% :     0.000001s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000001s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.57% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.74% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.50% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.27% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.89% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.51% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.13% :     0.000002s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.66% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.52% :     0.000002s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.55% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.39% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.67% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.12% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.27% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.80% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000520    12
54.10% :     0.000281s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.92% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.99% :     0.000213s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015027    72
 0.37% :     0.000056s :      1: add_recomputation
 0.44% :     0.000066s :      1: auto_monad
 0.14% :     0.000021s :      1: auto_monad_reorder
 2.08% :     0.000312s :      1: bootstrap
 0.18% :     0.000026s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000021s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.14% :     0.000021s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
31.47% :     0.004729s :      1: jit_opt_a
 1.00% :     0.000150s :      1: jit_opt_after_cconv
 0.32% :     0.000049s :      1: jit_opt_b
 3.08% :     0.000463s :      1: loop_unroll
 3.31% :     0.000498s :      1: mutable_eliminate
 5.19% :     0.000780s :     26: opt.transform.jit_opt_a
 0.29% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000023s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.31% :     0.000497s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.20% :     0.000030s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 1.41% :     0.000212s :      1: renormalize.infer
 1.44% :     0.000217s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000035s :      1: rewriter_after_opt_a
 0.48% :     0.000072s :      1: rewriter_before_opt_a
 0.52% :     0.000078s :      1: symbol_engine_optimizer
43.04% :     0.006468s :      1: type_inference

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.100.490 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda84a8a60> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.101.148 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0219435, [33]
[bootstrap]: 0.00019299
[type_inference]: 0.0102085
[event_method]: 4.557e-05
[auto_monad]: 9.081e-05
[graph_reusing]: 5.72001e-06
[pre_auto_parallel]: 2.46998e-06
[py_interpret_to_execute]: 2.238e-05
[rewriter_before_opt_a]: 8.638e-05
[expand_dump_flag]: 3.02002e-06
[jit_opt_a]: 0.00913224, [2]
    [Cycle 1]: 0.00511982, [27]
        [switch_simplify]: 6.921e-05
        [loop_unroll]: 3.872e-05
        [a_1]: 0.00090856
        [with_stream_mark]: 3.074e-05
        [recompute_prepare]: 2.354e-05
        [updatestate_depend_eliminate]: 8.84e-06
        [updatestate_assign_eliminate]: 7.36001e-06
        [updatestate_loads_eliminate]: 6.36e-06
        [parameter_eliminate]: 3.17002e-06
        [specialize_transform]: 1.447e-05
        [updatestate_useless_node_eliminater]: 1.249e-05
        [accelerated_algorithm]: 1.273e-05
        [meta_shard_fg_expand]: 3.46001e-06
        [get_grad_eliminate_]: 1.238e-05
        [merge_forward]: 9.34998e-06
        [cell_reuse_recompute_pass]: 8.70001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.808e-05
        [j_node_and_user_rematch]: 2.392e-05
        [meta_fg_expand]: 0.00177052
        [replace_old_param]: 6.133e-05
        [inline_without_move]: 5.984e-05
        [renormalize]: 0.00172285
        [add_forward_monad_depend]: 7.04001e-06
        [auto_monad_grad]: 2.37999e-06
        [auto_monad_eliminator]: 1.307e-05
        [cse]: 1.951e-05
        [replace_applicator]: 1.457e-05
    [Cycle 2]: 0.00029165, [27]
        [switch_simplify]: 4.72e-06
        [loop_unroll]: 4.03001e-06
        [a_1]: 4.31e-05
        [with_stream_mark]: 1.014e-05
        [recompute_prepare]: 3.65998e-06
        [updatestate_depend_eliminate]: 2.55002e-06
        [updatestate_assign_eliminate]: 2.65002e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 1.44e-06
        [specialize_transform]: 3.89002e-06
        [updatestate_useless_node_eliminater]: 3.33e-06
        [accelerated_algorithm]: 3.58999e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 3.11999e-06
        [merge_forward]: 2.83e-06
        [cell_reuse_recompute_pass]: 2.89001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.263e-05
        [j_node_and_user_rematch]: 5.46e-06
        [meta_fg_expand]: 3.66e-05
        [replace_old_param]: 4.05e-06
        [inline_without_move]: 3.46001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.71e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 4.12e-06
        [cse]: 9.36e-06
        [replace_applicator]: 3.8e-06
[py_interpret_to_execute_after_opt_a]: 1.114e-05
[rewriter_after_opt_a]: 3.325e-05
[convert_after_rewriter]: 5.99999e-06
[order_py_execute_after_rewriter]: 4.70999e-06
[mutable_eliminate]: 0.00065471
[jit_opt_b]: 5.102e-05, [1]
    [Cycle 1]: 4.392e-05, [2]
        [frontend_op_eliminate]: 2.04e-05
        [inline_after_opt_a]: 1.159e-05
[cconv]: 2.281e-05
[loop_unroll]: 0.00042915
[jit_opt_after_cconv]: 0.00011678, [1]
    [Cycle 1]: 0.00011039, [11]
        [c_1]: 1.129e-05
        [parameter_eliminate]: 2.48998e-06
        [updatestate_depend_eliminate]: 5.19998e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [cse]: 1.693e-05
        [call_graph_tuple_transform]: 1.199e-05
        [tuple_list_get_item_eliminator]: 3.73999e-06
        [none_parameter_eliminate]: 1.50999e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 3.88999e-06
[remove_dup_value]: 1.32e-05
[partial_unused_args_eliminate]: 2.48998e-06
[environ_conv]: 4.90999e-06
[add_recomputation]: 3.396e-05
[cse_after_recomputation]: 1.969e-05, [1]
    [Cycle 1]: 1.452e-05, [1]
        [cse]: 8.12e-06
[auto_monad_reorder]: 1.215e-05
[get_jit_bprop_graph]: 1.62999e-06
[rewriter_after_jit_bprop_graph]: 6.51e-06
[opt_after_jit_grad]: 0.00044648
[symbol_engine_optimizer]: 6.241e-05, [1]
    [Cycle 1]: 5.648e-05, [6]
        [build]: 3.35998e-06
        [elim_shapecalc]: 5.63002e-06
        [elim_not_effective]: 9.31e-06
        [opt_reshape]: 3.9e-06
        [fold_const_symbol]: 6.11e-06
        [renormalize]: 4.09986e-07
[validate]: 2.824e-05
[backend_pass]: 1.12e-06
[task_emit]: 2.089e-05
[execute]: 1.27e-06
Sums
    bootstrap                            :     0.000193s :  1.10%
    type_inference                       :     0.010208s : 58.05%
    event_method                         :     0.000046s :  0.26%
    auto_monad                           :     0.000091s :  0.52%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000022s :  0.13%
    rewriter_before_opt_a                :     0.000086s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.42%
    jit_opt_a.loop_unroll                :     0.000043s :  0.24%
    jit_opt_a.a_1                        :     0.000952s :  5.41%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000027s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.03%
    jit_opt_a.specialize_transform       :     0.000018s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.09%
    jit_opt_a.merge_forward              :     0.000012s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000041s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001807s : 10.28%
    jit_opt_a.replace_old_param          :     0.000065s :  0.37%
    jit_opt_a.inline_without_move        :     0.000063s :  0.36%
    jit_opt_a.renormalize                :     0.001723s :  9.80%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.10%
    jit_opt_a.cse                        :     0.000029s :  0.16%
    jit_opt_a.replace_applicator         :     0.000018s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000655s :  3.72%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.07%
    cconv                                :     0.000023s :  0.13%
    loop_unroll                          :     0.000429s :  2.44%
    jit_opt_after_cconv.c_1              :     0.000011s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000012s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000034s :  0.19%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000012s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000446s :  2.54%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.16%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.000021s :  0.12%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000287    50
 0.54% :     0.000002s :      1: substitution.elim_not_effective
 0.33% :     0.000001s :      1: substitution.fold_const_symbol
 1.22% :     0.000003s :      1: substitution.graph_param_transform
65.29% :     0.000188s :      6: substitution.inline
 7.92% :     0.000023s :      2: substitution.inline_without_move
 2.29% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.44% :     0.000004s :      2: substitution.minmaximum_grad
 2.02% :     0.000006s :      3: substitution.partial_eliminate
 2.78% :     0.000008s :      9: substitution.remove_not_recompute_node
 2.30% :     0.000007s :      2: substitution.replace_applicator
 1.72% :     0.000005s :      5: substitution.replace_old_param
 1.63% :     0.000005s :      1: substitution.set_cell_output_no_recompute
 2.55% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.81% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 6.17% :     0.000018s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010141     2
89.79% :     0.009105s :      1: type_inference.infer
10.21% :     0.001036s :      1: type_inference.specialize
------[replace.]   0.000073     8
78.55% :     0.000058s :      6: replace.inline
21.45% :     0.000016s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000195     8
94.34% :     0.000184s :      6: match.inline
 5.66% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000191  1109
 1.50% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.57% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.10% :     0.000002s :     18: predicate.addn_check_dump
 1.32% :     0.000003s :     18: predicate.addn_zero_filter
 2.10% :     0.000004s :     18: predicate.arithmetic_simplify
 1.21% :     0.000002s :     18: predicate.cast_eliminate
 0.10% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.11% :     0.000002s :     18: predicate.compare_switch_simplify
 7.00% :     0.000013s :     18: predicate.depend_value_elim
 1.24% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.37% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.43% :     0.000003s :     18: predicate.dict_set_item_eliminator
 0.53% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      1: predicate.elim_not_effective
 0.18% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.60% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.14% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.10% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.17% :     0.000002s :     18: predicate.environ_get_eliminate
 1.08% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.02% :     0.000002s :     12: predicate.get_grad_eliminate
 0.08% :     0.000000s :      1: predicate.graph_param_transform
 4.05% :     0.000008s :     28: predicate.inline
 3.11% :     0.000006s :     33: predicate.inline_without_move
 0.56% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.42% :     0.000003s :     12: predicate.less_batch_normalization
 1.43% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.43% :     0.000003s :     21: predicate.load_eliminater
 0.93% :     0.000002s :      1: predicate.loop_unroll_after_grad
 3.52% :     0.000007s :     44: predicate.loop_unroll_before_grad
 1.33% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000002s :     18: predicate.merge_addn
 1.16% :     0.000002s :     18: predicate.minmaximum_grad
 0.81% :     0.000002s :      1: predicate.mutable_eliminate
 0.13% :     0.000000s :      1: predicate.opt_reshape
 2.08% :     0.000004s :     21: predicate.partial_eliminate
 1.12% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.40% :     0.000003s :     18: predicate.reduce_eliminate
 1.97% :     0.000004s :     20: predicate.redundant_stop_gradient_eliminater
 0.72% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.27% :     0.000002s :     22: predicate.replace_applicator
 1.55% :     0.000003s :     33: predicate.replace_old_param
 0.14% :     0.000000s :      1: predicate.reset_defer_inline
 1.40% :     0.000003s :     18: predicate.reshape_eliminate
 1.92% :     0.000004s :     18: predicate.row_tensor_add_zeros_like
 0.61% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     18: predicate.same_eliminate
 0.67% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.49% :     0.000001s :      2: predicate.special_op_eliminate
 1.05% :     0.000002s :     12: predicate.specialize_transform
 1.49% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.79% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.11% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.80% :     0.000005s :     26: predicate.switch_defer_inline
 2.03% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.80% :     0.000013s :     71: predicate.switch_simplify
 1.43% :     0.000003s :     18: predicate.tile_eliminate
 1.26% :     0.000002s :     18: predicate.transpose_eliminate
 1.66% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.53% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.89% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.79% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.36% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000003s :     18: predicate.value_based_eliminate
 0.10% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.26% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001149    22
74.33% :     0.000854s :     13: func_graph_cloner_run.FuncGraphClonerGraph
25.67% :     0.000295s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025082    76
 0.15% :     0.000037s :      1: add_recomputation
 0.38% :     0.000095s :      1: auto_monad
 0.06% :     0.000015s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.82% :     0.000205s :      1: bootstrap
 0.10% :     0.000025s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.20% :     0.000050s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
36.42% :     0.009136s :      1: jit_opt_a
 0.48% :     0.000119s :      1: jit_opt_after_cconv
 0.21% :     0.000053s :      1: jit_opt_b
 1.74% :     0.000437s :      1: loop_unroll
 2.65% :     0.000664s :      1: mutable_eliminate
 5.36% :     0.001344s :     26: opt.transform.jit_opt_a
 0.11% :     0.000028s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000026s :      4: opt.transform.jit_opt_b
 0.04% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000022s :      4: opt.transform.symbol_engine_opt
 1.81% :     0.000454s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.10% :     0.000025s :      1: py_interpret_to_execute
 0.05% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000016s :      1: remove_dup_value
 5.30% :     0.001330s :      1: renormalize.infer
 1.53% :     0.000383s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000036s :      1: rewriter_after_opt_a
 0.36% :     0.000089s :      1: rewriter_before_opt_a
 0.26% :     0.000065s :      1: symbol_engine_optimizer
 0.10% :     0.000024s :      1: task_emit
40.75% :     0.010221s :      1: type_inference
 0.18% :     0.000045s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.110.833 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c3714c0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.111.473 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0171057, [33]
[bootstrap]: 0.00024903
[type_inference]: 0.00408267
[event_method]: 1.564e-05
[auto_monad]: 5.851e-05
[graph_reusing]: 5.52999e-06
[pre_auto_parallel]: 2.30002e-06
[py_interpret_to_execute]: 3.249e-05
[rewriter_before_opt_a]: 6.526e-05
[expand_dump_flag]: 3.40998e-06
[jit_opt_a]: 0.00483487, [2]
    [Cycle 1]: 0.00137744, [27]
        [switch_simplify]: 4.575e-05
        [loop_unroll]: 2.879e-05
        [a_1]: 0.00048289
        [with_stream_mark]: 1.098e-05
        [recompute_prepare]: 6.26e-06
        [updatestate_depend_eliminate]: 2.49999e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.76e-06
        [parameter_eliminate]: 8.10018e-07
        [specialize_transform]: 5.86e-06
        [updatestate_useless_node_eliminater]: 4.90001e-06
        [accelerated_algorithm]: 5.21998e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 5.38002e-06
        [merge_forward]: 2.21998e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.049e-05
        [j_node_and_user_rematch]: 8.55001e-06
        [meta_fg_expand]: 1.68002e-06
        [replace_old_param]: 7.93999e-06
        [inline_without_move]: 4.90001e-06
        [renormalize]: 0.00053685
        [add_forward_monad_depend]: 4.23001e-06
        [auto_monad_grad]: 1.07998e-06
        [auto_monad_eliminator]: 9.19e-06
        [cse]: 1.562e-05
        [replace_applicator]: 1.16e-05
    [Cycle 2]: 0.00030738, [27]
        [switch_simplify]: 5.30999e-06
        [loop_unroll]: 4.74002e-06
        [a_1]: 8.783e-05
        [with_stream_mark]: 7.8e-06
        [recompute_prepare]: 4.73001e-06
        [updatestate_depend_eliminate]: 2.26e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.93002e-06
        [parameter_eliminate]: 1.06002e-06
        [specialize_transform]: 4.77e-06
        [updatestate_useless_node_eliminater]: 4.61002e-06
        [accelerated_algorithm]: 4.98001e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 4.45e-06
        [merge_forward]: 2.36998e-06
        [cell_reuse_recompute_pass]: 2.09999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.13e-05
        [j_node_and_user_rematch]: 7.40998e-06
        [meta_fg_expand]: 1.55001e-06
        [replace_old_param]: 6.50997e-06
        [inline_without_move]: 4.55999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 9.30013e-07
        [auto_monad_eliminator]: 4.82e-06
        [cse]: 9.99999e-06
        [replace_applicator]: 4.71002e-06
[py_interpret_to_execute_after_opt_a]: 7.88001e-06
[rewriter_after_opt_a]: 2.72e-05
[convert_after_rewriter]: 6.17999e-06
[order_py_execute_after_rewriter]: 3.71999e-06
[mutable_eliminate]: 0.00051427
[jit_opt_b]: 4.53e-05, [1]
    [Cycle 1]: 3.974e-05, [2]
        [frontend_op_eliminate]: 1.478e-05
        [inline_after_opt_a]: 1.422e-05
[cconv]: 1.638e-05
[loop_unroll]: 0.00040303
[jit_opt_after_cconv]: 0.00012755, [1]
    [Cycle 1]: 0.00012198, [11]
        [c_1]: 1.915e-05
        [parameter_eliminate]: 2.68e-06
        [updatestate_depend_eliminate]: 4.45e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.01998e-06
        [cse]: 1.68e-05
        [call_graph_tuple_transform]: 1.478e-05
        [tuple_list_get_item_eliminator]: 4.97e-06
        [none_parameter_eliminate]: 1.24003e-06
        [renormalize]: 4.10015e-07
        [switch_simplify]: 5.08002e-06
[remove_dup_value]: 6.79001e-06
[partial_unused_args_eliminate]: 1.44e-06
[environ_conv]: 3.47002e-06
[add_recomputation]: 3.207e-05
[cse_after_recomputation]: 1.897e-05, [1]
    [Cycle 1]: 1.409e-05, [1]
        [cse]: 8.3e-06
[auto_monad_reorder]: 9.29e-06
[get_jit_bprop_graph]: 1.32999e-06
[rewriter_after_jit_bprop_graph]: 5.05001e-06
[opt_after_jit_grad]: 0.00045405
[symbol_engine_optimizer]: 6.959e-05, [1]
    [Cycle 1]: 6.376e-05, [6]
        [build]: 3.52997e-06
        [elim_shapecalc]: 7.28e-06
        [elim_not_effective]: 1.188e-05
        [opt_reshape]: 5.16998e-06
        [fold_const_symbol]: 8.60001e-06
        [renormalize]: 4.89992e-07
[validate]: 2.819e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.0057973
[execute]: 7.03e-06
Sums
    bootstrap                            :     0.000249s :  1.86%
    type_inference                       :     0.004083s : 30.46%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000059s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.24%
    rewriter_before_opt_a                :     0.000065s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000051s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000571s :  4.26%
    jit_opt_a.with_stream_mark           :     0.000019s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000014s :  0.11%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000537s :  4.01%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.10%
    jit_opt_a.cse                        :     0.000026s :  0.19%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000027s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000514s :  3.84%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000016s :  0.12%
    loop_unroll                          :     0.000403s :  3.01%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.11%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000007s :  0.05%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000003s :  0.03%
    add_recomputation                    :     0.000032s :  0.24%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000009s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000454s :  3.39%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005797s : 43.25%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000158    23
 0.87% :     0.000001s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.24% :     0.000004s :      3: substitution.graph_param_transform
85.39% :     0.000135s :      4: substitution.inline
 1.56% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.49% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.87% :     0.000003s :      2: substitution.replace_old_param
 4.91% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004026     2
80.38% :     0.003236s :      1: type_inference.infer
19.62% :     0.000790s :      1: type_inference.specialize
------[replace.]   0.000061     6
66.01% :     0.000040s :      4: replace.inline
33.99% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000139     6
95.19% :     0.000132s :      4: match.inline
 4.81% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.37% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.24% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.38% :     0.000002s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.67% :     0.000002s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000001s :     10: predicate.depend_value_elim
 1.36% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.30% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.06% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.02% :     0.000001s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.95% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.24% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.66% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000000s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.24% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.63% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.96% :     0.000002s :     18: predicate.replace_applicator
 0.53% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.44% :     0.000008s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.29% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.50% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.82% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.79% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000502    12
46.80% :     0.000235s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.55% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.65% :     0.000239s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018500    76
 0.19% :     0.000035s :      1: add_recomputation
 0.33% :     0.000062s :      1: auto_monad
 0.06% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.41% :     0.000261s :      1: bootstrap
 0.10% :     0.000019s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.15% :     0.004838s :      1: jit_opt_a
 0.70% :     0.000130s :      1: jit_opt_after_cconv
 0.26% :     0.000048s :      1: jit_opt_b
 2.21% :     0.000410s :      1: loop_unroll
 2.82% :     0.000521s :      1: mutable_eliminate
 4.11% :     0.000761s :     26: opt.transform.jit_opt_a
 0.22% :     0.000041s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000018s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.49% :     0.000461s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000036s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000009s :      1: remove_dup_value
 1.46% :     0.000270s :      1: renormalize.infer
 1.41% :     0.000262s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000030s :      1: rewriter_after_opt_a
 0.37% :     0.000068s :      1: rewriter_before_opt_a
 0.39% :     0.000072s :      1: symbol_engine_optimizer
31.40% :     0.005809s :      1: task_emit
22.13% :     0.004093s :      1: type_inference
 0.24% :     0.000045s :      1: validate


TotalTime = 0.0331221, [33]
[bootstrap]: 0.00021895
[type_inference]: 0.0105362
[event_method]: 4.811e-05
[auto_monad]: 9.386e-05
[graph_reusing]: 7.29001e-06
[pre_auto_parallel]: 2.98e-06
[py_interpret_to_execute]: 2.572e-05
[rewriter_before_opt_a]: 9.116e-05
[expand_dump_flag]: 3.42002e-06
[jit_opt_a]: 0.012197, [3]
    [Cycle 1]: 0.00593659, [27]
        [switch_simplify]: 6.942e-05
        [loop_unroll]: 3.833e-05
        [a_1]: 0.00085304
        [with_stream_mark]: 2.591e-05
        [recompute_prepare]: 1.872e-05
        [updatestate_depend_eliminate]: 8.39002e-06
        [updatestate_assign_eliminate]: 6.71999e-06
        [updatestate_loads_eliminate]: 6.28e-06
        [parameter_eliminate]: 2.93998e-06
        [specialize_transform]: 1.275e-05
        [updatestate_useless_node_eliminater]: 1.214e-05
        [accelerated_algorithm]: 1.241e-05
        [meta_shard_fg_expand]: 3.85e-06
        [get_grad_eliminate_]: 1.197e-05
        [merge_forward]: 8.3e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.66e-05
        [j_node_and_user_rematch]: 2.333e-05
        [meta_fg_expand]: 0.00152025
        [replace_old_param]: 5.336e-05
        [inline_without_move]: 5.332e-05
        [renormalize]: 0.00265711
        [add_forward_monad_depend]: 9.92001e-06
        [auto_monad_grad]: 5.34e-06
        [auto_monad_eliminator]: 5.474e-05
        [cse]: 0.00016094
        [replace_applicator]: 7.114e-05
    [Cycle 2]: 0.0023122, [27]
        [switch_simplify]: 3.704e-05
        [loop_unroll]: 3.533e-05
        [a_1]: 0.00112835
        [with_stream_mark]: 1.625e-05
        [recompute_prepare]: 9.92999e-06
        [updatestate_depend_eliminate]: 5.24003e-06
        [updatestate_assign_eliminate]: 4.15e-06
        [updatestate_loads_eliminate]: 3.83001e-06
        [parameter_eliminate]: 1.23002e-06
        [specialize_transform]: 9.49999e-06
        [updatestate_useless_node_eliminater]: 8.23001e-06
        [accelerated_algorithm]: 8.25e-06
        [meta_shard_fg_expand]: 2.71e-06
        [get_grad_eliminate_]: 8.77999e-06
        [merge_forward]: 4.97999e-06
        [cell_reuse_recompute_pass]: 1.67999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.956e-05
        [j_node_and_user_rematch]: 1.373e-05
        [meta_fg_expand]: 4.805e-05
        [replace_old_param]: 1.238e-05
        [inline_without_move]: 8.37998e-06
        [renormalize]: 0.00068554
        [add_forward_monad_depend]: 4.51002e-06
        [auto_monad_grad]: 2.03002e-06
        [auto_monad_eliminator]: 1.589e-05
        [cse]: 4.795e-05
        [replace_applicator]: 1.573e-05
    [Cycle 3]: 0.00049109, [27]
        [switch_simplify]: 8.70001e-06
        [loop_unroll]: 7.9e-06
        [a_1]: 0.00019213
        [with_stream_mark]: 1.124e-05
        [recompute_prepare]: 7.98001e-06
        [updatestate_depend_eliminate]: 4.67998e-06
        [updatestate_assign_eliminate]: 4.27e-06
        [updatestate_loads_eliminate]: 3.76001e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 8.33001e-06
        [updatestate_useless_node_eliminater]: 7.93001e-06
        [accelerated_algorithm]: 9.10999e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 7.76001e-06
        [merge_forward]: 4.35e-06
        [cell_reuse_recompute_pass]: 1.52001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.656e-05
        [j_node_and_user_rematch]: 1.372e-05
        [meta_fg_expand]: 2.68e-06
        [replace_old_param]: 9.91e-06
        [inline_without_move]: 7.83999e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 1.04e-06
        [auto_monad_eliminator]: 1.014e-05
        [cse]: 2.362e-05
        [replace_applicator]: 8.07e-06
[py_interpret_to_execute_after_opt_a]: 1.225e-05
[rewriter_after_opt_a]: 4.212e-05
[convert_after_rewriter]: 7.66999e-06
[order_py_execute_after_rewriter]: 5.32999e-06
[mutable_eliminate]: 0.0006478
[jit_opt_b]: 6.704e-05, [1]
    [Cycle 1]: 6.002e-05, [2]
        [frontend_op_eliminate]: 2.522e-05
        [inline_after_opt_a]: 2.344e-05
[cconv]: 1.714e-05
[loop_unroll]: 0.00042771
[jit_opt_after_cconv]: 0.00019209, [1]
    [Cycle 1]: 0.00018481, [11]
        [c_1]: 3.741e-05
        [parameter_eliminate]: 2.44001e-06
        [updatestate_depend_eliminate]: 7.82e-06
        [updatestate_assign_eliminate]: 4.60999e-06
        [updatestate_loads_eliminate]: 3.88999e-06
        [cse]: 3.39e-05
        [call_graph_tuple_transform]: 2.539e-05
        [tuple_list_get_item_eliminator]: 8.13999e-06
        [none_parameter_eliminate]: 1.14998e-06
        [renormalize]: 3.9002e-07
        [switch_simplify]: 8.54998e-06
[remove_dup_value]: 1.662e-05
[partial_unused_args_eliminate]: 1.59998e-06
[environ_conv]: 6.11998e-06
[add_recomputation]: 4.474e-05
[cse_after_recomputation]: 3.726e-05, [1]
    [Cycle 1]: 3.086e-05, [1]
        [cse]: 2.421e-05
[auto_monad_reorder]: 1.733e-05
[get_jit_bprop_graph]: 1.97999e-06
[rewriter_after_jit_bprop_graph]: 5.00999e-06
[opt_after_jit_grad]: 0.00047742
[symbol_engine_optimizer]: 9.922e-05, [1]
    [Cycle 1]: 9.254e-05, [6]
        [build]: 5.32001e-06
        [elim_shapecalc]: 1.259e-05
        [elim_not_effective]: 2.147e-05
        [opt_reshape]: 9.71e-06
        [fold_const_symbol]: 1.432e-05
        [renormalize]: 3.49974e-07
[validate]: 4.181e-05
[backend_pass]: 7.09988e-07
[task_emit]: 0.00752492
[execute]: 5.17999e-06
Sums
    bootstrap                            :     0.000219s :  0.76%
    type_inference                       :     0.010536s : 36.50%
    event_method                         :     0.000048s :  0.17%
    auto_monad                           :     0.000094s :  0.33%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.09%
    rewriter_before_opt_a                :     0.000091s :  0.32%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000115s :  0.40%
    jit_opt_a.loop_unroll                :     0.000082s :  0.28%
    jit_opt_a.a_1                        :     0.002174s :  7.53%
    jit_opt_a.with_stream_mark           :     0.000053s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000037s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000018s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000014s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000031s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000028s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000030s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000009s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000029s :  0.10%
    jit_opt_a.merge_forward              :     0.000018s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000063s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000051s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001571s :  5.44%
    jit_opt_a.replace_old_param          :     0.000076s :  0.26%
    jit_opt_a.inline_without_move        :     0.000070s :  0.24%
    jit_opt_a.renormalize                :     0.003343s : 11.58%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000081s :  0.28%
    jit_opt_a.cse                        :     0.000233s :  0.81%
    jit_opt_a.replace_applicator         :     0.000095s :  0.33%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000042s :  0.15%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000648s :  2.24%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000023s :  0.08%
    cconv                                :     0.000017s :  0.06%
    loop_unroll                          :     0.000428s :  1.48%
    jit_opt_after_cconv.c_1              :     0.000037s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000034s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.03%
    remove_dup_value                     :     0.000017s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000045s :  0.15%
    cse_after_recomputation.cse          :     0.000024s :  0.08%
    auto_monad_reorder                   :     0.000017s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000477s :  1.65%
    symbol_engine_optimizer.build        :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000013s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000021s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000010s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000042s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007525s : 26.07%
    execute                              :     0.000005s :  0.02%


Time group info:
------[substitution.]   0.000554   126
 7.08% :     0.000039s :      2: substitution.arithmetic_simplify
 0.65% :     0.000004s :      5: substitution.elim_not_effective
 0.36% :     0.000002s :      5: substitution.fold_const_symbol
 1.04% :     0.000006s :      6: substitution.graph_param_transform
59.45% :     0.000330s :     12: substitution.inline
 3.43% :     0.000019s :      2: substitution.inline_without_move
 1.91% :     0.000011s :     18: substitution.j_node_and_user_rematch
 1.82% :     0.000010s :      7: substitution.minmaximum_grad
 1.35% :     0.000007s :      5: substitution.partial_eliminate
 2.36% :     0.000013s :     18: substitution.remove_not_recompute_node
 4.82% :     0.000027s :     10: substitution.replace_applicator
 1.52% :     0.000008s :      7: substitution.replace_old_param
 0.50% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.84% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.64% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 7.24% :     0.000040s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010462     2
88.91% :     0.009302s :      1: type_inference.infer
11.09% :     0.001161s :      1: type_inference.specialize
------[replace.]   0.000145    19
65.05% :     0.000094s :     12: replace.inline
34.95% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000344    19
93.77% :     0.000323s :     12: match.inline
 6.23% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000410  2895
 1.59% :     0.000007s :     48: predicate.accumulaten_eliminater
 0.58% :     0.000002s :      6: predicate.ad_related_special_op_eliminate
 1.41% :     0.000006s :     48: predicate.addn_check_dump
 1.84% :     0.000008s :     48: predicate.addn_zero_filter
 2.11% :     0.000009s :     48: predicate.arithmetic_simplify
 1.83% :     0.000007s :     48: predicate.cast_eliminate
 0.26% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.39% :     0.000006s :     48: predicate.compare_switch_simplify
 1.61% :     0.000007s :     48: predicate.depend_value_elim
 1.41% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.44% :     0.000006s :     48: predicate.dict_get_item_eliminator
 1.47% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.39% :     0.000002s :      6: predicate.dumpgradient_eliminate
 0.15% :     0.000001s :      6: predicate.elim_not_effective
 0.33% :     0.000001s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.53% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.52% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.66% :     0.000007s :     48: predicate.environ_get_depend_swap
 1.63% :     0.000007s :     48: predicate.environ_get_eliminate
 1.41% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.11% :     0.000000s :      6: predicate.fold_const_symbol
 1.02% :     0.000004s :     23: predicate.get_grad_eliminate
 0.12% :     0.000000s :      6: predicate.graph_param_transform
 4.45% :     0.000018s :     79: predicate.inline
 1.74% :     0.000007s :     44: predicate.inline_without_move
 0.37% :     0.000002s :     23: predicate.j_node_and_user_rematch
 0.98% :     0.000004s :     23: predicate.less_batch_normalization
 1.79% :     0.000007s :     55: predicate.list_to_tuple_eliminator_
 1.92% :     0.000008s :     61: predicate.load_eliminater
 0.50% :     0.000002s :      6: predicate.loop_unroll_after_grad
 2.91% :     0.000012s :     90: predicate.loop_unroll_before_grad
 1.76% :     0.000007s :     54: predicate.make_slice_get_slice_eliminator
 1.42% :     0.000006s :     48: predicate.merge_addn
 1.44% :     0.000006s :     48: predicate.minmaximum_grad
 0.60% :     0.000002s :      6: predicate.mutable_eliminate
 0.28% :     0.000001s :      6: predicate.opt_reshape
 2.36% :     0.000010s :     61: predicate.partial_eliminate
 1.46% :     0.000006s :     48: predicate.print_const_string_wrapper
 1.84% :     0.000008s :     48: predicate.reduce_eliminate
 1.74% :     0.000007s :     55: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.40% :     0.000010s :    108: predicate.replace_applicator
 0.90% :     0.000004s :     44: predicate.replace_old_param
 0.14% :     0.000001s :      6: predicate.reset_defer_inline
 1.71% :     0.000007s :     48: predicate.reshape_eliminate
 1.44% :     0.000006s :     48: predicate.row_tensor_add_zeros_like
 0.34% :     0.000001s :      6: predicate.row_tensor_eliminate
 1.61% :     0.000007s :     48: predicate.same_eliminate
 0.47% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.58% :     0.000002s :     12: predicate.special_op_eliminate
 0.91% :     0.000004s :     23: predicate.specialize_transform
 1.64% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000006s :     48: predicate.stack_unstack_eliminate
 0.22% :     0.000001s :      6: predicate.switch_call_monad_eliminater
 2.60% :     0.000011s :     67: predicate.switch_defer_inline
 2.39% :     0.000010s :     67: predicate.switch_layer_defer_inline
 5.90% :     0.000024s :    163: predicate.switch_simplify
 1.48% :     0.000006s :     48: predicate.tile_eliminate
 1.47% :     0.000006s :     48: predicate.transpose_eliminate
 1.87% :     0.000008s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.68% :     0.000007s :     48: predicate.tuple_list_get_item_depend_reorder
 3.07% :     0.000013s :     67: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000007s :     48: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000007s :     55: predicate.tuple_to_list_eliminator_
 1.94% :     0.000008s :     61: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000012s :     84: predicate.updatestate_useless_node_eliminater
 1.87% :     0.000008s :     48: predicate.value_based_eliminate
 0.22% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.33% :     0.000001s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001457    29
58.08% :     0.000846s :     13: func_graph_cloner_run.FuncGraphClonerGraph
41.92% :     0.000611s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.039478    91
 0.12% :     0.000048s :      1: add_recomputation
 0.25% :     0.000098s :      1: auto_monad
 0.05% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.59% :     0.000231s :      1: bootstrap
 0.05% :     0.000020s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000040s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.13% :     0.000052s :      1: event_method
 0.02% :     0.000009s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
30.90% :     0.012201s :      1: jit_opt_a
 0.49% :     0.000195s :      1: jit_opt_after_cconv
 0.18% :     0.000070s :      1: jit_opt_b
 1.11% :     0.000436s :      1: loop_unroll
 1.66% :     0.000657s :      1: mutable_eliminate
 7.17% :     0.002832s :     39: opt.transform.jit_opt_a
 0.19% :     0.000076s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000043s :      4: opt.transform.jit_opt_b
 0.04% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000054s :      4: opt.transform.symbol_engine_opt
 1.23% :     0.000486s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000028s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 4.66% :     0.001841s :      2: renormalize.infer
 3.77% :     0.001487s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000045s :      1: rewriter_after_opt_a
 0.24% :     0.000094s :      1: rewriter_before_opt_a
 0.26% :     0.000102s :      1: symbol_engine_optimizer
19.09% :     0.007537s :      1: task_emit
26.72% :     0.010550s :      1: type_inference
 0.16% :     0.000063s :      1: validate


TotalTime = 0.0264385, [33]
[bootstrap]: 0.00021569
[type_inference]: 0.0145979
[event_method]: 4.953e-05
[auto_monad]: 9.727e-05
[graph_reusing]: 6.79999e-06
[pre_auto_parallel]: 3.06999e-06
[py_interpret_to_execute]: 3.425e-05
[rewriter_before_opt_a]: 9.451e-05
[expand_dump_flag]: 3.53e-06
[jit_opt_a]: 0.00915567, [2]
    [Cycle 1]: 0.00484716, [27]
        [switch_simplify]: 7.112e-05
        [loop_unroll]: 3.78e-05
        [a_1]: 0.00085066
        [with_stream_mark]: 2.749e-05
        [recompute_prepare]: 1.732e-05
        [updatestate_depend_eliminate]: 8.12e-06
        [updatestate_assign_eliminate]: 5.79999e-06
        [updatestate_loads_eliminate]: 6.12999e-06
        [parameter_eliminate]: 2.99001e-06
        [specialize_transform]: 1.271e-05
        [updatestate_useless_node_eliminater]: 1.144e-05
        [accelerated_algorithm]: 1.201e-05
        [meta_shard_fg_expand]: 2.99999e-06
        [get_grad_eliminate_]: 1.19e-05
        [merge_forward]: 8.53001e-06
        [cell_reuse_recompute_pass]: 9.10019e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.484e-05
        [j_node_and_user_rematch]: 2.219e-05
        [meta_fg_expand]: 0.00162278
        [replace_old_param]: 5.485e-05
        [inline_without_move]: 5.345e-05
        [renormalize]: 0.00171852
        [add_forward_monad_depend]: 1.055e-05
        [auto_monad_grad]: 2.69999e-06
        [auto_monad_eliminator]: 1.274e-05
        [cse]: 2.222e-05
        [replace_applicator]: 1.354e-05
    [Cycle 2]: 0.0002957, [27]
        [switch_simplify]: 4.58001e-06
        [loop_unroll]: 3.8e-06
        [a_1]: 4.363e-05
        [with_stream_mark]: 1.023e-05
        [recompute_prepare]: 3.76001e-06
        [updatestate_depend_eliminate]: 2.63e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 2.22001e-06
        [parameter_eliminate]: 1.65001e-06
        [specialize_transform]: 3.58e-06
        [updatestate_useless_node_eliminater]: 3.26001e-06
        [accelerated_algorithm]: 3.46999e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 3.16001e-06
        [merge_forward]: 2.91e-06
        [cell_reuse_recompute_pass]: 3.03e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.16e-05
        [j_node_and_user_rematch]: 6.02001e-06
        [meta_fg_expand]: 3.882e-05
        [replace_old_param]: 4.27998e-06
        [inline_without_move]: 3.2e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.81003e-06
        [auto_monad_grad]: 1.39e-06
        [auto_monad_eliminator]: 4.95999e-06
        [cse]: 9.42001e-06
        [replace_applicator]: 3.45e-06
[py_interpret_to_execute_after_opt_a]: 1.061e-05
[rewriter_after_opt_a]: 3.368e-05
[convert_after_rewriter]: 6.09999e-06
[order_py_execute_after_rewriter]: 4.48001e-06
[mutable_eliminate]: 0.00064457
[jit_opt_b]: 5.017e-05, [1]
    [Cycle 1]: 4.302e-05, [2]
        [frontend_op_eliminate]: 1.942e-05
        [inline_after_opt_a]: 1.128e-05
[cconv]: 2.446e-05
[loop_unroll]: 0.00042272
[jit_opt_after_cconv]: 0.00014149, [1]
    [Cycle 1]: 0.00013459, [11]
        [c_1]: 1.146e-05
        [parameter_eliminate]: 2.56e-06
        [updatestate_depend_eliminate]: 4.64002e-06
        [updatestate_assign_eliminate]: 1.85001e-06
        [updatestate_loads_eliminate]: 1.60001e-06
        [cse]: 1.684e-05
        [call_graph_tuple_transform]: 3.486e-05
        [tuple_list_get_item_eliminator]: 3.93999e-06
        [none_parameter_eliminate]: 1.77001e-06
        [renormalize]: 9.20001e-07
        [switch_simplify]: 3.61001e-06
[remove_dup_value]: 1.492e-05
[partial_unused_args_eliminate]: 2.32999e-06
[environ_conv]: 6.09001e-06
[add_recomputation]: 3.746e-05
[cse_after_recomputation]: 2.006e-05, [1]
    [Cycle 1]: 1.417e-05, [1]
        [cse]: 8.1e-06
[auto_monad_reorder]: 1.415e-05
[get_jit_bprop_graph]: 1.95001e-06
[rewriter_after_jit_bprop_graph]: 5.52001e-06
[opt_after_jit_grad]: 0.00044395
[symbol_engine_optimizer]: 6.418e-05, [1]
    [Cycle 1]: 5.715e-05, [6]
        [build]: 3.61999e-06
        [elim_shapecalc]: 5.35001e-06
        [elim_not_effective]: 9.77999e-06
        [opt_reshape]: 3.86999e-06
        [fold_const_symbol]: 6.12999e-06
        [renormalize]: 3.9002e-07
[validate]: 3.05e-05
[backend_pass]: 9.79984e-07
[task_emit]: 1.911e-05
[execute]: 1.48002e-06
Sums
    bootstrap                            :     0.000216s :  0.99%
    type_inference                       :     0.014598s : 66.94%
    event_method                         :     0.000050s :  0.23%
    auto_monad                           :     0.000097s :  0.45%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000034s :  0.16%
    rewriter_before_opt_a                :     0.000095s :  0.43%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000076s :  0.35%
    jit_opt_a.loop_unroll                :     0.000042s :  0.19%
    jit_opt_a.a_1                        :     0.000894s :  4.10%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000016s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000015s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.07%
    jit_opt_a.merge_forward              :     0.000011s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.001662s :  7.62%
    jit_opt_a.replace_old_param          :     0.000059s :  0.27%
    jit_opt_a.inline_without_move        :     0.000057s :  0.26%
    jit_opt_a.renormalize                :     0.001719s :  7.88%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.08%
    jit_opt_a.cse                        :     0.000032s :  0.15%
    jit_opt_a.replace_applicator         :     0.000017s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.05%
    rewriter_after_opt_a                 :     0.000034s :  0.15%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000645s :  2.96%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.05%
    cconv                                :     0.000024s :  0.11%
    loop_unroll                          :     0.000423s :  1.94%
    jit_opt_after_cconv.c_1              :     0.000011s :  0.05%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000035s :  0.16%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000015s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000037s :  0.17%
    cse_after_recomputation.cse          :     0.000008s :  0.04%
    auto_monad_reorder                   :     0.000014s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000444s :  2.04%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000005s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000030s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000019s :  0.09%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000289    50
 0.51% :     0.000001s :      1: substitution.elim_not_effective
 0.31% :     0.000001s :      1: substitution.fold_const_symbol
 8.49% :     0.000025s :      1: substitution.graph_param_transform
63.13% :     0.000183s :      6: substitution.inline
 6.59% :     0.000019s :      2: substitution.inline_without_move
 2.28% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.25% :     0.000004s :      2: substitution.minmaximum_grad
 1.37% :     0.000004s :      3: substitution.partial_eliminate
 2.39% :     0.000007s :      9: substitution.remove_not_recompute_node
 1.91% :     0.000006s :      2: substitution.replace_applicator
 1.70% :     0.000005s :      5: substitution.replace_old_param
 0.78% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.37% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.51% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 5.41% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.014514     2
64.27% :     0.009328s :      1: type_inference.infer
35.73% :     0.005186s :      1: type_inference.specialize
------[replace.]   0.000064     8
78.55% :     0.000050s :      6: replace.inline
21.45% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000190     8
94.33% :     0.000179s :      6: match.inline
 5.67% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000170  1109
 1.53% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.73% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.26% :     0.000002s :     18: predicate.addn_check_dump
 1.60% :     0.000003s :     18: predicate.addn_zero_filter
 2.46% :     0.000004s :     18: predicate.arithmetic_simplify
 1.58% :     0.000003s :     18: predicate.cast_eliminate
 0.09% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.31% :     0.000002s :     18: predicate.compare_switch_simplify
 1.49% :     0.000003s :     18: predicate.depend_value_elim
 1.27% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.53% :     0.000003s :     18: predicate.dict_set_item_eliminator
 0.58% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      1: predicate.elim_not_effective
 0.21% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.30% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.38% :     0.000002s :     18: predicate.environ_get_eliminate
 1.25% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.26% :     0.000002s :     12: predicate.get_grad_eliminate
 0.08% :     0.000000s :      1: predicate.graph_param_transform
 4.51% :     0.000008s :     28: predicate.inline
 3.16% :     0.000005s :     33: predicate.inline_without_move
 0.51% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.27% :     0.000002s :     12: predicate.less_batch_normalization
 1.61% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.70% :     0.000003s :     21: predicate.load_eliminater
 0.77% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.55% :     0.000006s :     44: predicate.loop_unroll_before_grad
 2.01% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000002s :     18: predicate.merge_addn
 1.31% :     0.000002s :     18: predicate.minmaximum_grad
 1.36% :     0.000002s :      1: predicate.mutable_eliminate
 0.15% :     0.000000s :      1: predicate.opt_reshape
 2.17% :     0.000004s :     21: predicate.partial_eliminate
 1.37% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.68% :     0.000003s :     18: predicate.reduce_eliminate
 1.60% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.67% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.45% :     0.000002s :     22: predicate.replace_applicator
 1.80% :     0.000003s :     33: predicate.replace_old_param
 0.15% :     0.000000s :      1: predicate.reset_defer_inline
 1.58% :     0.000003s :     18: predicate.reshape_eliminate
 1.38% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.57% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.54% :     0.000003s :     18: predicate.same_eliminate
 0.62% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.46% :     0.000001s :      2: predicate.special_op_eliminate
 1.14% :     0.000002s :     12: predicate.specialize_transform
 1.54% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.54% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.11% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.55% :     0.000004s :     26: predicate.switch_defer_inline
 2.43% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.59% :     0.000011s :     71: predicate.switch_simplify
 1.47% :     0.000002s :     18: predicate.tile_eliminate
 1.32% :     0.000002s :     18: predicate.transpose_eliminate
 1.54% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.47% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 2.87% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.73% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.56% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.86% :     0.000003s :     18: predicate.value_based_eliminate
 0.13% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.19% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001214    22
70.70% :     0.000858s :     13: func_graph_cloner_run.FuncGraphClonerGraph
29.30% :     0.000356s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.029509    76
 0.14% :     0.000040s :      1: add_recomputation
 0.34% :     0.000102s :      1: auto_monad
 0.06% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.80% :     0.000235s :      1: bootstrap
 0.09% :     0.000028s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.07% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.18% :     0.000054s :      1: event_method
 0.01% :     0.000004s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
31.04% :     0.009159s :      1: jit_opt_a
 0.49% :     0.000144s :      1: jit_opt_after_cconv
 0.18% :     0.000053s :      1: jit_opt_b
 1.46% :     0.000431s :      1: loop_unroll
 2.22% :     0.000654s :      1: mutable_eliminate
 4.27% :     0.001259s :     26: opt.transform.jit_opt_a
 0.17% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000024s :      4: opt.transform.jit_opt_b
 0.03% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000022s :      4: opt.transform.symbol_engine_opt
 1.53% :     0.000452s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000037s :      1: py_interpret_to_execute
 0.04% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 4.45% :     0.001312s :      1: renormalize.infer
 1.35% :     0.000397s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000037s :      1: rewriter_after_opt_a
 0.33% :     0.000098s :      1: rewriter_before_opt_a
 0.23% :     0.000067s :      1: symbol_engine_optimizer
 0.08% :     0.000023s :      1: task_emit
49.51% :     0.014610s :      1: type_inference
 0.16% :     0.000047s :      1: validate


TotalTime = 0.0116942, [30]
[bootstrap]: 0.00029021
[type_inference]: 0.00423071
[event_method]: 1.58e-05
[auto_monad]: 6.268e-05
[graph_reusing]: 5.92001e-06
[pre_auto_parallel]: 2.60002e-06
[py_interpret_to_execute]: 2.908e-05
[rewriter_before_opt_a]: 7.064e-05
[expand_dump_flag]: 2.99999e-06
[jit_opt_a]: 0.00498472, [2]
    [Cycle 1]: 0.00140574, [27]
        [switch_simplify]: 5.27e-05
        [loop_unroll]: 2.785e-05
        [a_1]: 0.00050324
        [with_stream_mark]: 1.74e-05
        [recompute_prepare]: 6.76999e-06
        [updatestate_depend_eliminate]: 3.23e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.71999e-06
        [parameter_eliminate]: 2.12999e-06
        [specialize_transform]: 5.71003e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 5.27001e-06
        [meta_shard_fg_expand]: 2.29001e-06
        [get_grad_eliminate_]: 5.05999e-06
        [merge_forward]: 3.52997e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.353e-05
        [j_node_and_user_rematch]: 8.99e-06
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 8.89998e-06
        [inline_without_move]: 5.24e-06
        [renormalize]: 0.00048969
        [add_forward_monad_depend]: 5.34998e-06
        [auto_monad_grad]: 2.14e-06
        [auto_monad_eliminator]: 1.258e-05
        [cse]: 2.745e-05
        [replace_applicator]: 1.058e-05
    [Cycle 2]: 0.00032679, [27]
        [switch_simplify]: 5.05001e-06
        [loop_unroll]: 4.88001e-06
        [a_1]: 0.00010144
        [with_stream_mark]: 8.54998e-06
        [recompute_prepare]: 4.79998e-06
        [updatestate_depend_eliminate]: 2.44001e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 4.97999e-06
        [updatestate_useless_node_eliminater]: 4.63999e-06
        [accelerated_algorithm]: 4.90001e-06
        [meta_shard_fg_expand]: 1.11002e-06
        [get_grad_eliminate_]: 4.34002e-06
        [merge_forward]: 2.53998e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.233e-05
        [j_node_and_user_rematch]: 7.38999e-06
        [meta_fg_expand]: 1.43002e-06
        [replace_old_param]: 7.18998e-06
        [inline_without_move]: 4.44998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 1.10001e-06
        [auto_monad_eliminator]: 6.04999e-06
        [cse]: 1.082e-05
        [replace_applicator]: 9.17001e-06
[py_interpret_to_execute_after_opt_a]: 9.07999e-06
[rewriter_after_opt_a]: 3.264e-05
[convert_after_rewriter]: 6.93998e-06
[order_py_execute_after_rewriter]: 4.74e-06
[mutable_eliminate]: 0.00051282
[jit_opt_b]: 5.106e-05, [1]
    [Cycle 1]: 4.423e-05, [2]
        [frontend_op_eliminate]: 1.732e-05
        [inline_after_opt_a]: 1.526e-05
[cconv]: 2.397e-05
[loop_unroll]: 0.00040144
[jit_opt_after_cconv]: 0.00013223, [1]
    [Cycle 1]: 0.0001262, [11]
        [c_1]: 1.943e-05
        [parameter_eliminate]: 2.46e-06
        [updatestate_depend_eliminate]: 5.32999e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.19999e-06
        [cse]: 1.759e-05
        [call_graph_tuple_transform]: 1.721e-05
        [tuple_list_get_item_eliminator]: 5.11002e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 4.97999e-06
[remove_dup_value]: 1.369e-05
[partial_unused_args_eliminate]: 2.33998e-06
[environ_conv]: 5.08002e-06
[add_recomputation]: 4.692e-05
[cse_after_recomputation]: 1.902e-05, [1]
    [Cycle 1]: 1.378e-05, [1]
        [cse]: 8.43999e-06
[auto_monad_reorder]: 1.393e-05
[get_jit_bprop_graph]: 1.43002e-06
[rewriter_after_jit_bprop_graph]: 5.61e-06
[opt_after_jit_grad]: 0.00045071
[symbol_engine_optimizer]: 6.985e-05, [1]
    [Cycle 1]: 6.412e-05, [6]
        [build]: 3.44001e-06
        [elim_shapecalc]: 7.31001e-06
        [elim_not_effective]: 1.198e-05
        [opt_reshape]: 5.92999e-06
        [fold_const_symbol]: 8e-06
        [renormalize]: 4.00003e-07
[validate]: 3.11e-05
Sums
    bootstrap                            :     0.000290s :  3.68%
    type_inference                       :     0.004231s : 53.69%
    event_method                         :     0.000016s :  0.20%
    auto_monad                           :     0.000063s :  0.80%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000029s :  0.37%
    rewriter_before_opt_a                :     0.000071s :  0.90%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000058s :  0.73%
    jit_opt_a.loop_unroll                :     0.000033s :  0.42%
    jit_opt_a.a_1                        :     0.000605s :  7.67%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000490s :  6.21%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.24%
    jit_opt_a.cse                        :     0.000038s :  0.49%
    jit_opt_a.replace_applicator         :     0.000020s :  0.25%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.12%
    rewriter_after_opt_a                 :     0.000033s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000513s :  6.51%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000024s :  0.30%
    loop_unroll                          :     0.000401s :  5.09%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000018s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000047s :  0.60%
    cse_after_recomputation.cse          :     0.000008s :  0.11%
    auto_monad_reorder                   :     0.000014s :  0.18%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000451s :  5.72%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000031s :  0.39%


Time group info:
------[substitution.]   0.000190    23
 1.10% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.75% :     0.000005s :      3: substitution.graph_param_transform
83.10% :     0.000158s :      4: substitution.inline
 1.49% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.05% :     0.000004s :      2: substitution.replace_old_param
 6.66% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004163     2
81.93% :     0.003411s :      1: type_inference.infer
18.07% :     0.000752s :      1: type_inference.specialize
------[replace.]   0.000062     6
68.68% :     0.000042s :      4: replace.inline
31.32% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000168     6
93.08% :     0.000156s :      4: match.inline
 6.92% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.50% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.58% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 1.89% :     0.000002s :     10: predicate.arithmetic_simplify
 1.14% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.62% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.61% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.22% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.33% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.63% :     0.000002s :      3: predicate.mutable_eliminate
 0.67% :     0.000001s :      3: predicate.opt_reshape
 2.21% :     0.000003s :     15: predicate.partial_eliminate
 1.48% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.81% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.62% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.61% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.48% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 1.97% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.73% :     0.000009s :     53: predicate.switch_simplify
 1.29% :     0.000001s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.47% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 2.05% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000533    12
52.68% :     0.000281s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.11% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.21% :     0.000220s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013050    72
 0.38% :     0.000050s :      1: add_recomputation
 0.50% :     0.000066s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.38% :     0.000310s :      1: bootstrap
 0.20% :     0.000026s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000021s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.22% :     0.004988s :      1: jit_opt_a
 1.03% :     0.000135s :      1: jit_opt_after_cconv
 0.41% :     0.000054s :      1: jit_opt_b
 3.14% :     0.000410s :      1: loop_unroll
 4.00% :     0.000521s :      1: mutable_eliminate
 6.18% :     0.000807s :     26: opt.transform.jit_opt_a
 0.33% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.51% :     0.000459s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000032s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 1.94% :     0.000254s :      1: renormalize.infer
 1.76% :     0.000229s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000036s :      1: rewriter_after_opt_a
 0.57% :     0.000074s :      1: rewriter_before_opt_a
 0.55% :     0.000072s :      1: symbol_engine_optimizer
32.52% :     0.004244s :      1: type_inference

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.143.362 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda84a8f70> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.143.964 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.145.604 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c371940> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.146.241 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0176691, [33]
[bootstrap]: 0.00028043
[type_inference]: 0.00431042
[event_method]: 1.626e-05
[auto_monad]: 6.36e-05
[graph_reusing]: 5.84999e-06
[pre_auto_parallel]: 2.39999e-06
[py_interpret_to_execute]: 3.378e-05
[rewriter_before_opt_a]: 7.137e-05
[expand_dump_flag]: 2.95002e-06
[jit_opt_a]: 0.00497432, [2]
    [Cycle 1]: 0.00147292, [27]
        [switch_simplify]: 5.202e-05
        [loop_unroll]: 2.826e-05
        [a_1]: 0.00050954
        [with_stream_mark]: 1.539e-05
        [recompute_prepare]: 6.96999e-06
        [updatestate_depend_eliminate]: 3.23e-06
        [updatestate_assign_eliminate]: 2.95002e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [parameter_eliminate]: 1.82999e-06
        [specialize_transform]: 5.86998e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 5.08002e-06
        [meta_shard_fg_expand]: 2.18998e-06
        [get_grad_eliminate_]: 4.85999e-06
        [merge_forward]: 3.45e-06
        [cell_reuse_recompute_pass]: 1.60001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.318e-05
        [j_node_and_user_rematch]: 8.61002e-06
        [meta_fg_expand]: 2.09e-06
        [replace_old_param]: 8.40001e-06
        [inline_without_move]: 5.02999e-06
        [renormalize]: 0.00055854
        [add_forward_monad_depend]: 5.79999e-06
        [auto_monad_grad]: 2.73e-06
        [auto_monad_eliminator]: 1.281e-05
        [cse]: 2.588e-05
        [replace_applicator]: 1.162e-05
    [Cycle 2]: 0.00030757, [27]
        [switch_simplify]: 5.07e-06
        [loop_unroll]: 4.66997e-06
        [a_1]: 8.681e-05
        [with_stream_mark]: 8.87e-06
        [recompute_prepare]: 4.94e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 1.92001e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 5.14e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 4.83001e-06
        [meta_shard_fg_expand]: 9.70002e-07
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.53e-06
        [cell_reuse_recompute_pass]: 1.16002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.199e-05
        [j_node_and_user_rematch]: 7.97998e-06
        [meta_fg_expand]: 1.64e-06
        [replace_old_param]: 7.38999e-06
        [inline_without_move]: 4.61002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.11002e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 4.97e-06
        [cse]: 9.76e-06
        [replace_applicator]: 5.05001e-06
[py_interpret_to_execute_after_opt_a]: 1.009e-05
[rewriter_after_opt_a]: 3.28e-05
[convert_after_rewriter]: 6.39001e-06
[order_py_execute_after_rewriter]: 4.70001e-06
[mutable_eliminate]: 0.00052398
[jit_opt_b]: 4.969e-05, [1]
    [Cycle 1]: 4.294e-05, [2]
        [frontend_op_eliminate]: 1.718e-05
        [inline_after_opt_a]: 1.487e-05
[cconv]: 2.39e-05
[loop_unroll]: 0.00040198
[jit_opt_after_cconv]: 0.00013257, [1]
    [Cycle 1]: 0.00012608, [11]
        [c_1]: 1.903e-05
        [parameter_eliminate]: 2.69001e-06
        [updatestate_depend_eliminate]: 5.44998e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.676e-05
        [call_graph_tuple_transform]: 1.837e-05
        [tuple_list_get_item_eliminator]: 4.95001e-06
        [none_parameter_eliminate]: 1.32999e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 5.04e-06
[remove_dup_value]: 1.3e-05
[partial_unused_args_eliminate]: 2.04e-06
[environ_conv]: 4.79e-06
[add_recomputation]: 4.518e-05
[cse_after_recomputation]: 1.913e-05, [1]
    [Cycle 1]: 1.394e-05, [1]
        [cse]: 8.28999e-06
[auto_monad_reorder]: 1.51e-05
[get_jit_bprop_graph]: 1.91e-06
[rewriter_after_jit_bprop_graph]: 3.93001e-06
[opt_after_jit_grad]: 0.00043759
[symbol_engine_optimizer]: 6.795e-05, [1]
    [Cycle 1]: 6.185e-05, [6]
        [build]: 3.17002e-06
        [elim_shapecalc]: 6.73e-06
        [elim_not_effective]: 1.143e-05
        [opt_reshape]: 5.18002e-06
        [fold_const_symbol]: 8.05e-06
        [renormalize]: 9.39996e-07
[validate]: 3.196e-05
[backend_pass]: 8.39995e-07
[task_emit]: 0.00588921
[execute]: 7.06999e-06
Sums
    bootstrap                            :     0.000280s :  2.02%
    type_inference                       :     0.004310s : 31.01%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000064s :  0.46%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000034s :  0.24%
    rewriter_before_opt_a                :     0.000071s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.41%
    jit_opt_a.loop_unroll                :     0.000033s :  0.24%
    jit_opt_a.a_1                        :     0.000596s :  4.29%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000559s :  4.02%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000036s :  0.26%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.24%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000524s :  3.77%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000402s :  2.89%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000045s :  0.33%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000438s :  3.15%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005889s : 42.37%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000181    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.67% :     0.000001s :      2: substitution.fold_const_symbol
 3.08% :     0.000006s :      3: substitution.graph_param_transform
82.43% :     0.000149s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.35% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.99% :     0.000004s :      2: substitution.replace_old_param
 6.93% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004245     2
81.12% :     0.003444s :      1: type_inference.infer
18.88% :     0.000801s :      1: type_inference.specialize
------[replace.]   0.000063     6
67.88% :     0.000043s :      4: replace.inline
32.12% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000158     6
92.76% :     0.000146s :      4: match.inline
 7.24% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.35% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.30% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000001s :     10: predicate.addn_zero_filter
 2.10% :     0.000002s :     10: predicate.arithmetic_simplify
 1.31% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000001s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.38% :     0.000002s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.17% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      6: predicate.less_batch_normalization
 1.72% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.85% :     0.000002s :     15: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.09% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.21% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.52% :     0.000002s :      3: predicate.mutable_eliminate
 0.41% :     0.000000s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.31% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.81% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000001s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.64% :     0.000003s :     16: predicate.switch_defer_inline
 2.20% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.21% :     0.000009s :     53: predicate.switch_simplify
 1.31% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.03% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.35% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000551    12
50.50% :     0.000278s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.63% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.88% :     0.000247s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019123    76
 0.25% :     0.000048s :      1: add_recomputation
 0.35% :     0.000067s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.54% :     0.000294s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.05% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.03% :     0.004977s :      1: jit_opt_a
 0.71% :     0.000135s :      1: jit_opt_after_cconv
 0.27% :     0.000052s :      1: jit_opt_b
 2.14% :     0.000410s :      1: loop_unroll
 2.78% :     0.000532s :      1: mutable_eliminate
 4.16% :     0.000796s :     26: opt.transform.jit_opt_a
 0.23% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.33% :     0.000445s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.43% :     0.000273s :      1: renormalize.infer
 1.46% :     0.000279s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000035s :      1: rewriter_after_opt_a
 0.39% :     0.000074s :      1: rewriter_before_opt_a
 0.37% :     0.000070s :      1: symbol_engine_optimizer
30.85% :     0.005899s :      1: task_emit
22.61% :     0.004325s :      1: type_inference
 0.26% :     0.000050s :      1: validate


TotalTime = 0.0111516, [30]
[bootstrap]: 0.00026721
[type_inference]: 0.00391717
[event_method]: 1.677e-05
[auto_monad]: 6.283e-05
[graph_reusing]: 6.03002e-06
[pre_auto_parallel]: 2.39001e-06
[py_interpret_to_execute]: 2.614e-05
[rewriter_before_opt_a]: 6.894e-05
[expand_dump_flag]: 2.83e-06
[jit_opt_a]: 0.00480459, [2]
    [Cycle 1]: 0.00144637, [27]
        [switch_simplify]: 4.854e-05
        [loop_unroll]: 2.77e-05
        [a_1]: 0.0005568
        [with_stream_mark]: 1.626e-05
        [recompute_prepare]: 7.45e-06
        [updatestate_depend_eliminate]: 3.23998e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.69999e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 5.59998e-06
        [updatestate_useless_node_eliminater]: 4.97999e-06
        [accelerated_algorithm]: 5.51998e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 5.21002e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 1.10999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.47e-05
        [j_node_and_user_rematch]: 9.08002e-06
        [meta_fg_expand]: 1.97001e-06
        [replace_old_param]: 8.94e-06
        [inline_without_move]: 5.29e-06
        [renormalize]: 0.00048289
        [add_forward_monad_depend]: 5.03002e-06
        [auto_monad_grad]: 1.92001e-06
        [auto_monad_eliminator]: 1.249e-05
        [cse]: 2.941e-05
        [replace_applicator]: 1.04e-05
    [Cycle 2]: 0.00030489, [27]
        [switch_simplify]: 5.32001e-06
        [loop_unroll]: 4.83001e-06
        [a_1]: 8.54e-05
        [with_stream_mark]: 8.40999e-06
        [recompute_prepare]: 4.72998e-06
        [updatestate_depend_eliminate]: 2.40002e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 9.09989e-07
        [specialize_transform]: 4.67e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 5.05001e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 4.44998e-06
        [merge_forward]: 2.19999e-06
        [cell_reuse_recompute_pass]: 1.13001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.184e-05
        [j_node_and_user_rematch]: 7.38999e-06
        [meta_fg_expand]: 1.34e-06
        [replace_old_param]: 7.4e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 4.47e-06
        [cse]: 9.64999e-06
        [replace_applicator]: 4.80999e-06
[py_interpret_to_execute_after_opt_a]: 8.27e-06
[rewriter_after_opt_a]: 3.07e-05
[convert_after_rewriter]: 6.99001e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00050099
[jit_opt_b]: 4.946e-05, [1]
    [Cycle 1]: 4.249e-05, [2]
        [frontend_op_eliminate]: 1.636e-05
        [inline_after_opt_a]: 1.45e-05
[cconv]: 2.398e-05
[loop_unroll]: 0.00040412
[jit_opt_after_cconv]: 0.00013264, [1]
    [Cycle 1]: 0.00012662, [11]
        [c_1]: 1.975e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 4.84e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [cse]: 1.685e-05
        [call_graph_tuple_transform]: 1.728e-05
        [tuple_list_get_item_eliminator]: 5.24e-06
        [none_parameter_eliminate]: 1.34998e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 5.09e-06
[remove_dup_value]: 1.347e-05
[partial_unused_args_eliminate]: 2.21998e-06
[environ_conv]: 5.15001e-06
[add_recomputation]: 4.586e-05
[cse_after_recomputation]: 1.998e-05, [1]
    [Cycle 1]: 1.434e-05, [1]
        [cse]: 8.45001e-06
[auto_monad_reorder]: 1.451e-05
[get_jit_bprop_graph]: 1.38002e-06
[rewriter_after_jit_bprop_graph]: 4.75999e-06
[opt_after_jit_grad]: 0.000447
[symbol_engine_optimizer]: 7.091e-05, [1]
    [Cycle 1]: 6.487e-05, [6]
        [build]: 3.87002e-06
        [elim_shapecalc]: 7.11999e-06
        [elim_not_effective]: 1.226e-05
        [opt_reshape]: 5.66e-06
        [fold_const_symbol]: 8.43999e-06
        [renormalize]: 4.19997e-07
[validate]: 3.116e-05
Sums
    bootstrap                            :     0.000267s :  3.54%
    type_inference                       :     0.003917s : 51.95%
    event_method                         :     0.000017s :  0.22%
    auto_monad                           :     0.000063s :  0.83%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000026s :  0.35%
    rewriter_before_opt_a                :     0.000069s :  0.91%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000054s :  0.71%
    jit_opt_a.loop_unroll                :     0.000033s :  0.43%
    jit_opt_a.a_1                        :     0.000642s :  8.52%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000010s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.22%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000483s :  6.40%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.22%
    jit_opt_a.cse                        :     0.000039s :  0.52%
    jit_opt_a.replace_applicator         :     0.000015s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000031s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000501s :  6.64%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.19%
    cconv                                :     0.000024s :  0.32%
    loop_unroll                          :     0.000404s :  5.36%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000013s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000046s :  0.61%
    cse_after_recomputation.cse          :     0.000008s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.19%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000447s :  5.93%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000031s :  0.41%


Time group info:
------[substitution.]   0.000179    23
 1.10% :     0.000002s :      2: substitution.elim_not_effective
 0.86% :     0.000002s :      2: substitution.fold_const_symbol
 2.76% :     0.000005s :      3: substitution.graph_param_transform
82.04% :     0.000147s :      4: substitution.inline
 1.90% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.39% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.56% :     0.000005s :      2: substitution.replace_old_param
 6.40% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003858     2
80.51% :     0.003106s :      1: type_inference.infer
19.49% :     0.000752s :      1: type_inference.specialize
------[replace.]   0.000060     6
66.45% :     0.000040s :      4: replace.inline
33.55% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000155     6
93.32% :     0.000145s :      4: match.inline
 6.68% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.79% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.38% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.52% :     0.000002s :     10: predicate.addn_zero_filter
 2.43% :     0.000003s :     10: predicate.arithmetic_simplify
 1.19% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000001s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.47% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.96% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.34% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.80% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.15% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.77% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000002s :     15: predicate.partial_eliminate
 1.41% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.50% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.55% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.50% :     0.000009s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.78% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.84% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000492    12
51.61% :     0.000254s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.81% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.58% :     0.000210s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012531    72
 0.39% :     0.000049s :      1: add_recomputation
 0.53% :     0.000066s :      1: auto_monad
 0.14% :     0.000017s :      1: auto_monad_reorder
 2.27% :     0.000285s :      1: bootstrap
 0.21% :     0.000026s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.07% :     0.000008s :      1: graph_reusing
38.36% :     0.004807s :      1: jit_opt_a
 1.08% :     0.000136s :      1: jit_opt_after_cconv
 0.41% :     0.000052s :      1: jit_opt_b
 3.28% :     0.000411s :      1: loop_unroll
 4.06% :     0.000509s :      1: mutable_eliminate
 6.68% :     0.000837s :     26: opt.transform.jit_opt_a
 0.35% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.64% :     0.000456s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.23% :     0.000029s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000016s :      1: remove_dup_value
 2.00% :     0.000251s :      1: renormalize.infer
 1.80% :     0.000225s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000034s :      1: rewriter_after_opt_a
 0.57% :     0.000072s :      1: rewriter_before_opt_a
 0.58% :     0.000073s :      1: symbol_engine_optimizer
31.35% :     0.003928s :      1: type_inference


TotalTime = 0.021113, [33]
[bootstrap]: 0.00021517
[type_inference]: 0.0104307
[event_method]: 4.534e-05
[auto_monad]: 7.579e-05
[graph_reusing]: 5.94e-06
[pre_auto_parallel]: 2.17999e-06
[py_interpret_to_execute]: 2.241e-05
[rewriter_before_opt_a]: 9.909e-05
[expand_dump_flag]: 2.59999e-06
[jit_opt_a]: 0.00820009, [2]
    [Cycle 1]: 0.00441875, [27]
        [switch_simplify]: 5.989e-05
        [loop_unroll]: 3.781e-05
        [a_1]: 0.00090259
        [with_stream_mark]: 2.24e-05
        [recompute_prepare]: 1.812e-05
        [updatestate_depend_eliminate]: 7.31999e-06
        [updatestate_assign_eliminate]: 5.72001e-06
        [updatestate_loads_eliminate]: 5.17999e-06
        [parameter_eliminate]: 2.19001e-06
        [specialize_transform]: 1.301e-05
        [updatestate_useless_node_eliminater]: 1.187e-05
        [accelerated_algorithm]: 1.218e-05
        [meta_shard_fg_expand]: 2.97002e-06
        [get_grad_eliminate_]: 1.19e-05
        [merge_forward]: 7.00998e-06
        [cell_reuse_recompute_pass]: 7.39994e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.463e-05
        [j_node_and_user_rematch]: 2.13e-05
        [meta_fg_expand]: 0.00140318
        [replace_old_param]: 5.105e-05
        [inline_without_move]: 5.071e-05
        [renormalize]: 0.00151523
        [add_forward_monad_depend]: 4.23999e-06
        [auto_monad_grad]: 1.60001e-06
        [auto_monad_eliminator]: 8.1e-06
        [cse]: 1.62e-05
        [replace_applicator]: 1.096e-05
    [Cycle 2]: 0.00033874, [27]
        [switch_simplify]: 4.2e-06
        [loop_unroll]: 3.85e-06
        [a_1]: 0.0001013
        [with_stream_mark]: 7.97998e-06
        [recompute_prepare]: 4.03999e-06
        [updatestate_depend_eliminate]: 2.19999e-06
        [updatestate_assign_eliminate]: 1.82001e-06
        [updatestate_loads_eliminate]: 1.52999e-06
        [parameter_eliminate]: 1.11997e-06
        [specialize_transform]: 3.76001e-06
        [updatestate_useless_node_eliminater]: 3.58e-06
        [accelerated_algorithm]: 3.21999e-06
        [meta_shard_fg_expand]: 9.89996e-07
        [get_grad_eliminate_]: 3.50998e-06
        [merge_forward]: 1.89e-06
        [cell_reuse_recompute_pass]: 1.79998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 8.79e-06
        [j_node_and_user_rematch]: 5.35999e-06
        [meta_fg_expand]: 3.538e-05
        [replace_old_param]: 3.79002e-06
        [inline_without_move]: 3.28998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 8.89995e-07
        [auto_monad_eliminator]: 4.50001e-06
        [cse]: 1.061e-05
        [replace_applicator]: 3.48e-06
[py_interpret_to_execute_after_opt_a]: 9.57001e-06
[rewriter_after_opt_a]: 3.277e-05
[convert_after_rewriter]: 6.99001e-06
[order_py_execute_after_rewriter]: 4.48001e-06
[mutable_eliminate]: 0.00053484
[jit_opt_b]: 4.137e-05, [1]
    [Cycle 1]: 3.412e-05, [2]
        [frontend_op_eliminate]: 1.158e-05
        [inline_after_opt_a]: 1.12e-05
[cconv]: 2.237e-05
[loop_unroll]: 0.00041421
[jit_opt_after_cconv]: 0.00011536, [1]
    [Cycle 1]: 0.00010949, [11]
        [c_1]: 1.178e-05
        [parameter_eliminate]: 2.32999e-06
        [updatestate_depend_eliminate]: 4.72e-06
        [updatestate_assign_eliminate]: 1.93002e-06
        [updatestate_loads_eliminate]: 1.64e-06
        [cse]: 1.558e-05
        [call_graph_tuple_transform]: 1.318e-05
        [tuple_list_get_item_eliminator]: 3.9e-06
        [none_parameter_eliminate]: 1.34e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 3.71001e-06
[remove_dup_value]: 1.458e-05
[partial_unused_args_eliminate]: 2.01e-06
[environ_conv]: 5.19e-06
[add_recomputation]: 3.256e-05
[cse_after_recomputation]: 1.958e-05, [1]
    [Cycle 1]: 1.378e-05, [1]
        [cse]: 8.17998e-06
[auto_monad_reorder]: 1.379e-05
[get_jit_bprop_graph]: 1.67999e-06
[rewriter_after_jit_bprop_graph]: 5.09e-06
[opt_after_jit_grad]: 0.00043838
[symbol_engine_optimizer]: 6.408e-05, [1]
    [Cycle 1]: 5.769e-05, [6]
        [build]: 3.00002e-06
        [elim_shapecalc]: 6.06e-06
        [elim_not_effective]: 9.97999e-06
        [opt_reshape]: 4.09997e-06
        [fold_const_symbol]: 5.66998e-06
        [renormalize]: 4.39992e-07
[validate]: 2.679e-05
[backend_pass]: 7.80012e-07
[task_emit]: 1.847e-05
[execute]: 1.15999e-06
Sums
    bootstrap                            :     0.000215s :  1.26%
    type_inference                       :     0.010431s : 61.15%
    event_method                         :     0.000045s :  0.27%
    auto_monad                           :     0.000076s :  0.44%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000022s :  0.13%
    rewriter_before_opt_a                :     0.000099s :  0.58%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.38%
    jit_opt_a.loop_unroll                :     0.000042s :  0.24%
    jit_opt_a.a_1                        :     0.001004s :  5.89%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000015s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.09%
    jit_opt_a.merge_forward              :     0.000009s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001439s :  8.43%
    jit_opt_a.replace_old_param          :     0.000055s :  0.32%
    jit_opt_a.inline_without_move        :     0.000054s :  0.32%
    jit_opt_a.renormalize                :     0.001515s :  8.88%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.07%
    jit_opt_a.cse                        :     0.000027s :  0.16%
    jit_opt_a.replace_applicator         :     0.000014s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.19%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000535s :  3.14%
    jit_opt_b.frontend_op_eliminate      :     0.000012s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000022s :  0.13%
    loop_unroll                          :     0.000414s :  2.43%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000033s :  0.19%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000438s :  2.57%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000018s :  0.11%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000274    50
 0.58% :     0.000002s :      1: substitution.elim_not_effective
 0.34% :     0.000001s :      1: substitution.fold_const_symbol
 1.56% :     0.000004s :      1: substitution.graph_param_transform
71.98% :     0.000197s :      6: substitution.inline
 5.70% :     0.000016s :      2: substitution.inline_without_move
 1.96% :     0.000005s :      9: substitution.j_node_and_user_rematch
 1.07% :     0.000003s :      2: substitution.minmaximum_grad
 1.26% :     0.000003s :      3: substitution.partial_eliminate
 2.52% :     0.000007s :      9: substitution.remove_not_recompute_node
 1.96% :     0.000005s :      2: substitution.replace_applicator
 1.53% :     0.000004s :      5: substitution.replace_old_param
 0.83% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.46% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.38% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 4.86% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010365     2
89.72% :     0.009300s :      1: type_inference.infer
10.28% :     0.001065s :      1: type_inference.specialize
------[replace.]   0.000064     8
78.15% :     0.000050s :      6: replace.inline
21.85% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000202     8
95.72% :     0.000194s :      6: match.inline
 4.28% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000165  1109
 1.57% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.61% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.28% :     0.000002s :     18: predicate.addn_check_dump
 1.56% :     0.000003s :     18: predicate.addn_zero_filter
 2.14% :     0.000004s :     18: predicate.arithmetic_simplify
 1.45% :     0.000002s :     18: predicate.cast_eliminate
 0.11% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.25% :     0.000002s :     18: predicate.compare_switch_simplify
 1.55% :     0.000003s :     18: predicate.depend_value_elim
 1.37% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.39% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.39% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.48% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.16% :     0.000000s :      1: predicate.elim_not_effective
 0.33% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.51% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.32% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.28% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.72% :     0.000003s :     18: predicate.environ_get_eliminate
 1.35% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.18% :     0.000002s :     12: predicate.get_grad_eliminate
 0.08% :     0.000000s :      1: predicate.graph_param_transform
 4.45% :     0.000007s :     28: predicate.inline
 3.36% :     0.000006s :     33: predicate.inline_without_move
 0.51% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.34% :     0.000002s :     12: predicate.less_batch_normalization
 1.53% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.67% :     0.000003s :     21: predicate.load_eliminater
 0.61% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.73% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.75% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.30% :     0.000002s :     18: predicate.merge_addn
 1.31% :     0.000002s :     18: predicate.minmaximum_grad
 0.74% :     0.000001s :      1: predicate.mutable_eliminate
 0.16% :     0.000000s :      1: predicate.opt_reshape
 2.11% :     0.000003s :     21: predicate.partial_eliminate
 1.39% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.80% :     0.000003s :     18: predicate.reduce_eliminate
 1.62% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.47% :     0.000002s :     22: predicate.replace_applicator
 1.63% :     0.000003s :     33: predicate.replace_old_param
 0.12% :     0.000000s :      1: predicate.reset_defer_inline
 1.54% :     0.000003s :     18: predicate.reshape_eliminate
 1.66% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.49% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.45% :     0.000002s :     18: predicate.same_eliminate
 0.64% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.31% :     0.000001s :      2: predicate.special_op_eliminate
 1.25% :     0.000002s :     12: predicate.specialize_transform
 1.57% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.13% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.75% :     0.000005s :     26: predicate.switch_defer_inline
 2.31% :     0.000004s :     26: predicate.switch_layer_defer_inline
 7.06% :     0.000012s :     71: predicate.switch_simplify
 1.45% :     0.000002s :     18: predicate.tile_eliminate
 1.37% :     0.000002s :     18: predicate.transpose_eliminate
 1.65% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.48% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 2.65% :     0.000004s :     22: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.59% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.87% :     0.000003s :     18: predicate.value_based_eliminate
 0.31% :     0.000001s :      1: predicate.virtual_view_grad_eliminate
 0.28% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001142    22
73.18% :     0.000836s :     13: func_graph_cloner_run.FuncGraphClonerGraph
26.82% :     0.000306s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024054    76
 0.15% :     0.000035s :      1: add_recomputation
 0.33% :     0.000080s :      1: auto_monad
 0.07% :     0.000016s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 1.04% :     0.000251s :      1: bootstrap
 0.10% :     0.000025s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.20% :     0.000049s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
34.10% :     0.008203s :      1: jit_opt_a
 0.49% :     0.000118s :      1: jit_opt_after_cconv
 0.18% :     0.000044s :      1: jit_opt_b
 1.75% :     0.000420s :      1: loop_unroll
 2.26% :     0.000543s :      1: mutable_eliminate
 5.62% :     0.001351s :     26: opt.transform.jit_opt_a
 0.12% :     0.000029s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000022s :      4: opt.transform.symbol_engine_opt
 1.85% :     0.000446s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.10% :     0.000025s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 4.85% :     0.001168s :      1: renormalize.infer
 1.42% :     0.000340s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000036s :      1: rewriter_after_opt_a
 0.43% :     0.000103s :      1: rewriter_before_opt_a
 0.28% :     0.000067s :      1: symbol_engine_optimizer
 0.09% :     0.000023s :      1: task_emit
43.42% :     0.010443s :      1: type_inference
 0.18% :     0.000042s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.174.349 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c371dc0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.174.966 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0357633, [33]
[bootstrap]: 0.0002034
[type_inference]: 0.00982611
[event_method]: 4.614e-05
[auto_monad]: 8.938e-05
[graph_reusing]: 6.41e-06
[pre_auto_parallel]: 2.96001e-06
[py_interpret_to_execute]: 2.589e-05
[rewriter_before_opt_a]: 8.798e-05
[expand_dump_flag]: 3.87998e-06
[jit_opt_a]: 0.0116157, [3]
    [Cycle 1]: 0.00575201, [27]
        [switch_simplify]: 6.095e-05
        [loop_unroll]: 3.924e-05
        [a_1]: 0.00083004
        [with_stream_mark]: 2.554e-05
        [recompute_prepare]: 1.791e-05
        [updatestate_depend_eliminate]: 7.54002e-06
        [updatestate_assign_eliminate]: 6.37001e-06
        [updatestate_loads_eliminate]: 6.29999e-06
        [parameter_eliminate]: 2.21e-06
        [specialize_transform]: 1.32e-05
        [updatestate_useless_node_eliminater]: 1.168e-05
        [accelerated_algorithm]: 1.206e-05
        [meta_shard_fg_expand]: 3.16001e-06
        [get_grad_eliminate_]: 1.2e-05
        [merge_forward]: 8.1e-06
        [cell_reuse_recompute_pass]: 8.80013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.614e-05
        [j_node_and_user_rematch]: 2.277e-05
        [meta_fg_expand]: 0.00139729
        [replace_old_param]: 5.088e-05
        [inline_without_move]: 5.091e-05
        [renormalize]: 0.0026627
        [add_forward_monad_depend]: 9.79e-06
        [auto_monad_grad]: 6.33002e-06
        [auto_monad_eliminator]: 5.272e-05
        [cse]: 0.00015296
        [replace_applicator]: 6.517e-05
    [Cycle 2]: 0.00217333, [27]
        [switch_simplify]: 3.789e-05
        [loop_unroll]: 3.626e-05
        [a_1]: 0.00109897
        [with_stream_mark]: 1.417e-05
        [recompute_prepare]: 9.88998e-06
        [updatestate_depend_eliminate]: 4.88001e-06
        [updatestate_assign_eliminate]: 4.03999e-06
        [updatestate_loads_eliminate]: 3.91999e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 8.20999e-06
        [updatestate_useless_node_eliminater]: 8.03001e-06
        [accelerated_algorithm]: 8.3e-06
        [meta_shard_fg_expand]: 1.74998e-06
        [get_grad_eliminate_]: 8.56002e-06
        [merge_forward]: 4.3e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.732e-05
        [j_node_and_user_rematch]: 1.403e-05
        [meta_fg_expand]: 4.002e-05
        [replace_old_param]: 1.191e-05
        [inline_without_move]: 8.48999e-06
        [renormalize]: 0.00059766
        [add_forward_monad_depend]: 4.47e-06
        [auto_monad_grad]: 1.28002e-06
        [auto_monad_eliminator]: 1.487e-05
        [cse]: 4.327e-05
        [replace_applicator]: 1.56e-05
    [Cycle 3]: 0.00048215, [27]
        [switch_simplify]: 8.80999e-06
        [loop_unroll]: 7.92e-06
        [a_1]: 0.00018829
        [with_stream_mark]: 9.74e-06
        [recompute_prepare]: 8.1e-06
        [updatestate_depend_eliminate]: 4.33999e-06
        [updatestate_assign_eliminate]: 4.23999e-06
        [updatestate_loads_eliminate]: 3.70998e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 8.22e-06
        [updatestate_useless_node_eliminater]: 7.92e-06
        [accelerated_algorithm]: 8.97999e-06
        [meta_shard_fg_expand]: 1.54e-06
        [get_grad_eliminate_]: 7.88999e-06
        [merge_forward]: 4.1e-06
        [cell_reuse_recompute_pass]: 1.37999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.572e-05
        [j_node_and_user_rematch]: 1.366e-05
        [meta_fg_expand]: 2.59999e-06
        [replace_old_param]: 1.035e-05
        [inline_without_move]: 8.05e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 9.71998e-06
        [cse]: 2.351e-05
        [replace_applicator]: 7.98001e-06
[py_interpret_to_execute_after_opt_a]: 1.225e-05
[rewriter_after_opt_a]: 4.04e-05
[convert_after_rewriter]: 7.46999e-06
[order_py_execute_after_rewriter]: 5.30999e-06
[mutable_eliminate]: 0.00059192
[jit_opt_b]: 6.656e-05, [1]
    [Cycle 1]: 5.925e-05, [2]
        [frontend_op_eliminate]: 2.512e-05
        [inline_after_opt_a]: 2.331e-05
[cconv]: 1.646e-05
[loop_unroll]: 0.00042222
[jit_opt_after_cconv]: 0.00018809, [1]
    [Cycle 1]: 0.00018139, [11]
        [c_1]: 3.482e-05
        [parameter_eliminate]: 2.89001e-06
        [updatestate_depend_eliminate]: 7.87e-06
        [updatestate_assign_eliminate]: 4.58001e-06
        [updatestate_loads_eliminate]: 4.08001e-06
        [cse]: 3.312e-05
        [call_graph_tuple_transform]: 2.385e-05
        [tuple_list_get_item_eliminator]: 8.33001e-06
        [none_parameter_eliminate]: 1.05999e-06
        [renormalize]: 3.4002e-07
        [switch_simplify]: 8.89998e-06
[remove_dup_value]: 1.674e-05
[partial_unused_args_eliminate]: 1.74e-06
[environ_conv]: 5.97001e-06
[add_recomputation]: 4.481e-05
[cse_after_recomputation]: 3.428e-05, [1]
    [Cycle 1]: 2.878e-05, [1]
        [cse]: 2.269e-05
[auto_monad_reorder]: 1.575e-05
[get_jit_bprop_graph]: 1.37e-06
[rewriter_after_jit_bprop_graph]: 4.54002e-06
[opt_after_jit_grad]: 0.00046112
[symbol_engine_optimizer]: 9.13e-05, [1]
    [Cycle 1]: 8.498e-05, [6]
        [build]: 3.87998e-06
        [elim_shapecalc]: 1.173e-05
        [elim_not_effective]: 1.828e-05
        [opt_reshape]: 8.95999e-06
        [fold_const_symbol]: 1.419e-05
        [renormalize]: 4.09986e-07
[validate]: 3.808e-05
[backend_pass]: 6.99976e-07
[task_emit]: 0.0115791
[execute]: 7.35e-06
Sums
    bootstrap                            :     0.000203s :  0.64%
    type_inference                       :     0.009826s : 30.93%
    event_method                         :     0.000046s :  0.15%
    auto_monad                           :     0.000089s :  0.28%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.08%
    rewriter_before_opt_a                :     0.000088s :  0.28%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000108s :  0.34%
    jit_opt_a.loop_unroll                :     0.000083s :  0.26%
    jit_opt_a.a_1                        :     0.002117s :  6.66%
    jit_opt_a.with_stream_mark           :     0.000049s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000036s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000014s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000030s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000028s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000029s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000028s :  0.09%
    jit_opt_a.merge_forward              :     0.000017s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000059s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000050s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001440s :  4.53%
    jit_opt_a.replace_old_param          :     0.000073s :  0.23%
    jit_opt_a.inline_without_move        :     0.000067s :  0.21%
    jit_opt_a.renormalize                :     0.003260s : 10.26%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000077s :  0.24%
    jit_opt_a.cse                        :     0.000220s :  0.69%
    jit_opt_a.replace_applicator         :     0.000089s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000040s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000592s :  1.86%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000023s :  0.07%
    cconv                                :     0.000016s :  0.05%
    loop_unroll                          :     0.000422s :  1.33%
    jit_opt_after_cconv.c_1              :     0.000035s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000033s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.03%
    remove_dup_value                     :     0.000017s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000045s :  0.14%
    cse_after_recomputation.cse          :     0.000023s :  0.07%
    auto_monad_reorder                   :     0.000016s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000461s :  1.45%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.011579s : 36.45%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000505   126
 6.84% :     0.000035s :      2: substitution.arithmetic_simplify
 0.48% :     0.000002s :      5: substitution.elim_not_effective
 0.42% :     0.000002s :      5: substitution.fold_const_symbol
 1.11% :     0.000006s :      6: substitution.graph_param_transform
59.30% :     0.000299s :     12: substitution.inline
 3.20% :     0.000016s :      2: substitution.inline_without_move
 2.03% :     0.000010s :     18: substitution.j_node_and_user_rematch
 1.90% :     0.000010s :      7: substitution.minmaximum_grad
 1.40% :     0.000007s :      5: substitution.partial_eliminate
 2.43% :     0.000012s :     18: substitution.remove_not_recompute_node
 4.81% :     0.000024s :     10: substitution.replace_applicator
 1.42% :     0.000007s :      7: substitution.replace_old_param
 0.53% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.90% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.60% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 7.63% :     0.000038s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009756     2
88.35% :     0.008620s :      1: type_inference.infer
11.65% :     0.001136s :      1: type_inference.specialize
------[replace.]   0.000140    19
64.67% :     0.000091s :     12: replace.inline
35.33% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000314    19
93.33% :     0.000293s :     12: match.inline
 6.67% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000404  2895
 1.68% :     0.000007s :     48: predicate.accumulaten_eliminater
 0.54% :     0.000002s :      6: predicate.ad_related_special_op_eliminate
 1.42% :     0.000006s :     48: predicate.addn_check_dump
 1.64% :     0.000007s :     48: predicate.addn_zero_filter
 1.98% :     0.000008s :     48: predicate.arithmetic_simplify
 1.67% :     0.000007s :     48: predicate.cast_eliminate
 0.25% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.42% :     0.000006s :     48: predicate.compare_switch_simplify
 1.59% :     0.000006s :     48: predicate.depend_value_elim
 1.43% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.53% :     0.000006s :     48: predicate.dict_get_item_eliminator
 1.46% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.35% :     0.000001s :      6: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      6: predicate.elim_not_effective
 0.28% :     0.000001s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.51% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.42% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.43% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.52% :     0.000006s :     48: predicate.environ_get_eliminate
 1.47% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.12% :     0.000000s :      6: predicate.fold_const_symbol
 0.91% :     0.000004s :     23: predicate.get_grad_eliminate
 0.13% :     0.000001s :      6: predicate.graph_param_transform
 4.26% :     0.000017s :     79: predicate.inline
 1.74% :     0.000007s :     44: predicate.inline_without_move
 0.39% :     0.000002s :     23: predicate.j_node_and_user_rematch
 1.02% :     0.000004s :     23: predicate.less_batch_normalization
 1.80% :     0.000007s :     55: predicate.list_to_tuple_eliminator_
 2.03% :     0.000008s :     61: predicate.load_eliminater
 0.58% :     0.000002s :      6: predicate.loop_unroll_after_grad
 3.16% :     0.000013s :     90: predicate.loop_unroll_before_grad
 1.89% :     0.000008s :     54: predicate.make_slice_get_slice_eliminator
 1.49% :     0.000006s :     48: predicate.merge_addn
 1.47% :     0.000006s :     48: predicate.minmaximum_grad
 0.66% :     0.000003s :      6: predicate.mutable_eliminate
 0.27% :     0.000001s :      6: predicate.opt_reshape
 2.30% :     0.000009s :     61: predicate.partial_eliminate
 1.61% :     0.000007s :     48: predicate.print_const_string_wrapper
 1.79% :     0.000007s :     48: predicate.reduce_eliminate
 1.76% :     0.000007s :     55: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.48% :     0.000010s :    108: predicate.replace_applicator
 0.92% :     0.000004s :     44: predicate.replace_old_param
 0.15% :     0.000001s :      6: predicate.reset_defer_inline
 1.80% :     0.000007s :     48: predicate.reshape_eliminate
 1.49% :     0.000006s :     48: predicate.row_tensor_add_zeros_like
 0.38% :     0.000002s :      6: predicate.row_tensor_eliminate
 1.55% :     0.000006s :     48: predicate.same_eliminate
 0.49% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.54% :     0.000002s :     12: predicate.special_op_eliminate
 0.93% :     0.000004s :     23: predicate.specialize_transform
 1.79% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.60% :     0.000006s :     48: predicate.stack_unstack_eliminate
 0.24% :     0.000001s :      6: predicate.switch_call_monad_eliminater
 2.52% :     0.000010s :     67: predicate.switch_defer_inline
 2.42% :     0.000010s :     67: predicate.switch_layer_defer_inline
 6.01% :     0.000024s :    163: predicate.switch_simplify
 1.49% :     0.000006s :     48: predicate.tile_eliminate
 1.51% :     0.000006s :     48: predicate.transpose_eliminate
 1.82% :     0.000007s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.68% :     0.000007s :     48: predicate.tuple_list_get_item_depend_reorder
 3.00% :     0.000012s :     67: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000007s :     48: predicate.tuple_list_set_item_eliminator
 1.75% :     0.000007s :     55: predicate.tuple_to_list_eliminator_
 1.90% :     0.000008s :     61: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000012s :     84: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000007s :     48: predicate.value_based_eliminate
 0.19% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.33% :     0.000001s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001412    29
58.53% :     0.000826s :     13: func_graph_cloner_run.FuncGraphClonerGraph
41.47% :     0.000585s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.041954    91
 0.11% :     0.000048s :      1: add_recomputation
 0.22% :     0.000093s :      1: auto_monad
 0.04% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.53% :     0.000221s :      1: bootstrap
 0.05% :     0.000019s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000036s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.12% :     0.000050s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
27.69% :     0.011619s :      1: jit_opt_a
 0.45% :     0.000191s :      1: jit_opt_after_cconv
 0.16% :     0.000069s :      1: jit_opt_b
 1.02% :     0.000430s :      1: loop_unroll
 1.43% :     0.000600s :      1: mutable_eliminate
 6.57% :     0.002756s :     39: opt.transform.jit_opt_a
 0.17% :     0.000073s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000050s :      4: opt.transform.symbol_engine_opt
 1.12% :     0.000470s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000028s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 4.32% :     0.001812s :      2: renormalize.infer
 3.42% :     0.001434s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000043s :      1: rewriter_after_opt_a
 0.22% :     0.000091s :      1: rewriter_before_opt_a
 0.22% :     0.000094s :      1: symbol_engine_optimizer
27.63% :     0.011592s :      1: task_emit
23.45% :     0.009839s :      1: type_inference
 0.15% :     0.000061s :      1: validate


TotalTime = 0.0172102, [33]
[bootstrap]: 0.00025089
[type_inference]: 0.00411667
[event_method]: 1.694e-05
[auto_monad]: 6.125e-05
[graph_reusing]: 5.71e-06
[pre_auto_parallel]: 2.51e-06
[py_interpret_to_execute]: 3.056e-05
[rewriter_before_opt_a]: 6.883e-05
[expand_dump_flag]: 3.14999e-06
[jit_opt_a]: 0.00517439, [2]
    [Cycle 1]: 0.00155506, [27]
        [switch_simplify]: 5.113e-05
        [loop_unroll]: 2.879e-05
        [a_1]: 0.00053363
        [with_stream_mark]: 1.786e-05
        [recompute_prepare]: 6.94001e-06
        [updatestate_depend_eliminate]: 3.06001e-06
        [updatestate_assign_eliminate]: 2.87002e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [parameter_eliminate]: 1.97999e-06
        [specialize_transform]: 5.69e-06
        [updatestate_useless_node_eliminater]: 4.76002e-06
        [accelerated_algorithm]: 5.17999e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 3.56001e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.385e-05
        [j_node_and_user_rematch]: 8.57e-06
        [meta_fg_expand]: 2.22999e-06
        [replace_old_param]: 8.30999e-06
        [inline_without_move]: 4.82998e-06
        [renormalize]: 0.0006206
        [add_forward_monad_depend]: 4.23001e-06
        [auto_monad_grad]: 2.44001e-06
        [auto_monad_eliminator]: 1.308e-05
        [cse]: 2.559e-05
        [replace_applicator]: 1.067e-05
    [Cycle 2]: 0.00031493, [27]
        [switch_simplify]: 5.67001e-06
        [loop_unroll]: 4.48001e-06
        [a_1]: 8.676e-05
        [with_stream_mark]: 9.66e-06
        [recompute_prepare]: 4.74e-06
        [updatestate_depend_eliminate]: 2.66e-06
        [updatestate_assign_eliminate]: 1.91e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 4.90999e-06
        [updatestate_useless_node_eliminater]: 4.64002e-06
        [accelerated_algorithm]: 5.17999e-06
        [meta_shard_fg_expand]: 1.32999e-06
        [get_grad_eliminate_]: 4.61002e-06
        [merge_forward]: 2.62001e-06
        [cell_reuse_recompute_pass]: 1.73002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.271e-05
        [j_node_and_user_rematch]: 7.53999e-06
        [meta_fg_expand]: 1.54998e-06
        [replace_old_param]: 8.06001e-06
        [inline_without_move]: 4.62e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.35001e-06
        [auto_monad_grad]: 9.09989e-07
        [auto_monad_eliminator]: 5.45001e-06
        [cse]: 9.13002e-06
        [replace_applicator]: 9.77999e-06
[py_interpret_to_execute_after_opt_a]: 9.87999e-06
[rewriter_after_opt_a]: 3.459e-05
[convert_after_rewriter]: 6.77002e-06
[order_py_execute_after_rewriter]: 4.77e-06
[mutable_eliminate]: 0.00054738
[jit_opt_b]: 5.135e-05, [1]
    [Cycle 1]: 4.413e-05, [2]
        [frontend_op_eliminate]: 1.482e-05
        [inline_after_opt_a]: 1.539e-05
[cconv]: 2.664e-05
[loop_unroll]: 0.00045744
[jit_opt_after_cconv]: 0.00013765, [1]
    [Cycle 1]: 0.00013083, [11]
        [c_1]: 2.148e-05
        [parameter_eliminate]: 2.91999e-06
        [updatestate_depend_eliminate]: 5.47001e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [cse]: 1.73e-05
        [call_graph_tuple_transform]: 1.826e-05
        [tuple_list_get_item_eliminator]: 5.09e-06
        [none_parameter_eliminate]: 1.35999e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 4.97e-06
[remove_dup_value]: 1.332e-05
[partial_unused_args_eliminate]: 2.32001e-06
[environ_conv]: 4.87998e-06
[add_recomputation]: 4.542e-05
[cse_after_recomputation]: 1.943e-05, [1]
    [Cycle 1]: 1.411e-05, [1]
        [cse]: 8.35999e-06
[auto_monad_reorder]: 1.36e-05
[get_jit_bprop_graph]: 1.96e-06
[rewriter_after_jit_bprop_graph]: 5.74e-06
[opt_after_jit_grad]: 0.00044974
[symbol_engine_optimizer]: 6.808e-05, [1]
    [Cycle 1]: 6.163e-05, [6]
        [build]: 2.65002e-06
        [elim_shapecalc]: 7.3e-06
        [elim_not_effective]: 1.14e-05
        [opt_reshape]: 5.34998e-06
        [fold_const_symbol]: 8.28999e-06
        [renormalize]: 4.50003e-07
[validate]: 3.243e-05
[backend_pass]: 8.29983e-07
[task_emit]: 0.00535616
[execute]: 6.33e-06
Sums
    bootstrap                            :     0.000251s :  1.88%
    type_inference                       :     0.004117s : 30.89%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000061s :  0.46%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.23%
    rewriter_before_opt_a                :     0.000069s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.43%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000620s :  4.65%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000009s :  0.07%
    jit_opt_a.renormalize                :     0.000621s :  4.66%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.14%
    jit_opt_a.cse                        :     0.000035s :  0.26%
    jit_opt_a.replace_applicator         :     0.000020s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.26%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000547s :  4.11%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000027s :  0.20%
    loop_unroll                          :     0.000457s :  3.43%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000045s :  0.34%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000450s :  3.37%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005356s : 40.19%
    execute                              :     0.000006s :  0.05%


Time group info:
------[substitution.]   0.000203    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.83% :     0.000006s :      3: substitution.graph_param_transform
83.71% :     0.000170s :      4: substitution.inline
 1.48% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.17% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.37% :     0.000005s :      2: substitution.replace_old_param
 5.95% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004063     2
81.45% :     0.003309s :      1: type_inference.infer
18.55% :     0.000754s :      1: type_inference.specialize
------[replace.]   0.000063     6
66.99% :     0.000042s :      4: replace.inline
33.01% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000178     6
93.81% :     0.000167s :      4: match.inline
 6.19% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.23% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
12.85% :     0.000016s :     10: predicate.addn_check_dump
 1.15% :     0.000001s :     10: predicate.addn_zero_filter
 1.57% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 0.99% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.10% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.70% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.26% :     0.000000s :      3: predicate.elim_not_effective
 0.48% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.05% :     0.000001s :     10: predicate.environ_get_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.06% :     0.000006s :     22: predicate.inline
 0.77% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.95% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.17% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.70% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.93% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 1.94% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.03% :     0.000003s :     15: predicate.partial_eliminate
 1.03% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.39% :     0.000002s :     10: predicate.reduce_eliminate
 1.26% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.54% :     0.000002s :     18: predicate.replace_applicator
 0.49% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.06% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000001s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.75% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.06% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.19% :     0.000003s :     16: predicate.switch_defer_inline
 2.01% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.63% :     0.000008s :     53: predicate.switch_simplify
 1.09% :     0.000001s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.20% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.05% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.23% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.21% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.40% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.51% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.50% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.35% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.53% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000588    12
54.14% :     0.000318s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.01% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
38.85% :     0.000228s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018759    76
 0.26% :     0.000048s :      1: add_recomputation
 0.34% :     0.000064s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.39% :     0.000260s :      1: bootstrap
 0.16% :     0.000029s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
27.60% :     0.005177s :      1: jit_opt_a
 0.75% :     0.000141s :      1: jit_opt_after_cconv
 0.29% :     0.000054s :      1: jit_opt_b
 2.49% :     0.000467s :      1: loop_unroll
 2.97% :     0.000557s :      1: mutable_eliminate
 4.39% :     0.000824s :     26: opt.transform.jit_opt_a
 0.25% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000023s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.44% :     0.000457s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000033s :      1: py_interpret_to_execute
 0.07% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.83% :     0.000342s :      1: renormalize.infer
 1.44% :     0.000271s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000038s :      1: rewriter_after_opt_a
 0.38% :     0.000071s :      1: rewriter_before_opt_a
 0.38% :     0.000071s :      1: symbol_engine_optimizer
28.62% :     0.005370s :      1: task_emit
22.02% :     0.004130s :      1: type_inference
 0.27% :     0.000051s :      1: validate


TotalTime = 0.0112523, [30]
[bootstrap]: 0.00026139
[type_inference]: 0.00392382
[event_method]: 1.639e-05
[auto_monad]: 6.202e-05
[graph_reusing]: 5.22e-06
[pre_auto_parallel]: 2.34001e-06
[py_interpret_to_execute]: 2.727e-05
[rewriter_before_opt_a]: 7.03e-05
[expand_dump_flag]: 2.73998e-06
[jit_opt_a]: 0.00485143, [2]
    [Cycle 1]: 0.00139715, [27]
        [switch_simplify]: 5.094e-05
        [loop_unroll]: 2.977e-05
        [a_1]: 0.00049565
        [with_stream_mark]: 1.66e-05
        [recompute_prepare]: 6.94001e-06
        [updatestate_depend_eliminate]: 3.23e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.70002e-06
        [parameter_eliminate]: 1.90001e-06
        [specialize_transform]: 5.77999e-06
        [updatestate_useless_node_eliminater]: 4.85999e-06
        [accelerated_algorithm]: 5.17e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 5.07e-06
        [merge_forward]: 4.07e-06
        [cell_reuse_recompute_pass]: 1.16002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.341e-05
        [j_node_and_user_rematch]: 8.84998e-06
        [meta_fg_expand]: 2.02999e-06
        [replace_old_param]: 8.33001e-06
        [inline_without_move]: 5.28002e-06
        [renormalize]: 0.00049104
        [add_forward_monad_depend]: 4.55001e-06
        [auto_monad_grad]: 2.05002e-06
        [auto_monad_eliminator]: 1.31e-05
        [cse]: 2.846e-05
        [replace_applicator]: 1.094e-05
    [Cycle 2]: 0.00031133, [27]
        [switch_simplify]: 5.14998e-06
        [loop_unroll]: 4.57e-06
        [a_1]: 8.734e-05
        [with_stream_mark]: 8.65999e-06
        [recompute_prepare]: 4.90001e-06
        [updatestate_depend_eliminate]: 2.39999e-06
        [updatestate_assign_eliminate]: 1.90001e-06
        [updatestate_loads_eliminate]: 1.96998e-06
        [parameter_eliminate]: 9.90025e-07
        [specialize_transform]: 5.06997e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 4.99998e-06
        [meta_shard_fg_expand]: 1.07998e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.51e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.162e-05
        [j_node_and_user_rematch]: 7.78001e-06
        [meta_fg_expand]: 1.39e-06
        [replace_old_param]: 7.7e-06
        [inline_without_move]: 4.50001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 4.80001e-06
        [cse]: 1.025e-05
        [replace_applicator]: 5.19998e-06
[py_interpret_to_execute_after_opt_a]: 9.72001e-06
[rewriter_after_opt_a]: 3.141e-05
[convert_after_rewriter]: 6.56e-06
[order_py_execute_after_rewriter]: 5.42999e-06
[mutable_eliminate]: 0.00051423
[jit_opt_b]: 5.13e-05, [1]
    [Cycle 1]: 4.418e-05, [2]
        [frontend_op_eliminate]: 1.664e-05
        [inline_after_opt_a]: 1.578e-05
[cconv]: 2.673e-05
[loop_unroll]: 0.0004139
[jit_opt_after_cconv]: 0.00013576, [1]
    [Cycle 1]: 0.00012903, [11]
        [c_1]: 1.994e-05
        [parameter_eliminate]: 2.52001e-06
        [updatestate_depend_eliminate]: 4.80001e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.13002e-06
        [cse]: 1.8e-05
        [call_graph_tuple_transform]: 1.79e-05
        [tuple_list_get_item_eliminator]: 5.07e-06
        [none_parameter_eliminate]: 1.50999e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 5.22e-06
[remove_dup_value]: 1.381e-05
[partial_unused_args_eliminate]: 2.54001e-06
[environ_conv]: 5.38002e-06
[add_recomputation]: 4.708e-05
[cse_after_recomputation]: 1.989e-05, [1]
    [Cycle 1]: 1.437e-05, [1]
        [cse]: 8.92e-06
[auto_monad_reorder]: 1.457e-05
[get_jit_bprop_graph]: 1.76e-06
[rewriter_after_jit_bprop_graph]: 4.79e-06
[opt_after_jit_grad]: 0.00046245
[symbol_engine_optimizer]: 7.15e-05, [1]
    [Cycle 1]: 6.543e-05, [6]
        [build]: 3.98999e-06
        [elim_shapecalc]: 7.23e-06
        [[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.189.957 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e93a0> should be wrapped by jit only once.
elim_not_effective]: 1.215e-05
        [opt_reshape]: 5.40001e-06
        [fold_const_symbol]: 8.62e-06
        [renormalize]: 4.00003e-07
[validate]: 3.222e-05
Sums
    bootstrap                            :     0.000261s :  3.46%
    type_inference                       :     0.003924s : 51.99%
    event_method                         :     0.000016s :  0.22%
    auto_monad                           :     0.000062s :  0.82%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000027s :  0.36%
    rewriter_before_opt_a                :     0.000070s :  0.93%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000056s :  0.74%
    jit_opt_a.loop_unroll                :     0.000034s :  0.46%
    jit_opt_a.a_1                        :     0.000583s :  7.72%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000007s :  0.09%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000491s :  6.51%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.24%
    jit_opt_a.cse                        :     0.000039s :  0.51%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.13%
    rewriter_after_opt_a                 :     0.000031s :  0.42%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000514s :  6.81%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.21%
    cconv                                :     0.000027s :  0.35%
    loop_unroll                          :     0.000414s :  5.48%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000018s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.18%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000047s :  0.62%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000015s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000462s :  6.13%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000032s :  0.43%


Time group info:
------[substitution.]   0.000180    23
 1.17% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.68% :     0.000005s :      3: substitution.graph_param_transform
82.87% :     0.000149s :      4: substitution.inline
 1.50% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.25% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.61% :     0.000005s :      2: substitution.replace_old_param
 6.24% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003864     2
81.37% :     0.003145s :      1: type_inference.infer
18.63% :     0.000720s :      1: type_inference.specialize
------[replace.]   0.000060     6
67.04% :     0.000040s :      4: replace.inline
32.96% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000157     6
93.52% :     0.000147s :      4: match.inline
 6.48% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.62% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.41% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000001s :     10: predicate.addn_zero_filter
 2.09% :     0.000002s :     10: predicate.arithmetic_simplify
 1.13% :     0.000001s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.72% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.41% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.47% :     0.000001s :      3: predicate.graph_param_transform
 5.67% :     0.000007s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.28% :     0.000001s :      6: predicate.less_batch_normalization
 1.39% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.54% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.89% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.54% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.53% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.48% :     0.000002s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.95% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.41% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.48% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.05% :     0.000009s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.25% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.53% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.71% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000486    12
51.53% :     0.000251s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.91% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.56% :     0.000207s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012589    72
 0.40% :     0.000050s :      1: add_recomputation
 0.52% :     0.000066s :      1: auto_monad
 0.14% :     0.000017s :      1: auto_monad_reorder
 2.20% :     0.000277s :      1: bootstrap
 0.23% :     0.000030s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
38.56% :     0.004854s :      1: jit_opt_a
 1.10% :     0.000138s :      1: jit_opt_after_cconv
 0.43% :     0.000054s :      1: jit_opt_b
 3.35% :     0.000421s :      1: loop_unroll
 4.16% :     0.000523s :      1: mutable_eliminate
 6.22% :     0.000784s :     26: opt.transform.jit_opt_a
 0.36% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.74% :     0.000471s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.04% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000030s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000016s :      1: remove_dup_value
 2.02% :     0.000254s :      1: renormalize.infer
 1.83% :     0.000230s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000036s :      1: rewriter_after_opt_a
 0.58% :     0.000073s :      1: rewriter_before_opt_a
 0.59% :     0.000074s :      1: symbol_engine_optimizer
31.27% :     0.003936s :      1: type_inference

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.190.662 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0226916, [33]
[bootstrap]: 0.00021083
[type_inference]: 0.0109112
[event_method]: 4.787e-05
[auto_monad]: 9.41e-05
[graph_reusing]: 6.85998e-06
[pre_auto_parallel]: 2.86999e-06
[py_interpret_to_execute]: 2.599e-05
[rewriter_before_opt_a]: 9.228e-05
[expand_dump_flag]: 3.01001e-06
[jit_opt_a]: 0.00911921, [2]
    [Cycle 1]: 0.00482793, [27]
        [switch_simplify]: 8.181e-05
        [loop_unroll]: 3.878e-05
        [a_1]: 0.00090223
        [with_stream_mark]: 3.011e-05
        [recompute_prepare]: 2.037e-05
        [updatestate_depend_eliminate]: 8.09002e-06
        [updatestate_assign_eliminate]: 6.38e-06
        [updatestate_loads_eliminate]: 6.02001e-06
        [parameter_eliminate]: 2.61e-06
        [specialize_transform]: 1.311e-05
        [updatestate_useless_node_eliminater]: 1.177e-05
        [accelerated_algorithm]: 1.307e-05
        [meta_shard_fg_expand]: 3.98001e-06
        [get_grad_eliminate_]: 1.227e-05
        [merge_forward]: 8.57e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.737e-05
        [j_node_and_user_rematch]: 2.319e-05
        [meta_fg_expand]: 0.00163468
        [replace_old_param]: 5.643e-05
        [inline_without_move]: 5.204e-05
        [renormalize]: 0.00160525
        [add_forward_monad_depend]: 5.30999e-06
        [auto_monad_grad]: 2.71e-06
        [auto_monad_eliminator]: 1.155e-05
        [cse]: 2.441e-05
        [replace_applicator]: 1.222e-05
    [Cycle 2]: 0.00028716, [27]
        [switch_simplify]: 5.06002e-06
        [loop_unroll]: 4.17998e-06
        [a_1]: 4.328e-05
        [with_stream_mark]: 9.71e-06
        [recompute_prepare]: 3.78001e-06
        [updatestate_depend_eliminate]: 2.15002e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [parameter_eliminate]: 1.32999e-06
        [specialize_transform]: 3.68e-06
        [updatestate_useless_node_eliminater]: 3.39001e-06
        [accelerated_algorithm]: 3.65e-06
        [meta_shard_fg_expand]: 1.81003e-06
        [get_grad_eliminate_]: 3.25e-06
        [merge_forward]: 2.64999e-06
        [cell_reuse_recompute_pass]: 2.49999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.135e-05
        [j_node_and_user_rematch]: 5.86e-06
        [meta_fg_expand]: 3.606e-05
        [replace_old_param]: 3.93001e-06
        [inline_without_move]: 3.4e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 4.38999e-06
        [cse]: 9.64e-06
        [replace_applicator]: 3.82002e-06
[py_interpret_to_execute_after_opt_a]: 9.64e-06
[rewriter_after_opt_a]: 3.201e-05
[convert_after_rewriter]: 5.94999e-06
[order_py_execute_after_rewriter]: 4.2e-06
[mutable_eliminate]: 0.00062223
[jit_opt_b]: 5.019e-05, [1]
    [Cycle 1]: 4.238e-05, [2]
        [frontend_op_eliminate]: 2.023e-05
        [inline_after_opt_a]: 1.058e-05
[cconv]: 2.15e-05
[loop_unroll]: 0.00042875
[jit_opt_after_cconv]: 0.00011865, [1]
    [Cycle 1]: 0.00011156, [11]
        [c_1]: 1.181e-05
        [parameter_eliminate]: 2.53003e-06
        [updatestate_depend_eliminate]: 5.00001e-06
        [updatestate_assign_eliminate]: 1.87999e-06
        [updatestate_loads_eliminate]: 1.55999e-06
        [cse]: 1.659e-05
        [call_graph_tuple_transform]: 1.407e-05
        [tuple_list_get_item_eliminator]: 3.73001e-06
        [none_parameter_eliminate]: 1.75001e-06
        [renormalize]: 6.79982e-07
        [switch_simplify]: 3.61999e-06
[remove_dup_value]: 1.273e-05
[partial_unused_args_eliminate]: 2.26e-06
[environ_conv]: 5.17e-06
[add_recomputation]: 3.346e-05
[cse_after_recomputation]: 2.013e-05, [1]
    [Cycle 1]: 1.437e-05, [1]
        [cse]: 8.33999e-06
[auto_monad_reorder]: 1.322e-05
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 4.68001e-06
[opt_after_jit_grad]: 0.00050479
[symbol_engine_optimizer]: 6.46e-05, [1]
    [Cycle 1]: 5.864e-05, [6]
        [build]: 3.11999e-06
        [elim_shapecalc]: 6.23e-06
        [elim_not_effective]: 9.36998e-06
        [opt_reshape]: 4.58001e-06
        [fold_const_symbol]: 6.94001e-06
        [renormalize]: 4.19997e-07
[validate]: 3.076e-05
[backend_pass]: 9.39996e-07
[task_emit]: 2.214e-05
[execute]: 1.58002e-06
Sums
    bootstrap                            :     0.000211s :  1.17%
    type_inference                       :     0.010911s : 60.36%
    event_method                         :     0.000048s :  0.26%
    auto_monad                           :     0.000094s :  0.52%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.14%
    rewriter_before_opt_a                :     0.000092s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000087s :  0.48%
    jit_opt_a.loop_unroll                :     0.000043s :  0.24%
    jit_opt_a.a_1                        :     0.000946s :  5.23%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000024s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.09%
    jit_opt_a.merge_forward              :     0.000011s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001671s :  9.24%
    jit_opt_a.replace_old_param          :     0.000060s :  0.33%
    jit_opt_a.inline_without_move        :     0.000055s :  0.31%
    jit_opt_a.renormalize                :     0.001605s :  8.88%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.09%
    jit_opt_a.cse                        :     0.000034s :  0.19%
    jit_opt_a.replace_applicator         :     0.000016s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.05%
    rewriter_after_opt_a                 :     0.000032s :  0.18%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000622s :  3.44%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.06%
    cconv                                :     0.000021s :  0.12%
    loop_unroll                          :     0.000429s :  2.37%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000013s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000033s :  0.19%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000505s :  2.79%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000007s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.17%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.000022s :  0.12%
    execute                              :     0.000002s :  0.01%


Time group info:
------[substitution.]   0.000290    50
 0.52% :     0.000002s :      1: substitution.elim_not_effective
 0.33% :     0.000001s :      1: substitution.fold_const_symbol
 1.41% :     0.000004s :      1: substitution.graph_param_transform
67.40% :     0.000196s :      6: substitution.inline
 6.09% :     0.000018s :      2: substitution.inline_without_move
 2.24% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.53% :     0.000004s :      2: substitution.minmaximum_grad
 1.64% :     0.000005s :      3: substitution.partial_eliminate
 2.53% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.11% :     0.000006s :      2: substitution.replace_applicator
 1.75% :     0.000005s :      5: substitution.replace_old_param
 1.04% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.36% :     0.000010s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.86% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 6.19% :     0.000018s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010837     2
88.49% :     0.009590s :      1: type_inference.infer
11.51% :     0.001247s :      1: type_inference.specialize
------[replace.]   0.000068     8
78.01% :     0.000053s :      6: replace.inline
21.99% :     0.000015s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000204     8
94.09% :     0.000192s :      6: match.inline
 5.91% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000174  1109
 1.85% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.68% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.31% :     0.000002s :     18: predicate.addn_check_dump
 1.64% :     0.000003s :     18: predicate.addn_zero_filter
 1.92% :     0.000003s :     18: predicate.arithmetic_simplify
 1.40% :     0.000002s :     18: predicate.cast_eliminate
 0.16% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.28% :     0.000002s :     18: predicate.compare_switch_simplify
 1.48% :     0.000003s :     18: predicate.depend_value_elim
 1.40% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.87% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.56% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.10% :     0.000000s :      1: predicate.elim_not_effective
 0.33% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.52% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.28% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.55% :     0.000003s :     18: predicate.environ_get_eliminate
 1.31% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.12% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.28% :     0.000007s :     28: predicate.inline
 3.19% :     0.000006s :     33: predicate.inline_without_move
 0.60% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :     12: predicate.less_batch_normalization
 1.49% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.65% :     0.000003s :     21: predicate.load_eliminater
 0.70% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.47% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.80% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.35% :     0.000002s :     18: predicate.merge_addn
 1.27% :     0.000002s :     18: predicate.minmaximum_grad
 0.67% :     0.000001s :      1: predicate.mutable_eliminate
 0.17% :     0.000000s :      1: predicate.opt_reshape
 2.01% :     0.000003s :     21: predicate.partial_eliminate
 1.43% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.94% :     0.000003s :     18: predicate.reduce_eliminate
 1.64% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.41% :     0.000002s :     22: predicate.replace_applicator
 1.69% :     0.000003s :     33: predicate.replace_old_param
 0.09% :     0.000000s :      1: predicate.reset_defer_inline
 1.56% :     0.000003s :     18: predicate.reshape_eliminate
 1.54% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.47% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.56% :     0.000003s :     18: predicate.same_eliminate
 0.60% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.38% :     0.000001s :      2: predicate.special_op_eliminate
 1.16% :     0.000002s :     12: predicate.specialize_transform
 1.61% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.56% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.14% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.57% :     0.000004s :     26: predicate.switch_defer_inline
 2.20% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.87% :     0.000012s :     71: predicate.switch_simplify
 1.67% :     0.000003s :     18: predicate.tile_eliminate
 1.50% :     0.000003s :     18: predicate.transpose_eliminate
 1.66% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 3.54% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.55% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000003s :     18: predicate.value_based_eliminate
 0.11% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.24% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001220    22
71.40% :     0.000871s :     13: func_graph_cloner_run.FuncGraphClonerGraph
28.60% :     0.000349s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025700    76
 0.14% :     0.000036s :      1: add_recomputation
 0.38% :     0.000099s :      1: auto_monad
 0.06% :     0.000016s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.86% :     0.000221s :      1: bootstrap
 0.09% :     0.000024s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.20% :     0.000052s :      1: event_method
 0.01% :     0.000004s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
35.50% :     0.009123s :      1: jit_opt_a
 0.47% :     0.000121s :      1: jit_opt_after_cconv
 0.21% :     0.000053s :      1: jit_opt_b
 1.70% :     0.000436s :      1: loop_unroll
 2.45% :     0.000628s :      1: mutable_eliminate
 5.18% :     0.001330s :     26: opt.transform.jit_opt_a
 0.12% :     0.000030s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000024s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000023s :      4: opt.transform.symbol_engine_opt
 2.00% :     0.000513s :      1: opt_after_jit_grad
 0.02% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000028s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000015s :      1: remove_dup_value
 4.64% :     0.001192s :      1: renormalize.infer
 1.58% :     0.000405s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000035s :      1: rewriter_after_opt_a
 0.37% :     0.000095s :      1: rewriter_before_opt_a
 0.26% :     0.000067s :      1: symbol_engine_optimizer
 0.10% :     0.000026s :      1: task_emit
42.51% :     0.010925s :      1: type_inference
 0.18% :     0.000047s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.204.924 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c36b280> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.205.667 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0176923, [33]
[bootstrap]: 0.00025228
[type_inference]: 0.00387734
[event_method]: 1.572e-05
[auto_monad]: 6.127e-05
[graph_reusing]: 5.76e-06
[pre_auto_parallel]: 2.16e-06
[py_interpret_to_execute]: 2.845e-05
[rewriter_before_opt_a]: 6.506e-05
[expand_dump_flag]: 3.08e-06
[jit_opt_a]: 0.00494724, [2]
    [Cycle 1]: 0.00150885, [27]
        [switch_simplify]: 4.837e-05
        [loop_unroll]: 2.965e-05
        [a_1]: 0.00050492
        [with_stream_mark]: 1.713e-05
        [recompute_prepare]: 8.97e-06
        [updatestate_depend_eliminate]: 3.28e-06
        [updatestate_assign_eliminate]: 3.12002e-06
        [updatestate_loads_eliminate]: 2.62001e-06
        [parameter_eliminate]: 2.28998e-06
        [specialize_transform]: 6.49999e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 5.46e-06
        [meta_shard_fg_expand]: 2.34999e-06
        [get_grad_eliminate_]: 5.33002e-06
        [merge_forward]: 3.4e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.286e-05
        [j_node_and_user_rematch]: 9.10999e-06
        [meta_fg_expand]: 2.61999e-06
        [replace_old_param]: 8.12e-06
        [inline_without_move]: 5.21998e-06
        [renormalize]: 0.00059993
        [add_forward_monad_depend]: 5.12e-06
        [auto_monad_grad]: 2.06e-06
        [auto_monad_eliminator]: 1.35e-05
        [cse]: 2.045e-05
        [replace_applicator]: 1.313e-05
    [Cycle 2]: 0.00040309, [27]
        [switch_simplify]: 5.36998e-06
        [loop_unroll]: 4.83001e-06
        [a_1]: 8.741e-05
        [with_stream_mark]: 8.74003e-06
        [recompute_prepare]: 4.99998e-06
        [updatestate_depend_eliminate]: 2.39001e-06
        [updatestate_assign_eliminate]: 1.98002e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 5.66003e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 4.95999e-06
        [meta_shard_fg_expand]: 1.24e-06
        [get_grad_eliminate_]: 4.40999e-06
        [merge_forward]: 3.31999e-06
        [cell_reuse_recompute_pass]: 2.16e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.311e-05
        [j_node_and_user_rematch]: 7.40003e-06
        [meta_fg_expand]: 1.69e-06
        [replace_old_param]: 7.19001e-06
        [inline_without_move]: 4.82e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.18001e-06
        [auto_monad_grad]: 7.40023e-07
        [auto_monad_eliminator]: 5.77001e-06
        [cse]: 1.171e-05
        [replace_applicator]: 4.90001e-06
[py_interpret_to_execute_after_opt_a]: 9.56e-06
[rewriter_after_opt_a]: 3.109e-05
[convert_after_rewriter]: 7.33e-06
[order_py_execute_after_rewriter]: 5.15999e-06
[mutable_eliminate]: 0.00055543
[jit_opt_b]: 5.156e-05, [1]
    [Cycle 1]: 4.292e-05, [2]
        [frontend_op_eliminate]: 1.653e-05
        [inline_after_opt_a]: 1.442e-05
[cconv]: 2.516e-05
[loop_unroll]: 0.00042113
[jit_opt_after_cconv]: 0.00013478, [1]
    [Cycle 1]: 0.00012863, [11]
        [c_1]: 2.056e-05
        [parameter_eliminate]: 2.56e-06
        [updatestate_depend_eliminate]: 5.87999e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 1.679e-05
        [call_graph_tuple_transform]: 1.843e-05
        [tuple_list_get_item_eliminator]: 5.09e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 4.84998e-06
[remove_dup_value]: 1.253e-05
[partial_unused_args_eliminate]: 2.19999e-06
[environ_conv]: 5.30001e-06
[add_recomputation]: 4.874e-05
[cse_after_recomputation]: 1.979e-05, [1]
    [Cycle 1]: 1.447e-05, [1]
        [cse]: 8.97999e-06
[auto_monad_reorder]: 1.389e-05
[get_jit_bprop_graph]: 1.68002e-06
[rewriter_after_jit_bprop_graph]: 5.52001e-06
[opt_after_jit_grad]: 0.00044999
[symbol_engine_optimizer]: 7.168e-05, [1]
    [Cycle 1]: 6.524e-05, [6]
        [build]: 3.3e-06
        [elim_shapecalc]: 7.13e-06
        [elim_not_effective]: 1.271e-05
        [opt_reshape]: 5.42999e-06
        [fold_const_symbol]: 8.72998e-06
        [renormalize]: 3.80009e-07
[validate]: 3.481e-05
[backend_pass]: 8.30012e-07
[task_emit]: 0.00634094
[execute]: 8.50001e-06
Sums
    bootstrap                            :     0.000252s :  1.80%
    type_inference                       :     0.003877s : 27.71%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000061s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.20%
    rewriter_before_opt_a                :     0.000065s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.38%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000592s :  4.23%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000600s :  4.29%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.14%
    jit_opt_a.cse                        :     0.000032s :  0.23%
    jit_opt_a.replace_applicator         :     0.000018s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000031s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000555s :  3.97%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000025s :  0.18%
    loop_unroll                          :     0.000421s :  3.01%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000049s :  0.35%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000450s :  3.22%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006341s : 45.31%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000188    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.80% :     0.000002s :      2: substitution.fold_const_symbol
 3.05% :     0.000006s :      3: substitution.graph_param_transform
81.33% :     0.000153s :      4: substitution.inline
 1.73% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.46% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.95% :     0.000004s :      2: substitution.replace_old_param
 7.66% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003829     2
81.61% :     0.003125s :      1: type_inference.infer
18.39% :     0.000704s :      1: type_inference.specialize
------[replace.]   0.000063     6
67.12% :     0.000042s :      4: replace.inline
32.88% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000164     6
91.89% :     0.000151s :      4: match.inline
 8.11% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.26% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.02% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     10: predicate.addn_check_dump
 1.44% :     0.000002s :     10: predicate.addn_zero_filter
 2.50% :     0.000003s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.61% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     10: predicate.compare_switch_simplify
 1.53% :     0.000002s :     10: predicate.depend_value_elim
 1.27% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.92% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.67% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.34% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.00% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.08% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.71% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000002s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.71% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.96% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.54% :     0.000002s :     10: predicate.reshape_eliminate
 1.37% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.18% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.27% :     0.000008s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.36% :     0.000002s :     10: predicate.transpose_eliminate
 1.34% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.75% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.00% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000491    12
47.43% :     0.000233s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.81% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.76% :     0.000230s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019195    76
 0.27% :     0.000052s :      1: add_recomputation
 0.34% :     0.000065s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.36% :     0.000262s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.79% :     0.004950s :      1: jit_opt_a
 0.72% :     0.000137s :      1: jit_opt_after_cconv
 0.28% :     0.000054s :      1: jit_opt_b
 2.23% :     0.000429s :      1: loop_unroll
 2.94% :     0.000565s :      1: mutable_eliminate
 4.14% :     0.000796s :     26: opt.transform.jit_opt_a
 0.24% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.39% :     0.000458s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000031s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.70% :     0.000327s :      1: renormalize.infer
 1.39% :     0.000266s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000034s :      1: rewriter_after_opt_a
 0.35% :     0.000068s :      1: rewriter_before_opt_a
 0.39% :     0.000074s :      1: symbol_engine_optimizer
33.11% :     0.006355s :      1: task_emit
20.25% :     0.003888s :      1: type_inference
 0.28% :     0.000054s :      1: validate


TotalTime = 0.0109767, [30]
[bootstrap]: 0.00026038
[type_inference]: 0.00393516
[event_method]: 1.634e-05
[auto_monad]: 6.106e-05
[graph_reusing]: 5.19e-06
[pre_auto_parallel]: 2.03997e-06
[py_interpret_to_execute]: 2.675e-05
[rewriter_before_opt_a]: 6.946e-05
[expand_dump_flag]: 2.66e-06
[jit_opt_a]: 0.00462435, [2]
    [Cycle 1]: 0.00132029, [27]
        [switch_simplify]: 5.069e-05
        [loop_unroll]: 2.84e-05
        [a_1]: 0.00049066
        [with_stream_mark]: 1.495e-05
        [recompute_prepare]: 7.48e-06
        [updatestate_depend_eliminate]: 3.75998e-06
        [updatestate_assign_eliminate]: 3.14999e-06
        [updatestate_loads_eliminate]: 2.61999e-06
        [parameter_eliminate]: 1.70001e-06
        [specialize_transform]: 6.45997e-06
        [updatestate_useless_node_eliminater]: 5.55001e-06
        [accelerated_algorithm]: 6.06e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 5.19e-06
        [merge_forward]: 3.41999e-06
        [cell_reuse_recompute_pass]: 9.80013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.357e-05
        [j_node_and_user_rematch]: 9.15999e-06
        [meta_fg_expand]: 2.12999e-06
        [replace_old_param]: 8.33001e-06
        [inline_without_move]: 5.55001e-06
        [renormalize]: 0.00042739
        [add_forward_monad_depend]: 4.75999e-06
        [auto_monad_grad]: 1.96e-06
        [auto_monad_eliminator]: 1.24e-05
        [cse]: 2.282e-05
        [replace_applicator]: 1.081e-05
    [Cycle 2]: 0.00030948, [27]
        [switch_simplify]: 5.25999e-06
        [loop_unroll]: 4.82998e-06
        [a_1]: 8.785e-05
        [with_stream_mark]: 7.58999e-06
        [recompute_prepare]: 4.85999e-06
        [updatestate_depend_eliminate]: 2.49001e-06
        [updatestate_assign_eliminate]: 2.00002e-06
        [updatestate_loads_eliminate]: 1.95001e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 4.90001e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 4.99e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.56002e-06
        [merge_forward]: 2.18998e-06
        [cell_reuse_recompute_pass]: 1.07998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.607e-05
        [j_node_and_user_rematch]: 7.73001e-06
        [meta_fg_expand]: 1.32999e-06
        [replace_old_param]: 6.93e-06
        [inline_without_move]: 4.60001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.05999e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 4.70001e-06
        [cse]: 9.96e-06
        [replace_applicator]: 6.58e-06
[py_interpret_to_execute_after_opt_a]: 7.79002e-06
[rewriter_after_opt_a]: 3.236e-05
[convert_after_rewriter]: 7.08e-06
[order_py_execute_after_rewriter]: 5.59998e-06
[mutable_eliminate]: 0.0004606
[jit_opt_b]: 4.941e-05, [1]
    [Cycle 1]: 4.252e-05, [2]
        [frontend_op_eliminate]: 1.575e-05
        [inline_after_opt_a]: 1.522e-05
[cconv]: 2.327e-05
[loop_unroll]: 0.00042769
[jit_opt_after_cconv]: 0.00013589, [1]
    [Cycle 1]: 0.00012949, [11]
        [c_1]: 1.977e-05
        [parameter_eliminate]: 2.62001e-06
        [updatestate_depend_eliminate]: 5.22e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 1.591e-05
        [call_graph_tuple_transform]: 1.842e-05
        [tuple_list_get_item_eliminator]: 5.35999e-06
        [none_parameter_eliminate]: 1.34e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 5.59998e-06
[remove_dup_value]: 1.41e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 5.12999e-06
[add_recomputation]: 4.408e-05
[cse_after_recomputation]: 2.035e-05, [1]
    [Cycle 1]: 1.47e-05, [1]
        [cse]: 8.89e-06
[auto_monad_reorder]: 1.406e-05
[get_jit_bprop_graph]: 1.54e-06
[rewriter_after_jit_bprop_graph]: 4.80001e-06
[opt_after_jit_grad]: 0.00046896
[symbol_engine_optimizer]: 6.84e-05, [1]
    [Cycle 1]: 6.215e-05, [6]
        [build]: 2.94999e-06
        [elim_shapecalc]: 7.48999e-06
        [elim_not_effective]: 1.143e-05
        [opt_reshape]: 5.22e-06
        [fold_const_symbol]: 8e-06
        [renormalize]: 5.19998e-07
[validate]: 2.815e-05
Sums
    bootstrap                            :     0.000260s :  3.50%
    type_inference                       :     0.003935s : 52.93%
    event_method                         :     0.000016s :  0.22%
    auto_monad                           :     0.000061s :  0.82%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000027s :  0.36%
    rewriter_before_opt_a                :     0.000069s :  0.93%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000056s :  0.75%
    jit_opt_a.loop_unroll                :     0.000033s :  0.45%
    jit_opt_a.a_1                        :     0.000579s :  7.78%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.40%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000427s :  5.75%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.23%
    jit_opt_a.cse                        :     0.000033s :  0.44%
    jit_opt_a.replace_applicator         :     0.000017s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.10%
    rewriter_after_opt_a                 :     0.000032s :  0.44%
    convert_after_rewriter               :     0.000007s :  0.10%
    order_py_execute_after_rewriter      :     0.000006s :  0.08%
    mutable_eliminate                    :     0.000461s :  6.20%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.20%
    cconv                                :     0.000023s :  0.31%
    loop_unroll                          :     0.000428s :  5.75%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.25%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.08%
    remove_dup_value                     :     0.000014s :  0.19%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000044s :  0.59%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000014s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000469s :  6.31%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000028s :  0.38%


Time group info:
------[substitution.]   0.000172    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.89% :     0.000005s :      3: substitution.graph_param_transform
80.25% :     0.000138s :      4: substitution.inline
 1.77% :     0.000003s :      4: substitution.j_node_and_user_rematch
 5.09% :     0.000009s :      4: substitution.remove_not_recompute_node
 1.89% :     0.000003s :      2: substitution.replace_old_param
 6.42% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003878     2
81.13% :     0.003146s :      1: type_inference.infer
18.87% :     0.000732s :      1: type_inference.specialize
------[replace.]   0.000058     6
67.03% :     0.000039s :      4: replace.inline
32.97% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000146     6
93.10% :     0.000136s :      4: match.inline
 6.90% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.83% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.23% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 1.92% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.58% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.33% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 5.34% :     0.000006s :     22: predicate.inline
 1.02% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.97% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.16% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.16% :     0.000001s :     10: predicate.minmaximum_grad
 1.70% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.42% :     0.000002s :     10: predicate.reduce_eliminate
 1.42% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.79% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.27% :     0.000001s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.22% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.72% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.46% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.87% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.58% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.88% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000483    12
51.92% :     0.000251s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.38% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.70% :     0.000206s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012255    72
 0.38% :     0.000047s :      1: add_recomputation
 0.52% :     0.000064s :      1: auto_monad
 0.14% :     0.000017s :      1: auto_monad_reorder
 2.23% :     0.000273s :      1: bootstrap
 0.21% :     0.000026s :      1: cconv
 0.08% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
37.75% :     0.004627s :      1: jit_opt_a
 1.13% :     0.000139s :      1: jit_opt_after_cconv
 0.43% :     0.000052s :      1: jit_opt_b
 3.55% :     0.000436s :      1: loop_unroll
 3.83% :     0.000469s :      1: mutable_eliminate
 6.41% :     0.000786s :     26: opt.transform.jit_opt_a
 0.37% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000029s :      4: opt.transform.symbol_engine_opt
 3.89% :     0.000477s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000029s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.14% :     0.000017s :      1: remove_dup_value
 1.71% :     0.000209s :      1: renormalize.infer
 1.72% :     0.000211s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.29% :     0.000035s :      1: rewriter_after_opt_a
 0.59% :     0.000072s :      1: rewriter_before_opt_a
 0.58% :     0.000071s :      1: symbol_engine_optimizer
32.20% :     0.003946s :      1: type_inference


TotalTime = 0.0332238, [33]
[bootstrap]: 0.00020683
[type_inference]: 0.00975808
[event_method]: 4.966e-05
[auto_monad]: 8.802e-05
[graph_reusing]: 6.61e-06
[pre_auto_parallel]: 2.14999e-06
[py_interpret_to_execute]: 2.369e-05
[rewriter_before_opt_a]: 8.743e-05
[expand_dump_flag]: 2.71e-06
[jit_opt_a]: 0.0123267, [3]
    [Cycle 1]: 0.00614435, [27]
        [switch_simplify]: 5.941e-05
        [loop_unroll]: 3.791e-05
        [a_1]: 0.00087371
        [with_stream_mark]: 2.372e-05
        [recompute_prepare]: 1.868e-05
        [updatestate_depend_eliminate]: 7.65e-06
        [updatestate_assign_eliminate]: 5.59e-06
        [updatestate_loads_eliminate]: 5.35999e-06
        [parameter_eliminate]: 2.22999e-06
        [specialize_transform]: 1.373e-05
        [updatestate_useless_node_eliminater]: 1.296e-05
        [accelerated_algorithm]: 1.415e-05
        [meta_shard_fg_expand]: 2.63998e-06
        [get_grad_eliminate_]: 1.309e-05
        [merge_forward]: 7.68999e-06
        [cell_reuse_recompute_pass]: 1.00999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.474e-05
        [j_node_and_user_rematch]: 2.317e-05
        [meta_fg_expand]: 0.00156113
        [replace_old_param]: 5.404e-05
        [inline_without_move]: 5.305e-05
        [renormalize]: 0.00279425
        [add_forward_monad_depend]: 1.225e-05
        [auto_monad_grad]: 5.87999e-06
        [auto_monad_eliminator]: 5.548e-05
        [cse]: 0.00015956
        [replace_applicator]: 7.727e-05
    [Cycle 2]: 0.00233735, [27]
        [switch_simplify]: 4.064e-05
        [loop_unroll]: 3.783e-05
        [a_1]: 0.00119955
        [with_stream_mark]: 1.813e-05
        [recompute_prepare]: 9.84001e-06
        [updatestate_depend_eliminate]: 4.90001e-06
        [updatestate_assign_eliminate]: 4.10998e-06
        [updatestate_loads_eliminate]: 3.83001e-06
        [parameter_eliminate]: 2.26e-06
        [specialize_transform]: 7.65e-06
        [updatestate_useless_node_eliminater]: 6.89001e-06
        [accelerated_algorithm]: 7.97e-06
        [meta_shard_fg_expand]: 2.59001e-06
        [get_grad_eliminate_]: 6.86999e-06
        [merge_forward]: 4.48999e-06
        [cell_reuse_recompute_pass]: 1.30001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.777e-05
        [j_node_and_user_rematch]: 1.176e-05
        [meta_fg_expand]: 5.578e-05
        [replace_old_param]: 1.257e-05
        [inline_without_move]: 7.30998e-06
        [renormalize]: 0.00064219
        [add_forward_monad_depend]: 5.86e-06
        [auto_monad_grad]: 2.32999e-06
        [auto_monad_eliminator]: 1.773e-05
        [cse]: 2.613e-05
        [replace_applicator]: 1.711e-05
    [Cycle 3]: 0.00045663, [27]
        [switch_simplify]: 9.42999e-06
        [loop_unroll]: 7.5e-06
        [a_1]: 0.00015403
        [with_stream_mark]: 1.072e-05
        [recompute_prepare]: 6.86001e-06
        [updatestate_depend_eliminate]: 3.93001e-06
        [updatestate_assign_eliminate]: 3.32002e-06
        [updatestate_loads_eliminate]: 3.22002e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 7.65e-06
        [updatestate_useless_node_eliminater]: 7.38e-06
        [accelerated_algorithm]: 7.68999e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 7.03e-06
        [merge_forward]: 3.63e-06
        [cell_reuse_recompute_pass]: 3.41999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.826e-05
        [j_node_and_user_rematch]: 1.237e-05
        [meta_fg_expand]: 2.56e-06
        [replace_old_param]: 9.87999e-06
        [inline_without_move]: 6.69999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 2.09e-06
        [auto_monad_grad]: 1.20999e-06
        [auto_monad_eliminator]: 8.32e-06
        [cse]: 1.88e-05
        [replace_applicator]: 6.72002e-06
[py_interpret_to_execute_after_opt_a]: 1.513e-05
[rewriter_after_opt_a]: 4.611e-05
[convert_after_rewriter]: 9.50001e-06
[order_py_execute_after_rewriter]: 6.17001e-06
[mutable_eliminate]: 0.00077658
[jit_opt_b]: 6.42e-05, [1]
    [Cycle 1]: 5.579e-05, [2]
        [frontend_op_eliminate]: 2.279e-05
        [inline_after_opt_a]: 2.108e-05
[cconv]: 2.448e-05
[loop_unroll]: 0.00044734
[jit_opt_after_cconv]: 0.00017108, [1]
    [Cycle 1]: 0.00016467, [11]
        [c_1]: 3.049e-05
        [parameter_eliminate]: 2.41998e-06
        [updatestate_depend_eliminate]: 6.71999e-06
        [updatestate_assign_eliminate]: 3.36001e-06
        [updatestate_loads_eliminate]: 3.03e-06
        [cse]: 2.888e-05
        [call_graph_tuple_transform]: 2.314e-05
        [tuple_list_get_item_eliminator]: 7.33e-06
        [none_parameter_eliminate]: 1.38002e-06
        [renormalize]: 5.10016e-07
        [switch_simplify]: 7.40998e-06
[remove_dup_value]: 1.709e-05
[partial_unused_args_eliminate]: 2.39001e-06
[environ_conv]: 7.38e-06
[add_recomputation]: 5.704e-05
[cse_after_recomputation]: 3.164e-05, [1]
    [Cycle 1]: 2.542e-05, [1]
        [cse]: 1.865e-05
[auto_monad_reorder]: 2.216e-05
[get_jit_bprop_graph]: 2.20002e-06
[rewriter_after_jit_bprop_graph]: 6.23e-06
[opt_after_jit_grad]: 0.00046615
[symbol_engine_optimizer]: 8.633e-05, [1]
    [Cycle 1]: 7.976e-05, [6]
        [build]: 4.01001e-06
        [elim_shapecalc]: 9.89001e-06
        [elim_not_effective]: 1.604e-05
        [opt_reshape]: 8.03999e-06
        [fold_const_symbol]: 1.265e-05
        [renormalize]: 9.20001e-07
[validate]: 4.36e-05
[backend_pass]: 8.60018e-07
[task_emit]: 0.00815083
[execute]: 7.7e-06
Sums
    bootstrap                            :     0.000207s :  0.71%
    type_inference                       :     0.009758s : 33.67%
    event_method                         :     0.000050s :  0.17%
    auto_monad                           :     0.000088s :  0.30%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000024s :  0.08%
    rewriter_before_opt_a                :     0.000087s :  0.30%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000109s :  0.38%
    jit_opt_a.loop_unroll                :     0.000083s :  0.29%
    jit_opt_a.a_1                        :     0.002227s :  7.68%
    jit_opt_a.with_stream_mark           :     0.000053s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000035s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.02%
    jit_opt_a.specialize_transform       :     0.000029s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000027s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000030s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000027s :  0.09%
    jit_opt_a.merge_forward              :     0.000016s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000061s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000047s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001619s :  5.59%
    jit_opt_a.replace_old_param          :     0.000076s :  0.26%
    jit_opt_a.inline_without_move        :     0.000067s :  0.23%
    jit_opt_a.renormalize                :     0.003437s : 11.86%
    jit_opt_a.add_forward_monad_depend   :     0.000020s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000082s :  0.28%
    jit_opt_a.cse                        :     0.000204s :  0.71%
    jit_opt_a.replace_applicator         :     0.000101s :  0.35%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.05%
    rewriter_after_opt_a                 :     0.000046s :  0.16%
    convert_after_rewriter               :     0.000010s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000777s :  2.68%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000021s :  0.07%
    cconv                                :     0.000024s :  0.08%
    loop_unroll                          :     0.000447s :  1.54%
    jit_opt_after_cconv.c_1              :     0.000030s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000017s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000057s :  0.20%
    cse_after_recomputation.cse          :     0.000019s :  0.06%
    auto_monad_reorder                   :     0.000022s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000466s :  1.61%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008151s : 28.12%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000547   118
 5.77% :     0.000032s :      1: substitution.arithmetic_simplify
 0.46% :     0.000003s :      4: substitution.elim_not_effective
 0.44% :     0.000002s :      4: substitution.fold_const_symbol
 1.28% :     0.000007s :      5: substitution.graph_param_transform
58.29% :     0.000319s :     12: substitution.inline
 3.49% :     0.000019s :      2: substitution.inline_without_move
 1.75% :     0.000010s :     16: substitution.j_node_and_user_rematch
 2.11% :     0.000012s :      7: substitution.minmaximum_grad
 1.40% :     0.000008s :      5: substitution.partial_eliminate
 2.24% :     0.000012s :     16: substitution.remove_not_recompute_node
 5.65% :     0.000031s :     10: substitution.replace_applicator
 1.63% :     0.000009s :      7: substitution.replace_old_param
 0.47% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.20% :     0.000023s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.89% :     0.000016s :      7: substitution.tuple_list_get_item_depend_reorder
 7.92% :     0.000043s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009691     2
87.83% :     0.008511s :      1: type_inference.infer
12.17% :     0.001180s :      1: type_inference.specialize
------[replace.]   0.000150    20
 3.76% :     0.000006s :      1: replace.arithmetic_simplify
62.26% :     0.000093s :     12: replace.inline
33.98% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000364    20
 8.43% :     0.000031s :      1: match.arithmetic_simplify
85.90% :     0.000313s :     12: match.inline
 5.67% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000410  2790
 1.67% :     0.000007s :     46: predicate.accumulaten_eliminater
 0.40% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.39% :     0.000006s :     46: predicate.addn_check_dump
 1.98% :     0.000008s :     46: predicate.addn_zero_filter
 2.32% :     0.000010s :     47: predicate.arithmetic_simplify
 1.64% :     0.000007s :     47: predicate.cast_eliminate
 0.25% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.36% :     0.000006s :     46: predicate.compare_switch_simplify
 1.67% :     0.000007s :     46: predicate.depend_value_elim
 1.42% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.59% :     0.000007s :     47: predicate.dict_get_item_eliminator
 1.52% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.29% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      5: predicate.elim_not_effective
 0.22% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.50% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.45% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.53% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.46% :     0.000006s :     47: predicate.environ_get_eliminate
 1.57% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      5: predicate.fold_const_symbol
 0.93% :     0.000004s :     21: predicate.get_grad_eliminate
 0.11% :     0.000000s :      5: predicate.graph_param_transform
 4.18% :     0.000017s :     76: predicate.inline
 1.66% :     0.000007s :     42: predicate.inline_without_move
 0.35% :     0.000001s :     21: predicate.j_node_and_user_rematch
 1.15% :     0.000005s :     21: predicate.less_batch_normalization
 1.83% :     0.000008s :     54: predicate.list_to_tuple_eliminator_
 1.98% :     0.000008s :     59: predicate.load_eliminater
 0.42% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.00% :     0.000012s :     89: predicate.loop_unroll_before_grad
 2.07% :     0.000008s :     52: predicate.make_slice_get_slice_eliminator
 1.45% :     0.000006s :     46: predicate.merge_addn
 1.45% :     0.000006s :     47: predicate.minmaximum_grad
 0.45% :     0.000002s :      5: predicate.mutable_eliminate
 0.22% :     0.000001s :      5: predicate.opt_reshape
 2.31% :     0.000009s :     59: predicate.partial_eliminate
 1.51% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.81% :     0.000007s :     47: predicate.reduce_eliminate
 1.76% :     0.000007s :     54: predicate.redundant_stop_gradient_eliminater
 0.47% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.57% :     0.000011s :    105: predicate.replace_applicator
 0.80% :     0.000003s :     42: predicate.replace_old_param
 0.13% :     0.000001s :      5: predicate.reset_defer_inline
 1.90% :     0.000008s :     47: predicate.reshape_eliminate
 1.51% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.33% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.52% :     0.000006s :     46: predicate.same_eliminate
 0.48% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.42% :     0.000002s :     10: predicate.special_op_eliminate
 0.88% :     0.000004s :     21: predicate.specialize_transform
 1.86% :     0.000008s :     46: predicate.split_environ_get_set_with_tuple_value
 1.54% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.19% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.56% :     0.000011s :     66: predicate.switch_defer_inline
 2.37% :     0.000010s :     66: predicate.switch_layer_defer_inline
 6.21% :     0.000025s :    160: predicate.switch_simplify
 1.62% :     0.000007s :     47: predicate.tile_eliminate
 1.59% :     0.000007s :     47: predicate.transpose_eliminate
 1.72% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.55% :     0.000006s :     47: predicate.tuple_list_get_item_depend_reorder
 3.02% :     0.000012s :     64: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.83% :     0.000008s :     54: predicate.tuple_to_list_eliminator_
 1.91% :     0.000008s :     59: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000012s :     80: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000007s :     46: predicate.value_based_eliminate
 0.16% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.28% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001451    29
57.73% :     0.000838s :     13: func_graph_cloner_run.FuncGraphClonerGraph
42.27% :     0.000613s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.039684    91
 0.15% :     0.000060s :      1: add_recomputation
 0.23% :     0.000092s :      1: auto_monad
 0.06% :     0.000025s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.55% :     0.000219s :      1: bootstrap
 0.07% :     0.000027s :      1: cconv
 0.03% :     0.000012s :      1: convert_after_rewriter
 0.09% :     0.000034s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.14% :     0.000054s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
31.07% :     0.012331s :      1: jit_opt_a
 0.44% :     0.000173s :      1: jit_opt_after_cconv
 0.17% :     0.000067s :      1: jit_opt_b
 1.15% :     0.000455s :      1: loop_unroll
 1.98% :     0.000787s :      1: mutable_eliminate
 7.25% :     0.002876s :     39: opt.transform.jit_opt_a
 0.16% :     0.000065s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000037s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000043s :      4: opt.transform.symbol_engine_opt
 1.20% :     0.000474s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000026s :      1: py_interpret_to_execute
 0.04% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000027s :      1: remove_dup_value
 4.73% :     0.001876s :      2: renormalize.infer
 3.89% :     0.001543s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000049s :      1: rewriter_after_opt_a
 0.23% :     0.000090s :      1: rewriter_before_opt_a
 0.22% :     0.000089s :      1: symbol_engine_optimizer
20.57% :     0.008163s :      1: task_emit
24.63% :     0.009772s :      1: type_inference
 0.17% :     0.000066s :      1: validate


TotalTime = 0.0236118, [33]
[bootstrap]: 0.00023322
[type_inference]: 0.0113981
[event_method]: 4.918e-05
[auto_monad]: 9.764e-05
[graph_reusing]: 7.13998e-06
[pre_auto_parallel]: 3.48999e-06
[py_interpret_to_execute]: 2.785e-05
[rewriter_before_opt_a]: 9.511e-05
[expand_dump_flag]: 3.13998e-06
[jit_opt_a]: 0.00932248, [2]
    [Cycle 1]: 0.0052925, [27]
        [switch_simplify]: 7.261e-05
        [loop_unroll]: 3.833e-05
        [a_1]: 0.00091014
        [with_stream_mark]: 2.83e-05
        [recompute_prepare]: 1.808e-05
        [updatestate_depend_eliminate]: 8.17998e-06
        [updatestate_assign_eliminate]: 6.19001e-06
        [updatestate_loads_eliminate]: 5.94999e-06
        [parameter_eliminate]: 2.99999e-06
        [specialize_transform]: 1.295e-05
        [updatestate_useless_node_eliminater]: 3.191e-05
        [accelerated_algorithm]: 1.285e-05
        [meta_shard_fg_expand]: 4.1e-06
        [get_grad_eliminate_]: 1.202e-05
        [merge_forward]: 9.66e-06
        [cell_reuse_recompute_pass]: 1.31002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.667e-05
        [j_node_and_user_rematch]: 2.453e-05
        [meta_fg_expand]: 0.00176094
        [replace_old_param]: 6.256e-05
        [inline_without_move]: 5.658e-05
        [renormalize]: 0.00191194
        [add_forward_monad_depend]: 5.44998e-06
        [auto_monad_grad]: 2.52001e-06
        [auto_monad_eliminator]: 1.486e-05
        [cse]: 2.763e-05
        [replace_applicator]: 1.42e-05
    [Cycle 2]: 0.00033862, [27]
        [switch_simplify]: 5.36998e-06
        [loop_unroll]: 4.51002e-06
        [a_1]: 5.216e-05
        [with_stream_mark]: 9.43002e-06
        [recompute_prepare]: 4.55001e-06
        [updatestate_depend_eliminate]: 2.19999e-06
        [updatestate_assign_eliminate]: 2.11998e-06
        [updatestate_loads_eliminate]: 1.44e-06
        [parameter_eliminate]: 1.33002e-06
        [specialize_transform]: 3.72002e-06
        [updatestate_useless_node_eliminater]: 3.98999e-06
        [accelerated_algorithm]: 3.91001e-06
        [meta_shard_fg_expand]: 1.42999e-06
        [get_grad_eliminate_]: 4.01001e-06
        [merge_forward]: 2.77002e-06
        [cell_reuse_recompute_pass]: 4.15e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.294e-05
        [j_node_and_user_rematch]: 7.45e-06
        [meta_fg_expand]: 4.714e-05
        [replace_old_param]: 5.33002e-06
        [inline_without_move]: 3.81001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.49e-06
        [auto_monad_grad]: 1.34e-06
        [auto_monad_eliminator]: 6.43998e-06
        [cse]: 1.277e-05
        [replace_applicator]: 5.61998e-06
[py_interpret_to_execute_after_opt_a]: 1.18e-05
[rewriter_after_opt_a]: 3.689e-05
[convert_after_rewriter]: 7.24001e-06
[order_py_execute_after_rewriter]: 4.67e-06
[mutable_eliminate]: 0.00074728
[jit_opt_b]: 6.103e-05, [1]
    [Cycle 1]: 5.296e-05, [2]
        [frontend_op_eliminate]: 2.854e-05
        [inline_after_opt_a]: 1.123e-05
[cconv]: 2.561e-05
[loop_unroll]: 0.00045728
[jit_opt_after_cconv]: 0.00012664, [1]
    [Cycle 1]: 0.0001194, [11]
        [c_1]: 1.263e-05
        [parameter_eliminate]: 2.73998e-06
        [updatestate_depend_eliminate]: 5.14e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.48002e-06
        [cse]: 1.946e-05
        [call_graph_tuple_transform]: 1.435e-05
        [tuple_list_get_item_eliminator]: 3.70998e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 4.67998e-06
[remove_dup_value]: 1.51e-05
[partial_unused_args_eliminate]: 2.22999e-06
[environ_conv]: 4.87998e-06
[add_recomputation]: 3.509e-05
[cse_after_recomputation]: 2.012e-05, [1]
    [Cycle 1]: 1.415e-05, [1]
        [cse]: 8e-06
[auto_monad_reorder]: 1.355e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 6.73e-06
[opt_after_jit_grad]: 0.00049388
[symbol_engine_optimizer]: 6.805e-05, [1]
    [Cycle 1]: 6.104e-05, [6]
        [build]: 3.81999e-06
        [elim_shapecalc]: 7.11999e-06
        [elim_not_effective]: 9.49999e-06
        [opt_reshape]: 4.95999e-06
        [fold_const_symbol]: 6.26e-06
        [renormalize]: 2.09984e-07
[validate]: 3.138e-05
[backend_pass]: 7.7e-07
[task_emit]: 2.546e-05
[execute]: 1.19e-06
Sums
    bootstrap                            :     0.000233s :  1.21%
    type_inference                       :     0.011398s : 59.12%
    event_method                         :     0.000049s :  0.26%
    auto_monad                           :     0.000098s :  0.51%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.14%
    rewriter_before_opt_a                :     0.000095s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000078s :  0.40%
    jit_opt_a.loop_unroll                :     0.000043s :  0.22%
    jit_opt_a.a_1                        :     0.000962s :  4.99%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000023s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000036s :  0.19%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.08%
    jit_opt_a.merge_forward              :     0.000012s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000032s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001808s :  9.38%
    jit_opt_a.replace_old_param          :     0.000068s :  0.35%
    jit_opt_a.inline_without_move        :     0.000060s :  0.31%
    jit_opt_a.renormalize                :     0.001912s :  9.92%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.11%
    jit_opt_a.cse                        :     0.000040s :  0.21%
    jit_opt_a.replace_applicator         :     0.000020s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.06%
    rewriter_after_opt_a                 :     0.000037s :  0.19%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000747s :  3.88%
    jit_opt_b.frontend_op_eliminate      :     0.000029s :  0.15%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.06%
    cconv                                :     0.000026s :  0.13%
    loop_unroll                          :     0.000457s :  2.37%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.02%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000035s :  0.18%
    cse_after_recomputation.cse          :     0.000008s :  0.04%
    auto_monad_reorder                   :     0.000014s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.03%
    opt_after_jit_grad                   :     0.000494s :  2.56%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000025s :  0.13%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000310    50
 0.48% :     0.000001s :      1: substitution.elim_not_effective
 0.34% :     0.000001s :      1: substitution.fold_const_symbol
 1.37% :     0.000004s :      1: substitution.graph_param_transform
69.70% :     0.000216s :      6: substitution.inline
 6.65% :     0.000021s :      2: substitution.inline_without_move
 2.46% :     0.000008s :      9: substitution.j_node_and_user_rematch
 1.20% :     0.000004s :      2: substitution.minmaximum_grad
 1.63% :     0.000005s :      3: substitution.partial_eliminate
 2.46% :     0.000008s :      9: substitution.remove_not_recompute_node
 1.94% :     0.000006s :      2: substitution.replace_applicator
 1.61% :     0.000005s :      5: substitution.replace_old_param
 0.78% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.33% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.51% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.54% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011325     2
88.95% :     0.010073s :      1: type_inference.infer
11.05% :     0.001252s :      1: type_inference.specialize
------[replace.]   0.000069     8
77.88% :     0.000054s :      6: replace.inline
22.12% :     0.000015s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000225     8
94.39% :     0.000213s :      6: match.inline
 5.61% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000182  1109
 1.69% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.70% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.45% :     0.000003s :     18: predicate.addn_check_dump
 1.69% :     0.000003s :     18: predicate.addn_zero_filter
 2.08% :     0.000004s :     18: predicate.arithmetic_simplify
 1.41% :     0.000003s :     18: predicate.cast_eliminate
 0.30% :     0.000001s :      1: predicate.check_bprop_eliminate
 1.32% :     0.000002s :     18: predicate.compare_switch_simplify
 1.67% :     0.000003s :     18: predicate.depend_value_elim
 1.34% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.40% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.38% :     0.000003s :     18: predicate.dict_set_item_eliminator
 0.38% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      1: predicate.elim_not_effective
 0.27% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.22% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.24% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.44% :     0.000003s :     18: predicate.environ_get_eliminate
 1.28% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.04% :     0.000000s :      1: predicate.fold_const_symbol
 1.16% :     0.000002s :     12: predicate.get_grad_eliminate
 0.18% :     0.000000s :      1: predicate.graph_param_transform
 4.52% :     0.000008s :     28: predicate.inline
 3.16% :     0.000006s :     33: predicate.inline_without_move
 0.47% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.22% :     0.000002s :     12: predicate.less_batch_normalization
 1.80% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.77% :     0.000003s :     21: predicate.load_eliminater
 0.62% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.57% :     0.000007s :     44: predicate.loop_unroll_before_grad
 1.82% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.32% :     0.000002s :     18: predicate.merge_addn
 1.31% :     0.000002s :     18: predicate.minmaximum_grad
 1.10% :     0.000002s :      1: predicate.mutable_eliminate
 0.44% :     0.000001s :      1: predicate.opt_reshape
 1.95% :     0.000004s :     21: predicate.partial_eliminate
 1.50% :     0.000003s :     18: predicate.print_const_string_wrapper
 1.72% :     0.000003s :     18: predicate.reduce_eliminate
 1.39% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.81% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.39% :     0.000003s :     22: predicate.replace_applicator
 1.80% :     0.000003s :     33: predicate.replace_old_param
 0.09% :     0.000000s :      1: predicate.reset_defer_inline
 1.75% :     0.000003s :     18: predicate.reshape_eliminate
 1.37% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.45% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     18: predicate.same_eliminate
 0.82% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.29% :     0.000001s :      2: predicate.special_op_eliminate
 1.06% :     0.000002s :     12: predicate.specialize_transform
 1.57% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.13% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.52% :     0.000005s :     26: predicate.switch_defer_inline
 2.47% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.17% :     0.000011s :     71: predicate.switch_simplify
 1.47% :     0.000003s :     18: predicate.tile_eliminate
 1.46% :     0.000003s :     18: predicate.transpose_eliminate
 1.74% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 3.07% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.83% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.55% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.04% :     0.000006s :     33: predicate.updatestate_useless_node_eliminater
 1.90% :     0.000003s :     18: predicate.value_based_eliminate
 0.09% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.46% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001278    22
72.94% :     0.000933s :     13: func_graph_cloner_run.FuncGraphClonerGraph
27.06% :     0.000346s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026984    76
 0.14% :     0.000038s :      1: add_recomputation
 0.38% :     0.000102s :      1: auto_monad
 0.06% :     0.000016s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.91% :     0.000245s :      1: bootstrap
 0.11% :     0.000028s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.20% :     0.000053s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
34.56% :     0.009326s :      1: jit_opt_a
 0.48% :     0.000129s :      1: jit_opt_after_cconv
 0.24% :     0.000064s :      1: jit_opt_b
 1.72% :     0.000464s :      1: loop_unroll
 2.81% :     0.000757s :      1: mutable_eliminate
 5.09% :     0.001374s :     26: opt.transform.jit_opt_a
 0.12% :     0.000032s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000033s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000024s :      4: opt.transform.symbol_engine_opt
 1.86% :     0.000503s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000030s :      1: py_interpret_to_execute
 0.05% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000018s :      1: remove_dup_value
 5.34% :     0.001442s :      1: renormalize.infer
 1.71% :     0.000461s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000040s :      1: rewriter_after_opt_a
 0.36% :     0.000098s :      1: rewriter_before_opt_a
 0.26% :     0.000071s :      1: symbol_engine_optimizer
 0.11% :     0.000030s :      1: task_emit
42.29% :     0.011411s :      1: type_inference
 0.18% :     0.000048s :      1: validate

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.234.259 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e98b0> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.235.003 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0191078, [33]
[bootstrap]: 0.00029881
[type_inference]: 0.00431671
[event_method]: 1.661e-05
[auto_monad]: 5.884e-05
[graph_reusing]: 5.74e-06
[pre_auto_parallel]: 2.17001e-06
[py_interpret_to_execute]: 2.942e-05
[rewriter_before_opt_a]: 7.35e-05
[expand_dump_flag]: 2.90998e-06
[jit_opt_a]: 0.00504208, [2]
    [Cycle 1]: 0.00146247, [27]
        [switch_simplify]: 4.67e-05
        [loop_unroll]: 2.869e-05
        [a_1]: 0.00049463
        [with_stream_mark]: 1.547e-05
        [recompute_prepare]: 7.41001e-06
        [updatestate_depend_eliminate]: 3.11001e-06
        [updatestate_assign_eliminate]: 2.90002e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 1.66002e-06
        [specialize_transform]: 5.97999e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.67999e-06
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 5.34998e-06
        [merge_forward]: 3.58e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.356e-05
        [j_node_and_user_rematch]: 9.31e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 8.23999e-06
        [inline_without_move]: 6.02999e-06
        [renormalize]: 0.00057428
        [add_forward_monad_depend]: 5.12e-06
        [auto_monad_grad]: 1.46998e-06
        [auto_monad_eliminator]: 1.091e-05
        [cse]: 1.74e-05
        [replace_applicator]: 1.309e-05
    [Cycle 2]: 0.00031655, [27]
        [switch_simplify]: 5.14e-06
        [loop_unroll]: 4.63001e-06
        [a_1]: 8.586e-05
        [with_stream_mark]: 9.99001e-06
        [recompute_prepare]: 5.37001e-06
        [updatestate_depend_eliminate]: 2.46998e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.85001e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 4.57e-06
        [updatestate_useless_node_eliminater]: 4.65001e-06
        [accelerated_algorithm]: 4.82998e-06
        [meta_shard_fg_expand]: 1.52001e-06
        [get_grad_eliminate_]: 4.25e-06
        [merge_forward]: 2.47001e-06
        [cell_reuse_recompute_pass]: 1.62001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.254e-05
        [j_node_and_user_rematch]: 7.7e-06
        [meta_fg_expand]: 1.52999e-06
        [replace_old_param]: 7.77e-06
        [inline_without_move]: 4.42e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 8.99978e-07
        [auto_monad_eliminator]: 5.33002e-06
        [cse]: 1.03e-05
        [replace_applicator]: 4.93001e-06
[py_interpret_to_execute_after_opt_a]: 8.15e-06
[rewriter_after_opt_a]: 2.708e-05
[convert_after_rewriter]: 7.05e-06
[order_py_execute_after_rewriter]: 4.04002e-06
[mutable_eliminate]: 0.00052768
[jit_opt_b]: 4.871e-05, [1]
    [Cycle 1]: 4.213e-05, [2]
        [frontend_op_eliminate]: 1.602e-05
        [inline_after_opt_a]: 1.429e-05
[cconv]: 2.757e-05
[loop_unroll]: 0.00041997
[jit_opt_after_cconv]: 0.00013109, [1]
    [Cycle 1]: 0.00012486, [11]
        [c_1]: 2.012e-05
        [parameter_eliminate]: 2.61e-06
        [updatestate_depend_eliminate]: 4.99998e-06
        [updatestate_assign_eliminate]: 2.55997e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.687e-05
        [call_graph_tuple_transform]: 1.566e-05
        [tuple_list_get_item_eliminator]: 5.18002e-06
        [none_parameter_eliminate]: 9.70002e-07
        [renormalize]: 5.39992e-07
        [switch_simplify]: 5.12e-06
[remove_dup_value]: 6.24001e-06
[partial_unused_args_eliminate]: 1.49998e-06
[environ_conv]: 3.90998e-06
[add_recomputation]: 3.409e-05
[cse_after_recomputation]: 1.948e-05, [1]
    [Cycle 1]: 1.428e-05, [1]
        [cse]: 8.69998e-06
[auto_monad_reorder]: 9.30001e-06
[get_jit_bprop_graph]: 1.38002e-06
[rewriter_after_jit_bprop_graph]: 5.44998e-06
[opt_after_jit_grad]: 0.00049968
[symbol_engine_optimizer]: 6.947e-05, [1]
    [Cycle 1]: 6.325e-05, [6]
        [build]: 3.03998e-06
        [elim_shapecalc]: 7.44002e-06
        [elim_not_effective]: 1.194e-05
        [opt_reshape]: 5.39e-06
        [fold_const_symbol]: 7.93999e-06
        [renormalize]: 8.70001e-07
[validate]: 2.639e-05
[backend_pass]: 7.50006e-07
[task_emit]: 0.00719113
[execute]: 8.53001e-06
Sums
    bootstrap                            :     0.000299s :  1.96%
    type_inference                       :     0.004317s : 28.29%
    event_method                         :     0.000017s :  0.11%
    auto_monad                           :     0.000059s :  0.39%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000029s :  0.19%
    rewriter_before_opt_a                :     0.000074s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000052s :  0.34%
    jit_opt_a.loop_unroll                :     0.000033s :  0.22%
    jit_opt_a.a_1                        :     0.000580s :  3.80%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.06%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.11%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000574s :  3.76%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.11%
    jit_opt_a.cse                        :     0.000028s :  0.18%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000027s :  0.18%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000528s :  3.46%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.10%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.09%
    cconv                                :     0.000028s :  0.18%
    loop_unroll                          :     0.000420s :  2.75%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000006s :  0.04%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000034s :  0.22%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000009s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000500s :  3.27%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000026s :  0.17%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007191s : 47.13%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000173    23
 0.74% :     0.000001s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 2.06% :     0.000004s :      3: substitution.graph_param_transform
83.70% :     0.000145s :      4: substitution.inline
 1.94% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.25% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.47% :     0.000004s :      2: substitution.replace_old_param
 6.35% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004256     2
81.66% :     0.003476s :      1: type_inference.infer
18.34% :     0.000780s :      1: type_inference.specialize
------[replace.]   0.000060     6
65.42% :     0.000040s :      4: replace.inline
34.58% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000153     6
93.46% :     0.000143s :      4: match.inline
 6.54% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.24% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.38% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000001s :     10: predicate.addn_zero_filter
 2.08% :     0.000002s :     10: predicate.arithmetic_simplify
 1.41% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000001s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.19% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.80% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.91% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.25% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.34% :     0.000006s :     22: predicate.inline
 1.08% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.37% :     0.000002s :      6: predicate.less_batch_normalization
 1.83% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.36% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.14% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.18% :     0.000001s :     10: predicate.minmaximum_grad
 1.67% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000002s :     15: predicate.partial_eliminate
 1.30% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.97% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.35% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.20% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.21% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.56% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.34% :     0.000008s :     53: predicate.switch_simplify
 1.31% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.21% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000523    12
48.52% :     0.000254s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.16% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.32% :     0.000242s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020563    76
 0.18% :     0.000037s :      1: add_recomputation
 0.30% :     0.000062s :      1: auto_monad
 0.06% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.52% :     0.000312s :      1: bootstrap
 0.15% :     0.000030s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.06% :     0.000013s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
24.53% :     0.005045s :      1: jit_opt_a
 0.65% :     0.000134s :      1: jit_opt_after_cconv
 0.25% :     0.000051s :      1: jit_opt_b
 2.08% :     0.000428s :      1: loop_unroll
 2.61% :     0.000536s :      1: mutable_eliminate
 3.79% :     0.000780s :     26: opt.transform.jit_opt_a
 0.21% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.47% :     0.000508s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000032s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000009s :      1: remove_dup_value
 1.47% :     0.000303s :      1: renormalize.infer
 1.29% :     0.000265s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000030s :      1: rewriter_after_opt_a
 0.37% :     0.000077s :      1: rewriter_before_opt_a
 0.35% :     0.000072s :      1: symbol_engine_optimizer
35.04% :     0.007205s :      1: task_emit
21.05% :     0.004329s :      1: type_inference
 0.21% :     0.000043s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.237.889 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c36b700> should be wrapped by jit only once.

TotalTime = 0.0108311, [30]
[bootstrap]: 0.00024924
[type_inference]: 0.00377009
[event_method]: 1.62e-05
[auto_monad]: 5.757e-05
[graph_reusing]: 5.41002e-06
[pre_auto_parallel]: 2.16e-06
[py_interpret_to_execute]: 2.584e-05
[rewriter_before_opt_a]: 6.783e-05
[expand_dump_flag]: 2.93e-06
[jit_opt_a]: 0.00464434, [2]
    [Cycle 1]: 0.00133567, [27]
        [switch_simplify]: 4.885e-05
        [loop_unroll]: 2.802e-05
        [a_1]: 0.00048361
        [with_stream_mark]: 1.473e-05
        [recompute_prepare]: 7.06001e-06
        [updatestate_depend_eliminate]: 3.18e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 1.76e-06
        [specialize_transform]: 6.15002e-06
        [updatestate_useless_node_eliminater]: 5.11002e-06
        [accelerated_algorithm]: 5.30001e-06
        [meta_shard_fg_expand]: 2.16e-06
        [get_grad_eliminate_]: 4.97e-06
        [merge_forward]: 3.33e-06
        [cell_reuse_recompute_pass]: 8.30012e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.352e-05
        [j_node_and_user_rematch]: 8.80001e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 8.38999e-06
        [inline_without_move]: 5.00999e-06
        [renormalize]: 0.00045743
        [add_forward_monad_depend]: 4.99998e-06
        [auto_monad_grad]: 2.62001e-06
        [auto_monad_eliminator]: 1.23e-05
        [cse]: 2.501e-05
        [replace_applicator]: 1.065e-05
    [Cycle 2]: 0.00030952, [27]
        [switch_simplify]: 5.30001e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 8.693e-05
        [with_stream_mark]: 7.88999e-06
        [recompute_prepare]: 5.07999e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 1.89999e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.11002e-06
        [specialize_transform]: 5.06997e-06
        [updatestate_useless_node_eliminater]: 4.84e-06
        [accelerated_algorithm]: 4.90001e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 2.57001e-06
        [cell_reuse_recompute_pass]: 1.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.173e-05
        [j_node_and_user_rematch]: 7.8e-06
        [meta_fg_expand]: 1.39e-06
        [replace_old_param]: 6.89999e-06
        [inline_without_move]: 9.18002e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.02e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 4.75999e-06
        [cse]: 9.70002e-06
        [replace_applicator]: 4.93001e-06
[py_interpret_to_execute_after_opt_a]: 8.12e-06
[rewriter_after_opt_a]: 3.309e-05
[convert_after_rewriter]: 6.54999e-06
[order_py_execute_after_rewriter]: 4.84998e-06
[mutable_eliminate]: 0.00048427
[jit_opt_b]: 4.613e-05, [1]
    [Cycle 1]: 3.969e-05, [2]
        [frontend_op_eliminate]: 1.508e-05
        [inline_after_opt_a]: 1.377e-05
[cconv]: 2.386e-05
[loop_unroll]: 0.00040753
[jit_opt_after_cconv]: 0.00014774, [1]
    [Cycle 1]: 0.00014175, [11]
        [c_1]: 1.995e-05
        [parameter_eliminate]: 2.36e-06
        [updatestate_depend_eliminate]: 5.12e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [cse]: 2.903e-05
        [call_graph_tuple_transform]: 1.777e-05
        [tuple_list_get_item_eliminator]: 5.34e-06
        [none_parameter_eliminate]: 1.69e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 5.12e-06
[remove_dup_value]: 1.376e-05
[partial_unused_args_eliminate]: 1.96e-06
[environ_conv]: 4.63999e-06
[add_recomputation]: 4.464e-05
[cse_after_recomputation]: 2.005e-05, [1]
    [Cycle 1]: 1.456e-05, [1]
        [cse]: 8.91002e-06
[auto_monad_reorder]: 1.426e-05
[get_jit_bprop_graph]: 1.60999e-06
[rewriter_after_jit_bprop_graph]: 4.82998e-06
[opt_after_jit_grad]: 0.00046753
[symbol_engine_optimizer]: 7.569e-05, [1]
    [Cycle 1]: 6.901e-05, [6]
        [build]: 3.31999e-06
        [elim_shapecalc]: 7.51001e-06
        [elim_not_effective]: 1.287e-05
        [opt_reshape]: 6.31e-06
        [fold_const_symbol]: 9.30001e-06
        [renormalize]: 4.19997e-07
[validate]: 2.973e-05
Sums
    bootstrap                            :     0.000249s :  3.42%
    type_inference                       :     0.003770s : 51.73%
    event_method                         :     0.000016s :  0.22%
    auto_monad                           :     0.000058s :  0.79%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000026s :  0.35%
    rewriter_before_opt_a                :     0.000068s :  0.93%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000054s :  0.74%
    jit_opt_a.loop_unroll                :     0.000033s :  0.45%
    jit_opt_a.a_1                        :     0.000571s :  7.83%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.21%
    jit_opt_a.inline_without_move        :     0.000014s :  0.19%
    jit_opt_a.renormalize                :     0.000457s :  6.28%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.23%
    jit_opt_a.cse                        :     0.000035s :  0.48%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000033s :  0.45%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000484s :  6.65%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.19%
    cconv                                :     0.000024s :  0.33%
    loop_unroll                          :     0.000408s :  5.59%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000029s :  0.40%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.19%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000045s :  0.61%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000014s :  0.20%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000468s :  6.42%
    symbol_engine_optimizer.build        :     0.000003s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.18%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.09%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.13%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000030s :  0.41%

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.238.708 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

Time group info:
------[substitution.]   0.000172    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.92% :     0.000002s :      2: substitution.fold_const_symbol
 3.04% :     0.000005s :      3: substitution.graph_param_transform
82.37% :     0.000142s :      4: substitution.inline
 1.62% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.44% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000004s :      2: substitution.replace_old_param
 6.56% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003715     2
80.70% :     0.002998s :      1: type_inference.infer
19.30% :     0.000717s :      1: type_inference.specialize
------[replace.]   0.000058     6
67.20% :     0.000039s :      4: replace.inline
32.80% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000150     6
93.13% :     0.000140s :      4: match.inline
 6.87% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.65% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.33% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000001s :     10: predicate.addn_zero_filter
 2.18% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.22% :     0.000001s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.60% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.54% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.32% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.94% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.26% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.62% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.50% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.28% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.56% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.33% :     0.000008s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.18% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.88% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.69% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000478    12
49.88% :     0.000238s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.96% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.16% :     0.000211s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012122    72
 0.39% :     0.000048s :      1: add_recomputation
 0.50% :     0.000061s :      1: auto_monad
 0.14% :     0.000017s :      1: auto_monad_reorder
 2.15% :     0.000261s :      1: bootstrap
 0.22% :     0.000027s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
38.33% :     0.004647s :      1: jit_opt_a
 1.24% :     0.000151s :      1: jit_opt_after_cconv
 0.40% :     0.000049s :      1: jit_opt_b
 3.42% :     0.000415s :      1: loop_unroll
 4.06% :     0.000492s :      1: mutable_eliminate
 6.37% :     0.000772s :     26: opt.transform.jit_opt_a
 0.37% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000023s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.27% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.92% :     0.000475s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.23% :     0.000028s :      1: py_interpret_to_execute
 0.09% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000016s :      1: remove_dup_value
 1.90% :     0.000230s :      1: renormalize.infer
 1.82% :     0.000221s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.30% :     0.000036s :      1: rewriter_after_opt_a
 0.58% :     0.000070s :      1: rewriter_before_opt_a
 0.65% :     0.000078s :      1: symbol_engine_optimizer
31.19% :     0.003781s :      1: type_inference


TotalTime = 0.0104774, [30]
[bootstrap]: 0.00023987
[type_inference]: 0.00367479
[event_method]: 1.561e-05
[auto_monad]: 5.67e-05
[graph_reusing]: 6.06998e-06
[pre_auto_parallel]: 2.37999e-06
[py_interpret_to_execute]: 2.728e-05
[rewriter_before_opt_a]: 6.565e-05
[expand_dump_flag]: 2.96999e-06
[jit_opt_a]: 0.00448381, [2]
    [Cycle 1]: 0.00126629, [27]
        [switch_simplify]: 4.647e-05
        [loop_unroll]: 2.783e-05
        [a_1]: 0.00047314
        [with_stream_mark]: 1.303e-05
        [recompute_prepare]: 7.01001e-06
        [updatestate_depend_eliminate]: 2.98e-06
        [updatestate_assign_eliminate]: 2.48002e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [parameter_eliminate]: 1.69e-06
        [specialize_transform]: 5.77001e-06
        [updatestate_useless_node_eliminater]: 4.91002e-06
        [accelerated_algorithm]: 5.01997e-06
        [meta_shard_fg_expand]: 1.82001e-06
        [get_grad_eliminate_]: 4.75999e-06
        [merge_forward]: 3.50003e-06
        [cell_reuse_recompute_pass]: 9.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.33e-05
        [j_node_and_user_rematch]: 8.32e-06
        [meta_fg_expand]: 2.00002e-06
        [replace_old_param]: 7.52002e-06
        [inline_without_move]: 5.22999e-06
        [renormalize]: 0.00041672
        [add_forward_monad_depend]: 4.67e-06
        [auto_monad_grad]: 1.99e-06
        [auto_monad_eliminator]: 1.221e-05
        [cse]: 2.207e-05
        [replace_applicator]: 1.031e-05
    [Cycle 2]: 0.00030779, [27]
        [switch_simplify]: 5.17e-06
        [loop_unroll]: 4.84998e-06
        [a_1]: 8.734e-05
        [with_stream_mark]: 7.87998e-06
        [recompute_prepare]: 5.10999e-06
        [updatestate_depend_eliminate]: 2.39001e-06
        [updatestate_assign_eliminate]: 1.91003e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.22999e-06
        [specialize_transform]: 4.87998e-06
        [updatestate_useless_node_eliminater]: 4.63999e-06
        [accelerated_algorithm]: 5.00999e-06
        [meta_shard_fg_expand]: 1.02998e-06
        [get_grad_eliminate_]: 4.41002e-06
        [merge_forward]: 2.19999e-06
        [cell_reuse_recompute_pass]: 1.24998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.184e-05
        [j_node_and_user_rematch]: 7.41999e-06
        [meta_fg_expand]: 1.36002e-06
        [replace_old_param]: 6.88998e-06
        [inline_without_move]: 4.82998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.03001e-06
        [auto_monad_grad]: 6.79982e-07
        [auto_monad_eliminator]: 4.55999e-06
        [cse]: 9.92999e-06
        [replace_applicator]: 4.84e-06
[py_interpret_to_execute_after_opt_a]: 8.31002e-06
[rewriter_after_opt_a]: 2.896e-05
[convert_after_rewriter]: 6.32001e-06
[order_py_execute_after_rewriter]: 4.78001e-06
[mutable_eliminate]: 0.00045661
[jit_opt_b]: 4.654e-05, [1]
    [Cycle 1]: 4.002e-05, [2]
        [frontend_op_eliminate]: 1.514e-05
        [inline_after_opt_a]: 1.394e-05
[cconv]: 2.158e-05
[loop_unroll]: 0.00039939
[jit_opt_after_cconv]: 0.0001311, [1]
    [Cycle 1]: 0.00012464, [11]
        [c_1]: 2.009e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 5.04e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.03002e-06
        [cse]: 1.527e-05
        [call_graph_tuple_transform]: 1.695e-05
        [tuple_list_get_item_eliminator]: 5.27001e-06
        [none_parameter_eliminate]: 1.29e-06
        [renormalize]: 4.90021e-07
        [switch_simplify]: 5.17999e-06
[remove_dup_value]: 1.327e-05
[partial_unused_args_eliminate]: 2.11e-06
[environ_conv]: 4.75999e-06
[add_recomputation]: 4.267e-05
[cse_after_recomputation]: 1.95e-05, [1]
    [Cycle 1]: 1.422e-05, [1]
        [cse]: 8.84e-06
[auto_monad_reorder]: 1.27e-05
[get_jit_bprop_graph]: 1.55999e-06
[rewriter_after_jit_bprop_graph]: 4.44998e-06
[opt_after_jit_grad]: 0.00043395
[symbol_engine_optimizer]: 6.919e-05, [1]
    [Cycle 1]: 6.298e-05, [6]
        [build]: 3.37002e-06
        [elim_shapecalc]: 7.58001e-06
        [elim_not_effective]: 1.159e-05
        [opt_reshape]: 5.46e-06
        [fold_const_symbol]: 8.25e-06
        [renormalize]: 3.60014e-07
[validate]: 4.896e-05
Sums
    bootstrap                            :     0.000240s :  3.41%
    type_inference                       :     0.003675s : 52.23%
    event_method                         :     0.000016s :  0.22%
    auto_monad                           :     0.000057s :  0.81%
    graph_reusing                        :     0.000006s :  0.09%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000027s :  0.39%
    rewriter_before_opt_a                :     0.000066s :  0.93%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000052s :  0.73%
    jit_opt_a.loop_unroll                :     0.000033s :  0.46%
    jit_opt_a.a_1                        :     0.000560s :  7.97%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.36%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.05%
    jit_opt_a.replace_old_param          :     0.000014s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000417s :  5.92%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.24%
    jit_opt_a.cse                        :     0.000032s :  0.45%
    jit_opt_a.replace_applicator         :     0.000015s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.12%
    rewriter_after_opt_a                 :     0.000029s :  0.41%
    convert_after_rewriter               :     0.000006s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000457s :  6.49%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.20%
    cconv                                :     0.000022s :  0.31%
    loop_unroll                          :     0.000399s :  5.68%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.29%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000015s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000013s :  0.19%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000043s :  0.61%
    cse_after_recomputation.cse          :     0.000009s :  0.13%
    auto_monad_reorder                   :     0.000013s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.06%
    opt_after_jit_grad                   :     0.000434s :  6.17%
    symbol_engine_optimizer.build        :     0.000003s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000049s :  0.70%


Time group info:
------[substitution.]   0.000163    23
 1.11% :     0.000002s :      2: substitution.elim_not_effective
 0.78% :     0.000001s :      2: substitution.fold_const_symbol
 2.77% :     0.000005s :      3: substitution.graph_param_transform
82.70% :     0.000135s :      4: substitution.inline
 1.66% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.90% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.79% :     0.000003s :      2: substitution.replace_old_param
 6.29% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003624     2
80.67% :     0.002923s :      1: type_inference.infer
19.33% :     0.000700s :      1: type_inference.specialize
------[replace.]   0.000059     6
67.92% :     0.000040s :      4: replace.inline
32.08% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000142     6
93.52% :     0.000133s :      4: match.inline
 6.48% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.36% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000001s :     10: predicate.addn_zero_filter
 2.17% :     0.000002s :     10: predicate.arithmetic_simplify
 1.19% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000001s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.53% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.48% :     0.000001s :      3: predicate.graph_param_transform
 5.84% :     0.000006s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.45% :     0.000000s :      3: predicate.opt_reshape
 2.35% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000001s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.84% :     0.000001s :      6: predicate.special_op_eliminate
 1.19% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.42% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.48% :     0.000008s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.57% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.73% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.02% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000458    12
49.88% :     0.000228s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.35% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.77% :     0.000205s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011688    72
 0.39% :     0.000045s :      1: add_recomputation
 0.51% :     0.000060s :      1: auto_monad
 0.13% :     0.000015s :      1: auto_monad_reorder
 2.15% :     0.000251s :      1: bootstrap
 0.21% :     0.000024s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000019s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.07% :     0.000008s :      1: graph_reusing
38.38% :     0.004486s :      1: jit_opt_a
 1.14% :     0.000134s :      1: jit_opt_after_cconv
 0.42% :     0.000049s :      1: jit_opt_b
 3.48% :     0.000407s :      1: loop_unroll
 3.98% :     0.000465s :      1: mutable_eliminate
 6.43% :     0.000751s :     26: opt.transform.jit_opt_a
 0.38% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000023s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.77% :     0.000441s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.04% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000030s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000015s :      1: remove_dup_value
 1.71% :     0.000199s :      1: renormalize.infer
 1.80% :     0.000211s :      1: renormalize.specialize
 0.06% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.29% :     0.000034s :      1: rewriter_after_opt_a
 0.58% :     0.000068s :      1: rewriter_before_opt_a
 0.61% :     0.000072s :      1: symbol_engine_optimizer
31.53% :     0.003685s :      1: type_inference


TotalTime = 0.0178039, [33]
[bootstrap]: 0.00029311
[type_inference]: 0.0043835
[event_method]: 1.643e-05
[auto_monad]: 5.416e-05
[graph_reusing]: 4.68999e-06
[pre_auto_parallel]: 2.04999e-06
[py_interpret_to_execute]: 3.029e-05
[rewriter_before_opt_a]: 6.761e-05
[expand_dump_flag]: 3.09999e-06
[jit_opt_a]: 0.00499276, [2]
    [Cycle 1]: 0.00138509, [27]
        [switch_simplify]: 4.542e-05
        [loop_unroll]: 2.809e-05
        [a_1]: 0.00046969
        [with_stream_mark]: 1.417e-05
        [recompute_prepare]: 6.77002e-06
        [updatestate_depend_eliminate]: 3.43e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [parameter_eliminate]: 1.46998e-06
        [specialize_transform]: 5.60001e-06
        [updatestate_useless_node_eliminater]: 4.75999e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 1.46002e-06
        [get_grad_eliminate_]: 4.95999e-06
        [merge_forward]: 2.99999e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.077e-05
        [j_node_and_user_rematch]: 8.82e-06
        [meta_fg_expand]: 1.83997e-06
        [replace_old_param]: 7.71999e-06
        [inline_without_move]: 5.32001e-06
        [renormalize]: 0.00053512
        [add_forward_monad_depend]: 4.97999e-06
        [auto_monad_grad]: 1.96e-06
        [auto_monad_eliminator]: 1.365e-05
        [cse]: 2.528e-05
        [replace_applicator]: 1.133e-05
    [Cycle 2]: 0.00030648, [27]
        [switch_simplify]: 5.22e-06
        [loop_unroll]: 4.89003e-06
        [a_1]: 8.66e-05
        [with_stream_mark]: 8.95001e-06
        [recompute_prepare]: 5.00999e-06
        [updatestate_depend_eliminate]: 2.48e-06
        [updatestate_assign_eliminate]: 1.95001e-06
        [updatestate_loads_eliminate]: 1.81998e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 5.09003e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 5.34e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.74e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 1.74e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.177e-05
        [j_node_and_user_rematch]: 7.66999e-06
        [meta_fg_expand]: 1.35999e-06
        [replace_old_param]: 6.86999e-06
        [inline_without_move]: 4.67e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 4.52998e-06
        [cse]: 8.83001e-06
        [replace_applicator]: 4.74e-06
[py_interpret_to_execute_after_opt_a]: 7.61999e-06
[rewriter_after_opt_a]: 3.173e-05
[convert_after_rewriter]: 7.29001e-06
[order_py_execute_after_rewriter]: 5.17e-06
[mutable_eliminate]: 0.00053036
[jit_opt_b]: 5.032e-05, [1]
    [Cycle 1]: 4.337e-05, [2]
        [frontend_op_eliminate]: 1.71e-05
        [inline_after_opt_a]: 1.523e-05
[cconv]: 2.301e-05
[loop_unroll]: 0.00046586
[jit_opt_after_cconv]: 0.00013497, [1]
    [Cycle 1]: 0.00012863, [11]
        [c_1]: 1.961e-05
        [parameter_eliminate]: 2.53e-06
        [updatestate_depend_eliminate]: 5.94999e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 2.11003e-06
        [cse]: 1.818e-05
        [call_graph_tuple_transform]: 1.801e-05
        [tuple_list_get_item_eliminator]: 5.32999e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 6.69999e-07
        [switch_simplify]: 5.14e-06
[remove_dup_value]: 1.302e-05
[partial_unused_args_eliminate]: 2.18998e-06
[environ_conv]: 4.69002e-06
[add_recomputation]: 4.501e-05
[cse_after_recomputation]: 1.953e-05, [1]
    [Cycle 1]: 1.418e-05, [1]
        [cse]: 8.64e-06
[auto_monad_reorder]: 1.488e-05
[get_jit_bprop_graph]: 1.59998e-06
[rewriter_after_jit_bprop_graph]: 4.72e-06
[opt_after_jit_grad]: 0.00045299
[symbol_engine_optimizer]: 6.887e-05, [1]
    [Cycle 1]: 6.295e-05, [6]
        [build]: 3.4e-06
        [elim_shapecalc]: 6.95998e-06
        [elim_not_effective]: 1.147e-05
        [opt_reshape]: 5.24998e-06
        [fold_const_symbol]: 8.60001e-06
        [renormalize]: 4.80009e-07
[validate]: 3.199e-05
[backend_pass]: 9.29984e-07
[task_emit]: 0.00585647
[execute]: 7.23999e-06
Sums
    bootstrap                            :     0.000293s :  2.10%
    type_inference                       :     0.004383s : 31.44%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000054s :  0.39%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000030s :  0.22%
    rewriter_before_opt_a                :     0.000068s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000051s :  0.36%
    jit_opt_a.loop_unroll                :     0.000033s :  0.24%
    jit_opt_a.a_1                        :     0.000556s :  3.99%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000535s :  3.84%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000034s :  0.24%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000032s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000530s :  3.80%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000466s :  3.34%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.14%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000045s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000453s :  3.25%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005856s : 42.01%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000159    23
 1.12% :     0.000002s :      2: substitution.elim_not_effective
 0.80% :     0.000001s :      2: substitution.fold_const_symbol
 3.37% :     0.000005s :      3: substitution.graph_param_transform
81.80% :     0.000130s :      4: substitution.inline
 2.13% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.42% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000003s :      2: substitution.replace_old_param
 6.34% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004321     2
82.78% :     0.003577s :      1: type_inference.infer
17.22% :     0.000744s :      1: type_inference.specialize
------[replace.]   0.000061     6
66.11% :     0.000041s :      4: replace.inline
33.89% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000137     6
93.36% :     0.000128s :      4: match.inline
 6.64% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.29% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.57% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000001s :     10: predicate.addn_zero_filter
 2.05% :     0.000002s :     10: predicate.arithmetic_simplify
 1.32% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.45% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.22% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.50% :     0.000002s :      6: predicate.less_batch_normalization
 1.86% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.61% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.66% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.07% :     0.000002s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.44% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.48% :     0.000001s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.29% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.30% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.62% :     0.000008s :     53: predicate.switch_simplify
 1.24% :     0.000001s :     10: predicate.tile_eliminate
 1.55% :     0.000002s :     10: predicate.transpose_eliminate
 1.99% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.40% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000502    12
49.27% :     0.000247s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.46% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.28% :     0.000227s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019195    76
 0.25% :     0.000048s :      1: add_recomputation
 0.30% :     0.000057s :      1: auto_monad
 0.09% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.60% :     0.000306s :      1: bootstrap
 0.13% :     0.000026s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.03% :     0.004996s :      1: jit_opt_a
 0.72% :     0.000138s :      1: jit_opt_after_cconv
 0.28% :     0.000053s :      1: jit_opt_b
 2.47% :     0.000473s :      1: loop_unroll
 2.80% :     0.000538s :      1: mutable_eliminate
 3.90% :     0.000748s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000026s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.40% :     0.000460s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.40% :     0.000269s :      1: renormalize.infer
 1.36% :     0.000261s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000035s :      1: rewriter_after_opt_a
 0.37% :     0.000071s :      1: rewriter_before_opt_a
 0.37% :     0.000071s :      1: symbol_engine_optimizer
30.57% :     0.005867s :      1: task_emit
22.89% :     0.004395s :      1: type_inference
 0.27% :     0.000051s :      1: validate


TotalTime = 0.0272808, [33]
[bootstrap]: 0.00025897
[type_inference]: 0.016225
[event_method]: 5.318e-05
[auto_monad]: 9.103e-05
[graph_reusing]: 6.74001e-06
[pre_auto_parallel]: 2.36e-06
[py_interpret_to_execute]: 2.569e-05
[rewriter_before_opt_a]: 8.668e-05
[expand_dump_flag]: 3.66001e-06
[jit_opt_a]: 0.00849172, [2]
    [Cycle 1]: 0.00459183, [27]
        [switch_simplify]: 5.762e-05
        [loop_unroll]: 3.867e-05
        [a_1]: 0.0008319
        [with_stream_mark]: 2.212e-05
        [recompute_prepare]: 1.815e-05
        [updatestate_depend_eliminate]: 7.73999e-06
        [updatestate_assign_eliminate]: 5.86e-06
        [updatestate_loads_eliminate]: 5.45001e-06
        [parameter_eliminate]: 1.78002e-06
        [specialize_transform]: 1.257e-05
        [updatestate_useless_node_eliminater]: 1.207e-05
        [accelerated_algorithm]: 1.203e-05
        [meta_shard_fg_expand]: 2.94001e-06
        [get_grad_eliminate_]: 1.238e-05
        [merge_forward]: 7.53999e-06
        [cell_reuse_recompute_pass]: 8.10018e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.351e-05
        [j_node_and_user_rematch]: 2.717e-05
        [meta_fg_expand]: 0.00164578
        [replace_old_param]: 5.224e-05
        [inline_without_move]: 5.251e-05
        [renormalize]: 0.00148967
        [add_forward_monad_depend]: 5.10999e-06
        [auto_monad_grad]: 2.13002e-06
        [auto_monad_eliminator]: 1.018e-05
        [cse]: 1.71e-05
        [replace_applicator]: 1.155e-05
    [Cycle 2]: 0.00028433, [27]
        [switch_simplify]: 4.28999e-06
        [loop_unroll]: 3.83001e-06
        [a_1]: 4.027e-05
        [with_stream_mark]: 8.56997e-06
        [recompute_prepare]: 3.64002e-06
        [updatestate_depend_eliminate]: 2.23998e-06
        [updatestate_assign_eliminate]: 1.91998e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.31998e-06
        [specialize_transform]: 3.51001e-06
        [updatestate_useless_node_eliminater]: 3.33998e-06
        [accelerated_algorithm]: 5.42999e-06
        [meta_shard_fg_expand]: 1.62001e-06
        [get_grad_eliminate_]: 4.12e-06
        [merge_forward]: 2.33002e-06
        [cell_reuse_recompute_pass]: 2.19001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.021e-05
        [j_node_and_user_rematch]: 6.33e-06
        [meta_fg_expand]: 3.188e-05
        [replace_old_param]: 4.26001e-06
        [inline_without_move]: 3.23e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.37999e-06
        [auto_monad_grad]: 9.80013e-07
        [auto_monad_eliminator]: 4.50001e-06
        [cse]: 1.042e-05
        [replace_applicator]: 4.02e-06
[py_interpret_to_execute_after_opt_a]: 7.80998e-06
[rewriter_after_opt_a]: 3.467e-05
[convert_after_rewriter]: 5.65001e-06
[order_py_execute_after_rewriter]: 4.43999e-06
[mutable_eliminate]: 0.00055249
[jit_opt_b]: 4.077e-05, [1]
    [Cycle 1]: 3.364e-05, [2]
        [frontend_op_eliminate]: 1.033e-05
        [inline_after_opt_a]: 1.12e-05
[cconv]: 1.83e-05
[loop_unroll]: 0.00043514
[jit_opt_after_cconv]: 0.00011622, [1]
    [Cycle 1]: 0.00011002, [11]
        [c_1]: 1.179e-05
        [parameter_eliminate]: 2.09e-06
        [updatestate_depend_eliminate]: 4.52e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.59e-06
        [cse]: 1.526e-05
        [call_graph_tuple_transform]: 1.317e-05
        [tuple_list_get_item_eliminator]: 3.58e-06
        [none_parameter_eliminate]: 1.63997e-06
        [renormalize]: 6.79982e-07
        [switch_simplify]: 3.71999e-06
[remove_dup_value]: 1.125e-05
[partial_unused_args_eliminate]: 1.82999e-06
[environ_conv]: 4.61002e-06
[add_recomputation]: 3.274e-05
[cse_after_recomputation]: 2.094e-05, [1]
    [Cycle 1]: 1.527e-05, [1]
        [cse]: 8.65001e-06
[auto_monad_reorder]: 1.04e-05
[get_jit_bprop_graph]: 1.69e-06
[rewriter_after_jit_bprop_graph]: 4.62e-06
[opt_after_jit_grad]: 0.00043895
[symbol_engine_optimizer]: 6.458e-05, [1]
    [Cycle 1]: 5.792e-05, [6]
        [build]: 3.05998e-06
        [elim_shapecalc]: 6.33e-06
        [elim_not_effective]: 9.67001e-06
        [opt_reshape]: 4.58999e-06
        [fold_const_symbol]: 6.29001e-06
        [renormalize]: 4.19997e-07
[validate]: 3.141e-05
[backend_pass]: 8.39995e-07
[task_emit]: 1.8e-05
[execute]: 1.21002e-06
Sums
    bootstrap                            :     0.000259s :  1.12%
    type_inference                       :     0.016225s : 70.41%
    event_method                         :     0.000053s :  0.23%
    auto_monad                           :     0.000091s :  0.40%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.11%
    rewriter_before_opt_a                :     0.000087s :  0.38%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.27%
    jit_opt_a.loop_unroll                :     0.000043s :  0.18%
    jit_opt_a.a_1                        :     0.000872s :  3.78%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.13%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000016s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.07%
    jit_opt_a.merge_forward              :     0.000010s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.15%
    jit_opt_a.j_node_and_user_rematch    :     0.000033s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001678s :  7.28%
    jit_opt_a.replace_old_param          :     0.000056s :  0.25%
    jit_opt_a.inline_without_move        :     0.000056s :  0.24%
    jit_opt_a.renormalize                :     0.001490s :  6.46%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.06%
    jit_opt_a.cse                        :     0.000028s :  0.12%
    jit_opt_a.replace_applicator         :     0.000016s :  0.07%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.03%
    rewriter_after_opt_a                 :     0.000035s :  0.15%
    convert_after_rewriter               :     0.000006s :  0.02%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000552s :  2.40%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.05%
    cconv                                :     0.000018s :  0.08%
    loop_unroll                          :     0.000435s :  1.89%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.05%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.07%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000011s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000033s :  0.14%
    cse_after_recomputation.cse          :     0.000009s :  0.04%
    auto_monad_reorder                   :     0.000010s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000439s :  1.90%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000018s :  0.08%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000228    50
 0.64% :     0.000001s :      1: substitution.elim_not_effective
 0.56% :     0.000001s :      1: substitution.fold_const_symbol
 1.60% :     0.000004s :      1: substitution.graph_param_transform
66.13% :     0.000150s :      6: substitution.inline
 7.35% :     0.000017s :      2: substitution.inline_without_move
 2.74% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.31% :     0.000003s :      2: substitution.minmaximum_grad
 1.61% :     0.000004s :      3: substitution.partial_eliminate
 2.87% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.01% :     0.000005s :      2: substitution.replace_applicator
 1.74% :     0.000004s :      5: substitution.replace_old_param
 1.07% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.87% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.74% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 5.75% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.016144     2
93.60% :     0.015110s :      1: type_inference.infer
 6.40% :     0.001034s :      1: type_inference.specialize
------[replace.]   0.000068     8
78.58% :     0.000054s :      6: replace.inline
21.42% :     0.000015s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000155     8
94.75% :     0.000147s :      6: match.inline
 5.25% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000169  1109
 1.78% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.59% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.40% :     0.000002s :     18: predicate.addn_check_dump
 1.55% :     0.000003s :     18: predicate.addn_zero_filter
 2.01% :     0.000003s :     18: predicate.arithmetic_simplify
 1.49% :     0.000003s :     18: predicate.cast_eliminate
 0.12% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.30% :     0.000002s :     18: predicate.compare_switch_simplify
 1.61% :     0.000003s :     18: predicate.depend_value_elim
 1.46% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.47% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.32% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.41% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      1: predicate.elim_not_effective
 0.36% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.56% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.33% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.31% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.34% :     0.000002s :     18: predicate.environ_get_eliminate
 1.26% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.46% :     0.000002s :     12: predicate.get_grad_eliminate
 0.11% :     0.000000s :      1: predicate.graph_param_transform
 4.25% :     0.000007s :     28: predicate.inline
 3.34% :     0.000006s :     33: predicate.inline_without_move
 0.52% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.39% :     0.000002s :     12: predicate.less_batch_normalization
 1.69% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.79% :     0.000003s :     21: predicate.load_eliminater
 0.55% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.88% :     0.000007s :     44: predicate.loop_unroll_before_grad
 1.94% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.33% :     0.000002s :     18: predicate.merge_addn
 1.35% :     0.000002s :     18: predicate.minmaximum_grad
 0.65% :     0.000001s :      1: predicate.mutable_eliminate
 0.23% :     0.000000s :      1: predicate.opt_reshape
 2.04% :     0.000003s :     21: predicate.partial_eliminate
 1.36% :     0.000002s :     18: predicate.print_const_string_wrapper
 2.10% :     0.000004s :     18: predicate.reduce_eliminate
 1.59% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.42% :     0.000002s :     22: predicate.replace_applicator
 1.68% :     0.000003s :     33: predicate.replace_old_param
 0.11% :     0.000000s :      1: predicate.reset_defer_inline
 1.55% :     0.000003s :     18: predicate.reshape_eliminate
 1.45% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.43% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.55% :     0.000003s :     18: predicate.same_eliminate
 0.63% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.27% :     0.000000s :      2: predicate.special_op_eliminate
 1.19% :     0.000002s :     12: predicate.specialize_transform
 1.57% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.48% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.61% :     0.000004s :     26: predicate.switch_defer_inline
 2.41% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.70% :     0.000011s :     71: predicate.switch_simplify
 1.47% :     0.000002s :     18: predicate.tile_eliminate
 1.40% :     0.000002s :     18: predicate.transpose_eliminate
 1.54% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.54% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.74% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.71% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.84% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.58% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000003s :     18: predicate.value_based_eliminate
 0.12% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.25% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001204    22
76.38% :     0.000920s :     13: func_graph_cloner_run.FuncGraphClonerGraph
23.62% :     0.000284s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.030062    76
 0.12% :     0.000036s :      1: add_recomputation
 0.32% :     0.000095s :      1: auto_monad
 0.04% :     0.000013s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.91% :     0.000273s :      1: bootstrap
 0.07% :     0.000021s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.08% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.19% :     0.000058s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000007s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
28.26% :     0.008495s :      1: jit_opt_a
 0.40% :     0.000119s :      1: jit_opt_after_cconv
 0.14% :     0.000043s :      1: jit_opt_b
 1.47% :     0.000443s :      1: loop_unroll
 1.87% :     0.000561s :      1: mutable_eliminate
 4.10% :     0.001231s :     26: opt.transform.jit_opt_a
 0.10% :     0.000029s :      4: opt.transform.jit_opt_after_cconv
 0.05% :     0.000015s :      4: opt.transform.jit_opt_b
 0.03% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000024s :      4: opt.transform.symbol_engine_opt
 1.49% :     0.000447s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.09% :     0.000028s :      1: py_interpret_to_execute
 0.03% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000014s :      1: remove_dup_value
 3.75% :     0.001129s :      1: renormalize.infer
 1.17% :     0.000353s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000038s :      1: rewriter_after_opt_a
 0.30% :     0.000089s :      1: rewriter_before_opt_a
 0.22% :     0.000067s :      1: symbol_engine_optimizer
 0.07% :     0.000022s :      1: task_emit
54.02% :     0.016239s :      1: type_inference
 0.16% :     0.000048s :      1: validate


TotalTime = 0.0340117, [33]
[bootstrap]: 0.00026651
[type_inference]: 0.0124316
[event_method]: 4.89e-05
[auto_monad]: 9.736e-05
[graph_reusing]: 7.18e-06
[pre_auto_parallel]: 2.69001e-06
[py_interpret_to_execute]: 2.721e-05
[rewriter_before_opt_a]: 9.507e-05
[expand_dump_flag]: 3.07002e-06
[jit_opt_a]: 0.0117556, [3]
    [Cycle 1]: 0.0057228, [27]
        [switch_simplify]: 6.277e-05
        [loop_unroll]: 3.737e-05
        [a_1]: 0.00082869
        [with_stream_mark]: 2.491e-05
        [recompute_prepare]: 1.744e-05
        [updatestate_depend_eliminate]: 7.98999e-06
        [updatestate_assign_eliminate]: 6.29999e-06
        [updatestate_loads_eliminate]: 5.99e-06
        [parameter_eliminate]: 2.55002e-06
        [specialize_transform]: 1.353e-05
        [updatestate_useless_node_eliminater]: 1.176e-05
        [accelerated_algorithm]: 1.312e-05
        [meta_shard_fg_expand]: 3.26001e-06
        [get_grad_eliminate_]: 1.153e-05
        [merge_forward]: 9.34e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.726e-05
        [j_node_and_user_rematch]: 2.247e-05
        [meta_fg_expand]: 0.00148746
        [replace_old_param]: 5.294e-05
        [inline_without_move]: 5.226e-05
        [renormalize]: 0.00250156
        [add_forward_monad_depend]: 9.38002e-06
        [auto_monad_grad]: 4.28999e-06
        [auto_monad_eliminator]: 5.037e-05
        [cse]: 0.00018186
        [replace_applicator]: 7.037e-05
    [Cycle 2]: 0.00214416, [27]
        [switch_simplify]: 3.739e-05
        [loop_unroll]: 3.654e-05
        [a_1]: 0.00112376
        [with_stream_mark]: 1.411e-05
        [recompute_prepare]: 8.68001e-06
        [updatestate_depend_eliminate]: 4.3e-06
        [updatestate_assign_eliminate]: 3.31001e-06
        [updatestate_loads_eliminate]: 2.98998e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 7.45998e-06
        [updatestate_useless_node_eliminater]: 7.3e-06
        [accelerated_algorithm]: 7.46001e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 7.03e-06
        [merge_forward]: 3.85e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.485e-05
        [j_node_and_user_rematch]: 1.14e-05
        [meta_fg_expand]: 4.436e-05
        [replace_old_param]: 1.02e-05
        [inline_without_move]: 6.84001e-06
        [renormalize]: 0.000584
        [add_forward_monad_depend]: 4.69998e-06
        [auto_monad_grad]: 1.25999e-06
        [auto_monad_eliminator]: 1.341e-05
        [cse]: 2.305e-05
        [replace_applicator]: 1.436e-05
    [Cycle 3]: 0.00040777, [27]
        [switch_simplify]: 7.33e-06
        [loop_unroll]: 6.82002e-06
        [a_1]: 0.00014454
        [with_stream_mark]: 1.007e-05
        [recompute_prepare]: 6.76999e-06
        [updatestate_depend_eliminate]: 3.61001e-06
        [updatestate_assign_eliminate]: 3.25e-06
        [updatestate_loads_eliminate]: 2.96001e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 6.79999e-06
        [updatestate_useless_node_eliminater]: 6.86001e-06
        [accelerated_algorithm]: 6.91001e-06
        [meta_shard_fg_expand]: 1.49998e-06
        [get_grad_eliminate_]: 6.54999e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 1.81e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.62e-05
        [j_node_and_user_rematch]: 1.169e-05
        [meta_fg_expand]: 2.26e-06
        [replace_old_param]: 8.65001e-06
        [inline_without_move]: 6.73e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.53002e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 7.13998e-06
        [cse]: 1.589e-05
        [replace_applicator]: 6.69999e-06
[py_interpret_to_execute_after_opt_a]: 1.17e-05
[rewriter_after_opt_a]: 4.154e-05
[convert_after_rewriter]: 7.82002e-06
[order_py_execute_after_rewriter]: 5.46e-06
[mutable_eliminate]: 0.00062496
[jit_opt_b]: 5.801e-05, [1]
    [Cycle 1]: 5.126e-05, [2]
        [frontend_op_eliminate]: 2.099e-05
        [inline_after_opt_a]: 1.896e-05
[cconv]: 1.932e-05
[loop_unroll]: 0.00041659
[jit_opt_after_cconv]: 0.00022539, [1]
    [Cycle 1]: 0.00021885, [11]
        [c_1]: 3.081e-05
        [parameter_eliminate]: 2.69001e-06
        [updatestate_depend_eliminate]: 6.39001e-06
        [updatestate_assign_eliminate]: 3.4e-06
        [updatestate_loads_eliminate]: 3.92998e-06
        [cse]: 2.712e-05
        [call_graph_tuple_transform]: 2.257e-05
        [tuple_list_get_item_eliminator]: 7.66999e-06
        [none_parameter_eliminate]: 1.00999e-06
        [renormalize]: 3.00002e-07
        [switch_simplify]: 7.12002e-06
[remove_dup_value]: 1.282e-05
[partial_unused_args_eliminate]: 1.77999e-06
[environ_conv]: 6.43e-06
[add_recomputation]: 4.792e-05
[cse_after_recomputation]: 2.958e-05, [1]
    [Cycle 1]: 2.319e-05, [1]
        [cse]: 1.694e-05
[auto_monad_reorder]: 1.829e-05
[get_jit_bprop_graph]: 1.77999e-06
[rewriter_after_jit_bprop_graph]: 5.30999e-06
[opt_after_jit_grad]: 0.00045755
[symbol_engine_optimizer]: 8.341e-05, [1]
    [Cycle 1]: 7.698e-05, [6]
        [build]: 3.39001e-06
        [elim_shapecalc]: 9.15001e-06
        [elim_not_effective]: 1.629e-05
        [opt_reshape]: 7.42002e-06
        [fold_const_symbol]: 1.282e-05
        [renormalize]: 4.80009e-07
[validate]: 3.813e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.00695553
[execute]: 5.52999e-06
Sums
    bootstrap                            :     0.000267s :  0.90%
    type_inference                       :     0.012432s : 41.87%
    event_method                         :     0.000049s :  0.16%
    auto_monad                           :     0.000097s :  0.33%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000027s :  0.09%
    rewriter_before_opt_a                :     0.000095s :  0.32%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000107s :  0.36%
    jit_opt_a.loop_unroll                :     0.000081s :  0.27%
    jit_opt_a.a_1                        :     0.002097s :  7.06%
    jit_opt_a.with_stream_mark           :     0.000049s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000028s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000027s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.08%
    jit_opt_a.merge_forward              :     0.000017s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000058s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000046s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001534s :  5.17%
    jit_opt_a.replace_old_param          :     0.000072s :  0.24%
    jit_opt_a.inline_without_move        :     0.000066s :  0.22%
    jit_opt_a.renormalize                :     0.003086s : 10.39%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000006s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000071s :  0.24%
    jit_opt_a.cse                        :     0.000221s :  0.74%
    jit_opt_a.replace_applicator         :     0.000091s :  0.31%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000042s :  0.14%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000625s :  2.10%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.06%
    cconv                                :     0.000019s :  0.07%
    loop_unroll                          :     0.000417s :  1.40%
    jit_opt_after_cconv.c_1              :     0.000031s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000013s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000048s :  0.16%
    cse_after_recomputation.cse          :     0.000017s :  0.06%
    auto_monad_reorder                   :     0.000018s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000458s :  1.54%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006956s : 23.43%
    execute                              :     0.000006s :  0.02%


Time group info:
------[substitution.]   0.000529   118
 5.48% :     0.000029s :      1: substitution.arithmetic_simplify
 0.46% :     0.000002s :      4: substitution.elim_not_effective
 0.42% :     0.000002s :      4: substitution.fold_const_symbol
 1.11% :     0.000006s :      5: substitution.graph_param_transform
60.49% :     0.000320s :     12: substitution.inline
 3.38% :     0.000018s :      2: substitution.inline_without_move
 1.72% :     0.000009s :     16: substitution.j_node_and_user_rematch
 1.93% :     0.000010s :      7: substitution.minmaximum_grad
 1.40% :     0.000007s :      5: substitution.partial_eliminate
 2.11% :     0.000011s :     16: substitution.remove_not_recompute_node
 5.38% :     0.000028s :     10: substitution.replace_applicator
 1.39% :     0.000007s :      7: substitution.replace_old_param
 0.47% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.14% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.66% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.46% :     0.000039s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012351     2
90.03% :     0.011119s :      1: type_inference.infer
 9.97% :     0.001232s :      1: type_inference.specialize
------[replace.]   0.000147    20
 3.62% :     0.000005s :      1: replace.arithmetic_simplify
62.41% :     0.000092s :     12: replace.inline
33.97% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000361    20
 7.66% :     0.000028s :      1: match.arithmetic_simplify
86.81% :     0.000314s :     12: match.inline
 5.53% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000390  2790
 1.59% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.45% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.41% :     0.000006s :     46: predicate.addn_check_dump
 1.66% :     0.000006s :     46: predicate.addn_zero_filter
 2.33% :     0.000009s :     47: predicate.arithmetic_simplify
 1.80% :     0.000007s :     47: predicate.cast_eliminate
 0.19% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.38% :     0.000005s :     46: predicate.compare_switch_simplify
 1.58% :     0.000006s :     46: predicate.depend_value_elim
 1.53% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.54% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.55% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.33% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      5: predicate.elim_not_effective
 0.23% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.48% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.43% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.46% :     0.000006s :     47: predicate.environ_get_eliminate
 1.47% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      5: predicate.fold_const_symbol
 0.90% :     0.000004s :     21: predicate.get_grad_eliminate
 0.15% :     0.000001s :      5: predicate.graph_param_transform
 4.29% :     0.000017s :     76: predicate.inline
 1.72% :     0.000007s :     42: predicate.inline_without_move
 0.37% :     0.000001s :     21: predicate.j_node_and_user_rematch
 1.05% :     0.000004s :     21: predicate.less_batch_normalization
 1.87% :     0.000007s :     54: predicate.list_to_tuple_eliminator_
 1.93% :     0.000008s :     59: predicate.load_eliminater
 0.56% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.17% :     0.000012s :     89: predicate.loop_unroll_before_grad
 1.75% :     0.000007s :     52: predicate.make_slice_get_slice_eliminator
 1.39% :     0.000005s :     46: predicate.merge_addn
 1.59% :     0.000006s :     47: predicate.minmaximum_grad
 0.52% :     0.000002s :      5: predicate.mutable_eliminate
 0.21% :     0.000001s :      5: predicate.opt_reshape
 2.40% :     0.000009s :     59: predicate.partial_eliminate
 1.48% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.89% :     0.000007s :     47: predicate.reduce_eliminate
 1.78% :     0.000007s :     54: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.52% :     0.000010s :    105: predicate.replace_applicator
 0.94% :     0.000004s :     42: predicate.replace_old_param
 0.12% :     0.000000s :      5: predicate.reset_defer_inline
 2.03% :     0.000008s :     47: predicate.reshape_eliminate
 1.44% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.33% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.56% :     0.000006s :     46: predicate.same_eliminate
 0.46% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.45% :     0.000002s :     10: predicate.special_op_eliminate
 0.86% :     0.000003s :     21: predicate.specialize_transform
 1.61% :     0.000006s :     46: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.20% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.58% :     0.000010s :     66: predicate.switch_defer_inline
 2.43% :     0.000009s :     66: predicate.switch_layer_defer_inline
 6.05% :     0.000024s :    160: predicate.switch_simplify
 1.52% :     0.000006s :     47: predicate.tile_eliminate
 1.55% :     0.000006s :     47: predicate.transpose_eliminate
 1.89% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.73% :     0.000007s :     47: predicate.tuple_list_get_item_depend_reorder
 3.08% :     0.000012s :     64: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.82% :     0.000007s :     54: predicate.tuple_to_list_eliminator_
 1.89% :     0.000007s :     59: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000011s :     80: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000007s :     46: predicate.value_based_eliminate
 0.19% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.24% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001591    29
61.46% :     0.000978s :     13: func_graph_cloner_run.FuncGraphClonerGraph
38.54% :     0.000613s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.039957    91
 0.13% :     0.000051s :      1: add_recomputation
 0.26% :     0.000102s :      1: auto_monad
 0.05% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.70% :     0.000279s :      1: bootstrap
 0.05% :     0.000022s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000032s :      1: cse_after_recomputation
 0.04% :     0.000014s :      1: environ_conv
 0.13% :     0.000053s :      1: event_method
 0.02% :     0.000009s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
29.43% :     0.011760s :      1: jit_opt_a
 0.57% :     0.000228s :      1: jit_opt_after_cconv
 0.15% :     0.000061s :      1: jit_opt_b
 1.06% :     0.000424s :      1: loop_unroll
 1.58% :     0.000633s :      1: mutable_eliminate
 6.79% :     0.002714s :     39: opt.transform.jit_opt_a
 0.16% :     0.000064s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000034s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000043s :      4: opt.transform.symbol_engine_opt
 1.16% :     0.000465s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000030s :      1: py_interpret_to_execute
 0.03% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000015s :      1: remove_dup_value
 4.31% :     0.001722s :      2: renormalize.infer
 3.38% :     0.001350s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000045s :      1: rewriter_after_opt_a
 0.24% :     0.000098s :      1: rewriter_before_opt_a
 0.21% :     0.000086s :      1: symbol_engine_optimizer
17.44% :     0.006967s :      1: task_emit
31.15% :     0.012446s :      1: type_inference
 0.15% :     0.000061s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.272.672 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c36bc10> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.273.307 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.277.071 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e9d30> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.277.705 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0100523, [30]
[bootstrap]: 0.00025134
[type_inference]: 0.00348738
[event_method]: 1.517e-05
[auto_monad]: 4.713e-05
[graph_reusing]: 4.19002e-06
[pre_auto_parallel]: 1.91998e-06
[py_interpret_to_execute]: 2.411e-05
[rewriter_before_opt_a]: 6.111e-05
[expand_dump_flag]: 2.46998e-06
[jit_opt_a]: 0.00426666, [2]
    [Cycle 1]: 0.00122304, [27]
        [switch_simplify]: 4.183e-05
        [loop_unroll]: 2.822e-05
        [a_1]: 0.00045402
        [with_stream_mark]: 1.144e-05
        [recompute_prepare]: 6.54001e-06
        [updatestate_depend_eliminate]: 2.89001e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 5.56e-06
        [updatestate_useless_node_eliminater]: 5.15999e-06
        [accelerated_algorithm]: 5.15999e-06
        [meta_shard_fg_expand]: 1.25001e-06
        [get_grad_eliminate_]: 5.04998e-06
        [merge_forward]: 2.52001e-06
        [cell_reuse_recompute_pass]: 8.50006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.174e-05
        [j_node_and_user_rematch]: 8.17e-06
        [meta_fg_expand]: 1.86e-06
        [replace_old_param]: 7.34002e-06
        [inline_without_move]: 5.04e-06
        [renormalize]: 0.00041497
        [add_forward_monad_depend]: 4.62998e-06
        [auto_monad_grad]: 1.27999e-06
        [auto_monad_eliminator]: 9.85002e-06
        [cse]: 1.644e-05
        [replace_applicator]: 1.019e-05
    [Cycle 2]: 0.00032025, [27]
        [switch_simplify]: 5.30999e-06
        [loop_unroll]: 4.82e-06
        [a_1]: 8.703e-05
        [with_stream_mark]: 8.08999e-06
        [recompute_prepare]: 5.04e-06
        [updatestate_depend_eliminate]: 2.34001e-06
        [updatestate_assign_eliminate]: 2.13002e-06
        [updatestate_loads_eliminate]: 1.79998e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.90999e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 4.90999e-06
        [meta_shard_fg_expand]: 1.04998e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.27001e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.138e-05
        [j_node_and_user_rematch]: 8.06001e-06
        [meta_fg_expand]: 1.60001e-06
        [replace_old_param]: 7.13e-06
        [inline_without_move]: 4.55999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.04e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 1.424e-05
        [cse]: 1.107e-05
        [replace_applicator]: 5.35999e-06
[py_interpret_to_execute_after_opt_a]: 7.3e-06
[rewriter_after_opt_a]: 2.596e-05
[convert_after_rewriter]: 5.45001e-06
[order_py_execute_after_rewriter]: 4.50001e-06
[mutable_eliminate]: 0.00046387
[jit_opt_b]: 5.144e-05, [1]
    [Cycle 1]: 4.444e-05, [2]
        [frontend_op_eliminate]: 1.877e-05
        [inline_after_opt_a]: 1.437e-05
[cconv]: 2.173e-05
[loop_unroll]: 0.00041294
[jit_opt_after_cconv]: 0.00013359, [1]
    [Cycle 1]: 0.00012672, [11]
        [c_1]: 2.035e-05
        [parameter_eliminate]: 2.71e-06
        [updatestate_depend_eliminate]: 5.07e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.625e-05
        [call_graph_tuple_transform]: 1.773e-05
        [tuple_list_get_item_eliminator]: 4.98001e-06
        [none_parameter_eliminate]: 1.15001e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 5.10001e-06
[remove_dup_value]: 8.89e-06
[partial_unused_args_eliminate]: 1.77001e-06
[environ_conv]: 3.93999e-06
[add_recomputation]: 3.497e-05
[cse_after_recomputation]: 2.097e-05, [1]
    [Cycle 1]: 1.534e-05, [1]
        [cse]: 9.39e-06
[auto_monad_reorder]: 1.117e-05
[get_jit_bprop_graph]: 1.57999e-06
[rewriter_after_jit_bprop_graph]: 4.85999e-06
[opt_after_jit_grad]: 0.00044002
[symbol_engine_optimizer]: 6.74e-05, [1]
    [Cycle 1]: 6.13e-05, [6]
        [build]: 2.83e-06
        [elim_shapecalc]: 7.1e-06
        [elim_not_effective]: 1.124e-05
        [opt_reshape]: 5.67001e-06
        [fold_const_symbol]: 7.88001e-06
        [renormalize]: 4.89992e-07
[validate]: 2.498e-05
Sums
    bootstrap                            :     0.000251s :  3.70%
    type_inference                       :     0.003487s : 51.30%
    event_method                         :     0.000015s :  0.22%
    auto_monad                           :     0.000047s :  0.69%
    graph_reusing                        :     0.000004s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000024s :  0.35%
    rewriter_before_opt_a                :     0.000061s :  0.90%
    expand_dump_flag                     :     0.000002s :  0.04%
    jit_opt_a.switch_simplify            :     0.000047s :  0.69%
    jit_opt_a.loop_unroll                :     0.000033s :  0.49%
    jit_opt_a.a_1                        :     0.000541s :  7.96%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000010s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.15%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.24%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.05%
    jit_opt_a.replace_old_param          :     0.000014s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000415s :  6.11%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.35%
    jit_opt_a.cse                        :     0.000028s :  0.40%
    jit_opt_a.replace_applicator         :     0.000016s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.11%
    rewriter_after_opt_a                 :     0.000026s :  0.38%
    convert_after_rewriter               :     0.000005s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000464s :  6.82%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.28%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.21%
    cconv                                :     0.000022s :  0.32%
    loop_unroll                          :     0.000413s :  6.07%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.30%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.26%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.08%
    remove_dup_value                     :     0.000009s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000035s :  0.51%
    cse_after_recomputation.cse          :     0.000009s :  0.14%
    auto_monad_reorder                   :     0.000011s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000440s :  6.47%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000025s :  0.37%


Time group info:
------[substitution.]   0.000148    23
 1.21% :     0.000002s :      2: substitution.elim_not_effective
 0.81% :     0.000001s :      2: substitution.fold_const_symbol
 3.33% :     0.000005s :      3: substitution.graph_param_transform
82.43% :     0.000122s :      4: substitution.inline
 1.78% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.65% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.10% :     0.000003s :      2: substitution.replace_old_param
 5.69% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003443     2
80.30% :     0.002765s :      1: type_inference.infer
19.70% :     0.000678s :      1: type_inference.specialize
------[replace.]   0.000058     6
66.08% :     0.000038s :      4: replace.inline
33.92% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000127     6
94.19% :     0.000119s :      4: match.inline
 5.81% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.40% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000001s :     10: predicate.addn_zero_filter
 2.29% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.16% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000001s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.26% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.22% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000000s :      3: predicate.graph_param_transform
 5.51% :     0.000006s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.14% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.24% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.16% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.57% :     0.000002s :      3: predicate.mutable_eliminate
 0.59% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000002s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.59% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.62% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.57% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.96% :     0.000009s :     53: predicate.switch_simplify
 1.32% :     0.000001s :     10: predicate.tile_eliminate
 1.29% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.54% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.73% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000454    12
49.34% :     0.000224s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.29% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.38% :     0.000206s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011270    72
 0.34% :     0.000038s :      1: add_recomputation
 0.45% :     0.000050s :      1: auto_monad
 0.12% :     0.000014s :      1: auto_monad_reorder
 2.33% :     0.000263s :      1: bootstrap
 0.22% :     0.000024s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.20% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000006s :      1: environ_conv
 0.16% :     0.000019s :      1: event_method
 0.04% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000006s :      1: graph_reusing
37.88% :     0.004269s :      1: jit_opt_a
 1.21% :     0.000136s :      1: jit_opt_after_cconv
 0.48% :     0.000054s :      1: jit_opt_b
 3.74% :     0.000422s :      1: loop_unroll
 4.19% :     0.000473s :      1: mutable_eliminate
 6.46% :     0.000728s :     26: opt.transform.jit_opt_a
 0.40% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.24% :     0.000027s :      4: opt.transform.jit_opt_b
 0.10% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000029s :      4: opt.transform.symbol_engine_opt
 3.97% :     0.000447s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.23% :     0.000026s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000011s :      1: remove_dup_value
 1.74% :     0.000196s :      1: renormalize.infer
 1.89% :     0.000213s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000029s :      1: rewriter_after_opt_a
 0.56% :     0.000064s :      1: rewriter_before_opt_a
 0.62% :     0.000070s :      1: symbol_engine_optimizer
31.03% :     0.003497s :      1: type_inference


TotalTime = 0.0180061, [33]
[bootstrap]: 0.00025094
[type_inference]: 0.00426895
[event_method]: 1.664e-05
[auto_monad]: 6.234e-05
[graph_reusing]: 5.74999e-06
[pre_auto_parallel]: 2.53998e-06
[py_interpret_to_execute]: 3.025e-05
[rewriter_before_opt_a]: 7.378e-05
[expand_dump_flag]: 3.38e-06
[jit_opt_a]: 0.00515058, [2]
    [Cycle 1]: 0.00153743, [27]
        [switch_simplify]: 5.516e-05
        [loop_unroll]: 2.898e-05
        [a_1]: 0.00051771
        [with_stream_mark]: 1.733e-05
        [recompute_prepare]: 7.73999e-06
        [updatestate_depend_eliminate]: 4.65999e-06
        [updatestate_assign_eliminate]: 3.55998e-06
        [updatestate_loads_eliminate]: 2.68998e-06
        [parameter_eliminate]: 2.27999e-06
        [specialize_transform]: 5.84999e-06
        [updatestate_useless_node_eliminater]: 4.77e-06
        [accelerated_algorithm]: 5.07999e-06
        [meta_shard_fg_expand]: 2.21998e-06
        [get_grad_eliminate_]: 4.95999e-06
        [merge_forward]: 3.6e-06
        [cell_reuse_recompute_pass]: 1.57999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.37e-05
        [j_node_and_user_rematch]: 8.84e-06
        [meta_fg_expand]: 2.36e-06
        [replace_old_param]: 8.28999e-06
        [inline_without_move]: 5.28002e-06
        [renormalize]: 0.00059849
        [add_forward_monad_depend]: 5.66998e-06
        [auto_monad_grad]: 2.41e-06
        [auto_monad_eliminator]: 1.414e-05
        [cse]: 2.721e-05
        [replace_applicator]: 1.295e-05
    [Cycle 2]: 0.00032523, [27]
        [switch_simplify]: 5.52001e-06
        [loop_unroll]: 5.12999e-06
        [a_1]: 9.442e-05
        [with_stream_mark]: 9.24998e-06
        [recompute_prepare]: 5.03002e-06
        [updatestate_depend_eliminate]: 2.96999e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.72001e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 4.79002e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.09e-06
        [meta_shard_fg_expand]: 1.20001e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 2.66999e-06
        [cell_reuse_recompute_pass]: 1.53002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.243e-05
        [j_node_and_user_rematch]: 7.84002e-06
        [meta_fg_expand]: 1.81e-06
        [replace_old_param]: 7.33e-06
        [inline_without_move]: 4.63999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.24003e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 5.19e-06
        [cse]: 1.096e-05
        [replace_applicator]: 4.92999e-06
[py_interpret_to_execute_after_opt_a]: 7.38e-06
[rewriter_after_opt_a]: 3.411e-05
[convert_after_rewriter]: 7e-06
[order_py_execute_after_rewriter]: 5.11002e-06
[mutable_eliminate]: 0.0005529
[jit_opt_b]: 5.382e-05, [1]
    [Cycle 1]: 4.6e-05, [2]
        [frontend_op_eliminate]: 1.943e-05
        [inline_after_opt_a]: 1.441e-05
[cconv]: 2.389e-05
[loop_unroll]: 0.00041748
[jit_opt_after_cconv]: 0.00013927, [1]
    [Cycle 1]: 0.00013274, [11]
        [c_1]: 2.251e-05
        [parameter_eliminate]: 2.86e-06
        [updatestate_depend_eliminate]: 5.42001e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [cse]: 1.853e-05
        [call_graph_tuple_transform]: 1.794e-05
        [tuple_list_get_item_eliminator]: 5.10001e-06
        [none_parameter_eliminate]: 1.38002e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 4.98001e-06
[remove_dup_value]: 1.397e-05
[partial_unused_args_eliminate]: 2.69999e-06
[environ_conv]: 5.99e-06
[add_recomputation]: 5.116e-05
[cse_after_recomputation]: 2.151e-05, [1]
    [Cycle 1]: 1.556e-05, [1]
        [cse]: 9.51e-06
[auto_monad_reorder]: 1.57e-05
[get_jit_bprop_graph]: 1.43002e-06
[rewriter_after_jit_bprop_graph]: 6.01e-06
[opt_after_jit_grad]: 0.00047748
[symbol_engine_optimizer]: 7.148e-05, [1]
    [Cycle 1]: 6.489e-05, [6]
        [build]: 4.08001e-06
        [elim_shapecalc]: 7.27002e-06
        [elim_not_effective]: 1.254e-05
        [opt_reshape]: 5.35999e-06
        [fold_const_symbol]: 8.10999e-06
        [renormalize]: 5.39992e-07
[validate]: 3.638e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.00595642
[execute]: 7.03998e-06
Sums
    bootstrap                            :     0.000251s :  1.78%
    type_inference                       :     0.004269s : 30.32%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000062s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.21%
    rewriter_before_opt_a                :     0.000074s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.43%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000612s :  4.35%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000599s :  4.25%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.14%
    jit_opt_a.cse                        :     0.000038s :  0.27%
    jit_opt_a.replace_applicator         :     0.000018s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.05%
    rewriter_after_opt_a                 :     0.000034s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000553s :  3.93%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000024s :  0.17%
    loop_unroll                          :     0.000417s :  2.97%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000014s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000051s :  0.36%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000016s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000477s :  3.39%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.26%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005956s : 42.31%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000193    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.86% :     0.000006s :      3: substitution.graph_param_transform
82.97% :     0.000160s :      4: substitution.inline
 1.51% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.31% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.17% :     0.000004s :      2: substitution.replace_old_param
 6.48% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004211     2
81.66% :     0.003438s :      1: type_inference.infer
18.34% :     0.000772s :      1: type_inference.specialize
------[replace.]   0.000065     6
68.71% :     0.000045s :      4: replace.inline
31.29% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000169     6
93.23% :     0.000158s :      4: match.inline
 6.77% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.26% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.11% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 2.08% :     0.000002s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.70% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.56% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.75% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.61% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.96% :     0.000001s :      6: predicate.less_batch_normalization
 1.69% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.06% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.43% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.14% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.57% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.36% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.82% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.84% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.01% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.10% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.48% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.58% :     0.000009s :     53: predicate.switch_simplify
 1.28% :     0.000001s :     10: predicate.tile_eliminate
 1.28% :     0.000001s :     10: predicate.transpose_eliminate
 1.34% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.94% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.72% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.35% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000516    12
49.23% :     0.000254s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.27% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.49% :     0.000235s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019533    76
 0.28% :     0.000054s :      1: add_recomputation
 0.34% :     0.000066s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.35% :     0.000264s :      1: bootstrap
 0.14% :     0.000027s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.38% :     0.005153s :      1: jit_opt_a
 0.73% :     0.000142s :      1: jit_opt_after_cconv
 0.29% :     0.000056s :      1: jit_opt_b
 2.18% :     0.000425s :      1: loop_unroll
 3.13% :     0.000611s :      1: mutable_eliminate
 4.19% :     0.000819s :     26: opt.transform.jit_opt_a
 0.24% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000028s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.49% :     0.000487s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.60% :     0.000312s :      1: renormalize.infer
 1.43% :     0.000279s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000037s :      1: rewriter_after_opt_a
 0.39% :     0.000077s :      1: rewriter_before_opt_a
 0.38% :     0.000074s :      1: symbol_engine_optimizer
30.55% :     0.005968s :      1: task_emit
21.92% :     0.004281s :      1: type_inference
 0.29% :     0.000057s :      1: validate


TotalTime = 0.022108, [33]
[bootstrap]: 0.00021747
[type_inference]: 0.0105529
[event_method]: 4.652e-05
[auto_monad]: 9.355e-05
[graph_reusing]: 6.19999e-06
[pre_auto_parallel]: 2.56e-06
[py_interpret_to_execute]: 2.52e-05
[rewriter_before_opt_a]: 9.229e-05
[expand_dump_flag]: 3.3e-06
[jit_opt_a]: 0.00879636, [2]
    [Cycle 1]: 0.0048409, [27]
        [switch_simplify]: 6.739e-05
        [loop_unroll]: 3.921e-05
        [a_1]: 0.00088215
        [with_stream_mark]: 2.616e-05
        [recompute_prepare]: 5.779e-05
        [updatestate_depend_eliminate]: 8.80001e-06
        [updatestate_assign_eliminate]: 6.28002e-06
        [updatestate_loads_eliminate]: 5.92001e-06
        [parameter_eliminate]: 2.93e-06
        [specialize_transform]: 1.401e-05
        [updatestate_useless_node_eliminater]: 1.237e-05
        [accelerated_algorithm]: 1.302e-05
        [meta_shard_fg_expand]: 3.66001e-06
        [get_grad_eliminate_]: 1.314e-05
        [merge_forward]: 7.76001e-06
        [cell_reuse_recompute_pass]: 9.5999e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.663e-05
        [j_node_and_user_rematch]: 2.491e-05
        [meta_fg_expand]: 0.00165551
        [replace_old_param]: 5.861e-05
        [inline_without_move]: 5.25e-05
        [renormalize]: 0.00158816
        [add_forward_monad_depend]: 5.82001e-06
        [auto_monad_grad]: 2.64001e-06
        [auto_monad_eliminator]: 1.33e-05
        [cse]: 2.064e-05
        [replace_applicator]: 1.299e-05
    [Cycle 2]: 0.00032513, [27]
        [switch_simplify]: 4.87e-06
        [loop_unroll]: 3.88001e-06
        [a_1]: 4.193e-05
        [with_stream_mark]: 1.104e-05
        [recompute_prepare]: 3.77002e-06
        [updatestate_depend_eliminate]: 2.61e-06
        [updatestate_assign_eliminate]: 1.74e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 1.48002e-06
        [specialize_transform]: 3.71001e-06
        [updatestate_useless_node_eliminater]: 3.91999e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 1.72999e-06
        [get_grad_eliminate_]: 4e-06
        [merge_forward]: 3.23998e-06
        [cell_reuse_recompute_pass]: 2.29999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.251e-05
        [j_node_and_user_rematch]: 5.83002e-06
        [meta_fg_expand]: 5.661e-05
        [replace_old_param]: 4.86002e-06
        [inline_without_move]: 3.43999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.40999e-06
        [auto_monad_grad]: 1.11002e-06
        [auto_monad_eliminator]: 4.67e-06
        [cse]: 1.177e-05
        [replace_applicator]: 3.93001e-06
[py_interpret_to_execute_after_opt_a]: 1.158e-05
[rewriter_after_opt_a]: 3.421e-05
[convert_after_rewriter]: 6.02001e-06
[order_py_execute_after_rewriter]: 4.27003e-06
[mutable_eliminate]: 0.00063979
[jit_opt_b]: 4.373e-05, [1]
    [Cycle 1]: 3.61e-05, [2]
        [frontend_op_eliminate]: 1.207e-05
        [inline_after_opt_a]: 1.224e-05
[cconv]: 2.341e-05
[loop_unroll]: 0.00045393
[jit_opt_after_cconv]: 0.0001294, [1]
    [Cycle 1]: 0.000122, [11]
        [c_1]: 1.285e-05
        [parameter_eliminate]: 3.44001e-06
        [updatestate_depend_eliminate]: 4.72998e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 2.074e-05
        [call_graph_tuple_transform]: 1.467e-05
        [tuple_list_get_item_eliminator]: 3.86999e-06
        [none_parameter_eliminate]: 1.62001e-06
        [renormalize]: 8.00006e-07
        [switch_simplify]: 4.34997e-06
[remove_dup_value]: 1.664e-05
[partial_unused_args_eliminate]: 2.71999e-06
[environ_conv]: 5.51e-06
[add_recomputation]: 3.893e-05
[cse_after_recomputation]: 2.295e-05, [1]
    [Cycle 1]: 1.651e-05, [1]
        [cse]: 9.20001e-06
[auto_monad_reorder]: 1.496e-05
[get_jit_bprop_graph]: 1.76e-06
[rewriter_after_jit_bprop_graph]: 6.68e-06
[opt_after_jit_grad]: 0.00050594
[symbol_engine_optimizer]: 7.009e-05, [1]
    [Cycle 1]: 6.33e-05, [6]
        [build]: 4.42998e-06
        [elim_shapecalc]: 6.72002e-06
        [elim_not_effective]: 1.087e-05
        [opt_reshape]: 4.85001e-06
        [fold_const_symbol]: 6.52001e-06
        [renormalize]: 7.00005e-07
[validate]: 3.223e-05
[backend_pass]: 8.99978e-07
[task_emit]: 2.52e-05
[execute]: 1.72999e-06
Sums
    bootstrap                            :     0.000217s :  1.22%
    type_inference                       :     0.010553s : 59.18%
    event_method                         :     0.000047s :  0.26%
    auto_monad                           :     0.000094s :  0.52%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000025s :  0.14%
    rewriter_before_opt_a                :     0.000092s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000072s :  0.41%
    jit_opt_a.loop_unroll                :     0.000043s :  0.24%
    jit_opt_a.a_1                        :     0.000924s :  5.18%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000062s :  0.35%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000018s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000019s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.10%
    jit_opt_a.merge_forward              :     0.000011s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000031s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001712s :  9.60%
    jit_opt_a.replace_old_param          :     0.000063s :  0.36%
    jit_opt_a.inline_without_move        :     0.000056s :  0.31%
    jit_opt_a.renormalize                :     0.001588s :  8.91%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.10%
    jit_opt_a.cse                        :     0.000032s :  0.18%
    jit_opt_a.replace_applicator         :     0.000017s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.06%
    rewriter_after_opt_a                 :     0.000034s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000640s :  3.59%
    jit_opt_b.frontend_op_eliminate      :     0.000012s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.07%
    cconv                                :     0.000023s :  0.13%
    loop_unroll                          :     0.000454s :  2.55%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000017s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000039s :  0.22%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000506s :  2.84%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000007s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.18%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.000025s :  0.14%
    execute                              :     0.000002s :  0.01%


Time group info:
------[substitution.]   0.000266    50
 0.70% :     0.000002s :      1: substitution.elim_not_effective
 0.50% :     0.000001s :      1: substitution.fold_const_symbol
 1.75% :     0.000005s :      1: substitution.graph_param_transform
65.85% :     0.000175s :      6: substitution.inline
 6.40% :     0.000017s :      2: substitution.inline_without_move
 2.70% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.33% :     0.000004s :      2: substitution.minmaximum_grad
 1.63% :     0.000004s :      3: substitution.partial_eliminate
 2.95% :     0.000008s :      9: substitution.remove_not_recompute_node
 2.42% :     0.000006s :      2: substitution.replace_applicator
 1.57% :     0.000004s :      5: substitution.replace_old_param
 1.38% :     0.000004s :      1: substitution.set_cell_output_no_recompute
 2.70% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.94% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 6.18% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010476     2
89.15% :     0.009340s :      1: type_inference.infer
10.85% :     0.001136s :      1: type_inference.specialize
------[replace.]   0.000067     8
77.41% :     0.000052s :      6: replace.inline
22.59% :     0.000015s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000183     8
93.90% :     0.000172s :      6: match.inline
 6.10% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000176  1109
 1.52% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.55% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.22% :     0.000002s :     18: predicate.addn_check_dump
 1.61% :     0.000003s :     18: predicate.addn_zero_filter
 1.89% :     0.000003s :     18: predicate.arithmetic_simplify
 1.34% :     0.000002s :     18: predicate.cast_eliminate
 0.17% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.26% :     0.000002s :     18: predicate.compare_switch_simplify
 1.60% :     0.000003s :     18: predicate.depend_value_elim
 1.25% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.46% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.62% :     0.000003s :     18: predicate.dict_set_item_eliminator
 0.64% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.23% :     0.000000s :      1: predicate.elim_not_effective
 0.44% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.76% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.21% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.26% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.36% :     0.000002s :     18: predicate.environ_get_eliminate
 1.19% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.07% :     0.000000s :      1: predicate.fold_const_symbol
 1.20% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.00% :     0.000007s :     28: predicate.inline
 3.33% :     0.000006s :     33: predicate.inline_without_move
 0.46% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :     12: predicate.less_batch_normalization
 1.48% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.64% :     0.000003s :     21: predicate.load_eliminater
 0.77% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.62% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.80% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.27% :     0.000002s :     18: predicate.merge_addn
 1.28% :     0.000002s :     18: predicate.minmaximum_grad
 0.80% :     0.000001s :      1: predicate.mutable_eliminate
 0.23% :     0.000000s :      1: predicate.opt_reshape
 2.04% :     0.000004s :     21: predicate.partial_eliminate
 1.33% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.78% :     0.000003s :     18: predicate.reduce_eliminate
 1.56% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.68% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.63% :     0.000003s :     22: predicate.replace_applicator
 1.75% :     0.000003s :     33: predicate.replace_old_param
 0.11% :     0.000000s :      1: predicate.reset_defer_inline
 1.69% :     0.000003s :     18: predicate.reshape_eliminate
 1.42% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.43% :     0.000003s :     18: predicate.same_eliminate
 0.62% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.37% :     0.000001s :      2: predicate.special_op_eliminate
 1.38% :     0.000002s :     12: predicate.specialize_transform
 1.58% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.55% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.22% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.84% :     0.000005s :     26: predicate.switch_defer_inline
 2.37% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.94% :     0.000012s :     71: predicate.switch_simplify
 1.47% :     0.000003s :     18: predicate.tile_eliminate
 1.31% :     0.000002s :     18: predicate.transpose_eliminate
 1.54% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.47% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.81% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.98% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.73% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000003s :     18: predicate.value_based_eliminate
 0.10% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.22% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001221    22
74.64% :     0.000911s :     13: func_graph_cloner_run.FuncGraphClonerGraph
25.36% :     0.000310s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025115    76
 0.17% :     0.000042s :      1: add_recomputation
 0.39% :     0.000098s :      1: auto_monad
 0.07% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.93% :     0.000235s :      1: bootstrap
 0.10% :     0.000026s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.10% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.20% :     0.000051s :      1: event_method
 0.02% :     0.000004s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
35.04% :     0.008800s :      1: jit_opt_a
 0.53% :     0.000132s :      1: jit_opt_after_cconv
 0.18% :     0.000046s :      1: jit_opt_b
 1.84% :     0.000462s :      1: loop_unroll
 2.58% :     0.000648s :      1: mutable_eliminate
 5.35% :     0.001345s :     26: opt.transform.jit_opt_a
 0.13% :     0.000032s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000018s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000018s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000025s :      4: opt.transform.symbol_engine_opt
 2.05% :     0.000515s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000028s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000019s :      1: remove_dup_value
 4.70% :     0.001180s :      1: renormalize.infer
 1.59% :     0.000399s :      1: renormalize.specialize
 0.03% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000037s :      1: rewriter_after_opt_a
 0.38% :     0.000096s :      1: rewriter_before_opt_a
 0.29% :     0.000073s :      1: symbol_engine_optimizer
 0.12% :     0.000029s :      1: task_emit
42.07% :     0.010565s :      1: type_inference
 0.20% :     0.000050s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.301.874 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c37b0d0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.302.520 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.00998555, [30]
[bootstrap]: 0.00022117
[type_inference]: 0.00349509
[event_method]: 1.489e-05
[auto_monad]: 4.317e-05
[graph_reusing]: 3.76999e-06
[pre_auto_parallel]: 1.60999e-06
[py_interpret_to_execute]: 2.421e-05
[rewriter_before_opt_a]: 6.092e-05
[expand_dump_flag]: 1.94999e-06
[jit_opt_a]: 0.00428556, [2]
    [Cycle 1]: 0.00125838, [27]
        [switch_simplify]: 4.219e-05
        [loop_unroll]: 2.809e-05
        [a_1]: 0.00049489
        [with_stream_mark]: 1.179e-05
        [recompute_prepare]: 6.71e-06
        [updatestate_depend_eliminate]: 3.11001e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 5.67999e-06
        [updatestate_useless_node_eliminater]: 5.09998e-06
        [accelerated_algorithm]: 5.57999e-06
        [meta_shard_fg_expand]: 1.27999e-06
        [get_grad_eliminate_]: 5.12999e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.177e-05
        [j_node_and_user_rematch]: 8.17e-06
        [meta_fg_expand]: 1.82999e-06
        [replace_old_param]: 7.25e-06
        [inline_without_move]: 5.42001e-06
        [renormalize]: 0.0004075
        [add_forward_monad_depend]: 4.32998e-06
        [auto_monad_grad]: 1.28002e-06
        [auto_monad_eliminator]: 9.07001e-06
        [cse]: 1.637e-05
        [replace_applicator]: 1.06e-05
    [Cycle 2]: 0.00031187, [27]
        [switch_simplify]: 5.36998e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 8.891e-05
        [with_stream_mark]: 7.55998e-06
        [recompute_prepare]: 4.84003e-06
        [updatestate_depend_eliminate]: 2.53e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 9.40025e-07
        [specialize_transform]: 5.09003e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 4.99998e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.48999e-06
        [merge_forward]: 2.32001e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.169e-05
        [j_node_and_user_rematch]: 7.58001e-06
        [meta_fg_expand]: 1.39998e-06
        [replace_old_param]: 7.21001e-06
        [inline_without_move]: 4.70001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.25999e-06
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 5.25999e-06
        [cse]: 1.044e-05
        [replace_applicator]: 5.04998e-06
[py_interpret_to_execute_after_opt_a]: 7.51999e-06
[rewriter_after_opt_a]: 2.444e-05
[convert_after_rewriter]: 5.22e-06
[order_py_execute_after_rewriter]: 4.17e-06
[mutable_eliminate]: 0.00045437
[jit_opt_b]: 4.776e-05, [1]
    [Cycle 1]: 4.153e-05, [2]
        [frontend_op_eliminate]: 1.561e-05
        [inline_after_opt_a]: 1.474e-05
[cconv]: 1.657e-05
[loop_unroll]: 0.00040001
[jit_opt_after_cconv]: 0.0001284, [1]
    [Cycle 1]: 0.00012258, [11]
        [c_1]: 1.993e-05
        [parameter_eliminate]: 2.36998e-06
        [updatestate_depend_eliminate]: 5.05999e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [cse]: 1.51e-05
        [call_graph_tuple_transform]: 1.568e-05
        [tuple_list_get_item_eliminator]: 5.27999e-06
        [none_parameter_eliminate]: 9.99979e-07
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.02999e-06
[remove_dup_value]: 9.30001e-06
[partial_unused_args_eliminate]: 1.61002e-06
[environ_conv]: 3.73999e-06
[add_recomputation]: 3.223e-05
[cse_after_recomputation]: 1.954e-05, [1]
    [Cycle 1]: 1.446e-05, [1]
        [cse]: 8.80999e-06
[auto_monad_reorder]: 1.052e-05
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 4.42e-06
[opt_after_jit_grad]: 0.00042865
[symbol_engine_optimizer]: 6.868e-05, [1]
    [Cycle 1]: 6.242e-05, [6]
        [build]: 2.66e-06
        [elim_shapecalc]: 7.28e-06
        [elim_not_effective]: 1.158e-05
        [opt_reshape]: 5.44e-06
        [fold_const_symbol]: 8.02e-06
        [renormalize]: 4.50003e-07
[validate]: 2.456e-05
Sums
    bootstrap                            :     0.000221s :  3.28%
    type_inference                       :     0.003495s : 51.82%
    event_method                         :     0.000015s :  0.22%
    auto_monad                           :     0.000043s :  0.64%
    graph_reusing                        :     0.000004s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000024s :  0.36%
    rewriter_before_opt_a                :     0.000061s :  0.90%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000048s :  0.71%
    jit_opt_a.loop_unroll                :     0.000033s :  0.49%
    jit_opt_a.a_1                        :     0.000584s :  8.66%
    jit_opt_a.with_stream_mark           :     0.000019s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.15%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.16%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.05%
    jit_opt_a.replace_old_param          :     0.000014s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.15%
    jit_opt_a.renormalize                :     0.000408s :  6.04%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.21%
    jit_opt_a.cse                        :     0.000027s :  0.40%
    jit_opt_a.replace_applicator         :     0.000016s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000024s :  0.36%
    convert_after_rewriter               :     0.000005s :  0.08%
    order_py_execute_after_rewriter      :     0.000004s :  0.06%
    mutable_eliminate                    :     0.000454s :  6.74%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.22%
    cconv                                :     0.000017s :  0.25%
    loop_unroll                          :     0.000400s :  5.93%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.30%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000015s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000009s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000032s :  0.48%
    cse_after_recomputation.cse          :     0.000009s :  0.13%
    auto_monad_reorder                   :     0.000011s :  0.16%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.07%
    opt_after_jit_grad                   :     0.000429s :  6.36%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000025s :  0.36%


Time group info:
------[substitution.]   0.000146    23
 1.12% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.36% :     0.000003s :      3: substitution.graph_param_transform
83.91% :     0.000122s :      4: substitution.inline
 1.69% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.82% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.90% :     0.000003s :      2: substitution.replace_old_param
 5.49% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003451     2
79.87% :     0.002757s :      1: type_inference.infer
20.13% :     0.000695s :      1: type_inference.specialize
------[replace.]   0.000058     6
66.14% :     0.000038s :      4: replace.inline
33.86% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000127     6
94.53% :     0.000120s :      4: match.inline
 5.47% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000107   691
 1.25% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.35% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     10: predicate.addn_check_dump
 1.37% :     0.000001s :     10: predicate.addn_zero_filter
 1.90% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000001s :     10: predicate.depend_value_elim
 1.24% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.27% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.72% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.65% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.41% :     0.000000s :      3: predicate.graph_param_transform
 5.14% :     0.000006s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.86% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.35% :     0.000003s :     15: predicate.partial_eliminate
 1.29% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.53% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.75% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 1.17% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.58% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.98% :     0.000009s :     53: predicate.switch_simplify
 1.28% :     0.000001s :     10: predicate.tile_eliminate
 1.20% :     0.000001s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.32% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.54% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.79% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000460    12
46.92% :     0.000216s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.23% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.85% :     0.000220s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011235    72
 0.31% :     0.000035s :      1: add_recomputation
 0.41% :     0.000046s :      1: auto_monad
 0.12% :     0.000013s :      1: auto_monad_reorder
 2.07% :     0.000232s :      1: bootstrap
 0.17% :     0.000019s :      1: cconv
 0.07% :     0.000007s :      1: convert_after_rewriter
 0.19% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000006s :      1: environ_conv
 0.16% :     0.000018s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000006s :      1: graph_reusing
38.17% :     0.004288s :      1: jit_opt_a
 1.17% :     0.000131s :      1: jit_opt_after_cconv
 0.45% :     0.000051s :      1: jit_opt_b
 3.63% :     0.000407s :      1: loop_unroll
 4.11% :     0.000462s :      1: mutable_eliminate
 6.88% :     0.000772s :     26: opt.transform.jit_opt_a
 0.38% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000024s :      4: opt.transform.jit_opt_b
 0.10% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000029s :      4: opt.transform.symbol_engine_opt
 3.88% :     0.000435s :      1: opt_after_jit_grad
 0.05% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000003s :      1: pre_auto_parallel
 0.24% :     0.000027s :      1: py_interpret_to_execute
 0.09% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000012s :      1: remove_dup_value
 1.70% :     0.000191s :      1: renormalize.infer
 1.88% :     0.000211s :      1: renormalize.specialize
 0.06% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000028s :      1: rewriter_after_opt_a
 0.57% :     0.000064s :      1: rewriter_before_opt_a
 0.63% :     0.000071s :      1: symbol_engine_optimizer
31.20% :     0.003505s :      1: type_inference


TotalTime = 0.0399308, [33]
[bootstrap]: 0.00019713
[type_inference]: 0.0159355
[event_method]: 4.649e-05
[auto_monad]: 9.331e-05
[graph_reusing]: 7.4e-06
[pre_auto_parallel]: 2.69001e-06
[py_interpret_to_execute]: 2.453e-05
[rewriter_before_opt_a]: 8.885e-05
[expand_dump_flag]: 3.09999e-06
[jit_opt_a]: 0.0119865, [3]
    [Cycle 1]: 0.00591747, [27]
        [switch_simplify]: 6.113e-05
        [loop_unroll]: 3.743e-05
        [a_1]: 0.00083006
        [with_stream_mark]: 2.897e-05
        [recompute_prepare]: 1.773e-05
        [updatestate_depend_eliminate]: 8.13001e-06
        [updatestate_assign_eliminate]: 6.23e-06
        [updatestate_loads_eliminate]: 5.51998e-06
        [parameter_eliminate]: 2.11e-06
        [specialize_transform]: 1.248e-05
        [updatestate_useless_node_eliminater]: 1.152e-05
        [accelerated_algorithm]: 1.226e-05
        [meta_shard_fg_expand]: 3.18e-06
        [get_grad_eliminate_]: 1.238e-05
        [merge_forward]: 8.18999e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.513e-05
        [j_node_and_user_rematch]: 2.349e-05
        [meta_fg_expand]: 0.00150629
        [replace_old_param]: 5.09e-05
        [inline_without_move]: 5.265e-05
        [renormalize]: 0.00269731
        [add_forward_monad_depend]: 9.83998e-06
        [auto_monad_grad]: 5.70001e-06
        [auto_monad_eliminator]: 5.245e-05
        [cse]: 0.00015995
        [replace_applicator]: 7.257e-05
    [Cycle 2]: 0.00224592, [27]
        [switch_simplify]: 3.728e-05
        [loop_unroll]: 3.627e-05
        [a_1]: 0.00113914
        [with_stream_mark]: 1.459e-05
        [recompute_prepare]: 9.63002e-06
        [updatestate_depend_eliminate]: 4.78001e-06
        [updatestate_assign_eliminate]: 3.8e-06
        [updatestate_loads_eliminate]: 3.76001e-06
        [parameter_eliminate]: 2.00002e-06
        [specialize_transform]: 7.2e-06
        [updatestate_useless_node_eliminater]: 7.01001e-06
        [accelerated_algorithm]: 7.62002e-06
        [meta_shard_fg_expand]: 2.37999e-06
        [get_grad_eliminate_]: 6.71999e-06
        [merge_forward]: 4.38999e-06
        [cell_reuse_recompute_pass]: 1.42e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.656e-05
        [j_node_and_user_rematch]: 1.178e-05
        [meta_fg_expand]: 4.617e-05
        [replace_old_param]: 1.096e-05
        [inline_without_move]: 6.78e-06
        [renormalize]: 0.00065614
        [add_forward_monad_depend]: 5.59e-06
        [auto_monad_grad]: 2.39999e-06
        [auto_monad_eliminator]: 1.422e-05
        [cse]: 2.62e-05
        [replace_applicator]: 1.318e-05
    [Cycle 3]: 0.00040883, [27]
        [switch_simplify]: 7.31001e-06
        [loop_unroll]: 7e-06
        [a_1]: 0.00014382
        [with_stream_mark]: 1.021e-05
        [recompute_prepare]: 6.76e-06
        [updatestate_depend_eliminate]: 3.83001e-06
        [updatestate_assign_eliminate]: 3.65e-06
        [updatestate_loads_eliminate]: 2.95998e-06
        [parameter_eliminate]: 9.90025e-07
        [specialize_transform]: 6.57002e-06
        [updatestate_useless_node_eliminater]: 6.50002e-06
        [accelerated_algorithm]: 6.83e-06
        [meta_shard_fg_expand]: 1.49998e-06
        [get_grad_eliminate_]: 6.36e-06
        [merge_forward]: 3.36001e-06
        [cell_reuse_recompute_pass]: 1.94e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.641e-05
        [j_node_and_user_rematch]: 1.125e-05
        [meta_fg_expand]: 2.36998e-06
        [replace_old_param]: 8.85001e-06
        [inline_without_move]: 6.53e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 6.70028e-07
        [auto_monad_eliminator]: 7.78001e-06
        [cse]: 1.629e-05
        [replace_applicator]: 6.84001e-06
[py_interpret_to_execute_after_opt_a]: 1.277e-05
[rewriter_after_opt_a]: 4.658e-05
[convert_after_rewriter]: 8.72998e-06
[order_py_execute_after_rewriter]: 5.92999e-06
[mutable_eliminate]: 0.00070489
[jit_opt_b]: 6.271e-05, [1]
    [Cycle 1]: 5.398e-05, [2]
        [frontend_op_eliminate]: 2.183e-05
        [inline_after_opt_a]: 2.006e-05
[cconv]: 2.315e-05
[loop_unroll]: 0.00043531
[jit_opt_after_cconv]: 0.00017348, [1]
    [Cycle 1]: 0.00016556, [11]
        [c_1]: 3.051e-05
        [parameter_eliminate]: 2.59999e-06
        [updatestate_depend_eliminate]: 7.3e-06
        [updatestate_assign_eliminate]: 3.09999e-06
        [updatestate_loads_eliminate]: 2.97002e-06
        [cse]: 2.923e-05
        [call_graph_tuple_transform]: 2.402e-05
        [tuple_list_get_item_eliminator]: 7.03998e-06
        [none_parameter_eliminate]: 1.74e-06
        [renormalize]: 7.79983e-07
        [switch_simplify]: 7.1e-06
[remove_dup_value]: 1.73e-05
[partial_unused_args_eliminate]: 2.34001e-06
[environ_conv]: 6.27001e-06
[add_recomputation]: 6.628e-05
[cse_after_recomputation]: 3.158e-05, [1]
    [Cycle 1]: 2.485e-05, [1]
        [cse]: 1.797e-05
[auto_monad_reorder]: 2.07e-05
[get_jit_bprop_graph]: 1.74e-06
[rewriter_after_jit_bprop_graph]: 6.39999e-06
[opt_after_jit_grad]: 0.00047298
[symbol_engine_optimizer]: 9.119e-05, [1]
    [Cycle 1]: 8.391e-05, [6]
        [build]: 4.92e-06
        [elim_shapecalc]: 1.036e-05
        [elim_not_effective]: 1.709e-05
        [opt_reshape]: 7.83001e-06
        [fold_const_symbol]: 1.498e-05
        [renormalize]: 5.8001e-07
[validate]: 4.739e-05
[backend_pass]: 1.22e-06
[task_emit]: 0.00908658
[execute]: 9.00001e-06
Sums
    bootstrap                            :     0.000197s :  0.55%
    type_inference                       :     0.015936s : 44.62%
    event_method                         :     0.000046s :  0.13%
    auto_monad                           :     0.000093s :  0.26%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000025s :  0.07%
    rewriter_before_opt_a                :     0.000089s :  0.25%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000106s :  0.30%
    jit_opt_a.loop_unroll                :     0.000081s :  0.23%
    jit_opt_a.a_1                        :     0.002113s :  5.92%
    jit_opt_a.with_stream_mark           :     0.000054s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000034s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000014s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.01%
    jit_opt_a.specialize_transform       :     0.000026s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000027s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.07%
    jit_opt_a.merge_forward              :     0.000016s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000058s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000047s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.001555s :  4.35%
    jit_opt_a.replace_old_param          :     0.000071s :  0.20%
    jit_opt_a.inline_without_move        :     0.000066s :  0.18%
    jit_opt_a.renormalize                :     0.003354s :  9.39%
    jit_opt_a.add_forward_monad_depend   :     0.000017s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000074s :  0.21%
    jit_opt_a.cse                        :     0.000202s :  0.57%
    jit_opt_a.replace_applicator         :     0.000093s :  0.26%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.04%
    rewriter_after_opt_a                 :     0.000047s :  0.13%
    convert_after_rewriter               :     0.000009s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000705s :  1.97%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.06%
    cconv                                :     0.000023s :  0.06%
    loop_unroll                          :     0.000435s :  1.22%
    jit_opt_after_cconv.c_1              :     0.000031s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000017s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000066s :  0.19%
    cse_after_recomputation.cse          :     0.000018s :  0.05%
    auto_monad_reorder                   :     0.000021s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000473s :  1.32%
    symbol_engine_optimizer.build        :     0.000005s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000047s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.009087s : 25.44%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000534   118
 4.90% :     0.000026s :      1: substitution.arithmetic_simplify
 0.51% :     0.000003s :      4: substitution.elim_not_effective
 0.58% :     0.000003s :      4: substitution.fold_const_symbol
 1.17% :     0.000006s :      5: substitution.graph_param_transform
61.09% :     0.000326s :     12: substitution.inline
 3.44% :     0.000018s :      2: substitution.inline_without_move
 1.97% :     0.000011s :     16: substitution.j_node_and_user_rematch
 2.03% :     0.000011s :      7: substitution.minmaximum_grad
 1.26% :     0.000007s :      5: substitution.partial_eliminate
 2.18% :     0.000012s :     16: substitution.remove_not_recompute_node
 5.18% :     0.000028s :     10: substitution.replace_applicator
 1.36% :     0.000007s :      7: substitution.replace_old_param
 0.46% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 3.81% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.77% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 7.29% :     0.000039s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.015866     2
93.09% :     0.014769s :      1: type_inference.infer
 6.91% :     0.001097s :      1: type_inference.specialize
------[replace.]   0.000151    20
 3.63% :     0.000005s :      1: replace.arithmetic_simplify
62.11% :     0.000094s :     12: replace.inline
34.25% :     0.000052s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000365    20
 6.93% :     0.000025s :      1: match.arithmetic_simplify
87.66% :     0.000320s :     12: match.inline
 5.41% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000391  2790
 1.63% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.44% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.38% :     0.000005s :     46: predicate.addn_check_dump
 1.75% :     0.000007s :     46: predicate.addn_zero_filter
 2.15% :     0.000008s :     47: predicate.arithmetic_simplify
 1.64% :     0.000006s :     47: predicate.cast_eliminate
 0.25% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.38% :     0.000005s :     46: predicate.compare_switch_simplify
 1.54% :     0.000006s :     46: predicate.depend_value_elim
 1.43% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.51% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.69% :     0.000007s :     47: predicate.dict_set_item_eliminator
 0.33% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      5: predicate.elim_not_effective
 0.29% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.42% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.61% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.49% :     0.000006s :     47: predicate.environ_get_eliminate
 1.45% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      5: predicate.fold_const_symbol
 0.87% :     0.000003s :     21: predicate.get_grad_eliminate
 0.16% :     0.000001s :      5: predicate.graph_param_transform
 4.16% :     0.000016s :     76: predicate.inline
 1.74% :     0.000007s :     42: predicate.inline_without_move
 0.35% :     0.000001s :     21: predicate.j_node_and_user_rematch
 1.03% :     0.000004s :     21: predicate.less_batch_normalization
 1.91% :     0.000007s :     54: predicate.list_to_tuple_eliminator_
 1.94% :     0.000008s :     59: predicate.load_eliminater
 0.55% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.18% :     0.000012s :     89: predicate.loop_unroll_before_grad
 1.87% :     0.000007s :     52: predicate.make_slice_get_slice_eliminator
 1.40% :     0.000005s :     46: predicate.merge_addn
 1.47% :     0.000006s :     47: predicate.minmaximum_grad
 0.58% :     0.000002s :      5: predicate.mutable_eliminate
 0.24% :     0.000001s :      5: predicate.opt_reshape
 2.42% :     0.000009s :     59: predicate.partial_eliminate
 1.56% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.89% :     0.000007s :     47: predicate.reduce_eliminate
 1.82% :     0.000007s :     54: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.48% :     0.000010s :    105: predicate.replace_applicator
 0.84% :     0.000003s :     42: predicate.replace_old_param
 0.13% :     0.000001s :      5: predicate.reset_defer_inline
 1.79% :     0.000007s :     47: predicate.reshape_eliminate
 1.49% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.39% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.43% :     0.000006s :     46: predicate.same_eliminate
 0.45% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.39% :     0.000002s :     10: predicate.special_op_eliminate
 0.84% :     0.000003s :     21: predicate.specialize_transform
 1.64% :     0.000006s :     46: predicate.split_environ_get_set_with_tuple_value
 1.52% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.21% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.62% :     0.000010s :     66: predicate.switch_defer_inline
 2.40% :     0.000009s :     66: predicate.switch_layer_defer_inline
 6.05% :     0.000024s :    160: predicate.switch_simplify
 1.69% :     0.000007s :     47: predicate.tile_eliminate
 1.52% :     0.000006s :     47: predicate.transpose_eliminate
 1.81% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.61% :     0.000006s :     47: predicate.tuple_list_get_item_depend_reorder
 2.97% :     0.000012s :     64: predicate.tuple_list_get_item_eliminator
 1.94% :     0.000008s :     47: predicate.tuple_list_set_item_eliminator
 1.85% :     0.000007s :     54: predicate.tuple_to_list_eliminator_
 1.98% :     0.000008s :     59: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000011s :     80: predicate.updatestate_useless_node_eliminater
 1.91% :     0.000007s :     46: predicate.value_based_eliminate
 0.20% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.25% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001440    29
59.18% :     0.000852s :     13: func_graph_cloner_run.FuncGraphClonerGraph
40.82% :     0.000588s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.046161    91
 0.15% :     0.000070s :      1: add_recomputation
 0.21% :     0.000097s :      1: auto_monad
 0.05% :     0.000024s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.46% :     0.000211s :      1: bootstrap
 0.06% :     0.000026s :      1: cconv
 0.02% :     0.000011s :      1: convert_after_rewriter
 0.07% :     0.000034s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.11% :     0.000051s :      1: event_method
 0.03% :     0.000014s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
25.97% :     0.011990s :      1: jit_opt_a
 0.38% :     0.000176s :      1: jit_opt_after_cconv
 0.14% :     0.000065s :      1: jit_opt_b
 0.96% :     0.000444s :      1: loop_unroll
 1.55% :     0.000714s :      1: mutable_eliminate
 5.91% :     0.002730s :     39: opt.transform.jit_opt_a
 0.14% :     0.000065s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000036s :      4: opt.transform.jit_opt_b
 0.03% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000047s :      4: opt.transform.symbol_engine_opt
 1.04% :     0.000481s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.06% :     0.000027s :      1: py_interpret_to_execute
 0.03% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000020s :      1: remove_dup_value
 4.13% :     0.001906s :      2: renormalize.infer
 3.10% :     0.001432s :      2: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000049s :      1: rewriter_after_opt_a
 0.20% :     0.000092s :      1: rewriter_before_opt_a
 0.20% :     0.000094s :      1: symbol_engine_optimizer
19.73% :     0.009106s :      1: task_emit
34.55% :     0.015948s :      1: type_inference
 0.15% :     0.000071s :      1: validate


TotalTime = 0.0100513, [30]
[bootstrap]: 0.00022683
[type_inference]: 0.00350646
[event_method]: 1.519e-05
[auto_monad]: 4.78e-05
[graph_reusing]: 4.94e-06
[pre_auto_parallel]: 1.64e-06
[py_interpret_to_execute]: 2.486e-05
[rewriter_before_opt_a]: 6.24e-05
[expand_dump_flag]: 2.50997e-06
[jit_opt_a]: 0.0042801, [2]
    [Cycle 1]: 0.00121132, [27]
        [switch_simplify]: 4.204e-05
        [loop_unroll]: 2.828e-05
        [a_1]: 0.00045204
        [with_stream_mark]: 1.263e-05
        [recompute_prepare]: 6.83998e-06
        [updatestate_depend_eliminate]: 3.01001e-06
        [updatestate_assign_eliminate]: 2.25002e-06
        [updatestate_loads_eliminate]: 1.94999e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 5.86e-06
        [updatestate_useless_node_eliminater]: 4.63001e-06
        [accelerated_algorithm]: 5.17e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 4.97e-06
        [merge_forward]: 2.56998e-06
        [cell_reuse_recompute_pass]: 9.20001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.173e-05
        [j_node_and_user_rematch]: 8.65999e-06
        [meta_fg_expand]: 2.12001e-06
        [replace_old_param]: 7.45e-06
        [inline_without_move]: 5.22e-06
        [renormalize]: 0.00040448
        [add_forward_monad_depend]: 4.33001e-06
        [auto_monad_grad]: 1.32e-06
        [auto_monad_eliminator]: 1.073e-05
        [cse]: 1.544e-05
        [replace_applicator]: 1.015e-05
    [Cycle 2]: 0.00030636, [27]
        [switch_simplify]: 5.20999e-06
        [loop_unroll]: 4.51002e-06
        [a_1]: 8.774e-05
        [with_stream_mark]: 7.53e-06
        [recompute_prepare]: 4.85001e-06
        [updatestate_depend_eliminate]: 2.24001e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 5.15999e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.32999e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.133e-05
        [j_node_and_user_rematch]: 7.50998e-06
        [meta_fg_expand]: 1.47999e-06
        [replace_old_param]: 6.62002e-06
        [inline_without_move]: 6.78e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 4.60999e-06
        [cse]: 9.77999e-06
        [replace_applicator]: 4.70001e-06
[py_interpret_to_execute_after_opt_a]: 7.12002e-06
[rewriter_after_opt_a]: 2.631e-05
[convert_after_rewriter]: 5.76998e-06
[order_py_execute_after_rewriter]: 4.82998e-06
[mutable_eliminate]: 0.00046537
[jit_opt_b]: 4.597e-05, [1]
    [Cycle 1]: 3.99e-05, [2]
        [frontend_op_eliminate]: 1.463e-05
        [inline_after_opt_a]: 1.426e-05
[cconv]: 1.939e-05
[loop_unroll]: 0.00041055
[jit_opt_after_cconv]: 0.00013205, [1]
    [Cycle 1]: 0.00012587, [11]
        [c_1]: 2.007e-05
        [parameter_eliminate]: 2.62001e-06
        [updatestate_depend_eliminate]: 5.33002e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.673e-05
        [call_graph_tuple_transform]: 1.625e-05
        [tuple_list_get_item_eliminator]: 5.43002e-06
        [none_parameter_eliminate]: 1.12e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.15999e-06
[remove_dup_value]: 1.012e-05
[partial_unused_args_eliminate]: 1.60999e-06
[environ_conv]: 3.85e-06
[add_recomputation]: 3.459e-05
[cse_after_recomputation]: 1.933e-05, [1]
    [Cycle 1]: 1.422e-05, [1]
        [cse]: 8.81002e-06
[auto_monad_reorder]: 1.035e-05
[get_jit_bprop_graph]: 1.60999e-06
[rewriter_after_jit_bprop_graph]: 4.37e-06
[opt_after_jit_grad]: 0.00043466
[symbol_engine_optimizer]: 6.969e-05, [1]
    [Cycle 1]: 6.371e-05, [6]
        [build]: 2.86e-06
        [elim_shapecalc]: 7.50998e-06
        [elim_not_effective]: 1.184e-05
        [opt_reshape]: 5.59e-06
        [fold_const_symbol]: 8.74e-06
        [renormalize]: 4.00003e-07
[validate]: 2.574e-05
Sums
    bootstrap                            :     0.000227s :  3.35%
    type_inference                       :     0.003506s : 51.84%
    event_method                         :     0.000015s :  0.22%
    auto_monad                           :     0.000048s :  0.71%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000025s :  0.37%
    rewriter_before_opt_a                :     0.000062s :  0.92%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000047s :  0.70%
    jit_opt_a.loop_unroll                :     0.000033s :  0.48%
    jit_opt_a.a_1                        :     0.000540s :  7.98%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.24%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000014s :  0.21%
    jit_opt_a.inline_without_move        :     0.000012s :  0.18%
    jit_opt_a.renormalize                :     0.000405s :  5.98%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.23%
    jit_opt_a.cse                        :     0.000025s :  0.37%
    jit_opt_a.replace_applicator         :     0.000015s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.11%
    rewriter_after_opt_a                 :     0.000026s :  0.39%
    convert_after_rewriter               :     0.000006s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000465s :  6.88%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.21%
    cconv                                :     0.000019s :  0.29%
    loop_unroll                          :     0.000411s :  6.07%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.30%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000017s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.08%
    remove_dup_value                     :     0.000010s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000035s :  0.51%
    cse_after_recomputation.cse          :     0.000009s :  0.13%
    auto_monad_reorder                   :     0.000010s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.06%
    opt_after_jit_grad                   :     0.000435s :  6.43%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.18%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.13%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000026s :  0.38%


Time group info:
------[substitution.]   0.000145    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.78% :     0.000001s :      2: substitution.fold_const_symbol
 2.52% :     0.000004s :      3: substitution.graph_param_transform
83.30% :     0.000120s :      4: substitution.inline
 1.74% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.67% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.83% :     0.000003s :      2: substitution.replace_old_param
 6.06% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003461     2
80.37% :     0.002782s :      1: type_inference.infer
19.63% :     0.000679s :      1: type_inference.specialize
------[replace.]   0.000057     6
67.22% :     0.000039s :      4: replace.inline
32.78% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000126     6
93.88% :     0.000118s :      4: match.inline
 6.12% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000108   691
 1.44% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 1.79% :     0.000002s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000001s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.32% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.31% :     0.000001s :     10: predicate.environ_get_eliminate
 1.22% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000000s :      3: predicate.graph_param_transform
 5.36% :     0.000006s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.38% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.52% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.83% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000002s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.57% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000001s :     10: predicate.reshape_eliminate
 1.25% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.94% :     0.000009s :     53: predicate.switch_simplify
 1.31% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.53% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000450    12
49.08% :     0.000221s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.31% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.61% :     0.000205s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011252    72
 0.33% :     0.000037s :      1: add_recomputation
 0.45% :     0.000051s :      1: auto_monad
 0.11% :     0.000013s :      1: auto_monad_reorder
 2.12% :     0.000239s :      1: bootstrap
 0.20% :     0.000022s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.19% :     0.000021s :      1: cse_after_recomputation
 0.06% :     0.000006s :      1: environ_conv
 0.17% :     0.000019s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
38.06% :     0.004282s :      1: jit_opt_a
 1.20% :     0.000135s :      1: jit_opt_after_cconv
 0.43% :     0.000049s :      1: jit_opt_b
 3.72% :     0.000419s :      1: loop_unroll
 4.22% :     0.000474s :      1: mutable_eliminate
 6.47% :     0.000728s :     26: opt.transform.jit_opt_a
 0.39% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000023s :      4: opt.transform.jit_opt_b
 0.10% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.27% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.93% :     0.000442s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000003s :      1: pre_auto_parallel
 0.24% :     0.000027s :      1: py_interpret_to_execute
 0.08% :     0.000009s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000013s :      1: remove_dup_value
 1.70% :     0.000191s :      1: renormalize.infer
 1.84% :     0.000207s :      1: renormalize.specialize
 0.06% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000029s :      1: rewriter_after_opt_a
 0.58% :     0.000065s :      1: rewriter_before_opt_a
 0.64% :     0.000072s :      1: symbol_engine_optimizer
31.25% :     0.003517s :      1: type_inference

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.326.975 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80ef1f0> should be wrapped by jit only once.

TotalTime = 0.0240849, [33]
[bootstrap]: 0.00020791
[type_inference]: 0.0111754
[event_method]: 5.189e-05
[auto_monad]: 9.997e-05
[graph_reusing]: 7.45998e-06
[pre_auto_parallel]: 3.11001e-06
[py_interpret_to_execute]: 2.993e-05
[rewriter_before_opt_a]: 9.827e-05
[expand_dump_flag]: 3.31001e-06
[jit_opt_a]: 0.0102268, [2]
    [Cycle 1]: 0.00549904, [27]
        [switch_simplify]: 8.242e-05
        [loop_unroll]: 4.031e-05
        [a_1]: 0.00094228
        [with_stream_mark]: 3.903e-05
        [recompute_prepare]: 2.55e-05
        [updatestate_depend_eliminate]: 9.02e-06
        [updatestate_assign_eliminate]: 7.33e-06
        [updatestate_loads_eliminate]: 6.38e-06
        [parameter_eliminate]: 3.70998e-06
        [specialize_transform]: 1.49e-05
        [updatestate_useless_node_eliminater]: 1.243e-05
        [accelerated_algorithm]: 1.404e-05
        [meta_shard_fg_expand]: 4.07e-06
        [get_grad_eliminate_]: 1.278e-05
        [merge_forward]: 8.85001e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 3.112e-05
        [j_node_and_user_rematch]: 2.666e-05
        [meta_fg_expand]: 0.002025
        [replace_old_param]: 6.479e-05
        [inline_without_move]: 5.619e-05
        [renormalize]: 0.00175489
        [add_forward_monad_depend]: 6.44001e-06
        [auto_monad_grad]: 2.46e-06
        [auto_monad_eliminator]: 1.305e-05
        [cse]: 2.604e-05
        [replace_applicator]: 1.421e-05
    [Cycle 2]: 0.00031385, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 4e-06
        [a_1]: 4.813e-05
        [with_stream_mark]: 1.073e-05
        [recompute_prepare]: 3.83001e-06
        [updatestate_depend_eliminate]: 3.22002e-06
        [updatestate_assign_eliminate]: 2.47001e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 1.74998e-06
        [specialize_transform]: 3.76999e-06
        [updatestate_useless_node_eliminater]: 3.78999e-06
        [accelerated_algorithm]: 6.33e-06
        [meta_shard_fg_expand]: 2.19999e-06
        [get_grad_eliminate_]: 3.48e-06
        [merge_forward]: 3.22002e-06
        [cell_reuse_recompute_pass]: 3.26001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.372e-05
        [j_node_and_user_rematch]: 6.33e-06
        [meta_fg_expand]: 3.78e-05
        [replace_old_param]: 4.2e-06
        [inline_without_move]: 3.34001e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.47001e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.19998e-06
        [cse]: 1.085e-05
        [replace_applicator]: 4.27e-06
[py_interpret_to_execute_after_opt_a]: 1.013e-05
[rewriter_after_opt_a]: 3.539e-05
[convert_after_rewriter]: 6.51e-06
[order_py_execute_after_rewriter]: 4.38999e-06
[mutable_eliminate]: 0.00066959
[jit_opt_b]: 4.439e-05, [1]
    [Cycle 1]: 3.664e-05, [2]
        [frontend_op_eliminate]: 1.216e-05
        [inline_after_opt_a]: 1.181e-05
[cconv]: 2.249e-05
[loop_unroll]: 0.00042834
[jit_opt_after_cconv]: 0.00011966, [1]
    [Cycle 1]: 0.0001136, [11]
        [c_1]: 1.288e-05
        [parameter_eliminate]: 2.19001e-06
        [updatestate_depend_eliminate]: 4.65001e-06
        [updatestate_assign_eliminate]: 2.05002e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [cse]: 1.579e-05
        [call_graph_tuple_transform]: 1.359e-05
        [tuple_list_get_item_eliminator]: 3.93001e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 3.9e-06
[remove_dup_value]: 1.47e-05
[partial_unused_args_eliminate]: 2.14999e-06
[environ_conv]: 5.21002e-06
[add_recomputation]: 3.482e-05
[cse_after_recomputation]: 2.09e-05, [1]
    [Cycle 1]: 1.49e-05, [1]
        [cse]: 8.64e-06
[auto_monad_reorder]: 1.366e-05
[get_jit_bprop_graph]: 1.64e-06
[rewriter_after_jit_bprop_graph]: 4.27e-06
[opt_after_jit_grad]: 0.00044168
[symbol_engine_optimizer]: 6.502e-05, [1]
    [Cycle 1]: 5.837e-05, [6]
        [build]: 2.75002e-06
        [elim_shapecalc]: 6.37001e-06
        [elim_not_effective]: 9.70002e-06
        [opt_reshape]: 4.28001e-06
        [fold_const_symbol]: 6.06e-06
        [renormalize]: 4.69998e-07
[validate]: 2.76e-05
[backend_pass]: 1.34e-06
[task_emit]: 2.238e-05
[execute]: 1.41998e-06
Sums
    bootstrap                            :     0.000208s :  1.09%
    type_inference                       :     0.011175s : 58.84%
    event_method                         :     0.000052s :  0.27%
    auto_monad                           :     0.000100s :  0.53%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.16%
    rewriter_before_opt_a                :     0.000098s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000088s :  0.46%
    jit_opt_a.loop_unroll                :     0.000044s :  0.23%
    jit_opt_a.a_1                        :     0.000990s :  5.22%
    jit_opt_a.with_stream_mark           :     0.000050s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000029s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000010s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.03%
    jit_opt_a.specialize_transform       :     0.000019s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000020s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.09%
    jit_opt_a.merge_forward              :     0.000012s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000045s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000033s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.002063s : 10.86%
    jit_opt_a.replace_old_param          :     0.000069s :  0.36%
    jit_opt_a.inline_without_move        :     0.000060s :  0.31%
    jit_opt_a.renormalize                :     0.001755s :  9.24%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.10%
    jit_opt_a.cse                        :     0.000037s :  0.19%
    jit_opt_a.replace_applicator         :     0.000018s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.05%
    rewriter_after_opt_a                 :     0.000035s :  0.19%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000670s :  3.53%
    jit_opt_b.frontend_op_eliminate      :     0.000012s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.06%
    cconv                                :     0.000022s :  0.12%
    loop_unroll                          :     0.000428s :  2.26%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000035s :  0.18%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.02%
    opt_after_jit_grad                   :     0.000442s :  2.33%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.15%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.000022s :  0.12%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000321    50
 0.47% :     0.000002s :      1: substitution.elim_not_effective
 0.33% :     0.000001s :      1: substitution.fold_const_symbol
 1.29% :     0.000004s :      1: substitution.graph_param_transform
69.27% :     0.000222s :      6: substitution.inline
 6.39% :     0.000020s :      2: substitution.inline_without_move
 2.47% :     0.000008s :      9: substitution.j_node_and_user_rematch
 1.29% :     0.000004s :      2: substitution.minmaximum_grad
 1.36% :     0.000004s :      3: substitution.partial_eliminate
 2.64% :     0.000008s :      9: substitution.remove_not_recompute_node
 1.99% :     0.000006s :      2: substitution.replace_applicator
 1.52% :     0.000005s :      5: substitution.replace_old_param
 1.24% :     0.000004s :      1: substitution.set_cell_output_no_recompute
 2.50% :     0.000008s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.50% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.72% :     0.000018s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011089     2
88.21% :     0.009782s :      1: type_inference.infer
11.79% :     0.001307s :      1: type_inference.specialize
------[replace.]   0.000073     8
79.94% :     0.000058s :      6: replace.inline
20.06% :     0.000015s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000231     8
94.40% :     0.000218s :      6: match.inline
 5.60% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000185  1109
 2.05% :     0.000004s :     18: predicate.accumulaten_eliminater
 0.75% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.11% :     0.000002s :     18: predicate.addn_check_dump
 1.44% :     0.000003s :     18: predicate.addn_zero_filter
 3.02% :     0.000006s :     18: predicate.arithmetic_simplify
 1.39% :     0.000003s :     18: predicate.cast_eliminate
 0.12% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.12% :     0.000002s :     18: predicate.compare_switch_simplify
 1.58% :     0.000003s :     18: predicate.depend_value_elim
 1.17% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.52% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.37% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.14% :     0.000000s :      1: predicate.elim_not_effective
 0.33% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.46% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.38% :     0.000003s :     18: predicate.environ_get_add_eliminate
 1.15% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     18: predicate.environ_get_eliminate
 1.20% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.09% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.39% :     0.000008s :     28: predicate.inline
 3.15% :     0.000006s :     33: predicate.inline_without_move
 0.50% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.52% :     0.000003s :     12: predicate.less_batch_normalization
 1.64% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 2.29% :     0.000004s :     21: predicate.load_eliminater
 0.64% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.47% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.63% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000002s :     18: predicate.merge_addn
 1.21% :     0.000002s :     18: predicate.minmaximum_grad
 0.85% :     0.000002s :      1: predicate.mutable_eliminate
 0.18% :     0.000000s :      1: predicate.opt_reshape
 2.05% :     0.000004s :     21: predicate.partial_eliminate
 1.27% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.82% :     0.000003s :     18: predicate.reduce_eliminate
 1.54% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.43% :     0.000003s :     22: predicate.replace_applicator
 1.91% :     0.000004s :     33: predicate.replace_old_param
 0.11% :     0.000000s :      1: predicate.reset_defer_inline
 1.67% :     0.000003s :     18: predicate.reshape_eliminate
 1.31% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.39% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     18: predicate.same_eliminate
 0.86% :     0.000002s :     12: predicate.set_cell_output_no_recompute
 0.43% :     0.000001s :      2: predicate.special_op_eliminate
 1.19% :     0.000002s :     12: predicate.specialize_transform
 1.54% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.13% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 3.20% :     0.000006s :     26: predicate.switch_defer_inline
 2.24% :     0.000004s :     26: predicate.switch_layer_defer_inline
 7.24% :     0.000013s :     71: predicate.switch_simplify
 1.29% :     0.000002s :     18: predicate.tile_eliminate
 1.23% :     0.000002s :     18: predicate.transpose_eliminate
 1.51% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 3.10% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.93% :     0.000004s :     18: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.54% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.77% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000003s :     18: predicate.value_based_eliminate
 0.10% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.43% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001350    22
73.44% :     0.000991s :     13: func_graph_cloner_run.FuncGraphClonerGraph
26.56% :     0.000358s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027313    76
 0.14% :     0.000037s :      1: add_recomputation
 0.38% :     0.000104s :      1: auto_monad
 0.06% :     0.000016s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.81% :     0.000222s :      1: bootstrap
 0.09% :     0.000025s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.21% :     0.000058s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000007s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
37.46% :     0.010230s :      1: jit_opt_a
 0.45% :     0.000122s :      1: jit_opt_after_cconv
 0.17% :     0.000047s :      1: jit_opt_b
 1.60% :     0.000436s :      1: loop_unroll
 2.48% :     0.000678s :      1: mutable_eliminate
 5.16% :     0.001410s :     26: opt.transform.jit_opt_a
 0.11% :     0.000031s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000017s :      4: opt.transform.jit_opt_b
 0.04% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000023s :      4: opt.transform.symbol_engine_opt
 1.64% :     0.000449s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000032s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 4.77% :     0.001303s :      1: renormalize.infer
 1.62% :     0.000442s :      1: renormalize.specialize
 0.02% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000038s :      1: rewriter_after_opt_a
 0.37% :     0.000101s :      1: rewriter_before_opt_a
 0.25% :     0.000068s :      1: symbol_engine_optimizer
 0.10% :     0.000027s :      1: task_emit
40.99% :     0.011195s :      1: type_inference
 0.16% :     0.000044s :      1: validate

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.327.676 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.333.177 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c36bd30> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.333.887 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0173346, [33]
[bootstrap]: 0.00026088
[type_inference]: 0.0040558
[event_method]: 1.63e-05
[auto_monad]: 6.103e-05
[graph_reusing]: 5.09003e-06
[pre_auto_parallel]: 2.31e-06
[py_interpret_to_execute]: 2.934e-05
[rewriter_before_opt_a]: 6.866e-05
[expand_dump_flag]: 3.25e-06
[jit_opt_a]: 0.00493075, [2]
    [Cycle 1]: 0.00155015, [27]
        [switch_simplify]: 4.828e-05
        [loop_unroll]: 2.861e-05
        [a_1]: 0.00048069
        [with_stream_mark]: 1.474e-05
        [recompute_prepare]: 6.88e-06
        [updatestate_depend_eliminate]: 3.35003e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 2.32001e-06
        [parameter_eliminate]: 1.70001e-06
        [specialize_transform]: 5.84999e-06
        [updatestate_useless_node_eliminater]: 4.74e-06
        [accelerated_algorithm]: 5.59e-06
        [meta_shard_fg_expand]: 2.84999e-06
        [get_grad_eliminate_]: 8.40999e-06
        [merge_forward]: 4.37998e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.471e-05
        [j_node_and_user_rematch]: 9.02999e-06
        [meta_fg_expand]: 2.29999e-06
        [replace_old_param]: 1.331e-05
        [inline_without_move]: 6.29001e-06
        [renormalize]: 0.00058229
        [add_forward_monad_depend]: 6.06e-06
        [auto_monad_grad]: 2.44001e-06
        [auto_monad_eliminator]: 1.326e-05
        [cse]: 2.448e-05
        [replace_applicator]: 1.079e-05
    [Cycle 2]: 0.00030284, [27]
        [switch_simplify]: 5.16998e-06
        [loop_unroll]: 4.81002e-06
        [a_1]: 8.605e-05
        [with_stream_mark]: 8.95999e-06
        [recompute_prepare]: 4.72998e-06
        [updatestate_depend_eliminate]: 2.39999e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.02998e-06
        [specialize_transform]: 4.77e-06
        [updatestate_useless_node_eliminater]: 4.68999e-06
        [accelerated_algorithm]: 4.60999e-06
        [meta_shard_fg_expand]: 1.00001e-06
        [get_grad_eliminate_]: 4.28001e-06
        [merge_forward]: 2.16998e-06
        [cell_reuse_recompute_pass]: 1.27e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.203e-05
        [j_node_and_user_rematch]: 7.43999e-06
        [meta_fg_expand]: 1.65001e-06
        [replace_old_param]: 6.99001e-06
        [inline_without_move]: 4.57998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 9.79984e-07
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 4.61002e-06
        [cse]: 9.76e-06
        [replace_applicator]: 5.52001e-06
[py_interpret_to_execute_after_opt_a]: 8.88002e-06
[rewriter_after_opt_a]: 3.307e-05
[convert_after_rewriter]: 6.85002e-06
[order_py_execute_after_rewriter]: 4.62e-06
[mutable_eliminate]: 0.00050498
[jit_opt_b]: 4.492e-05, [1]
    [Cycle 1]: 3.865e-05, [2]
        [frontend_op_eliminate]: 1.45e-05
        [inline_after_opt_a]: 1.351e-05
[cconv]: 2.271e-05
[loop_unroll]: 0.00040299
[jit_opt_after_cconv]: 0.00013616, [1]
    [Cycle 1]: 0.00012938, [11]
        [c_1]: 2.003e-05
        [parameter_eliminate]: 2.44001e-06
        [updatestate_depend_eliminate]: 5.58002e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [cse]: 1.747e-05
        [call_graph_tuple_transform]: 1.858e-05
        [tuple_list_get_item_eliminator]: 5.12e-06
        [none_parameter_eliminate]: 1.45001e-06
        [renormalize]: 5.19998e-07
        [switch_simplify]: 5.13002e-06
[remove_dup_value]: 1.238e-05
[partial_unused_args_eliminate]: 2.02999e-06
[environ_conv]: 4.70999e-06
[add_recomputation]: 4.575e-05
[cse_after_recomputation]: 2.016e-05, [1]
    [Cycle 1]: 1.497e-05, [1]
        [cse]: 9.08002e-06
[auto_monad_reorder]: 1.412e-05
[get_jit_bprop_graph]: 1.84998e-06
[rewriter_after_jit_bprop_graph]: 4.36002e-06
[opt_after_jit_grad]: 0.00043525
[symbol_engine_optimizer]: 6.992e-05, [1]
    [Cycle 1]: 6.4e-05, [6]
        [build]: 3.43e-06
        [elim_shapecalc]: 7.51001e-06
        [elim_not_effective]: 1.221e-05
        [opt_reshape]: 5.64998e-06
        [fold_const_symbol]: 8.06001e-06
        [renormalize]: 4.60015e-07
[validate]: 3.027e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00591767
[execute]: 7.07002e-06
Sums
    bootstrap                            :     0.000261s :  1.92%
    type_inference                       :     0.004056s : 29.78%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000061s :  0.45%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.22%
    rewriter_before_opt_a                :     0.000069s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.39%
    jit_opt_a.loop_unroll                :     0.000033s :  0.25%
    jit_opt_a.a_1                        :     0.000567s :  4.16%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000013s :  0.09%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000020s :  0.15%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.000582s :  4.28%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000034s :  0.25%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000505s :  3.71%
    jit_opt_b.frontend_op_eliminate      :     0.000014s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.10%
    cconv                                :     0.000023s :  0.17%
    loop_unroll                          :     0.000403s :  2.96%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000012s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000046s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000435s :  3.20%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000030s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005918s : 43.45%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000172    23
 1.29% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 3.39% :     0.000006s :      3: substitution.graph_param_transform
81.42% :     0.000140s :      4: substitution.inline
 1.55% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.78% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.22% :     0.000004s :      2: substitution.replace_old_param
 6.74% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004002     2
81.18% :     0.003249s :      1: type_inference.infer
18.82% :     0.000753s :      1: type_inference.specialize
------[replace.]   0.000060     6
67.18% :     0.000040s :      4: replace.inline
32.82% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000148     6
92.95% :     0.000138s :      4: match.inline
 7.05% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.23% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000001s :     10: predicate.addn_zero_filter
 1.96% :     0.000002s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.76% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000001s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000000s :      3: predicate.elim_not_effective
 0.67% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.27% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.42% :     0.000006s :     22: predicate.inline
 1.14% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.28% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.29% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.10% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.76% :     0.000002s :      3: predicate.mutable_eliminate
 0.60% :     0.000001s :      3: predicate.opt_reshape
 2.32% :     0.000003s :     15: predicate.partial_eliminate
 1.28% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.32% :     0.000001s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.48% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 1.18% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.53% :     0.000008s :     53: predicate.switch_simplify
 1.14% :     0.000001s :     10: predicate.tile_eliminate
 1.25% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.33% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000579    12
40.27% :     0.000233s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.61% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
55.12% :     0.000319s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018791    76
 0.26% :     0.000049s :      1: add_recomputation
 0.34% :     0.000064s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.45% :     0.000272s :      1: bootstrap
 0.13% :     0.000025s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000010s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.25% :     0.004933s :      1: jit_opt_a
 0.74% :     0.000139s :      1: jit_opt_after_cconv
 0.25% :     0.000047s :      1: jit_opt_b
 2.19% :     0.000411s :      1: loop_unroll
 2.74% :     0.000514s :      1: mutable_eliminate
 4.11% :     0.000772s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000022s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.36% :     0.000443s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.32% :     0.000248s :      1: renormalize.infer
 1.74% :     0.000327s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000036s :      1: rewriter_after_opt_a
 0.38% :     0.000071s :      1: rewriter_before_opt_a
 0.38% :     0.000072s :      1: symbol_engine_optimizer
31.54% :     0.005927s :      1: task_emit
21.64% :     0.004067s :      1: type_inference
 0.25% :     0.000048s :      1: validate


TotalTime = 0.0102107, [30]
[bootstrap]: 0.00022989
[type_inference]: 0.00359352
[event_method]: 1.575e-05
[auto_monad]: 4.781e-05
[graph_reusing]: 5.14e-06
[pre_auto_parallel]: 1.69998e-06
[py_interpret_to_execute]: 2.764e-05
[rewriter_before_opt_a]: 6.258e-05
[expand_dump_flag]: 2.32001e-06
[jit_opt_a]: 0.00434571, [2]
    [Cycle 1]: 0.00124259, [27]
        [switch_simplify]: 4.342e-05
        [loop_unroll]: 2.787e-05
        [a_1]: 0.00045798
        [with_stream_mark]: 1.276e-05
        [recompute_prepare]: 6.61999e-06
        [updatestate_depend_eliminate]: 3.03e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 2.39001e-06
        [parameter_eliminate]: 1.45001e-06
        [specialize_transform]: 5.71e-06
        [updatestate_useless_node_eliminater]: 5.30999e-06
        [accelerated_algorithm]: 5.55001e-06
        [meta_shard_fg_expand]: 1.46998e-06
        [get_grad_eliminate_]: 4.85001e-06
        [merge_forward]: 2.49001e-06
        [cell_reuse_recompute_pass]: 8.59989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.337e-05
        [j_node_and_user_rematch]: 8.22e-06
        [meta_fg_expand]: 1.97999e-06
        [replace_old_param]: 7.93999e-06
        [inline_without_move]: 5.07e-06
        [renormalize]: 0.00041951
        [add_forward_monad_depend]: 4.64998e-06
        [auto_monad_grad]: 1.40001e-06
        [auto_monad_eliminator]: 1.06e-05
        [cse]: 1.849e-05
        [replace_applicator]: 1.037e-05
    [Cycle 2]: 0.0003077, [27]
        [switch_simplify]: 5.40001e-06
        [loop_unroll]: 4.75999e-06
        [a_1]: 8.772e-05
        [with_stream_mark]: 7.7e-06
        [recompute_prepare]: 4.84998e-06
        [updatestate_depend_eliminate]: 2.36e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 9.60019e-07
        [specialize_transform]: 5.02999e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.13002e-06
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.25002e-06
        [cell_reuse_recompute_pass]: 1.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.159e-05
        [j_node_and_user_rematch]: 7.51999e-06
        [meta_fg_expand]: 1.44998e-06
        [replace_old_param]: 6.71e-06
        [inline_without_move]: 4.45e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 9.89996e-07
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 4.67998e-06
        [cse]: 1.004e-05
        [replace_applicator]: 4.77e-06
[py_interpret_to_execute_after_opt_a]: 7.00002e-06
[rewriter_after_opt_a]: 2.304e-05
[convert_after_rewriter]: 7.91001e-06
[order_py_execute_after_rewriter]: 4.43001e-06
[mutable_eliminate]: 0.00044836
[jit_opt_b]: 4.634e-05, [1]
    [Cycle 1]: 4.038e-05, [2]
        [frontend_op_eliminate]: 1.481e-05
        [inline_after_opt_a]: 1.43e-05
[cconv]: 1.916e-05
[loop_unroll]: 0.00040414
[jit_opt_after_cconv]: 0.00012829, [1]
    [Cycle 1]: 0.00012201, [11]
        [c_1]: 1.926e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 5.05001e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 1.513e-05
        [call_graph_tuple_transform]: 1.563e-05
        [tuple_list_get_item_eliminator]: 5.29e-06
        [none_parameter_eliminate]: 7.89994e-07
        [renormalize]: 4.69998e-07
        [switch_simplify]: 5.17e-06
[remove_dup_value]: 1.064e-05
[partial_unused_args_eliminate]: 2.10002e-06
[environ_conv]: 4.37e-06
[add_recomputation]: 3.725e-05
[cse_after_recomputation]: 2.034e-05, [1]
    [Cycle 1]: 1.515e-05, [1]
        [cse]: 9.32999e-06
[auto_monad_reorder]: 1.118e-05
[get_jit_bprop_graph]: 1.47999e-06
[rewriter_after_jit_bprop_graph]: 4.70999e-06
[opt_after_jit_grad]: 0.00044861
[symbol_engine_optimizer]: 6.979e-05, [1]
    [Cycle 1]: 6.318e-05, [6]
        [build]: 3.01999e-06
        [elim_shapecalc]: 7.2e-06
        [elim_not_effective]: 1.189e-05
        [opt_reshape]: 5.68997e-06
        [fold_const_symbol]: 8.15e-06
        [renormalize]: 4.09986e-07
[validate]: 2.741e-05
Sums
    bootstrap                            :     0.000230s :  3.34%
    type_inference                       :     0.003594s : 52.25%
    event_method                         :     0.000016s :  0.23%
    auto_monad                           :     0.000048s :  0.70%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.40%
    rewriter_before_opt_a                :     0.000063s :  0.91%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000049s :  0.71%
    jit_opt_a.loop_unroll                :     0.000033s :  0.47%
    jit_opt_a.a_1                        :     0.000546s :  7.94%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.15%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.16%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.14%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.36%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000420s :  6.10%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.22%
    jit_opt_a.cse                        :     0.000029s :  0.41%
    jit_opt_a.replace_applicator         :     0.000015s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.10%
    rewriter_after_opt_a                 :     0.000023s :  0.34%
    convert_after_rewriter               :     0.000008s :  0.12%
    order_py_execute_after_rewriter      :     0.000004s :  0.06%
    mutable_eliminate                    :     0.000448s :  6.52%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.21%
    cconv                                :     0.000019s :  0.28%
    loop_unroll                          :     0.000404s :  5.88%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000015s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.08%
    remove_dup_value                     :     0.000011s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000037s :  0.54%
    cse_after_recomputation.cse          :     0.000009s :  0.14%
    auto_monad_reorder                   :     0.000011s :  0.16%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000449s :  6.52%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000027s :  0.40%


Time group info:
------[substitution.]   0.000152    23
 1.13% :     0.000002s :      2: substitution.elim_not_effective
 0.79% :     0.000001s :      2: substitution.fold_const_symbol
 2.51% :     0.000004s :      3: substitution.graph_param_transform
83.17% :     0.000126s :      4: substitution.inline
 1.70% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.55% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.97% :     0.000003s :      2: substitution.replace_old_param
 6.18% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003546     2
80.69% :     0.002861s :      1: type_inference.infer
19.31% :     0.000685s :      1: type_inference.specialize
------[replace.]   0.000057     6
66.64% :     0.000038s :      4: replace.inline
33.36% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000132     6
93.70% :     0.000124s :      4: match.inline
 6.30% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.68% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.40% :     0.000002s :     10: predicate.addn_zero_filter
 2.16% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000001s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.31% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.22% :     0.000006s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.29% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.18% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.58% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.58% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000001s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.72% :     0.000008s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.26% :     0.000001s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.58% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.08% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000457    12
48.60% :     0.000222s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.25% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.14% :     0.000211s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011427    72
 0.35% :     0.000040s :      1: add_recomputation
 0.44% :     0.000051s :      1: auto_monad
 0.12% :     0.000014s :      1: auto_monad_reorder
 2.11% :     0.000241s :      1: bootstrap
 0.19% :     0.000022s :      1: cconv
 0.09% :     0.000010s :      1: convert_after_rewriter
 0.20% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.17% :     0.000019s :      1: event_method
 0.04% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
38.05% :     0.004348s :      1: jit_opt_a
 1.15% :     0.000131s :      1: jit_opt_after_cconv
 0.43% :     0.000049s :      1: jit_opt_b
 3.60% :     0.000411s :      1: loop_unroll
 4.00% :     0.000457s :      1: mutable_eliminate
 6.43% :     0.000734s :     26: opt.transform.jit_opt_a
 0.37% :     0.000042s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000023s :      4: opt.transform.jit_opt_b
 0.09% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.99% :     0.000456s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.04% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000003s :      1: pre_auto_parallel
 0.26% :     0.000030s :      1: py_interpret_to_execute
 0.08% :     0.000009s :      1: py_interpret_to_execute_after_opt_a
 0.21% :     0.000024s :      1: remove_dup_value
 1.75% :     0.000200s :      1: renormalize.infer
 1.87% :     0.000213s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000026s :      1: rewriter_after_opt_a
 0.57% :     0.000065s :      1: rewriter_before_opt_a
 0.63% :     0.000072s :      1: symbol_engine_optimizer
31.54% :     0.003604s :      1: type_inference


TotalTime = 0.0206082, [33]
[bootstrap]: 0.00019625
[type_inference]: 0.00984876
[event_method]: 4.824e-05
[auto_monad]: 8.808e-05
[graph_reusing]: 6.44999e-06
[pre_auto_parallel]: 2.43998e-06
[py_interpret_to_execute]: 2.607e-05
[rewriter_before_opt_a]: 8.8e-05
[expand_dump_flag]: 2.53e-06
[jit_opt_a]: 0.00830192, [2]
    [Cycle 1]: 0.00460479, [27]
        [switch_simplify]: 6.094e-05
        [loop_unroll]: 3.933e-05
        [a_1]: 0.00081673
        [with_stream_mark]: 2.052e-05
        [recompute_prepare]: 1.773e-05
        [updatestate_depend_eliminate]: 7.92e-06
        [updatestate_assign_eliminate]: 5.99e-06
        [updatestate_loads_eliminate]: 5.59998e-06
        [parameter_eliminate]: 1.97001e-06
        [specialize_transform]: 1.315e-05
        [updatestate_useless_node_eliminater]: 1.238e-05
        [accelerated_algorithm]: 1.203e-05
        [meta_shard_fg_expand]: 2.96001e-06
        [get_grad_eliminate_]: 1.208e-05
        [merge_forward]: 7.36001e-06
        [cell_reuse_recompute_pass]: 8.00006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.459e-05
        [j_node_and_user_rematch]: 2.223e-05
        [meta_fg_expand]: 0.00156563
        [replace_old_param]: 5.412e-05
        [inline_without_move]: 5.311e-05
        [renormalize]: 0.00160356
        [add_forward_monad_depend]: 5.04e-06
        [auto_monad_grad]: 2.12001e-06
        [auto_monad_eliminator]: 9.74999e-06
        [cse]: 1.78e-05
        [replace_applicator]: 1.075e-05
    [Cycle 2]: 0.00029197, [27]
        [switch_simplify]: 4.3e-06
        [loop_unroll]: 3.48e-06
        [a_1]: 4.302e-05
        [with_stream_mark]: 9.47999e-06
        [recompute_prepare]: 3.78999e-06
        [updatestate_depend_eliminate]: 2.04e-06
        [updatestate_assign_eliminate]: 1.69e-06
        [updatestate_loads_eliminate]: 1.80001e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 3.84002e-06
        [updatestate_useless_node_eliminater]: 3.63999e-06
        [accelerated_algorithm]: 3.68e-06
        [meta_shard_fg_expand]: 1.31002e-06
        [get_grad_eliminate_]: 3.7e-06
        [merge_forward]: 2.07999e-06
        [cell_reuse_recompute_pass]: 2.34999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.048e-05
        [j_node_and_user_rematch]: 6.24999e-06
        [meta_fg_expand]: 3.544e-05
        [replace_old_param]: 4.12003e-06
        [inline_without_move]: 3.6e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 5.10001e-06
        [cse]: 1.039e-05
        [replace_applicator]: 3.90998e-06
[py_interpret_to_execute_after_opt_a]: 8.97999e-06
[rewriter_after_opt_a]: 3.089e-05
[convert_after_rewriter]: 6.10002e-06
[order_py_execute_after_rewriter]: 4.53001e-06
[mutable_eliminate]: 0.00055368
[jit_opt_b]: 4.136e-05, [1]
    [Cycle 1]: 3.464e-05, [2]
        [frontend_op_eliminate]: 1.16e-05
        [inline_after_opt_a]: 1.157e-05
[cconv]: 1.825e-05
[loop_unroll]: 0.00042674
[jit_opt_after_cconv]: 0.00011609, [1]
    [Cycle 1]: 0.00010931, [11]
        [c_1]: 1.221e-05
        [parameter_eliminate]: 2.66e-06
        [updatestate_depend_eliminate]: 4.37e-06
        [updatestate_assign_eliminate]: 1.89e-06
        [updatestate_loads_eliminate]: 1.77999e-06
        [cse]: 1.545e-05
        [call_graph_tuple_transform]: 1.186e-05
        [tuple_list_get_item_eliminator]: 3.75003e-06
        [none_parameter_eliminate]: 1.35999e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 3.6e-06
[remove_dup_value]: 1.065e-05
[partial_unused_args_eliminate]: 1.87001e-06
[environ_conv]: 4.47998e-06
[add_recomputation]: 2.705e-05
[cse_after_recomputation]: 1.964e-05, [1]
    [Cycle 1]: 1.466e-05, [1]
        [cse]: 8.23999e-06
[auto_monad_reorder]: 9.25001e-06
[get_jit_bprop_graph]: 1.60001e-06
[rewriter_after_jit_bprop_graph]: 4.97e-06
[opt_after_jit_grad]: 0.00044276
[symbol_engine_optimizer]: 6.323e-05, [1]
    [Cycle 1]: 5.733e-05, [6]
        [build]: 2.46e-06
        [elim_shapecalc]: 6.71999e-06
        [elim_not_effective]: 8.79e-06
        [opt_reshape]: 4.48999e-06
        [fold_const_symbol]: 5.51e-06
        [renormalize]: 4.09986e-07
[validate]: 2.364e-05
[backend_pass]: 6.19999e-07
[task_emit]: 1.821e-05
[execute]: 1.23002e-06
Sums
    bootstrap                            :     0.000196s :  1.18%
    type_inference                       :     0.009849s : 59.33%
    event_method                         :     0.000048s :  0.29%
    auto_monad                           :     0.000088s :  0.53%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.16%
    rewriter_before_opt_a                :     0.000088s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.39%
    jit_opt_a.loop_unroll                :     0.000043s :  0.26%
    jit_opt_a.a_1                        :     0.000860s :  5.18%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.10%
    jit_opt_a.merge_forward              :     0.000009s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001601s :  9.64%
    jit_opt_a.replace_old_param          :     0.000058s :  0.35%
    jit_opt_a.inline_without_move        :     0.000057s :  0.34%
    jit_opt_a.renormalize                :     0.001604s :  9.66%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.09%
    jit_opt_a.cse                        :     0.000028s :  0.17%
    jit_opt_a.replace_applicator         :     0.000015s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.05%
    rewriter_after_opt_a                 :     0.000031s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000554s :  3.34%
    jit_opt_b.frontend_op_eliminate      :     0.000012s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.07%
    cconv                                :     0.000018s :  0.11%
    loop_unroll                          :     0.000427s :  2.57%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000012s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000011s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000027s :  0.16%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000009s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000443s :  2.67%
    symbol_engine_optimizer.build        :     0.000002s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000024s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000018s :  0.11%
    execute                              :     0.000001s :  0.01%

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.354.437 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd84048dc0> should be wrapped by jit only once.

Time group info:
------[substitution.]   0.000225    50
 0.59% :     0.000001s :      1: substitution.elim_not_effective
 0.29% :     0.000001s :      1: substitution.fold_const_symbol
 1.15% :     0.000003s :      1: substitution.graph_param_transform
63.98% :     0.000144s :      6: substitution.inline
 7.80% :     0.000018s :      2: substitution.inline_without_move
 2.66% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.27% :     0.000003s :      2: substitution.minmaximum_grad
 1.52% :     0.000003s :      3: substitution.partial_eliminate
 3.07% :     0.000007s :      9: substitution.remove_not_recompute_node
 1.98% :     0.000004s :      2: substitution.replace_applicator
 1.76% :     0.000004s :      5: substitution.replace_old_param
 1.05% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.63% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.54% :     0.000003s :      2: substitution.tuple_list_get_item_depend_reorder
 8.70% :     0.000020s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009779     2
88.60% :     0.008664s :      1: type_inference.infer
11.40% :     0.001115s :      1: type_inference.specialize
------[replace.]   0.000062     8
77.25% :     0.000048s :      6: replace.inline
22.75% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000155     8
90.55% :     0.000140s :      6: match.inline
 9.45% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000169  1109
 1.70% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.58% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.27% :     0.000002s :     18: predicate.addn_check_dump
 1.58% :     0.000003s :     18: predicate.addn_zero_filter
 2.24% :     0.000004s :     18: predicate.arithmetic_simplify
 1.47% :     0.000002s :     18: predicate.cast_eliminate
 0.14% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.29% :     0.000002s :     18: predicate.compare_switch_simplify
 1.62% :     0.000003s :     18: predicate.depend_value_elim
 1.28% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.48% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.43% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      1: predicate.elim_not_effective
 0.34% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.57% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.24% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.24% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.46% :     0.000002s :     18: predicate.environ_get_eliminate
 1.25% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.08% :     0.000000s :      1: predicate.fold_const_symbol
 1.20% :     0.000002s :     12: predicate.get_grad_eliminate
 0.07% :     0.000000s :      1: predicate.graph_param_transform
 4.36% :     0.000007s :     28: predicate.inline
 3.34% :     0.000006s :     33: predicate.inline_without_move
 0.52% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :     12: predicate.less_batch_normalization
 1.53% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.67% :     0.000003s :     21: predicate.load_eliminater
 0.70% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.54% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.78% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.28% :     0.000002s :     18: predicate.merge_addn
 1.30% :     0.000002s :     18: predicate.minmaximum_grad
 0.74% :     0.000001s :      1: predicate.mutable_eliminate
 0.20% :     0.000000s :      1: predicate.opt_reshape
 2.01% :     0.000003s :     21: predicate.partial_eliminate
 1.48% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.75% :     0.000003s :     18: predicate.reduce_eliminate
 1.59% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.54% :     0.000003s :     22: predicate.replace_applicator
 1.71% :     0.000003s :     33: predicate.replace_old_param
 0.12% :     0.000000s :      1: predicate.reset_defer_inline
 1.56% :     0.000003s :     18: predicate.reshape_eliminate
 1.49% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.46% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.63% :     0.000003s :     18: predicate.same_eliminate
 0.64% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.30% :     0.000001s :      2: predicate.special_op_eliminate
 1.23% :     0.000002s :     12: predicate.specialize_transform
 1.97% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.52% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.72% :     0.000005s :     26: predicate.switch_defer_inline
 2.44% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.62% :     0.000011s :     71: predicate.switch_simplify
 1.53% :     0.000003s :     18: predicate.tile_eliminate
 1.36% :     0.000002s :     18: predicate.transpose_eliminate
 1.55% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.58% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.83% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 2.04% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.52% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 2.00% :     0.000003s :     18: predicate.value_based_eliminate
 0.11% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.27% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001148    22
72.55% :     0.000833s :     13: func_graph_cloner_run.FuncGraphClonerGraph
27.45% :     0.000315s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023492    76
 0.13% :     0.000030s :      1: add_recomputation
 0.39% :     0.000092s :      1: auto_monad
 0.05% :     0.000012s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.88% :     0.000208s :      1: bootstrap
 0.09% :     0.000021s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.22% :     0.000052s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
35.35% :     0.008305s :      1: jit_opt_a
 0.51% :     0.000119s :      1: jit_opt_after_cconv
 0.19% :     0.000044s :      1: jit_opt_b
 1.85% :     0.000434s :      1: loop_unroll
 2.40% :     0.000563s :      1: mutable_eliminate
 5.19% :     0.001219s :     26: opt.transform.jit_opt_a
 0.12% :     0.000028s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000022s :      4: opt.transform.symbol_engine_opt
 1.91% :     0.000449s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000029s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000013s :      1: remove_dup_value
 5.10% :     0.001197s :      1: renormalize.infer
 1.70% :     0.000399s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000034s :      1: rewriter_after_opt_a
 0.39% :     0.000091s :      1: rewriter_before_opt_a
 0.28% :     0.000066s :      1: symbol_engine_optimizer
 0.10% :     0.000022s :      1: task_emit
41.97% :     0.009859s :      1: type_inference
 0.16% :     0.000039s :      1: validate


TotalTime = 0.0309706, [33]
[bootstrap]: 0.00019255
[type_inference]: 0.00954114
[event_method]: 4.485e-05
[auto_monad]: 9.118e-05
[graph_reusing]: 6.69999e-06
[pre_auto_parallel]: 2.88e-06
[py_interpret_to_execute]: 2.33e-05
[rewriter_before_opt_a]: 8.506e-05
[expand_dump_flag]: 3.79002e-06
[jit_opt_a]: 0.0110378, [3]
    [Cycle 1]: 0.00547865, [27]
        [switch_simplify]: 5.998e-05
        [loop_unroll]: 3.801e-05
        [a_1]: 0.00087249
        [with_stream_mark]: 2.426e-05
        [recompute_prepare]: 1.791e-05
        [updatestate_depend_eliminate]: 7.58001e-06
        [updatestate_assign_eliminate]: 6.39999e-06
        [updatestate_loads_eliminate]: 6.01998e-06
        [parameter_eliminate]: 2.49001e-06
        [specialize_transform]: 1.247e-05
        [updatestate_useless_node_eliminater]: 1.175e-05
        [accelerated_algorithm]: 1.204e-05
        [meta_shard_fg_expand]: 3.08e-06
        [get_grad_eliminate_]: 1.199e-05
        [merge_forward]: 8.35999e-06
        [cell_reuse_recompute_pass]: 8.80013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.604e-05
        [j_node_and_user_rematch]: 2.496e-05
        [meta_fg_expand]: 0.00135926
        [replace_old_param]: 5.071e-05
        [inline_without_move]: 5.036e-05
        [renormalize]: 0.00239005
        [add_forward_monad_depend]: 9.89001e-06
        [auto_monad_grad]: 5.64e-06
        [auto_monad_eliminator]: 5.212e-05
        [cse]: 0.00014839
        [replace_applicator]: 6.667e-05
    [Cycle 2]: 0.00203055, [27]
        [switch_simplify]: 3.771e-05
        [loop_unroll]: 3.556e-05
        [a_1]: 0.00110576
        [with_stream_mark]: 1.383e-05
        [recompute_prepare]: 8.87999e-06
        [updatestate_depend_eliminate]: 4.33001e-06
        [updatestate_assign_eliminate]: 3.65e-06
        [updatestate_loads_eliminate]: 3.49001e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 7.51001e-06
        [updatestate_useless_node_eliminater]: 7.05e-06
        [accelerated_algorithm]: 7.18e-06
        [meta_shard_fg_expand]: 2.44999e-06
        [get_grad_eliminate_]: 6.74999e-06
        [merge_forward]: 4.03999e-06
        [cell_reuse_recompute_pass]: 1.10999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.59e-05
        [j_node_and_user_rematch]: 1.206e-05
        [meta_fg_expand]: 3.671e-05
        [replace_old_param]: 1.026e-05
        [inline_without_move]: 6.74001e-06
        [renormalize]: 0.000498
        [add_forward_monad_depend]: 4.55001e-06
        [auto_monad_grad]: 1.45001e-06
        [auto_monad_eliminator]: 1.312e-05
        [cse]: 2.328e-05
        [replace_applicator]: 1.306e-05
    [Cycle 3]: 0.00040145, [27]
        [switch_simplify]: 7.43e-06
        [loop_unroll]: 6.65002e-06
        [a_1]: 0.00014205
        [with_stream_mark]: 8.79e-06
        [recompute_prepare]: 6.73e-06
        [updatestate_depend_eliminate]: 3.61001e-06
        [updatestate_assign_eliminate]: 3.2e-06
        [updatestate_loads_eliminate]: 2.92002e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 6.69001e-06
        [updatestate_useless_node_eliminater]: 6.64999e-06
        [accelerated_algorithm]: 6.71e-06
        [meta_shard_fg_expand]: 1.33002e-06
        [get_grad_eliminate_]: 6.34001e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 1.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.547e-05
        [j_node_and_user_rematch]: 1.133e-05
        [meta_fg_expand]: 2.33002e-06
        [replace_old_param]: 8.53001e-06
        [inline_without_move]: 6.59999e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.09003e-06
        [auto_monad_grad]: 6.40022e-07
        [auto_monad_eliminator]: 7.79997e-06
        [cse]: 1.612e-05
        [replace_applicator]: 7.56001e-06
[py_interpret_to_execute_after_opt_a]: 1.04e-05
[rewriter_after_opt_a]: 4.299e-05
[convert_after_rewriter]: 8.35001e-06
[order_py_execute_after_rewriter]: 5.56e-06
[mutable_eliminate]: 0.00059416
[jit_opt_b]: 5.912e-05, [1]
    [Cycle 1]: 5.209e-05, [2]
        [frontend_op_eliminate]: 2.142e-05
        [inline_after_opt_a]: 1.954e-05
[cconv]: 2.127e-05
[loop_unroll]: 0.00041343
[jit_opt_after_cconv]: 0.00016552, [1]
    [Cycle 1]: 0.00015856, [11]
        [c_1]: 3.017e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 6.38003e-06
        [updatestate_assign_eliminate]: 3.85e-06
        [updatestate_loads_eliminate]: 3.04999e-06
        [cse]: 2.543e-05
        [call_graph_tuple_transform]: 2.262e-05
        [tuple_list_get_item_eliminator]: 6.81001e-06
        [none_parameter_eliminate]: 1.99e-06
        [renormalize]: 3.69997e-07
        [switch_simplify]: 6.96001e-06
[remove_dup_value]: 1.825e-05
[partial_unused_args_eliminate]: 2.37999e-06
[environ_conv]: 6.99001e-06
[add_recomputation]: 5.254e-05
[cse_after_recomputation]: 3.34e-05, [1]
    [Cycle 1]: 2.233e-05, [1]
        [cse]: 1.614e-05
[auto_monad_reorder]: 2.215e-05
[get_jit_bprop_graph]: 1.67999e-06
[rewriter_after_jit_bprop_graph]: 5.20999e-06
[opt_after_jit_grad]: 0.00045731
[symbol_engine_optimizer]: 8.341e-05, [1]
    [Cycle 1]: 7.734e-05, [6]
        [build]: 3.78999e-06
        [elim_shapecalc]: 9.57001e-06
        [elim_not_effective]: 1.633e-05
        [opt_reshape]: 7.65e-06
        [fold_const_symbol]: 1.214e-05
        [renormalize]: 5.10016e-07
[validate]: 4.013e-05
[backend_pass]: 1.14e-06
[task_emit]: 0.00769748
[execute]: 8.00999e-06
Sums
    bootstrap                            :     0.000193s :  0.71%
    type_inference                       :     0.009541s : 35.24%
    event_method                         :     0.000045s :  0.17%
    auto_monad                           :     0.000091s :  0.34%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000023s :  0.09%
    rewriter_before_opt_a                :     0.000085s :  0.31%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000105s :  0.39%
    jit_opt_a.loop_unroll                :     0.000080s :  0.30%
    jit_opt_a.a_1                        :     0.002120s :  7.83%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000034s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000027s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000026s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.09%
    jit_opt_a.merge_forward              :     0.000016s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000057s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000048s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001398s :  5.16%
    jit_opt_a.replace_old_param          :     0.000070s :  0.26%
    jit_opt_a.inline_without_move        :     0.000064s :  0.24%
    jit_opt_a.renormalize                :     0.002888s : 10.67%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000073s :  0.27%
    jit_opt_a.cse                        :     0.000188s :  0.69%
    jit_opt_a.replace_applicator         :     0.000087s :  0.32%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.04%
    rewriter_after_opt_a                 :     0.000043s :  0.16%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000594s :  2.19%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.07%
    cconv                                :     0.000021s :  0.08%
    loop_unroll                          :     0.000413s :  1.53%
    jit_opt_after_cconv.c_1              :     0.000030s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000018s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000053s :  0.19%
    cse_after_recomputation.cse          :     0.000016s :  0.06%
    auto_monad_reorder                   :     0.000022s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000457s :  1.69%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000040s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007697s : 28.43%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000509   118
 5.18% :     0.000026s :      1: substitution.arithmetic_simplify
 0.49% :     0.000002s :      4: substitution.elim_not_effective
 0.38% :     0.000002s :      4: substitution.fold_const_symbol
 1.32% :     0.000007s :      5: substitution.graph_param_transform
60.05% :     0.000305s :     12: substitution.inline
 3.27% :     0.000017s :      2: substitution.inline_without_move
 2.03% :     0.000010s :     16: substitution.j_node_and_user_rematch
 1.85% :     0.000009s :      7: substitution.minmaximum_grad
 1.45% :     0.000007s :      5: substitution.partial_eliminate
 2.26% :     0.000011s :     16: substitution.remove_not_recompute_node
 4.99% :     0.000025s :     10: substitution.replace_applicator
 1.38% :     0.000007s :      7: substitution.replace_old_param
 0.50% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.29% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.73% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.83% :     0.000040s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009477     2
88.54% :     0.008390s :      1: type_inference.infer
11.46% :     0.001086s :      1: type_inference.specialize
------[replace.]   0.000140    20
 3.53% :     0.000005s :      1: replace.arithmetic_simplify
62.32% :     0.000087s :     12: replace.inline
34.15% :     0.000048s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000345    20
 7.34% :     0.000025s :      1: match.arithmetic_simplify
86.70% :     0.000299s :     12: match.inline
 5.96% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000383  2790
 1.56% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.42% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.41% :     0.000005s :     46: predicate.addn_check_dump
 1.67% :     0.000006s :     46: predicate.addn_zero_filter
 2.02% :     0.000008s :     47: predicate.arithmetic_simplify
 1.58% :     0.000006s :     47: predicate.cast_eliminate
 0.20% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.43% :     0.000005s :     46: predicate.compare_switch_simplify
 1.53% :     0.000006s :     46: predicate.depend_value_elim
 1.45% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.48% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.49% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.32% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.14% :     0.000001s :      5: predicate.elim_not_effective
 0.22% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.49% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.61% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.65% :     0.000006s :     47: predicate.environ_get_eliminate
 1.45% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      5: predicate.fold_const_symbol
 0.84% :     0.000003s :     21: predicate.get_grad_eliminate
 0.12% :     0.000000s :      5: predicate.graph_param_transform
 4.37% :     0.000017s :     76: predicate.inline
 1.74% :     0.000007s :     42: predicate.inline_without_move
 0.39% :     0.000002s :     21: predicate.j_node_and_user_rematch
 0.93% :     0.000004s :     21: predicate.less_batch_normalization
 1.85% :     0.000007s :     54: predicate.list_to_tuple_eliminator_
 1.99% :     0.000008s :     59: predicate.load_eliminater
 0.59% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.08% :     0.000012s :     89: predicate.loop_unroll_before_grad
 1.83% :     0.000007s :     52: predicate.make_slice_get_slice_eliminator
 1.45% :     0.000006s :     46: predicate.merge_addn
 1.50% :     0.000006s :     47: predicate.minmaximum_grad
 0.59% :     0.000002s :      5: predicate.mutable_eliminate
 0.22% :     0.000001s :      5: predicate.opt_reshape
 2.43% :     0.000009s :     59: predicate.partial_eliminate
 1.44% :     0.000006s :     46: predicate.print_const_string_wrapper
 2.01% :     0.000008s :     47: predicate.reduce_eliminate
 1.89% :     0.000007s :     54: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.57% :     0.000010s :    105: predicate.replace_applicator
 0.90% :     0.000003s :     42: predicate.replace_old_param
 0.14% :     0.000001s :      5: predicate.reset_defer_inline
 1.80% :     0.000007s :     47: predicate.reshape_eliminate
 1.46% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.39% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.46% :     0.000006s :     46: predicate.same_eliminate
 0.48% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.39% :     0.000002s :     10: predicate.special_op_eliminate
 0.91% :     0.000003s :     21: predicate.specialize_transform
 1.63% :     0.000006s :     46: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.20% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.63% :     0.000010s :     66: predicate.switch_defer_inline
 2.51% :     0.000010s :     66: predicate.switch_layer_defer_inline
 6.13% :     0.000024s :    160: predicate.switch_simplify
 1.50% :     0.000006s :     47: predicate.tile_eliminate
 1.57% :     0.000006s :     47: predicate.transpose_eliminate
 1.79% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.68% :     0.000006s :     47: predicate.tuple_list_get_item_depend_reorder
 3.28% :     0.000013s :     64: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.91% :     0.000007s :     54: predicate.tuple_to_list_eliminator_
 1.92% :     0.000007s :     59: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000011s :     80: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000006s :     46: predicate.value_based_eliminate
 0.17% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.28% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001359    29
59.12% :     0.000803s :     13: func_graph_cloner_run.FuncGraphClonerGraph
40.88% :     0.000555s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.036734    91
 0.15% :     0.000056s :      1: add_recomputation
 0.26% :     0.000095s :      1: auto_monad
 0.07% :     0.000025s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.56% :     0.000207s :      1: bootstrap
 0.06% :     0.000024s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000036s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.13% :     0.000049s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
30.06% :     0.011041s :      1: jit_opt_a
 0.46% :     0.000168s :      1: jit_opt_after_cconv
 0.17% :     0.000062s :      1: jit_opt_b
 1.15% :     0.000421s :      1: loop_unroll
 1.64% :     0.000603s :      1: mutable_eliminate
 7.42% :     0.002726s :     39: opt.transform.jit_opt_a
 0.17% :     0.000063s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000035s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000042s :      4: opt.transform.symbol_engine_opt
 1.27% :     0.000465s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000026s :      1: py_interpret_to_execute
 0.03% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000021s :      1: remove_dup_value
 4.28% :     0.001571s :      2: renormalize.infer
 3.55% :     0.001304s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000046s :      1: rewriter_after_opt_a
 0.24% :     0.000088s :      1: rewriter_before_opt_a
 0.23% :     0.000086s :      1: symbol_engine_optimizer
20.99% :     0.007711s :      1: task_emit
26.01% :     0.009553s :      1: type_inference
 0.17% :     0.000063s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.360.686 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c36b8b0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.361.298 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.366.663 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80ef670> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.367.290 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0101489, [30]
[bootstrap]: 0.00023121
[type_inference]: 0.00355579
[event_method]: 1.527e-05
[auto_monad]: 4.841e-05
[graph_reusing]: 4.83001e-06
[pre_auto_parallel]: 1.66e-06
[py_interpret_to_execute]: 2.569e-05
[rewriter_before_opt_a]: 6.436e-05
[expand_dump_flag]: 2.49001e-06
[jit_opt_a]: 0.00434478, [2]
    [Cycle 1]: 0.00123339, [27]
        [switch_simplify]: 4.298e-05
        [loop_unroll]: 2.776e-05
        [a_1]: 0.00045695
        [with_stream_mark]: 1.241e-05
        [recompute_prepare]: 6.59001e-06
        [updatestate_depend_eliminate]: 2.94001e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [parameter_eliminate]: 1.37999e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 4.91002e-06
        [accelerated_algorithm]: 5.23002e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 4.88001e-06
        [merge_forward]: 2.64001e-06
        [cell_reuse_recompute_pass]: 1.03001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.201e-05
        [j_node_and_user_rematch]: 8.05999e-06
        [meta_fg_expand]: 2.32999e-06
        [replace_old_param]: 7.31999e-06
        [inline_without_move]: 5.27001e-06
        [renormalize]: 0.00041311
        [add_forward_monad_depend]: 4.33001e-06
        [auto_monad_grad]: 1.65001e-06
        [auto_monad_eliminator]: 1.068e-05
        [cse]: 1.859e-05
        [replace_applicator]: 1.082e-05
    [Cycle 2]: 0.00032922, [27]
        [switch_simplify]: 5.39e-06
        [loop_unroll]: 4.82998e-06
        [a_1]: 8.767e-05
        [with_stream_mark]: 8.07e-06
        [recompute_prepare]: 4.95001e-06
        [updatestate_depend_eliminate]: 2.37999e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 9.99979e-07
        [specialize_transform]: 4.77e-06
        [updatestate_useless_node_eliminater]: 4.73001e-06
        [accelerated_algorithm]: 4.84998e-06
        [meta_shard_fg_expand]: 1.06002e-06
        [get_grad_eliminate_]: 4.26001e-06
        [merge_forward]: 2.31e-06
        [cell_reuse_recompute_pass]: 1.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.181e-05
        [j_node_and_user_rematch]: 7.39002e-06
        [meta_fg_expand]: 1.45999e-06
        [replace_old_param]: 7.32997e-06
        [inline_without_move]: 4.47e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.35999e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 5.45001e-06
        [cse]: 1.064e-05
        [replace_applicator]: 5.24e-06
[py_interpret_to_execute_after_opt_a]: 7.21999e-06
[rewriter_after_opt_a]: 2.575e-05
[convert_after_rewriter]: 5.67999e-06
[order_py_execute_after_rewriter]: 4.42e-06
[mutable_eliminate]: 0.00045167
[jit_opt_b]: 4.677e-05, [1]
    [Cycle 1]: 4.086e-05, [2]
        [frontend_op_eliminate]: 1.556e-05
        [inline_after_opt_a]: 1.398e-05
[cconv]: 1.94e-05
[loop_unroll]: 0.00040581
[jit_opt_after_cconv]: 0.00012885, [1]
    [Cycle 1]: 0.00012266, [11]
        [c_1]: 1.933e-05
        [parameter_eliminate]: 2.36e-06
        [updatestate_depend_eliminate]: 5.02999e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.519e-05
        [call_graph_tuple_transform]: 1.629e-05
        [tuple_list_get_item_eliminator]: 5.25999e-06
        [none_parameter_eliminate]: 8.60018e-07
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.19e-06
[remove_dup_value]: 9.98998e-06
[partial_unused_args_eliminate]: 1.86998e-06
[environ_conv]: 3.82998e-06
[add_recomputation]: 3.517e-05
[cse_after_recomputation]: 1.866e-05, [1]
    [Cycle 1]: 1.372e-05, [1]
        [cse]: 8.28999e-06
[auto_monad_reorder]: 1.083e-05
[get_jit_bprop_graph]: 1.48002e-06
[rewriter_after_jit_bprop_graph]: 4.63001e-06
[opt_after_jit_grad]: 0.00043682
[symbol_engine_optimizer]: 6.869e-05, [1]
    [Cycle 1]: 6.256e-05, [6]
        [build]: 3.13e-06
        [elim_shapecalc]: 7.05e-06
        [elim_not_effective]: 1.199e-05
        [opt_reshape]: 5.23002e-06
        [fold_const_symbol]: 8.03001e-06
        [renormalize]: 4.00003e-07
[validate]: 2.427e-05
Sums
    bootstrap                            :     0.000231s :  3.39%
    type_inference                       :     0.003556s : 52.14%
    event_method                         :     0.000015s :  0.22%
    auto_monad                           :     0.000048s :  0.71%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.38%
    rewriter_before_opt_a                :     0.000064s :  0.94%
    expand_dump_flag                     :     0.000002s :  0.04%
    jit_opt_a.switch_simplify            :     0.000048s :  0.71%
    jit_opt_a.loop_unroll                :     0.000033s :  0.48%
    jit_opt_a.a_1                        :     0.000545s :  7.99%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.13%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000015s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.06%
    jit_opt_a.replace_old_param          :     0.000015s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000413s :  6.06%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.24%
    jit_opt_a.cse                        :     0.000029s :  0.43%
    jit_opt_a.replace_applicator         :     0.000016s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.11%
    rewriter_after_opt_a                 :     0.000026s :  0.38%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000004s :  0.06%
    mutable_eliminate                    :     0.000452s :  6.62%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.21%
    cconv                                :     0.000019s :  0.28%
    loop_unroll                          :     0.000406s :  5.95%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000015s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.08%
    remove_dup_value                     :     0.000010s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000035s :  0.52%
    cse_after_recomputation.cse          :     0.000008s :  0.12%
    auto_monad_reorder                   :     0.000011s :  0.16%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000437s :  6.41%
    symbol_engine_optimizer.build        :     0.000003s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.18%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000024s :  0.36%


Time group info:
------[substitution.]   0.000149    23
 1.18% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.67% :     0.000004s :      3: substitution.graph_param_transform
82.87% :     0.000124s :      4: substitution.inline
 1.63% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.80% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.21% :     0.000003s :      2: substitution.replace_old_param
 5.91% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003511     2
80.48% :     0.002826s :      1: type_inference.infer
19.52% :     0.000685s :      1: type_inference.specialize
------[replace.]   0.000058     6
67.72% :     0.000039s :      4: replace.inline
32.28% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000129     6
94.05% :     0.000121s :      4: match.inline
 5.95% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000109   691
 1.54% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     10: predicate.addn_check_dump
 1.43% :     0.000002s :     10: predicate.addn_zero_filter
 2.09% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.50% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.23% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000000s :      3: predicate.graph_param_transform
 5.83% :     0.000006s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.48% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.26% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.03% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.20% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.62% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.91% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.21% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.60% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.65% :     0.000008s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.46% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.34% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000456    12
48.68% :     0.000222s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.46% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.85% :     0.000209s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011362    72
 0.33% :     0.000038s :      1: add_recomputation
 0.45% :     0.000052s :      1: auto_monad
 0.12% :     0.000013s :      1: auto_monad_reorder
 2.14% :     0.000243s :      1: bootstrap
 0.19% :     0.000022s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.18% :     0.000021s :      1: cse_after_recomputation
 0.06% :     0.000006s :      1: environ_conv
 0.16% :     0.000019s :      1: event_method
 0.04% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
38.26% :     0.004347s :      1: jit_opt_a
 1.16% :     0.000132s :      1: jit_opt_after_cconv
 0.43% :     0.000049s :      1: jit_opt_b
 3.64% :     0.000413s :      1: loop_unroll
 4.05% :     0.000460s :      1: mutable_eliminate
 6.44% :     0.000732s :     26: opt.transform.jit_opt_a
 0.38% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000024s :      4: opt.transform.jit_opt_b
 0.10% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000029s :      4: opt.transform.symbol_engine_opt
 3.91% :     0.000444s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000028s :      1: py_interpret_to_execute
 0.08% :     0.000009s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000012s :      1: remove_dup_value
 1.71% :     0.000194s :      1: renormalize.infer
 1.87% :     0.000213s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000029s :      1: rewriter_after_opt_a
 0.59% :     0.000067s :      1: rewriter_before_opt_a
 0.63% :     0.000071s :      1: symbol_engine_optimizer
31.39% :     0.003567s :      1: type_inference


TotalTime = 0.0223735, [33]
[bootstrap]: 0.00019834
[type_inference]: 0.0108405
[event_method]: 5.084e-05
[auto_monad]: 9.932e-05
[graph_reusing]: 7.55e-06
[pre_auto_parallel]: 3.01001e-06
[py_interpret_to_execute]: 2.929e-05
[rewriter_before_opt_a]: 9.503e-05
[expand_dump_flag]: 3.97e-06
[jit_opt_a]: 0.00895311, [2]
    [Cycle 1]: 0.00499846, [27]
        [switch_simplify]: 6.133e-05
        [loop_unroll]: 3.88e-05
        [a_1]: 0.00093427
        [with_stream_mark]: 2.508e-05
        [recompute_prepare]: 2.352e-05
        [updatestate_depend_eliminate]: 7.55e-06
        [updatestate_assign_eliminate]: 5.81e-06
        [updatestate_loads_eliminate]: 5.89e-06
        [parameter_eliminate]: 2.29001e-06
        [specialize_transform]: 1.372e-05
        [updatestate_useless_node_eliminater]: 1.254e-05
        [accelerated_algorithm]: 1.301e-05
        [meta_shard_fg_expand]: 3.71999e-06
        [get_grad_eliminate_]: 1.243e-05
        [merge_forward]: 8.13999e-06
        [cell_reuse_recompute_pass]: 1.01002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.522e-05
        [j_node_and_user_rematch]: 2.357e-05
        [meta_fg_expand]: 0.00163933
        [replace_old_param]: 5.766e-05
        [inline_without_move]: 5.452e-05
        [renormalize]: 0.00176599
        [add_forward_monad_depend]: 5.09e-06
        [auto_monad_grad]: 2.26e-06
        [auto_monad_eliminator]: 1.117e-05
        [cse]: 2.608e-05
        [replace_applicator]: 1.346e-05
    [Cycle 2]: 0.00030407, [27]
        [switch_simplify]: 4.44998e-06
        [loop_unroll]: 4.29002e-06
        [a_1]: 4.412e-05
        [with_stream_mark]: 1.026e-05
        [recompute_prepare]: 3.80998e-06
        [updatestate_depend_eliminate]: 2.57001e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 3.78999e-06
        [updatestate_useless_node_eliminater]: 3.3e-06
        [accelerated_algorithm]: 3.91001e-06
        [meta_shard_fg_expand]: 1.97001e-06
        [get_grad_eliminate_]: 3.78001e-06
        [merge_forward]: 3.18998e-06
        [cell_reuse_recompute_pass]: 2.68e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.214e-05
        [j_node_and_user_rematch]: 6.48e-06
        [meta_fg_expand]: 3.977e-05
        [replace_old_param]: 4.82e-06
        [inline_without_move]: 3.4e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.49e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 4.52e-06
        [cse]: 9.65002e-06
        [replace_applicator]: 3.91001e-06
[py_interpret_to_execute_after_opt_a]: 8.12e-06
[rewriter_after_opt_a]: 3.054e-05
[convert_after_rewriter]: 6.98998e-06
[order_py_execute_after_rewriter]: 4.78001e-06
[mutable_eliminate]: 0.00058936
[jit_opt_b]: 4.128e-05, [1]
    [Cycle 1]: 3.433e-05, [2]
        [frontend_op_eliminate]: 1.064e-05
        [inline_after_opt_a]: 1.158e-05
[cconv]: 2.185e-05
[loop_unroll]: 0.00042747
[jit_opt_after_cconv]: 0.00012034, [1]
    [Cycle 1]: 0.00011354, [11]
        [c_1]: 1.272e-05
        [parameter_eliminate]: 2.63998e-06
        [updatestate_depend_eliminate]: 3.96001e-06
        [updatestate_assign_eliminate]: 1.98002e-06
        [updatestate_loads_eliminate]: 1.64e-06
        [cse]: 1.611e-05
        [call_graph_tuple_transform]: 1.429e-05
        [tuple_list_get_item_eliminator]: 3.88001e-06
        [none_parameter_eliminate]: 1.92001e-06
        [renormalize]: 7.80012e-07
        [switch_simplify]: 3.90998e-06
[remove_dup_value]: 1.473e-05
[partial_unused_args_eliminate]: 2.54999e-06
[environ_conv]: 4.97e-06
[add_recomputation]: 3.114e-05
[cse_after_recomputation]: 1.958e-05, [1]
    [Cycle 1]: 1.44e-05, [1]
        [cse]: 8.32e-06
[auto_monad_reorder]: 1.259e-05
[get_jit_bprop_graph]: 1.97001e-06
[rewriter_after_jit_bprop_graph]: 4.85001e-06
[opt_after_jit_grad]: 0.00045923
[symbol_engine_optimizer]: 6.523e-05, [1]
    [Cycle 1]: 5.895e-05, [6]
        [build]: 2.93998e-06
        [elim_shapecalc]: 6.09001e-06
        [elim_not_effective]: 1.079e-05
        [opt_reshape]: 4.17998e-06
        [fold_const_symbol]: 6.11e-06
        [renormalize]: 5.19998e-07
[validate]: 2.795e-05
[backend_pass]: 8.99978e-07
[task_emit]: 1.956e-05
[execute]: 1.27e-06
Sums
    bootstrap                            :     0.000198s :  1.10%
    type_inference                       :     0.010840s : 59.89%
    event_method                         :     0.000051s :  0.28%
    auto_monad                           :     0.000099s :  0.55%
    graph_reusing                        :     0.000008s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.16%
    rewriter_before_opt_a                :     0.000095s :  0.52%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.36%
    jit_opt_a.loop_unroll                :     0.000043s :  0.24%
    jit_opt_a.a_1                        :     0.000978s :  5.40%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000027s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000018s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.09%
    jit_opt_a.merge_forward              :     0.000011s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000030s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001679s :  9.28%
    jit_opt_a.replace_old_param          :     0.000062s :  0.35%
    jit_opt_a.inline_without_move        :     0.000058s :  0.32%
    jit_opt_a.renormalize                :     0.001766s :  9.76%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.09%
    jit_opt_a.cse                        :     0.000036s :  0.20%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.04%
    rewriter_after_opt_a                 :     0.000031s :  0.17%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000589s :  3.26%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.06%
    cconv                                :     0.000022s :  0.12%
    loop_unroll                          :     0.000427s :  2.36%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000031s :  0.17%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000459s :  2.54%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000028s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000020s :  0.11%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000269    50
 0.54% :     0.000001s :      1: substitution.elim_not_effective
 0.42% :     0.000001s :      1: substitution.fold_const_symbol
 1.58% :     0.000004s :      1: substitution.graph_param_transform
67.17% :     0.000181s :      6: substitution.inline
 6.85% :     0.000018s :      2: substitution.inline_without_move
 2.65% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.51% :     0.000004s :      2: substitution.minmaximum_grad
 1.48% :     0.000004s :      3: substitution.partial_eliminate
 2.61% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.27% :     0.000006s :      2: substitution.replace_applicator
 1.79% :     0.000005s :      5: substitution.replace_old_param
 0.86% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.66% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.99% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.62% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010762     2
88.45% :     0.009519s :      1: type_inference.infer
11.55% :     0.001243s :      1: type_inference.specialize
------[replace.]   0.000071     8
79.56% :     0.000057s :      6: replace.inline
20.44% :     0.000015s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000187     8
94.58% :     0.000177s :      6: match.inline
 5.42% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000173  1109
 1.63% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.59% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.33% :     0.000002s :     18: predicate.addn_check_dump
 1.51% :     0.000003s :     18: predicate.addn_zero_filter
 2.26% :     0.000004s :     18: predicate.arithmetic_simplify
 1.43% :     0.000002s :     18: predicate.cast_eliminate
 0.10% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.26% :     0.000002s :     18: predicate.compare_switch_simplify
 1.71% :     0.000003s :     18: predicate.depend_value_elim
 1.25% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.46% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.50% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      1: predicate.elim_not_effective
 0.29% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.44% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.31% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.51% :     0.000003s :     18: predicate.environ_get_eliminate
 1.22% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.18% :     0.000002s :     12: predicate.get_grad_eliminate
 0.08% :     0.000000s :      1: predicate.graph_param_transform
 4.18% :     0.000007s :     28: predicate.inline
 3.23% :     0.000006s :     33: predicate.inline_without_move
 0.49% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.49% :     0.000003s :     12: predicate.less_batch_normalization
 1.58% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.72% :     0.000003s :     21: predicate.load_eliminater
 0.72% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.51% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.81% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.29% :     0.000002s :     18: predicate.merge_addn
 1.40% :     0.000002s :     18: predicate.minmaximum_grad
 0.72% :     0.000001s :      1: predicate.mutable_eliminate
 0.21% :     0.000000s :      1: predicate.opt_reshape
 2.04% :     0.000004s :     21: predicate.partial_eliminate
 1.52% :     0.000003s :     18: predicate.print_const_string_wrapper
 2.13% :     0.000004s :     18: predicate.reduce_eliminate
 1.57% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.69% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.50% :     0.000003s :     22: predicate.replace_applicator
 1.67% :     0.000003s :     33: predicate.replace_old_param
 0.21% :     0.000000s :      1: predicate.reset_defer_inline
 1.68% :     0.000003s :     18: predicate.reshape_eliminate
 1.83% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.40% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     18: predicate.same_eliminate
 0.65% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.36% :     0.000001s :      2: predicate.special_op_eliminate
 1.21% :     0.000002s :     12: predicate.specialize_transform
 1.83% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.48% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.13% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.69% :     0.000005s :     26: predicate.switch_defer_inline
 2.49% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.44% :     0.000011s :     71: predicate.switch_simplify
 1.50% :     0.000003s :     18: predicate.tile_eliminate
 1.40% :     0.000002s :     18: predicate.transpose_eliminate
 1.53% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.50% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.87% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.53% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.89% :     0.000003s :     18: predicate.value_based_eliminate
 0.14% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.22% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001249    22
72.13% :     0.000901s :     13: func_graph_cloner_run.FuncGraphClonerGraph
27.87% :     0.000348s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025558    76
 0.13% :     0.000034s :      1: add_recomputation
 0.41% :     0.000104s :      1: auto_monad
 0.06% :     0.000015s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.83% :     0.000211s :      1: bootstrap
 0.10% :     0.000024s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.22% :     0.000055s :      1: event_method
 0.01% :     0.000004s :      1: execute
 0.03% :     0.000007s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
35.04% :     0.008956s :      1: jit_opt_a
 0.48% :     0.000123s :      1: jit_opt_after_cconv
 0.17% :     0.000044s :      1: jit_opt_b
 1.70% :     0.000434s :      1: loop_unroll
 2.34% :     0.000597s :      1: mutable_eliminate
 5.30% :     0.001356s :     26: opt.transform.jit_opt_a
 0.12% :     0.000031s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000024s :      4: opt.transform.symbol_engine_opt
 1.83% :     0.000467s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000032s :      1: py_interpret_to_execute
 0.04% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 4.93% :     0.001260s :      1: renormalize.infer
 1.94% :     0.000495s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000034s :      1: rewriter_after_opt_a
 0.38% :     0.000098s :      1: rewriter_before_opt_a
 0.26% :     0.000068s :      1: symbol_engine_optimizer
 0.09% :     0.000024s :      1: task_emit
42.47% :     0.010855s :      1: type_inference
 0.17% :     0.000043s :      1: validate


TotalTime = 0.0105152, [30]
[bootstrap]: 0.00024421
[type_inference]: 0.00371861
[event_method]: 1.534e-05
[auto_monad]: 5.186e-05
[graph_reusing]: 5.49e-06
[pre_auto_parallel]: 2.06e-06
[py_interpret_to_execute]: 2.562e-05
[rewriter_before_opt_a]: 6.592e-05
[expand_dump_flag]: 3.11999e-06
[jit_opt_a]: 0.00447358, [2]
    [Cycle 1]: 0.001278, [27]
        [switch_simplify]: 4.596e-05
        [loop_unroll]: 2.795e-05
        [a_1]: 0.00046932
        [with_stream_mark]: 1.331e-05
        [recompute_prepare]: 6.97002e-06
        [updatestate_depend_eliminate]: 3.03e-06
        [updatestate_assign_eliminate]: 2.33998e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 1.59e-06
        [specialize_transform]: 5.61e-06
        [updatestate_useless_node_eliminater]: 4.72e-06
        [accelerated_algorithm]: 5.12999e-06
        [meta_shard_fg_expand]: 1.77001e-06
        [get_grad_eliminate_]: 4.80001e-06
        [merge_forward]: 3.18e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.305e-05
        [j_node_and_user_rematch]: 8.23001e-06
        [meta_fg_expand]: 2.36998e-06
        [replace_old_param]: 8.26002e-06
        [inline_without_move]: 5.10999e-06
        [renormalize]: 0.00043297
        [add_forward_monad_depend]: 4.87998e-06
        [auto_monad_grad]: 1.30999e-06
        [auto_monad_eliminator]: 1.11e-05
        [cse]: 2.337e-05
        [replace_applicator]: 1.067e-05
    [Cycle 2]: 0.00030832, [27]
        [switch_simplify]: 5.57001e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 8.772e-05
        [with_stream_mark]: 8.19998e-06
        [recompute_prepare]: 4.90001e-06
        [updatestate_depend_eliminate]: 2.34999e-06
        [updatestate_assign_eliminate]: 1.87001e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 4.77e-06
        [updatestate_useless_node_eliminater]: 4.53001e-06
        [accelerated_algorithm]: 4.91002e-06
        [meta_shard_fg_expand]: 1.05001e-06
        [get_grad_eliminate_]: 4.40999e-06
        [merge_forward]: 2.17999e-06
        [cell_reuse_recompute_pass]: 1.44e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.136e-05
        [j_node_and_user_rematch]: 7.51999e-06
        [meta_fg_expand]: 1.59998e-06
        [replace_old_param]: 7.24001e-06
        [inline_without_move]: 4.75999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 4.70001e-06
        [cse]: 9.35001e-06
        [replace_applicator]: 5.08002e-06
[py_interpret_to_execute_after_opt_a]: 7.95e-06
[rewriter_after_opt_a]: 2.901e-05
[convert_after_rewriter]: 5.67999e-06
[order_py_execute_after_rewriter]: 4.62e-06
[mutable_eliminate]: 0.00047015
[jit_opt_b]: 4.76e-05, [1]
    [Cycle 1]: 4.107e-05, [2]
        [frontend_op_eliminate]: 1.546e-05
        [inline_after_opt_a]: 1.437e-05
[cconv]: 2.185e-05
[loop_unroll]: 0.0004084
[jit_opt_after_cconv]: 0.00013143, [1]
    [Cycle 1]: 0.00012517, [11]
        [c_1]: 2.041e-05
        [parameter_eliminate]: 2.32999e-06
        [updatestate_depend_eliminate]: 5.03002e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.539e-05
        [call_graph_tuple_transform]: 1.752e-05
        [tuple_list_get_item_eliminator]: 5.13002e-06
        [none_parameter_eliminate]: 9.39996e-07
        [renormalize]: 4.60015e-07
        [switch_simplify]: 5.49e-06
[remove_dup_value]: 1.162e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 4.42998e-06
[add_recomputation]: 4.052e-05
[cse_after_recomputation]: 1.893e-05, [1]
    [Cycle 1]: 1.379e-05, [1]
        [cse]: 8.47e-06
[auto_monad_reorder]: 1.202e-05
[get_jit_bprop_graph]: 1.35001e-06
[rewriter_after_jit_bprop_graph]: 4.57e-06
[opt_after_jit_grad]: 0.0004421
[symbol_engine_optimizer]: 7.003e-05, [1]
    [Cycle 1]: 6.381e-05, [6]
        [build]: 3.28e-06
        [elim_shapecalc]: 7.26999e-06
        [elim_not_effective]: 1.216e-05
        [opt_reshape]: 5.51998e-06
        [fold_const_symbol]: 8.40001e-06
        [renormalize]: 4.09986e-07
[validate]: 2.743e-05
Sums
    bootstrap                            :     0.000244s :  3.44%
    type_inference                       :     0.003719s : 52.42%
    event_method                         :     0.000015s :  0.22%
    auto_monad                           :     0.000052s :  0.73%
    graph_reusing                        :     0.000005s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000026s :  0.36%
    rewriter_before_opt_a                :     0.000066s :  0.93%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000052s :  0.73%
    jit_opt_a.loop_unroll                :     0.000033s :  0.46%
    jit_opt_a.a_1                        :     0.000557s :  7.85%
    jit_opt_a.with_stream_mark           :     0.000022s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000010s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.13%
    jit_opt_a.merge_forward              :     0.000005s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.06%
    jit_opt_a.replace_old_param          :     0.000016s :  0.22%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000433s :  6.10%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.22%
    jit_opt_a.cse                        :     0.000033s :  0.46%
    jit_opt_a.replace_applicator         :     0.000016s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000029s :  0.41%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000470s :  6.63%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.20%
    cconv                                :     0.000022s :  0.31%
    loop_unroll                          :     0.000408s :  5.76%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.29%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000015s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.25%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.08%
    remove_dup_value                     :     0.000012s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000041s :  0.57%
    cse_after_recomputation.cse          :     0.000008s :  0.12%
    auto_monad_reorder                   :     0.000012s :  0.17%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000442s :  6.23%
    symbol_engine_optimizer.build        :     0.000003s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000027s :  0.39%


Time group info:
------[substitution.]   0.000162    23
 1.30% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.75% :     0.000004s :      3: substitution.graph_param_transform
82.68% :     0.000134s :      4: substitution.inline
 1.75% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.47% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.16% :     0.000003s :      2: substitution.replace_old_param
 6.18% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003665     2
80.41% :     0.002947s :      1: type_inference.infer
19.59% :     0.000718s :      1: type_inference.specialize
------[replace.]   0.000058     6
67.42% :     0.000039s :      4: replace.inline
32.58% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000141     6
93.64% :     0.000132s :      4: match.inline
 6.36% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000108   691
 1.66% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.13% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000001s :     10: predicate.addn_zero_filter
 1.74% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.12% :     0.000001s :     10: predicate.compare_switch_simplify
 1.36% :     0.000001s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.44% :     0.000000s :      3: predicate.graph_param_transform
 5.34% :     0.000006s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      6: predicate.less_batch_normalization
 1.65% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.82% :     0.000002s :     15: predicate.load_eliminater
 1.55% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.79% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.14% :     0.000001s :     10: predicate.minmaximum_grad
 1.50% :     0.000002s :      3: predicate.mutable_eliminate
 0.60% :     0.000001s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.75% :     0.000002s :     18: predicate.replace_applicator
 0.78% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.56% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.66% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.46% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.18% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.85% :     0.000009s :     53: predicate.switch_simplify
 1.35% :     0.000001s :     10: predicate.tile_eliminate
 1.27% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.63% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.78% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000473    12
49.09% :     0.000232s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.23% :     0.000025s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.68% :     0.000216s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011764    72
 0.37% :     0.000043s :      1: add_recomputation
 0.47% :     0.000055s :      1: auto_monad
 0.12% :     0.000015s :      1: auto_monad_reorder
 2.19% :     0.000258s :      1: bootstrap
 0.21% :     0.000024s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.18% :     0.000021s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000019s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000003s :      1: get_jit_bprop_graph
 0.07% :     0.000008s :      1: graph_reusing
38.05% :     0.004477s :      1: jit_opt_a
 1.14% :     0.000134s :      1: jit_opt_after_cconv
 0.43% :     0.000050s :      1: jit_opt_b
 3.54% :     0.000416s :      1: loop_unroll
 4.07% :     0.000479s :      1: mutable_eliminate
 6.37% :     0.000749s :     26: opt.transform.jit_opt_a
 0.38% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000024s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.82% :     0.000450s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000028s :      1: py_interpret_to_execute
 0.09% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000014s :      1: remove_dup_value
 1.78% :     0.000210s :      1: renormalize.infer
 1.85% :     0.000217s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000032s :      1: rewriter_after_opt_a
 0.58% :     0.000069s :      1: rewriter_before_opt_a
 0.62% :     0.000073s :      1: symbol_engine_optimizer
31.70% :     0.003729s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.389.998 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c36b4c0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.390.621 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0170118, [33]
[bootstrap]: 0.00025126
[type_inference]: 0.0039617
[event_method]: 1.58e-05
[auto_monad]: 4.453e-05
[graph_reusing]: 5.04998e-06
[pre_auto_parallel]: 2.16e-06
[py_interpret_to_execute]: 2.844e-05
[rewriter_before_opt_a]: 6.333e-05
[expand_dump_flag]: 2.69001e-06
[jit_opt_a]: 0.00467682, [2]
    [Cycle 1]: 0.00133763, [27]
        [switch_simplify]: 4.354e-05
        [loop_unroll]: 2.892e-05
        [a_1]: 0.00046191
        [with_stream_mark]: 1.2e-05
        [recompute_prepare]: 7.37997e-06
        [updatestate_depend_eliminate]: 2.98e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 5.53002e-06
        [updatestate_useless_node_eliminater]: 4.92999e-06
        [accelerated_algorithm]: 5.29998e-06
        [meta_shard_fg_expand]: 1.19e-06
        [get_grad_eliminate_]: 5.21998e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 9.30013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.215e-05
        [j_node_and_user_rematch]: 8.55999e-06
        [meta_fg_expand]: 1.69998e-06
        [replace_old_param]: 7.48e-06
        [inline_without_move]: 5.31998e-06
        [renormalize]: 0.00051072
        [add_forward_monad_depend]: 3.75e-06
        [auto_monad_grad]: 1.79e-06
        [auto_monad_eliminator]: 1.091e-05
        [cse]: 1.711e-05
        [replace_applicator]: 1.122e-05
    [Cycle 2]: 0.00030699, [27]
        [switch_simplify]: 5.54e-06
        [loop_unroll]: 5.03002e-06
        [a_1]: 8.695e-05
        [with_stream_mark]: 7.76001e-06
        [recompute_prepare]: 4.72998e-06
        [updatestate_depend_eliminate]: 2.37999e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.98002e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.89e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 4.77e-06
        [meta_shard_fg_expand]: 1.00999e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.24001e-06
        [cell_reuse_recompute_pass]: 1.68002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.14e-05
        [j_node_and_user_rematch]: 7.92e-06
        [meta_fg_expand]: 1.45999e-06
        [replace_old_param]: 7.03e-06
        [inline_without_move]: 4.44002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.05999e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 4.82e-06
        [cse]: 9.57001e-06
        [replace_applicator]: 4.77998e-06
[py_interpret_to_execute_after_opt_a]: 8.15e-06
[rewriter_after_opt_a]: 2.808e-05
[convert_after_rewriter]: 6.11e-06
[order_py_execute_after_rewriter]: 4.27e-06
[mutable_eliminate]: 0.00048173
[jit_opt_b]: 4.727e-05, [1]
    [Cycle 1]: 4.095e-05, [2]
        [frontend_op_eliminate]: 1.533e-05
        [inline_after_opt_a]: 1.43e-05
[cconv]: 1.634e-05
[loop_unroll]: 0.00046012
[jit_opt_after_cconv]: 0.00013528, [1]
    [Cycle 1]: 0.00012916, [11]
        [c_1]: 2.051e-05
        [parameter_eliminate]: 2.37999e-06
        [updatestate_depend_eliminate]: 5.12e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 1.91003e-06
        [cse]: 1.919e-05
        [call_graph_tuple_transform]: 1.619e-05
        [tuple_list_get_item_eliminator]: 5.19e-06
        [none_parameter_eliminate]: 1.07e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 4.92999e-06
[remove_dup_value]: 1.021e-05
[partial_unused_args_eliminate]: 1.94999e-06
[environ_conv]: 4.05e-06
[add_recomputation]: 3.902e-05
[cse_after_recomputation]: 2.077e-05, [1]
    [Cycle 1]: 1.526e-05, [1]
        [cse]: 9.28002e-06
[auto_monad_reorder]: 1.29e-05
[get_jit_bprop_graph]: 1.55001e-06
[rewriter_after_jit_bprop_graph]: 4.33001e-06
[opt_after_jit_grad]: 0.00046588
[symbol_engine_optimizer]: 7.06e-05, [1]
    [Cycle 1]: 6.393e-05, [6]
        [build]: 3.23e-06
        [elim_shapecalc]: 7.7e-06
        [elim_not_effective]: 1.137e-05
        [opt_reshape]: 5.24e-06
        [fold_const_symbol]: 8.16002e-06
        [renormalize]: 5.3001e-07
[validate]: 2.907e-05
[backend_pass]: 6.89994e-07
[task_emit]: 0.00593041
[execute]: 7.7e-06
Sums
    bootstrap                            :     0.000251s :  1.87%
    type_inference                       :     0.003962s : 29.53%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000045s :  0.33%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.21%
    rewriter_before_opt_a                :     0.000063s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000049s :  0.37%
    jit_opt_a.loop_unroll                :     0.000034s :  0.25%
    jit_opt_a.a_1                        :     0.000549s :  4.09%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000511s :  3.81%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.12%
    jit_opt_a.cse                        :     0.000027s :  0.20%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000028s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.05%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000482s :  3.59%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.11%
    cconv                                :     0.000016s :  0.12%
    loop_unroll                          :     0.000460s :  3.43%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000010s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000039s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000013s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000466s :  3.47%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000029s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005930s : 44.21%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000151    23
 1.19% :     0.000002s :      2: substitution.elim_not_effective
 0.87% :     0.000001s :      2: substitution.fold_const_symbol
 2.97% :     0.000004s :      3: substitution.graph_param_transform
82.97% :     0.000125s :      4: substitution.inline
 1.94% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.52% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.07% :     0.000003s :      2: substitution.replace_old_param
 5.45% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003909     2
81.56% :     0.003188s :      1: type_inference.infer
18.44% :     0.000721s :      1: type_inference.specialize
------[replace.]   0.000058     6
66.43% :     0.000039s :      4: replace.inline
33.57% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000130     6
94.45% :     0.000123s :      4: match.inline
 5.55% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000108   691
 1.36% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.08% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000001s :     10: predicate.addn_zero_filter
 2.27% :     0.000002s :     10: predicate.arithmetic_simplify
 1.33% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.11% :     0.000001s :     10: predicate.compare_switch_simplify
 1.23% :     0.000001s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 1.00% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.03% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.09% :     0.000006s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.42% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.33% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.26% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.43% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000002s :     15: predicate.partial_eliminate
 1.49% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.55% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000001s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.56% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 1.13% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.54% :     0.000008s :     53: predicate.switch_simplify
 1.27% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.33% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.84% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.80% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000481    12
48.21% :     0.000232s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.37% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.42% :     0.000223s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018366    76
 0.23% :     0.000042s :      1: add_recomputation
 0.26% :     0.000048s :      1: auto_monad
 0.08% :     0.000015s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.42% :     0.000262s :      1: bootstrap
 0.10% :     0.000019s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
25.48% :     0.004680s :      1: jit_opt_a
 0.75% :     0.000138s :      1: jit_opt_after_cconv
 0.27% :     0.000050s :      1: jit_opt_b
 2.55% :     0.000468s :      1: loop_unroll
 2.67% :     0.000490s :      1: mutable_eliminate
 4.03% :     0.000741s :     26: opt.transform.jit_opt_a
 0.24% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.58% :     0.000473s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000031s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000013s :      1: remove_dup_value
 1.38% :     0.000254s :      1: renormalize.infer
 1.36% :     0.000251s :      1: renormalize.specialize
 0.04% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000031s :      1: rewriter_after_opt_a
 0.36% :     0.000066s :      1: rewriter_before_opt_a
 0.40% :     0.000073s :      1: symbol_engine_optimizer
32.34% :     0.005940s :      1: task_emit
21.63% :     0.003972s :      1: type_inference
 0.26% :     0.000047s :      1: validate


TotalTime = 0.0320914, [33]
[bootstrap]: 0.00020718
[type_inference]: 0.0101976
[event_method]: 4.349e-05
[auto_monad]: 8.53e-05
[graph_reusing]: 6.69001e-06
[pre_auto_parallel]: 2.94999e-06
[py_interpret_to_execute]: 2.3e-05
[rewriter_before_opt_a]: 8.614e-05
[expand_dump_flag]: 3.14001e-06
[jit_opt_a]: 0.0114498, [3]
    [Cycle 1]: 0.00547179, [27]
        [switch_simplify]: 5.886e-05
        [loop_unroll]: 4.633e-05
        [a_1]: 0.0008151
        [with_stream_mark]: 2.362e-05
        [recompute_prepare]: 1.733e-05
        [updatestate_depend_eliminate]: 7.63999e-06
        [updatestate_assign_eliminate]: 5.66e-06
        [updatestate_loads_eliminate]: 5.81e-06
        [parameter_eliminate]: 2.16e-06
        [specialize_transform]: 1.271e-05
        [updatestate_useless_node_eliminater]: 1.165e-05
        [accelerated_algorithm]: 1.243e-05
        [meta_shard_fg_expand]: 2.37999e-06
        [get_grad_eliminate_]: 1.219e-05
        [merge_forward]: 7.6e-06
        [cell_reuse_recompute_pass]: 1.10001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.388e-05
        [j_node_and_user_rematch]: 2.209e-05
        [meta_fg_expand]: 0.00135297
        [replace_old_param]: 4.868e-05
        [inline_without_move]: 4.987e-05
        [renormalize]: 0.00244228
        [add_forward_monad_depend]: 8.86002e-06
        [auto_monad_grad]: 5.34998e-06
        [auto_monad_eliminator]: 4.871e-05
        [cse]: 0.00015614
        [replace_applicator]: 6.961e-05
    [Cycle 2]: 0.00212812, [27]
        [switch_simplify]: 3.71e-05
        [loop_unroll]: 3.621e-05
        [a_1]: 0.00112415
        [with_stream_mark]: 1.462e-05
        [recompute_prepare]: 8.89998e-06
        [updatestate_depend_eliminate]: 4.09002e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 3.54002e-06
        [parameter_eliminate]: 1.61002e-06
        [specialize_transform]: 7.23999e-06
        [updatestate_useless_node_eliminater]: 6.74001e-06
        [accelerated_algorithm]: 7.73999e-06
        [meta_shard_fg_expand]: 2.02999e-06
        [get_grad_eliminate_]: 7.26999e-06
        [merge_forward]: 4.50999e-06
        [cell_reuse_recompute_pass]: 1.07998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.626e-05
        [j_node_and_user_rematch]: 1.166e-05
        [meta_fg_expand]: 4.157e-05
        [replace_old_param]: 1.062e-05
        [inline_without_move]: 7.13e-06
        [renormalize]: 0.00056013
        [add_forward_monad_depend]: 5.29e-06
        [auto_monad_grad]: 1.91e-06
        [auto_monad_eliminator]: 1.403e-05
        [cse]: 2.504e-05
        [replace_applicator]: 1.468e-05
    [Cycle 3]: 0.00043408, [27]
        [switch_simplify]: 7.33e-06
        [loop_unroll]: 6.94999e-06
        [a_1]: 0.00016553
        [with_stream_mark]: 1.076e-05
        [recompute_prepare]: 7.28e-06
        [updatestate_depend_eliminate]: 3.9e-06
        [updatestate_assign_eliminate]: 3.38999e-06
        [updatestate_loads_eliminate]: 3.03e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 6.74999e-06
        [updatestate_useless_node_eliminater]: 6.83998e-06
        [accelerated_algorithm]: 6.89999e-06
        [meta_shard_fg_expand]: 1.50999e-06
        [get_grad_eliminate_]: 6.58e-06
        [merge_forward]: 3.56001e-06
        [cell_reuse_recompute_pass]: 1.82001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.662e-05
        [j_node_and_user_rematch]: 1.135e-05
        [meta_fg_expand]: 2.40002e-06
        [replace_old_param]: 8.82e-06
        [inline_without_move]: 6.45002e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.34998e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 8.45999e-06
        [cse]: 1.685e-05
        [replace_applicator]: 6.93998e-06
[py_interpret_to_execute_after_opt_a]: 1.351e-05
[rewriter_after_opt_a]: 4.232e-05
[convert_after_rewriter]: 9.07001e-06
[order_py_execute_after_rewriter]: 5.59998e-06
[mutable_eliminate]: 0.0006752
[jit_opt_b]: 5.895e-05, [1]
    [Cycle 1]: 5.147e-05, [2]
        [frontend_op_eliminate]: 2.169e-05
        [inline_after_opt_a]: 1.872e-05
[cconv]: 2.274e-05
[loop_unroll]: 0.00046464
[jit_opt_after_cconv]: 0.00016525, [1]
    [Cycle 1]: 0.00015881, [11]
        [c_1]: 2.978e-05
        [parameter_eliminate]: 2.69001e-06
        [updatestate_depend_eliminate]: 6.52001e-06
        [updatestate_assign_eliminate]: 3.23998e-06
        [updatestate_loads_eliminate]: 3.66999e-06
        [cse]: 2.545e-05
        [call_graph_tuple_transform]: 2.323e-05
        [tuple_list_get_item_eliminator]: 7.58999e-06
        [none_parameter_eliminate]: 1.48002e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 7.06001e-06
[remove_dup_value]: 1.628e-05
[partial_unused_args_eliminate]: 2.05002e-06
[environ_conv]: 5.98998e-06
[add_recomputation]: 4.894e-05
[cse_after_recomputation]: 5.539e-05, [1]
    [Cycle 1]: 4.934e-05, [1]
        [cse]: 4.221e-05
[auto_monad_reorder]: 1.961e-05
[get_jit_bprop_graph]: 1.60999e-06
[rewriter_after_jit_bprop_graph]: 5.57001e-06
[opt_after_jit_grad]: 0.00045564
[symbol_engine_optimizer]: 8.396e-05, [1]
    [Cycle 1]: 7.817e-05, [6]
        [build]: 3.47002e-06
        [elim_shapecalc]: 9.43997e-06
        [elim_not_effective]: 1.691e-05
        [opt_reshape]: 7.88001e-06
        [fold_const_symbol]: 1.236e-05
        [renormalize]: 5.69999e-07
[validate]: 3.904e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.00758375
[execute]: 8.3e-06
Sums
    bootstrap                            :     0.000207s :  0.74%
    type_inference                       :     0.010198s : 36.56%
    event_method                         :     0.000043s :  0.16%
    auto_monad                           :     0.000085s :  0.31%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000023s :  0.08%
    rewriter_before_opt_a                :     0.000086s :  0.31%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000103s :  0.37%
    jit_opt_a.loop_unroll                :     0.000089s :  0.32%
    jit_opt_a.a_1                        :     0.002105s :  7.55%
    jit_opt_a.with_stream_mark           :     0.000049s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000034s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000027s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000027s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000026s :  0.09%
    jit_opt_a.merge_forward              :     0.000016s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000057s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000045s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001397s :  5.01%
    jit_opt_a.replace_old_param          :     0.000068s :  0.24%
    jit_opt_a.inline_without_move        :     0.000063s :  0.23%
    jit_opt_a.renormalize                :     0.003002s : 10.76%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000071s :  0.26%
    jit_opt_a.cse                        :     0.000198s :  0.71%
    jit_opt_a.replace_applicator         :     0.000091s :  0.33%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.05%
    rewriter_after_opt_a                 :     0.000042s :  0.15%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000675s :  2.42%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.07%
    cconv                                :     0.000023s :  0.08%
    loop_unroll                          :     0.000465s :  1.67%
    jit_opt_after_cconv.c_1              :     0.000030s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000049s :  0.18%
    cse_after_recomputation.cse          :     0.000042s :  0.15%
    auto_monad_reorder                   :     0.000020s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000456s :  1.63%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007584s : 27.19%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000508   118
 5.40% :     0.000027s :      1: substitution.arithmetic_simplify
 0.56% :     0.000003s :      4: substitution.elim_not_effective
 0.38% :     0.000002s :      4: substitution.fold_const_symbol
 1.36% :     0.000007s :      5: substitution.graph_param_transform
59.81% :     0.000304s :     12: substitution.inline
 3.17% :     0.000016s :      2: substitution.inline_without_move
 1.85% :     0.000009s :     16: substitution.j_node_and_user_rematch
 2.11% :     0.000011s :      7: substitution.minmaximum_grad
 1.29% :     0.000007s :      5: substitution.partial_eliminate
 2.18% :     0.000011s :     16: substitution.remove_not_recompute_node
 5.33% :     0.000027s :     10: substitution.replace_applicator
 1.41% :     0.000007s :      7: substitution.replace_old_param
 0.47% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.12% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.85% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.71% :     0.000039s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010132     2
87.89% :     0.008905s :      1: type_inference.infer
12.11% :     0.001227s :      1: type_inference.specialize
------[replace.]   0.000146    20
 3.92% :     0.000006s :      1: replace.arithmetic_simplify
62.25% :     0.000091s :     12: replace.inline
33.83% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000343    20
 7.71% :     0.000026s :      1: match.arithmetic_simplify
86.65% :     0.000297s :     12: match.inline
 5.64% :     0.000019s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000395  2790
 1.54% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.45% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.40% :     0.000006s :     46: predicate.addn_check_dump
 1.76% :     0.000007s :     46: predicate.addn_zero_filter
 1.91% :     0.000008s :     47: predicate.arithmetic_simplify
 1.62% :     0.000006s :     47: predicate.cast_eliminate
 0.25% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.39% :     0.000005s :     46: predicate.compare_switch_simplify
 1.72% :     0.000007s :     46: predicate.depend_value_elim
 1.44% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.56% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.51% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.33% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.17% :     0.000001s :      5: predicate.elim_not_effective
 0.23% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.56% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.39% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.41% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.51% :     0.000006s :     47: predicate.environ_get_eliminate
 1.42% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.11% :     0.000000s :      5: predicate.fold_const_symbol
 0.98% :     0.000004s :     21: predicate.get_grad_eliminate
 0.11% :     0.000000s :      5: predicate.graph_param_transform
 4.11% :     0.000016s :     76: predicate.inline
 1.70% :     0.000007s :     42: predicate.inline_without_move
 0.37% :     0.000001s :     21: predicate.j_node_and_user_rematch
 1.13% :     0.000004s :     21: predicate.less_batch_normalization
 1.79% :     0.000007s :     54: predicate.list_to_tuple_eliminator_
 1.97% :     0.000008s :     59: predicate.load_eliminater
 0.56% :     0.000002s :      5: predicate.loop_unroll_after_grad
 5.10% :     0.000020s :     89: predicate.loop_unroll_before_grad
 1.82% :     0.000007s :     52: predicate.make_slice_get_slice_eliminator
 1.39% :     0.000006s :     46: predicate.merge_addn
 1.47% :     0.000006s :     47: predicate.minmaximum_grad
 0.64% :     0.000003s :      5: predicate.mutable_eliminate
 0.24% :     0.000001s :      5: predicate.opt_reshape
 2.26% :     0.000009s :     59: predicate.partial_eliminate
 1.62% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.82% :     0.000007s :     47: predicate.reduce_eliminate
 1.74% :     0.000007s :     54: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.46% :     0.000010s :    105: predicate.replace_applicator
 0.85% :     0.000003s :     42: predicate.replace_old_param
 0.12% :     0.000000s :      5: predicate.reset_defer_inline
 1.60% :     0.000006s :     47: predicate.reshape_eliminate
 1.48% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.33% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.43% :     0.000006s :     46: predicate.same_eliminate
 0.46% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.41% :     0.000002s :     10: predicate.special_op_eliminate
 0.88% :     0.000003s :     21: predicate.specialize_transform
 1.70% :     0.000007s :     46: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.20% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.58% :     0.000010s :     66: predicate.switch_defer_inline
 2.38% :     0.000009s :     66: predicate.switch_layer_defer_inline
 5.91% :     0.000023s :    160: predicate.switch_simplify
 1.60% :     0.000006s :     47: predicate.tile_eliminate
 1.47% :     0.000006s :     47: predicate.transpose_eliminate
 1.80% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.57% :     0.000006s :     47: predicate.tuple_list_get_item_depend_reorder
 2.93% :     0.000012s :     64: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000007s :     54: predicate.tuple_to_list_eliminator_
 1.91% :     0.000008s :     59: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000011s :     80: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000007s :     46: predicate.value_based_eliminate
 0.15% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.27% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001519    29
54.41% :     0.000827s :     13: func_graph_cloner_run.FuncGraphClonerGraph
45.59% :     0.000693s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.037963    91
 0.15% :     0.000058s :      1: add_recomputation
 0.24% :     0.000089s :      1: auto_monad
 0.06% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.58% :     0.000221s :      1: bootstrap
 0.07% :     0.000026s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.15% :     0.000058s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.13% :     0.000048s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
30.17% :     0.011453s :      1: jit_opt_a
 0.44% :     0.000168s :      1: jit_opt_after_cconv
 0.16% :     0.000062s :      1: jit_opt_b
 1.25% :     0.000473s :      1: loop_unroll
 1.80% :     0.000684s :      1: mutable_eliminate
 7.16% :     0.002720s :     39: opt.transform.jit_opt_a
 0.17% :     0.000064s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000035s :      4: opt.transform.jit_opt_b
 0.04% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000043s :      4: opt.transform.symbol_engine_opt
 1.22% :     0.000464s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000026s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 4.33% :     0.001645s :      2: renormalize.infer
 3.54% :     0.001344s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000045s :      1: rewriter_after_opt_a
 0.23% :     0.000089s :      1: rewriter_before_opt_a
 0.23% :     0.000086s :      1: symbol_engine_optimizer
20.01% :     0.007596s :      1: task_emit
26.89% :     0.010210s :      1: type_inference
 0.16% :     0.000061s :      1: validate

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.406.645 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80efaf0> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.407.223 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0527308, [30]
[bootstrap]: 0.00062571
[type_inference]: 0.0330991
[event_method]: 4.65e-05
[auto_monad]: 0.0001564
[graph_reusing]: 6.84999e-06
[pre_auto_parallel]: 1.1e-05
[py_interpret_to_execute]: 4.895e-05
[rewriter_before_opt_a]: 0.00012872
[expand_dump_flag]: 3.40998e-06
[jit_opt_a]: 0.0162799, [2]
    [Cycle 1]: 0.00521048, [27]
        [switch_simplify]: 9.947e-05
        [loop_unroll]: 3.817e-05
        [a_1]: 0.00093528
        [with_stream_mark]: 2.484e-05
        [recompute_prepare]: 1.848e-05
        [updatestate_depend_eliminate]: 8.1e-06
        [updatestate_assign_eliminate]: 5.79999e-06
        [updatestate_loads_eliminate]: 5.59998e-06
        [parameter_eliminate]: 2.05002e-06
        [specialize_transform]: 1.449e-05
        [updatestate_useless_node_eliminater]: 1.24e-05
        [accelerated_algorithm]: 1.256e-05
        [meta_shard_fg_expand]: 1.161e-05
        [get_grad_eliminate_]: 1.244e-05
        [merge_forward]: 8.28999e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.682e-05
        [j_node_and_user_rematch]: 3.37e-05
        [meta_fg_expand]: 0.00193461
        [replace_old_param]: 5.302e-05
        [inline_without_move]: 5.226e-05
        [renormalize]: 0.0016215
        [add_forward_monad_depend]: 2.337e-05
        [auto_monad_grad]: 2.55002e-06
        [auto_monad_eliminator]: 1.163e-05
        [cse]: 2.366e-05
        [replace_applicator]: 1.063e-05
    [Cycle 2]: 0.00027531, [27]
        [switch_simplify]: 4.75999e-06
        [loop_unroll]: 3.66001e-06
        [a_1]: 4.034e-05
        [with_stream_mark]: 8.99e-06
        [recompute_prepare]: 3.67002e-06
        [updatestate_depend_eliminate]: 2.48e-06
        [updatestate_assign_eliminate]: 1.66e-06
        [updatestate_loads_eliminate]: 1.67999e-06
        [parameter_eliminate]: 1.56998e-06
        [specialize_transform]: 3.76999e-06
        [updatestate_useless_node_eliminater]: 3.55003e-06
        [accelerated_algorithm]: 3.97e-06
        [meta_shard_fg_expand]: 1.58002e-06
        [get_grad_eliminate_]: 3.51999e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 2.04999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.036e-05
        [j_node_and_user_rematch]: 6.31e-06
        [meta_fg_expand]: 2.58e-05
        [replace_old_param]: 3.73001e-06
        [inline_without_move]: 3.26999e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.59998e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 3.71999e-06
        [cse]: 7.97e-06
        [replace_applicator]: 4e-06
[py_interpret_to_execute_after_opt_a]: 1.011e-05
[rewriter_after_opt_a]: 4.051e-05
[convert_after_rewriter]: 6.49001e-06
[order_py_execute_after_rewriter]: 4.59002e-06
[mutable_eliminate]: 0.00054966
[jit_opt_b]: 4.039e-05, [1]
    [Cycle 1]: 3.294e-05, [2]
        [frontend_op_eliminate]: 1.034e-05
        [inline_after_opt_a]: 1.131e-05
[cconv]: 2.068e-05
[loop_unroll]: 0.00041432
[jit_opt_after_cconv]: 0.00011994, [1]
    [Cycle 1]: 0.00011305, [11]
        [c_1]: 1.251e-05
        [parameter_eliminate]: 2.40002e-06
        [updatestate_depend_eliminate]: 4.67e-06
        [updatestate_assign_eliminate]: 1.73002e-06
        [updatestate_loads_eliminate]: 1.55999e-06
        [cse]: 1.543e-05
        [call_graph_tuple_transform]: 1.389e-05
        [tuple_list_get_item_eliminator]: 3.93001e-06
        [none_parameter_eliminate]: 1.47999e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 4.06001e-06
[remove_dup_value]: 1.343e-05
[partial_unused_args_eliminate]: 2.32001e-06
[environ_conv]: 8.95999e-06
[add_recomputation]: 3.857e-05
[cse_after_recomputation]: 2.059e-05, [1]
    [Cycle 1]: 1.481e-05, [1]
        [cse]: 8.57e-06
[auto_monad_reorder]: 1.922e-05
[get_jit_bprop_graph]: 1.93002e-06
[rewriter_after_jit_bprop_graph]: 0.0001793
[opt_after_jit_grad]: 0.00047005
[symbol_engine_optimizer]: 6.311e-05, [1]
    [Cycle 1]: 5.696e-05, [6]
        [build]: 3.14001e-06
        [elim_shapecalc]: 5.57999e-06
     
TotalTime = 0.0215035, [33]
[bootstrap]: 0.00024345
[type_inference]: 0.0106182
[event_method]: 4.833e-05
[auto_monad]: 9.44e-05
[graph_reusing]: 8.17e-06
[pre_auto_parallel]: 3.26001e-06
[py_interpret_to_execute]: 2.943e-05
[rewriter_before_opt_a]: 9.447e-05
[expand_dump_flag]: 2.82002e-06
[jit_opt_a]: 0.00839434, [2]
    [Cycle 1]: 0.00461635, [27]
        [switch_simplify]: 6.102e-05
        [loop_unroll]: 3.869e-05
        [a_1]: 0.0008319
        [with_stream_mark]: 2.227e-05
        [recompute_prepare]: 1.757e-05
        [updatestate_depend_eliminate]: 7.98999e-06
        [updatestate_assign_eliminate]: 5.84999e-06
        [updatestate_loads_eliminate]: 5.82001e-06
        [parameter_eliminate]: 1.72999e-06
        [specialize_transform]: 1.299e-05
        [updatestate_useless_node_eliminater]: 1.211e-05
        [accelerated_algorithm]: 1.283e-05
        [meta_shard_fg_expand]: 2.57001e-06
        [get_grad_eliminate_]: 1.225e-05
        [merge_forward]: 7.03998e-06
        [cell_reuse_recompute_pass]: 8.79983e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.4e-05
        [j_node_and_user_rematch]: 2.745e-05
        [meta_fg_expand]: 0.00147981
        [replace_old_param]: 5.231e-05
        [inline_without_move]: 5.201e-05
        [renormalize]: 0.00168745
        [add_forward_monad_depend]: 4.79e-06
        [auto_monad_grad]: 1.32999e-06
        [auto_monad_eliminator]: 8.55001e-06
        [cse]: 1.573e-05
        [replace_applicator]: 1.104e-05
    [Cycle 2]: 0.00029458, [27]
        [switch_simplify]: 5.26002e-06
        [loop_unroll]: 3.54002e-06
        [a_1]: 4.301e-05
        [with_stream_mark]: 8.52e-06
        [recompute_prepare]: 4.21001e-06
        [updatestate_depend_eliminate]: 2.09999e-06
        [updatestate_assign_eliminate]: 1.59e-06
        [updatestate_loads_eliminate]: 1.57001e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 3.80998e-06
        [updatestate_useless_node_eliminater]: 3.51001e-06
        [accelerated_algorithm]: 6.17001e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 3.61999e-06
        [merge_forward]: 2.22999e-06
        [cell_reuse_recompute_pass]: 2.17001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.049e-05
        [j_node_and_user_rematch]: 5.89999e-06
        [meta_fg_expand]: 3.811e-05
        [replace_old_param]: 4.22e-06
        [inline_without_move]: 3.65998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 8.10018e-07
        [auto_monad_eliminator]: 4.78001e-06
        [cse]: 1.016e-05
        [replace_applicator]: 4.07e-06
[py_interpret_to_execute_after_opt_a]: 8.22e-06
[rewriter_after_opt_a]: 2.565e-05
[convert_after_rewriter]: 4.90999e-06
[order_py_execute_after_rewriter]: 3.61999e-06
[mutable_eliminate]: 0.00053368
[jit_opt_b]: 4.024e-05, [1]
    [Cycle 1]: 3.322e-05, [2]
        [frontend_op_eliminate]: 1.039e-05
        [inline_after_opt_a]: 1.13e-05
[cconv]: 1.495e-05
[loop_unroll]: 0.00042056
[jit_opt_after_cconv]: 0.0001159, [1]
    [Cycle 1]: 0.00010881, [11]
        [c_1]: 1.181e-05
        [parameter_eliminate]: 2.59999e-06
        [updatestate_depend_eliminate]: 4.62e-06
        [updatestate_assign_eliminate]: 1.80001e-06
        [updatestate_loads_eliminate]: 1.72999e-06
        [cse]: 1.503e-05
        [call_graph_tuple_transform]: 1.161e-05
        [tuple_list_get_item_eliminator]: 3.71001e-06
        [none_parameter_eliminate]: 1.00001e-06
        [renormalize]: 4.10015e-07
        [switch_simplify]: 3.81999e-06
[remove_dup_value]: 8.28001e-06
[partial_unused_args_eliminate]: 1.55001e-06
[environ_conv]: 3.44001e-06
[add_recomputation]: 2.213e-05
[cse_after_recomputation]: 1.949e-05, [1]
    [Cycle 1]: 1.41e-05, [1]
        [cse]: 8.12998e-06
[auto_monad_reorder]: 8.08999e-06
[get_jit_bprop_graph]: 1.43002e-06
[rewriter_after_jit_bprop_graph]: 4.93001e-06
[opt_after_jit_grad]: 0.00045259
[symbol_engine_optimizer]: 6.346e-05, [1]
    [Cycle 1]: 5.734e-05, [6]
        [build]: 2.57001e-06
        [elim_shapecalc]:    [elim_not_effective]: 9.49e-06
        [opt_reshape]: 4.08999e-06
        [fold_const_symbol]: 6.07999e-06
        [renormalize]: 6.29982e-07
[validate]: 6.652e-05
Sums
    bootstrap                            :     0.000626s :  1.52%
    type_inference                       :     0.033099s : 80.23%
    event_method                         :     0.000047s :  0.11%
    auto_monad                           :     0.000156s :  0.38%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000011s :  0.03%
    py_interpret_to_execute              :     0.000049s :  0.12%
    rewriter_before_opt_a                :     0.000129s :  0.31%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000104s :  0.25%
    jit_opt_a.loop_unroll                :     0.000042s :  0.10%
    jit_opt_a.a_1                        :     0.000976s :  2.36%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.08%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.05%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.03%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.02%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.02%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000018s :  0.04%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.04%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.04%
    jit_opt_a.meta_shard_fg_expand       :     0.000013s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.04%
    jit_opt_a.merge_forward              :     0.000011s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.09%
    jit_opt_a.j_node_and_user_rematch    :     0.000040s :  0.10%
    jit_opt_a.meta_fg_expand             :     0.001960s :  4.75%
    jit_opt_a.replace_old_param          :     0.000057s :  0.14%
    jit_opt_a.inline_without_move        :     0.000056s :  0.13%
    jit_opt_a.renormalize                :     0.001622s :  3.93%
    jit_opt_a.add_forward_monad_depend   :     0.000025s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.04%
    jit_opt_a.cse                        :     0.000032s :  0.08%
    jit_opt_a.replace_applicator         :     0.000015s :  0.04%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.02%
    rewriter_after_opt_a                 :     0.000041s :  0.10%
    convert_after_rewriter               :     0.000006s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.01%
    mutable_eliminate                    :     0.000550s :  1.33%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.03%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.03%
    cconv                                :     0.000021s :  0.05%
    loop_unroll                          :     0.000414s :  1.00%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.03%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.01%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.00%
    jit_opt_after_cconv.cse              :     0.000015s :  0.04%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.03%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.01%
    remove_dup_value                     :     0.000013s :  0.03%
    partial_unused_args_eliminate        :     0.000002s :  0.06.06e-06
        [elim_not_effective]: 9.35001e-06
        [opt_reshape]: 4.43999e-06
        [fold_const_symbol]: 6.07001e-06
        [renormalize]: 5.40022e-07
[validate]: 2.138e-05
[backend_pass]: 1.04998e-06
[task_emit]: 1.869e-05
[execute]: 1.15999e-06
Sums
    bootstrap                            :     0.000243s :  1.40%
    type_inference                       :     0.010618s : 60.98%
    event_method                         :     0.000048s :  0.28%
    auto_monad                           :     0.000094s :  0.54%
    graph_reusing                        :     0.000008s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.17%
    rewriter_before_opt_a                :     0.000094s :  0.54%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.38%
    jit_opt_a.loop_unroll                :     0.000042s :  0.24%
    jit_opt_a.a_1                        :     0.000875s :  5.02%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000019s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.09%
    jit_opt_a.merge_forward              :     0.000009s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000033s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.001518s :  8.72%
    jit_opt_a.replace_old_param          :     0.000057s :  0.32%
    jit_opt_a.inline_without_move        :     0.000056s :  0.32%
    jit_opt_a.renormalize                :     0.001688s :  9.69%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.08%
    jit_opt_a.cse                        :     0.000026s :  0.15%
    jit_opt_a.replace_applicator         :     0.000015s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000026s :  0.15%
    convert_after_rewriter               :     0.000005s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000534s :  3.07%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.06%
    cconv                                :     0.000015s :  0.09%
    loop_unroll                          :     0.000421s :  2.42%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000012s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000008s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000003s :  0.02%
    add_recomputation                    :     0.000022s :  0.13%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000008s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000453s :  2.60%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000021s :  0.12%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.000019s :  0.11%
    execute                              :     0.000001s :  0.01%

1%
    environ_conv                         :     0.000009s :  0.02%
    add_recomputation                    :     0.000039s :  0.09%
    cse_after_recomputation.cse          :     0.000009s :  0.02%
    auto_monad_reorder                   :     0.000019s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000179s :  0.43%
    opt_after_jit_grad                   :     0.000470s :  1.14%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.01%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.02%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.01%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000067s :  0.16%


Time group info:
------[substitution.]   0.000293    50
 0.49% :     0.000001s :      1: substitution.elim_not_effective
 0.45% :     0.000001s :      1: substitution.fold_const_symbol
 1.45% :     0.000004s :      1: substitution.graph_param_transform
61.31% :     0.000180s :      6: substitution.inline
 5.67% :     0.000017s :      2: substitution.inline_without_move
 5.39% :     0.000016s :      9: substitution.j_node_and_user_rematch
 1.48% :     0.000004s :      2: substitution.minmaximum_grad
 6.96% :     0.000020s :      3: substitution.partial_eliminate
 2.39% :     0.000007s :      9: substitution.remove_not_recompute_node
 1.92% :     0.000006s :      2: substitution.replace_applicator
 1.46% :     0.000004s :      5: substitution.replace_old_param
 0.95% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.68% :     0.000008s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.70% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.70% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.033037     2
96.91% :     0.032017s :      1: type_inference.infer
 3.09% :     0.001020s :      1: type_inference.specialize
------[replace.]   0.000064     8
78.76% :     0.000051s :      6: replace.inline
21.24% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000188     8
93.74% :     0.000177s :      6: match.inline
 6.26% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000167  1109
 1.63% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.79% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.31% :     0.000002s :     18: predicate.addn_check_dump
 1.57% :     0.000003s :     18: predicate.addn_zero_filter
 2.01% :     0.000003s :     18: predicate.arithmetic_simplify
 1.55% :     0.000003s :     18: predicate.cast_eliminate
 0.13% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.32% :     0.000002s :     18: predicate.compare_switch_simplify
 1.48% :     0.000002s :     18: predicate.depend_value_elim
 1.30% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.42% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.39% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.14% :     0.000000s :      1: predicate.elim_not_effective
 0.25% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.28% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.31% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.42% :     0.000002s :     18: predicate.environ_get_eliminate
 1.28% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.07% :     0.000000s :      1: predicate.fold_const_symbol
 1.19% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.15% :     0.000007s :     28: predicate.inline
 3.37% :     0.000006s :     33: predicate.inline_without_move
 0.53% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :     12: predicate.less_batch_normalization
 1.73% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.73% :     0.000003s :     21: predicate.load_eliminater
 0.72% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.76% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.68% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.28% :     0.000002s :     18: predicate.merge_addn
 1.36% :     0.000002s :     18: predicate.minmaximum_grad
 0.70% :     0.000001s :      1: predicate.mutable_eliminate
 0.18% :     0.000000s :      1: predicate.opt_reshape
 2.07% :     0.000003s :     21: predicate.partial_eliminate
 1.60% :     0.000003s :     18: predicate.print_const_string_wrapper
 1.89% :     0.000003s :     18: predicate.reduce_eli
Time group info:
------[substitution.]   0.000230    50
 0.47% :     0.000001s :      1: substitution.elim_not_effective
 0.28% :     0.000001s :      1: substitution.fold_const_symbol
 0.91% :     0.000002s :      1: substitution.graph_param_transform
68.78% :     0.000158s :      6: substitution.inline
 7.28% :     0.000017s :      2: substitution.inline_without_move
 2.71% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.19% :     0.000003s :      2: substitution.minmaximum_grad
 1.44% :     0.000003s :      3: substitution.partial_eliminate
 3.10% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.03% :     0.000005s :      2: substitution.replace_applicator
 1.55% :     0.000004s :      5: substitution.replace_old_param
 0.96% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.46% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.69% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 5.14% :     0.000012s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010545     2
88.65% :     0.009348s :      1: type_inference.infer
11.35% :     0.001197s :      1: type_inference.specialize
------[replace.]   0.000069     8
78.22% :     0.000054s :      6: replace.inline
21.78% :     0.000015s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     8
95.52% :     0.000154s :      6: match.inline
 4.48% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000169  1109
 1.57% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.71% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.22% :     0.000002s :     18: predicate.addn_check_dump
 1.59% :     0.000003s :     18: predicate.addn_zero_filter
 2.30% :     0.000004s :     18: predicate.arithmetic_simplify
 1.56% :     0.000003s :     18: predicate.cast_eliminate
 0.18% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.23% :     0.000002s :     18: predicate.compare_switch_simplify
 1.46% :     0.000002s :     18: predicate.depend_value_elim
 1.23% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.45% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      1: predicate.elim_not_effective
 0.31% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.50% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.28% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.23% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.66% :     0.000003s :     18: predicate.environ_get_eliminate
 1.39% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.20% :     0.000002s :     12: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 4.41% :     0.000007s :     28: predicate.inline
 3.42% :     0.000006s :     33: predicate.inline_without_move
 0.51% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.46% :     0.000002s :     12: predicate.less_batch_normalization
 1.54% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.67% :     0.000003s :     21: predicate.load_eliminater
 0.82% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.54% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.71% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.40% :     0.000002s :     18: predicate.merge_addn
 1.26% :     0.000002s :     18: predicate.minmaximum_grad
 0.76% :     0.000001s :      1: predicate.mutable_eliminate
 0.15% :     0.000000s :      1: predicate.opt_reshape
 2.05% :     0.000003s :     21: predicate.partial_eliminate
 1.36% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.80% :     0.000003s :     18: predicate.reduce_eliminate
 1.59% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.49% :     0.000002s :     22: predicate.replace_applicator
 1.74% :     0.000003s :     33: predicate.replace_old_param
 0.11% :     0.000000s :      1: predicate.reset_defer_inline
 1.49% :     0.000003s :     18: predicate.reshape_eliminate
 1.45% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.51% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     18: predicate.same_eliminate
 0.66% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.41% :     0.000001s :      2: predicate.special_op_eliminate
 1.34% :     0.000002s :     12: predicate.specialize_transform
 1.62% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.16% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.74% :     0.000005s :     26: predicate.switch_defer_inline
 2.33% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.71% :     0.000011s :     71: predicate.switch_simplify
 1.41% :     0.000002s :     18: predicate.tile_eliminate
 1.42% :     0.000002s :     18: predicate.transpose_eliminate
 1.82% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.53% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.01% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.58% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.09% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000003s :     18: predicate.value_based_eliminate
 0.10% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.30% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001131    22
76.31% :     0.000863s :     13: func_graph_cloner_run.FuncGraphClonerGraph
23.69% :     0.000268s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.055682    72
 0.07% :     0.000042s :      1: add_recomputation
 0.29% :     0.000161s :      1: auto_monad
 0.04% :     0.000022s :      1: auto_monad_reorder
 1.16% :     0.000646s :      1: bootstrap
 0.04% :     0.000023s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.04% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000011s :      1: environ_conv
 0.09% :     0.000051s :      1: event_method
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
29.24% :     0.016282s :      1: jit_opt_a
 0.22% :     0.000123s :      1: jit_opt_after_cconv
 0.08% :     0.000043s :      1: jit_opt_b
 0.76% :     0.000421s :      1: loop_unroll
 1.00% :     0.000557s :      1: mutable_eliminate
 2.46% :     0.001368s :     26: opt.transform.jit_opt_a
 0.06% :     0.000031s :      4: opt.transform.jit_opt_after_cconv
 0.03% :     0.000015s :      4: opt.transform.jit_opt_b
 0.02% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.02% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.03% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.04% :     0.000022s :      4: opt.transform.symbol_engine_opt
 0.86% :     0.000477s :      1: opt_after_jit_grad
 0.01% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000013s :      1: pre_auto_parallel
 0.09% :     0.000052s :      1: py_interpret_to_execute
 0.02% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000016s :      1: remove_dup_value
 2.14% :     0.001191s :    minate
 1.48% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.45% :     0.000002s :     22: predicate.replace_applicator
 1.69% :     0.000003s :     33: predicate.replace_old_param
 0.11% :     0.000000s :      1: predicate.reset_defer_inline
 1.47% :     0.000002s :     18: predicate.reshape_eliminate
 1.74% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.51% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     18: predicate.same_eliminate
 0.71% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.37% :     0.000001s :      2: predicate.special_op_eliminate
 1.25% :     0.000002s :     12: predicate.specialize_transform
 1.84% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.52% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.11% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.87% :     0.000005s :     26: predicate.switch_defer_inline
 2.47% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.69% :     0.000011s :     71: predicate.switch_simplify
 1.46% :     0.000002s :     18: predicate.tile_eliminate
 1.30% :     0.000002s :     18: predicate.transpose_eliminate
 1.62% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.61% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.69% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.86% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.56% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000003s :     18: predicate.value_based_eliminate
 0.11% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.21% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001238    22
72.51% :     0.000898s :     13: func_graph_cloner_run.FuncGraphClonerGraph
27.49% :     0.000340s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024491    76
 0.10% :     0.000025s :      1: add_recomputation
 0.40% :     0.000098s :      1: auto_monad
 0.04% :     0.000011s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.04% :     0.000256s :      1: bootstrap
 0.07% :     0.000018s :      1: cconv
 0.03% :     0.000007s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.02% :     0.000006s :      1: environ_conv
 0.22% :     0.000054s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000011s :      1: graph_reusing
34.29% :     0.008398s :      1: jit_opt_a
 0.48% :     0.000119s :      1: jit_opt_after_cconv
 0.18% :     0.000043s :      1: jit_opt_b
 1.75% :     0.000428s :      1: loop_unroll
 2.21% :     0.000541s :      1: mutable_eliminate
 5.06% :     0.001239s :     26: opt.transform.jit_opt_a
 0.11% :     0.000028s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000023s :      4: opt.transform.symbol_engine_opt
 1.88% :     0.000460s :      1: opt_after_jit_grad
 0.02% :     0.000006s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000032s :      1: py_interpret_to_execute
 0.04% :     0.000011s :      1: py_interpret_to_execute_aft  1: renormalize.infer
 0.76% :     0.000422s :      1: renormalize.specialize
 0.33% :     0.000182s :      1: rewriter_after_jit_bprop_graph
 0.08% :     0.000044s :      1: rewriter_after_opt_a
 0.24% :     0.000131s :      1: rewriter_before_opt_a
 0.12% :     0.000066s :      1: symbol_engine_optimizer
59.55% :     0.033157s :      1: type_inference

er_opt_a
 0.04% :     0.000011s :      1: remove_dup_value
 5.18% :     0.001268s :      1: renormalize.infer
 1.68% :     0.000412s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000029s :      1: rewriter_after_opt_a
 0.40% :     0.000097s :      1: rewriter_before_opt_a
 0.27% :     0.000066s :      1: symbol_engine_optimizer
 0.09% :     0.000022s :      1: task_emit
43.41% :     0.010631s :      1: type_inference
 0.15% :     0.000038s :      1: validate


TotalTime = 0.0105305, [30]
[bootstrap]: 0.00023797
[type_inference]: 0.0036803
[event_method]: 1.475e-05
[auto_monad]: 5.125e-05
[graph_reusing]: 5.04e-06
[pre_auto_parallel]: 1.76e-06
[py_interpret_to_execute]: 2.507e-05
[rewriter_before_opt_a]: 6.392e-05
[expand_dump_flag]: 2.70002e-06
[jit_opt_a]: 0.00454517, [2]
    [Cycle 1]: 0.00130281, [27]
        [switch_simplify]: 4.565e-05
        [loop_unroll]: 2.839e-05
        [a_1]: 0.00046866
        [with_stream_mark]: 1.382e-05
        [recompute_prepare]: 6.94001e-06
        [updatestate_depend_eliminate]: 3.13e-06
        [updatestate_assign_eliminate]: 2.19999e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [parameter_eliminate]: 1.45999e-06
        [specialize_transform]: 6.19999e-06
        [updatestate_useless_node_eliminater]: 5.04998e-06
        [accelerated_algorithm]: 5.33002e-06
        [meta_shard_fg_expand]: 1.62001e-06
        [get_grad_eliminate_]: 5.04e-06
        [merge_forward]: 3.11001e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.235e-05
        [j_node_and_user_rematch]: 8.38001e-06
        [meta_fg_expand]: 1.98002e-06
        [replace_old_param]: 7.75e-06
        [inline_without_move]: 4.92999e-06
        [renormalize]: 0.00045543
        [add_forward_monad_depend]: 5.39e-06
        [auto_monad_grad]: 1.49e-06
        [auto_monad_eliminator]: 1.193e-05
        [cse]: 2.124e-05
        [replace_applicator]: 1.046e-05
    [Cycle 2]: 0.00036549, [27]
        [switch_simplify]: 5.76998e-06
        [loop_unroll]: 4.76002e-06
        [a_1]: 8.762e-05
        [with_stream_mark]: 7.61001e-06
        [recompute_prepare]: 4.57e-06
        [updatestate_depend_eliminate]: 2.27999e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 8.89995e-07
        [specialize_transform]: 4.95999e-06
        [updatestate_useless_node_eliminater]: 4.65001e-06
        [accelerated_algorithm]: 4.84998e-06
        [meta_shard_fg_expand]: 1.00001e-06
        [get_grad_eliminate_]: 4.31002e-06
        [merge_forward]: 2.27999e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.166e-05
        [j_node_and_user_rematch]: 7.21001e-06
        [meta_fg_expand]: 1.39e-06
        [replace_old_param]: 6.98e-06
        [inline_without_move]: 4.73001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.08001e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 4.3e-06
        [cse]: 6.632e-05
        [replace_applicator]: 5.42999e-06
[py_interpret_to_execute_after_opt_a]: 7.70998e-06
[rewriter_after_opt_a]: 2.721e-05
[convert_after_rewriter]: 6.33998e-06
[order_py_execute_after_rewriter]: 5.10999e-06
[mutable_eliminate]: 0.00046986
[jit_opt_b]: 5.284e-05, [1]
    [Cycle 1]: 4.586e-05, [2]
        [frontend_op_eliminate]: 1.708e-05
        [inline_after_opt_a]: 1.645e-05
[cconv]: 2.185e-05
[loop_unroll]: 0.00040405
[jit_opt_after_cconv]: 0.00013151, [1]
    [Cycle 1]: 0.00012532, [11]
        [c_1]: 1.976e-05
        [parameter_eliminate]: 2.48998e-06
        [updatestate_depend_eliminate]: 4.77e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.563e-05
        [call_graph_tuple_transform]: 1.818e-05
        [tuple_list_get_item_eliminator]: 5.13002e-06
        [none_parameter_eliminate]: 9.40025e-07
        [renormalize]: 4.00003e-07
        [switch_simplify]: 5.17999e-06
[remove_dup_value]: 1.158e-05
[partial_unused_args_eliminate]: 2.39001e-06
[environ_conv]: 4.64002e-06
[add_recomputation]: 4.195e-05
[cse_after_recomputation]: 1.929e-05, [1]
    [Cycle 1]: 1.406e-05, [1]
        [cse]: 8.78001e-06
[auto_monad_reorder]: 1.245e-05
[get_jit_bprop_graph]: 1.42999e-06
[rewriter_after_jit_bprop_graph]: 4.97999e-06
[opt_after_jit_grad]: 0.00043354
[symbol_engine_optimizer]: 6.989e-05, [1]
    [Cycle 1]: 6.364e-05, [6]
        [build]: 3.30998e-06
        [elim_shapecalc]: 7.17002e-06
        [elim_not_effective]: 1.209e-05
        [opt_reshape]: 5.64998e-06
        [fold_const_symbol]: 8.21002e-06
        [renormalize]: 2.89991e-07
[validate]: 2.779e-05
Sums
    bootstrap                            :     0.000238s :  3.35%
    type_inference                       :     0.003680s : 51.74%
    event_method                         :     0.000015s :  0.21%
    auto_monad                           :     0.000051s :  0.72%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000025s :  0.35%
    rewriter_before_opt_a                :     0.000064s :  0.90%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000051s :  0.72%
    jit_opt_a.loop_unroll                :     0.000033s :  0.47%
    jit_opt_a.a_1                        :     0.000556s :  7.82%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.16%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.14%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.13%
    jit_opt_a.merge_forward              :     0.000005s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.14%
    jit_opt_a.renormalize                :     0.000456s :  6.40%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.23%
    jit_opt_a.cse                        :     0.000088s :  1.23%
    jit_opt_a.replace_applicator         :     0.000016s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.11%
    rewriter_after_opt_a                 :     0.000027s :  0.38%
    convert_after_rewriter               :     0.000006s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000470s :  6.61%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.24%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.23%
    cconv                                :     0.000022s :  0.31%
    loop_unroll                          :     0.000404s :  5.68%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.26%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000012s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000042s :  0.59%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000012s :  0.18%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000434s :  6.09%
    symbol_engine_optimizer.build        :     0.000003s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.39%


Time group info:
------[substitution.]   0.000161    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.80% :     0.000001s :      2: substitution.fold_const_symbol
 3.12% :     0.000005s :      3: substitution.graph_param_transform
82.74% :     0.000133s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.54% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.91% :     0.000003s :      2: substitution.replace_old_param
 6.25% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003632     2
80.76% :     0.002933s :      1: type_inference.infer
19.24% :     0.000699s :      1: type_inference.specialize
------[replace.]   0.000058     6
66.95% :     0.000039s :      4: replace.inline
33.05% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000139     6
93.62% :     0.000130s :      4: match.inline
 6.38% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.42% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.10% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 1.86% :     0.000002s :     10: predicate.arithmetic_simplify
 1.19% :     0.000001s :     10: predicate.cast_eliminate
 0.55% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000001s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.44% :     0.000000s :      3: predicate.graph_param_transform
 5.35% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.20% :     0.000001s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.70% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 2.15% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.24% :     0.000002s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.83% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.44% :     0.000000s :      3: predicate.reset_defer_inline
 1.32% :     0.000001s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.10% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.51% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.51% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.62% :     0.000008s :     53: predicate.switch_simplify
 1.29% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.58% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.39% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000472    12
49.17% :     0.000232s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.68% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.15% :     0.000213s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011803    72
 0.38% :     0.000045s :      1: add_recomputation
 0.46% :     0.000054s :      1: auto_monad
 0.13% :     0.000015s :      1: auto_monad_reorder
 2.13% :     0.000252s :      1: bootstrap
 0.21% :     0.000025s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000021s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.15% :     0.000018s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
38.53% :     0.004548s :      1: jit_opt_a
 1.14% :     0.000134s :      1: jit_opt_after_cconv
 0.47% :     0.000055s :      1: jit_opt_b
 3.49% :     0.000411s :      1: loop_unroll
 4.05% :     0.000478s :      1: mutable_eliminate
 6.33% :     0.000747s :     26: opt.transform.jit_opt_a
 0.38% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.23% :     0.000027s :      4: opt.transform.jit_opt_b
 0.09% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.25% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.73% :     0.000441s :      1: opt_after_jit_grad
 0.06% :     0.000007s :      1: order_py_execute_after_rewriter
 0.04% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.23% :     0.000028s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000014s :      1: remove_dup_value
 1.93% :     0.000228s :      1: renormalize.infer
 1.87% :     0.000221s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000030s :      1: rewriter_after_opt_a
 0.56% :     0.000067s :      1: rewriter_before_opt_a
 0.61% :     0.000072s :      1: symbol_engine_optimizer
31.27% :     0.003691s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.418.608 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c375ee0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.419.257 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.423.323 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843c19d0> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.423.928 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0104745, [30]
[bootstrap]: 0.00023856
[type_inference]: 0.00371365
[event_method]: 1.526e-05
[auto_monad]: 5.529e-05
[graph_reusing]: 5.43002e-06
[pre_auto_parallel]: 1.89e-06
[py_interpret_to_execute]: 2.639e-05
[rewriter_before_opt_a]: 7.184e-05
[expand_dump_flag]: 2.73e-06
[jit_opt_a]: 0.00442585, [2]
    [Cycle 1]: 0.00128179, [27]
        [switch_simplify]: 4.5e-05
        [loop_unroll]: 2.772e-05
        [a_1]: 0.00047005
        [with_stream_mark]: 1.416e-05
        [recompute_prepare]: 7.19001e-06
        [updatestate_depend_eliminate]: 3.30998e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [parameter_eliminate]: 1.59998e-06
        [specialize_transform]: 5.74999e-06
        [updatestate_useless_node_eliminater]: 4.79998e-06
        [accelerated_algorithm]: 5.44e-06
        [meta_shard_fg_expand]: 1.64998e-06
        [get_grad_eliminate_]: 5.36998e-06
        [merge_forward]: 3.30998e-06
        [cell_reuse_recompute_pass]: 9.50007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.273e-05
        [j_node_and_user_rematch]: 8.64e-06
        [meta_fg_expand]: 2.02001e-06
        [replace_old_param]: 7.83999e-06
        [inline_without_move]: 5.25001e-06
        [renormalize]: 0.00043326
        [add_forward_monad_depend]: 4.95999e-06
        [auto_monad_grad]: 1.66e-06
        [auto_monad_eliminator]: 1.149e-05
        [cse]: 2.112e-05
        [replace_applicator]: 1.092e-05
    [Cycle 2]: 0.0003102, [27]
        [switch_simplify]: 5.34e-06
        [loop_unroll]: 4.82e-06
        [a_1]: 8.775e-05
        [with_stream_mark]: 8.40001e-06
        [recompute_prepare]: 4.82e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 1.98997e-06
        [updatestate_loads_eliminate]: 1.86998e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 4.82e-06
        [updatestate_useless_node_eliminater]: 4.67e-06
        [accelerated_algorithm]: 5.09998e-06
        [meta_shard_fg_expand]: 1.00999e-06
        [get_grad_eliminate_]: 4.53001e-06
        [merge_forward]: 2.21e-06
        [cell_reuse_recompute_pass]: 1.45001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.172e-05
        [j_node_and_user_rematch]: 7.26001e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 6.84999e-06
        [inline_without_move]: 9.14e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 4.75001e-06
        [cse]: 9.25999e-06
        [replace_applicator]: 4.68001e-06
[py_interpret_to_execute_after_opt_a]: 7.45e-06
[rewriter_after_opt_a]: 2.91e-05
[convert_after_rewriter]: 6.34001e-06
[order_py_execute_after_rewriter]: 5.40001e-06
[mutable_eliminate]: 0.00046973
[jit_opt_b]: 4.803e-05, [1]
    [Cycle 1]: 4.159e-05, [2]
        [frontend_op_eliminate]: 1.628e-05
        [inline_after_opt_a]: 1.42e-05
[cconv]: 2.122e-05
[loop_unroll]: 0.00041881
[jit_opt_after_cconv]: 0.0001338, [1]
    [Cycle 1]: 0.00012743, [11]
        [c_1]: 2.018e-05
        [parameter_eliminate]: 2.79999e-06
        [updatestate_depend_eliminate]: 5.57001e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.543e-05
        [call_graph_tuple_transform]: 1.777e-05
        [tuple_list_get_item_eliminator]: 5.44998e-06
        [none_parameter_eliminate]: 1.32e-06
        [renormalize]: 3.9002e-07
        [switch_simplify]: 5.07e-06
[remove_dup_value]: 1.16e-05
[partial_unused_args_eliminate]: 1.76e-06
[environ_conv]: 4.15999e-06
[add_recomputation]: 3.95e-05
[cse_after_recomputation]: 1.962e-05, [1]
    [Cycle 1]: 1.443e-05, [1]
        [cse]: 8.47998e-06
[auto_monad_reorder]: 1.288e-05
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 4.4e-06
[opt_after_jit_grad]: 0.00044107
[symbol_engine_optimizer]: 7.082e-05, [1]
    [Cycle 1]: 6.45e-05, [6]
        [build]: 3.33e-06
        [elim_shapecalc]: 7.47002e-06
        [elim_not_effective]: 1.215e-05
        [opt_reshape]: 5.47001e-06
        [fold_const_symbol]: 8.53001e-06
        [renormalize]: 3.49974e-07
[validate]: 2.668e-05
Sums
    bootstrap                            :     0.000239s :  3.36%
    type_inference                       :     0.003714s : 52.25%
    event_method                         :     0.000015s :  0.21%
    auto_monad                           :     0.000055s :  0.78%
    graph_reusing                        :     0.000005s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000026s :  0.37%
    rewriter_before_opt_a                :     0.000072s :  1.01%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000050s :  0.71%
    jit_opt_a.loop_unroll                :     0.000033s :  0.46%
    jit_opt_a.a_1                        :     0.000558s :  7.85%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.14%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.21%
    jit_opt_a.inline_without_move        :     0.000014s :  0.20%
    jit_opt_a.renormalize                :     0.000433s :  6.10%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.23%
    jit_opt_a.cse                        :     0.000030s :  0.43%
    jit_opt_a.replace_applicator         :     0.000016s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.10%
    rewriter_after_opt_a                 :     0.000029s :  0.41%
    convert_after_rewriter               :     0.000006s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.08%
    mutable_eliminate                    :     0.000470s :  6.61%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.20%
    cconv                                :     0.000021s :  0.30%
    loop_unroll                          :     0.000419s :  5.89%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000015s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.25%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.08%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000012s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.06%
    add_recomputation                    :     0.000039s :  0.56%
    cse_after_recomputation.cse          :     0.000008s :  0.12%
    auto_monad_reorder                   :     0.000013s :  0.18%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.06%
    opt_after_jit_grad                   :     0.000441s :  6.21%
    symbol_engine_optimizer.build        :     0.000003s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.38%


Time group info:
------[substitution.]   0.000161    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.75% :     0.000001s :      2: substitution.fold_const_symbol
 3.35% :     0.000005s :      3: substitution.graph_param_transform
82.36% :     0.000132s :      4: substitution.inline
 1.78% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.63% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.73% :     0.000003s :      2: substitution.replace_old_param
 6.33% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003662     2
80.49% :     0.002948s :      1: type_inference.infer
19.51% :     0.000715s :      1: type_inference.specialize
------[replace.]   0.000059     6
66.62% :     0.000039s :      4: replace.inline
33.38% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000139     6
93.44% :     0.000130s :      4: match.inline
 6.56% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.61% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000001s :     10: predicate.addn_zero_filter
 1.99% :     0.000002s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.48% :     0.000006s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000001s :      6: predicate.less_batch_normalization
 1.45% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.18% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.61% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.54% :     0.000003s :     15: predicate.partial_eliminate
 1.25% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.79% :     0.000001s :      6: predicate.replace_old_param
 0.35% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.97% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.47% :     0.000002s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.13% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.43% :     0.000003s :     16: predicate.switch_defer_inline
 2.01% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.49% :     0.000008s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.52% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.84% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.86% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000477    12
49.55% :     0.000236s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.12% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.33% :     0.000216s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.011731    72
 0.36% :     0.000042s :      1: add_recomputation
 0.50% :     0.000058s :      1: auto_monad
 0.13% :     0.000015s :      1: auto_monad_reorder
 2.14% :     0.000251s :      1: bootstrap
 0.20% :     0.000024s :      1: cconv
 0.07% :     0.000008s :      1: convert_after_rewriter
 0.18% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.16% :     0.000019s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.07% :     0.000008s :      1: graph_reusing
37.75% :     0.004428s :      1: jit_opt_a
 1.16% :     0.000137s :      1: jit_opt_after_cconv
 0.43% :     0.000051s :      1: jit_opt_b
 3.63% :     0.000426s :      1: loop_unroll
 4.07% :     0.000478s :      1: mutable_eliminate
 6.42% :     0.000753s :     26: opt.transform.jit_opt_a
 0.39% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000024s :      4: opt.transform.jit_opt_b
 0.10% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.83% :     0.000449s :      1: opt_after_jit_grad
 0.07% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000029s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000014s :      1: remove_dup_value
 1.77% :     0.000207s :      1: renormalize.infer
 1.87% :     0.000219s :      1: renormalize.specialize
 0.05% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000032s :      1: rewriter_after_opt_a
 0.64% :     0.000075s :      1: rewriter_before_opt_a
 0.62% :     0.000073s :      1: symbol_engine_optimizer
31.74% :     0.003724s :      1: type_inference


TotalTime = 0.0331206, [33]
[bootstrap]: 0.00018944
[type_inference]: 0.00978044
[event_method]: 4.588e-05
[auto_monad]: 9.394e-05
[graph_reusing]: 7.45e-06
[pre_auto_parallel]: 3.14001e-06
[py_interpret_to_execute]: 2.529e-05
[rewriter_before_opt_a]: 9.356e-05
[expand_dump_flag]: 3.23e-06
[jit_opt_a]: 0.0126356, [3]
    [Cycle 1]: 0.00616714, [27]
        [switch_simplify]: 6.757e-05
        [loop_unroll]: 4.007e-05
        [a_1]: 0.00091212
        [with_stream_mark]: 2.698e-05
        [recompute_prepare]: 2.051e-05
        [updatestate_depend_eliminate]: 8.06001e-06
        [updatestate_assign_eliminate]: 6.72002e-06
        [updatestate_loads_eliminate]: 6.01998e-06
        [parameter_eliminate]: 3.06001e-06
        [specialize_transform]: 1.384e-05
        [updatestate_useless_node_eliminater]: 1.276e-05
        [accelerated_algorithm]: 1.221e-05
        [meta_shard_fg_expand]: 3.27002e-06
        [get_grad_eliminate_]: 1.193e-05
        [merge_forward]: 8.43001e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.752e-05
        [j_node_and_user_rematch]: 2.366e-05
        [meta_fg_expand]: 0.00155518
        [replace_old_param]: 5.892e-05
        [inline_without_move]: 5.477e-05
        [renormalize]: 0.00274408
        [add_forward_monad_depend]: 1.283e-05
        [auto_monad_grad]: 6.21e-06
        [auto_monad_eliminator]: 5.717e-05
        [cse]: 0.00016843
        [replace_applicator]: 7.911e-05
    [Cycle 2]: 0.00260953, [27]
        [switch_simplify]: 3.753e-05
        [loop_unroll]: 3.625e-05
        [a_1]: 0.00120721
        [with_stream_mark]: 1.916e-05
        [recompute_prepare]: 9.76e-06
        [updatestate_depend_eliminate]: 4.53001e-06
        [updatestate_assign_eliminate]: 4.10998e-06
        [updatestate_loads_eliminate]: 3.83001e-06
        [parameter_eliminate]: 1.97999e-06
        [specialize_transform]: 7.25e-06
        [updatestate_useless_node_eliminater]: 7.23999e-06
        [accelerated_algorithm]: 7.95e-06
        [meta_shard_fg_expand]: 2.44001e-06
        [get_grad_eliminate_]: 7.38999e-06
        [merge_forward]: 4.77e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.737e-05
        [j_node_and_user_rematch]: 1.185e-05
        [meta_fg_expand]: 5.467e-05
        [replace_old_param]: 1.17e-05
        [inline_without_move]: 6.69999e-06
        [renormalize]: 0.00085026
        [add_forward_monad_depend]: 2.137e-05
        [auto_monad_grad]: 2.06998e-06
        [auto_monad_eliminator]: 3.219e-05
        [cse]: 3.052e-05
        [replace_applicator]: 1.601e-05
    [Cycle 3]: 0.00041773, [27]
        [switch_simplify]: 7.5e-06
        [loop_unroll]: 7.36999e-06
        [a_1]: 0.00014462
        [with_stream_mark]: 1.102e-05
        [recompute_prepare]: 6.95002e-06
        [updatestate_depend_eliminate]: 3.97998e-06
        [updatestate_assign_eliminate]: 3.35e-06
        [updatestate_loads_eliminate]: 3.08e-06
        [parameter_eliminate]: 1.45001e-06
        [specialize_transform]: 6.67002e-06
        [updatestate_useless_node_eliminater]: 6.73e-06
        [accelerated_algorithm]: 7.15e-06
        [meta_shard_fg_expand]: 2.23002e-06
        [get_grad_eliminate_]: 6.57002e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 2.30002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.841e-05
        [j_node_and_user_rematch]: 1.215e-05
        [meta_fg_expand]: 2.48002e-06
        [replace_old_param]: 9.56e-06
        [inline_without_move]: 6.68e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 7.99977e-07
        [auto_monad_eliminator]: 7.54002e-06
        [cse]: 1.624e-05
        [replace_applicator]: 7.27002e-06
[py_interpret_to_execute_after_opt_a]: 1.457e-05
[rewriter_after_opt_a]: 4.692e-05
[convert_after_rewriter]: 8.32e-06
[order_py_execute_after_rewriter]: 5.62001e-06
[mutable_eliminate]: 0.00072666
[jit_opt_b]: 5.995e-05, [1]
    [Cycle 1]: 5.192e-05, [2]
        [frontend_op_eliminate]: 2.105e-05
        [inline_after_opt_a]: 1.97e-05
[cconv]: 2.3e-05
[loop_unroll]: 0.00042291
[jit_opt_after_cconv]: 0.00016892, [1]
    [Cycle 1]: 0.00016235, [11]
        [c_1]: 2.988e-05
        [parameter_eliminate]: 2.84001e-06
        [updatestate_depend_eliminate]: 7.46001e-06
        [updatestate_assign_eliminate]: 4.3e-06
        [updatestate_loads_eliminate]: 3.08e-06
        [cse]: 2.847e-05
        [call_graph_tuple_transform]: 2.226e-05
        [tuple_list_get_item_eliminator]: 6.73e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 9.30013e-07
        [switch_simplify]: 7.64002e-06
[remove_dup_value]: 1.604e-05
[partial_unused_args_eliminate]: 2.04e-06
[environ_conv]: 1.209e-05
[add_recomputation]: 5.415e-05
[cse_after_recomputation]: 3.358e-05, [1]
    [Cycle 1]: 2.708e-05, [1]
        [cse]: 1.972e-05
[auto_monad_reorder]: 2.139e-05
[get_jit_bprop_graph]: 1.42999e-06
[rewriter_after_jit_bprop_graph]: 5.47999e-06
[opt_after_jit_grad]: 0.00045094
[symbol_engine_optimizer]: 8.311e-05, [1]
    [Cycle 1]: 7.686e-05, [6]
        [build]: 3.51999e-06
        [elim_shapecalc]: 9.94001e-06
        [elim_not_effective]: 1.594e-05
        [opt_reshape]: 8.00999e-06
        [fold_const_symbol]: 1.232e-05
        [renormalize]: 3.4002e-07
[validate]: 4.107e-05
[backend_pass]: 1.18001e-06
[task_emit]: 0.00782641
[execute]: 8.27998e-06
Sums
    bootstrap                            :     0.000189s :  0.66%
    type_inference                       :     0.009780s : 33.93%
    event_method                         :     0.000046s :  0.16%
    auto_monad                           :     0.000094s :  0.33%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000025s :  0.09%
    rewriter_before_opt_a                :     0.000094s :  0.32%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000113s :  0.39%
    jit_opt_a.loop_unroll                :     0.000084s :  0.29%
    jit_opt_a.a_1                        :     0.002264s :  7.85%
    jit_opt_a.with_stream_mark           :     0.000057s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000037s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000014s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000013s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.02%
    jit_opt_a.specialize_transform       :     0.000028s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000027s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000027s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000026s :  0.09%
    jit_opt_a.merge_forward              :     0.000017s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000063s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000048s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001612s :  5.59%
    jit_opt_a.replace_old_param          :     0.000080s :  0.28%
    jit_opt_a.inline_without_move        :     0.000068s :  0.24%
    jit_opt_a.renormalize                :     0.003594s : 12.47%
    jit_opt_a.add_forward_monad_depend   :     0.000036s :  0.12%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000097s :  0.34%
    jit_opt_a.cse                        :     0.000215s :  0.75%
    jit_opt_a.replace_applicator         :     0.000102s :  0.36%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.05%
    rewriter_after_opt_a                 :     0.000047s :  0.16%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000727s :  2.52%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.07%
    cconv                                :     0.000023s :  0.08%
    loop_unroll                          :     0.000423s :  1.47%
    jit_opt_after_cconv.c_1              :     0.000030s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000008s :  0.03%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000012s :  0.04%
    add_recomputation                    :     0.000054s :  0.19%
    cse_after_recomputation.cse          :     0.000020s :  0.07%
    auto_monad_reorder                   :     0.000021s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000451s :  1.56%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000041s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007826s : 27.15%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000581   118
 5.41% :     0.000031s :      1: substitution.arithmetic_simplify
 0.45% :     0.000003s :      4: substitution.elim_not_effective
 0.33% :     0.000002s :      4: substitution.fold_const_symbol
 1.04% :     0.000006s :      5: substitution.graph_param_transform
60.83% :     0.000354s :     12: substitution.inline
 3.31% :     0.000019s :      2: substitution.inline_without_move
 1.68% :     0.000010s :     16: substitution.j_node_and_user_rematch
 2.01% :     0.000012s :      7: substitution.minmaximum_grad
 1.35% :     0.000008s :      5: substitution.partial_eliminate
 2.25% :     0.000013s :     16: substitution.remove_not_recompute_node
 5.15% :     0.000030s :     10: substitution.replace_applicator
 1.45% :     0.000008s :      7: substitution.replace_old_param
 0.48% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.00% :     0.000023s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.63% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 7.63% :     0.000044s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009653     2
88.71% :     0.008563s :      1: type_inference.infer
11.29% :     0.001090s :      1: type_inference.specialize
------[replace.]   0.000163    20
 3.42% :     0.000006s :      1: replace.arithmetic_simplify
62.27% :     0.000102s :     12: replace.inline
34.31% :     0.000056s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000399    20
 7.63% :     0.000030s :      1: match.arithmetic_simplify
86.96% :     0.000347s :     12: match.inline
 5.41% :     0.000022s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000400  2790
 1.65% :     0.000007s :     46: predicate.accumulaten_eliminater
 0.41% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.39% :     0.000006s :     46: predicate.addn_check_dump
 1.70% :     0.000007s :     46: predicate.addn_zero_filter
 2.10% :     0.000008s :     47: predicate.arithmetic_simplify
 1.58% :     0.000006s :     47: predicate.cast_eliminate
 0.20% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.39% :     0.000006s :     46: predicate.compare_switch_simplify
 1.60% :     0.000006s :     46: predicate.depend_value_elim
 1.52% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.52% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.60% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.29% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      5: predicate.elim_not_effective
 0.30% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.50% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.42% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.44% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.51% :     0.000006s :     47: predicate.environ_get_eliminate
 1.46% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.11% :     0.000000s :      5: predicate.fold_const_symbol
 0.89% :     0.000004s :     21: predicate.get_grad_eliminate
 0.12% :     0.000000s :      5: predicate.graph_param_transform
 4.45% :     0.000018s :     76: predicate.inline
 1.68% :     0.000007s :     42: predicate.inline_without_move
 0.37% :     0.000001s :     21: predicate.j_node_and_user_rematch
 0.99% :     0.000004s :     21: predicate.less_batch_normalization
 1.91% :     0.000008s :     54: predicate.list_to_tuple_eliminator_
 1.95% :     0.000008s :     59: predicate.load_eliminater
 0.50% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.03% :     0.000012s :     89: predicate.loop_unroll_before_grad
 1.84% :     0.000007s :     52: predicate.make_slice_get_slice_eliminator
 1.43% :     0.000006s :     46: predicate.merge_addn
 1.49% :     0.000006s :     47: predicate.minmaximum_grad
 0.46% :     0.000002s :      5: predicate.mutable_eliminate
 0.22% :     0.000001s :      5: predicate.opt_reshape
 2.30% :     0.000009s :     59: predicate.partial_eliminate
 1.51% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.87% :     0.000007s :     47: predicate.reduce_eliminate
 1.82% :     0.000007s :     54: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.40% :     0.000010s :    105: predicate.replace_applicator
 0.91% :     0.000004s :     42: predicate.replace_old_param
 0.13% :     0.000001s :      5: predicate.reset_defer_inline
 1.69% :     0.000007s :     47: predicate.reshape_eliminate
 1.48% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.34% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.51% :     0.000006s :     46: predicate.same_eliminate
 0.46% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.55% :     0.000002s :     10: predicate.special_op_eliminate
 0.91% :     0.000004s :     21: predicate.specialize_transform
 1.62% :     0.000006s :     46: predicate.split_environ_get_set_with_tuple_value
 1.58% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.19% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.65% :     0.000011s :     66: predicate.switch_defer_inline
 2.52% :     0.000010s :     66: predicate.switch_layer_defer_inline
 6.13% :     0.000025s :    160: predicate.switch_simplify
 1.53% :     0.000006s :     47: predicate.tile_eliminate
 1.67% :     0.000007s :     47: predicate.transpose_eliminate
 1.79% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.62% :     0.000006s :     47: predicate.tuple_list_get_item_depend_reorder
 3.41% :     0.000014s :     64: predicate.tuple_list_get_item_eliminator
 1.91% :     0.000008s :     47: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000007s :     54: predicate.tuple_to_list_eliminator_
 1.88% :     0.000008s :     59: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000012s :     80: predicate.updatestate_useless_node_eliminater
 1.89% :     0.000008s :     46: predicate.value_based_eliminate
 0.19% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.24% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001503    29
58.08% :     0.000873s :     13: func_graph_cloner_run.FuncGraphClonerGraph
41.92% :     0.000630s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.039771    91
 0.15% :     0.000058s :      1: add_recomputation
 0.25% :     0.000099s :      1: auto_monad
 0.06% :     0.000024s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.51% :     0.000202s :      1: bootstrap
 0.06% :     0.000026s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000036s :      1: cse_after_recomputation
 0.04% :     0.000015s :      1: environ_conv
 0.13% :     0.000051s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
31.78% :     0.012640s :      1: jit_opt_a
 0.43% :     0.000172s :      1: jit_opt_after_cconv
 0.16% :     0.000062s :      1: jit_opt_b
 1.09% :     0.000432s :      1: loop_unroll
 1.85% :     0.000734s :      1: mutable_eliminate
 7.34% :     0.002918s :     39: opt.transform.jit_opt_a
 0.16% :     0.000063s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000034s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000043s :      4: opt.transform.symbol_engine_opt
 1.15% :     0.000459s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000006s :      1: pre_auto_parallel
 0.07% :     0.000028s :      1: py_interpret_to_execute
 0.04% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 4.57% :     0.001818s :      2: renormalize.infer
 4.43% :     0.001761s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000050s :      1: rewriter_after_opt_a
 0.24% :     0.000097s :      1: rewriter_before_opt_a
 0.21% :     0.000085s :      1: symbol_engine_optimizer
19.70% :     0.007836s :      1: task_emit
24.65% :     0.009804s :      1: type_inference
 0.16% :     0.000063s :      1: validate


TotalTime = 0.0226149, [33]
[bootstrap]: 0.00023062
[type_inference]: 0.0111751
[event_method]: 5.029e-05
[auto_monad]: 9.617e-05
[graph_reusing]: 7e-06
[pre_auto_parallel]: 2.81e-06
[py_interpret_to_execute]: 2.55e-05
[rewriter_before_opt_a]: 9.089e-05
[expand_dump_flag]: 3.29001e-06
[jit_opt_a]: 0.00886149, [2]
    [Cycle 1]: 0.00505013, [27]
        [switch_simplify]: 6.604e-05
        [loop_unroll]: 3.937e-05
        [a_1]: 0.00090563
        [with_stream_mark]: 2.671e-05
        [recompute_prepare]: 1.831e-05
        [updatestate_depend_eliminate]: 7.57002e-06
        [updatestate_assign_eliminate]: 6.34999e-06
        [updatestate_loads_eliminate]: 6.01e-06
        [parameter_eliminate]: 2.22001e-06
        [specialize_transform]: 1.343e-05
        [updatestate_useless_node_eliminater]: 1.221e-05
        [accelerated_algorithm]: 1.25e-05
        [meta_shard_fg_expand]: 3.66999e-06
        [get_grad_eliminate_]: 1.192e-05
        [merge_forward]: 8.12e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.671e-05
        [j_node_and_user_rematch]: 2.287e-05
        [meta_fg_expand]: 0.00165305
        [replace_old_param]: 5.694e-05
        [inline_without_move]: 5.481e-05
        [renormalize]: 0.00183642
        [add_forward_monad_depend]: 5.36002e-06
        [auto_monad_grad]: 2.61999e-06
        [auto_monad_eliminator]: 1.287e-05
        [cse]: 1.997e-05
        [replace_applicator]: 1.113e-05
    [Cycle 2]: 0.00029704, [27]
        [switch_simplify]: 4.73001e-06
        [loop_unroll]: 3.9e-06
        [a_1]: 4.55e-05
        [with_stream_mark]: 8.75999e-06
        [recompute_prepare]: 3.91999e-06
        [updatestate_depend_eliminate]: 2.19999e-06
        [updatestate_assign_eliminate]: 1.66e-06
        [updatestate_loads_eliminate]: 1.44998e-06
        [parameter_eliminate]: 1.04998e-06
        [specialize_transform]: 3.71999e-06
        [updatestate_useless_node_eliminater]: 3.86001e-06
        [accelerated_algorithm]: 3.93001e-06
        [meta_shard_fg_expand]: 1.24998e-06
        [get_grad_eliminate_]: 3.22002e-06
        [merge_forward]: 1.88997e-06
        [cell_reuse_recompute_pass]: 1.60001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.043e-05
        [j_node_and_user_rematch]: 6.29001e-06
        [meta_fg_expand]: 3.853e-05
        [replace_old_param]: 3.9e-06
        [inline_without_move]: 3.25e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 1.10001e-06
        [auto_monad_eliminator]: 5.07e-06
        [cse]: 1.088e-05
        [replace_applicator]: 4.07003e-06
[py_interpret_to_execute_after_opt_a]: 1.006e-05
[rewriter_after_opt_a]: 3.39e-05
[convert_after_rewriter]: 5.95002e-06
[order_py_execute_after_rewriter]: 4.90001e-06
[mutable_eliminate]: 0.00056013
[jit_opt_b]: 4.183e-05, [1]
    [Cycle 1]: 3.428e-05, [2]
        [frontend_op_eliminate]: 1.068e-05
        [inline_after_opt_a]: 1.187e-05
[cconv]: 2.009e-05
[loop_unroll]: 0.00042244
[jit_opt_after_cconv]: 0.00013433, [1]
    [Cycle 1]: 0.00012723, [11]
        [c_1]: 1.267e-05
        [parameter_eliminate]: 2.56998e-06
        [updatestate_depend_eliminate]: 4.68999e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [cse]: 1.624e-05
        [call_graph_tuple_transform]: 1.398e-05
        [tuple_list_get_item_eliminator]: 4.04002e-06
        [none_parameter_eliminate]: 1.67999e-06
        [renormalize]: 3.4002e-07
        [switch_simplify]: 1.659e-05
[remove_dup_value]: 1.428e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 5.26002e-06
[add_recomputation]: 3.171e-05
[cse_after_recomputation]: 2.109e-05, [1]
    [Cycle 1]: 1.526e-05, [1]
        [cse]: 8.38999e-06
[auto_monad_reorder]: 1.227e-05
[get_jit_bprop_graph]: 1.71e-06
[rewriter_after_jit_bprop_graph]: 4.91997e-06
[opt_after_jit_grad]: 0.0004519
[symbol_engine_optimizer]: 6.465e-05, [1]
    [Cycle 1]: 5.772e-05, [6]
        [build]: 3.20998e-06
        [elim_shapecalc]: 6.24999e-06
        [elim_not_effective]: 1.007e-05
        [opt_reshape]: 3.98999e-06
        [fold_const_symbol]: 6.38e-06
        [renormalize]: 4.50003e-07
[validate]: 2.828e-05
[backend_pass]: 8.89995e-07
[task_emit]: 2.2e-05
[execute]: 1.30999e-06
Sums
    bootstrap                            :     0.000231s :  1.25%
    type_inference                       :     0.011175s : 60.49%
    event_method                         :     0.000050s :  0.27%
    auto_monad                           :     0.000096s :  0.52%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.14%
    rewriter_before_opt_a                :     0.000091s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000071s :  0.38%
    jit_opt_a.loop_unroll                :     0.000043s :  0.23%
    jit_opt_a.a_1                        :     0.000951s :  5.15%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.08%
    jit_opt_a.merge_forward              :     0.000010s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001692s :  9.16%
    jit_opt_a.replace_old_param          :     0.000061s :  0.33%
    jit_opt_a.inline_without_move        :     0.000058s :  0.31%
    jit_opt_a.renormalize                :     0.001837s :  9.94%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.10%
    jit_opt_a.cse                        :     0.000031s :  0.17%
    jit_opt_a.replace_applicator         :     0.000015s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.05%
    rewriter_after_opt_a                 :     0.000034s :  0.18%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000560s :  3.03%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.06%
    cconv                                :     0.000020s :  0.11%
    loop_unroll                          :     0.000422s :  2.29%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000017s :  0.09%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000032s :  0.17%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000012s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000452s :  2.45%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000022s :  0.12%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000274    50
 0.54% :     0.000001s :      1: substitution.elim_not_effective
 0.35% :     0.000001s :      1: substitution.fold_const_symbol
 1.53% :     0.000004s :      1: substitution.graph_param_transform
67.57% :     0.000185s :      6: substitution.inline
 6.89% :     0.000019s :      2: substitution.inline_without_move
 2.44% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.38% :     0.000004s :      2: substitution.minmaximum_grad
 1.79% :     0.000005s :      3: substitution.partial_eliminate
 2.55% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.31% :     0.000006s :      2: substitution.replace_applicator
 1.46% :     0.000004s :      5: substitution.replace_old_param
 0.80% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.74% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.75% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.90% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011101     2
89.17% :     0.009899s :      1: type_inference.infer
10.83% :     0.001202s :      1: type_inference.specialize
------[replace.]   0.000069     8
77.61% :     0.000053s :      6: replace.inline
22.39% :     0.000015s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000193     8
94.01% :     0.000182s :      6: match.inline
 5.99% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000179  1109
 1.99% :     0.000004s :     18: predicate.accumulaten_eliminater
 0.66% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.28% :     0.000002s :     18: predicate.addn_check_dump
 1.75% :     0.000003s :     18: predicate.addn_zero_filter
 2.21% :     0.000004s :     18: predicate.arithmetic_simplify
 1.41% :     0.000003s :     18: predicate.cast_eliminate
 0.12% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.30% :     0.000002s :     18: predicate.compare_switch_simplify
 1.61% :     0.000003s :     18: predicate.depend_value_elim
 1.23% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.44% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.53% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      1: predicate.elim_not_effective
 0.34% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.30% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.27% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     18: predicate.environ_get_eliminate
 1.23% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.04% :     0.000000s :      1: predicate.fold_const_symbol
 1.19% :     0.000002s :     12: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 4.40% :     0.000008s :     28: predicate.inline
 3.15% :     0.000006s :     33: predicate.inline_without_move
 0.49% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.48% :     0.000003s :     12: predicate.less_batch_normalization
 1.73% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.58% :     0.000003s :     21: predicate.load_eliminater
 0.76% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.57% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.81% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.31% :     0.000002s :     18: predicate.merge_addn
 1.48% :     0.000003s :     18: predicate.minmaximum_grad
 0.68% :     0.000001s :      1: predicate.mutable_eliminate
 0.16% :     0.000000s :      1: predicate.opt_reshape
 2.03% :     0.000004s :     21: predicate.partial_eliminate
 1.56% :     0.000003s :     18: predicate.print_const_string_wrapper
 2.05% :     0.000004s :     18: predicate.reduce_eliminate
 1.56% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.53% :     0.000003s :     22: predicate.replace_applicator
 1.64% :     0.000003s :     33: predicate.replace_old_param
 0.11% :     0.000000s :      1: predicate.reset_defer_inline
 1.49% :     0.000003s :     18: predicate.reshape_eliminate
 1.56% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.36% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.44% :     0.000003s :     18: predicate.same_eliminate
 0.62% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.26% :     0.000000s :      2: predicate.special_op_eliminate
 1.24% :     0.000002s :     12: predicate.specialize_transform
 1.92% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.60% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.73% :     0.000005s :     26: predicate.switch_defer_inline
 2.38% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.39% :     0.000011s :     71: predicate.switch_simplify
 1.43% :     0.000003s :     18: predicate.tile_eliminate
 1.49% :     0.000003s :     18: predicate.transpose_eliminate
 1.63% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.47% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.94% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.74% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000003s :     18: predicate.value_based_eliminate
 0.10% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.24% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001298    22
72.33% :     0.000939s :     13: func_graph_cloner_run.FuncGraphClonerGraph
27.67% :     0.000359s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025837    76
 0.13% :     0.000034s :      1: add_recomputation
 0.39% :     0.000100s :      1: auto_monad
 0.06% :     0.000015s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.96% :     0.000247s :      1: bootstrap
 0.09% :     0.000023s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.21% :     0.000054s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
34.31% :     0.008865s :      1: jit_opt_a
 0.53% :     0.000137s :      1: jit_opt_after_cconv
 0.17% :     0.000045s :      1: jit_opt_b
 1.67% :     0.000431s :      1: loop_unroll
 2.20% :     0.000568s :      1: mutable_eliminate
 5.12% :     0.001322s :     26: opt.transform.jit_opt_a
 0.12% :     0.000032s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000023s :      4: opt.transform.symbol_engine_opt
 1.78% :     0.000459s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000028s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 5.29% :     0.001368s :      1: renormalize.infer
 1.78% :     0.000460s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000037s :      1: rewriter_after_opt_a
 0.36% :     0.000094s :      1: rewriter_before_opt_a
 0.26% :     0.000067s :      1: symbol_engine_optimizer
 0.10% :     0.000026s :      1: task_emit
43.30% :     0.011187s :      1: type_inference
 0.17% :     0.000045s :      1: validate


TotalTime = 0.0199467, [30]
[bootstrap]: 0.00018619
[type_inference]: 0.00964356
[event_method]: 3.917e-05
[auto_monad]: 8.069e-05
[graph_reusing]: 6.01e-06
[pre_auto_parallel]: 2.36998e-06
[py_interpret_to_execute]: 3.824e-05
[rewriter_before_opt_a]: 8.722e-05
[expand_dump_flag]: 2.48e-06
[jit_opt_a]: 0.00795233, [2]
    [Cycle 1]: 0.00441214, [27]
        [switch_simplify]: 5.948e-05
        [loop_unroll]: 3.865e-05
        [a_1]: 0.0008291
        [with_stream_mark]: 2.309e-05
        [recompute_prepare]: 1.796e-05
        [updatestate_depend_eliminate]: 7.97998e-06
        [updatestate_assign_eliminate]: 5.88002e-06
        [updatestate_loads_eliminate]: 5.66003e-06
        [parameter_eliminate]: 2.82002e-06
        [specialize_transform]: 1.384e-05
        [updatestate_useless_node_eliminater]: 1.235e-05
        [accelerated_algorithm]: 1.337e-05
        [meta_shard_fg_expand]: 3.03998e-06
        [get_grad_eliminate_]: 1.246e-05
        [merge_forward]: 7.35998e-06
        [cell_reuse_recompute_pass]: 9.50007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.414e-05
        [j_node_and_user_rematch]: 2.271e-05
        [meta_fg_expand]: 0.00152386
        [replace_old_param]: 5.223e-05
        [inline_without_move]: 5.211e-05
        [renormalize]: 0.00143864
        [add_forward_monad_depend]: 4.93001e-06
        [auto_monad_grad]: 1.67001e-06
        [auto_monad_eliminator]: 8.72e-06
        [cse]: 1.658e-05
        [replace_applicator]: 1.924e-05
    [Cycle 2]: 0.00027971, [27]
        [switch_simplify]: 5.02e-06
        [loop_unroll]: 3.74002e-06
        [a_1]: 4.053e-05
        [with_stream_mark]: 8.99e-06
        [recompute_prepare]: 3.95998e-06
        [updatestate_depend_eliminate]: 2.09e-06
        [updatestate_assign_eliminate]: 1.71e-06
        [updatestate_loads_eliminate]: 1.53002e-06
        [parameter_eliminate]: 1.72999e-06
        [specialize_transform]: 3.71001e-06
        [updatestate_useless_node_eliminater]: 3.23e-06
        [accelerated_algorithm]: 3.80998e-06
        [meta_shard_fg_expand]: 1.68002e-06
        [get_grad_eliminate_]: 3.63999e-06
        [merge_forward]: 2.25002e-06
        [cell_reuse_recompute_pass]: 1.97999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.035e-05
        [j_node_and_user_rematch]: 6.39001e-06
        [meta_fg_expand]: 2.556e-05
        [replace_old_param]: 3.77002e-06
        [inline_without_move]: 3.18e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.50999e-06
        [auto_monad_grad]: 8.99978e-07
        [auto_monad_eliminator]: 4.21001e-06
        [cse]: 1.122e-05
        [replace_applicator]: 4.20999e-06
[py_interpret_to_execute_after_opt_a]: 8.36002e-06
[rewriter_after_opt_a]: 2.657e-05
[convert_after_rewriter]: 5.30999e-06
[order_py_execute_after_rewriter]: 3.95e-06
[mutable_eliminate]: 0.00051439
[jit_opt_b]: 3.981e-05, [1]
    [Cycle 1]: 3.338e-05, [2]
        [frontend_op_eliminate]: 1.043e-05
        [inline_after_opt_a]: 1.136e-05
[cconv]: 1.736e-05
[loop_unroll]: 0.00041507
[jit_opt_after_cconv]: 0.00012086, [1]
    [Cycle 1]: 0.0001145, [11]
        [c_1]: 1.297e-05
        [parameter_eliminate]: 2.53e-06
        [updatestate_depend_eliminate]: 5.15999e-06
        [updatestate_assign_eliminate]: 1.94999e-06
        [updatestate_loads_eliminate]: 1.68002e-06
        [cse]: 1.662e-05
        [call_graph_tuple_transform]: 1.391e-05
        [tuple_list_get_item_eliminator]: 3.92002e-06
        [none_parameter_eliminate]: 1.19998e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 3.7e-06
[remove_dup_value]: 1.01e-05
[partial_unused_args_eliminate]: 2.04999e-06
[environ_conv]: 3.55e-06
[add_recomputation]: 2.437e-05
[cse_after_recomputation]: 1.967e-05, [1]
    [Cycle 1]: 1.433e-05, [1]
        [cse]: 8.55001e-06
[auto_monad_reorder]: 9.53002e-06
[get_jit_bprop_graph]: 1.35999e-06
[rewriter_after_jit_bprop_graph]: 3.78001e-06
[opt_after_jit_grad]: 0.00043851
[symbol_engine_optimizer]: 6.22e-05, [1]
    [Cycle 1]: 5.615e-05, [6]
        [build]: 2.63e-06
        [elim_shapecalc]: 5.72001e-06
        [elim_not_effective]: 9.75002e-06
        [opt_reshape]: 4.10998e-06
        [fold_const_symbol]: 5.84e-06
        [renormalize]: 6.00005e-07
[validate]: 2.435e-05
Sums
    bootstrap                            :     0.000186s :  1.16%
    type_inference                       :     0.009644s : 59.91%
    event_method                         :     0.000039s :  0.24%
    auto_monad                           :     0.000081s :  0.50%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000038s :  0.24%
    rewriter_before_opt_a                :     0.000087s :  0.54%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.40%
    jit_opt_a.loop_unroll                :     0.000042s :  0.26%
    jit_opt_a.a_1                        :     0.000870s :  5.40%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.03%
    jit_opt_a.specialize_transform       :     0.000018s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.10%
    jit_opt_a.merge_forward              :     0.000010s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001549s :  9.63%
    jit_opt_a.replace_old_param          :     0.000056s :  0.35%
    jit_opt_a.inline_without_move        :     0.000055s :  0.34%
    jit_opt_a.renormalize                :     0.001439s :  8.94%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.08%
    jit_opt_a.cse                        :     0.000028s :  0.17%
    jit_opt_a.replace_applicator         :     0.000023s :  0.15%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000027s :  0.17%
    convert_after_rewriter               :     0.000005s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000514s :  3.20%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000017s :  0.11%
    loop_unroll                          :     0.000415s :  2.58%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000010s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000024s :  0.15%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000010s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.02%
    opt_after_jit_grad                   :     0.000439s :  2.72%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000024s :  0.15%


Time group info:
------[substitution.]   0.000218    50
 0.67% :     0.000001s :      1: substitution.elim_not_effective
 0.33% :     0.000001s :      1: substitution.fold_const_symbol
 1.53% :     0.000003s :      1: substitution.graph_param_transform
64.97% :     0.000142s :      6: substitution.inline
 7.55% :     0.000016s :      2: substitution.inline_without_move
 2.88% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.61% :     0.000004s :      2: substitution.minmaximum_grad
 1.50% :     0.000003s :      3: substitution.partial_eliminate
 2.97% :     0.000006s :      9: substitution.remove_not_recompute_node
 2.24% :     0.000005s :      2: substitution.replace_applicator
 1.84% :     0.000004s :      5: substitution.replace_old_param
 1.20% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.92% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.71% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 6.10% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009584     2
88.19% :     0.008452s :      1: type_inference.infer
11.81% :     0.001132s :      1: type_inference.specialize
------[replace.]   0.000063     8
79.17% :     0.000050s :      6: replace.inline
20.83% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000147     8
94.25% :     0.000138s :      6: match.inline
 5.75% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000170  1109
 1.52% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.82% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.31% :     0.000002s :     18: predicate.addn_check_dump
 1.50% :     0.000003s :     18: predicate.addn_zero_filter
 2.04% :     0.000003s :     18: predicate.arithmetic_simplify
 1.44% :     0.000002s :     18: predicate.cast_eliminate
 0.13% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.28% :     0.000002s :     18: predicate.compare_switch_simplify
 1.46% :     0.000002s :     18: predicate.depend_value_elim
 1.32% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.53% :     0.000003s :     18: predicate.dict_set_item_eliminator
 0.43% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      1: predicate.elim_not_effective
 0.31% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.32% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.33% :     0.000002s :     18: predicate.environ_get_eliminate
 1.41% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.22% :     0.000002s :     12: predicate.get_grad_eliminate
 0.11% :     0.000000s :      1: predicate.graph_param_transform
 4.34% :     0.000007s :     28: predicate.inline
 3.35% :     0.000006s :     33: predicate.inline_without_move
 0.48% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.56% :     0.000003s :     12: predicate.less_batch_normalization
 1.87% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.71% :     0.000003s :     21: predicate.load_eliminater
 1.01% :     0.000002s :      1: predicate.loop_unroll_after_grad
 3.69% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.66% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.29% :     0.000002s :     18: predicate.merge_addn
 1.30% :     0.000002s :     18: predicate.minmaximum_grad
 0.95% :     0.000002s :      1: predicate.mutable_eliminate
 0.19% :     0.000000s :      1: predicate.opt_reshape
 2.04% :     0.000003s :     21: predicate.partial_eliminate
 1.39% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.83% :     0.000003s :     18: predicate.reduce_eliminate
 1.54% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.51% :     0.000003s :     22: predicate.replace_applicator
 1.67% :     0.000003s :     33: predicate.replace_old_param
 0.09% :     0.000000s :      1: predicate.reset_defer_inline
 1.45% :     0.000002s :     18: predicate.reshape_eliminate
 1.45% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.35% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     18: predicate.same_eliminate
 0.69% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.32% :     0.000001s :      2: predicate.special_op_eliminate
 1.30% :     0.000002s :     12: predicate.specialize_transform
 1.56% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.13% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.76% :     0.000005s :     26: predicate.switch_defer_inline
 2.45% :     0.000004s :     26: predicate.switch_layer_defer_inline
 7.05% :     0.000012s :     71: predicate.switch_simplify
 1.37% :     0.000002s :     18: predicate.tile_eliminate
 1.47% :     0.000002s :     18: predicate.transpose_eliminate
 1.67% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.53% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.66% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.64% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.12% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000003s :     18: predicate.value_based_eliminate
 0.12% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.25% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001145    22
70.71% :     0.000809s :     13: func_graph_cloner_run.FuncGraphClonerGraph
29.29% :     0.000335s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022646    72
 0.12% :     0.000027s :      1: add_recomputation
 0.37% :     0.000084s :      1: auto_monad
 0.05% :     0.000012s :      1: auto_monad_reorder
 0.87% :     0.000197s :      1: bootstrap
 0.09% :     0.000020s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.02% :     0.000006s :      1: environ_conv
 0.19% :     0.000044s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
35.13% :     0.007956s :      1: jit_opt_a
 0.55% :     0.000124s :      1: jit_opt_after_cconv
 0.19% :     0.000042s :      1: jit_opt_b
 1.87% :     0.000424s :      1: loop_unroll
 2.31% :     0.000523s :      1: mutable_eliminate
 5.45% :     0.001235s :     26: opt.transform.jit_opt_a
 0.14% :     0.000031s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000015s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000022s :      4: opt.transform.symbol_engine_opt
 1.97% :     0.000446s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.18% :     0.000041s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000013s :      1: remove_dup_value
 4.80% :     0.001088s :      1: renormalize.infer
 1.52% :     0.000344s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000030s :      1: rewriter_after_opt_a
 0.40% :     0.000090s :      1: rewriter_before_opt_a
 0.29% :     0.000065s :      1: symbol_engine_optimizer
42.63% :     0.009654s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.448.413 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c375ca0> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.448.733 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80efee0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.449.108 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.449.269 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.450.277 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843c1e50> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.450.762 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0111974, [30]
[bootstrap]: 0.00026756
[type_inference]: 0.00398532
[event_method]: 1.642e-05
[auto_monad]: 6.052e-05
[graph_reusing]: 5.15999e-06
[pre_auto_parallel]: 2.74999e-06
[py_interpret_to_execute]: 2.743e-05
[rewriter_before_opt_a]: 6.917e-05
[expand_dump_flag]: 2.86999e-06
[jit_opt_a]: 0.00480264, [2]
    [Cycle 1]: 0.0014128, [27]
        [switch_simplify]: 4.878e-05
        [loop_unroll]: 2.818e-05
        [a_1]: 0.00049019
        [with_stream_mark]: 1.652e-05
        [recompute_prepare]: 6.64001e-06
        [updatestate_depend_eliminate]: 3.14999e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 2.15002e-06
        [specialize_transform]: 5.77001e-06
        [updatestate_useless_node_eliminater]: 4.83001e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 2.09e-06
        [get_grad_eliminate_]: 4.73001e-06
        [merge_forward]: 3.71999e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.399e-05
        [j_node_and_user_rematch]: 8.08001e-06
        [meta_fg_expand]: 2.14999e-06
        [replace_old_param]: 8.48001e-06
        [inline_without_move]: 5.07e-06
        [renormalize]: 0.00051679
        [add_forward_monad_depend]: 4.77e-06
        [auto_monad_grad]: 2.11998e-06
        [auto_monad_eliminator]: 1.317e-05
        [cse]: 2.879e-05
        [replace_applicator]: 1.066e-05
    [Cycle 2]: 0.00030862, [27]
        [switch_simplify]: 5.62001e-06
        [loop_unroll]: 4.88001e-06
        [a_1]: 8.686e-05
        [with_stream_mark]: 7.63001e-06
        [recompute_prepare]: 5.04e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 1.90001e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 4.83001e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.18002e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.49002e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.33002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.204e-05
        [j_node_and_user_rematch]: 7.37997e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 6.82002e-06
        [inline_without_move]: 4.57e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 7.99977e-07
        [auto_monad_eliminator]: 4.62e-06
        [cse]: 9.69999e-06
        [replace_applicator]: 5.00001e-06
[py_interpret_to_execute_after_opt_a]: 9.32001e-06
[rewriter_after_opt_a]: 3.022e-05
[convert_after_rewriter]: 8.89e-06
[order_py_execute_after_rewriter]: 4.63001e-06
[mutable_eliminate]: 0.00046667
[jit_opt_b]: 4.662e-05, [1]
    [Cycle 1]: 3.94e-05, [2]
        [frontend_op_eliminate]: 1.507e-05
        [inline_after_opt_a]: 1.377e-05
[cconv]: 3.475e-05
[loop_unroll]: 0.00041532
[jit_opt_after_cconv]: 0.00013393, [1]
    [Cycle 1]: 0.00012764, [11]
        [c_1]: 1.931e-05
        [parameter_eliminate]: 2.46998e-06
        [updatestate_depend_eliminate]: 5.30001e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [cse]: 1.613e-05
        [call_graph_tuple_transform]: 1.804e-05
        [tuple_list_get_item_eliminator]: 5.02e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 9.5999e-07
        [switch_simplify]: 5.19e-06
[remove_dup_value]: 1.402e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 5.09e-06
[add_recomputation]: 4.744e-05
[cse_after_recomputation]: 1.919e-05, [1]
    [Cycle 1]: 1.385e-05, [1]
        [cse]: 8.34002e-06
[auto_monad_reorder]: 1.481e-05
[get_jit_bprop_graph]: 1.40001e-06
[rewriter_after_jit_bprop_graph]: 5.11002e-06
[opt_after_jit_grad]: 0.00044006
[symbol_engine_optimizer]: 7.049e-05, [1]
    [Cycle 1]: 6.431e-05, [6]
        [build]: 3.26999e-06
        [elim_shapecalc]: 7.17997e-06
        [elim_not_effective]: 1.201e-05
        [opt_reshape]: 5.59998e-06
        [fold_const_symbol]: 8.64e-06
        [renormalize]: 4.49974e-07
[validate]: 3.074e-05
Sums
    bootstrap                            :     0.000268s :  3.54%
    type_inference                       :     0.003985s : 52.73%
    event_method                         :     0.000016s :  0.22%
    auto_monad                           :     0.000061s :  0.80%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.04%
    py_interpret_to_execute              :     0.000027s :  0.36%
    rewriter_before_opt_a                :     0.000069s :  0.92%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000054s :  0.72%
    jit_opt_a.loop_unroll                :     0.000033s :  0.44%
    jit_opt_a.a_1                        :     0.000577s :  7.63%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000015s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000517s :  6.84%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.24%
    jit_opt_a.cse                        :     0.000038s :  0.51%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.12%
    rewriter_after_opt_a                 :     0.000030s :  0.40%
    convert_after_rewriter               :     0.000009s :  0.12%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000467s :  6.17%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.18%
    cconv                                :     0.000035s :  0.46%
    loop_unroll                          :     0.000415s :  5.49%
    jit_opt_after_cconv.c_1              :     0.000019s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.19%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000047s :  0.63%
    cse_after_recomputation.cse          :     0.000008s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.20%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000440s :  5.82%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000031s :  0.41%


Time group info:
------[substitution.]   0.000178    23
 0.98% :     0.000002s :      2: substitution.elim_not_effective
 0.91% :     0.000002s :      2: substitution.fold_const_symbol
 2.96% :     0.000005s :      3: substitution.graph_param_transform
82.76% :     0.000147s :      4: substitution.inline
 1.40% :     0.000002s :      4: substitution.j_node_and_user_rematch
 2.47% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000004s :      2: substitution.replace_old_param
 6.49% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003921     2
80.84% :     0.003170s :      1: type_inference.infer
19.16% :     0.000751s :      1: type_inference.specialize
------[replace.]   0.000060     6
67.61% :     0.000041s :      4: replace.inline
32.39% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000155     6
93.28% :     0.000145s :      4: match.inline
 6.72% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.61% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.16% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000001s :     10: predicate.addn_zero_filter
 2.11% :     0.000002s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.50% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.75% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.14% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.01% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.55% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 5.62% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.12% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.63% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000002s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.45% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.77% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.62% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.74% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 0.95% :     0.000001s :      6: predicate.specialize_transform
 1.63% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.57% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.45% :     0.000003s :     16: predicate.switch_defer_inline
 2.01% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.74% :     0.000009s :     53: predicate.switch_simplify
 1.20% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.26% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.49% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000530    12
49.72% :     0.000264s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.50% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.78% :     0.000237s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012544    72
 0.40% :     0.000051s :      1: add_recomputation
 0.51% :     0.000064s :      1: auto_monad
 0.14% :     0.000018s :      1: auto_monad_reorder
 2.26% :     0.000283s :      1: bootstrap
 0.30% :     0.000038s :      1: cconv
 0.09% :     0.000011s :      1: convert_after_rewriter
 0.17% :     0.000021s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000007s :      1: graph_reusing
38.31% :     0.004806s :      1: jit_opt_a
 1.09% :     0.000137s :      1: jit_opt_after_cconv
 0.39% :     0.000049s :      1: jit_opt_b
 3.37% :     0.000423s :      1: loop_unroll
 3.79% :     0.000475s :      1: mutable_eliminate
 6.16% :     0.000772s :     26: opt.transform.jit_opt_a
 0.35% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000023s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.57% :     0.000448s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.04% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000030s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000016s :      1: remove_dup_value
 2.07% :     0.000259s :      1: renormalize.infer
 2.00% :     0.000250s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000033s :      1: rewriter_after_opt_a
 0.57% :     0.000072s :      1: rewriter_before_opt_a
 0.58% :     0.000073s :      1: symbol_engine_optimizer
31.86% :     0.003996s :      1: type_inference


TotalTime = 0.0203506, [30]
[bootstrap]: 0.00018146
[type_inference]: 0.00930326
[event_method]: 4.169e-05
[auto_monad]: 6.439e-05
[graph_reusing]: 4.75001e-06
[pre_auto_parallel]: 2.11998e-06
[py_interpret_to_execute]: 2.313e-05
[rewriter_before_opt_a]: 8.263e-05
[expand_dump_flag]: 1.94999e-06
[jit_opt_a]: 0.00850937, [2]
    [Cycle 1]: 0.00444536, [27]
        [switch_simplify]: 6.008e-05
        [loop_unroll]: 5.176e-05
        [a_1]: 0.00087424
        [with_stream_mark]: 2.076e-05
        [recompute_prepare]: 1.808e-05
        [updatestate_depend_eliminate]: 7.72998e-06
        [updatestate_assign_eliminate]: 6.46e-06
        [updatestate_loads_eliminate]: 5.62999e-06
        [parameter_eliminate]: 1.97999e-06
        [specialize_transform]: 1.29e-05
        [updatestate_useless_node_eliminater]: 1.253e-05
        [accelerated_algorithm]: 1.243e-05
        [meta_shard_fg_expand]: 2.39001e-06
        [get_grad_eliminate_]: 1.218e-05
        [merge_forward]: 6.79001e-06
        [cell_reuse_recompute_pass]: 1.91998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 3.007e-05
        [j_node_and_user_rematch]: 2.241e-05
        [meta_fg_expand]: 0.00148108
        [replace_old_param]: 5.124e-05
        [inline_without_move]: 5.198e-05
        [renormalize]: 0.00144816
        [add_forward_monad_depend]: 6.36998e-06
        [auto_monad_grad]: 2.41e-06
        [auto_monad_eliminator]: 1.177e-05
        [cse]: 2.613e-05
        [replace_applicator]: 1.226e-05
    [Cycle 2]: 0.0002851, [27]
        [switch_simplify]: 4.69998e-06
        [loop_unroll]: 4.03001e-06
        [a_1]: 4.611e-05
        [with_stream_mark]: 9.06002e-06
        [recompute_prepare]: 3.91001e-06
        [updatestate_depend_eliminate]: 2.48e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.67001e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 3.38999e-06
        [updatestate_useless_node_eliminater]: 3.21001e-06
        [accelerated_algorithm]: 3.43e-06
        [meta_shard_fg_expand]: 1.66998e-06
        [get_grad_eliminate_]: 3.26001e-06
        [merge_forward]: 2.31e-06
        [cell_reuse_recompute_pass]: 2.32001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.114e-05
        [j_node_and_user_rematch]: 6.19999e-06
        [meta_fg_expand]: 3.324e-05
        [replace_old_param]: 3.77002e-06
        [inline_without_move]: 3.23e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 6.00005e-07
        [auto_monad_eliminator]: 4.13001e-06
        [cse]: 9.12001e-06
        [replace_applicator]: 3.55998e-06
[py_interpret_to_execute_after_opt_a]: 8.32e-06
[rewriter_after_opt_a]: 3.31e-05
[convert_after_rewriter]: 6.11e-06
[order_py_execute_after_rewriter]: 4.48999e-06
[mutable_eliminate]: 0.00071028
[jit_opt_b]: 3.953e-05, [1]
    [Cycle 1]: 3.298e-05, [2]
        [frontend_op_eliminate]: 1.074e-05
        [inline_after_opt_a]: 1.104e-05
[cconv]: 2.133e-05
[loop_unroll]: 0.00041833
[jit_opt_after_cconv]: 0.00011328, [1]
    [Cycle 1]: 0.00010739, [11]
        [c_1]: 1.127e-05
        [parameter_eliminate]: 1.96e-06
        [updatestate_depend_eliminate]: 3.87002e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.66e-06
        [cse]: 1.461e-05
        [call_graph_tuple_transform]: 1.252e-05
        [tuple_list_get_item_eliminator]: 3.81999e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 2.30008e-07
        [switch_simplify]: 3.65e-06
[remove_dup_value]: 1.457e-05
[partial_unused_args_eliminate]: 2.35002e-06
[environ_conv]: 4.80001e-06
[add_recomputation]: 3.34e-05
[cse_after_recomputation]: 1.953e-05, [1]
    [Cycle 1]: 1.409e-05, [1]
        [cse]: 7.75998e-06
[auto_monad_reorder]: 1.251e-05
[get_jit_bprop_graph]: 1.62001e-06
[rewriter_after_jit_bprop_graph]: 5.14e-06
[opt_after_jit_grad]: 0.00044313
[symbol_engine_optimizer]: 6.292e-05, [1]
    [Cycle 1]: 5.674e-05, [6]
        [build]: 2.81999e-06
        [elim_shapecalc]: 5.87001e-06
        [elim_not_effective]: 8.77999e-06
        [opt_reshape]: 4.12e-06
        [fold_const_symbol]: 5.91998e-06
        [renormalize]: 4.00003e-07
[validate]: 2.693e-05
Sums
    bootstrap                            :     0.000181s :  1.13%
    type_inference                       :     0.009303s : 58.18%
    event_method                         :     0.000042s :  0.26%
    auto_monad                           :     0.000064s :  0.40%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000023s :  0.14%
    rewriter_before_opt_a                :     0.000083s :  0.52%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000065s :  0.41%
    jit_opt_a.loop_unroll                :     0.000056s :  0.35%
    jit_opt_a.a_1                        :     0.000920s :  5.76%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000016s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.10%
    jit_opt_a.merge_forward              :     0.000009s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000041s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001514s :  9.47%
    jit_opt_a.replace_old_param          :     0.000055s :  0.34%
    jit_opt_a.inline_without_move        :     0.000055s :  0.35%
    jit_opt_a.renormalize                :     0.001448s :  9.06%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.10%
    jit_opt_a.cse                        :     0.000035s :  0.22%
    jit_opt_a.replace_applicator         :     0.000016s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000033s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000710s :  4.44%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000021s :  0.13%
    loop_unroll                          :     0.000418s :  2.62%
    jit_opt_after_cconv.c_1              :     0.000011s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000033s :  0.21%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000443s :  2.77%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.17%


Time group info:
------[substitution.]   0.000219    50
 0.65% :     0.000001s :      1: substitution.elim_not_effective
 0.39% :     0.000001s :      1: substitution.fold_const_symbol
 1.62% :     0.000004s :      1: substitution.graph_param_transform
65.19% :     0.000143s :      6: substitution.inline
 7.47% :     0.000016s :      2: substitution.inline_without_move
 2.51% :     0.000005s :      9: substitution.j_node_and_user_rematch
 1.30% :     0.000003s :      2: substitution.minmaximum_grad
 1.43% :     0.000003s :      3: substitution.partial_eliminate
 3.62% :     0.000008s :      9: substitution.remove_not_recompute_node
 4.26% :     0.000009s :      2: substitution.replace_applicator
 1.70% :     0.000004s :      5: substitution.replace_old_param
 1.00% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.47% :     0.000005s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.44% :     0.000003s :      2: substitution.tuple_list_get_item_depend_reorder
 4.95% :     0.000011s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009248     2
88.73% :     0.008205s :      1: type_inference.infer
11.27% :     0.001042s :      1: type_inference.specialize
------[replace.]   0.000060     8
77.18% :     0.000046s :      6: replace.inline
22.82% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000145     8
95.77% :     0.000139s :      6: match.inline
 4.23% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000172  1109
 1.61% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.51% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.29% :     0.000002s :     18: predicate.addn_check_dump
 1.57% :     0.000003s :     18: predicate.addn_zero_filter
 2.58% :     0.000004s :     18: predicate.arithmetic_simplify
 1.44% :     0.000002s :     18: predicate.cast_eliminate
 0.13% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.32% :     0.000002s :     18: predicate.compare_switch_simplify
 1.42% :     0.000002s :     18: predicate.depend_value_elim
 1.32% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.45% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.49% :     0.000003s :     18: predicate.dict_set_item_eliminator
 0.38% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.08% :     0.000000s :      1: predicate.elim_not_effective
 0.23% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.32% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.24% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.55% :     0.000003s :     18: predicate.environ_get_eliminate
 1.35% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.27% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.16% :     0.000007s :     28: predicate.inline
 3.31% :     0.000006s :     33: predicate.inline_without_move
 0.55% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :     12: predicate.less_batch_normalization
 1.59% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.64% :     0.000003s :     21: predicate.load_eliminater
 0.63% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.86% :     0.000007s :     44: predicate.loop_unroll_before_grad
 1.75% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.36% :     0.000002s :     18: predicate.merge_addn
 1.28% :     0.000002s :     18: predicate.minmaximum_grad
 0.71% :     0.000001s :      1: predicate.mutable_eliminate
 0.17% :     0.000000s :      1: predicate.opt_reshape
 2.01% :     0.000003s :     21: predicate.partial_eliminate
 1.45% :     0.000003s :     18: predicate.print_const_string_wrapper
 2.06% :     0.000004s :     18: predicate.reduce_eliminate
 1.77% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.74% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.45% :     0.000002s :     22: predicate.replace_applicator
 1.75% :     0.000003s :     33: predicate.replace_old_param
 0.10% :     0.000000s :      1: predicate.reset_defer_inline
 1.50% :     0.000003s :     18: predicate.reshape_eliminate
 1.42% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.35% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.46% :     0.000003s :     18: predicate.same_eliminate
 0.68% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.30% :     0.000001s :      2: predicate.special_op_eliminate
 1.18% :     0.000002s :     12: predicate.specialize_transform
 1.64% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.10% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.71% :     0.000005s :     26: predicate.switch_defer_inline
 2.35% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.72% :     0.000012s :     71: predicate.switch_simplify
 1.36% :     0.000002s :     18: predicate.tile_eliminate
 1.40% :     0.000002s :     18: predicate.transpose_eliminate
 1.74% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.65% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.07% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.63% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.93% :     0.000003s :     18: predicate.value_based_eliminate
 0.11% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.26% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001143    22
72.94% :     0.000834s :     13: func_graph_cloner_run.FuncGraphClonerGraph
27.06% :     0.000309s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023104    72
 0.16% :     0.000036s :      1: add_recomputation
 0.30% :     0.000069s :      1: auto_monad
 0.07% :     0.000015s :      1: auto_monad_reorder
 0.85% :     0.000196s :      1: bootstrap
 0.10% :     0.000024s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.19% :     0.000045s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
36.85% :     0.008513s :      1: jit_opt_a
 0.50% :     0.000116s :      1: jit_opt_after_cconv
 0.18% :     0.000042s :      1: jit_opt_b
 1.84% :     0.000425s :      1: loop_unroll
 3.10% :     0.000717s :      1: mutable_eliminate
 5.59% :     0.001292s :     26: opt.transform.jit_opt_a
 0.12% :     0.000028s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000015s :      4: opt.transform.jit_opt_b
 0.04% :     0.000008s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000010s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000013s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000021s :      4: opt.transform.symbol_engine_opt
 1.94% :     0.000449s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.11% :     0.000026s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 4.69% :     0.001083s :      1: renormalize.infer
 1.55% :     0.000358s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000036s :      1: rewriter_after_opt_a
 0.37% :     0.000086s :      1: rewriter_before_opt_a
 0.28% :     0.000065s :      1: symbol_engine_optimizer
40.31% :     0.009313s :      1: type_inference


TotalTime = 0.0226222, [33]
[bootstrap]: 0.00025541
[type_inference]: 0.0113569
[event_method]: 5.225e-05
[auto_monad]: 7.3e-05
[graph_reusing]: 5.20001e-06
[pre_auto_parallel]: 2.27999e-06
[py_interpret_to_execute]: 2.838e-05
[rewriter_before_opt_a]: 8.693e-05
[expand_dump_flag]: 2.36998e-06
[jit_opt_a]: 0.00877284, [2]
    [Cycle 1]: 0.00489453, [27]
        [switch_simplify]: 6.418e-05
        [loop_unroll]: 3.985e-05
        [a_1]: 0.0009171
        [with_stream_mark]: 2.085e-05
        [recompute_prepare]: 1.938e-05
        [updatestate_depend_eliminate]: 7.4e-06
        [updatestate_assign_eliminate]: 6.21e-06
        [updatestate_loads_eliminate]: 5.72999e-06
        [parameter_eliminate]: 1.59e-06
        [specialize_transform]: 1.392e-05
        [updatestate_useless_node_eliminater]: 1.244e-05
        [accelerated_algorithm]: 1.309e-05
        [meta_shard_fg_expand]: 3.59002e-06
        [get_grad_eliminate_]: 1.234e-05
        [merge_forward]: 7.1e-06
        [cell_reuse_recompute_pass]: 7.60017e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.359e-05
        [j_node_and_user_rematch]: 2.235e-05
        [meta_fg_expand]: 0.00151844
        [replace_old_param]: 5.503e-05
        [inline_without_move]: 5.392e-05
        [renormalize]: 0.00181725
        [add_forward_monad_depend]: 5.51e-06
        [auto_monad_grad]: 1.50001e-06
        [auto_monad_eliminator]: 9.39e-06
        [cse]: 1.777e-05
        [replace_applicator]: 1.416e-05
    [Cycle 2]: 0.00030884, [27]
        [switch_simplify]: 4.72e-06
        [loop_unroll]: 4.05998e-06
        [a_1]: 4.433e-05
        [with_stream_mark]: 9.00001e-06
        [recompute_prepare]: 4.1e-06
        [updatestate_depend_eliminate]: 2.23998e-06
        [updatestate_assign_eliminate]: 1.75001e-06
        [updatestate_loads_eliminate]: 1.57001e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 3.86001e-06
        [updatestate_useless_node_eliminater]: 3.98001e-06
        [accelerated_algorithm]: 7.98001e-06
        [meta_shard_fg_expand]: 1.52999e-06
        [get_grad_eliminate_]: 3.92998e-06
        [merge_forward]: 2.19001e-06
        [cell_reuse_recompute_pass]: 3.33998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.075e-05
        [j_node_and_user_rematch]: 6.24001e-06
        [meta_fg_expand]: 3.973e-05
        [replace_old_param]: 4.43999e-06
        [inline_without_move]: 3.27002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.62001e-06
        [auto_monad_grad]: 6.79982e-07
        [auto_monad_eliminator]: 5.14e-06
        [cse]: 1.118e-05
        [replace_applicator]: 4.20999e-06
[py_interpret_to_execute_after_opt_a]: 8.79e-06
[rewriter_after_opt_a]: 2.642e-05
[convert_after_rewriter]: 5.31998e-06
[order_py_execute_after_rewriter]: 3.7e-06
[mutable_eliminate]: 0.00055349
[jit_opt_b]: 4.071e-05, [1]
    [Cycle 1]: 3.415e-05, [2]
        [frontend_op_eliminate]: 1.099e-05
        [inline_after_opt_a]: 1.146e-05
[cconv]: 1.5e-05
[loop_unroll]: 0.00039654
[jit_opt_after_cconv]: 0.00011645, [1]
    [Cycle 1]: 0.00011029, [11]
        [c_1]: 1.248e-05
        [parameter_eliminate]: 2.26e-06
        [updatestate_depend_eliminate]: 4.53999e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.71e-06
        [cse]: 1.483e-05
        [call_graph_tuple_transform]: 1.222e-05
        [tuple_list_get_item_eliminator]: 3.95e-06
        [none_parameter_eliminate]: 1.13001e-06
        [renormalize]: 4.40021e-07
        [switch_simplify]: 3.91999e-06
[remove_dup_value]: 8.08999e-06
[partial_unused_args_eliminate]: 1.63002e-06
[environ_conv]: 3.75e-06
[add_recomputation]: 2.213e-05
[cse_after_recomputation]: 1.931e-05, [1]
    [Cycle 1]: 1.437e-05, [1]
        [cse]: 8.69e-06
[auto_monad_reorder]: 8.03001e-06
[get_jit_bprop_graph]: 1.64e-06
[rewriter_after_jit_bprop_graph]: 4.25e-06
[opt_after_jit_grad]: 0.00044652
[symbol_engine_optimizer]: 6.56e-05, [1]
    [Cycle 1]: 5.938e-05, [6]
        [build]: 3.04001e-06
        [elim_shapecalc]: 5.67999e-06
        [elim_not_effective]: 9.09998e-06
        [opt_reshape]: 4.85999e-06
        [fold_const_symbol]: 6.41998e-06
        [renormalize]: 7.7e-07
[validate]: 2.344e-05
[backend_pass]: 1.30999e-06
[task_emit]: 2.742e-05
[execute]: 1.17999e-06
Sums
    bootstrap                            :     0.000255s :  1.39%
    type_inference                       :     0.011357s : 61.68%
    event_method                         :     0.000052s :  0.28%
    auto_monad                           :     0.000073s :  0.40%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.15%
    rewriter_before_opt_a                :     0.000087s :  0.47%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000069s :  0.37%
    jit_opt_a.loop_unroll                :     0.000044s :  0.24%
    jit_opt_a.a_1                        :     0.000961s :  5.22%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000023s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000018s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000021s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.09%
    jit_opt_a.merge_forward              :     0.000009s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001558s :  8.46%
    jit_opt_a.replace_old_param          :     0.000059s :  0.32%
    jit_opt_a.inline_without_move        :     0.000057s :  0.31%
    jit_opt_a.renormalize                :     0.001817s :  9.87%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.08%
    jit_opt_a.cse                        :     0.000029s :  0.16%
    jit_opt_a.replace_applicator         :     0.000018s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.05%
    rewriter_after_opt_a                 :     0.000026s :  0.14%
    convert_after_rewriter               :     0.000005s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000553s :  3.01%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.06%
    cconv                                :     0.000015s :  0.08%
    loop_unroll                          :     0.000397s :  2.15%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000012s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000008s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000022s :  0.12%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000008s :  0.04%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.02%
    opt_after_jit_grad                   :     0.000447s :  2.42%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000023s :  0.13%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.000027s :  0.15%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000248    50
 0.39% :     0.000001s :      1: substitution.elim_not_effective
 0.63% :     0.000002s :      1: substitution.fold_const_symbol
 0.92% :     0.000002s :      1: substitution.graph_param_transform
66.81% :     0.000166s :      6: substitution.inline
 7.29% :     0.000018s :      2: substitution.inline_without_move
 2.40% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.15% :     0.000003s :      2: substitution.minmaximum_grad
 1.33% :     0.000003s :      3: substitution.partial_eliminate
 2.89% :     0.000007s :      9: substitution.remove_not_recompute_node
 1.97% :     0.000005s :      2: substitution.replace_applicator
 1.48% :     0.000004s :      5: substitution.replace_old_param
 0.93% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.91% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.53% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 7.39% :     0.000018s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011287     2
89.28% :     0.010077s :      1: type_inference.infer
10.72% :     0.001210s :      1: type_inference.specialize
------[replace.]   0.000067     8
75.57% :     0.000051s :      6: replace.inline
24.43% :     0.000016s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000175     8
92.24% :     0.000162s :      6: match.inline
 7.76% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000224  1109
 1.30% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.46% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.03% :     0.000002s :     18: predicate.addn_check_dump
 1.20% :     0.000003s :     18: predicate.addn_zero_filter
 1.62% :     0.000004s :     18: predicate.arithmetic_simplify
 1.20% :     0.000003s :     18: predicate.cast_eliminate
 0.10% :     0.000000s :      1: predicate.check_bprop_eliminate
 0.96% :     0.000002s :     18: predicate.compare_switch_simplify
 1.10% :     0.000002s :     18: predicate.depend_value_elim
 1.03% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.11% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.13% :     0.000003s :     18: predicate.dict_set_item_eliminator
 0.31% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.07% :     0.000000s :      1: predicate.elim_not_effective
 0.25% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000003s :     18: predicate.environ_add_const_eliminate
 0.98% :     0.000002s :     18: predicate.environ_get_add_eliminate
 0.96% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.07% :     0.000002s :     18: predicate.environ_get_eliminate
 0.96% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.04% :     0.000000s :      1: predicate.fold_const_symbol
 0.90% :     0.000002s :     12: predicate.get_grad_eliminate
 0.08% :     0.000000s :      1: predicate.graph_param_transform
 3.39% :     0.000008s :     28: predicate.inline
 2.58% :     0.000006s :     33: predicate.inline_without_move
 0.40% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.16% :     0.000003s :     12: predicate.less_batch_normalization
 1.29% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.49% :     0.000003s :     21: predicate.load_eliminater
 0.50% :     0.000001s :      1: predicate.loop_unroll_after_grad
 2.88% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.51% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000002s :     18: predicate.merge_addn
 1.06% :     0.000002s :     18: predicate.minmaximum_grad
 0.51% :     0.000001s :      1: predicate.mutable_eliminate
 0.22% :     0.000001s :      1: predicate.opt_reshape
 1.60% :     0.000004s :     21: predicate.partial_eliminate
 1.07% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.64% :     0.000004s :     18: predicate.reduce_eliminate
 1.14% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.20% :     0.000003s :     22: predicate.replace_applicator
 1.30% :     0.000003s :     33: predicate.replace_old_param
 0.08% :     0.000000s :      1: predicate.reset_defer_inline
 1.26% :     0.000003s :     18: predicate.reshape_eliminate
 1.16% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.40% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.04% :     0.000002s :     18: predicate.same_eliminate
 0.53% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.25% :     0.000001s :      2: predicate.special_op_eliminate
 0.98% :     0.000002s :     12: predicate.specialize_transform
 1.47% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.11% :     0.000000s :      1: predicate.switch_call_monad_eliminater
24.14% :     0.000054s :     26: predicate.switch_defer_inline
 1.82% :     0.000004s :     26: predicate.switch_layer_defer_inline
 4.94% :     0.000011s :     71: predicate.switch_simplify
 1.09% :     0.000002s :     18: predicate.tile_eliminate
 1.03% :     0.000002s :     18: predicate.transpose_eliminate
 1.25% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.14% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.36% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.49% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.24% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.21% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000004s :     18: predicate.value_based_eliminate
 0.09% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.17% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001297    22
71.48% :     0.000927s :     13: func_graph_cloner_run.FuncGraphClonerGraph
28.52% :     0.000370s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025832    76
 0.10% :     0.000025s :      1: add_recomputation
 0.30% :     0.000077s :      1: auto_monad
 0.04% :     0.000011s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.07% :     0.000276s :      1: bootstrap
 0.07% :     0.000018s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.08% :     0.000021s :      1: cse_after_recomputation
 0.02% :     0.000006s :      1: environ_conv
 0.22% :     0.000057s :      1: event_method
 0.02% :     0.000004s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
33.97% :     0.008776s :      1: jit_opt_a
 0.46% :     0.000119s :      1: jit_opt_after_cconv
 0.17% :     0.000043s :      1: jit_opt_b
 1.56% :     0.000403s :      1: loop_unroll
 2.18% :     0.000562s :      1: mutable_eliminate
 5.18% :     0.001338s :     26: opt.transform.jit_opt_a
 0.11% :     0.000029s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000014s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000022s :      4: opt.transform.symbol_engine_opt
 1.75% :     0.000453s :      1: opt_after_jit_grad
 0.02% :     0.000006s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000031s :      1: py_interpret_to_execute
 0.04% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000011s :      1: remove_dup_value
 5.01% :     0.001294s :      1: renormalize.infer
 1.99% :     0.000515s :      1: renormalize.specialize
 0.02% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000030s :      1: rewriter_after_opt_a
 0.35% :     0.000090s :      1: rewriter_before_opt_a
 0.26% :     0.000068s :      1: symbol_engine_optimizer
 0.12% :     0.000032s :      1: task_emit
44.01% :     0.011370s :      1: type_inference
 0.16% :     0.000040s :      1: validate

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.477.029 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843cd310> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.477.671 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.478.416 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c375700> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.479.016 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0179462, [33]
[bootstrap]: 0.00026824
[type_inference]: 0.00428441
[event_method]: 1.671e-05
[auto_monad]: 6.141e-05
[graph_reusing]: 5.66e-06
[pre_auto_parallel]: 2.63998e-06
[py_interpret_to_execute]: 3.078e-05
[rewriter_before_opt_a]: 6.852e-05
[expand_dump_flag]: 3.24001e-06
[jit_opt_a]: 0.00484647, [2]
    [Cycle 1]: 0.00133266, [27]
        [switch_simplify]: 4.097e-05
        [loop_unroll]: 2.98e-05
        [a_1]: 0.0004622
        [with_stream_mark]: 1.203e-05
        [recompute_prepare]: 6.62002e-06
        [updatestate_depend_eliminate]: 2.83e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.01998e-06
        [parameter_eliminate]: 1.14998e-06
        [specialize_transform]: 5.79999e-06
        [updatestate_useless_node_eliminater]: 5.50001e-06
        [accelerated_algorithm]: 5.05999e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 5.55001e-06
        [merge_forward]: 2.63e-06
        [cell_reuse_recompute_pass]: 1.00999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.223e-05
        [j_node_and_user_rematch]: 9.10001e-06
        [meta_fg_expand]: 1.76e-06
        [replace_old_param]: 8.13999e-06
        [inline_without_move]: 5.14e-06
        [renormalize]: 0.00051624
        [add_forward_monad_depend]: 4.54002e-06
        [auto_monad_grad]: 1.27e-06
        [auto_monad_eliminator]: 9.10001e-06
        [cse]: 1.6e-05
        [replace_applicator]: 9.77999e-06
    [Cycle 2]: 0.00031141, [27]
        [switch_simplify]: 5.30999e-06
        [loop_unroll]: 4.67e-06
        [a_1]: 8.754e-05
        [with_stream_mark]: 9.12999e-06
        [recompute_prepare]: 4.95999e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.95001e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 4.74002e-06
        [updatestate_useless_node_eliminater]: 5.28002e-06
        [accelerated_algorithm]: 4.79e-06
        [meta_shard_fg_expand]: 1.10999e-06
        [get_grad_eliminate_]: 4.79e-06
        [merge_forward]: 2.24001e-06
        [cell_reuse_recompute_pass]: 1.67999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.15e-05
        [j_node_and_user_rematch]: 7.74002e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 6.74001e-06
        [inline_without_move]: 4.67998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 6.79982e-07
        [auto_monad_eliminator]: 4.80999e-06
        [cse]: 1.01e-05
        [replace_applicator]: 4.84998e-06
[py_interpret_to_execute_after_opt_a]: 8.05999e-06
[rewriter_after_opt_a]: 2.652e-05
[convert_after_rewriter]: 5.81998e-06
[order_py_execute_after_rewriter]: 3.68e-06
[mutable_eliminate]: 0.00060475
[jit_opt_b]: 5.631e-05, [1]
    [Cycle 1]: 4.921e-05, [2]
        [frontend_op_eliminate]: 2.031e-05
        [inline_after_opt_a]: 1.605e-05
[cconv]: 1.612e-05
[loop_unroll]: 0.00043395
[jit_opt_after_cconv]: 0.00013904, [1]
    [Cycle 1]: 0.00013195, [11]
        [c_1]: 2.171e-05
        [parameter_eliminate]: 2.80002e-06
        [updatestate_depend_eliminate]: 6.04001e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [cse]: 1.789e-05
        [call_graph_tuple_transform]: 1.695e-05
        [tuple_list_get_item_eliminator]: 5.12e-06
        [none_parameter_eliminate]: 1.05999e-06
        [renormalize]: 5.79981e-07
        [switch_simplify]: 5.17e-06
[remove_dup_value]: 7.28999e-06
[partial_unused_args_eliminate]: 1.60999e-06
[environ_conv]: 3.49001e-06
[add_recomputation]: 3.547e-05
[cse_after_recomputation]: 2.025e-05, [1]
    [Cycle 1]: 1.482e-05, [1]
        [cse]: 9.02e-06
[auto_monad_reorder]: 1.319e-05
[get_jit_bprop_graph]: 2.09e-06
[rewriter_after_jit_bprop_graph]: 5.80002e-06
[opt_after_jit_grad]: 0.00045476
[symbol_engine_optimizer]: 7.008e-05, [1]
    [Cycle 1]: 6.374e-05, [6]
        [build]: 3.41001e-06
        [elim_shapecalc]: 7.31001e-06
        [elim_not_effective]: 1.239e-05
        [opt_reshape]: 5.26002e-06
        [fold_const_symbol]: 8.27998e-06
        [renormalize]: 5.09986e-07
[validate]: 3.199e-05
[backend_pass]: 9.50007e-07
[task_emit]: 0.00622604
[execute]: 9.13002e-06
Sums
    bootstrap                            :     0.000268s :  1.89%
    type_inference                       :     0.004284s : 30.23%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000061s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.22%
    rewriter_before_opt_a                :     0.000069s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000046s :  0.33%
    jit_opt_a.loop_unroll                :     0.000034s :  0.24%
    jit_opt_a.a_1                        :     0.000550s :  3.88%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.03%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000015s :  0.10%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000516s :  3.64%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.10%
    jit_opt_a.cse                        :     0.000026s :  0.18%
    jit_opt_a.replace_applicator         :     0.000015s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000027s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000605s :  4.27%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.14%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000016s :  0.11%
    loop_unroll                          :     0.000434s :  3.06%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000007s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000003s :  0.02%
    add_recomputation                    :     0.000035s :  0.25%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000013s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000455s :  3.21%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006226s : 43.92%
    execute                              :     0.000009s :  0.06%


Time group info:
------[substitution.]   0.000150    23
 1.41% :     0.000002s :      2: substitution.elim_not_effective
 1.01% :     0.000002s :      2: substitution.fold_const_symbol
 2.45% :     0.000004s :      3: substitution.graph_param_transform
83.22% :     0.000125s :      4: substitution.inline
 1.89% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.71% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.00% :     0.000003s :      2: substitution.replace_old_param
 5.30% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004231     2
82.21% :     0.003478s :      1: type_inference.infer
17.79% :     0.000753s :      1: type_inference.specialize
------[replace.]   0.000061     6
66.53% :     0.000040s :      4: replace.inline
33.47% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000130     6
94.68% :     0.000123s :      4: match.inline
 5.32% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.26% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.85% :     0.000002s :     10: predicate.addn_zero_filter
 2.35% :     0.000003s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.58% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000001s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.34% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.30% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.45% :     0.000002s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.39% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.76% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.90% :     0.000002s :     15: predicate.load_eliminater
 1.45% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.25% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.62% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000002s :     15: predicate.partial_eliminate
 1.28% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.46% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.41% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.31% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.31% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.74% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.13% :     0.000003s :     18: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.79% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.81% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000519    12
49.78% :     0.000258s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.48% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.74% :     0.000232s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019315    76
 0.20% :     0.000038s :      1: add_recomputation
 0.33% :     0.000064s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.46% :     0.000282s :      1: bootstrap
 0.10% :     0.000019s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.07% :     0.000013s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.10% :     0.004849s :      1: jit_opt_a
 0.73% :     0.000142s :      1: jit_opt_after_cconv
 0.30% :     0.000059s :      1: jit_opt_b
 2.29% :     0.000442s :      1: loop_unroll
 3.19% :     0.000615s :      1: mutable_eliminate
 3.84% :     0.000741s :     26: opt.transform.jit_opt_a
 0.24% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000030s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.40% :     0.000464s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.17% :     0.000034s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000009s :      1: remove_dup_value
 1.31% :     0.000253s :      1: renormalize.infer
 1.33% :     0.000258s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000030s :      1: rewriter_after_opt_a
 0.37% :     0.000071s :      1: rewriter_before_opt_a
 0.37% :     0.000072s :      1: symbol_engine_optimizer
32.30% :     0.006240s :      1: task_emit
22.23% :     0.004293s :      1: type_inference
 0.26% :     0.000050s :      1: validate


TotalTime = 0.0325089, [33]
[bootstrap]: 0.0001868
[type_inference]: 0.0102598
[event_method]: 4.997e-05
[auto_monad]: 0.00010053
[graph_reusing]: 7.37002e-06
[pre_auto_parallel]: 3.25e-06
[py_interpret_to_execute]: 2.65e-05
[rewriter_before_opt_a]: 9.401e-05
[expand_dump_flag]: 3.7e-06
[jit_opt_a]: 0.0123248, [3]
    [Cycle 1]: 0.0059817, [27]
        [switch_simplify]: 7.528e-05
        [loop_unroll]: 3.828e-05
        [a_1]: 0.00084017
        [with_stream_mark]: 2.813e-05
        [recompute_prepare]: 1.718e-05
        [updatestate_depend_eliminate]: 7.5e-06
        [updatestate_assign_eliminate]: 6.83e-06
        [updatestate_loads_eliminate]: 5.81e-06
        [parameter_eliminate]: 2.64999e-06
        [specialize_transform]: 1.245e-05
        [updatestate_useless_node_eliminater]: 1.125e-05
        [accelerated_algorithm]: 1.221e-05
        [meta_shard_fg_expand]: 3.8e-06
        [get_grad_eliminate_]: 1.213e-05
        [merge_forward]: 8.06001e-06
        [cell_reuse_recompute_pass]: 9.30013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.551e-05
        [j_node_and_user_rematch]: 2.274e-05
        [meta_fg_expand]: 0.00158569
        [replace_old_param]: 5.298e-05
        [inline_without_move]: 5.236e-05
        [renormalize]: 0.00265286
        [add_forward_monad_depend]: 9.46e-06
        [auto_monad_grad]: 4.99998e-06
        [auto_monad_eliminator]: 5.31e-05
        [cse]: 0.00015726
        [replace_applicator]: 7.326e-05
    [Cycle 2]: 0.00221215, [27]
        [switch_simplify]: 3.807e-05
        [loop_unroll]: 3.585e-05
        [a_1]: 0.00112603
        [with_stream_mark]: 1.261e-05
        [recompute_prepare]: 9.20001e-06
        [updatestate_depend_eliminate]: 3.72002e-06
        [updatestate_assign_eliminate]: 3.20002e-06
        [updatestate_loads_eliminate]: 3.05002e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 7.26999e-06
        [updatestate_useless_node_eliminater]: 6.74001e-06
        [accelerated_algorithm]: 7.26001e-06
        [meta_shard_fg_expand]: 2.07001e-06
        [get_grad_eliminate_]: 6.78998e-06
        [merge_forward]: 3.41999e-06
        [cell_reuse_recompute_pass]: 1.00999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.413e-05
        [j_node_and_user_rematch]: 1.177e-05
        [meta_fg_expand]: 4.946e-05
        [replace_old_param]: 1.067e-05
        [inline_without_move]: 7.7e-06
        [renormalize]: 0.00065101
        [add_forward_monad_depend]: 5.05999e-06
        [auto_monad_grad]: 1.31002e-06
        [auto_monad_eliminator]: 1.424e-05
        [cse]: 2.482e-05
        [replace_applicator]: 1.353e-05
    [Cycle 3]: 0.0004174, [27]
        [switch_simplify]: 7.51999e-06
        [loop_unroll]: 6.64001e-06
        [a_1]: 0.00014359
        [with_stream_mark]: 1.029e-05
        [recompute_prepare]: 7e-06
        [updatestate_depend_eliminate]: 3.64002e-06
        [updatestate_assign_eliminate]: 3.35e-06
        [updatestate_loads_eliminate]: 3.16999e-06
        [parameter_eliminate]: 1.42999e-06
        [specialize_transform]: 6.89999e-06
        [updatestate_useless_node_eliminater]: 6.73003e-06
        [accelerated_algorithm]: 7.08e-06
        [meta_shard_fg_expand]: 2.16e-06
        [get_grad_eliminate_]: 6.74999e-06
        [merge_forward]: 3.71001e-06
        [cell_reuse_recompute_pass]: 2.84001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.707e-05
        [j_node_and_user_rematch]: 1.208e-05
        [meta_fg_expand]: 2.43002e-06
        [replace_old_param]: 9.37001e-06
        [inline_without_move]: 6.77002e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 7.80012e-07
        [auto_monad_eliminator]: 8.50001e-06
        [cse]: 1.723e-05
        [replace_applicator]: 7.45e-06
[py_interpret_to_execute_after_opt_a]: 1.362e-05
[rewriter_after_opt_a]: 4.054e-05
[convert_after_rewriter]: 7.77e-06
[order_py_execute_after_rewriter]: 5.20001e-06
[mutable_eliminate]: 0.00074252
[jit_opt_b]: 6.291e-05, [1]
    [Cycle 1]: 5.565e-05, [2]
        [frontend_op_eliminate]: 2.29e-05
        [inline_after_opt_a]: 1.983e-05
[cconv]: 2.369e-05
[loop_unroll]: 0.00047077
[jit_opt_after_cconv]: 0.00018475, [1]
    [Cycle 1]: 0.00017653, [11]
        [c_1]: 3.067e-05
        [parameter_eliminate]: 3.65998e-06
        [updatestate_depend_eliminate]: 7.68999e-06
        [updatestate_assign_eliminate]: 4.68001e-06
        [updatestate_loads_eliminate]: 3.23e-06
        [cse]: 3.187e-05
        [call_graph_tuple_transform]: 2.553e-05
        [tuple_list_get_item_eliminator]: 7.68999e-06
        [none_parameter_eliminate]: 1.76998e-06
        [renormalize]: 5.19998e-07
        [switch_simplify]: 7.98001e-06
[remove_dup_value]: 1.664e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 6.54001e-06
[add_recomputation]: 7.766e-05
[cse_after_recomputation]: 3.199e-05, [1]
    [Cycle 1]: 2.557e-05, [1]
        [cse]: 1.894e-05
[auto_monad_reorder]: 1.521e-05
[get_jit_bprop_graph]: 1.28002e-06
[rewriter_after_jit_bprop_graph]: 7.17002e-06
[opt_after_jit_grad]: 0.00048507
[symbol_engine_optimizer]: 8.672e-05, [1]
    [Cycle 1]: 8.003e-05, [6]
        [build]: 4.09002e-06
        [elim_shapecalc]: 1.025e-05
        [elim_not_effective]: 1.638e-05
        [opt_reshape]: 8.15e-06
        [fold_const_symbol]: 1.307e-05
        [renormalize]: 4.69998e-07
[validate]: 3.606e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00692079
[execute]: 5.89e-06
Sums
    bootstrap                            :     0.000187s :  0.67%
    type_inference                       :     0.010260s : 36.64%
    event_method                         :     0.000050s :  0.18%
    auto_monad                           :     0.000101s :  0.36%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.09%
    rewriter_before_opt_a                :     0.000094s :  0.34%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000121s :  0.43%
    jit_opt_a.loop_unroll                :     0.000081s :  0.29%
    jit_opt_a.a_1                        :     0.002110s :  7.53%
    jit_opt_a.with_stream_mark           :     0.000051s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000027s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000027s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000026s :  0.09%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000057s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000047s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001638s :  5.85%
    jit_opt_a.replace_old_param          :     0.000073s :  0.26%
    jit_opt_a.inline_without_move        :     0.000067s :  0.24%
    jit_opt_a.renormalize                :     0.003304s : 11.80%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000076s :  0.27%
    jit_opt_a.cse                        :     0.000199s :  0.71%
    jit_opt_a.replace_applicator         :     0.000094s :  0.34%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.05%
    rewriter_after_opt_a                 :     0.000041s :  0.14%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000743s :  2.65%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.07%
    cconv                                :     0.000024s :  0.08%
    loop_unroll                          :     0.000471s :  1.68%
    jit_opt_after_cconv.c_1              :     0.000031s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000032s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000008s :  0.03%
    remove_dup_value                     :     0.000017s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000078s :  0.28%
    cse_after_recomputation.cse          :     0.000019s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.03%
    opt_after_jit_grad                   :     0.000485s :  1.73%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000036s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.006921s : 24.72%
    execute                              :     0.000006s :  0.02%


Time group info:
------[substitution.]   0.000530   118
 5.13% :     0.000027s :      1: substitution.arithmetic_simplify
 0.41% :     0.000002s :      4: substitution.elim_not_effective
 0.37% :     0.000002s :      4: substitution.fold_const_symbol
 1.18% :     0.000006s :      5: substitution.graph_param_transform
60.95% :     0.000323s :     12: substitution.inline
 3.17% :     0.000017s :      2: substitution.inline_without_move
 1.76% :     0.000009s :     16: substitution.j_node_and_user_rematch
 2.04% :     0.000011s :      7: substitution.minmaximum_grad
 1.43% :     0.000008s :      5: substitution.partial_eliminate
 2.16% :     0.000011s :     16: substitution.remove_not_recompute_node
 5.30% :     0.000028s :     10: substitution.replace_applicator
 1.53% :     0.000008s :      7: substitution.replace_old_param
 0.53% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.82% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.61% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.60% :     0.000040s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010178     2
88.87% :     0.009045s :      1: type_inference.infer
11.13% :     0.001133s :      1: type_inference.specialize
------[replace.]   0.000150    20
 3.42% :     0.000005s :      1: replace.arithmetic_simplify
62.46% :     0.000094s :     12: replace.inline
34.12% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000363    20
 7.26% :     0.000026s :      1: match.arithmetic_simplify
87.17% :     0.000317s :     12: match.inline
 5.56% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000434  2790
 1.46% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.41% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.26% :     0.000005s :     46: predicate.addn_check_dump
 1.53% :     0.000007s :     46: predicate.addn_zero_filter
 2.04% :     0.000009s :     47: predicate.arithmetic_simplify
 1.52% :     0.000007s :     47: predicate.cast_eliminate
 0.20% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.27% :     0.000006s :     46: predicate.compare_switch_simplify
 1.39% :     0.000006s :     46: predicate.depend_value_elim
 1.32% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.38% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.49% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.28% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.12% :     0.000001s :      5: predicate.elim_not_effective
 0.23% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.36% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.34% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.31% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.44% :     0.000006s :     47: predicate.environ_get_eliminate
 1.36% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      5: predicate.fold_const_symbol
 0.83% :     0.000004s :     21: predicate.get_grad_eliminate
 0.10% :     0.000000s :      5: predicate.graph_param_transform
 3.82% :     0.000017s :     76: predicate.inline
 1.57% :     0.000007s :     42: predicate.inline_without_move
 0.32% :     0.000001s :     21: predicate.j_node_and_user_rematch
 0.93% :     0.000004s :     21: predicate.less_batch_normalization
 1.62% :     0.000007s :     54: predicate.list_to_tuple_eliminator_
 1.87% :     0.000008s :     59: predicate.load_eliminater
 0.53% :     0.000002s :      5: predicate.loop_unroll_after_grad
 2.78% :     0.000012s :     89: predicate.loop_unroll_before_grad
 1.77% :     0.000008s :     52: predicate.make_slice_get_slice_eliminator
 1.28% :     0.000006s :     46: predicate.merge_addn
 1.34% :     0.000006s :     47: predicate.minmaximum_grad
10.11% :     0.000044s :      5: predicate.mutable_eliminate
 0.21% :     0.000001s :      5: predicate.opt_reshape
 2.09% :     0.000009s :     59: predicate.partial_eliminate
 1.35% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.67% :     0.000007s :     47: predicate.reduce_eliminate
 1.69% :     0.000007s :     54: predicate.redundant_stop_gradient_eliminater
 0.38% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.26% :     0.000010s :    105: predicate.replace_applicator
 0.79% :     0.000003s :     42: predicate.replace_old_param
 0.11% :     0.000000s :      5: predicate.reset_defer_inline
 1.65% :     0.000007s :     47: predicate.reshape_eliminate
 1.31% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.27% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.42% :     0.000006s :     46: predicate.same_eliminate
 0.43% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.40% :     0.000002s :     10: predicate.special_op_eliminate
 0.77% :     0.000003s :     21: predicate.specialize_transform
 1.49% :     0.000006s :     46: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.18% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.32% :     0.000010s :     66: predicate.switch_defer_inline
 2.17% :     0.000009s :     66: predicate.switch_layer_defer_inline
 5.55% :     0.000024s :    160: predicate.switch_simplify
 1.41% :     0.000006s :     47: predicate.tile_eliminate
 1.36% :     0.000006s :     47: predicate.transpose_eliminate
 1.69% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.52% :     0.000007s :     47: predicate.tuple_list_get_item_depend_reorder
 2.78% :     0.000012s :     64: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000008s :     54: predicate.tuple_to_list_eliminator_
 1.72% :     0.000007s :     59: predicate.updatestate_pure_node_eliminater
 2.63% :     0.000011s :     80: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000007s :     46: predicate.value_based_eliminate
 0.16% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.22% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001452    29
57.77% :     0.000839s :     13: func_graph_cloner_run.FuncGraphClonerGraph
42.23% :     0.000613s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038755    91
 0.21% :     0.000081s :      1: add_recomputation
 0.27% :     0.000105s :      1: auto_monad
 0.05% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.52% :     0.000200s :      1: bootstrap
 0.07% :     0.000026s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000034s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.14% :     0.000055s :      1: event_method
 0.02% :     0.000010s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000010s :      1: graph_reusing
31.81% :     0.012328s :      1: jit_opt_a
 0.48% :     0.000188s :      1: jit_opt_after_cconv
 0.17% :     0.000065s :      1: jit_opt_b
 1.24% :     0.000480s :      1: loop_unroll
 1.94% :     0.000753s :      1: mutable_eliminate
 7.08% :     0.002742s :     39: opt.transform.jit_opt_a
 0.17% :     0.000068s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000036s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.16% :     0.000061s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000044s :      4: opt.transform.symbol_engine_opt
 1.27% :     0.000494s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000029s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 4.56% :     0.001769s :      2: renormalize.infer
 3.93% :     0.001521s :      2: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000044s :      1: rewriter_after_opt_a
 0.25% :     0.000097s :      1: rewriter_before_opt_a
 0.23% :     0.000089s :      1: symbol_engine_optimizer
17.89% :     0.006934s :      1: task_emit
26.53% :     0.010281s :      1: type_inference
 0.15% :     0.000057s :      1: validate


TotalTime = 0.0111431, [30]
[bootstrap]: 0.00025482
[type_inference]: 0.00393324
[event_method]: 1.606e-05
[auto_monad]: 6.226e-05
[graph_reusing]: 5.96e-06
[pre_auto_parallel]: 2.34999e-06
[py_interpret_to_execute]: 3.035e-05
[rewriter_before_opt_a]: 6.982e-05
[expand_dump_flag]: 3.4e-06
[jit_opt_a]: 0.00479029, [2]
    [Cycle 1]: 0.00140836, [27]
        [switch_simplify]: 4.971e-05
        [loop_unroll]: 2.879e-05
        [a_1]: 0.0004955
        [with_stream_mark]: 1.626e-05
        [recompute_prepare]: 6.66999e-06
        [updatestate_depend_eliminate]: 3.29001e-06
        [updatestate_assign_eliminate]: 3.09001e-06
        [updatestate_loads_eliminate]: 2.48002e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 5.66e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.84998e-06
        [get_grad_eliminate_]: 4.66002e-06
        [merge_forward]: 3.26001e-06
        [cell_reuse_recompute_pass]: 1.02998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.339e-05
        [j_node_and_user_rematch]: 8.73001e-06
        [meta_fg_expand]: 1.97001e-06
        [replace_old_param]: 8.66002e-06
        [inline_without_move]: 5.53002e-06
        [renormalize]: 0.00051108
        [add_forward_monad_depend]: 4.50999e-06
        [auto_monad_grad]: 2.12999e-06
        [auto_monad_eliminator]: 1.336e-05
        [cse]: 2.963e-05
        [replace_applicator]: 1.061e-05
    [Cycle 2]: 0.00031118, [27]
        [switch_simplify]: 5.65001e-06
        [loop_unroll]: 4.85001e-06
        [a_1]: 8.734e-05
        [with_stream_mark]: 8.42e-06
        [recompute_prepare]: 5.02e-06
        [updatestate_depend_eliminate]: 2.43e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.84998e-06
        [updatestate_useless_node_eliminater]: 4.67998e-06
        [accelerated_algorithm]: 4.75001e-06
        [meta_shard_fg_expand]: 1.08001e-06
        [get_grad_eliminate_]: 4.25999e-06
        [merge_forward]: 2.32999e-06
        [cell_reuse_recompute_pass]: 1.22999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.159e-05
        [j_node_and_user_rematch]: 7.52002e-06
        [meta_fg_expand]: 1.40001e-06
        [replace_old_param]: 7.11001e-06
        [inline_without_move]: 4.60001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 4.95001e-06
        [cse]: 1.066e-05
        [replace_applicator]: 5.22e-06
[py_interpret_to_execute_after_opt_a]: 8.72e-06
[rewriter_after_opt_a]: 3.231e-05
[convert_after_rewriter]: 7.36001e-06
[order_py_execute_after_rewriter]: 4.79998e-06
[mutable_eliminate]: 0.00049944
[jit_opt_b]: 4.829e-05, [1]
    [Cycle 1]: 4.138e-05, [2]
        [frontend_op_eliminate]: 1.555e-05
        [inline_after_opt_a]: 1.486e-05
[cconv]: 2.376e-05
[loop_unroll]: 0.00040216
[jit_opt_after_cconv]: 0.00013054, [1]
    [Cycle 1]: 0.00012425, [11]
        [c_1]: 2.003e-05
        [parameter_eliminate]: 2.24001e-06
        [updatestate_depend_eliminate]: 4.90999e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 1.577e-05
        [call_graph_tuple_transform]: 1.752e-05
        [tuple_list_get_item_eliminator]: 4.95999e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 4.84e-06
[remove_dup_value]: 1.487e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 5.48002e-06
[add_recomputation]: 4.991e-05
[cse_after_recomputation]: 1.983e-05, [1]
    [Cycle 1]: 1.45e-05, [1]
        [cse]: 8.72998e-06
[auto_monad_reorder]: 1.453e-05
[get_jit_bprop_graph]: 1.79998e-06
[rewriter_after_jit_bprop_graph]: 5.24998e-06
[opt_after_jit_grad]: 0.00044411
[symbol_engine_optimizer]: 7.056e-05, [1]
    [Cycle 1]: 6.416e-05, [6]
        [build]: 3.56001e-06
        [elim_shapecalc]: 7.71001e-06
        [elim_not_effective]: 1.216e-05
        [opt_reshape]: 5.59e-06
        [fold_const_symbol]: 7.97e-06
        [renormalize]: 4.30009e-07
[validate]: 3.144e-05
Sums
    bootstrap                            :     0.000255s :  3.39%
    type_inference                       :     0.003933s : 52.30%
    event_method                         :     0.000016s :  0.21%
    auto_monad                           :     0.000062s :  0.83%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000030s :  0.40%
    rewriter_before_opt_a                :     0.000070s :  0.93%
    expand_dump_flag                     :     0.000003s :  0.05%
    jit_opt_a.switch_simplify            :     0.000055s :  0.74%
    jit_opt_a.loop_unroll                :     0.000034s :  0.45%
    jit_opt_a.a_1                        :     0.000583s :  7.75%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000511s :  6.80%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.24%
    jit_opt_a.cse                        :     0.000040s :  0.54%
    jit_opt_a.replace_applicator         :     0.000016s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.12%
    rewriter_after_opt_a                 :     0.000032s :  0.43%
    convert_after_rewriter               :     0.000007s :  0.10%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000499s :  6.64%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.20%
    cconv                                :     0.000024s :  0.32%
    loop_unroll                          :     0.000402s :  5.35%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.20%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000050s :  0.66%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000015s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000444s :  5.91%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000031s :  0.42%


Time group info:
------[substitution.]   0.000181    23
 1.01% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.71% :     0.000005s :      3: substitution.graph_param_transform
82.63% :     0.000150s :      4: substitution.inline
 1.56% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.36% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.52% :     0.000005s :      2: substitution.replace_old_param
 6.50% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003869     2
80.78% :     0.003125s :      1: type_inference.infer
19.22% :     0.000744s :      1: type_inference.specialize
------[replace.]   0.000062     6
67.85% :     0.000042s :      4: replace.inline
32.15% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000158     6
93.25% :     0.000148s :      4: match.inline
 6.75% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.65% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.79% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000001s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.15% :     0.000001s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000001s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.54% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.52% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.45% :     0.000001s :      3: predicate.graph_param_transform
 5.47% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000001s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 1.32% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.93% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.11% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.46% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000002s :     15: predicate.partial_eliminate
 1.43% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.41% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.61% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.52% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.63% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.56% :     0.000003s :     16: predicate.switch_defer_inline
 2.11% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.88% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.26% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000508    12
51.02% :     0.000259s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.42% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.56% :     0.000221s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.012492    72
 0.42% :     0.000053s :      1: add_recomputation
 0.52% :     0.000065s :      1: auto_monad
 0.14% :     0.000017s :      1: auto_monad_reorder
 2.17% :     0.000271s :      1: bootstrap
 0.21% :     0.000027s :      1: cconv
 0.08% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.07% :     0.000008s :      1: graph_reusing
38.36% :     0.004793s :      1: jit_opt_a
 1.07% :     0.000133s :      1: jit_opt_after_cconv
 0.41% :     0.000051s :      1: jit_opt_b
 3.28% :     0.000410s :      1: loop_unroll
 4.06% :     0.000507s :      1: mutable_eliminate
 6.24% :     0.000780s :     26: opt.transform.jit_opt_a
 0.35% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000024s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.62% :     0.000452s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000033s :      1: py_interpret_to_execute
 0.09% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.14% :     0.000017s :      1: remove_dup_value
 2.08% :     0.000260s :      1: renormalize.infer
 1.95% :     0.000244s :      1: renormalize.specialize
 0.06% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000036s :      1: rewriter_after_opt_a
 0.58% :     0.000073s :      1: rewriter_before_opt_a
 0.59% :     0.000073s :      1: symbol_engine_optimizer
31.58% :     0.003945s :      1: type_inference

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.490.283 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80ef9d0> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.490.930 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0215583, [30]
[bootstrap]: 0.00020397
[type_inference]: 0.0101322
[event_method]: 6.772e-05
[auto_monad]: 0.00010095
[graph_reusing]: 7.56999e-06
[pre_auto_parallel]: 2.83998e-06
[py_interpret_to_execute]: 2.885e-05
[rewriter_before_opt_a]: 9.468e-05
[expand_dump_flag]: 3.06001e-06
[jit_opt_a]: 0.00888341, [2]
    [Cycle 1]: 0.00467772, [27]
        [switch_simplify]: 7.655e-05
        [loop_unroll]: 4.359e-05
        [a_1]: 0.00089522
        [with_stream_mark]: 2.665e-05
        [recompute_prepare]: 1.817e-05
        [updatestate_depend_eliminate]: 8.37e-06
        [updatestate_assign_eliminate]: 6.49999e-06
        [updatestate_loads_eliminate]: 6.61999e-06
        [parameter_eliminate]: 2.81e-06
        [specialize_transform]: 1.306e-05
        [updatestate_useless_node_eliminater]: 1.221e-05
        [accelerated_algorithm]: 1.315e-05
        [meta_shard_fg_expand]: 3.68e-06
        [get_grad_eliminate_]: 1.173e-05
        [merge_forward]: 8.48999e-06
        [cell_reuse_recompute_pass]: 1.04998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.727e-05
        [j_node_and_user_rematch]: 2.238e-05
        [meta_fg_expand]: 0.00154457
        [replace_old_param]: 5.444e-05
        [inline_without_move]: 5.348e-05
        [renormalize]: 0.00154846
        [add_forward_monad_depend]: 5.08002e-06
        [auto_monad_grad]: 2.36998e-06
        [auto_monad_eliminator]: 1.237e-05
        [cse]: 2.693e-05
        [replace_applicator]: 1.251e-05
    [Cycle 2]: 0.00031541, [27]
        [switch_simplify]: 4.85999e-06
        [loop_unroll]: 3.85e-06
        [a_1]: 4.418e-05
        [with_stream_mark]: 9.79e-06
        [recompute_prepare]: 3.86999e-06
        [updatestate_depend_eliminate]: 2.52001e-06
        [updatestate_assign_eliminate]: 2.88e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.74998e-06
        [specialize_transform]: 3.66999e-06
        [updatestate_useless_node_eliminater]: 6.59999e-06
        [accelerated_algorithm]: 4.05e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 3.95998e-06
        [merge_forward]: 2.71e-06
        [cell_reuse_recompute_pass]: 1.86e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.438e-05
        [j_node_and_user_rematch]: 7.32002e-06
        [meta_fg_expand]: 3.224e-05
        [replace_old_param]: 4.04002e-06
        [inline_without_move]: 3.35e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.01998e-06
        [auto_monad_grad]: 8.60018e-07
        [auto_monad_eliminator]: 5.24e-06
        [cse]: 1.216e-05
        [replace_applicator]: 3.93001e-06
[py_interpret_to_execute_after_opt_a]: 9.65002e-06
[rewriter_after_opt_a]: 3.39e-05
[convert_after_rewriter]: 6.97002e-06
[order_py_execute_after_rewriter]: 4.63001e-06
[mutable_eliminate]: 0.00056687
[jit_opt_b]: 4.236e-05, [1]
    [Cycle 1]: 3.456e-05, [2]
        [frontend_op_eliminate]: 1.169e-05
        [inline_after_opt_a]: 1.155e-05
[cconv]: 2.234e-05
[loop_unroll]: 0.00042207
[jit_opt_after_cconv]: 0.00012228, [1]
    [Cycle 1]: 0.00011577, [11]
        [c_1]: 1.213e-05
        [parameter_eliminate]: 2.49001e-06
        [updatestate_depend_eliminate]: 4.94e-06
        [updatestate_assign_eliminate]: 1.93002e-06
        [updatestate_loads_eliminate]: 1.73002e-06
        [cse]: 1.808e-05
        [call_graph_tuple_transform]: 1.348e-05
        [tuple_list_get_item_eliminator]: 3.95e-06
        [none_parameter_eliminate]: 1.80001e-06
        [renormalize]: 7.99977e-07
        [switch_simplify]: 3.98001e-06
[remove_dup_value]: 1.479e-05
[partial_unused_args_eliminate]: 2.63e-06
[environ_conv]: 4.86997e-06
[add_recomputation]: 3.583e-05
[cse_after_recomputation]: 2.04e-05, [1]
    [Cycle 1]: 1.486e-05, [1]
        [cse]: 8.90999e-06
[auto_monad_reorder]: 1.479e-05
[get_jit_bprop_graph]: 1.84998e-06
[rewriter_after_jit_bprop_graph]: 5.52999e-06
[opt_after_jit_grad]: 0.00043947
[symbol_engine_optimizer]: 6.519e-05, [1]
    [Cycle 1]: 5.883e-05, [6]
        [build]: 3.03e-06
        [elim_shapecalc]: 6.12999e-06
        [elim_not_effective]: 9.97999e-06
        [opt_reshape]: 4.24997e-06
        [fold_const_symbol]: 5.92999e-06
        [renormalize]: 7.30011e-07
[validate]: 2.81e-05
Sums
    bootstrap                            :     0.000204s :  1.20%
    type_inference                       :     0.010132s : 59.47%
    event_method                         :     0.000068s :  0.40%
    auto_monad                           :     0.000101s :  0.59%
    graph_reusing                        :     0.000008s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.17%
    rewriter_before_opt_a                :     0.000095s :  0.56%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000081s :  0.48%
    jit_opt_a.loop_unroll                :     0.000047s :  0.28%
    jit_opt_a.a_1                        :     0.000939s :  5.51%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000009s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.03%
    jit_opt_a.specialize_transform       :     0.000017s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000019s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.09%
    jit_opt_a.merge_forward              :     0.000011s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000052s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000030s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001577s :  9.25%
    jit_opt_a.replace_old_param          :     0.000058s :  0.34%
    jit_opt_a.inline_without_move        :     0.000057s :  0.33%
    jit_opt_a.renormalize                :     0.001549s :  9.09%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.10%
    jit_opt_a.cse                        :     0.000039s :  0.23%
    jit_opt_a.replace_applicator         :     0.000016s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000034s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000567s :  3.33%
    jit_opt_b.frontend_op_eliminate      :     0.000012s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.07%
    cconv                                :     0.000022s :  0.13%
    loop_unroll                          :     0.000422s :  2.48%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000036s :  0.21%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000439s :  2.58%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000028s :  0.16%


Time group info:
------[substitution.]   0.000276    50
 0.62% :     0.000002s :      1: substitution.elim_not_effective
 0.34% :     0.000001s :      1: substitution.fold_const_symbol
 1.53% :     0.000004s :      1: substitution.graph_param_transform
67.50% :     0.000186s :      6: substitution.inline
 6.67% :     0.000018s :      2: substitution.inline_without_move
 2.46% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.40% :     0.000004s :      2: substitution.minmaximum_grad
 1.59% :     0.000004s :      3: substitution.partial_eliminate
 2.82% :     0.000008s :      9: substitution.remove_not_recompute_node
 2.32% :     0.000006s :      2: substitution.replace_applicator
 1.53% :     0.000004s :      5: substitution.replace_old_param
 1.01% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.75% :     0.000008s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.82% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.64% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010051     2
87.66% :     0.008810s :      1: type_inference.infer
12.34% :     0.001241s :      1: type_inference.specialize
------[replace.]   0.000069     8
77.89% :     0.000054s :      6: replace.inline
22.11% :     0.000015s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000193     8
94.41% :     0.000183s :      6: match.inline
 5.59% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000178  1109
 1.42% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.48% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.28% :     0.000002s :     18: predicate.addn_check_dump
 1.51% :     0.000003s :     18: predicate.addn_zero_filter
 2.28% :     0.000004s :     18: predicate.arithmetic_simplify
 1.58% :     0.000003s :     18: predicate.cast_eliminate
 0.15% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.21% :     0.000002s :     18: predicate.compare_switch_simplify
 1.42% :     0.000003s :     18: predicate.depend_value_elim
 1.21% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.57% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.38% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.10% :     0.000000s :      1: predicate.elim_not_effective
 0.24% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.41% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.31% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     18: predicate.environ_get_eliminate
 1.28% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.09% :     0.000002s :     12: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 4.18% :     0.000007s :     28: predicate.inline
 3.06% :     0.000005s :     33: predicate.inline_without_move
 0.48% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.67% :     0.000003s :     12: predicate.less_batch_normalization
 1.55% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 2.32% :     0.000004s :     21: predicate.load_eliminater
 0.53% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.79% :     0.000007s :     44: predicate.loop_unroll_before_grad
 1.69% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.29% :     0.000002s :     18: predicate.merge_addn
 1.37% :     0.000002s :     18: predicate.minmaximum_grad
 0.79% :     0.000001s :      1: predicate.mutable_eliminate
 0.19% :     0.000000s :      1: predicate.opt_reshape
 2.04% :     0.000004s :     21: predicate.partial_eliminate
 1.35% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.92% :     0.000003s :     18: predicate.reduce_eliminate
 1.76% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.65% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.72% :     0.000003s :     22: predicate.replace_applicator
 1.77% :     0.000003s :     33: predicate.replace_old_param
 0.08% :     0.000000s :      1: predicate.reset_defer_inline
 1.48% :     0.000003s :     18: predicate.reshape_eliminate
 1.41% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.42% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.47% :     0.000003s :     18: predicate.same_eliminate
 0.64% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.43% :     0.000001s :      2: predicate.special_op_eliminate
 1.17% :     0.000002s :     12: predicate.specialize_transform
 1.49% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.79% :     0.000005s :     26: predicate.switch_defer_inline
 2.33% :     0.000004s :     26: predicate.switch_layer_defer_inline
 7.11% :     0.000013s :     71: predicate.switch_simplify
 1.36% :     0.000002s :     18: predicate.tile_eliminate
 1.39% :     0.000002s :     18: predicate.transpose_eliminate
 1.69% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.54% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.38% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.53% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.05% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000003s :     18: predicate.value_based_eliminate
 0.15% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.40% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001182    22
70.85% :     0.000837s :     13: func_graph_cloner_run.FuncGraphClonerGraph
29.15% :     0.000345s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024450    72
 0.16% :     0.000039s :      1: add_recomputation
 0.43% :     0.000105s :      1: auto_monad
 0.07% :     0.000017s :      1: auto_monad_reorder
 0.89% :     0.000218s :      1: bootstrap
 0.10% :     0.000025s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.30% :     0.000074s :      1: event_method
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
36.35% :     0.008886s :      1: jit_opt_a
 0.51% :     0.000125s :      1: jit_opt_after_cconv
 0.18% :     0.000045s :      1: jit_opt_b
 1.76% :     0.000430s :      1: loop_unroll
 2.35% :     0.000575s :      1: mutable_eliminate
 5.42% :     0.001326s :     26: opt.transform.jit_opt_a
 0.12% :     0.000030s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000023s :      4: opt.transform.symbol_engine_opt
 1.83% :     0.000447s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000031s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 4.66% :     0.001140s :      1: renormalize.infer
 1.64% :     0.000400s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000037s :      1: rewriter_after_opt_a
 0.40% :     0.000097s :      1: rewriter_before_opt_a
 0.28% :     0.000068s :      1: symbol_engine_optimizer
41.51% :     0.010149s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.507.496 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843cd790> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.508.096 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0183161, [33]
[bootstrap]: 0.00028745
[type_inference]: 0.00446865
[event_method]: 1.631e-05
[auto_monad]: 6.163e-05
[graph_reusing]: 5.69999e-06
[pre_auto_parallel]: 2.43002e-06
[py_interpret_to_execute]: 3.088e-05
[rewriter_before_opt_a]: 7.139e-05
[expand_dump_flag]: 3.11999e-06
[jit_opt_a]: 0.00506995, [2]
    [Cycle 1]: 0.00141788, [27]
        [switch_simplify]: 4.427e-05
        [loop_unroll]: 3.031e-05
        [a_1]: 0.00049713
        [with_stream_mark]: 1.25e-05
        [recompute_prepare]: 7.01999e-06
        [updatestate_depend_eliminate]: 2.68e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 5.72999e-06
        [updatestate_useless_node_eliminater]: 5.23002e-06
        [accelerated_algorithm]: 5.22e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 4.90999e-06
        [merge_forward]: 2.71e-06
        [cell_reuse_recompute_pass]: 1.38002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.1e-05
        [j_node_and_user_rematch]: 8.79e-06
        [meta_fg_expand]: 1.74998e-06
        [replace_old_param]: 8.32e-06
        [inline_without_move]: 5.60001e-06
        [renormalize]: 0.00055406
        [add_forward_monad_depend]: 4.89e-06
        [auto_monad_grad]: 1.47001e-06
        [auto_monad_eliminator]: 1.044e-05
        [cse]: 1.657e-05
        [replace_applicator]: 1.167e-05
    [Cycle 2]: 0.0003139, [27]
        [switch_simplify]: 5.69e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 8.839e-05
        [with_stream_mark]: 8.99e-06
        [recompute_prepare]: 4.83001e-06
        [updatestate_depend_eliminate]: 2.56e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 4.76002e-06
        [updatestate_useless_node_eliminater]: 4.79998e-06
        [accelerated_algorithm]: 4.68999e-06
        [meta_shard_fg_expand]: 1.19e-06
        [get_grad_eliminate_]: 4.37e-06
        [merge_forward]: 2.61999e-06
        [cell_reuse_recompute_pass]: 1.60999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.274e-05
        [j_node_and_user_rematch]: 7.99997e-06
        [meta_fg_expand]: 1.47999e-06
        [replace_old_param]: 7.23e-06
        [inline_without_move]: 4.67e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 8.10018e-07
        [auto_monad_eliminator]: 5.10001e-06
        [cse]: 9.46e-06
        [replace_applicator]: 5.42999e-06
[py_interpret_to_execute_after_opt_a]: 8.82999e-06
[rewriter_after_opt_a]: 2.738e-05
[convert_after_rewriter]: 5.99999e-06
[order_py_execute_after_rewriter]: 4.52e-06
[mutable_eliminate]: 0.00052575
[jit_opt_b]: 4.928e-05, [1]
    [Cycle 1]: 4.261e-05, [2]
        [frontend_op_eliminate]: 1.601e-05
        [inline_after_opt_a]: 1.466e-05
[cconv]: 2.143e-05
[loop_unroll]: 0.00049759
[jit_opt_after_cconv]: 0.00013629, [1]
    [Cycle 1]: 0.00012997, [11]
        [c_1]: 2.094e-05
        [parameter_eliminate]: 2.58e-06
        [updatestate_depend_eliminate]: 5.33002e-06
        [updatestate_assign_eliminate]: 2.30002e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [cse]: 1.793e-05
        [call_graph_tuple_transform]: 1.744e-05
        [tuple_list_get_item_eliminator]: 5.27001e-06
        [none_parameter_eliminate]: 1.26997e-06
        [renormalize]: 5.50004e-07
        [switch_simplify]: 5.08002e-06
[remove_dup_value]: 9.57999e-06
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 4.52e-06
[add_recomputation]: 4.079e-05
[cse_after_recomputation]: 2.712e-05, [1]
    [Cycle 1]: 2.179e-05, [1]
        [cse]: 1.534e-05
[auto_monad_reorder]: 1.349e-05
[get_jit_bprop_graph]: 1.67001e-06
[rewriter_after_jit_bprop_graph]: 6.14001e-06
[opt_after_jit_grad]: 0.00047772
[symbol_engine_optimizer]: 7.051e-05, [1]
    [Cycle 1]: 6.445e-05, [6]
        [build]: 3.27002e-06
        [elim_shapecalc]: 7.31001e-06
        [elim_not_effective]: 1.168e-05
        [opt_reshape]: 5.54998e-06
        [fold_const_symbol]: 8.18001e-06
        [renormalize]: 1.09003e-06
[validate]: 3.054e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.00613664
[execute]: 7.41001e-06
Sums
    bootstrap                            :     0.000287s :  2.00%
    type_inference                       :     0.004469s : 31.04%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000062s :  0.43%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.21%
    rewriter_before_opt_a                :     0.000071s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000050s :  0.35%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000586s :  4.07%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.08%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000010s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.06%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.16%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.02%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000554s :  3.85%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.11%
    jit_opt_a.cse                        :     0.000026s :  0.18%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000027s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000526s :  3.65%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.10%
    cconv                                :     0.000021s :  0.15%
    loop_unroll                          :     0.000498s :  3.46%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000018s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.12%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000010s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000041s :  0.28%
    cse_after_recomputation.cse          :     0.000015s :  0.11%
    auto_monad_reorder                   :     0.000013s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000478s :  3.32%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.08%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000031s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006137s : 42.63%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000178    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.79% :     0.000005s :      3: substitution.graph_param_transform
82.98% :     0.000148s :      4: substitution.inline
 1.75% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.13% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.12% :     0.000004s :      2: substitution.replace_old_param
 6.50% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004411     2
81.93% :     0.003613s :      1: type_inference.infer
18.07% :     0.000797s :      1: type_inference.specialize
------[replace.]   0.000065     6
66.54% :     0.000043s :      4: replace.inline
33.46% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000156     6
93.25% :     0.000145s :      4: match.inline
 6.75% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000110   691
 1.32% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.08% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.12% :     0.000001s :     10: predicate.addn_check_dump
 1.84% :     0.000002s :     10: predicate.addn_zero_filter
 1.96% :     0.000002s :     10: predicate.arithmetic_simplify
 1.29% :     0.000001s :     10: predicate.cast_eliminate
 0.50% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000001s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.75% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.33% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.29% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.34% :     0.000006s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.69% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.86% :     0.000002s :     15: predicate.load_eliminater
 1.28% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.47% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.45% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000002s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.85% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.94% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.64% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000001s :     10: predicate.reshape_eliminate
 1.31% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 1.14% :     0.000001s :      6: predicate.specialize_transform
 1.38% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.65% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.41% :     0.000008s :     53: predicate.switch_simplify
 1.30% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.42% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.49% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.31% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.47% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000531    12
49.21% :     0.000261s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.55% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.23% :     0.000246s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019749    76
 0.22% :     0.000044s :      1: add_recomputation
 0.33% :     0.000065s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.54% :     0.000304s :      1: bootstrap
 0.12% :     0.000024s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.15% :     0.000029s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.05% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.69% :     0.005073s :      1: jit_opt_a
 0.70% :     0.000139s :      1: jit_opt_after_cconv
 0.26% :     0.000052s :      1: jit_opt_b
 2.56% :     0.000506s :      1: loop_unroll
 2.71% :     0.000535s :      1: mutable_eliminate
 3.96% :     0.000782s :     26: opt.transform.jit_opt_a
 0.23% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.12% :     0.000024s :      4: opt.transform.jit_opt_b
 0.06% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000029s :      4: opt.transform.symbol_engine_opt
 2.46% :     0.000486s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000034s :      1: py_interpret_to_execute
 0.06% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000012s :      1: remove_dup_value
 1.41% :     0.000278s :      1: renormalize.infer
 1.37% :     0.000270s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000030s :      1: rewriter_after_opt_a
 0.38% :     0.000075s :      1: rewriter_before_opt_a
 0.37% :     0.000073s :      1: symbol_engine_optimizer
31.12% :     0.006146s :      1: task_emit
22.69% :     0.004480s :      1: type_inference
 0.24% :     0.000048s :      1: validate


TotalTime = 0.0319754, [33]
[bootstrap]: 0.00020126
[type_inference]: 0.0199181
[event_method]: 5.149e-05
[auto_monad]: 0.00010126
[graph_reusing]: 7.01001e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 2.877e-05
[rewriter_before_opt_a]: 8.842e-05
[expand_dump_flag]: 2.83e-06
[jit_opt_a]: 0.00938388, [2]
    [Cycle 1]: 0.00516403, [27]
        [switch_simplify]: 7.078e-05
        [loop_unroll]: 4.335e-05
        [a_1]: 0.00092358
        [with_stream_mark]: 2.456e-05
        [recompute_prepare]: 1.942e-05
        [updatestate_depend_eliminate]: 7.66999e-06
        [updatestate_assign_eliminate]: 6.16e-06
        [updatestate_loads_eliminate]: 5.78997e-06
        [parameter_eliminate]: 2.42001e-06
        [specialize_transform]: 1.388e-05
        [updatestate_useless_node_eliminater]: 1.317e-05
        [accelerated_algorithm]: 1.888e-05
        [meta_shard_fg_expand]: 4.21001e-06
        [get_grad_eliminate_]: 1.301e-05
        [merge_forward]: 8.80999e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.722e-05
        [j_node_and_user_rematch]: 2.559e-05
        [meta_fg_expand]: 0.00175866
        [replace_old_param]: 5.727e-05
        [inline_without_move]: 5.548e-05
        [renormalize]: 0.00179914
        [add_forward_monad_depend]: 6.13998e-06
        [auto_monad_grad]: 2.00002e-06
        [auto_monad_eliminator]: 1.255e-05
        [cse]: 1.941e-05
        [replace_applicator]: 1.219e-05
    [Cycle 2]: 0.00030927, [27]
        [switch_simplify]: 4.77998e-06
        [loop_unroll]: 4.2e-06
        [a_1]: 4.451e-05
        [with_stream_mark]: 8.65999e-06
        [recompute_prepare]: 4.04997e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 1.77999e-06
        [updatestate_loads_eliminate]: 1.65001e-06
        [parameter_eliminate]: 1.74e-06
        [specialize_transform]: 4.07e-06
        [updatestate_useless_node_eliminater]: 3.71999e-06
        [accelerated_algorithm]: 4.05e-06
        [meta_shard_fg_expand]: 2.05002e-06
        [get_grad_eliminate_]: 3.36001e-06
        [merge_forward]: 2.26998e-06
        [cell_reuse_recompute_pass]: 2.46e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.142e-05
        [j_node_and_user_rematch]: 6.96999e-06
        [meta_fg_expand]: 4.037e-05
        [replace_old_param]: 4.35999e-06
        [inline_without_move]: 3.44001e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.58997e-06
        [auto_monad_grad]: 1.04003e-06
        [auto_monad_eliminator]: 4.62e-06
        [cse]: 1.129e-05
        [replace_applicator]: 4.25e-06
[py_interpret_to_execute_after_opt_a]: 9.62999e-06
[rewriter_after_opt_a]: 3.155e-05
[convert_after_rewriter]: 6.01998e-06
[order_py_execute_after_rewriter]: 4.62998e-06
[mutable_eliminate]: 0.00065958
[jit_opt_b]: 4.161e-05, [1]
    [Cycle 1]: 3.493e-05, [2]
        [frontend_op_eliminate]: 1.104e-05
        [inline_after_opt_a]: 1.143e-05
[cconv]: 2.083e-05
[loop_unroll]: 0.00044781
[jit_opt_after_cconv]: 0.00012264, [1]
    [Cycle 1]: 0.00011623, [11]
        [c_1]: 1.274e-05
        [parameter_eliminate]: 2.96999e-06
        [updatestate_depend_eliminate]: 5.06997e-06
        [updatestate_assign_eliminate]: 2.09999e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [cse]: 1.616e-05
        [call_graph_tuple_transform]: 1.453e-05
        [tuple_list_get_item_eliminator]: 3.98999e-06
        [none_parameter_eliminate]: 1.57999e-06
        [renormalize]: 4.39992e-07
        [switch_simplify]: 3.65e-06
[remove_dup_value]: 1.366e-05
[partial_unused_args_eliminate]: 1.97999e-06
[environ_conv]: 5.17e-06
[add_recomputation]: 3.153e-05
[cse_after_recomputation]: 2.058e-05, [1]
    [Cycle 1]: 1.501e-05, [1]
        [cse]: 8.85001e-06
[auto_monad_reorder]: 1.244e-05
[get_jit_bprop_graph]: 1.79e-06
[rewriter_after_jit_bprop_graph]: 5.34998e-06
[opt_after_jit_grad]: 0.0004577
[symbol_engine_optimizer]: 6.876e-05, [1]
    [Cycle 1]: 6.159e-05, [6]
        [build]: 3.03998e-06
        [elim_shapecalc]: 6.34999e-06
        [elim_not_effective]: 1.104e-05
        [opt_reshape]: 3.95e-06
        [fold_const_symbol]: 6.91001e-06
        [renormalize]: 4.00003e-07
[validate]: 2.688e-05
[backend_pass]: 8.79983e-07
[task_emit]: 1.955e-05
[execute]: 1.22999e-06
Sums
    bootstrap                            :     0.000201s :  0.73%
    type_inference                       :     0.019918s : 72.62%
    event_method                         :     0.000051s :  0.19%
    auto_monad                           :     0.000101s :  0.37%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000029s :  0.10%
    rewriter_before_opt_a                :     0.000088s :  0.32%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000076s :  0.28%
    jit_opt_a.loop_unroll                :     0.000048s :  0.17%
    jit_opt_a.a_1                        :     0.000968s :  3.53%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.12%
    jit_opt_a.recompute_prepare          :     0.000023s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000018s :  0.07%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000017s :  0.06%
    jit_opt_a.accelerated_algorithm      :     0.000023s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.06%
    jit_opt_a.merge_forward              :     0.000011s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.14%
    jit_opt_a.j_node_and_user_rematch    :     0.000033s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.001799s :  6.56%
    jit_opt_a.replace_old_param          :     0.000062s :  0.22%
    jit_opt_a.inline_without_move        :     0.000059s :  0.21%
    jit_opt_a.renormalize                :     0.001799s :  6.56%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.06%
    jit_opt_a.cse                        :     0.000031s :  0.11%
    jit_opt_a.replace_applicator         :     0.000016s :  0.06%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.04%
    rewriter_after_opt_a                 :     0.000032s :  0.12%
    convert_after_rewriter               :     0.000006s :  0.02%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000660s :  2.40%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.04%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.04%
    cconv                                :     0.000021s :  0.08%
    loop_unroll                          :     0.000448s :  1.63%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.05%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.06%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.05%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.01%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.01%
    remove_dup_value                     :     0.000014s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000032s :  0.11%
    cse_after_recomputation.cse          :     0.000009s :  0.03%
    auto_monad_reorder                   :     0.000012s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000458s :  1.67%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.02%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.04%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.01%
    symbol_engine_optimizer.fold_const_symbol :     0.000007s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.10%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000020s :  0.07%
    execute                              :     0.000001s :  0.00%


Time group info:
------[substitution.]   0.000280    50
 0.55% :     0.000002s :      1: substitution.elim_not_effective
 0.58% :     0.000002s :      1: substitution.fold_const_symbol
 1.63% :     0.000005s :      1: substitution.graph_param_transform
67.35% :     0.000189s :      6: substitution.inline
 6.29% :     0.000018s :      2: substitution.inline_without_move
 2.58% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.38% :     0.000004s :      2: substitution.minmaximum_grad
 1.63% :     0.000005s :      3: substitution.partial_eliminate
 2.55% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.37% :     0.000007s :      2: substitution.replace_applicator
 1.72% :     0.000005s :      5: substitution.replace_old_param
 1.02% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.54% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.90% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.92% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.019841     2
92.52% :     0.018356s :      1: type_inference.infer
 7.48% :     0.001485s :      1: type_inference.specialize
------[replace.]   0.000071     8
77.03% :     0.000054s :      6: replace.inline
22.97% :     0.000016s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000196     8
94.39% :     0.000185s :      6: match.inline
 5.61% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000186  1109
 1.63% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.61% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.31% :     0.000002s :     18: predicate.addn_check_dump
 1.51% :     0.000003s :     18: predicate.addn_zero_filter
 2.18% :     0.000004s :     18: predicate.arithmetic_simplify
 1.50% :     0.000003s :     18: predicate.cast_eliminate
 0.17% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.22% :     0.000002s :     18: predicate.compare_switch_simplify
 1.45% :     0.000003s :     18: predicate.depend_value_elim
 1.32% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.43% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.47% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.14% :     0.000000s :      1: predicate.elim_not_effective
 0.31% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.51% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.33% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.37% :     0.000003s :     18: predicate.environ_get_depend_swap
 1.29% :     0.000002s :     18: predicate.environ_get_eliminate
 1.22% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.07% :     0.000000s :      1: predicate.fold_const_symbol
 1.15% :     0.000002s :     12: predicate.get_grad_eliminate
 0.08% :     0.000000s :      1: predicate.graph_param_transform
 4.40% :     0.000008s :     28: predicate.inline
 3.32% :     0.000006s :     33: predicate.inline_without_move
 0.50% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.52% :     0.000003s :     12: predicate.less_batch_normalization
 1.53% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.75% :     0.000003s :     21: predicate.load_eliminater
 0.68% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.88% :     0.000007s :     44: predicate.loop_unroll_before_grad
 1.80% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.28% :     0.000002s :     18: predicate.merge_addn
 1.29% :     0.000002s :     18: predicate.minmaximum_grad
 0.73% :     0.000001s :      1: predicate.mutable_eliminate
 0.14% :     0.000000s :      1: predicate.opt_reshape
 1.95% :     0.000004s :     21: predicate.partial_eliminate
 1.33% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.88% :     0.000004s :     18: predicate.reduce_eliminate
 1.64% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.45% :     0.000003s :     22: predicate.replace_applicator
 1.64% :     0.000003s :     33: predicate.replace_old_param
 0.11% :     0.000000s :      1: predicate.reset_defer_inline
 1.46% :     0.000003s :     18: predicate.reshape_eliminate
 1.41% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.41% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.43% :     0.000003s :     18: predicate.same_eliminate
 0.66% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.31% :     0.000001s :      2: predicate.special_op_eliminate
 1.19% :     0.000002s :     12: predicate.specialize_transform
 1.77% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.92% :     0.000004s :     18: predicate.stack_unstack_eliminate
 0.13% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 3.31% :     0.000006s :     26: predicate.switch_defer_inline
 2.53% :     0.000005s :     26: predicate.switch_layer_defer_inline
 6.29% :     0.000012s :     71: predicate.switch_simplify
 1.51% :     0.000003s :     18: predicate.tile_eliminate
 1.33% :     0.000002s :     18: predicate.transpose_eliminate
 1.75% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.42% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.79% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.92% :     0.000004s :     18: predicate.tuple_list_set_item_eliminator
 1.83% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.58% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000006s :     33: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000003s :     18: predicate.value_based_eliminate
 0.10% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.31% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001416    22
67.04% :     0.000949s :     13: func_graph_cloner_run.FuncGraphClonerGraph
32.96% :     0.000467s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.035200    76
 0.10% :     0.000034s :      1: add_recomputation
 0.30% :     0.000105s :      1: auto_monad
 0.04% :     0.000015s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.60% :     0.000213s :      1: bootstrap
 0.07% :     0.000024s :      1: cconv
 0.02% :     0.000008s :      1: convert_after_rewriter
 0.06% :     0.000023s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.16% :     0.000055s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
26.67% :     0.009388s :      1: jit_opt_a
 0.36% :     0.000125s :      1: jit_opt_after_cconv
 0.13% :     0.000044s :      1: jit_opt_b
 1.29% :     0.000456s :      1: loop_unroll
 1.90% :     0.000668s :      1: mutable_eliminate
 3.88% :     0.001367s :     26: opt.transform.jit_opt_a
 0.09% :     0.000032s :      4: opt.transform.jit_opt_after_cconv
 0.04% :     0.000016s :      4: opt.transform.jit_opt_b
 0.03% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.07% :     0.000025s :      4: opt.transform.symbol_engine_opt
 1.32% :     0.000466s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.09% :     0.000031s :      1: py_interpret_to_execute
 0.03% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000016s :      1: remove_dup_value
 3.78% :     0.001330s :      1: renormalize.infer
 1.31% :     0.000460s :      1: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000035s :      1: rewriter_after_opt_a
 0.26% :     0.000091s :      1: rewriter_before_opt_a
 0.20% :     0.000071s :      1: symbol_engine_optimizer
 0.07% :     0.000024s :      1: task_emit
56.61% :     0.019928s :      1: type_inference
 0.12% :     0.000043s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.518.866 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c375310> should be wrapped by jit only once.

TotalTime = 0.0184462, [30]
[bootstrap]: 0.00025332
[type_inference]: 0.00405157
[event_method]: 1.69e-05
[auto_monad]: 6.115e-05
[graph_reusing]: 5.62001e-06
[pre_auto_parallel]: 2.21e-06
[py_interpret_to_execute]: 2.868e-05
[rewriter_before_opt_a]: 7.349e-05
[expand_dump_flag]: 3.58999e-06
[jit_opt_a]: 0.0117532, [2]
    [Cycle 1]: 0.00136558, [27]
        [switch_simplify]: 4.77e-05
        [loop_unroll]: 2.849e-05
        [a_1]: 0.00048592
        [with_stream_mark]: 1.627e-05
        [recompute_prepare]: 7.45998e-06
        [updatestate_depend_eliminate]: 3.46001e-06
        [updatestate_assign_eliminate]: 3.03998e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [parameter_eliminate]: 1.76998e-06
        [specialize_transform]: 5.69e-06
        [updatestate_useless_node_eliminater]: 4.81002e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.82001e-06
        [get_grad_eliminate_]: 4.89e-06
        [merge_forward]: 3.46001e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.373e-05
        [j_node_and_user_rematch]: 8.32e-06
        [meta_fg_expand]: 1.99999e-06
        [replace_old_param]: 7.8e-06
        [inline_without_move]: 5.02e-06
        [renormalize]: 0.00048005
        [add_forward_monad_depend]: 4.73001e-06
        [auto_monad_grad]: 1.64e-06
        [auto_monad_eliminator]: 1.315e-05
        [cse]: 2.62e-05
        [replace_applicator]: 1.053e-05
    [Cycle 2]: 0.0072943, [27]
        [switch_simplify]: 5.15999e-06
        [loop_unroll]: 1.918e-05
        [a_1]: 9.52e-05
        [with_stream_mark]: 1.945e-05
        [recompute_prepare]: 5.59998e-06
        [updatestate_depend_eliminate]: 3.10998e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [parameter_eliminate]: 1.74998e-06
        [specialize_transform]: 5.18002e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 5.89e-06
        [meta_shard_fg_expand]: 1.86998e-06
        [get_grad_eliminate_]: 4.37e-06
        [merge_forward]: 3.57002e-06
        [cell_reuse_recompute_pass]: 1.96e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.68e-05
        [j_node_and_user_rematch]: 8.62998e-06
        [meta_fg_expand]: 1.86998e-06
        [replace_old_param]: 1.561e-05
        [inline_without_move]: 4.78001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 2.66e-06
        [auto_monad_grad]: 1.57001e-06
        [auto_monad_eliminator]: 8.89e-06
        [cse]: 2.323e-05
        [replace_applicator]: 4.85001e-06
[py_interpret_to_execute_after_opt_a]: 1.743e-05
[rewriter_after_opt_a]: 3.718e-05
[convert_after_rewriter]: 6.83998e-06
[order_py_execute_after_rewriter]: 5.14e-06
[mutable_eliminate]: 0.00068973
[jit_opt_b]: 5.025e-05, [1]
    [Cycle 1]: 4.245e-05, [2]
        [frontend_op_eliminate]: 1.642e-05
        [inline_after_opt_a]: 1.463e-05
[cconv]: 3.059e-05
[loop_unroll]: 0.00041176
[jit_opt_after_cconv]: 0.0001356, [1]
    [Cycle 1]: 0.00012868, [11]
        [c_1]: 1.981e-05
        [parameter_eliminate]: 2.25002e-06
        [updatestate_depend_eliminate]: 5.60001e-06
        [updatestate_assign_eliminate]: 2.28998e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 1.77e-05
        [call_graph_tuple_transform]: 1.77e-05
        [tuple_list_get_item_eliminator]: 4.97999e-06
        [none_parameter_eliminate]: 1.25001e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 5.00001e-06
[remove_dup_value]: 1.409e-05
[partial_unused_args_eliminate]: 2.71e-06
[environ_conv]: 5.17999e-06
[add_recomputation]: 4.95e-05
[cse_after_recomputation]: 1.994e-05, [1]
    [Cycle 1]: 1.472e-05, [1]
        [cse]: 8.95001e-06
[auto_monad_reorder]: 1.371e-05
[get_jit_bprop_graph]: 1.82999e-06
[rewriter_after_jit_bprop_graph]: 5.52999e-06
[opt_after_jit_grad]: 0.00043368
[symbol_engine_optimizer]: 7.192e-05, [1]
    [Cycle 1]: 6.532e-05, [6]
        [build]: 4.1e-06
        [elim_shapecalc]: 7.28999e-06
        [elim_not_effective]: 1.194e-05
        [opt_reshape]: 5.60001e-06
        [fold_const_symbol]: 8.56002e-06
        [renormalize]: 3.69997e-07
[validate]: 3.361e-05
Sums
    bootstrap                            :     0.000253s :  3.22%
    type_inference                       :     0.004052s : 51.42%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000061s :  0.78%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000029s :  0.36%
    rewriter_before_opt_a                :     0.000073s :  0.93%
    expand_dump_flag                     :     0.000004s :  0.05%
    jit_opt_a.switch_simplify            :     0.000053s :  0.67%
    jit_opt_a.loop_unroll                :     0.000048s :  0.61%
    jit_opt_a.a_1                        :     0.000581s :  7.38%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.45%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.12%
    jit_opt_a.merge_forward              :     0.000007s :  0.09%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.39%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000023s :  0.30%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000480s :  6.09%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.28%
    jit_opt_a.cse                        :     0.000049s :  0.63%
    jit_opt_a.replace_applicator         :     0.000015s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.22%
    rewriter_after_opt_a                 :     0.000037s :  0.47%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000690s :  8.75%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000031s :  0.39%
    loop_unroll                          :     0.000412s :  5.23%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000018s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.18%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000050s :  0.63%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000014s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000434s :  5.50%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.43%


Time group info:
------[substitution.]   0.000177    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.86% :     0.000005s :      3: substitution.graph_param_transform
82.18% :     0.000145s :      4: substitution.inline
 1.94% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.85% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000004s :      2: substitution.replace_old_param
 6.39% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003994     2
80.73% :     0.003224s :      1: type_inference.infer
19.27% :     0.000770s :      1: type_inference.specialize
------[replace.]   0.000061     6
68.20% :     0.000041s :      4: replace.inline
31.80% :     0.000019s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000154     6
93.32% :     0.000143s :      4: match.inline
 6.68% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.73% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.23% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.25% :     0.000003s :     10: predicate.arithmetic_simplify
 1.14% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.14% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.14% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 6.30% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.47% :     0.000002s :      6: predicate.less_batch_normalization
 1.60% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.97% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.97% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.25% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.85% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000002s :     15: predicate.partial_eliminate
 1.24% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.61% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.09% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.71% :     0.000003s :     16: predicate.switch_defer_inline
 2.00% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.07% :     0.000008s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.38% :     0.000002s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.90% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 2.03% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.66% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000508    12
50.68% :     0.000257s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.78% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.54% :     0.000221s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019786    72
 0.26% :     0.000052s :      1: add_recomputation
 0.33% :     0.000065s :      1: auto_monad
 0.08% :     0.000016s :      1: auto_monad_reorder
 1.35% :     0.000268s :      1: bootstrap
 0.17% :     0.000033s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.10% :     0.000021s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
59.42% :     0.011757s :      1: jit_opt_a
 0.70% :     0.000139s :      1: jit_opt_after_cconv
 0.27% :     0.000053s :      1: jit_opt_b
 2.12% :     0.000419s :      1: loop_unroll
 3.53% :     0.000699s :      1: mutable_eliminate
 4.05% :     0.000802s :     26: opt.transform.jit_opt_a
 0.22% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.15% :     0.000030s :      4: opt.transform.symbol_engine_opt
 2.23% :     0.000441s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.16% :     0.000031s :      1: py_interpret_to_execute
 0.10% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.24% :     0.000244s :      1: renormalize.infer
 1.16% :     0.000229s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.20% :     0.000040s :      1: rewriter_after_opt_a
 0.39% :     0.000076s :      1: rewriter_before_opt_a
 0.38% :     0.000074s :      1: symbol_engine_optimizer
20.54% :     0.004064s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.519.493 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0321573, [33]
[bootstrap]: 0.0001947
[type_inference]: 0.00976372
[event_method]: 4.521e-05
[auto_monad]: 7.386e-05
[graph_reusing]: 4.95999e-06
[pre_auto_parallel]: 2.29001e-06
[py_interpret_to_execute]: 2.179e-05
[rewriter_before_opt_a]: 0.00018032
[expand_dump_flag]: 2.26e-06
[jit_opt_a]: 0.0117501, [3]
    [Cycle 1]: 0.0057304, [27]
        [switch_simplify]: 6.616e-05
        [loop_unroll]: 4.128e-05
        [a_1]: 0.00085919
        [with_stream_mark]: 2.742e-05
        [recompute_prepare]: 1.878e-05
        [updatestate_depend_eliminate]: 7.85e-06
        [updatestate_assign_eliminate]: 5.74e-06
        [updatestate_loads_eliminate]: 5.47001e-06
        [parameter_eliminate]: 2.12999e-06
        [specialize_transform]: 1.283e-05
        [updatestate_useless_node_eliminater]: 1.199e-05
        [accelerated_algorithm]: 1.289e-05
        [meta_shard_fg_expand]: 2.53e-06
        [get_grad_eliminate_]: 1.225e-05
        [merge_forward]: 7.13e-06
        [cell_reuse_recompute_pass]: 7.79983e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.477e-05
        [j_node_and_user_rematch]: 2.178e-05
        [meta_fg_expand]: 0.00137996
        [replace_old_param]: 4.901e-05
        [inline_without_move]: 5.118e-05
        [renormalize]: 0.00261748
        [add_forward_monad_depend]: 9.89999e-06
        [auto_monad_grad]: 4.99998e-06
        [auto_monad_eliminator]: 4.875e-05
        [cse]: 0.00015165
        [replace_applicator]: 6.871e-05
    [Cycle 2]: 0.00228544, [27]
        [switch_simplify]: 3.718e-05
        [loop_unroll]: 3.66e-05
        [a_1]: 0.00112843
        [with_stream_mark]: 1.295e-05
        [recompute_prepare]: 9.45001e-06
        [updatestate_depend_eliminate]: 4.45e-06
        [updatestate_assign_eliminate]: 3.28e-06
        [updatestate_loads_eliminate]: 3.09001e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 8.04997e-06
        [updatestate_useless_node_eliminater]: 8.02003e-06
        [accelerated_algorithm]: 8e-06
        [meta_shard_fg_expand]: 2.16e-06
        [get_grad_eliminate_]: 7.53999e-06
        [merge_forward]: 3.38e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.603e-05
        [j_node_and_user_rematch]: 1.245e-05
        [meta_fg_expand]: 4.834e-05
        [replace_old_param]: 1.108e-05
        [inline_without_move]: 7.26999e-06
        [renormalize]: 0.00070601
        [add_forward_monad_depend]: 4.84003e-06
        [auto_monad_grad]: 1.65001e-06
        [auto_monad_eliminator]: 1.392e-05
        [cse]: 2.563e-05
        [replace_applicator]: 1.451e-05
    [Cycle 3]: 0.00041591, [27]
        [switch_simplify]: 7.61999e-06
        [loop_unroll]: 7.25003e-06
        [a_1]: 0.00014688
        [with_stream_mark]: 9.77999e-06
        [recompute_prepare]: 6.80998e-06
        [updatestate_depend_eliminate]: 3.78001e-06
        [updatestate_assign_eliminate]: 3.21001e-06
        [updatestate_loads_eliminate]: 3.2e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 6.83998e-06
        [updatestate_useless_node_eliminater]: 6.71e-06
        [accelerated_algorithm]: 7.21001e-06
        [meta_shard_fg_expand]: 1.62999e-06
        [get_grad_eliminate_]: 6.85998e-06
        [merge_forward]: 3.34001e-06
        [cell_reuse_recompute_pass]: 1.94999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.541e-05
        [j_node_and_user_rematch]: 1.127e-05
        [meta_fg_expand]: 2.52001e-06
        [replace_old_param]: 8.59e-06
        [inline_without_move]: 6.59999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.12e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 8.13999e-06
        [cse]: 1.7e-05
        [replace_applicator]: 8.07e-06
[py_interpret_to_execute_after_opt_a]: 1.235e-05
[rewriter_after_opt_a]: 4.241e-05
[convert_after_rewriter]: 8.32e-06
[order_py_execute_after_rewriter]: 1.673e-05
[mutable_eliminate]: 0.00068605
[jit_opt_b]: 6.15e-05, [1]
    [Cycle 1]: 5.385e-05, [2]
        [frontend_op_eliminate]: 2.215e-05
        [inline_after_opt_a]: 1.98e-05
[cconv]: 2.403e-05
[loop_unroll]: 0.00042533
[jit_opt_after_cconv]: 0.00018271, [1]
    [Cycle 1]: 0.00016658, [11]
        [c_1]: 3.198e-05
        [parameter_eliminate]: 2.91999e-06
        [updatestate_depend_eliminate]: 6.81001e-06
        [updatestate_assign_eliminate]: 3.98999e-06
        [updatestate_loads_eliminate]: 3.14999e-06
        [cse]: 3.06e-05
        [call_graph_tuple_transform]: 2.282e-05
        [tuple_list_get_item_eliminator]: 6.93e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 6.79982e-07
        [switch_simplify]: 7.1e-06
[remove_dup_value]: 1.69e-05
[partial_unused_args_eliminate]: 2.83998e-06
[environ_conv]: 7.24001e-06
[add_recomputation]: 5.624e-05
[cse_after_recomputation]: 3.154e-05, [1]
    [Cycle 1]: 2.53e-05, [1]
        [cse]: 1.91e-05
[auto_monad_reorder]: 1.874e-05
[get_jit_bprop_graph]: 1.86e-06
[rewriter_after_jit_bprop_graph]: 5.02e-06
[opt_after_jit_grad]: 0.00046643
[symbol_engine_optimizer]: 8.672e-05, [1]
    [Cycle 1]: 8.035e-05, [6]
        [build]: 3.91001e-06
        [elim_shapecalc]: 1.044e-05
        [elim_not_effective]: 1.614e-05
        [opt_reshape]: 8.32e-06
        [fold_const_symbol]: 1.32e-05
        [renormalize]: 6.69999e-07
[validate]: 4.337e-05
[backend_pass]: 8.09989e-07
[task_emit]: 0.00771204
[execute]: 7.44002e-06
Sums
    bootstrap                            :     0.000195s :  0.69%
    type_inference                       :     0.009764s : 34.82%
    event_method                         :     0.000045s :  0.16%
    auto_monad                           :     0.000074s :  0.26%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000022s :  0.08%
    rewriter_before_opt_a                :     0.000180s :  0.64%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000111s :  0.40%
    jit_opt_a.loop_unroll                :     0.000085s :  0.30%
    jit_opt_a.a_1                        :     0.002134s :  7.61%
    jit_opt_a.with_stream_mark           :     0.000050s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000035s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000028s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000027s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000028s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000027s :  0.10%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000056s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000045s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001431s :  5.10%
    jit_opt_a.replace_old_param          :     0.000069s :  0.24%
    jit_opt_a.inline_without_move        :     0.000065s :  0.23%
    jit_opt_a.renormalize                :     0.003324s : 11.85%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000071s :  0.25%
    jit_opt_a.cse                        :     0.000194s :  0.69%
    jit_opt_a.replace_applicator         :     0.000091s :  0.33%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000042s :  0.15%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000017s :  0.06%
    mutable_eliminate                    :     0.000686s :  2.45%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.07%
    cconv                                :     0.000024s :  0.09%
    loop_unroll                          :     0.000425s :  1.52%
    jit_opt_after_cconv.c_1              :     0.000032s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000031s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000017s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000056s :  0.20%
    cse_after_recomputation.cse          :     0.000019s :  0.07%
    auto_monad_reorder                   :     0.000019s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000466s :  1.66%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000043s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007712s : 27.51%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000501   118
 5.53% :     0.000028s :      1: substitution.arithmetic_simplify
 0.55% :     0.000003s :      4: substitution.elim_not_effective
 0.55% :     0.000003s :      4: substitution.fold_const_symbol
 1.31% :     0.000007s :      5: substitution.graph_param_transform
60.88% :     0.000305s :     12: substitution.inline
 3.49% :     0.000017s :      2: substitution.inline_without_move
 1.74% :     0.000009s :     16: substitution.j_node_and_user_rematch
 1.95% :     0.000010s :      7: substitution.minmaximum_grad
 1.08% :     0.000005s :      5: substitution.partial_eliminate
 2.30% :     0.000012s :     16: substitution.remove_not_recompute_node
 5.03% :     0.000025s :     10: substitution.replace_applicator
 1.31% :     0.000007s :      7: substitution.replace_old_param
 0.59% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.93% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.51% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 7.25% :     0.000036s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009705     2
89.08% :     0.008646s :      1: type_inference.infer
10.92% :     0.001060s :      1: type_inference.specialize
------[replace.]   0.000146    20
 3.73% :     0.000005s :      1: replace.arithmetic_simplify
61.78% :     0.000090s :     12: replace.inline
34.50% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000342    20
 7.84% :     0.000027s :      1: match.arithmetic_simplify
87.06% :     0.000298s :     12: match.inline
 5.10% :     0.000017s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000399  2790
 1.55% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.49% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.38% :     0.000005s :     46: predicate.addn_check_dump
 1.77% :     0.000007s :     46: predicate.addn_zero_filter
 2.04% :     0.000008s :     47: predicate.arithmetic_simplify
 1.62% :     0.000006s :     47: predicate.cast_eliminate
 0.32% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.41% :     0.000006s :     46: predicate.compare_switch_simplify
 1.56% :     0.000006s :     46: predicate.depend_value_elim
 1.42% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.60% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.48% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.32% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      5: predicate.elim_not_effective
 0.26% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.53% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.48% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.44% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.52% :     0.000006s :     47: predicate.environ_get_eliminate
 1.39% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      5: predicate.fold_const_symbol
 1.04% :     0.000004s :     21: predicate.get_grad_eliminate
 0.11% :     0.000000s :      5: predicate.graph_param_transform
 4.17% :     0.000017s :     76: predicate.inline
 1.63% :     0.000007s :     42: predicate.inline_without_move
 0.36% :     0.000001s :     21: predicate.j_node_and_user_rematch
 1.23% :     0.000005s :     21: predicate.less_batch_normalization
 2.09% :     0.000008s :     54: predicate.list_to_tuple_eliminator_
 2.00% :     0.000008s :     59: predicate.load_eliminater
 0.62% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.32% :     0.000013s :     89: predicate.loop_unroll_before_grad
 1.80% :     0.000007s :     52: predicate.make_slice_get_slice_eliminator
 1.43% :     0.000006s :     46: predicate.merge_addn
 1.48% :     0.000006s :     47: predicate.minmaximum_grad
 0.69% :     0.000003s :      5: predicate.mutable_eliminate
 0.23% :     0.000001s :      5: predicate.opt_reshape
 2.31% :     0.000009s :     59: predicate.partial_eliminate
 1.47% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.80% :     0.000007s :     47: predicate.reduce_eliminate
 1.77% :     0.000007s :     54: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.54% :     0.000010s :    105: predicate.replace_applicator
 0.86% :     0.000003s :     42: predicate.replace_old_param
 0.15% :     0.000001s :      5: predicate.reset_defer_inline
 1.69% :     0.000007s :     47: predicate.reshape_eliminate
 1.53% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.33% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.54% :     0.000006s :     46: predicate.same_eliminate
 0.47% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.44% :     0.000002s :     10: predicate.special_op_eliminate
 0.88% :     0.000004s :     21: predicate.specialize_transform
 1.72% :     0.000007s :     46: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.20% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.58% :     0.000010s :     66: predicate.switch_defer_inline
 2.41% :     0.000010s :     66: predicate.switch_layer_defer_inline
 6.12% :     0.000024s :    160: predicate.switch_simplify
 1.53% :     0.000006s :     47: predicate.tile_eliminate
 1.57% :     0.000006s :     47: predicate.transpose_eliminate
 1.80% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.63% :     0.000007s :     47: predicate.tuple_list_get_item_depend_reorder
 2.94% :     0.000012s :     64: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.90% :     0.000008s :     54: predicate.tuple_to_list_eliminator_
 1.91% :     0.000008s :     59: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000012s :     80: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000007s :     46: predicate.value_based_eliminate
 0.20% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.27% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001494    29
54.92% :     0.000820s :     13: func_graph_cloner_run.FuncGraphClonerGraph
45.08% :     0.000673s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038395    91
 0.15% :     0.000059s :      1: add_recomputation
 0.20% :     0.000078s :      1: auto_monad
 0.06% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.55% :     0.000211s :      1: bootstrap
 0.07% :     0.000027s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000034s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.13% :     0.000049s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000004s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
30.61% :     0.011754s :      1: jit_opt_a
 0.48% :     0.000186s :      1: jit_opt_after_cconv
 0.17% :     0.000064s :      1: jit_opt_b
 1.13% :     0.000433s :      1: loop_unroll
 1.81% :     0.000694s :      1: mutable_eliminate
 7.19% :     0.002761s :     39: opt.transform.jit_opt_a
 0.17% :     0.000065s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000036s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000044s :      4: opt.transform.symbol_engine_opt
 1.24% :     0.000475s :      1: opt_after_jit_grad
 0.05% :     0.000019s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.06% :     0.000024s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000020s :      1: remove_dup_value
 4.68% :     0.001796s :      2: renormalize.infer
 3.95% :     0.001515s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000045s :      1: rewriter_after_opt_a
 0.48% :     0.000184s :      1: rewriter_before_opt_a
 0.23% :     0.000089s :      1: symbol_engine_optimizer
20.11% :     0.007723s :      1: task_emit
25.46% :     0.009776s :      1: type_inference
 0.18% :     0.000068s :      1: validate


TotalTime = 0.0202993, [30]
[bootstrap]: 0.00019093
[type_inference]: 0.00952409
[event_method]: 4.344e-05
[auto_monad]: 9.357e-05
[graph_reusing]: 6.65002e-06
[pre_auto_parallel]: 2.75997e-06
[py_interpret_to_execute]: 2.373e-05
[rewriter_before_opt_a]: 8.796e-05
[expand_dump_flag]: 2.98e-06
[jit_opt_a]: 0.00818458, [2]
    [Cycle 1]: 0.00456831, [27]
        [switch_simplify]: 6.385e-05
        [loop_unroll]: 3.941e-05
        [a_1]: 0.00083961
        [with_stream_mark]: 2.484e-05
        [recompute_prepare]: 1.81e-05
        [updatestate_depend_eliminate]: 7.75e-06
        [updatestate_assign_eliminate]: 6.21998e-06
        [updatestate_loads_eliminate]: 6.14001e-06
        [parameter_eliminate]: 2.41998e-06
        [specialize_transform]: 1.293e-05
        [updatestate_useless_node_eliminater]: 1.197e-05
        [accelerated_algorithm]: 1.249e-05
        [meta_shard_fg_expand]: 3.33e-06
        [get_grad_eliminate_]: 1.277e-05
        [merge_forward]: 8.01001e-06
        [cell_reuse_recompute_pass]: 1.12999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.632e-05
        [j_node_and_user_rematch]: 2.232e-05
        [meta_fg_expand]: 0.00141407
        [replace_old_param]: 5.287e-05
        [inline_without_move]: 5.272e-05
        [renormalize]: 0.00166475
        [add_forward_monad_depend]: 6.46e-06
        [auto_monad_grad]: 2.37999e-06
        [auto_monad_eliminator]: 1.314e-05
        [cse]: 2.614e-05
        [replace_applicator]: 1.249e-05
    [Cycle 2]: 0.00030879, [27]
        [switch_simplify]: 5.31002e-06
        [loop_unroll]: 4.4e-06
        [a_1]: 4.689e-05
        [with_stream_mark]: 9.58002e-06
        [recompute_prepare]: 4.20999e-06
        [updatestate_depend_eliminate]: 2.46998e-06
        [updatestate_assign_eliminate]: 2.30002e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 4.1e-06
        [updatestate_useless_node_eliminater]: 4.07e-06
        [accelerated_algorithm]: 7.3e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 4.35e-06
        [merge_forward]: 2.55002e-06
        [cell_reuse_recompute_pass]: 2.31e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.264e-05
        [j_node_and_user_rematch]: 6.69999e-06
        [meta_fg_expand]: 2.976e-05
        [replace_old_param]: 4.35999e-06
        [inline_without_move]: 3.93999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.68002e-06
        [auto_monad_grad]: 1.02998e-06
        [auto_monad_eliminator]: 4.57e-06
        [cse]: 1.084e-05
        [replace_applicator]: 4.13001e-06
[py_interpret_to_execute_after_opt_a]: 8.60999e-06
[rewriter_after_opt_a]: 3.424e-05
[convert_after_rewriter]: 6.63003e-06
[order_py_execute_after_rewriter]: 5.25001e-06
[mutable_eliminate]: 0.00060802
[jit_opt_b]: 4.247e-05, [1]
    [Cycle 1]: 3.489e-05, [2]
        [frontend_op_eliminate]: 1.014e-05
        [inline_after_opt_a]: 1.278e-05
[cconv]: 2.423e-05
[loop_unroll]: 0.00044832
[jit_opt_after_cconv]: 0.00012732, [1]
    [Cycle 1]: 0.00012045, [11]
        [c_1]: 1.243e-05
        [parameter_eliminate]: 3.10998e-06
        [updatestate_depend_eliminate]: 5.31998e-06
        [updatestate_assign_eliminate]: 1.91e-06
        [updatestate_loads_eliminate]: 1.74e-06
        [cse]: 1.929e-05
        [call_graph_tuple_transform]: 1.507e-05
        [tuple_list_get_item_eliminator]: 3.97e-06
        [none_parameter_eliminate]: 1.61002e-06
        [renormalize]: 1.01002e-06
        [switch_simplify]: 3.85e-06
[remove_dup_value]: 1.498e-05
[partial_unused_args_eliminate]: 2.56e-06
[environ_conv]: 5.50001e-06
[add_recomputation]: 3.51e-05
[cse_after_recomputation]: 2.021e-05, [1]
    [Cycle 1]: 1.423e-05, [1]
        [cse]: 8.15999e-06
[auto_monad_reorder]: 1.417e-05
[get_jit_bprop_graph]: 1.75001e-06
[rewriter_after_jit_bprop_graph]: 5.60001e-06
[opt_after_jit_grad]: 0.00047164
[symbol_engine_optimizer]: 6.764e-05, [1]
    [Cycle 1]: 6.053e-05, [6]
        [build]: 3.45998e-06
        [elim_shapecalc]: 6.66999e-06
        [elim_not_effective]: 1.078e-05
        [opt_reshape]: 4.40999e-06
        [fold_const_symbol]: 6.26998e-06
        [renormalize]: 5.50004e-07
[validate]: 3.232e-05
Sums
    bootstrap                            :     0.000191s :  1.17%
    type_inference                       :     0.009524s : 58.15%
    event_method                         :     0.000043s :  0.27%
    auto_monad                           :     0.000094s :  0.57%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000024s :  0.14%
    rewriter_before_opt_a                :     0.000088s :  0.54%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000069s :  0.42%
    jit_opt_a.loop_unroll                :     0.000044s :  0.27%
    jit_opt_a.a_1                        :     0.000886s :  5.41%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000017s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000020s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.10%
    jit_opt_a.merge_forward              :     0.000011s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001444s :  8.82%
    jit_opt_a.replace_old_param          :     0.000057s :  0.35%
    jit_opt_a.inline_without_move        :     0.000057s :  0.35%
    jit_opt_a.renormalize                :     0.001665s : 10.17%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.11%
    jit_opt_a.cse                        :     0.000037s :  0.23%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.05%
    rewriter_after_opt_a                 :     0.000034s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000608s :  3.71%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000013s :  0.08%
    cconv                                :     0.000024s :  0.15%
    loop_unroll                          :     0.000448s :  2.74%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000035s :  0.21%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000472s :  2.88%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.20%


Time group info:
------[substitution.]   0.000257    50
 0.59% :     0.000002s :      1: substitution.elim_not_effective
 0.42% :     0.000001s :      1: substitution.fold_const_symbol
 1.86% :     0.000005s :      1: substitution.graph_param_transform
66.15% :     0.000170s :      6: substitution.inline
 6.93% :     0.000018s :      2: substitution.inline_without_move
 2.62% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.33% :     0.000003s :      2: substitution.minmaximum_grad
 1.57% :     0.000004s :      3: substitution.partial_eliminate
 2.83% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.24% :     0.000006s :      2: substitution.replace_applicator
 1.60% :     0.000004s :      5: substitution.replace_old_param
 1.07% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.87% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.74% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 6.18% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009460     2
87.54% :     0.008280s :      1: type_inference.infer
12.46% :     0.001179s :      1: type_inference.specialize
------[replace.]   0.000064     8
79.34% :     0.000051s :      6: replace.inline
20.66% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000177     8
93.84% :     0.000166s :      6: match.inline
 6.16% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000175  1109
 1.52% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.73% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.38% :     0.000002s :     18: predicate.addn_check_dump
 1.60% :     0.000003s :     18: predicate.addn_zero_filter
 2.22% :     0.000004s :     18: predicate.arithmetic_simplify
 1.41% :     0.000002s :     18: predicate.cast_eliminate
 0.11% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.27% :     0.000002s :     18: predicate.compare_switch_simplify
 1.46% :     0.000003s :     18: predicate.depend_value_elim
 1.26% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.39% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.48% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      1: predicate.elim_not_effective
 0.27% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.36% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.27% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.26% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.46% :     0.000003s :     18: predicate.environ_get_eliminate
 1.27% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.40% :     0.000002s :     12: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 4.91% :     0.000009s :     28: predicate.inline
 3.08% :     0.000005s :     33: predicate.inline_without_move
 0.49% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :     12: predicate.less_batch_normalization
 1.63% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.72% :     0.000003s :     21: predicate.load_eliminater
 0.98% :     0.000002s :      1: predicate.loop_unroll_after_grad
 3.64% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.88% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.26% :     0.000002s :     18: predicate.merge_addn
 1.37% :     0.000002s :     18: predicate.minmaximum_grad
 0.95% :     0.000002s :      1: predicate.mutable_eliminate
 0.21% :     0.000000s :      1: predicate.opt_reshape
 2.08% :     0.000004s :     21: predicate.partial_eliminate
 1.42% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.75% :     0.000003s :     18: predicate.reduce_eliminate
 1.60% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.67% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.46% :     0.000003s :     22: predicate.replace_applicator
 1.70% :     0.000003s :     33: predicate.replace_old_param
 0.10% :     0.000000s :      1: predicate.reset_defer_inline
 1.43% :     0.000003s :     18: predicate.reshape_eliminate
 1.63% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.30% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     18: predicate.same_eliminate
 0.63% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.31% :     0.000001s :      2: predicate.special_op_eliminate
 1.22% :     0.000002s :     12: predicate.specialize_transform
 1.46% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.14% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.68% :     0.000005s :     26: predicate.switch_defer_inline
 2.32% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.61% :     0.000012s :     71: predicate.switch_simplify
 1.47% :     0.000003s :     18: predicate.tile_eliminate
 1.44% :     0.000003s :     18: predicate.transpose_eliminate
 1.70% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.44% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.24% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.86% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.64% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.51% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.04% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000003s :     18: predicate.value_based_eliminate
 0.10% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.29% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001182    22
66.96% :     0.000792s :     13: func_graph_cloner_run.FuncGraphClonerGraph
33.04% :     0.000391s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023242    72
 0.16% :     0.000038s :      1: add_recomputation
 0.42% :     0.000097s :      1: auto_monad
 0.07% :     0.000017s :      1: auto_monad_reorder
 0.87% :     0.000202s :      1: bootstrap
 0.12% :     0.000027s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.20% :     0.000047s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
35.23% :     0.008188s :      1: jit_opt_a
 0.56% :     0.000130s :      1: jit_opt_after_cconv
 0.19% :     0.000045s :      1: jit_opt_b
 1.96% :     0.000456s :      1: loop_unroll
 2.65% :     0.000617s :      1: mutable_eliminate
 5.41% :     0.001257s :     26: opt.transform.jit_opt_a
 0.14% :     0.000032s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000015s :      4: opt.transform.jit_opt_b
 0.05% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000025s :      4: opt.transform.symbol_engine_opt
 2.07% :     0.000480s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000027s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 5.35% :     0.001245s :      1: renormalize.infer
 1.77% :     0.000412s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000037s :      1: rewriter_after_opt_a
 0.39% :     0.000091s :      1: rewriter_before_opt_a
 0.30% :     0.000070s :      1: symbol_engine_optimizer
41.02% :     0.009535s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.534.374 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843cdc10> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.534.956 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.536.607 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80ef430> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.537.184 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0176417, [33]
[bootstrap]: 0.00024576
[type_inference]: 0.00401431
[event_method]: 1.696e-05
[auto_monad]: 5.83e-05
[graph_reusing]: 5.23002e-06
[pre_auto_parallel]: 2.55002e-06
[py_interpret_to_execute]: 2.994e-05
[rewriter_before_opt_a]: 7.04e-05
[expand_dump_flag]: 3.43e-06
[jit_opt_a]: 0.00504031, [2]
    [Cycle 1]: 0.00153815, [27]
        [switch_simplify]: 4.738e-05
        [loop_unroll]: 3.267e-05
        [a_1]: 0.00053412
        [with_stream_mark]: 1.478e-05
        [recompute_prepare]: 7.28e-06
        [updatestate_depend_eliminate]: 3.18e-06
        [updatestate_assign_eliminate]: 2.40002e-06
        [updatestate_loads_eliminate]: 2.13998e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 6.17999e-06
        [updatestate_useless_node_eliminater]: 5.12999e-06
        [accelerated_algorithm]: 5.54e-06
        [meta_shard_fg_expand]: 1.95001e-06
        [get_grad_eliminate_]: 5.23002e-06
        [merge_forward]: 3.14999e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.221e-05
        [j_node_and_user_rematch]: 8.72e-06
        [meta_fg_expand]: 2.19999e-06
        [replace_old_param]: 9.09e-06
        [inline_without_move]: 5.24e-06
        [renormalize]: 0.00060879
        [add_forward_monad_depend]: 4.80001e-06
        [auto_monad_grad]: 2.29001e-06
        [auto_monad_eliminator]: 1.13e-05
        [cse]: 2.08e-05
        [replace_applicator]: 1.242e-05
    [Cycle 2]: 0.00031542, [27]
        [switch_simplify]: 5.32001e-06
        [loop_unroll]: 5.08002e-06
        [a_1]: 8.914e-05
        [with_stream_mark]: 9.15999e-06
        [recompute_prepare]: 5.12e-06
        [updatestate_depend_eliminate]: 2.47001e-06
        [updatestate_assign_eliminate]: 2.08998e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 4.92999e-06
        [updatestate_useless_node_eliminater]: 4.75001e-06
        [accelerated_algorithm]: 4.84e-06
        [meta_shard_fg_expand]: 1.16997e-06
        [get_grad_eliminate_]: 4.60999e-06
        [merge_forward]: 2.90002e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.208e-05
        [j_node_and_user_rematch]: 7.75e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 7.53e-06
        [inline_without_move]: 4.63001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 8.89995e-07
        [auto_monad_eliminator]: 5.19e-06
        [cse]: 9.87999e-06
        [replace_applicator]: 4.94e-06
[py_interpret_to_execute_after_opt_a]: 1.024e-05
[rewriter_after_opt_a]: 3.262e-05
[convert_after_rewriter]: 6.53e-06
[order_py_execute_after_rewriter]: 4.75001e-06
[mutable_eliminate]: 0.00055679
[jit_opt_b]: 4.888e-05, [1]
    [Cycle 1]: 4.255e-05, [2]
        [frontend_op_eliminate]: 1.648e-05
        [inline_after_opt_a]: 1.468e-05
[cconv]: 2.185e-05
[loop_unroll]: 0.00041962
[jit_opt_after_cconv]: 0.00014151, [1]
    [Cycle 1]: 0.00013542, [11]
        [c_1]: 2.21e-05
        [parameter_eliminate]: 2.57001e-06
        [updatestate_depend_eliminate]: 5.54e-06
        [updatestate_assign_eliminate]: 2.51998e-06
        [updatestate_loads_eliminate]: 2.23002e-06
        [cse]: 1.845e-05
        [call_graph_tuple_transform]: 1.764e-05
        [tuple_list_get_item_eliminator]: 5.54e-06
        [none_parameter_eliminate]: 1.64998e-06
        [renormalize]: 5.3001e-07
        [switch_simplify]: 5.42001e-06
[remove_dup_value]: 1.085e-05
[partial_unused_args_eliminate]: 2.05002e-06
[environ_conv]: 4.48999e-06
[add_recomputation]: 4.38e-05
[cse_after_recomputation]: 2.109e-05, [1]
    [Cycle 1]: 1.56e-05, [1]
        [cse]: 9.77999e-06
[auto_monad_reorder]: 1.291e-05
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 4.70001e-06
[opt_after_jit_grad]: 0.00046275
[symbol_engine_optimizer]: 7.063e-05, [1]
    [Cycle 1]: 6.439e-05, [6]
        [build]: 2.96001e-06
        [elim_shapecalc]: 6.92002e-06
        [elim_not_effective]: 1.29e-05
        [opt_reshape]: 5.56e-06
        [fold_const_symbol]: 8.41002e-06
        [renormalize]: 5.3001e-07
[validate]: 3.026e-05
[backend_pass]: 7.2e-07
[task_emit]: 0.00606534
[execute]: 7.51001e-06
Sums
    bootstrap                            :     0.000246s :  1.77%
    type_inference                       :     0.004014s : 28.92%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000058s :  0.42%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.22%
    rewriter_before_opt_a                :     0.000070s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000053s :  0.38%
    jit_opt_a.loop_unroll                :     0.000038s :  0.27%
    jit_opt_a.a_1                        :     0.000623s :  4.49%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000609s :  4.39%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.12%
    jit_opt_a.cse                        :     0.000031s :  0.22%
    jit_opt_a.replace_applicator         :     0.000017s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000557s :  4.01%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000022s :  0.16%
    loop_unroll                          :     0.000420s :  3.02%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000011s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000044s :  0.32%
    cse_after_recomputation.cse          :     0.000010s :  0.07%
    auto_monad_reorder                   :     0.000013s :  0.09%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000463s :  3.33%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.05%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000030s :  0.22%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006065s : 43.69%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000195    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.19% :     0.000004s :      3: substitution.graph_param_transform
83.91% :     0.000164s :      4: substitution.inline
 1.34% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.12% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.27% :     0.000004s :      2: substitution.replace_old_param
 6.56% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003959     2
80.75% :     0.003197s :      1: type_inference.infer
19.25% :     0.000762s :      1: type_inference.specialize
------[replace.]   0.000070     6
65.84% :     0.000046s :      4: replace.inline
34.16% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000173     6
93.25% :     0.000161s :      4: match.inline
 6.75% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.22% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.43% :     0.000002s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.34% :     0.000002s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.69% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.34% :     0.000002s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.32% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.65% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.87% :     0.000002s :     15: predicate.load_eliminater
 1.78% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.67% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.58% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.53% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.40% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.91% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.41% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.62% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 1.14% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.51% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.64% :     0.000009s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.29% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000500    12
46.80% :     0.000234s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.58% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.62% :     0.000238s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019187    76
 0.24% :     0.000047s :      1: add_recomputation
 0.32% :     0.000061s :      1: auto_monad
 0.08% :     0.000015s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.34% :     0.000256s :      1: bootstrap
 0.13% :     0.000025s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
26.29% :     0.005043s :      1: jit_opt_a
 0.75% :     0.000144s :      1: jit_opt_after_cconv
 0.27% :     0.000051s :      1: jit_opt_b
 2.23% :     0.000428s :      1: loop_unroll
 2.95% :     0.000565s :      1: mutable_eliminate
 4.31% :     0.000827s :     26: opt.transform.jit_opt_a
 0.25% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.16% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.46% :     0.000471s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000032s :      1: py_interpret_to_execute
 0.07% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000013s :      1: remove_dup_value
 1.60% :     0.000307s :      1: renormalize.infer
 1.54% :     0.000295s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.18% :     0.000035s :      1: rewriter_after_opt_a
 0.38% :     0.000073s :      1: rewriter_before_opt_a
 0.38% :     0.000073s :      1: symbol_engine_optimizer
31.66% :     0.006075s :      1: task_emit
20.98% :     0.004026s :      1: type_inference
 0.25% :     0.000049s :      1: validate


TotalTime = 0.0226619, [33]
[bootstrap]: 0.00022168
[type_inference]: 0.0108277
[event_method]: 5.211e-05
[auto_monad]: 9.944e-05
[graph_reusing]: 7.38999e-06
[pre_auto_parallel]: 2.89001e-06
[py_interpret_to_execute]: 2.624e-05
[rewriter_before_opt_a]: 9.354e-05
[expand_dump_flag]: 3.60998e-06
[jit_opt_a]: 0.00906745, [2]
    [Cycle 1]: 0.005027, [27]
        [switch_simplify]: 7.429e-05
        [loop_unroll]: 3.911e-05
        [a_1]: 0.00090686
        [with_stream_mark]: 2.766e-05
        [recompute_prepare]: 1.797e-05
        [updatestate_depend_eliminate]: 8.13999e-06
        [updatestate_assign_eliminate]: 6.80002e-06
        [updatestate_loads_eliminate]: 6.21998e-06
        [parameter_eliminate]: 3.03e-06
        [specialize_transform]: 1.289e-05
        [updatestate_useless_node_eliminater]: 1.24e-05
        [accelerated_algorithm]: 1.927e-05
        [meta_shard_fg_expand]: 4.52e-06
        [get_grad_eliminate_]: 1.217e-05
        [merge_forward]: 8.82e-06
        [cell_reuse_recompute_pass]: 1.11002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.629e-05
        [j_node_and_user_rematch]: 2.277e-05
        [meta_fg_expand]: 0.00160336
        [replace_old_param]: 5.39e-05
        [inline_without_move]: 5.324e-05
        [renormalize]: 0.00183866
        [add_forward_monad_depend]: 6.74999e-06
        [auto_monad_grad]: 2.32999e-06
        [auto_monad_eliminator]: 1.18e-05
        [cse]: 2.419e-05
        [replace_applicator]: 1.248e-05
    [Cycle 2]: 0.00031031, [27]
        [switch_simplify]: 4.45e-06
        [loop_unroll]: 3.63999e-06
        [a_1]: 4.878e-05
        [with_stream_mark]: 9.13002e-06
        [recompute_prepare]: 3.7e-06
        [updatestate_depend_eliminate]: 2.36e-06
        [updatestate_assign_eliminate]: 1.66002e-06
        [updatestate_loads_eliminate]: 1.50001e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 3.77998e-06
        [updatestate_useless_node_eliminater]: 3.53e-06
        [accelerated_algorithm]: 7.32997e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 3.54002e-06
        [merge_forward]: 2.36998e-06
        [cell_reuse_recompute_pass]: 2.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.125e-05
        [j_node_and_user_rematch]: 5.81e-06
        [meta_fg_expand]: 4.433e-05
        [replace_old_param]: 4.28001e-06
        [inline_without_move]: 3.48e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.49998e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 4.94e-06
        [cse]: 1.121e-05
        [replace_applicator]: 3.93999e-06
[py_interpret_to_execute_after_opt_a]: 1.05e-05
[rewriter_after_opt_a]: 3.438e-05
[convert_after_rewriter]: 6.31998e-06
[order_py_execute_after_rewriter]: 4.67e-06
[mutable_eliminate]: 0.00070763
[jit_opt_b]: 4.232e-05, [1]
    [Cycle 1]: 3.477e-05, [2]
        [frontend_op_eliminate]: 1.108e-05
        [inline_after_opt_a]: 1.198e-05
[cconv]: 2.487e-05
[loop_unroll]: 0.00045328
[jit_opt_after_cconv]: 0.00012077, [1]
    [Cycle 1]: 0.00011453, [11]
        [c_1]: 1.356e-05
        [parameter_eliminate]: 2.16e-06
        [updatestate_depend_eliminate]: 4.42e-06
        [updatestate_assign_eliminate]: 1.81e-06
        [updatestate_loads_eliminate]: 1.66e-06
        [cse]: 1.59e-05
        [call_graph_tuple_transform]: 1.349e-05
        [tuple_list_get_item_eliminator]: 3.95e-06
        [none_parameter_eliminate]: 1.74e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 3.91001e-06
[remove_dup_value]: 1.473e-05
[partial_unused_args_eliminate]: 2.19001e-06
[environ_conv]: 5.04e-06
[add_recomputation]: 3.297e-05
[cse_after_recomputation]: 2.025e-05, [1]
    [Cycle 1]: 1.461e-05, [1]
        [cse]: 8.57e-06
[auto_monad_reorder]: 1.365e-05
[get_jit_bprop_graph]: 1.77001e-06
[rewriter_after_jit_bprop_graph]: 5.07999e-06
[opt_after_jit_grad]: 0.00046384
[symbol_engine_optimizer]: 6.748e-05, [1]
    [Cycle 1]: 6.146e-05, [6]
        [build]: 2.91e-06
        [elim_shapecalc]: 6.19999e-06
        [elim_not_effective]: 1.034e-05
        [opt_reshape]: 4.54002e-06
        [fold_const_symbol]: 7.25e-06
        [renormalize]: 5.50004e-07
[validate]: 2.9e-05
[backend_pass]: 1.02e-06
[task_emit]: 2.101e-05
[execute]: 1.52999e-06
Sums
    bootstrap                            :     0.000222s :  1.21%
    type_inference                       :     0.010828s : 59.15%
    event_method                         :     0.000052s :  0.28%
    auto_monad                           :     0.000099s :  0.54%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.14%
    rewriter_before_opt_a                :     0.000094s :  0.51%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000079s :  0.43%
    jit_opt_a.loop_unroll                :     0.000043s :  0.23%
    jit_opt_a.a_1                        :     0.000956s :  5.22%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000027s :  0.15%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.09%
    jit_opt_a.merge_forward              :     0.000011s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001648s :  9.00%
    jit_opt_a.replace_old_param          :     0.000058s :  0.32%
    jit_opt_a.inline_without_move        :     0.000057s :  0.31%
    jit_opt_a.renormalize                :     0.001839s : 10.05%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.09%
    jit_opt_a.cse                        :     0.000035s :  0.19%
    jit_opt_a.replace_applicator         :     0.000016s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000034s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000708s :  3.87%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.07%
    cconv                                :     0.000025s :  0.14%
    loop_unroll                          :     0.000453s :  2.48%
    jit_opt_after_cconv.c_1              :     0.000014s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000033s :  0.18%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000464s :  2.53%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000007s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000029s :  0.16%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.000021s :  0.11%
    execute                              :     0.000002s :  0.01%


Time group info:
------[substitution.]   0.000298    50
 0.71% :     0.000002s :      1: substitution.elim_not_effective
 0.44% :     0.000001s :      1: substitution.fold_const_symbol
 1.33% :     0.000004s :      1: substitution.graph_param_transform
70.13% :     0.000209s :      6: substitution.inline
 6.02% :     0.000018s :      2: substitution.inline_without_move
 2.08% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.33% :     0.000004s :      2: substitution.minmaximum_grad
 1.43% :     0.000004s :      3: substitution.partial_eliminate
 2.50% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.16% :     0.000006s :      2: substitution.replace_applicator
 1.42% :     0.000004s :      5: substitution.replace_old_param
 0.79% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.51% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.75% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.40% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010754     2
88.57% :     0.009525s :      1: type_inference.infer
11.43% :     0.001229s :      1: type_inference.specialize
------[replace.]   0.000072     8
78.10% :     0.000056s :      6: replace.inline
21.90% :     0.000016s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000216     8
94.74% :     0.000205s :      6: match.inline
 5.26% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000176  1109
 1.54% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.67% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.28% :     0.000002s :     18: predicate.addn_check_dump
 1.52% :     0.000003s :     18: predicate.addn_zero_filter
 2.20% :     0.000004s :     18: predicate.arithmetic_simplify
 1.70% :     0.000003s :     18: predicate.cast_eliminate
 0.23% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.34% :     0.000002s :     18: predicate.compare_switch_simplify
 1.61% :     0.000003s :     18: predicate.depend_value_elim
 1.49% :     0.000003s :     18: predicate.dict_get_item_const_eliminator
 1.43% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.48% :     0.000003s :     18: predicate.dict_set_item_eliminator
 0.44% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.08% :     0.000000s :      1: predicate.elim_not_effective
 0.18% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.25% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.37% :     0.000002s :     18: predicate.environ_get_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.10% :     0.000002s :     12: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 4.26% :     0.000008s :     28: predicate.inline
 3.11% :     0.000005s :     33: predicate.inline_without_move
 0.48% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.40% :     0.000002s :     12: predicate.less_batch_normalization
 1.52% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.95% :     0.000003s :     21: predicate.load_eliminater
 0.62% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.48% :     0.000006s :     44: predicate.loop_unroll_before_grad
 2.04% :     0.000004s :     19: predicate.make_slice_get_slice_eliminator
 1.29% :     0.000002s :     18: predicate.merge_addn
 1.25% :     0.000002s :     18: predicate.minmaximum_grad
 0.66% :     0.000001s :      1: predicate.mutable_eliminate
 0.20% :     0.000000s :      1: predicate.opt_reshape
 2.07% :     0.000004s :     21: predicate.partial_eliminate
 1.33% :     0.000002s :     18: predicate.print_const_string_wrapper
 2.16% :     0.000004s :     18: predicate.reduce_eliminate
 1.49% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.44% :     0.000003s :     22: predicate.replace_applicator
 1.60% :     0.000003s :     33: predicate.replace_old_param
 0.15% :     0.000000s :      1: predicate.reset_defer_inline
 1.56% :     0.000003s :     18: predicate.reshape_eliminate
 1.48% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.50% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     18: predicate.same_eliminate
 0.65% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.43% :     0.000001s :      2: predicate.special_op_eliminate
 1.17% :     0.000002s :     12: predicate.specialize_transform
 1.80% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.56% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.98% :     0.000005s :     26: predicate.switch_defer_inline
 2.22% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.52% :     0.000011s :     71: predicate.switch_simplify
 1.43% :     0.000003s :     18: predicate.tile_eliminate
 1.54% :     0.000003s :     18: predicate.transpose_eliminate
 1.82% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.60% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.08% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.57% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000003s :     18: predicate.value_based_eliminate
 0.11% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.18% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001263    22
70.92% :     0.000896s :     13: func_graph_cloner_run.FuncGraphClonerGraph
29.08% :     0.000367s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025903    76
 0.14% :     0.000036s :      1: add_recomputation
 0.40% :     0.000104s :      1: auto_monad
 0.06% :     0.000016s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.91% :     0.000235s :      1: bootstrap
 0.11% :     0.000027s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.22% :     0.000057s :      1: event_method
 0.01% :     0.000004s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
35.02% :     0.009071s :      1: jit_opt_a
 0.48% :     0.000124s :      1: jit_opt_after_cconv
 0.17% :     0.000045s :      1: jit_opt_b
 1.78% :     0.000461s :      1: loop_unroll
 2.77% :     0.000717s :      1: mutable_eliminate
 5.17% :     0.001340s :     26: opt.transform.jit_opt_a
 0.12% :     0.000031s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000024s :      4: opt.transform.symbol_engine_opt
 1.82% :     0.000472s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000029s :      1: py_interpret_to_execute
 0.05% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 5.21% :     0.001350s :      1: renormalize.infer
 1.85% :     0.000479s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000038s :      1: rewriter_after_opt_a
 0.37% :     0.000096s :      1: rewriter_before_opt_a
 0.27% :     0.000070s :      1: symbol_engine_optimizer
 0.10% :     0.000025s :      1: task_emit
41.85% :     0.010839s :      1: type_inference
 0.17% :     0.000045s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.550.357 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dc39820> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.550.976 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0159146, [30]
[bootstrap]: 0.00037061
[type_inference]: 0.00568517
[event_method]: 1.992e-05
[auto_monad]: 7.385e-05
[graph_reusing]: 6.26998e-06
[pre_auto_parallel]: 3.33998e-06
[py_interpret_to_execute]: 3.588e-05
[rewriter_before_opt_a]: 7.643e-05
[expand_dump_flag]: 3.3e-06
[jit_opt_a]: 0.00680728, [2]
    [Cycle 1]: 0.00210283, [27]
        [switch_simplify]: 6.858e-05
        [loop_unroll]: 3.112e-05
        [a_1]: 0.0006451
        [with_stream_mark]: 3.021e-05
        [recompute_prepare]: 1.177e-05
        [updatestate_depend_eliminate]: 4.33001e-06
        [updatestate_assign_eliminate]: 3.08998e-06
        [updatestate_loads_eliminate]: 2.51998e-06
        [parameter_eliminate]: 2.03002e-06
        [specialize_transform]: 6.34999e-06
        [updatestate_useless_node_eliminater]: 5.57999e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 2.17001e-06
        [get_grad_eliminate_]: 6.12001e-06
        [merge_forward]: 4.4e-06
        [cell_reuse_recompute_pass]: 1.82001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.799e-05
        [j_node_and_user_rematch]: 9.25999e-06
        [meta_fg_expand]: 3.06001e-06
        [replace_old_param]: 1.071e-05
        [inline_without_move]: 5.99999e-06
        [renormalize]: 0.0009
        [add_forward_monad_depend]: 9.24998e-06
        [auto_monad_grad]: 2.74001e-06
        [auto_monad_eliminator]: 2.055e-05
        [cse]: 3.673e-05
        [replace_applicator]: 1.926e-05
    [Cycle 2]: 0.00037503, [27]
        [switch_simplify]: 6.27001e-06
        [loop_unroll]: 5.35999e-06
        [a_1]: 9.01e-05
        [with_stream_mark]: 1.668e-05
        [recompute_prepare]: 5.95002e-06
        [updatestate_depend_eliminate]: 3.79002e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [parameter_eliminate]: 1.71e-06
        [specialize_transform]: 5.51e-06
        [updatestate_useless_node_eliminater]: 5.69999e-06
        [accelerated_algorithm]: 6.38998e-06
        [meta_shard_fg_expand]: 1.96e-06
        [get_grad_eliminate_]: 4.48999e-06
        [merge_forward]: 3.66001e-06
        [cell_reuse_recompute_pass]: 2.19001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.719e-05
        [j_node_and_user_rematch]: 8.84998e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 7.77e-06
        [inline_without_move]: 5.51e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.90002e-06
        [auto_monad_grad]: 1.37e-06
        [auto_monad_eliminator]: 8.04002e-06
        [cse]: 1.519e-05
        [replace_applicator]: 5.57001e-06
[py_interpret_to_execute_after_opt_a]: 1.75e-05
[rewriter_after_opt_a]: 4.482e-05
[convert_after_rewriter]: 8.40001e-06
[order_py_execute_after_rewriter]: 5.93002e-06
[mutable_eliminate]: 0.00084439
[jit_opt_b]: 5.813e-05, [1]
    [Cycle 1]: 4.901e-05, [2]
        [frontend_op_eliminate]: 1.823e-05
        [inline_after_opt_a]: 1.702e-05
[cconv]: 3.29e-05
[loop_unroll]: 0.00063607
[jit_opt_after_cconv]: 0.00016069, [1]
    [Cycle 1]: 0.00015238, [11]
        [c_1]: 2.043e-05
        [parameter_eliminate]: 4.37e-06
        [updatestate_depend_eliminate]: 1.064e-05
        [updatestate_assign_eliminate]: 3.03e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [cse]: 2.66e-05
        [call_graph_tuple_transform]: 2.193e-05
        [tuple_list_get_item_eliminator]: 5.39e-06
        [none_parameter_eliminate]: 1.51002e-06
        [renormalize]: 2.9002e-07
        [switch_simplify]: 5.36002e-06
[remove_dup_value]: 1.631e-05
[partial_unused_args_eliminate]: 2.90002e-06
[environ_conv]: 6.54999e-06
[add_recomputation]: 5.894e-05
[cse_after_recomputation]: 2.203e-05, [1]
    [Cycle 1]: 1.603e-05, [1]
        [cse]: 8.85999e-06
[auto_monad_reorder]: 1.643e-05
[get_jit_bprop_graph]: 2.60002e-06
[rewriter_after_jit_bprop_graph]: 9.04e-06
[opt_after_jit_grad]: 0.00055395
[symbol_engine_optimizer]: 7.66e-05, [1]
    [Cycle 1]: 6.919e-05, [6]
        [build]: 4.86002e-06
        [elim_shapecalc]: 7.28e-06
        [elim_not_effective]: 1.371e-05
        [opt_reshape]: 5.35999e-06
        [fold_const_symbol]: 8.70001e-06
        [renormalize]: 4.7998e-07
[validate]: 4.414e-05
Sums
    bootstrap                            :     0.000371s :  3.41%
    type_inference                       :     0.005685s : 52.31%
    event_method                         :     0.000020s :  0.18%
    auto_monad                           :     0.000074s :  0.68%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.33%
    rewriter_before_opt_a                :     0.000076s :  0.70%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000075s :  0.69%
    jit_opt_a.loop_unroll                :     0.000036s :  0.34%
    jit_opt_a.a_1                        :     0.000735s :  6.76%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.43%
    jit_opt_a.recompute_prepare          :     0.000018s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.10%
    jit_opt_a.merge_forward              :     0.000008s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000018s :  0.17%
    jit_opt_a.inline_without_move        :     0.000012s :  0.11%
    jit_opt_a.renormalize                :     0.000900s :  8.28%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.11%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000029s :  0.26%
    jit_opt_a.cse                        :     0.000052s :  0.48%
    jit_opt_a.replace_applicator         :     0.000025s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.16%
    rewriter_after_opt_a                 :     0.000045s :  0.41%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.05%
    mutable_eliminate                    :     0.000844s :  7.77%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.16%
    cconv                                :     0.000033s :  0.30%
    loop_unroll                          :     0.000636s :  5.85%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.19%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000011s :  0.10%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000027s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000016s :  0.15%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000007s :  0.06%
    add_recomputation                    :     0.000059s :  0.54%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000016s :  0.15%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.08%
    opt_after_jit_grad                   :     0.000554s :  5.10%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000044s :  0.41%


Time group info:
------[substitution.]   0.000279    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.46% :     0.000001s :      2: substitution.fold_const_symbol
 2.06% :     0.000006s :      3: substitution.graph_param_transform
85.62% :     0.000239s :      4: substitution.inline
 1.28% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.96% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.26% :     0.000006s :      2: substitution.replace_old_param
 5.61% :     0.000016s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005597     2
83.35% :     0.004665s :      1: type_inference.infer
16.65% :     0.000932s :      1: type_inference.specialize
------[replace.]   0.000084     6
71.37% :     0.000060s :      4: replace.inline
28.63% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000250     6
94.16% :     0.000235s :      4: match.inline
 5.84% :     0.000015s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000136   691
 1.39% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.78% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.92% :     0.000001s :     10: predicate.addn_check_dump
 1.59% :     0.000002s :     10: predicate.addn_zero_filter
 2.86% :     0.000004s :     10: predicate.arithmetic_simplify
 1.15% :     0.000002s :     10: predicate.cast_eliminate
 0.34% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.89% :     0.000001s :     10: predicate.compare_switch_simplify
 1.75% :     0.000002s :     10: predicate.depend_value_elim
 0.90% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.99% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.19% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.02% :     0.000001s :     10: predicate.environ_get_eliminate
 0.88% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.18% :     0.000000s :      3: predicate.fold_const_symbol
 0.77% :     0.000001s :      6: predicate.get_grad_eliminate
 0.39% :     0.000001s :      3: predicate.graph_param_transform
 6.14% :     0.000008s :     22: predicate.inline
 1.17% :     0.000002s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.46% :     0.000002s :      6: predicate.less_batch_normalization
 1.20% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.46% :     0.000002s :     15: predicate.load_eliminater
 2.47% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.68% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.87% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 2.91% :     0.000004s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 2.04% :     0.000003s :     15: predicate.partial_eliminate
 1.26% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.89% :     0.000003s :     10: predicate.reduce_eliminate
 1.15% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.73% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     18: predicate.replace_applicator
 0.54% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.59% :     0.000002s :     10: predicate.reshape_eliminate
 0.97% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.06% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.51% :     0.000002s :     10: predicate.same_eliminate
 0.64% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.50% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.89% :     0.000004s :     16: predicate.switch_defer_inline
 1.64% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.48% :     0.000010s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 0.95% :     0.000001s :     10: predicate.transpose_eliminate
 1.09% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.04% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.27% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.35% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.38% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.56% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.89% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000724    12
51.04% :     0.000370s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.92% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.04% :     0.000297s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017853    72
 0.35% :     0.000062s :      1: add_recomputation
 0.44% :     0.000078s :      1: auto_monad
 0.11% :     0.000020s :      1: auto_monad_reorder
 2.22% :     0.000397s :      1: bootstrap
 0.20% :     0.000036s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.14% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.14% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.15% :     0.006811s :      1: jit_opt_a
 0.92% :     0.000163s :      1: jit_opt_after_cconv
 0.34% :     0.000061s :      1: jit_opt_b
 3.63% :     0.000648s :      1: loop_unroll
 4.81% :     0.000859s :      1: mutable_eliminate
 5.51% :     0.000984s :     26: opt.transform.jit_opt_a
 0.27% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000028s :      4: opt.transform.jit_opt_b
 0.10% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.17% :     0.000565s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000038s :      1: py_interpret_to_execute
 0.11% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000019s :      1: remove_dup_value
 2.83% :     0.000505s :      1: renormalize.infer
 2.14% :     0.000382s :      1: renormalize.specialize
 0.06% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000049s :      1: rewriter_after_opt_a
 0.45% :     0.000080s :      1: rewriter_before_opt_a
 0.44% :     0.000079s :      1: symbol_engine_optimizer
31.97% :     0.005708s :      1: type_inference


TotalTime = 0.0212027, [30]
[bootstrap]: 0.00020725
[type_inference]: 0.00919927
[event_method]: 4.589e-05
[auto_monad]: 0.00070389
[graph_reusing]: 5.62001e-06
[pre_auto_parallel]: 2.11e-06
[py_interpret_to_execute]: 8.55e-05
[rewriter_before_opt_a]: 0.00018805
[expand_dump_flag]: 2.78998e-06
[jit_opt_a]: 0.00849841, [2]
    [Cycle 1]: 0.00460531, [27]
        [switch_simplify]: 0.00010584
        [loop_unroll]: 3.764e-05
        [a_1]: 0.00081674
        [with_stream_mark]: 2.374e-05
        [recompute_prepare]: 1.737e-05
        [updatestate_depend_eliminate]: 7.69002e-06
        [updatestate_assign_eliminate]: 5.66998e-06
        [updatestate_loads_eliminate]: 5.89e-06
        [parameter_eliminate]: 2.49999e-06
        [specialize_transform]: 1.48e-05
        [updatestate_useless_node_eliminater]: 1.263e-05
        [accelerated_algorithm]: 1.223e-05
        [meta_shard_fg_expand]: 2.64001e-06
        [get_grad_eliminate_]: 1.251e-05
        [merge_forward]: 7.02002e-06
        [cell_reuse_recompute_pass]: 7.39994e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.472e-05
        [j_node_and_user_rematch]: 2.44e-05
        [meta_fg_expand]: 0.00140268
        [replace_old_param]: 5.025e-05
        [inline_without_move]: 5.191e-05
        [renormalize]: 0.00170672
        [add_forward_monad_depend]: 5.12e-06
        [auto_monad_grad]: 2.32999e-06
        [auto_monad_eliminator]: 1.177e-05
        [cse]: 2.391e-05
        [replace_applicator]: 1.205e-05
    [Cycle 2]: 0.00030264, [27]
        [switch_simplify]: 5.18002e-06
        [loop_unroll]: 3.95e-06
        [a_1]: 4.645e-05
        [with_stream_mark]: 9.20999e-06
        [recompute_prepare]: 4.01001e-06
        [updatestate_depend_eliminate]: 2.51998e-06
        [updatestate_assign_eliminate]: 1.98002e-06
        [updatestate_loads_eliminate]: 2.53e-06
        [parameter_eliminate]: 1.62999e-06
        [specialize_transform]: 3.73001e-06
        [updatestate_useless_node_eliminater]: 3.91001e-06
        [accelerated_algorithm]: 6.58e-06
        [meta_shard_fg_expand]: 1.92001e-06
        [get_grad_eliminate_]: 4.12e-06
        [merge_forward]: 2.31998e-06
        [cell_reuse_recompute_pass]: 2.05002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.203e-05
        [j_node_and_user_rematch]: 6.29999e-06
        [meta_fg_expand]: 2.776e-05
        [replace_old_param]: 4.05e-06
        [inline_without_move]: 3.61999e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.57001e-06
        [auto_monad_grad]: 1.01997e-06
        [auto_monad_eliminator]: 4.24002e-06
        [cse]: 1.067e-05
        [replace_applicator]: 4.18001e-06
[py_interpret_to_execute_after_opt_a]: 1.019e-05
[rewriter_after_opt_a]: 3.349e-05
[convert_after_rewriter]: 5.44e-06
[order_py_execute_after_rewriter]: 4.65999e-06
[mutable_eliminate]: 0.00062171
[jit_opt_b]: 4.319e-05, [1]
    [Cycle 1]: 3.545e-05, [2]
        [frontend_op_eliminate]: 1.186e-05
        [inline_after_opt_a]: 1.145e-05
[cconv]: 2.209e-05
[loop_unroll]: 0.0004804
[jit_opt_after_cconv]: 0.00013373, [1]
    [Cycle 1]: 0.00012661, [11]
        [c_1]: 1.335e-05
        [parameter_eliminate]: 2.54999e-06
        [updatestate_depend_eliminate]: 6.04001e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [cse]: 2.042e-05
        [call_graph_tuple_transform]: 1.487e-05
        [tuple_list_get_item_eliminator]: 4.38001e-06
        [none_parameter_eliminate]: 1.32999e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 4.62998e-06
[remove_dup_value]: 1.366e-05
[partial_unused_args_eliminate]: 2.63e-06
[environ_conv]: 5.66e-06
[add_recomputation]: 3.327e-05
[cse_after_recomputation]: 2.082e-05, [1]
    [Cycle 1]: 1.533e-05, [1]
        [cse]: 8.32e-06
[auto_monad_reorder]: 1.241e-05
[get_jit_bprop_graph]: 1.54e-06
[rewriter_after_jit_bprop_graph]: 7.58999e-06
[opt_after_jit_grad]: 0.0004855
[symbol_engine_optimizer]: 6.785e-05, [1]
    [Cycle 1]: 6.145e-05, [6]
        [build]: 3.06999e-06
        [elim_shapecalc]: 6.05002e-06
        [elim_not_effective]: 9.94001e-06
        [opt_reshape]: 5.07e-06
        [fold_const_symbol]: 6.71e-06
        [renormalize]: 5.89993e-07
[validate]: 2.881e-05
Sums
    bootstrap                            :     0.000207s :  1.22%
    type_inference                       :     0.009199s : 54.36%
    event_method                         :     0.000046s :  0.27%
    auto_monad                           :     0.000704s :  4.16%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000086s :  0.51%
    rewriter_before_opt_a                :     0.000188s :  1.11%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000111s :  0.66%
    jit_opt_a.loop_unroll                :     0.000042s :  0.25%
    jit_opt_a.a_1                        :     0.000863s :  5.10%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000019s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000017s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000019s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.10%
    jit_opt_a.merge_forward              :     0.000009s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000031s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001430s :  8.45%
    jit_opt_a.replace_old_param          :     0.000054s :  0.32%
    jit_opt_a.inline_without_move        :     0.000056s :  0.33%
    jit_opt_a.renormalize                :     0.001707s : 10.09%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.09%
    jit_opt_a.cse                        :     0.000035s :  0.20%
    jit_opt_a.replace_applicator         :     0.000016s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.20%
    convert_after_rewriter               :     0.000005s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000622s :  3.67%
    jit_opt_b.frontend_op_eliminate      :     0.000012s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000022s :  0.13%
    loop_unroll                          :     0.000480s :  2.84%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000033s :  0.20%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000012s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000486s :  2.87%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000007s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000029s :  0.17%


Time group info:
------[substitution.]   0.000243    50
 0.62% :     0.000002s :      1: substitution.elim_not_effective
 0.35% :     0.000001s :      1: substitution.fold_const_symbol
 1.74% :     0.000004s :      1: substitution.graph_param_transform
66.31% :     0.000161s :      6: substitution.inline
 7.34% :     0.000018s :      2: substitution.inline_without_move
 2.67% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.46% :     0.000004s :      2: substitution.minmaximum_grad
 1.48% :     0.000004s :      3: substitution.partial_eliminate
 2.93% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.36% :     0.000006s :      2: substitution.replace_applicator
 1.55% :     0.000004s :      5: substitution.replace_old_param
 1.14% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.78% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.84% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 5.43% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009141     2
88.92% :     0.008128s :      1: type_inference.infer
11.08% :     0.001013s :      1: type_inference.specialize
------[replace.]   0.000061     8
77.33% :     0.000047s :      6: replace.inline
22.67% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000166     8
94.96% :     0.000158s :      6: match.inline
 5.04% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000217  1109
 1.10% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.48% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.02% :     0.000002s :     18: predicate.addn_check_dump
 1.25% :     0.000003s :     18: predicate.addn_zero_filter
 1.57% :     0.000003s :     18: predicate.arithmetic_simplify
 1.35% :     0.000003s :     18: predicate.cast_eliminate
 0.12% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.00% :     0.000002s :     18: predicate.compare_switch_simplify
 1.06% :     0.000002s :     18: predicate.depend_value_elim
 1.09% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.09% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.06% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.31% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.07% :     0.000000s :      1: predicate.elim_not_effective
 0.17% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.02% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.05% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.08% :     0.000002s :     18: predicate.environ_get_eliminate
 1.10% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.07% :     0.000000s :      1: predicate.fold_const_symbol
 1.03% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 3.32% :     0.000007s :     28: predicate.inline
 2.45% :     0.000005s :     33: predicate.inline_without_move
 0.42% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.11% :     0.000002s :     12: predicate.less_batch_normalization
 1.41% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.28% :     0.000003s :     21: predicate.load_eliminater
 0.58% :     0.000001s :      1: predicate.loop_unroll_after_grad
 2.81% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.25% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000002s :     18: predicate.merge_addn
 1.02% :     0.000002s :     18: predicate.minmaximum_grad
 0.81% :     0.000002s :      1: predicate.mutable_eliminate
 0.14% :     0.000000s :      1: predicate.opt_reshape
 1.56% :     0.000003s :     21: predicate.partial_eliminate
 1.05% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.51% :     0.000003s :     18: predicate.reduce_eliminate
 1.20% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.20% :     0.000003s :     22: predicate.replace_applicator
 1.30% :     0.000003s :     33: predicate.replace_old_param
 0.10% :     0.000000s :      1: predicate.reset_defer_inline
 1.40% :     0.000003s :     18: predicate.reshape_eliminate
 1.28% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.43% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.07% :     0.000002s :     18: predicate.same_eliminate
 0.48% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.36% :     0.000001s :      2: predicate.special_op_eliminate
 1.09% :     0.000002s :     12: predicate.specialize_transform
 1.29% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.09% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.13% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 1.96% :     0.000004s :     26: predicate.switch_defer_inline
 1.86% :     0.000004s :     26: predicate.switch_layer_defer_inline
26.80% :     0.000058s :     71: predicate.switch_simplify
 1.20% :     0.000003s :     18: predicate.tile_eliminate
 1.22% :     0.000003s :     18: predicate.transpose_eliminate
 1.45% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.29% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.18% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.26% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.47% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.30% :     0.000003s :     18: predicate.value_based_eliminate
 0.09% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.20% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001108    22
73.00% :     0.000809s :     13: func_graph_cloner_run.FuncGraphClonerGraph
27.00% :     0.000299s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024205    72
 0.15% :     0.000036s :      1: add_recomputation
 2.94% :     0.000711s :      1: auto_monad
 0.06% :     0.000015s :      1: auto_monad_reorder
 0.91% :     0.000221s :      1: bootstrap
 0.10% :     0.000025s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.21% :     0.000050s :      1: event_method
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
35.12% :     0.008502s :      1: jit_opt_a
 0.56% :     0.000137s :      1: jit_opt_after_cconv
 0.19% :     0.000046s :      1: jit_opt_b
 2.02% :     0.000489s :      1: loop_unroll
 2.61% :     0.000632s :      1: mutable_eliminate
 5.26% :     0.001273s :     26: opt.transform.jit_opt_a
 0.14% :     0.000034s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000017s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000014s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000024s :      4: opt.transform.symbol_engine_opt
 2.03% :     0.000492s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.37% :     0.000089s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 5.29% :     0.001281s :      1: renormalize.infer
 1.72% :     0.000417s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000037s :      1: rewriter_after_opt_a
 1.03% :     0.000249s :      1: rewriter_before_opt_a
 0.29% :     0.000070s :      1: symbol_engine_optimizer
38.06% :     0.009211s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.562.761 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d10d0> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.563.298 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0171899, [33]
[bootstrap]: 0.00026671
[type_inference]: 0.0042793
[event_method]: 1.813e-05
[auto_monad]: 6.432e-05
[graph_reusing]: 5.34e-06
[pre_auto_parallel]: 2.46998e-06
[py_interpret_to_execute]: 3.045e-05
[rewriter_before_opt_a]: 7.324e-05
[expand_dump_flag]: 3.78001e-06
[jit_opt_a]: 0.00500162, [2]
    [Cycle 1]: 0.00142061, [27]
        [switch_simplify]: 4.749e-05
        [loop_unroll]: 3.13e-05
        [a_1]: 0.0005096
        [with_stream_mark]: 1.242e-05
        [recompute_prepare]: 7.31001e-06
        [updatestate_depend_eliminate]: 3.44001e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.88002e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 5.65001e-06
        [updatestate_useless_node_eliminater]: 5.30001e-06
        [accelerated_algorithm]: 6.36998e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 6.01e-06
        [merge_forward]: 2.42001e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.071e-05
        [j_node_and_user_rematch]: 8.94e-06
        [meta_fg_expand]: 1.98002e-06
        [replace_old_param]: 9.03002e-06
        [inline_without_move]: 5.86e-06
        [renormalize]: 0.00052619
        [add_forward_monad_depend]: 4.28001e-06
        [auto_monad_grad]: 2.22999e-06
        [auto_monad_eliminator]: 9.94999e-06
        [cse]: 1.618e-05
        [replace_applicator]: 1.152e-05
    [Cycle 2]: 0.00031592, [27]
        [switch_simplify]: 5.45001e-06
        [loop_unroll]: 4.86002e-06
        [a_1]: 8.89e-05
        [with_stream_mark]: 7.77e-06
        [recompute_prepare]: 4.90001e-06
        [updatestate_depend_eliminate]: 2.69999e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 1.10999e-06
        [specialize_transform]: 4.92e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 4.75999e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.48999e-06
        [merge_forward]: 2.29001e-06
        [cell_reuse_recompute_pass]: 1.97999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.24e-05
        [j_node_and_user_rematch]: 7.90998e-06
        [meta_fg_expand]: 1.40999e-06
        [replace_old_param]: 7.03998e-06
        [inline_without_move]: 4.85001e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.13001e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 5.02e-06
        [cse]: 1.016e-05
        [replace_applicator]: 5.00001e-06
[py_interpret_to_execute_after_opt_a]: 8.15e-06
[rewriter_after_opt_a]: 2.648e-05
[convert_after_rewriter]: 5.56e-06
[order_py_execute_after_rewriter]: 4.26001e-06
[mutable_eliminate]: 0.00054674
[jit_opt_b]: 6.001e-05, [1]
    [Cycle 1]: 5.282e-05, [2]
        [frontend_op_eliminate]: 2.096e-05
        [inline_after_opt_a]: 1.686e-05
[cconv]: 1.805e-05
[loop_unroll]: 0.00050967
[jit_opt_after_cconv]: 0.00014344, [1]
    [Cycle 1]: 0.00013709, [11]
        [c_1]: 2.271e-05
        [parameter_eliminate]: 2.76999e-06
        [updatestate_depend_eliminate]: 6.07999e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.18002e-06
        [cse]: 1.916e-05
        [call_graph_tuple_transform]: 1.817e-05
        [tuple_list_get_item_eliminator]: 5.51e-06
        [none_parameter_eliminate]: 1.07e-06
        [renormalize]: 3.30008e-07
        [switch_simplify]: 5.32999e-06
[remove_dup_value]: 7.15e-06
[partial_unused_args_eliminate]: 1.86e-06
[environ_conv]: 3.91001e-06
[add_recomputation]: 3.824e-05
[cse_after_recomputation]: 2.042e-05, [1]
    [Cycle 1]: 1.502e-05, [1]
        [cse]: 9.49e-06
[auto_monad_reorder]: 9.84999e-06
[get_jit_bprop_graph]: 1.39998e-06
[rewriter_after_jit_bprop_graph]: 4.77e-06
[opt_after_jit_grad]: 0.000467
[symbol_engine_optimizer]: 7.238e-05, [1]
    [Cycle 1]: 6.587e-05, [6]
        [build]: 2.85002e-06
        [elim_shapecalc]: 8.02e-06
        [elim_not_effective]: 1.315e-05
        [opt_reshape]: 5.47001e-06
        [fold_const_symbol]: 8.35999e-06
        [renormalize]: 4.00003e-07
[validate]: 2.686e-05
[backend_pass]: 8.70001e-07
[task_emit]: 0.0052766
[execute]: 7.30003e-06
Sums
    bootstrap                            :     0.000267s :  2.00%
    type_inference                       :     0.004279s : 32.10%
    event_method                         :     0.000018s :  0.14%
    auto_monad                           :     0.000064s :  0.48%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.23%
    rewriter_before_opt_a                :     0.000073s :  0.55%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000053s :  0.40%
    jit_opt_a.loop_unroll                :     0.000036s :  0.27%
    jit_opt_a.a_1                        :     0.000598s :  4.49%
    jit_opt_a.with_stream_mark           :     0.000020s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000023s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.12%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.000526s :  3.95%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000026s :  0.20%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000026s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000547s :  4.10%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.16%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.13%
    cconv                                :     0.000018s :  0.14%
    loop_unroll                          :     0.000510s :  3.82%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.17%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000007s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000038s :  0.29%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000010s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000467s :  3.50%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.20%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005277s : 39.58%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000172    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.59% :     0.000004s :      3: substitution.graph_param_transform
84.35% :     0.000145s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.28% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.74% :     0.000003s :      2: substitution.replace_old_param
 5.85% :     0.000010s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004215     2
82.03% :     0.003458s :      1: type_inference.infer
17.97% :     0.000757s :      1: type_inference.specialize
------[replace.]   0.000064     6
65.10% :     0.000042s :      4: replace.inline
34.90% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000152     6
94.14% :     0.000143s :      4: match.inline
 5.86% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000161   691
 0.89% :     0.000001s :     10: predicate.accumulaten_eliminater
 0.93% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.76% :     0.000001s :     10: predicate.addn_check_dump
 1.01% :     0.000002s :     10: predicate.addn_zero_filter
 1.84% :     0.000003s :     10: predicate.arithmetic_simplify
 0.89% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.79% :     0.000001s :     10: predicate.compare_switch_simplify
 0.94% :     0.000002s :     10: predicate.depend_value_elim
 0.83% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 0.89% :     0.000001s :     10: predicate.dict_get_item_eliminator
 0.86% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.60% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000001s :      3: predicate.elim_not_effective
 0.40% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.84% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.78% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.78% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.98% :     0.000002s :     10: predicate.environ_get_eliminate
 0.76% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000002s :      6: predicate.get_grad_eliminate
 0.18% :     0.000000s :      3: predicate.graph_param_transform
 3.97% :     0.000006s :     22: predicate.inline
 0.75% :     0.000001s :      6: predicate.inline_without_move
 0.28% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.07% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.32% :     0.000002s :     15: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.23% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.78% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.77% :     0.000001s :     10: predicate.merge_addn
 0.78% :     0.000001s :     10: predicate.minmaximum_grad
26.94% :     0.000043s :      3: predicate.mutable_eliminate
 0.35% :     0.000001s :      3: predicate.opt_reshape
 1.52% :     0.000002s :     15: predicate.partial_eliminate
 0.90% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.07% :     0.000002s :     10: predicate.reduce_eliminate
 1.07% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.34% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.27% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000001s :      3: predicate.reset_defer_inline
 0.94% :     0.000002s :     10: predicate.reshape_eliminate
 0.88% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.61% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.85% :     0.000001s :     10: predicate.same_eliminate
 0.40% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.71% :     0.000001s :      6: predicate.special_op_eliminate
 0.76% :     0.000001s :      6: predicate.specialize_transform
 1.04% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 0.88% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.34% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 1.86% :     0.000003s :     16: predicate.switch_defer_inline
 1.70% :     0.000003s :     16: predicate.switch_layer_defer_inline
 5.54% :     0.000009s :     53: predicate.switch_simplify
 0.84% :     0.000001s :     10: predicate.tile_eliminate
 1.00% :     0.000002s :     10: predicate.transpose_eliminate
 1.05% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.02% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 2.97% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.14% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.05% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.25% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.20% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.12% :     0.000002s :     10: predicate.value_based_eliminate
 0.26% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000506    12
47.67% :     0.000241s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.89% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.44% :     0.000235s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018684    76
 0.22% :     0.000041s :      1: add_recomputation
 0.36% :     0.000068s :      1: auto_monad
 0.07% :     0.000013s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.49% :     0.000279s :      1: bootstrap
 0.11% :     0.000021s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.12% :     0.000022s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
26.78% :     0.005004s :      1: jit_opt_a
 0.78% :     0.000146s :      1: jit_opt_after_cconv
 0.34% :     0.000063s :      1: jit_opt_b
 2.77% :     0.000518s :      1: loop_unroll
 2.99% :     0.000558s :      1: mutable_eliminate
 4.29% :     0.000802s :     26: opt.transform.jit_opt_a
 0.26% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000030s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.34% :     0.000063s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.54% :     0.000475s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.18% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000010s :      1: remove_dup_value
 1.42% :     0.000266s :      1: renormalize.infer
 1.36% :     0.000254s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000029s :      1: rewriter_after_opt_a
 0.41% :     0.000076s :      1: rewriter_before_opt_a
 0.40% :     0.000075s :      1: symbol_engine_optimizer
28.29% :     0.005287s :      1: task_emit
22.97% :     0.004292s :      1: type_inference
 0.23% :     0.000044s :      1: validate


TotalTime = 0.0311443, [33]
[bootstrap]: 0.00018748
[type_inference]: 0.00944382
[event_method]: 4.447e-05
[auto_monad]: 6.847e-05
[graph_reusing]: 4.78001e-06
[pre_auto_parallel]: 1.96003e-06
[py_interpret_to_execute]: 2.074e-05
[rewriter_before_opt_a]: 8.866e-05
[expand_dump_flag]: 2.74999e-06
[jit_opt_a]: 0.0117603, [3]
    [Cycle 1]: 0.00566208, [27]
        [switch_simplify]: 6.028e-05
        [loop_unroll]: 4.043e-05
        [a_1]: 0.00078228
        [with_stream_mark]: 2.067e-05
        [recompute_prepare]: 1.688e-05
        [updatestate_depend_eliminate]: 7.1e-06
        [updatestate_assign_eliminate]: 5.42999e-06
        [updatestate_loads_eliminate]: 5.20001e-06
        [parameter_eliminate]: 1.59998e-06
        [specialize_transform]: 1.285e-05
        [updatestate_useless_node_eliminater]: 1.14e-05
        [accelerated_algorithm]: 1.217e-05
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 1.174e-05
        [merge_forward]: 7.25e-06
        [cell_reuse_recompute_pass]: 7.09988e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.298e-05
        [j_node_and_user_rematch]: 2.193e-05
        [meta_fg_expand]: 0.00133672
        [replace_old_param]: 5.08e-05
        [inline_without_move]: 5.137e-05
        [renormalize]: 0.00268338
        [add_forward_monad_depend]: 9.35001e-06
        [auto_monad_grad]: 5.18002e-06
        [auto_monad_eliminator]: 5.345e-05
        [cse]: 0.00015654
        [replace_applicator]: 7.572e-05
    [Cycle 2]: 0.00252798, [27]
        [switch_simplify]: 3.956e-05
        [loop_unroll]: 3.892e-05
        [a_1]: 0.00120261
        [with_stream_mark]: 1.525e-05
        [recompute_prepare]: 9.66998e-06
        [updatestate_depend_eliminate]: 4.28999e-06
        [updatestate_assign_eliminate]: 3.58e-06
        [updatestate_loads_eliminate]: 3.29001e-06
        [parameter_eliminate]: 1.30999e-06
        [specialize_transform]: 7.41999e-06
        [updatestate_useless_node_eliminater]: 7.05e-06
        [accelerated_algorithm]: 5.606e-05
        [meta_shard_fg_expand]: 2.16e-06
        [get_grad_eliminate_]: 7.46001e-06
        [merge_forward]: 4.82e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.678e-05
        [j_node_and_user_rematch]: 3.154e-05
        [meta_fg_expand]: 7.978e-05
        [replace_old_param]: 1.244e-05
        [inline_without_move]: 7.60998e-06
        [renormalize]: 0.00071792
        [add_forward_monad_depend]: 5.46e-06
        [auto_monad_grad]: 1.59998e-06
        [auto_monad_eliminator]: 1.585e-05
        [cse]: 2.961e-05
        [replace_applicator]: 1.68e-05
    [Cycle 3]: 0.00043395, [27]
        [switch_simplify]: 7.95998e-06
        [loop_unroll]: 7.95998e-06
        [a_1]: 0.00015365
        [with_stream_mark]: 1.076e-05
        [recompute_prepare]: 6.78e-06
        [updatestate_depend_eliminate]: 3.66999e-06
        [updatestate_assign_eliminate]: 3.3e-06
        [updatestate_loads_eliminate]: 3.18998e-06
        [parameter_eliminate]: 9.5999e-07
        [specialize_transform]: 6.74999e-06
        [updatestate_useless_node_eliminater]: 6.39999e-06
        [accelerated_algorithm]: 7.06001e-06
        [meta_shard_fg_expand]: 1.80001e-06
        [get_grad_eliminate_]: 6.71e-06
        [merge_forward]: 3.8e-06
        [cell_reuse_recompute_pass]: 2.04999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.702e-05
        [j_node_and_user_rematch]: 1.322e-05
        [meta_fg_expand]: 2.80002e-06
        [replace_old_param]: 1.091e-05
        [inline_without_move]: 6.96999e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.49003e-06
        [auto_monad_grad]: 6.89994e-07
        [auto_monad_eliminator]: 8.40001e-06
        [cse]: 1.792e-05
        [replace_applicator]: 6.65002e-06
[py_interpret_to_execute_after_opt_a]: 1.171e-05
[rewriter_after_opt_a]: 4.751e-05
[convert_after_rewriter]: 8.26002e-06
[order_py_execute_after_rewriter]: 5.82001e-06
[mutable_eliminate]: 0.00068159
[jit_opt_b]: 6.477e-05, [1]
    [Cycle 1]: 5.722e-05, [2]
        [frontend_op_eliminate]: 2.402e-05
        [inline_after_opt_a]: 2.143e-05
[cconv]: 2.321e-05
[loop_unroll]: 0.00041397
[jit_opt_after_cconv]: 0.00016728, [1]
    [Cycle 1]: 0.0001607, [11]
        [c_1]: 2.969e-05
        [parameter_eliminate]: 2.68e-06
        [updatestate_depend_eliminate]: 6.51e-06
        [updatestate_assign_eliminate]: 3.29001e-06
        [updatestate_loads_eliminate]: 3.21001e-06
        [cse]: 2.791e-05
        [call_graph_tuple_transform]: 2.29e-05
        [tuple_list_get_item_eliminator]: 6.98e-06
        [none_parameter_eliminate]: 1.28002e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 7.35e-06
[remove_dup_value]: 2.074e-05
[partial_unused_args_eliminate]: 1.96e-06
[environ_conv]: 5.91998e-06
[add_recomputation]: 5.388e-05
[cse_after_recomputation]: 3.002e-05, [1]
    [Cycle 1]: 2.409e-05, [1]
        [cse]: 1.773e-05
[auto_monad_reorder]: 1.883e-05
[get_jit_bprop_graph]: 1.59e-06
[rewriter_after_jit_bprop_graph]: 6.06e-06
[opt_after_jit_grad]: 0.00047026
[symbol_engine_optimizer]: 8.441e-05, [1]
    [Cycle 1]: 7.807e-05, [6]
        [build]: 4.43999e-06
        [elim_shapecalc]: 9.47001e-06
        [elim_not_effective]: 1.618e-05
        [opt_reshape]: 8.33999e-06
        [fold_const_symbol]: 1.181e-05
        [renormalize]: 4.19997e-07
[validate]: 3.955e-05
[backend_pass]: 8.50006e-07
[task_emit]: 0.00716045
[execute]: 4.28999e-06
Sums
    bootstrap                            :     0.000187s :  0.69%
    type_inference                       :     0.009444s : 34.73%
    event_method                         :     0.000044s :  0.16%
    auto_monad                           :     0.000068s :  0.25%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000021s :  0.08%
    rewriter_before_opt_a                :     0.000089s :  0.33%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000108s :  0.40%
    jit_opt_a.loop_unroll                :     0.000087s :  0.32%
    jit_opt_a.a_1                        :     0.002139s :  7.86%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000027s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000075s :  0.28%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000026s :  0.10%
    jit_opt_a.merge_forward              :     0.000016s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000057s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000067s :  0.25%
    jit_opt_a.meta_fg_expand             :     0.001419s :  5.22%
    jit_opt_a.replace_old_param          :     0.000074s :  0.27%
    jit_opt_a.inline_without_move        :     0.000066s :  0.24%
    jit_opt_a.renormalize                :     0.003401s : 12.51%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000078s :  0.29%
    jit_opt_a.cse                        :     0.000204s :  0.75%
    jit_opt_a.replace_applicator         :     0.000099s :  0.36%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000048s :  0.17%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000682s :  2.51%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000021s :  0.08%
    cconv                                :     0.000023s :  0.09%
    loop_unroll                          :     0.000414s :  1.52%
    jit_opt_after_cconv.c_1              :     0.000030s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000028s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000021s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000054s :  0.20%
    cse_after_recomputation.cse          :     0.000018s :  0.07%
    auto_monad_reorder                   :     0.000019s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000470s :  1.73%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000040s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007160s : 26.33%
    execute                              :     0.000004s :  0.02%


Time group info:
------[substitution.]   0.000500   118
 5.32% :     0.000027s :      1: substitution.arithmetic_simplify
 0.51% :     0.000003s :      4: substitution.elim_not_effective
 0.37% :     0.000002s :      4: substitution.fold_const_symbol
 1.14% :     0.000006s :      5: substitution.graph_param_transform
59.51% :     0.000298s :     12: substitution.inline
 3.35% :     0.000017s :      2: substitution.inline_without_move
 1.82% :     0.000009s :     16: substitution.j_node_and_user_rematch
 2.08% :     0.000010s :      7: substitution.minmaximum_grad
 1.20% :     0.000006s :      5: substitution.partial_eliminate
 2.32% :     0.000012s :     16: substitution.remove_not_recompute_node
 5.63% :     0.000028s :     10: substitution.replace_applicator
 1.82% :     0.000009s :      7: substitution.replace_old_param
 0.42% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.14% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.91% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 7.46% :     0.000037s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009385     2
89.00% :     0.008352s :      1: type_inference.infer
11.00% :     0.001033s :      1: type_inference.specialize
------[replace.]   0.000143    20
 3.71% :     0.000005s :      1: replace.arithmetic_simplify
60.91% :     0.000087s :     12: replace.inline
35.38% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000333    20
 7.69% :     0.000026s :      1: match.arithmetic_simplify
87.40% :     0.000291s :     12: match.inline
 4.91% :     0.000016s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000456  2790
 1.60% :     0.000007s :     46: predicate.accumulaten_eliminater
 0.48% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.27% :     0.000006s :     46: predicate.addn_check_dump
 1.59% :     0.000007s :     46: predicate.addn_zero_filter
 1.83% :     0.000008s :     47: predicate.arithmetic_simplify
 1.62% :     0.000007s :     47: predicate.cast_eliminate
 0.17% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.26% :     0.000006s :     46: predicate.compare_switch_simplify
 1.37% :     0.000006s :     46: predicate.depend_value_elim
 1.38% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.42% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.40% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.27% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.12% :     0.000001s :      5: predicate.elim_not_effective
 0.20% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.50% :     0.000007s :     47: predicate.environ_add_const_eliminate
 1.28% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.33% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.41% :     0.000006s :     47: predicate.environ_get_eliminate
 1.30% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      5: predicate.fold_const_symbol
 0.82% :     0.000004s :     21: predicate.get_grad_eliminate
 0.10% :     0.000000s :      5: predicate.graph_param_transform
 3.67% :     0.000017s :     76: predicate.inline
 1.50% :     0.000007s :     42: predicate.inline_without_move
 0.34% :     0.000002s :     21: predicate.j_node_and_user_rematch
11.37% :     0.000052s :     21: predicate.less_batch_normalization
 1.68% :     0.000008s :     54: predicate.list_to_tuple_eliminator_
 1.71% :     0.000008s :     59: predicate.load_eliminater
 0.50% :     0.000002s :      5: predicate.loop_unroll_after_grad
 2.82% :     0.000013s :     89: predicate.loop_unroll_before_grad
 1.64% :     0.000007s :     52: predicate.make_slice_get_slice_eliminator
 1.29% :     0.000006s :     46: predicate.merge_addn
 1.29% :     0.000006s :     47: predicate.minmaximum_grad
 0.48% :     0.000002s :      5: predicate.mutable_eliminate
 0.20% :     0.000001s :      5: predicate.opt_reshape
 2.03% :     0.000009s :     59: predicate.partial_eliminate
 1.33% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.66% :     0.000008s :     47: predicate.reduce_eliminate
 1.81% :     0.000008s :     54: predicate.redundant_stop_gradient_eliminater
 0.39% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.29% :     0.000010s :    105: predicate.replace_applicator
 0.75% :     0.000003s :     42: predicate.replace_old_param
 0.11% :     0.000001s :      5: predicate.reset_defer_inline
 1.54% :     0.000007s :     47: predicate.reshape_eliminate
 1.35% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.34% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.36% :     0.000006s :     46: predicate.same_eliminate
 0.40% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.35% :     0.000002s :     10: predicate.special_op_eliminate
 0.79% :     0.000004s :     21: predicate.specialize_transform
 1.57% :     0.000007s :     46: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.18% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.33% :     0.000011s :     66: predicate.switch_defer_inline
 2.16% :     0.000010s :     66: predicate.switch_layer_defer_inline
 5.38% :     0.000025s :    160: predicate.switch_simplify
 1.34% :     0.000006s :     47: predicate.tile_eliminate
 1.43% :     0.000007s :     47: predicate.transpose_eliminate
 1.60% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000007s :     47: predicate.tuple_list_get_item_depend_reorder
 2.62% :     0.000012s :     64: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000008s :     47: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000008s :     54: predicate.tuple_to_list_eliminator_
 1.67% :     0.000008s :     59: predicate.updatestate_pure_node_eliminater
 2.59% :     0.000012s :     80: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000008s :     46: predicate.value_based_eliminate
 0.16% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.26% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001392    29
56.35% :     0.000784s :     13: func_graph_cloner_run.FuncGraphClonerGraph
43.65% :     0.000608s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.037516    91
 0.15% :     0.000057s :      1: add_recomputation
 0.19% :     0.000073s :      1: auto_monad
 0.06% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.54% :     0.000203s :      1: bootstrap
 0.07% :     0.000026s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000032s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.13% :     0.000049s :      1: event_method
 0.02% :     0.000008s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
31.36% :     0.011764s :      1: jit_opt_a
 0.45% :     0.000170s :      1: jit_opt_after_cconv
 0.18% :     0.000067s :      1: jit_opt_b
 1.12% :     0.000422s :      1: loop_unroll
 1.84% :     0.000692s :      1: mutable_eliminate
 7.51% :     0.002817s :     39: opt.transform.jit_opt_a
 0.17% :     0.000064s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000039s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000043s :      4: opt.transform.symbol_engine_opt
 1.27% :     0.000478s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.06% :     0.000023s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000023s :      1: remove_dup_value
 4.66% :     0.001746s :      2: renormalize.infer
 4.37% :     0.001641s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000050s :      1: rewriter_after_opt_a
 0.24% :     0.000092s :      1: rewriter_before_opt_a
 0.23% :     0.000087s :      1: symbol_engine_optimizer
19.12% :     0.007171s :      1: task_emit
25.21% :     0.009459s :      1: type_inference
 0.16% :     0.000062s :      1: validate


TotalTime = 0.0228278, [33]
[bootstrap]: 0.00019889
[type_inference]: 0.0113245
[event_method]: 5.056e-05
[auto_monad]: 9.176e-05
[graph_reusing]: 6.25002e-06
[pre_auto_parallel]: 2.84999e-06
[py_interpret_to_execute]: 2.545e-05
[rewriter_before_opt_a]: 8.876e-05
[expand_dump_flag]: 3.18998e-06
[jit_opt_a]: 0.00881348, [2]
    [Cycle 1]: 0.00495394, [27]
        [switch_simplify]: 6.363e-05
        [loop_unroll]: 3.954e-05
        [a_1]: 0.00085444
        [with_stream_mark]: 2.448e-05
        [recompute_prepare]: 1.816e-05
        [updatestate_depend_eliminate]: 7.86001e-06
        [updatestate_assign_eliminate]: 6.59999e-06
        [updatestate_loads_eliminate]: 6.39999e-06
        [parameter_eliminate]: 2.37999e-06
        [specialize_transform]: 1.332e-05
        [updatestate_useless_node_eliminater]: 1.242e-05
        [accelerated_algorithm]: 1.234e-05
        [meta_shard_fg_expand]: 3.48e-06
        [get_grad_eliminate_]: 1.226e-05
        [merge_forward]: 8.28999e-06
        [cell_reuse_recompute_pass]: 8.89995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.668e-05
        [j_node_and_user_rematch]: 2.293e-05
        [meta_fg_expand]: 0.00149386
        [replace_old_param]: 6.399e-05
        [inline_without_move]: 5.447e-05
        [renormalize]: 0.00194221
        [add_forward_monad_depend]: 6.03002e-06
        [auto_monad_grad]: 2.31e-06
        [auto_monad_eliminator]: 1.222e-05
        [cse]: 2.133e-05
        [replace_applicator]: 1.461e-05
    [Cycle 2]: 0.00031297, [27]
        [switch_simplify]: 5.24e-06
        [loop_unroll]: 3.88999e-06
        [a_1]: 4.725e-05
        [with_stream_mark]: 9.50001e-06
        [recompute_prepare]: 4.09002e-06
        [updatestate_depend_eliminate]: 2.39001e-06
        [updatestate_assign_eliminate]: 1.81e-06
        [updatestate_loads_eliminate]: 1.82001e-06
        [parameter_eliminate]: 1.62001e-06
        [specialize_transform]: 3.58999e-06
        [updatestate_useless_node_eliminater]: 3.43e-06
        [accelerated_algorithm]: 7.70998e-06
        [meta_shard_fg_expand]: 2.13002e-06
        [get_grad_eliminate_]: 3.48999e-06
        [merge_forward]: 3.09999e-06
        [cell_reuse_recompute_pass]: 3.63e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.257e-05
        [j_node_and_user_rematch]: 6.98e-06
        [meta_fg_expand]: 4.094e-05
        [replace_old_param]: 4.23999e-06
        [inline_without_move]: 3.35e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.50001e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 4.54998e-06
        [cse]: 1.048e-05
        [replace_applicator]: 4.07998e-06
[py_interpret_to_execute_after_opt_a]: 1.032e-05
[rewriter_after_opt_a]: 3.41e-05
[convert_after_rewriter]: 6.28998e-06
[order_py_execute_after_rewriter]: 4.57998e-06
[mutable_eliminate]: 0.00062982
[jit_opt_b]: 4.087e-05, [1]
    [Cycle 1]: 3.335e-05, [2]
        [frontend_op_eliminate]: 1.079e-05
        [inline_after_opt_a]: 1.118e-05
[cconv]: 2.308e-05
[loop_unroll]: 0.00044081
[jit_opt_after_cconv]: 0.00011955, [1]
    [Cycle 1]: 0.00011309, [11]
        [c_1]: 1.313e-05
        [parameter_eliminate]: 2.70002e-06
        [updatestate_depend_eliminate]: 4.77e-06
        [updatestate_assign_eliminate]: 1.83002e-06
        [updatestate_loads_eliminate]: 1.54998e-06
        [cse]: 1.547e-05
        [call_graph_tuple_transform]: 1.362e-05
        [tuple_list_get_item_eliminator]: 3.92998e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 3.99002e-06
[remove_dup_value]: 1.415e-05
[partial_unused_args_eliminate]: 2.19001e-06
[environ_conv]: 5.36002e-06
[add_recomputation]: 3.311e-05
[cse_after_recomputation]: 2.041e-05, [1]
    [Cycle 1]: 1.499e-05, [1]
        [cse]: 8.65999e-06
[auto_monad_reorder]: 1.268e-05
[get_jit_bprop_graph]: 1.41002e-06
[rewriter_after_jit_bprop_graph]: 4.77e-06
[opt_after_jit_grad]: 0.00052928
[symbol_engine_optimizer]: 6.604e-05, [1]
    [Cycle 1]: 6.02e-05, [6]
        [build]: 3.18998e-06
        [elim_shapecalc]: 5.91e-06
        [elim_not_effective]: 1.037e-05
        [opt_reshape]: 4.65999e-06
        [fold_const_symbol]: 7.05e-06
        [renormalize]: 5.50004e-07
[validate]: 2.87e-05
[backend_pass]: 8.00006e-07
[task_emit]: 2.043e-05
[execute]: 1.50001e-06
Sums
    bootstrap                            :     0.000199s :  1.07%
    type_inference                       :     0.011325s : 60.68%
    event_method                         :     0.000051s :  0.27%
    auto_monad                           :     0.000092s :  0.49%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000025s :  0.14%
    rewriter_before_opt_a                :     0.000089s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000069s :  0.37%
    jit_opt_a.loop_unroll                :     0.000043s :  0.23%
    jit_opt_a.a_1                        :     0.000902s :  4.83%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000020s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.08%
    jit_opt_a.merge_forward              :     0.000011s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000030s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001535s :  8.22%
    jit_opt_a.replace_old_param          :     0.000068s :  0.37%
    jit_opt_a.inline_without_move        :     0.000058s :  0.31%
    jit_opt_a.renormalize                :     0.001942s : 10.41%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.09%
    jit_opt_a.cse                        :     0.000032s :  0.17%
    jit_opt_a.replace_applicator         :     0.000019s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000034s :  0.18%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000630s :  3.37%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.06%
    cconv                                :     0.000023s :  0.12%
    loop_unroll                          :     0.000441s :  2.36%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000033s :  0.18%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000529s :  2.84%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000007s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000029s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000020s :  0.11%
    execute                              :     0.000002s :  0.01%


Time group info:
------[substitution.]   0.000259    50
 0.56% :     0.000001s :      1: substitution.elim_not_effective
 0.50% :     0.000001s :      1: substitution.fold_const_symbol
 1.46% :     0.000004s :      1: substitution.graph_param_transform
66.19% :     0.000172s :      6: substitution.inline
 7.06% :     0.000018s :      2: substitution.inline_without_move
 2.82% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.57% :     0.000004s :      2: substitution.minmaximum_grad
 1.58% :     0.000004s :      3: substitution.partial_eliminate
 2.98% :     0.000008s :      9: substitution.remove_not_recompute_node
 2.05% :     0.000005s :      2: substitution.replace_applicator
 1.62% :     0.000004s :      5: substitution.replace_old_param
 0.89% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.84% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 2.07% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.82% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011259     2
89.29% :     0.010053s :      1: type_inference.infer
10.71% :     0.001205s :      1: type_inference.specialize
------[replace.]   0.000066     8
78.48% :     0.000051s :      6: replace.inline
21.52% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000179     8
94.23% :     0.000168s :      6: match.inline
 5.77% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000173  1109
 1.62% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.64% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.22% :     0.000002s :     18: predicate.addn_check_dump
 1.43% :     0.000002s :     18: predicate.addn_zero_filter
 2.15% :     0.000004s :     18: predicate.arithmetic_simplify
 1.56% :     0.000003s :     18: predicate.cast_eliminate
 0.14% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.26% :     0.000002s :     18: predicate.compare_switch_simplify
 1.43% :     0.000002s :     18: predicate.depend_value_elim
 1.41% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.41% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.42% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      1: predicate.elim_not_effective
 0.31% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.26% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.27% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.40% :     0.000002s :     18: predicate.environ_get_eliminate
 1.23% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.17% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.30% :     0.000007s :     28: predicate.inline
 3.30% :     0.000006s :     33: predicate.inline_without_move
 0.50% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.56% :     0.000003s :     12: predicate.less_batch_normalization
 1.73% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.77% :     0.000003s :     21: predicate.load_eliminater
 0.66% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.52% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.84% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.28% :     0.000002s :     18: predicate.merge_addn
 1.40% :     0.000002s :     18: predicate.minmaximum_grad
 0.81% :     0.000001s :      1: predicate.mutable_eliminate
 0.18% :     0.000000s :      1: predicate.opt_reshape
 2.04% :     0.000004s :     21: predicate.partial_eliminate
 1.40% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.96% :     0.000003s :     18: predicate.reduce_eliminate
 1.60% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.67% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.73% :     0.000003s :     22: predicate.replace_applicator
 1.66% :     0.000003s :     33: predicate.replace_old_param
 0.10% :     0.000000s :      1: predicate.reset_defer_inline
 1.63% :     0.000003s :     18: predicate.reshape_eliminate
 1.41% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.41% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     18: predicate.same_eliminate
 0.62% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.28% :     0.000000s :      2: predicate.special_op_eliminate
 1.18% :     0.000002s :     12: predicate.specialize_transform
 2.02% :     0.000004s :     18: predicate.split_environ_get_set_with_tuple_value
 1.48% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.45% :     0.000004s :     26: predicate.switch_defer_inline
 2.23% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.29% :     0.000011s :     71: predicate.switch_simplify
 1.56% :     0.000003s :     18: predicate.tile_eliminate
 1.65% :     0.000003s :     18: predicate.transpose_eliminate
 1.76% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.54% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.20% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.94% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.56% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000003s :     18: predicate.value_based_eliminate
 0.12% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.31% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001294    22
72.37% :     0.000937s :     13: func_graph_cloner_run.FuncGraphClonerGraph
27.63% :     0.000358s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026110    76
 0.14% :     0.000036s :      1: add_recomputation
 0.37% :     0.000096s :      1: auto_monad
 0.06% :     0.000015s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.80% :     0.000210s :      1: bootstrap
 0.10% :     0.000026s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.21% :     0.000055s :      1: event_method
 0.02% :     0.000005s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
33.77% :     0.008817s :      1: jit_opt_a
 0.47% :     0.000123s :      1: jit_opt_after_cconv
 0.17% :     0.000043s :      1: jit_opt_b
 1.72% :     0.000449s :      1: loop_unroll
 2.45% :     0.000639s :      1: mutable_eliminate
 4.89% :     0.001277s :     26: opt.transform.jit_opt_a
 0.12% :     0.000031s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000024s :      4: opt.transform.symbol_engine_opt
 2.06% :     0.000537s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000028s :      1: py_interpret_to_execute
 0.05% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000017s :      1: remove_dup_value
 5.50% :     0.001437s :      1: renormalize.infer
 1.90% :     0.000496s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000037s :      1: rewriter_after_opt_a
 0.35% :     0.000091s :      1: rewriter_before_opt_a
 0.26% :     0.000069s :      1: symbol_engine_optimizer
 0.09% :     0.000024s :      1: task_emit
43.42% :     0.011336s :      1: type_inference
 0.17% :     0.000045s :      1: validate

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.576.493 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda84a8af0> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.577.031 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.580.732 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c3680d0> should be wrapped by jit only once.

TotalTime = 0.0137581, [30]
[bootstrap]: 0.0003532
[type_inference]: 0.00479078
[event_method]: 1.915e-05
[auto_monad]: 6.839e-05
[graph_reusing]: 6.04999e-06
[pre_auto_parallel]: 2.78e-06
[py_interpret_to_execute]: 3.189e-05
[rewriter_before_opt_a]: 7.182e-05
[expand_dump_flag]: 3.04001e-06
[jit_opt_a]: 0.00603172, [2]
    [Cycle 1]: 0.00179288, [27]
        [switch_simplify]: 6.544e-05
        [loop_unroll]: 2.907e-05
        [a_1]: 0.00055453
        [with_stream_mark]: 2.532e-05
        [recompute_prepare]: 7.56001e-06
        [updatestate_depend_eliminate]: 4.20999e-06
        [updatestate_assign_eliminate]: 3.19001e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [parameter_eliminate]: 2.11003e-06
        [specialize_transform]: 5.84e-06
        [updatestate_useless_node_eliminater]: 4.65999e-06
        [accelerated_algorithm]: 5.52999e-06
        [meta_shard_fg_expand]: 2.39001e-06
        [get_grad_eliminate_]: 4.94e-06
        [merge_forward]: 3.60998e-06
        [cell_reuse_recompute_pass]: 1.07998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.541e-05
        [j_node_and_user_rematch]: 9.57999e-06
        [meta_fg_expand]: 2.13002e-06
        [replace_old_param]: 9.61998e-06
        [inline_without_move]: 5.52999e-06
        [renormalize]: 0.00076108
        [add_forward_monad_depend]: 6.64001e-06
        [auto_monad_grad]: 2.48002e-06
        [auto_monad_eliminator]: 1.758e-05
        [cse]: 3.107e-05
        [replace_applicator]: 1.374e-05
    [Cycle 2]: 0.00032748, [27]
        [switch_simplify]: 5.68002e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 9.128e-05
        [with_stream_mark]: 1.011e-05
        [recompute_prepare]: 4.80001e-06
        [updatestate_depend_eliminate]: 2.94001e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.10002e-06
        [parameter_eliminate]: 8.89995e-07
        [specialize_transform]: 4.70001e-06
        [updatestate_useless_node_eliminater]: 4.61002e-06
        [accelerated_algorithm]: 4.97e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 4.28999e-06
        [merge_forward]: 3.02002e-06
        [cell_reuse_recompute_pass]: 2.78998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.6e-05
        [j_node_and_user_rematch]: 8.86002e-06
        [meta_fg_expand]: 1.57001e-06
        [replace_old_param]: 7.96001e-06
        [inline_without_move]: 4.95999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 5.32999e-06
        [cse]: 1.091e-05
        [replace_applicator]: 5.19e-06
[py_interpret_to_execute_after_opt_a]: 1.461e-05
[rewriter_after_opt_a]: 3.801e-05
[convert_after_rewriter]: 7.05e-06
[order_py_execute_after_rewriter]: 5.07e-06
[mutable_eliminate]: 0.00070666
[jit_opt_b]: 5.108e-05, [1]
    [Cycle 1]: 4.332e-05, [2]
        [frontend_op_eliminate]: 1.627e-05
        [inline_after_opt_a]: 1.595e-05
[cconv]: 3.042e-05
[loop_unroll]: 0.00044423
[jit_opt_after_cconv]: 0.00014997, [1]
    [Cycle 1]: 0.00014299, [11]
        [c_1]: 1.951e-05
        [parameter_eliminate]: 3.46999e-06
        [updatestate_depend_eliminate]: 8.48999e-06
        [updatestate_assign_eliminate]: 2.54001e-06
        [updatestate_loads_eliminate]: 2.51e-06
        [cse]: 2.395e-05
        [call_graph_tuple_transform]: 2.112e-05
        [tuple_list_get_item_eliminator]: 5.30001e-06
        [none_parameter_eliminate]: 1.44998e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.59e-06
[remove_dup_value]: 1.571e-05
[partial_unused_args_eliminate]: 2.44999e-06
[environ_conv]: 5.61e-06
[add_recomputation]: 5.756e-05
[cse_after_recomputation]: 2.051e-05, [1]
    [Cycle 1]: 1.47e-05, [1]
        [cse]: 8.70001e-06
[auto_monad_reorder]: 1.761e-05
[get_jit_bprop_graph]: 2.11e-06
[rewriter_after_jit_bprop_graph]: 6.81001e-06
[opt_after_jit_grad]: 0.00048914
[symbol_engine_optimizer]: 7.63e-05, [1]
    [Cycle 1]: 6.892e-05, [6]
        [build]: 4.32e-06
        [elim_shapecalc]: 7.56999e-06
        [elim_not_effective]: 1.362e-05
        [opt_reshape]: 5.77999e-06
        [fold_const_symbol]: 9.20999e-06
        [renormalize]: 7.30011e-07
[validate]: 4.403e-05
Sums
    bootstrap                            :     0.000353s :  3.83%
    type_inference                       :     0.004791s : 51.95%
    event_method                         :     0.000019s :  0.21%
    auto_monad                           :     0.000068s :  0.74%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.35%
    rewriter_before_opt_a                :     0.000072s :  0.78%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000071s :  0.77%
    jit_opt_a.loop_unroll                :     0.000034s :  0.37%
    jit_opt_a.a_1                        :     0.000646s :  7.00%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.38%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000761s :  8.25%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.25%
    jit_opt_a.cse                        :     0.000042s :  0.46%
    jit_opt_a.replace_applicator         :     0.000019s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.16%
    rewriter_after_opt_a                 :     0.000038s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000707s :  7.66%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.17%
    cconv                                :     0.000030s :  0.33%
    loop_unroll                          :     0.000444s :  4.82%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.09%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000024s :  0.26%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000016s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000058s :  0.62%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000018s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000489s :  5.30%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.48%


Time group info:
------[substitution.]   0.000230    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000001s :      2: substitution.fold_const_symbol
 2.54% :     0.000006s :      3: substitution.graph_param_transform
83.08% :     0.000191s :      4: substitution.inline
 1.78% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.56% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.48% :     0.000006s :      2: substitution.replace_old_param
 5.85% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004711     2
82.56% :     0.003889s :      1: type_inference.infer
17.44% :     0.000821s :      1: type_inference.specialize
------[replace.]   0.000068     6
65.99% :     0.000045s :      4: replace.inline
34.01% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000201     6
93.84% :     0.000189s :      4: match.inline
 6.16% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.74% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.41% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 2.17% :     0.000003s :     10: predicate.addn_zero_filter
 1.93% :     0.000002s :     10: predicate.arithmetic_simplify
 1.09% :     0.000001s :     10: predicate.cast_eliminate
 0.53% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.54% :     0.000002s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.46% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.14% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.65% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.81% :     0.000001s :      6: predicate.get_grad_eliminate
 0.40% :     0.000000s :      3: predicate.graph_param_transform
 5.81% :     0.000007s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.44% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.62% :     0.000002s :     15: predicate.load_eliminater
 1.55% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.70% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 0.95% :     0.000001s :     10: predicate.minmaximum_grad
 1.86% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.33% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.70% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.54% :     0.000001s :      3: predicate.reset_defer_inline
 1.61% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.04% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.94% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.12% :     0.000003s :     16: predicate.switch_defer_inline
 1.89% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.45% :     0.000009s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.23% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.44% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.35% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000623    12
51.80% :     0.000323s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.08% :     0.000050s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.12% :     0.000250s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015433    72
 0.39% :     0.000061s :      1: add_recomputation
 0.46% :     0.000071s :      1: auto_monad
 0.13% :     0.000020s :      1: auto_monad_reorder
 2.44% :     0.000377s :      1: bootstrap
 0.21% :     0.000033s :      1: cconv
 0.07% :     0.000011s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.11% :     0.006035s :      1: jit_opt_a
 0.99% :     0.000153s :      1: jit_opt_after_cconv
 0.35% :     0.000054s :      1: jit_opt_b
 2.94% :     0.000453s :      1: loop_unroll
 4.66% :     0.000719s :      1: mutable_eliminate
 5.62% :     0.000868s :     26: opt.transform.jit_opt_a
 0.31% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.23% :     0.000498s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000035s :      1: py_interpret_to_execute
 0.11% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000019s :      1: remove_dup_value
 2.70% :     0.000417s :      1: renormalize.infer
 2.17% :     0.000335s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000042s :      1: rewriter_after_opt_a
 0.48% :     0.000075s :      1: rewriter_before_opt_a
 0.51% :     0.000079s :      1: symbol_engine_optimizer
31.16% :     0.004808s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.581.343 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0196026, [30]
[bootstrap]: 0.00017151
[type_inference]: 0.00914301
[event_method]: 4.396e-05
[auto_monad]: 9.068e-05
[graph_reusing]: 7.48e-06
[pre_auto_parallel]: 3.01001e-06
[py_interpret_to_execute]: 2.333e-05
[rewriter_before_opt_a]: 8.835e-05
[expand_dump_flag]: 2.98e-06
[jit_opt_a]: 0.00809193, [2]
    [Cycle 1]: 0.00462234, [27]
        [switch_simplify]: 5.728e-05
        [loop_unroll]: 3.718e-05
        [a_1]: 0.00082086
        [with_stream_mark]: 2.619e-05
        [recompute_prepare]: 1.714e-05
        [updatestate_depend_eliminate]: 6.86999e-06
        [updatestate_assign_eliminate]: 6.04001e-06
        [updatestate_loads_eliminate]: 6.08002e-06
        [parameter_eliminate]: 2.83e-06
        [specialize_transform]: 1.257e-05
        [updatestate_useless_node_eliminater]: 0.00032127
        [accelerated_algorithm]: 1.36e-05
        [meta_shard_fg_expand]: 3.33998e-06
        [get_grad_eliminate_]: 1.265e-05
        [merge_forward]: 9.19e-06
        [cell_reuse_recompute_pass]: 1.04998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.608e-05
        [j_node_and_user_rematch]: 2.332e-05
        [meta_fg_expand]: 0.00138846
        [replace_old_param]: 5.076e-05
        [inline_without_move]: 5.04e-05
        [renormalize]: 0.00147582
        [add_forward_monad_depend]: 5.20999e-06
        [auto_monad_grad]: 2.11e-06
        [auto_monad_eliminator]: 1.153e-05
        [cse]: 2.283e-05
        [replace_applicator]: 9.67999e-06
    [Cycle 2]: 0.00027225, [27]
        [switch_simplify]: 4.37998e-06
        [loop_unroll]: 3.77998e-06
        [a_1]: 3.736e-05
        [with_stream_mark]: 9.22999e-06
        [recompute_prepare]: 3.68e-06
        [updatestate_depend_eliminate]: 2.62001e-06
        [updatestate_assign_eliminate]: 1.69998e-06
        [updatestate_loads_eliminate]: 2.08002e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 3.48e-06
        [updatestate_useless_node_eliminater]: 3.38e-06
        [accelerated_algorithm]: 5.32999e-06
        [meta_shard_fg_expand]: 1.72999e-06
        [get_grad_eliminate_]: 3.16999e-06
        [merge_forward]: 1.99e-06
        [cell_reuse_recompute_pass]: 2.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.027e-05
        [j_node_and_user_rematch]: 5.85002e-06
        [meta_fg_expand]: 2.492e-05
        [replace_old_param]: 3.81999e-06
        [inline_without_move]: 3.23998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.43002e-06
        [auto_monad_grad]: 9.89996e-07
        [auto_monad_eliminator]: 4.62e-06
        [cse]: 1.101e-05
        [replace_applicator]: 3.76999e-06
[py_interpret_to_execute_after_opt_a]: 8.15e-06
[rewriter_after_opt_a]: 3.099e-05
[convert_after_rewriter]: 6.00002e-06
[order_py_execute_after_rewriter]: 4.34002e-06
[mutable_eliminate]: 0.00053298
[jit_opt_b]: 3.889e-05, [1]
    [Cycle 1]: 3.221e-05, [2]
        [frontend_op_eliminate]: 9.84001e-06
        [inline_after_opt_a]: 1.122e-05
[cconv]: 2.069e-05
[loop_unroll]: 0.00041024
[jit_opt_after_cconv]: 0.00011674, [1]
    [Cycle 1]: 0.00011022, [11]
        [c_1]: 1.164e-05
        [parameter_eliminate]: 2.91e-06
        [updatestate_depend_eliminate]: 4.66002e-06
        [updatestate_assign_eliminate]: 1.91998e-06
        [updatestate_loads_eliminate]: 1.69e-06
        [cse]: 1.559e-05
        [call_graph_tuple_transform]: 1.295e-05
        [tuple_list_get_item_eliminator]: 3.78001e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 3.71999e-06
[remove_dup_value]: 1.451e-05
[partial_unused_args_eliminate]: 2.35002e-06
[environ_conv]: 5.25999e-06
[add_recomputation]: 3.168e-05
[cse_after_recomputation]: 1.909e-05, [1]
    [Cycle 1]: 1.37e-05, [1]
        [cse]: 7.68999e-06
[auto_monad_reorder]: 1.178e-05
[get_jit_bprop_graph]: 1.42999e-06
[rewriter_after_jit_bprop_graph]: 5.39e-06
[opt_after_jit_grad]: 0.00043605
[symbol_engine_optimizer]: 6.2e-05, [1]
    [Cycle 1]: 5.612e-05, [6]
        [build]: 3.31001e-06
        [elim_shapecalc]: 6.17001e-06
        [elim_not_effective]: 9.74e-06
        [opt_reshape]: 3.86999e-06
        [fold_const_symbol]: 5.50001e-06
        [renormalize]: 4.2998e-07
[validate]: 2.694e-05
Sums
    bootstrap                            :     0.000172s :  1.08%
    type_inference                       :     0.009143s : 57.79%
    event_method                         :     0.000044s :  0.28%
    auto_monad                           :     0.000091s :  0.57%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000023s :  0.15%
    rewriter_before_opt_a                :     0.000088s :  0.56%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.39%
    jit_opt_a.loop_unroll                :     0.000041s :  0.26%
    jit_opt_a.a_1                        :     0.000858s :  5.42%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000016s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000325s :  2.05%
    jit_opt_a.accelerated_algorithm      :     0.000019s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.10%
    jit_opt_a.merge_forward              :     0.000011s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001413s :  8.93%
    jit_opt_a.replace_old_param          :     0.000055s :  0.35%
    jit_opt_a.inline_without_move        :     0.000054s :  0.34%
    jit_opt_a.renormalize                :     0.001476s :  9.33%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.10%
    jit_opt_a.cse                        :     0.000034s :  0.21%
    jit_opt_a.replace_applicator         :     0.000013s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000031s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000533s :  3.37%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000021s :  0.13%
    loop_unroll                          :     0.000410s :  2.59%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000032s :  0.20%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000012s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000436s :  2.76%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.17%


Time group info:
------[substitution.]   0.000240    50
 0.61% :     0.000001s :      1: substitution.elim_not_effective
 0.36% :     0.000001s :      1: substitution.fold_const_symbol
 1.51% :     0.000004s :      1: substitution.graph_param_transform
66.20% :     0.000159s :      6: substitution.inline
 6.81% :     0.000016s :      2: substitution.inline_without_move
 2.95% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.57% :     0.000004s :      2: substitution.minmaximum_grad
 1.48% :     0.000004s :      3: substitution.partial_eliminate
 2.90% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.26% :     0.000005s :      2: substitution.replace_applicator
 1.69% :     0.000004s :      5: substitution.replace_old_param
 1.00% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.71% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.88% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 6.06% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009078     2
88.40% :     0.008025s :      1: type_inference.infer
11.60% :     0.001053s :      1: type_inference.specialize
------[replace.]   0.000061     8
78.10% :     0.000048s :      6: replace.inline
21.90% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000166     8
94.11% :     0.000156s :      6: match.inline
 5.89% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000469  1109
 0.52% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.25% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 0.46% :     0.000002s :     18: predicate.addn_check_dump
 0.56% :     0.000003s :     18: predicate.addn_zero_filter
 0.73% :     0.000003s :     18: predicate.arithmetic_simplify
 0.51% :     0.000002s :     18: predicate.cast_eliminate
 0.06% :     0.000000s :      1: predicate.check_bprop_eliminate
 0.48% :     0.000002s :     18: predicate.compare_switch_simplify
 0.52% :     0.000002s :     18: predicate.depend_value_elim
 0.46% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 0.51% :     0.000002s :     18: predicate.dict_get_item_eliminator
 0.49% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.18% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.04% :     0.000000s :      1: predicate.elim_not_effective
 0.09% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 0.52% :     0.000002s :     18: predicate.environ_add_const_eliminate
 0.44% :     0.000002s :     18: predicate.environ_get_add_eliminate
 0.45% :     0.000002s :     18: predicate.environ_get_depend_swap
 0.47% :     0.000002s :     18: predicate.environ_get_eliminate
 0.44% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.02% :     0.000000s :      1: predicate.fold_const_symbol
 0.45% :     0.000002s :     12: predicate.get_grad_eliminate
 0.04% :     0.000000s :      1: predicate.graph_param_transform
 1.54% :     0.000007s :     28: predicate.inline
 1.12% :     0.000005s :     33: predicate.inline_without_move
 0.18% :     0.000001s :     12: predicate.j_node_and_user_rematch
 0.49% :     0.000002s :     12: predicate.less_batch_normalization
 0.59% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 0.65% :     0.000003s :     21: predicate.load_eliminater
 0.28% :     0.000001s :      1: predicate.loop_unroll_after_grad
 1.28% :     0.000006s :     44: predicate.loop_unroll_before_grad
 0.64% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 0.50% :     0.000002s :     18: predicate.merge_addn
 0.45% :     0.000002s :     18: predicate.minmaximum_grad
 0.37% :     0.000002s :      1: predicate.mutable_eliminate
 0.07% :     0.000000s :      1: predicate.opt_reshape
 0.73% :     0.000003s :     21: predicate.partial_eliminate
 0.47% :     0.000002s :     18: predicate.print_const_string_wrapper
 0.60% :     0.000003s :     18: predicate.reduce_eliminate
 0.55% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.23% :     0.000001s :     12: predicate.remove_not_recompute_node
 0.53% :     0.000002s :     22: predicate.replace_applicator
 0.60% :     0.000003s :     33: predicate.replace_old_param
 0.03% :     0.000000s :      1: predicate.reset_defer_inline
 0.59% :     0.000003s :     18: predicate.reshape_eliminate
 0.48% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.13% :     0.000001s :      1: predicate.row_tensor_eliminate
 0.48% :     0.000002s :     18: predicate.same_eliminate
 0.22% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.14% :     0.000001s :      2: predicate.special_op_eliminate
 0.42% :     0.000002s :     12: predicate.specialize_transform
 0.54% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 0.49% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.04% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 0.98% :     0.000005s :     26: predicate.switch_defer_inline
 0.83% :     0.000004s :     26: predicate.switch_layer_defer_inline
 2.40% :     0.000011s :     71: predicate.switch_simplify
 0.50% :     0.000002s :     18: predicate.tile_eliminate
 0.49% :     0.000002s :     18: predicate.transpose_eliminate
 0.56% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 0.56% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 1.10% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 0.56% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 0.57% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 0.56% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
66.09% :     0.000310s :     33: predicate.updatestate_useless_node_eliminater
 0.56% :     0.000003s :     18: predicate.value_based_eliminate
 0.04% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.09% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001079    22
72.31% :     0.000780s :     13: func_graph_cloner_run.FuncGraphClonerGraph
27.69% :     0.000299s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022616    72
 0.15% :     0.000034s :      1: add_recomputation
 0.42% :     0.000095s :      1: auto_monad
 0.06% :     0.000014s :      1: auto_monad_reorder
 0.80% :     0.000182s :      1: bootstrap
 0.10% :     0.000023s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.09% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.21% :     0.000048s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
35.79% :     0.008095s :      1: jit_opt_a
 0.53% :     0.000119s :      1: jit_opt_after_cconv
 0.18% :     0.000041s :      1: jit_opt_b
 1.85% :     0.000418s :      1: loop_unroll
 2.39% :     0.000541s :      1: mutable_eliminate
 6.70% :     0.001516s :     26: opt.transform.jit_opt_a
 0.13% :     0.000029s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000015s :      4: opt.transform.jit_opt_b
 0.04% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000014s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000022s :      4: opt.transform.symbol_engine_opt
 1.96% :     0.000444s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000026s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 4.98% :     0.001127s :      1: renormalize.infer
 1.51% :     0.000341s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000034s :      1: rewriter_after_opt_a
 0.40% :     0.000091s :      1: rewriter_before_opt_a
 0.29% :     0.000065s :      1: symbol_engine_optimizer
40.48% :     0.009154s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.588.828 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d1550> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.589.410 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0174429, [33]
[bootstrap]: 0.0002596
[type_inference]: 0.00404831
[event_method]: 1.704e-05
[auto_monad]: 5.768e-05
[graph_reusing]: 4.89e-06
[pre_auto_parallel]: 2.09999e-06
[py_interpret_to_execute]: 3.314e-05
[rewriter_before_opt_a]: 6.724e-05
[expand_dump_flag]: 3.08998e-06
[jit_opt_a]: 0.00459882, [2]
    [Cycle 1]: 0.00134761, [27]
        [switch_simplify]: 4.528e-05
        [loop_unroll]: 3.115e-05
        [a_1]: 0.00048925
        [with_stream_mark]: 1.239e-05
        [recompute_prepare]: 6.93e-06
        [updatestate_depend_eliminate]: 2.96001e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 2.01998e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 6.00002e-06
        [updatestate_useless_node_eliminater]: 5.00001e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 1.72999e-06
        [get_grad_eliminate_]: 4.97999e-06
        [merge_forward]: 2.56e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.124e-05
        [j_node_and_user_rematch]: 8.62e-06
        [meta_fg_expand]: 2.12001e-06
        [replace_old_param]: 7.98001e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.00048699
        [add_forward_monad_depend]: 4.12e-06
        [auto_monad_grad]: 1.79e-06
        [auto_monad_eliminator]: 9.66e-06
        [cse]: 1.689e-05
        [replace_applicator]: 1.112e-05
    [Cycle 2]: 0.00032893, [27]
        [switch_simplify]: 5.48002e-06
        [loop_unroll]: 4.84003e-06
        [a_1]: 9.261e-05
        [with_stream_mark]: 8.92999e-06
        [recompute_prepare]: 5.13002e-06
        [updatestate_depend_eliminate]: 2.54999e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 5.00999e-06
        [updatestate_useless_node_eliminater]: 4.94998e-06
        [accelerated_algorithm]: 5.52001e-06
        [meta_shard_fg_expand]: 1.55999e-06
        [get_grad_eliminate_]: 4.70001e-06
        [merge_forward]: 2.53e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.246e-05
        [j_node_and_user_rematch]: 8.20999e-06
        [meta_fg_expand]: 1.50999e-06
        [replace_old_param]: 7.84002e-06
        [inline_without_move]: 4.65001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 1.29e-06
        [auto_monad_eliminator]: 5.28002e-06
        [cse]: 1.089e-05
        [replace_applicator]: 5.61e-06
[py_interpret_to_execute_after_opt_a]: 9.39998e-06
[rewriter_after_opt_a]: 2.853e-05
[convert_after_rewriter]: 7.13e-06
[order_py_execute_after_rewriter]: 5.04e-06
[mutable_eliminate]: 0.00054206
[jit_opt_b]: 5.261e-05, [1]
    [Cycle 1]: 4.56e-05, [2]
        [frontend_op_eliminate]: 1.767e-05
        [inline_after_opt_a]: 1.558e-05
[cconv]: 2.247e-05
[loop_unroll]: 0.00043434
[jit_opt_after_cconv]: 0.00014544, [1]
    [Cycle 1]: 0.00013917, [11]
        [c_1]: 2.125e-05
        [parameter_eliminate]: 2.87002e-06
        [updatestate_depend_eliminate]: 7.09001e-06
        [updatestate_assign_eliminate]: 2.61999e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [cse]: 2.09e-05
        [call_graph_tuple_transform]: 1.784e-05
        [tuple_list_get_item_eliminator]: 5.87001e-06
        [none_parameter_eliminate]: 1.32999e-06
        [renormalize]: 5.79981e-07
        [switch_simplify]: 5.22999e-06
[remove_dup_value]: 1.132e-05
[partial_unused_args_eliminate]: 1.70001e-06
[environ_conv]: 5.08002e-06
[add_recomputation]: 4.222e-05
[cse_after_recomputation]: 2.086e-05, [1]
    [Cycle 1]: 1.54e-05, [1]
        [cse]: 9.38002e-06
[auto_monad_reorder]: 1.383e-05
[get_jit_bprop_graph]: 1.96e-06
[rewriter_after_jit_bprop_graph]: 5.42999e-06
[opt_after_jit_grad]: 0.00048203
[symbol_engine_optimizer]: 7.304e-05, [1]
    [Cycle 1]: 6.65e-05, [6]
        [build]: 3.54002e-06
        [elim_shapecalc]: 7.82e-06
        [elim_not_effective]: 1.328e-05
        [opt_reshape]: 5.37001e-06
        [fold_const_symbol]: 8.43999e-06
        [renormalize]: 6.69999e-07
[validate]: 3.404e-05
[backend_pass]: 8.59989e-07
[task_emit]: 0.00621636
[execute]: 7.49002e-06
Sums
    bootstrap                            :     0.000260s :  1.86%
    type_inference                       :     0.004048s : 29.06%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000058s :  0.41%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.24%
    rewriter_before_opt_a                :     0.000067s :  0.48%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000051s :  0.36%
    jit_opt_a.loop_unroll                :     0.000036s :  0.26%
    jit_opt_a.a_1                        :     0.000582s :  4.18%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.07%
    jit_opt_a.renormalize                :     0.000487s :  3.50%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.11%
    jit_opt_a.cse                        :     0.000028s :  0.20%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000029s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000542s :  3.89%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.13%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.11%
    cconv                                :     0.000022s :  0.16%
    loop_unroll                          :     0.000434s :  3.12%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000021s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000011s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000042s :  0.30%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000014s :  0.10%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000482s :  3.46%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000034s :  0.24%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006216s : 44.63%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000161    23
 1.30% :     0.000002s :      2: substitution.elim_not_effective
 0.79% :     0.000001s :      2: substitution.fold_const_symbol
 2.82% :     0.000005s :      3: substitution.graph_param_transform
82.01% :     0.000132s :      4: substitution.inline
 1.68% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.52% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.37% :     0.000004s :      2: substitution.replace_old_param
 6.51% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003994     2
81.16% :     0.003242s :      1: type_inference.infer
18.84% :     0.000752s :      1: type_inference.specialize
------[replace.]   0.000062     6
66.42% :     0.000041s :      4: replace.inline
33.58% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000139     6
93.38% :     0.000129s :      4: match.inline
 6.62% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.21% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.52% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 1.93% :     0.000002s :     10: predicate.arithmetic_simplify
 1.28% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.74% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.45% :     0.000001s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.14% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.35% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.66% :     0.000002s :      6: predicate.less_batch_normalization
 1.73% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.65% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.36% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.09% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.19% :     0.000001s :     10: predicate.minmaximum_grad
 2.05% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.17% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.15% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.23% :     0.000008s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.90% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.34% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.91% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.45% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000498    12
47.98% :     0.000239s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.35% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.67% :     0.000233s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018822    76
 0.24% :     0.000046s :      1: add_recomputation
 0.32% :     0.000061s :      1: auto_monad
 0.09% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.45% :     0.000273s :      1: bootstrap
 0.13% :     0.000025s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.11% :     0.000021s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
24.45% :     0.004602s :      1: jit_opt_a
 0.79% :     0.000148s :      1: jit_opt_after_cconv
 0.29% :     0.000055s :      1: jit_opt_b
 2.35% :     0.000442s :      1: loop_unroll
 2.94% :     0.000554s :      1: mutable_eliminate
 4.15% :     0.000781s :     26: opt.transform.jit_opt_a
 0.25% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.60% :     0.000490s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000036s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000014s :      1: remove_dup_value
 1.24% :     0.000234s :      1: renormalize.infer
 1.31% :     0.000246s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000031s :      1: rewriter_after_opt_a
 0.37% :     0.000070s :      1: rewriter_before_opt_a
 0.40% :     0.000075s :      1: symbol_engine_optimizer
33.08% :     0.006227s :      1: task_emit
21.56% :     0.004059s :      1: type_inference
 0.29% :     0.000054s :      1: validate


TotalTime = 0.0221245, [33]
[bootstrap]: 0.00019515
[type_inference]: 0.0106634
[event_method]: 5.434e-05
[auto_monad]: 9.449e-05
[graph_reusing]: 7.39002e-06
[pre_auto_parallel]: 3.38e-06
[py_interpret_to_execute]: 3.047e-05
[rewriter_before_opt_a]: 9.492e-05
[expand_dump_flag]: 3.72998e-06
[jit_opt_a]: 0.00897404, [2]
    [Cycle 1]: 0.00487843, [27]
        [switch_simplify]: 6.797e-05
        [loop_unroll]: 3.916e-05
        [a_1]: 0.00086803
        [with_stream_mark]: 2.537e-05
        [recompute_prepare]: 1.946e-05
        [updatestate_depend_eliminate]: 7.2e-06
        [updatestate_assign_eliminate]: 5.84e-06
        [updatestate_loads_eliminate]: 5.49e-06
        [parameter_eliminate]: 2.12001e-06
        [specialize_transform]: 1.365e-05
        [updatestate_useless_node_eliminater]: 1.228e-05
        [accelerated_algorithm]: 1.273e-05
        [meta_shard_fg_expand]: 3.67998e-06
        [get_grad_eliminate_]: 1.244e-05
        [merge_forward]: 7.05e-06
        [cell_reuse_recompute_pass]: 7.39994e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.421e-05
        [j_node_and_user_rematch]: 2.291e-05
        [meta_fg_expand]: 0.00159522
        [replace_old_param]: 5.299e-05
        [inline_without_move]: 5.275e-05
        [renormalize]: 0.00175722
        [add_forward_monad_depend]: 4.68001e-06
        [auto_monad_grad]: 1.73002e-06
        [auto_monad_eliminator]: 9.21998e-06
        [cse]: 1.589e-05
        [replace_applicator]: 1.123e-05
    [Cycle 2]: 0.00030142, [27]
        [switch_simplify]: 4.52e-06
        [loop_unroll]: 3.73001e-06
        [a_1]: 4.644e-05
        [with_stream_mark]: 9.19e-06
        [recompute_prepare]: 3.97e-06
        [updatestate_depend_eliminate]: 2.22999e-06
        [updatestate_assign_eliminate]: 1.70001e-06
        [updatestate_loads_eliminate]: 1.57999e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 4.2e-06
        [updatestate_useless_node_eliminater]: 3.71999e-06
        [accelerated_algorithm]: 7.38e-06
        [meta_shard_fg_expand]: 1.55001e-06
        [get_grad_eliminate_]: 3.83001e-06
        [merge_forward]: 2.22999e-06
        [cell_reuse_recompute_pass]: 1.97999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.016e-05
        [j_node_and_user_rematch]: 5.76e-06
        [meta_fg_expand]: 3.784e-05
        [replace_old_param]: 4.55999e-06
        [inline_without_move]: 3.32002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.50001e-06
        [auto_monad_grad]: 9.19972e-07
        [auto_monad_eliminator]: 4.82e-06
        [cse]: 1.047e-05
        [replace_applicator]: 3.58e-06
[py_interpret_to_execute_after_opt_a]: 8.47998e-06
[rewriter_after_opt_a]: 2.732e-05
[convert_after_rewriter]: 6.04999e-06
[order_py_execute_after_rewriter]: 3.9e-06
[mutable_eliminate]: 0.00055987
[jit_opt_b]: 4.181e-05, [1]
    [Cycle 1]: 3.532e-05, [2]
        [frontend_op_eliminate]: 1.178e-05
        [inline_after_opt_a]: 1.19e-05
[cconv]: 1.624e-05
[loop_unroll]: 0.00042807
[jit_opt_after_cconv]: 0.00011669, [1]
    [Cycle 1]: 0.00010998, [11]
        [c_1]: 1.248e-05
        [parameter_eliminate]: 2.64999e-06
        [updatestate_depend_eliminate]: 4.59998e-06
        [updatestate_assign_eliminate]: 1.85001e-06
        [updatestate_loads_eliminate]: 1.72001e-06
        [cse]: 1.57e-05
        [call_graph_tuple_transform]: 1.255e-05
        [tuple_list_get_item_eliminator]: 4.10998e-06
        [none_parameter_eliminate]: 1.07e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 3.97e-06
[remove_dup_value]: 9.76e-06
[partial_unused_args_eliminate]: 1.80001e-06
[environ_conv]: 4.02002e-06
[add_recomputation]: 2.534e-05
[cse_after_recomputation]: 1.976e-05, [1]
    [Cycle 1]: 1.453e-05, [1]
        [cse]: 8.60999e-06
[auto_monad_reorder]: 8.60999e-06
[get_jit_bprop_graph]: 1.70001e-06
[rewriter_after_jit_bprop_graph]: 5.32999e-06
[opt_after_jit_grad]: 0.00044233
[symbol_engine_optimizer]: 6.351e-05, [1]
    [Cycle 1]: 5.708e-05, [6]
        [build]: 2.59001e-06
        [elim_shapecalc]: 5.84999e-06
        [elim_not_effective]: 9.98998e-06
        [opt_reshape]: 4.03001e-06
        [fold_const_symbol]: 5.91998e-06
        [renormalize]: 5.59987e-07
[validate]: 2.291e-05
[backend_pass]: 7.39994e-07
[task_emit]: 1.723e-05
[execute]: 1.32999e-06
Sums
    bootstrap                            :     0.000195s :  1.10%
    type_inference                       :     0.010663s : 60.26%
    event_method                         :     0.000054s :  0.31%
    auto_monad                           :     0.000094s :  0.53%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.17%
    rewriter_before_opt_a                :     0.000095s :  0.54%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000072s :  0.41%
    jit_opt_a.loop_unroll                :     0.000043s :  0.24%
    jit_opt_a.a_1                        :     0.000914s :  5.17%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000023s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000018s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000020s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.09%
    jit_opt_a.merge_forward              :     0.000009s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001633s :  9.23%
    jit_opt_a.replace_old_param          :     0.000058s :  0.33%
    jit_opt_a.inline_without_move        :     0.000056s :  0.32%
    jit_opt_a.renormalize                :     0.001757s :  9.93%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.08%
    jit_opt_a.cse                        :     0.000026s :  0.15%
    jit_opt_a.replace_applicator         :     0.000015s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000027s :  0.15%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000560s :  3.16%
    jit_opt_b.frontend_op_eliminate      :     0.000012s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.07%
    cconv                                :     0.000016s :  0.09%
    loop_unroll                          :     0.000428s :  2.42%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000010s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000025s :  0.14%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000009s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000442s :  2.50%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000023s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000017s :  0.10%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000250    50
 0.63% :     0.000002s :      1: substitution.elim_not_effective
 0.29% :     0.000001s :      1: substitution.fold_const_symbol
 1.03% :     0.000003s :      1: substitution.graph_param_transform
69.12% :     0.000173s :      6: substitution.inline
 6.98% :     0.000017s :      2: substitution.inline_without_move
 2.38% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.22% :     0.000003s :      2: substitution.minmaximum_grad
 1.46% :     0.000004s :      3: substitution.partial_eliminate
 2.68% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.06% :     0.000005s :      2: substitution.replace_applicator
 1.67% :     0.000004s :      5: substitution.replace_old_param
 1.24% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.48% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.67% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 5.08% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010590     2
88.23% :     0.009344s :      1: type_inference.infer
11.77% :     0.001246s :      1: type_inference.specialize
------[replace.]   0.000065     8
78.31% :     0.000051s :      6: replace.inline
21.69% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000176     8
95.56% :     0.000169s :      6: match.inline
 4.44% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000178  1109
 1.82% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.67% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.19% :     0.000002s :     18: predicate.addn_check_dump
 1.40% :     0.000002s :     18: predicate.addn_zero_filter
 2.52% :     0.000004s :     18: predicate.arithmetic_simplify
 1.33% :     0.000002s :     18: predicate.cast_eliminate
 0.25% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.24% :     0.000002s :     18: predicate.compare_switch_simplify
 1.48% :     0.000003s :     18: predicate.depend_value_elim
 1.21% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.43% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.68% :     0.000003s :     18: predicate.dict_set_item_eliminator
 0.41% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      1: predicate.elim_not_effective
 0.30% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.46% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.41% :     0.000003s :     18: predicate.environ_get_add_eliminate
 1.20% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.40% :     0.000003s :     18: predicate.environ_get_eliminate
 1.19% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.18% :     0.000002s :     12: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 3.98% :     0.000007s :     28: predicate.inline
 3.03% :     0.000005s :     33: predicate.inline_without_move
 0.48% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.40% :     0.000002s :     12: predicate.less_batch_normalization
 1.70% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.69% :     0.000003s :     21: predicate.load_eliminater
 0.60% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.62% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.99% :     0.000004s :     19: predicate.make_slice_get_slice_eliminator
 1.63% :     0.000003s :     18: predicate.merge_addn
 1.52% :     0.000003s :     18: predicate.minmaximum_grad
 0.70% :     0.000001s :      1: predicate.mutable_eliminate
 0.20% :     0.000000s :      1: predicate.opt_reshape
 1.91% :     0.000003s :     21: predicate.partial_eliminate
 1.29% :     0.000002s :     18: predicate.print_const_string_wrapper
 2.01% :     0.000004s :     18: predicate.reduce_eliminate
 1.43% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.45% :     0.000003s :     22: predicate.replace_applicator
 1.57% :     0.000003s :     33: predicate.replace_old_param
 0.10% :     0.000000s :      1: predicate.reset_defer_inline
 1.40% :     0.000002s :     18: predicate.reshape_eliminate
 1.53% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.48% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.56% :     0.000003s :     18: predicate.same_eliminate
 0.74% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.30% :     0.000001s :      2: predicate.special_op_eliminate
 1.35% :     0.000002s :     12: predicate.specialize_transform
 1.86% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.13% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 3.28% :     0.000006s :     26: predicate.switch_defer_inline
 2.41% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.51% :     0.000012s :     71: predicate.switch_simplify
 1.35% :     0.000002s :     18: predicate.tile_eliminate
 1.35% :     0.000002s :     18: predicate.transpose_eliminate
 1.92% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.51% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.86% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.94% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.46% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000003s :     18: predicate.value_based_eliminate
 0.11% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.32% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001243    22
70.75% :     0.000879s :     13: func_graph_cloner_run.FuncGraphClonerGraph
29.25% :     0.000364s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025231    76
 0.11% :     0.000028s :      1: add_recomputation
 0.39% :     0.000099s :      1: auto_monad
 0.04% :     0.000011s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.82% :     0.000207s :      1: bootstrap
 0.08% :     0.000019s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.02% :     0.000006s :      1: environ_conv
 0.23% :     0.000059s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.03% :     0.000007s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
35.58% :     0.008977s :      1: jit_opt_a
 0.47% :     0.000120s :      1: jit_opt_after_cconv
 0.18% :     0.000044s :      1: jit_opt_b
 1.73% :     0.000436s :      1: loop_unroll
 2.25% :     0.000568s :      1: mutable_eliminate
 5.10% :     0.001286s :     26: opt.transform.jit_opt_a
 0.12% :     0.000029s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000017s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000022s :      4: opt.transform.symbol_engine_opt
 1.79% :     0.000450s :      1: opt_after_jit_grad
 0.02% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000033s :      1: py_interpret_to_execute
 0.04% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000012s :      1: remove_dup_value
 5.18% :     0.001307s :      1: renormalize.infer
 1.76% :     0.000443s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000030s :      1: rewriter_after_opt_a
 0.39% :     0.000098s :      1: rewriter_before_opt_a
 0.26% :     0.000066s :      1: symbol_engine_optimizer
 0.09% :     0.000022s :      1: task_emit
42.31% :     0.010675s :      1: type_inference
 0.15% :     0.000037s :      1: validate


TotalTime = 0.0310421, [33]
[bootstrap]: 0.00020018
[type_inference]: 0.00930043
[event_method]: 4.156e-05
[auto_monad]: 8.874e-05
[graph_reusing]: 6.01e-06
[pre_auto_parallel]: 2.49999e-06
[py_interpret_to_execute]: 2.438e-05
[rewriter_before_opt_a]: 8.655e-05
[expand_dump_flag]: 2.93e-06
[jit_opt_a]: 0.0118012, [3]
    [Cycle 1]: 0.00589959, [27]
        [switch_simplify]: 6.75e-05
        [loop_unroll]: 3.825e-05
        [a_1]: 0.0008399
        [with_stream_mark]: 2.62e-05
        [recompute_prepare]: 1.923e-05
        [updatestate_depend_eliminate]: 8.07e-06
        [updatestate_assign_eliminate]: 6.76e-06
        [updatestate_loads_eliminate]: 5.89999e-06
        [parameter_eliminate]: 2.40002e-06
        [specialize_transform]: 1.299e-05
        [updatestate_useless_node_eliminater]: 1.146e-05
        [accelerated_algorithm]: 1.209e-05
        [meta_shard_fg_expand]: 3.45e-06
        [get_grad_eliminate_]: 1.162e-05
        [merge_forward]: 8.1e-06
        [cell_reuse_recompute_pass]: 1.29e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.659e-05
        [j_node_and_user_rematch]: 2.265e-05
        [meta_fg_expand]: 0.00148533
        [replace_old_param]: 5.491e-05
        [inline_without_move]: 5.473e-05
        [renormalize]: 0.00266418
        [add_forward_monad_depend]: 1.14e-05
        [auto_monad_grad]: 5.66e-06
        [auto_monad_eliminator]: 5.397e-05
        [cse]: 0.0001552
        [replace_applicator]: 7.323e-05
    [Cycle 2]: 0.00219095, [27]
        [switch_simplify]: 3.829e-05
        [loop_unroll]: 3.563e-05
        [a_1]: 0.00114657
        [with_stream_mark]: 1.835e-05
        [recompute_prepare]: 1.085e-05
        [updatestate_depend_eliminate]: 5.07999e-06
        [updatestate_assign_eliminate]: 4.18001e-06
        [updatestate_loads_eliminate]: 3.73999e-06
        [parameter_eliminate]: 1.96998e-06
        [specialize_transform]: 7.92e-06
        [updatestate_useless_node_eliminater]: 6.84001e-06
        [accelerated_algorithm]: 7.43e-06
        [meta_shard_fg_expand]: 2.37999e-06
        [get_grad_eliminate_]: 6.68003e-06
        [merge_forward]: 4.83001e-06
        [cell_reuse_recompute_pass]: 1.92999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.951e-05
        [j_node_and_user_rematch]: 1.252e-05
        [meta_fg_expand]: 4.668e-05
        [replace_old_param]: 1.132e-05
        [inline_without_move]: 7.08998e-06
        [renormalize]: 0.00057491
        [add_forward_monad_depend]: 4.97e-06
        [auto_monad_grad]: 2.06e-06
        [auto_monad_eliminator]: 1.509e-05
        [cse]: 2.908e-05
        [replace_applicator]: 1.364e-05
    [Cycle 3]: 0.00040517, [27]
        [switch_simplify]: 7.43e-06
        [loop_unroll]: 6.72002e-06
        [a_1]: 0.00014279
        [with_stream_mark]: 1.04e-05
        [recompute_prepare]: 6.98998e-06
        [updatestate_depend_eliminate]: 3.71999e-06
        [updatestate_assign_eliminate]: 3.28998e-06
        [updatestate_loads_eliminate]: 2.99001e-06
        [parameter_eliminate]: 1.25999e-06
        [specialize_transform]: 6.43e-06
        [updatestate_useless_node_eliminater]: 6.29999e-06
        [accelerated_algorithm]: 6.73e-06
        [meta_shard_fg_expand]: 1.66e-06
        [get_grad_eliminate_]: 6.63e-06
        [merge_forward]: 3.51999e-06
        [cell_reuse_recompute_pass]: 1.81998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.547e-05
        [j_node_and_user_rematch]: 1.091e-05
        [meta_fg_expand]: 2.25002e-06
        [replace_old_param]: 8.79e-06
        [inline_without_move]: 6.41e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 9.89996e-07
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 8.09002e-06
        [cse]: 1.582e-05
        [replace_applicator]: 7.16001e-06
[py_interpret_to_execute_after_opt_a]: 1.23e-05
[rewriter_after_opt_a]: 4.404e-05
[convert_after_rewriter]: 7.9e-06
[order_py_execute_after_rewriter]: 6.77002e-06
[mutable_eliminate]: 0.00062185
[jit_opt_b]: 5.967e-05, [1]
    [Cycle 1]: 5.255e-05, [2]
        [frontend_op_eliminate]: 2.137e-05
        [inline_after_opt_a]: 2.014e-05
[cconv]: 2.076e-05
[loop_unroll]: 0.00042432
[jit_opt_after_cconv]: 0.00016435, [1]
    [Cycle 1]: 0.00015753, [11]
        [c_1]: 2.919e-05
        [parameter_eliminate]: 2.27999e-06
        [updatestate_depend_eliminate]: 6.76999e-06
        [updatestate_assign_eliminate]: 3.21999e-06
        [updatestate_loads_eliminate]: 3.18e-06
        [cse]: 2.541e-05
        [call_graph_tuple_transform]: 2.376e-05
        [tuple_list_get_item_eliminator]: 6.92002e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 6.69001e-06
[remove_dup_value]: 1.573e-05
[partial_unused_args_eliminate]: 7.26999e-06
[environ_conv]: 6.81001e-06
[add_recomputation]: 5.203e-05
[cse_after_recomputation]: 2.973e-05, [1]
    [Cycle 1]: 2.406e-05, [1]
        [cse]: 1.812e-05
[auto_monad_reorder]: 2.009e-05
[get_jit_bprop_graph]: 1.35999e-06
[rewriter_after_jit_bprop_graph]: 5.63002e-06
[opt_after_jit_grad]: 0.00045496
[symbol_engine_optimizer]: 8.319e-05, [1]
    [Cycle 1]: 7.704e-05, [6]
        [build]: 3.53e-06
        [elim_shapecalc]: 9.63002e-06
        [elim_not_effective]: 1.589e-05
        [opt_reshape]: 7.94002e-06
        [fold_const_symbol]: 1.257e-05
        [renormalize]: 5.89993e-07
[validate]: 3.744e-05
[backend_pass]: 1.09e-06
[task_emit]: 0.00721584
[execute]: 7.35e-06
Sums
    bootstrap                            :     0.000200s :  0.74%
    type_inference                       :     0.009300s : 34.50%
    event_method                         :     0.000042s :  0.15%
    auto_monad                           :     0.000089s :  0.33%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000024s :  0.09%
    rewriter_before_opt_a                :     0.000087s :  0.32%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000113s :  0.42%
    jit_opt_a.loop_unroll                :     0.000081s :  0.30%
    jit_opt_a.a_1                        :     0.002129s :  7.90%
    jit_opt_a.with_stream_mark           :     0.000055s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000037s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000014s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000013s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.02%
    jit_opt_a.specialize_transform       :     0.000027s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000026s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.09%
    jit_opt_a.merge_forward              :     0.000016s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000062s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000046s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001534s :  5.69%
    jit_opt_a.replace_old_param          :     0.000075s :  0.28%
    jit_opt_a.inline_without_move        :     0.000068s :  0.25%
    jit_opt_a.renormalize                :     0.003239s : 12.02%
    jit_opt_a.add_forward_monad_depend   :     0.000017s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000077s :  0.29%
    jit_opt_a.cse                        :     0.000200s :  0.74%
    jit_opt_a.replace_applicator         :     0.000094s :  0.35%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.05%
    rewriter_after_opt_a                 :     0.000044s :  0.16%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.03%
    mutable_eliminate                    :     0.000622s :  2.31%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.07%
    cconv                                :     0.000021s :  0.08%
    loop_unroll                          :     0.000424s :  1.57%
    jit_opt_after_cconv.c_1              :     0.000029s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000007s :  0.03%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000052s :  0.19%
    cse_after_recomputation.cse          :     0.000018s :  0.07%
    auto_monad_reorder                   :     0.000020s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000455s :  1.69%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000037s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007216s : 26.77%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000533   118
 5.55% :     0.000030s :      1: substitution.arithmetic_simplify
 0.47% :     0.000002s :      4: substitution.elim_not_effective
 0.43% :     0.000002s :      4: substitution.fold_const_symbol
 1.27% :     0.000007s :      5: substitution.graph_param_transform
58.52% :     0.000312s :     12: substitution.inline
 3.69% :     0.000020s :      2: substitution.inline_without_move
 1.79% :     0.000010s :     16: substitution.j_node_and_user_rematch
 2.09% :     0.000011s :      7: substitution.minmaximum_grad
 1.37% :     0.000007s :      5: substitution.partial_eliminate
 2.27% :     0.000012s :     16: substitution.remove_not_recompute_node
 5.31% :     0.000028s :     10: substitution.replace_applicator
 1.64% :     0.000009s :      7: substitution.replace_old_param
 0.65% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.26% :     0.000023s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.86% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 7.84% :     0.000042s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009232     2
88.64% :     0.008183s :      1: type_inference.infer
11.36% :     0.001049s :      1: type_inference.specialize
------[replace.]   0.000149    20
 3.41% :     0.000005s :      1: replace.arithmetic_simplify
62.29% :     0.000093s :     12: replace.inline
34.30% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000355    20
 7.98% :     0.000028s :      1: match.arithmetic_simplify
86.11% :     0.000305s :     12: match.inline
 5.91% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000392  2790
 1.64% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.45% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.41% :     0.000006s :     46: predicate.addn_check_dump
 1.86% :     0.000007s :     46: predicate.addn_zero_filter
 1.94% :     0.000008s :     47: predicate.arithmetic_simplify
 1.64% :     0.000006s :     47: predicate.cast_eliminate
 0.22% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.39% :     0.000005s :     46: predicate.compare_switch_simplify
 1.59% :     0.000006s :     46: predicate.depend_value_elim
 1.42% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.48% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.50% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.32% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.14% :     0.000001s :      5: predicate.elim_not_effective
 0.23% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.46% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.45% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.62% :     0.000006s :     47: predicate.environ_get_eliminate
 1.43% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      5: predicate.fold_const_symbol
 0.90% :     0.000004s :     21: predicate.get_grad_eliminate
 0.16% :     0.000001s :      5: predicate.graph_param_transform
 4.54% :     0.000018s :     76: predicate.inline
 1.84% :     0.000007s :     42: predicate.inline_without_move
 0.38% :     0.000001s :     21: predicate.j_node_and_user_rematch
 0.93% :     0.000004s :     21: predicate.less_batch_normalization
 1.91% :     0.000007s :     54: predicate.list_to_tuple_eliminator_
 2.02% :     0.000008s :     59: predicate.load_eliminater
 0.48% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.05% :     0.000012s :     89: predicate.loop_unroll_before_grad
 1.74% :     0.000007s :     52: predicate.make_slice_get_slice_eliminator
 1.46% :     0.000006s :     46: predicate.merge_addn
 1.52% :     0.000006s :     47: predicate.minmaximum_grad
 0.53% :     0.000002s :      5: predicate.mutable_eliminate
 0.22% :     0.000001s :      5: predicate.opt_reshape
 2.34% :     0.000009s :     59: predicate.partial_eliminate
 1.46% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.82% :     0.000007s :     47: predicate.reduce_eliminate
 1.88% :     0.000007s :     54: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.67% :     0.000010s :    105: predicate.replace_applicator
 0.94% :     0.000004s :     42: predicate.replace_old_param
 0.12% :     0.000000s :      5: predicate.reset_defer_inline
 1.67% :     0.000007s :     47: predicate.reshape_eliminate
 1.51% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.34% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.44% :     0.000006s :     46: predicate.same_eliminate
 0.47% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.46% :     0.000002s :     10: predicate.special_op_eliminate
 0.93% :     0.000004s :     21: predicate.specialize_transform
 1.68% :     0.000007s :     46: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.19% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.69% :     0.000011s :     66: predicate.switch_defer_inline
 2.40% :     0.000009s :     66: predicate.switch_layer_defer_inline
 6.07% :     0.000024s :    160: predicate.switch_simplify
 1.51% :     0.000006s :     47: predicate.tile_eliminate
 1.51% :     0.000006s :     47: predicate.transpose_eliminate
 1.83% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.67% :     0.000007s :     47: predicate.tuple_list_get_item_depend_reorder
 3.25% :     0.000013s :     64: predicate.tuple_list_get_item_eliminator
 1.96% :     0.000008s :     47: predicate.tuple_list_set_item_eliminator
 1.87% :     0.000007s :     54: predicate.tuple_to_list_eliminator_
 1.93% :     0.000008s :     59: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000011s :     80: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000007s :     46: predicate.value_based_eliminate
 0.18% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.24% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001414    29
59.02% :     0.000834s :     13: func_graph_cloner_run.FuncGraphClonerGraph
40.98% :     0.000579s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.037190    91
 0.15% :     0.000055s :      1: add_recomputation
 0.25% :     0.000093s :      1: auto_monad
 0.06% :     0.000023s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.57% :     0.000212s :      1: bootstrap
 0.06% :     0.000023s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000032s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.12% :     0.000046s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000003s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
31.74% :     0.011806s :      1: jit_opt_a
 0.45% :     0.000167s :      1: jit_opt_after_cconv
 0.17% :     0.000062s :      1: jit_opt_b
 1.16% :     0.000432s :      1: loop_unroll
 1.70% :     0.000630s :      1: mutable_eliminate
 7.43% :     0.002761s :     39: opt.transform.jit_opt_a
 0.17% :     0.000064s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000035s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000043s :      4: opt.transform.symbol_engine_opt
 1.24% :     0.000463s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.03% :     0.000009s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000027s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 4.87% :     0.001811s :      2: renormalize.infer
 3.80% :     0.001413s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000047s :      1: rewriter_after_opt_a
 0.24% :     0.000089s :      1: rewriter_before_opt_a
 0.23% :     0.000086s :      1: symbol_engine_optimizer
19.43% :     0.007227s :      1: task_emit
25.04% :     0.009313s :      1: type_inference
 0.16% :     0.000059s :      1: validate


TotalTime = 0.0198985, [30]
[bootstrap]: 0.00020687
[type_inference]: 0.00998713
[event_method]: 4.11e-05
[auto_monad]: 8.929e-05
[graph_reusing]: 6.44001e-06
[pre_auto_parallel]: 2.83e-06
[py_interpret_to_execute]: 2.24e-05
[rewriter_before_opt_a]: 8.515e-05
[expand_dump_flag]: 3.25e-06
[jit_opt_a]: 0.00757197, [2]
    [Cycle 1]: 0.00413323, [27]
        [switch_simplify]: 5.435e-05
        [loop_unroll]: 3.735e-05
        [a_1]: 0.00079672
        [with_stream_mark]: 2.093e-05
        [recompute_prepare]: 1.748e-05
        [updatestate_depend_eliminate]: 7.42002e-06
        [updatestate_assign_eliminate]: 5.77999e-06
        [updatestate_loads_eliminate]: 5.57001e-06
        [parameter_eliminate]: 1.95001e-06
        [specialize_transform]: 1.353e-05
        [updatestate_useless_node_eliminater]: 1.172e-05
        [accelerated_algorithm]: 1.167e-05
        [meta_shard_fg_expand]: 2.39001e-06
        [get_grad_eliminate_]: 1.181e-05
        [merge_forward]: 7.06001e-06
        [cell_reuse_recompute_pass]: 8.70001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.359e-05
        [j_node_and_user_rematch]: 2.173e-05
        [meta_fg_expand]: 0.0013427
        [replace_old_param]: 5.027e-05
        [inline_without_move]: 5e-05
        [renormalize]: 0.00140191
        [add_forward_monad_depend]: 4.85999e-06
        [auto_monad_grad]: 1.74998e-06
        [auto_monad_eliminator]: 1.07e-05
        [cse]: 1.73e-05
        [replace_applicator]: 1.165e-05
    [Cycle 2]: 0.00026306, [27]
        [switch_simplify]: 4.22003e-06
        [loop_unroll]: 3.76999e-06
        [a_1]: 3.67e-05
        [with_stream_mark]: 9.11002e-06
        [recompute_prepare]: 3.73999e-06
        [updatestate_depend_eliminate]: 2.19001e-06
        [updatestate_assign_eliminate]: 1.63002e-06
        [updatestate_loads_eliminate]: 1.62999e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 3.50998e-06
        [updatestate_useless_node_eliminater]: 3.53999e-06
        [accelerated_algorithm]: 3.56999e-06
        [meta_shard_fg_expand]: 1.55999e-06
        [get_grad_eliminate_]: 3.23e-06
        [merge_forward]: 2.40002e-06
        [cell_reuse_recompute_pass]: 1.28002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.61e-06
        [j_node_and_user_rematch]: 5.14998e-06
        [meta_fg_expand]: 2.461e-05
        [replace_old_param]: 3.63e-06
        [inline_without_move]: 3.04999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.42e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 4.05e-06
        [cse]: 1.041e-05
        [replace_applicator]: 4.27998e-06
[py_interpret_to_execute_after_opt_a]: 7.41999e-06
[rewriter_after_opt_a]: 2.848e-05
[convert_after_rewriter]: 6.02999e-06
[order_py_execute_after_rewriter]: 4.22e-06
[mutable_eliminate]: 0.00048325
[jit_opt_b]: 3.775e-05, [1]
    [Cycle 1]: 3.127e-05, [2]
        [frontend_op_eliminate]: 9.87999e-06
        [inline_after_opt_a]: 1.088e-05
[cconv]: 1.776e-05
[loop_unroll]: 0.00041039
[jit_opt_after_cconv]: 0.00011405, [1]
    [Cycle 1]: 0.00010793, [11]
        [c_1]: 1.111e-05
        [parameter_eliminate]: 2.53998e-06
        [updatestate_depend_eliminate]: 4.70001e-06
        [updatestate_assign_eliminate]: 1.92001e-06
        [updatestate_loads_eliminate]: 1.67001e-06
        [cse]: 1.572e-05
        [call_graph_tuple_transform]: 1.228e-05
        [tuple_list_get_item_eliminator]: 3.65e-06
        [none_parameter_eliminate]: 1.49998e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 3.80998e-06
[remove_dup_value]: 1.22e-05
[partial_unused_args_eliminate]: 1.76e-06
[environ_conv]: 4.47e-06
[add_recomputation]: 2.762e-05
[cse_after_recomputation]: 1.837e-05, [1]
    [Cycle 1]: 1.313e-05, [1]
        [cse]: 7.51001e-06
[auto_monad_reorder]: 1.089e-05
[get_jit_bprop_graph]: 1.27999e-06
[rewriter_after_jit_bprop_graph]: 4.77998e-06
[opt_after_jit_grad]: 0.00044937
[symbol_engine_optimizer]: 6.156e-05, [1]
    [Cycle 1]: 5.56e-05, [6]
        [build]: 3.04001e-06
        [elim_shapecalc]: 5.81e-06
        [elim_not_effective]: 9.86e-06
        [opt_reshape]: 3.9e-06
        [fold_const_symbol]: 5.86998e-06
        [renormalize]: 4.80009e-07
[validate]: 2.427e-05
Sums
    bootstrap                            :     0.000207s :  1.28%
    type_inference                       :     0.009987s : 61.84%
    event_method                         :     0.000041s :  0.25%
    auto_monad                           :     0.000089s :  0.55%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000022s :  0.14%
    rewriter_before_opt_a                :     0.000085s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.36%
    jit_opt_a.loop_unroll                :     0.000041s :  0.25%
    jit_opt_a.a_1                        :     0.000833s :  5.16%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000015s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.09%
    jit_opt_a.merge_forward              :     0.000009s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001367s :  8.47%
    jit_opt_a.replace_old_param          :     0.000054s :  0.33%
    jit_opt_a.inline_without_move        :     0.000053s :  0.33%
    jit_opt_a.renormalize                :     0.001402s :  8.68%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.09%
    jit_opt_a.cse                        :     0.000028s :  0.17%
    jit_opt_a.replace_applicator         :     0.000016s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.05%
    rewriter_after_opt_a                 :     0.000028s :  0.18%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000483s :  2.99%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000018s :  0.11%
    loop_unroll                          :     0.000410s :  2.54%
    jit_opt_after_cconv.c_1              :     0.000011s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000012s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000012s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000028s :  0.17%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000011s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000449s :  2.78%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000024s :  0.15%


Time group info:
------[substitution.]   0.000218    50
 0.76% :     0.000002s :      1: substitution.elim_not_effective
 0.31% :     0.000001s :      1: substitution.fold_const_symbol
 1.47% :     0.000003s :      1: substitution.graph_param_transform
64.36% :     0.000140s :      6: substitution.inline
 7.27% :     0.000016s :      2: substitution.inline_without_move
 2.60% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.30% :     0.000003s :      2: substitution.minmaximum_grad
 1.72% :     0.000004s :      3: substitution.partial_eliminate
 2.90% :     0.000006s :      9: substitution.remove_not_recompute_node
 2.07% :     0.000005s :      2: substitution.replace_applicator
 1.91% :     0.000004s :      5: substitution.replace_old_param
 1.07% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 3.34% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 2.23% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 6.70% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009921     2
89.47% :     0.008875s :      1: type_inference.infer
10.53% :     0.001045s :      1: type_inference.specialize
------[replace.]   0.000057     8
78.59% :     0.000045s :      6: replace.inline
21.41% :     0.000012s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000147     8
93.27% :     0.000137s :      6: match.inline
 6.73% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000162  1109
 1.48% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.68% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.29% :     0.000002s :     18: predicate.addn_check_dump
 1.63% :     0.000003s :     18: predicate.addn_zero_filter
 2.05% :     0.000003s :     18: predicate.arithmetic_simplify
 1.41% :     0.000002s :     18: predicate.cast_eliminate
 0.17% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.32% :     0.000002s :     18: predicate.compare_switch_simplify
 1.49% :     0.000002s :     18: predicate.depend_value_elim
 1.32% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.39% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.42% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.66% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.14% :     0.000000s :      1: predicate.elim_not_effective
 0.29% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.48% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.30% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.31% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.40% :     0.000002s :     18: predicate.environ_get_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.24% :     0.000002s :     12: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 4.30% :     0.000007s :     28: predicate.inline
 3.26% :     0.000005s :     33: predicate.inline_without_move
 0.50% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :     12: predicate.less_batch_normalization
 1.61% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.67% :     0.000003s :     21: predicate.load_eliminater
 0.75% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.58% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.75% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.31% :     0.000002s :     18: predicate.merge_addn
 1.34% :     0.000002s :     18: predicate.minmaximum_grad
 0.86% :     0.000001s :      1: predicate.mutable_eliminate
 0.20% :     0.000000s :      1: predicate.opt_reshape
 2.13% :     0.000003s :     21: predicate.partial_eliminate
 1.38% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.94% :     0.000003s :     18: predicate.reduce_eliminate
 1.57% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.73% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.46% :     0.000002s :     22: predicate.replace_applicator
 1.63% :     0.000003s :     33: predicate.replace_old_param
 0.12% :     0.000000s :      1: predicate.reset_defer_inline
 1.56% :     0.000003s :     18: predicate.reshape_eliminate
 1.64% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.43% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.56% :     0.000003s :     18: predicate.same_eliminate
 0.67% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.54% :     0.000001s :      2: predicate.special_op_eliminate
 1.29% :     0.000002s :     12: predicate.specialize_transform
 1.64% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.14% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.67% :     0.000004s :     26: predicate.switch_defer_inline
 2.45% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.77% :     0.000011s :     71: predicate.switch_simplify
 1.59% :     0.000003s :     18: predicate.tile_eliminate
 1.46% :     0.000002s :     18: predicate.transpose_eliminate
 1.56% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.52% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 2.91% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.62% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.06% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000003s :     18: predicate.value_based_eliminate
 0.13% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.33% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001103    22
73.34% :     0.000809s :     13: func_graph_cloner_run.FuncGraphClonerGraph
26.66% :     0.000294s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022499    72
 0.13% :     0.000030s :      1: add_recomputation
 0.41% :     0.000093s :      1: auto_monad
 0.06% :     0.000013s :      1: auto_monad_reorder
 0.99% :     0.000223s :      1: bootstrap
 0.09% :     0.000021s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.09% :     0.000020s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.20% :     0.000045s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000003s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
33.67% :     0.007574s :      1: jit_opt_a
 0.52% :     0.000117s :      1: jit_opt_after_cconv
 0.18% :     0.000040s :      1: jit_opt_b
 1.86% :     0.000418s :      1: loop_unroll
 2.19% :     0.000492s :      1: mutable_eliminate
 5.22% :     0.001174s :     26: opt.transform.jit_opt_a
 0.12% :     0.000028s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000015s :      4: opt.transform.jit_opt_b
 0.04% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000010s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000022s :      4: opt.transform.symbol_engine_opt
 2.03% :     0.000458s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000025s :      1: py_interpret_to_execute
 0.04% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000015s :      1: remove_dup_value
 4.77% :     0.001073s :      1: renormalize.infer
 1.43% :     0.000321s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000032s :      1: rewriter_after_opt_a
 0.39% :     0.000088s :      1: rewriter_before_opt_a
 0.28% :     0.000064s :      1: symbol_engine_optimizer
44.44% :     0.009999s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.610.255 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c3684c0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.610.859 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.014556, [30]
[bootstrap]: 0.00036518
[type_inference]: 0.00497966
[event_method]: 2.102e-05
[auto_monad]: 7.572e-05
[graph_reusing]: 6.26998e-06
[pre_auto_parallel]: 3.16999e-06
[py_interpret_to_execute]: 3.898e-05
[rewriter_before_opt_a]: 8.071e-05
[expand_dump_flag]: 3.31999e-06
[jit_opt_a]: 0.00619924, [2]
    [Cycle 1]: 0.00179505, [27]
        [switch_simplify]: 6.295e-05
        [loop_unroll]: 2.911e-05
        [a_1]: 0.00055785
        [with_stream_mark]: 2.186e-05
        [recompute_prepare]: 8.93002e-06
        [updatestate_depend_eliminate]: 4.93001e-06
        [updatestate_assign_eliminate]: 2.86999e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [parameter_eliminate]: 2.22001e-06
        [specialize_transform]: 6.31e-06
        [updatestate_useless_node_eliminater]: 5.01002e-06
        [accelerated_algorithm]: 5.65001e-06
        [meta_shard_fg_expand]: 2.44999e-06
        [get_grad_eliminate_]: 6.16e-06
        [merge_forward]: 4.38999e-06
        [cell_reuse_recompute_pass]: 1.66998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.649e-05
        [j_node_and_user_rematch]: 1.001e-05
        [meta_fg_expand]: 3.45e-06
        [replace_old_param]: 9.49e-06
        [inline_without_move]: 5.12e-06
        [renormalize]: 0.00072114
        [add_forward_monad_depend]: 9.14e-06
        [auto_monad_grad]: 2.26e-06
        [auto_monad_eliminator]: 1.998e-05
        [cse]: 3.564e-05
        [replace_applicator]: 1.598e-05
    [Cycle 2]: 0.00037268, [27]
        [switch_simplify]: 5.62999e-06
        [loop_unroll]: 4.72e-06
        [a_1]: 9.149e-05
        [with_stream_mark]: 1.43e-05
        [recompute_prepare]: 5.75001e-06
        [updatestate_depend_eliminate]: 2.92002e-06
        [updatestate_assign_eliminate]: 2.48e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [parameter_eliminate]: 1.66e-06
        [specialize_transform]: 5.52999e-06
        [updatestate_useless_node_eliminater]: 4.60001e-06
        [accelerated_algorithm]: 4.95001e-06
        [meta_shard_fg_expand]: 1.74998e-06
        [get_grad_eliminate_]: 4.58001e-06
        [merge_forward]: 3.39001e-06
        [cell_reuse_recompute_pass]: 2.39001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.756e-05
        [j_node_and_user_rematch]: 8.90999e-06
        [meta_fg_expand]: 2.93e-06
        [replace_old_param]: 8.87e-06
        [inline_without_move]: 5.46e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 3.00002e-06
        [auto_monad_grad]: 1.39998e-06
        [auto_monad_eliminator]: 1.04e-05
        [cse]: 1.462e-05
        [replace_applicator]: 7.95e-06
[py_interpret_to_execute_after_opt_a]: 1.506e-05
[rewriter_after_opt_a]: 4.289e-05
[convert_after_rewriter]: 7.68999e-06
[order_py_execute_after_rewriter]: 5.35999e-06
[mutable_eliminate]: 0.00078447
[jit_opt_b]: 9.772e-05, [1]
    [Cycle 1]: 8.911e-05, [2]
        [frontend_op_eliminate]: 1.835e-05
        [inline_after_opt_a]: 1.771e-05
[cconv]: 3.141e-05
[loop_unroll]: 0.00054892
[jit_opt_after_cconv]: 0.0001725, [1]
    [Cycle 1]: 0.00016401, [11]
        [c_1]: 2.216e-05
        [parameter_eliminate]: 4.48999e-06
        [updatestate_depend_eliminate]: 8.92e-06
        [updatestate_assign_eliminate]: 2.83998e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [cse]: 2.947e-05
        [call_graph_tuple_transform]: 2.295e-05
        [tuple_list_get_item_eliminator]: 5.87999e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 7.60017e-07
        [switch_simplify]: 5.57999e-06
[remove_dup_value]: 1.645e-05
[partial_unused_args_eliminate]: 2.37001e-06
[environ_conv]: 5.63002e-06
[add_recomputation]: 5.91e-05
[cse_after_recomputation]: 2.793e-05, [1]
    [Cycle 1]: 1.967e-05, [1]
        [cse]: 9.69e-06
[auto_monad_reorder]: 1.777e-05
[get_jit_bprop_graph]: 2.54999e-06
[rewriter_after_jit_bprop_graph]: 9.14998e-06
[opt_after_jit_grad]: 0.00060024
[symbol_engine_optimizer]: 8.177e-05, [1]
    [Cycle 1]: 7.371e-05, [6]
        [build]: 4.25999e-06
        [elim_shapecalc]: 7.64002e-06
        [elim_not_effective]: 1.623e-05
        [opt_reshape]: 5.86998e-06
        [fold_const_symbol]: 8.92e-06
        [renormalize]: 6.59988e-07
[validate]: 4.061e-05
Sums
    bootstrap                            :     0.000365s :  3.74%
    type_inference                       :     0.004980s : 50.95%
    event_method                         :     0.000021s :  0.22%
    auto_monad                           :     0.000076s :  0.77%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000039s :  0.40%
    rewriter_before_opt_a                :     0.000081s :  0.83%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000069s :  0.70%
    jit_opt_a.loop_unroll                :     0.000034s :  0.35%
    jit_opt_a.a_1                        :     0.000649s :  6.64%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.37%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.07%
    jit_opt_a.replace_old_param          :     0.000018s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000721s :  7.38%
    jit_opt_a.add_forward_monad_depend   :     0.000012s :  0.12%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000030s :  0.31%
    jit_opt_a.cse                        :     0.000050s :  0.51%
    jit_opt_a.replace_applicator         :     0.000024s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.15%
    rewriter_after_opt_a                 :     0.000043s :  0.44%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000784s :  8.03%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.18%
    cconv                                :     0.000031s :  0.32%
    loop_unroll                          :     0.000549s :  5.62%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.09%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000029s :  0.30%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000016s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000059s :  0.60%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000018s :  0.18%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.09%
    opt_after_jit_grad                   :     0.000600s :  6.14%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.42%


Time group info:
------[substitution.]   0.000226    23
 1.02% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 3.07% :     0.000007s :      3: substitution.graph_param_transform
82.30% :     0.000186s :      4: substitution.inline
 1.91% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.63% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.18% :     0.000005s :      2: substitution.replace_old_param
 6.30% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004894     2
80.58% :     0.003944s :      1: type_inference.infer
19.42% :     0.000950s :      1: type_inference.specialize
------[replace.]   0.000072     6
64.52% :     0.000046s :      4: replace.inline
35.48% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000196     6
93.31% :     0.000183s :      4: match.inline
 6.69% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000127   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.60% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.09% :     0.000001s :     10: predicate.addn_zero_filter
 2.71% :     0.000003s :     10: predicate.arithmetic_simplify
 1.79% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.46% :     0.000002s :     10: predicate.depend_value_elim
 0.96% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.01% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.28% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.50% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.48% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.01% :     0.000001s :     10: predicate.environ_get_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.40% :     0.000001s :      3: predicate.graph_param_transform
 5.29% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.47% :     0.000002s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.54% :     0.000002s :     15: predicate.load_eliminater
 2.09% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.60% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.94% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 2.05% :     0.000003s :      3: predicate.mutable_eliminate
 0.39% :     0.000000s :      3: predicate.opt_reshape
 2.45% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     10: predicate.reduce_eliminate
 1.27% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.76% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.80% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.46% :     0.000001s :      3: predicate.reset_defer_inline
 1.33% :     0.000002s :     10: predicate.reshape_eliminate
 1.02% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000002s :     10: predicate.same_eliminate
 0.65% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.25% :     0.000002s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.54% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.56% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.46% :     0.000003s :     16: predicate.switch_defer_inline
 1.86% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.40% :     0.000011s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.01% :     0.000001s :     10: predicate.transpose_eliminate
 1.19% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.02% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.29% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.70% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.46% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000614    12
51.09% :     0.000314s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.12% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.78% :     0.000263s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016226    72
 0.39% :     0.000063s :      1: add_recomputation
 0.49% :     0.000080s :      1: auto_monad
 0.13% :     0.000021s :      1: auto_monad_reorder
 2.39% :     0.000387s :      1: bootstrap
 0.21% :     0.000034s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.19% :     0.000030s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000026s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
38.23% :     0.006203s :      1: jit_opt_a
 1.08% :     0.000175s :      1: jit_opt_after_cconv
 0.62% :     0.000101s :      1: jit_opt_b
 3.46% :     0.000561s :      1: loop_unroll
 4.92% :     0.000799s :      1: mutable_eliminate
 5.44% :     0.000882s :     26: opt.transform.jit_opt_a
 0.32% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000028s :      4: opt.transform.jit_opt_b
 0.12% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.13% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.76% :     0.000611s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000042s :      1: py_interpret_to_execute
 0.11% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000019s :      1: remove_dup_value
 2.35% :     0.000382s :      1: renormalize.infer
 2.03% :     0.000329s :      1: renormalize.specialize
 0.07% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.29% :     0.000047s :      1: rewriter_after_opt_a
 0.52% :     0.000084s :      1: rewriter_before_opt_a
 0.52% :     0.000084s :      1: symbol_engine_optimizer
30.81% :     0.005000s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.614.631 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d19d0> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.615.911 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80cac10> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.615.993 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.616.420 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0168147, [33]
[bootstrap]: 0.00026629
[type_inference]: 0.0039064
[event_method]: 1.71e-05
[auto_monad]: 5.457e-05
[graph_reusing]: 4.88001e-06
[pre_auto_parallel]: 2.33002e-06
[py_interpret_to_execute]: 2.909e-05
[rewriter_before_opt_a]: 6.762e-05
[expand_dump_flag]: 2.93998e-06
[jit_opt_a]: 0.00456512, [2]
    [Cycle 1]: 0.00133585, [27]
        [switch_simplify]: 5.32e-05
        [loop_unroll]: 3.056e-05
        [a_1]: 0.00050386
        [with_stream_mark]: 1.125e-05
        [recompute_prepare]: 7.16001e-06
        [updatestate_depend_eliminate]: 3.01999e-06
        [updatestate_assign_eliminate]: 2.10002e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.54998e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 5.07e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 1.31002e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 2.56e-06
        [cell_reuse_recompute_pass]: 1.08001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.091e-05
        [j_node_and_user_rematch]: 8.55999e-06
        [meta_fg_expand]: 1.92001e-06
        [replace_old_param]: 7.73999e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 0.00046241
        [add_forward_monad_depend]: 3.49001e-06
        [auto_monad_grad]: 1.49e-06
        [auto_monad_eliminator]: 8.45999e-06
        [cse]: 1.515e-05
        [replace_applicator]: 1.109e-05
    [Cycle 2]: 0.00030935, [27]
        [switch_simplify]: 5.66e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 8.823e-05
        [with_stream_mark]: 7.98001e-06
        [recompute_prepare]: 4.87998e-06
        [updatestate_depend_eliminate]: 2.37001e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 4.94e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 4.60999e-06
        [meta_shard_fg_expand]: 1.18001e-06
        [get_grad_eliminate_]: 4.43999e-06
        [merge_forward]: 2.37999e-06
        [cell_reuse_recompute_pass]: 1.82001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.134e-05
        [j_node_and_user_rematch]: 7.97e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 6.98e-06
        [inline_without_move]: 4.97e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.90001e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 4.73001e-06
        [cse]: 9.64e-06
        [replace_applicator]: 5.01002e-06
[py_interpret_to_execute_after_opt_a]: 7.98001e-06
[rewriter_after_opt_a]: 2.486e-05
[convert_after_rewriter]: 5.44e-06
[order_py_execute_after_rewriter]: 4.27e-06
[mutable_eliminate]: 0.00046989
[jit_opt_b]: 5.597e-05, [1]
    [Cycle 1]: 4.204e-05, [2]
        [frontend_op_eliminate]: 1.597e-05
        [inline_after_opt_a]: 1.534e-05
[cconv]: 1.705e-05
[loop_unroll]: 0.00041274
[jit_opt_after_cconv]: 0.00013334, [1]
    [Cycle 1]: 0.00012744, [11]
        [c_1]: 2.096e-05
        [parameter_eliminate]: 1.83002e-06
        [updatestate_depend_eliminate]: 4.42998e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.08002e-06
        [cse]: 1.687e-05
        [call_graph_tuple_transform]: 1.716e-05
        [tuple_list_get_item_eliminator]: 5.38002e-06
        [none_parameter_eliminate]: 1.14998e-06
        [renormalize]: 4.60015e-07
        [switch_simplify]: 5.36002e-06
[remove_dup_value]: 8.33001e-06
[partial_unused_args_eliminate]: 1.84998e-06
[environ_conv]: 4.03999e-06
[add_recomputation]: 3.607e-05
[cse_after_recomputation]: 1.995e-05, [1]
    [Cycle 1]: 1.456e-05, [1]
        [cse]: 8.97e-06
[auto_monad_reorder]: 1.973e-05
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 3.97e-06
[opt_after_jit_grad]: 0.00043827
[symbol_engine_optimizer]: 8.126e-05, [1]
    [Cycle 1]: 7.5e-05, [6]
        [build]: 2.93998e-06
        [elim_shapecalc]: 7.76001e-06
        [elim_not_effective]: 1.197e-05
        [opt_reshape]: 5.82001e-06
        [fold_const_symbol]: 1.727e-05
        [renormalize]: 4.30009e-07
[validate]: 2.832e-05
[backend_pass]: 6.80011e-07
[task_emit]: 0.00594646
[execute]: 7.26001e-06
Sums
    bootstrap                            :     0.000266s :  2.00%
    type_inference                       :     0.003906s : 29.30%
    event_method                         :     0.000017s :  0.13%
    auto_monad                           :     0.000055s :  0.41%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.22%
    rewriter_before_opt_a                :     0.000068s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.44%
    jit_opt_a.loop_unroll                :     0.000035s :  0.27%
    jit_opt_a.a_1                        :     0.000592s :  4.44%
    jit_opt_a.with_stream_mark           :     0.000019s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.07%
    jit_opt_a.meta_shard_fg_expand       :     0.000002s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000005s :  0.04%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000022s :  0.17%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.11%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000462s :  3.47%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.10%
    jit_opt_a.cse                        :     0.000025s :  0.19%
    jit_opt_a.replace_applicator         :     0.000016s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000025s :  0.19%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000470s :  3.52%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.12%
    cconv                                :     0.000017s :  0.13%
    loop_unroll                          :     0.000413s :  3.10%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000008s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000036s :  0.27%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000020s :  0.15%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000438s :  3.29%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000017s :  0.13%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.21%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005946s : 44.61%
    execute                              :     0.000007s :  0.05%


Time group info:
------[substitution.]   0.000156    23
 0.90% :     0.000001s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.67% :     0.000004s :      3: substitution.graph_param_transform
82.92% :     0.000129s :      4: substitution.inline
 1.79% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.34% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.85% :     0.000003s :      2: substitution.replace_old_param
 6.83% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003856     2
79.89% :     0.003081s :      1: type_inference.infer
20.11% :     0.000776s :      1: type_inference.specialize
------[replace.]   0.000061     6
65.67% :     0.000040s :      4: replace.inline
34.33% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000137     6
92.97% :     0.000127s :      4: match.inline
 7.03% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   691
 1.01% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.13% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.85% :     0.000001s :     10: predicate.addn_check_dump
 1.00% :     0.000001s :     10: predicate.addn_zero_filter
 1.55% :     0.000002s :     10: predicate.arithmetic_simplify
 1.09% :     0.000002s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.85% :     0.000001s :     10: predicate.compare_switch_simplify
 1.06% :     0.000001s :     10: predicate.depend_value_elim
 0.90% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.04% :     0.000001s :     10: predicate.dict_get_item_eliminator
 0.93% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.72% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.43% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.02% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.87% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.98% :     0.000001s :     10: predicate.environ_get_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.80% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 4.29% :     0.000006s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000001s :      6: predicate.j_node_and_user_rematch
 0.90% :     0.000001s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
20.99% :     0.000029s :     15: predicate.load_eliminater
 1.21% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.59% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.59% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.90% :     0.000001s :     10: predicate.merge_addn
 0.85% :     0.000001s :     10: predicate.minmaximum_grad
 1.31% :     0.000002s :      3: predicate.mutable_eliminate
 0.38% :     0.000001s :      3: predicate.opt_reshape
 1.78% :     0.000002s :     15: predicate.partial_eliminate
 0.93% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.32% :     0.000002s :     10: predicate.reduce_eliminate
 1.24% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.44% :     0.000002s :     18: predicate.replace_applicator
 0.43% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.11% :     0.000002s :     10: predicate.reshape_eliminate
 0.95% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.64% :     0.000001s :      3: predicate.row_tensor_eliminate
 0.98% :     0.000001s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.71% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 0.98% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 1.89% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.30% :     0.000009s :     53: predicate.switch_simplify
 0.94% :     0.000001s :     10: predicate.tile_eliminate
 0.95% :     0.000001s :     10: predicate.transpose_eliminate
 1.21% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.06% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 2.76% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.26% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.24% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.38% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.43% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.29% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.52% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000488    12
46.72% :     0.000228s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.00% :     0.000024s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.28% :     0.000235s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018192    76
 0.21% :     0.000039s :      1: add_recomputation
 0.32% :     0.000058s :      1: auto_monad
 0.13% :     0.000023s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.53% :     0.000278s :      1: bootstrap
 0.11% :     0.000020s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
25.11% :     0.004568s :      1: jit_opt_a
 0.75% :     0.000136s :      1: jit_opt_after_cconv
 0.32% :     0.000059s :      1: jit_opt_b
 2.31% :     0.000420s :      1: loop_unroll
 2.63% :     0.000478s :      1: mutable_eliminate
 4.37% :     0.000796s :     26: opt.transform.jit_opt_a
 0.25% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000040s :      4: opt.transform.symbol_engine_opt
 2.45% :     0.000445s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000032s :      1: py_interpret_to_execute
 0.06% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000011s :      1: remove_dup_value
 1.25% :     0.000227s :      1: renormalize.infer
 1.27% :     0.000230s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000028s :      1: rewriter_after_opt_a
 0.39% :     0.000070s :      1: rewriter_before_opt_a
 0.46% :     0.000084s :      1: symbol_engine_optimizer
32.75% :     0.005957s :      1: task_emit
21.53% :     0.003917s :      1: type_inference
 0.24% :     0.000044s :      1: validate


TotalTime = 0.0215714, [33]
[bootstrap]: 0.00020871
[type_inference]: 0.0105356
[event_method]: 4.969e-05
[auto_monad]: 9.273e-05
[graph_reusing]: 6.59999e-06
[pre_auto_parallel]: 3.04001e-06
[py_interpret_to_execute]: 2.922e-05
[rewriter_before_opt_a]: 8.967e-05
[expand_dump_flag]: 3.30998e-06
[jit_opt_a]: 0.00855729, [2]
    [Cycle 1]: 0.00476265, [27]
        [switch_simplify]: 6.194e-05
        [loop_unroll]: 3.783e-05
        [a_1]: 0.00082736
        [with_stream_mark]: 2.413e-05
        [recompute_prepare]: 1.898e-05
        [updatestate_depend_eliminate]: 7.36001e-06
        [updatestate_assign_eliminate]: 6.43e-06
        [updatestate_loads_eliminate]: 5.41002e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 1.278e-05
        [updatestate_useless_node_eliminater]: 1.179e-05
        [accelerated_algorithm]: 1.348e-05
        [meta_shard_fg_expand]: 3.60998e-06
        [get_grad_eliminate_]: 1.219e-05
        [merge_forward]: 7.56999e-06
        [cell_reuse_recompute_pass]: 8.39995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.525e-05
        [j_node_and_user_rematch]: 2.313e-05
        [meta_fg_expand]: 0.00150997
        [replace_old_param]: 5.423e-05
        [inline_without_move]: 5.277e-05
        [renormalize]: 0.00179907
        [add_forward_monad_depend]: 5.17e-06
        [auto_monad_grad]: 2.19999e-06
        [auto_monad_eliminator]: 1.096e-05
        [cse]: 1.663e-05
        [replace_applicator]: 1.079e-05
    [Cycle 2]: 0.00028716, [27]
        [switch_simplify]: 4.62e-06
        [loop_unroll]: 3.71999e-06
        [a_1]: 4.019e-05
        [with_stream_mark]: 8.43001e-06
        [recompute_prepare]: 3.45e-06
        [updatestate_depend_eliminate]: 2.12001e-06
        [updatestate_assign_eliminate]: 1.60999e-06
        [updatestate_loads_eliminate]: 1.40999e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 3.78001e-06
        [updatestate_useless_node_eliminater]: 3.81001e-06
        [accelerated_algorithm]: 4.02e-06
        [meta_shard_fg_expand]: 1.02e-06
        [get_grad_eliminate_]: 3.25e-06
        [merge_forward]: 1.97999e-06
        [cell_reuse_recompute_pass]: 1.94e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.69999e-06
        [j_node_and_user_rematch]: 6.04999e-06
        [meta_fg_expand]: 3.803e-05
        [replace_old_param]: 3.98999e-06
        [inline_without_move]: 3.23998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.53002e-06
        [auto_monad_grad]: 1.07e-06
        [auto_monad_eliminator]: 4.52e-06
        [cse]: 1.125e-05
        [replace_applicator]: 4.1e-06
[py_interpret_to_execute_after_opt_a]: 1.003e-05
[rewriter_after_opt_a]: 3.21e-05
[convert_after_rewriter]: 5.96998e-06
[order_py_execute_after_rewriter]: 4.38999e-06
[mutable_eliminate]: 0.00052295
[jit_opt_b]: 3.988e-05, [1]
    [Cycle 1]: 3.306e-05, [2]
        [frontend_op_eliminate]: 1.062e-05
        [inline_after_opt_a]: 1.085e-05
[cconv]: 1.923e-05
[loop_unroll]: 0.00042458
[jit_opt_after_cconv]: 0.00011841, [1]
    [Cycle 1]: 0.00011166, [11]
        [c_1]: 1.212e-05
        [parameter_eliminate]: 2.71e-06
        [updatestate_depend_eliminate]: 4.79e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.62999e-06
        [cse]: 1.502e-05
        [call_graph_tuple_transform]: 1.282e-05
        [tuple_list_get_item_eliminator]: 3.76001e-06
        [none_parameter_eliminate]: 1.28002e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 4.07003e-06
[remove_dup_value]: 1.254e-05
[partial_unused_args_eliminate]: 2.17001e-06
[environ_conv]: 4.62e-06
[add_recomputation]: 2.974e-05
[cse_after_recomputation]: 1.876e-05, [1]
    [Cycle 1]: 1.337e-05, [1]
        [cse]: 7.76001e-06
[auto_monad_reorder]: 1.198e-05
[get_jit_bprop_graph]: 1.65001e-06
[rewriter_after_jit_bprop_graph]: 5.15001e-06
[opt_after_jit_grad]: 0.00044762
[symbol_engine_optimizer]: 6.42e-05, [1]
    [Cycle 1]: 5.809e-05, [6]
        [build]: 3.04999e-06
        [elim_shapecalc]: 6.16e-06
        [elim_not_effective]: 1.014e-05
        [opt_reshape]: 4.22e-06
        [fold_const_symbol]: 6.32001e-06
        [renormalize]: 5.69999e-07
[validate]: 2.569e-05
[backend_pass]: 7.7e-07
[task_emit]: 1.959e-05
[execute]: 1.00999e-06
Sums
    bootstrap                            :     0.000209s :  1.20%
    type_inference                       :     0.010536s : 60.36%
    event_method                         :     0.000050s :  0.28%
    auto_monad                           :     0.000093s :  0.53%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.17%
    rewriter_before_opt_a                :     0.000090s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.38%
    jit_opt_a.loop_unroll                :     0.000042s :  0.24%
    jit_opt_a.a_1                        :     0.000868s :  4.97%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.09%
    jit_opt_a.merge_forward              :     0.000010s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001548s :  8.87%
    jit_opt_a.replace_old_param          :     0.000058s :  0.33%
    jit_opt_a.inline_without_move        :     0.000056s :  0.32%
    jit_opt_a.renormalize                :     0.001799s : 10.31%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.09%
    jit_opt_a.cse                        :     0.000028s :  0.16%
    jit_opt_a.replace_applicator         :     0.000015s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000032s :  0.18%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000523s :  3.00%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.06%
    cconv                                :     0.000019s :  0.11%
    loop_unroll                          :     0.000425s :  2.43%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000013s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000030s :  0.17%
    cse_after_recomputation.cse          :     0.000008s :  0.04%
    auto_monad_reorder                   :     0.000012s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000448s :  2.56%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000026s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000020s :  0.11%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000249    50
 0.68% :     0.000002s :      1: substitution.elim_not_effective
 0.39% :     0.000001s :      1: substitution.fold_const_symbol
 1.39% :     0.000003s :      1: substitution.graph_param_transform
65.48% :     0.000163s :      6: substitution.inline
 6.85% :     0.000017s :      2: substitution.inline_without_move
 2.68% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.46% :     0.000004s :      2: substitution.minmaximum_grad
 1.81% :     0.000005s :      3: substitution.partial_eliminate
 2.70% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.42% :     0.000006s :      2: substitution.replace_applicator
 1.78% :     0.000004s :      5: substitution.replace_old_param
 1.20% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.58% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 2.07% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 6.51% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010463     2
87.88% :     0.009195s :      1: type_inference.infer
12.12% :     0.001268s :      1: type_inference.specialize
------[replace.]   0.000060     8
78.14% :     0.000047s :      6: replace.inline
21.86% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000171     8
93.44% :     0.000160s :      6: match.inline
 6.56% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000166  1109
 1.43% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.70% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.27% :     0.000002s :     18: predicate.addn_check_dump
 1.60% :     0.000003s :     18: predicate.addn_zero_filter
 2.12% :     0.000004s :     18: predicate.arithmetic_simplify
 1.48% :     0.000002s :     18: predicate.cast_eliminate
 0.24% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.30% :     0.000002s :     18: predicate.compare_switch_simplify
 1.52% :     0.000003s :     18: predicate.depend_value_elim
 1.32% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.55% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.14% :     0.000000s :      1: predicate.elim_not_effective
 0.31% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.40% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.28% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     18: predicate.environ_get_eliminate
 1.28% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.18% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.27% :     0.000007s :     28: predicate.inline
 3.24% :     0.000005s :     33: predicate.inline_without_move
 0.55% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.65% :     0.000003s :     12: predicate.less_batch_normalization
 1.58% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.71% :     0.000003s :     21: predicate.load_eliminater
 0.80% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.64% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.77% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.30% :     0.000002s :     18: predicate.merge_addn
 1.31% :     0.000002s :     18: predicate.minmaximum_grad
 0.72% :     0.000001s :      1: predicate.mutable_eliminate
 0.22% :     0.000000s :      1: predicate.opt_reshape
 2.03% :     0.000003s :     21: predicate.partial_eliminate
 1.37% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.95% :     0.000003s :     18: predicate.reduce_eliminate
 1.57% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.51% :     0.000002s :     22: predicate.replace_applicator
 1.67% :     0.000003s :     33: predicate.replace_old_param
 0.13% :     0.000000s :      1: predicate.reset_defer_inline
 1.52% :     0.000003s :     18: predicate.reshape_eliminate
 1.47% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.41% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.47% :     0.000002s :     18: predicate.same_eliminate
 0.65% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.40% :     0.000001s :      2: predicate.special_op_eliminate
 1.23% :     0.000002s :     12: predicate.specialize_transform
 1.81% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.53% :     0.000004s :     26: predicate.switch_defer_inline
 2.33% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.87% :     0.000011s :     71: predicate.switch_simplify
 1.50% :     0.000002s :     18: predicate.tile_eliminate
 1.36% :     0.000002s :     18: predicate.transpose_eliminate
 1.97% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.50% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 2.62% :     0.000004s :     22: predicate.tuple_list_get_item_eliminator
 1.97% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.63% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000003s :     18: predicate.value_based_eliminate
 0.13% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.29% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001227    22
71.03% :     0.000871s :     13: func_graph_cloner_run.FuncGraphClonerGraph
28.97% :     0.000355s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024660    76
 0.13% :     0.000032s :      1: add_recomputation
 0.39% :     0.000097s :      1: auto_monad
 0.06% :     0.000015s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.90% :     0.000221s :      1: bootstrap
 0.09% :     0.000022s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.08% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.22% :     0.000054s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
34.71% :     0.008560s :      1: jit_opt_a
 0.49% :     0.000121s :      1: jit_opt_after_cconv
 0.17% :     0.000043s :      1: jit_opt_b
 1.76% :     0.000433s :      1: loop_unroll
 2.15% :     0.000531s :      1: mutable_eliminate
 4.98% :     0.001229s :     26: opt.transform.jit_opt_a
 0.12% :     0.000029s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000023s :      4: opt.transform.symbol_engine_opt
 1.85% :     0.000455s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000032s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000015s :      1: remove_dup_value
 5.53% :     0.001363s :      1: renormalize.infer
 1.73% :     0.000426s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000035s :      1: rewriter_after_opt_a
 0.37% :     0.000092s :      1: rewriter_before_opt_a
 0.27% :     0.000067s :      1: symbol_engine_optimizer
 0.09% :     0.000023s :      1: task_emit
42.77% :     0.010548s :      1: type_inference
 0.17% :     0.000042s :      1: validate


TotalTime = 0.0192165, [30]
[bootstrap]: 0.00018223
[type_inference]: 0.00910832
[event_method]: 4.169e-05
[auto_monad]: 7.902e-05
[graph_reusing]: 5.54998e-06
[pre_auto_parallel]: 2.32001e-06
[py_interpret_to_execute]: 2.245e-05
[rewriter_before_opt_a]: 8.249e-05
[expand_dump_flag]: 3.04999e-06
[jit_opt_a]: 0.00782336, [2]
    [Cycle 1]: 0.00447436, [27]
        [switch_simplify]: 5.427e-05
        [loop_unroll]: 3.801e-05
        [a_1]: 0.00087896
        [with_stream_mark]: 2.043e-05
        [recompute_prepare]: 1.688e-05
        [updatestate_depend_eliminate]: 7.33e-06
        [updatestate_assign_eliminate]: 6.23998e-06
        [updatestate_loads_eliminate]: 5.28002e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 1.23e-05
        [updatestate_useless_node_eliminater]: 1.179e-05
        [accelerated_algorithm]: 1.188e-05
        [meta_shard_fg_expand]: 2.72001e-06
        [get_grad_eliminate_]: 1.179e-05
        [merge_forward]: 6.88e-06
        [cell_reuse_recompute_pass]: 7.39994e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.362e-05
        [j_node_and_user_rematch]: 4.727e-05
        [meta_fg_expand]: 0.0015754
        [replace_old_param]: 4.552e-05
        [inline_without_move]: 4.831e-05
        [renormalize]: 0.00141492
        [add_forward_monad_depend]: 3.99002e-06
        [auto_monad_grad]: 1.82999e-06
        [auto_monad_eliminator]: 8.84e-06
        [cse]: 1.664e-05
        [replace_applicator]: 9.91998e-06
    [Cycle 2]: 0.00028313, [27]
        [switch_simplify]: 4.54002e-06
        [loop_unroll]: 3.81999e-06
        [a_1]: 5.174e-05
        [with_stream_mark]: 8.90001e-06
        [recompute_prepare]: 4.09002e-06
        [updatestate_depend_eliminate]: 2.09e-06
        [updatestate_assign_eliminate]: 1.77999e-06
        [updatestate_loads_eliminate]: 1.39e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 3.48999e-06
        [updatestate_useless_node_eliminater]: 3.51001e-06
        [accelerated_algorithm]: 6.78e-06
        [meta_shard_fg_expand]: 1.59e-06
        [get_grad_eliminate_]: 3.07002e-06
        [merge_forward]: 2.34001e-06
        [cell_reuse_recompute_pass]: 1.71e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.79e-06
        [j_node_and_user_rematch]: 5.72001e-06
        [meta_fg_expand]: 2.396e-05
        [replace_old_param]: 3.63e-06
        [inline_without_move]: 3.30998e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.55001e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 4.70001e-06
        [cse]: 1.06e-05
        [replace_applicator]: 4.11001e-06
[py_interpret_to_execute_after_opt_a]: 8.15e-06
[rewriter_after_opt_a]: 2.621e-05
[convert_after_rewriter]: 5.42999e-06
[order_py_execute_after_rewriter]: 4.60999e-06
[mutable_eliminate]: 0.00050695
[jit_opt_b]: 3.862e-05, [1]
    [Cycle 1]: 3.227e-05, [2]
        [frontend_op_eliminate]: 1.052e-05
        [inline_after_opt_a]: 1.095e-05
[cconv]: 1.66e-05
[loop_unroll]: 0.00040768
[jit_opt_after_cconv]: 0.00010835, [1]
    [Cycle 1]: 0.0001026, [11]
        [c_1]: 1.077e-05
        [parameter_eliminate]: 2.11e-06
        [updatestate_depend_eliminate]: 3.91999e-06
        [updatestate_assign_eliminate]: 1.81e-06
        [updatestate_loads_eliminate]: 1.66998e-06
        [cse]: 1.334e-05
        [call_graph_tuple_transform]: 1.161e-05
        [tuple_list_get_item_eliminator]: 3.71999e-06
        [none_parameter_eliminate]: 9.49978e-07
        [renormalize]: 4.89992e-07
        [switch_simplify]: 3.6e-06
[remove_dup_value]: 1.057e-05
[partial_unused_args_eliminate]: 1.79e-06
[environ_conv]: 3.75e-06
[add_recomputation]: 2.428e-05
[cse_after_recomputation]: 1.821e-05, [1]
    [Cycle 1]: 1.316e-05, [1]
        [cse]: 7.45998e-06
[auto_monad_reorder]: 8.47998e-06
[get_jit_bprop_graph]: 1.30999e-06
[rewriter_after_jit_bprop_graph]: 4.58001e-06
[opt_after_jit_grad]: 0.00043673
[symbol_engine_optimizer]: 6.145e-05, [1]
    [Cycle 1]: 5.567e-05, [6]
        [build]: 2.76e-06
        [elim_shapecalc]: 5.72001e-06
        [elim_not_effective]: 8.67998e-06
        [opt_reshape]: 4.13999e-06
        [fold_const_symbol]: 5.90002e-06
        [renormalize]: 4.99975e-07
[validate]: 2.421e-05
Sums
    bootstrap                            :     0.000182s :  1.17%
    type_inference                       :     0.009108s : 58.45%
    event_method                         :     0.000042s :  0.27%
    auto_monad                           :     0.000079s :  0.51%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000022s :  0.14%
    rewriter_before_opt_a                :     0.000082s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.38%
    jit_opt_a.loop_unroll                :     0.000042s :  0.27%
    jit_opt_a.a_1                        :     0.000931s :  5.97%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000016s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000019s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.10%
    jit_opt_a.merge_forward              :     0.000009s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000053s :  0.34%
    jit_opt_a.meta_fg_expand             :     0.001599s : 10.26%
    jit_opt_a.replace_old_param          :     0.000049s :  0.32%
    jit_opt_a.inline_without_move        :     0.000052s :  0.33%
    jit_opt_a.renormalize                :     0.001415s :  9.08%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.09%
    jit_opt_a.cse                        :     0.000027s :  0.17%
    jit_opt_a.replace_applicator         :     0.000014s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000026s :  0.17%
    convert_after_rewriter               :     0.000005s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000507s :  3.25%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000017s :  0.11%
    loop_unroll                          :     0.000408s :  2.62%
    jit_opt_after_cconv.c_1              :     0.000011s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000013s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000012s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000011s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000024s :  0.16%
    cse_after_recomputation.cse          :     0.000007s :  0.05%
    auto_monad_reorder                   :     0.000008s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000437s :  2.80%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000024s :  0.16%


Time group info:
------[substitution.]   0.000254    50
 0.47% :     0.000001s :      1: substitution.elim_not_effective
 0.33% :     0.000001s :      1: substitution.fold_const_symbol
 1.20% :     0.000003s :      1: substitution.graph_param_transform
54.38% :     0.000138s :      6: substitution.inline
 5.32% :     0.000014s :      2: substitution.inline_without_move
11.81% :     0.000030s :      9: substitution.j_node_and_user_rematch
 1.27% :     0.000003s :      2: substitution.minmaximum_grad
 9.49% :     0.000024s :      3: substitution.partial_eliminate
 2.41% :     0.000006s :      9: substitution.remove_not_recompute_node
 1.68% :     0.000004s :      2: substitution.replace_applicator
 1.23% :     0.000003s :      5: substitution.replace_old_param
 0.87% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.88% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.56% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 5.11% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009048     2
88.27% :     0.007987s :      1: type_inference.infer
11.73% :     0.001061s :      1: type_inference.specialize
------[replace.]   0.000058     8
76.35% :     0.000044s :      6: replace.inline
23.65% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000143     8
94.27% :     0.000135s :      6: match.inline
 5.73% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000163  1109
 1.44% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.53% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.36% :     0.000002s :     18: predicate.addn_check_dump
 1.50% :     0.000002s :     18: predicate.addn_zero_filter
 2.31% :     0.000004s :     18: predicate.arithmetic_simplify
 1.59% :     0.000003s :     18: predicate.cast_eliminate
 0.17% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.28% :     0.000002s :     18: predicate.compare_switch_simplify
 1.51% :     0.000002s :     18: predicate.depend_value_elim
 1.31% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.46% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.39% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.10% :     0.000000s :      1: predicate.elim_not_effective
 0.21% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.38% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.31% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.30% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.44% :     0.000002s :     18: predicate.environ_get_eliminate
 1.42% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.20% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.18% :     0.000007s :     28: predicate.inline
 3.30% :     0.000005s :     33: predicate.inline_without_move
 0.56% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.60% :     0.000003s :     12: predicate.less_batch_normalization
 1.81% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.76% :     0.000003s :     21: predicate.load_eliminater
 0.45% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.73% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.60% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.30% :     0.000002s :     18: predicate.merge_addn
 1.47% :     0.000002s :     18: predicate.minmaximum_grad
 0.71% :     0.000001s :      1: predicate.mutable_eliminate
 0.20% :     0.000000s :      1: predicate.opt_reshape
 2.18% :     0.000004s :     21: predicate.partial_eliminate
 1.34% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.98% :     0.000003s :     18: predicate.reduce_eliminate
 1.55% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.52% :     0.000002s :     22: predicate.replace_applicator
 1.83% :     0.000003s :     33: predicate.replace_old_param
 0.09% :     0.000000s :      1: predicate.reset_defer_inline
 1.65% :     0.000003s :     18: predicate.reshape_eliminate
 1.54% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.49% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     18: predicate.same_eliminate
 0.66% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.37% :     0.000001s :      2: predicate.special_op_eliminate
 1.23% :     0.000002s :     12: predicate.specialize_transform
 1.71% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.63% :     0.000004s :     26: predicate.switch_defer_inline
 2.30% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.85% :     0.000011s :     71: predicate.switch_simplify
 1.46% :     0.000002s :     18: predicate.tile_eliminate
 1.39% :     0.000002s :     18: predicate.transpose_eliminate
 1.60% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.47% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 2.89% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.64% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000003s :     18: predicate.value_based_eliminate
 0.11% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.31% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001172    22
76.61% :     0.000898s :     13: func_graph_cloner_run.FuncGraphClonerGraph
23.39% :     0.000274s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021940    72
 0.12% :     0.000027s :      1: add_recomputation
 0.38% :     0.000083s :      1: auto_monad
 0.05% :     0.000011s :      1: auto_monad_reorder
 0.89% :     0.000194s :      1: bootstrap
 0.09% :     0.000019s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.09% :     0.000020s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.21% :     0.000045s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
35.67% :     0.007826s :      1: jit_opt_a
 0.51% :     0.000111s :      1: jit_opt_after_cconv
 0.19% :     0.000041s :      1: jit_opt_b
 1.88% :     0.000414s :      1: loop_unroll
 2.34% :     0.000513s :      1: mutable_eliminate
 5.89% :     0.001292s :     26: opt.transform.jit_opt_a
 0.12% :     0.000027s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000016s :      4: opt.transform.jit_opt_b
 0.03% :     0.000007s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000010s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000012s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000021s :      4: opt.transform.symbol_engine_opt
 2.02% :     0.000444s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.11% :     0.000025s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000013s :      1: remove_dup_value
 4.89% :     0.001072s :      1: renormalize.infer
 1.53% :     0.000336s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000029s :      1: rewriter_after_opt_a
 0.39% :     0.000085s :      1: rewriter_before_opt_a
 0.29% :     0.000064s :      1: symbol_engine_optimizer
41.57% :     0.009120s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.639.420 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c368940> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.640.066 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.640.481 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843c1a60> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.640.964 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0133234, [30]
[bootstrap]: 0.00030074
[type_inference]: 0.00441994
[event_method]: 1.871e-05
[auto_monad]: 6.517e-05
[graph_reusing]: 5.63002e-06
[pre_auto_parallel]: 2.64001e-06
[py_interpret_to_execute]: 3.163e-05
[rewriter_before_opt_a]: 7.303e-05
[expand_dump_flag]: 3.16001e-06
[jit_opt_a]: 0.00598279, [2]
    [Cycle 1]: 0.0017385, [27]
        [switch_simplify]: 5.992e-05
        [loop_unroll]: 2.931e-05
        [a_1]: 0.00056124
        [with_stream_mark]: 2.265e-05
        [recompute_prepare]: 8.56002e-06
        [updatestate_depend_eliminate]: 4.40999e-06
        [updatestate_assign_eliminate]: 3.24001e-06
        [updatestate_loads_eliminate]: 2.41998e-06
        [parameter_eliminate]: 1.84998e-06
        [specialize_transform]: 5.69e-06
        [updatestate_useless_node_eliminater]: 5.44e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 2.48998e-06
        [get_grad_eliminate_]: 5.30001e-06
        [merge_forward]: 3.72002e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.453e-05
        [j_node_and_user_rematch]: 8.74e-06
        [meta_fg_expand]: 2.46e-06
        [replace_old_param]: 8.28001e-06
        [inline_without_move]: 5.27001e-06
        [renormalize]: 0.00070526
        [add_forward_monad_depend]: 6.01e-06
        [auto_monad_grad]: 2.53e-06
        [auto_monad_eliminator]: 1.523e-05
        [cse]: 3.246e-05
        [replace_applicator]: 1.522e-05
    [Cycle 2]: 0.00034364, [27]
        [switch_simplify]: 5.36002e-06
        [loop_unroll]: 4.96002e-06
        [a_1]: 9.124e-05
        [with_stream_mark]: 1.216e-05
        [recompute_prepare]: 4.92e-06
        [updatestate_depend_eliminate]: 2.98e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [parameter_eliminate]: 1.20001e-06
        [specialize_transform]: 6.39001e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 5.92001e-06
        [meta_shard_fg_expand]: 1.46002e-06
        [get_grad_eliminate_]: 4.89998e-06
        [merge_forward]: 2.99999e-06
        [cell_reuse_recompute_pass]: 1.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.671e-05
        [j_node_and_user_rematch]: 8.97e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 8.16002e-06
        [inline_without_move]: 4.60001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.42e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 7.45e-06
        [cse]: 1.175e-05
        [replace_applicator]: 5.05001e-06
[py_interpret_to_execute_after_opt_a]: 1.343e-05
[rewriter_after_opt_a]: 3.997e-05
[convert_after_rewriter]: 7.16001e-06
[order_py_execute_after_rewriter]: 7.73999e-06
[mutable_eliminate]: 0.00076426
[jit_opt_b]: 5.421e-05, [1]
    [Cycle 1]: 4.505e-05, [2]
        [frontend_op_eliminate]: 1.684e-05
        [inline_after_opt_a]: 1.584e-05
[cconv]: 3.11e-05
[loop_unroll]: 0.00044795
[jit_opt_after_cconv]: 0.00014746, [1]
    [Cycle 1]: 0.00014115, [11]
        [c_1]: 2.031e-05
        [parameter_eliminate]: 2.91e-06
        [updatestate_depend_eliminate]: 6.88e-06
        [updatestate_assign_eliminate]: 2.66e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [cse]: 2.423e-05
        [call_graph_tuple_transform]: 1.917e-05
        [tuple_list_get_item_eliminator]: 5.23002e-06
        [none_parameter_eliminate]: 1.67001e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 5.40999e-06
[remove_dup_value]: 1.685e-05
[partial_unused_args_eliminate]: 2.38998e-06
[environ_conv]: 5.49e-06
[add_recomputation]: 5.415e-05
[cse_after_recomputation]: 1.992e-05, [1]
    [Cycle 1]: 1.439e-05, [1]
        [cse]: 8.75001e-06
[auto_monad_reorder]: 1.553e-05
[get_jit_bprop_graph]: 1.97999e-06
[rewriter_after_jit_bprop_graph]: 7.53999e-06
[opt_after_jit_grad]: 0.00048171
[symbol_engine_optimizer]: 7.408e-05, [1]
    [Cycle 1]: 6.771e-05, [6]
        [build]: 4.35e-06
        [elim_shapecalc]: 8.03999e-06
        [elim_not_effective]: 1.301e-05
        [opt_reshape]: 5.37999e-06
        [fold_const_symbol]: 9.05001e-06
        [renormalize]: 8.99978e-07
[validate]: 3.978e-05
Sums
    bootstrap                            :     0.000301s :  3.42%
    type_inference                       :     0.004420s : 50.27%
    event_method                         :     0.000019s :  0.21%
    auto_monad                           :     0.000065s :  0.74%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.36%
    rewriter_before_opt_a                :     0.000073s :  0.83%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000065s :  0.74%
    jit_opt_a.loop_unroll                :     0.000034s :  0.39%
    jit_opt_a.a_1                        :     0.000652s :  7.42%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.40%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.36%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000705s :  8.02%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.26%
    jit_opt_a.cse                        :     0.000044s :  0.50%
    jit_opt_a.replace_applicator         :     0.000020s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.15%
    rewriter_after_opt_a                 :     0.000040s :  0.45%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000008s :  0.09%
    mutable_eliminate                    :     0.000764s :  8.69%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.18%
    cconv                                :     0.000031s :  0.35%
    loop_unroll                          :     0.000448s :  5.09%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000024s :  0.28%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000017s :  0.19%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000054s :  0.62%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.09%
    opt_after_jit_grad                   :     0.000482s :  5.48%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.45%


Time group info:
------[substitution.]   0.000230    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000002s :      2: substitution.fold_const_symbol
 2.51% :     0.000006s :      3: substitution.graph_param_transform
83.93% :     0.000193s :      4: substitution.inline
 1.80% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.64% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.02% :     0.000005s :      2: substitution.replace_old_param
 5.55% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004349     2
81.54% :     0.003546s :      1: type_inference.infer
18.46% :     0.000803s :      1: type_inference.specialize
------[replace.]   0.000071     6
67.22% :     0.000048s :      4: replace.inline
32.78% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000202     6
94.24% :     0.000190s :      4: match.inline
 5.76% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.47% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 2.34% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.32% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.42% :     0.000002s :     10: predicate.depend_value_elim
 0.99% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.06% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.25% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.67% :     0.000001s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.41% :     0.000001s :      3: predicate.graph_param_transform
 5.62% :     0.000007s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.64% :     0.000002s :      6: predicate.less_batch_normalization
 1.32% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.57% :     0.000002s :     15: predicate.load_eliminater
 1.28% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.61% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.34% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.10% :     0.000003s :     15: predicate.partial_eliminate
 1.37% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.88% :     0.000002s :     10: predicate.reduce_eliminate
 1.66% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.79% :     0.000002s :     18: predicate.replace_applicator
 0.60% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.14% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.11% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.17% :     0.000003s :     16: predicate.switch_defer_inline
 1.91% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.33% :     0.000009s :     53: predicate.switch_simplify
 1.23% :     0.000002s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.60% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 2.01% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.92% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.76% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000567    12
50.93% :     0.000289s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.45% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.62% :     0.000236s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014952    72
 0.38% :     0.000057s :      1: add_recomputation
 0.46% :     0.000068s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.15% :     0.000322s :      1: bootstrap
 0.23% :     0.000034s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
40.04% :     0.005987s :      1: jit_opt_a
 1.01% :     0.000151s :      1: jit_opt_after_cconv
 0.38% :     0.000057s :      1: jit_opt_b
 3.05% :     0.000456s :      1: loop_unroll
 5.18% :     0.000775s :      1: mutable_eliminate
 5.84% :     0.000874s :     26: opt.transform.jit_opt_a
 0.31% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.29% :     0.000492s :      1: opt_after_jit_grad
 0.07% :     0.000010s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000034s :      1: py_interpret_to_execute
 0.10% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000020s :      1: remove_dup_value
 2.56% :     0.000383s :      1: renormalize.infer
 2.09% :     0.000313s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.29% :     0.000043s :      1: rewriter_after_opt_a
 0.51% :     0.000076s :      1: rewriter_before_opt_a
 0.51% :     0.000077s :      1: symbol_engine_optimizer
29.68% :     0.004438s :      1: type_inference


TotalTime = 0.0304363, [33]
[bootstrap]: 0.00017218
[type_inference]: 0.00921238
[event_method]: 4.195e-05
[auto_monad]: 6.177e-05
[graph_reusing]: 4.95001e-06
[pre_auto_parallel]: 1.99999e-06
[py_interpret_to_execute]: 2.064e-05
[rewriter_before_opt_a]: 7.892e-05
[expand_dump_flag]: 1.96e-06
[jit_opt_a]: 0.0106626, [3]
    [Cycle 1]: 0.00517208, [27]
        [switch_simplify]: 5.276e-05
        [loop_unroll]: 4.864e-05
        [a_1]: 0.00076929
        [with_stream_mark]: 1.889e-05
        [recompute_prepare]: 1.669e-05
        [updatestate_depend_eliminate]: 7.09001e-06
        [updatestate_assign_eliminate]: 5.66998e-06
        [updatestate_loads_eliminate]: 5.11002e-06
        [parameter_eliminate]: 1.58002e-06
        [specialize_transform]: 1.258e-05
        [updatestate_useless_node_eliminater]: 1.148e-05
        [accelerated_algorithm]: 1.193e-05
        [meta_shard_fg_expand]: 2.32001e-06
        [get_grad_eliminate_]: 1.205e-05
        [merge_forward]: 6.59001e-06
        [cell_reuse_recompute_pass]: 7.29982e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.387e-05
        [j_node_and_user_rematch]: 2.104e-05
        [meta_fg_expand]: 0.00122024
        [replace_old_param]: 4.831e-05
        [inline_without_move]: 4.98e-05
        [renormalize]: 0.00238355
        [add_forward_monad_depend]: 9.14998e-06
        [auto_monad_grad]: 4.48001e-06
        [auto_monad_eliminator]: 4.787e-05
        [cse]: 0.00013734
        [replace_applicator]: 6.261e-05
    [Cycle 2]: 0.00205344, [27]
        [switch_simplify]: 3.67e-05
        [loop_unroll]: 3.536e-05
        [a_1]: 0.00109728
        [with_stream_mark]: 1.345e-05
        [recompute_prepare]: 8.48999e-06
        [updatestate_depend_eliminate]: 3.78001e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 3.08e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 7.48e-06
        [updatestate_useless_node_eliminater]: 6.75002e-06
        [accelerated_algorithm]: 7.6e-06
        [meta_shard_fg_expand]: 1.81e-06
        [get_grad_eliminate_]: 6.88e-06
        [merge_forward]: 3.70998e-06
        [cell_reuse_recompute_pass]: 1.36002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.377e-05
        [j_node_and_user_rematch]: 1.137e-05
        [meta_fg_expand]: 4.075e-05
        [replace_old_param]: 1.109e-05
        [inline_without_move]: 6.95998e-06
        [renormalize]: 0.00053317
        [add_forward_monad_depend]: 4.53001e-06
        [auto_monad_grad]: 1.55001e-06
        [auto_monad_eliminator]: 1.362e-05
        [cse]: 2.448e-05
        [replace_applicator]: 1.322e-05
    [Cycle 3]: 0.00040729, [27]
        [switch_simplify]: 7.29001e-06
        [loop_unroll]: 6.96001e-06
        [a_1]: 0.00014192
        [with_stream_mark]: 1.035e-05
        [recompute_prepare]: 6.66e-06
        [updatestate_depend_eliminate]: 3.67998e-06
        [updatestate_assign_eliminate]: 3.26999e-06
        [updatestate_loads_eliminate]: 3.09999e-06
        [parameter_eliminate]: 8.89995e-07
        [specialize_transform]: 7.18e-06
        [updatestate_useless_node_eliminater]: 6.81999e-06
        [accelerated_algorithm]: 7e-06
        [meta_shard_fg_expand]: 1.59e-06
        [get_grad_eliminate_]: 6.47001e-06
        [merge_forward]: 3.38e-06
        [cell_reuse_recompute_pass]: 1.72999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.62e-05
        [j_node_and_user_rematch]: 1.117e-05
        [meta_fg_expand]: 2.49001e-06
        [replace_old_param]: 9.47001e-06
        [inline_without_move]: 6.81001e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.28002e-06
        [auto_monad_grad]: 1.02e-06
        [auto_monad_eliminator]: 7.56999e-06
        [cse]: 1.614e-05
        [replace_applicator]: 6.89001e-06
[py_interpret_to_execute_after_opt_a]: 1.194e-05
[rewriter_after_opt_a]: 3.896e-05
[convert_after_rewriter]: 7.23999e-06
[order_py_execute_after_rewriter]: 5.48002e-06
[mutable_eliminate]: 0.00063482
[jit_opt_b]: 5.959e-05, [1]
    [Cycle 1]: 5.254e-05, [2]
        [frontend_op_eliminate]: 2.164e-05
        [inline_after_opt_a]: 2.025e-05
[cconv]: 1.963e-05
[loop_unroll]: 0.0004149
[jit_opt_after_cconv]: 0.00016613, [1]
    [Cycle 1]: 0.00015943, [11]
        [c_1]: 3.049e-05
        [parameter_eliminate]: 2.53e-06
        [updatestate_depend_eliminate]: 6.01998e-06
        [updatestate_assign_eliminate]: 3.38e-06
        [updatestate_loads_eliminate]: 3.11001e-06
        [cse]: 2.702e-05
        [call_graph_tuple_transform]: 2.201e-05
        [tuple_list_get_item_eliminator]: 7.11001e-06
        [none_parameter_eliminate]: 1.40999e-06
        [renormalize]: 4.2998e-07
        [switch_simplify]: 7.80998e-06
[remove_dup_value]: 1.588e-05
[partial_unused_args_eliminate]: 1.91e-06
[environ_conv]: 6.61e-06
[add_recomputation]: 5.277e-05
[cse_after_recomputation]: 3.167e-05, [1]
    [Cycle 1]: 2.543e-05, [1]
        [cse]: 1.936e-05
[auto_monad_reorder]: 1.697e-05
[get_jit_bprop_graph]: 2.01003e-06
[rewriter_after_jit_bprop_graph]: 6.04999e-06
[opt_after_jit_grad]: 0.00046281
[symbol_engine_optimizer]: 8.532e-05, [1]
    [Cycle 1]: 7.899e-05, [6]
        [build]: 3.85998e-06
        [elim_shapecalc]: 9.44998e-06
        [elim_not_effective]: 1.615e-05
        [opt_reshape]: 8.15999e-06
        [fold_const_symbol]: 1.31e-05
        [renormalize]: 4.30009e-07
[validate]: 3.779e-05
[backend_pass]: 7.00005e-07
[task_emit]: 0.0078964
[execute]: 7.20998e-06
Sums
    bootstrap                            :     0.000172s :  0.65%
    type_inference                       :     0.009212s : 34.55%
    event_method                         :     0.000042s :  0.16%
    auto_monad                           :     0.000062s :  0.23%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000021s :  0.08%
    rewriter_before_opt_a                :     0.000079s :  0.30%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000097s :  0.36%
    jit_opt_a.loop_unroll                :     0.000091s :  0.34%
    jit_opt_a.a_1                        :     0.002008s :  7.53%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000027s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000027s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.10%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000054s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000044s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001263s :  4.74%
    jit_opt_a.replace_old_param          :     0.000069s :  0.26%
    jit_opt_a.inline_without_move        :     0.000064s :  0.24%
    jit_opt_a.renormalize                :     0.002917s : 10.94%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000069s :  0.26%
    jit_opt_a.cse                        :     0.000178s :  0.67%
    jit_opt_a.replace_applicator         :     0.000083s :  0.31%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000039s :  0.15%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000635s :  2.38%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.08%
    cconv                                :     0.000020s :  0.07%
    loop_unroll                          :     0.000415s :  1.56%
    jit_opt_after_cconv.c_1              :     0.000030s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000008s :  0.03%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000053s :  0.20%
    cse_after_recomputation.cse          :     0.000019s :  0.07%
    auto_monad_reorder                   :     0.000017s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000463s :  1.74%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007896s : 29.62%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000448   118
 5.37% :     0.000024s :      1: substitution.arithmetic_simplify
 0.48% :     0.000002s :      4: substitution.elim_not_effective
 0.50% :     0.000002s :      4: substitution.fold_const_symbol
 1.27% :     0.000006s :      5: substitution.graph_param_transform
58.62% :     0.000263s :     12: substitution.inline
 3.50% :     0.000016s :      2: substitution.inline_without_move
 1.71% :     0.000008s :     16: substitution.j_node_and_user_rematch
 2.08% :     0.000009s :      7: substitution.minmaximum_grad
 1.37% :     0.000006s :      5: substitution.partial_eliminate
 2.55% :     0.000011s :     16: substitution.remove_not_recompute_node
 4.92% :     0.000022s :     10: substitution.replace_applicator
 1.55% :     0.000007s :      7: substitution.replace_old_param
 0.46% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.21% :     0.000019s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.69% :     0.000012s :      7: substitution.tuple_list_get_item_depend_reorder
 8.71% :     0.000039s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009157     2
88.04% :     0.008062s :      1: type_inference.infer
11.96% :     0.001095s :      1: type_inference.specialize
------[replace.]   0.000136    20
 3.53% :     0.000005s :      1: replace.arithmetic_simplify
60.96% :     0.000083s :     12: replace.inline
35.51% :     0.000048s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000295    20
 7.89% :     0.000023s :      1: match.arithmetic_simplify
87.12% :     0.000257s :     12: match.inline
 4.99% :     0.000015s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000379  2790
 1.63% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.47% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.42% :     0.000005s :     46: predicate.addn_check_dump
 1.65% :     0.000006s :     46: predicate.addn_zero_filter
 2.07% :     0.000008s :     47: predicate.arithmetic_simplify
 1.70% :     0.000006s :     47: predicate.cast_eliminate
 0.23% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.46% :     0.000006s :     46: predicate.compare_switch_simplify
 1.57% :     0.000006s :     46: predicate.depend_value_elim
 1.50% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.56% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.56% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.31% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      5: predicate.elim_not_effective
 0.22% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.55% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.46% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.47% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.55% :     0.000006s :     47: predicate.environ_get_eliminate
 1.48% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      5: predicate.fold_const_symbol
 0.86% :     0.000003s :     21: predicate.get_grad_eliminate
 0.11% :     0.000000s :      5: predicate.graph_param_transform
 4.23% :     0.000016s :     76: predicate.inline
 1.76% :     0.000007s :     42: predicate.inline_without_move
 0.37% :     0.000001s :     21: predicate.j_node_and_user_rematch
 1.08% :     0.000004s :     21: predicate.less_batch_normalization
 1.80% :     0.000007s :     54: predicate.list_to_tuple_eliminator_
 1.98% :     0.000007s :     59: predicate.load_eliminater
 0.47% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.09% :     0.000012s :     89: predicate.loop_unroll_before_grad
 1.81% :     0.000007s :     52: predicate.make_slice_get_slice_eliminator
 1.46% :     0.000006s :     46: predicate.merge_addn
 1.52% :     0.000006s :     47: predicate.minmaximum_grad
 0.52% :     0.000002s :      5: predicate.mutable_eliminate
 0.21% :     0.000001s :      5: predicate.opt_reshape
 2.33% :     0.000009s :     59: predicate.partial_eliminate
 1.46% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.93% :     0.000007s :     47: predicate.reduce_eliminate
 1.84% :     0.000007s :     54: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.57% :     0.000010s :    105: predicate.replace_applicator
 0.89% :     0.000003s :     42: predicate.replace_old_param
 0.14% :     0.000001s :      5: predicate.reset_defer_inline
 1.66% :     0.000006s :     47: predicate.reshape_eliminate
 1.50% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.38% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.52% :     0.000006s :     46: predicate.same_eliminate
 0.49% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.40% :     0.000002s :     10: predicate.special_op_eliminate
 0.88% :     0.000003s :     21: predicate.specialize_transform
 1.65% :     0.000006s :     46: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.20% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.63% :     0.000010s :     66: predicate.switch_defer_inline
 2.58% :     0.000010s :     66: predicate.switch_layer_defer_inline
 6.12% :     0.000023s :    160: predicate.switch_simplify
 1.63% :     0.000006s :     47: predicate.tile_eliminate
 1.56% :     0.000006s :     47: predicate.transpose_eliminate
 1.84% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.67% :     0.000006s :     47: predicate.tuple_list_get_item_depend_reorder
 2.92% :     0.000011s :     64: predicate.tuple_list_get_item_eliminator
 1.95% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000007s :     54: predicate.tuple_to_list_eliminator_
 1.95% :     0.000007s :     59: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000011s :     80: predicate.updatestate_useless_node_eliminater
 1.89% :     0.000007s :     46: predicate.value_based_eliminate
 0.17% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.25% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001333    29
57.62% :     0.000768s :     13: func_graph_cloner_run.FuncGraphClonerGraph
42.38% :     0.000565s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.036106    91
 0.16% :     0.000056s :      1: add_recomputation
 0.18% :     0.000066s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.51% :     0.000183s :      1: bootstrap
 0.06% :     0.000022s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000034s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.13% :     0.000045s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000004s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
29.54% :     0.010666s :      1: jit_opt_a
 0.47% :     0.000169s :      1: jit_opt_after_cconv
 0.17% :     0.000062s :      1: jit_opt_b
 1.17% :     0.000423s :      1: loop_unroll
 1.78% :     0.000643s :      1: mutable_eliminate
 7.22% :     0.002608s :     39: opt.transform.jit_opt_a
 0.18% :     0.000064s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000036s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000044s :      4: opt.transform.symbol_engine_opt
 1.30% :     0.000471s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.06% :     0.000023s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 4.36% :     0.001574s :      2: renormalize.infer
 3.68% :     0.001330s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000042s :      1: rewriter_after_opt_a
 0.23% :     0.000082s :      1: rewriter_before_opt_a
 0.24% :     0.000088s :      1: symbol_engine_optimizer
21.90% :     0.007908s :      1: task_emit
25.54% :     0.009222s :      1: type_inference
 0.17% :     0.000060s :      1: validate


TotalTime = 0.0173265, [33]
[bootstrap]: 0.00026633
[type_inference]: 0.00425926
[event_method]: 1.641e-05
[auto_monad]: 6.09e-05
[graph_reusing]: 5.52001e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 3.021e-05
[rewriter_before_opt_a]: 6.844e-05
[expand_dump_flag]: 3.23998e-06
[jit_opt_a]: 0.00483279, [2]
    [Cycle 1]: 0.0014257, [27]
        [switch_simplify]: 4.66e-05
        [loop_unroll]: 3.016e-05
        [a_1]: 0.00050916
        [with_stream_mark]: 1.677e-05
        [recompute_prepare]: 7.73999e-06
        [updatestate_depend_eliminate]: 3.47997e-06
        [updatestate_assign_eliminate]: 3.21001e-06
        [updatestate_loads_eliminate]: 2.41998e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 6.17001e-06
        [updatestate_useless_node_eliminater]: 5.32999e-06
        [accelerated_algorithm]: 5.62999e-06
        [meta_shard_fg_expand]: 2.21e-06
        [get_grad_eliminate_]: 5.42001e-06
        [merge_forward]: 3.67998e-06
        [cell_reuse_recompute_pass]: 1.09998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.361e-05
        [j_node_and_user_rematch]: 9.14e-06
        [meta_fg_expand]: 2.41e-06
        [replace_old_param]: 8.32998e-06
        [inline_without_move]: 5.49998e-06
        [renormalize]: 0.00052556
        [add_forward_monad_depend]: 4.89e-06
        [auto_monad_grad]: 2.53998e-06
        [auto_monad_eliminator]: 1.267e-05
        [cse]: 1.674e-05
        [replace_applicator]: 1.136e-05
    [Cycle 2]: 0.00031956, [27]
        [switch_simplify]: 5.70001e-06
        [loop_unroll]: 5.24e-06
        [a_1]: 9.108e-05
        [with_stream_mark]: 8.94e-06
        [recompute_prepare]: 5.12999e-06
        [updatestate_depend_eliminate]: 2.56998e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 5.21998e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 4.85999e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 4.90001e-06
        [merge_forward]: 2.59999e-06
        [cell_reuse_recompute_pass]: 1.92999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.285e-05
        [j_node_and_user_rematch]: 7.77998e-06
        [meta_fg_expand]: 1.50999e-06
        [replace_old_param]: 7.18e-06
        [inline_without_move]: 5.00999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 5.56e-06
        [cse]: 1.011e-05
        [replace_applicator]: 5.54e-06
[py_interpret_to_execute_after_opt_a]: 9.52999e-06
[rewriter_after_opt_a]: 3.278e-05
[convert_after_rewriter]: 7.11999e-06
[order_py_execute_after_rewriter]: 4.87e-06
[mutable_eliminate]: 0.00054002
[jit_opt_b]: 7.503e-05, [1]
    [Cycle 1]: 6.705e-05, [2]
        [frontend_op_eliminate]: 1.528e-05
        [inline_after_opt_a]: 3.71e-05
[cconv]: 2.119e-05
[loop_unroll]: 0.00045279
[jit_opt_after_cconv]: 0.00014475, [1]
    [Cycle 1]: 0.00013792, [11]
        [c_1]: 2.136e-05
        [parameter_eliminate]: 2.21e-06
        [updatestate_depend_eliminate]: 4.58999e-06
        [updatestate_assign_eliminate]: 2.76e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [cse]: 2.083e-05
        [call_graph_tuple_transform]: 1.929e-05
        [tuple_list_get_item_eliminator]: 5.69e-06
        [none_parameter_eliminate]: 1.67999e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 5.21002e-06
[remove_dup_value]: 1.307e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 5.24e-06
[add_recomputation]: 4.335e-05
[cse_after_recomputation]: 2.016e-05, [1]
    [Cycle 1]: 1.491e-05, [1]
        [cse]: 9.32001e-06
[auto_monad_reorder]: 1.493e-05
[get_jit_bprop_graph]: 1.95001e-06
[rewriter_after_jit_bprop_graph]: 4.61002e-06
[opt_after_jit_grad]: 0.00044984
[symbol_engine_optimizer]: 7.275e-05, [1]
    [Cycle 1]: 6.65e-05, [6]
        [build]: 3.37002e-06
        [elim_shapecalc]: 7.78001e-06
        [elim_not_effective]: 1.25e-05
        [opt_reshape]: 5.67999e-06
        [fold_const_symbol]: 8.75999e-06
        [renormalize]: 4.39992e-07
[validate]: 3.107e-05
[backend_pass]: 8.09989e-07
[task_emit]: 0.00565277
[execute]: 5.64e-06
Sums
    bootstrap                            :     0.000266s :  1.95%
    type_inference                       :     0.004259s : 31.17%
    event_method                         :     0.000016s :  0.12%
    auto_monad                           :     0.000061s :  0.45%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.22%
    rewriter_before_opt_a                :     0.000068s :  0.50%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000052s :  0.38%
    jit_opt_a.loop_unroll                :     0.000035s :  0.26%
    jit_opt_a.a_1                        :     0.000600s :  4.39%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.08%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000016s :  0.11%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.000526s :  3.85%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.13%
    jit_opt_a.cse                        :     0.000027s :  0.20%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000033s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000540s :  3.95%
    jit_opt_b.frontend_op_eliminate      :     0.000015s :  0.11%
    jit_opt_b.inline_after_opt_a         :     0.000037s :  0.27%
    cconv                                :     0.000021s :  0.16%
    loop_unroll                          :     0.000453s :  3.31%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000043s :  0.32%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000450s :  3.29%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005653s : 41.37%
    execute                              :     0.000006s :  0.04%


Time group info:
------[substitution.]   0.000180    23
 1.13% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 3.25% :     0.000006s :      3: substitution.graph_param_transform
81.80% :     0.000147s :      4: substitution.inline
 1.47% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.47% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.86% :     0.000003s :      2: substitution.replace_old_param
 7.32% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004202     2
81.97% :     0.003444s :      1: type_inference.infer
18.03% :     0.000758s :      1: type_inference.specialize
------[replace.]   0.000066     6
66.92% :     0.000044s :      4: replace.inline
33.08% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000157     6
92.38% :     0.000145s :      4: match.inline
 7.62% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.21% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.20% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.35% :     0.000002s :     10: predicate.addn_zero_filter
 2.00% :     0.000002s :     10: predicate.arithmetic_simplify
 1.39% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.48% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.69% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.72% :     0.000007s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.40% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.94% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.51% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.05% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.32% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.92% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.70% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.83% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.19% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.62% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.96% :     0.000001s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.48% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.28% :     0.000008s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.38% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.78% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000533    12
50.15% :     0.000267s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.09% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.76% :     0.000238s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018766    76
 0.25% :     0.000046s :      1: add_recomputation
 0.34% :     0.000064s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.48% :     0.000278s :      1: bootstrap
 0.13% :     0.000024s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.05% :     0.000009s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.77% :     0.004835s :      1: jit_opt_a
 0.79% :     0.000148s :      1: jit_opt_after_cconv
 0.42% :     0.000078s :      1: jit_opt_b
 2.46% :     0.000462s :      1: loop_unroll
 2.92% :     0.000548s :      1: mutable_eliminate
 4.28% :     0.000803s :     26: opt.transform.jit_opt_a
 0.26% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.44% :     0.000458s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 1.42% :     0.000266s :      1: renormalize.infer
 1.34% :     0.000252s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000036s :      1: rewriter_after_opt_a
 0.38% :     0.000071s :      1: rewriter_before_opt_a
 0.40% :     0.000075s :      1: symbol_engine_optimizer
30.17% :     0.005662s :      1: task_emit
22.75% :     0.004270s :      1: type_inference
 0.26% :     0.000048s :      1: validate

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.655.196 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80caf70> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.655.788 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0199878, [30]
[bootstrap]: 0.00017314
[type_inference]: 0.00915978
[event_method]: 4.195e-05
[auto_monad]: 9.238e-05
[graph_reusing]: 6.71e-06
[pre_auto_parallel]: 2.76e-06
[py_interpret_to_execute]: 2.477e-05
[rewriter_before_opt_a]: 8.825e-05
[expand_dump_flag]: 3.51001e-06
[jit_opt_a]: 0.0084687, [2]
    [Cycle 1]: 0.00482175, [27]
        [switch_simplify]: 6.095e-05
        [loop_unroll]: 3.815e-05
        [a_1]: 0.00093302
        [with_stream_mark]: 3.085e-05
        [recompute_prepare]: 1.788e-05
        [updatestate_depend_eliminate]: 8.02998e-06
        [updatestate_assign_eliminate]: 6.82002e-06
        [updatestate_loads_eliminate]: 6.09999e-06
        [parameter_eliminate]: 2.93998e-06
        [specialize_transform]: 1.337e-05
        [updatestate_useless_node_eliminater]: 1.167e-05
        [accelerated_algorithm]: 1.222e-05
        [meta_shard_fg_expand]: 3.3e-06
        [get_grad_eliminate_]: 1.19e-05
        [merge_forward]: 8.89e-06
        [cell_reuse_recompute_pass]: 1.00001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.569e-05
        [j_node_and_user_rematch]: 2.328e-05
        [meta_fg_expand]: 0.00170405
        [replace_old_param]: 5.09e-05
        [inline_without_move]: 5.182e-05
        [renormalize]: 0.00154036
        [add_forward_monad_depend]: 5.28002e-06
        [auto_monad_grad]: 2.61e-06
        [auto_monad_eliminator]: 1.155e-05
        [cse]: 2.522e-05
        [replace_applicator]: 1.012e-05
    [Cycle 2]: 0.00026581, [27]
        [switch_simplify]: 4.51002e-06
        [loop_unroll]: 3.91999e-06
        [a_1]: 3.981e-05
        [with_stream_mark]: 8.2e-06
        [recompute_prepare]: 3.8e-06
        [updatestate_depend_eliminate]: 2.17999e-06
        [updatestate_assign_eliminate]: 1.91e-06
        [updatestate_loads_eliminate]: 1.52001e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 3.8e-06
        [updatestate_useless_node_eliminater]: 3.58999e-06
        [accelerated_algorithm]: 3.39001e-06
        [meta_shard_fg_expand]: 1.03001e-06
        [get_grad_eliminate_]: 3.12002e-06
        [merge_forward]: 1.92999e-06
        [cell_reuse_recompute_pass]: 1.74998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.87001e-06
        [j_node_and_user_rematch]: 5.20999e-06
        [meta_fg_expand]: 2.411e-05
        [replace_old_param]: 3.56999e-06
        [inline_without_move]: 3.21001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.28002e-06
        [auto_monad_grad]: 7.79983e-07
        [auto_monad_eliminator]: 4.25999e-06
        [cse]: 1.063e-05
        [replace_applicator]: 6.50002e-06
[py_interpret_to_execute_after_opt_a]: 1.112e-05
[rewriter_after_opt_a]: 3.503e-05
[convert_after_rewriter]: 7.09001e-06
[order_py_execute_after_rewriter]: 4.92e-06
[mutable_eliminate]: 0.00052237
[jit_opt_b]: 4.001e-05, [1]
    [Cycle 1]: 3.352e-05, [2]
        [frontend_op_eliminate]: 1.116e-05
        [inline_after_opt_a]: 1.098e-05
[cconv]: 2.082e-05
[loop_unroll]: 0.00040827
[jit_opt_after_cconv]: 0.00011174, [1]
    [Cycle 1]: 0.0001056, [11]
        [c_1]: 1.184e-05
        [parameter_eliminate]: 2.14999e-06
        [updatestate_depend_eliminate]: 3.63e-06
        [updatestate_assign_eliminate]: 1.80001e-06
        [updatestate_loads_eliminate]: 1.71998e-06
        [cse]: 1.383e-05
        [call_graph_tuple_transform]: 1.245e-05
        [tuple_list_get_item_eliminator]: 3.81999e-06
        [none_parameter_eliminate]: 1.37999e-06
        [renormalize]: 3.29979e-07
        [switch_simplify]: 3.82002e-06
[remove_dup_value]: 1.364e-05
[partial_unused_args_eliminate]: 2.21e-06
[environ_conv]: 4.87998e-06
[add_recomputation]: 3.159e-05
[cse_after_recomputation]: 1.988e-05, [1]
    [Cycle 1]: 1.447e-05, [1]
        [cse]: 8.00999e-06
[auto_monad_reorder]: 1.322e-05
[get_jit_bprop_graph]: 1.90001e-06
[rewriter_after_jit_bprop_graph]: 4.74998e-06
[opt_after_jit_grad]: 0.00043672
[symbol_engine_optimizer]: 6.174e-05, [1]
    [Cycle 1]: 5.559e-05, [6]
        [build]: 3.11999e-06
        [elim_shapecalc]: 5.85002e-06
TotalTime = 0.020941, [33]
[bootstrap]: 0.00021025
[type_inference]: 0.0102901
[event_method]: 4.77e-05
[auto_monad]: 6.584e-05
[graph_reusing]: 4.99e-06
[pre_auto_parallel]: 1.97001e-06
[py_interpret_to_execute]: 2.4e-05
[rewriter_before_opt_a]: 8.288e-05
[expand_dump_flag]: 1.99999e-06
[jit_opt_a]: 0.00828496, [2]
    [Cycle 1]: 0.00465419, [27]
        [switch_simplify]: 6e-05
        [loop_unroll]: 3.731e-05
        [a_1]: 0.00083003
        [with_stream_mark]: 2.108e-05
        [recompute_prepare]: 1.776e-05
        [updatestate_depend_eliminate]: 7.09001e-06
        [updatestate_assign_eliminate]: 5.44e-06
        [updatestate_loads_eliminate]: 5.37001e-06
        [parameter_eliminate]: 1.77001e-06
        [specialize_transform]: 1.303e-05
        [updatestate_useless_node_eliminater]: 1.283e-05
        [accelerated_algorithm]: 1.22e-05
        [meta_shard_fg_expand]: 3.08e-06
        [get_grad_eliminate_]: 1.179e-05
        [merge_forward]: 6.87002e-06
        [cell_reuse_recompute_pass]: 7.59988e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.345e-05
        [j_node_and_user_rematch]: 2.189e-05
        [meta_fg_expand]: 0.00141785
        [replace_old_param]: 5.21e-05
        [inline_without_move]: 5.018e-05
        [renormalize]: 0.00180281
        [add_forward_monad_depend]: 4.55001e-06
        [auto_monad_grad]: 1.77999e-06
        [auto_monad_eliminator]: 9.22001e-06
        [cse]: 1.514e-05
        [replace_applicator]: 1.109e-05
    [Cycle 2]: 0.00028959, [27]
        [switch_simplify]: 4.23999e-06
        [loop_unroll]: 3.6e-06
        [a_1]: 4.151e-05
        [with_stream_mark]: 8.30999e-06
        [recompute_prepare]: 3.78001e-06
        [updatestate_depend_eliminate]: 2.06003e-06
        [updatestate_assign_eliminate]: 1.80001e-06
        [updatestate_loads_eliminate]: 1.53002e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 3.56001e-06
        [updatestate_useless_node_eliminater]: 3.71999e-06
        [accelerated_algorithm]: 3.78999e-06
        [meta_shard_fg_expand]: 1.12e-06
        [get_grad_eliminate_]: 3.26001e-06
        [merge_forward]: 1.97001e-06
        [cell_reuse_recompute_pass]: 2.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.59999e-06
        [j_node_and_user_rematch]: 5.79999e-06
        [meta_fg_expand]: 3.883e-05
        [replace_old_param]: 3.89002e-06
        [inline_without_move]: 3.48e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 4.62998e-06
        [cse]: 1.083e-05
        [replace_applicator]: 4.33999e-06
[py_interpret_to_execute_after_opt_a]: 8.60001e-06
[rewriter_after_opt_a]: 2.581e-05
[convert_after_rewriter]: 5.00999e-06
[order_py_execute_after_rewriter]: 3.46999e-06
[mutable_eliminate]: 0.00051806
[jit_opt_b]: 3.93e-05, [1]
    [Cycle 1]: 3.268e-05, [2]
        [frontend_op_eliminate]: 9.84999e-06
        [inline_after_opt_a]: 1.107e-05
[cconv]: 1.419e-05
[loop_unroll]: 0.00041759
[jit_opt_after_cconv]: 0.00011398, [1]
    [Cycle 1]: 0.00010777, [11]
        [c_1]: 1.26e-05
        [parameter_eliminate]: 2.58e-06
        [updatestate_depend_eliminate]: 4.70001e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 1.59e-06
        [cse]: 1.595e-05
        [call_graph_tuple_transform]: 1.109e-05
        [tuple_list_get_item_eliminator]: 3.74002e-06
        [none_parameter_eliminate]: 9.49978e-07
        [renormalize]: 4.10015e-07
        [switch_simplify]: 3.66999e-06
[remove_dup_value]: 7.4e-06
[partial_unused_args_eliminate]: 1.45999e-06
[environ_conv]: 3.24001e-06
[add_recomputation]: 2.191e-05
[cse_after_recomputation]: 1.88e-05, [1]
    [Cycle 1]: 1.371e-05, [1]
        [cse]: 7.58999e-06
[auto_monad_reorder]: 7.61001e-06
[get_jit_bprop_graph]: 1.34e-06
[rewriter_after_jit_bprop_graph]: 4.35999e-06
[opt_after_jit_grad]: 0.00043885
[symbol_engine_optimizer]: 6.26e-05, [1]
    [Cycle 1]: 5.665e-05, [6]
        [build]: 2.79001e-06
        [elim_shapecalc]: 6.47001e-06
 
        [elim_not_effective]: 8.93002e-06
        [opt_reshape]: 4.07e-06
        [fold_const_symbol]: 5.82001e-06
        [renormalize]: 4.09986e-07
[validate]: 2.45e-05
Sums
    bootstrap                            :     0.000173s :  1.08%
    type_inference                       :     0.009160s : 57.16%
    event_method                         :     0.000042s :  0.26%
    auto_monad                           :     0.000092s :  0.58%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000025s :  0.15%
    rewriter_before_opt_a                :     0.000088s :  0.55%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.41%
    jit_opt_a.loop_unroll                :     0.000042s :  0.26%
    jit_opt_a.a_1                        :     0.000973s :  6.07%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000017s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.09%
    jit_opt_a.merge_forward              :     0.000011s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001728s : 10.78%
    jit_opt_a.replace_old_param          :     0.000054s :  0.34%
    jit_opt_a.inline_without_move        :     0.000055s :  0.34%
    jit_opt_a.renormalize                :     0.001540s :  9.61%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.10%
    jit_opt_a.cse                        :     0.000036s :  0.22%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000522s :  3.26%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000021s :  0.13%
    loop_unroll                          :     0.000408s :  2.55%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000014s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000012s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :       [elim_not_effective]: 8.79e-06
        [opt_reshape]: 4.20999e-06
        [fold_const_symbol]: 5.46002e-06
        [renormalize]: 4.19997e-07
[validate]: 2.172e-05
[backend_pass]: 8.2e-07
[task_emit]: 2.024e-05
[execute]: 1.42999e-06
Sums
    bootstrap                            :     0.000210s :  1.24%
    type_inference                       :     0.010290s : 60.54%
    event_method                         :     0.000048s :  0.28%
    auto_monad                           :     0.000066s :  0.39%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000024s :  0.14%
    rewriter_before_opt_a                :     0.000083s :  0.49%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000064s :  0.38%
    jit_opt_a.loop_unroll                :     0.000041s :  0.24%
    jit_opt_a.a_1                        :     0.000872s :  5.13%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000017s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.09%
    jit_opt_a.merge_forward              :     0.000009s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001457s :  8.57%
    jit_opt_a.replace_old_param          :     0.000056s :  0.33%
    jit_opt_a.inline_without_move        :     0.000054s :  0.32%
    jit_opt_a.renormalize                :     0.001803s : 10.61%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.08%
    jit_opt_a.cse                        :     0.000026s :  0.15%
    jit_opt_a.replace_applicator         :     0.000015s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.05%
    rewriter_after_opt_a                 :     0.000026s :  0.15%
    convert_after_rewriter               :     0.000005s :  0.03%
    order_py_execute_after_rewriter      :     0.000003s :  0.02%
    mutable_eliminate                    :     0.000518s :  3.05%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000014s :  0.08%
    loop_unroll                          :     0.000418s :  2.46%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000011s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000007  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000032s :  0.20%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000437s :  2.73%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000024s :  0.15%

s :  0.04%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000003s :  0.02%
    add_recomputation                    :     0.000022s :  0.13%
    cse_after_recomputation.cse          :     0.000008s :  0.04%
    auto_monad_reorder                   :     0.000008s :  0.04%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000439s :  2.58%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000005s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000022s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000020s :  0.12%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000243    50
 0.57% :     0.000001s :      1: substitution.elim_not_effective
 0.36% :     0.000001s :      1: substitution.fold_const_symbol
 1.61% :     0.000004s :      1: substitution.graph_param_transform
65.90% :     0.000160s :      6: substitution.inline
 6.54% :     0.000016s :      2: substitution.inline_without_move
 2.67% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.59% :     0.000004s :      2: substitution.minmaximum_grad
 1.74% :     0.000004s :      3: substitution.partial_eliminate
 2.68% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.21% :     0.000005s :      2: substitution.replace_applicator
 1.58% :     0.000004s :      5: substitution.replace_old_param
 1.04% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.00% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.93% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 6.57% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009097     2
87.91% :     0.007998s :      1: type_inference.infer
12.09% :     0.001100s :      1: type_inference.specialize
------[replace.]   0.000085     8
83.88% :     0.000071s :      6: replace.inline
16.12% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000168     8
93.60% :     0.000157s :      6: match.inline
 6.40% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000173  1109
 1.44% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.59% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.33% :     0.000002s :     18: predicate.addn_check_dump
 1.58% :     0.000003s :     18: predicate.addn_zero_filter
 2.29% :     0.000004s :     18: predicate.arithmetic_simplify
 1.31% :     0.000002s :     18: predicate.cast_eliminate
 0.47% :     0.000001s :      1: predicate.check_bprop_eliminate
 1.42% :     0.000002s :     18: predicate.compare_switch_simplify
 1.61% :     0.000003s :     18: predicate.depend_value_elim
 1.23% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.56% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.33% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.41% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      1: predicate.elim_not_effective
 0.28% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.27% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.22% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.38% :     0.000002s :     18: predicate.environ_get_eliminate
 1.32% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.08% :     0.000000s :      1: predicate.fold_const_symbol
 1.12% :     0.000002s :     12: predicate.get_grad_eliminate
 0.08% :     0.000000s :      1: predicate.graph_param_transform
 4.35% :     0.000008s :     28: predicate.inline
 3.23% :     0.000006s :     33: predicate.inline_without_move
 0.50% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.18% :     0.000002s :     12: predicate.less_batch_normalization
 1.74% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.77% :     0.000003s :     21: predicate.load_eliminater
 0.54% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.49% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.71% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.31% :     0.000002s :     18: predicate.merge_addn
 1.47% :     0.000003s :     18: predicate.minmaximum_grad
 0.76% :     0.000001s :      1: predicate.mutable_eliminate
 0.19% :     0.000000s :      1: predicate.opt_reshape
 2.19% :     0.000004s :     21: predicate.partial_eliminate
 1.41% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.94% :     0.000003s :     18: predicate.reduce_eli
Time group info:
------[substitution.]   0.000224    50
 0.40% :     0.000001s :      1: substitution.elim_not_effective
 0.29% :     0.000001s :      1: substitution.fold_const_symbol
 0.93% :     0.000002s :      1: substitution.graph_param_transform
68.37% :     0.000153s :      6: substitution.inline
 7.15% :     0.000016s :      2: substitution.inline_without_move
 2.64% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.22% :     0.000003s :      2: substitution.minmaximum_grad
 1.33% :     0.000003s :      3: substitution.partial_eliminate
 3.13% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.14% :     0.000005s :      2: substitution.replace_applicator
 1.83% :     0.000004s :      5: substitution.replace_old_param
 1.02% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.62% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.57% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 5.35% :     0.000012s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010230     2
89.01% :     0.009105s :      1: type_inference.infer
10.99% :     0.001124s :      1: type_inference.specialize
------[replace.]   0.000063     8
76.64% :     0.000048s :      6: replace.inline
23.36% :     0.000015s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000157     8
95.28% :     0.000150s :      6: match.inline
 4.72% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000171  1109
 1.65% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.60% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.31% :     0.000002s :     18: predicate.addn_check_dump
 1.59% :     0.000003s :     18: predicate.addn_zero_filter
 1.99% :     0.000003s :     18: predicate.arithmetic_simplify
 1.44% :     0.000002s :     18: predicate.cast_eliminate
 0.20% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.28% :     0.000002s :     18: predicate.compare_switch_simplify
 1.65% :     0.000003s :     18: predicate.depend_value_elim
 1.30% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.49% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.48% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.16% :     0.000000s :      1: predicate.elim_not_effective
 0.33% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.46% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.52% :     0.000003s :     18: predicate.environ_get_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.08% :     0.000000s :      1: predicate.fold_const_symbol
 1.21% :     0.000002s :     12: predicate.get_grad_eliminate
 0.12% :     0.000000s :      1: predicate.graph_param_transform
 4.19% :     0.000007s :     28: predicate.inline
 3.19% :     0.000005s :     33: predicate.inline_without_move
 0.50% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.20% :     0.000002s :     12: predicate.less_batch_normalization
 1.73% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.79% :     0.000003s :     21: predicate.load_eliminater
 0.72% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.41% :     0.000006s :     44: predicate.loop_unroll_before_grad
 2.23% :     0.000004s :     19: predicate.make_slice_get_slice_eliminator
 1.32% :     0.000002s :     18: predicate.merge_addn
 1.31% :     0.000002s :     18: predicate.minmaximum_grad
 0.82% :     0.000001s :      1: predicate.mutable_eliminate
 0.18% :     0.000000s :      1: predicate.opt_reshape
 1.97% :     0.000003s :     21: predicate.partial_eliminate
 1.29% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.90% :     0.000003s :     18: predicate.reduce_eliminate
 1.70% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.48% :     0.000003s :     22: predicate.replace_applicator
 1.58% :     0.000003s :     33: predicate.replace_old_param
 0.16% :     0.000000s :      1: predicate.reset_defer_inline
 1.60% :     0.000003s :     18: predicate.reshape_eliminate
 1.45% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.35% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.67% :     0.000003s :     18: predicate.same_eliminate
 0.65% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.31% :     0.000001s :      2: predicate.special_op_eliminate
 1.24% :     0.000002s :     12: predicate.specialize_transform
 1.64% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.61% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.82% :     0.000005s :     26: predicate.switch_defer_inline
 2.31% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.78% :     0.000012s :     71: predicate.switch_simplify
 1.44% :     0.000002s :     18: predicate.tile_eliminate
 1.35% :     0.000002s :     18: predicate.transpose_eliminate
 1.61% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.52% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.00% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.68% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000003s :     18: predicate.value_based_eliminate
 0.13% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.30% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001298    22
74.50% :     0.000967s :     13: func_graph_cloner_run.FuncGraphClonerGraph
25.50% :     0.000331s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022877    72
 0.15% :     0.000034s :      1: add_recomputation
 0.42% :     0.000096s :      1: auto_monad
 0.07% :     0.000016s :      1: auto_monad_reorder
 0.80% :     0.000182s :      1: bootstrap
 0.10% :     0.000024s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.20% :     0.000046s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
37.03% :     0.008472s :      1: jit_opt_a
 0.50% :     0.000115s :      1: jit_opt_after_cconv
 0.19% :     0.000043s :      1: jit_opt_b
 1.81% :     0.000415s :      1: loop_unroll
 2.32% :     0.000530s :      1: mutable_eliminate
 5.80% :     0.001327s :     26: opt.transform.jit_opt_a
 0.12% :     0.000029s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000010s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000014s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000021s :      4: opt.transform.symbol_engine_opt
 1.94% :     0.000444s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000027s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 5.11% :     0.001170s :    minate
 1.67% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.51% :     0.000003s :     22: predicate.replace_applicator
 1.67% :     0.000003s :     33: predicate.replace_old_param
 0.12% :     0.000000s :      1: predicate.reset_defer_inline
 1.56% :     0.000003s :     18: predicate.reshape_eliminate
 1.50% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.38% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.44% :     0.000002s :     18: predicate.same_eliminate
 0.57% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.32% :     0.000001s :      2: predicate.special_op_eliminate
 1.15% :     0.000002s :     12: predicate.specialize_transform
 1.72% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.55% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.78% :     0.000005s :     26: predicate.switch_defer_inline
 2.35% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.29% :     0.000011s :     71: predicate.switch_simplify
 1.47% :     0.000003s :     18: predicate.tile_eliminate
 1.36% :     0.000002s :     18: predicate.transpose_eliminate
 1.65% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.54% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.69% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.75% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.44% :     0.000006s :     33: predicate.updatestate_useless_node_eliminater
 2.07% :     0.000004s :     18: predicate.value_based_eliminate
 0.13% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.23% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001279    22
73.19% :     0.000936s :     13: func_graph_cloner_run.FuncGraphClonerGraph
26.81% :     0.000343s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024021    76
 0.10% :     0.000025s :      1: add_recomputation
 0.29% :     0.000069s :      1: auto_monad
 0.04% :     0.000010s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.93% :     0.000223s :      1: bootstrap
 0.07% :     0.000017s :      1: cconv
 0.03% :     0.000007s :      1: convert_after_rewriter
 0.09% :     0.000021s :      1: cse_after_recomputation
 0.02% :     0.000006s :      1: environ_conv
 0.22% :     0.000052s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000003s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
34.51% :     0.008288s :      1: jit_opt_a
 0.49% :     0.000117s :      1: jit_opt_after_cconv
 0.17% :     0.000042s :      1: jit_opt_b
 1.77% :     0.000425s :      1: loop_unroll
 2.19% :     0.000526s :      1: mutable_eliminate
 5.09% :     0.001222s :     26: opt.transform.jit_opt_a
 0.12% :     0.000028s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000015s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000021s :      4: opt.transform.symbol_engine_opt
 1.86% :     0.000447s :      1: opt_after_jit_grad
 0.02% :     0.000006s :      1: order_py_execute_after_rewriter
 0.01% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.11% :     0.000026s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_aft  1: renormalize.infer
 1.59% :     0.000363s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000038s :      1: rewriter_after_opt_a
 0.40% :     0.000091s :      1: rewriter_before_opt_a
 0.28% :     0.000064s :      1: symbol_engine_optimizer
40.09% :     0.009172s :      1: type_inference

er_opt_a
 0.04% :     0.000010s :      1: remove_dup_value
 5.70% :     0.001368s :      1: renormalize.infer
 1.77% :     0.000426s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000029s :      1: rewriter_after_opt_a
 0.36% :     0.000086s :      1: rewriter_before_opt_a
 0.27% :     0.000065s :      1: symbol_engine_optimizer
 0.10% :     0.000024s :      1: task_emit
42.89% :     0.010301s :      1: type_inference
 0.14% :     0.000034s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.667.719 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c368dc0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.668.323 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.668.822 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd84048820> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.669.343 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0134542, [30]
[bootstrap]: 0.00035655
[type_inference]: 0.00478978
[event_method]: 1.951e-05
[auto_monad]: 6.656e-05
[graph_reusing]: 6.21e-06
[pre_auto_parallel]: 2.56998e-06
[py_interpret_to_execute]: 3.096e-05
[rewriter_before_opt_a]: 7.698e-05
[expand_dump_flag]: 3.63999e-06
[jit_opt_a]: 0.0057414, [2]
    [Cycle 1]: 0.00159558, [27]
        [switch_simplify]: 5.671e-05
        [loop_unroll]: 2.755e-05
        [a_1]: 0.00052105
        [with_stream_mark]: 1.992e-05
        [recompute_prepare]: 7.18998e-06
        [updatestate_depend_eliminate]: 3.85998e-06
        [updatestate_assign_eliminate]: 3.26999e-06
        [updatestate_loads_eliminate]: 2.88998e-06
        [parameter_eliminate]: 2.09999e-06
        [specialize_transform]: 5.63002e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 5.51002e-06
        [meta_shard_fg_expand]: 2.01e-06
        [get_grad_eliminate_]: 4.85999e-06
        [merge_forward]: 4.09002e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.424e-05
        [j_node_and_user_rematch]: 9.31e-06
        [meta_fg_expand]: 2.24001e-06
        [replace_old_param]: 7.9e-06
        [inline_without_move]: 5.45001e-06
        [renormalize]: 0.00063216
        [add_forward_monad_depend]: 5.92001e-06
        [auto_monad_grad]: 2.83998e-06
        [auto_monad_eliminator]: 1.524e-05
        [cse]: 2.964e-05
        [replace_applicator]: 1.188e-05
    [Cycle 2]: 0.00031492, [27]
        [switch_simplify]: 5.40001e-06
        [loop_unroll]: 4.72998e-06
        [a_1]: 8.687e-05
        [with_stream_mark]: 9.74e-06
        [recompute_prepare]: 5.18002e-06
        [updatestate_depend_eliminate]: 2.62001e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.86998e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 4.90001e-06
        [updatestate_useless_node_eliminater]: 4.63001e-06
        [accelerated_algorithm]: 5.31002e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 4.35999e-06
        [merge_forward]: 3.08e-06
        [cell_reuse_recompute_pass]: 1.44998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.215e-05
        [j_node_and_user_rematch]: 7.5e-06
        [meta_fg_expand]: 1.62001e-06
        [replace_old_param]: 7.18e-06
        [inline_without_move]: 4.58999e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.55001e-06
        [auto_monad_grad]: 6.79982e-07
        [auto_monad_eliminator]: 5.62001e-06
        [cse]: 1.113e-05
        [replace_applicator]: 5.40999e-06
[py_interpret_to_execute_after_opt_a]: 1.147e-05
[rewriter_after_opt_a]: 3.591e-05
[convert_after_rewriter]: 7.54002e-06
[order_py_execute_after_rewriter]: 4.98001e-06
[mutable_eliminate]: 0.00064776
[jit_opt_b]: 5.251e-05, [1]
    [Cycle 1]: 4.435e-05, [2]
        [frontend_op_eliminate]: 1.739e-05
        [inline_after_opt_a]: 1.483e-05
[cconv]: 2.577e-05
[loop_unroll]: 0.00044843
[jit_opt_after_cconv]: 0.00014728, [1]
    [Cycle 1]: 0.00014072, [11]
        [c_1]: 2.013e-05
        [parameter_eliminate]: 2.36e-06
        [updatestate_depend_eliminate]: 6.23e-06
        [updatestate_assign_eliminate]: 3.26001e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [cse]: 2.321e-05
        [call_graph_tuple_transform]: 1.954e-05
        [tuple_list_get_item_eliminator]: 5.47001e-06
        [none_parameter_eliminate]: 1.64998e-06
        [renormalize]: 8.09989e-07
        [switch_simplify]: 4.97e-06
[remove_dup_value]: 1.526e-05
[partial_unused_args_eliminate]: 2.13002e-06
[environ_conv]: 5.91998e-06
[add_recomputation]: 0.0001219
[cse_after_recomputation]: 2.119e-05, [1]
    [Cycle 1]: 1.551e-05, [1]
        [cse]: 9.18002e-06
[auto_monad_reorder]: 1.59e-05
[get_jit_bprop_graph]: 2.32999e-06
[rewriter_after_jit_bprop_graph]: 6.36e-06
[opt_after_jit_grad]: 0.00048135
[symbol_engine_optimizer]: 7.966e-05, [1]
    [Cycle 1]: 7.219e-05, [6]
        [build]: 4.37e-06
        [elim_shapecalc]: 7.70998e-06
        [elim_not_effective]: 1.392e-05
        [opt_reshape]: 6.23998e-06
        [fold_const_symbol]: 9.46e-06
        [renormalize]: 4.30009e-07
[validate]: 3.803e-05
Sums
    bootstrap                            :     0.000357s :  3.96%
    type_inference                       :     0.004790s : 53.17%
    event_method                         :     0.000020s :  0.22%
    auto_monad                           :     0.000067s :  0.74%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000031s :  0.34%
    rewriter_before_opt_a                :     0.000077s :  0.85%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000062s :  0.69%
    jit_opt_a.loop_unroll                :     0.000032s :  0.36%
    jit_opt_a.a_1                        :     0.000608s :  6.75%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000015s :  0.17%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000632s :  7.02%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.23%
    jit_opt_a.cse                        :     0.000041s :  0.45%
    jit_opt_a.replace_applicator         :     0.000017s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.13%
    rewriter_after_opt_a                 :     0.000036s :  0.40%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000648s :  7.19%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000026s :  0.29%
    loop_unroll                          :     0.000448s :  4.98%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000023s :  0.26%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000122s :  1.35%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000481s :  5.34%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.42%


Time group info:
------[substitution.]   0.000200    23
 1.17% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000002s :      2: substitution.fold_const_symbol
 3.09% :     0.000006s :      3: substitution.graph_param_transform
82.67% :     0.000165s :      4: substitution.inline
 1.82% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.32% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.81% :     0.000004s :      2: substitution.replace_old_param
 6.37% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004710     2
81.59% :     0.003843s :      1: type_inference.infer
18.41% :     0.000867s :      1: type_inference.specialize
------[replace.]   0.000066     6
69.18% :     0.000046s :      4: replace.inline
30.82% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000174     6
93.39% :     0.000163s :      4: match.inline
 6.61% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.61% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.37% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.93% :     0.000002s :     10: predicate.addn_zero_filter
 1.93% :     0.000002s :     10: predicate.arithmetic_simplify
 1.14% :     0.000001s :     10: predicate.cast_eliminate
 0.60% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.13% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.50% :     0.000001s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.62% :     0.000007s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :      6: predicate.less_batch_normalization
 1.37% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.29% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.67% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.29% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.69% :     0.000002s :      3: predicate.mutable_eliminate
 0.60% :     0.000001s :      3: predicate.opt_reshape
 2.35% :     0.000003s :     15: predicate.partial_eliminate
 1.44% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.43% :     0.000002s :     10: predicate.reduce_eliminate
 1.35% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.77% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.60% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.36% :     0.000002s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.50% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.40% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.22% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.23% :     0.000009s :     53: predicate.switch_simplify
 1.06% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.28% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.06% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.19% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.92% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000566    12
51.58% :     0.000292s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.95% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.47% :     0.000246s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014948    72
 0.84% :     0.000125s :      1: add_recomputation
 0.47% :     0.000071s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 2.54% :     0.000379s :      1: bootstrap
 0.19% :     0.000028s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
38.43% :     0.005745s :      1: jit_opt_a
 1.00% :     0.000150s :      1: jit_opt_after_cconv
 0.37% :     0.000055s :      1: jit_opt_b
 3.06% :     0.000457s :      1: loop_unroll
 4.39% :     0.000657s :      1: mutable_eliminate
 5.44% :     0.000813s :     26: opt.transform.jit_opt_a
 0.31% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.27% :     0.000489s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.22% :     0.000033s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.18% :     0.000327s :      1: renormalize.infer
 1.99% :     0.000298s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000039s :      1: rewriter_after_opt_a
 0.54% :     0.000080s :      1: rewriter_before_opt_a
 0.55% :     0.000082s :      1: symbol_engine_optimizer
32.16% :     0.004808s :      1: type_inference


TotalTime = 0.0172118, [33]
[bootstrap]: 0.00023621
[type_inference]: 0.00395569
[event_method]: 1.65e-05
[auto_monad]: 6.255e-05
[graph_reusing]: 5.95002e-06
[pre_auto_parallel]: 2.29001e-06
[py_interpret_to_execute]: 3.203e-05
[rewriter_before_opt_a]: 7.004e-05
[expand_dump_flag]: 3.09999e-06
[jit_opt_a]: 0.00479985, [2]
    [Cycle 1]: 0.00142599, [27]
        [switch_simplify]: 4.769e-05
        [loop_unroll]: 3.074e-05
        [a_1]: 0.0005071
        [with_stream_mark]: 1.416e-05
        [recompute_prepare]: 7.33999e-06
        [updatestate_depend_eliminate]: 3.28e-06
        [updatestate_assign_eliminate]: 3.13e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.88002e-06
        [specialize_transform]: 5.66003e-06
        [updatestate_useless_node_eliminater]: 5.08002e-06
        [accelerated_algorithm]: 5.34998e-06
        [meta_shard_fg_expand]: 2.24999e-06
        [get_grad_eliminate_]: 5.22999e-06
        [merge_forward]: 3.54002e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.37e-05
        [j_node_and_user_rematch]: 9.31e-06
        [meta_fg_expand]: 2.53003e-06
        [replace_old_param]: 9.12001e-06
        [inline_without_move]: 5.12999e-06
        [renormalize]: 0.00051984
        [add_forward_monad_depend]: 5.27001e-06
        [auto_monad_grad]: 2.64001e-06
        [auto_monad_eliminator]: 1.307e-05
        [cse]: 2.3e-05
        [replace_applicator]: 1.166e-05
    [Cycle 2]: 0.00033705, [27]
        [switch_simplify]: 5.94999e-06
        [loop_unroll]: 4.96002e-06
        [a_1]: 9.434e-05
        [with_stream_mark]: 1.007e-05
        [recompute_prepare]: 5.34998e-06
        [updatestate_depend_eliminate]: 2.84001e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 4.87998e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 5.19e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 4.72e-06
        [merge_forward]: 2.76e-06
        [cell_reuse_recompute_pass]: 1.86e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.296e-05
        [j_node_and_user_rematch]: 8.40001e-06
        [meta_fg_expand]: 1.84e-06
        [replace_old_param]: 7.39002e-06
        [inline_without_move]: 5.08002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.60999e-06
        [auto_monad_grad]: 1.20001e-06
        [auto_monad_eliminator]: 5.77999e-06
        [cse]: 1.111e-05
        [replace_applicator]: 5.22999e-06
[py_interpret_to_execute_after_opt_a]: 9.09e-06
[rewriter_after_opt_a]: 3.23e-05
[convert_after_rewriter]: 6.61999e-06
[order_py_execute_after_rewriter]: 4.87e-06
[mutable_eliminate]: 0.00059592
[jit_opt_b]: 5.391e-05, [1]
    [Cycle 1]: 4.599e-05, [2]
        [frontend_op_eliminate]: 1.69e-05
        [inline_after_opt_a]: 1.711e-05
[cconv]: 2.61e-05
[loop_unroll]: 0.00045324
[jit_opt_after_cconv]: 0.00014709, [1]
    [Cycle 1]: 0.00013983, [11]
        [c_1]: 2.171e-05
        [parameter_eliminate]: 2.79001e-06
        [updatestate_depend_eliminate]: 5.48997e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 2.29001e-06
        [cse]: 2.131e-05
        [call_graph_tuple_transform]: 1.902e-05
        [tuple_list_get_item_eliminator]: 5.40999e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 6.29982e-07
        [switch_simplify]: 5.43002e-06
[remove_dup_value]: 1.273e-05
[partial_unused_args_eliminate]: 2.31e-06
[environ_conv]: 4.60999e-06
[add_recomputation]: 4.496e-05
[cse_after_recomputation]: 2.054e-05, [1]
    [Cycle 1]: 1.536e-05, [1]
        [cse]: 9.33002e-06
[auto_monad_reorder]: 1.516e-05
[get_jit_bprop_graph]: 1.45001e-06
[rewriter_after_jit_bprop_graph]: 5.15001e-06
[opt_after_jit_grad]: 0.0004479
[symbol_engine_optimizer]: 7.381e-05, [1]
    [Cycle 1]: 6.739e-05, [6]
        [build]: 3.11001e-06
        [elim_shapecalc]: 8.22998e-06
        [elim_not_effective]: 1.225e-05
        [opt_reshape]: 6.12999e-06
        [fold_const_symbol]: 9.17999e-06
        [renormalize]: 4.89992e-07
[validate]: 3.105e-05
[backend_pass]: 7.7e-07
[task_emit]: 0.00585711
[execute]: 7.58999e-06
Sums
    bootstrap                            :     0.000236s :  1.74%
    type_inference                       :     0.003956s : 29.11%
    event_method                         :     0.000017s :  0.12%
    auto_monad                           :     0.000063s :  0.46%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.24%
    rewriter_before_opt_a                :     0.000070s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000054s :  0.39%
    jit_opt_a.loop_unroll                :     0.000036s :  0.26%
    jit_opt_a.a_1                        :     0.000601s :  4.43%
    jit_opt_a.with_stream_mark           :     0.000024s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.07%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000006s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.13%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000017s :  0.12%
    jit_opt_a.inline_without_move        :     0.000010s :  0.08%
    jit_opt_a.renormalize                :     0.000520s :  3.83%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.14%
    jit_opt_a.cse                        :     0.000034s :  0.25%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.24%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000596s :  4.38%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.13%
    cconv                                :     0.000026s :  0.19%
    loop_unroll                          :     0.000453s :  3.33%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.16%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.16%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.14%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000045s :  0.33%
    cse_after_recomputation.cse          :     0.000009s :  0.07%
    auto_monad_reorder                   :     0.000015s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.04%
    opt_after_jit_grad                   :     0.000448s :  3.30%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.06%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.09%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.07%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.23%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.005857s : 43.10%
    execute                              :     0.000008s :  0.06%


Time group info:
------[substitution.]   0.000183    23
 1.03% :     0.000002s :      2: substitution.elim_not_effective
 0.83% :     0.000002s :      2: substitution.fold_const_symbol
 3.10% :     0.000006s :      3: substitution.graph_param_transform
81.26% :     0.000149s :      4: substitution.inline
 1.83% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.35% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.46% :     0.000005s :      2: substitution.replace_old_param
 7.13% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003901     2
80.58% :     0.003144s :      1: type_inference.infer
19.42% :     0.000758s :      1: type_inference.specialize
------[replace.]   0.000064     6
66.60% :     0.000043s :      4: replace.inline
33.40% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000158     6
92.47% :     0.000146s :      4: match.inline
 7.53% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 0.95% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 1.88% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000001s :     10: predicate.depend_value_elim
 1.05% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.18% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.82% :     0.000007s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.24% :     0.000001s :      6: predicate.less_batch_normalization
 1.89% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.50% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.22% :     0.000001s :     10: predicate.minmaximum_grad
 1.66% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.84% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.91% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.56% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.28% :     0.000002s :      6: predicate.special_op_eliminate
 1.09% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.28% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.88% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.41% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.61% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.13% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000498    12
47.57% :     0.000237s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.60% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.83% :     0.000233s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018649    76
 0.26% :     0.000048s :      1: add_recomputation
 0.35% :     0.000066s :      1: auto_monad
 0.09% :     0.000018s :      1: auto_monad_reorder
 0.02% :     0.000003s :      1: backend_pass
 1.31% :     0.000244s :      1: bootstrap
 0.16% :     0.000029s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.12% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000007s :      1: environ_conv
 0.11% :     0.000020s :      1: event_method
 0.06% :     0.000011s :      1: execute
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
25.75% :     0.004803s :      1: jit_opt_a
 0.80% :     0.000150s :      1: jit_opt_after_cconv
 0.30% :     0.000057s :      1: jit_opt_b
 2.47% :     0.000461s :      1: loop_unroll
 3.26% :     0.000608s :      1: mutable_eliminate
 4.32% :     0.000805s :     26: opt.transform.jit_opt_a
 0.26% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.14% :     0.000027s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.11% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.17% :     0.000032s :      4: opt.transform.symbol_engine_opt
 2.44% :     0.000455s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.19% :     0.000035s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.40% :     0.000261s :      1: renormalize.infer
 1.35% :     0.000251s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.19% :     0.000035s :      1: rewriter_after_opt_a
 0.39% :     0.000073s :      1: rewriter_before_opt_a
 0.41% :     0.000076s :      1: symbol_engine_optimizer
31.46% :     0.005868s :      1: task_emit
21.27% :     0.003966s :      1: type_inference
 0.26% :     0.000048s :      1: validate


TotalTime = 0.0307192, [33]
[bootstrap]: 0.00019406
[type_inference]: 0.0092717
[event_method]: 4.242e-05
[auto_monad]: 7.207e-05
[graph_reusing]: 4.55001e-06
[pre_auto_parallel]: 1.96e-06
[py_interpret_to_execute]: 2.095e-05
[rewriter_before_opt_a]: 8.008e-05
[expand_dump_flag]: 2.09e-06
[jit_opt_a]: 0.011161, [3]
    [Cycle 1]: 0.0057064, [27]
        [switch_simplify]: 5.719e-05
        [loop_unroll]: 3.783e-05
        [a_1]: 0.00086387
        [with_stream_mark]: 2.41e-05
        [recompute_prepare]: 1.913e-05
        [updatestate_depend_eliminate]: 7.68999e-06
        [updatestate_assign_eliminate]: 5.49e-06
        [updatestate_loads_eliminate]: 5.29e-06
        [parameter_eliminate]: 2.09999e-06
        [specialize_transform]: 1.436e-05
        [updatestate_useless_node_eliminater]: 1.257e-05
        [accelerated_algorithm]: 1.3e-05
        [meta_shard_fg_expand]: 2.68e-06
        [get_grad_eliminate_]: 1.311e-05
        [merge_forward]: 7.83001e-06
        [cell_reuse_recompute_pass]: 8.79983e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.469e-05
        [j_node_and_user_rematch]: 2.47e-05
        [meta_fg_expand]: 0.00143811
        [replace_old_param]: 4.867e-05
        [inline_without_move]: 4.912e-05
        [renormalize]: 0.00255123
        [add_forward_monad_depend]: 9.36998e-06
        [auto_monad_grad]: 5.49e-06
        [auto_monad_eliminator]: 5.166e-05
        [cse]: 0.00014367
        [replace_applicator]: 6.868e-05
    [Cycle 2]: 0.00204706, [27]
        [switch_simplify]: 3.7e-05
        [loop_unroll]: 3.637e-05
        [a_1]: 0.00110794
        [with_stream_mark]: 1.441e-05
        [recompute_prepare]: 8.67e-06
        [updatestate_depend_eliminate]: 4.36002e-06
        [updatestate_assign_eliminate]: 3.65e-06
        [updatestate_loads_eliminate]: 3.22002e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 7.26001e-06
        [updatestate_useless_node_eliminater]: 6.94999e-06
        [accelerated_algorithm]: 7.36001e-06
        [meta_shard_fg_expand]: 2.27001e-06
        [get_grad_eliminate_]: 6.76999e-06
        [merge_forward]: 3.9e-06
        [cell_reuse_recompute_pass]: 9.89996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.619e-05
        [j_node_and_user_rematch]: 1.253e-05
        [meta_fg_expand]: 4.135e-05
        [replace_old_param]: 1.068e-05
        [inline_without_move]: 6.96001e-06
        [renormalize]: 0.00050779
        [add_forward_monad_depend]: 4.84998e-06
        [auto_monad_grad]: 1.40001e-06
        [auto_monad_eliminator]: 1.335e-05
        [cse]: 2.235e-05
        [replace_applicator]: 1.441e-05
    [Cycle 3]: 0.00040429, [27]
        [switch_simplify]: 7.66999e-06
        [loop_unroll]: 6.66999e-06
        [a_1]: 0.00014226
        [with_stream_mark]: 1.014e-05
        [recompute_prepare]: 7.17002e-06
        [updatestate_depend_eliminate]: 3.6e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 3.01999e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 6.89999e-06
        [updatestate_useless_node_eliminater]: 6.69999e-06
        [accelerated_algorithm]: 7.21999e-06
        [meta_shard_fg_expand]: 1.58002e-06
        [get_grad_eliminate_]: 6.63998e-06
        [merge_forward]: 3.3e-06
        [cell_reuse_recompute_pass]: 1.14998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.504e-05
        [j_node_and_user_rematch]: 1.216e-05
        [meta_fg_expand]: 2.74001e-06
        [replace_old_param]: 9.10999e-06
        [inline_without_move]: 6.89999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.50001e-06
        [auto_monad_grad]: 6.99976e-07
        [auto_monad_eliminator]: 7.4e-06
        [cse]: 1.516e-05
        [replace_applicator]: 6.72002e-06
[py_interpret_to_execute_after_opt_a]: 1.092e-05
[rewriter_after_opt_a]: 4.271e-05
[convert_after_rewriter]: 7.40998e-06
[order_py_execute_after_rewriter]: 5.74e-06
[mutable_eliminate]: 0.00071935
[jit_opt_b]: 5.869e-05, [1]
    [Cycle 1]: 5.222e-05, [2]
        [frontend_op_eliminate]: 2.192e-05
        [inline_after_opt_a]: 1.954e-05
[cconv]: 2.237e-05
[loop_unroll]: 0.00042558
[jit_opt_after_cconv]: 0.00016239, [1]
    [Cycle 1]: 0.00015607, [11]
        [c_1]: 2.904e-05
        [parameter_eliminate]: 1.89999e-06
        [updatestate_depend_eliminate]: 6.09001e-06
        [updatestate_assign_eliminate]: 3.21999e-06
        [updatestate_loads_eliminate]: 2.98e-06
        [cse]: 2.592e-05
        [call_graph_tuple_transform]: 2.215e-05
        [tuple_list_get_item_eliminator]: 7.3e-06
        [none_parameter_eliminate]: 1.07e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 7.56001e-06
[remove_dup_value]: 1.716e-05
[partial_unused_args_eliminate]: 2.11e-06
[environ_conv]: 5.52001e-06
[add_recomputation]: 4.943e-05
[cse_after_recomputation]: 2.864e-05, [1]
    [Cycle 1]: 2.287e-05, [1]
        [cse]: 1.709e-05
[auto_monad_reorder]: 1.773e-05
[get_jit_bprop_graph]: 2.45002e-06
[rewriter_after_jit_bprop_graph]: 5.62001e-06
[opt_after_jit_grad]: 0.00045132
[symbol_engine_optimizer]: 8.275e-05, [1]
    [Cycle 1]: 7.661e-05, [6]
        [build]: 2.99999e-06
        [elim_shapecalc]: 9.15999e-06
        [elim_not_effective]: 1.594e-05
        [opt_reshape]: 8.00999e-06
        [fold_const_symbol]: 1.272e-05
        [renormalize]: 5.19998e-07
[validate]: 3.813e-05
[backend_pass]: 7.60017e-07
[task_emit]: 0.00751985
[execute]: 6.64999e-06
Sums
    bootstrap                            :     0.000194s :  0.72%
    type_inference                       :     0.009272s : 34.39%
    event_method                         :     0.000042s :  0.16%
    auto_monad                           :     0.000072s :  0.27%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000021s :  0.08%
    rewriter_before_opt_a                :     0.000080s :  0.30%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000102s :  0.38%
    jit_opt_a.loop_unroll                :     0.000081s :  0.30%
    jit_opt_a.a_1                        :     0.002114s :  7.84%
    jit_opt_a.with_stream_mark           :     0.000049s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000035s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000029s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000028s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000027s :  0.10%
    jit_opt_a.merge_forward              :     0.000015s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000056s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000049s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001482s :  5.50%
    jit_opt_a.replace_old_param          :     0.000068s :  0.25%
    jit_opt_a.inline_without_move        :     0.000063s :  0.23%
    jit_opt_a.renormalize                :     0.003059s : 11.35%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000072s :  0.27%
    jit_opt_a.cse                        :     0.000181s :  0.67%
    jit_opt_a.replace_applicator         :     0.000090s :  0.33%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.04%
    rewriter_after_opt_a                 :     0.000043s :  0.16%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000719s :  2.67%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.07%
    cconv                                :     0.000022s :  0.08%
    loop_unroll                          :     0.000426s :  1.58%
    jit_opt_after_cconv.c_1              :     0.000029s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000008s :  0.03%
    remove_dup_value                     :     0.000017s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000049s :  0.18%
    cse_after_recomputation.cse          :     0.000017s :  0.06%
    auto_monad_reorder                   :     0.000018s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000451s :  1.67%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007520s : 27.89%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000491   118
 5.21% :     0.000026s :      1: substitution.arithmetic_simplify
 0.56% :     0.000003s :      4: substitution.elim_not_effective
 0.38% :     0.000002s :      4: substitution.fold_const_symbol
 1.29% :     0.000006s :      5: substitution.graph_param_transform
59.21% :     0.000291s :     12: substitution.inline
 3.15% :     0.000015s :      2: substitution.inline_without_move
 2.17% :     0.000011s :     16: substitution.j_node_and_user_rematch
 2.16% :     0.000011s :      7: substitution.minmaximum_grad
 1.38% :     0.000007s :      5: substitution.partial_eliminate
 2.28% :     0.000011s :     16: substitution.remove_not_recompute_node
 5.43% :     0.000027s :     10: substitution.replace_applicator
 1.37% :     0.000007s :      7: substitution.replace_old_param
 0.59% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.22% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.84% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.76% :     0.000038s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009214     2
88.93% :     0.008194s :      1: type_inference.infer
11.07% :     0.001020s :      1: type_inference.specialize
------[replace.]   0.000144    20
 3.91% :     0.000006s :      1: replace.arithmetic_simplify
61.55% :     0.000088s :     12: replace.inline
34.54% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000326    20
 7.57% :     0.000025s :      1: match.arithmetic_simplify
86.96% :     0.000283s :     12: match.inline
 5.47% :     0.000018s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000388  2790
 1.60% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.43% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.45% :     0.000006s :     46: predicate.addn_check_dump
 1.83% :     0.000007s :     46: predicate.addn_zero_filter
 2.12% :     0.000008s :     47: predicate.arithmetic_simplify
 1.60% :     0.000006s :     47: predicate.cast_eliminate
 0.26% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.44% :     0.000006s :     46: predicate.compare_switch_simplify
 1.51% :     0.000006s :     46: predicate.depend_value_elim
 1.47% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.59% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.60% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.29% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.14% :     0.000001s :      5: predicate.elim_not_effective
 0.23% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.58% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.45% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.54% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.62% :     0.000006s :     47: predicate.environ_get_eliminate
 1.46% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.11% :     0.000000s :      5: predicate.fold_const_symbol
 0.94% :     0.000004s :     21: predicate.get_grad_eliminate
 0.12% :     0.000000s :      5: predicate.graph_param_transform
 4.18% :     0.000016s :     76: predicate.inline
 1.73% :     0.000007s :     42: predicate.inline_without_move
 0.38% :     0.000001s :     21: predicate.j_node_and_user_rematch
 1.13% :     0.000004s :     21: predicate.less_batch_normalization
 1.80% :     0.000007s :     54: predicate.list_to_tuple_eliminator_
 1.95% :     0.000008s :     59: predicate.load_eliminater
 0.49% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.09% :     0.000012s :     89: predicate.loop_unroll_before_grad
 1.74% :     0.000007s :     52: predicate.make_slice_get_slice_eliminator
 1.52% :     0.000006s :     46: predicate.merge_addn
 1.49% :     0.000006s :     47: predicate.minmaximum_grad
 0.50% :     0.000002s :      5: predicate.mutable_eliminate
 0.25% :     0.000001s :      5: predicate.opt_reshape
 2.33% :     0.000009s :     59: predicate.partial_eliminate
 1.46% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.91% :     0.000007s :     47: predicate.reduce_eliminate
 1.79% :     0.000007s :     54: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.52% :     0.000010s :    105: predicate.replace_applicator
 0.86% :     0.000003s :     42: predicate.replace_old_param
 0.13% :     0.000000s :      5: predicate.reset_defer_inline
 1.74% :     0.000007s :     47: predicate.reshape_eliminate
 1.47% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.39% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.53% :     0.000006s :     46: predicate.same_eliminate
 0.47% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.43% :     0.000002s :     10: predicate.special_op_eliminate
 0.94% :     0.000004s :     21: predicate.specialize_transform
 1.66% :     0.000006s :     46: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.21% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.65% :     0.000010s :     66: predicate.switch_defer_inline
 2.47% :     0.000010s :     66: predicate.switch_layer_defer_inline
 6.04% :     0.000023s :    160: predicate.switch_simplify
 1.49% :     0.000006s :     47: predicate.tile_eliminate
 1.58% :     0.000006s :     47: predicate.transpose_eliminate
 1.86% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.74% :     0.000007s :     47: predicate.tuple_list_get_item_depend_reorder
 3.09% :     0.000012s :     64: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000007s :     54: predicate.tuple_to_list_eliminator_
 1.94% :     0.000008s :     59: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000012s :     80: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000007s :     46: predicate.value_based_eliminate
 0.17% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.23% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001406    29
58.10% :     0.000817s :     13: func_graph_cloner_run.FuncGraphClonerGraph
41.90% :     0.000589s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.036644    91
 0.14% :     0.000053s :      1: add_recomputation
 0.21% :     0.000076s :      1: auto_monad
 0.06% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.56% :     0.000205s :      1: bootstrap
 0.07% :     0.000025s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000031s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.13% :     0.000047s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.01% :     0.000004s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
30.47% :     0.011164s :      1: jit_opt_a
 0.45% :     0.000165s :      1: jit_opt_after_cconv
 0.17% :     0.000061s :      1: jit_opt_b
 1.18% :     0.000433s :      1: loop_unroll
 1.99% :     0.000728s :      1: mutable_eliminate
 7.44% :     0.002726s :     39: opt.transform.jit_opt_a
 0.17% :     0.000063s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000035s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000043s :      4: opt.transform.symbol_engine_opt
 1.25% :     0.000458s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.06% :     0.000023s :      1: py_interpret_to_execute
 0.04% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000020s :      1: remove_dup_value
 4.31% :     0.001579s :      2: renormalize.infer
 4.00% :     0.001466s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000046s :      1: rewriter_after_opt_a
 0.23% :     0.000083s :      1: rewriter_before_opt_a
 0.23% :     0.000085s :      1: symbol_engine_optimizer
20.55% :     0.007530s :      1: task_emit
25.33% :     0.009284s :      1: type_inference
 0.16% :     0.000059s :      1: validate


TotalTime = 0.0210116, [33]
[bootstrap]: 0.00020188
[type_inference]: 0.010309
[event_method]: 4.822e-05
[auto_monad]: 6.381e-05
[graph_reusing]: 4.77998e-06
[pre_auto_parallel]: 2.34999e-06
[py_interpret_to_execute]: 2.4e-05
[rewriter_before_opt_a]: 8.428e-05
[expand_dump_flag]: 2.09e-06
[jit_opt_a]: 0.0082909, [2]
    [Cycle 1]: 0.00467096, [27]
        [switch_simplify]: 6.366e-05
        [loop_unroll]: 3.795e-05
        [a_1]: 0.00082045
        [with_stream_mark]: 2.163e-05
        [recompute_prepare]: 1.725e-05
        [updatestate_depend_eliminate]: 7.26999e-06
        [updatestate_assign_eliminate]: 5.32001e-06
        [updatestate_loads_eliminate]: 5.30001e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 1.263e-05
        [updatestate_useless_node_eliminater]: 1.194e-05
        [accelerated_algorithm]: 1.197e-05
        [meta_shard_fg_expand]: 3.16999e-06
        [get_grad_eliminate_]: 1.199e-05
        [merge_forward]: 7.01999e-06
        [cell_reuse_recompute_pass]: 7.09988e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.347e-05
        [j_node_and_user_rematch]: 2.165e-05
        [meta_fg_expand]: 0.00142786
        [replace_old_param]: 5.13e-05
        [inline_without_move]: 9.862e-05
        [renormalize]: 0.00176335
        [add_forward_monad_depend]: 5.03002e-06
        [auto_monad_grad]: 1.27e-06
        [auto_monad_eliminator]: 9.12999e-06
        [cse]: 1.638e-05
        [replace_applicator]: 1.062e-05
    [Cycle 2]: 0.00029096, [27]
        [switch_simplify]: 4.63001e-06
        [loop_unroll]: 4.20999e-06
        [a_1]: 4.284e-05
        [with_stream_mark]: 8.1e-06
        [recompute_prepare]: 3.81001e-06
        [updatestate_depend_eliminate]: 2.16e-06
        [updatestate_assign_eliminate]: 1.71e-06
        [updatestate_loads_eliminate]: 1.37999e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 3.38e-06
        [updatestate_useless_node_eliminater]: 3.41999e-06
        [accelerated_algorithm]: 3.52002e-06
        [meta_shard_fg_expand]: 1.19003e-06
        [get_grad_eliminate_]: 3.52997e-06
        [merge_forward]: 2.27999e-06
        [cell_reuse_recompute_pass]: 1.65001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.024e-05
        [j_node_and_user_rematch]: 6.12001e-06
        [meta_fg_expand]: 3.74e-05
        [replace_old_param]: 3.81999e-06
        [inline_without_move]: 3.30998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.37e-06
        [auto_monad_grad]: 1.04998e-06
        [auto_monad_eliminator]: 4.70999e-06
        [cse]: 1.085e-05
        [replace_applicator]: 3.98999e-06
[py_interpret_to_execute_after_opt_a]: 7.75998e-06
[rewriter_after_opt_a]: 2.459e-05
[convert_after_rewriter]: 4.74998e-06
[order_py_execute_after_rewriter]: 3.53e-06
[mutable_eliminate]: 0.00053853
[jit_opt_b]: 3.975e-05, [1]
    [Cycle 1]: 3.351e-05, [2]
        [frontend_op_eliminate]: 1.086e-05
        [inline_after_opt_a]: 1.139e-05
[cconv]: 1.396e-05
[loop_unroll]: 0.00041529
[jit_opt_after_cconv]: 0.00011364, [1]
    [Cycle 1]: 0.00010676, [11]
        [c_1]: 1.249e-05
        [parameter_eliminate]: 2.12999e-06
        [updatestate_depend_eliminate]: 4.45e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.62001e-06
        [cse]: 1.439e-05
        [call_graph_tuple_transform]: 1.056e-05
        [tuple_list_get_item_eliminator]: 3.91999e-06
        [none_parameter_eliminate]: 8.49977e-07
        [renormalize]: 4.39992e-07
        [switch_simplify]: 3.75e-06
[remove_dup_value]: 7.67002e-06
[partial_unused_args_eliminate]: 1.44e-06
[environ_conv]: 3.83999e-06
[add_recomputation]: 2.107e-05
[cse_after_recomputation]: 1.868e-05, [1]
    [Cycle 1]: 1.373e-05, [1]
        [cse]: 7.78001e-06
[auto_monad_reorder]: 6.94999e-06
[get_jit_bprop_graph]: 1.29e-06
[rewriter_after_jit_bprop_graph]: 3.98001e-06
[opt_after_jit_grad]: 0.00043915
[symbol_engine_optimizer]: 6.294e-05, [1]
    [Cycle 1]: 5.701e-05, [6]
        [build]: 2.53e-06
        [elim_shapecalc]: 6.99001e-06
        [elim_not_effective]: 8.40999e-06
        [opt_reshape]: 4.22e-06
        [fold_const_symbol]: 5.46e-06
        [renormalize]: 5.09986e-07
[validate]: 2.228e-05
[backend_pass]: 8.29983e-07
[task_emit]: 1.964e-05
[execute]: 1.14e-06
Sums
    bootstrap                            :     0.000202s :  1.19%
    type_inference                       :     0.010309s : 60.52%
    event_method                         :     0.000048s :  0.28%
    auto_monad                           :     0.000064s :  0.37%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000024s :  0.14%
    rewriter_before_opt_a                :     0.000084s :  0.49%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000068s :  0.40%
    jit_opt_a.loop_unroll                :     0.000042s :  0.25%
    jit_opt_a.a_1                        :     0.000863s :  5.07%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000016s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000015s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.09%
    jit_opt_a.merge_forward              :     0.000009s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001465s :  8.60%
    jit_opt_a.replace_old_param          :     0.000055s :  0.32%
    jit_opt_a.inline_without_move        :     0.000102s :  0.60%
    jit_opt_a.renormalize                :     0.001763s : 10.35%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.08%
    jit_opt_a.cse                        :     0.000027s :  0.16%
    jit_opt_a.replace_applicator         :     0.000015s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000025s :  0.14%
    convert_after_rewriter               :     0.000005s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000539s :  3.16%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000014s :  0.08%
    loop_unroll                          :     0.000415s :  2.44%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000014s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000011s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000008s :  0.05%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000021s :  0.12%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000007s :  0.04%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.02%
    opt_after_jit_grad                   :     0.000439s :  2.58%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000008s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000005s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000022s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000020s :  0.12%
    execute                              :     0.000001s :  0.01%


TotalTime = 0.0198919, [30]
[bootstrap]: 0.00019869
[type_inference]: 0.00891271
[event_method]: 4.208e-05
[auto_monad]: 9.197e-05
[graph_reusing]: 7.12002e-06
[pre_auto_parallel]: 2.47001e-06
[py_interpret_to_execute]: 2.435e-05
[rewriter_before_opt_a]: 8.73e-05
[expand_dump_flag]: 2.91999e-06
[jit_opt_a]: 0.00849241, [2]
    [Cycle 1]: 0.00479836, [27]
        [switch_simplify]: 5.923e-05
        [loop_unroll]: 3.855e-05
        [a_1]: 0.00096143
        [with_stream_mark]: 2.432e-05
        [recompute_prepare]: 1.796e-05
        [updatestate_depend_eliminate]: 7.51999e-06
        [updatestate_assign_eliminate]: 5.77999e-06
        [updatestate_loads_eliminate]: 5.96e-06
        [parameter_eliminate]: 2.71999e-06
        [specialize_transform]: 3.938e-05
        [updatestate_useless_node_eliminater]: 1.233e-05
        [accelerated_algorithm]: 1.261e-05
        [meta_shard_fg_expand]: 3.63e-06
        [get_grad_eliminate_]: 1.225e-05
        [merge_forward]: 7.58999e-06
        [cell_reuse_recompute_pass]: 9.30013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.551e-05
        [j_node_and_user_rematch]: 2.236e-05
        [meta_fg_expand]: 0.00169113
        [replace_old_param]: 5.235e-05
        [inline_without_move]: 5.279e-05
        [renormalize]: 0.00148843
        [add_forward_monad_depend]: 6.34999e-06
        [auto_monad_grad]: 2.49999e-06
        [auto_monad_eliminator]: 1.162e-05
        [cse]: 2.022e-05
        [replace_applicator]: 1.124e-05
    [Cycle 2]: 0.00027457, [27]
        [switch_simplify]: 4.69998e-06
        [loop_unroll]: 3.73001e-06
        [a_1]: 3.978e-05
        [with_stream_mark]: 8.67e-06
        [recompute_prepare]: 3.51999e-06
        [updatestate_depend_eliminate]: 2.34001e-06
        [updatestate_assign_eliminate]: 1.92001e-06
        [updatestate_loads_eliminate]: 1.63002e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 3.61001e-06
        [updatestate_useless_node_eliminater]: 3.51999e-06
        [accelerated_algorithm]: 3.65e-06
        [meta_shard_fg_expand]: 1.25001e-06
        [get_grad_eliminate_]: 3.21001e-06
        [merge_forward]: 2.02999e-06
        [cell_reuse_recompute_pass]: 1.69998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.206e-05
        [j_node_and_user_rematch]: 6.18002e-06
        [meta_fg_expand]: 2.58e-05
        [replace_old_param]: 3.64002e-06
        [inline_without_move]: 3.16001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 4.22e-06
        [cse]: 1.001e-05
        [replace_applicator]: 3.91001e-06
[py_interpret_to_execute_after_opt_a]: 1.048e-05
[rewriter_after_opt_a]: 3.228e-05
[convert_after_rewriter]: 6.44999e-06
[order_py_execute_after_rewriter]: 4.92e-06
[mutable_eliminate]: 0.00052636
[jit_opt_b]: 4.081e-05, [1]
    [Cycle 1]: 3.38e-05, [2]
        [frontend_op_eliminate]: 1.069e-05
        [inline_after_opt_a]: 1.176e-05
[cconv]: 2.222e-05
[loop_unroll]: 0.00041712
[jit_opt_after_cconv]: 0.00011977, [1]
    [Cycle 1]: 0.0001134, [11]
        [c_1]: 1.249e-05
        [parameter_eliminate]: 2.65002e-06
        [updatestate_depend_eliminate]: 4.83001e-06
        [updatestate_assign_eliminate]: 1.87001e-06
        [updatestate_loads_eliminate]: 1.75001e-06
        [cse]: 1.577e-05
        [call_graph_tuple_transform]: 1.393e-05
        [tuple_list_get_item_eliminator]: 3.99002e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 5.60016e-07
        [switch_simplify]: 3.79002e-06
[remove_dup_value]: 1.321e-05
[partial_unused_args_eliminate]: 2.33998e-06
[environ_conv]: 4.75001e-06
[add_recomputation]: 3.296e-05
[cse_after_recomputation]: 1.956e-05, [1]
    [Cycle 1]: 1.382e-05, [1]
        [cse]: 7.90998e-06
[auto_monad_reorder]: 1.276e-05
[get_jit_bprop_graph]: 1.76e-06
[rewriter_after_jit_bprop_graph]: 5.22e-06
[opt_after_jit_grad]: 0.0004495
[symbol_engine_optimizer]: 0.00013005, [1]
    [Cycle 1]: 0.00012386, [6]
        [build]: 3.51999e-06
        [elim_shapecalc]: 6.26e-06
        [elim_not_effective]: 1.137e-05
        [opt_reshape]: 4.85001e-06
        [fold_const_symbol]: 6.63e-06
        [renormalize]: 4.70027e-07
[validate]: 2.876e-05
Sums
    bootstrap                            :     0.000199s :  1.26%
    type_inference                       :     0.008913s : 56.33%
    event_method                         :     0.000042s :  0.27%
    auto_monad                           :     0.000092s :  0.58%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000024s :  0.15%
    rewriter_before_opt_a                :     0.000087s :  0.55%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.40%
    jit_opt_a.loop_unroll                :     0.000042s :  0.27%
    jit_opt_a.a_1                        :     0.001001s :  6.33%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000043s :  0.27%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.10%
    jit_opt_a.merge_forward              :     0.000010s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001717s : 10.85%
    jit_opt_a.replace_old_param          :     0.000056s :  0.35%
    jit_opt_a.inline_without_move        :     0.000056s :  0.35%
    jit_opt_a.renormalize                :     0.001489s :  9.41%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.10%
    jit_opt_a.cse                        :     0.000030s :  0.19%
    jit_opt_a.replace_applicator         :     0.000015s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000526s :  3.33%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.07%
    cconv                                :     0.000022s :  0.14%
    loop_unroll                          :     0.000417s :  2.64%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000033s :  0.21%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000450s :  2.84%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000007s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.18%


Time group info:
------[substitution.]   0.000228    50
 0.46% :     0.000001s :      1: substitution.elim_not_effective
 0.30% :     0.000001s :      1: substitution.fold_const_symbol
 0.96% :     0.000002s :      1: substitution.graph_param_transform
69.04% :     0.000157s :      6: substitution.inline
 7.38% :     0.000017s :      2: substitution.inline_without_move
 2.38% :     0.000005s :      9: substitution.j_node_and_user_rematch
 1.39% :     0.000003s :      2: substitution.minmaximum_grad
 1.58% :     0.000004s :      3: substitution.partial_eliminate
 2.93% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.10% :     0.000005s :      2: substitution.replace_applicator
 1.79% :     0.000004s :      5: substitution.replace_old_param
 0.76% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.44% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.48% :     0.000003s :      2: substitution.tuple_list_get_item_depend_reorder
 5.01% :     0.000011s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010249     2
89.05% :     0.009126s :      1: type_inference.infer
10.95% :     0.001123s :      1: type_inference.specialize
------[replace.]   0.000062     8
76.41% :     0.000047s :      6: replace.inline
23.59% :     0.000015s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000161     8
95.66% :     0.000154s :      6: match.inline
 4.34% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000166  1109
 1.55% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.60% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.32% :     0.000002s :     18: predicate.addn_check_dump
 1.48% :     0.000002s :     18: predicate.addn_zero_filter
 1.97% :     0.000003s :     18: predicate.arithmetic_simplify
 1.43% :     0.000002s :     18: predicate.cast_eliminate
 0.11% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.39% :     0.000002s :     18: predicate.compare_switch_simplify
 1.63% :     0.000003s :     18: predicate.depend_value_elim
 1.31% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.64% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.41% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.38% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      1: predicate.elim_not_effective
 0.33% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.31% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.45% :     0.000002s :     18: predicate.environ_get_eliminate
 1.30% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.07% :     0.000000s :      1: predicate.fold_const_symbol
 1.26% :     0.000002s :     12: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 4.14% :     0.000007s :     28: predicate.inline
 3.37% :     0.000006s :     33: predicate.inline_without_move
 0.52% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.22% :     0.000002s :     12: predicate.less_batch_normalization
 1.73% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.87% :     0.000003s :     21: predicate.load_eliminater
 0.58% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.63% :     0.000006s :     44: predicate.loop_unroll_before_grad
 2.14% :     0.000004s :     19: predicate.make_slice_get_slice_eliminator
 1.38% :     0.000002s :     18: predicate.merge_addn
 1.40% :     0.000002s :     18: predicate.minmaximum_grad
 0.67% :     0.000001s :      1: predicate.mutable_eliminate
 0.19% :     0.000000s :      1: predicate.opt_reshape
 2.04% :     0.000003s :     21: predicate.partial_eliminate
 1.46% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.84% :     0.000003s :     18: predicate.reduce_eliminate
 1.49% :     0.000002s :     20: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.46% :     0.000002s :     22: predicate.replace_applicator
 1.61% :     0.000003s :     33: predicate.replace_old_param
 0.13% :     0.000000s :      1: predicate.reset_defer_inline
 1.49% :     0.000002s :     18: predicate.reshape_eliminate
 1.80% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.38% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.47% :     0.000002s :     18: predicate.same_eliminate
 0.66% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.58% :     0.000001s :      2: predicate.special_op_eliminate
 1.19% :     0.000002s :     12: predicate.specialize_transform
 1.73% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.58% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.57% :     0.000004s :     26: predicate.switch_defer_inline
 2.30% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.72% :     0.000011s :     71: predicate.switch_simplify
 1.48% :     0.000002s :     18: predicate.tile_eliminate
 1.33% :     0.000002s :     18: predicate.transpose_eliminate
 1.72% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.42% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 2.72% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 2.07% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.61% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000003s :     18: predicate.value_based_eliminate
 0.11% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.23% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001181    22
71.21% :     0.000841s :     13: func_graph_cloner_run.FuncGraphClonerGraph
28.79% :     0.000340s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024095    76
 0.10% :     0.000024s :      1: add_recomputation
 0.28% :     0.000068s :      1: auto_monad
 0.04% :     0.000010s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.88% :     0.000211s :      1: bootstrap
 0.07% :     0.000017s :      1: cconv
 0.03% :     0.000007s :      1: convert_after_rewriter
 0.09% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.22% :     0.000052s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000003s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
34.42% :     0.008294s :      1: jit_opt_a
 0.48% :     0.000116s :      1: jit_opt_after_cconv
 0.18% :     0.000042s :      1: jit_opt_b
 1.75% :     0.000422s :      1: loop_unroll
 2.27% :     0.000546s :      1: mutable_eliminate
 5.24% :     0.001263s :     26: opt.transform.jit_opt_a
 0.11% :     0.000028s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000014s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000021s :      4: opt.transform.symbol_engine_opt
 2.05% :     0.000493s :      1: opt_after_jit_grad
 0.02% :     0.000006s :      1: order_py_execute_after_rewriter
 0.01% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.11% :     0.000026s :      1: py_interpret_to_execute
 0.04% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000010s :      1: remove_dup_value
 5.44% :     0.001310s :      1: renormalize.infer
 1.85% :     0.000445s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000028s :      1: rewriter_after_opt_a
 0.36% :     0.000087s :      1: rewriter_before_opt_a
 0.27% :     0.000066s :      1: symbol_engine_optimizer
 0.10% :     0.000023s :      1: task_emit
42.83% :     0.010320s :      1: type_inference
 0.15% :     0.000036s :      1: validate


Time group info:
------[substitution.]   0.000329    50
 0.45% :     0.000001s :      1: substitution.elim_not_effective
 0.35% :     0.000001s :      1: substitution.fold_const_symbol
 1.26% :     0.000004s :      1: substitution.graph_param_transform
75.02% :     0.000247s :      6: substitution.inline
 4.76% :     0.000016s :      2: substitution.inline_without_move
 1.90% :     0.000006s :      9: substitution.j_node_and_user_rematch
 0.94% :     0.000003s :      2: substitution.minmaximum_grad
 1.18% :     0.000004s :      3: substitution.partial_eliminate
 2.09% :     0.000007s :      9: substitution.remove_not_recompute_node
 1.77% :     0.000006s :      2: substitution.replace_applicator
 1.11% :     0.000004s :      5: substitution.replace_old_param
 0.83% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.32% :     0.000008s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.46% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 4.57% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008852     2
88.90% :     0.007869s :      1: type_inference.infer
11.10% :     0.000983s :      1: type_inference.specialize
------[replace.]   0.000065     8
76.69% :     0.000050s :      6: replace.inline
23.31% :     0.000015s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000254     8
95.89% :     0.000243s :      6: match.inline
 4.11% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000169  1109
 1.39% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.64% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.29% :     0.000002s :     18: predicate.addn_check_dump
 1.51% :     0.000003s :     18: predicate.addn_zero_filter
 2.14% :     0.000004s :     18: predicate.arithmetic_simplify
 1.56% :     0.000003s :     18: predicate.cast_eliminate
 0.20% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.24% :     0.000002s :     18: predicate.compare_switch_simplify
 1.32% :     0.000002s :     18: predicate.depend_value_elim
 1.28% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.53% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.35% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.38% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      1: predicate.elim_not_effective
 0.35% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.24% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.28% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.41% :     0.000002s :     18: predicate.environ_get_eliminate
 1.40% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.24% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.35% :     0.000007s :     28: predicate.inline
 3.37% :     0.000006s :     33: predicate.inline_without_move
 0.50% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :     12: predicate.less_batch_normalization
 1.90% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.67% :     0.000003s :     21: predicate.load_eliminater
 0.88% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.67% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.82% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.26% :     0.000002s :     18: predicate.merge_addn
 1.41% :     0.000002s :     18: predicate.minmaximum_grad
 0.79% :     0.000001s :      1: predicate.mutable_eliminate
 0.38% :     0.000001s :      1: predicate.opt_reshape
 2.08% :     0.000004s :     21: predicate.partial_eliminate
 1.30% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.94% :     0.000003s :     18: predicate.reduce_eliminate
 1.59% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.73% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.43% :     0.000002s :     22: predicate.replace_applicator
 1.67% :     0.000003s :     33: predicate.replace_old_param
 0.08% :     0.000000s :      1: predicate.reset_defer_inline
 1.50% :     0.000003s :     18: predicate.reshape_eliminate
 1.52% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.45% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     18: predicate.same_eliminate
 0.64% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.31% :     0.000001s :      2: predicate.special_op_eliminate
 1.27% :     0.000002s :     12: predicate.specialize_transform
 1.53% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.15% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.84% :     0.000005s :     26: predicate.switch_defer_inline
 2.38% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.86% :     0.000012s :     71: predicate.switch_simplify
 1.40% :     0.000002s :     18: predicate.tile_eliminate
 1.50% :     0.000003s :     18: predicate.transpose_eliminate
 1.76% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.47% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 3.12% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.69% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000003s :     18: predicate.value_based_eliminate
 0.11% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.28% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001142    22
76.04% :     0.000868s :     13: func_graph_cloner_run.FuncGraphClonerGraph
23.96% :     0.000274s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022789    72
 0.16% :     0.000036s :      1: add_recomputation
 0.42% :     0.000096s :      1: auto_monad
 0.07% :     0.000015s :      1: auto_monad_reorder
 0.91% :     0.000208s :      1: bootstrap
 0.11% :     0.000025s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.20% :     0.000046s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
37.28% :     0.008496s :      1: jit_opt_a
 0.54% :     0.000123s :      1: jit_opt_after_cconv
 0.19% :     0.000043s :      1: jit_opt_b
 1.86% :     0.000425s :      1: loop_unroll
 2.35% :     0.000535s :      1: mutable_eliminate
 6.07% :     0.001383s :     26: opt.transform.jit_opt_a
 0.13% :     0.000031s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000014s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000025s :      4: opt.transform.symbol_engine_opt
 2.01% :     0.000457s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000027s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 4.84% :     0.001103s :      1: renormalize.infer
 1.66% :     0.000379s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000036s :      1: rewriter_after_opt_a
 0.39% :     0.000090s :      1: rewriter_before_opt_a
 0.58% :     0.000133s :      1: symbol_engine_optimizer
39.15% :     0.008922s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.694.179 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d1670> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.694.188 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e39d0> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.694.745 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.694.746 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.696.014 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c37f280> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.696.665 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0134408, [30]
[bootstrap]: 0.00027344
[type_inference]: 0.00403174
[event_method]: 1.843e-05
[auto_monad]: 6.917e-05
[graph_reusing]: 6.37001e-06
[pre_auto_parallel]: 3.16999e-06
[py_interpret_to_execute]: 3.026e-05
[rewriter_before_opt_a]: 7.179e-05
[expand_dump_flag]: 3.16001e-06
[jit_opt_a]: 0.00598936, [2]
    [Cycle 1]: 0.00181674, [27]
        [switch_simplify]: 5.946e-05
        [loop_unroll]: 2.862e-05
        [a_1]: 0.00059218
        [with_stream_mark]: 2.346e-05
        [recompute_prepare]: 9.04e-06
        [updatestate_depend_eliminate]: 4.09002e-06
        [updatestate_assign_eliminate]: 3.26999e-06
        [updatestate_loads_eliminate]: 3.13e-06
        [parameter_eliminate]: 2.11998e-06
        [specialize_transform]: 6.45002e-06
        [updatestate_useless_node_eliminater]: 4.74998e-06
        [accelerated_algorithm]: 5.22999e-06
        [meta_shard_fg_expand]: 2.19001e-06
        [get_grad_eliminate_]: 5.59e-06
        [merge_forward]: 4.90001e-06
        [cell_reuse_recompute_pass]: 1.82001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.704e-05
        [j_node_and_user_rematch]: 1.03e-05
        [meta_fg_expand]: 2.59001e-06
        [replace_old_param]: 9.64e-06
        [inline_without_move]: 5.77999e-06
        [renormalize]: 0.0007256
        [add_forward_monad_depend]: 6.69001e-06
        [auto_monad_grad]: 2.71e-06
        [auto_monad_eliminator]: 1.695e-05
        [cse]: 3.503e-05
        [replace_applicator]: 1.621e-05
    [Cycle 2]: 0.00036736, [27]
        [switch_simplify]: 6.89999e-06
        [loop_unroll]: 5.15001e-06
        [a_1]: 9.738e-05
        [with_stream_mark]: 1.548e-05
        [recompute_prepare]: 5.87001e-06
        [updatestate_depend_eliminate]: 3.43999e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 1.35999e-06
        [specialize_transform]: 5.92001e-06
        [updatestate_useless_node_eliminater]: 4.97999e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.59998e-06
        [get_grad_eliminate_]: 5.19e-06
        [merge_forward]: 4.18999e-06
        [cell_reuse_recompute_pass]: 3.16999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.752e-05
        [j_node_and_user_rematch]: 8.61002e-06
        [meta_fg_expand]: 1.94e-06
        [replace_old_param]: 8.98002e-06
        [inline_without_move]: 5.02e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.75001e-06
        [auto_monad_grad]: 1.17e-06
        [auto_monad_eliminator]: 6.78998e-06
        [cse]: 1.199e-05
        [replace_applicator]: 5.19998e-06
[py_interpret_to_execute_after_opt_a]: 1.55e-05
[rewriter_after_opt_a]: 4.13e-05
[convert_after_rewriter]: 7.93001e-06
[order_py_execute_after_rewriter]: 5.54e-06
[mutable_eliminate]: 0.0008484
[jit_opt_b]: 5.774e-05, [1]
    [Cycle 1]: 4.847e-05, [2]
        [frontend_op_eliminate]: 1.851e-05
        [inline_after_opt_a]: 1.678e-05
[cconv]: 3.829e-05
[loop_unroll]: 0.00061947
[jit_opt_after_cconv]: 0.0001758, [1]
    [Cycle 1]: 0.00016705, [11]
        [c_1]: 2.465e-05
        [parameter_eliminate]: 4.08999e-06
        [updatestate_depend_eliminate]: 8.52e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 2.66e-06
        [cse]: 3.443e-05
        [call_graph_tuple_transform]: 2.361e-05
        [tuple_list_get_item_eliminator]: 5.14e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 7.50006e-07
        [switch_simplify]: 5.97999e-06
[remove_dup_value]: 1.629e-05
[partial_unused_args_eliminate]: 2.35997e-06
[environ_conv]: 6.93e-06
[add_recomputation]: 5.936e-05
[cse_after_recomputation]: 2.219e-05, [1]
    [Cycle 1]: 1.607e-05, [1]
        [cse]: 9.61e-06
[auto_monad_reorder]: 1.749e-05
[get_jit_bprop_graph]: 2.77002e-06
[rewriter_after_jit_bprop_graph]: 9.71e-06
[opt_after_jit_grad]: 0.00066249
[symbol_engine_optimizer]: 7.985e-05, [1]
    [Cycle 1]: 7.199e-05, [6]
        [build]: 4.28999e-06
        [elim_shapecalc]: 7.87e-06
        [elim_not_effective]: 1.492e-05
        [opt_reshape]: 5.47001e-06
        [fold_const_symbol]: 9.79e-06
        [renormalize]: 5.8001e-07
[validate]: 4.638e-05
Sums
    bootstrap                            :     0.000273s :  3.05%
    type_inference                       :     0.004032s : 45.03%
    event_method                         :     0.000018s :  0.21%
    auto_monad                           :     0.000069s :  0.77%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.04%
    py_interpret_to_execute              :     0.000030s :  0.34%
    rewriter_before_opt_a                :     0.000072s :  0.80%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000066s :  0.74%
    jit_opt_a.loop_unroll                :     0.000034s :  0.38%
    jit_opt_a.a_1                        :     0.000690s :  7.70%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.43%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.12%
    jit_opt_a.merge_forward              :     0.000009s :  0.10%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.06%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.39%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000019s :  0.21%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000726s :  8.10%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.27%
    jit_opt_a.cse                        :     0.000047s :  0.53%
    jit_opt_a.replace_applicator         :     0.000021s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.17%
    rewriter_after_opt_a                 :     0.000041s :  0.46%
    convert_after_rewriter               :     0.000008s :  0.09%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000848s :  9.47%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.19%
    cconv                                :     0.000038s :  0.43%
    loop_unroll                          :     0.000619s :  6.92%
    jit_opt_after_cconv.c_1              :     0.000025s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.10%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000034s :  0.38%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.26%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000016s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000007s :  0.08%
    add_recomputation                    :     0.000059s :  0.66%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000017s :  0.20%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.11%
    opt_after_jit_grad                   :     0.000662s :  7.40%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000046s :  0.52%


Time group info:
------[substitution.]   0.000217    23
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.86% :     0.000006s :      3: substitution.graph_param_transform
82.21% :     0.000179s :      4: substitution.inline
 1.80% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.52% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.78% :     0.000006s :      2: substitution.replace_old_param
 6.28% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003953     2
79.95% :     0.003161s :      1: type_inference.infer
20.05% :     0.000793s :      1: type_inference.specialize
------[replace.]   0.000070     6
67.41% :     0.000047s :      4: replace.inline
32.59% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000189     6
93.33% :     0.000176s :      4: match.inline
 6.67% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.23% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000002s :     10: predicate.addn_zero_filter
 1.99% :     0.000003s :     10: predicate.arithmetic_simplify
 1.13% :     0.000001s :     10: predicate.cast_eliminate
 0.64% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.83% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.16% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.27% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.12% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 5.59% :     0.000007s :     22: predicate.inline
 1.06% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.49% :     0.000002s :     15: predicate.load_eliminater
 1.91% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.65% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.12% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 0.95% :     0.000001s :     10: predicate.minmaximum_grad
 2.40% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.47% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.85% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.54% :     0.000002s :     18: predicate.replace_applicator
 0.84% :     0.000001s :      6: predicate.replace_old_param
 0.53% :     0.000001s :      3: predicate.reset_defer_inline
 1.59% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.16% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     10: predicate.same_eliminate
 0.64% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 1.37% :     0.000002s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.54% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.19% :     0.000003s :     16: predicate.switch_defer_inline
 1.92% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.19% :     0.000009s :     53: predicate.switch_simplify
 1.91% :     0.000002s :     10: predicate.tile_eliminate
 1.06% :     0.000001s :     10: predicate.transpose_eliminate
 1.26% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.66% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.51% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000536    12
46.61% :     0.000250s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.53% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.86% :     0.000257s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015149    72
 0.41% :     0.000063s :      1: add_recomputation
 0.48% :     0.000073s :      1: auto_monad
 0.13% :     0.000020s :      1: auto_monad_reorder
 1.94% :     0.000294s :      1: bootstrap
 0.27% :     0.000041s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000010s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.04% :     0.000006s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
39.56% :     0.005993s :      1: jit_opt_a
 1.18% :     0.000179s :      1: jit_opt_after_cconv
 0.40% :     0.000061s :      1: jit_opt_b
 4.17% :     0.000631s :      1: loop_unroll
 5.70% :     0.000864s :      1: mutable_eliminate
 6.07% :     0.000920s :     26: opt.transform.jit_opt_a
 0.36% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000028s :      4: opt.transform.jit_opt_b
 0.12% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.14% :     0.000021s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000035s :      4: opt.transform.symbol_engine_opt
 4.45% :     0.000674s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000033s :      1: py_interpret_to_execute
 0.12% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000019s :      1: remove_dup_value
 2.36% :     0.000358s :      1: renormalize.infer
 2.37% :     0.000359s :      1: renormalize.specialize
 0.08% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.31% :     0.000047s :      1: rewriter_after_opt_a
 0.49% :     0.000075s :      1: rewriter_before_opt_a
 0.54% :     0.000083s :      1: symbol_engine_optimizer
26.73% :     0.004050s :      1: type_inference


TotalTime = 0.0182338, [33]
[bootstrap]: 0.00025837
[type_inference]: 0.00400661
[event_method]: 1.609e-05
[auto_monad]: 6.07e-05
[graph_reusing]: 5.05001e-06
[pre_auto_parallel]: 2.32999e-06
[py_interpret_to_execute]: 3.049e-05
[rewriter_before_opt_a]: 6.8e-05
[expand_dump_flag]: 3.05002e-06
[jit_opt_a]: 0.00504195, [2]
    [Cycle 1]: 0.00155157, [27]
        [switch_simplify]: 4.974e-05
        [loop_unroll]: 2.973e-05
        [a_1]: 0.00052095
        [with_stream_mark]: 1.642e-05
        [recompute_prepare]: 7.66001e-06
        [updatestate_depend_eliminate]: 3.9e-06
        [updatestate_assign_eliminate]: 2.87002e-06
        [updatestate_loads_eliminate]: 2.14e-06
        [parameter_eliminate]: 1.52001e-06
        [specialize_transform]: 5.95002e-06
        [updatestate_useless_node_eliminater]: 5.45001e-06
        [accelerated_algorithm]: 5.77001e-06
        [meta_shard_fg_expand]: 1.96998e-06
        [get_grad_eliminate_]: 5.26002e-06
        [merge_forward]: 3.45e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.336e-05
        [j_node_and_user_rematch]: 8.72998e-06
        [meta_fg_expand]: 2.67001e-06
        [replace_old_param]: 8.75999e-06
        [inline_without_move]: 6.02999e-06
        [renormalize]: 0.00060449
        [add_forward_monad_depend]: 6.61e-06
        [auto_monad_grad]: 2.75002e-06
        [auto_monad_eliminator]: 1.56e-05
        [cse]: 2.75e-05
        [replace_applicator]: 1.403e-05
    [Cycle 2]: 0.00033731, [27]
        [switch_simplify]: 6.34001e-06
        [loop_unroll]: 4.94e-06
        [a_1]: 8.986e-05
        [with_stream_mark]: 1.059e-05
        [recompute_prepare]: 4.97999e-06
        [updatestate_depend_eliminate]: 2.88e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 5.06002e-06
        [updatestate_useless_node_eliminater]: 5.42999e-06
        [accelerated_algorithm]: 5.35001e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 4.63999e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 2.59001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.428e-05
        [j_node_and_user_rematch]: 8.08999e-06
        [meta_fg_expand]: 1.75001e-06
        [replace_old_param]: 1.514e-05
        [inline_without_move]: 5.22999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.39998e-06
        [auto_monad_grad]: 1.03001e-06
        [auto_monad_eliminator]: 5.77001e-06
        [cse]: 1.128e-05
        [replace_applicator]: 5.26002e-06
[py_interpret_to_execute_after_opt_a]: 1.257e-05
[rewriter_after_opt_a]: 3.836e-05
[convert_after_rewriter]: 7.06001e-06
[order_py_execute_after_rewriter]: 8.37e-06
[mutable_eliminate]: 0.00060268
[jit_opt_b]: 5.227e-05, [1]
    [Cycle 1]: 4.427e-05, [2]
        [frontend_op_eliminate]: 1.689e-05
        [inline_after_opt_a]: 1.542e-05
[cconv]: 2.486e-05
[loop_unroll]: 0.0004274
[jit_opt_after_cconv]: 0.00014514, [1]
    [Cycle 1]: 0.00013822, [11]
        [c_1]: 2.106e-05
        [parameter_eliminate]: 2.44001e-06
        [updatestate_depend_eliminate]: 5.71e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [cse]: 2.085e-05
        [call_graph_tuple_transform]: 1.898e-05
        [tuple_list_get_item_eliminator]: 5.76e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 1.01997e-06
        [switch_simplify]: 5.33002e-06
[remove_dup_value]: 1.263e-05
[partial_unused_args_eliminate]: 2.43998e-06
[environ_conv]: 5.44e-06
[add_recomputation]: 4.855e-05
[cse_after_recomputation]: 2.054e-05, [1]
    [Cycle 1]: 1.524e-05, [1]
        [cse]: 9.05999e-06
[auto_monad_reorder]: 1.733e-05
[get_jit_bprop_graph]: 2.01e-06
[rewriter_after_jit_bprop_graph]: 6.50997e-06
[opt_after_jit_grad]: 0.00046457
[symbol_engine_optimizer]: 0.00014793, [1]
    [Cycle 1]: 0.00014054, [6]
        [build]: 3.43e-06
        [elim_shapecalc]: 7.594e-05
        [elim_not_effective]: 1.515e-05
        [opt_reshape]: 6.10002e-06
        [fold_const_symbol]: 9.02e-06
        [renormalize]: 5.50004e-07
[validate]: 3.641e-05
[backend_pass]: 1.14e-06
[task_emit]: 0.00647055
[execute]: 7.96001e-06
Sums
    bootstrap                            :     0.000258s :  1.78%
    type_inference                       :     0.004007s : 27.66%
    event_method                         :     0.000016s :  0.11%
    auto_monad                           :     0.000061s :  0.42%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.21%
    rewriter_before_opt_a                :     0.000068s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000056s :  0.39%
    jit_opt_a.loop_unroll                :     0.000035s :  0.24%
    jit_opt_a.a_1                        :     0.000611s :  4.22%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.09%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.03%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.03%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.08%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.07%
    jit_opt_a.merge_forward              :     0.000007s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.12%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.03%
    jit_opt_a.replace_old_param          :     0.000024s :  0.17%
    jit_opt_a.inline_without_move        :     0.000011s :  0.08%
    jit_opt_a.renormalize                :     0.000605s :  4.17%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.15%
    jit_opt_a.cse                        :     0.000039s :  0.27%
    jit_opt_a.replace_applicator         :     0.000019s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.09%
    rewriter_after_opt_a                 :     0.000038s :  0.26%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000008s :  0.06%
    mutable_eliminate                    :     0.000603s :  4.16%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.12%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.11%
    cconv                                :     0.000025s :  0.17%
    loop_unroll                          :     0.000427s :  2.95%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.15%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.14%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.13%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.04%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000049s :  0.34%
    cse_after_recomputation.cse          :     0.000009s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000465s :  3.21%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000076s :  0.52%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.10%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.04%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.06%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.25%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.006471s : 44.68%
    execute                              :     0.000008s :  0.05%


Time group info:
------[substitution.]   0.000192    23
 1.27% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.86% :     0.000005s :      3: substitution.graph_param_transform
82.55% :     0.000158s :      4: substitution.inline
 1.61% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.29% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.30% :     0.000004s :      2: substitution.replace_old_param
 6.39% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003955     2
79.78% :     0.003155s :      1: type_inference.infer
20.22% :     0.000800s :      1: type_inference.specialize
------[replace.]   0.000068     6
66.76% :     0.000045s :      4: replace.inline
33.24% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000167     6
93.31% :     0.000156s :      4: match.inline
 6.69% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.20% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000001s :     10: predicate.addn_zero_filter
 1.99% :     0.000002s :     10: predicate.arithmetic_simplify
 1.16% :     0.000001s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.22% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.53% :     0.000001s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.37% :     0.000002s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.08% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.24% :     0.000006s :     22: predicate.inline
 1.11% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.56% :     0.000002s :      6: predicate.less_batch_normalization
 1.57% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.32% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.21% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.99% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.80% :     0.000002s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.35% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.82% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.42% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.03% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.26% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.21% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.47% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.98% :     0.000009s :     53: predicate.switch_simplify
 1.19% :     0.000001s :     10: predicate.tile_eliminate
 1.24% :     0.000001s :     10: predicate.transpose_eliminate
 1.43% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.99% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.43% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.12% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000512    12
46.90% :     0.000240s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.80% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.30% :     0.000242s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.019849    76
 0.26% :     0.000052s :      1: add_recomputation
 0.32% :     0.000064s :      1: auto_monad
 0.10% :     0.000021s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 1.35% :     0.000267s :      1: bootstrap
 0.14% :     0.000028s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.10% :     0.000019s :      1: event_method
 0.06% :     0.000012s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000007s :      1: graph_reusing
25.42% :     0.005045s :      1: jit_opt_a
 0.75% :     0.000148s :      1: jit_opt_after_cconv
 0.28% :     0.000055s :      1: jit_opt_b
 2.19% :     0.000435s :      1: loop_unroll
 3.08% :     0.000612s :      1: mutable_eliminate
 4.17% :     0.000829s :     26: opt.transform.jit_opt_a
 0.24% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000025s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.10% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.52% :     0.000102s :      4: opt.transform.symbol_engine_opt
 2.38% :     0.000473s :      1: opt_after_jit_grad
 0.05% :     0.000011s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.17% :     0.000033s :      1: py_interpret_to_execute
 0.08% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000015s :      1: remove_dup_value
 1.57% :     0.000312s :      1: renormalize.infer
 1.44% :     0.000285s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000041s :      1: rewriter_after_opt_a
 0.36% :     0.000071s :      1: rewriter_before_opt_a
 0.76% :     0.000150s :      1: symbol_engine_optimizer
32.66% :     0.006482s :      1: task_emit
20.23% :     0.004014s :      1: type_inference
 0.29% :     0.000058s :      1: validate


TotalTime = 0.0202821, [30]
[bootstrap]: 0.00020213
[type_inference]: 0.0098337
[event_method]: 4.459e-05
[auto_monad]: 9.372e-05
[graph_reusing]: 6.79001e-06
[pre_auto_parallel]: 2.75002e-06
[py_interpret_to_execute]: 2.508e-05
[rewriter_before_opt_a]: 9.064e-05
[expand_dump_flag]: 2.94999e-06
[jit_opt_a]: 0.00806876, [2]
    [Cycle 1]: 0.0045556, [27]
        [switch_simplify]: 6.001e-05
        [loop_unroll]: 3.965e-05
        [a_1]: 0.00094356
        [with_stream_mark]: 1.799e-05
        [recompute_prepare]: 1.783e-05
        [updatestate_depend_eliminate]: 7.21001e-06
        [updatestate_assign_eliminate]: 5.60001e-06
        [updatestate_loads_eliminate]: 5.72001e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 1.29e-05
        [updatestate_useless_node_eliminater]: 1.209e-05
        [accelerated_algorithm]: 1.233e-05
        [meta_shard_fg_expand]: 2.61999e-06
        [get_grad_eliminate_]: 1.222e-05
        [merge_forward]: 7.1e-06
        [cell_reuse_recompute_pass]: 6.39993e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.3e-05
        [j_node_and_user_rematch]: 2.239e-05
        [meta_fg_expand]: 0.00162115
        [replace_old_param]: 4.945e-05
        [inline_without_move]: 4.97e-05
        [renormalize]: 0.00139338
        [add_forward_monad_depend]: 5.45001e-06
        [auto_monad_grad]: 2.07001e-06
        [auto_monad_eliminator]: 1.099e-05
        [cse]: 1.906e-05
        [replace_applicator]: 1.022e-05
    [Cycle 2]: 0.0002886, [27]
        [switch_simplify]: 4.44002e-06
        [loop_unroll]: 3.7e-06
        [a_1]: 4.037e-05
        [with_stream_mark]: 8.87999e-06
        [recompute_prepare]: 3.71999e-06
        [updatestate_depend_eliminate]: 2.14e-06
        [updatestate_assign_eliminate]: 1.89999e-06
        [updatestate_loads_eliminate]: 1.60001e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 3.69002e-06
        [updatestate_useless_node_eliminater]: 3.8e-06
        [accelerated_algorithm]: 3.36001e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 3.38e-06
        [merge_forward]: 2.30002e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.164e-05
        [j_node_and_user_rematch]: 5.30999e-06
        [meta_fg_expand]: 2.543e-05
        [replace_old_param]: 3.69002e-06
        [inline_without_move]: 3.41001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.25001e-06
        [auto_monad_grad]: 1.46002e-06
        [auto_monad_eliminator]: 4.58999e-06
        [cse]: 1.142e-05
        [replace_applicator]: 4.32e-06
[py_interpret_to_execute_after_opt_a]: 9.66e-06
[rewriter_after_opt_a]: 3.309e-05
[convert_after_rewriter]: 6.32001e-06
[order_py_execute_after_rewriter]: 5.06002e-06
[mutable_eliminate]: 0.00050447
[jit_opt_b]: 3.969e-05, [1]
    [Cycle 1]: 3.25e-05, [2]
        [frontend_op_eliminate]: 1.022e-05
        [inline_after_opt_a]: 1.167e-05
[cconv]: 1.943e-05
[loop_unroll]: 0.00040889
[jit_opt_after_cconv]: 0.00011503, [1]
    [Cycle 1]: 0.00010867, [11]
        [c_1]: 1.158e-05
        [parameter_eliminate]: 2.46e-06
        [updatestate_depend_eliminate]: 4.42998e-06
        [updatestate_assign_eliminate]: 1.87001e-06
        [updatestate_loads_eliminate]: 1.64e-06
        [cse]: 1.451e-05
        [call_graph_tuple_transform]: 1.257e-05
        [tuple_list_get_item_eliminator]: 3.88001e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 5.79981e-07
        [switch_simplify]: 4.12e-06
[remove_dup_value]: 1.364e-05
[partial_unused_args_eliminate]: 2.31e-06
[environ_conv]: 4.97999e-06
[add_recomputation]: 3.03e-05
[cse_after_recomputation]: 1.929e-05, [1]
    [Cycle 1]: 1.399e-05, [1]
        [cse]: 7.96001e-06
[auto_monad_reorder]: 1.267e-05
[get_jit_bprop_graph]: 1.39003e-06
[rewriter_after_jit_bprop_graph]: 4.80999e-06
[opt_after_jit_grad]: 0.00043835
[symbol_engine_optimizer]: 6.192e-05, [1]
    [Cycle 1]: 5.59e-05, [6]
        [build]: 3.44001e-06
        [elim_shapecalc]: 5.37999e-06
        [elim_not_effective]: 9.04998e-06
        [opt_reshape]: 4.06001e-06
        [fold_const_symbol]: 5.97999e-06
        [renormalize]: 6.99976e-07
[validate]: 2.765e-05
Sums
    bootstrap                            :     0.000202s :  1.23%
    type_inference                       :     0.009834s : 59.71%
    event_method                         :     0.000045s :  0.27%
    auto_monad                           :     0.000094s :  0.57%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000025s :  0.15%
    rewriter_before_opt_a                :     0.000091s :  0.55%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.39%
    jit_opt_a.loop_unroll                :     0.000043s :  0.26%
    jit_opt_a.a_1                        :     0.000984s :  5.97%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.01%
    jit_opt_a.specialize_transform       :     0.000017s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.09%
    jit_opt_a.merge_forward              :     0.000009s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001647s : 10.00%
    jit_opt_a.replace_old_param          :     0.000053s :  0.32%
    jit_opt_a.inline_without_move        :     0.000053s :  0.32%
    jit_opt_a.renormalize                :     0.001393s :  8.46%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.09%
    jit_opt_a.cse                        :     0.000030s :  0.19%
    jit_opt_a.replace_applicator         :     0.000015s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000504s :  3.06%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.07%
    cconv                                :     0.000019s :  0.12%
    loop_unroll                          :     0.000409s :  2.48%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000030s :  0.18%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.08%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000438s :  2.66%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000028s :  0.17%


Time group info:
------[substitution.]   0.000229    50
 0.52% :     0.000001s :      1: substitution.elim_not_effective
 0.41% :     0.000001s :      1: substitution.fold_const_symbol
 1.80% :     0.000004s :      1: substitution.graph_param_transform
54.82% :     0.000126s :      6: substitution.inline
 6.10% :     0.000014s :      2: substitution.inline_without_move
 2.21% :     0.000005s :      9: substitution.j_node_and_user_rematch
14.76% :     0.000034s :      2: substitution.minmaximum_grad
 1.16% :     0.000003s :      3: substitution.partial_eliminate
 2.87% :     0.000007s :      9: substitution.remove_not_recompute_node
 1.97% :     0.000005s :      2: substitution.replace_applicator
 1.46% :     0.000003s :      5: substitution.replace_old_param
 1.10% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.97% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 2.02% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.83% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009765     2
89.01% :     0.008691s :      1: type_inference.infer
10.99% :     0.001073s :      1: type_inference.specialize
------[replace.]   0.000056     8
76.55% :     0.000043s :      6: replace.inline
23.45% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000130     8
93.68% :     0.000122s :      6: match.inline
 6.32% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000170  1109
 1.60% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.52% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.34% :     0.000002s :     18: predicate.addn_check_dump
 1.76% :     0.000003s :     18: predicate.addn_zero_filter
 2.08% :     0.000004s :     18: predicate.arithmetic_simplify
 1.52% :     0.000003s :     18: predicate.cast_eliminate
 0.16% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.25% :     0.000002s :     18: predicate.compare_switch_simplify
 1.53% :     0.000003s :     18: predicate.depend_value_elim
 1.26% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.48% :     0.000003s :     18: predicate.dict_set_item_eliminator
 0.37% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.09% :     0.000000s :      1: predicate.elim_not_effective
 0.21% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.27% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.28% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.57% :     0.000003s :     18: predicate.environ_get_eliminate
 1.26% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.25% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.18% :     0.000007s :     28: predicate.inline
 3.36% :     0.000006s :     33: predicate.inline_without_move
 0.50% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.44% :     0.000002s :     12: predicate.less_batch_normalization
 1.65% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.72% :     0.000003s :     21: predicate.load_eliminater
 0.56% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.82% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.60% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.39% :     0.000002s :     18: predicate.merge_addn
 1.30% :     0.000002s :     18: predicate.minmaximum_grad
 0.66% :     0.000001s :      1: predicate.mutable_eliminate
 0.18% :     0.000000s :      1: predicate.opt_reshape
 2.02% :     0.000003s :     21: predicate.partial_eliminate
 1.40% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.97% :     0.000003s :     18: predicate.reduce_eliminate
 1.59% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.51% :     0.000003s :     22: predicate.replace_applicator
 1.65% :     0.000003s :     33: predicate.replace_old_param
 0.12% :     0.000000s :      1: predicate.reset_defer_inline
 1.48% :     0.000003s :     18: predicate.reshape_eliminate
 1.67% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.41% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.57% :     0.000003s :     18: predicate.same_eliminate
 0.64% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.35% :     0.000001s :      2: predicate.special_op_eliminate
 1.27% :     0.000002s :     12: predicate.specialize_transform
 1.82% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.64% :     0.000004s :     26: predicate.switch_defer_inline
 2.34% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.94% :     0.000012s :     71: predicate.switch_simplify
 1.43% :     0.000002s :     18: predicate.tile_eliminate
 1.39% :     0.000002s :     18: predicate.transpose_eliminate
 1.73% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.57% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.76% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.86% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.64% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.26% :     0.000006s :     33: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000003s :     18: predicate.value_based_eliminate
 0.12% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.24% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001359    22
78.37% :     0.001065s :     13: func_graph_cloner_run.FuncGraphClonerGraph
21.63% :     0.000294s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023020    72
 0.14% :     0.000033s :      1: add_recomputation
 0.42% :     0.000097s :      1: auto_monad
 0.07% :     0.000015s :      1: auto_monad_reorder
 0.93% :     0.000214s :      1: bootstrap
 0.10% :     0.000022s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.21% :     0.000048s :      1: event_method
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
35.06% :     0.008072s :      1: jit_opt_a
 0.51% :     0.000118s :      1: jit_opt_after_cconv
 0.18% :     0.000042s :      1: jit_opt_b
 1.80% :     0.000415s :      1: loop_unroll
 2.22% :     0.000511s :      1: mutable_eliminate
 5.79% :     0.001332s :     26: opt.transform.jit_opt_a
 0.12% :     0.000028s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000015s :      4: opt.transform.jit_opt_b
 0.03% :     0.000008s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000010s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000013s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000021s :      4: opt.transform.symbol_engine_opt
 1.93% :     0.000445s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000028s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 4.50% :     0.001035s :      1: renormalize.infer
 1.53% :     0.000352s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000036s :      1: rewriter_after_opt_a
 0.41% :     0.000093s :      1: rewriter_before_opt_a
 0.28% :     0.000064s :      1: symbol_engine_optimizer
42.77% :     0.009846s :      1: type_inference


TotalTime = 0.0225932, [33]
[bootstrap]: 0.00030384
[type_inference]: 0.0115309
[event_method]: 4.978e-05
[auto_monad]: 8.303e-05
[graph_reusing]: 5.84999e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 2.613e-05
[rewriter_before_opt_a]: 0.00014789
[expand_dump_flag]: 3.63e-06
[jit_opt_a]: 0.00842847, [2]
    [Cycle 1]: 0.00467608, [27]
        [switch_simplify]: 6.122e-05
        [loop_unroll]: 3.755e-05
        [a_1]: 0.00082301
        [with_stream_mark]: 2.224e-05
        [recompute_prepare]: 1.803e-05
        [updatestate_depend_eliminate]: 7.45998e-06
        [updatestate_assign_eliminate]: 5.85002e-06
        [updatestate_loads_eliminate]: 5.16002e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 1.273e-05
        [updatestate_useless_node_eliminater]: 1.157e-05
        [accelerated_algorithm]: 1.2e-05
        [meta_shard_fg_expand]: 2.56998e-06
        [get_grad_eliminate_]: 1.18e-05
        [merge_forward]: 6.79999e-06
        [cell_reuse_recompute_pass]: 7.39994e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.277e-05
        [j_node_and_user_rematch]: 2.154e-05
        [meta_fg_expand]: 0.00141889
        [replace_old_param]: 5.106e-05
        [inline_without_move]: 4.941e-05
        [renormalize]: 0.00182619
        [add_forward_monad_depend]: 4.91997e-06
        [auto_monad_grad]: 1.45999e-06
        [auto_monad_eliminator]: 9.04e-06
        [cse]: 1.696e-05
        [replace_applicator]: 1.144e-05
    [Cycle 2]: 0.00029244, [27]
        [switch_simplify]: 5.17999e-06
        [loop_unroll]: 3.95998e-06
        [a_1]: 4.422e-05
        [with_stream_mark]: 8.05e-06
        [recompute_prepare]: 3.80998e-06
        [updatestate_depend_eliminate]: 2.09e-06
        [updatestate_assign_eliminate]: 1.69e-06
        [updatestate_loads_eliminate]: 1.58002e-06
        [parameter_eliminate]: 1.08001e-06
        [specialize_transform]: 3.89002e-06
        [updatestate_useless_node_eliminater]: 3.52002e-06
        [accelerated_algorithm]: 6.56e-06
        [meta_shard_fg_expand]: 1.32e-06
        [get_grad_eliminate_]: 4.05e-06
        [merge_forward]: 2.17001e-06
        [cell_reuse_recompute_pass]: 2.09e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.002e-05
        [j_node_and_user_rematch]: 5.70001e-06
        [meta_fg_expand]: 3.822e-05
        [replace_old_param]: 4.12998e-06
        [inline_without_move]: 3.29001e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.43002e-06
        [auto_monad_grad]: 7.90023e-07
        [auto_monad_eliminator]: 4.37e-06
        [cse]: 9.97999e-06
        [replace_applicator]: 3.78001e-06
[py_interpret_to_execute_after_opt_a]: 8e-06
[rewriter_after_opt_a]: 2.573e-05
[convert_after_rewriter]: 5.34998e-06
[order_py_execute_after_rewriter]: 3.55e-06
[mutable_eliminate]: 0.0005528
[jit_opt_b]: 4.073e-05, [1]
    [Cycle 1]: 3.406e-05, [2]
        [frontend_op_eliminate]: 1.114e-05
        [inline_after_opt_a]: 1.152e-05
[cconv]: 1.523e-05
[loop_unroll]: 0.00043392
[jit_opt_after_cconv]: 0.00011496, [1]
    [Cycle 1]: 0.00010874, [11]
        [c_1]: 1.211e-05
        [parameter_eliminate]: 2.44999e-06
        [updatestate_depend_eliminate]: 5.15999e-06
        [updatestate_assign_eliminate]: 1.81e-06
        [updatestate_loads_eliminate]: 1.59e-06
        [cse]: 1.5e-05
        [call_graph_tuple_transform]: 1.206e-05
        [tuple_list_get_item_eliminator]: 3.56999e-06
        [none_parameter_eliminate]: 8.59989e-07
        [renormalize]: 3.89991e-07
        [switch_simplify]: 3.77998e-06
[remove_dup_value]: 8.09997e-06
[partial_unused_args_eliminate]: 1.65001e-06
[environ_conv]: 3.81001e-06
[add_recomputation]: 2.303e-05
[cse_after_recomputation]: 1.996e-05, [1]
    [Cycle 1]: 1.472e-05, [1]
        [cse]: 8.60999e-06
[auto_monad_reorder]: 7.25e-06
[get_jit_bprop_graph]: 1.37e-06
[rewriter_after_jit_bprop_graph]: 4.50001e-06
[opt_after_jit_grad]: 0.00045681
[symbol_engine_optimizer]: 6.488e-05, [1]
    [Cycle 1]: 5.884e-05, [6]
        [build]: 2.89999e-06
        [elim_shapecalc]: 5.89999e-06
        [e[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.720.156 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d10d0> should be wrapped by jit only once.
lim_not_effective]: 9.52999e-06
        [opt_reshape]: 4.75999e-06
        [fold_const_symbol]: 5.86e-06
        [renormalize]: 4.60015e-07
[validate]: 2.277e-05
[backend_pass]: 9.20001e-07
[task_emit]: 1.783e-05
[execute]: 1.25001e-06
Sums
    bootstrap                            :     0.000304s :  1.64%
    type_inference                       :     0.011531s : 62.27%
    event_method                         :     0.000050s :  0.27%
    auto_monad                           :     0.000083s :  0.45%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.14%
    rewriter_before_opt_a                :     0.000148s :  0.80%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.36%
    jit_opt_a.loop_unroll                :     0.000042s :  0.22%
    jit_opt_a.a_1                        :     0.000867s :  4.68%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000019s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.09%
    jit_opt_a.merge_forward              :     0.000009s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001457s :  7.87%
    jit_opt_a.replace_old_param          :     0.000055s :  0.30%
    jit_opt_a.inline_without_move        :     0.000053s :  0.28%
    jit_opt_a.renormalize                :     0.001826s :  9.86%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.07%
    jit_opt_a.cse                        :     0.000027s :  0.15%
    jit_opt_a.replace_applicator         :     0.000015s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.04%
    rewriter_after_opt_a                 :     0.000026s :  0.14%
    convert_after_rewriter               :     0.000005s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000553s :  2.99%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.06%
    cconv                                :     0.000015s :  0.08%
    loop_unroll                          :     0.000434s :  2.34%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000012s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000008s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000023s :  0.12%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000007s :  0.04%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000457s :  2.47%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000023s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000018s :  0.10%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000239    50
 0.46% :     0.000001s :      1: substitution.elim_not_effective
 0.27% :     0.000001s :      1: substitution.fold_const_symbol
 1.05% :     0.000003s :      1: substitution.graph_param_transform
69.16% :     0.000165s :      6: substitution.inline
 6.37% :     0.000015s :      2: substitution.inline_without_move
 2.20% :     0.000005s :      9: substitution.j_node_and_user_rematch
 1.34% :     0.000003s :      2: substitution.minmaximum_grad
 1.52% :     0.000004s :      3: substitution.partial_eliminate
 2.79% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.31% :     0.000006s :      2: substitution.replace_applicator
 1.53% :     0.000004s :      5: substitution.replace_old_param
 0.97% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.67% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.59% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 5.77% :     0.000014s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011464     2
89.15% :     0.010220s :      1: type_inference.infer
10.85% :     0.001243s :      1: type_inference.specialize
------[replace.]   0.000063     8
76.92% :     0.000048s :      6: replace.inline
23.08% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000171     8
94.71% :     0.000162s :      6: match.inline
 5.29% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000169  1109
 1.49% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.77% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.32% :     0.000002s :     18: predicate.addn_check_dump
 1.68% :     0.000003s :     18: predicate.addn_zero_filter
 2.03% :     0.000003s :     18: predicate.arithmetic_simplify
 1.44% :     0.000002s :     18: predicate.cast_eliminate
 0.25% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.28% :     0.000002s :     18: predicate.compare_switch_simplify
 1.41% :     0.000002s :     18: predicate.depend_value_elim
 1.35% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.57% :     0.000003s :     18: predicate.dict_set_item_eliminator
 0.52% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      1: predicate.elim_not_effective
 0.28% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.33% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.49% :     0.000003s :     18: predicate.environ_get_eliminate
 1.31% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.08% :     0.000000s :      1: predicate.fold_const_symbol
 1.30% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.27% :     0.000007s :     28: predicate.inline
 3.31% :     0.000006s :     33: predicate.inline_without_move
 0.62% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :     12: predicate.less_batch_normalization
 1.68% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.70% :     0.000003s :     21: predicate.load_eliminater
 0.72% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.64% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.83% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.34% :     0.000002s :     18: predicate.merge_addn
 1.39% :     0.000002s :     18: predicate.minmaximum_grad
 0.85% :     0.000001s :      1: predicate.mutable_eliminate
 0.28% :     0.000000s :      1: predicate.opt_reshape
 1.98% :     0.000003s :     21: predicate.partial_eliminate
 1.36% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.70% :     0.000003s :     18: predicate.reduce_eliminate
 1.73% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.48% :     0.000002s :     22: predicate.replace_applicator
 1.60% :     0.000003s :     33: predicate.replace_old_param
 0.12% :     0.000000s :      1: predicate.reset_defer_inline
 1.55% :     0.000003s :     18: predicate.reshape_eliminate
 1.76% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.48% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     18: predicate.same_eliminate
 0.61% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.38% :     0.000001s :      2: predicate.special_op_eliminate
 1.19% :     0.000002s :     12: predicate.specialize_transform
 1.51% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.55% :     0.000004s :     26: predicate.switch_defer_inline
 2.33% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.72% :     0.000011s :     71: predicate.switch_simplify
 1.50% :     0.000003s :     18: predicate.tile_eliminate
 1.33% :     0.000002s :     18: predicate.transpose_eliminate
 1.70% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.54% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.90% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.67% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.86% :     0.000003s :     18: predicate.value_based_eliminate
 0.09% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.35% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001304    22
71.53% :     0.000933s :     13: func_graph_cloner_run.FuncGraphClonerGraph
28.47% :     0.000371s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025694    76
 0.10% :     0.000026s :      1: add_recomputation
 0.34% :     0.000087s :      1: auto_monad
 0.04% :     0.000010s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.22% :     0.000314s :      1: bootstrap
 0.07% :     0.000018s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.02% :     0.000006s :      1: environ_conv
 0.21% :     0.000054s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.03% :     0.000007s :      1: expand_dump_flag
 0.01% :     0.000003s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
32.82% :     0.008432s :      1: jit_opt_a
 0.46% :     0.000118s :      1: jit_opt_after_cconv
 0.17% :     0.000044s :      1: jit_opt_b
 1.72% :     0.000442s :      1: loop_unroll
 2.18% :     0.000561s :      1: mutable_eliminate
 4.74% :     0.001219s :     26: opt.transform.jit_opt_a
 0.11% :     0.000028s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000022s :      4: opt.transform.symbol_engine_opt
 1.81% :     0.000465s :      1: opt_after_jit_grad
 0.02% :     0.000006s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000029s :      1: py_interpret_to_execute
 0.04% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000011s :      1: remove_dup_value
 5.14% :     0.001322s :      1: renormalize.infer
 1.93% :     0.000495s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000029s :      1: rewriter_after_opt_a
 0.59% :     0.000152s :      1: rewriter_before_opt_a
 0.26% :     0.000068s :      1: symbol_engine_optimizer
 0.08% :     0.000022s :      1: task_emit
44.93% :     0.011543s :      1: type_inference
 0.14% :     0.000037s :      1: validate

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.720.726 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.726.806 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd6dcafee0> should be wrapped by jit only once.

TotalTime = 0.0318863, [33]
[bootstrap]: 0.00019565
[type_inference]: 0.00983407
[event_method]: 4.409e-05
[auto_monad]: 9.391e-05
[graph_reusing]: 7.02997e-06
[pre_auto_parallel]: 2.88e-06
[py_interpret_to_execute]: 2.515e-05
[rewriter_before_opt_a]: 9.125e-05
[expand_dump_flag]: 2.96001e-06
[jit_opt_a]: 0.0114134, [3]
    [Cycle 1]: 0.00569784, [27]
        [switch_simplify]: 5.994e-05
        [loop_unroll]: 3.667e-05
        [a_1]: 0.00089005
        [with_stream_mark]: 2.596e-05
        [recompute_prepare]: 1.744e-05
        [updatestate_depend_eliminate]: 6.83e-06
        [updatestate_assign_eliminate]: 5.91e-06
        [updatestate_loads_eliminate]: 5.67001e-06
        [parameter_eliminate]: 2.20002e-06
        [specialize_transform]: 1.284e-05
        [updatestate_useless_node_eliminater]: 1.178e-05
        [accelerated_algorithm]: 1.244e-05
        [meta_shard_fg_expand]: 2.92002e-06
        [get_grad_eliminate_]: 1.215e-05
        [merge_forward]: 8.53001e-06
        [cell_reuse_recompute_pass]: 8.89995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.704e-05
        [j_node_and_user_rematch]: 2.279e-05
        [meta_fg_expand]: 0.00143917
        [replace_old_param]: 5.097e-05
        [inline_without_move]: 5.094e-05
        [renormalize]: 0.00252738
        [add_forward_monad_depend]: 8.68001e-06
        [auto_monad_grad]: 4.90001e-06
        [auto_monad_eliminator]: 4.895e-05
        [cse]: 0.0001442
        [replace_applicator]: 6.476e-05
    [Cycle 2]: 0.00207984, [27]
        [switch_simplify]: 3.67e-05
        [loop_unroll]: 3.609e-05
        [a_1]: 0.00109997
        [with_stream_mark]: 1.22e-05
        [recompute_prepare]: 9.38997e-06
        [updatestate_depend_eliminate]: 3.67002e-06
        [updatestate_assign_eliminate]: 3.03e-06
        [updatestate_loads_eliminate]: 3.28e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 7.55e-06
        [updatestate_useless_node_eliminater]: 7.08e-06
        [accelerated_algorithm]: 7.53e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 6.73e-06
        [merge_forward]: 3.4e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.459e-05
        [j_node_and_user_rematch]: 1.146e-05
        [meta_fg_expand]: 4.267e-05
        [replace_old_param]: 9.97999e-06
        [inline_without_move]: 7.02002e-06
        [renormalize]: 0.00055566
        [add_forward_monad_depend]: 4.50001e-06
        [auto_monad_grad]: 1.17e-06
        [auto_monad_eliminator]: 1.299e-05
        [cse]: 2.39e-05
        [replace_applicator]: 1.278e-05
    [Cycle 3]: 0.00040612, [27]
        [switch_simplify]: 7.36001e-06
        [loop_unroll]: 6.73e-06
        [a_1]: 0.00014319
        [with_stream_mark]: 8.82e-06
        [recompute_prepare]: 7.05e-06
        [updatestate_depend_eliminate]: 3.85e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 3.33998e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 6.93e-06
        [updatestate_useless_node_eliminater]: 6.64001e-06
        [accelerated_algorithm]: 6.89001e-06
        [meta_shard_fg_expand]: 1.40001e-06
        [get_grad_eliminate_]: 6.56e-06
        [merge_forward]: 3.76001e-06
        [cell_reuse_recompute_pass]: 1.43002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.611e-05
        [j_node_and_user_rematch]: 1.122e-05
        [meta_fg_expand]: 2.67001e-06
        [replace_old_param]: 9.13002e-06
        [inline_without_move]: 6.69001e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.09003e-06
        [auto_monad_grad]: 7.90023e-07
        [auto_monad_eliminator]: 7.63001e-06
        [cse]: 1.592e-05
        [replace_applicator]: 6.91999e-06
[py_interpret_to_execute_after_opt_a]: 1.15e-05
[rewriter_after_opt_a]: 4.061e-05
[convert_after_rewriter]: 7.23e-06
[order_py_execute_after_rewriter]: 5.94999e-06
[mutable_eliminate]: 0.00064931
[jit_opt_b]: 5.857e-05, [1]
    [Cycle 1]: 5.201e-05, [2]
        [frontend_op_eliminate]: 2.077e-05
        [inline_after_opt_a]: 1.993e-05
[cconv]: 1.908e-05
[loop_unroll]: 0.00041837
[jit_opt_after_cconv]: 0.00016528, [1]
    [Cycle 1]: 0.00015903, [11]
        [c_1]: 3.017e-05
        [parameter_eliminate]: 2.93e-06
        [updatestate_depend_eliminate]: 6.66e-06
        [updatestate_assign_eliminate]: 3.26999e-06
        [updatestate_loads_eliminate]: 3.10002e-06
        [cse]: 2.536e-05
        [call_graph_tuple_transform]: 2.22e-05
        [tuple_list_get_item_eliminator]: 6.89999e-06
        [none_parameter_eliminate]: 1.22e-06
        [renormalize]: 5.60016e-07
        [switch_simplify]: 7.33e-06
[remove_dup_value]: 1.553e-05
[partial_unused_args_eliminate]: 1.92999e-06
[environ_conv]: 6.16e-06
[add_recomputation]: 4.563e-05
[cse_after_recomputation]: 2.839e-05, [1]
    [Cycle 1]: 2.278e-05, [1]
        [cse]: 1.726e-05
[auto_monad_reorder]: 1.658e-05
[get_jit_bprop_graph]: 1.35999e-06
[rewriter_after_jit_bprop_graph]: 1.28e-05
[opt_after_jit_grad]: 0.00046658
[symbol_engine_optimizer]: 8.819e-05, [1]
    [Cycle 1]: 8.149e-05, [6]
        [build]: 3.64002e-06
        [elim_shapecalc]: 1.075e-05
        [elim_not_effective]: 1.772e-05
        [opt_reshape]: 8.09002e-06
        [fold_const_symbol]: 1.257e-05
        [renormalize]: 4.00003e-07
[validate]: 3.775e-05
[backend_pass]: 7.30011e-07
[task_emit]: 0.00788398
[execute]: 7.45e-06
Sums
    bootstrap                            :     0.000196s :  0.70%
    type_inference                       :     0.009834s : 35.24%
    event_method                         :     0.000044s :  0.16%
    auto_monad                           :     0.000094s :  0.34%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000025s :  0.09%
    rewriter_before_opt_a                :     0.000091s :  0.33%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000104s :  0.37%
    jit_opt_a.loop_unroll                :     0.000079s :  0.28%
    jit_opt_a.a_1                        :     0.002133s :  7.65%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000034s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000014s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000027s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000027s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.09%
    jit_opt_a.merge_forward              :     0.000016s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000058s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000045s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001485s :  5.32%
    jit_opt_a.replace_old_param          :     0.000070s :  0.25%
    jit_opt_a.inline_without_move        :     0.000065s :  0.23%
    jit_opt_a.renormalize                :     0.003083s : 11.05%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000070s :  0.25%
    jit_opt_a.cse                        :     0.000184s :  0.66%
    jit_opt_a.replace_applicator         :     0.000084s :  0.30%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000041s :  0.15%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000649s :  2.33%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.07%
    cconv                                :     0.000019s :  0.07%
    loop_unroll                          :     0.000418s :  1.50%
    jit_opt_after_cconv.c_1              :     0.000030s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000016s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000046s :  0.16%
    cse_after_recomputation.cse          :     0.000017s :  0.06%
    auto_monad_reorder                   :     0.000017s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000013s :  0.05%
    opt_after_jit_grad                   :     0.000467s :  1.67%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007884s : 28.26%
    execute                              :     0.000007s :  0.03%

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.727.483 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

Time group info:
------[substitution.]   0.000503   118
 4.93% :     0.000025s :      1: substitution.arithmetic_simplify
 0.54% :     0.000003s :      4: substitution.elim_not_effective
 0.34% :     0.000002s :      4: substitution.fold_const_symbol
 1.19% :     0.000006s :      5: substitution.graph_param_transform
61.89% :     0.000311s :     12: substitution.inline
 3.35% :     0.000017s :      2: substitution.inline_without_move
 1.93% :     0.000010s :     16: substitution.j_node_and_user_rematch
 1.76% :     0.000009s :      7: substitution.minmaximum_grad
 1.32% :     0.000007s :      5: substitution.partial_eliminate
 2.41% :     0.000012s :     16: substitution.remove_not_recompute_node
 4.93% :     0.000025s :     10: substitution.replace_applicator
 1.41% :     0.000007s :      7: substitution.replace_old_param
 0.54% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.51% :     0.000018s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.41% :     0.000012s :      7: substitution.tuple_list_get_item_depend_reorder
 7.54% :     0.000038s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009766     2
89.00% :     0.008692s :      1: type_inference.infer
11.00% :     0.001074s :      1: type_inference.specialize
------[replace.]   0.000145    20
 3.70% :     0.000005s :      1: replace.arithmetic_simplify
61.77% :     0.000089s :     12: replace.inline
34.53% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000350    20
 6.89% :     0.000024s :      1: match.arithmetic_simplify
87.25% :     0.000305s :     12: match.inline
 5.86% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000387  2790
 1.61% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.48% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.42% :     0.000006s :     46: predicate.addn_check_dump
 1.64% :     0.000006s :     46: predicate.addn_zero_filter
 2.10% :     0.000008s :     47: predicate.arithmetic_simplify
 1.63% :     0.000006s :     47: predicate.cast_eliminate
 0.21% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.41% :     0.000005s :     46: predicate.compare_switch_simplify
 1.58% :     0.000006s :     46: predicate.depend_value_elim
 1.48% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.57% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.55% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.35% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      5: predicate.elim_not_effective
 0.25% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.58% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.50% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.50% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.56% :     0.000006s :     47: predicate.environ_get_eliminate
 1.47% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      5: predicate.fold_const_symbol
 0.85% :     0.000003s :     21: predicate.get_grad_eliminate
 0.10% :     0.000000s :      5: predicate.graph_param_transform
 4.26% :     0.000017s :     76: predicate.inline
 1.71% :     0.000007s :     42: predicate.inline_without_move
 0.37% :     0.000001s :     21: predicate.j_node_and_user_rematch
 1.03% :     0.000004s :     21: predicate.less_batch_normalization
 1.92% :     0.000007s :     54: predicate.list_to_tuple_eliminator_
 1.96% :     0.000008s :     59: predicate.load_eliminater
 0.50% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.10% :     0.000012s :     89: predicate.loop_unroll_before_grad
 1.89% :     0.000007s :     52: predicate.make_slice_get_slice_eliminator
 1.43% :     0.000006s :     46: predicate.merge_addn
 1.55% :     0.000006s :     47: predicate.minmaximum_grad
 0.59% :     0.000002s :      5: predicate.mutable_eliminate
 0.26% :     0.000001s :      5: predicate.opt_reshape
 2.35% :     0.000009s :     59: predicate.partial_eliminate
 1.63% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.80% :     0.000007s :     47: predicate.reduce_eliminate
 1.76% :     0.000007s :     54: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.51% :     0.000010s :    105: predicate.replace_applicator
 0.86% :     0.000003s :     42: predicate.replace_old_param
 0.15% :     0.000001s :      5: predicate.reset_defer_inline
 1.75% :     0.000007s :     47: predicate.reshape_eliminate
 1.48% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.31% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.53% :     0.000006s :     46: predicate.same_eliminate
 0.48% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.41% :     0.000002s :     10: predicate.special_op_eliminate
 0.90% :     0.000003s :     21: predicate.specialize_transform
 1.72% :     0.000007s :     46: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.19% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.74% :     0.000011s :     66: predicate.switch_defer_inline
 2.44% :     0.000009s :     66: predicate.switch_layer_defer_inline
 5.98% :     0.000023s :    160: predicate.switch_simplify
 1.53% :     0.000006s :     47: predicate.tile_eliminate
 1.50% :     0.000006s :     47: predicate.transpose_eliminate
 1.85% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.81% :     0.000007s :     47: predicate.tuple_list_get_item_depend_reorder
 3.04% :     0.000012s :     64: predicate.tuple_list_get_item_eliminator
 1.86% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.82% :     0.000007s :     54: predicate.tuple_to_list_eliminator_
 1.91% :     0.000007s :     59: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000012s :     80: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000007s :     46: predicate.value_based_eliminate
 0.19% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.25% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001494    29
59.35% :     0.000887s :     13: func_graph_cloner_run.FuncGraphClonerGraph
40.65% :     0.000607s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.037864    91
 0.13% :     0.000048s :      1: add_recomputation
 0.26% :     0.000098s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.55% :     0.000207s :      1: bootstrap
 0.06% :     0.000022s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000030s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.13% :     0.000048s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
30.15% :     0.011416s :      1: jit_opt_a
 0.44% :     0.000168s :      1: jit_opt_after_cconv
 0.16% :     0.000061s :      1: jit_opt_b
 1.13% :     0.000426s :      1: loop_unroll
 1.74% :     0.000658s :      1: mutable_eliminate
 7.23% :     0.002738s :     39: opt.transform.jit_opt_a
 0.17% :     0.000063s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000035s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000046s :      4: opt.transform.symbol_engine_opt
 1.25% :     0.000475s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000028s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000018s :      1: remove_dup_value
 4.36% :     0.001653s :      2: renormalize.infer
 3.75% :     0.001418s :      2: renormalize.specialize
 0.04% :     0.000015s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000043s :      1: rewriter_after_opt_a
 0.25% :     0.000094s :      1: rewriter_before_opt_a
 0.24% :     0.000091s :      1: symbol_engine_optimizer
20.86% :     0.007898s :      1: task_emit
26.00% :     0.009846s :      1: type_inference
 0.16% :     0.000061s :      1: validate


TotalTime = 0.0139205, [30]
[bootstrap]: 0.00030981
[type_inference]: 0.00447332
[event_method]: 1.741e-05
[auto_monad]: 6.697e-05
[graph_reusing]: 6.30997e-06
[pre_auto_parallel]: 2.79999e-06
[py_interpret_to_execute]: 3.129e-05
[rewriter_before_opt_a]: 7.705e-05
[expand_dump_flag]: 3.03e-06
[jit_opt_a]: 0.00638543, [2]
    [Cycle 1]: 0.0017961, [27]
        [switch_simplify]: 5.25e-05
        [loop_unroll]: 2.995e-05
        [a_1]: 0.00059931
        [with_stream_mark]: 2.24e-05
        [recompute_prepare]: 1.038e-05
        [updatestate_depend_eliminate]: 3.85e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.20999e-06
        [specialize_transform]: 6.41e-06
        [updatestate_useless_node_eliminater]: 5.31998e-06
        [accelerated_algorithm]: 6.75998e-06
        [meta_shard_fg_expand]: 1.57001e-06
        [get_grad_eliminate_]: 4.87998e-06
        [merge_forward]: 3.76001e-06
        [cell_reuse_recompute_pass]: 1.90001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.504e-05
        [j_node_and_user_rematch]: 9.84001e-06
        [meta_fg_expand]: 1.89e-06
        [replace_old_param]: 9.65002e-06
        [inline_without_move]: 5.20999e-06
        [renormalize]: 0.00074124
        [add_forward_monad_depend]: 7.08e-06
        [auto_monad_grad]: 3.19001e-06
        [auto_monad_eliminator]: 1.632e-05
        [cse]: 3.026e-05
        [replace_applicator]: 1.216e-05
    [Cycle 2]: 0.00037403, [27]
        [switch_simplify]: 6.44999e-06
        [loop_unroll]: 5.29998e-06
        [a_1]: 9.122e-05
        [with_stream_mark]: 1.376e-05
        [recompute_prepare]: 5.87999e-06
        [updatestate_depend_eliminate]: 3.43e-06
        [updatestate_assign_eliminate]: 2.96999e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 4.78001e-06
        [updatestate_useless_node_eliminater]: 4.68999e-06
        [accelerated_algorithm]: 5.42999e-06
        [meta_shard_fg_expand]: 1.51998e-06
        [get_grad_eliminate_]: 4.78001e-06
        [merge_forward]: 3.43999e-06
        [cell_reuse_recompute_pass]: 2.21e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.604e-05
        [j_node_and_user_rematch]: 9.54e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 9.05001e-06
        [inline_without_move]: 1.032e-05
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 3.6e-06
        [auto_monad_grad]: 1.47999e-06
        [auto_monad_eliminator]: 1.026e-05
        [cse]: 1.455e-05
        [replace_applicator]: 6.41e-06
[py_interpret_to_execute_after_opt_a]: 1.405e-05
[rewriter_after_opt_a]: 3.941e-05
[convert_after_rewriter]: 7.51001e-06
[order_py_execute_after_rewriter]: 5.89999e-06
[mutable_eliminate]: 0.00078638
[jit_opt_b]: 5.411e-05, [1]
    [Cycle 1]: 4.718e-05, [2]
        [frontend_op_eliminate]: 1.699e-05
        [inline_after_opt_a]: 1.64e-05
[cconv]: 2.744e-05
[loop_unroll]: 0.00052063
[jit_opt_after_cconv]: 0.00016124, [1]
    [Cycle 1]: 0.00015349, [11]
        [c_1]: 2.195e-05
        [parameter_eliminate]: 2.31998e-06
        [updatestate_depend_eliminate]: 6.83e-06
        [updatestate_assign_eliminate]: 2.83998e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [cse]: 2.717e-05
        [call_graph_tuple_transform]: 2.116e-05
        [tuple_list_get_item_eliminator]: 5.27001e-06
        [none_parameter_eliminate]: 1.14e-06
        [renormalize]: 9.70002e-07
        [switch_simplify]: 5.59e-06
[remove_dup_value]: 1.147e-05
[partial_unused_args_eliminate]: 1.99e-06
[environ_conv]: 5.25001e-06
[add_recomputation]: 4.358e-05
[cse_after_recomputation]: 2.337e-05, [1]
    [Cycle 1]: 1.774e-05, [1]
        [cse]: 9.76998e-06
[auto_monad_reorder]: 1.363e-05
[get_jit_bprop_graph]: 2.17999e-06
[rewriter_after_jit_bprop_graph]: 8.31002e-06
[opt_after_jit_grad]: 0.00052595
[symbol_engine_optimizer]: 7.401e-05, [1]
    [Cycle 1]: 6.625e-05, [6]
        [build]: 3.76001e-06
        [elim_shapecalc]: 7.28e-06
        [elim_not_effective]: 1.205e-05
        [opt_reshape]: 5.92999e-06
        [fold_const_symbol]: 8.22e-06
        [renormalize]: 4.40021e-07
[validate]: 3.13e-05
Sums
    bootstrap                            :     0.000310s :  3.42%
    type_inference                       :     0.004473s : 49.37%
    event_method                         :     0.000017s :  0.19%
    auto_monad                           :     0.000067s :  0.74%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000031s :  0.35%
    rewriter_before_opt_a                :     0.000077s :  0.85%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000059s :  0.65%
    jit_opt_a.loop_unroll                :     0.000035s :  0.39%
    jit_opt_a.a_1                        :     0.000691s :  7.62%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.40%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.18%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.21%
    jit_opt_a.inline_without_move        :     0.000016s :  0.17%
    jit_opt_a.renormalize                :     0.000741s :  8.18%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.12%
    jit_opt_a.auto_monad_grad            :     0.000005s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.29%
    jit_opt_a.cse                        :     0.000045s :  0.49%
    jit_opt_a.replace_applicator         :     0.000019s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.16%
    rewriter_after_opt_a                 :     0.000039s :  0.43%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.07%
    mutable_eliminate                    :     0.000786s :  8.68%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.18%
    cconv                                :     0.000027s :  0.30%
    loop_unroll                          :     0.000521s :  5.75%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000027s :  0.30%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000011s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000044s :  0.48%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000014s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.09%
    opt_after_jit_grad                   :     0.000526s :  5.80%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.35%


Time group info:
------[substitution.]   0.000247    23
 0.60% :     0.000001s :      2: substitution.elim_not_effective
 0.40% :     0.000001s :      2: substitution.fold_const_symbol
 2.01% :     0.000005s :      3: substitution.graph_param_transform
85.03% :     0.000210s :      4: substitution.inline
 1.65% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.11% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.53% :     0.000006s :      2: substitution.replace_old_param
 5.68% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004400     2
81.57% :     0.003589s :      1: type_inference.infer
18.43% :     0.000811s :      1: type_inference.specialize
------[replace.]   0.000079     6
63.84% :     0.000051s :      4: replace.inline
36.16% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000221     6
94.12% :     0.000208s :      4: match.inline
 5.88% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.02% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 3.05% :     0.000004s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.62% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.01% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.23% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.04% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.25% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.00% :     0.000001s :     10: predicate.environ_get_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000000s :      3: predicate.graph_param_transform
 5.04% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.63% :     0.000002s :      6: predicate.less_batch_normalization
 1.31% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.53% :     0.000002s :     15: predicate.load_eliminater
 1.61% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.33% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.80% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 2.27% :     0.000003s :      3: predicate.mutable_eliminate
 0.59% :     0.000001s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.87% :     0.000002s :     10: predicate.reduce_eliminate
 1.90% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.78% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.85% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.04% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.90% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.21% :     0.000002s :     10: predicate.same_eliminate
 0.66% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.09% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.25% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.57% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.57% :     0.000003s :     16: predicate.switch_defer_inline
 1.88% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.64% :     0.000010s :     53: predicate.switch_simplify
 1.23% :     0.000002s :     10: predicate.tile_eliminate
 1.04% :     0.000001s :     10: predicate.transpose_eliminate
 1.16% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.07% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.55% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.85% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000562    12
48.59% :     0.000273s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.78% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.63% :     0.000245s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015644    72
 0.30% :     0.000047s :      1: add_recomputation
 0.45% :     0.000070s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 2.10% :     0.000329s :      1: bootstrap
 0.20% :     0.000031s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000022s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
40.84% :     0.006389s :      1: jit_opt_a
 1.05% :     0.000164s :      1: jit_opt_after_cconv
 0.36% :     0.000057s :      1: jit_opt_b
 3.39% :     0.000530s :      1: loop_unroll
 5.10% :     0.000798s :      1: mutable_eliminate
 5.88% :     0.000920s :     26: opt.transform.jit_opt_a
 0.32% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.42% :     0.000535s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000034s :      1: py_interpret_to_execute
 0.11% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000014s :      1: remove_dup_value
 2.51% :     0.000393s :      1: renormalize.infer
 2.17% :     0.000340s :      1: renormalize.specialize
 0.07% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000043s :      1: rewriter_after_opt_a
 0.52% :     0.000081s :      1: rewriter_before_opt_a
 0.49% :     0.000076s :      1: symbol_engine_optimizer
28.72% :     0.004492s :      1: type_inference

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.734.532 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e3d30> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.735.104 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0209863, [30]
[bootstrap]: 0.00020185
[type_inference]: 0.0095992
[event_method]: 4.415e-05
[auto_monad]: 8.166e-05
[graph_reusing]: 6.06e-06
[pre_auto_parallel]: 2.49999e-06
[py_interpret_to_execute]: 2.4e-05
[rewriter_before_opt_a]: 8.835e-05
[expand_dump_flag]: 2.73e-06
[jit_opt_a]: 0.00862956, [2]
    [Cycle 1]: 0.00495692, [27]
        [switch_simplify]: 5.975e-05
        [loop_unroll]: 4.044e-05
        [a_1]: 0.00090713
        [with_stream_mark]: 2.841e-05
        [recompute_prepare]: 1.892e-05
        [updatestate_depend_eliminate]: 8e-06
        [updatestate_assign_eliminate]: 6.19999e-06
        [updatestate_loads_eliminate]: 5.77999e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 1.352e-05
        [updatestate_useless_node_eliminater]: 1.212e-05
        [accelerated_algorithm]: 1.212e-05
        [meta_shard_fg_expand]: 3.53e-06
        [get_grad_eliminate_]: 1.185e-05
        [merge_forward]: 7.97e-06
        [cell_reuse_recompute_pass]: 9.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.456e-05
        [j_node_and_user_rematch]: 2.306e-05
        [meta_fg_expand]: 0.00183491
        [replace_old_param]: 5.45e-05
        [inline_without_move]: 5.482e-05
        [renormalize]: 0.00154623
        [add_forward_monad_depend]: 6.29001e-06
        [auto_monad_grad]: 1.99e-06
        [auto_monad_eliminator]: 1.31e-05
        [cse]: 2.859e-05
        [replace_applicator]: 1.275e-05
    [Cycle 2]: 0.00031839, [27]
        [switch_simplify]: 4.68999e-06
        [loop_unroll]: 3.8e-06
        [a_1]: 4.537e-05
        [with_stream_mark]: 1.168e-05
        [recompute_prepare]: 4.03999e-06
        [updatestate_depend_eliminate]: 3.16001e-06
        [updatestate_assign_eliminate]: 2.59999e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 1.93002e-06
        [specialize_transform]: 4.19002e-06
        [updatestate_useless_node_eliminater]: 3.81999e-06
        [accelerated_algorithm]: 3.97e-06
        [meta_shard_fg_expand]: 2.24001e-06
        [get_grad_eliminate_]: 3.38e-06
        [merge_forward]: 2.99001e-06
        [cell_reuse_recompute_pass]: 1.95001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.453e-05
        [j_node_and_user_rematch]: 7.08e-06
        [meta_fg_expand]: 3.017e-05
        [replace_old_param]: 5.37999e-06
        [inline_without_move]: 3.49001e-06
        [renormalize]: 1.09983e-07
        [add_forward_monad_depend]: 2.54999e-06
        [auto_monad_grad]: 1.05999e-06
        [auto_monad_eliminator]: 5.67999e-06
        [cse]: 1.322e-05
        [replace_applicator]: 9.05001e-06
[py_interpret_to_execute_after_opt_a]: 1.036e-05
[rewriter_after_opt_a]: 3.481e-05
[convert_after_rewriter]: 7.2e-06
[order_py_execute_after_rewriter]: 4.57e-06
[mutable_eliminate]: 0.0006691
[jit_opt_b]: 4.274e-05, [1]
    [Cycle 1]: 3.556e-05, [2]
        [frontend_op_eliminate]: 1.119e-05
        [inline_after_opt_a]: 1.274e-05
[cconv]: 2.417e-05
[loop_unroll]: 0.00045842
[jit_opt_after_cconv]: 0.00013065, [1]
    [Cycle 1]: 0.00012274, [11]
        [c_1]: 1.329e-05
        [parameter_eliminate]: 2.88e-06
        [updatestate_depend_eliminate]: 5.25999e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [cse]: 1.995e-05
        [call_graph_tuple_transform]: 1.412e-05
        [tuple_list_get_item_eliminator]: 3.72002e-06
        [none_parameter_eliminate]: 1.27999e-06
        [renormalize]: 1.23002e-06
        [switch_simplify]: 4.79998e-06
[remove_dup_value]: 1.417e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 5.04998e-06
[add_recomputation]: 3.352e-05
[cse_after_recomputation]: 2.12e-05, [1]
    [Cycle 1]: 1.532e-05, [1]
        [cse]: 8.79e-06
[auto_monad_reorder]: 1.26e-05
[get_jit_bprop_graph]: 1.76e-06
[rewriter_after_jit_bprop_graph]: 5.14e-06
[opt_after_jit_grad]: 0.00055938
[symbol_engine_optimizer]: 6.843e-05, [1]
    [Cycle 1]: 6.142e-05, [6]
        [build]: 3.93001e-06
        [elim_shapecalc]: 6.68998e-06
        [elim_not_effective]: 1.091e-05
        [opt_reshape]: 4.84e-06
        [fold_const_symbol]: 5.82001e-06
        [renormalize]: 7.7e-07
[validate]: 3.201e-05
Sums
    bootstrap                            :     0.000202s :  1.19%
    type_inference                       :     0.009599s : 56.48%
    event_method                         :     0.000044s :  0.26%
    auto_monad                           :     0.000082s :  0.48%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000024s :  0.14%
    rewriter_before_opt_a                :     0.000088s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.38%
    jit_opt_a.loop_unroll                :     0.000044s :  0.26%
    jit_opt_a.a_1                        :     0.000953s :  5.60%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000023s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000018s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.09%
    jit_opt_a.merge_forward              :     0.000011s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000030s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001865s : 10.97%
    jit_opt_a.replace_old_param          :     0.000060s :  0.35%
    jit_opt_a.inline_without_move        :     0.000058s :  0.34%
    jit_opt_a.renormalize                :     0.001546s :  9.10%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.11%
    jit_opt_a.cse                        :     0.000042s :  0.25%
    jit_opt_a.replace_applicator         :     0.000022s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000035s :  0.20%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000669s :  3.94%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000013s :  0.07%
    cconv                                :     0.000024s :  0.14%
    loop_unroll                          :     0.000458s :  2.70%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000034s :  0.20%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000559s :  3.29%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.19%


Time group info:
------[substitution.]   0.000262    50
 0.66% :     0.000002s :      1: substitution.elim_not_effective
 0.30% :     0.000001s :      1: substitution.fold_const_symbol
 1.51% :     0.000004s :      1: substitution.graph_param_transform
56.68% :     0.000148s :      6: substitution.inline
 7.04% :     0.000018s :      2: substitution.inline_without_move
 2.80% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.34% :     0.000004s :      2: substitution.minmaximum_grad
12.46% :     0.000033s :      3: substitution.partial_eliminate
 3.02% :     0.000008s :      9: substitution.remove_not_recompute_node
 2.00% :     0.000005s :      2: substitution.replace_applicator
 1.67% :     0.000004s :      5: substitution.replace_old_param
 1.00% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.66% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.82% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.03% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009535     2
89.26% :     0.008511s :      1: type_inference.infer
10.74% :     0.001024s :      1: type_inference.specialize
------[replace.]   0.000061     8
78.54% :     0.000048s :      6: replace.inline
21.46% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000153     8
94.75% :     0.000145s :      6: match.inline
 5.25% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000177  1109
 1.52% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.79% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.23% :     0.000002s :     18: predicate.addn_check_dump
 1.45% :     0.000003s :     18: predicate.addn_zero_filter
 2.27% :     0.000004s :     18: predicate.arithmetic_simplify
 1.36% :     0.000002s :     18: predicate.cast_eliminate
 0.15% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.27% :     0.000002s :     18: predicate.compare_switch_simplify
 1.36% :     0.000002s :     18: predicate.depend_value_elim
 1.18% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.41% :     0.000003s :     18: predicate.dict_set_item_eliminator
 0.37% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      1: predicate.elim_not_effective
 0.27% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.26% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.33% :     0.000002s :     18: predicate.environ_get_eliminate
 1.32% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.17% :     0.000002s :     12: predicate.get_grad_eliminate
 0.20% :     0.000000s :      1: predicate.graph_param_transform
 4.42% :     0.000008s :     28: predicate.inline
 3.69% :     0.000007s :     33: predicate.inline_without_move
 0.50% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :     12: predicate.less_batch_normalization
 1.60% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.78% :     0.000003s :     21: predicate.load_eliminater
 1.05% :     0.000002s :      1: predicate.loop_unroll_after_grad
 3.84% :     0.000007s :     44: predicate.loop_unroll_before_grad
 1.70% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.47% :     0.000003s :     18: predicate.merge_addn
 1.23% :     0.000002s :     18: predicate.minmaximum_grad
 0.99% :     0.000002s :      1: predicate.mutable_eliminate
 0.27% :     0.000000s :      1: predicate.opt_reshape
 2.14% :     0.000004s :     21: predicate.partial_eliminate
 1.58% :     0.000003s :     18: predicate.print_const_string_wrapper
 1.74% :     0.000003s :     18: predicate.reduce_eliminate
 1.47% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.73% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.56% :     0.000003s :     22: predicate.replace_applicator
 1.93% :     0.000003s :     33: predicate.replace_old_param
 0.11% :     0.000000s :      1: predicate.reset_defer_inline
 1.54% :     0.000003s :     18: predicate.reshape_eliminate
 1.38% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.52% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     18: predicate.same_eliminate
 0.68% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.42% :     0.000001s :      2: predicate.special_op_eliminate
 1.30% :     0.000002s :     12: predicate.specialize_transform
 1.51% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.39% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.64% :     0.000005s :     26: predicate.switch_defer_inline
 2.18% :     0.000004s :     26: predicate.switch_layer_defer_inline
 7.18% :     0.000013s :     71: predicate.switch_simplify
 1.43% :     0.000003s :     18: predicate.tile_eliminate
 1.53% :     0.000003s :     18: predicate.transpose_eliminate
 1.58% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.48% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.76% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.48% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.07% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000003s :     18: predicate.value_based_eliminate
 0.12% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.30% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001288    22
77.67% :     0.001000s :     13: func_graph_cloner_run.FuncGraphClonerGraph
22.33% :     0.000288s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023885    72
 0.15% :     0.000037s :      1: add_recomputation
 0.36% :     0.000086s :      1: auto_monad
 0.06% :     0.000015s :      1: auto_monad_reorder
 0.89% :     0.000213s :      1: bootstrap
 0.11% :     0.000027s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.20% :     0.000048s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
36.14% :     0.008633s :      1: jit_opt_a
 0.56% :     0.000133s :      1: jit_opt_after_cconv
 0.19% :     0.000045s :      1: jit_opt_b
 1.96% :     0.000468s :      1: loop_unroll
 2.84% :     0.000679s :      1: mutable_eliminate
 5.55% :     0.001326s :     26: opt.transform.jit_opt_a
 0.14% :     0.000033s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000017s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000017s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000024s :      4: opt.transform.symbol_engine_opt
 2.38% :     0.000569s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.11% :     0.000027s :      1: py_interpret_to_execute
 0.05% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 4.84% :     0.001157s :      1: renormalize.infer
 1.59% :     0.000380s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000039s :      1: rewriter_after_opt_a
 0.38% :     0.000091s :      1: rewriter_before_opt_a
 0.30% :     0.000071s :      1: symbol_engine_optimizer
40.24% :     0.009612s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.747.199 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d1e50> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.747.744 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0233009, [33]
[bootstrap]: 0.00022669
[type_inference]: 0.0108169
[event_method]: 5.261e-05
[auto_monad]: 9.603e-05
[graph_reusing]: 7.26999e-06
[pre_auto_parallel]: 2.96001e-06
[py_interpret_to_execute]: 3.009e-05
[rewriter_before_opt_a]: 9.911e-05
[expand_dump_flag]: 3.24001e-06
[jit_opt_a]: 0.00974843, [2]
    [Cycle 1]: 0.00547896, [27]
        [switch_simplify]: 6.891e-05
        [loop_unroll]: 3.76e-05
        [a_1]: 0.00090263
        [with_stream_mark]: 3.426e-05
        [recompute_prepare]: 1.976e-05
        [updatestate_depend_eliminate]: 8.15e-06
        [updatestate_assign_eliminate]: 5.91998e-06
        [updatestate_loads_eliminate]: 5.81003e-06
        [parameter_eliminate]: 2.64001e-06
        [specialize_transform]: 1.32e-05
        [updatestate_useless_node_eliminater]: 1.16e-05
        [accelerated_algorithm]: 1.241e-05
        [meta_shard_fg_expand]: 4.19002e-06
        [get_grad_eliminate_]: 1.21e-05
        [merge_forward]: 8.83001e-06
        [cell_reuse_recompute_pass]: 9.29984e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.695e-05
        [j_node_and_user_rematch]: 2.388e-05
        [meta_fg_expand]: 0.00181355
        [replace_old_param]: 6.289e-05
        [inline_without_move]: 5.778e-05
        [renormalize]: 0.00207325
        [add_forward_monad_depend]: 6.26e-06
        [auto_monad_grad]: 2.41998e-06
        [auto_monad_eliminator]: 1.394e-05
        [cse]: 2.472e-05
        [replace_applicator]: 1.329e-05
    [Cycle 2]: 0.00031922, [27]
        [switch_simplify]: 4.74e-06
        [loop_unroll]: 4.01001e-06
        [a_1]: 5.147e-05
        [with_stream_mark]: 1.044e-05
        [recompute_prepare]: 3.85998e-06
        [updatestate_depend_eliminate]: 2.56e-06
        [updatestate_assign_eliminate]: 2.02999e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 1.75001e-06
        [specialize_transform]: 4.01001e-06
        [updatestate_useless_node_eliminater]: 3.6e-06
        [accelerated_algorithm]: 7.15e-06
        [meta_shard_fg_expand]: 2.19001e-06
        [get_grad_eliminate_]: 3.53999e-06
        [merge_forward]: 3.08e-06
        [cell_reuse_recompute_pass]: 3.14999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.277e-05
        [j_node_and_user_rematch]: 6.07999e-06
        [meta_fg_expand]: 4.45e-05
        [replace_old_param]: 5.03002e-06
        [inline_without_move]: 3.31001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.48002e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 4.27e-06
        [cse]: 1.009e-05
        [replace_applicator]: 3.97998e-06
[py_interpret_to_execute_after_opt_a]: 1.059e-05
[rewriter_after_opt_a]: 3.259e-05
[convert_after_rewriter]: 6.16e-06
[order_py_execute_after_rewriter]: 4.2e-06
[mutable_eliminate]: 0.00065731
[jit_opt_b]: 4.404e-05, [1]
    [Cycle 1]: 3.6e-05, [2]
        [frontend_op_eliminate]: 1.247e-05
        [inline_after_opt_a]: 1.176e-05
[cconv]: 2.363e-05
[loop_unroll]: 0.00044746
[jit_opt_after_cconv]: 0.00012196, [1]
    [Cycle 1]: 0.00011527, [11]
        [c_1]: 1.277e-05
        [parameter_eliminate]: 2.69001e-06
        [updatestate_depend_eliminate]: 4.95001e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 1.66e-06
        [cse]: 1.57e-05
        [call_graph_tuple_transform]: 1.302e-05
        [tuple_list_get_item_eliminator]: 3.78001e-06
        [none_parameter_eliminate]: 1.37e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 3.95e-06
[remove_dup_value]: 1.366e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 5.30999e-06
[add_recomputation]: 3.38e-05
[cse_after_recomputation]: 1.981e-05, [1]
    [Cycle 1]: 1.436e-05, [1]
        [cse]: 8.35001e-06
[auto_monad_reorder]: 1.29e-05
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 5.62001e-06
[opt_after_jit_grad]: 0.00045945
[symbol_engine_optimizer]: 7.052e-05, [1]
    [Cycle 1]: 6.347e-05, [6]
        [build]: 3.63e-06
        [elim_shapecalc]: 6.49001e-06
        [elim_not_effective]: 1.071e-05
        [opt_reshape]: 4.53001e-06
        [fold_const_symbol]: 6.64999e-06
        [renormalize]: 4.69998e-07
[validate]: 2.919e-05
[backend_pass]: 1.10001e-06
[task_emit]: 2.418e-05
[execute]: 1.34e-06
Sums
    bootstrap                            :     0.000227s :  1.21%
    type_inference                       :     0.010817s : 57.83%
    event_method                         :     0.000053s :  0.28%
    auto_monad                           :     0.000096s :  0.51%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.16%
    rewriter_before_opt_a                :     0.000099s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.39%
    jit_opt_a.loop_unroll                :     0.000042s :  0.22%
    jit_opt_a.a_1                        :     0.000954s :  5.10%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000024s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000020s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.08%
    jit_opt_a.merge_forward              :     0.000012s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000030s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001858s :  9.93%
    jit_opt_a.replace_old_param          :     0.000068s :  0.36%
    jit_opt_a.inline_without_move        :     0.000061s :  0.33%
    jit_opt_a.renormalize                :     0.002073s : 11.08%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.10%
    jit_opt_a.cse                        :     0.000035s :  0.19%
    jit_opt_a.replace_applicator         :     0.000017s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.17%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000657s :  3.51%
    jit_opt_b.frontend_op_eliminate      :     0.000012s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.06%
    cconv                                :     0.000024s :  0.13%
    loop_unroll                          :     0.000447s :  2.39%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000014s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000034s :  0.18%
    cse_after_recomputation.cse          :     0.000008s :  0.04%
    auto_monad_reorder                   :     0.000013s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000459s :  2.46%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000007s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.16%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.000024s :  0.13%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000308    50
 0.52% :     0.000002s :      1: substitution.elim_not_effective
 0.40% :     0.000001s :      1: substitution.fold_const_symbol
 1.11% :     0.000003s :      1: substitution.graph_param_transform
69.30% :     0.000213s :      6: substitution.inline
 7.38% :     0.000023s :      2: substitution.inline_without_move
 2.37% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.52% :     0.000005s :      2: substitution.minmaximum_grad
 1.56% :     0.000005s :      3: substitution.partial_eliminate
 2.53% :     0.000008s :      9: substitution.remove_not_recompute_node
 2.05% :     0.000006s :      2: substitution.replace_applicator
 1.57% :     0.000005s :      5: substitution.replace_old_param
 0.86% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.26% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.51% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.07% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010741     2
88.00% :     0.009452s :      1: type_inference.infer
12.00% :     0.001289s :      1: type_inference.specialize
------[replace.]   0.000069     8
78.21% :     0.000054s :      6: replace.inline
21.79% :     0.000015s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000221     8
95.04% :     0.000210s :      6: match.inline
 4.96% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000178  1109
 1.65% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.67% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.43% :     0.000003s :     18: predicate.addn_check_dump
 1.63% :     0.000003s :     18: predicate.addn_zero_filter
 2.36% :     0.000004s :     18: predicate.arithmetic_simplify
 1.43% :     0.000003s :     18: predicate.cast_eliminate
 0.19% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.30% :     0.000002s :     18: predicate.compare_switch_simplify
 1.41% :     0.000003s :     18: predicate.depend_value_elim
 1.21% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.46% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.53% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.16% :     0.000000s :      1: predicate.elim_not_effective
 0.30% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.82% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.24% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.23% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.41% :     0.000003s :     18: predicate.environ_get_eliminate
 1.25% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.07% :     0.000000s :      1: predicate.fold_const_symbol
 1.14% :     0.000002s :     12: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 4.24% :     0.000008s :     28: predicate.inline
 3.19% :     0.000006s :     33: predicate.inline_without_move
 0.47% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.44% :     0.000003s :     12: predicate.less_batch_normalization
 1.88% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.68% :     0.000003s :     21: predicate.load_eliminater
 0.75% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.46% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.88% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.33% :     0.000002s :     18: predicate.merge_addn
 1.43% :     0.000003s :     18: predicate.minmaximum_grad
 0.79% :     0.000001s :      1: predicate.mutable_eliminate
 0.17% :     0.000000s :      1: predicate.opt_reshape
 1.91% :     0.000003s :     21: predicate.partial_eliminate
 1.56% :     0.000003s :     18: predicate.print_const_string_wrapper
 1.94% :     0.000003s :     18: predicate.reduce_eliminate
 1.61% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.67% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.53% :     0.000003s :     22: predicate.replace_applicator
 1.83% :     0.000003s :     33: predicate.replace_old_param
 0.12% :     0.000000s :      1: predicate.reset_defer_inline
 1.74% :     0.000003s :     18: predicate.reshape_eliminate
 1.53% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     18: predicate.same_eliminate
 0.59% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.42% :     0.000001s :      2: predicate.special_op_eliminate
 1.17% :     0.000002s :     12: predicate.specialize_transform
 1.56% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.11% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.63% :     0.000005s :     26: predicate.switch_defer_inline
 2.27% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.40% :     0.000011s :     71: predicate.switch_simplify
 1.45% :     0.000003s :     18: predicate.tile_eliminate
 1.37% :     0.000002s :     18: predicate.transpose_eliminate
 1.61% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.37% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 3.11% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.67% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.61% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.83% :     0.000003s :     18: predicate.value_based_eliminate
 0.12% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.44% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001341    22
71.10% :     0.000953s :     13: func_graph_cloner_run.FuncGraphClonerGraph
28.90% :     0.000387s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026778    76
 0.14% :     0.000036s :      1: add_recomputation
 0.38% :     0.000100s :      1: auto_monad
 0.06% :     0.000016s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.90% :     0.000242s :      1: bootstrap
 0.10% :     0.000026s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.21% :     0.000057s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
36.42% :     0.009752s :      1: jit_opt_a
 0.46% :     0.000124s :      1: jit_opt_after_cconv
 0.17% :     0.000047s :      1: jit_opt_b
 1.70% :     0.000456s :      1: loop_unroll
 2.49% :     0.000666s :      1: mutable_eliminate
 5.02% :     0.001344s :     26: opt.transform.jit_opt_a
 0.11% :     0.000030s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000018s :      4: opt.transform.jit_opt_b
 0.04% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000017s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000025s :      4: opt.transform.symbol_engine_opt
 1.75% :     0.000468s :      1: opt_after_jit_grad
 0.02% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000033s :      1: py_interpret_to_execute
 0.05% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000016s :      1: remove_dup_value
 5.73% :     0.001535s :      1: renormalize.infer
 1.98% :     0.000529s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000036s :      1: rewriter_after_opt_a
 0.38% :     0.000102s :      1: rewriter_before_opt_a
 0.28% :     0.000074s :      1: symbol_engine_optimizer
 0.11% :     0.000028s :      1: task_emit
40.45% :     0.010831s :      1: type_inference
 0.17% :     0.000045s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.758.063 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c375280> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.758.742 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0191326, [30]
[bootstrap]: 0.00027979
[type_inference]: 0.00403883
[event_method]: 1.58e-05
[auto_monad]: 6.611e-05
[graph_reusing]: 5.69e-06
[pre_auto_parallel]: 2.36998e-06
[py_interpret_to_execute]: 2.805e-05
[rewriter_before_opt_a]: 7.145e-05
[expand_dump_flag]: 3.14001e-06
[jit_opt_a]: 0.011691, [2]
    [Cycle 1]: 0.00181014, [27]
        [switch_simplify]: 6.456e-05
        [loop_unroll]: 2.808e-05
        [a_1]: 0.0005978
        [with_stream_mark]: 2.482e-05
        [recompute_prepare]: 8.23999e-06
        [updatestate_depend_eliminate]: 4.03999e-06
        [updatestate_assign_eliminate]: 3.08e-06
        [updatestate_loads_eliminate]: 2.69001e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 6.13998e-06
        [updatestate_useless_node_eliminater]: 5.19003e-06
        [accelerated_algorithm]: 5.47001e-06
        [meta_shard_fg_expand]: 2.36e-06
        [get_grad_eliminate_]: 5.05999e-06
        [merge_forward]: 3.59002e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.54e-05
        [j_node_and_user_rematch]: 9.51e-06
        [meta_fg_expand]: 2.07001e-06
        [replace_old_param]: 9.72001e-06
        [inline_without_move]: 5.63002e-06
        [renormalize]: 0.00072999
        [add_forward_monad_depend]: 6.72002e-06
        [auto_monad_grad]: 3.25002e-06
        [auto_monad_eliminator]: 1.564e-05
        [cse]: 3.701e-05
        [replace_applicator]: 1.496e-05
    [Cycle 2]: 0.00032751, [27]
        [switch_simplify]: 5.93002e-06
        [loop_unroll]: 4.65999e-06
        [a_1]: 9.068e-05
        [with_stream_mark]: 1.092e-05
        [recompute_prepare]: 4.92e-06
        [updatestate_depend_eliminate]: 2.79001e-06
        [updatestate_assign_eliminate]: 2.23002e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [parameter_eliminate]: 9.30013e-07
        [specialize_transform]: 4.84e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.13002e-06
        [meta_shard_fg_expand]: 1.65001e-06
        [get_grad_eliminate_]: 4.28001e-06
        [merge_forward]: 2.93e-06
        [cell_reuse_recompute_pass]: 3.22002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.333e-05
        [j_node_and_user_rematch]: 7.36999e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 7.71999e-06
        [inline_without_move]: 4.70001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.87999e-06
        [auto_monad_grad]: 6.59988e-07
        [auto_monad_eliminator]: 6.36e-06
        [cse]: 1.124e-05
        [replace_applicator]: 5.02e-06
[py_interpret_to_execute_after_opt_a]: 1.37e-05
[rewriter_after_opt_a]: 4.339e-05
[convert_after_rewriter]: 8.39002e-06
[order_py_execute_after_rewriter]: 5.28002e-06
[mutable_eliminate]: 0.00090822
[jit_opt_b]: 5.655e-05, [1]
    [Cycle 1]: 4.765e-05, [2]
        [frontend_op_eliminate]: 1.837e-05
        [inline_after_opt_a]: 1.652e-05
[cconv]: 3.234e-05
[loop_unroll]: 0.00058024
[jit_opt_after_cconv]: 0.00017553, [1]
    [Cycle 1]: 0.00016613, [11]
        [c_1]: 2.198e-05
        [parameter_eliminate]: 5.46998e-06
        [updatestate_depend_eliminate]: 9.12001e-06
        [updatestate_assign_eliminate]: 2.63e-06
        [updatestate_loads_eliminate]: 2.23998e-06
        [cse]: 2.919e-05
        [call_graph_tuple_transform]: 2.37e-05
        [tuple_list_get_item_eliminator]: 5.74e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 9.49978e-07
        [switch_simplify]: 6.74001e-06
[remove_dup_value]: 1.834e-05
[partial_unused_args_eliminate]: 2.29001e-06
[environ_conv]: 6.44001e-06
[add_recomputation]: 5.973e-05
[cse_after_recomputation]: 2.603e-05, [1]
    [Cycle 1]: 1.985e-05, [1]
        [cse]: 1.116e-05
[auto_monad_reorder]: 1.952e-05
[get_jit_bprop_graph]: 2.22999e-06
[rewriter_after_jit_bprop_graph]: 1.011e-05
[opt_after_jit_grad]: 0.00063838
[symbol_engine_optimizer]: 8.728e-05, [1]
    [Cycle 1]: 7.853e-05, [6]
        [build]: 6.56e-06
        [elim_shapecalc]: 9.76e-06
        [elim_not_effective]: 1.585e-05
        [opt_reshape]: 6.34999e-06
        [fold_const_symbol]: 9.00001e-06
        [renormalize]: 2.60014e-07
[validate]: 4.771e-05
Sums
    bootstrap                            :     0.000280s :  3.13%
    type_inference                       :     0.004039s : 45.19%
    event_method                         :     0.000016s :  0.18%
    auto_monad                           :     0.000066s :  0.74%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000028s :  0.31%
    rewriter_before_opt_a                :     0.000071s :  0.80%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000070s :  0.79%
    jit_opt_a.loop_unroll                :     0.000033s :  0.37%
    jit_opt_a.a_1                        :     0.000688s :  7.70%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.40%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000005s :  0.05%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000730s :  8.17%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.25%
    jit_opt_a.cse                        :     0.000048s :  0.54%
    jit_opt_a.replace_applicator         :     0.000020s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.15%
    rewriter_after_opt_a                 :     0.000043s :  0.49%
    convert_after_rewriter               :     0.000008s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000908s : 10.16%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.18%
    cconv                                :     0.000032s :  0.36%
    loop_unroll                          :     0.000580s :  6.49%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.10%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000029s :  0.33%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.27%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.08%
    remove_dup_value                     :     0.000018s :  0.21%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000060s :  0.67%
    cse_after_recomputation.cse          :     0.000011s :  0.12%
    auto_monad_reorder                   :     0.000020s :  0.22%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.11%
    opt_after_jit_grad                   :     0.000638s :  7.14%
    symbol_engine_optimizer.build        :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.18%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000048s :  0.53%


Time group info:
------[substitution.]   0.000241    23
 1.19% :     0.000003s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.67% :     0.000006s :      3: substitution.graph_param_transform
84.68% :     0.000204s :      4: substitution.inline
 1.29% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.06% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.10% :     0.000005s :      2: substitution.replace_old_param
 5.43% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003972     2
80.37% :     0.003192s :      1: type_inference.infer
19.63% :     0.000780s :      1: type_inference.specialize
------[replace.]   0.000072     6
71.63% :     0.000051s :      4: replace.inline
28.37% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000214     6
94.39% :     0.000202s :      4: match.inline
 5.61% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.29% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.87% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.18% :     0.000002s :     10: predicate.addn_zero_filter
 2.01% :     0.000003s :     10: predicate.arithmetic_simplify
 1.21% :     0.000002s :     10: predicate.cast_eliminate
 0.66% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.53% :     0.000002s :     10: predicate.depend_value_elim
 1.04% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.38% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.66% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.98% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.39% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.79% :     0.000001s :      6: predicate.get_grad_eliminate
 0.43% :     0.000001s :      3: predicate.graph_param_transform
 5.36% :     0.000007s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.65% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.53% :     0.000002s :     15: predicate.load_eliminater
 2.63% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.34% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 2.45% :     0.000003s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000003s :     15: predicate.partial_eliminate
 1.51% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.78% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.43% :     0.000001s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.06% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.04% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     10: predicate.same_eliminate
 0.43% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.40% :     0.000002s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.25% :     0.000003s :     16: predicate.switch_defer_inline
 1.86% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.07% :     0.000009s :     53: predicate.switch_simplify
 1.24% :     0.000002s :     10: predicate.tile_eliminate
 1.09% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.11% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.79% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.54% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.47% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.81% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000542    12
48.26% :     0.000262s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.11% :     0.000039s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.63% :     0.000242s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020839    72
 0.30% :     0.000063s :      1: add_recomputation
 0.33% :     0.000069s :      1: auto_monad
 0.11% :     0.000023s :      1: auto_monad_reorder
 1.41% :     0.000294s :      1: bootstrap
 0.17% :     0.000035s :      1: cconv
 0.05% :     0.000011s :      1: convert_after_rewriter
 0.14% :     0.000029s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.10% :     0.000020s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
56.12% :     0.011695s :      1: jit_opt_a
 0.86% :     0.000179s :      1: jit_opt_after_cconv
 0.29% :     0.000059s :      1: jit_opt_b
 2.84% :     0.000591s :      1: loop_unroll
 4.42% :     0.000921s :      1: mutable_eliminate
 4.36% :     0.000909s :     26: opt.transform.jit_opt_a
 0.26% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.13% :     0.000028s :      4: opt.transform.jit_opt_b
 0.08% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.12% :     0.000650s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.15% :     0.000031s :      1: py_interpret_to_execute
 0.08% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000021s :      1: remove_dup_value
 1.91% :     0.000399s :      1: renormalize.infer
 1.55% :     0.000322s :      1: renormalize.specialize
 0.06% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.23% :     0.000049s :      1: rewriter_after_opt_a
 0.36% :     0.000074s :      1: rewriter_before_opt_a
 0.44% :     0.000091s :      1: symbol_engine_optimizer
19.44% :     0.004052s :      1: type_inference


TotalTime = 0.0197823, [30]
[bootstrap]: 0.00017133
[type_inference]: 0.00899342
[event_method]: 4.152e-05
[auto_monad]: 7.399e-05
[graph_reusing]: 5.11997e-06
[pre_auto_parallel]: 1.97001e-06
[py_interpret_to_execute]: 2.553e-05
[rewriter_before_opt_a]: 8.339e-05
[expand_dump_flag]: 2.32999e-06
[jit_opt_a]: 0.00840277, [2]
    [Cycle 1]: 0.00486328, [27]
        [switch_simplify]: 5.329e-05
        [loop_unroll]: 3.866e-05
        [a_1]: 0.00089673
        [with_stream_mark]: 2.401e-05
        [recompute_prepare]: 1.996e-05
        [updatestate_depend_eliminate]: 7.48e-06
        [updatestate_assign_eliminate]: 5.96e-06
        [updatestate_loads_eliminate]: 5.61e-06
        [parameter_eliminate]: 2.61999e-06
        [specialize_transform]: 1.317e-05
        [updatestate_useless_node_eliminater]: 1.181e-05
        [accelerated_algorithm]: 1.222e-05
        [meta_shard_fg_expand]: 3.86999e-06
        [get_grad_eliminate_]: 1.203e-05
        [merge_forward]: 7.87998e-06
        [cell_reuse_recompute_pass]: 1.41002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.533e-05
        [j_node_and_user_rematch]: 2.336e-05
        [meta_fg_expand]: 0.00182598
        [replace_old_param]: 5.448e-05
        [inline_without_move]: 5.318e-05
        [renormalize]: 0.00149287
        [add_forward_monad_depend]: 5.91e-06
        [auto_monad_grad]: 2.22001e-06
        [auto_monad_eliminator]: 1.008e-05
        [cse]: 1.468e-05
        [replace_applicator]: 1.147e-05
    [Cycle 2]: 0.00027204, [27]
        [switch_simplify]: 4.62998e-06
        [loop_unroll]: 3.73999e-06
        [a_1]: 3.806e-05
        [with_stream_mark]: 9.65002e-06
        [recompute_prepare]: 3.78999e-06
        [updatestate_depend_eliminate]: 2.01e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.32999e-06
        [specialize_transform]: 3.72998e-06
        [updatestate_useless_node_eliminater]: 3.27002e-06
        [accelerated_algorithm]: 3.96001e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 3.51999e-06
        [merge_forward]: 1.82001e-06
        [cell_reuse_recompute_pass]: 1.74e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.12e-05
        [j_node_and_user_rematch]: 5.52999e-06
        [meta_fg_expand]: 2.394e-05
        [replace_old_param]: 3.63999e-06
        [inline_without_move]: 3.23e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.47999e-06
        [auto_monad_grad]: 1.22e-06
        [auto_monad_eliminator]: 4.10998e-06
        [cse]: 9.76e-06
        [replace_applicator]: 4.69998e-06
[py_interpret_to_execute_after_opt_a]: 9.90002e-06
[rewriter_after_opt_a]: 3.198e-05
[convert_after_rewriter]: 6.23002e-06
[order_py_execute_after_rewriter]: 4.93001e-06
[mutable_eliminate]: 0.00055119
[jit_opt_b]: 4.157e-05, [1]
    [Cycle 1]: 3.387e-05, [2]
        [frontend_op_eliminate]: 1.084e-05
        [inline_after_opt_a]: 1.134e-05
[cconv]: 2.043e-05
[loop_unroll]: 0.00042394
[jit_opt_after_cconv]: 0.00011672, [1]
    [Cycle 1]: 0.00011026, [11]
        [c_1]: 1.195e-05
        [parameter_eliminate]: 2.74001e-06
        [updatestate_depend_eliminate]: 4.67e-06
        [updatestate_assign_eliminate]: 1.96e-06
        [updatestate_loads_eliminate]: 1.69e-06
        [cse]: 1.356e-05
        [call_graph_tuple_transform]: 1.275e-05
        [tuple_list_get_item_eliminator]: 4.18001e-06
        [none_parameter_eliminate]: 1.36998e-06
        [renormalize]: 8.70001e-07
        [switch_simplify]: 3.73999e-06
[remove_dup_value]: 1.232e-05
[partial_unused_args_eliminate]: 2.28002e-06
[environ_conv]: 3.83001e-06
[add_recomputation]: 2.873e-05
[cse_after_recomputation]: 1.914e-05, [1]
    [Cycle 1]: 1.399e-05, [1]
        [cse]: 7.7e-06
[auto_monad_reorder]: 9.92999e-06
[get_jit_bprop_graph]: 1.69e-06
[rewriter_after_jit_bprop_graph]: 4.20999e-06
[opt_after_jit_grad]: 0.00044409
[symbol_engine_optimizer]: 6.362e-05, [1]
    [Cycle 1]: 5.767e-05, [6]
        [build]: 2.86e-06
        [elim_shapecalc]: 6.38e-06
        [elim_not_effective]: 9.81e-06
        [opt_reshape]: 3.97e-06
        [fold_const_symbol]: 5.99e-06
        [renormalize]: 6.19999e-07
[validate]: 2.289e-05
Sums
    bootstrap                            :     0.000171s :  1.08%
    type_inference                       :     0.008993s : 56.60%
    event_method                         :     0.000042s :  0.26%
    auto_monad                           :     0.000074s :  0.47%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.16%
    rewriter_before_opt_a                :     0.000083s :  0.52%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000058s :  0.36%
    jit_opt_a.loop_unroll                :     0.000042s :  0.27%
    jit_opt_a.a_1                        :     0.000935s :  5.88%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000024s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.10%
    jit_opt_a.merge_forward              :     0.000010s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001850s : 11.64%
    jit_opt_a.replace_old_param          :     0.000058s :  0.37%
    jit_opt_a.inline_without_move        :     0.000056s :  0.36%
    jit_opt_a.renormalize                :     0.001493s :  9.40%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.09%
    jit_opt_a.cse                        :     0.000024s :  0.15%
    jit_opt_a.replace_applicator         :     0.000016s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000032s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000551s :  3.47%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000020s :  0.13%
    loop_unroll                          :     0.000424s :  2.67%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000014s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000012s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000029s :  0.18%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000010s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000444s :  2.80%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000023s :  0.14%


TotalTime = 0.0326496, [33]
[bootstrap]: 0.00020809
[type_inference]: 0.0104868
[event_method]: 4.631e-05
[auto_monad]: 9.36e-05
[graph_reusing]: 7.20998e-06
[pre_auto_parallel]: 3.01001e-06
[py_interpret_to_execute]: 2.591e-05
[rewriter_before_opt_a]: 9.077e-05
[expand_dump_flag]: 3.49001e-06
[jit_opt_a]: 0.0117606, [3]
    [Cycle 1]: 0.00579929, [27]
        [switch_simplify]: 6.066e-05
        [loop_unroll]: 4e-05
        [a_1]: 0.00082883
        [with_stream_mark]: 2.219e-05
        [recompute_prepare]: 1.872e-05
        [updatestate_depend_eliminate]: 8.04002e-06
        [updatestate_assign_eliminate]: 6.19001e-06
        [updatestate_loads_eliminate]: 5.51e-06
        [parameter_eliminate]: 2.37999e-06
        [specialize_transform]: 1.377e-05
        [updatestate_useless_node_eliminater]: 1.251e-05
        [accelerated_algorithm]: 1.332e-05
        [meta_shard_fg_expand]: 2.83003e-06
        [get_grad_eliminate_]: 1.247e-05
        [merge_forward]: 7.4e-06
        [cell_reuse_recompute_pass]: 8.40024e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.457e-05
        [j_node_and_user_rematch]: 2.388e-05
        [meta_fg_expand]: 0.00144231
        [replace_old_param]: 5.43e-05
        [inline_without_move]: 5.284e-05
        [renormalize]: 0.00265162
        [add_forward_monad_depend]: 9.87999e-06
        [auto_monad_grad]: 5.97999e-06
        [auto_monad_eliminator]: 5.211e-05
        [cse]: 0.00015376
        [replace_applicator]: 6.658e-05
    [Cycle 2]: 0.00212164, [27]
        [switch_simplify]: 3.699e-05
        [loop_unroll]: 3.589e-05
        [a_1]: 0.00112521
        [with_stream_mark]: 1.573e-05
        [recompute_prepare]: 9.59999e-06
        [updatestate_depend_eliminate]: 4.52998e-06
        [updatestate_assign_eliminate]: 3.66001e-06
        [updatestate_loads_eliminate]: 3.25e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 7.38e-06
        [updatestate_useless_node_eliminater]: 7.29001e-06
        [accelerated_algorithm]: 7.9e-06
        [meta_shard_fg_expand]: 2.73e-06
        [get_grad_eliminate_]: 7.00002e-06
        [merge_forward]: 4.58001e-06
        [cell_reuse_recompute_pass]: 1.42e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.916e-05
        [j_node_and_user_rematch]: 1.317e-05
        [meta_fg_expand]: 4.386e-05
        [replace_old_param]: 1.221e-05
        [inline_without_move]: 6.87002e-06
        [renormalize]: 0.00054294
        [add_forward_monad_depend]: 4.75001e-06
        [auto_monad_grad]: 1.20999e-06
        [auto_monad_eliminator]: 1.371e-05
        [cse]: 2.423e-05
        [replace_applicator]: 1.345e-05
    [Cycle 3]: 0.0004176, [27]
        [switch_simplify]: 7.3e-06
        [loop_unroll]: 6.96999e-06
        [a_1]: 0.00014438
        [with_stream_mark]: 9.42999e-06
        [recompute_prepare]: 6.96001e-06
        [updatestate_depend_eliminate]: 4.22e-06
        [updatestate_assign_eliminate]: 3.26001e-06
        [updatestate_loads_eliminate]: 3.03e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 6.95002e-06
        [updatestate_useless_node_eliminater]: 6.86999e-06
        [accelerated_algorithm]: 7.08e-06
        [meta_shard_fg_expand]: 1.55999e-06
        [get_grad_eliminate_]: 6.63e-06
        [merge_forward]: 3.3e-06
        [cell_reuse_recompute_pass]: 1.62999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.785e-05
        [j_node_and_user_rematch]: 1.214e-05
        [meta_fg_expand]: 2.66999e-06
        [replace_old_param]: 9.51e-06
        [inline_without_move]: 6.60997e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.20001e-06
        [auto_monad_grad]: 1.38002e-06
        [auto_monad_eliminator]: 7.85e-06
        [cse]: 1.709e-05
        [replace_applicator]: 7.04001e-06
[py_interpret_to_execute_after_opt_a]: 1.06e-05
[rewriter_after_opt_a]: 4.202e-05
[convert_after_rewriter]: 8.15999e-06
[order_py_execute_after_rewriter]: 5.86e-06
[mutable_eliminate]: 0.00062682
[jit_opt_b]: 5.991e-05, [1]
    [Cycle 1]: 5.294e-05, [2]
        [frontend_op_eliminate]: 2.192e-05
        [inline_after_opt_a]: 1.958e-05
[cconv]: 2.167e-05
[loop_unroll]: 0.00040286
[jit_opt_after_cconv]: 0.00016354, [1]
    [Cycle 1]: 0.00015689, [11]
        [c_1]: 3e-05
        [parameter_eliminate]: 2.74999e-06
        [updatestate_depend_eliminate]: 6.68e-06
        [updatestate_assign_eliminate]: 3.26001e-06
        [updatestate_loads_eliminate]: 2.99001e-06
        [cse]: 2.355e-05
        [call_graph_tuple_transform]: 2.245e-05
        [tuple_list_get_item_eliminator]: 7.44002e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 6.60017e-07
        [switch_simplify]: 6.94001e-06
[remove_dup_value]: 1.763e-05
[partial_unused_args_eliminate]: 2.74001e-06
[environ_conv]: 5.99e-06
[add_recomputation]: 5.04e-05
[cse_after_recomputation]: 2.884e-05, [1]
    [Cycle 1]: 2.318e-05, [1]
        [cse]: 1.735e-05
[auto_monad_reorder]: 1.91e-05
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 5.14e-06
[opt_after_jit_grad]: 0.00042874
[symbol_engine_optimizer]: 8.549e-05, [1]
    [Cycle 1]: 7.937e-05, [6]
        [build]: 3.53999e-06
        [elim_shapecalc]: 1.001e-05
        [elim_not_effective]: 1.703e-05
        [opt_reshape]: 7.85e-06
        [fold_const_symbol]: 1.247e-05
        [renormalize]: 6.49976e-07
[validate]: 4.21e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00769236
[execute]: 6.91001e-06
Sums
    bootstrap                            :     0.000208s :  0.73%
    type_inference                       :     0.010487s : 36.87%
    event_method                         :     0.000046s :  0.16%
    auto_monad                           :     0.000094s :  0.33%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.09%
    rewriter_before_opt_a                :     0.000091s :  0.32%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000105s :  0.37%
    jit_opt_a.loop_unroll                :     0.000083s :  0.29%
    jit_opt_a.a_1                        :     0.002098s :  7.38%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000035s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000028s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000027s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000028s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000026s :  0.09%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000062s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000049s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001489s :  5.23%
    jit_opt_a.replace_old_param          :     0.000076s :  0.27%
    jit_opt_a.inline_without_move        :     0.000066s :  0.23%
    jit_opt_a.renormalize                :     0.003195s : 11.23%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000074s :  0.26%
    jit_opt_a.cse                        :     0.000195s :  0.69%
    jit_opt_a.replace_applicator         :     0.000087s :  0.31%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.04%
    rewriter_after_opt_a                 :     0.000042s :  0.15%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000627s :  2.20%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.07%
    cconv                                :     0.000022s :  0.08%
    loop_unroll                          :     0.000403s :  1.42%
    jit_opt_after_cconv.c_1              :     0.000030s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000018s :  0.06%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000050s :  0.18%
    cse_after_recomputation.cse          :     0.000017s :  0.06%
    auto_monad_reorder                   :     0.000019s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000429s :  1.51%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007692s : 27.04%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000249    50
 0.55% :     0.000001s :      1: substitution.elim_not_effective
 0.29% :     0.000001s :      1: substitution.fold_const_symbol
 1.27% :     0.000003s :      1: substitution.graph_param_transform
68.32% :     0.000170s :      6: substitution.inline
 6.87% :     0.000017s :      2: substitution.inline_without_move
 2.30% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.37% :     0.000003s :      2: substitution.minmaximum_grad
 1.44% :     0.000004s :      3: substitution.partial_eliminate
 2.85% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.11% :     0.000005s :      2: substitution.replace_applicator
 1.73% :     0.000004s :      5: substitution.replace_old_param
 1.03% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.61% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.91% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.37% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008941     2
89.22% :     0.007977s :      1: type_inference.infer
10.78% :     0.000964s :      1: type_inference.specialize
------[replace.]   0.000064     8
79.07% :     0.000050s :      6: replace.inline
20.93% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000174     8
95.49% :     0.000166s :      6: match.inline
 4.51% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000169  1109
 1.45% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.58% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.40% :     0.000002s :     18: predicate.addn_check_dump
 1.44% :     0.000002s :     18: predicate.addn_zero_filter
 2.28% :     0.000004s :     18: predicate.arithmetic_simplify
 1.47% :     0.000002s :     18: predicate.cast_eliminate
 0.15% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.26% :     0.000002s :     18: predicate.compare_switch_simplify
 1.48% :     0.000002s :     18: predicate.depend_value_elim
 1.27% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.57% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.40% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.42% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.14% :     0.000000s :      1: predicate.elim_not_effective
 0.22% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.36% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.36% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.39% :     0.000002s :     18: predicate.environ_get_eliminate
 1.32% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.30% :     0.000002s :     12: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 4.32% :     0.000007s :     28: predicate.inline
 3.41% :     0.000006s :     33: predicate.inline_without_move
 0.55% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.51% :     0.000003s :     12: predicate.less_batch_normalization
 1.56% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.61% :     0.000003s :     21: predicate.load_eliminater
 0.62% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.74% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.82% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.29% :     0.000002s :     18: predicate.merge_addn
 1.39% :     0.000002s :     18: predicate.minmaximum_grad
 0.78% :     0.000001s :      1: predicate.mutable_eliminate
 0.19% :     0.000000s :      1: predicate.opt_reshape
 2.17% :     0.000004s :     21: predicate.partial_eliminate
 1.39% :     0.000002s :     18: predicate.print_const_string_wrapper
 2.03% :     0.000003s :     18: predicate.reduce_eliminate
 1.48% :     0.000002s :     20: predicate.redundant_stop_gradient_eliminater
 0.82% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.44% :     0.000002s :     22: predicate.replace_applicator
 1.66% :     0.000003s :     33: predicate.replace_old_param
 0.15% :     0.000000s :      1: predicate.reset_defer_inline
 1.49% :     0.000003s :     18: predicate.reshape_eliminate
 1.45% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.48% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.45% :     0.000002s :     18: predicate.same_eliminate
 0.65% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.31% :     0.000001s :      2: predicate.special_op_eliminate
 1.29% :     0.000002s :     12: predicate.specialize_transform
 1.90% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.15% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.91% :     0.000005s :     26: predicate.switch_defer_inline
 2.47% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.58% :     0.000011s :     71: predicate.switch_simplify
 1.47% :     0.000002s :     18: predicate.tile_eliminate
 1.45% :     0.000002s :     18: predicate.transpose_eliminate
 1.75% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.45% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 2.69% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.58% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000003s :     18: predicate.value_based_eliminate
 0.14% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.27% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001206    22
77.36% :     0.000933s :     13: func_graph_cloner_run.FuncGraphClonerGraph
22.64% :     0.000273s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022594    72
 0.14% :     0.000032s :      1: add_recomputation
 0.34% :     0.000077s :      1: auto_monad
 0.05% :     0.000012s :      1: auto_monad_reorder
 0.81% :     0.000183s :      1: bootstrap
 0.10% :     0.000023s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.20% :     0.000045s :      1: event_method
 0.08% :     0.000017s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
37.21% :     0.008406s :      1: jit_opt_a
 0.53% :     0.000119s :      1: jit_opt_after_cconv
 0.20% :     0.000045s :      1: jit_opt_b
 1.91% :     0.000431s :      1: loop_unroll
 2.48% :     0.000559s :      1: mutable_eliminate
 5.71% :     0.001291s :     26: opt.transform.jit_opt_a
 0.13% :     0.000029s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000014s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000023s :      4: opt.transform.symbol_engine_opt
 2.00% :     0.000451s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000028s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000015s :      1: remove_dup_value
 4.94% :     0.001116s :      1: renormalize.infer
 1.63% :     0.000368s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000035s :      1: rewriter_after_opt_a
 0.38% :     0.000086s :      1: rewriter_before_opt_a
 0.29% :     0.000066s :      1: symbol_engine_optimizer
39.84% :     0.009002s :      1: type_inference


Time group info:
------[substitution.]   0.000501   118
 5.36% :     0.000027s :      1: substitution.arithmetic_simplify
 0.53% :     0.000003s :      4: substitution.elim_not_effective
 0.36% :     0.000002s :      4: substitution.fold_const_symbol
 1.19% :     0.000006s :      5: substitution.graph_param_transform
58.75% :     0.000294s :     12: substitution.inline
 3.44% :     0.000017s :      2: substitution.inline_without_move
 2.02% :     0.000010s :     16: substitution.j_node_and_user_rematch
 2.19% :     0.000011s :      7: substitution.minmaximum_grad
 1.42% :     0.000007s :      5: substitution.partial_eliminate
 2.56% :     0.000013s :     16: substitution.remove_not_recompute_node
 5.26% :     0.000026s :     10: substitution.replace_applicator
 1.64% :     0.000008s :      7: substitution.replace_old_param
 0.60% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.08% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.72% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.86% :     0.000039s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010415     2
87.87% :     0.009152s :      1: type_inference.infer
12.13% :     0.001263s :      1: type_inference.specialize
------[replace.]   0.000140    20
 3.88% :     0.000005s :      1: replace.arithmetic_simplify
61.78% :     0.000087s :     12: replace.inline
34.34% :     0.000048s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000334    20
 7.77% :     0.000026s :      1: match.arithmetic_simplify
86.18% :     0.000288s :     12: match.inline
 6.05% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000395  2790
 1.62% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.42% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.44% :     0.000006s :     46: predicate.addn_check_dump
 1.70% :     0.000007s :     46: predicate.addn_zero_filter
 2.11% :     0.000008s :     47: predicate.arithmetic_simplify
 1.77% :     0.000007s :     47: predicate.cast_eliminate
 0.18% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.39% :     0.000005s :     46: predicate.compare_switch_simplify
 1.86% :     0.000007s :     46: predicate.depend_value_elim
 1.46% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.55% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.51% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.33% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.14% :     0.000001s :      5: predicate.elim_not_effective
 0.25% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.51% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.52% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.49% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.51% :     0.000006s :     47: predicate.environ_get_eliminate
 1.46% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      5: predicate.fold_const_symbol
 0.87% :     0.000003s :     21: predicate.get_grad_eliminate
 0.13% :     0.000001s :      5: predicate.graph_param_transform
 4.36% :     0.000017s :     76: predicate.inline
 1.71% :     0.000007s :     42: predicate.inline_without_move
 0.36% :     0.000001s :     21: predicate.j_node_and_user_rematch
 1.09% :     0.000004s :     21: predicate.less_batch_normalization
 1.76% :     0.000007s :     54: predicate.list_to_tuple_eliminator_
 1.96% :     0.000008s :     59: predicate.load_eliminater
 0.46% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.11% :     0.000012s :     89: predicate.loop_unroll_before_grad
 1.85% :     0.000007s :     52: predicate.make_slice_get_slice_eliminator
 1.41% :     0.000006s :     46: predicate.merge_addn
 1.68% :     0.000007s :     47: predicate.minmaximum_grad
 0.54% :     0.000002s :      5: predicate.mutable_eliminate
 0.22% :     0.000001s :      5: predicate.opt_reshape
 2.43% :     0.000010s :     59: predicate.partial_eliminate
 1.47% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.87% :     0.000007s :     47: predicate.reduce_eliminate
 1.78% :     0.000007s :     54: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.56% :     0.000010s :    105: predicate.replace_applicator
 0.95% :     0.000004s :     42: predicate.replace_old_param
 0.12% :     0.000000s :      5: predicate.reset_defer_inline
 1.72% :     0.000007s :     47: predicate.reshape_eliminate
 1.51% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.29% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.46% :     0.000006s :     46: predicate.same_eliminate
 0.48% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.39% :     0.000002s :     10: predicate.special_op_eliminate
 0.90% :     0.000004s :     21: predicate.specialize_transform
 1.60% :     0.000006s :     46: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.20% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.60% :     0.000010s :     66: predicate.switch_defer_inline
 2.49% :     0.000010s :     66: predicate.switch_layer_defer_inline
 6.07% :     0.000024s :    160: predicate.switch_simplify
 1.58% :     0.000006s :     47: predicate.tile_eliminate
 1.48% :     0.000006s :     47: predicate.transpose_eliminate
 1.82% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.68% :     0.000007s :     47: predicate.tuple_list_get_item_depend_reorder
 3.03% :     0.000012s :     64: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000007s :     54: predicate.tuple_to_list_eliminator_
 1.96% :     0.000008s :     59: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000012s :     80: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000007s :     46: predicate.value_based_eliminate
 0.17% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.32% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001580    29
56.30% :     0.000890s :     13: func_graph_cloner_run.FuncGraphClonerGraph
43.70% :     0.000690s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038723    91
 0.14% :     0.000053s :      1: add_recomputation
 0.25% :     0.000098s :      1: auto_monad
 0.06% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.56% :     0.000218s :      1: bootstrap
 0.06% :     0.000024s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000031s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.13% :     0.000051s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
30.38% :     0.011764s :      1: jit_opt_a
 0.45% :     0.000173s :      1: jit_opt_after_cconv
 0.16% :     0.000062s :      1: jit_opt_b
 1.06% :     0.000411s :      1: loop_unroll
 1.64% :     0.000636s :      1: mutable_eliminate
 7.04% :     0.002728s :     39: opt.transform.jit_opt_a
 0.16% :     0.000064s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000036s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000044s :      4: opt.transform.symbol_engine_opt
 1.13% :     0.000436s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000029s :      1: py_interpret_to_execute
 0.03% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000020s :      1: remove_dup_value
 4.33% :     0.001678s :      2: renormalize.infer
 3.88% :     0.001504s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000045s :      1: rewriter_after_opt_a
 0.24% :     0.000093s :      1: rewriter_before_opt_a
 0.23% :     0.000088s :      1: symbol_engine_optimizer
19.89% :     0.007702s :      1: task_emit
27.12% :     0.010502s :      1: type_inference
 0.17% :     0.000065s :      1: validate

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.773.747 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843cd310> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.774.282 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.775.156 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e31f0> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.775.750 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.023207, [33]
[bootstrap]: 0.00020645
[type_inference]: 0.0105911
[event_method]: 5.066e-05
[auto_monad]: 9.231e-05
[graph_reusing]: 7.46999e-06
[pre_auto_parallel]: 2.89999e-06
[py_interpret_to_execute]: 2.75e-05
[rewriter_before_opt_a]: 9.002e-05
[expand_dump_flag]: 3.31999e-06
[jit_opt_a]: 0.00950242, [2]
    [Cycle 1]: 0.00558837, [27]
        [switch_simplify]: 6.305e-05
        [loop_unroll]: 3.693e-05
        [a_1]: 0.00087051
        [with_stream_mark]: 3.034e-05
        [recompute_prepare]: 2.143e-05
        [updatestate_depend_eliminate]: 8.38999e-06
        [updatestate_assign_eliminate]: 6.21998e-06
        [updatestate_loads_eliminate]: 6.02999e-06
        [parameter_eliminate]: 3.51001e-06
        [specialize_transform]: 1.388e-05
        [updatestate_useless_node_eliminater]: 1.256e-05
        [accelerated_algorithm]: 1.303e-05
        [meta_shard_fg_expand]: 4.21001e-06
        [get_grad_eliminate_]: 1.232e-05
        [merge_forward]: 8.63001e-06
        [cell_reuse_recompute_pass]: 9.20001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.673e-05
        [j_node_and_user_rematch]: 2.315e-05
        [meta_fg_expand]: 0.00188572
        [replace_old_param]: 6.333e-05
        [inline_without_move]: 5.666e-05
        [renormalize]: 0.00210345
        [add_forward_monad_depend]: 8.43999e-06
        [auto_monad_grad]: 2.49001e-06
        [auto_monad_eliminator]: 1.496e-05
        [cse]: 3.023e-05
        [replace_applicator]: 1.983e-05
    [Cycle 2]: 0.00036106, [27]
        [switch_simplify]: 5.07e-06
        [loop_unroll]: 5.32001e-06
        [a_1]: 5.746e-05
        [with_stream_mark]: 1.265e-05
        [recompute_prepare]: 3.86001e-06
        [updatestate_depend_eliminate]: 2.54999e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [parameter_eliminate]: 1.86e-06
        [specialize_transform]: 4.07e-06
        [updatestate_useless_node_eliminater]: 3.7e-06
        [accelerated_algorithm]: 8.64998e-06
        [meta_shard_fg_expand]: 1.90001e-06
        [get_grad_eliminate_]: 3.28998e-06
        [merge_forward]: 3.43999e-06
        [cell_reuse_recompute_pass]: 4.70999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.605e-05
        [j_node_and_user_rematch]: 6.88e-06
        [meta_fg_expand]: 5.645e-05
        [replace_old_param]: 5.49e-06
        [inline_without_move]: 3.63999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.66e-06
        [auto_monad_grad]: 1.03001e-06
        [auto_monad_eliminator]: 5.51998e-06
        [cse]: 1.241e-05
        [replace_applicator]: 4.33999e-06
[py_interpret_to_execute_after_opt_a]: 1.294e-05
[rewriter_after_opt_a]: 3.647e-05
[convert_after_rewriter]: 7e-06
[order_py_execute_after_rewriter]: 4.99998e-06
[mutable_eliminate]: 0.00077338
[jit_opt_b]: 4.785e-05, [1]
    [Cycle 1]: 3.973e-05, [2]
        [frontend_op_eliminate]: 1.415e-05
        [inline_after_opt_a]: 1.321e-05
[cconv]: 2.434e-05
[loop_unroll]: 0.0004944
[jit_opt_after_cconv]: 0.00013877, [1]
    [Cycle 1]: 0.00013095, [11]
        [c_1]: 1.401e-05
        [parameter_eliminate]: 3.45e-06
        [updatestate_depend_eliminate]: 5.35999e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [cse]: 2.268e-05
        [call_graph_tuple_transform]: 1.608e-05
        [tuple_list_get_item_eliminator]: 4.53999e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 7.79983e-07
        [switch_simplify]: 5.17999e-06
[remove_dup_value]: 1.656e-05
[partial_unused_args_eliminate]: 2.34999e-06
[environ_conv]: 5.71e-06
[add_recomputation]: 3.633e-05
[cse_after_recomputation]: 2.324e-05, [1]
    [Cycle 1]: 1.751e-05, [1]
        [cse]: 9.76003e-06
[auto_monad_reorder]: 1.425e-05
[get_jit_bprop_graph]: 2.15002e-06
[rewriter_after_jit_bprop_graph]: 7.95e-06
[opt_after_jit_grad]: 0.00064459
[symbol_engine_optimizer]: 7.833e-05, [1]
    [Cycle 1]: 7.062e-05, [6]
        [build]: 4.40999e-06
        [elim_shapecalc]: 7.55998e-06
        [elim_not_effective]: 1.298e-05
        [opt_reshape]: 5.51e-06
        [fold_const_symbol]: 6.73e-06
        [renormalize]: 8.00006e-07
[validate]: 3.527e-05
[backend_pass]: 9.20001e-07
[task_emit]: 2.929e-05
[execute]: 1.55999e-06
Sums
    bootstrap                            :     0.000206s :  1.09%
    type_inference                       :     0.010591s : 55.86%
    event_method                         :     0.000051s :  0.27%
    auto_monad                           :     0.000092s :  0.49%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.15%
    rewriter_before_opt_a                :     0.000090s :  0.47%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.36%
    jit_opt_a.loop_unroll                :     0.000042s :  0.22%
    jit_opt_a.a_1                        :     0.000928s :  4.89%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000025s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.03%
    jit_opt_a.specialize_transform       :     0.000018s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000022s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.08%
    jit_opt_a.merge_forward              :     0.000012s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000006s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000043s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000030s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001942s : 10.24%
    jit_opt_a.replace_old_param          :     0.000069s :  0.36%
    jit_opt_a.inline_without_move        :     0.000060s :  0.32%
    jit_opt_a.renormalize                :     0.002104s : 11.10%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.11%
    jit_opt_a.cse                        :     0.000043s :  0.22%
    jit_opt_a.replace_applicator         :     0.000024s :  0.13%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.19%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000773s :  4.08%
    jit_opt_b.frontend_op_eliminate      :     0.000014s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000013s :  0.07%
    cconv                                :     0.000024s :  0.13%
    loop_unroll                          :     0.000494s :  2.61%
    jit_opt_after_cconv.c_1              :     0.000014s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000023s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.03%
    remove_dup_value                     :     0.000017s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000036s :  0.19%
    cse_after_recomputation.cse          :     0.000010s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.04%
    opt_after_jit_grad                   :     0.000645s :  3.40%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000007s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000035s :  0.19%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000029s :  0.15%
    execute                              :     0.000002s :  0.01%


Time group info:
------[substitution.]   0.000260    50
 0.65% :     0.000002s :      1: substitution.elim_not_effective
 0.43% :     0.000001s :      1: substitution.fold_const_symbol
 1.92% :     0.000005s :      1: substitution.graph_param_transform
64.31% :     0.000167s :      6: substitution.inline
 7.76% :     0.000020s :      2: substitution.inline_without_move
 2.92% :     0.000008s :      9: substitution.j_node_and_user_rematch
 1.31% :     0.000003s :      2: substitution.minmaximum_grad
 1.85% :     0.000005s :      3: substitution.partial_eliminate
 3.03% :     0.000008s :      9: substitution.remove_not_recompute_node
 2.41% :     0.000006s :      2: substitution.replace_applicator
 1.81% :     0.000005s :      5: substitution.replace_old_param
 1.38% :     0.000004s :      1: substitution.set_cell_output_no_recompute
 2.51% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.73% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 5.99% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010522     2
88.20% :     0.009280s :      1: type_inference.infer
11.80% :     0.001242s :      1: type_inference.specialize
------[replace.]   0.000062     8
78.77% :     0.000049s :      6: replace.inline
21.23% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000175     8
93.69% :     0.000164s :      6: match.inline
 6.31% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000184  1109
 1.77% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.87% :     0.000002s :      1: predicate.ad_related_special_op_eliminate
 1.17% :     0.000002s :     18: predicate.addn_check_dump
 1.47% :     0.000003s :     18: predicate.addn_zero_filter
 2.43% :     0.000004s :     18: predicate.arithmetic_simplify
 1.43% :     0.000003s :     18: predicate.cast_eliminate
 0.21% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.15% :     0.000002s :     18: predicate.compare_switch_simplify
 1.43% :     0.000003s :     18: predicate.depend_value_elim
 1.24% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.38% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.44% :     0.000003s :     18: predicate.dict_set_item_eliminator
 0.72% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.20% :     0.000000s :      1: predicate.elim_not_effective
 0.29% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.48% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.18% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.47% :     0.000003s :     18: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     18: predicate.environ_get_eliminate
 1.21% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.10% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.41% :     0.000008s :     28: predicate.inline
 3.07% :     0.000006s :     33: predicate.inline_without_move
 0.46% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.41% :     0.000003s :     12: predicate.less_batch_normalization
 1.53% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.62% :     0.000003s :     21: predicate.load_eliminater
 0.80% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.26% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.97% :     0.000004s :     19: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000002s :     18: predicate.merge_addn
 1.20% :     0.000002s :     18: predicate.minmaximum_grad
 0.82% :     0.000002s :      1: predicate.mutable_eliminate
 0.27% :     0.000001s :      1: predicate.opt_reshape
 1.97% :     0.000004s :     21: predicate.partial_eliminate
 1.39% :     0.000003s :     18: predicate.print_const_string_wrapper
 1.78% :     0.000003s :     18: predicate.reduce_eliminate
 1.63% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.76% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.70% :     0.000003s :     22: predicate.replace_applicator
 1.68% :     0.000003s :     33: predicate.replace_old_param
 0.11% :     0.000000s :      1: predicate.reset_defer_inline
 1.50% :     0.000003s :     18: predicate.reshape_eliminate
 1.56% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.97% :     0.000002s :      1: predicate.row_tensor_eliminate
 1.56% :     0.000003s :     18: predicate.same_eliminate
 0.59% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.53% :     0.000001s :      2: predicate.special_op_eliminate
 1.11% :     0.000002s :     12: predicate.specialize_transform
 1.74% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.21% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.60% :     0.000005s :     26: predicate.switch_defer_inline
 2.24% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.79% :     0.000013s :     71: predicate.switch_simplify
 1.50% :     0.000003s :     18: predicate.tile_eliminate
 1.28% :     0.000002s :     18: predicate.transpose_eliminate
 1.56% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.49% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.45% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.80% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.55% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000003s :     18: predicate.value_based_eliminate
 0.12% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.24% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001299    22
72.52% :     0.000942s :     13: func_graph_cloner_run.FuncGraphClonerGraph
27.48% :     0.000357s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026724    76
 0.15% :     0.000039s :      1: add_recomputation
 0.36% :     0.000096s :      1: auto_monad
 0.06% :     0.000017s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.81% :     0.000218s :      1: bootstrap
 0.10% :     0.000027s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000025s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.21% :     0.000056s :      1: event_method
 0.01% :     0.000004s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
35.57% :     0.009507s :      1: jit_opt_a
 0.53% :     0.000142s :      1: jit_opt_after_cconv
 0.19% :     0.000050s :      1: jit_opt_b
 1.89% :     0.000504s :      1: loop_unroll
 2.94% :     0.000785s :      1: mutable_eliminate
 4.96% :     0.001325s :     26: opt.transform.jit_opt_a
 0.13% :     0.000036s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000019s :      4: opt.transform.jit_opt_b
 0.05% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.07% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000028s :      4: opt.transform.symbol_engine_opt
 2.46% :     0.000656s :      1: opt_after_jit_grad
 0.03% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 5.68% :     0.001517s :      1: renormalize.infer
 2.16% :     0.000576s :      1: renormalize.specialize
 0.04% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000040s :      1: rewriter_after_opt_a
 0.35% :     0.000093s :      1: rewriter_before_opt_a
 0.31% :     0.000082s :      1: symbol_engine_optimizer
 0.13% :     0.000034s :      1: task_emit
39.68% :     0.010605s :      1: type_inference
 0.20% :     0.000052s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.790.106 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c375700> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.790.759 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.013793, [30]
[bootstrap]: 0.00038735
[type_inference]: 0.00504007
[event_method]: 1.966e-05
[auto_monad]: 6.975e-05
[graph_reusing]: 6.64999e-06
[pre_auto_parallel]: 3.4e-06
[py_interpret_to_execute]: 3.469e-05
[rewriter_before_opt_a]: 7.543e-05
[expand_dump_flag]: 3.45998e-06
[jit_opt_a]: 0.00589795, [2]
    [Cycle 1]: 0.00174427, [27]
        [switch_simplify]: 5.428e-05
        [loop_unroll]: 2.909e-05
        [a_1]: 0.0005769
        [with_stream_mark]: 2.034e-05
        [recompute_prepare]: 7.18998e-06
        [updatestate_depend_eliminate]: 3.17002e-06
        [updatestate_assign_eliminate]: 2.21998e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 5.76998e-06
        [updatestate_useless_node_eliminater]: 4.91002e-06
        [accelerated_algorithm]: 5.87999e-06
        [meta_shard_fg_expand]: 1.35001e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 2.76999e-06
        [cell_reuse_recompute_pass]: 9.80013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.159e-05
        [j_node_and_user_rematch]: 8.64998e-06
        [meta_fg_expand]: 1.64e-06
        [replace_old_param]: 8.28999e-06
        [inline_without_move]: 5.24e-06
        [renormalize]: 0.00075799
        [add_forward_monad_depend]: 4.81002e-06
        [auto_monad_grad]: 1.92001e-06
        [auto_monad_eliminator]: 1.27e-05
        [cse]: 2.286e-05
        [replace_applicator]: 1.159e-05
    [Cycle 2]: 0.00031506, [27]
        [switch_simplify]: 5.36002e-06
        [loop_unroll]: 5.06002e-06
        [a_1]: 8.794e-05
        [with_stream_mark]: 9.88002e-06
        [recompute_prepare]: 5.19998e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 1.97999e-06
        [updatestate_loads_eliminate]: 1.89999e-06
        [parameter_eliminate]: 1.14998e-06
        [specialize_transform]: 4.69998e-06
        [updatestate_useless_node_eliminater]: 4.70999e-06
        [accelerated_algorithm]: 4.84e-06
        [meta_shard_fg_expand]: 1.52999e-06
        [get_grad_eliminate_]: 4.33999e-06
        [merge_forward]: 2.73e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.23e-05
        [j_node_and_user_rematch]: 7.8e-06
        [meta_fg_expand]: 1.45001e-06
        [replace_old_param]: 6.94999e-06
        [inline_without_move]: 4.47e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.59e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 5.22999e-06
        [cse]: 1.038e-05
        [replace_applicator]: 5.12e-06
[py_interpret_to_execute_after_opt_a]: 1.085e-05
[rewriter_after_opt_a]: 3.581e-05
[convert_after_rewriter]: 7.23e-06
[order_py_execute_after_rewriter]: 5.44e-06
[mutable_eliminate]: 0.00063445
[jit_opt_b]: 5.038e-05, [1]
    [Cycle 1]: 4.288e-05, [2]
        [frontend_op_eliminate]: 1.639e-05
        [inline_after_opt_a]: 1.453e-05
[cconv]: 1.865e-05
[loop_unroll]: 0.00043094
[jit_opt_after_cconv]: 0.00014545, [1]
    [Cycle 1]: 0.00013857, [11]
        [c_1]: 1.994e-05
        [parameter_eliminate]: 4.08001e-06
        [updatestate_depend_eliminate]: 6.43e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 2.027e-05
        [call_graph_tuple_transform]: 2.039e-05
        [tuple_list_get_item_eliminator]: 5.38002e-06
        [none_parameter_eliminate]: 1.74e-06
        [renormalize]: 6.69999e-07
        [switch_simplify]: 5.17e-06
[remove_dup_value]: 1.511e-05
[partial_unused_args_eliminate]: 2.27001e-06
[environ_conv]: 5.74999e-06
[add_recomputation]: 5.213e-05
[cse_after_recomputation]: 2.023e-05, [1]
    [Cycle 1]: 1.46e-05, [1]
        [cse]: 8.62998e-06
[auto_monad_reorder]: 1.566e-05
[get_jit_bprop_graph]: 2.39001e-06
[rewriter_after_jit_bprop_graph]: 8.17e-06
[opt_after_jit_grad]: 0.00048581
[symbol_engine_optimizer]: 7.434e-05, [1]
    [Cycle 1]: 6.732e-05, [6]
        [build]: 4.63001e-06
        [elim_shapecalc]: 7.71999e-06
        [elim_not_effective]: 1.254e-05
        [opt_reshape]: 5.84e-06
        [fold_const_symbol]: 8.53001e-06
        [renormalize]: 8.80013e-07
[validate]: 4.406e-05
Sums
    bootstrap                            :     0.000387s :  4.14%
    type_inference                       :     0.005040s : 53.87%
    event_method                         :     0.000020s :  0.21%
    auto_monad                           :     0.000070s :  0.75%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.04%
    py_interpret_to_execute              :     0.000035s :  0.37%
    rewriter_before_opt_a                :     0.000075s :  0.81%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000060s :  0.64%
    jit_opt_a.loop_unroll                :     0.000034s :  0.37%
    jit_opt_a.a_1                        :     0.000665s :  7.11%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000010s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000005s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.26%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.03%
    jit_opt_a.replace_old_param          :     0.000015s :  0.16%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000758s :  8.10%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.19%
    jit_opt_a.cse                        :     0.000033s :  0.36%
    jit_opt_a.replace_applicator         :     0.000017s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000036s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000634s :  6.78%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000019s :  0.20%
    loop_unroll                          :     0.000431s :  4.61%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000052s :  0.56%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.09%
    opt_after_jit_grad                   :     0.000486s :  5.19%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.47%


Time group info:
------[substitution.]   0.000244    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.64% :     0.000002s :      2: substitution.fold_const_symbol
 2.35% :     0.000006s :      3: substitution.graph_param_transform
87.16% :     0.000212s :      4: substitution.inline
 1.16% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.64% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.42% :     0.000003s :      2: substitution.replace_old_param
 4.76% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004958     2
82.87% :     0.004109s :      1: type_inference.infer
17.13% :     0.000849s :      1: type_inference.specialize
------[replace.]   0.000071     6
67.09% :     0.000047s :      4: replace.inline
32.91% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000221     6
95.20% :     0.000210s :      4: match.inline
 4.80% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.79% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.71% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000001s :     10: predicate.addn_zero_filter
 2.11% :     0.000002s :     10: predicate.arithmetic_simplify
 1.11% :     0.000001s :     10: predicate.cast_eliminate
 0.77% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.46% :     0.000002s :     10: predicate.depend_value_elim
 1.21% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.55% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.10% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.72% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.46% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.10% :     0.000006s :     22: predicate.inline
 0.87% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.62% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.01% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.58% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000002s :     15: predicate.partial_eliminate
 1.35% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.84% :     0.000002s :     10: predicate.reduce_eliminate
 1.32% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.51% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.62% :     0.000002s :     10: predicate.reshape_eliminate
 1.32% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.04% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.43% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.52% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.77% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.47% :     0.000003s :     16: predicate.switch_defer_inline
 1.96% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.43% :     0.000009s :     53: predicate.switch_simplify
 1.37% :     0.000002s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.54% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.62% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000654    12
50.14% :     0.000328s :      5: func_graph_cloner_run.FuncGraphClonerGraph
10.37% :     0.000068s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.49% :     0.000258s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015458    72
 0.36% :     0.000055s :      1: add_recomputation
 0.47% :     0.000073s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.65% :     0.000409s :      1: bootstrap
 0.14% :     0.000021s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
38.18% :     0.005902s :      1: jit_opt_a
 0.96% :     0.000149s :      1: jit_opt_after_cconv
 0.34% :     0.000053s :      1: jit_opt_b
 2.84% :     0.000439s :      1: loop_unroll
 4.17% :     0.000644s :      1: mutable_eliminate
 5.62% :     0.000868s :     26: opt.transform.jit_opt_a
 0.31% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.19% :     0.000494s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000037s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.72% :     0.000421s :      1: renormalize.infer
 2.13% :     0.000330s :      1: renormalize.specialize
 0.07% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000039s :      1: rewriter_after_opt_a
 0.50% :     0.000078s :      1: rewriter_before_opt_a
 0.50% :     0.000077s :      1: symbol_engine_optimizer
32.73% :     0.005059s :      1: type_inference


TotalTime = 0.0216264, [30]
[bootstrap]: 0.00020136
[type_inference]: 0.0101635
[event_method]: 4.477e-05
[auto_monad]: 0.00010034
[graph_reusing]: 7.30003e-06
[pre_auto_parallel]: 3.25002e-06
[py_interpret_to_execute]: 3.027e-05
[rewriter_before_opt_a]: 9.365e-05
[expand_dump_flag]: 3.33e-06
[jit_opt_a]: 0.0090597, [2]
    [Cycle 1]: 0.00502716, [27]
        [switch_simplify]: 6.941e-05
        [loop_unroll]: 3.976e-05
        [a_1]: 0.00094855
        [with_stream_mark]: 3.288e-05
        [recompute_prepare]: 2.036e-05
        [updatestate_depend_eliminate]: 8.97e-06
        [updatestate_assign_eliminate]: 6.84999e-06
        [updatestate_loads_eliminate]: 6.18998e-06
        [parameter_eliminate]: 2.98e-06
        [specialize_transform]: 1.258e-05
        [updatestate_useless_node_eliminater]: 1.177e-05
        [accelerated_algorithm]: 1.189e-05
        [meta_shard_fg_expand]: 3.90998e-06
        [get_grad_eliminate_]: 1.226e-05
        [merge_forward]: 7.71001e-06
        [cell_reuse_recompute_pass]: 9.70002e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.599e-05
        [j_node_and_user_rematch]: 2.285e-05
        [meta_fg_expand]: 0.00184433
        [replace_old_param]: 5.237e-05
        [inline_without_move]: 5.126e-05
        [renormalize]: 0.00157119
        [add_forward_monad_depend]: 4.73001e-06
        [auto_monad_grad]: 1.87001e-06
        [auto_monad_eliminator]: 1.14e-05
        [cse]: 2.383e-05
        [replace_applicator]: 1.066e-05
    [Cycle 2]: 0.00026229, [27]
        [switch_simplify]: 4.92999e-06
        [loop_unroll]: 4.27e-06
        [a_1]: 3.738e-05
        [with_stream_mark]: 8.13999e-06
        [recompute_prepare]: 3.71001e-06
        [updatestate_depend_eliminate]: 2.07999e-06
        [updatestate_assign_eliminate]: 1.60001e-06
        [updatestate_loads_eliminate]: 1.47001e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 3.53e-06
        [updatestate_useless_node_eliminater]: 3.51001e-06
        [accelerated_algorithm]: 3.37002e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 3.53e-06
        [merge_forward]: 2.07999e-06
        [cell_reuse_recompute_pass]: 1.41998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.20999e-06
        [j_node_and_user_rematch]: 5.52001e-06
        [meta_fg_expand]: 2.455e-05
        [replace_old_param]: 3.65998e-06
        [inline_without_move]: 3.21999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.40999e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 4.27003e-06
        [cse]: 9.64e-06
        [replace_applicator]: 3.65e-06
[py_interpret_to_execute_after_opt_a]: 8.82e-06
[rewriter_after_opt_a]: 3.094e-05
[convert_after_rewriter]: 6.81001e-06
[order_py_execute_after_rewriter]: 5.00001e-06
[mutable_eliminate]: 0.00051274
[jit_opt_b]: 3.937e-05, [1]
    [Cycle 1]: 3.264e-05, [2]
        [frontend_op_eliminate]: 1.029e-05
        [inline_after_opt_a]: 1.08e-05
[cconv]: 1.838e-05
[loop_unroll]: 0.00040675
[jit_opt_after_cconv]: 0.00011477, [1]
    [Cycle 1]: 0.00010853, [11]
        [c_1]: 1.145e-05
        [parameter_eliminate]: 2.15002e-06
        [updatestate_depend_eliminate]: 4.03001e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 1.55001e-06
        [cse]: 1.526e-05
        [call_graph_tuple_transform]: 1.241e-05
        [tuple_list_get_item_eliminator]: 3.88999e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 3.8e-06
[remove_dup_value]: 1.426e-05
[partial_unused_args_eliminate]: 2.04999e-06
[environ_conv]: 4.95999e-06
[add_recomputation]: 3.111e-05
[cse_after_recomputation]: 1.928e-05, [1]
    [Cycle 1]: 1.38e-05, [1]
        [cse]: 7.83001e-06
[auto_monad_reorder]: 1.331e-05
[get_jit_bprop_graph]: 1.79e-06
[rewriter_after_jit_bprop_graph]: 4.68999e-06
[opt_after_jit_grad]: 0.00043159
[symbol_engine_optimizer]: 6.01e-05, [1]
    [Cycle 1]: 5.419e-05, [6]
        [build]: 3.23e-06
        [elim_shapecalc]: 5.49998e-06
        [elim_not_effective]: 8.32998e-06
        [opt_reshape]: 4.03001e-06
        [fold_const_symbol]: 5.81e-06
        [renormalize]: 3.30008e-07
[validate]: 2.646e-05
Sums
    bootstrap                            :     0.000201s :  1.17%
    type_inference                       :     0.010163s : 58.92%
    event_method                         :     0.000045s :  0.26%
    auto_monad                           :     0.000100s :  0.58%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.18%
    rewriter_before_opt_a                :     0.000094s :  0.54%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.43%
    jit_opt_a.loop_unroll                :     0.000044s :  0.26%
    jit_opt_a.a_1                        :     0.000986s :  5.72%
    jit_opt_a.with_stream_mark           :     0.000041s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000024s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000016s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000015s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.09%
    jit_opt_a.merge_forward              :     0.000010s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001869s : 10.83%
    jit_opt_a.replace_old_param          :     0.000056s :  0.32%
    jit_opt_a.inline_without_move        :     0.000054s :  0.32%
    jit_opt_a.renormalize                :     0.001571s :  9.11%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.09%
    jit_opt_a.cse                        :     0.000033s :  0.19%
    jit_opt_a.replace_applicator         :     0.000014s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.05%
    rewriter_after_opt_a                 :     0.000031s :  0.18%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000513s :  2.97%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.06%
    cconv                                :     0.000018s :  0.11%
    loop_unroll                          :     0.000407s :  2.36%
    jit_opt_after_cconv.c_1              :     0.000011s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000012s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000031s :  0.18%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000432s :  2.50%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000008s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000026s :  0.15%


Time group info:
------[substitution.]   0.000288    50
 0.49% :     0.000001s :      1: substitution.elim_not_effective
 0.29% :     0.000001s :      1: substitution.fold_const_symbol
 1.19% :     0.000003s :      1: substitution.graph_param_transform
61.97% :     0.000178s :      6: substitution.inline
 5.43% :     0.000016s :      2: substitution.inline_without_move
 2.11% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.24% :     0.000004s :      2: substitution.minmaximum_grad
 1.45% :     0.000004s :      3: substitution.partial_eliminate
 2.31% :     0.000007s :      9: substitution.remove_not_recompute_node
10.59% :     0.000030s :      2: substitution.replace_applicator
 1.66% :     0.000005s :      5: substitution.replace_old_param
 0.94% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.50% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.82% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 6.00% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010079     2
89.10% :     0.008980s :      1: type_inference.infer
10.90% :     0.001098s :      1: type_inference.specialize
------[replace.]   0.000078     8
81.85% :     0.000063s :      6: replace.inline
18.15% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000187     8
93.52% :     0.000175s :      6: match.inline
 6.48% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000170  1109
 1.78% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.46% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.38% :     0.000002s :     18: predicate.addn_check_dump
 1.53% :     0.000003s :     18: predicate.addn_zero_filter
 2.22% :     0.000004s :     18: predicate.arithmetic_simplify
 1.38% :     0.000002s :     18: predicate.cast_eliminate
 0.15% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.28% :     0.000002s :     18: predicate.compare_switch_simplify
 1.62% :     0.000003s :     18: predicate.depend_value_elim
 1.26% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.34% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.08% :     0.000000s :      1: predicate.elim_not_effective
 0.19% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     18: predicate.environ_get_eliminate
 1.38% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.19% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.60% :     0.000008s :     28: predicate.inline
 3.17% :     0.000005s :     33: predicate.inline_without_move
 0.50% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :     12: predicate.less_batch_normalization
 1.58% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.78% :     0.000003s :     21: predicate.load_eliminater
 0.63% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.68% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.95% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.25% :     0.000002s :     18: predicate.merge_addn
 1.29% :     0.000002s :     18: predicate.minmaximum_grad
 0.64% :     0.000001s :      1: predicate.mutable_eliminate
 0.15% :     0.000000s :      1: predicate.opt_reshape
 2.15% :     0.000004s :     21: predicate.partial_eliminate
 1.37% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.82% :     0.000003s :     18: predicate.reduce_eliminate
 1.49% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.44% :     0.000002s :     22: predicate.replace_applicator
 1.66% :     0.000003s :     33: predicate.replace_old_param
 0.09% :     0.000000s :      1: predicate.reset_defer_inline
 1.42% :     0.000002s :     18: predicate.reshape_eliminate
 1.51% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.40% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     18: predicate.same_eliminate
 0.61% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.37% :     0.000001s :      2: predicate.special_op_eliminate
 1.22% :     0.000002s :     12: predicate.specialize_transform
 1.71% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.74% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.79% :     0.000005s :     26: predicate.switch_defer_inline
 2.25% :     0.000004s :     26: predicate.switch_layer_defer_inline
 7.49% :     0.000013s :     71: predicate.switch_simplify
 1.46% :     0.000002s :     18: predicate.tile_eliminate
 1.34% :     0.000002s :     18: predicate.transpose_eliminate
 1.60% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 2.00% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.49% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.54% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000003s :     18: predicate.value_based_eliminate
 0.15% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.27% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001363    22
78.07% :     0.001064s :     13: func_graph_cloner_run.FuncGraphClonerGraph
21.93% :     0.000299s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024558    72
 0.14% :     0.000034s :      1: add_recomputation
 0.43% :     0.000106s :      1: auto_monad
 0.07% :     0.000016s :      1: auto_monad_reorder
 0.87% :     0.000215s :      1: bootstrap
 0.08% :     0.000021s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.20% :     0.000050s :      1: event_method
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
36.90% :     0.009062s :      1: jit_opt_a
 0.48% :     0.000118s :      1: jit_opt_after_cconv
 0.17% :     0.000042s :      1: jit_opt_b
 1.69% :     0.000414s :      1: loop_unroll
 2.12% :     0.000521s :      1: mutable_eliminate
 5.49% :     0.001348s :     26: opt.transform.jit_opt_a
 0.11% :     0.000028s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000015s :      4: opt.transform.jit_opt_b
 0.04% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000010s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000012s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000020s :      4: opt.transform.symbol_engine_opt
 1.78% :     0.000438s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000006s :      1: pre_auto_parallel
 0.14% :     0.000034s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 4.89% :     0.001202s :      1: renormalize.infer
 1.47% :     0.000362s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000034s :      1: rewriter_after_opt_a
 0.40% :     0.000097s :      1: rewriter_before_opt_a
 0.25% :     0.000062s :      1: symbol_engine_optimizer
41.45% :     0.010181s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.800.695 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843cd790> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.801.211 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0320448, [33]
[bootstrap]: 0.00020327
[type_inference]: 0.0104649
[event_method]: 4.73e-05
[auto_monad]: 8.582e-05
[graph_reusing]: 5.37999e-06
[pre_auto_parallel]: 2.91e-06
[py_interpret_to_execute]: 2.825e-05
[rewriter_before_opt_a]: 9.009e-05
[expand_dump_flag]: 2.91999e-06
[jit_opt_a]: 0.0113371, [3]
    [Cycle 1]: 0.00546296, [27]
        [switch_simplify]: 5.994e-05
        [loop_unroll]: 3.772e-05
        [a_1]: 0.00080412
        [with_stream_mark]: 1.98e-05
        [recompute_prepare]: 1.68e-05
        [updatestate_depend_eliminate]: 7.28e-06
        [updatestate_assign_eliminate]: 5.70001e-06
        [updatestate_loads_eliminate]: 5.14998e-06
        [parameter_eliminate]: 1.55001e-06
        [specialize_transform]: 1.268e-05
        [updatestate_useless_node_eliminater]: 1.13e-05
        [accelerated_algorithm]: 1.2e-05
        [meta_shard_fg_expand]: 2.95002e-06
        [get_grad_eliminate_]: 1.198e-05
        [merge_forward]: 6.78e-06
        [cell_reuse_recompute_pass]: 7.50006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.379e-05
        [j_node_and_user_rematch]: 2.231e-05
        [meta_fg_expand]: 0.00128806
        [replace_old_param]: 4.903e-05
        [inline_without_move]: 4.863e-05
        [renormalize]: 0.00256811
        [add_forward_monad_depend]: 7.61999e-06
        [auto_monad_grad]: 5.07e-06
        [auto_monad_eliminator]: 4.761e-05
        [cse]: 0.00014297
        [replace_applicator]: 6.343e-05
    [Cycle 2]: 0.00205207, [27]
        [switch_simplify]: 3.685e-05
        [loop_unroll]: 3.581e-05
        [a_1]: 0.00110425
        [with_stream_mark]: 1.256e-05
        [recompute_prepare]: 8.67998e-06
        [updatestate_depend_eliminate]: 4.05e-06
        [updatestate_assign_eliminate]: 3.16001e-06
        [updatestate_loads_eliminate]: 2.93998e-06
        [parameter_eliminate]: 8.89995e-07
        [specialize_transform]: 7.16001e-06
        [updatestate_useless_node_eliminater]: 6.91001e-06
        [accelerated_algorithm]: 7.68001e-06
        [meta_shard_fg_expand]: 1.72999e-06
        [get_grad_eliminate_]: 6.76e-06
        [merge_forward]: 3.33998e-06
        [cell_reuse_recompute_pass]: 1.04998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.426e-05
        [j_node_and_user_rematch]: 1.114e-05
        [meta_fg_expand]: 3.995e-05
        [replace_old_param]: 1.01e-05
        [inline_without_move]: 6.74001e-06
        [renormalize]: 0.00052742
        [add_forward_monad_depend]: 5.05999e-06
        [auto_monad_grad]: 2.02001e-06
        [auto_monad_eliminator]: 1.407e-05
        [cse]: 2.222e-05
        [replace_applicator]: 1.356e-05
    [Cycle 3]: 0.0004083, [27]
        [switch_simplify]: 7.58001e-06
        [loop_unroll]: 6.97002e-06
        [a_1]: 0.00014334
        [with_stream_mark]: 9.43002e-06
        [recompute_prepare]: 6.94001e-06
        [updatestate_depend_eliminate]: 3.6e-06
        [updatestate_assign_eliminate]: 3.2e-06
        [updatestate_loads_eliminate]: 3.09999e-06
        [parameter_eliminate]: 9.30013e-07
        [specialize_transform]: 7.03e-06
        [updatestate_useless_node_eliminater]: 6.69999e-06
        [accelerated_algorithm]: 7.34002e-06
        [meta_shard_fg_expand]: 1.67001e-06
        [get_grad_eliminate_]: 6.58e-06
        [merge_forward]: 3.31001e-06
        [cell_reuse_recompute_pass]: 1.87001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.525e-05
        [j_node_and_user_rematch]: 1.124e-05
        [meta_fg_expand]: 2.27999e-06
        [replace_old_param]: 8.72998e-06
        [inline_without_move]: 6.76e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 1.44e-06
        [auto_monad_eliminator]: 8.32e-06
        [cse]: 1.554e-05
        [replace_applicator]: 6.98e-06
[py_interpret_to_execute_after_opt_a]: 1.114e-05
[rewriter_after_opt_a]: 3.836e-05
[convert_after_rewriter]: 7.38e-06
[order_py_execute_after_rewriter]: 5.27999e-06
[mutable_eliminate]: 0.00057693
[jit_opt_b]: 5.989e-05, [1]
    [Cycle 1]: 5.279e-05, [2]
        [frontend_op_eliminate]: 2.207e-05
        [inline_after_opt_a]: 1.952e-05
[cconv]: 2.11e-05
[loop_unroll]: 0.00041493
[jit_opt_after_cconv]: 0.00017176, [1]
    [Cycle 1]: 0.00016591, [11]
        [c_1]: 2.964e-05
        [parameter_eliminate]: 3.4e-06
        [updatestate_depend_eliminate]: 6.68e-06
        [updatestate_assign_eliminate]: 3.48999e-06
        [updatestate_loads_eliminate]: 3.69002e-06
        [cse]: 2.924e-05
        [call_graph_tuple_transform]: 2.377e-05
        [tuple_list_get_item_eliminator]: 7.43e-06
        [none_parameter_eliminate]: 1.18001e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 7.15e-06
[remove_dup_value]: 2.048e-05
[partial_unused_args_eliminate]: 2.15002e-06
[environ_conv]: 5.61e-06
[add_recomputation]: 4.75e-05
[cse_after_recomputation]: 3.034e-05, [1]
    [Cycle 1]: 2.417e-05, [1]
        [cse]: 1.781e-05
[auto_monad_reorder]: 1.576e-05
[get_jit_bprop_graph]: 1.48002e-06
[rewriter_after_jit_bprop_graph]: 4.32e-06
[opt_after_jit_grad]: 0.00047042
[symbol_engine_optimizer]: 8.518e-05, [1]
    [Cycle 1]: 7.851e-05, [6]
        [build]: 4.02e-06
        [elim_shapecalc]: 9.72001e-06
        [elim_not_effective]: 1.67e-05
        [opt_reshape]: 7.70998e-06
        [fold_const_symbol]: 1.217e-05
        [renormalize]: 6.69999e-07
[validate]: 3.854e-05
[backend_pass]: 7.7e-07
[task_emit]: 0.00754207
[execute]: 6.86999e-06
Sums
    bootstrap                            :     0.000203s :  0.73%
    type_inference                       :     0.010465s : 37.54%
    event_method                         :     0.000047s :  0.17%
    auto_monad                           :     0.000086s :  0.31%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000028s :  0.10%
    rewriter_before_opt_a                :     0.000090s :  0.32%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000104s :  0.37%
    jit_opt_a.loop_unroll                :     0.000081s :  0.29%
    jit_opt_a.a_1                        :     0.002052s :  7.36%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000032s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000011s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.01%
    jit_opt_a.specialize_transform       :     0.000027s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000027s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.09%
    jit_opt_a.merge_forward              :     0.000013s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000053s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000045s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001330s :  4.77%
    jit_opt_a.replace_old_param          :     0.000068s :  0.24%
    jit_opt_a.inline_without_move        :     0.000062s :  0.22%
    jit_opt_a.renormalize                :     0.003096s : 11.10%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000070s :  0.25%
    jit_opt_a.cse                        :     0.000181s :  0.65%
    jit_opt_a.replace_applicator         :     0.000084s :  0.30%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.04%
    rewriter_after_opt_a                 :     0.000038s :  0.14%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000577s :  2.07%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.07%
    cconv                                :     0.000021s :  0.08%
    loop_unroll                          :     0.000415s :  1.49%
    jit_opt_after_cconv.c_1              :     0.000030s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000029s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000020s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000048s :  0.17%
    cse_after_recomputation.cse          :     0.000018s :  0.06%
    auto_monad_reorder                   :     0.000016s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.02%
    opt_after_jit_grad                   :     0.000470s :  1.69%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000039s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007542s : 27.05%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000473   118
 4.88% :     0.000023s :      1: substitution.arithmetic_simplify
 0.54% :     0.000003s :      4: substitution.elim_not_effective
 0.35% :     0.000002s :      4: substitution.fold_const_symbol
 1.24% :     0.000006s :      5: substitution.graph_param_transform
61.82% :     0.000292s :     12: substitution.inline
 3.08% :     0.000015s :      2: substitution.inline_without_move
 1.81% :     0.000009s :     16: substitution.j_node_and_user_rematch
 2.00% :     0.000009s :      7: substitution.minmaximum_grad
 1.28% :     0.000006s :      5: substitution.partial_eliminate
 2.36% :     0.000011s :     16: substitution.remove_not_recompute_node
 5.24% :     0.000025s :     10: substitution.replace_applicator
 1.21% :     0.000006s :      7: substitution.replace_old_param
 0.47% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 3.89% :     0.000018s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.63% :     0.000012s :      7: substitution.tuple_list_get_item_depend_reorder
 7.20% :     0.000034s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010387     2
86.85% :     0.009021s :      1: type_inference.infer
13.15% :     0.001366s :      1: type_inference.specialize
------[replace.]   0.000143    20
 3.75% :     0.000005s :      1: replace.arithmetic_simplify
60.98% :     0.000087s :     12: replace.inline
35.27% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000325    20
 6.90% :     0.000022s :      1: match.arithmetic_simplify
88.11% :     0.000286s :     12: match.inline
 4.99% :     0.000016s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000385  2790
 1.57% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.51% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.45% :     0.000006s :     46: predicate.addn_check_dump
 1.71% :     0.000007s :     46: predicate.addn_zero_filter
 2.17% :     0.000008s :     47: predicate.arithmetic_simplify
 1.54% :     0.000006s :     47: predicate.cast_eliminate
 0.21% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.43% :     0.000006s :     46: predicate.compare_switch_simplify
 1.61% :     0.000006s :     46: predicate.depend_value_elim
 1.49% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.53% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.61% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.38% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.14% :     0.000001s :      5: predicate.elim_not_effective
 0.24% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.54% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.46% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.48% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.50% :     0.000006s :     47: predicate.environ_get_eliminate
 1.48% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.11% :     0.000000s :      5: predicate.fold_const_symbol
 0.88% :     0.000003s :     21: predicate.get_grad_eliminate
 0.13% :     0.000001s :      5: predicate.graph_param_transform
 4.15% :     0.000016s :     76: predicate.inline
 1.71% :     0.000007s :     42: predicate.inline_without_move
 0.37% :     0.000001s :     21: predicate.j_node_and_user_rematch
 1.08% :     0.000004s :     21: predicate.less_batch_normalization
 1.78% :     0.000007s :     54: predicate.list_to_tuple_eliminator_
 1.99% :     0.000008s :     59: predicate.load_eliminater
 0.49% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.25% :     0.000013s :     89: predicate.loop_unroll_before_grad
 1.79% :     0.000007s :     52: predicate.make_slice_get_slice_eliminator
 1.49% :     0.000006s :     46: predicate.merge_addn
 1.58% :     0.000006s :     47: predicate.minmaximum_grad
 0.59% :     0.000002s :      5: predicate.mutable_eliminate
 0.21% :     0.000001s :      5: predicate.opt_reshape
 2.45% :     0.000009s :     59: predicate.partial_eliminate
 1.52% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.85% :     0.000007s :     47: predicate.reduce_eliminate
 1.80% :     0.000007s :     54: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.59% :     0.000010s :    105: predicate.replace_applicator
 0.87% :     0.000003s :     42: predicate.replace_old_param
 0.13% :     0.000000s :      5: predicate.reset_defer_inline
 1.66% :     0.000006s :     47: predicate.reshape_eliminate
 1.49% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.41% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.49% :     0.000006s :     46: predicate.same_eliminate
 0.45% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.40% :     0.000002s :     10: predicate.special_op_eliminate
 0.87% :     0.000003s :     21: predicate.specialize_transform
 1.71% :     0.000007s :     46: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.20% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.70% :     0.000010s :     66: predicate.switch_defer_inline
 2.47% :     0.000010s :     66: predicate.switch_layer_defer_inline
 6.11% :     0.000024s :    160: predicate.switch_simplify
 1.54% :     0.000006s :     47: predicate.tile_eliminate
 1.51% :     0.000006s :     47: predicate.transpose_eliminate
 1.87% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.68% :     0.000006s :     47: predicate.tuple_list_get_item_depend_reorder
 2.91% :     0.000011s :     64: predicate.tuple_list_get_item_eliminator
 1.73% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.80% :     0.000007s :     54: predicate.tuple_to_list_eliminator_
 1.94% :     0.000007s :     59: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000011s :     80: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000007s :     46: predicate.value_based_eliminate
 0.19% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.33% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001531    29
55.23% :     0.000846s :     13: func_graph_cloner_run.FuncGraphClonerGraph
44.77% :     0.000686s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.037944    91
 0.13% :     0.000051s :      1: add_recomputation
 0.24% :     0.000091s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.59% :     0.000223s :      1: bootstrap
 0.06% :     0.000024s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000033s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.14% :     0.000051s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
29.89% :     0.011341s :      1: jit_opt_a
 0.46% :     0.000175s :      1: jit_opt_after_cconv
 0.17% :     0.000063s :      1: jit_opt_b
 1.11% :     0.000423s :      1: loop_unroll
 1.54% :     0.000586s :      1: mutable_eliminate
 6.98% :     0.002648s :     39: opt.transform.jit_opt_a
 0.17% :     0.000064s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000036s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000043s :      4: opt.transform.symbol_engine_opt
 1.26% :     0.000478s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000031s :      1: py_interpret_to_execute
 0.04% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000023s :      1: remove_dup_value
 4.38% :     0.001662s :      2: renormalize.infer
 3.75% :     0.001422s :      2: renormalize.specialize
 0.02% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000041s :      1: rewriter_after_opt_a
 0.25% :     0.000093s :      1: rewriter_before_opt_a
 0.23% :     0.000088s :      1: symbol_engine_optimizer
19.90% :     0.007552s :      1: task_emit
27.63% :     0.010483s :      1: type_inference
 0.16% :     0.000059s :      1: validate


TotalTime = 0.0212654, [33]
[bootstrap]: 0.0002139
[type_inference]: 0.0103413
[event_method]: 4.742e-05
[auto_monad]: 9.208e-05
[graph_reusing]: 6.86001e-06
[pre_auto_parallel]: 2.93e-06
[py_interpret_to_execute]: 2.41e-05
[rewriter_before_opt_a]: 8.987e-05
[expand_dump_flag]: 3.16999e-06
[jit_opt_a]: 0.00840962, [2]
    [Cycle 1]: 0.00478311, [27]
        [switch_simplify]: 6.253e-05
        [loop_unroll]: 3.743e-05
        [a_1]: 0.00093115
        [with_stream_mark]: 2.361e-05
        [recompute_prepare]: 1.796e-05
        [updatestate_depend_eliminate]: 7.52002e-06
        [updatestate_assign_eliminate]: 6.43003e-06
        [updatestate_loads_eliminate]: 5.42999e-06
        [parameter_eliminate]: 2.91e-06
        [specialize_transform]: 1.241e-05
        [updatestate_useless_node_eliminater]: 1.141e-05
        [accelerated_algorithm]: 1.226e-05
        [meta_shard_fg_expand]: 3.63e-06
        [get_grad_eliminate_]: 1.19e-05
        [merge_forward]: 7.66999e-06
        [cell_reuse_recompute_pass]: 9.49978e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.577e-05
        [j_node_and_user_rematch]: 2.302e-05
        [meta_fg_expand]: 0.00145998
        [replace_old_param]: 5.259e-05
        [inline_without_move]: 4.992e-05
        [renormalize]: 0.00176102
        [add_forward_monad_depend]: 5.25001e-06
        [auto_monad_grad]: 2.41998e-06
        [auto_monad_eliminator]: 1.229e-05
        [cse]: 2.228e-05
        [replace_applicator]: 1.26e-05
    [Cycle 2]: 0.00029549, [27]
        [switch_simplify]: 4.37998e-06
        [loop_unroll]: 3.60998e-06
        [a_1]: 4.123e-05
        [with_stream_mark]: 9.09998e-06
        [recompute_prepare]: 3.65e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 1.72999e-06
        [updatestate_loads_eliminate]: 1.45001e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 3.50998e-06
        [updatestate_useless_node_eliminater]: 3.21999e-06
        [accelerated_algorithm]: 6.54999e-06
        [meta_shard_fg_expand]: 1.53002e-06
        [get_grad_eliminate_]: 3.09999e-06
        [merge_forward]: 2.26998e-06
        [cell_reuse_recompute_pass]: 2.26998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.268e-05
        [j_node_and_user_rematch]: 6.29999e-06
        [meta_fg_expand]: 4.009e-05
        [replace_old_param]: 4.03001e-06
        [inline_without_move]: 3.3e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 9.29984e-07
        [auto_monad_eliminator]: 4.61002e-06
        [cse]: 1.113e-05
        [replace_applicator]: 4.02e-06
[py_interpret_to_execute_after_opt_a]: 9.86e-06
[rewriter_after_opt_a]: 3.263e-05
[convert_after_rewriter]: 5.81e-06
[order_py_execute_after_rewriter]: 4.70999e-06
[mutable_eliminate]: 0.00055399
[jit_opt_b]: 3.92e-05, [1]
    [Cycle 1]: 3.223e-05, [2]
        [frontend_op_eliminate]: 1.019e-05
        [inline_after_opt_a]: 1.095e-05
[cconv]: 2.069e-05
[loop_unroll]: 0.00043
[jit_opt_after_cconv]: 0.0001147, [1]
    [Cycle 1]: 0.00010816, [11]
        [c_1]: 1.174e-05
        [parameter_eliminate]: 2.36e-06
        [updatestate_depend_eliminate]: 4.70999e-06
        [updatestate_assign_eliminate]: 1.78002e-06
        [updatestate_loads_eliminate]: 1.46998e-06
        [cse]: 1.467e-05
        [call_graph_tuple_transform]: 1.304e-05
        [tuple_list_get_item_eliminator]: 3.73001e-06
        [none_parameter_eliminate]: 1.79998e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 3.85e-06
[remove_dup_value]: 1.357e-05
[partial_unused_args_eliminate]: 2.16998e-06
[environ_conv]: 5.77999e-06
[add_recomputation]: 3.242e-05
[cse_after_recomputation]: 1.88e-05, [1]
    [Cycle 1]: 1.353e-05, [1]
        [cse]: 7.5e-06
[auto_monad_reorder]: 1.132e-05
[get_jit_bprop_graph]: 1.49e-06
[rewriter_after_jit_bprop_graph]: 5.32001e-06
[opt_after_jit_grad]: 0.0004467
[symbol_engine_optimizer]: 6.356e-05, [1]
    [Cycle 1]: 5.719e-05, [6]
        [build]: 2.73e-06
        [elim_shapecalc]: 6.17001e-06
        [elim_not_effective]: 9.60001e-06
        [opt_reshape]: 4.02998e-06
        [fold_const_symbol]: 6.34001e-06
        [renormalize]: 3.80009e-07
[validate]: 2.88e-05
[backend_pass]: 8.10018e-07
[task_emit]: 2.075e-05
[execute]: 1.21002e-06
Sums
    bootstrap                            :     0.000214s :  1.23%
    type_inference                       :     0.010341s : 59.68%
    event_method                         :     0.000047s :  0.27%
    auto_monad                           :     0.000092s :  0.53%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000024s :  0.14%
    rewriter_before_opt_a                :     0.000090s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.39%
    jit_opt_a.loop_unroll                :     0.000041s :  0.24%
    jit_opt_a.a_1                        :     0.000972s :  5.61%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000016s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000019s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.09%
    jit_opt_a.merge_forward              :     0.000010s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001500s :  8.66%
    jit_opt_a.replace_old_param          :     0.000057s :  0.33%
    jit_opt_a.inline_without_move        :     0.000053s :  0.31%
    jit_opt_a.renormalize                :     0.001761s : 10.16%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.10%
    jit_opt_a.cse                        :     0.000033s :  0.19%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000554s :  3.20%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.06%
    cconv                                :     0.000021s :  0.12%
    loop_unroll                          :     0.000430s :  2.48%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000032s :  0.19%
    cse_after_recomputation.cse          :     0.000007s :  0.04%
    auto_monad_reorder                   :     0.000011s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000447s :  2.58%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000029s :  0.17%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000021s :  0.12%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000253    50
 0.68% :     0.000002s :      1: substitution.elim_not_effective
 0.45% :     0.000001s :      1: substitution.fold_const_symbol
 1.49% :     0.000004s :      1: substitution.graph_param_transform
67.03% :     0.000169s :      6: substitution.inline
 6.33% :     0.000016s :      2: substitution.inline_without_move
 2.68% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.40% :     0.000004s :      2: substitution.minmaximum_grad
 1.57% :     0.000004s :      3: substitution.partial_eliminate
 3.04% :     0.000008s :      9: substitution.remove_not_recompute_node
 2.07% :     0.000005s :      2: substitution.replace_applicator
 1.65% :     0.000004s :      5: substitution.replace_old_param
 1.02% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.91% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.64% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 6.03% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010272     2
88.17% :     0.009057s :      1: type_inference.infer
11.83% :     0.001215s :      1: type_inference.specialize
------[replace.]   0.000065     8
77.07% :     0.000050s :      6: replace.inline
22.93% :     0.000015s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000176     8
94.07% :     0.000166s :      6: match.inline
 5.93% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000166  1109
 1.76% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.52% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.34% :     0.000002s :     18: predicate.addn_check_dump
 1.58% :     0.000003s :     18: predicate.addn_zero_filter
 1.93% :     0.000003s :     18: predicate.arithmetic_simplify
 1.51% :     0.000002s :     18: predicate.cast_eliminate
 0.11% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.33% :     0.000002s :     18: predicate.compare_switch_simplify
 1.43% :     0.000002s :     18: predicate.depend_value_elim
 1.41% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.56% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.35% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.45% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      1: predicate.elim_not_effective
 0.21% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.31% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.51% :     0.000003s :     18: predicate.environ_get_depend_swap
 1.40% :     0.000002s :     18: predicate.environ_get_eliminate
 1.31% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.13% :     0.000002s :     12: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 4.23% :     0.000007s :     28: predicate.inline
 3.21% :     0.000005s :     33: predicate.inline_without_move
 0.49% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :     12: predicate.less_batch_normalization
 1.60% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.73% :     0.000003s :     21: predicate.load_eliminater
 0.72% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.47% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.82% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.38% :     0.000002s :     18: predicate.merge_addn
 1.44% :     0.000002s :     18: predicate.minmaximum_grad
 0.95% :     0.000002s :      1: predicate.mutable_eliminate
 0.19% :     0.000000s :      1: predicate.opt_reshape
 2.17% :     0.000004s :     21: predicate.partial_eliminate
 1.41% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.93% :     0.000003s :     18: predicate.reduce_eliminate
 1.73% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.49% :     0.000002s :     22: predicate.replace_applicator
 1.55% :     0.000003s :     33: predicate.replace_old_param
 0.16% :     0.000000s :      1: predicate.reset_defer_inline
 1.79% :     0.000003s :     18: predicate.reshape_eliminate
 1.38% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.43% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     18: predicate.same_eliminate
 0.62% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.25% :     0.000000s :      2: predicate.special_op_eliminate
 1.18% :     0.000002s :     12: predicate.specialize_transform
 1.42% :     0.000002s :     18: predicate.split_environ_get_set_with_tuple_value
 1.77% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.15% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.72% :     0.000005s :     26: predicate.switch_defer_inline
 2.39% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.75% :     0.000011s :     71: predicate.switch_simplify
 1.47% :     0.000002s :     18: predicate.tile_eliminate
 1.41% :     0.000002s :     18: predicate.transpose_eliminate
 1.60% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.51% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.08% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.76% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.58% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.05% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000003s :     18: predicate.value_based_eliminate
 0.11% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.28% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001190    22
71.60% :     0.000852s :     13: func_graph_cloner_run.FuncGraphClonerGraph
28.40% :     0.000338s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024415    76
 0.14% :     0.000035s :      1: add_recomputation
 0.39% :     0.000096s :      1: auto_monad
 0.06% :     0.000014s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.93% :     0.000228s :      1: bootstrap
 0.10% :     0.000023s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.09% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.21% :     0.000051s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
34.46% :     0.008413s :      1: jit_opt_a
 0.48% :     0.000117s :      1: jit_opt_after_cconv
 0.17% :     0.000042s :      1: jit_opt_b
 1.79% :     0.000437s :      1: loop_unroll
 2.30% :     0.000561s :      1: mutable_eliminate
 5.45% :     0.001330s :     26: opt.transform.jit_opt_a
 0.12% :     0.000029s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000015s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000023s :      4: opt.transform.symbol_engine_opt
 1.86% :     0.000455s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000027s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 5.35% :     0.001305s :      1: renormalize.infer
 1.83% :     0.000447s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000036s :      1: rewriter_after_opt_a
 0.38% :     0.000093s :      1: rewriter_before_opt_a
 0.27% :     0.000066s :      1: symbol_engine_optimizer
 0.10% :     0.000025s :      1: task_emit
42.40% :     0.010353s :      1: type_inference
 0.18% :     0.000044s :      1: validate

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.815.260 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e61f0> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.816.215 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.819.069 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c375c10> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.819.664 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0200767, [30]
[bootstrap]: 0.0001837
[type_inference]: 0.00898964
[event_method]: 4.27e-05
[auto_monad]: 9.048e-05
[graph_reusing]: 6.68e-06
[pre_auto_parallel]: 2.67001e-06
[py_interpret_to_execute]: 2.498e-05
[rewriter_before_opt_a]: 8.757e-05
[expand_dump_flag]: 2.98e-06
[jit_opt_a]: 0.00860651, [2]
    [Cycle 1]: 0.00496888, [27]
        [switch_simplify]: 6.184e-05
        [loop_unroll]: 4.012e-05
        [a_1]: 0.00090912
        [with_stream_mark]: 2.581e-05
        [recompute_prepare]: 1.841e-05
        [updatestate_depend_eliminate]: 7.66999e-06
        [updatestate_assign_eliminate]: 6.39999e-06
        [updatestate_loads_eliminate]: 5.54e-06
        [parameter_eliminate]: 2.26998e-06
        [specialize_transform]: 1.272e-05
        [updatestate_useless_node_eliminater]: 1.216e-05
        [accelerated_algorithm]: 1.227e-05
        [meta_shard_fg_expand]: 3.28e-06
        [get_grad_eliminate_]: 1.213e-05
        [merge_forward]: 7.85e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.526e-05
        [j_node_and_user_rematch]: 2.331e-05
        [meta_fg_expand]: 0.00189054
        [replace_old_param]: 5.509e-05
        [inline_without_move]: 5.3e-05
        [renormalize]: 0.00152096
        [add_forward_monad_depend]: 6.19001e-06
        [auto_monad_grad]: 2.59999e-06
        [auto_monad_eliminator]: 1.135e-05
        [cse]: 2.436e-05
        [replace_applicator]: 1.024e-05
    [Cycle 2]: 0.00028582, [27]
        [switch_simplify]: 4.60999e-06
        [loop_unroll]: 4.09002e-06
        [a_1]: 4.029e-05
        [with_stream_mark]: 9.02e-06
        [recompute_prepare]: 3.68e-06
        [updatestate_depend_eliminate]: 2.39999e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 1.49e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 4.53001e-06
        [updatestate_useless_node_eliminater]: 3.44001e-06
        [accelerated_algorithm]: 4.51002e-06
        [meta_shard_fg_expand]: 1.84e-06
        [get_grad_eliminate_]: 3.78001e-06
        [merge_forward]: 2.50002e-06
        [cell_reuse_recompute_pass]: 1.85001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.208e-05
        [j_node_and_user_rematch]: 6.27001e-06
        [meta_fg_expand]: 2.492e-05
        [replace_old_param]: 3.54002e-06
        [inline_without_move]: 3.3e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.41002e-06
        [auto_monad_grad]: 9.5999e-07
        [auto_monad_eliminator]: 4.2e-06
        [cse]: 1.096e-05
        [replace_applicator]: 4.14002e-06
[py_interpret_to_execute_after_opt_a]: 1.058e-05
[rewriter_after_opt_a]: 3.472e-05
[convert_after_rewriter]: 6.59999e-06
[order_py_execute_after_rewriter]: 4.87998e-06
[mutable_eliminate]: 0.00055217
[jit_opt_b]: 3.948e-05, [1]
    [Cycle 1]: 3.289e-05, [2]
        [frontend_op_eliminate]: 1.043e-05
        [inline_after_opt_a]: 1.126e-05
[cconv]: 2.101e-05
[loop_unroll]: 0.00041695
[jit_opt_after_cconv]: 0.00011968, [1]
    [Cycle 1]: 0.00011287, [11]
        [c_1]: 1.268e-05
        [parameter_eliminate]: 2.10002e-06
        [updatestate_depend_eliminate]: 4.37998e-06
        [updatestate_assign_eliminate]: 1.92001e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [cse]: 1.685e-05
        [call_graph_tuple_transform]: 1.229e-05
        [tuple_list_get_item_eliminator]: 4.53001e-06
        [none_parameter_eliminate]: 1.34e-06
        [renormalize]: 9.40025e-07
        [switch_simplify]: 4.09002e-06
[remove_dup_value]: 1.388e-05
[partial_unused_args_eliminate]: 2.32001e-06
[environ_conv]: 4.93001e-06
[add_recomputation]: 3.236e-05
[cse_after_recomputation]: 1.887e-05, [1]
    [Cycle 1]: 1.37e-05, [1]
        [cse]: 7.90998e-06
[auto_monad_reorder]: 1.266e-05
[get_jit_bprop_graph]: 1.70001e-06
[rewriter_after_jit_bprop_graph]: 4.41002e-06
[opt_after_jit_grad]: 0.00048918
[symbol_engine_optimizer]: 6.708e-05, [1]
    [Cycle 1]: 6.063e-05, [6]
        [build]: 2.97002e-06
        [elim_shapecalc]: 5.74e-06
        [elim_not_effective]: 1.059e-05
        [opt_reshape]: 4.47e-06
        [fold_const_symbol]: 6.44001e-06
        [renormalize]: 7.39994e-07
[validate]: 2.793e-05
Sums
    bootstrap                            :     0.000184s :  1.14%
    type_inference                       :     0.008990s : 55.78%
    event_method                         :     0.000043s :  0.26%
    auto_monad                           :     0.000090s :  0.56%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000025s :  0.15%
    rewriter_before_opt_a                :     0.000088s :  0.54%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.41%
    jit_opt_a.loop_unroll                :     0.000044s :  0.27%
    jit_opt_a.a_1                        :     0.000949s :  5.89%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.10%
    jit_opt_a.merge_forward              :     0.000010s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000030s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001915s : 11.89%
    jit_opt_a.replace_old_param          :     0.000059s :  0.36%
    jit_opt_a.inline_without_move        :     0.000056s :  0.35%
    jit_opt_a.renormalize                :     0.001521s :  9.44%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.10%
    jit_opt_a.cse                        :     0.000035s :  0.22%
    jit_opt_a.replace_applicator         :     0.000014s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000552s :  3.43%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000021s :  0.13%
    loop_unroll                          :     0.000417s :  2.59%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000012s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000032s :  0.20%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000489s :  3.04%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000028s :  0.17%


Time group info:
------[substitution.]   0.000236    50
 0.68% :     0.000002s :      1: substitution.elim_not_effective
 0.46% :     0.000001s :      1: substitution.fold_const_symbol
 1.58% :     0.000004s :      1: substitution.graph_param_transform
64.50% :     0.000152s :      6: substitution.inline
 7.55% :     0.000018s :      2: substitution.inline_without_move
 2.83% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.42% :     0.000003s :      2: substitution.minmaximum_grad
 1.65% :     0.000004s :      3: substitution.partial_eliminate
 2.99% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.25% :     0.000005s :      2: substitution.replace_applicator
 1.80% :     0.000004s :      5: substitution.replace_old_param
 1.18% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.86% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.86% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 6.39% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008924     2
89.06% :     0.007947s :      1: type_inference.infer
10.94% :     0.000976s :      1: type_inference.specialize
------[replace.]   0.000089     8
85.26% :     0.000076s :      6: replace.inline
14.74% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000159     8
93.51% :     0.000149s :      6: match.inline
 6.49% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000170  1109
 1.38% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.69% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.24% :     0.000002s :     18: predicate.addn_check_dump
 1.51% :     0.000003s :     18: predicate.addn_zero_filter
 2.24% :     0.000004s :     18: predicate.arithmetic_simplify
 1.35% :     0.000002s :     18: predicate.cast_eliminate
 0.15% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.26% :     0.000002s :     18: predicate.compare_switch_simplify
 1.43% :     0.000002s :     18: predicate.depend_value_elim
 1.34% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.37% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.38% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      1: predicate.elim_not_effective
 0.23% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.27% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.44% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     18: predicate.environ_get_eliminate
 1.24% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.28% :     0.000002s :     12: predicate.get_grad_eliminate
 0.07% :     0.000000s :      1: predicate.graph_param_transform
 4.54% :     0.000008s :     28: predicate.inline
 3.23% :     0.000006s :     33: predicate.inline_without_move
 0.53% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :     12: predicate.less_batch_normalization
 1.62% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.73% :     0.000003s :     21: predicate.load_eliminater
 0.80% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.80% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.76% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.30% :     0.000002s :     18: predicate.merge_addn
 1.24% :     0.000002s :     18: predicate.minmaximum_grad
 0.97% :     0.000002s :      1: predicate.mutable_eliminate
 0.21% :     0.000000s :      1: predicate.opt_reshape
 2.21% :     0.000004s :     21: predicate.partial_eliminate
 1.47% :     0.000003s :     18: predicate.print_const_string_wrapper
 2.00% :     0.000003s :     18: predicate.reduce_eliminate
 1.54% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.72% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.53% :     0.000003s :     22: predicate.replace_applicator
 1.63% :     0.000003s :     33: predicate.replace_old_param
 0.12% :     0.000000s :      1: predicate.reset_defer_inline
 1.47% :     0.000003s :     18: predicate.reshape_eliminate
 1.44% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.38% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.41% :     0.000002s :     18: predicate.same_eliminate
 0.63% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.35% :     0.000001s :      2: predicate.special_op_eliminate
 1.21% :     0.000002s :     12: predicate.specialize_transform
 1.49% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.15% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.75% :     0.000005s :     26: predicate.switch_defer_inline
 2.39% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.88% :     0.000012s :     71: predicate.switch_simplify
 1.38% :     0.000002s :     18: predicate.tile_eliminate
 1.35% :     0.000002s :     18: predicate.transpose_eliminate
 1.81% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.58% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.20% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.60% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000003s :     18: predicate.value_based_eliminate
 0.18% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.48% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001218    22
78.38% :     0.000954s :     13: func_graph_cloner_run.FuncGraphClonerGraph
21.62% :     0.000263s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022935    72
 0.15% :     0.000035s :      1: add_recomputation
 0.41% :     0.000095s :      1: auto_monad
 0.07% :     0.000015s :      1: auto_monad_reorder
 0.85% :     0.000194s :      1: bootstrap
 0.10% :     0.000024s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.20% :     0.000047s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
37.54% :     0.008610s :      1: jit_opt_a
 0.53% :     0.000122s :      1: jit_opt_after_cconv
 0.18% :     0.000042s :      1: jit_opt_b
 1.85% :     0.000425s :      1: loop_unroll
 2.44% :     0.000560s :      1: mutable_eliminate
 5.73% :     0.001314s :     26: opt.transform.jit_opt_a
 0.13% :     0.000030s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000015s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000024s :      4: opt.transform.symbol_engine_opt
 2.17% :     0.000498s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000028s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 5.02% :     0.001152s :      1: renormalize.infer
 1.57% :     0.000361s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000038s :      1: rewriter_after_opt_a
 0.39% :     0.000090s :      1: rewriter_before_opt_a
 0.30% :     0.000070s :      1: symbol_engine_optimizer
39.25% :     0.009001s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.826.368 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843cdc10> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.826.905 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0208656, [33]
[bootstrap]: 0.00020524
[type_inference]: 0.0102049
[event_method]: 4.957e-05
[auto_monad]: 7.962e-05
[graph_reusing]: 4.97e-06
[pre_auto_parallel]: 2.63e-06
[py_interpret_to_execute]: 2.561e-05
[rewriter_before_opt_a]: 8.658e-05
[expand_dump_flag]: 2.57001e-06
[jit_opt_a]: 0.00824139, [2]
    [Cycle 1]: 0.0045841, [27]
        [switch_simplify]: 6.098e-05
        [loop_unroll]: 4.621e-05
        [a_1]: 0.00079377
        [with_stream_mark]: 2.085e-05
        [recompute_prepare]: 1.687e-05
        [updatestate_depend_eliminate]: 7.24001e-06
        [updatestate_assign_eliminate]: 5.79e-06
        [updatestate_loads_eliminate]: 6.10002e-06
        [parameter_eliminate]: 2.34001e-06
        [specialize_transform]: 1.287e-05
        [updatestate_useless_node_eliminater]: 1.17e-05
        [accelerated_algorithm]: 1.248e-05
        [meta_shard_fg_expand]: 3.23998e-06
        [get_grad_eliminate_]: 1.17e-05
        [merge_forward]: 7.16001e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.4e-05
        [j_node_and_user_rematch]: 2.201e-05
        [meta_fg_expand]: 0.0014286
        [replace_old_param]: 5.027e-05
        [inline_without_move]: 4.96e-05
        [renormalize]: 0.00175051
        [add_forward_monad_depend]: 4.53999e-06
        [auto_monad_grad]: 1.47999e-06
        [auto_monad_eliminator]: 9.27001e-06
        [cse]: 1.644e-05
        [replace_applicator]: 1.135e-05
    [Cycle 2]: 0.00029168, [27]
        [switch_simplify]: 4.32e-06
        [loop_unroll]: 3.76999e-06
        [a_1]: 3.987e-05
        [with_stream_mark]: 8.43001e-06
        [recompute_prepare]: 4.17e-06
        [updatestate_depend_eliminate]: 2.06e-06
        [updatestate_assign_eliminate]: 1.57001e-06
        [updatestate_loads_eliminate]: 1.41002e-06
        [parameter_eliminate]: 1.02998e-06
        [specialize_transform]: 3.49001e-06
        [updatestate_useless_node_eliminater]: 3.48e-06
        [accelerated_algorithm]: 6.69001e-06
        [meta_shard_fg_expand]: 1.20001e-06
        [get_grad_eliminate_]: 3.56001e-06
        [merge_forward]: 2.22999e-06
        [cell_reuse_recompute_pass]: 1.69e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.064e-05
        [j_node_and_user_rematch]: 5.74999e-06
        [meta_fg_expand]: 3.833e-05
        [replace_old_param]: 4e-06
        [inline_without_move]: 3.33e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.42999e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 4.95001e-06
        [cse]: 1.145e-05
        [replace_applicator]: 3.81999e-06
[py_interpret_to_execute_after_opt_a]: 8.07e-06
[rewriter_after_opt_a]: 2.504e-05
[convert_after_rewriter]: 4.85999e-06
[order_py_execute_after_rewriter]: 3.76999e-06
[mutable_eliminate]: 0.00052721
[jit_opt_b]: 3.94e-05, [1]
    [Cycle 1]: 3.302e-05, [2]
        [frontend_op_eliminate]: 1.064e-05
        [inline_after_opt_a]: 1.089e-05
[cconv]: 1.536e-05
[loop_unroll]: 0.00043848
[jit_opt_after_cconv]: 0.00011633, [1]
    [Cycle 1]: 0.00010986, [11]
        [c_1]: 1.105e-05
        [parameter_eliminate]: 2.53e-06
        [updatestate_depend_eliminate]: 4.94998e-06
        [updatestate_assign_eliminate]: 2.01998e-06
        [updatestate_loads_eliminate]: 1.66e-06
        [cse]: 1.706e-05
        [call_graph_tuple_transform]: 1.2e-05
        [tuple_list_get_item_eliminator]: 3.75998e-06
        [none_parameter_eliminate]: 9.50007e-07
        [renormalize]: 4.30009e-07
        [switch_simplify]: 3.61999e-06
[remove_dup_value]: 7.68001e-06
[partial_unused_args_eliminate]: 1.63002e-06
[environ_conv]: 3.25e-06
[add_recomputation]: 2.258e-05
[cse_after_recomputation]: 1.983e-05, [1]
    [Cycle 1]: 1.436e-05, [1]
        [cse]: 8.87999e-06
[auto_monad_reorder]: 7.43e-06
[get_jit_bprop_graph]: 1.34e-06
[rewriter_after_jit_bprop_graph]: 4.82e-06
[opt_after_jit_grad]: 0.00044544
[symbol_engine_optimizer]: 6.294e-05, [1]
    [Cycle 1]: 5.688e-05, [6]
        [build]: 2.61e-06
        [elim_shapecalc]: 5.92999e-06
        [elim_not_effective]: 9.48002e-06
        [opt_reshape]: 4.75001e-06
        [fold_const_symbol]: 6.16e-06
        [renormalize]: 4.7998e-07
[validate]: 2.259e-05
[backend_pass]: 6.39993e-07
[task_emit]: 1.652e-05
[execute]: 1.41002e-06
Sums
    bootstrap                            :     0.000205s :  1.21%
    type_inference                       :     0.010205s : 60.38%
    event_method                         :     0.000050s :  0.29%
    auto_monad                           :     0.000080s :  0.47%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.15%
    rewriter_before_opt_a                :     0.000087s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.39%
    jit_opt_a.loop_unroll                :     0.000050s :  0.30%
    jit_opt_a.a_1                        :     0.000834s :  4.93%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000016s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000019s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.09%
    jit_opt_a.merge_forward              :     0.000009s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001467s :  8.68%
    jit_opt_a.replace_old_param          :     0.000054s :  0.32%
    jit_opt_a.inline_without_move        :     0.000053s :  0.31%
    jit_opt_a.renormalize                :     0.001751s : 10.36%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.08%
    jit_opt_a.cse                        :     0.000028s :  0.17%
    jit_opt_a.replace_applicator         :     0.000015s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000025s :  0.15%
    convert_after_rewriter               :     0.000005s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000527s :  3.12%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.06%
    cconv                                :     0.000015s :  0.09%
    loop_unroll                          :     0.000438s :  2.59%
    jit_opt_after_cconv.c_1              :     0.000011s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000012s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000008s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000003s :  0.02%
    add_recomputation                    :     0.000023s :  0.13%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000007s :  0.04%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000445s :  2.64%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000023s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000017s :  0.10%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000221    50
 0.51% :     0.000001s :      1: substitution.elim_not_effective
 0.43% :     0.000001s :      1: substitution.fold_const_symbol
 1.05% :     0.000002s :      1: substitution.graph_param_transform
66.47% :     0.000147s :      6: substitution.inline
 7.04% :     0.000016s :      2: substitution.inline_without_move
 2.72% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.29% :     0.000003s :      2: substitution.minmaximum_grad
 1.54% :     0.000003s :      3: substitution.partial_eliminate
 3.07% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.37% :     0.000005s :      2: substitution.replace_applicator
 1.61% :     0.000004s :      5: substitution.replace_old_param
 0.98% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.93% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.73% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 6.26% :     0.000014s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010137     2
88.72% :     0.008993s :      1: type_inference.infer
11.28% :     0.001144s :      1: type_inference.specialize
------[replace.]   0.000057     8
77.47% :     0.000044s :      6: replace.inline
22.53% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000153     8
93.89% :     0.000144s :      6: match.inline
 6.11% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000164  1109
 1.62% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.63% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.35% :     0.000002s :     18: predicate.addn_check_dump
 1.51% :     0.000002s :     18: predicate.addn_zero_filter
 1.91% :     0.000003s :     18: predicate.arithmetic_simplify
 1.48% :     0.000002s :     18: predicate.cast_eliminate
 0.22% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.33% :     0.000002s :     18: predicate.compare_switch_simplify
 1.42% :     0.000002s :     18: predicate.depend_value_elim
 1.34% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.46% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.52% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      1: predicate.elim_not_effective
 0.25% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.41% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.34% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     18: predicate.environ_get_eliminate
 1.30% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.16% :     0.000002s :     12: predicate.get_grad_eliminate
 0.11% :     0.000000s :      1: predicate.graph_param_transform
 4.25% :     0.000007s :     28: predicate.inline
 3.23% :     0.000005s :     33: predicate.inline_without_move
 0.49% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.57% :     0.000003s :     12: predicate.less_batch_normalization
 1.62% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.80% :     0.000003s :     21: predicate.load_eliminater
 0.77% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.93% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.86% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.38% :     0.000002s :     18: predicate.merge_addn
 1.30% :     0.000002s :     18: predicate.minmaximum_grad
 0.85% :     0.000001s :      1: predicate.mutable_eliminate
 0.18% :     0.000000s :      1: predicate.opt_reshape
 1.99% :     0.000003s :     21: predicate.partial_eliminate
 1.38% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.87% :     0.000003s :     18: predicate.reduce_eliminate
 1.54% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.73% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.50% :     0.000002s :     22: predicate.replace_applicator
 1.66% :     0.000003s :     33: predicate.replace_old_param
 0.12% :     0.000000s :      1: predicate.reset_defer_inline
 1.64% :     0.000003s :     18: predicate.reshape_eliminate
 1.41% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.61% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     18: predicate.same_eliminate
 0.69% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.32% :     0.000001s :      2: predicate.special_op_eliminate
 1.19% :     0.000002s :     12: predicate.specialize_transform
 1.71% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.55% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.69% :     0.000004s :     26: predicate.switch_defer_inline
 2.35% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.74% :     0.000011s :     71: predicate.switch_simplify
 1.47% :     0.000002s :     18: predicate.tile_eliminate
 1.34% :     0.000002s :     18: predicate.transpose_eliminate
 1.65% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.47% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 2.71% :     0.000004s :     22: predicate.tuple_list_get_item_eliminator
 1.83% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.61% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.15% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000003s :     18: predicate.value_based_eliminate
 0.09% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.23% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001211    22
71.42% :     0.000865s :     13: func_graph_cloner_run.FuncGraphClonerGraph
28.58% :     0.000346s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023869    76
 0.11% :     0.000025s :      1: add_recomputation
 0.35% :     0.000083s :      1: auto_monad
 0.04% :     0.000010s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.91% :     0.000217s :      1: bootstrap
 0.07% :     0.000018s :      1: cconv
 0.03% :     0.000007s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.02% :     0.000005s :      1: environ_conv
 0.22% :     0.000054s :      1: event_method
 0.02% :     0.000004s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000003s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
34.54% :     0.008244s :      1: jit_opt_a
 0.50% :     0.000119s :      1: jit_opt_after_cconv
 0.18% :     0.000042s :      1: jit_opt_b
 1.87% :     0.000446s :      1: loop_unroll
 2.24% :     0.000535s :      1: mutable_eliminate
 5.00% :     0.001194s :     26: opt.transform.jit_opt_a
 0.11% :     0.000027s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000015s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000023s :      4: opt.transform.symbol_engine_opt
 1.90% :     0.000453s :      1: opt_after_jit_grad
 0.02% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000028s :      1: py_interpret_to_execute
 0.04% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000010s :      1: remove_dup_value
 5.51% :     0.001315s :      1: renormalize.infer
 1.79% :     0.000427s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000028s :      1: rewriter_after_opt_a
 0.37% :     0.000089s :      1: rewriter_before_opt_a
 0.27% :     0.000065s :      1: symbol_engine_optimizer
 0.09% :     0.000021s :      1: task_emit
42.80% :     0.010217s :      1: type_inference
 0.15% :     0.000036s :      1: validate


TotalTime = 0.0193371, [30]
[bootstrap]: 0.00018284
[type_inference]: 0.00868335
[event_method]: 4.17e-05
[auto_monad]: 8.204e-05
[graph_reusing]: 6.02999e-06
[pre_auto_parallel]: 2.36e-06
[py_interpret_to_execute]: 2.527e-05
[rewriter_before_opt_a]: 8.546e-05
[expand_dump_flag]: 2.88998e-06
[jit_opt_a]: 0.00823123, [2]
    [Cycle 1]: 0.00475203, [27]
        [switch_simplify]: 5.285e-05
        [loop_unroll]: 3.883e-05
        [a_1]: 0.00085058
        [with_stream_mark]: 1.845e-05
        [recompute_prepare]: 1.703e-05
        [updatestate_depend_eliminate]: 7.20998e-06
        [updatestate_assign_eliminate]: 5.91e-06
        [updatestate_loads_eliminate]: 5.59998e-06
        [parameter_eliminate]: 1.89999e-06
        [specialize_transform]: 4.046e-05
        [updatestate_useless_node_eliminater]: 1.246e-05
        [accelerated_algorithm]: 1.219e-05
        [meta_shard_fg_expand]: 2.77002e-06
        [get_grad_eliminate_]: 1.179e-05
        [merge_forward]: 7.71001e-06
        [cell_reuse_recompute_pass]: 8.00006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.358e-05
        [j_node_and_user_rematch]: 2.223e-05
        [meta_fg_expand]: 0.00182191
        [replace_old_param]: 5.223e-05
        [inline_without_move]: 5.127e-05
        [renormalize]: 0.00145069
        [add_forward_monad_depend]: 5.77999e-06
        [auto_monad_grad]: 2.64001e-06
        [auto_monad_eliminator]: 9.34e-06
        [cse]: 2.193e-05
        [replace_applicator]: 1.096e-05
    [Cycle 2]: 0.00026519, [27]
        [switch_simplify]: 4.53001e-06
        [loop_unroll]: 3.68e-06
        [a_1]: 3.888e-05
        [with_stream_mark]: 9.59999e-06
        [recompute_prepare]: 3.66999e-06
        [updatestate_depend_eliminate]: 2.39001e-06
        [updatestate_assign_eliminate]: 2.00002e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 1.35999e-06
        [specialize_transform]: 3.4e-06
        [updatestate_useless_node_eliminater]: 3.48e-06
        [accelerated_algorithm]: 3.7e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 3.21001e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.46002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.1e-05
        [j_node_and_user_rematch]: 5.77001e-06
        [meta_fg_expand]: 2.297e-05
        [replace_old_param]: 3.63e-06
        [inline_without_move]: 3.31999e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 3.88001e-06
        [cse]: 9.61998e-06
        [replace_applicator]: 3.78001e-06
[py_interpret_to_execute_after_opt_a]: 9.76e-06
[rewriter_after_opt_a]: 3.141e-05
[convert_after_rewriter]: 5.76e-06
[order_py_execute_after_rewriter]: 4.43999e-06
[mutable_eliminate]: 0.0005561
[jit_opt_b]: 4.115e-05, [1]
    [Cycle 1]: 3.388e-05, [2]
        [frontend_op_eliminate]: 1.08e-05
        [inline_after_opt_a]: 1.144e-05
[cconv]: 1.989e-05
[loop_unroll]: 0.00043974
[jit_opt_after_cconv]: 0.00011783, [1]
    [Cycle 1]: 0.00011123, [11]
        [c_1]: 1.279e-05
        [parameter_eliminate]: 3.5e-06
        [updatestate_depend_eliminate]: 4.08999e-06
        [updatestate_assign_eliminate]: 2.13998e-06
        [updatestate_loads_eliminate]: 1.66e-06
        [cse]: 1.37e-05
        [call_graph_tuple_transform]: 1.321e-05
        [tuple_list_get_item_eliminator]: 3.91999e-06
        [none_parameter_eliminate]: 1.34e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 3.68e-06
[remove_dup_value]: 1.36e-05
[partial_unused_args_eliminate]: 2.31998e-06
[environ_conv]: 4.27e-06
[add_recomputation]: 3.172e-05
[cse_after_recomputation]: 1.934e-05, [1]
    [Cycle 1]: 1.398e-05, [1]
        [cse]: 7.98999e-06
[auto_monad_reorder]: 1.122e-05
[get_jit_bprop_graph]: 1.78002e-06
[rewriter_after_jit_bprop_graph]: 4.29002e-06
[opt_after_jit_grad]: 0.00044227
[symbol_engine_optimizer]: 6.249e-05, [1]
    [Cycle 1]: 5.652e-05, [6]
        [build]: 2.80002e-06
        [elim_shapecalc]: 5.72999e-06
        [elim_not_effective]: 1.005e-05
        [opt_reshape]: 4.13001e-06
        [fold_const_symbol]: 5.60001e-06
        [renormalize]: 5.40022e-07
[validate]: 2.488e-05
Sums
    bootstrap                            :     0.000183s :  1.18%
    type_inference                       :     0.008683s : 55.85%
    event_method                         :     0.000042s :  0.27%
    auto_monad                           :     0.000082s :  0.53%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000025s :  0.16%
    rewriter_before_opt_a                :     0.000085s :  0.55%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000057s :  0.37%
    jit_opt_a.loop_unroll                :     0.000043s :  0.27%
    jit_opt_a.a_1                        :     0.000889s :  5.72%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000044s :  0.28%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.10%
    jit_opt_a.merge_forward              :     0.000010s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001845s : 11.87%
    jit_opt_a.replace_old_param          :     0.000056s :  0.36%
    jit_opt_a.inline_without_move        :     0.000055s :  0.35%
    jit_opt_a.renormalize                :     0.001451s :  9.33%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.09%
    jit_opt_a.cse                        :     0.000032s :  0.20%
    jit_opt_a.replace_applicator         :     0.000015s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000556s :  3.58%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000020s :  0.13%
    loop_unroll                          :     0.000440s :  2.83%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000014s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000032s :  0.20%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000011s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000442s :  2.84%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000025s :  0.16%


Time group info:
------[substitution.]   0.000230    50
 0.69% :     0.000002s :      1: substitution.elim_not_effective
 0.37% :     0.000001s :      1: substitution.fold_const_symbol
 1.50% :     0.000003s :      1: substitution.graph_param_transform
67.84% :     0.000156s :      6: substitution.inline
 7.22% :     0.000017s :      2: substitution.inline_without_move
 2.61% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.38% :     0.000003s :      2: substitution.minmaximum_grad
 1.56% :     0.000004s :      3: substitution.partial_eliminate
 2.78% :     0.000006s :      9: substitution.remove_not_recompute_node
 1.71% :     0.000004s :      2: substitution.replace_applicator
 1.79% :     0.000004s :      5: substitution.replace_old_param
 1.04% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.68% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.57% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 5.26% :     0.000012s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008625     2
89.20% :     0.007693s :      1: type_inference.infer
10.80% :     0.000932s :      1: type_inference.specialize
------[replace.]   0.000059     8
77.65% :     0.000046s :      6: replace.inline
22.35% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000160     8
95.41% :     0.000152s :      6: match.inline
 4.59% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000170  1109
 1.52% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.56% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.36% :     0.000002s :     18: predicate.addn_check_dump
 1.79% :     0.000003s :     18: predicate.addn_zero_filter
 2.08% :     0.000004s :     18: predicate.arithmetic_simplify
 1.40% :     0.000002s :     18: predicate.cast_eliminate
 0.25% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.36% :     0.000002s :     18: predicate.compare_switch_simplify
 1.40% :     0.000002s :     18: predicate.depend_value_elim
 1.27% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.47% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.33% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      1: predicate.elim_not_effective
 0.24% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.31% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.26% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.44% :     0.000002s :     18: predicate.environ_get_eliminate
 1.24% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.18% :     0.000002s :     12: predicate.get_grad_eliminate
 0.12% :     0.000000s :      1: predicate.graph_param_transform
 4.14% :     0.000007s :     28: predicate.inline
 3.17% :     0.000005s :     33: predicate.inline_without_move
 0.51% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :     12: predicate.less_batch_normalization
 1.52% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.73% :     0.000003s :     21: predicate.load_eliminater
 0.59% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.74% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.73% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.33% :     0.000002s :     18: predicate.merge_addn
 1.42% :     0.000002s :     18: predicate.minmaximum_grad
 0.77% :     0.000001s :      1: predicate.mutable_eliminate
 0.20% :     0.000000s :      1: predicate.opt_reshape
 2.07% :     0.000004s :     21: predicate.partial_eliminate
 1.37% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.88% :     0.000003s :     18: predicate.reduce_eliminate
 1.62% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.72% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.48% :     0.000003s :     22: predicate.replace_applicator
 1.68% :     0.000003s :     33: predicate.replace_old_param
 0.09% :     0.000000s :      1: predicate.reset_defer_inline
 1.52% :     0.000003s :     18: predicate.reshape_eliminate
 1.45% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.56% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     18: predicate.same_eliminate
 0.65% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.34% :     0.000001s :      2: predicate.special_op_eliminate
 1.27% :     0.000002s :     12: predicate.specialize_transform
 1.59% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.57% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.13% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.94% :     0.000005s :     26: predicate.switch_defer_inline
 2.47% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.66% :     0.000011s :     71: predicate.switch_simplify
 1.57% :     0.000003s :     18: predicate.tile_eliminate
 1.44% :     0.000002s :     18: predicate.transpose_eliminate
 1.69% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.60% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.00% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.55% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.24% :     0.000006s :     33: predicate.updatestate_useless_node_eliminater
 1.98% :     0.000003s :     18: predicate.value_based_eliminate
 0.16% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.26% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001178    22
77.75% :     0.000916s :     13: func_graph_cloner_run.FuncGraphClonerGraph
22.25% :     0.000262s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022072    72
 0.16% :     0.000034s :      1: add_recomputation
 0.39% :     0.000086s :      1: auto_monad
 0.06% :     0.000014s :      1: auto_monad_reorder
 0.88% :     0.000195s :      1: bootstrap
 0.10% :     0.000023s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.10% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.20% :     0.000045s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
37.31% :     0.008234s :      1: jit_opt_a
 0.55% :     0.000121s :      1: jit_opt_after_cconv
 0.20% :     0.000044s :      1: jit_opt_b
 2.03% :     0.000447s :      1: loop_unroll
 2.56% :     0.000564s :      1: mutable_eliminate
 5.71% :     0.001261s :     26: opt.transform.jit_opt_a
 0.14% :     0.000030s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000010s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000014s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000022s :      4: opt.transform.symbol_engine_opt
 2.04% :     0.000449s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.13% :     0.000028s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 4.99% :     0.001102s :      1: renormalize.infer
 1.55% :     0.000341s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000034s :      1: rewriter_after_opt_a
 0.40% :     0.000088s :      1: rewriter_before_opt_a
 0.29% :     0.000065s :      1: symbol_engine_optimizer
39.38% :     0.008692s :      1: type_inference

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.847.502 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c37f040> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.848.113 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0315586, [33]
[bootstrap]: 0.00018576
[type_inference]: 0.00968695
[event_method]: 4.424e-05
[auto_monad]: 8.974e-05
[graph_reusing]: 6.64001e-06
[pre_auto_parallel]: 3.14999e-06
[py_interpret_to_execute]: 2.434e-05
[rewriter_before_opt_a]: 8.812e-05
[expand_dump_flag]: 2.91999e-06
[jit_opt_a]: 0.0114111, [3]
    [Cycle 1]: 0.00570415, [27]
        [switch_simplify]: 6.427e-05
        [loop_unroll]: 3.748e-05
        [a_1]: 0.00082713
        [with_stream_mark]: 2.375e-05
        [recompute_prepare]: 1.767e-05
        [updatestate_depend_eliminate]: 7.09001e-06
        [updatestate_assign_eliminate]: 6.25002e-06
        [updatestate_loads_eliminate]: 5.77999e-06
        [parameter_eliminate]: 2.47001e-06
        [specialize_transform]: 1.283e-05
        [updatestate_useless_node_eliminater]: 1.137e-05
        [accelerated_algorithm]: 1.193e-05
        [meta_shard_fg_expand]: 3.23e-06
        [get_grad_eliminate_]: 1.155e-05
        [merge_forward]: 7.88999e-06
        [cell_reuse_recompute_pass]: 9.20001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.688e-05
        [j_node_and_user_rematch]: 2.233e-05
        [meta_fg_expand]: 0.00137262
        [replace_old_param]: 5.199e-05
        [inline_without_move]: 4.971e-05
        [renormalize]: 0.00264514
        [add_forward_monad_depend]: 8.55001e-06
        [auto_monad_grad]: 5.84e-06
        [auto_monad_eliminator]: 5.001e-05
        [cse]: 0.00016207
        [replace_applicator]: 6.332e-05
    [Cycle 2]: 0.00213444, [27]
        [switch_simplify]: 3.761e-05
        [loop_unroll]: 3.598e-05
        [a_1]: 0.00109003
        [with_stream_mark]: 1.254e-05
        [recompute_prepare]: 8.92999e-06
        [updatestate_depend_eliminate]: 4.25e-06
        [updatestate_assign_eliminate]: 3.23998e-06
        [updatestate_loads_eliminate]: 3.10002e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 7.68999e-06
        [updatestate_useless_node_eliminater]: 7.06001e-06
        [accelerated_algorithm]: 7.53e-06
        [meta_shard_fg_expand]: 2.04999e-06
        [get_grad_eliminate_]: 6.89999e-06
        [merge_forward]: 3.62998e-06
        [cell_reuse_recompute_pass]: 1.52999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.536e-05
        [j_node_and_user_rematch]: 1.168e-05
        [meta_fg_expand]: 4.127e-05
        [replace_old_param]: 1.031e-05
        [inline_without_move]: 7.12002e-06
        [renormalize]: 0.00061246
        [add_forward_monad_depend]: 4.41002e-06
        [auto_monad_grad]: 1.62999e-06
        [auto_monad_eliminator]: 1.283e-05
        [cse]: 2.482e-05
        [replace_applicator]: 1.377e-05
    [Cycle 3]: 0.00040175, [27]
        [switch_simplify]: 7.43e-06
        [loop_unroll]: 6.90998e-06
        [a_1]: 0.0001418
        [with_stream_mark]: 9.22999e-06
        [recompute_prepare]: 6.69001e-06
        [updatestate_depend_eliminate]: 3.74002e-06
        [updatestate_assign_eliminate]: 3.2e-06
        [updatestate_loads_eliminate]: 2.90002e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 6.78e-06
        [updatestate_useless_node_eliminater]: 6.73998e-06
        [accelerated_algorithm]: 7.09001e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 6.53998e-06
        [merge_forward]: 3.23e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.517e-05
        [j_node_and_user_rematch]: 1.169e-05
        [meta_fg_expand]: 2.41998e-06
        [replace_old_param]: 8.63001e-06
        [inline_without_move]: 6.84999e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 6.79982e-07
        [auto_monad_eliminator]: 7.66999e-06
        [cse]: 1.566e-05
        [replace_applicator]: 6.99001e-06
[py_interpret_to_execute_after_opt_a]: 9.00999e-06
[rewriter_after_opt_a]: 4.225e-05
[convert_after_rewriter]: 7.75e-06
[order_py_execute_after_rewriter]: 5.70001e-06
[mutable_eliminate]: 0.00057851
[jit_opt_b]: 6.127e-05, [1]
    [Cycle 1]: 5.379e-05, [2]
        [frontend_op_eliminate]: 2.247e-05
        [inline_after_opt_a]: 2.007e-05
[cconv]: 2.374e-05
[loop_unroll]: 0.00041521
[jit_opt_after_cconv]: 0.00016386, [1]
    [Cycle 1]: 0.00015766, [11]
        [c_1]: 3.047e-05
        [parameter_eliminate]: 2.56998e-06
        [updatestate_depend_eliminate]: 6.26e-06
        [updatestate_assign_eliminate]: 3.18998e-06
        [updatestate_loads_eliminate]: 3.08998e-06
        [cse]: 2.516e-05
        [call_graph_tuple_transform]: 2.225e-05
        [tuple_list_get_item_eliminator]: 7.25e-06
        [none_parameter_eliminate]: 1.35999e-06
        [renormalize]: 4.09986e-07
        [switch_simplify]: 6.86001e-06
[remove_dup_value]: 1.667e-05
[partial_unused_args_eliminate]: 2.20002e-06
[environ_conv]: 6.46e-06
[add_recomputation]: 5.202e-05
[cse_after_recomputation]: 2.844e-05, [1]
    [Cycle 1]: 2.302e-05, [1]
        [cse]: 1.679e-05
[auto_monad_reorder]: 1.933e-05
[get_jit_bprop_graph]: 1.54e-06
[rewriter_after_jit_bprop_graph]: 5.49e-06
[opt_after_jit_grad]: 0.00045214
[symbol_engine_optimizer]: 8.381e-05, [1]
    [Cycle 1]: 7.756e-05, [6]
        [build]: 3.49001e-06
        [elim_shapecalc]: 1.062e-05
        [elim_not_effective]: 1.515e-05
        [opt_reshape]: 7.87e-06
        [fold_const_symbol]: 1.202e-05
        [renormalize]: 5.29981e-07
[validate]: 3.822e-05
[backend_pass]: 7.59988e-07
[task_emit]: 0.00780473
[execute]: 6.60002e-06
Sums
    bootstrap                            :     0.000186s :  0.67%
    type_inference                       :     0.009687s : 35.06%
    event_method                         :     0.000044s :  0.16%
    auto_monad                           :     0.000090s :  0.32%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000024s :  0.09%
    rewriter_before_opt_a                :     0.000088s :  0.32%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000109s :  0.40%
    jit_opt_a.loop_unroll                :     0.000080s :  0.29%
    jit_opt_a.a_1                        :     0.002059s :  7.45%
    jit_opt_a.with_stream_mark           :     0.000046s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000027s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000027s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.09%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000057s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000046s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001416s :  5.13%
    jit_opt_a.replace_old_param          :     0.000071s :  0.26%
    jit_opt_a.inline_without_move        :     0.000064s :  0.23%
    jit_opt_a.renormalize                :     0.003258s : 11.79%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000071s :  0.26%
    jit_opt_a.cse                        :     0.000203s :  0.73%
    jit_opt_a.replace_applicator         :     0.000084s :  0.30%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.03%
    rewriter_after_opt_a                 :     0.000042s :  0.15%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000579s :  2.09%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.07%
    cconv                                :     0.000024s :  0.09%
    loop_unroll                          :     0.000415s :  1.50%
    jit_opt_after_cconv.c_1              :     0.000030s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000017s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000052s :  0.19%
    cse_after_recomputation.cse          :     0.000017s :  0.06%
    auto_monad_reorder                   :     0.000019s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000452s :  1.64%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.14%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007805s : 28.25%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000486   118
 5.00% :     0.000024s :      1: substitution.arithmetic_simplify
 0.51% :     0.000002s :      4: substitution.elim_not_effective
 0.39% :     0.000002s :      4: substitution.fold_const_symbol
 1.37% :     0.000007s :      5: substitution.graph_param_transform
60.90% :     0.000296s :     12: substitution.inline
 3.22% :     0.000016s :      2: substitution.inline_without_move
 1.89% :     0.000009s :     16: substitution.j_node_and_user_rematch
 1.96% :     0.000010s :      7: substitution.minmaximum_grad
 1.26% :     0.000006s :      5: substitution.partial_eliminate
 2.40% :     0.000012s :     16: substitution.remove_not_recompute_node
 4.68% :     0.000023s :     10: substitution.replace_applicator
 1.47% :     0.000007s :      7: substitution.replace_old_param
 0.47% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.13% :     0.000020s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.82% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.53% :     0.000037s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009618     2
88.02% :     0.008466s :      1: type_inference.infer
11.98% :     0.001153s :      1: type_inference.specialize
------[replace.]   0.000147    20
 3.34% :     0.000005s :      1: replace.arithmetic_simplify
62.58% :     0.000092s :     12: replace.inline
34.08% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000332    20
 7.05% :     0.000023s :      1: match.arithmetic_simplify
87.18% :     0.000290s :     12: match.inline
 5.78% :     0.000019s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000385  2790
 1.55% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.44% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.44% :     0.000006s :     46: predicate.addn_check_dump
 1.73% :     0.000007s :     46: predicate.addn_zero_filter
 2.00% :     0.000008s :     47: predicate.arithmetic_simplify
 1.70% :     0.000007s :     47: predicate.cast_eliminate
 0.28% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.46% :     0.000006s :     46: predicate.compare_switch_simplify
 1.59% :     0.000006s :     46: predicate.depend_value_elim
 1.45% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.55% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.48% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.30% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      5: predicate.elim_not_effective
 0.28% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.53% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.47% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.44% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.51% :     0.000006s :     47: predicate.environ_get_eliminate
 1.46% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      5: predicate.fold_const_symbol
 0.91% :     0.000003s :     21: predicate.get_grad_eliminate
 0.13% :     0.000001s :      5: predicate.graph_param_transform
 4.31% :     0.000017s :     76: predicate.inline
 1.74% :     0.000007s :     42: predicate.inline_without_move
 0.37% :     0.000001s :     21: predicate.j_node_and_user_rematch
 0.92% :     0.000004s :     21: predicate.less_batch_normalization
 1.76% :     0.000007s :     54: predicate.list_to_tuple_eliminator_
 1.96% :     0.000008s :     59: predicate.load_eliminater
 0.58% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.02% :     0.000012s :     89: predicate.loop_unroll_before_grad
 1.77% :     0.000007s :     52: predicate.make_slice_get_slice_eliminator
 1.40% :     0.000005s :     46: predicate.merge_addn
 1.51% :     0.000006s :     47: predicate.minmaximum_grad
 0.71% :     0.000003s :      5: predicate.mutable_eliminate
 0.23% :     0.000001s :      5: predicate.opt_reshape
 2.40% :     0.000009s :     59: predicate.partial_eliminate
 1.47% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.82% :     0.000007s :     47: predicate.reduce_eliminate
 1.82% :     0.000007s :     54: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.54% :     0.000010s :    105: predicate.replace_applicator
 0.93% :     0.000004s :     42: predicate.replace_old_param
 0.13% :     0.000000s :      5: predicate.reset_defer_inline
 1.72% :     0.000007s :     47: predicate.reshape_eliminate
 1.50% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.43% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.53% :     0.000006s :     46: predicate.same_eliminate
 0.46% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.46% :     0.000002s :     10: predicate.special_op_eliminate
 0.89% :     0.000003s :     21: predicate.specialize_transform
 1.64% :     0.000006s :     46: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.20% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.70% :     0.000010s :     66: predicate.switch_defer_inline
 2.44% :     0.000009s :     66: predicate.switch_layer_defer_inline
 6.14% :     0.000024s :    160: predicate.switch_simplify
 1.52% :     0.000006s :     47: predicate.tile_eliminate
 1.59% :     0.000006s :     47: predicate.transpose_eliminate
 1.87% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.65% :     0.000006s :     47: predicate.tuple_list_get_item_depend_reorder
 3.25% :     0.000013s :     64: predicate.tuple_list_get_item_eliminator
 1.83% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.90% :     0.000007s :     54: predicate.tuple_to_list_eliminator_
 1.95% :     0.000008s :     59: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000011s :     80: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000007s :     46: predicate.value_based_eliminate
 0.18% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.24% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001475    29
55.31% :     0.000816s :     13: func_graph_cloner_run.FuncGraphClonerGraph
44.69% :     0.000659s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.037629    91
 0.15% :     0.000055s :      1: add_recomputation
 0.25% :     0.000094s :      1: auto_monad
 0.06% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.53% :     0.000198s :      1: bootstrap
 0.07% :     0.000026s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000031s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.13% :     0.000048s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
30.33% :     0.011415s :      1: jit_opt_a
 0.44% :     0.000167s :      1: jit_opt_after_cconv
 0.17% :     0.000064s :      1: jit_opt_b
 1.12% :     0.000422s :      1: loop_unroll
 1.56% :     0.000587s :      1: mutable_eliminate
 7.09% :     0.002666s :     39: opt.transform.jit_opt_a
 0.17% :     0.000064s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000036s :      4: opt.transform.jit_opt_b
 0.04% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000043s :      4: opt.transform.symbol_engine_opt
 1.22% :     0.000459s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000027s :      1: py_interpret_to_execute
 0.03% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 4.61% :     0.001734s :      2: renormalize.infer
 4.01% :     0.001511s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000045s :      1: rewriter_after_opt_a
 0.24% :     0.000091s :      1: rewriter_before_opt_a
 0.23% :     0.000086s :      1: symbol_engine_optimizer
20.77% :     0.007815s :      1: task_emit
25.78% :     0.009701s :      1: type_inference
 0.16% :     0.000060s :      1: validate

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.851.307 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d60d0> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.851.819 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.855.600 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e6670> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.856.155 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0133108, [30]
[bootstrap]: 0.00033678
[type_inference]: 0.00471301
[event_method]: 1.762e-05
[auto_monad]: 6.528e-05
[graph_reusing]: 5.69999e-06
[pre_auto_parallel]: 2.50002e-06
[py_interpret_to_execute]: 3.036e-05
[rewriter_before_opt_a]: 7.233e-05
[expand_dump_flag]: 3.04999e-06
[jit_opt_a]: 0.00574547, [2]
    [Cycle 1]: 0.00171095, [27]
        [switch_simplify]: 5.552e-05
        [loop_unroll]: 2.83e-05
        [a_1]: 0.00054498
        [with_stream_mark]: 2.351e-05
        [recompute_prepare]: 8.15999e-06
        [updatestate_depend_eliminate]: 4e-06
        [updatestate_assign_eliminate]: 3.11999e-06
        [updatestate_loads_eliminate]: 2.41998e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 6.21e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.99e-06
        [meta_shard_fg_expand]: 1.89999e-06
        [get_grad_eliminate_]: 5.05999e-06
        [merge_forward]: 3.81999e-06
        [cell_reuse_recompute_pass]: 1.42e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.488e-05
        [j_node_and_user_rematch]: 9.47999e-06
        [meta_fg_expand]: 2.11998e-06
        [replace_old_param]: 9.31998e-06
        [inline_without_move]: 5.67999e-06
        [renormalize]: 0.00070636
        [add_forward_monad_depend]: 8.35001e-06
        [auto_monad_grad]: 2.98e-06
        [auto_monad_eliminator]: 1.786e-05
        [cse]: 3.32e-05
        [replace_applicator]: 1.343e-05
    [Cycle 2]: 0.00040863, [27]
        [switch_simplify]: 5.59e-06
        [loop_unroll]: 5.36998e-06
        [a_1]: 0.00014836
        [with_stream_mark]: 1.323e-05
        [recompute_prepare]: 6.04001e-06
        [updatestate_depend_eliminate]: 3.41001e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.47001e-06
        [updatestate_useless_node_eliminater]: 5.44e-06
        [accelerated_algorithm]: 5.39e-06
        [meta_shard_fg_expand]: 1.70001e-06
        [get_grad_eliminate_]: 5.92999e-06
        [merge_forward]: 3.28998e-06
        [cell_reuse_recompute_pass]: 2.36e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.696e-05
        [j_node_and_user_rematch]: 7.87e-06
        [meta_fg_expand]: 1.83002e-06
        [replace_old_param]: 7.15e-06
        [inline_without_move]: 5.13002e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.91998e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 7.16999e-06
        [cse]: 1.261e-05
        [replace_applicator]: 6.07001e-06
[py_interpret_to_execute_after_opt_a]: 1.535e-05
[rewriter_after_opt_a]: 4.241e-05
[convert_after_rewriter]: 6.94999e-06
[order_py_execute_after_rewriter]: 5.34e-06
[mutable_eliminate]: 0.00066624
[jit_opt_b]: 5.287e-05, [1]
    [Cycle 1]: 4.565e-05, [2]
        [frontend_op_eliminate]: 1.772e-05
        [inline_after_opt_a]: 1.529e-05
[cconv]: 3.001e-05
[loop_unroll]: 0.00044594
[jit_opt_after_cconv]: 0.00015143, [1]
    [Cycle 1]: 0.0001447, [11]
        [c_1]: 2.328e-05
        [parameter_eliminate]: 3.01001e-06
        [updatestate_depend_eliminate]: 6.53e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 2.044e-05
        [call_graph_tuple_transform]: 2.077e-05
        [tuple_list_get_item_eliminator]: 5.81998e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 6.12999e-06
[remove_dup_value]: 1.566e-05
[partial_unused_args_eliminate]: 2.41998e-06
[environ_conv]: 6.46e-06
[add_recomputation]: 5.452e-05
[cse_after_recomputation]: 2.225e-05, [1]
    [Cycle 1]: 1.627e-05, [1]
        [cse]: 9.61998e-06
[auto_monad_reorder]: 1.608e-05
[get_jit_bprop_graph]: 1.82001e-06
[rewriter_after_jit_bprop_graph]: 6.58e-06
[opt_after_jit_grad]: 0.00047391
[symbol_engine_optimizer]: 7.575e-05, [1]
    [Cycle 1]: 6.88e-05, [6]
        [build]: 3.75e-06
        [elim_shapecalc]: 7.70998e-06
        [elim_not_effective]: 1.313e-05
        [opt_reshape]: 6.07001e-06
        [fold_const_symbol]: 9.14e-06
        [renormalize]: 4.90021e-07
[validate]: 3.873e-05
Sums
    bootstrap                            :     0.000337s :  3.72%
    type_inference                       :     0.004713s : 52.02%
    event_method                         :     0.000018s :  0.19%
    auto_monad                           :     0.000065s :  0.72%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000030s :  0.34%
    rewriter_before_opt_a                :     0.000072s :  0.80%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000061s :  0.67%
    jit_opt_a.loop_unroll                :     0.000034s :  0.37%
    jit_opt_a.a_1                        :     0.000693s :  7.65%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.41%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.12%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.18%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000706s :  7.80%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.11%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.28%
    jit_opt_a.cse                        :     0.000046s :  0.51%
    jit_opt_a.replace_applicator         :     0.000020s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.17%
    rewriter_after_opt_a                 :     0.000042s :  0.47%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000666s :  7.35%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000030s :  0.33%
    loop_unroll                          :     0.000446s :  4.92%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000016s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000055s :  0.60%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000474s :  5.23%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000039s :  0.43%


Time group info:
------[substitution.]   0.000224    23
 0.88% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.74% :     0.000006s :      3: substitution.graph_param_transform
84.25% :     0.000188s :      4: substitution.inline
 1.53% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.17% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000004s :      2: substitution.replace_old_param
 5.75% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004645     2
83.12% :     0.003861s :      1: type_inference.infer
16.88% :     0.000784s :      1: type_inference.specialize
------[replace.]   0.000064     6
67.36% :     0.000043s :      4: replace.inline
32.64% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000198     6
94.03% :     0.000186s :      4: match.inline
 5.97% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.39% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.56% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.17% :     0.000001s :     10: predicate.addn_zero_filter
 2.21% :     0.000003s :     10: predicate.arithmetic_simplify
 1.36% :     0.000002s :     10: predicate.cast_eliminate
 0.67% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     10: predicate.compare_switch_simplify
 1.42% :     0.000002s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.06% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.06% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.19% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000001s :      3: predicate.graph_param_transform
 5.64% :     0.000007s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 1.56% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.69% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.45% :     0.000003s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.37% :     0.000003s :     15: predicate.partial_eliminate
 1.24% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.74% :     0.000002s :     10: predicate.reduce_eliminate
 1.35% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.97% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.52% :     0.000002s :     10: predicate.reshape_eliminate
 1.29% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.74% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.14% :     0.000003s :     16: predicate.switch_defer_inline
 1.93% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.65% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.11% :     0.000001s :     10: predicate.transpose_eliminate
 1.26% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.20% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.66% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.82% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000603    12
52.37% :     0.000316s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.53% :     0.000045s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.10% :     0.000242s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014983    72
 0.38% :     0.000057s :      1: add_recomputation
 0.46% :     0.000069s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 2.38% :     0.000357s :      1: bootstrap
 0.22% :     0.000033s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.15% :     0.000022s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.37% :     0.005749s :      1: jit_opt_a
 1.03% :     0.000154s :      1: jit_opt_after_cconv
 0.37% :     0.000056s :      1: jit_opt_b
 3.03% :     0.000454s :      1: loop_unroll
 4.52% :     0.000677s :      1: mutable_eliminate
 6.09% :     0.000913s :     26: opt.transform.jit_opt_a
 0.34% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000027s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.22% :     0.000482s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.22% :     0.000033s :      1: py_interpret_to_execute
 0.12% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000018s :      1: remove_dup_value
 2.60% :     0.000390s :      1: renormalize.infer
 2.06% :     0.000308s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.30% :     0.000045s :      1: rewriter_after_opt_a
 0.50% :     0.000075s :      1: rewriter_before_opt_a
 0.53% :     0.000079s :      1: symbol_engine_optimizer
31.56% :     0.004729s :      1: type_inference


TotalTime = 0.0222852, [33]
[bootstrap]: 0.00020866
[type_inference]: 0.0108684
[event_method]: 4.838e-05
[auto_monad]: 8.336e-05
[graph_reusing]: 5.71e-06
[pre_auto_parallel]: 2.24999e-06
[py_interpret_to_execute]: 2.518e-05
[rewriter_before_opt_a]: 8.671e-05
[expand_dump_flag]: 2.46998e-06
[jit_opt_a]: 0.00889337, [2]
    [Cycle 1]: 0.00490028, [27]
        [switch_simplify]: 7.053e-05
        [loop_unroll]: 4.052e-05
        [a_1]: 0.00088025
        [with_stream_mark]: 2.567e-05
        [recompute_prepare]: 1.889e-05
        [updatestate_depend_eliminate]: 7.63001e-06
        [updatestate_assign_eliminate]: 5.56e-06
        [updatestate_loads_eliminate]: 5.39e-06
        [parameter_eliminate]: 2.39001e-06
        [specialize_transform]: 1.393e-05
        [updatestate_useless_node_eliminater]: 1.28e-05
        [accelerated_algorithm]: 1.278e-05
        [meta_shard_fg_expand]: 3.01999e-06
        [get_grad_eliminate_]: 1.251e-05
        [merge_forward]: 7.75e-06
        [cell_reuse_recompute_pass]: 8.80013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.781e-05
        [j_node_and_user_rematch]: 2.411e-05
        [meta_fg_expand]: 0.00157405
        [replace_old_param]: 5.113e-05
        [inline_without_move]: 5.348e-05
        [renormalize]: 0.0018065
        [add_forward_monad_depend]: 4.90001e-06
        [auto_monad_grad]: 1.40001e-06
        [auto_monad_eliminator]: 9.37001e-06
        [cse]: 1.834e-05
        [replace_applicator]: 1.096e-05
    [Cycle 2]: 0.00029652, [27]
        [switch_simplify]: 4.79e-06
        [loop_unroll]: 3.97e-06
        [a_1]: 4.486e-05
        [with_stream_mark]: 8.45999e-06
        [recompute_prepare]: 3.60998e-06
        [updatestate_depend_eliminate]: 2.16e-06
        [updatestate_assign_eliminate]: 1.88002e-06
        [updatestate_loads_eliminate]: 1.37999e-06
        [parameter_eliminate]: 1.09998e-06
        [specialize_transform]: 3.69002e-06
        [updatestate_useless_node_eliminater]: 3.6e-06
        [accelerated_algorithm]: 3.97002e-06
        [meta_shard_fg_expand]: 1.48002e-06
        [get_grad_eliminate_]: 3.80998e-06
        [merge_forward]: 2.01e-06
        [cell_reuse_recompute_pass]: 2.68998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.84999e-06
        [j_node_and_user_rematch]: 5.49e-06
        [meta_fg_expand]: 4.306e-05
        [replace_old_param]: 4.23999e-06
        [inline_without_move]: 3.6e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 4.50999e-06
        [cse]: 9.69999e-06
        [replace_applicator]: 3.56999e-06
[py_interpret_to_execute_after_opt_a]: 7.92e-06
[rewriter_after_opt_a]: 2.917e-05
[convert_after_rewriter]: 5.20001e-06
[order_py_execute_after_rewriter]: 1.307e-05
[mutable_eliminate]: 0.00057254
[jit_opt_b]: 4.066e-05, [1]
    [Cycle 1]: 3.372e-05, [2]
        [frontend_op_eliminate]: 1.115e-05
        [inline_after_opt_a]: 1.115e-05
[cconv]: 1.749e-05
[loop_unroll]: 0.00043882
[jit_opt_after_cconv]: 0.00011958, [1]
    [Cycle 1]: 0.00011294, [11]
        [c_1]: 1.221e-05
        [parameter_eliminate]: 2.54999e-06
        [updatestate_depend_eliminate]: 4.63999e-06
        [updatestate_assign_eliminate]: 1.87001e-06
        [updatestate_loads_eliminate]: 1.62001e-06
        [cse]: 1.631e-05
        [call_graph_tuple_transform]: 1.297e-05
        [tuple_list_get_item_eliminator]: 3.94002e-06
        [none_parameter_eliminate]: 1.10001e-06
        [renormalize]: 7.10017e-07
        [switch_simplify]: 4.46002e-06
[remove_dup_value]: 1.103e-05
[partial_unused_args_eliminate]: 2.36998e-06
[environ_conv]: 4e-06
[add_recomputation]: 2.59e-05
[cse_after_recomputation]: 1.947e-05, [1]
    [Cycle 1]: 1.406e-05, [1]
        [cse]: 7.88999e-06
[auto_monad_reorder]: 9.19e-06
[get_jit_bprop_graph]: 1.60999e-06
[rewriter_after_jit_bprop_graph]: 4.94003e-06
[opt_after_jit_grad]: 0.00045765
[symbol_engine_optimizer]: 6.365e-05, [1]
    [Cycle 1]: 5.702e-05, [6]
        [build]: 3.71999e-06
        [elim_shapecalc]: 6.07999e-06
        [elim_not_effective]: 9.36e-06
        [opt_reshape]: 4.28001e-06
        [fold_const_symbol]: 5.39e-06
        [renormalize]: 5.90022e-07
[validate]: 2.245e-05
[backend_pass]: 7.59988e-07
[task_emit]: 1.912e-05
[execute]: 1.28002e-06
Sums
    bootstrap                            :     0.000209s :  1.16%
    type_inference                       :     0.010868s : 60.45%
    event_method                         :     0.000048s :  0.27%
    auto_monad                           :     0.000083s :  0.46%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000025s :  0.14%
    rewriter_before_opt_a                :     0.000087s :  0.48%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000075s :  0.42%
    jit_opt_a.loop_unroll                :     0.000044s :  0.25%
    jit_opt_a.a_1                        :     0.000925s :  5.15%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000018s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.09%
    jit_opt_a.merge_forward              :     0.000010s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000030s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001617s :  8.99%
    jit_opt_a.replace_old_param          :     0.000055s :  0.31%
    jit_opt_a.inline_without_move        :     0.000057s :  0.32%
    jit_opt_a.renormalize                :     0.001807s : 10.05%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.08%
    jit_opt_a.cse                        :     0.000028s :  0.16%
    jit_opt_a.replace_applicator         :     0.000015s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.04%
    rewriter_after_opt_a                 :     0.000029s :  0.16%
    convert_after_rewriter               :     0.000005s :  0.03%
    order_py_execute_after_rewriter      :     0.000013s :  0.07%
    mutable_eliminate                    :     0.000573s :  3.18%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.06%
    cconv                                :     0.000017s :  0.10%
    loop_unroll                          :     0.000439s :  2.44%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000011s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000026s :  0.14%
    cse_after_recomputation.cse          :     0.000008s :  0.04%
    auto_monad_reorder                   :     0.000009s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000458s :  2.55%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000005s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000022s :  0.12%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000019s :  0.11%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000259    50
 0.61% :     0.000002s :      1: substitution.elim_not_effective
 0.30% :     0.000001s :      1: substitution.fold_const_symbol
 1.19% :     0.000003s :      1: substitution.graph_param_transform
67.98% :     0.000176s :      6: substitution.inline
 6.58% :     0.000017s :      2: substitution.inline_without_move
 2.67% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.37% :     0.000004s :      2: substitution.minmaximum_grad
 1.55% :     0.000004s :      3: substitution.partial_eliminate
 3.00% :     0.000008s :      9: substitution.remove_not_recompute_node
 2.35% :     0.000006s :      2: substitution.replace_applicator
 1.37% :     0.000004s :      5: substitution.replace_old_param
 1.03% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.81% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.80% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.37% :     0.000014s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010798     2
88.59% :     0.009566s :      1: type_inference.infer
11.41% :     0.001232s :      1: type_inference.specialize
------[replace.]   0.000069     8
77.40% :     0.000053s :      6: replace.inline
22.60% :     0.000016s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000182     8
94.94% :     0.000173s :      6: match.inline
 5.06% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000178  1109
 1.68% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.68% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.31% :     0.000002s :     18: predicate.addn_check_dump
 1.77% :     0.000003s :     18: predicate.addn_zero_filter
 2.10% :     0.000004s :     18: predicate.arithmetic_simplify
 1.53% :     0.000003s :     18: predicate.cast_eliminate
 0.21% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.32% :     0.000002s :     18: predicate.compare_switch_simplify
 1.55% :     0.000003s :     18: predicate.depend_value_elim
 1.34% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.60% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.54% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      1: predicate.elim_not_effective
 0.40% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.41% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.39% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.39% :     0.000002s :     18: predicate.environ_get_eliminate
 1.25% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.04% :     0.000000s :      1: predicate.fold_const_symbol
 1.14% :     0.000002s :     12: predicate.get_grad_eliminate
 0.07% :     0.000000s :      1: predicate.graph_param_transform
 4.11% :     0.000007s :     28: predicate.inline
 3.29% :     0.000006s :     33: predicate.inline_without_move
 0.49% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :     12: predicate.less_batch_normalization
 1.87% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.72% :     0.000003s :     21: predicate.load_eliminater
 0.64% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.60% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.98% :     0.000004s :     19: predicate.make_slice_get_slice_eliminator
 1.30% :     0.000002s :     18: predicate.merge_addn
 1.30% :     0.000002s :     18: predicate.minmaximum_grad
 0.79% :     0.000001s :      1: predicate.mutable_eliminate
 0.15% :     0.000000s :      1: predicate.opt_reshape
 1.93% :     0.000003s :     21: predicate.partial_eliminate
 1.57% :     0.000003s :     18: predicate.print_const_string_wrapper
 2.21% :     0.000004s :     18: predicate.reduce_eliminate
 1.58% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.43% :     0.000003s :     22: predicate.replace_applicator
 1.46% :     0.000003s :     33: predicate.replace_old_param
 0.11% :     0.000000s :      1: predicate.reset_defer_inline
 1.51% :     0.000003s :     18: predicate.reshape_eliminate
 1.59% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.46% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.37% :     0.000002s :     18: predicate.same_eliminate
 0.62% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.35% :     0.000001s :      2: predicate.special_op_eliminate
 1.19% :     0.000002s :     12: predicate.specialize_transform
 1.62% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.57% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.13% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.52% :     0.000004s :     26: predicate.switch_defer_inline
 2.46% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.62% :     0.000012s :     71: predicate.switch_simplify
 1.45% :     0.000003s :     18: predicate.tile_eliminate
 1.32% :     0.000002s :     18: predicate.transpose_eliminate
 1.75% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.61% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.71% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.55% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.93% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 2.04% :     0.000004s :     18: predicate.value_based_eliminate
 0.09% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.29% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001246    22
71.38% :     0.000889s :     13: func_graph_cloner_run.FuncGraphClonerGraph
28.62% :     0.000357s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025453    76
 0.11% :     0.000029s :      1: add_recomputation
 0.34% :     0.000087s :      1: auto_monad
 0.05% :     0.000012s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.87% :     0.000221s :      1: bootstrap
 0.08% :     0.000020s :      1: cconv
 0.03% :     0.000007s :      1: convert_after_rewriter
 0.08% :     0.000021s :      1: cse_after_recomputation
 0.02% :     0.000006s :      1: environ_conv
 0.21% :     0.000053s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
34.96% :     0.008898s :      1: jit_opt_a
 0.48% :     0.000123s :      1: jit_opt_after_cconv
 0.17% :     0.000043s :      1: jit_opt_b
 1.75% :     0.000447s :      1: loop_unroll
 2.28% :     0.000581s :      1: mutable_eliminate
 5.11% :     0.001301s :     26: opt.transform.jit_opt_a
 0.12% :     0.000031s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000022s :      4: opt.transform.symbol_engine_opt
 1.83% :     0.000466s :      1: opt_after_jit_grad
 0.06% :     0.000016s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.11% :     0.000028s :      1: py_interpret_to_execute
 0.04% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000013s :      1: remove_dup_value
 5.37% :     0.001368s :      1: renormalize.infer
 1.69% :     0.000430s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000032s :      1: rewriter_after_opt_a
 0.35% :     0.000089s :      1: rewriter_before_opt_a
 0.26% :     0.000066s :      1: symbol_engine_optimizer
 0.09% :     0.000023s :      1: task_emit
42.75% :     0.010882s :      1: type_inference
 0.14% :     0.000037s :      1: validate


TotalTime = 0.0198121, [30]
[bootstrap]: 0.00018719
[type_inference]: 0.0088989
[event_method]: 4.489e-05
[auto_monad]: 6.953e-05
[graph_reusing]: 5.67001e-06
[pre_auto_parallel]: 2.23998e-06
[py_interpret_to_execute]: 2.356e-05
[rewriter_before_opt_a]: 0.00013027
[expand_dump_flag]: 2.34001e-06
[jit_opt_a]: 0.00858579, [2]
    [Cycle 1]: 0.00486777, [27]
        [switch_simplify]: 6.018e-05
        [loop_unroll]: 4.185e-05
        [a_1]: 0.00091311
        [with_stream_mark]: 2.021e-05
        [recompute_prepare]: 1.807e-05
        [updatestate_depend_eliminate]: 7.65998e-06
        [updatestate_assign_eliminate]: 5.97001e-06
        [updatestate_loads_eliminate]: 5.55001e-06
        [parameter_eliminate]: 1.40999e-06
        [specialize_transform]: 1.295e-05
        [updatestate_useless_node_eliminater]: 4.05e-05
        [accelerated_algorithm]: 1.277e-05
        [meta_shard_fg_expand]: 2.94999e-06
        [get_grad_eliminate_]: 1.2e-05
        [merge_forward]: 7.35e-06
        [cell_reuse_recompute_pass]: 9.29984e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.352e-05
        [j_node_and_user_rematch]: 2.203e-05
        [meta_fg_expand]: 0.00181395
        [replace_old_param]: 5.594e-05
        [inline_without_move]: 5.558e-05
        [renormalize]: 0.00148734
        [add_forward_monad_depend]: 3.91001e-06
        [auto_monad_grad]: 1.15999e-06
        [auto_monad_eliminator]: 8.33999e-06
        [cse]: 1.627e-05
        [replace_applicator]: 1.095e-05
    [Cycle 2]: 0.00027267, [27]
        [switch_simplify]: 4.45e-06
        [loop_unroll]: 3.58999e-06
        [a_1]: 3.939e-05
        [with_stream_mark]: 8.35999e-06
        [recompute_prepare]: 3.89002e-06
        [updatestate_depend_eliminate]: 2.16e-06
        [updatestate_assign_eliminate]: 1.82999e-06
        [updatestate_loads_eliminate]: 1.56998e-06
        [parameter_eliminate]: 1.21002e-06
        [specialize_transform]: 3.73001e-06
        [updatestate_useless_node_eliminater]: 4.07e-06
        [accelerated_algorithm]: 3.55e-06
        [meta_shard_fg_expand]: 1.18001e-06
        [get_grad_eliminate_]: 3.51999e-06
        [merge_forward]: 2.44999e-06
        [cell_reuse_recompute_pass]: 2.14e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.67001e-06
        [j_node_and_user_rematch]: 5.74e-06
        [meta_fg_expand]: 2.444e-05
        [replace_old_param]: 3.97e-06
        [inline_without_move]: 3.21001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.54998e-06
        [auto_monad_grad]: 7.00005e-07
        [auto_monad_eliminator]: 4.37e-06
        [cse]: 9.74e-06
        [replace_applicator]: 4.23999e-06
[py_interpret_to_execute_after_opt_a]: 7.98001e-06
[rewriter_after_opt_a]: 2.63e-05
[convert_after_rewriter]: 5.30001e-06
[order_py_execute_after_rewriter]: 3.90998e-06
[mutable_eliminate]: 0.00049724
[jit_opt_b]: 3.885e-05, [1]
    [Cycle 1]: 3.236e-05, [2]
        [frontend_op_eliminate]: 9.96e-06
        [inline_after_opt_a]: 1.139e-05
[cconv]: 1.423e-05
[loop_unroll]: 0.00041743
[jit_opt_after_cconv]: 0.00011366, [1]
    [Cycle 1]: 0.00010756, [11]
        [c_1]: 1.208e-05
        [parameter_eliminate]: 2.14e-06
        [updatestate_depend_eliminate]: 4.1e-06
        [updatestate_assign_eliminate]: 1.76998e-06
        [updatestate_loads_eliminate]: 1.60001e-06
        [cse]: 1.622e-05
        [call_graph_tuple_transform]: 1.15e-05
        [tuple_list_get_item_eliminator]: 3.77002e-06
        [none_parameter_eliminate]: 1.02e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 3.73001e-06
[remove_dup_value]: 7.98999e-06
[partial_unused_args_eliminate]: 1.76e-06
[environ_conv]: 3.81001e-06
[add_recomputation]: 2.164e-05
[cse_after_recomputation]: 1.965e-05, [1]
    [Cycle 1]: 1.437e-05, [1]
        [cse]: 8.38001e-06
[auto_monad_reorder]: 7.23999e-06
[get_jit_bprop_graph]: 1.27e-06
[rewriter_after_jit_bprop_graph]: 4.55999e-06
[opt_after_jit_grad]: 0.00043298
[symbol_engine_optimizer]: 6.064e-05, [1]
    [Cycle 1]: 5.486e-05, [6]
        [build]: 2.33998e-06
        [elim_shapecalc]: 5.95002e-06
        [elim_not_effective]: 8.92e-06
        [opt_reshape]: 3.94002e-06
        [fold_const_symbol]: 5.64998e-06
        [renormalize]: 5.19998e-07
[validate]: 2.262e-05
Sums
    bootstrap                            :     0.000187s :  1.19%
    type_inference                       :     0.008899s : 56.42%
    event_method                         :     0.000045s :  0.28%
    auto_monad                           :     0.000070s :  0.44%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000024s :  0.15%
    rewriter_before_opt_a                :     0.000130s :  0.83%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000065s :  0.41%
    jit_opt_a.loop_unroll                :     0.000045s :  0.29%
    jit_opt_a.a_1                        :     0.000952s :  6.04%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000045s :  0.28%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.10%
    jit_opt_a.merge_forward              :     0.000010s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001838s : 11.65%
    jit_opt_a.replace_old_param          :     0.000060s :  0.38%
    jit_opt_a.inline_without_move        :     0.000059s :  0.37%
    jit_opt_a.renormalize                :     0.001487s :  9.43%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.08%
    jit_opt_a.cse                        :     0.000026s :  0.16%
    jit_opt_a.replace_applicator         :     0.000015s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000026s :  0.17%
    convert_after_rewriter               :     0.000005s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000497s :  3.15%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000014s :  0.09%
    loop_unroll                          :     0.000417s :  2.65%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000012s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000008s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000022s :  0.14%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000007s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000433s :  2.74%
    symbol_engine_optimizer.build        :     0.000002s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000023s :  0.14%


Time group info:
------[substitution.]   0.000236    50
 0.41% :     0.000001s :      1: substitution.elim_not_effective
 0.26% :     0.000001s :      1: substitution.fold_const_symbol
 1.00% :     0.000002s :      1: substitution.graph_param_transform
70.01% :     0.000165s :      6: substitution.inline
 7.31% :     0.000017s :      2: substitution.inline_without_move
 2.38% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.11% :     0.000003s :      2: substitution.minmaximum_grad
 1.50% :     0.000004s :      3: substitution.partial_eliminate
 2.67% :     0.000006s :      9: substitution.remove_not_recompute_node
 1.76% :     0.000004s :      2: substitution.replace_applicator
 1.78% :     0.000004s :      5: substitution.replace_old_param
 0.97% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.35% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.70% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 4.81% :     0.000011s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008839     2
88.69% :     0.007839s :      1: type_inference.infer
11.31% :     0.000999s :      1: type_inference.specialize
------[replace.]   0.000062     8
77.58% :     0.000048s :      6: replace.inline
22.42% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000168     8
96.01% :     0.000162s :      6: match.inline
 3.99% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000179  1109
 1.44% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.50% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.28% :     0.000002s :     18: predicate.addn_check_dump
 1.55% :     0.000003s :     18: predicate.addn_zero_filter
 2.08% :     0.000004s :     18: predicate.arithmetic_simplify
 1.46% :     0.000003s :     18: predicate.cast_eliminate
 0.15% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.24% :     0.000002s :     18: predicate.compare_switch_simplify
 1.46% :     0.000003s :     18: predicate.depend_value_elim
 1.29% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.81% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.34% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.43% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.10% :     0.000000s :      1: predicate.elim_not_effective
 0.34% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.39% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.31% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.46% :     0.000003s :     18: predicate.environ_get_depend_swap
 1.56% :     0.000003s :     18: predicate.environ_get_eliminate
 1.60% :     0.000003s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.14% :     0.000002s :     12: predicate.get_grad_eliminate
 0.08% :     0.000000s :      1: predicate.graph_param_transform
 4.14% :     0.000007s :     28: predicate.inline
 3.29% :     0.000006s :     33: predicate.inline_without_move
 0.45% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.24% :     0.000002s :     12: predicate.less_batch_normalization
 1.84% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.88% :     0.000003s :     21: predicate.load_eliminater
 0.61% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.63% :     0.000006s :     44: predicate.loop_unroll_before_grad
 2.07% :     0.000004s :     19: predicate.make_slice_get_slice_eliminator
 1.33% :     0.000002s :     18: predicate.merge_addn
 1.37% :     0.000002s :     18: predicate.minmaximum_grad
 0.58% :     0.000001s :      1: predicate.mutable_eliminate
 0.16% :     0.000000s :      1: predicate.opt_reshape
 2.11% :     0.000004s :     21: predicate.partial_eliminate
 1.33% :     0.000002s :     18: predicate.print_const_string_wrapper
 2.09% :     0.000004s :     18: predicate.reduce_eliminate
 1.51% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.60% :     0.000003s :     22: predicate.replace_applicator
 1.75% :     0.000003s :     33: predicate.replace_old_param
 0.08% :     0.000000s :      1: predicate.reset_defer_inline
 1.55% :     0.000003s :     18: predicate.reshape_eliminate
 1.51% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.34% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     18: predicate.same_eliminate
 0.64% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.30% :     0.000001s :      2: predicate.special_op_eliminate
 1.23% :     0.000002s :     12: predicate.specialize_transform
 1.74% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.69% :     0.000005s :     26: predicate.switch_defer_inline
 2.33% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.97% :     0.000012s :     71: predicate.switch_simplify
 1.45% :     0.000003s :     18: predicate.tile_eliminate
 1.45% :     0.000003s :     18: predicate.transpose_eliminate
 1.70% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.56% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.69% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.67% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.19% :     0.000006s :     33: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000003s :     18: predicate.value_based_eliminate
 0.12% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.24% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001222    22
77.32% :     0.000945s :     13: func_graph_cloner_run.FuncGraphClonerGraph
22.68% :     0.000277s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022664    72
 0.11% :     0.000024s :      1: add_recomputation
 0.33% :     0.000074s :      1: auto_monad
 0.04% :     0.000010s :      1: auto_monad_reorder
 0.89% :     0.000202s :      1: bootstrap
 0.07% :     0.000017s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.21% :     0.000049s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
37.89% :     0.008589s :      1: jit_opt_a
 0.51% :     0.000116s :      1: jit_opt_after_cconv
 0.18% :     0.000041s :      1: jit_opt_b
 1.87% :     0.000424s :      1: loop_unroll
 2.22% :     0.000504s :      1: mutable_eliminate
 5.94% :     0.001346s :     26: opt.transform.jit_opt_a
 0.12% :     0.000028s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000015s :      4: opt.transform.jit_opt_b
 0.03% :     0.000008s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000008s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000014s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000021s :      4: opt.transform.symbol_engine_opt
 1.94% :     0.000440s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000026s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000010s :      1: remove_dup_value
 4.95% :     0.001121s :      1: renormalize.infer
 1.58% :     0.000359s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000029s :      1: rewriter_after_opt_a
 0.59% :     0.000134s :      1: rewriter_before_opt_a
 0.28% :     0.000063s :      1: symbol_engine_optimizer
39.31% :     0.008909s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.876.757 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d6550> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.877.310 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.878.671 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c37f670> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.879.281 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0322298, [33]
[bootstrap]: 0.00018753
[type_inference]: 0.00977086
[event_method]: 4.723e-05
[auto_monad]: 8.387e-05
[graph_reusing]: 5.79999e-06
[pre_auto_parallel]: 2.73e-06
[py_interpret_to_execute]: 2.269e-05
[rewriter_before_opt_a]: 8.42e-05
[expand_dump_flag]: 2.98e-06
[jit_opt_a]: 0.0115852, [3]
    [Cycle 1]: 0.00567685, [27]
        [switch_simplify]: 5.881e-05
        [loop_unroll]: 3.783e-05
        [a_1]: 0.00079869
        [with_stream_mark]: 1.927e-05
        [recompute_prepare]: 1.672e-05
        [updatestate_depend_eliminate]: 6.98e-06
        [updatestate_assign_eliminate]: 5.34998e-06
        [updatestate_loads_eliminate]: 5.61e-06
        [parameter_eliminate]: 1.84e-06
        [specialize_transform]: 1.305e-05
        [updatestate_useless_node_eliminater]: 1.193e-05
        [accelerated_algorithm]: 1.215e-05
        [meta_shard_fg_expand]: 2.36e-06
        [get_grad_eliminate_]: 1.212e-05
        [merge_forward]: 7.2e-06
        [cell_reuse_recompute_pass]: 7.2e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.317e-05
        [j_node_and_user_rematch]: 2.175e-05
        [meta_fg_expand]: 0.0013478
        [replace_old_param]: 4.868e-05
        [inline_without_move]: 5.318e-05
        [renormalize]: 0.00268511
        [add_forward_monad_depend]: 1.134e-05
        [auto_monad_grad]: 5.71998e-06
        [auto_monad_eliminator]: 5.188e-05
        [cse]: 0.00015384
        [replace_applicator]: 7.081e-05
    [Cycle 2]: 0.00229875, [27]
        [switch_simplify]: 3.704e-05
        [loop_unroll]: 3.557e-05
        [a_1]: 0.00114107
        [with_stream_mark]: 1.609e-05
        [recompute_prepare]: 9.77999e-06
        [updatestate_depend_eliminate]: 5.15001e-06
        [updatestate_assign_eliminate]: 3.81999e-06
        [updatestate_loads_eliminate]: 3.63e-06
        [parameter_eliminate]: 1.57999e-06
        [specialize_transform]: 7.63001e-06
        [updatestate_useless_node_eliminater]: 7.26001e-06
        [accelerated_algorithm]: 8.08999e-06
        [meta_shard_fg_expand]: 2.74001e-06
        [get_grad_eliminate_]: 6.66999e-06
        [merge_forward]: 4.4e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.65e-05
        [j_node_and_user_rematch]: 1.257e-05
        [meta_fg_expand]: 5.057e-05
        [replace_old_param]: 1.15e-05
        [inline_without_move]: 7.01999e-06
        [renormalize]: 0.00069375
        [add_forward_monad_depend]: 6.39001e-06
        [auto_monad_grad]: 1.94999e-06
        [auto_monad_eliminator]: 1.484e-05
        [cse]: 2.68e-05
        [replace_applicator]: 1.438e-05
    [Cycle 3]: 0.00041503, [27]
        [switch_simplify]: 7.97e-06
        [loop_unroll]: 6.60002e-06
        [a_1]: 0.00014552
        [with_stream_mark]: 9.46e-06
        [recompute_prepare]: 6.56999e-06
        [updatestate_depend_eliminate]: 3.78001e-06
        [updatestate_assign_eliminate]: 3.11001e-06
        [updatestate_loads_eliminate]: 3.09001e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 6.97002e-06
        [updatestate_useless_node_eliminater]: 6.90002e-06
        [accelerated_algorithm]: 7.01001e-06
        [meta_shard_fg_expand]: 1.87001e-06
        [get_grad_eliminate_]: 6.62002e-06
        [merge_forward]: 3.33e-06
        [cell_reuse_recompute_pass]: 2.07999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.666e-05
        [j_node_and_user_rematch]: 1.173e-05
        [meta_fg_expand]: 2.98e-06
        [replace_old_param]: 9.37999e-06
        [inline_without_move]: 7.95e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.31002e-06
        [auto_monad_grad]: 6.19999e-07
        [auto_monad_eliminator]: 7.45998e-06
        [cse]: 1.754e-05
        [replace_applicator]: 7.08e-06
[py_interpret_to_execute_after_opt_a]: 1.202e-05
[rewriter_after_opt_a]: 4.655e-05
[convert_after_rewriter]: 8.43999e-06
[order_py_execute_after_rewriter]: 5.66998e-06
[mutable_eliminate]: 0.00069997
[jit_opt_b]: 6.237e-05, [1]
    [Cycle 1]: 5.449e-05, [2]
        [frontend_op_eliminate]: 2.302e-05
        [inline_after_opt_a]: 2.05e-05
[cconv]: 2.521e-05
[loop_unroll]: 0.00042146
[jit_opt_after_cconv]: 0.00016526, [1]
    [Cycle 1]: 0.0001593, [11]
        [c_1]: 2.975e-05
        [parameter_eliminate]: 2.02999e-06
        [updatestate_depend_eliminate]: 7.28e-06
        [updatestate_assign_eliminate]: 3.39001e-06
        [updatestate_loads_eliminate]: 2.98998e-06
        [cse]: 2.591e-05
        [call_graph_tuple_transform]: 2.19e-05
        [tuple_list_get_item_eliminator]: 7.47002e-06
        [none_parameter_eliminate]: 1.68002e-06
        [renormalize]: 6.10016e-07
        [switch_simplify]: 7.26999e-06
[remove_dup_value]: 1.669e-05
[partial_unused_args_eliminate]: 1.007e-05
[environ_conv]: 6.41998e-06
[add_recomputation]: 5.26e-05
[cse_after_recomputation]: 3.043e-05, [1]
    [Cycle 1]: 2.464e-05, [1]
        [cse]: 1.799e-05
[auto_monad_reorder]: 1.91e-05
[get_jit_bprop_graph]: 2.26e-06
[rewriter_after_jit_bprop_graph]: 4.35999e-06
[opt_after_jit_grad]: 0.00045985
[symbol_engine_optimizer]: 8.402e-05, [1]
    [Cycle 1]: 7.776e-05, [6]
        [build]: 3.71999e-06
        [elim_shapecalc]: 9.72999e-06
        [elim_not_effective]: 1.603e-05
        [opt_reshape]: 7.66999e-06
        [fold_const_symbol]: 1.253e-05
        [renormalize]: 7.80012e-07
[validate]: 4.128e-05
[backend_pass]: 1.05001e-06
[task_emit]: 0.00806118
[execute]: 7.61999e-06
Sums
    bootstrap                            :     0.000188s :  0.66%
    type_inference                       :     0.009771s : 34.56%
    event_method                         :     0.000047s :  0.17%
    auto_monad                           :     0.000084s :  0.30%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000023s :  0.08%
    rewriter_before_opt_a                :     0.000084s :  0.30%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000104s :  0.37%
    jit_opt_a.loop_unroll                :     0.000080s :  0.28%
    jit_opt_a.a_1                        :     0.002085s :  7.38%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000012s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000028s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000027s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.09%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000056s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000046s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001401s :  4.96%
    jit_opt_a.replace_old_param          :     0.000070s :  0.25%
    jit_opt_a.inline_without_move        :     0.000068s :  0.24%
    jit_opt_a.renormalize                :     0.003379s : 11.95%
    jit_opt_a.add_forward_monad_depend   :     0.000019s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000074s :  0.26%
    jit_opt_a.cse                        :     0.000198s :  0.70%
    jit_opt_a.replace_applicator         :     0.000092s :  0.33%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000047s :  0.16%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000700s :  2.48%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000021s :  0.07%
    cconv                                :     0.000025s :  0.09%
    loop_unroll                          :     0.000421s :  1.49%
    jit_opt_after_cconv.c_1              :     0.000030s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000026s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000017s :  0.06%
    partial_unused_args_eliminate        :     0.000010s :  0.04%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000053s :  0.19%
    cse_after_recomputation.cse          :     0.000018s :  0.06%
    auto_monad_reorder                   :     0.000019s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.02%
    opt_after_jit_grad                   :     0.000460s :  1.63%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000013s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008061s : 28.52%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000498   118
 5.65% :     0.000028s :      1: substitution.arithmetic_simplify
 0.50% :     0.000002s :      4: substitution.elim_not_effective
 0.41% :     0.000002s :      4: substitution.fold_const_symbol
 1.34% :     0.000007s :      5: substitution.graph_param_transform
59.69% :     0.000297s :     12: substitution.inline
 3.54% :     0.000018s :      2: substitution.inline_without_move
 1.90% :     0.000009s :     16: substitution.j_node_and_user_rematch
 1.95% :     0.000010s :      7: substitution.minmaximum_grad
 1.24% :     0.000006s :      5: substitution.partial_eliminate
 2.24% :     0.000011s :     16: substitution.remove_not_recompute_node
 5.04% :     0.000025s :     10: substitution.replace_applicator
 1.46% :     0.000007s :      7: substitution.replace_old_param
 0.47% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.15% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.95% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 7.47% :     0.000037s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009709     2
87.84% :     0.008528s :      1: type_inference.infer
12.16% :     0.001181s :      1: type_inference.specialize
------[replace.]   0.000146    20
 3.89% :     0.000006s :      1: replace.arithmetic_simplify
60.77% :     0.000089s :     12: replace.inline
35.34% :     0.000052s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000336    20
 8.11% :     0.000027s :      1: match.arithmetic_simplify
86.49% :     0.000291s :     12: match.inline
 5.40% :     0.000018s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000390  2790
 1.59% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.44% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.39% :     0.000005s :     46: predicate.addn_check_dump
 1.65% :     0.000006s :     46: predicate.addn_zero_filter
 1.97% :     0.000008s :     47: predicate.arithmetic_simplify
 1.60% :     0.000006s :     47: predicate.cast_eliminate
 0.25% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.45% :     0.000006s :     46: predicate.compare_switch_simplify
 1.58% :     0.000006s :     46: predicate.depend_value_elim
 1.51% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.65% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.67% :     0.000007s :     47: predicate.dict_set_item_eliminator
 0.25% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      5: predicate.elim_not_effective
 0.24% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.52% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.46% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.53% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.68% :     0.000007s :     47: predicate.environ_get_eliminate
 1.47% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      5: predicate.fold_const_symbol
 0.89% :     0.000003s :     21: predicate.get_grad_eliminate
 0.11% :     0.000000s :      5: predicate.graph_param_transform
 4.18% :     0.000016s :     76: predicate.inline
 1.77% :     0.000007s :     42: predicate.inline_without_move
 0.36% :     0.000001s :     21: predicate.j_node_and_user_rematch
 0.98% :     0.000004s :     21: predicate.less_batch_normalization
 1.86% :     0.000007s :     54: predicate.list_to_tuple_eliminator_
 1.99% :     0.000008s :     59: predicate.load_eliminater
 0.52% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.03% :     0.000012s :     89: predicate.loop_unroll_before_grad
 1.83% :     0.000007s :     52: predicate.make_slice_get_slice_eliminator
 1.52% :     0.000006s :     46: predicate.merge_addn
 1.50% :     0.000006s :     47: predicate.minmaximum_grad
 0.65% :     0.000003s :      5: predicate.mutable_eliminate
 0.21% :     0.000001s :      5: predicate.opt_reshape
 2.30% :     0.000009s :     59: predicate.partial_eliminate
 1.53% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.85% :     0.000007s :     47: predicate.reduce_eliminate
 1.75% :     0.000007s :     54: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.52% :     0.000010s :    105: predicate.replace_applicator
 0.88% :     0.000003s :     42: predicate.replace_old_param
 0.12% :     0.000000s :      5: predicate.reset_defer_inline
 1.68% :     0.000007s :     47: predicate.reshape_eliminate
 1.51% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.34% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.47% :     0.000006s :     46: predicate.same_eliminate
 0.46% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.42% :     0.000002s :     10: predicate.special_op_eliminate
 0.89% :     0.000003s :     21: predicate.specialize_transform
 1.65% :     0.000006s :     46: predicate.split_environ_get_set_with_tuple_value
 1.55% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.20% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.86% :     0.000011s :     66: predicate.switch_defer_inline
 2.43% :     0.000009s :     66: predicate.switch_layer_defer_inline
 6.14% :     0.000024s :    160: predicate.switch_simplify
 1.49% :     0.000006s :     47: predicate.tile_eliminate
 1.56% :     0.000006s :     47: predicate.transpose_eliminate
 1.84% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.67% :     0.000006s :     47: predicate.tuple_list_get_item_depend_reorder
 3.20% :     0.000012s :     64: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000007s :     54: predicate.tuple_to_list_eliminator_
 1.93% :     0.000008s :     59: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000011s :     80: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000007s :     46: predicate.value_based_eliminate
 0.17% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.29% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001635    29
53.90% :     0.000881s :     13: func_graph_cloner_run.FuncGraphClonerGraph
46.10% :     0.000754s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038461    91
 0.14% :     0.000056s :      1: add_recomputation
 0.23% :     0.000088s :      1: auto_monad
 0.06% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.53% :     0.000204s :      1: bootstrap
 0.07% :     0.000028s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.08% :     0.000033s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.13% :     0.000051s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
30.13% :     0.011589s :      1: jit_opt_a
 0.44% :     0.000168s :      1: jit_opt_after_cconv
 0.17% :     0.000065s :      1: jit_opt_b
 1.12% :     0.000430s :      1: loop_unroll
 1.85% :     0.000710s :      1: mutable_eliminate
 7.02% :     0.002700s :     39: opt.transform.jit_opt_a
 0.16% :     0.000063s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000037s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000027s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000043s :      4: opt.transform.symbol_engine_opt
 1.22% :     0.000468s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000012s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000025s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000019s :      1: remove_dup_value
 4.57% :     0.001756s :      2: renormalize.infer
 4.19% :     0.001611s :      2: renormalize.specialize
 0.02% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000050s :      1: rewriter_after_opt_a
 0.23% :     0.000087s :      1: rewriter_before_opt_a
 0.22% :     0.000087s :      1: symbol_engine_optimizer
20.99% :     0.008074s :      1: task_emit
25.43% :     0.009781s :      1: type_inference
 0.17% :     0.000064s :      1: validate

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.896.443 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e6af0> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.897.026 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0199907, [30]
[bootstrap]: 0.0001936
[type_inference]: 0.00927114
[event_method]: 4.171e-05
[auto_monad]: 9.246e-05
[graph_reusing]: 7.03e-06
[pre_auto_parallel]: 2.43e-06
[py_interpret_to_execute]: 2.552e-05
[rewriter_before_opt_a]: 8.986e-05
[expand_dump_flag]: 3.11001e-06
[jit_opt_a]: 0.00833259, [2]
    [Cycle 1]: 0.00466829, [27]
        [switch_simplify]: 5.943e-05
        [loop_unroll]: 4.068e-05
        [a_1]: 0.00089495
        [with_stream_mark]: 2.414e-05
        [recompute_prepare]: 1.782e-05
        [updatestate_depend_eliminate]: 8.03001e-06
        [updatestate_assign_eliminate]: 5.87999e-06
        [updatestate_loads_eliminate]: 5.52999e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 1.287e-05
        [updatestate_useless_node_eliminater]: 1.292e-05
        [accelerated_algorithm]: 1.327e-05
        [meta_shard_fg_expand]: 2.86999e-06
        [get_grad_eliminate_]: 1.24e-05
        [merge_forward]: 8.37e-06
        [cell_reuse_recompute_pass]: 8.00006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.366e-05
        [j_node_and_user_rematch]: 2.273e-05
        [meta_fg_expand]: 0.00160132
        [replace_old_param]: 5.304e-05
        [inline_without_move]: 5.177e-05
        [renormalize]: 0.00151571
        [add_forward_monad_depend]: 5.75001e-06
        [auto_monad_grad]: 2.39001e-06
        [auto_monad_eliminator]: 1.106e-05
        [cse]: 2.05e-05
        [replace_applicator]: 1.062e-05
    [Cycle 2]: 0.00026854, [27]
        [switch_simplify]: 4.67e-06
        [loop_unroll]: 3.97002e-06
        [a_1]: 3.886e-05
        [with_stream_mark]: 8.93002e-06
        [recompute_prepare]: 3.53999e-06
        [updatestate_depend_eliminate]: 2.22001e-06
        [updatestate_assign_eliminate]: 1.71e-06
        [updatestate_loads_eliminate]: 1.61998e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 3.55e-06
        [updatestate_useless_node_eliminater]: 3.7e-06
        [accelerated_algorithm]: 3.5e-06
        [meta_shard_fg_expand]: 9.89996e-07
        [get_grad_eliminate_]: 3.5e-06
        [merge_forward]: 2.07001e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.133e-05
        [j_node_and_user_rematch]: 5.88998e-06
        [meta_fg_expand]: 2.368e-05
        [replace_old_param]: 3.83999e-06
        [inline_without_move]: 3.35003e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.64e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 4.13999e-06
        [cse]: 9.86998e-06
        [replace_applicator]: 4.04002e-06
[py_interpret_to_execute_after_opt_a]: 9.20001e-06
[rewriter_after_opt_a]: 3.166e-05
[convert_after_rewriter]: 6.41e-06
[order_py_execute_after_rewriter]: 4.68001e-06
[mutable_eliminate]: 0.00049228
[jit_opt_b]: 4.027e-05, [1]
    [Cycle 1]: 3.353e-05, [2]
        [frontend_op_eliminate]: 1.056e-05
        [inline_after_opt_a]: 1.173e-05
[cconv]: 2.215e-05
[loop_unroll]: 0.00040365
[jit_opt_after_cconv]: 0.0001165, [1]
    [Cycle 1]: 0.00010989, [11]
        [c_1]: 1.144e-05
        [parameter_eliminate]: 2.64999e-06
        [updatestate_depend_eliminate]: 4.38001e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.66002e-06
        [cse]: 1.494e-05
        [call_graph_tuple_transform]: 1.364e-05
        [tuple_list_get_item_eliminator]: 3.82998e-06
        [none_parameter_eliminate]: 1.74e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 3.75998e-06
[remove_dup_value]: 1.194e-05
[partial_unused_args_eliminate]: 1.99999e-06
[environ_conv]: 4.75001e-06
[add_recomputation]: 2.995e-05
[cse_after_recomputation]: 1.923e-05, [1]
    [Cycle 1]: 1.393e-05, [1]
        [cse]: 8.23001e-06
[auto_monad_reorder]: 1.028e-05
[get_jit_bprop_graph]: 1.99e-06
[rewriter_after_jit_bprop_graph]: 4.35e-06
[opt_after_jit_grad]: 0.00048049
[symbol_engine_optimizer]: 6.218e-05, [1]
    [Cycle 1]: 5.639e-05, [6]
        [build]: 2.58e-06
        [elim_shapecalc]: 5.99e-06
        [elim_not_effective]: 9.38002e-06
        [opt_reshape]: 3.95998e-06
        [fold_const_symbol]: 6.01003e-06
        [renormalize]: 5.39992e-07
[validate]: 2.506e-05
Sums
    bootstrap                            :     0.000194s :  1.21%
    type_inference                       :     0.009271s : 58.01%
    event_method                         :     0.000042s :  0.26%
    auto_monad                           :     0.000092s :  0.58%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.16%
    rewriter_before_opt_a                :     0.000090s :  0.56%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.40%
    jit_opt_a.loop_unroll                :     0.000045s :  0.28%
    jit_opt_a.a_1                        :     0.000934s :  5.84%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000016s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000017s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.10%
    jit_opt_a.merge_forward              :     0.000010s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001625s : 10.17%
    jit_opt_a.replace_old_param          :     0.000057s :  0.36%
    jit_opt_a.inline_without_move        :     0.000055s :  0.34%
    jit_opt_a.renormalize                :     0.001516s :  9.48%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.10%
    jit_opt_a.cse                        :     0.000030s :  0.19%
    jit_opt_a.replace_applicator         :     0.000015s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000032s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000492s :  3.08%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.07%
    cconv                                :     0.000022s :  0.14%
    loop_unroll                          :     0.000404s :  2.53%
    jit_opt_after_cconv.c_1              :     0.000011s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000012s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000030s :  0.19%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000010s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000480s :  3.01%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000025s :  0.16%


Time group info:
------[substitution.]   0.000264    50
 0.55% :     0.000001s :      1: substitution.elim_not_effective
 0.35% :     0.000001s :      1: substitution.fold_const_symbol
 1.60% :     0.000004s :      1: substitution.graph_param_transform
68.28% :     0.000180s :      6: substitution.inline
 6.25% :     0.000017s :      2: substitution.inline_without_move
 2.60% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.51% :     0.000004s :      2: substitution.minmaximum_grad
 1.43% :     0.000004s :      3: substitution.partial_eliminate
 2.40% :     0.000006s :      9: substitution.remove_not_recompute_node
 2.18% :     0.000006s :      2: substitution.replace_applicator
 1.60% :     0.000004s :      5: substitution.replace_old_param
 0.90% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.66% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.84% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.86% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009209     2
88.83% :     0.008180s :      1: type_inference.infer
11.17% :     0.001029s :      1: type_inference.specialize
------[replace.]   0.000066     8
77.24% :     0.000051s :      6: replace.inline
22.76% :     0.000015s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000188     8
94.38% :     0.000177s :      6: match.inline
 5.62% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000169  1109
 1.44% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.69% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.26% :     0.000002s :     18: predicate.addn_check_dump
 1.69% :     0.000003s :     18: predicate.addn_zero_filter
 2.35% :     0.000004s :     18: predicate.arithmetic_simplify
 1.50% :     0.000003s :     18: predicate.cast_eliminate
 0.17% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.24% :     0.000002s :     18: predicate.compare_switch_simplify
 1.72% :     0.000003s :     18: predicate.depend_value_elim
 1.27% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.37% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      1: predicate.elim_not_effective
 0.19% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.37% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.25% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.43% :     0.000002s :     18: predicate.environ_get_eliminate
 1.26% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.34% :     0.000002s :     12: predicate.get_grad_eliminate
 0.11% :     0.000000s :      1: predicate.graph_param_transform
 4.38% :     0.000007s :     28: predicate.inline
 3.39% :     0.000006s :     33: predicate.inline_without_move
 0.52% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.52% :     0.000003s :     12: predicate.less_batch_normalization
 1.49% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.66% :     0.000003s :     21: predicate.load_eliminater
 0.55% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.74% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.79% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.25% :     0.000002s :     18: predicate.merge_addn
 1.29% :     0.000002s :     18: predicate.minmaximum_grad
 0.69% :     0.000001s :      1: predicate.mutable_eliminate
 0.15% :     0.000000s :      1: predicate.opt_reshape
 2.11% :     0.000004s :     21: predicate.partial_eliminate
 1.33% :     0.000002s :     18: predicate.print_const_string_wrapper
 2.17% :     0.000004s :     18: predicate.reduce_eliminate
 1.49% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.67% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.50% :     0.000003s :     22: predicate.replace_applicator
 1.84% :     0.000003s :     33: predicate.replace_old_param
 0.10% :     0.000000s :      1: predicate.reset_defer_inline
 1.67% :     0.000003s :     18: predicate.reshape_eliminate
 1.53% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.42% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     18: predicate.same_eliminate
 0.63% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.42% :     0.000001s :      2: predicate.special_op_eliminate
 1.17% :     0.000002s :     12: predicate.specialize_transform
 1.54% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.62% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.91% :     0.000005s :     26: predicate.switch_defer_inline
 2.33% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.91% :     0.000012s :     71: predicate.switch_simplify
 1.43% :     0.000002s :     18: predicate.tile_eliminate
 1.43% :     0.000002s :     18: predicate.transpose_eliminate
 1.76% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.51% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.65% :     0.000004s :     22: predicate.tuple_list_get_item_eliminator
 1.83% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.53% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.16% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000003s :     18: predicate.value_based_eliminate
 0.12% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.27% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001098    22
74.71% :     0.000820s :     13: func_graph_cloner_run.FuncGraphClonerGraph
25.29% :     0.000278s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022818    72
 0.14% :     0.000033s :      1: add_recomputation
 0.42% :     0.000097s :      1: auto_monad
 0.06% :     0.000013s :      1: auto_monad_reorder
 0.89% :     0.000202s :      1: bootstrap
 0.11% :     0.000025s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.20% :     0.000045s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
36.53% :     0.008336s :      1: jit_opt_a
 0.52% :     0.000119s :      1: jit_opt_after_cconv
 0.19% :     0.000043s :      1: jit_opt_b
 1.80% :     0.000412s :      1: loop_unroll
 2.20% :     0.000501s :      1: mutable_eliminate
 5.66% :     0.001291s :     26: opt.transform.jit_opt_a
 0.13% :     0.000029s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000010s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000022s :      4: opt.transform.symbol_engine_opt
 2.14% :     0.000489s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000028s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000014s :      1: remove_dup_value
 5.05% :     0.001152s :      1: renormalize.infer
 1.55% :     0.000355s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000035s :      1: rewriter_after_opt_a
 0.41% :     0.000093s :      1: rewriter_before_opt_a
 0.28% :     0.000065s :      1: symbol_engine_optimizer
40.67% :     0.009281s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.902.710 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d69d0> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.903.227 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.023653, [33]
[bootstrap]: 0.00020542
[type_inference]: 0.0123978
[event_method]: 5.464e-05
[auto_monad]: 9.268e-05
[graph_reusing]: 6.41e-06
[pre_auto_parallel]: 2.61e-06
[py_interpret_to_execute]: 3.104e-05
[rewriter_before_opt_a]: 9.029e-05
[expand_dump_flag]: 3.06999e-06
[jit_opt_a]: 0.00860516, [2]
    [Cycle 1]: 0.00479695, [27]
        [switch_simplify]: 6.05e-05
        [loop_unroll]: 3.803e-05
        [a_1]: 0.00082442
        [with_stream_mark]: 2.11e-05
        [recompute_prepare]: 1.782e-05
        [updatestate_depend_eliminate]: 7.45e-06
        [updatestate_assign_eliminate]: 5.49e-06
        [updatestate_loads_eliminate]: 5.39998e-06
        [parameter_eliminate]: 1.85001e-06
        [specialize_transform]: 1.277e-05
        [updatestate_useless_node_eliminater]: 1.168e-05
        [accelerated_algorithm]: 1.224e-05
        [meta_shard_fg_expand]: 3.04001e-06
        [get_grad_eliminate_]: 1.241e-05
        [merge_forward]: 6.91001e-06
        [cell_reuse_recompute_pass]: 7.59988e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.364e-05
        [j_node_and_user_rematch]: 2.215e-05
        [meta_fg_expand]: 0.00155661
        [replace_old_param]: 5.138e-05
        [inline_without_move]: 5.09e-05
        [renormalize]: 0.00179133
        [add_forward_monad_depend]: 5.79e-06
        [auto_monad_grad]: 1.96998e-06
        [auto_monad_eliminator]: 1.174e-05
        [cse]: 2.11e-05
        [replace_applicator]: 1.29e-05
    [Cycle 2]: 0.00030217, [27]
        [switch_simplify]: 4.97999e-06
        [loop_unroll]: 4.58999e-06
        [a_1]: 4.261e-05
        [with_stream_mark]: 8.65999e-06
        [recompute_prepare]: 4.12998e-06
        [updatestate_depend_eliminate]: 2.39001e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 1.71e-06
        [parameter_eliminate]: 1.70001e-06
        [specialize_transform]: 4.03001e-06
        [updatestate_useless_node_eliminater]: 3.38e-06
        [accelerated_algorithm]: 3.71001e-06
        [meta_shard_fg_expand]: 1.87001e-06
        [get_grad_eliminate_]: 3.2e-06
        [merge_forward]: 2.79001e-06
        [cell_reuse_recompute_pass]: 2.59001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.121e-05
        [j_node_and_user_rematch]: 5.77999e-06
        [meta_fg_expand]: 3.987e-05
        [replace_old_param]: 3.97002e-06
        [inline_without_move]: 3.48e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.75001e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 5.49998e-06
        [cse]: 1.155e-05
        [replace_applicator]: 3.9e-06
[py_interpret_to_execute_after_opt_a]: 1.11e-05
[rewriter_after_opt_a]: 3.445e-05
[convert_after_rewriter]: 6.66999e-06
[order_py_execute_after_rewriter]: 4.55001e-06
[mutable_eliminate]: 0.00067003
[jit_opt_b]: 4.048e-05, [1]
    [Cycle 1]: 3.355e-05, [2]
        [frontend_op_eliminate]: 1.064e-05
        [inline_after_opt_a]: 1.08e-05
[cconv]: 2.079e-05
[loop_unroll]: 0.00043002
[jit_opt_after_cconv]: 0.0001169, [1]
    [Cycle 1]: 0.00011014, [11]
        [c_1]: 1.228e-05
        [parameter_eliminate]: 2.39001e-06
        [updatestate_depend_eliminate]: 4.66002e-06
        [updatestate_assign_eliminate]: 1.76e-06
        [updatestate_loads_eliminate]: 1.55999e-06
        [cse]: 1.549e-05
        [call_graph_tuple_transform]: 1.334e-05
        [tuple_list_get_item_eliminator]: 3.88999e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 3.65998e-06
[remove_dup_value]: 1.204e-05
[partial_unused_args_eliminate]: 2.25002e-06
[environ_conv]: 4.43999e-06
[add_recomputation]: 3.062e-05
[cse_after_recomputation]: 1.919e-05, [1]
    [Cycle 1]: 1.372e-05, [1]
        [cse]: 7.73001e-06
[auto_monad_reorder]: 1.156e-05
[get_jit_bprop_graph]: 1.64998e-06
[rewriter_after_jit_bprop_graph]: 5.09998e-06
[opt_after_jit_grad]: 0.0004545
[symbol_engine_optimizer]: 6.319e-05, [1]
    [Cycle 1]: 5.697e-05, [6]
        [build]: 2.72001e-06
        [elim_shapecalc]: 6.01e-06
        [elim_not_effective]: 9.76998e-06
        [opt_reshape]: 4.13001e-06
        [fold_const_symbol]: 6.04999e-06
        [renormalize]: 4.69998e-07
[validate]: 2.542e-05
[backend_pass]: 8.59989e-07
[task_emit]: 2.009e-05
[execute]: 1.44998e-06
Sums
    bootstrap                            :     0.000205s :  1.05%
    type_inference                       :     0.012398s : 63.50%
    event_method                         :     0.000055s :  0.28%
    auto_monad                           :     0.000093s :  0.47%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000031s :  0.16%
    rewriter_before_opt_a                :     0.000090s :  0.46%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.34%
    jit_opt_a.loop_unroll                :     0.000043s :  0.22%
    jit_opt_a.a_1                        :     0.000867s :  4.44%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.11%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.08%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.08%
    jit_opt_a.merge_forward              :     0.000010s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.14%
    jit_opt_a.meta_fg_expand             :     0.001596s :  8.18%
    jit_opt_a.replace_old_param          :     0.000055s :  0.28%
    jit_opt_a.inline_without_move        :     0.000054s :  0.28%
    jit_opt_a.renormalize                :     0.001791s :  9.17%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.09%
    jit_opt_a.cse                        :     0.000033s :  0.17%
    jit_opt_a.replace_applicator         :     0.000017s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000034s :  0.18%
    convert_after_rewriter               :     0.000007s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000670s :  3.43%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.06%
    cconv                                :     0.000021s :  0.11%
    loop_unroll                          :     0.000430s :  2.20%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000012s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000031s :  0.16%
    cse_after_recomputation.cse          :     0.000008s :  0.04%
    auto_monad_reorder                   :     0.000012s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000455s :  2.33%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000025s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000020s :  0.10%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000236    50
 0.68% :     0.000002s :      1: substitution.elim_not_effective
 0.45% :     0.000001s :      1: substitution.fold_const_symbol
 1.64% :     0.000004s :      1: substitution.graph_param_transform
66.61% :     0.000157s :      6: substitution.inline
 6.94% :     0.000016s :      2: substitution.inline_without_move
 2.81% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.30% :     0.000003s :      2: substitution.minmaximum_grad
 1.57% :     0.000004s :      3: substitution.partial_eliminate
 2.85% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.24% :     0.000005s :      2: substitution.replace_applicator
 1.65% :     0.000004s :      5: substitution.replace_old_param
 1.03% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.65% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.74% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 5.81% :     0.000014s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.012326     2
89.77% :     0.011065s :      1: type_inference.infer
10.23% :     0.001261s :      1: type_inference.specialize
------[replace.]   0.000065     8
78.69% :     0.000051s :      6: replace.inline
21.31% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     8
94.57% :     0.000154s :      6: match.inline
 5.43% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000169  1109
 1.69% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.75% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.33% :     0.000002s :     18: predicate.addn_check_dump
 1.76% :     0.000003s :     18: predicate.addn_zero_filter
 1.91% :     0.000003s :     18: predicate.arithmetic_simplify
 1.52% :     0.000003s :     18: predicate.cast_eliminate
 0.17% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.25% :     0.000002s :     18: predicate.compare_switch_simplify
 1.51% :     0.000003s :     18: predicate.depend_value_elim
 1.28% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.47% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.34% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.39% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.24% :     0.000000s :      1: predicate.elim_not_effective
 0.29% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.55% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.38% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.35% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.50% :     0.000003s :     18: predicate.environ_get_eliminate
 1.27% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.24% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.27% :     0.000007s :     28: predicate.inline
 3.28% :     0.000006s :     33: predicate.inline_without_move
 0.49% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.54% :     0.000003s :     12: predicate.less_batch_normalization
 1.58% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.58% :     0.000003s :     21: predicate.load_eliminater
 0.73% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.60% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.89% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.30% :     0.000002s :     18: predicate.merge_addn
 1.28% :     0.000002s :     18: predicate.minmaximum_grad
 0.76% :     0.000001s :      1: predicate.mutable_eliminate
 0.16% :     0.000000s :      1: predicate.opt_reshape
 1.99% :     0.000003s :     21: predicate.partial_eliminate
 1.36% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.81% :     0.000003s :     18: predicate.reduce_eliminate
 1.57% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.44% :     0.000002s :     22: predicate.replace_applicator
 1.65% :     0.000003s :     33: predicate.replace_old_param
 0.11% :     0.000000s :      1: predicate.reset_defer_inline
 1.52% :     0.000003s :     18: predicate.reshape_eliminate
 1.40% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.58% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     18: predicate.same_eliminate
 0.66% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.40% :     0.000001s :      2: predicate.special_op_eliminate
 1.27% :     0.000002s :     12: predicate.specialize_transform
 1.63% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.52% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.14% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.57% :     0.000004s :     26: predicate.switch_defer_inline
 2.52% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.46% :     0.000011s :     71: predicate.switch_simplify
 1.51% :     0.000003s :     18: predicate.tile_eliminate
 1.29% :     0.000002s :     18: predicate.transpose_eliminate
 1.57% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.42% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 3.12% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.95% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.75% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000003s :     18: predicate.value_based_eliminate
 0.21% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.47% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001277    22
69.99% :     0.000894s :     13: func_graph_cloner_run.FuncGraphClonerGraph
30.01% :     0.000383s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026725    76
 0.12% :     0.000033s :      1: add_recomputation
 0.36% :     0.000097s :      1: auto_monad
 0.05% :     0.000014s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.82% :     0.000219s :      1: bootstrap
 0.09% :     0.000023s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000021s :      1: cse_after_recomputation
 0.02% :     0.000007s :      1: environ_conv
 0.22% :     0.000059s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000009s :      1: graph_reusing
32.21% :     0.008609s :      1: jit_opt_a
 0.45% :     0.000120s :      1: jit_opt_after_cconv
 0.16% :     0.000043s :      1: jit_opt_b
 1.64% :     0.000438s :      1: loop_unroll
 2.54% :     0.000678s :      1: mutable_eliminate
 4.57% :     0.001222s :     26: opt.transform.jit_opt_a
 0.11% :     0.000030s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000015s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000023s :      4: opt.transform.symbol_engine_opt
 1.73% :     0.000462s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000034s :      1: py_interpret_to_execute
 0.05% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000015s :      1: remove_dup_value
 4.95% :     0.001324s :      1: renormalize.infer
 1.72% :     0.000460s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000038s :      1: rewriter_after_opt_a
 0.35% :     0.000094s :      1: rewriter_before_opt_a
 0.25% :     0.000066s :      1: symbol_engine_optimizer
 0.09% :     0.000024s :      1: task_emit
46.43% :     0.012409s :      1: type_inference
 0.15% :     0.000041s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.910.275 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c37faf0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.910.880 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0124448, [30]
[bootstrap]: 0.00030427
[type_inference]: 0.00438035
[event_method]: 1.751e-05
[auto_monad]: 6.85e-05
[graph_reusing]: 6.34999e-06
[pre_auto_parallel]: 2.76999e-06
[py_interpret_to_execute]: 3.227e-05
[rewriter_before_opt_a]: 7.898e-05
[expand_dump_flag]: 3.37002e-06
[jit_opt_a]: 0.00543737, [2]
    [Cycle 1]: 0.00159655, [27]
        [switch_simplify]: 5.659e-05
        [loop_unroll]: 2.808e-05
        [a_1]: 0.00053644
        [with_stream_mark]: 1.951e-05
        [recompute_prepare]: 6.83e-06
        [updatestate_depend_eliminate]: 3.48e-06
        [updatestate_assign_eliminate]: 3.21999e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [parameter_eliminate]: 2.07001e-06
        [specialize_transform]: 5.88998e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 2.07999e-06
        [get_grad_eliminate_]: 4.84e-06
        [merge_forward]: 4e-06
        [cell_reuse_recompute_pass]: 1.18001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.503e-05
        [j_node_and_user_rematch]: 8.98002e-06
        [meta_fg_expand]: 2.36e-06
        [replace_old_param]: 8.22998e-06
        [inline_without_move]: 4.95001e-06
        [renormalize]: 0.00063202
        [add_forward_monad_depend]: 5.01002e-06
        [auto_monad_grad]: 2.33998e-06
        [auto_monad_eliminator]: 1.426e-05
        [cse]: 2.987e-05
        [replace_applicator]: 1.239e-05
    [Cycle 2]: 0.00031286, [27]
        [switch_simplify]: 5.87001e-06
        [loop_unroll]: 4.58999e-06
        [a_1]: 8.714e-05
        [with_stream_mark]: 9.14998e-06
        [recompute_prepare]: 5.05001e-06
        [updatestate_depend_eliminate]: 2.48e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 9.50007e-07
        [specialize_transform]: 4.95001e-06
        [updatestate_useless_node_eliminater]: 4.60001e-06
        [accelerated_algorithm]: 4.77e-06
        [meta_shard_fg_expand]: 1.05001e-06
        [get_grad_eliminate_]: 4.31002e-06
        [merge_forward]: 2.83e-06
        [cell_reuse_recompute_pass]: 1.79e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.263e-05
        [j_node_and_user_rematch]: 7.35003e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 6.79999e-06
        [inline_without_move]: 4.45999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.42999e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 4.76002e-06
        [cse]: 1.069e-05
        [replace_applicator]: 4.82e-06
[py_interpret_to_execute_after_opt_a]: 9.97999e-06
[rewriter_after_opt_a]: 3.679e-05
[convert_after_rewriter]: 6.68e-06
[order_py_execute_after_rewriter]: 5.42001e-06
[mutable_eliminate]: 0.00056092
[jit_opt_b]: 4.962e-05, [1]
    [Cycle 1]: 4.262e-05, [2]
        [frontend_op_eliminate]: 1.589e-05
        [inline_after_opt_a]: 1.499e-05
[cconv]: 2.581e-05
[loop_unroll]: 0.00040875
[jit_opt_after_cconv]: 0.00013461, [1]
    [Cycle 1]: 0.00012841, [11]
        [c_1]: 2.007e-05
        [parameter_eliminate]: 2.31e-06
        [updatestate_depend_eliminate]: 5.47001e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [cse]: 1.718e-05
        [call_graph_tuple_transform]: 1.794e-05
        [tuple_list_get_item_eliminator]: 5.06002e-06
        [none_parameter_eliminate]: 1.37e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 5.02999e-06
[remove_dup_value]: 1.56e-05
[partial_unused_args_eliminate]: 2.89999e-06
[environ_conv]: 5.37001e-06
[add_recomputation]: 7.775e-05
[cse_after_recomputation]: 2.172e-05, [1]
    [Cycle 1]: 1.597e-05, [1]
        [cse]: 9.72001e-06
[auto_monad_reorder]: 1.586e-05
[get_jit_bprop_graph]: 1.96e-06
[rewriter_after_jit_bprop_graph]: 5.41002e-06
[opt_after_jit_grad]: 0.00044597
[symbol_engine_optimizer]: 7.112e-05, [1]
    [Cycle 1]: 6.434e-05, [6]
        [build]: 3.93999e-06
        [elim_shapecalc]: 7.58001e-06
        [elim_not_effective]: 1.237e-05
        [opt_reshape]: 5.24e-06
        [fold_const_symbol]: 8.23999e-06
        [renormalize]: 4.69998e-07
[validate]: 3.483e-05
Sums
    bootstrap                            :     0.000304s :  3.65%
    type_inference                       :     0.004380s : 52.56%
    event_method                         :     0.000018s :  0.21%
    auto_monad                           :     0.000068s :  0.82%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.39%
    rewriter_before_opt_a                :     0.000079s :  0.95%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000062s :  0.75%
    jit_opt_a.loop_unroll                :     0.000033s :  0.39%
    jit_opt_a.a_1                        :     0.000624s :  7.48%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.34%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.18%
    jit_opt_a.inline_without_move        :     0.000009s :  0.11%
    jit_opt_a.renormalize                :     0.000632s :  7.59%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.23%
    jit_opt_a.cse                        :     0.000041s :  0.49%
    jit_opt_a.replace_applicator         :     0.000017s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000037s :  0.44%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000561s :  6.73%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000026s :  0.31%
    loop_unroll                          :     0.000409s :  4.91%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000016s :  0.19%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000078s :  0.93%
    cse_after_recomputation.cse          :     0.000010s :  0.12%
    auto_monad_reorder                   :     0.000016s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000446s :  5.35%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000035s :  0.42%


Time group info:
------[substitution.]   0.000216    23
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.47% :     0.000005s :      3: substitution.graph_param_transform
85.38% :     0.000184s :      4: substitution.inline
 1.31% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.04% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.75% :     0.000004s :      2: substitution.replace_old_param
 5.41% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004307     2
81.15% :     0.003496s :      1: type_inference.infer
18.85% :     0.000812s :      1: type_inference.specialize
------[replace.]   0.000064     6
67.84% :     0.000043s :      4: replace.inline
32.16% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     6
94.50% :     0.000182s :      4: match.inline
 5.50% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.38% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000001s :     10: predicate.addn_zero_filter
 1.74% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.55% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000001s :     10: predicate.depend_value_elim
 1.26% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.08% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.20% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.65% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.25% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.90% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.02% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.34% :     0.000003s :     15: predicate.partial_eliminate
 1.35% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.71% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.53% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.25% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.54% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.04% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.06% :     0.000009s :     53: predicate.switch_simplify
 1.28% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.31% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.95% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.67% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.78% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.62% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000575    12
50.60% :     0.000291s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.14% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.26% :     0.000243s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013949    72
 0.58% :     0.000081s :      1: add_recomputation
 0.52% :     0.000072s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 2.32% :     0.000324s :      1: bootstrap
 0.20% :     0.000028s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000024s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.15% :     0.000022s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
39.00% :     0.005441s :      1: jit_opt_a
 0.99% :     0.000137s :      1: jit_opt_after_cconv
 0.37% :     0.000052s :      1: jit_opt_b
 2.98% :     0.000416s :      1: loop_unroll
 4.08% :     0.000569s :      1: mutable_eliminate
 5.93% :     0.000827s :     26: opt.transform.jit_opt_a
 0.32% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.25% :     0.000453s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.04% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000035s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000018s :      1: remove_dup_value
 2.42% :     0.000337s :      1: renormalize.infer
 2.06% :     0.000287s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.29% :     0.000040s :      1: rewriter_after_opt_a
 0.59% :     0.000082s :      1: rewriter_before_opt_a
 0.53% :     0.000074s :      1: symbol_engine_optimizer
31.52% :     0.004397s :      1: type_inference


TotalTime = 0.0192936, [30]
[bootstrap]: 0.00018451
[type_inference]: 0.00876329
[event_method]: 4.269e-05
[auto_monad]: 8.314e-05
[graph_reusing]: 6.36e-06
[pre_auto_parallel]: 2.44999e-06
[py_interpret_to_execute]: 2.638e-05
[rewriter_before_opt_a]: 9.281e-05
[expand_dump_flag]: 2.79999e-06
[jit_opt_a]: 0.00818924, [2]
    [Cycle 1]: 0.00464295, [27]
        [switch_simplify]: 5.437e-05
        [loop_unroll]: 3.939e-05
        [a_1]: 0.0008441
        [with_stream_mark]: 2.156e-05
        [recompute_prepare]: 1.725e-05
        [updatestate_depend_eliminate]: 7.34002e-06
        [updatestate_assign_eliminate]: 6.16998e-06
        [updatestate_loads_eliminate]: 5.89999e-06
        [parameter_eliminate]: 2.26e-06
        [specialize_transform]: 1.393e-05
        [updatestate_useless_node_eliminater]: 1.226e-05
        [accelerated_algorithm]: 1.21e-05
        [meta_shard_fg_expand]: 3.61999e-06
        [get_grad_eliminate_]: 1.18e-05
        [merge_forward]: 7.85998e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.447e-05
        [j_node_and_user_rematch]: 2.255e-05
        [meta_fg_expand]: 0.00176904
        [replace_old_param]: 4.802e-05
        [inline_without_move]: 5.133e-05
        [renormalize]: 0.00141574
        [add_forward_monad_depend]: 3.80998e-06
        [auto_monad_grad]: 1.07e-06
        [auto_monad_eliminator]: 7.3e-06
        [cse]: 1.351e-05
        [replace_applicator]: 8.99e-06
    [Cycle 2]: 0.00026419, [27]
        [switch_simplify]: 4.37998e-06
        [loop_unroll]: 3.55e-06
        [a_1]: 3.909e-05
        [with_stream_mark]: 7.66999e-06
        [recompute_prepare]: 3.63e-06
        [updatestate_depend_eliminate]: 2.10002e-06
        [updatestate_assign_eliminate]: 1.66002e-06
        [updatestate_loads_eliminate]: 1.47001e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 3.97998e-06
        [updatestate_useless_node_eliminater]: 3.43e-06
        [accelerated_algorithm]: 3.70998e-06
        [meta_shard_fg_expand]: 1.02e-06
        [get_grad_eliminate_]: 3.67998e-06
        [merge_forward]: 1.96998e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.37999e-06
        [j_node_and_user_rematch]: 5.25001e-06
        [meta_fg_expand]: 2.332e-05
        [replace_old_param]: 3.76001e-06
        [inline_without_move]: 3.31999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.52999e-06
        [auto_monad_grad]: 7.80012e-07
        [auto_monad_eliminator]: 4.08999e-06
        [cse]: 9.10999e-06
        [replace_applicator]: 3.73999e-06
[py_interpret_to_execute_after_opt_a]: 8.37e-06
[rewriter_after_opt_a]: 2.545e-05
[convert_after_rewriter]: 5.46e-06
[order_py_execute_after_rewriter]: 3.50998e-06
[mutable_eliminate]: 0.00052193
[jit_opt_b]: 3.939e-05, [1]
    [Cycle 1]: 3.319e-05, [2]
        [frontend_op_eliminate]: 1.071e-05
        [inline_after_opt_a]: 1.127e-05
[cconv]: 1.472e-05
[loop_unroll]: 0.00042125
[jit_opt_after_cconv]: 0.00011564, [1]
    [Cycle 1]: 0.00010928, [11]
        [c_1]: 1.288e-05
        [parameter_eliminate]: 2.51e-06
        [updatestate_depend_eliminate]: 4.1e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [cse]: 1.498e-05
        [call_graph_tuple_transform]: 1.136e-05
        [tuple_list_get_item_eliminator]: 4.02e-06
        [none_parameter_eliminate]: 1.10999e-06
        [renormalize]: 5.50004e-07
        [switch_simplify]: 3.62002e-06
[remove_dup_value]: 8.42998e-06
[partial_unused_args_eliminate]: 1.76e-06
[environ_conv]: 3.6e-06
[add_recomputation]: 2.289e-05
[cse_after_recomputation]: 1.95e-05, [1]
    [Cycle 1]: 1.412e-05, [1]
        [cse]: 8.06001e-06
[auto_monad_reorder]: 8.10999e-06
[get_jit_bprop_graph]: 1.36002e-06
[rewriter_after_jit_bprop_graph]: 4.42998e-06
[opt_after_jit_grad]: 0.00044288
[symbol_engine_optimizer]: 6.112e-05, [1]
    [Cycle 1]: 5.564e-05, [6]
        [build]: 2.44001e-06
        [elim_shapecalc]: 5.67999e-06
        [elim_not_effective]: 9.29e-06
        [opt_reshape]: 4.22998e-06
        [fold_const_symbol]: 5.59e-06
        [renormalize]: 5.3001e-07
[validate]: 2.156e-05
Sums
    bootstrap                            :     0.000185s :  1.20%
    type_inference                       :     0.008763s : 56.86%
    event_method                         :     0.000043s :  0.28%
    auto_monad                           :     0.000083s :  0.54%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.17%
    rewriter_before_opt_a                :     0.000093s :  0.60%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.38%
    jit_opt_a.loop_unroll                :     0.000043s :  0.28%
    jit_opt_a.a_1                        :     0.000883s :  5.73%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000018s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.10%
    jit_opt_a.merge_forward              :     0.000010s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001792s : 11.63%
    jit_opt_a.replace_old_param          :     0.000052s :  0.34%
    jit_opt_a.inline_without_move        :     0.000055s :  0.35%
    jit_opt_a.renormalize                :     0.001416s :  9.19%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000011s :  0.07%
    jit_opt_a.cse                        :     0.000023s :  0.15%
    jit_opt_a.replace_applicator         :     0.000013s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000025s :  0.17%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000522s :  3.39%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000015s :  0.10%
    loop_unroll                          :     0.000421s :  2.73%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000011s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000008s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000023s :  0.15%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000008s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000443s :  2.87%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000022s :  0.14%


Time group info:
------[substitution.]   0.000209    50
 0.66% :     0.000001s :      1: substitution.elim_not_effective
 0.38% :     0.000001s :      1: substitution.fold_const_symbol
 1.07% :     0.000002s :      1: substitution.graph_param_transform
65.74% :     0.000138s :      6: substitution.inline
 7.76% :     0.000016s :      2: substitution.inline_without_move
 2.92% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.41% :     0.000003s :      2: substitution.minmaximum_grad
 1.58% :     0.000003s :      3: substitution.partial_eliminate
 3.07% :     0.000006s :      9: substitution.remove_not_recompute_node
 1.78% :     0.000004s :      2: substitution.replace_applicator
 1.69% :     0.000004s :      5: substitution.replace_old_param
 0.99% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.94% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.87% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 6.13% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008705     2
88.78% :     0.007729s :      1: type_inference.infer
11.22% :     0.000977s :      1: type_inference.specialize
------[replace.]   0.000059     8
76.72% :     0.000045s :      6: replace.inline
23.28% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000143     8
94.09% :     0.000134s :      6: match.inline
 5.91% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000166  1109
 1.39% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.70% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.30% :     0.000002s :     18: predicate.addn_check_dump
 1.65% :     0.000003s :     18: predicate.addn_zero_filter
 2.27% :     0.000004s :     18: predicate.arithmetic_simplify
 1.46% :     0.000002s :     18: predicate.cast_eliminate
 0.27% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.29% :     0.000002s :     18: predicate.compare_switch_simplify
 1.44% :     0.000002s :     18: predicate.depend_value_elim
 1.31% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.44% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.41% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.37% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      1: predicate.elim_not_effective
 0.20% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.27% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.40% :     0.000002s :     18: predicate.environ_get_eliminate
 1.27% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.27% :     0.000002s :     12: predicate.get_grad_eliminate
 0.14% :     0.000000s :      1: predicate.graph_param_transform
 4.29% :     0.000007s :     28: predicate.inline
 3.31% :     0.000006s :     33: predicate.inline_without_move
 0.50% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :     12: predicate.less_batch_normalization
 1.63% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.69% :     0.000003s :     21: predicate.load_eliminater
 0.55% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.73% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.81% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.29% :     0.000002s :     18: predicate.merge_addn
 1.41% :     0.000002s :     18: predicate.minmaximum_grad
 0.77% :     0.000001s :      1: predicate.mutable_eliminate
 0.22% :     0.000000s :      1: predicate.opt_reshape
 1.99% :     0.000003s :     21: predicate.partial_eliminate
 1.38% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.94% :     0.000003s :     18: predicate.reduce_eliminate
 1.54% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.52% :     0.000003s :     22: predicate.replace_applicator
 1.59% :     0.000003s :     33: predicate.replace_old_param
 0.11% :     0.000000s :      1: predicate.reset_defer_inline
 1.52% :     0.000003s :     18: predicate.reshape_eliminate
 1.56% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.36% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     18: predicate.same_eliminate
 0.67% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.43% :     0.000001s :      2: predicate.special_op_eliminate
 1.33% :     0.000002s :     12: predicate.specialize_transform
 1.69% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.13% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.82% :     0.000005s :     26: predicate.switch_defer_inline
 2.37% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.70% :     0.000011s :     71: predicate.switch_simplify
 1.48% :     0.000002s :     18: predicate.tile_eliminate
 1.57% :     0.000003s :     18: predicate.transpose_eliminate
 1.70% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.57% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.89% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.59% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.10% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.91% :     0.000003s :     18: predicate.value_based_eliminate
 0.10% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.28% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001169    22
77.14% :     0.000902s :     13: func_graph_cloner_run.FuncGraphClonerGraph
22.86% :     0.000267s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021960    72
 0.12% :     0.000026s :      1: add_recomputation
 0.40% :     0.000087s :      1: auto_monad
 0.05% :     0.000011s :      1: auto_monad_reorder
 0.89% :     0.000196s :      1: bootstrap
 0.08% :     0.000017s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.21% :     0.000047s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
37.31% :     0.008193s :      1: jit_opt_a
 0.54% :     0.000118s :      1: jit_opt_after_cconv
 0.19% :     0.000042s :      1: jit_opt_b
 1.95% :     0.000429s :      1: loop_unroll
 2.41% :     0.000529s :      1: mutable_eliminate
 5.58% :     0.001225s :     26: opt.transform.jit_opt_a
 0.13% :     0.000028s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000015s :      4: opt.transform.jit_opt_b
 0.04% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000009s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000021s :      4: opt.transform.symbol_engine_opt
 2.05% :     0.000450s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.13% :     0.000029s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000011s :      1: remove_dup_value
 4.89% :     0.001073s :      1: renormalize.infer
 1.53% :     0.000336s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000029s :      1: rewriter_after_opt_a
 0.43% :     0.000095s :      1: rewriter_before_opt_a
 0.29% :     0.000064s :      1: symbol_engine_optimizer
39.95% :     0.008772s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.928.223 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d6e50> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.928.762 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0319905, [33]
[bootstrap]: 0.00025064
[type_inference]: 0.00980965
[event_method]: 4.342e-05
[auto_monad]: 8.954e-05
[graph_reusing]: 6.71e-06
[pre_auto_parallel]: 2.87002e-06
[py_interpret_to_execute]: 2.377e-05
[rewriter_before_opt_a]: 8.757e-05
[expand_dump_flag]: 3.04001e-06
[jit_opt_a]: 0.0114802, [3]
    [Cycle 1]: 0.00576558, [27]
        [switch_simplify]: 6.564e-05
        [loop_unroll]: 3.794e-05
        [a_1]: 0.00082661
        [with_stream_mark]: 2.279e-05
        [recompute_prepare]: 1.782e-05
        [updatestate_depend_eliminate]: 7.43e-06
        [updatestate_assign_eliminate]: 6.44001e-06
        [updatestate_loads_eliminate]: 6.34001e-06
        [parameter_eliminate]: 2.19001e-06
        [specialize_transform]: 1.273e-05
        [updatestate_useless_node_eliminater]: 1.223e-05
        [accelerated_algorithm]: 1.208e-05
        [meta_shard_fg_expand]: 3.2e-06
        [get_grad_eliminate_]: 1.177e-05
        [merge_forward]: 7.58001e-06
        [cell_reuse_recompute_pass]: 9.50007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.454e-05
        [j_node_and_user_rematch]: 2.241e-05
        [meta_fg_expand]: 0.00144492
        [replace_old_param]: 5.061e-05
        [inline_without_move]: 4.988e-05
        [renormalize]: 0.00264012
        [add_forward_monad_depend]: 9.32001e-06
        [auto_monad_grad]: 6.04001e-06
        [auto_monad_eliminator]: 5.147e-05
        [cse]: 0.00016062
        [replace_applicator]: 6.126e-05
    [Cycle 2]: 0.00211569, [27]
        [switch_simplify]: 3.705e-05
        [loop_unroll]: 3.585e-05
        [a_1]: 0.00113552
        [with_stream_mark]: 1.197e-05
        [recompute_prepare]: 9.45001e-06
        [updatestate_depend_eliminate]: 4.05e-06
        [updatestate_assign_eliminate]: 3.40003e-06
        [updatestate_loads_eliminate]: 3.03e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 7.43e-06
        [updatestate_useless_node_eliminater]: 6.98e-06
        [accelerated_algorithm]: 7.46001e-06
        [meta_shard_fg_expand]: 1.87001e-06
        [get_grad_eliminate_]: 6.76e-06
        [merge_forward]: 3.61999e-06
        [cell_reuse_recompute_pass]: 9.79984e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.468e-05
        [j_node_and_user_rematch]: 1.127e-05
        [meta_fg_expand]: 4.137e-05
        [replace_old_param]: 1.023e-05
        [inline_without_move]: 7.01001e-06
        [renormalize]: 0.00055276
        [add_forward_monad_depend]: 3.91999e-06
        [auto_monad_grad]: 1.42e-06
        [auto_monad_eliminator]: 1.2e-05
        [cse]: 2.53e-05
        [replace_applicator]: 1.426e-05
    [Cycle 3]: 0.00040267, [27]
        [switch_simplify]: 7.73001e-06
        [loop_unroll]: 6.91001e-06
        [a_1]: 0.00014194
        [with_stream_mark]: 9.07999e-06
        [recompute_prepare]: 6.66999e-06
        [updatestate_depend_eliminate]: 3.61999e-06
        [updatestate_assign_eliminate]: 3.10998e-06
        [updatestate_loads_eliminate]: 2.99999e-06
        [parameter_eliminate]: 9.30013e-07
        [specialize_transform]: 6.91999e-06
        [updatestate_useless_node_eliminater]: 6.93e-06
        [accelerated_algorithm]: 6.94001e-06
        [meta_shard_fg_expand]: 1.32999e-06
        [get_grad_eliminate_]: 6.44999e-06
        [merge_forward]: 3.37997e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.521e-05
        [j_node_and_user_rematch]: 1.132e-05
        [meta_fg_expand]: 2.39001e-06
        [replace_old_param]: 8.42e-06
        [inline_without_move]: 6.61e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.07998e-06
        [auto_monad_grad]: 9.29984e-07
        [auto_monad_eliminator]: 7.7e-06
        [cse]: 1.67e-05
        [replace_applicator]: 6.73e-06
[py_interpret_to_execute_after_opt_a]: 1.157e-05
[rewriter_after_opt_a]: 4.426e-05
[convert_after_rewriter]: 7.94002e-06
[order_py_execute_after_rewriter]: 5.95002e-06
[mutable_eliminate]: 0.00059412
[jit_opt_b]: 6.129e-05, [1]
    [Cycle 1]: 5.34e-05, [2]
        [frontend_op_eliminate]: 2.214e-05
        [inline_after_opt_a]: 1.96e-05
[cconv]: 1.991e-05
[loop_unroll]: 0.00041898
[jit_opt_after_cconv]: 0.00016209, [1]
    [Cycle 1]: 0.00015497, [11]
        [c_1]: 2.935e-05
        [parameter_eliminate]: 2.24999e-06
        [updatestate_depend_eliminate]: 6.49001e-06
        [updatestate_assign_eliminate]: 3.16001e-06
        [updatestate_loads_eliminate]: 2.99999e-06
        [cse]: 2.416e-05
        [call_graph_tuple_transform]: 2.294e-05
        [tuple_list_get_item_eliminator]: 6.94999e-06
        [none_parameter_eliminate]: 1.30001e-06
        [renormalize]: 7.80012e-07
        [switch_simplify]: 7.07002e-06
[remove_dup_value]: 1.753e-05
[partial_unused_args_eliminate]: 2.02999e-06
[environ_conv]: 6.38e-06
[add_recomputation]: 5.411e-05
[cse_after_recomputation]: 2.94e-05, [1]
    [Cycle 1]: 2.301e-05, [1]
        [cse]: 1.689e-05
[auto_monad_reorder]: 1.931e-05
[get_jit_bprop_graph]: 1.86e-06
[rewriter_after_jit_bprop_graph]: 4.51002e-06
[opt_after_jit_grad]: 0.00045318
[symbol_engine_optimizer]: 8.158e-05, [1]
    [Cycle 1]: 7.576e-05, [6]
        [build]: 3.61001e-06
        [elim_shapecalc]: 9.12999e-06
        [elim_not_effective]: 1.593e-05
        [opt_reshape]: 7.83999e-06
        [fold_const_symbol]: 1.207e-05
        [renormalize]: 5.10016e-07
[validate]: 3.781e-05
[backend_pass]: 9.80013e-07
[task_emit]: 0.00796396
[execute]: 8.15e-06
Sums
    bootstrap                            :     0.000251s :  0.89%
    type_inference                       :     0.009810s : 34.98%
    event_method                         :     0.000043s :  0.15%
    auto_monad                           :     0.000090s :  0.32%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000024s :  0.08%
    rewriter_before_opt_a                :     0.000088s :  0.31%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000110s :  0.39%
    jit_opt_a.loop_unroll                :     0.000081s :  0.29%
    jit_opt_a.a_1                        :     0.002104s :  7.50%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000034s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000027s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000026s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.09%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000054s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000045s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001489s :  5.31%
    jit_opt_a.replace_old_param          :     0.000069s :  0.25%
    jit_opt_a.inline_without_move        :     0.000064s :  0.23%
    jit_opt_a.renormalize                :     0.003193s : 11.39%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000071s :  0.25%
    jit_opt_a.cse                        :     0.000203s :  0.72%
    jit_opt_a.replace_applicator         :     0.000082s :  0.29%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000044s :  0.16%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000594s :  2.12%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000020s :  0.07%
    cconv                                :     0.000020s :  0.07%
    loop_unroll                          :     0.000419s :  1.49%
    jit_opt_after_cconv.c_1              :     0.000029s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000024s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000018s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000054s :  0.19%
    cse_after_recomputation.cse          :     0.000017s :  0.06%
    auto_monad_reorder                   :     0.000019s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000453s :  1.62%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007964s : 28.40%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000480   118
 5.66% :     0.000027s :      1: substitution.arithmetic_simplify
 0.58% :     0.000003s :      4: substitution.elim_not_effective
 0.39% :     0.000002s :      4: substitution.fold_const_symbol
 1.37% :     0.000007s :      5: substitution.graph_param_transform
60.01% :     0.000288s :     12: substitution.inline
 3.31% :     0.000016s :      2: substitution.inline_without_move
 1.93% :     0.000009s :     16: substitution.j_node_and_user_rematch
 1.96% :     0.000009s :      7: substitution.minmaximum_grad
 1.27% :     0.000006s :      5: substitution.partial_eliminate
 2.34% :     0.000011s :     16: substitution.remove_not_recompute_node
 4.98% :     0.000024s :     10: substitution.replace_applicator
 1.50% :     0.000007s :      7: substitution.replace_old_param
 0.50% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.04% :     0.000019s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.69% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 7.46% :     0.000036s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009744     2
88.15% :     0.008590s :      1: type_inference.infer
11.85% :     0.001154s :      1: type_inference.specialize
------[replace.]   0.000144    20
 3.54% :     0.000005s :      1: replace.arithmetic_simplify
61.67% :     0.000089s :     12: replace.inline
34.79% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000327    20
 8.04% :     0.000026s :      1: match.arithmetic_simplify
85.97% :     0.000281s :     12: match.inline
 5.99% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000385  2790
 1.60% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.41% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.43% :     0.000006s :     46: predicate.addn_check_dump
 1.64% :     0.000006s :     46: predicate.addn_zero_filter
 2.04% :     0.000008s :     47: predicate.arithmetic_simplify
 1.68% :     0.000006s :     47: predicate.cast_eliminate
 0.20% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.42% :     0.000005s :     46: predicate.compare_switch_simplify
 1.62% :     0.000006s :     46: predicate.depend_value_elim
 1.48% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.60% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.62% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.35% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      5: predicate.elim_not_effective
 0.22% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.50% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.52% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.46% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.69% :     0.000007s :     47: predicate.environ_get_eliminate
 1.44% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.11% :     0.000000s :      5: predicate.fold_const_symbol
 0.88% :     0.000003s :     21: predicate.get_grad_eliminate
 0.12% :     0.000000s :      5: predicate.graph_param_transform
 4.25% :     0.000016s :     76: predicate.inline
 1.74% :     0.000007s :     42: predicate.inline_without_move
 0.37% :     0.000001s :     21: predicate.j_node_and_user_rematch
 0.97% :     0.000004s :     21: predicate.less_batch_normalization
 1.79% :     0.000007s :     54: predicate.list_to_tuple_eliminator_
 2.08% :     0.000008s :     59: predicate.load_eliminater
 0.46% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.16% :     0.000012s :     89: predicate.loop_unroll_before_grad
 1.95% :     0.000008s :     52: predicate.make_slice_get_slice_eliminator
 1.47% :     0.000006s :     46: predicate.merge_addn
 1.60% :     0.000006s :     47: predicate.minmaximum_grad
 0.57% :     0.000002s :      5: predicate.mutable_eliminate
 0.27% :     0.000001s :      5: predicate.opt_reshape
 2.40% :     0.000009s :     59: predicate.partial_eliminate
 1.47% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.83% :     0.000007s :     47: predicate.reduce_eliminate
 1.81% :     0.000007s :     54: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.48% :     0.000010s :    105: predicate.replace_applicator
 0.84% :     0.000003s :     42: predicate.replace_old_param
 0.13% :     0.000001s :      5: predicate.reset_defer_inline
 1.63% :     0.000006s :     47: predicate.reshape_eliminate
 1.50% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.39% :     0.000002s :      5: predicate.row_tensor_eliminate
 1.55% :     0.000006s :     46: predicate.same_eliminate
 0.47% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.44% :     0.000002s :     10: predicate.special_op_eliminate
 0.90% :     0.000003s :     21: predicate.specialize_transform
 1.61% :     0.000006s :     46: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.20% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.65% :     0.000010s :     66: predicate.switch_defer_inline
 2.42% :     0.000009s :     66: predicate.switch_layer_defer_inline
 6.27% :     0.000024s :    160: predicate.switch_simplify
 1.62% :     0.000006s :     47: predicate.tile_eliminate
 1.54% :     0.000006s :     47: predicate.transpose_eliminate
 1.76% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.64% :     0.000006s :     47: predicate.tuple_list_get_item_depend_reorder
 3.02% :     0.000012s :     64: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000007s :     54: predicate.tuple_to_list_eliminator_
 1.94% :     0.000007s :     59: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000012s :     80: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000007s :     46: predicate.value_based_eliminate
 0.18% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.25% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001509    29
57.49% :     0.000868s :     13: func_graph_cloner_run.FuncGraphClonerGraph
42.51% :     0.000642s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038041    91
 0.15% :     0.000057s :      1: add_recomputation
 0.25% :     0.000093s :      1: auto_monad
 0.06% :     0.000022s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.68% :     0.000259s :      1: bootstrap
 0.06% :     0.000022s :      1: cconv
 0.03% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000032s :      1: cse_after_recomputation
 0.04% :     0.000016s :      1: environ_conv
 0.12% :     0.000047s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
30.19% :     0.011484s :      1: jit_opt_a
 0.43% :     0.000165s :      1: jit_opt_after_cconv
 0.17% :     0.000064s :      1: jit_opt_b
 1.12% :     0.000427s :      1: loop_unroll
 1.59% :     0.000603s :      1: mutable_eliminate
 7.12% :     0.002707s :     39: opt.transform.jit_opt_a
 0.17% :     0.000063s :      4: opt.transform.jit_opt_after_cconv
 0.09% :     0.000036s :      4: opt.transform.jit_opt_b
 0.04% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000042s :      4: opt.transform.symbol_engine_opt
 1.21% :     0.000460s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000026s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000020s :      1: remove_dup_value
 4.51% :     0.001717s :      2: renormalize.infer
 3.84% :     0.001462s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000047s :      1: rewriter_after_opt_a
 0.24% :     0.000090s :      1: rewriter_before_opt_a
 0.22% :     0.000084s :      1: symbol_engine_optimizer
20.96% :     0.007975s :      1: task_emit
25.82% :     0.009822s :      1: type_inference
 0.16% :     0.000060s :      1: validate


TotalTime = 0.0209016, [33]
[bootstrap]: 0.00019663
[type_inference]: 0.00987779
[event_method]: 4.594e-05
[auto_monad]: 8.664e-05
[graph_reusing]: 5.69999e-06
[pre_auto_parallel]: 2.37001e-06
[py_interpret_to_execute]: 2.447e-05
[rewriter_before_opt_a]: 8.642e-05
[expand_dump_flag]: 2.89999e-06
[jit_opt_a]: 0.00849005, [2]
    [Cycle 1]: 0.00494661, [27]
        [switch_simplify]: 5.626e-05
        [loop_unroll]: 3.82e-05
        [a_1]: 0.00083504
        [with_stream_mark]: 2.45e-05
        [recompute_prepare]: 1.969e-05
        [updatestate_depend_eliminate]: 7.28e-06
        [updatestate_assign_eliminate]: 5.67999e-06
        [updatestate_loads_eliminate]: 6.04001e-06
        [parameter_eliminate]: 2.12999e-06
        [specialize_transform]: 1.884e-05
        [updatestate_useless_node_eliminater]: 1.236e-05
        [accelerated_algorithm]: 1.236e-05
        [meta_shard_fg_expand]: 3.54002e-06
        [get_grad_eliminate_]: 1.238e-05
        [merge_forward]: 7.67002e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.503e-05
        [j_node_and_user_rematch]: 2.199e-05
        [meta_fg_expand]: 0.00162864
        [replace_old_param]: 5.795e-05
        [inline_without_move]: 5.396e-05
        [renormalize]: 0.00184702
        [add_forward_monad_depend]: 5.91e-06
        [auto_monad_grad]: 2.72001e-06
        [auto_monad_eliminator]: 1.221e-05
        [cse]: 1.825e-05
        [replace_applicator]: 1.197e-05
    [Cycle 2]: 0.0002928, [27]
        [switch_simplify]: 4.48999e-06
        [loop_unroll]: 3.69002e-06
        [a_1]: 4.236e-05
        [with_stream_mark]: 9.66e-06
        [recompute_prepare]: 3.89002e-06
        [updatestate_depend_eliminate]: 2.32999e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 1.65001e-06
        [parameter_eliminate]: 1.58002e-06
        [specialize_transform]: 3.61999e-06
        [updatestate_useless_node_eliminater]: 3.42997e-06
        [accelerated_algorithm]: 4.2e-06
        [meta_shard_fg_expand]: 1.41998e-06
        [get_grad_eliminate_]: 3.28998e-06
        [merge_forward]: 2.83e-06
        [cell_reuse_recompute_pass]: 2.82002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.152e-05
        [j_node_and_user_rematch]: 5.85002e-06
        [meta_fg_expand]: 3.727e-05
        [replace_old_param]: 3.91999e-06
        [inline_without_move]: 3.40998e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.66e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 4.52e-06
        [cse]: 1.032e-05
        [replace_applicator]: 4.09002e-06
[py_interpret_to_execute_after_opt_a]: 1.063e-05
[rewriter_after_opt_a]: 3.201e-05
[convert_after_rewriter]: 6.34999e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00058322
[jit_opt_b]: 3.915e-05, [1]
    [Cycle 1]: 3.238e-05, [2]
        [frontend_op_eliminate]: 1.024e-05
        [inline_after_opt_a]: 1.091e-05
[cconv]: 2.16e-05
[loop_unroll]: 0.00042198
[jit_opt_after_cconv]: 0.00011566, [1]
    [Cycle 1]: 0.00010946, [11]
        [c_1]: 1.188e-05
        [parameter_eliminate]: 2.41998e-06
        [updatestate_depend_eliminate]: 4.71002e-06
        [updatestate_assign_eliminate]: 1.83002e-06
        [updatestate_loads_eliminate]: 1.57999e-06
        [cse]: 1.495e-05
        [call_graph_tuple_transform]: 1.327e-05
        [tuple_list_get_item_eliminator]: 3.83001e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 3.61001e-06
[remove_dup_value]: 1.298e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 4.84e-06
[add_recomputation]: 3.241e-05
[cse_after_recomputation]: 1.934e-05, [1]
    [Cycle 1]: 1.408e-05, [1]
        [cse]: 7.96001e-06
[auto_monad_reorder]: 1.121e-05
[get_jit_bprop_graph]: 2.17001e-06
[rewriter_after_jit_bprop_graph]: 5.29e-06
[opt_after_jit_grad]: 0.00044636
[symbol_engine_optimizer]: 6.237e-05, [1]
    [Cycle 1]: 5.662e-05, [6]
        [build]: 3.18998e-06
        [elim_shapecalc]: 5.37999e-06
        [elim_not_effective]: 1.002e-05
        [opt_reshape]: 4.19002e-06
        [fold_const_symbol]: 5.87001e-06
        [renormalize]: 3.89991e-07
[validate]: 2.794e-05
[backend_pass]: 8.50006e-07
[task_emit]: 1.954e-05
[execute]: 1.30999e-06
Sums
    bootstrap                            :     0.000197s :  1.16%
    type_inference                       :     0.009878s : 58.03%
    event_method                         :     0.000046s :  0.27%
    auto_monad                           :     0.000087s :  0.51%
    graph_reusing                        :     0.000006s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000024s :  0.14%
    rewriter_before_opt_a                :     0.000086s :  0.51%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.36%
    jit_opt_a.loop_unroll                :     0.000042s :  0.25%
    jit_opt_a.a_1                        :     0.000877s :  5.15%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000024s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000022s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.09%
    jit_opt_a.merge_forward              :     0.000011s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001666s :  9.79%
    jit_opt_a.replace_old_param          :     0.000062s :  0.36%
    jit_opt_a.inline_without_move        :     0.000057s :  0.34%
    jit_opt_a.renormalize                :     0.001847s : 10.85%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.10%
    jit_opt_a.cse                        :     0.000029s :  0.17%
    jit_opt_a.replace_applicator         :     0.000016s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.06%
    rewriter_after_opt_a                 :     0.000032s :  0.19%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000583s :  3.43%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.06%
    cconv                                :     0.000022s :  0.13%
    loop_unroll                          :     0.000422s :  2.48%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000032s :  0.19%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000011s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000446s :  2.62%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000028s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000020s :  0.11%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000245    50
 0.66% :     0.000002s :      1: substitution.elim_not_effective
 0.36% :     0.000001s :      1: substitution.fold_const_symbol
 1.65% :     0.000004s :      1: substitution.graph_param_transform
65.23% :     0.000160s :      6: substitution.inline
 7.65% :     0.000019s :      2: substitution.inline_without_move
 2.30% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.35% :     0.000003s :      2: substitution.minmaximum_grad
 1.71% :     0.000004s :      3: substitution.partial_eliminate
 2.77% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.20% :     0.000005s :      2: substitution.replace_applicator
 1.89% :     0.000005s :      5: substitution.replace_old_param
 1.26% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.94% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.80% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 6.22% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009813     2
89.19% :     0.008753s :      1: type_inference.infer
10.81% :     0.001061s :      1: type_inference.specialize
------[replace.]   0.000059     8
76.81% :     0.000046s :      6: replace.inline
23.19% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000167     8
93.76% :     0.000157s :      6: match.inline
 6.24% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000172  1109
 1.77% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.67% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.28% :     0.000002s :     18: predicate.addn_check_dump
 1.44% :     0.000002s :     18: predicate.addn_zero_filter
 2.17% :     0.000004s :     18: predicate.arithmetic_simplify
 1.62% :     0.000003s :     18: predicate.cast_eliminate
 0.13% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.35% :     0.000002s :     18: predicate.compare_switch_simplify
 1.46% :     0.000003s :     18: predicate.depend_value_elim
 1.25% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.54% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.32% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.53% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      1: predicate.elim_not_effective
 0.23% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.24% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.33% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.40% :     0.000002s :     18: predicate.environ_get_eliminate
 1.26% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.20% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.36% :     0.000008s :     28: predicate.inline
 3.13% :     0.000005s :     33: predicate.inline_without_move
 0.51% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.36% :     0.000002s :     12: predicate.less_batch_normalization
 1.53% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.76% :     0.000003s :     21: predicate.load_eliminater
 0.72% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.58% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.79% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.30% :     0.000002s :     18: predicate.merge_addn
 1.29% :     0.000002s :     18: predicate.minmaximum_grad
 0.77% :     0.000001s :      1: predicate.mutable_eliminate
 0.17% :     0.000000s :      1: predicate.opt_reshape
 2.22% :     0.000004s :     21: predicate.partial_eliminate
 1.38% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.93% :     0.000003s :     18: predicate.reduce_eliminate
 1.51% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.47% :     0.000003s :     22: predicate.replace_applicator
 1.72% :     0.000003s :     33: predicate.replace_old_param
 0.10% :     0.000000s :      1: predicate.reset_defer_inline
 1.71% :     0.000003s :     18: predicate.reshape_eliminate
 1.54% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.41% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.54% :     0.000003s :     18: predicate.same_eliminate
 0.63% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.35% :     0.000001s :      2: predicate.special_op_eliminate
 1.19% :     0.000002s :     12: predicate.specialize_transform
 1.72% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.55% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.11% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.61% :     0.000005s :     26: predicate.switch_defer_inline
 2.32% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.56% :     0.000011s :     71: predicate.switch_simplify
 1.42% :     0.000002s :     18: predicate.tile_eliminate
 1.80% :     0.000003s :     18: predicate.transpose_eliminate
 1.62% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.53% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.93% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.84% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.61% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.95% :     0.000003s :     18: predicate.value_based_eliminate
 0.11% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.26% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001184    22
73.20% :     0.000867s :     13: func_graph_cloner_run.FuncGraphClonerGraph
26.80% :     0.000317s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024052    76
 0.15% :     0.000035s :      1: add_recomputation
 0.38% :     0.000091s :      1: auto_monad
 0.06% :     0.000014s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.86% :     0.000208s :      1: bootstrap
 0.10% :     0.000024s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.21% :     0.000050s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
35.31% :     0.008493s :      1: jit_opt_a
 0.49% :     0.000118s :      1: jit_opt_after_cconv
 0.17% :     0.000042s :      1: jit_opt_b
 1.79% :     0.000429s :      1: loop_unroll
 2.46% :     0.000591s :      1: mutable_eliminate
 5.17% :     0.001244s :     26: opt.transform.jit_opt_a
 0.12% :     0.000029s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000015s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000022s :      4: opt.transform.symbol_engine_opt
 1.89% :     0.000454s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000059s :      1: py_interpret_to_execute
 0.05% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000015s :      1: remove_dup_value
 5.78% :     0.001390s :      1: renormalize.infer
 1.86% :     0.000448s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000035s :      1: rewriter_after_opt_a
 0.37% :     0.000090s :      1: rewriter_before_opt_a
 0.27% :     0.000065s :      1: symbol_engine_optimizer
 0.10% :     0.000024s :      1: task_emit
41.11% :     0.009889s :      1: type_inference
 0.18% :     0.000043s :      1: validate

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.936.808 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e6ee0> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.937.341 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.938.944 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c3680d0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.939.495 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0200732, [30]
[bootstrap]: 0.00018568
[type_inference]: 0.00915924
[event_method]: 4.475e-05
[auto_monad]: 9.434e-05
[graph_reusing]: 7.11001e-06
[pre_auto_parallel]: 2.96001e-06
[py_interpret_to_execute]: 2.628e-05
[rewriter_before_opt_a]: 9.039e-05
[expand_dump_flag]: 3.42002e-06
[jit_opt_a]: 0.00841452, [2]
    [Cycle 1]: 0.00482184, [27]
        [switch_simplify]: 6.131e-05
        [loop_unroll]: 3.898e-05
        [a_1]: 0.00088819
        [with_stream_mark]: 2.368e-05
        [recompute_prepare]: 1.862e-05
        [updatestate_depend_eliminate]: 7.82998e-06
        [updatestate_assign_eliminate]: 7.16999e-06
        [updatestate_loads_eliminate]: 6.59001e-06
        [parameter_eliminate]: 2.54001e-06
        [specialize_transform]: 1.281e-05
        [updatestate_useless_node_eliminater]: 1.202e-05
        [accelerated_algorithm]: 1.232e-05
        [meta_shard_fg_expand]: 3.53999e-06
        [get_grad_eliminate_]: 1.191e-05
        [merge_forward]: 8.99e-06
        [cell_reuse_recompute_pass]: 9.90025e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.73e-05
        [j_node_and_user_rematch]: 2.272e-05
        [meta_fg_expand]: 0.00179494
        [replace_old_param]: 5.316e-05
        [inline_without_move]: 5.123e-05
        [renormalize]: 0.00147187
        [add_forward_monad_depend]: 5.10999e-06
        [auto_monad_grad]: 2.27999e-06
        [auto_monad_eliminator]: 1.139e-05
        [cse]: 2.694e-05
        [replace_applicator]: 1.025e-05
    [Cycle 2]: 0.00026922, [27]
        [switch_simplify]: 4.61002e-06
        [loop_unroll]: 3.9e-06
        [a_1]: 4.119e-05
        [with_stream_mark]: 8.70001e-06
        [recompute_prepare]: 3.7e-06
        [updatestate_depend_eliminate]: 2.32001e-06
        [updatestate_assign_eliminate]: 1.62001e-06
        [updatestate_loads_eliminate]: 1.49998e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 3.53e-06
        [updatestate_useless_node_eliminater]: 3.39001e-06
        [accelerated_algorithm]: 3.63e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 3.38e-06
        [merge_forward]: 2.16e-06
        [cell_reuse_recompute_pass]: 1.34998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.066e-05
        [j_node_and_user_rematch]: 5.77999e-06
        [meta_fg_expand]: 2.598e-05
        [replace_old_param]: 3.78001e-06
        [inline_without_move]: 3.08998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.55001e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 4.08999e-06
        [cse]: 9.89999e-06
        [replace_applicator]: 3.8e-06
[py_interpret_to_execute_after_opt_a]: 9.39e-06
[rewriter_after_opt_a]: 3.26e-05
[convert_after_rewriter]: 6.43e-06
[order_py_execute_after_rewriter]: 4.70001e-06
[mutable_eliminate]: 0.0005053
[jit_opt_b]: 4.073e-05, [1]
    [Cycle 1]: 3.387e-05, [2]
        [frontend_op_eliminate]: 1.091e-05
        [inline_after_opt_a]: 1.186e-05
[cconv]: 1.995e-05
[loop_unroll]: 0.0004586
[jit_opt_after_cconv]: 0.00013198, [1]
    [Cycle 1]: 0.00012456, [11]
        [c_1]: 1.249e-05
        [parameter_eliminate]: 3.28998e-06
        [updatestate_depend_eliminate]: 5.99999e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [cse]: 1.962e-05
        [call_graph_tuple_transform]: 1.562e-05
        [tuple_list_get_item_eliminator]: 4.94e-06
        [none_parameter_eliminate]: 1.68002e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 4.37e-06
[remove_dup_value]: 1.422e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 5.17999e-06
[add_recomputation]: 3.42e-05
[cse_after_recomputation]: 2.146e-05, [1]
    [Cycle 1]: 1.545e-05, [1]
        [cse]: 8.69998e-06
[auto_monad_reorder]: 1.312e-05
[get_jit_bprop_graph]: 1.29e-06
[rewriter_after_jit_bprop_graph]: 5.61998e-06
[opt_after_jit_grad]: 0.00048782
[symbol_engine_optimizer]: 6.439e-05, [1]
    [Cycle 1]: 5.832e-05, [6]
        [build]: 3.61999e-06
        [elim_shapecalc]: 6.04001e-06
        [elim_not_effective]: 9.67999e-06
        [opt_reshape]: 4.25e-06
        [fold_const_symbol]: 5.99e-06
        [renormalize]: 5.69999e-07
[validate]: 2.857e-05
Sums
    bootstrap                            :     0.000186s :  1.15%
    type_inference                       :     0.009159s : 56.79%
    event_method                         :     0.000045s :  0.28%
    auto_monad                           :     0.000094s :  0.58%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.16%
    rewriter_before_opt_a                :     0.000090s :  0.56%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.41%
    jit_opt_a.loop_unroll                :     0.000043s :  0.27%
    jit_opt_a.a_1                        :     0.000929s :  5.76%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000016s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.09%
    jit_opt_a.merge_forward              :     0.000011s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001821s : 11.29%
    jit_opt_a.replace_old_param          :     0.000057s :  0.35%
    jit_opt_a.inline_without_move        :     0.000054s :  0.34%
    jit_opt_a.renormalize                :     0.001472s :  9.13%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.10%
    jit_opt_a.cse                        :     0.000037s :  0.23%
    jit_opt_a.replace_applicator         :     0.000014s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000505s :  3.13%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.07%
    cconv                                :     0.000020s :  0.12%
    loop_unroll                          :     0.000459s :  2.84%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000034s :  0.21%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.08%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000488s :  3.02%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000029s :  0.18%


Time group info:
------[substitution.]   0.000239    50
 0.60% :     0.000001s :      1: substitution.elim_not_effective
 0.37% :     0.000001s :      1: substitution.fold_const_symbol
 1.67% :     0.000004s :      1: substitution.graph_param_transform
65.23% :     0.000156s :      6: substitution.inline
 6.68% :     0.000016s :      2: substitution.inline_without_move
 2.71% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.47% :     0.000004s :      2: substitution.minmaximum_grad
 1.71% :     0.000004s :      3: substitution.partial_eliminate
 2.94% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.33% :     0.000006s :      2: substitution.replace_applicator
 1.72% :     0.000004s :      5: substitution.replace_old_param
 1.06% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.89% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.91% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 6.70% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009095     2
88.48% :     0.008048s :      1: type_inference.infer
11.52% :     0.001047s :      1: type_inference.specialize
------[replace.]   0.000087     8
84.65% :     0.000074s :      6: replace.inline
15.35% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000163     8
93.14% :     0.000152s :      6: match.inline
 6.86% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000174  1109
 1.44% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.71% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.29% :     0.000002s :     18: predicate.addn_check_dump
 1.63% :     0.000003s :     18: predicate.addn_zero_filter
 2.31% :     0.000004s :     18: predicate.arithmetic_simplify
 1.41% :     0.000002s :     18: predicate.cast_eliminate
 0.18% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.26% :     0.000002s :     18: predicate.compare_switch_simplify
 1.44% :     0.000003s :     18: predicate.depend_value_elim
 1.25% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.55% :     0.000003s :     18: predicate.dict_set_item_eliminator
 0.52% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.21% :     0.000000s :      1: predicate.elim_not_effective
 0.26% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.34% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.25% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.32% :     0.000002s :     18: predicate.environ_get_eliminate
 1.36% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.44% :     0.000003s :     12: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 4.46% :     0.000008s :     28: predicate.inline
 3.13% :     0.000005s :     33: predicate.inline_without_move
 0.49% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :     12: predicate.less_batch_normalization
 1.57% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.64% :     0.000003s :     21: predicate.load_eliminater
 1.09% :     0.000002s :      1: predicate.loop_unroll_after_grad
 3.53% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.99% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.29% :     0.000002s :     18: predicate.merge_addn
 1.30% :     0.000002s :     18: predicate.minmaximum_grad
 0.70% :     0.000001s :      1: predicate.mutable_eliminate
 0.21% :     0.000000s :      1: predicate.opt_reshape
 2.04% :     0.000004s :     21: predicate.partial_eliminate
 1.40% :     0.000002s :     18: predicate.print_const_string_wrapper
 2.04% :     0.000004s :     18: predicate.reduce_eliminate
 1.59% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.50% :     0.000003s :     22: predicate.replace_applicator
 1.69% :     0.000003s :     33: predicate.replace_old_param
 0.08% :     0.000000s :      1: predicate.reset_defer_inline
 1.59% :     0.000003s :     18: predicate.reshape_eliminate
 1.61% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.45% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     18: predicate.same_eliminate
 0.64% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.30% :     0.000001s :      2: predicate.special_op_eliminate
 1.19% :     0.000002s :     12: predicate.specialize_transform
 1.61% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.15% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.58% :     0.000004s :     26: predicate.switch_defer_inline
 2.38% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.45% :     0.000011s :     71: predicate.switch_simplify
 1.42% :     0.000002s :     18: predicate.tile_eliminate
 1.61% :     0.000003s :     18: predicate.transpose_eliminate
 1.74% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.53% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.08% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.68% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000003s :     18: predicate.value_based_eliminate
 0.10% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.34% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001191    22
77.17% :     0.000919s :     13: func_graph_cloner_run.FuncGraphClonerGraph
22.83% :     0.000272s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022857    72
 0.16% :     0.000037s :      1: add_recomputation
 0.43% :     0.000098s :      1: auto_monad
 0.07% :     0.000016s :      1: auto_monad_reorder
 0.86% :     0.000197s :      1: bootstrap
 0.10% :     0.000022s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.21% :     0.000048s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000003s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
36.83% :     0.008417s :      1: jit_opt_a
 0.59% :     0.000135s :      1: jit_opt_after_cconv
 0.19% :     0.000043s :      1: jit_opt_b
 2.05% :     0.000468s :      1: loop_unroll
 2.24% :     0.000512s :      1: mutable_eliminate
 5.62% :     0.001284s :     26: opt.transform.jit_opt_a
 0.15% :     0.000034s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000016s :      4: opt.transform.jit_opt_b
 0.06% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000010s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000022s :      4: opt.transform.symbol_engine_opt
 2.18% :     0.000498s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000029s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 4.97% :     0.001137s :      1: renormalize.infer
 1.43% :     0.000327s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000036s :      1: rewriter_after_opt_a
 0.41% :     0.000093s :      1: rewriter_before_opt_a
 0.29% :     0.000067s :      1: symbol_engine_optimizer
40.12% :     0.009169s :      1: type_inference


TotalTime = 0.0142017, [30]
[bootstrap]: 0.00031616
[type_inference]: 0.00505715
[event_method]: 1.851e-05
[auto_monad]: 6.953e-05
[graph_reusing]: 6.53e-06
[pre_auto_parallel]: 3.26001e-06
[py_interpret_to_execute]: 3.469e-05
[rewriter_before_opt_a]: 7.318e-05
[expand_dump_flag]: 3.30003e-06
[jit_opt_a]: 0.0062518, [2]
    [Cycle 1]: 0.00189514, [27]
        [switch_simplify]: 6.294e-05
        [loop_unroll]: 2.825e-05
        [a_1]: 0.00060632
        [with_stream_mark]: 2.709e-05
        [recompute_prepare]: 8.60001e-06
        [updatestate_depend_eliminate]: 4.63001e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 2.62001e-06
        [parameter_eliminate]: 2.30002e-06
        [specialize_transform]: 6.16e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.60001e-06
        [meta_shard_fg_expand]: 2.68e-06
        [get_grad_eliminate_]: 4.99e-06
        [merge_forward]: 3.98001e-06
        [cell_reuse_recompute_pass]: 1.30999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.612e-05
        [j_node_and_user_rematch]: 9.46e-06
        [meta_fg_expand]: 2.65002e-06
        [replace_old_param]: 9.25999e-06
        [inline_without_move]: 5.69e-06
        [renormalize]: 0.00078863
        [add_forward_monad_depend]: 6.73e-06
        [auto_monad_grad]: 2.54999e-06
        [auto_monad_eliminator]: 1.706e-05
        [cse]: 3.483e-05
        [replace_applicator]: 1.517e-05
    [Cycle 2]: 0.00033258, [27]
        [switch_simplify]: 5.47999e-06
        [loop_unroll]: 5.09e-06
        [a_1]: 9.316e-05
        [with_stream_mark]: 1.077e-05
        [recompute_prepare]: 5.74e-06
        [updatestate_depend_eliminate]: 2.78e-06
        [updatestate_assign_eliminate]: 1.89e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 9.30013e-07
        [specialize_transform]: 5.02e-06
        [updatestate_useless_node_eliminater]: 4.80999e-06
        [accelerated_algorithm]: 5.02999e-06
        [meta_shard_fg_expand]: 1.66e-06
        [get_grad_eliminate_]: 4.43001e-06
        [merge_forward]: 2.64001e-06
        [cell_reuse_recompute_pass]: 2.21e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.506e-05
        [j_node_and_user_rematch]: 7.87e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 8.03999e-06
        [inline_without_move]: 9.39998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.52001e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 5.87999e-06
        [cse]: 1.139e-05
        [replace_applicator]: 5.27001e-06
[py_interpret_to_execute_after_opt_a]: 1.336e-05
[rewriter_after_opt_a]: 4.058e-05
[convert_after_rewriter]: 6.86001e-06
[order_py_execute_after_rewriter]: 5.46e-06
[mutable_eliminate]: 0.00077389
[jit_opt_b]: 5.24e-05, [1]
    [Cycle 1]: 4.41e-05, [2]
        [frontend_op_eliminate]: 1.716e-05
        [inline_after_opt_a]: 1.551e-05
[cconv]: 3.189e-05
[loop_unroll]: 0.00043451
[jit_opt_after_cconv]: 0.00014608, [1]
    [Cycle 1]: 0.00013941, [11]
        [c_1]: 2.046e-05
        [parameter_eliminate]: 3.56999e-06
        [updatestate_depend_eliminate]: 6.90002e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [cse]: 2.093e-05
        [call_graph_tuple_transform]: 2.014e-05
        [tuple_list_get_item_eliminator]: 5.40999e-06
        [none_parameter_eliminate]: 1.84e-06
        [renormalize]: 6.89994e-07
        [switch_simplify]: 5.17999e-06
[remove_dup_value]: 1.554e-05
[partial_unused_args_eliminate]: 2.31e-06
[environ_conv]: 5.62001e-06
[add_recomputation]: 5.451e-05
[cse_after_recomputation]: 2.049e-05, [1]
    [Cycle 1]: 1.508e-05, [1]
        [cse]: 9.09e-06
[auto_monad_reorder]: 1.566e-05
[get_jit_bprop_graph]: 2.22999e-06
[rewriter_after_jit_bprop_graph]: 6.35002e-06
[opt_after_jit_grad]: 0.00044479
[symbol_engine_optimizer]: 7.473e-05, [1]
    [Cycle 1]: 6.786e-05, [6]
        [build]: 4.85999e-06
        [elim_shapecalc]: 7.12002e-06
        [elim_not_effective]: 1.325e-05
        [opt_reshape]: 5.40999e-06
        [fold_const_symbol]: 8.47998e-06
        [renormalize]: 7.7e-07
[validate]: 3.875e-05
Sums
    bootstrap                            :     0.000316s :  3.31%
    type_inference                       :     0.005057s : 52.95%
    event_method                         :     0.000019s :  0.19%
    auto_monad                           :     0.000070s :  0.73%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.36%
    rewriter_before_opt_a                :     0.000073s :  0.77%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000068s :  0.72%
    jit_opt_a.loop_unroll                :     0.000033s :  0.35%
    jit_opt_a.a_1                        :     0.000699s :  7.32%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.40%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.18%
    jit_opt_a.inline_without_move        :     0.000015s :  0.16%
    jit_opt_a.renormalize                :     0.000789s :  8.26%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.24%
    jit_opt_a.cse                        :     0.000046s :  0.48%
    jit_opt_a.replace_applicator         :     0.000020s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.14%
    rewriter_after_opt_a                 :     0.000041s :  0.42%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000774s :  8.10%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.16%
    cconv                                :     0.000032s :  0.33%
    loop_unroll                          :     0.000435s :  4.55%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000016s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000055s :  0.57%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000445s :  4.66%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.41%


Time group info:
------[substitution.]   0.000263    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 2.18% :     0.000006s :      3: substitution.graph_param_transform
85.84% :     0.000226s :      4: substitution.inline
 1.46% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.96% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.14% :     0.000006s :      2: substitution.replace_old_param
 5.06% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004970     2
81.78% :     0.004064s :      1: type_inference.infer
18.22% :     0.000906s :      1: type_inference.specialize
------[replace.]   0.000075     6
67.50% :     0.000050s :      4: replace.inline
32.50% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000236     6
94.78% :     0.000223s :      4: match.inline
 5.22% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000129   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.23% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 2.48% :     0.000003s :     10: predicate.addn_zero_filter
 2.13% :     0.000003s :     10: predicate.arithmetic_simplify
 1.06% :     0.000001s :     10: predicate.cast_eliminate
 0.31% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.63% :     0.000002s :     10: predicate.depend_value_elim
 0.94% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.97% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.69% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.02% :     0.000001s :     10: predicate.environ_get_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.41% :     0.000001s :      3: predicate.graph_param_transform
 5.94% :     0.000008s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.44% :     0.000002s :      6: predicate.less_batch_normalization
 1.28% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.52% :     0.000002s :     15: predicate.load_eliminater
 1.39% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.61% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.78% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 0.92% :     0.000001s :     10: predicate.minmaximum_grad
 2.71% :     0.000003s :      3: predicate.mutable_eliminate
 0.38% :     0.000000s :      3: predicate.opt_reshape
 2.01% :     0.000003s :     15: predicate.partial_eliminate
 1.51% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.97% :     0.000003s :     10: predicate.reduce_eliminate
 1.27% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.50% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.03% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.31% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.42% :     0.000002s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.55% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.28% :     0.000003s :     16: predicate.switch_defer_inline
 1.90% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.55% :     0.000010s :     53: predicate.switch_simplify
 1.28% :     0.000002s :     10: predicate.tile_eliminate
 1.01% :     0.000001s :     10: predicate.transpose_eliminate
 1.17% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.10% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.23% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.23% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.47% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.44% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.98% :     0.000003s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000657    12
50.60% :     0.000332s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.62% :     0.000057s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.78% :     0.000268s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015968    72
 0.36% :     0.000057s :      1: add_recomputation
 0.46% :     0.000073s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 2.10% :     0.000335s :      1: bootstrap
 0.22% :     0.000034s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
39.17% :     0.006255s :      1: jit_opt_a
 0.93% :     0.000149s :      1: jit_opt_after_cconv
 0.34% :     0.000055s :      1: jit_opt_b
 2.77% :     0.000442s :      1: loop_unroll
 4.91% :     0.000784s :      1: mutable_eliminate
 5.81% :     0.000927s :     26: opt.transform.jit_opt_a
 0.30% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.19% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.83% :     0.000452s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000037s :      1: py_interpret_to_execute
 0.10% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.68% :     0.000428s :      1: renormalize.infer
 2.19% :     0.000350s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000044s :      1: rewriter_after_opt_a
 0.48% :     0.000076s :      1: rewriter_before_opt_a
 0.48% :     0.000077s :      1: symbol_engine_optimizer
31.80% :     0.005078s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.955.307 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d6ee0> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.955.840 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0207727, [33]
[bootstrap]: 0.00020077
[type_inference]: 0.0100045
[event_method]: 4.79e-05
[auto_monad]: 7.978e-05
[graph_reusing]: 6.02999e-06
[pre_auto_parallel]: 2.22999e-06
[py_interpret_to_execute]: 2.557e-05
[rewriter_before_opt_a]: 8.973e-05
[expand_dump_flag]: 2.61999e-06
[jit_opt_a]: 0.00819656, [2]
    [Cycle 1]: 0.0044627, [27]
        [switch_simplify]: 5.863e-05
        [loop_unroll]: 3.78e-05
        [a_1]: 0.00083013
        [with_stream_mark]: 2.142e-05
        [recompute_prepare]: 1.751e-05
        [updatestate_depend_eliminate]: 7.83001e-06
        [updatestate_assign_eliminate]: 5.65001e-06
        [updatestate_loads_eliminate]: 1.083e-05
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 1.32e-05
        [updatestate_useless_node_eliminater]: 1.165e-05
        [accelerated_algorithm]: 1.243e-05
        [meta_shard_fg_expand]: 2.77002e-06
        [get_grad_eliminate_]: 1.235e-05
        [merge_forward]: 6.91999e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.376e-05
        [j_node_and_user_rematch]: 2.154e-05
        [meta_fg_expand]: 0.00135674
        [replace_old_param]: 5.015e-05
        [inline_without_move]: 4.974e-05
        [renormalize]: 0.00166754
        [add_forward_monad_depend]: 4.55001e-06
        [auto_monad_grad]: 1.64998e-06
        [auto_monad_eliminator]: 8.75001e-06
        [cse]: 1.788e-05
        [replace_applicator]: 1.041e-05
    [Cycle 2]: 0.00028921, [27]
        [switch_simplify]: 4.60001e-06
        [loop_unroll]: 3.85998e-06
        [a_1]: 4.034e-05
        [with_stream_mark]: 9.32999e-06
        [recompute_prepare]: 3.81001e-06
        [updatestate_depend_eliminate]: 2.11e-06
        [updatestate_assign_eliminate]: 1.67999e-06
        [updatestate_loads_eliminate]: 1.44998e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 3.58e-06
        [updatestate_useless_node_eliminater]: 3.43999e-06
        [accelerated_algorithm]: 6.48e-06
        [meta_shard_fg_expand]: 1.09998e-06
        [get_grad_eliminate_]: 3.42002e-06
        [merge_forward]: 1.99e-06
        [cell_reuse_recompute_pass]: 2.01e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.044e-05
        [j_node_and_user_rematch]: 5.57999e-06
        [meta_fg_expand]: 3.342e-05
        [replace_old_param]: 4.26001e-06
        [inline_without_move]: 3.66999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.76003e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 5.34998e-06
        [cse]: 1.169e-05
        [replace_applicator]: 3.98001e-06
[py_interpret_to_execute_after_opt_a]: 1.096e-05
[rewriter_after_opt_a]: 3.594e-05
[convert_after_rewriter]: 6.39999e-06
[order_py_execute_after_rewriter]: 4.48999e-06
[mutable_eliminate]: 0.0005621
[jit_opt_b]: 4.212e-05, [1]
    [Cycle 1]: 3.47e-05, [2]
        [frontend_op_eliminate]: 1.167e-05
        [inline_after_opt_a]: 1.141e-05
[cconv]: 2.118e-05
[loop_unroll]: 0.00043283
[jit_opt_after_cconv]: 0.00011995, [1]
    [Cycle 1]: 0.00011322, [11]
        [c_1]: 1.235e-05
        [parameter_eliminate]: 2.41e-06
        [updatestate_depend_eliminate]: 4.46002e-06
        [updatestate_assign_eliminate]: 1.87001e-06
        [updatestate_loads_eliminate]: 1.76003e-06
        [cse]: 1.663e-05
        [call_graph_tuple_transform]: 1.399e-05
        [tuple_list_get_item_eliminator]: 3.97002e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 3.30008e-07
        [switch_simplify]: 4.08999e-06
[remove_dup_value]: 1.432e-05
[partial_unused_args_eliminate]: 2.04e-06
[environ_conv]: 4.80999e-06
[add_recomputation]: 3.446e-05
[cse_after_recomputation]: 2.07e-05, [1]
    [Cycle 1]: 1.525e-05, [1]
        [cse]: 9.04998e-06
[auto_monad_reorder]: 1.467e-05
[get_jit_bprop_graph]: 1.99999e-06
[rewriter_after_jit_bprop_graph]: 5.99e-06
[opt_after_jit_grad]: 0.00047787
[symbol_engine_optimizer]: 6.469e-05, [1]
    [Cycle 1]: 5.848e-05, [6]
        [build]: 3.08998e-06
        [elim_shapecalc]: 5.61e-06
        [elim_not_effective]: 1.011e-05
        [opt_reshape]: 4.64002e-06
        [fold_const_symbol]: 6.01003e-06
        [renormalize]: 8.79983e-07
[validate]: 2.963e-05
[backend_pass]: 9.60019e-07
[task_emit]: 2.423e-05
[execute]: 1.55999e-06
Sums
    bootstrap                            :     0.000201s :  1.20%
    type_inference                       :     0.010004s : 59.88%
    event_method                         :     0.000048s :  0.29%
    auto_monad                           :     0.000080s :  0.48%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.15%
    rewriter_before_opt_a                :     0.000090s :  0.54%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.38%
    jit_opt_a.loop_unroll                :     0.000042s :  0.25%
    jit_opt_a.a_1                        :     0.000870s :  5.21%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.07%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000019s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.09%
    jit_opt_a.merge_forward              :     0.000009s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001390s :  8.32%
    jit_opt_a.replace_old_param          :     0.000054s :  0.33%
    jit_opt_a.inline_without_move        :     0.000053s :  0.32%
    jit_opt_a.renormalize                :     0.001668s :  9.98%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000014s :  0.08%
    jit_opt_a.cse                        :     0.000030s :  0.18%
    jit_opt_a.replace_applicator         :     0.000014s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000562s :  3.36%
    jit_opt_b.frontend_op_eliminate      :     0.000012s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000021s :  0.13%
    loop_unroll                          :     0.000433s :  2.59%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000034s :  0.21%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000015s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000478s :  2.86%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000030s :  0.18%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.000024s :  0.15%
    execute                              :     0.000002s :  0.01%


Time group info:
------[substitution.]   0.000233    50
 0.63% :     0.000001s :      1: substitution.elim_not_effective
 0.50% :     0.000001s :      1: substitution.fold_const_symbol
 1.80% :     0.000004s :      1: substitution.graph_param_transform
68.13% :     0.000159s :      6: substitution.inline
 6.56% :     0.000015s :      2: substitution.inline_without_move
 2.34% :     0.000005s :      9: substitution.j_node_and_user_rematch
 1.22% :     0.000003s :      2: substitution.minmaximum_grad
 1.45% :     0.000003s :      3: substitution.partial_eliminate
 2.78% :     0.000006s :      9: substitution.remove_not_recompute_node
 2.03% :     0.000005s :      2: substitution.replace_applicator
 1.68% :     0.000004s :      5: substitution.replace_old_param
 0.96% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.41% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.74% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 5.77% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009938     2
88.35% :     0.008780s :      1: type_inference.infer
11.65% :     0.001158s :      1: type_inference.specialize
------[replace.]   0.000059     8
78.21% :     0.000046s :      6: replace.inline
21.79% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000164     8
94.94% :     0.000155s :      6: match.inline
 5.06% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000167  1109
 1.61% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.56% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.29% :     0.000002s :     18: predicate.addn_check_dump
 1.56% :     0.000003s :     18: predicate.addn_zero_filter
 2.17% :     0.000004s :     18: predicate.arithmetic_simplify
 1.54% :     0.000003s :     18: predicate.cast_eliminate
 0.23% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.40% :     0.000002s :     18: predicate.compare_switch_simplify
 1.62% :     0.000003s :     18: predicate.depend_value_elim
 1.30% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.47% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.09% :     0.000000s :      1: predicate.elim_not_effective
 0.24% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.41% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.26% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.25% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.49% :     0.000002s :     18: predicate.environ_get_eliminate
 1.26% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.15% :     0.000002s :     12: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 4.45% :     0.000007s :     28: predicate.inline
 3.34% :     0.000006s :     33: predicate.inline_without_move
 0.53% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :     12: predicate.less_batch_normalization
 1.62% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.70% :     0.000003s :     21: predicate.load_eliminater
 0.72% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.68% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.83% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.27% :     0.000002s :     18: predicate.merge_addn
 1.32% :     0.000002s :     18: predicate.minmaximum_grad
 0.75% :     0.000001s :      1: predicate.mutable_eliminate
 0.17% :     0.000000s :      1: predicate.opt_reshape
 2.02% :     0.000003s :     21: predicate.partial_eliminate
 1.63% :     0.000003s :     18: predicate.print_const_string_wrapper
 1.83% :     0.000003s :     18: predicate.reduce_eliminate
 1.56% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.47% :     0.000002s :     22: predicate.replace_applicator
 1.58% :     0.000003s :     33: predicate.replace_old_param
 0.13% :     0.000000s :      1: predicate.reset_defer_inline
 1.59% :     0.000003s :     18: predicate.reshape_eliminate
 1.58% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.48% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.50% :     0.000003s :     18: predicate.same_eliminate
 0.67% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.40% :     0.000001s :      2: predicate.special_op_eliminate
 1.27% :     0.000002s :     12: predicate.specialize_transform
 1.60% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.58% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.14% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.58% :     0.000004s :     26: predicate.switch_defer_inline
 2.36% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.63% :     0.000011s :     71: predicate.switch_simplify
 1.46% :     0.000002s :     18: predicate.tile_eliminate
 1.33% :     0.000002s :     18: predicate.transpose_eliminate
 1.62% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.49% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.91% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.89% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.60% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.90% :     0.000003s :     18: predicate.value_based_eliminate
 0.10% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.29% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001205    22
69.76% :     0.000840s :     13: func_graph_cloner_run.FuncGraphClonerGraph
30.24% :     0.000364s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023721    76
 0.16% :     0.000037s :      1: add_recomputation
 0.36% :     0.000084s :      1: auto_monad
 0.07% :     0.000017s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 0.90% :     0.000213s :      1: bootstrap
 0.10% :     0.000024s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.22% :     0.000052s :      1: event_method
 0.02% :     0.000004s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
34.57% :     0.008200s :      1: jit_opt_a
 0.52% :     0.000123s :      1: jit_opt_after_cconv
 0.19% :     0.000045s :      1: jit_opt_b
 1.86% :     0.000441s :      1: loop_unroll
 2.41% :     0.000571s :      1: mutable_eliminate
 5.15% :     0.001221s :     26: opt.transform.jit_opt_a
 0.13% :     0.000031s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000017s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000023s :      4: opt.transform.symbol_engine_opt
 2.05% :     0.000486s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000028s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 5.36% :     0.001272s :      1: renormalize.infer
 1.64% :     0.000388s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000039s :      1: rewriter_after_opt_a
 0.39% :     0.000093s :      1: rewriter_before_opt_a
 0.28% :     0.000067s :      1: symbol_engine_optimizer
 0.12% :     0.000029s :      1: task_emit
42.23% :     0.010017s :      1: type_inference
 0.20% :     0.000047s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.968.179 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c368430> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.968.940 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.032621, [33]
[bootstrap]: 0.00018089
[type_inference]: 0.0093569
[event_method]: 4.538e-05
[auto_monad]: 8.708e-05
[graph_reusing]: 5.97999e-06
[pre_auto_parallel]: 2.27001e-06
[py_interpret_to_execute]: 2.142e-05
[rewriter_before_opt_a]: 8.586e-05
[expand_dump_flag]: 2.91999e-06
[jit_opt_a]: 0.0116703, [3]
    [Cycle 1]: 0.00591545, [27]
        [switch_simplify]: 6.367e-05
        [loop_unroll]: 3.784e-05
        [a_1]: 0.00082063
        [with_stream_mark]: 2.427e-05
        [recompute_prepare]: 1.781e-05
        [updatestate_depend_eliminate]: 7.65e-06
        [updatestate_assign_eliminate]: 6.34999e-06
        [updatestate_loads_eliminate]: 5.42999e-06
        [parameter_eliminate]: 2.32999e-06
        [specialize_transform]: 1.256e-05
        [updatestate_useless_node_eliminater]: 1.155e-05
        [accelerated_algorithm]: 1.259e-05
        [meta_shard_fg_expand]: 3.34001e-06
        [get_grad_eliminate_]: 1.195e-05
        [merge_forward]: 8.13999e-06
        [cell_reuse_recompute_pass]: 9.60019e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.55e-05
        [j_node_and_user_rematch]: 2.266e-05
        [meta_fg_expand]: 0.00146945
        [replace_old_param]: 5.257e-05
        [inline_without_move]: 5.018e-05
        [renormalize]: 0.00276252
        [add_forward_monad_depend]: 8.80999e-06
        [auto_monad_grad]: 5.77001e-06
        [auto_monad_eliminator]: 5.31e-05
        [cse]: 0.00015951
        [replace_applicator]: 6.462e-05
    [Cycle 2]: 0.00211492, [27]
        [switch_simplify]: 3.71e-05
        [loop_unroll]: 3.589e-05
        [a_1]: 0.00112627
        [with_stream_mark]: 1.138e-05
        [recompute_prepare]: 8.79e-06
        [updatestate_depend_eliminate]: 4.37e-06
        [updatestate_assign_eliminate]: 3.2e-06
        [updatestate_loads_eliminate]: 3.36999e-06
        [parameter_eliminate]: 1.60001e-06
        [specialize_transform]: 7.49002e-06
        [updatestate_useless_node_eliminater]: 7.06999e-06
        [accelerated_algorithm]: 7.5e-06
        [meta_shard_fg_expand]: 1.82999e-06
        [get_grad_eliminate_]: 6.63e-06
        [merge_forward]: 3.75e-06
        [cell_reuse_recompute_pass]: 1.61998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.507e-05
        [j_node_and_user_rematch]: 1.165e-05
        [meta_fg_expand]: 4.635e-05
        [replace_old_param]: 1.037e-05
        [inline_without_move]: 7.13e-06
        [renormalize]: 0.00055226
        [add_forward_monad_depend]: 4.20999e-06
        [auto_monad_grad]: 1.15999e-06
        [auto_monad_eliminator]: 1.366e-05
        [cse]: 2.372e-05
        [replace_applicator]: 1.437e-05
    [Cycle 3]: 0.00041259, [27]
        [switch_simplify]: 7.43e-06
        [loop_unroll]: 7.3e-06
        [a_1]: 0.00014609
        [with_stream_mark]: 1.007e-05
        [recompute_prepare]: 6.84001e-06
        [updatestate_depend_eliminate]: 3.82002e-06
        [updatestate_assign_eliminate]: 3.35e-06
        [updatestate_loads_eliminate]: 3.14001e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 6.96999e-06
        [updatestate_useless_node_eliminater]: 6.68998e-06
        [accelerated_algorithm]: 6.98e-06
        [meta_shard_fg_expand]: 1.44998e-06
        [get_grad_eliminate_]: 6.54001e-06
        [merge_forward]: 3.38e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.593e-05
        [j_node_and_user_rematch]: 1.132e-05
        [meta_fg_expand]: 2.29001e-06
        [replace_old_param]: 9.74999e-06
        [inline_without_move]: 6.53e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.26997e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 7.65e-06
        [cse]: 1.669e-05
        [replace_applicator]: 6.88e-06
[py_interpret_to_execute_after_opt_a]: 1.105e-05
[rewriter_after_opt_a]: 3.841e-05
[convert_after_rewriter]: 6.91999e-06
[order_py_execute_after_rewriter]: 5.86998e-06
[mutable_eliminate]: 0.00061109
[jit_opt_b]: 6.334e-05, [1]
    [Cycle 1]: 5.607e-05, [2]
        [frontend_op_eliminate]: 2.311e-05
        [inline_after_opt_a]: 2.078e-05
[cconv]: 1.611e-05
[loop_unroll]: 0.00042569
[jit_opt_after_cconv]: 0.00017436, [1]
    [Cycle 1]: 0.00016797, [11]
        [c_1]: 3.214e-05
        [parameter_eliminate]: 1.96998e-06
        [updatestate_depend_eliminate]: 6.81001e-06
        [updatestate_assign_eliminate]: 3.48e-06
        [updatestate_loads_eliminate]: 3.18e-06
        [cse]: 2.711e-05
        [call_graph_tuple_transform]: 2.27e-05
        [tuple_list_get_item_eliminator]: 7.97e-06
        [none_parameter_eliminate]: 8.50006e-07
        [renormalize]: 5.40022e-07
        [switch_simplify]: 7.5e-06
[remove_dup_value]: 1.256e-05
[partial_unused_args_eliminate]: 1.76e-06
[environ_conv]: 6.09001e-06
[add_recomputation]: 4.622e-05
[cse_after_recomputation]: 3.391e-05, [1]
    [Cycle 1]: 2.775e-05, [1]
        [cse]: 1.993e-05
[auto_monad_reorder]: 1.669e-05
[get_jit_bprop_graph]: 1.60999e-06
[rewriter_after_jit_bprop_graph]: 5.40999e-06
[opt_after_jit_grad]: 0.00055481
[symbol_engine_optimizer]: 8.941e-05, [1]
    [Cycle 1]: 8.287e-05, [6]
        [build]: 3.12002e-06
        [elim_shapecalc]: 1.175e-05
        [elim_not_effective]: 1.698e-05
        [opt_reshape]: 7.93001e-06
        [fold_const_symbol]: 1.352e-05
        [renormalize]: 5.19998e-07
[validate]: 3.768e-05
[backend_pass]: 7.50006e-07
[task_emit]: 0.00878896
[execute]: 7.78999e-06
Sums
    bootstrap                            :     0.000181s :  0.63%
    type_inference                       :     0.009357s : 32.71%
    event_method                         :     0.000045s :  0.16%
    auto_monad                           :     0.000087s :  0.30%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000021s :  0.07%
    rewriter_before_opt_a                :     0.000086s :  0.30%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000108s :  0.38%
    jit_opt_a.loop_unroll                :     0.000081s :  0.28%
    jit_opt_a.a_1                        :     0.002093s :  7.32%
    jit_opt_a.with_stream_mark           :     0.000046s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000033s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000027s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000027s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000025s :  0.09%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000056s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000046s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001518s :  5.31%
    jit_opt_a.replace_old_param          :     0.000073s :  0.25%
    jit_opt_a.inline_without_move        :     0.000064s :  0.22%
    jit_opt_a.renormalize                :     0.003315s : 11.59%
    jit_opt_a.add_forward_monad_depend   :     0.000014s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000074s :  0.26%
    jit_opt_a.cse                        :     0.000200s :  0.70%
    jit_opt_a.replace_applicator         :     0.000086s :  0.30%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.04%
    rewriter_after_opt_a                 :     0.000038s :  0.13%
    convert_after_rewriter               :     0.000007s :  0.02%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000611s :  2.14%
    jit_opt_b.frontend_op_eliminate      :     0.000023s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000021s :  0.07%
    cconv                                :     0.000016s :  0.06%
    loop_unroll                          :     0.000426s :  1.49%
    jit_opt_after_cconv.c_1              :     0.000032s :  0.11%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000027s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.03%
    remove_dup_value                     :     0.000013s :  0.04%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000046s :  0.16%
    cse_after_recomputation.cse          :     0.000020s :  0.07%
    auto_monad_reorder                   :     0.000017s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000555s :  1.94%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000038s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008789s : 30.72%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000501   118
 5.06% :     0.000025s :      1: substitution.arithmetic_simplify
 0.54% :     0.000003s :      4: substitution.elim_not_effective
 0.41% :     0.000002s :      4: substitution.fold_const_symbol
 1.07% :     0.000005s :      5: substitution.graph_param_transform
60.80% :     0.000305s :     12: substitution.inline
 3.24% :     0.000016s :      2: substitution.inline_without_move
 1.91% :     0.000010s :     16: substitution.j_node_and_user_rematch
 1.98% :     0.000010s :      7: substitution.minmaximum_grad
 1.22% :     0.000006s :      5: substitution.partial_eliminate
 2.32% :     0.000012s :     16: substitution.remove_not_recompute_node
 4.92% :     0.000025s :     10: substitution.replace_applicator
 1.45% :     0.000007s :      7: substitution.replace_old_param
 0.53% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.28% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.63% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 7.66% :     0.000038s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009297     2
88.50% :     0.008228s :      1: type_inference.infer
11.50% :     0.001069s :      1: type_inference.specialize
------[replace.]   0.000149    20
 3.67% :     0.000005s :      1: replace.arithmetic_simplify
61.55% :     0.000092s :     12: replace.inline
34.78% :     0.000052s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000342    20
 7.16% :     0.000024s :      1: match.arithmetic_simplify
87.13% :     0.000298s :     12: match.inline
 5.72% :     0.000020s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000404  2790
 1.63% :     0.000007s :     46: predicate.accumulaten_eliminater
 0.65% :     0.000003s :      5: predicate.ad_related_special_op_eliminate
 1.41% :     0.000006s :     46: predicate.addn_check_dump
 1.66% :     0.000007s :     46: predicate.addn_zero_filter
 4.76% :     0.000019s :     47: predicate.arithmetic_simplify
 1.64% :     0.000007s :     47: predicate.cast_eliminate
 0.23% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.36% :     0.000005s :     46: predicate.compare_switch_simplify
 1.51% :     0.000006s :     46: predicate.depend_value_elim
 1.40% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.52% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.50% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.31% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.17% :     0.000001s :      5: predicate.elim_not_effective
 0.36% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.55% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.43% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.42% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.46% :     0.000006s :     47: predicate.environ_get_eliminate
 1.41% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      5: predicate.fold_const_symbol
 0.85% :     0.000003s :     21: predicate.get_grad_eliminate
 0.13% :     0.000001s :      5: predicate.graph_param_transform
 4.04% :     0.000016s :     76: predicate.inline
 1.65% :     0.000007s :     42: predicate.inline_without_move
 0.36% :     0.000001s :     21: predicate.j_node_and_user_rematch
 0.92% :     0.000004s :     21: predicate.less_batch_normalization
 1.73% :     0.000007s :     54: predicate.list_to_tuple_eliminator_
 1.98% :     0.000008s :     59: predicate.load_eliminater
 0.49% :     0.000002s :      5: predicate.loop_unroll_after_grad
 2.95% :     0.000012s :     89: predicate.loop_unroll_before_grad
 1.82% :     0.000007s :     52: predicate.make_slice_get_slice_eliminator
 1.43% :     0.000006s :     46: predicate.merge_addn
 1.44% :     0.000006s :     47: predicate.minmaximum_grad
 0.59% :     0.000002s :      5: predicate.mutable_eliminate
 0.21% :     0.000001s :      5: predicate.opt_reshape
 2.29% :     0.000009s :     59: predicate.partial_eliminate
 1.51% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.80% :     0.000007s :     47: predicate.reduce_eliminate
 1.79% :     0.000007s :     54: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.51% :     0.000010s :    105: predicate.replace_applicator
 0.85% :     0.000003s :     42: predicate.replace_old_param
 0.12% :     0.000000s :      5: predicate.reset_defer_inline
 1.57% :     0.000006s :     47: predicate.reshape_eliminate
 1.54% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.36% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.42% :     0.000006s :     46: predicate.same_eliminate
 0.45% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.45% :     0.000002s :     10: predicate.special_op_eliminate
 0.82% :     0.000003s :     21: predicate.specialize_transform
 1.55% :     0.000006s :     46: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.20% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.54% :     0.000010s :     66: predicate.switch_defer_inline
 2.39% :     0.000010s :     66: predicate.switch_layer_defer_inline
 5.88% :     0.000024s :    160: predicate.switch_simplify
 1.45% :     0.000006s :     47: predicate.tile_eliminate
 1.59% :     0.000006s :     47: predicate.transpose_eliminate
 1.86% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.60% :     0.000006s :     47: predicate.tuple_list_get_item_depend_reorder
 2.96% :     0.000012s :     64: predicate.tuple_list_get_item_eliminator
 1.88% :     0.000008s :     47: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000007s :     54: predicate.tuple_to_list_eliminator_
 1.87% :     0.000008s :     59: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000011s :     80: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000007s :     46: predicate.value_based_eliminate
 0.18% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.32% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001468    29
57.63% :     0.000846s :     13: func_graph_cloner_run.FuncGraphClonerGraph
42.37% :     0.000622s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038796    91
 0.13% :     0.000050s :      1: add_recomputation
 0.23% :     0.000091s :      1: auto_monad
 0.05% :     0.000019s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.50% :     0.000196s :      1: bootstrap
 0.05% :     0.000019s :      1: cconv
 0.02% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000036s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.13% :     0.000049s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
30.09% :     0.011674s :      1: jit_opt_a
 0.46% :     0.000177s :      1: jit_opt_after_cconv
 0.17% :     0.000066s :      1: jit_opt_b
 1.12% :     0.000433s :      1: loop_unroll
 1.60% :     0.000620s :      1: mutable_eliminate
 6.97% :     0.002705s :     39: opt.transform.jit_opt_a
 0.17% :     0.000067s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000037s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000047s :      4: opt.transform.symbol_engine_opt
 1.45% :     0.000563s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.06% :     0.000024s :      1: py_interpret_to_execute
 0.03% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000016s :      1: remove_dup_value
 4.61% :     0.001790s :      2: renormalize.infer
 3.90% :     0.001512s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000041s :      1: rewriter_after_opt_a
 0.23% :     0.000089s :      1: rewriter_before_opt_a
 0.24% :     0.000092s :      1: symbol_engine_optimizer
22.69% :     0.008802s :      1: task_emit
24.14% :     0.009367s :      1: type_inference
 0.15% :     0.000060s :      1: validate


TotalTime = 0.0213342, [30]
[bootstrap]: 0.00018203
[type_inference]: 0.00955832
[event_method]: 4.505e-05
[auto_monad]: 9.822e-05
[graph_reusing]: 7.35e-06
[pre_auto_parallel]: 3.56999e-06
[py_interpret_to_execute]: 2.923e-05
[rewriter_before_opt_a]: 9.385e-05
[expand_dump_flag]: 3.46001e-06
[jit_opt_a]: 0.00916772, [2]
    [Cycle 1]: 0.00505945, [27]
        [switch_simplify]: 7.419e-05
        [loop_unroll]: 3.901e-05
        [a_1]: 0.00087735
        [with_stream_mark]: 3.329e-05
        [recompute_prepare]: 2.243e-05
        [updatestate_depend_eliminate]: 9.52999e-06
        [updatestate_assign_eliminate]: 6.63e-06
        [updatestate_loads_eliminate]: 6.46999e-06
        [parameter_eliminate]: 2.71999e-06
        [specialize_transform]: 1.414e-05
        [updatestate_useless_node_eliminater]: 1.25e-05
        [accelerated_algorithm]: 1.305e-05
        [meta_shard_fg_expand]: 4.22e-06
        [get_grad_eliminate_]: 1.252e-05
        [merge_forward]: 9.40001e-06
        [cell_reuse_recompute_pass]: 8.99978e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.575e-05
        [j_node_and_user_rematch]: 2.245e-05
        [meta_fg_expand]: 0.00168872
        [replace_old_param]: 5.889e-05
        [inline_without_move]: 5.343e-05
        [renormalize]: 0.00177766
        [add_forward_monad_depend]: 7.23999e-06
        [auto_monad_grad]: 2.51998e-06
        [auto_monad_eliminator]: 1.288e-05
        [cse]: 2.735e-05
        [replace_applicator]: 1.512e-05
    [Cycle 2]: 0.00029818, [27]
        [switch_simplify]: 4.58001e-06
        [loop_unroll]: 4.31002e-06
        [a_1]: 4.674e-05
        [with_stream_mark]: 1.026e-05
        [recompute_prepare]: 3.97e-06
        [updatestate_depend_eliminate]: 2.59999e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [parameter_eliminate]: 1.73002e-06
        [specialize_transform]: 3.83001e-06
        [updatestate_useless_node_eliminater]: 3.37002e-06
        [accelerated_algorithm]: 3.63999e-06
        [meta_shard_fg_expand]: 1.89999e-06
        [get_grad_eliminate_]: 4.03999e-06
        [merge_forward]: 3.51999e-06
        [cell_reuse_recompute_pass]: 2.07001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.301e-05
        [j_node_and_user_rematch]: 6.86001e-06
        [meta_fg_expand]: 2.958e-05
        [replace_old_param]: 3.93001e-06
        [inline_without_move]: 3.34001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.71002e-06
        [auto_monad_grad]: 9.09989e-07
        [auto_monad_eliminator]: 4.14997e-06
        [cse]: 1.134e-05
        [replace_applicator]: 3.90998e-06
[py_interpret_to_execute_after_opt_a]: 1.011e-05
[rewriter_after_opt_a]: 3.611e-05
[convert_after_rewriter]: 5.96e-06
[order_py_execute_after_rewriter]: 4.67e-06
[mutable_eliminate]: 0.00064743
[jit_opt_b]: 4.384e-05, [1]
    [Cycle 1]: 3.622e-05, [2]
        [frontend_op_eliminate]: 1.257e-05
        [inline_after_opt_a]: 1.144e-05
[cconv]: 2.428e-05
[loop_unroll]: 0.00042168
[jit_opt_after_cconv]: 0.00012462, [1]
    [Cycle 1]: 0.00011799, [11]
        [c_1]: 1.252e-05
        [parameter_eliminate]: 3.07002e-06
        [updatestate_depend_eliminate]: 5.07e-06
        [updatestate_assign_eliminate]: 1.95001e-06
        [updatestate_loads_eliminate]: 1.71e-06
        [cse]: 1.722e-05
        [call_graph_tuple_transform]: 1.484e-05
        [tuple_list_get_item_eliminator]: 4.07e-06
        [none_parameter_eliminate]: 1.41002e-06
        [renormalize]: 1.12999e-06
        [switch_simplify]: 3.93999e-06
[remove_dup_value]: 1.442e-05
[partial_unused_args_eliminate]: 2.37001e-06
[environ_conv]: 5.23002e-06
[add_recomputation]: 3.421e-05
[cse_after_recomputation]: 2.031e-05, [1]
    [Cycle 1]: 1.474e-05, [1]
        [cse]: 8.34998e-06
[auto_monad_reorder]: 1.249e-05
[get_jit_bprop_graph]: 1.81e-06
[rewriter_after_jit_bprop_graph]: 4.79e-06
[opt_after_jit_grad]: 0.00046076
[symbol_engine_optimizer]: 6.464e-05, [1]
    [Cycle 1]: 5.844e-05, [6]
        [build]: 3.60003e-06
        [elim_shapecalc]: 5.96e-06
        [elim_not_effective]: 1.022e-05
        [opt_reshape]: 4.16001e-06
        [fold_const_symbol]: 6.11e-06
        [renormalize]: 4.50003e-07
[validate]: 4.291e-05
Sums
    bootstrap                            :     0.000182s :  1.08%
    type_inference                       :     0.009558s : 56.59%
    event_method                         :     0.000045s :  0.27%
    auto_monad                           :     0.000098s :  0.58%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000004s :  0.02%
    py_interpret_to_execute              :     0.000029s :  0.17%
    rewriter_before_opt_a                :     0.000094s :  0.56%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000079s :  0.47%
    jit_opt_a.loop_unroll                :     0.000043s :  0.26%
    jit_opt_a.a_1                        :     0.000924s :  5.47%
    jit_opt_a.with_stream_mark           :     0.000044s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000026s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000012s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000018s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000017s :  0.10%
    jit_opt_a.merge_forward              :     0.000013s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001718s : 10.17%
    jit_opt_a.replace_old_param          :     0.000063s :  0.37%
    jit_opt_a.inline_without_move        :     0.000057s :  0.34%
    jit_opt_a.renormalize                :     0.001778s : 10.53%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.10%
    jit_opt_a.cse                        :     0.000039s :  0.23%
    jit_opt_a.replace_applicator         :     0.000019s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000036s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000647s :  3.83%
    jit_opt_b.frontend_op_eliminate      :     0.000013s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000024s :  0.14%
    loop_unroll                          :     0.000422s :  2.50%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000034s :  0.20%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000012s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000461s :  2.73%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000043s :  0.25%


Time group info:
------[substitution.]   0.000266    50
 0.56% :     0.000001s :      1: substitution.elim_not_effective
 0.37% :     0.000001s :      1: substitution.fold_const_symbol
 1.59% :     0.000004s :      1: substitution.graph_param_transform
65.91% :     0.000175s :      6: substitution.inline
 6.82% :     0.000018s :      2: substitution.inline_without_move
 2.53% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.31% :     0.000003s :      2: substitution.minmaximum_grad
 1.82% :     0.000005s :      3: substitution.partial_eliminate
 2.74% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.12% :     0.000006s :      2: substitution.replace_applicator
 1.79% :     0.000005s :      5: substitution.replace_old_param
 1.36% :     0.000004s :      1: substitution.set_cell_output_no_recompute
 2.90% :     0.000008s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.87% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 6.30% :     0.000017s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009487     2
88.47% :     0.008393s :      1: type_inference.infer
11.53% :     0.001094s :      1: type_inference.specialize
------[replace.]   0.000066     8
77.07% :     0.000051s :      6: replace.inline
22.93% :     0.000015s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000183     8
93.44% :     0.000171s :      6: match.inline
 6.56% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000181  1109
 1.66% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.58% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.26% :     0.000002s :     18: predicate.addn_check_dump
 1.53% :     0.000003s :     18: predicate.addn_zero_filter
 2.28% :     0.000004s :     18: predicate.arithmetic_simplify
 1.45% :     0.000003s :     18: predicate.cast_eliminate
 0.23% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.23% :     0.000002s :     18: predicate.compare_switch_simplify
 1.56% :     0.000003s :     18: predicate.depend_value_elim
 1.54% :     0.000003s :     18: predicate.dict_get_item_const_eliminator
 1.49% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.43% :     0.000003s :     18: predicate.dict_set_item_eliminator
 0.42% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.10% :     0.000000s :      1: predicate.elim_not_effective
 0.24% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.51% :     0.000003s :     18: predicate.environ_add_const_eliminate
 1.27% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.18% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.37% :     0.000002s :     18: predicate.environ_get_eliminate
 1.26% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.07% :     0.000000s :      1: predicate.fold_const_symbol
 1.22% :     0.000002s :     12: predicate.get_grad_eliminate
 0.18% :     0.000000s :      1: predicate.graph_param_transform
 4.62% :     0.000008s :     28: predicate.inline
 3.04% :     0.000006s :     33: predicate.inline_without_move
 0.48% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :     12: predicate.less_batch_normalization
 1.55% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.61% :     0.000003s :     21: predicate.load_eliminater
 0.67% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.41% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.65% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000002s :     18: predicate.merge_addn
 1.28% :     0.000002s :     18: predicate.minmaximum_grad
 0.98% :     0.000002s :      1: predicate.mutable_eliminate
 0.20% :     0.000000s :      1: predicate.opt_reshape
 2.00% :     0.000004s :     21: predicate.partial_eliminate
 1.51% :     0.000003s :     18: predicate.print_const_string_wrapper
 2.01% :     0.000004s :     18: predicate.reduce_eliminate
 1.69% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.42% :     0.000003s :     22: predicate.replace_applicator
 1.91% :     0.000003s :     33: predicate.replace_old_param
 0.11% :     0.000000s :      1: predicate.reset_defer_inline
 1.48% :     0.000003s :     18: predicate.reshape_eliminate
 1.30% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.55% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.42% :     0.000003s :     18: predicate.same_eliminate
 0.61% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.34% :     0.000001s :      2: predicate.special_op_eliminate
 1.20% :     0.000002s :     12: predicate.specialize_transform
 1.44% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.13% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.71% :     0.000005s :     26: predicate.switch_defer_inline
 2.22% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.83% :     0.000012s :     71: predicate.switch_simplify
 1.48% :     0.000003s :     18: predicate.tile_eliminate
 1.39% :     0.000003s :     18: predicate.transpose_eliminate
 1.65% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.91% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.44% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.50% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000003s :     18: predicate.value_based_eliminate
 0.12% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.23% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001106    22
74.02% :     0.000819s :     13: func_graph_cloner_run.FuncGraphClonerGraph
25.98% :     0.000287s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.024430    72
 0.15% :     0.000037s :      1: add_recomputation
 0.42% :     0.000102s :      1: auto_monad
 0.06% :     0.000015s :      1: auto_monad_reorder
 0.80% :     0.000194s :      1: bootstrap
 0.11% :     0.000027s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.09% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.20% :     0.000049s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
37.54% :     0.009171s :      1: jit_opt_a
 0.52% :     0.000127s :      1: jit_opt_after_cconv
 0.19% :     0.000046s :      1: jit_opt_b
 1.76% :     0.000430s :      1: loop_unroll
 2.69% :     0.000656s :      1: mutable_eliminate
 5.36% :     0.001311s :     26: opt.transform.jit_opt_a
 0.13% :     0.000032s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000018s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000023s :      4: opt.transform.symbol_engine_opt
 1.92% :     0.000469s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000006s :      1: pre_auto_parallel
 0.13% :     0.000032s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 5.53% :     0.001352s :      1: renormalize.infer
 1.70% :     0.000416s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000039s :      1: rewriter_after_opt_a
 0.40% :     0.000097s :      1: rewriter_before_opt_a
 0.27% :     0.000067s :      1: symbol_engine_optimizer
39.18% :     0.009571s :      1: type_inference

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.979.003 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e31f0> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:21.979.574 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0133649, [30]
[bootstrap]: 0.00036045
[type_inference]: 0.00496129
[event_method]: 1.9e-05
[auto_monad]: 7.011e-05
[graph_reusing]: 6.28e-06
[pre_auto_parallel]: 3.21999e-06
[py_interpret_to_execute]: 3.259e-05
[rewriter_before_opt_a]: 7.5e-05
[expand_dump_flag]: 3.18998e-06
[jit_opt_a]: 0.00564129, [2]
    [Cycle 1]: 0.00165826, [27]
        [switch_simplify]: 5.851e-05
        [loop_unroll]: 2.885e-05
        [a_1]: 0.00054358
        [with_stream_mark]: 2.09e-05
        [recompute_prepare]: 6.84999e-06
        [updatestate_depend_eliminate]: 3.81001e-06
        [updatestate_assign_eliminate]: 3.20002e-06
        [updatestate_loads_eliminate]: 2.67001e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 5.62999e-06
        [updatestate_useless_node_eliminater]: 4.74998e-06
        [accelerated_algorithm]: 5.56e-06
        [meta_shard_fg_expand]: 2.46e-06
        [get_grad_eliminate_]: 5.05001e-06
        [merge_forward]: 3.86999e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.503e-05
        [j_node_and_user_rematch]: 9.32999e-06
        [meta_fg_expand]: 2.69001e-06
        [replace_old_param]: 9.05999e-06
        [inline_without_move]: 5.32999e-06
        [renormalize]: 0.00066694
        [add_forward_monad_depend]: 6.63e-06
        [auto_monad_grad]: 2.65997e-06
        [auto_monad_eliminator]: 2.87e-05
        [cse]: 2.826e-05
        [replace_applicator]: 1.18e-05
    [Cycle 2]: 0.00031551, [27]
        [switch_simplify]: 5.59e-06
        [loop_unroll]: 4.72e-06
        [a_1]: 8.682e-05
        [with_stream_mark]: 9.97999e-06
        [recompute_prepare]: 4.58999e-06
        [updatestate_depend_eliminate]: 2.46998e-06
        [updatestate_assign_eliminate]: 2.11e-06
        [updatestate_loads_eliminate]: 1.77001e-06
        [parameter_eliminate]: 1.04e-06
        [specialize_transform]: 5.57001e-06
        [updatestate_useless_node_eliminater]: 4.57e-06
        [accelerated_algorithm]: 5.13002e-06
        [meta_shard_fg_expand]: 1.34998e-06
        [get_grad_eliminate_]: 4.32e-06
        [merge_forward]: 2.81e-06
        [cell_reuse_recompute_pass]: 1.15001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.254e-05
        [j_node_and_user_rematch]: 7.61999e-06
        [meta_fg_expand]: 1.57999e-06
        [replace_old_param]: 7.3e-06
        [inline_without_move]: 4.59002e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.14e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 4.62e-06
        [cse]: 9.96e-06
        [replace_applicator]: 5.20001e-06
[py_interpret_to_execute_after_opt_a]: 1.094e-05
[rewriter_after_opt_a]: 3.672e-05
[convert_after_rewriter]: 6.60997e-06
[order_py_execute_after_rewriter]: 4.75999e-06
[mutable_eliminate]: 0.00058295
[jit_opt_b]: 5.221e-05, [1]
    [Cycle 1]: 4.471e-05, [2]
        [frontend_op_eliminate]: 1.83e-05
        [inline_after_opt_a]: 1.424e-05
[cconv]: 2.796e-05
[loop_unroll]: 0.00044608
[jit_opt_after_cconv]: 0.00014201, [1]
    [Cycle 1]: 0.0001354, [11]
        [c_1]: 2.047e-05
        [parameter_eliminate]: 3.32002e-06
        [updatestate_depend_eliminate]: 6.04999e-06
        [updatestate_assign_eliminate]: 2.32001e-06
        [updatestate_loads_eliminate]: 2.04e-06
        [cse]: 1.969e-05
        [call_graph_tuple_transform]: 1.959e-05
        [tuple_list_get_item_eliminator]: 5.46e-06
        [none_parameter_eliminate]: 1.69998e-06
        [renormalize]: 6.40022e-07
        [switch_simplify]: 5.27999e-06
[remove_dup_value]: 1.46e-05
[partial_unused_args_eliminate]: 2.51e-06
[environ_conv]: 6.44999e-06
[add_recomputation]: 5.374e-05
[cse_after_recomputation]: 2.034e-05, [1]
    [Cycle 1]: 1.494e-05, [1]
        [cse]: 9.29e-06
[auto_monad_reorder]: 1.66e-05
[get_jit_bprop_graph]: 2.52001e-06
[rewriter_after_jit_bprop_graph]: 6.58e-06
[opt_after_jit_grad]: 0.00045822
[symbol_engine_optimizer]: 7.418e-05, [1]
    [Cycle 1]: 6.738e-05, [6]
        [build]: 4.15e-06
        [elim_shapecalc]: 7.82998e-06
        [elim_not_effective]: 1.268e-05
        [opt_reshape]: 5.72999e-06
        [fold_const_symbol]: 8.57e-06
        [renormalize]: 4.39992e-07
[validate]: 3.753e-05
Sums
    bootstrap                            :     0.000360s :  3.96%
    type_inference                       :     0.004961s : 54.53%
    event_method                         :     0.000019s :  0.21%
    auto_monad                           :     0.000070s :  0.77%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.04%
    py_interpret_to_execute              :     0.000033s :  0.36%
    rewriter_before_opt_a                :     0.000075s :  0.82%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000064s :  0.70%
    jit_opt_a.loop_unroll                :     0.000034s :  0.37%
    jit_opt_a.a_1                        :     0.000630s :  6.93%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.34%
    jit_opt_a.recompute_prepare          :     0.000011s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.18%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000667s :  7.33%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000033s :  0.37%
    jit_opt_a.cse                        :     0.000038s :  0.42%
    jit_opt_a.replace_applicator         :     0.000017s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000037s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000583s :  6.41%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000014s :  0.16%
    cconv                                :     0.000028s :  0.31%
    loop_unroll                          :     0.000446s :  4.90%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000054s :  0.59%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000017s :  0.18%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000458s :  5.04%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000038s :  0.41%


Time group info:
------[substitution.]   0.000220    23
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000002s :      2: substitution.fold_const_symbol
 2.41% :     0.000005s :      3: substitution.graph_param_transform
84.76% :     0.000187s :      4: substitution.inline
 1.38% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.03% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.03% :     0.000004s :      2: substitution.replace_old_param
 5.77% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004884     2
82.08% :     0.004009s :      1: type_inference.infer
17.92% :     0.000875s :      1: type_inference.specialize
------[replace.]   0.000066     6
68.02% :     0.000045s :      4: replace.inline
31.98% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000196     6
94.04% :     0.000184s :      4: match.inline
 5.96% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.18% :     0.000001s :     10: predicate.addn_zero_filter
 2.23% :     0.000003s :     10: predicate.arithmetic_simplify
 1.12% :     0.000001s :     10: predicate.cast_eliminate
 0.73% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.25% :     0.000001s :     10: predicate.depend_value_elim
 1.03% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.73% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.45% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.10% :     0.000001s :     10: predicate.environ_get_eliminate
 1.37% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.65% :     0.000001s :      3: predicate.graph_param_transform
 5.41% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.42% :     0.000002s :      6: predicate.less_batch_normalization
 1.36% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.47% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.75% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.04% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.39% :     0.000003s :      3: predicate.mutable_eliminate
 0.67% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.41% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     10: predicate.reduce_eliminate
 1.36% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.67% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.44% :     0.000002s :     10: predicate.reshape_eliminate
 1.09% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.45% :     0.000002s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.10% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.14% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.69% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.28% :     0.000009s :     53: predicate.switch_simplify
 1.41% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.33% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.09% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.81% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.56% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.56% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000639    12
53.28% :     0.000341s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.82% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.90% :     0.000255s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014923    72
 0.39% :     0.000058s :      1: add_recomputation
 0.50% :     0.000074s :      1: auto_monad
 0.13% :     0.000019s :      1: auto_monad_reorder
 2.57% :     0.000384s :      1: bootstrap
 0.20% :     0.000030s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
37.82% :     0.005644s :      1: jit_opt_a
 0.97% :     0.000145s :      1: jit_opt_after_cconv
 0.37% :     0.000055s :      1: jit_opt_b
 3.04% :     0.000454s :      1: loop_unroll
 3.96% :     0.000591s :      1: mutable_eliminate
 5.63% :     0.000839s :     26: opt.transform.jit_opt_a
 0.32% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.13% :     0.000466s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000035s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 2.46% :     0.000368s :      1: renormalize.infer
 1.95% :     0.000291s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000040s :      1: rewriter_after_opt_a
 0.53% :     0.000078s :      1: rewriter_before_opt_a
 0.51% :     0.000076s :      1: symbol_engine_optimizer
33.38% :     0.004981s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.983.614 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d69d0> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:21.984.181 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0233026, [33]
[bootstrap]: 0.00031132
[type_inference]: 0.0112531
[event_method]: 5.03e-05
[auto_monad]: 9.517e-05
[graph_reusing]: 7.5e-06
[pre_auto_parallel]: 2.87002e-06
[py_interpret_to_execute]: 2.715e-05
[rewriter_before_opt_a]: 9.314e-05
[expand_dump_flag]: 3.06999e-06
[jit_opt_a]: 0.00903479, [2]
    [Cycle 1]: 0.00496611, [27]
        [switch_simplify]: 7.196e-05
        [loop_unroll]: 4.125e-05
        [a_1]: 0.00087667
        [with_stream_mark]: 2.627e-05
        [recompute_prepare]: 1.954e-05
        [updatestate_depend_eliminate]: 8.53001e-06
        [updatestate_assign_eliminate]: 6.85002e-06
        [updatestate_loads_eliminate]: 6.10002e-06
        [parameter_eliminate]: 2.83e-06
        [specialize_transform]: 1.31e-05
        [updatestate_useless_node_eliminater]: 1.233e-05
        [accelerated_algorithm]: 1.259e-05
        [meta_shard_fg_expand]: 4.43999e-06
        [get_grad_eliminate_]: 1.199e-05
        [merge_forward]: 8.77e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.682e-05
        [j_node_and_user_rematch]: 2.233e-05
        [meta_fg_expand]: 0.00150339
        [replace_old_param]: 5.408e-05
        [inline_without_move]: 5.178e-05
        [renormalize]: 0.00188742
        [add_forward_monad_depend]: 6.48e-06
        [auto_monad_grad]: 2.40002e-06
        [auto_monad_eliminator]: 1.306e-05
        [cse]: 3.106e-05
        [replace_applicator]: 1.443e-05
    [Cycle 2]: 0.00033224, [27]
        [switch_simplify]: 4.63999e-06
        [loop_unroll]: 4.15999e-06
        [a_1]: 4.42e-05
        [with_stream_mark]: 1.102e-05
        [recompute_prepare]: 4.25999e-06
        [updatestate_depend_eliminate]: 2.55997e-06
        [updatestate_assign_eliminate]: 2.00002e-06
        [updatestate_loads_eliminate]: 1.66002e-06
        [parameter_eliminate]: 1.17999e-06
        [specialize_transform]: 4.57e-06
        [updatestate_useless_node_eliminater]: 4.22003e-06
        [accelerated_algorithm]: 3.85998e-06
        [meta_shard_fg_expand]: 1.39998e-06
        [get_grad_eliminate_]: 3.31999e-06
        [merge_forward]: 2.49001e-06
        [cell_reuse_recompute_pass]: 2.57001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.297e-05
        [j_node_and_user_rematch]: 6.07001e-06
        [meta_fg_expand]: 4.328e-05
        [replace_old_param]: 4.87e-06
        [inline_without_move]: 3.48999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 3.13e-06
        [auto_monad_grad]: 1.09e-06
        [auto_monad_eliminator]: 7.77002e-06
        [cse]: 1.672e-05
        [replace_applicator]: 5.46002e-06
[py_interpret_to_execute_after_opt_a]: 1.403e-05
[rewriter_after_opt_a]: 3.603e-05
[convert_after_rewriter]: 7.1e-06
[order_py_execute_after_rewriter]: 5.16002e-06
[mutable_eliminate]: 0.00066608
[jit_opt_b]: 4.56e-05, [1]
    [Cycle 1]: 3.71e-05, [2]
        [frontend_op_eliminate]: 1.156e-05
        [inline_after_opt_a]: 1.315e-05
[cconv]: 2.732e-05
[loop_unroll]: 0.00052257
[jit_opt_after_cconv]: 0.00013599, [1]
    [Cycle 1]: 0.00012771, [11]
        [c_1]: 1.283e-05
        [parameter_eliminate]: 3.5e-06
        [updatestate_depend_eliminate]: 5.45001e-06
        [updatestate_assign_eliminate]: 2.20002e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [cse]: 2.475e-05
        [call_graph_tuple_transform]: 1.54e-05
        [tuple_list_get_item_eliminator]: 3.8e-06
        [none_parameter_eliminate]: 1.60001e-06
        [renormalize]: 6.29982e-07
        [switch_simplify]: 3.81001e-06
[remove_dup_value]: 1.621e-05
[partial_unused_args_eliminate]: 1.97999e-06
[environ_conv]: 5.25001e-06
[add_recomputation]: 3.861e-05
[cse_after_recomputation]: 2.166e-05, [1]
    [Cycle 1]: 1.504e-05, [1]
        [cse]: 9.05999e-06
[auto_monad_reorder]: 1.377e-05
[get_jit_bprop_graph]: 1.92001e-06
[rewriter_after_jit_bprop_graph]: 6.04001e-06
[opt_after_jit_grad]: 0.00052188
[symbol_engine_optimizer]: 7.016e-05, [1]
    [Cycle 1]: 6.302e-05, [6]
        [build]: 3.66999e-06
        [elim_shapecalc]: 6.40002e-06
        [elim_not_effective]: 1.027e-05
        [opt_reshape]: 4.87e-06
        [fold_const_symbol]: 6.49001e-06
        [renormalize]: 6.10016e-07
[validate]: 3.645e-05
[backend_pass]: 9.00007e-07
[task_emit]: 2.957e-05
[execute]: 1.30999e-06
Sums
    bootstrap                            :     0.000311s :  1.65%
    type_inference                       :     0.011253s : 59.61%
    event_method                         :     0.000050s :  0.27%
    auto_monad                           :     0.000095s :  0.50%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000027s :  0.14%
    rewriter_before_opt_a                :     0.000093s :  0.49%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000077s :  0.41%
    jit_opt_a.loop_unroll                :     0.000045s :  0.24%
    jit_opt_a.a_1                        :     0.000921s :  4.88%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000024s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000018s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000017s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.08%
    jit_opt_a.merge_forward              :     0.000011s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000040s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001547s :  8.19%
    jit_opt_a.replace_old_param          :     0.000059s :  0.31%
    jit_opt_a.inline_without_move        :     0.000055s :  0.29%
    jit_opt_a.renormalize                :     0.001888s : 10.00%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.11%
    jit_opt_a.cse                        :     0.000048s :  0.25%
    jit_opt_a.replace_applicator         :     0.000020s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.07%
    rewriter_after_opt_a                 :     0.000036s :  0.19%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000666s :  3.53%
    jit_opt_b.frontend_op_eliminate      :     0.000012s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000013s :  0.07%
    cconv                                :     0.000027s :  0.14%
    loop_unroll                          :     0.000523s :  2.77%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000016s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000039s :  0.20%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.03%
    opt_after_jit_grad                   :     0.000522s :  2.76%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000036s :  0.19%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000030s :  0.16%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000275    50
 0.49% :     0.000001s :      1: substitution.elim_not_effective
 0.41% :     0.000001s :      1: substitution.fold_const_symbol
 1.74% :     0.000005s :      1: substitution.graph_param_transform
67.42% :     0.000185s :      6: substitution.inline
 6.36% :     0.000017s :      2: substitution.inline_without_move
 2.61% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.40% :     0.000004s :      2: substitution.minmaximum_grad
 1.49% :     0.000004s :      3: substitution.partial_eliminate
 2.74% :     0.000008s :      9: substitution.remove_not_recompute_node
 2.26% :     0.000006s :      2: substitution.replace_applicator
 1.60% :     0.000004s :      5: substitution.replace_old_param
 1.12% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.68% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.76% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.91% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011178     2
88.98% :     0.009946s :      1: type_inference.infer
11.02% :     0.001232s :      1: type_inference.specialize
------[replace.]   0.000068     8
76.85% :     0.000052s :      6: replace.inline
23.15% :     0.000016s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000193     8
94.02% :     0.000182s :      6: match.inline
 5.98% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000178  1109
 1.45% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.63% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.32% :     0.000002s :     18: predicate.addn_check_dump
 1.42% :     0.000003s :     18: predicate.addn_zero_filter
 2.27% :     0.000004s :     18: predicate.arithmetic_simplify
 1.56% :     0.000003s :     18: predicate.cast_eliminate
 0.13% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.27% :     0.000002s :     18: predicate.compare_switch_simplify
 1.78% :     0.000003s :     18: predicate.depend_value_elim
 1.21% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.54% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.22% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.44% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.19% :     0.000000s :      1: predicate.elim_not_effective
 0.34% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.28% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.40% :     0.000002s :     18: predicate.environ_get_eliminate
 1.20% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.05% :     0.000002s :     12: predicate.get_grad_eliminate
 0.18% :     0.000000s :      1: predicate.graph_param_transform
 4.40% :     0.000008s :     28: predicate.inline
 2.93% :     0.000005s :     33: predicate.inline_without_move
 0.49% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.37% :     0.000002s :     12: predicate.less_batch_normalization
 1.51% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.80% :     0.000003s :     21: predicate.load_eliminater
 0.79% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.92% :     0.000007s :     44: predicate.loop_unroll_before_grad
 1.94% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.21% :     0.000002s :     18: predicate.merge_addn
 1.27% :     0.000002s :     18: predicate.minmaximum_grad
 1.05% :     0.000002s :      1: predicate.mutable_eliminate
 0.27% :     0.000000s :      1: predicate.opt_reshape
 2.11% :     0.000004s :     21: predicate.partial_eliminate
 1.34% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.68% :     0.000003s :     18: predicate.reduce_eliminate
 1.54% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.80% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.54% :     0.000003s :     22: predicate.replace_applicator
 1.74% :     0.000003s :     33: predicate.replace_old_param
 0.16% :     0.000000s :      1: predicate.reset_defer_inline
 1.42% :     0.000003s :     18: predicate.reshape_eliminate
 1.31% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.52% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.25% :     0.000002s :     18: predicate.same_eliminate
 0.59% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.50% :     0.000001s :      2: predicate.special_op_eliminate
 1.10% :     0.000002s :     12: predicate.specialize_transform
 1.47% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.22% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.73% :     0.000005s :     26: predicate.switch_defer_inline
 2.27% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.85% :     0.000012s :     71: predicate.switch_simplify
 1.37% :     0.000002s :     18: predicate.tile_eliminate
 1.20% :     0.000002s :     18: predicate.transpose_eliminate
 1.60% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 3.71% :     0.000007s :     22: predicate.tuple_list_get_item_eliminator
 1.93% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.58% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.39% :     0.000006s :     33: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000003s :     18: predicate.value_based_eliminate
 0.13% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.26% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001269    22
71.12% :     0.000902s :     13: func_graph_cloner_run.FuncGraphClonerGraph
28.88% :     0.000366s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.026561    76
 0.16% :     0.000042s :      1: add_recomputation
 0.37% :     0.000099s :      1: auto_monad
 0.06% :     0.000016s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 1.24% :     0.000329s :      1: bootstrap
 0.12% :     0.000031s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.09% :     0.000024s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.20% :     0.000054s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
34.03% :     0.009038s :      1: jit_opt_a
 0.52% :     0.000139s :      1: jit_opt_after_cconv
 0.18% :     0.000048s :      1: jit_opt_b
 2.01% :     0.000533s :      1: loop_unroll
 2.55% :     0.000677s :      1: mutable_eliminate
 4.90% :     0.001301s :     26: opt.transform.jit_opt_a
 0.12% :     0.000033s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000017s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000017s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000024s :      4: opt.transform.symbol_engine_opt
 2.00% :     0.000532s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.11% :     0.000030s :      1: py_interpret_to_execute
 0.06% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000019s :      1: remove_dup_value
 5.16% :     0.001372s :      1: renormalize.infer
 1.91% :     0.000507s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.15% :     0.000040s :      1: rewriter_after_opt_a
 0.36% :     0.000096s :      1: rewriter_before_opt_a
 0.27% :     0.000073s :      1: symbol_engine_optimizer
 0.13% :     0.000034s :      1: task_emit
42.42% :     0.011267s :      1: type_inference
 0.21% :     0.000055s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:21.999.794 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c3688b0> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:22.458. [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0202323, [30]
[bootstrap]: 0.00018438
[type_inference]: 0.0097192
[event_method]: 4.382e-05
[auto_monad]: 8.892e-05
[graph_reusing]: 7.01999e-06
[pre_auto_parallel]: 2.59999e-06
[py_interpret_to_execute]: 2.589e-05
[rewriter_before_opt_a]: 8.669e-05
[expand_dump_flag]: 2.92002e-06
[jit_opt_a]: 0.0080654, [2]
    [Cycle 1]: 0.00438793, [27]
        [switch_simplify]: 6.112e-05
        [loop_unroll]: 3.914e-05
        [a_1]: 0.00081935
        [with_stream_mark]: 2.424e-05
        [recompute_prepare]: 1.805e-05
        [updatestate_depend_eliminate]: 8.66002e-06
        [updatestate_assign_eliminate]: 6.58e-06
        [updatestate_loads_eliminate]: 6.02001e-06
        [parameter_eliminate]: 2.63998e-06
        [specialize_transform]: 1.281e-05
        [updatestate_useless_node_eliminater]: 1.208e-05
        [accelerated_algorithm]: 1.34e-05
        [meta_shard_fg_expand]: 3.61999e-06
        [get_grad_eliminate_]: 1.182e-05
        [merge_forward]: 7.95e-06
        [cell_reuse_recompute_pass]: 9.39996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.619e-05
        [j_node_and_user_rematch]: 2.229e-05
        [meta_fg_expand]: 0.00146222
        [replace_old_param]: 4.817e-05
        [inline_without_move]: 5.097e-05
        [renormalize]: 0.00148732
        [add_forward_monad_depend]: 4.35e-06
        [auto_monad_grad]: 1.67001e-06
        [auto_monad_eliminator]: 8.54e-06
        [cse]: 2.16e-05
        [replace_applicator]: 9.52001e-06
    [Cycle 2]: 0.00027452, [27]
        [switch_simplify]: 4.40999e-06
        [loop_unroll]: 3.93001e-06
        [a_1]: 3.999e-05
        [with_stream_mark]: 8.30999e-06
        [recompute_prepare]: 3.69002e-06
        [updatestate_depend_eliminate]: 2.16998e-06
        [updatestate_assign_eliminate]: 1.61998e-06
        [updatestate_loads_eliminate]: 1.59e-06
        [parameter_eliminate]: 1.25999e-06
        [specialize_transform]: 4.13001e-06
        [updatestate_useless_node_eliminater]: 3.73999e-06
        [accelerated_algorithm]: 3.9e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 3.61999e-06
        [merge_forward]: 2.13002e-06
        [cell_reuse_recompute_pass]: 1.52999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.62999e-06
        [j_node_and_user_rematch]: 5.94e-06
        [meta_fg_expand]: 2.562e-05
        [replace_old_param]: 3.92998e-06
        [inline_without_move]: 3.24001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.42999e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 4.50001e-06
        [cse]: 1.091e-05
        [replace_applicator]: 3.80998e-06
[py_interpret_to_execute_after_opt_a]: 9.74e-06
[rewriter_after_opt_a]: 3.359e-05
[convert_after_rewriter]: 6.91001e-06
[order_py_execute_after_rewriter]: 4.63001e-06
[mutable_eliminate]: 0.00058668
[jit_opt_b]: 3.808e-05, [1]
    [Cycle 1]: 3.091e-05, [2]
        [frontend_op_eliminate]: 9.65002e-06
        [inline_after_opt_a]: 1.056e-05
[cconv]: 2.317e-05
[loop_unroll]: 0.00041752
[jit_opt_after_cconv]: 0.00012044, [1]
    [Cycle 1]: 0.00011405, [11]
        [c_1]: 1.249e-05
        [parameter_eliminate]: 2.70002e-06
        [updatestate_depend_eliminate]: 4.37e-06
        [updatestate_assign_eliminate]: 1.95001e-06
        [updatestate_loads_eliminate]: 1.67001e-06
        [cse]: 1.7e-05
        [call_graph_tuple_transform]: 1.362e-05
        [tuple_list_get_item_eliminator]: 3.92998e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 4.12e-06
[remove_dup_value]: 1.445e-05
[partial_unused_args_eliminate]: 2.37001e-06
[environ_conv]: 4.99e-06
[add_recomputation]: 3.471e-05
[cse_after_recomputation]: 2.035e-05, [1]
    [Cycle 1]: 1.471e-05, [1]
        [cse]: 8.58001e-06
[auto_monad_reorder]: 1.36e-05
[get_jit_bprop_graph]: 1.71998e-06
[rewriter_after_jit_bprop_graph]: 5.28002e-06
[opt_after_jit_grad]: 0.0004164
[symbol_engine_optimizer]: 6.552e-05, [1]
    [Cycle 1]: 5.958e-05, [6]
        [build]: 3.66999e-06
        [elim_shapecalc]: 5.52999e-06
        [elim_not_effective]: 1.099e-05
        [opt_reshape]: 4.65999e-06
        [fold_const_symbol]: 6.39001e-06
        [renormalize]: 4.80009e-07
[validate]: 3.15e-05
Sums
    bootstrap                            :     0.000184s :  1.14%
    type_inference                       :     0.009719s : 59.84%
    event_method                         :     0.000044s :  0.27%
    auto_monad                           :     0.000089s :  0.55%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.16%
    rewriter_before_opt_a                :     0.000087s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000066s :  0.40%
    jit_opt_a.loop_unroll                :     0.000043s :  0.27%
    jit_opt_a.a_1                        :     0.000859s :  5.29%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000017s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.10%
    jit_opt_a.merge_forward              :     0.000010s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001488s :  9.16%
    jit_opt_a.replace_old_param          :     0.000052s :  0.32%
    jit_opt_a.inline_without_move        :     0.000054s :  0.33%
    jit_opt_a.renormalize                :     0.001487s :  9.16%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.08%
    jit_opt_a.cse                        :     0.000033s :  0.20%
    jit_opt_a.replace_applicator         :     0.000013s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000034s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000587s :  3.61%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000023s :  0.14%
    loop_unroll                          :     0.000418s :  2.57%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000035s :  0.21%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000416s :  2.56%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.19%


Time group info:
------[substitution.]   0.000238    50
 0.68% :     0.000002s :      1: substitution.elim_not_effective
 0.48% :     0.000001s :      1: substitution.fold_const_symbol
 1.67% :     0.000004s :      1: substitution.graph_param_transform
64.85% :     0.000154s :      6: substitution.inline
 7.03% :     0.000017s :      2: substitution.inline_without_move
 2.70% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.47% :     0.000003s :      2: substitution.minmaximum_grad
 1.72% :     0.000004s :      3: substitution.partial_eliminate
 2.87% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.39% :     0.000006s :      2: substitution.replace_applicator
 1.65% :     0.000004s :      5: substitution.replace_old_param
 1.03% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.99% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 2.05% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 6.43% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009646     2
87.74% :     0.008463s :      1: type_inference.infer
12.26% :     0.001183s :      1: type_inference.specialize
------[replace.]   0.000059     8
75.98% :     0.000045s :      6: replace.inline
24.02% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000161     8
93.36% :     0.000151s :      6: match.inline
 6.64% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000168  1109
 1.55% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.66% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.27% :     0.000002s :     18: predicate.addn_check_dump
 1.51% :     0.000003s :     18: predicate.addn_zero_filter
 1.94% :     0.000003s :     18: predicate.arithmetic_simplify
 1.45% :     0.000002s :     18: predicate.cast_eliminate
 0.15% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.24% :     0.000002s :     18: predicate.compare_switch_simplify
 1.56% :     0.000003s :     18: predicate.depend_value_elim
 1.27% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.32% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.32% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.10% :     0.000000s :      1: predicate.elim_not_effective
 0.27% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.33% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.26% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.43% :     0.000002s :     18: predicate.environ_get_eliminate
 1.27% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.07% :     0.000000s :      1: predicate.fold_const_symbol
 1.25% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.32% :     0.000007s :     28: predicate.inline
 3.20% :     0.000005s :     33: predicate.inline_without_move
 0.52% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.75% :     0.000003s :     12: predicate.less_batch_normalization
 1.56% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.69% :     0.000003s :     21: predicate.load_eliminater
 0.78% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.80% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.91% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.27% :     0.000002s :     18: predicate.merge_addn
 1.44% :     0.000002s :     18: predicate.minmaximum_grad
 0.72% :     0.000001s :      1: predicate.mutable_eliminate
 0.44% :     0.000001s :      1: predicate.opt_reshape
 2.10% :     0.000004s :     21: predicate.partial_eliminate
 1.35% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.88% :     0.000003s :     18: predicate.reduce_eliminate
 1.55% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.53% :     0.000003s :     22: predicate.replace_applicator
 1.73% :     0.000003s :     33: predicate.replace_old_param
 0.07% :     0.000000s :      1: predicate.reset_defer_inline
 1.47% :     0.000002s :     18: predicate.reshape_eliminate
 1.53% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.34% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     18: predicate.same_eliminate
 0.66% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.31% :     0.000001s :      2: predicate.special_op_eliminate
 1.24% :     0.000002s :     12: predicate.specialize_transform
 1.59% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.79% :     0.000005s :     26: predicate.switch_defer_inline
 2.34% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.86% :     0.000012s :     71: predicate.switch_simplify
 1.46% :     0.000002s :     18: predicate.tile_eliminate
 1.59% :     0.000003s :     18: predicate.transpose_eliminate
 1.71% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.49% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.17% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.73% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.63% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.09% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.82% :     0.000003s :     18: predicate.value_based_eliminate
 0.12% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.22% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001087    22
71.71% :     0.000779s :     13: func_graph_cloner_run.FuncGraphClonerGraph
28.29% :     0.000308s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022946    72
 0.16% :     0.000037s :      1: add_recomputation
 0.40% :     0.000092s :      1: auto_monad
 0.07% :     0.000016s :      1: auto_monad_reorder
 0.87% :     0.000199s :      1: bootstrap
 0.11% :     0.000026s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.21% :     0.000047s :      1: event_method
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
35.16% :     0.008068s :      1: jit_opt_a
 0.54% :     0.000123s :      1: jit_opt_after_cconv
 0.18% :     0.000041s :      1: jit_opt_b
 1.85% :     0.000425s :      1: loop_unroll
 2.58% :     0.000593s :      1: mutable_eliminate
 5.28% :     0.001211s :     26: opt.transform.jit_opt_a
 0.13% :     0.000031s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000014s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000009s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000013s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000024s :      4: opt.transform.symbol_engine_opt
 1.84% :     0.000422s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000028s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 4.93% :     0.001131s :      1: renormalize.infer
 1.52% :     0.000349s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000037s :      1: rewriter_after_opt_a
 0.39% :     0.000089s :      1: rewriter_before_opt_a
 0.30% :     0.000068s :      1: symbol_engine_optimizer
42.40% :     0.009730s :      1: type_inference


TotalTime = 0.0133382, [30]
[bootstrap]: 0.00030305
[type_inference]: 0.00445779
[event_method]: 1.818e-05
[auto_monad]: 6.634e-05
[graph_reusing]: 5.86e-06
[pre_auto_parallel]: 2.82002e-06
[py_interpret_to_execute]: 3.191e-05
[rewriter_before_opt_a]: 7.571e-05
[expand_dump_flag]: 3.41999e-06
[jit_opt_a]: 0.00602078, [2]
    [Cycle 1]: 0.0017946, [27]
        [switch_simplify]: 6.039e-05
        [loop_unroll]: 2.843e-05
        [a_1]: 0.00058152
        [with_stream_mark]: 2.384e-05
        [recompute_prepare]: 9.01998e-06
        [updatestate_depend_eliminate]: 4.05e-06
        [updatestate_assign_eliminate]: 3.18998e-06
        [updatestate_loads_eliminate]: 3.01999e-06
        [parameter_eliminate]: 2.26e-06
        [specialize_transform]: 5.89999e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.50001e-06
        [meta_shard_fg_expand]: 2.44999e-06
        [get_grad_eliminate_]: 5.19998e-06
        [merge_forward]: 4.38001e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.697e-05
        [j_node_and_user_rematch]: 9.91998e-06
        [meta_fg_expand]: 2.58003e-06
        [replace_old_param]: 9.22999e-06
        [inline_without_move]: 5.34e-06
        [renormalize]: 0.00072163
        [add_forward_monad_depend]: 6.84001e-06
        [auto_monad_grad]: 2.64001e-06
        [auto_monad_eliminator]: 1.763e-05
        [cse]: 3.471e-05
        [replace_applicator]: 1.39e-05
    [Cycle 2]: 0.00035567, [27]
        [switch_simplify]: 6.09999e-06
        [loop_unroll]: 5.64998e-06
        [a_1]: 9.502e-05
        [with_stream_mark]: 1.177e-05
        [recompute_prepare]: 6.17001e-06
        [updatestate_depend_eliminate]: 2.94001e-06
        [updatestate_assign_eliminate]: 2.41998e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 5.04e-06
        [updatestate_useless_node_eliminater]: 4.65001e-06
        [accelerated_algorithm]: 5.19e-06
        [meta_shard_fg_expand]: 1.76e-06
        [get_grad_eliminate_]: 4.33001e-06
        [merge_forward]: 3.45e-06
        [cell_reuse_recompute_pass]: 1.41998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.507e-05
        [j_node_and_user_rematch]: 8.10999e-06
        [meta_fg_expand]: 2.76999e-06
        [replace_old_param]: 8.07e-06
        [inline_without_move]: 9.66998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.52001e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 6.64999e-06
        [cse]: 1.466e-05
        [replace_applicator]: 6.19999e-06
[py_interpret_to_execute_after_opt_a]: 1.432e-05
[rewriter_after_opt_a]: 4.29e-05
[convert_after_rewriter]: 6.99001e-06
[order_py_execute_after_rewriter]: 5.61e-06
[mutable_eliminate]: 0.00063918
[jit_opt_b]: 5.665e-05, [1]
    [Cycle 1]: 4.77e-05, [2]
        [frontend_op_eliminate]: 1.817e-05
        [inline_after_opt_a]: 1.673e-05
[cconv]: 2.793e-05
[loop_unroll]: 0.00047845
[jit_opt_after_cconv]: 0.00014626, [1]
    [Cycle 1]: 0.00013878, [11]
        [c_1]: 1.962e-05
        [parameter_eliminate]: 3.5e-06
        [updatestate_depend_eliminate]: 7.01001e-06
        [updatestate_assign_eliminate]: 2.64999e-06
        [updatestate_loads_eliminate]: 2.33998e-06
        [cse]: 2.073e-05
        [call_graph_tuple_transform]: 2.002e-05
        [tuple_list_get_item_eliminator]: 5.50001e-06
        [none_parameter_eliminate]: 1.65001e-06
        [renormalize]: 6.29982e-07
        [switch_simplify]: 5.02e-06
[remove_dup_value]: 1.529e-05
[partial_unused_args_eliminate]: 2.25002e-06
[environ_conv]: 5.75001e-06
[add_recomputation]: 5.579e-05
[cse_after_recomputation]: 2.076e-05, [1]
    [Cycle 1]: 1.474e-05, [1]
        [cse]: 8.69e-06
[auto_monad_reorder]: 1.59e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 8.15e-06
[opt_after_jit_grad]: 0.00049817
[symbol_engine_optimizer]: 7.57e-05, [1]
    [Cycle 1]: 6.853e-05, [6]
        [build]: 3.98001e-06
        [elim_shapecalc]: 7.46001e-06
        [elim_not_effective]: 1.309e-05
        [opt_reshape]: 5.66e-06
        [fold_const_symbol]: 9.10001e-06
        [renormalize]: 5.60016e-07
[validate]: 3.892e-05
Sums
    bootstrap                            :     0.000303s :  3.44%
    type_inference                       :     0.004458s : 50.57%
    event_method                         :     0.000018s :  0.21%
    auto_monad                           :     0.000066s :  0.75%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.36%
    rewriter_before_opt_a                :     0.000076s :  0.86%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000066s :  0.75%
    jit_opt_a.loop_unroll                :     0.000034s :  0.39%
    jit_opt_a.a_1                        :     0.000677s :  7.68%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.40%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000008s :  0.09%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.36%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.06%
    jit_opt_a.replace_old_param          :     0.000017s :  0.20%
    jit_opt_a.inline_without_move        :     0.000015s :  0.17%
    jit_opt_a.renormalize                :     0.000722s :  8.19%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.28%
    jit_opt_a.cse                        :     0.000049s :  0.56%
    jit_opt_a.replace_applicator         :     0.000020s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.16%
    rewriter_after_opt_a                 :     0.000043s :  0.49%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000639s :  7.25%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.19%
    cconv                                :     0.000028s :  0.32%
    loop_unroll                          :     0.000478s :  5.43%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000021s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000056s :  0.63%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.09%
    opt_after_jit_grad                   :     0.000498s :  5.65%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.44%


Time group info:
------[substitution.]   0.000244    23
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 2.28% :     0.000006s :      3: substitution.graph_param_transform
84.66% :     0.000207s :      4: substitution.inline
 1.63% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.19% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.05% :     0.000005s :      2: substitution.replace_old_param
 5.79% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004384     2
82.02% :     0.003596s :      1: type_inference.infer
17.98% :     0.000788s :      1: type_inference.specialize
------[replace.]   0.000075     6
66.96% :     0.000050s :      4: replace.inline
33.04% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000217     6
94.11% :     0.000204s :      4: match.inline
 5.89% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.29% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 2.18% :     0.000003s :     10: predicate.addn_zero_filter
 2.08% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000002s :     10: predicate.cast_eliminate
 0.65% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 1.82% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.05% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.08% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.83% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.94% :     0.000007s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.34% :     0.000002s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.57% :     0.000002s :     15: predicate.load_eliminater
 1.40% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.57% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 2.18% :     0.000003s :      3: predicate.mutable_eliminate
 0.59% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     15: predicate.partial_eliminate
 1.29% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.43% :     0.000002s :     10: predicate.reduce_eliminate
 1.35% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.36% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.22% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.65% :     0.000003s :     16: predicate.switch_defer_inline
 1.83% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.11% :     0.000010s :     53: predicate.switch_simplify
 1.01% :     0.000001s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.14% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.05% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.64% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.96% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.57% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.43% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.12% :     0.000003s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.89% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000588    12
50.06% :     0.000294s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.34% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.59% :     0.000244s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015022    72
 0.39% :     0.000059s :      1: add_recomputation
 0.47% :     0.000070s :      1: auto_monad
 0.13% :     0.000019s :      1: auto_monad_reorder
 2.15% :     0.000322s :      1: bootstrap
 0.20% :     0.000031s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
40.10% :     0.006025s :      1: jit_opt_a
 0.99% :     0.000149s :      1: jit_opt_after_cconv
 0.40% :     0.000059s :      1: jit_opt_b
 3.25% :     0.000489s :      1: loop_unroll
 4.34% :     0.000653s :      1: mutable_eliminate
 6.02% :     0.000905s :     26: opt.transform.jit_opt_a
 0.31% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000028s :      4: opt.transform.jit_opt_b
 0.10% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.13% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.39% :     0.000509s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000035s :      1: py_interpret_to_execute
 0.12% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000018s :      1: remove_dup_value
 2.70% :     0.000406s :      1: renormalize.infer
 2.04% :     0.000306s :      1: renormalize.specialize
 0.07% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.31% :     0.000046s :      1: rewriter_after_opt_a
 0.52% :     0.000079s :      1: rewriter_before_opt_a
 0.52% :     0.000079s :      1: symbol_engine_optimizer
29.79% :     0.004475s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.123.75 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d6430> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.130.46 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0341177, [33]
[bootstrap]: 0.00020132
[type_inference]: 0.0102696
[event_method]: 5.644e-05
[auto_monad]: 9.856e-05
[graph_reusing]: 6.73e-06
[pre_auto_parallel]: 2.72001e-06
[py_interpret_to_execute]: 2.731e-05
[rewriter_before_opt_a]: 9.28e-05
[expand_dump_flag]: 3.33e-06
[jit_opt_a]: 0.0126053, [3]
    [Cycle 1]: 0.00630643, [27]
        [switch_simplify]: 6.879e-05
        [loop_unroll]: 3.831e-05
        [a_1]: 0.00084616
        [with_stream_mark]: 2.412e-05
        [recompute_prepare]: 1.924e-05
        [updatestate_depend_eliminate]: 8.22e-06
        [updatestate_assign_eliminate]: 6.02001e-06
        [updatestate_loads_eliminate]: 5.89999e-06
        [parameter_eliminate]: 2.30002e-06
        [specialize_transform]: 1.337e-05
        [updatestate_useless_node_eliminater]: 1.161e-05
        [accelerated_algorithm]: 1.269e-05
        [meta_shard_fg_expand]: 3.88001e-06
        [get_grad_eliminate_]: 1.242e-05
        [merge_forward]: 8.03001e-06
        [cell_reuse_recompute_pass]: 7.40023e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.586e-05
        [j_node_and_user_rematch]: 2.249e-05
        [meta_fg_expand]: 0.00153267
        [replace_old_param]: 6.932e-05
        [inline_without_move]: 5.598e-05
        [renormalize]: 0.00299792
        [add_forward_monad_depend]: 9.67999e-06
        [auto_monad_grad]: 4.94003e-06
        [auto_monad_eliminator]: 5.385e-05
        [cse]: 0.00016719
        [replace_applicator]: 6.77e-05
    [Cycle 2]: 0.00222786, [27]
        [switch_simplify]: 3.825e-05
        [loop_unroll]: 3.617e-05
        [a_1]: 0.0011685
        [with_stream_mark]: 1.391e-05
        [recompute_prepare]: 8.74e-06
        [updatestate_depend_eliminate]: 4.3e-06
        [updatestate_assign_eliminate]: 3.79002e-06
        [updatestate_loads_eliminate]: 3.09001e-06
        [parameter_eliminate]: 1.45999e-06
        [specialize_transform]: 7.43e-06
        [updatestate_useless_node_eliminater]: 7.15998e-06
        [accelerated_algorithm]: 8.13999e-06
        [meta_shard_fg_expand]: 1.92999e-06
        [get_grad_eliminate_]: 7.47998e-06
        [merge_forward]: 3.76999e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.514e-05
        [j_node_and_user_rematch]: 1.203e-05
        [meta_fg_expand]: 5.659e-05
        [replace_old_param]: 1.096e-05
        [inline_without_move]: 7.45998e-06
        [renormalize]: 0.00061464
        [add_forward_monad_depend]: 3.78001e-06
        [auto_monad_grad]: 1.10999e-06
        [auto_monad_eliminator]: 1.055e-05
        [cse]: 2.096e-05
        [replace_applicator]: 1.209e-05
    [Cycle 3]: 0.00041271, [27]
        [switch_simplify]: 8.1e-06
        [loop_unroll]: 7.08e-06
        [a_1]: 0.00014672
        [with_stream_mark]: 9.10999e-06
        [recompute_prepare]: 6.91001e-06
        [updatestate_depend_eliminate]: 3.51999e-06
        [updatestate_assign_eliminate]: 3.26999e-06
        [updatestate_loads_eliminate]: 2.90998e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 6.89999e-06
        [updatestate_useless_node_eliminater]: 6.73e-06
        [accelerated_algorithm]: 7.26001e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 6.78e-06
        [merge_forward]: 3.54002e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.546e-05
        [j_node_and_user_rematch]: 1.145e-05
        [meta_fg_expand]: 2.49999e-06
        [replace_old_param]: 9.48002e-06
        [inline_without_move]: 6.88998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 7.82e-06
        [cse]: 1.688e-05
        [replace_applicator]: 6.74999e-06
[py_interpret_to_execute_after_opt_a]: 1.137e-05
[rewriter_after_opt_a]: 3.68e-05
[convert_after_rewriter]: 7.94997e-06
[order_py_execute_after_rewriter]: 4.95001e-06
[mutable_eliminate]: 0.00060882
[jit_opt_b]: 5.832e-05, [1]
    [Cycle 1]: 5.166e-05, [2]
        [frontend_op_eliminate]: 2.053e-05
        [inline_after_opt_a]: 1.942e-05
[cconv]: 1.604e-05
[loop_unroll]: 0.00042359
[jit_opt_after_cconv]: 0.00016134, [1]
    [Cycle 1]: 0.00015523, [11]
        [c_1]: 3.026e-05
        [parameter_eliminate]: 3.33e-06
        [updatestate_depend_eliminate]: 5.54998e-06
        [updatestate_assign_eliminate]: 3.3e-06
        [updatestate_loads_eliminate]: 3.35e-06
        [cse]: 2.507e-05
        [call_graph_tuple_transform]: 1.964e-05
        [tuple_list_get_item_eliminator]: 7.89002e-06
        [none_parameter_eliminate]: 1.08001e-06
        [renormalize]: 3.10014e-07
        [switch_simplify]: 6.98e-06
[remove_dup_value]: 1.109e-05
[partial_unused_args_eliminate]: 1.49e-06
[environ_conv]: 5.16002e-06
[add_recomputation]: 4.102e-05
[cse_after_recomputation]: 2.919e-05, [1]
    [Cycle 1]: 2.43e-05, [1]
        [cse]: 1.708e-05
[auto_monad_reorder]: 1.529e-05
[get_jit_bprop_graph]: 1.30999e-06
[rewriter_after_jit_bprop_graph]: 5.15001e-06
[opt_after_jit_grad]: 0.00048049
[symbol_engine_optimizer]: 8.633e-05, [1]
    [Cycle 1]: 7.963e-05, [6]
        [build]: 3.72998e-06
        [elim_shapecalc]: 1.059e-05
        [elim_not_effective]: 1.651e-05
        [opt_reshape]: 8e-06
        [fold_const_symbol]: 1.171e-05
        [renormalize]: 5.89993e-07
[validate]: 0.00012737
[backend_pass]: 8.40024e-07
[task_emit]: 0.00840561
[execute]: 8.1e-06
Sums
    bootstrap                            :     0.000201s :  0.68%
    type_inference                       :     0.010270s : 34.63%
    event_method                         :     0.000056s :  0.19%
    auto_monad                           :     0.000099s :  0.33%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000027s :  0.09%
    rewriter_before_opt_a                :     0.000093s :  0.31%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000115s :  0.39%
    jit_opt_a.loop_unroll                :     0.000082s :  0.27%
    jit_opt_a.a_1                        :     0.002161s :  7.29%
    jit_opt_a.with_stream_mark           :     0.000047s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000035s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000013s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000028s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000025s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000028s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000027s :  0.09%
    jit_opt_a.merge_forward              :     0.000015s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000056s :  0.19%
    jit_opt_a.j_node_and_user_rematch    :     0.000046s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001592s :  5.37%
    jit_opt_a.replace_old_param          :     0.000090s :  0.30%
    jit_opt_a.inline_without_move        :     0.000070s :  0.24%
    jit_opt_a.renormalize                :     0.003613s : 12.18%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000072s :  0.24%
    jit_opt_a.cse                        :     0.000205s :  0.69%
    jit_opt_a.replace_applicator         :     0.000087s :  0.29%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.04%
    rewriter_after_opt_a                 :     0.000037s :  0.12%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000609s :  2.05%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000019s :  0.07%
    cconv                                :     0.000016s :  0.05%
    loop_unroll                          :     0.000424s :  1.43%
    jit_opt_after_cconv.c_1              :     0.000030s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.cse              :     0.000025s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.02%
    remove_dup_value                     :     0.000011s :  0.04%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000005s :  0.02%
    add_recomputation                    :     0.000041s :  0.14%
    cse_after_recomputation.cse          :     0.000017s :  0.06%
    auto_monad_reorder                   :     0.000015s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000480s :  1.62%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000017s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000012s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000127s :  0.43%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008406s : 28.34%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000533   118
 5.44% :     0.000029s :      1: substitution.arithmetic_simplify
 0.39% :     0.000002s :      4: substitution.elim_not_effective
 0.29% :     0.000002s :      4: substitution.fold_const_symbol
 0.93% :     0.000005s :      5: substitution.graph_param_transform
60.52% :     0.000322s :     12: substitution.inline
 4.09% :     0.000022s :      2: substitution.inline_without_move
 1.78% :     0.000009s :     16: substitution.j_node_and_user_rematch
 1.92% :     0.000010s :      7: substitution.minmaximum_grad
 1.38% :     0.000007s :      5: substitution.partial_eliminate
 2.19% :     0.000012s :     16: substitution.remove_not_recompute_node
 4.75% :     0.000025s :     10: substitution.replace_applicator
 1.58% :     0.000008s :      7: substitution.replace_old_param
 0.62% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.10% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.65% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.35% :     0.000039s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010194     2
86.87% :     0.008856s :      1: type_inference.infer
13.13% :     0.001338s :      1: type_inference.specialize
------[replace.]   0.000150    20
 3.62% :     0.000005s :      1: replace.arithmetic_simplify
62.17% :     0.000093s :     12: replace.inline
34.20% :     0.000051s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000365    20
 7.66% :     0.000028s :      1: match.arithmetic_simplify
86.55% :     0.000316s :     12: match.inline
 5.78% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000400  2790
 1.50% :     0.000006s :     46: predicate.accumulaten_eliminater
 0.52% :     0.000002s :      5: predicate.ad_related_special_op_eliminate
 1.49% :     0.000006s :     46: predicate.addn_check_dump
 1.74% :     0.000007s :     46: predicate.addn_zero_filter
 2.05% :     0.000008s :     47: predicate.arithmetic_simplify
 1.74% :     0.000007s :     47: predicate.cast_eliminate
 0.20% :     0.000001s :      5: predicate.check_bprop_eliminate
 1.40% :     0.000006s :     46: predicate.compare_switch_simplify
 1.56% :     0.000006s :     46: predicate.depend_value_elim
 1.52% :     0.000006s :     47: predicate.dict_get_item_const_eliminator
 1.52% :     0.000006s :     47: predicate.dict_get_item_eliminator
 1.48% :     0.000006s :     47: predicate.dict_set_item_eliminator
 0.31% :     0.000001s :      5: predicate.dumpgradient_eliminate
 0.19% :     0.000001s :      5: predicate.elim_not_effective
 0.24% :     0.000001s :      5: predicate.elim_shapecalc_of_broadcastargs
 1.49% :     0.000006s :     47: predicate.environ_add_const_eliminate
 1.42% :     0.000006s :     47: predicate.environ_get_add_eliminate
 1.51% :     0.000006s :     47: predicate.environ_get_depend_swap
 1.59% :     0.000006s :     47: predicate.environ_get_eliminate
 1.48% :     0.000006s :     47: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      5: predicate.fold_const_symbol
 0.94% :     0.000004s :     21: predicate.get_grad_eliminate
 0.10% :     0.000000s :      5: predicate.graph_param_transform
 4.09% :     0.000016s :     76: predicate.inline
 1.66% :     0.000007s :     42: predicate.inline_without_move
 0.36% :     0.000001s :     21: predicate.j_node_and_user_rematch
 1.08% :     0.000004s :     21: predicate.less_batch_normalization
 1.81% :     0.000007s :     54: predicate.list_to_tuple_eliminator_
 1.96% :     0.000008s :     59: predicate.load_eliminater
 0.54% :     0.000002s :      5: predicate.loop_unroll_after_grad
 3.31% :     0.000013s :     89: predicate.loop_unroll_before_grad
 1.94% :     0.000008s :     52: predicate.make_slice_get_slice_eliminator
 1.45% :     0.000006s :     46: predicate.merge_addn
 1.52% :     0.000006s :     47: predicate.minmaximum_grad
 0.52% :     0.000002s :      5: predicate.mutable_eliminate
 0.23% :     0.000001s :      5: predicate.opt_reshape
 2.40% :     0.000010s :     59: predicate.partial_eliminate
 1.44% :     0.000006s :     46: predicate.print_const_string_wrapper
 1.95% :     0.000008s :     47: predicate.reduce_eliminate
 1.86% :     0.000007s :     54: predicate.redundant_stop_gradient_eliminater
 0.44% :     0.000002s :     21: predicate.remove_not_recompute_node
 2.54% :     0.000010s :    105: predicate.replace_applicator
 1.07% :     0.000004s :     42: predicate.replace_old_param
 0.12% :     0.000000s :      5: predicate.reset_defer_inline
 1.70% :     0.000007s :     47: predicate.reshape_eliminate
 1.51% :     0.000006s :     46: predicate.row_tensor_add_zeros_like
 0.29% :     0.000001s :      5: predicate.row_tensor_eliminate
 1.50% :     0.000006s :     46: predicate.same_eliminate
 0.46% :     0.000002s :     21: predicate.set_cell_output_no_recompute
 0.48% :     0.000002s :     10: predicate.special_op_eliminate
 0.85% :     0.000003s :     21: predicate.specialize_transform
 1.64% :     0.000007s :     46: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000006s :     46: predicate.stack_unstack_eliminate
 0.19% :     0.000001s :      5: predicate.switch_call_monad_eliminater
 2.74% :     0.000011s :     66: predicate.switch_defer_inline
 2.39% :     0.000010s :     66: predicate.switch_layer_defer_inline
 6.13% :     0.000025s :    160: predicate.switch_simplify
 1.50% :     0.000006s :     47: predicate.tile_eliminate
 1.53% :     0.000006s :     47: predicate.transpose_eliminate
 1.82% :     0.000007s :     47: predicate.tuple_list_convert_item_index_to_positive
 1.69% :     0.000007s :     47: predicate.tuple_list_get_item_depend_reorder
 3.02% :     0.000012s :     64: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000007s :     47: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000007s :     54: predicate.tuple_to_list_eliminator_
 1.90% :     0.000008s :     59: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000012s :     80: predicate.updatestate_useless_node_eliminater
 2.00% :     0.000008s :     46: predicate.value_based_eliminate
 0.17% :     0.000001s :      5: predicate.virtual_view_grad_eliminate
 0.23% :     0.000001s :      5: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001549    29
58.41% :     0.000905s :     13: func_graph_cloner_run.FuncGraphClonerGraph
41.59% :     0.000644s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.040667    91
 0.11% :     0.000044s :      1: add_recomputation
 0.25% :     0.000103s :      1: auto_monad
 0.04% :     0.000018s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.53% :     0.000216s :      1: bootstrap
 0.05% :     0.000019s :      1: cconv
 0.02% :     0.000010s :      1: convert_after_rewriter
 0.08% :     0.000031s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.15% :     0.000061s :      1: event_method
 0.03% :     0.000012s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000003s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
31.01% :     0.012609s :      1: jit_opt_a
 0.40% :     0.000164s :      1: jit_opt_after_cconv
 0.15% :     0.000061s :      1: jit_opt_b
 1.06% :     0.000432s :      1: loop_unroll
 1.51% :     0.000615s :      1: mutable_eliminate
 6.89% :     0.002802s :     39: opt.transform.jit_opt_a
 0.15% :     0.000061s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000034s :      4: opt.transform.jit_opt_b
 0.04% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000043s :      4: opt.transform.symbol_engine_opt
 1.20% :     0.000487s :      1: opt_after_jit_grad
 0.02% :     0.000007s :      1: order_py_execute_after_rewriter
 0.01% :     0.000003s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000030s :      1: py_interpret_to_execute
 0.03% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.03% :     0.000013s :      1: remove_dup_value
 5.06% :     0.002057s :      2: renormalize.infer
 3.79% :     0.001541s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000040s :      1: rewriter_after_opt_a
 0.23% :     0.000096s :      1: rewriter_before_opt_a
 0.22% :     0.000089s :      1: symbol_engine_optimizer
20.70% :     0.008419s :      1: task_emit
25.29% :     0.010284s :      1: type_inference
 0.38% :     0.000153s :      1: validate

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.218.39 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e3310> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.224.15 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0225433, [33]
[bootstrap]: 0.00021436
[type_inference]: 0.010758
[event_method]: 5.406e-05
[auto_monad]: 0.00010555
[graph_reusing]: 8.00999e-06
[pre_auto_parallel]: 3.19001e-06
[py_interpret_to_execute]: 3.177e-05
[rewriter_before_opt_a]: 9.741e-05
[expand_dump_flag]: 3.33998e-06
[jit_opt_a]: 0.00907686, [2]
    [Cycle 1]: 0.00477755, [27]
        [switch_simplify]: 6.86e-05
        [loop_unroll]: 3.748e-05
        [a_1]: 0.00085385
        [with_stream_mark]: 2.891e-05
        [recompute_prepare]: 1.883e-05
        [updatestate_depend_eliminate]: 8.69e-06
        [updatestate_assign_eliminate]: 6.64001e-06
        [updatestate_loads_eliminate]: 6.73e-06
        [parameter_eliminate]: 2.89999e-06
        [specialize_transform]: 1.303e-05
        [updatestate_useless_node_eliminater]: 1.158e-05
        [accelerated_algorithm]: 1.244e-05
        [meta_shard_fg_expand]: 3.96001e-06
        [get_grad_eliminate_]: 1.274e-05
        [merge_forward]: 8.60001e-06
        [cell_reuse_recompute_pass]: 1.09998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.673e-05
        [j_node_and_user_rematch]: 2.298e-05
        [meta_fg_expand]: 0.00156603
        [replace_old_param]: 4.936e-05
        [inline_without_move]: 4.91e-05
        [renormalize]: 0.00169954
        [add_forward_monad_depend]: 5.37999e-06
        [auto_monad_grad]: 2.51e-06
        [auto_monad_eliminator]: 1.241e-05
        [cse]: 2.634e-05
        [replace_applicator]: 1.264e-05
    [Cycle 2]: 0.0002929, [27]
        [switch_simplify]: 4.23999e-06
        [loop_unroll]: 3.83001e-06
        [a_1]: 4.228e-05
        [with_stream_mark]: 9.66e-06
        [recompute_prepare]: 3.91999e-06
        [updatestate_depend_eliminate]: 1.94e-06
        [updatestate_assign_eliminate]: 1.60999e-06
        [updatestate_loads_eliminate]: 1.69e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 3.83999e-06
        [updatestate_useless_node_eliminater]: 3.48999e-06
        [accelerated_algorithm]: 3.84002e-06
        [meta_shard_fg_expand]: 1.00999e-06
        [get_grad_eliminate_]: 3.69002e-06
        [merge_forward]: 2.11998e-06
        [cell_reuse_recompute_pass]: 1.69e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.043e-05
        [j_node_and_user_rematch]: 6.14001e-06
        [meta_fg_expand]: 3.92e-05
        [replace_old_param]: 4.06001e-06
        [inline_without_move]: 3.25e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.27e-06
        [auto_monad_grad]: 1.12999e-06
        [auto_monad_eliminator]: 4.48999e-06
        [cse]: 1.081e-05
        [replace_applicator]: 3.58999e-06
[py_interpret_to_execute_after_opt_a]: 9.91998e-06
[rewriter_after_opt_a]: 3.295e-05
[convert_after_rewriter]: 6.15002e-06
[order_py_execute_after_rewriter]: 4.53999e-06
[mutable_eliminate]: 0.0006679
[jit_opt_b]: 4.122e-05, [1]
    [Cycle 1]: 3.392e-05, [2]
        [frontend_op_eliminate]: 1.11e-05
        [inline_after_opt_a]: 1.162e-05
[cconv]: 2.129e-05
[loop_unroll]: 0.00043031
[jit_opt_after_cconv]: 0.00010959, [1]
    [Cycle 1]: 0.00010327, [11]
        [c_1]: 1.124e-05
        [parameter_eliminate]: 1.80001e-06
        [updatestate_depend_eliminate]: 3.73001e-06
        [updatestate_assign_eliminate]: 1.64998e-06
        [updatestate_loads_eliminate]: 1.49e-06
        [cse]: 1.348e-05
        [call_graph_tuple_transform]: 1.199e-05
        [tuple_list_get_item_eliminator]: 3.46999e-06
        [none_parameter_eliminate]: 1.35001e-06
        [renormalize]: 7.00005e-07
        [switch_simplify]: 3.6e-06
[remove_dup_value]: 1.508e-05
[partial_unused_args_eliminate]: 1.99e-06
[environ_conv]: 5.12e-06
[add_recomputation]: 3.499e-05
[cse_after_recomputation]: 1.906e-05, [1]
    [Cycle 1]: 1.346e-05, [1]
        [cse]: 7.19001e-06
[auto_monad_reorder]: 1.312e-05
[get_jit_bprop_graph]: 2.51e-06
[rewriter_after_jit_bprop_graph]: 5.35999e-06
[opt_after_jit_grad]: 0.00046464
[symbol_engine_optimizer]: 6.637e-05, [1]
    [Cycle 1]: 6.016e-05, [6]
        [build]: 3.36999e-06
        [elim_shapecalc]: 5.54e-06
        [elim_not_effective]: 1.021e-05
        [opt_reshape]: 4.58999e-06
        [fold_const_symbol]: 6.42001e-06
        [renormalize]: 6.59988e-07
[validate]: 2.897e-05
[backend_pass]: 9.20001e-07
[task_emit]: 2.108e-05
[execute]: 1.26997e-06
Sums
    bootstrap                            :     0.000214s :  1.20%
    type_inference                       :     0.010758s : 60.09%
    event_method                         :     0.000054s :  0.30%
    auto_monad                           :     0.000106s :  0.59%
    graph_reusing                        :     0.000008s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000032s :  0.18%
    rewriter_before_opt_a                :     0.000097s :  0.54%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000073s :  0.41%
    jit_opt_a.loop_unroll                :     0.000041s :  0.23%
    jit_opt_a.a_1                        :     0.000896s :  5.01%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000023s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.09%
    jit_opt_a.merge_forward              :     0.000011s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001605s :  8.97%
    jit_opt_a.replace_old_param          :     0.000053s :  0.30%
    jit_opt_a.inline_without_move        :     0.000052s :  0.29%
    jit_opt_a.renormalize                :     0.001700s :  9.49%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.09%
    jit_opt_a.cse                        :     0.000037s :  0.21%
    jit_opt_a.replace_applicator         :     0.000016s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000033s :  0.18%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000668s :  3.73%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.06%
    cconv                                :     0.000021s :  0.12%
    loop_unroll                          :     0.000430s :  2.40%
    jit_opt_after_cconv.c_1              :     0.000011s :  0.06%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.cse              :     0.000013s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000012s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000003s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000015s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000035s :  0.20%
    cse_after_recomputation.cse          :     0.000007s :  0.04%
    auto_monad_reorder                   :     0.000013s :  0.07%
    get_jit_bprop_graph                  :     0.000003s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000465s :  2.60%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000029s :  0.16%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.000021s :  0.12%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000267    50
 0.61% :     0.000002s :      1: substitution.elim_not_effective
 0.38% :     0.000001s :      1: substitution.fold_const_symbol
 1.40% :     0.000004s :      1: substitution.graph_param_transform
67.78% :     0.000181s :      6: substitution.inline
 5.79% :     0.000015s :      2: substitution.inline_without_move
 2.48% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.35% :     0.000004s :      2: substitution.minmaximum_grad
 1.80% :     0.000005s :      3: substitution.partial_eliminate
 2.50% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.24% :     0.000006s :      2: substitution.replace_applicator
 1.81% :     0.000005s :      5: substitution.replace_old_param
 1.14% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.92% :     0.000008s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.86% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.93% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010679     2
88.64% :     0.009466s :      1: type_inference.infer
11.36% :     0.001213s :      1: type_inference.specialize
------[replace.]   0.000066     8
76.56% :     0.000051s :      6: replace.inline
23.44% :     0.000016s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000189     8
94.01% :     0.000178s :      6: match.inline
 5.99% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000171  1109
 1.75% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.63% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.43% :     0.000002s :     18: predicate.addn_check_dump
 1.69% :     0.000003s :     18: predicate.addn_zero_filter
 2.17% :     0.000004s :     18: predicate.arithmetic_simplify
 1.61% :     0.000003s :     18: predicate.cast_eliminate
 0.22% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.43% :     0.000002s :     18: predicate.compare_switch_simplify
 1.41% :     0.000002s :     18: predicate.depend_value_elim
 1.25% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.54% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.40% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.54% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.10% :     0.000000s :      1: predicate.elim_not_effective
 0.25% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.36% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.27% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.48% :     0.000003s :     18: predicate.environ_get_eliminate
 1.27% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.34% :     0.000002s :     12: predicate.get_grad_eliminate
 0.11% :     0.000000s :      1: predicate.graph_param_transform
 4.47% :     0.000008s :     28: predicate.inline
 3.07% :     0.000005s :     33: predicate.inline_without_move
 0.50% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.58% :     0.000003s :     12: predicate.less_batch_normalization
 1.52% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.69% :     0.000003s :     21: predicate.load_eliminater
 0.54% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.59% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.82% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.33% :     0.000002s :     18: predicate.merge_addn
 1.31% :     0.000002s :     18: predicate.minmaximum_grad
 0.88% :     0.000001s :      1: predicate.mutable_eliminate
 0.30% :     0.000001s :      1: predicate.opt_reshape
 2.06% :     0.000004s :     21: predicate.partial_eliminate
 1.38% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.70% :     0.000003s :     18: predicate.reduce_eliminate
 1.68% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.70% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.47% :     0.000003s :     22: predicate.replace_applicator
 1.73% :     0.000003s :     33: predicate.replace_old_param
 0.09% :     0.000000s :      1: predicate.reset_defer_inline
 1.52% :     0.000003s :     18: predicate.reshape_eliminate
 1.36% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.53% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     18: predicate.same_eliminate
 0.60% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.50% :     0.000001s :      2: predicate.special_op_eliminate
 1.12% :     0.000002s :     12: predicate.specialize_transform
 1.52% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.52% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.64% :     0.000005s :     26: predicate.switch_defer_inline
 2.30% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.66% :     0.000011s :     71: predicate.switch_simplify
 1.40% :     0.000002s :     18: predicate.tile_eliminate
 1.31% :     0.000002s :     18: predicate.transpose_eliminate
 1.59% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.49% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.23% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.78% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.56% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.80% :     0.000003s :     18: predicate.value_based_eliminate
 0.12% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.23% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001197    22
72.09% :     0.000863s :     13: func_graph_cloner_run.FuncGraphClonerGraph
27.91% :     0.000334s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025553    76
 0.15% :     0.000038s :      1: add_recomputation
 0.43% :     0.000111s :      1: auto_monad
 0.06% :     0.000016s :      1: auto_monad_reorder
 0.02% :     0.000005s :      1: backend_pass
 0.89% :     0.000228s :      1: bootstrap
 0.09% :     0.000024s :      1: cconv
 0.03% :     0.000009s :      1: convert_after_rewriter
 0.08% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.23% :     0.000059s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000011s :      1: graph_reusing
35.54% :     0.009081s :      1: jit_opt_a
 0.44% :     0.000112s :      1: jit_opt_after_cconv
 0.17% :     0.000044s :      1: jit_opt_b
 1.71% :     0.000437s :      1: loop_unroll
 2.65% :     0.000677s :      1: mutable_eliminate
 4.91% :     0.001255s :     26: opt.transform.jit_opt_a
 0.11% :     0.000027s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000016s :      4: opt.transform.jit_opt_b
 0.03% :     0.000008s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000023s :      4: opt.transform.symbol_engine_opt
 1.85% :     0.000473s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000034s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 5.00% :     0.001277s :      1: renormalize.infer
 1.62% :     0.000414s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000036s :      1: rewriter_after_opt_a
 0.40% :     0.000102s :      1: rewriter_before_opt_a
 0.27% :     0.000069s :      1: symbol_engine_optimizer
 0.10% :     0.000025s :      1: task_emit
42.17% :     0.010776s :      1: type_inference
 0.18% :     0.000045s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:22.306.85 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c368d30> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:22.313.54 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.020266, [30]
[bootstrap]: 0.00030406
[type_inference]: 0.00984962
[event_method]: 4.467e-05
[auto_monad]: 9.471e-05
[graph_reusing]: 6.89999e-06
[pre_auto_parallel]: 2.51e-06
[py_interpret_to_execute]: 2.542e-05
[rewriter_before_opt_a]: 8.769e-05
[expand_dump_flag]: 3.85e-06
[jit_opt_a]: 0.00775802, [2]
    [Cycle 1]: 0.00428121, [27]
        [switch_simplify]: 5.847e-05
        [loop_unroll]: 3.849e-05
        [a_1]: 0.00081898
        [with_stream_mark]: 2.158e-05
        [recompute_prepare]: 1.776e-05
        [updatestate_depend_eliminate]: 7.56999e-06
        [updatestate_assign_eliminate]: 5.95002e-06
        [updatestate_loads_eliminate]: 5.37999e-06
        [parameter_eliminate]: 2.69999e-06
        [specialize_transform]: 1.301e-05
        [updatestate_useless_node_eliminater]: 1.193e-05
        [accelerated_algorithm]: 1.22e-05
        [meta_shard_fg_expand]: 3.46001e-06
        [get_grad_eliminate_]: 1.166e-05
        [merge_forward]: 8.18999e-06
        [cell_reuse_recompute_pass]: 7.7e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.448e-05
        [j_node_and_user_rematch]: 2.182e-05
        [meta_fg_expand]: 0.00140912
        [replace_old_param]: 5.248e-05
        [inline_without_move]: 5.179e-05
        [renormalize]: 0.00143715
        [add_forward_monad_depend]: 5.42999e-06
        [auto_monad_grad]: 2.54999e-06
        [auto_monad_eliminator]: 1.185e-05
        [cse]: 1.937e-05
        [replace_applicator]: 1.05e-05
    [Cycle 2]: 0.00028276, [27]
        [switch_simplify]: 5.10999e-06
        [loop_unroll]: 4.18001e-06
        [a_1]: 4.393e-05
        [with_stream_mark]: 9.04998e-06
        [recompute_prepare]: 3.93001e-06
        [updatestate_depend_eliminate]: 2.39001e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 1.69e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 3.85e-06
        [updatestate_useless_node_eliminater]: 3.38e-06
        [accelerated_algorithm]: 3.52997e-06
        [meta_shard_fg_expand]: 1.30999e-06
        [get_grad_eliminate_]: 3.71001e-06
        [merge_forward]: 1.96e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.55001e-06
        [j_node_and_user_rematch]: 5.87999e-06
        [meta_fg_expand]: 2.547e-05
        [replace_old_param]: 4.41002e-06
        [inline_without_move]: 3.48e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 6.39993e-07
        [auto_monad_eliminator]: 4.18999e-06
        [cse]: 9.40001e-06
        [replace_applicator]: 3.95e-06
[py_interpret_to_execute_after_opt_a]: 9.92999e-06
[rewriter_after_opt_a]: 3.465e-05
[convert_after_rewriter]: 6.19999e-06
[order_py_execute_after_rewriter]: 4.55001e-06
[mutable_eliminate]: 0.00056994
[jit_opt_b]: 4.234e-05, [1]
    [Cycle 1]: 3.453e-05, [2]
        [frontend_op_eliminate]: 1.087e-05
        [inline_after_opt_a]: 1.189e-05
[cconv]: 2.112e-05
[loop_unroll]: 0.00044751
[jit_opt_after_cconv]: 0.0001155, [1]
    [Cycle 1]: 0.00010908, [11]
        [c_1]: 1.128e-05
        [parameter_eliminate]: 2.14e-06
        [updatestate_depend_eliminate]: 3.75e-06
        [updatestate_assign_eliminate]: 1.92001e-06
        [updatestate_loads_eliminate]: 1.61002e-06
        [cse]: 1.651e-05
        [call_graph_tuple_transform]: 1.227e-05
        [tuple_list_get_item_eliminator]: 3.90998e-06
        [none_parameter_eliminate]: 1.24e-06
        [renormalize]: 8.89995e-07
        [switch_simplify]: 4.03001e-06
[remove_dup_value]: 1.417e-05
[partial_unused_args_eliminate]: 2.05002e-06
[environ_conv]: 4.38999e-06
[add_recomputation]: 2.991e-05
[cse_after_recomputation]: 2.023e-05, [1]
    [Cycle 1]: 1.457e-05, [1]
        [cse]: 8.36002e-06
[auto_monad_reorder]: 1.269e-05
[get_jit_bprop_graph]: 1.37999e-06
[rewriter_after_jit_bprop_graph]: 4.92999e-06
[opt_after_jit_grad]: 0.00049334
[symbol_engine_optimizer]: 6.339e-05, [1]
    [Cycle 1]: 5.754e-05, [6]
        [build]: 3.31001e-06
        [elim_shapecalc]: 6.09999e-06
        [elim_not_effective]: 9.42001e-06
        [opt_reshape]: 4.07e-06
        [fold_const_symbol]: 5.72999e-06
        [renormalize]: 4.30009e-07
[validate]: 2.747e-05
Sums
    bootstrap                            :     0.000304s :  1.85%
    type_inference                       :     0.009850s : 59.81%
    event_method                         :     0.000045s :  0.27%
    auto_monad                           :     0.000095s :  0.58%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000025s :  0.15%
    rewriter_before_opt_a                :     0.000088s :  0.53%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.39%
    jit_opt_a.loop_unroll                :     0.000043s :  0.26%
    jit_opt_a.a_1                        :     0.000863s :  5.24%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.09%
    jit_opt_a.merge_forward              :     0.000010s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001435s :  8.71%
    jit_opt_a.replace_old_param          :     0.000057s :  0.35%
    jit_opt_a.inline_without_move        :     0.000055s :  0.34%
    jit_opt_a.renormalize                :     0.001437s :  8.73%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.10%
    jit_opt_a.cse                        :     0.000029s :  0.17%
    jit_opt_a.replace_applicator         :     0.000014s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000035s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000570s :  3.46%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.07%
    cconv                                :     0.000021s :  0.13%
    loop_unroll                          :     0.000448s :  2.72%
    jit_opt_after_cconv.c_1              :     0.000011s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000012s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000030s :  0.18%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.08%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000493s :  3.00%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.17%


Time group info:
------[substitution.]   0.000240    50
 0.60% :     0.000001s :      1: substitution.elim_not_effective
 0.40% :     0.000001s :      1: substitution.fold_const_symbol
 1.60% :     0.000004s :      1: substitution.graph_param_transform
65.49% :     0.000157s :      6: substitution.inline
 7.16% :     0.000017s :      2: substitution.inline_without_move
 2.53% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.37% :     0.000003s :      2: substitution.minmaximum_grad
 1.76% :     0.000004s :      3: substitution.partial_eliminate
 2.94% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.31% :     0.000006s :      2: substitution.replace_applicator
 1.78% :     0.000004s :      5: substitution.replace_old_param
 0.98% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 3.05% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.82% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 6.20% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009780     2
89.34% :     0.008737s :      1: type_inference.infer
10.66% :     0.001043s :      1: type_inference.specialize
------[replace.]   0.000063     8
77.70% :     0.000049s :      6: replace.inline
22.30% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000164     8
93.70% :     0.000154s :      6: match.inline
 6.30% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000169  1109
 1.40% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.84% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.27% :     0.000002s :     18: predicate.addn_check_dump
 1.55% :     0.000003s :     18: predicate.addn_zero_filter
 2.28% :     0.000004s :     18: predicate.arithmetic_simplify
 1.53% :     0.000003s :     18: predicate.cast_eliminate
 0.15% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.27% :     0.000002s :     18: predicate.compare_switch_simplify
 1.53% :     0.000003s :     18: predicate.depend_value_elim
 1.31% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.47% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.40% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.47% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      1: predicate.elim_not_effective
 0.41% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.45% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.41% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.40% :     0.000002s :     18: predicate.environ_get_eliminate
 1.32% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.07% :     0.000000s :      1: predicate.fold_const_symbol
 1.21% :     0.000002s :     12: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 4.09% :     0.000007s :     28: predicate.inline
 3.22% :     0.000005s :     33: predicate.inline_without_move
 0.51% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.57% :     0.000003s :     12: predicate.less_batch_normalization
 1.77% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.70% :     0.000003s :     21: predicate.load_eliminater
 0.78% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.60% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.68% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.31% :     0.000002s :     18: predicate.merge_addn
 1.28% :     0.000002s :     18: predicate.minmaximum_grad
 0.82% :     0.000001s :      1: predicate.mutable_eliminate
 0.20% :     0.000000s :      1: predicate.opt_reshape
 2.19% :     0.000004s :     21: predicate.partial_eliminate
 1.35% :     0.000002s :     18: predicate.print_const_string_wrapper
 2.04% :     0.000003s :     18: predicate.reduce_eliminate
 1.58% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.58% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.48% :     0.000003s :     22: predicate.replace_applicator
 1.54% :     0.000003s :     33: predicate.replace_old_param
 0.15% :     0.000000s :      1: predicate.reset_defer_inline
 1.54% :     0.000003s :     18: predicate.reshape_eliminate
 1.76% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.39% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     18: predicate.same_eliminate
 0.57% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.49% :     0.000001s :      2: predicate.special_op_eliminate
 1.22% :     0.000002s :     12: predicate.specialize_transform
 1.64% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.70% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.11% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.54% :     0.000004s :     26: predicate.switch_defer_inline
 2.25% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.79% :     0.000011s :     71: predicate.switch_simplify
 1.51% :     0.000003s :     18: predicate.tile_eliminate
 1.38% :     0.000002s :     18: predicate.transpose_eliminate
 1.62% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.47% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 3.00% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.67% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.65% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000003s :     18: predicate.value_based_eliminate
 0.14% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.28% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001094    22
74.45% :     0.000815s :     13: func_graph_cloner_run.FuncGraphClonerGraph
25.55% :     0.000280s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022936    72
 0.14% :     0.000033s :      1: add_recomputation
 0.43% :     0.000099s :      1: auto_monad
 0.07% :     0.000015s :      1: auto_monad_reorder
 1.39% :     0.000319s :      1: bootstrap
 0.10% :     0.000024s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.21% :     0.000049s :      1: event_method
 0.03% :     0.000007s :      1: expand_dump_flag
 0.02% :     0.000003s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
33.84% :     0.007762s :      1: jit_opt_a
 0.51% :     0.000118s :      1: jit_opt_after_cconv
 0.20% :     0.000045s :      1: jit_opt_b
 1.98% :     0.000454s :      1: loop_unroll
 2.53% :     0.000579s :      1: mutable_eliminate
 5.29% :     0.001214s :     26: opt.transform.jit_opt_a
 0.12% :     0.000028s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000016s :      4: opt.transform.jit_opt_b
 0.04% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000022s :      4: opt.transform.symbol_engine_opt
 2.19% :     0.000503s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000028s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 4.79% :     0.001099s :      1: renormalize.infer
 1.44% :     0.000330s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000038s :      1: rewriter_after_opt_a
 0.40% :     0.000091s :      1: rewriter_before_opt_a
 0.29% :     0.000066s :      1: symbol_engine_optimizer
42.99% :     0.009861s :      1: type_inference


TotalTime = 0.0135961, [30]
[bootstrap]: 0.00030891
[type_inference]: 0.00460611
[event_method]: 1.869e-05
[auto_monad]: 6.918e-05
[graph_reusing]: 6.07999e-06
[pre_auto_parallel]: 3.03998e-06
[py_interpret_to_execute]: 3.318e-05
[rewriter_before_opt_a]: 7.591e-05
[expand_dump_flag]: 3.22002e-06
[jit_opt_a]: 0.00609281, [2]
    [Cycle 1]: 0.00189595, [27]
        [switch_simplify]: 6.661e-05
        [loop_unroll]: 3.076e-05
        [a_1]: 0.00059491
        [with_stream_mark]: 2.629e-05
        [recompute_prepare]: 7.75998e-06
        [updatestate_depend_eliminate]: 4.33001e-06
        [updatestate_assign_eliminate]: 3.46999e-06
        [updatestate_loads_eliminate]: 2.45002e-06
        [parameter_eliminate]: 2.17999e-06
        [specialize_transform]: 5.70001e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 5.98002e-06
        [meta_shard_fg_expand]: 1.92001e-06
        [get_grad_eliminate_]: 5.64e-06
        [merge_forward]: 4.05998e-06
        [cell_reuse_recompute_pass]: 1.37999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.511e-05
        [j_node_and_user_rematch]: 8.84003e-06
        [meta_fg_expand]: 2.69999e-06
        [replace_old_param]: 9.34998e-06
        [inline_without_move]: 5.40999e-06
        [renormalize]: 0.0008034
        [add_forward_monad_depend]: 7.09001e-06
        [auto_monad_grad]: 2.93e-06
        [auto_monad_eliminator]: 1.74e-05
        [cse]: 3.585e-05
        [replace_applicator]: 1.358e-05
    [Cycle 2]: 0.00032527, [27]
        [switch_simplify]: 5.50001e-06
        [loop_unroll]: 4.97999e-06
        [a_1]: 9.077e-05
        [with_stream_mark]: 1.089e-05
        [recompute_prepare]: 5.54e-06
        [updatestate_depend_eliminate]: 2.40002e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [parameter_eliminate]: 1.25999e-06
        [specialize_transform]: 5.10001e-06
        [updatestate_useless_node_eliminater]: 4.90999e-06
        [accelerated_algorithm]: 5.40999e-06
        [meta_shard_fg_expand]: 1.24e-06
        [get_grad_eliminate_]: 4.58001e-06
        [merge_forward]: 2.61999e-06
        [cell_reuse_recompute_pass]: 2.11e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.425e-05
        [j_node_and_user_rematch]: 8.17e-06
        [meta_fg_expand]: 1.49998e-06
        [replace_old_param]: 7.38e-06
        [inline_without_move]: 4.74e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.00001e-06
        [auto_monad_grad]: 9.00007e-07
        [auto_monad_eliminator]: 5.02999e-06
        [cse]: 1.096e-05
        [replace_applicator]: 4.98001e-06
[py_interpret_to_execute_after_opt_a]: 1.35e-05
[rewriter_after_opt_a]: 3.861e-05
[convert_after_rewriter]: 9.22001e-06
[order_py_execute_after_rewriter]: 5.15001e-06
[mutable_eliminate]: 0.00074497
[jit_opt_b]: 5.282e-05, [1]
    [Cycle 1]: 4.376e-05, [2]
        [frontend_op_eliminate]: 1.642e-05
        [inline_after_opt_a]: 1.493e-05
[cconv]: 3.249e-05
[loop_unroll]: 0.00043745
[jit_opt_after_cconv]: 0.00015041, [1]
    [Cycle 1]: 0.0001439, [11]
        [c_1]: 2.032e-05
        [parameter_eliminate]: 2.86999e-06
        [updatestate_depend_eliminate]: 7.22002e-06
        [updatestate_assign_eliminate]: 2.61999e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 2.463e-05
        [call_graph_tuple_transform]: 2.096e-05
        [tuple_list_get_item_eliminator]: 5.28002e-06
        [none_parameter_eliminate]: 1.82001e-06
        [renormalize]: 6.99976e-07
        [switch_simplify]: 5.15999e-06
[remove_dup_value]: 1.618e-05
[partial_unused_args_eliminate]: 2.49999e-06
[environ_conv]: 5.80002e-06
[add_recomputation]: 5.634e-05
[cse_after_recomputation]: 2.201e-05, [1]
    [Cycle 1]: 1.615e-05, [1]
        [cse]: 9.44e-06
[auto_monad_reorder]: 1.616e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 7.55998e-06
[opt_after_jit_grad]: 0.00046308
[symbol_engine_optimizer]: 7.265e-05, [1]
    [Cycle 1]: 6.615e-05, [6]
        [build]: 4.75999e-06
        [elim_shapecalc]: 7.38e-06
        [elim_not_effective]: 1.227e-05
        [opt_reshape]: 5.74999e-06
        [fold_const_symbol]: 8.3e-06
        [renormalize]: 6.30011e-07
[validate]: 4.051e-05
Sums
    bootstrap                            :     0.000309s :  3.40%
    type_inference                       :     0.004606s : 50.67%
    event_method                         :     0.000019s :  0.21%
    auto_monad                           :     0.000069s :  0.76%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.37%
    rewriter_before_opt_a                :     0.000076s :  0.84%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000072s :  0.79%
    jit_opt_a.loop_unroll                :     0.000036s :  0.39%
    jit_opt_a.a_1                        :     0.000686s :  7.54%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.41%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.18%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000803s :  8.84%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.25%
    jit_opt_a.cse                        :     0.000047s :  0.51%
    jit_opt_a.replace_applicator         :     0.000019s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.15%
    rewriter_after_opt_a                 :     0.000039s :  0.42%
    convert_after_rewriter               :     0.000009s :  0.10%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000745s :  8.20%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000032s :  0.36%
    loop_unroll                          :     0.000437s :  4.81%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000025s :  0.27%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000016s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000056s :  0.62%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.08%
    opt_after_jit_grad                   :     0.000463s :  5.09%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.45%


Time group info:
------[substitution.]   0.000250    23
 0.76% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.31% :     0.000006s :      3: substitution.graph_param_transform
85.94% :     0.000215s :      4: substitution.inline
 1.34% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.96% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000005s :      2: substitution.replace_old_param
 5.18% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004528     2
81.10% :     0.003672s :      1: type_inference.infer
18.90% :     0.000856s :      1: type_inference.specialize
------[replace.]   0.000070     6
69.01% :     0.000048s :      4: replace.inline
30.99% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000224     6
94.70% :     0.000212s :      4: match.inline
 5.30% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.68% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.57% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 2.17% :     0.000003s :     10: predicate.arithmetic_simplify
 1.16% :     0.000001s :     10: predicate.cast_eliminate
 0.56% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.72% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.37% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.84% :     0.000001s :      6: predicate.get_grad_eliminate
 0.66% :     0.000001s :      3: predicate.graph_param_transform
 5.55% :     0.000007s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.37% :     0.000002s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.59% :     0.000002s :     15: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.86% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 2.10% :     0.000003s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.33% :     0.000003s :     15: predicate.partial_eliminate
 1.30% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.86% :     0.000002s :     10: predicate.reduce_eliminate
 1.33% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.57% :     0.000002s :     10: predicate.reshape_eliminate
 1.10% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.64% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 1.11% :     0.000001s :      6: predicate.specialize_transform
 1.72% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.10% :     0.000003s :     16: predicate.switch_defer_inline
 1.87% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.73% :     0.000010s :     53: predicate.switch_simplify
 1.32% :     0.000002s :     10: predicate.tile_eliminate
 1.05% :     0.000001s :     10: predicate.transpose_eliminate
 1.53% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.00% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.76% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.04% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.57% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.84% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000609    12
49.58% :     0.000302s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.80% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.61% :     0.000253s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015354    72
 0.39% :     0.000060s :      1: add_recomputation
 0.47% :     0.000073s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 2.14% :     0.000329s :      1: bootstrap
 0.23% :     0.000035s :      1: cconv
 0.07% :     0.000011s :      1: convert_after_rewriter
 0.16% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.71% :     0.006097s :      1: jit_opt_a
 1.00% :     0.000153s :      1: jit_opt_after_cconv
 0.36% :     0.000056s :      1: jit_opt_b
 2.91% :     0.000447s :      1: loop_unroll
 4.92% :     0.000756s :      1: mutable_eliminate
 5.93% :     0.000910s :     26: opt.transform.jit_opt_a
 0.31% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.07% :     0.000472s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000036s :      1: py_interpret_to_execute
 0.10% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000019s :      1: remove_dup_value
 3.03% :     0.000464s :      1: renormalize.infer
 2.15% :     0.000330s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000042s :      1: rewriter_after_opt_a
 0.51% :     0.000079s :      1: rewriter_before_opt_a
 0.49% :     0.000075s :      1: symbol_engine_optimizer
30.13% :     0.004627s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.394.29 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d1ca0> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.399.59 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0202039, [33]
[bootstrap]: 0.00021396
[type_inference]: 0.0100322
[event_method]: 4.512e-05
[auto_monad]: 8.809e-05
[graph_reusing]: 6.72002e-06
[pre_auto_parallel]: 2.29001e-06
[py_interpret_to_execute]: 2.519e-05
[rewriter_before_opt_a]: 8.653e-05
[expand_dump_flag]: 3.45003e-06
[jit_opt_a]: 0.00781254, [2]
    [Cycle 1]: 0.00425937, [27]
        [switch_simplify]: 5.629e-05
        [loop_unroll]: 3.744e-05
        [a_1]: 0.00078191
        [with_stream_mark]: 1.972e-05
        [recompute_prepare]: 1.768e-05
        [updatestate_depend_eliminate]: 6.93998e-06
        [updatestate_assign_eliminate]: 5.35999e-06
        [updatestate_loads_eliminate]: 5.20999e-06
        [parameter_eliminate]: 1.78002e-06
        [specialize_transform]: 1.296e-05
        [updatestate_useless_node_eliminater]: 1.167e-05
        [accelerated_algorithm]: 1.198e-05
        [meta_shard_fg_expand]: 2.62001e-06
        [get_grad_eliminate_]: 1.19e-05
        [merge_forward]: 7.11001e-06
        [cell_reuse_recompute_pass]: 6.80011e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.322e-05
        [j_node_and_user_rematch]: 2.218e-05
        [meta_fg_expand]: 0.00134047
        [replace_old_param]: 4.937e-05
        [inline_without_move]: 4.988e-05
        [renormalize]: 0.00154886
        [add_forward_monad_depend]: 4.85001e-06
        [auto_monad_grad]: 1.50999e-06
        [auto_monad_eliminator]: 1.035e-05
        [cse]: 1.599e-05
        [replace_applicator]: 1.021e-05
    [Cycle 2]: 0.00027371, [27]
        [switch_simplify]: 4.43001e-06
        [loop_unroll]: 3.63e-06
        [a_1]: 3.772e-05
        [with_stream_mark]: 7.75e-06
        [recompute_prepare]: 3.7e-06
        [updatestate_depend_eliminate]: 2.23002e-06
        [updatestate_assign_eliminate]: 1.59e-06
        [updatestate_loads_eliminate]: 1.70001e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 3.95e-06
        [updatestate_useless_node_eliminater]: 3.48999e-06
        [accelerated_algorithm]: 5.66e-06
        [meta_shard_fg_expand]: 9.89996e-07
        [get_grad_eliminate_]: 3.58e-06
        [merge_forward]: 2.02001e-06
        [cell_reuse_recompute_pass]: 1.71002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.56e-06
        [j_node_and_user_rematch]: 6.07001e-06
        [meta_fg_expand]: 2.918e-05
        [replace_old_param]: 3.8e-06
        [inline_without_move]: 3.11999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.45001e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 4.68001e-06
        [cse]: 1.064e-05
        [replace_applicator]: 3.70998e-06
[py_interpret_to_execute_after_opt_a]: 7.43e-06
[rewriter_after_opt_a]: 2.763e-05
[convert_after_rewriter]: 5.57001e-06
[order_py_execute_after_rewriter]: 4.00998e-06
[mutable_eliminate]: 0.00049232
[jit_opt_b]: 3.971e-05, [1]
    [Cycle 1]: 3.259e-05, [2]
        [frontend_op_eliminate]: 1.039e-05
        [inline_after_opt_a]: 1.115e-05
[cconv]: 1.745e-05
[loop_unroll]: 0.00040164
[jit_opt_after_cconv]: 0.00011246, [1]
    [Cycle 1]: 0.00010562, [11]
        [c_1]: 1.162e-05
        [parameter_eliminate]: 2.28998e-06
        [updatestate_depend_eliminate]: 4.48999e-06
        [updatestate_assign_eliminate]: 1.89999e-06
        [updatestate_loads_eliminate]: 1.64e-06
        [cse]: 1.49e-05
        [call_graph_tuple_transform]: 1.164e-05
        [tuple_list_get_item_eliminator]: 3.69002e-06
        [none_parameter_eliminate]: 1.06997e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 3.79002e-06
[remove_dup_value]: 9.46998e-06
[partial_unused_args_eliminate]: 1.77999e-06
[environ_conv]: 4.2e-06
[add_recomputation]: 2.628e-05
[cse_after_recomputation]: 1.912e-05, [1]
    [Cycle 1]: 1.39e-05, [1]
        [cse]: 8.09002e-06
[auto_monad_reorder]: 9.89999e-06
[get_jit_bprop_graph]: 1.27e-06
[rewriter_after_jit_bprop_graph]: 4.33001e-06
[opt_after_jit_grad]: 0.0004338
[symbol_engine_optimizer]: 6.094e-05, [1]
    [Cycle 1]: 5.504e-05, [6]
        [build]: 2.99999e-06
        [elim_shapecalc]: 5.69e-06
        [elim_not_effective]: 9.13002e-06
        [opt_reshape]: 3.71001e-06
        [fold_const_symbol]: 5.68002e-06
        [renormalize]: 3.69997e-07
[validate]: 2.304e-05
[backend_pass]: 1.00999e-06
[task_emit]: 1.791e-05
[execute]: 1.22e-06
Sums
    bootstrap                            :     0.000214s :  1.31%
    type_inference                       :     0.010032s : 61.42%
    event_method                         :     0.000045s :  0.28%
    auto_monad                           :     0.000088s :  0.54%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000025s :  0.15%
    rewriter_before_opt_a                :     0.000087s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.37%
    jit_opt_a.loop_unroll                :     0.000041s :  0.25%
    jit_opt_a.a_1                        :     0.000820s :  5.02%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.09%
    jit_opt_a.merge_forward              :     0.000009s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001370s :  8.38%
    jit_opt_a.replace_old_param          :     0.000053s :  0.33%
    jit_opt_a.inline_without_move        :     0.000053s :  0.32%
    jit_opt_a.renormalize                :     0.001549s :  9.48%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.09%
    jit_opt_a.cse                        :     0.000027s :  0.16%
    jit_opt_a.replace_applicator         :     0.000014s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.05%
    rewriter_after_opt_a                 :     0.000028s :  0.17%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000492s :  3.01%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000017s :  0.11%
    loop_unroll                          :     0.000402s :  2.46%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000012s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000009s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000026s :  0.16%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000010s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000434s :  2.66%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000023s :  0.14%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.000018s :  0.11%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000214    50
 0.65% :     0.000001s :      1: substitution.elim_not_effective
 0.38% :     0.000001s :      1: substitution.fold_const_symbol
 1.25% :     0.000003s :      1: substitution.graph_param_transform
65.73% :     0.000141s :      6: substitution.inline
 7.17% :     0.000015s :      2: substitution.inline_without_move
 3.05% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.34% :     0.000003s :      2: substitution.minmaximum_grad
 1.59% :     0.000003s :      3: substitution.partial_eliminate
 3.07% :     0.000007s :      9: substitution.remove_not_recompute_node
 1.93% :     0.000004s :      2: substitution.replace_applicator
 1.78% :     0.000004s :      5: substitution.replace_old_param
 1.18% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.72% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.84% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 6.31% :     0.000014s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009968     2
89.45% :     0.008917s :      1: type_inference.infer
10.55% :     0.001051s :      1: type_inference.specialize
------[replace.]   0.000055     8
77.23% :     0.000043s :      6: replace.inline
22.77% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000146     8
93.90% :     0.000137s :      6: match.inline
 6.10% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000162  1109
 1.53% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.59% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.42% :     0.000002s :     18: predicate.addn_check_dump
 1.95% :     0.000003s :     18: predicate.addn_zero_filter
 2.02% :     0.000003s :     18: predicate.arithmetic_simplify
 1.61% :     0.000003s :     18: predicate.cast_eliminate
 0.20% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.33% :     0.000002s :     18: predicate.compare_switch_simplify
 1.53% :     0.000002s :     18: predicate.depend_value_elim
 1.32% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.51% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.34% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.53% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      1: predicate.elim_not_effective
 0.21% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.43% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.36% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.34% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.38% :     0.000002s :     18: predicate.environ_get_eliminate
 1.32% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.19% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.45% :     0.000007s :     28: predicate.inline
 3.20% :     0.000005s :     33: predicate.inline_without_move
 0.52% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.30% :     0.000002s :     12: predicate.less_batch_normalization
 1.58% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.81% :     0.000003s :     21: predicate.load_eliminater
 0.72% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.69% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.83% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.40% :     0.000002s :     18: predicate.merge_addn
 1.34% :     0.000002s :     18: predicate.minmaximum_grad
 0.75% :     0.000001s :      1: predicate.mutable_eliminate
 0.15% :     0.000000s :      1: predicate.opt_reshape
 2.21% :     0.000004s :     21: predicate.partial_eliminate
 1.42% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.76% :     0.000003s :     18: predicate.reduce_eliminate
 1.61% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.59% :     0.000003s :     22: predicate.replace_applicator
 1.60% :     0.000003s :     33: predicate.replace_old_param
 0.12% :     0.000000s :      1: predicate.reset_defer_inline
 1.47% :     0.000002s :     18: predicate.reshape_eliminate
 1.39% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.43% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.45% :     0.000002s :     18: predicate.same_eliminate
 0.68% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.34% :     0.000001s :      2: predicate.special_op_eliminate
 1.18% :     0.000002s :     12: predicate.specialize_transform
 1.64% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.59% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.15% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.63% :     0.000004s :     26: predicate.switch_defer_inline
 2.32% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.66% :     0.000011s :     71: predicate.switch_simplify
 1.35% :     0.000002s :     18: predicate.tile_eliminate
 1.40% :     0.000002s :     18: predicate.transpose_eliminate
 1.62% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.55% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 2.84% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.63% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.87% :     0.000003s :     18: predicate.value_based_eliminate
 0.15% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.30% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001083    22
73.95% :     0.000801s :     13: func_graph_cloner_run.FuncGraphClonerGraph
26.05% :     0.000282s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022972    76
 0.13% :     0.000029s :      1: add_recomputation
 0.40% :     0.000092s :      1: auto_monad
 0.05% :     0.000012s :      1: auto_monad_reorder
 0.02% :     0.000004s :      1: backend_pass
 0.98% :     0.000225s :      1: bootstrap
 0.09% :     0.000020s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.09% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.21% :     0.000049s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.03% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000003s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
34.02% :     0.007815s :      1: jit_opt_a
 0.50% :     0.000115s :      1: jit_opt_after_cconv
 0.18% :     0.000042s :      1: jit_opt_b
 1.78% :     0.000409s :      1: loop_unroll
 2.18% :     0.000500s :      1: mutable_eliminate
 5.06% :     0.001163s :     26: opt.transform.jit_opt_a
 0.12% :     0.000028s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000015s :      4: opt.transform.jit_opt_b
 0.04% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000014s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000021s :      4: opt.transform.symbol_engine_opt
 1.92% :     0.000442s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000028s :      1: py_interpret_to_execute
 0.04% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000012s :      1: remove_dup_value
 5.05% :     0.001159s :      1: renormalize.infer
 1.67% :     0.000383s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000031s :      1: rewriter_after_opt_a
 0.39% :     0.000089s :      1: rewriter_before_opt_a
 0.28% :     0.000063s :      1: symbol_engine_optimizer
 0.10% :     0.000022s :      1: task_emit
43.73% :     0.010045s :      1: type_inference
 0.16% :     0.000037s :      1: validate


TotalTime = 0.0323915, [33]
[bootstrap]: 0.00020269
[type_inference]: 0.00967821
[event_method]: 4.53e-05
[auto_monad]: 8.838e-05
[graph_reusing]: 6.63998e-06
[pre_auto_parallel]: 2.71e-06
[py_interpret_to_execute]: 2.198e-05
[rewriter_before_opt_a]: 8.51e-05
[expand_dump_flag]: 3.66999e-06
[jit_opt_a]: 0.0114867, [3]
    [Cycle 1]: 0.00571565, [27]
        [switch_simplify]: 5.837e-05
        [loop_unroll]: 3.767e-05
        [a_1]: 0.00082265
        [with_stream_mark]: 2.216e-05
        [recompute_prepare]: 1.725e-05
        [updatestate_depend_eliminate]: 7.65e-06
        [updatestate_assign_eliminate]: 5.96e-06
        [updatestate_loads_eliminate]: 5.32001e-06
        [parameter_eliminate]: 2.32999e-06
        [specialize_transform]: 1.334e-05
        [updatestate_useless_node_eliminater]: 1.166e-05
        [accelerated_algorithm]: 1.217e-05
        [meta_shard_fg_expand]: 3.20002e-06
        [get_grad_eliminate_]: 1.24e-05
        [merge_forward]: 7.63001e-06
        [cell_reuse_recompute_pass]: 9.09989e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.433e-05
        [j_node_and_user_rematch]: 2.225e-05
        [meta_fg_expand]: 0.00144229
        [replace_old_param]: 4.934e-05
        [inline_without_move]: 5.015e-05
        [renormalize]: 0.00253727
        [add_forward_monad_depend]: 9.71e-06
        [auto_monad_grad]: 5.57999e-06
        [auto_monad_eliminator]: 5.262e-05
        [cse]: 0.00022318
        [replace_applicator]: 6.585e-05
    [Cycle 2]: 0.00217012, [27]
        [switch_simplify]: 3.648e-05
        [loop_unroll]: 3.583e-05
        [a_1]: 0.00109994
        [with_stream_mark]: 1.374e-05
        [recompute_prepare]: 9.87999e-06
        [updatestate_depend_eliminate]: 4.79e-06
        [updatestate_assign_eliminate]: 4.42e-06
        [updatestate_loads_eliminate]: 4.40999e-06
        [parameter_eliminate]: 1.50999e-06
        [specialize_transform]: 8.48999e-06
        [updatestate_useless_node_eliminater]: 7.85998e-06
        [accelerated_algorithm]: 7.97e-06
        [meta_shard_fg_expand]: 2.44001e-06
        [get_grad_eliminate_]: 8.13999e-06
        [merge_forward]: 4.81002e-06
        [cell_reuse_recompute_pass]: 1.04003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.677e-05
        [j_node_and_user_rematch]: 1.374e-05
        [meta_fg_expand]: 4.072e-05
        [replace_old_param]: 1.133e-05
        [inline_without_move]: 8.15e-06
        [renormalize]: 0.00059809
        [add_forward_monad_depend]: 4.59002e-06
        [auto_monad_grad]: 1.24e-06
        [auto_monad_eliminator]: 1.571e-05
        [cse]: 4.722e-05
        [replace_applicator]: 1.576e-05
    [Cycle 3]: 0.00048536, [27]
        [switch_simplify]: 8.65999e-06
        [loop_unroll]: 7.89997e-06
        [a_1]: 0.00018993
        [with_stream_mark]: 1.046e-05
        [recompute_prepare]: 7.78001e-06
        [updatestate_depend_eliminate]: 4.38001e-06
        [updatestate_assign_eliminate]: 4.22e-06
        [updatestate_loads_eliminate]: 3.68e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 8.35001e-06
        [updatestate_useless_node_eliminater]: 7.81001e-06
        [accelerated_algorithm]: 8.90001e-06
        [meta_shard_fg_expand]: 2.15002e-06
        [get_grad_eliminate_]: 7.63999e-06
        [merge_forward]: 4.03001e-06
        [cell_reuse_recompute_pass]: 1.15999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.68e-05
        [j_node_and_user_rematch]: 1.401e-05
        [meta_fg_expand]: 2.66e-06
        [replace_old_param]: 1.032e-05
        [inline_without_move]: 7.72002e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.49e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 1.014e-05
        [cse]: 2.27e-05
        [replace_applicator]: 8.03001e-06
[py_interpret_to_execute_after_opt_a]: 1.148e-05
[rewriter_after_opt_a]: 4.595e-05
[convert_after_rewriter]: 8.60001e-06
[order_py_execute_after_rewriter]: 6.65998e-06
[mutable_eliminate]: 0.00057505
[jit_opt_b]: 6.609e-05, [1]
    [Cycle 1]: 5.816e-05, [2]
        [frontend_op_eliminate]: 2.39e-05
        [inline_after_opt_a]: 2.323e-05
[cconv]: 2.339e-05
[loop_unroll]: 0.0004316
[jit_opt_after_cconv]: 0.00018609, [1]
    [Cycle 1]: 0.00017901, [11]
        [c_1]: 3.421e-05
        [parameter_eliminate]: 2.74001e-06
        [updatestate_depend_eliminate]: 7.48e-06
        [updatestate_assign_eliminate]: 4.37998e-06
        [updatestate_loads_eliminate]: 4.42e-06
        [cse]: 3.114e-05
        [call_graph_tuple_transform]: 2.537e-05
        [tuple_list_get_item_eliminator]: 8.30999e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 5.10016e-07
        [switch_simplify]: 8.36002e-06
[remove_dup_value]: 2.204e-05
[partial_unused_args_eliminate]: 2.53e-06
[environ_conv]: 7.18998e-06
[add_recomputation]: 5.486e-05
[cse_after_recomputation]: 3.376e-05, [1]
    [Cycle 1]: 2.811e-05, [1]
        [cse]: 2.233e-05
[auto_monad_reorder]: 2.146e-05
[get_jit_bprop_graph]: 1.27999e-06
[rewriter_after_jit_bprop_graph]: 4.78001e-06
[opt_after_jit_grad]: 0.00045635
[symbol_engine_optimizer]: 9.21e-05, [1]
    [Cycle 1]: 8.587e-05, [6]
        [build]: 3.93001e-06
        [elim_shapecalc]: 1.131e-05
        [elim_not_effective]: 1.858e-05
        [opt_reshape]: 8.94e-06
        [fold_const_symbol]: 1.51e-05
        [renormalize]: 8.10018e-07
[validate]: 4.211e-05
[backend_pass]: 9.39996e-07
[task_emit]: 0.00848128
[execute]: 7.15e-06
Sums
    bootstrap                            :     0.000203s :  0.71%
    type_inference                       :     0.009678s : 33.94%
    event_method                         :     0.000045s :  0.16%
    auto_monad                           :     0.000088s :  0.31%
    graph_reusing                        :     0.000007s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000022s :  0.08%
    rewriter_before_opt_a                :     0.000085s :  0.30%
    expand_dump_flag                     :     0.000004s :  0.01%
    jit_opt_a.switch_simplify            :     0.000104s :  0.36%
    jit_opt_a.loop_unroll                :     0.000081s :  0.29%
    jit_opt_a.a_1                        :     0.002113s :  7.41%
    jit_opt_a.with_stream_mark           :     0.000046s :  0.16%
    jit_opt_a.recompute_prepare          :     0.000035s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000013s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000030s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000027s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000029s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000028s :  0.10%
    jit_opt_a.merge_forward              :     0.000016s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000058s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000050s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001486s :  5.21%
    jit_opt_a.replace_old_param          :     0.000071s :  0.25%
    jit_opt_a.inline_without_move        :     0.000066s :  0.23%
    jit_opt_a.renormalize                :     0.003135s : 10.99%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000078s :  0.28%
    jit_opt_a.cse                        :     0.000293s :  1.03%
    jit_opt_a.replace_applicator         :     0.000090s :  0.31%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.04%
    rewriter_after_opt_a                 :     0.000046s :  0.16%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000575s :  2.02%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000023s :  0.08%
    cconv                                :     0.000023s :  0.08%
    loop_unroll                          :     0.000432s :  1.51%
    jit_opt_after_cconv.c_1              :     0.000034s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.cse              :     0.000031s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000008s :  0.03%
    remove_dup_value                     :     0.000022s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000055s :  0.19%
    cse_after_recomputation.cse          :     0.000022s :  0.08%
    auto_monad_reorder                   :     0.000021s :  0.08%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000456s :  1.60%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008481s : 29.74%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000510   126
 6.92% :     0.000035s :      2: substitution.arithmetic_simplify
 0.60% :     0.000003s :      5: substitution.elim_not_effective
 0.41% :     0.000002s :      5: substitution.fold_const_symbol
 1.35% :     0.000007s :      6: substitution.graph_param_transform
58.66% :     0.000299s :     12: substitution.inline
 3.10% :     0.000016s :      2: substitution.inline_without_move
 2.03% :     0.000010s :     18: substitution.j_node_and_user_rematch
 2.00% :     0.000010s :      7: substitution.minmaximum_grad
 1.41% :     0.000007s :      5: substitution.partial_eliminate
 2.36% :     0.000012s :     18: substitution.remove_not_recompute_node
 4.94% :     0.000025s :     10: substitution.replace_applicator
 1.38% :     0.000007s :      7: substitution.replace_old_param
 0.49% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.18% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.67% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.49% :     0.000038s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009618     2
88.17% :     0.008481s :      1: type_inference.infer
11.83% :     0.001138s :      1: type_inference.specialize
------[replace.]   0.000138    19
65.25% :     0.000090s :     12: replace.inline
34.75% :     0.000048s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000313    19
93.43% :     0.000293s :     12: match.inline
 6.57% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000402  2895
 1.66% :     0.000007s :     48: predicate.accumulaten_eliminater
 0.54% :     0.000002s :      6: predicate.ad_related_special_op_eliminate
 1.45% :     0.000006s :     48: predicate.addn_check_dump
 1.75% :     0.000007s :     48: predicate.addn_zero_filter
 2.06% :     0.000008s :     48: predicate.arithmetic_simplify
 1.61% :     0.000006s :     48: predicate.cast_eliminate
 0.21% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.44% :     0.000006s :     48: predicate.compare_switch_simplify
 1.54% :     0.000006s :     48: predicate.depend_value_elim
 1.48% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.54% :     0.000006s :     48: predicate.dict_get_item_eliminator
 1.50% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.35% :     0.000001s :      6: predicate.dumpgradient_eliminate
 0.15% :     0.000001s :      6: predicate.elim_not_effective
 0.30% :     0.000001s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.52% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.44% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.43% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.62% :     0.000007s :     48: predicate.environ_get_eliminate
 1.46% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.11% :     0.000000s :      6: predicate.fold_const_symbol
 0.94% :     0.000004s :     23: predicate.get_grad_eliminate
 0.13% :     0.000001s :      6: predicate.graph_param_transform
 4.09% :     0.000016s :     79: predicate.inline
 1.77% :     0.000007s :     44: predicate.inline_without_move
 0.40% :     0.000002s :     23: predicate.j_node_and_user_rematch
 1.01% :     0.000004s :     23: predicate.less_batch_normalization
 1.82% :     0.000007s :     55: predicate.list_to_tuple_eliminator_
 1.97% :     0.000008s :     61: predicate.load_eliminater
 0.70% :     0.000003s :      6: predicate.loop_unroll_after_grad
 3.02% :     0.000012s :     90: predicate.loop_unroll_before_grad
 1.87% :     0.000008s :     54: predicate.make_slice_get_slice_eliminator
 1.46% :     0.000006s :     48: predicate.merge_addn
 1.50% :     0.000006s :     48: predicate.minmaximum_grad
 0.63% :     0.000003s :      6: predicate.mutable_eliminate
 0.30% :     0.000001s :      6: predicate.opt_reshape
 2.31% :     0.000009s :     61: predicate.partial_eliminate
 1.49% :     0.000006s :     48: predicate.print_const_string_wrapper
 1.82% :     0.000007s :     48: predicate.reduce_eliminate
 1.83% :     0.000007s :     55: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.51% :     0.000010s :    108: predicate.replace_applicator
 0.86% :     0.000003s :     44: predicate.replace_old_param
 0.12% :     0.000000s :      6: predicate.reset_defer_inline
 1.62% :     0.000007s :     48: predicate.reshape_eliminate
 1.62% :     0.000007s :     48: predicate.row_tensor_add_zeros_like
 0.33% :     0.000001s :      6: predicate.row_tensor_eliminate
 1.53% :     0.000006s :     48: predicate.same_eliminate
 0.47% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.50% :     0.000002s :     12: predicate.special_op_eliminate
 0.92% :     0.000004s :     23: predicate.specialize_transform
 1.72% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.47% :     0.000006s :     48: predicate.stack_unstack_eliminate
 0.24% :     0.000001s :      6: predicate.switch_call_monad_eliminater
 2.58% :     0.000010s :     67: predicate.switch_defer_inline
 2.39% :     0.000010s :     67: predicate.switch_layer_defer_inline
 5.93% :     0.000024s :    163: predicate.switch_simplify
 1.49% :     0.000006s :     48: predicate.tile_eliminate
 1.46% :     0.000006s :     48: predicate.transpose_eliminate
 1.89% :     0.000008s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.66% :     0.000007s :     48: predicate.tuple_list_get_item_depend_reorder
 3.07% :     0.000012s :     67: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000007s :     48: predicate.tuple_list_set_item_eliminator
 1.83% :     0.000007s :     55: predicate.tuple_to_list_eliminator_
 2.03% :     0.000008s :     61: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000012s :     84: predicate.updatestate_useless_node_eliminater
 1.88% :     0.000008s :     48: predicate.value_based_eliminate
 0.20% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.31% :     0.000001s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001440    29
57.88% :     0.000833s :     13: func_graph_cloner_run.FuncGraphClonerGraph
42.12% :     0.000606s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038446    91
 0.15% :     0.000058s :      1: add_recomputation
 0.24% :     0.000092s :      1: auto_monad
 0.06% :     0.000024s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.56% :     0.000214s :      1: bootstrap
 0.07% :     0.000026s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000036s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.13% :     0.000049s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
29.88% :     0.011490s :      1: jit_opt_a
 0.49% :     0.000189s :      1: jit_opt_after_cconv
 0.18% :     0.000069s :      1: jit_opt_b
 1.14% :     0.000439s :      1: loop_unroll
 1.52% :     0.000583s :      1: mutable_eliminate
 7.13% :     0.002742s :     39: opt.transform.jit_opt_a
 0.19% :     0.000073s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000041s :      4: opt.transform.jit_opt_b
 0.05% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000050s :      4: opt.transform.symbol_engine_opt
 1.21% :     0.000464s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.06% :     0.000024s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000025s :      1: remove_dup_value
 4.46% :     0.001715s :      2: renormalize.infer
 3.66% :     0.001406s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000049s :      1: rewriter_after_opt_a
 0.23% :     0.000088s :      1: rewriter_before_opt_a
 0.25% :     0.000095s :      1: symbol_engine_optimizer
22.09% :     0.008493s :      1: task_emit
25.20% :     0.009689s :      1: type_inference
 0.17% :     0.000066s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:22.593.46 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c3821f0> should be wrapped by jit only once.

TotalTime = 0.0195681, [30]
[bootstrap]: 0.0001839
[type_inference]: 0.00895213
[event_method]: 8.116e-05
[auto_monad]: 8.715e-05
[graph_reusing]: 6.17999e-06
[pre_auto_parallel]: 2.37999e-06
[py_interpret_to_execute]: 2.636e-05
[rewriter_before_opt_a]: 8.522e-05
[expand_dump_flag]: 3.4e-06
[jit_opt_a]: 0.00796818, [2]
    [Cycle 1]: 0.00462718, [27]
        [switch_simplify]: 5.681e-05
        [loop_unroll]: 3.968e-05
        [a_1]: 0.00088774
        [with_stream_mark]: 2.595e-05
        [recompute_prepare]: 1.809e-05
        [updatestate_depend_eliminate]: 7.79002e-06
        [updatestate_assign_eliminate]: 5.72001e-06
        [updatestate_loads_eliminate]: 5.62001e-06
        [parameter_eliminate]: 2.11e-06
        [specialize_transform]: 1.293e-05
        [updatestate_useless_node_eliminater]: 1.212e-05
        [accelerated_algorithm]: 1.252e-05
        [meta_shard_fg_expand]: 3.03e-06
        [get_grad_eliminate_]: 1.233e-05
        [merge_forward]: 8.37e-06
        [cell_reuse_recompute_pass]: 7.90023e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.507e-05
        [j_node_and_user_rematch]: 2.337e-05
        [meta_fg_expand]: 0.00150587
        [replace_old_param]: 5.334e-05
        [inline_without_move]: 5.142e-05
        [renormalize]: 0.0015858
        [add_forward_monad_depend]: 6.82002e-06
        [auto_monad_grad]: 2.46998e-06
        [auto_monad_eliminator]: 1.589e-05
        [cse]: 2.548e-05
        [replace_applicator]: 1.319e-05
    [Cycle 2]: 0.00033239, [27]
        [switch_simplify]: 5.30999e-06
        [loop_unroll]: 3.88999e-06
        [a_1]: 3.931e-05
        [with_stream_mark]: 1.015e-05
        [recompute_prepare]: 3.7e-06
        [updatestate_depend_eliminate]: 2.66999e-06
        [updatestate_assign_eliminate]: 2.57001e-06
        [updatestate_loads_eliminate]: 1.74e-06
        [parameter_eliminate]: 1.66e-06
        [specialize_transform]: 3.91999e-06
        [updatestate_useless_node_eliminater]: 6.49001e-06
        [accelerated_algorithm]: 3.44001e-06
        [meta_shard_fg_expand]: 1.89e-06
        [get_grad_eliminate_]: 3.59002e-06
        [merge_forward]: 2.45002e-06
        [cell_reuse_recompute_pass]: 1.84998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.353e-05
        [j_node_and_user_rematch]: 6.24001e-06
        [meta_fg_expand]: 2.519e-05
        [replace_old_param]: 3.76999e-06
        [inline_without_move]: 3.24001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.57001e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 5.84999e-06
        [cse]: 1.237e-05
        [replace_applicator]: 4.04997e-06
[py_interpret_to_execute_after_opt_a]: 1.152e-05
[rewriter_after_opt_a]: 3.509e-05
[convert_after_rewriter]: 5.91998e-06
[order_py_execute_after_rewriter]: 4.90001e-06
[mutable_eliminate]: 0.00059534
[jit_opt_b]: 4.278e-05, [1]
    [Cycle 1]: 3.496e-05, [2]
        [frontend_op_eliminate]: 1.073e-05
        [inline_after_opt_a]: 1.233e-05
[cconv]: 2.319e-05
[loop_unroll]: 0.00045526
[jit_opt_after_cconv]: 0.00013144, [1]
    [Cycle 1]: 0.00012425, [11]
        [c_1]: 1.306e-05
        [parameter_eliminate]: 3.33998e-06
        [updatestate_depend_eliminate]: 6.14001e-06
        [updatestate_assign_eliminate]: 2.24999e-06
        [updatestate_loads_eliminate]: 1.59e-06
        [cse]: 2.011e-05
        [call_graph_tuple_transform]: 1.5e-05
        [tuple_list_get_item_eliminator]: 3.87002e-06
        [none_parameter_eliminate]: 1.52001e-06
        [renormalize]: 9.89996e-07
        [switch_simplify]: 3.95e-06
[remove_dup_value]: 1.49e-05
[partial_unused_args_eliminate]: 2.32999e-06
[environ_conv]: 6.46e-06
[add_recomputation]: 3.617e-05
[cse_after_recomputation]: 2.115e-05, [1]
    [Cycle 1]: 1.492e-05, [1]
        [cse]: 8.52998e-06
[auto_monad_reorder]: 1.334e-05
[get_jit_bprop_graph]: 1.93997e-06
[rewriter_after_jit_bprop_graph]: 5.61e-06
[opt_after_jit_grad]: 0.00050196
[symbol_engine_optimizer]: 6.769e-05, [1]
    [Cycle 1]: 6.095e-05, [6]
        [build]: 2.88e-06
        [elim_shapecalc]: 6.28e-06
        [WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:22.600.48 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[elim_not_effective]: 1.044e-05
        [opt_reshape]: 4.29002e-06
        [fold_const_symbol]: 6.06e-06
        [renormalize]: 5.09986e-07
[validate]: 3.197e-05
Sums
    bootstrap                            :     0.000184s :  1.16%
    type_inference                       :     0.008952s : 56.31%
    event_method                         :     0.000081s :  0.51%
    auto_monad                           :     0.000087s :  0.55%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.17%
    rewriter_before_opt_a                :     0.000085s :  0.54%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.39%
    jit_opt_a.loop_unroll                :     0.000044s :  0.27%
    jit_opt_a.a_1                        :     0.000927s :  5.83%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000019s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.10%
    jit_opt_a.merge_forward              :     0.000011s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000030s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.001531s :  9.63%
    jit_opt_a.replace_old_param          :     0.000057s :  0.36%
    jit_opt_a.inline_without_move        :     0.000055s :  0.34%
    jit_opt_a.renormalize                :     0.001586s :  9.97%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.14%
    jit_opt_a.cse                        :     0.000038s :  0.24%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000035s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000595s :  3.74%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.08%
    cconv                                :     0.000023s :  0.15%
    loop_unroll                          :     0.000455s :  2.86%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.13%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.04%
    add_recomputation                    :     0.000036s :  0.23%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000502s :  3.16%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000032s :  0.20%


Time group info:
------[substitution.]   0.000230    50
 0.67% :     0.000002s :      1: substitution.elim_not_effective
 0.49% :     0.000001s :      1: substitution.fold_const_symbol
 1.80% :     0.000004s :      1: substitution.graph_param_transform
63.54% :     0.000146s :      6: substitution.inline
 7.42% :     0.000017s :      2: substitution.inline_without_move
 2.95% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.55% :     0.000004s :      2: substitution.minmaximum_grad
 1.63% :     0.000004s :      3: substitution.partial_eliminate
 2.98% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.19% :     0.000005s :      2: substitution.replace_applicator
 1.91% :     0.000004s :      5: substitution.replace_old_param
 1.23% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.75% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 2.16% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 6.73% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008893     2
88.82% :     0.007899s :      1: type_inference.infer
11.18% :     0.000994s :      1: type_inference.specialize
------[replace.]   0.000060     8
78.19% :     0.000047s :      6: replace.inline
21.81% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000153     8
93.29% :     0.000143s :      6: match.inline
 6.71% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000244  1109
 1.04% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.48% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 0.94% :     0.000002s :     18: predicate.addn_check_dump
 1.14% :     0.000003s :     18: predicate.addn_zero_filter
 1.38% :     0.000003s :     18: predicate.arithmetic_simplify
 0.99% :     0.000002s :     18: predicate.cast_eliminate
 0.11% :     0.000000s :      1: predicate.check_bprop_eliminate
 0.90% :     0.000002s :     18: predicate.compare_switch_simplify
 0.99% :     0.000002s :     18: predicate.depend_value_elim
 0.90% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.13% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.04% :     0.000003s :     18: predicate.dict_set_item_eliminator
 0.42% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.06% :     0.000000s :      1: predicate.elim_not_effective
 0.24% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000003s :     18: predicate.environ_add_const_eliminate
 0.92% :     0.000002s :     18: predicate.environ_get_add_eliminate
 0.93% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.12% :     0.000003s :     18: predicate.environ_get_eliminate
 0.96% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.04% :     0.000000s :      1: predicate.fold_const_symbol
 0.99% :     0.000002s :     12: predicate.get_grad_eliminate
 0.14% :     0.000000s :      1: predicate.graph_param_transform
 2.99% :     0.000007s :     28: predicate.inline
 2.17% :     0.000005s :     33: predicate.inline_without_move
 0.37% :     0.000001s :     12: predicate.j_node_and_user_rematch
 0.95% :     0.000002s :     12: predicate.less_batch_normalization
 1.11% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.26% :     0.000003s :     21: predicate.load_eliminater
 0.68% :     0.000002s :      1: predicate.loop_unroll_after_grad
 2.56% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.22% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 0.94% :     0.000002s :     18: predicate.merge_addn
 0.93% :     0.000002s :     18: predicate.minmaximum_grad
 0.60% :     0.000001s :      1: predicate.mutable_eliminate
 0.14% :     0.000000s :      1: predicate.opt_reshape
 1.47% :     0.000004s :     21: predicate.partial_eliminate
 1.08% :     0.000003s :     18: predicate.print_const_string_wrapper
 1.48% :     0.000004s :     18: predicate.reduce_eliminate
 1.12% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.19% :     0.000003s :     22: predicate.replace_applicator
 1.19% :     0.000003s :     33: predicate.replace_old_param
 0.18% :     0.000000s :      1: predicate.reset_defer_inline
 1.09% :     0.000003s :     18: predicate.reshape_eliminate
 1.09% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.37% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.07% :     0.000003s :     18: predicate.same_eliminate
 0.43% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.22% :     0.000001s :      2: predicate.special_op_eliminate
 0.82% :     0.000002s :     12: predicate.specialize_transform
 1.13% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 0.94% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.18% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 1.95% :     0.000005s :     26: predicate.switch_defer_inline
 1.74% :     0.000004s :     26: predicate.switch_layer_defer_inline
 4.91% :     0.000012s :     71: predicate.switch_simplify
 0.95% :     0.000002s :     18: predicate.tile_eliminate
 0.96% :     0.000002s :     18: predicate.transpose_eliminate
 1.20% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.23% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.18% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.12% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.15% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.14% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
29.69% :     0.000072s :     18: predicate.value_based_eliminate
 0.07% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.14% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001065    22
74.71% :     0.000796s :     13: func_graph_cloner_run.FuncGraphClonerGraph
25.29% :     0.000269s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022468    72
 0.17% :     0.000039s :      1: add_recomputation
 0.41% :     0.000092s :      1: auto_monad
 0.07% :     0.000016s :      1: auto_monad_reorder
 0.87% :     0.000196s :      1: bootstrap
 0.11% :     0.000026s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000009s :      1: environ_conv
 0.38% :     0.000086s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
35.48% :     0.007971s :      1: jit_opt_a
 0.60% :     0.000134s :      1: jit_opt_after_cconv
 0.20% :     0.000045s :      1: jit_opt_b
 2.07% :     0.000465s :      1: loop_unroll
 2.69% :     0.000605s :      1: mutable_eliminate
 5.74% :     0.001289s :     26: opt.transform.jit_opt_a
 0.14% :     0.000032s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000016s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000023s :      4: opt.transform.symbol_engine_opt
 2.27% :     0.000511s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.13% :     0.000029s :      1: py_interpret_to_execute
 0.06% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 5.40% :     0.001213s :      1: renormalize.infer
 1.62% :     0.000364s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000039s :      1: rewriter_after_opt_a
 0.39% :     0.000089s :      1: rewriter_before_opt_a
 0.31% :     0.000070s :      1: symbol_engine_optimizer
39.89% :     0.008962s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.651.36 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d1a60> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.655.46 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e34c0> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.656.99 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.660.81 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0142349, [30]
[bootstrap]: 0.00034621
[type_inference]: 0.00489832
[event_method]: 1.901e-05
[auto_monad]: 6.789e-05
[graph_reusing]: 6.46e-06
[pre_auto_parallel]: 2.49001e-06
[py_interpret_to_execute]: 3.273e-05
[rewriter_before_opt_a]: 7.288e-05
[expand_dump_flag]: 3.28998e-06
[jit_opt_a]: 0.0060792, [2]
    [Cycle 1]: 0.00184622, [27]
        [switch_simplify]: 6.358e-05
        [loop_unroll]: 3.132e-05
        [a_1]: 0.00059359
        [with_stream_mark]: 2.564e-05
        [recompute_prepare]: 1.091e-05
        [updatestate_depend_eliminate]: 4.45999e-06
        [updatestate_assign_eliminate]: 3.39001e-06
        [updatestate_loads_eliminate]: 2.91e-06
        [parameter_eliminate]: 2.26998e-06
        [specialize_transform]: 6.58e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 5.72999e-06
        [meta_shard_fg_expand]: 2.43e-06
        [get_grad_eliminate_]: 5.44e-06
        [merge_forward]: 4.07e-06
        [cell_reuse_recompute_pass]: 1.55001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.647e-05
        [j_node_and_user_rematch]: 9.55001e-06
        [meta_fg_expand]: 3.01999e-06
        [replace_old_param]: 9.71998e-06
        [inline_without_move]: 5.57001e-06
        [renormalize]: 0.00073437
        [add_forward_monad_depend]: 7.91001e-06
        [auto_monad_grad]: 2.49999e-06
        [auto_monad_eliminator]: 1.812e-05
        [cse]: 3.219e-05
        [replace_applicator]: 1.57e-05
    [Cycle 2]: 0.00034515, [27]
        [switch_simplify]: 5.34e-06
        [loop_unroll]: 4.61002e-06
        [a_1]: 8.943e-05
        [with_stream_mark]: 1.298e-05
        [recompute_prepare]: 6.21e-06
        [updatestate_depend_eliminate]: 2.88998e-06
        [updatestate_assign_eliminate]: 1.99999e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [parameter_eliminate]: 1.51998e-06
        [specialize_transform]: 5.94e-06
        [updatestate_useless_node_eliminater]: 4.63001e-06
        [accelerated_algorithm]: 5.02e-06
        [meta_shard_fg_expand]: 1.35999e-06
        [get_grad_eliminate_]: 5.12999e-06
        [merge_forward]: 3.31001e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.617e-05
        [j_node_and_user_rematch]: 8.55999e-06
        [meta_fg_expand]: 2.96999e-06
        [replace_old_param]: 7.73001e-06
        [inline_without_move]: 4.55001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.40999e-06
        [auto_monad_grad]: 8.80013e-07
        [auto_monad_eliminator]: 7.55e-06
        [cse]: 1.2e-05
        [replace_applicator]: 6.81001e-06
[py_interpret_to_execute_after_opt_a]: 1.413e-05
[rewriter_after_opt_a]: 4.244e-05
[convert_after_rewriter]: 8.65001e-06
[order_py_execute_after_rewriter]: 5.74e-06
[mutable_eliminate]: 0.00075715
[jit_opt_b]: 5.865e-05, [1]
    [Cycle 1]: 4.842e-05, [2]
        [frontend_op_eliminate]: 1.869e-05
        [inline_after_opt_a]: 1.711e-05
[cconv]: 3.062e-05
[loop_unroll]: 0.00056323
[jit_opt_after_cconv]: 0.0001583, [1]
    [Cycle 1]: 0.00015002, [11]
        [c_1]: 2.014e-05
        [parameter_eliminate]: 3.97e-06
        [updatestate_depend_eliminate]: 7.95998e-06
        [updatestate_assign_eliminate]: 2.57001e-06
        [updatestate_loads_eliminate]: 2.76e-06
        [cse]: 2.4e-05
        [call_graph_tuple_transform]: 2.075e-05
        [tuple_list_get_item_eliminator]: 5.51e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 8.49977e-07
        [switch_simplify]: 5.74999e-06
[remove_dup_value]: 1.676e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 6.13002e-06
[add_recomputation]: 5.807e-05
[cse_after_recomputation]: 2.495e-05, [1]
    [Cycle 1]: 1.776e-05, [1]
        [cse]: 1.009e-05
[auto_monad_reorder]: 1.757e-05
[get_jit_bprop_graph]: 2.07001e-06
[rewriter_after_jit_bprop_graph]: 7.44002e-06
[opt_after_jit_grad]: 0.00059209
[symbol_engine_optimizer]: 7.971e-05, [1]
    [Cycle 1]: 7.068e-05, [6]
        [build]: 5.68002e-06
        [elim_shapecalc]: 6.88e-06
        [elim_not_effective]: 1.508e-05
        [opt_reshape]: 5.27999e-06
        [fold_const_symbol]: 8.62998e-06
        [renormalize]: 5.00004e-07
[validate]: 4.322e-05
Sums
    bootstrap                            :     0.000346s :  3.59%
    type_inference                       :     0.004898s : 50.77%
    event_method                         :     0.000019s :  0.20%
    auto_monad                           :     0.000068s :  0.70%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.34%
    rewriter_before_opt_a                :     0.000073s :  0.76%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000069s :  0.71%
    jit_opt_a.loop_unroll                :     0.000036s :  0.37%
    jit_opt_a.a_1                        :     0.000683s :  7.08%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.40%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.18%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000013s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000006s :  0.06%
    jit_opt_a.replace_old_param          :     0.000017s :  0.18%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000734s :  7.61%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.27%
    jit_opt_a.cse                        :     0.000044s :  0.46%
    jit_opt_a.replace_applicator         :     0.000023s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.15%
    rewriter_after_opt_a                 :     0.000042s :  0.44%
    convert_after_rewriter               :     0.000009s :  0.09%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000757s :  7.85%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.18%
    cconv                                :     0.000031s :  0.32%
    loop_unroll                          :     0.000563s :  5.84%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000024s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000017s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000058s :  0.60%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000018s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.08%
    opt_after_jit_grad                   :     0.000592s :  6.14%
    symbol_engine_optimizer.build        :     0.000006s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.45%


Time group info:
------[substitution.]   0.000245    23
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 2.34% :     0.000006s :      3: substitution.graph_param_transform
84.51% :     0.000207s :      4: substitution.inline
 1.83% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.07% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.21% :     0.000005s :      2: substitution.replace_old_param
 5.59% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004823     2
81.86% :     0.003948s :      1: type_inference.infer
18.14% :     0.000875s :      1: type_inference.specialize
------[replace.]   0.000077     6
64.89% :     0.000050s :      4: replace.inline
35.11% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000216     6
94.31% :     0.000204s :      4: match.inline
 5.69% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.41% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.78% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.92% :     0.000001s :     10: predicate.addn_check_dump
 2.45% :     0.000003s :     10: predicate.addn_zero_filter
 2.26% :     0.000003s :     10: predicate.arithmetic_simplify
 1.01% :     0.000001s :     10: predicate.cast_eliminate
 0.35% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.92% :     0.000001s :     10: predicate.compare_switch_simplify
 1.60% :     0.000002s :     10: predicate.depend_value_elim
 0.92% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.98% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.65% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.98% :     0.000001s :     10: predicate.environ_get_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.73% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.60% :     0.000007s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.27% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.52% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.75% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.82% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.94% :     0.000001s :     10: predicate.merge_addn
 0.92% :     0.000001s :     10: predicate.minmaximum_grad
 2.23% :     0.000003s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 1.85% :     0.000002s :     15: predicate.partial_eliminate
 0.94% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     10: predicate.reduce_eliminate
 1.26% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.47% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.24% :     0.000002s :     10: predicate.reshape_eliminate
 0.98% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.20% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.85% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.41% :     0.000002s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.19% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 1.79% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.85% :     0.000010s :     53: predicate.switch_simplify
 1.68% :     0.000002s :     10: predicate.tile_eliminate
 1.17% :     0.000002s :     10: predicate.transpose_eliminate
 1.22% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.07% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 5.22% :     0.000007s :     18: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.40% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.67% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.95% :     0.000003s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.08% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000637    12
51.46% :     0.000328s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.55% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.99% :     0.000261s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015947    72
 0.39% :     0.000062s :      1: add_recomputation
 0.45% :     0.000071s :      1: auto_monad
 0.14% :     0.000022s :      1: auto_monad_reorder
 2.33% :     0.000371s :      1: bootstrap
 0.21% :     0.000033s :      1: cconv
 0.07% :     0.000011s :      1: convert_after_rewriter
 0.17% :     0.000027s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
38.14% :     0.006083s :      1: jit_opt_a
 1.01% :     0.000161s :      1: jit_opt_after_cconv
 0.38% :     0.000061s :      1: jit_opt_b
 3.62% :     0.000578s :      1: loop_unroll
 4.84% :     0.000772s :      1: mutable_eliminate
 5.74% :     0.000916s :     26: opt.transform.jit_opt_a
 0.31% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000028s :      4: opt.transform.jit_opt_b
 0.11% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.14% :     0.000023s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.81% :     0.000607s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.22% :     0.000035s :      1: py_interpret_to_execute
 0.10% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000019s :      1: remove_dup_value
 2.50% :     0.000398s :      1: renormalize.infer
 2.04% :     0.000325s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.29% :     0.000047s :      1: rewriter_after_opt_a
 0.47% :     0.000076s :      1: rewriter_before_opt_a
 0.52% :     0.000082s :      1: symbol_engine_optimizer
30.84% :     0.004918s :      1: type_inference


TotalTime = 0.0185738, [30]
[bootstrap]: 0.0002026
[type_inference]: 0.00856921
[event_method]: 4.227e-05
[auto_monad]: 7.219e-05
[graph_reusing]: 5.14e-06
[pre_auto_parallel]: 1.94999e-06
[py_interpret_to_execute]: 2.565e-05
[rewriter_before_opt_a]: 9.136e-05
[expand_dump_flag]: 2.37999e-06
[jit_opt_a]: 0.00762513, [2]
    [Cycle 1]: 0.00424507, [27]
        [switch_simplify]: 6.093e-05
        [loop_unroll]: 3.812e-05
        [a_1]: 0.00081766
        [with_stream_mark]: 2.445e-05
        [recompute_prepare]: 1.728e-05
        [updatestate_depend_eliminate]: 7.53999e-06
        [updatestate_assign_eliminate]: 6.18002e-06
        [updatestate_loads_eliminate]: 6.40002e-06
        [parameter_eliminate]: 2.46e-06
        [specialize_transform]: 1.278e-05
        [updatestate_useless_node_eliminater]: 1.197e-05
        [accelerated_algorithm]: 1.215e-05
        [meta_shard_fg_expand]: 2.84001e-06
        [get_grad_eliminate_]: 1.201e-05
        [merge_forward]: 7.86001e-06
        [cell_reuse_recompute_pass]: 7.49977e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.533e-05
        [j_node_and_user_rematch]: 2.254e-05
        [meta_fg_expand]: 0.00136696
        [replace_old_param]: 5.132e-05
        [inline_without_move]: 5.071e-05
        [renormalize]: 0.00143317
        [add_forward_monad_depend]: 5.52001e-06
        [auto_monad_grad]: 2.52001e-06
        [auto_monad_eliminator]: 1.22e-05
        [cse]: 2.469e-05
        [replace_applicator]: 1.084e-05
    [Cycle 2]: 0.00027423, [27]
        [switch_simplify]: 4.52e-06
        [loop_unroll]: 4.00998e-06
        [a_1]: 3.755e-05
        [with_stream_mark]: 8.48001e-06
        [recompute_prepare]: 3.73001e-06
        [updatestate_depend_eliminate]: 2.22999e-06
        [updatestate_assign_eliminate]: 1.49998e-06
        [updatestate_loads_eliminate]: 1.40999e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 3.6e-06
        [updatestate_useless_node_eliminater]: 3.66001e-06
        [accelerated_algorithm]: 3.41001e-06
        [meta_shard_fg_expand]: 1.14e-06
        [get_grad_eliminate_]: 3.25e-06
        [merge_forward]: 1.97999e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.60001e-06
        [j_node_and_user_rematch]: 5.77001e-06
        [meta_fg_expand]: 3.025e-05
        [replace_old_param]: 3.80998e-06
        [inline_without_move]: 3.23e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.37999e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 4.65999e-06
        [cse]: 9.78002e-06
        [replace_applicator]: 4.11001e-06
[py_interpret_to_execute_after_opt_a]: 8.60001e-06
[rewriter_after_opt_a]: 3.087e-05
[convert_after_rewriter]: 6.85998e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00048384
[jit_opt_b]: 3.915e-05, [1]
    [Cycle 1]: 3.24e-05, [2]
        [frontend_op_eliminate]: 9.72001e-06
        [inline_after_opt_a]: 1.12e-05
[cconv]: 2.088e-05
[loop_unroll]: 0.00042258
[jit_opt_after_cconv]: 0.00011665, [1]
    [Cycle 1]: 0.00011015, [11]
        [c_1]: 1.273e-05
        [parameter_eliminate]: 2.74001e-06
        [updatestate_depend_eliminate]: 5.07999e-06
        [updatestate_assign_eliminate]: 1.95001e-06
        [updatestate_loads_eliminate]: 1.62999e-06
        [cse]: 1.463e-05
        [call_graph_tuple_transform]: 1.281e-05
        [tuple_list_get_item_eliminator]: 3.88001e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 3.73999e-06
[remove_dup_value]: 1.491e-05
[partial_unused_args_eliminate]: 2.77002e-06
[environ_conv]: 5.48002e-06
[add_recomputation]: 3.386e-05
[cse_after_recomputation]: 1.89e-05, [1]
    [Cycle 1]: 1.381e-05, [1]
        [cse]: 7.93999e-06
[auto_monad_reorder]: 1.3e-05
[get_jit_bprop_graph]: 1.67001e-06
[rewriter_after_jit_bprop_graph]: 4.95999e-06
[opt_after_jit_grad]: 0.00045242
[symbol_engine_optimizer]: 6.265e-05, [1]
    [Cycle 1]: 5.654e-05, [6]
        [build]: 3.13998e-06
        [elim_shapecalc]: 5.55001e-06
        [elim_not_effective]: 9.91e-06
        [opt_reshape]: 4.13999e-06
        [fold_const_symbol]: 5.64e-06
        [renormalize]: 5.60016e-07
[validate]: 2.538e-05
Sums
    bootstrap                            :     0.000203s :  1.36%
    type_inference                       :     0.008569s : 57.63%
    event_method                         :     0.000042s :  0.28%
    auto_monad                           :     0.000072s :  0.49%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000026s :  0.17%
    rewriter_before_opt_a                :     0.000091s :  0.61%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.44%
    jit_opt_a.loop_unroll                :     0.000042s :  0.28%
    jit_opt_a.a_1                        :     0.000855s :  5.75%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000016s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.10%
    jit_opt_a.merge_forward              :     0.000010s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.001397s :  9.40%
    jit_opt_a.replace_old_param          :     0.000055s :  0.37%
    jit_opt_a.inline_without_move        :     0.000054s :  0.36%
    jit_opt_a.renormalize                :     0.001433s :  9.64%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.11%
    jit_opt_a.cse                        :     0.000034s :  0.23%
    jit_opt_a.replace_applicator         :     0.000015s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000031s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000484s :  3.25%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.08%
    cconv                                :     0.000021s :  0.14%
    loop_unroll                          :     0.000423s :  2.84%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.03%
    remove_dup_value                     :     0.000015s :  0.10%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.04%
    add_recomputation                    :     0.000034s :  0.23%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000452s :  3.04%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000025s :  0.17%


TotalTime = 0.0242433, [33]
[bootstrap]: 0.00022753
[type_inference]: 0.0108438
[event_method]: 4.48e-05
[auto_monad]: 9.282e-05
[graph_reusing]: 7.29001e-06
[pre_auto_parallel]: 3.74002e-06
[py_interpret_to_execute]: 2.661e-05
[rewriter_before_opt_a]: 9.064e-05
[expand_dump_flag]: 3.14999e-06
[jit_opt_a]: 0.0110293, [2]
    [Cycle 1]: 0.00439461, [27]
        [switch_simplify]: 6.281e-05
        [loop_unroll]: 3.773e-05
        [a_1]: 0.00081652
        [with_stream_mark]: 2.41e-05
        [recompute_prepare]: 1.732e-05
        [updatestate_depend_eliminate]: 7.09001e-06
        [updatestate_assign_eliminate]: 6.89999e-06
        [updatestate_loads_eliminate]: 6.01e-06
        [parameter_eliminate]: 2.56998e-06
        [specialize_transform]: 1.241e-05
        [updatestate_useless_node_eliminater]: 1.159e-05
        [accelerated_algorithm]: 1.246e-05
        [meta_shard_fg_expand]: 3.38999e-06
        [get_grad_eliminate_]: 1.195e-05
        [merge_forward]: 8.48001e-06
        [cell_reuse_recompute_pass]: 9.20001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.508e-05
        [j_node_and_user_rematch]: 2.274e-05
        [meta_fg_expand]: 0.00136596
        [replace_old_param]: 5.092e-05
        [inline_without_move]: 5.08e-05
        [renormalize]: 0.00159556
        [add_forward_monad_depend]: 4.43999e-06
        [auto_monad_grad]: 1.13001e-06
        [auto_monad_eliminator]: 8.28001e-06
        [cse]: 1.804e-05
        [replace_applicator]: 1.118e-05
    [Cycle 2]: 0.00027381, [27]
        [switch_simplify]: 4.42e-06
        [loop_unroll]: 3.78001e-06
        [a_1]: 3.833e-05
        [with_stream_mark]: 7.71001e-06
        [recompute_prepare]: 3.73999e-06
        [updatestate_depend_eliminate]: 1.94999e-06
        [updatestate_assign_eliminate]: 1.62999e-06
        [updatestate_loads_eliminate]: 1.44e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 3.43999e-06
        [updatestate_useless_node_eliminater]: 3.35e-06
        [accelerated_algorithm]: 5.52999e-06
        [meta_shard_fg_expand]: 1.03001e-06
        [get_grad_eliminate_]: 3.74002e-06
        [merge_forward]: 1.89999e-06
        [cell_reuse_recompute_pass]: 1.79998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.88998e-06
        [j_node_and_user_rematch]: 5.45001e-06
        [meta_fg_expand]: 3.052e-05
        [replace_old_param]: 4.4e-06
        [inline_without_move]: 3.21001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.36002e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 4.69998e-06
        [cse]: 1.061e-05
        [replace_applicator]: 3.81999e-06
[py_interpret_to_execute_after_opt_a]: 7.82998e-06
[rewriter_after_opt_a]: 2.324e-05
[convert_after_rewriter]: 5.11997e-06
[order_py_execute_after_rewriter]: 3.30998e-06
[mutable_eliminate]: 0.00047206
[jit_opt_b]: 3.906e-05, [1]
    [Cycle 1]: 3.252e-05, [2]
        [frontend_op_eliminate]: 1.026e-05
        [inline_after_opt_a]: 1.099e-05
[cconv]: 1.522e-05
[loop_unroll]: 0.00040695
[jit_opt_after_cconv]: 0.00011418, [1]
    [Cycle 1]: 0.00010769, [11]
        [c_1]: 1.124e-05
        [parameter_eliminate]: 2.59001e-06
        [updatestate_depend_eliminate]: 5.37001e-06
        [updatestate_assign_eliminate]: 1.86003e-06
        [updatestate_loads_eliminate]: 1.55001e-06
        [cse]: 1.529e-05
        [call_graph_tuple_transform]: 1.127e-05
        [tuple_list_get_item_eliminator]: 3.56999e-06
        [none_parameter_eliminate]: 8.29983e-07
        [renormalize]: 4.20026e-07
        [switch_simplify]: 3.65e-06
[remove_dup_value]: 7.6e-06
[partial_unused_args_eliminate]: 1.39e-06
[environ_conv]: 3.41999e-06
[add_recomputation]: 2.188e-05
[cse_after_recomputation]: 1.896e-05, [1]
    [Cycle 1]: 1.404e-05, [1]
        [cse]: 8.37998e-06
[auto_monad_reorder]: 7.71001e-06
[get_jit_bprop_graph]: 1.35999e-06
[rewriter_after_jit_bprop_graph]: 4.59002e-06
[opt_after_jit_grad]: 0.00044151
[symbol_engine_optimizer]: 6.063e-05, [1]
    [Cycle 1]: 5.479e-05, [6]
        [build]: 2.27001e-06
        [elim_shapecalc]: 5.64998e-06
        [elim_not_effective]: 9.15001e-06
        [opt_reshape]: 4.29997e-06
        [fold_const_symbol]: 5.40999e-06
        [renormalize]: 5.70028e-07
[validate]: 2.157e-05
[backend_pass]: 9.10019e-07
[task_emit]: 2.468e-05
[execute]: 1.39998e-06
Sums
    bootstrap                            :     0.000228s :  1.32%
    type_inference                       :     0.010844s : 62.74%
    event_method                         :     0.000045s :  0.26%
    auto_monad                           :     0.000093s :  0.54%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000004s :  0.02%
    py_interpret_to_execute              :     0.000027s :  0.15%
    rewriter_before_opt_a                :     0.000091s :  0.52%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000067s :  0.39%
    jit_opt_a.loop_unroll                :     0.000042s :  0.24%
    jit_opt_a.a_1                        :     0.000855s :  4.95%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000016s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000016s :  0.09%
    jit_opt_a.merge_forward              :     0.000010s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001396s :  8.08%
    jit_opt_a.replace_old_param          :     0.000055s :  0.32%
    jit_opt_a.inline_without_move        :     0.000054s :  0.31%
    jit_opt_a.renormalize                :     0.001596s :  9.23%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.08%
    jit_opt_a.cse                        :     0.000029s :  0.17%
    jit_opt_a.replace_applicator         :     0.000015s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000023s :  0.13%
    convert_after_rewriter               :     0.000005s :  0.03%
    order_py_execute_after_rewriter      :     0.000003s :  0.02%
    mutable_eliminate                    :     0.000472s :  2.73%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.06%
    cconv                                :     0.000015s :  0.09%
    loop_unroll                          :     0.000407s :  2.35%
    jit_opt_after_cconv.c_1              :     0.000011s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000011s :  0.07%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000008s :  0.04%
    partial_unused_args_eliminate        :     0.000001s :  0.01%
    environ_conv                         :     0.000003s :  0.02%
    add_recomputation                    :     0.000022s :  0.13%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000008s :  0.04%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000442s :  2.55%
    symbol_engine_optimizer.build        :     0.000002s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000005s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000022s :  0.12%
    backend_pass                         :     0.000001s :  0.01%
    task_emit                            :     0.000025s :  0.14%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000234    50
 0.62% :     0.000001s :      1: substitution.elim_not_effective
 0.38% :     0.000001s :      1: substitution.fold_const_symbol
 1.48% :     0.000003s :      1: substitution.graph_param_transform
66.05% :     0.000155s :      6: substitution.inline
 6.40% :     0.000015s :      2: substitution.inline_without_move
 2.83% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.55% :     0.000004s :      2: substitution.minmaximum_grad
 1.81% :     0.000004s :      3: substitution.partial_eliminate
 2.98% :     0.000007s :      9: substitution.remove_not_recompute_node
 1.81% :     0.000004s :      2: substitution.replace_applicator
 1.51% :     0.000004s :      5: substitution.replace_old_param
 0.98% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 3.00% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.98% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 6.61% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008511     2
88.35% :     0.007520s :      1: type_inference.infer
11.65% :     0.000991s :      1: type_inference.specialize
------[replace.]   0.000060     8
79.54% :     0.000048s :      6: replace.inline
20.46% :     0.000012s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000162     8
93.32% :     0.000151s :      6: match.inline
 6.68% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000167  1109
 1.46% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.61% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.33% :     0.000002s :     18: predicate.addn_check_dump
 1.52% :     0.000003s :     18: predicate.addn_zero_filter
 2.09% :     0.000003s :     18: predicate.arithmetic_simplify
 1.44% :     0.000002s :     18: predicate.cast_eliminate
 0.18% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.32% :     0.000002s :     18: predicate.compare_switch_simplify
 1.43% :     0.000002s :     18: predicate.depend_value_elim
 1.30% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.40% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.48% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.14% :     0.000000s :      1: predicate.elim_not_effective
 0.22% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.42% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.36% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.50% :     0.000003s :     18: predicate.environ_get_eliminate
 1.34% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.07% :     0.000000s :      1: predicate.fold_const_symbol
 1.28% :     0.000002s :     12: predicate.get_grad_eliminate
 0.11% :     0.000000s :      1: predicate.graph_param_transform
 4.58% :     0.000008s :     28: predicate.inline
 3.32% :     0.000006s :     33: predicate.inline_without_move
 0.51% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :     12: predicate.less_batch_normalization
 1.50% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.72% :     0.000003s :     21: predicate.load_eliminater
 0.78% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.63% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.80% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.32% :     0.000002s :     18: predicate.merge_addn
 1.35% :     0.000002s :     18: predicate.minmaximum_grad
 0.75% :     0.000001s :      1: predicate.mutable_eliminate
 0.29% :     0.000000s :      1: predicate.opt_reshape
 2.09% :     0.000003s :     21: predicate.partial_eliminate
 1.88% :     0.000003s :     18: predicate.print_const_string_wrapper
 1.70% :     0.000003s :     18: predicate.reduce_eliminate
 1.53% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.52% :     0.000003s :     22: predicate.replace_applicator
 1.71% :     0.000003s :     33: predicate.replace_old_param
 0.13% :     0.000000s :      1: predicate.reset_defer_inline
 1.71% :     0.000003s :     18: predicate.reshape_eliminate
 1.39% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.37% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.45% :     0.000002s :     18: predicate.same_eliminate
 0.65% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.40% :     0.000001s :      2: predicate.special_op_eliminate
 1.21% :     0.000002s :     12: predicate.specialize_transform
 1.48% :     0.000002s :     18: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.13% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.63% :     0.000004s :     26: predicate.switch_defer_inline
 2.38% :     0.000004s :     26: predicate.switch_layer_defer_inline
 7.14% :     0.000012s :     71: predicate.switch_simplify
 1.40% :     0.000002s :     18: predicate.tile_eliminate
 1.34% :     0.000002s :     18: predicate.transpose_eliminate
 1.67% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.53% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.23% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.58% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.08% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000003s :     18: predicate.value_based_eliminate
 0.13% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.26% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001028    22
73.61% :     0.000757s :     13: func_graph_cloner_run.FuncGraphClonerGraph
26.39% :     0.000271s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021234    72
 0.17% :     0.000037s :      1: add_recomputation
 0.36% :     0.000077s :      1: auto_monad
 0.07% :     0.000015s :      1: auto_monad_reorder
 1.03% :     0.000219s :      1: bootstrap
 0.11% :     0.000023s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000021s :      1: cse_after_recomputation
 0.04% :     0.000008s :      1: environ_conv
 0.22% :     0.000047s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
35.92% :     0.007628s :      1: jit_opt_a
 0.56% :     0.000120s :      1: jit_opt_after_cconv
 0.20% :     0.000042s :      1: jit_opt_b
 2.03% :     0.000431s :      1: loop_unroll
 2.32% :     0.000492s :      1: mutable_eliminate
 5.68% :     0.001205s :     26: opt.transform.jit_opt_a
 0.14% :     0.000030s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000015s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000022s :      4: opt.transform.symbol_engine_opt
 2.17% :     0.000461s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.13% :     0.000029s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000018s :      1: remove_dup_value
 5.18% :     0.001101s :      1: renormalize.infer
 1.53% :     0.000325s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000034s :      1: rewriter_after_opt_a
 0.45% :     0.000095s :      1: rewriter_before_opt_a
 0.31% :     0.000065s :      1: symbol_engine_optimizer
40.41% :     0.008581s :      1: type_inference


Time group info:
------[substitution.]   0.000235    50
 0.38% :     0.000001s :      1: substitution.elim_not_effective
 0.28% :     0.000001s :      1: substitution.fold_const_symbol
 0.94% :     0.000002s :      1: substitution.graph_param_transform
68.15% :     0.000160s :      6: substitution.inline
 6.78% :     0.000016s :      2: substitution.inline_without_move
 2.72% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.36% :     0.000003s :      2: substitution.minmaximum_grad
 1.26% :     0.000003s :      3: substitution.partial_eliminate
 2.83% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.21% :     0.000005s :      2: substitution.replace_applicator
 1.81% :     0.000004s :      5: substitution.replace_old_param
 1.02% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.20% :     0.000005s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.41% :     0.000003s :      2: substitution.tuple_list_get_item_depend_reorder
 6.64% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010771     2
90.38% :     0.009734s :      1: type_inference.infer
 9.62% :     0.001036s :      1: type_inference.specialize
------[replace.]   0.000062     8
77.53% :     0.000048s :      6: replace.inline
22.47% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000168     8
93.45% :     0.000157s :      6: match.inline
 6.55% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000166  1109
 1.56% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.66% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.29% :     0.000002s :     18: predicate.addn_check_dump
 1.71% :     0.000003s :     18: predicate.addn_zero_filter
 2.10% :     0.000004s :     18: predicate.arithmetic_simplify
 1.45% :     0.000002s :     18: predicate.cast_eliminate
 0.14% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.36% :     0.000002s :     18: predicate.compare_switch_simplify
 1.47% :     0.000002s :     18: predicate.depend_value_elim
 1.24% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.45% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.55% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      1: predicate.elim_not_effective
 0.19% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.27% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.50% :     0.000002s :     18: predicate.environ_get_eliminate
 1.36% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.32% :     0.000002s :     12: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 4.17% :     0.000007s :     28: predicate.inline
 3.29% :     0.000005s :     33: predicate.inline_without_move
 0.50% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :     12: predicate.less_batch_normalization
 1.59% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.89% :     0.000003s :     21: predicate.load_eliminater
 0.74% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.59% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.65% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.34% :     0.000002s :     18: predicate.merge_addn
 1.35% :     0.000002s :     18: predicate.minmaximum_grad
 0.88% :     0.000001s :      1: predicate.mutable_eliminate
 0.25% :     0.000000s :      1: predicate.opt_reshape
 1.97% :     0.000003s :     21: predicate.partial_eliminate
 1.46% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.92% :     0.000003s :     18: predicate.reduce_eliminate
 1.54% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.50% :     0.000002s :     22: predicate.replace_applicator
 1.67% :     0.000003s :     33: predicate.replace_old_param
 0.10% :     0.000000s :      1: predicate.reset_defer_inline
 1.53% :     0.000003s :     18: predicate.reshape_eliminate
 1.41% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.68% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.44% :     0.000002s :     18: predicate.same_eliminate
 0.62% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.26% :     0.000000s :      2: predicate.special_op_eliminate
 1.17% :     0.000002s :     12: predicate.specialize_transform
 1.63% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.57% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.14% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.40% :     0.000004s :     26: predicate.switch_defer_inline
 2.32% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.91% :     0.000011s :     71: predicate.switch_simplify
 1.42% :     0.000002s :     18: predicate.tile_eliminate
 1.41% :     0.000002s :     18: predicate.transpose_eliminate
 1.74% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.45% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 3.41% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.80% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.60% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000003s :     18: predicate.value_based_eliminate
 0.14% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.32% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001126    22
75.05% :     0.000845s :     13: func_graph_cloner_run.FuncGraphClonerGraph
24.95% :     0.000281s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.027103    76
 0.09% :     0.000025s :      1: add_recomputation
 0.36% :     0.000097s :      1: auto_monad
 0.04% :     0.000010s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.89% :     0.000240s :      1: bootstrap
 0.07% :     0.000018s :      1: cconv
 0.03% :     0.000007s :      1: convert_after_rewriter
 0.08% :     0.000021s :      1: cse_after_recomputation
 0.02% :     0.000006s :      1: environ_conv
 0.18% :     0.000049s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000003s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
40.71% :     0.011033s :      1: jit_opt_a
 0.43% :     0.000117s :      1: jit_opt_after_cconv
 0.15% :     0.000042s :      1: jit_opt_b
 1.53% :     0.000415s :      1: loop_unroll
 1.77% :     0.000481s :      1: mutable_eliminate
 4.46% :     0.001209s :     26: opt.transform.jit_opt_a
 0.10% :     0.000027s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000015s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000015s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000021s :      4: opt.transform.symbol_engine_opt
 1.66% :     0.000449s :      1: opt_after_jit_grad
 0.02% :     0.000005s :      1: order_py_execute_after_rewriter
 0.01% :     0.000003s :      1: partial_unused_args_eliminate
 0.02% :     0.000006s :      1: pre_auto_parallel
 0.11% :     0.000029s :      1: py_interpret_to_execute
 0.04% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000010s :      1: remove_dup_value
 4.31% :     0.001168s :      1: renormalize.infer
 1.55% :     0.000419s :      1: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.10% :     0.000026s :      1: rewriter_after_opt_a
 0.34% :     0.000093s :      1: rewriter_before_opt_a
 0.23% :     0.000063s :      1: symbol_engine_optimizer
 0.11% :     0.000029s :      1: task_emit
40.06% :     0.010857s :      1: type_inference
 0.13% :     0.000035s :      1: validate

[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:22.917.50 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c382670> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:22.923.55 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.931.93 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d15e0> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.937.24 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0305228, [33]
[bootstrap]: 0.00018596
[type_inference]: 0.00910521
[event_method]: 4.32e-05
[auto_monad]: 6.405e-05
[graph_reusing]: 5.06002e-06
[pre_auto_parallel]: 2.07001e-06
[py_interpret_to_execute]: 2.292e-05
[rewriter_before_opt_a]: 8.971e-05
[expand_dump_flag]: 2.07001e-06
[jit_opt_a]: 0.0108189, [3]
    [Cycle 1]: 0.00518439, [27]
        [switch_simplify]: 5.514e-05
        [loop_unroll]: 3.772e-05
        [a_1]: 0.00078129
        [with_stream_mark]: 1.805e-05
        [recompute_prepare]: 1.71e-05
        [updatestate_depend_eliminate]: 6.89999e-06
        [updatestate_assign_eliminate]: 5.72999e-06
        [updatestate_loads_eliminate]: 5.32999e-06
        [parameter_eliminate]: 1.74e-06
        [specialize_transform]: 1.233e-05
        [updatestate_useless_node_eliminater]: 1.147e-05
        [accelerated_algorithm]: 1.201e-05
        [meta_shard_fg_expand]: 2.36e-06
        [get_grad_eliminate_]: 1.162e-05
        [merge_forward]: 6.57002e-06
        [cell_reuse_recompute_pass]: 7.50006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.29e-05
        [j_node_and_user_rematch]: 2.117e-05
        [meta_fg_expand]: 0.00124438
        [replace_old_param]: 4.823e-05
        [inline_without_move]: 4.905e-05
        [renormalize]: 0.00235157
        [add_forward_monad_depend]: 8.98002e-06
        [auto_monad_grad]: 4.78001e-06
        [auto_monad_eliminator]: 4.717e-05
        [cse]: 0.00014379
        [replace_applicator]: 6.424e-05
    [Cycle 2]: 0.0022014, [27]
        [switch_simplify]: 3.762e-05
        [loop_unroll]: 3.555e-05
        [a_1]: 0.00109618
        [with_stream_mark]: 1.17e-05
        [recompute_prepare]: 9.56e-06
        [updatestate_depend_eliminate]: 4.58001e-06
        [updatestate_assign_eliminate]: 4.17e-06
        [updatestate_loads_eliminate]: 3.74002e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 8.65001e-06
        [updatestate_useless_node_eliminater]: 8.48001e-06
        [accelerated_algorithm]: 8.53001e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 8.82e-06
        [merge_forward]: 4.22e-06
        [cell_reuse_recompute_pass]: 1.03001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.739e-05
        [j_node_and_user_rematch]: 1.34e-05
        [meta_fg_expand]: 0.00010032
        [replace_old_param]: 1.296e-05
        [inline_without_move]: 8.94e-06
        [renormalize]: 0.00057132
        [add_forward_monad_depend]: 4.43999e-06
        [auto_monad_grad]: 1.24e-06
        [auto_monad_eliminator]: 1.466e-05
        [cse]: 4.501e-05
        [replace_applicator]: 1.489e-05
    [Cycle 3]: 0.00047701, [27]
        [switch_simplify]: 8.57e-06
        [loop_unroll]: 7.77998e-06
        [a_1]: 0.00018567
        [with_stream_mark]: 1.024e-05
        [recompute_prepare]: 8.00999e-06
        [updatestate_depend_eliminate]: 4.44002e-06
        [updatestate_assign_eliminate]: 4.55001e-06
        [updatestate_loads_eliminate]: 3.59002e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 7.93001e-06
        [updatestate_useless_node_eliminater]: 7.58999e-06
        [accelerated_algorithm]: 8.90001e-06
        [meta_shard_fg_expand]: 1.50999e-06
        [get_grad_eliminate_]: 7.54002e-06
        [merge_forward]: 4.31002e-06
        [cell_reuse_recompute_pass]: 1.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.623e-05
        [j_node_and_user_rematch]: 1.352e-05
        [meta_fg_expand]: 2.74001e-06
        [replace_old_param]: 9.88998e-06
        [inline_without_move]: 7.68999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.12999e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 1.027e-05
        [cse]: 2.26e-05
        [replace_applicator]: 7.88001e-06
[py_interpret_to_execute_after_opt_a]: 1.176e-05
[rewriter_after_opt_a]: 4.199e-05
[convert_after_rewriter]: 8.44998e-06
[order_py_execute_after_rewriter]: 5.44e-06
[mutable_eliminate]: 0.00056281
[jit_opt_b]: 6.522e-05, [1]
    [Cycle 1]: 5.814e-05, [2]
        [frontend_op_eliminate]: 2.427e-05
        [inline_after_opt_a]: 2.277e-05
[cconv]: 1.812e-05
[loop_unroll]: 0.0004054
[jit_opt_after_cconv]: 0.00018409, [1]
    [Cycle 1]: 0.00017765, [11]
        [c_1]: 3.43e-05
        [parameter_eliminate]: 2.84001e-06
        [updatestate_depend_eliminate]: 6.94999e-06
        [updatestate_assign_eliminate]: 3.91999e-06
        [updatestate_loads_eliminate]: 3.58e-06
        [cse]: 3.196e-05
        [call_graph_tuple_transform]: 2.484e-05
        [tuple_list_get_item_eliminator]: 8.39002e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 5.69999e-07
        [switch_simplify]: 8.77999e-06
[remove_dup_value]: 1.85e-05
[partial_unused_args_eliminate]: 1.92001e-06
[environ_conv]: 6.09999e-06
[add_recomputation]: 4.747e-05
[cse_after_recomputation]: 3.382e-05, [1]
    [Cycle 1]: 2.772e-05, [1]
        [cse]: 2.182e-05
[auto_monad_reorder]: 1.767e-05
[get_jit_bprop_graph]: 1.58002e-06
[rewriter_after_jit_bprop_graph]: 5.37001e-06
[opt_after_jit_grad]: 0.00045232
[symbol_engine_optimizer]: 8.902e-05, [1]
    [Cycle 1]: 8.296e-05, [6]
        [build]: 3.88999e-06
        [elim_shapecalc]: 1.059e-05
        [elim_not_effective]: 1.831e-05
        [opt_reshape]: 8.70001e-06
        [fold_const_symbol]: 1.417e-05
        [renormalize]: 4.39992e-07
[validate]: 4.089e-05
[backend_pass]: 6.89994e-07
[task_emit]: 0.00796795
[execute]: 7.29001e-06
Sums
    bootstrap                            :     0.000186s :  0.69%
    type_inference                       :     0.009105s : 33.97%
    event_method                         :     0.000043s :  0.16%
    auto_monad                           :     0.000064s :  0.24%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000023s :  0.09%
    rewriter_before_opt_a                :     0.000090s :  0.33%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000101s :  0.38%
    jit_opt_a.loop_unroll                :     0.000081s :  0.30%
    jit_opt_a.a_1                        :     0.002063s :  7.70%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000035s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000014s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000013s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000029s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000028s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000029s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000028s :  0.10%
    jit_opt_a.merge_forward              :     0.000015s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000057s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000048s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001347s :  5.03%
    jit_opt_a.replace_old_param          :     0.000071s :  0.27%
    jit_opt_a.inline_without_move        :     0.000066s :  0.25%
    jit_opt_a.renormalize                :     0.002923s : 10.91%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000072s :  0.27%
    jit_opt_a.cse                        :     0.000211s :  0.79%
    jit_opt_a.replace_applicator         :     0.000087s :  0.32%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.04%
    rewriter_after_opt_a                 :     0.000042s :  0.16%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000005s :  0.02%
    mutable_eliminate                    :     0.000563s :  2.10%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000023s :  0.08%
    cconv                                :     0.000018s :  0.07%
    loop_unroll                          :     0.000405s :  1.51%
    jit_opt_after_cconv.c_1              :     0.000034s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000032s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.03%
    remove_dup_value                     :     0.000018s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000047s :  0.18%
    cse_after_recomputation.cse          :     0.000022s :  0.08%
    auto_monad_reorder                   :     0.000018s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000452s :  1.69%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000041s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007968s : 29.73%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000465   126
 7.12% :     0.000033s :      2: substitution.arithmetic_simplify
 0.62% :     0.000003s :      5: substitution.elim_not_effective
 0.46% :     0.000002s :      5: substitution.fold_const_symbol
 1.36% :     0.000006s :      6: substitution.graph_param_transform
59.20% :     0.000275s :     12: substitution.inline
 3.32% :     0.000015s :      2: substitution.inline_without_move
 1.86% :     0.000009s :     18: substitution.j_node_and_user_rematch
 1.87% :     0.000009s :      7: substitution.minmaximum_grad
 1.28% :     0.000006s :      5: substitution.partial_eliminate
 2.66% :     0.000012s :     18: substitution.remove_not_recompute_node
 4.73% :     0.000022s :     10: substitution.replace_applicator
 1.46% :     0.000007s :      7: substitution.replace_old_param
 0.44% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 3.78% :     0.000018s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.74% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 7.11% :     0.000033s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009051     2
89.03% :     0.008058s :      1: type_inference.infer
10.97% :     0.000993s :      1: type_inference.specialize
------[replace.]   0.000134    19
64.22% :     0.000086s :     12: replace.inline
35.78% :     0.000048s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000284    19
94.66% :     0.000269s :     12: match.inline
 5.34% :     0.000015s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000398  2895
 1.63% :     0.000006s :     48: predicate.accumulaten_eliminater
 0.52% :     0.000002s :      6: predicate.ad_related_special_op_eliminate
 1.46% :     0.000006s :     48: predicate.addn_check_dump
 1.74% :     0.000007s :     48: predicate.addn_zero_filter
 1.98% :     0.000008s :     48: predicate.arithmetic_simplify
 1.60% :     0.000006s :     48: predicate.cast_eliminate
 0.26% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.43% :     0.000006s :     48: predicate.compare_switch_simplify
 1.69% :     0.000007s :     48: predicate.depend_value_elim
 1.48% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.54% :     0.000006s :     48: predicate.dict_get_item_eliminator
 1.52% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.36% :     0.000001s :      6: predicate.dumpgradient_eliminate
 0.15% :     0.000001s :      6: predicate.elim_not_effective
 0.25% :     0.000001s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.50% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.47% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.46% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.55% :     0.000006s :     48: predicate.environ_get_eliminate
 1.48% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.12% :     0.000000s :      6: predicate.fold_const_symbol
 0.97% :     0.000004s :     23: predicate.get_grad_eliminate
 0.14% :     0.000001s :      6: predicate.graph_param_transform
 4.20% :     0.000017s :     79: predicate.inline
 1.79% :     0.000007s :     44: predicate.inline_without_move
 0.39% :     0.000002s :     23: predicate.j_node_and_user_rematch
 0.98% :     0.000004s :     23: predicate.less_batch_normalization
 1.78% :     0.000007s :     55: predicate.list_to_tuple_eliminator_
 1.98% :     0.000008s :     61: predicate.load_eliminater
 0.62% :     0.000002s :      6: predicate.loop_unroll_after_grad
 3.08% :     0.000012s :     90: predicate.loop_unroll_before_grad
 1.81% :     0.000007s :     54: predicate.make_slice_get_slice_eliminator
 1.52% :     0.000006s :     48: predicate.merge_addn
 1.48% :     0.000006s :     48: predicate.minmaximum_grad
 0.61% :     0.000002s :      6: predicate.mutable_eliminate
 0.24% :     0.000001s :      6: predicate.opt_reshape
 2.31% :     0.000009s :     61: predicate.partial_eliminate
 1.46% :     0.000006s :     48: predicate.print_const_string_wrapper
 1.80% :     0.000007s :     48: predicate.reduce_eliminate
 1.81% :     0.000007s :     55: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.49% :     0.000010s :    108: predicate.replace_applicator
 0.91% :     0.000004s :     44: predicate.replace_old_param
 0.16% :     0.000001s :      6: predicate.reset_defer_inline
 1.67% :     0.000007s :     48: predicate.reshape_eliminate
 1.52% :     0.000006s :     48: predicate.row_tensor_add_zeros_like
 0.39% :     0.000002s :      6: predicate.row_tensor_eliminate
 1.50% :     0.000006s :     48: predicate.same_eliminate
 0.49% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.52% :     0.000002s :     12: predicate.special_op_eliminate
 0.94% :     0.000004s :     23: predicate.specialize_transform
 1.78% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000006s :     48: predicate.stack_unstack_eliminate
 0.24% :     0.000001s :      6: predicate.switch_call_monad_eliminater
 2.54% :     0.000010s :     67: predicate.switch_defer_inline
 2.40% :     0.000010s :     67: predicate.switch_layer_defer_inline
 5.92% :     0.000024s :    163: predicate.switch_simplify
 1.57% :     0.000006s :     48: predicate.tile_eliminate
 1.49% :     0.000006s :     48: predicate.transpose_eliminate
 1.81% :     0.000007s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.62% :     0.000006s :     48: predicate.tuple_list_get_item_depend_reorder
 3.09% :     0.000012s :     67: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000007s :     48: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000007s :     55: predicate.tuple_to_list_eliminator_
 1.94% :     0.000008s :     61: predicate.updatestate_pure_node_eliminater
 3.03% :     0.000012s :     84: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000007s :     48: predicate.value_based_eliminate
 0.21% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.29% :     0.000001s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001360    29
59.03% :     0.000803s :     13: func_graph_cloner_run.FuncGraphClonerGraph
40.97% :     0.000557s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.036311    91
 0.14% :     0.000051s :      1: add_recomputation
 0.19% :     0.000068s :      1: auto_monad
 0.06% :     0.000020s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.54% :     0.000198s :      1: bootstrap
 0.06% :     0.000021s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000036s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.13% :     0.000047s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000004s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
29.80% :     0.010822s :      1: jit_opt_a
 0.51% :     0.000187s :      1: jit_opt_after_cconv
 0.19% :     0.000068s :      1: jit_opt_b
 1.14% :     0.000413s :      1: loop_unroll
 1.57% :     0.000571s :      1: mutable_eliminate
 7.39% :     0.002684s :     39: opt.transform.jit_opt_a
 0.20% :     0.000073s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000041s :      4: opt.transform.jit_opt_b
 0.05% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000049s :      4: opt.transform.symbol_engine_opt
 1.27% :     0.000461s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000026s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000021s :      1: remove_dup_value
 4.36% :     0.001584s :      2: renormalize.infer
 3.66% :     0.001328s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000045s :      1: rewriter_after_opt_a
 0.26% :     0.000093s :      1: rewriter_before_opt_a
 0.25% :     0.000091s :      1: symbol_engine_optimizer
21.98% :     0.007980s :      1: task_emit
25.11% :     0.009117s :      1: type_inference
 0.18% :     0.000065s :      1: validate


TotalTime = 0.012431, [30]
[bootstrap]: 0.00029095
[type_inference]: 0.00433113
[event_method]: 1.745e-05
[auto_monad]: 6.547e-05
[graph_reusing]: 6.26e-06
[pre_auto_parallel]: 2.49001e-06
[py_interpret_to_execute]: 3.352e-05
[rewriter_before_opt_a]: 7.353e-05
[expand_dump_flag]: 3.14999e-06
[jit_opt_a]: 0.00543649, [2]
    [Cycle 1]: 0.00164675, [27]
        [switch_simplify]: 5.553e-05
        [loop_unroll]: 2.808e-05
        [a_1]: 0.00054247
        [with_stream_mark]: 2.249e-05
        [recompute_prepare]: 7.27002e-06
        [updatestate_depend_eliminate]: 3.69002e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 5.60001e-06
        [updatestate_useless_node_eliminater]: 4.86002e-06
        [accelerated_algorithm]: 5.49e-06
        [meta_shard_fg_expand]: 2.23998e-06
        [get_grad_eliminate_]: 5.03002e-06
        [merge_forward]: 4.13001e-06
        [cell_reuse_recompute_pass]: 1.01002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.523e-05
        [j_node_and_user_rematch]: 8.99e-06
        [meta_fg_expand]: 2.08998e-06
        [replace_old_param]: 8.92999e-06
        [inline_without_move]: 5.52999e-06
        [renormalize]: 0.0006603
        [add_forward_monad_depend]: 6.57002e-06
        [auto_monad_grad]: 2.60002e-06
        [auto_monad_eliminator]: 1.476e-05
        [cse]: 3.1e-05
        [replace_applicator]: 1.073e-05
    [Cycle 2]: 0.00034061, [27]
        [switch_simplify]: 5.69e-06
        [loop_unroll]: 4.77e-06
        [a_1]: 8.759e-05
        [with_stream_mark]: 1.087e-05
        [recompute_prepare]: 4.73001e-06
        [updatestate_depend_eliminate]: 2.78e-06
        [updatestate_assign_eliminate]: 2.08002e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 5.10999e-06
        [updatestate_useless_node_eliminater]: 5.13002e-06
        [accelerated_algorithm]: 4.94998e-06
        [meta_shard_fg_expand]: 1.47001e-06
        [get_grad_eliminate_]: 4.56002e-06
        [merge_forward]: 2.48e-06
        [cell_reuse_recompute_pass]: 1.47999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.322e-05
        [j_node_and_user_rematch]: 7.60998e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 7.06001e-06
        [inline_without_move]: 4.92999e-06
        [renormalize]: 2.60014e-07
        [add_forward_monad_depend]: 1.42e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 7e-06
        [cse]: 1.152e-05
        [replace_applicator]: 5.71e-06
[py_interpret_to_execute_after_opt_a]: 1.13e-05
[rewriter_after_opt_a]: 3.659e-05
[convert_after_rewriter]: 6.38e-06
[order_py_execute_after_rewriter]: 5.17e-06
[mutable_eliminate]: 0.0005747
[jit_opt_b]: 5.134e-05, [1]
    [Cycle 1]: 4.396e-05, [2]
        [frontend_op_eliminate]: 1.668e-05
        [inline_after_opt_a]: 1.533e-05
[cconv]: 2.921e-05
[loop_unroll]: 0.00043162
[jit_opt_after_cconv]: 0.00014719, [1]
    [Cycle 1]: 0.0001411, [11]
        [c_1]: 2.166e-05
        [parameter_eliminate]: 3.50998e-06
        [updatestate_depend_eliminate]: 6.48e-06
        [updatestate_assign_eliminate]: 2.76e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [cse]: 2.071e-05
        [call_graph_tuple_transform]: 1.987e-05
        [tuple_list_get_item_eliminator]: 5.11002e-06
        [none_parameter_eliminate]: 1.59e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 5.09e-06
[remove_dup_value]: 1.549e-05
[partial_unused_args_eliminate]: 2.87002e-06
[environ_conv]: 5.57001e-06
[add_recomputation]: 5.405e-05
[cse_after_recomputation]: 2.057e-05, [1]
    [Cycle 1]: 1.465e-05, [1]
        [cse]: 8.44002e-06
[auto_monad_reorder]: 1.505e-05
[get_jit_bprop_graph]: 2.24999e-06
[rewriter_after_jit_bprop_graph]: 5.81998e-06
[opt_after_jit_grad]: 0.00046776
[symbol_engine_optimizer]: 7.134e-05, [1]
    [Cycle 1]: 6.488e-05, [6]
        [build]: 3.99002e-06
        [elim_shapecalc]: 7.14001e-06
        [elim_not_effective]: 1.208e-05
        [opt_reshape]: 5.71e-06
        [fold_const_symbol]: 8.31002e-06
        [renormalize]: 4.7998e-07
[validate]: 3.732e-05
Sums
    bootstrap                            :     0.000291s :  3.48%
    type_inference                       :     0.004331s : 51.79%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000065s :  0.78%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.40%
    rewriter_before_opt_a                :     0.000074s :  0.88%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000061s :  0.73%
    jit_opt_a.loop_unroll                :     0.000033s :  0.39%
    jit_opt_a.a_1                        :     0.000630s :  7.53%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.40%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000661s :  7.90%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.26%
    jit_opt_a.cse                        :     0.000043s :  0.51%
    jit_opt_a.replace_applicator         :     0.000016s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.14%
    rewriter_after_opt_a                 :     0.000037s :  0.44%
    convert_after_rewriter               :     0.000006s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000575s :  6.87%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000029s :  0.35%
    loop_unroll                          :     0.000432s :  5.16%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000021s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.19%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000054s :  0.65%
    cse_after_recomputation.cse          :     0.000008s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000468s :  5.59%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000037s :  0.45%


Time group info:
------[substitution.]   0.000225    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.61% :     0.000006s :      3: substitution.graph_param_transform
84.94% :     0.000191s :      4: substitution.inline
 1.49% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.26% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000004s :      2: substitution.replace_old_param
 5.38% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004261     2
81.48% :     0.003472s :      1: type_inference.infer
18.52% :     0.000789s :      1: type_inference.specialize
------[replace.]   0.000064     6
67.20% :     0.000043s :      4: replace.inline
32.80% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000200     6
94.45% :     0.000189s :      4: match.inline
 5.55% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.61% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.55% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000001s :     10: predicate.addn_zero_filter
 2.04% :     0.000002s :     10: predicate.arithmetic_simplify
 1.19% :     0.000001s :     10: predicate.cast_eliminate
 0.73% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.54% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.02% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000000s :      3: predicate.graph_param_transform
 5.55% :     0.000007s :     22: predicate.inline
 1.11% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 1.67% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.81% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 1.99% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.53% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.61% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.47% :     0.000001s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.11% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.44% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.38% :     0.000002s :      6: predicate.special_op_eliminate
 1.10% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.44% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.86% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.29% :     0.000003s :     16: predicate.switch_defer_inline
 2.02% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.38% :     0.000009s :     53: predicate.switch_simplify
 1.08% :     0.000001s :     10: predicate.tile_eliminate
 1.08% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.48% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000575    12
51.13% :     0.000294s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.94% :     0.000046s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.93% :     0.000235s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013978    72
 0.41% :     0.000057s :      1: add_recomputation
 0.49% :     0.000069s :      1: auto_monad
 0.12% :     0.000017s :      1: auto_monad_reorder
 2.22% :     0.000311s :      1: bootstrap
 0.23% :     0.000032s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.15% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.91% :     0.005440s :      1: jit_opt_a
 1.07% :     0.000150s :      1: jit_opt_after_cconv
 0.39% :     0.000054s :      1: jit_opt_b
 3.14% :     0.000439s :      1: loop_unroll
 4.18% :     0.000585s :      1: mutable_eliminate
 5.99% :     0.000837s :     26: opt.transform.jit_opt_a
 0.34% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.41% :     0.000477s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.04% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000036s :      1: py_interpret_to_execute
 0.10% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000018s :      1: remove_dup_value
 2.59% :     0.000362s :      1: renormalize.infer
 2.07% :     0.000289s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000040s :      1: rewriter_after_opt_a
 0.54% :     0.000076s :      1: rewriter_before_opt_a
 0.53% :     0.000074s :      1: symbol_engine_optimizer
31.10% :     0.004348s :      1: type_inference

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.105.439 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80caaf0> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.106.039 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0187021, [30]
[bootstrap]: 0.00017485
[type_inference]: 0.0087149
[event_method]: 4.191e-05
[auto_monad]: 9.06e-05
[graph_reusing]: 6.92002e-06
[pre_auto_parallel]: 3.28e-06
[py_interpret_to_execute]: 2.568e-05
[rewriter_before_opt_a]: 9.275e-05
[expand_dump_flag]: 3.16001e-06
[jit_opt_a]: 0.00761831, [2]
    [Cycle 1]: 0.00422773, [27]
        [switch_simplify]: 5.94e-05
        [loop_unroll]: 3.825e-05
        [a_1]: 0.00079915
        [with_stream_mark]: 2.455e-05
        [recompute_prepare]: 1.804e-05
        [updatestate_depend_eliminate]: 7.68999e-06
        [updatestate_assign_eliminate]: 6.48998e-06
        [updatestate_loads_eliminate]: 6.06e-06
        [parameter_eliminate]: 2.96999e-06
        [specialize_transform]: 1.361e-05
        [updatestate_useless_node_eliminater]: 1.156e-05
        [accelerated_algorithm]: 1.195e-05
        [meta_shard_fg_expand]: 3.46999e-06
        [get_grad_eliminate_]: 1.173e-05
        [merge_forward]: 8.08001e-06
        [cell_reuse_recompute_pass]: 9.20001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.525e-05
        [j_node_and_user_rematch]: 2.242e-05
        [meta_fg_expand]: 0.00139636
        [replace_old_param]: 5.044e-05
        [inline_without_move]: 5.084e-05
        [renormalize]: 0.00140941
        [add_forward_monad_depend]: 5.56e-06
        [auto_monad_grad]: 2.88e-06
        [auto_monad_eliminator]: 1.206e-05
        [cse]: 2.606e-05
        [replace_applicator]: 1.068e-05
    [Cycle 2]: 0.00026931, [27]
        [switch_simplify]: 4.28001e-06
        [loop_unroll]: 3.84002e-06
        [a_1]: 3.682e-05
        [with_stream_mark]: 8.31002e-06
        [recompute_prepare]: 4.18999e-06
        [updatestate_depend_eliminate]: 2.37999e-06
        [updatestate_assign_eliminate]: 1.81e-06
        [updatestate_loads_eliminate]: 1.54998e-06
        [parameter_eliminate]: 1.22e-06
        [specialize_transform]: 3.66001e-06
        [updatestate_useless_node_eliminater]: 3.81999e-06
        [accelerated_algorithm]: 3.53999e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 3.11001e-06
        [merge_forward]: 2.16e-06
        [cell_reuse_recompute_pass]: 1.70001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.31e-06
        [j_node_and_user_rematch]: 5.19e-06
        [meta_fg_expand]: 2.446e-05
        [replace_old_param]: 3.46999e-06
        [inline_without_move]: 3.09999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.48002e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 4.24997e-06
        [cse]: 1.067e-05
        [replace_applicator]: 4.38001e-06
[py_interpret_to_execute_after_opt_a]: 9.52999e-06
[rewriter_after_opt_a]: 3.411e-05
[convert_after_rewriter]: 6.87002e-06
[order_py_execute_after_rewriter]: 5.20999e-06
[mutable_eliminate]: 0.00049464
[jit_opt_b]: 4.101e-05, [1]
    [Cycle 1]: 3.427e-05, [2]
        [frontend_op_eliminate]: 1.093e-05
        [inline_after_opt_a]: 1.174e-05
[cconv]: 1.966e-05
[loop_unroll]: 0.00041609
[jit_opt_after_cconv]: 0.00011436, [1]
    [Cycle 1]: 0.00010806, [11]
        [c_1]: 1.15e-05
        [parameter_eliminate]: 2.39999e-06
        [updatestate_depend_eliminate]: 4.58001e-06
        [updatestate_assign_eliminate]: 1.91998e-06
        [updatestate_loads_eliminate]: 1.47001e-06
        [cse]: 1.419e-05
        [call_graph_tuple_transform]: 1.342e-05
        [tuple_list_get_item_eliminator]: 3.81999e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 3.80998e-06
[remove_dup_value]: 1.367e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 5.14003e-06
[add_recomputation]: 3.225e-05
[cse_after_recomputation]: 1.979e-05, [1]
    [Cycle 1]: 1.425e-05, [1]
        [cse]: 8.08999e-06
[auto_monad_reorder]: 1.296e-05
[get_jit_bprop_graph]: 1.78002e-06
[rewriter_after_jit_bprop_graph]: 5.82999e-06
[opt_after_jit_grad]: 0.00045155
[symbol_engine_optimizer]: 6.297e-05, [1]
    [Cycle 1]: 5.685e-05, [6]
        [build]: 3.10002e-06
        [elim_shapecalc]: 5.83002e-06
        [elim_not_effective]: 9.39e-06
        [opt_reshape]: 4.25e-06
        [fold_const_symbol]: 5.71e-06
        [renormalize]: 3.89991e-07
[validate]: 2.599e-05
Sums
    bootstrap                            :     0.000175s :  1.17%
    type_inference                       :     0.008715s : 58.11%
    event_method                         :     0.000042s :  0.28%
    auto_monad                           :     0.000091s :  0.60%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.17%
    rewriter_before_opt_a                :     0.000093s :  0.62%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000064s :  0.42%
    jit_opt_a.loop_unroll                :     0.000042s :  0.28%
    jit_opt_a.a_1                        :     0.000836s :  5.57%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000017s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000015s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.10%
    jit_opt_a.merge_forward              :     0.000010s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001421s :  9.47%
    jit_opt_a.replace_old_param          :     0.000054s :  0.36%
    jit_opt_a.inline_without_move        :     0.000054s :  0.36%
    jit_opt_a.renormalize                :     0.001409s :  9.40%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.11%
    jit_opt_a.cse                        :     0.000037s :  0.24%
    jit_opt_a.replace_applicator         :     0.000015s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000034s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000495s :  3.30%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.08%
    cconv                                :     0.000020s :  0.13%
    loop_unroll                          :     0.000416s :  2.77%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.cse              :     0.000014s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.03%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000032s :  0.22%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000452s :  3.01%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000026s :  0.17%


Time group info:
------[substitution.]   0.000226    50
 0.65% :     0.000001s :      1: substitution.elim_not_effective
 0.42% :     0.000001s :      1: substitution.fold_const_symbol
 1.84% :     0.000004s :      1: substitution.graph_param_transform
63.47% :     0.000143s :      6: substitution.inline
 7.33% :     0.000017s :      2: substitution.inline_without_move
 2.54% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.61% :     0.000004s :      2: substitution.minmaximum_grad
 1.73% :     0.000004s :      3: substitution.partial_eliminate
 2.98% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.35% :     0.000005s :      2: substitution.replace_applicator
 1.81% :     0.000004s :      5: substitution.replace_old_param
 1.18% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.23% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.96% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 6.91% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008653     2
88.37% :     0.007647s :      1: type_inference.infer
11.63% :     0.001006s :      1: type_inference.specialize
------[replace.]   0.000058     8
78.72% :     0.000046s :      6: replace.inline
21.28% :     0.000012s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000151     8
92.69% :     0.000140s :      6: match.inline
 7.31% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000165  1109
 1.58% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.61% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.38% :     0.000002s :     18: predicate.addn_check_dump
 1.49% :     0.000002s :     18: predicate.addn_zero_filter
 2.20% :     0.000004s :     18: predicate.arithmetic_simplify
 1.44% :     0.000002s :     18: predicate.cast_eliminate
 0.17% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.38% :     0.000002s :     18: predicate.compare_switch_simplify
 1.44% :     0.000002s :     18: predicate.depend_value_elim
 1.29% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.50% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.45% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.46% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      1: predicate.elim_not_effective
 0.29% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.39% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.30% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.47% :     0.000002s :     18: predicate.environ_get_eliminate
 1.28% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.14% :     0.000002s :     12: predicate.get_grad_eliminate
 0.12% :     0.000000s :      1: predicate.graph_param_transform
 4.54% :     0.000007s :     28: predicate.inline
 3.24% :     0.000005s :     33: predicate.inline_without_move
 0.54% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.40% :     0.000002s :     12: predicate.less_batch_normalization
 1.60% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.66% :     0.000003s :     21: predicate.load_eliminater
 0.73% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.58% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.71% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.32% :     0.000002s :     18: predicate.merge_addn
 1.34% :     0.000002s :     18: predicate.minmaximum_grad
 1.06% :     0.000002s :      1: predicate.mutable_eliminate
 0.31% :     0.000001s :      1: predicate.opt_reshape
 2.07% :     0.000003s :     21: predicate.partial_eliminate
 1.34% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.71% :     0.000003s :     18: predicate.reduce_eliminate
 1.58% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.51% :     0.000002s :     22: predicate.replace_applicator
 1.66% :     0.000003s :     33: predicate.replace_old_param
 0.13% :     0.000000s :      1: predicate.reset_defer_inline
 1.54% :     0.000003s :     18: predicate.reshape_eliminate
 1.37% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     18: predicate.same_eliminate
 0.68% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.32% :     0.000001s :      2: predicate.special_op_eliminate
 1.21% :     0.000002s :     12: predicate.specialize_transform
 1.54% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.62% :     0.000004s :     26: predicate.switch_defer_inline
 2.42% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.74% :     0.000011s :     71: predicate.switch_simplify
 1.43% :     0.000002s :     18: predicate.tile_eliminate
 1.40% :     0.000002s :     18: predicate.transpose_eliminate
 1.71% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.61% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.11% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.73% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.66% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000003s :     18: predicate.value_based_eliminate
 0.13% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.29% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001029    22
73.00% :     0.000751s :     13: func_graph_cloner_run.FuncGraphClonerGraph
27.00% :     0.000278s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021321    72
 0.16% :     0.000035s :      1: add_recomputation
 0.44% :     0.000094s :      1: auto_monad
 0.07% :     0.000016s :      1: auto_monad_reorder
 0.87% :     0.000186s :      1: bootstrap
 0.11% :     0.000022s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.22% :     0.000046s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
35.74% :     0.007621s :      1: jit_opt_a
 0.55% :     0.000117s :      1: jit_opt_after_cconv
 0.20% :     0.000044s :      1: jit_opt_b
 1.99% :     0.000424s :      1: loop_unroll
 2.36% :     0.000503s :      1: mutable_eliminate
 5.55% :     0.001184s :     26: opt.transform.jit_opt_a
 0.14% :     0.000029s :      4: opt.transform.jit_opt_after_cconv
 0.08% :     0.000016s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000014s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000022s :      4: opt.transform.symbol_engine_opt
 2.15% :     0.000459s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000028s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 5.02% :     0.001069s :      1: renormalize.infer
 1.56% :     0.000332s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000037s :      1: rewriter_after_opt_a
 0.45% :     0.000096s :      1: rewriter_before_opt_a
 0.31% :     0.000065s :      1: symbol_engine_optimizer
40.93% :     0.008728s :      1: type_inference


TotalTime = 0.0229272, [33]
[bootstrap]: 0.00024171
[type_inference]: 0.0113909
[event_method]: 4.543e-05
[auto_monad]: 6.767e-05
[graph_reusing]: 5.25999e-06
[pre_auto_parallel]: 2.31998e-06
[py_interpret_to_execute]: 2.186e-05
[rewriter_before_opt_a]: 8.785e-05
[expand_dump_flag]: 2.34999e-06
[jit_opt_a]: 0.00919134, [2]
    [Cycle 1]: 0.0043337, [27]
        [switch_simplify]: 5.833e-05
        [loop_unroll]: 3.787e-05
        [a_1]: 0.00084802
        [with_stream_mark]: 2.11e-05
        [recompute_prepare]: 1.714e-05
        [updatestate_depend_eliminate]: 6.95002e-06
        [updatestate_assign_eliminate]: 5.62999e-06
        [updatestate_loads_eliminate]: 5.46998e-06
        [parameter_eliminate]: 2.26998e-06
        [specialize_transform]: 1.264e-05
        [updatestate_useless_node_eliminater]: 1.155e-05
        [accelerated_algorithm]: 1.248e-05
        [meta_shard_fg_expand]: 2.94001e-06
        [get_grad_eliminate_]: 1.164e-05
        [merge_forward]: 7.31999e-06
        [cell_reuse_recompute_pass]: 1.00001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.6e-05
        [j_node_and_user_rematch]: 2.329e-05
        [meta_fg_expand]: 0.00135062
        [replace_old_param]: 4.989e-05
        [inline_without_move]: 4.839e-05
        [renormalize]: 0.00152906
        [add_forward_monad_depend]: 4.2e-06
        [auto_monad_grad]: 1.29e-06
        [auto_monad_eliminator]: 8.57e-06
        [cse]: 1.724e-05
        [replace_applicator]: 1.148e-05
    [Cycle 2]: 0.00032294, [27]
        [switch_simplify]: 5.71e-06
        [loop_unroll]: 4.55001e-06
        [a_1]: 4.039e-05
        [with_stream_mark]: 8.77999e-06
        [recompute_prepare]: 3.79002e-06
        [updatestate_depend_eliminate]: 2.02001e-06
        [updatestate_assign_eliminate]: 1.54e-06
        [updatestate_loads_eliminate]: 1.44e-06
        [parameter_eliminate]: 1.00001e-06
        [specialize_transform]: 3.5e-06
        [updatestate_useless_node_eliminater]: 3.3e-06
        [accelerated_algorithm]: 6.04001e-06
        [meta_shard_fg_expand]: 1.10001e-06
        [get_grad_eliminate_]: 3.43999e-06
        [merge_forward]: 2.29999e-06
        [cell_reuse_recompute_pass]: 1.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.46998e-06
        [j_node_and_user_rematch]: 5.86e-06
        [meta_fg_expand]: 3.355e-05
        [replace_old_param]: 3.93999e-06
        [inline_without_move]: 3.28e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.37999e-06
        [auto_monad_grad]: 1.12999e-06
        [auto_monad_eliminator]: 4.53999e-06
        [cse]: 1.079e-05
        [replace_applicator]: 3.65e-06
[py_interpret_to_execute_after_opt_a]: 8.17998e-06
[rewriter_after_opt_a]: 2.615e-05
[convert_after_rewriter]: 5.60001e-06
[order_py_execute_after_rewriter]: 3.88999e-06
[mutable_eliminate]: 0.00048709
[jit_opt_b]: 3.789e-05, [1]
    [Cycle 1]: 3.165e-05, [2]
        [frontend_op_eliminate]: 9.65002e-06
        [inline_after_opt_a]: 1.086e-05
[cconv]: 1.637e-05
[loop_unroll]: 0.00040549
[jit_opt_after_cconv]: 0.00011457, [1]
    [Cycle 1]: 0.00010795, [11]
        [c_1]: 1.232e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 4.68001e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.62001e-06
        [cse]: 1.502e-05
        [call_graph_tuple_transform]: 1.149e-05
        [tuple_list_get_item_eliminator]: 4.01001e-06
        [none_parameter_eliminate]: 9.5999e-07
        [renormalize]: 5.19998e-07
        [switch_simplify]: 3.73001e-06
[remove_dup_value]: 9.04e-06
[partial_unused_args_eliminate]: 1.75001e-06
[environ_conv]: 3.76999e-06
[add_recomputation]: 2.528e-05
[cse_after_recomputation]: 1.846e-05, [1]
    [Cycle 1]: 1.343e-05, [1]
        [cse]: 7.71999e-06
[auto_monad_reorder]: 8.90001e-06
[get_jit_bprop_graph]: 1.34e-06
[rewriter_after_jit_bprop_graph]: 4.23999e-06
[opt_after_jit_grad]: 0.00042599
[symbol_engine_optimizer]: 6.036e-05, [1]
    [Cycle 1]: 5.463e-05, [6]
        [build]: 2.58e-06
        [elim_shapecalc]: 5.49998e-06
        [elim_not_effective]: 8.96002e-06
        [opt_reshape]: 3.85e-06
        [fold_const_symbol]: 5.84e-06
        [renormalize]: 6.09987e-07
[validate]: 2.311e-05
[backend_pass]: 7.30011e-07
[task_emit]: 1.7e-05
[execute]: 1.01002e-06
Sums
    bootstrap                            :     0.000242s :  1.36%
    type_inference                       :     0.011391s : 64.15%
    event_method                         :     0.000045s :  0.26%
    auto_monad                           :     0.000068s :  0.38%
    graph_reusing                        :     0.000005s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000022s :  0.12%
    rewriter_before_opt_a                :     0.000088s :  0.49%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000064s :  0.36%
    jit_opt_a.loop_unroll                :     0.000042s :  0.24%
    jit_opt_a.a_1                        :     0.000888s :  5.00%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.04%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000016s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000019s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.08%
    jit_opt_a.merge_forward              :     0.000010s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001384s :  7.80%
    jit_opt_a.replace_old_param          :     0.000054s :  0.30%
    jit_opt_a.inline_without_move        :     0.000052s :  0.29%
    jit_opt_a.renormalize                :     0.001529s :  8.61%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.03%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.07%
    jit_opt_a.cse                        :     0.000028s :  0.16%
    jit_opt_a.replace_applicator         :     0.000015s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000026s :  0.15%
    convert_after_rewriter               :     0.000006s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000487s :  2.74%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.05%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.06%
    cconv                                :     0.000016s :  0.09%
    loop_unroll                          :     0.000405s :  2.28%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.08%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000011s :  0.06%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000009s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.02%
    add_recomputation                    :     0.000025s :  0.14%
    cse_after_recomputation.cse          :     0.000008s :  0.04%
    auto_monad_reorder                   :     0.000009s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.02%
    opt_after_jit_grad                   :     0.000426s :  2.40%
    symbol_engine_optimizer.build        :     0.000003s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000023s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000017s :  0.10%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000219    50
 0.51% :     0.000001s :      1: substitution.elim_not_effective
 0.44% :     0.000001s :      1: substitution.fold_const_symbol
 1.08% :     0.000002s :      1: substitution.graph_param_transform
67.39% :     0.000148s :      6: substitution.inline
 6.77% :     0.000015s :      2: substitution.inline_without_move
 2.68% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.17% :     0.000003s :      2: substitution.minmaximum_grad
 1.86% :     0.000004s :      3: substitution.partial_eliminate
 3.17% :     0.000007s :      9: substitution.remove_not_recompute_node
 1.89% :     0.000004s :      2: substitution.replace_applicator
 1.72% :     0.000004s :      5: substitution.replace_old_param
 0.98% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.85% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.72% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 5.76% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.011332     2
91.23% :     0.010338s :      1: type_inference.infer
 8.77% :     0.000994s :      1: type_inference.specialize
------[replace.]   0.000061     8
78.65% :     0.000048s :      6: replace.inline
21.35% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000153     8
94.68% :     0.000144s :      6: match.inline
 5.32% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000164  1109
 1.64% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.52% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.36% :     0.000002s :     18: predicate.addn_check_dump
 1.72% :     0.000003s :     18: predicate.addn_zero_filter
 2.29% :     0.000004s :     18: predicate.arithmetic_simplify
 1.55% :     0.000003s :     18: predicate.cast_eliminate
 0.13% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.36% :     0.000002s :     18: predicate.compare_switch_simplify
 1.56% :     0.000003s :     18: predicate.depend_value_elim
 1.34% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.57% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.45% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.42% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      1: predicate.elim_not_effective
 0.23% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.48% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.37% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.28% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.47% :     0.000002s :     18: predicate.environ_get_eliminate
 1.31% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.13% :     0.000002s :     12: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 4.38% :     0.000007s :     28: predicate.inline
 3.16% :     0.000005s :     33: predicate.inline_without_move
 0.50% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :     12: predicate.less_batch_normalization
 1.63% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.80% :     0.000003s :     21: predicate.load_eliminater
 0.68% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.58% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.92% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.37% :     0.000002s :     18: predicate.merge_addn
 1.31% :     0.000002s :     18: predicate.minmaximum_grad
 0.60% :     0.000001s :      1: predicate.mutable_eliminate
 0.24% :     0.000000s :      1: predicate.opt_reshape
 2.03% :     0.000003s :     21: predicate.partial_eliminate
 1.52% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.69% :     0.000003s :     18: predicate.reduce_eliminate
 1.64% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.53% :     0.000003s :     22: predicate.replace_applicator
 1.59% :     0.000003s :     33: predicate.replace_old_param
 0.13% :     0.000000s :      1: predicate.reset_defer_inline
 1.62% :     0.000003s :     18: predicate.reshape_eliminate
 1.46% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.52% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     18: predicate.same_eliminate
 0.64% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.30% :     0.000000s :      2: predicate.special_op_eliminate
 1.18% :     0.000002s :     12: predicate.specialize_transform
 1.64% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.14% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.64% :     0.000004s :     26: predicate.switch_defer_inline
 2.38% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.98% :     0.000011s :     71: predicate.switch_simplify
 1.42% :     0.000002s :     18: predicate.tile_eliminate
 1.44% :     0.000002s :     18: predicate.transpose_eliminate
 1.60% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.47% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 2.80% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.72% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.89% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.92% :     0.000003s :     18: predicate.value_based_eliminate
 0.12% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.25% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001091    22
72.96% :     0.000796s :     13: func_graph_cloner_run.FuncGraphClonerGraph
27.04% :     0.000295s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.025750    76
 0.11% :     0.000028s :      1: add_recomputation
 0.28% :     0.000072s :      1: auto_monad
 0.04% :     0.000011s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.99% :     0.000254s :      1: bootstrap
 0.07% :     0.000019s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.08% :     0.000020s :      1: cse_after_recomputation
 0.02% :     0.000006s :      1: environ_conv
 0.19% :     0.000050s :      1: event_method
 0.01% :     0.000003s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000008s :      1: graph_reusing
35.71% :     0.009194s :      1: jit_opt_a
 0.46% :     0.000117s :      1: jit_opt_after_cconv
 0.16% :     0.000040s :      1: jit_opt_b
 1.60% :     0.000413s :      1: loop_unroll
 1.92% :     0.000495s :      1: mutable_eliminate
 4.81% :     0.001238s :     26: opt.transform.jit_opt_a
 0.11% :     0.000028s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000014s :      4: opt.transform.jit_opt_b
 0.04% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.05% :     0.000014s :      1: opt.transform.opt_after_jit_grad
 0.08% :     0.000021s :      4: opt.transform.symbol_engine_opt
 1.68% :     0.000433s :      1: opt_after_jit_grad
 0.02% :     0.000006s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.10% :     0.000025s :      1: py_interpret_to_execute
 0.04% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.04% :     0.000011s :      1: remove_dup_value
 4.49% :     0.001157s :      1: renormalize.infer
 1.42% :     0.000365s :      1: renormalize.specialize
 0.02% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.11% :     0.000029s :      1: rewriter_after_opt_a
 0.35% :     0.000091s :      1: rewriter_before_opt_a
 0.24% :     0.000063s :      1: symbol_engine_optimizer
 0.08% :     0.000021s :      1: task_emit
44.28% :     0.011403s :      1: type_inference
 0.14% :     0.000037s :      1: validate

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.118.483 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d18b0> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.119.036 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:22.122.402 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd4c368a60> should be wrapped by jit only once.
[WARNING] ME(27375:281473069915952,ForkProcess-1):2026-01-29-17:46:22.123.016 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0132781, [30]
[bootstrap]: 0.00031706
[type_inference]: 0.004453
[event_method]: 1.81e-05
[auto_monad]: 6.898e-05
[graph_reusing]: 6.31e-06
[pre_auto_parallel]: 3.13e-06
[py_interpret_to_execute]: 3.158e-05
[rewriter_before_opt_a]: 7.296e-05
[expand_dump_flag]: 3.16001e-06
[jit_opt_a]: 0.00574571, [2]
    [Cycle 1]: 0.00179895, [27]
        [switch_simplify]: 5.942e-05
        [loop_unroll]: 2.926e-05
        [a_1]: 0.00057819
        [with_stream_mark]: 2.52e-05
        [recompute_prepare]: 7.55e-06
        [updatestate_depend_eliminate]: 4.13999e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 2.58003e-06
        [parameter_eliminate]: 2.17999e-06
        [specialize_transform]: 6.35002e-06
        [updatestate_useless_node_eliminater]: 5.30001e-06
        [accelerated_algorithm]: 5.82001e-06
        [meta_shard_fg_expand]: 2.53e-06
        [get_grad_eliminate_]: 4.95999e-06
        [merge_forward]: 3.83001e-06
        [cell_reuse_recompute_pass]: 1.12999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.593e-05
        [j_node_and_user_rematch]: 9.75002e-06
        [meta_fg_expand]: 2.08002e-06
        [replace_old_param]: 1.053e-05
        [inline_without_move]: 5.39998e-06
        [renormalize]: 0.00072851
        [add_forward_monad_depend]: 8.35999e-06
        [auto_monad_grad]: 2.76999e-06
        [auto_monad_eliminator]: 2.016e-05
        [cse]: 3.553e-05
        [replace_applicator]: 1.649e-05
    [Cycle 2]: 0.00033837, [27]
        [switch_simplify]: 5.82001e-06
        [loop_unroll]: 5.50001e-06
        [a_1]: 9.316e-05
        [with_stream_mark]: 1.353e-05
        [recompute_prepare]: 5.71e-06
        [updatestate_depend_eliminate]: 3.13e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 1.98002e-06
        [parameter_eliminate]: 1.31002e-06
        [specialize_transform]: 5.20999e-06
        [updatestate_useless_node_eliminater]: 4.76002e-06
        [accelerated_algorithm]: 5.00999e-06
        [meta_shard_fg_expand]: 1.96e-06
        [get_grad_eliminate_]: 4.54998e-06
        [merge_forward]: 2.92002e-06
        [cell_reuse_recompute_pass]: 1.52001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.751e-05
        [j_node_and_user_rematch]: 8.37e-06
        [meta_fg_expand]: 1.58002e-06
        [replace_old_param]: 8.17e-06
        [inline_without_move]: 4.62998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.86e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 5.51998e-06
        [cse]: 1.145e-05
        [replace_applicator]: 5.56e-06
[py_interpret_to_execute_after_opt_a]: 1.454e-05
[rewriter_after_opt_a]: 3.818e-05
[convert_after_rewriter]: 6.73998e-06
[order_py_execute_after_rewriter]: 5.08002e-06
[mutable_eliminate]: 0.00080792
[jit_opt_b]: 5.414e-05, [1]
    [Cycle 1]: 4.43e-05, [2]
        [frontend_op_eliminate]: 1.634e-05
        [inline_after_opt_a]: 1.564e-05
[cconv]: 3.22e-05
[loop_unroll]: 0.00049213
[jit_opt_after_cconv]: 0.00016272, [1]
    [Cycle 1]: 0.00015464, [11]
        [c_1]: 2.268e-05
        [parameter_eliminate]: 4.02002e-06
        [updatestate_depend_eliminate]: 8.67e-06
        [updatestate_assign_eliminate]: 2.82002e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [cse]: 2.747e-05
        [call_graph_tuple_transform]: 2.196e-05
        [tuple_list_get_item_eliminator]: 5.69e-06
        [none_parameter_eliminate]: 1.59998e-06
        [renormalize]: 8.29983e-07
        [switch_simplify]: 5.14e-06
[remove_dup_value]: 1.585e-05
[partial_unused_args_eliminate]: 2.74001e-06
[environ_conv]: 7.18998e-06
[add_recomputation]: 6.196e-05
[cse_after_recomputation]: 2.223e-05, [1]
    [Cycle 1]: 1.651e-05, [1]
        [cse]: 9.68997e-06
[auto_monad_reorder]: 1.634e-05
[get_jit_bprop_graph]: 1.98002e-06
[rewriter_after_jit_bprop_graph]: 8.45999e-06
[opt_after_jit_grad]: 0.00050804
[symbol_engine_optimizer]: 7.431e-05, [1]
    [Cycle 1]: 6.816e-05, [6]
        [build]: 5.55001e-06
        [elim_shapecalc]: 7.63999e-06
        [elim_not_effective]: 1.272e-05
        [opt_reshape]: 5.39e-06
        [fold_const_symbol]: 8.42e-06
        [renormalize]: 5.09986e-07
[validate]: 4.106e-05
Sums
    bootstrap                            :     0.000317s :  3.51%
    type_inference                       :     0.004453s : 49.25%
    event_method                         :     0.000018s :  0.20%
    auto_monad                           :     0.000069s :  0.76%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.35%
    rewriter_before_opt_a                :     0.000073s :  0.81%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000065s :  0.72%
    jit_opt_a.loop_unroll                :     0.000035s :  0.38%
    jit_opt_a.a_1                        :     0.000671s :  7.43%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.43%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.37%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000729s :  8.06%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.11%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000026s :  0.28%
    jit_opt_a.cse                        :     0.000047s :  0.52%
    jit_opt_a.replace_applicator         :     0.000022s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.16%
    rewriter_after_opt_a                 :     0.000038s :  0.42%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000808s :  8.94%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.17%
    cconv                                :     0.000032s :  0.36%
    loop_unroll                          :     0.000492s :  5.44%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.10%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000027s :  0.30%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000016s :  0.18%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000007s :  0.08%
    add_recomputation                    :     0.000062s :  0.69%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.09%
    opt_after_jit_grad                   :     0.000508s :  5.62%
    symbol_engine_optimizer.build        :     0.000006s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000041s :  0.45%


Time group info:
------[substitution.]   0.000242    23
 0.77% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.44% :     0.000006s :      3: substitution.graph_param_transform
84.11% :     0.000203s :      4: substitution.inline
 1.58% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.50% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.40% :     0.000006s :      2: substitution.replace_old_param
 5.71% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004382     2
82.01% :     0.003594s :      1: type_inference.infer
17.99% :     0.000788s :      1: type_inference.specialize
------[replace.]   0.000069     6
66.96% :     0.000046s :      4: replace.inline
33.04% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000213     6
94.08% :     0.000201s :      4: match.inline
 5.92% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.11% :     0.000001s :     10: predicate.addn_zero_filter
 1.96% :     0.000003s :     10: predicate.arithmetic_simplify
 1.01% :     0.000001s :     10: predicate.cast_eliminate
 0.69% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.96% :     0.000001s :     10: predicate.compare_switch_simplify
 1.13% :     0.000001s :     10: predicate.depend_value_elim
 1.00% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.60% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.94% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.50% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000002s :     10: predicate.environ_get_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.80% :     0.000001s :      6: predicate.get_grad_eliminate
 0.41% :     0.000001s :      3: predicate.graph_param_transform
 5.47% :     0.000007s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.55% :     0.000002s :      6: predicate.less_batch_normalization
 2.06% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.52% :     0.000002s :     15: predicate.load_eliminater
 2.11% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.63% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.63% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 2.89% :     0.000004s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.30% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.86% :     0.000002s :     10: predicate.reduce_eliminate
 1.29% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.72% :     0.000002s :     10: predicate.reshape_eliminate
 1.33% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.22% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000002s :     10: predicate.same_eliminate
 0.69% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.69% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.25% :     0.000003s :     16: predicate.switch_defer_inline
 1.87% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.41% :     0.000010s :     53: predicate.switch_simplify
 1.02% :     0.000001s :     10: predicate.tile_eliminate
 1.59% :     0.000002s :     10: predicate.transpose_eliminate
 1.19% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.00% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.26% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.48% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.97% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000586    12
50.58% :     0.000297s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.11% :     0.000048s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.31% :     0.000242s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014960    72
 0.43% :     0.000065s :      1: add_recomputation
 0.49% :     0.000073s :      1: auto_monad
 0.13% :     0.000019s :      1: auto_monad_reorder
 2.25% :     0.000337s :      1: bootstrap
 0.23% :     0.000035s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.43% :     0.005749s :      1: jit_opt_a
 1.11% :     0.000166s :      1: jit_opt_after_cconv
 0.38% :     0.000057s :      1: jit_opt_b
 3.35% :     0.000502s :      1: loop_unroll
 5.48% :     0.000820s :      1: mutable_eliminate
 6.00% :     0.000897s :     26: opt.transform.jit_opt_a
 0.34% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000025s :      4: opt.transform.jit_opt_b
 0.10% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.13% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.46% :     0.000517s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000035s :      1: py_interpret_to_execute
 0.11% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000018s :      1: remove_dup_value
 2.69% :     0.000402s :      1: renormalize.infer
 2.11% :     0.000316s :      1: renormalize.specialize
 0.07% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000041s :      1: rewriter_after_opt_a
 0.51% :     0.000076s :      1: rewriter_before_opt_a
 0.51% :     0.000077s :      1: symbol_engine_optimizer
29.87% :     0.004468s :      1: type_inference


TotalTime = 0.0185861, [30]
[bootstrap]: 0.00019572
[type_inference]: 0.00874834
[event_method]: 3.83e-05
[auto_monad]: 8.918e-05
[graph_reusing]: 6.62002e-06
[pre_auto_parallel]: 2.46998e-06
[py_interpret_to_execute]: 2.227e-05
[rewriter_before_opt_a]: 8.465e-05
[expand_dump_flag]: 2.93e-06
[jit_opt_a]: 0.00746066, [2]
    [Cycle 1]: 0.00414159, [27]
        [switch_simplify]: 5.463e-05
        [loop_unroll]: 3.745e-05
        [a_1]: 0.00078002
        [with_stream_mark]: 2.058e-05
        [recompute_prepare]: 1.711e-05
        [updatestate_depend_eliminate]: 6.98998e-06
        [updatestate_assign_eliminate]: 5.61998e-06
        [updatestate_loads_eliminate]: 5.45001e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 1.257e-05
        [updatestate_useless_node_eliminater]: 1.162e-05
        [accelerated_algorithm]: 1.18e-05
        [meta_shard_fg_expand]: 2.75997e-06
        [get_grad_eliminate_]: 1.164e-05
        [merge_forward]: 7.38999e-06
        [cell_reuse_recompute_pass]: 7.80012e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.426e-05
        [j_node_and_user_rematch]: 2.204e-05
        [meta_fg_expand]: 0.00139206
        [replace_old_param]: 5.063e-05
        [inline_without_move]: 4.924e-05
        [renormalize]: 0.00137997
        [add_forward_monad_depend]: 5.32001e-06
        [auto_monad_grad]: 1.90001e-06
        [auto_monad_eliminator]: 1.111e-05
        [cse]: 1.653e-05
        [replace_applicator]: 1.148e-05
    [Cycle 2]: 0.00026035, [27]
        [switch_simplify]: 4.25999e-06
        [loop_unroll]: 3.71001e-06
        [a_1]: 3.662e-05
        [with_stream_mark]: 8.31002e-06
        [recompute_prepare]: 3.70998e-06
        [updatestate_depend_eliminate]: 2.22001e-06
        [updatestate_assign_eliminate]: 1.69998e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 1.09e-06
        [specialize_transform]: 3.43e-06
        [updatestate_useless_node_eliminater]: 3.58e-06
        [accelerated_algorithm]: 3.43e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 3.56001e-06
        [merge_forward]: 2.32001e-06
        [cell_reuse_recompute_pass]: 1.45001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.073e-05
        [j_node_and_user_rematch]: 5.09e-06
        [meta_fg_expand]: 2.134e-05
        [replace_old_param]: 3.73001e-06
        [inline_without_move]: 3.10002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.50999e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 4.04002e-06
        [cse]: 9.68997e-06
        [replace_applicator]: 3.71999e-06
[py_interpret_to_execute_after_opt_a]: 8.43001e-06
[rewriter_after_opt_a]: 2.99e-05
[convert_after_rewriter]: 5.40999e-06
[order_py_execute_after_rewriter]: 4.48999e-06
[mutable_eliminate]: 0.00048525
[jit_opt_b]: 3.774e-05, [1]
    [Cycle 1]: 3.15e-05, [2]
        [frontend_op_eliminate]: 9.92999e-06
        [inline_after_opt_a]: 1.052e-05
[cconv]: 2.025e-05
[loop_unroll]: 0.00042327
[jit_opt_after_cconv]: 0.00012274, [1]
    [Cycle 1]: 0.00011658, [11]
        [c_1]: 1.19e-05
        [parameter_eliminate]: 2.56e-06
        [updatestate_depend_eliminate]: 4.71002e-06
        [updatestate_assign_eliminate]: 1.99999e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [cse]: 1.567e-05
        [call_graph_tuple_transform]: 1.413e-05
        [tuple_list_get_item_eliminator]: 4.15999e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 6.19999e-07
        [switch_simplify]: 3.76999e-06
[remove_dup_value]: 1.253e-05
[partial_unused_args_eliminate]: 2.44999e-06
[environ_conv]: 5.05999e-06
[add_recomputation]: 3.065e-05
[cse_after_recomputation]: 1.987e-05, [1]
    [Cycle 1]: 1.462e-05, [1]
        [cse]: 8.40001e-06
[auto_monad_reorder]: 1.094e-05
[get_jit_bprop_graph]: 1.34e-06
[rewriter_after_jit_bprop_graph]: 4.87998e-06
[opt_after_jit_grad]: 0.00046607
[symbol_engine_optimizer]: 6.36e-05, [1]
    [Cycle 1]: 5.677e-05, [6]
        [build]: 3.56001e-06
        [elim_shapecalc]: 5.66e-06
        [elim_not_effective]: 9.64e-06
        [opt_reshape]: 4.05e-06
        [fold_const_symbol]: 5.82001e-06
        [renormalize]: 4.09986e-07
[validate]: 2.678e-05
Sums
    bootstrap                            :     0.000196s :  1.31%
    type_inference                       :     0.008748s : 58.52%
    event_method                         :     0.000038s :  0.26%
    auto_monad                           :     0.000089s :  0.60%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000022s :  0.15%
    rewriter_before_opt_a                :     0.000085s :  0.57%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000059s :  0.39%
    jit_opt_a.loop_unroll                :     0.000041s :  0.28%
    jit_opt_a.a_1                        :     0.000817s :  5.46%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000016s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000015s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.10%
    jit_opt_a.merge_forward              :     0.000010s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001413s :  9.46%
    jit_opt_a.replace_old_param          :     0.000054s :  0.36%
    jit_opt_a.inline_without_move        :     0.000052s :  0.35%
    jit_opt_a.renormalize                :     0.001380s :  9.23%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.10%
    jit_opt_a.cse                        :     0.000026s :  0.18%
    jit_opt_a.replace_applicator         :     0.000015s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000030s :  0.20%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000485s :  3.25%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000020s :  0.14%
    loop_unroll                          :     0.000423s :  2.83%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000016s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.03%
    remove_dup_value                     :     0.000013s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000031s :  0.21%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000011s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000466s :  3.12%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.18%


TotalTime = 0.0316511, [33]
[bootstrap]: 0.00018885
[type_inference]: 0.0092617
[event_method]: 4.19e-05
[auto_monad]: 8.632e-05
[graph_reusing]: 6.36e-06
[pre_auto_parallel]: 2.56998e-06
[py_interpret_to_execute]: 2.28e-05
[rewriter_before_opt_a]: 8.312e-05
[expand_dump_flag]: 3.22002e-06
[jit_opt_a]: 0.0114484, [3]
    [Cycle 1]: 0.00570959, [27]
        [switch_simplify]: 6.008e-05
        [loop_unroll]: 3.812e-05
        [a_1]: 0.00083234
        [with_stream_mark]: 2.544e-05
        [recompute_prepare]: 1.803e-05
        [updatestate_depend_eliminate]: 7.33e-06
        [updatestate_assign_eliminate]: 5.96e-06
        [updatestate_loads_eliminate]: 5.51e-06
        [parameter_eliminate]: 2.61e-06
        [specialize_transform]: 1.276e-05
        [updatestate_useless_node_eliminater]: 1.179e-05
        [accelerated_algorithm]: 1.21e-05
        [meta_shard_fg_expand]: 3.45e-06
        [get_grad_eliminate_]: 1.178e-05
        [merge_forward]: 7.98001e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.543e-05
        [j_node_and_user_rematch]: 2.175e-05
        [meta_fg_expand]: 0.00145947
        [replace_old_param]: 5.326e-05
        [inline_without_move]: 5.201e-05
        [renormalize]: 0.00254813
        [add_forward_monad_depend]: 9.84001e-06
        [auto_monad_grad]: 5.28002e-06
        [auto_monad_eliminator]: 5.237e-05
        [cse]: 0.00015579
        [replace_applicator]: 6.853e-05
    [Cycle 2]: 0.00217394, [27]
        [switch_simplify]: 3.703e-05
        [loop_unroll]: 3.583e-05
        [a_1]: 0.00114328
        [with_stream_mark]: 1.43e-05
        [recompute_prepare]: 1.024e-05
        [updatestate_depend_eliminate]: 5.29e-06
        [updatestate_assign_eliminate]: 4.33999e-06
        [updatestate_loads_eliminate]: 4.26001e-06
        [parameter_eliminate]: 1.55001e-06
        [specialize_transform]: 8.55001e-06
        [updatestate_useless_node_eliminater]: 8.04002e-06
        [accelerated_algorithm]: 8.49002e-06
        [meta_shard_fg_expand]: 2.53998e-06
        [get_grad_eliminate_]: 8.07e-06
        [merge_forward]: 5.05001e-06
        [cell_reuse_recompute_pass]: 1.00001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.905e-05
        [j_node_and_user_rematch]: 1.38e-05
        [meta_fg_expand]: 3.966e-05
        [replace_old_param]: 1.244e-05
        [inline_without_move]: 8e-06
        [renormalize]: 0.00055407
        [add_forward_monad_depend]: 4.18999e-06
        [auto_monad_grad]: 1.60999e-06
        [auto_monad_eliminator]: 1.414e-05
        [cse]: 4.487e-05
        [replace_applicator]: 1.39e-05
    [Cycle 3]: 0.00048448, [27]
        [switch_simplify]: 8.88002e-06
        [loop_unroll]: 7.93999e-06
        [a_1]: 0.00018723
        [with_stream_mark]: 1.116e-05
        [recompute_prepare]: 8.34002e-06
        [updatestate_depend_eliminate]: 4.37e-06
        [updatestate_assign_eliminate]: 4.03001e-06
        [updatestate_loads_eliminate]: 3.57002e-06
        [parameter_eliminate]: 1.11002e-06
        [specialize_transform]: 8.2e-06
        [updatestate_useless_node_eliminater]: 7.73001e-06
        [accelerated_algorithm]: 8.52998e-06
        [meta_shard_fg_expand]: 1.73997e-06
        [get_grad_eliminate_]: 7.65e-06
        [merge_forward]: 4.07e-06
        [cell_reuse_recompute_pass]: 1.11002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.633e-05
        [j_node_and_user_rematch]: 1.327e-05
        [meta_fg_expand]: 2.83998e-06
        [replace_old_param]: 9.91e-06
        [inline_without_move]: 7.74002e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 7.80012e-07
        [auto_monad_eliminator]: 1.034e-05
        [cse]: 2.369e-05
        [replace_applicator]: 9.00001e-06
[py_interpret_to_execute_after_opt_a]: 1.356e-05
[rewriter_after_opt_a]: 4.977e-05
[convert_after_rewriter]: 9.12999e-06
[order_py_execute_after_rewriter]: 6.31998e-06
[mutable_eliminate]: 0.00065591
[jit_opt_b]: 6.713e-05, [1]
    [Cycle 1]: 5.981e-05, [2]
        [frontend_op_eliminate]: 2.518e-05
        [inline_after_opt_a]: 2.324e-05
[cconv]: 2.318e-05
[loop_unroll]: 0.00041437
[jit_opt_after_cconv]: 0.00018656, [1]
    [Cycle 1]: 0.00017998, [11]
        [c_1]: 3.427e-05
        [parameter_eliminate]: 2.81e-06
        [updatestate_depend_eliminate]: 7.21999e-06
        [updatestate_assign_eliminate]: 4.36002e-06
        [updatestate_loads_eliminate]: 4.13999e-06
        [cse]: 3.312e-05
        [call_graph_tuple_transform]: 2.533e-05
        [tuple_list_get_item_eliminator]: 8.26002e-06
        [none_parameter_eliminate]: 1.66e-06
        [renormalize]: 6.59988e-07
        [switch_simplify]: 8.62e-06
[remove_dup_value]: 2.207e-05
[partial_unused_args_eliminate]: 2.17001e-06
[environ_conv]: 7.11001e-06
[add_recomputation]: 5.647e-05
[cse_after_recomputation]: 3.445e-05, [1]
    [Cycle 1]: 2.864e-05, [1]
        [cse]: 2.264e-05
[auto_monad_reorder]: 2.193e-05
[get_jit_bprop_graph]: 1.35001e-06
[rewriter_after_jit_bprop_graph]: 4.77998e-06
[opt_after_jit_grad]: 0.00053974
[symbol_engine_optimizer]: 9.073e-05, [1]
    [Cycle 1]: 8.447e-05, [6]
        [build]: 4.13001e-06
        [elim_shapecalc]: 1.162e-05
        [elim_not_effective]: 1.794e-05
        [opt_reshape]: 8.82999e-06
        [fold_const_symbol]: 1.478e-05
        [renormalize]: 4.30009e-07
[validate]: 4.355e-05
[backend_pass]: 8.59989e-07
[task_emit]: 0.00805995
[execute]: 7.14001e-06
Sums
    bootstrap                            :     0.000189s :  0.68%
    type_inference                       :     0.009262s : 33.32%
    event_method                         :     0.000042s :  0.15%
    auto_monad                           :     0.000086s :  0.31%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000023s :  0.08%
    rewriter_before_opt_a                :     0.000083s :  0.30%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000106s :  0.38%
    jit_opt_a.loop_unroll                :     0.000082s :  0.29%
    jit_opt_a.a_1                        :     0.002163s :  7.78%
    jit_opt_a.with_stream_mark           :     0.000051s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000037s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000014s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000013s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000030s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000028s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000029s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000028s :  0.10%
    jit_opt_a.merge_forward              :     0.000017s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000061s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000049s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001502s :  5.40%
    jit_opt_a.replace_old_param          :     0.000076s :  0.27%
    jit_opt_a.inline_without_move        :     0.000068s :  0.24%
    jit_opt_a.renormalize                :     0.003102s : 11.16%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000077s :  0.28%
    jit_opt_a.cse                        :     0.000224s :  0.81%
    jit_opt_a.replace_applicator         :     0.000091s :  0.33%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.05%
    rewriter_after_opt_a                 :     0.000050s :  0.18%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000656s :  2.36%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000023s :  0.08%
    cconv                                :     0.000023s :  0.08%
    loop_unroll                          :     0.000414s :  1.49%
    jit_opt_after_cconv.c_1              :     0.000034s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000033s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.03%
    remove_dup_value                     :     0.000022s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000056s :  0.20%
    cse_after_recomputation.cse          :     0.000023s :  0.08%
    auto_monad_reorder                   :     0.000022s :  0.08%
    get_jit_bprop_graph                  :     0.000001s :  0.00%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000540s :  1.94%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000012s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000044s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008060s : 28.99%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000213    50
 0.67% :     0.000001s :      1: substitution.elim_not_effective
 0.44% :     0.000001s :      1: substitution.fold_const_symbol
 1.46% :     0.000003s :      1: substitution.graph_param_transform
64.96% :     0.000138s :      6: substitution.inline
 7.10% :     0.000015s :      2: substitution.inline_without_move
 2.81% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.59% :     0.000003s :      2: substitution.minmaximum_grad
 1.71% :     0.000004s :      3: substitution.partial_eliminate
 3.18% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.22% :     0.000005s :      2: substitution.replace_applicator
 1.83% :     0.000004s :      5: substitution.replace_old_param
 1.18% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.86% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.79% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 6.20% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008688     2
89.19% :     0.007749s :      1: type_inference.infer
10.81% :     0.000939s :      1: type_inference.specialize
------[replace.]   0.000056     8
78.56% :     0.000044s :      6: replace.inline
21.44% :     0.000012s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000144     8
94.05% :     0.000135s :      6: match.inline
 5.95% :     0.000009s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000163  1109
 1.57% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.73% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.28% :     0.000002s :     18: predicate.addn_check_dump
 1.51% :     0.000002s :     18: predicate.addn_zero_filter
 2.08% :     0.000003s :     18: predicate.arithmetic_simplify
 1.43% :     0.000002s :     18: predicate.cast_eliminate
 0.20% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.27% :     0.000002s :     18: predicate.compare_switch_simplify
 1.51% :     0.000002s :     18: predicate.depend_value_elim
 1.39% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.35% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.53% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.14% :     0.000000s :      1: predicate.elim_not_effective
 0.22% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.39% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.39% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.34% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.33% :     0.000002s :     18: predicate.environ_get_eliminate
 1.33% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.49% :     0.000002s :     12: predicate.get_grad_eliminate
 0.13% :     0.000000s :      1: predicate.graph_param_transform
 4.27% :     0.000007s :     28: predicate.inline
 3.34% :     0.000005s :     33: predicate.inline_without_move
 0.52% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :     12: predicate.less_batch_normalization
 1.66% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.70% :     0.000003s :     21: predicate.load_eliminater
 0.80% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.61% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.70% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.34% :     0.000002s :     18: predicate.merge_addn
 1.33% :     0.000002s :     18: predicate.minmaximum_grad
 0.90% :     0.000001s :      1: predicate.mutable_eliminate
 0.22% :     0.000000s :      1: predicate.opt_reshape
 2.03% :     0.000003s :     21: predicate.partial_eliminate
 1.40% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.75% :     0.000003s :     18: predicate.reduce_eliminate
 1.63% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.70% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.55% :     0.000003s :     22: predicate.replace_applicator
 1.64% :     0.000003s :     33: predicate.replace_old_param
 0.09% :     0.000000s :      1: predicate.reset_defer_inline
 1.58% :     0.000003s :     18: predicate.reshape_eliminate
 1.72% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.46% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.45% :     0.000002s :     18: predicate.same_eliminate
 0.65% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.44% :     0.000001s :      2: predicate.special_op_eliminate
 1.19% :     0.000002s :     12: predicate.specialize_transform
 1.53% :     0.000002s :     18: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.14% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.65% :     0.000004s :     26: predicate.switch_defer_inline
 2.37% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.87% :     0.000011s :     71: predicate.switch_simplify
 1.43% :     0.000002s :     18: predicate.tile_eliminate
 1.39% :     0.000002s :     18: predicate.transpose_eliminate
 1.56% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.52% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 2.81% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     20: predicate.tuple_to_list_eliminator_
 1.61% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.11% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.98% :     0.000003s :     18: predicate.value_based_eliminate
 0.17% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.29% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001018    22
74.86% :     0.000762s :     13: func_graph_cloner_run.FuncGraphClonerGraph
25.14% :     0.000256s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021147    72
 0.16% :     0.000034s :      1: add_recomputation
 0.44% :     0.000093s :      1: auto_monad
 0.06% :     0.000013s :      1: auto_monad_reorder
 0.98% :     0.000207s :      1: bootstrap
 0.11% :     0.000023s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.20% :     0.000042s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
35.29% :     0.007463s :      1: jit_opt_a
 0.59% :     0.000125s :      1: jit_opt_after_cconv
 0.19% :     0.000040s :      1: jit_opt_b
 2.04% :     0.000432s :      1: loop_unroll
 2.33% :     0.000493s :      1: mutable_eliminate
 5.46% :     0.001156s :     26: opt.transform.jit_opt_a
 0.14% :     0.000030s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000014s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000022s :      4: opt.transform.symbol_engine_opt
 2.24% :     0.000475s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000025s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000015s :      1: remove_dup_value
 5.03% :     0.001065s :      1: renormalize.infer
 1.46% :     0.000309s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000033s :      1: rewriter_after_opt_a
 0.41% :     0.000087s :      1: rewriter_before_opt_a
 0.31% :     0.000066s :      1: symbol_engine_optimizer
41.42% :     0.008759s :      1: type_inference


Time group info:
------[substitution.]   0.000529   126
 6.81% :     0.000036s :      2: substitution.arithmetic_simplify
 0.51% :     0.000003s :      5: substitution.elim_not_effective
 0.48% :     0.000003s :      5: substitution.fold_const_symbol
 1.37% :     0.000007s :      6: substitution.graph_param_transform
58.80% :     0.000311s :     12: substitution.inline
 3.36% :     0.000018s :      2: substitution.inline_without_move
 1.85% :     0.000010s :     18: substitution.j_node_and_user_rematch
 1.85% :     0.000010s :      7: substitution.minmaximum_grad
 1.43% :     0.000008s :      5: substitution.partial_eliminate
 2.42% :     0.000013s :     18: substitution.remove_not_recompute_node
 4.73% :     0.000025s :     10: substitution.replace_applicator
 1.55% :     0.000008s :      7: substitution.replace_old_param
 0.41% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.34% :     0.000023s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.79% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 7.29% :     0.000039s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009205     2
89.42% :     0.008231s :      1: type_inference.infer
10.58% :     0.000974s :      1: type_inference.specialize
------[replace.]   0.000143    19
65.68% :     0.000094s :     12: replace.inline
34.32% :     0.000049s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000323    19
94.20% :     0.000305s :     12: match.inline
 5.80% :     0.000019s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000405  2895
 1.66% :     0.000007s :     48: predicate.accumulaten_eliminater
 0.52% :     0.000002s :      6: predicate.ad_related_special_op_eliminate
 1.48% :     0.000006s :     48: predicate.addn_check_dump
 1.71% :     0.000007s :     48: predicate.addn_zero_filter
 2.17% :     0.000009s :     48: predicate.arithmetic_simplify
 1.59% :     0.000006s :     48: predicate.cast_eliminate
 0.24% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.45% :     0.000006s :     48: predicate.compare_switch_simplify
 1.60% :     0.000006s :     48: predicate.depend_value_elim
 1.72% :     0.000007s :     48: predicate.dict_get_item_const_eliminator
 1.55% :     0.000006s :     48: predicate.dict_get_item_eliminator
 1.52% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.34% :     0.000001s :      6: predicate.dumpgradient_eliminate
 0.15% :     0.000001s :      6: predicate.elim_not_effective
 0.38% :     0.000002s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.51% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.46% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.43% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.63% :     0.000007s :     48: predicate.environ_get_eliminate
 1.44% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.12% :     0.000001s :      6: predicate.fold_const_symbol
 0.88% :     0.000004s :     23: predicate.get_grad_eliminate
 0.13% :     0.000001s :      6: predicate.graph_param_transform
 4.19% :     0.000017s :     79: predicate.inline
 1.70% :     0.000007s :     44: predicate.inline_without_move
 0.37% :     0.000002s :     23: predicate.j_node_and_user_rematch
 1.03% :     0.000004s :     23: predicate.less_batch_normalization
 1.84% :     0.000007s :     55: predicate.list_to_tuple_eliminator_
 1.95% :     0.000008s :     61: predicate.load_eliminater
 0.50% :     0.000002s :      6: predicate.loop_unroll_after_grad
 3.12% :     0.000013s :     90: predicate.loop_unroll_before_grad
 1.85% :     0.000008s :     54: predicate.make_slice_get_slice_eliminator
 1.46% :     0.000006s :     48: predicate.merge_addn
 1.75% :     0.000007s :     48: predicate.minmaximum_grad
 0.58% :     0.000002s :      6: predicate.mutable_eliminate
 0.22% :     0.000001s :      6: predicate.opt_reshape
 2.28% :     0.000009s :     61: predicate.partial_eliminate
 1.49% :     0.000006s :     48: predicate.print_const_string_wrapper
 1.73% :     0.000007s :     48: predicate.reduce_eliminate
 1.77% :     0.000007s :     55: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.45% :     0.000010s :    108: predicate.replace_applicator
 0.88% :     0.000004s :     44: predicate.replace_old_param
 0.15% :     0.000001s :      6: predicate.reset_defer_inline
 1.62% :     0.000007s :     48: predicate.reshape_eliminate
 1.62% :     0.000007s :     48: predicate.row_tensor_add_zeros_like
 0.39% :     0.000002s :      6: predicate.row_tensor_eliminate
 1.51% :     0.000006s :     48: predicate.same_eliminate
 0.53% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.46% :     0.000002s :     12: predicate.special_op_eliminate
 0.91% :     0.000004s :     23: predicate.specialize_transform
 1.68% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000006s :     48: predicate.stack_unstack_eliminate
 0.22% :     0.000001s :      6: predicate.switch_call_monad_eliminater
 2.55% :     0.000010s :     67: predicate.switch_defer_inline
 2.42% :     0.000010s :     67: predicate.switch_layer_defer_inline
 5.88% :     0.000024s :    163: predicate.switch_simplify
 1.49% :     0.000006s :     48: predicate.tile_eliminate
 1.52% :     0.000006s :     48: predicate.transpose_eliminate
 1.79% :     0.000007s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.64% :     0.000007s :     48: predicate.tuple_list_get_item_depend_reorder
 3.12% :     0.000013s :     67: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000007s :     48: predicate.tuple_list_set_item_eliminator
 1.79% :     0.000007s :     55: predicate.tuple_to_list_eliminator_
 1.89% :     0.000008s :     61: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000012s :     84: predicate.updatestate_useless_node_eliminater
 1.71% :     0.000007s :     48: predicate.value_based_eliminate
 0.21% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.35% :     0.000001s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001378    29
60.14% :     0.000829s :     13: func_graph_cloner_run.FuncGraphClonerGraph
39.86% :     0.000549s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.037734    91
 0.16% :     0.000060s :      1: add_recomputation
 0.24% :     0.000090s :      1: auto_monad
 0.06% :     0.000024s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.53% :     0.000201s :      1: bootstrap
 0.07% :     0.000026s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000036s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.12% :     0.000046s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
30.35% :     0.011452s :      1: jit_opt_a
 0.50% :     0.000189s :      1: jit_opt_after_cconv
 0.19% :     0.000070s :      1: jit_opt_b
 1.12% :     0.000422s :      1: loop_unroll
 1.76% :     0.000664s :      1: mutable_eliminate
 7.42% :     0.002802s :     39: opt.transform.jit_opt_a
 0.19% :     0.000073s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000050s :      4: opt.transform.symbol_engine_opt
 1.45% :     0.000548s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000025s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000025s :      1: remove_dup_value
 4.52% :     0.001707s :      2: renormalize.infer
 3.66% :     0.001382s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000053s :      1: rewriter_after_opt_a
 0.23% :     0.000086s :      1: rewriter_before_opt_a
 0.25% :     0.000093s :      1: symbol_engine_optimizer
21.39% :     0.008070s :      1: task_emit
24.57% :     0.009273s :      1: type_inference
 0.18% :     0.000068s :      1: validate

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.143.473 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843c1820> should be wrapped by jit only once.

TotalTime = 0.0199594, [33]
[bootstrap]: 0.00020523
[type_inference]: 0.0093412
[event_method]: 4.502e-05
[auto_monad]: 7.615e-05
[graph_reusing]: 5.92001e-06
[pre_auto_parallel]: 2.22001e-06
[py_interpret_to_execute]: 2.291e-05
[rewriter_before_opt_a]: 8.377e-05
[expand_dump_flag]: 2.39001e-06
[jit_opt_a]: 0.00790183, [2]
    [Cycle 1]: 0.00440782, [27]
        [switch_simplify]: 5.986e-05
        [loop_unroll]: 3.75e-05
        [a_1]: 0.00081789
        [with_stream_mark]: 2.154e-05
        [recompute_prepare]: 1.842e-05
        [updatestate_depend_eliminate]: 7.5e-06
        [updatestate_assign_eliminate]: 5.66e-06
        [updatestate_loads_eliminate]: 5.46e-06
        [parameter_eliminate]: 1.80001e-06
        [specialize_transform]: 1.265e-05
        [updatestate_useless_node_eliminater]: 2.234e-05
        [accelerated_algorithm]: 1.256e-05
        [meta_shard_fg_expand]: 2.29999e-06
        [get_grad_eliminate_]: 1.175e-05
        [merge_forward]: 7.23e-06
        [cell_reuse_recompute_pass]: 7.50006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.33e-05
        [j_node_and_user_rematch]: 2.182e-05
        [meta_fg_expand]: 0.00132645
        [replace_old_param]: 4.654e-05
        [inline_without_move]: 4.878e-05
        [renormalize]: 0.00164578
        [add_forward_monad_depend]: 5.34998e-06
        [auto_monad_grad]: 2.02999e-06
        [auto_monad_eliminator]: 9.71e-06
        [cse]: 1.872e-05
        [replace_applicator]: 1.139e-05
    [Cycle 2]: 0.00028862, [27]
        [switch_simplify]: 4.39998e-06
        [loop_unroll]: 3.55e-06
        [a_1]: 4.198e-05
        [with_stream_mark]: 8.53001e-06
        [recompute_prepare]: 3.63e-06
        [updatestate_depend_eliminate]: 2.21e-06
        [updatestate_assign_eliminate]: 1.72001e-06
        [updatestate_loads_eliminate]: 1.55999e-06
        [parameter_eliminate]: 1.35001e-06
        [specialize_transform]: 3.51001e-06
        [updatestate_useless_node_eliminater]: 3.43e-06
        [accelerated_algorithm]: 5.66e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 3.39001e-06
        [merge_forward]: 2.46e-06
        [cell_reuse_recompute_pass]: 2.56998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.069e-05
        [j_node_and_user_rematch]: 5.91998e-06
        [meta_fg_expand]: 3.702e-05
        [replace_old_param]: 3.87998e-06
        [inline_without_move]: 3.16999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.58002e-06
        [auto_monad_grad]: 1.04e-06
        [auto_monad_eliminator]: 4.87e-06
        [cse]: 1.042e-05
        [replace_applicator]: 3.87002e-06
[py_interpret_to_execute_after_opt_a]: 9.25999e-06
[rewriter_after_opt_a]: 2.98e-05
[convert_after_rewriter]: 5.36998e-06
[order_py_execute_after_rewriter]: 4.35e-06
[mutable_eliminate]: 0.0006279
[jit_opt_b]: 3.812e-05, [1]
    [Cycle 1]: 3.159e-05, [2]
        [frontend_op_eliminate]: 1.044e-05
        [inline_after_opt_a]: 9.97001e-06
[cconv]: 1.953e-05
[loop_unroll]: 0.00047607
[jit_opt_after_cconv]: 0.00012426, [1]
    [Cycle 1]: 0.00011724, [11]
        [c_1]: 1.315e-05
        [parameter_eliminate]: 2.30002e-06
        [updatestate_depend_eliminate]: 4.11001e-06
        [updatestate_assign_eliminate]: 1.79998e-06
        [updatestate_loads_eliminate]: 1.84e-06
        [cse]: 1.739e-05
        [call_graph_tuple_transform]: 1.392e-05
        [tuple_list_get_item_eliminator]: 3.83001e-06
        [none_parameter_eliminate]: 1.25999e-06
        [renormalize]: 4.80009e-07
        [switch_simplify]: 4.37e-06
[remove_dup_value]: 1.357e-05
[partial_unused_args_eliminate]: 2.53998e-06
[environ_conv]: 5.33002e-06
[add_recomputation]: 3.235e-05
[cse_after_recomputation]: 2.022e-05, [1]
    [Cycle 1]: 1.48e-05, [1]
        [cse]: 8.53001e-06
[auto_monad_reorder]: 1.116e-05
[get_jit_bprop_graph]: 2.01e-06
[rewriter_after_jit_bprop_graph]: 5.77999e-06
[opt_after_jit_grad]: 0.00053398
[symbol_engine_optimizer]: 7.063e-05, [1]
    [Cycle 1]: 6.391e-05, [6]
        [build]: 4.4e-06
        [elim_shapecalc]: 6.44999e-06
        [elim_not_effective]: 1.128e-05
        [opt_reshape]: 4.89e-06
        [fold_const_symbol]: 6.78e-06
        [renormalize]: 4.30009e-07
[validate]: 3.358e-05
[backend_pass]: 6.69999e-07
[task_emit]: 2.482e-05
[execute]: 1.42999e-06
Sums
    bootstrap                            :     0.000205s :  1.27%
    type_inference                       :     0.009341s : 57.91%
    event_method                         :     0.000045s :  0.28%
    auto_monad                           :     0.000076s :  0.47%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000023s :  0.14%
    rewriter_before_opt_a                :     0.000084s :  0.52%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000064s :  0.40%
    jit_opt_a.loop_unroll                :     0.000041s :  0.25%
    jit_opt_a.a_1                        :     0.000860s :  5.33%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000016s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000026s :  0.16%
    jit_opt_a.accelerated_algorithm      :     0.000018s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.09%
    jit_opt_a.merge_forward              :     0.000010s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001363s :  8.45%
    jit_opt_a.replace_old_param          :     0.000050s :  0.31%
    jit_opt_a.inline_without_move        :     0.000052s :  0.32%
    jit_opt_a.renormalize                :     0.001646s : 10.20%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.09%
    jit_opt_a.cse                        :     0.000029s :  0.18%
    jit_opt_a.replace_applicator         :     0.000015s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000030s :  0.18%
    convert_after_rewriter               :     0.000005s :  0.03%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000628s :  3.89%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000010s :  0.06%
    cconv                                :     0.000020s :  0.12%
    loop_unroll                          :     0.000476s :  2.95%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.03%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000032s :  0.20%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000011s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000534s :  3.31%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000007s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000034s :  0.21%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.000025s :  0.15%
    execute                              :     0.000001s :  0.01%


Time group info:
------[substitution.]   0.000225    50
 0.83% :     0.000002s :      1: substitution.elim_not_effective
 0.59% :     0.000001s :      1: substitution.fold_const_symbol
 1.55% :     0.000003s :      1: substitution.graph_param_transform
66.63% :     0.000150s :      6: substitution.inline
 6.75% :     0.000015s :      2: substitution.inline_without_move
 2.67% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.36% :     0.000003s :      2: substitution.minmaximum_grad
 1.52% :     0.000003s :      3: substitution.partial_eliminate
 2.77% :     0.000006s :      9: substitution.remove_not_recompute_node
 2.34% :     0.000005s :      2: substitution.replace_applicator
 1.59% :     0.000004s :      5: substitution.replace_old_param
 1.32% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.80% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.85% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 5.43% :     0.000012s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009279     2
89.38% :     0.008293s :      1: type_inference.infer
10.62% :     0.000986s :      1: type_inference.specialize
------[replace.]   0.000060     8
77.07% :     0.000047s :      6: replace.inline
22.93% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000154     8
95.10% :     0.000147s :      6: match.inline
 4.90% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000180  1109
 1.62% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.61% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.30% :     0.000002s :     18: predicate.addn_check_dump
 1.54% :     0.000003s :     18: predicate.addn_zero_filter
 1.97% :     0.000004s :     18: predicate.arithmetic_simplify
 1.58% :     0.000003s :     18: predicate.cast_eliminate
 0.16% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.30% :     0.000002s :     18: predicate.compare_switch_simplify
 1.36% :     0.000002s :     18: predicate.depend_value_elim
 1.29% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.23% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.52% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.17% :     0.000000s :      1: predicate.elim_not_effective
 0.30% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.38% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.23% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.21% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.53% :     0.000003s :     18: predicate.environ_get_eliminate
 1.23% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.11% :     0.000002s :     12: predicate.get_grad_eliminate
 0.13% :     0.000000s :      1: predicate.graph_param_transform
 3.93% :     0.000007s :     28: predicate.inline
 2.86% :     0.000005s :     33: predicate.inline_without_move
 0.46% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.19% :     0.000002s :     12: predicate.less_batch_normalization
 1.50% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.75% :     0.000003s :     21: predicate.load_eliminater
 0.81% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.31% :     0.000006s :     44: predicate.loop_unroll_before_grad
 2.12% :     0.000004s :     19: predicate.make_slice_get_slice_eliminator
 1.32% :     0.000002s :     18: predicate.merge_addn
 1.29% :     0.000002s :     18: predicate.minmaximum_grad
 0.62% :     0.000001s :      1: predicate.mutable_eliminate
 0.27% :     0.000000s :      1: predicate.opt_reshape
 1.90% :     0.000003s :     21: predicate.partial_eliminate
 1.34% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.61% :     0.000003s :     18: predicate.reduce_eliminate
 1.65% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.40% :     0.000003s :     22: predicate.replace_applicator
 1.41% :     0.000003s :     33: predicate.replace_old_param
 0.08% :     0.000000s :      1: predicate.reset_defer_inline
 1.53% :     0.000003s :     18: predicate.reshape_eliminate
 1.43% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.37% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     18: predicate.same_eliminate
 0.57% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.32% :     0.000001s :      2: predicate.special_op_eliminate
 1.15% :     0.000002s :     12: predicate.specialize_transform
 1.65% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.48% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.11% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.39% :     0.000004s :     26: predicate.switch_defer_inline
 2.31% :     0.000004s :     26: predicate.switch_layer_defer_inline
 5.96% :     0.000011s :     71: predicate.switch_simplify
 1.30% :     0.000002s :     18: predicate.tile_eliminate
 1.36% :     0.000002s :     18: predicate.transpose_eliminate
 1.50% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.49% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.69% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.70% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.67% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 8.34% :     0.000015s :     33: predicate.updatestate_useless_node_eliminater
 1.79% :     0.000003s :     18: predicate.value_based_eliminate
 0.09% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.21% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001077    22
72.18% :     0.000778s :     13: func_graph_cloner_run.FuncGraphClonerGraph
27.82% :     0.000300s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022888    76
 0.15% :     0.000035s :      1: add_recomputation
 0.35% :     0.000081s :      1: auto_monad
 0.06% :     0.000015s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.96% :     0.000219s :      1: bootstrap
 0.10% :     0.000022s :      1: cconv
 0.03% :     0.000008s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.22% :     0.000049s :      1: event_method
 0.02% :     0.000003s :      1: execute
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
34.54% :     0.007905s :      1: jit_opt_a
 0.56% :     0.000127s :      1: jit_opt_after_cconv
 0.18% :     0.000041s :      1: jit_opt_b
 2.12% :     0.000484s :      1: loop_unroll
 2.77% :     0.000635s :      1: mutable_eliminate
 5.31% :     0.001216s :     26: opt.transform.jit_opt_a
 0.14% :     0.000032s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000015s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000018s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000026s :      4: opt.transform.symbol_engine_opt
 2.38% :     0.000545s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.11% :     0.000025s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 5.36% :     0.001227s :      1: renormalize.infer
 1.80% :     0.000412s :      1: renormalize.specialize
 0.03% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000033s :      1: rewriter_after_opt_a
 0.38% :     0.000086s :      1: rewriter_before_opt_a
 0.32% :     0.000073s :      1: symbol_engine_optimizer
 0.13% :     0.000029s :      1: task_emit
40.86% :     0.009353s :      1: type_inference
 0.23% :     0.000053s :      1: validate

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.144.051 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.146.777 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda84a8c10> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.147.347 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0132337, [30]
[bootstrap]: 0.00030299
[type_inference]: 0.00449019
[event_method]: 1.866e-05
[auto_monad]: 7.602e-05
[graph_reusing]: 6.74001e-06
[pre_auto_parallel]: 3.17002e-06
[py_interpret_to_execute]: 3.189e-05
[rewriter_before_opt_a]: 7.407e-05
[expand_dump_flag]: 2.99999e-06
[jit_opt_a]: 0.00587135, [2]
    [Cycle 1]: 0.00179929, [27]
        [switch_simplify]: 6.183e-05
        [loop_unroll]: 2.9e-05
        [a_1]: 0.00058664
        [with_stream_mark]: 2.691e-05
        [recompute_prepare]: 8.11002e-06
        [updatestate_depend_eliminate]: 4.24002e-06
        [updatestate_assign_eliminate]: 3.23e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 5.98998e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.90002e-06
        [meta_shard_fg_expand]: 2.22999e-06
        [get_grad_eliminate_]: 5.11002e-06
        [merge_forward]: 3.9e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.469e-05
        [j_node_and_user_rematch]: 9.34998e-06
        [meta_fg_expand]: 2.61e-06
        [replace_old_param]: 1.039e-05
        [inline_without_move]: 6.28e-06
        [renormalize]: 0.00073875
        [add_forward_monad_depend]: 5.82001e-06
        [auto_monad_grad]: 2.80997e-06
        [auto_monad_eliminator]: 1.504e-05
        [cse]: 2.991e-05
        [replace_applicator]: 1.366e-05
    [Cycle 2]: 0.00032265, [27]
        [switch_simplify]: 5.64e-06
        [loop_unroll]: 5.62001e-06
        [a_1]: 8.874e-05
        [with_stream_mark]: 1.326e-05
        [recompute_prepare]: 5.17999e-06
        [updatestate_depend_eliminate]: 2.71e-06
        [updatestate_assign_eliminate]: 2.18998e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 8.79983e-07
        [specialize_transform]: 4.72998e-06
        [updatestate_useless_node_eliminater]: 4.68001e-06
        [accelerated_algorithm]: 4.77e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 4.35999e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 2.37001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.378e-05
        [j_node_and_user_rematch]: 7.68001e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 7.58999e-06
        [inline_without_move]: 4.48001e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.42e-06
        [auto_monad_grad]: 1.35001e-06
        [auto_monad_eliminator]: 5.20001e-06
        [cse]: 1.016e-05
        [replace_applicator]: 6.89999e-06
[py_interpret_to_execute_after_opt_a]: 1.314e-05
[rewriter_after_opt_a]: 3.725e-05
[convert_after_rewriter]: 7.03e-06
[order_py_execute_after_rewriter]: 4.87e-06
[mutable_eliminate]: 0.00067419
[jit_opt_b]: 5.231e-05, [1]
    [Cycle 1]: 4.418e-05, [2]
        [frontend_op_eliminate]: 1.703e-05
        [inline_after_opt_a]: 1.552e-05
[cconv]: 3.01e-05
[loop_unroll]: 0.00049562
[jit_opt_after_cconv]: 0.00014769, [1]
    [Cycle 1]: 0.00014112, [11]
        [c_1]: 2.075e-05
        [parameter_eliminate]: 4.07e-06
        [updatestate_depend_eliminate]: 6.78e-06
        [updatestate_assign_eliminate]: 2.74999e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 2.278e-05
        [call_graph_tuple_transform]: 1.982e-05
        [tuple_list_get_item_eliminator]: 5.22999e-06
        [none_parameter_eliminate]: 1.52999e-06
        [renormalize]: 6.29982e-07
        [switch_simplify]: 5.07e-06
[remove_dup_value]: 1.43e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 5.76e-06
[add_recomputation]: 5.207e-05
[cse_after_recomputation]: 1.97e-05, [1]
    [Cycle 1]: 1.435e-05, [1]
        [cse]: 8.67998e-06
[auto_monad_reorder]: 1.521e-05
[get_jit_bprop_graph]: 2.04e-06
[rewriter_after_jit_bprop_graph]: 7.12002e-06
[opt_after_jit_grad]: 0.00047504
[symbol_engine_optimizer]: 7.614e-05, [1]
    [Cycle 1]: 6.976e-05, [6]
        [build]: 4.57998e-06
        [elim_shapecalc]: 8.13001e-06
        [elim_not_effective]: 1.317e-05
        [opt_reshape]: 5.73997e-06
        [fold_const_symbol]: 8.59998e-06
        [renormalize]: 5.8001e-07
[validate]: 3.95e-05
Sums
    bootstrap                            :     0.000303s :  3.42%
    type_inference                       :     0.004490s : 50.62%
    event_method                         :     0.000019s :  0.21%
    auto_monad                           :     0.000076s :  0.86%
    graph_reusing                        :     0.000007s :  0.08%
    pre_auto_parallel                    :     0.000003s :  0.04%
    py_interpret_to_execute              :     0.000032s :  0.36%
    rewriter_before_opt_a                :     0.000074s :  0.84%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000067s :  0.76%
    jit_opt_a.loop_unroll                :     0.000035s :  0.39%
    jit_opt_a.a_1                        :     0.000675s :  7.61%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.45%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000018s :  0.20%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000739s :  8.33%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.23%
    jit_opt_a.cse                        :     0.000040s :  0.45%
    jit_opt_a.replace_applicator         :     0.000021s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.15%
    rewriter_after_opt_a                 :     0.000037s :  0.42%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000674s :  7.60%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.17%
    cconv                                :     0.000030s :  0.34%
    loop_unroll                          :     0.000496s :  5.59%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.26%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000052s :  0.59%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.08%
    opt_after_jit_grad                   :     0.000475s :  5.36%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.45%


Time group info:
------[substitution.]   0.000246    23
 0.86% :     0.000002s :      2: substitution.elim_not_effective
 0.53% :     0.000001s :      2: substitution.fold_const_symbol
 2.35% :     0.000006s :      3: substitution.graph_param_transform
85.30% :     0.000210s :      4: substitution.inline
 1.54% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.87% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.91% :     0.000005s :      2: substitution.replace_old_param
 5.64% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004414     2
80.92% :     0.003571s :      1: type_inference.infer
19.08% :     0.000842s :      1: type_inference.specialize
------[replace.]   0.000072     6
67.97% :     0.000049s :      4: replace.inline
32.03% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000220     6
94.29% :     0.000207s :      4: match.inline
 5.71% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.61% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.58% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.56% :     0.000002s :     10: predicate.addn_zero_filter
 2.69% :     0.000003s :     10: predicate.arithmetic_simplify
 1.06% :     0.000001s :     10: predicate.cast_eliminate
 0.63% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.20% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.01% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.39% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.07% :     0.000001s :     10: predicate.environ_get_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.83% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000001s :      3: predicate.graph_param_transform
 5.69% :     0.000007s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.26% :     0.000002s :      6: predicate.less_batch_normalization
 1.35% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.57% :     0.000002s :     15: predicate.load_eliminater
 1.81% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.65% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 2.00% :     0.000003s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.68% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.90% :     0.000002s :     10: predicate.reduce_eliminate
 1.31% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.63% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.06% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.17% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.52% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.42% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.52% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.55% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.67% :     0.000010s :     53: predicate.switch_simplify
 1.28% :     0.000002s :     10: predicate.tile_eliminate
 1.06% :     0.000001s :     10: predicate.transpose_eliminate
 1.17% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.05% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.60% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.88% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.58% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.87% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000620    12
49.08% :     0.000304s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.96% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.96% :     0.000266s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014923    72
 0.37% :     0.000055s :      1: add_recomputation
 0.54% :     0.000080s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.17% :     0.000323s :      1: bootstrap
 0.22% :     0.000033s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
39.36% :     0.005874s :      1: jit_opt_a
 1.01% :     0.000150s :      1: jit_opt_after_cconv
 0.37% :     0.000055s :      1: jit_opt_b
 3.38% :     0.000504s :      1: loop_unroll
 4.59% :     0.000685s :      1: mutable_eliminate
 6.01% :     0.000897s :     26: opt.transform.jit_opt_a
 0.31% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.24% :     0.000484s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000035s :      1: py_interpret_to_execute
 0.10% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 2.69% :     0.000401s :      1: renormalize.infer
 2.20% :     0.000329s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000040s :      1: rewriter_after_opt_a
 0.51% :     0.000077s :      1: rewriter_before_opt_a
 0.53% :     0.000079s :      1: symbol_engine_optimizer
30.20% :     0.004507s :      1: type_inference

.
TotalTime = 0.0206315, [30]
[bootstrap]: 0.00021255
[type_inference]: 0.00979027
[event_method]: 4.302e-05
[auto_monad]: 9.323e-05
[graph_reusing]: 6.89001e-06
[pre_auto_parallel]: 2.98e-06
[py_interpret_to_execute]: 2.792e-05
[rewriter_before_opt_a]: 8.886e-05
[expand_dump_flag]: 2.85998e-06
[jit_opt_a]: 0.00825854, [2]
    [Cycle 1]: 0.00460612, [27]
        [switch_simplify]: 6.089e-05
        [loop_unroll]: 3.782e-05
        [a_1]: 0.00082048
        [with_stream_mark]: 2.366e-05
        [recompute_prepare]: 1.794e-05
        [updatestate_depend_eliminate]: 7.37002e-06
        [updatestate_assign_eliminate]: 6.12999e-06
        [updatestate_loads_eliminate]: 5.22e-06
        [parameter_eliminate]: 2.71e-06
        [specialize_transform]: 1.384e-05
        [updatestate_useless_node_eliminater]: 1.19e-05
        [accelerated_algorithm]: 1.226e-05
        [meta_shard_fg_expand]: 2.91999e-06
        [get_grad_eliminate_]: 1.164e-05
        [merge_forward]: 7.85e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.511e-05
        [j_node_and_user_rematch]: 2.212e-05
        [meta_fg_expand]: 0.00145305
        [replace_old_param]: 5.763e-05
        [inline_without_move]: 5.315e-05
        [renormalize]: 0.00168749
        [add_forward_monad_depend]: 5.71998e-06
        [auto_monad_grad]: 2.64999e-06
        [auto_monad_eliminator]: 1.316e-05
        [cse]: 2.737e-05
        [replace_applicator]: 1.164e-05
    [Cycle 2]: 0.00025983, [27]
        [switch_simplify]: 4.31002e-06
        [loop_unroll]: 3.89002e-06
        [a_1]: 3.877e-05
        [with_stream_mark]: 8.23999e-06
        [recompute_prepare]: 3.65998e-06
        [updatestate_depend_eliminate]: 2.14999e-06
        [updatestate_assign_eliminate]: 1.62001e-06
        [updatestate_loads_eliminate]: 1.39e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 3.44001e-06
        [updatestate_useless_node_eliminater]: 3.3e-06
        [accelerated_algorithm]: 3.29001e-06
        [meta_shard_fg_expand]: 1.08001e-06
        [get_grad_eliminate_]: 3.25e-06
        [merge_forward]: 1.93002e-06
        [cell_reuse_recompute_pass]: 1.19e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.57001e-06
        [j_node_and_user_rematch]: 5.14e-06
        [meta_fg_expand]: 2.357e-05
        [replace_old_param]: 3.56001e-06
        [inline_without_move]: 3.23e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.60999e-06
        [auto_monad_grad]: 8.2e-07
        [auto_monad_eliminator]: 4.96002e-06
        [cse]: 9.49999e-06
        [replace_applicator]: 3.87998e-06
[py_interpret_to_execute_after_opt_a]: 1.042e-05
[rewriter_after_opt_a]: 3.386e-05
[convert_after_rewriter]: 6.41998e-06
[order_py_execute_after_rewriter]: 5.02999e-06
[mutable_eliminate]: 0.00053782
[jit_opt_b]: 4.055e-05, [1]
    [Cycle 1]: 3.344e-05, [2]
        [frontend_op_eliminate]: 1.111e-05
        [inline_after_opt_a]: 1.117e-05
[cconv]: 2.056e-05
[loop_unroll]: 0.00044324
[jit_opt_after_cconv]: 0.00012911, [1]
    [Cycle 1]: 0.00012123, [11]
        [c_1]: 1.281e-05
        [parameter_eliminate]: 3.18e-06
        [updatestate_depend_eliminate]: 5.15001e-06
        [updatestate_assign_eliminate]: 1.97001e-06
        [updatestate_loads_eliminate]: 1.84998e-06
        [cse]: 1.975e-05
        [call_graph_tuple_transform]: 1.487e-05
        [tuple_list_get_item_eliminator]: 4.09002e-06
        [none_parameter_eliminate]: 1.29e-06
        [renormalize]: 1.12e-06
        [switch_simplify]: 4.03001e-06
[remove_dup_value]: 1.431e-05
[partial_unused_args_eliminate]: 2.64001e-06
[environ_conv]: 4.80001e-06
[add_recomputation]: 3.491e-05
[cse_after_recomputation]: 1.997e-05, [1]
    [Cycle 1]: 1.346e-05, [1]
        [cse]: 7.8e-06
[auto_monad_reorder]: 1.309e-05
[get_jit_bprop_graph]: 1.77999e-06
[rewriter_after_jit_bprop_graph]: 6.46e-06
[opt_after_jit_grad]: 0.00051916
[symbol_engine_optimizer]: 6.872e-05, [1]
    [Cycle 1]: 6.148e-05, [6]
        [build]: 4.28999e-06
        [elim_shapecalc]: 6.56e-06
        [elim_not_effective]: 1.124e-05
        [opt_reshape]: 4.23001e-06
        [fold_const_symbol]: 6.23002e-06
        [renormalize]: 1.06002e-06
[validate]: 2.821e-05
Sums
    bootstrap                            :     0.000213s :  1.28%
    type_inference                       :     0.009790s : 58.85%
    event_method                         :     0.000043s :  0.26%
    auto_monad                           :     0.000093s :  0.56%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.17%
    rewriter_before_opt_a                :     0.000089s :  0.53%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.39%
    jit_opt_a.loop_unroll                :     0.000042s :  0.25%
    jit_opt_a.a_1                        :     0.000859s :  5.16%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.09%
    jit_opt_a.merge_forward              :     0.000010s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000035s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.16%
    jit_opt_a.meta_fg_expand             :     0.001477s :  8.88%
    jit_opt_a.replace_old_param          :     0.000061s :  0.37%
    jit_opt_a.inline_without_move        :     0.000056s :  0.34%
    jit_opt_a.renormalize                :     0.001688s : 10.14%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.11%
    jit_opt_a.cse                        :     0.000037s :  0.22%
    jit_opt_a.replace_applicator         :     0.000016s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000034s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000538s :  3.23%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000021s :  0.12%
    loop_unroll                          :     0.000443s :  2.66%
    jit_opt_after_cconv.c_1              :     0.000013s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000015s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000035s :  0.21%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000519s :  3.12%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000028s :  0.17%


Time group info:
------[substitution.]   0.000247    50
 0.59% :     0.000001s :      1: substitution.elim_not_effective
 0.40% :     0.000001s :      1: substitution.fold_const_symbol
 1.67% :     0.000004s :      1: substitution.graph_param_transform
65.16% :     0.000161s :      6: substitution.inline
 7.71% :     0.000019s :      2: substitution.inline_without_move
 2.44% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.42% :     0.000004s :      2: substitution.minmaximum_grad
 1.63% :     0.000004s :      3: substitution.partial_eliminate
 2.69% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.34% :     0.000006s :      2: substitution.replace_applicator
 1.97% :     0.000005s :      5: substitution.replace_old_param
 1.10% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.95% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.81% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 6.11% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009719     2
88.65% :     0.008616s :      1: type_inference.infer
11.35% :     0.001103s :      1: type_inference.specialize
------[replace.]   0.000061     8
77.82% :     0.000048s :      6: replace.inline
22.18% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000168     8
93.83% :     0.000157s :      6: match.inline
 6.17% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000169  1109
 1.43% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.66% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.37% :     0.000002s :     18: predicate.addn_check_dump
 1.62% :     0.000003s :     18: predicate.addn_zero_filter
 2.06% :     0.000003s :     18: predicate.arithmetic_simplify
 1.33% :     0.000002s :     18: predicate.cast_eliminate
 0.19% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.42% :     0.000002s :     18: predicate.compare_switch_simplify
 1.43% :     0.000002s :     18: predicate.depend_value_elim
 1.32% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.48% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.60% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      1: predicate.elim_not_effective
 0.47% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.31% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.25% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.34% :     0.000002s :     18: predicate.environ_get_eliminate
 1.32% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.11% :     0.000002s :     12: predicate.get_grad_eliminate
 0.18% :     0.000000s :      1: predicate.graph_param_transform
 4.27% :     0.000007s :     28: predicate.inline
 3.29% :     0.000006s :     33: predicate.inline_without_move
 0.55% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.46% :     0.000002s :     12: predicate.less_batch_normalization
 1.58% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.74% :     0.000003s :     21: predicate.load_eliminater
 0.97% :     0.000002s :      1: predicate.loop_unroll_after_grad
 3.59% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.89% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.35% :     0.000002s :     18: predicate.merge_addn
 1.37% :     0.000002s :     18: predicate.minmaximum_grad
 0.75% :     0.000001s :      1: predicate.mutable_eliminate
 0.19% :     0.000000s :      1: predicate.opt_reshape
 2.06% :     0.000003s :     21: predicate.partial_eliminate
 1.43% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.72% :     0.000003s :     18: predicate.reduce_eliminate
 1.57% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.46% :     0.000002s :     22: predicate.replace_applicator
 1.68% :     0.000003s :     33: predicate.replace_old_param
 0.14% :     0.000000s :      1: predicate.reset_defer_inline
 1.58% :     0.000003s :     18: predicate.reshape_eliminate
 1.37% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.49% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.49% :     0.000003s :     18: predicate.same_eliminate
 0.62% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.44% :     0.000001s :      2: predicate.special_op_eliminate
 1.18% :     0.000002s :     12: predicate.specialize_transform
 1.52% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.39% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.24% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.61% :     0.000004s :     26: predicate.switch_defer_inline
 2.39% :     0.000004s :     26: predicate.switch_layer_defer_inline
 7.43% :     0.000013s :     71: predicate.switch_simplify
 1.43% :     0.000002s :     18: predicate.tile_eliminate
 1.42% :     0.000002s :     18: predicate.transpose_eliminate
 1.57% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.43% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 2.79% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.77% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.61% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.63% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.11% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000003s :     18: predicate.value_based_eliminate
 0.10% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.25% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001124    22
72.34% :     0.000813s :     13: func_graph_cloner_run.FuncGraphClonerGraph
27.66% :     0.000311s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023567    72
 0.16% :     0.000038s :      1: add_recomputation
 0.41% :     0.000097s :      1: auto_monad
 0.07% :     0.000016s :      1: auto_monad_reorder
 0.95% :     0.000225s :      1: bootstrap
 0.10% :     0.000023s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.20% :     0.000048s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
35.06% :     0.008262s :      1: jit_opt_a
 0.56% :     0.000132s :      1: jit_opt_after_cconv
 0.18% :     0.000043s :      1: jit_opt_b
 1.92% :     0.000453s :      1: loop_unroll
 2.32% :     0.000546s :      1: mutable_eliminate
 5.17% :     0.001217s :     26: opt.transform.jit_opt_a
 0.14% :     0.000032s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000016s :      4: opt.transform.jit_opt_b
 0.05% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000025s :      4: opt.transform.symbol_engine_opt
 2.24% :     0.000529s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000031s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000017s :      1: remove_dup_value
 5.47% :     0.001290s :      1: renormalize.infer
 1.65% :     0.000390s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000037s :      1: rewriter_after_opt_a
 0.39% :     0.000092s :      1: rewriter_before_opt_a
 0.30% :     0.000071s :      1: symbol_engine_optimizer
41.60% :     0.009804s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.171.020 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843c1e50> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.171.600 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0339987, [33]
[bootstrap]: 0.00019778
[type_inference]: 0.0100849
[event_method]: 4.278e-05
[auto_monad]: 8.223e-05
[graph_reusing]: 5.61e-06
[pre_auto_parallel]: 2.82002e-06
[py_interpret_to_execute]: 2.428e-05
[rewriter_before_opt_a]: 8.639e-05
[expand_dump_flag]: 3.33e-06
[jit_opt_a]: 0.0124551, [3]
    [Cycle 1]: 0.00597869, [27]
        [switch_simplify]: 6.242e-05
        [loop_unroll]: 3.755e-05
        [a_1]: 0.00080054
        [with_stream_mark]: 2.624e-05
        [recompute_prepare]: 2.076e-05
        [updatestate_depend_eliminate]: 8.08001e-06
        [updatestate_assign_eliminate]: 6.12001e-06
        [updatestate_loads_eliminate]: 5.89999e-06
        [parameter_eliminate]: 2.28998e-06
        [specialize_transform]: 1.329e-05
        [updatestate_useless_node_eliminater]: 1.143e-05
        [accelerated_algorithm]: 1.207e-05
        [meta_shard_fg_expand]: 3.3e-06
        [get_grad_eliminate_]: 1.19e-05
        [merge_forward]: 8.24002e-06
        [cell_reuse_recompute_pass]: 7.30011e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.642e-05
        [j_node_and_user_rematch]: 2.3e-05
        [meta_fg_expand]: 0.00147705
        [replace_old_param]: 5.117e-05
        [inline_without_move]: 5.315e-05
        [renormalize]: 0.00278203
        [add_forward_monad_depend]: 1.281e-05
        [auto_monad_grad]: 5.51e-06
        [auto_monad_eliminator]: 5.834e-05
        [cse]: 0.00015931
        [replace_applicator]: 7.729e-05
    [Cycle 2]: 0.00264838, [27]
        [switch_simplify]: 3.808e-05
        [loop_unroll]: 3.581e-05
        [a_1]: 0.00121555
        [with_stream_mark]: 2.643e-05
        [recompute_prepare]: 1.395e-05
        [updatestate_depend_eliminate]: 6.06e-06
        [updatestate_assign_eliminate]: 4.72e-06
        [updatestate_loads_eliminate]: 5.12999e-06
        [parameter_eliminate]: 2.02001e-06
        [specialize_transform]: 9.32999e-06
        [updatestate_useless_node_eliminater]: 8.74003e-06
        [accelerated_algorithm]: 8.32e-06
        [meta_shard_fg_expand]: 3.04999e-06
        [get_grad_eliminate_]: 9.48002e-06
        [merge_forward]: 5.67999e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.27e-05
        [j_node_and_user_rematch]: 1.555e-05
        [meta_fg_expand]: 5.802e-05
        [replace_old_param]: 1.521e-05
        [inline_without_move]: 8.30999e-06
        [renormalize]: 0.00085449
        [add_forward_monad_depend]: 7.35998e-06
        [auto_monad_grad]: 1.86e-06
        [auto_monad_eliminator]: 1.901e-05
        [cse]: 5.253e-05
        [replace_applicator]: 1.784e-05
    [Cycle 3]: 0.0005224, [27]
        [switch_simplify]: 8.72e-06
        [loop_unroll]: 7.93999e-06
        [a_1]: 0.00019578
        [with_stream_mark]: 1.452e-05
        [recompute_prepare]: 8.25999e-06
        [updatestate_depend_eliminate]: 4.82e-06
        [updatestate_assign_eliminate]: 5.20001e-06
        [updatestate_loads_eliminate]: 4.12e-06
        [parameter_eliminate]: 1.37e-06
        [specialize_transform]: 7.9e-06
        [updatestate_useless_node_eliminater]: 8.02e-06
        [accelerated_algorithm]: 8.85999e-06
        [meta_shard_fg_expand]: 1.61998e-06
        [get_grad_eliminate_]: 8.15999e-06
        [merge_forward]: 4.68999e-06
        [cell_reuse_recompute_pass]: 1.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.952e-05
        [j_node_and_user_rematch]: 1.397e-05
        [meta_fg_expand]: 2.91e-06
        [replace_old_param]: 1.148e-05
        [inline_without_move]: 8.10999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.23002e-06
        [auto_monad_grad]: 9.89996e-07
        [auto_monad_eliminator]: 1.204e-05
        [cse]: 2.788e-05
        [replace_applicator]: 9.68997e-06
[py_interpret_to_execute_after_opt_a]: 1.788e-05
[rewriter_after_opt_a]: 5.253e-05
[convert_after_rewriter]: 9.76e-06
[order_py_execute_after_rewriter]: 6.59999e-06
[mutable_eliminate]: 0.00074192
[jit_opt_b]: 6.699e-05, [1]
    [Cycle 1]: 5.856e-05, [2]
        [frontend_op_eliminate]: 2.399e-05
        [inline_after_opt_a]: 2.362e-05
[cconv]: 2.6e-05
[loop_unroll]: 0.00042572
[jit_opt_after_cconv]: 0.00021373, [1]
    [Cycle 1]: 0.00020655, [11]
        [c_1]: 3.712e-05
        [parameter_eliminate]: 4.63001e-06
        [updatestate_depend_eliminate]: 8.75001e-06
        [updatestate_assign_eliminate]: 4.75001e-06
        [updatestate_loads_eliminate]: 4.26001e-06
        [cse]: 4.593e-05
        [call_graph_tuple_transform]: 2.96e-05
        [tuple_list_get_item_eliminator]: 8.43999e-06
        [none_parameter_eliminate]: 1.54e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 8.80999e-06
[remove_dup_value]: 2.505e-05
[partial_unused_args_eliminate]: 2.66e-06
[environ_conv]: 8.27998e-06
[add_recomputation]: 5.976e-05
[cse_after_recomputation]: 3.315e-05, [1]
    [Cycle 1]: 2.703e-05, [1]
        [cse]: 2.089e-05
[auto_monad_reorder]: 2.088e-05
[get_jit_bprop_graph]: 2.06e-06
[rewriter_after_jit_bprop_graph]: 6.31998e-06
[opt_after_jit_grad]: 0.00048209
[symbol_engine_optimizer]: 9.012e-05, [1]
    [Cycle 1]: 8.373e-05, [6]
        [build]: 4.35e-06
        [elim_shapecalc]: 1.143e-05
        [elim_not_effective]: 1.824e-05
        [opt_reshape]: 9.04e-06
        [fold_const_symbol]: 1.38e-05
        [renormalize]: 6.39993e-07
[validate]: 4.386e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00845836
[execute]: 8.95001e-06
Sums
    bootstrap                            :     0.000198s :  0.66%
    type_inference                       :     0.010085s : 33.80%
    event_method                         :     0.000043s :  0.14%
    auto_monad                           :     0.000082s :  0.28%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000024s :  0.08%
    rewriter_before_opt_a                :     0.000086s :  0.29%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000109s :  0.37%
    jit_opt_a.loop_unroll                :     0.000081s :  0.27%
    jit_opt_a.a_1                        :     0.002212s :  7.41%
    jit_opt_a.with_stream_mark           :     0.000067s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000043s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000019s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000016s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000015s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000006s :  0.02%
    jit_opt_a.specialize_transform       :     0.000031s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000028s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000029s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000030s :  0.10%
    jit_opt_a.merge_forward              :     0.000019s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000069s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000053s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001538s :  5.16%
    jit_opt_a.replace_old_param          :     0.000078s :  0.26%
    jit_opt_a.inline_without_move        :     0.000070s :  0.23%
    jit_opt_a.renormalize                :     0.003637s : 12.19%
    jit_opt_a.add_forward_monad_depend   :     0.000022s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000089s :  0.30%
    jit_opt_a.cse                        :     0.000240s :  0.80%
    jit_opt_a.replace_applicator         :     0.000105s :  0.35%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.06%
    rewriter_after_opt_a                 :     0.000053s :  0.18%
    convert_after_rewriter               :     0.000010s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000742s :  2.49%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.08%
    cconv                                :     0.000026s :  0.09%
    loop_unroll                          :     0.000426s :  1.43%
    jit_opt_after_cconv.c_1              :     0.000037s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000046s :  0.15%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000030s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.03%
    remove_dup_value                     :     0.000025s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.01%
    environ_conv                         :     0.000008s :  0.03%
    add_recomputation                    :     0.000060s :  0.20%
    cse_after_recomputation.cse          :     0.000021s :  0.07%
    auto_monad_reorder                   :     0.000021s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000482s :  1.62%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008458s : 28.35%
    execute                              :     0.000009s :  0.03%


Time group info:
------[substitution.]   0.000589   126
 7.44% :     0.000044s :      2: substitution.arithmetic_simplify
 0.47% :     0.000003s :      5: substitution.elim_not_effective
 0.35% :     0.000002s :      5: substitution.fold_const_symbol
 1.33% :     0.000008s :      6: substitution.graph_param_transform
58.50% :     0.000345s :     12: substitution.inline
 3.27% :     0.000019s :      2: substitution.inline_without_move
 1.70% :     0.000010s :     18: substitution.j_node_and_user_rematch
 1.93% :     0.000011s :      7: substitution.minmaximum_grad
 1.41% :     0.000008s :      5: substitution.partial_eliminate
 2.38% :     0.000014s :     18: substitution.remove_not_recompute_node
 5.28% :     0.000031s :     10: substitution.replace_applicator
 1.72% :     0.000010s :      7: substitution.replace_old_param
 0.75% :     0.000004s :      1: substitution.set_cell_output_no_recompute
 3.81% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.45% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.22% :     0.000043s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010023     2
89.13% :     0.008933s :      1: type_inference.infer
10.87% :     0.001090s :      1: type_inference.specialize
------[replace.]   0.000150    19
64.04% :     0.000096s :     12: replace.inline
35.96% :     0.000054s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000358    19
94.14% :     0.000337s :     12: match.inline
 5.86% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000412  2895
 1.60% :     0.000007s :     48: predicate.accumulaten_eliminater
 0.49% :     0.000002s :      6: predicate.ad_related_special_op_eliminate
 1.45% :     0.000006s :     48: predicate.addn_check_dump
 1.64% :     0.000007s :     48: predicate.addn_zero_filter
 2.15% :     0.000009s :     48: predicate.arithmetic_simplify
 1.65% :     0.000007s :     48: predicate.cast_eliminate
 0.24% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.45% :     0.000006s :     48: predicate.compare_switch_simplify
 1.65% :     0.000007s :     48: predicate.depend_value_elim
 1.43% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.47% :     0.000006s :     48: predicate.dict_get_item_eliminator
 1.46% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.33% :     0.000001s :      6: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      6: predicate.elim_not_effective
 0.28% :     0.000001s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.43% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.45% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.52% :     0.000006s :     48: predicate.environ_get_eliminate
 1.54% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.11% :     0.000000s :      6: predicate.fold_const_symbol
 1.09% :     0.000005s :     23: predicate.get_grad_eliminate
 0.18% :     0.000001s :      6: predicate.graph_param_transform
 4.39% :     0.000018s :     79: predicate.inline
 1.72% :     0.000007s :     44: predicate.inline_without_move
 0.36% :     0.000001s :     23: predicate.j_node_and_user_rematch
 1.03% :     0.000004s :     23: predicate.less_batch_normalization
 1.91% :     0.000008s :     55: predicate.list_to_tuple_eliminator_
 2.07% :     0.000009s :     61: predicate.load_eliminater
 0.62% :     0.000003s :      6: predicate.loop_unroll_after_grad
 2.95% :     0.000012s :     90: predicate.loop_unroll_before_grad
 1.93% :     0.000008s :     54: predicate.make_slice_get_slice_eliminator
 1.45% :     0.000006s :     48: predicate.merge_addn
 1.48% :     0.000006s :     48: predicate.minmaximum_grad
 0.65% :     0.000003s :      6: predicate.mutable_eliminate
 0.21% :     0.000001s :      6: predicate.opt_reshape
 2.44% :     0.000010s :     61: predicate.partial_eliminate
 1.41% :     0.000006s :     48: predicate.print_const_string_wrapper
 1.81% :     0.000007s :     48: predicate.reduce_eliminate
 1.79% :     0.000007s :     55: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.49% :     0.000010s :    108: predicate.replace_applicator
 0.93% :     0.000004s :     44: predicate.replace_old_param
 0.13% :     0.000001s :      6: predicate.reset_defer_inline
 1.53% :     0.000006s :     48: predicate.reshape_eliminate
 1.49% :     0.000006s :     48: predicate.row_tensor_add_zeros_like
 0.36% :     0.000001s :      6: predicate.row_tensor_eliminate
 1.46% :     0.000006s :     48: predicate.same_eliminate
 0.52% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.46% :     0.000002s :     12: predicate.special_op_eliminate
 0.99% :     0.000004s :     23: predicate.specialize_transform
 1.89% :     0.000008s :     48: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000006s :     48: predicate.stack_unstack_eliminate
 0.22% :     0.000001s :      6: predicate.switch_call_monad_eliminater
 2.62% :     0.000011s :     67: predicate.switch_defer_inline
 2.29% :     0.000009s :     67: predicate.switch_layer_defer_inline
 5.89% :     0.000024s :    163: predicate.switch_simplify
 1.52% :     0.000006s :     48: predicate.tile_eliminate
 1.44% :     0.000006s :     48: predicate.transpose_eliminate
 1.70% :     0.000007s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.57% :     0.000006s :     48: predicate.tuple_list_get_item_depend_reorder
 3.21% :     0.000013s :     67: predicate.tuple_list_get_item_eliminator
 1.81% :     0.000007s :     48: predicate.tuple_list_set_item_eliminator
 1.73% :     0.000007s :     55: predicate.tuple_to_list_eliminator_
 2.28% :     0.000009s :     61: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000012s :     84: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000007s :     48: predicate.value_based_eliminate
 0.18% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.28% :     0.000001s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001529    29
58.83% :     0.000899s :     13: func_graph_cloner_run.FuncGraphClonerGraph
41.17% :     0.000629s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.040699    91
 0.15% :     0.000063s :      1: add_recomputation
 0.21% :     0.000087s :      1: auto_monad
 0.06% :     0.000023s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.52% :     0.000211s :      1: bootstrap
 0.07% :     0.000029s :      1: cconv
 0.03% :     0.000012s :      1: convert_after_rewriter
 0.09% :     0.000035s :      1: cse_after_recomputation
 0.03% :     0.000010s :      1: environ_conv
 0.12% :     0.000047s :      1: event_method
 0.03% :     0.000013s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000005s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
30.61% :     0.012458s :      1: jit_opt_a
 0.53% :     0.000217s :      1: jit_opt_after_cconv
 0.17% :     0.000070s :      1: jit_opt_b
 1.07% :     0.000434s :      1: loop_unroll
 1.85% :     0.000753s :      1: mutable_eliminate
 7.09% :     0.002885s :     39: opt.transform.jit_opt_a
 0.20% :     0.000080s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000041s :      4: opt.transform.jit_opt_b
 0.05% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.12% :     0.000049s :      4: opt.transform.symbol_engine_opt
 1.20% :     0.000490s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000027s :      1: py_interpret_to_execute
 0.05% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000028s :      1: remove_dup_value
 5.08% :     0.002069s :      2: renormalize.infer
 3.81% :     0.001551s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000056s :      1: rewriter_after_opt_a
 0.22% :     0.000089s :      1: rewriter_before_opt_a
 0.23% :     0.000093s :      1: symbol_engine_optimizer
20.82% :     0.008475s :      1: task_emit
24.81% :     0.010098s :      1: type_inference
 0.17% :     0.000068s :      1: validate


TotalTime = 0.0145478, [30]
[bootstrap]: 0.00037041
[type_inference]: 0.00503387
[event_method]: 1.856e-05
[auto_monad]: 7.055e-05
[graph_reusing]: 5.67001e-06
[pre_auto_parallel]: 2.66999e-06
[py_interpret_to_execute]: 3.416e-05
[rewriter_before_opt_a]: 7.607e-05
[expand_dump_flag]: 3.65998e-06
[jit_opt_a]: 0.00626333, [2]
    [Cycle 1]: 0.00202317, [27]
        [switch_simplify]: 6.943e-05
        [loop_unroll]: 2.92e-05
        [a_1]: 0.00062718
        [with_stream_mark]: 2.79e-05
        [recompute_prepare]: 9.92999e-06
        [updatestate_depend_eliminate]: 4.55999e-06
        [updatestate_assign_eliminate]: 3.07002e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [parameter_eliminate]: 2.07999e-06
        [specialize_transform]: 6.59999e-06
        [updatestate_useless_node_eliminater]: 5.31002e-06
        [accelerated_algorithm]: 5.693e-05
        [meta_shard_fg_expand]: 2.61e-06
        [get_grad_eliminate_]: 5.59e-06
        [merge_forward]: 4.79998e-06
        [cell_reuse_recompute_pass]: 1.34998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.713e-05
        [j_node_and_user_rematch]: 9.52999e-06
        [meta_fg_expand]: 2.86e-06
        [replace_old_param]: 9.47999e-06
        [inline_without_move]: 6.14001e-06
        [renormalize]: 0.00083911
        [add_forward_monad_depend]: 8.2e-06
        [auto_monad_grad]: 2.56e-06
        [auto_monad_eliminator]: 1.652e-05
        [cse]: 3.012e-05
        [replace_applicator]: 1.672e-05
    [Cycle 2]: 0.00034598, [27]
        [switch_simplify]: 5.24e-06
        [loop_unroll]: 4.98001e-06
        [a_1]: 9.278e-05
        [with_stream_mark]: 1.408e-05
        [recompute_prepare]: 5.27001e-06
        [updatestate_depend_eliminate]: 3.28e-06
        [updatestate_assign_eliminate]: 2.28998e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 1.25001e-06
        [specialize_transform]: 5.19e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.62999e-06
        [meta_shard_fg_expand]: 1.96e-06
        [get_grad_eliminate_]: 5.12999e-06
        [merge_forward]: 3.94002e-06
        [cell_reuse_recompute_pass]: 2.17999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.655e-05
        [j_node_and_user_rematch]: 9.07999e-06
        [meta_fg_expand]: 1.79e-06
        [replace_old_param]: 7.4e-06
        [inline_without_move]: 9.56e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.45001e-06
        [auto_monad_grad]: 9.5999e-07
        [auto_monad_eliminator]: 5.80002e-06
        [cse]: 1.184e-05
        [replace_applicator]: 5.22e-06
[py_interpret_to_execute_after_opt_a]: 1.803e-05
[rewriter_after_opt_a]: 4.142e-05
[convert_after_rewriter]: 7.52002e-06
[order_py_execute_after_rewriter]: 5.85002e-06
[mutable_eliminate]: 0.00077582
[jit_opt_b]: 5.644e-05, [1]
    [Cycle 1]: 4.751e-05, [2]
        [frontend_op_eliminate]: 1.806e-05
        [inline_after_opt_a]: 1.663e-05
[cconv]: 3.483e-05
[loop_unroll]: 0.00049493
[jit_opt_after_cconv]: 0.00019149, [1]
    [Cycle 1]: 0.00018378, [11]
        [c_1]: 2.218e-05
        [parameter_eliminate]: 4.87e-06
        [updatestate_depend_eliminate]: 9.37999e-06
        [updatestate_assign_eliminate]: 3.48e-06
        [updatestate_loads_eliminate]: 2.51998e-06
        [cse]: 4.267e-05
        [call_graph_tuple_transform]: 2.67e-05
        [tuple_list_get_item_eliminator]: 6.01998e-06
        [none_parameter_eliminate]: 1.69e-06
        [renormalize]: 7.7e-07
        [switch_simplify]: 6.05002e-06
[remove_dup_value]: 1.774e-05
[partial_unused_args_eliminate]: 2.74999e-06
[environ_conv]: 7.86001e-06
[add_recomputation]: 5.848e-05
[cse_after_recomputation]: 2.257e-05, [1]
    [Cycle 1]: 1.63e-05, [1]
        [cse]: 9.71e-06
[auto_monad_reorder]: 1.632e-05
[get_jit_bprop_graph]: 2.31e-06
[rewriter_after_jit_bprop_graph]: 1.032e-05
[opt_after_jit_grad]: 0.00054508
[symbol_engine_optimizer]: 8.695e-05, [1]
    [Cycle 1]: 7.981e-05, [6]
        [build]: 6.96999e-06
        [elim_shapecalc]: 9.17001e-06
        [elim_not_effective]: 1.457e-05
        [opt_reshape]: 7.51999e-06
        [fold_const_symbol]: 1.065e-05
        [renormalize]: 6.89994e-07
[validate]: 6.405e-05
Sums
    bootstrap                            :     0.000370s :  3.71%
    type_inference                       :     0.005034s : 50.41%
    event_method                         :     0.000019s :  0.19%
    auto_monad                           :     0.000071s :  0.71%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.34%
    rewriter_before_opt_a                :     0.000076s :  0.76%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000075s :  0.75%
    jit_opt_a.loop_unroll                :     0.000034s :  0.34%
    jit_opt_a.a_1                        :     0.000720s :  7.21%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.42%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000063s :  0.63%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000009s :  0.09%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.17%
    jit_opt_a.inline_without_move        :     0.000016s :  0.16%
    jit_opt_a.renormalize                :     0.000839s :  8.40%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.22%
    jit_opt_a.cse                        :     0.000042s :  0.42%
    jit_opt_a.replace_applicator         :     0.000022s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.18%
    rewriter_after_opt_a                 :     0.000041s :  0.41%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000776s :  7.77%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.17%
    cconv                                :     0.000035s :  0.35%
    loop_unroll                          :     0.000495s :  4.96%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000005s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.09%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000043s :  0.43%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.27%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000018s :  0.18%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000008s :  0.08%
    add_recomputation                    :     0.000058s :  0.59%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.10%
    opt_after_jit_grad                   :     0.000545s :  5.46%
    symbol_engine_optimizer.build        :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000064s :  0.64%


Time group info:
------[substitution.]   0.000266    23
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000002s :      2: substitution.fold_const_symbol
 2.72% :     0.000007s :      3: substitution.graph_param_transform
84.89% :     0.000226s :      4: substitution.inline
 1.45% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.92% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.89% :     0.000005s :      2: substitution.replace_old_param
 5.77% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004953     2
82.78% :     0.004101s :      1: type_inference.infer
17.22% :     0.000853s :      1: type_inference.specialize
------[replace.]   0.000082     6
65.97% :     0.000054s :      4: replace.inline
34.03% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000237     6
94.08% :     0.000223s :      4: match.inline
 5.92% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000135   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.49% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.22% :     0.000002s :     10: predicate.addn_check_dump
 1.60% :     0.000002s :     10: predicate.addn_zero_filter
 2.77% :     0.000004s :     10: predicate.arithmetic_simplify
 1.01% :     0.000001s :     10: predicate.cast_eliminate
 0.31% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.90% :     0.000001s :     10: predicate.compare_switch_simplify
 1.14% :     0.000002s :     10: predicate.depend_value_elim
 0.89% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.32% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.94% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.40% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.44% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.90% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.05% :     0.000001s :     10: predicate.environ_get_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000001s :      3: predicate.graph_param_transform
 5.50% :     0.000007s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.38% :     0.000002s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.51% :     0.000002s :     15: predicate.load_eliminater
 1.90% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.59% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.92% :     0.000001s :     10: predicate.merge_addn
 1.25% :     0.000002s :     10: predicate.minmaximum_grad
 2.27% :     0.000003s :      3: predicate.mutable_eliminate
 0.77% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     15: predicate.partial_eliminate
 1.26% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.72% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.74% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.45% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.02% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.33% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.31% :     0.000002s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.47% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.31% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.86% :     0.000011s :     53: predicate.switch_simplify
 1.33% :     0.000002s :     10: predicate.tile_eliminate
 0.99% :     0.000001s :     10: predicate.transpose_eliminate
 1.23% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.01% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.76% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.42% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.42% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.90% :     0.000003s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000664    12
51.48% :     0.000342s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.42% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.10% :     0.000273s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016440    72
 0.38% :     0.000062s :      1: add_recomputation
 0.46% :     0.000075s :      1: auto_monad
 0.12% :     0.000020s :      1: auto_monad_reorder
 2.40% :     0.000395s :      1: bootstrap
 0.23% :     0.000037s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000011s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.12% :     0.006267s :      1: jit_opt_a
 1.19% :     0.000195s :      1: jit_opt_after_cconv
 0.36% :     0.000060s :      1: jit_opt_b
 3.07% :     0.000504s :      1: loop_unroll
 4.80% :     0.000789s :      1: mutable_eliminate
 6.15% :     0.001011s :     26: opt.transform.jit_opt_a
 0.34% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000027s :      4: opt.transform.jit_opt_b
 0.10% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000038s :      4: opt.transform.symbol_engine_opt
 3.38% :     0.000555s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000037s :      1: py_interpret_to_execute
 0.12% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000021s :      1: remove_dup_value
 2.79% :     0.000459s :      1: renormalize.infer
 2.25% :     0.000370s :      1: renormalize.specialize
 0.08% :     0.000013s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000045s :      1: rewriter_after_opt_a
 0.48% :     0.000079s :      1: rewriter_before_opt_a
 0.55% :     0.000090s :      1: symbol_engine_optimizer
30.73% :     0.005052s :      1: type_inference

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.191.122 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e6c10> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.191.738 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.02056, [30]
[bootstrap]: 0.00019677
[type_inference]: 0.00956258
[event_method]: 4.185e-05
[auto_monad]: 9.564e-05
[graph_reusing]: 6.90998e-06
[pre_auto_parallel]: 3.16001e-06
[py_interpret_to_execute]: 2.648e-05
[rewriter_before_opt_a]: 9.124e-05
[expand_dump_flag]: 3.02002e-06
[jit_opt_a]: 0.0084198, [2]
    [Cycle 1]: 0.00460635, [27]
        [switch_simplify]: 6.929e-05
        [loop_unroll]: 3.798e-05
        [a_1]: 0.00084723
        [with_stream_mark]: 2.79e-05
        [recompute_prepare]: 1.761e-05
        [updatestate_depend_eliminate]: 8.17e-06
        [updatestate_assign_eliminate]: 6.54001e-06
        [updatestate_loads_eliminate]: 5.91e-06
        [parameter_eliminate]: 2.69001e-06
        [specialize_transform]: 1.272e-05
        [updatestate_useless_node_eliminater]: 1.182e-05
        [accelerated_algorithm]: 1.223e-05
        [meta_shard_fg_expand]: 3.35e-06
        [get_grad_eliminate_]: 1.158e-05
        [merge_forward]: 8.86002e-06
        [cell_reuse_recompute_pass]: 1.03001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.623e-05
        [j_node_and_user_rematch]: 2.259e-05
        [meta_fg_expand]: 0.00149798
        [replace_old_param]: 5.203e-05
        [inline_without_move]: 5.534e-05
        [renormalize]: 0.0015915
        [add_forward_monad_depend]: 6.11998e-06
        [auto_monad_grad]: 3.14001e-06
        [auto_monad_eliminator]: 1.335e-05
        [cse]: 2.904e-05
        [replace_applicator]: 1.214e-05
    [Cycle 2]: 0.00028478, [27]
        [switch_simplify]: 5.13002e-06
        [loop_unroll]: 4.03001e-06
        [a_1]: 3.957e-05
        [with_stream_mark]: 9.68002e-06
        [recompute_prepare]: 3.65e-06
        [updatestate_depend_eliminate]: 2.32999e-06
        [updatestate_assign_eliminate]: 1.86e-06
        [updatestate_loads_eliminate]: 2.26998e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 3.80998e-06
        [updatestate_useless_node_eliminater]: 3.6e-06
        [accelerated_algorithm]: 3.97e-06
        [meta_shard_fg_expand]: 1.45001e-06
        [get_grad_eliminate_]: 3.77998e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.188e-05
        [j_node_and_user_rematch]: 5.52001e-06
        [meta_fg_expand]: 2.581e-05
        [replace_old_param]: 4.20999e-06
        [inline_without_move]: 3.66001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.44998e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 4.85999e-06
        [cse]: 1.157e-05
        [replace_applicator]: 4.50001e-06
[py_interpret_to_execute_after_opt_a]: 1.223e-05
[rewriter_after_opt_a]: 3.73e-05
[convert_after_rewriter]: 6.88998e-06
[order_py_execute_after_rewriter]: 4.98001e-06
[mutable_eliminate]: 0.00055371
[jit_opt_b]: 4.134e-05, [1]
    [Cycle 1]: 3.352e-05, [2]
        [frontend_op_eliminate]: 1.046e-05
        [inline_after_opt_a]: 1.167e-05
[cconv]: 2.522e-05
[loop_unroll]: 0.00045254
[jit_opt_after_cconv]: 0.00012766, [1]
    [Cycle 1]: 0.00011955, [11]
        [c_1]: 1.231e-05
        [parameter_eliminate]: 3.43e-06
        [updatestate_depend_eliminate]: 5.99e-06
        [updatestate_assign_eliminate]: 2.01998e-06
        [updatestate_loads_eliminate]: 1.59e-06
        [cse]: 1.95e-05
        [call_graph_tuple_transform]: 1.438e-05
        [tuple_list_get_item_eliminator]: 3.57002e-06
        [none_parameter_eliminate]: 1.41998e-06
        [renormalize]: 7.89994e-07
        [switch_simplify]: 4.27e-06
[remove_dup_value]: 1.484e-05
[partial_unused_args_eliminate]: 2.64999e-06
[environ_conv]: 5.74e-06
[add_recomputation]: 3.634e-05
[cse_after_recomputation]: 2.121e-05, [1]
    [Cycle 1]: 1.553e-05, [1]
        [cse]: 8.73001e-06
[auto_monad_reorder]: 1.414e-05
[get_jit_bprop_graph]: 2.04999e-06
[rewriter_after_jit_bprop_graph]: 4.72e-06
[opt_after_jit_grad]: 0.00049204
[symbol_engine_optimizer]: 6.534e-05, [1]
    [Cycle 1]: 5.844e-05, [6]
        [build]: 3.14001e-06
        [elim_shapecalc]: 5.99999e-06
        [elim_not_effective]: 1.095e-05
        [opt_reshape]: 4.08001e-06
        [fold_const_symbol]: 5.74e-06
        [renormalize]: 6.50005e-07
[validate]: 3.06e-05
Sums
    bootstrap                            :     0.000197s :  1.20%
    type_inference                       :     0.009563s : 58.26%
    event_method                         :     0.000042s :  0.25%
    auto_monad                           :     0.000096s :  0.58%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.16%
    rewriter_before_opt_a                :     0.000091s :  0.56%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000074s :  0.45%
    jit_opt_a.loop_unroll                :     0.000042s :  0.26%
    jit_opt_a.a_1                        :     0.000887s :  5.40%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.09%
    jit_opt_a.merge_forward              :     0.000011s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001524s :  9.28%
    jit_opt_a.replace_old_param          :     0.000056s :  0.34%
    jit_opt_a.inline_without_move        :     0.000059s :  0.36%
    jit_opt_a.renormalize                :     0.001592s :  9.70%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.11%
    jit_opt_a.cse                        :     0.000041s :  0.25%
    jit_opt_a.replace_applicator         :     0.000017s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.07%
    rewriter_after_opt_a                 :     0.000037s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000554s :  3.37%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.07%
    cconv                                :     0.000025s :  0.15%
    loop_unroll                          :     0.000453s :  2.76%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.04%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000020s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.03%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000036s :  0.22%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000492s :  3.00%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.03%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000031s :  0.19%


Time group info:
------[substitution.]   0.000261    50
 0.63% :     0.000002s :      1: substitution.elim_not_effective
 0.33% :     0.000001s :      1: substitution.fold_const_symbol
 1.58% :     0.000004s :      1: substitution.graph_param_transform
67.86% :     0.000177s :      6: substitution.inline
 6.67% :     0.000017s :      2: substitution.inline_without_move
 2.46% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.38% :     0.000004s :      2: substitution.minmaximum_grad
 1.58% :     0.000004s :      3: substitution.partial_eliminate
 2.64% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.05% :     0.000005s :      2: substitution.replace_applicator
 1.59% :     0.000004s :      5: substitution.replace_old_param
 0.87% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.66% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.81% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.87% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009492     2
87.79% :     0.008333s :      1: type_inference.infer
12.21% :     0.001159s :      1: type_inference.specialize
------[replace.]   0.000064     8
77.56% :     0.000050s :      6: replace.inline
22.44% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000184     8
94.25% :     0.000174s :      6: match.inline
 5.75% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000174  1109
 1.45% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.77% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.30% :     0.000002s :     18: predicate.addn_check_dump
 1.54% :     0.000003s :     18: predicate.addn_zero_filter
 1.94% :     0.000003s :     18: predicate.arithmetic_simplify
 1.40% :     0.000002s :     18: predicate.cast_eliminate
 0.17% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.26% :     0.000002s :     18: predicate.compare_switch_simplify
 1.53% :     0.000003s :     18: predicate.depend_value_elim
 1.28% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.49% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.42% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.41% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      1: predicate.elim_not_effective
 0.25% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.31% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.30% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.33% :     0.000002s :     18: predicate.environ_get_eliminate
 1.34% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.10% :     0.000002s :     12: predicate.get_grad_eliminate
 0.18% :     0.000000s :      1: predicate.graph_param_transform
 4.77% :     0.000008s :     28: predicate.inline
 3.34% :     0.000006s :     33: predicate.inline_without_move
 0.49% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.50% :     0.000003s :     12: predicate.less_batch_normalization
 1.56% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.59% :     0.000003s :     21: predicate.load_eliminater
 0.98% :     0.000002s :      1: predicate.loop_unroll_after_grad
 3.60% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.68% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.26% :     0.000002s :     18: predicate.merge_addn
 1.34% :     0.000002s :     18: predicate.minmaximum_grad
 1.04% :     0.000002s :      1: predicate.mutable_eliminate
 0.16% :     0.000000s :      1: predicate.opt_reshape
 2.13% :     0.000004s :     21: predicate.partial_eliminate
 1.38% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.82% :     0.000003s :     18: predicate.reduce_eliminate
 1.64% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.72% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.51% :     0.000003s :     22: predicate.replace_applicator
 1.61% :     0.000003s :     33: predicate.replace_old_param
 0.21% :     0.000000s :      1: predicate.reset_defer_inline
 1.49% :     0.000003s :     18: predicate.reshape_eliminate
 1.52% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.51% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.46% :     0.000003s :     18: predicate.same_eliminate
 0.63% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.44% :     0.000001s :      2: predicate.special_op_eliminate
 1.14% :     0.000002s :     12: predicate.specialize_transform
 1.49% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.24% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.62% :     0.000005s :     26: predicate.switch_defer_inline
 2.29% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.95% :     0.000012s :     71: predicate.switch_simplify
 1.62% :     0.000003s :     18: predicate.tile_eliminate
 1.37% :     0.000002s :     18: predicate.transpose_eliminate
 1.54% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.37% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 3.26% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.73% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.54% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000003s :     18: predicate.value_based_eliminate
 0.11% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.32% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001089    22
73.31% :     0.000798s :     13: func_graph_cloner_run.FuncGraphClonerGraph
26.69% :     0.000291s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023432    72
 0.17% :     0.000039s :      1: add_recomputation
 0.43% :     0.000100s :      1: auto_monad
 0.07% :     0.000017s :      1: auto_monad_reorder
 0.89% :     0.000209s :      1: bootstrap
 0.12% :     0.000028s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.20% :     0.000046s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
35.95% :     0.008423s :      1: jit_opt_a
 0.56% :     0.000130s :      1: jit_opt_after_cconv
 0.19% :     0.000044s :      1: jit_opt_b
 1.97% :     0.000462s :      1: loop_unroll
 2.40% :     0.000563s :      1: mutable_eliminate
 5.35% :     0.001254s :     26: opt.transform.jit_opt_a
 0.13% :     0.000031s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000016s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000023s :      4: opt.transform.symbol_engine_opt
 2.13% :     0.000500s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000029s :      1: py_interpret_to_execute
 0.06% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000018s :      1: remove_dup_value
 5.17% :     0.001212s :      1: renormalize.infer
 1.58% :     0.000371s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000041s :      1: rewriter_after_opt_a
 0.40% :     0.000094s :      1: rewriter_before_opt_a
 0.29% :     0.000068s :      1: symbol_engine_optimizer
40.86% :     0.009575s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.198.169 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843db280> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.198.738 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.018578, [30]
[bootstrap]: 0.0002026
[type_inference]: 0.00911903
[event_method]: 4.021e-05
[auto_monad]: 7.376e-05
[graph_reusing]: 5.45001e-06
[pre_auto_parallel]: 2.23002e-06
[py_interpret_to_execute]: 2.33e-05
[rewriter_before_opt_a]: 8.355e-05
[expand_dump_flag]: 2.64001e-06
[jit_opt_a]: 0.00719158, [2]
    [Cycle 1]: 0.00390277, [27]
        [switch_simplify]: 5.338e-05
        [loop_unroll]: 3.723e-05
        [a_1]: 0.00077068
        [with_stream_mark]: 1.983e-05
        [recompute_prepare]: 1.687e-05
        [updatestate_depend_eliminate]: 7.36999e-06
        [updatestate_assign_eliminate]: 5.54e-06
        [updatestate_loads_eliminate]: 5.24998e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 1.29e-05
        [updatestate_useless_node_eliminater]: 1.146e-05
        [accelerated_algorithm]: 1.218e-05
        [meta_shard_fg_expand]: 2.36e-06
        [get_grad_eliminate_]: 1.177e-05
        [merge_forward]: 6.53e-06
        [cell_reuse_recompute_pass]: 7.39994e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.279e-05
        [j_node_and_user_rematch]: 2.179e-05
        [meta_fg_expand]: 0.00129077
        [replace_old_param]: 4.858e-05
        [inline_without_move]: 4.903e-05
        [renormalize]: 0.00126045
        [add_forward_monad_depend]: 5.03002e-06
        [auto_monad_grad]: 1.71e-06
        [auto_monad_eliminator]: 8.06001e-06
        [cse]: 1.567e-05
        [replace_applicator]: 1.036e-05
    [Cycle 2]: 0.0002563, [27]
        [switch_simplify]: 4.14002e-06
        [loop_unroll]: 3.59002e-06
        [a_1]: 3.647e-05
        [with_stream_mark]: 8.2e-06
        [recompute_prepare]: 3.68e-06
        [updatestate_depend_eliminate]: 2.04e-06
        [updatestate_assign_eliminate]: 1.64e-06
        [updatestate_loads_eliminate]: 1.39e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 3.51999e-06
        [updatestate_useless_node_eliminater]: 3.36999e-06
        [accelerated_algorithm]: 3.53999e-06
        [meta_shard_fg_expand]: 1.06997e-06
        [get_grad_eliminate_]: 3.13e-06
        [merge_forward]: 2.18002e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.39e-06
        [j_node_and_user_rematch]: 5.46e-06
        [meta_fg_expand]: 2.061e-05
        [replace_old_param]: 3.5e-06
        [inline_without_move]: 3.20998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.53002e-06
        [auto_monad_grad]: 8.60018e-07
        [auto_monad_eliminator]: 4.44998e-06
        [cse]: 9.98998e-06
        [replace_applicator]: 3.58999e-06
[py_interpret_to_execute_after_opt_a]: 8.60999e-06
[rewriter_after_opt_a]: 2.379e-05
[convert_after_rewriter]: 5.27001e-06
[order_py_execute_after_rewriter]: 3.87002e-06
[mutable_eliminate]: 0.00046172
[jit_opt_b]: 3.74e-05, [1]
    [Cycle 1]: 3.134e-05, [2]
        [frontend_op_eliminate]: 9.46998e-06
        [inline_after_opt_a]: 1.09e-05
[cconv]: 1.585e-05
[loop_unroll]: 0.00040687
[jit_opt_after_cconv]: 0.00011142, [1]
    [Cycle 1]: 0.00010528, [11]
        [c_1]: 1.129e-05
        [parameter_eliminate]: 2.42001e-06
        [updatestate_depend_eliminate]: 4.62e-06
        [updatestate_assign_eliminate]: 1.84e-06
        [updatestate_loads_eliminate]: 1.64e-06
        [cse]: 1.471e-05
        [call_graph_tuple_transform]: 1.123e-05
        [tuple_list_get_item_eliminator]: 3.50998e-06
        [none_parameter_eliminate]: 9.00007e-07
        [renormalize]: 4.7998e-07
        [switch_simplify]: 3.71999e-06
[remove_dup_value]: 8.14002e-06
[partial_unused_args_eliminate]: 1.79e-06
[environ_conv]: 3.25998e-06
[add_recomputation]: 2.069e-05
[cse_after_recomputation]: 1.811e-05, [1]
    [Cycle 1]: 1.308e-05, [1]
        [cse]: 7.76001e-06
[auto_monad_reorder]: 7.17002e-06
[get_jit_bprop_graph]: 1.30001e-06
[rewriter_after_jit_bprop_graph]: 4.1e-06
[opt_after_jit_grad]: 0.00046081
[symbol_engine_optimizer]: 6.09e-05, [1]
    [Cycle 1]: 5.521e-05, [6]
        [build]: 2.81e-06
        [elim_shapecalc]: 5.39e-06
        [elim_not_effective]: 9.52999e-06
        [opt_reshape]: 4.03999e-06
        [fold_const_symbol]: 5.57001e-06
        [renormalize]: 5.00004e-07
[validate]: 1.954e-05
Sums
    bootstrap                            :     0.000203s :  1.35%
    type_inference                       :     0.009119s : 60.91%
    event_method                         :     0.000040s :  0.27%
    auto_monad                           :     0.000074s :  0.49%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000023s :  0.16%
    rewriter_before_opt_a                :     0.000084s :  0.56%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.38%
    jit_opt_a.loop_unroll                :     0.000041s :  0.27%
    jit_opt_a.a_1                        :     0.000807s :  5.39%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000016s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.10%
    jit_opt_a.merge_forward              :     0.000009s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001311s :  8.76%
    jit_opt_a.replace_old_param          :     0.000052s :  0.35%
    jit_opt_a.inline_without_move        :     0.000052s :  0.35%
    jit_opt_a.renormalize                :     0.001261s :  8.42%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000013s :  0.08%
    jit_opt_a.cse                        :     0.000026s :  0.17%
    jit_opt_a.replace_applicator         :     0.000014s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000024s :  0.16%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000462s :  3.08%
    jit_opt_b.frontend_op_eliminate      :     0.000009s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000016s :  0.11%
    loop_unroll                          :     0.000407s :  2.72%
    jit_opt_after_cconv.c_1              :     0.000011s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000011s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000008s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000003s :  0.02%
    add_recomputation                    :     0.000021s :  0.14%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000007s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000461s :  3.08%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000020s :  0.13%


Time group info:
------[substitution.]   0.000197    50
 0.55% :     0.000001s :      1: substitution.elim_not_effective
 0.34% :     0.000001s :      1: substitution.fold_const_symbol
 1.11% :     0.000002s :      1: substitution.graph_param_transform
65.26% :     0.000129s :      6: substitution.inline
 7.80% :     0.000015s :      2: substitution.inline_without_move
 2.81% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.31% :     0.000003s :      2: substitution.minmaximum_grad
 1.56% :     0.000003s :      3: substitution.partial_eliminate
 3.22% :     0.000006s :      9: substitution.remove_not_recompute_node
 2.22% :     0.000004s :      2: substitution.replace_applicator
 1.69% :     0.000003s :      5: substitution.replace_old_param
 1.12% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 2.76% :     0.000005s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.71% :     0.000003s :      2: substitution.tuple_list_get_item_depend_reorder
 6.52% :     0.000013s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009060     2
89.67% :     0.008124s :      1: type_inference.infer
10.33% :     0.000936s :      1: type_inference.specialize
------[replace.]   0.000055     8
76.48% :     0.000042s :      6: replace.inline
23.52% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000134     8
93.87% :     0.000125s :      6: match.inline
 6.13% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000163  1109
 1.51% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.64% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.32% :     0.000002s :     18: predicate.addn_check_dump
 1.59% :     0.000003s :     18: predicate.addn_zero_filter
 1.97% :     0.000003s :     18: predicate.arithmetic_simplify
 1.50% :     0.000002s :     18: predicate.cast_eliminate
 0.10% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.35% :     0.000002s :     18: predicate.compare_switch_simplify
 1.68% :     0.000003s :     18: predicate.depend_value_elim
 1.30% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.46% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.51% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      1: predicate.elim_not_effective
 0.18% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.31% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.51% :     0.000002s :     18: predicate.environ_get_eliminate
 1.34% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.16% :     0.000002s :     12: predicate.get_grad_eliminate
 0.12% :     0.000000s :      1: predicate.graph_param_transform
 4.31% :     0.000007s :     28: predicate.inline
 3.34% :     0.000005s :     33: predicate.inline_without_move
 0.53% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.37% :     0.000002s :     12: predicate.less_batch_normalization
 1.86% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.75% :     0.000003s :     21: predicate.load_eliminater
 0.76% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.56% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.70% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.46% :     0.000002s :     18: predicate.merge_addn
 1.38% :     0.000002s :     18: predicate.minmaximum_grad
 0.86% :     0.000001s :      1: predicate.mutable_eliminate
 0.17% :     0.000000s :      1: predicate.opt_reshape
 2.08% :     0.000003s :     21: predicate.partial_eliminate
 1.38% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.66% :     0.000003s :     18: predicate.reduce_eliminate
 1.59% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.56% :     0.000003s :     22: predicate.replace_applicator
 1.72% :     0.000003s :     33: predicate.replace_old_param
 0.19% :     0.000000s :      1: predicate.reset_defer_inline
 1.54% :     0.000003s :     18: predicate.reshape_eliminate
 1.61% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.36% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     18: predicate.same_eliminate
 0.66% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.38% :     0.000001s :      2: predicate.special_op_eliminate
 1.24% :     0.000002s :     12: predicate.specialize_transform
 1.54% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.15% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.73% :     0.000004s :     26: predicate.switch_defer_inline
 2.52% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.95% :     0.000011s :     71: predicate.switch_simplify
 1.38% :     0.000002s :     18: predicate.tile_eliminate
 1.42% :     0.000002s :     18: predicate.transpose_eliminate
 1.69% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.53% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 2.75% :     0.000004s :     22: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.65% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.65% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.09% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000003s :     18: predicate.value_based_eliminate
 0.09% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.28% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001034    22
75.49% :     0.000781s :     13: func_graph_cloner_run.FuncGraphClonerGraph
24.51% :     0.000253s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021039    72
 0.11% :     0.000023s :      1: add_recomputation
 0.37% :     0.000078s :      1: auto_monad
 0.05% :     0.000010s :      1: auto_monad_reorder
 1.05% :     0.000221s :      1: bootstrap
 0.09% :     0.000018s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.10% :     0.000020s :      1: cse_after_recomputation
 0.03% :     0.000005s :      1: environ_conv
 0.21% :     0.000044s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000003s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
34.19% :     0.007194s :      1: jit_opt_a
 0.54% :     0.000114s :      1: jit_opt_after_cconv
 0.19% :     0.000040s :      1: jit_opt_b
 1.97% :     0.000414s :      1: loop_unroll
 2.23% :     0.000470s :      1: mutable_eliminate
 5.42% :     0.001140s :     26: opt.transform.jit_opt_a
 0.13% :     0.000027s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000014s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.22% :     0.000046s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000021s :      4: opt.transform.symbol_engine_opt
 2.23% :     0.000468s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000026s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000010s :      1: remove_dup_value
 4.56% :     0.000960s :      1: renormalize.infer
 1.40% :     0.000294s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000027s :      1: rewriter_after_opt_a
 0.41% :     0.000086s :      1: rewriter_before_opt_a
 0.30% :     0.000063s :      1: symbol_engine_optimizer
43.40% :     0.009130s :      1: type_inference


TotalTime = 0.0147469, [30]
[bootstrap]: 0.00033277
[type_inference]: 0.00549939
[event_method]: 1.996e-05
[auto_monad]: 7.169e-05
[graph_reusing]: 6.59001e-06
[pre_auto_parallel]: 3.61999e-06
[py_interpret_to_execute]: 4.215e-05
[rewriter_before_opt_a]: 8.114e-05
[expand_dump_flag]: 3.88999e-06
[jit_opt_a]: 0.00612623, [2]
    [Cycle 1]: 0.00200444, [27]
        [switch_simplify]: 6.376e-05
        [loop_unroll]: 7.513e-05
        [a_1]: 0.00060727
        [with_stream_mark]: 2.657e-05
        [recompute_prepare]: 7.45998e-06
        [updatestate_depend_eliminate]: 4.32e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.61999e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 5.99999e-06
        [updatestate_useless_node_eliminater]: 5.12999e-06
        [accelerated_algorithm]: 5.82001e-06
        [meta_shard_fg_expand]: 2.29001e-06
        [get_grad_eliminate_]: 4.95999e-06
        [merge_forward]: 4.31002e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.486e-05
        [j_node_and_user_rematch]: 9.89999e-06
        [meta_fg_expand]: 2.65997e-06
        [replace_old_param]: 9.16998e-06
        [inline_without_move]: 5.10999e-06
        [renormalize]: 0.00084918
        [add_forward_monad_depend]: 8.28001e-06
        [auto_monad_grad]: 2.36e-06
        [auto_monad_eliminator]: 1.962e-05
        [cse]: 3.617e-05
        [replace_applicator]: 1.724e-05
    [Cycle 2]: 0.0003464, [27]
        [switch_simplify]: 6.06e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 9.483e-05
        [with_stream_mark]: 1.287e-05
        [recompute_prepare]: 5.47001e-06
        [updatestate_depend_eliminate]: 3.13e-06
        [updatestate_assign_eliminate]: 2.22999e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [parameter_eliminate]: 1.19e-06
        [specialize_transform]: 5.29e-06
        [updatestate_useless_node_eliminater]: 4.97e-06
        [accelerated_algorithm]: 5.76e-06
        [meta_shard_fg_expand]: 1.64e-06
        [get_grad_eliminate_]: 4.60999e-06
        [merge_forward]: 3.41999e-06
        [cell_reuse_recompute_pass]: 2.49999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.64e-05
        [j_node_and_user_rematch]: 8.17e-06
        [meta_fg_expand]: 1.99e-06
        [replace_old_param]: 8.10999e-06
        [inline_without_move]: 5.31002e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.72999e-06
        [auto_monad_grad]: 1.27999e-06
        [auto_monad_eliminator]: 5.40001e-06
        [cse]: 1.116e-05
        [replace_applicator]: 4.79002e-06
[py_interpret_to_execute_after_opt_a]: 1.468e-05
[rewriter_after_opt_a]: 3.861e-05
[convert_after_rewriter]: 7.78999e-06
[order_py_execute_after_rewriter]: 5.18002e-06
[mutable_eliminate]: 0.0007875
[jit_opt_b]: 5.876e-05, [1]
    [Cycle 1]: 4.857e-05, [2]
        [frontend_op_eliminate]: 1.922e-05
        [inline_after_opt_a]: 1.637e-05
[cconv]: 3.387e-05
[loop_unroll]: 0.0004534
[jit_opt_after_cconv]: 0.00015339, [1]
    [Cycle 1]: 0.00014648, [11]
        [c_1]: 2.189e-05
        [parameter_eliminate]: 3.27002e-06
        [updatestate_depend_eliminate]: 8.35999e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [cse]: 2.393e-05
        [call_graph_tuple_transform]: 2.056e-05
        [tuple_list_get_item_eliminator]: 5.63002e-06
        [none_parameter_eliminate]: 1.94e-06
        [renormalize]: 6.60017e-07
        [switch_simplify]: 5.62999e-06
[remove_dup_value]: 1.444e-05
[partial_unused_args_eliminate]: 2.39999e-06
[environ_conv]: 6.52001e-06
[add_recomputation]: 5.659e-05
[cse_after_recomputation]: 2.104e-05, [1]
    [Cycle 1]: 1.49e-05, [1]
        [cse]: 9.32999e-06
[auto_monad_reorder]: 1.61e-05
[get_jit_bprop_graph]: 2.05002e-06
[rewriter_after_jit_bprop_graph]: 8.32998e-06
[opt_after_jit_grad]: 0.00054083
[symbol_engine_optimizer]: 7.957e-05, [1]
    [Cycle 1]: 7.226e-05, [6]
        [build]: 4.70001e-06
        [elim_shapecalc]: 9.05001e-06
        [elim_not_effective]: 1.314e-05
        [opt_reshape]: 5.61003e-06
        [fold_const_symbol]: 1.133e-05
        [renormalize]: 7.29982e-07
[validate]: 4.458e-05
Sums
    bootstrap                            :     0.000333s :  3.23%
    type_inference                       :     0.005499s : 53.38%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000072s :  0.70%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000004s :  0.04%
    py_interpret_to_execute              :     0.000042s :  0.41%
    rewriter_before_opt_a                :     0.000081s :  0.79%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000070s :  0.68%
    jit_opt_a.loop_unroll                :     0.000080s :  0.78%
    jit_opt_a.a_1                        :     0.000702s :  6.81%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.38%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.17%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000849s :  8.24%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.24%
    jit_opt_a.cse                        :     0.000047s :  0.46%
    jit_opt_a.replace_applicator         :     0.000022s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.14%
    rewriter_after_opt_a                 :     0.000039s :  0.37%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000788s :  7.64%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.16%
    cconv                                :     0.000034s :  0.33%
    loop_unroll                          :     0.000453s :  4.40%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000014s :  0.14%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000007s :  0.06%
    add_recomputation                    :     0.000057s :  0.55%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.08%
    opt_after_jit_grad                   :     0.000541s :  5.25%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000011s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000045s :  0.43%


Time group info:
------[substitution.]   0.000265    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.84% :     0.000002s :      2: substitution.fold_const_symbol
 2.21% :     0.000006s :      3: substitution.graph_param_transform
85.84% :     0.000227s :      4: substitution.inline
 1.59% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.82% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000005s :      2: substitution.replace_old_param
 5.07% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005412     2
83.45% :     0.004517s :      1: type_inference.infer
16.55% :     0.000896s :      1: type_inference.specialize
------[replace.]   0.000075     6
69.86% :     0.000052s :      4: replace.inline
30.14% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000237     6
94.82% :     0.000225s :      4: match.inline
 5.18% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.35% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.67% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.92% :     0.000001s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 3.03% :     0.000004s :     10: predicate.arithmetic_simplify
 0.97% :     0.000001s :     10: predicate.cast_eliminate
 0.65% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.93% :     0.000001s :     10: predicate.compare_switch_simplify
 1.11% :     0.000001s :     10: predicate.depend_value_elim
 0.97% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.33% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.01% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.43% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.56% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.08% :     0.000001s :     10: predicate.environ_get_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.78% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000001s :      3: predicate.graph_param_transform
 6.54% :     0.000009s :     22: predicate.inline
 0.83% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.64% :     0.000002s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.50% :     0.000002s :     15: predicate.load_eliminater
 1.68% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.64% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.93% :     0.000001s :     10: predicate.merge_addn
 0.91% :     0.000001s :     10: predicate.minmaximum_grad
 2.15% :     0.000003s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 1.27% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.04% :     0.000003s :     10: predicate.reduce_eliminate
 1.29% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.59% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.51% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     10: predicate.reshape_eliminate
 1.05% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     10: predicate.same_eliminate
 0.45% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000002s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.55% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.24% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.17% :     0.000003s :     16: predicate.switch_defer_inline
 1.81% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.59% :     0.000010s :     53: predicate.switch_simplify
 1.30% :     0.000002s :     10: predicate.tile_eliminate
 1.27% :     0.000002s :     10: predicate.transpose_eliminate
 1.13% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.15% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.93% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.52% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.88% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000699    12
52.64% :     0.000368s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.54% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
39.82% :     0.000278s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016613    72
 0.36% :     0.000060s :      1: add_recomputation
 0.45% :     0.000075s :      1: auto_monad
 0.11% :     0.000019s :      1: auto_monad_reorder
 2.14% :     0.000356s :      1: bootstrap
 0.22% :     0.000036s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
36.90% :     0.006130s :      1: jit_opt_a
 0.94% :     0.000156s :      1: jit_opt_after_cconv
 0.37% :     0.000062s :      1: jit_opt_b
 2.78% :     0.000462s :      1: loop_unroll
 4.81% :     0.000799s :      1: mutable_eliminate
 5.86% :     0.000974s :     26: opt.transform.jit_opt_a
 0.30% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000028s :      4: opt.transform.jit_opt_b
 0.09% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.31% :     0.000550s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000045s :      1: py_interpret_to_execute
 0.10% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.82% :     0.000468s :      1: renormalize.infer
 2.24% :     0.000372s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000042s :      1: rewriter_after_opt_a
 0.51% :     0.000085s :      1: rewriter_before_opt_a
 0.50% :     0.000083s :      1: symbol_engine_optimizer
33.24% :     0.005522s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.223.411 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843db670> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.223.947 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0324287, [33]
[bootstrap]: 0.00020822
[type_inference]: 0.0107423
[event_method]: 4.43e-05
[auto_monad]: 9.337e-05
[graph_reusing]: 7.28e-06
[pre_auto_parallel]: 3.04001e-06
[py_interpret_to_execute]: 2.733e-05
[rewriter_before_opt_a]: 9.074e-05
[expand_dump_flag]: 3.16001e-06
[jit_opt_a]: 0.0112521, [3]
    [Cycle 1]: 0.00539427, [27]
        [switch_simplify]: 5.959e-05
        [loop_unroll]: 3.718e-05
        [a_1]: 0.00080668
        [with_stream_mark]: 2.512e-05
        [recompute_prepare]: 1.676e-05
        [updatestate_depend_eliminate]: 7.57998e-06
        [updatestate_assign_eliminate]: 6.43e-06
        [updatestate_loads_eliminate]: 6.02999e-06
        [parameter_eliminate]: 2.69001e-06
        [specialize_transform]: 1.253e-05
        [updatestate_useless_node_eliminater]: 1.127e-05
        [accelerated_algorithm]: 1.172e-05
        [meta_shard_fg_expand]: 3.08e-06
        [get_grad_eliminate_]: 1.189e-05
        [merge_forward]: 8.27e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.545e-05
        [j_node_and_user_rematch]: 2.251e-05
        [meta_fg_expand]: 0.0013486
        [replace_old_param]: 4.931e-05
        [inline_without_move]: 5.02e-05
        [renormalize]: 0.0024002
        [add_forward_monad_depend]: 9.62001e-06
        [auto_monad_grad]: 5.44998e-06
        [auto_monad_eliminator]: 5.003e-05
        [cse]: 0.0001492
        [replace_applicator]: 6.332e-05
    [Cycle 2]: 0.00207255, [27]
        [switch_simplify]: 3.762e-05
        [loop_unroll]: 3.581e-05
        [a_1]: 0.0010876
        [with_stream_mark]: 1.218e-05
        [recompute_prepare]: 1.003e-05
        [updatestate_depend_eliminate]: 4.72e-06
        [updatestate_assign_eliminate]: 3.93001e-06
        [updatestate_loads_eliminate]: 3.67002e-06
        [parameter_eliminate]: 1.10001e-06
        [specialize_transform]: 8.48999e-06
        [updatestate_useless_node_eliminater]: 7.92003e-06
        [accelerated_algorithm]: 7.91001e-06
        [meta_shard_fg_expand]: 1.71e-06
        [get_grad_eliminate_]: 7.76001e-06
        [merge_forward]: 3.83999e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.648e-05
        [j_node_and_user_rematch]: 1.337e-05
        [meta_fg_expand]: 4.556e-05
        [replace_old_param]: 1.231e-05
        [inline_without_move]: 8.22998e-06
        [renormalize]: 0.00051872
        [add_forward_monad_depend]: 4.12e-06
        [auto_monad_grad]: 1.30001e-06
        [auto_monad_eliminator]: 1.398e-05
        [cse]: 4.373e-05
        [replace_applicator]: 1.411e-05
    [Cycle 3]: 0.00047219, [27]
        [switch_simplify]: 8.38999e-06
        [loop_unroll]: 7.82e-06
        [a_1]: 0.00018439
        [with_stream_mark]: 1.03e-05
        [recompute_prepare]: 8.09002e-06
        [updatestate_depend_eliminate]: 4.41002e-06
        [updatestate_assign_eliminate]: 4.13001e-06
        [updatestate_loads_eliminate]: 3.56001e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 8.17e-06
        [updatestate_useless_node_eliminater]: 7.80998e-06
        [accelerated_algorithm]: 8.87e-06
        [meta_shard_fg_expand]: 1.54e-06
        [get_grad_eliminate_]: 7.56001e-06
        [merge_forward]: 4.01001e-06
        [cell_reuse_recompute_pass]: 1.16002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.579e-05
        [j_node_and_user_rematch]: 1.322e-05
        [meta_fg_expand]: 2.63003e-06
        [replace_old_param]: 1.013e-05
        [inline_without_move]: 7.58001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.09e-06
        [auto_monad_grad]: 7.89994e-07
        [auto_monad_eliminator]: 9.66998e-06
        [cse]: 2.244e-05
        [replace_applicator]: 7.63999e-06
[py_interpret_to_execute_after_opt_a]: 1.141e-05
[rewriter_after_opt_a]: 4.519e-05
[convert_after_rewriter]: 8.43999e-06
[order_py_execute_after_rewriter]: 6.39999e-06
[mutable_eliminate]: 0.00051102
[jit_opt_b]: 6.587e-05, [1]
    [Cycle 1]: 5.927e-05, [2]
        [frontend_op_eliminate]: 2.492e-05
        [inline_after_opt_a]: 2.291e-05
[cconv]: 2.197e-05
[loop_unroll]: 0.00040775
[jit_opt_after_cconv]: 0.00018529, [1]
    [Cycle 1]: 0.00017928, [11]
        [c_1]: 3.424e-05
        [parameter_eliminate]: 2.84001e-06
        [updatestate_depend_eliminate]: 7.53999e-06
        [updatestate_assign_eliminate]: 4.37998e-06
        [updatestate_loads_eliminate]: 4.03999e-06
        [cse]: 3.13e-05
        [call_graph_tuple_transform]: 2.538e-05
        [tuple_list_get_item_eliminator]: 8.12003e-06
        [none_parameter_eliminate]: 1.50999e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 8.25999e-06
[remove_dup_value]: 2.182e-05
[partial_unused_args_eliminate]: 2.29001e-06
[environ_conv]: 6.33e-06
[add_recomputation]: 5.38e-05
[cse_after_recomputation]: 3.354e-05, [1]
    [Cycle 1]: 2.808e-05, [1]
        [cse]: 2.203e-05
[auto_monad_reorder]: 2.109e-05
[get_jit_bprop_graph]: 1.72999e-06
[rewriter_after_jit_bprop_graph]: 4.82e-06
[opt_after_jit_grad]: 0.00046298
[symbol_engine_optimizer]: 8.936e-05, [1]
    [Cycle 1]: 8.383e-05, [6]
        [build]: 3.78999e-06
        [elim_shapecalc]: 1.103e-05
        [elim_not_effective]: 1.886e-05
        [opt_reshape]: 8.89e-06
        [fold_const_symbol]: 1.377e-05
        [renormalize]: 4.89992e-07
[validate]: 4.137e-05
[backend_pass]: 9.20001e-07
[task_emit]: 0.00775124
[execute]: 7.58999e-06
Sums
    bootstrap                            :     0.000208s :  0.73%
    type_inference                       :     0.010742s : 37.89%
    event_method                         :     0.000044s :  0.16%
    auto_monad                           :     0.000093s :  0.33%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000027s :  0.10%
    rewriter_before_opt_a                :     0.000091s :  0.32%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000106s :  0.37%
    jit_opt_a.loop_unroll                :     0.000081s :  0.29%
    jit_opt_a.a_1                        :     0.002079s :  7.33%
    jit_opt_a.with_stream_mark           :     0.000048s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000035s :  0.12%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000014s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000013s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000029s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000027s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000029s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000027s :  0.10%
    jit_opt_a.merge_forward              :     0.000016s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000058s :  0.20%
    jit_opt_a.j_node_and_user_rematch    :     0.000049s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001397s :  4.93%
    jit_opt_a.replace_old_param          :     0.000072s :  0.25%
    jit_opt_a.inline_without_move        :     0.000066s :  0.23%
    jit_opt_a.renormalize                :     0.002919s : 10.29%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000074s :  0.26%
    jit_opt_a.cse                        :     0.000215s :  0.76%
    jit_opt_a.replace_applicator         :     0.000085s :  0.30%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.04%
    rewriter_after_opt_a                 :     0.000045s :  0.16%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000511s :  1.80%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000023s :  0.08%
    cconv                                :     0.000022s :  0.08%
    loop_unroll                          :     0.000408s :  1.44%
    jit_opt_after_cconv.c_1              :     0.000034s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000031s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000008s :  0.03%
    remove_dup_value                     :     0.000022s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000054s :  0.19%
    cse_after_recomputation.cse          :     0.000022s :  0.08%
    auto_monad_reorder                   :     0.000021s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000463s :  1.63%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000041s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007751s : 27.34%
    execute                              :     0.000008s :  0.03%


Time group info:
------[substitution.]   0.000488   126
 6.78% :     0.000033s :      2: substitution.arithmetic_simplify
 0.57% :     0.000003s :      5: substitution.elim_not_effective
 0.44% :     0.000002s :      5: substitution.fold_const_symbol
 1.49% :     0.000007s :      6: substitution.graph_param_transform
57.77% :     0.000282s :     12: substitution.inline
 3.20% :     0.000016s :      2: substitution.inline_without_move
 2.05% :     0.000010s :     18: substitution.j_node_and_user_rematch
 2.10% :     0.000010s :      7: substitution.minmaximum_grad
 1.29% :     0.000006s :      5: substitution.partial_eliminate
 2.48% :     0.000012s :     18: substitution.remove_not_recompute_node
 5.06% :     0.000025s :     10: substitution.replace_applicator
 1.46% :     0.000007s :      7: substitution.replace_old_param
 0.50% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.30% :     0.000021s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.69% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 7.83% :     0.000038s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010665     2
89.07% :     0.009499s :      1: type_inference.infer
10.93% :     0.001166s :      1: type_inference.specialize
------[replace.]   0.000134    19
63.73% :     0.000085s :     12: replace.inline
36.27% :     0.000049s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000296    19
92.86% :     0.000275s :     12: match.inline
 7.14% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000397  2895
 1.64% :     0.000006s :     48: predicate.accumulaten_eliminater
 0.56% :     0.000002s :      6: predicate.ad_related_special_op_eliminate
 1.45% :     0.000006s :     48: predicate.addn_check_dump
 1.75% :     0.000007s :     48: predicate.addn_zero_filter
 2.05% :     0.000008s :     48: predicate.arithmetic_simplify
 1.61% :     0.000006s :     48: predicate.cast_eliminate
 0.26% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.44% :     0.000006s :     48: predicate.compare_switch_simplify
 1.55% :     0.000006s :     48: predicate.depend_value_elim
 1.43% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.49% :     0.000006s :     48: predicate.dict_get_item_eliminator
 1.48% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.31% :     0.000001s :      6: predicate.dumpgradient_eliminate
 0.14% :     0.000001s :      6: predicate.elim_not_effective
 0.29% :     0.000001s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.48% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.49% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.48% :     0.000006s :     48: predicate.environ_get_eliminate
 1.51% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.11% :     0.000000s :      6: predicate.fold_const_symbol
 0.90% :     0.000004s :     23: predicate.get_grad_eliminate
 0.11% :     0.000000s :      6: predicate.graph_param_transform
 4.22% :     0.000017s :     79: predicate.inline
 1.83% :     0.000007s :     44: predicate.inline_without_move
 0.40% :     0.000002s :     23: predicate.j_node_and_user_rematch
 0.95% :     0.000004s :     23: predicate.less_batch_normalization
 1.85% :     0.000007s :     55: predicate.list_to_tuple_eliminator_
 1.98% :     0.000008s :     61: predicate.load_eliminater
 0.52% :     0.000002s :      6: predicate.loop_unroll_after_grad
 3.00% :     0.000012s :     90: predicate.loop_unroll_before_grad
 1.83% :     0.000007s :     54: predicate.make_slice_get_slice_eliminator
 1.48% :     0.000006s :     48: predicate.merge_addn
 1.50% :     0.000006s :     48: predicate.minmaximum_grad
 0.70% :     0.000003s :      6: predicate.mutable_eliminate
 0.24% :     0.000001s :      6: predicate.opt_reshape
 2.38% :     0.000009s :     61: predicate.partial_eliminate
 1.50% :     0.000006s :     48: predicate.print_const_string_wrapper
 1.86% :     0.000007s :     48: predicate.reduce_eliminate
 1.83% :     0.000007s :     55: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.46% :     0.000010s :    108: predicate.replace_applicator
 0.90% :     0.000004s :     44: predicate.replace_old_param
 0.15% :     0.000001s :      6: predicate.reset_defer_inline
 1.65% :     0.000007s :     48: predicate.reshape_eliminate
 1.51% :     0.000006s :     48: predicate.row_tensor_add_zeros_like
 0.35% :     0.000001s :      6: predicate.row_tensor_eliminate
 1.52% :     0.000006s :     48: predicate.same_eliminate
 0.47% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.57% :     0.000002s :     12: predicate.special_op_eliminate
 0.92% :     0.000004s :     23: predicate.specialize_transform
 1.74% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000006s :     48: predicate.stack_unstack_eliminate
 0.23% :     0.000001s :      6: predicate.switch_call_monad_eliminater
 2.57% :     0.000010s :     67: predicate.switch_defer_inline
 2.42% :     0.000010s :     67: predicate.switch_layer_defer_inline
 6.10% :     0.000024s :    163: predicate.switch_simplify
 1.51% :     0.000006s :     48: predicate.tile_eliminate
 1.53% :     0.000006s :     48: predicate.transpose_eliminate
 1.81% :     0.000007s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.63% :     0.000006s :     48: predicate.tuple_list_get_item_depend_reorder
 3.22% :     0.000013s :     67: predicate.tuple_list_get_item_eliminator
 1.72% :     0.000007s :     48: predicate.tuple_list_set_item_eliminator
 1.74% :     0.000007s :     55: predicate.tuple_to_list_eliminator_
 1.95% :     0.000008s :     61: predicate.updatestate_pure_node_eliminater
 3.09% :     0.000012s :     84: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000007s :     48: predicate.value_based_eliminate
 0.20% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.27% :     0.000001s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001443    29
60.64% :     0.000875s :     13: func_graph_cloner_run.FuncGraphClonerGraph
39.36% :     0.000568s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038225    91
 0.15% :     0.000057s :      1: add_recomputation
 0.26% :     0.000098s :      1: auto_monad
 0.06% :     0.000024s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.58% :     0.000222s :      1: bootstrap
 0.07% :     0.000025s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000036s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.13% :     0.000049s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
29.45% :     0.011255s :      1: jit_opt_a
 0.49% :     0.000188s :      1: jit_opt_after_cconv
 0.18% :     0.000068s :      1: jit_opt_b
 1.09% :     0.000415s :      1: loop_unroll
 1.36% :     0.000519s :      1: mutable_eliminate
 7.06% :     0.002700s :     39: opt.transform.jit_opt_a
 0.19% :     0.000073s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000049s :      4: opt.transform.symbol_engine_opt
 1.23% :     0.000470s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.08% :     0.000030s :      1: py_interpret_to_execute
 0.04% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000024s :      1: remove_dup_value
 4.27% :     0.001633s :      2: renormalize.infer
 3.33% :     0.001273s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000048s :      1: rewriter_after_opt_a
 0.25% :     0.000094s :      1: rewriter_before_opt_a
 0.24% :     0.000092s :      1: symbol_engine_optimizer
20.30% :     0.007761s :      1: task_emit
28.15% :     0.010760s :      1: type_inference
 0.17% :     0.000066s :      1: validate

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.231.602 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e6790> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.232.123 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0182499, [30]
[bootstrap]: 0.00017852
[type_inference]: 0.00858457
[event_method]: 3.983e-05
[auto_monad]: 7.144e-05
[graph_reusing]: 4.38999e-06
[pre_auto_parallel]: 1.99999e-06
[py_interpret_to_execute]: 2.219e-05
[rewriter_before_opt_a]: 8.239e-05
[expand_dump_flag]: 2.07999e-06
[jit_opt_a]: 0.00740232, [2]
    [Cycle 1]: 0.00410631, [27]
        [switch_simplify]: 5.902e-05
        [loop_unroll]: 3.737e-05
        [a_1]: 0.00079399
        [with_stream_mark]: 1.97e-05
        [recompute_prepare]: 1.68e-05
        [updatestate_depend_eliminate]: 7.15e-06
        [updatestate_assign_eliminate]: 5.64998e-06
        [updatestate_loads_eliminate]: 5.22e-06
        [parameter_eliminate]: 1.92001e-06
        [specialize_transform]: 1.293e-05
        [updatestate_useless_node_eliminater]: 1.226e-05
        [accelerated_algorithm]: 1.249e-05
        [meta_shard_fg_expand]: 2.40002e-06
        [get_grad_eliminate_]: 1.189e-05
        [merge_forward]: 6.84999e-06
        [cell_reuse_recompute_pass]: 7.90023e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.265e-05
        [j_node_and_user_rematch]: 2.191e-05
        [meta_fg_expand]: 0.00137409
        [replace_old_param]: 4.825e-05
        [inline_without_move]: 4.943e-05
        [renormalize]: 0.00130145
        [add_forward_monad_depend]: 4.71002e-06
        [auto_monad_grad]: 1.31002e-06
        [auto_monad_eliminator]: 7.97998e-06
        [cse]: 1.673e-05
        [replace_applicator]: 1.014e-05
    [Cycle 2]: 0.00025418, [27]
        [switch_simplify]: 4.2e-06
        [loop_unroll]: 3.65e-06
        [a_1]: 3.563e-05
        [with_stream_mark]: 8.15e-06
        [recompute_prepare]: 3.75e-06
        [updatestate_depend_eliminate]: 2.04999e-06
        [updatestate_assign_eliminate]: 1.67001e-06
        [updatestate_loads_eliminate]: 1.34998e-06
        [parameter_eliminate]: 9.30013e-07
        [specialize_transform]: 3.71001e-06
        [updatestate_useless_node_eliminater]: 3.35998e-06
        [accelerated_algorithm]: 3.55e-06
        [meta_shard_fg_expand]: 1.08001e-06
        [get_grad_eliminate_]: 3.03e-06
        [merge_forward]: 1.90001e-06
        [cell_reuse_recompute_pass]: 1.46998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.10999e-06
        [j_node_and_user_rematch]: 5.21002e-06
        [meta_fg_expand]: 2.015e-05
        [replace_old_param]: 3.43999e-06
        [inline_without_move]: 3.16999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.41002e-06
        [auto_monad_grad]: 6.80011e-07
        [auto_monad_eliminator]: 4.32e-06
        [cse]: 9.89001e-06
        [replace_applicator]: 6.64999e-06
[py_interpret_to_execute_after_opt_a]: 8.07998e-06
[rewriter_after_opt_a]: 2.606e-05
[convert_after_rewriter]: 5.49998e-06
[order_py_execute_after_rewriter]: 3.52002e-06
[mutable_eliminate]: 0.00047042
[jit_opt_b]: 3.926e-05, [1]
    [Cycle 1]: 3.227e-05, [2]
        [frontend_op_eliminate]: 1.022e-05
        [inline_after_opt_a]: 1.08e-05
[cconv]: 1.543e-05
[loop_unroll]: 0.00044862
[jit_opt_after_cconv]: 0.00011347, [1]
    [Cycle 1]: 0.00010743, [11]
        [c_1]: 1.163e-05
        [parameter_eliminate]: 2.79999e-06
        [updatestate_depend_eliminate]: 4.56002e-06
        [updatestate_assign_eliminate]: 1.92999e-06
        [updatestate_loads_eliminate]: 1.60999e-06
        [cse]: 1.475e-05
        [call_graph_tuple_transform]: 1.185e-05
        [tuple_list_get_item_eliminator]: 3.53e-06
        [none_parameter_eliminate]: 1.03001e-06
        [renormalize]: 7.30011e-07
        [switch_simplify]: 3.77998e-06
[remove_dup_value]: 7.53e-06
[partial_unused_args_eliminate]: 1.87001e-06
[environ_conv]: 3.29001e-06
[add_recomputation]: 2.118e-05
[cse_after_recomputation]: 1.97e-05, [1]
    [Cycle 1]: 1.421e-05, [1]
        [cse]: 8.09002e-06
[auto_monad_reorder]: 8e-06
[get_jit_bprop_graph]: 1.43002e-06
[rewriter_after_jit_bprop_graph]: 4.53001e-06
[opt_after_jit_grad]: 0.00043358
[symbol_engine_optimizer]: 5.95e-05, [1]
    [Cycle 1]: 5.387e-05, [6]
        [build]: 2.27999e-06
        [elim_shapecalc]: 5.94999e-06
        [elim_not_effective]: 8.94e-06
        [opt_reshape]: 3.97e-06
        [fold_const_symbol]: 5.32999e-06
        [renormalize]: 3.89991e-07
[validate]: 2.043e-05
Sums
    bootstrap                            :     0.000179s :  1.22%
    type_inference                       :     0.008585s : 58.84%
    event_method                         :     0.000040s :  0.27%
    auto_monad                           :     0.000071s :  0.49%
    graph_reusing                        :     0.000004s :  0.03%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000022s :  0.15%
    rewriter_before_opt_a                :     0.000082s :  0.56%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000063s :  0.43%
    jit_opt_a.loop_unroll                :     0.000041s :  0.28%
    jit_opt_a.a_1                        :     0.000830s :  5.69%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.10%
    jit_opt_a.merge_forward              :     0.000009s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.001394s :  9.56%
    jit_opt_a.replace_old_param          :     0.000052s :  0.35%
    jit_opt_a.inline_without_move        :     0.000053s :  0.36%
    jit_opt_a.renormalize                :     0.001302s :  8.92%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000012s :  0.08%
    jit_opt_a.cse                        :     0.000027s :  0.18%
    jit_opt_a.replace_applicator         :     0.000017s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.06%
    rewriter_after_opt_a                 :     0.000026s :  0.18%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.02%
    mutable_eliminate                    :     0.000470s :  3.22%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000015s :  0.11%
    loop_unroll                          :     0.000449s :  3.07%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000012s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.03%
    remove_dup_value                     :     0.000008s :  0.05%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000003s :  0.02%
    add_recomputation                    :     0.000021s :  0.15%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000008s :  0.05%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000434s :  2.97%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000005s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000020s :  0.14%


Time group info:
------[substitution.]   0.000217    50
 0.48% :     0.000001s :      1: substitution.elim_not_effective
 0.29% :     0.000001s :      1: substitution.fold_const_symbol
 0.98% :     0.000002s :      1: substitution.graph_param_transform
64.90% :     0.000141s :      6: substitution.inline
 6.99% :     0.000015s :      2: substitution.inline_without_move
 2.65% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.55% :     0.000003s :      2: substitution.minmaximum_grad
 1.76% :     0.000004s :      3: substitution.partial_eliminate
 2.88% :     0.000006s :      9: substitution.remove_not_recompute_node
 2.33% :     0.000005s :      2: substitution.replace_applicator
 1.74% :     0.000004s :      5: substitution.replace_old_param
 1.10% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 3.33% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 2.08% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 6.91% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008521     2
89.19% :     0.007600s :      1: type_inference.infer
10.81% :     0.000921s :      1: type_inference.specialize
------[replace.]   0.000058     8
78.55% :     0.000046s :      6: replace.inline
21.45% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000148     8
92.93% :     0.000138s :      6: match.inline
 7.07% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000164  1109
 1.57% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.69% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.34% :     0.000002s :     18: predicate.addn_check_dump
 1.76% :     0.000003s :     18: predicate.addn_zero_filter
 2.30% :     0.000004s :     18: predicate.arithmetic_simplify
 1.48% :     0.000002s :     18: predicate.cast_eliminate
 0.12% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.31% :     0.000002s :     18: predicate.compare_switch_simplify
 1.56% :     0.000003s :     18: predicate.depend_value_elim
 1.34% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.38% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      1: predicate.elim_not_effective
 0.20% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.42% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.35% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.34% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.38% :     0.000002s :     18: predicate.environ_get_eliminate
 1.32% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.24% :     0.000002s :     12: predicate.get_grad_eliminate
 0.08% :     0.000000s :      1: predicate.graph_param_transform
 4.33% :     0.000007s :     28: predicate.inline
 3.25% :     0.000005s :     33: predicate.inline_without_move
 0.54% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.40% :     0.000002s :     12: predicate.less_batch_normalization
 1.64% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.68% :     0.000003s :     21: predicate.load_eliminater
 0.75% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.71% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.74% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.34% :     0.000002s :     18: predicate.merge_addn
 1.35% :     0.000002s :     18: predicate.minmaximum_grad
 0.82% :     0.000001s :      1: predicate.mutable_eliminate
 0.20% :     0.000000s :      1: predicate.opt_reshape
 1.99% :     0.000003s :     21: predicate.partial_eliminate
 1.34% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.82% :     0.000003s :     18: predicate.reduce_eliminate
 1.57% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.56% :     0.000003s :     22: predicate.replace_applicator
 1.67% :     0.000003s :     33: predicate.replace_old_param
 0.20% :     0.000000s :      1: predicate.reset_defer_inline
 1.57% :     0.000003s :     18: predicate.reshape_eliminate
 1.58% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.62% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     18: predicate.same_eliminate
 0.65% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.51% :     0.000001s :      2: predicate.special_op_eliminate
 1.22% :     0.000002s :     12: predicate.specialize_transform
 1.53% :     0.000002s :     18: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.13% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.56% :     0.000004s :     26: predicate.switch_defer_inline
 2.37% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.93% :     0.000011s :     71: predicate.switch_simplify
 1.43% :     0.000002s :     18: predicate.tile_eliminate
 1.46% :     0.000002s :     18: predicate.transpose_eliminate
 1.67% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.56% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.72% :     0.000004s :     22: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.65% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.01% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.84% :     0.000003s :     18: predicate.value_based_eliminate
 0.12% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.21% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000993    22
74.01% :     0.000735s :     13: func_graph_cloner_run.FuncGraphClonerGraph
25.99% :     0.000258s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020755    72
 0.12% :     0.000024s :      1: add_recomputation
 0.36% :     0.000075s :      1: auto_monad
 0.05% :     0.000011s :      1: auto_monad_reorder
 0.93% :     0.000193s :      1: bootstrap
 0.09% :     0.000018s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000005s :      1: environ_conv
 0.21% :     0.000044s :      1: event_method
 0.02% :     0.000004s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000007s :      1: graph_reusing
35.68% :     0.007405s :      1: jit_opt_a
 0.56% :     0.000116s :      1: jit_opt_after_cconv
 0.20% :     0.000042s :      1: jit_opt_b
 2.20% :     0.000457s :      1: loop_unroll
 2.31% :     0.000479s :      1: mutable_eliminate
 5.65% :     0.001172s :     26: opt.transform.jit_opt_a
 0.13% :     0.000028s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000015s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000014s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000021s :      4: opt.transform.symbol_engine_opt
 2.13% :     0.000441s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000025s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000010s :      1: remove_dup_value
 4.81% :     0.000999s :      1: renormalize.infer
 1.43% :     0.000296s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000029s :      1: rewriter_after_opt_a
 0.41% :     0.000085s :      1: rewriter_before_opt_a
 0.30% :     0.000062s :      1: symbol_engine_optimizer
41.41% :     0.008594s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.248.223 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843dbaf0> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.248.739 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0155677, [30]
[bootstrap]: 0.00038618
[type_inference]: 0.00554835
[event_method]: 2.048e-05
[auto_monad]: 0.00011384
[graph_reusing]: 5.74999e-06
[pre_auto_parallel]: 3.21001e-06
[py_interpret_to_execute]: 4.146e-05
[rewriter_before_opt_a]: 8.126e-05
[expand_dump_flag]: 3.94002e-06
[jit_opt_a]: 0.00660544, [2]
    [Cycle 1]: 0.00198753, [27]
        [switch_simplify]: 6.851e-05
        [loop_unroll]: 3.075e-05
        [a_1]: 0.00061505
        [with_stream_mark]: 2.877e-05
        [recompute_prepare]: 8.07e-06
        [updatestate_depend_eliminate]: 3.89997e-06
        [updatestate_assign_eliminate]: 3.35e-06
        [updatestate_loads_eliminate]: 2.74001e-06
        [parameter_eliminate]: 2.43998e-06
        [specialize_transform]: 6.76e-06
        [updatestate_useless_node_eliminater]: 5.75001e-06
        [accelerated_algorithm]: 6.43998e-06
        [meta_shard_fg_expand]: 2.36e-06
        [get_grad_eliminate_]: 5.39e-06
        [merge_forward]: 4.1e-06
        [cell_reuse_recompute_pass]: 1.08001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.762e-05
        [j_node_and_user_rematch]: 9.42999e-06
        [meta_fg_expand]: 2.53e-06
        [replace_old_param]: 1.046e-05
        [inline_without_move]: 5.59e-06
        [renormalize]: 0.00086247
        [add_forward_monad_depend]: 8e-06
        [auto_monad_grad]: 2.46e-06
        [auto_monad_eliminator]: 1.945e-05
        [cse]: 3.331e-05
        [replace_applicator]: 1.41e-05
    [Cycle 2]: 0.00035312, [27]
        [switch_simplify]: 5.44e-06
        [loop_unroll]: 4.97999e-06
        [a_1]: 9.45e-05
        [with_stream_mark]: 1.356e-05
        [recompute_prepare]: 5.85002e-06
        [updatestate_depend_eliminate]: 3.64002e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.24998e-06
        [specialize_transform]: 5.29e-06
        [updatestate_useless_node_eliminater]: 5.19998e-06
        [accelerated_algorithm]: 5.94e-06
        [meta_shard_fg_expand]: 1.69e-06
        [get_grad_eliminate_]: 4.78001e-06
        [merge_forward]: 3.23e-06
        [cell_reuse_recompute_pass]: 2.34001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.63e-05
        [j_node_and_user_rematch]: 1.035e-05
        [meta_fg_expand]: 2.12999e-06
        [replace_old_param]: 8.70001e-06
        [inline_without_move]: 5.07e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 2.65997e-06
        [auto_monad_grad]: 1.02998e-06
        [auto_monad_eliminator]: 6.00002e-06
        [cse]: 1.163e-05
        [replace_applicator]: 5.15001e-06
[py_interpret_to_execute_after_opt_a]: 1.659e-05
[rewriter_after_opt_a]: 4.098e-05
[convert_after_rewriter]: 6.49999e-06
[order_py_execute_after_rewriter]: 5.22999e-06
[mutable_eliminate]: 0.0008564
[jit_opt_b]: 5.617e-05, [1]
    [Cycle 1]: 4.771e-05, [2]
        [frontend_op_eliminate]: 1.803e-05
        [inline_after_opt_a]: 1.721e-05
[cconv]: 3.459e-05
[loop_unroll]: 0.00051482
[jit_opt_after_cconv]: 0.00017743, [1]
    [Cycle 1]: 0.00016973, [11]
        [c_1]: 2.22e-05
        [parameter_eliminate]: 6.16e-06
        [updatestate_depend_eliminate]: 9.62001e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 3.354e-05
        [call_graph_tuple_transform]: 2.356e-05
        [tuple_list_get_item_eliminator]: 5.57001e-06
        [none_parameter_eliminate]: 1.89999e-06
        [renormalize]: 6.50005e-07
        [switch_simplify]: 7.71001e-06
[remove_dup_value]: 1.568e-05
[partial_unused_args_eliminate]: 2.78998e-06
[environ_conv]: 6.66e-06
[add_recomputation]: 6.244e-05
[cse_after_recomputation]: 2.375e-05, [1]
    [Cycle 1]: 1.722e-05, [1]
        [cse]: 1.018e-05
[auto_monad_reorder]: 1.81e-05
[get_jit_bprop_graph]: 2.56e-06
[rewriter_after_jit_bprop_graph]: 1.215e-05
[opt_after_jit_grad]: 0.00054891
[symbol_engine_optimizer]: 8.25e-05, [1]
    [Cycle 1]: 7.454e-05, [6]
        [build]: 5.37999e-06
        [elim_shapecalc]: 8.64e-06
        [elim_not_effective]: 1.448e-05
        [opt_reshape]: 6.02001e-06
        [fold_const_symbol]: 1.021e-05
        [renormalize]: 5.59987e-07
[validate]: 4.808e-05
Sums
    bootstrap                            :     0.000386s :  3.64%
    type_inference                       :     0.005548s : 52.27%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000114s :  1.07%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000041s :  0.39%
    rewriter_before_opt_a                :     0.000081s :  0.77%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000074s :  0.70%
    jit_opt_a.loop_unroll                :     0.000036s :  0.34%
    jit_opt_a.a_1                        :     0.000710s :  6.68%
    jit_opt_a.with_stream_mark           :     0.000042s :  0.40%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000034s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.18%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000863s :  8.13%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000025s :  0.24%
    jit_opt_a.cse                        :     0.000045s :  0.42%
    jit_opt_a.replace_applicator         :     0.000019s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.16%
    rewriter_after_opt_a                 :     0.000041s :  0.39%
    convert_after_rewriter               :     0.000006s :  0.06%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000856s :  8.07%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.16%
    cconv                                :     0.000035s :  0.33%
    loop_unroll                          :     0.000515s :  4.85%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.09%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000034s :  0.32%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000008s :  0.07%
    remove_dup_value                     :     0.000016s :  0.15%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000007s :  0.06%
    add_recomputation                    :     0.000062s :  0.59%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000018s :  0.17%
    get_jit_bprop_graph                  :     0.000003s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000012s :  0.11%
    opt_after_jit_grad                   :     0.000549s :  5.17%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000048s :  0.45%


Time group info:
------[substitution.]   0.000275    23
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.81% :     0.000002s :      2: substitution.fold_const_symbol
 2.19% :     0.000006s :      3: substitution.graph_param_transform
85.53% :     0.000235s :      4: substitution.inline
 1.65% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.12% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.07% :     0.000006s :      2: substitution.replace_old_param
 4.91% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005458     2
81.78% :     0.004463s :      1: type_inference.infer
18.22% :     0.000995s :      1: type_inference.specialize
------[replace.]   0.000073     6
69.17% :     0.000051s :      4: replace.inline
30.83% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000245     6
95.01% :     0.000233s :      4: match.inline
 4.99% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   691
 1.51% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.79% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.90% :     0.000001s :     10: predicate.addn_check_dump
 1.52% :     0.000002s :     10: predicate.addn_zero_filter
 1.99% :     0.000003s :     10: predicate.arithmetic_simplify
 1.12% :     0.000001s :     10: predicate.cast_eliminate
 0.37% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.73% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.54% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.35% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.67% :     0.000001s :      3: predicate.elim_not_effective
 0.71% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.58% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.98% :     0.000001s :     10: predicate.environ_get_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.80% :     0.000001s :      6: predicate.get_grad_eliminate
 0.38% :     0.000001s :      3: predicate.graph_param_transform
 5.31% :     0.000007s :     22: predicate.inline
 0.86% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.41% :     0.000002s :      6: predicate.less_batch_normalization
 1.26% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.42% :     0.000002s :     15: predicate.load_eliminater
 2.37% :     0.000003s :      3: predicate.loop_unroll_after_grad
 4.10% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.63% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.91% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 2.38% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.26% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.70% :     0.000002s :     10: predicate.reduce_eliminate
 1.24% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.76% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.43% :     0.000001s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.06% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.64% :     0.000002s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.50% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.80% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.27% :     0.000003s :     16: predicate.switch_defer_inline
 1.88% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.99% :     0.000011s :     53: predicate.switch_simplify
 1.02% :     0.000001s :     10: predicate.tile_eliminate
 1.02% :     0.000001s :     10: predicate.transpose_eliminate
 1.15% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.08% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.50% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.59% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.61% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000731    12
50.94% :     0.000372s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.69% :     0.000056s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.37% :     0.000302s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017431    72
 0.38% :     0.000067s :      1: add_recomputation
 0.68% :     0.000119s :      1: auto_monad
 0.12% :     0.000021s :      1: auto_monad_reorder
 2.36% :     0.000411s :      1: bootstrap
 0.22% :     0.000038s :      1: cconv
 0.05% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000011s :      1: environ_conv
 0.14% :     0.000025s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000006s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
37.92% :     0.006609s :      1: jit_opt_a
 1.04% :     0.000181s :      1: jit_opt_after_cconv
 0.34% :     0.000059s :      1: jit_opt_b
 3.01% :     0.000525s :      1: loop_unroll
 4.98% :     0.000868s :      1: mutable_eliminate
 5.44% :     0.000948s :     26: opt.transform.jit_opt_a
 0.32% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.15% :     0.000027s :      4: opt.transform.jit_opt_b
 0.09% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.22% :     0.000561s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000006s :      1: pre_auto_parallel
 0.25% :     0.000044s :      1: py_interpret_to_execute
 0.11% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000018s :      1: remove_dup_value
 2.77% :     0.000483s :      1: renormalize.infer
 2.12% :     0.000370s :      1: renormalize.specialize
 0.08% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000044s :      1: rewriter_after_opt_a
 0.49% :     0.000085s :      1: rewriter_before_opt_a
 0.49% :     0.000086s :      1: symbol_engine_optimizer
31.97% :     0.005573s :      1: type_inference


TotalTime = 0.0307517, [33]
[bootstrap]: 0.00018552
[type_inference]: 0.00906594
[event_method]: 3.933e-05
[auto_monad]: 8.164e-05
[graph_reusing]: 6.16e-06
[pre_auto_parallel]: 2.36e-06
[py_interpret_to_execute]: 2.178e-05
[rewriter_before_opt_a]: 8.315e-05
[expand_dump_flag]: 2.54001e-06
[jit_opt_a]: 0.0108453, [3]
    [Cycle 1]: 0.00534545, [27]
        [switch_simplify]: 5.486e-05
        [loop_unroll]: 3.729e-05
        [a_1]: 0.00084611
        [with_stream_mark]: 2.081e-05
        [recompute_prepare]: 1.769e-05
        [updatestate_depend_eliminate]: 7.51001e-06
        [updatestate_assign_eliminate]: 6.04001e-06
        [updatestate_loads_eliminate]: 5.53002e-06
        [parameter_eliminate]: 1.91998e-06
        [specialize_transform]: 1.299e-05
        [updatestate_useless_node_eliminater]: 1.144e-05
        [accelerated_algorithm]: 1.168e-05
        [meta_shard_fg_expand]: 2.55002e-06
        [get_grad_eliminate_]: 1.195e-05
        [merge_forward]: 7.28e-06
        [cell_reuse_recompute_pass]: 7.39994e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.34e-05
        [j_node_and_user_rematch]: 2.116e-05
        [meta_fg_expand]: 0.00131517
        [replace_old_param]: 4.855e-05
        [inline_without_move]: 4.902e-05
        [renormalize]: 0.00236644
        [add_forward_monad_depend]: 9.31e-06
        [auto_monad_grad]: 5.32999e-06
        [auto_monad_eliminator]: 5.084e-05
        [cse]: 0.00014261
        [replace_applicator]: 6.499e-05
    [Cycle 2]: 0.00211422, [27]
        [switch_simplify]: 3.732e-05
        [loop_unroll]: 3.572e-05
        [a_1]: 0.00111147
        [with_stream_mark]: 1.434e-05
        [recompute_prepare]: 9.65002e-06
        [updatestate_depend_eliminate]: 4.90001e-06
        [updatestate_assign_eliminate]: 4.48001e-06
        [updatestate_loads_eliminate]: 4.15e-06
        [parameter_eliminate]: 1.00999e-06
        [specialize_transform]: 8.32e-06
        [updatestate_useless_node_eliminater]: 8.04002e-06
        [accelerated_algorithm]: 7.9e-06
        [meta_shard_fg_expand]: 2.59999e-06
        [get_grad_eliminate_]: 8.62e-06
        [merge_forward]: 4.22e-06
        [cell_reuse_recompute_pass]: 8.89995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.768e-05
        [j_node_and_user_rematch]: 1.324e-05
        [meta_fg_expand]: 3.493e-05
        [replace_old_param]: 1.121e-05
        [inline_without_move]: 8.47e-06
        [renormalize]: 0.00053217
        [add_forward_monad_depend]: 5.08002e-06
        [auto_monad_grad]: 1.64e-06
        [auto_monad_eliminator]: 1.44e-05
        [cse]: 4.58e-05
        [replace_applicator]: 1.634e-05
    [Cycle 3]: 0.0004814, [27]
        [switch_simplify]: 9.81e-06
        [loop_unroll]: 8.35001e-06
        [a_1]: 0.00018611
        [with_stream_mark]: 1.025e-05
        [recompute_prepare]: 8.18001e-06
        [updatestate_depend_eliminate]: 4.60999e-06
        [updatestate_assign_eliminate]: 4.31002e-06
        [updatestate_loads_eliminate]: 3.71999e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 8.22998e-06
        [updatestate_useless_node_eliminater]: 7.67002e-06
        [accelerated_algorithm]: 8.47e-06
        [meta_shard_fg_expand]: 1.48002e-06
        [get_grad_eliminate_]: 7.93001e-06
        [merge_forward]: 4.11001e-06
        [cell_reuse_recompute_pass]: 1.41998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.658e-05
        [j_node_and_user_rematch]: 1.305e-05
        [meta_fg_expand]: 2.86999e-06
        [replace_old_param]: 9.81e-06
        [inline_without_move]: 7.74002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.66e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 1.035e-05
        [cse]: 2.337e-05
        [replace_applicator]: 7.94002e-06
[py_interpret_to_execute_after_opt_a]: 1.345e-05
[rewriter_after_opt_a]: 4.766e-05
[convert_after_rewriter]: 8.54e-06
[order_py_execute_after_rewriter]: 6.11e-06
[mutable_eliminate]: 0.00057642
[jit_opt_b]: 6.436e-05, [1]
    [Cycle 1]: 5.652e-05, [2]
        [frontend_op_eliminate]: 2.355e-05
        [inline_after_opt_a]: 2.236e-05
[cconv]: 2.283e-05
[loop_unroll]: 0.00041174
[jit_opt_after_cconv]: 0.00019287, [1]
    [Cycle 1]: 0.00018659, [11]
        [c_1]: 3.57e-05
        [parameter_eliminate]: 2.68e-06
        [updatestate_depend_eliminate]: 7.08e-06
        [updatestate_assign_eliminate]: 4.43001e-06
        [updatestate_loads_eliminate]: 4.08999e-06
        [cse]: 3.19e-05
        [call_graph_tuple_transform]: 2.662e-05
        [tuple_list_get_item_eliminator]: 9.51e-06
        [none_parameter_eliminate]: 1.55001e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 9.06002e-06
[remove_dup_value]: 2.063e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 6.63e-06
[add_recomputation]: 5.383e-05
[cse_after_recomputation]: 3.333e-05, [1]
    [Cycle 1]: 2.768e-05, [1]
        [cse]: 2.105e-05
[auto_monad_reorder]: 1.956e-05
[get_jit_bprop_graph]: 1.94e-06
[rewriter_after_jit_bprop_graph]: 5.35999e-06
[opt_after_jit_grad]: 0.00046688
[symbol_engine_optimizer]: 8.837e-05, [1]
    [Cycle 1]: 8.195e-05, [6]
        [build]: 3.65e-06
        [elim_shapecalc]: 1.104e-05
        [elim_not_effective]: 1.784e-05
        [opt_reshape]: 8.80001e-06
        [fold_const_symbol]: 1.372e-05
        [renormalize]: 4.09986e-07
[validate]: 4.096e-05
[backend_pass]: 9.09989e-07
[task_emit]: 0.00813272
[execute]: 7.41999e-06
Sums
    bootstrap                            :     0.000186s :  0.69%
    type_inference                       :     0.009066s : 33.48%
    event_method                         :     0.000039s :  0.15%
    auto_monad                           :     0.000082s :  0.30%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000022s :  0.08%
    rewriter_before_opt_a                :     0.000083s :  0.31%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000102s :  0.38%
    jit_opt_a.loop_unroll                :     0.000081s :  0.30%
    jit_opt_a.a_1                        :     0.002144s :  7.92%
    jit_opt_a.with_stream_mark           :     0.000045s :  0.17%
    jit_opt_a.recompute_prepare          :     0.000036s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000013s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000030s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000027s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000028s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000029s :  0.11%
    jit_opt_a.merge_forward              :     0.000016s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000058s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000047s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001353s :  5.00%
    jit_opt_a.replace_old_param          :     0.000070s :  0.26%
    jit_opt_a.inline_without_move        :     0.000065s :  0.24%
    jit_opt_a.renormalize                :     0.002899s : 10.71%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000076s :  0.28%
    jit_opt_a.cse                        :     0.000212s :  0.78%
    jit_opt_a.replace_applicator         :     0.000089s :  0.33%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.05%
    rewriter_after_opt_a                 :     0.000048s :  0.18%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000576s :  2.13%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000022s :  0.08%
    cconv                                :     0.000023s :  0.08%
    loop_unroll                          :     0.000412s :  1.52%
    jit_opt_after_cconv.c_1              :     0.000036s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.cse              :     0.000032s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000027s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000010s :  0.04%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.03%
    remove_dup_value                     :     0.000021s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.02%
    add_recomputation                    :     0.000054s :  0.20%
    cse_after_recomputation.cse          :     0.000021s :  0.08%
    auto_monad_reorder                   :     0.000020s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000467s :  1.72%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000041s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008133s : 30.03%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000486   126
 7.19% :     0.000035s :      2: substitution.arithmetic_simplify
 0.54% :     0.000003s :      5: substitution.elim_not_effective
 0.40% :     0.000002s :      5: substitution.fold_const_symbol
 1.42% :     0.000007s :      6: substitution.graph_param_transform
58.76% :     0.000286s :     12: substitution.inline
 3.10% :     0.000015s :      2: substitution.inline_without_move
 1.89% :     0.000009s :     18: substitution.j_node_and_user_rematch
 2.06% :     0.000010s :      7: substitution.minmaximum_grad
 1.16% :     0.000006s :      5: substitution.partial_eliminate
 2.62% :     0.000013s :     18: substitution.remove_not_recompute_node
 4.83% :     0.000023s :     10: substitution.replace_applicator
 1.45% :     0.000007s :      7: substitution.replace_old_param
 0.48% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 3.91% :     0.000019s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.85% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.33% :     0.000036s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009009     2
89.40% :     0.008054s :      1: type_inference.infer
10.60% :     0.000955s :      1: type_inference.specialize
------[replace.]   0.000135    19
65.31% :     0.000088s :     12: replace.inline
34.69% :     0.000047s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000296    19
94.18% :     0.000279s :     12: match.inline
 5.82% :     0.000017s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000458  2895
 1.44% :     0.000007s :     48: predicate.accumulaten_eliminater
 0.45% :     0.000002s :      6: predicate.ad_related_special_op_eliminate
 1.26% :     0.000006s :     48: predicate.addn_check_dump
 1.57% :     0.000007s :     48: predicate.addn_zero_filter
 1.69% :     0.000008s :     48: predicate.arithmetic_simplify
 1.46% :     0.000007s :     48: predicate.cast_eliminate
 0.23% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.25% :     0.000006s :     48: predicate.compare_switch_simplify
 1.39% :     0.000006s :     48: predicate.depend_value_elim
 1.29% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.29% :     0.000006s :     48: predicate.dict_get_item_eliminator
 1.39% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.30% :     0.000001s :      6: predicate.dumpgradient_eliminate
 0.13% :     0.000001s :      6: predicate.elim_not_effective
 0.21% :     0.000001s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.27% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.29% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.36% :     0.000006s :     48: predicate.environ_get_eliminate
 1.25% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.10% :     0.000000s :      6: predicate.fold_const_symbol
 0.81% :     0.000004s :     23: predicate.get_grad_eliminate
 0.11% :     0.000001s :      6: predicate.graph_param_transform
 3.63% :     0.000017s :     79: predicate.inline
 1.60% :     0.000007s :     44: predicate.inline_without_move
 0.33% :     0.000002s :     23: predicate.j_node_and_user_rematch
 0.85% :     0.000004s :     23: predicate.less_batch_normalization
 1.62% :     0.000007s :     55: predicate.list_to_tuple_eliminator_
 1.71% :     0.000008s :     61: predicate.load_eliminater
 0.48% :     0.000002s :      6: predicate.loop_unroll_after_grad
 2.62% :     0.000012s :     90: predicate.loop_unroll_before_grad
 1.58% :     0.000007s :     54: predicate.make_slice_get_slice_eliminator
 1.28% :     0.000006s :     48: predicate.merge_addn
 1.32% :     0.000006s :     48: predicate.minmaximum_grad
 0.50% :     0.000002s :      6: predicate.mutable_eliminate
 0.23% :     0.000001s :      6: predicate.opt_reshape
 2.07% :     0.000009s :     61: predicate.partial_eliminate
 1.31% :     0.000006s :     48: predicate.print_const_string_wrapper
 1.55% :     0.000007s :     48: predicate.reduce_eliminate
 1.60% :     0.000007s :     55: predicate.redundant_stop_gradient_eliminater
 0.41% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.17% :     0.000010s :    108: predicate.replace_applicator
 0.80% :     0.000004s :     44: predicate.replace_old_param
 0.11% :     0.000001s :      6: predicate.reset_defer_inline
 1.47% :     0.000007s :     48: predicate.reshape_eliminate
 1.30% :     0.000006s :     48: predicate.row_tensor_add_zeros_like
 0.28% :     0.000001s :      6: predicate.row_tensor_eliminate
 1.29% :     0.000006s :     48: predicate.same_eliminate
 0.43% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.39% :     0.000002s :     12: predicate.special_op_eliminate
 0.85% :     0.000004s :     23: predicate.specialize_transform
 1.49% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.32% :     0.000006s :     48: predicate.stack_unstack_eliminate
 0.20% :     0.000001s :      6: predicate.switch_call_monad_eliminater
15.40% :     0.000070s :     67: predicate.switch_defer_inline
 2.09% :     0.000010s :     67: predicate.switch_layer_defer_inline
 5.27% :     0.000024s :    163: predicate.switch_simplify
 1.30% :     0.000006s :     48: predicate.tile_eliminate
 1.34% :     0.000006s :     48: predicate.transpose_eliminate
 1.58% :     0.000007s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.46% :     0.000007s :     48: predicate.tuple_list_get_item_depend_reorder
 2.62% :     0.000012s :     67: predicate.tuple_list_get_item_eliminator
 1.50% :     0.000007s :     48: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000007s :     55: predicate.tuple_to_list_eliminator_
 1.69% :     0.000008s :     61: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000012s :     84: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000007s :     48: predicate.value_based_eliminate
 0.17% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.25% :     0.000001s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001298    29
59.64% :     0.000774s :     13: func_graph_cloner_run.FuncGraphClonerGraph
40.36% :     0.000524s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.036595    91
 0.16% :     0.000057s :      1: add_recomputation
 0.23% :     0.000086s :      1: auto_monad
 0.06% :     0.000023s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.55% :     0.000202s :      1: bootstrap
 0.07% :     0.000025s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000036s :      1: cse_after_recomputation
 0.03% :     0.000009s :      1: environ_conv
 0.12% :     0.000043s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
29.64% :     0.010848s :      1: jit_opt_a
 0.53% :     0.000196s :      1: jit_opt_after_cconv
 0.18% :     0.000067s :      1: jit_opt_b
 1.15% :     0.000419s :      1: loop_unroll
 1.60% :     0.000585s :      1: mutable_eliminate
 7.56% :     0.002765s :     39: opt.transform.jit_opt_a
 0.21% :     0.000077s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000040s :      4: opt.transform.jit_opt_b
 0.04% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000048s :      4: opt.transform.symbol_engine_opt
 1.30% :     0.000475s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000024s :      1: py_interpret_to_execute
 0.04% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000024s :      1: remove_dup_value
 4.18% :     0.001531s :      2: renormalize.infer
 3.70% :     0.001356s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000050s :      1: rewriter_after_opt_a
 0.24% :     0.000086s :      1: rewriter_before_opt_a
 0.25% :     0.000091s :      1: symbol_engine_optimizer
22.25% :     0.008144s :      1: task_emit
24.81% :     0.009077s :      1: type_inference
 0.18% :     0.000065s :      1: validate


TotalTime = 0.0186783, [30]
[bootstrap]: 0.00018132
[type_inference]: 0.00861403
[event_method]: 3.824e-05
[auto_monad]: 8.828e-05
[graph_reusing]: 6.74999e-06
[pre_auto_parallel]: 2.36e-06
[py_interpret_to_execute]: 2.295e-05
[rewriter_before_opt_a]: 8.665e-05
[expand_dump_flag]: 3.04999e-06
[jit_opt_a]: 0.00770627, [2]
    [Cycle 1]: 0.00434939, [27]
        [switch_simplify]: 5.815e-05
        [loop_unroll]: 3.743e-05
        [a_1]: 0.00084469
        [with_stream_mark]: 2.2e-05
        [recompute_prepare]: 1.735e-05
        [updatestate_depend_eliminate]: 7.46999e-06
        [updatestate_assign_eliminate]: 6.41e-06
        [updatestate_loads_eliminate]: 5.97001e-06
        [parameter_eliminate]: 2.45997e-06
        [specialize_transform]: 1.282e-05
        [updatestate_useless_node_eliminater]: 1.174e-05
        [accelerated_algorithm]: 1.211e-05
        [meta_shard_fg_expand]: 3.01001e-06
        [get_grad_eliminate_]: 1.201e-05
        [merge_forward]: 6.88998e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.436e-05
        [j_node_and_user_rematch]: 2.231e-05
        [meta_fg_expand]: 0.00142227
        [replace_old_param]: 5.43e-05
        [inline_without_move]: 5.262e-05
        [renormalize]: 0.00138849
        [add_forward_monad_depend]: 5.62001e-06
        [auto_monad_grad]: 2.27999e-06
        [auto_monad_eliminator]: 1.138e-05
        [cse]: 2.289e-05
        [replace_applicator]: 1.136e-05
    [Cycle 2]: 0.00027188, [27]
        [switch_simplify]: 4.53001e-06
        [loop_unroll]: 3.73999e-06
        [a_1]: 3.886e-05
        [with_stream_mark]: 9.90002e-06
        [recompute_prepare]: 3.78001e-06
        [updatestate_depend_eliminate]: 2.41998e-06
        [updatestate_assign_eliminate]: 1.87999e-06
        [updatestate_loads_eliminate]: 1.77999e-06
        [parameter_eliminate]: 1.26997e-06
        [specialize_transform]: 3.56999e-06
        [updatestate_useless_node_eliminater]: 4.13999e-06
        [accelerated_algorithm]: 3.73001e-06
        [meta_shard_fg_expand]: 1.39998e-06
        [get_grad_eliminate_]: 3.16999e-06
        [merge_forward]: 2.09e-06
        [cell_reuse_recompute_pass]: 1.60001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.193e-05
        [j_node_and_user_rematch]: 5.51002e-06
        [meta_fg_expand]: 2.323e-05
        [replace_old_param]: 3.96001e-06
        [inline_without_move]: 3.38e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.22999e-06
        [auto_monad_grad]: 8.29983e-07
        [auto_monad_eliminator]: 4.42e-06
        [cse]: 1.101e-05
        [replace_applicator]: 4.06001e-06
[py_interpret_to_execute_after_opt_a]: 1.074e-05
[rewriter_after_opt_a]: 3.218e-05
[convert_after_rewriter]: 6.56999e-06
[order_py_execute_after_rewriter]: 4.28999e-06
[mutable_eliminate]: 0.000519
[jit_opt_b]: 3.936e-05, [1]
    [Cycle 1]: 3.246e-05, [2]
        [frontend_op_eliminate]: 9.94999e-06
        [inline_after_opt_a]: 1.107e-05
[cconv]: 2.083e-05
[loop_unroll]: 0.00041465
[jit_opt_after_cconv]: 0.0001161, [1]
    [Cycle 1]: 0.00010917, [11]
        [c_1]: 1.154e-05
        [parameter_eliminate]: 2.32999e-06
        [updatestate_depend_eliminate]: 4.92e-06
        [updatestate_assign_eliminate]: 1.87001e-06
        [updatestate_loads_eliminate]: 1.56998e-06
        [cse]: 1.523e-05
        [call_graph_tuple_transform]: 1.317e-05
        [tuple_list_get_item_eliminator]: 3.63999e-06
        [none_parameter_eliminate]: 2.00002e-06
        [renormalize]: 8.70001e-07
        [switch_simplify]: 3.45998e-06
[remove_dup_value]: 1.366e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 4.64002e-06
[add_recomputation]: 3.28e-05
[cse_after_recomputation]: 1.892e-05, [1]
    [Cycle 1]: 1.373e-05, [1]
        [cse]: 8.18001e-06
[auto_monad_reorder]: 1.263e-05
[get_jit_bprop_graph]: 1.74998e-06
[rewriter_after_jit_bprop_graph]: 4.47998e-06
[opt_after_jit_grad]: 0.00043553
[symbol_engine_optimizer]: 6.223e-05, [1]
    [Cycle 1]: 5.603e-05, [6]
        [build]: 2.99999e-06
        [elim_shapecalc]: 5.71998e-06
        [elim_not_effective]: 9.89999e-06
        [opt_reshape]: 3.91999e-06
        [fold_const_symbol]: 5.55001e-06
        [renormalize]: 5.09986e-07
[validate]: 2.611e-05
Sums
    bootstrap                            :     0.000181s :  1.21%
    type_inference                       :     0.008614s : 57.65%
    event_method                         :     0.000038s :  0.26%
    auto_monad                           :     0.000088s :  0.59%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000023s :  0.15%
    rewriter_before_opt_a                :     0.000087s :  0.58%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000063s :  0.42%
    jit_opt_a.loop_unroll                :     0.000041s :  0.28%
    jit_opt_a.a_1                        :     0.000884s :  5.91%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.21%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000016s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.10%
    jit_opt_a.merge_forward              :     0.000009s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000036s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.001446s :  9.67%
    jit_opt_a.replace_old_param          :     0.000058s :  0.39%
    jit_opt_a.inline_without_move        :     0.000056s :  0.37%
    jit_opt_a.renormalize                :     0.001389s :  9.29%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.11%
    jit_opt_a.cse                        :     0.000034s :  0.23%
    jit_opt_a.replace_applicator         :     0.000015s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000032s :  0.22%
    convert_after_rewriter               :     0.000007s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000519s :  3.47%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000021s :  0.14%
    loop_unroll                          :     0.000415s :  2.78%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000003s :  0.02%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000033s :  0.22%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000436s :  2.92%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000026s :  0.17%


Time group info:
------[substitution.]   0.000229    50
 0.61% :     0.000001s :      1: substitution.elim_not_effective
 0.36% :     0.000001s :      1: substitution.fold_const_symbol
 1.60% :     0.000004s :      1: substitution.graph_param_transform
64.15% :     0.000147s :      6: substitution.inline
 7.30% :     0.000017s :      2: substitution.inline_without_move
 2.84% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.50% :     0.000003s :      2: substitution.minmaximum_grad
 1.72% :     0.000004s :      3: substitution.partial_eliminate
 2.94% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.37% :     0.000005s :      2: substitution.replace_applicator
 1.79% :     0.000004s :      5: substitution.replace_old_param
 1.24% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.15% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 2.06% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 6.37% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008556     2
88.92% :     0.007608s :      1: type_inference.infer
11.08% :     0.000948s :      1: type_inference.specialize
------[replace.]   0.000058     8
77.41% :     0.000045s :      6: replace.inline
22.59% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000153     8
93.64% :     0.000144s :      6: match.inline
 6.36% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000173  1109
 1.44% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.54% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.34% :     0.000002s :     18: predicate.addn_check_dump
 1.58% :     0.000003s :     18: predicate.addn_zero_filter
 1.93% :     0.000003s :     18: predicate.arithmetic_simplify
 1.45% :     0.000003s :     18: predicate.cast_eliminate
 0.18% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.38% :     0.000002s :     18: predicate.compare_switch_simplify
 1.41% :     0.000002s :     18: predicate.depend_value_elim
 1.28% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.51% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.45% :     0.000003s :     18: predicate.dict_set_item_eliminator
 0.41% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      1: predicate.elim_not_effective
 0.20% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.36% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.31% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.33% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.39% :     0.000002s :     18: predicate.environ_get_eliminate
 1.26% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.18% :     0.000002s :     12: predicate.get_grad_eliminate
 0.19% :     0.000000s :      1: predicate.graph_param_transform
 4.61% :     0.000008s :     28: predicate.inline
 3.20% :     0.000006s :     33: predicate.inline_without_move
 0.48% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :     12: predicate.less_batch_normalization
 1.61% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.74% :     0.000003s :     21: predicate.load_eliminater
 0.68% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.48% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.67% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.33% :     0.000002s :     18: predicate.merge_addn
 1.31% :     0.000002s :     18: predicate.minmaximum_grad
 0.83% :     0.000001s :      1: predicate.mutable_eliminate
 0.18% :     0.000000s :      1: predicate.opt_reshape
 2.38% :     0.000004s :     21: predicate.partial_eliminate
 1.70% :     0.000003s :     18: predicate.print_const_string_wrapper
 1.81% :     0.000003s :     18: predicate.reduce_eliminate
 1.70% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.73% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.51% :     0.000003s :     22: predicate.replace_applicator
 1.75% :     0.000003s :     33: predicate.replace_old_param
 0.16% :     0.000000s :      1: predicate.reset_defer_inline
 1.54% :     0.000003s :     18: predicate.reshape_eliminate
 1.61% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.27% :     0.000000s :      1: predicate.row_tensor_eliminate
 1.90% :     0.000003s :     18: predicate.same_eliminate
 0.61% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.29% :     0.000000s :      2: predicate.special_op_eliminate
 1.09% :     0.000002s :     12: predicate.specialize_transform
 1.58% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.82% :     0.000005s :     26: predicate.switch_defer_inline
 2.38% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.71% :     0.000012s :     71: predicate.switch_simplify
 1.33% :     0.000002s :     18: predicate.tile_eliminate
 1.30% :     0.000002s :     18: predicate.transpose_eliminate
 1.79% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.55% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.24% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.59% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.63% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.09% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.75% :     0.000003s :     18: predicate.value_based_eliminate
 0.11% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.25% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000996    22
74.37% :     0.000741s :     13: func_graph_cloner_run.FuncGraphClonerGraph
25.63% :     0.000255s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021327    72
 0.17% :     0.000035s :      1: add_recomputation
 0.43% :     0.000092s :      1: auto_monad
 0.07% :     0.000015s :      1: auto_monad_reorder
 0.90% :     0.000192s :      1: bootstrap
 0.11% :     0.000023s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.20% :     0.000043s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
36.15% :     0.007709s :      1: jit_opt_a
 0.56% :     0.000119s :      1: jit_opt_after_cconv
 0.20% :     0.000042s :      1: jit_opt_b
 1.98% :     0.000422s :      1: loop_unroll
 2.47% :     0.000527s :      1: mutable_eliminate
 5.80% :     0.001236s :     26: opt.transform.jit_opt_a
 0.13% :     0.000029s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000015s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000014s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000022s :      4: opt.transform.symbol_engine_opt
 2.08% :     0.000444s :      1: opt_after_jit_grad
 0.03% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000025s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000016s :      1: remove_dup_value
 4.93% :     0.001052s :      1: renormalize.infer
 1.54% :     0.000329s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000035s :      1: rewriter_after_opt_a
 0.42% :     0.000089s :      1: rewriter_before_opt_a
 0.30% :     0.000065s :      1: symbol_engine_optimizer
40.44% :     0.008625s :      1: type_inference

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.271.793 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80e63a0> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.272.387 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.272.871 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843dd0d0> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.273.323 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0154921, [30]
[bootstrap]: 0.00035066
[type_inference]: 0.00520146
[event_method]: 2.043e-05
[auto_monad]: 7.228e-05
[graph_reusing]: 6.61999e-06
[pre_auto_parallel]: 2.93e-06
[py_interpret_to_execute]: 3.971e-05
[rewriter_before_opt_a]: 7.79e-05
[expand_dump_flag]: 3.38e-06
[jit_opt_a]: 0.00654681, [2]
    [Cycle 1]: 0.00206626, [27]
        [switch_simplify]: 7.125e-05
        [loop_unroll]: 2.975e-05
        [a_1]: 0.00063227
        [with_stream_mark]: 2.826e-05
        [recompute_prepare]: 9.12001e-06
        [updatestate_depend_eliminate]: 4.41002e-06
        [updatestate_assign_eliminate]: 2.83e-06
        [updatestate_loads_eliminate]: 2.46998e-06
        [parameter_eliminate]: 2.33002e-06
        [specialize_transform]: 7.00998e-06
        [updatestate_useless_node_eliminater]: 5.42999e-06
        [accelerated_algorithm]: 5.80002e-06
        [meta_shard_fg_expand]: 2.46e-06
        [get_grad_eliminate_]: 5.86e-06
        [merge_forward]: 4.52998e-06
        [cell_reuse_recompute_pass]: 1.00001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.636e-05
        [j_node_and_user_rematch]: 1.105e-05
        [meta_fg_expand]: 2.24999e-06
        [replace_old_param]: 1.028e-05
        [inline_without_move]: 6.26e-06
        [renormalize]: 0.00090787
        [add_forward_monad_depend]: 9.12999e-06
        [auto_monad_grad]: 2.70002e-06
        [auto_monad_eliminator]: 1.846e-05
        [cse]: 3.676e-05
        [replace_applicator]: 1.893e-05
    [Cycle 2]: 0.00035993, [27]
        [switch_simplify]: 6.04999e-06
        [loop_unroll]: 5.51998e-06
        [a_1]: 9.8e-05
        [with_stream_mark]: 1.511e-05
        [recompute_prepare]: 5.44e-06
        [updatestate_depend_eliminate]: 4.11001e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 1.53002e-06
        [specialize_transform]: 5.01002e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 5.71e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 4.00998e-06
        [cell_reuse_recompute_pass]: 2.88e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.078e-05
        [j_node_and_user_rematch]: 9.53002e-06
        [meta_fg_expand]: 1.73002e-06
        [replace_old_param]: 8.12e-06
        [inline_without_move]: 5.15999e-06
        [renormalize]: 1.29978e-07
        [add_forward_monad_depend]: 1.97001e-06
        [auto_monad_grad]: 1.28002e-06
        [auto_monad_eliminator]: 5.63002e-06
        [cse]: 1.215e-05
        [replace_applicator]: 5.34e-06
[py_interpret_to_execute_after_opt_a]: 1.653e-05
[rewriter_after_opt_a]: 4.208e-05
[convert_after_rewriter]: 8.56002e-06
[order_py_execute_after_rewriter]: 6.09001e-06
[mutable_eliminate]: 0.00101622
[jit_opt_b]: 5.949e-05, [1]
    [Cycle 1]: 4.937e-05, [2]
        [frontend_op_eliminate]: 1.934e-05
        [inline_after_opt_a]: 1.687e-05
[cconv]: 3.588e-05
[loop_unroll]: 0.00057828
[jit_opt_after_cconv]: 0.00018664, [1]
    [Cycle 1]: 0.00017775, [11]
        [c_1]: 2.226e-05
        [parameter_eliminate]: 5.59998e-06
        [updatestate_depend_eliminate]: 8.27e-06
        [updatestate_assign_eliminate]: 3.82998e-06
        [updatestate_loads_eliminate]: 2.93e-06
        [cse]: 3.788e-05
        [call_graph_tuple_transform]: 2.635e-05
        [tuple_list_get_item_eliminator]: 5.90002e-06
        [none_parameter_eliminate]: 2.10002e-06
        [renormalize]: 9.00007e-07
        [switch_simplify]: 5.79e-06
[remove_dup_value]: 1.781e-05
[partial_unused_args_eliminate]: 2.39999e-06
[environ_conv]: 6.68003e-06
[add_recomputation]: 5.977e-05
[cse_after_recomputation]: 2.584e-05, [1]
    [Cycle 1]: 1.856e-05, [1]
        [cse]: 1.035e-05
[auto_monad_reorder]: 1.796e-05
[get_jit_bprop_graph]: 2.11e-06
[rewriter_after_jit_bprop_graph]: 1.116e-05
[opt_after_jit_grad]: 0.0007044
[symbol_engine_optimizer]: 9.109e-05, [1]
    [Cycle 1]: 8.269e-05, [6]
        [build]: 7.25e-06
        [elim_shapecalc]: 9.10001e-06
        [elim_not_effective]: 1.646e-05
        [opt_reshape]: 7.56001e-06
        [fold_const_symbol]: 9.59e-06
        [renormalize]: 1.02e-06
[validate]: 5.475e-05
Sums
    bootstrap                            :     0.000351s :  3.29%
    type_inference                       :     0.005201s : 48.74%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000072s :  0.68%
    graph_reusing                        :     0.000007s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000040s :  0.37%
    rewriter_before_opt_a                :     0.000078s :  0.73%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000077s :  0.72%
    jit_opt_a.loop_unroll                :     0.000035s :  0.33%
    jit_opt_a.a_1                        :     0.000730s :  6.84%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.41%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000009s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.17%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000908s :  8.51%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.23%
    jit_opt_a.cse                        :     0.000049s :  0.46%
    jit_opt_a.replace_applicator         :     0.000024s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.15%
    rewriter_after_opt_a                 :     0.000042s :  0.39%
    convert_after_rewriter               :     0.000009s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.001016s :  9.52%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.16%
    cconv                                :     0.000036s :  0.34%
    loop_unroll                          :     0.000578s :  5.42%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000038s :  0.35%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.25%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000018s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000007s :  0.06%
    add_recomputation                    :     0.000060s :  0.56%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000018s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.10%
    opt_after_jit_grad                   :     0.000704s :  6.60%
    symbol_engine_optimizer.build        :     0.000007s :  0.07%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000008s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000055s :  0.51%


Time group info:
------[substitution.]   0.000287    23
 1.01% :     0.000003s :      2: substitution.elim_not_effective
 0.48% :     0.000001s :      2: substitution.fold_const_symbol
 2.17% :     0.000006s :      3: substitution.graph_param_transform
84.51% :     0.000242s :      4: substitution.inline
 1.66% :     0.000005s :      4: substitution.j_node_and_user_rematch
 3.73% :     0.000011s :      4: substitution.remove_not_recompute_node
 1.97% :     0.000006s :      2: substitution.replace_old_param
 4.47% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005117     2
81.55% :     0.004173s :      1: type_inference.infer
18.45% :     0.000944s :      1: type_inference.specialize
------[replace.]   0.000079     6
68.30% :     0.000054s :      4: replace.inline
31.70% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000251     6
95.35% :     0.000239s :      4: match.inline
 4.65% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000140   691
 1.23% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.61% :     0.000004s :      3: predicate.ad_related_special_op_eliminate
 0.87% :     0.000001s :     10: predicate.addn_check_dump
 1.02% :     0.000001s :     10: predicate.addn_zero_filter
 2.23% :     0.000003s :     10: predicate.arithmetic_simplify
 1.47% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.58% :     0.000002s :     10: predicate.depend_value_elim
 0.93% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.08% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.50% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.47% :     0.000001s :      3: predicate.elim_not_effective
 0.72% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.89% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.95% :     0.000001s :     10: predicate.environ_get_eliminate
 0.91% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.52% :     0.000001s :      3: predicate.graph_param_transform
 5.68% :     0.000008s :     22: predicate.inline
 1.31% :     0.000002s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.43% :     0.000002s :     15: predicate.load_eliminater
 1.92% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.59% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.87% :     0.000001s :     10: predicate.merge_addn
 0.92% :     0.000001s :     10: predicate.minmaximum_grad
 2.87% :     0.000004s :      3: predicate.mutable_eliminate
 0.82% :     0.000001s :      3: predicate.opt_reshape
 2.12% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.73% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.44% :     0.000001s :      3: predicate.reset_defer_inline
 1.25% :     0.000002s :     10: predicate.reshape_eliminate
 0.95% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.20% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.08% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.35% :     0.000002s :      6: predicate.special_op_eliminate
 1.18% :     0.000002s :      6: predicate.specialize_transform
 1.16% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.45% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 1.86% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.80% :     0.000011s :     53: predicate.switch_simplify
 1.29% :     0.000002s :     10: predicate.tile_eliminate
 1.06% :     0.000001s :     10: predicate.transpose_eliminate
 1.18% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.00% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.01% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.98% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.40% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.46% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000685    12
49.36% :     0.000338s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.80% :     0.000060s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.84% :     0.000287s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017443    72
 0.36% :     0.000063s :      1: add_recomputation
 0.44% :     0.000076s :      1: auto_monad
 0.12% :     0.000021s :      1: auto_monad_reorder
 2.12% :     0.000371s :      1: bootstrap
 0.23% :     0.000040s :      1: cconv
 0.07% :     0.000012s :      1: convert_after_rewriter
 0.16% :     0.000028s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
37.55% :     0.006551s :      1: jit_opt_a
 1.09% :     0.000189s :      1: jit_opt_after_cconv
 0.36% :     0.000063s :      1: jit_opt_b
 3.40% :     0.000593s :      1: loop_unroll
 5.92% :     0.001032s :      1: mutable_eliminate
 5.62% :     0.000981s :     26: opt.transform.jit_opt_a
 0.32% :     0.000056s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000029s :      4: opt.transform.jit_opt_b
 0.11% :     0.000019s :      1: opt.transform.loop_unroll_optimizer
 0.14% :     0.000024s :      1: opt.transform.mutable_eliminate
 0.18% :     0.000032s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000039s :      4: opt.transform.symbol_engine_opt
 4.13% :     0.000720s :      1: opt_after_jit_grad
 0.05% :     0.000009s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000043s :      1: py_interpret_to_execute
 0.11% :     0.000019s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000021s :      1: remove_dup_value
 2.84% :     0.000495s :      1: renormalize.infer
 2.31% :     0.000402s :      1: renormalize.specialize
 0.08% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000046s :      1: rewriter_after_opt_a
 0.46% :     0.000081s :      1: rewriter_before_opt_a
 0.54% :     0.000094s :      1: symbol_engine_optimizer
29.93% :     0.005221s :      1: type_inference


TotalTime = 0.0187409, [30]
[bootstrap]: 0.00016928
[type_inference]: 0.00818248
[event_method]: 3.794e-05
[auto_monad]: 5.954e-05
[graph_reusing]: 6.09001e-06
[pre_auto_parallel]: 2.22999e-06
[py_interpret_to_execute]: 2.319e-05
[rewriter_before_opt_a]: 8.543e-05
[expand_dump_flag]: 3.16999e-06
[jit_opt_a]: 0.0081475, [2]
    [Cycle 1]: 0.00454957, [27]
        [switch_simplify]: 6.082e-05
        [loop_unroll]: 3.687e-05
        [a_1]: 0.00081045
        [with_stream_mark]: 2.8e-05
        [recompute_prepare]: 1.854e-05
        [updatestate_depend_eliminate]: 7.61001e-06
        [updatestate_assign_eliminate]: 6.06998e-06
        [updatestate_loads_eliminate]: 5.69e-06
        [parameter_eliminate]: 2.66999e-06
        [specialize_transform]: 1.225e-05
        [updatestate_useless_node_eliminater]: 1.193e-05
        [accelerated_algorithm]: 1.29e-05
        [meta_shard_fg_expand]: 3.31001e-06
        [get_grad_eliminate_]: 1.147e-05
        [merge_forward]: 8.54002e-06
        [cell_reuse_recompute_pass]: 9.00007e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.616e-05
        [j_node_and_user_rematch]: 2.34e-05
        [meta_fg_expand]: 0.00152839
        [replace_old_param]: 5.507e-05
        [inline_without_move]: 5.276e-05
        [renormalize]: 0.0015655
        [add_forward_monad_depend]: 6.56e-06
        [auto_monad_grad]: 2.22999e-06
        [auto_monad_eliminator]: 1.168e-05
        [cse]: 1.927e-05
        [replace_applicator]: 1.061e-05
    [Cycle 2]: 0.00028129, [27]
        [switch_simplify]: 4.65001e-06
        [loop_unroll]: 3.7e-06
        [a_1]: 3.996e-05
        [with_stream_mark]: 1.009e-05
        [recompute_prepare]: 3.95e-06
        [updatestate_depend_eliminate]: 2.17001e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.51002e-06
        [specialize_transform]: 3.63999e-06
        [updatestate_useless_node_eliminater]: 3.7e-06
        [accelerated_algorithm]: 6.17999e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 3.26001e-06
        [merge_forward]: 2.95002e-06
        [cell_reuse_recompute_pass]: 2.22001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.134e-05
        [j_node_and_user_rematch]: 6.03002e-06
        [meta_fg_expand]: 2.502e-05
        [replace_old_param]: 3.76001e-06
        [inline_without_move]: 3.63999e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 4.33999e-06
        [cse]: 1.037e-05
        [replace_applicator]: 4.1e-06
[py_interpret_to_execute_after_opt_a]: 1.087e-05
[rewriter_after_opt_a]: 3.37e-05
[convert_after_rewriter]: 7.13e-06
[order_py_execute_after_rewriter]: 5.34998e-06
[mutable_eliminate]: 0.00057045
[jit_opt_b]: 3.972e-05, [1]
    [Cycle 1]: 3.257e-05, [2]
        [frontend_op_eliminate]: 1.066e-05
        [inline_after_opt_a]: 1.103e-05
[cconv]: 2.169e-05
[loop_unroll]: 0.00043892
[jit_opt_after_cconv]: 0.00011444, [1]
    [Cycle 1]: 0.00010814, [11]
        [c_1]: 1.156e-05
        [parameter_eliminate]: 2.45002e-06
        [updatestate_depend_eliminate]: 4.30999e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 1.67001e-06
        [cse]: 1.364e-05
        [call_graph_tuple_transform]: 1.225e-05
        [tuple_list_get_item_eliminator]: 3.76999e-06
        [none_parameter_eliminate]: 1.40001e-06
        [renormalize]: 5.19998e-07
        [switch_simplify]: 3.6e-06
[remove_dup_value]: 1.361e-05
[partial_unused_args_eliminate]: 2.17999e-06
[environ_conv]: 4.68999e-06
[add_recomputation]: 3.221e-05
[cse_after_recomputation]: 1.894e-05, [1]
    [Cycle 1]: 1.347e-05, [1]
        [cse]: 7.61001e-06
[auto_monad_reorder]: 1.33e-05
[get_jit_bprop_graph]: 1.81e-06
[rewriter_after_jit_bprop_graph]: 4.90001e-06
[opt_after_jit_grad]: 0.00045556
[symbol_engine_optimizer]: 6.246e-05, [1]
    [Cycle 1]: 5.625e-05, [6]
        [build]: 2.59001e-06
        [elim_shapecalc]: 5.16998e-06
        [elim_not_effective]: 9.29e-06
        [opt_reshape]: 4.77e-06
        [fold_const_symbol]: 5.72999e-06
        [renormalize]: 4.19997e-07
[validate]: 2.659e-05
Sums
    bootstrap                            :     0.000169s :  1.14%
    type_inference                       :     0.008182s : 55.17%
    event_method                         :     0.000038s :  0.26%
    auto_monad                           :     0.000060s :  0.40%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000023s :  0.16%
    rewriter_before_opt_a                :     0.000085s :  0.58%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.44%
    jit_opt_a.loop_unroll                :     0.000041s :  0.27%
    jit_opt_a.a_1                        :     0.000850s :  5.73%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000016s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000016s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000019s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.10%
    jit_opt_a.merge_forward              :     0.000011s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.25%
    jit_opt_a.j_node_and_user_rematch    :     0.000029s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.001553s : 10.47%
    jit_opt_a.replace_old_param          :     0.000059s :  0.40%
    jit_opt_a.inline_without_move        :     0.000056s :  0.38%
    jit_opt_a.renormalize                :     0.001566s : 10.56%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.11%
    jit_opt_a.cse                        :     0.000030s :  0.20%
    jit_opt_a.replace_applicator         :     0.000015s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.07%
    rewriter_after_opt_a                 :     0.000034s :  0.23%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.04%
    mutable_eliminate                    :     0.000570s :  3.85%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000022s :  0.15%
    loop_unroll                          :     0.000439s :  2.96%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000014s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000012s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000014s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000032s :  0.22%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000456s :  3.07%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000005s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.18%


Time group info:
------[substitution.]   0.000243    50
 0.58% :     0.000001s :      1: substitution.elim_not_effective
 0.41% :     0.000001s :      1: substitution.fold_const_symbol
 1.55% :     0.000004s :      1: substitution.graph_param_transform
64.76% :     0.000158s :      6: substitution.inline
 7.63% :     0.000019s :      2: substitution.inline_without_move
 2.68% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.48% :     0.000004s :      2: substitution.minmaximum_grad
 1.73% :     0.000004s :      3: substitution.partial_eliminate
 2.98% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.02% :     0.000005s :      2: substitution.replace_applicator
 1.96% :     0.000005s :      5: substitution.replace_old_param
 1.16% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.83% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 2.07% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 6.17% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008138     2
88.79% :     0.007226s :      1: type_inference.infer
11.21% :     0.000912s :      1: type_inference.specialize
------[replace.]   0.000059     8
78.85% :     0.000047s :      6: replace.inline
21.15% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000165     8
93.54% :     0.000154s :      6: match.inline
 6.46% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000168  1109
 1.40% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.58% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.34% :     0.000002s :     18: predicate.addn_check_dump
 1.65% :     0.000003s :     18: predicate.addn_zero_filter
 2.23% :     0.000004s :     18: predicate.arithmetic_simplify
 1.50% :     0.000003s :     18: predicate.cast_eliminate
 0.11% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.28% :     0.000002s :     18: predicate.compare_switch_simplify
 1.59% :     0.000003s :     18: predicate.depend_value_elim
 1.39% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.76% :     0.000003s :     18: predicate.dict_get_item_eliminator
 1.38% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.39% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.09% :     0.000000s :      1: predicate.elim_not_effective
 0.25% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.35% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     18: predicate.environ_get_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.22% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.37% :     0.000007s :     28: predicate.inline
 3.46% :     0.000006s :     33: predicate.inline_without_move
 0.51% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.35% :     0.000002s :     12: predicate.less_batch_normalization
 1.63% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.84% :     0.000003s :     21: predicate.load_eliminater
 0.66% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.54% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.78% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.26% :     0.000002s :     18: predicate.merge_addn
 1.33% :     0.000002s :     18: predicate.minmaximum_grad
 0.94% :     0.000002s :      1: predicate.mutable_eliminate
 0.28% :     0.000000s :      1: predicate.opt_reshape
 2.02% :     0.000003s :     21: predicate.partial_eliminate
 1.43% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.89% :     0.000003s :     18: predicate.reduce_eliminate
 1.56% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.72% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.45% :     0.000002s :     22: predicate.replace_applicator
 1.92% :     0.000003s :     33: predicate.replace_old_param
 0.14% :     0.000000s :      1: predicate.reset_defer_inline
 1.56% :     0.000003s :     18: predicate.reshape_eliminate
 1.35% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.34% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     18: predicate.same_eliminate
 0.63% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.47% :     0.000001s :      2: predicate.special_op_eliminate
 1.17% :     0.000002s :     12: predicate.specialize_transform
 1.57% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.34% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.11% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.57% :     0.000004s :     26: predicate.switch_defer_inline
 2.32% :     0.000004s :     26: predicate.switch_layer_defer_inline
 7.08% :     0.000012s :     71: predicate.switch_simplify
 1.54% :     0.000003s :     18: predicate.tile_eliminate
 1.31% :     0.000002s :     18: predicate.transpose_eliminate
 1.87% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.52% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.15% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.60% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.53% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.50% :     0.000003s :     18: predicate.value_based_eliminate
 0.15% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.38% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001004    22
74.71% :     0.000750s :     13: func_graph_cloner_run.FuncGraphClonerGraph
25.29% :     0.000254s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.021536    72
 0.16% :     0.000035s :      1: add_recomputation
 0.29% :     0.000063s :      1: auto_monad
 0.07% :     0.000016s :      1: auto_monad_reorder
 0.83% :     0.000180s :      1: bootstrap
 0.11% :     0.000024s :      1: cconv
 0.05% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.19% :     0.000041s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
37.85% :     0.008151s :      1: jit_opt_a
 0.54% :     0.000117s :      1: jit_opt_after_cconv
 0.20% :     0.000042s :      1: jit_opt_b
 2.07% :     0.000446s :      1: loop_unroll
 2.68% :     0.000578s :      1: mutable_eliminate
 5.62% :     0.001210s :     26: opt.transform.jit_opt_a
 0.13% :     0.000027s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000015s :      4: opt.transform.jit_opt_b
 0.04% :     0.000008s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000010s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000013s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000022s :      4: opt.transform.symbol_engine_opt
 2.15% :     0.000464s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000026s :      1: py_interpret_to_execute
 0.06% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 5.60% :     0.001205s :      1: renormalize.infer
 1.64% :     0.000353s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000037s :      1: rewriter_after_opt_a
 0.41% :     0.000089s :      1: rewriter_before_opt_a
 0.30% :     0.000065s :      1: symbol_engine_optimizer
38.03% :     0.008191s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.299.085 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843c1940> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.299.655 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.03198, [33]
[bootstrap]: 0.00020077
[type_inference]: 0.00931188
[event_method]: 4.54e-05
[auto_monad]: 8.718e-05
[graph_reusing]: 6.22001e-06
[pre_auto_parallel]: 2.43e-06
[py_interpret_to_execute]: 2.308e-05
[rewriter_before_opt_a]: 8.521e-05
[expand_dump_flag]: 3.27002e-06
[jit_opt_a]: 0.0116186, [3]
    [Cycle 1]: 0.00565508, [27]
        [switch_simplify]: 6.154e-05
        [loop_unroll]: 3.752e-05
        [a_1]: 0.0008101
        [with_stream_mark]: 2.783e-05
        [recompute_prepare]: 1.834e-05
        [updatestate_depend_eliminate]: 7.2e-06
        [updatestate_assign_eliminate]: 6.26e-06
        [updatestate_loads_eliminate]: 5.48002e-06
        [parameter_eliminate]: 2.71e-06
        [specialize_transform]: 1.268e-05
        [updatestate_useless_node_eliminater]: 1.151e-05
        [accelerated_algorithm]: 1.317e-05
        [meta_shard_fg_expand]: 2.68e-06
        [get_grad_eliminate_]: 1.163e-05
        [merge_forward]: 8.22998e-06
        [cell_reuse_recompute_pass]: 8.90024e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.622e-05
        [j_node_and_user_rematch]: 2.353e-05
        [meta_fg_expand]: 0.00152826
        [replace_old_param]: 5.51e-05
        [inline_without_move]: 5.16e-05
        [renormalize]: 0.00246276
        [add_forward_monad_depend]: 7.99002e-06
        [auto_monad_grad]: 4.94e-06
        [auto_monad_eliminator]: 4.803e-05
        [cse]: 0.00014259
        [replace_applicator]: 6.389e-05
    [Cycle 2]: 0.00215068, [27]
        [switch_simplify]: 3.688e-05
        [loop_unroll]: 3.586e-05
        [a_1]: 0.00111849
        [with_stream_mark]: 1.305e-05
        [recompute_prepare]: 9.07001e-06
        [updatestate_depend_eliminate]: 4.60001e-06
        [updatestate_assign_eliminate]: 4.12e-06
        [updatestate_loads_eliminate]: 3.60998e-06
        [parameter_eliminate]: 1.25999e-06
        [specialize_transform]: 9.10001e-06
        [updatestate_useless_node_eliminater]: 8.22e-06
        [accelerated_algorithm]: 9.25999e-06
        [meta_shard_fg_expand]: 1.84e-06
        [get_grad_eliminate_]: 8.22003e-06
        [merge_forward]: 4.07e-06
        [cell_reuse_recompute_pass]: 1.19003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.765e-05
        [j_node_and_user_rematch]: 1.42e-05
        [meta_fg_expand]: 3.829e-05
        [replace_old_param]: 1.333e-05
        [inline_without_move]: 8.48999e-06
        [renormalize]: 0.00055849
        [add_forward_monad_depend]: 4.99003e-06
        [auto_monad_grad]: 1.46002e-06
        [auto_monad_eliminator]: 1.553e-05
        [cse]: 4.43e-05
        [replace_applicator]: 1.494e-05
    [Cycle 3]: 0.00049481, [27]
        [switch_simplify]: 8.69e-06
        [loop_unroll]: 7.85998e-06
        [a_1]: 0.00019326
        [with_stream_mark]: 1.026e-05
        [recompute_prepare]: 8.29002e-06
        [updatestate_depend_eliminate]: 4.94998e-06
        [updatestate_assign_eliminate]: 4.28999e-06
        [updatestate_loads_eliminate]: 3.71001e-06
        [parameter_eliminate]: 1.02998e-06
        [specialize_transform]: 7.93001e-06
        [updatestate_useless_node_eliminater]: 7.82e-06
        [accelerated_algorithm]: 8.79e-06
        [meta_shard_fg_expand]: 1.74e-06
        [get_grad_eliminate_]: 7.71999e-06
        [merge_forward]: 4.15e-06
        [cell_reuse_recompute_pass]: 1.64e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.804e-05
        [j_node_and_user_rematch]: 1.369e-05
        [meta_fg_expand]: 2.88998e-06
        [replace_old_param]: 1.066e-05
        [inline_without_move]: 7.48999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.60999e-06
        [auto_monad_grad]: 8.90024e-07
        [auto_monad_eliminator]: 1.069e-05
        [cse]: 2.32e-05
        [replace_applicator]: 8.16002e-06
[py_interpret_to_execute_after_opt_a]: 1.286e-05
[rewriter_after_opt_a]: 4.506e-05
[convert_after_rewriter]: 8.70001e-06
[order_py_execute_after_rewriter]: 5.67999e-06
[mutable_eliminate]: 0.00056749
[jit_opt_b]: 6.722e-05, [1]
    [Cycle 1]: 6.005e-05, [2]
        [frontend_op_eliminate]: 2.466e-05
        [inline_after_opt_a]: 2.398e-05
[cconv]: 1.964e-05
[loop_unroll]: 0.00041739
[jit_opt_after_cconv]: 0.00018669, [1]
    [Cycle 1]: 0.00018004, [11]
        [c_1]: 3.449e-05
        [parameter_eliminate]: 2.46998e-06
        [updatestate_depend_eliminate]: 8.18999e-06
        [updatestate_assign_eliminate]: 4.08999e-06
        [updatestate_loads_eliminate]: 4.05e-06
        [cse]: 3.118e-05
        [call_graph_tuple_transform]: 2.462e-05
        [tuple_list_get_item_eliminator]: 8.55999e-06
        [none_parameter_eliminate]: 1.33002e-06
        [renormalize]: 4.7998e-07
        [switch_simplify]: 8.25e-06
[remove_dup_value]: 1.88e-05
[partial_unused_args_eliminate]: 1.77999e-06
[environ_conv]: 5.85002e-06
[add_recomputation]: 4.904e-05
[cse_after_recomputation]: 3.211e-05, [1]
    [Cycle 1]: 2.659e-05, [1]
        [cse]: 2.064e-05
[auto_monad_reorder]: 1.798e-05
[get_jit_bprop_graph]: 1.40001e-06
[rewriter_after_jit_bprop_graph]: 5.29998e-06
[opt_after_jit_grad]: 0.00046612
[symbol_engine_optimizer]: 9.031e-05, [1]
    [Cycle 1]: 8.389e-05, [6]
        [build]: 3.73001e-06
        [elim_shapecalc]: 1.07e-05
        [elim_not_effective]: 1.801e-05
        [opt_reshape]: 8.77e-06
        [fold_const_symbol]: 1.482e-05
        [renormalize]: 5.59987e-07
[validate]: 4.24e-05
[backend_pass]: 7.89994e-07
[task_emit]: 0.00833171
[execute]: 6.43e-06
Sums
    bootstrap                            :     0.000201s :  0.72%
    type_inference                       :     0.009312s : 33.40%
    event_method                         :     0.000045s :  0.16%
    auto_monad                           :     0.000087s :  0.31%
    graph_reusing                        :     0.000006s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000023s :  0.08%
    rewriter_before_opt_a                :     0.000085s :  0.31%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000107s :  0.38%
    jit_opt_a.loop_unroll                :     0.000081s :  0.29%
    jit_opt_a.a_1                        :     0.002122s :  7.61%
    jit_opt_a.with_stream_mark           :     0.000051s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000036s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000013s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000030s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000028s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000031s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000028s :  0.10%
    jit_opt_a.merge_forward              :     0.000016s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000062s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000051s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001569s :  5.63%
    jit_opt_a.replace_old_param          :     0.000079s :  0.28%
    jit_opt_a.inline_without_move        :     0.000068s :  0.24%
    jit_opt_a.renormalize                :     0.003021s : 10.84%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000074s :  0.27%
    jit_opt_a.cse                        :     0.000210s :  0.75%
    jit_opt_a.replace_applicator         :     0.000087s :  0.31%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.05%
    rewriter_after_opt_a                 :     0.000045s :  0.16%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000567s :  2.04%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000024s :  0.09%
    cconv                                :     0.000020s :  0.07%
    loop_unroll                          :     0.000417s :  1.50%
    jit_opt_after_cconv.c_1              :     0.000034s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000031s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000008s :  0.03%
    remove_dup_value                     :     0.000019s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000049s :  0.18%
    cse_after_recomputation.cse          :     0.000021s :  0.07%
    auto_monad_reorder                   :     0.000018s :  0.06%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.02%
    opt_after_jit_grad                   :     0.000466s :  1.67%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000042s :  0.15%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008332s : 29.89%
    execute                              :     0.000006s :  0.02%


Time group info:
------[substitution.]   0.000508   126
 6.75% :     0.000034s :      2: substitution.arithmetic_simplify
 0.50% :     0.000003s :      5: substitution.elim_not_effective
 0.40% :     0.000002s :      5: substitution.fold_const_symbol
 1.14% :     0.000006s :      6: substitution.graph_param_transform
59.42% :     0.000302s :     12: substitution.inline
 3.50% :     0.000018s :      2: substitution.inline_without_move
 1.95% :     0.000010s :     18: substitution.j_node_and_user_rematch
 1.93% :     0.000010s :      7: substitution.minmaximum_grad
 1.09% :     0.000006s :      5: substitution.partial_eliminate
 2.65% :     0.000013s :     18: substitution.remove_not_recompute_node
 4.95% :     0.000025s :     10: substitution.replace_applicator
 1.69% :     0.000009s :      7: substitution.replace_old_param
 0.51% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.74% :     0.000019s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.41% :     0.000012s :      7: substitution.tuple_list_get_item_depend_reorder
 7.38% :     0.000037s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009247     2
89.15% :     0.008244s :      1: type_inference.infer
10.85% :     0.001004s :      1: type_inference.specialize
------[replace.]   0.000137    19
64.17% :     0.000088s :     12: replace.inline
35.83% :     0.000049s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000314    19
94.14% :     0.000295s :     12: match.inline
 5.86% :     0.000018s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000406  2895
 1.67% :     0.000007s :     48: predicate.accumulaten_eliminater
 0.52% :     0.000002s :      6: predicate.ad_related_special_op_eliminate
 1.45% :     0.000006s :     48: predicate.addn_check_dump
 1.69% :     0.000007s :     48: predicate.addn_zero_filter
 2.30% :     0.000009s :     48: predicate.arithmetic_simplify
 1.52% :     0.000006s :     48: predicate.cast_eliminate
 0.24% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.43% :     0.000006s :     48: predicate.compare_switch_simplify
 1.60% :     0.000007s :     48: predicate.depend_value_elim
 1.45% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.53% :     0.000006s :     48: predicate.dict_get_item_eliminator
 1.53% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.36% :     0.000001s :      6: predicate.dumpgradient_eliminate
 0.19% :     0.000001s :      6: predicate.elim_not_effective
 0.25% :     0.000001s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.51% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.44% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.45% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.59% :     0.000006s :     48: predicate.environ_get_eliminate
 1.48% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.12% :     0.000000s :      6: predicate.fold_const_symbol
 0.95% :     0.000004s :     23: predicate.get_grad_eliminate
 0.12% :     0.000000s :      6: predicate.graph_param_transform
 4.05% :     0.000016s :     79: predicate.inline
 1.77% :     0.000007s :     44: predicate.inline_without_move
 0.38% :     0.000002s :     23: predicate.j_node_and_user_rematch
 1.51% :     0.000006s :     23: predicate.less_batch_normalization
 1.84% :     0.000007s :     55: predicate.list_to_tuple_eliminator_
 1.98% :     0.000008s :     61: predicate.load_eliminater
 0.50% :     0.000002s :      6: predicate.loop_unroll_after_grad
 2.95% :     0.000012s :     90: predicate.loop_unroll_before_grad
 1.95% :     0.000008s :     54: predicate.make_slice_get_slice_eliminator
 1.46% :     0.000006s :     48: predicate.merge_addn
 1.46% :     0.000006s :     48: predicate.minmaximum_grad
 0.50% :     0.000002s :      6: predicate.mutable_eliminate
 0.24% :     0.000001s :      6: predicate.opt_reshape
 2.30% :     0.000009s :     61: predicate.partial_eliminate
 1.56% :     0.000006s :     48: predicate.print_const_string_wrapper
 1.76% :     0.000007s :     48: predicate.reduce_eliminate
 1.82% :     0.000007s :     55: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.47% :     0.000010s :    108: predicate.replace_applicator
 1.00% :     0.000004s :     44: predicate.replace_old_param
 0.13% :     0.000001s :      6: predicate.reset_defer_inline
 1.77% :     0.000007s :     48: predicate.reshape_eliminate
 1.51% :     0.000006s :     48: predicate.row_tensor_add_zeros_like
 0.36% :     0.000001s :      6: predicate.row_tensor_eliminate
 1.58% :     0.000006s :     48: predicate.same_eliminate
 0.50% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.50% :     0.000002s :     12: predicate.special_op_eliminate
 0.93% :     0.000004s :     23: predicate.specialize_transform
 1.70% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000006s :     48: predicate.stack_unstack_eliminate
 0.23% :     0.000001s :      6: predicate.switch_call_monad_eliminater
 2.53% :     0.000010s :     67: predicate.switch_defer_inline
 2.43% :     0.000010s :     67: predicate.switch_layer_defer_inline
 5.94% :     0.000024s :    163: predicate.switch_simplify
 1.53% :     0.000006s :     48: predicate.tile_eliminate
 1.53% :     0.000006s :     48: predicate.transpose_eliminate
 1.85% :     0.000008s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.59% :     0.000006s :     48: predicate.tuple_list_get_item_depend_reorder
 2.84% :     0.000012s :     67: predicate.tuple_list_get_item_eliminator
 1.74% :     0.000007s :     48: predicate.tuple_list_set_item_eliminator
 1.75% :     0.000007s :     55: predicate.tuple_to_list_eliminator_
 1.93% :     0.000008s :     61: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000012s :     84: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000007s :     48: predicate.value_based_eliminate
 0.20% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.35% :     0.000001s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001333    29
59.60% :     0.000794s :     13: func_graph_cloner_run.FuncGraphClonerGraph
40.40% :     0.000538s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.037948    91
 0.14% :     0.000052s :      1: add_recomputation
 0.24% :     0.000091s :      1: auto_monad
 0.05% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000003s :      1: backend_pass
 0.57% :     0.000216s :      1: bootstrap
 0.06% :     0.000022s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.09% :     0.000034s :      1: cse_after_recomputation
 0.02% :     0.000008s :      1: environ_conv
 0.13% :     0.000049s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.01% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000009s :      1: graph_reusing
30.63% :     0.011622s :      1: jit_opt_a
 0.50% :     0.000189s :      1: jit_opt_after_cconv
 0.18% :     0.000070s :      1: jit_opt_b
 1.12% :     0.000425s :      1: loop_unroll
 1.52% :     0.000577s :      1: mutable_eliminate
 7.29% :     0.002767s :     39: opt.transform.jit_opt_a
 0.19% :     0.000072s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000049s :      4: opt.transform.symbol_engine_opt
 1.25% :     0.000475s :      1: opt_after_jit_grad
 0.02% :     0.000008s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000026s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000021s :      1: remove_dup_value
 4.35% :     0.001649s :      2: renormalize.infer
 3.58% :     0.001359s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000048s :      1: rewriter_after_opt_a
 0.23% :     0.000088s :      1: rewriter_before_opt_a
 0.24% :     0.000093s :      1: symbol_engine_optimizer
21.98% :     0.008343s :      1: task_emit
24.57% :     0.009325s :      1: type_inference
 0.18% :     0.000067s :      1: validate

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.314.563 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80f4160> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.315.159 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0197624, [30]
[bootstrap]: 0.00019845
[type_inference]: 0.00923838
[event_method]: 5.279e-05
[auto_monad]: 9.257e-05
[graph_reusing]: 7.16999e-06
[pre_auto_parallel]: 2.63e-06
[py_interpret_to_execute]: 2.634e-05
[rewriter_before_opt_a]: 8.988e-05
[expand_dump_flag]: 3.39001e-06
[jit_opt_a]: 0.00799207, [2]
    [Cycle 1]: 0.00439148, [27]
        [switch_simplify]: 6.131e-05
        [loop_unroll]: 3.846e-05
        [a_1]: 0.0008941
        [with_stream_mark]: 2.888e-05
        [recompute_prepare]: 1.957e-05
        [updatestate_depend_eliminate]: 8.37e-06
        [updatestate_assign_eliminate]: 7e-06
        [updatestate_loads_eliminate]: 5.91e-06
        [parameter_eliminate]: 2.69999e-06
        [specialize_transform]: 1.281e-05
        [updatestate_useless_node_eliminater]: 1.151e-05
        [accelerated_algorithm]: 1.19e-05
        [meta_shard_fg_expand]: 3.26001e-06
        [get_grad_eliminate_]: 1.173e-05
        [merge_forward]: 8.13999e-06
        [cell_reuse_recompute_pass]: 1.02998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.641e-05
        [j_node_and_user_rematch]: 2.241e-05
        [meta_fg_expand]: 0.0014334
        [replace_old_param]: 5.082e-05
        [inline_without_move]: 5.179e-05
        [renormalize]: 0.00141405
        [add_forward_monad_depend]: 6.14999e-06
        [auto_monad_grad]: 2.86e-06
        [auto_monad_eliminator]: 1.366e-05
        [cse]: 2.738e-05
        [replace_applicator]: 1.16e-05
    [Cycle 2]: 0.00036349, [27]
        [switch_simplify]: 4.16001e-06
        [loop_unroll]: 3.55998e-06
        [a_1]: 3.949e-05
        [with_stream_mark]: 8.85001e-06
        [recompute_prepare]: 3.91999e-06
        [updatestate_depend_eliminate]: 2.26e-06
        [updatestate_assign_eliminate]: 7.844e-05
        [updatestate_loads_eliminate]: 1.97001e-06
        [parameter_eliminate]: 1.17999e-06
        [specialize_transform]: 4.07e-06
        [updatestate_useless_node_eliminater]: 3.56999e-06
        [accelerated_algorithm]: 3.95e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 3.6e-06
        [merge_forward]: 2.29001e-06
        [cell_reuse_recompute_pass]: 2.01e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.244e-05
        [j_node_and_user_rematch]: 6.01e-06
        [meta_fg_expand]: 2.414e-05
        [replace_old_param]: 4.33001e-06
        [inline_without_move]: 3.28e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.74e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 5.20999e-06
        [cse]: 1.228e-05
        [replace_applicator]: 4.43999e-06
[py_interpret_to_execute_after_opt_a]: 9.99999e-06
[rewriter_after_opt_a]: 3.445e-05
[convert_after_rewriter]: 6.49999e-06
[order_py_execute_after_rewriter]: 4.95999e-06
[mutable_eliminate]: 0.00056868
[jit_opt_b]: 4.099e-05, [1]
    [Cycle 1]: 3.35e-05, [2]
        [frontend_op_eliminate]: 1.028e-05
        [inline_after_opt_a]: 1.172e-05
[cconv]: 2.273e-05
[loop_unroll]: 0.00043152
[jit_opt_after_cconv]: 0.00011987, [1]
    [Cycle 1]: 0.00011268, [11]
        [c_1]: 1.133e-05
        [parameter_eliminate]: 2.61999e-06
        [updatestate_depend_eliminate]: 4.89003e-06
        [updatestate_assign_eliminate]: 1.82999e-06
        [updatestate_loads_eliminate]: 1.59998e-06
        [cse]: 1.685e-05
        [call_graph_tuple_transform]: 1.378e-05
        [tuple_list_get_item_eliminator]: 3.73001e-06
        [none_parameter_eliminate]: 1.42999e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 3.85998e-06
[remove_dup_value]: 1.634e-05
[partial_unused_args_eliminate]: 2.47001e-06
[environ_conv]: 5.52001e-06
[add_recomputation]: 3.423e-05
[cse_after_recomputation]: 1.993e-05, [1]
    [Cycle 1]: 1.438e-05, [1]
        [cse]: 8.35001e-06
[auto_monad_reorder]: 1.447e-05
[get_jit_bprop_graph]: 1.86998e-06
[rewriter_after_jit_bprop_graph]: 5.15999e-06
[opt_after_jit_grad]: 0.00046602
[symbol_engine_optimizer]: 6.525e-05, [1]
    [Cycle 1]: 5.796e-05, [6]
        [build]: 2.88e-06
        [elim_shapecalc]: 5.87999e-06
        [elim_not_effective]: 1.005e-05
        [opt_reshape]: 4.01001e-06
        [fold_const_symbol]: 5.92001e-06
        [renormalize]: 5.50004e-07
[validate]: 2.924e-05
Sums
    bootstrap                            :     0.000198s :  1.25%
    type_inference                       :     0.009238s : 58.05%
    event_method                         :     0.000053s :  0.33%
    auto_monad                           :     0.000093s :  0.58%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.17%
    rewriter_before_opt_a                :     0.000090s :  0.56%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000065s :  0.41%
    jit_opt_a.loop_unroll                :     0.000042s :  0.26%
    jit_opt_a.a_1                        :     0.000934s :  5.87%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.24%
    jit_opt_a.recompute_prepare          :     0.000023s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000085s :  0.54%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.10%
    jit_opt_a.merge_forward              :     0.000010s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001458s :  9.16%
    jit_opt_a.replace_old_param          :     0.000055s :  0.35%
    jit_opt_a.inline_without_move        :     0.000055s :  0.35%
    jit_opt_a.renormalize                :     0.001414s :  8.89%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.12%
    jit_opt_a.cse                        :     0.000040s :  0.25%
    jit_opt_a.replace_applicator         :     0.000016s :  0.10%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000034s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000569s :  3.57%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000012s :  0.07%
    cconv                                :     0.000023s :  0.14%
    loop_unroll                          :     0.000432s :  2.71%
    jit_opt_after_cconv.c_1              :     0.000011s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000014s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000016s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.03%
    add_recomputation                    :     0.000034s :  0.22%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000014s :  0.09%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000466s :  2.93%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000029s :  0.18%


Time group info:
------[substitution.]   0.000308    50
 0.61% :     0.000002s :      1: substitution.elim_not_effective
 0.35% :     0.000001s :      1: substitution.fold_const_symbol
 1.32% :     0.000004s :      1: substitution.graph_param_transform
51.77% :     0.000159s :      6: substitution.inline
 5.60% :     0.000017s :      2: substitution.inline_without_move
 2.00% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.25% :     0.000004s :      2: substitution.minmaximum_grad
 1.44% :     0.000004s :      3: substitution.partial_eliminate
 2.41% :     0.000007s :      9: substitution.remove_not_recompute_node
 1.88% :     0.000006s :      2: substitution.replace_applicator
 1.39% :     0.000004s :      5: substitution.replace_old_param
 0.95% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.09% :     0.000006s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.52% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
25.41% :     0.000078s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009173     2
89.37% :     0.008198s :      1: type_inference.infer
10.63% :     0.000975s :      1: type_inference.specialize
------[replace.]   0.000063     8
78.75% :     0.000049s :      6: replace.inline
21.25% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000167     8
93.13% :     0.000156s :      6: match.inline
 6.87% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000168  1109
 1.49% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.52% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.28% :     0.000002s :     18: predicate.addn_check_dump
 1.75% :     0.000003s :     18: predicate.addn_zero_filter
 2.13% :     0.000004s :     18: predicate.arithmetic_simplify
 1.41% :     0.000002s :     18: predicate.cast_eliminate
 0.15% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.30% :     0.000002s :     18: predicate.compare_switch_simplify
 1.55% :     0.000003s :     18: predicate.depend_value_elim
 1.27% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.42% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.40% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      1: predicate.elim_not_effective
 0.24% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.35% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.29% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.36% :     0.000002s :     18: predicate.environ_get_eliminate
 1.27% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.13% :     0.000002s :     12: predicate.get_grad_eliminate
 0.10% :     0.000000s :      1: predicate.graph_param_transform
 4.38% :     0.000007s :     28: predicate.inline
 3.21% :     0.000005s :     33: predicate.inline_without_move
 0.52% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :     12: predicate.less_batch_normalization
 1.63% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.69% :     0.000003s :     21: predicate.load_eliminater
 1.13% :     0.000002s :      1: predicate.loop_unroll_after_grad
 3.63% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.68% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.32% :     0.000002s :     18: predicate.merge_addn
 1.29% :     0.000002s :     18: predicate.minmaximum_grad
 0.90% :     0.000002s :      1: predicate.mutable_eliminate
 0.21% :     0.000000s :      1: predicate.opt_reshape
 2.16% :     0.000004s :     21: predicate.partial_eliminate
 1.34% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.80% :     0.000003s :     18: predicate.reduce_eliminate
 1.56% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.77% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.44% :     0.000002s :     22: predicate.replace_applicator
 1.75% :     0.000003s :     33: predicate.replace_old_param
 0.11% :     0.000000s :      1: predicate.reset_defer_inline
 1.47% :     0.000002s :     18: predicate.reshape_eliminate
 1.71% :     0.000003s :     18: predicate.row_tensor_add_zeros_like
 0.45% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.51% :     0.000003s :     18: predicate.same_eliminate
 0.64% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.39% :     0.000001s :      2: predicate.special_op_eliminate
 1.16% :     0.000002s :     12: predicate.specialize_transform
 1.56% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000003s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.61% :     0.000004s :     26: predicate.switch_defer_inline
 2.67% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.85% :     0.000012s :     71: predicate.switch_simplify
 1.32% :     0.000002s :     18: predicate.tile_eliminate
 1.57% :     0.000003s :     18: predicate.transpose_eliminate
 1.61% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.49% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 3.39% :     0.000006s :     22: predicate.tuple_list_get_item_eliminator
 1.68% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.53% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.59% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 2.96% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000003s :     18: predicate.value_based_eliminate
 0.14% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.32% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001035    22
75.14% :     0.000778s :     13: func_graph_cloner_run.FuncGraphClonerGraph
24.86% :     0.000257s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022489    72
 0.16% :     0.000037s :      1: add_recomputation
 0.43% :     0.000096s :      1: auto_monad
 0.08% :     0.000017s :      1: auto_monad_reorder
 0.94% :     0.000212s :      1: bootstrap
 0.11% :     0.000025s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000008s :      1: environ_conv
 0.26% :     0.000058s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000009s :      1: graph_reusing
35.55% :     0.007995s :      1: jit_opt_a
 0.55% :     0.000123s :      1: jit_opt_after_cconv
 0.19% :     0.000044s :      1: jit_opt_b
 1.95% :     0.000439s :      1: loop_unroll
 2.57% :     0.000578s :      1: mutable_eliminate
 5.73% :     0.001290s :     26: opt.transform.jit_opt_a
 0.13% :     0.000029s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000015s :      4: opt.transform.jit_opt_b
 0.05% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000023s :      4: opt.transform.symbol_engine_opt
 2.11% :     0.000475s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.13% :     0.000029s :      1: py_interpret_to_execute
 0.06% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000019s :      1: remove_dup_value
 4.78% :     0.001076s :      1: renormalize.infer
 1.47% :     0.000330s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000038s :      1: rewriter_after_opt_a
 0.41% :     0.000093s :      1: rewriter_before_opt_a
 0.30% :     0.000068s :      1: symbol_engine_optimizer
41.13% :     0.009250s :      1: type_inference


TotalTime = 0.0160786, [30]
[bootstrap]: 0.00041107
[type_inference]: 0.00539548
[event_method]: 1.814e-05
[auto_monad]: 7.285e-05
[graph_reusing]: 5.87999e-06
[pre_auto_parallel]: 2.94999e-06
[py_interpret_to_execute]: 3.684e-05
[rewriter_before_opt_a]: 7.898e-05
[expand_dump_flag]: 3.48999e-06
[jit_opt_a]: 0.00705406, [2]
    [Cycle 1]: 0.0021983, [27]
        [switch_simplify]: 7.175e-05
        [loop_unroll]: 3.161e-05
        [a_1]: 0.00066986
        [with_stream_mark]: 3.124e-05
        [recompute_prepare]: 1.118e-05
        [updatestate_depend_eliminate]: 4.44002e-06
        [updatestate_assign_eliminate]: 3.72998e-06
        [updatestate_loads_eliminate]: 2.63998e-06
        [parameter_eliminate]: 2.39001e-06
        [specialize_transform]: 6.83e-06
        [updatestate_useless_node_eliminater]: 5.59e-06
        [accelerated_algorithm]: 6.83e-06
        [meta_shard_fg_expand]: 2.89999e-06
        [get_grad_eliminate_]: 5.47001e-06
        [merge_forward]: 4.86002e-06
        [cell_reuse_recompute_pass]: 1.54e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.901e-05
        [j_node_and_user_rematch]: 1.092e-05
        [meta_fg_expand]: 3.01999e-06
        [replace_old_param]: 1.048e-05
        [inline_without_move]: 5.96e-06
        [renormalize]: 0.00094195
        [add_forward_monad_depend]: 1.07e-05
        [auto_monad_grad]: 3.37002e-06
        [auto_monad_eliminator]: 2.142e-05
        [cse]: 3.786e-05
        [replace_applicator]: 2.041e-05
    [Cycle 2]: 0.00040881, [27]
        [switch_simplify]: 6.94001e-06
        [loop_unroll]: 5.10999e-06
        [a_1]: 9.909e-05
        [with_stream_mark]: 1.761e-05
        [recompute_prepare]: 5.62999e-06
        [updatestate_depend_eliminate]: 4.31002e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 2.63998e-06
        [parameter_eliminate]: 1.60999e-06
        [specialize_transform]: 4.99003e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 6.21998e-06
        [meta_shard_fg_expand]: 2.12001e-06
        [get_grad_eliminate_]: 5.14e-06
        [merge_forward]: 4.35999e-06
        [cell_reuse_recompute_pass]: 2.73e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.006e-05
        [j_node_and_user_rematch]: 1.129e-05
        [meta_fg_expand]: 4.07e-06
        [replace_old_param]: 1.7e-05
        [inline_without_move]: 4.90999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.34001e-06
        [auto_monad_grad]: 1.09e-06
        [auto_monad_eliminator]: 9.35001e-06
        [cse]: 1.785e-05
        [replace_applicator]: 7.73999e-06
[py_interpret_to_execute_after_opt_a]: 2.147e-05
[rewriter_after_opt_a]: 4.759e-05
[convert_after_rewriter]: 8.33999e-06
[order_py_execute_after_rewriter]: 7.29001e-06
[mutable_eliminate]: 0.00097996
[jit_opt_b]: 6.351e-05, [1]
    [Cycle 1]: 5.179e-05, [2]
        [frontend_op_eliminate]: 1.919e-05
        [inline_after_opt_a]: 1.811e-05
[cconv]: 3.563e-05
[loop_unroll]: 0.0006177
[jit_opt_after_cconv]: 0.00017942, [1]
    [Cycle 1]: 0.00017083, [11]
        [c_1]: 2.366e-05
        [parameter_eliminate]: 5.65001e-06
        [updatestate_depend_eliminate]: 1.03e-05
        [updatestate_assign_eliminate]: 2.96001e-06
        [updatestate_loads_eliminate]: 3.09999e-06
        [cse]: 3.439e-05
        [call_graph_tuple_transform]: 2.313e-05
        [tuple_list_get_item_eliminator]: 6.54999e-06
        [none_parameter_eliminate]: 1.67999e-06
        [renormalize]: 1.69995e-07
        [switch_simplify]: 5.49998e-06
[remove_dup_value]: 1.764e-05
[partial_unused_args_eliminate]: 2.40002e-06
[environ_conv]: 5.72999e-06
[add_recomputation]: 5.962e-05
[cse_after_recomputation]: 2.195e-05, [1]
    [Cycle 1]: 1.551e-05, [1]
        [cse]: 9.29998e-06
[auto_monad_reorder]: 1.799e-05
[get_jit_bprop_graph]: 2.31e-06
[rewriter_after_jit_bprop_graph]: 9.97001e-06
[opt_after_jit_grad]: 0.00054785
[symbol_engine_optimizer]: 7.944e-05, [1]
    [Cycle 1]: 7.151e-05, [6]
        [build]: 5.17e-06
        [elim_shapecalc]: 7.91001e-06
        [elim_not_effective]: 1.387e-05
        [opt_reshape]: 5.94e-06
        [fold_const_symbol]: 8.82999e-06
        [renormalize]: 6.30011e-07
[validate]: 4.632e-05
Sums
    bootstrap                            :     0.000411s :  3.78%
    type_inference                       :     0.005395s : 49.59%
    event_method                         :     0.000018s :  0.17%
    auto_monad                           :     0.000073s :  0.67%
    graph_reusing                        :     0.000006s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000037s :  0.34%
    rewriter_before_opt_a                :     0.000079s :  0.73%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000079s :  0.72%
    jit_opt_a.loop_unroll                :     0.000037s :  0.34%
    jit_opt_a.a_1                        :     0.000769s :  7.07%
    jit_opt_a.with_stream_mark           :     0.000049s :  0.45%
    jit_opt_a.recompute_prepare          :     0.000017s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.10%
    jit_opt_a.merge_forward              :     0.000009s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000039s :  0.36%
    jit_opt_a.j_node_and_user_rematch    :     0.000022s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000007s :  0.07%
    jit_opt_a.replace_old_param          :     0.000027s :  0.25%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000942s :  8.66%
    jit_opt_a.add_forward_monad_depend   :     0.000013s :  0.12%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000031s :  0.28%
    jit_opt_a.cse                        :     0.000056s :  0.51%
    jit_opt_a.replace_applicator         :     0.000028s :  0.26%
    py_interpret_to_execute_after_opt_a  :     0.000021s :  0.20%
    rewriter_after_opt_a                 :     0.000048s :  0.44%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000007s :  0.07%
    mutable_eliminate                    :     0.000980s :  9.01%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.17%
    cconv                                :     0.000036s :  0.33%
    loop_unroll                          :     0.000618s :  5.68%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000006s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.09%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000034s :  0.32%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000018s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.05%
    add_recomputation                    :     0.000060s :  0.55%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000018s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000010s :  0.09%
    opt_after_jit_grad                   :     0.000548s :  5.04%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.05%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.08%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000046s :  0.43%


Time group info:
------[substitution.]   0.000293    23
 0.64% :     0.000002s :      2: substitution.elim_not_effective
 0.45% :     0.000001s :      2: substitution.fold_const_symbol
 2.07% :     0.000006s :      3: substitution.graph_param_transform
85.54% :     0.000251s :      4: substitution.inline
 1.87% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.17% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.40% :     0.000007s :      2: substitution.replace_old_param
 4.86% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005311     2
81.40% :     0.004323s :      1: type_inference.infer
18.60% :     0.000988s :      1: type_inference.specialize
------[replace.]   0.000089     6
65.16% :     0.000058s :      4: replace.inline
34.84% :     0.000031s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000260     6
94.95% :     0.000247s :      4: match.inline
 5.05% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000146   691
 1.09% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.06% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.86% :     0.000001s :     10: predicate.addn_check_dump
 1.76% :     0.000003s :     10: predicate.addn_zero_filter
 2.24% :     0.000003s :     10: predicate.arithmetic_simplify
 1.19% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.82% :     0.000001s :     10: predicate.compare_switch_simplify
 1.96% :     0.000003s :     10: predicate.depend_value_elim
 0.87% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.49% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.60% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.86% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.83% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.95% :     0.000001s :     10: predicate.environ_get_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.53% :     0.000001s :      3: predicate.graph_param_transform
 6.43% :     0.000009s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.50% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.75% :     0.000003s :      6: predicate.less_batch_normalization
 1.10% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.40% :     0.000002s :     15: predicate.load_eliminater
 1.97% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.61% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.82% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.85% :     0.000001s :     10: predicate.merge_addn
 0.85% :     0.000001s :     10: predicate.minmaximum_grad
 2.80% :     0.000004s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 1.93% :     0.000003s :     15: predicate.partial_eliminate
 1.36% :     0.000002s :     10: predicate.print_const_string_wrapper
 2.08% :     0.000003s :     10: predicate.reduce_eliminate
 1.17% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.86% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.21% :     0.000000s :      3: predicate.reset_defer_inline
 1.26% :     0.000002s :     10: predicate.reshape_eliminate
 0.91% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.15% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.12% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 1.14% :     0.000002s :      6: predicate.specialize_transform
 1.19% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.34% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.59% :     0.000004s :     16: predicate.switch_defer_inline
 1.69% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.31% :     0.000012s :     53: predicate.switch_simplify
 1.44% :     0.000002s :     10: predicate.tile_eliminate
 0.88% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.02% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.54% :     0.000007s :     18: predicate.tuple_list_get_item_eliminator
 1.88% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.63% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.39% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.33% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.76% :     0.000003s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000721    12
49.74% :     0.000359s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 8.74% :     0.000063s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.52% :     0.000300s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.018124    72
 0.35% :     0.000063s :      1: add_recomputation
 0.42% :     0.000077s :      1: auto_monad
 0.11% :     0.000020s :      1: auto_monad_reorder
 2.43% :     0.000440s :      1: bootstrap
 0.21% :     0.000039s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.14% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.13% :     0.000023s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
38.95% :     0.007059s :      1: jit_opt_a
 1.01% :     0.000183s :      1: jit_opt_after_cconv
 0.37% :     0.000067s :      1: jit_opt_b
 3.49% :     0.000633s :      1: loop_unroll
 5.53% :     0.001001s :      1: mutable_eliminate
 5.74% :     0.001040s :     26: opt.transform.jit_opt_a
 0.30% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000029s :      4: opt.transform.jit_opt_b
 0.10% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.14% :     0.000026s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.18% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.06% :     0.000555s :      1: opt_after_jit_grad
 0.06% :     0.000010s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.22% :     0.000040s :      1: py_interpret_to_execute
 0.13% :     0.000024s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000020s :      1: remove_dup_value
 2.89% :     0.000523s :      1: renormalize.infer
 2.25% :     0.000407s :      1: renormalize.specialize
 0.07% :     0.000012s :      1: rewriter_after_jit_bprop_graph
 0.30% :     0.000054s :      1: rewriter_after_opt_a
 0.45% :     0.000082s :      1: rewriter_before_opt_a
 0.45% :     0.000082s :      1: symbol_engine_optimizer
29.87% :     0.005415s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.325.629 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd84048820> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.326.192 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0193022, [30]
[bootstrap]: 0.00019208
[type_inference]: 0.00892843
[event_method]: 3.956e-05
[auto_monad]: 8.842e-05
[graph_reusing]: 7.27002e-06
[pre_auto_parallel]: 2.35002e-06
[py_interpret_to_execute]: 2.417e-05
[rewriter_before_opt_a]: 8.599e-05
[expand_dump_flag]: 3.25e-06
[jit_opt_a]: 0.00787507, [2]
    [Cycle 1]: 0.00442493, [27]
        [switch_simplify]: 5.735e-05
        [loop_unroll]: 4.089e-05
        [a_1]: 0.00080949
        [with_stream_mark]: 2.592e-05
        [recompute_prepare]: 1.921e-05
        [updatestate_depend_eliminate]: 7.15e-06
        [updatestate_assign_eliminate]: 5.96998e-06
        [updatestate_loads_eliminate]: 5.24e-06
        [parameter_eliminate]: 1.97001e-06
        [specialize_transform]: 1.318e-05
        [updatestate_useless_node_eliminater]: 1.282e-05
        [accelerated_algorithm]: 1.227e-05
        [meta_shard_fg_expand]: 2.84999e-06
        [get_grad_eliminate_]: 1.185e-05
        [merge_forward]: 8.37998e-06
        [cell_reuse_recompute_pass]: 9.30013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.542e-05
        [j_node_and_user_rematch]: 2.218e-05
        [meta_fg_expand]: 0.00150532
        [replace_old_param]: 5.958e-05
        [inline_without_move]: 5.686e-05
        [renormalize]: 0.00145726
        [add_forward_monad_depend]: 7.4e-06
        [auto_monad_grad]: 2.24001e-06
        [auto_monad_eliminator]: 1.181e-05
        [cse]: 2.27e-05
        [replace_applicator]: 1.369e-05
    [Cycle 2]: 0.00027267, [27]
        [switch_simplify]: 4.71002e-06
        [loop_unroll]: 4.06001e-06
        [a_1]: 3.852e-05
        [with_stream_mark]: 1.034e-05
        [recompute_prepare]: 3.7e-06
        [updatestate_depend_eliminate]: 2.56e-06
        [updatestate_assign_eliminate]: 2.58998e-06
        [updatestate_loads_eliminate]: 1.42e-06
        [parameter_eliminate]: 1.24e-06
        [specialize_transform]: 3.55e-06
        [updatestate_useless_node_eliminater]: 3.77998e-06
        [accelerated_algorithm]: 3.16999e-06
        [meta_shard_fg_expand]: 1.42e-06
        [get_grad_eliminate_]: 3.09001e-06
        [merge_forward]: 2.93e-06
        [cell_reuse_recompute_pass]: 1.99e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.236e-05
        [j_node_and_user_rematch]: 6.08998e-06
        [meta_fg_expand]: 2.527e-05
        [replace_old_param]: 3.7e-06
        [inline_without_move]: 3.35e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.20999e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 4.46002e-06
        [cse]: 9.74999e-06
        [replace_applicator]: 3.58e-06
[py_interpret_to_execute_after_opt_a]: 8.22e-06
[rewriter_after_opt_a]: 3.326e-05
[convert_after_rewriter]: 7e-06
[order_py_execute_after_rewriter]: 4.84e-06
[mutable_eliminate]: 0.0005976
[jit_opt_b]: 3.845e-05, [1]
    [Cycle 1]: 3.172e-05, [2]
        [frontend_op_eliminate]: 1.053e-05
        [inline_after_opt_a]: 1.045e-05
[cconv]: 2.312e-05
[loop_unroll]: 0.00041722
[jit_opt_after_cconv]: 0.00011826, [1]
    [Cycle 1]: 0.00011225, [11]
        [c_1]: 1.139e-05
        [parameter_eliminate]: 2.59001e-06
        [updatestate_depend_eliminate]: 4.3e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.58002e-06
        [cse]: 1.662e-05
        [call_graph_tuple_transform]: 1.312e-05
        [tuple_list_get_item_eliminator]: 4.11001e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 9.29984e-07
        [switch_simplify]: 3.65e-06
[remove_dup_value]: 1.348e-05
[partial_unused_args_eliminate]: 2.22999e-06
[environ_conv]: 4.95999e-06
[add_recomputation]: 3.236e-05
[cse_after_recomputation]: 1.894e-05, [1]
    [Cycle 1]: 1.375e-05, [1]
        [cse]: 7.78001e-06
[auto_monad_reorder]: 1.206e-05
[get_jit_bprop_graph]: 1.35999e-06
[rewriter_after_jit_bprop_graph]: 4.74e-06
[opt_after_jit_grad]: 0.00047491
[symbol_engine_optimizer]: 6.251e-05, [1]
    [Cycle 1]: 5.662e-05, [6]
        [build]: 2.81e-06
        [elim_shapecalc]: 6.09001e-06
        [elim_not_effective]: 9.86e-06
        [opt_reshape]: 4.22998e-06
        [fold_const_symbol]: 5.81e-06
        [renormalize]: 5.00004e-07
[validate]: 2.835e-05
Sums
    bootstrap                            :     0.000192s :  1.24%
    type_inference                       :     0.008928s : 57.47%
    event_method                         :     0.000040s :  0.25%
    auto_monad                           :     0.000088s :  0.57%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000024s :  0.16%
    rewriter_before_opt_a                :     0.000086s :  0.55%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000062s :  0.40%
    jit_opt_a.loop_unroll                :     0.000045s :  0.29%
    jit_opt_a.a_1                        :     0.000848s :  5.46%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.23%
    jit_opt_a.recompute_prepare          :     0.000023s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000010s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000017s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000015s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.10%
    jit_opt_a.merge_forward              :     0.000011s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.24%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001531s :  9.85%
    jit_opt_a.replace_old_param          :     0.000063s :  0.41%
    jit_opt_a.inline_without_move        :     0.000060s :  0.39%
    jit_opt_a.renormalize                :     0.001457s :  9.38%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.10%
    jit_opt_a.cse                        :     0.000032s :  0.21%
    jit_opt_a.replace_applicator         :     0.000017s :  0.11%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000033s :  0.21%
    convert_after_rewriter               :     0.000007s :  0.05%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000598s :  3.85%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000010s :  0.07%
    cconv                                :     0.000023s :  0.15%
    loop_unroll                          :     0.000417s :  2.69%
    jit_opt_after_cconv.c_1              :     0.000011s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000013s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000032s :  0.21%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000012s :  0.08%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.03%
    opt_after_jit_grad                   :     0.000475s :  3.06%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000010s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000028s :  0.18%


Time group info:
------[substitution.]   0.000237    50
 0.76% :     0.000002s :      1: substitution.elim_not_effective
 0.41% :     0.000001s :      1: substitution.fold_const_symbol
 1.52% :     0.000004s :      1: substitution.graph_param_transform
63.80% :     0.000151s :      6: substitution.inline
 8.21% :     0.000019s :      2: substitution.inline_without_move
 2.72% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.36% :     0.000003s :      2: substitution.minmaximum_grad
 1.54% :     0.000004s :      3: substitution.partial_eliminate
 3.18% :     0.000008s :      9: substitution.remove_not_recompute_node
 2.30% :     0.000005s :      2: substitution.replace_applicator
 1.79% :     0.000004s :      5: substitution.replace_old_param
 1.19% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.00% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.63% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 6.60% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008868     2
89.52% :     0.007939s :      1: type_inference.infer
10.48% :     0.000929s :      1: type_inference.specialize
------[replace.]   0.000060     8
78.59% :     0.000047s :      6: replace.inline
21.41% :     0.000013s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000159     8
93.20% :     0.000148s :      6: match.inline
 6.80% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000168  1109
 1.50% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.59% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.35% :     0.000002s :     18: predicate.addn_check_dump
 1.56% :     0.000003s :     18: predicate.addn_zero_filter
 2.27% :     0.000004s :     18: predicate.arithmetic_simplify
 1.37% :     0.000002s :     18: predicate.cast_eliminate
 0.24% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.25% :     0.000002s :     18: predicate.compare_switch_simplify
 1.49% :     0.000003s :     18: predicate.depend_value_elim
 1.33% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.41% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.32% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.45% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.12% :     0.000000s :      1: predicate.elim_not_effective
 0.22% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.36% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.37% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.41% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.41% :     0.000002s :     18: predicate.environ_get_eliminate
 1.30% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.14% :     0.000002s :     12: predicate.get_grad_eliminate
 0.13% :     0.000000s :      1: predicate.graph_param_transform
 4.24% :     0.000007s :     28: predicate.inline
 3.48% :     0.000006s :     33: predicate.inline_without_move
 0.49% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :     12: predicate.less_batch_normalization
 1.54% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.69% :     0.000003s :     21: predicate.load_eliminater
 0.73% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.93% :     0.000007s :     44: predicate.loop_unroll_before_grad
 1.84% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.36% :     0.000002s :     18: predicate.merge_addn
 1.34% :     0.000002s :     18: predicate.minmaximum_grad
 0.95% :     0.000002s :      1: predicate.mutable_eliminate
 0.23% :     0.000000s :      1: predicate.opt_reshape
 2.10% :     0.000004s :     21: predicate.partial_eliminate
 1.36% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.67% :     0.000003s :     18: predicate.reduce_eliminate
 1.54% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.71% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.52% :     0.000003s :     22: predicate.replace_applicator
 1.80% :     0.000003s :     33: predicate.replace_old_param
 0.12% :     0.000000s :      1: predicate.reset_defer_inline
 1.55% :     0.000003s :     18: predicate.reshape_eliminate
 1.46% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.42% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     18: predicate.same_eliminate
 0.58% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.43% :     0.000001s :      2: predicate.special_op_eliminate
 1.21% :     0.000002s :     12: predicate.specialize_transform
 1.56% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.15% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.72% :     0.000005s :     26: predicate.switch_defer_inline
 2.38% :     0.000004s :     26: predicate.switch_layer_defer_inline
 7.13% :     0.000012s :     71: predicate.switch_simplify
 1.44% :     0.000002s :     18: predicate.tile_eliminate
 1.33% :     0.000002s :     18: predicate.transpose_eliminate
 1.58% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 3.00% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.66% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.78% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.27% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000003s :     18: predicate.value_based_eliminate
 0.14% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.26% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001022    22
74.56% :     0.000762s :     13: func_graph_cloner_run.FuncGraphClonerGraph
25.44% :     0.000260s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022000    72
 0.16% :     0.000035s :      1: add_recomputation
 0.42% :     0.000092s :      1: auto_monad
 0.07% :     0.000014s :      1: auto_monad_reorder
 0.93% :     0.000205s :      1: bootstrap
 0.12% :     0.000026s :      1: cconv
 0.04% :     0.000010s :      1: convert_after_rewriter
 0.10% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.20% :     0.000043s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
35.81% :     0.007878s :      1: jit_opt_a
 0.55% :     0.000121s :      1: jit_opt_after_cconv
 0.19% :     0.000041s :      1: jit_opt_b
 1.93% :     0.000424s :      1: loop_unroll
 2.75% :     0.000605s :      1: mutable_eliminate
 5.54% :     0.001219s :     26: opt.transform.jit_opt_a
 0.13% :     0.000029s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000015s :      4: opt.transform.jit_opt_b
 0.04% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000023s :      4: opt.transform.symbol_engine_opt
 2.19% :     0.000483s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000027s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 5.09% :     0.001119s :      1: renormalize.infer
 1.50% :     0.000330s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000036s :      1: rewriter_after_opt_a
 0.40% :     0.000089s :      1: rewriter_before_opt_a
 0.30% :     0.000065s :      1: symbol_engine_optimizer
40.63% :     0.008939s :      1: type_inference


TotalTime = 0.0327751, [33]
[bootstrap]: 0.00017922
[type_inference]: 0.00984593
[event_method]: 4.413e-05
[auto_monad]: 9.597e-05
[graph_reusing]: 7.20998e-06
[pre_auto_parallel]: 2.83e-06
[py_interpret_to_execute]: 2.522e-05
[rewriter_before_opt_a]: 9.072e-05
[expand_dump_flag]: 3.25e-06
[jit_opt_a]: 0.0114015, [3]
    [Cycle 1]: 0.00551565, [27]
        [switch_simplify]: 6.268e-05
        [loop_unroll]: 3.791e-05
        [a_1]: 0.0008862
        [with_stream_mark]: 2.598e-05
        [recompute_prepare]: 1.84e-05
        [updatestate_depend_eliminate]: 8e-06
        [updatestate_assign_eliminate]: 6.23e-06
        [updatestate_loads_eliminate]: 5.57999e-06
        [parameter_eliminate]: 2.54999e-06
        [specialize_transform]: 1.266e-05
        [updatestate_useless_node_eliminater]: 1.182e-05
        [accelerated_algorithm]: 1.241e-05
        [meta_shard_fg_expand]: 3.09001e-06
        [get_grad_eliminate_]: 1.157e-05
        [merge_forward]: 8.21002e-06
        [cell_reuse_recompute_pass]: 1.05999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.615e-05
        [j_node_and_user_rematch]: 2.176e-05
        [meta_fg_expand]: 0.00133244
        [replace_old_param]: 4.956e-05
        [inline_without_move]: 5.029e-05
        [renormalize]: 0.00244049
        [add_forward_monad_depend]: 1.045e-05
        [auto_monad_grad]: 5.97999e-06
        [auto_monad_eliminator]: 5.016e-05
        [cse]: 0.00014895
        [replace_applicator]: 6.717e-05
    [Cycle 2]: 0.00215762, [27]
        [switch_simplify]: 3.679e-05
        [loop_unroll]: 3.631e-05
        [a_1]: 0.00110298
        [with_stream_mark]: 1.309e-05
        [recompute_prepare]: 1.046e-05
        [updatestate_depend_eliminate]: 5.19e-06
        [updatestate_assign_eliminate]: 4.57e-06
        [updatestate_loads_eliminate]: 4.13001e-06
        [parameter_eliminate]: 1.57999e-06
        [specialize_transform]: 8.72e-06
        [updatestate_useless_node_eliminater]: 8.12e-06
        [accelerated_algorithm]: 8.13999e-06
        [meta_shard_fg_expand]: 2.77002e-06
        [get_grad_eliminate_]: 7.97e-06
        [merge_forward]: 5.09998e-06
        [cell_reuse_recompute_pass]: 9.89996e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.822e-05
        [j_node_and_user_rematch]: 1.414e-05
        [meta_fg_expand]: 4.09e-05
        [replace_old_param]: 1.238e-05
        [inline_without_move]: 8.44998e-06
        [renormalize]: 0.00057073
        [add_forward_monad_depend]: 5.41998e-06
        [auto_monad_grad]: 1.64e-06
        [auto_monad_eliminator]: 1.558e-05
        [cse]: 4.772e-05
        [replace_applicator]: 1.707e-05
    [Cycle 3]: 0.00048933, [27]
        [switch_simplify]: 9.41998e-06
        [loop_unroll]: 8.33999e-06
        [a_1]: 0.00019129
        [with_stream_mark]: 1.203e-05
        [recompute_prepare]: 8.03001e-06
        [updatestate_depend_eliminate]: 4.36002e-06
        [updatestate_assign_eliminate]: 4.47e-06
        [updatestate_loads_eliminate]: 3.73001e-06
        [parameter_eliminate]: 1.29998e-06
        [specialize_transform]: 8.15e-06
        [updatestate_useless_node_eliminater]: 7.78001e-06
        [accelerated_algorithm]: 8.62998e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 7.73001e-06
        [merge_forward]: 4.33001e-06
        [cell_reuse_recompute_pass]: 2.41e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.634e-05
        [j_node_and_user_rematch]: 1.331e-05
        [meta_fg_expand]: 2.91e-06
        [replace_old_param]: 9.88002e-06
        [inline_without_move]: 7.75998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.31002e-06
        [auto_monad_grad]: 1.06002e-06
        [auto_monad_eliminator]: 1.005e-05
        [cse]: 2.234e-05
        [replace_applicator]: 7.98999e-06
[py_interpret_to_execute_after_opt_a]: 1.27e-05
[rewriter_after_opt_a]: 5.027e-05
[convert_after_rewriter]: 9.57999e-06
[order_py_execute_after_rewriter]: 7.08e-06
[mutable_eliminate]: 0.00070584
[jit_opt_b]: 6.739e-05, [1]
    [Cycle 1]: 5.944e-05, [2]
        [frontend_op_eliminate]: 2.469e-05
        [inline_after_opt_a]: 2.347e-05
[cconv]: 2.581e-05
[loop_unroll]: 0.0004039
[jit_opt_after_cconv]: 0.00019479, [1]
    [Cycle 1]: 0.00018726, [11]
        [c_1]: 3.446e-05
        [parameter_eliminate]: 2.71e-06
        [updatestate_depend_eliminate]: 8.03001e-06
        [updatestate_assign_eliminate]: 4.44998e-06
        [updatestate_loads_eliminate]: 3.91999e-06
        [cse]: 3.568e-05
        [call_graph_tuple_transform]: 2.646e-05
        [tuple_list_get_item_eliminator]: 8.47998e-06
        [none_parameter_eliminate]: 1.72999e-06
        [renormalize]: 6.39993e-07
        [switch_simplify]: 8.68001e-06
[remove_dup_value]: 2.21e-05
[partial_unused_args_eliminate]: 2.24999e-06
[environ_conv]: 7.33e-06
[add_recomputation]: 5.914e-05
[cse_after_recomputation]: 3.482e-05, [1]
    [Cycle 1]: 2.889e-05, [1]
        [cse]: 2.243e-05
[auto_monad_reorder]: 2.236e-05
[get_jit_bprop_graph]: 1.90001e-06
[rewriter_after_jit_bprop_graph]: 6.39001e-06
[opt_after_jit_grad]: 0.00046915
[symbol_engine_optimizer]: 9.28e-05, [1]
    [Cycle 1]: 8.612e-05, [6]
        [build]: 4.46002e-06
        [elim_shapecalc]: 1.146e-05
        [elim_not_effective]: 1.795e-05
        [opt_reshape]: 8.72998e-06
        [fold_const_symbol]: 1.491e-05
        [renormalize]: 6.69999e-07
[validate]: 4.566e-05
[backend_pass]: 9.00007e-07
[task_emit]: 0.00862847
[execute]: 7.46999e-06
Sums
    bootstrap                            :     0.000179s :  0.62%
    type_inference                       :     0.009846s : 34.25%
    event_method                         :     0.000044s :  0.15%
    auto_monad                           :     0.000096s :  0.33%
    graph_reusing                        :     0.000007s :  0.03%
    pre_auto_parallel                    :     0.000003s :  0.01%
    py_interpret_to_execute              :     0.000025s :  0.09%
    rewriter_before_opt_a                :     0.000091s :  0.32%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000109s :  0.38%
    jit_opt_a.loop_unroll                :     0.000083s :  0.29%
    jit_opt_a.a_1                        :     0.002180s :  7.58%
    jit_opt_a.with_stream_mark           :     0.000051s :  0.18%
    jit_opt_a.recompute_prepare          :     0.000037s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000018s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000013s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000005s :  0.02%
    jit_opt_a.specialize_transform       :     0.000030s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000028s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000029s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000008s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000027s :  0.09%
    jit_opt_a.merge_forward              :     0.000018s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000061s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000049s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001376s :  4.79%
    jit_opt_a.replace_old_param          :     0.000072s :  0.25%
    jit_opt_a.inline_without_move        :     0.000066s :  0.23%
    jit_opt_a.renormalize                :     0.003011s : 10.47%
    jit_opt_a.add_forward_monad_depend   :     0.000017s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000009s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000076s :  0.26%
    jit_opt_a.cse                        :     0.000219s :  0.76%
    jit_opt_a.replace_applicator         :     0.000092s :  0.32%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.04%
    rewriter_after_opt_a                 :     0.000050s :  0.17%
    convert_after_rewriter               :     0.000010s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000706s :  2.46%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000023s :  0.08%
    cconv                                :     0.000026s :  0.09%
    loop_unroll                          :     0.000404s :  1.40%
    jit_opt_after_cconv.c_1              :     0.000034s :  0.12%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000036s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000008s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.03%
    remove_dup_value                     :     0.000022s :  0.08%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000007s :  0.03%
    add_recomputation                    :     0.000059s :  0.21%
    cse_after_recomputation.cse          :     0.000022s :  0.08%
    auto_monad_reorder                   :     0.000022s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000469s :  1.63%
    symbol_engine_optimizer.build        :     0.000004s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000018s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000015s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000046s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008628s : 30.01%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000527   126
 7.65% :     0.000040s :      2: substitution.arithmetic_simplify
 0.58% :     0.000003s :      5: substitution.elim_not_effective
 0.46% :     0.000002s :      5: substitution.fold_const_symbol
 1.40% :     0.000007s :      6: substitution.graph_param_transform
57.67% :     0.000304s :     12: substitution.inline
 3.05% :     0.000016s :      2: substitution.inline_without_move
 1.88% :     0.000010s :     18: substitution.j_node_and_user_rematch
 1.97% :     0.000010s :      7: substitution.minmaximum_grad
 1.32% :     0.000007s :      5: substitution.partial_eliminate
 2.47% :     0.000013s :     18: substitution.remove_not_recompute_node
 4.83% :     0.000025s :     10: substitution.replace_applicator
 1.40% :     0.000007s :      7: substitution.replace_old_param
 0.59% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 4.19% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.85% :     0.000015s :      7: substitution.tuple_list_get_item_depend_reorder
 7.68% :     0.000040s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009765     2
89.19% :     0.008710s :      1: type_inference.infer
10.81% :     0.001055s :      1: type_inference.specialize
------[replace.]   0.000140    19
65.33% :     0.000092s :     12: replace.inline
34.67% :     0.000049s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000318    19
93.49% :     0.000297s :     12: match.inline
 6.51% :     0.000021s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000458  2895
 1.46% :     0.000007s :     48: predicate.accumulaten_eliminater
 0.38% :     0.000002s :      6: predicate.ad_related_special_op_eliminate
13.99% :     0.000064s :     48: predicate.addn_check_dump
 1.52% :     0.000007s :     48: predicate.addn_zero_filter
 1.75% :     0.000008s :     48: predicate.arithmetic_simplify
 1.41% :     0.000006s :     48: predicate.cast_eliminate
 0.22% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.24% :     0.000006s :     48: predicate.compare_switch_simplify
 1.41% :     0.000006s :     48: predicate.depend_value_elim
 1.24% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.29% :     0.000006s :     48: predicate.dict_get_item_eliminator
 1.30% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.29% :     0.000001s :      6: predicate.dumpgradient_eliminate
 0.11% :     0.000001s :      6: predicate.elim_not_effective
 0.26% :     0.000001s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.32% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.25% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.27% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.35% :     0.000006s :     48: predicate.environ_get_eliminate
 1.28% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.09% :     0.000000s :      6: predicate.fold_const_symbol
 0.80% :     0.000004s :     23: predicate.get_grad_eliminate
 0.11% :     0.000001s :      6: predicate.graph_param_transform
 3.86% :     0.000018s :     79: predicate.inline
 1.51% :     0.000007s :     44: predicate.inline_without_move
 0.33% :     0.000002s :     23: predicate.j_node_and_user_rematch
 0.90% :     0.000004s :     23: predicate.less_batch_normalization
 1.68% :     0.000008s :     55: predicate.list_to_tuple_eliminator_
 1.76% :     0.000008s :     61: predicate.load_eliminater
 0.48% :     0.000002s :      6: predicate.loop_unroll_after_grad
 2.65% :     0.000012s :     90: predicate.loop_unroll_before_grad
 1.53% :     0.000007s :     54: predicate.make_slice_get_slice_eliminator
 1.28% :     0.000006s :     48: predicate.merge_addn
 1.31% :     0.000006s :     48: predicate.minmaximum_grad
 0.54% :     0.000002s :      6: predicate.mutable_eliminate
 0.23% :     0.000001s :      6: predicate.opt_reshape
 2.06% :     0.000009s :     61: predicate.partial_eliminate
 1.28% :     0.000006s :     48: predicate.print_const_string_wrapper
 1.61% :     0.000007s :     48: predicate.reduce_eliminate
 1.61% :     0.000007s :     55: predicate.redundant_stop_gradient_eliminater
 0.40% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.27% :     0.000010s :    108: predicate.replace_applicator
 0.83% :     0.000004s :     44: predicate.replace_old_param
 0.17% :     0.000001s :      6: predicate.reset_defer_inline
 1.45% :     0.000007s :     48: predicate.reshape_eliminate
 1.30% :     0.000006s :     48: predicate.row_tensor_add_zeros_like
 0.33% :     0.000002s :      6: predicate.row_tensor_eliminate
 1.33% :     0.000006s :     48: predicate.same_eliminate
 0.41% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.46% :     0.000002s :     12: predicate.special_op_eliminate
 0.85% :     0.000004s :     23: predicate.specialize_transform
 1.42% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.30% :     0.000006s :     48: predicate.stack_unstack_eliminate
 0.20% :     0.000001s :      6: predicate.switch_call_monad_eliminater
 2.23% :     0.000010s :     67: predicate.switch_defer_inline
 2.10% :     0.000010s :     67: predicate.switch_layer_defer_inline
 5.18% :     0.000024s :    163: predicate.switch_simplify
 1.27% :     0.000006s :     48: predicate.tile_eliminate
 1.31% :     0.000006s :     48: predicate.transpose_eliminate
 1.56% :     0.000007s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.39% :     0.000006s :     48: predicate.tuple_list_get_item_depend_reorder
 2.83% :     0.000013s :     67: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000007s :     48: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000007s :     55: predicate.tuple_to_list_eliminator_
 1.72% :     0.000008s :     61: predicate.updatestate_pure_node_eliminater
 2.60% :     0.000012s :     84: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000007s :     48: predicate.value_based_eliminate
 0.20% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.26% :     0.000001s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001346    29
58.73% :     0.000790s :     13: func_graph_cloner_run.FuncGraphClonerGraph
41.27% :     0.000555s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.038784    91
 0.16% :     0.000062s :      1: add_recomputation
 0.26% :     0.000100s :      1: auto_monad
 0.06% :     0.000025s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.49% :     0.000191s :      1: bootstrap
 0.07% :     0.000028s :      1: cconv
 0.03% :     0.000012s :      1: convert_after_rewriter
 0.10% :     0.000037s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.13% :     0.000049s :      1: event_method
 0.03% :     0.000011s :      1: execute
 0.01% :     0.000005s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.03% :     0.000010s :      1: graph_reusing
29.41% :     0.011405s :      1: jit_opt_a
 0.51% :     0.000198s :      1: jit_opt_after_cconv
 0.18% :     0.000070s :      1: jit_opt_b
 1.06% :     0.000411s :      1: loop_unroll
 1.84% :     0.000715s :      1: mutable_eliminate
 7.27% :     0.002820s :     39: opt.transform.jit_opt_a
 0.19% :     0.000075s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000042s :      4: opt.transform.jit_opt_b
 0.04% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.08% :     0.000030s :      1: opt.transform.opt_after_jit_grad
 0.13% :     0.000050s :      4: opt.transform.symbol_engine_opt
 1.23% :     0.000476s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000005s :      1: pre_auto_parallel
 0.07% :     0.000028s :      1: py_interpret_to_execute
 0.04% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000025s :      1: remove_dup_value
 4.22% :     0.001636s :      2: renormalize.infer
 3.51% :     0.001361s :      2: renormalize.specialize
 0.02% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000053s :      1: rewriter_after_opt_a
 0.24% :     0.000093s :      1: rewriter_before_opt_a
 0.25% :     0.000095s :      1: symbol_engine_optimizer
22.28% :     0.008639s :      1: task_emit
25.43% :     0.009864s :      1: type_inference
 0.19% :     0.000072s :      1: validate

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.351.946 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d1040> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.352.573 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0145361, [30]
[bootstrap]: 0.00034184
[type_inference]: 0.00509947
[event_method]: 1.9e-05
[auto_monad]: 7.137e-05
[graph_reusing]: 5.83002e-06
[pre_auto_parallel]: 3.01001e-06
[py_interpret_to_execute]: 3.609e-05
[rewriter_before_opt_a]: 7.944e-05
[expand_dump_flag]: 4.28001e-06
[jit_opt_a]: 0.00616925, [2]
    [Cycle 1]: 0.00196661, [27]
        [switch_simplify]: 6.488e-05
        [loop_unroll]: 3.12e-05
        [a_1]: 0.00061007
        [with_stream_mark]: 2.801e-05
        [recompute_prepare]: 9.34998e-06
        [updatestate_depend_eliminate]: 4.50999e-06
        [updatestate_assign_eliminate]: 3.23998e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 2.36e-06
        [specialize_transform]: 6.44001e-06
        [updatestate_useless_node_eliminater]: 5.36002e-06
        [accelerated_algorithm]: 5.78997e-06
        [meta_shard_fg_expand]: 2.88e-06
        [get_grad_eliminate_]: 5.09e-06
        [merge_forward]: 4.07e-06
        [cell_reuse_recompute_pass]: 1.84998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.51e-05
        [j_node_and_user_rematch]: 9.22001e-06
        [meta_fg_expand]: 2.47001e-06
        [replace_old_param]: 9.25001e-06
        [inline_without_move]: 5.51998e-06
        [renormalize]: 0.00082921
        [add_forward_monad_depend]: 8.38001e-06
        [auto_monad_grad]: 2.38002e-06
        [auto_monad_eliminator]: 1.881e-05
        [cse]: 3.673e-05
        [replace_applicator]: 1.482e-05
    [Cycle 2]: 0.00035479, [27]
        [switch_simplify]: 5.72999e-06
        [loop_unroll]: 5.00001e-06
        [a_1]: 9.217e-05
        [with_stream_mark]: 1.513e-05
        [recompute_prepare]: 5.41998e-06
        [updatestate_depend_eliminate]: 3.41999e-06
        [updatestate_assign_eliminate]: 2.73998e-06
        [updatestate_loads_eliminate]: 1.96998e-06
        [parameter_eliminate]: 8.89995e-07
        [specialize_transform]: 5.20001e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 1.79e-06
        [get_grad_eliminate_]: 4.65999e-06
        [merge_forward]: 3.36999e-06
        [cell_reuse_recompute_pass]: 1.82001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.612e-05
        [j_node_and_user_rematch]: 8.70999e-06
        [meta_fg_expand]: 1.74e-06
        [replace_old_param]: 8.07e-06
        [inline_without_move]: 4.60001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.92999e-06
        [auto_monad_grad]: 8.99978e-07
        [auto_monad_eliminator]: 8.23001e-06
        [cse]: 1.266e-05
        [replace_applicator]: 5.91003e-06
[py_interpret_to_execute_after_opt_a]: 1.571e-05
[rewriter_after_opt_a]: 4.102e-05
[convert_after_rewriter]: 7.73999e-06
[order_py_execute_after_rewriter]: 6.09999e-06
[mutable_eliminate]: 0.00076981
[jit_opt_b]: 5.623e-05, [1]
    [Cycle 1]: 4.657e-05, [2]
        [frontend_op_eliminate]: 1.799e-05
        [inline_after_opt_a]: 1.605e-05
[cconv]: 2.937e-05
[loop_unroll]: 0.00056098
[jit_opt_after_cconv]: 0.00016361, [1]
    [Cycle 1]: 0.0001552, [11]
        [c_1]: 2.303e-05
        [parameter_eliminate]: 3.84002e-06
        [updatestate_depend_eliminate]: 9.56998e-06
        [updatestate_assign_eliminate]: 2.69001e-06
        [updatestate_loads_eliminate]: 2.26e-06
        [cse]: 2.522e-05
        [call_graph_tuple_transform]: 2.196e-05
        [tuple_list_get_item_eliminator]: 5.30001e-06
        [none_parameter_eliminate]: 1.77001e-06
        [renormalize]: 1.09998e-06
        [switch_simplify]: 5.84e-06
[remove_dup_value]: 1.599e-05
[partial_unused_args_eliminate]: 2.19001e-06
[environ_conv]: 6.43e-06
[add_recomputation]: 5.515e-05
[cse_after_recomputation]: 2.51e-05, [1]
    [Cycle 1]: 1.805e-05, [1]
        [cse]: 1.02e-05
[auto_monad_reorder]: 1.781e-05
[get_jit_bprop_graph]: 2.46e-06
[rewriter_after_jit_bprop_graph]: 7.92e-06
[opt_after_jit_grad]: 0.00057431
[symbol_engine_optimizer]: 8.653e-05, [1]
    [Cycle 1]: 7.79e-05, [6]
        [build]: 4.58999e-06
        [elim_shapecalc]: 9.59999e-06
        [elim_not_effective]: 1.586e-05
        [opt_reshape]: 5.69e-06
        [fold_const_symbol]: 8.89e-06
        [renormalize]: 4.39992e-07
[validate]: 4.548e-05
Sums
    bootstrap                            :     0.000342s :  3.43%
    type_inference                       :     0.005099s : 51.10%
    event_method                         :     0.000019s :  0.19%
    auto_monad                           :     0.000071s :  0.72%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.36%
    rewriter_before_opt_a                :     0.000079s :  0.80%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000071s :  0.71%
    jit_opt_a.loop_unroll                :     0.000036s :  0.36%
    jit_opt_a.a_1                        :     0.000702s :  7.04%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.43%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.17%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000829s :  8.31%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000027s :  0.27%
    jit_opt_a.cse                        :     0.000049s :  0.49%
    jit_opt_a.replace_applicator         :     0.000021s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000016s :  0.16%
    rewriter_after_opt_a                 :     0.000041s :  0.41%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000770s :  7.71%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.16%
    cconv                                :     0.000029s :  0.29%
    loop_unroll                          :     0.000561s :  5.62%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000010s :  0.10%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000025s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000016s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000055s :  0.55%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000018s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.08%
    opt_after_jit_grad                   :     0.000574s :  5.76%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000010s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000045s :  0.46%


Time group info:
------[substitution.]   0.000255    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000002s :      2: substitution.fold_const_symbol
 2.40% :     0.000006s :      3: substitution.graph_param_transform
85.11% :     0.000217s :      4: substitution.inline
 1.36% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.08% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.84% :     0.000005s :      2: substitution.replace_old_param
 5.78% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005021     2
81.49% :     0.004092s :      1: type_inference.infer
18.51% :     0.000929s :      1: type_inference.specialize
------[replace.]   0.000078     6
64.44% :     0.000050s :      4: replace.inline
35.56% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000227     6
94.28% :     0.000214s :      4: match.inline
 5.72% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000134   691
 1.17% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.55% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.11% :     0.000001s :     10: predicate.addn_zero_filter
 2.77% :     0.000004s :     10: predicate.arithmetic_simplify
 1.25% :     0.000002s :     10: predicate.cast_eliminate
 0.62% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 0.91% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.04% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.43% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000001s :      3: predicate.elim_not_effective
 0.69% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.93% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.04% :     0.000001s :     10: predicate.environ_get_eliminate
 0.95% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 0.81% :     0.000001s :      6: predicate.get_grad_eliminate
 0.36% :     0.000000s :      3: predicate.graph_param_transform
 5.72% :     0.000008s :     22: predicate.inline
 0.85% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000002s :      6: predicate.less_batch_normalization
 1.81% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.58% :     0.000002s :     15: predicate.load_eliminater
 2.84% :     0.000004s :      3: predicate.loop_unroll_after_grad
 3.87% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.33% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.89% :     0.000003s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.42% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.89% :     0.000003s :     10: predicate.reduce_eliminate
 1.28% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.73% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.57% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.42% :     0.000001s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.01% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.07% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000002s :     10: predicate.same_eliminate
 0.61% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.26% :     0.000002s :      6: predicate.special_op_eliminate
 0.92% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.53% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.51% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 1.86% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.57% :     0.000010s :     53: predicate.switch_simplify
 0.99% :     0.000001s :     10: predicate.tile_eliminate
 1.04% :     0.000001s :     10: predicate.transpose_eliminate
 1.24% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.04% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.21% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.41% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.55% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.69% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000668    12
50.76% :     0.000339s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.78% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.46% :     0.000277s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016356    72
 0.36% :     0.000059s :      1: add_recomputation
 0.46% :     0.000075s :      1: auto_monad
 0.13% :     0.000022s :      1: auto_monad_reorder
 2.23% :     0.000366s :      1: bootstrap
 0.20% :     0.000032s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000028s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.04% :     0.000007s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
37.75% :     0.006174s :      1: jit_opt_a
 1.03% :     0.000168s :      1: jit_opt_after_cconv
 0.36% :     0.000059s :      1: jit_opt_b
 3.50% :     0.000573s :      1: loop_unroll
 4.79% :     0.000783s :      1: mutable_eliminate
 5.71% :     0.000933s :     26: opt.transform.jit_opt_a
 0.32% :     0.000052s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000027s :      4: opt.transform.jit_opt_b
 0.11% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.58% :     0.000586s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000039s :      1: py_interpret_to_execute
 0.11% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.84% :     0.000464s :      1: renormalize.infer
 2.17% :     0.000354s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000044s :      1: rewriter_after_opt_a
 0.51% :     0.000083s :      1: rewriter_before_opt_a
 0.54% :     0.000089s :      1: symbol_engine_optimizer
31.31% :     0.005120s :      1: type_inference

[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.358.317 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80f45e0> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.358.912 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0209492, [30]
[bootstrap]: 0.00022966
[type_inference]: 0.0104183
[event_method]: 4.253e-05
[auto_monad]: 9.377e-05
[graph_reusing]: 7.07002e-06
[pre_auto_parallel]: 3.43e-06
[py_interpret_to_execute]: 2.572e-05
[rewriter_before_opt_a]: 9.921e-05
[expand_dump_flag]: 3.40998e-06
[jit_opt_a]: 0.00805051, [2]
    [Cycle 1]: 0.00434375, [27]
        [switch_simplify]: 6.335e-05
        [loop_unroll]: 3.849e-05
        [a_1]: 0.00080967
        [with_stream_mark]: 2.7e-05
        [recompute_prepare]: 1.792e-05
        [updatestate_depend_eliminate]: 8.47e-06
        [updatestate_assign_eliminate]: 6.81001e-06
        [updatestate_loads_eliminate]: 5.91998e-06
        [parameter_eliminate]: 2.76e-06
        [specialize_transform]: 1.257e-05
        [updatestate_useless_node_eliminater]: 1.157e-05
        [accelerated_algorithm]: 1.205e-05
        [meta_shard_fg_expand]: 3.63999e-06
        [get_grad_eliminate_]: 1.187e-05
        [merge_forward]: 8.07e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 2.617e-05
        [j_node_and_user_rematch]: 2.201e-05
        [meta_fg_expand]: 0.00142164
        [replace_old_param]: 5.149e-05
        [inline_without_move]: 5.012e-05
        [renormalize]: 0.00146702
        [add_forward_monad_depend]: 5.62001e-06
        [auto_monad_grad]: 2.40002e-06
        [auto_monad_eliminator]: 1.189e-05
        [cse]: 2.656e-05
        [replace_applicator]: 1.109e-05
    [Cycle 2]: 0.0002666, [27]
        [switch_simplify]: 4.37998e-06
        [loop_unroll]: 3.88999e-06
        [a_1]: 3.827e-05
        [with_stream_mark]: 9.71998e-06
        [recompute_prepare]: 3.76001e-06
        [updatestate_depend_eliminate]: 2.49001e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.71998e-06
        [parameter_eliminate]: 1.47001e-06
        [specialize_transform]: 3.56001e-06
        [updatestate_useless_node_eliminater]: 3.38e-06
        [accelerated_algorithm]: 3.58e-06
        [meta_shard_fg_expand]: 1.25001e-06
        [get_grad_eliminate_]: 3.36999e-06
        [merge_forward]: 2.53e-06
        [cell_reuse_recompute_pass]: 1.36002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.168e-05
        [j_node_and_user_rematch]: 5.99e-06
        [meta_fg_expand]: 2.279e-05
        [replace_old_param]: 3.63e-06
        [inline_without_move]: 3.27002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.49e-06
        [auto_monad_grad]: 8.49977e-07
        [auto_monad_eliminator]: 4.35999e-06
        [cse]: 1.04e-05
        [replace_applicator]: 3.74002e-06
[py_interpret_to_execute_after_opt_a]: 9.74e-06
[rewriter_after_opt_a]: 3.455e-05
[convert_after_rewriter]: 6.41998e-06
[order_py_execute_after_rewriter]: 4.60999e-06
[mutable_eliminate]: 0.00051838
[jit_opt_b]: 3.849e-05, [1]
    [Cycle 1]: 3.127e-05, [2]
        [frontend_op_eliminate]: 9.80002e-06
        [inline_after_opt_a]: 1.088e-05
[cconv]: 2.187e-05
[loop_unroll]: 0.00040203
[jit_opt_after_cconv]: 0.0001145, [1]
    [Cycle 1]: 0.00010839, [11]
        [c_1]: 1.158e-05
        [parameter_eliminate]: 2.13002e-06
        [updatestate_depend_eliminate]: 4.52e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.60001e-06
        [cse]: 1.474e-05
        [call_graph_tuple_transform]: 1.308e-05
        [tuple_list_get_item_eliminator]: 3.91999e-06
        [none_parameter_eliminate]: 1.39003e-06
        [renormalize]: 4.69998e-07
        [switch_simplify]: 3.66001e-06
[remove_dup_value]: 1.372e-05
[partial_unused_args_eliminate]: 2.61e-06
[environ_conv]: 4.65001e-06
[add_recomputation]: 3.202e-05
[cse_after_recomputation]: 1.943e-05, [1]
    [Cycle 1]: 1.385e-05, [1]
        [cse]: 7.93999e-06
[auto_monad_reorder]: 1.223e-05
[get_jit_bprop_graph]: 1.25999e-06
[rewriter_after_jit_bprop_graph]: 4.31002e-06
[opt_after_jit_grad]: 0.00048203
[symbol_engine_optimizer]: 6.189e-05, [1]
    [Cycle 1]: 5.56e-05, [6]
        [build]: 2.84999e-06
        [elim_shapecalc]: 6.34001e-06
        [elim_not_effective]: 8.42998e-06
        [opt_reshape]: 4.03999e-06
        [fold_const_symbol]: 6.07999e-06
        [renormalize]: 4.50003e-07
[validate]: 2.625e-05
Sums
    bootstrap                            :     0.000230s :  1.36%
    type_inference                       :     0.010418s : 61.62%
    event_method                         :     0.000043s :  0.25%
    auto_monad                           :     0.000094s :  0.55%
    graph_reusing                        :     0.000007s :  0.04%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000026s :  0.15%
    rewriter_before_opt_a                :     0.000099s :  0.59%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000068s :  0.40%
    jit_opt_a.loop_unroll                :     0.000042s :  0.25%
    jit_opt_a.a_1                        :     0.000848s :  5.01%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000022s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000011s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000009s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000008s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000016s :  0.10%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.09%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.09%
    jit_opt_a.merge_forward              :     0.000011s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000038s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.001444s :  8.54%
    jit_opt_a.replace_old_param          :     0.000055s :  0.33%
    jit_opt_a.inline_without_move        :     0.000053s :  0.32%
    jit_opt_a.renormalize                :     0.001467s :  8.68%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.10%
    jit_opt_a.cse                        :     0.000037s :  0.22%
    jit_opt_a.replace_applicator         :     0.000015s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.06%
    rewriter_after_opt_a                 :     0.000035s :  0.20%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000518s :  3.07%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.06%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.06%
    cconv                                :     0.000022s :  0.13%
    loop_unroll                          :     0.000402s :  2.38%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.07%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000015s :  0.09%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000014s :  0.08%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000032s :  0.19%
    cse_after_recomputation.cse          :     0.000008s :  0.05%
    auto_monad_reorder                   :     0.000012s :  0.07%
    get_jit_bprop_graph                  :     0.000001s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000482s :  2.85%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000008s :  0.05%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.02%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000026s :  0.16%


Time group info:
------[substitution.]   0.000234    50
 0.58% :     0.000001s :      1: substitution.elim_not_effective
 0.52% :     0.000001s :      1: substitution.fold_const_symbol
 1.68% :     0.000004s :      1: substitution.graph_param_transform
63.49% :     0.000148s :      6: substitution.inline
 6.70% :     0.000016s :      2: substitution.inline_without_move
 2.83% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.86% :     0.000004s :      2: substitution.minmaximum_grad
 1.85% :     0.000004s :      3: substitution.partial_eliminate
 3.15% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.21% :     0.000005s :      2: substitution.replace_applicator
 1.76% :     0.000004s :      5: substitution.replace_old_param
 1.25% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.21% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 2.12% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 6.77% :     0.000016s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.010341     2
88.77% :     0.009179s :      1: type_inference.infer
11.23% :     0.001162s :      1: type_inference.specialize
------[replace.]   0.000060     8
76.76% :     0.000046s :      6: replace.inline
23.24% :     0.000014s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000155     8
92.92% :     0.000144s :      6: match.inline
 7.08% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000163  1109
 1.66% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.54% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.35% :     0.000002s :     18: predicate.addn_check_dump
 1.64% :     0.000003s :     18: predicate.addn_zero_filter
 2.01% :     0.000003s :     18: predicate.arithmetic_simplify
 1.51% :     0.000002s :     18: predicate.cast_eliminate
 0.16% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.23% :     0.000002s :     18: predicate.compare_switch_simplify
 1.42% :     0.000002s :     18: predicate.depend_value_elim
 1.28% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.43% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.44% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.36% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.13% :     0.000000s :      1: predicate.elim_not_effective
 0.63% :     0.000001s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.36% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.35% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.43% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.38% :     0.000002s :     18: predicate.environ_get_eliminate
 1.27% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.06% :     0.000000s :      1: predicate.fold_const_symbol
 1.18% :     0.000002s :     12: predicate.get_grad_eliminate
 0.20% :     0.000000s :      1: predicate.graph_param_transform
 4.36% :     0.000007s :     28: predicate.inline
 3.38% :     0.000006s :     33: predicate.inline_without_move
 0.53% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.44% :     0.000002s :     12: predicate.less_batch_normalization
 1.57% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.68% :     0.000003s :     21: predicate.load_eliminater
 0.67% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.52% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.70% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.32% :     0.000002s :     18: predicate.merge_addn
 1.34% :     0.000002s :     18: predicate.minmaximum_grad
 0.81% :     0.000001s :      1: predicate.mutable_eliminate
 0.22% :     0.000000s :      1: predicate.opt_reshape
 2.22% :     0.000004s :     21: predicate.partial_eliminate
 1.41% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.87% :     0.000003s :     18: predicate.reduce_eliminate
 1.57% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.71% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.57% :     0.000003s :     22: predicate.replace_applicator
 1.67% :     0.000003s :     33: predicate.replace_old_param
 0.12% :     0.000000s :      1: predicate.reset_defer_inline
 1.64% :     0.000003s :     18: predicate.reshape_eliminate
 1.45% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.41% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.39% :     0.000002s :     18: predicate.same_eliminate
 0.64% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.31% :     0.000000s :      2: predicate.special_op_eliminate
 1.20% :     0.000002s :     12: predicate.specialize_transform
 1.55% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.13% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.66% :     0.000004s :     26: predicate.switch_defer_inline
 2.43% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.94% :     0.000011s :     71: predicate.switch_simplify
 1.41% :     0.000002s :     18: predicate.tile_eliminate
 1.36% :     0.000002s :     18: predicate.transpose_eliminate
 1.75% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.49% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 3.31% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.52% :     0.000002s :     20: predicate.tuple_to_list_eliminator_
 1.57% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.10% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.67% :     0.000003s :     18: predicate.value_based_eliminate
 0.12% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.32% :     0.000001s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001173    22
73.97% :     0.000868s :     13: func_graph_cloner_run.FuncGraphClonerGraph
26.03% :     0.000305s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.023637    72
 0.15% :     0.000035s :      1: add_recomputation
 0.41% :     0.000098s :      1: auto_monad
 0.06% :     0.000015s :      1: auto_monad_reorder
 1.04% :     0.000246s :      1: bootstrap
 0.10% :     0.000025s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.09% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.20% :     0.000047s :      1: event_method
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000003s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
34.07% :     0.008054s :      1: jit_opt_a
 0.50% :     0.000117s :      1: jit_opt_after_cconv
 0.17% :     0.000041s :      1: jit_opt_b
 1.73% :     0.000409s :      1: loop_unroll
 2.23% :     0.000527s :      1: mutable_eliminate
 5.08% :     0.001200s :     26: opt.transform.jit_opt_a
 0.12% :     0.000029s :      4: opt.transform.jit_opt_after_cconv
 0.06% :     0.000014s :      4: opt.transform.jit_opt_b
 0.04% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000013s :      1: opt.transform.opt_after_jit_grad
 0.09% :     0.000022s :      4: opt.transform.symbol_engine_opt
 2.07% :     0.000490s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000005s :      1: partial_unused_args_eliminate
 0.02% :     0.000006s :      1: pre_auto_parallel
 0.12% :     0.000028s :      1: py_interpret_to_execute
 0.05% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000016s :      1: remove_dup_value
 4.69% :     0.001110s :      1: renormalize.infer
 1.48% :     0.000349s :      1: renormalize.specialize
 0.03% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000038s :      1: rewriter_after_opt_a
 0.43% :     0.000102s :      1: rewriter_before_opt_a
 0.27% :     0.000064s :      1: symbol_engine_optimizer
44.15% :     0.010436s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.379.573 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d11f0> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.380.123 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0151913, [30]
[bootstrap]: 0.00041181
[type_inference]: 0.00552083
[event_method]: 2.024e-05
[auto_monad]: 6.826e-05
[graph_reusing]: 6.48003e-06
[pre_auto_parallel]: 3.03e-06
[py_interpret_to_execute]: 4.002e-05
[rewriter_before_opt_a]: 8.112e-05
[expand_dump_flag]: 3.19001e-06
[jit_opt_a]: 0.00636176, [2]
    [Cycle 1]: 0.00194464, [27]
        [switch_simplify]: 6.553e-05
        [loop_unroll]: 3.07e-05
        [a_1]: 0.00061398
        [with_stream_mark]: 2.586e-05
        [recompute_prepare]: 8.62998e-06
        [updatestate_depend_eliminate]: 4.55999e-06
        [updatestate_assign_eliminate]: 3.06999e-06
        [updatestate_loads_eliminate]: 2.87002e-06
        [parameter_eliminate]: 2.04e-06
        [specialize_transform]: 6.44999e-06
        [updatestate_useless_node_eliminater]: 5.29e-06
        [accelerated_algorithm]: 6.41e-06
        [meta_shard_fg_expand]: 2.22001e-06
        [get_grad_eliminate_]: 5.49e-06
        [merge_forward]: 4.05e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.472e-05
        [j_node_and_user_rematch]: 9.26002e-06
        [meta_fg_expand]: 2.14999e-06
        [replace_old_param]: 9.32001e-06
        [inline_without_move]: 5.47001e-06
        [renormalize]: 0.00084117
        [add_forward_monad_depend]: 6.19999e-06
        [auto_monad_grad]: 3.04999e-06
        [auto_monad_eliminator]: 1.651e-05
        [cse]: 3.527e-05
        [replace_applicator]: 1.479e-05
    [Cycle 2]: 0.00034743, [27]
        [switch_simplify]: 6.07999e-06
        [loop_unroll]: 5.17e-06
        [a_1]: 9.671e-05
        [with_stream_mark]: 1.219e-05
        [recompute_prepare]: 5.52999e-06
        [updatestate_depend_eliminate]: 2.78e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 2.24001e-06
        [parameter_eliminate]: 1.18001e-06
        [specialize_transform]: 4.97e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 5.70001e-06
        [meta_shard_fg_expand]: 1.21002e-06
        [get_grad_eliminate_]: 4.84e-06
        [merge_forward]: 3.09001e-06
        [cell_reuse_recompute_pass]: 1.99999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.527e-05
        [j_node_and_user_rematch]: 8.60999e-06
        [meta_fg_expand]: 2.03002e-06
        [replace_old_param]: 8.45999e-06
        [inline_without_move]: 5.10001e-06
        [renormalize]: 1.09983e-07
        [add_forward_monad_depend]: 1.23002e-06
        [auto_monad_grad]: 8.00006e-07
        [auto_monad_eliminator]: 5.51998e-06
        [cse]: 1.141e-05
        [replace_applicator]: 5.11002e-06
[py_interpret_to_execute_after_opt_a]: 1.353e-05
[rewriter_after_opt_a]: 3.888e-05
[convert_after_rewriter]: 7.29001e-06
[order_py_execute_after_rewriter]: 5.34998e-06
[mutable_eliminate]: 0.00074581
[jit_opt_b]: 5.722e-05, [1]
    [Cycle 1]: 4.76e-05, [2]
        [frontend_op_eliminate]: 1.886e-05
        [inline_after_opt_a]: 1.601e-05
[cconv]: 3.058e-05
[loop_unroll]: 0.00057426
[jit_opt_after_cconv]: 0.00017335, [1]
    [Cycle 1]: 0.00016553, [11]
        [c_1]: 2.234e-05
        [parameter_eliminate]: 3.21999e-06
        [updatestate_depend_eliminate]: 8.95999e-06
        [updatestate_assign_eliminate]: 3.29001e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [cse]: 2.741e-05
        [call_graph_tuple_transform]: 2.601e-05
        [tuple_list_get_item_eliminator]: 6.56e-06
        [none_parameter_eliminate]: 1.92001e-06
        [renormalize]: 8.99978e-07
        [switch_simplify]: 7.03e-06
[remove_dup_value]: 1.609e-05
[partial_unused_args_eliminate]: 2.67001e-06
[environ_conv]: 6.82002e-06
[add_recomputation]: 5.68e-05
[cse_after_recomputation]: 2.265e-05, [1]
    [Cycle 1]: 1.699e-05, [1]
        [cse]: 1.039e-05
[auto_monad_reorder]: 1.673e-05
[get_jit_bprop_graph]: 2.11998e-06
[rewriter_after_jit_bprop_graph]: 8.82e-06
[opt_after_jit_grad]: 0.00056
[symbol_engine_optimizer]: 8.399e-05, [1]
    [Cycle 1]: 7.629e-05, [6]
        [build]: 5.87001e-06
        [elim_shapecalc]: 9.44e-06
        [elim_not_effective]: 1.483e-05
        [opt_reshape]: 6.46e-06
        [fold_const_symbol]: 9.34e-06
        [renormalize]: 6.69999e-07
[validate]: 4.629e-05
Sums
    bootstrap                            :     0.000412s :  3.94%
    type_inference                       :     0.005521s : 52.78%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000068s :  0.65%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000040s :  0.38%
    rewriter_before_opt_a                :     0.000081s :  0.78%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000072s :  0.68%
    jit_opt_a.loop_unroll                :     0.000036s :  0.34%
    jit_opt_a.a_1                        :     0.000711s :  6.79%
    jit_opt_a.with_stream_mark           :     0.000038s :  0.36%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.17%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.17%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000841s :  8.04%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.21%
    jit_opt_a.cse                        :     0.000047s :  0.45%
    jit_opt_a.replace_applicator         :     0.000020s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.13%
    rewriter_after_opt_a                 :     0.000039s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000746s :  7.13%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.15%
    cconv                                :     0.000031s :  0.29%
    loop_unroll                          :     0.000574s :  5.49%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.09%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000027s :  0.26%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000026s :  0.25%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000007s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000007s :  0.07%
    remove_dup_value                     :     0.000016s :  0.15%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000007s :  0.07%
    add_recomputation                    :     0.000057s :  0.54%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000017s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.08%
    opt_after_jit_grad                   :     0.000560s :  5.35%
    symbol_engine_optimizer.build        :     0.000006s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000046s :  0.44%


Time group info:
------[substitution.]   0.000262    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000002s :      2: substitution.fold_const_symbol
 2.49% :     0.000007s :      3: substitution.graph_param_transform
85.82% :     0.000225s :      4: substitution.inline
 1.30% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.80% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.05% :     0.000005s :      2: substitution.replace_old_param
 5.16% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005442     2
81.62% :     0.004442s :      1: type_inference.infer
18.38% :     0.001000s :      1: type_inference.specialize
------[replace.]   0.000075     6
68.34% :     0.000052s :      4: replace.inline
31.66% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000235     6
94.71% :     0.000222s :      4: match.inline
 5.29% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000128   691
 1.62% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.56% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.25% :     0.000002s :     10: predicate.addn_zero_filter
 1.87% :     0.000002s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.73% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.29% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.48% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.17% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000002s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.32% :     0.000007s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.51% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.59% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.73% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.91% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 1.72% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.15% :     0.000003s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.03% :     0.000003s :     10: predicate.reduce_eliminate
 1.30% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.45% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.51% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.20% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.31% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.28% :     0.000002s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.77% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.71% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 1.94% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.93% :     0.000010s :     53: predicate.switch_simplify
 1.64% :     0.000002s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.42% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.64% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.59% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000691    12
50.96% :     0.000352s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.66% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
41.38% :     0.000286s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017032    72
 0.35% :     0.000060s :      1: add_recomputation
 0.42% :     0.000072s :      1: auto_monad
 0.11% :     0.000019s :      1: auto_monad_reorder
 2.56% :     0.000436s :      1: bootstrap
 0.19% :     0.000033s :      1: cconv
 0.07% :     0.000012s :      1: convert_after_rewriter
 0.15% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000010s :      1: environ_conv
 0.15% :     0.000026s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
37.37% :     0.006366s :      1: jit_opt_a
 1.04% :     0.000177s :      1: jit_opt_after_cconv
 0.36% :     0.000061s :      1: jit_opt_b
 3.42% :     0.000583s :      1: loop_unroll
 4.45% :     0.000757s :      1: mutable_eliminate
 5.53% :     0.000941s :     26: opt.transform.jit_opt_a
 0.34% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000028s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.34% :     0.000568s :      1: opt_after_jit_grad
 0.04% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000043s :      1: py_interpret_to_execute
 0.09% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000019s :      1: remove_dup_value
 2.70% :     0.000460s :      1: renormalize.infer
 2.19% :     0.000373s :      1: renormalize.specialize
 0.07% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000042s :      1: rewriter_after_opt_a
 0.49% :     0.000084s :      1: rewriter_before_opt_a
 0.51% :     0.000086s :      1: symbol_engine_optimizer
32.54% :     0.005542s :      1: type_inference


TotalTime = 0.0379865, [33]
[bootstrap]: 0.00020554
[type_inference]: 0.014901
[event_method]: 4.215e-05
[auto_monad]: 7.052e-05
[graph_reusing]: 4.46002e-06
[pre_auto_parallel]: 2.03002e-06
[py_interpret_to_execute]: 2.219e-05
[rewriter_before_opt_a]: 8.23e-05
[expand_dump_flag]: 2.08998e-06
[jit_opt_a]: 0.011606, [3]
    [Cycle 1]: 0.00554851, [27]
        [switch_simplify]: 6.124e-05
        [loop_unroll]: 3.721e-05
        [a_1]: 0.00079427
        [with_stream_mark]: 2.095e-05
        [recompute_prepare]: 1.804e-05
        [updatestate_depend_eliminate]: 7.97998e-06
        [updatestate_assign_eliminate]: 5.79e-06
        [updatestate_loads_eliminate]: 5.45001e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 1.269e-05
        [updatestate_useless_node_eliminater]: 1.192e-05
        [accelerated_algorithm]: 1.195e-05
        [meta_shard_fg_expand]: 2.29001e-06
        [get_grad_eliminate_]: 1.176e-05
        [merge_forward]: 7.05998e-06
        [cell_reuse_recompute_pass]: 8.80013e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.352e-05
        [j_node_and_user_rematch]: 2.306e-05
        [meta_fg_expand]: 0.00137448
        [replace_old_param]: 5.023e-05
        [inline_without_move]: 5.012e-05
        [renormalize]: 0.00253035
        [add_forward_monad_depend]: 1.034e-05
        [auto_monad_grad]: 5.46e-06
        [auto_monad_eliminator]: 5.354e-05
        [cse]: 0.00014986
        [replace_applicator]: 7.106e-05
    [Cycle 2]: 0.00226462, [27]
        [switch_simplify]: 3.84e-05
        [loop_unroll]: 3.783e-05
        [a_1]: 0.00113427
        [with_stream_mark]: 1.433e-05
        [recompute_prepare]: 9.39e-06
        [updatestate_depend_eliminate]: 4.53999e-06
        [updatestate_assign_eliminate]: 4.86997e-06
        [updatestate_loads_eliminate]: 4.2e-06
        [parameter_eliminate]: 1.49998e-06
        [specialize_transform]: 8.67998e-06
        [updatestate_useless_node_eliminater]: 8e-06
        [accelerated_algorithm]: 8.35999e-06
        [meta_shard_fg_expand]: 2.94999e-06
        [get_grad_eliminate_]: 8.37e-06
        [merge_forward]: 5.35999e-06
        [cell_reuse_recompute_pass]: 1.30001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.971e-05
        [j_node_and_user_rematch]: 1.393e-05
        [meta_fg_expand]: 4.161e-05
        [replace_old_param]: 1.215e-05
        [inline_without_move]: 8.27e-06
        [renormalize]: 0.00064333
        [add_forward_monad_depend]: 4.33999e-06
        [auto_monad_grad]: 1.96998e-06
        [auto_monad_eliminator]: 1.538e-05
        [cse]: 4.732e-05
        [replace_applicator]: 1.663e-05
    [Cycle 3]: 0.00050882, [27]
        [switch_simplify]: 8.77999e-06
        [loop_unroll]: 8.67998e-06
        [a_1]: 0.00019806
        [with_stream_mark]: 1.111e-05
        [recompute_prepare]: 7.92e-06
        [updatestate_depend_eliminate]: 4.45e-06
        [updatestate_assign_eliminate]: 4.63999e-06
        [updatestate_loads_eliminate]: 3.8e-06
        [parameter_eliminate]: 9.60019e-07
        [specialize_transform]: 8.66002e-06
        [updatestate_useless_node_eliminater]: 8.31002e-06
        [accelerated_algorithm]: 9.99001e-06
        [meta_shard_fg_expand]: 1.75001e-06
        [get_grad_eliminate_]: 8.55001e-06
        [merge_forward]: 4.80999e-06
        [cell_reuse_recompute_pass]: 1.77999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.774e-05
        [j_node_and_user_rematch]: 1.473e-05
        [meta_fg_expand]: 2.86999e-06
        [replace_old_param]: 1.095e-05
        [inline_without_move]: 8.22e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.57001e-06
        [auto_monad_grad]: 7.50006e-07
        [auto_monad_eliminator]: 1.044e-05
        [cse]: 2.324e-05
        [replace_applicator]: 8.29998e-06
[py_interpret_to_execute_after_opt_a]: 1.522e-05
[rewriter_after_opt_a]: 5.038e-05
[convert_after_rewriter]: 9.15001e-06
[order_py_execute_after_rewriter]: 6.59001e-06
[mutable_eliminate]: 0.00069644
[jit_opt_b]: 6.904e-05, [1]
    [Cycle 1]: 6.09e-05, [2]
        [frontend_op_eliminate]: 2.625e-05
        [inline_after_opt_a]: 2.283e-05
[cconv]: 2.206e-05
[loop_unroll]: 0.00042187
[jit_opt_after_cconv]: 0.00019008, [1]
    [Cycle 1]: 0.00018346, [11]
        [c_1]: 3.444e-05
        [parameter_eliminate]: 2.30002e-06
        [updatestate_depend_eliminate]: 6.94999e-06
        [updatestate_assign_eliminate]: 3.88999e-06
        [updatestate_loads_eliminate]: 3.71999e-06
        [cse]: 3.262e-05
        [call_graph_tuple_transform]: 2.828e-05
        [tuple_list_get_item_eliminator]: 8.85001e-06
        [none_parameter_eliminate]: 1.40001e-06
        [renormalize]: 8.30012e-07
        [switch_simplify]: 8.55999e-06
[remove_dup_value]: 2.13e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 6.46e-06
[add_recomputation]: 5.673e-05
[cse_after_recomputation]: 3.433e-05, [1]
    [Cycle 1]: 2.782e-05, [1]
        [cse]: 2.19e-05
[auto_monad_reorder]: 2.074e-05
[get_jit_bprop_graph]: 1.81e-06
[rewriter_after_jit_bprop_graph]: 5.07e-06
[opt_after_jit_grad]: 0.00045755
[symbol_engine_optimizer]: 9.028e-05, [1]
    [Cycle 1]: 8.411e-05, [6]
        [build]: 3.58999e-06
        [elim_shapecalc]: 1.091e-05
        [elim_not_effective]: 1.871e-05
        [opt_reshape]: 8.82e-06
        [fold_const_symbol]: 1.416e-05
        [renormalize]: 7.60017e-07
[validate]: 4.407e-05
[backend_pass]: 1.22999e-06
[task_emit]: 0.00862505
[execute]: 7.28999e-06
Sums
    bootstrap                            :     0.000206s :  0.61%
    type_inference                       :     0.014901s : 43.92%
    event_method                         :     0.000042s :  0.12%
    auto_monad                           :     0.000071s :  0.21%
    graph_reusing                        :     0.000004s :  0.01%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000022s :  0.07%
    rewriter_before_opt_a                :     0.000082s :  0.24%
    expand_dump_flag                     :     0.000002s :  0.01%
    jit_opt_a.switch_simplify            :     0.000108s :  0.32%
    jit_opt_a.loop_unroll                :     0.000084s :  0.25%
    jit_opt_a.a_1                        :     0.002127s :  6.27%
    jit_opt_a.with_stream_mark           :     0.000046s :  0.14%
    jit_opt_a.recompute_prepare          :     0.000035s :  0.10%
    jit_opt_a.updatestate_depend_eliminate :     0.000017s :  0.05%
    jit_opt_a.updatestate_assign_eliminate :     0.000015s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000013s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000030s :  0.09%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000028s :  0.08%
    jit_opt_a.accelerated_algorithm      :     0.000030s :  0.09%
    jit_opt_a.meta_shard_fg_expand       :     0.000007s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000029s :  0.08%
    jit_opt_a.merge_forward              :     0.000017s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000061s :  0.18%
    jit_opt_a.j_node_and_user_rematch    :     0.000052s :  0.15%
    jit_opt_a.meta_fg_expand             :     0.001419s :  4.18%
    jit_opt_a.replace_old_param          :     0.000073s :  0.22%
    jit_opt_a.inline_without_move        :     0.000067s :  0.20%
    jit_opt_a.renormalize                :     0.003174s :  9.36%
    jit_opt_a.add_forward_monad_depend   :     0.000016s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000008s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000079s :  0.23%
    jit_opt_a.cse                        :     0.000220s :  0.65%
    jit_opt_a.replace_applicator         :     0.000096s :  0.28%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.04%
    rewriter_after_opt_a                 :     0.000050s :  0.15%
    convert_after_rewriter               :     0.000009s :  0.03%
    order_py_execute_after_rewriter      :     0.000007s :  0.02%
    mutable_eliminate                    :     0.000696s :  2.05%
    jit_opt_b.frontend_op_eliminate      :     0.000026s :  0.08%
    jit_opt_b.inline_after_opt_a         :     0.000023s :  0.07%
    cconv                                :     0.000022s :  0.07%
    loop_unroll                          :     0.000422s :  1.24%
    jit_opt_after_cconv.c_1              :     0.000034s :  0.10%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.01%
    jit_opt_after_cconv.cse              :     0.000033s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000028s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.00%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.03%
    remove_dup_value                     :     0.000021s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000057s :  0.17%
    cse_after_recomputation.cse          :     0.000022s :  0.06%
    auto_monad_reorder                   :     0.000021s :  0.06%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.01%
    opt_after_jit_grad                   :     0.000458s :  1.35%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.03%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000044s :  0.13%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.008625s : 25.42%
    execute                              :     0.000007s :  0.02%


Time group info:
------[substitution.]   0.000521   126
 7.28% :     0.000038s :      2: substitution.arithmetic_simplify
 0.61% :     0.000003s :      5: substitution.elim_not_effective
 0.46% :     0.000002s :      5: substitution.fold_const_symbol
 1.49% :     0.000008s :      6: substitution.graph_param_transform
57.68% :     0.000300s :     12: substitution.inline
 3.12% :     0.000016s :      2: substitution.inline_without_move
 2.03% :     0.000011s :     18: substitution.j_node_and_user_rematch
 2.01% :     0.000010s :      7: substitution.minmaximum_grad
 1.38% :     0.000007s :      5: substitution.partial_eliminate
 2.49% :     0.000013s :     18: substitution.remove_not_recompute_node
 5.30% :     0.000028s :     10: substitution.replace_applicator
 1.46% :     0.000008s :      7: substitution.replace_old_param
 0.46% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 4.26% :     0.000022s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.74% :     0.000014s :      7: substitution.tuple_list_get_item_depend_reorder
 7.23% :     0.000038s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.014845     2
92.90% :     0.013791s :      1: type_inference.infer
 7.10% :     0.001054s :      1: type_inference.specialize
------[replace.]   0.000137    19
63.81% :     0.000088s :     12: replace.inline
36.19% :     0.000050s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000311    19
94.54% :     0.000294s :     12: match.inline
 5.46% :     0.000017s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000402  2895
 1.63% :     0.000007s :     48: predicate.accumulaten_eliminater
 0.48% :     0.000002s :      6: predicate.ad_related_special_op_eliminate
 1.44% :     0.000006s :     48: predicate.addn_check_dump
 1.66% :     0.000007s :     48: predicate.addn_zero_filter
 2.06% :     0.000008s :     48: predicate.arithmetic_simplify
 1.69% :     0.000007s :     48: predicate.cast_eliminate
 0.32% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.44% :     0.000006s :     48: predicate.compare_switch_simplify
 1.65% :     0.000007s :     48: predicate.depend_value_elim
 1.41% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.52% :     0.000006s :     48: predicate.dict_get_item_eliminator
 1.53% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.33% :     0.000001s :      6: predicate.dumpgradient_eliminate
 0.14% :     0.000001s :      6: predicate.elim_not_effective
 0.27% :     0.000001s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.47% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.45% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.46% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.50% :     0.000006s :     48: predicate.environ_get_eliminate
 1.44% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.11% :     0.000000s :      6: predicate.fold_const_symbol
 0.98% :     0.000004s :     23: predicate.get_grad_eliminate
 0.13% :     0.000001s :      6: predicate.graph_param_transform
 4.29% :     0.000017s :     79: predicate.inline
 1.73% :     0.000007s :     44: predicate.inline_without_move
 0.39% :     0.000002s :     23: predicate.j_node_and_user_rematch
 1.13% :     0.000005s :     23: predicate.less_batch_normalization
 1.88% :     0.000008s :     55: predicate.list_to_tuple_eliminator_
 1.99% :     0.000008s :     61: predicate.load_eliminater
 0.50% :     0.000002s :      6: predicate.loop_unroll_after_grad
 3.10% :     0.000012s :     90: predicate.loop_unroll_before_grad
 1.81% :     0.000007s :     54: predicate.make_slice_get_slice_eliminator
 1.45% :     0.000006s :     48: predicate.merge_addn
 1.52% :     0.000006s :     48: predicate.minmaximum_grad
 0.54% :     0.000002s :      6: predicate.mutable_eliminate
 0.24% :     0.000001s :      6: predicate.opt_reshape
 2.33% :     0.000009s :     61: predicate.partial_eliminate
 1.54% :     0.000006s :     48: predicate.print_const_string_wrapper
 1.89% :     0.000008s :     48: predicate.reduce_eliminate
 1.74% :     0.000007s :     55: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.50% :     0.000010s :    108: predicate.replace_applicator
 0.91% :     0.000004s :     44: predicate.replace_old_param
 0.14% :     0.000001s :      6: predicate.reset_defer_inline
 1.75% :     0.000007s :     48: predicate.reshape_eliminate
 1.55% :     0.000006s :     48: predicate.row_tensor_add_zeros_like
 0.33% :     0.000001s :      6: predicate.row_tensor_eliminate
 1.51% :     0.000006s :     48: predicate.same_eliminate
 0.49% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.51% :     0.000002s :     12: predicate.special_op_eliminate
 0.98% :     0.000004s :     23: predicate.specialize_transform
 1.72% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000006s :     48: predicate.stack_unstack_eliminate
 0.22% :     0.000001s :      6: predicate.switch_call_monad_eliminater
 2.54% :     0.000010s :     67: predicate.switch_defer_inline
 2.35% :     0.000009s :     67: predicate.switch_layer_defer_inline
 5.99% :     0.000024s :    163: predicate.switch_simplify
 1.51% :     0.000006s :     48: predicate.tile_eliminate
 1.54% :     0.000006s :     48: predicate.transpose_eliminate
 1.86% :     0.000007s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.61% :     0.000006s :     48: predicate.tuple_list_get_item_depend_reorder
 3.11% :     0.000012s :     67: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000007s :     48: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000007s :     55: predicate.tuple_to_list_eliminator_
 1.94% :     0.000008s :     61: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000012s :     84: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000007s :     48: predicate.value_based_eliminate
 0.22% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.28% :     0.000001s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.002952    29
80.85% :     0.002387s :     13: func_graph_cloner_run.FuncGraphClonerGraph
19.15% :     0.000566s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.044124    91
 0.14% :     0.000060s :      1: add_recomputation
 0.17% :     0.000074s :      1: auto_monad
 0.05% :     0.000024s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.50% :     0.000222s :      1: bootstrap
 0.06% :     0.000025s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.08% :     0.000037s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.10% :     0.000046s :      1: event_method
 0.02% :     0.000011s :      1: execute
 0.01% :     0.000004s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000007s :      1: graph_reusing
26.31% :     0.011609s :      1: jit_opt_a
 0.44% :     0.000193s :      1: jit_opt_after_cconv
 0.16% :     0.000072s :      1: jit_opt_b
 0.97% :     0.000430s :      1: loop_unroll
 1.60% :     0.000706s :      1: mutable_eliminate
 6.30% :     0.002779s :     39: opt.transform.jit_opt_a
 0.17% :     0.000077s :      4: opt.transform.jit_opt_after_cconv
 0.10% :     0.000043s :      4: opt.transform.jit_opt_b
 0.04% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.04% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.11% :     0.000049s :      4: opt.transform.symbol_engine_opt
 1.05% :     0.000465s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000005s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.06% :     0.000025s :      1: py_interpret_to_execute
 0.04% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000024s :      1: remove_dup_value
 3.83% :     0.001691s :      2: renormalize.infer
 3.33% :     0.001469s :      2: renormalize.specialize
 0.02% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.12% :     0.000054s :      1: rewriter_after_opt_a
 0.19% :     0.000085s :      1: rewriter_before_opt_a
 0.21% :     0.000093s :      1: symbol_engine_optimizer
19.57% :     0.008636s :      1: task_emit
33.79% :     0.014911s :      1: type_inference
 0.16% :     0.000070s :      1: validate


TotalTime = 0.0194103, [30]
[bootstrap]: 0.00017775
[type_inference]: 0.00911243
[event_method]: 4.243e-05
[auto_monad]: 9.62e-05
[graph_reusing]: 7.16001e-06
[pre_auto_parallel]: 2.78e-06
[py_interpret_to_execute]: 2.309e-05
[rewriter_before_opt_a]: 9.171e-05
[expand_dump_flag]: 3.76001e-06
[jit_opt_a]: 0.00774514, [2]
    [Cycle 1]: 0.00428955, [27]
        [switch_simplify]: 5.702e-05
        [loop_unroll]: 3.759e-05
        [a_1]: 0.00084192
        [with_stream_mark]: 2.392e-05
        [recompute_prepare]: 1.719e-05
        [updatestate_depend_eliminate]: 7.03e-06
        [updatestate_assign_eliminate]: 6.04001e-06
        [updatestate_loads_eliminate]: 5.52999e-06
        [parameter_eliminate]: 2.41e-06
        [specialize_transform]: 1.279e-05
        [updatestate_useless_node_eliminater]: 1.159e-05
        [accelerated_algorithm]: 1.261e-05
        [meta_shard_fg_expand]: 3.25e-06
        [get_grad_eliminate_]: 1.186e-05
        [merge_forward]: 7.3e-06
        [cell_reuse_recompute_pass]: 9.20001e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.519e-05
        [j_node_and_user_rematch]: 2.244e-05
        [meta_fg_expand]: 0.00137265
        [replace_old_param]: 5.396e-05
        [inline_without_move]: 5.256e-05
        [renormalize]: 0.00145681
        [add_forward_monad_depend]: 6.38998e-06
        [auto_monad_grad]: 2.50002e-06
        [auto_monad_eliminator]: 1.175e-05
        [cse]: 1.957e-05
        [replace_applicator]: 1.164e-05
    [Cycle 2]: 0.00033236, [27]
        [switch_simplify]: 4.32e-06
        [loop_unroll]: 4.87e-06
        [a_1]: 3.976e-05
        [with_stream_mark]: 1.042e-05
        [recompute_prepare]: 3.7e-06
        [updatestate_depend_eliminate]: 2.19001e-06
        [updatestate_assign_eliminate]: 2.20002e-06
        [updatestate_loads_eliminate]: 1.57999e-06
        [parameter_eliminate]: 1.57999e-06
        [specialize_transform]: 3.69002e-06
        [updatestate_useless_node_eliminater]: 3.36999e-06
        [accelerated_algorithm]: 3.61999e-06
        [meta_shard_fg_expand]: 1.52001e-06
        [get_grad_eliminate_]: 3.33e-06
        [merge_forward]: 2.12001e-06
        [cell_reuse_recompute_pass]: 1.97999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.153e-05
        [j_node_and_user_rematch]: 5.84999e-06
        [meta_fg_expand]: 2.542e-05
        [replace_old_param]: 4.12998e-06
        [inline_without_move]: 3.38e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.71e-06
        [auto_monad_grad]: 1.30001e-06
        [auto_monad_eliminator]: 5.01002e-06
        [cse]: 1.178e-05
        [replace_applicator]: 6.79001e-06
[py_interpret_to_execute_after_opt_a]: 1.2e-05
[rewriter_after_opt_a]: 3.452e-05
[convert_after_rewriter]: 6.35002e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00057406
[jit_opt_b]: 4.312e-05, [1]
    [Cycle 1]: 3.518e-05, [2]
        [frontend_op_eliminate]: 1.088e-05
        [inline_after_opt_a]: 1.136e-05
[cconv]: 2.315e-05
[loop_unroll]: 0.00044721
[jit_opt_after_cconv]: 0.00012823, [1]
    [Cycle 1]: 0.0001208, [11]
        [c_1]: 1.188e-05
        [parameter_eliminate]: 3.39001e-06
        [updatestate_depend_eliminate]: 5.33002e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.54998e-06
        [cse]: 1.946e-05
        [call_graph_tuple_transform]: 1.567e-05
        [tuple_list_get_item_eliminator]: 3.65e-06
        [none_parameter_eliminate]: 1.40999e-06
        [renormalize]: 6.99976e-07
        [switch_simplify]: 3.61999e-06
[remove_dup_value]: 1.473e-05
[partial_unused_args_eliminate]: 2.24001e-06
[environ_conv]: 4.90999e-06
[add_recomputation]: 3.341e-05
[cse_after_recomputation]: 2.094e-05, [1]
    [Cycle 1]: 1.468e-05, [1]
        [cse]: 8.54e-06
[auto_monad_reorder]: 1.26e-05
[get_jit_bprop_graph]: 1.86e-06
[rewriter_after_jit_bprop_graph]: 6.79001e-06
[opt_after_jit_grad]: 0.00048417
[symbol_engine_optimizer]: 6.457e-05, [1]
    [Cycle 1]: 5.806e-05, [6]
        [build]: 3.21999e-06
        [elim_shapecalc]: 5.61998e-06
        [elim_not_effective]: 1.077e-05
        [opt_reshape]: 4.10998e-06
        [fold_const_symbol]: 5.74e-06
        [renormalize]: 6.50005e-07
[validate]: 2.797e-05
Sums
    bootstrap                            :     0.000178s :  1.14%
    type_inference                       :     0.009112s : 58.27%
    event_method                         :     0.000042s :  0.27%
    auto_monad                           :     0.000096s :  0.62%
    graph_reusing                        :     0.000007s :  0.05%
    pre_auto_parallel                    :     0.000003s :  0.02%
    py_interpret_to_execute              :     0.000023s :  0.15%
    rewriter_before_opt_a                :     0.000092s :  0.59%
    expand_dump_flag                     :     0.000004s :  0.02%
    jit_opt_a.switch_simplify            :     0.000061s :  0.39%
    jit_opt_a.loop_unroll                :     0.000042s :  0.27%
    jit_opt_a.a_1                        :     0.000882s :  5.64%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.22%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000008s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000016s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.10%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.10%
    jit_opt_a.merge_forward              :     0.000009s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.02%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.23%
    jit_opt_a.j_node_and_user_rematch    :     0.000028s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001398s :  8.94%
    jit_opt_a.replace_old_param          :     0.000058s :  0.37%
    jit_opt_a.inline_without_move        :     0.000056s :  0.36%
    jit_opt_a.renormalize                :     0.001457s :  9.32%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.11%
    jit_opt_a.cse                        :     0.000031s :  0.20%
    jit_opt_a.replace_applicator         :     0.000018s :  0.12%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.08%
    rewriter_after_opt_a                 :     0.000035s :  0.22%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000005s :  0.03%
    mutable_eliminate                    :     0.000574s :  3.67%
    jit_opt_b.frontend_op_eliminate      :     0.000011s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000023s :  0.15%
    loop_unroll                          :     0.000447s :  2.86%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000019s :  0.12%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.10%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.02%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.02%
    remove_dup_value                     :     0.000015s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000005s :  0.03%
    add_recomputation                    :     0.000033s :  0.21%
    cse_after_recomputation.cse          :     0.000009s :  0.05%
    auto_monad_reorder                   :     0.000013s :  0.08%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.04%
    opt_after_jit_grad                   :     0.000484s :  3.10%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000011s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000028s :  0.18%


Time group info:
------[substitution.]   0.000231    50
 0.78% :     0.000002s :      1: substitution.elim_not_effective
 0.47% :     0.000001s :      1: substitution.fold_const_symbol
 1.82% :     0.000004s :      1: substitution.graph_param_transform
64.25% :     0.000148s :      6: substitution.inline
 7.21% :     0.000017s :      2: substitution.inline_without_move
 2.91% :     0.000007s :      9: substitution.j_node_and_user_rematch
 1.33% :     0.000003s :      2: substitution.minmaximum_grad
 1.58% :     0.000004s :      3: substitution.partial_eliminate
 3.00% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.21% :     0.000005s :      2: substitution.replace_applicator
 2.00% :     0.000005s :      5: substitution.replace_old_param
 1.10% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 2.89% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 1.84% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 6.59% :     0.000015s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.009050     2
88.93% :     0.008049s :      1: type_inference.infer
11.07% :     0.001002s :      1: type_inference.specialize
------[replace.]   0.000058     8
79.60% :     0.000046s :      6: replace.inline
20.40% :     0.000012s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000156     8
93.25% :     0.000145s :      6: match.inline
 6.75% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000169  1109
 1.74% :     0.000003s :     18: predicate.accumulaten_eliminater
 0.66% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.30% :     0.000002s :     18: predicate.addn_check_dump
 1.46% :     0.000002s :     18: predicate.addn_zero_filter
 2.13% :     0.000004s :     18: predicate.arithmetic_simplify
 1.39% :     0.000002s :     18: predicate.cast_eliminate
 0.15% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.30% :     0.000002s :     18: predicate.compare_switch_simplify
 1.39% :     0.000002s :     18: predicate.depend_value_elim
 1.28% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.40% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.45% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.10% :     0.000000s :      1: predicate.elim_not_effective
 0.20% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.34% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.25% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.35% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.36% :     0.000002s :     18: predicate.environ_get_eliminate
 1.27% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.17% :     0.000002s :     12: predicate.get_grad_eliminate
 0.20% :     0.000000s :      1: predicate.graph_param_transform
 4.47% :     0.000008s :     28: predicate.inline
 3.34% :     0.000006s :     33: predicate.inline_without_move
 0.53% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :     12: predicate.less_batch_normalization
 1.59% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.66% :     0.000003s :     21: predicate.load_eliminater
 0.89% :     0.000002s :      1: predicate.loop_unroll_after_grad
 3.56% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.85% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.33% :     0.000002s :     18: predicate.merge_addn
 1.49% :     0.000003s :     18: predicate.minmaximum_grad
 0.82% :     0.000001s :      1: predicate.mutable_eliminate
 0.27% :     0.000000s :      1: predicate.opt_reshape
 1.97% :     0.000003s :     21: predicate.partial_eliminate
 1.37% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.85% :     0.000003s :     18: predicate.reduce_eliminate
 1.82% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.73% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.41% :     0.000002s :     22: predicate.replace_applicator
 1.94% :     0.000003s :     33: predicate.replace_old_param
 0.12% :     0.000000s :      1: predicate.reset_defer_inline
 1.74% :     0.000003s :     18: predicate.reshape_eliminate
 1.42% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.58% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.35% :     0.000002s :     18: predicate.same_eliminate
 0.64% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.46% :     0.000001s :      2: predicate.special_op_eliminate
 1.22% :     0.000002s :     12: predicate.specialize_transform
 1.41% :     0.000002s :     18: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.12% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.55% :     0.000004s :     26: predicate.switch_defer_inline
 2.47% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.74% :     0.000011s :     71: predicate.switch_simplify
 1.43% :     0.000002s :     18: predicate.tile_eliminate
 1.68% :     0.000003s :     18: predicate.transpose_eliminate
 1.56% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.45% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 3.22% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.68% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.57% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.13% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000003s :     18: predicate.value_based_eliminate
 0.12% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.28% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001020    22
73.99% :     0.000755s :     13: func_graph_cloner_run.FuncGraphClonerGraph
26.01% :     0.000265s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.022135    72
 0.16% :     0.000036s :      1: add_recomputation
 0.46% :     0.000101s :      1: auto_monad
 0.07% :     0.000015s :      1: auto_monad_reorder
 0.84% :     0.000187s :      1: bootstrap
 0.12% :     0.000026s :      1: cconv
 0.04% :     0.000009s :      1: convert_after_rewriter
 0.11% :     0.000023s :      1: cse_after_recomputation
 0.03% :     0.000007s :      1: environ_conv
 0.21% :     0.000047s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000010s :      1: graph_reusing
35.00% :     0.007748s :      1: jit_opt_a
 0.59% :     0.000131s :      1: jit_opt_after_cconv
 0.21% :     0.000046s :      1: jit_opt_b
 2.07% :     0.000458s :      1: loop_unroll
 2.64% :     0.000585s :      1: mutable_eliminate
 5.59% :     0.001238s :     26: opt.transform.jit_opt_a
 0.14% :     0.000031s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000016s :      4: opt.transform.jit_opt_b
 0.05% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.06% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000016s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000023s :      4: opt.transform.symbol_engine_opt
 2.23% :     0.000493s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000005s :      1: pre_auto_parallel
 0.12% :     0.000026s :      1: py_interpret_to_execute
 0.07% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000017s :      1: remove_dup_value
 5.04% :     0.001116s :      1: renormalize.infer
 1.50% :     0.000333s :      1: renormalize.specialize
 0.04% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.17% :     0.000038s :      1: rewriter_after_opt_a
 0.43% :     0.000094s :      1: rewriter_before_opt_a
 0.30% :     0.000067s :      1: symbol_engine_optimizer
41.22% :     0.009123s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.405.282 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843d1160> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.405.886 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.407.456 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffda80f4a60> should be wrapped by jit only once.
[WARNING] ME(28432:281473069915952,ForkProcess-7):2026-01-29-17:46:22.408.584 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0133636, [30]
[bootstrap]: 0.00038615
[type_inference]: 0.00518357
[event_method]: 1.736e-05
[auto_monad]: 6.24e-05
[graph_reusing]: 6.17999e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 4.674e-05
[rewriter_before_opt_a]: 7.516e-05
[expand_dump_flag]: 3.13998e-06
[jit_opt_a]: 0.00550394, [2]
    [Cycle 1]: 0.00159037, [27]
        [switch_simplify]: 5.406e-05
        [loop_unroll]: 3.057e-05
        [a_1]: 0.00052429
        [with_stream_mark]: 1.423e-05
        [recompute_prepare]: 6.93e-06
        [updatestate_depend_eliminate]: 1.353e-05
        [updatestate_assign_eliminate]: 3.10002e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 2.27001e-06
        [specialize_transform]: 6.96999e-06
        [updatestate_useless_node_eliminater]: 5.14998e-06
        [accelerated_algorithm]: 5.91998e-06
        [meta_shard_fg_expand]: 1.80001e-06
        [get_grad_eliminate_]: 5.35999e-06
        [merge_forward]: 3.78999e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.526e-05
        [j_node_and_user_rematch]: 9.34e-06
        [meta_fg_expand]: 2.09999e-06
        [replace_old_param]: 8.85999e-06
        [inline_without_move]: 5.19e-06
        [renormalize]: 0.0006428
        [add_forward_monad_depend]: 4.96997e-06
        [auto_monad_grad]: 2.21e-06
        [auto_monad_eliminator]: 1.295e-05
        [cse]: 2.641e-05
        [replace_applicator]: 1.079e-05
    [Cycle 2]: 0.00032996, [27]
        [switch_simplify]: 5.84999e-06
        [loop_unroll]: 5.02999e-06
        [a_1]: 9.153e-05
        [with_stream_mark]: 8.54e-06
        [recompute_prepare]: 5.31002e-06
        [updatestate_depend_eliminate]: 2.51e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 5.14e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 5.32999e-06
        [meta_shard_fg_expand]: 1.13001e-06
        [get_grad_eliminate_]: 4.74e-06
        [merge_forward]: 2.41e-06
        [cell_reuse_recompute_pass]: 1.39e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.317e-05
        [j_node_and_user_rematch]: 7.69002e-06
        [meta_fg_expand]: 1.97001e-06
        [replace_old_param]: 7.25e-06
        [inline_without_move]: 4.80999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.07998e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 5.34998e-06
        [cse]: 1.242e-05
        [replace_applicator]: 5.61e-06
[py_interpret_to_execute_after_opt_a]: 9.07999e-06
[rewriter_after_opt_a]: 3.372e-05
[convert_after_rewriter]: 7.82998e-06
[order_py_execute_after_rewriter]: 4.88001e-06
[mutable_eliminate]: 0.00051367
[jit_opt_b]: 5.09e-05, [1]
    [Cycle 1]: 4.379e-05, [2]
        [frontend_op_eliminate]: 1.743e-05
        [inline_after_opt_a]: 1.478e-05
[cconv]: 2.294e-05
[loop_unroll]: 0.00041481
[jit_opt_after_cconv]: 0.00014055, [1]
    [Cycle 1]: 0.00013353, [11]
        [c_1]: 2.095e-05
        [parameter_eliminate]: 2.44001e-06
        [updatestate_depend_eliminate]: 5.40999e-06
        [updatestate_assign_eliminate]: 2.29999e-06
        [updatestate_loads_eliminate]: 2.12999e-06
        [cse]: 1.74e-05
        [call_graph_tuple_transform]: 1.93e-05
        [tuple_list_get_item_eliminator]: 6.02999e-06
        [none_parameter_eliminate]: 1.34e-06
        [renormalize]: 3.89991e-07
        [switch_simplify]: 5.42001e-06
[remove_dup_value]: 1.401e-05
[partial_unused_args_eliminate]: 2.12999e-06
[environ_conv]: 5.33002e-06
[add_recomputation]: 4.685e-05
[cse_after_recomputation]: 2.105e-05, [1]
    [Cycle 1]: 1.566e-05, [1]
        [cse]: 9.69e-06
[auto_monad_reorder]: 1.452e-05
[get_jit_bprop_graph]: 1.49e-06
[rewriter_after_jit_bprop_graph]: 4.97999e-06
[opt_after_jit_grad]: 0.00047982
[symbol_engine_optimizer]: 7.394e-05, [1]
    [Cycle 1]: 6.758e-05, [6]
        [build]: 3.45e-06
        [elim_shapecalc]: 8.58001e-06
        [elim_not_effective]: 1.225e-05
        [opt_reshape]: 6.02001e-06
        [fold_const_symbol]: 8.42e-06
        [renormalize]: 8.39995e-07
[validate]: 3.246e-05
Sums
    bootstrap                            :     0.000386s :  4.20%
    type_inference                       :     0.005184s : 56.42%
    event_method                         :     0.000017s :  0.19%
    auto_monad                           :     0.000062s :  0.68%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000047s :  0.51%
    rewriter_before_opt_a                :     0.000075s :  0.82%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000060s :  0.65%
    jit_opt_a.loop_unroll                :     0.000036s :  0.39%
    jit_opt_a.a_1                        :     0.000616s :  6.70%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.25%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.17%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.18%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000643s :  7.00%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.20%
    jit_opt_a.cse                        :     0.000039s :  0.42%
    jit_opt_a.replace_applicator         :     0.000016s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.10%
    rewriter_after_opt_a                 :     0.000034s :  0.37%
    convert_after_rewriter               :     0.000008s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000514s :  5.59%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000023s :  0.25%
    loop_unroll                          :     0.000415s :  4.51%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000047s :  0.51%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.16%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000480s :  5.22%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000032s :  0.35%


Time group info:
------[substitution.]   0.000192    23
 0.94% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 2.84% :     0.000005s :      3: substitution.graph_param_transform
82.45% :     0.000158s :      4: substitution.inline
 1.53% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.57% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.19% :     0.000004s :      2: substitution.replace_old_param
 6.82% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005114     2
81.69% :     0.004178s :      1: type_inference.infer
18.31% :     0.000936s :      1: type_inference.specialize
------[replace.]   0.000068     6
66.81% :     0.000046s :      4: replace.inline
33.19% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000168     6
92.86% :     0.000156s :      4: match.inline
 7.14% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.32% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.37% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.16% :     0.000001s :     10: predicate.addn_zero_filter
 2.09% :     0.000002s :     10: predicate.arithmetic_simplify
 1.27% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.56% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.27% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.17% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.13% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.21% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.31% :     0.000002s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.57% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.12% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.57% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.36% :     0.000003s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.74% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.66% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.60% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.47% :     0.000010s :     53: predicate.switch_simplify
 1.12% :     0.000001s :     10: predicate.tile_eliminate
 1.12% :     0.000001s :     10: predicate.transpose_eliminate
 1.63% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.16% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.61% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.57% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000657    12
51.98% :     0.000341s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.24% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.78% :     0.000288s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014889    72
 0.33% :     0.000050s :      1: add_recomputation
 0.44% :     0.000066s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 2.75% :     0.000410s :      1: bootstrap
 0.17% :     0.000026s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000021s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
36.99% :     0.005508s :      1: jit_opt_a
 0.96% :     0.000144s :      1: jit_opt_after_cconv
 0.36% :     0.000053s :      1: jit_opt_b
 2.84% :     0.000422s :      1: loop_unroll
 3.50% :     0.000522s :      1: mutable_eliminate
 5.55% :     0.000826s :     26: opt.transform.jit_opt_a
 0.32% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.27% :     0.000487s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.34% :     0.000050s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 2.21% :     0.000328s :      1: renormalize.infer
 2.06% :     0.000307s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000037s :      1: rewriter_after_opt_a
 0.52% :     0.000078s :      1: rewriter_before_opt_a
 0.51% :     0.000077s :      1: symbol_engine_optimizer
34.91% :     0.005198s :      1: type_inference


TotalTime = 0.0179689, [30]
[bootstrap]: 0.00018778
[type_inference]: 0.008518
[event_method]: 3.978e-05
[auto_monad]: 7.823e-05
[graph_reusing]: 5.39e-06
[pre_auto_parallel]: 2.22999e-06
[py_interpret_to_execute]: 2.293e-05
[rewriter_before_opt_a]: 8.368e-05
[expand_dump_flag]: 2.39001e-06
[jit_opt_a]: 0.00726479, [2]
    [Cycle 1]: 0.00407021, [27]
        [switch_simplify]: 5.378e-05
        [loop_unroll]: 3.728e-05
        [a_1]: 0.0007724
        [with_stream_mark]: 2.081e-05
        [recompute_prepare]: 1.752e-05
        [updatestate_depend_eliminate]: 7.00002e-06
        [updatestate_assign_eliminate]: 5.44e-06
        [updatestate_loads_eliminate]: 5.29003e-06
        [parameter_eliminate]: 2.02001e-06
        [specialize_transform]: 1.299e-05
        [updatestate_useless_node_eliminater]: 1.152e-05
        [accelerated_algorithm]: 1.209e-05
        [meta_shard_fg_expand]: 2.57001e-06
        [get_grad_eliminate_]: 1.201e-05
        [merge_forward]: 6.89001e-06
        [cell_reuse_recompute_pass]: 7.50006e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.248e-05
        [j_node_and_user_rematch]: 2.15e-05
        [meta_fg_expand]: 0.001349
        [replace_old_param]: 4.972e-05
        [inline_without_move]: 4.947e-05
        [renormalize]: 0.00137627
        [add_forward_monad_depend]: 4.25e-06
        [auto_monad_grad]: 1.22e-06
        [auto_monad_eliminator]: 8e-06
        [cse]: 1.512e-05
        [replace_applicator]: 7.88999e-06
    [Cycle 2]: 0.0002494, [27]
        [switch_simplify]: 4.11001e-06
        [loop_unroll]: 3.7e-06
        [a_1]: 3.682e-05
        [with_stream_mark]: 6.43998e-06
        [recompute_prepare]: 3.48e-06
        [updatestate_depend_eliminate]: 1.79998e-06
        [updatestate_assign_eliminate]: 1.51998e-06
        [updatestate_loads_eliminate]: 1.39998e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 3.55e-06
        [updatestate_useless_node_eliminater]: 3.26999e-06
        [accelerated_algorithm]: 3.65e-06
        [meta_shard_fg_expand]: 1.05999e-06
        [get_grad_eliminate_]: 3.28998e-06
        [merge_forward]: 1.79998e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 8.41002e-06
        [j_node_and_user_rematch]: 5.52999e-06
        [meta_fg_expand]: 2.183e-05
        [replace_old_param]: 3.42002e-06
        [inline_without_move]: 3.09001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 7.2e-07
        [auto_monad_eliminator]: 4.13001e-06
        [cse]: 9.51e-06
        [replace_applicator]: 3.23998e-06
[py_interpret_to_execute_after_opt_a]: 7.63001e-06
[rewriter_after_opt_a]: 2.53e-05
[convert_after_rewriter]: 5.29e-06
[order_py_execute_after_rewriter]: 4.4e-06
[mutable_eliminate]: 0.00043054
[jit_opt_b]: 3.852e-05, [1]
    [Cycle 1]: 3.229e-05, [2]
        [frontend_op_eliminate]: 1.045e-05
        [inline_after_opt_a]: 1.078e-05
[cconv]: 1.459e-05
[loop_unroll]: 0.00039978
[jit_opt_after_cconv]: 0.00010935, [1]
    [Cycle 1]: 0.00010334, [11]
        [c_1]: 1.095e-05
        [parameter_eliminate]: 2.09999e-06
        [updatestate_depend_eliminate]: 3.51999e-06
        [updatestate_assign_eliminate]: 1.96998e-06
        [updatestate_loads_eliminate]: 1.49e-06
        [cse]: 1.389e-05
        [call_graph_tuple_transform]: 1.149e-05
        [tuple_list_get_item_eliminator]: 3.81001e-06
        [none_parameter_eliminate]: 8.50006e-07
        [renormalize]: 3.9002e-07
        [switch_simplify]: 3.70998e-06
[remove_dup_value]: 8.28001e-06
[partial_unused_args_eliminate]: 1.95001e-06
[environ_conv]: 3.85998e-06
[add_recomputation]: 2.298e-05
[cse_after_recomputation]: 1.864e-05, [1]
    [Cycle 1]: 1.365e-05, [1]
        [cse]: 8.07e-06
[auto_monad_reorder]: 7.92e-06
[get_jit_bprop_graph]: 1.51998e-06
[rewriter_after_jit_bprop_graph]: 4.18001e-06
[opt_after_jit_grad]: 0.00042923
[symbol_engine_optimizer]: 5.925e-05, [1]
    [Cycle 1]: 5.379e-05, [6]
        [build]: 2.37999e-06
        [elim_shapecalc]: 5.30999e-06
        [elim_not_effective]: 8.47e-06
        [opt_reshape]: 4.15e-06
        [fold_const_symbol]: 5.69e-06
        [renormalize]: 5.00004e-07
[validate]: 2.13e-05
Sums
    bootstrap                            :     0.000188s :  1.30%
    type_inference                       :     0.008518s : 58.89%
    event_method                         :     0.000040s :  0.28%
    auto_monad                           :     0.000078s :  0.54%
    graph_reusing                        :     0.000005s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000023s :  0.16%
    rewriter_before_opt_a                :     0.000084s :  0.58%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.40%
    jit_opt_a.loop_unroll                :     0.000041s :  0.28%
    jit_opt_a.a_1                        :     0.000809s :  5.59%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.19%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.11%
    jit_opt_a.merge_forward              :     0.000009s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.001371s :  9.48%
    jit_opt_a.replace_old_param          :     0.000053s :  0.37%
    jit_opt_a.inline_without_move        :     0.000053s :  0.36%
    jit_opt_a.renormalize                :     0.001376s :  9.52%
    jit_opt_a.add_forward_monad_depend   :     0.000005s :  0.04%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.01%
    jit_opt_a.auto_monad_eliminator      :     0.000012s :  0.08%
    jit_opt_a.cse                        :     0.000025s :  0.17%
    jit_opt_a.replace_applicator         :     0.000011s :  0.08%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.05%
    rewriter_after_opt_a                 :     0.000025s :  0.17%
    convert_after_rewriter               :     0.000005s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000431s :  2.98%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000015s :  0.10%
    loop_unroll                          :     0.000400s :  2.76%
    jit_opt_after_cconv.c_1              :     0.000011s :  0.08%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.cse              :     0.000014s :  0.10%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000011s :  0.08%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.03%
    remove_dup_value                     :     0.000008s :  0.06%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000023s :  0.16%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000008s :  0.05%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000004s :  0.03%
    opt_after_jit_grad                   :     0.000429s :  2.97%
    symbol_engine_optimizer.build        :     0.000002s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000008s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000021s :  0.15%


Time group info:
------[substitution.]   0.000200    50
 0.50% :     0.000001s :      1: substitution.elim_not_effective
 0.29% :     0.000001s :      1: substitution.fold_const_symbol
 1.36% :     0.000003s :      1: substitution.graph_param_transform
63.39% :     0.000127s :      6: substitution.inline
 7.58% :     0.000015s :      2: substitution.inline_without_move
 2.83% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.72% :     0.000003s :      2: substitution.minmaximum_grad
 1.82% :     0.000004s :      3: substitution.partial_eliminate
 3.07% :     0.000006s :      9: substitution.remove_not_recompute_node
 2.84% :     0.000006s :      2: substitution.replace_applicator
 1.86% :     0.000004s :      5: substitution.replace_old_param
 1.19% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 3.41% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 2.25% :     0.000005s :      2: substitution.tuple_list_get_item_depend_reorder
 5.88% :     0.000012s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008461     2
89.04% :     0.007534s :      1: type_inference.infer
10.96% :     0.000927s :      1: type_inference.specialize
------[replace.]   0.000055     8
77.95% :     0.000043s :      6: replace.inline
22.05% :     0.000012s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000131     8
94.55% :     0.000124s :      6: match.inline
 5.45% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000160  1109
 1.49% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.56% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.38% :     0.000002s :     18: predicate.addn_check_dump
 1.57% :     0.000003s :     18: predicate.addn_zero_filter
 2.01% :     0.000003s :     18: predicate.arithmetic_simplify
 1.43% :     0.000002s :     18: predicate.cast_eliminate
 0.16% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.37% :     0.000002s :     18: predicate.compare_switch_simplify
 1.51% :     0.000002s :     18: predicate.depend_value_elim
 1.36% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.40% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.49% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      1: predicate.elim_not_effective
 0.29% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.38% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.55% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.33% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.46% :     0.000002s :     18: predicate.environ_get_eliminate
 1.36% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.07% :     0.000000s :      1: predicate.fold_const_symbol
 1.20% :     0.000002s :     12: predicate.get_grad_eliminate
 0.12% :     0.000000s :      1: predicate.graph_param_transform
 4.03% :     0.000006s :     28: predicate.inline
 3.43% :     0.000006s :     33: predicate.inline_without_move
 0.55% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :     12: predicate.less_batch_normalization
 1.66% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.69% :     0.000003s :     21: predicate.load_eliminater
 0.60% :     0.000001s :      1: predicate.loop_unroll_after_grad
 3.71% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.75% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.33% :     0.000002s :     18: predicate.merge_addn
 1.35% :     0.000002s :     18: predicate.minmaximum_grad
 0.75% :     0.000001s :      1: predicate.mutable_eliminate
 0.28% :     0.000000s :      1: predicate.opt_reshape
 2.16% :     0.000003s :     21: predicate.partial_eliminate
 1.44% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.94% :     0.000003s :     18: predicate.reduce_eliminate
 1.78% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.47% :     0.000002s :     22: predicate.replace_applicator
 1.66% :     0.000003s :     33: predicate.replace_old_param
 0.11% :     0.000000s :      1: predicate.reset_defer_inline
 1.57% :     0.000003s :     18: predicate.reshape_eliminate
 1.46% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.55% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.46% :     0.000002s :     18: predicate.same_eliminate
 0.64% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.30% :     0.000000s :      2: predicate.special_op_eliminate
 1.27% :     0.000002s :     12: predicate.specialize_transform
 1.53% :     0.000002s :     18: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.13% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.68% :     0.000004s :     26: predicate.switch_defer_inline
 2.36% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.85% :     0.000011s :     71: predicate.switch_simplify
 1.45% :     0.000002s :     18: predicate.tile_eliminate
 1.43% :     0.000002s :     18: predicate.transpose_eliminate
 1.74% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.61% :     0.000003s :     18: predicate.tuple_list_get_item_depend_reorder
 2.81% :     0.000005s :     22: predicate.tuple_list_get_item_eliminator
 1.76% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.64% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.81% :     0.000003s :     18: predicate.value_based_eliminate
 0.11% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.28% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000972    22
73.97% :     0.000719s :     13: func_graph_cloner_run.FuncGraphClonerGraph
26.03% :     0.000253s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020512    72
 0.12% :     0.000026s :      1: add_recomputation
 0.40% :     0.000082s :      1: auto_monad
 0.05% :     0.000011s :      1: auto_monad_reorder
 0.98% :     0.000202s :      1: bootstrap
 0.08% :     0.000017s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.10% :     0.000021s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.22% :     0.000044s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
35.43% :     0.007267s :      1: jit_opt_a
 0.54% :     0.000112s :      1: jit_opt_after_cconv
 0.20% :     0.000041s :      1: jit_opt_b
 1.99% :     0.000407s :      1: loop_unroll
 2.14% :     0.000438s :      1: mutable_eliminate
 5.56% :     0.001141s :     26: opt.transform.jit_opt_a
 0.13% :     0.000027s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000015s :      4: opt.transform.jit_opt_b
 0.04% :     0.000009s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.06% :     0.000013s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000020s :      4: opt.transform.symbol_engine_opt
 2.12% :     0.000436s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000026s :      1: py_interpret_to_execute
 0.05% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.05% :     0.000011s :      1: remove_dup_value
 5.20% :     0.001066s :      1: renormalize.infer
 1.48% :     0.000304s :      1: renormalize.specialize
 0.03% :     0.000006s :      1: rewriter_after_jit_bprop_graph
 0.14% :     0.000028s :      1: rewriter_after_opt_a
 0.42% :     0.000086s :      1: rewriter_before_opt_a
 0.30% :     0.000062s :      1: symbol_engine_optimizer
41.58% :     0.008529s :      1: type_inference

[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.431.351 [mindspore/graph/api.py:1147] The func <function _Grad.__call__.<locals>.after_grad at 0xfffd843dbb80> should be wrapped by jit only once.
[WARNING] ME(27782:281473069915952,ForkProcess-4):2026-01-29-17:46:22.431.860 [mindspore/graph/api.py:128] The function "after_grad" at the file "/home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/graph/api.py", line 1152 has been compiled again. Try to reuse the function object decorated by @jit to reduce the compile time. For more details, get instructions about `jit` at https://www.mindspore.cn/search?inputValue=jit. 

TotalTime = 0.0303536, [33]
[bootstrap]: 0.00034564
[type_inference]: 0.00888556
[event_method]: 4.118e-05
[auto_monad]: 6.926e-05
[graph_reusing]: 5.17e-06
[pre_auto_parallel]: 2.21e-06
[py_interpret_to_execute]: 2.175e-05
[rewriter_before_opt_a]: 8.126e-05
[expand_dump_flag]: 3.09999e-06
[jit_opt_a]: 0.010736, [3]
    [Cycle 1]: 0.00521757, [27]
        [switch_simplify]: 5.352e-05
        [loop_unroll]: 3.717e-05
        [a_1]: 0.00078496
        [with_stream_mark]: 1.682e-05
        [recompute_prepare]: 1.591e-05
        [updatestate_depend_eliminate]: 6.61999e-06
        [updatestate_assign_eliminate]: 5.43002e-06
        [updatestate_loads_eliminate]: 4.91002e-06
        [parameter_eliminate]: 1.44998e-06
        [specialize_transform]: 1.26e-05
        [updatestate_useless_node_eliminater]: 1.18e-05
        [accelerated_algorithm]: 1.179e-05
        [meta_shard_fg_expand]: 2.15002e-06
        [get_grad_eliminate_]: 1.154e-05
        [merge_forward]: 6.21998e-06
        [cell_reuse_recompute_pass]: 7.00005e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.274e-05
        [j_node_and_user_rematch]: 2.141e-05
        [meta_fg_expand]: 0.00124118
        [replace_old_param]: 4.537e-05
        [inline_without_move]: 4.9e-05
        [renormalize]: 0.00240148
        [add_forward_monad_depend]: 9.51e-06
        [auto_monad_grad]: 4.99e-06
        [auto_monad_eliminator]: 4.862e-05
        [cse]: 0.00014131
        [replace_applicator]: 6.686e-05
    [Cycle 2]: 0.00210315, [27]
        [switch_simplify]: 3.725e-05
        [loop_unroll]: 3.625e-05
        [a_1]: 0.00109251
        [with_stream_mark]: 1.251e-05
        [recompute_prepare]: 9.76998e-06
        [updatestate_depend_eliminate]: 5.15001e-06
        [updatestate_assign_eliminate]: 4.07e-06
        [updatestate_loads_eliminate]: 3.64002e-06
        [parameter_eliminate]: 1.17e-06
        [specialize_transform]: 8.49002e-06
        [updatestate_useless_node_eliminater]: 7.92003e-06
        [accelerated_algorithm]: 8.27e-06
        [meta_shard_fg_expand]: 1.98002e-06
        [get_grad_eliminate_]: 7.95e-06
        [merge_forward]: 4.13001e-06
        [cell_reuse_recompute_pass]: 7.60017e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.793e-05
        [j_node_and_user_rematch]: 1.357e-05
        [meta_fg_expand]: 3.784e-05
        [replace_old_param]: 1.149e-05
        [inline_without_move]: 8.01001e-06
        [renormalize]: 0.00055252
        [add_forward_monad_depend]: 4.18999e-06
        [auto_monad_grad]: 1.37e-06
        [auto_monad_eliminator]: 1.467e-05
        [cse]: 4.18e-05
        [replace_applicator]: 1.503e-05
    [Cycle 3]: 0.00047759, [27]
        [switch_simplify]: 8.38999e-06
        [loop_unroll]: 7.94002e-06
        [a_1]: 0.00018643
        [with_stream_mark]: 1.088e-05
        [recompute_prepare]: 8.12e-06
        [updatestate_depend_eliminate]: 4.28999e-06
        [updatestate_assign_eliminate]: 4.14002e-06
        [updatestate_loads_eliminate]: 3.56999e-06
        [parameter_eliminate]: 1.13001e-06
        [specialize_transform]: 8.3e-06
        [updatestate_useless_node_eliminater]: 7.61999e-06
        [accelerated_algorithm]: 8.32e-06
        [meta_shard_fg_expand]: 1.54e-06
        [get_grad_eliminate_]: 7.56999e-06
        [merge_forward]: 4.02e-06
        [cell_reuse_recompute_pass]: 1.41998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.632e-05
        [j_node_and_user_rematch]: 1.33e-05
        [meta_fg_expand]: 2.89999e-06
        [replace_old_param]: 9.77999e-06
        [inline_without_move]: 7.45e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.35999e-06
        [auto_monad_grad]: 8.39995e-07
        [auto_monad_eliminator]: 1.019e-05
        [cse]: 2.301e-05
        [replace_applicator]: 7.96001e-06
[py_interpret_to_execute_after_opt_a]: 1.394e-05
[rewriter_after_opt_a]: 4.533e-05
[convert_after_rewriter]: 8.2e-06
[order_py_execute_after_rewriter]: 6.29999e-06
[mutable_eliminate]: 0.00067879
[jit_opt_b]: 6.491e-05, [1]
    [Cycle 1]: 5.75e-05, [2]
        [frontend_op_eliminate]: 2.383e-05
        [inline_after_opt_a]: 2.281e-05
[cconv]: 2.176e-05
[loop_unroll]: 0.00040844
[jit_opt_after_cconv]: 0.00018349, [1]
    [Cycle 1]: 0.00017675, [11]
        [c_1]: 3.362e-05
        [parameter_eliminate]: 2.37001e-06
        [updatestate_depend_eliminate]: 7e-06
        [updatestate_assign_eliminate]: 4.68001e-06
        [updatestate_loads_eliminate]: 4.23999e-06
        [cse]: 3.026e-05
        [call_graph_tuple_transform]: 2.453e-05
        [tuple_list_get_item_eliminator]: 8.57e-06
        [none_parameter_eliminate]: 1.34e-06
        [renormalize]: 4.10015e-07
        [switch_simplify]: 8.67e-06
[remove_dup_value]: 1.908e-05
[partial_unused_args_eliminate]: 2.00002e-06
[environ_conv]: 6.48e-06
[add_recomputation]: 5.064e-05
[cse_after_recomputation]: 3.314e-05, [1]
    [Cycle 1]: 2.717e-05, [1]
        [cse]: 2.137e-05
[auto_monad_reorder]: 1.815e-05
[get_jit_bprop_graph]: 1.55999e-06
[rewriter_after_jit_bprop_graph]: 5.66e-06
[opt_after_jit_grad]: 0.00045338
[symbol_engine_optimizer]: 9.023e-05, [1]
    [Cycle 1]: 8.382e-05, [6]
        [build]: 3.58e-06
        [elim_shapecalc]: 1.079e-05
        [elim_not_effective]: 1.866e-05
        [opt_reshape]: 8.92e-06
        [fold_const_symbol]: 1.423e-05
        [renormalize]: 5.39992e-07
[validate]: 4.14e-05
[backend_pass]: 8.59989e-07
[task_emit]: 0.00781912
[execute]: 6.78998e-06
Sums
    bootstrap                            :     0.000346s :  1.30%
    type_inference                       :     0.008886s : 33.31%
    event_method                         :     0.000041s :  0.15%
    auto_monad                           :     0.000069s :  0.26%
    graph_reusing                        :     0.000005s :  0.02%
    pre_auto_parallel                    :     0.000002s :  0.01%
    py_interpret_to_execute              :     0.000022s :  0.08%
    rewriter_before_opt_a                :     0.000081s :  0.30%
    expand_dump_flag                     :     0.000003s :  0.01%
    jit_opt_a.switch_simplify            :     0.000099s :  0.37%
    jit_opt_a.loop_unroll                :     0.000081s :  0.30%
    jit_opt_a.a_1                        :     0.002064s :  7.74%
    jit_opt_a.with_stream_mark           :     0.000040s :  0.15%
    jit_opt_a.recompute_prepare          :     0.000034s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000016s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000014s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000012s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.01%
    jit_opt_a.specialize_transform       :     0.000029s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000027s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000028s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000006s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000027s :  0.10%
    jit_opt_a.merge_forward              :     0.000014s :  0.05%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000057s :  0.21%
    jit_opt_a.j_node_and_user_rematch    :     0.000048s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.001282s :  4.81%
    jit_opt_a.replace_old_param          :     0.000067s :  0.25%
    jit_opt_a.inline_without_move        :     0.000064s :  0.24%
    jit_opt_a.renormalize                :     0.002954s : 11.07%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000007s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000073s :  0.28%
    jit_opt_a.cse                        :     0.000206s :  0.77%
    jit_opt_a.replace_applicator         :     0.000090s :  0.34%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.05%
    rewriter_after_opt_a                 :     0.000045s :  0.17%
    convert_after_rewriter               :     0.000008s :  0.03%
    order_py_execute_after_rewriter      :     0.000006s :  0.02%
    mutable_eliminate                    :     0.000679s :  2.54%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.09%
    jit_opt_b.inline_after_opt_a         :     0.000023s :  0.09%
    cconv                                :     0.000022s :  0.08%
    loop_unroll                          :     0.000408s :  1.53%
    jit_opt_after_cconv.c_1              :     0.000034s :  0.13%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000005s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000004s :  0.02%
    jit_opt_after_cconv.cse              :     0.000030s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000009s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000009s :  0.03%
    remove_dup_value                     :     0.000019s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000006s :  0.02%
    add_recomputation                    :     0.000051s :  0.19%
    cse_after_recomputation.cse          :     0.000021s :  0.08%
    auto_monad_reorder                   :     0.000018s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.02%
    opt_after_jit_grad                   :     0.000453s :  1.70%
    symbol_engine_optimizer.build        :     0.000004s :  0.01%
    symbol_engine_optimizer.elim_shapecalc :     0.000011s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000019s :  0.07%
    symbol_engine_optimizer.opt_reshape  :     0.000009s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000014s :  0.05%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.16%
    backend_pass                         :     0.000001s :  0.00%
    task_emit                            :     0.007819s : 29.31%
    execute                              :     0.000007s :  0.03%


Time group info:
------[substitution.]   0.000462   126
 7.48% :     0.000035s :      2: substitution.arithmetic_simplify
 0.70% :     0.000003s :      5: substitution.elim_not_effective
 0.45% :     0.000002s :      5: substitution.fold_const_symbol
 1.41% :     0.000007s :      6: substitution.graph_param_transform
57.69% :     0.000267s :     12: substitution.inline
 3.17% :     0.000015s :      2: substitution.inline_without_move
 1.94% :     0.000009s :     18: substitution.j_node_and_user_rematch
 2.27% :     0.000011s :      7: substitution.minmaximum_grad
 1.16% :     0.000005s :      5: substitution.partial_eliminate
 2.66% :     0.000012s :     18: substitution.remove_not_recompute_node
 5.13% :     0.000024s :     10: substitution.replace_applicator
 1.39% :     0.000006s :      7: substitution.replace_old_param
 0.42% :     0.000002s :      1: substitution.set_cell_output_no_recompute
 3.87% :     0.000018s :      7: substitution.tuple_list_convert_item_index_to_positive
 2.91% :     0.000013s :      7: substitution.tuple_list_get_item_depend_reorder
 7.35% :     0.000034s :     14: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008826     2
88.44% :     0.007805s :      1: type_inference.infer
11.56% :     0.001020s :      1: type_inference.specialize
------[replace.]   0.000130    19
63.31% :     0.000082s :     12: replace.inline
36.69% :     0.000048s :      7: replace.tuple_list_get_item_eliminator
------[match.]   0.000276    19
94.08% :     0.000260s :     12: match.inline
 5.92% :     0.000016s :      7: match.tuple_list_get_item_eliminator
------[predicate.]   0.000397  2895
 1.65% :     0.000007s :     48: predicate.accumulaten_eliminater
 0.61% :     0.000002s :      6: predicate.ad_related_special_op_eliminate
 1.44% :     0.000006s :     48: predicate.addn_check_dump
 1.73% :     0.000007s :     48: predicate.addn_zero_filter
 2.07% :     0.000008s :     48: predicate.arithmetic_simplify
 1.50% :     0.000006s :     48: predicate.cast_eliminate
 0.23% :     0.000001s :      6: predicate.check_bprop_eliminate
 1.41% :     0.000006s :     48: predicate.compare_switch_simplify
 1.55% :     0.000006s :     48: predicate.depend_value_elim
 1.50% :     0.000006s :     48: predicate.dict_get_item_const_eliminator
 1.50% :     0.000006s :     48: predicate.dict_get_item_eliminator
 1.59% :     0.000006s :     48: predicate.dict_set_item_eliminator
 0.38% :     0.000001s :      6: predicate.dumpgradient_eliminate
 0.14% :     0.000001s :      6: predicate.elim_not_effective
 0.25% :     0.000001s :      6: predicate.elim_shapecalc_of_broadcastargs
 1.54% :     0.000006s :     48: predicate.environ_add_const_eliminate
 1.46% :     0.000006s :     48: predicate.environ_get_add_eliminate
 1.47% :     0.000006s :     48: predicate.environ_get_depend_swap
 1.62% :     0.000006s :     48: predicate.environ_get_eliminate
 1.55% :     0.000006s :     48: predicate.environ_get_set_eliminate
 0.12% :     0.000000s :      6: predicate.fold_const_symbol
 0.89% :     0.000004s :     23: predicate.get_grad_eliminate
 0.12% :     0.000000s :      6: predicate.graph_param_transform
 4.17% :     0.000017s :     79: predicate.inline
 1.73% :     0.000007s :     44: predicate.inline_without_move
 0.40% :     0.000002s :     23: predicate.j_node_and_user_rematch
 0.96% :     0.000004s :     23: predicate.less_batch_normalization
 1.85% :     0.000007s :     55: predicate.list_to_tuple_eliminator_
 2.04% :     0.000008s :     61: predicate.load_eliminater
 0.54% :     0.000002s :      6: predicate.loop_unroll_after_grad
 3.02% :     0.000012s :     90: predicate.loop_unroll_before_grad
 1.81% :     0.000007s :     54: predicate.make_slice_get_slice_eliminator
 1.47% :     0.000006s :     48: predicate.merge_addn
 1.48% :     0.000006s :     48: predicate.minmaximum_grad
 0.55% :     0.000002s :      6: predicate.mutable_eliminate
 0.25% :     0.000001s :      6: predicate.opt_reshape
 2.35% :     0.000009s :     61: predicate.partial_eliminate
 1.52% :     0.000006s :     48: predicate.print_const_string_wrapper
 1.84% :     0.000007s :     48: predicate.reduce_eliminate
 1.77% :     0.000007s :     55: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000002s :     23: predicate.remove_not_recompute_node
 2.54% :     0.000010s :    108: predicate.replace_applicator
 0.90% :     0.000004s :     44: predicate.replace_old_param
 0.16% :     0.000001s :      6: predicate.reset_defer_inline
 1.77% :     0.000007s :     48: predicate.reshape_eliminate
 1.48% :     0.000006s :     48: predicate.row_tensor_add_zeros_like
 0.36% :     0.000001s :      6: predicate.row_tensor_eliminate
 1.55% :     0.000006s :     48: predicate.same_eliminate
 0.49% :     0.000002s :     23: predicate.set_cell_output_no_recompute
 0.52% :     0.000002s :     12: predicate.special_op_eliminate
 0.95% :     0.000004s :     23: predicate.specialize_transform
 1.67% :     0.000007s :     48: predicate.split_environ_get_set_with_tuple_value
 1.51% :     0.000006s :     48: predicate.stack_unstack_eliminate
 0.23% :     0.000001s :      6: predicate.switch_call_monad_eliminater
 2.57% :     0.000010s :     67: predicate.switch_defer_inline
 2.49% :     0.000010s :     67: predicate.switch_layer_defer_inline
 5.95% :     0.000024s :    163: predicate.switch_simplify
 1.63% :     0.000006s :     48: predicate.tile_eliminate
 1.50% :     0.000006s :     48: predicate.transpose_eliminate
 1.79% :     0.000007s :     48: predicate.tuple_list_convert_item_index_to_positive
 1.60% :     0.000006s :     48: predicate.tuple_list_get_item_depend_reorder
 2.93% :     0.000012s :     67: predicate.tuple_list_get_item_eliminator
 1.75% :     0.000007s :     48: predicate.tuple_list_set_item_eliminator
 1.82% :     0.000007s :     55: predicate.tuple_to_list_eliminator_
 2.05% :     0.000008s :     61: predicate.updatestate_pure_node_eliminater
 3.00% :     0.000012s :     84: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000007s :     48: predicate.value_based_eliminate
 0.21% :     0.000001s :      6: predicate.virtual_view_grad_eliminate
 0.27% :     0.000001s :      6: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.001402    29
55.00% :     0.000771s :     13: func_graph_cloner_run.FuncGraphClonerGraph
45.00% :     0.000631s :     16: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.036166    91
 0.15% :     0.000054s :      1: add_recomputation
 0.20% :     0.000073s :      1: auto_monad
 0.06% :     0.000021s :      1: auto_monad_reorder
 0.01% :     0.000004s :      1: backend_pass
 0.99% :     0.000357s :      1: bootstrap
 0.07% :     0.000024s :      1: cconv
 0.03% :     0.000011s :      1: convert_after_rewriter
 0.10% :     0.000035s :      1: cse_after_recomputation
 0.02% :     0.000009s :      1: environ_conv
 0.12% :     0.000045s :      1: event_method
 0.03% :     0.000010s :      1: execute
 0.02% :     0.000006s :      1: expand_dump_flag
 0.01% :     0.000004s :      1: get_jit_bprop_graph
 0.02% :     0.000008s :      1: graph_reusing
29.69% :     0.010739s :      1: jit_opt_a
 0.51% :     0.000186s :      1: jit_opt_after_cconv
 0.19% :     0.000068s :      1: jit_opt_b
 1.15% :     0.000416s :      1: loop_unroll
 1.90% :     0.000687s :      1: mutable_eliminate
 7.41% :     0.002679s :     39: opt.transform.jit_opt_a
 0.20% :     0.000072s :      4: opt.transform.jit_opt_after_cconv
 0.11% :     0.000041s :      4: opt.transform.jit_opt_b
 0.04% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.09% :     0.000031s :      1: opt.transform.opt_after_jit_grad
 0.14% :     0.000049s :      4: opt.transform.symbol_engine_opt
 1.28% :     0.000461s :      1: opt_after_jit_grad
 0.02% :     0.000009s :      1: order_py_execute_after_rewriter
 0.01% :     0.000004s :      1: partial_unused_args_eliminate
 0.01% :     0.000004s :      1: pre_auto_parallel
 0.07% :     0.000024s :      1: py_interpret_to_execute
 0.05% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000022s :      1: remove_dup_value
 4.22% :     0.001526s :      2: renormalize.infer
 3.92% :     0.001417s :      2: renormalize.specialize
 0.02% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.13% :     0.000048s :      1: rewriter_after_opt_a
 0.23% :     0.000084s :      1: rewriter_before_opt_a
 0.26% :     0.000092s :      1: symbol_engine_optimizer
21.65% :     0.007830s :      1: task_emit
24.60% :     0.008895s :      1: type_inference
 0.18% :     0.000065s :      1: validate


TotalTime = 0.0117128, [30]
[bootstrap]: 0.00028398
[type_inference]: 0.00413736
[event_method]: 1.681e-05
[auto_monad]: 6.272e-05
[graph_reusing]: 5.44e-06
[pre_auto_parallel]: 2.31e-06
[py_interpret_to_execute]: 3.371e-05
[rewriter_before_opt_a]: 7.055e-05
[expand_dump_flag]: 2.91e-06
[jit_opt_a]: 0.00501866, [2]
    [Cycle 1]: 0.00147154, [27]
        [switch_simplify]: 4.792e-05
        [loop_unroll]: 3.017e-05
        [a_1]: 0.0005119
        [with_stream_mark]: 1.449e-05
        [recompute_prepare]: 7.16001e-06
        [updatestate_depend_eliminate]: 3.00002e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [parameter_eliminate]: 1.45001e-06
        [specialize_transform]: 5.75001e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.95002e-06
        [meta_shard_fg_expand]: 1.89999e-06
        [get_grad_eliminate_]: 5.20999e-06
        [merge_forward]: 3.28e-06
        [cell_reuse_recompute_pass]: 8.89995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.423e-05
        [j_node_and_user_rematch]: 9.22999e-06
        [meta_fg_expand]: 2.59001e-06
        [replace_old_param]: 8.66002e-06
        [inline_without_move]: 5.19e-06
        [renormalize]: 0.00057223
        [add_forward_monad_depend]: 5.59998e-06
        [auto_monad_grad]: 1.87999e-06
        [auto_monad_eliminator]: 1.227e-05
        [cse]: 1.809e-05
        [replace_applicator]: 1.156e-05
    [Cycle 2]: 0.00033043, [27]
        [switch_simplify]: 6.21e-06
        [loop_unroll]: 5.30999e-06
        [a_1]: 9.499e-05
        [with_stream_mark]: 8.99e-06
        [recompute_prepare]: 4.85001e-06
        [updatestate_depend_eliminate]: 2.49999e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [parameter_eliminate]: 1.34e-06
        [specialize_transform]: 4.82998e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 5.24e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.4e-06
        [merge_forward]: 2.49999e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.197e-05
        [j_node_and_user_rematch]: 7.82e-06
        [meta_fg_expand]: 1.81e-06
        [replace_old_param]: 1.505e-05
        [inline_without_move]: 4.80001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 5.29e-06
        [cse]: 1.047e-05
        [replace_applicator]: 5.19998e-06
[py_interpret_to_execute_after_opt_a]: 1.195e-05
[rewriter_after_opt_a]: 2.902e-05
[convert_after_rewriter]: 7.13e-06
[order_py_execute_after_rewriter]: 5.49998e-06
[mutable_eliminate]: 0.00055696
[jit_opt_b]: 5.113e-05, [1]
    [Cycle 1]: 4.467e-05, [2]
        [frontend_op_eliminate]: 1.724e-05
        [inline_after_opt_a]: 1.541e-05
[cconv]: 2.395e-05
[loop_unroll]: 0.000425
[jit_opt_after_cconv]: 0.00014256, [1]
    [Cycle 1]: 0.00013588, [11]
        [c_1]: 2.198e-05
        [parameter_eliminate]: 3.03e-06
        [updatestate_depend_eliminate]: 5.45001e-06
        [updatestate_assign_eliminate]: 2.28002e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [cse]: 1.894e-05
        [call_graph_tuple_transform]: 1.834e-05
        [tuple_list_get_item_eliminator]: 5.62001e-06
        [none_parameter_eliminate]: 1.62001e-06
        [renormalize]: 6.09987e-07
        [switch_simplify]: 5.39998e-06
[remove_dup_value]: 1.314e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 4.26001e-06
[add_recomputation]: 4.326e-05
[cse_after_recomputation]: 2.018e-05, [1]
    [Cycle 1]: 1.506e-05, [1]
        [cse]: 9.16002e-06
[auto_monad_reorder]: 1.444e-05
[get_jit_bprop_graph]: 1.47999e-06
[rewriter_after_jit_bprop_graph]: 4.65001e-06
[opt_after_jit_grad]: 0.00045951
[symbol_engine_optimizer]: 7.346e-05, [1]
    [Cycle 1]: 6.653e-05, [6]
        [build]: 3.23e-06
        [elim_shapecalc]: 7.63999e-06
        [elim_not_effective]: 1.281e-05
        [opt_reshape]: 6.06998e-06
        [fold_const_symbol]: 8.81002e-06
        [renormalize]: 6.00005e-07
[validate]: 3.228e-05
Sums
    bootstrap                            :     0.000284s :  3.57%
    type_inference                       :     0.004137s : 52.08%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000063s :  0.79%
    graph_reusing                        :     0.000005s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.42%
    rewriter_before_opt_a                :     0.000071s :  0.89%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000054s :  0.68%
    jit_opt_a.loop_unroll                :     0.000035s :  0.45%
    jit_opt_a.a_1                        :     0.000607s :  7.64%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.06%
    jit_opt_a.replace_old_param          :     0.000024s :  0.30%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000572s :  7.20%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.22%
    jit_opt_a.cse                        :     0.000029s :  0.36%
    jit_opt_a.replace_applicator         :     0.000017s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.15%
    rewriter_after_opt_a                 :     0.000029s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000557s :  7.01%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000024s :  0.30%
    loop_unroll                          :     0.000425s :  5.35%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.28%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000019s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000013s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000043s :  0.54%
    cse_after_recomputation.cse          :     0.000009s :  0.12%
    auto_monad_reorder                   :     0.000014s :  0.18%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000460s :  5.78%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000032s :  0.41%


Time group info:
------[substitution.]   0.000183    23
 1.07% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.77% :     0.000005s :      3: substitution.graph_param_transform
82.92% :     0.000152s :      4: substitution.inline
 1.80% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.49% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.88% :     0.000003s :      2: substitution.replace_old_param
 6.38% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004084     2
79.75% :     0.003257s :      1: type_inference.infer
20.25% :     0.000827s :      1: type_inference.specialize
------[replace.]   0.000065     6
67.39% :     0.000044s :      4: replace.inline
32.61% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000160     6
93.39% :     0.000149s :      4: match.inline
 6.61% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.27% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.36% :     0.000002s :     10: predicate.addn_zero_filter
 2.23% :     0.000003s :     10: predicate.arithmetic_simplify
 1.26% :     0.000001s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.38% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.83% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.61% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.19% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 4.99% :     0.000006s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.89% :     0.000002s :     15: predicate.load_eliminater
 1.23% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.17% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.07% :     0.000001s :     10: predicate.merge_addn
 1.07% :     0.000001s :     10: predicate.minmaximum_grad
 1.57% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.83% :     0.000002s :     10: predicate.reduce_eliminate
 1.57% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.25% :     0.000000s :      3: predicate.reset_defer_inline
 1.20% :     0.000001s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.20% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.55% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.10% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.50% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.67% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.68% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.45% :     0.000002s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.18% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.87% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.74% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.40% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.70% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000572    12
48.37% :     0.000277s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.86% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.77% :     0.000268s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013215    72
 0.35% :     0.000046s :      1: add_recomputation
 0.50% :     0.000066s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.24% :     0.000296s :      1: bootstrap
 0.20% :     0.000026s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.00% :     0.005021s :      1: jit_opt_a
 1.10% :     0.000145s :      1: jit_opt_after_cconv
 0.41% :     0.000054s :      1: jit_opt_b
 3.27% :     0.000432s :      1: loop_unroll
 4.28% :     0.000566s :      1: mutable_eliminate
 6.18% :     0.000817s :     26: opt.transform.jit_opt_a
 0.36% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.48% :     0.000063s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.54% :     0.000468s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000036s :      1: py_interpret_to_execute
 0.11% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000015s :      1: remove_dup_value
 2.05% :     0.000271s :      1: renormalize.infer
 2.22% :     0.000294s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000032s :      1: rewriter_after_opt_a
 0.56% :     0.000073s :      1: rewriter_before_opt_a
 0.58% :     0.000076s :      1: symbol_engine_optimizer
31.40% :     0.004149s :      1: type_inference


TotalTime = 0.018253, [30]
[bootstrap]: 0.0001637
[type_inference]: 0.00846818
[event_method]: 3.894e-05
[auto_monad]: 8.005e-05
[graph_reusing]: 5.84999e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 2.274e-05
[rewriter_before_opt_a]: 8.424e-05
[expand_dump_flag]: 2.63e-06
[jit_opt_a]: 0.00750473, [2]
    [Cycle 1]: 0.00420823, [27]
        [switch_simplify]: 5.391e-05
        [loop_unroll]: 3.719e-05
        [a_1]: 0.00077833
        [with_stream_mark]: 2.086e-05
        [recompute_prepare]: 1.745e-05
        [updatestate_depend_eliminate]: 6.78e-06
        [updatestate_assign_eliminate]: 5.64e-06
        [updatestate_loads_eliminate]: 5.42001e-06
        [parameter_eliminate]: 2.14e-06
        [specialize_transform]: 1.283e-05
        [updatestate_useless_node_eliminater]: 1.138e-05
        [accelerated_algorithm]: 1.234e-05
        [meta_shard_fg_expand]: 2.37001e-06
        [get_grad_eliminate_]: 1.162e-05
        [merge_forward]: 6.89001e-06
        [cell_reuse_recompute_pass]: 7.00005e-07
        [cell_reuse_handle_not_recompute_node_pass]: 2.341e-05
        [j_node_and_user_rematch]: 2.192e-05
        [meta_fg_expand]: 0.0014297
        [replace_old_param]: 4.85e-05
        [inline_without_move]: 4.918e-05
        [renormalize]: 0.00141468
        [add_forward_monad_depend]: 5.64e-06
        [auto_monad_grad]: 1.62999e-06
        [auto_monad_eliminator]: 1.03e-05
        [cse]: 1.619e-05
        [replace_applicator]: 1.024e-05
    [Cycle 2]: 0.00026243, [27]
        [switch_simplify]: 4.57e-06
        [loop_unroll]: 3.47002e-06
        [a_1]: 3.769e-05
        [with_stream_mark]: 8.13001e-06
        [recompute_prepare]: 3.6e-06
        [updatestate_depend_eliminate]: 2.34001e-06
        [updatestate_assign_eliminate]: 1.69e-06
        [updatestate_loads_eliminate]: 1.51002e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 3.76999e-06
        [updatestate_useless_node_eliminater]: 3.50998e-06
        [accelerated_algorithm]: 3.66999e-06
        [meta_shard_fg_expand]: 1.08001e-06
        [get_grad_eliminate_]: 3.24001e-06
        [merge_forward]: 1.94e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 9.52999e-06
        [j_node_and_user_rematch]: 5.35999e-06
        [meta_fg_expand]: 2.374e-05
        [replace_old_param]: 3.68999e-06
        [inline_without_move]: 3.06999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.30001e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 4.34002e-06
        [cse]: 9.59e-06
        [replace_applicator]: 3.66999e-06
[py_interpret_to_execute_after_opt_a]: 8.60001e-06
[rewriter_after_opt_a]: 3.032e-05
[convert_after_rewriter]: 5.84e-06
[order_py_execute_after_rewriter]: 4.43001e-06
[mutable_eliminate]: 0.00048556
[jit_opt_b]: 3.867e-05, [1]
    [Cycle 1]: 3.195e-05, [2]
        [frontend_op_eliminate]: 9.82999e-06
        [inline_after_opt_a]: 1.092e-05
[cconv]: 1.821e-05
[loop_unroll]: 0.00042441
[jit_opt_after_cconv]: 0.00012003, [1]
    [Cycle 1]: 0.00011374, [11]
        [c_1]: 1.249e-05
        [parameter_eliminate]: 2.63003e-06
        [updatestate_depend_eliminate]: 4.90001e-06
        [updatestate_assign_eliminate]: 1.94999e-06
        [updatestate_loads_eliminate]: 1.79e-06
        [cse]: 1.672e-05
        [call_graph_tuple_transform]: 1.284e-05
        [tuple_list_get_item_eliminator]: 3.66999e-06
        [none_parameter_eliminate]: 1.22e-06
        [renormalize]: 6.39993e-07
        [switch_simplify]: 3.86001e-06
[remove_dup_value]: 1.064e-05
[partial_unused_args_eliminate]: 2.09e-06
[environ_conv]: 4.07e-06
[add_recomputation]: 2.652e-05
[cse_after_recomputation]: 1.959e-05, [1]
    [Cycle 1]: 1.432e-05, [1]
        [cse]: 8.25e-06
[auto_monad_reorder]: 9.84001e-06
[get_jit_bprop_graph]: 1.56998e-06
[rewriter_after_jit_bprop_graph]: 5.87999e-06
[opt_after_jit_grad]: 0.00042951
[symbol_engine_optimizer]: 6.118e-05, [1]
    [Cycle 1]: 5.493e-05, [6]
        [build]: 3.02002e-06
        [elim_shapecalc]: 5.64e-06
        [elim_not_effective]: 9.14e-06
        [opt_reshape]: 4.00998e-06
        [fold_const_symbol]: 5.58002e-06
        [renormalize]: 4.39992e-07
[validate]: 2.449e-05
Sums
    bootstrap                            :     0.000164s :  1.12%
    type_inference                       :     0.008468s : 57.82%
    event_method                         :     0.000039s :  0.27%
    auto_monad                           :     0.000080s :  0.55%
    graph_reusing                        :     0.000006s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000023s :  0.16%
    rewriter_before_opt_a                :     0.000084s :  0.58%
    expand_dump_flag                     :     0.000003s :  0.02%
    jit_opt_a.switch_simplify            :     0.000058s :  0.40%
    jit_opt_a.loop_unroll                :     0.000041s :  0.28%
    jit_opt_a.a_1                        :     0.000816s :  5.57%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.20%
    jit_opt_a.recompute_prepare          :     0.000021s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000007s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.02%
    jit_opt_a.specialize_transform       :     0.000017s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000015s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000016s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.02%
    jit_opt_a.get_grad_eliminate_        :     0.000015s :  0.10%
    jit_opt_a.merge_forward              :     0.000009s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.01%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.22%
    jit_opt_a.j_node_and_user_rematch    :     0.000027s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.001453s :  9.92%
    jit_opt_a.replace_old_param          :     0.000052s :  0.36%
    jit_opt_a.inline_without_move        :     0.000052s :  0.36%
    jit_opt_a.renormalize                :     0.001415s :  9.66%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.05%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.10%
    jit_opt_a.cse                        :     0.000026s :  0.18%
    jit_opt_a.replace_applicator         :     0.000014s :  0.09%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.06%
    rewriter_after_opt_a                 :     0.000030s :  0.21%
    convert_after_rewriter               :     0.000006s :  0.04%
    order_py_execute_after_rewriter      :     0.000004s :  0.03%
    mutable_eliminate                    :     0.000486s :  3.32%
    jit_opt_b.frontend_op_eliminate      :     0.000010s :  0.07%
    jit_opt_b.inline_after_opt_a         :     0.000011s :  0.07%
    cconv                                :     0.000018s :  0.12%
    loop_unroll                          :     0.000424s :  2.90%
    jit_opt_after_cconv.c_1              :     0.000012s :  0.09%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.03%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.cse              :     0.000017s :  0.11%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000013s :  0.09%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000004s :  0.03%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000004s :  0.03%
    remove_dup_value                     :     0.000011s :  0.07%
    partial_unused_args_eliminate        :     0.000002s :  0.01%
    environ_conv                         :     0.000004s :  0.03%
    add_recomputation                    :     0.000027s :  0.18%
    cse_after_recomputation.cse          :     0.000008s :  0.06%
    auto_monad_reorder                   :     0.000010s :  0.07%
    get_jit_bprop_graph                  :     0.000002s :  0.01%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.04%
    opt_after_jit_grad                   :     0.000430s :  2.93%
    symbol_engine_optimizer.build        :     0.000003s :  0.02%
    symbol_engine_optimizer.elim_shapecalc :     0.000006s :  0.04%
    symbol_engine_optimizer.elim_not_effective :     0.000009s :  0.06%
    symbol_engine_optimizer.opt_reshape  :     0.000004s :  0.03%
    symbol_engine_optimizer.fold_const_symbol :     0.000006s :  0.04%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000024s :  0.17%


Time group info:
------[substitution.]   0.000209    50
 0.60% :     0.000001s :      1: substitution.elim_not_effective
 0.40% :     0.000001s :      1: substitution.fold_const_symbol
 1.55% :     0.000003s :      1: substitution.graph_param_transform
63.98% :     0.000134s :      6: substitution.inline
 7.19% :     0.000015s :      2: substitution.inline_without_move
 2.78% :     0.000006s :      9: substitution.j_node_and_user_rematch
 1.61% :     0.000003s :      2: substitution.minmaximum_grad
 1.79% :     0.000004s :      3: substitution.partial_eliminate
 3.19% :     0.000007s :      9: substitution.remove_not_recompute_node
 2.41% :     0.000005s :      2: substitution.replace_applicator
 1.95% :     0.000004s :      5: substitution.replace_old_param
 1.24% :     0.000003s :      1: substitution.set_cell_output_no_recompute
 3.26% :     0.000007s :      2: substitution.tuple_list_convert_item_index_to_positive
 2.11% :     0.000004s :      2: substitution.tuple_list_get_item_depend_reorder
 5.95% :     0.000012s :      4: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.008414     2
89.03% :     0.007491s :      1: type_inference.infer
10.97% :     0.000923s :      1: type_inference.specialize
------[replace.]   0.000056     8
77.65% :     0.000043s :      6: replace.inline
22.35% :     0.000012s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000139     8
94.05% :     0.000130s :      6: match.inline
 5.95% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000160  1109
 1.55% :     0.000002s :     18: predicate.accumulaten_eliminater
 0.53% :     0.000001s :      1: predicate.ad_related_special_op_eliminate
 1.35% :     0.000002s :     18: predicate.addn_check_dump
 1.53% :     0.000002s :     18: predicate.addn_zero_filter
 1.95% :     0.000003s :     18: predicate.arithmetic_simplify
 1.49% :     0.000002s :     18: predicate.cast_eliminate
 0.16% :     0.000000s :      1: predicate.check_bprop_eliminate
 1.27% :     0.000002s :     18: predicate.compare_switch_simplify
 1.52% :     0.000002s :     18: predicate.depend_value_elim
 1.33% :     0.000002s :     18: predicate.dict_get_item_const_eliminator
 1.49% :     0.000002s :     18: predicate.dict_get_item_eliminator
 1.43% :     0.000002s :     18: predicate.dict_set_item_eliminator
 0.44% :     0.000001s :      1: predicate.dumpgradient_eliminate
 0.11% :     0.000000s :      1: predicate.elim_not_effective
 0.20% :     0.000000s :      1: predicate.elim_shapecalc_of_broadcastargs
 1.44% :     0.000002s :     18: predicate.environ_add_const_eliminate
 1.39% :     0.000002s :     18: predicate.environ_get_add_eliminate
 1.33% :     0.000002s :     18: predicate.environ_get_depend_swap
 1.40% :     0.000002s :     18: predicate.environ_get_eliminate
 1.35% :     0.000002s :     18: predicate.environ_get_set_eliminate
 0.05% :     0.000000s :      1: predicate.fold_const_symbol
 1.17% :     0.000002s :     12: predicate.get_grad_eliminate
 0.09% :     0.000000s :      1: predicate.graph_param_transform
 4.32% :     0.000007s :     28: predicate.inline
 3.28% :     0.000005s :     33: predicate.inline_without_move
 0.54% :     0.000001s :     12: predicate.j_node_and_user_rematch
 1.44% :     0.000002s :     12: predicate.less_batch_normalization
 1.59% :     0.000003s :     20: predicate.list_to_tuple_eliminator_
 1.85% :     0.000003s :     21: predicate.load_eliminater
 0.94% :     0.000002s :      1: predicate.loop_unroll_after_grad
 3.63% :     0.000006s :     44: predicate.loop_unroll_before_grad
 1.78% :     0.000003s :     19: predicate.make_slice_get_slice_eliminator
 1.33% :     0.000002s :     18: predicate.merge_addn
 1.42% :     0.000002s :     18: predicate.minmaximum_grad
 0.80% :     0.000001s :      1: predicate.mutable_eliminate
 0.21% :     0.000000s :      1: predicate.opt_reshape
 2.02% :     0.000003s :     21: predicate.partial_eliminate
 1.44% :     0.000002s :     18: predicate.print_const_string_wrapper
 1.85% :     0.000003s :     18: predicate.reduce_eliminate
 1.59% :     0.000003s :     20: predicate.redundant_stop_gradient_eliminater
 0.64% :     0.000001s :     12: predicate.remove_not_recompute_node
 1.55% :     0.000002s :     22: predicate.replace_applicator
 1.69% :     0.000003s :     33: predicate.replace_old_param
 0.09% :     0.000000s :      1: predicate.reset_defer_inline
 1.75% :     0.000003s :     18: predicate.reshape_eliminate
 1.44% :     0.000002s :     18: predicate.row_tensor_add_zeros_like
 0.38% :     0.000001s :      1: predicate.row_tensor_eliminate
 1.55% :     0.000002s :     18: predicate.same_eliminate
 0.64% :     0.000001s :     12: predicate.set_cell_output_no_recompute
 0.42% :     0.000001s :      2: predicate.special_op_eliminate
 1.22% :     0.000002s :     12: predicate.specialize_transform
 1.59% :     0.000003s :     18: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     18: predicate.stack_unstack_eliminate
 0.16% :     0.000000s :      1: predicate.switch_call_monad_eliminater
 2.69% :     0.000004s :     26: predicate.switch_defer_inline
 2.40% :     0.000004s :     26: predicate.switch_layer_defer_inline
 6.74% :     0.000011s :     71: predicate.switch_simplify
 1.44% :     0.000002s :     18: predicate.tile_eliminate
 1.41% :     0.000002s :     18: predicate.transpose_eliminate
 1.75% :     0.000003s :     18: predicate.tuple_list_convert_item_index_to_positive
 1.49% :     0.000002s :     18: predicate.tuple_list_get_item_depend_reorder
 2.71% :     0.000004s :     22: predicate.tuple_list_get_item_eliminator
 1.63% :     0.000003s :     18: predicate.tuple_list_set_item_eliminator
 1.70% :     0.000003s :     20: predicate.tuple_to_list_eliminator_
 1.72% :     0.000003s :     21: predicate.updatestate_pure_node_eliminater
 3.09% :     0.000005s :     33: predicate.updatestate_useless_node_eliminater
 1.77% :     0.000003s :     18: predicate.value_based_eliminate
 0.13% :     0.000000s :      1: predicate.virtual_view_grad_eliminate
 0.22% :     0.000000s :      1: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000998    22
73.04% :     0.000729s :     13: func_graph_cloner_run.FuncGraphClonerGraph
26.96% :     0.000269s :      9: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.020847    72
 0.14% :     0.000029s :      1: add_recomputation
 0.40% :     0.000084s :      1: auto_monad
 0.06% :     0.000013s :      1: auto_monad_reorder
 0.83% :     0.000173s :      1: bootstrap
 0.10% :     0.000021s :      1: cconv
 0.04% :     0.000008s :      1: convert_after_rewriter
 0.10% :     0.000022s :      1: cse_after_recomputation
 0.03% :     0.000006s :      1: environ_conv
 0.21% :     0.000043s :      1: event_method
 0.02% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.04% :     0.000008s :      1: graph_reusing
36.01% :     0.007507s :      1: jit_opt_a
 0.59% :     0.000123s :      1: jit_opt_after_cconv
 0.20% :     0.000041s :      1: jit_opt_b
 2.08% :     0.000433s :      1: loop_unroll
 2.37% :     0.000494s :      1: mutable_eliminate
 5.53% :     0.001152s :     26: opt.transform.jit_opt_a
 0.14% :     0.000029s :      4: opt.transform.jit_opt_after_cconv
 0.07% :     0.000015s :      4: opt.transform.jit_opt_b
 0.06% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.05% :     0.000011s :      1: opt.transform.mutable_eliminate
 0.07% :     0.000014s :      1: opt.transform.opt_after_jit_grad
 0.10% :     0.000021s :      4: opt.transform.symbol_engine_opt
 2.10% :     0.000437s :      1: opt_after_jit_grad
 0.03% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.12% :     0.000025s :      1: py_interpret_to_execute
 0.05% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.06% :     0.000013s :      1: remove_dup_value
 5.14% :     0.001071s :      1: renormalize.infer
 1.61% :     0.000336s :      1: renormalize.specialize
 0.04% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.16% :     0.000033s :      1: rewriter_after_opt_a
 0.42% :     0.000087s :      1: rewriter_before_opt_a
 0.31% :     0.000064s :      1: symbol_engine_optimizer
40.67% :     0.008479s :      1: type_inference


TotalTime = 0.0118472, [30]
[bootstrap]: 0.00027624
[type_inference]: 0.00416196
[event_method]: 1.657e-05
[auto_monad]: 6.209e-05
[graph_reusing]: 5.81e-06
[pre_auto_parallel]: 2.21e-06
[py_interpret_to_execute]: 3.403e-05
[rewriter_before_opt_a]: 7.228e-05
[expand_dump_flag]: 2.71999e-06
[jit_opt_a]: 0.0052121, [2]
    [Cycle 1]: 0.00151823, [27]
        [switch_simplify]: 4.994e-05
        [loop_unroll]: 2.98e-05
        [a_1]: 0.00050947
        [with_stream_mark]: 1.446e-05
        [recompute_prepare]: 7.43999e-06
        [updatestate_depend_eliminate]: 3.09001e-06
        [updatestate_assign_eliminate]: 2.77002e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 6.51e-06
        [updatestate_useless_node_eliminater]: 5.15999e-06
        [accelerated_algorithm]: 5.34998e-06
        [meta_shard_fg_expand]: 1.92999e-06
        [get_grad_eliminate_]: 5.64e-06
        [merge_forward]: 3.81999e-06
        [cell_reuse_recompute_pass]: 1.02998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.416e-05
        [j_node_and_user_rematch]: 9.51e-06
        [meta_fg_expand]: 2.49001e-06
        [replace_old_param]: 8.55999e-06
        [inline_without_move]: 5.27001e-06
        [renormalize]: 0.0006105
        [add_forward_monad_depend]: 5.56002e-06
        [auto_monad_grad]: 1.71e-06
        [auto_monad_eliminator]: 1.251e-05
        [cse]: 2.321e-05
        [replace_applicator]: 1.096e-05
    [Cycle 2]: 0.00031957, [27]
        [switch_simplify]: 5.81e-06
        [loop_unroll]: 4.74998e-06
        [a_1]: 9.093e-05
        [with_stream_mark]: 8.73001e-06
        [recompute_prepare]: 5.15999e-06
        [updatestate_depend_eliminate]: 2.49999e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 1.02e-06
        [specialize_transform]: 5.14e-06
        [updatestate_useless_node_eliminater]: 4.79e-06
        [accelerated_algorithm]: 4.95001e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 4.63001e-06
        [merge_forward]: 2.27001e-06
        [cell_reuse_recompute_pass]: 1.18001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.18e-05
        [j_node_and_user_rematch]: 7.65998e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 6.93e-06
        [inline_without_move]: 4.80999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.07e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 4.60999e-06
        [cse]: 1.016e-05
        [replace_applicator]: 4.92999e-06
[py_interpret_to_execute_after_opt_a]: 8.87e-06
[rewriter_after_opt_a]: 3.098e-05
[convert_after_rewriter]: 6.75998e-06
[order_py_execute_after_rewriter]: 5.49e-06
[mutable_eliminate]: 0.00048241
[jit_opt_b]: 4.898e-05, [1]
    [Cycle 1]: 4.215e-05, [2]
        [frontend_op_eliminate]: 1.613e-05
        [inline_after_opt_a]: 1.482e-05
[cconv]: 2.434e-05
[loop_unroll]: 0.00043491
[jit_opt_after_cconv]: 0.0001354, [1]
    [Cycle 1]: 0.00012877, [11]
        [c_1]: 2.057e-05
        [parameter_eliminate]: 2.48e-06
        [updatestate_depend_eliminate]: 5.19e-06
        [updatestate_assign_eliminate]: 2.41e-06
        [updatestate_loads_eliminate]: 2.09e-06
        [cse]: 1.609e-05
        [call_graph_tuple_transform]: 1.837e-05
        [tuple_list_get_item_eliminator]: 5.27999e-06
        [none_parameter_eliminate]: 1.68997e-06
        [renormalize]: 4.40021e-07
        [switch_simplify]: 5.39998e-06
[remove_dup_value]: 1.364e-05
[partial_unused_args_eliminate]: 2.51e-06
[environ_conv]: 4.70001e-06
[add_recomputation]: 4.444e-05
[cse_after_recomputation]: 1.953e-05, [1]
    [Cycle 1]: 1.415e-05, [1]
        [cse]: 8.59e-06
[auto_monad_reorder]: 1.462e-05
[get_jit_bprop_graph]: 1.40999e-06
[rewriter_after_jit_bprop_graph]: 4.83001e-06
[opt_after_jit_grad]: 0.00045043
[symbol_engine_optimizer]: 7.222e-05, [1]
    [Cycle 1]: 6.579e-05, [6]
        [build]: 3.04999e-06
        [elim_shapecalc]: 7.88001e-06
        [elim_not_effective]: 1.243e-05
        [opt_reshape]: 5.59998e-06
        [fold_const_symbol]: 8.74e-06
        [renormalize]: 5.00004e-07
[validate]: 3.182e-05
Sums
    bootstrap                            :     0.000276s :  3.49%
    type_inference                       :     0.004162s : 52.61%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000062s :  0.78%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.43%
    rewriter_before_opt_a                :     0.000072s :  0.91%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000056s :  0.70%
    jit_opt_a.loop_unroll                :     0.000035s :  0.44%
    jit_opt_a.a_1                        :     0.000600s :  7.59%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.15%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.06%
    jit_opt_a.replace_old_param          :     0.000015s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000611s :  7.72%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.22%
    jit_opt_a.cse                        :     0.000033s :  0.42%
    jit_opt_a.replace_applicator         :     0.000016s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000031s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000482s :  6.10%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000024s :  0.31%
    loop_unroll                          :     0.000435s :  5.50%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000044s :  0.56%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.18%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000450s :  5.69%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000032s :  0.40%


Time group info:
------[substitution.]   0.000183    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000001s :      2: substitution.fold_const_symbol
 2.83% :     0.000005s :      3: substitution.graph_param_transform
83.29% :     0.000153s :      4: substitution.inline
 1.51% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.36% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.16% :     0.000004s :      2: substitution.replace_old_param
 6.15% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004106     2
79.63% :     0.003270s :      1: type_inference.infer
20.37% :     0.000837s :      1: type_inference.specialize
------[replace.]   0.000066     6
68.27% :     0.000045s :      4: replace.inline
31.73% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000160     6
93.66% :     0.000150s :      4: match.inline
 6.34% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.37% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.27% :     0.000001s :     10: predicate.addn_zero_filter
 1.87% :     0.000002s :     10: predicate.arithmetic_simplify
 1.40% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.28% :     0.000001s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.19% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.27% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000000s :      3: predicate.graph_param_transform
 5.65% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.41% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.12% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.36% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.31% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.73% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.32% :     0.000001s :     10: predicate.reshape_eliminate
 1.42% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.44% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.50% :     0.000003s :     16: predicate.switch_defer_inline
 2.19% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.59% :     0.000009s :     53: predicate.switch_simplify
 1.17% :     0.000001s :     10: predicate.tile_eliminate
 1.19% :     0.000001s :     10: predicate.transpose_eliminate
 1.51% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.42% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.06% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.68% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.08% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.64% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000585    12
49.70% :     0.000291s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.86% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.44% :     0.000266s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013314    72
 0.36% :     0.000047s :      1: add_recomputation
 0.49% :     0.000065s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.17% :     0.000289s :      1: bootstrap
 0.20% :     0.000027s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
39.17% :     0.005216s :      1: jit_opt_a
 1.04% :     0.000138s :      1: jit_opt_after_cconv
 0.39% :     0.000052s :      1: jit_opt_b
 3.33% :     0.000443s :      1: loop_unroll
 3.69% :     0.000491s :      1: mutable_eliminate
 6.02% :     0.000802s :     26: opt.transform.jit_opt_a
 0.35% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.44% :     0.000458s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.28% :     0.000037s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.40% :     0.000319s :      1: renormalize.infer
 2.14% :     0.000285s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000034s :      1: rewriter_after_opt_a
 0.56% :     0.000075s :      1: rewriter_before_opt_a
 0.56% :     0.000075s :      1: symbol_engine_optimizer
31.35% :     0.004174s :      1: type_inference


TotalTime = 0.0122529, [30]
[bootstrap]: 0.00030092
[type_inference]: 0.00424406
[event_method]: 1.807e-05
[auto_monad]: 6.483e-05
[graph_reusing]: 6.31e-06
[pre_auto_parallel]: 2.51e-06
[py_interpret_to_execute]: 3.363e-05
[rewriter_before_opt_a]: 7.765e-05
[expand_dump_flag]: 3.16999e-06
[jit_opt_a]: 0.00527459, [2]
    [Cycle 1]: 0.00155154, [27]
        [switch_simplify]: 5.142e-05
        [loop_unroll]: 2.935e-05
        [a_1]: 0.00050547
        [with_stream_mark]: 1.561e-05
        [recompute_prepare]: 6.78e-06
        [updatestate_depend_eliminate]: 3.46001e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.86999e-06
        [parameter_eliminate]: 1.75001e-06
        [specialize_transform]: 6.24001e-06
        [updatestate_useless_node_eliminater]: 5.56e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 5.28002e-06
        [merge_forward]: 3.73999e-06
        [cell_reuse_recompute_pass]: 1.05001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.402e-05
        [j_node_and_user_rematch]: 8.83001e-06
        [meta_fg_expand]: 2.14e-06
        [replace_old_param]: 8.66997e-06
        [inline_without_move]: 5.32999e-06
        [renormalize]: 0.00063333
        [add_forward_monad_depend]: 6.11998e-06
        [auto_monad_grad]: 2.48e-06
        [auto_monad_eliminator]: 1.348e-05
        [cse]: 2.598e-05
        [replace_applicator]: 1.319e-05
    [Cycle 2]: 0.00032917, [27]
        [switch_simplify]: 5.67999e-06
        [loop_unroll]: 5.17e-06
        [a_1]: 9.365e-05
        [with_stream_mark]: 9.38002e-06
        [recompute_prepare]: 5.03002e-06
        [updatestate_depend_eliminate]: 2.66999e-06
        [updatestate_assign_eliminate]: 2.06998e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 8.79983e-07
        [specialize_transform]: 4.94e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 5.09e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.55001e-06
        [merge_forward]: 2.57001e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.25e-05
        [j_node_and_user_rematch]: 8.27e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 7.40998e-06
        [inline_without_move]: 4.72e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 5.59998e-06
        [cse]: 1.092e-05
        [replace_applicator]: 5.00001e-06
[py_interpret_to_execute_after_opt_a]: 1.028e-05
[rewriter_after_opt_a]: 3.297e-05
[convert_after_rewriter]: 6.98e-06
[order_py_execute_after_rewriter]: 5.36002e-06
[mutable_eliminate]: 0.00055064
[jit_opt_b]: 5.181e-05, [1]
    [Cycle 1]: 4.394e-05, [2]
        [frontend_op_eliminate]: 1.665e-05
        [inline_after_opt_a]: 1.515e-05
[cconv]: 2.758e-05
[loop_unroll]: 0.00042703
[jit_opt_after_cconv]: 0.00018473, [1]
    [Cycle 1]: 0.00017789, [11]
        [c_1]: 2.118e-05
        [parameter_eliminate]: 3.28998e-06
        [updatestate_depend_eliminate]: 5.87001e-06
        [updatestate_assign_eliminate]: 2.71e-06
        [updatestate_loads_eliminate]: 2.16e-06
        [cse]: 1.954e-05
        [call_graph_tuple_transform]: 1.891e-05
        [tuple_list_get_item_eliminator]: 5.61e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 8.09989e-07
        [switch_simplify]: 4.572e-05
[remove_dup_value]: 1.517e-05
[partial_unused_args_eliminate]: 2.32001e-06
[environ_conv]: 5.96e-06
[add_recomputation]: 9.358e-05
[cse_after_recomputation]: 2.357e-05, [1]
    [Cycle 1]: 1.696e-05, [1]
        [cse]: 1.08e-05
[auto_monad_reorder]: 1.353e-05
[get_jit_bprop_graph]: 1.84e-06
[rewriter_after_jit_bprop_graph]: 6.33998e-06
[opt_after_jit_grad]: 0.00048134
[symbol_engine_optimizer]: 7.661e-05, [1]
    [Cycle 1]: 6.884e-05, [6]
        [build]: 3.61001e-06
        [elim_shapecalc]: 8e-06
        [elim_not_effective]: 1.355e-05
        [opt_reshape]: 5.86998e-06
        [fold_const_symbol]: 8.70001e-06
        [renormalize]: 8.09989e-07
[validate]: 3.648e-05
Sums
    bootstrap                            :     0.000301s :  3.64%
    type_inference                       :     0.004244s : 51.31%
    event_method                         :     0.000018s :  0.22%
    auto_monad                           :     0.000065s :  0.78%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.41%
    rewriter_before_opt_a                :     0.000078s :  0.94%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000057s :  0.69%
    jit_opt_a.loop_unroll                :     0.000035s :  0.42%
    jit_opt_a.a_1                        :     0.000599s :  7.24%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000633s :  7.66%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.23%
    jit_opt_a.cse                        :     0.000037s :  0.45%
    jit_opt_a.replace_applicator         :     0.000018s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000033s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000551s :  6.66%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000028s :  0.33%
    loop_unroll                          :     0.000427s :  5.16%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000020s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000046s :  0.55%
    remove_dup_value                     :     0.000015s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000094s :  1.13%
    cse_after_recomputation.cse          :     0.000011s :  0.13%
    auto_monad_reorder                   :     0.000014s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.08%
    opt_after_jit_grad                   :     0.000481s :  5.82%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000036s :  0.44%


Time group info:
------[substitution.]   0.000178    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.71% :     0.000001s :      2: substitution.fold_const_symbol
 2.96% :     0.000005s :      3: substitution.graph_param_transform
82.30% :     0.000147s :      4: substitution.inline
 1.95% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.45% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.21% :     0.000004s :      2: substitution.replace_old_param
 6.43% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004183     2
78.83% :     0.003297s :      1: type_inference.infer
21.17% :     0.000886s :      1: type_inference.specialize
------[replace.]   0.000063     6
64.33% :     0.000040s :      4: replace.inline
35.67% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000155     6
93.28% :     0.000144s :      4: match.inline
 6.72% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.45% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.42% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 1.95% :     0.000002s :     10: predicate.arithmetic_simplify
 1.37% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.85% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.52% :     0.000001s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.21% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.29% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.31% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.74% :     0.000002s :     15: predicate.load_eliminater
 1.83% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.20% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.88% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.55% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.35% :     0.000003s :     15: predicate.partial_eliminate
 1.23% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.67% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.76% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.41% :     0.000002s :     10: predicate.reshape_eliminate
 1.30% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.69% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.29% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.91% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.36% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.62% :     0.000003s :     16: predicate.switch_defer_inline
 2.17% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.41% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.40% :     0.000002s :     10: predicate.transpose_eliminate
 1.50% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.88% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.60% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000580    12
46.72% :     0.000271s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.87% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
48.42% :     0.000281s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013785    72
 0.71% :     0.000097s :      1: add_recomputation
 0.50% :     0.000069s :      1: auto_monad
 0.12% :     0.000016s :      1: auto_monad_reorder
 2.29% :     0.000315s :      1: bootstrap
 0.22% :     0.000030s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.19% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.16% :     0.000022s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
38.29% :     0.005278s :      1: jit_opt_a
 1.36% :     0.000188s :      1: jit_opt_after_cconv
 0.39% :     0.000054s :      1: jit_opt_b
 3.16% :     0.000436s :      1: loop_unroll
 4.06% :     0.000560s :      1: mutable_eliminate
 5.83% :     0.000804s :     26: opt.transform.jit_opt_a
 0.63% :     0.000087s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.55% :     0.000489s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000037s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000018s :      1: remove_dup_value
 2.26% :     0.000311s :      1: renormalize.infer
 2.28% :     0.000314s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000039s :      1: rewriter_after_opt_a
 0.59% :     0.000081s :      1: rewriter_before_opt_a
 0.57% :     0.000079s :      1: symbol_engine_optimizer
30.88% :     0.004256s :      1: type_inference


TotalTime = 0.013091, [30]
[bootstrap]: 0.00027213
[type_inference]: 0.00450325
[event_method]: 6.106e-05
[auto_monad]: 8.737e-05
[graph_reusing]: 5.28002e-06
[pre_auto_parallel]: 2.56998e-06
[py_interpret_to_execute]: 4.139e-05
[rewriter_before_opt_a]: 7.841e-05
[expand_dump_flag]: 3.50998e-06
[jit_opt_a]: 0.00579994, [2]
    [Cycle 1]: 0.00174705, [27]
        [switch_simplify]: 5.371e-05
        [loop_unroll]: 2.851e-05
        [a_1]: 0.00054511
        [with_stream_mark]: 1.892e-05
        [recompute_prepare]: 7.43e-06
        [updatestate_depend_eliminate]: 3.66999e-06
        [updatestate_assign_eliminate]: 3.13e-06
        [updatestate_loads_eliminate]: 2.31998e-06
        [parameter_eliminate]: 1.87001e-06
        [specialize_transform]: 6.11e-06
        [updatestate_useless_node_eliminater]: 5.25999e-06
        [accelerated_algorithm]: 5.57001e-06
        [meta_shard_fg_expand]: 2.69999e-06
        [get_grad_eliminate_]: 5.22e-06
        [merge_forward]: 3.51001e-06
        [cell_reuse_recompute_pass]: 1.28002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.497e-05
        [j_node_and_user_rematch]: 9.68997e-06
        [meta_fg_expand]: 2.47001e-06
        [replace_old_param]: 8.97e-06
        [inline_without_move]: 5.61003e-06
        [renormalize]: 0.00078705
        [add_forward_monad_depend]: 6.63e-06
        [auto_monad_grad]: 2.83e-06
        [auto_monad_eliminator]: 1.336e-05
        [cse]: 2.546e-05
        [replace_applicator]: 1.262e-05
    [Cycle 2]: 0.0003285, [27]
        [switch_simplify]: 5.57001e-06
        [loop_unroll]: 4.90001e-06
        [a_1]: 9.23e-05
        [with_stream_mark]: 9.86e-06
        [recompute_prepare]: 5.02999e-06
        [updatestate_depend_eliminate]: 2.67001e-06
        [updatestate_assign_eliminate]: 2.06e-06
        [updatestate_loads_eliminate]: 1.86003e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.14998e-06
        [updatestate_useless_node_eliminater]: 4.89003e-06
        [accelerated_algorithm]: 5.32999e-06
        [meta_shard_fg_expand]: 1.55999e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.27999e-06
        [cell_reuse_recompute_pass]: 1.25001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.314e-05
        [j_node_and_user_rematch]: 8.03999e-06
        [meta_fg_expand]: 1.81998e-06
        [replace_old_param]: 7.45e-06
        [inline_without_move]: 4.68001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.57001e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 4.88001e-06
        [cse]: 1.112e-05
        [replace_applicator]: 5.08002e-06
[py_interpret_to_execute_after_opt_a]: 1.175e-05
[rewriter_after_opt_a]: 3.348e-05
[convert_after_rewriter]: 6.48e-06
[order_py_execute_after_rewriter]: 5.53002e-06
[mutable_eliminate]: 0.00058907
[jit_opt_b]: 5.077e-05, [1]
    [Cycle 1]: 4.349e-05, [2]
        [frontend_op_eliminate]: 1.689e-05
        [inline_after_opt_a]: 1.551e-05
[cconv]: 2.6e-05
[loop_unroll]: 0.0004334
[jit_opt_after_cconv]: 0.00014086, [1]
    [Cycle 1]: 0.00013434, [11]
        [c_1]: 2.093e-05
        [parameter_eliminate]: 3.26001e-06
        [updatestate_depend_eliminate]: 5.81e-06
        [updatestate_assign_eliminate]: 2.79001e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 1.949e-05
        [call_graph_tuple_transform]: 1.871e-05
        [tuple_list_get_item_eliminator]: 5.48002e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 6.79982e-07
        [switch_simplify]: 5.29e-06
[remove_dup_value]: 1.382e-05
[partial_unused_args_eliminate]: 2.56998e-06
[environ_conv]: 5.91e-06
[add_recomputation]: 4.845e-05
[cse_after_recomputation]: 2.157e-05, [1]
    [Cycle 1]: 1.564e-05, [1]
        [cse]: 9.55001e-06
[auto_monad_reorder]: 1.592e-05
[get_jit_bprop_graph]: 2.24001e-06
[rewriter_after_jit_bprop_graph]: 5.03002e-06
[opt_after_jit_grad]: 0.00052063
[symbol_engine_optimizer]: 7.387e-05, [1]
    [Cycle 1]: 6.674e-05, [6]
        [build]: 3.34001e-06
        [elim_shapecalc]: 7.65e-06
        [elim_not_effective]: 1.3e-05
        [opt_reshape]: 5.57999e-06
        [fold_const_symbol]: 8.82999e-06
        [renormalize]: 4.39992e-07
[validate]: 3.655e-05
Sums
    bootstrap                            :     0.000272s :  3.10%
    type_inference                       :     0.004503s : 51.34%
    event_method                         :     0.000061s :  0.70%
    auto_monad                           :     0.000087s :  1.00%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000041s :  0.47%
    rewriter_before_opt_a                :     0.000078s :  0.89%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000059s :  0.68%
    jit_opt_a.loop_unroll                :     0.000033s :  0.38%
    jit_opt_a.a_1                        :     0.000637s :  7.27%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000787s :  8.97%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.21%
    jit_opt_a.cse                        :     0.000037s :  0.42%
    jit_opt_a.replace_applicator         :     0.000018s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.13%
    rewriter_after_opt_a                 :     0.000033s :  0.38%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000589s :  6.72%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.18%
    cconv                                :     0.000026s :  0.30%
    loop_unroll                          :     0.000433s :  4.94%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.16%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000048s :  0.55%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000521s :  5.94%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000037s :  0.42%


Time group info:
------[substitution.]   0.000214    23
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.32% :     0.000005s :      3: substitution.graph_param_transform
84.98% :     0.000182s :      4: substitution.inline
 1.50% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.42% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.89% :     0.000004s :      2: substitution.replace_old_param
 5.48% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004349     2
79.68% :     0.003465s :      1: type_inference.infer
20.32% :     0.000884s :      1: type_inference.specialize
------[replace.]   0.000067     6
67.47% :     0.000045s :      4: replace.inline
32.53% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000190     6
94.41% :     0.000180s :      4: match.inline
 5.59% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.32% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.48% :     0.000002s :     10: predicate.addn_zero_filter
 2.37% :     0.000003s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.43% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.97% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.38% :     0.000000s :      3: predicate.elim_not_effective
 0.77% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.28% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.25% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.87% :     0.000002s :     15: predicate.load_eliminater
 1.70% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.24% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 1.76% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.63% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.55% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.56% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.36% :     0.000002s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.46% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.70% :     0.000009s :     53: predicate.switch_simplify
 1.27% :     0.000001s :     10: predicate.tile_eliminate
 1.34% :     0.000002s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.24% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.69% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.58% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.76% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.63% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.66% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000611    12
46.34% :     0.000283s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.51% :     0.000046s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.14% :     0.000282s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014771    72
 0.35% :     0.000052s :      1: add_recomputation
 0.62% :     0.000092s :      1: auto_monad
 0.13% :     0.000019s :      1: auto_monad_reorder
 1.93% :     0.000285s :      1: bootstrap
 0.19% :     0.000029s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000024s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.45% :     0.000066s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.29% :     0.005803s :      1: jit_opt_a
 0.97% :     0.000144s :      1: jit_opt_after_cconv
 0.36% :     0.000053s :      1: jit_opt_b
 2.99% :     0.000442s :      1: loop_unroll
 4.05% :     0.000598s :      1: mutable_eliminate
 5.72% :     0.000845s :     26: opt.transform.jit_opt_a
 0.32% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.58% :     0.000529s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000044s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000016s :      1: remove_dup_value
 2.84% :     0.000420s :      1: renormalize.infer
 2.43% :     0.000358s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000036s :      1: rewriter_after_opt_a
 0.55% :     0.000081s :      1: rewriter_before_opt_a
 0.52% :     0.000076s :      1: symbol_engine_optimizer
30.70% :     0.004535s :      1: type_inference


TotalTime = 0.013531, [30]
[bootstrap]: 0.00034866
[type_inference]: 0.0044829
[event_method]: 1.652e-05
[auto_monad]: 6.295e-05
[graph_reusing]: 5.56002e-06
[pre_auto_parallel]: 2.31e-06
[py_interpret_to_execute]: 3.283e-05
[rewriter_before_opt_a]: 7.106e-05
[expand_dump_flag]: 3.08998e-06
[jit_opt_a]: 0.00580887, [2]
    [Cycle 1]: 0.00206209, [27]
        [switch_simplify]: 0.00010101
        [loop_unroll]: 0.00014147
        [a_1]: 0.00063449
        [with_stream_mark]: 2.142e-05
        [recompute_prepare]: 8.94e-06
        [updatestate_depend_eliminate]: 4.45e-06
        [updatestate_assign_eliminate]: 3.59002e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [parameter_eliminate]: 2.02999e-06
        [specialize_transform]: 6.02999e-06
        [updatestate_useless_node_eliminater]: 5.35001e-06
        [accelerated_algorithm]: 6.11e-06
        [meta_shard_fg_expand]: 2.50002e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 3.9e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.625e-05
        [j_node_and_user_rematch]: 9.20999e-06
        [meta_fg_expand]: 2.79001e-06
        [replace_old_param]: 1.059e-05
        [inline_without_move]: 5.59e-06
        [renormalize]: 0.00080251
        [add_forward_monad_depend]: 6.89001e-06
        [auto_monad_grad]: 2.60002e-06
        [auto_monad_eliminator]: 1.505e-05
        [cse]: 3.233e-05
        [replace_applicator]: 1.577e-05
    [Cycle 2]: 0.00035654, [27]
        [switch_simplify]: 5.94e-06
        [loop_unroll]: 6.07001e-06
        [a_1]: 9.427e-05
        [with_stream_mark]: 1.378e-05
        [recompute_prepare]: 5.39e-06
        [updatestate_depend_eliminate]: 3.3e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.43e-06
        [parameter_eliminate]: 1.18001e-06
        [specialize_transform]: 5.47001e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 5.37999e-06
        [meta_shard_fg_expand]: 1.94e-06
        [get_grad_eliminate_]: 4.53999e-06
        [merge_forward]: 3.2e-06
        [cell_reuse_recompute_pass]: 2.01e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.465e-05
        [j_node_and_user_rematch]: 7.87998e-06
        [meta_fg_expand]: 1.97999e-06
        [replace_old_param]: 8.47e-06
        [inline_without_move]: 5.00999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.65001e-06
        [auto_monad_grad]: 1.13001e-06
        [auto_monad_eliminator]: 5.56e-06
        [cse]: 1.139e-05
        [replace_applicator]: 5.09e-06
[py_interpret_to_execute_after_opt_a]: 1.155e-05
[rewriter_after_opt_a]: 3.642e-05
[convert_after_rewriter]: 7.23999e-06
[order_py_execute_after_rewriter]: 4.77998e-06
[mutable_eliminate]: 0.00064502
[jit_opt_b]: 5.079e-05, [1]
    [Cycle 1]: 4.382e-05, [2]
        [frontend_op_eliminate]: 1.655e-05
        [inline_after_opt_a]: 1.519e-05
[cconv]: 2.834e-05
[loop_unroll]: 0.00041848
[jit_opt_after_cconv]: 0.00022076, [1]
    [Cycle 1]: 0.00021377, [11]
        [c_1]: 2.162e-05
        [parameter_eliminate]: 2.43e-06
        [updatestate_depend_eliminate]: 5.64998e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.21e-06
        [cse]: 2.039e-05
        [call_graph_tuple_transform]: 4.999e-05
        [tuple_list_get_item_eliminator]: 5.81e-06
        [none_parameter_eliminate]: 2.071e-05
        [renormalize]: 6.89994e-07
        [switch_simplify]: 5.40999e-06
[remove_dup_value]: 1.46e-05
[partial_unused_args_eliminate]: 2.34001e-06
[environ_conv]: 2.241e-05
[add_recomputation]: 0.00012018
[cse_after_recomputation]: 4.044e-05, [1]
    [Cycle 1]: 3.428e-05, [1]
        [cse]: 9.81e-06
[auto_monad_reorder]: 4.818e-05
[get_jit_bprop_graph]: 1.84e-06
[rewriter_after_jit_bprop_graph]: 5.40001e-06
[opt_after_jit_grad]: 0.00071041
[symbol_engine_optimizer]: 7.446e-05, [1]
    [Cycle 1]: 6.777e-05, [6]
        [build]: 3.16999e-06
        [elim_shapecalc]: 7.76001e-06
        [elim_not_effective]: 1.349e-05
        [opt_reshape]: 5.76e-06
        [fold_const_symbol]: 9.14e-06
        [renormalize]: 5.3001e-07
[validate]: 3.704e-05
Sums
    bootstrap                            :     0.000349s :  3.69%
    type_inference                       :     0.004483s : 47.42%
    event_method                         :     0.000017s :  0.17%
    auto_monad                           :     0.000063s :  0.67%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000033s :  0.35%
    rewriter_before_opt_a                :     0.000071s :  0.75%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000107s :  1.13%
    jit_opt_a.loop_unroll                :     0.000148s :  1.56%
    jit_opt_a.a_1                        :     0.000729s :  7.71%
    jit_opt_a.with_stream_mark           :     0.000035s :  0.37%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000019s :  0.20%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000803s :  8.49%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.22%
    jit_opt_a.cse                        :     0.000044s :  0.46%
    jit_opt_a.replace_applicator         :     0.000021s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.12%
    rewriter_after_opt_a                 :     0.000036s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000645s :  6.82%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000028s :  0.30%
    loop_unroll                          :     0.000418s :  4.43%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000050s :  0.53%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000021s :  0.22%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000022s :  0.24%
    add_recomputation                    :     0.000120s :  1.27%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000048s :  0.51%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000710s :  7.51%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000037s :  0.39%


Time group info:
------[substitution.]   0.000245    23
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 2.38% :     0.000006s :      3: substitution.graph_param_transform
85.11% :     0.000208s :      4: substitution.inline
 1.18% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.22% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.62% :     0.000006s :      2: substitution.replace_old_param
 5.14% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004424     2
80.54% :     0.003563s :      1: type_inference.infer
19.46% :     0.000861s :      1: type_inference.specialize
------[replace.]   0.000083     6
66.90% :     0.000056s :      4: replace.inline
33.10% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000216     6
94.68% :     0.000205s :      4: match.inline
 5.32% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000138   691
 1.19% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.05% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.88% :     0.000001s :     10: predicate.addn_check_dump
 1.15% :     0.000002s :     10: predicate.addn_zero_filter
 1.74% :     0.000002s :     10: predicate.arithmetic_simplify
 1.11% :     0.000002s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.86% :     0.000001s :     10: predicate.compare_switch_simplify
 1.17% :     0.000002s :     10: predicate.depend_value_elim
 0.97% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.11% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.76% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.28% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.00% :     0.000001s :     10: predicate.environ_get_eliminate
 1.29% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
14.81% :     0.000020s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.35% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.60% :     0.000002s :     15: predicate.load_eliminater
 1.15% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.96% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.63% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 1.50% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 1.98% :     0.000003s :     15: predicate.partial_eliminate
 1.27% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.13% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.57% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.53% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.22% :     0.000000s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     10: predicate.reshape_eliminate
 1.41% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.86% :     0.000001s :      6: predicate.special_op_eliminate
 0.86% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.35% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.67% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 1.96% :     0.000003s :     16: predicate.switch_layer_defer_inline
 6.93% :     0.000010s :     53: predicate.switch_simplify
 1.34% :     0.000002s :     10: predicate.tile_eliminate
 1.04% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.30% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.26% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.36% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.81% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.26% :     0.000002s :     10: predicate.value_based_eliminate
 0.27% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000653    12
47.34% :     0.000309s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.23% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.43% :     0.000303s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015526    72
 0.79% :     0.000123s :      1: add_recomputation
 0.43% :     0.000067s :      1: auto_monad
 0.33% :     0.000051s :      1: auto_monad_reorder
 2.36% :     0.000367s :      1: bootstrap
 0.20% :     0.000031s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.28% :     0.000043s :      1: cse_after_recomputation
 0.16% :     0.000025s :      1: environ_conv
 0.13% :     0.000021s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
37.44% :     0.005812s :      1: jit_opt_a
 1.44% :     0.000224s :      1: jit_opt_after_cconv
 0.34% :     0.000053s :      1: jit_opt_b
 2.74% :     0.000426s :      1: loop_unroll
 4.21% :     0.000654s :      1: mutable_eliminate
 7.13% :     0.001107s :     26: opt.transform.jit_opt_a
 0.51% :     0.000079s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000033s :      4: opt.transform.symbol_engine_opt
 4.63% :     0.000719s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.23% :     0.000036s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.24% :     0.000038s :      1: remove_dup_value
 2.53% :     0.000392s :      1: renormalize.infer
 2.59% :     0.000402s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000040s :      1: rewriter_after_opt_a
 0.48% :     0.000074s :      1: rewriter_before_opt_a
 0.50% :     0.000077s :      1: symbol_engine_optimizer
28.95% :     0.004495s :      1: type_inference


TotalTime = 0.0143965, [30]
[bootstrap]: 0.00028992
[type_inference]: 0.00455021
[event_method]: 1.888e-05
[auto_monad]: 6.722e-05
[graph_reusing]: 5.26002e-06
[pre_auto_parallel]: 2.47001e-06
[py_interpret_to_execute]: 3.854e-05
[rewriter_before_opt_a]: 7.401e-05
[expand_dump_flag]: 2.98e-06
[jit_opt_a]: 0.00665367, [2]
    [Cycle 1]: 0.0022448, [27]
        [switch_simplify]: 6.327e-05
        [loop_unroll]: 3.006e-05
        [a_1]: 0.00064116
        [with_stream_mark]: 2.488e-05
        [recompute_prepare]: 9.44998e-06
        [updatestate_depend_eliminate]: 4.70001e-06
        [updatestate_assign_eliminate]: 3.20998e-06
        [updatestate_loads_eliminate]: 2.79001e-06
        [parameter_eliminate]: 2.06e-06
        [specialize_transform]: 7.2e-06
        [updatestate_useless_node_eliminater]: 6.22001e-06
        [accelerated_algorithm]: 5.66998e-06
        [meta_shard_fg_expand]: 2.80997e-06
        [get_grad_eliminate_]: 5.94e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.719e-05
        [j_node_and_user_rematch]: 1.001e-05
        [meta_fg_expand]: 2.46e-06
        [replace_old_param]: 1.187e-05
        [inline_without_move]: 5.66e-06
        [renormalize]: 0.00107717
        [add_forward_monad_depend]: 1.159e-05
        [auto_monad_grad]: 2.78e-06
        [auto_monad_eliminator]: 2.12e-05
        [cse]: 3.231e-05
        [replace_applicator]: 2.182e-05
    [Cycle 2]: 0.00042452, [27]
        [switch_simplify]: 6.54999e-06
        [loop_unroll]: 5.39e-06
        [a_1]: 0.00010056
        [with_stream_mark]: 1.806e-05
        [recompute_prepare]: 6.54001e-06
        [updatestate_depend_eliminate]: 3.91001e-06
        [updatestate_assign_eliminate]: 3.66001e-06
        [updatestate_loads_eliminate]: 2.51998e-06
        [parameter_eliminate]: 1.92999e-06
        [specialize_transform]: 6.38e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 6.07999e-06
        [meta_shard_fg_expand]: 2.53e-06
        [get_grad_eliminate_]: 4.76997e-06
        [merge_forward]: 4.17e-06
        [cell_reuse_recompute_pass]: 2.78e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.966e-05
        [j_node_and_user_rematch]: 9.81e-06
        [meta_fg_expand]: 2.41e-06
        [replace_old_param]: 9.89001e-06
        [inline_without_move]: 5.35999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 3.8e-06
        [auto_monad_grad]: 1.04e-06
        [auto_monad_eliminator]: 8.47e-06
        [cse]: 1.676e-05
        [replace_applicator]: 1.123e-05
[py_interpret_to_execute_after_opt_a]: 1.876e-05
[rewriter_after_opt_a]: 4.483e-05
[convert_after_rewriter]: 7.3e-06
[order_py_execute_after_rewriter]: 5.01002e-06
[mutable_eliminate]: 0.00079934
[jit_opt_b]: 5.863e-05, [1]
    [Cycle 1]: 4.961e-05, [2]
        [frontend_op_eliminate]: 1.96e-05
        [inline_after_opt_a]: 1.727e-05
[cconv]: 3.352e-05
[loop_unroll]: 0.0004972
[jit_opt_after_cconv]: 0.00017799, [1]
    [Cycle 1]: 0.00017044, [11]
        [c_1]: 2.404e-05
        [parameter_eliminate]: 6.97002e-06
        [updatestate_depend_eliminate]: 9.38002e-06
        [updatestate_assign_eliminate]: 3.61001e-06
        [updatestate_loads_eliminate]: 2.48998e-06
        [cse]: 2.893e-05
        [call_graph_tuple_transform]: 2.475e-05
        [tuple_list_get_item_eliminator]: 6.29001e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 9.30013e-07
        [switch_simplify]: 6.32001e-06
[remove_dup_value]: 1.594e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 6.23e-06
[add_recomputation]: 5.761e-05
[cse_after_recomputation]: 2.503e-05, [1]
    [Cycle 1]: 1.808e-05, [1]
        [cse]: 1.027e-05
[auto_monad_reorder]: 1.619e-05
[get_jit_bprop_graph]: 1.76e-06
[rewriter_after_jit_bprop_graph]: 1.11e-05
[opt_after_jit_grad]: 0.00056944
[symbol_engine_optimizer]: 8.557e-05, [1]
    [Cycle 1]: 7.671e-05, [6]
        [build]: 5.67999e-06
        [elim_shapecalc]: 8.26002e-06
        [elim_not_effective]: 1.48e-05
        [opt_reshape]: 6.60002e-06
        [fold_const_symbol]: 9.04e-06
        [renormalize]: 1.11002e-06
[validate]: 4.505e-05
Sums
    bootstrap                            :     0.000290s :  2.99%
    type_inference                       :     0.004550s : 46.97%
    event_method                         :     0.000019s :  0.19%
    auto_monad                           :     0.000067s :  0.69%
    graph_reusing                        :     0.000005s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000039s :  0.40%
    rewriter_before_opt_a                :     0.000074s :  0.76%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000070s :  0.72%
    jit_opt_a.loop_unroll                :     0.000035s :  0.37%
    jit_opt_a.a_1                        :     0.000742s :  7.66%
    jit_opt_a.with_stream_mark           :     0.000043s :  0.44%
    jit_opt_a.recompute_prepare          :     0.000016s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000009s :  0.09%
    jit_opt_a.updatestate_assign_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000014s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.06%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000037s :  0.38%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000022s :  0.22%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.001077s : 11.12%
    jit_opt_a.add_forward_monad_depend   :     0.000015s :  0.16%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000030s :  0.31%
    jit_opt_a.cse                        :     0.000049s :  0.51%
    jit_opt_a.replace_applicator         :     0.000033s :  0.34%
    py_interpret_to_execute_after_opt_a  :     0.000019s :  0.19%
    rewriter_after_opt_a                 :     0.000045s :  0.46%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000799s :  8.25%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.18%
    cconv                                :     0.000034s :  0.35%
    loop_unroll                          :     0.000497s :  5.13%
    jit_opt_after_cconv.c_1              :     0.000024s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.10%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000029s :  0.30%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.26%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000016s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000058s :  0.59%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000011s :  0.11%
    opt_after_jit_grad                   :     0.000569s :  5.88%
    symbol_engine_optimizer.build        :     0.000006s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000045s :  0.47%


Time group info:
------[substitution.]   0.000245    23
 0.73% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000002s :      2: substitution.fold_const_symbol
 2.81% :     0.000007s :      3: substitution.graph_param_transform
83.14% :     0.000204s :      4: substitution.inline
 1.77% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.38% :     0.000006s :      4: substitution.remove_not_recompute_node
 3.01% :     0.000007s :      2: substitution.replace_old_param
 5.52% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004479     2
78.66% :     0.003523s :      1: type_inference.infer
21.34% :     0.000956s :      1: type_inference.specialize
------[replace.]   0.000080     6
66.77% :     0.000054s :      4: replace.inline
33.23% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000214     6
94.18% :     0.000201s :      4: match.inline
 5.82% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000173   691
 0.97% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.01% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.71% :     0.000001s :     10: predicate.addn_check_dump
 0.85% :     0.000001s :     10: predicate.addn_zero_filter
 1.59% :     0.000003s :     10: predicate.arithmetic_simplify
 0.77% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.70% :     0.000001s :     10: predicate.compare_switch_simplify
 1.24% :     0.000002s :     10: predicate.depend_value_elim
 0.99% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.13% :     0.000002s :     10: predicate.dict_get_item_eliminator
 0.82% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.63% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000001s :      3: predicate.elim_not_effective
 0.45% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.88% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.91% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.76% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.80% :     0.000001s :     10: predicate.environ_get_eliminate
 0.88% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.15% :     0.000000s :      3: predicate.fold_const_symbol
 0.73% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000001s :      3: predicate.graph_param_transform
 4.08% :     0.000007s :     22: predicate.inline
 0.94% :     0.000002s :      6: predicate.inline_without_move
 0.24% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000002s :      6: predicate.less_batch_normalization
 0.99% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.19% :     0.000002s :     15: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 2.84% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.49% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.73% :     0.000001s :     10: predicate.merge_addn
 0.73% :     0.000001s :     10: predicate.minmaximum_grad
 1.87% :     0.000003s :      3: predicate.mutable_eliminate
 0.33% :     0.000001s :      3: predicate.opt_reshape
 1.68% :     0.000003s :     15: predicate.partial_eliminate
 0.97% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.35% :     0.000002s :     10: predicate.reduce_eliminate
 1.14% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.56% :     0.000003s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.18% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.08% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000002s :      3: predicate.row_tensor_eliminate
 0.80% :     0.000001s :     10: predicate.same_eliminate
 0.39% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.84% :     0.000001s :      6: predicate.special_op_eliminate
 0.80% :     0.000001s :      6: predicate.specialize_transform
 0.98% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 0.79% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 1.96% :     0.000003s :     16: predicate.switch_defer_inline
 1.66% :     0.000003s :     16: predicate.switch_layer_defer_inline
 5.52% :     0.000010s :     53: predicate.switch_simplify
 0.89% :     0.000002s :     10: predicate.tile_eliminate
 0.86% :     0.000002s :     10: predicate.transpose_eliminate
25.05% :     0.000043s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.04% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.00% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.10% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 1.98% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.34% :     0.000002s :     10: predicate.value_based_eliminate
 0.25% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.55% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000685    12
42.92% :     0.000294s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.92% :     0.000041s :      1: func_graph_cloner_run.FuncGraphClonerNode
51.17% :     0.000351s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016518    72
 0.37% :     0.000062s :      1: add_recomputation
 0.43% :     0.000071s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 1.84% :     0.000305s :      1: bootstrap
 0.22% :     0.000036s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.17% :     0.000028s :      1: cse_after_recomputation
 0.06% :     0.000010s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
40.31% :     0.006659s :      1: jit_opt_a
 1.10% :     0.000182s :      1: jit_opt_after_cconv
 0.37% :     0.000062s :      1: jit_opt_b
 3.07% :     0.000507s :      1: loop_unroll
 4.90% :     0.000810s :      1: mutable_eliminate
 6.02% :     0.000994s :     26: opt.transform.jit_opt_a
 0.35% :     0.000058s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000029s :      4: opt.transform.jit_opt_b
 0.09% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000020s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.52% :     0.000582s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000041s :      1: py_interpret_to_execute
 0.13% :     0.000022s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000019s :      1: remove_dup_value
 3.32% :     0.000548s :      1: renormalize.infer
 3.14% :     0.000518s :      1: renormalize.specialize
 0.08% :     0.000014s :      1: rewriter_after_jit_bprop_graph
 0.30% :     0.000050s :      1: rewriter_after_opt_a
 0.47% :     0.000077s :      1: rewriter_before_opt_a
 0.53% :     0.000088s :      1: symbol_engine_optimizer
27.65% :     0.004568s :      1: type_inference


TotalTime = 0.0149872, [30]
[bootstrap]: 0.00028668
[type_inference]: 0.00544858
[event_method]: 2.087e-05
[auto_monad]: 6.929e-05
[graph_reusing]: 6.12001e-06
[pre_auto_parallel]: 3.14999e-06
[py_interpret_to_execute]: 4.673e-05
[rewriter_before_opt_a]: 8.156e-05
[expand_dump_flag]: 3.19001e-06
[jit_opt_a]: 0.00647104, [2]
    [Cycle 1]: 0.00197237, [27]
        [switch_simplify]: 6.204e-05
        [loop_unroll]: 2.872e-05
        [a_1]: 0.0005716
        [with_stream_mark]: 2.023e-05
        [recompute_prepare]: 8.41002e-06
        [updatestate_depend_eliminate]: 3.93999e-06
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 2.89999e-06
        [parameter_eliminate]: 1.94e-06
        [specialize_transform]: 6.04999e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 6.12001e-06
        [meta_shard_fg_expand]: 2.16998e-06
        [get_grad_eliminate_]: 5.77999e-06
        [merge_forward]: 3.55003e-06
        [cell_reuse_recompute_pass]: 1.28002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 5.149e-05
        [j_node_and_user_rematch]: 1.224e-05
        [meta_fg_expand]: 2.84001e-06
        [replace_old_param]: 9.60001e-06
        [inline_without_move]: 5.89e-06
        [renormalize]: 0.00088639
        [add_forward_monad_depend]: 7.41001e-06
        [auto_monad_grad]: 2.29001e-06
        [auto_monad_eliminator]: 1.813e-05
        [cse]: 2.987e-05
        [replace_applicator]: 1.709e-05
    [Cycle 2]: 0.00035997, [27]
        [switch_simplify]: 6.32001e-06
        [loop_unroll]: 5.77001e-06
        [a_1]: 9.839e-05
        [with_stream_mark]: 1.258e-05
        [recompute_prepare]: 5.20999e-06
        [updatestate_depend_eliminate]: 3.79002e-06
        [updatestate_assign_eliminate]: 2.20002e-06
        [updatestate_loads_eliminate]: 1.87001e-06
        [parameter_eliminate]: 1.17999e-06
        [specialize_transform]: 5.46002e-06
        [updatestate_useless_node_eliminater]: 5.52001e-06
        [accelerated_algorithm]: 6.38e-06
        [meta_shard_fg_expand]: 1.59e-06
        [get_grad_eliminate_]: 5.27001e-06
        [merge_forward]: 3.36001e-06
        [cell_reuse_recompute_pass]: 2.11003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.572e-05
        [j_node_and_user_rematch]: 8.90001e-06
        [meta_fg_expand]: 1.85001e-06
        [replace_old_param]: 9.05999e-06
        [inline_without_move]: 5.24998e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 2.63e-06
        [auto_monad_grad]: 9.89996e-07
        [auto_monad_eliminator]: 5.42999e-06
        [cse]: 1.293e-05
        [replace_applicator]: 5.29e-06
[py_interpret_to_execute_after_opt_a]: 1.503e-05
[rewriter_after_opt_a]: 3.905e-05
[convert_after_rewriter]: 7.4e-06
[order_py_execute_after_rewriter]: 5.66e-06
[mutable_eliminate]: 0.00072977
[jit_opt_b]: 6.366e-05, [1]
    [Cycle 1]: 5.499e-05, [2]
        [frontend_op_eliminate]: 2.478e-05
        [inline_after_opt_a]: 1.69e-05
[cconv]: 3.048e-05
[loop_unroll]: 0.00051755
[jit_opt_after_cconv]: 0.00015857, [1]
    [Cycle 1]: 0.00015094, [11]
        [c_1]: 2.216e-05
        [parameter_eliminate]: 3.45e-06
        [updatestate_depend_eliminate]: 6.64001e-06
        [updatestate_assign_eliminate]: 2.53e-06
        [updatestate_loads_eliminate]: 2.30002e-06
        [cse]: 2.607e-05
        [call_graph_tuple_transform]: 2.125e-05
        [tuple_list_get_item_eliminator]: 5.54e-06
        [none_parameter_eliminate]: 1.49e-06
        [renormalize]: 5.09986e-07
        [switch_simplify]: 5.74e-06
[remove_dup_value]: 1.696e-05
[partial_unused_args_eliminate]: 2.71999e-06
[environ_conv]: 6.26998e-06
[add_recomputation]: 5.41e-05
[cse_after_recomputation]: 2.312e-05, [1]
    [Cycle 1]: 1.679e-05, [1]
        [cse]: 9.49e-06
[auto_monad_reorder]: 1.571e-05
[get_jit_bprop_graph]: 2.44999e-06
[rewriter_after_jit_bprop_graph]: 7.38999e-06
[opt_after_jit_grad]: 0.00051583
[symbol_engine_optimizer]: 8.003e-05, [1]
    [Cycle 1]: 7.226e-05, [6]
        [build]: 5.00001e-06
        [elim_shapecalc]: 7.85e-06
        [elim_not_effective]: 1.437e-05
        [opt_reshape]: 6.63e-06
        [fold_const_symbol]: 9.34998e-06
        [renormalize]: 6.19999e-07
[validate]: 4.312e-05
Sums
    bootstrap                            :     0.000287s :  2.82%
    type_inference                       :     0.005449s : 53.53%
    event_method                         :     0.000021s :  0.21%
    auto_monad                           :     0.000069s :  0.68%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000047s :  0.46%
    rewriter_before_opt_a                :     0.000082s :  0.80%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000068s :  0.67%
    jit_opt_a.loop_unroll                :     0.000034s :  0.34%
    jit_opt_a.a_1                        :     0.000670s :  6.58%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000067s :  0.66%
    jit_opt_a.j_node_and_user_rematch    :     0.000021s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000019s :  0.18%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000886s :  8.71%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.23%
    jit_opt_a.cse                        :     0.000043s :  0.42%
    jit_opt_a.replace_applicator         :     0.000022s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.15%
    rewriter_after_opt_a                 :     0.000039s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000730s :  7.17%
    jit_opt_b.frontend_op_eliminate      :     0.000025s :  0.24%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.17%
    cconv                                :     0.000030s :  0.30%
    loop_unroll                          :     0.000518s :  5.08%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000026s :  0.26%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000017s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000054s :  0.53%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000516s :  5.07%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000043s :  0.42%


Time group info:
------[substitution.]   0.000223    23
 0.94% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.67% :     0.000006s :      3: substitution.graph_param_transform
83.14% :     0.000186s :      4: substitution.inline
 2.11% :     0.000005s :      4: substitution.j_node_and_user_rematch
 2.46% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.54% :     0.000006s :      2: substitution.replace_old_param
 5.55% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005368     2
78.84% :     0.004232s :      1: type_inference.infer
21.16% :     0.001136s :      1: type_inference.specialize
------[replace.]   0.000073     6
66.02% :     0.000048s :      4: replace.inline
33.98% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000195     6
94.15% :     0.000183s :      4: match.inline
 5.85% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.27% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.15% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.99% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 2.11% :     0.000003s :     10: predicate.arithmetic_simplify
 1.20% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.09% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.29% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.37% :     0.000002s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.33% :     0.000002s :      6: predicate.get_grad_eliminate
 0.40% :     0.000001s :      3: predicate.graph_param_transform
 5.22% :     0.000007s :     22: predicate.inline
 1.00% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.21% :     0.000002s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.85% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.87% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.94% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.97% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 2.19% :     0.000003s :      3: predicate.mutable_eliminate
 0.58% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000003s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.80% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.51% :     0.000002s :     18: predicate.replace_applicator
 0.66% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.53% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.17% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.10% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.52% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.85% :     0.000010s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.35% :     0.000002s :     10: predicate.transpose_eliminate
 1.25% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.21% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.57% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.90% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000746    12
48.72% :     0.000364s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.34% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.95% :     0.000350s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016836    72
 0.34% :     0.000058s :      1: add_recomputation
 0.44% :     0.000074s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 1.80% :     0.000304s :      1: bootstrap
 0.20% :     0.000033s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000026s :      1: cse_after_recomputation
 0.06% :     0.000010s :      1: environ_conv
 0.15% :     0.000026s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.46% :     0.006475s :      1: jit_opt_a
 0.96% :     0.000162s :      1: jit_opt_after_cconv
 0.40% :     0.000067s :      1: jit_opt_b
 3.12% :     0.000526s :      1: loop_unroll
 4.40% :     0.000741s :      1: mutable_eliminate
 5.58% :     0.000940s :     26: opt.transform.jit_opt_a
 0.30% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000034s :      4: opt.transform.jit_opt_b
 0.09% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.12% :     0.000526s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000050s :      1: py_interpret_to_execute
 0.11% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000019s :      1: remove_dup_value
 2.53% :     0.000426s :      1: renormalize.infer
 2.53% :     0.000427s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000042s :      1: rewriter_after_opt_a
 0.51% :     0.000085s :      1: rewriter_before_opt_a
 0.49% :     0.000083s :      1: symbol_engine_optimizer
32.48% :     0.005468s :      1: type_inference


TotalTime = 0.0134761, [30]
[bootstrap]: 0.00031125
[type_inference]: 0.00449058
[event_method]: 1.824e-05
[auto_monad]: 6.609e-05
[graph_reusing]: 5.89999e-06
[pre_auto_parallel]: 2.33998e-06
[py_interpret_to_execute]: 3.358e-05
[rewriter_before_opt_a]: 7.143e-05
[expand_dump_flag]: 3.07002e-06
[jit_opt_a]: 0.00594576, [2]
    [Cycle 1]: 0.00174587, [27]
        [switch_simplify]: 5.795e-05
        [loop_unroll]: 2.887e-05
        [a_1]: 0.00054957
        [with_stream_mark]: 1.8e-05
        [recompute_prepare]: 8.86002e-06
        [updatestate_depend_eliminate]: 3.43e-06
        [updatestate_assign_eliminate]: 3.36999e-06
        [updatestate_loads_eliminate]: 2.69001e-06
        [parameter_eliminate]: 1.74e-06
        [specialize_transform]: 5.77001e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 6.34001e-06
        [meta_shard_fg_expand]: 2.12001e-06
        [get_grad_eliminate_]: 6.25002e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 1.17e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.559e-05
        [j_node_and_user_rematch]: 9.85002e-06
        [meta_fg_expand]: 2.28998e-06
        [replace_old_param]: 9.64e-06
        [inline_without_move]: 5.66e-06
        [renormalize]: 0.00074426
        [add_forward_monad_depend]: 7.13e-06
        [auto_monad_grad]: 2.69999e-06
        [auto_monad_eliminator]: 1.669e-05
        [cse]: 2.688e-05
        [replace_applicator]: 1.434e-05
    [Cycle 2]: 0.00035652, [27]
        [switch_simplify]: 5.30999e-06
        [loop_unroll]: 5.00001e-06
        [a_1]: 9.338e-05
        [with_stream_mark]: 1.226e-05
        [recompute_prepare]: 5.38002e-06
        [updatestate_depend_eliminate]: 2.59999e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.11e-06
        [parameter_eliminate]: 1.45001e-06
        [specialize_transform]: 5.37001e-06
        [updatestate_useless_node_eliminater]: 5.15001e-06
        [accelerated_algorithm]: 5.44e-06
        [meta_shard_fg_expand]: 1.43002e-06
        [get_grad_eliminate_]: 4.90001e-06
        [merge_forward]: 2.71e-06
        [cell_reuse_recompute_pass]: 1.88002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.289e-05
        [j_node_and_user_rematch]: 8.3e-06
        [meta_fg_expand]: 1.64998e-06
        [replace_old_param]: 7.92e-06
        [inline_without_move]: 5.19998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 1.27999e-06
        [auto_monad_eliminator]: 6.93e-06
        [cse]: 1.494e-05
        [replace_applicator]: 6.38e-06
[py_interpret_to_execute_after_opt_a]: 1.169e-05
[rewriter_after_opt_a]: 3.798e-05
[convert_after_rewriter]: 6.93e-06
[order_py_execute_after_rewriter]: 5.58002e-06
[mutable_eliminate]: 0.0006454
[jit_opt_b]: 9.514e-05, [1]
    [Cycle 1]: 8.649e-05, [2]
        [frontend_op_eliminate]: 5.555e-05
        [inline_after_opt_a]: 1.676e-05
[cconv]: 3.14e-05
[loop_unroll]: 0.00051894
[jit_opt_after_cconv]: 0.00016794, [1]
    [Cycle 1]: 0.00016052, [11]
        [c_1]: 2.188e-05
        [parameter_eliminate]: 3.24001e-06
        [updatestate_depend_eliminate]: 7.36001e-06
        [updatestate_assign_eliminate]: 2.71999e-06
        [updatestate_loads_eliminate]: 2.58e-06
        [cse]: 2.744e-05
        [call_graph_tuple_transform]: 2.492e-05
        [tuple_list_get_item_eliminator]: 6.26998e-06
        [none_parameter_eliminate]: 1.69e-06
        [renormalize]: 1.12e-06
        [switch_simplify]: 5.30999e-06
[remove_dup_value]: 1.483e-05
[partial_unused_args_eliminate]: 2.47001e-06
[environ_conv]: 5.54998e-06
[add_recomputation]: 4.884e-05
[cse_after_recomputation]: 2.258e-05, [1]
    [Cycle 1]: 1.639e-05, [1]
        [cse]: 9.71e-06
[auto_monad_reorder]: 1.497e-05
[get_jit_bprop_graph]: 1.59998e-06
[rewriter_after_jit_bprop_graph]: 8.97e-06
[opt_after_jit_grad]: 0.0005721
[symbol_engine_optimizer]: 7.96e-05, [1]
    [Cycle 1]: 7.169e-05, [6]
        [build]: 4.97e-06
        [elim_shapecalc]: 8.35999e-06
        [elim_not_effective]: 1.488e-05
        [opt_reshape]: 5.47001e-06
        [fold_const_symbol]: 8.55999e-06
        [renormalize]: 1.14e-06
[validate]: 4.358e-05
Sums
    bootstrap                            :     0.000311s :  3.46%
    type_inference                       :     0.004491s : 49.97%
    event_method                         :     0.000018s :  0.20%
    auto_monad                           :     0.000066s :  0.74%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.37%
    rewriter_before_opt_a                :     0.000071s :  0.79%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000063s :  0.70%
    jit_opt_a.loop_unroll                :     0.000034s :  0.38%
    jit_opt_a.a_1                        :     0.000643s :  7.15%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.34%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.20%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000744s :  8.28%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.26%
    jit_opt_a.cse                        :     0.000042s :  0.47%
    jit_opt_a.replace_applicator         :     0.000021s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.13%
    rewriter_after_opt_a                 :     0.000038s :  0.42%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000645s :  7.18%
    jit_opt_b.frontend_op_eliminate      :     0.000056s :  0.62%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.19%
    cconv                                :     0.000031s :  0.35%
    loop_unroll                          :     0.000519s :  5.77%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000027s :  0.31%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000025s :  0.28%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000049s :  0.54%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.10%
    opt_after_jit_grad                   :     0.000572s :  6.37%
    symbol_engine_optimizer.build        :     0.000005s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.48%


Time group info:
------[substitution.]   0.000208    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.63% :     0.000001s :      2: substitution.fold_const_symbol
 2.95% :     0.000006s :      3: substitution.graph_param_transform
82.95% :     0.000172s :      4: substitution.inline
 1.93% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.38% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.37% :     0.000005s :      2: substitution.replace_old_param
 5.82% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004426     2
79.99% :     0.003541s :      1: type_inference.infer
20.01% :     0.000885s :      1: type_inference.specialize
------[replace.]   0.000076     6
65.89% :     0.000050s :      4: replace.inline
34.11% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000180     6
94.04% :     0.000170s :      4: match.inline
 5.96% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.88% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.12% :     0.000001s :     10: predicate.addn_zero_filter
 1.99% :     0.000003s :     10: predicate.arithmetic_simplify
 1.01% :     0.000001s :     10: predicate.cast_eliminate
 0.79% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.97% :     0.000001s :     10: predicate.compare_switch_simplify
 1.52% :     0.000002s :     10: predicate.depend_value_elim
 1.01% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.55% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.45% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.88% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.53% :     0.000001s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.48% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.11% :     0.000001s :     10: predicate.environ_get_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.70% :     0.000002s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.37% :     0.000007s :     22: predicate.inline
 0.84% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.40% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.70% :     0.000002s :     15: predicate.load_eliminater
 2.33% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.82% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.33% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.01% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 2.16% :     0.000003s :      3: predicate.mutable_eliminate
 0.40% :     0.000000s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     15: predicate.partial_eliminate
 0.98% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.54% :     0.000002s :     10: predicate.reduce_eliminate
 1.33% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.84% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.86% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
 1.55% :     0.000002s :     10: predicate.reshape_eliminate
 1.07% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.95% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.52% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.27% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.56% :     0.000003s :     16: predicate.switch_defer_inline
 1.88% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.04% :     0.000009s :     53: predicate.switch_simplify
 1.44% :     0.000002s :     10: predicate.tile_eliminate
 1.23% :     0.000002s :     10: predicate.transpose_eliminate
 1.29% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.85% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.08% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.55% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.48% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.56% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.51% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000647    12
48.31% :     0.000313s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.25% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.44% :     0.000301s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015182    72
 0.34% :     0.000052s :      1: add_recomputation
 0.46% :     0.000069s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.15% :     0.000326s :      1: bootstrap
 0.22% :     0.000034s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000025s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.19% :     0.005950s :      1: jit_opt_a
 1.13% :     0.000171s :      1: jit_opt_after_cconv
 0.64% :     0.000098s :      1: jit_opt_b
 3.48% :     0.000529s :      1: loop_unroll
 4.33% :     0.000658s :      1: mutable_eliminate
 5.70% :     0.000865s :     26: opt.transform.jit_opt_a
 0.36% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.43% :     0.000065s :      4: opt.transform.jit_opt_b
 0.12% :     0.000018s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.83% :     0.000582s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000036s :      1: py_interpret_to_execute
 0.10% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 2.49% :     0.000378s :      1: renormalize.infer
 2.35% :     0.000357s :      1: renormalize.specialize
 0.07% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000041s :      1: rewriter_after_opt_a
 0.49% :     0.000074s :      1: rewriter_before_opt_a
 0.54% :     0.000082s :      1: symbol_engine_optimizer
29.67% :     0.004504s :      1: type_inference


TotalTime = 0.0153386, [30]
[bootstrap]: 0.00038002
[type_inference]: 0.00576191
[event_method]: 2.071e-05
[auto_monad]: 7.07e-05
[graph_reusing]: 6.98e-06
[pre_auto_parallel]: 2.67001e-06
[py_interpret_to_execute]: 5.045e-05
[rewriter_before_opt_a]: 8.421e-05
[expand_dump_flag]: 3.56001e-06
[jit_opt_a]: 0.00640488, [2]
    [Cycle 1]: 0.00198706, [27]
        [switch_simplify]: 6.026e-05
        [loop_unroll]: 2.906e-05
        [a_1]: 0.00060463
        [with_stream_mark]: 2.141e-05
        [recompute_prepare]: 9.54e-06
        [updatestate_depend_eliminate]: 3.73001e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.91e-06
        [parameter_eliminate]: 1.80001e-06
        [specialize_transform]: 6.14001e-06
        [updatestate_useless_node_eliminater]: 5.04998e-06
        [accelerated_algorithm]: 5.74e-06
        [meta_shard_fg_expand]: 1.94999e-06
        [get_grad_eliminate_]: 5.96998e-06
        [merge_forward]: 4.25e-06
        [cell_reuse_recompute_pass]: 1.73002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.554e-05
        [j_node_and_user_rematch]: 9.75002e-06
        [meta_fg_expand]: 2.39999e-06
        [replace_old_param]: 1.009e-05
        [inline_without_move]: 6.21998e-06
        [renormalize]: 0.00091852
        [add_forward_monad_depend]: 7.93001e-06
        [auto_monad_grad]: 2.46e-06
        [auto_monad_eliminator]: 1.628e-05
        [cse]: 2.778e-05
        [replace_applicator]: 1.555e-05
    [Cycle 2]: 0.00040755, [27]
        [switch_simplify]: 5.99999e-06
        [loop_unroll]: 5.06997e-06
        [a_1]: 9.683e-05
        [with_stream_mark]: 1.076e-05
        [recompute_prepare]: 5.74e-06
        [updatestate_depend_eliminate]: 2.63e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.07999e-06
        [parameter_eliminate]: 1.82999e-06
        [specialize_transform]: 6.36e-06
        [updatestate_useless_node_eliminater]: 5.51998e-06
        [accelerated_algorithm]: 5.79e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 4.67e-06
        [merge_forward]: 2.59999e-06
        [cell_reuse_recompute_pass]: 2.12999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.617e-05
        [j_node_and_user_rematch]: 9.02e-06
        [meta_fg_expand]: 1.84e-06
        [replace_old_param]: 8.65001e-06
        [inline_without_move]: 4.85001e-06
        [renormalize]: 1.00001e-07
        [add_forward_monad_depend]: 2.68e-06
        [auto_monad_grad]: 1.14e-06
        [auto_monad_eliminator]: 7.27002e-06
        [cse]: 1.573e-05
        [replace_applicator]: 6.98e-06
[py_interpret_to_execute_after_opt_a]: 1.353e-05
[rewriter_after_opt_a]: 3.913e-05
[convert_after_rewriter]: 8.33001e-06
[order_py_execute_after_rewriter]: 5.05001e-06
[mutable_eliminate]: 0.00072733
[jit_opt_b]: 5.563e-05, [1]
    [Cycle 1]: 4.741e-05, [2]
        [frontend_op_eliminate]: 1.901e-05
        [inline_after_opt_a]: 1.632e-05
[cconv]: 2.987e-05
[loop_unroll]: 0.00051133
[jit_opt_after_cconv]: 0.00015813, [1]
    [Cycle 1]: 0.00014984, [11]
        [c_1]: 2.208e-05
        [parameter_eliminate]: 3.66001e-06
        [updatestate_depend_eliminate]: 6.80998e-06
        [updatestate_assign_eliminate]: 2.44999e-06
        [updatestate_loads_eliminate]: 2.39001e-06
        [cse]: 2.309e-05
        [call_graph_tuple_transform]: 2.115e-05
        [tuple_list_get_item_eliminator]: 5.92999e-06
        [none_parameter_eliminate]: 1.57999e-06
        [renormalize]: 9.70002e-07
        [switch_simplify]: 5.50001e-06
[remove_dup_value]: 1.424e-05
[partial_unused_args_eliminate]: 2.74999e-06
[environ_conv]: 5.15999e-06
[add_recomputation]: 5.101e-05
[cse_after_recomputation]: 2.439e-05, [1]
    [Cycle 1]: 1.767e-05, [1]
        [cse]: 1.004e-05
[auto_monad_reorder]: 1.633e-05
[get_jit_bprop_graph]: 2.32999e-06
[rewriter_after_jit_bprop_graph]: 6.82002e-06
[opt_after_jit_grad]: 0.00054945
[symbol_engine_optimizer]: 7.901e-05, [1]
    [Cycle 1]: 7.087e-05, [6]
        [build]: 3.71001e-06
        [elim_shapecalc]: 7.71001e-06
        [elim_not_effective]: 1.463e-05
        [opt_reshape]: 6.54999e-06
        [fold_const_symbol]: 9.35001e-06
        [renormalize]: 9.20001e-07
[validate]: 3.885e-05
Sums
    bootstrap                            :     0.000380s :  3.58%
    type_inference                       :     0.005762s : 54.24%
    event_method                         :     0.000021s :  0.19%
    auto_monad                           :     0.000071s :  0.67%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000050s :  0.47%
    rewriter_before_opt_a                :     0.000084s :  0.79%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000066s :  0.62%
    jit_opt_a.loop_unroll                :     0.000034s :  0.32%
    jit_opt_a.a_1                        :     0.000701s :  6.60%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.03%
    jit_opt_a.specialize_transform       :     0.000013s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.18%
    jit_opt_a.inline_without_move        :     0.000011s :  0.10%
    jit_opt_a.renormalize                :     0.000919s :  8.65%
    jit_opt_a.add_forward_monad_depend   :     0.000011s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.22%
    jit_opt_a.cse                        :     0.000044s :  0.41%
    jit_opt_a.replace_applicator         :     0.000023s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.13%
    rewriter_after_opt_a                 :     0.000039s :  0.37%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000727s :  6.85%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.15%
    cconv                                :     0.000030s :  0.28%
    loop_unroll                          :     0.000511s :  4.81%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000014s :  0.13%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000051s :  0.48%
    cse_after_recomputation.cse          :     0.000010s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.06%
    opt_after_jit_grad                   :     0.000549s :  5.17%
    symbol_engine_optimizer.build        :     0.000004s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.37%


Time group info:
------[substitution.]   0.000249    23
 0.71% :     0.000002s :      2: substitution.elim_not_effective
 0.80% :     0.000002s :      2: substitution.fold_const_symbol
 2.09% :     0.000005s :      3: substitution.graph_param_transform
85.29% :     0.000213s :      4: substitution.inline
 1.61% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.91% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.20% :     0.000005s :      2: substitution.replace_old_param
 5.38% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005676     2
78.05% :     0.004430s :      1: type_inference.infer
21.95% :     0.001246s :      1: type_inference.specialize
------[replace.]   0.000080     6
64.69% :     0.000052s :      4: replace.inline
35.31% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000222     6
94.56% :     0.000210s :      4: match.inline
 5.44% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.22% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.97% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 2.42% :     0.000003s :     10: predicate.arithmetic_simplify
 1.19% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.46% :     0.000002s :     10: predicate.depend_value_elim
 1.60% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.12% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.78% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.46% :     0.000001s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.30% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.43% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.38% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.83% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.54% :     0.000007s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.18% :     0.000001s :      6: predicate.less_batch_normalization
 1.38% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.24% :     0.000003s :     15: predicate.load_eliminater
 1.80% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.76% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.86% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 1.74% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000003s :     15: predicate.partial_eliminate
 1.08% :     0.000001s :     10: predicate.print_const_string_wrapper
 2.19% :     0.000003s :     10: predicate.reduce_eliminate
 1.55% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.78% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.68% :     0.000002s :     18: predicate.replace_applicator
 0.70% :     0.000001s :      6: predicate.replace_old_param
 0.44% :     0.000001s :      3: predicate.reset_defer_inline
 1.92% :     0.000002s :     10: predicate.reshape_eliminate
 1.09% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.98% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     10: predicate.same_eliminate
 0.68% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 1.20% :     0.000002s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.10% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.79% :     0.000003s :     16: predicate.switch_defer_inline
 1.86% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.28% :     0.000009s :     53: predicate.switch_simplify
 0.96% :     0.000001s :     10: predicate.tile_eliminate
 1.03% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.48% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.37% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.33% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.72% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.85% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.81% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000808    12
47.58% :     0.000384s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.10% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.31% :     0.000374s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017249    72
 0.32% :     0.000055s :      1: add_recomputation
 0.43% :     0.000074s :      1: auto_monad
 0.11% :     0.000019s :      1: auto_monad_reorder
 2.31% :     0.000399s :      1: bootstrap
 0.19% :     0.000033s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.15% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
37.16% :     0.006409s :      1: jit_opt_a
 0.94% :     0.000161s :      1: jit_opt_after_cconv
 0.34% :     0.000058s :      1: jit_opt_b
 3.03% :     0.000523s :      1: loop_unroll
 4.29% :     0.000740s :      1: mutable_eliminate
 5.41% :     0.000933s :     26: opt.transform.jit_opt_a
 0.30% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000028s :      4: opt.transform.jit_opt_b
 0.09% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.25% :     0.000560s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.31% :     0.000054s :      1: py_interpret_to_execute
 0.10% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.90% :     0.000501s :      1: renormalize.infer
 2.37% :     0.000409s :      1: renormalize.specialize
 0.05% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000043s :      1: rewriter_after_opt_a
 0.52% :     0.000090s :      1: rewriter_before_opt_a
 0.47% :     0.000081s :      1: symbol_engine_optimizer
33.51% :     0.005780s :      1: type_inference


TotalTime = 0.0139753, [30]
[bootstrap]: 0.00032496
[type_inference]: 0.00495125
[event_method]: 1.907e-05
[auto_monad]: 6.046e-05
[graph_reusing]: 5.40999e-06
[pre_auto_parallel]: 2.04e-06
[py_interpret_to_execute]: 4.129e-05
[rewriter_before_opt_a]: 7.432e-05
[expand_dump_flag]: 2.92002e-06
[jit_opt_a]: 0.00620771, [2]
    [Cycle 1]: 0.00205486, [27]
        [switch_simplify]: 6.452e-05
        [loop_unroll]: 2.91e-05
        [a_1]: 0.00085303
        [with_stream_mark]: 2.247e-05
        [recompute_prepare]: 7.51001e-06
        [updatestate_depend_eliminate]: 4.06001e-06
        [updatestate_assign_eliminate]: 2.42001e-06
        [updatestate_loads_eliminate]: 2.31e-06
        [parameter_eliminate]: 1.75001e-06
        [specialize_transform]: 5.89999e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 5.34998e-06
        [meta_shard_fg_expand]: 1.99e-06
        [get_grad_eliminate_]: 5.29998e-06
        [merge_forward]: 3.43e-06
        [cell_reuse_recompute_pass]: 1.48002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.53e-05
        [j_node_and_user_rematch]: 9.02e-06
        [meta_fg_expand]: 2.36998e-06
        [replace_old_param]: 9.57999e-06
        [inline_without_move]: 5.14e-06
        [renormalize]: 0.00076772
        [add_forward_monad_depend]: 4.90001e-06
        [auto_monad_grad]: 1.42e-06
        [auto_monad_eliminator]: 1.177e-05
        [cse]: 2.057e-05
        [replace_applicator]: 1.19e-05
    [Cycle 2]: 0.00031912, [27]
        [switch_simplify]: 5.97001e-06
        [loop_unroll]: 4.72e-06
        [a_1]: 9.124e-05
        [with_stream_mark]: 8.59e-06
        [recompute_prepare]: 5.12999e-06
        [updatestate_depend_eliminate]: 2.44999e-06
        [updatestate_assign_eliminate]: 2.09e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [parameter_eliminate]: 9.60019e-07
        [specialize_transform]: 4.95999e-06
        [updatestate_useless_node_eliminater]: 4.57e-06
        [accelerated_algorithm]: 5.19998e-06
        [meta_shard_fg_expand]: 1.07e-06
        [get_grad_eliminate_]: 4.49002e-06
        [merge_forward]: 2.66e-06
        [cell_reuse_recompute_pass]: 1.76e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.189e-05
        [j_node_and_user_rematch]: 7.78999e-06
        [meta_fg_expand]: 1.44e-06
        [replace_old_param]: 1.072e-05
        [inline_without_move]: 4.57e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.11002e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 4.92e-06
        [cse]: 1.044e-05
        [replace_applicator]: 5.03002e-06
[py_interpret_to_execute_after_opt_a]: 1.028e-05
[rewriter_after_opt_a]: 3.5e-05
[convert_after_rewriter]: 6.64001e-06
[order_py_execute_after_rewriter]: 4.50001e-06
[mutable_eliminate]: 0.00062555
[jit_opt_b]: 4.879e-05, [1]
    [Cycle 1]: 4.233e-05, [2]
        [frontend_op_eliminate]: 1.66e-05
        [inline_after_opt_a]: 1.454e-05
[cconv]: 1.711e-05
[loop_unroll]: 0.00043699
[jit_opt_after_cconv]: 0.00014332, [1]
    [Cycle 1]: 0.00013625, [11]
        [c_1]: 2.235e-05
        [parameter_eliminate]: 2.39999e-06
        [updatestate_depend_eliminate]: 5.52999e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 2.23002e-06
        [cse]: 2.02e-05
        [call_graph_tuple_transform]: 1.799e-05
        [tuple_list_get_item_eliminator]: 5.67999e-06
        [none_parameter_eliminate]: 1.14e-06
        [renormalize]: 4.59986e-07
        [switch_simplify]: 5.35001e-06
[remove_dup_value]: 9.60001e-06
[partial_unused_args_eliminate]: 1.69998e-06
[environ_conv]: 4.37e-06
[add_recomputation]: 4.773e-05
[cse_after_recomputation]: 2.389e-05, [1]
    [Cycle 1]: 1.756e-05, [1]
        [cse]: 1.07e-05
[auto_monad_reorder]: 1.167e-05
[get_jit_bprop_graph]: 1.65001e-06
[rewriter_after_jit_bprop_graph]: 7.15e-06
[opt_after_jit_grad]: 0.00055182
[symbol_engine_optimizer]: 7.831e-05, [1]
    [Cycle 1]: 7.042e-05, [6]
        [build]: 4.72998e-06
        [elim_shapecalc]: 7.95998e-06
        [elim_not_effective]: 1.358e-05
        [opt_reshape]: 6.44999e-06
        [fold_const_symbol]: 9.20999e-06
        [renormalize]: 6.60017e-07
[validate]: 3.385e-05
Sums
    bootstrap                            :     0.000325s :  3.41%
    type_inference                       :     0.004951s : 51.90%
    event_method                         :     0.000019s :  0.20%
    auto_monad                           :     0.000060s :  0.63%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000041s :  0.43%
    rewriter_before_opt_a                :     0.000074s :  0.78%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000070s :  0.74%
    jit_opt_a.loop_unroll                :     0.000034s :  0.35%
    jit_opt_a.a_1                        :     0.000944s :  9.90%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000009s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000020s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.10%
    jit_opt_a.renormalize                :     0.000768s :  8.05%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.02%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.17%
    jit_opt_a.cse                        :     0.000031s :  0.33%
    jit_opt_a.replace_applicator         :     0.000017s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.11%
    rewriter_after_opt_a                 :     0.000035s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000626s :  6.56%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.17%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.15%
    cconv                                :     0.000017s :  0.18%
    loop_unroll                          :     0.000437s :  4.58%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000010s :  0.10%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000048s :  0.50%
    cse_after_recomputation.cse          :     0.000011s :  0.11%
    auto_monad_reorder                   :     0.000012s :  0.12%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000552s :  5.78%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000034s :  0.35%


Time group info:
------[substitution.]   0.000290    23
 0.51% :     0.000001s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 1.54% :     0.000004s :      3: substitution.graph_param_transform
62.52% :     0.000181s :      4: substitution.inline
 0.93% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.74% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.76% :     0.000008s :      2: substitution.replace_old_param
29.53% :     0.000086s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004874     2
80.50% :     0.003923s :      1: type_inference.infer
19.50% :     0.000951s :      1: type_inference.specialize
------[replace.]   0.000081     6
63.52% :     0.000051s :      4: replace.inline
36.48% :     0.000030s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000263     6
67.94% :     0.000179s :      4: match.inline
32.06% :     0.000084s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000132   691
 1.20% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.20% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.91% :     0.000001s :     10: predicate.addn_check_dump
 1.19% :     0.000002s :     10: predicate.addn_zero_filter
 3.16% :     0.000004s :     10: predicate.arithmetic_simplify
 1.46% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.92% :     0.000001s :     10: predicate.compare_switch_simplify
 1.67% :     0.000002s :     10: predicate.depend_value_elim
 0.97% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.08% :     0.000001s :     10: predicate.dict_get_item_eliminator
10.88% :     0.000014s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.01% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.96% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.08% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.87% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 4.74% :     0.000006s :     22: predicate.inline
 0.81% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.03% :     0.000001s :      6: predicate.less_batch_normalization
 1.36% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.78% :     0.000002s :     15: predicate.load_eliminater
 1.20% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.39% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.20% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.93% :     0.000001s :     10: predicate.merge_addn
 0.95% :     0.000001s :     10: predicate.minmaximum_grad
 1.54% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 1.86% :     0.000002s :     15: predicate.partial_eliminate
 1.05% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.89% :     0.000003s :     10: predicate.reduce_eliminate
 1.25% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.48% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.13% :     0.000001s :     10: predicate.reshape_eliminate
 1.20% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.83% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.08% :     0.000001s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.20% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.01% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 1.86% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.98% :     0.000009s :     53: predicate.switch_simplify
 1.22% :     0.000002s :     10: predicate.tile_eliminate
 1.03% :     0.000001s :     10: predicate.transpose_eliminate
 1.19% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.29% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.42% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.36% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.45% :     0.000002s :     10: predicate.value_based_eliminate
 0.30% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.78% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000646    12
48.37% :     0.000313s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.00% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.63% :     0.000301s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015963    72
 0.32% :     0.000051s :      1: add_recomputation
 0.40% :     0.000064s :      1: auto_monad
 0.09% :     0.000014s :      1: auto_monad_reorder
 2.15% :     0.000342s :      1: bootstrap
 0.12% :     0.000020s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000027s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.91% :     0.006212s :      1: jit_opt_a
 0.92% :     0.000146s :      1: jit_opt_after_cconv
 0.32% :     0.000051s :      1: jit_opt_b
 2.79% :     0.000446s :      1: loop_unroll
 3.97% :     0.000634s :      1: mutable_eliminate
 7.29% :     0.001164s :     26: opt.transform.jit_opt_a
 0.30% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.52% :     0.000561s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.02% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000044s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.08% :     0.000012s :      1: remove_dup_value
 2.54% :     0.000405s :      1: renormalize.infer
 2.23% :     0.000355s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000038s :      1: rewriter_after_opt_a
 0.49% :     0.000078s :      1: rewriter_before_opt_a
 0.51% :     0.000081s :      1: symbol_engine_optimizer
31.12% :     0.004968s :      1: type_inference


TotalTime = 0.0141503, [30]
[bootstrap]: 0.00031389
[type_inference]: 0.00464328
[event_method]: 1.771e-05
[auto_monad]: 6.68e-05
[graph_reusing]: 5.56e-06
[pre_auto_parallel]: 2.66e-06
[py_interpret_to_execute]: 3.462e-05
[rewriter_before_opt_a]: 7.312e-05
[expand_dump_flag]: 3.09001e-06
[jit_opt_a]: 0.00669375, [2]
    [Cycle 1]: 0.00224183, [27]
        [switch_simplify]: 6.28e-05
        [loop_unroll]: 3.065e-05
        [a_1]: 0.00060771
        [with_stream_mark]: 2.248e-05
        [recompute_prepare]: 7.77e-06
        [updatestate_depend_eliminate]: 4.08999e-06
        [updatestate_assign_eliminate]: 2.93e-06
        [updatestate_loads_eliminate]: 2.54001e-06
        [parameter_eliminate]: 2.09e-06
        [specialize_transform]: 6.43e-06
        [updatestate_useless_node_eliminater]: 5.12e-06
        [accelerated_algorithm]: 5.87999e-06
        [meta_shard_fg_expand]: 2.12001e-06
        [get_grad_eliminate_]: 6.07999e-06
        [merge_forward]: 3.48e-06
        [cell_reuse_recompute_pass]: 1.26002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.389e-05
        [j_node_and_user_rematch]: 9.34998e-06
        [meta_fg_expand]: 2.04e-06
        [replace_old_param]: 1.007e-05
        [inline_without_move]: 5.51e-06
        [renormalize]: 0.00115995
        [add_forward_monad_depend]: 6.74001e-06
        [auto_monad_grad]: 2.66999e-06
        [auto_monad_eliminator]: 1.488e-05
        [cse]: 3.336e-05
        [replace_applicator]: 1.419e-05
    [Cycle 2]: 0.00032604, [27]
        [switch_simplify]: 5.66998e-06
        [loop_unroll]: 5.34e-06
        [a_1]: 8.907e-05
        [with_stream_mark]: 1.096e-05
        [recompute_prepare]: 5.17e-06
        [updatestate_depend_eliminate]: 2.59999e-06
        [updatestate_assign_eliminate]: 2.19999e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 8.89995e-07
        [specialize_transform]: 5.06002e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 5.32999e-06
        [meta_shard_fg_expand]: 1.54e-06
        [get_grad_eliminate_]: 4.95999e-06
        [merge_forward]: 3.01999e-06
        [cell_reuse_recompute_pass]: 1.27999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.292e-05
        [j_node_and_user_rematch]: 7.75998e-06
        [meta_fg_expand]: 1.59e-06
        [replace_old_param]: 7.46999e-06
        [inline_without_move]: 4.75001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.23002e-06
        [auto_monad_grad]: 7.09988e-07
        [auto_monad_eliminator]: 4.85001e-06
        [cse]: 1.122e-05
        [replace_applicator]: 4.96002e-06
[py_interpret_to_execute_after_opt_a]: 1.235e-05
[rewriter_after_opt_a]: 3.794e-05
[convert_after_rewriter]: 7.41999e-06
[order_py_execute_after_rewriter]: 5.15999e-06
[mutable_eliminate]: 0.00069636
[jit_opt_b]: 5e-05, [1]
    [Cycle 1]: 4.275e-05, [2]
        [frontend_op_eliminate]: 1.691e-05
        [inline_after_opt_a]: 1.501e-05
[cconv]: 2.709e-05
[loop_unroll]: 0.0004344
[jit_opt_after_cconv]: 0.00014096, [1]
    [Cycle 1]: 0.00013446, [11]
        [c_1]: 1.972e-05
        [parameter_eliminate]: 2.71999e-06
        [updatestate_depend_eliminate]: 6.33e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 2.07001e-06
        [cse]: 1.992e-05
        [call_graph_tuple_transform]: 1.858e-05
        [tuple_list_get_item_eliminator]: 5.67001e-06
        [none_parameter_eliminate]: 1.86e-06
        [renormalize]: 5.09986e-07
        [switch_simplify]: 5.22e-06
[remove_dup_value]: 1.342e-05
[partial_unused_args_eliminate]: 2.64999e-06
[environ_conv]: 5.79999e-06
[add_recomputation]: 5.182e-05
[cse_after_recomputation]: 1.93e-05, [1]
    [Cycle 1]: 1.385e-05, [1]
        [cse]: 8.40001e-06
[auto_monad_reorder]: 1.62e-05
[get_jit_bprop_graph]: 2.24999e-06
[rewriter_after_jit_bprop_graph]: 5.24998e-06
[opt_after_jit_grad]: 0.00047067
[symbol_engine_optimizer]: 7.578e-05, [1]
    [Cycle 1]: 6.849e-05, [6]
        [build]: 3.98999e-06
        [elim_shapecalc]: 7.73999e-06
        [elim_not_effective]: 1.343e-05
        [opt_reshape]: 6.15002e-06
        [fold_const_symbol]: 9.23002e-06
        [renormalize]: 4.69998e-07
[validate]: 3.947e-05
Sums
    bootstrap                            :     0.000314s :  3.34%
    type_inference                       :     0.004643s : 49.35%
    event_method                         :     0.000018s :  0.19%
    auto_monad                           :     0.000067s :  0.71%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.37%
    rewriter_before_opt_a                :     0.000073s :  0.78%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000068s :  0.73%
    jit_opt_a.loop_unroll                :     0.000036s :  0.38%
    jit_opt_a.a_1                        :     0.000697s :  7.41%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.36%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.001160s : 12.33%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.21%
    jit_opt_a.cse                        :     0.000045s :  0.47%
    jit_opt_a.replace_applicator         :     0.000019s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.13%
    rewriter_after_opt_a                 :     0.000038s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000696s :  7.40%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000027s :  0.29%
    loop_unroll                          :     0.000434s :  4.62%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.21%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000013s :  0.14%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000052s :  0.55%
    cse_after_recomputation.cse          :     0.000008s :  0.09%
    auto_monad_reorder                   :     0.000016s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000471s :  5.00%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000039s :  0.42%


Time group info:
------[substitution.]   0.000261    23
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000002s :      2: substitution.fold_const_symbol
 2.05% :     0.000005s :      3: substitution.graph_param_transform
86.60% :     0.000226s :      4: substitution.inline
 1.24% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.88% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.68% :     0.000004s :      2: substitution.replace_old_param
 5.16% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004575     2
80.86% :     0.003700s :      1: type_inference.infer
19.14% :     0.000876s :      1: type_inference.specialize
------[replace.]   0.000077     6
68.02% :     0.000053s :      4: replace.inline
31.98% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000236     6
94.76% :     0.000223s :      4: match.inline
 5.24% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.30% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.17% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.14% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.06% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.46% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.93% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.77% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.30% :     0.000002s :     10: predicate.environ_get_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.88% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 6.21% :     0.000007s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.55% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.53% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.12% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.03% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.86% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.23% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.82% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.26% :     0.000000s :      3: predicate.reset_defer_inline
 1.30% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.23% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.14% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.33% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.42% :     0.000003s :     16: predicate.switch_defer_inline
 2.31% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.83% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.27% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.29% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.62% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.39% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000725    12
48.66% :     0.000353s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.46% :     0.000054s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.88% :     0.000318s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016270    72
 0.34% :     0.000055s :      1: add_recomputation
 0.43% :     0.000070s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 2.05% :     0.000334s :      1: bootstrap
 0.18% :     0.000030s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.13% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.13% :     0.000022s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
41.17% :     0.006698s :      1: jit_opt_a
 0.88% :     0.000144s :      1: jit_opt_after_cconv
 0.32% :     0.000053s :      1: jit_opt_b
 2.72% :     0.000442s :      1: loop_unroll
 4.34% :     0.000706s :      1: mutable_eliminate
 5.64% :     0.000917s :     26: opt.transform.jit_opt_a
 0.28% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.12% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000033s :      4: opt.transform.symbol_engine_opt
 2.94% :     0.000478s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000037s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000016s :      1: remove_dup_value
 4.38% :     0.000712s :      1: renormalize.infer
 2.69% :     0.000438s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000041s :      1: rewriter_after_opt_a
 0.47% :     0.000076s :      1: rewriter_before_opt_a
 0.48% :     0.000078s :      1: symbol_engine_optimizer
28.63% :     0.004658s :      1: type_inference


TotalTime = 0.0150235, [30]
[bootstrap]: 0.00038019
[type_inference]: 0.00507497
[event_method]: 2.177e-05
[auto_monad]: 6.765e-05
[graph_reusing]: 5.97001e-06
[pre_auto_parallel]: 2.83998e-06
[py_interpret_to_execute]: 4.287e-05
[rewriter_before_opt_a]: 8.47e-05
[expand_dump_flag]: 3.41999e-06
[jit_opt_a]: 0.00678138, [2]
    [Cycle 1]: 0.00188499, [27]
        [switch_simplify]: 5.986e-05
        [loop_unroll]: 3.046e-05
        [a_1]: 0.00061534
        [with_stream_mark]: 2.012e-05
        [recompute_prepare]: 8.80999e-06
        [updatestate_depend_eliminate]: 4.27998e-06
        [updatestate_assign_eliminate]: 3.31001e-06
        [updatestate_loads_eliminate]: 2.66e-06
        [parameter_eliminate]: 2.22001e-06
        [specialize_transform]: 6.10002e-06
        [updatestate_useless_node_eliminater]: 5.37001e-06
        [accelerated_algorithm]: 6.36e-06
        [meta_shard_fg_expand]: 2.22999e-06
        [get_grad_eliminate_]: 5.93002e-06
        [merge_forward]: 4.42e-06
        [cell_reuse_recompute_pass]: 1.46002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.546e-05
        [j_node_and_user_rematch]: 9.53002e-06
        [meta_fg_expand]: 2.76e-06
        [replace_old_param]: 9.98998e-06
        [inline_without_move]: 6.04001e-06
        [renormalize]: 0.00079858
        [add_forward_monad_depend]: 6.29999e-06
        [auto_monad_grad]: 2.39999e-06
        [auto_monad_eliminator]: 1.473e-05
        [cse]: 2.929e-05
        [replace_applicator]: 1.439e-05
    [Cycle 2]: 0.00038642, [27]
        [switch_simplify]: 5.61e-06
        [loop_unroll]: 5.13002e-06
        [a_1]: 9.334e-05
        [with_stream_mark]: 1.247e-05
        [recompute_prepare]: 5.34e-06
        [updatestate_depend_eliminate]: 3.06001e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.08002e-06
        [parameter_eliminate]: 1.33002e-06
        [specialize_transform]: 5.67001e-06
        [updatestate_useless_node_eliminater]: 5.24e-06
        [accelerated_algorithm]: 6.12999e-06
        [meta_shard_fg_expand]: 1.27999e-06
        [get_grad_eliminate_]: 4.99e-06
        [merge_forward]: 2.70997e-06
        [cell_reuse_recompute_pass]: 1.67999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.405e-05
        [j_node_and_user_rematch]: 7.88001e-06
        [meta_fg_expand]: 1.87001e-06
        [replace_old_param]: 7.65998e-06
        [inline_without_move]: 3.965e-05
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 2.21998e-06
        [auto_monad_grad]: 9.70002e-07
        [auto_monad_eliminator]: 7.98999e-06
        [cse]: 1.43e-05
        [replace_applicator]: 5.94999e-06
[py_interpret_to_execute_after_opt_a]: 1.393e-05
[rewriter_after_opt_a]: 4.484e-05
[convert_after_rewriter]: 7.8e-06
[order_py_execute_after_rewriter]: 5.57999e-06
[mutable_eliminate]: 0.00067666
[jit_opt_b]: 5.84e-05, [1]
    [Cycle 1]: 4.966e-05, [2]
        [frontend_op_eliminate]: 1.915e-05
        [inline_after_opt_a]: 1.673e-05
[cconv]: 3.098e-05
[loop_unroll]: 0.00053369
[jit_opt_after_cconv]: 0.00015815, [1]
    [Cycle 1]: 0.00015023, [11]
        [c_1]: 2.337e-05
        [parameter_eliminate]: 2.98e-06
        [updatestate_depend_eliminate]: 8.23001e-06
        [updatestate_assign_eliminate]: 2.94001e-06
        [updatestate_loads_eliminate]: 2.20002e-06
        [cse]: 2.352e-05
        [call_graph_tuple_transform]: 2.161e-05
        [tuple_list_get_item_eliminator]: 5.69999e-06
        [none_parameter_eliminate]: 1.70001e-06
        [renormalize]: 7.89994e-07
        [switch_simplify]: 5.47999e-06
[remove_dup_value]: 1.474e-05
[partial_unused_args_eliminate]: 2.66e-06
[environ_conv]: 5.30999e-06
[add_recomputation]: 5.046e-05
[cse_after_recomputation]: 2.248e-05, [1]
    [Cycle 1]: 1.652e-05, [1]
        [cse]: 9.84001e-06
[auto_monad_reorder]: 1.429e-05
[get_jit_bprop_graph]: 2.32001e-06
[rewriter_after_jit_bprop_graph]: 8.08001e-06
[opt_after_jit_grad]: 0.00056085
[symbol_engine_optimizer]: 8.419e-05, [1]
    [Cycle 1]: 7.551e-05, [6]
        [build]: 5.00999e-06
        [elim_shapecalc]: 8.36002e-06
        [elim_not_effective]: 1.41e-05
        [opt_reshape]: 6.05002e-06
        [fold_const_symbol]: 1.015e-05
        [renormalize]: 9.00007e-07
[validate]: 4.393e-05
Sums
    bootstrap                            :     0.000380s :  3.86%
    type_inference                       :     0.005075s : 51.58%
    event_method                         :     0.000022s :  0.22%
    auto_monad                           :     0.000068s :  0.69%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000043s :  0.44%
    rewriter_before_opt_a                :     0.000085s :  0.86%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000065s :  0.67%
    jit_opt_a.loop_unroll                :     0.000036s :  0.36%
    jit_opt_a.a_1                        :     0.000709s :  7.20%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000018s :  0.18%
    jit_opt_a.inline_without_move        :     0.000046s :  0.46%
    jit_opt_a.renormalize                :     0.000799s :  8.12%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.23%
    jit_opt_a.cse                        :     0.000044s :  0.44%
    jit_opt_a.replace_applicator         :     0.000020s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.14%
    rewriter_after_opt_a                 :     0.000045s :  0.46%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000677s :  6.88%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.17%
    cconv                                :     0.000031s :  0.31%
    loop_unroll                          :     0.000534s :  5.42%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.24%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.15%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000050s :  0.51%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000014s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.08%
    opt_after_jit_grad                   :     0.000561s :  5.70%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.45%


Time group info:
------[substitution.]   0.000244    23
 0.91% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.64% :     0.000006s :      3: substitution.graph_param_transform
84.68% :     0.000207s :      4: substitution.inline
 1.35% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.36% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000005s :      2: substitution.replace_old_param
 5.45% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004989     2
78.86% :     0.003934s :      1: type_inference.infer
21.14% :     0.001055s :      1: type_inference.specialize
------[replace.]   0.000081     6
65.47% :     0.000053s :      4: replace.inline
34.53% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000215     6
94.38% :     0.000203s :      4: match.inline
 5.62% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000125   691
 1.66% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.45% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 2.31% :     0.000003s :     10: predicate.arithmetic_simplify
 1.11% :     0.000001s :     10: predicate.cast_eliminate
 0.58% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.40% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.11% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.10% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.57% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.20% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.98% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.20% :     0.000000s :      3: predicate.fold_const_symbol
 1.09% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.73% :     0.000007s :     22: predicate.inline
 1.17% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.54% :     0.000002s :      6: predicate.less_batch_normalization
 1.99% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.60% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.88% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.30% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 1.13% :     0.000001s :     10: predicate.minmaximum_grad
 1.44% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 1.97% :     0.000002s :     15: predicate.partial_eliminate
 0.98% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.62% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.43% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.70% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.96% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.15% :     0.000001s :      6: predicate.special_op_eliminate
 1.14% :     0.000001s :      6: predicate.specialize_transform
 1.19% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.05% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.34% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.63% :     0.000003s :     16: predicate.switch_defer_inline
 2.39% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.44% :     0.000009s :     53: predicate.switch_simplify
 1.07% :     0.000001s :     10: predicate.tile_eliminate
 1.06% :     0.000001s :     10: predicate.transpose_eliminate
 2.47% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.38% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.13% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.66% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.54% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.39% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.86% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000744    12
48.21% :     0.000359s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.39% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.40% :     0.000345s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016839    72
 0.32% :     0.000054s :      1: add_recomputation
 0.43% :     0.000072s :      1: auto_monad
 0.10% :     0.000017s :      1: auto_monad_reorder
 2.38% :     0.000401s :      1: bootstrap
 0.20% :     0.000034s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000009s :      1: environ_conv
 0.16% :     0.000027s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
40.30% :     0.006786s :      1: jit_opt_a
 0.96% :     0.000162s :      1: jit_opt_after_cconv
 0.37% :     0.000062s :      1: jit_opt_b
 3.24% :     0.000545s :      1: loop_unroll
 4.09% :     0.000688s :      1: mutable_eliminate
 5.75% :     0.000968s :     26: opt.transform.jit_opt_a
 0.31% :     0.000053s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000028s :      4: opt.transform.jit_opt_b
 0.09% :     0.000015s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.40% :     0.000572s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000046s :      1: py_interpret_to_execute
 0.10% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.35% :     0.000396s :      1: renormalize.infer
 2.33% :     0.000392s :      1: renormalize.specialize
 0.06% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.29% :     0.000049s :      1: rewriter_after_opt_a
 0.53% :     0.000090s :      1: rewriter_before_opt_a
 0.52% :     0.000087s :      1: symbol_engine_optimizer
30.26% :     0.005095s :      1: type_inference


TotalTime = 0.0127041, [30]
[bootstrap]: 0.00034205
[type_inference]: 0.00457871
[event_method]: 1.721e-05
[auto_monad]: 6.329e-05
[graph_reusing]: 5.32999e-06
[pre_auto_parallel]: 2.51e-06
[py_interpret_to_execute]: 3.606e-05
[rewriter_before_opt_a]: 8.565e-05
[expand_dump_flag]: 3.21001e-06
[jit_opt_a]: 0.00555364, [2]
    [Cycle 1]: 0.00177193, [27]
        [switch_simplify]: 4.864e-05
        [loop_unroll]: 2.923e-05
        [a_1]: 0.00059247
        [with_stream_mark]: 1.71e-05
        [recompute_prepare]: 7.55998e-06
        [updatestate_depend_eliminate]: 3.04001e-06
        [updatestate_assign_eliminate]: 2.14999e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.33002e-06
        [specialize_transform]: 6.38e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 5.67999e-06
        [meta_shard_fg_expand]: 1.64998e-06
        [get_grad_eliminate_]: 2.188e-05
        [merge_forward]: 3.21001e-06
        [cell_reuse_recompute_pass]: 9.99979e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.259e-05
        [j_node_and_user_rematch]: 9.02e-06
        [meta_fg_expand]: 2.10002e-06
        [replace_old_param]: 9.76e-06
        [inline_without_move]: 5.38002e-06
        [renormalize]: 0.00074802
        [add_forward_monad_depend]: 4.79e-06
        [auto_monad_grad]: 2.90002e-06
        [auto_monad_eliminator]: 1.089e-05
        [cse]: 1.925e-05
        [replace_applicator]: 1.17e-05
    [Cycle 2]: 0.00031811, [27]
        [switch_simplify]: 6.02999e-06
        [loop_unroll]: 4.93001e-06
        [a_1]: 8.857e-05
        [with_stream_mark]: 9.22999e-06
        [recompute_prepare]: 4.93001e-06
        [updatestate_depend_eliminate]: 2.41998e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.90001e-06
        [parameter_eliminate]: 1.22999e-06
        [specialize_transform]: 4.90999e-06
        [updatestate_useless_node_eliminater]: 5.42999e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.26002e-06
        [get_grad_eliminate_]: 4.57998e-06
        [merge_forward]: 2.21998e-06
        [cell_reuse_recompute_pass]: 1.47001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.198e-05
        [j_node_and_user_rematch]: 7.38e-06
        [meta_fg_expand]: 1.42e-06
        [replace_old_param]: 1.392e-05
        [inline_without_move]: 4.70999e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.10999e-06
        [auto_monad_grad]: 8.09989e-07
        [auto_monad_eliminator]: 5.20999e-06
        [cse]: 1.063e-05
        [replace_applicator]: 5.04003e-06
[py_interpret_to_execute_after_opt_a]: 8.92999e-06
[rewriter_after_opt_a]: 2.953e-05
[convert_after_rewriter]: 6.01e-06
[order_py_execute_after_rewriter]: 4.74998e-06
[mutable_eliminate]: 0.00054279
[jit_opt_b]: 4.779e-05, [1]
    [Cycle 1]: 4.147e-05, [2]
        [frontend_op_eliminate]: 1.562e-05
        [inline_after_opt_a]: 1.489e-05
[cconv]: 2.029e-05
[loop_unroll]: 0.00041649
[jit_opt_after_cconv]: 0.00013357, [1]
    [Cycle 1]: 0.00012686, [11]
        [c_1]: 2.029e-05
        [parameter_eliminate]: 2.87002e-06
        [updatestate_depend_eliminate]: 5.05001e-06
        [updatestate_assign_eliminate]: 2.23002e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 1.627e-05
        [call_graph_tuple_transform]: 1.673e-05
        [tuple_list_get_item_eliminator]: 5.36998e-06
        [none_parameter_eliminate]: 1.50001e-06
        [renormalize]: 5.00004e-07
        [switch_simplify]: 5.22999e-06
[remove_dup_value]: 1.094e-05
[partial_unused_args_eliminate]: 1.58002e-06
[environ_conv]: 3.90998e-06
[add_recomputation]: 3.49e-05
[cse_after_recomputation]: 1.953e-05, [1]
    [Cycle 1]: 1.431e-05, [1]
        [cse]: 8.62e-06
[auto_monad_reorder]: 1.157e-05
[get_jit_bprop_graph]: 1.50999e-06
[rewriter_after_jit_bprop_graph]: 4.86002e-06
[opt_after_jit_grad]: 0.00045147
[symbol_engine_optimizer]: 7.098e-05, [1]
    [Cycle 1]: 6.455e-05, [6]
        [build]: 3.16999e-06
        [elim_shapecalc]: 7.28e-06
        [elim_not_effective]: 1.255e-05
        [opt_reshape]: 5.50001e-06
        [fold_const_symbol]: 8.37e-06
        [renormalize]: 4.50003e-07
[validate]: 2.995e-05
Sums
    bootstrap                            :     0.000342s :  3.95%
    type_inference                       :     0.004579s : 52.86%
    event_method                         :     0.000017s :  0.20%
    auto_monad                           :     0.000063s :  0.73%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.42%
    rewriter_before_opt_a                :     0.000086s :  0.99%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000055s :  0.63%
    jit_opt_a.loop_unroll                :     0.000034s :  0.39%
    jit_opt_a.a_1                        :     0.000681s :  7.86%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000026s :  0.31%
    jit_opt_a.merge_forward              :     0.000005s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.28%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000024s :  0.27%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000748s :  8.64%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.19%
    jit_opt_a.cse                        :     0.000030s :  0.34%
    jit_opt_a.replace_applicator         :     0.000017s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.10%
    rewriter_after_opt_a                 :     0.000030s :  0.34%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000543s :  6.27%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000020s :  0.23%
    loop_unroll                          :     0.000416s :  4.81%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000011s :  0.13%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.05%
    add_recomputation                    :     0.000035s :  0.40%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000012s :  0.13%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000451s :  5.21%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000030s :  0.35%


Time group info:
------[substitution.]   0.000205    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.54% :     0.000001s :      2: substitution.fold_const_symbol
 1.91% :     0.000004s :      3: substitution.graph_param_transform
81.68% :     0.000167s :      4: substitution.inline
 1.43% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.06% :     0.000004s :      4: substitution.remove_not_recompute_node
 5.49% :     0.000011s :      2: substitution.replace_old_param
 5.82% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004518     2
80.53% :     0.003638s :      1: type_inference.infer
19.47% :     0.000879s :      1: type_inference.specialize
------[replace.]   0.000071     6
65.14% :     0.000046s :      4: replace.inline
34.86% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000175     6
93.86% :     0.000165s :      4: match.inline
 6.14% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000137   691
 1.09% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.02% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 0.86% :     0.000001s :     10: predicate.addn_check_dump
 1.08% :     0.000001s :     10: predicate.addn_zero_filter
 1.71% :     0.000002s :     10: predicate.arithmetic_simplify
 0.98% :     0.000001s :     10: predicate.cast_eliminate
 0.35% :     0.000000s :      3: predicate.check_bprop_eliminate
 0.88% :     0.000001s :     10: predicate.compare_switch_simplify
 1.17% :     0.000002s :     10: predicate.depend_value_elim
 0.92% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.04% :     0.000001s :     10: predicate.dict_get_item_eliminator
 0.99% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.72% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 0.96% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.89% :     0.000001s :     10: predicate.environ_get_depend_swap
 0.93% :     0.000001s :     10: predicate.environ_get_eliminate
 1.18% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 0.86% :     0.000001s :      6: predicate.get_grad_eliminate
 0.21% :     0.000000s :      3: predicate.graph_param_transform
 4.33% :     0.000006s :     22: predicate.inline
 0.80% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 0.96% :     0.000001s :      6: predicate.less_batch_normalization
 1.22% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.48% :     0.000002s :     15: predicate.load_eliminater
 1.25% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.41% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.91% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.90% :     0.000001s :     10: predicate.merge_addn
 0.90% :     0.000001s :     10: predicate.minmaximum_grad
 1.44% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000001s :      3: predicate.opt_reshape
 1.75% :     0.000002s :     15: predicate.partial_eliminate
 1.12% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.42% :     0.000002s :     10: predicate.reduce_eliminate
 1.21% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.42% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.36% :     0.000002s :     18: predicate.replace_applicator
 0.45% :     0.000001s :      6: predicate.replace_old_param
 0.23% :     0.000000s :      3: predicate.reset_defer_inline
19.48% :     0.000027s :     10: predicate.reshape_eliminate
 1.01% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.65% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.11% :     0.000002s :     10: predicate.same_eliminate
 0.42% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.72% :     0.000001s :      6: predicate.special_op_eliminate
 0.96% :     0.000001s :      6: predicate.specialize_transform
 1.04% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 1.96% :     0.000003s :     16: predicate.switch_defer_inline
 1.68% :     0.000002s :     16: predicate.switch_layer_defer_inline
 6.22% :     0.000009s :     53: predicate.switch_simplify
 0.95% :     0.000001s :     10: predicate.tile_eliminate
 0.97% :     0.000001s :     10: predicate.transpose_eliminate
 1.15% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.02% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 2.74% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.51% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.39% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.33% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000632    12
49.30% :     0.000312s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.85% :     0.000037s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.86% :     0.000284s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014414    72
 0.26% :     0.000038s :      1: add_recomputation
 0.46% :     0.000067s :      1: auto_monad
 0.10% :     0.000014s :      1: auto_monad_reorder
 2.47% :     0.000357s :      1: bootstrap
 0.16% :     0.000023s :      1: cconv
 0.06% :     0.000008s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.14% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.02% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.56% :     0.005557s :      1: jit_opt_a
 0.95% :     0.000136s :      1: jit_opt_after_cconv
 0.35% :     0.000051s :      1: jit_opt_b
 2.94% :     0.000424s :      1: loop_unroll
 3.82% :     0.000551s :      1: mutable_eliminate
 6.30% :     0.000908s :     26: opt.transform.jit_opt_a
 0.31% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.19% :     0.000459s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000003s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000039s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.09% :     0.000013s :      1: remove_dup_value
 2.79% :     0.000402s :      1: renormalize.infer
 2.36% :     0.000340s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000032s :      1: rewriter_after_opt_a
 0.62% :     0.000089s :      1: rewriter_before_opt_a
 0.51% :     0.000074s :      1: symbol_engine_optimizer
31.85% :     0.004592s :      1: type_inference


TotalTime = 0.0134804, [30]
[bootstrap]: 0.00034143
[type_inference]: 0.00476113
[event_method]: 1.789e-05
[auto_monad]: 6.263e-05
[graph_reusing]: 6.15002e-06
[pre_auto_parallel]: 3.12002e-06
[py_interpret_to_execute]: 3.74e-05
[rewriter_before_opt_a]: 7.136e-05
[expand_dump_flag]: 3.68999e-06
[jit_opt_a]: 0.00603111, [2]
    [Cycle 1]: 0.00191086, [27]
        [switch_simplify]: 5.562e-05
        [loop_unroll]: 2.873e-05
        [a_1]: 0.00056755
        [with_stream_mark]: 1.905e-05
        [recompute_prepare]: 6.94999e-06
        [updatestate_depend_eliminate]: 3.42002e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 1.62999e-06
        [specialize_transform]: 6.01e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 6.26e-06
        [meta_shard_fg_expand]: 1.79e-06
        [get_grad_eliminate_]: 5.82001e-06
        [merge_forward]: 3.58999e-06
        [cell_reuse_recompute_pass]: 1.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.374e-05
        [j_node_and_user_rematch]: 9.10999e-06
        [meta_fg_expand]: 2.07001e-06
        [replace_old_param]: 8.85999e-06
        [inline_without_move]: 5.55001e-06
        [renormalize]: 0.00092062
        [add_forward_monad_depend]: 6.64001e-06
        [auto_monad_grad]: 2.56998e-06
        [auto_monad_eliminator]: 1.419e-05
        [cse]: 2.584e-05
        [replace_applicator]: 1.384e-05
    [Cycle 2]: 0.00033775, [27]
        [switch_simplify]: 6.32001e-06
        [loop_unroll]: 5.39e-06
        [a_1]: 9.294e-05
        [with_stream_mark]: 1.095e-05
        [recompute_prepare]: 4.85001e-06
        [updatestate_depend_eliminate]: 3.12002e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 2.14999e-06
        [parameter_eliminate]: 1.11997e-06
        [specialize_transform]: 5.10001e-06
        [updatestate_useless_node_eliminater]: 4.68001e-06
        [accelerated_algorithm]: 5.76998e-06
        [meta_shard_fg_expand]: 1.22e-06
        [get_grad_eliminate_]: 4.94998e-06
        [merge_forward]: 2.91999e-06
        [cell_reuse_recompute_pass]: 1.67001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.332e-05
        [j_node_and_user_rematch]: 7.87e-06
        [meta_fg_expand]: 2.12001e-06
        [replace_old_param]: 7.69002e-06
        [inline_without_move]: 4.70001e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.24998e-06
        [auto_monad_grad]: 9.60019e-07
        [auto_monad_eliminator]: 5.52999e-06
        [cse]: 1.079e-05
        [replace_applicator]: 4.91002e-06
[py_interpret_to_execute_after_opt_a]: 1.125e-05
[rewriter_after_opt_a]: 3.453e-05
[convert_after_rewriter]: 6.49999e-06
[order_py_execute_after_rewriter]: 5.76e-06
[mutable_eliminate]: 0.00060393
[jit_opt_b]: 5.17e-05, [1]
    [Cycle 1]: 4.464e-05, [2]
        [frontend_op_eliminate]: 1.824e-05
        [inline_after_opt_a]: 1.526e-05
[cconv]: 2.537e-05
[loop_unroll]: 0.00041791
[jit_opt_after_cconv]: 0.00013752, [1]
    [Cycle 1]: 0.00013126, [11]
        [c_1]: 1.977e-05
        [parameter_eliminate]: 2.36e-06
        [updatestate_depend_eliminate]: 5.16002e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 1.99999e-06
        [cse]: 1.763e-05
        [call_graph_tuple_transform]: 1.869e-05
        [tuple_list_get_item_eliminator]: 5.15001e-06
        [none_parameter_eliminate]: 1.39e-06
        [renormalize]: 5.89993e-07
        [switch_simplify]: 5.35999e-06
[remove_dup_value]: 1.378e-05
[partial_unused_args_eliminate]: 2.44999e-06
[environ_conv]: 5.20001e-06
[add_recomputation]: 4.226e-05
[cse_after_recomputation]: 2.069e-05, [1]
    [Cycle 1]: 1.501e-05, [1]
        [cse]: 9.10999e-06
[auto_monad_reorder]: 1.385e-05
[get_jit_bprop_graph]: 1.92001e-06
[rewriter_after_jit_bprop_graph]: 5.46002e-06
[opt_after_jit_grad]: 0.00046286
[symbol_engine_optimizer]: 7.502e-05, [1]
    [Cycle 1]: 6.794e-05, [6]
        [build]: 3.79002e-06
        [elim_shapecalc]: 7.71999e-06
        [elim_not_effective]: 1.38e-05
        [opt_reshape]: 5.98002e-06
        [fold_const_symbol]: 8.65001e-06
        [renormalize]: 5.3001e-07
[validate]: 3.561e-05
Sums
    bootstrap                            :     0.000341s :  3.75%
    type_inference                       :     0.004761s : 52.25%
    event_method                         :     0.000018s :  0.20%
    auto_monad                           :     0.000063s :  0.69%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000037s :  0.41%
    rewriter_before_opt_a                :     0.000071s :  0.78%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000062s :  0.68%
    jit_opt_a.loop_unroll                :     0.000034s :  0.37%
    jit_opt_a.a_1                        :     0.000660s :  7.25%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.12%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.18%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000921s : 10.10%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.22%
    jit_opt_a.cse                        :     0.000037s :  0.40%
    jit_opt_a.replace_applicator         :     0.000019s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000035s :  0.38%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000604s :  6.63%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000025s :  0.28%
    loop_unroll                          :     0.000418s :  4.59%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000018s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000042s :  0.46%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000014s :  0.15%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000463s :  5.08%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000036s :  0.39%


Time group info:
------[substitution.]   0.000226    23
 0.72% :     0.000002s :      2: substitution.elim_not_effective
 0.58% :     0.000001s :      2: substitution.fold_const_symbol
 2.39% :     0.000005s :      3: substitution.graph_param_transform
85.93% :     0.000194s :      4: substitution.inline
 1.36% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.22% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.79% :     0.000004s :      2: substitution.replace_old_param
 5.02% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004692     2
80.74% :     0.003788s :      1: type_inference.infer
19.26% :     0.000904s :      1: type_inference.specialize
------[replace.]   0.000074     6
66.64% :     0.000050s :      4: replace.inline
33.36% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000202     6
94.88% :     0.000192s :      4: match.inline
 5.12% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000119   691
 1.24% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.30% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.41% :     0.000002s :     10: predicate.addn_zero_filter
 2.69% :     0.000003s :     10: predicate.arithmetic_simplify
 1.26% :     0.000002s :     10: predicate.cast_eliminate
 0.66% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.59% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.59% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.33% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 4.84% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.66% :     0.000002s :      6: predicate.less_batch_normalization
 1.84% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.33% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.89% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.75% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 1.95% :     0.000002s :      3: predicate.mutable_eliminate
 0.46% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.42% :     0.000002s :     10: predicate.reduce_eliminate
 1.78% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.63% :     0.000002s :     18: predicate.replace_applicator
 0.82% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.34% :     0.000002s :     10: predicate.reshape_eliminate
 1.44% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.93% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.14% :     0.000001s :     10: predicate.same_eliminate
 0.58% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 1.00% :     0.000001s :      6: predicate.specialize_transform
 1.23% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.37% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.59% :     0.000009s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.70% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.36% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.68% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.92% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000689    12
49.41% :     0.000340s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.25% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.33% :     0.000305s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015329    72
 0.29% :     0.000045s :      1: add_recomputation
 0.43% :     0.000066s :      1: auto_monad
 0.11% :     0.000016s :      1: auto_monad_reorder
 2.33% :     0.000357s :      1: bootstrap
 0.18% :     0.000028s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000022s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
39.37% :     0.006034s :      1: jit_opt_a
 0.92% :     0.000140s :      1: jit_opt_after_cconv
 0.35% :     0.000054s :      1: jit_opt_b
 2.77% :     0.000425s :      1: loop_unroll
 3.99% :     0.000612s :      1: mutable_eliminate
 5.70% :     0.000874s :     26: opt.transform.jit_opt_a
 0.30% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000027s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.07% :     0.000470s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000040s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000016s :      1: remove_dup_value
 3.32% :     0.000509s :      1: renormalize.infer
 2.63% :     0.000404s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000037s :      1: rewriter_after_opt_a
 0.48% :     0.000074s :      1: rewriter_before_opt_a
 0.50% :     0.000077s :      1: symbol_engine_optimizer
31.15% :     0.004775s :      1: type_inference


TotalTime = 0.0117807, [30]
[bootstrap]: 0.00026467
[type_inference]: 0.00426294
[event_method]: 1.61e-05
[auto_monad]: 4.442e-05
[graph_reusing]: 4.08999e-06
[pre_auto_parallel]: 1.76e-06
[py_interpret_to_execute]: 2.985e-05
[rewriter_before_opt_a]: 6.407e-05
[expand_dump_flag]: 1.86998e-06
[jit_opt_a]: 0.00510712, [2]
    [Cycle 1]: 0.00149607, [27]
        [switch_simplify]: 4.712e-05
        [loop_unroll]: 2.89e-05
        [a_1]: 0.00047974
        [with_stream_mark]: 1.287e-05
        [recompute_prepare]: 7.42002e-06
        [updatestate_depend_eliminate]: 3.22997e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.73998e-06
        [parameter_eliminate]: 2.15002e-06
        [specialize_transform]: 6.28e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.10999e-06
        [meta_shard_fg_expand]: 1.96e-06
        [get_grad_eliminate_]: 5.07e-06
        [merge_forward]: 3.78999e-06
        [cell_reuse_recompute_pass]: 1.04003e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.346e-05
        [j_node_and_user_rematch]: 8.70001e-06
        [meta_fg_expand]: 2.37999e-06
        [replace_old_param]: 8.17e-06
        [inline_without_move]: 5.36002e-06
        [renormalize]: 0.00062913
        [add_forward_monad_depend]: 5.12999e-06
        [auto_monad_grad]: 1.99e-06
        [auto_monad_eliminator]: 1.264e-05
        [cse]: 2.552e-05
        [replace_applicator]: 1.056e-05
    [Cycle 2]: 0.0003185, [27]
        [switch_simplify]: 5.19e-06
        [loop_unroll]: 5.11002e-06
        [a_1]: 9.3e-05
        [with_stream_mark]: 8.08001e-06
        [recompute_prepare]: 4.99e-06
        [updatestate_depend_eliminate]: 2.61e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 4.82998e-06
        [updatestate_useless_node_eliminater]: 4.62998e-06
        [accelerated_algorithm]: 5.09e-06
        [meta_shard_fg_expand]: 1.25001e-06
        [get_grad_eliminate_]: 4.43001e-06
        [merge_forward]: 2.29001e-06
        [cell_reuse_recompute_pass]: 1.63002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.285e-05
        [j_node_and_user_rematch]: 8.05e-06
        [meta_fg_expand]: 1.56998e-06
        [replace_old_param]: 7.32002e-06
        [inline_without_move]: 4.75999e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 6.69999e-07
        [auto_monad_eliminator]: 4.82998e-06
        [cse]: 1.016e-05
        [replace_applicator]: 4.92999e-06
[py_interpret_to_execute_after_opt_a]: 8.57998e-06
[rewriter_after_opt_a]: 3.095e-05
[convert_after_rewriter]: 6.54001e-06
[order_py_execute_after_rewriter]: 7.75e-06
[mutable_eliminate]: 0.00048151
[jit_opt_b]: 4.838e-05, [1]
    [Cycle 1]: 4.196e-05, [2]
        [frontend_op_eliminate]: 1.622e-05
        [inline_after_opt_a]: 1.451e-05
[cconv]: 2.328e-05
[loop_unroll]: 0.00040631
[jit_opt_after_cconv]: 0.00014356, [1]
    [Cycle 1]: 0.00012852, [11]
        [c_1]: 2.144e-05
        [parameter_eliminate]: 2.27999e-06
        [updatestate_depend_eliminate]: 5.07999e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 2.06998e-06
        [cse]: 1.604e-05
        [call_graph_tuple_transform]: 1.734e-05
        [tuple_list_get_item_eliminator]: 5.47999e-06
        [none_parameter_eliminate]: 1.47999e-06
        [renormalize]: 2.80008e-07
        [switch_simplify]: 5.30999e-06
[remove_dup_value]: 1.393e-05
[partial_unused_args_eliminate]: 2.34001e-06
[environ_conv]: 5.22999e-06
[add_recomputation]: 4.512e-05
[cse_after_recomputation]: 2.08e-05, [1]
    [Cycle 1]: 1.517e-05, [1]
        [cse]: 9.64999e-06
[auto_monad_reorder]: 1.46e-05
[get_jit_bprop_graph]: 1.47001e-06
[rewriter_after_jit_bprop_graph]: 4.97999e-06
[opt_after_jit_grad]: 0.00045645
[symbol_engine_optimizer]: 7.104e-05, [1]
    [Cycle 1]: 6.485e-05, [6]
        [build]: 3.46001e-06
        [elim_shapecalc]: 7.65e-06
        [elim_not_effective]: 1.223e-05
        [opt_reshape]: 5.59998e-06
        [fold_const_symbol]: 8.22e-06
        [renormalize]: 3.69997e-07
[validate]: 3.139e-05
Sums
    bootstrap                            :     0.000265s :  3.34%
    type_inference                       :     0.004263s : 53.76%
    event_method                         :     0.000016s :  0.20%
    auto_monad                           :     0.000044s :  0.56%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000030s :  0.38%
    rewriter_before_opt_a                :     0.000064s :  0.81%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000052s :  0.66%
    jit_opt_a.loop_unroll                :     0.000034s :  0.43%
    jit_opt_a.a_1                        :     0.000573s :  7.22%
    jit_opt_a.with_stream_mark           :     0.000021s :  0.26%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000015s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.13%
    jit_opt_a.renormalize                :     0.000629s :  7.93%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.22%
    jit_opt_a.cse                        :     0.000036s :  0.45%
    jit_opt_a.replace_applicator         :     0.000015s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000031s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000008s :  0.10%
    mutable_eliminate                    :     0.000482s :  6.07%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000023s :  0.29%
    loop_unroll                          :     0.000406s :  5.12%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000016s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.07%
    add_recomputation                    :     0.000045s :  0.57%
    cse_after_recomputation.cse          :     0.000010s :  0.12%
    auto_monad_reorder                   :     0.000015s :  0.18%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000456s :  5.76%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000031s :  0.40%


Time group info:
------[substitution.]   0.000161    23
 1.16% :     0.000002s :      2: substitution.elim_not_effective
 0.78% :     0.000001s :      2: substitution.fold_const_symbol
 2.84% :     0.000005s :      3: substitution.graph_param_transform
84.50% :     0.000136s :      4: substitution.inline
 1.82% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.70% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.82% :     0.000003s :      2: substitution.replace_old_param
 4.38% :     0.000007s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004206     2
80.61% :     0.003390s :      1: type_inference.infer
19.39% :     0.000816s :      1: type_inference.specialize
------[replace.]   0.000061     6
66.32% :     0.000041s :      4: replace.inline
33.68% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000140     6
95.73% :     0.000134s :      4: match.inline
 4.27% :     0.000006s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.24% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.30% :     0.000001s :     10: predicate.addn_zero_filter
 2.11% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.45% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.40% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.21% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.84% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.24% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.34% :     0.000006s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.55% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.89% :     0.000002s :     15: predicate.load_eliminater
 1.41% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.07% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     10: predicate.merge_addn
 1.15% :     0.000001s :     10: predicate.minmaximum_grad
 1.69% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.26% :     0.000002s :     15: predicate.partial_eliminate
 1.36% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.92% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.31% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.46% :     0.000002s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.36% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.46% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.33% :     0.000008s :     53: predicate.switch_simplify
 1.26% :     0.000001s :     10: predicate.tile_eliminate
 1.61% :     0.000002s :     10: predicate.transpose_eliminate
 1.45% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.58% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.54% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.76% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.71% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000582    12
48.19% :     0.000281s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.00% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.81% :     0.000273s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013233    72
 0.36% :     0.000048s :      1: add_recomputation
 0.36% :     0.000048s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.08% :     0.000276s :      1: bootstrap
 0.20% :     0.000026s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000007s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000006s :      1: graph_reusing
38.62% :     0.005110s :      1: jit_opt_a
 1.11% :     0.000146s :      1: jit_opt_after_cconv
 0.38% :     0.000051s :      1: jit_opt_b
 3.12% :     0.000413s :      1: loop_unroll
 3.70% :     0.000490s :      1: mutable_eliminate
 5.82% :     0.000771s :     26: opt.transform.jit_opt_a
 0.35% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.51% :     0.000464s :      1: opt_after_jit_grad
 0.08% :     0.000010s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000032s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.35% :     0.000311s :      1: renormalize.infer
 2.35% :     0.000311s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000034s :      1: rewriter_after_opt_a
 0.51% :     0.000067s :      1: rewriter_before_opt_a
 0.56% :     0.000074s :      1: symbol_engine_optimizer
32.31% :     0.004276s :      1: type_inference


TotalTime = 0.0141875, [30]
[bootstrap]: 0.00032597
[type_inference]: 0.00476599
[event_method]: 1.987e-05
[auto_monad]: 6.887e-05
[graph_reusing]: 5.47999e-06
[pre_auto_parallel]: 2.39999e-06
[py_interpret_to_execute]: 3.91e-05
[rewriter_before_opt_a]: 7.576e-05
[expand_dump_flag]: 3.02002e-06
[jit_opt_a]: 0.00634188, [2]
    [Cycle 1]: 0.00204882, [27]
        [switch_simplify]: 6.222e-05
        [loop_unroll]: 2.94e-05
        [a_1]: 0.00064994
        [with_stream_mark]: 2.028e-05
        [recompute_prepare]: 9.77999e-06
        [updatestate_depend_eliminate]: 4.08999e-06
        [updatestate_assign_eliminate]: 2.78e-06
        [updatestate_loads_eliminate]: 2.59999e-06
        [parameter_eliminate]: 2.22001e-06
        [specialize_transform]: 6.21e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 6.60997e-06
        [meta_shard_fg_expand]: 2.21e-06
        [get_grad_eliminate_]: 5.67999e-06
        [merge_forward]: 3.56999e-06
        [cell_reuse_recompute_pass]: 1.77001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.562e-05
        [j_node_and_user_rematch]: 9.98998e-06
        [meta_fg_expand]: 2.85998e-06
        [replace_old_param]: 9.72999e-06
        [inline_without_move]: 5.47999e-06
        [renormalize]: 0.00091388
        [add_forward_monad_depend]: 5.87999e-06
        [auto_monad_grad]: 2.52001e-06
        [auto_monad_eliminator]: 1.646e-05
        [cse]: 2.942e-05
        [replace_applicator]: 1.572e-05
    [Cycle 2]: 0.00035839, [27]
        [switch_simplify]: 6.33e-06
        [loop_unroll]: 5.86998e-06
        [a_1]: 9.642e-05
        [with_stream_mark]: 1.179e-05
        [recompute_prepare]: 5.05001e-06
        [updatestate_depend_eliminate]: 3.38e-06
        [updatestate_assign_eliminate]: 2.39001e-06
        [updatestate_loads_eliminate]: 2.39999e-06
        [parameter_eliminate]: 1.21002e-06
        [specialize_transform]: 5.34998e-06
        [updatestate_useless_node_eliminater]: 5.53002e-06
        [accelerated_algorithm]: 6.01998e-06
        [meta_shard_fg_expand]: 1.40001e-06
        [get_grad_eliminate_]: 5.10999e-06
        [merge_forward]: 3.09001e-06
        [cell_reuse_recompute_pass]: 1.99999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.552e-05
        [j_node_and_user_rematch]: 8.17998e-06
        [meta_fg_expand]: 2.06998e-06
        [replace_old_param]: 9.51e-06
        [inline_without_move]: 4.76002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.27999e-06
        [auto_monad_grad]: 9.39996e-07
        [auto_monad_eliminator]: 5.92999e-06
        [cse]: 1.227e-05
        [replace_applicator]: 5.98998e-06
[py_interpret_to_execute_after_opt_a]: 1.184e-05
[rewriter_after_opt_a]: 3.733e-05
[convert_after_rewriter]: 7.03998e-06
[order_py_execute_after_rewriter]: 5.81e-06
[mutable_eliminate]: 0.00066576
[jit_opt_b]: 5.462e-05, [1]
    [Cycle 1]: 4.671e-05, [2]
        [frontend_op_eliminate]: 1.824e-05
        [inline_after_opt_a]: 1.635e-05
[cconv]: 3.323e-05
[loop_unroll]: 0.00050994
[jit_opt_after_cconv]: 0.00016585, [1]
    [Cycle 1]: 0.00015809, [11]
        [c_1]: 2.291e-05
        [parameter_eliminate]: 4.13001e-06
        [updatestate_depend_eliminate]: 8.34998e-06
        [updatestate_assign_eliminate]: 2.68e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [cse]: 2.873e-05
        [call_graph_tuple_transform]: 2.32e-05
        [tuple_list_get_item_eliminator]: 5.47999e-06
        [none_parameter_eliminate]: 1.49998e-06
        [renormalize]: 9.29984e-07
        [switch_simplify]: 5.82001e-06
[remove_dup_value]: 1.516e-05
[partial_unused_args_eliminate]: 2.66999e-06
[environ_conv]: 5.82999e-06
[add_recomputation]: 5.337e-05
[cse_after_recomputation]: 2.498e-05, [1]
    [Cycle 1]: 1.804e-05, [1]
        [cse]: 1.114e-05
[auto_monad_reorder]: 1.61e-05
[get_jit_bprop_graph]: 2.27001e-06
[rewriter_after_jit_bprop_graph]: 8.25e-06
[opt_after_jit_grad]: 0.00058313
[symbol_engine_optimizer]: 8.32e-05, [1]
    [Cycle 1]: 7.676e-05, [6]
        [build]: 4.86002e-06
        [elim_shapecalc]: 8.73001e-06
        [elim_not_effective]: 1.627e-05
        [opt_reshape]: 6.63e-06
        [fold_const_symbol]: 9.99999e-06
        [renormalize]: 9.60019e-07
[validate]: 4.429e-05
Sums
    bootstrap                            :     0.000326s :  3.40%
    type_inference                       :     0.004766s : 49.75%
    event_method                         :     0.000020s :  0.21%
    auto_monad                           :     0.000069s :  0.72%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000039s :  0.41%
    rewriter_before_opt_a                :     0.000076s :  0.79%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000069s :  0.72%
    jit_opt_a.loop_unroll                :     0.000035s :  0.37%
    jit_opt_a.a_1                        :     0.000746s :  7.79%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000013s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000019s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000914s :  9.54%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.23%
    jit_opt_a.cse                        :     0.000042s :  0.44%
    jit_opt_a.replace_applicator         :     0.000022s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.12%
    rewriter_after_opt_a                 :     0.000037s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000666s :  6.95%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.17%
    cconv                                :     0.000033s :  0.35%
    loop_unroll                          :     0.000510s :  5.32%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.09%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000029s :  0.30%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000023s :  0.24%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000053s :  0.56%
    cse_after_recomputation.cse          :     0.000011s :  0.12%
    auto_monad_reorder                   :     0.000016s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.09%
    opt_after_jit_grad                   :     0.000583s :  6.09%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000016s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000044s :  0.46%


Time group info:
------[substitution.]   0.000269    23
 1.15% :     0.000003s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.17% :     0.000006s :      3: substitution.graph_param_transform
84.78% :     0.000228s :      4: substitution.inline
 1.39% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.13% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.35% :     0.000006s :      2: substitution.replace_old_param
 5.52% :     0.000015s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004689     2
79.27% :     0.003717s :      1: type_inference.infer
20.73% :     0.000972s :      1: type_inference.specialize
------[replace.]   0.000091     6
65.06% :     0.000059s :      4: replace.inline
34.94% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000239     6
94.22% :     0.000225s :      4: match.inline
 5.78% :     0.000014s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000133   691
 1.11% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.58% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.94% :     0.000001s :     10: predicate.addn_check_dump
 1.19% :     0.000002s :     10: predicate.addn_zero_filter
 2.22% :     0.000003s :     10: predicate.arithmetic_simplify
 1.53% :     0.000002s :     10: predicate.cast_eliminate
 0.49% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.92% :     0.000001s :     10: predicate.compare_switch_simplify
 1.96% :     0.000003s :     10: predicate.depend_value_elim
 1.17% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.04% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.01% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.17% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.74% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.29% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.50% :     0.000002s :     10: predicate.environ_get_add_eliminate
 0.92% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.03% :     0.000001s :     10: predicate.environ_get_eliminate
 0.94% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.19% :     0.000000s :      3: predicate.fold_const_symbol
 1.25% :     0.000002s :      6: predicate.get_grad_eliminate
 0.22% :     0.000000s :      3: predicate.graph_param_transform
 5.51% :     0.000007s :     22: predicate.inline
 0.79% :     0.000001s :      6: predicate.inline_without_move
 0.32% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.46% :     0.000002s :      6: predicate.less_batch_normalization
 1.97% :     0.000003s :     12: predicate.list_to_tuple_eliminator_
 1.56% :     0.000002s :     15: predicate.load_eliminater
 1.86% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.86% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.23% :     0.000002s :     10: predicate.merge_addn
 0.95% :     0.000001s :     10: predicate.minmaximum_grad
 2.28% :     0.000003s :      3: predicate.mutable_eliminate
 0.65% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     15: predicate.partial_eliminate
 0.98% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.81% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.56% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.47% :     0.000002s :     18: predicate.replace_applicator
 0.76% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.16% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.85% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.45% :     0.000002s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.04% :     0.000001s :      6: predicate.special_op_eliminate
 0.89% :     0.000001s :      6: predicate.specialize_transform
 1.39% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.38% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.86% :     0.000004s :     16: predicate.switch_defer_inline
 1.86% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.06% :     0.000011s :     53: predicate.switch_simplify
 1.13% :     0.000002s :     10: predicate.tile_eliminate
 1.02% :     0.000001s :     10: predicate.transpose_eliminate
 1.66% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.02% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.02% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.65% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.41% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.44% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.73% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.32% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000754    12
48.11% :     0.000363s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.98% :     0.000053s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.91% :     0.000339s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016142    72
 0.35% :     0.000057s :      1: add_recomputation
 0.45% :     0.000072s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 2.14% :     0.000345s :      1: bootstrap
 0.22% :     0.000036s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000027s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000025s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.31% :     0.006345s :      1: jit_opt_a
 1.05% :     0.000169s :      1: jit_opt_after_cconv
 0.36% :     0.000058s :      1: jit_opt_b
 3.22% :     0.000520s :      1: loop_unroll
 4.20% :     0.000678s :      1: mutable_eliminate
 6.06% :     0.000978s :     26: opt.transform.jit_opt_a
 0.34% :     0.000054s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000028s :      4: opt.transform.jit_opt_b
 0.08% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.18% :     0.000029s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000038s :      4: opt.transform.symbol_engine_opt
 3.68% :     0.000595s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000041s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 2.88% :     0.000465s :      1: renormalize.infer
 2.74% :     0.000442s :      1: renormalize.specialize
 0.07% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000041s :      1: rewriter_after_opt_a
 0.49% :     0.000079s :      1: rewriter_before_opt_a
 0.54% :     0.000086s :      1: symbol_engine_optimizer
29.68% :     0.004791s :      1: type_inference


TotalTime = 0.0121035, [30]
[bootstrap]: 0.0003085
[type_inference]: 0.00438364
[event_method]: 1.691e-05
[auto_monad]: 6.306e-05
[graph_reusing]: 5.69e-06
[pre_auto_parallel]: 2.36e-06
[py_interpret_to_execute]: 3.194e-05
[rewriter_before_opt_a]: 7.094e-05
[expand_dump_flag]: 3.16999e-06
[jit_opt_a]: 0.00520799, [2]
    [Cycle 1]: 0.00147157, [27]
        [switch_simplify]: 5.287e-05
        [loop_unroll]: 2.86e-05
        [a_1]: 0.00051089
        [with_stream_mark]: 1.476e-05
        [recompute_prepare]: 6.62002e-06
        [updatestate_depend_eliminate]: 3.38999e-06
        [updatestate_assign_eliminate]: 2.99001e-06
        [updatestate_loads_eliminate]: 2.48e-06
        [parameter_eliminate]: 1.94999e-06
        [specialize_transform]: 6.00002e-06
        [updatestate_useless_node_eliminater]: 4.95001e-06
        [accelerated_algorithm]: 5.56e-06
        [meta_shard_fg_expand]: 1.89999e-06
        [get_grad_eliminate_]: 5.21998e-06
        [merge_forward]: 3.45e-06
        [cell_reuse_recompute_pass]: 1.02e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.466e-05
        [j_node_and_user_rematch]: 9.09e-06
        [meta_fg_expand]: 2.11e-06
        [replace_old_param]: 9.34998e-06
        [inline_without_move]: 5.32001e-06
        [renormalize]: 0.00056045
        [add_forward_monad_depend]: 4.72e-06
        [auto_monad_grad]: 2.51e-06
        [auto_monad_eliminator]: 1.296e-05
        [cse]: 2.537e-05
        [replace_applicator]: 1.096e-05
    [Cycle 2]: 0.00031222, [27]
        [switch_simplify]: 5.66998e-06
        [loop_unroll]: 5.02e-06
        [a_1]: 8.771e-05
        [with_stream_mark]: 8.21002e-06
        [recompute_prepare]: 4.99e-06
        [updatestate_depend_eliminate]: 2.37999e-06
        [updatestate_assign_eliminate]: 1.96003e-06
        [updatestate_loads_eliminate]: 1.91003e-06
        [parameter_eliminate]: 8.00006e-07
        [specialize_transform]: 4.82998e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 5.54998e-06
        [meta_shard_fg_expand]: 1.15001e-06
        [get_grad_eliminate_]: 4.39002e-06
        [merge_forward]: 2.34999e-06
        [cell_reuse_recompute_pass]: 1.34998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.23e-05
        [j_node_and_user_rematch]: 7.5e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 7.10002e-06
        [inline_without_move]: 4.80001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.08001e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 4.44002e-06
        [cse]: 9.97001e-06
        [replace_applicator]: 5.02999e-06
[py_interpret_to_execute_after_opt_a]: 7.37002e-06
[rewriter_after_opt_a]: 3.1e-05
[convert_after_rewriter]: 6.66e-06
[order_py_execute_after_rewriter]: 5.22e-06
[mutable_eliminate]: 0.00048801
[jit_opt_b]: 5.008e-05, [1]
    [Cycle 1]: 4.294e-05, [2]
        [frontend_op_eliminate]: 1.658e-05
        [inline_after_opt_a]: 1.453e-05
[cconv]: 2.397e-05
[loop_unroll]: 0.00040545
[jit_opt_after_cconv]: 0.00013251, [1]
    [Cycle 1]: 0.00012613, [11]
        [c_1]: 1.982e-05
        [parameter_eliminate]: 2.39999e-06
        [updatestate_depend_eliminate]: 4.97e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [cse]: 1.64e-05
        [call_graph_tuple_transform]: 1.719e-05
        [tuple_list_get_item_eliminator]: 5.40999e-06
        [none_parameter_eliminate]: 1.40999e-06
        [renormalize]: 4.00003e-07
        [switch_simplify]: 5.44e-06
[remove_dup_value]: 1.402e-05
[partial_unused_args_eliminate]: 2.11e-06
[environ_conv]: 5.17e-06
[add_recomputation]: 4.424e-05
[cse_after_recomputation]: 1.909e-05, [1]
    [Cycle 1]: 1.361e-05, [1]
        [cse]: 8.07e-06
[auto_monad_reorder]: 1.396e-05
[get_jit_bprop_graph]: 1.45999e-06
[rewriter_after_jit_bprop_graph]: 4.88001e-06
[opt_after_jit_grad]: 0.00048828
[symbol_engine_optimizer]: 7.094e-05, [1]
    [Cycle 1]: 6.469e-05, [6]
        [build]: 3.14999e-06
        [elim_shapecalc]: 7.2e-06
        [elim_not_effective]: 1.288e-05
        [opt_reshape]: 5.47001e-06
        [fold_const_symbol]: 8.03001e-06
        [renormalize]: 4.80009e-07
[validate]: 2.985e-05
Sums
    bootstrap                            :     0.000309s :  3.80%
    type_inference                       :     0.004384s : 53.99%
    event_method                         :     0.000017s :  0.21%
    auto_monad                           :     0.000063s :  0.78%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.39%
    rewriter_before_opt_a                :     0.000071s :  0.87%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000059s :  0.72%
    jit_opt_a.loop_unroll                :     0.000034s :  0.41%
    jit_opt_a.a_1                        :     0.000599s :  7.37%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000561s :  6.90%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000017s :  0.21%
    jit_opt_a.cse                        :     0.000035s :  0.44%
    jit_opt_a.replace_applicator         :     0.000016s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000007s :  0.09%
    rewriter_after_opt_a                 :     0.000031s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000488s :  6.01%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000024s :  0.30%
    loop_unroll                          :     0.000405s :  4.99%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.20%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000044s :  0.54%
    cse_after_recomputation.cse          :     0.000008s :  0.10%
    auto_monad_reorder                   :     0.000014s :  0.17%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000488s :  6.01%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000030s :  0.37%


Time group info:
------[substitution.]   0.000187    23
 1.16% :     0.000002s :      2: substitution.elim_not_effective
 0.61% :     0.000001s :      2: substitution.fold_const_symbol
 2.68% :     0.000005s :      3: substitution.graph_param_transform
83.05% :     0.000155s :      4: substitution.inline
 1.63% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.56% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.85% :     0.000003s :      2: substitution.replace_old_param
 6.45% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004322     2
79.62% :     0.003442s :      1: type_inference.infer
20.38% :     0.000881s :      1: type_inference.specialize
------[replace.]   0.000065     6
66.01% :     0.000043s :      4: replace.inline
33.99% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000164     6
93.32% :     0.000153s :      4: match.inline
 6.68% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.19% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.46% :     0.000002s :     10: predicate.addn_zero_filter
 2.04% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.53% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.17% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.19% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.53% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.00% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.62% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.19% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.76% :     0.000002s :     15: predicate.load_eliminater
 1.43% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 2.17% :     0.000002s :      3: predicate.mutable_eliminate
 0.42% :     0.000000s :      3: predicate.opt_reshape
 2.27% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.56% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 1.15% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.32% :     0.000001s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.26% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.56% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.32% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.47% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.75% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.80% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000618    12
51.31% :     0.000317s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.32% :     0.000027s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.37% :     0.000274s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013526    72
 0.35% :     0.000047s :      1: add_recomputation
 0.49% :     0.000067s :      1: auto_monad
 0.12% :     0.000016s :      1: auto_monad_reorder
 2.41% :     0.000326s :      1: bootstrap
 0.20% :     0.000027s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000021s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.53% :     0.005211s :      1: jit_opt_a
 1.00% :     0.000135s :      1: jit_opt_after_cconv
 0.39% :     0.000053s :      1: jit_opt_b
 3.05% :     0.000412s :      1: loop_unroll
 3.67% :     0.000497s :      1: mutable_eliminate
 5.93% :     0.000803s :     26: opt.transform.jit_opt_a
 0.33% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.66% :     0.000496s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000034s :      1: py_interpret_to_execute
 0.07% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.09% :     0.000282s :      1: renormalize.infer
 2.00% :     0.000271s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000034s :      1: rewriter_after_opt_a
 0.54% :     0.000074s :      1: rewriter_before_opt_a
 0.54% :     0.000073s :      1: symbol_engine_optimizer
32.50% :     0.004396s :      1: type_inference


TotalTime = 0.0140751, [30]
[bootstrap]: 0.00032923
[type_inference]: 0.00490301
[event_method]: 1.891e-05
[auto_monad]: 6.856e-05
[graph_reusing]: 6.36e-06
[pre_auto_parallel]: 2.90002e-06
[py_interpret_to_execute]: 4.121e-05
[rewriter_before_opt_a]: 7.396e-05
[expand_dump_flag]: 3.08e-06
[jit_opt_a]: 0.00614078, [2]
    [Cycle 1]: 0.00191834, [27]
        [switch_simplify]: 6.185e-05
        [loop_unroll]: 2.941e-05
        [a_1]: 0.00058937
        [with_stream_mark]: 2.112e-05
        [recompute_prepare]: 9.19e-06
        [updatestate_depend_eliminate]: 4.00998e-06
        [updatestate_assign_eliminate]: 3.49001e-06
        [updatestate_loads_eliminate]: 2.76e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 6.93e-06
        [updatestate_useless_node_eliminater]: 5.09998e-06
        [accelerated_algorithm]: 5.91e-06
        [meta_shard_fg_expand]: 2.10002e-06
        [get_grad_eliminate_]: 5.79999e-06
        [merge_forward]: 3.91001e-06
        [cell_reuse_recompute_pass]: 1.54e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.711e-05
        [j_node_and_user_rematch]: 9.77001e-06
        [meta_fg_expand]: 2.71999e-06
        [replace_old_param]: 9.84001e-06
        [inline_without_move]: 5.47999e-06
        [renormalize]: 0.0008497
        [add_forward_monad_depend]: 7.6e-06
        [auto_monad_grad]: 2.69999e-06
        [auto_monad_eliminator]: 1.676e-05
        [cse]: 3.135e-05
        [replace_applicator]: 1.499e-05
    [Cycle 2]: 0.00035624, [27]
        [switch_simplify]: 6.37001e-06
        [loop_unroll]: 5.23002e-06
        [a_1]: 9.476e-05
        [with_stream_mark]: 1.152e-05
        [recompute_prepare]: 5.41002e-06
        [updatestate_depend_eliminate]: 2.96999e-06
        [updatestate_assign_eliminate]: 2.29001e-06
        [updatestate_loads_eliminate]: 2.29999e-06
        [parameter_eliminate]: 1.47001e-06
        [specialize_transform]: 5.29e-06
        [updatestate_useless_node_eliminater]: 5.22e-06
        [accelerated_algorithm]: 5.32999e-06
        [meta_shard_fg_expand]: 1.44e-06
        [get_grad_eliminate_]: 4.95001e-06
        [merge_forward]: 3.70998e-06
        [cell_reuse_recompute_pass]: 1.94e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.601e-05
        [j_node_and_user_rematch]: 8.70001e-06
        [meta_fg_expand]: 2.02001e-06
        [replace_old_param]: 8.67e-06
        [inline_without_move]: 5.05999e-06
        [renormalize]: 8.9989e-08
        [add_forward_monad_depend]: 1.45999e-06
        [auto_monad_grad]: 7.99977e-07
        [auto_monad_eliminator]: 6.09001e-06
        [cse]: 1.18e-05
        [replace_applicator]: 4.80001e-06
[py_interpret_to_execute_after_opt_a]: 1.309e-05
[rewriter_after_opt_a]: 3.839e-05
[convert_after_rewriter]: 9.29e-06
[order_py_execute_after_rewriter]: 5.73002e-06
[mutable_eliminate]: 0.0006444
[jit_opt_b]: 6.188e-05, [1]
    [Cycle 1]: 5.338e-05, [2]
        [frontend_op_eliminate]: 2.371e-05
        [inline_after_opt_a]: 1.704e-05
[cconv]: 2.549e-05
[loop_unroll]: 0.00055941
[jit_opt_after_cconv]: 0.000147, [1]
    [Cycle 1]: 0.00013903, [11]
        [c_1]: 2.167e-05
        [parameter_eliminate]: 2.89001e-06
        [updatestate_depend_eliminate]: 5.94e-06
        [updatestate_assign_eliminate]: 2.16998e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [cse]: 2.003e-05
        [call_graph_tuple_transform]: 1.909e-05
        [tuple_list_get_item_eliminator]: 5.74999e-06
        [none_parameter_eliminate]: 1.67001e-06
        [renormalize]: 7.10017e-07
        [switch_simplify]: 5.59998e-06
[remove_dup_value]: 1.623e-05
[partial_unused_args_eliminate]: 2.51998e-06
[environ_conv]: 5.19e-06
[add_recomputation]: 4.939e-05
[cse_after_recomputation]: 2.347e-05, [1]
    [Cycle 1]: 1.654e-05, [1]
        [cse]: 9.83002e-06
[auto_monad_reorder]: 1.547e-05
[get_jit_bprop_graph]: 2.02001e-06
[rewriter_after_jit_bprop_graph]: 6.83e-06
[opt_after_jit_grad]: 0.00054593
[symbol_engine_optimizer]: 7.833e-05, [1]
    [Cycle 1]: 7.013e-05, [6]
        [build]: 4.22998e-06
        [elim_shapecalc]: 8.05e-06
        [elim_not_effective]: 1.35e-05
        [opt_reshape]: 5.87999e-06
        [fold_const_symbol]: 8.94e-06
        [renormalize]: 6.89994e-07
[validate]: 3.917e-05
Sums
    bootstrap                            :     0.000329s :  3.45%
    type_inference                       :     0.004903s : 51.32%
    event_method                         :     0.000019s :  0.20%
    auto_monad                           :     0.000069s :  0.72%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000041s :  0.43%
    rewriter_before_opt_a                :     0.000074s :  0.77%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000068s :  0.71%
    jit_opt_a.loop_unroll                :     0.000035s :  0.36%
    jit_opt_a.a_1                        :     0.000684s :  7.16%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.34%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.11%
    jit_opt_a.merge_forward              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000033s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000019s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000850s :  8.89%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.24%
    jit_opt_a.cse                        :     0.000043s :  0.45%
    jit_opt_a.replace_applicator         :     0.000020s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.14%
    rewriter_after_opt_a                 :     0.000038s :  0.40%
    convert_after_rewriter               :     0.000009s :  0.10%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000644s :  6.75%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.25%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.18%
    cconv                                :     0.000025s :  0.27%
    loop_unroll                          :     0.000559s :  5.86%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000016s :  0.17%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000049s :  0.52%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000546s :  5.71%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.41%


Time group info:
------[substitution.]   0.000239    23
 0.82% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.07% :     0.000005s :      3: substitution.graph_param_transform
84.22% :     0.000202s :      4: substitution.inline
 1.48% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.44% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.58% :     0.000006s :      2: substitution.replace_old_param
 5.88% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004833     2
79.54% :     0.003844s :      1: type_inference.infer
20.46% :     0.000989s :      1: type_inference.specialize
------[replace.]   0.000077     6
62.85% :     0.000049s :      4: replace.inline
37.15% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000211     6
94.11% :     0.000198s :      4: match.inline
 5.89% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000126   691
 1.26% :     0.000002s :     10: predicate.accumulaten_eliminater
 2.09% :     0.000003s :      3: predicate.ad_related_special_op_eliminate
 0.95% :     0.000001s :     10: predicate.addn_check_dump
 1.17% :     0.000001s :     10: predicate.addn_zero_filter
 2.08% :     0.000003s :     10: predicate.arithmetic_simplify
 1.18% :     0.000001s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.95% :     0.000001s :     10: predicate.compare_switch_simplify
 1.76% :     0.000002s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.07% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.18% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000001s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.31% :     0.000002s :     10: predicate.environ_add_const_eliminate
 0.99% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.09% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.02% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.74% :     0.000007s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.33% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.36% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.13% :     0.000003s :     15: predicate.load_eliminater
 2.30% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.73% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.78% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 0.99% :     0.000001s :     10: predicate.minmaximum_grad
 2.38% :     0.000003s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000003s :     15: predicate.partial_eliminate
 1.19% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.52% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.63% :     0.000001s :      3: predicate.reset_defer_inline
 1.17% :     0.000001s :     10: predicate.reshape_eliminate
 1.10% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.18% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.04% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.37% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.18% :     0.000004s :     16: predicate.switch_defer_inline
 1.97% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.56% :     0.000010s :     53: predicate.switch_simplify
 1.04% :     0.000001s :     10: predicate.tile_eliminate
 1.01% :     0.000001s :     10: predicate.transpose_eliminate
 1.26% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.61% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.50% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.30% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.89% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.52% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.65% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.75% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000722    12
48.88% :     0.000353s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.47% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.65% :     0.000322s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015908    72
 0.33% :     0.000053s :      1: add_recomputation
 0.46% :     0.000073s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 2.19% :     0.000348s :      1: bootstrap
 0.18% :     0.000028s :      1: cconv
 0.07% :     0.000011s :      1: convert_after_rewriter
 0.16% :     0.000026s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000009s :      1: graph_reusing
38.62% :     0.006144s :      1: jit_opt_a
 0.94% :     0.000150s :      1: jit_opt_after_cconv
 0.41% :     0.000065s :      1: jit_opt_b
 3.58% :     0.000570s :      1: loop_unroll
 4.13% :     0.000658s :      1: mutable_eliminate
 5.75% :     0.000914s :     26: opt.transform.jit_opt_a
 0.31% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000034s :      4: opt.transform.jit_opt_b
 0.10% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.18% :     0.000028s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.51% :     0.000558s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.28% :     0.000044s :      1: py_interpret_to_execute
 0.10% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000019s :      1: remove_dup_value
 2.81% :     0.000448s :      1: renormalize.infer
 2.46% :     0.000392s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000042s :      1: rewriter_after_opt_a
 0.48% :     0.000077s :      1: rewriter_before_opt_a
 0.51% :     0.000081s :      1: symbol_engine_optimizer
30.93% :     0.004920s :      1: type_inference


TotalTime = 0.0135252, [30]
[bootstrap]: 0.00030871
[type_inference]: 0.00451628
[event_method]: 1.697e-05
[auto_monad]: 6.255e-05
[graph_reusing]: 5.91e-06
[pre_auto_parallel]: 2.32001e-06
[py_interpret_to_execute]: 3.409e-05
[rewriter_before_opt_a]: 7.396e-05
[expand_dump_flag]: 3.8e-06
[jit_opt_a]: 0.00628503, [2]
    [Cycle 1]: 0.00202038, [27]
        [switch_simplify]: 6.079e-05
        [loop_unroll]: 3.173e-05
        [a_1]: 0.00058231
        [with_stream_mark]: 2.168e-05
        [recompute_prepare]: 7.55e-06
        [updatestate_depend_eliminate]: 3.62002e-06
        [updatestate_assign_eliminate]: 3.41001e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [parameter_eliminate]: 1.75001e-06
        [specialize_transform]: 6.26998e-06
        [updatestate_useless_node_eliminater]: 4.87998e-06
        [accelerated_algorithm]: 5.89999e-06
        [meta_shard_fg_expand]: 2.26998e-06
        [get_grad_eliminate_]: 5.59e-06
        [merge_forward]: 3.58e-06
        [cell_reuse_recompute_pass]: 1.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.437e-05
        [j_node_and_user_rematch]: 9.57999e-06
        [meta_fg_expand]: 1.96998e-06
        [replace_old_param]: 1.014e-05
        [inline_without_move]: 5.14998e-06
        [renormalize]: 0.00098665
        [add_forward_monad_depend]: 6.19001e-06
        [auto_monad_grad]: 2.66999e-06
        [auto_monad_eliminator]: 1.426e-05
        [cse]: 3.017e-05
        [replace_applicator]: 1.189e-05
    [Cycle 2]: 0.00032574, [27]
        [switch_simplify]: 5.47001e-06
        [loop_unroll]: 4.78001e-06
        [a_1]: 9.246e-05
        [with_stream_mark]: 1.117e-05
        [recompute_prepare]: 4.92999e-06
        [updatestate_depend_eliminate]: 2.81999e-06
        [updatestate_assign_eliminate]: 2.14e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.12e-06
        [specialize_transform]: 5.34998e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 5.56e-06
        [meta_shard_fg_expand]: 1.54998e-06
        [get_grad_eliminate_]: 4.70999e-06
        [merge_forward]: 2.89001e-06
        [cell_reuse_recompute_pass]: 2.00002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.298e-05
        [j_node_and_user_rematch]: 8.15e-06
        [meta_fg_expand]: 1.88002e-06
        [replace_old_param]: 8.04997e-06
        [inline_without_move]: 4.71002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 1.00001e-06
        [auto_monad_eliminator]: 4.88001e-06
        [cse]: 1.046e-05
        [replace_applicator]: 5.05999e-06
[py_interpret_to_execute_after_opt_a]: 1.152e-05
[rewriter_after_opt_a]: 3.716e-05
[convert_after_rewriter]: 6.76e-06
[order_py_execute_after_rewriter]: 4.89e-06
[mutable_eliminate]: 0.00059733
[jit_opt_b]: 5.728e-05, [1]
    [Cycle 1]: 4.932e-05, [2]
        [frontend_op_eliminate]: 1.966e-05
        [inline_after_opt_a]: 1.761e-05
[cconv]: 2.624e-05
[loop_unroll]: 0.00043438
[jit_opt_after_cconv]: 0.00014279, [1]
    [Cycle 1]: 0.00013651, [11]
        [c_1]: 2.134e-05
        [parameter_eliminate]: 2.35002e-06
        [updatestate_depend_eliminate]: 5.49e-06
        [updatestate_assign_eliminate]: 2.68998e-06
        [updatestate_loads_eliminate]: 1.93997e-06
        [cse]: 1.925e-05
        [call_graph_tuple_transform]: 1.934e-05
        [tuple_list_get_item_eliminator]: 5.30001e-06
        [none_parameter_eliminate]: 1.66002e-06
        [renormalize]: 8.2e-07
        [switch_simplify]: 5.55001e-06
[remove_dup_value]: 1.559e-05
[partial_unused_args_eliminate]: 2.37001e-06
[environ_conv]: 5.76e-06
[add_recomputation]: 4.97e-05
[cse_after_recomputation]: 2.1e-05, [1]
    [Cycle 1]: 1.516e-05, [1]
        [cse]: 9.14e-06
[auto_monad_reorder]: 1.428e-05
[get_jit_bprop_graph]: 2.11e-06
[rewriter_after_jit_bprop_graph]: 5.49998e-06
[opt_after_jit_grad]: 0.00047295
[symbol_engine_optimizer]: 7.899e-05, [1]
    [Cycle 1]: 7.106e-05, [6]
        [build]: 3.94002e-06
        [elim_shapecalc]: 7.62998e-06
        [elim_not_effective]: 1.426e-05
        [opt_reshape]: 5.81003e-06
        [fold_const_symbol]: 8.87e-06
        [renormalize]: 5.09986e-07
[validate]: 3.934e-05
Sums
    bootstrap                            :     0.000309s :  3.44%
    type_inference                       :     0.004516s : 50.32%
    event_method                         :     0.000017s :  0.19%
    auto_monad                           :     0.000063s :  0.70%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.38%
    rewriter_before_opt_a                :     0.000074s :  0.82%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000066s :  0.74%
    jit_opt_a.loop_unroll                :     0.000037s :  0.41%
    jit_opt_a.a_1                        :     0.000675s :  7.52%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.37%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000987s : 11.00%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.21%
    jit_opt_a.cse                        :     0.000041s :  0.45%
    jit_opt_a.replace_applicator         :     0.000017s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.13%
    rewriter_after_opt_a                 :     0.000037s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000597s :  6.66%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000018s :  0.20%
    cconv                                :     0.000026s :  0.29%
    loop_unroll                          :     0.000434s :  4.84%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000016s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000050s :  0.55%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000014s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000473s :  5.27%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.44%


Time group info:
------[substitution.]   0.000246    23
 1.13% :     0.000003s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 2.27% :     0.000006s :      3: substitution.graph_param_transform
85.28% :     0.000210s :      4: substitution.inline
 1.64% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.96% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.08% :     0.000005s :      2: substitution.replace_old_param
 5.13% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004446     2
79.80% :     0.003548s :      1: type_inference.infer
20.20% :     0.000898s :      1: type_inference.specialize
------[replace.]   0.000076     6
68.22% :     0.000052s :      4: replace.inline
31.78% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000219     6
94.76% :     0.000207s :      4: match.inline
 5.24% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000002s :     10: predicate.addn_zero_filter
 2.04% :     0.000002s :     10: predicate.arithmetic_simplify
 1.13% :     0.000001s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.76% :     0.000002s :     10: predicate.depend_value_elim
 1.07% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.07% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.16% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.15% :     0.000001s :     10: predicate.environ_get_eliminate
 1.26% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.26% :     0.000000s :      3: predicate.fold_const_symbol
 1.22% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.63% :     0.000007s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.74% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.22% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.79% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.85% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.79% :     0.000002s :     10: predicate.reshape_eliminate
 1.29% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.98% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.70% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.20% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.92% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.28% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.87% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.53% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.42% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.79% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.58% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000663    12
45.42% :     0.000301s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.91% :     0.000052s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.67% :     0.000310s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015450    72
 0.34% :     0.000053s :      1: add_recomputation
 0.43% :     0.000066s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 2.09% :     0.000323s :      1: bootstrap
 0.19% :     0.000029s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000021s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
40.70% :     0.006289s :      1: jit_opt_a
 0.94% :     0.000146s :      1: jit_opt_after_cconv
 0.39% :     0.000060s :      1: jit_opt_b
 2.87% :     0.000443s :      1: loop_unroll
 3.92% :     0.000606s :      1: mutable_eliminate
 5.79% :     0.000895s :     26: opt.transform.jit_opt_a
 0.31% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000030s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.12% :     0.000482s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000037s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000018s :      1: remove_dup_value
 3.65% :     0.000564s :      1: renormalize.infer
 2.68% :     0.000414s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000040s :      1: rewriter_after_opt_a
 0.50% :     0.000077s :      1: rewriter_before_opt_a
 0.53% :     0.000082s :      1: symbol_engine_optimizer
29.33% :     0.004532s :      1: type_inference


TotalTime = 0.0142133, [30]
[bootstrap]: 0.00033208
[type_inference]: 0.00533045
[event_method]: 1.928e-05
[auto_monad]: 6.74e-05
[graph_reusing]: 6.85998e-06
[pre_auto_parallel]: 3.19001e-06
[py_interpret_to_execute]: 3.77e-05
[rewriter_before_opt_a]: 7.601e-05
[expand_dump_flag]: 3.35e-06
[jit_opt_a]: 0.00596524, [2]
    [Cycle 1]: 0.00178853, [27]
        [switch_simplify]: 5.944e-05
        [loop_unroll]: 2.876e-05
        [a_1]: 0.00055659
        [with_stream_mark]: 1.877e-05
        [recompute_prepare]: 9.79999e-06
        [updatestate_depend_eliminate]: 3.73999e-06
        [updatestate_assign_eliminate]: 3.04001e-06
        [updatestate_loads_eliminate]: 2.49001e-06
        [parameter_eliminate]: 1.78002e-06
        [specialize_transform]: 6.71999e-06
        [updatestate_useless_node_eliminater]: 5.58002e-06
        [accelerated_algorithm]: 6.34999e-06
        [meta_shard_fg_expand]: 2.21e-06
        [get_grad_eliminate_]: 5.27999e-06
        [merge_forward]: 3.84002e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.55e-05
        [j_node_and_user_rematch]: 9.02e-06
        [meta_fg_expand]: 2.86e-06
        [replace_old_param]: 8.97999e-06
        [inline_without_move]: 5.57001e-06
        [renormalize]: 0.00078324
        [add_forward_monad_depend]: 6.06e-06
        [auto_monad_grad]: 2.17999e-06
        [auto_monad_eliminator]: 1.504e-05
        [cse]: 2.891e-05
        [replace_applicator]: 1.394e-05
    [Cycle 2]: 0.0003355, [27]
        [switch_simplify]: 5.72001e-06
        [loop_unroll]: 5.16002e-06
        [a_1]: 9.236e-05
        [with_stream_mark]: 9.74e-06
        [recompute_prepare]: 5.26998e-06
        [updatestate_depend_eliminate]: 3.13e-06
        [updatestate_assign_eliminate]: 2.04999e-06
        [updatestate_loads_eliminate]: 2.12001e-06
        [parameter_eliminate]: 9.30013e-07
        [specialize_transform]: 5.16998e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 1.55001e-06
        [get_grad_eliminate_]: 4.4e-06
        [merge_forward]: 3.04001e-06
        [cell_reuse_recompute_pass]: 1.37999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.348e-05
        [j_node_and_user_rematch]: 8.15999e-06
        [meta_fg_expand]: 2.05002e-06
        [replace_old_param]: 8.36002e-06
        [inline_without_move]: 4.80999e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.24998e-06
        [auto_monad_grad]: 7.60017e-07
        [auto_monad_eliminator]: 4.74002e-06
        [cse]: 1.133e-05
        [replace_applicator]: 1.471e-05
[py_interpret_to_execute_after_opt_a]: 1.202e-05
[rewriter_after_opt_a]: 3.643e-05
[convert_after_rewriter]: 7.50998e-06
[order_py_execute_after_rewriter]: 4.93001e-06
[mutable_eliminate]: 0.00060566
[jit_opt_b]: 6.041e-05, [1]
    [Cycle 1]: 5.131e-05, [2]
        [frontend_op_eliminate]: 2.243e-05
        [inline_after_opt_a]: 1.658e-05
[cconv]: 2.617e-05
[loop_unroll]: 0.00050888
[jit_opt_after_cconv]: 0.00014508, [1]
    [Cycle 1]: 0.0001374, [11]
        [c_1]: 2.197e-05
        [parameter_eliminate]: 2.93998e-06
        [updatestate_depend_eliminate]: 6.82002e-06
        [updatestate_assign_eliminate]: 2.26e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [cse]: 2.04e-05
        [call_graph_tuple_transform]: 1.875e-05
        [tuple_list_get_item_eliminator]: 5.24e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 5.3001e-07
        [switch_simplify]: 5.93998e-06
[remove_dup_value]: 1.476e-05
[partial_unused_args_eliminate]: 2.48e-06
[environ_conv]: 6.27001e-06
[add_recomputation]: 4.848e-05
[cse_after_recomputation]: 2.197e-05, [1]
    [Cycle 1]: 1.599e-05, [1]
        [cse]: 9.62001e-06
[auto_monad_reorder]: 1.522e-05
[get_jit_bprop_graph]: 2.01e-06
[rewriter_after_jit_bprop_graph]: 5.63997e-06
[opt_after_jit_grad]: 0.000529
[symbol_engine_optimizer]: 7.85e-05, [1]
    [Cycle 1]: 6.942e-05, [6]
        [build]: 4.33001e-06
        [elim_shapecalc]: 8.00999e-06
        [elim_not_effective]: 1.313e-05
        [opt_reshape]: 6.01e-06
        [fold_const_symbol]: 8.38001e-06
        [renormalize]: 4.80009e-07
[validate]: 3.935e-05
Sums
    bootstrap                            :     0.000332s :  3.41%
    type_inference                       :     0.005330s : 54.68%
    event_method                         :     0.000019s :  0.20%
    auto_monad                           :     0.000067s :  0.69%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000038s :  0.39%
    rewriter_before_opt_a                :     0.000076s :  0.78%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000065s :  0.67%
    jit_opt_a.loop_unroll                :     0.000034s :  0.35%
    jit_opt_a.a_1                        :     0.000649s :  6.66%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.18%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000783s :  8.04%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.20%
    jit_opt_a.cse                        :     0.000040s :  0.41%
    jit_opt_a.replace_applicator         :     0.000029s :  0.29%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.12%
    rewriter_after_opt_a                 :     0.000036s :  0.37%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000606s :  6.21%
    jit_opt_b.frontend_op_eliminate      :     0.000022s :  0.23%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.17%
    cconv                                :     0.000026s :  0.27%
    loop_unroll                          :     0.000509s :  5.22%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000015s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000048s :  0.50%
    cse_after_recomputation.cse          :     0.000010s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000529s :  5.43%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.13%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000039s :  0.40%


Time group info:
------[substitution.]   0.000214    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.57% :     0.000001s :      2: substitution.fold_const_symbol
 2.43% :     0.000005s :      3: substitution.graph_param_transform
84.06% :     0.000180s :      4: substitution.inline
 1.57% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.01% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.43% :     0.000005s :      2: substitution.replace_old_param
 5.96% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005254     2
80.53% :     0.004231s :      1: type_inference.infer
19.47% :     0.001023s :      1: type_inference.specialize
------[replace.]   0.000077     6
67.47% :     0.000052s :      4: replace.inline
32.53% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000189     6
93.82% :     0.000177s :      4: match.inline
 6.18% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.18% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.72% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.19% :     0.000001s :     10: predicate.addn_zero_filter
 2.72% :     0.000003s :     10: predicate.arithmetic_simplify
 1.80% :     0.000002s :     10: predicate.cast_eliminate
 0.40% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 2.06% :     0.000003s :     10: predicate.depend_value_elim
 1.17% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.10% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.43% :     0.000001s :      3: predicate.elim_not_effective
 0.73% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.00% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 5.22% :     0.000006s :     22: predicate.inline
 0.91% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.43% :     0.000002s :      6: predicate.less_batch_normalization
 1.33% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.65% :     0.000002s :     15: predicate.load_eliminater
 1.84% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.90% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 2.02% :     0.000002s :      3: predicate.mutable_eliminate
 0.63% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.35% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.79% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.76% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.43% :     0.000002s :     10: predicate.same_eliminate
 0.69% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.79% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.18% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.00% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.18% :     0.000004s :     16: predicate.switch_defer_inline
 1.89% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.39% :     0.000009s :     53: predicate.switch_simplify
 1.07% :     0.000001s :     10: predicate.tile_eliminate
 1.04% :     0.000001s :     10: predicate.transpose_eliminate
 1.30% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.70% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.86% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.66% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.36% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.55% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.61% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000728    12
51.90% :     0.000378s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.38% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.71% :     0.000318s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015938    72
 0.32% :     0.000052s :      1: add_recomputation
 0.44% :     0.000071s :      1: auto_monad
 0.12% :     0.000019s :      1: auto_monad_reorder
 2.21% :     0.000353s :      1: bootstrap
 0.18% :     0.000029s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
37.45% :     0.005968s :      1: jit_opt_a
 0.93% :     0.000148s :      1: jit_opt_after_cconv
 0.40% :     0.000063s :      1: jit_opt_b
 3.26% :     0.000520s :      1: loop_unroll
 3.87% :     0.000616s :      1: mutable_eliminate
 5.52% :     0.000879s :     26: opt.transform.jit_opt_a
 0.31% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000032s :      4: opt.transform.jit_opt_b
 0.10% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.40% :     0.000542s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000040s :      1: py_interpret_to_execute
 0.09% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 2.54% :     0.000404s :      1: renormalize.infer
 2.33% :     0.000371s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000040s :      1: rewriter_after_opt_a
 0.49% :     0.000079s :      1: rewriter_before_opt_a
 0.51% :     0.000081s :      1: symbol_engine_optimizer
33.57% :     0.005350s :      1: type_inference


TotalTime = 0.0131186, [30]
[bootstrap]: 0.0003247
[type_inference]: 0.00490897
[event_method]: 1.839e-05
[auto_monad]: 7.101e-05
[graph_reusing]: 5.71998e-06
[pre_auto_parallel]: 2.58998e-06
[py_interpret_to_execute]: 3.518e-05
[rewriter_before_opt_a]: 7.554e-05
[expand_dump_flag]: 4e-06
[jit_opt_a]: 0.00566362, [2]
    [Cycle 1]: 0.00160353, [27]
        [switch_simplify]: 5.373e-05
        [loop_unroll]: 2.818e-05
        [a_1]: 0.00050732
        [with_stream_mark]: 1.794e-05
        [recompute_prepare]: 6.54999e-06
        [updatestate_depend_eliminate]: 3.28998e-06
        [updatestate_assign_eliminate]: 2.90998e-06
        [updatestate_loads_eliminate]: 2.56998e-06
        [parameter_eliminate]: 1.71e-06
        [specialize_transform]: 5.92999e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.53002e-06
        [meta_shard_fg_expand]: 2.14e-06
        [get_grad_eliminate_]: 5.26002e-06
        [merge_forward]: 3.48e-06
        [cell_reuse_recompute_pass]: 1.22e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.376e-05
        [j_node_and_user_rematch]: 8.84e-06
        [meta_fg_expand]: 2.26998e-06
        [replace_old_param]: 8.68001e-06
        [inline_without_move]: 5.49e-06
        [renormalize]: 0.00068196
        [add_forward_monad_depend]: 5.44e-06
        [auto_monad_grad]: 2.63003e-06
        [auto_monad_eliminator]: 1.332e-05
        [cse]: 2.842e-05
        [replace_applicator]: 1.182e-05
    [Cycle 2]: 0.00031201, [27]
        [switch_simplify]: 5.19e-06
        [loop_unroll]: 4.84e-06
        [a_1]: 8.826e-05
        [with_stream_mark]: 8.98002e-06
        [recompute_prepare]: 5.07999e-06
        [updatestate_depend_eliminate]: 2.34001e-06
        [updatestate_assign_eliminate]: 1.93002e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 8.89995e-07
        [specialize_transform]: 4.62e-06
        [updatestate_useless_node_eliminater]: 4.70001e-06
        [accelerated_algorithm]: 5.12999e-06
        [meta_shard_fg_expand]: 1.40999e-06
        [get_grad_eliminate_]: 4.53001e-06
        [merge_forward]: 2.26998e-06
        [cell_reuse_recompute_pass]: 1.24e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.169e-05
        [j_node_and_user_rematch]: 7.83999e-06
        [meta_fg_expand]: 1.97999e-06
        [replace_old_param]: 7.06999e-06
        [inline_without_move]: 1.007e-05
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.08001e-06
        [auto_monad_grad]: 7.40023e-07
        [auto_monad_eliminator]: 4.92999e-06
        [cse]: 9.82999e-06
        [replace_applicator]: 4.63999e-06
[py_interpret_to_execute_after_opt_a]: 9.15001e-06
[rewriter_after_opt_a]: 3.259e-05
[convert_after_rewriter]: 6.52001e-06
[order_py_execute_after_rewriter]: 4.82e-06
[mutable_eliminate]: 0.00049466
[jit_opt_b]: 4.831e-05, [1]
    [Cycle 1]: 4.159e-05, [2]
        [frontend_op_eliminate]: 1.561e-05
        [inline_after_opt_a]: 1.475e-05
[cconv]: 2.464e-05
[loop_unroll]: 0.0004186
[jit_opt_after_cconv]: 0.00013466, [1]
    [Cycle 1]: 0.00012833, [11]
        [c_1]: 2.035e-05
        [parameter_eliminate]: 2.37001e-06
        [updatestate_depend_eliminate]: 5.59998e-06
        [updatestate_assign_eliminate]: 2.27001e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 1.662e-05
        [call_graph_tuple_transform]: 1.765e-05
        [tuple_list_get_item_eliminator]: 5.24e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 4.49974e-07
        [switch_simplify]: 5.24e-06
[remove_dup_value]: 1.462e-05
[partial_unused_args_eliminate]: 2.06e-06
[environ_conv]: 5.15001e-06
[add_recomputation]: 4.576e-05
[cse_after_recomputation]: 1.909e-05, [1]
    [Cycle 1]: 1.388e-05, [1]
        [cse]: 8.55999e-06
[auto_monad_reorder]: 1.461e-05
[get_jit_bprop_graph]: 1.98997e-06
[rewriter_after_jit_bprop_graph]: 4.55999e-06
[opt_after_jit_grad]: 0.00044975
[symbol_engine_optimizer]: 7.065e-05, [1]
    [Cycle 1]: 6.43e-05, [6]
        [build]: 3.36001e-06
        [elim_shapecalc]: 7.4e-06
        [elim_not_effective]: 1.244e-05
        [opt_reshape]: 5.54e-06
        [fold_const_symbol]: 8.13999e-06
        [renormalize]: 4.50003e-07
[validate]: 3.299e-05
Sums
    bootstrap                            :     0.000325s :  3.69%
    type_inference                       :     0.004909s : 55.77%
    event_method                         :     0.000018s :  0.21%
    auto_monad                           :     0.000071s :  0.81%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.40%
    rewriter_before_opt_a                :     0.000076s :  0.86%
    expand_dump_flag                     :     0.000004s :  0.05%
    jit_opt_a.switch_simplify            :     0.000059s :  0.67%
    jit_opt_a.loop_unroll                :     0.000033s :  0.38%
    jit_opt_a.a_1                        :     0.000596s :  6.77%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000025s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.18%
    jit_opt_a.inline_without_move        :     0.000016s :  0.18%
    jit_opt_a.renormalize                :     0.000682s :  7.75%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.21%
    jit_opt_a.cse                        :     0.000038s :  0.43%
    jit_opt_a.replace_applicator         :     0.000016s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.10%
    rewriter_after_opt_a                 :     0.000033s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000495s :  5.62%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000025s :  0.28%
    loop_unroll                          :     0.000419s :  4.76%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000046s :  0.52%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000450s :  5.11%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000033s :  0.37%


Time group info:
------[substitution.]   0.000185    23
 1.04% :     0.000002s :      2: substitution.elim_not_effective
 0.69% :     0.000001s :      2: substitution.fold_const_symbol
 2.49% :     0.000005s :      3: substitution.graph_param_transform
82.76% :     0.000153s :      4: substitution.inline
 1.64% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.39% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.05% :     0.000004s :      2: substitution.replace_old_param
 6.94% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004838     2
79.88% :     0.003865s :      1: type_inference.infer
20.12% :     0.000974s :      1: type_inference.specialize
------[replace.]   0.000065     6
66.67% :     0.000043s :      4: replace.inline
33.33% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000163     6
92.79% :     0.000151s :      4: match.inline
 7.21% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.26% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.15% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000001s :     10: predicate.addn_zero_filter
 2.03% :     0.000002s :     10: predicate.arithmetic_simplify
 1.42% :     0.000002s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.13% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000001s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.47% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.58% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.23% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.72% :     0.000006s :     22: predicate.inline
 0.97% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.17% :     0.000001s :      6: predicate.less_batch_normalization
 1.52% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.12% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.06% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.14% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.74% :     0.000002s :      3: predicate.mutable_eliminate
 0.56% :     0.000001s :      3: predicate.opt_reshape
 2.69% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.72% :     0.000002s :     10: predicate.reduce_eliminate
 1.52% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.40% :     0.000002s :     10: predicate.reshape_eliminate
 1.13% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.46% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.85% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.30% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.35% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.74% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.21% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.30% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.11% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.56% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.73% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000670    12
52.20% :     0.000350s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.19% :     0.000028s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.62% :     0.000292s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014652    72
 0.33% :     0.000049s :      1: add_recomputation
 0.51% :     0.000075s :      1: auto_monad
 0.12% :     0.000017s :      1: auto_monad_reorder
 2.33% :     0.000342s :      1: bootstrap
 0.19% :     0.000027s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.14% :     0.000021s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000022s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.67% :     0.005667s :      1: jit_opt_a
 0.94% :     0.000137s :      1: jit_opt_after_cconv
 0.35% :     0.000051s :      1: jit_opt_b
 2.91% :     0.000426s :      1: loop_unroll
 3.43% :     0.000502s :      1: mutable_eliminate
 5.48% :     0.000803s :     26: opt.transform.jit_opt_a
 0.31% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.12% :     0.000457s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000038s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.36% :     0.000345s :      1: renormalize.infer
 2.25% :     0.000330s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000035s :      1: rewriter_after_opt_a
 0.54% :     0.000079s :      1: rewriter_before_opt_a
 0.50% :     0.000073s :      1: symbol_engine_optimizer
33.62% :     0.004926s :      1: type_inference


TotalTime = 0.0120091, [30]
[bootstrap]: 0.0002665
[type_inference]: 0.00392484
[event_method]: 1.546e-05
[auto_monad]: 4.147e-05
[graph_reusing]: 3.5e-06
[pre_auto_parallel]: 1.77001e-06
[py_interpret_to_execute]: 2.831e-05
[rewriter_before_opt_a]: 6.26e-05
[expand_dump_flag]: 1.97999e-06
[jit_opt_a]: 0.00506633, [2]
    [Cycle 1]: 0.00152056, [27]
        [switch_simplify]: 4.453e-05
        [loop_unroll]: 2.875e-05
        [a_1]: 0.00047161
        [with_stream_mark]: 1.321e-05
        [recompute_prepare]: 7.11999e-06
        [updatestate_depend_eliminate]: 3.08e-06
        [updatestate_assign_eliminate]: 2.13002e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.05999e-06
        [specialize_transform]: 5.92999e-06
        [updatestate_useless_node_eliminater]: 5.00999e-06
        [accelerated_algorithm]: 5.20001e-06
        [meta_shard_fg_expand]: 1.37999e-06
        [get_grad_eliminate_]: 5.03002e-06
        [merge_forward]: 2.34999e-06
        [cell_reuse_recompute_pass]: 7.39994e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.133e-05
        [j_node_and_user_rematch]: 9.54999e-06
        [meta_fg_expand]: 1.86e-06
        [replace_old_param]: 8.10999e-06
        [inline_without_move]: 5.53002e-06
        [renormalize]: 0.00064776
        [add_forward_monad_depend]: 6.51e-06
        [auto_monad_grad]: 2.49999e-06
        [auto_monad_eliminator]: 1.631e-05
        [cse]: 3.009e-05
        [replace_applicator]: 1.471e-05
    [Cycle 2]: 0.00037018, [27]
        [switch_simplify]: 6.57002e-06
        [loop_unroll]: 5.23002e-06
        [a_1]: 9.587e-05
        [with_stream_mark]: 1.279e-05
        [recompute_prepare]: 5.44998e-06
        [updatestate_depend_eliminate]: 2.94999e-06
        [updatestate_assign_eliminate]: 2.02001e-06
        [updatestate_loads_eliminate]: 1.96e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 5.26002e-06
        [updatestate_useless_node_eliminater]: 4.68999e-06
        [accelerated_algorithm]: 5.95002e-06
        [meta_shard_fg_expand]: 1.81e-06
        [get_grad_eliminate_]: 4.62e-06
        [merge_forward]: 3.21999e-06
        [cell_reuse_recompute_pass]: 1.36998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.551e-05
        [j_node_and_user_rematch]: 8.67e-06
        [meta_fg_expand]: 1.99e-06
        [replace_old_param]: 8.35999e-06
        [inline_without_move]: 5.09998e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.94999e-06
        [auto_monad_grad]: 1.32e-06
        [auto_monad_eliminator]: 7.16999e-06
        [cse]: 1.626e-05
        [replace_applicator]: 6.84999e-06
[py_interpret_to_execute_after_opt_a]: 1.443e-05
[rewriter_after_opt_a]: 4.301e-05
[convert_after_rewriter]: 1.603e-05
[order_py_execute_after_rewriter]: 5.46e-06
[mutable_eliminate]: 0.00081549
[jit_opt_b]: 5.96e-05, [1]
    [Cycle 1]: 5.007e-05, [2]
        [frontend_op_eliminate]: 1.992e-05
        [inline_after_opt_a]: 1.634e-05
[cconv]: 3.623e-05
[loop_unroll]: 0.00052079
[jit_opt_after_cconv]: 0.00015453, [1]
    [Cycle 1]: 0.0001471, [11]
        [c_1]: 2.196e-05
        [parameter_eliminate]: 4.16001e-06
        [updatestate_depend_eliminate]: 6.14999e-06
        [updatestate_assign_eliminate]: 2.89001e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 2.442e-05
        [call_graph_tuple_transform]: 2.152e-05
        [tuple_list_get_item_eliminator]: 5.38002e-06
        [none_parameter_eliminate]: 1.61002e-06
        [renormalize]: 6.20028e-07
        [switch_simplify]: 5.87001e-06
[remove_dup_value]: 1.775e-05
[partial_unused_args_eliminate]: 2.64999e-06
[environ_conv]: 5.79e-06
[add_recomputation]: 5.715e-05
[cse_after_recomputation]: 2.252e-05, [1]
    [Cycle 1]: 1.629e-05, [1]
        [cse]: 9.59e-06
[auto_monad_reorder]: 1.681e-05
[get_jit_bprop_graph]: 2.66999e-06
[rewriter_after_jit_bprop_graph]: 8.18999e-06
[opt_after_jit_grad]: 0.00048517
[symbol_engine_optimizer]: 8.397e-05, [1]
    [Cycle 1]: 7.575e-05, [6]
        [build]: 4.68001e-06
        [elim_shapecalc]: 8.65999e-06
        [elim_not_effective]: 1.549e-05
        [opt_reshape]: 6.98998e-06
        [fold_const_symbol]: 9.57999e-06
        [renormalize]: 6.30011e-07
[validate]: 4.62e-05
Sums
    bootstrap                            :     0.000266s :  3.24%
    type_inference                       :     0.003925s : 47.77%
    event_method                         :     0.000015s :  0.19%
    auto_monad                           :     0.000041s :  0.50%
    graph_reusing                        :     0.000003s :  0.04%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000028s :  0.34%
    rewriter_before_opt_a                :     0.000063s :  0.76%
    expand_dump_flag                     :     0.000002s :  0.02%
    jit_opt_a.switch_simplify            :     0.000051s :  0.62%
    jit_opt_a.loop_unroll                :     0.000034s :  0.41%
    jit_opt_a.a_1                        :     0.000567s :  6.91%
    jit_opt_a.with_stream_mark           :     0.000026s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.14%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.14%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.22%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.20%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000648s :  7.89%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.29%
    jit_opt_a.cse                        :     0.000046s :  0.56%
    jit_opt_a.replace_applicator         :     0.000022s :  0.26%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.18%
    rewriter_after_opt_a                 :     0.000043s :  0.52%
    convert_after_rewriter               :     0.000016s :  0.20%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000815s :  9.93%
    jit_opt_b.frontend_op_eliminate      :     0.000020s :  0.24%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.20%
    cconv                                :     0.000036s :  0.44%
    loop_unroll                          :     0.000521s :  6.34%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.05%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.30%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000022s :  0.26%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000018s :  0.22%
    partial_unused_args_eliminate        :     0.000003s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000057s :  0.70%
    cse_after_recomputation.cse          :     0.000010s :  0.12%
    auto_monad_reorder                   :     0.000017s :  0.20%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.10%
    opt_after_jit_grad                   :     0.000485s :  5.91%
    symbol_engine_optimizer.build        :     0.000005s :  0.06%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.19%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.09%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000046s :  0.56%


Time group info:
------[substitution.]   0.000157    23
 1.29% :     0.000002s :      2: substitution.elim_not_effective
 0.87% :     0.000001s :      2: substitution.fold_const_symbol
 3.81% :     0.000006s :      3: substitution.graph_param_transform
80.92% :     0.000127s :      4: substitution.inline
 2.45% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.74% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.71% :     0.000004s :      2: substitution.replace_old_param
 5.21% :     0.000008s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.003873     2
80.25% :     0.003108s :      1: type_inference.infer
19.75% :     0.000765s :      1: type_inference.specialize
------[replace.]   0.000060     6
66.44% :     0.000040s :      4: replace.inline
33.56% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000132     6
94.66% :     0.000125s :      4: match.inline
 5.34% :     0.000007s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000002s :     10: predicate.addn_zero_filter
 2.12% :     0.000003s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.41% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.02% :     0.000001s :     10: predicate.compare_switch_simplify
 1.48% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.19% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.00% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.79% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.48% :     0.000002s :     10: predicate.environ_get_eliminate
 1.22% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.45% :     0.000007s :     22: predicate.inline
 1.12% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.33% :     0.000002s :      6: predicate.less_batch_normalization
 1.58% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.95% :     0.000002s :     15: predicate.load_eliminater
 1.44% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 3.41% :     0.000004s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.05% :     0.000002s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.48% :     0.000002s :     10: predicate.reduce_eliminate
 1.44% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.70% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.98% :     0.000002s :     18: predicate.replace_applicator
 0.57% :     0.000001s :      6: predicate.replace_old_param
 0.47% :     0.000001s :      3: predicate.reset_defer_inline
 1.45% :     0.000002s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.72% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.34% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.07% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.21% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.41% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.40% :     0.000003s :     16: predicate.switch_defer_inline
 2.13% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.20% :     0.000009s :     53: predicate.switch_simplify
 1.26% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.48% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.68% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.34% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.67% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.95% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.93% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000546    12
48.75% :     0.000266s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.81% :     0.000026s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.44% :     0.000253s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013490    72
 0.45% :     0.000060s :      1: add_recomputation
 0.33% :     0.000045s :      1: auto_monad
 0.14% :     0.000019s :      1: auto_monad_reorder
 2.07% :     0.000280s :      1: bootstrap
 0.29% :     0.000039s :      1: cconv
 0.14% :     0.000019s :      1: convert_after_rewriter
 0.19% :     0.000025s :      1: cse_after_recomputation
 0.07% :     0.000010s :      1: environ_conv
 0.14% :     0.000019s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.05% :     0.000006s :      1: get_jit_bprop_graph
 0.04% :     0.000006s :      1: graph_reusing
37.58% :     0.005070s :      1: jit_opt_a
 1.17% :     0.000157s :      1: jit_opt_after_cconv
 0.46% :     0.000062s :      1: jit_opt_b
 3.93% :     0.000530s :      1: loop_unroll
 6.14% :     0.000828s :      1: mutable_eliminate
 5.73% :     0.000774s :     26: opt.transform.jit_opt_a
 0.38% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.22% :     0.000029s :      4: opt.transform.jit_opt_b
 0.10% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.14% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.27% :     0.000037s :      4: opt.transform.symbol_engine_opt
 3.65% :     0.000493s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.23% :     0.000031s :      1: py_interpret_to_execute
 0.13% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.15% :     0.000020s :      1: remove_dup_value
 2.38% :     0.000321s :      1: renormalize.infer
 2.37% :     0.000320s :      1: renormalize.specialize
 0.08% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.35% :     0.000048s :      1: rewriter_after_opt_a
 0.48% :     0.000065s :      1: rewriter_before_opt_a
 0.64% :     0.000086s :      1: symbol_engine_optimizer
29.18% :     0.003936s :      1: type_inference


TotalTime = 0.0126532, [30]
[bootstrap]: 0.00030638
[type_inference]: 0.00452143
[event_method]: 1.718e-05
[auto_monad]: 6.243e-05
[graph_reusing]: 5.59e-06
[pre_auto_parallel]: 2.31998e-06
[py_interpret_to_execute]: 3.283e-05
[rewriter_before_opt_a]: 7.069e-05
[expand_dump_flag]: 3.07002e-06
[jit_opt_a]: 0.00560375, [2]
    [Cycle 1]: 0.0016646, [27]
        [switch_simplify]: 5.546e-05
        [loop_unroll]: 2.857e-05
        [a_1]: 0.00054269
        [with_stream_mark]: 1.763e-05
        [recompute_prepare]: 7.68999e-06
        [updatestate_depend_eliminate]: 3.56001e-06
        [updatestate_assign_eliminate]: 3.01001e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [parameter_eliminate]: 1.80001e-06
        [specialize_transform]: 5.69e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.64e-06
        [meta_shard_fg_expand]: 1.93002e-06
        [get_grad_eliminate_]: 4.90001e-06
        [merge_forward]: 3.50003e-06
        [cell_reuse_recompute_pass]: 1.04e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.395e-05
        [j_node_and_user_rematch]: 8.99998e-06
        [meta_fg_expand]: 2.02999e-06
        [replace_old_param]: 9.57999e-06
        [inline_without_move]: 5.30999e-06
        [renormalize]: 0.0007
        [add_forward_monad_depend]: 5.70001e-06
        [auto_monad_grad]: 3.06001e-06
        [auto_monad_eliminator]: 1.325e-05
        [cse]: 2.989e-05
        [replace_applicator]: 1.368e-05
    [Cycle 2]: 0.00032893, [27]
        [switch_simplify]: 6.71e-06
        [loop_unroll]: 5.61e-06
        [a_1]: 9.344e-05
        [with_stream_mark]: 9.00001e-06
        [recompute_prepare]: 5.29e-06
        [updatestate_depend_eliminate]: 2.84999e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 2.17999e-06
        [parameter_eliminate]: 1.07e-06
        [specialize_transform]: 5.25999e-06
        [updatestate_useless_node_eliminater]: 5.01997e-06
        [accelerated_algorithm]: 5.52001e-06
        [meta_shard_fg_expand]: 1.13001e-06
        [get_grad_eliminate_]: 4.28001e-06
        [merge_forward]: 2.41998e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.271e-05
        [j_node_and_user_rematch]: 7.61001e-06
        [meta_fg_expand]: 1.62999e-06
        [replace_old_param]: 7.24001e-06
        [inline_without_move]: 4.50999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.30999e-06
        [auto_monad_grad]: 8.70001e-07
        [auto_monad_eliminator]: 4.62e-06
        [cse]: 1.058e-05
        [replace_applicator]: 4.89998e-06
[py_interpret_to_execute_after_opt_a]: 9.40001e-06
[rewriter_after_opt_a]: 3.535e-05
[convert_after_rewriter]: 7.23e-06
[order_py_execute_after_rewriter]: 4.82e-06
[mutable_eliminate]: 0.00052947
[jit_opt_b]: 4.909e-05, [1]
    [Cycle 1]: 4.243e-05, [2]
        [frontend_op_eliminate]: 1.625e-05
        [inline_after_opt_a]: 1.49e-05
[cconv]: 2.427e-05
[loop_unroll]: 0.00040761
[jit_opt_after_cconv]: 0.00013295, [1]
    [Cycle 1]: 0.00012682, [11]
        [c_1]: 1.988e-05
        [parameter_eliminate]: 2.24001e-06
        [updatestate_depend_eliminate]: 5.42999e-06
        [updatestate_assign_eliminate]: 2.38998e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [cse]: 1.641e-05
        [call_graph_tuple_transform]: 1.834e-05
        [tuple_list_get_item_eliminator]: 5.30001e-06
        [none_parameter_eliminate]: 1.38002e-06
        [renormalize]: 5.50004e-07
        [switch_simplify]: 4.92999e-06
[remove_dup_value]: 1.394e-05
[partial_unused_args_eliminate]: 2.36e-06
[environ_conv]: 4.80999e-06
[add_recomputation]: 4.475e-05
[cse_after_recomputation]: 2.017e-05, [1]
    [Cycle 1]: 1.449e-05, [1]
        [cse]: 8.67e-06
[auto_monad_reorder]: 1.586e-05
[get_jit_bprop_graph]: 1.97999e-06
[rewriter_after_jit_bprop_graph]: 5.00001e-06
[opt_after_jit_grad]: 0.00044464
[symbol_engine_optimizer]: 7.075e-05, [1]
    [Cycle 1]: 6.402e-05, [6]
        [build]: 3.42002e-06
        [elim_shapecalc]: 7.24001e-06
        [elim_not_effective]: 1.213e-05
        [opt_reshape]: 5.51e-06
        [fold_const_symbol]: 8.48999e-06
        [renormalize]: 4.99975e-07
[validate]: 3.327e-05
Sums
    bootstrap                            :     0.000306s :  3.62%
    type_inference                       :     0.004521s : 53.39%
    event_method                         :     0.000017s :  0.20%
    auto_monad                           :     0.000062s :  0.74%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.39%
    rewriter_before_opt_a                :     0.000071s :  0.83%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000062s :  0.73%
    jit_opt_a.loop_unroll                :     0.000034s :  0.40%
    jit_opt_a.a_1                        :     0.000636s :  7.51%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000700s :  8.27%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.21%
    jit_opt_a.cse                        :     0.000040s :  0.48%
    jit_opt_a.replace_applicator         :     0.000019s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000035s :  0.42%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000529s :  6.25%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000024s :  0.29%
    loop_unroll                          :     0.000408s :  4.81%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000045s :  0.53%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000445s :  5.25%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000033s :  0.39%


Time group info:
------[substitution.]   0.000201    23
 0.99% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.58% :     0.000005s :      3: substitution.graph_param_transform
83.93% :     0.000168s :      4: substitution.inline
 1.52% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.07% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.29% :     0.000005s :      2: substitution.replace_old_param
 5.96% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004462     2
80.71% :     0.003601s :      1: type_inference.infer
19.29% :     0.000861s :      1: type_inference.specialize
------[replace.]   0.000070     6
65.12% :     0.000046s :      4: replace.inline
34.88% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000177     6
93.85% :     0.000166s :      4: match.inline
 6.15% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000121   691
 1.29% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.04% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.18% :     0.000001s :     10: predicate.addn_check_dump
 1.70% :     0.000002s :     10: predicate.addn_zero_filter
 2.11% :     0.000003s :     10: predicate.arithmetic_simplify
 1.52% :     0.000002s :     10: predicate.cast_eliminate
 0.55% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.68% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.86% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.31% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.23% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.35% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.54% :     0.000002s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.79% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.54% :     0.000007s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.46% :     0.000002s :      6: predicate.less_batch_normalization
 1.42% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.32% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.01% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.12% :     0.000001s :     10: predicate.merge_addn
 1.40% :     0.000002s :     10: predicate.minmaximum_grad
 1.45% :     0.000002s :      3: predicate.mutable_eliminate
 0.51% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000003s :     15: predicate.partial_eliminate
 1.31% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.68% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.59% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.57% :     0.000002s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.92% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.38% :     0.000002s :     10: predicate.same_eliminate
 0.48% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.83% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.15% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.77% :     0.000009s :     53: predicate.switch_simplify
 1.27% :     0.000002s :     10: predicate.tile_eliminate
 1.47% :     0.000002s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.87% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.82% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.38% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.82% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.33% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000626    12
51.67% :     0.000323s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.91% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.41% :     0.000272s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014249    72
 0.33% :     0.000047s :      1: add_recomputation
 0.46% :     0.000066s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 2.28% :     0.000325s :      1: bootstrap
 0.19% :     0.000027s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.15% :     0.000021s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.35% :     0.005607s :      1: jit_opt_a
 0.95% :     0.000136s :      1: jit_opt_after_cconv
 0.36% :     0.000052s :      1: jit_opt_b
 2.91% :     0.000415s :      1: loop_unroll
 3.77% :     0.000537s :      1: mutable_eliminate
 5.95% :     0.000847s :     26: opt.transform.jit_opt_a
 0.32% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.17% :     0.000452s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000036s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000016s :      1: remove_dup_value
 2.55% :     0.000363s :      1: renormalize.infer
 2.32% :     0.000330s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000039s :      1: rewriter_after_opt_a
 0.52% :     0.000073s :      1: rewriter_before_opt_a
 0.51% :     0.000073s :      1: symbol_engine_optimizer
31.83% :     0.004535s :      1: type_inference


TotalTime = 0.0128304, [30]
[bootstrap]: 0.00029847
[type_inference]: 0.00436069
[event_method]: 1.852e-05
[auto_monad]: 6.466e-05
[graph_reusing]: 5.54e-06
[pre_auto_parallel]: 2.36998e-06
[py_interpret_to_execute]: 3.075e-05
[rewriter_before_opt_a]: 7.708e-05
[expand_dump_flag]: 3.41001e-06
[jit_opt_a]: 0.00563187, [2]
    [Cycle 1]: 0.00182389, [27]
        [switch_simplify]: 5.929e-05
        [loop_unroll]: 3.161e-05
        [a_1]: 0.00054662
        [with_stream_mark]: 1.714e-05
        [recompute_prepare]: 7.84002e-06
        [updatestate_depend_eliminate]: 3.43e-06
        [updatestate_assign_eliminate]: 2.93998e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 1.86003e-06
        [specialize_transform]: 5.81e-06
        [updatestate_useless_node_eliminater]: 5.39e-06
        [accelerated_algorithm]: 5.40999e-06
        [meta_shard_fg_expand]: 1.96998e-06
        [get_grad_eliminate_]: 5.12999e-06
        [merge_forward]: 3.48e-06
        [cell_reuse_recompute_pass]: 1.28002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.513e-05
        [j_node_and_user_rematch]: 1.024e-05
        [meta_fg_expand]: 2.24999e-06
        [replace_old_param]: 9.92001e-06
        [inline_without_move]: 5.12999e-06
        [renormalize]: 0.00083486
        [add_forward_monad_depend]: 6.48e-06
        [auto_monad_grad]: 2.94001e-06
        [auto_monad_eliminator]: 1.384e-05
        [cse]: 2.773e-05
        [replace_applicator]: 1.298e-05
    [Cycle 2]: 0.00033216, [27]
        [switch_simplify]: 5.96e-06
        [loop_unroll]: 5.76e-06
        [a_1]: 9.041e-05
        [with_stream_mark]: 1.085e-05
        [recompute_prepare]: 5.07999e-06
        [updatestate_depend_eliminate]: 2.60002e-06
        [updatestate_assign_eliminate]: 2.06998e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 1.30001e-06
        [specialize_transform]: 5.07e-06
        [updatestate_useless_node_eliminater]: 5.29998e-06
        [accelerated_algorithm]: 5.37999e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 4.29002e-06
        [merge_forward]: 2.93998e-06
        [cell_reuse_recompute_pass]: 1.52001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.296e-05
        [j_node_and_user_rematch]: 8.04002e-06
        [meta_fg_expand]: 1.82001e-06
        [replace_old_param]: 8.48001e-06
        [inline_without_move]: 4.48999e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 9.60019e-07
        [auto_monad_eliminator]: 5.27001e-06
        [cse]: 1.186e-05
        [replace_applicator]: 5.41002e-06
[py_interpret_to_execute_after_opt_a]: 1.26e-05
[rewriter_after_opt_a]: 3.739e-05
[convert_after_rewriter]: 7.36999e-06
[order_py_execute_after_rewriter]: 5.14998e-06
[mutable_eliminate]: 0.00063861
[jit_opt_b]: 5.377e-05, [1]
    [Cycle 1]: 4.655e-05, [2]
        [frontend_op_eliminate]: 1.786e-05
        [inline_after_opt_a]: 1.527e-05
[cconv]: 2.685e-05
[loop_unroll]: 0.00045009
[jit_opt_after_cconv]: 0.0001486, [1]
    [Cycle 1]: 0.00014122, [11]
        [c_1]: 2.24e-05
        [parameter_eliminate]: 3.9e-06
        [updatestate_depend_eliminate]: 6.23e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.05002e-06
        [cse]: 2.143e-05
        [call_graph_tuple_transform]: 2.007e-05
        [tuple_list_get_item_eliminator]: 5.09e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 5.61998e-06
[remove_dup_value]: 1.514e-05
[partial_unused_args_eliminate]: 2.49001e-06
[environ_conv]: 5.71998e-06
[add_recomputation]: 4.899e-05
[cse_after_recomputation]: 2.161e-05, [1]
    [Cycle 1]: 1.558e-05, [1]
        [cse]: 9.46e-06
[auto_monad_reorder]: 1.503e-05
[get_jit_bprop_graph]: 2.22999e-06
[rewriter_after_jit_bprop_graph]: 5.27001e-06
[opt_after_jit_grad]: 0.00052082
[symbol_engine_optimizer]: 8.175e-05, [1]
    [Cycle 1]: 7.33e-05, [6]
        [build]: 4.48999e-06
        [elim_shapecalc]: 9.31998e-06
        [elim_not_effective]: 1.446e-05
        [opt_reshape]: 5.84999e-06
        [fold_const_symbol]: 9.46e-06
        [renormalize]: 1.04e-06
[validate]: 4.501e-05
Sums
    bootstrap                            :     0.000298s :  3.42%
    type_inference                       :     0.004361s : 49.92%
    event_method                         :     0.000019s :  0.21%
    auto_monad                           :     0.000065s :  0.74%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000031s :  0.35%
    rewriter_before_opt_a                :     0.000077s :  0.88%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000065s :  0.75%
    jit_opt_a.loop_unroll                :     0.000037s :  0.43%
    jit_opt_a.a_1                        :     0.000637s :  7.29%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.32%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000009s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000018s :  0.21%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000835s :  9.56%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.22%
    jit_opt_a.cse                        :     0.000040s :  0.45%
    jit_opt_a.replace_applicator         :     0.000018s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.14%
    rewriter_after_opt_a                 :     0.000037s :  0.43%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000639s :  7.31%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000027s :  0.31%
    loop_unroll                          :     0.000450s :  5.15%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000021s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000049s :  0.56%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000521s :  5.96%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.11%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000045s :  0.52%


Time group info:
------[substitution.]   0.000212    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.76% :     0.000002s :      2: substitution.fold_const_symbol
 2.41% :     0.000005s :      3: substitution.graph_param_transform
83.31% :     0.000177s :      4: substitution.inline
 1.58% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.51% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.73% :     0.000006s :      2: substitution.replace_old_param
 5.86% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004302     2
80.22% :     0.003451s :      1: type_inference.infer
19.78% :     0.000851s :      1: type_inference.specialize
------[replace.]   0.000073     6
67.60% :     0.000049s :      4: replace.inline
32.40% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000186     6
93.93% :     0.000174s :      4: match.inline
 6.07% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.28% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.60% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.24% :     0.000001s :     10: predicate.addn_zero_filter
 1.78% :     0.000002s :     10: predicate.arithmetic_simplify
 1.59% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.37% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.62% :     0.000001s :      3: predicate.elim_not_effective
 0.55% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.12% :     0.000001s :     10: predicate.environ_get_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.64% :     0.000007s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.44% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 1.86% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.22% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.84% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.06% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.84% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 1.98% :     0.000002s :     15: predicate.partial_eliminate
 1.15% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.58% :     0.000002s :     18: predicate.replace_applicator
 0.65% :     0.000001s :      6: predicate.replace_old_param
 0.50% :     0.000001s :      3: predicate.reset_defer_inline
 1.31% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.87% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.07% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.17% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.33% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.46% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.39% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.76% :     0.000003s :     16: predicate.switch_defer_inline
 2.63% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.47% :     0.000009s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.20% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.73% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.56% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.92% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.72% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.37% :     0.000002s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000700    12
44.93% :     0.000314s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.87% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
50.20% :     0.000351s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014569    72
 0.36% :     0.000053s :      1: add_recomputation
 0.47% :     0.000068s :      1: auto_monad
 0.12% :     0.000017s :      1: auto_monad_reorder
 2.16% :     0.000315s :      1: bootstrap
 0.20% :     0.000030s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.16% :     0.000024s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.15% :     0.000022s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.68% :     0.005635s :      1: jit_opt_a
 1.04% :     0.000152s :      1: jit_opt_after_cconv
 0.39% :     0.000057s :      1: jit_opt_b
 3.15% :     0.000459s :      1: loop_unroll
 4.47% :     0.000651s :      1: mutable_eliminate
 5.89% :     0.000858s :     26: opt.transform.jit_opt_a
 0.34% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000027s :      4: opt.transform.jit_opt_b
 0.10% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.17% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.24% :     0.000035s :      4: opt.transform.symbol_engine_opt
 3.65% :     0.000532s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.23% :     0.000033s :      1: py_interpret_to_execute
 0.10% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000018s :      1: remove_dup_value
 2.66% :     0.000387s :      1: renormalize.infer
 3.02% :     0.000440s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000040s :      1: rewriter_after_opt_a
 0.55% :     0.000080s :      1: rewriter_before_opt_a
 0.58% :     0.000084s :      1: symbol_engine_optimizer
30.02% :     0.004373s :      1: type_inference


TotalTime = 0.0138821, [30]
[bootstrap]: 0.00034217
[type_inference]: 0.00512869
[event_method]: 1.712e-05
[auto_monad]: 6.242e-05
[graph_reusing]: 5.37999e-06
[pre_auto_parallel]: 2.47001e-06
[py_interpret_to_execute]: 3.394e-05
[rewriter_before_opt_a]: 7.162e-05
[expand_dump_flag]: 2.91e-06
[jit_opt_a]: 0.00608578, [2]
    [Cycle 1]: 0.00179308, [27]
        [switch_simplify]: 5.496e-05
        [loop_unroll]: 2.847e-05
        [a_1]: 0.00056516
        [with_stream_mark]: 2.133e-05
        [recompute_prepare]: 7.66999e-06
        [updatestate_depend_eliminate]: 3.65998e-06
        [updatestate_assign_eliminate]: 3.16001e-06
        [updatestate_loads_eliminate]: 2.39001e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 6.23e-06
        [updatestate_useless_node_eliminater]: 5.18002e-06
        [accelerated_algorithm]: 5.65001e-06
        [meta_shard_fg_expand]: 1.81998e-06
        [get_grad_eliminate_]: 5.35001e-06
        [merge_forward]: 3.41999e-06
        [cell_reuse_recompute_pass]: 1.40999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.436e-05
        [j_node_and_user_rematch]: 9.96e-06
        [meta_fg_expand]: 2.18002e-06
        [replace_old_param]: 9.81e-06
        [inline_without_move]: 5.25999e-06
        [renormalize]: 0.00075721
        [add_forward_monad_depend]: 4.273e-05
        [auto_monad_grad]: 2.52001e-06
        [auto_monad_eliminator]: 1.446e-05
        [cse]: 2.974e-05
        [replace_applicator]: 1.36e-05
    [Cycle 2]: 0.00033374, [27]
        [switch_simplify]: 6.47001e-06
        [loop_unroll]: 4.80999e-06
        [a_1]: 9.003e-05
        [with_stream_mark]: 1.185e-05
        [recompute_prepare]: 5.34e-06
        [updatestate_depend_eliminate]: 2.83e-06
        [updatestate_assign_eliminate]: 2.04e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [parameter_eliminate]: 1.27e-06
        [specialize_transform]: 5.26998e-06
        [updatestate_useless_node_eliminater]: 4.79002e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.91e-06
        [get_grad_eliminate_]: 4.39002e-06
        [merge_forward]: 2.21e-06
        [cell_reuse_recompute_pass]: 1.91998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.358e-05
        [j_node_and_user_rematch]: 8.18001e-06
        [meta_fg_expand]: 2.00002e-06
        [replace_old_param]: 7.67998e-06
        [inline_without_move]: 4.57e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.42e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 4.97e-06
        [cse]: 1.073e-05
        [replace_applicator]: 5.52001e-06
[py_interpret_to_execute_after_opt_a]: 1.262e-05
[rewriter_after_opt_a]: 3.676e-05
[convert_after_rewriter]: 6.82002e-06
[order_py_execute_after_rewriter]: 5.17e-06
[mutable_eliminate]: 0.00059609
[jit_opt_b]: 5e-05, [1]
    [Cycle 1]: 4.31e-05, [2]
        [frontend_op_eliminate]: 1.732e-05
        [inline_after_opt_a]: 1.476e-05
[cconv]: 2.604e-05
[loop_unroll]: 0.00041165
[jit_opt_after_cconv]: 0.00014119, [1]
    [Cycle 1]: 0.00013521, [11]
        [c_1]: 2.09e-05
        [parameter_eliminate]: 3.23e-06
        [updatestate_depend_eliminate]: 5.71e-06
        [updatestate_assign_eliminate]: 2.56e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [cse]: 1.958e-05
        [call_graph_tuple_transform]: 1.882e-05
        [tuple_list_get_item_eliminator]: 5.19998e-06
        [none_parameter_eliminate]: 1.44e-06
        [renormalize]: 8.09989e-07
        [switch_simplify]: 5.00001e-06
[remove_dup_value]: 1.415e-05
[partial_unused_args_eliminate]: 2.39001e-06
[environ_conv]: 5.74999e-06
[add_recomputation]: 4.752e-05
[cse_after_recomputation]: 2.008e-05, [1]
    [Cycle 1]: 1.478e-05, [1]
        [cse]: 8.84e-06
[auto_monad_reorder]: 1.456e-05
[get_jit_bprop_graph]: 1.72001e-06
[rewriter_after_jit_bprop_graph]: 4.77998e-06
[opt_after_jit_grad]: 0.00044524
[symbol_engine_optimizer]: 7.226e-05, [1]
    [Cycle 1]: 6.581e-05, [6]
        [build]: 3.74002e-06
        [elim_shapecalc]: 7.1e-06
        [elim_not_effective]: 1.365e-05
        [opt_reshape]: 5.50001e-06
        [fold_const_symbol]: 8.43001e-06
        [renormalize]: 5.39992e-07
[validate]: 3.496e-05
Sums
    bootstrap                            :     0.000342s :  3.67%
    type_inference                       :     0.005129s : 54.97%
    event_method                         :     0.000017s :  0.18%
    auto_monad                           :     0.000062s :  0.67%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.36%
    rewriter_before_opt_a                :     0.000072s :  0.77%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000061s :  0.66%
    jit_opt_a.loop_unroll                :     0.000033s :  0.36%
    jit_opt_a.a_1                        :     0.000655s :  7.02%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.36%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000006s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000757s :  8.12%
    jit_opt_a.add_forward_monad_depend   :     0.000044s :  0.47%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.21%
    jit_opt_a.cse                        :     0.000040s :  0.43%
    jit_opt_a.replace_applicator         :     0.000019s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.14%
    rewriter_after_opt_a                 :     0.000037s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.07%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000596s :  6.39%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000026s :  0.28%
    loop_unroll                          :     0.000412s :  4.41%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.05%
    remove_dup_value                     :     0.000014s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000048s :  0.51%
    cse_after_recomputation.cse          :     0.000009s :  0.09%
    auto_monad_reorder                   :     0.000015s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000445s :  4.77%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000035s :  0.37%


Time group info:
------[substitution.]   0.000214    23
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.59% :     0.000001s :      2: substitution.fold_const_symbol
 2.60% :     0.000006s :      3: substitution.graph_param_transform
83.97% :     0.000180s :      4: substitution.inline
 1.63% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.98% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.49% :     0.000005s :      2: substitution.replace_old_param
 5.85% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005064     2
81.58% :     0.004131s :      1: type_inference.infer
18.42% :     0.000933s :      1: type_inference.specialize
------[replace.]   0.000089     6
75.00% :     0.000067s :      4: replace.inline
25.00% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000189     6
93.95% :     0.000177s :      4: match.inline
 6.05% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.24% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.41% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.44% :     0.000002s :     10: predicate.addn_zero_filter
 1.92% :     0.000002s :     10: predicate.arithmetic_simplify
 1.17% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.26% :     0.000001s :     10: predicate.depend_value_elim
 1.25% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.25% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.12% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.82% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.21% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.49% :     0.000002s :     10: predicate.environ_get_eliminate
 1.24% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.90% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.65% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.32% :     0.000002s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.99% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000001s :      3: predicate.loop_unroll_after_grad
 3.99% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.78% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.05% :     0.000001s :     10: predicate.minmaximum_grad
 2.00% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.29% :     0.000003s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.47% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.88% :     0.000002s :     18: predicate.replace_applicator
 0.72% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.49% :     0.000002s :     10: predicate.reshape_eliminate
 1.26% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.66% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.22% :     0.000001s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 0.99% :     0.000001s :      6: predicate.specialize_transform
 1.58% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.71% :     0.000003s :     16: predicate.switch_defer_inline
 2.22% :     0.000003s :     16: predicate.switch_layer_defer_inline
 8.01% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.75% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.69% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.50% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.80% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.98% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.70% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000697    12
50.44% :     0.000352s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.37% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.19% :     0.000301s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015557    72
 0.32% :     0.000050s :      1: add_recomputation
 0.42% :     0.000066s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 2.31% :     0.000359s :      1: bootstrap
 0.18% :     0.000029s :      1: cconv
 0.09% :     0.000014s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.13% :     0.000021s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.14% :     0.006090s :      1: jit_opt_a
 0.93% :     0.000144s :      1: jit_opt_after_cconv
 0.34% :     0.000053s :      1: jit_opt_b
 2.70% :     0.000420s :      1: loop_unroll
 3.89% :     0.000606s :      1: mutable_eliminate
 5.58% :     0.000868s :     26: opt.transform.jit_opt_a
 0.30% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000026s :      4: opt.transform.jit_opt_b
 0.07% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.91% :     0.000453s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.23% :     0.000036s :      1: py_interpret_to_execute
 0.10% :     0.000015s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000016s :      1: remove_dup_value
 2.52% :     0.000392s :      1: renormalize.infer
 2.30% :     0.000358s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000040s :      1: rewriter_after_opt_a
 0.48% :     0.000075s :      1: rewriter_before_opt_a
 0.48% :     0.000075s :      1: symbol_engine_optimizer
33.06% :     0.005143s :      1: type_inference


TotalTime = 0.0137186, [30]
[bootstrap]: 0.00030462
[type_inference]: 0.00448429
[event_method]: 1.866e-05
[auto_monad]: 0.00011988
[graph_reusing]: 6.38e-06
[pre_auto_parallel]: 2.63e-06
[py_interpret_to_execute]: 4.346e-05
[rewriter_before_opt_a]: 7.7e-05
[expand_dump_flag]: 3.11999e-06
[jit_opt_a]: 0.00622355, [2]
    [Cycle 1]: 0.00184751, [27]
        [switch_simplify]: 6.343e-05
        [loop_unroll]: 3.122e-05
        [a_1]: 0.00058764
        [with_stream_mark]: 2.213e-05
        [recompute_prepare]: 9.62001e-06
        [updatestate_depend_eliminate]: 4.21001e-06
        [updatestate_assign_eliminate]: 2.86999e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [parameter_eliminate]: 1.80001e-06
        [specialize_transform]: 6.88e-06
        [updatestate_useless_node_eliminater]: 5.04e-06
        [accelerated_algorithm]: 5.74e-06
        [meta_shard_fg_expand]: 2.88e-06
        [get_grad_eliminate_]: 5.90002e-06
        [merge_forward]: 3.54002e-06
        [cell_reuse_recompute_pass]: 1.20999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.549e-05
        [j_node_and_user_rematch]: 8.99e-06
        [meta_fg_expand]: 2.51998e-06
        [replace_old_param]: 9.74e-06
        [inline_without_move]: 5.80002e-06
        [renormalize]: 0.00078273
        [add_forward_monad_depend]: 6.05002e-06
        [auto_monad_grad]: 2.64999e-06
        [auto_monad_eliminator]: 1.583e-05
        [cse]: 2.956e-05
        [replace_applicator]: 1.376e-05
    [Cycle 2]: 0.00034736, [27]
        [switch_simplify]: 6.28e-06
        [loop_unroll]: 5.00999e-06
        [a_1]: 9.157e-05
        [with_stream_mark]: 1.169e-05
        [recompute_prepare]: 5.45001e-06
        [updatestate_depend_eliminate]: 3.06001e-06
        [updatestate_assign_eliminate]: 2.21e-06
        [updatestate_loads_eliminate]: 2.36e-06
        [parameter_eliminate]: 1.56002e-06
        [specialize_transform]: 5.03002e-06
        [updatestate_useless_node_eliminater]: 5.26002e-06
        [accelerated_algorithm]: 5.19e-06
        [meta_shard_fg_expand]: 1.60999e-06
        [get_grad_eliminate_]: 4.50999e-06
        [merge_forward]: 2.69999e-06
        [cell_reuse_recompute_pass]: 2.37999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.492e-05
        [j_node_and_user_rematch]: 8.94003e-06
        [meta_fg_expand]: 1.83002e-06
        [replace_old_param]: 8.1e-06
        [inline_without_move]: 5.37999e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.67001e-06
        [auto_monad_grad]: 1.25999e-06
        [auto_monad_eliminator]: 4.52998e-06
        [cse]: 1.145e-05
        [replace_applicator]: 1.036e-05
[py_interpret_to_execute_after_opt_a]: 1.116e-05
[rewriter_after_opt_a]: 3.711e-05
[convert_after_rewriter]: 7.1e-06
[order_py_execute_after_rewriter]: 5.40999e-06
[mutable_eliminate]: 0.00060652
[jit_opt_b]: 6.122e-05, [1]
    [Cycle 1]: 5.292e-05, [2]
        [frontend_op_eliminate]: 2.359e-05
        [inline_after_opt_a]: 1.67e-05
[cconv]: 3.302e-05
[loop_unroll]: 0.00048678
[jit_opt_after_cconv]: 0.00015331, [1]
    [Cycle 1]: 0.00014624, [11]
        [c_1]: 2.254e-05
        [parameter_eliminate]: 3.25002e-06
        [updatestate_depend_eliminate]: 6.37001e-06
        [updatestate_assign_eliminate]: 2.84999e-06
        [updatestate_loads_eliminate]: 2.16998e-06
        [cse]: 2.294e-05
        [call_graph_tuple_transform]: 2.006e-05
        [tuple_list_get_item_eliminator]: 5.35001e-06
        [none_parameter_eliminate]: 1.65001e-06
        [renormalize]: 1.03001e-06
        [switch_simplify]: 5.52001e-06
[remove_dup_value]: 1.446e-05
[partial_unused_args_eliminate]: 2.15002e-06
[environ_conv]: 5.56e-06
[add_recomputation]: 0.00010684
[cse_after_recomputation]: 2.642e-05, [1]
    [Cycle 1]: 1.893e-05, [1]
        [cse]: 1.119e-05
[auto_monad_reorder]: 1.648e-05
[get_jit_bprop_graph]: 1.99999e-06
[rewriter_after_jit_bprop_graph]: 5.72999e-06
[opt_after_jit_grad]: 0.00053698
[symbol_engine_optimizer]: 7.809e-05, [1]
    [Cycle 1]: 7.081e-05, [6]
        [build]: 4.55001e-06
        [elim_shapecalc]: 7.42002e-06
        [elim_not_effective]: 1.443e-05
        [opt_reshape]: 5.94e-06
        [fold_const_symbol]: 9.69999e-06
        [renormalize]: 5.79981e-07
[validate]: 3.807e-05
Sums
    bootstrap                            :     0.000305s :  3.37%
    type_inference                       :     0.004484s : 49.62%
    event_method                         :     0.000019s :  0.21%
    auto_monad                           :     0.000120s :  1.33%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000043s :  0.48%
    rewriter_before_opt_a                :     0.000077s :  0.85%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000070s :  0.77%
    jit_opt_a.loop_unroll                :     0.000036s :  0.40%
    jit_opt_a.a_1                        :     0.000679s :  7.52%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.37%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000012s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000018s :  0.20%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000783s :  8.66%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.23%
    jit_opt_a.cse                        :     0.000041s :  0.45%
    jit_opt_a.replace_applicator         :     0.000024s :  0.27%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000037s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000607s :  6.71%
    jit_opt_b.frontend_op_eliminate      :     0.000024s :  0.26%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.18%
    cconv                                :     0.000033s :  0.37%
    loop_unroll                          :     0.000487s :  5.39%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000014s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000107s :  1.18%
    cse_after_recomputation.cse          :     0.000011s :  0.12%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.06%
    opt_after_jit_grad                   :     0.000537s :  5.94%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000038s :  0.42%


Time group info:
------[substitution.]   0.000233    23
 0.85% :     0.000002s :      2: substitution.elim_not_effective
 0.55% :     0.000001s :      2: substitution.fold_const_symbol
 2.05% :     0.000005s :      3: substitution.graph_param_transform
85.16% :     0.000199s :      4: substitution.inline
 1.33% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.13% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.16% :     0.000005s :      2: substitution.replace_old_param
 5.76% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004409     2
78.76% :     0.003472s :      1: type_inference.infer
21.24% :     0.000936s :      1: type_inference.specialize
------[replace.]   0.000080     6
63.86% :     0.000051s :      4: replace.inline
36.14% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000208     6
94.22% :     0.000196s :      4: match.inline
 5.78% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.21% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.65% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.98% :     0.000001s :     10: predicate.addn_check_dump
 1.23% :     0.000002s :     10: predicate.addn_zero_filter
 1.60% :     0.000002s :     10: predicate.arithmetic_simplify
 1.09% :     0.000001s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.15% :     0.000001s :     10: predicate.compare_switch_simplify
 1.37% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.02% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.01% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.98% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.07% :     0.000001s :     10: predicate.environ_get_eliminate
 1.44% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.24% :     0.000000s :      3: predicate.graph_param_transform
 4.93% :     0.000006s :     22: predicate.inline
 0.93% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.34% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.64% :     0.000002s :     15: predicate.load_eliminater
 2.31% :     0.000003s :      3: predicate.loop_unroll_after_grad
 3.80% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 0.97% :     0.000001s :     10: predicate.minmaximum_grad
 2.10% :     0.000003s :      3: predicate.mutable_eliminate
 0.41% :     0.000001s :      3: predicate.opt_reshape
 1.97% :     0.000002s :     15: predicate.partial_eliminate
 1.18% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.31% :     0.000002s :     10: predicate.reduce_eliminate
 1.66% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.69% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.82% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.62% :     0.000002s :     10: predicate.reshape_eliminate
 1.07% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.22% :     0.000002s :      3: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     10: predicate.same_eliminate
 0.46% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.05% :     0.000001s :      6: predicate.special_op_eliminate
 0.90% :     0.000001s :      6: predicate.specialize_transform
 1.22% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.43% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.63% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.25% :     0.000010s :     53: predicate.switch_simplify
 1.85% :     0.000002s :     10: predicate.tile_eliminate
 1.10% :     0.000001s :     10: predicate.transpose_eliminate
 1.31% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.68% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 2.26% :     0.000003s :     10: predicate.tuple_list_set_item_eliminator
 1.35% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.60% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.57% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.35% :     0.000003s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000643    12
48.10% :     0.000309s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.54% :     0.000042s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.37% :     0.000292s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015480    72
 0.71% :     0.000110s :      1: add_recomputation
 0.80% :     0.000124s :      1: auto_monad
 0.13% :     0.000020s :      1: auto_monad_reorder
 2.09% :     0.000324s :      1: bootstrap
 0.23% :     0.000036s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.18% :     0.000029s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
40.23% :     0.006227s :      1: jit_opt_a
 1.01% :     0.000156s :      1: jit_opt_after_cconv
 0.41% :     0.000064s :      1: jit_opt_b
 3.21% :     0.000497s :      1: loop_unroll
 3.99% :     0.000617s :      1: mutable_eliminate
 5.89% :     0.000912s :     26: opt.transform.jit_opt_a
 0.32% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000032s :      4: opt.transform.jit_opt_b
 0.10% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.54% :     0.000548s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.30% :     0.000046s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 2.60% :     0.000403s :      1: renormalize.infer
 2.40% :     0.000372s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000041s :      1: rewriter_after_opt_a
 0.52% :     0.000081s :      1: rewriter_before_opt_a
 0.52% :     0.000081s :      1: symbol_engine_optimizer
29.09% :     0.004502s :      1: type_inference


TotalTime = 0.0131638, [30]
[bootstrap]: 0.00033076
[type_inference]: 0.00461354
[event_method]: 1.618e-05
[auto_monad]: 6.265e-05
[graph_reusing]: 5.05999e-06
[pre_auto_parallel]: 2.21e-06
[py_interpret_to_execute]: 3.09e-05
[rewriter_before_opt_a]: 7.046e-05
[expand_dump_flag]: 3.57002e-06
[jit_opt_a]: 0.00582482, [2]
    [Cycle 1]: 0.00187178, [27]
        [switch_simplify]: 5.601e-05
        [loop_unroll]: 2.865e-05
        [a_1]: 0.00056499
        [with_stream_mark]: 2.233e-05
        [recompute_prepare]: 7.09001e-06
        [updatestate_depend_eliminate]: 3.64002e-06
        [updatestate_assign_eliminate]: 2.63e-06
        [updatestate_loads_eliminate]: 2.50002e-06
        [parameter_eliminate]: 2.26e-06
        [specialize_transform]: 6.07001e-06
        [updatestate_useless_node_eliminater]: 5.02999e-06
        [accelerated_algorithm]: 5.48002e-06
        [meta_shard_fg_expand]: 2.06e-06
        [get_grad_eliminate_]: 5.31002e-06
        [merge_forward]: 3.76001e-06
        [cell_reuse_recompute_pass]: 1.31998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.522e-05
        [j_node_and_user_rematch]: 9.71e-06
        [meta_fg_expand]: 2.78998e-06
        [replace_old_param]: 8.94e-06
        [inline_without_move]: 5.62001e-06
        [renormalize]: 0.00087528
        [add_forward_monad_depend]: 6.12999e-06
        [auto_monad_grad]: 2.69001e-06
        [auto_monad_eliminator]: 1.388e-05
        [cse]: 2.211e-05
        [replace_applicator]: 1.379e-05
    [Cycle 2]: 0.00032089, [27]
        [switch_simplify]: 5.84e-06
        [loop_unroll]: 5.31002e-06
        [a_1]: 8.863e-05
        [with_stream_mark]: 1.093e-05
        [recompute_prepare]: 4.77e-06
        [updatestate_depend_eliminate]: 2.79999e-06
        [updatestate_assign_eliminate]: 1.94e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.09998e-06
        [specialize_transform]: 4.82998e-06
        [updatestate_useless_node_eliminater]: 4.60001e-06
        [accelerated_algorithm]: 5.67001e-06
        [meta_shard_fg_expand]: 9.89996e-07
        [get_grad_eliminate_]: 4.4e-06
        [merge_forward]: 2.74001e-06
        [cell_reuse_recompute_pass]: 1.83002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.158e-05
        [j_node_and_user_rematch]: 7.56001e-06
        [meta_fg_expand]: 1.49e-06
        [replace_old_param]: 7.16001e-06
        [inline_without_move]: 4.82998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 1.02998e-06
        [auto_monad_eliminator]: 4.95999e-06
        [cse]: 1.03e-05
        [replace_applicator]: 5.05001e-06
[py_interpret_to_execute_after_opt_a]: 1.031e-05
[rewriter_after_opt_a]: 3.317e-05
[convert_after_rewriter]: 6.97002e-06
[order_py_execute_after_rewriter]: 5.09e-06
[mutable_eliminate]: 0.00060654
[jit_opt_b]: 4.915e-05, [1]
    [Cycle 1]: 4.222e-05, [2]
        [frontend_op_eliminate]: 1.599e-05
        [inline_after_opt_a]: 1.492e-05
[cconv]: 2.664e-05
[loop_unroll]: 0.00043984
[jit_opt_after_cconv]: 0.00014053, [1]
    [Cycle 1]: 0.00013372, [11]
        [c_1]: 2.026e-05
        [parameter_eliminate]: 3.28998e-06
        [updatestate_depend_eliminate]: 5.76998e-06
        [updatestate_assign_eliminate]: 2.17999e-06
        [updatestate_loads_eliminate]: 2.02999e-06
        [cse]: 1.952e-05
        [call_graph_tuple_transform]: 1.872e-05
        [tuple_list_get_item_eliminator]: 5.57999e-06
        [none_parameter_eliminate]: 1.49998e-06
        [renormalize]: 5.20027e-07
        [switch_simplify]: 5.40999e-06
[remove_dup_value]: 1.474e-05
[partial_unused_args_eliminate]: 2.23002e-06
[environ_conv]: 5.10001e-06
[add_recomputation]: 4.673e-05
[cse_after_recomputation]: 2.021e-05, [1]
    [Cycle 1]: 1.465e-05, [1]
        [cse]: 9.07999e-06
[auto_monad_reorder]: 1.448e-05
[get_jit_bprop_graph]: 2.21e-06
[rewriter_after_jit_bprop_graph]: 6.40997e-06
[opt_after_jit_grad]: 0.00047758
[symbol_engine_optimizer]: 7.752e-05, [1]
    [Cycle 1]: 7.063e-05, [6]
        [build]: 3.86001e-06
        [elim_shapecalc]: 7.85e-06
        [elim_not_effective]: 1.339e-05
        [opt_reshape]: 7.01001e-06
        [fold_const_symbol]: 9.17999e-06
        [renormalize]: 7.00005e-07
[validate]: 3.964e-05
Sums
    bootstrap                            :     0.000331s :  3.70%
    type_inference                       :     0.004614s : 51.61%
    event_method                         :     0.000016s :  0.18%
    auto_monad                           :     0.000063s :  0.70%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.02%
    py_interpret_to_execute              :     0.000031s :  0.35%
    rewriter_before_opt_a                :     0.000070s :  0.79%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000062s :  0.69%
    jit_opt_a.loop_unroll                :     0.000034s :  0.38%
    jit_opt_a.a_1                        :     0.000654s :  7.31%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.37%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.18%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000875s :  9.79%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.21%
    jit_opt_a.cse                        :     0.000032s :  0.36%
    jit_opt_a.replace_applicator         :     0.000019s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000033s :  0.37%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000607s :  6.79%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000027s :  0.30%
    loop_unroll                          :     0.000440s :  4.92%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000047s :  0.52%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000014s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000478s :  5.34%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.44%


Time group info:
------[substitution.]   0.000227    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.56% :     0.000001s :      2: substitution.fold_const_symbol
 2.33% :     0.000005s :      3: substitution.graph_param_transform
85.36% :     0.000194s :      4: substitution.inline
 1.40% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.16% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.89% :     0.000004s :      2: substitution.replace_old_param
 5.30% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004547     2
80.84% :     0.003676s :      1: type_inference.infer
19.16% :     0.000871s :      1: type_inference.specialize
------[replace.]   0.000071     6
67.87% :     0.000048s :      4: replace.inline
32.13% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000202     6
94.55% :     0.000191s :      4: match.inline
 5.45% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000117   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.15% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.04% :     0.000001s :     10: predicate.addn_check_dump
 1.43% :     0.000002s :     10: predicate.addn_zero_filter
 1.84% :     0.000002s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.52% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.35% :     0.000002s :     10: predicate.depend_value_elim
 1.19% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.35% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.15% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.72% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.26% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.24% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.07% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.71% :     0.000007s :     22: predicate.inline
 0.92% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.09% :     0.000001s :      6: predicate.less_batch_normalization
 1.75% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.59% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.08% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.15% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 1.90% :     0.000002s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.06% :     0.000002s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.50% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.62% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.64% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.63% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.20% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.08% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.28% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.56% :     0.000003s :     16: predicate.switch_defer_inline
 2.33% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.63% :     0.000009s :     53: predicate.switch_simplify
 1.25% :     0.000001s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.22% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.65% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.63% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.09% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.74% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.72% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000657    12
48.69% :     0.000320s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.17% :     0.000047s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.13% :     0.000290s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014950    72
 0.33% :     0.000050s :      1: add_recomputation
 0.44% :     0.000066s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 2.33% :     0.000349s :      1: bootstrap
 0.20% :     0.000029s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.13% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000007s :      1: graph_reusing
38.99% :     0.005828s :      1: jit_opt_a
 0.96% :     0.000143s :      1: jit_opt_after_cconv
 0.35% :     0.000052s :      1: jit_opt_b
 3.00% :     0.000448s :      1: loop_unroll
 4.12% :     0.000616s :      1: mutable_eliminate
 5.78% :     0.000864s :     26: opt.transform.jit_opt_a
 0.31% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000034s :      4: opt.transform.symbol_engine_opt
 3.26% :     0.000487s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.22% :     0.000033s :      1: py_interpret_to_execute
 0.08% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 3.24% :     0.000484s :      1: renormalize.infer
 2.56% :     0.000383s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000036s :      1: rewriter_after_opt_a
 0.49% :     0.000073s :      1: rewriter_before_opt_a
 0.53% :     0.000080s :      1: symbol_engine_optimizer
30.95% :     0.004626s :      1: type_inference


TotalTime = 0.013692, [30]
[bootstrap]: 0.00031447
[type_inference]: 0.00478499
[event_method]: 1.783e-05
[auto_monad]: 6.33e-05
[graph_reusing]: 6.03998e-06
[pre_auto_parallel]: 2.44001e-06
[py_interpret_to_execute]: 3.773e-05
[rewriter_before_opt_a]: 7.333e-05
[expand_dump_flag]: 3.78001e-06
[jit_opt_a]: 0.00597226, [2]
    [Cycle 1]: 0.00188334, [27]
        [switch_simplify]: 6.009e-05
        [loop_unroll]: 2.91e-05
        [a_1]: 0.00058935
        [with_stream_mark]: 2.114e-05
        [recompute_prepare]: 9.24e-06
        [updatestate_depend_eliminate]: 3.85e-06
        [updatestate_assign_eliminate]: 2.96999e-06
        [updatestate_loads_eliminate]: 2.44001e-06
        [parameter_eliminate]: 1.87999e-06
        [specialize_transform]: 6.58998e-06
        [updatestate_useless_node_eliminater]: 6.54001e-06
        [accelerated_algorithm]: 6.16e-06
        [meta_shard_fg_expand]: 2.49001e-06
        [get_grad_eliminate_]: 6.64001e-06
        [merge_forward]: 4.29002e-06
        [cell_reuse_recompute_pass]: 1.70001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.588e-05
        [j_node_and_user_rematch]: 9.79e-06
        [meta_fg_expand]: 2.32001e-06
        [replace_old_param]: 9.92001e-06
        [inline_without_move]: 6.01e-06
        [renormalize]: 0.00081376
        [add_forward_monad_depend]: 7.26999e-06
        [auto_monad_grad]: 3.26001e-06
        [auto_monad_eliminator]: 1.454e-05
        [cse]: 2.828e-05
        [replace_applicator]: 1.386e-05
    [Cycle 2]: 0.00033576, [27]
        [switch_simplify]: 5.84e-06
        [loop_unroll]: 4.95999e-06
        [a_1]: 9.145e-05
        [with_stream_mark]: 1.084e-05
        [recompute_prepare]: 5.17e-06
        [updatestate_depend_eliminate]: 2.58e-06
        [updatestate_assign_eliminate]: 2.12001e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [parameter_eliminate]: 1.29e-06
        [specialize_transform]: 5.34e-06
        [updatestate_useless_node_eliminater]: 5.10001e-06
        [accelerated_algorithm]: 4.95999e-06
        [meta_shard_fg_expand]: 2.15002e-06
        [get_grad_eliminate_]: 4.62e-06
        [merge_forward]: 2.88e-06
        [cell_reuse_recompute_pass]: 1.61998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.449e-05
        [j_node_and_user_rematch]: 8.10999e-06
        [meta_fg_expand]: 1.29003e-06
        [replace_old_param]: 7.33999e-06
        [inline_without_move]: 4.84e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.65001e-06
        [auto_monad_grad]: 1.05001e-06
        [auto_monad_eliminator]: 5.05001e-06
        [cse]: 1.199e-05
        [replace_applicator]: 4.94998e-06
[py_interpret_to_execute_after_opt_a]: 1.279e-05
[rewriter_after_opt_a]: 3.483e-05
[convert_after_rewriter]: 8.58001e-06
[order_py_execute_after_rewriter]: 4.68001e-06
[mutable_eliminate]: 0.00066034
[jit_opt_b]: 5.425e-05, [1]
    [Cycle 1]: 4.591e-05, [2]
        [frontend_op_eliminate]: 1.709e-05
        [inline_after_opt_a]: 1.624e-05
[cconv]: 2.942e-05
[loop_unroll]: 0.00048991
[jit_opt_after_cconv]: 0.0001522, [1]
    [Cycle 1]: 0.00014449, [11]
        [c_1]: 2.181e-05
        [parameter_eliminate]: 3.78001e-06
        [updatestate_depend_eliminate]: 6.46e-06
        [updatestate_assign_eliminate]: 2.66999e-06
        [updatestate_loads_eliminate]: 2.25002e-06
        [cse]: 2.171e-05
        [call_graph_tuple_transform]: 2.075e-05
        [tuple_list_get_item_eliminator]: 5.59998e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 9.20001e-07
        [switch_simplify]: 5.60001e-06
[remove_dup_value]: 1.61e-05
[partial_unused_args_eliminate]: 2.22999e-06
[environ_conv]: 5.95002e-06
[add_recomputation]: 4.766e-05
[cse_after_recomputation]: 2.285e-05, [1]
    [Cycle 1]: 1.645e-05, [1]
        [cse]: 9.34e-06
[auto_monad_reorder]: 1.535e-05
[get_jit_bprop_graph]: 1.99999e-06
[rewriter_after_jit_bprop_graph]: 1.403e-05
[opt_after_jit_grad]: 0.00053337
[symbol_engine_optimizer]: 7.802e-05, [1]
    [Cycle 1]: 6.971e-05, [6]
        [build]: 3.97e-06
        [elim_shapecalc]: 7.31001e-06
        [elim_not_effective]: 1.411e-05
        [opt_reshape]: 5.66e-06
        [fold_const_symbol]: 9.44e-06
        [renormalize]: 8.39995e-07
[validate]: 3.771e-05
Sums
    bootstrap                            :     0.000314s :  3.39%
    type_inference                       :     0.004785s : 51.51%
    event_method                         :     0.000018s :  0.19%
    auto_monad                           :     0.000063s :  0.68%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000038s :  0.41%
    rewriter_before_opt_a                :     0.000073s :  0.79%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000066s :  0.71%
    jit_opt_a.loop_unroll                :     0.000034s :  0.37%
    jit_opt_a.a_1                        :     0.000681s :  7.33%
    jit_opt_a.with_stream_mark           :     0.000032s :  0.34%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000012s :  0.13%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000005s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.12%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.33%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000814s :  8.76%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.21%
    jit_opt_a.cse                        :     0.000040s :  0.43%
    jit_opt_a.replace_applicator         :     0.000019s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000013s :  0.14%
    rewriter_after_opt_a                 :     0.000035s :  0.37%
    convert_after_rewriter               :     0.000009s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000660s :  7.11%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.17%
    cconv                                :     0.000029s :  0.32%
    loop_unroll                          :     0.000490s :  5.27%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.06%
    remove_dup_value                     :     0.000016s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000048s :  0.51%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000014s :  0.15%
    opt_after_jit_grad                   :     0.000533s :  5.74%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000038s :  0.41%


Time group info:
------[substitution.]   0.000238    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000002s :      2: substitution.fold_const_symbol
 2.30% :     0.000005s :      3: substitution.graph_param_transform
85.13% :     0.000203s :      4: substitution.inline
 1.44% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.24% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000005s :      2: substitution.replace_old_param
 5.27% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004715     2
80.26% :     0.003784s :      1: type_inference.infer
19.74% :     0.000930s :      1: type_inference.specialize
------[replace.]   0.000081     6
66.25% :     0.000054s :      4: replace.inline
33.75% :     0.000027s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000211     6
94.58% :     0.000200s :      4: match.inline
 5.42% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000118   691
 1.39% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.34% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.34% :     0.000002s :     10: predicate.addn_zero_filter
 1.67% :     0.000002s :     10: predicate.arithmetic_simplify
 1.65% :     0.000002s :     10: predicate.cast_eliminate
 0.42% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.85% :     0.000002s :     10: predicate.depend_value_elim
 1.09% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.15% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.28% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.50% :     0.000001s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.16% :     0.000001s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.99% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.65% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.08% :     0.000001s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.66% :     0.000002s :     15: predicate.load_eliminater
 1.73% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.80% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.70% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.04% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 2.59% :     0.000003s :      3: predicate.mutable_eliminate
 0.52% :     0.000001s :      3: predicate.opt_reshape
 2.16% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.78% :     0.000002s :     10: predicate.reduce_eliminate
 1.41% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.78% :     0.000002s :     18: predicate.replace_applicator
 0.43% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.56% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.31% :     0.000002s :     10: predicate.same_eliminate
 0.51% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.21% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.40% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.88% :     0.000003s :     16: predicate.switch_defer_inline
 1.95% :     0.000002s :     16: predicate.switch_layer_defer_inline
 8.11% :     0.000010s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.29% :     0.000002s :     10: predicate.transpose_eliminate
 1.74% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.33% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.88% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.73% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.58% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.08% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.85% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000709    12
51.15% :     0.000363s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.71% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.14% :     0.000306s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015472    72
 0.33% :     0.000051s :      1: add_recomputation
 0.43% :     0.000067s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.14% :     0.000332s :      1: bootstrap
 0.21% :     0.000032s :      1: cconv
 0.07% :     0.000011s :      1: convert_after_rewriter
 0.16% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.14% :     0.000022s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.63% :     0.005976s :      1: jit_opt_a
 1.00% :     0.000155s :      1: jit_opt_after_cconv
 0.37% :     0.000057s :      1: jit_opt_b
 3.23% :     0.000500s :      1: loop_unroll
 4.34% :     0.000672s :      1: mutable_eliminate
 5.86% :     0.000906s :     26: opt.transform.jit_opt_a
 0.32% :     0.000050s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000026s :      4: opt.transform.jit_opt_b
 0.09% :     0.000014s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000025s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.51% :     0.000543s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.26% :     0.000041s :      1: py_interpret_to_execute
 0.10% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000019s :      1: remove_dup_value
 2.68% :     0.000415s :      1: renormalize.infer
 2.53% :     0.000391s :      1: renormalize.specialize
 0.11% :     0.000017s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000038s :      1: rewriter_after_opt_a
 0.49% :     0.000076s :      1: rewriter_before_opt_a
 0.52% :     0.000081s :      1: symbol_engine_optimizer
31.03% :     0.004800s :      1: type_inference


TotalTime = 0.0136318, [30]
[bootstrap]: 0.00032354
[type_inference]: 0.00461999
[event_method]: 1.958e-05
[auto_monad]: 6.878e-05
[graph_reusing]: 5.86e-06
[pre_auto_parallel]: 2.65997e-06
[py_interpret_to_execute]: 3.929e-05
[rewriter_before_opt_a]: 7.697e-05
[expand_dump_flag]: 3.49001e-06
[jit_opt_a]: 0.00629167, [2]
    [Cycle 1]: 0.00203823, [27]
        [switch_simplify]: 6.426e-05
        [loop_unroll]: 3.095e-05
        [a_1]: 0.00063788
        [with_stream_mark]: 2.346e-05
        [recompute_prepare]: 8.79e-06
        [updatestate_depend_eliminate]: 4.28999e-06
        [updatestate_assign_eliminate]: 2.76e-06
        [updatestate_loads_eliminate]: 2.64001e-06
        [parameter_eliminate]: 1.87001e-06
        [specialize_transform]: 6.06998e-06
        [updatestate_useless_node_eliminater]: 5.76e-06
        [accelerated_algorithm]: 6.89001e-06
        [meta_shard_fg_expand]: 2.29001e-06
        [get_grad_eliminate_]: 5.35001e-06
        [merge_forward]: 4.40999e-06
        [cell_reuse_recompute_pass]: 1.49e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.712e-05
        [j_node_and_user_rematch]: 1.002e-05
        [meta_fg_expand]: 2.31998e-06
        [replace_old_param]: 1.143e-05
        [inline_without_move]: 5.82001e-06
        [renormalize]: 0.00092681
        [add_forward_monad_depend]: 5.52001e-06
        [auto_monad_grad]: 2.78e-06
        [auto_monad_eliminator]: 1.416e-05
        [cse]: 2.855e-05
        [replace_applicator]: 1.15e-05
    [Cycle 2]: 0.00032537, [27]
        [switch_simplify]: 5.60001e-06
        [loop_unroll]: 5.12999e-06
        [a_1]: 8.988e-05
        [with_stream_mark]: 9.31e-06
        [recompute_prepare]: 4.88001e-06
        [updatestate_depend_eliminate]: 2.69999e-06
        [updatestate_assign_eliminate]: 1.87001e-06
        [updatestate_loads_eliminate]: 1.83002e-06
        [parameter_eliminate]: 9.70002e-07
        [specialize_transform]: 4.75001e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 5.33002e-06
        [meta_shard_fg_expand]: 1.03001e-06
        [get_grad_eliminate_]: 4.25e-06
        [merge_forward]: 2.47001e-06
        [cell_reuse_recompute_pass]: 1.59e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.471e-05
        [j_node_and_user_rematch]: 8.75999e-06
        [meta_fg_expand]: 1.47001e-06
        [replace_old_param]: 7.56999e-06
        [inline_without_move]: 5.49e-06
        [renormalize]: 9.00181e-08
        [add_forward_monad_depend]: 1.32e-06
        [auto_monad_grad]: 1.03001e-06
        [auto_monad_eliminator]: 5.07999e-06
        [cse]: 1.037e-05
        [replace_applicator]: 4.85999e-06
[py_interpret_to_execute_after_opt_a]: 1.123e-05
[rewriter_after_opt_a]: 3.424e-05
[convert_after_rewriter]: 7.32997e-06
[order_py_execute_after_rewriter]: 5.59998e-06
[mutable_eliminate]: 0.00064411
[jit_opt_b]: 4.995e-05, [1]
    [Cycle 1]: 4.272e-05, [2]
        [frontend_op_eliminate]: 1.655e-05
        [inline_after_opt_a]: 1.508e-05
[cconv]: 2.599e-05
[loop_unroll]: 0.00041769
[jit_opt_after_cconv]: 0.00013672, [1]
    [Cycle 1]: 0.00013024, [11]
        [c_1]: 2.041e-05
        [parameter_eliminate]: 2.51e-06
        [updatestate_depend_eliminate]: 5.42001e-06
        [updatestate_assign_eliminate]: 2.15002e-06
        [updatestate_loads_eliminate]: 1.99e-06
        [cse]: 1.736e-05
        [call_graph_tuple_transform]: 1.9e-05
        [tuple_list_get_item_eliminator]: 5.32001e-06
        [none_parameter_eliminate]: 1.45999e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.20001e-06
[remove_dup_value]: 1.416e-05
[partial_unused_args_eliminate]: 2.30002e-06
[environ_conv]: 4.97e-06
[add_recomputation]: 4.925e-05
[cse_after_recomputation]: 2.031e-05, [1]
    [Cycle 1]: 1.448e-05, [1]
        [cse]: 8.66002e-06
[auto_monad_reorder]: 1.413e-05
[get_jit_bprop_graph]: 2.37999e-06
[rewriter_after_jit_bprop_graph]: 4.97e-06
[opt_after_jit_grad]: 0.00045074
[symbol_engine_optimizer]: 7.255e-05, [1]
    [Cycle 1]: 6.559e-05, [6]
        [build]: 3.65003e-06
        [elim_shapecalc]: 7e-06
        [elim_not_effective]: 1.327e-05
        [opt_reshape]: 5.49e-06
        [fold_const_symbol]: 8.13999e-06
        [renormalize]: 4.09986e-07
[validate]: 3.31e-05
Sums
    bootstrap                            :     0.000324s :  3.56%
    type_inference                       :     0.004620s : 50.80%
    event_method                         :     0.000020s :  0.22%
    auto_monad                           :     0.000069s :  0.76%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000039s :  0.43%
    rewriter_before_opt_a                :     0.000077s :  0.85%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000070s :  0.77%
    jit_opt_a.loop_unroll                :     0.000036s :  0.40%
    jit_opt_a.a_1                        :     0.000728s :  8.00%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.36%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000012s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.21%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000927s : 10.19%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.21%
    jit_opt_a.cse                        :     0.000039s :  0.43%
    jit_opt_a.replace_applicator         :     0.000016s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000034s :  0.38%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000644s :  7.08%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000026s :  0.29%
    loop_unroll                          :     0.000418s :  4.59%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000049s :  0.54%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000014s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.05%
    opt_after_jit_grad                   :     0.000451s :  4.96%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000033s :  0.36%


Time group info:
------[substitution.]   0.000263    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.47% :     0.000001s :      2: substitution.fold_const_symbol
 1.99% :     0.000005s :      3: substitution.graph_param_transform
85.60% :     0.000225s :      4: substitution.inline
 1.39% :     0.000004s :      4: substitution.j_node_and_user_rematch
 1.89% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.31% :     0.000006s :      2: substitution.replace_old_param
 5.51% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004551     2
79.59% :     0.003622s :      1: type_inference.infer
20.41% :     0.000929s :      1: type_inference.specialize
------[replace.]   0.000090     6
61.12% :     0.000055s :      4: replace.inline
38.88% :     0.000035s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000235     6
94.76% :     0.000222s :      4: match.inline
 5.24% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.53% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.20% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000002s :     10: predicate.addn_zero_filter
 1.87% :     0.000002s :     10: predicate.arithmetic_simplify
 1.31% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.00% :     0.000001s :     10: predicate.compare_switch_simplify
 2.10% :     0.000003s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.49% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.26% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.99% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.33% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.04% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.54% :     0.000002s :     10: predicate.environ_get_depend_swap
 1.45% :     0.000002s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.85% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 6.05% :     0.000007s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000002s :      6: predicate.less_batch_normalization
 1.36% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.96% :     0.000002s :     15: predicate.load_eliminater
 1.52% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.96% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.75% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.02% :     0.000001s :     10: predicate.merge_addn
 0.98% :     0.000001s :     10: predicate.minmaximum_grad
 1.54% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000001s :      3: predicate.opt_reshape
 2.11% :     0.000003s :     15: predicate.partial_eliminate
 1.11% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.83% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.60% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.66% :     0.000002s :     10: predicate.reshape_eliminate
 1.18% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.81% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.13% :     0.000001s :     10: predicate.same_eliminate
 0.89% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.35% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.15% :     0.000004s :     16: predicate.switch_defer_inline
 2.08% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.61% :     0.000009s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.41% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.14% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.87% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.44% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.69% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.44% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000696    12
46.27% :     0.000322s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.30% :     0.000051s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.43% :     0.000323s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015569    72
 0.33% :     0.000052s :      1: add_recomputation
 0.47% :     0.000073s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 2.19% :     0.000341s :      1: bootstrap
 0.18% :     0.000029s :      1: cconv
 0.06% :     0.000010s :      1: convert_after_rewriter
 0.14% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000024s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
40.43% :     0.006295s :      1: jit_opt_a
 0.90% :     0.000139s :      1: jit_opt_after_cconv
 0.34% :     0.000052s :      1: jit_opt_b
 2.73% :     0.000425s :      1: loop_unroll
 4.20% :     0.000653s :      1: mutable_eliminate
 6.15% :     0.000958s :     26: opt.transform.jit_opt_a
 0.30% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.16% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000031s :      4: opt.transform.symbol_engine_opt
 2.94% :     0.000458s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000042s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000016s :      1: remove_dup_value
 3.26% :     0.000508s :      1: renormalize.infer
 2.64% :     0.000411s :      1: renormalize.specialize
 0.04% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000039s :      1: rewriter_after_opt_a
 0.52% :     0.000080s :      1: rewriter_before_opt_a
 0.48% :     0.000075s :      1: symbol_engine_optimizer
29.78% :     0.004637s :      1: type_inference


TotalTime = 0.0122006, [30]
[bootstrap]: 0.00028704
[type_inference]: 0.0043075
[event_method]: 1.752e-05
[auto_monad]: 6.326e-05
[graph_reusing]: 5.52999e-06
[pre_auto_parallel]: 3.01001e-06
[py_interpret_to_execute]: 3.337e-05
[rewriter_before_opt_a]: 7.239e-05
[expand_dump_flag]: 3.04001e-06
[jit_opt_a]: 0.00522882, [2]
    [Cycle 1]: 0.00148857, [27]
        [switch_simplify]: 5.023e-05
        [loop_unroll]: 2.847e-05
        [a_1]: 0.00050767
        [with_stream_mark]: 1.837e-05
        [recompute_prepare]: 6.91999e-06
        [updatestate_depend_eliminate]: 3.58999e-06
        [updatestate_assign_eliminate]: 3.04001e-06
        [updatestate_loads_eliminate]: 2.71e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 5.72999e-06
        [updatestate_useless_node_eliminater]: 4.93001e-06
        [accelerated_algorithm]: 5.93002e-06
        [meta_shard_fg_expand]: 2.22001e-06
        [get_grad_eliminate_]: 5.34e-06
        [merge_forward]: 3.53e-06
        [cell_reuse_recompute_pass]: 1.23002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.398e-05
        [j_node_and_user_rematch]: 8.92e-06
        [meta_fg_expand]: 2.14e-06
        [replace_old_param]: 8.81002e-06
        [inline_without_move]: 5.26002e-06
        [renormalize]: 0.00057737
        [add_forward_monad_depend]: 5.40999e-06
        [auto_monad_grad]: 2.22999e-06
        [auto_monad_eliminator]: 1.322e-05
        [cse]: 2.676e-05
        [replace_applicator]: 1.12e-05
    [Cycle 2]: 0.00031464, [27]
        [switch_simplify]: 5.81e-06
        [loop_unroll]: 4.79e-06
        [a_1]: 8.731e-05
        [with_stream_mark]: 9.19e-06
        [recompute_prepare]: 4.92e-06
        [updatestate_depend_eliminate]: 2.37999e-06
        [updatestate_assign_eliminate]: 2.46e-06
        [updatestate_loads_eliminate]: 2.06e-06
        [parameter_eliminate]: 1.14e-06
        [specialize_transform]: 4.67e-06
        [updatestate_useless_node_eliminater]: 4.72e-06
        [accelerated_algorithm]: 5.03002e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.38999e-06
        [merge_forward]: 2.32001e-06
        [cell_reuse_recompute_pass]: 1.40001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.27e-05
        [j_node_and_user_rematch]: 7.92998e-06
        [meta_fg_expand]: 1.80001e-06
        [replace_old_param]: 6.98e-06
        [inline_without_move]: 7.16999e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.29e-06
        [auto_monad_grad]: 7.7e-07
        [auto_monad_eliminator]: 4.87e-06
        [cse]: 1.104e-05
        [replace_applicator]: 4.79e-06
[py_interpret_to_execute_after_opt_a]: 9.24e-06
[rewriter_after_opt_a]: 3.376e-05
[convert_after_rewriter]: 6.96001e-06
[order_py_execute_after_rewriter]: 4.87998e-06
[mutable_eliminate]: 0.00053752
[jit_opt_b]: 5.419e-05, [1]
    [Cycle 1]: 4.672e-05, [2]
        [frontend_op_eliminate]: 1.71e-05
        [inline_after_opt_a]: 1.744e-05
[cconv]: 2.782e-05
[loop_unroll]: 0.00052182
[jit_opt_after_cconv]: 0.00013943, [1]
    [Cycle 1]: 0.00013299, [11]
        [c_1]: 2.111e-05
        [parameter_eliminate]: 2.63998e-06
        [updatestate_depend_eliminate]: 5.30999e-06
        [updatestate_assign_eliminate]: 2.28002e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 1.86e-05
        [call_graph_tuple_transform]: 1.847e-05
        [tuple_list_get_item_eliminator]: 5.47001e-06
        [none_parameter_eliminate]: 1.42e-06
        [renormalize]: 5.00004e-07
        [switch_simplify]: 5.40999e-06
[remove_dup_value]: 1.611e-05
[partial_unused_args_eliminate]: 2.34001e-06
[environ_conv]: 4.72998e-06
[add_recomputation]: 4.679e-05
[cse_after_recomputation]: 2.13e-05, [1]
    [Cycle 1]: 1.506e-05, [1]
        [cse]: 9.37001e-06
[auto_monad_reorder]: 1.476e-05
[get_jit_bprop_graph]: 1.79e-06
[rewriter_after_jit_bprop_graph]: 5.37999e-06
[opt_after_jit_grad]: 0.00046038
[symbol_engine_optimizer]: 7.29e-05, [1]
    [Cycle 1]: 6.687e-05, [6]
        [build]: 3.66999e-06
        [elim_shapecalc]: 7.52998e-06
        [elim_not_effective]: 1.238e-05
        [opt_reshape]: 5.62001e-06
        [fold_const_symbol]: 9.04998e-06
        [renormalize]: 6.90023e-07
[validate]: 3.414e-05
Sums
    bootstrap                            :     0.000287s :  3.49%
    type_inference                       :     0.004307s : 52.42%
    event_method                         :     0.000018s :  0.21%
    auto_monad                           :     0.000063s :  0.77%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.04%
    py_interpret_to_execute              :     0.000033s :  0.41%
    rewriter_before_opt_a                :     0.000072s :  0.88%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000056s :  0.68%
    jit_opt_a.loop_unroll                :     0.000033s :  0.40%
    jit_opt_a.a_1                        :     0.000595s :  7.24%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.34%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000010s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.19%
    jit_opt_a.inline_without_move        :     0.000012s :  0.15%
    jit_opt_a.renormalize                :     0.000577s :  7.03%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.22%
    jit_opt_a.cse                        :     0.000038s :  0.46%
    jit_opt_a.replace_applicator         :     0.000016s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000034s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000538s :  6.54%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.21%
    cconv                                :     0.000028s :  0.34%
    loop_unroll                          :     0.000522s :  6.35%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.07%
    remove_dup_value                     :     0.000016s :  0.20%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000047s :  0.57%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.07%
    opt_after_jit_grad                   :     0.000460s :  5.60%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000034s :  0.42%


Time group info:
------[substitution.]   0.000188    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.87% :     0.000002s :      2: substitution.fold_const_symbol
 2.78% :     0.000005s :      3: substitution.graph_param_transform
83.05% :     0.000156s :      4: substitution.inline
 1.66% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.40% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.98% :     0.000004s :      2: substitution.replace_old_param
 6.27% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004247     2
79.68% :     0.003384s :      1: type_inference.infer
20.32% :     0.000863s :      1: type_inference.specialize
------[replace.]   0.000064     6
67.85% :     0.000043s :      4: replace.inline
32.15% :     0.000020s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000164     6
93.49% :     0.000154s :      4: match.inline
 6.51% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.52% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.06% :     0.000001s :     10: predicate.addn_check_dump
 1.16% :     0.000001s :     10: predicate.addn_zero_filter
 1.79% :     0.000002s :     10: predicate.arithmetic_simplify
 1.20% :     0.000001s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.34% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.23% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.77% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.15% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000001s :     10: predicate.environ_get_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 1.06% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.45% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.25% :     0.000001s :      6: predicate.less_batch_normalization
 1.59% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.81% :     0.000002s :     15: predicate.load_eliminater
 1.73% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.85% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 2.64% :     0.000003s :      3: predicate.mutable_eliminate
 0.54% :     0.000001s :      3: predicate.opt_reshape
 2.22% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.61% :     0.000002s :     10: predicate.reduce_eliminate
 1.71% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.62% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.69% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.42% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     10: predicate.reshape_eliminate
 1.21% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.25% :     0.000001s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.87% :     0.000001s :      6: predicate.special_op_eliminate
 1.05% :     0.000001s :      6: predicate.specialize_transform
 1.24% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.41% :     0.000003s :     16: predicate.switch_defer_inline
 2.07% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.92% :     0.000009s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.49% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.50% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.05% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.42% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.44% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.80% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.84% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000610    12
51.90% :     0.000316s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.87% :     0.000030s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.23% :     0.000264s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013635    72
 0.37% :     0.000050s :      1: add_recomputation
 0.49% :     0.000067s :      1: auto_monad
 0.13% :     0.000017s :      1: auto_monad_reorder
 2.20% :     0.000301s :      1: bootstrap
 0.22% :     0.000030s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.36% :     0.005231s :      1: jit_opt_a
 1.04% :     0.000142s :      1: jit_opt_after_cconv
 0.42% :     0.000057s :      1: jit_opt_b
 3.89% :     0.000530s :      1: loop_unroll
 4.01% :     0.000547s :      1: mutable_eliminate
 5.85% :     0.000798s :     26: opt.transform.jit_opt_a
 0.34% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.20% :     0.000027s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.43% :     0.000468s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.04% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000036s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000018s :      1: remove_dup_value
 2.13% :     0.000290s :      1: renormalize.infer
 2.05% :     0.000280s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000037s :      1: rewriter_after_opt_a
 0.55% :     0.000075s :      1: rewriter_before_opt_a
 0.55% :     0.000075s :      1: symbol_engine_optimizer
31.68% :     0.004320s :      1: type_inference


TotalTime = 0.0126579, [30]
[bootstrap]: 0.00030038
[type_inference]: 0.00445805
[event_method]: 1.894e-05
[auto_monad]: 6.506e-05
[graph_reusing]: 5.82001e-06
[pre_auto_parallel]: 2.67001e-06
[py_interpret_to_execute]: 3.522e-05
[rewriter_before_opt_a]: 7.257e-05
[expand_dump_flag]: 3.29001e-06
[jit_opt_a]: 0.00554562, [2]
    [Cycle 1]: 0.00165383, [27]
        [switch_simplify]: 5.249e-05
        [loop_unroll]: 2.951e-05
        [a_1]: 0.00052964
        [with_stream_mark]: 1.824e-05
        [recompute_prepare]: 7.77e-06
        [updatestate_depend_eliminate]: 4.50001e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.34001e-06
        [parameter_eliminate]: 1.92001e-06
        [specialize_transform]: 5.98002e-06
        [updatestate_useless_node_eliminater]: 5.04003e-06
        [accelerated_algorithm]: 5.61998e-06
        [meta_shard_fg_expand]: 2.20002e-06
        [get_grad_eliminate_]: 5.15001e-06
        [merge_forward]: 3.83001e-06
        [cell_reuse_recompute_pass]: 1.12e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.383e-05
        [j_node_and_user_rematch]: 9.02e-06
        [meta_fg_expand]: 2.04999e-06
        [replace_old_param]: 8.55999e-06
        [inline_without_move]: 5.37001e-06
        [renormalize]: 0.00070845
        [add_forward_monad_depend]: 5.89999e-06
        [auto_monad_grad]: 2.51e-06
        [auto_monad_eliminator]: 1.263e-05
        [cse]: 2.74e-05
        [replace_applicator]: 1.213e-05
    [Cycle 2]: 0.00031782, [27]
        [switch_simplify]: 5.41002e-06
        [loop_unroll]: 4.92e-06
        [a_1]: 8.786e-05
        [with_stream_mark]: 9.54e-06
        [recompute_prepare]: 5.17999e-06
        [updatestate_depend_eliminate]: 2.58e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 1.01002e-06
        [specialize_transform]: 4.87e-06
        [updatestate_useless_node_eliminater]: 4.57e-06
        [accelerated_algorithm]: 4.92e-06
        [meta_shard_fg_expand]: 1.34e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 2.41e-06
        [cell_reuse_recompute_pass]: 1.66e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.282e-05
        [j_node_and_user_rematch]: 7.65998e-06
        [meta_fg_expand]: 1.91e-06
        [replace_old_param]: 7.23e-06
        [inline_without_move]: 5.04e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 9.50007e-07
        [auto_monad_eliminator]: 5.23002e-06
        [cse]: 1.056e-05
        [replace_applicator]: 5.30999e-06
[py_interpret_to_execute_after_opt_a]: 9.47001e-06
[rewriter_after_opt_a]: 3.368e-05
[convert_after_rewriter]: 6.79999e-06
[order_py_execute_after_rewriter]: 4.79998e-06
[mutable_eliminate]: 0.00053349
[jit_opt_b]: 4.918e-05, [1]
    [Cycle 1]: 4.219e-05, [2]
        [frontend_op_eliminate]: 1.606e-05
        [inline_after_opt_a]: 1.501e-05
[cconv]: 2.577e-05
[loop_unroll]: 0.00042769
[jit_opt_after_cconv]: 0.00017849, [1]
    [Cycle 1]: 0.00017196, [11]
        [c_1]: 2.134e-05
        [parameter_eliminate]: 2.71e-06
        [updatestate_depend_eliminate]: 6.21e-06
        [updatestate_assign_eliminate]: 2.51e-06
        [updatestate_loads_eliminate]: 2.09999e-06
        [cse]: 1.977e-05
        [call_graph_tuple_transform]: 1.781e-05
        [tuple_list_get_item_eliminator]: 5.39e-06
        [none_parameter_eliminate]: 1.50999e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 5.74e-06
[remove_dup_value]: 1.571e-05
[partial_unused_args_eliminate]: 2.27999e-06
[environ_conv]: 5.54e-06
[add_recomputation]: 4.719e-05
[cse_after_recomputation]: 2.133e-05, [1]
    [Cycle 1]: 1.576e-05, [1]
        [cse]: 9.70002e-06
[auto_monad_reorder]: 1.58e-05
[get_jit_bprop_graph]: 2.16e-06
[rewriter_after_jit_bprop_graph]: 5.73002e-06
[opt_after_jit_grad]: 0.0004798
[symbol_engine_optimizer]: 7.408e-05, [1]
    [Cycle 1]: 6.749e-05, [6]
        [build]: 4.25e-06
        [elim_shapecalc]: 7.55e-06
        [elim_not_effective]: 1.308e-05
        [opt_reshape]: 5.57999e-06
        [fold_const_symbol]: 8.94e-06
        [renormalize]: 4.59986e-07
[validate]: 3.435e-05
Sums
    bootstrap                            :     0.000300s :  3.55%
    type_inference                       :     0.004458s : 52.65%
    event_method                         :     0.000019s :  0.22%
    auto_monad                           :     0.000065s :  0.77%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000035s :  0.42%
    rewriter_before_opt_a                :     0.000073s :  0.86%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000058s :  0.68%
    jit_opt_a.loop_unroll                :     0.000034s :  0.41%
    jit_opt_a.a_1                        :     0.000618s :  7.29%
    jit_opt_a.with_stream_mark           :     0.000028s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000709s :  8.37%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.21%
    jit_opt_a.cse                        :     0.000038s :  0.45%
    jit_opt_a.replace_applicator         :     0.000017s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.11%
    rewriter_after_opt_a                 :     0.000034s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000533s :  6.30%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000026s :  0.30%
    loop_unroll                          :     0.000428s :  5.05%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000016s :  0.19%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000047s :  0.56%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000480s :  5.67%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000034s :  0.41%


Time group info:
------[substitution.]   0.000200    23
 0.90% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.63% :     0.000005s :      3: substitution.graph_param_transform
83.72% :     0.000168s :      4: substitution.inline
 1.45% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.40% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.02% :     0.000004s :      2: substitution.replace_old_param
 6.19% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004390     2
78.88% :     0.003463s :      1: type_inference.infer
21.12% :     0.000927s :      1: type_inference.specialize
------[replace.]   0.000069     6
66.88% :     0.000046s :      4: replace.inline
33.12% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000176     6
93.62% :     0.000165s :      4: match.inline
 6.38% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.22% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.31% :     0.000001s :     10: predicate.addn_zero_filter
 1.65% :     0.000002s :     10: predicate.arithmetic_simplify
 1.36% :     0.000002s :     10: predicate.cast_eliminate
 0.39% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.83% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.21% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.15% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.09% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.44% :     0.000001s :      3: predicate.elim_not_effective
 0.70% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.89% :     0.000001s :      6: predicate.get_grad_eliminate
 0.44% :     0.000001s :      3: predicate.graph_param_transform
 5.70% :     0.000006s :     22: predicate.inline
 1.02% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.34% :     0.000002s :      6: predicate.less_batch_normalization
 1.53% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.69% :     0.000002s :     15: predicate.load_eliminater
 1.63% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.00% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.05% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.11% :     0.000001s :     10: predicate.minmaximum_grad
 1.80% :     0.000002s :      3: predicate.mutable_eliminate
 0.48% :     0.000001s :      3: predicate.opt_reshape
 2.19% :     0.000002s :     15: predicate.partial_eliminate
 1.19% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.64% :     0.000002s :     10: predicate.reduce_eliminate
 1.48% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.68% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.75% :     0.000001s :      6: predicate.replace_old_param
 0.29% :     0.000000s :      3: predicate.reset_defer_inline
 1.57% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.76% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.92% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.40% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.11% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.53% :     0.000003s :     16: predicate.switch_defer_inline
 2.14% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.82% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.18% :     0.000001s :     10: predicate.transpose_eliminate
 1.40% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.23% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.93% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.36% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.02% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.65% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.63% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000635    12
47.84% :     0.000304s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.86% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.29% :     0.000300s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014238    72
 0.35% :     0.000050s :      1: add_recomputation
 0.48% :     0.000068s :      1: auto_monad
 0.13% :     0.000019s :      1: auto_monad_reorder
 2.24% :     0.000319s :      1: bootstrap
 0.20% :     0.000028s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000023s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.97% :     0.005549s :      1: jit_opt_a
 1.28% :     0.000182s :      1: jit_opt_after_cconv
 0.36% :     0.000052s :      1: jit_opt_b
 3.06% :     0.000436s :      1: loop_unroll
 3.80% :     0.000542s :      1: mutable_eliminate
 5.78% :     0.000822s :     26: opt.transform.jit_opt_a
 0.33% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.42% :     0.000488s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000038s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000018s :      1: remove_dup_value
 2.52% :     0.000359s :      1: renormalize.infer
 2.40% :     0.000342s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000037s :      1: rewriter_after_opt_a
 0.53% :     0.000075s :      1: rewriter_before_opt_a
 0.54% :     0.000077s :      1: symbol_engine_optimizer
31.41% :     0.004472s :      1: type_inference


TotalTime = 0.0120205, [30]
[bootstrap]: 0.00028354
[type_inference]: 0.00406397
[event_method]: 1.686e-05
[auto_monad]: 4.631e-05
[graph_reusing]: 4.16001e-06
[pre_auto_parallel]: 2.16e-06
[py_interpret_to_execute]: 3.024e-05
[rewriter_before_opt_a]: 6.505e-05
[expand_dump_flag]: 2.18998e-06
[jit_opt_a]: 0.00515204, [2]
    [Cycle 1]: 0.00161653, [27]
        [switch_simplify]: 4.76e-05
        [loop_unroll]: 2.902e-05
        [a_1]: 0.00050702
        [with_stream_mark]: 1.592e-05
        [recompute_prepare]: 7.5e-06
        [updatestate_depend_eliminate]: 3.31999e-06
        [updatestate_assign_eliminate]: 2.19001e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 1.25999e-06
        [specialize_transform]: 5.77001e-06
        [updatestate_useless_node_eliminater]: 4.85001e-06
        [accelerated_algorithm]: 5.50001e-06
        [meta_shard_fg_expand]: 1.49e-06
        [get_grad_eliminate_]: 5.82001e-06
        [merge_forward]: 2.71e-06
        [cell_reuse_recompute_pass]: 1.12999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.253e-05
        [j_node_and_user_rematch]: 8.72e-06
        [meta_fg_expand]: 2.50002e-06
        [replace_old_param]: 8.10999e-06
        [inline_without_move]: 5.77001e-06
        [renormalize]: 0.00070633
        [add_forward_monad_depend]: 5.93998e-06
        [auto_monad_grad]: 2.19001e-06
        [auto_monad_eliminator]: 1.484e-05
        [cse]: 2.588e-05
        [replace_applicator]: 1.175e-05
    [Cycle 2]: 0.00032648, [27]
        [switch_simplify]: 5.35001e-06
        [loop_unroll]: 5.05001e-06
        [a_1]: 9.199e-05
        [with_stream_mark]: 1.109e-05
        [recompute_prepare]: 5.10001e-06
        [updatestate_depend_eliminate]: 2.51e-06
        [updatestate_assign_eliminate]: 2.01e-06
        [updatestate_loads_eliminate]: 1.81e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.88001e-06
        [updatestate_useless_node_eliminater]: 4.82998e-06
        [accelerated_algorithm]: 5.20999e-06
        [meta_shard_fg_expand]: 1.20999e-06
        [get_grad_eliminate_]: 4.65999e-06
        [merge_forward]: 2.74999e-06
        [cell_reuse_recompute_pass]: 1.64998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.347e-05
        [j_node_and_user_rematch]: 7.7e-06
        [meta_fg_expand]: 1.67999e-06
        [replace_old_param]: 1.482e-05
        [inline_without_move]: 5.09e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.17e-06
        [auto_monad_grad]: 7.79983e-07
        [auto_monad_eliminator]: 5.07e-06
        [cse]: 1.082e-05
        [replace_applicator]: 4.67e-06
[py_interpret_to_execute_after_opt_a]: 1.13e-05
[rewriter_after_opt_a]: 3.65e-05
[convert_after_rewriter]: 7.33999e-06
[order_py_execute_after_rewriter]: 4.74e-06
[mutable_eliminate]: 0.00067372
[jit_opt_b]: 5.52e-05, [1]
    [Cycle 1]: 4.668e-05, [2]
        [frontend_op_eliminate]: 1.821e-05
        [inline_after_opt_a]: 1.64e-05
[cconv]: 2.983e-05
[loop_unroll]: 0.0004646
[jit_opt_after_cconv]: 0.00015763, [1]
    [Cycle 1]: 0.00014971, [11]
        [c_1]: 2.207e-05
        [parameter_eliminate]: 3.4e-06
        [updatestate_depend_eliminate]: 7.97998e-06
        [updatestate_assign_eliminate]: 3.2e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 2.428e-05
        [call_graph_tuple_transform]: 2.075e-05
        [tuple_list_get_item_eliminator]: 6.04999e-06
        [none_parameter_eliminate]: 1.55001e-06
        [renormalize]: 2.50002e-07
        [switch_simplify]: 5.65001e-06
[remove_dup_value]: 1.464e-05
[partial_unused_args_eliminate]: 2.14e-06
[environ_conv]: 5.29e-06
[add_recomputation]: 5.133e-05
[cse_after_recomputation]: 2.147e-05, [1]
    [Cycle 1]: 1.533e-05, [1]
        [cse]: 9.10001e-06
[auto_monad_reorder]: 1.581e-05
[get_jit_bprop_graph]: 2.21e-06
[rewriter_after_jit_bprop_graph]: 8.49998e-06
[opt_after_jit_grad]: 0.00049777
[symbol_engine_optimizer]: 7.912e-05, [1]
    [Cycle 1]: 7.155e-05, [6]
        [build]: 4.05998e-06
        [elim_shapecalc]: 8.38999e-06
        [elim_not_effective]: 1.428e-05
        [opt_reshape]: 7.04001e-06
        [fold_const_symbol]: 9.49999e-06
        [renormalize]: 5.60016e-07
[validate]: 4.043e-05
Sums
    bootstrap                            :     0.000284s :  3.45%
    type_inference                       :     0.004064s : 49.38%
    event_method                         :     0.000017s :  0.20%
    auto_monad                           :     0.000046s :  0.56%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000030s :  0.37%
    rewriter_before_opt_a                :     0.000065s :  0.79%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000053s :  0.64%
    jit_opt_a.loop_unroll                :     0.000034s :  0.41%
    jit_opt_a.a_1                        :     0.000599s :  7.28%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.13%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000016s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000023s :  0.28%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000706s :  8.58%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000020s :  0.24%
    jit_opt_a.cse                        :     0.000037s :  0.45%
    jit_opt_a.replace_applicator         :     0.000016s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.14%
    rewriter_after_opt_a                 :     0.000037s :  0.44%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000674s :  8.19%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.20%
    cconv                                :     0.000030s :  0.36%
    loop_unroll                          :     0.000465s :  5.65%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.27%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.10%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.30%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000021s :  0.25%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000015s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000051s :  0.62%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.10%
    opt_after_jit_grad                   :     0.000498s :  6.05%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.17%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.09%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.12%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.49%


Time group info:
------[substitution.]   0.000186    23
 1.08% :     0.000002s :      2: substitution.elim_not_effective
 0.65% :     0.000001s :      2: substitution.fold_const_symbol
 3.03% :     0.000006s :      3: substitution.graph_param_transform
80.63% :     0.000150s :      4: substitution.inline
 1.46% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.27% :     0.000004s :      4: substitution.remove_not_recompute_node
 5.82% :     0.000011s :      2: substitution.replace_old_param
 5.07% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004014     2
80.39% :     0.003227s :      1: type_inference.infer
19.61% :     0.000787s :      1: type_inference.specialize
------[replace.]   0.000065     6
66.37% :     0.000043s :      4: replace.inline
33.63% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000156     6
94.70% :     0.000147s :      4: match.inline
 5.30% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.28% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.31% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.28% :     0.000001s :     10: predicate.addn_zero_filter
 2.24% :     0.000003s :     10: predicate.arithmetic_simplify
 1.23% :     0.000001s :     10: predicate.cast_eliminate
 0.36% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.31% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.20% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.34% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.11% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.29% :     0.000001s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.28% :     0.000001s :      6: predicate.get_grad_eliminate
 0.30% :     0.000000s :      3: predicate.graph_param_transform
 5.19% :     0.000006s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.07% :     0.000001s :      6: predicate.less_batch_normalization
 1.71% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.72% :     0.000002s :     15: predicate.load_eliminater
 1.35% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.85% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.98% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.26% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.89% :     0.000002s :      3: predicate.mutable_eliminate
 1.05% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000002s :     15: predicate.partial_eliminate
 1.12% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     10: predicate.reduce_eliminate
 1.60% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.67% :     0.000002s :     18: predicate.replace_applicator
 0.52% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.48% :     0.000002s :     10: predicate.reshape_eliminate
 1.22% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.03% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.42% :     0.000002s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.36% :     0.000002s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.43% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.37% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.53% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.35% :     0.000008s :     53: predicate.switch_simplify
 1.11% :     0.000001s :     10: predicate.tile_eliminate
 1.33% :     0.000002s :     10: predicate.transpose_eliminate
 1.24% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.31% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.83% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.62% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.65% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.68% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000582    12
47.71% :     0.000278s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.92% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.37% :     0.000270s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013592    72
 0.40% :     0.000054s :      1: add_recomputation
 0.36% :     0.000049s :      1: auto_monad
 0.14% :     0.000019s :      1: auto_monad_reorder
 2.20% :     0.000298s :      1: bootstrap
 0.24% :     0.000032s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000024s :      1: cse_after_recomputation
 0.06% :     0.000009s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000006s :      1: graph_reusing
37.93% :     0.005155s :      1: jit_opt_a
 1.18% :     0.000161s :      1: jit_opt_after_cconv
 0.42% :     0.000058s :      1: jit_opt_b
 3.49% :     0.000474s :      1: loop_unroll
 5.03% :     0.000684s :      1: mutable_eliminate
 5.94% :     0.000808s :     26: opt.transform.jit_opt_a
 0.38% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000028s :      4: opt.transform.jit_opt_b
 0.09% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.26% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.72% :     0.000506s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000033s :      1: py_interpret_to_execute
 0.10% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000017s :      1: remove_dup_value
 2.64% :     0.000359s :      1: renormalize.infer
 2.50% :     0.000340s :      1: renormalize.specialize
 0.08% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.29% :     0.000040s :      1: rewriter_after_opt_a
 0.50% :     0.000068s :      1: rewriter_before_opt_a
 0.60% :     0.000082s :      1: symbol_engine_optimizer
29.98% :     0.004074s :      1: type_inference


TotalTime = 0.0142436, [30]
[bootstrap]: 0.00032485
[type_inference]: 0.00476057
[event_method]: 2.11e-05
[auto_monad]: 6.938e-05
[graph_reusing]: 5.99e-06
[pre_auto_parallel]: 2.83998e-06
[py_interpret_to_execute]: 4.038e-05
[rewriter_before_opt_a]: 8.121e-05
[expand_dump_flag]: 3.83999e-06
[jit_opt_a]: 0.00645086, [2]
    [Cycle 1]: 0.00203107, [27]
        [switch_simplify]: 6.657e-05
        [loop_unroll]: 3.076e-05
        [a_1]: 0.00063288
        [with_stream_mark]: 2.322e-05
        [recompute_prepare]: 7.41001e-06
        [updatestate_depend_eliminate]: 3.86001e-06
        [updatestate_assign_eliminate]: 2.99999e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [parameter_eliminate]: 1.96e-06
        [specialize_transform]: 6.24001e-06
        [updatestate_useless_node_eliminater]: 4.78001e-06
        [accelerated_algorithm]: 5.68997e-06
        [meta_shard_fg_expand]: 2.09e-06
        [get_grad_eliminate_]: 5.86e-06
        [merge_forward]: 3.61999e-06
        [cell_reuse_recompute_pass]: 1.25999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.408e-05
        [j_node_and_user_rematch]: 9.03002e-06
        [meta_fg_expand]: 2.09999e-06
        [replace_old_param]: 9.51e-06
        [inline_without_move]: 5.71998e-06
        [renormalize]: 0.00091539
        [add_forward_monad_depend]: 6.21e-06
        [auto_monad_grad]: 2.86999e-06
        [auto_monad_eliminator]: 1.528e-05
        [cse]: 3.118e-05
        [replace_applicator]: 1.54e-05
    [Cycle 2]: 0.0003454, [27]
        [switch_simplify]: 5.67001e-06
        [loop_unroll]: 4.96002e-06
        [a_1]: 9.463e-05
        [with_stream_mark]: 1.338e-05
        [recompute_prepare]: 4.96002e-06
        [updatestate_depend_eliminate]: 2.74001e-06
        [updatestate_assign_eliminate]: 2.28002e-06
        [updatestate_loads_eliminate]: 2.32001e-06
        [parameter_eliminate]: 9.80013e-07
        [specialize_transform]: 5.05999e-06
        [updatestate_useless_node_eliminater]: 5.20999e-06
        [accelerated_algorithm]: 5.62999e-06
        [meta_shard_fg_expand]: 1.52001e-06
        [get_grad_eliminate_]: 4.42e-06
        [merge_forward]: 3.21001e-06
        [cell_reuse_recompute_pass]: 2.48e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.511e-05
        [j_node_and_user_rematch]: 8.05999e-06
        [meta_fg_expand]: 2.05002e-06
        [replace_old_param]: 8.64e-06
        [inline_without_move]: 4.94998e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 1.14003e-06
        [auto_monad_grad]: 1.35001e-06
        [auto_monad_eliminator]: 5.86998e-06
        [cse]: 1.369e-05
        [replace_applicator]: 5.72999e-06
[py_interpret_to_execute_after_opt_a]: 1.391e-05
[rewriter_after_opt_a]: 3.689e-05
[convert_after_rewriter]: 1.037e-05
[order_py_execute_after_rewriter]: 4.80001e-06
[mutable_eliminate]: 0.00073304
[jit_opt_b]: 5.439e-05, [1]
    [Cycle 1]: 4.583e-05, [2]
        [frontend_op_eliminate]: 1.896e-05
        [inline_after_opt_a]: 1.538e-05
[cconv]: 2.759e-05
[loop_unroll]: 0.00047384
[jit_opt_after_cconv]: 0.00015344, [1]
    [Cycle 1]: 0.00014657, [11]
        [c_1]: 2.17e-05
        [parameter_eliminate]: 3.90998e-06
        [updatestate_depend_eliminate]: 6.76999e-06
        [updatestate_assign_eliminate]: 2.46998e-06
        [updatestate_loads_eliminate]: 2.19999e-06
        [cse]: 2.526e-05
        [call_graph_tuple_transform]: 1.992e-05
        [tuple_list_get_item_eliminator]: 5.54e-06
        [none_parameter_eliminate]: 1.80001e-06
        [renormalize]: 9.20001e-07
        [switch_simplify]: 5.49998e-06
[remove_dup_value]: 1.447e-05
[partial_unused_args_eliminate]: 2.07001e-06
[environ_conv]: 5.71003e-06
[add_recomputation]: 4.955e-05
[cse_after_recomputation]: 2.171e-05, [1]
    [Cycle 1]: 1.536e-05, [1]
        [cse]: 9.12001e-06
[auto_monad_reorder]: 1.512e-05
[get_jit_bprop_graph]: 1.86e-06
[rewriter_after_jit_bprop_graph]: 7.61001e-06
[opt_after_jit_grad]: 0.0005378
[symbol_engine_optimizer]: 7.825e-05, [1]
    [Cycle 1]: 7.015e-05, [6]
        [build]: 4.58999e-06
        [elim_shapecalc]: 8e-06
        [elim_not_effective]: 1.392e-05
        [opt_reshape]: 5.89e-06
        [fold_const_symbol]: 8.69e-06
        [renormalize]: 8.39995e-07
[validate]: 3.976e-05
Sums
    bootstrap                            :     0.000325s :  3.41%
    type_inference                       :     0.004761s : 50.00%
    event_method                         :     0.000021s :  0.22%
    auto_monad                           :     0.000069s :  0.73%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000040s :  0.42%
    rewriter_before_opt_a                :     0.000081s :  0.85%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000072s :  0.76%
    jit_opt_a.loop_unroll                :     0.000036s :  0.38%
    jit_opt_a.a_1                        :     0.000728s :  7.64%
    jit_opt_a.with_stream_mark           :     0.000037s :  0.38%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000029s :  0.31%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.19%
    jit_opt_a.inline_without_move        :     0.000011s :  0.11%
    jit_opt_a.renormalize                :     0.000915s :  9.62%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.22%
    jit_opt_a.cse                        :     0.000045s :  0.47%
    jit_opt_a.replace_applicator         :     0.000021s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.15%
    rewriter_after_opt_a                 :     0.000037s :  0.39%
    convert_after_rewriter               :     0.000010s :  0.11%
    order_py_execute_after_rewriter      :     0.000005s :  0.05%
    mutable_eliminate                    :     0.000733s :  7.70%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.16%
    cconv                                :     0.000028s :  0.29%
    loop_unroll                          :     0.000474s :  4.98%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000025s :  0.27%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.15%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000050s :  0.52%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.08%
    opt_after_jit_grad                   :     0.000538s :  5.65%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000040s :  0.42%


Time group info:
------[substitution.]   0.000239    23
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.62% :     0.000001s :      2: substitution.fold_const_symbol
 2.36% :     0.000006s :      3: substitution.graph_param_transform
84.91% :     0.000203s :      4: substitution.inline
 1.46% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.26% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.33% :     0.000006s :      2: substitution.replace_old_param
 5.28% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004684     2
78.63% :     0.003683s :      1: type_inference.infer
21.37% :     0.001001s :      1: type_inference.specialize
------[replace.]   0.000127     6
80.09% :     0.000102s :      4: replace.inline
19.91% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000211     6
94.59% :     0.000200s :      4: match.inline
 5.41% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000123   691
 1.42% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.15% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.44% :     0.000002s :     10: predicate.addn_zero_filter
 2.51% :     0.000003s :     10: predicate.arithmetic_simplify
 1.39% :     0.000002s :     10: predicate.cast_eliminate
 0.71% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.98% :     0.000001s :     10: predicate.compare_switch_simplify
 1.27% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.20% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.38% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.21% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.62% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.18% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.40% :     0.000002s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.83% :     0.000001s :      6: predicate.get_grad_eliminate
 0.31% :     0.000000s :      3: predicate.graph_param_transform
 5.35% :     0.000007s :     22: predicate.inline
 0.89% :     0.000001s :      6: predicate.inline_without_move
 0.35% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000002s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.83% :     0.000002s :     15: predicate.load_eliminater
 1.27% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.16% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.93% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.98% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.97% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.09% :     0.000003s :     15: predicate.partial_eliminate
 1.14% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.32% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.46% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.49% :     0.000002s :     18: predicate.replace_applicator
 0.50% :     0.000001s :      6: predicate.replace_old_param
 0.27% :     0.000000s :      3: predicate.reset_defer_inline
 1.63% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.20% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.06% :     0.000001s :      6: predicate.special_op_eliminate
 0.97% :     0.000001s :      6: predicate.specialize_transform
 1.23% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.19% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.36% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.58% :     0.000003s :     16: predicate.switch_defer_inline
 2.10% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.64% :     0.000009s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.40% :     0.000002s :     10: predicate.transpose_eliminate
 1.54% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.29% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.16% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.84% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.53% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.91% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.66% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.67% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000683    12
47.57% :     0.000325s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.58% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.85% :     0.000320s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.016165    72
 0.33% :     0.000053s :      1: add_recomputation
 0.45% :     0.000074s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 2.14% :     0.000346s :      1: bootstrap
 0.19% :     0.000030s :      1: cconv
 0.08% :     0.000012s :      1: convert_after_rewriter
 0.15% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.16% :     0.000026s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.93% :     0.006454s :      1: jit_opt_a
 0.97% :     0.000156s :      1: jit_opt_after_cconv
 0.35% :     0.000057s :      1: jit_opt_b
 2.99% :     0.000483s :      1: loop_unroll
 4.60% :     0.000743s :      1: mutable_eliminate
 5.92% :     0.000957s :     26: opt.transform.jit_opt_a
 0.31% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000028s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.39% :     0.000548s :      1: opt_after_jit_grad
 0.04% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.27% :     0.000043s :      1: py_interpret_to_execute
 0.10% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000017s :      1: remove_dup_value
 2.94% :     0.000475s :      1: renormalize.infer
 2.67% :     0.000432s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000040s :      1: rewriter_after_opt_a
 0.53% :     0.000085s :      1: rewriter_before_opt_a
 0.50% :     0.000081s :      1: symbol_engine_optimizer
29.57% :     0.004780s :      1: type_inference


TotalTime = 0.0121869, [30]
[bootstrap]: 0.0003087
[type_inference]: 0.00434259
[event_method]: 1.633e-05
[auto_monad]: 4.921e-05
[graph_reusing]: 3.94002e-06
[pre_auto_parallel]: 2.31998e-06
[py_interpret_to_execute]: 3.132e-05
[rewriter_before_opt_a]: 6.537e-05
[expand_dump_flag]: 2.08002e-06
[jit_opt_a]: 0.00524818, [2]
    [Cycle 1]: 0.00157783, [27]
        [switch_simplify]: 4.953e-05
        [loop_unroll]: 2.894e-05
        [a_1]: 0.0004973
        [with_stream_mark]: 1.496e-05
        [recompute_prepare]: 6.88998e-06
        [updatestate_depend_eliminate]: 2.99999e-06
        [updatestate_assign_eliminate]: 2.27999e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [parameter_eliminate]: 1.28002e-06
        [specialize_transform]: 5.77001e-06
        [updatestate_useless_node_eliminater]: 4.83001e-06
        [accelerated_algorithm]: 5.48002e-06
        [meta_shard_fg_expand]: 1.97001e-06
        [get_grad_eliminate_]: 5.30999e-06
        [merge_forward]: 2.99999e-06
        [cell_reuse_recompute_pass]: 8.89995e-07
        [cell_reuse_handle_not_recompute_node_pass]: 1.124e-05
        [j_node_and_user_rematch]: 9.12001e-06
        [meta_fg_expand]: 1.77999e-06
        [replace_old_param]: 8.54002e-06
        [inline_without_move]: 5.19998e-06
        [renormalize]: 0.00068318
        [add_forward_monad_depend]: 6.14999e-06
        [auto_monad_grad]: 2.36e-06
        [auto_monad_eliminator]: 1.298e-05
        [cse]: 2.799e-05
        [replace_applicator]: 1.133e-05
    [Cycle 2]: 0.00030894, [27]
        [switch_simplify]: 5.54998e-06
        [loop_unroll]: 4.80999e-06
        [a_1]: 8.853e-05
        [with_stream_mark]: 8.38999e-06
        [recompute_prepare]: 4.84e-06
        [updatestate_depend_eliminate]: 2.41998e-06
        [updatestate_assign_eliminate]: 2.06998e-06
        [updatestate_loads_eliminate]: 1.96998e-06
        [parameter_eliminate]: 8.00006e-07
        [specialize_transform]: 4.77e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 4.85001e-06
        [meta_shard_fg_expand]: 1.19e-06
        [get_grad_eliminate_]: 4.52e-06
        [merge_forward]: 2.43e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.248e-05
        [j_node_and_user_rematch]: 7.75e-06
        [meta_fg_expand]: 1.43002e-06
        [replace_old_param]: 8.18999e-06
        [inline_without_move]: 4.57998e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.16997e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 4.55999e-06
        [cse]: 9.84999e-06
        [replace_applicator]: 4.74002e-06
[py_interpret_to_execute_after_opt_a]: 1.537e-05
[rewriter_after_opt_a]: 3.441e-05
[convert_after_rewriter]: 6.91001e-06
[order_py_execute_after_rewriter]: 5.44e-06
[mutable_eliminate]: 0.00053948
[jit_opt_b]: 5.386e-05, [1]
    [Cycle 1]: 4.728e-05, [2]
        [frontend_op_eliminate]: 2.058e-05
        [inline_after_opt_a]: 1.543e-05
[cconv]: 2.452e-05
[loop_unroll]: 0.00042164
[jit_opt_after_cconv]: 0.00013591, [1]
    [Cycle 1]: 0.00012975, [11]
        [c_1]: 2.048e-05
        [parameter_eliminate]: 2.31e-06
        [updatestate_depend_eliminate]: 5.55001e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 2.08002e-06
        [cse]: 1.777e-05
        [call_graph_tuple_transform]: 1.803e-05
        [tuple_list_get_item_eliminator]: 5.46e-06
        [none_parameter_eliminate]: 1.74998e-06
        [renormalize]: 4.50003e-07
        [switch_simplify]: 5.02999e-06
[remove_dup_value]: 1.412e-05
[partial_unused_args_eliminate]: 2.11e-06
[environ_conv]: 4.90999e-06
[add_recomputation]: 4.694e-05
[cse_after_recomputation]: 2.146e-05, [1]
    [Cycle 1]: 1.537e-05, [1]
        [cse]: 8.79003e-06
[auto_monad_reorder]: 1.623e-05
[get_jit_bprop_graph]: 2.14999e-06
[rewriter_after_jit_bprop_graph]: 4.94e-06
[opt_after_jit_grad]: 0.00048524
[symbol_engine_optimizer]: 7.396e-05, [1]
    [Cycle 1]: 6.71e-05, [6]
        [build]: 4.23001e-06
        [elim_shapecalc]: 7.81001e-06
        [elim_not_effective]: 1.308e-05
        [opt_reshape]: 5.57999e-06
        [fold_const_symbol]: 8.69e-06
        [renormalize]: 5.00004e-07
[validate]: 3.61e-05
Sums
    bootstrap                            :     0.000309s :  3.74%
    type_inference                       :     0.004343s : 52.57%
    event_method                         :     0.000016s :  0.20%
    auto_monad                           :     0.000049s :  0.60%
    graph_reusing                        :     0.000004s :  0.05%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000031s :  0.38%
    rewriter_before_opt_a                :     0.000065s :  0.79%
    expand_dump_flag                     :     0.000002s :  0.03%
    jit_opt_a.switch_simplify            :     0.000055s :  0.67%
    jit_opt_a.loop_unroll                :     0.000034s :  0.41%
    jit_opt_a.a_1                        :     0.000586s :  7.09%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.28%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000005s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000005s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000683s :  8.27%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.21%
    jit_opt_a.cse                        :     0.000038s :  0.46%
    jit_opt_a.replace_applicator         :     0.000016s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000015s :  0.19%
    rewriter_after_opt_a                 :     0.000034s :  0.42%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.07%
    mutable_eliminate                    :     0.000539s :  6.53%
    jit_opt_b.frontend_op_eliminate      :     0.000021s :  0.25%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.19%
    cconv                                :     0.000025s :  0.30%
    loop_unroll                          :     0.000422s :  5.10%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000018s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000047s :  0.57%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.20%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000485s :  5.87%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000036s :  0.44%


Time group info:
------[substitution.]   0.000175    23
 1.09% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000001s :      2: substitution.fold_const_symbol
 2.99% :     0.000005s :      3: substitution.graph_param_transform
83.50% :     0.000146s :      4: substitution.inline
 1.78% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.37% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.61% :     0.000005s :      2: substitution.replace_old_param
 4.94% :     0.000009s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004286     2
79.88% :     0.003424s :      1: type_inference.infer
20.12% :     0.000862s :      1: type_inference.specialize
------[replace.]   0.000065     6
66.09% :     0.000043s :      4: replace.inline
33.91% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000151     6
94.95% :     0.000143s :      4: match.inline
 5.05% :     0.000008s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.46% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.27% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.16% :     0.000001s :     10: predicate.addn_zero_filter
 2.11% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.51% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.58% :     0.000002s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.30% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.04% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.72% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.24% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.15% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.22% :     0.000001s :     10: predicate.environ_get_eliminate
 1.14% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 1.25% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.60% :     0.000006s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.04% :     0.000001s :      6: predicate.less_batch_normalization
 1.61% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.67% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.96% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.90% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.17% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.70% :     0.000002s :      3: predicate.mutable_eliminate
 0.50% :     0.000001s :      3: predicate.opt_reshape
 2.14% :     0.000002s :     15: predicate.partial_eliminate
 1.37% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.50% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.73% :     0.000002s :     18: predicate.replace_applicator
 0.59% :     0.000001s :      6: predicate.replace_old_param
 0.34% :     0.000000s :      3: predicate.reset_defer_inline
 1.51% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.74% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.40% :     0.000002s :     10: predicate.same_eliminate
 0.52% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.99% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.49% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.16% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.51% :     0.000008s :     53: predicate.switch_simplify
 1.35% :     0.000002s :     10: predicate.tile_eliminate
 1.16% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.51% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.38% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.75% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.99% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.55% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.77% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000604    12
48.65% :     0.000294s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.09% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.26% :     0.000280s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013713    72
 0.36% :     0.000050s :      1: add_recomputation
 0.38% :     0.000052s :      1: auto_monad
 0.14% :     0.000019s :      1: auto_monad_reorder
 2.38% :     0.000327s :      1: bootstrap
 0.20% :     0.000027s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.17% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.14% :     0.000020s :      1: event_method
 0.03% :     0.000004s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000006s :      1: graph_reusing
38.30% :     0.005251s :      1: jit_opt_a
 1.01% :     0.000139s :      1: jit_opt_after_cconv
 0.41% :     0.000057s :      1: jit_opt_b
 3.13% :     0.000430s :      1: loop_unroll
 3.99% :     0.000548s :      1: mutable_eliminate
 5.74% :     0.000787s :     26: opt.transform.jit_opt_a
 0.33% :     0.000046s :      4: opt.transform.jit_opt_after_cconv
 0.21% :     0.000029s :      4: opt.transform.jit_opt_b
 0.09% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.60% :     0.000493s :      1: opt_after_jit_grad
 0.06% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.25% :     0.000034s :      1: py_interpret_to_execute
 0.13% :     0.000018s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.49% :     0.000342s :      1: renormalize.infer
 2.44% :     0.000335s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000037s :      1: rewriter_after_opt_a
 0.50% :     0.000068s :      1: rewriter_before_opt_a
 0.56% :     0.000076s :      1: symbol_engine_optimizer
31.76% :     0.004355s :      1: type_inference


TotalTime = 0.0129104, [30]
[bootstrap]: 0.00031905
[type_inference]: 0.00458644
[event_method]: 1.832e-05
[auto_monad]: 6.776e-05
[graph_reusing]: 6.62002e-06
[pre_auto_parallel]: 2.48e-06
[py_interpret_to_execute]: 3.636e-05
[rewriter_before_opt_a]: 7.375e-05
[expand_dump_flag]: 3.33998e-06
[jit_opt_a]: 0.00563761, [2]
    [Cycle 1]: 0.00170206, [27]
        [switch_simplify]: 5.735e-05
        [loop_unroll]: 3.072e-05
        [a_1]: 0.00055323
        [with_stream_mark]: 1.946e-05
        [recompute_prepare]: 6.73998e-06
        [updatestate_depend_eliminate]: 3.85e-06
        [updatestate_assign_eliminate]: 2.98e-06
        [updatestate_loads_eliminate]: 2.66e-06
        [parameter_eliminate]: 2.11e-06
        [specialize_transform]: 5.96e-06
        [updatestate_useless_node_eliminater]: 5.02999e-06
        [accelerated_algorithm]: 5.75001e-06
        [meta_shard_fg_expand]: 1.86e-06
        [get_grad_eliminate_]: 5.56e-06
        [merge_forward]: 3.60003e-06
        [cell_reuse_recompute_pass]: 1.20001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.39e-05
        [j_node_and_user_rematch]: 9.12001e-06
        [meta_fg_expand]: 2.18998e-06
        [replace_old_param]: 9.76e-06
        [inline_without_move]: 5.43002e-06
        [renormalize]: 0.00073884
        [add_forward_monad_depend]: 4.41002e-06
        [auto_monad_grad]: 1.60999e-06
        [auto_monad_eliminator]: 1.016e-05
        [cse]: 1.699e-05
        [replace_applicator]: 1.176e-05
    [Cycle 2]: 0.00032242, [27]
        [switch_simplify]: 5.56e-06
        [loop_unroll]: 4.99e-06
        [a_1]: 8.96e-05
        [with_stream_mark]: 9.16998e-06
        [recompute_prepare]: 5.14e-06
        [updatestate_depend_eliminate]: 2.34001e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.82999e-06
        [parameter_eliminate]: 9.20001e-07
        [specialize_transform]: 4.70999e-06
        [updatestate_useless_node_eliminater]: 5.10999e-06
        [accelerated_algorithm]: 4.72998e-06
        [meta_shard_fg_expand]: 1.18001e-06
        [get_grad_eliminate_]: 4.41002e-06
        [merge_forward]: 2.31e-06
        [cell_reuse_recompute_pass]: 1.96e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.232e-05
        [j_node_and_user_rematch]: 7.51999e-06
        [meta_fg_expand]: 1.50999e-06
        [replace_old_param]: 8.05e-06
        [inline_without_move]: 4.75001e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.19e-06
        [auto_monad_grad]: 9.20001e-07
        [auto_monad_eliminator]: 5.05001e-06
        [cse]: 1.022e-05
        [replace_applicator]: 5.27001e-06
[py_interpret_to_execute_after_opt_a]: 8.22e-06
[rewriter_after_opt_a]: 2.708e-05
[convert_after_rewriter]: 7.01999e-06
[order_py_execute_after_rewriter]: 4.48999e-06
[mutable_eliminate]: 0.00062668
[jit_opt_b]: 5.087e-05, [1]
    [Cycle 1]: 4.382e-05, [2]
        [frontend_op_eliminate]: 1.663e-05
        [inline_after_opt_a]: 1.46e-05
[cconv]: 1.746e-05
[loop_unroll]: 0.00043209
[jit_opt_after_cconv]: 0.00013814, [1]
    [Cycle 1]: 0.00013203, [11]
        [c_1]: 2.047e-05
        [parameter_eliminate]: 2.47001e-06
        [updatestate_depend_eliminate]: 5.88002e-06
        [updatestate_assign_eliminate]: 2.91e-06
        [updatestate_loads_eliminate]: 2.03002e-06
        [cse]: 1.905e-05
        [call_graph_tuple_transform]: 1.757e-05
        [tuple_list_get_item_eliminator]: 5.49e-06
        [none_parameter_eliminate]: 1.00001e-06
        [renormalize]: 6.40022e-07
        [switch_simplify]: 5.12e-06
[remove_dup_value]: 7.8e-06
[partial_unused_args_eliminate]: 1.64e-06
[environ_conv]: 3.38e-06
[add_recomputation]: 3.365e-05
[cse_after_recomputation]: 2.104e-05, [1]
    [Cycle 1]: 1.536e-05, [1]
        [cse]: 9.33002e-06
[auto_monad_reorder]: 9.32999e-06
[get_jit_bprop_graph]: 1.42999e-06
[rewriter_after_jit_bprop_graph]: 5.89e-06
[opt_after_jit_grad]: 0.00047644
[symbol_engine_optimizer]: 7.627e-05, [1]
    [Cycle 1]: 6.9e-05, [6]
        [build]: 3.95998e-06
        [elim_shapecalc]: 8.28001e-06
        [elim_not_effective]: 1.328e-05
        [opt_reshape]: 5.89999e-06
        [fold_const_symbol]: 9.09e-06
        [renormalize]: 5.69999e-07
[validate]: 3.558e-05
Sums
    bootstrap                            :     0.000319s :  3.66%
    type_inference                       :     0.004586s : 52.65%
    event_method                         :     0.000018s :  0.21%
    auto_monad                           :     0.000068s :  0.78%
    graph_reusing                        :     0.000007s :  0.08%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.42%
    rewriter_before_opt_a                :     0.000074s :  0.85%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000063s :  0.72%
    jit_opt_a.loop_unroll                :     0.000036s :  0.41%
    jit_opt_a.a_1                        :     0.000643s :  7.38%
    jit_opt_a.with_stream_mark           :     0.000029s :  0.33%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000018s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000739s :  8.48%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.17%
    jit_opt_a.cse                        :     0.000027s :  0.31%
    jit_opt_a.replace_applicator         :     0.000017s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.09%
    rewriter_after_opt_a                 :     0.000027s :  0.31%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000627s :  7.19%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000017s :  0.20%
    loop_unroll                          :     0.000432s :  4.96%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000008s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000003s :  0.04%
    add_recomputation                    :     0.000034s :  0.39%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000009s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000476s :  5.47%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000036s :  0.41%


Time group info:
------[substitution.]   0.000225    23
 0.74% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000002s :      2: substitution.fold_const_symbol
 1.60% :     0.000004s :      3: substitution.graph_param_transform
85.76% :     0.000193s :      4: substitution.inline
 1.40% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.92% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.24% :     0.000005s :      2: substitution.replace_old_param
 5.67% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004517     2
79.87% :     0.003607s :      1: type_inference.infer
20.13% :     0.000909s :      1: type_inference.specialize
------[replace.]   0.000067     6
66.28% :     0.000044s :      4: replace.inline
33.72% :     0.000022s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000202     6
94.22% :     0.000191s :      4: match.inline
 5.78% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.26% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 1.80% :     0.000002s :     10: predicate.arithmetic_simplify
 1.24% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.06% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.24% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.40% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.24% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000001s :     10: predicate.environ_get_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.98% :     0.000001s :      6: predicate.get_grad_eliminate
 0.29% :     0.000000s :      3: predicate.graph_param_transform
 5.99% :     0.000007s :     22: predicate.inline
 0.99% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.01% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 2.10% :     0.000002s :     15: predicate.load_eliminater
 1.50% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.81% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.13% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 2.03% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.25% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.58% :     0.000002s :     10: predicate.reduce_eliminate
 1.45% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.55% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.68% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.27% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.78% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000001s :     10: predicate.same_eliminate
 0.49% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.00% :     0.000001s :      6: predicate.special_op_eliminate
 1.04% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.36% :     0.000003s :     16: predicate.switch_defer_inline
 2.09% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.81% :     0.000009s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.27% :     0.000001s :     10: predicate.transpose_eliminate
 1.46% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.26% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.68% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.85% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.88% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.52% :     0.000002s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.74% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000642    12
48.30% :     0.000310s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.83% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.86% :     0.000288s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014564    72
 0.25% :     0.000037s :      1: add_recomputation
 0.49% :     0.000071s :      1: auto_monad
 0.08% :     0.000012s :      1: auto_monad_reorder
 2.29% :     0.000334s :      1: bootstrap
 0.14% :     0.000021s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000023s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.16% :     0.000023s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
38.73% :     0.005641s :      1: jit_opt_a
 0.97% :     0.000141s :      1: jit_opt_after_cconv
 0.37% :     0.000053s :      1: jit_opt_b
 3.02% :     0.000440s :      1: loop_unroll
 4.37% :     0.000636s :      1: mutable_eliminate
 5.88% :     0.000856s :     26: opt.transform.jit_opt_a
 0.31% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.33% :     0.000485s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000039s :      1: py_interpret_to_execute
 0.07% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000010s :      1: remove_dup_value
 2.70% :     0.000393s :      1: renormalize.infer
 2.33% :     0.000339s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.21% :     0.000031s :      1: rewriter_after_opt_a
 0.53% :     0.000077s :      1: rewriter_before_opt_a
 0.54% :     0.000079s :      1: symbol_engine_optimizer
31.61% :     0.004604s :      1: type_inference


TotalTime = 0.01301, [30]
[bootstrap]: 0.00029675
[type_inference]: 0.00440763
[event_method]: 1.874e-05
[auto_monad]: 6.65e-05
[graph_reusing]: 6.36e-06
[pre_auto_parallel]: 2.61999e-06
[py_interpret_to_execute]: 3.607e-05
[rewriter_before_opt_a]: 7.251e-05
[expand_dump_flag]: 3.66001e-06
[jit_opt_a]: 0.00582207, [2]
    [Cycle 1]: 0.00183761, [27]
        [switch_simplify]: 5.917e-05
        [loop_unroll]: 2.924e-05
        [a_1]: 0.0005554
        [with_stream_mark]: 2.052e-05
        [recompute_prepare]: 8.03999e-06
        [updatestate_depend_eliminate]: 4.26001e-06
        [updatestate_assign_eliminate]: 3.25e-06
        [updatestate_loads_eliminate]: 2.66e-06
        [parameter_eliminate]: 1.86998e-06
        [specialize_transform]: 6.27001e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.77999e-06
        [meta_shard_fg_expand]: 1.91998e-06
        [get_grad_eliminate_]: 5.51e-06
        [merge_forward]: 3.71999e-06
        [cell_reuse_recompute_pass]: 1.37e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.456e-05
        [j_node_and_user_rematch]: 9.44e-06
        [meta_fg_expand]: 2.77002e-06
        [replace_old_param]: 8.62998e-06
        [inline_without_move]: 5.39998e-06
        [renormalize]: 0.00082584
        [add_forward_monad_depend]: 5.77999e-06
        [auto_monad_grad]: 2.91e-06
        [auto_monad_eliminator]: 1.455e-05
        [cse]: 2.949e-05
        [replace_applicator]: 1.457e-05
    [Cycle 2]: 0.00032112, [27]
        [switch_simplify]: 5.91003e-06
        [loop_unroll]: 5.39998e-06
        [a_1]: 9.162e-05
        [with_stream_mark]: 9.77001e-06
        [recompute_prepare]: 5.54e-06
        [updatestate_depend_eliminate]: 2.43e-06
        [updatestate_assign_eliminate]: 2.17001e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 4.72e-06
        [updatestate_useless_node_eliminater]: 4.87e-06
        [accelerated_algorithm]: 5.34e-06
        [meta_shard_fg_expand]: 1.51998e-06
        [get_grad_eliminate_]: 4.33001e-06
        [merge_forward]: 2.69001e-06
        [cell_reuse_recompute_pass]: 1.49998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.341e-05
        [j_node_and_user_rematch]: 7.75998e-06
        [meta_fg_expand]: 1.82999e-06
        [replace_old_param]: 7.82e-06
        [inline_without_move]: 4.79e-06
        [renormalize]: 6.00121e-08
        [add_forward_monad_depend]: 1.44e-06
        [auto_monad_grad]: 9.5999e-07
        [auto_monad_eliminator]: 4.87e-06
        [cse]: 1.069e-05
        [replace_applicator]: 4.73001e-06
[py_interpret_to_execute_after_opt_a]: 1.053e-05
[rewriter_after_opt_a]: 3.547e-05
[convert_after_rewriter]: 6.81999e-06
[order_py_execute_after_rewriter]: 4.98001e-06
[mutable_eliminate]: 0.00064902
[jit_opt_b]: 5.427e-05, [1]
    [Cycle 1]: 4.658e-05, [2]
        [frontend_op_eliminate]: 1.949e-05
        [inline_after_opt_a]: 1.563e-05
[cconv]: 2.727e-05
[loop_unroll]: 0.0004436
[jit_opt_after_cconv]: 0.00014315, [1]
    [Cycle 1]: 0.00013625, [11]
        [c_1]: 2.082e-05
        [parameter_eliminate]: 3.00998e-06
        [updatestate_depend_eliminate]: 5.82999e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [cse]: 2.041e-05
        [call_graph_tuple_transform]: 1.931e-05
        [tuple_list_get_item_eliminator]: 5.29e-06
        [none_parameter_eliminate]: 1.62001e-06
        [renormalize]: 6.30011e-07
        [switch_simplify]: 5.37001e-06
[remove_dup_value]: 1.504e-05
[partial_unused_args_eliminate]: 2.28998e-06
[environ_conv]: 5.49998e-06
[add_recomputation]: 5.185e-05
[cse_after_recomputation]: 2.131e-05, [1]
    [Cycle 1]: 1.487e-05, [1]
        [cse]: 8.87e-06
[auto_monad_reorder]: 1.576e-05
[get_jit_bprop_graph]: 1.94999e-06
[rewriter_after_jit_bprop_graph]: 6.89999e-06
[opt_after_jit_grad]: 0.00048092
[symbol_engine_optimizer]: 7.651e-05, [1]
    [Cycle 1]: 6.943e-05, [6]
        [build]: 4e-06
        [elim_shapecalc]: 7.87998e-06
        [elim_not_effective]: 1.395e-05
        [opt_reshape]: 5.54998e-06
        [fold_const_symbol]: 9.00999e-06
        [renormalize]: 7.59988e-07
[validate]: 3.814e-05
Sums
    bootstrap                            :     0.000297s :  3.40%
    type_inference                       :     0.004408s : 50.46%
    event_method                         :     0.000019s :  0.21%
    auto_monad                           :     0.000066s :  0.76%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.41%
    rewriter_before_opt_a                :     0.000073s :  0.83%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000065s :  0.74%
    jit_opt_a.loop_unroll                :     0.000035s :  0.40%
    jit_opt_a.a_1                        :     0.000647s :  7.41%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000826s :  9.45%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.22%
    jit_opt_a.cse                        :     0.000040s :  0.46%
    jit_opt_a.replace_applicator         :     0.000019s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000035s :  0.41%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000649s :  7.43%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.22%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.18%
    cconv                                :     0.000027s :  0.31%
    loop_unroll                          :     0.000444s :  5.08%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000020s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000052s :  0.59%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.08%
    opt_after_jit_grad                   :     0.000481s :  5.51%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000038s :  0.44%


Time group info:
------[substitution.]   0.000218    23
 1.11% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.63% :     0.000006s :      3: substitution.graph_param_transform
84.20% :     0.000184s :      4: substitution.inline
 1.45% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.23% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000004s :      2: substitution.replace_old_param
 5.73% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004337     2
79.82% :     0.003462s :      1: type_inference.infer
20.18% :     0.000875s :      1: type_inference.specialize
------[replace.]   0.000073     6
67.65% :     0.000049s :      4: replace.inline
32.35% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000193     6
94.13% :     0.000181s :      4: match.inline
 5.87% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.27% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.32% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.08% :     0.000001s :     10: predicate.addn_check_dump
 1.50% :     0.000002s :     10: predicate.addn_zero_filter
 1.93% :     0.000002s :     10: predicate.arithmetic_simplify
 1.38% :     0.000002s :     10: predicate.cast_eliminate
 0.55% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.22% :     0.000001s :     10: predicate.depend_value_elim
 1.27% :     0.000002s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.29% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.95% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.33% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.91% :     0.000001s :      6: predicate.get_grad_eliminate
 0.32% :     0.000000s :      3: predicate.graph_param_transform
 5.40% :     0.000006s :     22: predicate.inline
 1.01% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.22% :     0.000001s :      6: predicate.less_batch_normalization
 1.56% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.31% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.90% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.95% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.03% :     0.000001s :     10: predicate.minmaximum_grad
 1.92% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.30% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.59% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.49% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.69% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.23% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.05% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.47% :     0.000002s :      6: predicate.special_op_eliminate
 1.07% :     0.000001s :      6: predicate.specialize_transform
 1.32% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.67% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.34% :     0.000003s :     16: predicate.switch_defer_inline
 2.03% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.59% :     0.000009s :     53: predicate.switch_simplify
 1.23% :     0.000001s :     10: predicate.tile_eliminate
 1.17% :     0.000001s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.12% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.22% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.41% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.80% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.59% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.83% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.47% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.12% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000621    12
49.66% :     0.000308s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.27% :     0.000033s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.07% :     0.000280s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014747    72
 0.37% :     0.000055s :      1: add_recomputation
 0.47% :     0.000070s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.10% :     0.000310s :      1: bootstrap
 0.20% :     0.000030s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000023s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
39.51% :     0.005826s :      1: jit_opt_a
 0.99% :     0.000146s :      1: jit_opt_after_cconv
 0.39% :     0.000057s :      1: jit_opt_b
 3.07% :     0.000452s :      1: loop_unroll
 4.46% :     0.000658s :      1: mutable_eliminate
 5.86% :     0.000864s :     26: opt.transform.jit_opt_a
 0.32% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000028s :      4: opt.transform.jit_opt_b
 0.09% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.32% :     0.000489s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000039s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.82% :     0.000417s :      1: renormalize.infer
 2.72% :     0.000402s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000038s :      1: rewriter_after_opt_a
 0.51% :     0.000076s :      1: rewriter_before_opt_a
 0.54% :     0.000079s :      1: symbol_engine_optimizer
30.00% :     0.004424s :      1: type_inference


TotalTime = 0.0123958, [30]
[bootstrap]: 0.0003262
[type_inference]: 0.00450759
[event_method]: 1.716e-05
[auto_monad]: 6.279e-05
[graph_reusing]: 5.52999e-06
[pre_auto_parallel]: 2.90002e-06
[py_interpret_to_execute]: 3.18e-05
[rewriter_before_opt_a]: 7.155e-05
[expand_dump_flag]: 2.88e-06
[jit_opt_a]: 0.00530082, [2]
    [Cycle 1]: 0.00161414, [27]
        [switch_simplify]: 5.53e-05
        [loop_unroll]: 2.929e-05
        [a_1]: 0.00051927
        [with_stream_mark]: 1.651e-05
        [recompute_prepare]: 7.49002e-06
        [updatestate_depend_eliminate]: 3.56999e-06
        [updatestate_assign_eliminate]: 3.21001e-06
        [updatestate_loads_eliminate]: 2.39001e-06
        [parameter_eliminate]: 1.81e-06
        [specialize_transform]: 6.06e-06
        [updatestate_useless_node_eliminater]: 5.09998e-06
        [accelerated_algorithm]: 5.68997e-06
        [meta_shard_fg_expand]: 1.90001e-06
        [get_grad_eliminate_]: 5.22999e-06
        [merge_forward]: 3.78999e-06
        [cell_reuse_recompute_pass]: 1.24998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.504e-05
        [j_node_and_user_rematch]: 9.94999e-06
        [meta_fg_expand]: 2.16998e-06
        [replace_old_param]: 9.66998e-06
        [inline_without_move]: 5.17999e-06
        [renormalize]: 0.00067084
        [add_forward_monad_depend]: 5.11002e-06
        [auto_monad_grad]: 1.95001e-06
        [auto_monad_eliminator]: 1.373e-05
        [cse]: 2.797e-05
        [replace_applicator]: 1.122e-05
    [Cycle 2]: 0.00031827, [27]
        [switch_simplify]: 5.55001e-06
        [loop_unroll]: 4.79e-06
        [a_1]: 8.923e-05
        [with_stream_mark]: 8.32e-06
        [recompute_prepare]: 5.17999e-06
        [updatestate_depend_eliminate]: 2.48998e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 2.01998e-06
        [parameter_eliminate]: 1.07998e-06
        [specialize_transform]: 4.87e-06
        [updatestate_useless_node_eliminater]: 5.05999e-06
        [accelerated_algorithm]: 5.15999e-06
        [meta_shard_fg_expand]: 1.20001e-06
        [get_grad_eliminate_]: 4.41002e-06
        [merge_forward]: 2.26e-06
        [cell_reuse_recompute_pass]: 1.71998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.22e-05
        [j_node_and_user_rematch]: 7.98001e-06
        [meta_fg_expand]: 1.42999e-06
        [replace_old_param]: 6.88e-06
        [inline_without_move]: 4.66002e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.34e-06
        [auto_monad_grad]: 9.5999e-07
        [auto_monad_eliminator]: 4.87998e-06
        [cse]: 1.034e-05
        [replace_applicator]: 5.03002e-06
[py_interpret_to_execute_after_opt_a]: 8.22998e-06
[rewriter_after_opt_a]: 3.281e-05
[convert_after_rewriter]: 6.79001e-06
[order_py_execute_after_rewriter]: 4.74e-06
[mutable_eliminate]: 0.00053816
[jit_opt_b]: 4.95e-05, [1]
    [Cycle 1]: 4.269e-05, [2]
        [frontend_op_eliminate]: 1.619e-05
        [inline_after_opt_a]: 1.493e-05
[cconv]: 2.489e-05
[loop_unroll]: 0.00041887
[jit_opt_after_cconv]: 0.00013781, [1]
    [Cycle 1]: 0.0001316, [11]
        [c_1]: 2.064e-05
        [parameter_eliminate]: 2.32001e-06
        [updatestate_depend_eliminate]: 5.53002e-06
        [updatestate_assign_eliminate]: 2.37999e-06
        [updatestate_loads_eliminate]: 2.22999e-06
        [cse]: 1.787e-05
        [call_graph_tuple_transform]: 1.859e-05
        [tuple_list_get_item_eliminator]: 5.46e-06
        [none_parameter_eliminate]: 1.39998e-06
        [renormalize]: 4.89992e-07
        [switch_simplify]: 5.37001e-06
[remove_dup_value]: 1.457e-05
[partial_unused_args_eliminate]: 2.44001e-06
[environ_conv]: 5.57001e-06
[add_recomputation]: 4.689e-05
[cse_after_recomputation]: 2.054e-05, [1]
    [Cycle 1]: 1.464e-05, [1]
        [cse]: 8.95001e-06
[auto_monad_reorder]: 1.541e-05
[get_jit_bprop_graph]: 1.56998e-06
[rewriter_after_jit_bprop_graph]: 5.58997e-06
[opt_after_jit_grad]: 0.0004516
[symbol_engine_optimizer]: 7.236e-05, [1]
    [Cycle 1]: 6.578e-05, [6]
        [build]: 3.6e-06
        [elim_shapecalc]: 7.25998e-06
        [elim_not_effective]: 1.227e-05
        [opt_reshape]: 5.62001e-06
        [fold_const_symbol]: 8.70999e-06
        [renormalize]: 5.8001e-07
[validate]: 3.645e-05
Sums
    bootstrap                            :     0.000326s :  3.86%
    type_inference                       :     0.004508s : 53.37%
    event_method                         :     0.000017s :  0.20%
    auto_monad                           :     0.000063s :  0.74%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000032s :  0.38%
    rewriter_before_opt_a                :     0.000072s :  0.85%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000061s :  0.72%
    jit_opt_a.loop_unroll                :     0.000034s :  0.40%
    jit_opt_a.a_1                        :     0.000609s :  7.21%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.29%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000027s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000017s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000671s :  7.94%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.22%
    jit_opt_a.cse                        :     0.000038s :  0.45%
    jit_opt_a.replace_applicator         :     0.000016s :  0.19%
    py_interpret_to_execute_after_opt_a  :     0.000008s :  0.10%
    rewriter_after_opt_a                 :     0.000033s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000538s :  6.37%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000025s :  0.29%
    loop_unroll                          :     0.000419s :  4.96%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000018s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000047s :  0.56%
    cse_after_recomputation.cse          :     0.000009s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000452s :  5.35%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000036s :  0.43%


Time group info:
------[substitution.]   0.000192    23
 1.00% :     0.000002s :      2: substitution.elim_not_effective
 0.72% :     0.000001s :      2: substitution.fold_const_symbol
 2.95% :     0.000006s :      3: substitution.graph_param_transform
81.87% :     0.000157s :      4: substitution.inline
 1.85% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.43% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.44% :     0.000005s :      2: substitution.replace_old_param
 6.73% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004448     2
81.14% :     0.003609s :      1: type_inference.infer
18.86% :     0.000839s :      1: type_inference.specialize
------[replace.]   0.000068     6
66.76% :     0.000045s :      4: replace.inline
33.24% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000166     6
92.88% :     0.000154s :      4: match.inline
 7.12% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.38% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.07% :     0.000001s :     10: predicate.addn_check_dump
 1.52% :     0.000002s :     10: predicate.addn_zero_filter
 1.89% :     0.000002s :     10: predicate.arithmetic_simplify
 1.32% :     0.000002s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.08% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000002s :     10: predicate.depend_value_elim
 1.16% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.29% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.89% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.36% :     0.000000s :      3: predicate.elim_not_effective
 0.63% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.25% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.19% :     0.000001s :     10: predicate.environ_get_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.25% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.94% :     0.000007s :     22: predicate.inline
 0.96% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.16% :     0.000001s :      6: predicate.less_batch_normalization
 1.64% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.71% :     0.000002s :     15: predicate.load_eliminater
 1.64% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.12% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.97% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.71% :     0.000002s :      3: predicate.mutable_eliminate
 0.53% :     0.000001s :      3: predicate.opt_reshape
 2.34% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.51% :     0.000002s :     10: predicate.reduce_eliminate
 1.59% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.66% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.24% :     0.000000s :      3: predicate.reset_defer_inline
 1.37% :     0.000002s :     10: predicate.reshape_eliminate
 1.28% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.73% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.56% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.88% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.29% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.49% :     0.000003s :     16: predicate.switch_defer_inline
 2.26% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.71% :     0.000009s :     53: predicate.switch_simplify
 1.16% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.39% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.25% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.71% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.71% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.87% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.78% :     0.000002s :     10: predicate.value_based_eliminate
 0.35% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000620    12
52.04% :     0.000323s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.97% :     0.000031s :      1: func_graph_cloner_run.FuncGraphClonerNode
42.99% :     0.000267s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013929    72
 0.36% :     0.000050s :      1: add_recomputation
 0.48% :     0.000066s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 2.49% :     0.000346s :      1: bootstrap
 0.20% :     0.000027s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.16% :     0.000023s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.15% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.08% :     0.005304s :      1: jit_opt_a
 1.01% :     0.000141s :      1: jit_opt_after_cconv
 0.37% :     0.000052s :      1: jit_opt_b
 3.06% :     0.000426s :      1: loop_unroll
 3.92% :     0.000546s :      1: mutable_eliminate
 5.87% :     0.000817s :     26: opt.transform.jit_opt_a
 0.34% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000024s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000031s :      4: opt.transform.symbol_engine_opt
 3.29% :     0.000458s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.25% :     0.000035s :      1: py_interpret_to_execute
 0.08% :     0.000010s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.62% :     0.000364s :      1: renormalize.infer
 2.15% :     0.000299s :      1: renormalize.specialize
 0.06% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000036s :      1: rewriter_after_opt_a
 0.53% :     0.000074s :      1: rewriter_before_opt_a
 0.54% :     0.000075s :      1: symbol_engine_optimizer
32.45% :     0.004519s :      1: type_inference


TotalTime = 0.0129911, [30]
[bootstrap]: 0.00034878
[type_inference]: 0.00473836
[event_method]: 1.916e-05
[auto_monad]: 6.632e-05
[graph_reusing]: 5.72999e-06
[pre_auto_parallel]: 2.46e-06
[py_interpret_to_execute]: 3.637e-05
[rewriter_before_opt_a]: 7.276e-05
[expand_dump_flag]: 3.48e-06
[jit_opt_a]: 0.00569182, [2]
    [Cycle 1]: 0.00170321, [27]
        [switch_simplify]: 5.625e-05
        [loop_unroll]: 2.893e-05
        [a_1]: 0.00053037
        [with_stream_mark]: 1.867e-05
        [recompute_prepare]: 6.64999e-06
        [updatestate_depend_eliminate]: 3.63e-06
        [updatestate_assign_eliminate]: 2.86999e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 2.01e-06
        [specialize_transform]: 5.74999e-06
        [updatestate_useless_node_eliminater]: 5.03002e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 2.48002e-06
        [get_grad_eliminate_]: 4.93001e-06
        [merge_forward]: 3.51999e-06
        [cell_reuse_recompute_pass]: 1.14998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.399e-05
        [j_node_and_user_rematch]: 9.10999e-06
        [meta_fg_expand]: 2.08002e-06
        [replace_old_param]: 8.97999e-06
        [inline_without_move]: 5.46e-06
        [renormalize]: 0.00076147
        [add_forward_monad_depend]: 4.79002e-06
        [auto_monad_grad]: 2.67001e-06
        [auto_monad_eliminator]: 1.211e-05
        [cse]: 2.722e-05
        [replace_applicator]: 1.113e-05
    [Cycle 2]: 0.00031249, [27]
        [switch_simplify]: 5.51e-06
        [loop_unroll]: 4.95001e-06
        [a_1]: 8.792e-05
        [with_stream_mark]: 8.48999e-06
        [recompute_prepare]: 4.92e-06
        [updatestate_depend_eliminate]: 2.43002e-06
        [updatestate_assign_eliminate]: 1.96998e-06
        [updatestate_loads_eliminate]: 1.86e-06
        [parameter_eliminate]: 9.20001e-07
        [specialize_transform]: 5.15999e-06
        [updatestate_useless_node_eliminater]: 5.05001e-06
        [accelerated_algorithm]: 5.29998e-06
        [meta_shard_fg_expand]: 1.12999e-06
        [get_grad_eliminate_]: 4.57e-06
        [merge_forward]: 2.45002e-06
        [cell_reuse_recompute_pass]: 1.34e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.155e-05
        [j_node_and_user_rematch]: 7.75e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 7.33999e-06
        [inline_without_move]: 4.55001e-06
        [renormalize]: 5.00004e-08
        [add_forward_monad_depend]: 1.24e-06
        [auto_monad_grad]: 7.30011e-07
        [auto_monad_eliminator]: 4.38999e-06
        [cse]: 1.031e-05
        [replace_applicator]: 4.92e-06
[py_interpret_to_execute_after_opt_a]: 9.91003e-06
[rewriter_after_opt_a]: 3.182e-05
[convert_after_rewriter]: 6.64999e-06
[order_py_execute_after_rewriter]: 7.53e-06
[mutable_eliminate]: 0.00049689
[jit_opt_b]: 4.957e-05, [1]
    [Cycle 1]: 4.349e-05, [2]
        [frontend_op_eliminate]: 1.699e-05
        [inline_after_opt_a]: 1.514e-05
[cconv]: 2.402e-05
[loop_unroll]: 0.00041726
[jit_opt_after_cconv]: 0.00013222, [1]
    [Cycle 1]: 0.00012574, [11]
        [c_1]: 1.988e-05
        [parameter_eliminate]: 2.19001e-06
        [updatestate_depend_eliminate]: 5.07e-06
        [updatestate_assign_eliminate]: 2.32999e-06
        [updatestate_loads_eliminate]: 1.97001e-06
        [cse]: 1.688e-05
        [call_graph_tuple_transform]: 1.76e-05
        [tuple_list_get_item_eliminator]: 4.98001e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 4.30009e-07
        [switch_simplify]: 4.97e-06
[remove_dup_value]: 1.41e-05
[partial_unused_args_eliminate]: 2.16e-06
[environ_conv]: 4.77e-06
[add_recomputation]: 4.336e-05
[cse_after_recomputation]: 1.986e-05, [1]
    [Cycle 1]: 1.46e-05, [1]
        [cse]: 8.80999e-06
[auto_monad_reorder]: 1.443e-05
[get_jit_bprop_graph]: 2.14999e-06
[rewriter_after_jit_bprop_graph]: 5.39998e-06
[opt_after_jit_grad]: 0.00044698
[symbol_engine_optimizer]: 7.058e-05, [1]
    [Cycle 1]: 6.385e-05, [6]
        [build]: 3.45e-06
        [elim_shapecalc]: 7.40998e-06
        [elim_not_effective]: 1.19e-05
        [opt_reshape]: 5.05999e-06
        [fold_const_symbol]: 8.37e-06
        [renormalize]: 4.59986e-07
[validate]: 3.359e-05
Sums
    bootstrap                            :     0.000349s :  3.99%
    type_inference                       :     0.004738s : 54.17%
    event_method                         :     0.000019s :  0.22%
    auto_monad                           :     0.000066s :  0.76%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000036s :  0.42%
    rewriter_before_opt_a                :     0.000073s :  0.83%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000062s :  0.71%
    jit_opt_a.loop_unroll                :     0.000034s :  0.39%
    jit_opt_a.a_1                        :     0.000618s :  7.07%
    jit_opt_a.with_stream_mark           :     0.000027s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.13%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000762s :  8.71%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.19%
    jit_opt_a.cse                        :     0.000038s :  0.43%
    jit_opt_a.replace_applicator         :     0.000016s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.11%
    rewriter_after_opt_a                 :     0.000032s :  0.36%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000008s :  0.09%
    mutable_eliminate                    :     0.000497s :  5.68%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000024s :  0.27%
    loop_unroll                          :     0.000417s :  4.77%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000017s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.20%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000043s :  0.50%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000014s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000447s :  5.11%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000034s :  0.38%


Time group info:
------[substitution.]   0.000207    23
 0.87% :     0.000002s :      2: substitution.elim_not_effective
 0.73% :     0.000002s :      2: substitution.fold_const_symbol
 2.32% :     0.000005s :      3: substitution.graph_param_transform
84.70% :     0.000176s :      4: substitution.inline
 1.51% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.12% :     0.000004s :      4: substitution.remove_not_recompute_node
 1.95% :     0.000004s :      2: substitution.replace_old_param
 5.80% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004666     2
80.91% :     0.003776s :      1: type_inference.infer
19.09% :     0.000891s :      1: type_inference.specialize
------[replace.]   0.000069     6
65.30% :     0.000045s :      4: replace.inline
34.70% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000184     6
94.02% :     0.000173s :      4: match.inline
 5.98% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000111   691
 1.20% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.30% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.11% :     0.000001s :     10: predicate.addn_check_dump
 1.39% :     0.000002s :     10: predicate.addn_zero_filter
 2.31% :     0.000003s :     10: predicate.arithmetic_simplify
 1.25% :     0.000001s :     10: predicate.cast_eliminate
 0.48% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.10% :     0.000001s :     10: predicate.compare_switch_simplify
 1.42% :     0.000002s :     10: predicate.depend_value_elim
 1.15% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.32% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.77% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.37% :     0.000000s :      3: predicate.elim_not_effective
 0.60% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.14% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.06% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.21% :     0.000001s :     10: predicate.environ_get_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.96% :     0.000001s :      6: predicate.get_grad_eliminate
 0.23% :     0.000000s :      3: predicate.graph_param_transform
 5.45% :     0.000006s :     22: predicate.inline
 0.90% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.06% :     0.000001s :      6: predicate.less_batch_normalization
 1.54% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.77% :     0.000002s :     15: predicate.load_eliminater
 1.46% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.18% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.92% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.36% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.63% :     0.000003s :     15: predicate.partial_eliminate
 1.22% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.67% :     0.000002s :     10: predicate.reduce_eliminate
 1.40% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.54% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.61% :     0.000001s :      6: predicate.replace_old_param
 0.31% :     0.000000s :      3: predicate.reset_defer_inline
 1.31% :     0.000001s :     10: predicate.reshape_eliminate
 1.24% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.91% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000001s :     10: predicate.same_eliminate
 0.57% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.30% :     0.000001s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.35% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.20% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.46% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.54% :     0.000003s :     16: predicate.switch_defer_inline
 2.08% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.91% :     0.000009s :     53: predicate.switch_simplify
 1.21% :     0.000001s :     10: predicate.tile_eliminate
 1.47% :     0.000002s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.40% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.11% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.73% :     0.000002s :     10: predicate.value_based_eliminate
 0.42% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000652    12
50.21% :     0.000327s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.79% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.00% :     0.000287s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014624    72
 0.31% :     0.000046s :      1: add_recomputation
 0.48% :     0.000070s :      1: auto_monad
 0.11% :     0.000017s :      1: auto_monad_reorder
 2.52% :     0.000369s :      1: bootstrap
 0.18% :     0.000027s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000007s :      1: environ_conv
 0.16% :     0.000024s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.95% :     0.005696s :      1: jit_opt_a
 0.92% :     0.000135s :      1: jit_opt_after_cconv
 0.36% :     0.000052s :      1: jit_opt_b
 2.90% :     0.000424s :      1: loop_unroll
 3.45% :     0.000504s :      1: mutable_eliminate
 5.64% :     0.000824s :     26: opt.transform.jit_opt_a
 0.30% :     0.000044s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.08% :     0.000012s :      1: opt.transform.mutable_eliminate
 0.13% :     0.000019s :      1: opt.transform.opt_after_jit_grad
 0.20% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.10% :     0.000454s :      1: opt_after_jit_grad
 0.07% :     0.000010s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000039s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000016s :      1: remove_dup_value
 2.92% :     0.000426s :      1: renormalize.infer
 2.24% :     0.000328s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.24% :     0.000035s :      1: rewriter_after_opt_a
 0.52% :     0.000076s :      1: rewriter_before_opt_a
 0.50% :     0.000073s :      1: symbol_engine_optimizer
32.52% :     0.004756s :      1: type_inference


TotalTime = 0.012299, [30]
[bootstrap]: 0.00029046
[type_inference]: 0.00429636
[event_method]: 1.699e-05
[auto_monad]: 6.628e-05
[graph_reusing]: 5.61998e-06
[pre_auto_parallel]: 2.96999e-06
[py_interpret_to_execute]: 3.13e-05
[rewriter_before_opt_a]: 7.371e-05
[expand_dump_flag]: 3.25998e-06
[jit_opt_a]: 0.00542538, [2]
    [Cycle 1]: 0.0017288, [27]
        [switch_simplify]: 5.669e-05
        [loop_unroll]: 2.826e-05
        [a_1]: 0.0005309
        [with_stream_mark]: 1.577e-05
        [recompute_prepare]: 7.31001e-06
        [updatestate_depend_eliminate]: 3.39001e-06
        [updatestate_assign_eliminate]: 3.16999e-06
        [updatestate_loads_eliminate]: 2.52001e-06
        [parameter_eliminate]: 2.04999e-06
        [specialize_transform]: 5.91e-06
        [updatestate_useless_node_eliminater]: 4.92e-06
        [accelerated_algorithm]: 5.51e-06
        [meta_shard_fg_expand]: 1.94999e-06
        [get_grad_eliminate_]: 5.29e-06
        [merge_forward]: 3.53e-06
        [cell_reuse_recompute_pass]: 1.06002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.454e-05
        [j_node_and_user_rematch]: 9.22999e-06
        [meta_fg_expand]: 2.16998e-06
        [replace_old_param]: 9.44998e-06
        [inline_without_move]: 5.72001e-06
        [renormalize]: 0.00077955
        [add_forward_monad_depend]: 5.49e-06
        [auto_monad_grad]: 2.53998e-06
        [auto_monad_eliminator]: 1.426e-05
        [cse]: 2.679e-05
        [replace_applicator]: 1.333e-05
    [Cycle 2]: 0.0003185, [27]
        [switch_simplify]: 5.52001e-06
        [loop_unroll]: 4.78001e-06
        [a_1]: 8.93e-05
        [with_stream_mark]: 9.22999e-06
        [recompute_prepare]: 4.97999e-06
        [updatestate_depend_eliminate]: 2.63e-06
        [updatestate_assign_eliminate]: 2.07999e-06
        [updatestate_loads_eliminate]: 1.80001e-06
        [parameter_eliminate]: 1.15999e-06
        [specialize_transform]: 4.90001e-06
        [updatestate_useless_node_eliminater]: 4.77998e-06
        [accelerated_algorithm]: 4.98001e-06
        [meta_shard_fg_expand]: 1.04e-06
        [get_grad_eliminate_]: 4.57e-06
        [merge_forward]: 2.41e-06
        [cell_reuse_recompute_pass]: 1.49998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.393e-05
        [j_node_and_user_rematch]: 8.22003e-06
        [meta_fg_expand]: 1.53002e-06
        [replace_old_param]: 9.92999e-06
        [inline_without_move]: 4.84003e-06
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 1.10001e-06
        [auto_monad_grad]: 9.30013e-07
        [auto_monad_eliminator]: 5.15999e-06
        [cse]: 1.046e-05
        [replace_applicator]: 4.70001e-06
[py_interpret_to_execute_after_opt_a]: 1.029e-05
[rewriter_after_opt_a]: 3.376e-05
[convert_after_rewriter]: 7.28e-06
[order_py_execute_after_rewriter]: 5.35001e-06
[mutable_eliminate]: 0.00056663
[jit_opt_b]: 4.975e-05, [1]
    [Cycle 1]: 4.265e-05, [2]
        [frontend_op_eliminate]: 1.644e-05
        [inline_after_opt_a]: 1.469e-05
[cconv]: 2.63e-05
[loop_unroll]: 0.00041389
[jit_opt_after_cconv]: 0.00013905, [1]
    [Cycle 1]: 0.00013266, [11]
        [c_1]: 2.099e-05
        [parameter_eliminate]: 2.64001e-06
        [updatestate_depend_eliminate]: 5.52999e-06
        [updatestate_assign_eliminate]: 2.22001e-06
        [updatestate_loads_eliminate]: 2.02001e-06
        [cse]: 1.853e-05
        [call_graph_tuple_transform]: 1.909e-05
        [tuple_list_get_item_eliminator]: 5.29e-06
        [none_parameter_eliminate]: 1.47001e-06
        [renormalize]: 6.00005e-07
        [switch_simplify]: 5.56e-06
[remove_dup_value]: 1.405e-05
[partial_unused_args_eliminate]: 2.44999e-06
[environ_conv]: 5.61003e-06
[add_recomputation]: 4.725e-05
[cse_after_recomputation]: 1.945e-05, [1]
    [Cycle 1]: 1.398e-05, [1]
        [cse]: 8.58001e-06
[auto_monad_reorder]: 1.497e-05
[get_jit_bprop_graph]: 2.03997e-06
[rewriter_after_jit_bprop_graph]: 5.15001e-06
[opt_after_jit_grad]: 0.00044332
[symbol_engine_optimizer]: 7.539e-05, [1]
    [Cycle 1]: 6.835e-05, [6]
        [build]: 4.07998e-06
        [elim_shapecalc]: 7.3e-06
        [elim_not_effective]: 1.295e-05
        [opt_reshape]: 6.67002e-06
        [fold_const_symbol]: 9.10001e-06
        [renormalize]: 6.50005e-07
[validate]: 3.51e-05
Sums
    bootstrap                            :     0.000290s :  3.48%
    type_inference                       :     0.004296s : 51.44%
    event_method                         :     0.000017s :  0.20%
    auto_monad                           :     0.000066s :  0.79%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.04%
    py_interpret_to_execute              :     0.000031s :  0.37%
    rewriter_before_opt_a                :     0.000074s :  0.88%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000062s :  0.74%
    jit_opt_a.loop_unroll                :     0.000033s :  0.40%
    jit_opt_a.a_1                        :     0.000620s :  7.43%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.30%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.34%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.04%
    jit_opt_a.replace_old_param          :     0.000019s :  0.23%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000780s :  9.33%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.23%
    jit_opt_a.cse                        :     0.000037s :  0.45%
    jit_opt_a.replace_applicator         :     0.000018s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000034s :  0.40%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000567s :  6.78%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000026s :  0.31%
    loop_unroll                          :     0.000414s :  4.96%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000019s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000047s :  0.57%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000443s :  5.31%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.08%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000035s :  0.42%


Time group info:
------[substitution.]   0.000210    23
 1.05% :     0.000002s :      2: substitution.elim_not_effective
 0.60% :     0.000001s :      2: substitution.fold_const_symbol
 2.76% :     0.000006s :      3: substitution.graph_param_transform
83.19% :     0.000175s :      4: substitution.inline
 1.55% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.29% :     0.000005s :      4: substitution.remove_not_recompute_node
 3.27% :     0.000007s :      2: substitution.replace_old_param
 5.29% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004241     2
80.33% :     0.003407s :      1: type_inference.infer
19.67% :     0.000834s :      1: type_inference.specialize
------[replace.]   0.000071     6
66.99% :     0.000047s :      4: replace.inline
33.01% :     0.000023s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000183     6
94.54% :     0.000173s :      4: match.inline
 5.46% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000113   691
 1.33% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.19% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.32% :     0.000001s :     10: predicate.addn_zero_filter
 2.07% :     0.000002s :     10: predicate.arithmetic_simplify
 1.30% :     0.000001s :     10: predicate.cast_eliminate
 0.47% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.09% :     0.000001s :     10: predicate.compare_switch_simplify
 1.56% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.56% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.20% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.79% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.30% :     0.000000s :      3: predicate.elim_not_effective
 0.56% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.17% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.12% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.17% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.94% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.54% :     0.000006s :     22: predicate.inline
 0.94% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.11% :     0.000001s :      6: predicate.less_batch_normalization
 1.50% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.84% :     0.000002s :     15: predicate.load_eliminater
 1.42% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.96% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.11% :     0.000001s :     10: predicate.merge_addn
 1.08% :     0.000001s :     10: predicate.minmaximum_grad
 1.56% :     0.000002s :      3: predicate.mutable_eliminate
 0.57% :     0.000001s :      3: predicate.opt_reshape
 2.50% :     0.000003s :     15: predicate.partial_eliminate
 1.23% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.47% :     0.000002s :     10: predicate.reduce_eliminate
 1.62% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.95% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.71% :     0.000002s :     18: predicate.replace_applicator
 0.83% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.39% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.86% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.24% :     0.000001s :     10: predicate.same_eliminate
 0.50% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.06% :     0.000001s :      6: predicate.specialize_transform
 1.66% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.17% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.43% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.39% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.57% :     0.000009s :     53: predicate.switch_simplify
 1.22% :     0.000001s :     10: predicate.tile_eliminate
 1.15% :     0.000001s :     10: predicate.transpose_eliminate
 1.65% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.17% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.21% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.49% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.59% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.85% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.54% :     0.000002s :     10: predicate.value_based_eliminate
 0.34% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.79% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000620    12
50.88% :     0.000316s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.44% :     0.000034s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.68% :     0.000271s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013962    72
 0.36% :     0.000050s :      1: add_recomputation
 0.50% :     0.000070s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 2.20% :     0.000307s :      1: bootstrap
 0.21% :     0.000029s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.15% :     0.000021s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
38.88% :     0.005428s :      1: jit_opt_a
 1.02% :     0.000142s :      1: jit_opt_after_cconv
 0.38% :     0.000053s :      1: jit_opt_b
 3.02% :     0.000421s :      1: loop_unroll
 4.11% :     0.000574s :      1: mutable_eliminate
 5.96% :     0.000832s :     26: opt.transform.jit_opt_a
 0.34% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000025s :      4: opt.transform.jit_opt_b
 0.08% :     0.000011s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000020s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.23% :     0.000451s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.24% :     0.000034s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.59% :     0.000362s :      1: renormalize.infer
 2.93% :     0.000410s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000037s :      1: rewriter_after_opt_a
 0.56% :     0.000078s :      1: rewriter_before_opt_a
 0.56% :     0.000078s :      1: symbol_engine_optimizer
30.85% :     0.004307s :      1: type_inference


TotalTime = 0.0139431, [30]
[bootstrap]: 0.00028398
[type_inference]: 0.00445545
[event_method]: 1.832e-05
[auto_monad]: 7.104e-05
[graph_reusing]: 6.86001e-06
[pre_auto_parallel]: 2.75002e-06
[py_interpret_to_execute]: 4.3e-05
[rewriter_before_opt_a]: 8.86e-05
[expand_dump_flag]: 3.87998e-06
[jit_opt_a]: 0.00661395, [2]
    [Cycle 1]: 0.00199485, [27]
        [switch_simplify]: 6.236e-05
        [loop_unroll]: 3.15e-05
        [a_1]: 0.00059548
        [with_stream_mark]: 2.144e-05
        [recompute_prepare]: 9.82999e-06
        [updatestate_depend_eliminate]: 3.5e-06
        [updatestate_assign_eliminate]: 3.2e-06
        [updatestate_loads_eliminate]: 2.56e-06
        [parameter_eliminate]: 2.46e-06
        [specialize_transform]: 6.21e-06
        [updatestate_useless_node_eliminater]: 4.94998e-06
        [accelerated_algorithm]: 5.89999e-06
        [meta_shard_fg_expand]: 2.68e-06
        [get_grad_eliminate_]: 6.07999e-06
        [merge_forward]: 3.35e-06
        [cell_reuse_recompute_pass]: 1.45999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.655e-05
        [j_node_and_user_rematch]: 1.026e-05
        [meta_fg_expand]: 2.71e-06
        [replace_old_param]: 1.03e-05
        [inline_without_move]: 5.82001e-06
        [renormalize]: 0.00091689
        [add_forward_monad_depend]: 7.53999e-06
        [auto_monad_grad]: 2.89999e-06
        [auto_monad_eliminator]: 1.595e-05
        [cse]: 3.055e-05
        [replace_applicator]: 1.612e-05
    [Cycle 2]: 0.00034583, [27]
        [switch_simplify]: 5.54e-06
        [loop_unroll]: 4.97e-06
        [a_1]: 9.489e-05
        [with_stream_mark]: 1.293e-05
        [recompute_prepare]: 5.40001e-06
        [updatestate_depend_eliminate]: 2.91999e-06
        [updatestate_assign_eliminate]: 2.36e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [parameter_eliminate]: 1.27999e-06
        [specialize_transform]: 4.89998e-06
        [updatestate_useless_node_eliminater]: 4.94e-06
        [accelerated_algorithm]: 5.40001e-06
        [meta_shard_fg_expand]: 1.66e-06
        [get_grad_eliminate_]: 4.62e-06
        [merge_forward]: 2.79001e-06
        [cell_reuse_recompute_pass]: 2.36998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.514e-05
        [j_node_and_user_rematch]: 7.9e-06
        [meta_fg_expand]: 1.37999e-06
        [replace_old_param]: 8.38999e-06
        [inline_without_move]: 4.99e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.22e-06
        [auto_monad_grad]: 1.00999e-06
        [auto_monad_eliminator]: 5.22e-06
        [cse]: 1.107e-05
        [replace_applicator]: 5.39e-06
[py_interpret_to_execute_after_opt_a]: 1.366e-05
[rewriter_after_opt_a]: 3.711e-05
[convert_after_rewriter]: 8.69e-06
[order_py_execute_after_rewriter]: 5.46e-06
[mutable_eliminate]: 0.00068981
[jit_opt_b]: 5.336e-05, [1]
    [Cycle 1]: 4.505e-05, [2]
        [frontend_op_eliminate]: 1.764e-05
        [inline_after_opt_a]: 1.537e-05
[cconv]: 2.745e-05
[loop_unroll]: 0.0004382
[jit_opt_after_cconv]: 0.00014598, [1]
    [Cycle 1]: 0.00013955, [11]
        [c_1]: 2.061e-05
        [parameter_eliminate]: 3.21001e-06
        [updatestate_depend_eliminate]: 6.94001e-06
        [updatestate_assign_eliminate]: 2.74999e-06
        [updatestate_loads_eliminate]: 2.54999e-06
        [cse]: 2.065e-05
        [call_graph_tuple_transform]: 1.89e-05
        [tuple_list_get_item_eliminator]: 5.35999e-06
        [none_parameter_eliminate]: 1.66002e-06
        [renormalize]: 8.00006e-07
        [switch_simplify]: 5.37001e-06
[remove_dup_value]: 1.45e-05
[partial_unused_args_eliminate]: 2.44001e-06
[environ_conv]: 5.30001e-06
[add_recomputation]: 4.902e-05
[cse_after_recomputation]: 2.176e-05, [1]
    [Cycle 1]: 1.588e-05, [1]
        [cse]: 9.68002e-06
[auto_monad_reorder]: 1.502e-05
[get_jit_bprop_graph]: 1.89e-06
[rewriter_after_jit_bprop_graph]: 6.62002e-06
[opt_after_jit_grad]: 0.00050795
[symbol_engine_optimizer]: 7.904e-05, [1]
    [Cycle 1]: 7.023e-05, [6]
        [build]: 3.84002e-06
        [elim_shapecalc]: 7.73001e-06
        [elim_not_effective]: 1.316e-05
        [opt_reshape]: 6.04001e-06
        [fold_const_symbol]: 9.01002e-06
        [renormalize]: 8.40024e-07
[validate]: 3.856e-05
Sums
    bootstrap                            :     0.000284s :  3.15%
    type_inference                       :     0.004455s : 49.36%
    event_method                         :     0.000018s :  0.20%
    auto_monad                           :     0.000071s :  0.79%
    graph_reusing                        :     0.000007s :  0.08%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000043s :  0.48%
    rewriter_before_opt_a                :     0.000089s :  0.98%
    expand_dump_flag                     :     0.000004s :  0.04%
    jit_opt_a.switch_simplify            :     0.000068s :  0.75%
    jit_opt_a.loop_unroll                :     0.000036s :  0.40%
    jit_opt_a.a_1                        :     0.000690s :  7.65%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.38%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.17%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000004s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000011s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000032s :  0.35%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000019s :  0.21%
    jit_opt_a.inline_without_move        :     0.000011s :  0.12%
    jit_opt_a.renormalize                :     0.000917s : 10.16%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000021s :  0.23%
    jit_opt_a.cse                        :     0.000042s :  0.46%
    jit_opt_a.replace_applicator         :     0.000022s :  0.24%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.15%
    rewriter_after_opt_a                 :     0.000037s :  0.41%
    convert_after_rewriter               :     0.000009s :  0.10%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000690s :  7.64%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000027s :  0.30%
    loop_unroll                          :     0.000438s :  4.85%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.cse              :     0.000021s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000049s :  0.54%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.07%
    opt_after_jit_grad                   :     0.000508s :  5.63%
    symbol_engine_optimizer.build        :     0.000004s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.43%


Time group info:
------[substitution.]   0.000237    23
 0.89% :     0.000002s :      2: substitution.elim_not_effective
 0.70% :     0.000002s :      2: substitution.fold_const_symbol
 2.27% :     0.000005s :      3: substitution.graph_param_transform
84.17% :     0.000200s :      4: substitution.inline
 1.76% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.47% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.06% :     0.000005s :      2: substitution.replace_old_param
 5.68% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004376     2
78.92% :     0.003454s :      1: type_inference.infer
21.08% :     0.000922s :      1: type_inference.specialize
------[replace.]   0.000081     6
65.74% :     0.000053s :      4: replace.inline
34.26% :     0.000028s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000209     6
94.16% :     0.000197s :      4: match.inline
 5.84% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000122   691
 1.22% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.39% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.00% :     0.000001s :     10: predicate.addn_check_dump
 1.22% :     0.000001s :     10: predicate.addn_zero_filter
 2.30% :     0.000003s :     10: predicate.arithmetic_simplify
 1.09% :     0.000001s :     10: predicate.cast_eliminate
 0.38% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.14% :     0.000001s :     10: predicate.compare_switch_simplify
 1.39% :     0.000002s :     10: predicate.depend_value_elim
 1.13% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.31% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.07% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.87% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.48% :     0.000001s :      3: predicate.elim_not_effective
 0.51% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.10% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.32% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.26% :     0.000002s :     10: predicate.environ_get_eliminate
 1.18% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.82% :     0.000001s :      6: predicate.get_grad_eliminate
 0.26% :     0.000000s :      3: predicate.graph_param_transform
 5.46% :     0.000007s :     22: predicate.inline
 0.88% :     0.000001s :      6: predicate.inline_without_move
 0.34% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.56% :     0.000002s :      6: predicate.less_batch_normalization
 1.43% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.67% :     0.000002s :     15: predicate.load_eliminater
 1.34% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.25% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.80% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.05% :     0.000001s :     10: predicate.merge_addn
 1.01% :     0.000001s :     10: predicate.minmaximum_grad
 1.89% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.07% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.63% :     0.000002s :     10: predicate.reduce_eliminate
 1.39% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.78% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.89% :     0.000002s :     18: predicate.replace_applicator
 0.56% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.47% :     0.000002s :     10: predicate.reshape_eliminate
 1.16% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.80% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.19% :     0.000001s :     10: predicate.same_eliminate
 0.70% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.13% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.45% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.27% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.38% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 3.33% :     0.000004s :     16: predicate.switch_defer_inline
 1.89% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.21% :     0.000009s :     53: predicate.switch_simplify
 1.15% :     0.000001s :     10: predicate.tile_eliminate
 1.06% :     0.000001s :     10: predicate.transpose_eliminate
 2.11% :     0.000003s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.18% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.95% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.78% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.58% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.68% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 2.16% :     0.000003s :     10: predicate.value_based_eliminate
 0.38% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.69% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000656    12
48.37% :     0.000317s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.10% :     0.000040s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.53% :     0.000299s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015833    72
 0.33% :     0.000052s :      1: add_recomputation
 0.48% :     0.000076s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 1.90% :     0.000301s :      1: bootstrap
 0.19% :     0.000030s :      1: cconv
 0.07% :     0.000011s :      1: convert_after_rewriter
 0.15% :     0.000024s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000022s :      1: event_method
 0.04% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000010s :      1: graph_reusing
41.79% :     0.006617s :      1: jit_opt_a
 0.94% :     0.000149s :      1: jit_opt_after_cconv
 0.36% :     0.000056s :      1: jit_opt_b
 2.82% :     0.000446s :      1: loop_unroll
 4.43% :     0.000701s :      1: mutable_eliminate
 5.82% :     0.000922s :     26: opt.transform.jit_opt_a
 0.30% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000027s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000018s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.28% :     0.000519s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.29% :     0.000046s :      1: py_interpret_to_execute
 0.10% :     0.000016s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 3.05% :     0.000483s :      1: renormalize.infer
 2.69% :     0.000426s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000040s :      1: rewriter_after_opt_a
 0.59% :     0.000093s :      1: rewriter_before_opt_a
 0.52% :     0.000082s :      1: symbol_engine_optimizer
28.25% :     0.004473s :      1: type_inference


TotalTime = 0.012887, [30]
[bootstrap]: 0.00031498
[type_inference]: 0.00488027
[event_method]: 1.805e-05
[auto_monad]: 5.855e-05
[graph_reusing]: 5.21002e-06
[pre_auto_parallel]: 2.85998e-06
[py_interpret_to_execute]: 4.016e-05
[rewriter_before_opt_a]: 7.503e-05
[expand_dump_flag]: 3.03e-06
[jit_opt_a]: 0.00555244, [2]
    [Cycle 1]: 0.00159936, [27]
        [switch_simplify]: 5.222e-05
        [loop_unroll]: 2.949e-05
        [a_1]: 0.00054629
        [with_stream_mark]: 1.481e-05
        [recompute_prepare]: 7.33e-06
        [updatestate_depend_eliminate]: 3.11999e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 2.00002e-06
        [parameter_eliminate]: 1.05001e-06
        [specialize_transform]: 5.98002e-06
        [updatestate_useless_node_eliminater]: 5.81e-06
        [accelerated_algorithm]: 5.65001e-06
        [meta_shard_fg_expand]: 1.52001e-06
        [get_grad_eliminate_]: 5.54e-06
        [merge_forward]: 2.76e-06
        [cell_reuse_recompute_pass]: 1.60999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.178e-05
        [j_node_and_user_rematch]: 9.27001e-06
        [meta_fg_expand]: 1.99e-06
        [replace_old_param]: 8.13999e-06
        [inline_without_move]: 5.22e-06
        [renormalize]: 0.00066374
        [add_forward_monad_depend]: 4.4e-06
        [auto_monad_grad]: 1.59e-06
        [auto_monad_eliminator]: 1.024e-05
        [cse]: 1.879e-05
        [replace_applicator]: 1.06e-05
    [Cycle 2]: 0.00030983, [27]
        [switch_simplify]: 5.54e-06
        [loop_unroll]: 4.77998e-06
        [a_1]: 8.858e-05
        [with_stream_mark]: 8.32998e-06
        [recompute_prepare]: 5.05001e-06
        [updatestate_depend_eliminate]: 2.45002e-06
        [updatestate_assign_eliminate]: 1.97999e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 9.39996e-07
        [specialize_transform]: 5.00001e-06
        [updatestate_useless_node_eliminater]: 4.89998e-06
        [accelerated_algorithm]: 4.98001e-06
        [meta_shard_fg_expand]: 1.14998e-06
        [get_grad_eliminate_]: 4.33001e-06
        [merge_forward]: 2.14e-06
        [cell_reuse_recompute_pass]: 1.38002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.175e-05
        [j_node_and_user_rematch]: 7.63999e-06
        [meta_fg_expand]: 1.47999e-06
        [replace_old_param]: 1.182e-05
        [inline_without_move]: 4.71002e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.14003e-06
        [auto_monad_grad]: 7.59988e-07
        [auto_monad_eliminator]: 4.33999e-06
        [cse]: 9.71e-06
        [replace_applicator]: 4.68001e-06
[py_interpret_to_execute_after_opt_a]: 8.57e-06
[rewriter_after_opt_a]: 2.826e-05
[convert_after_rewriter]: 5.47999e-06
[order_py_execute_after_rewriter]: 4.02998e-06
[mutable_eliminate]: 0.00049128
[jit_opt_b]: 4.803e-05, [1]
    [Cycle 1]: 4.203e-05, [2]
        [frontend_op_eliminate]: 1.62e-05
        [inline_after_opt_a]: 1.454e-05
[cconv]: 1.592e-05
[loop_unroll]: 0.00040264
[jit_opt_after_cconv]: 0.00013122, [1]
    [Cycle 1]: 0.00012517, [11]
        [c_1]: 2.033e-05
        [parameter_eliminate]: 2.27999e-06
        [updatestate_depend_eliminate]: 5.35999e-06
        [updatestate_assign_eliminate]: 2.34999e-06
        [updatestate_loads_eliminate]: 1.97999e-06
        [cse]: 1.619e-05
        [call_graph_tuple_transform]: 1.553e-05
        [tuple_list_get_item_eliminator]: 5.18002e-06
        [none_parameter_eliminate]: 1.04998e-06
        [renormalize]: 4.19997e-07
        [switch_simplify]: 5.19998e-06
[remove_dup_value]: 7.68001e-06
[partial_unused_args_eliminate]: 1.74998e-06
[environ_conv]: 3.63999e-06
[add_recomputation]: 3.028e-05
[cse_after_recomputation]: 2.03e-05, [1]
    [Cycle 1]: 1.471e-05, [1]
        [cse]: 9.04e-06
[auto_monad_reorder]: 9.61e-06
[get_jit_bprop_graph]: 1.49998e-06
[rewriter_after_jit_bprop_graph]: 5.09e-06
[opt_after_jit_grad]: 0.00045156
[symbol_engine_optimizer]: 7.011e-05, [1]
    [Cycle 1]: 6.334e-05, [6]
        [build]: 2.69001e-06
        [elim_shapecalc]: 7.83999e-06
        [elim_not_effective]: 1.229e-05
        [opt_reshape]: 5.16998e-06
        [fold_const_symbol]: 8.37e-06
        [renormalize]: 3.60014e-07
[validate]: 2.66e-05
Sums
    bootstrap                            :     0.000315s :  3.63%
    type_inference                       :     0.004880s : 56.23%
    event_method                         :     0.000018s :  0.21%
    auto_monad                           :     0.000059s :  0.67%
    graph_reusing                        :     0.000005s :  0.06%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000040s :  0.46%
    rewriter_before_opt_a                :     0.000075s :  0.86%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000058s :  0.67%
    jit_opt_a.loop_unroll                :     0.000034s :  0.39%
    jit_opt_a.a_1                        :     0.000635s :  7.31%
    jit_opt_a.with_stream_mark           :     0.000023s :  0.27%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_assign_eliminate :     0.000004s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.04%
    jit_opt_a.parameter_eliminate        :     0.000002s :  0.02%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.03%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000005s :  0.06%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000024s :  0.27%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.04%
    jit_opt_a.replace_old_param          :     0.000020s :  0.23%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000664s :  7.65%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.06%
    jit_opt_a.auto_monad_grad            :     0.000002s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000015s :  0.17%
    jit_opt_a.cse                        :     0.000028s :  0.33%
    jit_opt_a.replace_applicator         :     0.000015s :  0.18%
    py_interpret_to_execute_after_opt_a  :     0.000009s :  0.10%
    rewriter_after_opt_a                 :     0.000028s :  0.33%
    convert_after_rewriter               :     0.000005s :  0.06%
    order_py_execute_after_rewriter      :     0.000004s :  0.05%
    mutable_eliminate                    :     0.000491s :  5.66%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000016s :  0.18%
    loop_unroll                          :     0.000403s :  4.64%
    jit_opt_after_cconv.c_1              :     0.000020s :  0.23%
    jit_opt_after_cconv.parameter_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000005s :  0.06%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000016s :  0.19%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000016s :  0.18%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.01%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000008s :  0.09%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000004s :  0.04%
    add_recomputation                    :     0.000030s :  0.35%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000010s :  0.11%
    get_jit_bprop_graph                  :     0.000001s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000452s :  5.20%
    symbol_engine_optimizer.build        :     0.000003s :  0.03%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.14%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000027s :  0.31%


Time group info:
------[substitution.]   0.000207    23
 0.84% :     0.000002s :      2: substitution.elim_not_effective
 0.51% :     0.000001s :      2: substitution.fold_const_symbol
 1.51% :     0.000003s :      3: substitution.graph_param_transform
84.53% :     0.000175s :      4: substitution.inline
 1.21% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.05% :     0.000004s :      4: substitution.remove_not_recompute_node
 3.93% :     0.000008s :      2: substitution.replace_old_param
 5.42% :     0.000011s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004813     2
79.02% :     0.003803s :      1: type_inference.infer
20.98% :     0.001010s :      1: type_inference.specialize
------[replace.]   0.000073     6
65.21% :     0.000047s :      4: replace.inline
34.79% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000183     6
94.42% :     0.000173s :      4: match.inline
 5.58% :     0.000010s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000116   691
 1.36% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.22% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.42% :     0.000002s :     10: predicate.addn_zero_filter
 2.01% :     0.000002s :     10: predicate.arithmetic_simplify
 1.40% :     0.000002s :     10: predicate.cast_eliminate
 0.44% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.59% :     0.000002s :     10: predicate.depend_value_elim
 1.12% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.18% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.25% :     0.000001s :     10: predicate.dict_set_item_eliminator
 0.90% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.29% :     0.000000s :      3: predicate.elim_not_effective
 0.64% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.44% :     0.000002s :     10: predicate.environ_add_const_eliminate
 1.22% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.73% :     0.000002s :     10: predicate.environ_get_eliminate
 1.09% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.92% :     0.000001s :      6: predicate.get_grad_eliminate
 0.25% :     0.000000s :      3: predicate.graph_param_transform
 5.32% :     0.000006s :     22: predicate.inline
 0.95% :     0.000001s :      6: predicate.inline_without_move
 0.39% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.13% :     0.000001s :      6: predicate.less_batch_normalization
 1.77% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.94% :     0.000002s :     15: predicate.load_eliminater
 1.21% :     0.000001s :      3: predicate.loop_unroll_after_grad
 4.76% :     0.000006s :     34: predicate.loop_unroll_before_grad
 2.54% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.44% :     0.000001s :      3: predicate.opt_reshape
 2.40% :     0.000003s :     15: predicate.partial_eliminate
 1.13% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.52% :     0.000002s :     10: predicate.reduce_eliminate
 1.66% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.51% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.70% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.30% :     0.000000s :      3: predicate.reset_defer_inline
 1.25% :     0.000001s :     10: predicate.reshape_eliminate
 1.31% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 0.67% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.17% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.12% :     0.000001s :      6: predicate.special_op_eliminate
 0.98% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.31% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.41% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.76% :     0.000003s :     16: predicate.switch_defer_inline
 2.40% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.77% :     0.000009s :     53: predicate.switch_simplify
 1.13% :     0.000001s :     10: predicate.tile_eliminate
 1.13% :     0.000001s :     10: predicate.transpose_eliminate
 1.47% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.13% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.36% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.52% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.57% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.61% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.75% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.29% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.65% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000740    12
55.00% :     0.000407s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.92% :     0.000036s :      1: func_graph_cloner_run.FuncGraphClonerNode
40.07% :     0.000296s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014446    72
 0.23% :     0.000033s :      1: add_recomputation
 0.43% :     0.000062s :      1: auto_monad
 0.08% :     0.000012s :      1: auto_monad_reorder
 2.32% :     0.000335s :      1: bootstrap
 0.13% :     0.000018s :      1: cconv
 0.05% :     0.000008s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.04% :     0.000006s :      1: environ_conv
 0.15% :     0.000022s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
38.46% :     0.005556s :      1: jit_opt_a
 0.93% :     0.000134s :      1: jit_opt_after_cconv
 0.35% :     0.000051s :      1: jit_opt_b
 2.84% :     0.000410s :      1: loop_unroll
 3.45% :     0.000499s :      1: mutable_eliminate
 5.83% :     0.000842s :     26: opt.transform.jit_opt_a
 0.30% :     0.000043s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000025s :      4: opt.transform.jit_opt_b
 0.07% :     0.000010s :      1: opt.transform.loop_unroll_optimizer
 0.09% :     0.000013s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.18% :     0.000459s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.30% :     0.000043s :      1: py_interpret_to_execute
 0.08% :     0.000011s :      1: py_interpret_to_execute_after_opt_a
 0.07% :     0.000010s :      1: remove_dup_value
 2.36% :     0.000342s :      1: renormalize.infer
 2.18% :     0.000315s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000031s :      1: rewriter_after_opt_a
 0.54% :     0.000078s :      1: rewriter_before_opt_a
 0.50% :     0.000073s :      1: symbol_engine_optimizer
33.88% :     0.004894s :      1: type_inference


TotalTime = 0.0127213, [30]
[bootstrap]: 0.00028842
[type_inference]: 0.004245
[event_method]: 1.789e-05
[auto_monad]: 6.368e-05
[graph_reusing]: 5.82001e-06
[pre_auto_parallel]: 2.41e-06
[py_interpret_to_execute]: 3.664e-05
[rewriter_before_opt_a]: 7.233e-05
[expand_dump_flag]: 2.96001e-06
[jit_opt_a]: 0.00574885, [2]
    [Cycle 1]: 0.00184086, [27]
        [switch_simplify]: 5.978e-05
        [loop_unroll]: 2.913e-05
        [a_1]: 0.0005551
        [with_stream_mark]: 2.183e-05
        [recompute_prepare]: 7.15e-06
        [updatestate_depend_eliminate]: 3.9e-06
        [updatestate_assign_eliminate]: 3.4e-06
        [updatestate_loads_eliminate]: 2.24999e-06
        [parameter_eliminate]: 1.92999e-06
        [specialize_transform]: 6.43003e-06
        [updatestate_useless_node_eliminater]: 5.14e-06
        [accelerated_algorithm]: 5.75001e-06
        [meta_shard_fg_expand]: 2.12999e-06
        [get_grad_eliminate_]: 5.60001e-06
        [merge_forward]: 3.7e-06
        [cell_reuse_recompute_pass]: 1.67999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.405e-05
        [j_node_and_user_rematch]: 9.52001e-06
        [meta_fg_expand]: 1.92999e-06
        [replace_old_param]: 8.33001e-06
        [inline_without_move]: 5.68002e-06
        [renormalize]: 0.0008621
        [add_forward_monad_depend]: 4.99e-06
        [auto_monad_grad]: 1.77999e-06
        [auto_monad_eliminator]: 1.101e-05
        [cse]: 1.896e-05
        [replace_applicator]: 1.37e-05
    [Cycle 2]: 0.000322, [27]
        [switch_simplify]: 5.63002e-06
        [loop_unroll]: 5.27999e-06
        [a_1]: 9.024e-05
        [with_stream_mark]: 8.79998e-06
        [recompute_prepare]: 5.09e-06
        [updatestate_depend_eliminate]: 2.32999e-06
        [updatestate_assign_eliminate]: 2.07001e-06
        [updatestate_loads_eliminate]: 1.92999e-06
        [parameter_eliminate]: 8.70001e-07
        [specialize_transform]: 4.90999e-06
        [updatestate_useless_node_eliminater]: 4.80001e-06
        [accelerated_algorithm]: 5.12e-06
        [meta_shard_fg_expand]: 1.01002e-06
        [get_grad_eliminate_]: 4.47e-06
        [merge_forward]: 2.44001e-06
        [cell_reuse_recompute_pass]: 1.68002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.193e-05
        [j_node_and_user_rematch]: 7.41001e-06
        [meta_fg_expand]: 1.56998e-06
        [replace_old_param]: 7.7e-06
        [inline_without_move]: 4.52e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.33002e-06
        [auto_monad_grad]: 1.36002e-06
        [auto_monad_eliminator]: 5.36002e-06
        [cse]: 1.052e-05
        [replace_applicator]: 5.14e-06
[py_interpret_to_execute_after_opt_a]: 9.72999e-06
[rewriter_after_opt_a]: 2.951e-05
[convert_after_rewriter]: 5.75001e-06
[order_py_execute_after_rewriter]: 3.8e-06
[mutable_eliminate]: 0.00061224
[jit_opt_b]: 5.386e-05, [1]
    [Cycle 1]: 4.671e-05, [2]
        [frontend_op_eliminate]: 1.824e-05
        [inline_after_opt_a]: 1.686e-05
[cconv]: 2.526e-05
[loop_unroll]: 0.00043295
[jit_opt_after_cconv]: 0.00014702, [1]
    [Cycle 1]: 0.00013949, [11]
        [c_1]: 2.212e-05
        [parameter_eliminate]: 2.51e-06
        [updatestate_depend_eliminate]: 5.59998e-06
        [updatestate_assign_eliminate]: 2.34001e-06
        [updatestate_loads_eliminate]: 2.15002e-06
        [cse]: 2.139e-05
        [call_graph_tuple_transform]: 1.784e-05
        [tuple_list_get_item_eliminator]: 5.28002e-06
        [none_parameter_eliminate]: 1.75001e-06
        [renormalize]: 5.59987e-07
        [switch_simplify]: 5.32001e-06
[remove_dup_value]: 1.376e-05
[partial_unused_args_eliminate]: 1.67001e-06
[environ_conv]: 5.56e-06
[add_recomputation]: 4.808e-05
[cse_after_recomputation]: 2.306e-05, [1]
    [Cycle 1]: 1.653e-05, [1]
        [cse]: 9.67999e-06
[auto_monad_reorder]: 1.628e-05
[get_jit_bprop_graph]: 2.35002e-06
[rewriter_after_jit_bprop_graph]: 6.99001e-06
[opt_after_jit_grad]: 0.00050828
[symbol_engine_optimizer]: 7.823e-05, [1]
    [Cycle 1]: 7.012e-05, [6]
        [build]: 4.16001e-06
        [elim_shapecalc]: 8.17e-06
        [elim_not_effective]: 1.34e-05
        [opt_reshape]: 6.33e-06
        [fold_const_symbol]: 9.13002e-06
        [renormalize]: 5.89993e-07
[validate]: 3.541e-05
Sums
    bootstrap                            :     0.000288s :  3.38%
    type_inference                       :     0.004245s : 49.73%
    event_method                         :     0.000018s :  0.21%
    auto_monad                           :     0.000064s :  0.75%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000037s :  0.43%
    rewriter_before_opt_a                :     0.000072s :  0.85%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000065s :  0.77%
    jit_opt_a.loop_unroll                :     0.000034s :  0.40%
    jit_opt_a.a_1                        :     0.000645s :  7.56%
    jit_opt_a.with_stream_mark           :     0.000031s :  0.36%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.30%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000003s :  0.04%
    jit_opt_a.replace_old_param          :     0.000016s :  0.19%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000862s : 10.10%
    jit_opt_a.add_forward_monad_depend   :     0.000006s :  0.07%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000016s :  0.19%
    jit_opt_a.cse                        :     0.000029s :  0.35%
    jit_opt_a.replace_applicator         :     0.000019s :  0.22%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.11%
    rewriter_after_opt_a                 :     0.000030s :  0.35%
    convert_after_rewriter               :     0.000006s :  0.07%
    order_py_execute_after_rewriter      :     0.000004s :  0.04%
    mutable_eliminate                    :     0.000612s :  7.17%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.20%
    cconv                                :     0.000025s :  0.30%
    loop_unroll                          :     0.000433s :  5.07%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000021s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000018s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000048s :  0.56%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000016s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.08%
    opt_after_jit_grad                   :     0.000508s :  5.95%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.10%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.11%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000035s :  0.41%


Time group info:
------[substitution.]   0.000220    23
 0.71% :     0.000002s :      2: substitution.elim_not_effective
 0.43% :     0.000001s :      2: substitution.fold_const_symbol
 1.66% :     0.000004s :      3: substitution.graph_param_transform
86.44% :     0.000190s :      4: substitution.inline
 1.29% :     0.000003s :      4: substitution.j_node_and_user_rematch
 1.94% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.01% :     0.000004s :      2: substitution.replace_old_param
 5.52% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004178     2
78.72% :     0.003289s :      1: type_inference.infer
21.28% :     0.000889s :      1: type_inference.specialize
------[replace.]   0.000073     6
66.16% :     0.000048s :      4: replace.inline
33.84% :     0.000025s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000199     6
94.41% :     0.000188s :      4: match.inline
 5.59% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000114   691
 1.26% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.41% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.03% :     0.000001s :     10: predicate.addn_check_dump
 1.17% :     0.000001s :     10: predicate.addn_zero_filter
 1.80% :     0.000002s :     10: predicate.arithmetic_simplify
 1.22% :     0.000001s :     10: predicate.cast_eliminate
 0.56% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.87% :     0.000002s :     10: predicate.depend_value_elim
 1.14% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.36% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.18% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.24% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.54% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.13% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.23% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.26% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.47% :     0.000002s :     10: predicate.environ_get_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.24% :     0.000000s :      3: predicate.fold_const_symbol
 1.16% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.33% :     0.000006s :     22: predicate.inline
 1.03% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.15% :     0.000001s :      6: predicate.less_batch_normalization
 1.49% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.79% :     0.000002s :     15: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.08% :     0.000001s :     10: predicate.merge_addn
 1.06% :     0.000001s :     10: predicate.minmaximum_grad
 1.69% :     0.000002s :      3: predicate.mutable_eliminate
 0.43% :     0.000000s :      3: predicate.opt_reshape
 2.10% :     0.000002s :     15: predicate.partial_eliminate
 1.21% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.76% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.52% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.74% :     0.000002s :     18: predicate.replace_applicator
 0.58% :     0.000001s :      6: predicate.replace_old_param
 0.39% :     0.000000s :      3: predicate.reset_defer_inline
 1.46% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.94% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.38% :     0.000002s :      6: predicate.special_op_eliminate
 1.02% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.22% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 2.26% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.43% :     0.000008s :     53: predicate.switch_simplify
 1.30% :     0.000001s :     10: predicate.tile_eliminate
 1.30% :     0.000001s :     10: predicate.transpose_eliminate
 1.29% :     0.000001s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.21% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.38% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.43% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.77% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.70% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 3.08% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.59% :     0.000002s :     10: predicate.value_based_eliminate
 0.37% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 1.19% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000626    12
46.36% :     0.000290s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.93% :     0.000043s :      1: func_graph_cloner_run.FuncGraphClonerNode
46.71% :     0.000292s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014502    72
 0.35% :     0.000051s :      1: add_recomputation
 0.46% :     0.000067s :      1: auto_monad
 0.13% :     0.000019s :      1: auto_monad_reorder
 2.07% :     0.000301s :      1: bootstrap
 0.19% :     0.000028s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.16% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
39.67% :     0.005752s :      1: jit_opt_a
 1.03% :     0.000150s :      1: jit_opt_after_cconv
 0.39% :     0.000056s :      1: jit_opt_b
 3.05% :     0.000442s :      1: loop_unroll
 4.28% :     0.000621s :      1: mutable_eliminate
 5.94% :     0.000861s :     26: opt.transform.jit_opt_a
 0.33% :     0.000047s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000028s :      4: opt.transform.jit_opt_b
 0.09% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000015s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000023s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.57% :     0.000517s :      1: opt_after_jit_grad
 0.04% :     0.000006s :      1: order_py_execute_after_rewriter
 0.02% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.27% :     0.000040s :      1: py_interpret_to_execute
 0.08% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000016s :      1: remove_dup_value
 3.15% :     0.000457s :      1: renormalize.infer
 2.74% :     0.000398s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.22% :     0.000032s :      1: rewriter_after_opt_a
 0.52% :     0.000075s :      1: rewriter_before_opt_a
 0.56% :     0.000081s :      1: symbol_engine_optimizer
29.38% :     0.004260s :      1: type_inference


TotalTime = 0.0130984, [30]
[bootstrap]: 0.00027039
[type_inference]: 0.0042432
[event_method]: 1.777e-05
[auto_monad]: 7.546e-05
[graph_reusing]: 5.81e-06
[pre_auto_parallel]: 2.26998e-06
[py_interpret_to_execute]: 3.352e-05
[rewriter_before_opt_a]: 6.963e-05
[expand_dump_flag]: 3.02002e-06
[jit_opt_a]: 0.00606972, [2]
    [Cycle 1]: 0.00198018, [27]
        [switch_simplify]: 6.243e-05
        [loop_unroll]: 3.051e-05
        [a_1]: 0.00058022
        [with_stream_mark]: 2.102e-05
        [recompute_prepare]: 9.09e-06
        [updatestate_depend_eliminate]: 2.94999e-06
        [updatestate_assign_eliminate]: 3.16001e-06
        [updatestate_loads_eliminate]: 2.94001e-06
        [parameter_eliminate]: 1.94999e-06
        [specialize_transform]: 6.13998e-06
        [updatestate_useless_node_eliminater]: 4.95999e-06
        [accelerated_algorithm]: 5.99e-06
        [meta_shard_fg_expand]: 3.06001e-06
        [get_grad_eliminate_]: 5.81e-06
        [merge_forward]: 3.86001e-06
        [cell_reuse_recompute_pass]: 1.50001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.666e-05
        [j_node_and_user_rematch]: 1.001e-05
        [meta_fg_expand]: 2.33998e-06
        [replace_old_param]: 9.92001e-06
        [inline_without_move]: 5.82001e-06
        [renormalize]: 0.00092557
        [add_forward_monad_depend]: 7.8e-06
        [auto_monad_grad]: 2.88e-06
        [auto_monad_eliminator]: 1.455e-05
        [cse]: 2.248e-05
        [replace_applicator]: 1.375e-05
    [Cycle 2]: 0.00035768, [27]
        [switch_simplify]: 5.87999e-06
        [loop_unroll]: 4.87998e-06
        [a_1]: 9.423e-05
        [with_stream_mark]: 1.245e-05
        [recompute_prepare]: 5.08002e-06
        [updatestate_depend_eliminate]: 3.25998e-06
        [updatestate_assign_eliminate]: 2.31e-06
        [updatestate_loads_eliminate]: 1.91e-06
        [parameter_eliminate]: 1.49e-06
        [specialize_transform]: 5.14e-06
        [updatestate_useless_node_eliminater]: 5.55001e-06
        [accelerated_algorithm]: 5.35999e-06
        [meta_shard_fg_expand]: 1.27e-06
        [get_grad_eliminate_]: 4.66002e-06
        [merge_forward]: 3.29001e-06
        [cell_reuse_recompute_pass]: 1.90001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.458e-05
        [j_node_and_user_rematch]: 8.38001e-06
        [meta_fg_expand]: 1.70001e-06
        [replace_old_param]: 8.00999e-06
        [inline_without_move]: 5.32001e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 2.39999e-06
        [auto_monad_grad]: 1.15001e-06
        [auto_monad_eliminator]: 7.01999e-06
        [cse]: 1.318e-05
        [replace_applicator]: 6.01003e-06
[py_interpret_to_execute_after_opt_a]: 1.427e-05
[rewriter_after_opt_a]: 3.722e-05
[convert_after_rewriter]: 6.96999e-06
[order_py_execute_after_rewriter]: 5.05001e-06
[mutable_eliminate]: 0.00067287
[jit_opt_b]: 5.334e-05, [1]
    [Cycle 1]: 4.478e-05, [2]
        [frontend_op_eliminate]: 1.659e-05
        [inline_after_opt_a]: 1.563e-05
[cconv]: 2.896e-05
[loop_unroll]: 0.00044016
[jit_opt_after_cconv]: 0.00014655, [1]
    [Cycle 1]: 0.00013966, [11]
        [c_1]: 2.103e-05
        [parameter_eliminate]: 3.18e-06
        [updatestate_depend_eliminate]: 6.24999e-06
        [updatestate_assign_eliminate]: 2.69001e-06
        [updatestate_loads_eliminate]: 2.01e-06
        [cse]: 2.286e-05
        [call_graph_tuple_transform]: 1.913e-05
        [tuple_list_get_item_eliminator]: 5.88002e-06
        [none_parameter_eliminate]: 1.57001e-06
        [renormalize]: 5.39992e-07
        [switch_simplify]: 5.72999e-06
[remove_dup_value]: 1.488e-05
[partial_unused_args_eliminate]: 2.41e-06
[environ_conv]: 5.52999e-06
[add_recomputation]: 4.765e-05
[cse_after_recomputation]: 2.048e-05, [1]
    [Cycle 1]: 1.47e-05, [1]
        [cse]: 8.90999e-06
[auto_monad_reorder]: 1.584e-05
[get_jit_bprop_graph]: 2.06e-06
[rewriter_after_jit_bprop_graph]: 7.10998e-06
[opt_after_jit_grad]: 0.0004944
[symbol_engine_optimizer]: 7.686e-05, [1]
    [Cycle 1]: 6.975e-05, [6]
        [build]: 4.35e-06
        [elim_shapecalc]: 7.92e-06
        [elim_not_effective]: 1.361e-05
        [opt_reshape]: 5.75001e-06
        [fold_const_symbol]: 9.15999e-06
        [renormalize]: 5.8001e-07
[validate]: 3.908e-05
Sums
    bootstrap                            :     0.000270s :  3.10%
    type_inference                       :     0.004243s : 48.59%
    event_method                         :     0.000018s :  0.20%
    auto_monad                           :     0.000075s :  0.86%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000034s :  0.38%
    rewriter_before_opt_a                :     0.000070s :  0.80%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000068s :  0.78%
    jit_opt_a.loop_unroll                :     0.000035s :  0.41%
    jit_opt_a.a_1                        :     0.000674s :  7.72%
    jit_opt_a.with_stream_mark           :     0.000033s :  0.38%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.16%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.36%
    jit_opt_a.j_node_and_user_rematch    :     0.000018s :  0.21%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000018s :  0.21%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000926s : 10.60%
    jit_opt_a.add_forward_monad_depend   :     0.000010s :  0.12%
    jit_opt_a.auto_monad_grad            :     0.000004s :  0.05%
    jit_opt_a.auto_monad_eliminator      :     0.000022s :  0.25%
    jit_opt_a.cse                        :     0.000036s :  0.41%
    jit_opt_a.replace_applicator         :     0.000020s :  0.23%
    py_interpret_to_execute_after_opt_a  :     0.000014s :  0.16%
    rewriter_after_opt_a                 :     0.000037s :  0.43%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000673s :  7.71%
    jit_opt_b.frontend_op_eliminate      :     0.000017s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.18%
    cconv                                :     0.000029s :  0.33%
    loop_unroll                          :     0.000440s :  5.04%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000023s :  0.26%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000019s :  0.22%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.07%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000048s :  0.55%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000007s :  0.08%
    opt_after_jit_grad                   :     0.000494s :  5.66%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000014s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000039s :  0.45%


Time group info:
------[substitution.]   0.000226    23
 0.79% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000002s :      2: substitution.fold_const_symbol
 2.52% :     0.000006s :      3: substitution.graph_param_transform
83.70% :     0.000189s :      4: substitution.inline
 1.85% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.09% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.14% :     0.000005s :      2: substitution.replace_old_param
 6.18% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004182     2
79.32% :     0.003317s :      1: type_inference.infer
20.68% :     0.000865s :      1: type_inference.specialize
------[replace.]   0.000085     6
62.49% :     0.000053s :      4: replace.inline
37.51% :     0.000032s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000199     6
93.56% :     0.000187s :      4: match.inline
 6.44% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.23% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.14% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.01% :     0.000001s :     10: predicate.addn_check_dump
 1.13% :     0.000001s :     10: predicate.addn_zero_filter
 1.91% :     0.000002s :     10: predicate.arithmetic_simplify
 1.08% :     0.000001s :     10: predicate.cast_eliminate
 0.56% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.99% :     0.000001s :     10: predicate.compare_switch_simplify
 1.30% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.13% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.13% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.13% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.32% :     0.000000s :      3: predicate.elim_not_effective
 0.59% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.07% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.75% :     0.000002s :     10: predicate.environ_get_add_eliminate
 1.03% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.63% :     0.000002s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 1.13% :     0.000001s :      6: predicate.get_grad_eliminate
 0.42% :     0.000001s :      3: predicate.graph_param_transform
 5.38% :     0.000006s :     22: predicate.inline
 1.19% :     0.000001s :      6: predicate.inline_without_move
 0.63% :     0.000001s :      6: predicate.j_node_and_user_rematch
 1.18% :     0.000001s :      6: predicate.less_batch_normalization
 1.44% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.30% :     0.000002s :      3: predicate.loop_unroll_after_grad
 4.13% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.00% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.00% :     0.000001s :     10: predicate.merge_addn
 1.02% :     0.000001s :     10: predicate.minmaximum_grad
 1.99% :     0.000002s :      3: predicate.mutable_eliminate
 0.47% :     0.000001s :      3: predicate.opt_reshape
 2.49% :     0.000003s :     15: predicate.partial_eliminate
 1.20% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.39% :     0.000002s :     10: predicate.reduce_eliminate
 1.46% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.77% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.81% :     0.000002s :     18: predicate.replace_applicator
 0.74% :     0.000001s :      6: predicate.replace_old_param
 0.28% :     0.000000s :      3: predicate.reset_defer_inline
 1.43% :     0.000002s :     10: predicate.reshape_eliminate
 1.12% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.97% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.11% :     0.000001s :     10: predicate.same_eliminate
 0.78% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.02% :     0.000001s :      6: predicate.special_op_eliminate
 0.94% :     0.000001s :      6: predicate.specialize_transform
 1.21% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.59% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.38% :     0.000003s :     16: predicate.switch_defer_inline
 2.20% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.20% :     0.000009s :     53: predicate.switch_simplify
 1.33% :     0.000002s :     10: predicate.tile_eliminate
 1.38% :     0.000002s :     10: predicate.transpose_eliminate
 1.36% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.08% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.10% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.48% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 2.34% :     0.000003s :     12: predicate.tuple_to_list_eliminator_
 1.55% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.69% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.43% :     0.000002s :     10: predicate.value_based_eliminate
 0.36% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.83% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000632    12
46.45% :     0.000294s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.98% :     0.000038s :      1: func_graph_cloner_run.FuncGraphClonerNode
47.57% :     0.000301s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014977    72
 0.34% :     0.000051s :      1: add_recomputation
 0.53% :     0.000079s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 1.89% :     0.000283s :      1: bootstrap
 0.21% :     0.000031s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000023s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.14% :     0.000021s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
40.55% :     0.006074s :      1: jit_opt_a
 1.00% :     0.000149s :      1: jit_opt_after_cconv
 0.37% :     0.000056s :      1: jit_opt_b
 2.99% :     0.000448s :      1: loop_unroll
 4.57% :     0.000685s :      1: mutable_eliminate
 6.03% :     0.000904s :     26: opt.transform.jit_opt_a
 0.32% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000026s :      4: opt.transform.jit_opt_b
 0.08% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.36% :     0.000503s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000036s :      1: py_interpret_to_execute
 0.11% :     0.000017s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000017s :      1: remove_dup_value
 3.23% :     0.000484s :      1: renormalize.infer
 2.88% :     0.000432s :      1: renormalize.specialize
 0.06% :     0.000009s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000042s :      1: rewriter_after_opt_a
 0.48% :     0.000072s :      1: rewriter_before_opt_a
 0.53% :     0.000080s :      1: symbol_engine_optimizer
28.41% :     0.004255s :      1: type_inference


TotalTime = 0.0139656, [30]
[bootstrap]: 0.00031773
[type_inference]: 0.00463759
[event_method]: 1.849e-05
[auto_monad]: 7.057e-05
[graph_reusing]: 6.94999e-06
[pre_auto_parallel]: 2.74999e-06
[py_interpret_to_execute]: 3.882e-05
[rewriter_before_opt_a]: 7.682e-05
[expand_dump_flag]: 3.17002e-06
[jit_opt_a]: 0.00618072, [2]
    [Cycle 1]: 0.0020572, [27]
        [switch_simplify]: 6.309e-05
        [loop_unroll]: 2.957e-05
        [a_1]: 0.00060166
        [with_stream_mark]: 2.325e-05
        [recompute_prepare]: 8.52e-06
        [updatestate_depend_eliminate]: 4.68001e-06
        [updatestate_assign_eliminate]: 3.36999e-06
        [updatestate_loads_eliminate]: 2.46e-06
        [parameter_eliminate]: 1.89e-06
        [specialize_transform]: 5.62999e-06
        [updatestate_useless_node_eliminater]: 5.51e-06
        [accelerated_algorithm]: 6.38e-06
        [meta_shard_fg_expand]: 2.69001e-06
        [get_grad_eliminate_]: 5.25001e-06
        [merge_forward]: 4.43001e-06
        [cell_reuse_recompute_pass]: 1.32e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.449e-05
        [j_node_and_user_rematch]: 9.89999e-06
        [meta_fg_expand]: 2.69999e-06
        [replace_old_param]: 1.05e-05
        [inline_without_move]: 5.04e-06
        [renormalize]: 0.00097493
        [add_forward_monad_depend]: 7.20003e-06
        [auto_monad_grad]: 2.36998e-06
        [auto_monad_eliminator]: 1.686e-05
        [cse]: 3.041e-05
        [replace_applicator]: 1.353e-05
    [Cycle 2]: 0.0003769, [27]
        [switch_simplify]: 5.77001e-06
        [loop_unroll]: 5.45001e-06
        [a_1]: 9.452e-05
        [with_stream_mark]: 1.29e-05
        [recompute_prepare]: 5.10999e-06
        [updatestate_depend_eliminate]: 2.85998e-06
        [updatestate_assign_eliminate]: 2.44001e-06
        [updatestate_loads_eliminate]: 1.92001e-06
        [parameter_eliminate]: 1.31002e-06
        [specialize_transform]: 4.94e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 1.961e-05
        [meta_shard_fg_expand]: 1.67001e-06
        [get_grad_eliminate_]: 4.77998e-06
        [merge_forward]: 3.65e-06
        [cell_reuse_recompute_pass]: 2.02001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.563e-05
        [j_node_and_user_rematch]: 9.10999e-06
        [meta_fg_expand]: 1.66e-06
        [replace_old_param]: 1.848e-05
        [inline_without_move]: 5.11002e-06
        [renormalize]: 5.9983e-08
        [add_forward_monad_depend]: 2.09e-06
        [auto_monad_grad]: 1.07e-06
        [auto_monad_eliminator]: 6.13998e-06
        [cse]: 1.313e-05
        [replace_applicator]: 6.44999e-06
[py_interpret_to_execute_after_opt_a]: 1.756e-05
[rewriter_after_opt_a]: 4.244e-05
[convert_after_rewriter]: 8.03999e-06
[order_py_execute_after_rewriter]: 6.21e-06
[mutable_eliminate]: 0.00068949
[jit_opt_b]: 5.598e-05, [1]
    [Cycle 1]: 4.741e-05, [2]
        [frontend_op_eliminate]: 1.795e-05
        [inline_after_opt_a]: 1.695e-05
[cconv]: 3.051e-05
[loop_unroll]: 0.00051434
[jit_opt_after_cconv]: 0.00016803, [1]
    [Cycle 1]: 0.00016021, [11]
        [c_1]: 2.328e-05
        [parameter_eliminate]: 4.08001e-06
        [updatestate_depend_eliminate]: 8.57998e-06
        [updatestate_assign_eliminate]: 2.73e-06
        [updatestate_loads_eliminate]: 2.19001e-06
        [cse]: 2.731e-05
        [call_graph_tuple_transform]: 2.369e-05
        [tuple_list_get_item_eliminator]: 5.86e-06
        [none_parameter_eliminate]: 1.69e-06
        [renormalize]: 8.29983e-07
        [switch_simplify]: 6.41e-06
[remove_dup_value]: 1.532e-05
[partial_unused_args_eliminate]: 2.31998e-06
[environ_conv]: 6.16998e-06
[add_recomputation]: 5.48e-05
[cse_after_recomputation]: 2.421e-05, [1]
    [Cycle 1]: 1.832e-05, [1]
        [cse]: 1.024e-05
[auto_monad_reorder]: 1.547e-05
[get_jit_bprop_graph]: 2.12999e-06
[rewriter_after_jit_bprop_graph]: 8.75999e-06
[opt_after_jit_grad]: 0.00060998
[symbol_engine_optimizer]: 8.476e-05, [1]
    [Cycle 1]: 7.652e-05, [6]
        [build]: 5.13002e-06
        [elim_shapecalc]: 8.67998e-06
        [elim_not_effective]: 1.542e-05
        [opt_reshape]: 6.59001e-06
        [fold_const_symbol]: 9.88998e-06
        [renormalize]: 1.12e-06
[validate]: 4.815e-05
Sums
    bootstrap                            :     0.000318s :  3.33%
    type_inference                       :     0.004638s : 48.54%
    event_method                         :     0.000018s :  0.19%
    auto_monad                           :     0.000071s :  0.74%
    graph_reusing                        :     0.000007s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000039s :  0.41%
    rewriter_before_opt_a                :     0.000077s :  0.80%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000069s :  0.72%
    jit_opt_a.loop_unroll                :     0.000035s :  0.37%
    jit_opt_a.a_1                        :     0.000696s :  7.29%
    jit_opt_a.with_stream_mark           :     0.000036s :  0.38%
    jit_opt_a.recompute_prepare          :     0.000014s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.08%
    jit_opt_a.updatestate_assign_eliminate :     0.000006s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.11%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000011s :  0.11%
    jit_opt_a.accelerated_algorithm      :     0.000026s :  0.27%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.05%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.10%
    jit_opt_a.merge_forward              :     0.000008s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000030s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000029s :  0.30%
    jit_opt_a.inline_without_move        :     0.000010s :  0.11%
    jit_opt_a.renormalize                :     0.000975s : 10.21%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.10%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000023s :  0.24%
    jit_opt_a.cse                        :     0.000044s :  0.46%
    jit_opt_a.replace_applicator         :     0.000020s :  0.21%
    py_interpret_to_execute_after_opt_a  :     0.000018s :  0.18%
    rewriter_after_opt_a                 :     0.000042s :  0.44%
    convert_after_rewriter               :     0.000008s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.07%
    mutable_eliminate                    :     0.000689s :  7.22%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.19%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.18%
    cconv                                :     0.000031s :  0.32%
    loop_unroll                          :     0.000514s :  5.38%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000004s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000009s :  0.09%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000027s :  0.29%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000024s :  0.25%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000006s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.07%
    remove_dup_value                     :     0.000015s :  0.16%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000006s :  0.06%
    add_recomputation                    :     0.000055s :  0.57%
    cse_after_recomputation.cse          :     0.000010s :  0.11%
    auto_monad_reorder                   :     0.000015s :  0.16%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000009s :  0.09%
    opt_after_jit_grad                   :     0.000610s :  6.38%
    symbol_engine_optimizer.build        :     0.000005s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000009s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.16%
    symbol_engine_optimizer.opt_reshape  :     0.000007s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000010s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000048s :  0.50%


Time group info:
------[substitution.]   0.000253    23
 0.93% :     0.000002s :      2: substitution.elim_not_effective
 0.74% :     0.000002s :      2: substitution.fold_const_symbol
 2.77% :     0.000007s :      3: substitution.graph_param_transform
84.34% :     0.000213s :      4: substitution.inline
 1.59% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.01% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.26% :     0.000006s :      2: substitution.replace_old_param
 5.37% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004562     2
80.03% :     0.003651s :      1: type_inference.infer
19.97% :     0.000911s :      1: type_inference.specialize
------[replace.]   0.000083     6
68.65% :     0.000057s :      4: replace.inline
31.35% :     0.000026s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000223     6
94.35% :     0.000210s :      4: match.inline
 5.65% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000124   691
 1.25% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.56% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.96% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000002s :     10: predicate.addn_zero_filter
 1.98% :     0.000002s :     10: predicate.arithmetic_simplify
 1.33% :     0.000002s :     10: predicate.cast_eliminate
 0.46% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.04% :     0.000001s :     10: predicate.compare_switch_simplify
 1.21% :     0.000001s :     10: predicate.depend_value_elim
 1.02% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.38% :     0.000002s :     10: predicate.dict_get_item_eliminator
 1.09% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.23% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.41% :     0.000001s :      3: predicate.elim_not_effective
 0.66% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.08% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.02% :     0.000001s :     10: predicate.environ_get_add_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.27% :     0.000002s :     10: predicate.environ_get_eliminate
 1.05% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 1.04% :     0.000001s :      6: predicate.get_grad_eliminate
 0.37% :     0.000000s :      3: predicate.graph_param_transform
 5.94% :     0.000007s :     22: predicate.inline
 1.02% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.35% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 1.94% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.67% :     0.000005s :     34: predicate.loop_unroll_before_grad
 2.19% :     0.000003s :     13: predicate.make_slice_get_slice_eliminator
 0.99% :     0.000001s :     10: predicate.merge_addn
 1.12% :     0.000001s :     10: predicate.minmaximum_grad
 2.45% :     0.000003s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.51% :     0.000003s :     15: predicate.partial_eliminate
 1.43% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.49% :     0.000002s :     10: predicate.reduce_eliminate
 1.38% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.63% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.55% :     0.000002s :     18: predicate.replace_applicator
 0.67% :     0.000001s :      6: predicate.replace_old_param
 0.42% :     0.000001s :      3: predicate.reset_defer_inline
 1.15% :     0.000001s :     10: predicate.reshape_eliminate
 1.14% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.79% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.15% :     0.000001s :     10: predicate.same_eliminate
 0.47% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 0.91% :     0.000001s :      6: predicate.specialize_transform
 1.21% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.10% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.53% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 3.06% :     0.000004s :     16: predicate.switch_defer_inline
 1.98% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.57% :     0.000009s :     53: predicate.switch_simplify
 1.18% :     0.000001s :     10: predicate.tile_eliminate
 1.30% :     0.000002s :     10: predicate.transpose_eliminate
 1.58% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 4.50% :     0.000006s :     18: predicate.tuple_list_get_item_eliminator
 1.40% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.32% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.56% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.74% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.87% :     0.000002s :     10: predicate.value_based_eliminate
 0.41% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.64% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000689    12
47.97% :     0.000330s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.38% :     0.000044s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.66% :     0.000314s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.015940    72
 0.36% :     0.000058s :      1: add_recomputation
 0.47% :     0.000074s :      1: auto_monad
 0.11% :     0.000018s :      1: auto_monad_reorder
 2.12% :     0.000337s :      1: bootstrap
 0.21% :     0.000033s :      1: cconv
 0.07% :     0.000011s :      1: convert_after_rewriter
 0.17% :     0.000027s :      1: cse_after_recomputation
 0.06% :     0.000010s :      1: environ_conv
 0.14% :     0.000023s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
38.80% :     0.006184s :      1: jit_opt_a
 1.07% :     0.000170s :      1: jit_opt_after_cconv
 0.37% :     0.000059s :      1: jit_opt_b
 3.29% :     0.000525s :      1: loop_unroll
 4.41% :     0.000703s :      1: mutable_eliminate
 5.94% :     0.000947s :     26: opt.transform.jit_opt_a
 0.35% :     0.000055s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000027s :      4: opt.transform.jit_opt_b
 0.11% :     0.000017s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.16% :     0.000026s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.90% :     0.000622s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000041s :      1: py_interpret_to_execute
 0.13% :     0.000021s :      1: py_interpret_to_execute_after_opt_a
 0.11% :     0.000018s :      1: remove_dup_value
 3.27% :     0.000521s :      1: renormalize.infer
 2.78% :     0.000443s :      1: renormalize.specialize
 0.07% :     0.000011s :      1: rewriter_after_jit_bprop_graph
 0.29% :     0.000047s :      1: rewriter_after_opt_a
 0.50% :     0.000080s :      1: rewriter_before_opt_a
 0.55% :     0.000087s :      1: symbol_engine_optimizer
29.21% :     0.004657s :      1: type_inference


TotalTime = 0.0153282, [30]
[bootstrap]: 0.00038183
[type_inference]: 0.00562439
[event_method]: 1.978e-05
[auto_monad]: 6.999e-05
[graph_reusing]: 5.87999e-06
[pre_auto_parallel]: 3.56001e-06
[py_interpret_to_execute]: 4.114e-05
[rewriter_before_opt_a]: 7.977e-05
[expand_dump_flag]: 3.54002e-06
[jit_opt_a]: 0.00649615, [2]
    [Cycle 1]: 0.00206748, [27]
        [switch_simplify]: 6.288e-05
        [loop_unroll]: 2.988e-05
        [a_1]: 0.00066774
        [with_stream_mark]: 2.492e-05
        [recompute_prepare]: 9.94999e-06
        [updatestate_depend_eliminate]: 4.52998e-06
        [updatestate_assign_eliminate]: 3.04001e-06
        [updatestate_loads_eliminate]: 2.61e-06
        [parameter_eliminate]: 2.44001e-06
        [specialize_transform]: 7.38999e-06
        [updatestate_useless_node_eliminater]: 5.10001e-06
        [accelerated_algorithm]: 5.37001e-06
        [meta_shard_fg_expand]: 2.57001e-06
        [get_grad_eliminate_]: 5.52999e-06
        [merge_forward]: 4.35999e-06
        [cell_reuse_recompute_pass]: 1.35999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.593e-05
        [j_node_and_user_rematch]: 1.06e-05
        [meta_fg_expand]: 3.04999e-06
        [replace_old_param]: 1.047e-05
        [inline_without_move]: 5.90002e-06
        [renormalize]: 0.00091306
        [add_forward_monad_depend]: 6.74999e-06
        [auto_monad_grad]: 2.42001e-06
        [auto_monad_eliminator]: 1.636e-05
        [cse]: 3.122e-05
        [replace_applicator]: 1.35e-05
    [Cycle 2]: 0.00037757, [27]
        [switch_simplify]: 6.09999e-06
        [loop_unroll]: 5.74e-06
        [a_1]: 9.604e-05
        [with_stream_mark]: 1.448e-05
        [recompute_prepare]: 5.40001e-06
        [updatestate_depend_eliminate]: 3.07002e-06
        [updatestate_assign_eliminate]: 2.28998e-06
        [updatestate_loads_eliminate]: 2.42001e-06
        [parameter_eliminate]: 1.03001e-06
        [specialize_transform]: 5.10001e-06
        [updatestate_useless_node_eliminater]: 5.02e-06
        [accelerated_algorithm]: 6.06e-06
        [meta_shard_fg_expand]: 1.69e-06
        [get_grad_eliminate_]: 4.50001e-06
        [merge_forward]: 4.45999e-06
        [cell_reuse_recompute_pass]: 2.41998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.518e-05
        [j_node_and_user_rematch]: 8.32e-06
        [meta_fg_expand]: 1.94999e-06
        [replace_old_param]: 7.70998e-06
        [inline_without_move]: 1.686e-05
        [renormalize]: 6.99947e-08
        [add_forward_monad_depend]: 2.49001e-06
        [auto_monad_grad]: 1.05001e-06
        [auto_monad_eliminator]: 7.41999e-06
        [cse]: 1.58e-05
        [replace_applicator]: 7.9e-06
[py_interpret_to_execute_after_opt_a]: 1.654e-05
[rewriter_after_opt_a]: 4.25e-05
[convert_after_rewriter]: 8.72e-06
[order_py_execute_after_rewriter]: 5.94999e-06
[mutable_eliminate]: 0.00069117
[jit_opt_b]: 5.885e-05, [1]
    [Cycle 1]: 5.01e-05, [2]
        [frontend_op_eliminate]: 1.869e-05
        [inline_after_opt_a]: 1.736e-05
[cconv]: 3.002e-05
[loop_unroll]: 0.00051553
[jit_opt_after_cconv]: 0.00015873, [1]
    [Cycle 1]: 0.00015106, [11]
        [c_1]: 2.294e-05
        [parameter_eliminate]: 3.36001e-06
        [updatestate_depend_eliminate]: 7.66001e-06
        [updatestate_assign_eliminate]: 2.43e-06
        [updatestate_loads_eliminate]: 2.32999e-06
        [cse]: 2.435e-05
        [call_graph_tuple_transform]: 2.045e-05
        [tuple_list_get_item_eliminator]: 5.34e-06
        [none_parameter_eliminate]: 1.62999e-06
        [renormalize]: 7.80012e-07
        [switch_simplify]: 5.65001e-06
[remove_dup_value]: 1.545e-05
[partial_unused_args_eliminate]: 2.56e-06
[environ_conv]: 6.71e-06
[add_recomputation]: 5.013e-05
[cse_after_recomputation]: 2.209e-05, [1]
    [Cycle 1]: 1.595e-05, [1]
        [cse]: 9.00001e-06
[auto_monad_reorder]: 7.622e-05
[get_jit_bprop_graph]: 2.81e-06
[rewriter_after_jit_bprop_graph]: 7.88999e-06
[opt_after_jit_grad]: 0.00055366
[symbol_engine_optimizer]: 8.154e-05, [1]
    [Cycle 1]: 7.334e-05, [6]
        [build]: 4.67e-06
        [elim_shapecalc]: 7.93001e-06
        [elim_not_effective]: 1.541e-05
        [opt_reshape]: 6.29999e-06
        [fold_const_symbol]: 9.24998e-06
        [renormalize]: 5.10016e-07
[validate]: 4.129e-05
Sums
    bootstrap                            :     0.000382s :  3.60%
    type_inference                       :     0.005624s : 53.05%
    event_method                         :     0.000020s :  0.19%
    auto_monad                           :     0.000070s :  0.66%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000004s :  0.03%
    py_interpret_to_execute              :     0.000041s :  0.39%
    rewriter_before_opt_a                :     0.000080s :  0.75%
    expand_dump_flag                     :     0.000004s :  0.03%
    jit_opt_a.switch_simplify            :     0.000069s :  0.65%
    jit_opt_a.loop_unroll                :     0.000036s :  0.34%
    jit_opt_a.a_1                        :     0.000764s :  7.20%
    jit_opt_a.with_stream_mark           :     0.000039s :  0.37%
    jit_opt_a.recompute_prepare          :     0.000015s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.05%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000012s :  0.12%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.10%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.11%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.09%
    jit_opt_a.merge_forward              :     0.000009s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000004s :  0.04%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000031s :  0.29%
    jit_opt_a.j_node_and_user_rematch    :     0.000019s :  0.18%
    jit_opt_a.meta_fg_expand             :     0.000005s :  0.05%
    jit_opt_a.replace_old_param          :     0.000018s :  0.17%
    jit_opt_a.inline_without_move        :     0.000023s :  0.21%
    jit_opt_a.renormalize                :     0.000913s :  8.61%
    jit_opt_a.add_forward_monad_depend   :     0.000009s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.03%
    jit_opt_a.auto_monad_eliminator      :     0.000024s :  0.22%
    jit_opt_a.cse                        :     0.000047s :  0.44%
    jit_opt_a.replace_applicator         :     0.000021s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000017s :  0.16%
    rewriter_after_opt_a                 :     0.000043s :  0.40%
    convert_after_rewriter               :     0.000009s :  0.08%
    order_py_execute_after_rewriter      :     0.000006s :  0.06%
    mutable_eliminate                    :     0.000691s :  6.52%
    jit_opt_b.frontend_op_eliminate      :     0.000019s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000017s :  0.16%
    cconv                                :     0.000030s :  0.28%
    loop_unroll                          :     0.000516s :  4.86%
    jit_opt_after_cconv.c_1              :     0.000023s :  0.22%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000008s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000024s :  0.23%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.19%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.05%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000006s :  0.05%
    remove_dup_value                     :     0.000015s :  0.15%
    partial_unused_args_eliminate        :     0.000003s :  0.02%
    environ_conv                         :     0.000007s :  0.06%
    add_recomputation                    :     0.000050s :  0.47%
    cse_after_recomputation.cse          :     0.000009s :  0.08%
    auto_monad_reorder                   :     0.000076s :  0.72%
    get_jit_bprop_graph                  :     0.000003s :  0.03%
    rewriter_after_jit_bprop_graph       :     0.000008s :  0.07%
    opt_after_jit_grad                   :     0.000554s :  5.22%
    symbol_engine_optimizer.build        :     0.000005s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.07%
    symbol_engine_optimizer.elim_not_effective :     0.000015s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.09%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.00%
    validate                             :     0.000041s :  0.39%


Time group info:
------[substitution.]   0.000256    23
 1.04% :     0.000003s :      2: substitution.elim_not_effective
 0.49% :     0.000001s :      2: substitution.fold_const_symbol
 2.24% :     0.000006s :      3: substitution.graph_param_transform
84.93% :     0.000217s :      4: substitution.inline
 1.33% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.16% :     0.000006s :      4: substitution.remove_not_recompute_node
 2.18% :     0.000006s :      2: substitution.replace_old_param
 5.63% :     0.000014s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.005544     2
81.04% :     0.004492s :      1: type_inference.infer
18.96% :     0.001051s :      1: type_inference.specialize
------[replace.]   0.000082     6
64.92% :     0.000053s :      4: replace.inline
35.08% :     0.000029s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000227     6
94.14% :     0.000214s :      4: match.inline
 5.86% :     0.000013s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000131   691
 1.18% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.48% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 0.92% :     0.000001s :     10: predicate.addn_check_dump
 1.47% :     0.000002s :     10: predicate.addn_zero_filter
 2.45% :     0.000003s :     10: predicate.arithmetic_simplify
 1.26% :     0.000002s :     10: predicate.cast_eliminate
 0.45% :     0.000001s :      3: predicate.check_bprop_eliminate
 0.94% :     0.000001s :     10: predicate.compare_switch_simplify
 1.22% :     0.000002s :     10: predicate.depend_value_elim
 1.11% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.12% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.30% :     0.000002s :     10: predicate.dict_set_item_eliminator
 1.38% :     0.000002s :      3: predicate.dumpgradient_eliminate
 0.40% :     0.000001s :      3: predicate.elim_not_effective
 0.68% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.06% :     0.000001s :     10: predicate.environ_add_const_eliminate
 0.97% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.31% :     0.000002s :     10: predicate.environ_get_eliminate
 1.64% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.81% :     0.000001s :      6: predicate.get_grad_eliminate
 0.36% :     0.000000s :      3: predicate.graph_param_transform
 6.20% :     0.000008s :     22: predicate.inline
 1.12% :     0.000001s :      6: predicate.inline_without_move
 0.31% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.00% :     0.000001s :      6: predicate.less_batch_normalization
 1.41% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.68% :     0.000002s :     15: predicate.load_eliminater
 2.68% :     0.000004s :      3: predicate.loop_unroll_after_grad
 4.43% :     0.000006s :     34: predicate.loop_unroll_before_grad
 1.80% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 0.94% :     0.000001s :     10: predicate.merge_addn
 1.04% :     0.000001s :     10: predicate.minmaximum_grad
 2.75% :     0.000004s :      3: predicate.mutable_eliminate
 0.74% :     0.000001s :      3: predicate.opt_reshape
 2.08% :     0.000003s :     15: predicate.partial_eliminate
 1.16% :     0.000002s :     10: predicate.print_const_string_wrapper
 1.59% :     0.000002s :     10: predicate.reduce_eliminate
 1.35% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.77% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.51% :     0.000002s :     18: predicate.replace_applicator
 0.46% :     0.000001s :      6: predicate.replace_old_param
 0.43% :     0.000001s :      3: predicate.reset_defer_inline
 1.15% :     0.000002s :     10: predicate.reshape_eliminate
 1.17% :     0.000002s :     10: predicate.row_tensor_add_zeros_like
 1.06% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.27% :     0.000002s :     10: predicate.same_eliminate
 0.44% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.82% :     0.000001s :      6: predicate.special_op_eliminate
 0.93% :     0.000001s :      6: predicate.specialize_transform
 1.19% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.16% :     0.000002s :     10: predicate.stack_unstack_eliminate
 0.52% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.18% :     0.000003s :     16: predicate.switch_defer_inline
 1.84% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.19% :     0.000009s :     53: predicate.switch_simplify
 1.32% :     0.000002s :     10: predicate.tile_eliminate
 1.36% :     0.000002s :     10: predicate.transpose_eliminate
 1.23% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.25% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.52% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.25% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.72% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.49% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.90% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.60% :     0.000002s :     10: predicate.value_based_eliminate
 0.53% :     0.000001s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000789    12
50.36% :     0.000397s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 6.15% :     0.000049s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.49% :     0.000343s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.017304    72
 0.31% :     0.000053s :      1: add_recomputation
 0.43% :     0.000074s :      1: auto_monad
 0.47% :     0.000081s :      1: auto_monad_reorder
 2.35% :     0.000406s :      1: bootstrap
 0.19% :     0.000033s :      1: cconv
 0.06% :     0.000011s :      1: convert_after_rewriter
 0.14% :     0.000024s :      1: cse_after_recomputation
 0.06% :     0.000010s :      1: environ_conv
 0.14% :     0.000024s :      1: event_method
 0.03% :     0.000006s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
37.56% :     0.006500s :      1: jit_opt_a
 0.94% :     0.000162s :      1: jit_opt_after_cconv
 0.36% :     0.000062s :      1: jit_opt_b
 3.04% :     0.000526s :      1: loop_unroll
 4.06% :     0.000702s :      1: mutable_eliminate
 5.83% :     0.001009s :     26: opt.transform.jit_opt_a
 0.29% :     0.000051s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000029s :      4: opt.transform.jit_opt_b
 0.09% :     0.000016s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000019s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000024s :      1: opt.transform.opt_after_jit_grad
 0.21% :     0.000036s :      4: opt.transform.symbol_engine_opt
 3.26% :     0.000563s :      1: opt_after_jit_grad
 0.05% :     0.000008s :      1: order_py_execute_after_rewriter
 0.03% :     0.000005s :      1: partial_unused_args_eliminate
 0.03% :     0.000006s :      1: pre_auto_parallel
 0.25% :     0.000044s :      1: py_interpret_to_execute
 0.12% :     0.000020s :      1: py_interpret_to_execute_after_opt_a
 0.10% :     0.000018s :      1: remove_dup_value
 2.88% :     0.000498s :      1: renormalize.infer
 2.34% :     0.000404s :      1: renormalize.specialize
 0.06% :     0.000010s :      1: rewriter_after_jit_bprop_graph
 0.27% :     0.000046s :      1: rewriter_after_opt_a
 0.48% :     0.000083s :      1: rewriter_before_opt_a
 0.49% :     0.000085s :      1: symbol_engine_optimizer
32.61% :     0.005643s :      1: type_inference


TotalTime = 0.0121522, [30]
[bootstrap]: 0.00028498
[type_inference]: 0.00430117
[event_method]: 1.64e-05
[auto_monad]: 6.344e-05
[graph_reusing]: 6.16e-06
[pre_auto_parallel]: 2.86e-06
[py_interpret_to_execute]: 3.238e-05
[rewriter_before_opt_a]: 7.024e-05
[expand_dump_flag]: 3.16999e-06
[jit_opt_a]: 0.00529494, [2]
    [Cycle 1]: 0.00155638, [27]
        [switch_simplify]: 5.361e-05
        [loop_unroll]: 2.872e-05
        [a_1]: 0.00051334
        [with_stream_mark]: 1.624e-05
        [recompute_prepare]: 6.64999e-06
        [updatestate_depend_eliminate]: 3.45998e-06
        [updatestate_assign_eliminate]: 2.81e-06
        [updatestate_loads_eliminate]: 2.66999e-06
        [parameter_eliminate]: 1.79e-06
        [specialize_transform]: 5.82999e-06
        [updatestate_useless_node_eliminater]: 4.79002e-06
        [accelerated_algorithm]: 5.57001e-06
        [meta_shard_fg_expand]: 1.87001e-06
        [get_grad_eliminate_]: 5.66e-06
        [merge_forward]: 3.68e-06
        [cell_reuse_recompute_pass]: 1.01002e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.4e-05
        [j_node_and_user_rematch]: 9.47999e-06
        [meta_fg_expand]: 2.31e-06
        [replace_old_param]: 9.02e-06
        [inline_without_move]: 5.31002e-06
        [renormalize]: 0.00063197
        [add_forward_monad_depend]: 5.79e-06
        [auto_monad_grad]: 2.07999e-06
        [auto_monad_eliminator]: 1.257e-05
        [cse]: 2.566e-05
        [replace_applicator]: 1.154e-05
    [Cycle 2]: 0.00031333, [27]
        [switch_simplify]: 5.37001e-06
        [loop_unroll]: 5.02e-06
        [a_1]: 8.833e-05
        [with_stream_mark]: 8.78001e-06
        [recompute_prepare]: 5.04e-06
        [updatestate_depend_eliminate]: 2.46e-06
        [updatestate_assign_eliminate]: 1.99e-06
        [updatestate_loads_eliminate]: 1.94e-06
        [parameter_eliminate]: 9.89996e-07
        [specialize_transform]: 4.92e-06
        [updatestate_useless_node_eliminater]: 5.09e-06
        [accelerated_algorithm]: 4.90999e-06
        [meta_shard_fg_expand]: 1.09e-06
        [get_grad_eliminate_]: 4.34997e-06
        [merge_forward]: 2.51e-06
        [cell_reuse_recompute_pass]: 1.44998e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.184e-05
        [j_node_and_user_rematch]: 7.20998e-06
        [meta_fg_expand]: 1.54e-06
        [replace_old_param]: 7.13e-06
        [inline_without_move]: 4.65001e-06
        [renormalize]: 7.00238e-08
        [add_forward_monad_depend]: 1.15001e-06
        [auto_monad_grad]: 8.50006e-07
        [auto_monad_eliminator]: 5.17e-06
        [cse]: 9.70002e-06
        [replace_applicator]: 4.86002e-06
[py_interpret_to_execute_after_opt_a]: 9.64999e-06
[rewriter_after_opt_a]: 3.514e-05
[convert_after_rewriter]: 7.06001e-06
[order_py_execute_after_rewriter]: 4.97e-06
[mutable_eliminate]: 0.00050589
[jit_opt_b]: 4.907e-05, [1]
    [Cycle 1]: 4.207e-05, [2]
        [frontend_op_eliminate]: 1.619e-05
        [inline_after_opt_a]: 1.469e-05
[cconv]: 2.509e-05
[loop_unroll]: 0.00041174
[jit_opt_after_cconv]: 0.00013616, [1]
    [Cycle 1]: 0.00012984, [11]
        [c_1]: 2.092e-05
        [parameter_eliminate]: 2.61e-06
        [updatestate_depend_eliminate]: 5.61e-06
        [updatestate_assign_eliminate]: 2.16e-06
        [updatestate_loads_eliminate]: 2.41e-06
        [cse]: 1.764e-05
        [call_graph_tuple_transform]: 1.75e-05
        [tuple_list_get_item_eliminator]: 5.20001e-06
        [none_parameter_eliminate]: 1.35001e-06
        [renormalize]: 5.8001e-07
        [switch_simplify]: 5.07e-06
[remove_dup_value]: 1.387e-05
[partial_unused_args_eliminate]: 2.15002e-06
[environ_conv]: 5.05999e-06
[add_recomputation]: 0.00010186
[cse_after_recomputation]: 2.159e-05, [1]
    [Cycle 1]: 1.589e-05, [1]
        [cse]: 9.67001e-06
[auto_monad_reorder]: 1.541e-05
[get_jit_bprop_graph]: 1.75001e-06
[rewriter_after_jit_bprop_graph]: 5.13002e-06
[opt_after_jit_grad]: 0.00044975
[symbol_engine_optimizer]: 6.941e-05, [1]
    [Cycle 1]: 6.309e-05, [6]
        [build]: 3.11001e-06
        [elim_shapecalc]: 7.51999e-06
        [elim_not_effective]: 1.203e-05
        [opt_reshape]: 5.12e-06
        [fold_const_symbol]: 8.03001e-06
        [renormalize]: 4.89992e-07
[validate]: 3.41e-05
Sums
    bootstrap                            :     0.000285s :  3.50%
    type_inference                       :     0.004301s : 52.75%
    event_method                         :     0.000016s :  0.20%
    auto_monad                           :     0.000063s :  0.78%
    graph_reusing                        :     0.000006s :  0.08%
    pre_auto_parallel                    :     0.000003s :  0.04%
    py_interpret_to_execute              :     0.000032s :  0.40%
    rewriter_before_opt_a                :     0.000070s :  0.86%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000059s :  0.72%
    jit_opt_a.loop_unroll                :     0.000034s :  0.41%
    jit_opt_a.a_1                        :     0.000602s :  7.38%
    jit_opt_a.with_stream_mark           :     0.000025s :  0.31%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000010s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000002s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000026s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.20%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000016s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000632s :  7.75%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000018s :  0.22%
    jit_opt_a.cse                        :     0.000035s :  0.43%
    jit_opt_a.replace_applicator         :     0.000016s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000010s :  0.12%
    rewriter_after_opt_a                 :     0.000035s :  0.43%
    convert_after_rewriter               :     0.000007s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000506s :  6.20%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.20%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.18%
    cconv                                :     0.000025s :  0.31%
    loop_unroll                          :     0.000412s :  5.05%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.26%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000018s :  0.22%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000017s :  0.21%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000001s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000014s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000005s :  0.06%
    add_recomputation                    :     0.000102s :  1.25%
    cse_after_recomputation.cse          :     0.000010s :  0.12%
    auto_monad_reorder                   :     0.000015s :  0.19%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000450s :  5.52%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000012s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000005s :  0.06%
    symbol_engine_optimizer.fold_const_symbol :     0.000008s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.01%
    validate                             :     0.000034s :  0.42%


Time group info:
------[substitution.]   0.000189    23
 0.97% :     0.000002s :      2: substitution.elim_not_effective
 0.68% :     0.000001s :      2: substitution.fold_const_symbol
 2.61% :     0.000005s :      3: substitution.graph_param_transform
83.07% :     0.000157s :      4: substitution.inline
 1.83% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.34% :     0.000004s :      4: substitution.remove_not_recompute_node
 2.15% :     0.000004s :      2: substitution.replace_old_param
 6.36% :     0.000012s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004242     2
80.51% :     0.003415s :      1: type_inference.infer
19.49% :     0.000827s :      1: type_inference.specialize
------[replace.]   0.000066     6
67.35% :     0.000044s :      4: replace.inline
32.65% :     0.000021s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000166     6
93.37% :     0.000155s :      4: match.inline
 6.63% :     0.000011s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000112   691
 1.34% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.52% :     0.000002s :      3: predicate.ad_related_special_op_eliminate
 1.09% :     0.000001s :     10: predicate.addn_check_dump
 1.33% :     0.000001s :     10: predicate.addn_zero_filter
 1.82% :     0.000002s :     10: predicate.arithmetic_simplify
 1.32% :     0.000001s :     10: predicate.cast_eliminate
 0.54% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.07% :     0.000001s :     10: predicate.compare_switch_simplify
 1.32% :     0.000001s :     10: predicate.depend_value_elim
 1.18% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.26% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.24% :     0.000001s :     10: predicate.dict_set_item_eliminator
 1.03% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.35% :     0.000000s :      3: predicate.elim_not_effective
 0.78% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.19% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.11% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.25% :     0.000001s :     10: predicate.environ_get_eliminate
 1.10% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.21% :     0.000000s :      3: predicate.fold_const_symbol
 0.95% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.82% :     0.000007s :     22: predicate.inline
 0.98% :     0.000001s :      6: predicate.inline_without_move
 0.37% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.12% :     0.000001s :      6: predicate.less_batch_normalization
 1.92% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.75% :     0.000002s :     15: predicate.load_eliminater
 1.53% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.98% :     0.000004s :     34: predicate.loop_unroll_before_grad
 1.87% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.10% :     0.000001s :     10: predicate.merge_addn
 1.10% :     0.000001s :     10: predicate.minmaximum_grad
 1.59% :     0.000002s :      3: predicate.mutable_eliminate
 0.49% :     0.000001s :      3: predicate.opt_reshape
 2.28% :     0.000003s :     15: predicate.partial_eliminate
 1.09% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.60% :     0.000002s :     10: predicate.reduce_eliminate
 1.51% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.53% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.72% :     0.000002s :     18: predicate.replace_applicator
 0.99% :     0.000001s :      6: predicate.replace_old_param
 0.33% :     0.000000s :      3: predicate.reset_defer_inline
 1.29% :     0.000001s :     10: predicate.reshape_eliminate
 1.17% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.77% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.32% :     0.000001s :     10: predicate.same_eliminate
 0.53% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.97% :     0.000001s :      6: predicate.special_op_eliminate
 1.01% :     0.000001s :      6: predicate.specialize_transform
 1.34% :     0.000002s :     10: predicate.split_environ_get_set_with_tuple_value
 1.18% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.44% :     0.000000s :      3: predicate.switch_call_monad_eliminater
 2.37% :     0.000003s :     16: predicate.switch_defer_inline
 2.21% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.78% :     0.000009s :     53: predicate.switch_simplify
 1.27% :     0.000001s :     10: predicate.tile_eliminate
 1.30% :     0.000001s :     10: predicate.transpose_eliminate
 1.44% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.19% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 3.99% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.51% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.69% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.66% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.95% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.53% :     0.000002s :     10: predicate.value_based_eliminate
 0.44% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.60% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000593    12
50.69% :     0.000301s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 4.91% :     0.000029s :      1: func_graph_cloner_run.FuncGraphClonerNode
44.40% :     0.000263s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.013639    72
 0.77% :     0.000105s :      1: add_recomputation
 0.49% :     0.000067s :      1: auto_monad
 0.13% :     0.000018s :      1: auto_monad_reorder
 2.21% :     0.000301s :      1: bootstrap
 0.20% :     0.000028s :      1: cconv
 0.07% :     0.000009s :      1: convert_after_rewriter
 0.18% :     0.000024s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.15% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.06% :     0.000009s :      1: graph_reusing
38.84% :     0.005298s :      1: jit_opt_a
 1.02% :     0.000139s :      1: jit_opt_after_cconv
 0.38% :     0.000052s :      1: jit_opt_b
 3.07% :     0.000419s :      1: loop_unroll
 3.77% :     0.000515s :      1: mutable_eliminate
 5.90% :     0.000805s :     26: opt.transform.jit_opt_a
 0.33% :     0.000045s :      4: opt.transform.jit_opt_after_cconv
 0.18% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000012s :      1: opt.transform.loop_unroll_optimizer
 0.12% :     0.000017s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000030s :      4: opt.transform.symbol_engine_opt
 3.35% :     0.000457s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.04% :     0.000005s :      1: pre_auto_parallel
 0.26% :     0.000035s :      1: py_interpret_to_execute
 0.09% :     0.000012s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000016s :      1: remove_dup_value
 2.30% :     0.000314s :      1: renormalize.infer
 2.28% :     0.000311s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.28% :     0.000038s :      1: rewriter_after_opt_a
 0.53% :     0.000073s :      1: rewriter_before_opt_a
 0.53% :     0.000072s :      1: symbol_engine_optimizer
31.64% :     0.004315s :      1: type_inference


TotalTime = 0.0129742, [30]
[bootstrap]: 0.00034048
[type_inference]: 0.00445645
[event_method]: 1.829e-05
[auto_monad]: 6.602e-05
[graph_reusing]: 5.51002e-06
[pre_auto_parallel]: 2.22999e-06
[py_interpret_to_execute]: 3.306e-05
[rewriter_before_opt_a]: 7.243e-05
[expand_dump_flag]: 3.02002e-06
[jit_opt_a]: 0.00576352, [2]
    [Cycle 1]: 0.00181875, [27]
        [switch_simplify]: 5.889e-05
        [loop_unroll]: 2.911e-05
        [a_1]: 0.00056904
        [with_stream_mark]: 2.396e-05
        [recompute_prepare]: 8.13999e-06
        [updatestate_depend_eliminate]: 4.12998e-06
        [updatestate_assign_eliminate]: 3.13998e-06
        [updatestate_loads_eliminate]: 2.97002e-06
        [parameter_eliminate]: 1.72999e-06
        [specialize_transform]: 6.34001e-06
        [updatestate_useless_node_eliminater]: 5.30001e-06
        [accelerated_algorithm]: 6.44001e-06
        [meta_shard_fg_expand]: 1.92999e-06
        [get_grad_eliminate_]: 5.75001e-06
        [merge_forward]: 3.70998e-06
        [cell_reuse_recompute_pass]: 1.65001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.568e-05
        [j_node_and_user_rematch]: 1.194e-05
        [meta_fg_expand]: 2.74999e-06
        [replace_old_param]: 1.014e-05
        [inline_without_move]: 6.44999e-06
        [renormalize]: 0.0007858
        [add_forward_monad_depend]: 5.81998e-06
        [auto_monad_grad]: 2.48e-06
        [auto_monad_eliminator]: 1.293e-05
        [cse]: 3.683e-05
        [replace_applicator]: 1.241e-05
    [Cycle 2]: 0.00033088, [27]
        [switch_simplify]: 5.31002e-06
        [loop_unroll]: 5.09e-06
        [a_1]: 8.886e-05
        [with_stream_mark]: 9.66e-06
        [recompute_prepare]: 5.11002e-06
        [updatestate_depend_eliminate]: 2.42001e-06
        [updatestate_assign_eliminate]: 2.12999e-06
        [updatestate_loads_eliminate]: 1.89e-06
        [parameter_eliminate]: 1.15001e-06
        [specialize_transform]: 4.99e-06
        [updatestate_useless_node_eliminater]: 4.82e-06
        [accelerated_algorithm]: 5.04e-06
        [meta_shard_fg_expand]: 1.29e-06
        [get_grad_eliminate_]: 4.45e-06
        [merge_forward]: 2.56998e-06
        [cell_reuse_recompute_pass]: 1.30001e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.214e-05
        [j_node_and_user_rematch]: 8.15e-06
        [meta_fg_expand]: 1.72999e-06
        [replace_old_param]: 7.46999e-06
        [inline_without_move]: 4.52e-06
        [renormalize]: 7.99773e-08
        [add_forward_monad_depend]: 1.39e-06
        [auto_monad_grad]: 7.39994e-07
        [auto_monad_eliminator]: 6.14001e-06
        [cse]: 1.049e-05
        [replace_applicator]: 5.15999e-06
[py_interpret_to_execute_after_opt_a]: 1.09e-05
[rewriter_after_opt_a]: 3.429e-05
[convert_after_rewriter]: 6.72002e-06
[order_py_execute_after_rewriter]: 7.38e-06
[mutable_eliminate]: 0.00057898
[jit_opt_b]: 5.166e-05, [1]
    [Cycle 1]: 4.328e-05, [2]
        [frontend_op_eliminate]: 1.619e-05
        [inline_after_opt_a]: 1.494e-05
[cconv]: 2.645e-05
[loop_unroll]: 0.00050353
[jit_opt_after_cconv]: 0.00014416, [1]
    [Cycle 1]: 0.00013672, [11]
        [c_1]: 2.143e-05
        [parameter_eliminate]: 2.79001e-06
        [updatestate_depend_eliminate]: 6.33e-06
        [updatestate_assign_eliminate]: 2.63e-06
        [updatestate_loads_eliminate]: 2.37999e-06
        [cse]: 1.827e-05
        [call_graph_tuple_transform]: 2.017e-05
        [tuple_list_get_item_eliminator]: 5.34e-06
        [none_parameter_eliminate]: 1.60999e-06
        [renormalize]: 3.19997e-07
        [switch_simplify]: 5.07999e-06
[remove_dup_value]: 1.463e-05
[partial_unused_args_eliminate]: 2.09999e-06
[environ_conv]: 4.70999e-06
[add_recomputation]: 4.85e-05
[cse_after_recomputation]: 2.013e-05, [1]
    [Cycle 1]: 1.449e-05, [1]
        [cse]: 8.82999e-06
[auto_monad_reorder]: 1.507e-05
[get_jit_bprop_graph]: 2.02001e-06
[rewriter_after_jit_bprop_graph]: 4.81997e-06
[opt_after_jit_grad]: 0.000458
[symbol_engine_optimizer]: 7.316e-05, [1]
    [Cycle 1]: 6.603e-05, [6]
        [build]: 3.41999e-06
        [elim_shapecalc]: 7.26999e-06
        [elim_not_effective]: 1.299e-05
        [opt_reshape]: 6.19001e-06
        [fold_const_symbol]: 8.82e-06
        [renormalize]: 3.19997e-07
[validate]: 3.466e-05
Sums
    bootstrap                            :     0.000340s :  3.89%
    type_inference                       :     0.004456s : 50.90%
    event_method                         :     0.000018s :  0.21%
    auto_monad                           :     0.000066s :  0.75%
    graph_reusing                        :     0.000006s :  0.06%
    pre_auto_parallel                    :     0.000002s :  0.03%
    py_interpret_to_execute              :     0.000033s :  0.38%
    rewriter_before_opt_a                :     0.000072s :  0.83%
    expand_dump_flag                     :     0.000003s :  0.03%
    jit_opt_a.switch_simplify            :     0.000064s :  0.73%
    jit_opt_a.loop_unroll                :     0.000034s :  0.39%
    jit_opt_a.a_1                        :     0.000658s :  7.51%
    jit_opt_a.with_stream_mark           :     0.000034s :  0.38%
    jit_opt_a.recompute_prepare          :     0.000013s :  0.15%
    jit_opt_a.updatestate_depend_eliminate :     0.000007s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000005s :  0.06%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.03%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.13%
    jit_opt_a.meta_shard_fg_expand       :     0.000003s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.12%
    jit_opt_a.merge_forward              :     0.000006s :  0.07%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000020s :  0.23%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000018s :  0.20%
    jit_opt_a.inline_without_move        :     0.000011s :  0.13%
    jit_opt_a.renormalize                :     0.000786s :  8.98%
    jit_opt_a.add_forward_monad_depend   :     0.000007s :  0.08%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.22%
    jit_opt_a.cse                        :     0.000047s :  0.54%
    jit_opt_a.replace_applicator         :     0.000018s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000011s :  0.12%
    rewriter_after_opt_a                 :     0.000034s :  0.39%
    convert_after_rewriter               :     0.000007s :  0.08%
    order_py_execute_after_rewriter      :     0.000007s :  0.08%
    mutable_eliminate                    :     0.000579s :  6.61%
    jit_opt_b.frontend_op_eliminate      :     0.000016s :  0.18%
    jit_opt_b.inline_after_opt_a         :     0.000015s :  0.17%
    cconv                                :     0.000026s :  0.30%
    loop_unroll                          :     0.000504s :  5.75%
    jit_opt_after_cconv.c_1              :     0.000021s :  0.24%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000003s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.cse              :     0.000018s :  0.21%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000000s :  0.00%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000015s :  0.17%
    partial_unused_args_eliminate        :     0.000002s :  0.02%
    environ_conv                         :     0.000005s :  0.05%
    add_recomputation                    :     0.000048s :  0.55%
    cse_after_recomputation.cse          :     0.000009s :  0.10%
    auto_monad_reorder                   :     0.000015s :  0.17%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000005s :  0.06%
    opt_after_jit_grad                   :     0.000458s :  5.23%
    symbol_engine_optimizer.build        :     0.000003s :  0.04%
    symbol_engine_optimizer.elim_shapecalc :     0.000007s :  0.08%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000000s :  0.00%
    validate                             :     0.000035s :  0.40%


Time group info:
------[substitution.]   0.000218    23
 0.83% :     0.000002s :      2: substitution.elim_not_effective
 0.66% :     0.000001s :      2: substitution.fold_const_symbol
 2.81% :     0.000006s :      3: substitution.graph_param_transform
83.58% :     0.000182s :      4: substitution.inline
 1.74% :     0.000004s :      4: substitution.j_node_and_user_rematch
 2.20% :     0.000005s :      4: substitution.remove_not_recompute_node
 2.18% :     0.000005s :      2: substitution.replace_old_param
 6.00% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004393     2
79.26% :     0.003482s :      1: type_inference.infer
20.74% :     0.000911s :      1: type_inference.specialize
------[replace.]   0.000073     6
66.99% :     0.000049s :      4: replace.inline
33.01% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000192     6
93.75% :     0.000180s :      4: match.inline
 6.25% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000120   691
 1.39% :     0.000002s :     10: predicate.accumulaten_eliminater
 1.18% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.05% :     0.000001s :     10: predicate.addn_check_dump
 1.53% :     0.000002s :     10: predicate.addn_zero_filter
 2.14% :     0.000003s :     10: predicate.arithmetic_simplify
 1.21% :     0.000001s :     10: predicate.cast_eliminate
 0.43% :     0.000001s :      3: predicate.check_bprop_eliminate
 1.05% :     0.000001s :     10: predicate.compare_switch_simplify
 1.90% :     0.000002s :     10: predicate.depend_value_elim
 1.08% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.22% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.31% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.98% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.28% :     0.000000s :      3: predicate.elim_not_effective
 0.72% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.22% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.16% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.08% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.20% :     0.000001s :     10: predicate.environ_get_eliminate
 1.41% :     0.000002s :     10: predicate.environ_get_set_eliminate
 0.22% :     0.000000s :      3: predicate.fold_const_symbol
 0.97% :     0.000001s :      6: predicate.get_grad_eliminate
 0.28% :     0.000000s :      3: predicate.graph_param_transform
 5.48% :     0.000007s :     22: predicate.inline
 1.08% :     0.000001s :      6: predicate.inline_without_move
 0.38% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.28% :     0.000002s :      6: predicate.less_batch_normalization
 1.47% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.73% :     0.000002s :     15: predicate.load_eliminater
 1.48% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.82% :     0.000005s :     34: predicate.loop_unroll_before_grad
 1.89% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.00% :     0.000001s :     10: predicate.minmaximum_grad
 1.74% :     0.000002s :      3: predicate.mutable_eliminate
 0.55% :     0.000001s :      3: predicate.opt_reshape
 2.18% :     0.000003s :     15: predicate.partial_eliminate
 1.17% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.57% :     0.000002s :     10: predicate.reduce_eliminate
 1.43% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.48% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.89% :     0.000001s :      6: predicate.replace_old_param
 0.32% :     0.000000s :      3: predicate.reset_defer_inline
 1.70% :     0.000002s :     10: predicate.reshape_eliminate
 1.15% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 1.00% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.54% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 1.03% :     0.000001s :      6: predicate.special_op_eliminate
 1.08% :     0.000001s :      6: predicate.specialize_transform
 1.18% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.23% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.42% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.44% :     0.000003s :     16: predicate.switch_defer_inline
 2.12% :     0.000003s :     16: predicate.switch_layer_defer_inline
 7.73% :     0.000009s :     53: predicate.switch_simplify
 1.33% :     0.000002s :     10: predicate.tile_eliminate
 1.53% :     0.000002s :     10: predicate.transpose_eliminate
 1.37% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.15% :     0.000001s :     10: predicate.tuple_list_get_item_depend_reorder
 4.18% :     0.000005s :     18: predicate.tuple_list_get_item_eliminator
 1.35% :     0.000002s :     10: predicate.tuple_list_set_item_eliminator
 1.48% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.64% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.94% :     0.000004s :     21: predicate.updatestate_useless_node_eliminater
 1.58% :     0.000002s :     10: predicate.value_based_eliminate
 0.32% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.68% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000634    12
50.96% :     0.000323s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 5.07% :     0.000032s :      1: func_graph_cloner_run.FuncGraphClonerNode
43.96% :     0.000279s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014694    72
 0.35% :     0.000051s :      1: add_recomputation
 0.47% :     0.000070s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.42% :     0.000356s :      1: bootstrap
 0.20% :     0.000029s :      1: cconv
 0.06% :     0.000009s :      1: convert_after_rewriter
 0.15% :     0.000022s :      1: cse_after_recomputation
 0.05% :     0.000008s :      1: environ_conv
 0.15% :     0.000022s :      1: event_method
 0.03% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000004s :      1: get_jit_bprop_graph
 0.05% :     0.000008s :      1: graph_reusing
39.25% :     0.005767s :      1: jit_opt_a
 1.00% :     0.000147s :      1: jit_opt_after_cconv
 0.37% :     0.000054s :      1: jit_opt_b
 3.49% :     0.000512s :      1: loop_unroll
 4.00% :     0.000587s :      1: mutable_eliminate
 5.97% :     0.000878s :     26: opt.transform.jit_opt_a
 0.33% :     0.000049s :      4: opt.transform.jit_opt_after_cconv
 0.17% :     0.000025s :      4: opt.transform.jit_opt_b
 0.09% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.10% :     0.000014s :      1: opt.transform.mutable_eliminate
 0.14% :     0.000021s :      1: opt.transform.opt_after_jit_grad
 0.22% :     0.000032s :      4: opt.transform.symbol_engine_opt
 3.17% :     0.000466s :      1: opt_after_jit_grad
 0.06% :     0.000010s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000004s :      1: pre_auto_parallel
 0.24% :     0.000036s :      1: py_interpret_to_execute
 0.09% :     0.000013s :      1: py_interpret_to_execute_after_opt_a
 0.12% :     0.000017s :      1: remove_dup_value
 2.82% :     0.000414s :      1: renormalize.infer
 2.47% :     0.000363s :      1: renormalize.specialize
 0.05% :     0.000007s :      1: rewriter_after_jit_bprop_graph
 0.25% :     0.000037s :      1: rewriter_after_opt_a
 0.51% :     0.000075s :      1: rewriter_before_opt_a
 0.51% :     0.000076s :      1: symbol_engine_optimizer
30.43% :     0.004471s :      1: type_inference


TotalTime = 0.012916, [30]
[bootstrap]: 0.00032121
[type_inference]: 0.00438918
[event_method]: 1.651e-05
[auto_monad]: 6.454e-05
[graph_reusing]: 5.97001e-06
[pre_auto_parallel]: 2.71e-06
[py_interpret_to_execute]: 3.144e-05
[rewriter_before_opt_a]: 7.222e-05
[expand_dump_flag]: 3.42002e-06
[jit_opt_a]: 0.00573454, [2]
    [Cycle 1]: 0.00180096, [27]
        [switch_simplify]: 5.68e-05
        [loop_unroll]: 2.921e-05
        [a_1]: 0.0005574
        [with_stream_mark]: 2.049e-05
        [recompute_prepare]: 7.07002e-06
        [updatestate_depend_eliminate]: 3.86001e-06
        [updatestate_assign_eliminate]: 2.64001e-06
        [updatestate_loads_eliminate]: 2.38998e-06
        [parameter_eliminate]: 2.02001e-06
        [specialize_transform]: 6.13002e-06
        [updatestate_useless_node_eliminater]: 5.19998e-06
        [accelerated_algorithm]: 5.69999e-06
        [meta_shard_fg_expand]: 2.44999e-06
        [get_grad_eliminate_]: 5.28002e-06
        [merge_forward]: 4.03001e-06
        [cell_reuse_recompute_pass]: 1.07e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.415e-05
        [j_node_and_user_rematch]: 9.15999e-06
        [meta_fg_expand]: 2.27999e-06
        [replace_old_param]: 9.39e-06
        [inline_without_move]: 5.39998e-06
        [renormalize]: 0.00081219
        [add_forward_monad_depend]: 6.44001e-06
        [auto_monad_grad]: 2.25002e-06
        [auto_monad_eliminator]: 1.39e-05
        [cse]: 2.846e-05
        [replace_applicator]: 1.214e-05
    [Cycle 2]: 0.0003249, [27]
        [switch_simplify]: 5.92001e-06
        [loop_unroll]: 5.08002e-06
        [a_1]: 9.02e-05
        [with_stream_mark]: 9.74e-06
        [recompute_prepare]: 5.22e-06
        [updatestate_depend_eliminate]: 2.52001e-06
        [updatestate_assign_eliminate]: 2.24001e-06
        [updatestate_loads_eliminate]: 1.87999e-06
        [parameter_eliminate]: 1.36002e-06
        [specialize_transform]: 4.93001e-06
        [updatestate_useless_node_eliminater]: 4.89e-06
        [accelerated_algorithm]: 4.99e-06
        [meta_shard_fg_expand]: 1.39e-06
        [get_grad_eliminate_]: 4.56002e-06
        [merge_forward]: 2.56e-06
        [cell_reuse_recompute_pass]: 1.52999e-06
        [cell_reuse_handle_not_recompute_node_pass]: 1.344e-05
        [j_node_and_user_rematch]: 7.6e-06
        [meta_fg_expand]: 1.92001e-06
        [replace_old_param]: 7.66999e-06
        [inline_without_move]: 4.65999e-06
        [renormalize]: 8.00064e-08
        [add_forward_monad_depend]: 1.62999e-06
        [auto_monad_grad]: 9.80013e-07
        [auto_monad_eliminator]: 5.52001e-06
        [cse]: 1.047e-05
        [replace_applicator]: 5.57001e-06
[py_interpret_to_execute_after_opt_a]: 1.166e-05
[rewriter_after_opt_a]: 3.514e-05
[convert_after_rewriter]: 7.56999e-06
[order_py_execute_after_rewriter]: 4.94e-06
[mutable_eliminate]: 0.00064434
[jit_opt_b]: 5.358e-05, [1]
    [Cycle 1]: 4.607e-05, [2]
        [frontend_op_eliminate]: 1.788e-05
        [inline_after_opt_a]: 1.622e-05
[cconv]: 2.727e-05
[loop_unroll]: 0.00044632
[jit_opt_after_cconv]: 0.00014703, [1]
    [Cycle 1]: 0.00014038, [11]
        [c_1]: 2.158e-05
        [parameter_eliminate]: 3.41001e-06
        [updatestate_depend_eliminate]: 6.78e-06
        [updatestate_assign_eliminate]: 2.45997e-06
        [updatestate_loads_eliminate]: 2.04999e-06
        [cse]: 2.181e-05
        [call_graph_tuple_transform]: 1.969e-05
        [tuple_list_get_item_eliminator]: 5.04e-06
        [none_parameter_eliminate]: 1.72999e-06
        [renormalize]: 5.39992e-07
        [switch_simplify]: 5.40001e-06
[remove_dup_value]: 1.591e-05
[partial_unused_args_eliminate]: 2.31e-06
[environ_conv]: 5.72001e-06
[add_recomputation]: 5.123e-05
[cse_after_recomputation]: 2.28e-05, [1]
    [Cycle 1]: 1.624e-05, [1]
        [cse]: 1.004e-05
[auto_monad_reorder]: 1.571e-05
[get_jit_bprop_graph]: 1.97999e-06
[rewriter_after_jit_bprop_graph]: 5.67999e-06
[opt_after_jit_grad]: 0.00048187
[symbol_engine_optimizer]: 7.824e-05, [1]
    [Cycle 1]: 7.065e-05, [6]
        [build]: 4.43001e-06
        [elim_shapecalc]: 8.23001e-06
        [elim_not_effective]: 1.333e-05
        [opt_reshape]: 6.24999e-06
        [fold_const_symbol]: 9.07999e-06
        [renormalize]: 7.39994e-07
[validate]: 4.188e-05
Sums
    bootstrap                            :     0.000321s :  3.68%
    type_inference                       :     0.004389s : 50.34%
    event_method                         :     0.000017s :  0.19%
    auto_monad                           :     0.000065s :  0.74%
    graph_reusing                        :     0.000006s :  0.07%
    pre_auto_parallel                    :     0.000003s :  0.03%
    py_interpret_to_execute              :     0.000031s :  0.36%
    rewriter_before_opt_a                :     0.000072s :  0.83%
    expand_dump_flag                     :     0.000003s :  0.04%
    jit_opt_a.switch_simplify            :     0.000063s :  0.72%
    jit_opt_a.loop_unroll                :     0.000034s :  0.39%
    jit_opt_a.a_1                        :     0.000648s :  7.43%
    jit_opt_a.with_stream_mark           :     0.000030s :  0.35%
    jit_opt_a.recompute_prepare          :     0.000012s :  0.14%
    jit_opt_a.updatestate_depend_eliminate :     0.000006s :  0.07%
    jit_opt_a.updatestate_assign_eliminate :     0.000005s :  0.06%
    jit_opt_a.updatestate_loads_eliminate :     0.000004s :  0.05%
    jit_opt_a.parameter_eliminate        :     0.000003s :  0.04%
    jit_opt_a.specialize_transform       :     0.000011s :  0.13%
    jit_opt_a.updatestate_useless_node_eliminater :     0.000010s :  0.12%
    jit_opt_a.accelerated_algorithm      :     0.000011s :  0.12%
    jit_opt_a.meta_shard_fg_expand       :     0.000004s :  0.04%
    jit_opt_a.get_grad_eliminate_        :     0.000010s :  0.11%
    jit_opt_a.merge_forward              :     0.000007s :  0.08%
    jit_opt_a.cell_reuse_recompute_pass  :     0.000003s :  0.03%
    jit_opt_a.cell_reuse_handle_not_recompute_node_pass :     0.000028s :  0.32%
    jit_opt_a.j_node_and_user_rematch    :     0.000017s :  0.19%
    jit_opt_a.meta_fg_expand             :     0.000004s :  0.05%
    jit_opt_a.replace_old_param          :     0.000017s :  0.20%
    jit_opt_a.inline_without_move        :     0.000010s :  0.12%
    jit_opt_a.renormalize                :     0.000812s :  9.32%
    jit_opt_a.add_forward_monad_depend   :     0.000008s :  0.09%
    jit_opt_a.auto_monad_grad            :     0.000003s :  0.04%
    jit_opt_a.auto_monad_eliminator      :     0.000019s :  0.22%
    jit_opt_a.cse                        :     0.000039s :  0.45%
    jit_opt_a.replace_applicator         :     0.000018s :  0.20%
    py_interpret_to_execute_after_opt_a  :     0.000012s :  0.13%
    rewriter_after_opt_a                 :     0.000035s :  0.40%
    convert_after_rewriter               :     0.000008s :  0.09%
    order_py_execute_after_rewriter      :     0.000005s :  0.06%
    mutable_eliminate                    :     0.000644s :  7.39%
    jit_opt_b.frontend_op_eliminate      :     0.000018s :  0.21%
    jit_opt_b.inline_after_opt_a         :     0.000016s :  0.19%
    cconv                                :     0.000027s :  0.31%
    loop_unroll                          :     0.000446s :  5.12%
    jit_opt_after_cconv.c_1              :     0.000022s :  0.25%
    jit_opt_after_cconv.parameter_eliminate :     0.000003s :  0.04%
    jit_opt_after_cconv.updatestate_depend_eliminate :     0.000007s :  0.08%
    jit_opt_after_cconv.updatestate_assign_eliminate :     0.000002s :  0.03%
    jit_opt_after_cconv.updatestate_loads_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.cse              :     0.000022s :  0.25%
    jit_opt_after_cconv.call_graph_tuple_transform :     0.000020s :  0.23%
    jit_opt_after_cconv.tuple_list_get_item_eliminator :     0.000005s :  0.06%
    jit_opt_after_cconv.none_parameter_eliminate :     0.000002s :  0.02%
    jit_opt_after_cconv.renormalize      :     0.000001s :  0.01%
    jit_opt_after_cconv.switch_simplify  :     0.000005s :  0.06%
    remove_dup_value                     :     0.000016s :  0.18%
    partial_unused_args_eliminate        :     0.000002s :  0.03%
    environ_conv                         :     0.000006s :  0.07%
    add_recomputation                    :     0.000051s :  0.59%
    cse_after_recomputation.cse          :     0.000010s :  0.12%
    auto_monad_reorder                   :     0.000016s :  0.18%
    get_jit_bprop_graph                  :     0.000002s :  0.02%
    rewriter_after_jit_bprop_graph       :     0.000006s :  0.07%
    opt_after_jit_grad                   :     0.000482s :  5.53%
    symbol_engine_optimizer.build        :     0.000004s :  0.05%
    symbol_engine_optimizer.elim_shapecalc :     0.000008s :  0.09%
    symbol_engine_optimizer.elim_not_effective :     0.000013s :  0.15%
    symbol_engine_optimizer.opt_reshape  :     0.000006s :  0.07%
    symbol_engine_optimizer.fold_const_symbol :     0.000009s :  0.10%
    symbol_engine_optimizer.renormalize  :     0.000001s :  0.01%
    validate                             :     0.000042s :  0.48%


Time group info:
------[substitution.]   0.000224    23
 0.92% :     0.000002s :      2: substitution.elim_not_effective
 0.88% :     0.000002s :      2: substitution.fold_const_symbol
 2.48% :     0.000006s :      3: substitution.graph_param_transform
84.38% :     0.000189s :      4: substitution.inline
 1.34% :     0.000003s :      4: substitution.j_node_and_user_rematch
 2.09% :     0.000005s :      4: substitution.remove_not_recompute_node
 1.96% :     0.000004s :      2: substitution.replace_old_param
 5.96% :     0.000013s :      2: substitution.tuple_list_get_item_eliminator
------[type_inference.]   0.004332     2
80.42% :     0.003484s :      1: type_inference.infer
19.58% :     0.000848s :      1: type_inference.specialize
------[replace.]   0.000070     6
66.35% :     0.000047s :      4: replace.inline
33.65% :     0.000024s :      2: replace.tuple_list_get_item_eliminator
------[match.]   0.000199     6
93.78% :     0.000186s :      4: match.inline
 6.22% :     0.000012s :      2: match.tuple_list_get_item_eliminator
------[predicate.]   0.000115   691
 1.25% :     0.000001s :     10: predicate.accumulaten_eliminater
 1.25% :     0.000001s :      3: predicate.ad_related_special_op_eliminate
 1.02% :     0.000001s :     10: predicate.addn_check_dump
 1.29% :     0.000001s :     10: predicate.addn_zero_filter
 2.07% :     0.000002s :     10: predicate.arithmetic_simplify
 1.37% :     0.000002s :     10: predicate.cast_eliminate
 0.41% :     0.000000s :      3: predicate.check_bprop_eliminate
 1.03% :     0.000001s :     10: predicate.compare_switch_simplify
 1.33% :     0.000002s :     10: predicate.depend_value_elim
 1.10% :     0.000001s :     10: predicate.dict_get_item_const_eliminator
 1.28% :     0.000001s :     10: predicate.dict_get_item_eliminator
 1.55% :     0.000002s :     10: predicate.dict_set_item_eliminator
 0.91% :     0.000001s :      3: predicate.dumpgradient_eliminate
 0.27% :     0.000000s :      3: predicate.elim_not_effective
 0.83% :     0.000001s :      3: predicate.elim_shapecalc_of_broadcastargs
 1.09% :     0.000001s :     10: predicate.environ_add_const_eliminate
 1.07% :     0.000001s :     10: predicate.environ_get_add_eliminate
 1.17% :     0.000001s :     10: predicate.environ_get_depend_swap
 1.30% :     0.000001s :     10: predicate.environ_get_eliminate
 1.13% :     0.000001s :     10: predicate.environ_get_set_eliminate
 0.23% :     0.000000s :      3: predicate.fold_const_symbol
 0.93% :     0.000001s :      6: predicate.get_grad_eliminate
 0.27% :     0.000000s :      3: predicate.graph_param_transform
 5.85% :     0.000007s :     22: predicate.inline
 1.02% :     0.000001s :      6: predicate.inline_without_move
 0.36% :     0.000000s :      6: predicate.j_node_and_user_rematch
 1.23% :     0.000001s :      6: predicate.less_batch_normalization
 1.51% :     0.000002s :     12: predicate.list_to_tuple_eliminator_
 1.80% :     0.000002s :     15: predicate.load_eliminater
 1.60% :     0.000002s :      3: predicate.loop_unroll_after_grad
 3.91% :     0.000004s :     34: predicate.loop_unroll_before_grad
 2.06% :     0.000002s :     13: predicate.make_slice_get_slice_eliminator
 1.09% :     0.000001s :     10: predicate.merge_addn
 1.09% :     0.000001s :     10: predicate.minmaximum_grad
 2.38% :     0.000003s :      3: predicate.mutable_eliminate
 0.57% :     0.000001s :      3: predicate.opt_reshape
 2.49% :     0.000003s :     15: predicate.partial_eliminate
 1.10% :     0.000001s :     10: predicate.print_const_string_wrapper
 1.66% :     0.000002s :     10: predicate.reduce_eliminate
 1.49% :     0.000002s :     12: predicate.redundant_stop_gradient_eliminater
 0.60% :     0.000001s :      6: predicate.remove_not_recompute_node
 1.65% :     0.000002s :     18: predicate.replace_applicator
 0.73% :     0.000001s :      6: predicate.replace_old_param
 0.36% :     0.000000s :      3: predicate.reset_defer_inline
 1.38% :     0.000002s :     10: predicate.reshape_eliminate
 1.20% :     0.000001s :     10: predicate.row_tensor_add_zeros_like
 0.88% :     0.000001s :      3: predicate.row_tensor_eliminate
 1.18% :     0.000001s :     10: predicate.same_eliminate
 0.67% :     0.000001s :      6: predicate.set_cell_output_no_recompute
 0.89% :     0.000001s :      6: predicate.special_op_eliminate
 1.03% :     0.000001s :      6: predicate.specialize_transform
 1.29% :     0.000001s :     10: predicate.split_environ_get_set_with_tuple_value
 1.25% :     0.000001s :     10: predicate.stack_unstack_eliminate
 0.45% :     0.000001s :      3: predicate.switch_call_monad_eliminater
 2.68% :     0.000003s :     16: predicate.switch_defer_inline
 2.05% :     0.000002s :     16: predicate.switch_layer_defer_inline
 7.61% :     0.000009s :     53: predicate.switch_simplify
 1.44% :     0.000002s :     10: predicate.tile_eliminate
 1.14% :     0.000001s :     10: predicate.transpose_eliminate
 1.35% :     0.000002s :     10: predicate.tuple_list_convert_item_index_to_positive
 1.35% :     0.000002s :     10: predicate.tuple_list_get_item_depend_reorder
 3.84% :     0.000004s :     18: predicate.tuple_list_get_item_eliminator
 1.30% :     0.000001s :     10: predicate.tuple_list_set_item_eliminator
 1.46% :     0.000002s :     12: predicate.tuple_to_list_eliminator_
 1.62% :     0.000002s :     15: predicate.updatestate_pure_node_eliminater
 2.84% :     0.000003s :     21: predicate.updatestate_useless_node_eliminater
 1.49% :     0.000002s :     10: predicate.value_based_eliminate
 0.31% :     0.000000s :      3: predicate.virtual_view_grad_eliminate
 0.62% :     0.000001s :      3: predicate.zero_like_fill_zero
------[func_graph_cloner_run.]   0.000629    12
47.32% :     0.000298s :      5: func_graph_cloner_run.FuncGraphClonerGraph
 7.32% :     0.000046s :      1: func_graph_cloner_run.FuncGraphClonerNode
45.36% :     0.000285s :      6: func_graph_cloner_run.FuncGraphSpecializer
------[meta_graph.]   0.000000     0
------[manager.]   0.000000     0
------[pynative]   0.000000     0
------[others.]   0.014640    72
 0.37% :     0.000054s :      1: add_recomputation
 0.46% :     0.000068s :      1: auto_monad
 0.12% :     0.000018s :      1: auto_monad_reorder
 2.29% :     0.000336s :      1: bootstrap
 0.20% :     0.000030s :      1: cconv
 0.07% :     0.000010s :      1: convert_after_rewriter
 0.17% :     0.000025s :      1: cse_after_recomputation
 0.06% :     0.000008s :      1: environ_conv
 0.14% :     0.000020s :      1: event_method
 0.04% :     0.000005s :      1: expand_dump_flag
 0.03% :     0.000005s :      1: get_jit_bprop_graph
 0.06% :     0.000008s :      1: graph_reusing
39.19% :     0.005738s :      1: jit_opt_a
 1.02% :     0.000150s :      1: jit_opt_after_cconv
 0.38% :     0.000056s :      1: jit_opt_b
 3.10% :     0.000454s :      1: loop_unroll
 4.47% :     0.000654s :      1: mutable_eliminate
 5.87% :     0.000859s :     26: opt.transform.jit_opt_a
 0.33% :     0.000048s :      4: opt.transform.jit_opt_after_cconv
 0.19% :     0.000028s :      4: opt.transform.jit_opt_b
 0.09% :     0.000013s :      1: opt.transform.loop_unroll_optimizer
 0.11% :     0.000016s :      1: opt.transform.mutable_eliminate
 0.15% :     0.000022s :      1: opt.transform.opt_after_jit_grad
 0.23% :     0.000033s :      4: opt.transform.symbol_engine_opt
 3.36% :     0.000492s :      1: opt_after_jit_grad
 0.05% :     0.000007s :      1: order_py_execute_after_rewriter
 0.03% :     0.000004s :      1: partial_unused_args_eliminate
 0.03% :     0.000005s :      1: pre_auto_parallel
 0.23% :     0.000034s :      1: py_interpret_to_execute
 0.09% :     0.000014s :      1: py_interpret_to_execute_after_opt_a
 0.13% :     0.000018s :      1: remove_dup_value
 2.87% :     0.000420s :      1: renormalize.infer
 2.62% :     0.000384s :      1: renormalize.specialize
 0.05% :     0.000008s :      1: rewriter_after_jit_bprop_graph
 0.26% :     0.000038s :      1: rewriter_after_opt_a
 0.51% :     0.000075s :      1: rewriter_before_opt_a
 0.55% :     0.000081s :      1: symbol_engine_optimizer
30.07% :     0.004402s :      1: type_inference

.
op_name: mint.nn.GELU, mode:pynative, test_op_reference...

op_name: mint.nn.functional.logsigmoid, mode:pynative, test_op_reference...

op_name: mint.nn.ReLU6, mode:pynative, test_op_reference...

op_name: mint.erfc, mode:pynative, test_op_reference...

op_name: mint.neg, mode:pynative, test_op_reference...

op_name: Tensor.log, mode:pynative, test_op_reference...

op_name: mint.sin, mode:pynative, test_op_reference...

op_name: Tensor.half, mode:pynative, test_op_reference...

op_name: mint.cosh, mode:pynative, test_op_reference...

op_name: mint.imag, mode:pynative, test_op_reference...

op_name: mint.reciprocal, mode:pynative, test_op_reference...

op_name: mint.clone, mode:pynative, test_op_reference...

op_name: mint.nn.PReLU, mode:pynative, test_op_reference...

op_name: mint.erfc, mode:pynative, test_op_reference...

op_name: mint.sigmoid, mode:pynative, test_op_reference...

op_name: Tensor.sqrt, mode:pynative, test_op_reference...

test_op_reference failed:
op_name: Tensor.sqrt
mode: pynative
dtype: Float64
OpSampleInput(
op_input=Tensor(shape=(3, 3, 20), dtype=Float64, mean=4.71981515393741, max=8.965200085553702, min=0.04815172576929516),
op_args=tuple(),
op_kwargs=dict(),
sample_name=str(Tensor.sqrt),
loss=NoneType(None))
============================================ERROR============================================
test_unary_op_reference_backward fail ('pynative', 'Tensor.sqrt') {}
Traceback (most recent call last):
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/common/multi_process_actuator.py", line 41, in case_actuators
    func(*args, **kwargs)
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/test_unary_ops.py", line 67, in test_unary_op_reference_backward
    fact.test_op_reference(grad_cmp=True)
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 1010, in test_op_reference
    raise e
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 991, in test_op_reference
    self.compare_with_torch(sample_inputs=sample_input, grad_cmp=True)
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 906, in compare_with_torch
    self.assert_equal(
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 397, in assert_equal
    default_golden_compare(expect, actual, rtol, atol)
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 372, in default_golden_compare
    allclose_nparray(expect, actual, rtol, atol)
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 311, in allclose_nparray
    _count_unequal_element(expect, actual, rtol, atol)
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 305, in _count_unequal_element
    assert (loss_count / total_count) < rtol, \
AssertionError: 
data_expected_std:[2.27857892 0.40160788 0.32694866 0.17475991 0.30968021 0.1866338
 0.26831569 0.17329099 0.34713248 0.22986098 0.16921578 0.22982845
 0.1749223  0.31559295 0.17921687 0.24892748 0.32052111 0.23109432
 0.20888909 0.48728868 0.16725693 0.18422527 0.17227653 0.17372795
 0.27982703 0.1819155  0.44461532 0.18384712 0.39506881 0.66047171
 0.25416264 0.31318248 0.18246081 0.36141517 0.16770981 0.42775326
 0.25798219 0.24207294 0.29829802]
data_me_error:[1.01583233e-026 1.80377712e-094 2.86666435e-152 7.34197722e+089
 6.49158597e+041 1.12647195e+037 5.56307629e-132 1.47254190e-181
 5.22401257e-191 1.31995812e-210 1.28799836e+079 9.49629864e-172
 5.10173658e-254 8.30921643e-177 7.16211113e-201 2.21142403e+079
 4.69943747e-225 8.08394408e-225 1.10612441e+079 1.28318383e-234
 1.40864643e-181 7.86280492e-153 7.28232489e-248 1.44838659e-176
 4.95401029e-142 1.94561548e-162 2.82053211e-195 4.95401028e-142
 1.94561548e-162 4.69956114e-225 2.94347278e-037 6.88153681e-225
 5.46451044e-201 6.47786493e-142 2.26950843e+027 1.80823170e+056
 5.09083108e-254 1.08233373e-282 2.12980584e+084]
loss:[2.27857892e+00 4.01607881e-01 3.26948660e-01 7.34197722e+89
 6.49158597e+41 1.12647195e+37 2.68315692e-01 1.73290986e-01
 3.47132483e-01 2.29860983e-01 1.28799836e+79 2.29828452e-01
 1.74922303e-01 3.15592951e-01 1.79216870e-01 2.21142403e+79
 3.20521108e-01 2.31094319e-01 1.10612441e+79 4.87288679e-01
 1.67256933e-01 1.84225272e-01 1.72276526e-01 1.73727946e-01
 2.79827026e-01 1.81915504e-01 4.44615323e-01 1.83847121e-01
 3.95068813e-01 6.60471714e-01 2.54162636e-01 3.13182481e-01
 1.82460813e-01 3.61415174e-01 2.26950843e+27 1.80823170e+56
 2.57982189e-01 2.42072944e-01 2.12980584e+84]

=======================================ERROR PRINT END=======================================

op_name: mint.nn.functional.leaky_relu, mode:pynative, test_op_reference...

op_name: mint.tanh, mode:pynative, test_op_reference...

op_name: mint.sqrt, mode:pynative, test_op_reference...

op_name: Tensor.sqrt, mode:pynative, test_op_reference...

op_name: mint.sinc, mode:pynative, test_op_reference...

op_name: Tensor.float, mode:pynative, test_op_reference...

op_name: Tensor.logical_not, mode:pynative, test_op_reference...

op_name: mint.asin, mode:pynative, test_op_reference...

op_name: mint.cos, mode:pynative, test_op_reference...

op_name: Tensor.reciprocal, mode:pynative, test_op_reference...

op_name: Tensor.byte, mode:pynative, test_op_reference...

op_name: Tensor.is_contiguous, mode:pynative, test_op_reference...

op_name: Tensor.to, mode:pynative, test_op_reference...

op_name: mint.nn.functional.hardsigmoid, mode:pynative, test_op_reference...

op_name: mint.nn.Identity, mode:pynative, test_op_reference...

op_name: mint.nn.ReLU, mode:pynative, test_op_reference...

op_name: mint.nn.Tanh, mode:pynative, test_op_reference...

op_name: mint.erfinv, mode:pynative, test_op_reference...

op_name: Tensor.floor, mode:pynative, test_op_reference...

op_name: Tensor.abs, mode:pynative, test_op_reference...

op_name: mint.sinc, mode:pynative, test_op_reference...

op_name: Tensor.half, mode:pynative, test_op_reference...

op_name: mint.nn.functional.selu, mode:pynative, test_op_reference...

op_name: mint.atan, mode:pynative, test_op_reference...

op_name: mint.cosh, mode:pynative, test_op_reference...

op_name: mint.reciprocal, mode:pynative, test_op_reference...

op_name: mint.sign, mode:pynative, test_op_reference...

op_name: mint.tile, mode:kbk, test_op_reference...

op_name: mint.clone, mode:kbk, test_op_reference...

op_name: mint.acosh, mode:kbk, test_op_reference...

op_name: mint.nn.functional.hardsigmoid, mode:pynative, test_op_reference...

op_name: mint.nn.ReLU, mode:pynative, test_op_reference...

op_name: mint.ceil, mode:pynative, test_op_reference...

op_name: mint.abs, mode:pynative, test_op_reference...

op_name: Tensor.square, mode:pynative, test_op_reference...

op_name: Tensor.long, mode:pynative, test_op_reference...

op_name: mint.asinh, mode:pynative, test_op_reference...

op_name: Tensor.numpy, mode:pynative, test_op_reference...

op_name: mint.empty_like, mode:pynative, test_op_reference...

op_name: mint.nn.GLU, mode:pynative, test_op_reference...

op_name: mint.nn.LogSigmoid, mode:pynative, test_op_reference...

op_name: mint.nn.ReLU6, mode:pynative, test_op_reference...

op_name: mint.erf, mode:pynative, test_op_reference...

op_name: mint.exp, mode:pynative, test_op_reference...

op_name: mint.abs, mode:pynative, test_op_reference...

op_name: Tensor.sigmoid, mode:pynative, test_op_reference...

op_name: mint.zeros_like, mode:pynative, test_op_reference...

op_name: Tensor.sin, mode:pynative, test_op_reference...

op_name: Tensor.float, mode:pynative, test_op_reference...

op_name: Tensor.logical_not, mode:pynative, test_op_reference...

op_name: mint.acos, mode:pynative, test_op_reference...

op_name: mint.acosh, mode:pynative, test_op_reference...

op_name: Tensor.cos, mode:pynative, test_op_reference...

op_name: mint.trunc, mode:pynative, test_op_reference...

op_name: mint.nn.functional.relu, mode:kbk, test_op_reference...

op_name: Tensor.contiguous, mode:kbk, test_op_reference...

op_name: mint.nn.functional.hardswish, mode:pynative, test_op_reference...

op_name: mint.nn.functional.relu, mode:pynative, test_op_reference...

op_name: mint.floor, mode:pynative, test_op_reference...

op_name: Tensor.floor, mode:pynative, test_op_reference...

op_name: mint.ones_like, mode:pynative, test_op_reference...

op_name: Tensor.bool, mode:pynative, test_op_reference...

op_name: mint.nn.functional.selu, mode:pynative, test_op_reference...

op_name: mint.acosh, mode:pynative, test_op_reference...

op_name: mint.real, mode:pynative, test_op_reference...

op_name: mint.expm1, mode:pynative, test_op_reference...

op_name: Tensor.clone, mode:pynative, test_op_reference...

op_name: mint.nn.GELU, mode:pynative, test_op_reference...

op_name: mint.nn.functional.leaky_relu, mode:pynative, test_op_reference...

op_name: Tensor.tanh, mode:pynative, test_op_reference...

op_name: mint.neg, mode:pynative, test_op_reference...

op_name: Tensor.ceil, mode:pynative, test_op_reference...

op_name: Tensor.neg, mode:pynative, test_op_reference...

op_name: mint.ones_like, mode:pynative, test_op_reference...

op_name: Tensor.bfloat16, mode:pynative, test_op_reference...

op_name: Tensor.bool, mode:pynative, test_op_reference...

op_name: mint.logical_not, mode:pynative, test_op_reference...

op_name: mint.tile, mode:pynative, test_op_reference...

op_name: mint.atanh, mode:pynative, test_op_reference...

op_name: mint.asinh, mode:pynative, test_op_reference...

op_name: mint.tanh, mode:kbk, test_op_reference...

op_name: mint.tanh, mode:kbk, test_op_reference...

op_name: mint.nn.functional.relu, mode:kbk, test_op_reference...

op_name: mint.tile, mode:kbk, test_op_reference...

op_name: mint.clone, mode:kbk, test_op_reference...

op_name: mint.empty_like, mode:pynative, test_op_reference...

op_name: mint.nn.Identity, mode:pynative, test_op_reference...

op_name: mint.erf, mode:pynative, test_op_reference...

op_name: mint.exp, mode:pynative, test_op_reference...

op_name: Tensor.ceil, mode:pynative, test_op_reference...

op_name: Tensor.sigmoid, mode:pynative, test_op_reference...

op_name: Tensor.bfloat16, mode:pynative, test_op_reference...

op_name: mint.tile, mode:pynative, test_op_reference...

op_name: mint.acos, mode:pynative, test_op_reference...

op_name: mint.sinh, mode:pynative, test_op_reference...

op_name: mint.trunc, mode:pynative, test_op_reference...

op_name: Tensor.astype, mode:pynative, test_op_reference...

op_name: Tensor.contiguous, mode:pynative, test_op_reference...

op_name: mint.empty, mode:pynative, test_op_reference...

op_name: mint.nn.functional.hardswish, mode:pynative, test_op_reference...

op_name: mint.nn.functional.relu, mode:pynative, test_op_reference...

op_name: mint.tanh, mode:pynative, test_op_reference...

op_name: mint.ceil, mode:pynative, test_op_reference...

op_name: mint.log, mode:pynative, test_op_reference...

op_name: Tensor.exp, mode:pynative, test_op_reference...

op_name: Tensor.log, mode:pynative, test_op_reference...

op_name: Tensor.square, mode:pynative, test_op_reference...

op_name: mint.sin, mode:pynative, test_op_reference...

op_name: Tensor.double, mode:pynative, test_op_reference...

op_name: Tensor.int, mode:pynative, test_op_reference...

op_name: Tensor.tile, mode:pynative, test_op_reference...

op_name: mint.sinh, mode:pynative, test_op_reference...

op_name: mint.cos, mode:pynative, test_op_reference...

op_name: Tensor.reciprocal, mode:pynative, test_op_reference...

op_name: mint.rsqrt, mode:pynative, test_op_reference...

test_op_reference failed:
op_name: mint.rsqrt
mode: pynative
dtype: Float64
OpSampleInput(
op_input=Tensor(shape=(3, 3, 20), dtype=Float64, mean=4.71986271154681, max=8.965200472219419, min=0.04825119075011995),
op_args=tuple(),
op_kwargs=dict(),
sample_name=str(mint.rsqrt),
loss=NoneType(None))
============================================ERROR============================================
test_unary_op_reference_backward fail ('pynative', 'mint.rsqrt') {}
Traceback (most recent call last):
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/common/multi_process_actuator.py", line 41, in case_actuators
    func(*args, **kwargs)
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/test_unary_ops.py", line 67, in test_unary_op_reference_backward
    fact.test_op_reference(grad_cmp=True)
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 1010, in test_op_reference
    raise e
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 991, in test_op_reference
    self.compare_with_torch(sample_inputs=sample_input, grad_cmp=True)
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 906, in compare_with_torch
    self.assert_equal(
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 397, in assert_equal
    default_golden_compare(expect, actual, rtol, atol)
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 372, in default_golden_compare
    allclose_nparray(expect, actual, rtol, atol)
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 311, in allclose_nparray
    _count_unequal_element(expect, actual, rtol, atol)
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 305, in _count_unequal_element
    assert (loss_count / total_count) < rtol, \
AssertionError: 
data_expected_std:[-0.11879074 -0.02600334 -0.07726569 -0.02081552 -0.16730988 -0.04857908
 -0.01938128 -0.04855845 -0.02140892 -0.12572544 -0.02302478 -0.06169779
 -0.13170746 -0.04936523 -0.03645887 -0.46276907 -0.01871597 -0.02500956
 -0.0204521  -0.02097338 -0.08764271 -0.02408062 -0.35153535 -0.02485587
 -0.24662937 -1.15216914 -0.0656728  -0.12286666 -0.02429782 -0.18882189
 -0.01886841 -0.31303978 -0.06867822 -0.05674018 -0.10616837 -0.02126521
 -0.04398739 -0.03483827 -0.02161986 -0.05709893]
data_me_error:[-2.44855962e+221 -2.76978373e+020 -2.79175291e-082 -7.33338332e-272
 -1.32553920e-141 -5.07592408e-257             -inf             -inf
             -inf             -inf             -inf             -inf
 -8.26511521e-225             -inf             -inf             -inf
 -1.34135873e-239             -inf             -inf             -inf
             -inf             -inf             -inf             -inf
 -8.06512032e+265             -inf -5.08372125e-081             -inf
             -inf -4.26553699e-226             -inf             -inf
             -inf -0.00000000e+000             -inf -1.05691173e-170
             -inf             -inf             -inf -6.46933999e-255]
loss:[2.44855962e+221 2.76978373e+020 7.72656899e-002 2.08155227e-002
 1.67309883e-001 4.85790751e-002             inf             inf
             inf             inf             inf             inf
 1.31707457e-001             inf             inf             inf
 1.87159695e-002             inf             inf             inf
             inf             inf             inf             inf
 8.06512032e+265             inf 6.56727976e-002             inf
             inf 1.88821891e-001             inf             inf
             inf 5.67401808e-002             inf 2.12652113e-002
             inf             inf             inf 5.70989262e-002]

=======================================ERROR PRINT END=======================================

op_name: mint.empty, mode:pynative, test_op_reference...

op_name: mint.nn.functional.gelu, mode:pynative, test_op_reference...

op_name: mint.nn.GLU, mode:pynative, test_op_reference...

op_name: mint.nn.PReLU, mode:pynative, test_op_reference...

op_name: mint.nn.Tanh, mode:pynative, test_op_reference...

op_name: mint.exp2, mode:pynative, test_op_reference...

op_name: mint.sigmoid, mode:pynative, test_op_reference...

op_name: Tensor.neg, mode:pynative, test_op_reference...

op_name: Tensor.abs, mode:pynative, test_op_reference...

op_name: Tensor.sin, mode:pynative, test_op_reference...

op_name: Tensor.double, mode:pynative, test_op_reference...

op_name: Tensor.tile, mode:pynative, test_op_reference...

op_name: mint.atanh, mode:pynative, test_op_reference...

op_name: mint.atan, mode:pynative, test_op_reference...

op_name: mint.sign, mode:pynative, test_op_reference...

op_name: mint.nn.functional.gelu, mode:pynative, test_op_reference...

op_name: mint.nn.functional.glu, mode:pynative, test_op_reference...

op_name: mint.nn.functional.logsigmoid, mode:pynative, test_op_reference...

op_name: mint.nn.functional.relu6, mode:pynative, test_op_reference...

op_name: mint.floor, mode:pynative, test_op_reference...

op_name: mint.exp2, mode:pynative, test_op_reference...

op_name: mint.sqrt, mode:pynative, test_op_reference...

test_op_reference failed:
op_name: mint.sqrt
mode: pynative
dtype: Float64
OpSampleInput(
op_input=Tensor(shape=(3, 3, 20), dtype=Float64, mean=4.71981515393741, max=8.965200085553702, min=0.04815172576929516),
op_args=tuple(),
op_kwargs=dict(),
sample_name=str(mint.sqrt),
loss=NoneType(None))
============================================ERROR============================================
test_unary_op_reference_backward fail ('pynative', 'mint.sqrt') {}
Traceback (most recent call last):
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/common/multi_process_actuator.py", line 41, in case_actuators
    func(*args, **kwargs)
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/test_unary_ops.py", line 67, in test_unary_op_reference_backward
    fact.test_op_reference(grad_cmp=True)
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 1010, in test_op_reference
    raise e
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 991, in test_op_reference
    self.compare_with_torch(sample_inputs=sample_input, grad_cmp=True)
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 906, in compare_with_torch
    self.assert_equal(
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 397, in assert_equal
    default_golden_compare(expect, actual, rtol, atol)
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 372, in default_golden_compare
    allclose_nparray(expect, actual, rtol, atol)
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 311, in allclose_nparray
    _count_unequal_element(expect, actual, rtol, atol)
  File "/home/jenkins/mindspore/testcases/testcases/tests/st/ops/share/_internal/meta.py", line 305, in _count_unequal_element
    assert (loss_count / total_count) < rtol, \
AssertionError: 
data_expected_std:[2.27857892 0.40160788 0.32694866 0.17475991 0.30968021 0.1866338
 0.26831569 0.17329099 0.34713248 0.22986098 0.16921578 0.22982845
 0.1749223  0.31559295 0.17921687 0.24892748 0.32052111 0.23109432
 0.20888909 0.48728868 0.16725693 0.18422527 0.17227653 0.17372795
 0.27982703 0.1819155  0.44461532 0.18384712 0.39506881 0.66047171
 0.25416264 0.31318248 0.18246081 0.36141517 0.16770981 0.42775326
 0.25798219 0.24207294 0.29829802]
data_me_error:[1.01583233e-026 1.80377712e-094 2.86666435e-152 7.34197722e+089
 6.49158597e+041 1.36182102e+061 5.56848635e-132 1.47254190e-181
 5.22401257e-191 1.31995812e-210 1.28799836e+079 9.49629864e-172
 5.10173658e-254 8.30921643e-177 7.16211113e-201 2.21142403e+079
 4.69943747e-225 8.08394408e-225 1.10612441e+079 1.28318383e-234
 1.40864643e-181 7.86280492e-153 7.28232489e-248 1.44838659e-176
 4.95401029e-142 1.94561548e-162 2.82053211e-195 4.95401028e-142
 1.94561548e-162 4.69956114e-225 2.94347278e-037 6.88153681e-225
 5.46451044e-201 6.47786493e-142 2.26950843e+027 1.80823170e+056
 5.09083108e-254 1.08233373e-282 2.12980584e+084]
loss:[2.27857892e+00 4.01607881e-01 3.26948660e-01 7.34197722e+89
 6.49158597e+41 1.36182102e+61 2.68315692e-01 1.73290986e-01
 3.47132483e-01 2.29860983e-01 1.28799836e+79 2.29828452e-01
 1.74922303e-01 3.15592951e-01 1.79216870e-01 2.21142403e+79
 3.20521108e-01 2.31094319e-01 1.10612441e+79 4.87288679e-01
 1.67256933e-01 1.84225272e-01 1.72276526e-01 1.73727946e-01
 2.79827026e-01 1.81915504e-01 4.44615323e-01 1.83847121e-01
 3.95068813e-01 6.60471714e-01 2.54162636e-01 3.13182481e-01
 1.82460813e-01 3.61415174e-01 2.26950843e+27 1.80823170e+56
 2.57982189e-01 2.42072944e-01 2.12980584e+84]

=======================================ERROR PRINT END=======================================

op_name: mint.nn.functional.relu6, mode:pynative, test_op_reference...

op_name: mint.erfinv, mode:pynative, test_op_reference...

op_name: Tensor.exp, mode:pynative, test_op_reference...

op_name: mint.zeros_like, mode:pynative, test_op_reference...

op_name: Tensor.int, mode:pynative, test_op_reference...

op_name: mint.rsqrt, mode:pynative, test_op_reference...

op_name: mint.imag, mode:pynative, test_op_reference...

op_name: mint.expm1, mode:pynative, test_op_reference...

op_name: mint.acosh, mode:kbk, test_op_reference...

op_name: mint.nn.functional.glu, mode:pynative, test_op_reference...

op_name: mint.nn.LogSigmoid, mode:pynative, test_op_reference...

op_name: Tensor.tanh, mode:pynative, test_op_reference...

op_name: mint.log, mode:pynative, test_op_reference...

op_name: mint.tan, mode:pynative, test_op_reference...

op_name: mint.logical_not, mode:pynative, test_op_reference...

op_name: Tensor.cos, mode:pynative, test_op_reference...

op_name: mint.tan, mode:pynative, test_op_reference...

op_name: Tensor.long, mode:pynative, test_op_reference...

op_name: mint.asin, mode:pynative, test_op_reference...

op_name: mint.real, mode:pynative, test_op_reference...

op_name: Tensor.contiguous, mode:kbk, test_op_reference...
group_cases_0 have all been run, results of sub cases are below:
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.byte') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.clone') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.empty') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.clone') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.contiguous') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.nn.functional.gelu') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.numpy') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.astype') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.to') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.nn.GELU') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.empty_like') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.nn.functional.glu') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.nn.GLU') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.nn.functional.hardswish') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.nn.functional.hardsigmoid') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.is_contiguous') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.nn.functional.logsigmoid') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.nn.functional.leaky_relu') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.nn.LogSigmoid') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.nn.PReLU') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.nn.functional.relu') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.nn.ReLU') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.nn.Identity') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.nn.ReLU6') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.nn.functional.relu6') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.nn.Tanh') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.erf') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.tanh') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.erfc') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.exp2') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.tanh') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.ceil') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.floor') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.exp') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.erfinv') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.neg') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.sigmoid') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.ceil') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.sqrt') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.neg') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.abs') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.floor') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.exp') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.log') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.log') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.sqrt') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.abs') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.sigmoid') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.ones_like') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.sin') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.sinc') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.sin') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.zeros_like') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.square') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.tan') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.float') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.bool') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.bfloat16') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.double') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.tile') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.tile') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.atanh') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.acos') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.nn.functional.selu') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.long') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.logical_not') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.half') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.asin') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.logical_not') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.cosh') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.acosh') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.cos') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.imag') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.sinh') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.real') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.int') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.atan') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.trunc') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.reciprocal') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.expm1') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.astype') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.byte') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.asinh') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.is_contiguous') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.reciprocal') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.contiguous') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.numpy') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.sign') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.clone') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.nn.functional.gelu') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.empty_like') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.to') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.empty') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.nn.GELU') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.nn.functional.glu') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.nn.functional.hardsigmoid') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.nn.GLU') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.clone') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.nn.functional.hardswish') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.nn.Identity') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.nn.functional.logsigmoid') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.nn.LogSigmoid') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.nn.functional.relu') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.nn.functional.leaky_relu') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.nn.ReLU') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.nn.functional.relu6') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.tanh') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.nn.ReLU6') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.nn.PReLU') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.nn.Tanh') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.floor') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.erf') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.ceil') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.tanh') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.erfc') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.exp2') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.exp') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.erfinv') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.neg') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.log') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.exp') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.ceil') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.abs') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.sigmoid') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.floor') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.log') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.neg') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.sigmoid') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'Tensor.cos') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.square') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.abs') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.zeros_like') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.ones_like') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.bfloat16') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.sin') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.sin') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.sinc') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.double') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.tan') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.bool') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.float') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.half') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.logical_not') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.int') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.logical_not') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.tile') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.nn.functional.selu') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.tile') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.acos') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.atanh') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.long') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.atan') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.sinh') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.acosh') {} pass.
case: <function test_unary_op_reference_forward at 0xffff7d3a2e50> ('pynative', 'mint.rsqrt') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.asin') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.cos') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.cosh') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.cos') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.imag') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.reciprocal') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.reciprocal') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.asinh') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.trunc') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.sign') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.expm1') {} pass.
case: <function test_unary_op_reference_forward_kbk at 0xffff7d369040> ('kbk', 'mint.tile') {} pass.
case: <function test_unary_op_reference_forward_kbk at 0xffff7d369040> ('kbk', 'mint.nn.functional.relu') {} pass.
case: <function test_unary_op_reference_forward_kbk at 0xffff7d369040> ('kbk', 'mint.tanh') {} pass.
case: <function test_unary_op_reference_backward_kbk at 0xffff7d3690d0> ('kbk', 'mint.tanh') {} pass.
case: <function test_unary_op_reference_backward_kbk at 0xffff7d3690d0> ('kbk', 'mint.nn.functional.relu') {} pass.
case: <function test_unary_op_reference_forward_kbk at 0xffff7d369040> ('kbk', 'mint.clone') {} pass.
case: <function test_unary_op_reference_backward_kbk at 0xffff7d3690d0> ('kbk', 'mint.tile') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.real') {} pass.
case: <function test_unary_op_reference_forward_kbk at 0xffff7d369040> ('kbk', 'mint.acosh') {} pass.
case: <function test_unary_op_reference_backward_kbk at 0xffff7d3690d0> ('kbk', 'mint.clone') {} pass.
case: <function test_unary_op_reference_backward_kbk at 0xffff7d3690d0> ('kbk', 'mint.acosh') {} pass.
case: <function test_unary_op_reference_backward_kbk at 0xffff7d3690d0> ('kbk', 'Tensor.contiguous') {} pass.
case: <function test_unary_op_reference_forward_kbk at 0xffff7d369040> ('kbk', 'Tensor.contiguous') {} pass.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.sqrt') {} fail.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'Tensor.sqrt') {} fail.
case: <function test_unary_op_reference_backward at 0xffff7d3a2f70> ('pynative', 'mint.rsqrt') {} fail.

F
[hook] pytest_runtest_teardown:test_unary_op_group_case_level0

tests/st/ops/allcases_onecard/test_ops_group_cases.py::test_unary_op_group_case_level0,max_mem:2.0M


=================================== FAILURES ===================================
_______________________ test_unary_op_group_case_level0 ________________________

    @arg_mark(plat_marks=['platform_ascend910b',
                          'platform_gpu',
                          'cpu_linux', 'cpu_windows', 'cpu_macos'],
              level_mark='level0',
              card_mark='onecard',
              essential_mark='essential', mem_peak=0.01)
    @pytest.mark.no_taskset
    def test_unary_op_group_case_level0():
        """
        Feature: Test all level0 unary_op cases.
        Description: Run level0 unary_op cases on the gate.
        Expectation: success
        """
        platform = get_local_platform()
>       ops_group_cases_process(platform, "level0", "unary_op")

test_ops_group_cases.py:511: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
test_ops_group_cases.py:422: in ops_group_cases_process
    run_group_cases(group_id, group_cases, memory_used, memory_threshold)
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

group_id = 0
group_cases = [(<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.astype'), (<function test_unary_op...ensor.is_contiguous'), (<function test_unary_op_reference_forward at 0xffff7d3a2e50>, 'pynative', 'Tensor.numpy'), ...]
memory_used = 2912, memory_threshold = 51200

    def run_group_cases(group_id, group_cases, memory_used, memory_threshold):
        msg = f"\nops group_cases_{group_id} with {len(group_cases)} cases start to running, all cases are below:\n"
        for case in group_cases:
            msg += f"case: {case}\n"
        msg += f"ops group_cases_{group_id} total running memory: {memory_used}M, memory threshold: {memory_threshold}M"
        print(msg)
    
        all_result = True
        result = run_cases_multi_process(group_cases)
        msg = f"group_cases_{group_id} have all been run, results of sub cases are below:\n"
        for ret in result:
            if ret[1]:
                msg += f"case: {ret[0]} pass.\n"
            else:
                msg += f"case: {ret[0]} fail.\n"
                all_result = False
        print(msg)
>       assert all_result
E       assert False

test_ops_group_cases.py:350: AssertionError
=============================== warnings summary ===============================
../../../../../../../../../usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/classifier/transdata/transdata_classifier.py:222
  /usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/classifier/transdata/transdata_classifier.py:222: DeprecationWarning: invalid escape sequence \B
    """

../../../../../../../../../usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:143
  /usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:143: DeprecationWarning: invalid escape sequence \c
    """

../../../../../../../../../usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:170
  /usr/local/Ascend/cann-8.5.0/python/site-packages/tbe/dsl/unify_schedule/vector/transdata/common/graph/transdata_graph_info.py:170: DeprecationWarning: invalid escape sequence \c
    """

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for <class 'numpy.float64'> type is zero.
    setattr(self, word, getattr(machar, word).flat[0])

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class 'numpy.float64'> type is zero.
    return self._float_to_str(self.smallest_subnormal)

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for <class 'numpy.float32'> type is zero.
    setattr(self, word, getattr(machar, word).flat[0])

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class 'numpy.float32'> type is zero.
    return self._float_to_str(self.smallest_subnormal)

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2.py:57
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2.py:57: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute
    @fusion_manager.register("batchnorm_fold2")

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad.py:56
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad.py:56: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute
    @fusion_manager.register("batchnorm_fold2_grad")

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad_reduce.py:48
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/batchnorm_fold2_grad_reduce.py:48: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute
    @fusion_manager.register("batchnorm_fold2_grad_reduce")

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul.py:51
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul.py:51: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute
    @fusion_manager.register("correction_mul")

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:51
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:51: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute
    @fusion_manager.register("correction_mul_grad")

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:143
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/correction_mul_grad.py:143: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute
    @fusion_manager.register("correction_mul_grad_reduce")

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer.py:50
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute
    @fusion_manager.register("fake_learned_scale_quant_perlayer")

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad.py:92
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad.py:92: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute
    @fusion_manager.register("fake_learned_scale_quant_perlayer_grad_d")

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad_reduce.py:49
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perlayer_grad_reduce.py:49: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute
    @fusion_manager.register("fake_learned_scale_quant_perlayer_grad_d_reduce")

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel.py:50
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute
    @fusion_manager.register("fake_learned_scale_quant_perchannel")

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad.py:91
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad.py:91: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute
    @fusion_manager.register("fake_learned_scale_quant_perchannel_grad_d")

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad_reduce.py:48
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_learned_scale_quant_perchannel_grad_reduce.py:48: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute
    @fusion_manager.register("fake_learned_scale_quant_perchannel_grad_d_reduce")

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel.py:52
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel.py:52: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute
    @fusion_manager.register("fake_quant_perchannel")

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel_grad.py:81
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perchannel_grad.py:81: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute
    @fusion_manager.register("fake_quant_perchannel_grad")

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer.py:54
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer.py:54: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute
    @fusion_manager.register("fake_quant_per_layer")

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer_grad.py:81
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/fake_quant_perlayer_grad.py:81: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute
    @fusion_manager.register("fake_quant_per_layer_grad")

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perchannel.py:50
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perchannel.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute
    @fusion_manager.register("minmax_update_perchannel")

../../../../../../../anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perlayer.py:50
  /home/jenkins/anaconda3/envs/ci39/lib/python3.9/site-packages/mindspore/ops/_op_impl/_custom_op/minmax_update_perlayer.py:50: DeprecationWarning: te_fusion.fusion_manager.fusion_manager.register is deprecated,please replace it with tbe.common.register.register_op_compute
    @fusion_manager.register("minmax_update_perlayer")

-- Docs: https://docs.pytest.org/en/stable/warnings.html
=========================== short test summary info ============================
FAILED test_ops_group_cases.py::test_unary_op_group_case_level0 - assert False
================== 1 failed, 25 warnings in 597.88s (0:09:57) ==================